2781 files changed, 105252 insertions, 44746 deletions

diff --git a/Documentation/ABI/stable/sysfs-driver-ib_srp b/Documentation/ABI/stable/sysfs-driver-ib_srp
index 5c53d28f775c..b9688de8455b 100644
--- a/Documentation/ABI/stable/sysfs-driver-ib_srp
+++ b/Documentation/ABI/stable/sysfs-driver-ib_srp
@@ -61,6 +61,12 @@ Description:	Interface for making ib_srp connect to a new target.
 		  interrupt is handled by a different CPU then the comp_vector
 		  parameter can be used to spread the SRP completion workload
 		  over multiple CPU's.
+		* tl_retry_count, a number in the range 2..7 specifying the
+		  IB RC retry count.
+		* queue_size, the maximum number of commands that the
+		  initiator is allowed to queue per SCSI host. The default
+		  value for this parameter is 62. The lowest supported value
+		  is 2.
 
 What:		/sys/class/infiniband_srp/srp-<hca>-<port_number>/ibdev
 Date:		January 2, 2006
@@ -153,6 +159,13 @@ Contact:	linux-rdma@vger.kernel.org
 Description:	InfiniBand service ID used for establishing communication with
 		the SRP	target.
 
+What:		/sys/class/scsi_host/host<n>/sgid
+Date:		February 1, 2014
+KernelVersion:	3.13
+Contact:	linux-rdma@vger.kernel.org
+Description:	InfiniBand GID of the source port used for communication with
+		the SRP target.
+
 What:		/sys/class/scsi_host/host<n>/zero_req_lim
 Date:		September 20, 2006
 KernelVersion:	2.6.18
diff --git a/Documentation/ABI/stable/sysfs-transport-srp b/Documentation/ABI/stable/sysfs-transport-srp
index b36fb0dc13c8..ec7af69fea0a 100644
--- a/Documentation/ABI/stable/sysfs-transport-srp
+++ b/Documentation/ABI/stable/sysfs-transport-srp
@@ -5,6 +5,24 @@ Contact:	linux-scsi@vger.kernel.org, linux-rdma@vger.kernel.org
 Description:	Instructs an SRP initiator to disconnect from a target and to
 		remove all LUNs imported from that target.
 
+What:		/sys/class/srp_remote_ports/port-<h>:<n>/dev_loss_tmo
+Date:		February 1, 2014
+KernelVersion:	3.13
+Contact:	linux-scsi@vger.kernel.org, linux-rdma@vger.kernel.org
+Description:	Number of seconds the SCSI layer will wait after a transport
+		layer error has been observed before removing a target port.
+		Zero means immediate removal. Setting this attribute to "off"
+		will disable the dev_loss timer.
+
+What:		/sys/class/srp_remote_ports/port-<h>:<n>/fast_io_fail_tmo
+Date:		February 1, 2014
+KernelVersion:	3.13
+Contact:	linux-scsi@vger.kernel.org, linux-rdma@vger.kernel.org
+Description:	Number of seconds the SCSI layer will wait after a transport
+		layer error has been observed before failing I/O. Zero means
+		failing I/O immediately. Setting this attribute to "off" will
+		disable the fast_io_fail timer.
+
 What:		/sys/class/srp_remote_ports/port-<h>:<n>/port_id
 Date:		June 27, 2007
 KernelVersion:	2.6.24
@@ -12,8 +30,29 @@ Contact:	linux-scsi@vger.kernel.org
 Description:	16-byte local SRP port identifier in hexadecimal format. An
 		example: 4c:49:4e:55:58:20:56:49:4f:00:00:00:00:00:00:00.
 
+What:		/sys/class/srp_remote_ports/port-<h>:<n>/reconnect_delay
+Date:		February 1, 2014
+KernelVersion:	3.13
+Contact:	linux-scsi@vger.kernel.org, linux-rdma@vger.kernel.org
+Description:	Number of seconds the SCSI layer will wait after a reconnect
+		attempt failed before retrying. Setting this attribute to
+		"off" will disable time-based reconnecting.
+
 What:		/sys/class/srp_remote_ports/port-<h>:<n>/roles
 Date:		June 27, 2007
 KernelVersion:	2.6.24
 Contact:	linux-scsi@vger.kernel.org
 Description:	Role of the remote port. Either "SRP Initiator" or "SRP Target".
+
+What:		/sys/class/srp_remote_ports/port-<h>:<n>/state
+Date:		February 1, 2014
+KernelVersion:	3.13
+Contact:	linux-scsi@vger.kernel.org, linux-rdma@vger.kernel.org
+Description:	State of the transport layer used for communication with the
+		remote port. "running" if the transport layer is operational;
+		"blocked" if a transport layer error has been encountered but
+		the fast_io_fail_tmo timer has not yet fired; "fail-fast"
+		after the fast_io_fail_tmo timer has fired and before the
+		"dev_loss_tmo" timer has fired; "lost" after the
+		"dev_loss_tmo" timer has fired and before the port is finally
+		removed.
diff --git a/Documentation/ABI/testing/sysfs-driver-hid-roccat-ryos b/Documentation/ABI/testing/sysfs-driver-hid-roccat-ryos
new file mode 100644
index 000000000000..1d6a8cf9dc0a
--- /dev/null
+++ b/Documentation/ABI/testing/sysfs-driver-hid-roccat-ryos
@@ -0,0 +1,178 @@
+What:		/sys/bus/usb/devices/<busnum>-<devnum>:<config num>.<interface num>/<hid-bus>:<vendor-id>:<product-id>.<num>/ryos/roccatryos<minor>/control
+Date:		October 2013
+Contact:	Stefan Achatz <erazor_de@users.sourceforge.net>
+Description:	When written, this file lets one select which data from which
+		profile will be	read next. The data has to be 3 bytes long.
+		This file is writeonly.
+Users:		http://roccat.sourceforge.net
+
+What:		/sys/bus/usb/devices/<busnum>-<devnum>:<config num>.<interface num>/<hid-bus>:<vendor-id>:<product-id>.<num>/ryos/roccatryos<minor>/profile
+Date:		October 2013
+Contact:	Stefan Achatz <erazor_de@users.sourceforge.net>
+Description:	The mouse can store 5 profiles which can be switched by the
+		press of a button. profile holds index of actual profile.
+		This value is persistent, so its value determines the profile
+		that's active when the device is powered on next time.
+		When written, the device activates the set profile immediately.
+		The data has to be 3 bytes long.
+		The device will reject invalid data.
+Users:		http://roccat.sourceforge.net
+
+What:		/sys/bus/usb/devices/<busnum>-<devnum>:<config num>.<interface num>/<hid-bus>:<vendor-id>:<product-id>.<num>/ryos/roccatryos<minor>/keys_primary
+Date:		October 2013
+Contact:	Stefan Achatz <erazor_de@users.sourceforge.net>
+Description:	When written, this file lets one set the default of all keys for
+		a specific profile. Profile index is included in written data.
+		The data has to be 125 bytes long.
+		Before reading this file, control has to be written to select
+		which profile to read.
+Users:		http://roccat.sourceforge.net
+
+What:		/sys/bus/usb/devices/<busnum>-<devnum>:<config num>.<interface num>/<hid-bus>:<vendor-id>:<product-id>.<num>/ryos/roccatryos<minor>/keys_function
+Date:		October 2013
+Contact:	Stefan Achatz <erazor_de@users.sourceforge.net>
+Description:	When written, this file lets one set the function of the
+		function keys for a specific profile. Profile index is included
+		in written data. The data has to be 95 bytes long.
+		Before reading this file, control has to be written to select
+		which profile to read.
+Users:		http://roccat.sourceforge.net
+
+What:		/sys/bus/usb/devices/<busnum>-<devnum>:<config num>.<interface num>/<hid-bus>:<vendor-id>:<product-id>.<num>/ryos/roccatryos<minor>/keys_macro
+Date:		October 2013
+Contact:	Stefan Achatz <erazor_de@users.sourceforge.net>
+Description:	When written, this file lets one set the function of the macro
+		keys for a specific profile. Profile index is included in
+		written data. The data has to be 35 bytes long.
+		Before reading this file, control has to be written to select
+		which profile to read.
+Users:		http://roccat.sourceforge.net
+
+What:		/sys/bus/usb/devices/<busnum>-<devnum>:<config num>.<interface num>/<hid-bus>:<vendor-id>:<product-id>.<num>/ryos/roccatryos<minor>/keys_thumbster
+Date:		October 2013
+Contact:	Stefan Achatz <erazor_de@users.sourceforge.net>
+Description:	When written, this file lets one set the function of the
+		thumbster keys for a specific profile. Profile index is included
+		in written data. The data has to be 23 bytes long.
+		Before reading this file, control has to be written to select
+		which profile to read.
+Users:		http://roccat.sourceforge.net
+
+What:		/sys/bus/usb/devices/<busnum>-<devnum>:<config num>.<interface num>/<hid-bus>:<vendor-id>:<product-id>.<num>/ryos/roccatryos<minor>/keys_extra
+Date:		October 2013
+Contact:	Stefan Achatz <erazor_de@users.sourceforge.net>
+Description:	When written, this file lets one set the function of the
+		capslock and function keys for a specific profile. Profile index
+		is included in written data. The data has to be 8 bytes long.
+		Before reading this file, control has to be written to select
+		which profile to read.
+Users:		http://roccat.sourceforge.net
+
+What:		/sys/bus/usb/devices/<busnum>-<devnum>:<config num>.<interface num>/<hid-bus>:<vendor-id>:<product-id>.<num>/ryos/roccatryos<minor>/keys_easyzone
+Date:		October 2013
+Contact:	Stefan Achatz <erazor_de@users.sourceforge.net>
+Description:	When written, this file lets one set the function of the
+		easyzone keys for a specific profile. Profile index is included
+		in written data. The data has to be 294 bytes long.
+		Before reading this file, control has to be written to select
+		which profile to read.
+Users:		http://roccat.sourceforge.net
+
+What:		/sys/bus/usb/devices/<busnum>-<devnum>:<config num>.<interface num>/<hid-bus>:<vendor-id>:<product-id>.<num>/ryos/roccatryos<minor>/key_mask
+Date:		October 2013
+Contact:	Stefan Achatz <erazor_de@users.sourceforge.net>
+Description:	When written, this file lets one deactivate certain keys like
+		windows and application keys, to prevent accidental presses.
+		Profile index for which this settings occur is included in
+		written data. The data has to be 6 bytes long.
+		Before reading this file, control has to be written to select
+		which profile to read.
+Users:		http://roccat.sourceforge.net
+
+What:		/sys/bus/usb/devices/<busnum>-<devnum>:<config num>.<interface num>/<hid-bus>:<vendor-id>:<product-id>.<num>/ryos/roccatryos<minor>/light
+Date:		October 2013
+Contact:	Stefan Achatz <erazor_de@users.sourceforge.net>
+Description:	When written, this file lets one set the backlight intensity for
+		a specific profile. Profile index is included in written data.
+		This attribute is only valid for the glow and pro variant.
+		The data has to be 16 bytes long.
+		Before reading this file, control has to be written to select
+		which profile to read.
+Users:		http://roccat.sourceforge.net
+
+What:		/sys/bus/usb/devices/<busnum>-<devnum>:<config num>.<interface num>/<hid-bus>:<vendor-id>:<product-id>.<num>/ryos/roccatryos<minor>/macro
+Date:		October 2013
+Contact:	Stefan Achatz <erazor_de@users.sourceforge.net>
+Description:	When written, this file lets one store macros with max 480
+		keystrokes for a specific button for a specific profile.
+		Button and profile indexes are included in written data.
+		The data has to be 2002 bytes long.
+		Before reading this file, control has to be written to select
+		which profile and key to read.
+Users:		http://roccat.sourceforge.net
+
+What:		/sys/bus/usb/devices/<busnum>-<devnum>:<config num>.<interface num>/<hid-bus>:<vendor-id>:<product-id>.<num>/ryos/roccatryos<minor>/info
+Date:		October 2013
+Contact:	Stefan Achatz <erazor_de@users.sourceforge.net>
+Description:	When read, this file returns general data like firmware version.
+		The data is 8 bytes long.
+		This file is readonly.
+Users:		http://roccat.sourceforge.net
+
+What:		/sys/bus/usb/devices/<busnum>-<devnum>:<config num>.<interface num>/<hid-bus>:<vendor-id>:<product-id>.<num>/ryos/roccatryos<minor>/reset
+Date:		October 2013
+Contact:	Stefan Achatz <erazor_de@users.sourceforge.net>
+Description:	When written, this file lets one reset the device.
+		The data has to be 3 bytes long.
+		This file is writeonly.
+Users:		http://roccat.sourceforge.net
+
+What:		/sys/bus/usb/devices/<busnum>-<devnum>:<config num>.<interface num>/<hid-bus>:<vendor-id>:<product-id>.<num>/ryos/roccatryos<minor>/talk
+Date:		October 2013
+Contact:	Stefan Achatz <erazor_de@users.sourceforge.net>
+Description:	When written, this file lets one trigger easyshift functionality
+		from the host.
+		The data has to be 16 bytes long.
+		This file is writeonly.
+Users:		http://roccat.sourceforge.net
+
+What:		/sys/bus/usb/devices/<busnum>-<devnum>:<config num>.<interface num>/<hid-bus>:<vendor-id>:<product-id>.<num>/ryos/roccatryos<minor>/light_control
+Date:		October 2013
+Contact:	Stefan Achatz <erazor_de@users.sourceforge.net>
+Description:	When written, this file lets one switch between stored and custom
+		light settings.
+		This attribute is only valid for the pro variant.
+		The data has to be 8 bytes long.
+		This file is writeonly.
+Users:		http://roccat.sourceforge.net
+
+What:		/sys/bus/usb/devices/<busnum>-<devnum>:<config num>.<interface num>/<hid-bus>:<vendor-id>:<product-id>.<num>/ryos/roccatryos<minor>/stored_lights
+Date:		October 2013
+Contact:	Stefan Achatz <erazor_de@users.sourceforge.net>
+Description:	When written, this file lets one set per-key lighting for different
+		layers.
+		This attribute is only valid for the pro variant.
+		The data has to be 1382 bytes long.
+		Before reading this file, control has to be written to select
+		which profile to read.
+Users:		http://roccat.sourceforge.net
+
+What:		/sys/bus/usb/devices/<busnum>-<devnum>:<config num>.<interface num>/<hid-bus>:<vendor-id>:<product-id>.<num>/ryos/roccatryos<minor>/custom_lights
+Date:		October 2013
+Contact:	Stefan Achatz <erazor_de@users.sourceforge.net>
+Description:	When written, this file lets one set the actual per-key lighting.
+		This attribute is only valid for the pro variant.
+		The data has to be 20 bytes long.
+		This file is writeonly.
+Users:		http://roccat.sourceforge.net
+
+What:		/sys/bus/usb/devices/<busnum>-<devnum>:<config num>.<interface num>/<hid-bus>:<vendor-id>:<product-id>.<num>/ryos/roccatryos<minor>/light_macro
+Date:		October 2013
+Contact:	Stefan Achatz <erazor_de@users.sourceforge.net>
+Description:	When written, this file lets one set a light macro that is looped
+		whenever the device gets in dimness mode.
+		This attribute is only valid for the pro variant.
+		The data has to be 2002 bytes long.
+		Before reading this file, control has to be written to select
+		which profile to read.
+Users:		http://roccat.sourceforge.net
diff --git a/Documentation/ABI/testing/sysfs-driver-hid-wiimote b/Documentation/ABI/testing/sysfs-driver-hid-wiimote
index ed5dd567d397..39dfa5cb1cc5 100644
--- a/Documentation/ABI/testing/sysfs-driver-hid-wiimote
+++ b/Documentation/ABI/testing/sysfs-driver-hid-wiimote
@@ -57,3 +57,21 @@ Description:	This attribute is only provided if the device was detected as a
 		Calibration data is already applied by the kernel to all input
 		values but may be used by user-space to perform other
 		transformations.
+
+What:		/sys/bus/hid/drivers/wiimote/<dev>/pro_calib
+Date:		October 2013
+KernelVersion:	3.13
+Contact:	David Herrmann <dh.herrmann@gmail.com>
+Description:	This attribute is only provided if the device was detected as a
+		pro-controller. It provides a single line with 4 calibration
+		values for all 4 analog sticks. Format is: "x1:y1 x2:y2". Data
+		is prefixed with a +/-. Each value is a signed 16bit number.
+		Data is encoded as decimal numbers and specifies the offsets of
+		the analog sticks of the pro-controller.
+		Calibration data is already applied by the kernel to all input
+		values but may be used by user-space to perform other
+		transformations.
+		Calibration data is detected by the kernel during device setup.
+		You can write "scan\n" into this file to re-trigger calibration.
+		You can also write data directly in the form "x1:y1 x2:y2" to
+		set the calibration values manually.
diff --git a/Documentation/DMA-attributes.txt b/Documentation/DMA-attributes.txt
index e59480db9ee0..cc2450d80310 100644
--- a/Documentation/DMA-attributes.txt
+++ b/Documentation/DMA-attributes.txt
@@ -13,7 +13,7 @@ all pending DMA writes to complete, and thus provides a mechanism to
 strictly order DMA from a device across all intervening busses and
 bridges.  This barrier is not specific to a particular type of
 interconnect, it applies to the system as a whole, and so its
-implementation must account for the idiosyncracies of the system all
+implementation must account for the idiosyncrasies of the system all
 the way from the DMA device to memory.
 
 As an example of a situation where DMA_ATTR_WRITE_BARRIER would be
@@ -60,7 +60,7 @@ such mapping is non-trivial task and consumes very limited resources
 Buffers allocated with this attribute can be only passed to user space
 by calling dma_mmap_attrs(). By using this API, you are guaranteeing
 that you won't dereference the pointer returned by dma_alloc_attr(). You
-can threat it as a cookie that must be passed to dma_mmap_attrs() and
+can treat it as a cookie that must be passed to dma_mmap_attrs() and
 dma_free_attrs(). Make sure that both of these also get this attribute
 set on each call.
 
@@ -82,7 +82,7 @@ to 'device' domain, what synchronizes CPU caches for the given region
 (usually it means that the cache has been flushed or invalidated
 depending on the dma direction). However, next calls to
 dma_map_{single,page,sg}() for other devices will perform exactly the
-same sychronization operation on the CPU cache. CPU cache sychronization
+same synchronization operation on the CPU cache. CPU cache synchronization
 might be a time consuming operation, especially if the buffers are
 large, so it is highly recommended to avoid it if possible.
 DMA_ATTR_SKIP_CPU_SYNC allows platform code to skip synchronization of
diff --git a/Documentation/DocBook/kernel-locking.tmpl b/Documentation/DocBook/kernel-locking.tmpl
index 09e884e5b9f5..19f2a5a5a5b4 100644
--- a/Documentation/DocBook/kernel-locking.tmpl
+++ b/Documentation/DocBook/kernel-locking.tmpl
@@ -1958,7 +1958,7 @@ machines due to caching.
   <chapter id="apiref-mutex">
    <title>Mutex API reference</title>
 !Iinclude/linux/mutex.h
-!Ekernel/mutex.c
+!Ekernel/locking/mutex.c
   </chapter>
 
   <chapter id="apiref-futex">
diff --git a/Documentation/devicetree/bindings/arc/pmu.txt b/Documentation/devicetree/bindings/arc/pmu.txt
new file mode 100644
index 000000000000..49d517340de3
--- /dev/null
+++ b/Documentation/devicetree/bindings/arc/pmu.txt
@@ -0,0 +1,24 @@
+* ARC Performance Monitor Unit
+
+The ARC 700 can be configured with a pipeline performance monitor for counting
+CPU and cache events like cache misses and hits.
+
+Note that:
+ * ARC 700 refers to a family of ARC processor cores;
+   - There is only one type of PMU available for the whole family;
+   - The PMU may support different sets of events; supported events are probed
+     at boot time, as required by the reference manual.
+
+ * The ARC 700 PMU does not support interrupts; although HW events may be
+   counted, the HW events themselves cannot serve as a trigger for a sample.
+
+Required properties:
+
+- compatible : should contain
+	"snps,arc700-pmu"
+
+Example:
+
+pmu {
+        compatible = "snps,arc700-pmu";
+};
diff --git a/Documentation/devicetree/bindings/arm/calxeda/mem-ctrlr.txt b/Documentation/devicetree/bindings/arm/calxeda/mem-ctrlr.txt
index f770ac0893d4..049675944b78 100644
--- a/Documentation/devicetree/bindings/arm/calxeda/mem-ctrlr.txt
+++ b/Documentation/devicetree/bindings/arm/calxeda/mem-ctrlr.txt
@@ -1,7 +1,9 @@
 Calxeda DDR memory controller
 
 Properties:
-- compatible : Should be "calxeda,hb-ddr-ctrl"
+- compatible : Should be:
+  - "calxeda,hb-ddr-ctrl" for ECX-1000
+  - "calxeda,ecx-2000-ddr-ctrl" for ECX-2000
 - reg : Address and size for DDR controller registers.
 - interrupts : Interrupt for DDR controller.
 
diff --git a/Documentation/devicetree/bindings/crypto/omap-des.txt b/Documentation/devicetree/bindings/crypto/omap-des.txt
new file mode 100644
index 000000000000..e8c63bf2e16d
--- /dev/null
+++ b/Documentation/devicetree/bindings/crypto/omap-des.txt
@@ -0,0 +1,30 @@
+OMAP SoC DES crypto Module
+
+Required properties:
+
+- compatible : Should contain "ti,omap4-des"
+- ti,hwmods: Name of the hwmod associated with the DES module
+- reg : Offset and length of the register set for the module
+- interrupts : the interrupt-specifier for the DES module
+- clocks : A phandle to the functional clock node of the DES module
+           corresponding to each entry in clock-names
+- clock-names : Name of the functional clock, should be "fck"
+
+Optional properties:
+- dmas: DMA specifiers for tx and rx dma. See the DMA client binding,
+	Documentation/devicetree/bindings/dma/dma.txt
+	Each entry corresponds to an entry in dma-names
+- dma-names: DMA request names should include "tx" and "rx" if present
+
+Example:
+	/* DRA7xx SoC */
+	des: des@480a5000 {
+		compatible = "ti,omap4-des";
+		ti,hwmods = "des";
+		reg = <0x480a5000 0xa0>;
+		interrupts = <GIC_SPI 82 IRQ_TYPE_LEVEL_HIGH>;
+		dmas = <&sdma 117>, <&sdma 116>;
+		dma-names = "tx", "rx";
+		clocks = <&l3_iclk_div>;
+		clock-names = "fck";
+	};
diff --git a/Documentation/devicetree/bindings/crypto/omap-sham.txt b/Documentation/devicetree/bindings/crypto/omap-sham.txt
index f839acd6f0ee..ad9115569611 100644
--- a/Documentation/devicetree/bindings/crypto/omap-sham.txt
+++ b/Documentation/devicetree/bindings/crypto/omap-sham.txt
@@ -6,7 +6,7 @@ Required properties:
   SHAM versions:
   - "ti,omap2-sham" for OMAP2 & OMAP3.
   - "ti,omap4-sham" for OMAP4 and AM33XX.
-  Note that these two versions are incompatible.
+  - "ti,omap5-sham" for OMAP5, DRA7 and AM43XX.
 - ti,hwmods: Name of the hwmod associated with the SHAM module
 - reg : Offset and length of the register set for the module
 - interrupts : the interrupt-specifier for the SHAM module.
diff --git a/Documentation/devicetree/bindings/dma/atmel-dma.txt b/Documentation/devicetree/bindings/dma/atmel-dma.txt
index e1f343c7a34b..f69bcf5a6343 100644
--- a/Documentation/devicetree/bindings/dma/atmel-dma.txt
+++ b/Documentation/devicetree/bindings/dma/atmel-dma.txt
@@ -28,7 +28,7 @@ The three cells in order are:
 dependent:
   - bit 7-0: peripheral identifier for the hardware handshaking interface. The
   identifier can be different for tx and rx.
-  - bit 11-8: FIFO configuration. 0 for half FIFO, 1 for ALAP, 1 for ASAP.
+  - bit 11-8: FIFO configuration. 0 for half FIFO, 1 for ALAP, 2 for ASAP.
 
 Example:
 
diff --git a/Documentation/devicetree/bindings/hwmon/lm90.txt b/Documentation/devicetree/bindings/hwmon/lm90.txt
new file mode 100644
index 000000000000..e8632486b9ef
--- /dev/null
+++ b/Documentation/devicetree/bindings/hwmon/lm90.txt
@@ -0,0 +1,44 @@
+* LM90 series thermometer.
+
+Required node properties:
+- compatible: manufacturer and chip name, one of
+		"adi,adm1032"
+		"adi,adt7461"
+		"adi,adt7461a"
+		"gmt,g781"
+		"national,lm90"
+		"national,lm86"
+		"national,lm89"
+		"national,lm99"
+		"dallas,max6646"
+		"dallas,max6647"
+		"dallas,max6649"
+		"dallas,max6657"
+		"dallas,max6658"
+		"dallas,max6659"
+		"dallas,max6680"
+		"dallas,max6681"
+		"dallas,max6695"
+		"dallas,max6696"
+		"onnn,nct1008"
+		"winbond,w83l771"
+		"nxp,sa56004"
+
+- reg: I2C bus address of the device
+
+- vcc-supply: vcc regulator for the supply voltage.
+
+Optional properties:
+- interrupts: Contains a single interrupt specifier which describes the
+              LM90 "-ALERT" pin output.
+              See interrupt-controller/interrupts.txt for the format.
+
+Example LM90 node:
+
+temp-sensor {
+	compatible = "onnn,nct1008";
+	reg = <0x4c>;
+	vcc-supply = <&palmas_ldo6_reg>;
+	interrupt-parent = <&gpio>;
+	interrupts = <TEGRA_GPIO(O, 4) IRQ_TYPE_LEVEL_LOW>;
+}
diff --git a/Documentation/devicetree/bindings/i2c/i2c-bcm-kona.txt b/Documentation/devicetree/bindings/i2c/i2c-bcm-kona.txt
new file mode 100644
index 000000000000..1b87b741fa8e
--- /dev/null
+++ b/Documentation/devicetree/bindings/i2c/i2c-bcm-kona.txt
@@ -0,0 +1,35 @@
+Broadcom Kona Family I2C
+=========================
+
+This I2C controller is used in the following Broadcom SoCs:
+
+  BCM11130
+  BCM11140
+  BCM11351
+  BCM28145
+  BCM28155
+
+Required Properties
+-------------------
+- compatible: "brcm,bcm11351-i2c", "brcm,kona-i2c"
+- reg: Physical base address and length of controller registers
+- interrupts: The interrupt number used by the controller
+- clocks: clock specifier for the kona i2c external clock
+- clock-frequency: The I2C bus frequency in Hz
+- #address-cells: Should be <1>
+- #size-cells: Should be <0>
+
+Refer to clocks/clock-bindings.txt for generic clock consumer
+properties.
+
+Example:
+
+i2c@3e016000 {
+	compatible = "brcm,bcm11351-i2c","brcm,kona-i2c";
+	reg = <0x3e016000 0x80>;
+	interrupts = <GIC_SPI 103 IRQ_TYPE_LEVEL_HIGH>;
+	clocks = <&bsc1_clk>;
+	clock-frequency = <400000>;
+	#address-cells = <1>;
+	#size-cells = <0>;
+};
diff --git a/Documentation/devicetree/bindings/i2c/i2c-exynos5.txt b/Documentation/devicetree/bindings/i2c/i2c-exynos5.txt
new file mode 100644
index 000000000000..056732cfdcee
--- /dev/null
+++ b/Documentation/devicetree/bindings/i2c/i2c-exynos5.txt
@@ -0,0 +1,44 @@
+* Samsung's High Speed I2C controller
+
+The Samsung's High Speed I2C controller is used to interface with I2C devices
+at various speeds ranging from 100khz to 3.4Mhz.
+
+Required properties:
+  - compatible: value should be.
+      -> "samsung,exynos5-hsi2c", for i2c compatible with exynos5 hsi2c.
+  - reg: physical base address of the controller and length of memory mapped
+    region.
+  - interrupts: interrupt number to the cpu.
+  - #address-cells: always 1 (for i2c addresses)
+  - #size-cells: always 0
+
+  - Pinctrl:
+    - pinctrl-0: Pin control group to be used for this controller.
+    - pinctrl-names: Should contain only one value - "default".
+
+Optional properties:
+  - clock-frequency: Desired operating frequency in Hz of the bus.
+    -> If not specified, the bus operates in fast-speed mode at
+       at 100khz.
+    -> If specified, the bus operates in high-speed mode only if the
+       clock-frequency is >= 1Mhz.
+
+Example:
+
+hsi2c@12ca0000 {
+	compatible = "samsung,exynos5-hsi2c";
+	reg = <0x12ca0000 0x100>;
+	interrupts = <56>;
+	clock-frequency = <100000>;
+
+	pinctrl-0 = <&i2c4_bus>;
+	pinctrl-names = "default";
+
+	#address-cells = <1>;
+	#size-cells = <0>;
+
+	s2mps11_pmic@66 {
+		compatible = "samsung,s2mps11-pmic";
+		reg = <0x66>;
+	};
+};
diff --git a/Documentation/devicetree/bindings/i2c/i2c-rcar.txt b/Documentation/devicetree/bindings/i2c/i2c-rcar.txt
new file mode 100644
index 000000000000..897cfcd5ce92
--- /dev/null
+++ b/Documentation/devicetree/bindings/i2c/i2c-rcar.txt
@@ -0,0 +1,23 @@
+I2C for R-Car platforms
+
+Required properties:
+- compatible: Must be one of
+	"renesas,i2c-rcar"
+	"renesas,i2c-r8a7778"
+	"renesas,i2c-r8a7779"
+	"renesas,i2c-r8a7790"
+- reg: physical base address of the controller and length of memory mapped
+  region.
+- interrupts: interrupt specifier.
+
+Optional properties:
+- clock-frequency: desired I2C bus clock frequency in Hz. The absence of this
+  propoerty indicates the default frequency 100 kHz.
+
+Examples :
+
+i2c0: i2c@e6500000 {
+	compatible = "renesas,i2c-rcar-h2";
+	reg = <0 0xe6500000 0 0x428>;
+	interrupts = <0 174 0x4>;
+};
diff --git a/Documentation/devicetree/bindings/i2c/i2c-st.txt b/Documentation/devicetree/bindings/i2c/i2c-st.txt
new file mode 100644
index 000000000000..437e0db3823c
--- /dev/null
+++ b/Documentation/devicetree/bindings/i2c/i2c-st.txt
@@ -0,0 +1,41 @@
+ST SSC binding, for I2C mode operation
+
+Required properties :
+- compatible : Must be "st,comms-ssc-i2c" or "st,comms-ssc4-i2c"
+- reg : Offset and length of the register set for the device
+- interrupts : the interrupt specifier
+- clock-names: Must contain "ssc".
+- clocks: Must contain an entry for each name in clock-names. See the common
+  clock bindings.
+- A pinctrl state named "default" must be defined to set pins in mode of
+  operation for I2C transfer.
+
+Optional properties :
+- clock-frequency : Desired I2C bus clock frequency in Hz. If not specified,
+  the default 100 kHz frequency will be used. As only Normal and Fast modes
+  are supported, possible values are 100000 and 400000.
+- st,i2c-min-scl-pulse-width-us : The minimum valid SCL pulse width that is
+  allowed through the deglitch circuit. In units of us.
+- st,i2c-min-sda-pulse-width-us : The minimum valid SDA pulse width that is
+  allowed through the deglitch circuit. In units of us.
+- A pinctrl state named "idle" could be defined to set pins in idle state
+  when I2C instance is not performing a transfer.
+- A pinctrl state named "sleep" could be defined to set pins in sleep state
+  when driver enters in suspend.
+
+
+
+Example :
+
+i2c0: i2c@fed40000 {
+	compatible	= "st,comms-ssc4-i2c";
+	reg		= <0xfed40000 0x110>;
+	interrupts	=  <GIC_SPI 187 IRQ_TYPE_LEVEL_HIGH>;
+	clocks		= <&CLK_S_ICN_REG_0>;
+	clock-names	= "ssc";
+	clock-frequency = <400000>;
+	pinctrl-names	= "default";
+	pinctrl-0	= <&pinctrl_i2c0_default>;
+	st,i2c-min-scl-pulse-width-us = <0>;
+	st,i2c-min-sda-pulse-width-us = <5>;
+};
diff --git a/Documentation/devicetree/bindings/input/touchscreen/ti-tsc-adc.txt b/Documentation/devicetree/bindings/input/touchscreen/ti-tsc-adc.txt
index 491c97b78384..878549ba814d 100644
--- a/Documentation/devicetree/bindings/input/touchscreen/ti-tsc-adc.txt
+++ b/Documentation/devicetree/bindings/input/touchscreen/ti-tsc-adc.txt
@@ -6,7 +6,7 @@ Required properties:
 	ti,wires: Wires refer to application modes i.e. 4/5/8 wire touchscreen
 		  support on the platform.
 	ti,x-plate-resistance: X plate resistance
-	ti,coordiante-readouts: The sequencer supports a total of 16
+	ti,coordinate-readouts: The sequencer supports a total of 16
 				programmable steps each step is used to
 				read a single coordinate. A single
                                 readout is enough but multiple reads can
diff --git a/Documentation/devicetree/bindings/media/st-rc.txt b/Documentation/devicetree/bindings/media/st-rc.txt
new file mode 100644
index 000000000000..05c432d08bca
--- /dev/null
+++ b/Documentation/devicetree/bindings/media/st-rc.txt
@@ -0,0 +1,29 @@
+Device-Tree bindings for ST IRB IP
+
+Required properties:
+	- compatible: Should contain "st,comms-irb".
+	- reg: Base physical address of the controller and length of memory
+	  mapped region.
+	- interrupts: interrupt-specifier for the sole interrupt generated by
+	  the device. The interrupt specifier format depends on the interrupt
+	  controller parent.
+	- rx-mode: can be "infrared" or "uhf". This property specifies the L1
+	  protocol used for receiving remote control signals. rx-mode should
+	  be present iff the rx pins are wired up.
+	- tx-mode: should be "infrared". This property specifies the L1
+	  protocol used for transmitting remote control signals. tx-mode should
+	  be present iff the tx pins are wired up.
+
+Optional properties:
+	- pinctrl-names, pinctrl-0: the pincontrol settings to configure muxing
+	  properly for IRB pins.
+	- clocks : phandle with clock-specifier pair for IRB.
+
+Example node:
+
+	rc: rc@fe518000 {
+		compatible	= "st,comms-irb";
+		reg		= <0xfe518000 0x234>;
+		interrupts	= <0 203 0>;
+		rx-mode		= "infrared";
+	};
diff --git a/Documentation/devicetree/bindings/mfd/as3722.txt b/Documentation/devicetree/bindings/mfd/as3722.txt
new file mode 100644
index 000000000000..fc2191ecfd6b
--- /dev/null
+++ b/Documentation/devicetree/bindings/mfd/as3722.txt
@@ -0,0 +1,194 @@
+* ams AS3722 Power management IC.
+
+Required properties:
+-------------------
+- compatible: Must be "ams,as3722".
+- reg: I2C device address.
+- interrupt-controller: AS3722 has internal interrupt controller which takes the
+  interrupt request from internal sub-blocks like RTC, regulators, GPIOs as well
+  as external input.
+- #interrupt-cells: Should be set to 2 for IRQ number and flags.
+  The first cell is the IRQ number. IRQ numbers for different interrupt source
+  of AS3722 are defined at dt-bindings/mfd/as3722.h
+  The second cell is the flags, encoded as the trigger masks from binding document
+	interrupts.txt, using dt-bindings/irq.
+
+Optional submodule and their properties:
+=======================================
+
+Pinmux and GPIO:
+===============
+Device has 8 GPIO pins which can be configured as GPIO as well as the special IO
+functions.
+
+Please refer to pinctrl-bindings.txt in this directory for details of the
+common pinctrl bindings used by client devices, including the meaning of the
+phrase "pin configuration node".
+
+Following are properties which is needed if GPIO and pinmux functionality
+is required:
+    Required properties:
+    -------------------
+	- gpio-controller: Marks the device node as a GPIO controller.
+	- #gpio-cells: Number of GPIO cells. Refer to binding document
+			gpio/gpio.txt
+
+    Optional properties:
+    --------------------
+	Following properties are require if pin control setting is required
+	at boot.
+	- pinctrl-names: A pinctrl state named "default" be defined, using the
+		bindings in pinctrl/pinctrl-binding.txt.
+	- pinctrl[0...n]: Properties to contain the phandle that refer to
+		different nodes of pin control settings. These nodes represents
+		the pin control setting of state 0 to state n. Each of these
+		nodes contains different subnodes to represents some desired
+		configuration for a list of pins. This configuration can
+		include the mux function to select on those pin(s), and
+		various pin configuration parameters, such as pull-up,
+		open drain.
+
+		Each subnode have following properties:
+		Required properties:
+		    - pins: List of pins. Valid values of pins properties are:
+				gpio0, gpio1, gpio2, gpio3, gpio4, gpio5,
+				gpio6, gpio7
+
+		Optional properties:
+			function, bias-disable, bias-pull-up, bias-pull-down,
+			bias-high-impedance, drive-open-drain.
+
+			Valid values for function properties are:
+				gpio, interrupt-out, gpio-in-interrupt,
+				vsup-vbat-low-undebounce-out,
+				vsup-vbat-low-debounce-out,
+				voltage-in-standby, oc-pg-sd0, oc-pg-sd6,
+				powergood-out, pwm-in, pwm-out, clk32k-out,
+				watchdog-in, soft-reset-in
+
+Regulators:
+===========
+Device has multiple DCDC and LDOs. The node "regulators" is require if regulator
+functionality is needed.
+
+Following are properties of regulator subnode.
+
+    Optional properties:
+    -------------------
+	The input supply of regulators are the optional properties on the
+	regulator node. The input supply of these regulators are provided
+	through following properties:
+		vsup-sd2-supply: Input supply for SD2.
+		vsup-sd3-supply: Input supply for SD3.
+		vsup-sd4-supply: Input supply for SD4.
+		vsup-sd5-supply: Input supply for SD5.
+		vin-ldo0-supply: Input supply for LDO0.
+		vin-ldo1-6-supply: Input supply for LDO1 and LDO6.
+		vin-ldo2-5-7-supply: Input supply for LDO2, LDO5 and LDO7.
+		vin-ldo3-4-supply: Input supply for LDO3 and LDO4.
+		vin-ldo9-10-supply: Input supply for LDO9 and LDO10.
+		vin-ldo11-supply: Input supply for LDO11.
+
+    Optional sub nodes for regulators:
+    ---------------------------------
+	The subnodes name is the name of regulator and it must be one of:
+	sd[0-6], ldo[0-7], ldo[9-11]
+
+	Each sub-node should contain the constraints and initialization
+	information for that regulator. See regulator.txt for a description
+	of standard properties for these sub-nodes.
+	Additional optional custom properties  are listed below.
+		ams,ext-control: External control of the rail. The option of
+			this properties will tell which external input is
+			controlling this rail. Valid values are 0, 1, 2 ad 3.
+			0: There is no external control of this rail.
+			1: Rail is controlled by ENABLE1 input pin.
+			2: Rail is controlled by ENABLE2 input pin.
+			3: Rail is controlled by ENABLE3 input pin.
+			Missing this property on DT will be assume as no
+			external control. The external control pin macros
+			are defined @dt-bindings/mfd/as3722.h
+
+		ams,enable-tracking: Enable tracking with SD1, only supported
+			by LDO3.
+
+Example:
+--------
+#include <dt-bindings/mfd/as3722.h>
+...
+ams3722 {
+	compatible = "ams,as3722";
+	reg = <0x48>;
+
+	interrupt-parent = <&intc>;
+	interrupt-controller;
+	#interrupt-cells = <2>;
+
+	gpio-controller;
+	#gpio-cells = <2>;
+
+	pinctrl-names = "default";
+	pinctrl-0 = <&as3722_default>;
+
+	as3722_default: pinmux {
+			gpio0 {
+				pins = "gpio0";
+				function = "gpio";
+				bias-pull-down;
+			};
+
+			gpio1_2_4_7 {
+				pins = "gpio1", "gpio2", "gpio4", "gpio7";
+				function = "gpio";
+				bias-pull-up;
+			};
+
+			gpio5 {
+				pins = "gpio5";
+				function = "clk32k_out";
+			};
+	}
+
+	regulators {
+			vsup-sd2-supply = <...>;
+			...
+
+			sd0 {
+				regulator-name = "vdd_cpu";
+				regulator-min-microvolt = <700000>;
+				regulator-max-microvolt = <1400000>;
+				regulator-always-on;
+				ams,ext-control = <2>;
+			};
+
+			sd1 {
+				regulator-name = "vdd_core";
+				regulator-min-microvolt = <700000>;
+				regulator-max-microvolt = <1400000>;
+				regulator-always-on;
+				ams,ext-control = <1>;
+			};
+
+			sd2 {
+				regulator-name = "vddio_ddr";
+				regulator-min-microvolt = <1350000>;
+				regulator-max-microvolt = <1350000>;
+				regulator-always-on;
+			};
+
+			sd4 {
+				regulator-name = "avdd-hdmi-pex";
+				regulator-min-microvolt = <1050000>;
+				regulator-max-microvolt = <1050000>;
+				regulator-always-on;
+			};
+
+			sd5 {
+				regulator-name = "vdd-1v8";
+				regulator-min-microvolt = <1800000>;
+				regulator-max-microvolt = <1800000>;
+				regulator-always-on;
+			};
+			....
+	};
+};
diff --git a/Documentation/devicetree/bindings/mfd/s2mps11.txt b/Documentation/devicetree/bindings/mfd/s2mps11.txt
index c9332c626021..78a840d7510d 100644
--- a/Documentation/devicetree/bindings/mfd/s2mps11.txt
+++ b/Documentation/devicetree/bindings/mfd/s2mps11.txt
@@ -1,10 +1,10 @@
 
 * Samsung S2MPS11 Voltage and Current Regulator
 
-The Samsung S2MP211 is a multi-function device which includes voltage and
+The Samsung S2MPS11 is a multi-function device which includes voltage and
 current regulators, RTC, charger controller and other sub-blocks. It is
-interfaced to the host controller using a I2C interface. Each sub-block is
-addressed by the host system using different I2C slave address.
+interfaced to the host controller using an I2C interface. Each sub-block is
+addressed by the host system using different I2C slave addresses.
 
 Required properties:
 - compatible: Should be "samsung,s2mps11-pmic".
@@ -43,7 +43,8 @@ sub-node should be of the format as listed below.
 
  BUCK[2/3/4/6] supports disabling ramp delay on hardware, so explictly
  regulator-ramp-delay = <0> can be used for them to disable ramp delay.
- In absence of regulator-ramp-delay property, default ramp delay will be used.
+ In the absence of the regulator-ramp-delay property, the default ramp
+ delay will be used.
 
 NOTE: Some BUCKs share the ramp rate setting i.e. same ramp value will be set
 for a particular group of BUCKs. So provide same regulator-ramp-delay<value>.
@@ -58,10 +59,10 @@ supports. Note: The 'n' in LDOn and BUCKn represents the LDO or BUCK number
 as per the datasheet of s2mps11.
 
 	- LDOn
-		  - valid values for n are 1 to 28
+		  - valid values for n are 1 to 38
 		  - Example: LDO0, LD01, LDO28
 	- BUCKn
-		  - valid values for n are 1 to 9.
+		  - valid values for n are 1 to 10.
 		  - Example: BUCK1, BUCK2, BUCK9
 
 Example:
diff --git a/Documentation/devicetree/bindings/mmc/fsl-imx-esdhc.txt b/Documentation/devicetree/bindings/mmc/fsl-imx-esdhc.txt
index 1dd622546d06..9046ba06c47a 100644
--- a/Documentation/devicetree/bindings/mmc/fsl-imx-esdhc.txt
+++ b/Documentation/devicetree/bindings/mmc/fsl-imx-esdhc.txt
@@ -12,6 +12,11 @@ Required properties:
 Optional properties:
 - fsl,cd-controller : Indicate to use controller internal card detection
 - fsl,wp-controller : Indicate to use controller internal write protection
+- fsl,delay-line : Specify the number of delay cells for override mode.
+  This is used to set the clock delay for DLL(Delay Line) on override mode
+  to select a proper data sampling window in case the clock quality is not good
+  due to signal path is too long on the board. Please refer to eSDHC/uSDHC
+  chapter, DLL (Delay Line) section in RM for details.
 
 Examples:
 
diff --git a/Documentation/devicetree/bindings/mmc/synopsys-dw-mshc.txt b/Documentation/devicetree/bindings/mmc/synopsys-dw-mshc.txt
index 066a78b034ca..8f3f13315358 100644
--- a/Documentation/devicetree/bindings/mmc/synopsys-dw-mshc.txt
+++ b/Documentation/devicetree/bindings/mmc/synopsys-dw-mshc.txt
@@ -52,6 +52,9 @@ Optional properties:
   is specified and the ciu clock is specified then we'll try to set the ciu
   clock to this at probe time.
 
+* clock-freq-min-max: Minimum and Maximum clock frequency for card output
+  clock(cclk_out). If it's not specified, max is 200MHZ and min is 400KHz by default.
+
 * num-slots: specifies the number of slots supported by the controller.
   The number of physical slots actually used could be equal or less than the
   value specified by num-slots. If this property is not specified, the value
@@ -66,6 +69,10 @@ Optional properties:
 
 * supports-highspeed: Enables support for high speed cards (up to 50MHz)
 
+* caps2-mmc-hs200-1_8v: Supports mmc HS200 SDR 1.8V mode
+
+* caps2-mmc-hs200-1_2v: Supports mmc HS200 SDR 1.2V mode
+
 * broken-cd: as documented in mmc core bindings.
 
 * vmmc-supply: The phandle to the regulator to use for vmmc.  If this is
@@ -93,8 +100,10 @@ board specific portions as listed below.
 
 	dwmmc0@12200000 {
 		clock-frequency = <400000000>;
+		clock-freq-min-max = <400000 200000000>;
 		num-slots = <1>;
 		supports-highspeed;
+		caps2-mmc-hs200-1_8v;
 		broken-cd;
 		fifo-depth = <0x80>;
 		card-detect-delay = <200>;
diff --git a/Documentation/devicetree/bindings/power/twl-charger.txt b/Documentation/devicetree/bindings/power/twl-charger.txt
new file mode 100644
index 000000000000..d5c706216df5
--- /dev/null
+++ b/Documentation/devicetree/bindings/power/twl-charger.txt
@@ -0,0 +1,20 @@
+TWL BCI (Battery Charger Interface)
+
+Required properties:
+- compatible:
+  - "ti,twl4030-bci"
+- interrupts: two interrupt lines from the TWL SIH (secondary
+  interrupt handler) - interrupts 9 and 2.
+
+Optional properties:
+- ti,bb-uvolt: microvolts for charging the backup battery.
+- ti,bb-uamp: microamps for charging the backup battery.
+
+Examples:
+
+bci {
+   compatible = "ti,twl4030-bci";
+   interrupts = <9>, <2>;
+   ti,bb-uvolt = <3200000>;
+   ti,bb-uamp = <150>;
+};
diff --git a/Documentation/devicetree/bindings/power_supply/ti,bq24735.txt b/Documentation/devicetree/bindings/power_supply/ti,bq24735.txt
new file mode 100644
index 000000000000..4f6a550184d0
--- /dev/null
+++ b/Documentation/devicetree/bindings/power_supply/ti,bq24735.txt
@@ -0,0 +1,32 @@
+TI BQ24735 Charge Controller
+~~~~~~~~~~
+
+Required properties :
+ - compatible : "ti,bq24735"
+
+Optional properties :
+ - interrupts : Specify the interrupt to be used to trigger when the AC
+   adapter is either plugged in or removed.
+ - ti,ac-detect-gpios : This GPIO is optionally used to read the AC adapter
+   presence. This is a Host GPIO that is configured as an input and
+   connected to the bq24735.
+ - ti,charge-current : Used to control and set the charging current. This value
+   must be between 128mA and 8.128A with a 64mA step resolution. The POR value
+   is 0x0000h. This number is in mA (e.g. 8192), see spec for more information
+   about the ChargeCurrent (0x14h) register.
+ - ti,charge-voltage : Used to control and set the charging voltage. This value
+   must be between 1.024V and 19.2V with a 16mV step resolution. The POR value
+   is 0x0000h. This number is in mV (e.g. 19200), see spec for more information
+   about the ChargeVoltage (0x15h) register.
+ - ti,input-current : Used to control and set the charger input current. This
+   value must be between 128mA and 8.064A with a 128mA step resolution. The
+   POR value is 0x1000h. This number is in mA (e.g. 8064), see the spec for
+   more information about the InputCurrent (0x3fh) register.
+
+Example:
+
+	bq24735@9 {
+		compatible = "ti,bq24735";
+		reg = <0x9>;
+		ti,ac-detect-gpios = <&gpio 72 0x1>;
+	}
diff --git a/Documentation/devicetree/bindings/powerpc/fsl/dma.txt b/Documentation/devicetree/bindings/powerpc/fsl/dma.txt
index 2a4b4bce6110..7fc1b010fa75 100644
--- a/Documentation/devicetree/bindings/powerpc/fsl/dma.txt
+++ b/Documentation/devicetree/bindings/powerpc/fsl/dma.txt
@@ -1,33 +1,30 @@
-* Freescale 83xx DMA Controller
+* Freescale DMA Controllers
 
-Freescale PowerPC 83xx have on chip general purpose DMA controllers.
+** Freescale Elo DMA Controller
+   This is a little-endian 4-channel DMA controller, used in Freescale mpc83xx
+   series chips such as mpc8315, mpc8349, mpc8379 etc.
 
 Required properties:
 
-- compatible        : compatible list, contains 2 entries, first is
-		 "fsl,CHIP-dma", where CHIP is the processor
-		 (mpc8349, mpc8360, etc.) and the second is
-		 "fsl,elo-dma"
-- reg               : <registers mapping for DMA general status reg>
-- ranges		: Should be defined as specified in 1) to describe the
-		  DMA controller channels.
+- compatible        : must include "fsl,elo-dma"
+- reg               : DMA General Status Register, i.e. DGSR which contains
+                      status for all the 4 DMA channels
+- ranges            : describes the mapping between the address space of the
+                      DMA channels and the address space of the DMA controller
 - cell-index        : controller index.  0 for controller @ 0x8100
-- interrupts        : <interrupt mapping for DMA IRQ>
+- interrupts        : interrupt specifier for DMA IRQ
 - interrupt-parent  : optional, if needed for interrupt mapping
 
-
 - DMA channel nodes:
-        - compatible        : compatible list, contains 2 entries, first is
-			 "fsl,CHIP-dma-channel", where CHIP is the processor
-			 (mpc8349, mpc8350, etc.) and the second is
-			 "fsl,elo-dma-channel". However, see note below.
-        - reg               : <registers mapping for channel>
-        - cell-index        : dma channel index starts at 0.
+        - compatible        : must include "fsl,elo-dma-channel"
+                              However, see note below.
+        - reg               : DMA channel specific registers
+        - cell-index        : DMA channel index starts at 0.
 
 Optional properties:
-        - interrupts        : <interrupt mapping for DMA channel IRQ>
-			  (on 83xx this is expected to be identical to
-			   the interrupts property of the parent node)
+        - interrupts        : interrupt specifier for DMA channel IRQ
+                              (on 83xx this is expected to be identical to
+                              the interrupts property of the parent node)
         - interrupt-parent  : optional, if needed for interrupt mapping
 
 Example:
@@ -70,30 +67,27 @@ Example:
 		};
 	};
 
-* Freescale 85xx/86xx DMA Controller
-
-Freescale PowerPC 85xx/86xx have on chip general purpose DMA controllers.
+** Freescale EloPlus DMA Controller
+   This is a 4-channel DMA controller with extended addresses and chaining,
+   mainly used in Freescale mpc85xx/86xx, Pxxx and BSC series chips, such as
+   mpc8540, mpc8641 p4080, bsc9131 etc.
 
 Required properties:
 
-- compatible        : compatible list, contains 2 entries, first is
-		 "fsl,CHIP-dma", where CHIP is the processor
-		 (mpc8540, mpc8540, etc.) and the second is
-		 "fsl,eloplus-dma"
-- reg               : <registers mapping for DMA general status reg>
+- compatible        : must include "fsl,eloplus-dma"
+- reg               : DMA General Status Register, i.e. DGSR which contains
+                      status for all the 4 DMA channels
 - cell-index        : controller index.  0 for controller @ 0x21000,
                                          1 for controller @ 0xc000
-- ranges		: Should be defined as specified in 1) to describe the
-		  DMA controller channels.
+- ranges            : describes the mapping between the address space of the
+                      DMA channels and the address space of the DMA controller
 
 - DMA channel nodes:
-        - compatible        : compatible list, contains 2 entries, first is
-			 "fsl,CHIP-dma-channel", where CHIP is the processor
-			 (mpc8540, mpc8560, etc.) and the second is
-			 "fsl,eloplus-dma-channel". However, see note below.
-        - cell-index        : dma channel index starts at 0.
-        - reg               : <registers mapping for channel>
-        - interrupts        : <interrupt mapping for DMA channel IRQ>
+        - compatible        : must include "fsl,eloplus-dma-channel"
+                              However, see note below.
+        - cell-index        : DMA channel index starts at 0.
+        - reg               : DMA channel specific registers
+        - interrupts        : interrupt specifier for DMA channel IRQ
         - interrupt-parent  : optional, if needed for interrupt mapping
 
 Example:
@@ -134,6 +128,76 @@ Example:
 		};
 	};
 
+** Freescale Elo3 DMA Controller
+   DMA controller which has same function as EloPlus except that Elo3 has 8
+   channels while EloPlus has only 4, it is used in Freescale Txxx and Bxxx
+   series chips, such as t1040, t4240, b4860.
+
+Required properties:
+
+- compatible        : must include "fsl,elo3-dma"
+- reg               : contains two entries for DMA General Status Registers,
+                      i.e. DGSR0 which includes status for channel 1~4, and
+                      DGSR1 for channel 5~8
+- ranges            : describes the mapping between the address space of the
+                      DMA channels and the address space of the DMA controller
+
+- DMA channel nodes:
+        - compatible        : must include "fsl,eloplus-dma-channel"
+        - reg               : DMA channel specific registers
+        - interrupts        : interrupt specifier for DMA channel IRQ
+        - interrupt-parent  : optional, if needed for interrupt mapping
+
+Example:
+dma@100300 {
+	#address-cells = <1>;
+	#size-cells = <1>;
+	compatible = "fsl,elo3-dma";
+	reg = <0x100300 0x4>,
+	      <0x100600 0x4>;
+	ranges = <0x0 0x100100 0x500>;
+	dma-channel@0 {
+		compatible = "fsl,eloplus-dma-channel";
+		reg = <0x0 0x80>;
+		interrupts = <28 2 0 0>;
+	};
+	dma-channel@80 {
+		compatible = "fsl,eloplus-dma-channel";
+		reg = <0x80 0x80>;
+		interrupts = <29 2 0 0>;
+	};
+	dma-channel@100 {
+		compatible = "fsl,eloplus-dma-channel";
+		reg = <0x100 0x80>;
+		interrupts = <30 2 0 0>;
+	};
+	dma-channel@180 {
+		compatible = "fsl,eloplus-dma-channel";
+		reg = <0x180 0x80>;
+		interrupts = <31 2 0 0>;
+	};
+	dma-channel@300 {
+		compatible = "fsl,eloplus-dma-channel";
+		reg = <0x300 0x80>;
+		interrupts = <76 2 0 0>;
+	};
+	dma-channel@380 {
+		compatible = "fsl,eloplus-dma-channel";
+		reg = <0x380 0x80>;
+		interrupts = <77 2 0 0>;
+	};
+	dma-channel@400 {
+		compatible = "fsl,eloplus-dma-channel";
+		reg = <0x400 0x80>;
+		interrupts = <78 2 0 0>;
+	};
+	dma-channel@480 {
+		compatible = "fsl,eloplus-dma-channel";
+		reg = <0x480 0x80>;
+		interrupts = <79 2 0 0>;
+	};
+};
+
 Note on DMA channel compatible properties: The compatible property must say
 "fsl,elo-dma-channel" or "fsl,eloplus-dma-channel" to be used by the Elo DMA
 driver (fsldma).  Any DMA channel used by fsldma cannot be used by another
diff --git a/Documentation/devicetree/bindings/pwm/pwm-samsung.txt b/Documentation/devicetree/bindings/pwm/pwm-samsung.txt
index d61fccd40bad..5538de9c2007 100644
--- a/Documentation/devicetree/bindings/pwm/pwm-samsung.txt
+++ b/Documentation/devicetree/bindings/pwm/pwm-samsung.txt
@@ -15,7 +15,7 @@ Required properties:
     samsung,s5pc100-pwm - for 32-bit timers present on S5PC100, S5PV210,
 			  Exynos4210 rev0 SoCs
     samsung,exynos4210-pwm - for 32-bit timers present on Exynos4210,
-                          Exynos4x12 and Exynos5250 SoCs
+                          Exynos4x12, Exynos5250 and Exynos5420 SoCs
 - reg: base address and size of register area
 - interrupts: list of timer interrupts (one interrupt per timer, starting at
   timer 0)
diff --git a/Documentation/devicetree/bindings/spi/omap-spi.txt b/Documentation/devicetree/bindings/spi/omap-spi.txt
index 4c85c4c69584..2ba5f9c023ac 100644
--- a/Documentation/devicetree/bindings/spi/omap-spi.txt
+++ b/Documentation/devicetree/bindings/spi/omap-spi.txt
@@ -2,8 +2,8 @@ OMAP2+ McSPI device
 
 Required properties:
 - compatible :
-  - "ti,omap2-spi" for OMAP2 & OMAP3.
-  - "ti,omap4-spi" for OMAP4+.
+  - "ti,omap2-mcspi" for OMAP2 & OMAP3.
+  - "ti,omap4-mcspi" for OMAP4+.
 - ti,spi-num-cs : Number of chipselect supported  by the instance.
 - ti,hwmods: Name of the hwmod associated to the McSPI
 - ti,pindir-d0-out-d1-in: Select the D0 pin as output and D1 as
diff --git a/Documentation/devicetree/bindings/video/atmel,lcdc.txt b/Documentation/devicetree/bindings/video/atmel,lcdc.txt
new file mode 100644
index 000000000000..1ec175eddca8
--- /dev/null
+++ b/Documentation/devicetree/bindings/video/atmel,lcdc.txt
@@ -0,0 +1,75 @@
+Atmel LCDC Framebuffer
+-----------------------------------------------------
+
+Required properties:
+- compatible :
+	"atmel,at91sam9261-lcdc" , 
+	"atmel,at91sam9263-lcdc" ,
+	"atmel,at91sam9g10-lcdc" ,
+	"atmel,at91sam9g45-lcdc" ,
+	"atmel,at91sam9g45es-lcdc" ,
+	"atmel,at91sam9rl-lcdc" ,
+	"atmel,at32ap-lcdc"
+- reg : Should contain 1 register ranges(address and length)
+- interrupts : framebuffer controller interrupt
+- display: a phandle pointing to the display node
+
+Required nodes:
+- display: a display node is required to initialize the lcd panel
+	This should be in the board dts.
+- default-mode: a videomode within the display with timing parameters
+	as specified below.
+
+Example:
+
+	fb0: fb@0x00500000 {
+		compatible = "atmel,at91sam9g45-lcdc";
+		reg = <0x00500000 0x1000>;
+		interrupts = <23 3 0>;
+		pinctrl-names = "default";
+		pinctrl-0 = <&pinctrl_fb>;
+		display = <&display0>;
+		status = "okay";
+		#address-cells = <1>;
+		#size-cells = <1>;
+
+	};
+
+Atmel LCDC Display
+-----------------------------------------------------
+Required properties (as per of_videomode_helper):
+
+ - atmel,dmacon: dma controler configuration
+ - atmel,lcdcon2: lcd controler configuration
+ - atmel,guard-time: lcd guard time (Delay in frame periods)
+ - bits-per-pixel: lcd panel bit-depth.
+
+Optional properties (as per of_videomode_helper):
+ - atmel,lcdcon-backlight: enable backlight
+ - atmel,lcd-wiring-mode: lcd wiring mode "RGB" or "BRG"
+ - atmel,power-control-gpio: gpio to power on or off the LCD (as many as needed)
+
+Example:
+	display0: display {
+		bits-per-pixel = <32>;
+		atmel,lcdcon-backlight;
+		atmel,dmacon = <0x1>;
+		atmel,lcdcon2 = <0x80008002>;
+		atmel,guard-time = <9>;
+		atmel,lcd-wiring-mode = <1>;
+
+		display-timings {
+			native-mode = <&timing0>;
+			timing0: timing0 {
+				clock-frequency = <9000000>;
+				hactive = <480>;
+				vactive = <272>;
+				hback-porch = <1>;
+				hfront-porch = <1>;
+				vback-porch = <40>;
+				vfront-porch = <1>;
+				hsync-len = <45>;
+				vsync-len = <1>;
+			};
+		};
+	};
diff --git a/Documentation/devicetree/bindings/video/backlight/pwm-backlight.txt b/Documentation/devicetree/bindings/video/backlight/pwm-backlight.txt
index 1e4fc727f3b1..764db86d441a 100644
--- a/Documentation/devicetree/bindings/video/backlight/pwm-backlight.txt
+++ b/Documentation/devicetree/bindings/video/backlight/pwm-backlight.txt
@@ -10,12 +10,16 @@ Required properties:
       last value in the array represents a 100% duty cycle (brightest).
   - default-brightness-level: the default brightness level (index into the
       array defined by the "brightness-levels" property)
+  - power-supply: regulator for supply voltage
 
 Optional properties:
   - pwm-names: a list of names for the PWM devices specified in the
                "pwms" property (see PWM binding[0])
+  - enable-gpios: contains a single GPIO specifier for the GPIO which enables
+                  and disables the backlight (see GPIO binding[1])
 
 [0]: Documentation/devicetree/bindings/pwm/pwm.txt
+[1]: Documentation/devicetree/bindings/gpio/gpio.txt
 
 Example:
 
@@ -25,4 +29,7 @@ Example:
 
 		brightness-levels = <0 4 8 16 32 64 128 255>;
 		default-brightness-level = <6>;
+
+		power-supply = <&vdd_bl_reg>;
+		enable-gpios = <&gpio 58 0>;
 	};
diff --git a/Documentation/devicetree/bindings/watchdog/dw_wdt.txt b/Documentation/devicetree/bindings/watchdog/dw_wdt.txt
new file mode 100644
index 000000000000..08e16f684f2d
--- /dev/null
+++ b/Documentation/devicetree/bindings/watchdog/dw_wdt.txt
@@ -0,0 +1,21 @@
+Synopsys Designware Watchdog Timer
+
+Required Properties:
+
+- compatible	: Should contain "snps,dw-wdt"
+- reg		: Base address and size of the watchdog timer registers.
+- clocks	: phandle + clock-specifier for the clock that drives the
+		watchdog timer.
+
+Optional Properties:
+
+- interrupts	: The interrupt used for the watchdog timeout warning.
+
+Example:
+
+	watchdog0: wd@ffd02000 {
+		compatible = "snps,dw-wdt";
+		reg = <0xffd02000 0x1000>;
+		interrupts = <0 171 4>;
+		clocks = <&per_base_clk>;
+	};
diff --git a/Documentation/devicetree/bindings/gpio/men-a021-wdt.txt b/Documentation/devicetree/bindings/watchdog/men-a021-wdt.txt
index 370dee3226d9..370dee3226d9 100644
--- a/Documentation/devicetree/bindings/gpio/men-a021-wdt.txt
+++ b/Documentation/devicetree/bindings/watchdog/men-a021-wdt.txt
diff --git a/Documentation/devicetree/bindings/watchdog/moxa,moxart-watchdog.txt b/Documentation/devicetree/bindings/watchdog/moxa,moxart-watchdog.txt
new file mode 100644
index 000000000000..1169857d1d12
--- /dev/null
+++ b/Documentation/devicetree/bindings/watchdog/moxa,moxart-watchdog.txt
@@ -0,0 +1,15 @@
+MOXA ART Watchdog timer
+
+Required properties:
+
+- compatible : Must be "moxa,moxart-watchdog"
+- reg : Should contain registers location and length
+- clocks : Should contain phandle for the clock that drives the counter
+
+Example:
+
+	watchdog: watchdog@98500000 {
+		compatible = "moxa,moxart-watchdog";
+		reg = <0x98500000 0x10>;
+		clocks = <&coreclk>;
+	};
diff --git a/Documentation/devicetree/bindings/watchdog/rt2880-wdt.txt b/Documentation/devicetree/bindings/watchdog/rt2880-wdt.txt
new file mode 100644
index 000000000000..d7bab3db9d1f
--- /dev/null
+++ b/Documentation/devicetree/bindings/watchdog/rt2880-wdt.txt
@@ -0,0 +1,19 @@
+Ralink Watchdog Timers
+
+Required properties:
+- compatible: must be "ralink,rt2880-wdt"
+- reg: physical base address of the controller and length of the register range
+
+Optional properties:
+- interrupt-parent: phandle to the INTC device node
+- interrupts: Specify the INTC interrupt number
+
+Example:
+
+	watchdog@120 {
+		compatible = "ralink,rt2880-wdt";
+		reg = <0x120 0x10>;
+
+		interrupt-parent = <&intc>;
+		interrupts = <1>;
+	};
diff --git a/Documentation/devicetree/bindings/watchdog/sirfsoc_wdt.txt b/Documentation/devicetree/bindings/watchdog/sirfsoc_wdt.txt
new file mode 100644
index 000000000000..9cbc76c89b2b
--- /dev/null
+++ b/Documentation/devicetree/bindings/watchdog/sirfsoc_wdt.txt
@@ -0,0 +1,14 @@
+SiRFSoC Timer and Watchdog Timer(WDT) Controller
+
+Required properties:
+- compatible: "sirf,prima2-tick"
+- reg: Address range of tick timer/WDT register set
+- interrupts: interrupt number to the cpu
+
+Example:
+
+timer@b0020000 {
+	compatible = "sirf,prima2-tick";
+	reg = <0xb0020000 0x1000>;
+	interrupts = <0>;
+};
diff --git a/Documentation/dmatest.txt b/Documentation/dmatest.txt
index a2b5663eae26..dd77a81bdb80 100644
--- a/Documentation/dmatest.txt
+++ b/Documentation/dmatest.txt
@@ -15,39 +15,48 @@ be built as module or inside kernel. Let's consider those cases.
 
 	Part 2 - When dmatest is built as a module...
 
-After mounting debugfs and loading the module, the /sys/kernel/debug/dmatest
-folder with nodes will be created. There are two important files located. First
-is the 'run' node that controls run and stop phases of the test, and the second
-one, 'results', is used to get the test case results.
-
-Note that in this case test will not run on load automatically.
-
 Example of usage:
+	% modprobe dmatest channel=dma0chan0 timeout=2000 iterations=1 run=1
+
+...or:
+	% modprobe dmatest
 	% echo dma0chan0 > /sys/module/dmatest/parameters/channel
 	% echo 2000 > /sys/module/dmatest/parameters/timeout
 	% echo 1 > /sys/module/dmatest/parameters/iterations
-	% echo 1 > /sys/kernel/debug/dmatest/run
+	% echo 1 > /sys/module/dmatest/parameters/run
+
+...or on the kernel command line:
+
+	dmatest.channel=dma0chan0 dmatest.timeout=2000 dmatest.iterations=1 dmatest.run=1
 
 Hint: available channel list could be extracted by running the following
 command:
 	% ls -1 /sys/class/dma/
 
-After a while you will start to get messages about current status or error like
-in the original code.
+Once started a message like "dmatest: Started 1 threads using dma0chan0" is
+emitted.  After that only test failure messages are reported until the test
+stops.
 
 Note that running a new test will not stop any in progress test.
 
-The following command should return actual state of the test.
-	% cat /sys/kernel/debug/dmatest/run
-
-To wait for test done the user may perform a busy loop that checks the state.
-
-	% while [ $(cat /sys/kernel/debug/dmatest/run) = "Y" ]
-	> do
-	> 	echo -n "."
-	> 	sleep 1
-	> done
-	> echo
+The following command returns the state of the test.
+	% cat /sys/module/dmatest/parameters/run
+
+To wait for test completion userpace can poll 'run' until it is false, or use
+the wait parameter.  Specifying 'wait=1' when loading the module causes module
+initialization to pause until a test run has completed, while reading
+/sys/module/dmatest/parameters/wait waits for any running test to complete
+before returning.  For example, the following scripts wait for 42 tests
+to complete before exiting.  Note that if 'iterations' is set to 'infinite' then
+waiting is disabled.
+
+Example:
+	% modprobe dmatest run=1 iterations=42 wait=1
+	% modprobe -r dmatest
+...or:
+	% modprobe dmatest run=1 iterations=42
+	% cat /sys/module/dmatest/parameters/wait
+	% modprobe -r dmatest
 
 	Part 3 - When built-in in the kernel...
 
@@ -62,21 +71,22 @@ case. You always could check them at run-time by running
 
 	Part 4 - Gathering the test results
 
-The module provides a storage for the test results in the memory. The gathered
-data could be used after test is done.
+Test results are printed to the kernel log buffer with the format:
 
-The special file 'results' in the debugfs represents gathered data of the in
-progress test. The messages collected are printed to the kernel log as well.
+"dmatest: result <channel>: <test id>: '<error msg>' with src_off=<val> dst_off=<val> len=<val> (<err code>)"
 
 Example of output:
-	% cat /sys/kernel/debug/dmatest/results
-	dma0chan0-copy0: #1: No errors with src_off=0x7bf dst_off=0x8ad len=0x3fea (0)
+	% dmesg | tail -n 1
+	dmatest: result dma0chan0-copy0: #1: No errors with src_off=0x7bf dst_off=0x8ad len=0x3fea (0)
 
 The message format is unified across the different types of errors. A number in
 the parens represents additional information, e.g. error code, error counter,
-or status.
+or status.  A test thread also emits a summary line at completion listing the
+number of tests executed, number that failed, and a result code.
 
-Comparison between buffers is stored to the dedicated structure.
+Example:
+	% dmesg | tail -n 1
+	dmatest: dma0chan0-copy0: summary 1 test, 0 failures 1000 iops 100000 KB/s (0)
 
-Note that the verify result is now accessible only via file 'results' in the
-debugfs.
+The details of a data miscompare error are also emitted, but do not follow the
+above format.
diff --git a/Documentation/hwmon/lm90 b/Documentation/hwmon/lm90
index b466974e142f..ab81013cc390 100644
--- a/Documentation/hwmon/lm90
+++ b/Documentation/hwmon/lm90
@@ -122,6 +122,12 @@ Supported chips:
     Prefix: 'g781'
     Addresses scanned: I2C 0x4c, 0x4d
     Datasheet: Not publicly available from GMT
+  * Texas Instruments TMP451
+    Prefix: 'tmp451'
+    Addresses scanned: I2C 0x4c
+    Datasheet: Publicly available at TI website
+               http://www.ti.com/litv/pdf/sbos686
+
 
 Author: Jean Delvare <khali@linux-fr.org>
 
diff --git a/Documentation/i2c/busses/i2c-i801 b/Documentation/i2c/busses/i2c-i801
index d29dea0f3232..7b0dcdb57173 100644
--- a/Documentation/i2c/busses/i2c-i801
+++ b/Documentation/i2c/busses/i2c-i801
@@ -25,6 +25,7 @@ Supported adapters:
   * Intel Avoton (SOC)
   * Intel Wellsburg (PCH)
   * Intel Coleto Creek (PCH)
+  * Intel Wildcat Point-LP (PCH)
    Datasheets: Publicly available at the Intel website
 
 On Intel Patsburg and later chipsets, both the normal host SMBus controller
diff --git a/Documentation/input/gamepad.txt b/Documentation/input/gamepad.txt
index 8002c894c6b0..31bb6a4029ef 100644
--- a/Documentation/input/gamepad.txt
+++ b/Documentation/input/gamepad.txt
@@ -122,12 +122,14 @@ D-Pad:
       BTN_DPAD_*
     Analog buttons are reported as:
       ABS_HAT0X and ABS_HAT0Y
+      (for ABS values negative is left/up, positive is right/down)
 
 Analog-Sticks:
   The left analog-stick is reported as ABS_X, ABS_Y. The right analog stick is
   reported as ABS_RX, ABS_RY. Zero, one or two sticks may be present.
   If analog-sticks provide digital buttons, they are mapped accordingly as
   BTN_THUMBL (first/left) and BTN_THUMBR (second/right).
+    (for ABS values negative is left/up, positive is right/down)
 
 Triggers:
   Trigger buttons can be available as digital or analog buttons or both. User-
@@ -138,6 +140,7 @@ Triggers:
   ABS_HAT2X (right/ZR) and BTN_TL2 or ABS_HAT2Y (left/ZL).
   If only one trigger-button combination is present (upper+lower), they are
   reported as "right" triggers (BTN_TR/ABS_HAT1X).
+    (ABS trigger values start at 0, pressure is reported as positive values)
 
 Menu-Pad:
   Menu buttons are always digital and are mapped according to their location
diff --git a/Documentation/kbuild/kconfig.txt b/Documentation/kbuild/kconfig.txt
index 8ef6dbb6a462..bbc99c0c1094 100644
--- a/Documentation/kbuild/kconfig.txt
+++ b/Documentation/kbuild/kconfig.txt
@@ -20,16 +20,9 @@ symbols have been introduced.
 To see a list of new config symbols when using "make oldconfig", use
 
 	cp user/some/old.config .config
-	yes "" | make oldconfig >conf.new
+	make listnewconfig
 
-and the config program will list as (NEW) any new symbols that have
-unknown values.  Of course, the .config file is also updated with
-new (default) values, so you can use:
-
-	grep "(NEW)" conf.new
-
-to see the new config symbols or you can use diffconfig to see the
-differences between the previous and new .config files:
+and the config program will list any new symbols, one per line.
 
 	scripts/diffconfig .config.old .config | less
 
diff --git a/Documentation/lockstat.txt b/Documentation/lockstat.txt
index dd2f7b26ca30..72d010689751 100644
--- a/Documentation/lockstat.txt
+++ b/Documentation/lockstat.txt
@@ -46,16 +46,14 @@ With these hooks we provide the following statistics:
  contentions       - number of lock acquisitions that had to wait
  wait time min     - shortest (non-0) time we ever had to wait for a lock
            max     - longest time we ever had to wait for a lock
-           total   - total time we spend waiting on this lock
+	   total   - total time we spend waiting on this lock
+	   avg     - average time spent waiting on this lock
  acq-bounces       - number of lock acquisitions that involved x-cpu data
  acquisitions      - number of times we took the lock
  hold time min     - shortest (non-0) time we ever held the lock
-           max     - longest time we ever held the lock
-           total   - total time this lock was held
-
-From these number various other statistics can be derived, such as:
-
- hold time average = hold time total / acquisitions
+	   max     - longest time we ever held the lock
+	   total   - total time this lock was held
+	   avg     - average time this lock was held
 
 These numbers are gathered per lock class, per read/write state (when
 applicable).
@@ -84,37 +82,38 @@ Look at the current lock statistics:
 
 # less /proc/lock_stat
 
-01 lock_stat version 0.3
-02 -----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
-03                               class name    con-bounces    contentions   waittime-min   waittime-max waittime-total    acq-bounces   acquisitions   holdtime-min   holdtime-max holdtime-total
-04 -----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
+01 lock_stat version 0.4
+02-----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
+03                              class name    con-bounces    contentions   waittime-min   waittime-max waittime-total   waittime-avg    acq-bounces   acquisitions   holdtime-min   holdtime-max holdtime-total   holdtime-avg
+04-----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
 05
-06                          &mm->mmap_sem-W:           233            538 18446744073708       22924.27      607243.51           1342          45806           1.71        8595.89     1180582.34
-07                          &mm->mmap_sem-R:           205            587 18446744073708       28403.36      731975.00           1940         412426           0.58      187825.45     6307502.88
-08                          ---------------
-09                            &mm->mmap_sem            487          [<ffffffff8053491f>] do_page_fault+0x466/0x928
-10                            &mm->mmap_sem            179          [<ffffffff802a6200>] sys_mprotect+0xcd/0x21d
-11                            &mm->mmap_sem            279          [<ffffffff80210a57>] sys_mmap+0x75/0xce
-12                            &mm->mmap_sem             76          [<ffffffff802a490b>] sys_munmap+0x32/0x59
-13                          ---------------
-14                            &mm->mmap_sem            270          [<ffffffff80210a57>] sys_mmap+0x75/0xce
-15                            &mm->mmap_sem            431          [<ffffffff8053491f>] do_page_fault+0x466/0x928
-16                            &mm->mmap_sem            138          [<ffffffff802a490b>] sys_munmap+0x32/0x59
-17                            &mm->mmap_sem            145          [<ffffffff802a6200>] sys_mprotect+0xcd/0x21d
+06                         &mm->mmap_sem-W:            46             84           0.26         939.10       16371.53         194.90          47291        2922365           0.16     2220301.69 17464026916.32        5975.99
+07                         &mm->mmap_sem-R:            37            100           1.31      299502.61      325629.52        3256.30         212344       34316685           0.10        7744.91    95016910.20           2.77
+08                         ---------------
+09                           &mm->mmap_sem              1          [<ffffffff811502a7>] khugepaged_scan_mm_slot+0x57/0x280
+19                           &mm->mmap_sem             96          [<ffffffff815351c4>] __do_page_fault+0x1d4/0x510
+11                           &mm->mmap_sem             34          [<ffffffff81113d77>] vm_mmap_pgoff+0x87/0xd0
+12                           &mm->mmap_sem             17          [<ffffffff81127e71>] vm_munmap+0x41/0x80
+13                         ---------------
+14                           &mm->mmap_sem              1          [<ffffffff81046fda>] dup_mmap+0x2a/0x3f0
+15                           &mm->mmap_sem             60          [<ffffffff81129e29>] SyS_mprotect+0xe9/0x250
+16                           &mm->mmap_sem             41          [<ffffffff815351c4>] __do_page_fault+0x1d4/0x510
+17                           &mm->mmap_sem             68          [<ffffffff81113d77>] vm_mmap_pgoff+0x87/0xd0
 18
-19 ...............................................................................................................................................................................................
+19.............................................................................................................................................................................................................................
 20
-21                              dcache_lock:           621            623           0.52         118.26        1053.02           6745          91930           0.29         316.29      118423.41
-22                              -----------
-23                              dcache_lock            179          [<ffffffff80378274>] _atomic_dec_and_lock+0x34/0x54
-24                              dcache_lock            113          [<ffffffff802cc17b>] d_alloc+0x19a/0x1eb
-25                              dcache_lock             99          [<ffffffff802ca0dc>] d_rehash+0x1b/0x44
-26                              dcache_lock            104          [<ffffffff802cbca0>] d_instantiate+0x36/0x8a
-27                              -----------
-28                              dcache_lock            192          [<ffffffff80378274>] _atomic_dec_and_lock+0x34/0x54
-29                              dcache_lock             98          [<ffffffff802ca0dc>] d_rehash+0x1b/0x44
-30                              dcache_lock             72          [<ffffffff802cc17b>] d_alloc+0x19a/0x1eb
-31                              dcache_lock            112          [<ffffffff802cbca0>] d_instantiate+0x36/0x8a
+21                         unix_table_lock:           110            112           0.21          49.24         163.91           1.46          21094          66312           0.12         624.42       31589.81           0.48
+22                         ---------------
+23                         unix_table_lock             45          [<ffffffff8150ad8e>] unix_create1+0x16e/0x1b0
+24                         unix_table_lock             47          [<ffffffff8150b111>] unix_release_sock+0x31/0x250
+25                         unix_table_lock             15          [<ffffffff8150ca37>] unix_find_other+0x117/0x230
+26                         unix_table_lock              5          [<ffffffff8150a09f>] unix_autobind+0x11f/0x1b0
+27                         ---------------
+28                         unix_table_lock             39          [<ffffffff8150b111>] unix_release_sock+0x31/0x250
+29                         unix_table_lock             49          [<ffffffff8150ad8e>] unix_create1+0x16e/0x1b0
+30                         unix_table_lock             20          [<ffffffff8150ca37>] unix_find_other+0x117/0x230
+31                         unix_table_lock              4          [<ffffffff8150a09f>] unix_autobind+0x11f/0x1b0
+
 
 This excerpt shows the first two lock class statistics. Line 01 shows the
 output version - each time the format changes this will be updated. Line 02-04
@@ -131,30 +130,30 @@ The integer part of the time values is in us.
 
 Dealing with nested locks, subclasses may appear:
 
-32...............................................................................................................................................................................................
+32...........................................................................................................................................................................................................................
 33
-34                               &rq->lock:         13128          13128           0.43         190.53      103881.26          97454        3453404           0.00         401.11    13224683.11
+34                               &rq->lock:       13128          13128           0.43         190.53      103881.26           7.91          97454        3453404           0.00         401.11    13224683.11           3.82
 35                               ---------
-36                               &rq->lock            645          [<ffffffff8103bfc4>] task_rq_lock+0x43/0x75
-37                               &rq->lock            297          [<ffffffff8104ba65>] try_to_wake_up+0x127/0x25a
-38                               &rq->lock            360          [<ffffffff8103c4c5>] select_task_rq_fair+0x1f0/0x74a
-39                               &rq->lock            428          [<ffffffff81045f98>] scheduler_tick+0x46/0x1fb
+36                               &rq->lock          645          [<ffffffff8103bfc4>] task_rq_lock+0x43/0x75
+37                               &rq->lock          297          [<ffffffff8104ba65>] try_to_wake_up+0x127/0x25a
+38                               &rq->lock          360          [<ffffffff8103c4c5>] select_task_rq_fair+0x1f0/0x74a
+39                               &rq->lock          428          [<ffffffff81045f98>] scheduler_tick+0x46/0x1fb
 40                               ---------
-41                               &rq->lock             77          [<ffffffff8103bfc4>] task_rq_lock+0x43/0x75
-42                               &rq->lock            174          [<ffffffff8104ba65>] try_to_wake_up+0x127/0x25a
-43                               &rq->lock           4715          [<ffffffff8103ed4b>] double_rq_lock+0x42/0x54
-44                               &rq->lock            893          [<ffffffff81340524>] schedule+0x157/0x7b8
+41                               &rq->lock           77          [<ffffffff8103bfc4>] task_rq_lock+0x43/0x75
+42                               &rq->lock          174          [<ffffffff8104ba65>] try_to_wake_up+0x127/0x25a
+43                               &rq->lock         4715          [<ffffffff8103ed4b>] double_rq_lock+0x42/0x54
+44                               &rq->lock          893          [<ffffffff81340524>] schedule+0x157/0x7b8
 45
-46...............................................................................................................................................................................................
+46...........................................................................................................................................................................................................................
 47
-48                             &rq->lock/1:         11526          11488           0.33         388.73      136294.31          21461          38404           0.00          37.93      109388.53
+48                             &rq->lock/1:        1526          11488           0.33         388.73      136294.31          11.86          21461          38404           0.00          37.93      109388.53           2.84
 49                             -----------
-50                             &rq->lock/1          11526          [<ffffffff8103ed58>] double_rq_lock+0x4f/0x54
+50                             &rq->lock/1        11526          [<ffffffff8103ed58>] double_rq_lock+0x4f/0x54
 51                             -----------
-52                             &rq->lock/1           5645          [<ffffffff8103ed4b>] double_rq_lock+0x42/0x54
-53                             &rq->lock/1           1224          [<ffffffff81340524>] schedule+0x157/0x7b8
-54                             &rq->lock/1           4336          [<ffffffff8103ed58>] double_rq_lock+0x4f/0x54
-55                             &rq->lock/1            181          [<ffffffff8104ba65>] try_to_wake_up+0x127/0x25a
+52                             &rq->lock/1         5645          [<ffffffff8103ed4b>] double_rq_lock+0x42/0x54
+53                             &rq->lock/1         1224          [<ffffffff81340524>] schedule+0x157/0x7b8
+54                             &rq->lock/1         4336          [<ffffffff8103ed58>] double_rq_lock+0x4f/0x54
+55                             &rq->lock/1          181          [<ffffffff8104ba65>] try_to_wake_up+0x127/0x25a
 
 Line 48 shows statistics for the second subclass (/1) of &rq->lock class
 (subclass starts from 0), since in this case, as line 50 suggests,
@@ -163,16 +162,16 @@ double_rq_lock actually acquires a nested lock of two spinlocks.
 View the top contending locks:
 
 # grep : /proc/lock_stat | head
-              &inode->i_data.tree_lock-W:            15          21657           0.18     1093295.30 11547131054.85             58          10415           0.16          87.51        6387.60
-              &inode->i_data.tree_lock-R:             0              0           0.00           0.00           0.00          23302         231198           0.25           8.45       98023.38
-                             dcache_lock:          1037           1161           0.38          45.32         774.51           6611         243371           0.15         306.48       77387.24
-                         &inode->i_mutex:           161            286 18446744073709       62882.54     1244614.55           3653          20598 18446744073709       62318.60     1693822.74
-                         &zone->lru_lock:            94             94           0.53           7.33          92.10           4366          32690           0.29          59.81       16350.06
-              &inode->i_data.i_mmap_mutex:            79             79           0.40           3.77          53.03          11779          87755           0.28         116.93       29898.44
-                        &q->__queue_lock:            48             50           0.52          31.62          86.31            774          13131           0.17         113.08       12277.52
-                        &rq->rq_lock_key:            43             47           0.74          68.50         170.63           3706          33929           0.22         107.99       17460.62
-                      &rq->rq_lock_key#2:            39             46           0.75           6.68          49.03           2979          32292           0.17         125.17       17137.63
-                         tasklist_lock-W:            15             15           1.45          10.87          32.70           1201           7390           0.58          62.55       13648.47
+			clockevents_lock:       2926159        2947636           0.15       46882.81  1784540466.34         605.41        3381345        3879161           0.00        2260.97    53178395.68          13.71
+		     tick_broadcast_lock:        346460         346717           0.18        2257.43    39364622.71         113.54        3642919        4242696           0.00        2263.79    49173646.60          11.59
+		  &mapping->i_mmap_mutex:        203896         203899           3.36      645530.05 31767507988.39      155800.21        3361776        8893984           0.17        2254.15    14110121.02           1.59
+			       &rq->lock:        135014         136909           0.18         606.09      842160.68           6.15        1540728       10436146           0.00         728.72    17606683.41           1.69
+	       &(&zone->lru_lock)->rlock:         93000          94934           0.16          59.18      188253.78           1.98        1199912        3809894           0.15         391.40     3559518.81           0.93
+			 tasklist_lock-W:         40667          41130           0.23        1189.42      428980.51          10.43         270278         510106           0.16         653.51     3939674.91           7.72
+			 tasklist_lock-R:         21298          21305           0.20        1310.05      215511.12          10.12         186204         241258           0.14        1162.33     1179779.23           4.89
+			      rcu_node_1:         47656          49022           0.16         635.41      193616.41           3.95         844888        1865423           0.00         764.26     1656226.96           0.89
+       &(&dentry->d_lockref.lock)->rlock:         39791          40179           0.15        1302.08       88851.96           2.21        2790851       12527025           0.10        1910.75     3379714.27           0.27
+			      rcu_node_0:         29203          30064           0.16         786.55     1555573.00          51.74          88963         244254           0.00         398.87      428872.51           1.76
 
 Clear the statistics:
 
diff --git a/Documentation/mutex-design.txt b/Documentation/mutex-design.txt
index 38c10fd7f411..1dfe62c3641d 100644
--- a/Documentation/mutex-design.txt
+++ b/Documentation/mutex-design.txt
@@ -116,11 +116,11 @@ using mutexes at the moment, please let me know if you find any. ]
 Implementation of mutexes
 -------------------------
 
-'struct mutex' is the new mutex type, defined in include/linux/mutex.h
-and implemented in kernel/mutex.c. It is a counter-based mutex with a
-spinlock and a wait-list. The counter has 3 states: 1 for "unlocked",
-0 for "locked" and negative numbers (usually -1) for "locked, potential
-waiters queued".
+'struct mutex' is the new mutex type, defined in include/linux/mutex.h and
+implemented in kernel/locking/mutex.c. It is a counter-based mutex with a
+spinlock and a wait-list. The counter has 3 states: 1 for "unlocked", 0 for
+"locked" and negative numbers (usually -1) for "locked, potential waiters
+queued".
 
 the APIs of 'struct mutex' have been streamlined:
 
diff --git a/Documentation/networking/ip-sysctl.txt b/Documentation/networking/ip-sysctl.txt
index 8b8a05787641..3c12d9a7ed00 100644
--- a/Documentation/networking/ip-sysctl.txt
+++ b/Documentation/networking/ip-sysctl.txt
@@ -577,9 +577,6 @@ tcp_limit_output_bytes - INTEGER
 	typical pfifo_fast qdiscs.
 	tcp_limit_output_bytes limits the number of bytes on qdisc
 	or device to reduce artificial RTT/cwnd and reduce bufferbloat.
-	Note: For GSO/TSO enabled flows, we try to have at least two
-	packets in flight. Reducing tcp_limit_output_bytes might also
-	reduce the size of individual GSO packet (64KB being the max)
 	Default: 131072
 
 tcp_challenge_ack_limit - INTEGER
diff --git a/Documentation/power/power_supply_class.txt b/Documentation/power/power_supply_class.txt
index 3f10b39b0346..89a8816990ff 100644
--- a/Documentation/power/power_supply_class.txt
+++ b/Documentation/power/power_supply_class.txt
@@ -135,11 +135,11 @@ CAPACITY_LEVEL - capacity level. This corresponds to
 POWER_SUPPLY_CAPACITY_LEVEL_*.
 
 TEMP - temperature of the power supply.
-TEMP_ALERT_MIN - minimum battery temperature alert value in milli centigrade.
-TEMP_ALERT_MAX - maximum battery temperature alert value in milli centigrade.
+TEMP_ALERT_MIN - minimum battery temperature alert.
+TEMP_ALERT_MAX - maximum battery temperature alert.
 TEMP_AMBIENT - ambient temperature.
-TEMP_AMBIENT_ALERT_MIN - minimum ambient temperature alert value in milli centigrade.
-TEMP_AMBIENT_ALERT_MAX - maximum ambient temperature alert value in milli centigrade.
+TEMP_AMBIENT_ALERT_MIN - minimum ambient temperature alert.
+TEMP_AMBIENT_ALERT_MAX - maximum ambient temperature alert.
 
 TIME_TO_EMPTY - seconds left for battery to be considered empty (i.e.
 while battery powers a load)
diff --git a/Documentation/pwm.txt b/Documentation/pwm.txt
index 1039b68fe9c6..93cb97974986 100644
--- a/Documentation/pwm.txt
+++ b/Documentation/pwm.txt
@@ -39,7 +39,7 @@ New users should use the pwm_get() function and pass to it the consumer
 device or a consumer name. pwm_put() is used to free the PWM device. Managed
 variants of these functions, devm_pwm_get() and devm_pwm_put(), also exist.
 
-After being requested a PWM has to be configured using:
+After being requested, a PWM has to be configured using:
 
 int pwm_config(struct pwm_device *pwm, int duty_ns, int period_ns);
 
@@ -94,7 +94,7 @@ for new drivers to use the generic PWM framework.
 A new PWM controller/chip can be added using pwmchip_add() and removed
 again with pwmchip_remove(). pwmchip_add() takes a filled in struct
 pwm_chip as argument which provides a description of the PWM chip, the
-number of PWM devices provider by the chip and the chip-specific
+number of PWM devices provided by the chip and the chip-specific
 implementation of the supported PWM operations to the framework.
 
 Locking
diff --git a/Documentation/timers/00-INDEX b/Documentation/timers/00-INDEX
index a9248da5cdbc..ef2ccbf77fa2 100644
--- a/Documentation/timers/00-INDEX
+++ b/Documentation/timers/00-INDEX
@@ -8,5 +8,9 @@ hpet_example.c
 	- sample hpet timer test program
 hrtimers.txt
 	- subsystem for high-resolution kernel timers
+NO_HZ.txt
+	- Summary of the different methods for the scheduler clock-interrupts management.
+timers-howto.txt
+	- how to insert delays in the kernel the right (tm) way.
 timer_stats.txt
 	- timer usage statistics
diff --git a/Documentation/usb/gadget_configfs.txt b/Documentation/usb/gadget_configfs.txt
index 8ec2a67c39b7..4cf53e406613 100644
--- a/Documentation/usb/gadget_configfs.txt
+++ b/Documentation/usb/gadget_configfs.txt
@@ -26,7 +26,7 @@ Linux provides a number of functions for gadgets to use.
 Creating a gadget means deciding what configurations there will be
 and which functions each configuration will provide.
 
-Configfs (please see Documentation/filesystems/configfs/*) lends itslef nicely
+Configfs (please see Documentation/filesystems/configfs/*) lends itself nicely
 for the purpose of telling the kernel about the above mentioned decision.
 This document is about how to do it.
 
@@ -99,7 +99,7 @@ directories must be created:
 $ mkdir configs/<name>.<number>
 
 where <name> can be any string which is legal in a filesystem and the
-<numebr> is the configuration's number, e.g.:
+<number> is the configuration's number, e.g.:
 
 $ mkdir configs/c.1
 
@@ -327,7 +327,7 @@ from the buffer to the cs), but it is up to the implementer of the
 two functions to decide what they actually do.
 
 typedef struct configured_structure cs;
-typedef struc specific_attribute sa;
+typedef struct specific_attribute sa;
 
                                        sa
                        +----------------------------------+
diff --git a/Documentation/virtual/kvm/00-INDEX b/Documentation/virtual/kvm/00-INDEX
new file mode 100644
index 000000000000..641ec9220179
--- /dev/null
+++ b/Documentation/virtual/kvm/00-INDEX
@@ -0,0 +1,24 @@
+00-INDEX
+	- this file.
+api.txt
+	- KVM userspace API.
+cpuid.txt
+	- KVM-specific cpuid leaves (x86).
+devices/
+	- KVM_CAP_DEVICE_CTRL userspace API.
+hypercalls.txt
+	- KVM hypercalls.
+locking.txt
+	- notes on KVM locks.
+mmu.txt
+	- the x86 kvm shadow mmu.
+msr.txt
+	- KVM-specific MSRs (x86).
+nested-vmx.txt
+	- notes on nested virtualization for Intel x86 processors.
+ppc-pv.txt
+	- the paravirtualization interface on PowerPC.
+review-checklist.txt
+	- review checklist for KVM patches.
+timekeeping.txt
+	- timekeeping virtualization for x86-based architectures.
diff --git a/Documentation/virtual/kvm/api.txt b/Documentation/virtual/kvm/api.txt
index 858aecf21db2..a30035dd4c26 100644
--- a/Documentation/virtual/kvm/api.txt
+++ b/Documentation/virtual/kvm/api.txt
@@ -1122,9 +1122,9 @@ struct kvm_cpuid2 {
 	struct kvm_cpuid_entry2 entries[0];
 };
 
-#define KVM_CPUID_FLAG_SIGNIFCANT_INDEX 1
-#define KVM_CPUID_FLAG_STATEFUL_FUNC    2
-#define KVM_CPUID_FLAG_STATE_READ_NEXT  4
+#define KVM_CPUID_FLAG_SIGNIFCANT_INDEX		BIT(0)
+#define KVM_CPUID_FLAG_STATEFUL_FUNC		BIT(1)
+#define KVM_CPUID_FLAG_STATE_READ_NEXT		BIT(2)
 
 struct kvm_cpuid_entry2 {
 	__u32 function;
@@ -1810,6 +1810,50 @@ registers, find a list below:
   PPC   | KVM_REG_PPC_TLB3PS	| 32
   PPC   | KVM_REG_PPC_EPTCFG	| 32
   PPC   | KVM_REG_PPC_ICP_STATE | 64
+  PPC   | KVM_REG_PPC_TB_OFFSET	| 64
+  PPC   | KVM_REG_PPC_SPMC1	| 32
+  PPC   | KVM_REG_PPC_SPMC2	| 32
+  PPC   | KVM_REG_PPC_IAMR	| 64
+  PPC   | KVM_REG_PPC_TFHAR	| 64
+  PPC   | KVM_REG_PPC_TFIAR	| 64
+  PPC   | KVM_REG_PPC_TEXASR	| 64
+  PPC   | KVM_REG_PPC_FSCR	| 64
+  PPC   | KVM_REG_PPC_PSPB	| 32
+  PPC   | KVM_REG_PPC_EBBHR	| 64
+  PPC   | KVM_REG_PPC_EBBRR	| 64
+  PPC   | KVM_REG_PPC_BESCR	| 64
+  PPC   | KVM_REG_PPC_TAR	| 64
+  PPC   | KVM_REG_PPC_DPDES	| 64
+  PPC   | KVM_REG_PPC_DAWR	| 64
+  PPC   | KVM_REG_PPC_DAWRX	| 64
+  PPC   | KVM_REG_PPC_CIABR	| 64
+  PPC   | KVM_REG_PPC_IC	| 64
+  PPC   | KVM_REG_PPC_VTB	| 64
+  PPC   | KVM_REG_PPC_CSIGR	| 64
+  PPC   | KVM_REG_PPC_TACR	| 64
+  PPC   | KVM_REG_PPC_TCSCR	| 64
+  PPC   | KVM_REG_PPC_PID	| 64
+  PPC   | KVM_REG_PPC_ACOP	| 64
+  PPC   | KVM_REG_PPC_VRSAVE	| 32
+  PPC   | KVM_REG_PPC_LPCR	| 64
+  PPC   | KVM_REG_PPC_PPR	| 64
+  PPC   | KVM_REG_PPC_ARCH_COMPAT 32
+  PPC   | KVM_REG_PPC_TM_GPR0	| 64
+          ...
+  PPC   | KVM_REG_PPC_TM_GPR31	| 64
+  PPC   | KVM_REG_PPC_TM_VSR0	| 128
+          ...
+  PPC   | KVM_REG_PPC_TM_VSR63	| 128
+  PPC   | KVM_REG_PPC_TM_CR	| 64
+  PPC   | KVM_REG_PPC_TM_LR	| 64
+  PPC   | KVM_REG_PPC_TM_CTR	| 64
+  PPC   | KVM_REG_PPC_TM_FPSCR	| 64
+  PPC   | KVM_REG_PPC_TM_AMR	| 64
+  PPC   | KVM_REG_PPC_TM_PPR	| 64
+  PPC   | KVM_REG_PPC_TM_VRSAVE	| 64
+  PPC   | KVM_REG_PPC_TM_VSCR	| 32
+  PPC   | KVM_REG_PPC_TM_DSCR	| 64
+  PPC   | KVM_REG_PPC_TM_TAR	| 64
 
 ARM registers are mapped using the lower 32 bits.  The upper 16 of that
 is the register group type, or coprocessor number:
@@ -2304,7 +2348,31 @@ Possible features:
 	  Depends on KVM_CAP_ARM_EL1_32BIT (arm64 only).
 
 
-4.83 KVM_GET_REG_LIST
+4.83 KVM_ARM_PREFERRED_TARGET
+
+Capability: basic
+Architectures: arm, arm64
+Type: vm ioctl
+Parameters: struct struct kvm_vcpu_init (out)
+Returns: 0 on success; -1 on error
+Errors:
+  ENODEV:    no preferred target available for the host
+
+This queries KVM for preferred CPU target type which can be emulated
+by KVM on underlying host.
+
+The ioctl returns struct kvm_vcpu_init instance containing information
+about preferred CPU target type and recommended features for it.  The
+kvm_vcpu_init->features bitmap returned will have feature bits set if
+the preferred target recommends setting these features, but this is
+not mandatory.
+
+The information returned by this ioctl can be used to prepare an instance
+of struct kvm_vcpu_init for KVM_ARM_VCPU_INIT ioctl which will result in
+in VCPU matching underlying host.
+
+
+4.84 KVM_GET_REG_LIST
 
 Capability: basic
 Architectures: arm, arm64
@@ -2323,8 +2391,7 @@ struct kvm_reg_list {
 This ioctl returns the guest registers that are supported for the
 KVM_GET_ONE_REG/KVM_SET_ONE_REG calls.
 
-
-4.84 KVM_ARM_SET_DEVICE_ADDR
+4.85 KVM_ARM_SET_DEVICE_ADDR
 
 Capability: KVM_CAP_ARM_SET_DEVICE_ADDR
 Architectures: arm, arm64
@@ -2362,7 +2429,7 @@ must be called after calling KVM_CREATE_IRQCHIP, but before calling
 KVM_RUN on any of the VCPUs.  Calling this ioctl twice for any of the
 base addresses will return -EEXIST.
 
-4.85 KVM_PPC_RTAS_DEFINE_TOKEN
+4.86 KVM_PPC_RTAS_DEFINE_TOKEN
 
 Capability: KVM_CAP_PPC_RTAS
 Architectures: ppc
@@ -2661,6 +2728,77 @@ and usually define the validity of a groups of registers. (e.g. one bit
 };
 
 
+4.81 KVM_GET_EMULATED_CPUID
+
+Capability: KVM_CAP_EXT_EMUL_CPUID
+Architectures: x86
+Type: system ioctl
+Parameters: struct kvm_cpuid2 (in/out)
+Returns: 0 on success, -1 on error
+
+struct kvm_cpuid2 {
+	__u32 nent;
+	__u32 flags;
+	struct kvm_cpuid_entry2 entries[0];
+};
+
+The member 'flags' is used for passing flags from userspace.
+
+#define KVM_CPUID_FLAG_SIGNIFCANT_INDEX		BIT(0)
+#define KVM_CPUID_FLAG_STATEFUL_FUNC		BIT(1)
+#define KVM_CPUID_FLAG_STATE_READ_NEXT		BIT(2)
+
+struct kvm_cpuid_entry2 {
+	__u32 function;
+	__u32 index;
+	__u32 flags;
+	__u32 eax;
+	__u32 ebx;
+	__u32 ecx;
+	__u32 edx;
+	__u32 padding[3];
+};
+
+This ioctl returns x86 cpuid features which are emulated by
+kvm.Userspace can use the information returned by this ioctl to query
+which features are emulated by kvm instead of being present natively.
+
+Userspace invokes KVM_GET_EMULATED_CPUID by passing a kvm_cpuid2
+structure with the 'nent' field indicating the number of entries in
+the variable-size array 'entries'. If the number of entries is too low
+to describe the cpu capabilities, an error (E2BIG) is returned. If the
+number is too high, the 'nent' field is adjusted and an error (ENOMEM)
+is returned. If the number is just right, the 'nent' field is adjusted
+to the number of valid entries in the 'entries' array, which is then
+filled.
+
+The entries returned are the set CPUID bits of the respective features
+which kvm emulates, as returned by the CPUID instruction, with unknown
+or unsupported feature bits cleared.
+
+Features like x2apic, for example, may not be present in the host cpu
+but are exposed by kvm in KVM_GET_SUPPORTED_CPUID because they can be
+emulated efficiently and thus not included here.
+
+The fields in each entry are defined as follows:
+
+  function: the eax value used to obtain the entry
+  index: the ecx value used to obtain the entry (for entries that are
+         affected by ecx)
+  flags: an OR of zero or more of the following:
+        KVM_CPUID_FLAG_SIGNIFCANT_INDEX:
+           if the index field is valid
+        KVM_CPUID_FLAG_STATEFUL_FUNC:
+           if cpuid for this function returns different values for successive
+           invocations; there will be several entries with the same function,
+           all with this flag set
+        KVM_CPUID_FLAG_STATE_READ_NEXT:
+           for KVM_CPUID_FLAG_STATEFUL_FUNC entries, set if this entry is
+           the first entry to be read by a cpu
+   eax, ebx, ecx, edx: the values returned by the cpuid instruction for
+         this function/index combination
+
+
 6. Capabilities that can be enabled
 -----------------------------------
 
diff --git a/Documentation/virtual/kvm/cpuid.txt b/Documentation/virtual/kvm/cpuid.txt
index 22ff659bc0fb..3c65feb83010 100644
--- a/Documentation/virtual/kvm/cpuid.txt
+++ b/Documentation/virtual/kvm/cpuid.txt
@@ -43,6 +43,13 @@ KVM_FEATURE_CLOCKSOURCE2           ||     3 || kvmclock available at msrs
 KVM_FEATURE_ASYNC_PF               ||     4 || async pf can be enabled by
                                    ||       || writing to msr 0x4b564d02
 ------------------------------------------------------------------------------
+KVM_FEATURE_STEAL_TIME             ||     5 || steal time can be enabled by
+                                   ||       || writing to msr 0x4b564d03.
+------------------------------------------------------------------------------
+KVM_FEATURE_PV_EOI                 ||     6 || paravirtualized end of interrupt
+                                   ||       || handler can be enabled by writing
+                                   ||       || to msr 0x4b564d04.
+------------------------------------------------------------------------------
 KVM_FEATURE_PV_UNHALT              ||     7 || guest checks this feature bit
                                    ||       || before enabling paravirtualized
                                    ||       || spinlock support.
diff --git a/Documentation/virtual/kvm/devices/vfio.txt b/Documentation/virtual/kvm/devices/vfio.txt
new file mode 100644
index 000000000000..ef51740c67ca
--- /dev/null
+++ b/Documentation/virtual/kvm/devices/vfio.txt
@@ -0,0 +1,22 @@
+VFIO virtual device
+===================
+
+Device types supported:
+  KVM_DEV_TYPE_VFIO
+
+Only one VFIO instance may be created per VM.  The created device
+tracks VFIO groups in use by the VM and features of those groups
+important to the correctness and acceleration of the VM.  As groups
+are enabled and disabled for use by the VM, KVM should be updated
+about their presence.  When registered with KVM, a reference to the
+VFIO-group is held by KVM.
+
+Groups:
+  KVM_DEV_VFIO_GROUP
+
+KVM_DEV_VFIO_GROUP attributes:
+  KVM_DEV_VFIO_GROUP_ADD: Add a VFIO group to VFIO-KVM device tracking
+  KVM_DEV_VFIO_GROUP_DEL: Remove a VFIO group from VFIO-KVM device tracking
+
+For each, kvm_device_attr.addr points to an int32_t file descriptor
+for the VFIO group.
diff --git a/Documentation/virtual/kvm/locking.txt b/Documentation/virtual/kvm/locking.txt
index 41b7ac9884b5..f8869410d40c 100644
--- a/Documentation/virtual/kvm/locking.txt
+++ b/Documentation/virtual/kvm/locking.txt
@@ -132,10 +132,14 @@ See the comments in spte_has_volatile_bits() and mmu_spte_update().
 ------------
 
 Name:		kvm_lock
-Type:		raw_spinlock
+Type:		spinlock_t
 Arch:		any
 Protects:	- vm_list
-		- hardware virtualization enable/disable
+
+Name:		kvm_count_lock
+Type:		raw_spinlock_t
+Arch:		any
+Protects:	- hardware virtualization enable/disable
 Comment:	'raw' because hardware enabling/disabling must be atomic /wrt
 		migration.
 
@@ -151,3 +155,14 @@ Type:		spinlock_t
 Arch:		any
 Protects:	-shadow page/shadow tlb entry
 Comment:	it is a spinlock since it is used in mmu notifier.
+
+Name:		kvm->srcu
+Type:		srcu lock
+Arch:		any
+Protects:	- kvm->memslots
+		- kvm->buses
+Comment:	The srcu read lock must be held while accessing memslots (e.g.
+		when using gfn_to_* functions) and while accessing in-kernel
+		MMIO/PIO address->device structure mapping (kvm->buses).
+		The srcu index can be stored in kvm_vcpu->srcu_idx per vcpu
+		if it is needed by multiple functions.
diff --git a/Documentation/vm/00-INDEX b/Documentation/vm/00-INDEX
index 5481c8ba3412..a39d06680e1c 100644
--- a/Documentation/vm/00-INDEX
+++ b/Documentation/vm/00-INDEX
@@ -4,10 +4,12 @@ active_mm.txt
 	- An explanation from Linus about tsk->active_mm vs tsk->mm.
 balance
 	- various information on memory balancing.
-hugepage-mmap.c
-	- Example app using huge page memory with the mmap system call.
-hugepage-shm.c
-	- Example app using huge page memory with Sys V shared memory system calls.
+cleancache.txt
+	- Intro to cleancache and page-granularity victim cache.
+frontswap.txt
+	- Outline frontswap, part of the transcendent memory frontend.
+highmem.txt
+	- Outline of highmem and common issues.
 hugetlbpage.txt
 	- a brief summary of hugetlbpage support in the Linux kernel.
 hwpoison.txt
@@ -16,21 +18,23 @@ ksm.txt
 	- how to use the Kernel Samepage Merging feature.
 locking
 	- info on how locking and synchronization is done in the Linux vm code.
-map_hugetlb.c
-	- an example program that uses the MAP_HUGETLB mmap flag.
 numa
 	- information about NUMA specific code in the Linux vm.
 numa_memory_policy.txt
 	- documentation of concepts and APIs of the 2.6 memory policy support.
 overcommit-accounting
 	- description of the Linux kernels overcommit handling modes.
-page-types.c
-	- Tool for querying page flags
 page_migration
 	- description of page migration in NUMA systems.
 pagemap.txt
 	- pagemap, from the userspace perspective
 slub.txt
 	- a short users guide for SLUB.
+soft-dirty.txt
+	- short explanation for soft-dirty PTEs
+transhuge.txt
+	- Transparent Hugepage Support, alternative way of using hugepages.
 unevictable-lru.txt
 	- Unevictable LRU infrastructure
+zswap.txt
+	- Intro to compressed cache for swap pages
diff --git a/Documentation/vm/split_page_table_lock b/Documentation/vm/split_page_table_lock
new file mode 100644
index 000000000000..7521d367f21d
--- /dev/null
+++ b/Documentation/vm/split_page_table_lock
@@ -0,0 +1,94 @@
+Split page table lock
+=====================
+
+Originally, mm->page_table_lock spinlock protected all page tables of the
+mm_struct. But this approach leads to poor page fault scalability of
+multi-threaded applications due high contention on the lock. To improve
+scalability, split page table lock was introduced.
+
+With split page table lock we have separate per-table lock to serialize
+access to the table. At the moment we use split lock for PTE and PMD
+tables. Access to higher level tables protected by mm->page_table_lock.
+
+There are helpers to lock/unlock a table and other accessor functions:
+ - pte_offset_map_lock()
+	maps pte and takes PTE table lock, returns pointer to the taken
+	lock;
+ - pte_unmap_unlock()
+	unlocks and unmaps PTE table;
+ - pte_alloc_map_lock()
+	allocates PTE table if needed and take the lock, returns pointer
+	to taken lock or NULL if allocation failed;
+ - pte_lockptr()
+	returns pointer to PTE table lock;
+ - pmd_lock()
+	takes PMD table lock, returns pointer to taken lock;
+ - pmd_lockptr()
+	returns pointer to PMD table lock;
+
+Split page table lock for PTE tables is enabled compile-time if
+CONFIG_SPLIT_PTLOCK_CPUS (usually 4) is less or equal to NR_CPUS.
+If split lock is disabled, all tables guaded by mm->page_table_lock.
+
+Split page table lock for PMD tables is enabled, if it's enabled for PTE
+tables and the architecture supports it (see below).
+
+Hugetlb and split page table lock
+---------------------------------
+
+Hugetlb can support several page sizes. We use split lock only for PMD
+level, but not for PUD.
+
+Hugetlb-specific helpers:
+ - huge_pte_lock()
+	takes pmd split lock for PMD_SIZE page, mm->page_table_lock
+	otherwise;
+ - huge_pte_lockptr()
+	returns pointer to table lock;
+
+Support of split page table lock by an architecture
+---------------------------------------------------
+
+There's no need in special enabling of PTE split page table lock:
+everything required is done by pgtable_page_ctor() and pgtable_page_dtor(),
+which must be called on PTE table allocation / freeing.
+
+Make sure the architecture doesn't use slab allocator for page table
+allocation: slab uses page->slab_cache and page->first_page for its pages.
+These fields share storage with page->ptl.
+
+PMD split lock only makes sense if you have more than two page table
+levels.
+
+PMD split lock enabling requires pgtable_pmd_page_ctor() call on PMD table
+allocation and pgtable_pmd_page_dtor() on freeing.
+
+Allocation usually happens in pmd_alloc_one(), freeing in pmd_free(), but
+make sure you cover all PMD table allocation / freeing paths: i.e X86_PAE
+preallocate few PMDs on pgd_alloc().
+
+With everything in place you can set CONFIG_ARCH_ENABLE_SPLIT_PMD_PTLOCK.
+
+NOTE: pgtable_page_ctor() and pgtable_pmd_page_ctor() can fail -- it must
+be handled properly.
+
+page->ptl
+---------
+
+page->ptl is used to access split page table lock, where 'page' is struct
+page of page containing the table. It shares storage with page->private
+(and few other fields in union).
+
+To avoid increasing size of struct page and have best performance, we use a
+trick:
+ - if spinlock_t fits into long, we use page->ptr as spinlock, so we
+   can avoid indirect access and save a cache line.
+ - if size of spinlock_t is bigger then size of long, we use page->ptl as
+   pointer to spinlock_t and allocate it dynamically. This allows to use
+   split lock with enabled DEBUG_SPINLOCK or DEBUG_LOCK_ALLOC, but costs
+   one more cache line for indirect access;
+
+The spinlock_t allocated in pgtable_page_ctor() for PTE table and in
+pgtable_pmd_page_ctor() for PMD table.
+
+Please, never access page->ptl directly -- use appropriate helper.
diff --git a/MAINTAINERS b/MAINTAINERS
index f3ef1d1f6029..63f30484932b 100644
--- a/MAINTAINERS
+++ b/MAINTAINERS
@@ -1070,7 +1070,6 @@ S:	Maintained
 ARM/NOMADIK ARCHITECTURE
 M:	Alessandro Rubini <rubini@unipv.it>
 M:	Linus Walleij <linus.walleij@linaro.org>
-M:	STEricsson <STEricsson_nomadik_linux@list.st.com>
 L:	linux-arm-kernel@lists.infradead.org (moderated for non-subscribers)
 S:	Maintained
 F:	arch/arm/mach-nomadik/
@@ -1426,7 +1425,7 @@ M:	Wolfram Sang <wsa@the-dreams.de>
 L:	linux-i2c@vger.kernel.org
 S:	Maintained
 F:	drivers/misc/eeprom/at24.c
-F:	include/linux/i2c/at24.h
+F:	include/linux/platform_data/at24.h
 
 ATA OVER ETHERNET (AOE) DRIVER
 M:	"Ed L. Cashin" <ecashin@coraid.com>
@@ -2469,7 +2468,7 @@ S:	Maintained
 F:	drivers/media/dvb-frontends/cxd2820r*
 
 CXGB3 ETHERNET DRIVER (CXGB3)
-M:	Divy Le Ray <divy@chelsio.com>
+M:	Santosh Raspatur <santosh@chelsio.com>
 L:	netdev@vger.kernel.org
 W:	http://www.chelsio.com
 S:	Supported
@@ -2849,7 +2848,9 @@ L:	dri-devel@lists.freedesktop.org
 L:	linux-tegra@vger.kernel.org
 T:	git git://anongit.freedesktop.org/tegra/linux.git
 S:	Supported
+F:	drivers/gpu/drm/tegra/
 F:	drivers/gpu/host1x/
+F:	include/linux/host1x.h
 F:	include/uapi/drm/tegra_drm.h
 F:	Documentation/devicetree/bindings/gpu/nvidia,tegra20-host1x.txt
 
@@ -3062,6 +3063,14 @@ W:	bluesmoke.sourceforge.net
 S:	Maintained
 F:	drivers/edac/amd64_edac*
 
+EDAC-CALXEDA
+M:	Doug Thompson <dougthompson@xmission.com>
+M:	Robert Richter <rric@kernel.org>
+L:	linux-edac@vger.kernel.org
+W:	bluesmoke.sourceforge.net
+S:	Maintained
+F:	drivers/edac/highbank*
+
 EDAC-CAVIUM
 M:	Ralf Baechle <ralf@linux-mips.org>
 M:	David Daney <david.daney@cavium.com>
@@ -3143,6 +3152,13 @@ W:	bluesmoke.sourceforge.net
 S:	Maintained
 F:	drivers/edac/i82975x_edac.c
 
+EDAC-MPC85XX
+M:	Johannes Thumshirn <johannes.thumshirn@men.de>
+L:	linux-edac@vger.kernel.org
+W:	bluesmoke.sourceforge.net
+S:	Maintained
+F:	drivers/edac/mpc85xx_edac.[ch]
+
 EDAC-PASEMI
 M:	Egor Martovetsky <egor@pasemi.com>
 L:	linux-edac@vger.kernel.org
@@ -4806,9 +4822,10 @@ S:	Maintained
 F:	drivers/staging/ktap/
 
 KCONFIG
-M:	Michal Marek <mmarek@suse.cz>
+M:	"Yann E. MORIN" <yann.morin.1998@free.fr>
 L:	linux-kbuild@vger.kernel.org
-S:	Odd Fixes
+T:	git://gitorious.org/linux-kconfig/linux-kconfig
+S:	Maintained
 F:	Documentation/kbuild/kconfig-language.txt
 F:	scripts/kconfig/
 
@@ -4871,7 +4888,8 @@ KERNEL VIRTUAL MACHINE (KVM)
 M:	Gleb Natapov <gleb@redhat.com>
 M:	Paolo Bonzini <pbonzini@redhat.com>
 L:	kvm@vger.kernel.org
-W:	http://linux-kvm.org
+W:	http://www.linux-kvm.org
+T:	git git://git.kernel.org/pub/scm/virt/kvm/kvm.git
 S:	Supported
 F:	Documentation/*/kvm*.txt
 F:	Documentation/virtual/kvm/
@@ -5211,6 +5229,7 @@ M:	Jean Delvare <khali@linux-fr.org>
 L:	lm-sensors@lm-sensors.org
 S:	Maintained
 F:	Documentation/hwmon/lm90
+F:	Documentation/devicetree/bindings/hwmon/lm90.txt
 F:	drivers/hwmon/lm90.c
 
 LM95234 HARDWARE MONITOR DRIVER
@@ -6780,8 +6799,7 @@ PWM SUBSYSTEM
 M:	Thierry Reding <thierry.reding@gmail.com>
 L:	linux-pwm@vger.kernel.org
 S:	Maintained
-W:	http://gitorious.org/linux-pwm
-T:	git git://gitorious.org/linux-pwm/linux-pwm.git
+T:	git git://git.kernel.org/pub/scm/linux/kernel/git/thierry.reding/linux-pwm.git
 F:	Documentation/pwm.txt
 F:	Documentation/devicetree/bindings/pwm/
 F:	include/linux/pwm.h
@@ -8942,8 +8960,8 @@ USB PEGASUS DRIVER
 M:	Petko Manolov <petkan@nucleusys.com>
 L:	linux-usb@vger.kernel.org
 L:	netdev@vger.kernel.org
-T:	git git://git.code.sf.net/p/pegasus2/git
-W:	http://pegasus2.sourceforge.net/
+T:	git git://github.com/petkan/pegasus.git
+W:	https://github.com/petkan/pegasus
 S:	Maintained
 F:	drivers/net/usb/pegasus.*
 
@@ -8964,8 +8982,8 @@ USB RTL8150 DRIVER
 M:	Petko Manolov <petkan@nucleusys.com>
 L:	linux-usb@vger.kernel.org
 L:	netdev@vger.kernel.org
-T:	git git://git.code.sf.net/p/pegasus2/git
-W:	http://pegasus2.sourceforge.net/
+T:	git git://github.com/petkan/rtl8150.git
+W:	https://github.com/petkan/rtl8150
 S:	Maintained
 F:	drivers/net/usb/rtl8150.c
 
diff --git a/Makefile b/Makefile
index 606a66cdcdb8..920ad07180c9 100644
--- a/Makefile
+++ b/Makefile
@@ -22,6 +22,9 @@ LC_COLLATE=C
 LC_NUMERIC=C
 export LC_COLLATE LC_NUMERIC
 
+# Avoid interference with shell env settings
+unexport GREP_OPTIONS
+
 # We are using a recursive build, so we need to do a little thinking
 # to get the ordering right.
 #
@@ -659,6 +662,12 @@ KBUILD_CFLAGS	+= $(call cc-option,-fno-strict-overflow)
 # conserve stack if available
 KBUILD_CFLAGS   += $(call cc-option,-fconserve-stack)
 
+# disallow errors like 'EXPORT_GPL(foo);' with missing header
+KBUILD_CFLAGS   += $(call cc-option,-Werror=implicit-int)
+
+# require functions to have arguments in prototypes, not empty 'int foo()'
+KBUILD_CFLAGS   += $(call cc-option,-Werror=strict-prototypes)
+
 # use the deterministic mode of AR if available
 KBUILD_ARFLAGS := $(call ar-option,D)
 
diff --git a/arch/Kconfig b/arch/Kconfig
index ded747c7b74c..f1cf895c040f 100644
--- a/arch/Kconfig
+++ b/arch/Kconfig
@@ -207,9 +207,6 @@ config HAVE_DMA_ATTRS
 config HAVE_DMA_CONTIGUOUS
 	bool
 
-config USE_GENERIC_SMP_HELPERS
-	bool
-
 config GENERIC_SMP_IDLE_THREAD
        bool
 
diff --git a/arch/alpha/Kconfig b/arch/alpha/Kconfig
index 35a300d4a9fb..135c674eaf9e 100644
--- a/arch/alpha/Kconfig
+++ b/arch/alpha/Kconfig
@@ -1,6 +1,7 @@
 config ALPHA
 	bool
 	default y
+	select ARCH_MIGHT_HAVE_PC_PARPORT
 	select HAVE_AOUT
 	select HAVE_IDE
 	select HAVE_OPROFILE
@@ -522,7 +523,6 @@ config ARCH_MAY_HAVE_PC_FDC
 config SMP
 	bool "Symmetric multi-processing support"
 	depends on ALPHA_SABLE || ALPHA_LYNX || ALPHA_RAWHIDE || ALPHA_DP264 || ALPHA_WILDFIRE || ALPHA_TITAN || ALPHA_GENERIC || ALPHA_SHARK || ALPHA_MARVEL
-	select USE_GENERIC_SMP_HELPERS
 	---help---
 	  This enables support for systems with more than one CPU. If you have
 	  a system with only one CPU, like most personal computers, say N. If
diff --git a/arch/alpha/include/asm/pgalloc.h b/arch/alpha/include/asm/pgalloc.h
index bc2a0daf2d92..aab14a019c20 100644
--- a/arch/alpha/include/asm/pgalloc.h
+++ b/arch/alpha/include/asm/pgalloc.h
@@ -72,7 +72,10 @@ pte_alloc_one(struct mm_struct *mm, unsigned long address)
 	if (!pte)
 		return NULL;
 	page = virt_to_page(pte);
-	pgtable_page_ctor(page);
+	if (!pgtable_page_ctor(page)) {
+		__free_page(page);
+		return NULL;
+	}
 	return page;
 }
 
diff --git a/arch/alpha/include/asm/thread_info.h b/arch/alpha/include/asm/thread_info.h
index 52cd2a4a3ff4..453597b91f3a 100644
--- a/arch/alpha/include/asm/thread_info.h
+++ b/arch/alpha/include/asm/thread_info.h
@@ -58,8 +58,6 @@ register struct thread_info *__current_thread_info __asm__("$8");
 #define THREAD_SIZE_ORDER 1
 #define THREAD_SIZE (2*PAGE_SIZE)
 
-#define PREEMPT_ACTIVE		0x40000000
-
 /*
  * Thread information flags:
  * - these are process state flags and used from assembly
diff --git a/arch/arc/Kconfig b/arch/arc/Kconfig
index 5ede5460c806..2ee0c9bfd032 100644
--- a/arch/arc/Kconfig
+++ b/arch/arc/Kconfig
@@ -125,7 +125,6 @@ config ARC_PLAT_NEEDS_CPU_TO_DMA
 config SMP
 	bool "Symmetric Multi-Processing (Incomplete)"
 	default n
-	select USE_GENERIC_SMP_HELPERS
 	help
 	  This enables support for systems with more than one CPU. If you have
 	  a system with only one CPU, like most personal computers, say N. If
diff --git a/arch/arc/boot/dts/abilis_tb100.dtsi b/arch/arc/boot/dts/abilis_tb100.dtsi
index d9f8249aa66e..3942634f805a 100644
--- a/arch/arc/boot/dts/abilis_tb100.dtsi
+++ b/arch/arc/boot/dts/abilis_tb100.dtsi
@@ -43,124 +43,124 @@
 		iomux: iomux@FF10601c {
 			/* Port 1 */
 			pctl_tsin_s0: pctl-tsin-s0 {   /* Serial TS-in 0 */
-				pingrp = "mis0_pins";
+				abilis,function = "mis0";
 			};
 			pctl_tsin_s1: pctl-tsin-s1 {   /* Serial TS-in 1 */
-				pingrp = "mis1_pins";
+				abilis,function = "mis1";
 			};
 			pctl_gpio_a: pctl-gpio-a {     /* GPIO bank A */
-				pingrp = "gpioa_pins";
+				abilis,function = "gpioa";
 			};
 			pctl_tsin_p1: pctl-tsin-p1 {   /* Parallel TS-in 1 */
-				pingrp = "mip1_pins";
+				abilis,function = "mip1";
 			};
 			/* Port 2 */
 			pctl_tsin_s2: pctl-tsin-s2 {   /* Serial TS-in 2 */
-				pingrp = "mis2_pins";
+				abilis,function = "mis2";
 			};
 			pctl_tsin_s3: pctl-tsin-s3 {   /* Serial TS-in 3 */
-				pingrp = "mis3_pins";
+				abilis,function = "mis3";
 			};
 			pctl_gpio_c: pctl-gpio-c {     /* GPIO bank C */
-				pingrp = "gpioc_pins";
+				abilis,function = "gpioc";
 			};
 			pctl_tsin_p3: pctl-tsin-p3 {   /* Parallel TS-in 3 */
-				pingrp = "mip3_pins";
+				abilis,function = "mip3";
 			};
 			/* Port 3 */
 			pctl_tsin_s4: pctl-tsin-s4 {   /* Serial TS-in 4 */
-				pingrp = "mis4_pins";
+				abilis,function = "mis4";
 			};
 			pctl_tsin_s5: pctl-tsin-s5 {   /* Serial TS-in 5 */
-				pingrp = "mis5_pins";
+				abilis,function = "mis5";
 			};
 			pctl_gpio_e: pctl-gpio-e {     /* GPIO bank E */
-				pingrp = "gpioe_pins";
+				abilis,function = "gpioe";
 			};
 			pctl_tsin_p5: pctl-tsin-p5 {   /* Parallel TS-in 5 */
-				pingrp = "mip5_pins";
+				abilis,function = "mip5";
 			};
 			/* Port 4 */
 			pctl_tsin_s6: pctl-tsin-s6 {   /* Serial TS-in 6 */
-				pingrp = "mis6_pins";
+				abilis,function = "mis6";
 			};
 			pctl_tsin_s7: pctl-tsin-s7 {   /* Serial TS-in 7 */
-				pingrp = "mis7_pins";
+				abilis,function = "mis7";
 			};
 			pctl_gpio_g: pctl-gpio-g {     /* GPIO bank G */
-				pingrp = "gpiog_pins";
+				abilis,function = "gpiog";
 			};
 			pctl_tsin_p7: pctl-tsin-p7 {   /* Parallel TS-in 7 */
-				pingrp = "mip7_pins";
+				abilis,function = "mip7";
 			};
 			/* Port 5 */
 			pctl_gpio_j: pctl-gpio-j {     /* GPIO bank J */
-				pingrp = "gpioj_pins";
+				abilis,function = "gpioj";
 			};
 			pctl_gpio_k: pctl-gpio-k {     /* GPIO bank K */
-				pingrp = "gpiok_pins";
+				abilis,function = "gpiok";
 			};
 			pctl_ciplus: pctl-ciplus {     /* CI+ interface */
-				pingrp = "ciplus_pins";
+				abilis,function = "ciplus";
 			};
 			pctl_mcard: pctl-mcard {       /* M-Card interface */
-				pingrp = "mcard_pins";
+				abilis,function = "mcard";
 			};
 			/* Port 6 */
 			pctl_tsout_p: pctl-tsout-p {   /* Parallel TS-out */
-				pingrp = "mop_pins";
+				abilis,function = "mop";
 			};
 			pctl_tsout_s0: pctl-tsout-s0 { /* Serial TS-out 0 */
-				pingrp = "mos0_pins";
+				abilis,function = "mos0";
 			};
 			pctl_tsout_s1: pctl-tsout-s1 { /* Serial TS-out 1 */
-				pingrp = "mos1_pins";
+				abilis,function = "mos1";
 			};
 			pctl_tsout_s2: pctl-tsout-s2 { /* Serial TS-out 2 */
-				pingrp = "mos2_pins";
+				abilis,function = "mos2";
 			};
 			pctl_tsout_s3: pctl-tsout-s3 { /* Serial TS-out 3 */
-				pingrp = "mos3_pins";
+				abilis,function = "mos3";
 			};
 			/* Port 7 */
 			pctl_uart0: pctl-uart0 {       /* UART 0 */
-				pingrp = "uart0_pins";
+				abilis,function = "uart0";
 			};
 			pctl_uart1: pctl-uart1 {       /* UART 1 */
-				pingrp = "uart1_pins";
+				abilis,function = "uart1";
 			};
 			pctl_gpio_l: pctl-gpio-l {     /* GPIO bank L */
-				pingrp = "gpiol_pins";
+				abilis,function = "gpiol";
 			};
 			pctl_gpio_m: pctl-gpio-m {     /* GPIO bank M */
-				pingrp = "gpiom_pins";
+				abilis,function = "gpiom";
 			};
 			/* Port 8 */
 			pctl_spi3: pctl-spi3 {
-				pingrp = "spi3_pins";
+				abilis,function = "spi3";
 			};
 			/* Port 9 */
 			pctl_spi1: pctl-spi1 {
-				pingrp = "spi1_pins";
+				abilis,function = "spi1";
 			};
 			pctl_gpio_n: pctl-gpio-n {
-				pingrp = "gpion_pins";
+				abilis,function = "gpion";
 			};
 			/* Unmuxed GPIOs */
 			pctl_gpio_b: pctl-gpio-b {
-				pingrp = "gpiob_pins";
+				abilis,function = "gpiob";
 			};
 			pctl_gpio_d: pctl-gpio-d {
-				pingrp = "gpiod_pins";
+				abilis,function = "gpiod";
 			};
 			pctl_gpio_f: pctl-gpio-f {
-				pingrp = "gpiof_pins";
+				abilis,function = "gpiof";
 			};
 			pctl_gpio_h: pctl-gpio-h {
-				pingrp = "gpioh_pins";
+				abilis,function = "gpioh";
 			};
 			pctl_gpio_i: pctl-gpio-i {
-				pingrp = "gpioi_pins";
+				abilis,function = "gpioi";
 			};
 		};
 
@@ -172,9 +172,10 @@
 			interrupts = <27 2>;
 			reg = <0xFF140000 0x1000>;
 			gpio-controller;
-			#gpio-cells = <1>;
-			gpio-base  = <0>;
-			gpio-pins = <&pctl_gpio_a>;
+			#gpio-cells = <2>;
+			abilis,ngpio = <3>;
+			gpio-ranges = <&iomux 0 0 0>;
+			gpio-ranges-group-names = "gpioa";
 		};
 		gpiob: gpio@FF141000 {
 			compatible = "abilis,tb10x-gpio";
@@ -184,9 +185,10 @@
 			interrupts = <27 2>;
 			reg = <0xFF141000 0x1000>;
 			gpio-controller;
-			#gpio-cells = <1>;
-			gpio-base  = <3>;
-			gpio-pins = <&pctl_gpio_b>;
+			#gpio-cells = <2>;
+			abilis,ngpio = <2>;
+			gpio-ranges = <&iomux 0 0 0>;
+			gpio-ranges-group-names = "gpiob";
 		};
 		gpioc: gpio@FF142000 {
 			compatible = "abilis,tb10x-gpio";
@@ -196,9 +198,10 @@
 			interrupts = <27 2>;
 			reg = <0xFF142000 0x1000>;
 			gpio-controller;
-			#gpio-cells = <1>;
-			gpio-base  = <5>;
-			gpio-pins = <&pctl_gpio_c>;
+			#gpio-cells = <2>;
+			abilis,ngpio = <3>;
+			gpio-ranges = <&iomux 0 0 0>;
+			gpio-ranges-group-names = "gpioc";
 		};
 		gpiod: gpio@FF143000 {
 			compatible = "abilis,tb10x-gpio";
@@ -208,9 +211,10 @@
 			interrupts = <27 2>;
 			reg = <0xFF143000 0x1000>;
 			gpio-controller;
-			#gpio-cells = <1>;
-			gpio-base  = <8>;
-			gpio-pins = <&pctl_gpio_d>;
+			#gpio-cells = <2>;
+			abilis,ngpio = <2>;
+			gpio-ranges = <&iomux 0 0 0>;
+			gpio-ranges-group-names = "gpiod";
 		};
 		gpioe: gpio@FF144000 {
 			compatible = "abilis,tb10x-gpio";
@@ -220,9 +224,10 @@
 			interrupts = <27 2>;
 			reg = <0xFF144000 0x1000>;
 			gpio-controller;
-			#gpio-cells = <1>;
-			gpio-base  = <10>;
-			gpio-pins = <&pctl_gpio_e>;
+			#gpio-cells = <2>;
+			abilis,ngpio = <3>;
+			gpio-ranges = <&iomux 0 0 0>;
+			gpio-ranges-group-names = "gpioe";
 		};
 		gpiof: gpio@FF145000 {
 			compatible = "abilis,tb10x-gpio";
@@ -232,9 +237,10 @@
 			interrupts = <27 2>;
 			reg = <0xFF145000 0x1000>;
 			gpio-controller;
-			#gpio-cells = <1>;
-			gpio-base  = <13>;
-			gpio-pins = <&pctl_gpio_f>;
+			#gpio-cells = <2>;
+			abilis,ngpio = <2>;
+			gpio-ranges = <&iomux 0 0 0>;
+			gpio-ranges-group-names = "gpiof";
 		};
 		gpiog: gpio@FF146000 {
 			compatible = "abilis,tb10x-gpio";
@@ -244,9 +250,10 @@
 			interrupts = <27 2>;
 			reg = <0xFF146000 0x1000>;
 			gpio-controller;
-			#gpio-cells = <1>;
-			gpio-base  = <15>;
-			gpio-pins = <&pctl_gpio_g>;
+			#gpio-cells = <2>;
+			abilis,ngpio = <3>;
+			gpio-ranges = <&iomux 0 0 0>;
+			gpio-ranges-group-names = "gpiog";
 		};
 		gpioh: gpio@FF147000 {
 			compatible = "abilis,tb10x-gpio";
@@ -256,9 +263,10 @@
 			interrupts = <27 2>;
 			reg = <0xFF147000 0x1000>;
 			gpio-controller;
-			#gpio-cells = <1>;
-			gpio-base  = <18>;
-			gpio-pins = <&pctl_gpio_h>;
+			#gpio-cells = <2>;
+			abilis,ngpio = <2>;
+			gpio-ranges = <&iomux 0 0 0>;
+			gpio-ranges-group-names = "gpioh";
 		};
 		gpioi: gpio@FF148000 {
 			compatible = "abilis,tb10x-gpio";
@@ -268,9 +276,10 @@
 			interrupts = <27 2>;
 			reg = <0xFF148000 0x1000>;
 			gpio-controller;
-			#gpio-cells = <1>;
-			gpio-base  = <20>;
-			gpio-pins = <&pctl_gpio_i>;
+			#gpio-cells = <2>;
+			abilis,ngpio = <12>;
+			gpio-ranges = <&iomux 0 0 0>;
+			gpio-ranges-group-names = "gpioi";
 		};
 		gpioj: gpio@FF149000 {
 			compatible = "abilis,tb10x-gpio";
@@ -280,9 +289,10 @@
 			interrupts = <27 2>;
 			reg = <0xFF149000 0x1000>;
 			gpio-controller;
-			#gpio-cells = <1>;
-			gpio-base  = <32>;
-			gpio-pins = <&pctl_gpio_j>;
+			#gpio-cells = <2>;
+			abilis,ngpio = <32>;
+			gpio-ranges = <&iomux 0 0 0>;
+			gpio-ranges-group-names = "gpioj";
 		};
 		gpiok: gpio@FF14a000 {
 			compatible = "abilis,tb10x-gpio";
@@ -292,9 +302,10 @@
 			interrupts = <27 2>;
 			reg = <0xFF14A000 0x1000>;
 			gpio-controller;
-			#gpio-cells = <1>;
-			gpio-base  = <64>;
-			gpio-pins = <&pctl_gpio_k>;
+			#gpio-cells = <2>;
+			abilis,ngpio = <22>;
+			gpio-ranges = <&iomux 0 0 0>;
+			gpio-ranges-group-names = "gpiok";
 		};
 		gpiol: gpio@FF14b000 {
 			compatible = "abilis,tb10x-gpio";
@@ -304,9 +315,10 @@
 			interrupts = <27 2>;
 			reg = <0xFF14B000 0x1000>;
 			gpio-controller;
-			#gpio-cells = <1>;
-			gpio-base  = <86>;
-			gpio-pins = <&pctl_gpio_l>;
+			#gpio-cells = <2>;
+			abilis,ngpio = <4>;
+			gpio-ranges = <&iomux 0 0 0>;
+			gpio-ranges-group-names = "gpiol";
 		};
 		gpiom: gpio@FF14c000 {
 			compatible = "abilis,tb10x-gpio";
@@ -316,9 +328,10 @@
 			interrupts = <27 2>;
 			reg = <0xFF14C000 0x1000>;
 			gpio-controller;
-			#gpio-cells = <1>;
-			gpio-base  = <90>;
-			gpio-pins = <&pctl_gpio_m>;
+			#gpio-cells = <2>;
+			abilis,ngpio = <4>;
+			gpio-ranges = <&iomux 0 0 0>;
+			gpio-ranges-group-names = "gpiom";
 		};
 		gpion: gpio@FF14d000 {
 			compatible = "abilis,tb10x-gpio";
@@ -328,9 +341,10 @@
 			interrupts = <27 2>;
 			reg = <0xFF14D000 0x1000>;
 			gpio-controller;
-			#gpio-cells = <1>;
-			gpio-base  = <94>;
-			gpio-pins = <&pctl_gpio_n>;
+			#gpio-cells = <2>;
+			abilis,ngpio = <5>;
+			gpio-ranges = <&iomux 0 0 0>;
+			gpio-ranges-group-names = "gpion";
 		};
 	};
 };
diff --git a/arch/arc/boot/dts/abilis_tb100_dvk.dts b/arch/arc/boot/dts/abilis_tb100_dvk.dts
index ebc313a9f5b2..3dd6ed941464 100644
--- a/arch/arc/boot/dts/abilis_tb100_dvk.dts
+++ b/arch/arc/boot/dts/abilis_tb100_dvk.dts
@@ -64,62 +64,62 @@
 			compatible = "gpio-leds";
 			power {
 				label = "Power";
-				gpios = <&gpioi 0>;
+				gpios = <&gpioi 0 0>;
 				linux,default-trigger = "default-on";
 			};
 			heartbeat {
 				label = "Heartbeat";
-				gpios = <&gpioi 1>;
+				gpios = <&gpioi 1 0>;
 				linux,default-trigger = "heartbeat";
 			};
 			led2 {
 				label = "LED2";
-				gpios = <&gpioi 2>;
+				gpios = <&gpioi 2 0>;
 				default-state = "off";
 			};
 			led3 {
 				label = "LED3";
-				gpios = <&gpioi 3>;
+				gpios = <&gpioi 3 0>;
 				default-state = "off";
 			};
 			led4 {
 				label = "LED4";
-				gpios = <&gpioi 4>;
+				gpios = <&gpioi 4 0>;
 				default-state = "off";
 			};
 			led5 {
 				label = "LED5";
-				gpios = <&gpioi 5>;
+				gpios = <&gpioi 5 0>;
 				default-state = "off";
 			};
 			led6 {
 				label = "LED6";
-				gpios = <&gpioi 6>;
+				gpios = <&gpioi 6 0>;
 				default-state = "off";
 			};
 			led7 {
 				label = "LED7";
-				gpios = <&gpioi 7>;
+				gpios = <&gpioi 7 0>;
 				default-state = "off";
 			};
 			led8 {
 				label = "LED8";
-				gpios = <&gpioi 8>;
+				gpios = <&gpioi 8 0>;
 				default-state = "off";
 			};
 			led9 {
 				label = "LED9";
-				gpios = <&gpioi 9>;
+				gpios = <&gpioi 9 0>;
 				default-state = "off";
 			};
 			led10 {
 				label = "LED10";
-				gpios = <&gpioi 10>;
+				gpios = <&gpioi 10 0>;
 				default-state = "off";
 			};
 			led11 {
 				label = "LED11";
-				gpios = <&gpioi 11>;
+				gpios = <&gpioi 11 0>;
 				default-state = "off";
 			};
 		};
diff --git a/arch/arc/boot/dts/abilis_tb101.dtsi b/arch/arc/boot/dts/abilis_tb101.dtsi
index da8ca7941e67..b0467229a5c4 100644
--- a/arch/arc/boot/dts/abilis_tb101.dtsi
+++ b/arch/arc/boot/dts/abilis_tb101.dtsi
@@ -43,133 +43,133 @@
 		iomux: iomux@FF10601c {
 			/* Port 1 */
 			pctl_tsin_s0: pctl-tsin-s0 {   /* Serial TS-in 0 */
-				pingrp = "mis0_pins";
+				abilis,function = "mis0";
 			};
 			pctl_tsin_s1: pctl-tsin-s1 {   /* Serial TS-in 1 */
-				pingrp = "mis1_pins";
+				abilis,function = "mis1";
 			};
 			pctl_gpio_a: pctl-gpio-a {     /* GPIO bank A */
-				pingrp = "gpioa_pins";
+				abilis,function = "gpioa";
 			};
 			pctl_tsin_p1: pctl-tsin-p1 {   /* Parallel TS-in 1 */
-				pingrp = "mip1_pins";
+				abilis,function = "mip1";
 			};
 			/* Port 2 */
 			pctl_tsin_s2: pctl-tsin-s2 {   /* Serial TS-in 2 */
-				pingrp = "mis2_pins";
+				abilis,function = "mis2";
 			};
 			pctl_tsin_s3: pctl-tsin-s3 {   /* Serial TS-in 3 */
-				pingrp = "mis3_pins";
+				abilis,function = "mis3";
 			};
 			pctl_gpio_c: pctl-gpio-c {     /* GPIO bank C */
-				pingrp = "gpioc_pins";
+				abilis,function = "gpioc";
 			};
 			pctl_tsin_p3: pctl-tsin-p3 {   /* Parallel TS-in 3 */
-				pingrp = "mip3_pins";
+				abilis,function = "mip3";
 			};
 			/* Port 3 */
 			pctl_tsin_s4: pctl-tsin-s4 {   /* Serial TS-in 4 */
-				pingrp = "mis4_pins";
+				abilis,function = "mis4";
 			};
 			pctl_tsin_s5: pctl-tsin-s5 {   /* Serial TS-in 5 */
-				pingrp = "mis5_pins";
+				abilis,function = "mis5";
 			};
 			pctl_gpio_e: pctl-gpio-e {     /* GPIO bank E */
-				pingrp = "gpioe_pins";
+				abilis,function = "gpioe";
 			};
 			pctl_tsin_p5: pctl-tsin-p5 {   /* Parallel TS-in 5 */
-				pingrp = "mip5_pins";
+				abilis,function = "mip5";
 			};
 			/* Port 4 */
 			pctl_tsin_s6: pctl-tsin-s6 {   /* Serial TS-in 6 */
-				pingrp = "mis6_pins";
+				abilis,function = "mis6";
 			};
 			pctl_tsin_s7: pctl-tsin-s7 {   /* Serial TS-in 7 */
-				pingrp = "mis7_pins";
+				abilis,function = "mis7";
 			};
 			pctl_gpio_g: pctl-gpio-g {     /* GPIO bank G */
-				pingrp = "gpiog_pins";
+				abilis,function = "gpiog";
 			};
 			pctl_tsin_p7: pctl-tsin-p7 {   /* Parallel TS-in 7 */
-				pingrp = "mip7_pins";
+				abilis,function = "mip7";
 			};
 			/* Port 5 */
 			pctl_gpio_j: pctl-gpio-j {     /* GPIO bank J */
-				pingrp = "gpioj_pins";
+				abilis,function = "gpioj";
 			};
 			pctl_gpio_k: pctl-gpio-k {     /* GPIO bank K */
-				pingrp = "gpiok_pins";
+				abilis,function = "gpiok";
 			};
 			pctl_ciplus: pctl-ciplus {     /* CI+ interface */
-				pingrp = "ciplus_pins";
+				abilis,function = "ciplus";
 			};
 			pctl_mcard: pctl-mcard {       /* M-Card interface */
-				pingrp = "mcard_pins";
+				abilis,function = "mcard";
 			};
 			pctl_stc0: pctl-stc0 {         /* Smart card I/F 0 */
-				pingrp = "stc0_pins";
+				abilis,function = "stc0";
 			};
 			pctl_stc1: pctl-stc1 {         /* Smart card I/F 1 */
-				pingrp = "stc1_pins";
+				abilis,function = "stc1";
 			};
 			/* Port 6 */
 			pctl_tsout_p: pctl-tsout-p {   /* Parallel TS-out */
-				pingrp = "mop_pins";
+				abilis,function = "mop";
 			};
 			pctl_tsout_s0: pctl-tsout-s0 { /* Serial TS-out 0 */
-				pingrp = "mos0_pins";
+				abilis,function = "mos0";
 			};
 			pctl_tsout_s1: pctl-tsout-s1 { /* Serial TS-out 1 */
-				pingrp = "mos1_pins";
+				abilis,function = "mos1";
 			};
 			pctl_tsout_s2: pctl-tsout-s2 { /* Serial TS-out 2 */
-				pingrp = "mos2_pins";
+				abilis,function = "mos2";
 			};
 			pctl_tsout_s3: pctl-tsout-s3 { /* Serial TS-out 3 */
-				pingrp = "mos3_pins";
+				abilis,function = "mos3";
 			};
 			/* Port 7 */
 			pctl_uart0: pctl-uart0 {       /* UART 0 */
-				pingrp = "uart0_pins";
+				abilis,function = "uart0";
 			};
 			pctl_uart1: pctl-uart1 {       /* UART 1 */
-				pingrp = "uart1_pins";
+				abilis,function = "uart1";
 			};
 			pctl_gpio_l: pctl-gpio-l {     /* GPIO bank L */
-				pingrp = "gpiol_pins";
+				abilis,function = "gpiol";
 			};
 			pctl_gpio_m: pctl-gpio-m {     /* GPIO bank M */
-				pingrp = "gpiom_pins";
+				abilis,function = "gpiom";
 			};
 			/* Port 8 */
 			pctl_spi3: pctl-spi3 {
-				pingrp = "spi3_pins";
+				abilis,function = "spi3";
 			};
 			pctl_jtag: pctl-jtag {
-				pingrp = "jtag_pins";
+				abilis,function = "jtag";
 			};
 			/* Port 9 */
 			pctl_spi1: pctl-spi1 {
-				pingrp = "spi1_pins";
+				abilis,function = "spi1";
 			};
 			pctl_gpio_n: pctl-gpio-n {
-				pingrp = "gpion_pins";
+				abilis,function = "gpion";
 			};
 			/* Unmuxed GPIOs */
 			pctl_gpio_b: pctl-gpio-b {
-				pingrp = "gpiob_pins";
+				abilis,function = "gpiob";
 			};
 			pctl_gpio_d: pctl-gpio-d {
-				pingrp = "gpiod_pins";
+				abilis,function = "gpiod";
 			};
 			pctl_gpio_f: pctl-gpio-f {
-				pingrp = "gpiof_pins";
+				abilis,function = "gpiof";
 			};
 			pctl_gpio_h: pctl-gpio-h {
-				pingrp = "gpioh_pins";
+				abilis,function = "gpioh";
 			};
 			pctl_gpio_i: pctl-gpio-i {
-				pingrp = "gpioi_pins";
+				abilis,function = "gpioi";
 			};
 		};
 
@@ -181,9 +181,10 @@
 			interrupts = <27 2>;
 			reg = <0xFF140000 0x1000>;
 			gpio-controller;
-			#gpio-cells = <1>;
-			gpio-base  = <0>;
-			gpio-pins = <&pctl_gpio_a>;
+			#gpio-cells = <2>;
+			abilis,ngpio = <3>;
+			gpio-ranges = <&iomux 0 0 0>;
+			gpio-ranges-group-names = "gpioa";
 		};
 		gpiob: gpio@FF141000 {
 			compatible = "abilis,tb10x-gpio";
@@ -193,9 +194,10 @@
 			interrupts = <27 2>;
 			reg = <0xFF141000 0x1000>;
 			gpio-controller;
-			#gpio-cells = <1>;
-			gpio-base  = <3>;
-			gpio-pins = <&pctl_gpio_b>;
+			#gpio-cells = <2>;
+			abilis,ngpio = <2>;
+			gpio-ranges = <&iomux 0 0 0>;
+			gpio-ranges-group-names = "gpiob";
 		};
 		gpioc: gpio@FF142000 {
 			compatible = "abilis,tb10x-gpio";
@@ -205,9 +207,10 @@
 			interrupts = <27 2>;
 			reg = <0xFF142000 0x1000>;
 			gpio-controller;
-			#gpio-cells = <1>;
-			gpio-base  = <5>;
-			gpio-pins = <&pctl_gpio_c>;
+			#gpio-cells = <2>;
+			abilis,ngpio = <3>;
+			gpio-ranges = <&iomux 0 0 0>;
+			gpio-ranges-group-names = "gpioc";
 		};
 		gpiod: gpio@FF143000 {
 			compatible = "abilis,tb10x-gpio";
@@ -217,9 +220,10 @@
 			interrupts = <27 2>;
 			reg = <0xFF143000 0x1000>;
 			gpio-controller;
-			#gpio-cells = <1>;
-			gpio-base  = <8>;
-			gpio-pins = <&pctl_gpio_d>;
+			#gpio-cells = <2>;
+			abilis,ngpio = <2>;
+			gpio-ranges = <&iomux 0 0 0>;
+			gpio-ranges-group-names = "gpiod";
 		};
 		gpioe: gpio@FF144000 {
 			compatible = "abilis,tb10x-gpio";
@@ -229,9 +233,10 @@
 			interrupts = <27 2>;
 			reg = <0xFF144000 0x1000>;
 			gpio-controller;
-			#gpio-cells = <1>;
-			gpio-base  = <10>;
-			gpio-pins = <&pctl_gpio_e>;
+			#gpio-cells = <2>;
+			abilis,ngpio = <3>;
+			gpio-ranges = <&iomux 0 0 0>;
+			gpio-ranges-group-names = "gpioe";
 		};
 		gpiof: gpio@FF145000 {
 			compatible = "abilis,tb10x-gpio";
@@ -241,9 +246,10 @@
 			interrupts = <27 2>;
 			reg = <0xFF145000 0x1000>;
 			gpio-controller;
-			#gpio-cells = <1>;
-			gpio-base  = <13>;
-			gpio-pins = <&pctl_gpio_f>;
+			#gpio-cells = <2>;
+			abilis,ngpio = <2>;
+			gpio-ranges = <&iomux 0 0 0>;
+			gpio-ranges-group-names = "gpiof";
 		};
 		gpiog: gpio@FF146000 {
 			compatible = "abilis,tb10x-gpio";
@@ -253,9 +259,10 @@
 			interrupts = <27 2>;
 			reg = <0xFF146000 0x1000>;
 			gpio-controller;
-			#gpio-cells = <1>;
-			gpio-base  = <15>;
-			gpio-pins = <&pctl_gpio_g>;
+			#gpio-cells = <2>;
+			abilis,ngpio = <3>;
+			gpio-ranges = <&iomux 0 0 0>;
+			gpio-ranges-group-names = "gpiog";
 		};
 		gpioh: gpio@FF147000 {
 			compatible = "abilis,tb10x-gpio";
@@ -265,9 +272,10 @@
 			interrupts = <27 2>;
 			reg = <0xFF147000 0x1000>;
 			gpio-controller;
-			#gpio-cells = <1>;
-			gpio-base  = <18>;
-			gpio-pins = <&pctl_gpio_h>;
+			#gpio-cells = <2>;
+			abilis,ngpio = <2>;
+			gpio-ranges = <&iomux 0 0 0>;
+			gpio-ranges-group-names = "gpioh";
 		};
 		gpioi: gpio@FF148000 {
 			compatible = "abilis,tb10x-gpio";
@@ -277,9 +285,10 @@
 			interrupts = <27 2>;
 			reg = <0xFF148000 0x1000>;
 			gpio-controller;
-			#gpio-cells = <1>;
-			gpio-base  = <20>;
-			gpio-pins = <&pctl_gpio_i>;
+			#gpio-cells = <2>;
+			abilis,ngpio = <12>;
+			gpio-ranges = <&iomux 0 0 0>;
+			gpio-ranges-group-names = "gpioi";
 		};
 		gpioj: gpio@FF149000 {
 			compatible = "abilis,tb10x-gpio";
@@ -289,9 +298,10 @@
 			interrupts = <27 2>;
 			reg = <0xFF149000 0x1000>;
 			gpio-controller;
-			#gpio-cells = <1>;
-			gpio-base  = <32>;
-			gpio-pins = <&pctl_gpio_j>;
+			#gpio-cells = <2>;
+			abilis,ngpio = <32>;
+			gpio-ranges = <&iomux 0 0 0>;
+			gpio-ranges-group-names = "gpioj";
 		};
 		gpiok: gpio@FF14a000 {
 			compatible = "abilis,tb10x-gpio";
@@ -301,9 +311,10 @@
 			interrupts = <27 2>;
 			reg = <0xFF14A000 0x1000>;
 			gpio-controller;
-			#gpio-cells = <1>;
-			gpio-base  = <64>;
-			gpio-pins = <&pctl_gpio_k>;
+			#gpio-cells = <2>;
+			abilis,ngpio = <22>;
+			gpio-ranges = <&iomux 0 0 0>;
+			gpio-ranges-group-names = "gpiok";
 		};
 		gpiol: gpio@FF14b000 {
 			compatible = "abilis,tb10x-gpio";
@@ -313,9 +324,10 @@
 			interrupts = <27 2>;
 			reg = <0xFF14B000 0x1000>;
 			gpio-controller;
-			#gpio-cells = <1>;
-			gpio-base  = <86>;
-			gpio-pins = <&pctl_gpio_l>;
+			#gpio-cells = <2>;
+			abilis,ngpio = <4>;
+			gpio-ranges = <&iomux 0 0 0>;
+			gpio-ranges-group-names = "gpiol";
 		};
 		gpiom: gpio@FF14c000 {
 			compatible = "abilis,tb10x-gpio";
@@ -325,9 +337,10 @@
 			interrupts = <27 2>;
 			reg = <0xFF14C000 0x1000>;
 			gpio-controller;
-			#gpio-cells = <1>;
-			gpio-base  = <90>;
-			gpio-pins = <&pctl_gpio_m>;
+			#gpio-cells = <2>;
+			abilis,ngpio = <4>;
+			gpio-ranges = <&iomux 0 0 0>;
+			gpio-ranges-group-names = "gpiom";
 		};
 		gpion: gpio@FF14d000 {
 			compatible = "abilis,tb10x-gpio";
@@ -337,9 +350,10 @@
 			interrupts = <27 2>;
 			reg = <0xFF14D000 0x1000>;
 			gpio-controller;
-			#gpio-cells = <1>;
-			gpio-base  = <94>;
-			gpio-pins = <&pctl_gpio_n>;
+			#gpio-cells = <2>;
+			abilis,ngpio = <5>;
+			gpio-ranges = <&iomux 0 0 0>;
+			gpio-ranges-group-names = "gpion";
 		};
 	};
 };
diff --git a/arch/arc/boot/dts/abilis_tb101_dvk.dts b/arch/arc/boot/dts/abilis_tb101_dvk.dts
index b204657993aa..1cf51c280f28 100644
--- a/arch/arc/boot/dts/abilis_tb101_dvk.dts
+++ b/arch/arc/boot/dts/abilis_tb101_dvk.dts
@@ -64,62 +64,62 @@
 			compatible = "gpio-leds";
 			power {
 				label = "Power";
-				gpios = <&gpioi 0>;
+				gpios = <&gpioi 0 0>;
 				linux,default-trigger = "default-on";
 			};
 			heartbeat {
 				label = "Heartbeat";
-				gpios = <&gpioi 1>;
+				gpios = <&gpioi 1 0>;
 				linux,default-trigger = "heartbeat";
 			};
 			led2 {
 				label = "LED2";
-				gpios = <&gpioi 2>;
+				gpios = <&gpioi 2 0>;
 				default-state = "off";
 			};
 			led3 {
 				label = "LED3";
-				gpios = <&gpioi 3>;
+				gpios = <&gpioi 3 0>;
 				default-state = "off";
 			};
 			led4 {
 				label = "LED4";
-				gpios = <&gpioi 4>;
+				gpios = <&gpioi 4 0>;
 				default-state = "off";
 			};
 			led5 {
 				label = "LED5";
-				gpios = <&gpioi 5>;
+				gpios = <&gpioi 5 0>;
 				default-state = "off";
 			};
 			led6 {
 				label = "LED6";
-				gpios = <&gpioi 6>;
+				gpios = <&gpioi 6 0>;
 				default-state = "off";
 			};
 			led7 {
 				label = "LED7";
-				gpios = <&gpioi 7>;
+				gpios = <&gpioi 7 0>;
 				default-state = "off";
 			};
 			led8 {
 				label = "LED8";
-				gpios = <&gpioi 8>;
+				gpios = <&gpioi 8 0>;
 				default-state = "off";
 			};
 			led9 {
 				label = "LED9";
-				gpios = <&gpioi 9>;
+				gpios = <&gpioi 9 0>;
 				default-state = "off";
 			};
 			led10 {
 				label = "LED10";
-				gpios = <&gpioi 10>;
+				gpios = <&gpioi 10 0>;
 				default-state = "off";
 			};
 			led11 {
 				label = "LED11";
-				gpios = <&gpioi 11>;
+				gpios = <&gpioi 11 0>;
 				default-state = "off";
 			};
 		};
diff --git a/arch/arc/boot/dts/abilis_tb10x.dtsi b/arch/arc/boot/dts/abilis_tb10x.dtsi
index edf56f4749e1..a098d7c05e96 100644
--- a/arch/arc/boot/dts/abilis_tb10x.dtsi
+++ b/arch/arc/boot/dts/abilis_tb10x.dtsi
@@ -62,9 +62,8 @@
 		};
 
 		iomux: iomux@FF10601c {
-			#address-cells = <1>;
-			#size-cells = <1>;
 			compatible = "abilis,tb10x-iomux";
+			#gpio-range-cells = <3>;
 			reg = <0xFF10601c 0x4>;
 		};
 
diff --git a/arch/arc/boot/dts/angel4.dts b/arch/arc/boot/dts/angel4.dts
index 4fb2d6f655bd..bcf662d21a57 100644
--- a/arch/arc/boot/dts/angel4.dts
+++ b/arch/arc/boot/dts/angel4.dts
@@ -67,5 +67,9 @@
 				reg = <1>;
 			};
 		};
+
+		arcpmu0: pmu {
+			compatible = "snps,arc700-pmu";
+		};
 	};
 };
diff --git a/arch/arc/configs/fpga_noramfs_defconfig b/arch/arc/configs/fpga_noramfs_defconfig
new file mode 100644
index 000000000000..5276a52f6a2f
--- /dev/null
+++ b/arch/arc/configs/fpga_noramfs_defconfig
@@ -0,0 +1,64 @@
+CONFIG_CROSS_COMPILE="arc-linux-uclibc-"
+# CONFIG_LOCALVERSION_AUTO is not set
+CONFIG_DEFAULT_HOSTNAME="ARCLinux"
+# CONFIG_SWAP is not set
+CONFIG_HIGH_RES_TIMERS=y
+CONFIG_IKCONFIG=y
+CONFIG_IKCONFIG_PROC=y
+CONFIG_NAMESPACES=y
+# CONFIG_UTS_NS is not set
+# CONFIG_PID_NS is not set
+CONFIG_BLK_DEV_INITRD=y
+CONFIG_KALLSYMS_ALL=y
+CONFIG_EMBEDDED=y
+# CONFIG_SLUB_DEBUG is not set
+# CONFIG_COMPAT_BRK is not set
+CONFIG_KPROBES=y
+CONFIG_MODULES=y
+# CONFIG_LBDAF is not set
+# CONFIG_BLK_DEV_BSG is not set
+# CONFIG_IOSCHED_DEADLINE is not set
+# CONFIG_IOSCHED_CFQ is not set
+CONFIG_ARC_PLAT_FPGA_LEGACY=y
+CONFIG_ARC_BOARD_ML509=y
+# CONFIG_ARC_HAS_RTSC is not set
+CONFIG_ARC_BUILTIN_DTB_NAME="angel4"
+CONFIG_PREEMPT=y
+# CONFIG_COMPACTION is not set
+# CONFIG_CROSS_MEMORY_ATTACH is not set
+CONFIG_NET=y
+CONFIG_PACKET=y
+CONFIG_UNIX=y
+CONFIG_UNIX_DIAG=y
+CONFIG_NET_KEY=y
+CONFIG_INET=y
+# CONFIG_IPV6 is not set
+# CONFIG_STANDALONE is not set
+# CONFIG_PREVENT_FIRMWARE_BUILD is not set
+# CONFIG_FIRMWARE_IN_KERNEL is not set
+# CONFIG_BLK_DEV is not set
+CONFIG_NETDEVICES=y
+CONFIG_ARC_EMAC=y
+CONFIG_LXT_PHY=y
+# CONFIG_INPUT_MOUSEDEV_PSAUX is not set
+# CONFIG_INPUT_KEYBOARD is not set
+# CONFIG_INPUT_MOUSE is not set
+# CONFIG_SERIO is not set
+# CONFIG_LEGACY_PTYS is not set
+# CONFIG_DEVKMEM is not set
+CONFIG_SERIAL_ARC=y
+CONFIG_SERIAL_ARC_CONSOLE=y
+# CONFIG_HW_RANDOM is not set
+# CONFIG_HWMON is not set
+# CONFIG_VGA_CONSOLE is not set
+# CONFIG_HID is not set
+# CONFIG_USB_SUPPORT is not set
+# CONFIG_IOMMU_SUPPORT is not set
+CONFIG_EXT2_FS=y
+CONFIG_EXT2_FS_XATTR=y
+CONFIG_TMPFS=y
+# CONFIG_MISC_FILESYSTEMS is not set
+CONFIG_NFS_FS=y
+# CONFIG_ENABLE_WARN_DEPRECATED is not set
+# CONFIG_ENABLE_MUST_CHECK is not set
+CONFIG_XZ_DEC=y
diff --git a/arch/arc/include/asm/perf_event.h b/arch/arc/include/asm/perf_event.h
index 115ad96480e6..cbf755e32a03 100644
--- a/arch/arc/include/asm/perf_event.h
+++ b/arch/arc/include/asm/perf_event.h
@@ -1,5 +1,7 @@
 /*
- * Copyright (C) 2011-2012 Synopsys, Inc. (www.synopsys.com)
+ * Linux performance counter support for ARC
+ *
+ * Copyright (C) 2011-2013 Synopsys, Inc. (www.synopsys.com)
  *
  * This program is free software; you can redistribute it and/or modify
  * it under the terms of the GNU General Public License version 2 as
@@ -10,4 +12,204 @@
 #ifndef __ASM_PERF_EVENT_H
 #define __ASM_PERF_EVENT_H
 
+/* real maximum varies per CPU, this is the maximum supported by the driver */
+#define ARC_PMU_MAX_HWEVENTS	64
+
+#define ARC_REG_CC_BUILD	0xF6
+#define ARC_REG_CC_INDEX	0x240
+#define ARC_REG_CC_NAME0	0x241
+#define ARC_REG_CC_NAME1	0x242
+
+#define ARC_REG_PCT_BUILD	0xF5
+#define ARC_REG_PCT_COUNTL	0x250
+#define ARC_REG_PCT_COUNTH	0x251
+#define ARC_REG_PCT_SNAPL	0x252
+#define ARC_REG_PCT_SNAPH	0x253
+#define ARC_REG_PCT_CONFIG	0x254
+#define ARC_REG_PCT_CONTROL	0x255
+#define ARC_REG_PCT_INDEX	0x256
+
+#define ARC_REG_PCT_CONTROL_CC	(1 << 16)	/* clear counts */
+#define ARC_REG_PCT_CONTROL_SN	(1 << 17)	/* snapshot */
+
+struct arc_reg_pct_build {
+#ifdef CONFIG_CPU_BIG_ENDIAN
+	unsigned int m:8, c:8, r:6, s:2, v:8;
+#else
+	unsigned int v:8, s:2, r:6, c:8, m:8;
+#endif
+};
+
+struct arc_reg_cc_build {
+#ifdef CONFIG_CPU_BIG_ENDIAN
+	unsigned int c:16, r:8, v:8;
+#else
+	unsigned int v:8, r:8, c:16;
+#endif
+};
+
+#define PERF_COUNT_ARC_DCLM	(PERF_COUNT_HW_MAX + 0)
+#define PERF_COUNT_ARC_DCSM	(PERF_COUNT_HW_MAX + 1)
+#define PERF_COUNT_ARC_ICM	(PERF_COUNT_HW_MAX + 2)
+#define PERF_COUNT_ARC_BPOK	(PERF_COUNT_HW_MAX + 3)
+#define PERF_COUNT_ARC_EDTLB	(PERF_COUNT_HW_MAX + 4)
+#define PERF_COUNT_ARC_EITLB	(PERF_COUNT_HW_MAX + 5)
+#define PERF_COUNT_ARC_HW_MAX	(PERF_COUNT_HW_MAX + 6)
+
+/*
+ * The "generalized" performance events seem to really be a copy
+ * of the available events on x86 processors; the mapping to ARC
+ * events is not always possible 1-to-1. Fortunately, there doesn't
+ * seem to be an exact definition for these events, so we can cheat
+ * a bit where necessary.
+ *
+ * In particular, the following PERF events may behave a bit differently
+ * compared to other architectures:
+ *
+ * PERF_COUNT_HW_CPU_CYCLES
+ *	Cycles not in halted state
+ *
+ * PERF_COUNT_HW_REF_CPU_CYCLES
+ *	Reference cycles not in halted state, same as PERF_COUNT_HW_CPU_CYCLES
+ *	for now as we don't do Dynamic Voltage/Frequency Scaling (yet)
+ *
+ * PERF_COUNT_HW_BUS_CYCLES
+ *	Unclear what this means, Intel uses 0x013c, which according to
+ *	their datasheet means "unhalted reference cycles". It sounds similar
+ *	to PERF_COUNT_HW_REF_CPU_CYCLES, and we use the same counter for it.
+ *
+ * PERF_COUNT_HW_STALLED_CYCLES_BACKEND
+ * PERF_COUNT_HW_STALLED_CYCLES_FRONTEND
+ *	The ARC 700 can either measure stalls per pipeline stage, or all stalls
+ *	combined; for now we assign all stalls to STALLED_CYCLES_BACKEND
+ *	and all pipeline flushes (e.g. caused by mispredicts, etc.) to
+ *	STALLED_CYCLES_FRONTEND.
+ *
+ *	We could start multiple performance counters and combine everything
+ *	afterwards, but that makes it complicated.
+ *
+ *	Note that I$ cache misses aren't counted by either of the two!
+ */
+
+static const char * const arc_pmu_ev_hw_map[] = {
+	[PERF_COUNT_HW_CPU_CYCLES] = "crun",
+	[PERF_COUNT_HW_REF_CPU_CYCLES] = "crun",
+	[PERF_COUNT_HW_BUS_CYCLES] = "crun",
+	[PERF_COUNT_HW_INSTRUCTIONS] = "iall",
+	[PERF_COUNT_HW_BRANCH_MISSES] = "bpfail",
+	[PERF_COUNT_HW_BRANCH_INSTRUCTIONS] = "ijmp",
+	[PERF_COUNT_HW_STALLED_CYCLES_FRONTEND] = "bflush",
+	[PERF_COUNT_HW_STALLED_CYCLES_BACKEND] = "bstall",
+	[PERF_COUNT_ARC_DCLM] = "dclm",
+	[PERF_COUNT_ARC_DCSM] = "dcsm",
+	[PERF_COUNT_ARC_ICM] = "icm",
+	[PERF_COUNT_ARC_BPOK] = "bpok",
+	[PERF_COUNT_ARC_EDTLB] = "edtlb",
+	[PERF_COUNT_ARC_EITLB] = "eitlb",
+};
+
+#define C(_x)			PERF_COUNT_HW_CACHE_##_x
+#define CACHE_OP_UNSUPPORTED	0xffff
+
+static const unsigned arc_pmu_cache_map[C(MAX)][C(OP_MAX)][C(RESULT_MAX)] = {
+	[C(L1D)] = {
+		[C(OP_READ)] = {
+			[C(RESULT_ACCESS)]	= CACHE_OP_UNSUPPORTED,
+			[C(RESULT_MISS)]	= PERF_COUNT_ARC_DCLM,
+		},
+		[C(OP_WRITE)] = {
+			[C(RESULT_ACCESS)]	= CACHE_OP_UNSUPPORTED,
+			[C(RESULT_MISS)]	= PERF_COUNT_ARC_DCSM,
+		},
+		[C(OP_PREFETCH)] = {
+			[C(RESULT_ACCESS)]	= CACHE_OP_UNSUPPORTED,
+			[C(RESULT_MISS)]	= CACHE_OP_UNSUPPORTED,
+		},
+	},
+	[C(L1I)] = {
+		[C(OP_READ)] = {
+			[C(RESULT_ACCESS)]	= CACHE_OP_UNSUPPORTED,
+			[C(RESULT_MISS)]	= PERF_COUNT_ARC_ICM,
+		},
+		[C(OP_WRITE)] = {
+			[C(RESULT_ACCESS)]	= CACHE_OP_UNSUPPORTED,
+			[C(RESULT_MISS)]	= CACHE_OP_UNSUPPORTED,
+		},
+		[C(OP_PREFETCH)] = {
+			[C(RESULT_ACCESS)]	= CACHE_OP_UNSUPPORTED,
+			[C(RESULT_MISS)]	= CACHE_OP_UNSUPPORTED,
+		},
+	},
+	[C(LL)] = {
+		[C(OP_READ)] = {
+			[C(RESULT_ACCESS)]	= CACHE_OP_UNSUPPORTED,
+			[C(RESULT_MISS)]	= CACHE_OP_UNSUPPORTED,
+		},
+		[C(OP_WRITE)] = {
+			[C(RESULT_ACCESS)]	= CACHE_OP_UNSUPPORTED,
+			[C(RESULT_MISS)]	= CACHE_OP_UNSUPPORTED,
+		},
+		[C(OP_PREFETCH)] = {
+			[C(RESULT_ACCESS)]	= CACHE_OP_UNSUPPORTED,
+			[C(RESULT_MISS)]	= CACHE_OP_UNSUPPORTED,
+		},
+	},
+	[C(DTLB)] = {
+		[C(OP_READ)] = {
+			[C(RESULT_ACCESS)]	= CACHE_OP_UNSUPPORTED,
+			[C(RESULT_MISS)]	= PERF_COUNT_ARC_EDTLB,
+		},
+		[C(OP_WRITE)] = {
+			[C(RESULT_ACCESS)]	= CACHE_OP_UNSUPPORTED,
+			[C(RESULT_MISS)]	= CACHE_OP_UNSUPPORTED,
+		},
+		[C(OP_PREFETCH)] = {
+			[C(RESULT_ACCESS)]	= CACHE_OP_UNSUPPORTED,
+			[C(RESULT_MISS)]	= CACHE_OP_UNSUPPORTED,
+		},
+	},
+	[C(ITLB)] = {
+		[C(OP_READ)] = {
+			[C(RESULT_ACCESS)]	= CACHE_OP_UNSUPPORTED,
+			[C(RESULT_MISS)]	= PERF_COUNT_ARC_EITLB,
+		},
+		[C(OP_WRITE)] = {
+			[C(RESULT_ACCESS)]	= CACHE_OP_UNSUPPORTED,
+			[C(RESULT_MISS)]	= CACHE_OP_UNSUPPORTED,
+		},
+		[C(OP_PREFETCH)] = {
+			[C(RESULT_ACCESS)]	= CACHE_OP_UNSUPPORTED,
+			[C(RESULT_MISS)]	= CACHE_OP_UNSUPPORTED,
+		},
+	},
+	[C(BPU)] = {
+		[C(OP_READ)] = {
+			[C(RESULT_ACCESS)] = PERF_COUNT_HW_BRANCH_INSTRUCTIONS,
+			[C(RESULT_MISS)]	= PERF_COUNT_HW_BRANCH_MISSES,
+		},
+		[C(OP_WRITE)] = {
+			[C(RESULT_ACCESS)]	= CACHE_OP_UNSUPPORTED,
+			[C(RESULT_MISS)]	= CACHE_OP_UNSUPPORTED,
+		},
+		[C(OP_PREFETCH)] = {
+			[C(RESULT_ACCESS)]	= CACHE_OP_UNSUPPORTED,
+			[C(RESULT_MISS)]	= CACHE_OP_UNSUPPORTED,
+		},
+	},
+	[C(NODE)] = {
+		[C(OP_READ)] = {
+			[C(RESULT_ACCESS)]	= CACHE_OP_UNSUPPORTED,
+			[C(RESULT_MISS)]	= CACHE_OP_UNSUPPORTED,
+		},
+		[C(OP_WRITE)] = {
+			[C(RESULT_ACCESS)]	= CACHE_OP_UNSUPPORTED,
+			[C(RESULT_MISS)]	= CACHE_OP_UNSUPPORTED,
+		},
+		[C(OP_PREFETCH)] = {
+			[C(RESULT_ACCESS)]	= CACHE_OP_UNSUPPORTED,
+			[C(RESULT_MISS)]	= CACHE_OP_UNSUPPORTED,
+		},
+	},
+};
+
 #endif /* __ASM_PERF_EVENT_H */
diff --git a/arch/arc/include/asm/pgalloc.h b/arch/arc/include/asm/pgalloc.h
index 36a9f20c21a3..81208bfd9dcb 100644
--- a/arch/arc/include/asm/pgalloc.h
+++ b/arch/arc/include/asm/pgalloc.h
@@ -105,11 +105,16 @@ static inline pgtable_t
 pte_alloc_one(struct mm_struct *mm, unsigned long address)
 {
 	pgtable_t pte_pg;
+	struct page *page;
 
 	pte_pg = __get_free_pages(GFP_KERNEL | __GFP_REPEAT, __get_order_pte());
-	if (pte_pg) {
-		memzero((void *)pte_pg, PTRS_PER_PTE * 4);
-		pgtable_page_ctor(virt_to_page(pte_pg));
+	if (!pte_pg)
+		return 0;
+	memzero((void *)pte_pg, PTRS_PER_PTE * 4);
+	page = virt_to_page(pte_pg);
+	if (!pgtable_page_ctor(page)) {
+		__free_page(page);
+		return 0;
 	}
 
 	return pte_pg;
diff --git a/arch/arc/include/asm/thread_info.h b/arch/arc/include/asm/thread_info.h
index 2d50a4cdd7f3..45be21672011 100644
--- a/arch/arc/include/asm/thread_info.h
+++ b/arch/arc/include/asm/thread_info.h
@@ -80,8 +80,6 @@ static inline __attribute_const__ struct thread_info *current_thread_info(void)
 
 #endif /* !__ASSEMBLY__ */
 
-#define PREEMPT_ACTIVE      0x10000000
-
 /*
  * thread information flags
  * - these are process state flags that various assembly files may need to
diff --git a/arch/arc/kernel/Makefile b/arch/arc/kernel/Makefile
index c242ef07ba70..8004b4fa6461 100644
--- a/arch/arc/kernel/Makefile
+++ b/arch/arc/kernel/Makefile
@@ -19,6 +19,7 @@ obj-$(CONFIG_KPROBES)      		+= kprobes.o
 obj-$(CONFIG_ARC_MISALIGN_ACCESS) 	+= unaligned.o
 obj-$(CONFIG_KGDB)			+= kgdb.o
 obj-$(CONFIG_ARC_METAWARE_HLINK)	+= arc_hostlink.o
+obj-$(CONFIG_PERF_EVENTS)		+= perf_event.o
 
 obj-$(CONFIG_ARC_FPU_SAVE_RESTORE)	+= fpu.o
 CFLAGS_fpu.o   += -mdpfp
diff --git a/arch/arc/kernel/kprobes.c b/arch/arc/kernel/kprobes.c
index eb1c2ee5eaf0..42b05046fad9 100644
--- a/arch/arc/kernel/kprobes.c
+++ b/arch/arc/kernel/kprobes.c
@@ -327,7 +327,7 @@ int __kprobes kprobe_fault_handler(struct pt_regs *regs, unsigned long trapnr)
 		 */
 
 		/* We increment the nmissed count for accounting,
-		 * we can also use npre/npostfault count for accouting
+		 * we can also use npre/npostfault count for accounting
 		 * these specific fault cases.
 		 */
 		kprobes_inc_nmissed_count(cur);
diff --git a/arch/arc/kernel/perf_event.c b/arch/arc/kernel/perf_event.c
new file mode 100644
index 000000000000..e46d81f70979
--- /dev/null
+++ b/arch/arc/kernel/perf_event.c
@@ -0,0 +1,326 @@
+/*
+ * Linux performance counter support for ARC700 series
+ *
+ * Copyright (C) 2013 Synopsys, Inc. (www.synopsys.com)
+ *
+ * This code is inspired by the perf support of various other architectures.
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ *
+ */
+#include <linux/errno.h>
+#include <linux/module.h>
+#include <linux/of.h>
+#include <linux/perf_event.h>
+#include <linux/platform_device.h>
+#include <asm/arcregs.h>
+
+struct arc_pmu {
+	struct pmu	pmu;
+	int		counter_size;	/* in bits */
+	int		n_counters;
+	unsigned long	used_mask[BITS_TO_LONGS(ARC_PMU_MAX_HWEVENTS)];
+	int		ev_hw_idx[PERF_COUNT_ARC_HW_MAX];
+};
+
+/* read counter #idx; note that counter# != event# on ARC! */
+static uint64_t arc_pmu_read_counter(int idx)
+{
+	uint32_t tmp;
+	uint64_t result;
+
+	/*
+	 * ARC supports making 'snapshots' of the counters, so we don't
+	 * need to care about counters wrapping to 0 underneath our feet
+	 */
+	write_aux_reg(ARC_REG_PCT_INDEX, idx);
+	tmp = read_aux_reg(ARC_REG_PCT_CONTROL);
+	write_aux_reg(ARC_REG_PCT_CONTROL, tmp | ARC_REG_PCT_CONTROL_SN);
+	result = (uint64_t) (read_aux_reg(ARC_REG_PCT_SNAPH)) << 32;
+	result |= read_aux_reg(ARC_REG_PCT_SNAPL);
+
+	return result;
+}
+
+static void arc_perf_event_update(struct perf_event *event,
+				  struct hw_perf_event *hwc, int idx)
+{
+	struct arc_pmu *arc_pmu = container_of(event->pmu, struct arc_pmu, pmu);
+	uint64_t prev_raw_count, new_raw_count;
+	int64_t delta;
+
+	do {
+		prev_raw_count = local64_read(&hwc->prev_count);
+		new_raw_count = arc_pmu_read_counter(idx);
+	} while (local64_cmpxchg(&hwc->prev_count, prev_raw_count,
+				 new_raw_count) != prev_raw_count);
+
+	delta = (new_raw_count - prev_raw_count) &
+		((1ULL << arc_pmu->counter_size) - 1ULL);
+
+	local64_add(delta, &event->count);
+	local64_sub(delta, &hwc->period_left);
+}
+
+static void arc_pmu_read(struct perf_event *event)
+{
+	arc_perf_event_update(event, &event->hw, event->hw.idx);
+}
+
+static int arc_pmu_cache_event(u64 config)
+{
+	unsigned int cache_type, cache_op, cache_result;
+	int ret;
+
+	cache_type	= (config >>  0) & 0xff;
+	cache_op	= (config >>  8) & 0xff;
+	cache_result	= (config >> 16) & 0xff;
+	if (cache_type >= PERF_COUNT_HW_CACHE_MAX)
+		return -EINVAL;
+	if (cache_type >= PERF_COUNT_HW_CACHE_OP_MAX)
+		return -EINVAL;
+	if (cache_type >= PERF_COUNT_HW_CACHE_RESULT_MAX)
+		return -EINVAL;
+
+	ret = arc_pmu_cache_map[cache_type][cache_op][cache_result];
+
+	if (ret == CACHE_OP_UNSUPPORTED)
+		return -ENOENT;
+
+	return ret;
+}
+
+/* initializes hw_perf_event structure if event is supported */
+static int arc_pmu_event_init(struct perf_event *event)
+{
+	struct arc_pmu *arc_pmu = container_of(event->pmu, struct arc_pmu, pmu);
+	struct hw_perf_event *hwc = &event->hw;
+	int ret;
+
+	/* ARC 700 PMU does not support sampling events */
+	if (is_sampling_event(event))
+		return -ENOENT;
+
+	switch (event->attr.type) {
+	case PERF_TYPE_HARDWARE:
+		if (event->attr.config >= PERF_COUNT_HW_MAX)
+			return -ENOENT;
+		if (arc_pmu->ev_hw_idx[event->attr.config] < 0)
+			return -ENOENT;
+		hwc->config = arc_pmu->ev_hw_idx[event->attr.config];
+		pr_debug("initializing event %d with cfg %d\n",
+			 (int) event->attr.config, (int) hwc->config);
+		return 0;
+	case PERF_TYPE_HW_CACHE:
+		ret = arc_pmu_cache_event(event->attr.config);
+		if (ret < 0)
+			return ret;
+		hwc->config = arc_pmu->ev_hw_idx[ret];
+		return 0;
+	default:
+		return -ENOENT;
+	}
+}
+
+/* starts all counters */
+static void arc_pmu_enable(struct pmu *pmu)
+{
+	uint32_t tmp;
+	tmp = read_aux_reg(ARC_REG_PCT_CONTROL);
+	write_aux_reg(ARC_REG_PCT_CONTROL, (tmp & 0xffff0000) | 0x1);
+}
+
+/* stops all counters */
+static void arc_pmu_disable(struct pmu *pmu)
+{
+	uint32_t tmp;
+	tmp = read_aux_reg(ARC_REG_PCT_CONTROL);
+	write_aux_reg(ARC_REG_PCT_CONTROL, (tmp & 0xffff0000) | 0x0);
+}
+
+/*
+ * Assigns hardware counter to hardware condition.
+ * Note that there is no separate start/stop mechanism;
+ * stopping is achieved by assigning the 'never' condition
+ */
+static void arc_pmu_start(struct perf_event *event, int flags)
+{
+	struct hw_perf_event *hwc = &event->hw;
+	int idx = hwc->idx;
+
+	if (WARN_ON_ONCE(idx == -1))
+		return;
+
+	if (flags & PERF_EF_RELOAD)
+		WARN_ON_ONCE(!(event->hw.state & PERF_HES_UPTODATE));
+
+	event->hw.state = 0;
+
+	/* enable ARC pmu here */
+	write_aux_reg(ARC_REG_PCT_INDEX, idx);
+	write_aux_reg(ARC_REG_PCT_CONFIG, hwc->config);
+}
+
+static void arc_pmu_stop(struct perf_event *event, int flags)
+{
+	struct hw_perf_event *hwc = &event->hw;
+	int idx = hwc->idx;
+
+	if (!(event->hw.state & PERF_HES_STOPPED)) {
+		/* stop ARC pmu here */
+		write_aux_reg(ARC_REG_PCT_INDEX, idx);
+
+		/* condition code #0 is always "never" */
+		write_aux_reg(ARC_REG_PCT_CONFIG, 0);
+
+		event->hw.state |= PERF_HES_STOPPED;
+	}
+
+	if ((flags & PERF_EF_UPDATE) &&
+	    !(event->hw.state & PERF_HES_UPTODATE)) {
+		arc_perf_event_update(event, &event->hw, idx);
+		event->hw.state |= PERF_HES_UPTODATE;
+	}
+}
+
+static void arc_pmu_del(struct perf_event *event, int flags)
+{
+	struct arc_pmu *arc_pmu = container_of(event->pmu, struct arc_pmu, pmu);
+
+	arc_pmu_stop(event, PERF_EF_UPDATE);
+	__clear_bit(event->hw.idx, arc_pmu->used_mask);
+
+	perf_event_update_userpage(event);
+}
+
+/* allocate hardware counter and optionally start counting */
+static int arc_pmu_add(struct perf_event *event, int flags)
+{
+	struct arc_pmu *arc_pmu = container_of(event->pmu, struct arc_pmu, pmu);
+	struct hw_perf_event *hwc = &event->hw;
+	int idx = hwc->idx;
+
+	if (__test_and_set_bit(idx, arc_pmu->used_mask)) {
+		idx = find_first_zero_bit(arc_pmu->used_mask,
+					  arc_pmu->n_counters);
+		if (idx == arc_pmu->n_counters)
+			return -EAGAIN;
+
+		__set_bit(idx, arc_pmu->used_mask);
+		hwc->idx = idx;
+	}
+
+	write_aux_reg(ARC_REG_PCT_INDEX, idx);
+	write_aux_reg(ARC_REG_PCT_CONFIG, 0);
+	write_aux_reg(ARC_REG_PCT_COUNTL, 0);
+	write_aux_reg(ARC_REG_PCT_COUNTH, 0);
+	local64_set(&hwc->prev_count, 0);
+
+	hwc->state = PERF_HES_UPTODATE | PERF_HES_STOPPED;
+	if (flags & PERF_EF_START)
+		arc_pmu_start(event, PERF_EF_RELOAD);
+
+	perf_event_update_userpage(event);
+
+	return 0;
+}
+
+static int arc_pmu_device_probe(struct platform_device *pdev)
+{
+	struct arc_pmu *arc_pmu;
+	struct arc_reg_pct_build pct_bcr;
+	struct arc_reg_cc_build cc_bcr;
+	int i, j, ret;
+
+	union cc_name {
+		struct {
+			uint32_t word0, word1;
+			char sentinel;
+		} indiv;
+		char str[9];
+	} cc_name;
+
+
+	READ_BCR(ARC_REG_PCT_BUILD, pct_bcr);
+	if (!pct_bcr.v) {
+		pr_err("This core does not have performance counters!\n");
+		return -ENODEV;
+	}
+
+	arc_pmu = devm_kzalloc(&pdev->dev, sizeof(struct arc_pmu),
+			       GFP_KERNEL);
+	if (!arc_pmu)
+		return -ENOMEM;
+
+	arc_pmu->n_counters = pct_bcr.c;
+	BUG_ON(arc_pmu->n_counters > ARC_PMU_MAX_HWEVENTS);
+
+	arc_pmu->counter_size = 32 + (pct_bcr.s << 4);
+	pr_info("ARC PMU found with %d counters of size %d bits\n",
+		arc_pmu->n_counters, arc_pmu->counter_size);
+
+	READ_BCR(ARC_REG_CC_BUILD, cc_bcr);
+
+	if (!cc_bcr.v)
+		pr_err("Strange! Performance counters exist, but no countable conditions?\n");
+
+	pr_info("ARC PMU has %d countable conditions\n", cc_bcr.c);
+
+	cc_name.str[8] = 0;
+	for (i = 0; i < PERF_COUNT_HW_MAX; i++)
+		arc_pmu->ev_hw_idx[i] = -1;
+
+	for (j = 0; j < cc_bcr.c; j++) {
+		write_aux_reg(ARC_REG_CC_INDEX, j);
+		cc_name.indiv.word0 = read_aux_reg(ARC_REG_CC_NAME0);
+		cc_name.indiv.word1 = read_aux_reg(ARC_REG_CC_NAME1);
+		for (i = 0; i < ARRAY_SIZE(arc_pmu_ev_hw_map); i++) {
+			if (arc_pmu_ev_hw_map[i] &&
+			    !strcmp(arc_pmu_ev_hw_map[i], cc_name.str) &&
+			    strlen(arc_pmu_ev_hw_map[i])) {
+				pr_debug("mapping %d to idx %d with name %s\n",
+					 i, j, cc_name.str);
+				arc_pmu->ev_hw_idx[i] = j;
+			}
+		}
+	}
+
+	arc_pmu->pmu = (struct pmu) {
+		.pmu_enable	= arc_pmu_enable,
+		.pmu_disable	= arc_pmu_disable,
+		.event_init	= arc_pmu_event_init,
+		.add		= arc_pmu_add,
+		.del		= arc_pmu_del,
+		.start		= arc_pmu_start,
+		.stop		= arc_pmu_stop,
+		.read		= arc_pmu_read,
+	};
+
+	ret = perf_pmu_register(&arc_pmu->pmu, pdev->name, PERF_TYPE_RAW);
+
+	return ret;
+}
+
+#ifdef CONFIG_OF
+static const struct of_device_id arc_pmu_match[] = {
+	{ .compatible = "snps,arc700-pmu" },
+	{},
+};
+MODULE_DEVICE_TABLE(of, arc_pmu_match);
+#endif
+
+static struct platform_driver arc_pmu_driver = {
+	.driver	= {
+		.name		= "arc700-pmu",
+		.of_match_table = of_match_ptr(arc_pmu_match),
+	},
+	.probe		= arc_pmu_device_probe,
+};
+
+module_platform_driver(arc_pmu_driver);
+
+MODULE_LICENSE("GPL");
+MODULE_AUTHOR("Mischa Jonker <mjonker@synopsys.com>");
+MODULE_DESCRIPTION("ARC PMU driver");
diff --git a/arch/arc/plat-tb10x/Kconfig b/arch/arc/plat-tb10x/Kconfig
index 1ab386bb5da8..6994c188dc88 100644
--- a/arch/arc/plat-tb10x/Kconfig
+++ b/arch/arc/plat-tb10x/Kconfig
@@ -20,8 +20,10 @@ menuconfig ARC_PLAT_TB10X
 	bool "Abilis TB10x"
 	select COMMON_CLK
 	select PINCTRL
+	select PINCTRL_TB10X
 	select PINMUX
 	select ARCH_REQUIRE_GPIOLIB
+	select GPIO_TB10X
 	select TB10X_IRQC
 	help
 	  Support for platforms based on the TB10x home media gateway SOC by
diff --git a/arch/arm/Kconfig b/arch/arm/Kconfig
index 603d661b445d..214b698cefea 100644
--- a/arch/arm/Kconfig
+++ b/arch/arm/Kconfig
@@ -5,6 +5,7 @@ config ARM
 	select ARCH_HAS_ATOMIC64_DEC_IF_POSITIVE
 	select ARCH_HAS_TICK_BROADCAST if GENERIC_CLOCKEVENTS_BROADCAST
 	select ARCH_HAVE_CUSTOM_GPIO_H
+	select ARCH_MIGHT_HAVE_PC_PARPORT
 	select ARCH_USE_CMPXCHG_LOCKREF
 	select ARCH_WANT_IPC_PARSE_VERSION
 	select BUILDTIME_EXTABLE_SORT if MMU
@@ -1432,7 +1433,6 @@ config SMP
 	depends on GENERIC_CLOCKEVENTS
 	depends on HAVE_SMP
 	depends on MMU || ARM_MPU
-	select USE_GENERIC_SMP_HELPERS
 	help
 	  This enables support for systems with more than one CPU. If you have
 	  a system with only one CPU, like most personal computers, say N. If
@@ -1863,6 +1863,12 @@ config CC_STACKPROTECTOR
 	  neutralized via a kernel panic.
 	  This feature requires gcc version 4.2 or above.
 
+config SWIOTLB
+	def_bool y
+
+config IOMMU_HELPER
+	def_bool SWIOTLB
+
 config XEN_DOM0
 	def_bool y
 	depends on XEN
@@ -1873,6 +1879,7 @@ config XEN
 	depends on CPU_V7 && !CPU_V6
 	depends on !GENERIC_ATOMIC64
 	select ARM_PSCI
+	select SWIOTLB_XEN
 	help
 	  Say Y if you want to run Linux in a Virtual Machine on Xen on ARM.
 
diff --git a/arch/arm/boot/dts/am335x-evm.dts b/arch/arm/boot/dts/am335x-evm.dts
index 987429436171..7e6c64ed966d 100644
--- a/arch/arm/boot/dts/am335x-evm.dts
+++ b/arch/arm/boot/dts/am335x-evm.dts
@@ -630,7 +630,7 @@
 	tsc {
 		ti,wires = <4>;
 		ti,x-plate-resistance = <200>;
-		ti,coordiante-readouts = <5>;
+		ti,coordinate-readouts = <5>;
 		ti,wire-config = <0x00 0x11 0x22 0x33>;
 	};
 
diff --git a/arch/arm/boot/dts/am335x-evmsk.dts b/arch/arm/boot/dts/am335x-evmsk.dts
index 03febf85fd2f..4718ec4a4dbf 100644
--- a/arch/arm/boot/dts/am335x-evmsk.dts
+++ b/arch/arm/boot/dts/am335x-evmsk.dts
@@ -485,3 +485,13 @@
 		tx-num-evt = <1>;
 		rx-num-evt = <1>;
 };
+
+&tscadc {
+	status = "okay";
+	tsc {
+		ti,wires = <4>;
+		ti,x-plate-resistance = <200>;
+		ti,coordinate-readouts = <5>;
+		ti,wire-config = <0x00 0x11 0x22 0x33>;
+	};
+};
diff --git a/arch/arm/boot/dts/ecx-2000.dts b/arch/arm/boot/dts/ecx-2000.dts
index 139b40cc3a23..2ccbb57fbfa8 100644
--- a/arch/arm/boot/dts/ecx-2000.dts
+++ b/arch/arm/boot/dts/ecx-2000.dts
@@ -85,6 +85,12 @@
 				<1 10 0xf08>;
 		};
 
+		memory-controller@fff00000 {
+			compatible = "calxeda,ecx-2000-ddr-ctrl";
+			reg = <0xfff00000 0x1000>;
+			interrupts = <0 91 4>;
+		};
+
 		intc: interrupt-controller@fff11000 {
 			compatible = "arm,cortex-a15-gic";
 			#interrupt-cells = <3>;
diff --git a/arch/arm/boot/dts/ecx-common.dtsi b/arch/arm/boot/dts/ecx-common.dtsi
index bc22557d7a6a..b90045a8f8e3 100644
--- a/arch/arm/boot/dts/ecx-common.dtsi
+++ b/arch/arm/boot/dts/ecx-common.dtsi
@@ -53,12 +53,6 @@
 			status = "disabled";
 		};
 
-		memory-controller@fff00000 {
-			compatible = "calxeda,hb-ddr-ctrl";
-			reg = <0xfff00000 0x1000>;
-			interrupts = <0 91 4>;
-		};
-
 		ipc@fff20000 {
 			compatible = "arm,pl320", "arm,primecell";
 			reg = <0xfff20000 0x1000>;
diff --git a/arch/arm/boot/dts/highbank.dts b/arch/arm/boot/dts/highbank.dts
index 6aad34ad9517..ed14aeac0566 100644
--- a/arch/arm/boot/dts/highbank.dts
+++ b/arch/arm/boot/dts/highbank.dts
@@ -86,6 +86,12 @@
 	soc {
 		ranges = <0x00000000 0x00000000 0xffffffff>;
 
+		memory-controller@fff00000 {
+			compatible = "calxeda,hb-ddr-ctrl";
+			reg = <0xfff00000 0x1000>;
+			interrupts = <0 91 4>;
+		};
+
 		timer@fff10600 {
 			compatible = "arm,cortex-a9-twd-timer";
 			reg = <0xfff10600 0x20>;
diff --git a/arch/arm/boot/dts/imx51.dtsi b/arch/arm/boot/dts/imx51.dtsi
index f4dcff3a9969..4bcdd3ad15e5 100644
--- a/arch/arm/boot/dts/imx51.dtsi
+++ b/arch/arm/boot/dts/imx51.dtsi
@@ -190,7 +190,7 @@
 
 			usbphy0: usbphy@0 {
 				compatible = "usb-nop-xceiv";
-				clocks = <&clks 124>;
+				clocks = <&clks 75>;
 				clock-names = "main_clk";
 				status = "okay";
 			};
diff --git a/arch/arm/boot/dts/twl4030.dtsi b/arch/arm/boot/dts/twl4030.dtsi
index fb1b2ec8eaa9..4217096ee677 100644
--- a/arch/arm/boot/dts/twl4030.dtsi
+++ b/arch/arm/boot/dts/twl4030.dtsi
@@ -19,6 +19,12 @@
 		interrupts = <11>;
 	};
 
+	charger: bci {
+		compatible = "ti,twl4030-bci";
+		interrupts = <9>, <2>;
+		bci3v1-supply = <&vusb3v1>;
+	};
+
 	watchdog {
 		compatible = "ti,twl4030-wdt";
 	};
diff --git a/arch/arm/common/edma.c b/arch/arm/common/edma.c
index 8e1a0245907f..41bca32409fc 100644
--- a/arch/arm/common/edma.c
+++ b/arch/arm/common/edma.c
@@ -404,7 +404,7 @@ static irqreturn_t dma_irq_handler(int irq, void *data)
 					BIT(slot));
 			if (edma_cc[ctlr]->intr_data[channel].callback)
 				edma_cc[ctlr]->intr_data[channel].callback(
-					channel, DMA_COMPLETE,
+					channel, EDMA_DMA_COMPLETE,
 					edma_cc[ctlr]->intr_data[channel].data);
 		}
 	} while (sh_ipr);
@@ -459,7 +459,7 @@ static irqreturn_t dma_ccerr_handler(int irq, void *data)
 								callback) {
 						edma_cc[ctlr]->intr_data[k].
 						callback(k,
-						DMA_CC_ERROR,
+						EDMA_DMA_CC_ERROR,
 						edma_cc[ctlr]->intr_data
 						[k].data);
 					}
diff --git a/arch/arm/configs/prima2_defconfig b/arch/arm/configs/prima2_defconfig
index 002a1ceadceb..23591dba47a0 100644
--- a/arch/arm/configs/prima2_defconfig
+++ b/arch/arm/configs/prima2_defconfig
@@ -39,6 +39,7 @@ CONFIG_SPI=y
 CONFIG_SPI_SIRF=y
 CONFIG_SPI_SPIDEV=y
 # CONFIG_HWMON is not set
+CONFIG_WATCHDOG=y
 CONFIG_USB_GADGET=y
 CONFIG_USB_MASS_STORAGE=m
 CONFIG_MMC=y
diff --git a/arch/arm/configs/vt8500_v6_v7_defconfig b/arch/arm/configs/vt8500_v6_v7_defconfig
new file mode 100644
index 000000000000..f0520176acd0
--- /dev/null
+++ b/arch/arm/configs/vt8500_v6_v7_defconfig
@@ -0,0 +1,90 @@
+CONFIG_IRQ_DOMAIN_DEBUG=y
+CONFIG_NO_HZ=y
+CONFIG_HIGH_RES_TIMERS=y
+CONFIG_BLK_DEV_INITRD=y
+CONFIG_ARCH_MULTI_V6=y
+CONFIG_ARCH_WM8750=y
+CONFIG_ARCH_WM8850=y
+CONFIG_ARM_ERRATA_720789=y
+CONFIG_ARM_ERRATA_754322=y
+CONFIG_ARM_ERRATA_775420=y
+CONFIG_HAVE_ARM_ARCH_TIMER=y
+CONFIG_AEABI=y
+CONFIG_HIGHMEM=y
+CONFIG_HIGHPTE=y
+CONFIG_ARM_APPENDED_DTB=y
+CONFIG_ARM_ATAG_DTB_COMPAT=y
+CONFIG_VFP=y
+CONFIG_NEON=y
+CONFIG_PM_RUNTIME=y
+CONFIG_NET=y
+CONFIG_UNIX=y
+CONFIG_INET=y
+CONFIG_IP_PNP=y
+CONFIG_IP_PNP_DHCP=y
+CONFIG_DEVTMPFS=y
+CONFIG_DEVTMPFS_MOUNT=y
+CONFIG_PROC_DEVICETREE=y
+CONFIG_EEPROM_93CX6=y
+CONFIG_SCSI=y
+CONFIG_BLK_DEV_SD=y
+CONFIG_NETDEVICES=y
+# CONFIG_NET_CADENCE is not set
+# CONFIG_NET_VENDOR_BROADCOM is not set
+# CONFIG_NET_VENDOR_CIRRUS is not set
+# CONFIG_NET_VENDOR_FARADAY is not set
+# CONFIG_NET_VENDOR_INTEL is not set
+# CONFIG_NET_VENDOR_MARVELL is not set
+# CONFIG_NET_VENDOR_MICREL is not set
+# CONFIG_NET_VENDOR_NATSEMI is not set
+# CONFIG_NET_VENDOR_SEEQ is not set
+# CONFIG_NET_VENDOR_SMSC is not set
+# CONFIG_NET_VENDOR_STMICRO is not set
+CONFIG_VIA_VELOCITY=y
+# CONFIG_NET_VENDOR_WIZNET is not set
+CONFIG_PHYLIB=y
+CONFIG_INPUT_MATRIXKMAP=y
+CONFIG_SERIAL_VT8500=y
+CONFIG_SERIAL_VT8500_CONSOLE=y
+CONFIG_I2C=y
+CONFIG_I2C_WMT=y
+CONFIG_PINCTRL_SINGLE=y
+CONFIG_PINCTRL_WM8750=y
+CONFIG_GPIO_GENERIC_PLATFORM=y
+CONFIG_POWER_SUPPLY=y
+CONFIG_POWER_RESET=y
+CONFIG_MFD_SYSCON=y
+CONFIG_REGULATOR=y
+CONFIG_REGULATOR_FIXED_VOLTAGE=y
+CONFIG_REGULATOR_GPIO=y
+CONFIG_USB=y
+CONFIG_USB_EHCI_HCD=y
+CONFIG_USB_EHCI_HCD_PLATFORM=y
+CONFIG_USB_UHCI_HCD=y
+CONFIG_USB_STORAGE=y
+CONFIG_NOP_USB_XCEIV=y
+CONFIG_USB_GPIO_VBUS=y
+CONFIG_USB_ULPI=y
+CONFIG_MMC=y
+CONFIG_MMC_DEBUG=y
+CONFIG_NEW_LEDS=y
+CONFIG_LEDS_CLASS=y
+CONFIG_LEDS_TRIGGERS=y
+CONFIG_RTC_CLASS=y
+CONFIG_RTC_DRV_VT8500=y
+CONFIG_DMADEVICES=y
+CONFIG_COMMON_CLK_DEBUG=y
+# CONFIG_IOMMU_SUPPORT is not set
+CONFIG_PWM=y
+CONFIG_PWM_VT8500=y
+CONFIG_RESET_CONTROLLER=y
+CONFIG_GENERIC_PHY=y
+CONFIG_EXT4_FS=y
+CONFIG_TMPFS=y
+CONFIG_NFS_FS=y
+CONFIG_NFS_V3_ACL=y
+CONFIG_NFS_V4=y
+CONFIG_ROOT_NFS=y
+CONFIG_PRINTK_TIME=y
+CONFIG_DEBUG_KERNEL=y
+CONFIG_LOCKUP_DETECTOR=y
diff --git a/arch/arm/include/asm/dma-mapping.h b/arch/arm/include/asm/dma-mapping.h
index 863cd84eb1a2..e701a4d9aa59 100644
--- a/arch/arm/include/asm/dma-mapping.h
+++ b/arch/arm/include/asm/dma-mapping.h
@@ -11,17 +11,28 @@
 #include <asm-generic/dma-coherent.h>
 #include <asm/memory.h>
 
+#include <xen/xen.h>
+#include <asm/xen/hypervisor.h>
+
 #define DMA_ERROR_CODE	(~0)
 extern struct dma_map_ops arm_dma_ops;
 extern struct dma_map_ops arm_coherent_dma_ops;
 
-static inline struct dma_map_ops *get_dma_ops(struct device *dev)
+static inline struct dma_map_ops *__generic_dma_ops(struct device *dev)
 {
 	if (dev && dev->archdata.dma_ops)
 		return dev->archdata.dma_ops;
 	return &arm_dma_ops;
 }
 
+static inline struct dma_map_ops *get_dma_ops(struct device *dev)
+{
+	if (xen_initial_domain())
+		return xen_dma_ops;
+	else
+		return __generic_dma_ops(dev);
+}
+
 static inline void set_dma_ops(struct device *dev, struct dma_map_ops *ops)
 {
 	BUG_ON(!dev);
@@ -94,6 +105,39 @@ static inline unsigned long dma_max_pfn(struct device *dev)
 }
 #define dma_max_pfn(dev) dma_max_pfn(dev)
 
+static inline dma_addr_t phys_to_dma(struct device *dev, phys_addr_t paddr)
+{
+	unsigned int offset = paddr & ~PAGE_MASK;
+	return pfn_to_dma(dev, __phys_to_pfn(paddr)) + offset;
+}
+
+static inline phys_addr_t dma_to_phys(struct device *dev, dma_addr_t dev_addr)
+{
+	unsigned int offset = dev_addr & ~PAGE_MASK;
+	return __pfn_to_phys(dma_to_pfn(dev, dev_addr)) + offset;
+}
+
+static inline bool dma_capable(struct device *dev, dma_addr_t addr, size_t size)
+{
+	u64 limit, mask;
+
+	if (!dev->dma_mask)
+		return 0;
+
+	mask = *dev->dma_mask;
+
+	limit = (mask + 1) & ~mask;
+	if (limit && size > limit)
+		return 0;
+
+	if ((addr | (addr + size - 1)) & ~mask)
+		return 0;
+
+	return 1;
+}
+
+static inline void dma_mark_clean(void *addr, size_t size) { }
+
 /*
  * DMA errors are defined by all-bits-set in the DMA address.
  */
diff --git a/arch/arm/include/asm/hardware/iop3xx-adma.h b/arch/arm/include/asm/hardware/iop3xx-adma.h
index 9b28f1243bdc..240b29ef17db 100644
--- a/arch/arm/include/asm/hardware/iop3xx-adma.h
+++ b/arch/arm/include/asm/hardware/iop3xx-adma.h
@@ -393,36 +393,6 @@ static inline int iop_chan_zero_sum_slot_count(size_t len, int src_cnt,
 	return slot_cnt;
 }
 
-static inline int iop_desc_is_pq(struct iop_adma_desc_slot *desc)
-{
-	return 0;
-}
-
-static inline u32 iop_desc_get_dest_addr(struct iop_adma_desc_slot *desc,
-					struct iop_adma_chan *chan)
-{
-	union iop3xx_desc hw_desc = { .ptr = desc->hw_desc, };
-
-	switch (chan->device->id) {
-	case DMA0_ID:
-	case DMA1_ID:
-		return hw_desc.dma->dest_addr;
-	case AAU_ID:
-		return hw_desc.aau->dest_addr;
-	default:
-		BUG();
-	}
-	return 0;
-}
-
-
-static inline u32 iop_desc_get_qdest_addr(struct iop_adma_desc_slot *desc,
-					  struct iop_adma_chan *chan)
-{
-	BUG();
-	return 0;
-}
-
 static inline u32 iop_desc_get_byte_count(struct iop_adma_desc_slot *desc,
 					struct iop_adma_chan *chan)
 {
diff --git a/arch/arm/include/asm/hardware/iop_adma.h b/arch/arm/include/asm/hardware/iop_adma.h
index 122f86d8c991..250760e08103 100644
--- a/arch/arm/include/asm/hardware/iop_adma.h
+++ b/arch/arm/include/asm/hardware/iop_adma.h
@@ -82,8 +82,6 @@ struct iop_adma_chan {
  * @slot_cnt: total slots used in an transaction (group of operations)
  * @slots_per_op: number of slots per operation
  * @idx: pool index
- * @unmap_src_cnt: number of xor sources
- * @unmap_len: transaction bytecount
  * @tx_list: list of descriptors that are associated with one operation
  * @async_tx: support for the async_tx api
  * @group_list: list of slots that make up a multi-descriptor transaction
@@ -99,8 +97,6 @@ struct iop_adma_desc_slot {
 	u16 slot_cnt;
 	u16 slots_per_op;
 	u16 idx;
-	u16 unmap_src_cnt;
-	size_t unmap_len;
 	struct list_head tx_list;
 	struct dma_async_tx_descriptor async_tx;
 	union {
diff --git a/arch/arm/include/asm/io.h b/arch/arm/include/asm/io.h
index d070741b2b37..3c597c222ef2 100644
--- a/arch/arm/include/asm/io.h
+++ b/arch/arm/include/asm/io.h
@@ -24,9 +24,11 @@
 #ifdef __KERNEL__
 
 #include <linux/types.h>
+#include <linux/blk_types.h>
 #include <asm/byteorder.h>
 #include <asm/memory.h>
 #include <asm-generic/pci_iomap.h>
+#include <xen/xen.h>
 
 /*
  * ISA I/O bus memory addresses are 1:1 with the physical address.
@@ -372,6 +374,13 @@ extern void pci_iounmap(struct pci_dev *dev, void __iomem *addr);
 #define BIOVEC_MERGEABLE(vec1, vec2)	\
 	((bvec_to_phys((vec1)) + (vec1)->bv_len) == bvec_to_phys((vec2)))
 
+struct bio_vec;
+extern bool xen_biovec_phys_mergeable(const struct bio_vec *vec1,
+				      const struct bio_vec *vec2);
+#define BIOVEC_PHYS_MERGEABLE(vec1, vec2)				\
+	(__BIOVEC_PHYS_MERGEABLE(vec1, vec2) &&				\
+	 (!xen_domain() || xen_biovec_phys_mergeable(vec1, vec2)))
+
 #ifdef CONFIG_MMU
 #define ARCH_HAS_VALID_PHYS_ADDR_RANGE
 extern int valid_phys_addr_range(phys_addr_t addr, size_t size);
diff --git a/arch/arm/include/asm/kvm_arm.h b/arch/arm/include/asm/kvm_arm.h
index 64e96960de29..1d3153c7eb41 100644
--- a/arch/arm/include/asm/kvm_arm.h
+++ b/arch/arm/include/asm/kvm_arm.h
@@ -57,6 +57,7 @@
  * TSC:		Trap SMC
  * TSW:		Trap cache operations by set/way
  * TWI:		Trap WFI
+ * TWE:		Trap WFE
  * TIDCP:	Trap L2CTLR/L2ECTLR
  * BSU_IS:	Upgrade barriers to the inner shareable domain
  * FB:		Force broadcast of all maintainance operations
@@ -67,7 +68,7 @@
  */
 #define HCR_GUEST_MASK (HCR_TSC | HCR_TSW | HCR_TWI | HCR_VM | HCR_BSU_IS | \
 			HCR_FB | HCR_TAC | HCR_AMO | HCR_IMO | HCR_FMO | \
-			HCR_SWIO | HCR_TIDCP)
+			HCR_TWE | HCR_SWIO | HCR_TIDCP)
 #define HCR_VIRT_EXCP_MASK (HCR_VA | HCR_VI | HCR_VF)
 
 /* System Control Register (SCTLR) bits */
@@ -95,12 +96,12 @@
 #define TTBCR_IRGN1	(3 << 24)
 #define TTBCR_EPD1	(1 << 23)
 #define TTBCR_A1	(1 << 22)
-#define TTBCR_T1SZ	(3 << 16)
+#define TTBCR_T1SZ	(7 << 16)
 #define TTBCR_SH0	(3 << 12)
 #define TTBCR_ORGN0	(3 << 10)
 #define TTBCR_IRGN0	(3 << 8)
 #define TTBCR_EPD0	(1 << 7)
-#define TTBCR_T0SZ	3
+#define TTBCR_T0SZ	(7 << 0)
 #define HTCR_MASK	(TTBCR_T0SZ | TTBCR_IRGN0 | TTBCR_ORGN0 | TTBCR_SH0)
 
 /* Hyp System Trap Register */
@@ -208,6 +209,8 @@
 #define HSR_EC_DABT	(0x24)
 #define HSR_EC_DABT_HYP	(0x25)
 
+#define HSR_WFI_IS_WFE		(1U << 0)
+
 #define HSR_HVC_IMM_MASK	((1UL << 16) - 1)
 
 #define HSR_DABT_S1PTW		(1U << 7)
diff --git a/arch/arm/include/asm/kvm_asm.h b/arch/arm/include/asm/kvm_asm.h
index a2f43ddcc300..661da11f76f4 100644
--- a/arch/arm/include/asm/kvm_asm.h
+++ b/arch/arm/include/asm/kvm_asm.h
@@ -39,7 +39,7 @@
 #define c6_IFAR		17	/* Instruction Fault Address Register */
 #define c7_PAR		18	/* Physical Address Register */
 #define c7_PAR_high	19	/* PAR top 32 bits */
-#define c9_L2CTLR	20	/* Cortex A15 L2 Control Register */
+#define c9_L2CTLR	20	/* Cortex A15/A7 L2 Control Register */
 #define c10_PRRR	21	/* Primary Region Remap Register */
 #define c10_NMRR	22	/* Normal Memory Remap Register */
 #define c12_VBAR	23	/* Vector Base Address Register */
diff --git a/arch/arm/include/asm/kvm_emulate.h b/arch/arm/include/asm/kvm_emulate.h
index a464e8d7b6c5..0fa90c962ac8 100644
--- a/arch/arm/include/asm/kvm_emulate.h
+++ b/arch/arm/include/asm/kvm_emulate.h
@@ -157,4 +157,55 @@ static inline u32 kvm_vcpu_hvc_get_imm(struct kvm_vcpu *vcpu)
 	return kvm_vcpu_get_hsr(vcpu) & HSR_HVC_IMM_MASK;
 }
 
+static inline unsigned long kvm_vcpu_get_mpidr(struct kvm_vcpu *vcpu)
+{
+	return vcpu->arch.cp15[c0_MPIDR];
+}
+
+static inline void kvm_vcpu_set_be(struct kvm_vcpu *vcpu)
+{
+	*vcpu_cpsr(vcpu) |= PSR_E_BIT;
+}
+
+static inline bool kvm_vcpu_is_be(struct kvm_vcpu *vcpu)
+{
+	return !!(*vcpu_cpsr(vcpu) & PSR_E_BIT);
+}
+
+static inline unsigned long vcpu_data_guest_to_host(struct kvm_vcpu *vcpu,
+						    unsigned long data,
+						    unsigned int len)
+{
+	if (kvm_vcpu_is_be(vcpu)) {
+		switch (len) {
+		case 1:
+			return data & 0xff;
+		case 2:
+			return be16_to_cpu(data & 0xffff);
+		default:
+			return be32_to_cpu(data);
+		}
+	}
+
+	return data;		/* Leave LE untouched */
+}
+
+static inline unsigned long vcpu_data_host_to_guest(struct kvm_vcpu *vcpu,
+						    unsigned long data,
+						    unsigned int len)
+{
+	if (kvm_vcpu_is_be(vcpu)) {
+		switch (len) {
+		case 1:
+			return data & 0xff;
+		case 2:
+			return cpu_to_be16(data & 0xffff);
+		default:
+			return cpu_to_be32(data);
+		}
+	}
+
+	return data;		/* Leave LE untouched */
+}
+
 #endif /* __ARM_KVM_EMULATE_H__ */
diff --git a/arch/arm/include/asm/kvm_host.h b/arch/arm/include/asm/kvm_host.h
index 7d22517d8071..8a6f6db14ee4 100644
--- a/arch/arm/include/asm/kvm_host.h
+++ b/arch/arm/include/asm/kvm_host.h
@@ -38,11 +38,6 @@
 
 #define KVM_VCPU_MAX_FEATURES 1
 
-/* We don't currently support large pages. */
-#define KVM_HPAGE_GFN_SHIFT(x)	0
-#define KVM_NR_PAGE_SIZES	1
-#define KVM_PAGES_PER_HPAGE(x)	(1UL<<31)
-
 #include <kvm/arm_vgic.h>
 
 struct kvm_vcpu;
@@ -154,6 +149,7 @@ struct kvm_vcpu_stat {
 struct kvm_vcpu_init;
 int kvm_vcpu_set_target(struct kvm_vcpu *vcpu,
 			const struct kvm_vcpu_init *init);
+int kvm_vcpu_preferred_target(struct kvm_vcpu_init *init);
 unsigned long kvm_arm_num_regs(struct kvm_vcpu *vcpu);
 int kvm_arm_copy_reg_indices(struct kvm_vcpu *vcpu, u64 __user *indices);
 struct kvm_one_reg;
diff --git a/arch/arm/include/asm/kvm_mmu.h b/arch/arm/include/asm/kvm_mmu.h
index 9b28c41f4ba9..77de4a41cc50 100644
--- a/arch/arm/include/asm/kvm_mmu.h
+++ b/arch/arm/include/asm/kvm_mmu.h
@@ -62,6 +62,12 @@ phys_addr_t kvm_get_idmap_vector(void);
 int kvm_mmu_init(void);
 void kvm_clear_hyp_idmap(void);
 
+static inline void kvm_set_pmd(pmd_t *pmd, pmd_t new_pmd)
+{
+	*pmd = new_pmd;
+	flush_pmd_entry(pmd);
+}
+
 static inline void kvm_set_pte(pte_t *pte, pte_t new_pte)
 {
 	*pte = new_pte;
@@ -103,9 +109,15 @@ static inline void kvm_set_s2pte_writable(pte_t *pte)
 	pte_val(*pte) |= L_PTE_S2_RDWR;
 }
 
+static inline void kvm_set_s2pmd_writable(pmd_t *pmd)
+{
+	pmd_val(*pmd) |= L_PMD_S2_RDWR;
+}
+
 struct kvm;
 
-static inline void coherent_icache_guest_page(struct kvm *kvm, gfn_t gfn)
+static inline void coherent_icache_guest_page(struct kvm *kvm, hva_t hva,
+					      unsigned long size)
 {
 	/*
 	 * If we are going to insert an instruction page and the icache is
@@ -120,8 +132,7 @@ static inline void coherent_icache_guest_page(struct kvm *kvm, gfn_t gfn)
 	 * need any kind of flushing (DDI 0406C.b - Page B3-1392).
 	 */
 	if (icache_is_pipt()) {
-		unsigned long hva = gfn_to_hva(kvm, gfn);
-		__cpuc_coherent_user_range(hva, hva + PAGE_SIZE);
+		__cpuc_coherent_user_range(hva, hva + size);
 	} else if (!icache_is_vivt_asid_tagged()) {
 		/* any kind of VIPT cache */
 		__flush_icache_all();
diff --git a/arch/arm/include/asm/pgalloc.h b/arch/arm/include/asm/pgalloc.h
index 943504f53f57..78a779361682 100644
--- a/arch/arm/include/asm/pgalloc.h
+++ b/arch/arm/include/asm/pgalloc.h
@@ -102,12 +102,14 @@ pte_alloc_one(struct mm_struct *mm, unsigned long addr)
 #else
 	pte = alloc_pages(PGALLOC_GFP, 0);
 #endif
-	if (pte) {
-		if (!PageHighMem(pte))
-			clean_pte_table(page_address(pte));
-		pgtable_page_ctor(pte);
+	if (!pte)
+		return NULL;
+	if (!PageHighMem(pte))
+		clean_pte_table(page_address(pte));
+	if (!pgtable_page_ctor(pte)) {
+		__free_page(pte);
+		return NULL;
 	}
-
 	return pte;
 }
 
diff --git a/arch/arm/include/asm/pgtable-3level.h b/arch/arm/include/asm/pgtable-3level.h
index 39c54cfa03e9..4f9503908dca 100644
--- a/arch/arm/include/asm/pgtable-3level.h
+++ b/arch/arm/include/asm/pgtable-3level.h
@@ -126,6 +126,8 @@
 #define L_PTE_S2_RDONLY		 (_AT(pteval_t, 1) << 6)   /* HAP[1]   */
 #define L_PTE_S2_RDWR		 (_AT(pteval_t, 3) << 6)   /* HAP[2:1] */
 
+#define L_PMD_S2_RDWR		 (_AT(pmdval_t, 3) << 6)   /* HAP[2:1] */
+
 /*
  * Hyp-mode PL2 PTE definitions for LPAE.
  */
diff --git a/arch/arm/include/asm/thread_info.h b/arch/arm/include/asm/thread_info.h
index df5e13d64f2c..71a06b293489 100644
--- a/arch/arm/include/asm/thread_info.h
+++ b/arch/arm/include/asm/thread_info.h
@@ -141,12 +141,6 @@ extern int vfp_restore_user_hwstate(struct user_vfp __user *,
 #endif
 
 /*
- * We use bit 30 of the preempt_count to indicate that kernel
- * preemption is occurring.  See <asm/hardirq.h>.
- */
-#define PREEMPT_ACTIVE	0x40000000
-
-/*
  * thread information flags:
  *  TIF_SYSCALL_TRACE	- syscall trace active
  *  TIF_SYSCAL_AUDIT	- syscall auditing active
diff --git a/arch/arm/include/asm/xen/hypervisor.h b/arch/arm/include/asm/xen/hypervisor.h
index d7ab99a0c9eb..1317ee40f4df 100644
--- a/arch/arm/include/asm/xen/hypervisor.h
+++ b/arch/arm/include/asm/xen/hypervisor.h
@@ -16,4 +16,6 @@ static inline enum paravirt_lazy_mode paravirt_get_lazy_mode(void)
 	return PARAVIRT_LAZY_NONE;
 }
 
+extern struct dma_map_ops *xen_dma_ops;
+
 #endif /* _ASM_ARM_XEN_HYPERVISOR_H */
diff --git a/arch/arm/include/asm/xen/page-coherent.h b/arch/arm/include/asm/xen/page-coherent.h
new file mode 100644
index 000000000000..1109017499e5
--- /dev/null
+++ b/arch/arm/include/asm/xen/page-coherent.h
@@ -0,0 +1,50 @@
+#ifndef _ASM_ARM_XEN_PAGE_COHERENT_H
+#define _ASM_ARM_XEN_PAGE_COHERENT_H
+
+#include <asm/page.h>
+#include <linux/dma-attrs.h>
+#include <linux/dma-mapping.h>
+
+static inline void *xen_alloc_coherent_pages(struct device *hwdev, size_t size,
+		dma_addr_t *dma_handle, gfp_t flags,
+		struct dma_attrs *attrs)
+{
+	return __generic_dma_ops(hwdev)->alloc(hwdev, size, dma_handle, flags, attrs);
+}
+
+static inline void xen_free_coherent_pages(struct device *hwdev, size_t size,
+		void *cpu_addr, dma_addr_t dma_handle,
+		struct dma_attrs *attrs)
+{
+	__generic_dma_ops(hwdev)->free(hwdev, size, cpu_addr, dma_handle, attrs);
+}
+
+static inline void xen_dma_map_page(struct device *hwdev, struct page *page,
+	     unsigned long offset, size_t size, enum dma_data_direction dir,
+	     struct dma_attrs *attrs)
+{
+	__generic_dma_ops(hwdev)->map_page(hwdev, page, offset, size, dir, attrs);
+}
+
+static inline void xen_dma_unmap_page(struct device *hwdev, dma_addr_t handle,
+		size_t size, enum dma_data_direction dir,
+		struct dma_attrs *attrs)
+{
+	if (__generic_dma_ops(hwdev)->unmap_page)
+		__generic_dma_ops(hwdev)->unmap_page(hwdev, handle, size, dir, attrs);
+}
+
+static inline void xen_dma_sync_single_for_cpu(struct device *hwdev,
+		dma_addr_t handle, size_t size, enum dma_data_direction dir)
+{
+	if (__generic_dma_ops(hwdev)->sync_single_for_cpu)
+		__generic_dma_ops(hwdev)->sync_single_for_cpu(hwdev, handle, size, dir);
+}
+
+static inline void xen_dma_sync_single_for_device(struct device *hwdev,
+		dma_addr_t handle, size_t size, enum dma_data_direction dir)
+{
+	if (__generic_dma_ops(hwdev)->sync_single_for_device)
+		__generic_dma_ops(hwdev)->sync_single_for_device(hwdev, handle, size, dir);
+}
+#endif /* _ASM_ARM_XEN_PAGE_COHERENT_H */
diff --git a/arch/arm/include/asm/xen/page.h b/arch/arm/include/asm/xen/page.h
index 359a7b50b158..75579a9d6f76 100644
--- a/arch/arm/include/asm/xen/page.h
+++ b/arch/arm/include/asm/xen/page.h
@@ -6,12 +6,12 @@
 
 #include <linux/pfn.h>
 #include <linux/types.h>
+#include <linux/dma-mapping.h>
 
+#include <xen/xen.h>
 #include <xen/interface/grant_table.h>
 
-#define pfn_to_mfn(pfn)			(pfn)
 #define phys_to_machine_mapping_valid(pfn) (1)
-#define mfn_to_pfn(mfn)			(mfn)
 #define mfn_to_virt(m)			(__va(mfn_to_pfn(m) << PAGE_SHIFT))
 
 #define pte_mfn	    pte_pfn
@@ -32,6 +32,38 @@ typedef struct xpaddr {
 
 #define INVALID_P2M_ENTRY      (~0UL)
 
+unsigned long __pfn_to_mfn(unsigned long pfn);
+unsigned long __mfn_to_pfn(unsigned long mfn);
+extern struct rb_root phys_to_mach;
+
+static inline unsigned long pfn_to_mfn(unsigned long pfn)
+{
+	unsigned long mfn;
+
+	if (phys_to_mach.rb_node != NULL) {
+		mfn = __pfn_to_mfn(pfn);
+		if (mfn != INVALID_P2M_ENTRY)
+			return mfn;
+	}
+
+	return pfn;
+}
+
+static inline unsigned long mfn_to_pfn(unsigned long mfn)
+{
+	unsigned long pfn;
+
+	if (phys_to_mach.rb_node != NULL) {
+		pfn = __mfn_to_pfn(mfn);
+		if (pfn != INVALID_P2M_ENTRY)
+			return pfn;
+	}
+
+	return mfn;
+}
+
+#define mfn_to_local_pfn(mfn) mfn_to_pfn(mfn)
+
 static inline xmaddr_t phys_to_machine(xpaddr_t phys)
 {
 	unsigned offset = phys.paddr & ~PAGE_MASK;
@@ -76,11 +108,9 @@ static inline int m2p_remove_override(struct page *page, bool clear_pte)
 	return 0;
 }
 
-static inline bool __set_phys_to_machine(unsigned long pfn, unsigned long mfn)
-{
-	BUG_ON(pfn != mfn && mfn != INVALID_P2M_ENTRY);
-	return true;
-}
+bool __set_phys_to_machine(unsigned long pfn, unsigned long mfn);
+bool __set_phys_to_machine_multi(unsigned long pfn, unsigned long mfn,
+		unsigned long nr_pages);
 
 static inline bool set_phys_to_machine(unsigned long pfn, unsigned long mfn)
 {
diff --git a/arch/arm/include/uapi/asm/kvm.h b/arch/arm/include/uapi/asm/kvm.h
index c1ee007523d7..c498b60c0505 100644
--- a/arch/arm/include/uapi/asm/kvm.h
+++ b/arch/arm/include/uapi/asm/kvm.h
@@ -63,7 +63,8 @@ struct kvm_regs {
 
 /* Supported Processor Types */
 #define KVM_ARM_TARGET_CORTEX_A15	0
-#define KVM_ARM_NUM_TARGETS		1
+#define KVM_ARM_TARGET_CORTEX_A7	1
+#define KVM_ARM_NUM_TARGETS		2
 
 /* KVM_ARM_SET_DEVICE_ADDR ioctl id encoding */
 #define KVM_ARM_DEVICE_TYPE_SHIFT	0
diff --git a/arch/arm/kvm/Kconfig b/arch/arm/kvm/Kconfig
index ebf5015508b5..466bd299b1a8 100644
--- a/arch/arm/kvm/Kconfig
+++ b/arch/arm/kvm/Kconfig
@@ -20,6 +20,7 @@ config KVM
 	bool "Kernel-based Virtual Machine (KVM) support"
 	select PREEMPT_NOTIFIERS
 	select ANON_INODES
+	select HAVE_KVM_CPU_RELAX_INTERCEPT
 	select KVM_MMIO
 	select KVM_ARM_HOST
 	depends on ARM_VIRT_EXT && ARM_LPAE
diff --git a/arch/arm/kvm/Makefile b/arch/arm/kvm/Makefile
index d99bee4950e5..789bca9e64a7 100644
--- a/arch/arm/kvm/Makefile
+++ b/arch/arm/kvm/Makefile
@@ -19,6 +19,6 @@ kvm-arm-y = $(KVM)/kvm_main.o $(KVM)/coalesced_mmio.o
 
 obj-y += kvm-arm.o init.o interrupts.o
 obj-y += arm.o handle_exit.o guest.o mmu.o emulate.o reset.o
-obj-y += coproc.o coproc_a15.o mmio.o psci.o perf.o
+obj-y += coproc.o coproc_a15.o coproc_a7.o mmio.o psci.o perf.o
 obj-$(CONFIG_KVM_ARM_VGIC) += $(KVM)/arm/vgic.o
 obj-$(CONFIG_KVM_ARM_TIMER) += $(KVM)/arm/arch_timer.o
diff --git a/arch/arm/kvm/arm.c b/arch/arm/kvm/arm.c
index aea7ccb8d397..2a700e00528d 100644
--- a/arch/arm/kvm/arm.c
+++ b/arch/arm/kvm/arm.c
@@ -152,12 +152,13 @@ int kvm_arch_vcpu_fault(struct kvm_vcpu *vcpu, struct vm_fault *vmf)
 	return VM_FAULT_SIGBUS;
 }
 
-void kvm_arch_free_memslot(struct kvm_memory_slot *free,
+void kvm_arch_free_memslot(struct kvm *kvm, struct kvm_memory_slot *free,
 			   struct kvm_memory_slot *dont)
 {
 }
 
-int kvm_arch_create_memslot(struct kvm_memory_slot *slot, unsigned long npages)
+int kvm_arch_create_memslot(struct kvm *kvm, struct kvm_memory_slot *slot,
+			    unsigned long npages)
 {
 	return 0;
 }
@@ -797,6 +798,19 @@ long kvm_arch_vm_ioctl(struct file *filp,
 			return -EFAULT;
 		return kvm_vm_ioctl_set_device_addr(kvm, &dev_addr);
 	}
+	case KVM_ARM_PREFERRED_TARGET: {
+		int err;
+		struct kvm_vcpu_init init;
+
+		err = kvm_vcpu_preferred_target(&init);
+		if (err)
+			return err;
+
+		if (copy_to_user(argp, &init, sizeof(init)))
+			return -EFAULT;
+
+		return 0;
+	}
 	default:
 		return -EINVAL;
 	}
diff --git a/arch/arm/kvm/coproc.c b/arch/arm/kvm/coproc.c
index db9cf692d4dd..78c0885d6501 100644
--- a/arch/arm/kvm/coproc.c
+++ b/arch/arm/kvm/coproc.c
@@ -71,6 +71,98 @@ int kvm_handle_cp14_access(struct kvm_vcpu *vcpu, struct kvm_run *run)
 	return 1;
 }
 
+static void reset_mpidr(struct kvm_vcpu *vcpu, const struct coproc_reg *r)
+{
+	/*
+	 * Compute guest MPIDR. We build a virtual cluster out of the
+	 * vcpu_id, but we read the 'U' bit from the underlying
+	 * hardware directly.
+	 */
+	vcpu->arch.cp15[c0_MPIDR] = ((read_cpuid_mpidr() & MPIDR_SMP_BITMASK) |
+				     ((vcpu->vcpu_id >> 2) << MPIDR_LEVEL_BITS) |
+				     (vcpu->vcpu_id & 3));
+}
+
+/* TRM entries A7:4.3.31 A15:4.3.28 - RO WI */
+static bool access_actlr(struct kvm_vcpu *vcpu,
+			 const struct coproc_params *p,
+			 const struct coproc_reg *r)
+{
+	if (p->is_write)
+		return ignore_write(vcpu, p);
+
+	*vcpu_reg(vcpu, p->Rt1) = vcpu->arch.cp15[c1_ACTLR];
+	return true;
+}
+
+/* TRM entries A7:4.3.56, A15:4.3.60 - R/O. */
+static bool access_cbar(struct kvm_vcpu *vcpu,
+			const struct coproc_params *p,
+			const struct coproc_reg *r)
+{
+	if (p->is_write)
+		return write_to_read_only(vcpu, p);
+	return read_zero(vcpu, p);
+}
+
+/* TRM entries A7:4.3.49, A15:4.3.48 - R/O WI */
+static bool access_l2ctlr(struct kvm_vcpu *vcpu,
+			  const struct coproc_params *p,
+			  const struct coproc_reg *r)
+{
+	if (p->is_write)
+		return ignore_write(vcpu, p);
+
+	*vcpu_reg(vcpu, p->Rt1) = vcpu->arch.cp15[c9_L2CTLR];
+	return true;
+}
+
+static void reset_l2ctlr(struct kvm_vcpu *vcpu, const struct coproc_reg *r)
+{
+	u32 l2ctlr, ncores;
+
+	asm volatile("mrc p15, 1, %0, c9, c0, 2\n" : "=r" (l2ctlr));
+	l2ctlr &= ~(3 << 24);
+	ncores = atomic_read(&vcpu->kvm->online_vcpus) - 1;
+	/* How many cores in the current cluster and the next ones */
+	ncores -= (vcpu->vcpu_id & ~3);
+	/* Cap it to the maximum number of cores in a single cluster */
+	ncores = min(ncores, 3U);
+	l2ctlr |= (ncores & 3) << 24;
+
+	vcpu->arch.cp15[c9_L2CTLR] = l2ctlr;
+}
+
+static void reset_actlr(struct kvm_vcpu *vcpu, const struct coproc_reg *r)
+{
+	u32 actlr;
+
+	/* ACTLR contains SMP bit: make sure you create all cpus first! */
+	asm volatile("mrc p15, 0, %0, c1, c0, 1\n" : "=r" (actlr));
+	/* Make the SMP bit consistent with the guest configuration */
+	if (atomic_read(&vcpu->kvm->online_vcpus) > 1)
+		actlr |= 1U << 6;
+	else
+		actlr &= ~(1U << 6);
+
+	vcpu->arch.cp15[c1_ACTLR] = actlr;
+}
+
+/*
+ * TRM entries: A7:4.3.50, A15:4.3.49
+ * R/O WI (even if NSACR.NS_L2ERR, a write of 1 is ignored).
+ */
+static bool access_l2ectlr(struct kvm_vcpu *vcpu,
+			   const struct coproc_params *p,
+			   const struct coproc_reg *r)
+{
+	if (p->is_write)
+		return ignore_write(vcpu, p);
+
+	*vcpu_reg(vcpu, p->Rt1) = 0;
+	return true;
+}
+
 /* See note at ARM ARM B1.14.4 */
 static bool access_dcsw(struct kvm_vcpu *vcpu,
 			const struct coproc_params *p,
@@ -153,10 +245,22 @@ static bool pm_fake(struct kvm_vcpu *vcpu,
  *            registers preceding 32-bit ones.
  */
 static const struct coproc_reg cp15_regs[] = {
+	/* MPIDR: we use VMPIDR for guest access. */
+	{ CRn( 0), CRm( 0), Op1( 0), Op2( 5), is32,
+			NULL, reset_mpidr, c0_MPIDR },
+
 	/* CSSELR: swapped by interrupt.S. */
 	{ CRn( 0), CRm( 0), Op1( 2), Op2( 0), is32,
 			NULL, reset_unknown, c0_CSSELR },
 
+	/* ACTLR: trapped by HCR.TAC bit. */
+	{ CRn( 1), CRm( 0), Op1( 0), Op2( 1), is32,
+			access_actlr, reset_actlr, c1_ACTLR },
+
+	/* CPACR: swapped by interrupt.S. */
+	{ CRn( 1), CRm( 0), Op1( 0), Op2( 2), is32,
+			NULL, reset_val, c1_CPACR, 0x00000000 },
+
 	/* TTBR0/TTBR1: swapped by interrupt.S. */
 	{ CRm64( 2), Op1( 0), is64, NULL, reset_unknown64, c2_TTBR0 },
 	{ CRm64( 2), Op1( 1), is64, NULL, reset_unknown64, c2_TTBR1 },
@@ -195,6 +299,13 @@ static const struct coproc_reg cp15_regs[] = {
 	{ CRn( 7), CRm(10), Op1( 0), Op2( 2), is32, access_dcsw},
 	{ CRn( 7), CRm(14), Op1( 0), Op2( 2), is32, access_dcsw},
 	/*
+	 * L2CTLR access (guest wants to know #CPUs).
+	 */
+	{ CRn( 9), CRm( 0), Op1( 1), Op2( 2), is32,
+			access_l2ctlr, reset_l2ctlr, c9_L2CTLR },
+	{ CRn( 9), CRm( 0), Op1( 1), Op2( 3), is32, access_l2ectlr},
+
+	/*
 	 * Dummy performance monitor implementation.
 	 */
 	{ CRn( 9), CRm(12), Op1( 0), Op2( 0), is32, access_pmcr},
@@ -234,6 +345,9 @@ static const struct coproc_reg cp15_regs[] = {
 	/* CNTKCTL: swapped by interrupt.S. */
 	{ CRn(14), CRm( 1), Op1( 0), Op2( 0), is32,
 			NULL, reset_val, c14_CNTKCTL, 0x00000000 },
+
+	/* The Configuration Base Address Register. */
+	{ CRn(15), CRm( 0), Op1( 4), Op2( 0), is32, access_cbar},
 };
 
 /* Target specific emulation tables */
@@ -241,6 +355,12 @@ static struct kvm_coproc_target_table *target_tables[KVM_ARM_NUM_TARGETS];
 
 void kvm_register_target_coproc_table(struct kvm_coproc_target_table *table)
 {
+	unsigned int i;
+
+	for (i = 1; i < table->num; i++)
+		BUG_ON(cmp_reg(&table->table[i-1],
+			       &table->table[i]) >= 0);
+
 	target_tables[table->target] = table;
 }
 
diff --git a/arch/arm/kvm/coproc_a15.c b/arch/arm/kvm/coproc_a15.c
index cf93472b9dd6..bb0cac1410cc 100644
--- a/arch/arm/kvm/coproc_a15.c
+++ b/arch/arm/kvm/coproc_a15.c
@@ -17,101 +17,12 @@
  * Foundation, 51 Franklin Street, Fifth Floor, Boston, MA  02110-1301, USA.
  */
 #include <linux/kvm_host.h>
-#include <asm/cputype.h>
-#include <asm/kvm_arm.h>
-#include <asm/kvm_host.h>
-#include <asm/kvm_emulate.h>
 #include <asm/kvm_coproc.h>
+#include <asm/kvm_emulate.h>
 #include <linux/init.h>
 
-static void reset_mpidr(struct kvm_vcpu *vcpu, const struct coproc_reg *r)
-{
-	/*
-	 * Compute guest MPIDR:
-	 * (Even if we present only one VCPU to the guest on an SMP
-	 * host we don't set the U bit in the MPIDR, or vice versa, as
-	 * revealing the underlying hardware properties is likely to
-	 * be the best choice).
-	 */
-	vcpu->arch.cp15[c0_MPIDR] = (read_cpuid_mpidr() & ~MPIDR_LEVEL_MASK)
-		| (vcpu->vcpu_id & MPIDR_LEVEL_MASK);
-}
-
 #include "coproc.h"
 
-/* A15 TRM 4.3.28: RO WI */
-static bool access_actlr(struct kvm_vcpu *vcpu,
-			 const struct coproc_params *p,
-			 const struct coproc_reg *r)
-{
-	if (p->is_write)
-		return ignore_write(vcpu, p);
-
-	*vcpu_reg(vcpu, p->Rt1) = vcpu->arch.cp15[c1_ACTLR];
-	return true;
-}
-
-/* A15 TRM 4.3.60: R/O. */
-static bool access_cbar(struct kvm_vcpu *vcpu,
-			const struct coproc_params *p,
-			const struct coproc_reg *r)
-{
-	if (p->is_write)
-		return write_to_read_only(vcpu, p);
-	return read_zero(vcpu, p);
-}
-
-/* A15 TRM 4.3.48: R/O WI. */
-static bool access_l2ctlr(struct kvm_vcpu *vcpu,
-			  const struct coproc_params *p,
-			  const struct coproc_reg *r)
-{
-	if (p->is_write)
-		return ignore_write(vcpu, p);
-
-	*vcpu_reg(vcpu, p->Rt1) = vcpu->arch.cp15[c9_L2CTLR];
-	return true;
-}
-
-static void reset_l2ctlr(struct kvm_vcpu *vcpu, const struct coproc_reg *r)
-{
-	u32 l2ctlr, ncores;
-
-	asm volatile("mrc p15, 1, %0, c9, c0, 2\n" : "=r" (l2ctlr));
-	l2ctlr &= ~(3 << 24);
-	ncores = atomic_read(&vcpu->kvm->online_vcpus) - 1;
-	l2ctlr |= (ncores & 3) << 24;
-
-	vcpu->arch.cp15[c9_L2CTLR] = l2ctlr;
-}
-
-static void reset_actlr(struct kvm_vcpu *vcpu, const struct coproc_reg *r)
-{
-	u32 actlr;
-
-	/* ACTLR contains SMP bit: make sure you create all cpus first! */
-	asm volatile("mrc p15, 0, %0, c1, c0, 1\n" : "=r" (actlr));
-	/* Make the SMP bit consistent with the guest configuration */
-	if (atomic_read(&vcpu->kvm->online_vcpus) > 1)
-		actlr |= 1U << 6;
-	else
-		actlr &= ~(1U << 6);
-
-	vcpu->arch.cp15[c1_ACTLR] = actlr;
-}
-
-/* A15 TRM 4.3.49: R/O WI (even if NSACR.NS_L2ERR, a write of 1 is ignored). */
-static bool access_l2ectlr(struct kvm_vcpu *vcpu,
-			   const struct coproc_params *p,
-			   const struct coproc_reg *r)
-{
-	if (p->is_write)
-		return ignore_write(vcpu, p);
-
-	*vcpu_reg(vcpu, p->Rt1) = 0;
-	return true;
-}
-
 /*
  * A15-specific CP15 registers.
  * CRn denotes the primary register number, but is copied to the CRm in the
@@ -121,29 +32,9 @@ static bool access_l2ectlr(struct kvm_vcpu *vcpu,
  *            registers preceding 32-bit ones.
  */
 static const struct coproc_reg a15_regs[] = {
-	/* MPIDR: we use VMPIDR for guest access. */
-	{ CRn( 0), CRm( 0), Op1( 0), Op2( 5), is32,
-			NULL, reset_mpidr, c0_MPIDR },
-
 	/* SCTLR: swapped by interrupt.S. */
 	{ CRn( 1), CRm( 0), Op1( 0), Op2( 0), is32,
 			NULL, reset_val, c1_SCTLR, 0x00C50078 },
-	/* ACTLR: trapped by HCR.TAC bit. */
-	{ CRn( 1), CRm( 0), Op1( 0), Op2( 1), is32,
-			access_actlr, reset_actlr, c1_ACTLR },
-	/* CPACR: swapped by interrupt.S. */
-	{ CRn( 1), CRm( 0), Op1( 0), Op2( 2), is32,
-			NULL, reset_val, c1_CPACR, 0x00000000 },
-
-	/*
-	 * L2CTLR access (guest wants to know #CPUs).
-	 */
-	{ CRn( 9), CRm( 0), Op1( 1), Op2( 2), is32,
-			access_l2ctlr, reset_l2ctlr, c9_L2CTLR },
-	{ CRn( 9), CRm( 0), Op1( 1), Op2( 3), is32, access_l2ectlr},
-
-	/* The Configuration Base Address Register. */
-	{ CRn(15), CRm( 0), Op1( 4), Op2( 0), is32, access_cbar},
 };
 
 static struct kvm_coproc_target_table a15_target_table = {
@@ -154,12 +45,6 @@ static struct kvm_coproc_target_table a15_target_table = {
 
 static int __init coproc_a15_init(void)
 {
-	unsigned int i;
-
-	for (i = 1; i < ARRAY_SIZE(a15_regs); i++)
-		BUG_ON(cmp_reg(&a15_regs[i-1],
-			       &a15_regs[i]) >= 0);
-
 	kvm_register_target_coproc_table(&a15_target_table);
 	return 0;
 }
diff --git a/arch/arm/kvm/coproc_a7.c b/arch/arm/kvm/coproc_a7.c
new file mode 100644
index 000000000000..1df767331588
--- /dev/null
+++ b/arch/arm/kvm/coproc_a7.c
@@ -0,0 +1,54 @@
+/*
+ * Copyright (C) 2012 - Virtual Open Systems and Columbia University
+ * Copyright (C) 2013 - ARM Ltd
+ *
+ * Authors: Rusty Russell <rusty@rustcorp.au>
+ *          Christoffer Dall <c.dall@virtualopensystems.com>
+ *          Jonathan Austin <jonathan.austin@arm.com>
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License, version 2, as
+ * published by the Free Software Foundation.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+ * GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program; if not, write to the Free Software
+ * Foundation, 51 Franklin Street, Fifth Floor, Boston, MA  02110-1301, USA.
+ */
+#include <linux/kvm_host.h>
+#include <asm/kvm_coproc.h>
+#include <asm/kvm_emulate.h>
+#include <linux/init.h>
+
+#include "coproc.h"
+
+/*
+ * Cortex-A7 specific CP15 registers.
+ * CRn denotes the primary register number, but is copied to the CRm in the
+ * user space API for 64-bit register access in line with the terminology used
+ * in the ARM ARM.
+ * Important: Must be sorted ascending by CRn, CRM, Op1, Op2 and with 64-bit
+ *            registers preceding 32-bit ones.
+ */
+static const struct coproc_reg a7_regs[] = {
+	/* SCTLR: swapped by interrupt.S. */
+	{ CRn( 1), CRm( 0), Op1( 0), Op2( 0), is32,
+			NULL, reset_val, c1_SCTLR, 0x00C50878 },
+};
+
+static struct kvm_coproc_target_table a7_target_table = {
+	.target = KVM_ARM_TARGET_CORTEX_A7,
+	.table = a7_regs,
+	.num = ARRAY_SIZE(a7_regs),
+};
+
+static int __init coproc_a7_init(void)
+{
+	kvm_register_target_coproc_table(&a7_target_table);
+	return 0;
+}
+late_initcall(coproc_a7_init);
diff --git a/arch/arm/kvm/emulate.c b/arch/arm/kvm/emulate.c
index bdede9e7da51..d6c005283678 100644
--- a/arch/arm/kvm/emulate.c
+++ b/arch/arm/kvm/emulate.c
@@ -354,7 +354,7 @@ static void inject_abt(struct kvm_vcpu *vcpu, bool is_pabt, unsigned long addr)
 	*vcpu_pc(vcpu) = exc_vector_base(vcpu) + vect_offset;
 
 	if (is_pabt) {
-		/* Set DFAR and DFSR */
+		/* Set IFAR and IFSR */
 		vcpu->arch.cp15[c6_IFAR] = addr;
 		is_lpae = (vcpu->arch.cp15[c2_TTBCR] >> 31);
 		/* Always give debug fault for now - should give guest a clue */
diff --git a/arch/arm/kvm/guest.c b/arch/arm/kvm/guest.c
index 152d03612181..20f8d97904af 100644
--- a/arch/arm/kvm/guest.c
+++ b/arch/arm/kvm/guest.c
@@ -190,6 +190,8 @@ int __attribute_const__ kvm_target_cpu(void)
 		return -EINVAL;
 
 	switch (part_number) {
+	case ARM_CPU_PART_CORTEX_A7:
+		return KVM_ARM_TARGET_CORTEX_A7;
 	case ARM_CPU_PART_CORTEX_A15:
 		return KVM_ARM_TARGET_CORTEX_A15;
 	default:
@@ -202,7 +204,7 @@ int kvm_vcpu_set_target(struct kvm_vcpu *vcpu,
 {
 	unsigned int i;
 
-	/* We can only do a cortex A15 for now. */
+	/* We can only cope with guest==host and only on A15/A7 (for now). */
 	if (init->target != kvm_target_cpu())
 		return -EINVAL;
 
@@ -222,6 +224,26 @@ int kvm_vcpu_set_target(struct kvm_vcpu *vcpu,
 	return kvm_reset_vcpu(vcpu);
 }
 
+int kvm_vcpu_preferred_target(struct kvm_vcpu_init *init)
+{
+	int target = kvm_target_cpu();
+
+	if (target < 0)
+		return -ENODEV;
+
+	memset(init, 0, sizeof(*init));
+
+	/*
+	 * For now, we don't return any features.
+	 * In future, we might use features to return target
+	 * specific features available for the preferred
+	 * target type.
+	 */
+	init->target = (__u32)target;
+
+	return 0;
+}
+
 int kvm_arch_vcpu_ioctl_get_fpu(struct kvm_vcpu *vcpu, struct kvm_fpu *fpu)
 {
 	return -EINVAL;
diff --git a/arch/arm/kvm/handle_exit.c b/arch/arm/kvm/handle_exit.c
index df4c82d47ad7..a92079011a83 100644
--- a/arch/arm/kvm/handle_exit.c
+++ b/arch/arm/kvm/handle_exit.c
@@ -73,23 +73,29 @@ static int handle_dabt_hyp(struct kvm_vcpu *vcpu, struct kvm_run *run)
 }
 
 /**
- * kvm_handle_wfi - handle a wait-for-interrupts instruction executed by a guest
+ * kvm_handle_wfx - handle a WFI or WFE instructions trapped in guests
  * @vcpu:	the vcpu pointer
  * @run:	the kvm_run structure pointer
  *
- * Simply sets the wait_for_interrupts flag on the vcpu structure, which will
- * halt execution of world-switches and schedule other host processes until
- * there is an incoming IRQ or FIQ to the VM.
+ * WFE: Yield the CPU and come back to this vcpu when the scheduler
+ * decides to.
+ * WFI: Simply call kvm_vcpu_block(), which will halt execution of
+ * world-switches and schedule other host processes until there is an
+ * incoming IRQ or FIQ to the VM.
  */
-static int kvm_handle_wfi(struct kvm_vcpu *vcpu, struct kvm_run *run)
+static int kvm_handle_wfx(struct kvm_vcpu *vcpu, struct kvm_run *run)
 {
 	trace_kvm_wfi(*vcpu_pc(vcpu));
-	kvm_vcpu_block(vcpu);
+	if (kvm_vcpu_get_hsr(vcpu) & HSR_WFI_IS_WFE)
+		kvm_vcpu_on_spin(vcpu);
+	else
+		kvm_vcpu_block(vcpu);
+
 	return 1;
 }
 
 static exit_handle_fn arm_exit_handlers[] = {
-	[HSR_EC_WFI]		= kvm_handle_wfi,
+	[HSR_EC_WFI]		= kvm_handle_wfx,
 	[HSR_EC_CP15_32]	= kvm_handle_cp15_32,
 	[HSR_EC_CP15_64]	= kvm_handle_cp15_64,
 	[HSR_EC_CP14_MR]	= kvm_handle_cp14_access,
diff --git a/arch/arm/kvm/mmio.c b/arch/arm/kvm/mmio.c
index 0c25d9487d53..4cb5a93182e9 100644
--- a/arch/arm/kvm/mmio.c
+++ b/arch/arm/kvm/mmio.c
@@ -23,6 +23,68 @@
 
 #include "trace.h"
 
+static void mmio_write_buf(char *buf, unsigned int len, unsigned long data)
+{
+	void *datap = NULL;
+	union {
+		u8	byte;
+		u16	hword;
+		u32	word;
+		u64	dword;
+	} tmp;
+
+	switch (len) {
+	case 1:
+		tmp.byte	= data;
+		datap		= &tmp.byte;
+		break;
+	case 2:
+		tmp.hword	= data;
+		datap		= &tmp.hword;
+		break;
+	case 4:
+		tmp.word	= data;
+		datap		= &tmp.word;
+		break;
+	case 8:
+		tmp.dword	= data;
+		datap		= &tmp.dword;
+		break;
+	}
+
+	memcpy(buf, datap, len);
+}
+
+static unsigned long mmio_read_buf(char *buf, unsigned int len)
+{
+	unsigned long data = 0;
+	union {
+		u16	hword;
+		u32	word;
+		u64	dword;
+	} tmp;
+
+	switch (len) {
+	case 1:
+		data = buf[0];
+		break;
+	case 2:
+		memcpy(&tmp.hword, buf, len);
+		data = tmp.hword;
+		break;
+	case 4:
+		memcpy(&tmp.word, buf, len);
+		data = tmp.word;
+		break;
+	case 8:
+		memcpy(&tmp.dword, buf, len);
+		data = tmp.dword;
+		break;
+	}
+
+	return data;
+}
+
 /**
  * kvm_handle_mmio_return -- Handle MMIO loads after user space emulation
  * @vcpu: The VCPU pointer
@@ -33,28 +95,27 @@
  */
 int kvm_handle_mmio_return(struct kvm_vcpu *vcpu, struct kvm_run *run)
 {
-	unsigned long *dest;
+	unsigned long data;
 	unsigned int len;
 	int mask;
 
 	if (!run->mmio.is_write) {
-		dest = vcpu_reg(vcpu, vcpu->arch.mmio_decode.rt);
-		*dest = 0;
-
 		len = run->mmio.len;
 		if (len > sizeof(unsigned long))
 			return -EINVAL;
 
-		memcpy(dest, run->mmio.data, len);
-
-		trace_kvm_mmio(KVM_TRACE_MMIO_READ, len, run->mmio.phys_addr,
-				*((u64 *)run->mmio.data));
+		data = mmio_read_buf(run->mmio.data, len);
 
 		if (vcpu->arch.mmio_decode.sign_extend &&
 		    len < sizeof(unsigned long)) {
 			mask = 1U << ((len * 8) - 1);
-			*dest = (*dest ^ mask) - mask;
+			data = (data ^ mask) - mask;
 		}
+
+		trace_kvm_mmio(KVM_TRACE_MMIO_READ, len, run->mmio.phys_addr,
+			       data);
+		data = vcpu_data_host_to_guest(vcpu, data, len);
+		*vcpu_reg(vcpu, vcpu->arch.mmio_decode.rt) = data;
 	}
 
 	return 0;
@@ -105,6 +166,7 @@ int io_mem_abort(struct kvm_vcpu *vcpu, struct kvm_run *run,
 		 phys_addr_t fault_ipa)
 {
 	struct kvm_exit_mmio mmio;
+	unsigned long data;
 	unsigned long rt;
 	int ret;
 
@@ -125,13 +187,15 @@ int io_mem_abort(struct kvm_vcpu *vcpu, struct kvm_run *run,
 	}
 
 	rt = vcpu->arch.mmio_decode.rt;
+	data = vcpu_data_guest_to_host(vcpu, *vcpu_reg(vcpu, rt), mmio.len);
+
 	trace_kvm_mmio((mmio.is_write) ? KVM_TRACE_MMIO_WRITE :
 					 KVM_TRACE_MMIO_READ_UNSATISFIED,
 			mmio.len, fault_ipa,
-			(mmio.is_write) ? *vcpu_reg(vcpu, rt) : 0);
+			(mmio.is_write) ? data : 0);
 
 	if (mmio.is_write)
-		memcpy(mmio.data, vcpu_reg(vcpu, rt), mmio.len);
+		mmio_write_buf(mmio.data, mmio.len, data);
 
 	if (vgic_handle_mmio(vcpu, run, &mmio))
 		return 1;
diff --git a/arch/arm/kvm/mmu.c b/arch/arm/kvm/mmu.c
index b0de86b56c13..371958370de4 100644
--- a/arch/arm/kvm/mmu.c
+++ b/arch/arm/kvm/mmu.c
@@ -19,6 +19,7 @@
 #include <linux/mman.h>
 #include <linux/kvm_host.h>
 #include <linux/io.h>
+#include <linux/hugetlb.h>
 #include <trace/events/kvm.h>
 #include <asm/pgalloc.h>
 #include <asm/cacheflush.h>
@@ -41,6 +42,8 @@ static unsigned long hyp_idmap_start;
 static unsigned long hyp_idmap_end;
 static phys_addr_t hyp_idmap_vector;
 
+#define kvm_pmd_huge(_x)	(pmd_huge(_x) || pmd_trans_huge(_x))
+
 static void kvm_tlb_flush_vmid_ipa(struct kvm *kvm, phys_addr_t ipa)
 {
 	/*
@@ -93,19 +96,29 @@ static bool page_empty(void *ptr)
 
 static void clear_pud_entry(struct kvm *kvm, pud_t *pud, phys_addr_t addr)
 {
-	pmd_t *pmd_table = pmd_offset(pud, 0);
-	pud_clear(pud);
-	kvm_tlb_flush_vmid_ipa(kvm, addr);
-	pmd_free(NULL, pmd_table);
+	if (pud_huge(*pud)) {
+		pud_clear(pud);
+		kvm_tlb_flush_vmid_ipa(kvm, addr);
+	} else {
+		pmd_t *pmd_table = pmd_offset(pud, 0);
+		pud_clear(pud);
+		kvm_tlb_flush_vmid_ipa(kvm, addr);
+		pmd_free(NULL, pmd_table);
+	}
 	put_page(virt_to_page(pud));
 }
 
 static void clear_pmd_entry(struct kvm *kvm, pmd_t *pmd, phys_addr_t addr)
 {
-	pte_t *pte_table = pte_offset_kernel(pmd, 0);
-	pmd_clear(pmd);
-	kvm_tlb_flush_vmid_ipa(kvm, addr);
-	pte_free_kernel(NULL, pte_table);
+	if (kvm_pmd_huge(*pmd)) {
+		pmd_clear(pmd);
+		kvm_tlb_flush_vmid_ipa(kvm, addr);
+	} else {
+		pte_t *pte_table = pte_offset_kernel(pmd, 0);
+		pmd_clear(pmd);
+		kvm_tlb_flush_vmid_ipa(kvm, addr);
+		pte_free_kernel(NULL, pte_table);
+	}
 	put_page(virt_to_page(pmd));
 }
 
@@ -136,18 +149,32 @@ static void unmap_range(struct kvm *kvm, pgd_t *pgdp,
 			continue;
 		}
 
+		if (pud_huge(*pud)) {
+			/*
+			 * If we are dealing with a huge pud, just clear it and
+			 * move on.
+			 */
+			clear_pud_entry(kvm, pud, addr);
+			addr = pud_addr_end(addr, end);
+			continue;
+		}
+
 		pmd = pmd_offset(pud, addr);
 		if (pmd_none(*pmd)) {
 			addr = pmd_addr_end(addr, end);
 			continue;
 		}
 
-		pte = pte_offset_kernel(pmd, addr);
-		clear_pte_entry(kvm, pte, addr);
-		next = addr + PAGE_SIZE;
+		if (!kvm_pmd_huge(*pmd)) {
+			pte = pte_offset_kernel(pmd, addr);
+			clear_pte_entry(kvm, pte, addr);
+			next = addr + PAGE_SIZE;
+		}
 
-		/* If we emptied the pte, walk back up the ladder */
-		if (page_empty(pte)) {
+		/*
+		 * If the pmd entry is to be cleared, walk back up the ladder
+		 */
+		if (kvm_pmd_huge(*pmd) || page_empty(pte)) {
 			clear_pmd_entry(kvm, pmd, addr);
 			next = pmd_addr_end(addr, end);
 			if (page_empty(pmd) && !page_empty(pud)) {
@@ -420,29 +447,71 @@ void kvm_free_stage2_pgd(struct kvm *kvm)
 	kvm->arch.pgd = NULL;
 }
 
-
-static int stage2_set_pte(struct kvm *kvm, struct kvm_mmu_memory_cache *cache,
-			  phys_addr_t addr, const pte_t *new_pte, bool iomap)
+static pmd_t *stage2_get_pmd(struct kvm *kvm, struct kvm_mmu_memory_cache *cache,
+			     phys_addr_t addr)
 {
 	pgd_t *pgd;
 	pud_t *pud;
 	pmd_t *pmd;
-	pte_t *pte, old_pte;
 
-	/* Create 2nd stage page table mapping - Level 1 */
 	pgd = kvm->arch.pgd + pgd_index(addr);
 	pud = pud_offset(pgd, addr);
 	if (pud_none(*pud)) {
 		if (!cache)
-			return 0; /* ignore calls from kvm_set_spte_hva */
+			return NULL;
 		pmd = mmu_memory_cache_alloc(cache);
 		pud_populate(NULL, pud, pmd);
 		get_page(virt_to_page(pud));
 	}
 
-	pmd = pmd_offset(pud, addr);
+	return pmd_offset(pud, addr);
+}
+
+static int stage2_set_pmd_huge(struct kvm *kvm, struct kvm_mmu_memory_cache
+			       *cache, phys_addr_t addr, const pmd_t *new_pmd)
+{
+	pmd_t *pmd, old_pmd;
+
+	pmd = stage2_get_pmd(kvm, cache, addr);
+	VM_BUG_ON(!pmd);
+
+	/*
+	 * Mapping in huge pages should only happen through a fault.  If a
+	 * page is merged into a transparent huge page, the individual
+	 * subpages of that huge page should be unmapped through MMU
+	 * notifiers before we get here.
+	 *
+	 * Merging of CompoundPages is not supported; they should become
+	 * splitting first, unmapped, merged, and mapped back in on-demand.
+	 */
+	VM_BUG_ON(pmd_present(*pmd) && pmd_pfn(*pmd) != pmd_pfn(*new_pmd));
+
+	old_pmd = *pmd;
+	kvm_set_pmd(pmd, *new_pmd);
+	if (pmd_present(old_pmd))
+		kvm_tlb_flush_vmid_ipa(kvm, addr);
+	else
+		get_page(virt_to_page(pmd));
+	return 0;
+}
+
+static int stage2_set_pte(struct kvm *kvm, struct kvm_mmu_memory_cache *cache,
+			  phys_addr_t addr, const pte_t *new_pte, bool iomap)
+{
+	pmd_t *pmd;
+	pte_t *pte, old_pte;
 
-	/* Create 2nd stage page table mapping - Level 2 */
+	/* Create stage-2 page table mapping - Level 1 */
+	pmd = stage2_get_pmd(kvm, cache, addr);
+	if (!pmd) {
+		/*
+		 * Ignore calls from kvm_set_spte_hva for unallocated
+		 * address ranges.
+		 */
+		return 0;
+	}
+
+	/* Create stage-2 page mappings - Level 2 */
 	if (pmd_none(*pmd)) {
 		if (!cache)
 			return 0; /* ignore calls from kvm_set_spte_hva */
@@ -507,16 +576,60 @@ out:
 	return ret;
 }
 
+static bool transparent_hugepage_adjust(pfn_t *pfnp, phys_addr_t *ipap)
+{
+	pfn_t pfn = *pfnp;
+	gfn_t gfn = *ipap >> PAGE_SHIFT;
+
+	if (PageTransCompound(pfn_to_page(pfn))) {
+		unsigned long mask;
+		/*
+		 * The address we faulted on is backed by a transparent huge
+		 * page.  However, because we map the compound huge page and
+		 * not the individual tail page, we need to transfer the
+		 * refcount to the head page.  We have to be careful that the
+		 * THP doesn't start to split while we are adjusting the
+		 * refcounts.
+		 *
+		 * We are sure this doesn't happen, because mmu_notifier_retry
+		 * was successful and we are holding the mmu_lock, so if this
+		 * THP is trying to split, it will be blocked in the mmu
+		 * notifier before touching any of the pages, specifically
+		 * before being able to call __split_huge_page_refcount().
+		 *
+		 * We can therefore safely transfer the refcount from PG_tail
+		 * to PG_head and switch the pfn from a tail page to the head
+		 * page accordingly.
+		 */
+		mask = PTRS_PER_PMD - 1;
+		VM_BUG_ON((gfn & mask) != (pfn & mask));
+		if (pfn & mask) {
+			*ipap &= PMD_MASK;
+			kvm_release_pfn_clean(pfn);
+			pfn &= ~mask;
+			kvm_get_pfn(pfn);
+			*pfnp = pfn;
+		}
+
+		return true;
+	}
+
+	return false;
+}
+
 static int user_mem_abort(struct kvm_vcpu *vcpu, phys_addr_t fault_ipa,
-			  gfn_t gfn, struct kvm_memory_slot *memslot,
+			  struct kvm_memory_slot *memslot,
 			  unsigned long fault_status)
 {
-	pte_t new_pte;
-	pfn_t pfn;
 	int ret;
-	bool write_fault, writable;
+	bool write_fault, writable, hugetlb = false, force_pte = false;
 	unsigned long mmu_seq;
+	gfn_t gfn = fault_ipa >> PAGE_SHIFT;
+	unsigned long hva = gfn_to_hva(vcpu->kvm, gfn);
+	struct kvm *kvm = vcpu->kvm;
 	struct kvm_mmu_memory_cache *memcache = &vcpu->arch.mmu_page_cache;
+	struct vm_area_struct *vma;
+	pfn_t pfn;
 
 	write_fault = kvm_is_write_fault(kvm_vcpu_get_hsr(vcpu));
 	if (fault_status == FSC_PERM && !write_fault) {
@@ -524,6 +637,26 @@ static int user_mem_abort(struct kvm_vcpu *vcpu, phys_addr_t fault_ipa,
 		return -EFAULT;
 	}
 
+	/* Let's check if we will get back a huge page backed by hugetlbfs */
+	down_read(&current->mm->mmap_sem);
+	vma = find_vma_intersection(current->mm, hva, hva + 1);
+	if (is_vm_hugetlb_page(vma)) {
+		hugetlb = true;
+		gfn = (fault_ipa & PMD_MASK) >> PAGE_SHIFT;
+	} else {
+		/*
+		 * Pages belonging to VMAs not aligned to the PMD mapping
+		 * granularity cannot be mapped using block descriptors even
+		 * if the pages belong to a THP for the process, because the
+		 * stage-2 block descriptor will cover more than a single THP
+		 * and we loose atomicity for unmapping, updates, and splits
+		 * of the THP or other pages in the stage-2 block range.
+		 */
+		if (vma->vm_start & ~PMD_MASK)
+			force_pte = true;
+	}
+	up_read(&current->mm->mmap_sem);
+
 	/* We need minimum second+third level pages */
 	ret = mmu_topup_memory_cache(memcache, 2, KVM_NR_MEM_OBJS);
 	if (ret)
@@ -541,26 +674,40 @@ static int user_mem_abort(struct kvm_vcpu *vcpu, phys_addr_t fault_ipa,
 	 */
 	smp_rmb();
 
-	pfn = gfn_to_pfn_prot(vcpu->kvm, gfn, write_fault, &writable);
+	pfn = gfn_to_pfn_prot(kvm, gfn, write_fault, &writable);
 	if (is_error_pfn(pfn))
 		return -EFAULT;
 
-	new_pte = pfn_pte(pfn, PAGE_S2);
-	coherent_icache_guest_page(vcpu->kvm, gfn);
-
-	spin_lock(&vcpu->kvm->mmu_lock);
-	if (mmu_notifier_retry(vcpu->kvm, mmu_seq))
+	spin_lock(&kvm->mmu_lock);
+	if (mmu_notifier_retry(kvm, mmu_seq))
 		goto out_unlock;
-	if (writable) {
-		kvm_set_s2pte_writable(&new_pte);
-		kvm_set_pfn_dirty(pfn);
+	if (!hugetlb && !force_pte)
+		hugetlb = transparent_hugepage_adjust(&pfn, &fault_ipa);
+
+	if (hugetlb) {
+		pmd_t new_pmd = pfn_pmd(pfn, PAGE_S2);
+		new_pmd = pmd_mkhuge(new_pmd);
+		if (writable) {
+			kvm_set_s2pmd_writable(&new_pmd);
+			kvm_set_pfn_dirty(pfn);
+		}
+		coherent_icache_guest_page(kvm, hva & PMD_MASK, PMD_SIZE);
+		ret = stage2_set_pmd_huge(kvm, memcache, fault_ipa, &new_pmd);
+	} else {
+		pte_t new_pte = pfn_pte(pfn, PAGE_S2);
+		if (writable) {
+			kvm_set_s2pte_writable(&new_pte);
+			kvm_set_pfn_dirty(pfn);
+		}
+		coherent_icache_guest_page(kvm, hva, PAGE_SIZE);
+		ret = stage2_set_pte(kvm, memcache, fault_ipa, &new_pte, false);
 	}
-	stage2_set_pte(vcpu->kvm, memcache, fault_ipa, &new_pte, false);
+
 
 out_unlock:
-	spin_unlock(&vcpu->kvm->mmu_lock);
+	spin_unlock(&kvm->mmu_lock);
 	kvm_release_pfn_clean(pfn);
-	return 0;
+	return ret;
 }
 
 /**
@@ -629,7 +776,7 @@ int kvm_handle_guest_abort(struct kvm_vcpu *vcpu, struct kvm_run *run)
 
 	memslot = gfn_to_memslot(vcpu->kvm, gfn);
 
-	ret = user_mem_abort(vcpu, fault_ipa, gfn, memslot, fault_status);
+	ret = user_mem_abort(vcpu, fault_ipa, memslot, fault_status);
 	if (ret == 0)
 		ret = 1;
 out_unlock:
diff --git a/arch/arm/kvm/psci.c b/arch/arm/kvm/psci.c
index 86a693a02ba3..0881bf169fbc 100644
--- a/arch/arm/kvm/psci.c
+++ b/arch/arm/kvm/psci.c
@@ -18,6 +18,7 @@
 #include <linux/kvm_host.h>
 #include <linux/wait.h>
 
+#include <asm/cputype.h>
 #include <asm/kvm_emulate.h>
 #include <asm/kvm_psci.h>
 
@@ -34,22 +35,30 @@ static void kvm_psci_vcpu_off(struct kvm_vcpu *vcpu)
 static unsigned long kvm_psci_vcpu_on(struct kvm_vcpu *source_vcpu)
 {
 	struct kvm *kvm = source_vcpu->kvm;
-	struct kvm_vcpu *vcpu;
+	struct kvm_vcpu *vcpu = NULL, *tmp;
 	wait_queue_head_t *wq;
 	unsigned long cpu_id;
+	unsigned long mpidr;
 	phys_addr_t target_pc;
+	int i;
 
 	cpu_id = *vcpu_reg(source_vcpu, 1);
 	if (vcpu_mode_is_32bit(source_vcpu))
 		cpu_id &= ~((u32) 0);
 
-	if (cpu_id >= atomic_read(&kvm->online_vcpus))
+	kvm_for_each_vcpu(i, tmp, kvm) {
+		mpidr = kvm_vcpu_get_mpidr(tmp);
+		if ((mpidr & MPIDR_HWID_BITMASK) == (cpu_id & MPIDR_HWID_BITMASK)) {
+			vcpu = tmp;
+			break;
+		}
+	}
+
+	if (!vcpu)
 		return KVM_PSCI_RET_INVAL;
 
 	target_pc = *vcpu_reg(source_vcpu, 2);
 
-	vcpu = kvm_get_vcpu(kvm, cpu_id);
-
 	wq = kvm_arch_vcpu_wq(vcpu);
 	if (!waitqueue_active(wq))
 		return KVM_PSCI_RET_INVAL;
@@ -62,6 +71,10 @@ static unsigned long kvm_psci_vcpu_on(struct kvm_vcpu *source_vcpu)
 		vcpu_set_thumb(vcpu);
 	}
 
+	/* Propagate caller endianness */
+	if (kvm_vcpu_is_be(source_vcpu))
+		kvm_vcpu_set_be(vcpu);
+
 	*vcpu_pc(vcpu) = target_pc;
 	vcpu->arch.pause = false;
 	smp_mb();		/* Make sure the above is visible */
diff --git a/arch/arm/kvm/reset.c b/arch/arm/kvm/reset.c
index c02ba4af599f..f558c073c023 100644
--- a/arch/arm/kvm/reset.c
+++ b/arch/arm/kvm/reset.c
@@ -30,16 +30,14 @@
 #include <kvm/arm_arch_timer.h>
 
 /******************************************************************************
- * Cortex-A15 Reset Values
+ * Cortex-A15 and Cortex-A7 Reset Values
  */
 
-static const int a15_max_cpu_idx = 3;
-
-static struct kvm_regs a15_regs_reset = {
+static struct kvm_regs cortexa_regs_reset = {
 	.usr_regs.ARM_cpsr = SVC_MODE | PSR_A_BIT | PSR_I_BIT | PSR_F_BIT,
 };
 
-static const struct kvm_irq_level a15_vtimer_irq = {
+static const struct kvm_irq_level cortexa_vtimer_irq = {
 	{ .irq = 27 },
 	.level = 1,
 };
@@ -62,12 +60,11 @@ int kvm_reset_vcpu(struct kvm_vcpu *vcpu)
 	const struct kvm_irq_level *cpu_vtimer_irq;
 
 	switch (vcpu->arch.target) {
+	case KVM_ARM_TARGET_CORTEX_A7:
 	case KVM_ARM_TARGET_CORTEX_A15:
-		if (vcpu->vcpu_id > a15_max_cpu_idx)
-			return -EINVAL;
-		reset_regs = &a15_regs_reset;
+		reset_regs = &cortexa_regs_reset;
 		vcpu->arch.midr = read_cpuid_id();
-		cpu_vtimer_irq = &a15_vtimer_irq;
+		cpu_vtimer_irq = &cortexa_vtimer_irq;
 		break;
 	default:
 		return -ENODEV;
diff --git a/arch/arm/mach-at91/Makefile b/arch/arm/mach-at91/Makefile
index c1b737097c95..90aab2d5a07f 100644
--- a/arch/arm/mach-at91/Makefile
+++ b/arch/arm/mach-at91/Makefile
@@ -2,7 +2,7 @@
 # Makefile for the linux kernel.
 #
 
-obj-y		:= irq.o gpio.o setup.o
+obj-y		:= irq.o gpio.o setup.o sysirq_mask.o
 obj-m		:=
 obj-n		:=
 obj-		:=
diff --git a/arch/arm/mach-at91/at91sam9260.c b/arch/arm/mach-at91/at91sam9260.c
index f8629a3fa245..d6a1fa85371d 100644
--- a/arch/arm/mach-at91/at91sam9260.c
+++ b/arch/arm/mach-at91/at91sam9260.c
@@ -351,6 +351,8 @@ static void __init at91sam9260_initialize(void)
 	arm_pm_idle = at91sam9_idle;
 	arm_pm_restart = at91sam9_alt_restart;
 
+	at91_sysirq_mask_rtt(AT91SAM9260_BASE_RTT);
+
 	/* Register GPIO subsystem */
 	at91_gpio_init(at91sam9260_gpio, 3);
 }
diff --git a/arch/arm/mach-at91/at91sam9261.c b/arch/arm/mach-at91/at91sam9261.c
index 1f3867a17a28..23ba1d8a1531 100644
--- a/arch/arm/mach-at91/at91sam9261.c
+++ b/arch/arm/mach-at91/at91sam9261.c
@@ -293,6 +293,8 @@ static void __init at91sam9261_initialize(void)
 	arm_pm_idle = at91sam9_idle;
 	arm_pm_restart = at91sam9_alt_restart;
 
+	at91_sysirq_mask_rtt(AT91SAM9261_BASE_RTT);
+
 	/* Register GPIO subsystem */
 	at91_gpio_init(at91sam9261_gpio, 3);
 }
diff --git a/arch/arm/mach-at91/at91sam9261_devices.c b/arch/arm/mach-at91/at91sam9261_devices.c
index 629ea5fc95cf..b2a34740146a 100644
--- a/arch/arm/mach-at91/at91sam9261_devices.c
+++ b/arch/arm/mach-at91/at91sam9261_devices.c
@@ -465,7 +465,7 @@ void __init at91_add_device_spi(struct spi_board_info *devices, int nr_devices)
 
 #if defined(CONFIG_FB_ATMEL) || defined(CONFIG_FB_ATMEL_MODULE)
 static u64 lcdc_dmamask = DMA_BIT_MASK(32);
-static struct atmel_lcdfb_info lcdc_data;
+static struct atmel_lcdfb_pdata lcdc_data;
 
 static struct resource lcdc_resources[] = {
 	[0] = {
@@ -498,7 +498,7 @@ static struct platform_device at91_lcdc_device = {
 	.num_resources	= ARRAY_SIZE(lcdc_resources),
 };
 
-void __init at91_add_device_lcdc(struct atmel_lcdfb_info *data)
+void __init at91_add_device_lcdc(struct atmel_lcdfb_pdata *data)
 {
 	if (!data) {
 		return;
@@ -559,7 +559,7 @@ void __init at91_add_device_lcdc(struct atmel_lcdfb_info *data)
 	platform_device_register(&at91_lcdc_device);
 }
 #else
-void __init at91_add_device_lcdc(struct atmel_lcdfb_info *data) {}
+void __init at91_add_device_lcdc(struct atmel_lcdfb_pdata *data) {}
 #endif
 
 
diff --git a/arch/arm/mach-at91/at91sam9263.c b/arch/arm/mach-at91/at91sam9263.c
index 90d455d294a1..7eccb0fc57bc 100644
--- a/arch/arm/mach-at91/at91sam9263.c
+++ b/arch/arm/mach-at91/at91sam9263.c
@@ -330,6 +330,9 @@ static void __init at91sam9263_initialize(void)
 	arm_pm_idle = at91sam9_idle;
 	arm_pm_restart = at91sam9_alt_restart;
 
+	at91_sysirq_mask_rtt(AT91SAM9263_BASE_RTT0);
+	at91_sysirq_mask_rtt(AT91SAM9263_BASE_RTT1);
+
 	/* Register GPIO subsystem */
 	at91_gpio_init(at91sam9263_gpio, 5);
 }
diff --git a/arch/arm/mach-at91/at91sam9263_devices.c b/arch/arm/mach-at91/at91sam9263_devices.c
index 858c8aac2daf..4aeadddbc181 100644
--- a/arch/arm/mach-at91/at91sam9263_devices.c
+++ b/arch/arm/mach-at91/at91sam9263_devices.c
@@ -832,7 +832,7 @@ void __init at91_add_device_can(struct at91_can_data *data) {}
 
 #if defined(CONFIG_FB_ATMEL) || defined(CONFIG_FB_ATMEL_MODULE)
 static u64 lcdc_dmamask = DMA_BIT_MASK(32);
-static struct atmel_lcdfb_info lcdc_data;
+static struct atmel_lcdfb_pdata lcdc_data;
 
 static struct resource lcdc_resources[] = {
 	[0] = {
@@ -859,7 +859,7 @@ static struct platform_device at91_lcdc_device = {
 	.num_resources	= ARRAY_SIZE(lcdc_resources),
 };
 
-void __init at91_add_device_lcdc(struct atmel_lcdfb_info *data)
+void __init at91_add_device_lcdc(struct atmel_lcdfb_pdata *data)
 {
 	if (!data)
 		return;
@@ -891,7 +891,7 @@ void __init at91_add_device_lcdc(struct atmel_lcdfb_info *data)
 	platform_device_register(&at91_lcdc_device);
 }
 #else
-void __init at91_add_device_lcdc(struct atmel_lcdfb_info *data) {}
+void __init at91_add_device_lcdc(struct atmel_lcdfb_pdata *data) {}
 #endif
 
 
diff --git a/arch/arm/mach-at91/at91sam9g45.c b/arch/arm/mach-at91/at91sam9g45.c
index e9bf0b8f40eb..9405aa08b104 100644
--- a/arch/arm/mach-at91/at91sam9g45.c
+++ b/arch/arm/mach-at91/at91sam9g45.c
@@ -379,6 +379,9 @@ static void __init at91sam9g45_initialize(void)
 	arm_pm_idle = at91sam9_idle;
 	arm_pm_restart = at91sam9g45_restart;
 
+	at91_sysirq_mask_rtc(AT91SAM9G45_BASE_RTC);
+	at91_sysirq_mask_rtt(AT91SAM9G45_BASE_RTT);
+
 	/* Register GPIO subsystem */
 	at91_gpio_init(at91sam9g45_gpio, 5);
 }
diff --git a/arch/arm/mach-at91/at91sam9g45_devices.c b/arch/arm/mach-at91/at91sam9g45_devices.c
index acb703e13331..cb36fa872d30 100644
--- a/arch/arm/mach-at91/at91sam9g45_devices.c
+++ b/arch/arm/mach-at91/at91sam9g45_devices.c
@@ -965,7 +965,7 @@ void __init at91_add_device_isi(struct isi_platform_data *data,
 
 #if defined(CONFIG_FB_ATMEL) || defined(CONFIG_FB_ATMEL_MODULE)
 static u64 lcdc_dmamask = DMA_BIT_MASK(32);
-static struct atmel_lcdfb_info lcdc_data;
+static struct atmel_lcdfb_pdata lcdc_data;
 
 static struct resource lcdc_resources[] = {
 	[0] = {
@@ -991,7 +991,7 @@ static struct platform_device at91_lcdc_device = {
 	.num_resources	= ARRAY_SIZE(lcdc_resources),
 };
 
-void __init at91_add_device_lcdc(struct atmel_lcdfb_info *data)
+void __init at91_add_device_lcdc(struct atmel_lcdfb_pdata *data)
 {
 	if (!data)
 		return;
@@ -1037,7 +1037,7 @@ void __init at91_add_device_lcdc(struct atmel_lcdfb_info *data)
 	platform_device_register(&at91_lcdc_device);
 }
 #else
-void __init at91_add_device_lcdc(struct atmel_lcdfb_info *data) {}
+void __init at91_add_device_lcdc(struct atmel_lcdfb_pdata *data) {}
 #endif
 
 
diff --git a/arch/arm/mach-at91/at91sam9n12.c b/arch/arm/mach-at91/at91sam9n12.c
index 2d895a297739..388ec3aec4b9 100644
--- a/arch/arm/mach-at91/at91sam9n12.c
+++ b/arch/arm/mach-at91/at91sam9n12.c
@@ -224,7 +224,13 @@ static void __init at91sam9n12_map_io(void)
 	at91_init_sram(0, AT91SAM9N12_SRAM_BASE, AT91SAM9N12_SRAM_SIZE);
 }
 
+static void __init at91sam9n12_initialize(void)
+{
+	at91_sysirq_mask_rtc(AT91SAM9N12_BASE_RTC);
+}
+
 AT91_SOC_START(at91sam9n12)
 	.map_io = at91sam9n12_map_io,
 	.register_clocks = at91sam9n12_register_clocks,
+	.init = at91sam9n12_initialize,
 AT91_SOC_END
diff --git a/arch/arm/mach-at91/at91sam9rl.c b/arch/arm/mach-at91/at91sam9rl.c
index 88995af09c04..0750ffb7e6b1 100644
--- a/arch/arm/mach-at91/at91sam9rl.c
+++ b/arch/arm/mach-at91/at91sam9rl.c
@@ -296,6 +296,9 @@ static void __init at91sam9rl_initialize(void)
 	arm_pm_idle = at91sam9_idle;
 	arm_pm_restart = at91sam9_alt_restart;
 
+	at91_sysirq_mask_rtc(AT91SAM9RL_BASE_RTC);
+	at91_sysirq_mask_rtt(AT91SAM9RL_BASE_RTT);
+
 	/* Register GPIO subsystem */
 	at91_gpio_init(at91sam9rl_gpio, 4);
 }
diff --git a/arch/arm/mach-at91/at91sam9rl_devices.c b/arch/arm/mach-at91/at91sam9rl_devices.c
index 352468f265a9..a698bdab2cce 100644
--- a/arch/arm/mach-at91/at91sam9rl_devices.c
+++ b/arch/arm/mach-at91/at91sam9rl_devices.c
@@ -498,7 +498,7 @@ void __init at91_add_device_ac97(struct ac97c_platform_data *data) {}
 
 #if defined(CONFIG_FB_ATMEL) || defined(CONFIG_FB_ATMEL_MODULE)
 static u64 lcdc_dmamask = DMA_BIT_MASK(32);
-static struct atmel_lcdfb_info lcdc_data;
+static struct atmel_lcdfb_pdata lcdc_data;
 
 static struct resource lcdc_resources[] = {
 	[0] = {
@@ -525,7 +525,7 @@ static struct platform_device at91_lcdc_device = {
 	.num_resources	= ARRAY_SIZE(lcdc_resources),
 };
 
-void __init at91_add_device_lcdc(struct atmel_lcdfb_info *data)
+void __init at91_add_device_lcdc(struct atmel_lcdfb_pdata *data)
 {
 	if (!data) {
 		return;
@@ -557,7 +557,7 @@ void __init at91_add_device_lcdc(struct atmel_lcdfb_info *data)
 	platform_device_register(&at91_lcdc_device);
 }
 #else
-void __init at91_add_device_lcdc(struct atmel_lcdfb_info *data) {}
+void __init at91_add_device_lcdc(struct atmel_lcdfb_pdata *data) {}
 #endif
 
 
diff --git a/arch/arm/mach-at91/at91sam9x5.c b/arch/arm/mach-at91/at91sam9x5.c
index 916e5a142917..e8a2e075a1b8 100644
--- a/arch/arm/mach-at91/at91sam9x5.c
+++ b/arch/arm/mach-at91/at91sam9x5.c
@@ -322,6 +322,11 @@ static void __init at91sam9x5_map_io(void)
 	at91_init_sram(0, AT91SAM9X5_SRAM_BASE, AT91SAM9X5_SRAM_SIZE);
 }
 
+static void __init at91sam9x5_initialize(void)
+{
+	at91_sysirq_mask_rtc(AT91SAM9X5_BASE_RTC);
+}
+
 /* --------------------------------------------------------------------
  *  Interrupt initialization
  * -------------------------------------------------------------------- */
@@ -329,4 +334,5 @@ static void __init at91sam9x5_map_io(void)
 AT91_SOC_START(at91sam9x5)
 	.map_io = at91sam9x5_map_io,
 	.register_clocks = at91sam9x5_register_clocks,
+	.init = at91sam9x5_initialize,
 AT91_SOC_END
diff --git a/arch/arm/mach-at91/board-sam9260ek.c b/arch/arm/mach-at91/board-sam9260ek.c
index 0b153c87521d..f4f8735315da 100644
--- a/arch/arm/mach-at91/board-sam9260ek.c
+++ b/arch/arm/mach-at91/board-sam9260ek.c
@@ -28,7 +28,7 @@
 #include <linux/spi/spi.h>
 #include <linux/spi/at73c213.h>
 #include <linux/clk.h>
-#include <linux/i2c/at24.h>
+#include <linux/platform_data/at24.h>
 #include <linux/gpio_keys.h>
 #include <linux/input.h>
 
diff --git a/arch/arm/mach-at91/board-sam9261ek.c b/arch/arm/mach-at91/board-sam9261ek.c
index d3437624ca4e..473546b9408b 100644
--- a/arch/arm/mach-at91/board-sam9261ek.c
+++ b/arch/arm/mach-at91/board-sam9261ek.c
@@ -389,7 +389,7 @@ static struct fb_monspecs at91fb_default_stn_monspecs = {
 					| ATMEL_LCDC_IFWIDTH_4 \
 					| ATMEL_LCDC_SCANMOD_SINGLE)
 
-static void at91_lcdc_stn_power_control(int on)
+static void at91_lcdc_stn_power_control(struct atmel_lcdfb_pdata *pdata, int on)
 {
 	/* backlight */
 	if (on) {	/* power up */
@@ -401,7 +401,7 @@ static void at91_lcdc_stn_power_control(int on)
 	}
 }
 
-static struct atmel_lcdfb_info __initdata ek_lcdc_data = {
+static struct atmel_lcdfb_pdata __initdata ek_lcdc_data = {
 	.default_bpp			= 1,
 	.default_dmacon			= ATMEL_LCDC_DMAEN,
 	.default_lcdcon2		= AT91SAM9261_DEFAULT_STN_LCDCON2,
@@ -445,7 +445,7 @@ static struct fb_monspecs at91fb_default_tft_monspecs = {
 					| ATMEL_LCDC_DISTYPE_TFT    \
 					| ATMEL_LCDC_CLKMOD_ALWAYSACTIVE)
 
-static void at91_lcdc_tft_power_control(int on)
+static void at91_lcdc_tft_power_control(struct atmel_lcdfb_pdata *pdata, int on)
 {
 	if (on)
 		at91_set_gpio_value(AT91_PIN_PA12, 0);	/* power up */
@@ -453,7 +453,7 @@ static void at91_lcdc_tft_power_control(int on)
 		at91_set_gpio_value(AT91_PIN_PA12, 1);	/* power down */
 }
 
-static struct atmel_lcdfb_info __initdata ek_lcdc_data = {
+static struct atmel_lcdfb_pdata __initdata ek_lcdc_data = {
 	.lcdcon_is_backlight		= true,
 	.default_bpp			= 16,
 	.default_dmacon			= ATMEL_LCDC_DMAEN,
@@ -465,7 +465,7 @@ static struct atmel_lcdfb_info __initdata ek_lcdc_data = {
 #endif
 
 #else
-static struct atmel_lcdfb_info __initdata ek_lcdc_data;
+static struct atmel_lcdfb_pdata __initdata ek_lcdc_data;
 #endif
 
 
diff --git a/arch/arm/mach-at91/board-sam9263ek.c b/arch/arm/mach-at91/board-sam9263ek.c
index 3284df05df14..2f931915c80c 100644
--- a/arch/arm/mach-at91/board-sam9263ek.c
+++ b/arch/arm/mach-at91/board-sam9263ek.c
@@ -27,7 +27,7 @@
 #include <linux/platform_device.h>
 #include <linux/spi/spi.h>
 #include <linux/spi/ads7846.h>
-#include <linux/i2c/at24.h>
+#include <linux/platform_data/at24.h>
 #include <linux/fb.h>
 #include <linux/gpio_keys.h>
 #include <linux/input.h>
@@ -275,13 +275,13 @@ static struct fb_monspecs at91fb_default_monspecs = {
 					| ATMEL_LCDC_DISTYPE_TFT \
 					| ATMEL_LCDC_CLKMOD_ALWAYSACTIVE)
 
-static void at91_lcdc_power_control(int on)
+static void at91_lcdc_power_control(struct atmel_lcdfb_pdata *pdata, int on)
 {
 	at91_set_gpio_value(AT91_PIN_PA30, on);
 }
 
 /* Driver datas */
-static struct atmel_lcdfb_info __initdata ek_lcdc_data = {
+static struct atmel_lcdfb_pdata __initdata ek_lcdc_data = {
 	.lcdcon_is_backlight		= true,
 	.default_bpp			= 16,
 	.default_dmacon			= ATMEL_LCDC_DMAEN,
@@ -292,7 +292,7 @@ static struct atmel_lcdfb_info __initdata ek_lcdc_data = {
 };
 
 #else
-static struct atmel_lcdfb_info __initdata ek_lcdc_data;
+static struct atmel_lcdfb_pdata __initdata ek_lcdc_data;
 #endif
 
 
diff --git a/arch/arm/mach-at91/board-sam9m10g45ek.c b/arch/arm/mach-at91/board-sam9m10g45ek.c
index 2a94896a1375..ef39078c8ce2 100644
--- a/arch/arm/mach-at91/board-sam9m10g45ek.c
+++ b/arch/arm/mach-at91/board-sam9m10g45ek.c
@@ -284,7 +284,7 @@ static struct fb_monspecs at91fb_default_monspecs = {
 					| ATMEL_LCDC_CLKMOD_ALWAYSACTIVE)
 
 /* Driver datas */
-static struct atmel_lcdfb_info __initdata ek_lcdc_data = {
+static struct atmel_lcdfb_pdata __initdata ek_lcdc_data = {
 	.lcdcon_is_backlight		= true,
 	.default_bpp			= 32,
 	.default_dmacon			= ATMEL_LCDC_DMAEN,
@@ -295,7 +295,7 @@ static struct atmel_lcdfb_info __initdata ek_lcdc_data = {
 };
 
 #else
-static struct atmel_lcdfb_info __initdata ek_lcdc_data;
+static struct atmel_lcdfb_pdata __initdata ek_lcdc_data;
 #endif
 
 
diff --git a/arch/arm/mach-at91/board-sam9rlek.c b/arch/arm/mach-at91/board-sam9rlek.c
index aa265dcf2128..604eecf6cd70 100644
--- a/arch/arm/mach-at91/board-sam9rlek.c
+++ b/arch/arm/mach-at91/board-sam9rlek.c
@@ -170,7 +170,7 @@ static struct fb_monspecs at91fb_default_monspecs = {
 					| ATMEL_LCDC_DISTYPE_TFT \
 					| ATMEL_LCDC_CLKMOD_ALWAYSACTIVE)
 
-static void at91_lcdc_power_control(int on)
+static void at91_lcdc_power_control(struct atmel_lcdfb_pdata *pdata, int on)
 {
 	if (on)
 		at91_set_gpio_value(AT91_PIN_PC1, 0);	/* power up */
@@ -179,7 +179,7 @@ static void at91_lcdc_power_control(int on)
 }
 
 /* Driver datas */
-static struct atmel_lcdfb_info __initdata ek_lcdc_data = {
+static struct atmel_lcdfb_pdata __initdata ek_lcdc_data = {
 	.lcdcon_is_backlight            = true,
 	.default_bpp			= 16,
 	.default_dmacon			= ATMEL_LCDC_DMAEN,
@@ -191,7 +191,7 @@ static struct atmel_lcdfb_info __initdata ek_lcdc_data = {
 };
 
 #else
-static struct atmel_lcdfb_info __initdata ek_lcdc_data;
+static struct atmel_lcdfb_pdata __initdata ek_lcdc_data;
 #endif
 
 
diff --git a/arch/arm/mach-at91/board.h b/arch/arm/mach-at91/board.h
index 4a234fb2ab3b..6c08b341167d 100644
--- a/arch/arm/mach-at91/board.h
+++ b/arch/arm/mach-at91/board.h
@@ -107,8 +107,8 @@ extern void __init at91_add_device_pwm(u32 mask);
 extern void __init at91_add_device_ssc(unsigned id, unsigned pins);
 
  /* LCD Controller */
-struct atmel_lcdfb_info;
-extern void __init at91_add_device_lcdc(struct atmel_lcdfb_info *data);
+struct atmel_lcdfb_pdata;
+extern void __init at91_add_device_lcdc(struct atmel_lcdfb_pdata *data);
 
  /* AC97 */
 extern void __init at91_add_device_ac97(struct ac97c_platform_data *data);
diff --git a/arch/arm/mach-at91/generic.h b/arch/arm/mach-at91/generic.h
index dc6e2f5f804d..26dee3ce9397 100644
--- a/arch/arm/mach-at91/generic.h
+++ b/arch/arm/mach-at91/generic.h
@@ -34,6 +34,8 @@ extern int  __init at91_aic_of_init(struct device_node *node,
 				    struct device_node *parent);
 extern int  __init at91_aic5_of_init(struct device_node *node,
 				    struct device_node *parent);
+extern void __init at91_sysirq_mask_rtc(u32 rtc_base);
+extern void __init at91_sysirq_mask_rtt(u32 rtt_base);
 
 
  /* Timer */
diff --git a/arch/arm/mach-at91/include/mach/at91sam9n12.h b/arch/arm/mach-at91/include/mach/at91sam9n12.h
index d374b87c0459..0151bcf6163c 100644
--- a/arch/arm/mach-at91/include/mach/at91sam9n12.h
+++ b/arch/arm/mach-at91/include/mach/at91sam9n12.h
@@ -49,6 +49,11 @@
 #define AT91SAM9N12_BASE_USART3	0xf8028000
 
 /*
+ * System Peripherals
+ */
+#define AT91SAM9N12_BASE_RTC	0xfffffeb0
+
+/*
  * Internal Memory.
  */
 #define AT91SAM9N12_SRAM_BASE	0x00300000	/* Internal SRAM base address */
diff --git a/arch/arm/mach-at91/include/mach/at91sam9x5.h b/arch/arm/mach-at91/include/mach/at91sam9x5.h
index c75ee19b58d3..2fc76c49e97c 100644
--- a/arch/arm/mach-at91/include/mach/at91sam9x5.h
+++ b/arch/arm/mach-at91/include/mach/at91sam9x5.h
@@ -55,6 +55,11 @@
 #define AT91SAM9X5_BASE_USART2	0xf8024000
 
 /*
+ * System Peripherals
+ */
+#define AT91SAM9X5_BASE_RTC	0xfffffeb0
+
+/*
  * Internal Memory.
  */
 #define AT91SAM9X5_SRAM_BASE	0x00300000	/* Internal SRAM base address */
diff --git a/arch/arm/mach-at91/include/mach/sama5d3.h b/arch/arm/mach-at91/include/mach/sama5d3.h
index 31096a8aaf1d..25613d8c6dcd 100644
--- a/arch/arm/mach-at91/include/mach/sama5d3.h
+++ b/arch/arm/mach-at91/include/mach/sama5d3.h
@@ -73,6 +73,11 @@
 #define SAMA5D3_BASE_USART3	0xf8024000
 
 /*
+ * System Peripherals
+ */
+#define SAMA5D3_BASE_RTC	0xfffffeb0
+
+/*
  * Internal Memory
  */
 #define SAMA5D3_SRAM_BASE	0x00300000	/* Internal SRAM base address */
diff --git a/arch/arm/mach-at91/sama5d3.c b/arch/arm/mach-at91/sama5d3.c
index 401279715ab1..3ea86428ee09 100644
--- a/arch/arm/mach-at91/sama5d3.c
+++ b/arch/arm/mach-at91/sama5d3.c
@@ -371,7 +371,13 @@ static void __init sama5d3_map_io(void)
 	at91_init_sram(0, SAMA5D3_SRAM_BASE, SAMA5D3_SRAM_SIZE);
 }
 
+static void __init sama5d3_initialize(void)
+{
+	at91_sysirq_mask_rtc(SAMA5D3_BASE_RTC);
+}
+
 AT91_SOC_START(sama5d3)
 	.map_io = sama5d3_map_io,
 	.register_clocks = sama5d3_register_clocks,
+	.init = sama5d3_initialize,
 AT91_SOC_END
diff --git a/arch/arm/mach-at91/sysirq_mask.c b/arch/arm/mach-at91/sysirq_mask.c
new file mode 100644
index 000000000000..2ba694f9626b
--- /dev/null
+++ b/arch/arm/mach-at91/sysirq_mask.c
@@ -0,0 +1,71 @@
+/*
+ * sysirq_mask.c - System-interrupt masking
+ *
+ * Copyright (C) 2013 Johan Hovold <jhovold@gmail.com>
+ *
+ * Functions to disable system interrupts from backup-powered peripherals.
+ *
+ * The RTC and RTT-peripherals are generally powered by backup power (VDDBU)
+ * and are not reset on wake-up, user, watchdog or software reset. This means
+ * that their interrupts may be enabled during early boot (e.g. after a user
+ * reset).
+ *
+ * As the RTC and RTT share the system-interrupt line with the PIT, an
+ * interrupt occurring before a handler has been installed would lead to the
+ * system interrupt being disabled and prevent the system from booting.
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation; either version 2 of the License, or
+ * (at your option) any later version.
+ */
+
+#include <linux/io.h>
+#include <mach/at91_rtt.h>
+
+#include "generic.h"
+
+#define AT91_RTC_IDR	0x24	/* Interrupt Disable Register */
+#define AT91_RTC_IMR	0x28	/* Interrupt Mask Register */
+
+void __init at91_sysirq_mask_rtc(u32 rtc_base)
+{
+	void __iomem *base;
+	u32 mask;
+
+	base = ioremap(rtc_base, 64);
+	if (!base)
+		return;
+
+	mask = readl_relaxed(base + AT91_RTC_IMR);
+	if (mask) {
+		pr_info("AT91: Disabling rtc irq\n");
+		writel_relaxed(mask, base + AT91_RTC_IDR);
+		(void)readl_relaxed(base + AT91_RTC_IMR);	/* flush */
+	}
+
+	iounmap(base);
+}
+
+void __init at91_sysirq_mask_rtt(u32 rtt_base)
+{
+	void __iomem *base;
+	void __iomem *reg;
+	u32 mode;
+
+	base = ioremap(rtt_base, 16);
+	if (!base)
+		return;
+
+	reg = base + AT91_RTT_MR;
+
+	mode = readl_relaxed(reg);
+	if (mode & (AT91_RTT_ALMIEN | AT91_RTT_RTTINCIEN)) {
+		pr_info("AT91: Disabling rtt irq\n");
+		mode &= ~(AT91_RTT_ALMIEN | AT91_RTT_RTTINCIEN);
+		writel_relaxed(mode, reg);
+		(void)readl_relaxed(reg);			/* flush */
+	}
+
+	iounmap(base);
+}
diff --git a/arch/arm/mach-davinci/board-da830-evm.c b/arch/arm/mach-davinci/board-da830-evm.c
index 40f15f133c55..d1f45af7a530 100644
--- a/arch/arm/mach-davinci/board-da830-evm.c
+++ b/arch/arm/mach-davinci/board-da830-evm.c
@@ -17,7 +17,7 @@
 #include <linux/platform_device.h>
 #include <linux/i2c.h>
 #include <linux/i2c/pcf857x.h>
-#include <linux/i2c/at24.h>
+#include <linux/platform_data/at24.h>
 #include <linux/mtd/mtd.h>
 #include <linux/mtd/partitions.h>
 #include <linux/spi/spi.h>
diff --git a/arch/arm/mach-davinci/board-da850-evm.c b/arch/arm/mach-davinci/board-da850-evm.c
index df16cb88a26b..e0af0eccde8f 100644
--- a/arch/arm/mach-davinci/board-da850-evm.c
+++ b/arch/arm/mach-davinci/board-da850-evm.c
@@ -18,7 +18,7 @@
 #include <linux/init.h>
 #include <linux/kernel.h>
 #include <linux/i2c.h>
-#include <linux/i2c/at24.h>
+#include <linux/platform_data/at24.h>
 #include <linux/platform_data/pca953x.h>
 #include <linux/input.h>
 #include <linux/input/tps6507x-ts.h>
diff --git a/arch/arm/mach-davinci/board-dm365-evm.c b/arch/arm/mach-davinci/board-dm365-evm.c
index f4a6c18912ea..e08a8684ead2 100644
--- a/arch/arm/mach-davinci/board-dm365-evm.c
+++ b/arch/arm/mach-davinci/board-dm365-evm.c
@@ -18,7 +18,7 @@
 #include <linux/i2c.h>
 #include <linux/io.h>
 #include <linux/clk.h>
-#include <linux/i2c/at24.h>
+#include <linux/platform_data/at24.h>
 #include <linux/leds.h>
 #include <linux/mtd/mtd.h>
 #include <linux/mtd/partitions.h>
diff --git a/arch/arm/mach-davinci/board-dm644x-evm.c b/arch/arm/mach-davinci/board-dm644x-evm.c
index 9cc32c283b8b..987605b78556 100644
--- a/arch/arm/mach-davinci/board-dm644x-evm.c
+++ b/arch/arm/mach-davinci/board-dm644x-evm.c
@@ -15,7 +15,7 @@
 #include <linux/gpio.h>
 #include <linux/i2c.h>
 #include <linux/i2c/pcf857x.h>
-#include <linux/i2c/at24.h>
+#include <linux/platform_data/at24.h>
 #include <linux/mtd/mtd.h>
 #include <linux/mtd/nand.h>
 #include <linux/mtd/partitions.h>
diff --git a/arch/arm/mach-davinci/board-dm646x-evm.c b/arch/arm/mach-davinci/board-dm646x-evm.c
index 44b20191a9fe..13d0801fd6b1 100644
--- a/arch/arm/mach-davinci/board-dm646x-evm.c
+++ b/arch/arm/mach-davinci/board-dm646x-evm.c
@@ -22,7 +22,7 @@
 #include <linux/gpio.h>
 #include <linux/platform_device.h>
 #include <linux/i2c.h>
-#include <linux/i2c/at24.h>
+#include <linux/platform_data/at24.h>
 #include <linux/i2c/pcf857x.h>
 
 #include <media/tvp514x.h>
diff --git a/arch/arm/mach-davinci/board-mityomapl138.c b/arch/arm/mach-davinci/board-mityomapl138.c
index cd0f58730c2b..7aa105b1fd0f 100644
--- a/arch/arm/mach-davinci/board-mityomapl138.c
+++ b/arch/arm/mach-davinci/board-mityomapl138.c
@@ -15,7 +15,7 @@
 #include <linux/mtd/partitions.h>
 #include <linux/regulator/machine.h>
 #include <linux/i2c.h>
-#include <linux/i2c/at24.h>
+#include <linux/platform_data/at24.h>
 #include <linux/etherdevice.h>
 #include <linux/spi/spi.h>
 #include <linux/spi/flash.h>
diff --git a/arch/arm/mach-davinci/board-sffsdr.c b/arch/arm/mach-davinci/board-sffsdr.c
index d84360148100..41c7c9615791 100644
--- a/arch/arm/mach-davinci/board-sffsdr.c
+++ b/arch/arm/mach-davinci/board-sffsdr.c
@@ -26,7 +26,7 @@
 #include <linux/init.h>
 #include <linux/platform_device.h>
 #include <linux/i2c.h>
-#include <linux/i2c/at24.h>
+#include <linux/platform_data/at24.h>
 #include <linux/mtd/mtd.h>
 #include <linux/mtd/nand.h>
 #include <linux/mtd/partitions.h>
diff --git a/arch/arm/mach-highbank/Kconfig b/arch/arm/mach-highbank/Kconfig
index 08332d841440..0aded64a9ebc 100644
--- a/arch/arm/mach-highbank/Kconfig
+++ b/arch/arm/mach-highbank/Kconfig
@@ -7,7 +7,7 @@ config ARCH_HIGHBANK
 	select ARCH_SUPPORTS_BIG_ENDIAN
 	select ARCH_WANT_OPTIONAL_GPIOLIB
 	select ARM_AMBA
-	select ARM_ERRATA_764369
+	select ARM_ERRATA_764369 if SMP
 	select ARM_ERRATA_775420
 	select ARM_ERRATA_798181 if SMP
 	select ARM_GIC
diff --git a/arch/arm/mach-imx/Makefile b/arch/arm/mach-imx/Makefile
index bbe1f5bb799c..1789e2b31903 100644
--- a/arch/arm/mach-imx/Makefile
+++ b/arch/arm/mach-imx/Makefile
@@ -102,8 +102,8 @@ obj-$(CONFIG_SOC_IMX6SL) += clk-imx6sl.o mach-imx6sl.o
 
 ifeq ($(CONFIG_PM),y)
 obj-$(CONFIG_SOC_IMX6Q) += pm-imx6q.o headsmp.o
-# i.MX6SL reuses pm-imx6q.c
-obj-$(CONFIG_SOC_IMX6SL) += pm-imx6q.o
+# i.MX6SL reuses i.MX6Q code
+obj-$(CONFIG_SOC_IMX6SL) += pm-imx6q.o headsmp.o
 endif
 
 # i.MX5 based machines
diff --git a/arch/arm/mach-imx/clk-imx6q.c b/arch/arm/mach-imx/clk-imx6q.c
index d756d91fd741..04cfd0fcb0e5 100644
--- a/arch/arm/mach-imx/clk-imx6q.c
+++ b/arch/arm/mach-imx/clk-imx6q.c
@@ -122,13 +122,14 @@ static struct clk_div_table clk_enet_ref_table[] = {
 	{ .val = 1, .div = 10, },
 	{ .val = 2, .div = 5, },
 	{ .val = 3, .div = 4, },
+	{ /* sentinel */ }
 };
 
 static struct clk_div_table post_div_table[] = {
 	{ .val = 2, .div = 1, },
 	{ .val = 1, .div = 2, },
 	{ .val = 0, .div = 4, },
-	{ }
+	{ /* sentinel */ }
 };
 
 static struct clk_div_table video_div_table[] = {
@@ -136,7 +137,7 @@ static struct clk_div_table video_div_table[] = {
 	{ .val = 1, .div = 2, },
 	{ .val = 2, .div = 1, },
 	{ .val = 3, .div = 4, },
-	{ }
+	{ /* sentinel */ }
 };
 
 static void __init imx6q_clocks_init(struct device_node *ccm_node)
@@ -298,7 +299,7 @@ static void __init imx6q_clocks_init(struct device_node *ccm_node)
 	clk[asrc_podf]        = imx_clk_divider("asrc_podf",        "asrc_pred",         base + 0x30, 9,  3);
 	clk[spdif_pred]       = imx_clk_divider("spdif_pred",       "spdif_sel",         base + 0x30, 25, 3);
 	clk[spdif_podf]       = imx_clk_divider("spdif_podf",       "spdif_pred",        base + 0x30, 22, 3);
-	clk[can_root]         = imx_clk_divider("can_root",         "pll3_usb_otg",      base + 0x20, 2,  6);
+	clk[can_root]         = imx_clk_divider("can_root",         "pll3_60m",          base + 0x20, 2,  6);
 	clk[ecspi_root]       = imx_clk_divider("ecspi_root",       "pll3_60m",          base + 0x38, 19, 6);
 	clk[gpu2d_core_podf]  = imx_clk_divider("gpu2d_core_podf",  "gpu2d_core_sel",    base + 0x18, 23, 3);
 	clk[gpu3d_core_podf]  = imx_clk_divider("gpu3d_core_podf",  "gpu3d_core_sel",    base + 0x18, 26, 3);
diff --git a/arch/arm/mach-imx/clk-pllv3.c b/arch/arm/mach-imx/clk-pllv3.c
index f6640b6a7b31..61364050fccd 100644
--- a/arch/arm/mach-imx/clk-pllv3.c
+++ b/arch/arm/mach-imx/clk-pllv3.c
@@ -12,6 +12,7 @@
 
 #include <linux/clk.h>
 #include <linux/clk-provider.h>
+#include <linux/delay.h>
 #include <linux/io.h>
 #include <linux/slab.h>
 #include <linux/jiffies.h>
@@ -45,33 +46,49 @@ struct clk_pllv3 {
 
 #define to_clk_pllv3(_hw) container_of(_hw, struct clk_pllv3, hw)
 
+static int clk_pllv3_wait_lock(struct clk_pllv3 *pll)
+{
+	unsigned long timeout = jiffies + msecs_to_jiffies(10);
+	u32 val = readl_relaxed(pll->base) & BM_PLL_POWER;
+
+	/* No need to wait for lock when pll is not powered up */
+	if ((pll->powerup_set && !val) || (!pll->powerup_set && val))
+		return 0;
+
+	/* Wait for PLL to lock */
+	do {
+		if (readl_relaxed(pll->base) & BM_PLL_LOCK)
+			break;
+		if (time_after(jiffies, timeout))
+			break;
+		usleep_range(50, 500);
+	} while (1);
+
+	return readl_relaxed(pll->base) & BM_PLL_LOCK ? 0 : -ETIMEDOUT;
+}
+
 static int clk_pllv3_prepare(struct clk_hw *hw)
 {
 	struct clk_pllv3 *pll = to_clk_pllv3(hw);
-	unsigned long timeout;
 	u32 val;
+	int ret;
 
 	val = readl_relaxed(pll->base);
-	val &= ~BM_PLL_BYPASS;
 	if (pll->powerup_set)
 		val |= BM_PLL_POWER;
 	else
 		val &= ~BM_PLL_POWER;
 	writel_relaxed(val, pll->base);
 
-	timeout = jiffies + msecs_to_jiffies(10);
-	/* Wait for PLL to lock */
-	do {
-		if (readl_relaxed(pll->base) & BM_PLL_LOCK)
-			break;
-		if (time_after(jiffies, timeout))
-			break;
-	} while (1);
+	ret = clk_pllv3_wait_lock(pll);
+	if (ret)
+		return ret;
 
-	if (readl_relaxed(pll->base) & BM_PLL_LOCK)
-		return 0;
-	else
-		return -ETIMEDOUT;
+	val = readl_relaxed(pll->base);
+	val &= ~BM_PLL_BYPASS;
+	writel_relaxed(val, pll->base);
+
+	return 0;
 }
 
 static void clk_pllv3_unprepare(struct clk_hw *hw)
@@ -146,7 +163,7 @@ static int clk_pllv3_set_rate(struct clk_hw *hw, unsigned long rate,
 	val |= div;
 	writel_relaxed(val, pll->base);
 
-	return 0;
+	return clk_pllv3_wait_lock(pll);
 }
 
 static const struct clk_ops clk_pllv3_ops = {
@@ -202,7 +219,7 @@ static int clk_pllv3_sys_set_rate(struct clk_hw *hw, unsigned long rate,
 	val |= div;
 	writel_relaxed(val, pll->base);
 
-	return 0;
+	return clk_pllv3_wait_lock(pll);
 }
 
 static const struct clk_ops clk_pllv3_sys_ops = {
@@ -276,7 +293,7 @@ static int clk_pllv3_av_set_rate(struct clk_hw *hw, unsigned long rate,
 	writel_relaxed(mfn, pll->base + PLL_NUM_OFFSET);
 	writel_relaxed(mfd, pll->base + PLL_DENOM_OFFSET);
 
-	return 0;
+	return clk_pllv3_wait_lock(pll);
 }
 
 static const struct clk_ops clk_pllv3_av_ops = {
diff --git a/arch/arm/mach-imx/common.h b/arch/arm/mach-imx/common.h
index 7cbe22d0c6e9..24a7899e36a8 100644
--- a/arch/arm/mach-imx/common.h
+++ b/arch/arm/mach-imx/common.h
@@ -127,11 +127,6 @@ static inline void imx_smp_prepare(void) {}
 static inline void imx_scu_standby_enable(void) {}
 #endif
 void imx_src_init(void);
-#ifdef CONFIG_HAVE_IMX_SRC
-void imx_src_prepare_restart(void);
-#else
-static inline void imx_src_prepare_restart(void) {}
-#endif
 void imx_gpc_init(void);
 void imx_gpc_pre_suspend(void);
 void imx_gpc_post_resume(void);
diff --git a/arch/arm/mach-imx/mach-pca100.c b/arch/arm/mach-imx/mach-pca100.c
index 19bb6441a7d4..c5f95674e9b7 100644
--- a/arch/arm/mach-imx/mach-pca100.c
+++ b/arch/arm/mach-imx/mach-pca100.c
@@ -20,7 +20,7 @@
 #include <linux/platform_device.h>
 #include <linux/io.h>
 #include <linux/i2c.h>
-#include <linux/i2c/at24.h>
+#include <linux/platform_data/at24.h>
 #include <linux/dma-mapping.h>
 #include <linux/spi/spi.h>
 #include <linux/spi/eeprom.h>
diff --git a/arch/arm/mach-imx/mach-pcm037.c b/arch/arm/mach-imx/mach-pcm037.c
index 45303bd62902..639a3dfb0092 100644
--- a/arch/arm/mach-imx/mach-pcm037.c
+++ b/arch/arm/mach-imx/mach-pcm037.c
@@ -23,7 +23,7 @@
 #include <linux/smsc911x.h>
 #include <linux/interrupt.h>
 #include <linux/i2c.h>
-#include <linux/i2c/at24.h>
+#include <linux/platform_data/at24.h>
 #include <linux/delay.h>
 #include <linux/spi/spi.h>
 #include <linux/irq.h>
diff --git a/arch/arm/mach-imx/mach-pcm038.c b/arch/arm/mach-imx/mach-pcm038.c
index e805ac273e9c..592ddbe031ac 100644
--- a/arch/arm/mach-imx/mach-pcm038.c
+++ b/arch/arm/mach-imx/mach-pcm038.c
@@ -18,7 +18,7 @@
  */
 
 #include <linux/i2c.h>
-#include <linux/i2c/at24.h>
+#include <linux/platform_data/at24.h>
 #include <linux/io.h>
 #include <linux/mtd/plat-ram.h>
 #include <linux/mtd/physmap.h>
diff --git a/arch/arm/mach-imx/mach-pcm043.c b/arch/arm/mach-imx/mach-pcm043.c
index b726cb1c5fdd..ac504b67326b 100644
--- a/arch/arm/mach-imx/mach-pcm043.c
+++ b/arch/arm/mach-imx/mach-pcm043.c
@@ -24,7 +24,7 @@
 #include <linux/interrupt.h>
 #include <linux/delay.h>
 #include <linux/i2c.h>
-#include <linux/i2c/at24.h>
+#include <linux/platform_data/at24.h>
 #include <linux/usb/otg.h>
 #include <linux/usb/ulpi.h>
 
diff --git a/arch/arm/mach-imx/mach-vpr200.c b/arch/arm/mach-imx/mach-vpr200.c
index 0910761e8280..8825d1217d18 100644
--- a/arch/arm/mach-imx/mach-vpr200.c
+++ b/arch/arm/mach-imx/mach-vpr200.c
@@ -29,7 +29,7 @@
 #include <asm/mach/time.h>
 
 #include <linux/i2c.h>
-#include <linux/i2c/at24.h>
+#include <linux/platform_data/at24.h>
 #include <linux/mfd/mc13xxx.h>
 
 #include "common.h"
diff --git a/arch/arm/mach-imx/src.c b/arch/arm/mach-imx/src.c
index 4754373e7e7d..45f7f4e0a447 100644
--- a/arch/arm/mach-imx/src.c
+++ b/arch/arm/mach-imx/src.c
@@ -115,21 +115,6 @@ void imx_set_cpu_arg(int cpu, u32 arg)
 	writel_relaxed(arg, src_base + SRC_GPR1 + cpu * 8 + 4);
 }
 
-void imx_src_prepare_restart(void)
-{
-	u32 val;
-
-	/* clear enable bits of secondary cores */
-	spin_lock(&scr_lock);
-	val = readl_relaxed(src_base + SRC_SCR);
-	val &= ~(0x7 << BP_SRC_SCR_CORE1_ENABLE);
-	writel_relaxed(val, src_base + SRC_SCR);
-	spin_unlock(&scr_lock);
-
-	/* clear persistent entry register of primary core */
-	writel_relaxed(0, src_base + SRC_GPR1);
-}
-
 void __init imx_src_init(void)
 {
 	struct device_node *np;
diff --git a/arch/arm/mach-imx/system.c b/arch/arm/mach-imx/system.c
index e6edcd38b282..5e3027d3692f 100644
--- a/arch/arm/mach-imx/system.c
+++ b/arch/arm/mach-imx/system.c
@@ -42,9 +42,6 @@ void mxc_restart(enum reboot_mode mode, const char *cmd)
 {
 	unsigned int wcr_enable;
 
-	if (cpu_is_imx6q() || cpu_is_imx6dl())
-		imx_src_prepare_restart();
-
 	if (wdog_clk)
 		clk_enable(wdog_clk);
 
@@ -55,7 +52,14 @@ void mxc_restart(enum reboot_mode mode, const char *cmd)
 
 	/* Assert SRS signal */
 	__raw_writew(wcr_enable, wdog_base);
-	/* write twice to ensure the request will not get ignored */
+	/*
+	 * Due to imx6q errata ERR004346 (WDOG: WDOG SRS bit requires to be
+	 * written twice), we add another two writes to ensure there must be at
+	 * least two writes happen in the same one 32kHz clock period.  We save
+	 * the target check here, since the writes shouldn't be a huge burden
+	 * for other platforms.
+	 */
+	__raw_writew(wcr_enable, wdog_base);
 	__raw_writew(wcr_enable, wdog_base);
 
 	/* wait for reset to assert... */
diff --git a/arch/arm/mach-integrator/integrator_cp.c b/arch/arm/mach-integrator/integrator_cp.c
index 1df6e7602cad..4fc0a195de01 100644
--- a/arch/arm/mach-integrator/integrator_cp.c
+++ b/arch/arm/mach-integrator/integrator_cp.c
@@ -198,7 +198,8 @@ static struct mmci_platform_data mmc_data = {
 static void cp_clcd_enable(struct clcd_fb *fb)
 {
 	struct fb_var_screeninfo *var = &fb->fb.var;
-	u32 val = CM_CTRL_STATIC1 | CM_CTRL_STATIC2;
+	u32 val = CM_CTRL_STATIC1 | CM_CTRL_STATIC2
+			| CM_CTRL_LCDEN0 | CM_CTRL_LCDEN1;
 
 	if (var->bits_per_pixel <= 8 ||
 	    (var->bits_per_pixel == 16 && var->green.length == 5))
diff --git a/arch/arm/mach-iop13xx/include/mach/adma.h b/arch/arm/mach-iop13xx/include/mach/adma.h
index 6d3782d85a9f..a86fd0ed7757 100644
--- a/arch/arm/mach-iop13xx/include/mach/adma.h
+++ b/arch/arm/mach-iop13xx/include/mach/adma.h
@@ -218,20 +218,6 @@ iop_chan_xor_slot_count(size_t len, int src_cnt, int *slots_per_op)
 #define iop_chan_pq_slot_count iop_chan_xor_slot_count
 #define iop_chan_pq_zero_sum_slot_count iop_chan_xor_slot_count
 
-static inline u32 iop_desc_get_dest_addr(struct iop_adma_desc_slot *desc,
-					struct iop_adma_chan *chan)
-{
-	struct iop13xx_adma_desc_hw *hw_desc = desc->hw_desc;
-	return hw_desc->dest_addr;
-}
-
-static inline u32 iop_desc_get_qdest_addr(struct iop_adma_desc_slot *desc,
-					  struct iop_adma_chan *chan)
-{
-	struct iop13xx_adma_desc_hw *hw_desc = desc->hw_desc;
-	return hw_desc->q_dest_addr;
-}
-
 static inline u32 iop_desc_get_byte_count(struct iop_adma_desc_slot *desc,
 					struct iop_adma_chan *chan)
 {
@@ -350,18 +336,6 @@ iop_desc_init_pq(struct iop_adma_desc_slot *desc, int src_cnt,
 	hw_desc->desc_ctrl = u_desc_ctrl.value;
 }
 
-static inline int iop_desc_is_pq(struct iop_adma_desc_slot *desc)
-{
-	struct iop13xx_adma_desc_hw *hw_desc = desc->hw_desc;
-	union {
-		u32 value;
-		struct iop13xx_adma_desc_ctrl field;
-	} u_desc_ctrl;
-
-	u_desc_ctrl.value = hw_desc->desc_ctrl;
-	return u_desc_ctrl.field.pq_xfer_en;
-}
-
 static inline void
 iop_desc_init_pq_zero_sum(struct iop_adma_desc_slot *desc, int src_cnt,
 			  unsigned long flags)
diff --git a/arch/arm/mach-kirkwood/lacie_v2-common.c b/arch/arm/mach-kirkwood/lacie_v2-common.c
index 489495976fcd..8e3e4331c380 100644
--- a/arch/arm/mach-kirkwood/lacie_v2-common.c
+++ b/arch/arm/mach-kirkwood/lacie_v2-common.c
@@ -12,7 +12,7 @@
 #include <linux/spi/flash.h>
 #include <linux/spi/spi.h>
 #include <linux/i2c.h>
-#include <linux/i2c/at24.h>
+#include <linux/platform_data/at24.h>
 #include <linux/gpio.h>
 #include <asm/mach/time.h>
 #include <mach/kirkwood.h>
diff --git a/arch/arm/mach-mmp/ttc_dkb.c b/arch/arm/mach-mmp/ttc_dkb.c
index 702232996c8c..cfadd974f5ce 100644
--- a/arch/arm/mach-mmp/ttc_dkb.c
+++ b/arch/arm/mach-mmp/ttc_dkb.c
@@ -191,7 +191,6 @@ static struct pxa3xx_nand_platform_data dkb_nand_info = {
 #define SCLK_SOURCE_SELECT(x)  (x << 30) /* 0x0 ~ 0x3 */
 /* link config */
 #define CFG_DUMBMODE(mode)     (mode << 28) /* 0x0 ~ 0x6*/
-#define CFG_GRA_SWAPRB(x)      (x << 0) /* 1: rbswap enabled */
 static struct mmp_mach_path_config dkb_disp_config[] = {
 	[0] = {
 		.name = "mmp-parallel",
@@ -199,8 +198,7 @@ static struct mmp_mach_path_config dkb_disp_config[] = {
 		.output_type = PATH_OUT_PARALLEL,
 		.path_config = CFG_IOPADMODE(0x1)
 			| SCLK_SOURCE_SELECT(0x1),
-		.link_config = CFG_DUMBMODE(0x2)
-			| CFG_GRA_SWAPRB(0x1),
+		.link_config = CFG_DUMBMODE(0x2),
 	},
 };
 
diff --git a/arch/arm/mach-omap1/board-osk.c b/arch/arm/mach-omap1/board-osk.c
index a7ce69286688..d68909b095f1 100644
--- a/arch/arm/mach-omap1/board-osk.c
+++ b/arch/arm/mach-omap1/board-osk.c
@@ -300,7 +300,7 @@ static struct omap_lcd_config osk_lcd_config __initdata = {
 #ifdef	CONFIG_OMAP_OSK_MISTRAL
 
 #include <linux/input.h>
-#include <linux/i2c/at24.h>
+#include <linux/platform_data/at24.h>
 #include <linux/spi/spi.h>
 #include <linux/spi/ads7846.h>
 
diff --git a/arch/arm/mach-omap2/Makefile b/arch/arm/mach-omap2/Makefile
index e15ac005ef17..1f25f3e99c05 100644
--- a/arch/arm/mach-omap2/Makefile
+++ b/arch/arm/mach-omap2/Makefile
@@ -40,7 +40,7 @@ omap-4-5-common				=  omap4-common.o omap-wakeupgen.o
 obj-$(CONFIG_ARCH_OMAP4)		+= $(omap-4-5-common) $(smp-y) sleep44xx.o
 obj-$(CONFIG_SOC_OMAP5)			+= $(omap-4-5-common) $(smp-y) sleep44xx.o
 obj-$(CONFIG_SOC_AM43XX)		+= $(omap-4-5-common)
-obj-$(CONFIG_SOC_DRA7XX)		+= $(omap-4-5-common) $(smp-y)
+obj-$(CONFIG_SOC_DRA7XX)		+= $(omap-4-5-common) $(smp-y) sleep44xx.o
 
 plus_sec := $(call as-instr,.arch_extension sec,+sec)
 AFLAGS_omap-headsmp.o			:=-Wa,-march=armv7-a$(plus_sec)
diff --git a/arch/arm/mach-omap2/board-cm-t35.c b/arch/arm/mach-omap2/board-cm-t35.c
index 33d159e2386e..8dd0ec858cf1 100644
--- a/arch/arm/mach-omap2/board-cm-t35.c
+++ b/arch/arm/mach-omap2/board-cm-t35.c
@@ -25,7 +25,7 @@
 #include <linux/gpio.h>
 #include <linux/platform_data/gpio-omap.h>
 
-#include <linux/i2c/at24.h>
+#include <linux/platform_data/at24.h>
 #include <linux/i2c/twl.h>
 #include <linux/regulator/fixed.h>
 #include <linux/regulator/machine.h>
diff --git a/arch/arm/mach-omap2/board-h4.c b/arch/arm/mach-omap2/board-h4.c
index 87e41a8b8d46..f7808349a734 100644
--- a/arch/arm/mach-omap2/board-h4.c
+++ b/arch/arm/mach-omap2/board-h4.c
@@ -20,7 +20,7 @@
 #include <linux/delay.h>
 #include <linux/workqueue.h>
 #include <linux/i2c.h>
-#include <linux/i2c/at24.h>
+#include <linux/platform_data/at24.h>
 #include <linux/input.h>
 #include <linux/err.h>
 #include <linux/clk.h>
diff --git a/arch/arm/mach-omap2/board-omap3beagle.c b/arch/arm/mach-omap2/board-omap3beagle.c
index a516c1bda141..d6ed819ff15c 100644
--- a/arch/arm/mach-omap2/board-omap3beagle.c
+++ b/arch/arm/mach-omap2/board-omap3beagle.c
@@ -510,7 +510,7 @@ static int __init beagle_opp_init(void)
 		mpu_dev = get_cpu_device(0);
 		iva_dev = omap_device_get_by_hwmod_name("iva");
 
-		if (IS_ERR(mpu_dev) || IS_ERR(iva_dev)) {
+		if (!mpu_dev || IS_ERR(iva_dev)) {
 			pr_err("%s: Aiee.. no mpu/dsp devices? %p %p\n",
 				__func__, mpu_dev, iva_dev);
 			return -ENODEV;
diff --git a/arch/arm/mach-omap2/board-omap3stalker.c b/arch/arm/mach-omap2/board-omap3stalker.c
index ba8342fef799..119efaf5808a 100644
--- a/arch/arm/mach-omap2/board-omap3stalker.c
+++ b/arch/arm/mach-omap2/board-omap3stalker.c
@@ -32,7 +32,7 @@
 #include <linux/spi/spi.h>
 #include <linux/interrupt.h>
 #include <linux/smsc911x.h>
-#include <linux/i2c/at24.h>
+#include <linux/platform_data/at24.h>
 #include <linux/usb/phy.h>
 
 #include <asm/mach-types.h>
diff --git a/arch/arm/mach-omap2/cclock3xxx_data.c b/arch/arm/mach-omap2/cclock3xxx_data.c
index 03a2829beb8e..3b05aea56d1f 100644
--- a/arch/arm/mach-omap2/cclock3xxx_data.c
+++ b/arch/arm/mach-omap2/cclock3xxx_data.c
@@ -381,6 +381,42 @@ static struct clk_hw_omap dpll4_ck_hw = {
 
 DEFINE_STRUCT_CLK(dpll4_ck, dpll3_ck_parent_names, dpll4_ck_ops);
 
+static const struct clk_div_table dpll4_mx_ck_div_table[] = {
+	{ .div = 1, .val = 1 },
+	{ .div = 2, .val = 2 },
+	{ .div = 3, .val = 3 },
+	{ .div = 4, .val = 4 },
+	{ .div = 5, .val = 5 },
+	{ .div = 6, .val = 6 },
+	{ .div = 7, .val = 7 },
+	{ .div = 8, .val = 8 },
+	{ .div = 9, .val = 9 },
+	{ .div = 10, .val = 10 },
+	{ .div = 11, .val = 11 },
+	{ .div = 12, .val = 12 },
+	{ .div = 13, .val = 13 },
+	{ .div = 14, .val = 14 },
+	{ .div = 15, .val = 15 },
+	{ .div = 16, .val = 16 },
+	{ .div = 17, .val = 17 },
+	{ .div = 18, .val = 18 },
+	{ .div = 19, .val = 19 },
+	{ .div = 20, .val = 20 },
+	{ .div = 21, .val = 21 },
+	{ .div = 22, .val = 22 },
+	{ .div = 23, .val = 23 },
+	{ .div = 24, .val = 24 },
+	{ .div = 25, .val = 25 },
+	{ .div = 26, .val = 26 },
+	{ .div = 27, .val = 27 },
+	{ .div = 28, .val = 28 },
+	{ .div = 29, .val = 29 },
+	{ .div = 30, .val = 30 },
+	{ .div = 31, .val = 31 },
+	{ .div = 32, .val = 32 },
+	{ .div = 0 },
+};
+
 DEFINE_CLK_DIVIDER(dpll4_m5_ck, "dpll4_ck", &dpll4_ck, 0x0,
 		   OMAP_CM_REGADDR(OMAP3430_CAM_MOD, CM_CLKSEL),
 		   OMAP3430_CLKSEL_CAM_SHIFT, OMAP3630_CLKSEL_CAM_WIDTH,
@@ -524,10 +560,10 @@ static const struct clksel_rate clkout2_src_54m_rates[] = {
 	{ .div = 0 }
 };
 
-DEFINE_CLK_DIVIDER(dpll4_m3_ck, "dpll4_ck", &dpll4_ck, 0x0,
+DEFINE_CLK_DIVIDER_TABLE(dpll4_m3_ck, "dpll4_ck", &dpll4_ck, 0x0,
 		   OMAP_CM_REGADDR(OMAP3430_DSS_MOD, CM_CLKSEL),
 		   OMAP3430_CLKSEL_TV_SHIFT, OMAP3630_CLKSEL_TV_WIDTH,
-		   CLK_DIVIDER_ONE_BASED, NULL);
+		   0, dpll4_mx_ck_div_table, NULL);
 
 static struct clk dpll4_m3x2_ck;
 
@@ -847,10 +883,10 @@ static struct clk dpll3_m3x2_ck_3630 = {
 
 DEFINE_CLK_FIXED_FACTOR(dpll3_x2_ck, "dpll3_ck", &dpll3_ck, 0x0, 2, 1);
 
-DEFINE_CLK_DIVIDER(dpll4_m4_ck, "dpll4_ck", &dpll4_ck, 0x0,
+DEFINE_CLK_DIVIDER_TABLE(dpll4_m4_ck, "dpll4_ck", &dpll4_ck, 0x0,
 		   OMAP_CM_REGADDR(OMAP3430_DSS_MOD, CM_CLKSEL),
 		   OMAP3430_CLKSEL_DSS1_SHIFT, OMAP3630_CLKSEL_DSS1_WIDTH,
-		   CLK_DIVIDER_ONE_BASED, NULL);
+		   0, dpll4_mx_ck_div_table, NULL);
 
 static struct clk dpll4_m4x2_ck;
 
@@ -869,7 +905,8 @@ static struct clk_hw_omap dpll4_m4x2_ck_hw = {
 	.clkdm_name	= "dpll4_clkdm",
 };
 
-DEFINE_STRUCT_CLK(dpll4_m4x2_ck, dpll4_m4x2_ck_parent_names, dpll4_m5x2_ck_ops);
+DEFINE_STRUCT_CLK_FLAGS(dpll4_m4x2_ck, dpll4_m4x2_ck_parent_names,
+		dpll4_m5x2_ck_ops, CLK_SET_RATE_PARENT);
 
 static struct clk dpll4_m4x2_ck_3630 = {
 	.name		= "dpll4_m4x2_ck",
@@ -877,6 +914,7 @@ static struct clk dpll4_m4x2_ck_3630 = {
 	.parent_names	= dpll4_m4x2_ck_parent_names,
 	.num_parents	= ARRAY_SIZE(dpll4_m4x2_ck_parent_names),
 	.ops		= &dpll4_m5x2_ck_3630_ops,
+	.flags		= CLK_SET_RATE_PARENT,
 };
 
 DEFINE_CLK_DIVIDER(dpll4_m6_ck, "dpll4_ck", &dpll4_ck, 0x0,
@@ -968,8 +1006,9 @@ static struct clk_hw_omap dss1_alwon_fck_3430es1_hw = {
 	.clkdm_name	= "dss_clkdm",
 };
 
-DEFINE_STRUCT_CLK(dss1_alwon_fck_3430es1, dss1_alwon_fck_3430es1_parent_names,
-		  aes2_ick_ops);
+DEFINE_STRUCT_CLK_FLAGS(dss1_alwon_fck_3430es1,
+		dss1_alwon_fck_3430es1_parent_names, aes2_ick_ops,
+		CLK_SET_RATE_PARENT);
 
 static struct clk dss1_alwon_fck_3430es2;
 
@@ -983,8 +1022,9 @@ static struct clk_hw_omap dss1_alwon_fck_3430es2_hw = {
 	.clkdm_name	= "dss_clkdm",
 };
 
-DEFINE_STRUCT_CLK(dss1_alwon_fck_3430es2, dss1_alwon_fck_3430es1_parent_names,
-		  aes2_ick_ops);
+DEFINE_STRUCT_CLK_FLAGS(dss1_alwon_fck_3430es2,
+		dss1_alwon_fck_3430es1_parent_names, aes2_ick_ops,
+		CLK_SET_RATE_PARENT);
 
 static struct clk dss2_alwon_fck;
 
diff --git a/arch/arm/mach-omap2/cclock44xx_data.c b/arch/arm/mach-omap2/cclock44xx_data.c
index b237950eb8a3..ec0dc0b1755e 100644
--- a/arch/arm/mach-omap2/cclock44xx_data.c
+++ b/arch/arm/mach-omap2/cclock44xx_data.c
@@ -830,7 +830,8 @@ DEFINE_CLK_GATE(dss_tv_clk, "extalt_clkin_ck", &extalt_clkin_ck, 0x0,
 		OMAP4430_CM_DSS_DSS_CLKCTRL,
 		OMAP4430_OPTFCLKEN_TV_CLK_SHIFT, 0x0, NULL);
 
-DEFINE_CLK_GATE(dss_dss_clk, "dpll_per_m5x2_ck", &dpll_per_m5x2_ck, 0x0,
+DEFINE_CLK_GATE(dss_dss_clk, "dpll_per_m5x2_ck", &dpll_per_m5x2_ck,
+		CLK_SET_RATE_PARENT,
 		OMAP4430_CM_DSS_DSS_CLKCTRL, OMAP4430_OPTFCLKEN_DSSCLK_SHIFT,
 		0x0, NULL);
 
diff --git a/arch/arm/mach-omap2/gpmc-smsc911x.c b/arch/arm/mach-omap2/gpmc-smsc911x.c
index ef990118d32b..2757504a13c4 100644
--- a/arch/arm/mach-omap2/gpmc-smsc911x.c
+++ b/arch/arm/mach-omap2/gpmc-smsc911x.c
@@ -83,7 +83,7 @@ void __init gpmc_smsc911x_init(struct omap_smsc911x_platform_data *gpmc_cfg)
 	pdev = platform_device_register_resndata(NULL, "smsc911x", gpmc_cfg->id,
 		 gpmc_smsc911x_resources, ARRAY_SIZE(gpmc_smsc911x_resources),
 		 &gpmc_smsc911x_config, sizeof(gpmc_smsc911x_config));
-	if (!pdev) {
+	if (IS_ERR(pdev)) {
 		pr_err("Unable to register platform device\n");
 		gpio_free(gpmc_cfg->gpio_reset);
 		goto free2;
diff --git a/arch/arm/mach-omap2/omap_device.c b/arch/arm/mach-omap2/omap_device.c
index b69dd9abb50a..53f0735817bb 100644
--- a/arch/arm/mach-omap2/omap_device.c
+++ b/arch/arm/mach-omap2/omap_device.c
@@ -621,6 +621,7 @@ static int _od_suspend_noirq(struct device *dev)
 
 	if (!ret && !pm_runtime_status_suspended(dev)) {
 		if (pm_generic_runtime_suspend(dev) == 0) {
+			pm_runtime_set_suspended(dev);
 			omap_device_idle(pdev);
 			od->flags |= OMAP_DEVICE_SUSPENDED;
 		}
@@ -634,10 +635,18 @@ static int _od_resume_noirq(struct device *dev)
 	struct platform_device *pdev = to_platform_device(dev);
 	struct omap_device *od = to_omap_device(pdev);
 
-	if ((od->flags & OMAP_DEVICE_SUSPENDED) &&
-	    !pm_runtime_status_suspended(dev)) {
+	if (od->flags & OMAP_DEVICE_SUSPENDED) {
 		od->flags &= ~OMAP_DEVICE_SUSPENDED;
 		omap_device_enable(pdev);
+		/*
+		 * XXX: we run before core runtime pm has resumed itself. At
+		 * this point in time, we just restore the runtime pm state and
+		 * considering symmetric operations in resume, we donot expect
+		 * to fail. If we failed, something changed in core runtime_pm
+		 * framework OR some device driver messed things up, hence, WARN
+		 */
+		WARN(pm_runtime_set_active(dev),
+		     "Could not set %s runtime state active\n", dev_name(dev));
 		pm_generic_runtime_resume(dev);
 	}
 
diff --git a/arch/arm/mach-omap2/prm44xx_54xx.h b/arch/arm/mach-omap2/prm44xx_54xx.h
index a085d9cc1f5d..7a976065e138 100644
--- a/arch/arm/mach-omap2/prm44xx_54xx.h
+++ b/arch/arm/mach-omap2/prm44xx_54xx.h
@@ -42,7 +42,8 @@ extern u32 omap4_prm_vcvp_read(u8 offset);
 extern void omap4_prm_vcvp_write(u32 val, u8 offset);
 extern u32 omap4_prm_vcvp_rmw(u32 mask, u32 bits, u8 offset);
 
-#if defined(CONFIG_ARCH_OMAP4) || defined(CONFIG_SOC_OMAP5)
+#if defined(CONFIG_ARCH_OMAP4) || defined(CONFIG_SOC_OMAP5) || \
+	defined(CONFIG_SOC_DRA7XX)
 void omap44xx_prm_reconfigure_io_chain(void);
 #else
 static inline void omap44xx_prm_reconfigure_io_chain(void)
diff --git a/arch/arm/mach-pxa/cm-x300.c b/arch/arm/mach-pxa/cm-x300.c
index f9423493ed36..584439bfa59f 100644
--- a/arch/arm/mach-pxa/cm-x300.c
+++ b/arch/arm/mach-pxa/cm-x300.c
@@ -310,6 +310,7 @@ static struct platform_pwm_backlight_data cm_x300_backlight_data = {
 	.max_brightness	= 100,
 	.dft_brightness	= 100,
 	.pwm_period_ns	= 10000,
+	.enable_gpio	= -1,
 };
 
 static struct platform_device cm_x300_backlight_device = {
diff --git a/arch/arm/mach-pxa/colibri-pxa270-income.c b/arch/arm/mach-pxa/colibri-pxa270-income.c
index 2d4a7b4d5d78..3aa264640c9d 100644
--- a/arch/arm/mach-pxa/colibri-pxa270-income.c
+++ b/arch/arm/mach-pxa/colibri-pxa270-income.c
@@ -189,6 +189,7 @@ static struct platform_pwm_backlight_data income_backlight_data = {
 	.max_brightness	= 0x3ff,
 	.dft_brightness	= 0x1ff,
 	.pwm_period_ns	= 1000000,
+	.enable_gpio	= -1,
 };
 
 static struct platform_device income_backlight = {
diff --git a/arch/arm/mach-pxa/ezx.c b/arch/arm/mach-pxa/ezx.c
index fe2eb8394dff..ab93441e596e 100644
--- a/arch/arm/mach-pxa/ezx.c
+++ b/arch/arm/mach-pxa/ezx.c
@@ -54,6 +54,7 @@ static struct platform_pwm_backlight_data ezx_backlight_data = {
 	.max_brightness	= 1023,
 	.dft_brightness	= 1023,
 	.pwm_period_ns	= 78770,
+	.enable_gpio	= -1,
 };
 
 static struct platform_device ezx_backlight_device = {
diff --git a/arch/arm/mach-pxa/hx4700.c b/arch/arm/mach-pxa/hx4700.c
index 133109ec7332..a7c30eb0c8db 100644
--- a/arch/arm/mach-pxa/hx4700.c
+++ b/arch/arm/mach-pxa/hx4700.c
@@ -561,6 +561,7 @@ static struct platform_pwm_backlight_data backlight_data = {
 	.max_brightness = 200,
 	.dft_brightness = 100,
 	.pwm_period_ns  = 30923,
+	.enable_gpio    = -1,
 };
 
 static struct platform_device backlight = {
diff --git a/arch/arm/mach-pxa/lpd270.c b/arch/arm/mach-pxa/lpd270.c
index 1255ee00f3d1..9f6ec167902a 100644
--- a/arch/arm/mach-pxa/lpd270.c
+++ b/arch/arm/mach-pxa/lpd270.c
@@ -269,6 +269,7 @@ static struct platform_pwm_backlight_data lpd270_backlight_data = {
 	.max_brightness	= 1,
 	.dft_brightness	= 1,
 	.pwm_period_ns	= 78770,
+	.enable_gpio	= -1,
 };
 
 static struct platform_device lpd270_backlight_device = {
diff --git a/arch/arm/mach-pxa/magician.c b/arch/arm/mach-pxa/magician.c
index f44532fc648b..fab30d666cc7 100644
--- a/arch/arm/mach-pxa/magician.c
+++ b/arch/arm/mach-pxa/magician.c
@@ -378,6 +378,7 @@ static struct platform_pwm_backlight_data backlight_data = {
 	.max_brightness = 272,
 	.dft_brightness = 100,
 	.pwm_period_ns  = 30923,
+	.enable_gpio    = -1,
 	.init           = magician_backlight_init,
 	.notify         = magician_backlight_notify,
 	.exit           = magician_backlight_exit,
diff --git a/arch/arm/mach-pxa/mainstone.c b/arch/arm/mach-pxa/mainstone.c
index dd70343c8708..08ccc0718f31 100644
--- a/arch/arm/mach-pxa/mainstone.c
+++ b/arch/arm/mach-pxa/mainstone.c
@@ -338,6 +338,7 @@ static struct platform_pwm_backlight_data mainstone_backlight_data = {
 	.max_brightness	= 1023,
 	.dft_brightness	= 1023,
 	.pwm_period_ns	= 78770,
+	.enable_gpio	= -1,
 };
 
 static struct platform_device mainstone_backlight_device = {
diff --git a/arch/arm/mach-pxa/mioa701.c b/arch/arm/mach-pxa/mioa701.c
index acc9d3cc0762..f70583fee59f 100644
--- a/arch/arm/mach-pxa/mioa701.c
+++ b/arch/arm/mach-pxa/mioa701.c
@@ -186,6 +186,7 @@ static struct platform_pwm_backlight_data mioa701_backlight_data = {
 	.max_brightness	= 100,
 	.dft_brightness	= 50,
 	.pwm_period_ns	= 4000 * 1024,	/* Fl = 250kHz */
+	.enable_gpio	= -1,
 };
 
 /*
diff --git a/arch/arm/mach-pxa/palm27x.c b/arch/arm/mach-pxa/palm27x.c
index 17d4c53017ca..e54a296fb81f 100644
--- a/arch/arm/mach-pxa/palm27x.c
+++ b/arch/arm/mach-pxa/palm27x.c
@@ -322,6 +322,7 @@ static struct platform_pwm_backlight_data palm27x_backlight_data = {
 	.max_brightness	= 0xfe,
 	.dft_brightness	= 0x7e,
 	.pwm_period_ns	= 3500 * 1024,
+	.enable_gpio	= -1,
 	.init		= palm27x_backlight_init,
 	.notify		= palm27x_backlight_notify,
 	.exit		= palm27x_backlight_exit,
diff --git a/arch/arm/mach-pxa/palmtc.c b/arch/arm/mach-pxa/palmtc.c
index 100b176f7e88..7691c974ca4b 100644
--- a/arch/arm/mach-pxa/palmtc.c
+++ b/arch/arm/mach-pxa/palmtc.c
@@ -166,45 +166,12 @@ static inline void palmtc_keys_init(void) {}
  * Backlight
  ******************************************************************************/
 #if defined(CONFIG_BACKLIGHT_PWM) || defined(CONFIG_BACKLIGHT_PWM_MODULE)
-static int palmtc_backlight_init(struct device *dev)
-{
-	int ret;
-
-	ret = gpio_request(GPIO_NR_PALMTC_BL_POWER, "BL POWER");
-	if (ret)
-		goto err;
-	ret = gpio_direction_output(GPIO_NR_PALMTC_BL_POWER, 1);
-	if (ret)
-		goto err2;
-
-	return 0;
-
-err2:
-	gpio_free(GPIO_NR_PALMTC_BL_POWER);
-err:
-	return ret;
-}
-
-static int palmtc_backlight_notify(struct device *dev, int brightness)
-{
-	/* backlight is on when GPIO16 AF0 is high */
-	gpio_set_value(GPIO_NR_PALMTC_BL_POWER, brightness);
-	return brightness;
-}
-
-static void palmtc_backlight_exit(struct device *dev)
-{
-	gpio_free(GPIO_NR_PALMTC_BL_POWER);
-}
-
 static struct platform_pwm_backlight_data palmtc_backlight_data = {
 	.pwm_id		= 1,
 	.max_brightness	= PALMTC_MAX_INTENSITY,
 	.dft_brightness	= PALMTC_MAX_INTENSITY,
 	.pwm_period_ns	= PALMTC_PERIOD_NS,
-	.init		= palmtc_backlight_init,
-	.notify		= palmtc_backlight_notify,
-	.exit		= palmtc_backlight_exit,
+	.enable_gpio	= GPIO_NR_PALMTC_BL_POWER,
 };
 
 static struct platform_device palmtc_backlight = {
diff --git a/arch/arm/mach-pxa/palmte2.c b/arch/arm/mach-pxa/palmte2.c
index 0742721ced2d..956fd24ee6fd 100644
--- a/arch/arm/mach-pxa/palmte2.c
+++ b/arch/arm/mach-pxa/palmte2.c
@@ -165,6 +165,7 @@ static struct platform_pwm_backlight_data palmte2_backlight_data = {
 	.max_brightness	= PALMTE2_MAX_INTENSITY,
 	.dft_brightness	= PALMTE2_MAX_INTENSITY,
 	.pwm_period_ns	= PALMTE2_PERIOD_NS,
+	.enable_gpio	= -1,
 	.init		= palmte2_backlight_init,
 	.notify		= palmte2_backlight_notify,
 	.exit		= palmte2_backlight_exit,
diff --git a/arch/arm/mach-pxa/pcm990-baseboard.c b/arch/arm/mach-pxa/pcm990-baseboard.c
index 3133ba82c508..9a4e470f162b 100644
--- a/arch/arm/mach-pxa/pcm990-baseboard.c
+++ b/arch/arm/mach-pxa/pcm990-baseboard.c
@@ -153,6 +153,7 @@ static struct platform_pwm_backlight_data pcm990_backlight_data = {
 	.max_brightness	= 1023,
 	.dft_brightness	= 1023,
 	.pwm_period_ns	= 78770,
+	.enable_gpio	= -1,
 };
 
 static struct platform_device pcm990_backlight_device = {
diff --git a/arch/arm/mach-pxa/raumfeld.c b/arch/arm/mach-pxa/raumfeld.c
index 969b0ba7fa70..8386dc30b3e4 100644
--- a/arch/arm/mach-pxa/raumfeld.c
+++ b/arch/arm/mach-pxa/raumfeld.c
@@ -539,6 +539,7 @@ static struct platform_pwm_backlight_data raumfeld_pwm_backlight_data = {
 	.dft_brightness	= 100,
 	/* 10000 ns = 10 ms ^= 100 kHz */
 	.pwm_period_ns	= 10000,
+	.enable_gpio	= -1,
 };
 
 static struct platform_device raumfeld_pwm_backlight_device = {
diff --git a/arch/arm/mach-pxa/stargate2.c b/arch/arm/mach-pxa/stargate2.c
index 62aea3e835f3..01de542432a6 100644
--- a/arch/arm/mach-pxa/stargate2.c
+++ b/arch/arm/mach-pxa/stargate2.c
@@ -27,7 +27,7 @@
 
 #include <linux/i2c/pxa-i2c.h>
 #include <linux/i2c/pcf857x.h>
-#include <linux/i2c/at24.h>
+#include <linux/platform_data/at24.h>
 #include <linux/smc91x.h>
 #include <linux/gpio.h>
 #include <linux/leds.h>
diff --git a/arch/arm/mach-pxa/tavorevb.c b/arch/arm/mach-pxa/tavorevb.c
index 4680efe55345..a71da84e784b 100644
--- a/arch/arm/mach-pxa/tavorevb.c
+++ b/arch/arm/mach-pxa/tavorevb.c
@@ -175,6 +175,7 @@ static struct platform_pwm_backlight_data tavorevb_backlight_data[] = {
 		.max_brightness	= 100,
 		.dft_brightness	= 100,
 		.pwm_period_ns	= 100000,
+		.enable_gpio	= -1,
 	},
 	[1] = {
 		/* secondary backlight */
@@ -182,6 +183,7 @@ static struct platform_pwm_backlight_data tavorevb_backlight_data[] = {
 		.max_brightness	= 100,
 		.dft_brightness	= 100,
 		.pwm_period_ns	= 100000,
+		.enable_gpio	= -1,
 	},
 };
 
diff --git a/arch/arm/mach-pxa/viper.c b/arch/arm/mach-pxa/viper.c
index 9c363c081d3f..29905b127ad9 100644
--- a/arch/arm/mach-pxa/viper.c
+++ b/arch/arm/mach-pxa/viper.c
@@ -401,6 +401,7 @@ static struct platform_pwm_backlight_data viper_backlight_data = {
 	.max_brightness	= 100,
 	.dft_brightness	= 100,
 	.pwm_period_ns	= 1000000,
+	.enable_gpio	= -1,
 	.init		= viper_backlight_init,
 	.notify		= viper_backlight_notify,
 	.exit		= viper_backlight_exit,
diff --git a/arch/arm/mach-pxa/z2.c b/arch/arm/mach-pxa/z2.c
index 2513d8f4931f..e1a121b36cfa 100644
--- a/arch/arm/mach-pxa/z2.c
+++ b/arch/arm/mach-pxa/z2.c
@@ -206,6 +206,7 @@ static struct platform_pwm_backlight_data z2_backlight_data[] = {
 		.max_brightness	= 1023,
 		.dft_brightness	= 0,
 		.pwm_period_ns	= 1260320,
+		.enable_gpio	= -1,
 	},
 	[1] = {
 		/* LCD Backlight */
@@ -213,6 +214,7 @@ static struct platform_pwm_backlight_data z2_backlight_data[] = {
 		.max_brightness	= 1023,
 		.dft_brightness	= 512,
 		.pwm_period_ns	= 1260320,
+		.enable_gpio	= -1,
 	},
 };
 
diff --git a/arch/arm/mach-pxa/zylonite.c b/arch/arm/mach-pxa/zylonite.c
index 36cf7cf95ec1..77daea478e88 100644
--- a/arch/arm/mach-pxa/zylonite.c
+++ b/arch/arm/mach-pxa/zylonite.c
@@ -125,6 +125,7 @@ static struct platform_pwm_backlight_data zylonite_backlight_data = {
 	.max_brightness	= 100,
 	.dft_brightness	= 100,
 	.pwm_period_ns	= 10000,
+	.enable_gpio	= -1,
 };
 
 static struct platform_device zylonite_backlight_device = {
diff --git a/arch/arm/mach-s3c24xx/mach-h1940.c b/arch/arm/mach-s3c24xx/mach-h1940.c
index 74dd47988b41..952b6a040d1f 100644
--- a/arch/arm/mach-s3c24xx/mach-h1940.c
+++ b/arch/arm/mach-s3c24xx/mach-h1940.c
@@ -504,6 +504,7 @@ static struct platform_pwm_backlight_data backlight_data = {
 	.dft_brightness = 50,
 	/* tcnt = 0x31 */
 	.pwm_period_ns  = 36296,
+	.enable_gpio    = -1,
 	.init           = h1940_backlight_init,
 	.notify		= h1940_backlight_notify,
 	.exit           = h1940_backlight_exit,
diff --git a/arch/arm/mach-s3c24xx/mach-mini2440.c b/arch/arm/mach-s3c24xx/mach-mini2440.c
index a83db46320bc..4a18d49a63e0 100644
--- a/arch/arm/mach-s3c24xx/mach-mini2440.c
+++ b/arch/arm/mach-s3c24xx/mach-mini2440.c
@@ -24,7 +24,7 @@
 #include <linux/io.h>
 #include <linux/serial_core.h>
 #include <linux/dm9000.h>
-#include <linux/i2c/at24.h>
+#include <linux/platform_data/at24.h>
 #include <linux/platform_device.h>
 #include <linux/gpio_keys.h>
 #include <linux/i2c.h>
diff --git a/arch/arm/mach-s3c24xx/mach-rx1950.c b/arch/arm/mach-s3c24xx/mach-rx1950.c
index 206b1f7546d1..034b7fe45c49 100644
--- a/arch/arm/mach-s3c24xx/mach-rx1950.c
+++ b/arch/arm/mach-s3c24xx/mach-rx1950.c
@@ -522,6 +522,7 @@ static struct platform_pwm_backlight_data rx1950_backlight_data = {
 	.max_brightness = 24,
 	.dft_brightness = 4,
 	.pwm_period_ns = 48000,
+	.enable_gpio = -1,
 	.init = rx1950_backlight_init,
 	.notify = rx1950_backlight_notify,
 	.exit = rx1950_backlight_exit,
diff --git a/arch/arm/mach-s3c64xx/mach-crag6410.c b/arch/arm/mach-s3c64xx/mach-crag6410.c
index aca7d16e195d..758e31b26550 100644
--- a/arch/arm/mach-s3c64xx/mach-crag6410.c
+++ b/arch/arm/mach-s3c64xx/mach-crag6410.c
@@ -114,6 +114,7 @@ static struct platform_pwm_backlight_data crag6410_backlight_data = {
 	.max_brightness	= 1000,
 	.dft_brightness	= 600,
 	.pwm_period_ns	= 100000,	/* about 1kHz */
+	.enable_gpio	= -1,
 };
 
 static struct platform_device crag6410_backlight_device = {
diff --git a/arch/arm/mach-s3c64xx/mach-hmt.c b/arch/arm/mach-s3c64xx/mach-hmt.c
index e8064044ef79..614a03a92cf7 100644
--- a/arch/arm/mach-s3c64xx/mach-hmt.c
+++ b/arch/arm/mach-s3c64xx/mach-hmt.c
@@ -114,6 +114,7 @@ static struct platform_pwm_backlight_data hmt_backlight_data = {
 	.max_brightness	= 100 * 256,
 	.dft_brightness	= 40 * 256,
 	.pwm_period_ns	= 1000000000 / (100 * 256 * 20),
+	.enable_gpio	= -1,
 	.init		= hmt_bl_init,
 	.notify		= hmt_bl_notify,
 	.exit		= hmt_bl_exit,
diff --git a/arch/arm/mach-s3c64xx/mach-smartq.c b/arch/arm/mach-s3c64xx/mach-smartq.c
index 0f47237be3b2..a6b338fd0470 100644
--- a/arch/arm/mach-s3c64xx/mach-smartq.c
+++ b/arch/arm/mach-s3c64xx/mach-smartq.c
@@ -151,6 +151,7 @@ static struct platform_pwm_backlight_data smartq_backlight_data = {
 	.max_brightness	= 1000,
 	.dft_brightness	= 600,
 	.pwm_period_ns	= 1000000000 / (1000 * 20),
+	.enable_gpio	= -1,
 	.init		= smartq_bl_init,
 };
 
diff --git a/arch/arm/mach-s3c64xx/mach-smdk6410.c b/arch/arm/mach-s3c64xx/mach-smdk6410.c
index 2a7b32ca5c96..d5ea938cc9a1 100644
--- a/arch/arm/mach-s3c64xx/mach-smdk6410.c
+++ b/arch/arm/mach-s3c64xx/mach-smdk6410.c
@@ -625,6 +625,7 @@ static struct samsung_bl_gpio_info smdk6410_bl_gpio_info = {
 
 static struct platform_pwm_backlight_data smdk6410_bl_data = {
 	.pwm_id = 1,
+	.enable_gpio = -1,
 };
 
 static struct s3c_hsotg_plat smdk6410_hsotg_pdata;
diff --git a/arch/arm/mach-s5p64x0/mach-smdk6440.c b/arch/arm/mach-s5p64x0/mach-smdk6440.c
index 0b00304c1e91..9efdcc03df3b 100644
--- a/arch/arm/mach-s5p64x0/mach-smdk6440.c
+++ b/arch/arm/mach-s5p64x0/mach-smdk6440.c
@@ -223,6 +223,7 @@ static struct samsung_bl_gpio_info smdk6440_bl_gpio_info = {
 
 static struct platform_pwm_backlight_data smdk6440_bl_data = {
 	.pwm_id = 1,
+	.enable_gpio = -1,
 };
 
 static void __init smdk6440_map_io(void)
diff --git a/arch/arm/mach-s5p64x0/mach-smdk6450.c b/arch/arm/mach-s5p64x0/mach-smdk6450.c
index 5949296e88fd..c3cacc067efe 100644
--- a/arch/arm/mach-s5p64x0/mach-smdk6450.c
+++ b/arch/arm/mach-s5p64x0/mach-smdk6450.c
@@ -242,6 +242,7 @@ static struct samsung_bl_gpio_info smdk6450_bl_gpio_info = {
 
 static struct platform_pwm_backlight_data smdk6450_bl_data = {
 	.pwm_id = 1,
+	.enable_gpio = -1,
 };
 
 static void __init smdk6450_map_io(void)
diff --git a/arch/arm/mach-s5pc100/mach-smdkc100.c b/arch/arm/mach-s5pc100/mach-smdkc100.c
index 7c57a221785e..9e256b9fc930 100644
--- a/arch/arm/mach-s5pc100/mach-smdkc100.c
+++ b/arch/arm/mach-s5pc100/mach-smdkc100.c
@@ -216,6 +216,7 @@ static struct samsung_bl_gpio_info smdkc100_bl_gpio_info = {
 
 static struct platform_pwm_backlight_data smdkc100_bl_data = {
 	.pwm_id = 0,
+	.enable_gpio = -1,
 };
 
 static void __init smdkc100_map_io(void)
diff --git a/arch/arm/mach-s5pv210/mach-smdkv210.c b/arch/arm/mach-s5pv210/mach-smdkv210.c
index 6d72bb992e38..f52cc15c2d85 100644
--- a/arch/arm/mach-s5pv210/mach-smdkv210.c
+++ b/arch/arm/mach-s5pv210/mach-smdkv210.c
@@ -279,6 +279,7 @@ static struct samsung_bl_gpio_info smdkv210_bl_gpio_info = {
 static struct platform_pwm_backlight_data smdkv210_bl_data = {
 	.pwm_id = 3,
 	.pwm_period_ns = 1000,
+	.enable_gpio = -1,
 };
 
 static void __init smdkv210_map_io(void)
diff --git a/arch/arm/mach-shmobile/board-armadillo800eva.c b/arch/arm/mach-shmobile/board-armadillo800eva.c
index 8bc8e4c58847..958e3cbf0ac2 100644
--- a/arch/arm/mach-shmobile/board-armadillo800eva.c
+++ b/arch/arm/mach-shmobile/board-armadillo800eva.c
@@ -423,6 +423,7 @@ static struct platform_pwm_backlight_data pwm_backlight_data = {
 	.max_brightness = 255,
 	.dft_brightness = 255,
 	.pwm_period_ns = 33333, /* 30kHz */
+	.enable_gpio = -1,
 };
 
 static struct platform_device pwm_backlight_device = {
diff --git a/arch/arm/mach-sti/Kconfig b/arch/arm/mach-sti/Kconfig
index 835833e3c4f8..d71654bc8d54 100644
--- a/arch/arm/mach-sti/Kconfig
+++ b/arch/arm/mach-sti/Kconfig
@@ -12,7 +12,7 @@ menuconfig ARCH_STI
 	select HAVE_ARM_SCU if SMP
 	select ARCH_REQUIRE_GPIOLIB
 	select ARM_ERRATA_754322
-	select ARM_ERRATA_764369
+	select ARM_ERRATA_764369 if SMP
 	select ARM_ERRATA_775420
 	select PL310_ERRATA_753970 if CACHE_PL310
 	select PL310_ERRATA_769419 if CACHE_PL310
@@ -30,7 +30,7 @@ config SOC_STIH415
 	default y
 	help
 	  This enables support for STMicroelectronics Digital Consumer
-	  Electronics family StiH415 parts, primarily targetted at set-top-box
+	  Electronics family StiH415 parts, primarily targeted at set-top-box
 	  and other digital audio/video applications using Flattned Device
 	  Trees.
 
@@ -39,7 +39,7 @@ config SOC_STIH416
 	default y
 	help
 	  This enables support for STMicroelectronics Digital Consumer
-	  Electronics family StiH416 parts, primarily targetted at set-top-box
+	  Electronics family StiH416 parts, primarily targeted at set-top-box
 	  and other digital audio/video applications using Flattened Device
 	  Trees.
 
diff --git a/arch/arm/mach-tegra/apbio.c b/arch/arm/mach-tegra/apbio.c
index d7aa52ea6cfc..bc471973cf04 100644
--- a/arch/arm/mach-tegra/apbio.c
+++ b/arch/arm/mach-tegra/apbio.c
@@ -114,7 +114,7 @@ static int do_dma_transfer(unsigned long apb_add,
 	dma_desc->callback = apb_dma_complete;
 	dma_desc->callback_param = NULL;
 
-	INIT_COMPLETION(tegra_apb_wait);
+	reinit_completion(&tegra_apb_wait);
 
 	dmaengine_submit(dma_desc);
 	dma_async_issue_pending(tegra_apb_dma_chan);
diff --git a/arch/arm/mach-tegra/tegra.c b/arch/arm/mach-tegra/tegra.c
index ce553d557c31..73368176c6e8 100644
--- a/arch/arm/mach-tegra/tegra.c
+++ b/arch/arm/mach-tegra/tegra.c
@@ -90,9 +90,9 @@ static void __init tegra_init_cache(void)
 
 static void __init tegra_init_early(void)
 {
-	tegra_cpu_reset_handler_init();
 	tegra_apb_io_init();
 	tegra_init_fuse();
+	tegra_cpu_reset_handler_init();
 	tegra_init_cache();
 	tegra_powergate_init();
 	tegra_hotplug_init();
diff --git a/arch/arm/mm/fault-armv.c b/arch/arm/mm/fault-armv.c
index 2a5907b5c8d2..ff379ac115df 100644
--- a/arch/arm/mm/fault-armv.c
+++ b/arch/arm/mm/fault-armv.c
@@ -65,7 +65,7 @@ static int do_adjust_pte(struct vm_area_struct *vma, unsigned long address,
 	return ret;
 }
 
-#if USE_SPLIT_PTLOCKS
+#if USE_SPLIT_PTE_PTLOCKS
 /*
  * If we are using split PTE locks, then we need to take the page
  * lock here.  Otherwise we are using shared mm->page_table_lock
@@ -84,10 +84,10 @@ static inline void do_pte_unlock(spinlock_t *ptl)
 {
 	spin_unlock(ptl);
 }
-#else /* !USE_SPLIT_PTLOCKS */
+#else /* !USE_SPLIT_PTE_PTLOCKS */
 static inline void do_pte_lock(spinlock_t *ptl) {}
 static inline void do_pte_unlock(spinlock_t *ptl) {}
-#endif /* USE_SPLIT_PTLOCKS */
+#endif /* USE_SPLIT_PTE_PTLOCKS */
 
 static int adjust_pte(struct vm_area_struct *vma, unsigned long address,
 	unsigned long pfn)
diff --git a/arch/arm/plat-samsung/dev-backlight.c b/arch/arm/plat-samsung/dev-backlight.c
index d51f9565567c..be4ad0b21c08 100644
--- a/arch/arm/plat-samsung/dev-backlight.c
+++ b/arch/arm/plat-samsung/dev-backlight.c
@@ -70,6 +70,7 @@ static struct samsung_bl_drvdata samsung_dfl_bl_data __initdata = {
 		.max_brightness = 255,
 		.dft_brightness = 255,
 		.pwm_period_ns  = 78770,
+		.enable_gpio    = -1,
 		.init           = samsung_bl_init,
 		.exit           = samsung_bl_exit,
 	},
@@ -121,6 +122,10 @@ void __init samsung_bl_set(struct samsung_bl_gpio_info *gpio_info,
 		samsung_bl_data->lth_brightness = bl_data->lth_brightness;
 	if (bl_data->pwm_period_ns)
 		samsung_bl_data->pwm_period_ns = bl_data->pwm_period_ns;
+	if (bl_data->enable_gpio >= 0)
+		samsung_bl_data->enable_gpio = bl_data->enable_gpio;
+	if (bl_data->enable_gpio_flags)
+		samsung_bl_data->enable_gpio_flags = bl_data->enable_gpio_flags;
 	if (bl_data->init)
 		samsung_bl_data->init = bl_data->init;
 	if (bl_data->notify)
diff --git a/arch/arm/xen/Makefile b/arch/arm/xen/Makefile
index 43841033afd3..12969523414c 100644
--- a/arch/arm/xen/Makefile
+++ b/arch/arm/xen/Makefile
@@ -1 +1 @@
-obj-y		:= enlighten.o hypercall.o grant-table.o
+obj-y		:= enlighten.o hypercall.o grant-table.o p2m.o mm.o
diff --git a/arch/arm/xen/mm.c b/arch/arm/xen/mm.c
new file mode 100644
index 000000000000..b0e77de99148
--- /dev/null
+++ b/arch/arm/xen/mm.c
@@ -0,0 +1,65 @@
+#include <linux/bootmem.h>
+#include <linux/gfp.h>
+#include <linux/export.h>
+#include <linux/slab.h>
+#include <linux/types.h>
+#include <linux/dma-mapping.h>
+#include <linux/vmalloc.h>
+#include <linux/swiotlb.h>
+
+#include <xen/xen.h>
+#include <xen/interface/memory.h>
+#include <xen/swiotlb-xen.h>
+
+#include <asm/cacheflush.h>
+#include <asm/xen/page.h>
+#include <asm/xen/hypercall.h>
+#include <asm/xen/interface.h>
+
+int xen_create_contiguous_region(phys_addr_t pstart, unsigned int order,
+				 unsigned int address_bits,
+				 dma_addr_t *dma_handle)
+{
+	if (!xen_initial_domain())
+		return -EINVAL;
+
+	/* we assume that dom0 is mapped 1:1 for now */
+	*dma_handle = pstart;
+	return 0;
+}
+EXPORT_SYMBOL_GPL(xen_create_contiguous_region);
+
+void xen_destroy_contiguous_region(phys_addr_t pstart, unsigned int order)
+{
+	return;
+}
+EXPORT_SYMBOL_GPL(xen_destroy_contiguous_region);
+
+struct dma_map_ops *xen_dma_ops;
+EXPORT_SYMBOL_GPL(xen_dma_ops);
+
+static struct dma_map_ops xen_swiotlb_dma_ops = {
+	.mapping_error = xen_swiotlb_dma_mapping_error,
+	.alloc = xen_swiotlb_alloc_coherent,
+	.free = xen_swiotlb_free_coherent,
+	.sync_single_for_cpu = xen_swiotlb_sync_single_for_cpu,
+	.sync_single_for_device = xen_swiotlb_sync_single_for_device,
+	.sync_sg_for_cpu = xen_swiotlb_sync_sg_for_cpu,
+	.sync_sg_for_device = xen_swiotlb_sync_sg_for_device,
+	.map_sg = xen_swiotlb_map_sg_attrs,
+	.unmap_sg = xen_swiotlb_unmap_sg_attrs,
+	.map_page = xen_swiotlb_map_page,
+	.unmap_page = xen_swiotlb_unmap_page,
+	.dma_supported = xen_swiotlb_dma_supported,
+	.set_dma_mask = xen_swiotlb_set_dma_mask,
+};
+
+int __init xen_mm_init(void)
+{
+	if (!xen_initial_domain())
+		return 0;
+	xen_swiotlb_init(1, false);
+	xen_dma_ops = &xen_swiotlb_dma_ops;
+	return 0;
+}
+arch_initcall(xen_mm_init);
diff --git a/arch/arm/xen/p2m.c b/arch/arm/xen/p2m.c
new file mode 100644
index 000000000000..23732cdff551
--- /dev/null
+++ b/arch/arm/xen/p2m.c
@@ -0,0 +1,208 @@
+#include <linux/bootmem.h>
+#include <linux/gfp.h>
+#include <linux/export.h>
+#include <linux/rwlock.h>
+#include <linux/slab.h>
+#include <linux/types.h>
+#include <linux/dma-mapping.h>
+#include <linux/vmalloc.h>
+#include <linux/swiotlb.h>
+
+#include <xen/xen.h>
+#include <xen/interface/memory.h>
+#include <xen/swiotlb-xen.h>
+
+#include <asm/cacheflush.h>
+#include <asm/xen/page.h>
+#include <asm/xen/hypercall.h>
+#include <asm/xen/interface.h>
+
+struct xen_p2m_entry {
+	unsigned long pfn;
+	unsigned long mfn;
+	unsigned long nr_pages;
+	struct rb_node rbnode_mach;
+	struct rb_node rbnode_phys;
+};
+
+rwlock_t p2m_lock;
+struct rb_root phys_to_mach = RB_ROOT;
+static struct rb_root mach_to_phys = RB_ROOT;
+
+static int xen_add_phys_to_mach_entry(struct xen_p2m_entry *new)
+{
+	struct rb_node **link = &phys_to_mach.rb_node;
+	struct rb_node *parent = NULL;
+	struct xen_p2m_entry *entry;
+	int rc = 0;
+
+	while (*link) {
+		parent = *link;
+		entry = rb_entry(parent, struct xen_p2m_entry, rbnode_phys);
+
+		if (new->mfn == entry->mfn)
+			goto err_out;
+		if (new->pfn == entry->pfn)
+			goto err_out;
+
+		if (new->pfn < entry->pfn)
+			link = &(*link)->rb_left;
+		else
+			link = &(*link)->rb_right;
+	}
+	rb_link_node(&new->rbnode_phys, parent, link);
+	rb_insert_color(&new->rbnode_phys, &phys_to_mach);
+	goto out;
+
+err_out:
+	rc = -EINVAL;
+	pr_warn("%s: cannot add pfn=%pa -> mfn=%pa: pfn=%pa -> mfn=%pa already exists\n",
+			__func__, &new->pfn, &new->mfn, &entry->pfn, &entry->mfn);
+out:
+	return rc;
+}
+
+unsigned long __pfn_to_mfn(unsigned long pfn)
+{
+	struct rb_node *n = phys_to_mach.rb_node;
+	struct xen_p2m_entry *entry;
+	unsigned long irqflags;
+
+	read_lock_irqsave(&p2m_lock, irqflags);
+	while (n) {
+		entry = rb_entry(n, struct xen_p2m_entry, rbnode_phys);
+		if (entry->pfn <= pfn &&
+				entry->pfn + entry->nr_pages > pfn) {
+			read_unlock_irqrestore(&p2m_lock, irqflags);
+			return entry->mfn + (pfn - entry->pfn);
+		}
+		if (pfn < entry->pfn)
+			n = n->rb_left;
+		else
+			n = n->rb_right;
+	}
+	read_unlock_irqrestore(&p2m_lock, irqflags);
+
+	return INVALID_P2M_ENTRY;
+}
+EXPORT_SYMBOL_GPL(__pfn_to_mfn);
+
+static int xen_add_mach_to_phys_entry(struct xen_p2m_entry *new)
+{
+	struct rb_node **link = &mach_to_phys.rb_node;
+	struct rb_node *parent = NULL;
+	struct xen_p2m_entry *entry;
+	int rc = 0;
+
+	while (*link) {
+		parent = *link;
+		entry = rb_entry(parent, struct xen_p2m_entry, rbnode_mach);
+
+		if (new->mfn == entry->mfn)
+			goto err_out;
+		if (new->pfn == entry->pfn)
+			goto err_out;
+
+		if (new->mfn < entry->mfn)
+			link = &(*link)->rb_left;
+		else
+			link = &(*link)->rb_right;
+	}
+	rb_link_node(&new->rbnode_mach, parent, link);
+	rb_insert_color(&new->rbnode_mach, &mach_to_phys);
+	goto out;
+
+err_out:
+	rc = -EINVAL;
+	pr_warn("%s: cannot add pfn=%pa -> mfn=%pa: pfn=%pa -> mfn=%pa already exists\n",
+			__func__, &new->pfn, &new->mfn, &entry->pfn, &entry->mfn);
+out:
+	return rc;
+}
+
+unsigned long __mfn_to_pfn(unsigned long mfn)
+{
+	struct rb_node *n = mach_to_phys.rb_node;
+	struct xen_p2m_entry *entry;
+	unsigned long irqflags;
+
+	read_lock_irqsave(&p2m_lock, irqflags);
+	while (n) {
+		entry = rb_entry(n, struct xen_p2m_entry, rbnode_mach);
+		if (entry->mfn <= mfn &&
+				entry->mfn + entry->nr_pages > mfn) {
+			read_unlock_irqrestore(&p2m_lock, irqflags);
+			return entry->pfn + (mfn - entry->mfn);
+		}
+		if (mfn < entry->mfn)
+			n = n->rb_left;
+		else
+			n = n->rb_right;
+	}
+	read_unlock_irqrestore(&p2m_lock, irqflags);
+
+	return INVALID_P2M_ENTRY;
+}
+EXPORT_SYMBOL_GPL(__mfn_to_pfn);
+
+bool __set_phys_to_machine_multi(unsigned long pfn,
+		unsigned long mfn, unsigned long nr_pages)
+{
+	int rc;
+	unsigned long irqflags;
+	struct xen_p2m_entry *p2m_entry;
+	struct rb_node *n = phys_to_mach.rb_node;
+
+	if (mfn == INVALID_P2M_ENTRY) {
+		write_lock_irqsave(&p2m_lock, irqflags);
+		while (n) {
+			p2m_entry = rb_entry(n, struct xen_p2m_entry, rbnode_phys);
+			if (p2m_entry->pfn <= pfn &&
+					p2m_entry->pfn + p2m_entry->nr_pages > pfn) {
+				rb_erase(&p2m_entry->rbnode_mach, &mach_to_phys);
+				rb_erase(&p2m_entry->rbnode_phys, &phys_to_mach);
+				write_unlock_irqrestore(&p2m_lock, irqflags);
+				kfree(p2m_entry);
+				return true;
+			}
+			if (pfn < p2m_entry->pfn)
+				n = n->rb_left;
+			else
+				n = n->rb_right;
+		}
+		write_unlock_irqrestore(&p2m_lock, irqflags);
+		return true;
+	}
+
+	p2m_entry = kzalloc(sizeof(struct xen_p2m_entry), GFP_NOWAIT);
+	if (!p2m_entry) {
+		pr_warn("cannot allocate xen_p2m_entry\n");
+		return false;
+	}
+	p2m_entry->pfn = pfn;
+	p2m_entry->nr_pages = nr_pages;
+	p2m_entry->mfn = mfn;
+
+	write_lock_irqsave(&p2m_lock, irqflags);
+	if ((rc = xen_add_phys_to_mach_entry(p2m_entry) < 0) ||
+		(rc = xen_add_mach_to_phys_entry(p2m_entry) < 0)) {
+		write_unlock_irqrestore(&p2m_lock, irqflags);
+		return false;
+	}
+	write_unlock_irqrestore(&p2m_lock, irqflags);
+	return true;
+}
+EXPORT_SYMBOL_GPL(__set_phys_to_machine_multi);
+
+bool __set_phys_to_machine(unsigned long pfn, unsigned long mfn)
+{
+	return __set_phys_to_machine_multi(pfn, mfn, 1);
+}
+EXPORT_SYMBOL_GPL(__set_phys_to_machine);
+
+int p2m_init(void)
+{
+	rwlock_init(&p2m_lock);
+	return 0;
+}
+arch_initcall(p2m_init);
diff --git a/arch/arm64/Kconfig b/arch/arm64/Kconfig
index bb0bf1bfc05d..88c8b6c1341a 100644
--- a/arch/arm64/Kconfig
+++ b/arch/arm64/Kconfig
@@ -143,7 +143,6 @@ config CPU_BIG_ENDIAN
 
 config SMP
 	bool "Symmetric Multi-Processing"
-	select USE_GENERIC_SMP_HELPERS
 	help
 	  This enables support for systems with more than one CPU.  If
 	  you say N here, the kernel will run on single and
@@ -221,6 +220,7 @@ config XEN_DOM0
 config XEN
 	bool "Xen guest support on ARM64 (EXPERIMENTAL)"
 	depends on ARM64 && OF
+	select SWIOTLB_XEN
 	help
 	  Say Y if you want to run Linux in a Virtual Machine on Xen on ARM64.
 
diff --git a/arch/arm64/include/asm/dma-mapping.h b/arch/arm64/include/asm/dma-mapping.h
index 8d1810001aef..fd0c0c0e447a 100644
--- a/arch/arm64/include/asm/dma-mapping.h
+++ b/arch/arm64/include/asm/dma-mapping.h
@@ -23,11 +23,15 @@
 
 #include <asm-generic/dma-coherent.h>
 
+#include <xen/xen.h>
+#include <asm/xen/hypervisor.h>
+
 #define ARCH_HAS_DMA_GET_REQUIRED_MASK
 
+#define DMA_ERROR_CODE	(~(dma_addr_t)0)
 extern struct dma_map_ops *dma_ops;
 
-static inline struct dma_map_ops *get_dma_ops(struct device *dev)
+static inline struct dma_map_ops *__generic_dma_ops(struct device *dev)
 {
 	if (unlikely(!dev) || !dev->archdata.dma_ops)
 		return dma_ops;
@@ -35,6 +39,14 @@ static inline struct dma_map_ops *get_dma_ops(struct device *dev)
 		return dev->archdata.dma_ops;
 }
 
+static inline struct dma_map_ops *get_dma_ops(struct device *dev)
+{
+	if (xen_initial_domain())
+		return xen_dma_ops;
+	else
+		return __generic_dma_ops(dev);
+}
+
 #include <asm-generic/dma-mapping-common.h>
 
 static inline dma_addr_t phys_to_dma(struct device *dev, phys_addr_t paddr)
diff --git a/arch/arm64/include/asm/io.h b/arch/arm64/include/asm/io.h
index b56e5b5df881..4cc813eddacb 100644
--- a/arch/arm64/include/asm/io.h
+++ b/arch/arm64/include/asm/io.h
@@ -22,11 +22,14 @@
 #ifdef __KERNEL__
 
 #include <linux/types.h>
+#include <linux/blk_types.h>
 
 #include <asm/byteorder.h>
 #include <asm/barrier.h>
 #include <asm/pgtable.h>
 
+#include <xen/xen.h>
+
 /*
  * Generic IO read/write.  These perform native-endian accesses.
  */
@@ -263,5 +266,12 @@ extern int devmem_is_allowed(unsigned long pfn);
  */
 #define xlate_dev_kmem_ptr(p)	p
 
+struct bio_vec;
+extern bool xen_biovec_phys_mergeable(const struct bio_vec *vec1,
+				      const struct bio_vec *vec2);
+#define BIOVEC_PHYS_MERGEABLE(vec1, vec2)				\
+	(__BIOVEC_PHYS_MERGEABLE(vec1, vec2) &&				\
+	 (!xen_domain() || xen_biovec_phys_mergeable(vec1, vec2)))
+
 #endif	/* __KERNEL__ */
 #endif	/* __ASM_IO_H */
diff --git a/arch/arm64/include/asm/kvm_arm.h b/arch/arm64/include/asm/kvm_arm.h
index a5f28e2720c7..c98ef4771c73 100644
--- a/arch/arm64/include/asm/kvm_arm.h
+++ b/arch/arm64/include/asm/kvm_arm.h
@@ -63,6 +63,7 @@
  * TAC:		Trap ACTLR
  * TSC:		Trap SMC
  * TSW:		Trap cache operations by set/way
+ * TWE:		Trap WFE
  * TWI:		Trap WFI
  * TIDCP:	Trap L2CTLR/L2ECTLR
  * BSU_IS:	Upgrade barriers to the inner shareable domain
@@ -72,8 +73,9 @@
  * FMO:		Override CPSR.F and enable signaling with VF
  * SWIO:	Turn set/way invalidates into set/way clean+invalidate
  */
-#define HCR_GUEST_FLAGS (HCR_TSC | HCR_TSW | HCR_TWI | HCR_VM | HCR_BSU_IS | \
-			 HCR_FB | HCR_TAC | HCR_AMO | HCR_IMO | HCR_FMO | \
+#define HCR_GUEST_FLAGS (HCR_TSC | HCR_TSW | HCR_TWE | HCR_TWI | HCR_VM | \
+			 HCR_BSU_IS | HCR_FB | HCR_TAC | \
+			 HCR_AMO | HCR_IMO | HCR_FMO | \
 			 HCR_SWIO | HCR_TIDCP | HCR_RW)
 #define HCR_VIRT_EXCP_MASK (HCR_VA | HCR_VI | HCR_VF)
 
@@ -242,4 +244,6 @@
 
 #define ESR_EL2_EC_xABT_xFSR_EXTABT	0x10
 
+#define ESR_EL2_EC_WFI_ISS_WFE	(1 << 0)
+
 #endif /* __ARM64_KVM_ARM_H__ */
diff --git a/arch/arm64/include/asm/kvm_emulate.h b/arch/arm64/include/asm/kvm_emulate.h
index eec073875218..dd8ecfc3f995 100644
--- a/arch/arm64/include/asm/kvm_emulate.h
+++ b/arch/arm64/include/asm/kvm_emulate.h
@@ -177,4 +177,65 @@ static inline u8 kvm_vcpu_trap_get_fault(const struct kvm_vcpu *vcpu)
 	return kvm_vcpu_get_hsr(vcpu) & ESR_EL2_FSC_TYPE;
 }
 
+static inline unsigned long kvm_vcpu_get_mpidr(struct kvm_vcpu *vcpu)
+{
+	return vcpu_sys_reg(vcpu, MPIDR_EL1);
+}
+
+static inline void kvm_vcpu_set_be(struct kvm_vcpu *vcpu)
+{
+	if (vcpu_mode_is_32bit(vcpu))
+		*vcpu_cpsr(vcpu) |= COMPAT_PSR_E_BIT;
+	else
+		vcpu_sys_reg(vcpu, SCTLR_EL1) |= (1 << 25);
+}
+
+static inline bool kvm_vcpu_is_be(struct kvm_vcpu *vcpu)
+{
+	if (vcpu_mode_is_32bit(vcpu))
+		return !!(*vcpu_cpsr(vcpu) & COMPAT_PSR_E_BIT);
+
+	return !!(vcpu_sys_reg(vcpu, SCTLR_EL1) & (1 << 25));
+}
+
+static inline unsigned long vcpu_data_guest_to_host(struct kvm_vcpu *vcpu,
+						    unsigned long data,
+						    unsigned int len)
+{
+	if (kvm_vcpu_is_be(vcpu)) {
+		switch (len) {
+		case 1:
+			return data & 0xff;
+		case 2:
+			return be16_to_cpu(data & 0xffff);
+		case 4:
+			return be32_to_cpu(data & 0xffffffff);
+		default:
+			return be64_to_cpu(data);
+		}
+	}
+
+	return data;		/* Leave LE untouched */
+}
+
+static inline unsigned long vcpu_data_host_to_guest(struct kvm_vcpu *vcpu,
+						    unsigned long data,
+						    unsigned int len)
+{
+	if (kvm_vcpu_is_be(vcpu)) {
+		switch (len) {
+		case 1:
+			return data & 0xff;
+		case 2:
+			return cpu_to_be16(data & 0xffff);
+		case 4:
+			return cpu_to_be32(data & 0xffffffff);
+		default:
+			return cpu_to_be64(data);
+		}
+	}
+
+	return data;		/* Leave LE untouched */
+}
+
 #endif /* __ARM64_KVM_EMULATE_H__ */
diff --git a/arch/arm64/include/asm/kvm_host.h b/arch/arm64/include/asm/kvm_host.h
index 0859a4ddd1e7..5d85a02d1231 100644
--- a/arch/arm64/include/asm/kvm_host.h
+++ b/arch/arm64/include/asm/kvm_host.h
@@ -36,11 +36,6 @@
 
 #define KVM_VCPU_MAX_FEATURES 2
 
-/* We don't currently support large pages. */
-#define KVM_HPAGE_GFN_SHIFT(x)	0
-#define KVM_NR_PAGE_SIZES	1
-#define KVM_PAGES_PER_HPAGE(x)	(1UL<<31)
-
 struct kvm_vcpu;
 int kvm_target_cpu(void);
 int kvm_reset_vcpu(struct kvm_vcpu *vcpu);
@@ -151,6 +146,7 @@ struct kvm_vcpu_stat {
 struct kvm_vcpu_init;
 int kvm_vcpu_set_target(struct kvm_vcpu *vcpu,
 			const struct kvm_vcpu_init *init);
+int kvm_vcpu_preferred_target(struct kvm_vcpu_init *init);
 unsigned long kvm_arm_num_regs(struct kvm_vcpu *vcpu);
 int kvm_arm_copy_reg_indices(struct kvm_vcpu *vcpu, u64 __user *indices);
 struct kvm_one_reg;
diff --git a/arch/arm64/include/asm/kvm_mmu.h b/arch/arm64/include/asm/kvm_mmu.h
index efe609c6a3c9..680f74e67497 100644
--- a/arch/arm64/include/asm/kvm_mmu.h
+++ b/arch/arm64/include/asm/kvm_mmu.h
@@ -91,6 +91,7 @@ int kvm_mmu_init(void);
 void kvm_clear_hyp_idmap(void);
 
 #define	kvm_set_pte(ptep, pte)		set_pte(ptep, pte)
+#define	kvm_set_pmd(pmdp, pmd)		set_pmd(pmdp, pmd)
 
 static inline bool kvm_is_write_fault(unsigned long esr)
 {
@@ -116,13 +117,18 @@ static inline void kvm_set_s2pte_writable(pte_t *pte)
 	pte_val(*pte) |= PTE_S2_RDWR;
 }
 
+static inline void kvm_set_s2pmd_writable(pmd_t *pmd)
+{
+	pmd_val(*pmd) |= PMD_S2_RDWR;
+}
+
 struct kvm;
 
-static inline void coherent_icache_guest_page(struct kvm *kvm, gfn_t gfn)
+static inline void coherent_icache_guest_page(struct kvm *kvm, hva_t hva,
+					      unsigned long size)
 {
 	if (!icache_is_aliasing()) {		/* PIPT */
-		unsigned long hva = gfn_to_hva(kvm, gfn);
-		flush_icache_range(hva, hva + PAGE_SIZE);
+		flush_icache_range(hva, hva + size);
 	} else if (!icache_is_aivivt()) {	/* non ASID-tagged VIVT */
 		/* any kind of VIPT cache */
 		__flush_icache_all();
diff --git a/arch/arm64/include/asm/pgalloc.h b/arch/arm64/include/asm/pgalloc.h
index f214069ec5d5..9bea6e74a001 100644
--- a/arch/arm64/include/asm/pgalloc.h
+++ b/arch/arm64/include/asm/pgalloc.h
@@ -63,9 +63,12 @@ pte_alloc_one(struct mm_struct *mm, unsigned long addr)
 	struct page *pte;
 
 	pte = alloc_pages(PGALLOC_GFP, 0);
-	if (pte)
-		pgtable_page_ctor(pte);
-
+	if (!pte)
+		return NULL;
+	if (!pgtable_page_ctor(pte)) {
+		__free_page(pte);
+		return NULL;
+	}
 	return pte;
 }
 
diff --git a/arch/arm64/include/asm/pgtable-hwdef.h b/arch/arm64/include/asm/pgtable-hwdef.h
index d57e66845c86..755f86143320 100644
--- a/arch/arm64/include/asm/pgtable-hwdef.h
+++ b/arch/arm64/include/asm/pgtable-hwdef.h
@@ -85,6 +85,8 @@
 #define PTE_S2_RDONLY		(_AT(pteval_t, 1) << 6)   /* HAP[2:1] */
 #define PTE_S2_RDWR		(_AT(pteval_t, 3) << 6)   /* HAP[2:1] */
 
+#define PMD_S2_RDWR		(_AT(pmdval_t, 3) << 6)   /* HAP[2:1] */
+
 /*
  * Memory Attribute override for Stage-2 (MemAttr[3:0])
  */
diff --git a/arch/arm64/include/asm/thread_info.h b/arch/arm64/include/asm/thread_info.h
index 23a3c4791d86..720e70b66ffd 100644
--- a/arch/arm64/include/asm/thread_info.h
+++ b/arch/arm64/include/asm/thread_info.h
@@ -89,12 +89,6 @@ static inline struct thread_info *current_thread_info(void)
 #endif
 
 /*
- * We use bit 30 of the preempt_count to indicate that kernel
- * preemption is occurring.  See <asm/hardirq.h>.
- */
-#define PREEMPT_ACTIVE	0x40000000
-
-/*
  * thread information flags:
  *  TIF_SYSCALL_TRACE	- syscall trace active
  *  TIF_SIGPENDING	- signal pending
diff --git a/arch/arm64/include/asm/xen/page-coherent.h b/arch/arm64/include/asm/xen/page-coherent.h
new file mode 100644
index 000000000000..2820f1a6eebe
--- /dev/null
+++ b/arch/arm64/include/asm/xen/page-coherent.h
@@ -0,0 +1,47 @@
+#ifndef _ASM_ARM64_XEN_PAGE_COHERENT_H
+#define _ASM_ARM64_XEN_PAGE_COHERENT_H
+
+#include <asm/page.h>
+#include <linux/dma-attrs.h>
+#include <linux/dma-mapping.h>
+
+static inline void *xen_alloc_coherent_pages(struct device *hwdev, size_t size,
+		dma_addr_t *dma_handle, gfp_t flags,
+		struct dma_attrs *attrs)
+{
+	return __generic_dma_ops(hwdev)->alloc(hwdev, size, dma_handle, flags, attrs);
+}
+
+static inline void xen_free_coherent_pages(struct device *hwdev, size_t size,
+		void *cpu_addr, dma_addr_t dma_handle,
+		struct dma_attrs *attrs)
+{
+	__generic_dma_ops(hwdev)->free(hwdev, size, cpu_addr, dma_handle, attrs);
+}
+
+static inline void xen_dma_map_page(struct device *hwdev, struct page *page,
+	     unsigned long offset, size_t size, enum dma_data_direction dir,
+	     struct dma_attrs *attrs)
+{
+	__generic_dma_ops(hwdev)->map_page(hwdev, page, offset, size, dir, attrs);
+}
+
+static inline void xen_dma_unmap_page(struct device *hwdev, dma_addr_t handle,
+		size_t size, enum dma_data_direction dir,
+		struct dma_attrs *attrs)
+{
+	__generic_dma_ops(hwdev)->unmap_page(hwdev, handle, size, dir, attrs);
+}
+
+static inline void xen_dma_sync_single_for_cpu(struct device *hwdev,
+		dma_addr_t handle, size_t size, enum dma_data_direction dir)
+{
+	__generic_dma_ops(hwdev)->sync_single_for_cpu(hwdev, handle, size, dir);
+}
+
+static inline void xen_dma_sync_single_for_device(struct device *hwdev,
+		dma_addr_t handle, size_t size, enum dma_data_direction dir)
+{
+	__generic_dma_ops(hwdev)->sync_single_for_device(hwdev, handle, size, dir);
+}
+#endif /* _ASM_ARM64_XEN_PAGE_COHERENT_H */
diff --git a/arch/arm64/kvm/Kconfig b/arch/arm64/kvm/Kconfig
index 21e90820bd23..4480ab339a00 100644
--- a/arch/arm64/kvm/Kconfig
+++ b/arch/arm64/kvm/Kconfig
@@ -21,6 +21,7 @@ config KVM
 	select MMU_NOTIFIER
 	select PREEMPT_NOTIFIERS
 	select ANON_INODES
+	select HAVE_KVM_CPU_RELAX_INTERCEPT
 	select KVM_MMIO
 	select KVM_ARM_HOST
 	select KVM_ARM_VGIC
diff --git a/arch/arm64/kvm/guest.c b/arch/arm64/kvm/guest.c
index 2c3ff67a8ecb..3f0731e53274 100644
--- a/arch/arm64/kvm/guest.c
+++ b/arch/arm64/kvm/guest.c
@@ -248,6 +248,26 @@ int kvm_vcpu_set_target(struct kvm_vcpu *vcpu,
 	return kvm_reset_vcpu(vcpu);
 }
 
+int kvm_vcpu_preferred_target(struct kvm_vcpu_init *init)
+{
+	int target = kvm_target_cpu();
+
+	if (target < 0)
+		return -ENODEV;
+
+	memset(init, 0, sizeof(*init));
+
+	/*
+	 * For now, we don't return any features.
+	 * In future, we might use features to return target
+	 * specific features available for the preferred
+	 * target type.
+	 */
+	init->target = (__u32)target;
+
+	return 0;
+}
+
 int kvm_arch_vcpu_ioctl_get_fpu(struct kvm_vcpu *vcpu, struct kvm_fpu *fpu)
 {
 	return -EINVAL;
diff --git a/arch/arm64/kvm/handle_exit.c b/arch/arm64/kvm/handle_exit.c
index 9beaca033437..8da56067c304 100644
--- a/arch/arm64/kvm/handle_exit.c
+++ b/arch/arm64/kvm/handle_exit.c
@@ -47,21 +47,29 @@ static int handle_smc(struct kvm_vcpu *vcpu, struct kvm_run *run)
 }
 
 /**
- * kvm_handle_wfi - handle a wait-for-interrupts instruction executed by a guest
+ * kvm_handle_wfx - handle a wait-for-interrupts or wait-for-event
+ *		    instruction executed by a guest
+ *
  * @vcpu:	the vcpu pointer
  *
- * Simply call kvm_vcpu_block(), which will halt execution of
+ * WFE: Yield the CPU and come back to this vcpu when the scheduler
+ * decides to.
+ * WFI: Simply call kvm_vcpu_block(), which will halt execution of
  * world-switches and schedule other host processes until there is an
  * incoming IRQ or FIQ to the VM.
  */
-static int kvm_handle_wfi(struct kvm_vcpu *vcpu, struct kvm_run *run)
+static int kvm_handle_wfx(struct kvm_vcpu *vcpu, struct kvm_run *run)
 {
-	kvm_vcpu_block(vcpu);
+	if (kvm_vcpu_get_hsr(vcpu) & ESR_EL2_EC_WFI_ISS_WFE)
+		kvm_vcpu_on_spin(vcpu);
+	else
+		kvm_vcpu_block(vcpu);
+
 	return 1;
 }
 
 static exit_handle_fn arm_exit_handlers[] = {
-	[ESR_EL2_EC_WFI]	= kvm_handle_wfi,
+	[ESR_EL2_EC_WFI]	= kvm_handle_wfx,
 	[ESR_EL2_EC_CP15_32]	= kvm_handle_cp15_32,
 	[ESR_EL2_EC_CP15_64]	= kvm_handle_cp15_64,
 	[ESR_EL2_EC_CP14_MR]	= kvm_handle_cp14_access,
diff --git a/arch/arm64/xen/Makefile b/arch/arm64/xen/Makefile
index be240404ba96..74a8d87e542b 100644
--- a/arch/arm64/xen/Makefile
+++ b/arch/arm64/xen/Makefile
@@ -1,2 +1,2 @@
-xen-arm-y	+= $(addprefix ../../arm/xen/, enlighten.o grant-table.o)
+xen-arm-y	+= $(addprefix ../../arm/xen/, enlighten.o grant-table.o p2m.o mm.o)
 obj-y		:= xen-arm.o hypercall.o
diff --git a/arch/avr32/boards/atngw100/evklcd10x.c b/arch/avr32/boards/atngw100/evklcd10x.c
index 20388750d564..64919b0da7aa 100644
--- a/arch/avr32/boards/atngw100/evklcd10x.c
+++ b/arch/avr32/boards/atngw100/evklcd10x.c
@@ -58,7 +58,7 @@ static struct fb_monspecs __initdata atevklcd10x_default_monspecs = {
 	.dclkmax		= 28330000,
 };
 
-static struct atmel_lcdfb_info __initdata atevklcd10x_lcdc_data = {
+static struct atmel_lcdfb_pdata __initdata atevklcd10x_lcdc_data = {
 	.default_bpp		= 16,
 	.default_dmacon		= ATMEL_LCDC_DMAEN | ATMEL_LCDC_DMA2DEN,
 	.default_lcdcon2	= (ATMEL_LCDC_DISTYPE_TFT
@@ -96,7 +96,7 @@ static struct fb_monspecs __initdata atevklcd10x_default_monspecs = {
 	.dclkmax		= 7000000,
 };
 
-static struct atmel_lcdfb_info __initdata atevklcd10x_lcdc_data = {
+static struct atmel_lcdfb_pdata __initdata atevklcd10x_lcdc_data = {
 	.default_bpp		= 16,
 	.default_dmacon		= ATMEL_LCDC_DMAEN | ATMEL_LCDC_DMA2DEN,
 	.default_lcdcon2	= (ATMEL_LCDC_DISTYPE_TFT
@@ -134,7 +134,7 @@ static struct fb_monspecs __initdata atevklcd10x_default_monspecs = {
 	.dclkmax		= 6400000,
 };
 
-static struct atmel_lcdfb_info __initdata atevklcd10x_lcdc_data = {
+static struct atmel_lcdfb_pdata __initdata atevklcd10x_lcdc_data = {
 	.default_bpp		= 16,
 	.default_dmacon		= ATMEL_LCDC_DMAEN | ATMEL_LCDC_DMA2DEN,
 	.default_lcdcon2	= (ATMEL_LCDC_DISTYPE_TFT
@@ -145,7 +145,7 @@ static struct atmel_lcdfb_info __initdata atevklcd10x_lcdc_data = {
 };
 #endif
 
-static void atevklcd10x_lcdc_power_control(int on)
+static void atevklcd10x_lcdc_power_control(struct atmel_lcdfb_pdata *pdata, int on)
 {
 	gpio_set_value(GPIO_PIN_PB(15), on);
 }
diff --git a/arch/avr32/boards/atngw100/mrmt.c b/arch/avr32/boards/atngw100/mrmt.c
index 7de083d19b7e..1ba09e4c02b1 100644
--- a/arch/avr32/boards/atngw100/mrmt.c
+++ b/arch/avr32/boards/atngw100/mrmt.c
@@ -83,7 +83,7 @@ static struct fb_monspecs __initdata lcd_fb_default_monspecs = {
 	.dclkmax		= 9260000,
 };
 
-static struct atmel_lcdfb_info __initdata rmt_lcdc_data = {
+static struct atmel_lcdfb_pdata __initdata rmt_lcdc_data = {
 	.default_bpp		= 24,
 	.default_dmacon		= ATMEL_LCDC_DMAEN | ATMEL_LCDC_DMA2DEN,
 	.default_lcdcon2	= (ATMEL_LCDC_DISTYPE_TFT
@@ -126,7 +126,7 @@ static struct fb_monspecs __initdata lcd_fb_default_monspecs = {
 	.dclkmax		= 9260000,
 };
 
-static struct atmel_lcdfb_info __initdata rmt_lcdc_data = {
+static struct atmel_lcdfb_pdata __initdata rmt_lcdc_data = {
 	.default_bpp		= 24,
 	.default_dmacon		= ATMEL_LCDC_DMAEN | ATMEL_LCDC_DMA2DEN,
 	.default_lcdcon2	= (ATMEL_LCDC_DISTYPE_TFT
diff --git a/arch/avr32/boards/atstk1000/atstk1000.h b/arch/avr32/boards/atstk1000/atstk1000.h
index 9392d3252865..653cc09e536c 100644
--- a/arch/avr32/boards/atstk1000/atstk1000.h
+++ b/arch/avr32/boards/atstk1000/atstk1000.h
@@ -10,7 +10,7 @@
 #ifndef __ARCH_AVR32_BOARDS_ATSTK1000_ATSTK1000_H
 #define __ARCH_AVR32_BOARDS_ATSTK1000_ATSTK1000_H
 
-extern struct atmel_lcdfb_info atstk1000_lcdc_data;
+extern struct atmel_lcdfb_pdata atstk1000_lcdc_data;
 
 void atstk1000_setup_j2_leds(void);
 
diff --git a/arch/avr32/boards/atstk1000/setup.c b/arch/avr32/boards/atstk1000/setup.c
index 2d6b560115d9..b6b88f5e0b43 100644
--- a/arch/avr32/boards/atstk1000/setup.c
+++ b/arch/avr32/boards/atstk1000/setup.c
@@ -55,7 +55,7 @@ static struct fb_monspecs __initdata atstk1000_default_monspecs = {
 	.dclkmax		= 30000000,
 };
 
-struct atmel_lcdfb_info __initdata atstk1000_lcdc_data = {
+struct atmel_lcdfb_pdata __initdata atstk1000_lcdc_data = {
 	.default_bpp		= 24,
 	.default_dmacon		= ATMEL_LCDC_DMAEN | ATMEL_LCDC_DMA2DEN,
 	.default_lcdcon2	= (ATMEL_LCDC_DISTYPE_TFT
diff --git a/arch/avr32/boards/favr-32/setup.c b/arch/avr32/boards/favr-32/setup.c
index 27bd6fbe21cb..7b1f2cd85400 100644
--- a/arch/avr32/boards/favr-32/setup.c
+++ b/arch/avr32/boards/favr-32/setup.c
@@ -125,7 +125,7 @@ static struct fb_monspecs __initdata favr32_default_monspecs = {
 	.dclkmax		= 28000000,
 };
 
-struct atmel_lcdfb_info __initdata favr32_lcdc_data = {
+struct atmel_lcdfb_pdata __initdata favr32_lcdc_data = {
 	.default_bpp		= 16,
 	.default_dmacon		= ATMEL_LCDC_DMAEN | ATMEL_LCDC_DMA2DEN,
 	.default_lcdcon2	= (ATMEL_LCDC_DISTYPE_TFT
diff --git a/arch/avr32/boards/hammerhead/setup.c b/arch/avr32/boards/hammerhead/setup.c
index 9d1efd1cd425..dc0e317f2ecd 100644
--- a/arch/avr32/boards/hammerhead/setup.c
+++ b/arch/avr32/boards/hammerhead/setup.c
@@ -77,7 +77,7 @@ static struct fb_monspecs __initdata hammerhead_hda350t_monspecs = {
 	.dclkmax		= 10000000,
 };
 
-struct atmel_lcdfb_info __initdata hammerhead_lcdc_data = {
+struct atmel_lcdfb_pdata __initdata hammerhead_lcdc_data = {
 	.default_bpp		= 24,
 	.default_dmacon		= ATMEL_LCDC_DMAEN | ATMEL_LCDC_DMA2DEN,
 	.default_lcdcon2	= (ATMEL_LCDC_DISTYPE_TFT
diff --git a/arch/avr32/boards/merisc/display.c b/arch/avr32/boards/merisc/display.c
index 85a543cd4abc..e7683ee7ed40 100644
--- a/arch/avr32/boards/merisc/display.c
+++ b/arch/avr32/boards/merisc/display.c
@@ -45,7 +45,7 @@ static struct fb_monspecs merisc_fb_monspecs = {
 	.dclkmax	= 30000000,
 };
 
-struct atmel_lcdfb_info merisc_lcdc_data = {
+struct atmel_lcdfb_pdata merisc_lcdc_data = {
 	.default_bpp		= 24,
 	.default_dmacon		= ATMEL_LCDC_DMAEN | ATMEL_LCDC_DMA2DEN,
 	.default_lcdcon2	= (ATMEL_LCDC_DISTYPE_TFT
diff --git a/arch/avr32/boards/mimc200/setup.c b/arch/avr32/boards/mimc200/setup.c
index 05358aa5ef7d..1cb8e9cc5cfa 100644
--- a/arch/avr32/boards/mimc200/setup.c
+++ b/arch/avr32/boards/mimc200/setup.c
@@ -8,7 +8,7 @@
  * published by the Free Software Foundation.
  */
 
-extern struct atmel_lcdfb_info mimc200_lcdc_data;
+extern struct atmel_lcdfb_pdata mimc200_lcdc_data;
 
 #include <linux/clk.h>
 #include <linux/etherdevice.h>
@@ -71,7 +71,7 @@ static struct fb_monspecs __initdata mimc200_default_monspecs = {
 	.dclkmax		= 25200000,
 };
 
-struct atmel_lcdfb_info __initdata mimc200_lcdc_data = {
+struct atmel_lcdfb_pdata __initdata mimc200_lcdc_data = {
 	.default_bpp		= 16,
 	.default_dmacon		= ATMEL_LCDC_DMAEN | ATMEL_LCDC_DMA2DEN,
 	.default_lcdcon2	= (ATMEL_LCDC_DISTYPE_TFT
diff --git a/arch/avr32/include/asm/pgalloc.h b/arch/avr32/include/asm/pgalloc.h
index bc7e8ae479ee..1aba19d68c5e 100644
--- a/arch/avr32/include/asm/pgalloc.h
+++ b/arch/avr32/include/asm/pgalloc.h
@@ -68,7 +68,10 @@ static inline pgtable_t pte_alloc_one(struct mm_struct *mm,
 		return NULL;
 
 	page = virt_to_page(pg);
-	pgtable_page_ctor(page);
+	if (!pgtable_page_ctor(page)) {
+		quicklist_free(QUICK_PT, NULL, pg);
+		return NULL;
+	}
 
 	return page;
 }
diff --git a/arch/avr32/include/asm/thread_info.h b/arch/avr32/include/asm/thread_info.h
index 6dc62e1f94c7..a978f3fe7c25 100644
--- a/arch/avr32/include/asm/thread_info.h
+++ b/arch/avr32/include/asm/thread_info.h
@@ -66,8 +66,6 @@ static inline struct thread_info *current_thread_info(void)
 
 #endif /* !__ASSEMBLY__ */
 
-#define PREEMPT_ACTIVE		0x40000000
-
 /*
  * Thread information flags
  * - these are process state flags that various assembly files may need to access
diff --git a/arch/avr32/mach-at32ap/at32ap700x.c b/arch/avr32/mach-at32ap/at32ap700x.c
index a68f3cf7c3c1..a1f4d1e91b52 100644
--- a/arch/avr32/mach-at32ap/at32ap700x.c
+++ b/arch/avr32/mach-at32ap/at32ap700x.c
@@ -1439,7 +1439,7 @@ fail:
  *  LCDC
  * -------------------------------------------------------------------- */
 #if defined(CONFIG_CPU_AT32AP7000) || defined(CONFIG_CPU_AT32AP7002)
-static struct atmel_lcdfb_info atmel_lcdfb0_data;
+static struct atmel_lcdfb_pdata atmel_lcdfb0_data;
 static struct resource atmel_lcdfb0_resource[] = {
 	{
 		.start		= 0xff000000,
@@ -1467,12 +1467,12 @@ static struct clk atmel_lcdfb0_pixclk = {
 };
 
 struct platform_device *__init
-at32_add_device_lcdc(unsigned int id, struct atmel_lcdfb_info *data,
+at32_add_device_lcdc(unsigned int id, struct atmel_lcdfb_pdata *data,
 		     unsigned long fbmem_start, unsigned long fbmem_len,
 		     u64 pin_mask)
 {
 	struct platform_device *pdev;
-	struct atmel_lcdfb_info *info;
+	struct atmel_lcdfb_pdata *info;
 	struct fb_monspecs *monspecs;
 	struct fb_videomode *modedb;
 	unsigned int modedb_size;
@@ -1529,7 +1529,7 @@ at32_add_device_lcdc(unsigned int id, struct atmel_lcdfb_info *data,
 	}
 
 	info = pdev->dev.platform_data;
-	memcpy(info, data, sizeof(struct atmel_lcdfb_info));
+	memcpy(info, data, sizeof(struct atmel_lcdfb_pdata));
 	info->default_monspecs = monspecs;
 
 	pdev->name = "at32ap-lcdfb";
diff --git a/arch/avr32/mach-at32ap/include/mach/board.h b/arch/avr32/mach-at32ap/include/mach/board.h
index d485b0391357..f1a316d52c73 100644
--- a/arch/avr32/mach-at32ap/include/mach/board.h
+++ b/arch/avr32/mach-at32ap/include/mach/board.h
@@ -44,9 +44,9 @@ struct platform_device *
 at32_add_device_spi(unsigned int id, struct spi_board_info *b, unsigned int n);
 void at32_spi_setup_slaves(unsigned int bus_num, struct spi_board_info *b, unsigned int n);
 
-struct atmel_lcdfb_info;
+struct atmel_lcdfb_pdata;
 struct platform_device *
-at32_add_device_lcdc(unsigned int id, struct atmel_lcdfb_info *data,
+at32_add_device_lcdc(unsigned int id, struct atmel_lcdfb_pdata *data,
 		     unsigned long fbmem_start, unsigned long fbmem_len,
 		     u64 pin_mask);
 
diff --git a/arch/blackfin/Kconfig b/arch/blackfin/Kconfig
index e887b57c3176..9ceccef9c649 100644
--- a/arch/blackfin/Kconfig
+++ b/arch/blackfin/Kconfig
@@ -34,7 +34,6 @@ config BLACKFIN
 	select ARCH_WANT_IPC_PARSE_VERSION
 	select GENERIC_ATOMIC64
 	select GENERIC_IRQ_PROBE
-	select USE_GENERIC_SMP_HELPERS if SMP
 	select HAVE_NMI_WATCHDOG if NMI_WATCHDOG
 	select GENERIC_SMP_IDLE_THREAD
 	select ARCH_USES_GETTIMEOFFSET if !GENERIC_CLOCKEVENTS
diff --git a/arch/blackfin/configs/BF609-EZKIT_defconfig b/arch/blackfin/configs/BF609-EZKIT_defconfig
index 13eb73231a9a..4ca39ab6b2bf 100644
--- a/arch/blackfin/configs/BF609-EZKIT_defconfig
+++ b/arch/blackfin/configs/BF609-EZKIT_defconfig
@@ -102,7 +102,7 @@ CONFIG_I2C_CHARDEV=y
 CONFIG_I2C_BLACKFIN_TWI=y
 CONFIG_I2C_BLACKFIN_TWI_CLK_KHZ=100
 CONFIG_SPI=y
-CONFIG_SPI_BFIN6XX=y
+CONFIG_SPI_BFIN_V3=y
 CONFIG_GPIOLIB=y
 CONFIG_GPIO_SYSFS=y
 # CONFIG_HWMON is not set
diff --git a/arch/blackfin/include/asm/hardirq.h b/arch/blackfin/include/asm/hardirq.h
index c078dd78d998..58b54a6d5a16 100644
--- a/arch/blackfin/include/asm/hardirq.h
+++ b/arch/blackfin/include/asm/hardirq.h
@@ -12,9 +12,6 @@
 extern void ack_bad_irq(unsigned int irq);
 #define ack_bad_irq ack_bad_irq
 
-/* Define until common code gets sane defaults */
-#define HARDIRQ_BITS 9
-
 #include <asm-generic/hardirq.h>
 
 #endif
diff --git a/arch/blackfin/include/asm/irq.h b/arch/blackfin/include/asm/irq.h
index 4ae1144a4578..2fd04f10cc26 100644
--- a/arch/blackfin/include/asm/irq.h
+++ b/arch/blackfin/include/asm/irq.h
@@ -23,8 +23,7 @@
 /*
  * pm save bfin pint registers
  */
-struct bfin_pm_pint_save {
-	u32 mask_set;
+struct adi_pm_pint_save {
 	u32 assign;
 	u32 edge_set;
 	u32 invert_set;
diff --git a/arch/blackfin/include/asm/irq_handler.h b/arch/blackfin/include/asm/irq_handler.h
index 4fbf83575db1..4b2a992794d7 100644
--- a/arch/blackfin/include/asm/irq_handler.h
+++ b/arch/blackfin/include/asm/irq_handler.h
@@ -12,11 +12,11 @@
 #include <mach/irq.h>
 
 /* init functions only */
-extern int __init init_arch_irq(void);
+extern int init_arch_irq(void);
 extern void init_exception_vectors(void);
-extern void __init program_IAR(void);
+extern void program_IAR(void);
 #ifdef init_mach_irq
-extern void __init init_mach_irq(void);
+extern void init_mach_irq(void);
 #else
 # define init_mach_irq()
 #endif
diff --git a/arch/blackfin/include/asm/thread_info.h b/arch/blackfin/include/asm/thread_info.h
index 3894005337ba..55f473bdad36 100644
--- a/arch/blackfin/include/asm/thread_info.h
+++ b/arch/blackfin/include/asm/thread_info.h
@@ -88,8 +88,6 @@ static inline struct thread_info *current_thread_info(void)
 #define TI_CPU		12
 #define TI_PREEMPT	16
 
-#define	PREEMPT_ACTIVE	0x4000000
-
 /*
  * thread information flag bit numbers
  */
diff --git a/arch/blackfin/kernel/bfin_gpio.c b/arch/blackfin/kernel/bfin_gpio.c
index ed978f1c5cb9..a017359c1826 100644
--- a/arch/blackfin/kernel/bfin_gpio.c
+++ b/arch/blackfin/kernel/bfin_gpio.c
@@ -11,11 +11,8 @@
 #include <linux/err.h>
 #include <linux/proc_fs.h>
 #include <linux/seq_file.h>
-#include <asm/blackfin.h>
-#include <asm/gpio.h>
-#include <asm/portmux.h>
+#include <linux/gpio.h>
 #include <linux/irq.h>
-#include <asm/irq_handler.h>
 
 #if ANOMALY_05000311 || ANOMALY_05000323
 enum {
@@ -58,19 +55,6 @@ static struct gpio_port_t * const gpio_array[] = {
 	(struct gpio_port_t *) FIO0_FLAG_D,
 	(struct gpio_port_t *) FIO1_FLAG_D,
 	(struct gpio_port_t *) FIO2_FLAG_D,
-#elif defined(CONFIG_BF54x) || defined(CONFIG_BF60x) 
-	(struct gpio_port_t *)PORTA_FER,
-	(struct gpio_port_t *)PORTB_FER,
-	(struct gpio_port_t *)PORTC_FER,
-	(struct gpio_port_t *)PORTD_FER,
-	(struct gpio_port_t *)PORTE_FER,
-	(struct gpio_port_t *)PORTF_FER,
-	(struct gpio_port_t *)PORTG_FER,
-# if defined(CONFIG_BF54x)
-	(struct gpio_port_t *)PORTH_FER,
-	(struct gpio_port_t *)PORTI_FER,
-	(struct gpio_port_t *)PORTJ_FER,
-# endif
 #else
 # error no gpio arrays defined
 #endif
@@ -169,12 +153,6 @@ DECLARE_RESERVED_MAP(gpio_irq, GPIO_BANK_NUM);
 
 inline int check_gpio(unsigned gpio)
 {
-#if defined(CONFIG_BF54x)
-	if (gpio == GPIO_PB15 || gpio == GPIO_PC14 || gpio == GPIO_PC15
-	    || gpio == GPIO_PH14 || gpio == GPIO_PH15
-	    || gpio == GPIO_PJ14 || gpio == GPIO_PJ15)
-		return -EINVAL;
-#endif
 	if (gpio >= MAX_BLACKFIN_GPIOS)
 		return -EINVAL;
 	return 0;
@@ -212,12 +190,6 @@ static void port_setup(unsigned gpio, unsigned short usage)
 	else
 		*port_fer[gpio_bank(gpio)] |= gpio_bit(gpio);
 	SSYNC();
-#elif defined(CONFIG_BF54x) || defined(CONFIG_BF60x)
-	if (usage == GPIO_USAGE)
-		gpio_array[gpio_bank(gpio)]->port_fer &= ~gpio_bit(gpio);
-	else
-		gpio_array[gpio_bank(gpio)]->port_fer |= gpio_bit(gpio);
-	SSYNC();
 #endif
 }
 
@@ -255,7 +227,7 @@ static int portmux_group_check(unsigned short per)
 	u16 ident = P_IDENT(per);
 	u16 function = P_FUNCT2MUX(per);
 	s8 offset = port_mux[ident];
-	u16 m, pmux, pfunc;
+	u16 m, pmux, pfunc, mask;
 
 	if (offset < 0)
 		return 0;
@@ -270,10 +242,12 @@ static int portmux_group_check(unsigned short per)
 			continue;
 
 		if (offset == 1)
-			pfunc = (pmux >> offset) & 3;
+			mask = 3;
 		else
-			pfunc = (pmux >> offset) & 1;
-		if (pfunc != function) {
+			mask = 1;
+
+		pfunc = (pmux >> offset) & mask;
+		if (pfunc != (function & mask)) {
 			pr_err("pin group conflict! request pin %d func %d conflict with pin %d func %d\n",
 				ident, function, m, pfunc);
 			return -EINVAL;
@@ -288,43 +262,21 @@ static void portmux_setup(unsigned short per)
 	u16 ident = P_IDENT(per);
 	u16 function = P_FUNCT2MUX(per);
 	s8 offset = port_mux[ident];
-	u16 pmux;
+	u16 pmux, mask;
 
 	if (offset == -1)
 		return;
 
 	pmux = bfin_read_PORT_MUX();
-	if (offset != 1)
-		pmux &= ~(1 << offset);
+	if (offset == 1)
+		mask = 3;
 	else
-		pmux &= ~(3 << 1);
-	pmux |= (function << offset);
-	bfin_write_PORT_MUX(pmux);
-}
-#elif defined(CONFIG_BF54x) || defined(CONFIG_BF60x)
-inline void portmux_setup(unsigned short per)
-{
-	u16 ident = P_IDENT(per);
-	u16 function = P_FUNCT2MUX(per);
-	u32 pmux;
+		mask = 1;
 
-	pmux = gpio_array[gpio_bank(ident)]->port_mux;
+	pmux &= ~(mask << offset);
+	pmux |= ((function & mask) << offset);
 
-	pmux &= ~(0x3 << (2 * gpio_sub_n(ident)));
-	pmux |= (function & 0x3) << (2 * gpio_sub_n(ident));
-
-	gpio_array[gpio_bank(ident)]->port_mux = pmux;
-}
-
-inline u16 get_portmux(unsigned short per)
-{
-	u16 ident = P_IDENT(per);
-	u32 pmux = gpio_array[gpio_bank(ident)]->port_mux;
-	return (pmux >> (2 * gpio_sub_n(ident)) & 0x3);
-}
-static int portmux_group_check(unsigned short per)
-{
-	return 0;
+	bfin_write_PORT_MUX(pmux);
 }
 #elif defined(CONFIG_BF52x) || defined(CONFIG_BF51x)
 static int portmux_group_check(unsigned short per)
@@ -379,7 +331,6 @@ static int portmux_group_check(unsigned short per)
 }
 #endif
 
-#if !(defined(CONFIG_BF54x) || defined(CONFIG_BF60x))
 /***********************************************************
 *
 * FUNCTIONS: Blackfin General Purpose Ports Access Functions
@@ -572,7 +523,7 @@ static const unsigned int sic_iwr_irqs[] = {
 *************************************************************
 * MODIFICATION HISTORY :
 **************************************************************/
-int gpio_pm_wakeup_ctrl(unsigned gpio, unsigned ctrl)
+int bfin_gpio_pm_wakeup_ctrl(unsigned gpio, unsigned ctrl)
 {
 	unsigned long flags;
 
@@ -591,7 +542,7 @@ int gpio_pm_wakeup_ctrl(unsigned gpio, unsigned ctrl)
 	return 0;
 }
 
-int bfin_pm_standby_ctrl(unsigned ctrl)
+int bfin_gpio_pm_standby_ctrl(unsigned ctrl)
 {
 	u16 bank, mask, i;
 
@@ -682,53 +633,6 @@ void bfin_gpio_pm_hibernate_restore(void)
 
 
 #endif
-#else /* CONFIG_BF54x || CONFIG_BF60x */
-#ifdef CONFIG_PM
-
-int bfin_pm_standby_ctrl(unsigned ctrl)
-{
-	return 0;
-}
-
-void bfin_gpio_pm_hibernate_suspend(void)
-{
-	int i, bank;
-
-	for (i = 0; i < MAX_BLACKFIN_GPIOS; i += GPIO_BANKSIZE) {
-		bank = gpio_bank(i);
-
-		gpio_bank_saved[bank].fer = gpio_array[bank]->port_fer;
-		gpio_bank_saved[bank].mux = gpio_array[bank]->port_mux;
-		gpio_bank_saved[bank].data = gpio_array[bank]->data;
-		gpio_bank_saved[bank].inen = gpio_array[bank]->inen;
-		gpio_bank_saved[bank].dir = gpio_array[bank]->dir_set;
-	}
-}
-
-void bfin_gpio_pm_hibernate_restore(void)
-{
-	int i, bank;
-
-	for (i = 0; i < MAX_BLACKFIN_GPIOS; i += GPIO_BANKSIZE) {
-		bank = gpio_bank(i);
-
-		gpio_array[bank]->port_mux = gpio_bank_saved[bank].mux;
-		gpio_array[bank]->port_fer = gpio_bank_saved[bank].fer;
-		gpio_array[bank]->inen = gpio_bank_saved[bank].inen;
-		gpio_array[bank]->data_set = gpio_bank_saved[bank].data
-						& gpio_bank_saved[bank].dir;
-		gpio_array[bank]->dir_set = gpio_bank_saved[bank].dir;
-	}
-}
-#endif
-
-unsigned short get_gpio_dir(unsigned gpio)
-{
-	return (0x01 & (gpio_array[gpio_bank(gpio)]->dir_clear >> gpio_sub_n(gpio)));
-}
-EXPORT_SYMBOL(get_gpio_dir);
-
-#endif /* CONFIG_BF54x || CONFIG_BF60x */
 
 /***********************************************************
 *
@@ -785,11 +689,7 @@ int peripheral_request(unsigned short per, const char *label)
 		 * be requested and used by several drivers
 		 */
 
-#if defined(CONFIG_BF54x) || defined(CONFIG_BF60x)
-		if (!((per & P_MAYSHARE) && get_portmux(per) == P_FUNCT2MUX(per))) {
-#else
 		if (!(per & P_MAYSHARE)) {
-#endif
 			/*
 			 * Allow that the identical pin function can
 			 * be requested from the same driver twice
@@ -938,12 +838,9 @@ int bfin_gpio_request(unsigned gpio, const char *label)
 	if (unlikely(is_reserved(gpio_irq, gpio, 1))) {
 		printk(KERN_NOTICE "bfin-gpio: GPIO %d is already reserved as gpio-irq!"
 		       " (Documentation/blackfin/bfin-gpio-notes.txt)\n", gpio);
-	}
-#if !(defined(CONFIG_BF54x) || defined(CONFIG_BF60x))
-	else {	/* Reset POLAR setting when acquiring a gpio for the first time */
+	} else {	/* Reset POLAR setting when acquiring a gpio for the first time */
 		set_gpio_polar(gpio, 0);
 	}
-#endif
 
 	reserve(gpio, gpio);
 	set_label(gpio, label);
@@ -1112,11 +1009,7 @@ void bfin_gpio_irq_free(unsigned gpio)
 
 static inline void __bfin_gpio_direction_input(unsigned gpio)
 {
-#if defined(CONFIG_BF54x) || defined(CONFIG_BF60x)
-	gpio_array[gpio_bank(gpio)]->dir_clear = gpio_bit(gpio);
-#else
 	gpio_array[gpio_bank(gpio)]->dir &= ~gpio_bit(gpio);
-#endif
 	gpio_array[gpio_bank(gpio)]->inen |= gpio_bit(gpio);
 }
 
@@ -1140,17 +1033,7 @@ EXPORT_SYMBOL(bfin_gpio_direction_input);
 
 void bfin_gpio_irq_prepare(unsigned gpio)
 {
-#if defined(CONFIG_BF54x) || defined(CONFIG_BF60x)
-	unsigned long flags;
-#endif
-
 	port_setup(gpio, GPIO_USAGE);
-
-#if defined(CONFIG_BF54x) || defined(CONFIG_BF60x)
-	flags = hard_local_irq_save();
-	__bfin_gpio_direction_input(gpio);
-	hard_local_irq_restore(flags);
-#endif
 }
 
 void bfin_gpio_set_value(unsigned gpio, int arg)
@@ -1175,11 +1058,7 @@ int bfin_gpio_direction_output(unsigned gpio, int value)
 
 	gpio_array[gpio_bank(gpio)]->inen &= ~gpio_bit(gpio);
 	gpio_set_value(gpio, value);
-#if defined(CONFIG_BF54x) || defined(CONFIG_BF60x)
-	gpio_array[gpio_bank(gpio)]->dir_set = gpio_bit(gpio);
-#else
 	gpio_array[gpio_bank(gpio)]->dir |= gpio_bit(gpio);
-#endif
 
 	AWA_DUMMY_READ(dir);
 	hard_local_irq_restore(flags);
@@ -1190,9 +1069,6 @@ EXPORT_SYMBOL(bfin_gpio_direction_output);
 
 int bfin_gpio_get_value(unsigned gpio)
 {
-#if defined(CONFIG_BF54x) || defined(CONFIG_BF60x)
-	return (1 & (gpio_array[gpio_bank(gpio)]->data >> gpio_sub_n(gpio)));
-#else
 	unsigned long flags;
 
 	if (unlikely(get_gpio_edge(gpio))) {
@@ -1205,7 +1081,6 @@ int bfin_gpio_get_value(unsigned gpio)
 		return ret;
 	} else
 		return get_gpio_data(gpio);
-#endif
 }
 EXPORT_SYMBOL(bfin_gpio_get_value);
 
diff --git a/arch/blackfin/mach-bf548/Kconfig b/arch/blackfin/mach-bf548/Kconfig
index 94acb586832e..334ec7b12188 100644
--- a/arch/blackfin/mach-bf548/Kconfig
+++ b/arch/blackfin/mach-bf548/Kconfig
@@ -377,40 +377,6 @@ config IRQ_PINT3
 
 endmenu
 
-comment "Pin Interrupt to Port Assignment"
-menu "Assignment"
-
-config PINTx_REASSIGN
-	bool "Reprogram PINT Assignment"
-	default y
-	help
-	  The interrupt assignment registers controls the pin-to-interrupt
-	  assignment in a byte-wide manner. Each option allows you to select
-	  a set of pins (High/Low Byte) of an specific Port being mapped
-	  to one of the four PIN Interrupts IRQ_PINTx.
-
-	  You shouldn't change any of these unless you know exactly what you're doing.
-	  Please consult the Blackfin BF54x Processor Hardware Reference Manual.
-
-config PINT0_ASSIGN
-	hex "PINT0_ASSIGN"
-	depends on PINTx_REASSIGN
-	default 0x00000101
-config PINT1_ASSIGN
-	hex "PINT1_ASSIGN"
-	depends on PINTx_REASSIGN
-	default 0x01010000
-config PINT2_ASSIGN
-	hex "PINT2_ASSIGN"
-	depends on PINTx_REASSIGN
-	default 0x07000101
-config PINT3_ASSIGN
-	hex "PINT3_ASSIGN"
-	depends on PINTx_REASSIGN
-	default 0x02020303
-
-endmenu
-
 endmenu
 
 endif
diff --git a/arch/blackfin/mach-bf548/boards/ezkit.c b/arch/blackfin/mach-bf548/boards/ezkit.c
index 372eb54944ef..d495000b81a0 100644
--- a/arch/blackfin/mach-bf548/boards/ezkit.c
+++ b/arch/blackfin/mach-bf548/boards/ezkit.c
@@ -17,6 +17,9 @@
 #include <linux/i2c.h>
 #include <linux/interrupt.h>
 #include <linux/usb/musb.h>
+#include <linux/pinctrl/machine.h>
+#include <linux/pinctrl/pinconf-generic.h>
+#include <linux/platform_data/pinctrl-adi2.h>
 #include <asm/bfin5xx_spi.h>
 #include <asm/dma.h>
 #include <asm/gpio.h>
@@ -241,6 +244,13 @@ static struct resource bfin_uart0_resources[] = {
 		.end = UART0_RBR+2,
 		.flags = IORESOURCE_MEM,
 	},
+#ifdef CONFIG_EARLY_PRINTK
+	{
+		.start = PORTE_FER,
+		.end = PORTE_FER+2,
+		.flags = IORESOURCE_REG,
+	},
+#endif
 	{
 		.start = IRQ_UART0_TX,
 		.end = IRQ_UART0_TX,
@@ -289,6 +299,13 @@ static struct resource bfin_uart1_resources[] = {
 		.end = UART1_RBR+2,
 		.flags = IORESOURCE_MEM,
 	},
+#ifdef CONFIG_EARLY_PRINTK
+	{
+		.start = PORTH_FER,
+		.end = PORTH_FER+2,
+		.flags = IORESOURCE_REG,
+	},
+#endif
 	{
 		.start = IRQ_UART1_TX,
 		.end = IRQ_UART1_TX,
@@ -353,6 +370,13 @@ static struct resource bfin_uart2_resources[] = {
 		.end = UART2_RBR+2,
 		.flags = IORESOURCE_MEM,
 	},
+#ifdef CONFIG_EARLY_PRINTK
+	{
+		.start = PORTB_FER,
+		.end = PORTB_FER+2,
+		.flags = IORESOURCE_REG,
+	},
+#endif
 	{
 		.start = IRQ_UART2_TX,
 		.end = IRQ_UART2_TX,
@@ -401,6 +425,13 @@ static struct resource bfin_uart3_resources[] = {
 		.end = UART3_RBR+2,
 		.flags = IORESOURCE_MEM,
 	},
+#ifdef CONFIG_EARLY_PRINTK
+	{
+		.start = PORTB_FER,
+		.end = PORTB_FER+2,
+		.flags = IORESOURCE_REG,
+	},
+#endif
 	{
 		.start = IRQ_UART3_TX,
 		.end = IRQ_UART3_TX,
@@ -1058,6 +1089,411 @@ static const struct ad7877_platform_data bfin_ad7877_ts_info = {
 };
 #endif
 
+#ifdef CONFIG_PINCTRL_ADI2
+
+# define ADI_PINT_DEVNAME "adi-gpio-pint"
+# define ADI_GPIO_DEVNAME "adi-gpio"
+# define ADI_PINCTRL_DEVNAME "pinctrl-adi2"
+
+static struct platform_device bfin_pinctrl_device = {
+	.name = ADI_PINCTRL_DEVNAME,
+	.id = 0,
+};
+
+static struct resource bfin_pint0_resources[] = {
+	{
+		.start = PINT0_MASK_SET,
+		.end = PINT0_LATCH + 3,
+		.flags = IORESOURCE_MEM,
+	},
+	{
+		.start = IRQ_PINT0,
+		.end = IRQ_PINT0,
+		.flags = IORESOURCE_IRQ,
+	},
+};
+
+static struct platform_device bfin_pint0_device = {
+	.name = ADI_PINT_DEVNAME,
+	.id = 0,
+	.num_resources = ARRAY_SIZE(bfin_pint0_resources),
+	.resource = bfin_pint0_resources,
+};
+
+static struct resource bfin_pint1_resources[] = {
+	{
+		.start = PINT1_MASK_SET,
+		.end = PINT1_LATCH + 3,
+		.flags = IORESOURCE_MEM,
+	},
+	{
+		.start = IRQ_PINT1,
+		.end = IRQ_PINT1,
+		.flags = IORESOURCE_IRQ,
+	},
+};
+
+static struct platform_device bfin_pint1_device = {
+	.name = ADI_PINT_DEVNAME,
+	.id = 1,
+	.num_resources = ARRAY_SIZE(bfin_pint1_resources),
+	.resource = bfin_pint1_resources,
+};
+
+static struct resource bfin_pint2_resources[] = {
+	{
+		.start = PINT2_MASK_SET,
+		.end = PINT2_LATCH + 3,
+		.flags = IORESOURCE_MEM,
+	},
+	{
+		.start = IRQ_PINT2,
+		.end = IRQ_PINT2,
+		.flags = IORESOURCE_IRQ,
+	},
+};
+
+static struct platform_device bfin_pint2_device = {
+	.name = ADI_PINT_DEVNAME,
+	.id = 2,
+	.num_resources = ARRAY_SIZE(bfin_pint2_resources),
+	.resource = bfin_pint2_resources,
+};
+
+static struct resource bfin_pint3_resources[] = {
+	{
+		.start = PINT3_MASK_SET,
+		.end = PINT3_LATCH + 3,
+		.flags = IORESOURCE_MEM,
+	},
+	{
+		.start = IRQ_PINT3,
+		.end = IRQ_PINT3,
+		.flags = IORESOURCE_IRQ,
+	},
+};
+
+static struct platform_device bfin_pint3_device = {
+	.name = ADI_PINT_DEVNAME,
+	.id = 3,
+	.num_resources = ARRAY_SIZE(bfin_pint3_resources),
+	.resource = bfin_pint3_resources,
+};
+
+static struct resource bfin_gpa_resources[] = {
+	{
+		.start = PORTA_FER,
+		.end = PORTA_MUX + 3,
+		.flags = IORESOURCE_MEM,
+	},
+	{	/* optional */
+		.start = IRQ_PA0,
+		.end = IRQ_PA0,
+		.flags = IORESOURCE_IRQ,
+	},
+};
+
+static struct adi_pinctrl_gpio_platform_data bfin_gpa_pdata = {
+	.port_gpio_base	= GPIO_PA0,	/* Optional */
+	.port_pin_base	= GPIO_PA0,
+	.port_width	= GPIO_BANKSIZE,
+	.pint_id	= 0,		/* PINT0 */
+	.pint_assign	= true,		/* PINT upper 16 bit */
+	.pint_map	= 0,		/* mapping mask in PINT */
+};
+
+static struct platform_device bfin_gpa_device = {
+	.name = ADI_GPIO_DEVNAME,
+	.id = 0,
+	.num_resources = ARRAY_SIZE(bfin_gpa_resources),
+	.resource = bfin_gpa_resources,
+	.dev = {
+		.platform_data = &bfin_gpa_pdata, /* Passed to driver */
+	},
+};
+
+static struct resource bfin_gpb_resources[] = {
+	{
+		.start = PORTB_FER,
+		.end = PORTB_MUX + 3,
+		.flags = IORESOURCE_MEM,
+	},
+	{
+		.start = IRQ_PB0,
+		.end = IRQ_PB0,
+		.flags = IORESOURCE_IRQ,
+	},
+};
+
+static struct adi_pinctrl_gpio_platform_data bfin_gpb_pdata = {
+	.port_gpio_base	= GPIO_PB0,
+	.port_pin_base	= GPIO_PB0,
+	.port_width	= 15,
+	.pint_id	= 0,
+	.pint_assign	= true,
+	.pint_map	= 1,
+};
+
+static struct platform_device bfin_gpb_device = {
+	.name = ADI_GPIO_DEVNAME,
+	.id = 1,
+	.num_resources = ARRAY_SIZE(bfin_gpb_resources),
+	.resource = bfin_gpb_resources,
+	.dev = {
+		.platform_data = &bfin_gpb_pdata, /* Passed to driver */
+	},
+};
+
+static struct resource bfin_gpc_resources[] = {
+	{
+		.start = PORTC_FER,
+		.end = PORTC_MUX + 3,
+		.flags = IORESOURCE_MEM,
+	},
+	{
+		.start = IRQ_PC0,
+		.end = IRQ_PC0,
+		.flags = IORESOURCE_IRQ,
+	},
+};
+
+static struct adi_pinctrl_gpio_platform_data bfin_gpc_pdata = {
+	.port_gpio_base	= GPIO_PC0,
+	.port_pin_base	= GPIO_PC0,
+	.port_width	= 14,
+	.pint_id	= 2,
+	.pint_assign	= true,
+	.pint_map	= 0,
+};
+
+static struct platform_device bfin_gpc_device = {
+	.name = ADI_GPIO_DEVNAME,
+	.id = 2,
+	.num_resources = ARRAY_SIZE(bfin_gpc_resources),
+	.resource = bfin_gpc_resources,
+	.dev = {
+		.platform_data = &bfin_gpc_pdata, /* Passed to driver */
+	},
+};
+
+static struct resource bfin_gpd_resources[] = {
+	{
+		.start = PORTD_FER,
+		.end = PORTD_MUX + 3,
+		.flags = IORESOURCE_MEM,
+	},
+	{
+		.start = IRQ_PD0,
+		.end = IRQ_PD0,
+		.flags = IORESOURCE_IRQ,
+	},
+};
+
+static struct adi_pinctrl_gpio_platform_data bfin_gpd_pdata = {
+	.port_gpio_base	= GPIO_PD0,
+	.port_pin_base	= GPIO_PD0,
+	.port_width	= GPIO_BANKSIZE,
+	.pint_id	= 2,
+	.pint_assign	= false,
+	.pint_map	= 1,
+};
+
+static struct platform_device bfin_gpd_device = {
+	.name = ADI_GPIO_DEVNAME,
+	.id = 3,
+	.num_resources = ARRAY_SIZE(bfin_gpd_resources),
+	.resource = bfin_gpd_resources,
+	.dev = {
+		.platform_data = &bfin_gpd_pdata, /* Passed to driver */
+	},
+};
+
+static struct resource bfin_gpe_resources[] = {
+	{
+		.start = PORTE_FER,
+		.end = PORTE_MUX + 3,
+		.flags = IORESOURCE_MEM,
+	},
+	{
+		.start = IRQ_PE0,
+		.end = IRQ_PE0,
+		.flags = IORESOURCE_IRQ,
+	},
+};
+
+static struct adi_pinctrl_gpio_platform_data bfin_gpe_pdata = {
+	.port_gpio_base	= GPIO_PE0,
+	.port_pin_base	= GPIO_PE0,
+	.port_width	= GPIO_BANKSIZE,
+	.pint_id	= 3,
+	.pint_assign	= true,
+	.pint_map	= 2,
+};
+
+static struct platform_device bfin_gpe_device = {
+	.name = ADI_GPIO_DEVNAME,
+	.id = 4,
+	.num_resources = ARRAY_SIZE(bfin_gpe_resources),
+	.resource = bfin_gpe_resources,
+	.dev = {
+		.platform_data = &bfin_gpe_pdata, /* Passed to driver */
+	},
+};
+
+static struct resource bfin_gpf_resources[] = {
+	{
+		.start = PORTF_FER,
+		.end = PORTF_MUX + 3,
+		.flags = IORESOURCE_MEM,
+	},
+	{
+		.start = IRQ_PF0,
+		.end = IRQ_PF0,
+		.flags = IORESOURCE_IRQ,
+	},
+};
+
+static struct adi_pinctrl_gpio_platform_data bfin_gpf_pdata = {
+	.port_gpio_base	= GPIO_PF0,
+	.port_pin_base	= GPIO_PF0,
+	.port_width	= GPIO_BANKSIZE,
+	.pint_id	= 3,
+	.pint_assign	= false,
+	.pint_map	= 3,
+};
+
+static struct platform_device bfin_gpf_device = {
+	.name = ADI_GPIO_DEVNAME,
+	.id = 5,
+	.num_resources = ARRAY_SIZE(bfin_gpf_resources),
+	.resource = bfin_gpf_resources,
+	.dev = {
+		.platform_data = &bfin_gpf_pdata, /* Passed to driver */
+	},
+};
+
+static struct resource bfin_gpg_resources[] = {
+	{
+		.start = PORTG_FER,
+		.end = PORTG_MUX + 3,
+		.flags = IORESOURCE_MEM,
+	},
+	{
+		.start = IRQ_PG0,
+		.end = IRQ_PG0,
+		.flags = IORESOURCE_IRQ,
+	},
+};
+
+static struct adi_pinctrl_gpio_platform_data bfin_gpg_pdata = {
+	.port_gpio_base	= GPIO_PG0,
+	.port_pin_base	= GPIO_PG0,
+	.port_width	= GPIO_BANKSIZE,
+	.pint_id	= -1,
+};
+
+static struct platform_device bfin_gpg_device = {
+	.name = ADI_GPIO_DEVNAME,
+	.id = 6,
+	.num_resources = ARRAY_SIZE(bfin_gpg_resources),
+	.resource = bfin_gpg_resources,
+	.dev = {
+		.platform_data = &bfin_gpg_pdata, /* Passed to driver */
+	},
+};
+
+static struct resource bfin_gph_resources[] = {
+	{
+		.start = PORTH_FER,
+		.end = PORTH_MUX + 3,
+		.flags = IORESOURCE_MEM,
+	},
+	{
+		.start = IRQ_PH0,
+		.end = IRQ_PH0,
+		.flags = IORESOURCE_IRQ,
+	},
+};
+
+static struct adi_pinctrl_gpio_platform_data bfin_gph_pdata = {
+	.port_gpio_base	= GPIO_PH0,
+	.port_pin_base	= GPIO_PH0,
+	.port_width	= 14,
+	.pint_id	= -1,
+};
+
+static struct platform_device bfin_gph_device = {
+	.name = ADI_GPIO_DEVNAME,
+	.id = 7,
+	.num_resources = ARRAY_SIZE(bfin_gph_resources),
+	.resource = bfin_gph_resources,
+	.dev = {
+		.platform_data = &bfin_gph_pdata, /* Passed to driver */
+	},
+};
+
+static struct resource bfin_gpi_resources[] = {
+	{
+		.start = PORTI_FER,
+		.end = PORTI_MUX + 3,
+		.flags = IORESOURCE_MEM,
+	},
+	{
+		.start = IRQ_PI0,
+		.end = IRQ_PI0,
+		.flags = IORESOURCE_IRQ,
+	},
+};
+
+static struct adi_pinctrl_gpio_platform_data bfin_gpi_pdata = {
+	.port_gpio_base	= GPIO_PI0,
+	.port_pin_base	= GPIO_PI0,
+	.port_width	= GPIO_BANKSIZE,
+	.pint_id	= -1,
+};
+
+static struct platform_device bfin_gpi_device = {
+	.name = ADI_GPIO_DEVNAME,
+	.id = 8,
+	.num_resources = ARRAY_SIZE(bfin_gpi_resources),
+	.resource = bfin_gpi_resources,
+	.dev = {
+		.platform_data = &bfin_gpi_pdata, /* Passed to driver */
+	},
+};
+
+static struct resource bfin_gpj_resources[] = {
+	{
+		.start = PORTJ_FER,
+		.end = PORTJ_MUX + 3,
+		.flags = IORESOURCE_MEM,
+	},
+	{
+		.start = IRQ_PJ0,
+		.end = IRQ_PJ0,
+		.flags = IORESOURCE_IRQ,
+	},
+};
+
+static struct adi_pinctrl_gpio_platform_data bfin_gpj_pdata = {
+	.port_gpio_base	= GPIO_PJ0,
+	.port_pin_base	= GPIO_PJ0,
+	.port_width	= 14,
+	.pint_id	= -1,
+};
+
+static struct platform_device bfin_gpj_device = {
+	.name = ADI_GPIO_DEVNAME,
+	.id = 9,
+	.num_resources = ARRAY_SIZE(bfin_gpj_resources),
+	.resource = bfin_gpj_resources,
+	.dev = {
+		.platform_data = &bfin_gpj_pdata, /* Passed to driver */
+	},
+};
+
+#endif
+
 static struct spi_board_info bfin_spi_board_info[] __initdata = {
 #if defined(CONFIG_MTD_M25P80) \
 	|| defined(CONFIG_MTD_M25P80_MODULE)
@@ -1066,7 +1502,7 @@ static struct spi_board_info bfin_spi_board_info[] __initdata = {
 		.modalias = "m25p80", /* Name of spi_driver for this device */
 		.max_speed_hz = 25000000,     /* max spi clock (SCK) speed in HZ */
 		.bus_num = 0, /* Framework bus number */
-		.chip_select = 1, /* SPI_SSEL1*/
+		.chip_select = MAX_CTRL_CS + GPIO_PE4, /* SPI_SSEL1*/
 		.platform_data = &bfin_spi_flash_data,
 		.controller_data = &spi_flash_chip_info,
 		.mode = SPI_MODE_3,
@@ -1078,7 +1514,7 @@ static struct spi_board_info bfin_spi_board_info[] __initdata = {
 		.modalias = "ad183x",
 		.max_speed_hz = 3125000,     /* max spi clock (SCK) speed in HZ */
 		.bus_num = 1,
-		.chip_select = 4,
+		.chip_select = MAX_CTRL_CS + GPIO_PG6, /* SPI_SSEL2 */
 	},
 #endif
 #if defined(CONFIG_TOUCHSCREEN_AD7877) || defined(CONFIG_TOUCHSCREEN_AD7877_MODULE)
@@ -1088,7 +1524,7 @@ static struct spi_board_info bfin_spi_board_info[] __initdata = {
 		.irq			= IRQ_PB4,	/* old boards (<=Rev 1.3) use IRQ_PJ11 */
 		.max_speed_hz		= 12500000,     /* max spi clock (SCK) speed in HZ */
 		.bus_num		= 0,
-		.chip_select  		= 2,
+		.chip_select		= MAX_CTRL_CS + GPIO_PE5, /* SPI_SSEL2 */
 	},
 #endif
 #if defined(CONFIG_SPI_SPIDEV) || defined(CONFIG_SPI_SPIDEV_MODULE)
@@ -1096,7 +1532,7 @@ static struct spi_board_info bfin_spi_board_info[] __initdata = {
 		.modalias = "spidev",
 		.max_speed_hz = 3125000,     /* max spi clock (SCK) speed in HZ */
 		.bus_num = 0,
-		.chip_select = 1,
+		.chip_select = MAX_CTRL_CS + GPIO_PE4, /* SPI_SSEL1 */
 	},
 #endif
 #if defined(CONFIG_INPUT_ADXL34X_SPI) || defined(CONFIG_INPUT_ADXL34X_SPI_MODULE)
@@ -1106,7 +1542,7 @@ static struct spi_board_info bfin_spi_board_info[] __initdata = {
 		.irq			= IRQ_PC5,
 		.max_speed_hz		= 5000000,     /* max spi clock (SCK) speed in HZ */
 		.bus_num		= 1,
-		.chip_select  		= 2,
+		.chip_select		= MAX_CTRL_CS + GPIO_PG6, /* SPI_SSEL2 */
 		.mode = SPI_MODE_3,
 	},
 #endif
@@ -1152,7 +1588,7 @@ static struct resource bfin_spi1_resource[] = {
 
 /* SPI controller data */
 static struct bfin5xx_spi_master bf54x_spi_master_info0 = {
-	.num_chipselect = 4,
+	.num_chipselect = MAX_CTRL_CS + MAX_BLACKFIN_GPIOS,
 	.enable_dma = 1,  /* master has the ability to do dma transfer */
 	.pin_req = {P_SPI0_SCK, P_SPI0_MISO, P_SPI0_MOSI, 0},
 };
@@ -1168,7 +1604,7 @@ static struct platform_device bf54x_spi_master0 = {
 };
 
 static struct bfin5xx_spi_master bf54x_spi_master_info1 = {
-	.num_chipselect = 4,
+	.num_chipselect = MAX_CTRL_CS + MAX_BLACKFIN_GPIOS,
 	.enable_dma = 1,  /* master has the ability to do dma transfer */
 	.pin_req = {P_SPI1_SCK, P_SPI1_MISO, P_SPI1_MOSI, 0},
 };
@@ -1508,6 +1944,23 @@ static struct platform_device bfin_ac97 = {
 static struct platform_device *ezkit_devices[] __initdata = {
 
 	&bfin_dpmc,
+#if defined(CONFIG_PINCTRL_ADI2)
+	&bfin_pinctrl_device,
+	&bfin_pint0_device,
+	&bfin_pint1_device,
+	&bfin_pint2_device,
+	&bfin_pint3_device,
+	&bfin_gpa_device,
+	&bfin_gpb_device,
+	&bfin_gpc_device,
+	&bfin_gpd_device,
+	&bfin_gpe_device,
+	&bfin_gpf_device,
+	&bfin_gpg_device,
+	&bfin_gph_device,
+	&bfin_gpi_device,
+	&bfin_gpj_device,
+#endif
 
 #if defined(CONFIG_RTC_DRV_BFIN) || defined(CONFIG_RTC_DRV_BFIN_MODULE)
 	&rtc_device,
@@ -1644,10 +2097,66 @@ static struct platform_device *ezkit_devices[] __initdata = {
 #endif
 };
 
+/* Pin control settings */
+static struct pinctrl_map __initdata bfin_pinmux_map[] = {
+	/* per-device maps */
+	PIN_MAP_MUX_GROUP_DEFAULT("bfin-uart.0",  "pinctrl-adi2.0", NULL, "uart0"),
+	PIN_MAP_MUX_GROUP_DEFAULT("bfin-uart.1",  "pinctrl-adi2.0", NULL, "uart1"),
+#ifdef CONFIG_BFIN_UART1_CTSRTS
+	PIN_MAP_MUX_GROUP_DEFAULT("bfin-uart.1",  "pinctrl-adi2.0", NULL, "uart1_ctsrts"),
+#endif
+	PIN_MAP_MUX_GROUP_DEFAULT("bfin-uart.2",  "pinctrl-adi2.0", NULL, "uart2"),
+	PIN_MAP_MUX_GROUP_DEFAULT("bfin-uart.3",  "pinctrl-adi2.0", NULL, "uart3"),
+#ifdef CONFIG_BFIN_UART3_CTSRTS
+	PIN_MAP_MUX_GROUP_DEFAULT("bfin-uart.3",  "pinctrl-adi2.0", NULL, "uart3_ctsrts"),
+#endif
+	PIN_MAP_MUX_GROUP_DEFAULT("bfin_sir.0",  "pinctrl-adi2.0", NULL, "uart0"),
+	PIN_MAP_MUX_GROUP_DEFAULT("bfin_sir.1",  "pinctrl-adi2.0", NULL, "uart1"),
+	PIN_MAP_MUX_GROUP_DEFAULT("bfin_sir.2",  "pinctrl-adi2.0", NULL, "uart2"),
+	PIN_MAP_MUX_GROUP_DEFAULT("bfin_sir.3",  "pinctrl-adi2.0", NULL, "uart3"),
+	PIN_MAP_MUX_GROUP_DEFAULT("bfin-sdh.0",  "pinctrl-adi2.0", NULL, "rsi0"),
+	PIN_MAP_MUX_GROUP_DEFAULT("bfin-spi.0",  "pinctrl-adi2.0", NULL, "spi0"),
+	PIN_MAP_MUX_GROUP_DEFAULT("bfin-spi.1",  "pinctrl-adi2.0", NULL, "spi1"),
+	PIN_MAP_MUX_GROUP_DEFAULT("i2c-bfin-twi.0",  "pinctrl-adi2.0", NULL, "twi0"),
+#if !defined(CONFIG_BF542)	/* The BF542 only has 1 TWI */
+	PIN_MAP_MUX_GROUP_DEFAULT("i2c-bfin-twi.1",  "pinctrl-adi2.0", NULL, "twi1"),
+#endif
+	PIN_MAP_MUX_GROUP_DEFAULT("bfin-rotary",  "pinctrl-adi2.0", NULL, "rotary"),
+	PIN_MAP_MUX_GROUP_DEFAULT("bfin_can.0",  "pinctrl-adi2.0", NULL, "can0"),
+	PIN_MAP_MUX_GROUP_DEFAULT("bfin_can.1",  "pinctrl-adi2.0", NULL, "can1"),
+	PIN_MAP_MUX_GROUP_DEFAULT("bf54x-lq043",  "pinctrl-adi2.0", NULL, "ppi0_24b"),
+	PIN_MAP_MUX_GROUP_DEFAULT("bfin-i2s.0",  "pinctrl-adi2.0", NULL, "sport0"),
+	PIN_MAP_MUX_GROUP_DEFAULT("bfin-tdm.0",  "pinctrl-adi2.0", NULL, "sport0"),
+	PIN_MAP_MUX_GROUP_DEFAULT("bfin-ac97.0",  "pinctrl-adi2.0", NULL, "sport0"),
+	PIN_MAP_MUX_GROUP_DEFAULT("bfin-i2s.1",  "pinctrl-adi2.0", NULL, "sport1"),
+	PIN_MAP_MUX_GROUP_DEFAULT("bfin-tdm.1",  "pinctrl-adi2.0", NULL, "sport1"),
+	PIN_MAP_MUX_GROUP_DEFAULT("bfin-ac97.1",  "pinctrl-adi2.0", NULL, "sport1"),
+	PIN_MAP_MUX_GROUP_DEFAULT("bfin-i2s.2",  "pinctrl-adi2.0", NULL, "sport2"),
+	PIN_MAP_MUX_GROUP_DEFAULT("bfin-tdm.2",  "pinctrl-adi2.0", NULL, "sport2"),
+	PIN_MAP_MUX_GROUP_DEFAULT("bfin-ac97.2",  "pinctrl-adi2.0", NULL, "sport2"),
+	PIN_MAP_MUX_GROUP_DEFAULT("bfin-i2s.3",  "pinctrl-adi2.0", NULL, "sport3"),
+	PIN_MAP_MUX_GROUP_DEFAULT("bfin-tdm.3",  "pinctrl-adi2.0", NULL, "sport3"),
+	PIN_MAP_MUX_GROUP_DEFAULT("bfin-ac97.3",  "pinctrl-adi2.0", NULL, "sport3"),
+	PIN_MAP_MUX_GROUP_DEFAULT("bfin-sport-uart.0",  "pinctrl-adi2.0", NULL, "sport0"),
+	PIN_MAP_MUX_GROUP_DEFAULT("bfin-sport-uart.1",  "pinctrl-adi2.0", NULL, "sport1"),
+	PIN_MAP_MUX_GROUP_DEFAULT("bfin-sport-uart.2",  "pinctrl-adi2.0", NULL, "sport2"),
+	PIN_MAP_MUX_GROUP_DEFAULT("bfin-sport-uart.3",  "pinctrl-adi2.0", NULL, "sport3"),
+	PIN_MAP_MUX_GROUP_DEFAULT("pata-bf54x",  "pinctrl-adi2.0", NULL, "atapi"),
+#ifdef CONFIG_BF548_ATAPI_ALTERNATIVE_PORT
+	PIN_MAP_MUX_GROUP_DEFAULT("pata-bf54x",  "pinctrl-adi2.0", NULL, "atapi_alter"),
+#endif
+	PIN_MAP_MUX_GROUP_DEFAULT("bf5xx-nand.0",  "pinctrl-adi2.0", NULL, "nfc0"),
+	PIN_MAP_MUX_GROUP_DEFAULT("bf54x-keys",  "pinctrl-adi2.0", NULL, "keys_4x4"),
+};
+
 static int __init ezkit_init(void)
 {
 	printk(KERN_INFO "%s(): registering device resources\n", __func__);
 
+	/* Initialize pinmuxing */
+	pinctrl_register_mappings(bfin_pinmux_map,
+				ARRAY_SIZE(bfin_pinmux_map));
+
 	i2c_register_board_info(0, bfin_i2c_board_info0,
 				ARRAY_SIZE(bfin_i2c_board_info0));
 #if !defined(CONFIG_BF542)	/* The BF542 only has 1 TWI */
@@ -1679,21 +2188,6 @@ static struct platform_device *ezkit_early_devices[] __initdata = {
 	&bfin_uart3_device,
 #endif
 #endif
-
-#if defined(CONFIG_SERIAL_BFIN_SPORT_CONSOLE)
-#ifdef CONFIG_SERIAL_BFIN_SPORT0_UART
-	&bfin_sport0_uart_device,
-#endif
-#ifdef CONFIG_SERIAL_BFIN_SPORT1_UART
-	&bfin_sport1_uart_device,
-#endif
-#ifdef CONFIG_SERIAL_BFIN_SPORT2_UART
-	&bfin_sport2_uart_device,
-#endif
-#ifdef CONFIG_SERIAL_BFIN_SPORT3_UART
-	&bfin_sport3_uart_device,
-#endif
-#endif
 };
 
 void __init native_machine_early_platform_add_devices(void)
diff --git a/arch/blackfin/mach-bf548/include/mach/gpio.h b/arch/blackfin/mach-bf548/include/mach/gpio.h
index be9edb28f96b..006da1edcf84 100644
--- a/arch/blackfin/mach-bf548/include/mach/gpio.h
+++ b/arch/blackfin/mach-bf548/include/mach/gpio.h
@@ -194,14 +194,6 @@ struct gpio_port_t {
 	unsigned int port_mux;
 };
 
-struct gpio_port_s {
-	unsigned short fer;
-	unsigned short data;
-	unsigned short dir;
-	unsigned short inen;
-	unsigned int mux;
-};
-
 #endif
 
 #include <mach-common/ports-a.h>
diff --git a/arch/blackfin/mach-bf548/include/mach/irq.h b/arch/blackfin/mach-bf548/include/mach/irq.h
index 10dc142c518d..cf7cb725cfa2 100644
--- a/arch/blackfin/mach-bf548/include/mach/irq.h
+++ b/arch/blackfin/mach-bf548/include/mach/irq.h
@@ -433,7 +433,7 @@
 #include <linux/types.h>
 
 /*
- * bfin pint registers layout
+ * gpio pint registers layout
  */
 struct bfin_pint_regs {
 	u32 mask_set;
diff --git a/arch/blackfin/mach-bf609/Kconfig b/arch/blackfin/mach-bf609/Kconfig
index 2bcbf94b1edf..b0fca44110b0 100644
--- a/arch/blackfin/mach-bf609/Kconfig
+++ b/arch/blackfin/mach-bf609/Kconfig
@@ -9,48 +9,6 @@ source "arch/blackfin/mach-bf609/boards/Kconfig"
 
 menu "BF609 Specific Configuration"
 
-comment "Pin Interrupt to Port Assignment"
-menu "Assignment"
-
-config PINTx_REASSIGN
-	bool "Reprogram PINT Assignment"
-	default y
-	help
-	  The interrupt assignment registers controls the pin-to-interrupt
-	  assignment in a byte-wide manner. Each option allows you to select
-	  a set of pins (High/Low Byte) of an specific Port being mapped
-	  to one of the four PIN Interrupts IRQ_PINTx.
-
-	  You shouldn't change any of these unless you know exactly what you're doing.
-	  Please consult the Blackfin BF60x Processor Hardware Reference Manual.
-
-config PINT0_ASSIGN
-	hex "PINT0_ASSIGN"
-	depends on PINTx_REASSIGN
-	default 0x00000101
-config PINT1_ASSIGN
-	hex "PINT1_ASSIGN"
-	depends on PINTx_REASSIGN
-	default 0x00000101
-config PINT2_ASSIGN
-	hex "PINT2_ASSIGN"
-	depends on PINTx_REASSIGN
-	default 0x00000101
-config PINT3_ASSIGN
-	hex "PINT3_ASSIGN"
-	depends on PINTx_REASSIGN
-	default 0x00000101
-config PINT4_ASSIGN
-	hex "PINT3_ASSIGN"
-	depends on PINTx_REASSIGN
-	default 0x00000101
-config PINT5_ASSIGN
-	hex "PINT3_ASSIGN"
-	depends on PINTx_REASSIGN
-	default 0x00000101
-
-endmenu
-
 config SEC_IRQ_PRIORITY_LEVELS
 	int "SEC interrupt priority levels"
 	default 7
diff --git a/arch/blackfin/mach-bf609/boards/ezkit.c b/arch/blackfin/mach-bf609/boards/ezkit.c
index d56a55ad83a7..82beedd953f6 100644
--- a/arch/blackfin/mach-bf609/boards/ezkit.c
+++ b/arch/blackfin/mach-bf609/boards/ezkit.c
@@ -17,6 +17,9 @@
 #include <linux/i2c.h>
 #include <linux/interrupt.h>
 #include <linux/usb/musb.h>
+#include <linux/pinctrl/machine.h>
+#include <linux/pinctrl/pinconf-generic.h>
+#include <linux/platform_data/pinctrl-adi2.h>
 #include <asm/bfin_spi3.h>
 #include <asm/dma.h>
 #include <asm/gpio.h>
@@ -106,8 +109,6 @@ static struct platform_device bfin_rotary_device = {
 #include <linux/stmmac.h>
 #include <linux/phy.h>
 
-static unsigned short pins[] = P_RMII0;
-
 static struct stmmac_mdio_bus_data phy_private_data = {
 	.phy_mask = 1,
 };
@@ -212,6 +213,18 @@ static struct resource bfin_uart0_resources[] = {
 		.end = UART0_RXDIV+4,
 		.flags = IORESOURCE_MEM,
 	},
+#ifdef CONFIG_EARLY_PRINTK
+	{
+		.start = PORTD_FER,
+		.end = PORTD_FER+2,
+		.flags = IORESOURCE_REG,
+	},
+	{
+		.start = PORTD_MUX,
+		.end = PORTD_MUX+3,
+		.flags = IORESOURCE_REG,
+	},
+#endif
 	{
 		.start = IRQ_UART0_TX,
 		.end = IRQ_UART0_TX,
@@ -276,6 +289,13 @@ static struct resource bfin_uart1_resources[] = {
 		.end = UART1_RXDIV+4,
 		.flags = IORESOURCE_MEM,
 	},
+#ifdef CONFIG_EARLY_PRINTK
+	{
+		.start = PORTG_FER_SET,
+		.end = PORTG_FER_SET+2,
+		.flags = IORESOURCE_REG,
+	},
+#endif
 	{
 		.start = IRQ_UART1_TX,
 		.end = IRQ_UART1_TX,
@@ -674,17 +694,12 @@ static struct mtd_partition ezkit_partitions[] = {
 	},
 };
 
-int bf609_nor_flash_init(struct platform_device *dev)
+int bf609_nor_flash_init(struct platform_device *pdev)
 {
 #define CONFIG_SMC_GCTL_VAL     0x00000010
-	const unsigned short pins[] = {
-		P_A3, P_A4, P_A5, P_A6, P_A7, P_A8, P_A9, P_A10, P_A11, P_A12,
-		P_A13, P_A14, P_A15, P_A16, P_A17, P_A18, P_A19, P_A20, P_A21,
-		P_A22, P_A23, P_A24, P_A25, P_NORCK, 0,
-	};
-
-	peripheral_request_list(pins, "smc0");
 
+	if (!devm_pinctrl_get_select_default(&pdev->dev))
+		return -EBUSY;
 	bfin_write32(SMC_GCTL, CONFIG_SMC_GCTL_VAL);
 	bfin_write32(SMC_B0CTL, 0x01002011);
 	bfin_write32(SMC_B0TIM, 0x08170977);
@@ -692,16 +707,9 @@ int bf609_nor_flash_init(struct platform_device *dev)
 	return 0;
 }
 
-void bf609_nor_flash_exit(struct platform_device *dev)
+void bf609_nor_flash_exit(struct platform_device *pdev)
 {
-	const unsigned short pins[] = {
-		P_A3, P_A4, P_A5, P_A6, P_A7, P_A8, P_A9, P_A10, P_A11, P_A12,
-		P_A13, P_A14, P_A15, P_A16, P_A17, P_A18, P_A19, P_A20, P_A21,
-		P_A22, P_A23, P_A24, P_A25, P_NORCK, 0,
-	};
-
-	peripheral_free_list(pins);
-
+	devm_pinctrl_put(pdev->dev.pins->p);
 	bfin_write32(SMC_GCTL, 0);
 }
 
@@ -1319,6 +1327,356 @@ static const struct ad7877_platform_data bfin_ad7877_ts_info = {
 };
 #endif
 
+#ifdef CONFIG_PINCTRL_ADI2
+
+# define ADI_PINT_DEVNAME "adi-gpio-pint"
+# define ADI_GPIO_DEVNAME "adi-gpio"
+# define ADI_PINCTRL_DEVNAME "pinctrl-adi2"
+
+static struct platform_device bfin_pinctrl_device = {
+	.name = ADI_PINCTRL_DEVNAME,
+	.id = 0,
+};
+
+static struct resource bfin_pint0_resources[] = {
+	{
+		.start = PINT0_MASK_SET,
+		.end = PINT0_LATCH + 3,
+		.flags = IORESOURCE_MEM,
+	},
+	{
+		.start = IRQ_PINT0,
+		.end = IRQ_PINT0,
+		.flags = IORESOURCE_IRQ,
+	},
+};
+
+static struct platform_device bfin_pint0_device = {
+	.name = ADI_PINT_DEVNAME,
+	.id = 0,
+	.num_resources = ARRAY_SIZE(bfin_pint0_resources),
+	.resource = bfin_pint0_resources,
+};
+
+static struct resource bfin_pint1_resources[] = {
+	{
+		.start = PINT1_MASK_SET,
+		.end = PINT1_LATCH + 3,
+		.flags = IORESOURCE_MEM,
+	},
+	{
+		.start = IRQ_PINT1,
+		.end = IRQ_PINT1,
+		.flags = IORESOURCE_IRQ,
+	},
+};
+
+static struct platform_device bfin_pint1_device = {
+	.name = ADI_PINT_DEVNAME,
+	.id = 1,
+	.num_resources = ARRAY_SIZE(bfin_pint1_resources),
+	.resource = bfin_pint1_resources,
+};
+
+static struct resource bfin_pint2_resources[] = {
+	{
+		.start = PINT2_MASK_SET,
+		.end = PINT2_LATCH + 3,
+		.flags = IORESOURCE_MEM,
+	},
+	{
+		.start = IRQ_PINT2,
+		.end = IRQ_PINT2,
+		.flags = IORESOURCE_IRQ,
+	},
+};
+
+static struct platform_device bfin_pint2_device = {
+	.name = ADI_PINT_DEVNAME,
+	.id = 2,
+	.num_resources = ARRAY_SIZE(bfin_pint2_resources),
+	.resource = bfin_pint2_resources,
+};
+
+static struct resource bfin_pint3_resources[] = {
+	{
+		.start = PINT3_MASK_SET,
+		.end = PINT3_LATCH + 3,
+		.flags = IORESOURCE_MEM,
+	},
+	{
+		.start = IRQ_PINT3,
+		.end = IRQ_PINT3,
+		.flags = IORESOURCE_IRQ,
+	},
+};
+
+static struct platform_device bfin_pint3_device = {
+	.name = ADI_PINT_DEVNAME,
+	.id = 3,
+	.num_resources = ARRAY_SIZE(bfin_pint3_resources),
+	.resource = bfin_pint3_resources,
+};
+
+static struct resource bfin_pint4_resources[] = {
+	{
+		.start = PINT4_MASK_SET,
+		.end = PINT4_LATCH + 3,
+		.flags = IORESOURCE_MEM,
+	},
+	{
+		.start = IRQ_PINT4,
+		.end = IRQ_PINT4,
+		.flags = IORESOURCE_IRQ,
+	},
+};
+
+static struct platform_device bfin_pint4_device = {
+	.name = ADI_PINT_DEVNAME,
+	.id = 4,
+	.num_resources = ARRAY_SIZE(bfin_pint4_resources),
+	.resource = bfin_pint4_resources,
+};
+
+static struct resource bfin_pint5_resources[] = {
+	{
+		.start = PINT5_MASK_SET,
+		.end = PINT5_LATCH + 3,
+		.flags = IORESOURCE_MEM,
+	},
+	{
+		.start = IRQ_PINT5,
+		.end = IRQ_PINT5,
+		.flags = IORESOURCE_IRQ,
+	},
+};
+
+static struct platform_device bfin_pint5_device = {
+	.name = ADI_PINT_DEVNAME,
+	.id = 5,
+	.num_resources = ARRAY_SIZE(bfin_pint5_resources),
+	.resource = bfin_pint5_resources,
+};
+
+static struct resource bfin_gpa_resources[] = {
+	{
+		.start = PORTA_FER,
+		.end = PORTA_MUX + 3,
+		.flags = IORESOURCE_MEM,
+	},
+	{	/* optional */
+		.start = IRQ_PA0,
+		.end = IRQ_PA0,
+		.flags = IORESOURCE_IRQ,
+	},
+};
+
+static struct adi_pinctrl_gpio_platform_data bfin_gpa_pdata = {
+	.port_pin_base	= GPIO_PA0,
+	.port_width	= GPIO_BANKSIZE,
+	.pint_id	= 0,		/* PINT0 */
+	.pint_assign	= true,		/* PINT upper 16 bit */
+	.pint_map	= 0,		/* mapping mask in PINT */
+};
+
+static struct platform_device bfin_gpa_device = {
+	.name = ADI_GPIO_DEVNAME,
+	.id = 0,
+	.num_resources = ARRAY_SIZE(bfin_gpa_resources),
+	.resource = bfin_gpa_resources,
+	.dev = {
+		.platform_data = &bfin_gpa_pdata, /* Passed to driver */
+	},
+};
+
+static struct resource bfin_gpb_resources[] = {
+	{
+		.start = PORTB_FER,
+		.end = PORTB_MUX + 3,
+		.flags = IORESOURCE_MEM,
+	},
+	{
+		.start = IRQ_PB0,
+		.end = IRQ_PB0,
+		.flags = IORESOURCE_IRQ,
+	},
+};
+
+static struct adi_pinctrl_gpio_platform_data bfin_gpb_pdata = {
+	.port_pin_base	= GPIO_PB0,
+	.port_width	= GPIO_BANKSIZE,
+	.pint_id	= 0,
+	.pint_assign	= false,
+	.pint_map	= 1,
+};
+
+static struct platform_device bfin_gpb_device = {
+	.name = ADI_GPIO_DEVNAME,
+	.id = 1,
+	.num_resources = ARRAY_SIZE(bfin_gpb_resources),
+	.resource = bfin_gpb_resources,
+	.dev = {
+		.platform_data = &bfin_gpb_pdata, /* Passed to driver */
+	},
+};
+
+static struct resource bfin_gpc_resources[] = {
+	{
+		.start = PORTC_FER,
+		.end = PORTC_MUX + 3,
+		.flags = IORESOURCE_MEM,
+	},
+	{
+		.start = IRQ_PC0,
+		.end = IRQ_PC0,
+		.flags = IORESOURCE_IRQ,
+	},
+};
+
+static struct adi_pinctrl_gpio_platform_data bfin_gpc_pdata = {
+	.port_pin_base	= GPIO_PC0,
+	.port_width	= GPIO_BANKSIZE,
+	.pint_id	= 1,
+	.pint_assign	= false,
+	.pint_map	= 1,
+};
+
+static struct platform_device bfin_gpc_device = {
+	.name = ADI_GPIO_DEVNAME,
+	.id = 2,
+	.num_resources = ARRAY_SIZE(bfin_gpc_resources),
+	.resource = bfin_gpc_resources,
+	.dev = {
+		.platform_data = &bfin_gpc_pdata, /* Passed to driver */
+	},
+};
+
+static struct resource bfin_gpd_resources[] = {
+	{
+		.start = PORTD_FER,
+		.end = PORTD_MUX + 3,
+		.flags = IORESOURCE_MEM,
+	},
+	{
+		.start = IRQ_PD0,
+		.end = IRQ_PD0,
+		.flags = IORESOURCE_IRQ,
+	},
+};
+
+static struct adi_pinctrl_gpio_platform_data bfin_gpd_pdata = {
+	.port_pin_base	= GPIO_PD0,
+	.port_width	= GPIO_BANKSIZE,
+	.pint_id	= 2,
+	.pint_assign	= false,
+	.pint_map	= 1,
+};
+
+static struct platform_device bfin_gpd_device = {
+	.name = ADI_GPIO_DEVNAME,
+	.id = 3,
+	.num_resources = ARRAY_SIZE(bfin_gpd_resources),
+	.resource = bfin_gpd_resources,
+	.dev = {
+		.platform_data = &bfin_gpd_pdata, /* Passed to driver */
+	},
+};
+
+static struct resource bfin_gpe_resources[] = {
+	{
+		.start = PORTE_FER,
+		.end = PORTE_MUX + 3,
+		.flags = IORESOURCE_MEM,
+	},
+	{
+		.start = IRQ_PE0,
+		.end = IRQ_PE0,
+		.flags = IORESOURCE_IRQ,
+	},
+};
+
+static struct adi_pinctrl_gpio_platform_data bfin_gpe_pdata = {
+	.port_pin_base	= GPIO_PE0,
+	.port_width	= GPIO_BANKSIZE,
+	.pint_id	= 3,
+	.pint_assign	= false,
+	.pint_map	= 1,
+};
+
+static struct platform_device bfin_gpe_device = {
+	.name = ADI_GPIO_DEVNAME,
+	.id = 4,
+	.num_resources = ARRAY_SIZE(bfin_gpe_resources),
+	.resource = bfin_gpe_resources,
+	.dev = {
+		.platform_data = &bfin_gpe_pdata, /* Passed to driver */
+	},
+};
+
+static struct resource bfin_gpf_resources[] = {
+	{
+		.start = PORTF_FER,
+		.end = PORTF_MUX + 3,
+		.flags = IORESOURCE_MEM,
+	},
+	{
+		.start = IRQ_PF0,
+		.end = IRQ_PF0,
+		.flags = IORESOURCE_IRQ,
+	},
+};
+
+static struct adi_pinctrl_gpio_platform_data bfin_gpf_pdata = {
+	.port_pin_base	= GPIO_PF0,
+	.port_width	= GPIO_BANKSIZE,
+	.pint_id	= 4,
+	.pint_assign	= false,
+	.pint_map	= 1,
+};
+
+static struct platform_device bfin_gpf_device = {
+	.name = ADI_GPIO_DEVNAME,
+	.id = 5,
+	.num_resources = ARRAY_SIZE(bfin_gpf_resources),
+	.resource = bfin_gpf_resources,
+	.dev = {
+		.platform_data = &bfin_gpf_pdata, /* Passed to driver */
+	},
+};
+
+static struct resource bfin_gpg_resources[] = {
+	{
+		.start = PORTG_FER,
+		.end = PORTG_MUX + 3,
+		.flags = IORESOURCE_MEM,
+	},
+	{
+		.start = IRQ_PG0,
+		.end = IRQ_PG0,
+		.flags = IORESOURCE_IRQ,
+	},
+};
+
+static struct adi_pinctrl_gpio_platform_data bfin_gpg_pdata = {
+	.port_pin_base	= GPIO_PG0,
+	.port_width	= GPIO_BANKSIZE,
+	.pint_id	= 5,
+	.pint_assign	= false,
+	.pint_map	= 1,
+};
+
+static struct platform_device bfin_gpg_device = {
+	.name = ADI_GPIO_DEVNAME,
+	.id = 6,
+	.num_resources = ARRAY_SIZE(bfin_gpg_resources),
+	.resource = bfin_gpg_resources,
+	.dev = {
+		.platform_data = &bfin_gpg_pdata, /* Passed to driver */
+	},
+};
+
+#endif
+
 #if defined(CONFIG_KEYBOARD_GPIO) || defined(CONFIG_KEYBOARD_GPIO_MODULE)
 #include <linux/input.h>
 #include <linux/gpio_keys.h>
@@ -1349,7 +1707,7 @@ static struct spi_board_info bfin_spi_board_info[] __initdata = {
 		.modalias = "m25p80", /* Name of spi_driver for this device */
 		.max_speed_hz = 25000000,     /* max spi clock (SCK) speed in HZ */
 		.bus_num = 0, /* Framework bus number */
-		.chip_select = 1, /* SPI_SSEL1*/
+		.chip_select = MAX_CTRL_CS + GPIO_PD11, /* SPI_SSEL1*/
 		.platform_data = &bfin_spi_flash_data,
 		.controller_data = &spi_flash_chip_info,
 		.mode = SPI_MODE_3,
@@ -1362,7 +1720,7 @@ static struct spi_board_info bfin_spi_board_info[] __initdata = {
 		.irq			= IRQ_PD9,
 		.max_speed_hz		= 12500000,     /* max spi clock (SCK) speed in HZ */
 		.bus_num		= 0,
-		.chip_select  		= 4,
+		.chip_select		= MAX_CTRL_CS + GPIO_PC15, /* SPI_SSEL4 */
 	},
 #endif
 #if defined(CONFIG_SPI_SPIDEV) || defined(CONFIG_SPI_SPIDEV_MODULE)
@@ -1370,7 +1728,7 @@ static struct spi_board_info bfin_spi_board_info[] __initdata = {
 		.modalias = "spidev",
 		.max_speed_hz = 3125000,     /* max spi clock (SCK) speed in HZ */
 		.bus_num = 0,
-		.chip_select = 1,
+		.chip_select = MAX_CTRL_CS + GPIO_PD11, /* SPI_SSEL1*/
 		.controller_data = &spidev_chip_info,
 	},
 #endif
@@ -1565,6 +1923,22 @@ static struct platform_device bfin_dpmc = {
 static struct platform_device *ezkit_devices[] __initdata = {
 
 	&bfin_dpmc,
+#if defined(CONFIG_PINCTRL_ADI2)
+	&bfin_pinctrl_device,
+	&bfin_pint0_device,
+	&bfin_pint1_device,
+	&bfin_pint2_device,
+	&bfin_pint3_device,
+	&bfin_pint4_device,
+	&bfin_pint5_device,
+	&bfin_gpa_device,
+	&bfin_gpb_device,
+	&bfin_gpc_device,
+	&bfin_gpd_device,
+	&bfin_gpe_device,
+	&bfin_gpf_device,
+	&bfin_gpg_device,
+#endif
 
 #if defined(CONFIG_RTC_DRV_BFIN) || defined(CONFIG_RTC_DRV_BFIN_MODULE)
 	&rtc_device,
@@ -1681,20 +2055,52 @@ static struct platform_device *ezkit_devices[] __initdata = {
 
 };
 
+/* Pin control settings */
+static struct pinctrl_map __initdata bfin_pinmux_map[] = {
+	/* per-device maps */
+	PIN_MAP_MUX_GROUP_DEFAULT("bfin-uart.0",  "pinctrl-adi2.0", NULL, "uart0"),
+	PIN_MAP_MUX_GROUP_DEFAULT("bfin-uart.1",  "pinctrl-adi2.0", NULL, "uart1"),
+	PIN_MAP_MUX_GROUP_DEFAULT("bfin_sir.0",  "pinctrl-adi2.0", NULL, "uart0"),
+	PIN_MAP_MUX_GROUP_DEFAULT("bfin_sir.1",  "pinctrl-adi2.0", NULL, "uart1"),
+	PIN_MAP_MUX_GROUP_DEFAULT("bfin-sdh.0",  "pinctrl-adi2.0", NULL, "rsi0"),
+	PIN_MAP_MUX_GROUP_DEFAULT("stmmaceth.0",  "pinctrl-adi2.0", NULL, "eth0"),
+	PIN_MAP_MUX_GROUP_DEFAULT("bfin-spi3.0",  "pinctrl-adi2.0", NULL, "spi0"),
+	PIN_MAP_MUX_GROUP_DEFAULT("bfin-spi3.1",  "pinctrl-adi2.0", NULL, "spi1"),
+	PIN_MAP_MUX_GROUP_DEFAULT("i2c-bfin-twi.0",  "pinctrl-adi2.0", NULL, "twi0"),
+	PIN_MAP_MUX_GROUP_DEFAULT("i2c-bfin-twi.1",  "pinctrl-adi2.0", NULL, "twi1"),
+	PIN_MAP_MUX_GROUP_DEFAULT("bfin-rotary",  "pinctrl-adi2.0", NULL, "rotary"),
+	PIN_MAP_MUX_GROUP_DEFAULT("bfin_can.0",  "pinctrl-adi2.0", NULL, "can0"),
+	PIN_MAP_MUX_GROUP_DEFAULT("physmap-flash.0",  "pinctrl-adi2.0", NULL, "smc0"),
+	PIN_MAP_MUX_GROUP_DEFAULT("bf609_nl8048.2",  "pinctrl-adi2.0", NULL, "ppi2_16b"),
+	PIN_MAP_MUX_GROUP_DEFAULT("bfin_display.0",  "pinctrl-adi2.0", NULL, "ppi0_16b"),
+#if defined(CONFIG_VIDEO_MT9M114) || defined(CONFIG_VIDEO_MT9M114_MODULE)
+	PIN_MAP_MUX_GROUP_DEFAULT("bfin_capture.0",  "pinctrl-adi2.0", NULL, "ppi0_8b"),
+#elif defined(CONFIG_VIDEO_VS6624) || defined(CONFIG_VIDEO_VS6624_MODULE)
+	PIN_MAP_MUX_GROUP_DEFAULT("bfin_capture.0",  "pinctrl-adi2.0", NULL, "ppi0_16b"),
+#else
+	PIN_MAP_MUX_GROUP_DEFAULT("bfin_capture.0",  "pinctrl-adi2.0", NULL, "ppi0_24b"),
+#endif
+	PIN_MAP_MUX_GROUP_DEFAULT("bfin-i2s.0",  "pinctrl-adi2.0", NULL, "sport0"),
+	PIN_MAP_MUX_GROUP_DEFAULT("bfin-tdm.0",  "pinctrl-adi2.0", NULL, "sport0"),
+	PIN_MAP_MUX_GROUP_DEFAULT("bfin-i2s.1",  "pinctrl-adi2.0", NULL, "sport1"),
+	PIN_MAP_MUX_GROUP_DEFAULT("bfin-tdm.1",  "pinctrl-adi2.0", NULL, "sport1"),
+	PIN_MAP_MUX_GROUP_DEFAULT("bfin-i2s.2",  "pinctrl-adi2.0", NULL, "sport2"),
+	PIN_MAP_MUX_GROUP_DEFAULT("bfin-tdm.2",  "pinctrl-adi2.0", NULL, "sport2"),
+};
+
 static int __init ezkit_init(void)
 {
 	printk(KERN_INFO "%s(): registering device resources\n", __func__);
 
+	/* Initialize pinmuxing */
+	pinctrl_register_mappings(bfin_pinmux_map,
+				ARRAY_SIZE(bfin_pinmux_map));
+
 	i2c_register_board_info(0, bfin_i2c_board_info0,
 				ARRAY_SIZE(bfin_i2c_board_info0));
 	i2c_register_board_info(1, bfin_i2c_board_info1,
 				ARRAY_SIZE(bfin_i2c_board_info1));
 
-#if defined(CONFIG_STMMAC_ETH) || defined(CONFIG_STMMAC_ETH_MODULE)
-	if (!peripheral_request_list(pins, "emac0"))
-		printk(KERN_ERR "%s(): request emac pins failed\n", __func__);
-#endif
-
 	platform_add_devices(ezkit_devices, ARRAY_SIZE(ezkit_devices));
 
 	spi_register_board_info(bfin_spi_board_info, ARRAY_SIZE(bfin_spi_board_info));
@@ -1713,18 +2119,6 @@ static struct platform_device *ezkit_early_devices[] __initdata = {
 	&bfin_uart1_device,
 #endif
 #endif
-
-#if defined(CONFIG_SERIAL_BFIN_SPORT_CONSOLE)
-#ifdef CONFIG_SERIAL_BFIN_SPORT0_UART
-	&bfin_sport0_uart_device,
-#endif
-#ifdef CONFIG_SERIAL_BFIN_SPORT1_UART
-	&bfin_sport1_uart_device,
-#endif
-#ifdef CONFIG_SERIAL_BFIN_SPORT2_UART
-	&bfin_sport2_uart_device,
-#endif
-#endif
 };
 
 void __init native_machine_early_platform_add_devices(void)
diff --git a/arch/blackfin/mach-bf609/include/mach/gpio.h b/arch/blackfin/mach-bf609/include/mach/gpio.h
index c32c8cc8db2e..07182513e794 100644
--- a/arch/blackfin/mach-bf609/include/mach/gpio.h
+++ b/arch/blackfin/mach-bf609/include/mach/gpio.h
@@ -152,14 +152,6 @@ struct gpio_port_t {
 	unsigned long revid;
 };
 
-struct gpio_port_s {
-	unsigned short fer;
-	unsigned short data;
-	unsigned short dir;
-	unsigned short inen;
-	unsigned int mux;
-};
-
 #endif
 
 #include <mach-common/ports-a.h>
diff --git a/arch/blackfin/mach-bf609/include/mach/irq.h b/arch/blackfin/mach-bf609/include/mach/irq.h
index fa0843d5d77a..d1cb6a86f80a 100644
--- a/arch/blackfin/mach-bf609/include/mach/irq.h
+++ b/arch/blackfin/mach-bf609/include/mach/irq.h
@@ -298,7 +298,7 @@
 extern u8 sec_int_priority[];
 
 /*
- * bfin pint registers layout
+ * gpio pint registers layout
  */
 struct bfin_pint_regs {
 	u32 mask_set;
diff --git a/arch/blackfin/mach-bf609/include/mach/portmux.h b/arch/blackfin/mach-bf609/include/mach/portmux.h
index fe34191eef0b..c48bb71a55ce 100644
--- a/arch/blackfin/mach-bf609/include/mach/portmux.h
+++ b/arch/blackfin/mach-bf609/include/mach/portmux.h
@@ -19,6 +19,7 @@
 #define P_MII0_CRS	(P_DEFINED | P_IDENT(GPIO_PC5) | P_FUNCT(0))
 #define P_MII0_ERxER	(P_DEFINED | P_IDENT(GPIO_PC4) | P_FUNCT(0))
 #define P_MII0_TxCLK	(P_DEFINED | P_IDENT(GPIO_PB14) | P_FUNCT(0))
+#define P_MII0_PTPPPS	(P_DEFINED | P_IDENT(GPIO_PB15) | P_FUNCT(0))
 
 #define P_RMII0 {\
 	P_MII0_ETxD0, \
@@ -30,6 +31,7 @@
 	P_MII0_TxCLK, \
 	P_MII0_PHYINT, \
 	P_MII0_CRS, \
+	P_MII0_PTPPPS, \
 	P_MII0_MDC, \
 	P_MII0_MDIO, 0}
 
@@ -44,6 +46,7 @@
 #define P_MII1_CRS	(P_DEFINED | P_IDENT(GPIO_PE13) | P_FUNCT(0))
 #define P_MII1_ERxER	(P_DEFINED | P_IDENT(GPIO_PE14) | P_FUNCT(0))
 #define P_MII1_TxCLK	(P_DEFINED | P_IDENT(GPIO_PG6) | P_FUNCT(0))
+#define P_MII1_PTPPPS	(P_DEFINED | P_IDENT(GPIO_PC9) | P_FUNCT(0))
 
 #define P_RMII1 {\
 	P_MII1_ETxD0, \
@@ -55,6 +58,7 @@
 	P_MII1_TxCLK, \
 	P_MII1_PHYINT, \
 	P_MII1_CRS, \
+	P_MII1_PTPPPS, \
 	P_MII1_MDC, \
 	P_MII1_MDIO, 0}
 
diff --git a/arch/blackfin/mach-common/ints-priority.c b/arch/blackfin/mach-common/ints-priority.c
index d143fd8d2bc5..ca75613231c8 100644
--- a/arch/blackfin/mach-common/ints-priority.c
+++ b/arch/blackfin/mach-common/ints-priority.c
@@ -704,10 +704,9 @@ static inline void bfin_set_irq_handler(unsigned irq, irq_flow_handler_t handle)
 	__irq_set_handler_locked(irq, handle);
 }
 
-static DECLARE_BITMAP(gpio_enabled, MAX_BLACKFIN_GPIOS);
-extern void bfin_gpio_irq_prepare(unsigned gpio);
+#ifdef CONFIG_GPIO_ADI
 
-#if !BFIN_GPIO_PINT
+static DECLARE_BITMAP(gpio_enabled, MAX_BLACKFIN_GPIOS);
 
 static void bfin_gpio_ack_irq(struct irq_data *d)
 {
@@ -821,15 +820,6 @@ static int bfin_gpio_irq_type(struct irq_data *d, unsigned int type)
 	return 0;
 }
 
-#ifdef CONFIG_PM
-static int bfin_gpio_set_wake(struct irq_data *d, unsigned int state)
-{
-	return gpio_pm_wakeup_ctrl(irq_to_gpio(d->irq), state);
-}
-#else
-# define bfin_gpio_set_wake NULL
-#endif
-
 static void bfin_demux_gpio_block(unsigned int irq)
 {
 	unsigned int gpio, mask;
@@ -896,279 +886,40 @@ void bfin_demux_gpio_irq(unsigned int inta_irq,
 	bfin_demux_gpio_block(irq);
 }
 
-#else
-
-#define NR_PINT_BITS		32
-#define IRQ_NOT_AVAIL		0xFF
-
-#define PINT_2_BANK(x)		((x) >> 5)
-#define PINT_2_BIT(x)		((x) & 0x1F)
-#define PINT_BIT(x)		(1 << (PINT_2_BIT(x)))
-
-static unsigned char irq2pint_lut[NR_PINTS];
-static unsigned char pint2irq_lut[NR_PINT_SYS_IRQS * NR_PINT_BITS];
-
-static struct bfin_pint_regs * const pint[NR_PINT_SYS_IRQS] = {
-	(struct bfin_pint_regs *)PINT0_MASK_SET,
-	(struct bfin_pint_regs *)PINT1_MASK_SET,
-	(struct bfin_pint_regs *)PINT2_MASK_SET,
-	(struct bfin_pint_regs *)PINT3_MASK_SET,
-#ifdef CONFIG_BF60x
-	(struct bfin_pint_regs *)PINT4_MASK_SET,
-	(struct bfin_pint_regs *)PINT5_MASK_SET,
-#endif
-};
-
-inline unsigned int get_irq_base(u32 bank, u8 bmap)
-{
-	unsigned int irq_base;
-
-#ifndef CONFIG_BF60x
-	if (bank < 2) {		/*PA-PB */
-		irq_base = IRQ_PA0 + bmap * 16;
-	} else {		/*PC-PJ */
-		irq_base = IRQ_PC0 + bmap * 16;
-	}
-#else
-	irq_base = IRQ_PA0 + bank * 16 + bmap * 16;
-#endif
-	return irq_base;
-}
-
-	/* Whenever PINTx_ASSIGN is altered init_pint_lut() must be executed! */
-void init_pint_lut(void)
-{
-	u16 bank, bit, irq_base, bit_pos;
-	u32 pint_assign;
-	u8 bmap;
-
-	memset(irq2pint_lut, IRQ_NOT_AVAIL, sizeof(irq2pint_lut));
-
-	for (bank = 0; bank < NR_PINT_SYS_IRQS; bank++) {
-
-		pint_assign = pint[bank]->assign;
-
-		for (bit = 0; bit < NR_PINT_BITS; bit++) {
-
-			bmap = (pint_assign >> ((bit / 8) * 8)) & 0xFF;
-
-			irq_base = get_irq_base(bank, bmap);
-
-			irq_base += (bit % 8) + ((bit / 8) & 1 ? 8 : 0);
-			bit_pos = bit + bank * NR_PINT_BITS;
-
-			pint2irq_lut[bit_pos] = irq_base - SYS_IRQS;
-			irq2pint_lut[irq_base - SYS_IRQS] = bit_pos;
-		}
-	}
-}
-
-static void bfin_gpio_ack_irq(struct irq_data *d)
-{
-	u32 pint_val = irq2pint_lut[d->irq - SYS_IRQS];
-	u32 pintbit = PINT_BIT(pint_val);
-	u32 bank = PINT_2_BANK(pint_val);
-
-	if (irqd_get_trigger_type(d) == IRQ_TYPE_EDGE_BOTH) {
-		if (pint[bank]->invert_set & pintbit)
-			pint[bank]->invert_clear = pintbit;
-		else
-			pint[bank]->invert_set = pintbit;
-	}
-	pint[bank]->request = pintbit;
-
-}
-
-static void bfin_gpio_mask_ack_irq(struct irq_data *d)
-{
-	u32 pint_val = irq2pint_lut[d->irq - SYS_IRQS];
-	u32 pintbit = PINT_BIT(pint_val);
-	u32 bank = PINT_2_BANK(pint_val);
-
-	if (irqd_get_trigger_type(d) == IRQ_TYPE_EDGE_BOTH) {
-		if (pint[bank]->invert_set & pintbit)
-			pint[bank]->invert_clear = pintbit;
-		else
-			pint[bank]->invert_set = pintbit;
-	}
-
-	pint[bank]->request = pintbit;
-	pint[bank]->mask_clear = pintbit;
-}
-
-static void bfin_gpio_mask_irq(struct irq_data *d)
-{
-	u32 pint_val = irq2pint_lut[d->irq - SYS_IRQS];
-
-	pint[PINT_2_BANK(pint_val)]->mask_clear = PINT_BIT(pint_val);
-}
-
-static void bfin_gpio_unmask_irq(struct irq_data *d)
-{
-	u32 pint_val = irq2pint_lut[d->irq - SYS_IRQS];
-	u32 pintbit = PINT_BIT(pint_val);
-	u32 bank = PINT_2_BANK(pint_val);
-
-	pint[bank]->mask_set = pintbit;
-}
-
-static unsigned int bfin_gpio_irq_startup(struct irq_data *d)
-{
-	unsigned int irq = d->irq;
-	u32 gpionr = irq_to_gpio(irq);
-	u32 pint_val = irq2pint_lut[irq - SYS_IRQS];
-
-	if (pint_val == IRQ_NOT_AVAIL) {
-		printk(KERN_ERR
-		"GPIO IRQ %d :Not in PINT Assign table "
-		"Reconfigure Interrupt to Port Assignemt\n", irq);
-		return -ENODEV;
-	}
-
-	if (__test_and_set_bit(gpionr, gpio_enabled))
-		bfin_gpio_irq_prepare(gpionr);
-
-	bfin_gpio_unmask_irq(d);
-
-	return 0;
-}
-
-static void bfin_gpio_irq_shutdown(struct irq_data *d)
-{
-	u32 gpionr = irq_to_gpio(d->irq);
-
-	bfin_gpio_mask_irq(d);
-	__clear_bit(gpionr, gpio_enabled);
-	bfin_gpio_irq_free(gpionr);
-}
-
-static int bfin_gpio_irq_type(struct irq_data *d, unsigned int type)
-{
-	unsigned int irq = d->irq;
-	int ret;
-	char buf[16];
-	u32 gpionr = irq_to_gpio(irq);
-	u32 pint_val = irq2pint_lut[irq - SYS_IRQS];
-	u32 pintbit = PINT_BIT(pint_val);
-	u32 bank = PINT_2_BANK(pint_val);
-
-	if (pint_val == IRQ_NOT_AVAIL)
-		return -ENODEV;
-
-	if (type == IRQ_TYPE_PROBE) {
-		/* only probe unenabled GPIO interrupt lines */
-		if (test_bit(gpionr, gpio_enabled))
-			return 0;
-		type = IRQ_TYPE_EDGE_RISING | IRQ_TYPE_EDGE_FALLING;
-	}
-
-	if (type & (IRQ_TYPE_EDGE_RISING | IRQ_TYPE_EDGE_FALLING |
-		    IRQ_TYPE_LEVEL_HIGH | IRQ_TYPE_LEVEL_LOW)) {
-
-		snprintf(buf, 16, "gpio-irq%d", irq);
-		ret = bfin_gpio_irq_request(gpionr, buf);
-		if (ret)
-			return ret;
-
-		if (__test_and_set_bit(gpionr, gpio_enabled))
-			bfin_gpio_irq_prepare(gpionr);
-
-	} else {
-		__clear_bit(gpionr, gpio_enabled);
-		return 0;
-	}
-
-	if ((type & (IRQ_TYPE_EDGE_FALLING | IRQ_TYPE_LEVEL_LOW)))
-		pint[bank]->invert_set = pintbit;	/* low or falling edge denoted by one */
-	else
-		pint[bank]->invert_clear = pintbit;	/* high or rising edge denoted by zero */
-
-	if ((type & (IRQ_TYPE_EDGE_RISING | IRQ_TYPE_EDGE_FALLING))
-	    == (IRQ_TYPE_EDGE_RISING | IRQ_TYPE_EDGE_FALLING)) {
-		if (gpio_get_value(gpionr))
-			pint[bank]->invert_set = pintbit;
-		else
-			pint[bank]->invert_clear = pintbit;
-	}
-
-	if (type & (IRQ_TYPE_EDGE_RISING | IRQ_TYPE_EDGE_FALLING)) {
-		pint[bank]->edge_set = pintbit;
-		bfin_set_irq_handler(irq, handle_edge_irq);
-	} else {
-		pint[bank]->edge_clear = pintbit;
-		bfin_set_irq_handler(irq, handle_level_irq);
-	}
-
-	return 0;
-}
-
 #ifdef CONFIG_PM
-static struct bfin_pm_pint_save save_pint_reg[NR_PINT_SYS_IRQS];
-static u32 save_pint_sec_ctl[NR_PINT_SYS_IRQS];
 
 static int bfin_gpio_set_wake(struct irq_data *d, unsigned int state)
 {
-	u32 pint_irq;
-	u32 pint_val = irq2pint_lut[d->irq - SYS_IRQS];
-	u32 bank = PINT_2_BANK(pint_val);
-
-	switch (bank) {
-	case 0:
-		pint_irq = IRQ_PINT0;
-		break;
-	case 2:
-		pint_irq = IRQ_PINT2;
-		break;
-	case 3:
-		pint_irq = IRQ_PINT3;
-		break;
-	case 1:
-		pint_irq = IRQ_PINT1;
-		break;
-#ifdef CONFIG_BF60x
-	case 4:
-		pint_irq = IRQ_PINT4;
-		break;
-	case 5:
-		pint_irq = IRQ_PINT5;
-		break;
-#endif
-	default:
-		return -EINVAL;
-	}
+	return bfin_gpio_pm_wakeup_ctrl(irq_to_gpio(d->irq), state);
+}
 
-#ifndef SEC_GCTL
-	bfin_internal_set_wake(pint_irq, state);
-#endif
+#else
 
-	return 0;
-}
+# define bfin_gpio_set_wake NULL
 
-void bfin_pint_suspend(void)
-{
-	u32 bank;
+#endif
 
-	for (bank = 0; bank < NR_PINT_SYS_IRQS; bank++) {
-		save_pint_reg[bank].mask_set = pint[bank]->mask_set;
-		save_pint_reg[bank].assign = pint[bank]->assign;
-		save_pint_reg[bank].edge_set = pint[bank]->edge_set;
-		save_pint_reg[bank].invert_set = pint[bank]->invert_set;
-	}
-}
+static struct irq_chip bfin_gpio_irqchip = {
+	.name = "GPIO",
+	.irq_ack = bfin_gpio_ack_irq,
+	.irq_mask = bfin_gpio_mask_irq,
+	.irq_mask_ack = bfin_gpio_mask_ack_irq,
+	.irq_unmask = bfin_gpio_unmask_irq,
+	.irq_disable = bfin_gpio_mask_irq,
+	.irq_enable = bfin_gpio_unmask_irq,
+	.irq_set_type = bfin_gpio_irq_type,
+	.irq_startup = bfin_gpio_irq_startup,
+	.irq_shutdown = bfin_gpio_irq_shutdown,
+	.irq_set_wake = bfin_gpio_set_wake,
+};
 
-void bfin_pint_resume(void)
-{
-	u32 bank;
+#endif
 
-	for (bank = 0; bank < NR_PINT_SYS_IRQS; bank++) {
-		pint[bank]->mask_set = save_pint_reg[bank].mask_set;
-		pint[bank]->assign = save_pint_reg[bank].assign;
-		pint[bank]->edge_set = save_pint_reg[bank].edge_set;
-		pint[bank]->invert_set = save_pint_reg[bank].invert_set;
-	}
-}
+#ifdef CONFIG_PM
 
 #ifdef SEC_GCTL
+static u32 save_pint_sec_ctl[NR_PINT_SYS_IRQS];
+
 static int sec_suspend(void)
 {
 	u32 bank;
@@ -1195,92 +946,10 @@ static struct syscore_ops sec_pm_syscore_ops = {
 	.suspend = sec_suspend,
 	.resume = sec_resume,
 };
-
-#endif
-#else
-# define bfin_gpio_set_wake NULL
-#endif
-
-void bfin_demux_gpio_irq(unsigned int inta_irq,
-			struct irq_desc *desc)
-{
-	u32 bank, pint_val;
-	u32 request, irq;
-	u32 level_mask;
-	int umask = 0;
-	struct irq_chip *chip = irq_desc_get_chip(desc);
-
-	if (chip->irq_mask_ack) {
-		chip->irq_mask_ack(&desc->irq_data);
-	} else {
-		chip->irq_mask(&desc->irq_data);
-		if (chip->irq_ack)
-			chip->irq_ack(&desc->irq_data);
-	}
-
-	switch (inta_irq) {
-	case IRQ_PINT0:
-		bank = 0;
-		break;
-	case IRQ_PINT2:
-		bank = 2;
-		break;
-	case IRQ_PINT3:
-		bank = 3;
-		break;
-	case IRQ_PINT1:
-		bank = 1;
-		break;
-#ifdef CONFIG_BF60x
-	case IRQ_PINT4:
-		bank = 4;
-		break;
-	case IRQ_PINT5:
-		bank = 5;
-		break;
 #endif
-	default:
-		return;
-	}
-
-	pint_val = bank * NR_PINT_BITS;
-
-	request = pint[bank]->request;
-
-	level_mask = pint[bank]->edge_set & request;
-
-	while (request) {
-		if (request & 1) {
-			irq = pint2irq_lut[pint_val] + SYS_IRQS;
-			if (level_mask & PINT_BIT(pint_val)) {
-				umask = 1;
-				chip->irq_unmask(&desc->irq_data);
-			}
-			bfin_handle_irq(irq);
-		}
-		pint_val++;
-		request >>= 1;
-	}
 
-	if (!umask)
-		chip->irq_unmask(&desc->irq_data);
-}
 #endif
 
-static struct irq_chip bfin_gpio_irqchip = {
-	.name = "GPIO",
-	.irq_ack = bfin_gpio_ack_irq,
-	.irq_mask = bfin_gpio_mask_irq,
-	.irq_mask_ack = bfin_gpio_mask_ack_irq,
-	.irq_unmask = bfin_gpio_unmask_irq,
-	.irq_disable = bfin_gpio_mask_irq,
-	.irq_enable = bfin_gpio_unmask_irq,
-	.irq_set_type = bfin_gpio_irq_type,
-	.irq_startup = bfin_gpio_irq_startup,
-	.irq_shutdown = bfin_gpio_irq_shutdown,
-	.irq_set_wake = bfin_gpio_set_wake,
-};
-
 void init_exception_vectors(void)
 {
 	/* cannot program in software:
@@ -1331,17 +1000,6 @@ int __init init_arch_irq(void)
 
 	local_irq_disable();
 
-#if BFIN_GPIO_PINT
-# ifdef CONFIG_PINTx_REASSIGN
-	pint[0]->assign = CONFIG_PINT0_ASSIGN;
-	pint[1]->assign = CONFIG_PINT1_ASSIGN;
-	pint[2]->assign = CONFIG_PINT2_ASSIGN;
-	pint[3]->assign = CONFIG_PINT3_ASSIGN;
-# endif
-	/* Whenever PINTx_ASSIGN is altered init_pint_lut() must be executed! */
-	init_pint_lut();
-#endif
-
 	for (irq = 0; irq <= SYS_IRQS; irq++) {
 		if (irq <= IRQ_CORETMR)
 			irq_set_chip(irq, &bfin_core_irqchip);
@@ -1349,12 +1007,8 @@ int __init init_arch_irq(void)
 			irq_set_chip(irq, &bfin_internal_irqchip);
 
 		switch (irq) {
-#if BFIN_GPIO_PINT
-		case IRQ_PINT0:
-		case IRQ_PINT1:
-		case IRQ_PINT2:
-		case IRQ_PINT3:
-#elif defined(BF537_FAMILY)
+#if !BFIN_GPIO_PINT
+#if defined(BF537_FAMILY)
 		case IRQ_PH_INTA_MAC_RX:
 		case IRQ_PF_INTA_PG_INTA:
 #elif defined(BF533_FAMILY)
@@ -1372,6 +1026,7 @@ int __init init_arch_irq(void)
 #endif
 			irq_set_chained_handler(irq, bfin_demux_gpio_irq);
 			break;
+#endif
 #if defined(CONFIG_BFIN_MAC) || defined(CONFIG_BFIN_MAC_MODULE)
 		case IRQ_MAC_ERROR:
 			irq_set_chained_handler(irq,
@@ -1419,10 +1074,12 @@ int __init init_arch_irq(void)
 					 handle_level_irq);
 #endif
 	/* if configured as edge, then will be changed to do_edge_IRQ */
+#ifdef CONFIG_GPIO_ADI
 	for (irq = GPIO_IRQ_BASE;
 		irq < (GPIO_IRQ_BASE + MAX_BLACKFIN_GPIOS); irq++)
 		irq_set_chip_and_handler(irq, &bfin_gpio_irqchip,
 					 handle_level_irq);
+#endif
 	bfin_write_IMASK(0);
 	CSYNC();
 	ilat = bfin_read_ILAT();
@@ -1525,19 +1182,6 @@ int __init init_arch_irq(void)
 
 	local_irq_disable();
 
-#if BFIN_GPIO_PINT
-# ifdef CONFIG_PINTx_REASSIGN
-	pint[0]->assign = CONFIG_PINT0_ASSIGN;
-	pint[1]->assign = CONFIG_PINT1_ASSIGN;
-	pint[2]->assign = CONFIG_PINT2_ASSIGN;
-	pint[3]->assign = CONFIG_PINT3_ASSIGN;
-	pint[4]->assign = CONFIG_PINT4_ASSIGN;
-	pint[5]->assign = CONFIG_PINT5_ASSIGN;
-# endif
-	/* Whenever PINTx_ASSIGN is altered init_pint_lut() must be executed! */
-	init_pint_lut();
-#endif
-
 	for (irq = 0; irq <= SYS_IRQS; irq++) {
 		if (irq <= IRQ_CORETMR) {
 			irq_set_chip_and_handler(irq, &bfin_core_irqchip,
@@ -1546,9 +1190,6 @@ int __init init_arch_irq(void)
 			if (irq == IRQ_CORETMR)
 				irq_set_handler(irq, handle_percpu_irq);
 #endif
-		} else if (irq >= BFIN_IRQ(21) && irq <= BFIN_IRQ(26)) {
-			irq_set_chip(irq, &bfin_sec_irqchip);
-			irq_set_chained_handler(irq, bfin_demux_gpio_irq);
 		} else if (irq >= BFIN_IRQ(34) && irq <= BFIN_IRQ(37)) {
 			irq_set_chip_and_handler(irq, &bfin_sec_irqchip,
 				handle_percpu_irq);
@@ -1563,10 +1204,6 @@ int __init init_arch_irq(void)
 			__irq_set_preflow_handler(irq, bfin_sec_preflow_handler);
 		}
 	}
-	for (irq = GPIO_IRQ_BASE;
-		irq < (GPIO_IRQ_BASE + MAX_BLACKFIN_GPIOS); irq++)
-		irq_set_chip_and_handler(irq, &bfin_gpio_irqchip,
-					handle_level_irq);
 
 	bfin_write_IMASK(0);
 	CSYNC();
diff --git a/arch/blackfin/mach-common/pm.c b/arch/blackfin/mach-common/pm.c
index 87bfe549ad3f..1387a94bcfd5 100644
--- a/arch/blackfin/mach-common/pm.c
+++ b/arch/blackfin/mach-common/pm.c
@@ -27,7 +27,7 @@ struct bfin_cpu_pm_fns *bfin_cpu_pm;
 
 void bfin_pm_suspend_standby_enter(void)
 {
-#ifndef CONFIG_BF60x
+#if !BFIN_GPIO_PINT
 	bfin_pm_standby_setup();
 #endif
 
@@ -41,7 +41,7 @@ void bfin_pm_suspend_standby_enter(void)
 # endif
 #endif
 
-#ifndef CONFIG_BF60x
+#if !BFIN_GPIO_PINT
 	bfin_pm_standby_restore();
 #endif
 
@@ -128,6 +128,7 @@ static void flushinv_all_dcache(void)
 					if ((status & 0x3) != 0x3)
 						continue;
 
+
 					/* construct the address using the tag */
 					addr = (status & 0xFFFFC800) | (subbank << 12) | (set << 5);
 
@@ -140,11 +141,14 @@ static void flushinv_all_dcache(void)
 
 int bfin_pm_suspend_mem_enter(void)
 {
-	int wakeup, ret;
+	int ret;
+#ifndef CONFIG_BF60x
+	int wakeup;
+#endif
 
 	unsigned char *memptr = kmalloc(L1_CODE_LENGTH + L1_DATA_A_LENGTH
 					 + L1_DATA_B_LENGTH + L1_SCRATCH_LENGTH,
-					  GFP_KERNEL);
+					  GFP_ATOMIC);
 
 	if (memptr == NULL) {
 		panic("bf53x_suspend_l1_mem malloc failed");
@@ -170,10 +174,8 @@ int bfin_pm_suspend_mem_enter(void)
 		return ret;
 	}
 
+#ifdef CONFIG_GPIO_ADI
 	bfin_gpio_pm_hibernate_suspend();
-
-#if BFIN_GPIO_PINT
-	bfin_pint_suspend();
 #endif
 
 #if defined(CONFIG_BFIN_EXTMEM_WRITEBACK) || defined(CONFIG_BFIN_L2_WRITEBACK)
@@ -194,11 +196,9 @@ int bfin_pm_suspend_mem_enter(void)
 	_enable_icplb();
 	_enable_dcplb();
 
-#if BFIN_GPIO_PINT
-	bfin_pint_resume();
-#endif
-
+#ifdef CONFIG_GPIO_ADI
 	bfin_gpio_pm_hibernate_restore();
+#endif
 	blackfin_dma_resume();
 
 	kfree(memptr);
diff --git a/arch/blackfin/mach-common/smp.c b/arch/blackfin/mach-common/smp.c
index 82f301c117a5..2bbae0783819 100644
--- a/arch/blackfin/mach-common/smp.c
+++ b/arch/blackfin/mach-common/smp.c
@@ -146,6 +146,7 @@ static irqreturn_t ipi_handler_int1(int irq, void *dev_instance)
 
 	platform_clear_ipi(cpu, IRQ_SUPPLE_1);
 
+	smp_rmb();
 	bfin_ipi_data = &__get_cpu_var(bfin_ipi);
 	while ((pending = atomic_xchg(&bfin_ipi_data->bits, 0)) != 0) {
 		msg = 0;
@@ -161,18 +162,20 @@ static irqreturn_t ipi_handler_int1(int irq, void *dev_instance)
 			case BFIN_IPI_CALL_FUNC:
 				generic_smp_call_function_interrupt();
 				break;
-
 			case BFIN_IPI_CALL_FUNC_SINGLE:
 				generic_smp_call_function_single_interrupt();
 				break;
-
 			case BFIN_IPI_CPU_STOP:
 				ipi_cpu_stop(cpu);
 				break;
+			default:
+				goto out;
 			}
 			atomic_dec(&bfin_ipi_data->count);
 		} while (msg < BITS_PER_LONG);
+
 	}
+out:
 	return IRQ_HANDLED;
 }
 
@@ -198,10 +201,11 @@ void send_ipi(const struct cpumask *cpumask, enum ipi_message_type msg)
 		bfin_ipi_data = &per_cpu(bfin_ipi, cpu);
 		atomic_set_mask((1 << msg), &bfin_ipi_data->bits);
 		atomic_inc(&bfin_ipi_data->count);
-		platform_send_ipi_cpu(cpu, IRQ_SUPPLE_1);
 	}
-
 	local_irq_restore(flags);
+	smp_wmb();
+	for_each_cpu(cpu, cpumask)
+		platform_send_ipi_cpu(cpu, IRQ_SUPPLE_1);
 }
 
 void arch_send_call_function_single_ipi(int cpu)
diff --git a/arch/c6x/Kconfig b/arch/c6x/Kconfig
index 957dd00ea561..77ea09b8bce1 100644
--- a/arch/c6x/Kconfig
+++ b/arch/c6x/Kconfig
@@ -36,9 +36,6 @@ config GENERIC_HWEIGHT
 config GENERIC_BUG
 	def_bool y
 
-config COMMON_CLKDEV
-	def_bool y
-
 config C6X_BIG_KERNEL
 	bool "Build a big kernel"
 	help
@@ -105,10 +102,6 @@ menu "Processor type and features"
 
 source "arch/c6x/platforms/Kconfig"
 
-config TMS320C6X_CACHES_ON
-	bool "L2 cache support"
-	default y
-
 config KERNEL_RAM_BASE_ADDRESS
 	hex "Virtual address of memory base"
 	default 0xe0000000 if SOC_TMS320C6455
diff --git a/arch/c6x/include/asm/thread_info.h b/arch/c6x/include/asm/thread_info.h
index 4c8dc562bd90..d4e9ef87076d 100644
--- a/arch/c6x/include/asm/thread_info.h
+++ b/arch/c6x/include/asm/thread_info.h
@@ -84,8 +84,6 @@ struct thread_info *current_thread_info(void)
 #define put_thread_info(ti)	put_task_struct((ti)->task)
 #endif /* __ASSEMBLY__ */
 
-#define	PREEMPT_ACTIVE	0x10000000
-
 /*
  * thread information flag bit numbers
  * - pending work-to-be-done flags are in LSW
diff --git a/arch/cris/include/asm/hardirq.h b/arch/cris/include/asm/hardirq.h
index 17bb12d760b2..04126f7bfab2 100644
--- a/arch/cris/include/asm/hardirq.h
+++ b/arch/cris/include/asm/hardirq.h
@@ -2,18 +2,6 @@
 #define __ASM_HARDIRQ_H
 
 #include <asm/irq.h>
-
-#define HARDIRQ_BITS	8
-
-/*
- * The hardirq mask has to be large enough to have
- * space for potentially all IRQ sources in the system
- * nesting on a single CPU:
- */
-#if (1 << HARDIRQ_BITS) < NR_IRQS
-# error HARDIRQ_BITS is too low!
-#endif
-
 #include <asm-generic/hardirq.h>
 
 #endif /* __ASM_HARDIRQ_H */
diff --git a/arch/cris/include/asm/pgalloc.h b/arch/cris/include/asm/pgalloc.h
index 6da975db112f..235ece437ddd 100644
--- a/arch/cris/include/asm/pgalloc.h
+++ b/arch/cris/include/asm/pgalloc.h
@@ -32,7 +32,12 @@ static inline pgtable_t pte_alloc_one(struct mm_struct *mm, unsigned long addres
 {
 	struct page *pte;
 	pte = alloc_pages(GFP_KERNEL|__GFP_REPEAT|__GFP_ZERO, 0);
-	pgtable_page_ctor(pte);
+	if (!pte)
+		return NULL;
+	if (!pgtable_page_ctor(pte)) {
+		__free_page(pte);
+		return NULL;
+	}
 	return pte;
 }
 
diff --git a/arch/cris/include/asm/thread_info.h b/arch/cris/include/asm/thread_info.h
index 07c8c40c52b3..55dede18c032 100644
--- a/arch/cris/include/asm/thread_info.h
+++ b/arch/cris/include/asm/thread_info.h
@@ -44,8 +44,6 @@ struct thread_info {
 
 #endif
 
-#define PREEMPT_ACTIVE		0x10000000
-
 /*
  * macros/functions for gaining access to the thread information structure
  */
diff --git a/arch/frv/include/asm/thread_info.h b/arch/frv/include/asm/thread_info.h
index bebd7eadc772..af29e17c0181 100644
--- a/arch/frv/include/asm/thread_info.h
+++ b/arch/frv/include/asm/thread_info.h
@@ -52,8 +52,6 @@ struct thread_info {
 
 #endif
 
-#define PREEMPT_ACTIVE		0x10000000
-
 /*
  * macros/functions for gaining access to the thread information structure
  */
diff --git a/arch/frv/mm/pgalloc.c b/arch/frv/mm/pgalloc.c
index f6084bc524e8..41907d25ed38 100644
--- a/arch/frv/mm/pgalloc.c
+++ b/arch/frv/mm/pgalloc.c
@@ -37,11 +37,15 @@ pgtable_t pte_alloc_one(struct mm_struct *mm, unsigned long address)
 #else
 	page = alloc_pages(GFP_KERNEL|__GFP_REPEAT, 0);
 #endif
-	if (page) {
-		clear_highpage(page);
-		pgtable_page_ctor(page);
-		flush_dcache_page(page);
+	if (!page)
+		return NULL;
+
+	clear_highpage(page);
+	if (!pgtable_page_ctor(page)) {
+		__free_page(page);
+		return NULL;
 	}
+	flush_dcache_page(page);
 	return page;
 }
 
diff --git a/arch/hexagon/Kconfig b/arch/hexagon/Kconfig
index 99041b07e610..09df2608f40a 100644
--- a/arch/hexagon/Kconfig
+++ b/arch/hexagon/Kconfig
@@ -4,7 +4,6 @@ comment "Linux Kernel Configuration for Hexagon"
 config HEXAGON
 	def_bool y
 	select HAVE_OPROFILE
-	select USE_GENERIC_SMP_HELPERS if SMP
 	# Other pending projects/to-do items.
 	# select HAVE_REGS_AND_STACK_ACCESS_API
 	# select HAVE_HW_BREAKPOINT if PERF_EVENTS
diff --git a/arch/hexagon/include/asm/pgalloc.h b/arch/hexagon/include/asm/pgalloc.h
index 679bf6d66487..4c9d382d7798 100644
--- a/arch/hexagon/include/asm/pgalloc.h
+++ b/arch/hexagon/include/asm/pgalloc.h
@@ -65,10 +65,12 @@ static inline struct page *pte_alloc_one(struct mm_struct *mm,
 	struct page *pte;
 
 	pte = alloc_page(GFP_KERNEL | __GFP_REPEAT | __GFP_ZERO);
-
-	if (pte)
-		pgtable_page_ctor(pte);
-
+	if (!pte)
+		return NULL;
+	if (!pgtable_page_ctor(pte)) {
+		__free_page(pte);
+		return NULL;
+	}
 	return pte;
 }
 
diff --git a/arch/hexagon/include/asm/thread_info.h b/arch/hexagon/include/asm/thread_info.h
index f7c32406a711..a59dad3b3695 100644
--- a/arch/hexagon/include/asm/thread_info.h
+++ b/arch/hexagon/include/asm/thread_info.h
@@ -73,10 +73,6 @@ struct thread_info {
 
 #endif  /* __ASSEMBLY__  */
 
-/*  looks like "linux/hardirq.h" uses this.  */
-
-#define PREEMPT_ACTIVE		0x10000000
-
 #ifndef __ASSEMBLY__
 
 #define INIT_THREAD_INFO(tsk)                   \
diff --git a/arch/ia64/Kconfig b/arch/ia64/Kconfig
index 7740ab10a171..4e4119b0e691 100644
--- a/arch/ia64/Kconfig
+++ b/arch/ia64/Kconfig
@@ -6,6 +6,7 @@ menu "Processor type and features"
 
 config IA64
 	bool
+	select ARCH_MIGHT_HAVE_PC_PARPORT
 	select PCI if (!IA64_HP_SIM)
 	select ACPI if (!IA64_HP_SIM)
 	select PM if (!IA64_HP_SIM)
@@ -343,7 +344,6 @@ config FORCE_MAX_ZONEORDER
 
 config SMP
 	bool "Symmetric multi-processing support"
-	select USE_GENERIC_SMP_HELPERS
 	help
 	  This enables support for systems with more than one CPU. If you have
 	  a system with only one CPU, say N.  If you have a system with more
diff --git a/arch/ia64/include/asm/kvm_host.h b/arch/ia64/include/asm/kvm_host.h
index 989dd3fe8de1..db95f570705f 100644
--- a/arch/ia64/include/asm/kvm_host.h
+++ b/arch/ia64/include/asm/kvm_host.h
@@ -234,10 +234,6 @@ struct kvm_vm_data {
 #define KVM_REQ_PTC_G		32
 #define KVM_REQ_RESUME		33
 
-#define KVM_HPAGE_GFN_SHIFT(x)	0
-#define KVM_NR_PAGE_SIZES	1
-#define KVM_PAGES_PER_HPAGE(x)	1
-
 struct kvm;
 struct kvm_vcpu;
 
@@ -480,7 +476,7 @@ struct kvm_arch {
 
 	struct list_head assigned_dev_head;
 	struct iommu_domain *iommu_domain;
-	int iommu_flags;
+	bool iommu_noncoherent;
 
 	unsigned long irq_sources_bitmap;
 	unsigned long irq_states[KVM_IOAPIC_NUM_PINS];
diff --git a/arch/ia64/include/asm/pgalloc.h b/arch/ia64/include/asm/pgalloc.h
index 96a8d927db28..5767cdfc08db 100644
--- a/arch/ia64/include/asm/pgalloc.h
+++ b/arch/ia64/include/asm/pgalloc.h
@@ -91,7 +91,10 @@ static inline pgtable_t pte_alloc_one(struct mm_struct *mm, unsigned long addr)
 	if (!pg)
 		return NULL;
 	page = virt_to_page(pg);
-	pgtable_page_ctor(page);
+	if (!pgtable_page_ctor(page)) {
+		quicklist_free(0, NULL, pg);
+		return NULL;
+	}
 	return page;
 }
 
diff --git a/arch/ia64/include/asm/thread_info.h b/arch/ia64/include/asm/thread_info.h
index cade13dd0299..5957cf61f898 100644
--- a/arch/ia64/include/asm/thread_info.h
+++ b/arch/ia64/include/asm/thread_info.h
@@ -11,9 +11,6 @@
 #include <asm/processor.h>
 #include <asm/ptrace.h>
 
-#define PREEMPT_ACTIVE_BIT 30
-#define PREEMPT_ACTIVE	(1 << PREEMPT_ACTIVE_BIT)
-
 #ifndef __ASSEMBLY__
 
 /*
diff --git a/arch/ia64/include/asm/xen/page-coherent.h b/arch/ia64/include/asm/xen/page-coherent.h
new file mode 100644
index 000000000000..96e42f97fa1f
--- /dev/null
+++ b/arch/ia64/include/asm/xen/page-coherent.h
@@ -0,0 +1,38 @@
+#ifndef _ASM_IA64_XEN_PAGE_COHERENT_H
+#define _ASM_IA64_XEN_PAGE_COHERENT_H
+
+#include <asm/page.h>
+#include <linux/dma-attrs.h>
+#include <linux/dma-mapping.h>
+
+static inline void *xen_alloc_coherent_pages(struct device *hwdev, size_t size,
+		dma_addr_t *dma_handle, gfp_t flags,
+		struct dma_attrs *attrs)
+{
+	void *vstart = (void*)__get_free_pages(flags, get_order(size));
+	*dma_handle = virt_to_phys(vstart);
+	return vstart;
+}
+
+static inline void xen_free_coherent_pages(struct device *hwdev, size_t size,
+		void *cpu_addr, dma_addr_t dma_handle,
+		struct dma_attrs *attrs)
+{
+	free_pages((unsigned long) cpu_addr, get_order(size));
+}
+
+static inline void xen_dma_map_page(struct device *hwdev, struct page *page,
+	     unsigned long offset, size_t size, enum dma_data_direction dir,
+	     struct dma_attrs *attrs) { }
+
+static inline void xen_dma_unmap_page(struct device *hwdev, dma_addr_t handle,
+		size_t size, enum dma_data_direction dir,
+		struct dma_attrs *attrs) { }
+
+static inline void xen_dma_sync_single_for_cpu(struct device *hwdev,
+		dma_addr_t handle, size_t size, enum dma_data_direction dir) { }
+
+static inline void xen_dma_sync_single_for_device(struct device *hwdev,
+		dma_addr_t handle, size_t size, enum dma_data_direction dir) { }
+
+#endif /* _ASM_IA64_XEN_PAGE_COHERENT_H */
diff --git a/arch/ia64/kernel/entry.S b/arch/ia64/kernel/entry.S
index 7a53530f22c2..ddea607f948a 100644
--- a/arch/ia64/kernel/entry.S
+++ b/arch/ia64/kernel/entry.S
@@ -1169,21 +1169,8 @@ skip_rbs_switch:
 .work_pending:
 	tbit.z p6,p0=r31,TIF_NEED_RESCHED	// is resched not needed?
 (p6)	br.cond.sptk.few .notify
-#ifdef CONFIG_PREEMPT
-(pKStk) dep r21=-1,r0,PREEMPT_ACTIVE_BIT,1
-	;;
-(pKStk) st4 [r20]=r21
-#endif
-	SSM_PSR_I(p0, p6, r2)	// enable interrupts
-	br.call.spnt.many rp=schedule
+	br.call.spnt.many rp=preempt_schedule_irq
 .ret9:	cmp.eq p6,p0=r0,r0	// p6 <- 1 (re-check)
-	RSM_PSR_I(p0, r2, r20)	// disable interrupts
-	;;
-#ifdef CONFIG_PREEMPT
-(pKStk)	adds r20=TI_PRE_COUNT+IA64_TASK_SIZE,r13
-	;;
-(pKStk)	st4 [r20]=r0		// preempt_count() <- 0
-#endif
 (pLvSys)br.cond.sptk.few  __paravirt_pending_syscall_end
 	br.cond.sptk.many .work_processed_kernel
 
diff --git a/arch/ia64/kernel/kprobes.c b/arch/ia64/kernel/kprobes.c
index f8280a766a78..074fde49c9e6 100644
--- a/arch/ia64/kernel/kprobes.c
+++ b/arch/ia64/kernel/kprobes.c
@@ -947,7 +947,7 @@ int __kprobes kprobe_fault_handler(struct pt_regs *regs, int trapnr)
 	case KPROBE_HIT_SSDONE:
 		/*
 		 * We increment the nmissed count for accounting,
-		 * we can also use npre/npostfault count for accouting
+		 * we can also use npre/npostfault count for accounting
 		 * these specific fault cases.
 		 */
 		kprobes_inc_nmissed_count(cur);
diff --git a/arch/ia64/kvm/kvm-ia64.c b/arch/ia64/kvm/kvm-ia64.c
index bdfd8789b376..985bf80c622e 100644
--- a/arch/ia64/kvm/kvm-ia64.c
+++ b/arch/ia64/kvm/kvm-ia64.c
@@ -1550,12 +1550,13 @@ int kvm_arch_vcpu_fault(struct kvm_vcpu *vcpu, struct vm_fault *vmf)
 	return VM_FAULT_SIGBUS;
 }
 
-void kvm_arch_free_memslot(struct kvm_memory_slot *free,
+void kvm_arch_free_memslot(struct kvm *kvm, struct kvm_memory_slot *free,
 			   struct kvm_memory_slot *dont)
 {
 }
 
-int kvm_arch_create_memslot(struct kvm_memory_slot *slot, unsigned long npages)
+int kvm_arch_create_memslot(struct kvm *kvm, struct kvm_memory_slot *slot,
+			    unsigned long npages)
 {
 	return 0;
 }
diff --git a/arch/m32r/Kconfig b/arch/m32r/Kconfig
index 75661fbf4529..09ef94a8a7c3 100644
--- a/arch/m32r/Kconfig
+++ b/arch/m32r/Kconfig
@@ -275,7 +275,6 @@ source "kernel/Kconfig.preempt"
 
 config SMP
 	bool "Symmetric multi-processing support"
-	select USE_GENERIC_SMP_HELPERS
 	---help---
 	  This enables support for systems with more than one CPU. If you have
 	  a system with only one CPU, like most personal computers, say N. If
diff --git a/arch/m32r/include/asm/hardirq.h b/arch/m32r/include/asm/hardirq.h
index 4c31c0ae215e..5f2ac4f64ddf 100644
--- a/arch/m32r/include/asm/hardirq.h
+++ b/arch/m32r/include/asm/hardirq.h
@@ -3,22 +3,6 @@
 #define __ASM_HARDIRQ_H
 
 #include <asm/irq.h>
-
-#if NR_IRQS > 256
-#define HARDIRQ_BITS	9
-#else
-#define HARDIRQ_BITS	8
-#endif
-
-/*
- * The hardirq mask has to be large enough to have
- * space for potentially all IRQ sources in the system
- * nesting on a single CPU:
- */
-#if (1 << HARDIRQ_BITS) < NR_IRQS
-# error HARDIRQ_BITS is too low!
-#endif
-
 #include <asm-generic/hardirq.h>
 
 #endif /* __ASM_HARDIRQ_H */
diff --git a/arch/m32r/include/asm/mmu_context.h b/arch/m32r/include/asm/mmu_context.h
index a979a4198168..9fc78fc44445 100644
--- a/arch/m32r/include/asm/mmu_context.h
+++ b/arch/m32r/include/asm/mmu_context.h
@@ -45,7 +45,7 @@ static inline void get_new_mmu_context(struct mm_struct *mm)
 		   Flush all TLB and start new cycle. */
 		local_flush_tlb_all();
 		/* Fix version if needed.
-		   Note that we avoid version #0 to distingush NO_CONTEXT. */
+		   Note that we avoid version #0 to distinguish NO_CONTEXT. */
 		if (!mc)
 			mmu_context_cache = mc = MMU_CONTEXT_FIRST_VERSION;
 	}
diff --git a/arch/m32r/include/asm/pgalloc.h b/arch/m32r/include/asm/pgalloc.h
index 0fc736198979..2d55a064ccac 100644
--- a/arch/m32r/include/asm/pgalloc.h
+++ b/arch/m32r/include/asm/pgalloc.h
@@ -43,7 +43,12 @@ static __inline__ pgtable_t pte_alloc_one(struct mm_struct *mm,
 {
 	struct page *pte = alloc_page(GFP_KERNEL|__GFP_ZERO);
 
-	pgtable_page_ctor(pte);
+	if (!pte)
+		return NULL;
+	if (!pgtable_page_ctor(pte)) {
+		__free_page(pte);
+		return NULL;
+	}
 	return pte;
 }
 
diff --git a/arch/m32r/include/asm/thread_info.h b/arch/m32r/include/asm/thread_info.h
index c074f4c2e858..00171703402f 100644
--- a/arch/m32r/include/asm/thread_info.h
+++ b/arch/m32r/include/asm/thread_info.h
@@ -53,8 +53,6 @@ struct thread_info {
 
 #endif
 
-#define PREEMPT_ACTIVE		0x10000000
-
 #define THREAD_SIZE		(PAGE_SIZE << 1)
 #define THREAD_SIZE_ORDER	1
 /*
diff --git a/arch/m32r/kernel/entry.S b/arch/m32r/kernel/entry.S
index 0c01543f10cd..7c3db9940ce1 100644
--- a/arch/m32r/kernel/entry.S
+++ b/arch/m32r/kernel/entry.S
@@ -182,13 +182,7 @@ need_resched:
 	ld	r4, PSW(sp)		; interrupts off (exception path) ?
 	and3	r4, r4, #0x4000
 	beqz	r4, restore_all
-	LDIMM	(r4, PREEMPT_ACTIVE)
-	st	r4, @(TI_PRE_COUNT, r8)
-	ENABLE_INTERRUPTS(r4)
-	bl	schedule
-	ldi	r4, #0
-	st	r4, @(TI_PRE_COUNT, r8)
-	DISABLE_INTERRUPTS(r4)
+	bl	preempt_schedule_irq
 	bra	need_resched
 #endif
 
diff --git a/arch/m68k/Kconfig b/arch/m68k/Kconfig
index 311a300d48cc..75f25a8e3001 100644
--- a/arch/m68k/Kconfig
+++ b/arch/m68k/Kconfig
@@ -1,6 +1,7 @@
 config M68K
 	bool
 	default y
+	select ARCH_MIGHT_HAVE_PC_PARPORT if ISA
 	select HAVE_IDE
 	select HAVE_AOUT if MMU
 	select HAVE_DEBUG_BUGVERBOSE
diff --git a/arch/m68k/include/asm/hardirq.h b/arch/m68k/include/asm/hardirq.h
index db30ed276878..6c618529d9b9 100644
--- a/arch/m68k/include/asm/hardirq.h
+++ b/arch/m68k/include/asm/hardirq.h
@@ -5,17 +5,6 @@
 #include <linux/cache.h>
 #include <asm/irq.h>
 
-#define HARDIRQ_BITS	8
-
-/*
- * The hardirq mask has to be large enough to have
- * space for potentially all IRQ sources in the system
- * nesting on a single CPU:
- */
-#if (1 << HARDIRQ_BITS) < NR_IRQS
-# error HARDIRQ_BITS is too low!
-#endif
-
 #ifdef CONFIG_MMU
 
 static inline void ack_bad_irq(unsigned int irq)
diff --git a/arch/m68k/include/asm/mcf_pgalloc.h b/arch/m68k/include/asm/mcf_pgalloc.h
index 313f3dd23cdc..f9924fbcfe42 100644
--- a/arch/m68k/include/asm/mcf_pgalloc.h
+++ b/arch/m68k/include/asm/mcf_pgalloc.h
@@ -56,6 +56,10 @@ static inline struct page *pte_alloc_one(struct mm_struct *mm,
 
 	if (!page)
 		return NULL;
+	if (!pgtable_page_ctor(page)) {
+		__free_page(page);
+		return NULL;
+	}
 
 	pte = kmap(page);
 	if (pte) {
diff --git a/arch/m68k/include/asm/motorola_pgalloc.h b/arch/m68k/include/asm/motorola_pgalloc.h
index 2f02f264e694..24bcba496c75 100644
--- a/arch/m68k/include/asm/motorola_pgalloc.h
+++ b/arch/m68k/include/asm/motorola_pgalloc.h
@@ -29,18 +29,22 @@ static inline void pte_free_kernel(struct mm_struct *mm, pte_t *pte)
 
 static inline pgtable_t pte_alloc_one(struct mm_struct *mm, unsigned long address)
 {
-	struct page *page = alloc_pages(GFP_KERNEL|__GFP_REPEAT|__GFP_ZERO, 0);
+	struct page *page;
 	pte_t *pte;
 
+	page = alloc_pages(GFP_KERNEL|__GFP_REPEAT|__GFP_ZERO, 0);
 	if(!page)
 		return NULL;
+	if (!pgtable_page_ctor(page)) {
+		__free_page(page);
+		return NULL;
+	}
 
 	pte = kmap(page);
 	__flush_page_to_ram(pte);
 	flush_tlb_kernel_page(pte);
 	nocache_page(pte);
 	kunmap(page);
-	pgtable_page_ctor(page);
 	return page;
 }
 
diff --git a/arch/m68k/include/asm/sun3_pgalloc.h b/arch/m68k/include/asm/sun3_pgalloc.h
index 48d80d5a666f..f868506e3350 100644
--- a/arch/m68k/include/asm/sun3_pgalloc.h
+++ b/arch/m68k/include/asm/sun3_pgalloc.h
@@ -59,7 +59,10 @@ static inline pgtable_t pte_alloc_one(struct mm_struct *mm,
 		return NULL;
 
 	clear_highpage(page);
-	pgtable_page_ctor(page);
+	if (!pgtable_page_ctor(page)) {
+		__free_page(page);
+		return NULL;
+	}
 	return page;
 
 }
diff --git a/arch/m68k/include/asm/thread_info.h b/arch/m68k/include/asm/thread_info.h
index 126131f94a2c..21a4784ca5a1 100644
--- a/arch/m68k/include/asm/thread_info.h
+++ b/arch/m68k/include/asm/thread_info.h
@@ -35,8 +35,6 @@ struct thread_info {
 };
 #endif /* __ASSEMBLY__ */
 
-#define PREEMPT_ACTIVE		0x4000000
-
 #define INIT_THREAD_INFO(tsk)			\
 {						\
 	.task		= &tsk,			\
diff --git a/arch/m68k/kernel/entry.S b/arch/m68k/kernel/entry.S
index a78f5649e8de..b54ac7aba850 100644
--- a/arch/m68k/kernel/entry.S
+++ b/arch/m68k/kernel/entry.S
@@ -45,7 +45,7 @@
 .globl system_call, buserr, trap, resume
 .globl sys_call_table
 .globl __sys_fork, __sys_clone, __sys_vfork
-.globl ret_from_interrupt, bad_interrupt
+.globl bad_interrupt
 .globl auto_irqhandler_fixup
 .globl user_irqvec_fixup
 
@@ -275,8 +275,6 @@ do_delayed_trace:
 ENTRY(auto_inthandler)
 	SAVE_ALL_INT
 	GET_CURRENT(%d0)
-	movel	%d0,%a1
-	addqb	#1,%a1@(TINFO_PREEMPT+1)
 					|  put exception # in d0
 	bfextu	%sp@(PT_OFF_FORMATVEC){#4,#10},%d0
 	subw	#VEC_SPUR,%d0
@@ -286,32 +284,13 @@ ENTRY(auto_inthandler)
 auto_irqhandler_fixup = . + 2
 	jsr	do_IRQ			|  process the IRQ
 	addql	#8,%sp			|  pop parameters off stack
-
-ret_from_interrupt:
-	movel	%curptr@(TASK_STACK),%a1
-	subqb	#1,%a1@(TINFO_PREEMPT+1)
-	jeq	ret_from_last_interrupt
-2:	RESTORE_ALL
-
-	ALIGN
-ret_from_last_interrupt:
-	moveq	#(~ALLOWINT>>8)&0xff,%d0
-	andb	%sp@(PT_OFF_SR),%d0
-	jne	2b
-
-	/* check if we need to do software interrupts */
-	tstl	irq_stat+CPUSTAT_SOFTIRQ_PENDING
-	jeq	.Lret_from_exception
-	pea	ret_from_exception
-	jra	do_softirq
+	jra	ret_from_exception
 
 /* Handler for user defined interrupt vectors */
 
 ENTRY(user_inthandler)
 	SAVE_ALL_INT
 	GET_CURRENT(%d0)
-	movel	%d0,%a1
-	addqb	#1,%a1@(TINFO_PREEMPT+1)
 					|  put exception # in d0
 	bfextu	%sp@(PT_OFF_FORMATVEC){#4,#10},%d0
 user_irqvec_fixup = . + 2
@@ -321,29 +300,18 @@ user_irqvec_fixup = . + 2
 	movel	%d0,%sp@-		|  put vector # on stack
 	jsr	do_IRQ			|  process the IRQ
 	addql	#8,%sp			|  pop parameters off stack
-
-	movel	%curptr@(TASK_STACK),%a1
-	subqb	#1,%a1@(TINFO_PREEMPT+1)
-	jeq	ret_from_last_interrupt
-	RESTORE_ALL
+	jra	ret_from_exception
 
 /* Handler for uninitialized and spurious interrupts */
 
 ENTRY(bad_inthandler)
 	SAVE_ALL_INT
 	GET_CURRENT(%d0)
-	movel	%d0,%a1
-	addqb	#1,%a1@(TINFO_PREEMPT+1)
 
 	movel	%sp,%sp@-
 	jsr	handle_badint
 	addql	#4,%sp
-
-	movel	%curptr@(TASK_STACK),%a1
-	subqb	#1,%a1@(TINFO_PREEMPT+1)
-	jeq	ret_from_last_interrupt
-	RESTORE_ALL
-
+	jra	ret_from_exception
 
 resume:
 	/*
diff --git a/arch/m68k/kernel/ints.c b/arch/m68k/kernel/ints.c
index 4d7da384eea0..077d3a70fed1 100644
--- a/arch/m68k/kernel/ints.c
+++ b/arch/m68k/kernel/ints.c
@@ -58,12 +58,6 @@ void __init init_IRQ(void)
 {
 	int i;
 
-	/* assembly irq entry code relies on this... */
-	if (HARDIRQ_MASK != 0x00ff0000) {
-		extern void hardirq_mask_is_broken(void);
-		hardirq_mask_is_broken();
-	}
-
 	for (i = IRQ_AUTO_1; i <= IRQ_AUTO_7; i++)
 		irq_set_chip_and_handler(i, &auto_irq_chip, handle_simple_irq);
 
diff --git a/arch/m68k/platform/68000/entry.S b/arch/m68k/platform/68000/entry.S
index 7f91c2fde509..23ac054c6e1a 100644
--- a/arch/m68k/platform/68000/entry.S
+++ b/arch/m68k/platform/68000/entry.S
@@ -27,7 +27,6 @@
 .globl ret_from_exception
 .globl ret_from_signal
 .globl sys_call_table
-.globl ret_from_interrupt
 .globl bad_interrupt
 .globl inthandler1
 .globl inthandler2
@@ -137,7 +136,7 @@ inthandler1:
 	movel	#65,%sp@- 		/*  put vector # on stack*/
 	jbsr	process_int		/*  process the IRQ*/
 3:     	addql	#8,%sp			/*  pop parameters off stack*/
-	bra	ret_from_interrupt
+	bra	ret_from_exception
 
 inthandler2:
 	SAVE_ALL_INT
@@ -148,7 +147,7 @@ inthandler2:
 	movel	#66,%sp@- 		/*  put vector # on stack*/
 	jbsr	process_int		/*  process the IRQ*/
 3:     	addql	#8,%sp			/*  pop parameters off stack*/
-	bra	ret_from_interrupt
+	bra	ret_from_exception
 
 inthandler3:
 	SAVE_ALL_INT
@@ -159,7 +158,7 @@ inthandler3:
 	movel	#67,%sp@- 		/*  put vector # on stack*/
 	jbsr	process_int		/*  process the IRQ*/
 3:     	addql	#8,%sp			/*  pop parameters off stack*/
-	bra	ret_from_interrupt
+	bra	ret_from_exception
 
 inthandler4:
 	SAVE_ALL_INT
@@ -170,7 +169,7 @@ inthandler4:
 	movel	#68,%sp@- 		/*  put vector # on stack*/
 	jbsr	process_int		/*  process the IRQ*/
 3:     	addql	#8,%sp			/*  pop parameters off stack*/
-	bra	ret_from_interrupt
+	bra	ret_from_exception
 
 inthandler5:
 	SAVE_ALL_INT
@@ -181,7 +180,7 @@ inthandler5:
 	movel	#69,%sp@- 		/*  put vector # on stack*/
 	jbsr	process_int		/*  process the IRQ*/
 3:     	addql	#8,%sp			/*  pop parameters off stack*/
-	bra	ret_from_interrupt
+	bra	ret_from_exception
 
 inthandler6:
 	SAVE_ALL_INT
@@ -192,7 +191,7 @@ inthandler6:
 	movel	#70,%sp@- 		/*  put vector # on stack*/
 	jbsr	process_int		/*  process the IRQ*/
 3:     	addql	#8,%sp			/*  pop parameters off stack*/
-	bra	ret_from_interrupt
+	bra	ret_from_exception
 
 inthandler7:
 	SAVE_ALL_INT
@@ -203,7 +202,7 @@ inthandler7:
 	movel	#71,%sp@- 		/*  put vector # on stack*/
 	jbsr	process_int		/*  process the IRQ*/
 3:     	addql	#8,%sp			/*  pop parameters off stack*/
-	bra	ret_from_interrupt
+	bra	ret_from_exception
 
 inthandler:
 	SAVE_ALL_INT
@@ -214,23 +213,7 @@ inthandler:
 	movel	%d0,%sp@- 		/*  put vector # on stack*/
 	jbsr	process_int		/*  process the IRQ*/
 3:     	addql	#8,%sp			/*  pop parameters off stack*/
-	bra	ret_from_interrupt
-
-ret_from_interrupt:
-	jeq	1f
-2:
-	RESTORE_ALL
-1:
-	moveb	%sp@(PT_OFF_SR), %d0
-	and	#7, %d0
-	jhi	2b
-
-	/* check if we need to do software interrupts */
-	jeq	ret_from_exception
-
-	pea	ret_from_exception
-	jra	do_softirq
-
+	bra	ret_from_exception
 
 /*
  * Handler for uninitialized and spurious interrupts.
diff --git a/arch/m68k/platform/68360/entry.S b/arch/m68k/platform/68360/entry.S
index 904fd9a4af4e..447c33ef37fd 100644
--- a/arch/m68k/platform/68360/entry.S
+++ b/arch/m68k/platform/68360/entry.S
@@ -29,7 +29,6 @@
 .globl ret_from_exception
 .globl ret_from_signal
 .globl sys_call_table
-.globl ret_from_interrupt
 .globl bad_interrupt
 .globl inthandler
 
@@ -132,26 +131,9 @@ inthandler:
 
 	movel	%sp,%sp@-
 	movel	%d0,%sp@- 		/*  put vector # on stack*/
-	jbsr	do_IRQ			/*  process the IRQ*/
-3:     	addql	#8,%sp			/*  pop parameters off stack*/
-	bra	ret_from_interrupt
-
-ret_from_interrupt:
-	jeq	1f
-2:
-	RESTORE_ALL
-1:
-	moveb	%sp@(PT_OFF_SR), %d0
-	and	#7, %d0
-	jhi	2b
-	/* check if we need to do software interrupts */
-
-	movel	irq_stat+CPUSTAT_SOFTIRQ_PENDING,%d0
-	jeq	ret_from_exception
-
-	pea	ret_from_exception
-	jra	do_softirq
-
+	jbsr	do_IRQ			/*  process the IRQ */
+	addql	#8,%sp			/*  pop parameters off stack*/
+	jra	ret_from_exception
 
 /*
  * Handler for uninitialized and spurious interrupts.
diff --git a/arch/metag/Kconfig b/arch/metag/Kconfig
index 36368eb07e13..e56abd2c1b4f 100644
--- a/arch/metag/Kconfig
+++ b/arch/metag/Kconfig
@@ -111,7 +111,6 @@ config METAG_META21
 config SMP
 	bool "Symmetric multi-processing support"
 	depends on METAG_META21 && METAG_META21_MMU
-	select USE_GENERIC_SMP_HELPERS
 	help
 	  This enables support for systems with more than one thread running
 	  Linux. If you have a system with only one thread running Linux,
diff --git a/arch/metag/include/asm/pgalloc.h b/arch/metag/include/asm/pgalloc.h
index 275d9285141c..3104df0a4822 100644
--- a/arch/metag/include/asm/pgalloc.h
+++ b/arch/metag/include/asm/pgalloc.h
@@ -52,8 +52,12 @@ static inline pgtable_t pte_alloc_one(struct mm_struct *mm,
 {
 	struct page *pte;
 	pte = alloc_pages(GFP_KERNEL | __GFP_REPEAT | __GFP_ZERO, 0);
-	if (pte)
-		pgtable_page_ctor(pte);
+	if (!pte)
+		return NULL;
+	if (!pgtable_page_ctor(pte)) {
+		__free_page(pte);
+		return NULL;
+	}
 	return pte;
 }
 
diff --git a/arch/metag/include/asm/thread_info.h b/arch/metag/include/asm/thread_info.h
index 7c4a33006142..b19e9c588a16 100644
--- a/arch/metag/include/asm/thread_info.h
+++ b/arch/metag/include/asm/thread_info.h
@@ -46,8 +46,6 @@ struct thread_info {
 
 #endif
 
-#define PREEMPT_ACTIVE		0x10000000
-
 #ifdef CONFIG_4KSTACKS
 #define THREAD_SHIFT		12
 #else
diff --git a/arch/microblaze/Kconfig b/arch/microblaze/Kconfig
index 655e1cadf692..e23cccde9c27 100644
--- a/arch/microblaze/Kconfig
+++ b/arch/microblaze/Kconfig
@@ -1,5 +1,6 @@
 config MICROBLAZE
 	def_bool y
+	select ARCH_MIGHT_HAVE_PC_PARPORT
 	select HAVE_MEMBLOCK
 	select HAVE_MEMBLOCK_NODE_MAP
 	select HAVE_FUNCTION_TRACER
diff --git a/arch/microblaze/include/asm/pgalloc.h b/arch/microblaze/include/asm/pgalloc.h
index ebd35792482c..7fdf7fabc7d7 100644
--- a/arch/microblaze/include/asm/pgalloc.h
+++ b/arch/microblaze/include/asm/pgalloc.h
@@ -122,8 +122,13 @@ static inline struct page *pte_alloc_one(struct mm_struct *mm,
 #endif
 
 	ptepage = alloc_pages(flags, 0);
-	if (ptepage)
-		clear_highpage(ptepage);
+	if (!ptepage)
+		return NULL;
+	clear_highpage(ptepage);
+	if (!pgtable_page_ctor(ptepage)) {
+		__free_page(ptepage);
+		return NULL;
+	}
 	return ptepage;
 }
 
@@ -158,8 +163,9 @@ extern inline void pte_free_slow(struct page *ptepage)
 	__free_page(ptepage);
 }
 
-extern inline void pte_free(struct mm_struct *mm, struct page *ptepage)
+static inline void pte_free(struct mm_struct *mm, struct page *ptepage)
 {
+	pgtable_page_dtor(ptepage);
 	__free_page(ptepage);
 }
 
diff --git a/arch/microblaze/include/asm/thread_info.h b/arch/microblaze/include/asm/thread_info.h
index de26ea6373de..8c9d36591a03 100644
--- a/arch/microblaze/include/asm/thread_info.h
+++ b/arch/microblaze/include/asm/thread_info.h
@@ -106,8 +106,6 @@ static inline struct thread_info *current_thread_info(void)
 /* thread information allocation */
 #endif /* __ASSEMBLY__ */
 
-#define PREEMPT_ACTIVE		0x10000000
-
 /*
  * thread information flags
  * - these are process state flags that various assembly files may
diff --git a/arch/mips/Kconfig b/arch/mips/Kconfig
index 17cc7ff8458c..650de3976e7a 100644
--- a/arch/mips/Kconfig
+++ b/arch/mips/Kconfig
@@ -1,6 +1,7 @@
 config MIPS
 	bool
 	default y
+	select ARCH_MIGHT_HAVE_PC_PARPORT
 	select HAVE_CONTEXT_TRACKING
 	select HAVE_GENERIC_DMA_COHERENT
 	select HAVE_IDE
@@ -2125,7 +2126,6 @@ source "mm/Kconfig"
 config SMP
 	bool "Multi-Processing support"
 	depends on SYS_SUPPORTS_SMP
-	select USE_GENERIC_SMP_HELPERS
 	help
 	  This enables support for systems with more than one CPU. If you have
 	  a system with only one CPU, like most personal computers, say N. If
diff --git a/arch/mips/configs/db1235_defconfig b/arch/mips/configs/db1235_defconfig
index e2b4ad55462f..28e49f226dc0 100644
--- a/arch/mips/configs/db1235_defconfig
+++ b/arch/mips/configs/db1235_defconfig
@@ -351,7 +351,6 @@ CONFIG_USB_OHCI_HCD=y
 CONFIG_USB_OHCI_HCD_PLATFORM=y
 CONFIG_USB_STORAGE=y
 CONFIG_MMC=y
-CONFIG_MMC_CLKGATE=y
 CONFIG_MMC_AU1X=y
 CONFIG_NEW_LEDS=y
 CONFIG_LEDS_CLASS=y
diff --git a/arch/mips/include/asm/kvm_host.h b/arch/mips/include/asm/kvm_host.h
index 4d6fa0bf1305..32966969f2f9 100644
--- a/arch/mips/include/asm/kvm_host.h
+++ b/arch/mips/include/asm/kvm_host.h
@@ -27,13 +27,6 @@
 
 #define KVM_COALESCED_MMIO_PAGE_OFFSET 1
 
-/* Don't support huge pages */
-#define KVM_HPAGE_GFN_SHIFT(x)	0
-
-/* We don't currently support large pages. */
-#define KVM_NR_PAGE_SIZES	1
-#define KVM_PAGES_PER_HPAGE(x)	1
-
 
 
 /* Special address that contains the comm page, used for reducing # of traps */
diff --git a/arch/mips/include/asm/octeon/cvmx-pip.h b/arch/mips/include/asm/octeon/cvmx-pip.h
index a76fe5a57a9f..df69bfd2b006 100644
--- a/arch/mips/include/asm/octeon/cvmx-pip.h
+++ b/arch/mips/include/asm/octeon/cvmx-pip.h
@@ -192,13 +192,13 @@ typedef struct {
 	/* Number of packets processed by PIP */
 	uint32_t packets;
 	/*
-	 * Number of indentified L2 multicast packets.	Does not
+	 * Number of identified L2 multicast packets.	Does not
 	 * include broadcast packets.  Only includes packets whose
 	 * parse mode is SKIP_TO_L2
 	 */
 	uint32_t multicast_packets;
 	/*
-	 * Number of indentified L2 broadcast packets.	Does not
+	 * Number of identified L2 broadcast packets.	Does not
 	 * include multicast packets.  Only includes packets whose
 	 * parse mode is SKIP_TO_L2
 	 */
diff --git a/arch/mips/include/asm/pgalloc.h b/arch/mips/include/asm/pgalloc.h
index 881d18b4e298..b336037e8768 100644
--- a/arch/mips/include/asm/pgalloc.h
+++ b/arch/mips/include/asm/pgalloc.h
@@ -80,9 +80,12 @@ static inline struct page *pte_alloc_one(struct mm_struct *mm,
 	struct page *pte;
 
 	pte = alloc_pages(GFP_KERNEL | __GFP_REPEAT, PTE_ORDER);
-	if (pte) {
-		clear_highpage(pte);
-		pgtable_page_ctor(pte);
+	if (!pte)
+		return NULL;
+	clear_highpage(pte);
+	if (!pgtable_page_ctor(pte)) {
+		__free_page(pte);
+		return NULL;
 	}
 	return pte;
 }
diff --git a/arch/mips/include/asm/thread_info.h b/arch/mips/include/asm/thread_info.h
index f9b24bfbdbae..4f58ef6d0eed 100644
--- a/arch/mips/include/asm/thread_info.h
+++ b/arch/mips/include/asm/thread_info.h
@@ -92,8 +92,6 @@ static inline struct thread_info *current_thread_info(void)
 
 #define STACK_WARN	(THREAD_SIZE / 8)
 
-#define PREEMPT_ACTIVE		0x10000000
-
 /*
  * thread information flags
  * - these are process state flags that various assembly files may need to
diff --git a/arch/mips/kvm/kvm_mips.c b/arch/mips/kvm/kvm_mips.c
index a7b044536de4..73b34827826c 100644
--- a/arch/mips/kvm/kvm_mips.c
+++ b/arch/mips/kvm/kvm_mips.c
@@ -198,12 +198,13 @@ kvm_arch_dev_ioctl(struct file *filp, unsigned int ioctl, unsigned long arg)
 	return -ENOIOCTLCMD;
 }
 
-void kvm_arch_free_memslot(struct kvm_memory_slot *free,
+void kvm_arch_free_memslot(struct kvm *kvm, struct kvm_memory_slot *free,
 			   struct kvm_memory_slot *dont)
 {
 }
 
-int kvm_arch_create_memslot(struct kvm_memory_slot *slot, unsigned long npages)
+int kvm_arch_create_memslot(struct kvm *kvm, struct kvm_memory_slot *slot,
+			    unsigned long npages)
 {
 	return 0;
 }
diff --git a/arch/mn10300/Kconfig b/arch/mn10300/Kconfig
index 6aaa1607001a..8bde9237d13b 100644
--- a/arch/mn10300/Kconfig
+++ b/arch/mn10300/Kconfig
@@ -181,7 +181,6 @@ endmenu
 config SMP
 	bool "Symmetric multi-processing support"
 	default y
-	select USE_GENERIC_SMP_HELPERS
 	depends on MN10300_PROC_MN2WS0038 || MN10300_PROC_MN2WS0050
 	---help---
 	  This enables support for systems with more than one CPU. If you have
diff --git a/arch/mn10300/include/asm/mmu_context.h b/arch/mn10300/include/asm/mmu_context.h
index c67c2b5365a6..75dbe696f830 100644
--- a/arch/mn10300/include/asm/mmu_context.h
+++ b/arch/mn10300/include/asm/mmu_context.h
@@ -71,7 +71,7 @@ static inline unsigned long allocate_mmu_context(struct mm_struct *mm)
 		local_flush_tlb_all();
 
 		/* fix the TLB version if needed (we avoid version #0 so as to
-		 * distingush MMU_NO_CONTEXT) */
+		 * distinguish MMU_NO_CONTEXT) */
 		if (!mc)
 			*pmc = mc = MMU_CONTEXT_FIRST_VERSION;
 	}
diff --git a/arch/mn10300/include/asm/pgalloc.h b/arch/mn10300/include/asm/pgalloc.h
index 146bacf193ea..0f25d5fa86f3 100644
--- a/arch/mn10300/include/asm/pgalloc.h
+++ b/arch/mn10300/include/asm/pgalloc.h
@@ -46,6 +46,7 @@ static inline void pte_free_kernel(struct mm_struct *mm, pte_t *pte)
 
 static inline void pte_free(struct mm_struct *mm, struct page *pte)
 {
+	pgtable_page_dtor(pte);
 	__free_page(pte);
 }
 
diff --git a/arch/mn10300/include/asm/thread_info.h b/arch/mn10300/include/asm/thread_info.h
index 224b4262486d..bf280eaccd36 100644
--- a/arch/mn10300/include/asm/thread_info.h
+++ b/arch/mn10300/include/asm/thread_info.h
@@ -16,8 +16,6 @@
 
 #include <asm/page.h>
 
-#define PREEMPT_ACTIVE		0x10000000
-
 #ifdef CONFIG_4KSTACKS
 #define THREAD_SIZE		(4096)
 #define THREAD_SIZE_ORDER	(0)
diff --git a/arch/mn10300/mm/pgtable.c b/arch/mn10300/mm/pgtable.c
index bd9ada693f95..e77a7c728081 100644
--- a/arch/mn10300/mm/pgtable.c
+++ b/arch/mn10300/mm/pgtable.c
@@ -78,8 +78,13 @@ struct page *pte_alloc_one(struct mm_struct *mm, unsigned long address)
 #else
 	pte = alloc_pages(GFP_KERNEL|__GFP_REPEAT, 0);
 #endif
-	if (pte)
-		clear_highpage(pte);
+	if (!pte)
+		return NULL;
+	clear_highpage(pte);
+	if (!pgtable_page_ctor(pte)) {
+		__free_page(pte);
+		return NULL;
+	}
 	return pte;
 }
 
diff --git a/arch/openrisc/Makefile b/arch/openrisc/Makefile
index 4739b8302a58..89076a66eee2 100644
--- a/arch/openrisc/Makefile
+++ b/arch/openrisc/Makefile
@@ -24,7 +24,7 @@ OBJCOPYFLAGS    := -O binary -R .note -R .comment -S
 LDFLAGS_vmlinux :=
 LIBGCC 		:= $(shell $(CC) $(KBUILD_CFLAGS) -print-libgcc-file-name)
 
-KBUILD_CFLAGS	+= -pipe -ffixed-r10
+KBUILD_CFLAGS	+= -pipe -ffixed-r10 -D__linux__
 
 ifeq ($(CONFIG_OPENRISC_HAVE_INST_MUL),y)
 	KBUILD_CFLAGS += $(call cc-option,-mhard-mul)
diff --git a/arch/openrisc/configs/or1ksim_defconfig b/arch/openrisc/configs/or1ksim_defconfig
index ea172bdfa36a..42fe5303a370 100644
--- a/arch/openrisc/configs/or1ksim_defconfig
+++ b/arch/openrisc/configs/or1ksim_defconfig
@@ -1,9 +1,9 @@
 CONFIG_CROSS_COMPILE="or32-linux-"
+CONFIG_NO_HZ=y
 CONFIG_LOG_BUF_SHIFT=14
 CONFIG_BLK_DEV_INITRD=y
 # CONFIG_RD_GZIP is not set
 CONFIG_EXPERT=y
-# CONFIG_SYSCTL_SYSCALL is not set
 # CONFIG_KALLSYMS is not set
 # CONFIG_EPOLL is not set
 # CONFIG_TIMERFD is not set
@@ -15,7 +15,6 @@ CONFIG_SLOB=y
 CONFIG_MODULES=y
 # CONFIG_BLOCK is not set
 CONFIG_OPENRISC_BUILTIN_DTB="or1ksim"
-CONFIG_NO_HZ=y
 CONFIG_HZ_100=y
 CONFIG_NET=y
 CONFIG_PACKET=y
@@ -39,11 +38,8 @@ CONFIG_DEVTMPFS_MOUNT=y
 # CONFIG_FW_LOADER is not set
 CONFIG_PROC_DEVICETREE=y
 CONFIG_NETDEVICES=y
-CONFIG_MICREL_PHY=y
-CONFIG_NET_ETHERNET=y
 CONFIG_ETHOC=y
-# CONFIG_NETDEV_1000 is not set
-# CONFIG_NETDEV_10000 is not set
+CONFIG_MICREL_PHY=y
 # CONFIG_WLAN is not set
 # CONFIG_INPUT is not set
 # CONFIG_SERIO is not set
@@ -55,11 +51,9 @@ CONFIG_SERIAL_8250_CONSOLE=y
 CONFIG_SERIAL_OF_PLATFORM=y
 # CONFIG_HW_RANDOM is not set
 # CONFIG_HWMON is not set
-# CONFIG_MFD_SUPPORT is not set
 # CONFIG_USB_SUPPORT is not set
 # CONFIG_DNOTIFY is not set
 CONFIG_TMPFS=y
 CONFIG_NFS_FS=y
-CONFIG_NFS_V3=y
 # CONFIG_ENABLE_WARN_DEPRECATED is not set
 # CONFIG_ENABLE_MUST_CHECK is not set
diff --git a/arch/openrisc/include/asm/Kbuild b/arch/openrisc/include/asm/Kbuild
index 78405625e799..da1951a22907 100644
--- a/arch/openrisc/include/asm/Kbuild
+++ b/arch/openrisc/include/asm/Kbuild
@@ -65,6 +65,7 @@ generic-y += trace_clock.h
 generic-y += types.h
 generic-y += ucontext.h
 generic-y += user.h
+generic-y += vga.h
 generic-y += word-at-a-time.h
 generic-y += xor.h
 generic-y += preempt.h
diff --git a/arch/openrisc/include/asm/pgalloc.h b/arch/openrisc/include/asm/pgalloc.h
index 05c39ecd2efd..21484e5b9e9a 100644
--- a/arch/openrisc/include/asm/pgalloc.h
+++ b/arch/openrisc/include/asm/pgalloc.h
@@ -78,8 +78,13 @@ static inline struct page *pte_alloc_one(struct mm_struct *mm,
 {
 	struct page *pte;
 	pte = alloc_pages(GFP_KERNEL|__GFP_REPEAT, 0);
-	if (pte)
-		clear_page(page_address(pte));
+	if (!pte)
+		return NULL;
+	clear_page(page_address(pte));
+	if (!pgtable_page_ctor(pte)) {
+		__free_page(pte);
+		return NULL;
+	}
 	return pte;
 }
 
@@ -90,6 +95,7 @@ static inline void pte_free_kernel(struct mm_struct *mm, pte_t *pte)
 
 static inline void pte_free(struct mm_struct *mm, struct page *pte)
 {
+	pgtable_page_dtor(pte);
 	__free_page(pte);
 }
 
diff --git a/arch/openrisc/kernel/module.c b/arch/openrisc/kernel/module.c
index 10ff50f0202a..ef872ae4c878 100644
--- a/arch/openrisc/kernel/module.c
+++ b/arch/openrisc/kernel/module.c
@@ -47,12 +47,10 @@ int apply_relocate_add(Elf32_Shdr *sechdrs,
 			*location = value;
 			break;
 		case R_OR32_CONST:
-			location = (uint16_t *)location + 1;
-			*((uint16_t *)location) = (uint16_t) (value);
+			*((uint16_t *)location + 1) = value;
 			break;
 		case R_OR32_CONSTH:
-			location = (uint16_t *)location + 1;
-			*((uint16_t *)location) = (uint16_t) (value >> 16);
+			*((uint16_t *)location + 1) = value >> 16;
 			break;
 		case R_OR32_JUMPTARG:
 			value -= (uint32_t)location;
diff --git a/arch/openrisc/kernel/setup.c b/arch/openrisc/kernel/setup.c
index 09a769b69572..4fc7ccc0a2cf 100644
--- a/arch/openrisc/kernel/setup.c
+++ b/arch/openrisc/kernel/setup.c
@@ -40,6 +40,7 @@
 #include <linux/device.h>
 #include <linux/of_platform.h>
 
+#include <asm/sections.h>
 #include <asm/segment.h>
 #include <asm/pgtable.h>
 #include <asm/types.h>
@@ -75,7 +76,7 @@ static unsigned long __init setup_memory(void)
 
 	ram_start_pfn = PFN_UP(memory_start);
 	/* free_ram_start_pfn is first page after kernel */
-	free_ram_start_pfn = PFN_UP(__pa(&_end));
+	free_ram_start_pfn = PFN_UP(__pa(_end));
 	ram_end_pfn = PFN_DOWN(memblock_end_of_DRAM());
 
 	max_pfn = ram_end_pfn;
@@ -207,15 +208,15 @@ void __init setup_cpuinfo(void)
  * Falls back on built-in device tree in case null pointer is passed.
  */
 
-void __init or32_early_setup(unsigned int fdt)
+void __init or32_early_setup(void *fdt)
 {
-	if (fdt) {
-		early_init_devtree((void*) fdt);
-		printk(KERN_INFO "FDT at 0x%08x\n", fdt);
-	} else {
-		early_init_devtree(__dtb_start);
-		printk(KERN_INFO "Compiled-in FDT at %p\n", __dtb_start);
+	if (fdt)
+		pr_info("FDT at %p\n", fdt);
+	else {
+		fdt = __dtb_start;
+		pr_info("Compiled-in FDT at %p\n", fdt);
 	}
+	early_init_devtree(fdt);
 }
 
 static int __init openrisc_device_probe(void)
@@ -288,10 +289,10 @@ void __init setup_arch(char **cmdline_p)
 	setup_cpuinfo();
 
 	/* process 1's initial memory region is the kernel code/data */
-	init_mm.start_code = (unsigned long)&_stext;
-	init_mm.end_code = (unsigned long)&_etext;
-	init_mm.end_data = (unsigned long)&_edata;
-	init_mm.brk = (unsigned long)&_end;
+	init_mm.start_code = (unsigned long)_stext;
+	init_mm.end_code = (unsigned long)_etext;
+	init_mm.end_data = (unsigned long)_edata;
+	init_mm.brk = (unsigned long)_end;
 
 #ifdef CONFIG_BLK_DEV_INITRD
 	initrd_start = (unsigned long)&__initrd_start;
diff --git a/arch/openrisc/kernel/vmlinux.h b/arch/openrisc/kernel/vmlinux.h
index ee842a2d3f36..70b9ce41835c 100644
--- a/arch/openrisc/kernel/vmlinux.h
+++ b/arch/openrisc/kernel/vmlinux.h
@@ -1,10 +1,8 @@
 #ifndef __OPENRISC_VMLINUX_H_
 #define __OPENRISC_VMLINUX_H_
 
-extern char _stext, _etext, _edata, _end;
 #ifdef CONFIG_BLK_DEV_INITRD
 extern char __initrd_start, __initrd_end;
-extern char __initramfs_start;
 #endif
 
 extern u32 __dtb_start[];
diff --git a/arch/parisc/Kconfig b/arch/parisc/Kconfig
index 7dcde539d61e..b5f1858baf33 100644
--- a/arch/parisc/Kconfig
+++ b/arch/parisc/Kconfig
@@ -1,6 +1,7 @@
 config PARISC
 	def_bool y
 	select ARCH_HAS_DEBUG_STRICT_USER_COPY_CHECKS
+	select ARCH_MIGHT_HAVE_PC_PARPORT
 	select HAVE_IDE
 	select HAVE_OPROFILE
 	select HAVE_FUNCTION_TRACER if 64BIT
@@ -226,7 +227,6 @@ endchoice
 
 config SMP
 	bool "Symmetric multi-processing support"
-	select USE_GENERIC_SMP_HELPERS
 	---help---
 	  This enables support for systems with more than one CPU. If you have
 	  a system with only one CPU, like most personal computers, say N. If
diff --git a/arch/parisc/include/asm/pgalloc.h b/arch/parisc/include/asm/pgalloc.h
index fc987a1c12a8..f213f5b4c423 100644
--- a/arch/parisc/include/asm/pgalloc.h
+++ b/arch/parisc/include/asm/pgalloc.h
@@ -121,8 +121,12 @@ static inline pgtable_t
 pte_alloc_one(struct mm_struct *mm, unsigned long address)
 {
 	struct page *page = alloc_page(GFP_KERNEL|__GFP_REPEAT|__GFP_ZERO);
-	if (page)
-		pgtable_page_ctor(page);
+	if (!page)
+		return NULL;
+	if (!pgtable_page_ctor(page)) {
+		__free_page(page);
+		return NULL;
+	}
 	return page;
 }
 
diff --git a/arch/parisc/include/asm/thread_info.h b/arch/parisc/include/asm/thread_info.h
index bc7cf120106b..d5f97ea3a4e1 100644
--- a/arch/parisc/include/asm/thread_info.h
+++ b/arch/parisc/include/asm/thread_info.h
@@ -46,9 +46,6 @@ struct thread_info {
 #define THREAD_SIZE             (PAGE_SIZE << THREAD_SIZE_ORDER)
 #define THREAD_SHIFT            (PAGE_SHIFT + THREAD_SIZE_ORDER)
 
-#define PREEMPT_ACTIVE_BIT	28
-#define PREEMPT_ACTIVE		(1 << PREEMPT_ACTIVE_BIT)
-
 /*
  * thread information flags
  */
diff --git a/arch/powerpc/Kconfig b/arch/powerpc/Kconfig
index 2f898d63eb96..b44b52c0a8f0 100644
--- a/arch/powerpc/Kconfig
+++ b/arch/powerpc/Kconfig
@@ -85,6 +85,7 @@ config GENERIC_HWEIGHT
 config PPC
 	bool
 	default y
+	select ARCH_MIGHT_HAVE_PC_PARPORT
 	select BINFMT_ELF
 	select OF
 	select OF_EARLY_FLATTREE
@@ -106,7 +107,6 @@ config PPC
 	select HAVE_MEMBLOCK_NODE_MAP
 	select HAVE_DMA_ATTRS
 	select HAVE_DMA_API_DEBUG
-	select USE_GENERIC_SMP_HELPERS if SMP
 	select HAVE_OPROFILE
 	select HAVE_DEBUG_KMEMLEAK
 	select GENERIC_ATOMIC64 if PPC32
diff --git a/arch/powerpc/boot/dts/fsl/b4si-post.dtsi b/arch/powerpc/boot/dts/fsl/b4si-post.dtsi
index 4c617bf8cdb2..4f6e48277c46 100644
--- a/arch/powerpc/boot/dts/fsl/b4si-post.dtsi
+++ b/arch/powerpc/boot/dts/fsl/b4si-post.dtsi
@@ -223,13 +223,13 @@
 		reg = <0xe2000 0x1000>;
 	};
 
-/include/ "qoriq-dma-0.dtsi"
+/include/ "elo3-dma-0.dtsi"
 	dma@100300 {
 		fsl,iommu-parent = <&pamu0>;
 		fsl,liodn-reg = <&guts 0x580>; /* DMA1LIODNR */
 	};
 
-/include/ "qoriq-dma-1.dtsi"
+/include/ "elo3-dma-1.dtsi"
 	dma@101300 {
 		fsl,iommu-parent = <&pamu0>;
 		fsl,liodn-reg = <&guts 0x584>; /* DMA2LIODNR */
diff --git a/arch/powerpc/boot/dts/fsl/elo3-dma-0.dtsi b/arch/powerpc/boot/dts/fsl/elo3-dma-0.dtsi
new file mode 100644
index 000000000000..3c210e0d5201
--- /dev/null
+++ b/arch/powerpc/boot/dts/fsl/elo3-dma-0.dtsi
@@ -0,0 +1,82 @@
+/*
+ * QorIQ Elo3 DMA device tree stub [ controller @ offset 0x100000 ]
+ *
+ * Copyright 2013 Freescale Semiconductor Inc.
+ *
+ * Redistribution and use in source and binary forms, with or without
+ * modification, are permitted provided that the following conditions are met:
+ *     * Redistributions of source code must retain the above copyright
+ *       notice, this list of conditions and the following disclaimer.
+ *     * Redistributions in binary form must reproduce the above copyright
+ *       notice, this list of conditions and the following disclaimer in the
+ *       documentation and/or other materials provided with the distribution.
+ *     * Neither the name of Freescale Semiconductor nor the
+ *       names of its contributors may be used to endorse or promote products
+ *       derived from this software without specific prior written permission.
+ *
+ *
+ * ALTERNATIVELY, this software may be distributed under the terms of the
+ * GNU General Public License ("GPL") as published by the Free Software
+ * Foundation, either version 2 of that License or (at your option) any
+ * later version.
+ *
+ * THIS SOFTWARE IS PROVIDED BY Freescale Semiconductor ``AS IS'' AND ANY
+ * EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED
+ * WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE
+ * DISCLAIMED. IN NO EVENT SHALL Freescale Semiconductor BE LIABLE FOR ANY
+ * DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES
+ * (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES;
+ * LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND
+ * ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT
+ * (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF THIS
+ * SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
+ */
+
+dma0: dma@100300 {
+	#address-cells = <1>;
+	#size-cells = <1>;
+	compatible = "fsl,elo3-dma";
+	reg = <0x100300 0x4>,
+	      <0x100600 0x4>;
+	ranges = <0x0 0x100100 0x500>;
+	dma-channel@0 {
+		compatible = "fsl,eloplus-dma-channel";
+		reg = <0x0 0x80>;
+		interrupts = <28 2 0 0>;
+	};
+	dma-channel@80 {
+		compatible = "fsl,eloplus-dma-channel";
+		reg = <0x80 0x80>;
+		interrupts = <29 2 0 0>;
+	};
+	dma-channel@100 {
+		compatible = "fsl,eloplus-dma-channel";
+		reg = <0x100 0x80>;
+		interrupts = <30 2 0 0>;
+	};
+	dma-channel@180 {
+		compatible = "fsl,eloplus-dma-channel";
+		reg = <0x180 0x80>;
+		interrupts = <31 2 0 0>;
+	};
+	dma-channel@300 {
+		compatible = "fsl,eloplus-dma-channel";
+		reg = <0x300 0x80>;
+		interrupts = <76 2 0 0>;
+	};
+	dma-channel@380 {
+		compatible = "fsl,eloplus-dma-channel";
+		reg = <0x380 0x80>;
+		interrupts = <77 2 0 0>;
+	};
+	dma-channel@400 {
+		compatible = "fsl,eloplus-dma-channel";
+		reg = <0x400 0x80>;
+		interrupts = <78 2 0 0>;
+	};
+	dma-channel@480 {
+		compatible = "fsl,eloplus-dma-channel";
+		reg = <0x480 0x80>;
+		interrupts = <79 2 0 0>;
+	};
+};
diff --git a/arch/powerpc/boot/dts/fsl/elo3-dma-1.dtsi b/arch/powerpc/boot/dts/fsl/elo3-dma-1.dtsi
new file mode 100644
index 000000000000..cccf3bb38224
--- /dev/null
+++ b/arch/powerpc/boot/dts/fsl/elo3-dma-1.dtsi
@@ -0,0 +1,82 @@
+/*
+ * QorIQ Elo3 DMA device tree stub [ controller @ offset 0x101000 ]
+ *
+ * Copyright 2013 Freescale Semiconductor Inc.
+ *
+ * Redistribution and use in source and binary forms, with or without
+ * modification, are permitted provided that the following conditions are met:
+ *     * Redistributions of source code must retain the above copyright
+ *       notice, this list of conditions and the following disclaimer.
+ *     * Redistributions in binary form must reproduce the above copyright
+ *       notice, this list of conditions and the following disclaimer in the
+ *       documentation and/or other materials provided with the distribution.
+ *     * Neither the name of Freescale Semiconductor nor the
+ *       names of its contributors may be used to endorse or promote products
+ *       derived from this software without specific prior written permission.
+ *
+ *
+ * ALTERNATIVELY, this software may be distributed under the terms of the
+ * GNU General Public License ("GPL") as published by the Free Software
+ * Foundation, either version 2 of that License or (at your option) any
+ * later version.
+ *
+ * THIS SOFTWARE IS PROVIDED BY Freescale Semiconductor ``AS IS'' AND ANY
+ * EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED
+ * WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE
+ * DISCLAIMED. IN NO EVENT SHALL Freescale Semiconductor BE LIABLE FOR ANY
+ * DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES
+ * (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES;
+ * LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND
+ * ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT
+ * (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF THIS
+ * SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
+ */
+
+dma1: dma@101300 {
+	#address-cells = <1>;
+	#size-cells = <1>;
+	compatible = "fsl,elo3-dma";
+	reg = <0x101300 0x4>,
+	      <0x101600 0x4>;
+	ranges = <0x0 0x101100 0x500>;
+	dma-channel@0 {
+		compatible = "fsl,eloplus-dma-channel";
+		reg = <0x0 0x80>;
+		interrupts = <32 2 0 0>;
+	};
+	dma-channel@80 {
+		compatible = "fsl,eloplus-dma-channel";
+		reg = <0x80 0x80>;
+		interrupts = <33 2 0 0>;
+	};
+	dma-channel@100 {
+		compatible = "fsl,eloplus-dma-channel";
+		reg = <0x100 0x80>;
+		interrupts = <34 2 0 0>;
+	};
+	dma-channel@180 {
+		compatible = "fsl,eloplus-dma-channel";
+		reg = <0x180 0x80>;
+		interrupts = <35 2 0 0>;
+	};
+	dma-channel@300 {
+		compatible = "fsl,eloplus-dma-channel";
+		reg = <0x300 0x80>;
+		interrupts = <80 2 0 0>;
+	};
+	dma-channel@380 {
+		compatible = "fsl,eloplus-dma-channel";
+		reg = <0x380 0x80>;
+		interrupts = <81 2 0 0>;
+	};
+	dma-channel@400 {
+		compatible = "fsl,eloplus-dma-channel";
+		reg = <0x400 0x80>;
+		interrupts = <82 2 0 0>;
+	};
+	dma-channel@480 {
+		compatible = "fsl,eloplus-dma-channel";
+		reg = <0x480 0x80>;
+		interrupts = <83 2 0 0>;
+	};
+};
diff --git a/arch/powerpc/boot/dts/fsl/t4240si-post.dtsi b/arch/powerpc/boot/dts/fsl/t4240si-post.dtsi
index 510afa362de1..4143a9733cd0 100644
--- a/arch/powerpc/boot/dts/fsl/t4240si-post.dtsi
+++ b/arch/powerpc/boot/dts/fsl/t4240si-post.dtsi
@@ -387,8 +387,8 @@
 		reg	   = <0xea000 0x4000>;
 	};
 
-/include/ "qoriq-dma-0.dtsi"
-/include/ "qoriq-dma-1.dtsi"
+/include/ "elo3-dma-0.dtsi"
+/include/ "elo3-dma-1.dtsi"
 
 /include/ "qoriq-espi-0.dtsi"
 	spi@110000 {
diff --git a/arch/powerpc/include/asm/disassemble.h b/arch/powerpc/include/asm/disassemble.h
index 9b198d1b3b2b..856f8deb557a 100644
--- a/arch/powerpc/include/asm/disassemble.h
+++ b/arch/powerpc/include/asm/disassemble.h
@@ -77,4 +77,8 @@ static inline unsigned int get_d(u32 inst)
 	return inst & 0xffff;
 }
 
+static inline unsigned int get_oc(u32 inst)
+{
+	return (inst >> 11) & 0x7fff;
+}
 #endif /* __ASM_PPC_DISASSEMBLE_H__ */
diff --git a/arch/powerpc/include/asm/exception-64s.h b/arch/powerpc/include/asm/exception-64s.h
index cca12f084842..894662a5d4d5 100644
--- a/arch/powerpc/include/asm/exception-64s.h
+++ b/arch/powerpc/include/asm/exception-64s.h
@@ -198,12 +198,27 @@ END_FTR_SECTION_NESTED(ftr,ftr,943)
 	cmpwi	r10,0;							\
 	bne	do_kvm_##n
 
+#ifdef CONFIG_KVM_BOOK3S_HV_POSSIBLE
+/*
+ * If hv is possible, interrupts come into to the hv version
+ * of the kvmppc_interrupt code, which then jumps to the PR handler,
+ * kvmppc_interrupt_pr, if the guest is a PR guest.
+ */
+#define kvmppc_interrupt kvmppc_interrupt_hv
+#else
+#define kvmppc_interrupt kvmppc_interrupt_pr
+#endif
+
 #define __KVM_HANDLER(area, h, n)					\
 do_kvm_##n:								\
 	BEGIN_FTR_SECTION_NESTED(947)					\
 	ld	r10,area+EX_CFAR(r13);					\
 	std	r10,HSTATE_CFAR(r13);					\
 	END_FTR_SECTION_NESTED(CPU_FTR_CFAR,CPU_FTR_CFAR,947);		\
+	BEGIN_FTR_SECTION_NESTED(948)					\
+	ld	r10,area+EX_PPR(r13);					\
+	std	r10,HSTATE_PPR(r13);					\
+	END_FTR_SECTION_NESTED(CPU_FTR_HAS_PPR,CPU_FTR_HAS_PPR,948);	\
 	ld	r10,area+EX_R10(r13);					\
 	stw	r9,HSTATE_SCRATCH1(r13);				\
 	ld	r9,area+EX_R9(r13);					\
@@ -217,6 +232,10 @@ do_kvm_##n:								\
 	ld	r10,area+EX_R10(r13);					\
 	beq	89f;							\
 	stw	r9,HSTATE_SCRATCH1(r13);			\
+	BEGIN_FTR_SECTION_NESTED(948)					\
+	ld	r9,area+EX_PPR(r13);					\
+	std	r9,HSTATE_PPR(r13);					\
+	END_FTR_SECTION_NESTED(CPU_FTR_HAS_PPR,CPU_FTR_HAS_PPR,948);	\
 	ld	r9,area+EX_R9(r13);					\
 	std	r12,HSTATE_SCRATCH0(r13);			\
 	li	r12,n;							\
@@ -236,7 +255,7 @@ do_kvm_##n:								\
 #define KVM_HANDLER_SKIP(area, h, n)
 #endif
 
-#ifdef CONFIG_KVM_BOOK3S_PR
+#ifdef CONFIG_KVM_BOOK3S_PR_POSSIBLE
 #define KVMTEST_PR(n)			__KVMTEST(n)
 #define KVM_HANDLER_PR(area, h, n)	__KVM_HANDLER(area, h, n)
 #define KVM_HANDLER_PR_SKIP(area, h, n)	__KVM_HANDLER_SKIP(area, h, n)
diff --git a/arch/powerpc/include/asm/kvm_asm.h b/arch/powerpc/include/asm/kvm_asm.h
index 851bac7afa4b..1bd92fd43cfb 100644
--- a/arch/powerpc/include/asm/kvm_asm.h
+++ b/arch/powerpc/include/asm/kvm_asm.h
@@ -123,6 +123,8 @@
 #define BOOK3S_HFLAG_SLB			0x2
 #define BOOK3S_HFLAG_PAIRED_SINGLE		0x4
 #define BOOK3S_HFLAG_NATIVE_PS			0x8
+#define BOOK3S_HFLAG_MULTI_PGSIZE		0x10
+#define BOOK3S_HFLAG_NEW_TLBIE			0x20
 
 #define RESUME_FLAG_NV          (1<<0)  /* Reload guest nonvolatile state? */
 #define RESUME_FLAG_HOST        (1<<1)  /* Resume host? */
@@ -136,6 +138,8 @@
 #define KVM_GUEST_MODE_NONE	0
 #define KVM_GUEST_MODE_GUEST	1
 #define KVM_GUEST_MODE_SKIP	2
+#define KVM_GUEST_MODE_GUEST_HV	3
+#define KVM_GUEST_MODE_HOST_HV	4
 
 #define KVM_INST_FETCH_FAILED	-1
 
diff --git a/arch/powerpc/include/asm/kvm_book3s.h b/arch/powerpc/include/asm/kvm_book3s.h
index fa19e2f1a874..4a594b76674d 100644
--- a/arch/powerpc/include/asm/kvm_book3s.h
+++ b/arch/powerpc/include/asm/kvm_book3s.h
@@ -58,16 +58,18 @@ struct hpte_cache {
 	struct hlist_node list_pte_long;
 	struct hlist_node list_vpte;
 	struct hlist_node list_vpte_long;
+#ifdef CONFIG_PPC_BOOK3S_64
+	struct hlist_node list_vpte_64k;
+#endif
 	struct rcu_head rcu_head;
 	u64 host_vpn;
 	u64 pfn;
 	ulong slot;
 	struct kvmppc_pte pte;
+	int pagesize;
 };
 
 struct kvmppc_vcpu_book3s {
-	struct kvm_vcpu vcpu;
-	struct kvmppc_book3s_shadow_vcpu *shadow_vcpu;
 	struct kvmppc_sid_map sid_map[SID_MAP_NUM];
 	struct {
 		u64 esid;
@@ -99,6 +101,9 @@ struct kvmppc_vcpu_book3s {
 	struct hlist_head hpte_hash_pte_long[HPTEG_HASH_NUM_PTE_LONG];
 	struct hlist_head hpte_hash_vpte[HPTEG_HASH_NUM_VPTE];
 	struct hlist_head hpte_hash_vpte_long[HPTEG_HASH_NUM_VPTE_LONG];
+#ifdef CONFIG_PPC_BOOK3S_64
+	struct hlist_head hpte_hash_vpte_64k[HPTEG_HASH_NUM_VPTE_64K];
+#endif
 	int hpte_cache_count;
 	spinlock_t mmu_lock;
 };
@@ -107,8 +112,9 @@ struct kvmppc_vcpu_book3s {
 #define CONTEXT_GUEST		1
 #define CONTEXT_GUEST_END	2
 
-#define VSID_REAL	0x0fffffffffc00000ULL
-#define VSID_BAT	0x0fffffffffb00000ULL
+#define VSID_REAL	0x07ffffffffc00000ULL
+#define VSID_BAT	0x07ffffffffb00000ULL
+#define VSID_64K	0x0800000000000000ULL
 #define VSID_1T		0x1000000000000000ULL
 #define VSID_REAL_DR	0x2000000000000000ULL
 #define VSID_REAL_IR	0x4000000000000000ULL
@@ -118,11 +124,12 @@ extern void kvmppc_mmu_pte_flush(struct kvm_vcpu *vcpu, ulong ea, ulong ea_mask)
 extern void kvmppc_mmu_pte_vflush(struct kvm_vcpu *vcpu, u64 vp, u64 vp_mask);
 extern void kvmppc_mmu_pte_pflush(struct kvm_vcpu *vcpu, ulong pa_start, ulong pa_end);
 extern void kvmppc_set_msr(struct kvm_vcpu *vcpu, u64 new_msr);
-extern void kvmppc_set_pvr(struct kvm_vcpu *vcpu, u32 pvr);
 extern void kvmppc_mmu_book3s_64_init(struct kvm_vcpu *vcpu);
 extern void kvmppc_mmu_book3s_32_init(struct kvm_vcpu *vcpu);
 extern void kvmppc_mmu_book3s_hv_init(struct kvm_vcpu *vcpu);
-extern int kvmppc_mmu_map_page(struct kvm_vcpu *vcpu, struct kvmppc_pte *pte);
+extern int kvmppc_mmu_map_page(struct kvm_vcpu *vcpu, struct kvmppc_pte *pte,
+			       bool iswrite);
+extern void kvmppc_mmu_unmap_page(struct kvm_vcpu *vcpu, struct kvmppc_pte *pte);
 extern int kvmppc_mmu_map_segment(struct kvm_vcpu *vcpu, ulong eaddr);
 extern void kvmppc_mmu_flush_segment(struct kvm_vcpu *vcpu, ulong eaddr, ulong seg_size);
 extern void kvmppc_mmu_flush_segments(struct kvm_vcpu *vcpu);
@@ -134,6 +141,7 @@ extern long kvmppc_hv_find_lock_hpte(struct kvm *kvm, gva_t eaddr,
 
 extern void kvmppc_mmu_hpte_cache_map(struct kvm_vcpu *vcpu, struct hpte_cache *pte);
 extern struct hpte_cache *kvmppc_mmu_hpte_cache_next(struct kvm_vcpu *vcpu);
+extern void kvmppc_mmu_hpte_cache_free(struct hpte_cache *pte);
 extern void kvmppc_mmu_hpte_destroy(struct kvm_vcpu *vcpu);
 extern int kvmppc_mmu_hpte_init(struct kvm_vcpu *vcpu);
 extern void kvmppc_mmu_invalidate_pte(struct kvm_vcpu *vcpu, struct hpte_cache *pte);
@@ -151,7 +159,8 @@ extern void kvmppc_set_bat(struct kvm_vcpu *vcpu, struct kvmppc_bat *bat,
 			   bool upper, u32 val);
 extern void kvmppc_giveup_ext(struct kvm_vcpu *vcpu, ulong msr);
 extern int kvmppc_emulate_paired_single(struct kvm_run *run, struct kvm_vcpu *vcpu);
-extern pfn_t kvmppc_gfn_to_pfn(struct kvm_vcpu *vcpu, gfn_t gfn);
+extern pfn_t kvmppc_gfn_to_pfn(struct kvm_vcpu *vcpu, gfn_t gfn, bool writing,
+			bool *writable);
 extern void kvmppc_add_revmap_chain(struct kvm *kvm, struct revmap_entry *rev,
 			unsigned long *rmap, long pte_index, int realmode);
 extern void kvmppc_invalidate_hpte(struct kvm *kvm, unsigned long *hptep,
@@ -172,6 +181,8 @@ extern long kvmppc_do_h_remove(struct kvm *kvm, unsigned long flags,
 			unsigned long *hpret);
 extern long kvmppc_hv_get_dirty_log(struct kvm *kvm,
 			struct kvm_memory_slot *memslot, unsigned long *map);
+extern void kvmppc_update_lpcr(struct kvm *kvm, unsigned long lpcr,
+			unsigned long mask);
 
 extern void kvmppc_entry_trampoline(void);
 extern void kvmppc_hv_entry_trampoline(void);
@@ -184,11 +195,9 @@ extern int kvmppc_h_pr(struct kvm_vcpu *vcpu, unsigned long cmd);
 
 static inline struct kvmppc_vcpu_book3s *to_book3s(struct kvm_vcpu *vcpu)
 {
-	return container_of(vcpu, struct kvmppc_vcpu_book3s, vcpu);
+	return vcpu->arch.book3s;
 }
 
-extern void kvm_return_point(void);
-
 /* Also add subarch specific defines */
 
 #ifdef CONFIG_KVM_BOOK3S_32_HANDLER
@@ -198,203 +207,6 @@ extern void kvm_return_point(void);
 #include <asm/kvm_book3s_64.h>
 #endif
 
-#ifdef CONFIG_KVM_BOOK3S_PR
-
-static inline unsigned long kvmppc_interrupt_offset(struct kvm_vcpu *vcpu)
-{
-	return to_book3s(vcpu)->hior;
-}
-
-static inline void kvmppc_update_int_pending(struct kvm_vcpu *vcpu,
-			unsigned long pending_now, unsigned long old_pending)
-{
-	if (pending_now)
-		vcpu->arch.shared->int_pending = 1;
-	else if (old_pending)
-		vcpu->arch.shared->int_pending = 0;
-}
-
-static inline void kvmppc_set_gpr(struct kvm_vcpu *vcpu, int num, ulong val)
-{
-	if ( num < 14 ) {
-		struct kvmppc_book3s_shadow_vcpu *svcpu = svcpu_get(vcpu);
-		svcpu->gpr[num] = val;
-		svcpu_put(svcpu);
-		to_book3s(vcpu)->shadow_vcpu->gpr[num] = val;
-	} else
-		vcpu->arch.gpr[num] = val;
-}
-
-static inline ulong kvmppc_get_gpr(struct kvm_vcpu *vcpu, int num)
-{
-	if ( num < 14 ) {
-		struct kvmppc_book3s_shadow_vcpu *svcpu = svcpu_get(vcpu);
-		ulong r = svcpu->gpr[num];
-		svcpu_put(svcpu);
-		return r;
-	} else
-		return vcpu->arch.gpr[num];
-}
-
-static inline void kvmppc_set_cr(struct kvm_vcpu *vcpu, u32 val)
-{
-	struct kvmppc_book3s_shadow_vcpu *svcpu = svcpu_get(vcpu);
-	svcpu->cr = val;
-	svcpu_put(svcpu);
-	to_book3s(vcpu)->shadow_vcpu->cr = val;
-}
-
-static inline u32 kvmppc_get_cr(struct kvm_vcpu *vcpu)
-{
-	struct kvmppc_book3s_shadow_vcpu *svcpu = svcpu_get(vcpu);
-	u32 r;
-	r = svcpu->cr;
-	svcpu_put(svcpu);
-	return r;
-}
-
-static inline void kvmppc_set_xer(struct kvm_vcpu *vcpu, u32 val)
-{
-	struct kvmppc_book3s_shadow_vcpu *svcpu = svcpu_get(vcpu);
-	svcpu->xer = val;
-	to_book3s(vcpu)->shadow_vcpu->xer = val;
-	svcpu_put(svcpu);
-}
-
-static inline u32 kvmppc_get_xer(struct kvm_vcpu *vcpu)
-{
-	struct kvmppc_book3s_shadow_vcpu *svcpu = svcpu_get(vcpu);
-	u32 r;
-	r = svcpu->xer;
-	svcpu_put(svcpu);
-	return r;
-}
-
-static inline void kvmppc_set_ctr(struct kvm_vcpu *vcpu, ulong val)
-{
-	struct kvmppc_book3s_shadow_vcpu *svcpu = svcpu_get(vcpu);
-	svcpu->ctr = val;
-	svcpu_put(svcpu);
-}
-
-static inline ulong kvmppc_get_ctr(struct kvm_vcpu *vcpu)
-{
-	struct kvmppc_book3s_shadow_vcpu *svcpu = svcpu_get(vcpu);
-	ulong r;
-	r = svcpu->ctr;
-	svcpu_put(svcpu);
-	return r;
-}
-
-static inline void kvmppc_set_lr(struct kvm_vcpu *vcpu, ulong val)
-{
-	struct kvmppc_book3s_shadow_vcpu *svcpu = svcpu_get(vcpu);
-	svcpu->lr = val;
-	svcpu_put(svcpu);
-}
-
-static inline ulong kvmppc_get_lr(struct kvm_vcpu *vcpu)
-{
-	struct kvmppc_book3s_shadow_vcpu *svcpu = svcpu_get(vcpu);
-	ulong r;
-	r = svcpu->lr;
-	svcpu_put(svcpu);
-	return r;
-}
-
-static inline void kvmppc_set_pc(struct kvm_vcpu *vcpu, ulong val)
-{
-	struct kvmppc_book3s_shadow_vcpu *svcpu = svcpu_get(vcpu);
-	svcpu->pc = val;
-	svcpu_put(svcpu);
-}
-
-static inline ulong kvmppc_get_pc(struct kvm_vcpu *vcpu)
-{
-	struct kvmppc_book3s_shadow_vcpu *svcpu = svcpu_get(vcpu);
-	ulong r;
-	r = svcpu->pc;
-	svcpu_put(svcpu);
-	return r;
-}
-
-static inline u32 kvmppc_get_last_inst(struct kvm_vcpu *vcpu)
-{
-	ulong pc = kvmppc_get_pc(vcpu);
-	struct kvmppc_book3s_shadow_vcpu *svcpu = svcpu_get(vcpu);
-	u32 r;
-
-	/* Load the instruction manually if it failed to do so in the
-	 * exit path */
-	if (svcpu->last_inst == KVM_INST_FETCH_FAILED)
-		kvmppc_ld(vcpu, &pc, sizeof(u32), &svcpu->last_inst, false);
-
-	r = svcpu->last_inst;
-	svcpu_put(svcpu);
-	return r;
-}
-
-/*
- * Like kvmppc_get_last_inst(), but for fetching a sc instruction.
- * Because the sc instruction sets SRR0 to point to the following
- * instruction, we have to fetch from pc - 4.
- */
-static inline u32 kvmppc_get_last_sc(struct kvm_vcpu *vcpu)
-{
-	ulong pc = kvmppc_get_pc(vcpu) - 4;
-	struct kvmppc_book3s_shadow_vcpu *svcpu = svcpu_get(vcpu);
-	u32 r;
-
-	/* Load the instruction manually if it failed to do so in the
-	 * exit path */
-	if (svcpu->last_inst == KVM_INST_FETCH_FAILED)
-		kvmppc_ld(vcpu, &pc, sizeof(u32), &svcpu->last_inst, false);
-
-	r = svcpu->last_inst;
-	svcpu_put(svcpu);
-	return r;
-}
-
-static inline ulong kvmppc_get_fault_dar(struct kvm_vcpu *vcpu)
-{
-	struct kvmppc_book3s_shadow_vcpu *svcpu = svcpu_get(vcpu);
-	ulong r;
-	r = svcpu->fault_dar;
-	svcpu_put(svcpu);
-	return r;
-}
-
-static inline bool kvmppc_critical_section(struct kvm_vcpu *vcpu)
-{
-	ulong crit_raw = vcpu->arch.shared->critical;
-	ulong crit_r1 = kvmppc_get_gpr(vcpu, 1);
-	bool crit;
-
-	/* Truncate crit indicators in 32 bit mode */
-	if (!(vcpu->arch.shared->msr & MSR_SF)) {
-		crit_raw &= 0xffffffff;
-		crit_r1 &= 0xffffffff;
-	}
-
-	/* Critical section when crit == r1 */
-	crit = (crit_raw == crit_r1);
-	/* ... and we're in supervisor mode */
-	crit = crit && !(vcpu->arch.shared->msr & MSR_PR);
-
-	return crit;
-}
-#else /* CONFIG_KVM_BOOK3S_PR */
-
-static inline unsigned long kvmppc_interrupt_offset(struct kvm_vcpu *vcpu)
-{
-	return 0;
-}
-
-static inline void kvmppc_update_int_pending(struct kvm_vcpu *vcpu,
-			unsigned long pending_now, unsigned long old_pending)
-{
-}
-
 static inline void kvmppc_set_gpr(struct kvm_vcpu *vcpu, int num, ulong val)
 {
 	vcpu->arch.gpr[num] = val;
@@ -489,12 +301,6 @@ static inline ulong kvmppc_get_fault_dar(struct kvm_vcpu *vcpu)
 	return vcpu->arch.fault_dar;
 }
 
-static inline bool kvmppc_critical_section(struct kvm_vcpu *vcpu)
-{
-	return false;
-}
-#endif
-
 /* Magic register values loaded into r3 and r4 before the 'sc' assembly
  * instruction for the OSI hypercalls */
 #define OSI_SC_MAGIC_R3			0x113724FA
diff --git a/arch/powerpc/include/asm/kvm_book3s_32.h b/arch/powerpc/include/asm/kvm_book3s_32.h
index ce0ef6ce8f86..c720e0b3238d 100644
--- a/arch/powerpc/include/asm/kvm_book3s_32.h
+++ b/arch/powerpc/include/asm/kvm_book3s_32.h
@@ -22,7 +22,7 @@
 
 static inline struct kvmppc_book3s_shadow_vcpu *svcpu_get(struct kvm_vcpu *vcpu)
 {
-	return to_book3s(vcpu)->shadow_vcpu;
+	return vcpu->arch.shadow_vcpu;
 }
 
 static inline void svcpu_put(struct kvmppc_book3s_shadow_vcpu *svcpu)
diff --git a/arch/powerpc/include/asm/kvm_book3s_64.h b/arch/powerpc/include/asm/kvm_book3s_64.h
index 86d638a3b359..bf0fa8b0a883 100644
--- a/arch/powerpc/include/asm/kvm_book3s_64.h
+++ b/arch/powerpc/include/asm/kvm_book3s_64.h
@@ -20,7 +20,7 @@
 #ifndef __ASM_KVM_BOOK3S_64_H__
 #define __ASM_KVM_BOOK3S_64_H__
 
-#ifdef CONFIG_KVM_BOOK3S_PR
+#ifdef CONFIG_KVM_BOOK3S_PR_POSSIBLE
 static inline struct kvmppc_book3s_shadow_vcpu *svcpu_get(struct kvm_vcpu *vcpu)
 {
 	preempt_disable();
@@ -35,7 +35,7 @@ static inline void svcpu_put(struct kvmppc_book3s_shadow_vcpu *svcpu)
 
 #define SPAPR_TCE_SHIFT		12
 
-#ifdef CONFIG_KVM_BOOK3S_64_HV
+#ifdef CONFIG_KVM_BOOK3S_HV_POSSIBLE
 #define KVM_DEFAULT_HPT_ORDER	24	/* 16MB HPT by default */
 extern unsigned long kvm_rma_pages;
 #endif
@@ -278,7 +278,7 @@ static inline int is_vrma_hpte(unsigned long hpte_v)
 		(HPTE_V_1TB_SEG | (VRMA_VSID << (40 - 16)));
 }
 
-#ifdef CONFIG_KVM_BOOK3S_64_HV
+#ifdef CONFIG_KVM_BOOK3S_HV_POSSIBLE
 /*
  * Note modification of an HPTE; set the HPTE modified bit
  * if anyone is interested.
@@ -289,6 +289,6 @@ static inline void note_hpte_modification(struct kvm *kvm,
 	if (atomic_read(&kvm->arch.hpte_mod_interest))
 		rev->guest_rpte |= HPTE_GR_MODIFIED;
 }
-#endif /* CONFIG_KVM_BOOK3S_64_HV */
+#endif /* CONFIG_KVM_BOOK3S_HV_POSSIBLE */
 
 #endif /* __ASM_KVM_BOOK3S_64_H__ */
diff --git a/arch/powerpc/include/asm/kvm_book3s_asm.h b/arch/powerpc/include/asm/kvm_book3s_asm.h
index 9039d3c97eec..0bd9348a4db9 100644
--- a/arch/powerpc/include/asm/kvm_book3s_asm.h
+++ b/arch/powerpc/include/asm/kvm_book3s_asm.h
@@ -83,7 +83,7 @@ struct kvmppc_host_state {
 	u8 restore_hid5;
 	u8 napping;
 
-#ifdef CONFIG_KVM_BOOK3S_64_HV
+#ifdef CONFIG_KVM_BOOK3S_HV_POSSIBLE
 	u8 hwthread_req;
 	u8 hwthread_state;
 	u8 host_ipi;
@@ -101,6 +101,7 @@ struct kvmppc_host_state {
 #endif
 #ifdef CONFIG_PPC_BOOK3S_64
 	u64 cfar;
+	u64 ppr;
 #endif
 };
 
@@ -108,14 +109,14 @@ struct kvmppc_book3s_shadow_vcpu {
 	ulong gpr[14];
 	u32 cr;
 	u32 xer;
-
-	u32 fault_dsisr;
-	u32 last_inst;
 	ulong ctr;
 	ulong lr;
 	ulong pc;
+
 	ulong shadow_srr1;
 	ulong fault_dar;
+	u32 fault_dsisr;
+	u32 last_inst;
 
 #ifdef CONFIG_PPC_BOOK3S_32
 	u32     sr[16];			/* Guest SRs */
diff --git a/arch/powerpc/include/asm/kvm_booke.h b/arch/powerpc/include/asm/kvm_booke.h
index d3c1eb34c986..dd8f61510dfd 100644
--- a/arch/powerpc/include/asm/kvm_booke.h
+++ b/arch/powerpc/include/asm/kvm_booke.h
@@ -26,7 +26,12 @@
 /* LPIDs we support with this build -- runtime limit may be lower */
 #define KVMPPC_NR_LPIDS                        64
 
-#define KVMPPC_INST_EHPRIV	0x7c00021c
+#define KVMPPC_INST_EHPRIV		0x7c00021c
+#define EHPRIV_OC_SHIFT			11
+/* "ehpriv 1" : ehpriv with OC = 1 is used for debug emulation */
+#define EHPRIV_OC_DEBUG			1
+#define KVMPPC_INST_EHPRIV_DEBUG	(KVMPPC_INST_EHPRIV | \
+					 (EHPRIV_OC_DEBUG << EHPRIV_OC_SHIFT))
 
 static inline void kvmppc_set_gpr(struct kvm_vcpu *vcpu, int num, ulong val)
 {
diff --git a/arch/powerpc/include/asm/kvm_host.h b/arch/powerpc/include/asm/kvm_host.h
index 33283532e9d8..237d1d25b448 100644
--- a/arch/powerpc/include/asm/kvm_host.h
+++ b/arch/powerpc/include/asm/kvm_host.h
@@ -63,20 +63,17 @@ extern void kvm_set_spte_hva(struct kvm *kvm, unsigned long hva, pte_t pte);
 
 #endif
 
-/* We don't currently support large pages. */
-#define KVM_HPAGE_GFN_SHIFT(x)	0
-#define KVM_NR_PAGE_SIZES	1
-#define KVM_PAGES_PER_HPAGE(x)	(1UL<<31)
-
 #define HPTEG_CACHE_NUM			(1 << 15)
 #define HPTEG_HASH_BITS_PTE		13
 #define HPTEG_HASH_BITS_PTE_LONG	12
 #define HPTEG_HASH_BITS_VPTE		13
 #define HPTEG_HASH_BITS_VPTE_LONG	5
+#define HPTEG_HASH_BITS_VPTE_64K	11
 #define HPTEG_HASH_NUM_PTE		(1 << HPTEG_HASH_BITS_PTE)
 #define HPTEG_HASH_NUM_PTE_LONG		(1 << HPTEG_HASH_BITS_PTE_LONG)
 #define HPTEG_HASH_NUM_VPTE		(1 << HPTEG_HASH_BITS_VPTE)
 #define HPTEG_HASH_NUM_VPTE_LONG	(1 << HPTEG_HASH_BITS_VPTE_LONG)
+#define HPTEG_HASH_NUM_VPTE_64K		(1 << HPTEG_HASH_BITS_VPTE_64K)
 
 /* Physical Address Mask - allowed range of real mode RAM access */
 #define KVM_PAM			0x0fffffffffffffffULL
@@ -89,6 +86,9 @@ struct lppaca;
 struct slb_shadow;
 struct dtl_entry;
 
+struct kvmppc_vcpu_book3s;
+struct kvmppc_book3s_shadow_vcpu;
+
 struct kvm_vm_stat {
 	u32 remote_tlb_flush;
 };
@@ -224,15 +224,15 @@ struct revmap_entry {
 #define KVMPPC_GOT_PAGE		0x80
 
 struct kvm_arch_memory_slot {
-#ifdef CONFIG_KVM_BOOK3S_64_HV
+#ifdef CONFIG_KVM_BOOK3S_HV_POSSIBLE
 	unsigned long *rmap;
 	unsigned long *slot_phys;
-#endif /* CONFIG_KVM_BOOK3S_64_HV */
+#endif /* CONFIG_KVM_BOOK3S_HV_POSSIBLE */
 };
 
 struct kvm_arch {
 	unsigned int lpid;
-#ifdef CONFIG_KVM_BOOK3S_64_HV
+#ifdef CONFIG_KVM_BOOK3S_HV_POSSIBLE
 	unsigned long hpt_virt;
 	struct revmap_entry *revmap;
 	unsigned int host_lpid;
@@ -256,7 +256,10 @@ struct kvm_arch {
 	cpumask_t need_tlb_flush;
 	struct kvmppc_vcore *vcores[KVM_MAX_VCORES];
 	int hpt_cma_alloc;
-#endif /* CONFIG_KVM_BOOK3S_64_HV */
+#endif /* CONFIG_KVM_BOOK3S_HV_POSSIBLE */
+#ifdef CONFIG_KVM_BOOK3S_PR_POSSIBLE
+	struct mutex hpt_mutex;
+#endif
 #ifdef CONFIG_PPC_BOOK3S_64
 	struct list_head spapr_tce_tables;
 	struct list_head rtas_tokens;
@@ -267,6 +270,7 @@ struct kvm_arch {
 #ifdef CONFIG_KVM_XICS
 	struct kvmppc_xics *xics;
 #endif
+	struct kvmppc_ops *kvm_ops;
 };
 
 /*
@@ -294,6 +298,10 @@ struct kvmppc_vcore {
 	u64 stolen_tb;
 	u64 preempt_tb;
 	struct kvm_vcpu *runner;
+	u64 tb_offset;		/* guest timebase - host timebase */
+	ulong lpcr;
+	u32 arch_compat;
+	ulong pcr;
 };
 
 #define VCORE_ENTRY_COUNT(vc)	((vc)->entry_exit_count & 0xff)
@@ -328,6 +336,7 @@ struct kvmppc_pte {
 	bool may_read		: 1;
 	bool may_write		: 1;
 	bool may_execute	: 1;
+	u8 page_size;		/* MMU_PAGE_xxx */
 };
 
 struct kvmppc_mmu {
@@ -340,7 +349,8 @@ struct kvmppc_mmu {
 	/* book3s */
 	void (*mtsrin)(struct kvm_vcpu *vcpu, u32 srnum, ulong value);
 	u32  (*mfsrin)(struct kvm_vcpu *vcpu, u32 srnum);
-	int  (*xlate)(struct kvm_vcpu *vcpu, gva_t eaddr, struct kvmppc_pte *pte, bool data);
+	int  (*xlate)(struct kvm_vcpu *vcpu, gva_t eaddr,
+		      struct kvmppc_pte *pte, bool data, bool iswrite);
 	void (*reset_msr)(struct kvm_vcpu *vcpu);
 	void (*tlbie)(struct kvm_vcpu *vcpu, ulong addr, bool large);
 	int  (*esid_to_vsid)(struct kvm_vcpu *vcpu, ulong esid, u64 *vsid);
@@ -360,6 +370,7 @@ struct kvmppc_slb {
 	bool large	: 1;	/* PTEs are 16MB */
 	bool tb		: 1;	/* 1TB segment */
 	bool class	: 1;
+	u8 base_page_size;	/* MMU_PAGE_xxx */
 };
 
 # ifdef CONFIG_PPC_FSL_BOOK3E
@@ -377,17 +388,6 @@ struct kvmppc_slb {
 #define KVMPPC_EPR_USER		1 /* exit to userspace to fill EPR */
 #define KVMPPC_EPR_KERNEL	2 /* in-kernel irqchip */
 
-struct kvmppc_booke_debug_reg {
-	u32 dbcr0;
-	u32 dbcr1;
-	u32 dbcr2;
-#ifdef CONFIG_KVM_E500MC
-	u32 dbcr4;
-#endif
-	u64 iac[KVMPPC_BOOKE_MAX_IAC];
-	u64 dac[KVMPPC_BOOKE_MAX_DAC];
-};
-
 #define KVMPPC_IRQ_DEFAULT	0
 #define KVMPPC_IRQ_MPIC		1
 #define KVMPPC_IRQ_XICS		2
@@ -402,6 +402,10 @@ struct kvm_vcpu_arch {
 	int slb_max;		/* 1 + index of last valid entry in slb[] */
 	int slb_nr;		/* total number of entries in SLB */
 	struct kvmppc_mmu mmu;
+	struct kvmppc_vcpu_book3s *book3s;
+#endif
+#ifdef CONFIG_PPC_BOOK3S_32
+	struct kvmppc_book3s_shadow_vcpu *shadow_vcpu;
 #endif
 
 	ulong gpr[32];
@@ -463,6 +467,8 @@ struct kvm_vcpu_arch {
 	u32 ctrl;
 	ulong dabr;
 	ulong cfar;
+	ulong ppr;
+	ulong shadow_srr1;
 #endif
 	u32 vrsave; /* also USPRG0 */
 	u32 mmucr;
@@ -498,6 +504,8 @@ struct kvm_vcpu_arch {
 
 	u64 mmcr[3];
 	u32 pmc[8];
+	u64 siar;
+	u64 sdar;
 
 #ifdef CONFIG_KVM_EXIT_TIMING
 	struct mutex exit_timing_lock;
@@ -531,7 +539,10 @@ struct kvm_vcpu_arch {
 	u32 eptcfg;
 	u32 epr;
 	u32 crit_save;
-	struct kvmppc_booke_debug_reg dbg_reg;
+	/* guest debug registers*/
+	struct debug_reg dbg_reg;
+	/* hardware visible debug registers when in guest state */
+	struct debug_reg shadow_dbg_reg;
 #endif
 	gpa_t paddr_accessed;
 	gva_t vaddr_accessed;
@@ -582,7 +593,7 @@ struct kvm_vcpu_arch {
 	struct kvmppc_icp *icp; /* XICS presentation controller */
 #endif
 
-#ifdef CONFIG_KVM_BOOK3S_64_HV
+#ifdef CONFIG_KVM_BOOK3S_HV_POSSIBLE
 	struct kvm_vcpu_arch_shared shregs;
 
 	unsigned long pgfault_addr;
diff --git a/arch/powerpc/include/asm/kvm_ppc.h b/arch/powerpc/include/asm/kvm_ppc.h
index b15554a26c20..c8317fbf92c4 100644
--- a/arch/powerpc/include/asm/kvm_ppc.h
+++ b/arch/powerpc/include/asm/kvm_ppc.h
@@ -106,13 +106,6 @@ extern void kvmppc_core_queue_external(struct kvm_vcpu *vcpu,
                                        struct kvm_interrupt *irq);
 extern void kvmppc_core_dequeue_external(struct kvm_vcpu *vcpu);
 extern void kvmppc_core_flush_tlb(struct kvm_vcpu *vcpu);
-
-extern int kvmppc_core_emulate_op(struct kvm_run *run, struct kvm_vcpu *vcpu,
-                                  unsigned int op, int *advance);
-extern int kvmppc_core_emulate_mtspr(struct kvm_vcpu *vcpu, int sprn,
-				     ulong val);
-extern int kvmppc_core_emulate_mfspr(struct kvm_vcpu *vcpu, int sprn,
-				     ulong *val);
 extern int kvmppc_core_check_requests(struct kvm_vcpu *vcpu);
 
 extern int kvmppc_booke_init(void);
@@ -135,17 +128,17 @@ extern long kvm_vm_ioctl_create_spapr_tce(struct kvm *kvm,
 				struct kvm_create_spapr_tce *args);
 extern long kvmppc_h_put_tce(struct kvm_vcpu *vcpu, unsigned long liobn,
 			     unsigned long ioba, unsigned long tce);
-extern long kvm_vm_ioctl_allocate_rma(struct kvm *kvm,
-				struct kvm_allocate_rma *rma);
 extern struct kvm_rma_info *kvm_alloc_rma(void);
 extern void kvm_release_rma(struct kvm_rma_info *ri);
 extern struct page *kvm_alloc_hpt(unsigned long nr_pages);
 extern void kvm_release_hpt(struct page *page, unsigned long nr_pages);
 extern int kvmppc_core_init_vm(struct kvm *kvm);
 extern void kvmppc_core_destroy_vm(struct kvm *kvm);
-extern void kvmppc_core_free_memslot(struct kvm_memory_slot *free,
+extern void kvmppc_core_free_memslot(struct kvm *kvm,
+				     struct kvm_memory_slot *free,
 				     struct kvm_memory_slot *dont);
-extern int kvmppc_core_create_memslot(struct kvm_memory_slot *slot,
+extern int kvmppc_core_create_memslot(struct kvm *kvm,
+				      struct kvm_memory_slot *slot,
 				      unsigned long npages);
 extern int kvmppc_core_prepare_memory_region(struct kvm *kvm,
 				struct kvm_memory_slot *memslot,
@@ -177,6 +170,72 @@ extern int kvmppc_xics_get_xive(struct kvm *kvm, u32 irq, u32 *server,
 extern int kvmppc_xics_int_on(struct kvm *kvm, u32 irq);
 extern int kvmppc_xics_int_off(struct kvm *kvm, u32 irq);
 
+union kvmppc_one_reg {
+	u32	wval;
+	u64	dval;
+	vector128 vval;
+	u64	vsxval[2];
+	struct {
+		u64	addr;
+		u64	length;
+	}	vpaval;
+};
+
+struct kvmppc_ops {
+	struct module *owner;
+	int (*get_sregs)(struct kvm_vcpu *vcpu, struct kvm_sregs *sregs);
+	int (*set_sregs)(struct kvm_vcpu *vcpu, struct kvm_sregs *sregs);
+	int (*get_one_reg)(struct kvm_vcpu *vcpu, u64 id,
+			   union kvmppc_one_reg *val);
+	int (*set_one_reg)(struct kvm_vcpu *vcpu, u64 id,
+			   union kvmppc_one_reg *val);
+	void (*vcpu_load)(struct kvm_vcpu *vcpu, int cpu);
+	void (*vcpu_put)(struct kvm_vcpu *vcpu);
+	void (*set_msr)(struct kvm_vcpu *vcpu, u64 msr);
+	int (*vcpu_run)(struct kvm_run *run, struct kvm_vcpu *vcpu);
+	struct kvm_vcpu *(*vcpu_create)(struct kvm *kvm, unsigned int id);
+	void (*vcpu_free)(struct kvm_vcpu *vcpu);
+	int (*check_requests)(struct kvm_vcpu *vcpu);
+	int (*get_dirty_log)(struct kvm *kvm, struct kvm_dirty_log *log);
+	void (*flush_memslot)(struct kvm *kvm, struct kvm_memory_slot *memslot);
+	int (*prepare_memory_region)(struct kvm *kvm,
+				     struct kvm_memory_slot *memslot,
+				     struct kvm_userspace_memory_region *mem);
+	void (*commit_memory_region)(struct kvm *kvm,
+				     struct kvm_userspace_memory_region *mem,
+				     const struct kvm_memory_slot *old);
+	int (*unmap_hva)(struct kvm *kvm, unsigned long hva);
+	int (*unmap_hva_range)(struct kvm *kvm, unsigned long start,
+			   unsigned long end);
+	int (*age_hva)(struct kvm *kvm, unsigned long hva);
+	int (*test_age_hva)(struct kvm *kvm, unsigned long hva);
+	void (*set_spte_hva)(struct kvm *kvm, unsigned long hva, pte_t pte);
+	void (*mmu_destroy)(struct kvm_vcpu *vcpu);
+	void (*free_memslot)(struct kvm_memory_slot *free,
+			     struct kvm_memory_slot *dont);
+	int (*create_memslot)(struct kvm_memory_slot *slot,
+			      unsigned long npages);
+	int (*init_vm)(struct kvm *kvm);
+	void (*destroy_vm)(struct kvm *kvm);
+	int (*get_smmu_info)(struct kvm *kvm, struct kvm_ppc_smmu_info *info);
+	int (*emulate_op)(struct kvm_run *run, struct kvm_vcpu *vcpu,
+			  unsigned int inst, int *advance);
+	int (*emulate_mtspr)(struct kvm_vcpu *vcpu, int sprn, ulong spr_val);
+	int (*emulate_mfspr)(struct kvm_vcpu *vcpu, int sprn, ulong *spr_val);
+	void (*fast_vcpu_kick)(struct kvm_vcpu *vcpu);
+	long (*arch_vm_ioctl)(struct file *filp, unsigned int ioctl,
+			      unsigned long arg);
+
+};
+
+extern struct kvmppc_ops *kvmppc_hv_ops;
+extern struct kvmppc_ops *kvmppc_pr_ops;
+
+static inline bool is_kvmppc_hv_enabled(struct kvm *kvm)
+{
+	return kvm->arch.kvm_ops == kvmppc_hv_ops;
+}
+
 /*
  * Cuts out inst bits with ordering according to spec.
  * That means the leftmost bit is zero. All given bits are included.
@@ -210,17 +269,6 @@ static inline u32 kvmppc_set_field(u64 inst, int msb, int lsb, int value)
 	return r;
 }
 
-union kvmppc_one_reg {
-	u32	wval;
-	u64	dval;
-	vector128 vval;
-	u64	vsxval[2];
-	struct {
-		u64	addr;
-		u64	length;
-	}	vpaval;
-};
-
 #define one_reg_size(id)	\
 	(1ul << (((id) & KVM_REG_SIZE_MASK) >> KVM_REG_SIZE_SHIFT))
 
@@ -245,10 +293,10 @@ union kvmppc_one_reg {
 	__v;					\
 })
 
-void kvmppc_core_get_sregs(struct kvm_vcpu *vcpu, struct kvm_sregs *sregs);
+int kvmppc_core_get_sregs(struct kvm_vcpu *vcpu, struct kvm_sregs *sregs);
 int kvmppc_core_set_sregs(struct kvm_vcpu *vcpu, struct kvm_sregs *sregs);
 
-void kvmppc_get_sregs_ivor(struct kvm_vcpu *vcpu, struct kvm_sregs *sregs);
+int kvmppc_get_sregs_ivor(struct kvm_vcpu *vcpu, struct kvm_sregs *sregs);
 int kvmppc_set_sregs_ivor(struct kvm_vcpu *vcpu, struct kvm_sregs *sregs);
 
 int kvm_vcpu_ioctl_get_one_reg(struct kvm_vcpu *vcpu, struct kvm_one_reg *reg);
@@ -260,7 +308,7 @@ void kvmppc_set_pid(struct kvm_vcpu *vcpu, u32 pid);
 
 struct openpic;
 
-#ifdef CONFIG_KVM_BOOK3S_64_HV
+#ifdef CONFIG_KVM_BOOK3S_HV_POSSIBLE
 extern void kvm_cma_reserve(void) __init;
 static inline void kvmppc_set_xics_phys(int cpu, unsigned long addr)
 {
@@ -269,10 +317,10 @@ static inline void kvmppc_set_xics_phys(int cpu, unsigned long addr)
 
 static inline u32 kvmppc_get_xics_latch(void)
 {
-	u32 xirr = get_paca()->kvm_hstate.saved_xirr;
+	u32 xirr;
 
+	xirr = get_paca()->kvm_hstate.saved_xirr;
 	get_paca()->kvm_hstate.saved_xirr = 0;
-
 	return xirr;
 }
 
@@ -281,7 +329,10 @@ static inline void kvmppc_set_host_ipi(int cpu, u8 host_ipi)
 	paca[cpu].kvm_hstate.host_ipi = host_ipi;
 }
 
-extern void kvmppc_fast_vcpu_kick(struct kvm_vcpu *vcpu);
+static inline void kvmppc_fast_vcpu_kick(struct kvm_vcpu *vcpu)
+{
+	vcpu->kvm->arch.kvm_ops->fast_vcpu_kick(vcpu);
+}
 
 #else
 static inline void __init kvm_cma_reserve(void)
diff --git a/arch/powerpc/include/asm/paca.h b/arch/powerpc/include/asm/paca.h
index a5954cebbc55..b6ea9e068c13 100644
--- a/arch/powerpc/include/asm/paca.h
+++ b/arch/powerpc/include/asm/paca.h
@@ -166,7 +166,7 @@ struct paca_struct {
 	struct dtl_entry *dtl_curr;	/* pointer corresponding to dtl_ridx */
 
 #ifdef CONFIG_KVM_BOOK3S_HANDLER
-#ifdef CONFIG_KVM_BOOK3S_PR
+#ifdef CONFIG_KVM_BOOK3S_PR_POSSIBLE
 	/* We use this to store guest state in */
 	struct kvmppc_book3s_shadow_vcpu shadow_vcpu;
 #endif
diff --git a/arch/powerpc/include/asm/pgalloc-64.h b/arch/powerpc/include/asm/pgalloc-64.h
index f65e27b09bd3..16cb92d215d2 100644
--- a/arch/powerpc/include/asm/pgalloc-64.h
+++ b/arch/powerpc/include/asm/pgalloc-64.h
@@ -91,7 +91,10 @@ static inline pgtable_t pte_alloc_one(struct mm_struct *mm,
 	if (!pte)
 		return NULL;
 	page = virt_to_page(pte);
-	pgtable_page_ctor(page);
+	if (!pgtable_page_ctor(page)) {
+		__free_page(page);
+		return NULL;
+	}
 	return page;
 }
 
diff --git a/arch/powerpc/include/asm/processor.h b/arch/powerpc/include/asm/processor.h
index 7794b2b04eb2..fc14a38c7ccf 100644
--- a/arch/powerpc/include/asm/processor.h
+++ b/arch/powerpc/include/asm/processor.h
@@ -208,6 +208,7 @@ struct debug_reg {
 
 struct thread_struct {
 	unsigned long	ksp;		/* Kernel stack pointer */
+
 #ifdef CONFIG_PPC64
 	unsigned long	ksp_vsid;
 #endif
@@ -221,6 +222,7 @@ struct thread_struct {
 	void		*pgdir;		/* root of page-table tree */
 	unsigned long	ksp_limit;	/* if ksp <= ksp_limit stack overflow */
 #endif
+	/* Debug Registers */
 	struct debug_reg debug;
 	struct thread_fp_state	fp_state;
 	struct thread_fp_state	*fp_save_area;
diff --git a/arch/powerpc/include/asm/pte-book3e.h b/arch/powerpc/include/asm/pte-book3e.h
index 0156702ba24e..576ad88104cb 100644
--- a/arch/powerpc/include/asm/pte-book3e.h
+++ b/arch/powerpc/include/asm/pte-book3e.h
@@ -40,7 +40,7 @@
 #define _PAGE_U1	0x010000
 #define _PAGE_U0	0x020000
 #define _PAGE_ACCESSED	0x040000
-#define _PAGE_LENDIAN	0x080000
+#define _PAGE_ENDIAN	0x080000
 #define _PAGE_GUARDED	0x100000
 #define _PAGE_COHERENT	0x200000 /* M: enforce memory coherence */
 #define _PAGE_NO_CACHE	0x400000 /* I: cache inhibit */
diff --git a/arch/powerpc/include/asm/reg.h b/arch/powerpc/include/asm/reg.h
index 126f6e98f84d..5c45787d551e 100644
--- a/arch/powerpc/include/asm/reg.h
+++ b/arch/powerpc/include/asm/reg.h
@@ -248,6 +248,7 @@
 #define SPRN_TBRU	0x10D	/* Time Base Read Upper Register (user, R/O) */
 #define SPRN_TBWL	0x11C	/* Time Base Lower Register (super, R/W) */
 #define SPRN_TBWU	0x11D	/* Time Base Upper Register (super, R/W) */
+#define SPRN_TBU40	0x11E	/* Timebase upper 40 bits (hyper, R/W) */
 #define SPRN_SPURR	0x134	/* Scaled PURR */
 #define SPRN_HSPRG0	0x130	/* Hypervisor Scratch 0 */
 #define SPRN_HSPRG1	0x131	/* Hypervisor Scratch 1 */
@@ -288,6 +289,7 @@
 #define   LPCR_ISL	(1ul << (63-2))
 #define   LPCR_VC_SH	(63-2)
 #define   LPCR_DPFD_SH	(63-11)
+#define   LPCR_DPFD	(7ul << LPCR_DPFD_SH)
 #define   LPCR_VRMASD	(0x1ful << (63-16))
 #define   LPCR_VRMA_L	(1ul << (63-12))
 #define   LPCR_VRMA_LP0	(1ul << (63-15))
@@ -304,6 +306,7 @@
 #define     LPCR_PECE2	0x00001000	/* machine check etc can cause exit */
 #define   LPCR_MER	0x00000800	/* Mediated External Exception */
 #define   LPCR_MER_SH	11
+#define   LPCR_TC      0x00000200	/* Translation control */
 #define   LPCR_LPES    0x0000000c
 #define   LPCR_LPES0   0x00000008      /* LPAR Env selector 0 */
 #define   LPCR_LPES1   0x00000004      /* LPAR Env selector 1 */
@@ -316,6 +319,10 @@
 #define   LPID_RSVD	0x3ff		/* Reserved LPID for partn switching */
 #define	SPRN_HMER	0x150	/* Hardware m? error recovery */
 #define	SPRN_HMEER	0x151	/* Hardware m? enable error recovery */
+#define SPRN_PCR	0x152	/* Processor compatibility register */
+#define   PCR_VEC_DIS	(1ul << (63-0))	/* Vec. disable (bit NA since POWER8) */
+#define   PCR_VSX_DIS	(1ul << (63-1))	/* VSX disable (bit NA since POWER8) */
+#define   PCR_ARCH_205	0x2		/* Architecture 2.05 */
 #define	SPRN_HEIR	0x153	/* Hypervisor Emulated Instruction Register */
 #define SPRN_TLBINDEXR	0x154	/* P7 TLB control register */
 #define SPRN_TLBVPNR	0x155	/* P7 TLB control register */
@@ -425,6 +432,7 @@
 #define	 HID4_RMLS2_SH	 (63 - 2)	/* Real mode limit bottom 2 bits */
 #define	 HID4_LPID5_SH	 (63 - 6)	/* partition ID bottom 4 bits */
 #define	 HID4_RMOR_SH	 (63 - 22)	/* real mode offset (16 bits) */
+#define  HID4_RMOR	 (0xFFFFul << HID4_RMOR_SH)
 #define  HID4_LPES1	 (1 << (63-57))	/* LPAR env. sel. bit 1 */
 #define  HID4_RMLS0_SH	 (63 - 58)	/* Real mode limit top bit */
 #define	 HID4_LPID1_SH	 0		/* partition ID top 2 bits */
@@ -1107,6 +1115,13 @@
 #define PVR_BE		0x0070
 #define PVR_PA6T	0x0090
 
+/* "Logical" PVR values defined in PAPR, representing architecture levels */
+#define PVR_ARCH_204	0x0f000001
+#define PVR_ARCH_205	0x0f000002
+#define PVR_ARCH_206	0x0f000003
+#define PVR_ARCH_206p	0x0f100003
+#define PVR_ARCH_207	0x0f000004
+
 /* Macros for setting and retrieving special purpose registers */
 #ifndef __ASSEMBLY__
 #define mfmsr()		({unsigned long rval; \
diff --git a/arch/powerpc/include/asm/thread_info.h b/arch/powerpc/include/asm/thread_info.h
index ba7b1973866e..8fd6cf6dcee8 100644
--- a/arch/powerpc/include/asm/thread_info.h
+++ b/arch/powerpc/include/asm/thread_info.h
@@ -82,8 +82,6 @@ static inline struct thread_info *current_thread_info(void)
 
 #endif /* __ASSEMBLY__ */
 
-#define PREEMPT_ACTIVE		0x10000000
-
 /*
  * thread information flag bit numbers
  */
diff --git a/arch/powerpc/include/uapi/asm/kvm.h b/arch/powerpc/include/uapi/asm/kvm.h
index 0fb1a6e9ff90..6836ec79a830 100644
--- a/arch/powerpc/include/uapi/asm/kvm.h
+++ b/arch/powerpc/include/uapi/asm/kvm.h
@@ -27,6 +27,7 @@
 #define __KVM_HAVE_PPC_SMT
 #define __KVM_HAVE_IRQCHIP
 #define __KVM_HAVE_IRQ_LINE
+#define __KVM_HAVE_GUEST_DEBUG
 
 struct kvm_regs {
 	__u64 pc;
@@ -269,7 +270,24 @@ struct kvm_fpu {
 	__u64 fpr[32];
 };
 
+/*
+ * Defines for h/w breakpoint, watchpoint (read, write or both) and
+ * software breakpoint.
+ * These are used as "type" in KVM_SET_GUEST_DEBUG ioctl and "status"
+ * for KVM_DEBUG_EXIT.
+ */
+#define KVMPPC_DEBUG_NONE		0x0
+#define KVMPPC_DEBUG_BREAKPOINT		(1UL << 1)
+#define KVMPPC_DEBUG_WATCH_WRITE	(1UL << 2)
+#define KVMPPC_DEBUG_WATCH_READ		(1UL << 3)
 struct kvm_debug_exit_arch {
+	__u64 address;
+	/*
+	 * exiting to userspace because of h/w breakpoint, watchpoint
+	 * (read, write or both) and software breakpoint.
+	 */
+	__u32 status;
+	__u32 reserved;
 };
 
 /* for KVM_SET_GUEST_DEBUG */
@@ -281,10 +299,6 @@ struct kvm_guest_debug_arch {
 		 * Type denotes h/w breakpoint, read watchpoint, write
 		 * watchpoint or watchpoint (both read and write).
 		 */
-#define KVMPPC_DEBUG_NONE		0x0
-#define KVMPPC_DEBUG_BREAKPOINT		(1UL << 1)
-#define KVMPPC_DEBUG_WATCH_WRITE	(1UL << 2)
-#define KVMPPC_DEBUG_WATCH_READ		(1UL << 3)
 		__u32 type;
 		__u32 reserved;
 	} bp[16];
@@ -429,6 +443,11 @@ struct kvm_get_htab_header {
 #define KVM_REG_PPC_MMCR0	(KVM_REG_PPC | KVM_REG_SIZE_U64 | 0x10)
 #define KVM_REG_PPC_MMCR1	(KVM_REG_PPC | KVM_REG_SIZE_U64 | 0x11)
 #define KVM_REG_PPC_MMCRA	(KVM_REG_PPC | KVM_REG_SIZE_U64 | 0x12)
+#define KVM_REG_PPC_MMCR2	(KVM_REG_PPC | KVM_REG_SIZE_U64 | 0x13)
+#define KVM_REG_PPC_MMCRS	(KVM_REG_PPC | KVM_REG_SIZE_U64 | 0x14)
+#define KVM_REG_PPC_SIAR	(KVM_REG_PPC | KVM_REG_SIZE_U64 | 0x15)
+#define KVM_REG_PPC_SDAR	(KVM_REG_PPC | KVM_REG_SIZE_U64 | 0x16)
+#define KVM_REG_PPC_SIER	(KVM_REG_PPC | KVM_REG_SIZE_U64 | 0x17)
 
 #define KVM_REG_PPC_PMC1	(KVM_REG_PPC | KVM_REG_SIZE_U32 | 0x18)
 #define KVM_REG_PPC_PMC2	(KVM_REG_PPC | KVM_REG_SIZE_U32 | 0x19)
@@ -499,6 +518,65 @@ struct kvm_get_htab_header {
 #define KVM_REG_PPC_TLB3PS	(KVM_REG_PPC | KVM_REG_SIZE_U32 | 0x9a)
 #define KVM_REG_PPC_EPTCFG	(KVM_REG_PPC | KVM_REG_SIZE_U32 | 0x9b)
 
+/* Timebase offset */
+#define KVM_REG_PPC_TB_OFFSET	(KVM_REG_PPC | KVM_REG_SIZE_U64 | 0x9c)
+
+/* POWER8 registers */
+#define KVM_REG_PPC_SPMC1	(KVM_REG_PPC | KVM_REG_SIZE_U32 | 0x9d)
+#define KVM_REG_PPC_SPMC2	(KVM_REG_PPC | KVM_REG_SIZE_U32 | 0x9e)
+#define KVM_REG_PPC_IAMR	(KVM_REG_PPC | KVM_REG_SIZE_U64 | 0x9f)
+#define KVM_REG_PPC_TFHAR	(KVM_REG_PPC | KVM_REG_SIZE_U64 | 0xa0)
+#define KVM_REG_PPC_TFIAR	(KVM_REG_PPC | KVM_REG_SIZE_U64 | 0xa1)
+#define KVM_REG_PPC_TEXASR	(KVM_REG_PPC | KVM_REG_SIZE_U64 | 0xa2)
+#define KVM_REG_PPC_FSCR	(KVM_REG_PPC | KVM_REG_SIZE_U64 | 0xa3)
+#define KVM_REG_PPC_PSPB	(KVM_REG_PPC | KVM_REG_SIZE_U32 | 0xa4)
+#define KVM_REG_PPC_EBBHR	(KVM_REG_PPC | KVM_REG_SIZE_U64 | 0xa5)
+#define KVM_REG_PPC_EBBRR	(KVM_REG_PPC | KVM_REG_SIZE_U64 | 0xa6)
+#define KVM_REG_PPC_BESCR	(KVM_REG_PPC | KVM_REG_SIZE_U64 | 0xa7)
+#define KVM_REG_PPC_TAR		(KVM_REG_PPC | KVM_REG_SIZE_U64 | 0xa8)
+#define KVM_REG_PPC_DPDES	(KVM_REG_PPC | KVM_REG_SIZE_U64 | 0xa9)
+#define KVM_REG_PPC_DAWR	(KVM_REG_PPC | KVM_REG_SIZE_U64 | 0xaa)
+#define KVM_REG_PPC_DAWRX	(KVM_REG_PPC | KVM_REG_SIZE_U64 | 0xab)
+#define KVM_REG_PPC_CIABR	(KVM_REG_PPC | KVM_REG_SIZE_U64 | 0xac)
+#define KVM_REG_PPC_IC		(KVM_REG_PPC | KVM_REG_SIZE_U64 | 0xad)
+#define KVM_REG_PPC_VTB		(KVM_REG_PPC | KVM_REG_SIZE_U64 | 0xae)
+#define KVM_REG_PPC_CSIGR	(KVM_REG_PPC | KVM_REG_SIZE_U64 | 0xaf)
+#define KVM_REG_PPC_TACR	(KVM_REG_PPC | KVM_REG_SIZE_U64 | 0xb0)
+#define KVM_REG_PPC_TCSCR	(KVM_REG_PPC | KVM_REG_SIZE_U64 | 0xb1)
+#define KVM_REG_PPC_PID		(KVM_REG_PPC | KVM_REG_SIZE_U64 | 0xb2)
+#define KVM_REG_PPC_ACOP	(KVM_REG_PPC | KVM_REG_SIZE_U64 | 0xb3)
+
+#define KVM_REG_PPC_VRSAVE	(KVM_REG_PPC | KVM_REG_SIZE_U32 | 0xb4)
+#define KVM_REG_PPC_LPCR	(KVM_REG_PPC | KVM_REG_SIZE_U32 | 0xb5)
+#define KVM_REG_PPC_PPR		(KVM_REG_PPC | KVM_REG_SIZE_U64 | 0xb6)
+
+/* Architecture compatibility level */
+#define KVM_REG_PPC_ARCH_COMPAT	(KVM_REG_PPC | KVM_REG_SIZE_U32 | 0xb7)
+
+/* Transactional Memory checkpointed state:
+ * This is all GPRs, all VSX regs and a subset of SPRs
+ */
+#define KVM_REG_PPC_TM		(KVM_REG_PPC | 0x80000000)
+/* TM GPRs */
+#define KVM_REG_PPC_TM_GPR0	(KVM_REG_PPC_TM | KVM_REG_SIZE_U64 | 0)
+#define KVM_REG_PPC_TM_GPR(n)	(KVM_REG_PPC_TM_GPR0 + (n))
+#define KVM_REG_PPC_TM_GPR31	(KVM_REG_PPC_TM | KVM_REG_SIZE_U64 | 0x1f)
+/* TM VSX */
+#define KVM_REG_PPC_TM_VSR0	(KVM_REG_PPC_TM | KVM_REG_SIZE_U128 | 0x20)
+#define KVM_REG_PPC_TM_VSR(n)	(KVM_REG_PPC_TM_VSR0 + (n))
+#define KVM_REG_PPC_TM_VSR63	(KVM_REG_PPC_TM | KVM_REG_SIZE_U128 | 0x5f)
+/* TM SPRS */
+#define KVM_REG_PPC_TM_CR	(KVM_REG_PPC_TM | KVM_REG_SIZE_U64 | 0x60)
+#define KVM_REG_PPC_TM_LR	(KVM_REG_PPC_TM | KVM_REG_SIZE_U64 | 0x61)
+#define KVM_REG_PPC_TM_CTR	(KVM_REG_PPC_TM | KVM_REG_SIZE_U64 | 0x62)
+#define KVM_REG_PPC_TM_FPSCR	(KVM_REG_PPC_TM | KVM_REG_SIZE_U64 | 0x63)
+#define KVM_REG_PPC_TM_AMR	(KVM_REG_PPC_TM | KVM_REG_SIZE_U64 | 0x64)
+#define KVM_REG_PPC_TM_PPR	(KVM_REG_PPC_TM | KVM_REG_SIZE_U64 | 0x65)
+#define KVM_REG_PPC_TM_VRSAVE	(KVM_REG_PPC_TM | KVM_REG_SIZE_U64 | 0x66)
+#define KVM_REG_PPC_TM_VSCR	(KVM_REG_PPC_TM | KVM_REG_SIZE_U32 | 0x67)
+#define KVM_REG_PPC_TM_DSCR	(KVM_REG_PPC_TM | KVM_REG_SIZE_U64 | 0x68)
+#define KVM_REG_PPC_TM_TAR	(KVM_REG_PPC_TM | KVM_REG_SIZE_U64 | 0x69)
+
 /* PPC64 eXternal Interrupt Controller Specification */
 #define KVM_DEV_XICS_GRP_SOURCES	1	/* 64-bit source attributes */
 
diff --git a/arch/powerpc/kernel/asm-offsets.c b/arch/powerpc/kernel/asm-offsets.c
index e60a3697932c..2ea5cc033ec8 100644
--- a/arch/powerpc/kernel/asm-offsets.c
+++ b/arch/powerpc/kernel/asm-offsets.c
@@ -439,7 +439,7 @@ int main(void)
 	DEFINE(VCPU_LR, offsetof(struct kvm_vcpu, arch.lr));
 	DEFINE(VCPU_CR, offsetof(struct kvm_vcpu, arch.cr));
 	DEFINE(VCPU_PC, offsetof(struct kvm_vcpu, arch.pc));
-#ifdef CONFIG_KVM_BOOK3S_64_HV
+#ifdef CONFIG_KVM_BOOK3S_HV_POSSIBLE
 	DEFINE(VCPU_MSR, offsetof(struct kvm_vcpu, arch.shregs.msr));
 	DEFINE(VCPU_SRR0, offsetof(struct kvm_vcpu, arch.shregs.srr0));
 	DEFINE(VCPU_SRR1, offsetof(struct kvm_vcpu, arch.shregs.srr1));
@@ -470,7 +470,7 @@ int main(void)
 	DEFINE(KVM_LPID, offsetof(struct kvm, arch.lpid));
 
 	/* book3s */
-#ifdef CONFIG_KVM_BOOK3S_64_HV
+#ifdef CONFIG_KVM_BOOK3S_HV_POSSIBLE
 	DEFINE(KVM_SDR1, offsetof(struct kvm, arch.sdr1));
 	DEFINE(KVM_HOST_LPID, offsetof(struct kvm, arch.host_lpid));
 	DEFINE(KVM_HOST_LPCR, offsetof(struct kvm, arch.host_lpcr));
@@ -502,6 +502,8 @@ int main(void)
 	DEFINE(VCPU_PRODDED, offsetof(struct kvm_vcpu, arch.prodded));
 	DEFINE(VCPU_MMCR, offsetof(struct kvm_vcpu, arch.mmcr));
 	DEFINE(VCPU_PMC, offsetof(struct kvm_vcpu, arch.pmc));
+	DEFINE(VCPU_SIAR, offsetof(struct kvm_vcpu, arch.siar));
+	DEFINE(VCPU_SDAR, offsetof(struct kvm_vcpu, arch.sdar));
 	DEFINE(VCPU_SLB, offsetof(struct kvm_vcpu, arch.slb));
 	DEFINE(VCPU_SLB_MAX, offsetof(struct kvm_vcpu, arch.slb_max));
 	DEFINE(VCPU_SLB_NR, offsetof(struct kvm_vcpu, arch.slb_nr));
@@ -511,18 +513,22 @@ int main(void)
 	DEFINE(VCPU_TRAP, offsetof(struct kvm_vcpu, arch.trap));
 	DEFINE(VCPU_PTID, offsetof(struct kvm_vcpu, arch.ptid));
 	DEFINE(VCPU_CFAR, offsetof(struct kvm_vcpu, arch.cfar));
+	DEFINE(VCPU_PPR, offsetof(struct kvm_vcpu, arch.ppr));
+	DEFINE(VCPU_SHADOW_SRR1, offsetof(struct kvm_vcpu, arch.shadow_srr1));
 	DEFINE(VCORE_ENTRY_EXIT, offsetof(struct kvmppc_vcore, entry_exit_count));
 	DEFINE(VCORE_NAP_COUNT, offsetof(struct kvmppc_vcore, nap_count));
 	DEFINE(VCORE_IN_GUEST, offsetof(struct kvmppc_vcore, in_guest));
 	DEFINE(VCORE_NAPPING_THREADS, offsetof(struct kvmppc_vcore, napping_threads));
-	DEFINE(VCPU_SVCPU, offsetof(struct kvmppc_vcpu_book3s, shadow_vcpu) -
-			   offsetof(struct kvmppc_vcpu_book3s, vcpu));
+	DEFINE(VCORE_TB_OFFSET, offsetof(struct kvmppc_vcore, tb_offset));
+	DEFINE(VCORE_LPCR, offsetof(struct kvmppc_vcore, lpcr));
+	DEFINE(VCORE_PCR, offsetof(struct kvmppc_vcore, pcr));
 	DEFINE(VCPU_SLB_E, offsetof(struct kvmppc_slb, orige));
 	DEFINE(VCPU_SLB_V, offsetof(struct kvmppc_slb, origv));
 	DEFINE(VCPU_SLB_SIZE, sizeof(struct kvmppc_slb));
 
 #ifdef CONFIG_PPC_BOOK3S_64
-#ifdef CONFIG_KVM_BOOK3S_PR
+#ifdef CONFIG_KVM_BOOK3S_PR_POSSIBLE
+	DEFINE(PACA_SVCPU, offsetof(struct paca_struct, shadow_vcpu));
 # define SVCPU_FIELD(x, f)	DEFINE(x, offsetof(struct paca_struct, shadow_vcpu.f))
 #else
 # define SVCPU_FIELD(x, f)
@@ -574,7 +580,7 @@ int main(void)
 	HSTATE_FIELD(HSTATE_RESTORE_HID5, restore_hid5);
 	HSTATE_FIELD(HSTATE_NAPPING, napping);
 
-#ifdef CONFIG_KVM_BOOK3S_64_HV
+#ifdef CONFIG_KVM_BOOK3S_HV_POSSIBLE
 	HSTATE_FIELD(HSTATE_HWTHREAD_REQ, hwthread_req);
 	HSTATE_FIELD(HSTATE_HWTHREAD_STATE, hwthread_state);
 	HSTATE_FIELD(HSTATE_KVM_VCPU, kvm_vcpu);
@@ -590,10 +596,11 @@ int main(void)
 	HSTATE_FIELD(HSTATE_DABR, dabr);
 	HSTATE_FIELD(HSTATE_DECEXP, dec_expires);
 	DEFINE(IPI_PRIORITY, IPI_PRIORITY);
-#endif /* CONFIG_KVM_BOOK3S_64_HV */
+#endif /* CONFIG_KVM_BOOK3S_HV_POSSIBLE */
 
 #ifdef CONFIG_PPC_BOOK3S_64
 	HSTATE_FIELD(HSTATE_CFAR, cfar);
+	HSTATE_FIELD(HSTATE_PPR, ppr);
 #endif /* CONFIG_PPC_BOOK3S_64 */
 
 #else /* CONFIG_PPC_BOOK3S */
diff --git a/arch/powerpc/kernel/exceptions-64s.S b/arch/powerpc/kernel/exceptions-64s.S
index 3a9ed6ac224b..9f905e40922e 100644
--- a/arch/powerpc/kernel/exceptions-64s.S
+++ b/arch/powerpc/kernel/exceptions-64s.S
@@ -126,7 +126,7 @@ BEGIN_FTR_SECTION
 	bgt	cr1,.
 	GET_PACA(r13)
 
-#ifdef CONFIG_KVM_BOOK3S_64_HV
+#ifdef CONFIG_KVM_BOOK3S_HV_POSSIBLE
 	li	r0,KVM_HWTHREAD_IN_KERNEL
 	stb	r0,HSTATE_HWTHREAD_STATE(r13)
 	/* Order setting hwthread_state vs. testing hwthread_req */
@@ -425,7 +425,7 @@ data_access_check_stab:
 	mfspr	r9,SPRN_DSISR
 	srdi	r10,r10,60
 	rlwimi	r10,r9,16,0x20
-#ifdef CONFIG_KVM_BOOK3S_PR
+#ifdef CONFIG_KVM_BOOK3S_PR_POSSIBLE
 	lbz	r9,HSTATE_IN_GUEST(r13)
 	rlwimi	r10,r9,8,0x300
 #endif
@@ -650,6 +650,32 @@ slb_miss_user_pseries:
 	b	.				/* prevent spec. execution */
 #endif /* __DISABLED__ */
 
+#ifdef CONFIG_KVM_BOOK3S_64_HANDLER
+kvmppc_skip_interrupt:
+	/*
+	 * Here all GPRs are unchanged from when the interrupt happened
+	 * except for r13, which is saved in SPRG_SCRATCH0.
+	 */
+	mfspr	r13, SPRN_SRR0
+	addi	r13, r13, 4
+	mtspr	SPRN_SRR0, r13
+	GET_SCRATCH0(r13)
+	rfid
+	b	.
+
+kvmppc_skip_Hinterrupt:
+	/*
+	 * Here all GPRs are unchanged from when the interrupt happened
+	 * except for r13, which is saved in SPRG_SCRATCH0.
+	 */
+	mfspr	r13, SPRN_HSRR0
+	addi	r13, r13, 4
+	mtspr	SPRN_HSRR0, r13
+	GET_SCRATCH0(r13)
+	hrfid
+	b	.
+#endif
+
 /*
  * Code from here down to __end_handlers is invoked from the
  * exception prologs above.  Because the prologs assemble the
diff --git a/arch/powerpc/kernel/idle_power7.S b/arch/powerpc/kernel/idle_power7.S
index e11863f4e595..847e40e62fce 100644
--- a/arch/powerpc/kernel/idle_power7.S
+++ b/arch/powerpc/kernel/idle_power7.S
@@ -84,7 +84,7 @@ _GLOBAL(power7_nap)
 	std	r9,_MSR(r1)
 	std	r1,PACAR1(r13)
 
-#ifdef CONFIG_KVM_BOOK3S_64_HV
+#ifdef CONFIG_KVM_BOOK3S_HV_POSSIBLE
 	/* Tell KVM we're napping */
 	li	r4,KVM_HWTHREAD_IN_NAP
 	stb	r4,HSTATE_HWTHREAD_STATE(r13)
diff --git a/arch/powerpc/kernel/kprobes.c b/arch/powerpc/kernel/kprobes.c
index 2156ea90eb54..90fab64d911d 100644
--- a/arch/powerpc/kernel/kprobes.c
+++ b/arch/powerpc/kernel/kprobes.c
@@ -429,7 +429,7 @@ int __kprobes kprobe_fault_handler(struct pt_regs *regs, int trapnr)
 	case KPROBE_HIT_SSDONE:
 		/*
 		 * We increment the nmissed count for accounting,
-		 * we can also use npre/npostfault count for accouting
+		 * we can also use npre/npostfault count for accounting
 		 * these specific fault cases.
 		 */
 		kprobes_inc_nmissed_count(cur);
diff --git a/arch/powerpc/kernel/traps.c b/arch/powerpc/kernel/traps.c
index 62c3dd8c69f2..907a472f9a9e 100644
--- a/arch/powerpc/kernel/traps.c
+++ b/arch/powerpc/kernel/traps.c
@@ -1529,7 +1529,7 @@ static void handle_debug(struct pt_regs *regs, unsigned long debug_status)
 	 * back on or not.
 	 */
 	if (DBCR_ACTIVE_EVENTS(current->thread.debug.dbcr0,
-	    current->thread.debug.dbcr1))
+			       current->thread.debug.dbcr1))
 		regs->msr |= MSR_DE;
 	else
 		/* Make sure the IDM flag is off */
diff --git a/arch/powerpc/kvm/44x.c b/arch/powerpc/kvm/44x.c
index 2f5c6b6d6877..93221e87b911 100644
--- a/arch/powerpc/kvm/44x.c
+++ b/arch/powerpc/kvm/44x.c
@@ -31,13 +31,13 @@
 #include "44x_tlb.h"
 #include "booke.h"
 
-void kvmppc_core_vcpu_load(struct kvm_vcpu *vcpu, int cpu)
+static void kvmppc_core_vcpu_load_44x(struct kvm_vcpu *vcpu, int cpu)
 {
 	kvmppc_booke_vcpu_load(vcpu, cpu);
 	kvmppc_44x_tlb_load(vcpu);
 }
 
-void kvmppc_core_vcpu_put(struct kvm_vcpu *vcpu)
+static void kvmppc_core_vcpu_put_44x(struct kvm_vcpu *vcpu)
 {
 	kvmppc_44x_tlb_put(vcpu);
 	kvmppc_booke_vcpu_put(vcpu);
@@ -114,29 +114,32 @@ int kvmppc_core_vcpu_translate(struct kvm_vcpu *vcpu,
 	return 0;
 }
 
-void kvmppc_core_get_sregs(struct kvm_vcpu *vcpu, struct kvm_sregs *sregs)
+static int kvmppc_core_get_sregs_44x(struct kvm_vcpu *vcpu,
+				      struct kvm_sregs *sregs)
 {
-	kvmppc_get_sregs_ivor(vcpu, sregs);
+	return kvmppc_get_sregs_ivor(vcpu, sregs);
 }
 
-int kvmppc_core_set_sregs(struct kvm_vcpu *vcpu, struct kvm_sregs *sregs)
+static int kvmppc_core_set_sregs_44x(struct kvm_vcpu *vcpu,
+				     struct kvm_sregs *sregs)
 {
 	return kvmppc_set_sregs_ivor(vcpu, sregs);
 }
 
-int kvmppc_get_one_reg(struct kvm_vcpu *vcpu, u64 id,
-			union kvmppc_one_reg *val)
+static int kvmppc_get_one_reg_44x(struct kvm_vcpu *vcpu, u64 id,
+				  union kvmppc_one_reg *val)
 {
 	return -EINVAL;
 }
 
-int kvmppc_set_one_reg(struct kvm_vcpu *vcpu, u64 id,
-		       union kvmppc_one_reg *val)
+static int kvmppc_set_one_reg_44x(struct kvm_vcpu *vcpu, u64 id,
+				  union kvmppc_one_reg *val)
 {
 	return -EINVAL;
 }
 
-struct kvm_vcpu *kvmppc_core_vcpu_create(struct kvm *kvm, unsigned int id)
+static struct kvm_vcpu *kvmppc_core_vcpu_create_44x(struct kvm *kvm,
+						    unsigned int id)
 {
 	struct kvmppc_vcpu_44x *vcpu_44x;
 	struct kvm_vcpu *vcpu;
@@ -167,7 +170,7 @@ out:
 	return ERR_PTR(err);
 }
 
-void kvmppc_core_vcpu_free(struct kvm_vcpu *vcpu)
+static void kvmppc_core_vcpu_free_44x(struct kvm_vcpu *vcpu)
 {
 	struct kvmppc_vcpu_44x *vcpu_44x = to_44x(vcpu);
 
@@ -176,28 +179,53 @@ void kvmppc_core_vcpu_free(struct kvm_vcpu *vcpu)
 	kmem_cache_free(kvm_vcpu_cache, vcpu_44x);
 }
 
-int kvmppc_core_init_vm(struct kvm *kvm)
+static int kvmppc_core_init_vm_44x(struct kvm *kvm)
 {
 	return 0;
 }
 
-void kvmppc_core_destroy_vm(struct kvm *kvm)
+static void kvmppc_core_destroy_vm_44x(struct kvm *kvm)
 {
 }
 
+static struct kvmppc_ops kvm_ops_44x = {
+	.get_sregs = kvmppc_core_get_sregs_44x,
+	.set_sregs = kvmppc_core_set_sregs_44x,
+	.get_one_reg = kvmppc_get_one_reg_44x,
+	.set_one_reg = kvmppc_set_one_reg_44x,
+	.vcpu_load   = kvmppc_core_vcpu_load_44x,
+	.vcpu_put    = kvmppc_core_vcpu_put_44x,
+	.vcpu_create = kvmppc_core_vcpu_create_44x,
+	.vcpu_free   = kvmppc_core_vcpu_free_44x,
+	.mmu_destroy  = kvmppc_mmu_destroy_44x,
+	.init_vm = kvmppc_core_init_vm_44x,
+	.destroy_vm = kvmppc_core_destroy_vm_44x,
+	.emulate_op = kvmppc_core_emulate_op_44x,
+	.emulate_mtspr = kvmppc_core_emulate_mtspr_44x,
+	.emulate_mfspr = kvmppc_core_emulate_mfspr_44x,
+};
+
 static int __init kvmppc_44x_init(void)
 {
 	int r;
 
 	r = kvmppc_booke_init();
 	if (r)
-		return r;
+		goto err_out;
+
+	r = kvm_init(NULL, sizeof(struct kvmppc_vcpu_44x), 0, THIS_MODULE);
+	if (r)
+		goto err_out;
+	kvm_ops_44x.owner = THIS_MODULE;
+	kvmppc_pr_ops = &kvm_ops_44x;
 
-	return kvm_init(NULL, sizeof(struct kvmppc_vcpu_44x), 0, THIS_MODULE);
+err_out:
+	return r;
 }
 
 static void __exit kvmppc_44x_exit(void)
 {
+	kvmppc_pr_ops = NULL;
 	kvmppc_booke_exit();
 }
 
diff --git a/arch/powerpc/kvm/44x_emulate.c b/arch/powerpc/kvm/44x_emulate.c
index 35ec0a8547da..92c9ab4bcfec 100644
--- a/arch/powerpc/kvm/44x_emulate.c
+++ b/arch/powerpc/kvm/44x_emulate.c
@@ -91,8 +91,8 @@ static int emulate_mfdcr(struct kvm_vcpu *vcpu, int rt, int dcrn)
 	return EMULATE_DONE;
 }
 
-int kvmppc_core_emulate_op(struct kvm_run *run, struct kvm_vcpu *vcpu,
-                           unsigned int inst, int *advance)
+int kvmppc_core_emulate_op_44x(struct kvm_run *run, struct kvm_vcpu *vcpu,
+			       unsigned int inst, int *advance)
 {
 	int emulated = EMULATE_DONE;
 	int dcrn = get_dcrn(inst);
@@ -152,7 +152,7 @@ int kvmppc_core_emulate_op(struct kvm_run *run, struct kvm_vcpu *vcpu,
 	return emulated;
 }
 
-int kvmppc_core_emulate_mtspr(struct kvm_vcpu *vcpu, int sprn, ulong spr_val)
+int kvmppc_core_emulate_mtspr_44x(struct kvm_vcpu *vcpu, int sprn, ulong spr_val)
 {
 	int emulated = EMULATE_DONE;
 
@@ -172,7 +172,7 @@ int kvmppc_core_emulate_mtspr(struct kvm_vcpu *vcpu, int sprn, ulong spr_val)
 	return emulated;
 }
 
-int kvmppc_core_emulate_mfspr(struct kvm_vcpu *vcpu, int sprn, ulong *spr_val)
+int kvmppc_core_emulate_mfspr_44x(struct kvm_vcpu *vcpu, int sprn, ulong *spr_val)
 {
 	int emulated = EMULATE_DONE;
 
diff --git a/arch/powerpc/kvm/44x_tlb.c b/arch/powerpc/kvm/44x_tlb.c
index ed0385448148..0deef1082e02 100644
--- a/arch/powerpc/kvm/44x_tlb.c
+++ b/arch/powerpc/kvm/44x_tlb.c
@@ -268,7 +268,7 @@ static void kvmppc_44x_shadow_release(struct kvmppc_vcpu_44x *vcpu_44x,
 	trace_kvm_stlb_inval(stlb_index);
 }
 
-void kvmppc_mmu_destroy(struct kvm_vcpu *vcpu)
+void kvmppc_mmu_destroy_44x(struct kvm_vcpu *vcpu)
 {
 	struct kvmppc_vcpu_44x *vcpu_44x = to_44x(vcpu);
 	int i;
diff --git a/arch/powerpc/kvm/Kconfig b/arch/powerpc/kvm/Kconfig
index e593ff257bd3..141b2027189a 100644
--- a/arch/powerpc/kvm/Kconfig
+++ b/arch/powerpc/kvm/Kconfig
@@ -35,17 +35,20 @@ config KVM_BOOK3S_64_HANDLER
 	bool
 	select KVM_BOOK3S_HANDLER
 
-config KVM_BOOK3S_PR
+config KVM_BOOK3S_PR_POSSIBLE
 	bool
 	select KVM_MMIO
 	select MMU_NOTIFIER
 
+config KVM_BOOK3S_HV_POSSIBLE
+	bool
+
 config KVM_BOOK3S_32
 	tristate "KVM support for PowerPC book3s_32 processors"
 	depends on PPC_BOOK3S_32 && !SMP && !PTE_64BIT
 	select KVM
 	select KVM_BOOK3S_32_HANDLER
-	select KVM_BOOK3S_PR
+	select KVM_BOOK3S_PR_POSSIBLE
 	---help---
 	  Support running unmodified book3s_32 guest kernels
 	  in virtual machines on book3s_32 host processors.
@@ -60,6 +63,7 @@ config KVM_BOOK3S_64
 	depends on PPC_BOOK3S_64
 	select KVM_BOOK3S_64_HANDLER
 	select KVM
+	select KVM_BOOK3S_PR_POSSIBLE if !KVM_BOOK3S_HV_POSSIBLE
 	---help---
 	  Support running unmodified book3s_64 and book3s_32 guest kernels
 	  in virtual machines on book3s_64 host processors.
@@ -70,8 +74,9 @@ config KVM_BOOK3S_64
 	  If unsure, say N.
 
 config KVM_BOOK3S_64_HV
-	bool "KVM support for POWER7 and PPC970 using hypervisor mode in host"
+	tristate "KVM support for POWER7 and PPC970 using hypervisor mode in host"
 	depends on KVM_BOOK3S_64
+	select KVM_BOOK3S_HV_POSSIBLE
 	select MMU_NOTIFIER
 	select CMA
 	---help---
@@ -90,9 +95,20 @@ config KVM_BOOK3S_64_HV
 	  If unsure, say N.
 
 config KVM_BOOK3S_64_PR
-	def_bool y
-	depends on KVM_BOOK3S_64 && !KVM_BOOK3S_64_HV
-	select KVM_BOOK3S_PR
+	tristate "KVM support without using hypervisor mode in host"
+	depends on KVM_BOOK3S_64
+	select KVM_BOOK3S_PR_POSSIBLE
+	---help---
+	  Support running guest kernels in virtual machines on processors
+	  without using hypervisor mode in the host, by running the
+	  guest in user mode (problem state) and emulating all
+	  privileged instructions and registers.
+
+	  This is not as fast as using hypervisor mode, but works on
+	  machines where hypervisor mode is not available or not usable,
+	  and can emulate processors that are different from the host
+	  processor, including emulating 32-bit processors on a 64-bit
+	  host.
 
 config KVM_BOOKE_HV
 	bool
diff --git a/arch/powerpc/kvm/Makefile b/arch/powerpc/kvm/Makefile
index 6646c952c5e3..ce569b6bf4d8 100644
--- a/arch/powerpc/kvm/Makefile
+++ b/arch/powerpc/kvm/Makefile
@@ -53,41 +53,51 @@ kvm-e500mc-objs := \
 	e500_emulate.o
 kvm-objs-$(CONFIG_KVM_E500MC) := $(kvm-e500mc-objs)
 
-kvm-book3s_64-objs-$(CONFIG_KVM_BOOK3S_64_PR) := \
-	$(KVM)/coalesced_mmio.o \
+kvm-book3s_64-builtin-objs-$(CONFIG_KVM_BOOK3S_64_HANDLER) := \
+	book3s_64_vio_hv.o
+
+kvm-pr-y := \
 	fpu.o \
 	book3s_paired_singles.o \
 	book3s_pr.o \
 	book3s_pr_papr.o \
-	book3s_64_vio_hv.o \
 	book3s_emulate.o \
 	book3s_interrupts.o \
 	book3s_mmu_hpte.o \
 	book3s_64_mmu_host.o \
 	book3s_64_mmu.o \
 	book3s_32_mmu.o
-kvm-book3s_64-builtin-objs-$(CONFIG_KVM_BOOK3S_64_PR) := \
+
+ifdef CONFIG_KVM_BOOK3S_PR_POSSIBLE
+kvm-book3s_64-module-objs := \
+	$(KVM)/coalesced_mmio.o
+
+kvm-book3s_64-builtin-objs-$(CONFIG_KVM_BOOK3S_64_HANDLER) += \
 	book3s_rmhandlers.o
+endif
 
-kvm-book3s_64-objs-$(CONFIG_KVM_BOOK3S_64_HV) := \
+kvm-hv-y += \
 	book3s_hv.o \
 	book3s_hv_interrupts.o \
 	book3s_64_mmu_hv.o
+
 kvm-book3s_64-builtin-xics-objs-$(CONFIG_KVM_XICS) := \
 	book3s_hv_rm_xics.o
-kvm-book3s_64-builtin-objs-$(CONFIG_KVM_BOOK3S_64_HV) := \
+
+ifdef CONFIG_KVM_BOOK3S_HV_POSSIBLE
+kvm-book3s_64-builtin-objs-$(CONFIG_KVM_BOOK3S_64_HANDLER) += \
 	book3s_hv_rmhandlers.o \
 	book3s_hv_rm_mmu.o \
-	book3s_64_vio_hv.o \
 	book3s_hv_ras.o \
 	book3s_hv_builtin.o \
 	book3s_hv_cma.o \
 	$(kvm-book3s_64-builtin-xics-objs-y)
+endif
 
 kvm-book3s_64-objs-$(CONFIG_KVM_XICS) += \
 	book3s_xics.o
 
-kvm-book3s_64-module-objs := \
+kvm-book3s_64-module-objs += \
 	$(KVM)/kvm_main.o \
 	$(KVM)/eventfd.o \
 	powerpc.o \
@@ -123,4 +133,7 @@ obj-$(CONFIG_KVM_E500MC) += kvm.o
 obj-$(CONFIG_KVM_BOOK3S_64) += kvm.o
 obj-$(CONFIG_KVM_BOOK3S_32) += kvm.o
 
+obj-$(CONFIG_KVM_BOOK3S_64_PR) += kvm-pr.o
+obj-$(CONFIG_KVM_BOOK3S_64_HV) += kvm-hv.o
+
 obj-y += $(kvm-book3s_64-builtin-objs-y)
diff --git a/arch/powerpc/kvm/book3s.c b/arch/powerpc/kvm/book3s.c
index 700df6f1d32c..8912608b7e1b 100644
--- a/arch/powerpc/kvm/book3s.c
+++ b/arch/powerpc/kvm/book3s.c
@@ -34,6 +34,7 @@
 #include <linux/vmalloc.h>
 #include <linux/highmem.h>
 
+#include "book3s.h"
 #include "trace.h"
 
 #define VCPU_STAT(x) offsetof(struct kvm_vcpu, stat.x), KVM_STAT_VCPU
@@ -69,6 +70,50 @@ void kvmppc_core_load_guest_debugstate(struct kvm_vcpu *vcpu)
 {
 }
 
+static inline unsigned long kvmppc_interrupt_offset(struct kvm_vcpu *vcpu)
+{
+	if (!is_kvmppc_hv_enabled(vcpu->kvm))
+		return to_book3s(vcpu)->hior;
+	return 0;
+}
+
+static inline void kvmppc_update_int_pending(struct kvm_vcpu *vcpu,
+			unsigned long pending_now, unsigned long old_pending)
+{
+	if (is_kvmppc_hv_enabled(vcpu->kvm))
+		return;
+	if (pending_now)
+		vcpu->arch.shared->int_pending = 1;
+	else if (old_pending)
+		vcpu->arch.shared->int_pending = 0;
+}
+
+static inline bool kvmppc_critical_section(struct kvm_vcpu *vcpu)
+{
+	ulong crit_raw;
+	ulong crit_r1;
+	bool crit;
+
+	if (is_kvmppc_hv_enabled(vcpu->kvm))
+		return false;
+
+	crit_raw = vcpu->arch.shared->critical;
+	crit_r1 = kvmppc_get_gpr(vcpu, 1);
+
+	/* Truncate crit indicators in 32 bit mode */
+	if (!(vcpu->arch.shared->msr & MSR_SF)) {
+		crit_raw &= 0xffffffff;
+		crit_r1 &= 0xffffffff;
+	}
+
+	/* Critical section when crit == r1 */
+	crit = (crit_raw == crit_r1);
+	/* ... and we're in supervisor mode */
+	crit = crit && !(vcpu->arch.shared->msr & MSR_PR);
+
+	return crit;
+}
+
 void kvmppc_inject_interrupt(struct kvm_vcpu *vcpu, int vec, u64 flags)
 {
 	vcpu->arch.shared->srr0 = kvmppc_get_pc(vcpu);
@@ -126,28 +171,32 @@ void kvmppc_book3s_queue_irqprio(struct kvm_vcpu *vcpu, unsigned int vec)
 	printk(KERN_INFO "Queueing interrupt %x\n", vec);
 #endif
 }
-
+EXPORT_SYMBOL_GPL(kvmppc_book3s_queue_irqprio);
 
 void kvmppc_core_queue_program(struct kvm_vcpu *vcpu, ulong flags)
 {
 	/* might as well deliver this straight away */
 	kvmppc_inject_interrupt(vcpu, BOOK3S_INTERRUPT_PROGRAM, flags);
 }
+EXPORT_SYMBOL_GPL(kvmppc_core_queue_program);
 
 void kvmppc_core_queue_dec(struct kvm_vcpu *vcpu)
 {
 	kvmppc_book3s_queue_irqprio(vcpu, BOOK3S_INTERRUPT_DECREMENTER);
 }
+EXPORT_SYMBOL_GPL(kvmppc_core_queue_dec);
 
 int kvmppc_core_pending_dec(struct kvm_vcpu *vcpu)
 {
 	return test_bit(BOOK3S_IRQPRIO_DECREMENTER, &vcpu->arch.pending_exceptions);
 }
+EXPORT_SYMBOL_GPL(kvmppc_core_pending_dec);
 
 void kvmppc_core_dequeue_dec(struct kvm_vcpu *vcpu)
 {
 	kvmppc_book3s_dequeue_irqprio(vcpu, BOOK3S_INTERRUPT_DECREMENTER);
 }
+EXPORT_SYMBOL_GPL(kvmppc_core_dequeue_dec);
 
 void kvmppc_core_queue_external(struct kvm_vcpu *vcpu,
                                 struct kvm_interrupt *irq)
@@ -285,8 +334,10 @@ int kvmppc_core_prepare_to_enter(struct kvm_vcpu *vcpu)
 
 	return 0;
 }
+EXPORT_SYMBOL_GPL(kvmppc_core_prepare_to_enter);
 
-pfn_t kvmppc_gfn_to_pfn(struct kvm_vcpu *vcpu, gfn_t gfn)
+pfn_t kvmppc_gfn_to_pfn(struct kvm_vcpu *vcpu, gfn_t gfn, bool writing,
+			bool *writable)
 {
 	ulong mp_pa = vcpu->arch.magic_page_pa;
 
@@ -302,20 +353,23 @@ pfn_t kvmppc_gfn_to_pfn(struct kvm_vcpu *vcpu, gfn_t gfn)
 
 		pfn = (pfn_t)virt_to_phys((void*)shared_page) >> PAGE_SHIFT;
 		get_page(pfn_to_page(pfn));
+		if (writable)
+			*writable = true;
 		return pfn;
 	}
 
-	return gfn_to_pfn(vcpu->kvm, gfn);
+	return gfn_to_pfn_prot(vcpu->kvm, gfn, writing, writable);
 }
+EXPORT_SYMBOL_GPL(kvmppc_gfn_to_pfn);
 
 static int kvmppc_xlate(struct kvm_vcpu *vcpu, ulong eaddr, bool data,
-			 struct kvmppc_pte *pte)
+			bool iswrite, struct kvmppc_pte *pte)
 {
 	int relocated = (vcpu->arch.shared->msr & (data ? MSR_DR : MSR_IR));
 	int r;
 
 	if (relocated) {
-		r = vcpu->arch.mmu.xlate(vcpu, eaddr, pte, data);
+		r = vcpu->arch.mmu.xlate(vcpu, eaddr, pte, data, iswrite);
 	} else {
 		pte->eaddr = eaddr;
 		pte->raddr = eaddr & KVM_PAM;
@@ -361,7 +415,7 @@ int kvmppc_st(struct kvm_vcpu *vcpu, ulong *eaddr, int size, void *ptr,
 
 	vcpu->stat.st++;
 
-	if (kvmppc_xlate(vcpu, *eaddr, data, &pte))
+	if (kvmppc_xlate(vcpu, *eaddr, data, true, &pte))
 		return -ENOENT;
 
 	*eaddr = pte.raddr;
@@ -374,6 +428,7 @@ int kvmppc_st(struct kvm_vcpu *vcpu, ulong *eaddr, int size, void *ptr,
 
 	return EMULATE_DONE;
 }
+EXPORT_SYMBOL_GPL(kvmppc_st);
 
 int kvmppc_ld(struct kvm_vcpu *vcpu, ulong *eaddr, int size, void *ptr,
 		      bool data)
@@ -383,7 +438,7 @@ int kvmppc_ld(struct kvm_vcpu *vcpu, ulong *eaddr, int size, void *ptr,
 
 	vcpu->stat.ld++;
 
-	if (kvmppc_xlate(vcpu, *eaddr, data, &pte))
+	if (kvmppc_xlate(vcpu, *eaddr, data, false, &pte))
 		goto nopte;
 
 	*eaddr = pte.raddr;
@@ -404,6 +459,7 @@ nopte:
 mmio:
 	return EMULATE_DO_MMIO;
 }
+EXPORT_SYMBOL_GPL(kvmppc_ld);
 
 int kvm_arch_vcpu_setup(struct kvm_vcpu *vcpu)
 {
@@ -419,6 +475,18 @@ void kvmppc_subarch_vcpu_uninit(struct kvm_vcpu *vcpu)
 {
 }
 
+int kvm_arch_vcpu_ioctl_get_sregs(struct kvm_vcpu *vcpu,
+				  struct kvm_sregs *sregs)
+{
+	return vcpu->kvm->arch.kvm_ops->get_sregs(vcpu, sregs);
+}
+
+int kvm_arch_vcpu_ioctl_set_sregs(struct kvm_vcpu *vcpu,
+				  struct kvm_sregs *sregs)
+{
+	return vcpu->kvm->arch.kvm_ops->set_sregs(vcpu, sregs);
+}
+
 int kvm_arch_vcpu_ioctl_get_regs(struct kvm_vcpu *vcpu, struct kvm_regs *regs)
 {
 	int i;
@@ -495,8 +563,7 @@ int kvm_vcpu_ioctl_get_one_reg(struct kvm_vcpu *vcpu, struct kvm_one_reg *reg)
 	if (size > sizeof(val))
 		return -EINVAL;
 
-	r = kvmppc_get_one_reg(vcpu, reg->id, &val);
-
+	r = vcpu->kvm->arch.kvm_ops->get_one_reg(vcpu, reg->id, &val);
 	if (r == -EINVAL) {
 		r = 0;
 		switch (reg->id) {
@@ -528,6 +595,9 @@ int kvm_vcpu_ioctl_get_one_reg(struct kvm_vcpu *vcpu, struct kvm_one_reg *reg)
 			}
 			val = get_reg_val(reg->id, vcpu->arch.vscr.u[3]);
 			break;
+		case KVM_REG_PPC_VRSAVE:
+			val = get_reg_val(reg->id, vcpu->arch.vrsave);
+			break;
 #endif /* CONFIG_ALTIVEC */
 		case KVM_REG_PPC_DEBUG_INST: {
 			u32 opcode = INS_TW;
@@ -572,8 +642,7 @@ int kvm_vcpu_ioctl_set_one_reg(struct kvm_vcpu *vcpu, struct kvm_one_reg *reg)
 	if (copy_from_user(&val, (char __user *)(unsigned long)reg->addr, size))
 		return -EFAULT;
 
-	r = kvmppc_set_one_reg(vcpu, reg->id, &val);
-
+	r = vcpu->kvm->arch.kvm_ops->set_one_reg(vcpu, reg->id, &val);
 	if (r == -EINVAL) {
 		r = 0;
 		switch (reg->id) {
@@ -605,6 +674,13 @@ int kvm_vcpu_ioctl_set_one_reg(struct kvm_vcpu *vcpu, struct kvm_one_reg *reg)
 			}
 			vcpu->arch.vscr.u[3] = set_reg_val(reg->id, val);
 			break;
+		case KVM_REG_PPC_VRSAVE:
+			if (!cpu_has_feature(CPU_FTR_ALTIVEC)) {
+				r = -ENXIO;
+				break;
+			}
+			vcpu->arch.vrsave = set_reg_val(reg->id, val);
+			break;
 #endif /* CONFIG_ALTIVEC */
 #ifdef CONFIG_KVM_XICS
 		case KVM_REG_PPC_ICP_STATE:
@@ -625,6 +701,27 @@ int kvm_vcpu_ioctl_set_one_reg(struct kvm_vcpu *vcpu, struct kvm_one_reg *reg)
 	return r;
 }
 
+void kvmppc_core_vcpu_load(struct kvm_vcpu *vcpu, int cpu)
+{
+	vcpu->kvm->arch.kvm_ops->vcpu_load(vcpu, cpu);
+}
+
+void kvmppc_core_vcpu_put(struct kvm_vcpu *vcpu)
+{
+	vcpu->kvm->arch.kvm_ops->vcpu_put(vcpu);
+}
+
+void kvmppc_set_msr(struct kvm_vcpu *vcpu, u64 msr)
+{
+	vcpu->kvm->arch.kvm_ops->set_msr(vcpu, msr);
+}
+EXPORT_SYMBOL_GPL(kvmppc_set_msr);
+
+int kvmppc_vcpu_run(struct kvm_run *kvm_run, struct kvm_vcpu *vcpu)
+{
+	return vcpu->kvm->arch.kvm_ops->vcpu_run(kvm_run, vcpu);
+}
+
 int kvm_arch_vcpu_ioctl_translate(struct kvm_vcpu *vcpu,
                                   struct kvm_translation *tr)
 {
@@ -644,3 +741,141 @@ void kvmppc_decrementer_func(unsigned long data)
 	kvmppc_core_queue_dec(vcpu);
 	kvm_vcpu_kick(vcpu);
 }
+
+struct kvm_vcpu *kvmppc_core_vcpu_create(struct kvm *kvm, unsigned int id)
+{
+	return kvm->arch.kvm_ops->vcpu_create(kvm, id);
+}
+
+void kvmppc_core_vcpu_free(struct kvm_vcpu *vcpu)
+{
+	vcpu->kvm->arch.kvm_ops->vcpu_free(vcpu);
+}
+
+int kvmppc_core_check_requests(struct kvm_vcpu *vcpu)
+{
+	return vcpu->kvm->arch.kvm_ops->check_requests(vcpu);
+}
+
+int kvm_vm_ioctl_get_dirty_log(struct kvm *kvm, struct kvm_dirty_log *log)
+{
+	return kvm->arch.kvm_ops->get_dirty_log(kvm, log);
+}
+
+void kvmppc_core_free_memslot(struct kvm *kvm, struct kvm_memory_slot *free,
+			      struct kvm_memory_slot *dont)
+{
+	kvm->arch.kvm_ops->free_memslot(free, dont);
+}
+
+int kvmppc_core_create_memslot(struct kvm *kvm, struct kvm_memory_slot *slot,
+			       unsigned long npages)
+{
+	return kvm->arch.kvm_ops->create_memslot(slot, npages);
+}
+
+void kvmppc_core_flush_memslot(struct kvm *kvm, struct kvm_memory_slot *memslot)
+{
+	kvm->arch.kvm_ops->flush_memslot(kvm, memslot);
+}
+
+int kvmppc_core_prepare_memory_region(struct kvm *kvm,
+				struct kvm_memory_slot *memslot,
+				struct kvm_userspace_memory_region *mem)
+{
+	return kvm->arch.kvm_ops->prepare_memory_region(kvm, memslot, mem);
+}
+
+void kvmppc_core_commit_memory_region(struct kvm *kvm,
+				struct kvm_userspace_memory_region *mem,
+				const struct kvm_memory_slot *old)
+{
+	kvm->arch.kvm_ops->commit_memory_region(kvm, mem, old);
+}
+
+int kvm_unmap_hva(struct kvm *kvm, unsigned long hva)
+{
+	return kvm->arch.kvm_ops->unmap_hva(kvm, hva);
+}
+EXPORT_SYMBOL_GPL(kvm_unmap_hva);
+
+int kvm_unmap_hva_range(struct kvm *kvm, unsigned long start, unsigned long end)
+{
+	return kvm->arch.kvm_ops->unmap_hva_range(kvm, start, end);
+}
+
+int kvm_age_hva(struct kvm *kvm, unsigned long hva)
+{
+	return kvm->arch.kvm_ops->age_hva(kvm, hva);
+}
+
+int kvm_test_age_hva(struct kvm *kvm, unsigned long hva)
+{
+	return kvm->arch.kvm_ops->test_age_hva(kvm, hva);
+}
+
+void kvm_set_spte_hva(struct kvm *kvm, unsigned long hva, pte_t pte)
+{
+	kvm->arch.kvm_ops->set_spte_hva(kvm, hva, pte);
+}
+
+void kvmppc_mmu_destroy(struct kvm_vcpu *vcpu)
+{
+	vcpu->kvm->arch.kvm_ops->mmu_destroy(vcpu);
+}
+
+int kvmppc_core_init_vm(struct kvm *kvm)
+{
+
+#ifdef CONFIG_PPC64
+	INIT_LIST_HEAD(&kvm->arch.spapr_tce_tables);
+	INIT_LIST_HEAD(&kvm->arch.rtas_tokens);
+#endif
+
+	return kvm->arch.kvm_ops->init_vm(kvm);
+}
+
+void kvmppc_core_destroy_vm(struct kvm *kvm)
+{
+	kvm->arch.kvm_ops->destroy_vm(kvm);
+
+#ifdef CONFIG_PPC64
+	kvmppc_rtas_tokens_free(kvm);
+	WARN_ON(!list_empty(&kvm->arch.spapr_tce_tables));
+#endif
+}
+
+int kvmppc_core_check_processor_compat(void)
+{
+	/*
+	 * We always return 0 for book3s. We check
+	 * for compatability while loading the HV
+	 * or PR module
+	 */
+	return 0;
+}
+
+static int kvmppc_book3s_init(void)
+{
+	int r;
+
+	r = kvm_init(NULL, sizeof(struct kvm_vcpu), 0, THIS_MODULE);
+	if (r)
+		return r;
+#ifdef CONFIG_KVM_BOOK3S_32
+	r = kvmppc_book3s_init_pr();
+#endif
+	return r;
+
+}
+
+static void kvmppc_book3s_exit(void)
+{
+#ifdef CONFIG_KVM_BOOK3S_32
+	kvmppc_book3s_exit_pr();
+#endif
+	kvm_exit();
+}
+
+module_init(kvmppc_book3s_init);
+module_exit(kvmppc_book3s_exit);
diff --git a/arch/powerpc/kvm/book3s.h b/arch/powerpc/kvm/book3s.h
new file mode 100644
index 000000000000..4bf956cf94d6
--- /dev/null
+++ b/arch/powerpc/kvm/book3s.h
@@ -0,0 +1,34 @@
+/*
+ * Copyright IBM Corporation, 2013
+ * Author Aneesh Kumar K.V <aneesh.kumar@linux.vnet.ibm.com>
+ *
+ * This program is free software; you can redistribute it and/or
+ * modify it under the terms of the GNU General Public License as
+ * published by the Free Software Foundation; either version 2 of the
+ * License or (at your optional) any later version of the license.
+ *
+ */
+
+#ifndef __POWERPC_KVM_BOOK3S_H__
+#define __POWERPC_KVM_BOOK3S_H__
+
+extern void kvmppc_core_flush_memslot_hv(struct kvm *kvm,
+					 struct kvm_memory_slot *memslot);
+extern int kvm_unmap_hva_hv(struct kvm *kvm, unsigned long hva);
+extern int kvm_unmap_hva_range_hv(struct kvm *kvm, unsigned long start,
+				  unsigned long end);
+extern int kvm_age_hva_hv(struct kvm *kvm, unsigned long hva);
+extern int kvm_test_age_hva_hv(struct kvm *kvm, unsigned long hva);
+extern void kvm_set_spte_hva_hv(struct kvm *kvm, unsigned long hva, pte_t pte);
+
+extern void kvmppc_mmu_destroy_pr(struct kvm_vcpu *vcpu);
+extern int kvmppc_core_emulate_op_pr(struct kvm_run *run, struct kvm_vcpu *vcpu,
+				     unsigned int inst, int *advance);
+extern int kvmppc_core_emulate_mtspr_pr(struct kvm_vcpu *vcpu,
+					int sprn, ulong spr_val);
+extern int kvmppc_core_emulate_mfspr_pr(struct kvm_vcpu *vcpu,
+					int sprn, ulong *spr_val);
+extern int kvmppc_book3s_init_pr(void);
+extern void kvmppc_book3s_exit_pr(void);
+
+#endif
diff --git a/arch/powerpc/kvm/book3s_32_mmu.c b/arch/powerpc/kvm/book3s_32_mmu.c
index c8cefdd15fd8..76a64ce6a5b6 100644
--- a/arch/powerpc/kvm/book3s_32_mmu.c
+++ b/arch/powerpc/kvm/book3s_32_mmu.c
@@ -84,7 +84,8 @@ static inline bool sr_nx(u32 sr_raw)
 }
 
 static int kvmppc_mmu_book3s_32_xlate_bat(struct kvm_vcpu *vcpu, gva_t eaddr,
-					  struct kvmppc_pte *pte, bool data);
+					  struct kvmppc_pte *pte, bool data,
+					  bool iswrite);
 static int kvmppc_mmu_book3s_32_esid_to_vsid(struct kvm_vcpu *vcpu, ulong esid,
 					     u64 *vsid);
 
@@ -99,7 +100,7 @@ static u64 kvmppc_mmu_book3s_32_ea_to_vp(struct kvm_vcpu *vcpu, gva_t eaddr,
 	u64 vsid;
 	struct kvmppc_pte pte;
 
-	if (!kvmppc_mmu_book3s_32_xlate_bat(vcpu, eaddr, &pte, data))
+	if (!kvmppc_mmu_book3s_32_xlate_bat(vcpu, eaddr, &pte, data, false))
 		return pte.vpage;
 
 	kvmppc_mmu_book3s_32_esid_to_vsid(vcpu, eaddr >> SID_SHIFT, &vsid);
@@ -111,10 +112,11 @@ static void kvmppc_mmu_book3s_32_reset_msr(struct kvm_vcpu *vcpu)
 	kvmppc_set_msr(vcpu, 0);
 }
 
-static hva_t kvmppc_mmu_book3s_32_get_pteg(struct kvmppc_vcpu_book3s *vcpu_book3s,
+static hva_t kvmppc_mmu_book3s_32_get_pteg(struct kvm_vcpu *vcpu,
 				      u32 sre, gva_t eaddr,
 				      bool primary)
 {
+	struct kvmppc_vcpu_book3s *vcpu_book3s = to_book3s(vcpu);
 	u32 page, hash, pteg, htabmask;
 	hva_t r;
 
@@ -132,7 +134,7 @@ static hva_t kvmppc_mmu_book3s_32_get_pteg(struct kvmppc_vcpu_book3s *vcpu_book3
 		kvmppc_get_pc(&vcpu_book3s->vcpu), eaddr, vcpu_book3s->sdr1, pteg,
 		sr_vsid(sre));
 
-	r = gfn_to_hva(vcpu_book3s->vcpu.kvm, pteg >> PAGE_SHIFT);
+	r = gfn_to_hva(vcpu->kvm, pteg >> PAGE_SHIFT);
 	if (kvm_is_error_hva(r))
 		return r;
 	return r | (pteg & ~PAGE_MASK);
@@ -145,7 +147,8 @@ static u32 kvmppc_mmu_book3s_32_get_ptem(u32 sre, gva_t eaddr, bool primary)
 }
 
 static int kvmppc_mmu_book3s_32_xlate_bat(struct kvm_vcpu *vcpu, gva_t eaddr,
-					  struct kvmppc_pte *pte, bool data)
+					  struct kvmppc_pte *pte, bool data,
+					  bool iswrite)
 {
 	struct kvmppc_vcpu_book3s *vcpu_book3s = to_book3s(vcpu);
 	struct kvmppc_bat *bat;
@@ -186,8 +189,7 @@ static int kvmppc_mmu_book3s_32_xlate_bat(struct kvm_vcpu *vcpu, gva_t eaddr,
 				printk(KERN_INFO "BAT is not readable!\n");
 				continue;
 			}
-			if (!pte->may_write) {
-				/* let's treat r/o BATs as not-readable for now */
+			if (iswrite && !pte->may_write) {
 				dprintk_pte("BAT is read-only!\n");
 				continue;
 			}
@@ -201,9 +203,8 @@ static int kvmppc_mmu_book3s_32_xlate_bat(struct kvm_vcpu *vcpu, gva_t eaddr,
 
 static int kvmppc_mmu_book3s_32_xlate_pte(struct kvm_vcpu *vcpu, gva_t eaddr,
 				     struct kvmppc_pte *pte, bool data,
-				     bool primary)
+				     bool iswrite, bool primary)
 {
-	struct kvmppc_vcpu_book3s *vcpu_book3s = to_book3s(vcpu);
 	u32 sre;
 	hva_t ptegp;
 	u32 pteg[16];
@@ -218,7 +219,7 @@ static int kvmppc_mmu_book3s_32_xlate_pte(struct kvm_vcpu *vcpu, gva_t eaddr,
 
 	pte->vpage = kvmppc_mmu_book3s_32_ea_to_vp(vcpu, eaddr, data);
 
-	ptegp = kvmppc_mmu_book3s_32_get_pteg(vcpu_book3s, sre, eaddr, primary);
+	ptegp = kvmppc_mmu_book3s_32_get_pteg(vcpu, sre, eaddr, primary);
 	if (kvm_is_error_hva(ptegp)) {
 		printk(KERN_INFO "KVM: Invalid PTEG!\n");
 		goto no_page_found;
@@ -258,9 +259,6 @@ static int kvmppc_mmu_book3s_32_xlate_pte(struct kvm_vcpu *vcpu, gva_t eaddr,
 					break;
 			}
 
-			if ( !pte->may_read )
-				continue;
-
 			dprintk_pte("MMU: Found PTE -> %x %x - %x\n",
 				    pteg[i], pteg[i+1], pp);
 			found = 1;
@@ -271,19 +269,23 @@ static int kvmppc_mmu_book3s_32_xlate_pte(struct kvm_vcpu *vcpu, gva_t eaddr,
 	/* Update PTE C and A bits, so the guest's swapper knows we used the
 	   page */
 	if (found) {
-		u32 oldpte = pteg[i+1];
-
-		if (pte->may_read)
-			pteg[i+1] |= PTEG_FLAG_ACCESSED;
-		if (pte->may_write)
-			pteg[i+1] |= PTEG_FLAG_DIRTY;
-		else
-			dprintk_pte("KVM: Mapping read-only page!\n");
-
-		/* Write back into the PTEG */
-		if (pteg[i+1] != oldpte)
-			copy_to_user((void __user *)ptegp, pteg, sizeof(pteg));
-
+		u32 pte_r = pteg[i+1];
+		char __user *addr = (char __user *) &pteg[i+1];
+
+		/*
+		 * Use single-byte writes to update the HPTE, to
+		 * conform to what real hardware does.
+		 */
+		if (pte->may_read && !(pte_r & PTEG_FLAG_ACCESSED)) {
+			pte_r |= PTEG_FLAG_ACCESSED;
+			put_user(pte_r >> 8, addr + 2);
+		}
+		if (iswrite && pte->may_write && !(pte_r & PTEG_FLAG_DIRTY)) {
+			pte_r |= PTEG_FLAG_DIRTY;
+			put_user(pte_r, addr + 3);
+		}
+		if (!pte->may_read || (iswrite && !pte->may_write))
+			return -EPERM;
 		return 0;
 	}
 
@@ -302,12 +304,14 @@ no_page_found:
 }
 
 static int kvmppc_mmu_book3s_32_xlate(struct kvm_vcpu *vcpu, gva_t eaddr,
-				      struct kvmppc_pte *pte, bool data)
+				      struct kvmppc_pte *pte, bool data,
+				      bool iswrite)
 {
 	int r;
 	ulong mp_ea = vcpu->arch.magic_page_ea;
 
 	pte->eaddr = eaddr;
+	pte->page_size = MMU_PAGE_4K;
 
 	/* Magic page override */
 	if (unlikely(mp_ea) &&
@@ -323,11 +327,13 @@ static int kvmppc_mmu_book3s_32_xlate(struct kvm_vcpu *vcpu, gva_t eaddr,
 		return 0;
 	}
 
-	r = kvmppc_mmu_book3s_32_xlate_bat(vcpu, eaddr, pte, data);
+	r = kvmppc_mmu_book3s_32_xlate_bat(vcpu, eaddr, pte, data, iswrite);
 	if (r < 0)
-	       r = kvmppc_mmu_book3s_32_xlate_pte(vcpu, eaddr, pte, data, true);
+		r = kvmppc_mmu_book3s_32_xlate_pte(vcpu, eaddr, pte,
+						   data, iswrite, true);
 	if (r < 0)
-	       r = kvmppc_mmu_book3s_32_xlate_pte(vcpu, eaddr, pte, data, false);
+		r = kvmppc_mmu_book3s_32_xlate_pte(vcpu, eaddr, pte,
+						   data, iswrite, false);
 
 	return r;
 }
@@ -347,7 +353,12 @@ static void kvmppc_mmu_book3s_32_mtsrin(struct kvm_vcpu *vcpu, u32 srnum,
 
 static void kvmppc_mmu_book3s_32_tlbie(struct kvm_vcpu *vcpu, ulong ea, bool large)
 {
-	kvmppc_mmu_pte_flush(vcpu, ea, 0x0FFFF000);
+	int i;
+	struct kvm_vcpu *v;
+
+	/* flush this VA on all cpus */
+	kvm_for_each_vcpu(i, v, vcpu->kvm)
+		kvmppc_mmu_pte_flush(v, ea, 0x0FFFF000);
 }
 
 static int kvmppc_mmu_book3s_32_esid_to_vsid(struct kvm_vcpu *vcpu, ulong esid,
diff --git a/arch/powerpc/kvm/book3s_32_mmu_host.c b/arch/powerpc/kvm/book3s_32_mmu_host.c
index 00e619bf608e..3a0abd2e5a15 100644
--- a/arch/powerpc/kvm/book3s_32_mmu_host.c
+++ b/arch/powerpc/kvm/book3s_32_mmu_host.c
@@ -138,7 +138,8 @@ static u32 *kvmppc_mmu_get_pteg(struct kvm_vcpu *vcpu, u32 vsid, u32 eaddr,
 
 extern char etext[];
 
-int kvmppc_mmu_map_page(struct kvm_vcpu *vcpu, struct kvmppc_pte *orig_pte)
+int kvmppc_mmu_map_page(struct kvm_vcpu *vcpu, struct kvmppc_pte *orig_pte,
+			bool iswrite)
 {
 	pfn_t hpaddr;
 	u64 vpn;
@@ -152,9 +153,11 @@ int kvmppc_mmu_map_page(struct kvm_vcpu *vcpu, struct kvmppc_pte *orig_pte)
 	bool evict = false;
 	struct hpte_cache *pte;
 	int r = 0;
+	bool writable;
 
 	/* Get host physical address for gpa */
-	hpaddr = kvmppc_gfn_to_pfn(vcpu, orig_pte->raddr >> PAGE_SHIFT);
+	hpaddr = kvmppc_gfn_to_pfn(vcpu, orig_pte->raddr >> PAGE_SHIFT,
+				   iswrite, &writable);
 	if (is_error_noslot_pfn(hpaddr)) {
 		printk(KERN_INFO "Couldn't get guest page for gfn %lx!\n",
 				 orig_pte->eaddr);
@@ -204,7 +207,7 @@ next_pteg:
 		(primary ? 0 : PTE_SEC);
 	pteg1 = hpaddr | PTE_M | PTE_R | PTE_C;
 
-	if (orig_pte->may_write) {
+	if (orig_pte->may_write && writable) {
 		pteg1 |= PP_RWRW;
 		mark_page_dirty(vcpu->kvm, orig_pte->raddr >> PAGE_SHIFT);
 	} else {
@@ -259,6 +262,11 @@ out:
 	return r;
 }
 
+void kvmppc_mmu_unmap_page(struct kvm_vcpu *vcpu, struct kvmppc_pte *pte)
+{
+	kvmppc_mmu_pte_vflush(vcpu, pte->vpage, 0xfffffffffULL);
+}
+
 static struct kvmppc_sid_map *create_sid_map(struct kvm_vcpu *vcpu, u64 gvsid)
 {
 	struct kvmppc_sid_map *map;
@@ -341,7 +349,7 @@ void kvmppc_mmu_flush_segments(struct kvm_vcpu *vcpu)
 	svcpu_put(svcpu);
 }
 
-void kvmppc_mmu_destroy(struct kvm_vcpu *vcpu)
+void kvmppc_mmu_destroy_pr(struct kvm_vcpu *vcpu)
 {
 	int i;
 
diff --git a/arch/powerpc/kvm/book3s_64_mmu.c b/arch/powerpc/kvm/book3s_64_mmu.c
index 7e345e00661a..83da1f868fd5 100644
--- a/arch/powerpc/kvm/book3s_64_mmu.c
+++ b/arch/powerpc/kvm/book3s_64_mmu.c
@@ -107,9 +107,20 @@ static u64 kvmppc_mmu_book3s_64_ea_to_vp(struct kvm_vcpu *vcpu, gva_t eaddr,
 	return kvmppc_slb_calc_vpn(slb, eaddr);
 }
 
+static int mmu_pagesize(int mmu_pg)
+{
+	switch (mmu_pg) {
+	case MMU_PAGE_64K:
+		return 16;
+	case MMU_PAGE_16M:
+		return 24;
+	}
+	return 12;
+}
+
 static int kvmppc_mmu_book3s_64_get_pagesize(struct kvmppc_slb *slbe)
 {
-	return slbe->large ? 24 : 12;
+	return mmu_pagesize(slbe->base_page_size);
 }
 
 static u32 kvmppc_mmu_book3s_64_get_page(struct kvmppc_slb *slbe, gva_t eaddr)
@@ -119,11 +130,11 @@ static u32 kvmppc_mmu_book3s_64_get_page(struct kvmppc_slb *slbe, gva_t eaddr)
 	return ((eaddr & kvmppc_slb_offset_mask(slbe)) >> p);
 }
 
-static hva_t kvmppc_mmu_book3s_64_get_pteg(
-				struct kvmppc_vcpu_book3s *vcpu_book3s,
+static hva_t kvmppc_mmu_book3s_64_get_pteg(struct kvm_vcpu *vcpu,
 				struct kvmppc_slb *slbe, gva_t eaddr,
 				bool second)
 {
+	struct kvmppc_vcpu_book3s *vcpu_book3s = to_book3s(vcpu);
 	u64 hash, pteg, htabsize;
 	u32 ssize;
 	hva_t r;
@@ -148,10 +159,10 @@ static hva_t kvmppc_mmu_book3s_64_get_pteg(
 
 	/* When running a PAPR guest, SDR1 contains a HVA address instead
            of a GPA */
-	if (vcpu_book3s->vcpu.arch.papr_enabled)
+	if (vcpu->arch.papr_enabled)
 		r = pteg;
 	else
-		r = gfn_to_hva(vcpu_book3s->vcpu.kvm, pteg >> PAGE_SHIFT);
+		r = gfn_to_hva(vcpu->kvm, pteg >> PAGE_SHIFT);
 
 	if (kvm_is_error_hva(r))
 		return r;
@@ -166,18 +177,38 @@ static u64 kvmppc_mmu_book3s_64_get_avpn(struct kvmppc_slb *slbe, gva_t eaddr)
 	avpn = kvmppc_mmu_book3s_64_get_page(slbe, eaddr);
 	avpn |= slbe->vsid << (kvmppc_slb_sid_shift(slbe) - p);
 
-	if (p < 24)
-		avpn >>= ((80 - p) - 56) - 8;
+	if (p < 16)
+		avpn >>= ((80 - p) - 56) - 8;	/* 16 - p */
 	else
-		avpn <<= 8;
+		avpn <<= p - 16;
 
 	return avpn;
 }
 
+/*
+ * Return page size encoded in the second word of a HPTE, or
+ * -1 for an invalid encoding for the base page size indicated by
+ * the SLB entry.  This doesn't handle mixed pagesize segments yet.
+ */
+static int decode_pagesize(struct kvmppc_slb *slbe, u64 r)
+{
+	switch (slbe->base_page_size) {
+	case MMU_PAGE_64K:
+		if ((r & 0xf000) == 0x1000)
+			return MMU_PAGE_64K;
+		break;
+	case MMU_PAGE_16M:
+		if ((r & 0xff000) == 0)
+			return MMU_PAGE_16M;
+		break;
+	}
+	return -1;
+}
+
 static int kvmppc_mmu_book3s_64_xlate(struct kvm_vcpu *vcpu, gva_t eaddr,
-				struct kvmppc_pte *gpte, bool data)
+				      struct kvmppc_pte *gpte, bool data,
+				      bool iswrite)
 {
-	struct kvmppc_vcpu_book3s *vcpu_book3s = to_book3s(vcpu);
 	struct kvmppc_slb *slbe;
 	hva_t ptegp;
 	u64 pteg[16];
@@ -189,6 +220,7 @@ static int kvmppc_mmu_book3s_64_xlate(struct kvm_vcpu *vcpu, gva_t eaddr,
 	u8 pp, key = 0;
 	bool found = false;
 	bool second = false;
+	int pgsize;
 	ulong mp_ea = vcpu->arch.magic_page_ea;
 
 	/* Magic page override */
@@ -202,6 +234,7 @@ static int kvmppc_mmu_book3s_64_xlate(struct kvm_vcpu *vcpu, gva_t eaddr,
 		gpte->may_execute = true;
 		gpte->may_read = true;
 		gpte->may_write = true;
+		gpte->page_size = MMU_PAGE_4K;
 
 		return 0;
 	}
@@ -222,8 +255,12 @@ static int kvmppc_mmu_book3s_64_xlate(struct kvm_vcpu *vcpu, gva_t eaddr,
 	v_mask = SLB_VSID_B | HPTE_V_AVPN | HPTE_V_LARGE | HPTE_V_VALID |
 		HPTE_V_SECONDARY;
 
+	pgsize = slbe->large ? MMU_PAGE_16M : MMU_PAGE_4K;
+
+	mutex_lock(&vcpu->kvm->arch.hpt_mutex);
+
 do_second:
-	ptegp = kvmppc_mmu_book3s_64_get_pteg(vcpu_book3s, slbe, eaddr, second);
+	ptegp = kvmppc_mmu_book3s_64_get_pteg(vcpu, slbe, eaddr, second);
 	if (kvm_is_error_hva(ptegp))
 		goto no_page_found;
 
@@ -240,6 +277,13 @@ do_second:
 	for (i=0; i<16; i+=2) {
 		/* Check all relevant fields of 1st dword */
 		if ((pteg[i] & v_mask) == v_val) {
+			/* If large page bit is set, check pgsize encoding */
+			if (slbe->large &&
+			    (vcpu->arch.hflags & BOOK3S_HFLAG_MULTI_PGSIZE)) {
+				pgsize = decode_pagesize(slbe, pteg[i+1]);
+				if (pgsize < 0)
+					continue;
+			}
 			found = true;
 			break;
 		}
@@ -256,13 +300,15 @@ do_second:
 	v = pteg[i];
 	r = pteg[i+1];
 	pp = (r & HPTE_R_PP) | key;
-	eaddr_mask = 0xFFF;
+	if (r & HPTE_R_PP0)
+		pp |= 8;
 
 	gpte->eaddr = eaddr;
 	gpte->vpage = kvmppc_mmu_book3s_64_ea_to_vp(vcpu, eaddr, data);
-	if (slbe->large)
-		eaddr_mask = 0xFFFFFF;
+
+	eaddr_mask = (1ull << mmu_pagesize(pgsize)) - 1;
 	gpte->raddr = (r & HPTE_R_RPN & ~eaddr_mask) | (eaddr & eaddr_mask);
+	gpte->page_size = pgsize;
 	gpte->may_execute = ((r & HPTE_R_N) ? false : true);
 	gpte->may_read = false;
 	gpte->may_write = false;
@@ -277,6 +323,7 @@ do_second:
 	case 3:
 	case 5:
 	case 7:
+	case 10:
 		gpte->may_read = true;
 		break;
 	}
@@ -287,30 +334,37 @@ do_second:
 
 	/* Update PTE R and C bits, so the guest's swapper knows we used the
 	 * page */
-	if (gpte->may_read) {
-		/* Set the accessed flag */
+	if (gpte->may_read && !(r & HPTE_R_R)) {
+		/*
+		 * Set the accessed flag.
+		 * We have to write this back with a single byte write
+		 * because another vcpu may be accessing this on
+		 * non-PAPR platforms such as mac99, and this is
+		 * what real hardware does.
+		 */
+		char __user *addr = (char __user *) &pteg[i+1];
 		r |= HPTE_R_R;
+		put_user(r >> 8, addr + 6);
 	}
-	if (data && gpte->may_write) {
-		/* Set the dirty flag -- XXX even if not writing */
+	if (iswrite && gpte->may_write && !(r & HPTE_R_C)) {
+		/* Set the dirty flag */
+		/* Use a single byte write */
+		char __user *addr = (char __user *) &pteg[i+1];
 		r |= HPTE_R_C;
+		put_user(r, addr + 7);
 	}
 
-	/* Write back into the PTEG */
-	if (pteg[i+1] != r) {
-		pteg[i+1] = r;
-		copy_to_user((void __user *)ptegp, pteg, sizeof(pteg));
-	}
+	mutex_unlock(&vcpu->kvm->arch.hpt_mutex);
 
-	if (!gpte->may_read)
+	if (!gpte->may_read || (iswrite && !gpte->may_write))
 		return -EPERM;
 	return 0;
 
 no_page_found:
+	mutex_unlock(&vcpu->kvm->arch.hpt_mutex);
 	return -ENOENT;
 
 no_seg_found:
-
 	dprintk("KVM MMU: Trigger segment fault\n");
 	return -EINVAL;
 }
@@ -345,6 +399,21 @@ static void kvmppc_mmu_book3s_64_slbmte(struct kvm_vcpu *vcpu, u64 rs, u64 rb)
 	slbe->nx    = (rs & SLB_VSID_N) ? 1 : 0;
 	slbe->class = (rs & SLB_VSID_C) ? 1 : 0;
 
+	slbe->base_page_size = MMU_PAGE_4K;
+	if (slbe->large) {
+		if (vcpu->arch.hflags & BOOK3S_HFLAG_MULTI_PGSIZE) {
+			switch (rs & SLB_VSID_LP) {
+			case SLB_VSID_LP_00:
+				slbe->base_page_size = MMU_PAGE_16M;
+				break;
+			case SLB_VSID_LP_01:
+				slbe->base_page_size = MMU_PAGE_64K;
+				break;
+			}
+		} else
+			slbe->base_page_size = MMU_PAGE_16M;
+	}
+
 	slbe->orige = rb & (ESID_MASK | SLB_ESID_V);
 	slbe->origv = rs;
 
@@ -460,14 +529,45 @@ static void kvmppc_mmu_book3s_64_tlbie(struct kvm_vcpu *vcpu, ulong va,
 				       bool large)
 {
 	u64 mask = 0xFFFFFFFFFULL;
+	long i;
+	struct kvm_vcpu *v;
 
 	dprintk("KVM MMU: tlbie(0x%lx)\n", va);
 
-	if (large)
-		mask = 0xFFFFFF000ULL;
-	kvmppc_mmu_pte_vflush(vcpu, va >> 12, mask);
+	/*
+	 * The tlbie instruction changed behaviour starting with
+	 * POWER6.  POWER6 and later don't have the large page flag
+	 * in the instruction but in the RB value, along with bits
+	 * indicating page and segment sizes.
+	 */
+	if (vcpu->arch.hflags & BOOK3S_HFLAG_NEW_TLBIE) {
+		/* POWER6 or later */
+		if (va & 1) {		/* L bit */
+			if ((va & 0xf000) == 0x1000)
+				mask = 0xFFFFFFFF0ULL;	/* 64k page */
+			else
+				mask = 0xFFFFFF000ULL;	/* 16M page */
+		}
+	} else {
+		/* older processors, e.g. PPC970 */
+		if (large)
+			mask = 0xFFFFFF000ULL;
+	}
+	/* flush this VA on all vcpus */
+	kvm_for_each_vcpu(i, v, vcpu->kvm)
+		kvmppc_mmu_pte_vflush(v, va >> 12, mask);
 }
 
+#ifdef CONFIG_PPC_64K_PAGES
+static int segment_contains_magic_page(struct kvm_vcpu *vcpu, ulong esid)
+{
+	ulong mp_ea = vcpu->arch.magic_page_ea;
+
+	return mp_ea && !(vcpu->arch.shared->msr & MSR_PR) &&
+		(mp_ea >> SID_SHIFT) == esid;
+}
+#endif
+
 static int kvmppc_mmu_book3s_64_esid_to_vsid(struct kvm_vcpu *vcpu, ulong esid,
 					     u64 *vsid)
 {
@@ -475,11 +575,13 @@ static int kvmppc_mmu_book3s_64_esid_to_vsid(struct kvm_vcpu *vcpu, ulong esid,
 	struct kvmppc_slb *slb;
 	u64 gvsid = esid;
 	ulong mp_ea = vcpu->arch.magic_page_ea;
+	int pagesize = MMU_PAGE_64K;
 
 	if (vcpu->arch.shared->msr & (MSR_DR|MSR_IR)) {
 		slb = kvmppc_mmu_book3s_64_find_slbe(vcpu, ea);
 		if (slb) {
 			gvsid = slb->vsid;
+			pagesize = slb->base_page_size;
 			if (slb->tb) {
 				gvsid <<= SID_SHIFT_1T - SID_SHIFT;
 				gvsid |= esid & ((1ul << (SID_SHIFT_1T - SID_SHIFT)) - 1);
@@ -490,28 +592,41 @@ static int kvmppc_mmu_book3s_64_esid_to_vsid(struct kvm_vcpu *vcpu, ulong esid,
 
 	switch (vcpu->arch.shared->msr & (MSR_DR|MSR_IR)) {
 	case 0:
-		*vsid = VSID_REAL | esid;
+		gvsid = VSID_REAL | esid;
 		break;
 	case MSR_IR:
-		*vsid = VSID_REAL_IR | gvsid;
+		gvsid |= VSID_REAL_IR;
 		break;
 	case MSR_DR:
-		*vsid = VSID_REAL_DR | gvsid;
+		gvsid |= VSID_REAL_DR;
 		break;
 	case MSR_DR|MSR_IR:
 		if (!slb)
 			goto no_slb;
 
-		*vsid = gvsid;
 		break;
 	default:
 		BUG();
 		break;
 	}
 
+#ifdef CONFIG_PPC_64K_PAGES
+	/*
+	 * Mark this as a 64k segment if the host is using
+	 * 64k pages, the host MMU supports 64k pages and
+	 * the guest segment page size is >= 64k,
+	 * but not if this segment contains the magic page.
+	 */
+	if (pagesize >= MMU_PAGE_64K &&
+	    mmu_psize_defs[MMU_PAGE_64K].shift &&
+	    !segment_contains_magic_page(vcpu, esid))
+		gvsid |= VSID_64K;
+#endif
+
 	if (vcpu->arch.shared->msr & MSR_PR)
-		*vsid |= VSID_PR;
+		gvsid |= VSID_PR;
 
+	*vsid = gvsid;
 	return 0;
 
 no_slb:
diff --git a/arch/powerpc/kvm/book3s_64_mmu_host.c b/arch/powerpc/kvm/book3s_64_mmu_host.c
index e5240524bf6c..0d513af62bba 100644
--- a/arch/powerpc/kvm/book3s_64_mmu_host.c
+++ b/arch/powerpc/kvm/book3s_64_mmu_host.c
@@ -27,14 +27,14 @@
 #include <asm/machdep.h>
 #include <asm/mmu_context.h>
 #include <asm/hw_irq.h>
-#include "trace.h"
+#include "trace_pr.h"
 
 #define PTE_SIZE 12
 
 void kvmppc_mmu_invalidate_pte(struct kvm_vcpu *vcpu, struct hpte_cache *pte)
 {
 	ppc_md.hpte_invalidate(pte->slot, pte->host_vpn,
-			       MMU_PAGE_4K, MMU_PAGE_4K, MMU_SEGSIZE_256M,
+			       pte->pagesize, pte->pagesize, MMU_SEGSIZE_256M,
 			       false);
 }
 
@@ -78,7 +78,8 @@ static struct kvmppc_sid_map *find_sid_vsid(struct kvm_vcpu *vcpu, u64 gvsid)
 	return NULL;
 }
 
-int kvmppc_mmu_map_page(struct kvm_vcpu *vcpu, struct kvmppc_pte *orig_pte)
+int kvmppc_mmu_map_page(struct kvm_vcpu *vcpu, struct kvmppc_pte *orig_pte,
+			bool iswrite)
 {
 	unsigned long vpn;
 	pfn_t hpaddr;
@@ -90,16 +91,26 @@ int kvmppc_mmu_map_page(struct kvm_vcpu *vcpu, struct kvmppc_pte *orig_pte)
 	int attempt = 0;
 	struct kvmppc_sid_map *map;
 	int r = 0;
+	int hpsize = MMU_PAGE_4K;
+	bool writable;
+	unsigned long mmu_seq;
+	struct kvm *kvm = vcpu->kvm;
+	struct hpte_cache *cpte;
+	unsigned long gfn = orig_pte->raddr >> PAGE_SHIFT;
+	unsigned long pfn;
+
+	/* used to check for invalidations in progress */
+	mmu_seq = kvm->mmu_notifier_seq;
+	smp_rmb();
 
 	/* Get host physical address for gpa */
-	hpaddr = kvmppc_gfn_to_pfn(vcpu, orig_pte->raddr >> PAGE_SHIFT);
-	if (is_error_noslot_pfn(hpaddr)) {
-		printk(KERN_INFO "Couldn't get guest page for gfn %lx!\n", orig_pte->eaddr);
+	pfn = kvmppc_gfn_to_pfn(vcpu, gfn, iswrite, &writable);
+	if (is_error_noslot_pfn(pfn)) {
+		printk(KERN_INFO "Couldn't get guest page for gfn %lx!\n", gfn);
 		r = -EINVAL;
 		goto out;
 	}
-	hpaddr <<= PAGE_SHIFT;
-	hpaddr |= orig_pte->raddr & (~0xfffULL & ~PAGE_MASK);
+	hpaddr = pfn << PAGE_SHIFT;
 
 	/* and write the mapping ea -> hpa into the pt */
 	vcpu->arch.mmu.esid_to_vsid(vcpu, orig_pte->eaddr >> SID_SHIFT, &vsid);
@@ -117,20 +128,39 @@ int kvmppc_mmu_map_page(struct kvm_vcpu *vcpu, struct kvmppc_pte *orig_pte)
 		goto out;
 	}
 
-	vsid = map->host_vsid;
-	vpn = hpt_vpn(orig_pte->eaddr, vsid, MMU_SEGSIZE_256M);
+	vpn = hpt_vpn(orig_pte->eaddr, map->host_vsid, MMU_SEGSIZE_256M);
 
-	if (!orig_pte->may_write)
-		rflags |= HPTE_R_PP;
-	else
-		mark_page_dirty(vcpu->kvm, orig_pte->raddr >> PAGE_SHIFT);
+	kvm_set_pfn_accessed(pfn);
+	if (!orig_pte->may_write || !writable)
+		rflags |= PP_RXRX;
+	else {
+		mark_page_dirty(vcpu->kvm, gfn);
+		kvm_set_pfn_dirty(pfn);
+	}
 
 	if (!orig_pte->may_execute)
 		rflags |= HPTE_R_N;
 	else
-		kvmppc_mmu_flush_icache(hpaddr >> PAGE_SHIFT);
+		kvmppc_mmu_flush_icache(pfn);
+
+	/*
+	 * Use 64K pages if possible; otherwise, on 64K page kernels,
+	 * we need to transfer 4 more bits from guest real to host real addr.
+	 */
+	if (vsid & VSID_64K)
+		hpsize = MMU_PAGE_64K;
+	else
+		hpaddr |= orig_pte->raddr & (~0xfffULL & ~PAGE_MASK);
+
+	hash = hpt_hash(vpn, mmu_psize_defs[hpsize].shift, MMU_SEGSIZE_256M);
 
-	hash = hpt_hash(vpn, PTE_SIZE, MMU_SEGSIZE_256M);
+	cpte = kvmppc_mmu_hpte_cache_next(vcpu);
+
+	spin_lock(&kvm->mmu_lock);
+	if (!cpte || mmu_notifier_retry(kvm, mmu_seq)) {
+		r = -EAGAIN;
+		goto out_unlock;
+	}
 
 map_again:
 	hpteg = ((hash & htab_hash_mask) * HPTES_PER_GROUP);
@@ -139,11 +169,11 @@ map_again:
 	if (attempt > 1)
 		if (ppc_md.hpte_remove(hpteg) < 0) {
 			r = -1;
-			goto out;
+			goto out_unlock;
 		}
 
 	ret = ppc_md.hpte_insert(hpteg, vpn, hpaddr, rflags, vflags,
-				 MMU_PAGE_4K, MMU_PAGE_4K, MMU_SEGSIZE_256M);
+				 hpsize, hpsize, MMU_SEGSIZE_256M);
 
 	if (ret < 0) {
 		/* If we couldn't map a primary PTE, try a secondary */
@@ -152,8 +182,6 @@ map_again:
 		attempt++;
 		goto map_again;
 	} else {
-		struct hpte_cache *pte = kvmppc_mmu_hpte_cache_next(vcpu);
-
 		trace_kvm_book3s_64_mmu_map(rflags, hpteg,
 					    vpn, hpaddr, orig_pte);
 
@@ -164,19 +192,37 @@ map_again:
 			hpteg = ((hash & htab_hash_mask) * HPTES_PER_GROUP);
 		}
 
-		pte->slot = hpteg + (ret & 7);
-		pte->host_vpn = vpn;
-		pte->pte = *orig_pte;
-		pte->pfn = hpaddr >> PAGE_SHIFT;
+		cpte->slot = hpteg + (ret & 7);
+		cpte->host_vpn = vpn;
+		cpte->pte = *orig_pte;
+		cpte->pfn = pfn;
+		cpte->pagesize = hpsize;
 
-		kvmppc_mmu_hpte_cache_map(vcpu, pte);
+		kvmppc_mmu_hpte_cache_map(vcpu, cpte);
+		cpte = NULL;
 	}
-	kvm_release_pfn_clean(hpaddr >> PAGE_SHIFT);
+
+out_unlock:
+	spin_unlock(&kvm->mmu_lock);
+	kvm_release_pfn_clean(pfn);
+	if (cpte)
+		kvmppc_mmu_hpte_cache_free(cpte);
 
 out:
 	return r;
 }
 
+void kvmppc_mmu_unmap_page(struct kvm_vcpu *vcpu, struct kvmppc_pte *pte)
+{
+	u64 mask = 0xfffffffffULL;
+	u64 vsid;
+
+	vcpu->arch.mmu.esid_to_vsid(vcpu, pte->eaddr >> SID_SHIFT, &vsid);
+	if (vsid & VSID_64K)
+		mask = 0xffffffff0ULL;
+	kvmppc_mmu_pte_vflush(vcpu, pte->vpage, mask);
+}
+
 static struct kvmppc_sid_map *create_sid_map(struct kvm_vcpu *vcpu, u64 gvsid)
 {
 	struct kvmppc_sid_map *map;
@@ -291,6 +337,12 @@ int kvmppc_mmu_map_segment(struct kvm_vcpu *vcpu, ulong eaddr)
 	slb_vsid &= ~SLB_VSID_KP;
 	slb_esid |= slb_index;
 
+#ifdef CONFIG_PPC_64K_PAGES
+	/* Set host segment base page size to 64K if possible */
+	if (gvsid & VSID_64K)
+		slb_vsid |= mmu_psize_defs[MMU_PAGE_64K].sllp;
+#endif
+
 	svcpu->slb[slb_index].esid = slb_esid;
 	svcpu->slb[slb_index].vsid = slb_vsid;
 
@@ -326,7 +378,7 @@ void kvmppc_mmu_flush_segments(struct kvm_vcpu *vcpu)
 	svcpu_put(svcpu);
 }
 
-void kvmppc_mmu_destroy(struct kvm_vcpu *vcpu)
+void kvmppc_mmu_destroy_pr(struct kvm_vcpu *vcpu)
 {
 	kvmppc_mmu_hpte_destroy(vcpu);
 	__destroy_context(to_book3s(vcpu)->context_id[0]);
diff --git a/arch/powerpc/kvm/book3s_64_mmu_hv.c b/arch/powerpc/kvm/book3s_64_mmu_hv.c
index 043eec8461e7..f3ff587a8b7d 100644
--- a/arch/powerpc/kvm/book3s_64_mmu_hv.c
+++ b/arch/powerpc/kvm/book3s_64_mmu_hv.c
@@ -260,10 +260,6 @@ int kvmppc_mmu_hv_init(void)
 	return 0;
 }
 
-void kvmppc_mmu_destroy(struct kvm_vcpu *vcpu)
-{
-}
-
 static void kvmppc_mmu_book3s_64_hv_reset_msr(struct kvm_vcpu *vcpu)
 {
 	kvmppc_set_msr(vcpu, MSR_SF | MSR_ME);
@@ -451,7 +447,7 @@ static unsigned long kvmppc_mmu_get_real_addr(unsigned long v, unsigned long r,
 }
 
 static int kvmppc_mmu_book3s_64_hv_xlate(struct kvm_vcpu *vcpu, gva_t eaddr,
-			struct kvmppc_pte *gpte, bool data)
+			struct kvmppc_pte *gpte, bool data, bool iswrite)
 {
 	struct kvm *kvm = vcpu->kvm;
 	struct kvmppc_slb *slbe;
@@ -906,21 +902,22 @@ static int kvm_unmap_rmapp(struct kvm *kvm, unsigned long *rmapp,
 	return 0;
 }
 
-int kvm_unmap_hva(struct kvm *kvm, unsigned long hva)
+int kvm_unmap_hva_hv(struct kvm *kvm, unsigned long hva)
 {
 	if (kvm->arch.using_mmu_notifiers)
 		kvm_handle_hva(kvm, hva, kvm_unmap_rmapp);
 	return 0;
 }
 
-int kvm_unmap_hva_range(struct kvm *kvm, unsigned long start, unsigned long end)
+int kvm_unmap_hva_range_hv(struct kvm *kvm, unsigned long start, unsigned long end)
 {
 	if (kvm->arch.using_mmu_notifiers)
 		kvm_handle_hva_range(kvm, start, end, kvm_unmap_rmapp);
 	return 0;
 }
 
-void kvmppc_core_flush_memslot(struct kvm *kvm, struct kvm_memory_slot *memslot)
+void kvmppc_core_flush_memslot_hv(struct kvm *kvm,
+				  struct kvm_memory_slot *memslot)
 {
 	unsigned long *rmapp;
 	unsigned long gfn;
@@ -994,7 +991,7 @@ static int kvm_age_rmapp(struct kvm *kvm, unsigned long *rmapp,
 	return ret;
 }
 
-int kvm_age_hva(struct kvm *kvm, unsigned long hva)
+int kvm_age_hva_hv(struct kvm *kvm, unsigned long hva)
 {
 	if (!kvm->arch.using_mmu_notifiers)
 		return 0;
@@ -1032,14 +1029,14 @@ static int kvm_test_age_rmapp(struct kvm *kvm, unsigned long *rmapp,
 	return ret;
 }
 
-int kvm_test_age_hva(struct kvm *kvm, unsigned long hva)
+int kvm_test_age_hva_hv(struct kvm *kvm, unsigned long hva)
 {
 	if (!kvm->arch.using_mmu_notifiers)
 		return 0;
 	return kvm_handle_hva(kvm, hva, kvm_test_age_rmapp);
 }
 
-void kvm_set_spte_hva(struct kvm *kvm, unsigned long hva, pte_t pte)
+void kvm_set_spte_hva_hv(struct kvm *kvm, unsigned long hva, pte_t pte)
 {
 	if (!kvm->arch.using_mmu_notifiers)
 		return;
@@ -1512,9 +1509,8 @@ static ssize_t kvm_htab_write(struct file *file, const char __user *buf,
 
 				kvm->arch.vrma_slb_v = senc | SLB_VSID_B_1T |
 					(VRMA_VSID << SLB_VSID_SHIFT_1T);
-				lpcr = kvm->arch.lpcr & ~LPCR_VRMASD;
-				lpcr |= senc << (LPCR_VRMASD_SH - 4);
-				kvm->arch.lpcr = lpcr;
+				lpcr = senc << (LPCR_VRMASD_SH - 4);
+				kvmppc_update_lpcr(kvm, lpcr, LPCR_VRMASD);
 				rma_setup = 1;
 			}
 			++i;
diff --git a/arch/powerpc/kvm/book3s_64_vio_hv.c b/arch/powerpc/kvm/book3s_64_vio_hv.c
index 30c2f3b134c6..2c25f5412bdb 100644
--- a/arch/powerpc/kvm/book3s_64_vio_hv.c
+++ b/arch/powerpc/kvm/book3s_64_vio_hv.c
@@ -74,3 +74,4 @@ long kvmppc_h_put_tce(struct kvm_vcpu *vcpu, unsigned long liobn,
 	/* Didn't find the liobn, punt it to userspace */
 	return H_TOO_HARD;
 }
+EXPORT_SYMBOL_GPL(kvmppc_h_put_tce);
diff --git a/arch/powerpc/kvm/book3s_emulate.c b/arch/powerpc/kvm/book3s_emulate.c
index 360ce68c9809..99d40f8977e8 100644
--- a/arch/powerpc/kvm/book3s_emulate.c
+++ b/arch/powerpc/kvm/book3s_emulate.c
@@ -86,8 +86,8 @@ static bool spr_allowed(struct kvm_vcpu *vcpu, enum priv_level level)
 	return true;
 }
 
-int kvmppc_core_emulate_op(struct kvm_run *run, struct kvm_vcpu *vcpu,
-                           unsigned int inst, int *advance)
+int kvmppc_core_emulate_op_pr(struct kvm_run *run, struct kvm_vcpu *vcpu,
+			      unsigned int inst, int *advance)
 {
 	int emulated = EMULATE_DONE;
 	int rt = get_rt(inst);
@@ -172,7 +172,7 @@ int kvmppc_core_emulate_op(struct kvm_run *run, struct kvm_vcpu *vcpu,
 			vcpu->arch.mmu.tlbie(vcpu, addr, large);
 			break;
 		}
-#ifdef CONFIG_KVM_BOOK3S_64_PR
+#ifdef CONFIG_PPC_BOOK3S_64
 		case OP_31_XOP_FAKE_SC1:
 		{
 			/* SC 1 papr hypercalls */
@@ -267,12 +267,9 @@ int kvmppc_core_emulate_op(struct kvm_run *run, struct kvm_vcpu *vcpu,
 
 			r = kvmppc_st(vcpu, &addr, 32, zeros, true);
 			if ((r == -ENOENT) || (r == -EPERM)) {
-				struct kvmppc_book3s_shadow_vcpu *svcpu;
-
-				svcpu = svcpu_get(vcpu);
 				*advance = 0;
 				vcpu->arch.shared->dar = vaddr;
-				svcpu->fault_dar = vaddr;
+				vcpu->arch.fault_dar = vaddr;
 
 				dsisr = DSISR_ISSTORE;
 				if (r == -ENOENT)
@@ -281,8 +278,7 @@ int kvmppc_core_emulate_op(struct kvm_run *run, struct kvm_vcpu *vcpu,
 					dsisr |= DSISR_PROTFAULT;
 
 				vcpu->arch.shared->dsisr = dsisr;
-				svcpu->fault_dsisr = dsisr;
-				svcpu_put(svcpu);
+				vcpu->arch.fault_dsisr = dsisr;
 
 				kvmppc_book3s_queue_irqprio(vcpu,
 					BOOK3S_INTERRUPT_DATA_STORAGE);
@@ -349,7 +345,7 @@ static struct kvmppc_bat *kvmppc_find_bat(struct kvm_vcpu *vcpu, int sprn)
 	return bat;
 }
 
-int kvmppc_core_emulate_mtspr(struct kvm_vcpu *vcpu, int sprn, ulong spr_val)
+int kvmppc_core_emulate_mtspr_pr(struct kvm_vcpu *vcpu, int sprn, ulong spr_val)
 {
 	int emulated = EMULATE_DONE;
 
@@ -472,7 +468,7 @@ unprivileged:
 	return emulated;
 }
 
-int kvmppc_core_emulate_mfspr(struct kvm_vcpu *vcpu, int sprn, ulong *spr_val)
+int kvmppc_core_emulate_mfspr_pr(struct kvm_vcpu *vcpu, int sprn, ulong *spr_val)
 {
 	int emulated = EMULATE_DONE;
 
diff --git a/arch/powerpc/kvm/book3s_exports.c b/arch/powerpc/kvm/book3s_exports.c
index 7057a02f0906..852989a9bad3 100644
--- a/arch/powerpc/kvm/book3s_exports.c
+++ b/arch/powerpc/kvm/book3s_exports.c
@@ -20,9 +20,10 @@
 #include <linux/export.h>
 #include <asm/kvm_book3s.h>
 
-#ifdef CONFIG_KVM_BOOK3S_64_HV
+#ifdef CONFIG_KVM_BOOK3S_HV_POSSIBLE
 EXPORT_SYMBOL_GPL(kvmppc_hv_entry_trampoline);
-#else
+#endif
+#ifdef CONFIG_KVM_BOOK3S_PR_POSSIBLE
 EXPORT_SYMBOL_GPL(kvmppc_entry_trampoline);
 EXPORT_SYMBOL_GPL(kvmppc_load_up_fpu);
 #ifdef CONFIG_ALTIVEC
diff --git a/arch/powerpc/kvm/book3s_hv.c b/arch/powerpc/kvm/book3s_hv.c
index 62a2b5ab08ed..072287f1c3bc 100644
--- a/arch/powerpc/kvm/book3s_hv.c
+++ b/arch/powerpc/kvm/book3s_hv.c
@@ -52,6 +52,9 @@
 #include <linux/vmalloc.h>
 #include <linux/highmem.h>
 #include <linux/hugetlb.h>
+#include <linux/module.h>
+
+#include "book3s.h"
 
 /* #define EXIT_DEBUG */
 /* #define EXIT_DEBUG_SIMPLE */
@@ -66,7 +69,7 @@
 static void kvmppc_end_cede(struct kvm_vcpu *vcpu);
 static int kvmppc_hv_setup_htab_rma(struct kvm_vcpu *vcpu);
 
-void kvmppc_fast_vcpu_kick(struct kvm_vcpu *vcpu)
+static void kvmppc_fast_vcpu_kick_hv(struct kvm_vcpu *vcpu)
 {
 	int me;
 	int cpu = vcpu->cpu;
@@ -125,7 +128,7 @@ void kvmppc_fast_vcpu_kick(struct kvm_vcpu *vcpu)
  * purely defensive; they should never fail.)
  */
 
-void kvmppc_core_vcpu_load(struct kvm_vcpu *vcpu, int cpu)
+static void kvmppc_core_vcpu_load_hv(struct kvm_vcpu *vcpu, int cpu)
 {
 	struct kvmppc_vcore *vc = vcpu->arch.vcore;
 
@@ -143,7 +146,7 @@ void kvmppc_core_vcpu_load(struct kvm_vcpu *vcpu, int cpu)
 	spin_unlock(&vcpu->arch.tbacct_lock);
 }
 
-void kvmppc_core_vcpu_put(struct kvm_vcpu *vcpu)
+static void kvmppc_core_vcpu_put_hv(struct kvm_vcpu *vcpu)
 {
 	struct kvmppc_vcore *vc = vcpu->arch.vcore;
 
@@ -155,17 +158,46 @@ void kvmppc_core_vcpu_put(struct kvm_vcpu *vcpu)
 	spin_unlock(&vcpu->arch.tbacct_lock);
 }
 
-void kvmppc_set_msr(struct kvm_vcpu *vcpu, u64 msr)
+static void kvmppc_set_msr_hv(struct kvm_vcpu *vcpu, u64 msr)
 {
 	vcpu->arch.shregs.msr = msr;
 	kvmppc_end_cede(vcpu);
 }
 
-void kvmppc_set_pvr(struct kvm_vcpu *vcpu, u32 pvr)
+void kvmppc_set_pvr_hv(struct kvm_vcpu *vcpu, u32 pvr)
 {
 	vcpu->arch.pvr = pvr;
 }
 
+int kvmppc_set_arch_compat(struct kvm_vcpu *vcpu, u32 arch_compat)
+{
+	unsigned long pcr = 0;
+	struct kvmppc_vcore *vc = vcpu->arch.vcore;
+
+	if (arch_compat) {
+		if (!cpu_has_feature(CPU_FTR_ARCH_206))
+			return -EINVAL;	/* 970 has no compat mode support */
+
+		switch (arch_compat) {
+		case PVR_ARCH_205:
+			pcr = PCR_ARCH_205;
+			break;
+		case PVR_ARCH_206:
+		case PVR_ARCH_206p:
+			break;
+		default:
+			return -EINVAL;
+		}
+	}
+
+	spin_lock(&vc->lock);
+	vc->arch_compat = arch_compat;
+	vc->pcr = pcr;
+	spin_unlock(&vc->lock);
+
+	return 0;
+}
+
 void kvmppc_dump_regs(struct kvm_vcpu *vcpu)
 {
 	int r;
@@ -195,7 +227,7 @@ void kvmppc_dump_regs(struct kvm_vcpu *vcpu)
 		pr_err("  ESID = %.16llx VSID = %.16llx\n",
 		       vcpu->arch.slb[r].orige, vcpu->arch.slb[r].origv);
 	pr_err("lpcr = %.16lx sdr1 = %.16lx last_inst = %.8x\n",
-	       vcpu->kvm->arch.lpcr, vcpu->kvm->arch.sdr1,
+	       vcpu->arch.vcore->lpcr, vcpu->kvm->arch.sdr1,
 	       vcpu->arch.last_inst);
 }
 
@@ -489,7 +521,7 @@ static void kvmppc_create_dtl_entry(struct kvm_vcpu *vcpu,
 	memset(dt, 0, sizeof(struct dtl_entry));
 	dt->dispatch_reason = 7;
 	dt->processor_id = vc->pcpu + vcpu->arch.ptid;
-	dt->timebase = now;
+	dt->timebase = now + vc->tb_offset;
 	dt->enqueue_to_dispatch_time = stolen;
 	dt->srr0 = kvmppc_get_pc(vcpu);
 	dt->srr1 = vcpu->arch.shregs.msr;
@@ -538,6 +570,15 @@ int kvmppc_pseries_do_hcall(struct kvm_vcpu *vcpu)
 		}
 		break;
 	case H_CONFER:
+		target = kvmppc_get_gpr(vcpu, 4);
+		if (target == -1)
+			break;
+		tvcpu = kvmppc_find_vcpu(vcpu->kvm, target);
+		if (!tvcpu) {
+			ret = H_PARAMETER;
+			break;
+		}
+		kvm_vcpu_yield_to(tvcpu);
 		break;
 	case H_REGISTER_VPA:
 		ret = do_h_register_vpa(vcpu, kvmppc_get_gpr(vcpu, 4),
@@ -576,8 +617,8 @@ int kvmppc_pseries_do_hcall(struct kvm_vcpu *vcpu)
 	return RESUME_GUEST;
 }
 
-static int kvmppc_handle_exit(struct kvm_run *run, struct kvm_vcpu *vcpu,
-			      struct task_struct *tsk)
+static int kvmppc_handle_exit_hv(struct kvm_run *run, struct kvm_vcpu *vcpu,
+				 struct task_struct *tsk)
 {
 	int r = RESUME_HOST;
 
@@ -671,16 +712,16 @@ static int kvmppc_handle_exit(struct kvm_run *run, struct kvm_vcpu *vcpu,
 		printk(KERN_EMERG "trap=0x%x | pc=0x%lx | msr=0x%llx\n",
 			vcpu->arch.trap, kvmppc_get_pc(vcpu),
 			vcpu->arch.shregs.msr);
+		run->hw.hardware_exit_reason = vcpu->arch.trap;
 		r = RESUME_HOST;
-		BUG();
 		break;
 	}
 
 	return r;
 }
 
-int kvm_arch_vcpu_ioctl_get_sregs(struct kvm_vcpu *vcpu,
-				  struct kvm_sregs *sregs)
+static int kvm_arch_vcpu_ioctl_get_sregs_hv(struct kvm_vcpu *vcpu,
+					    struct kvm_sregs *sregs)
 {
 	int i;
 
@@ -694,12 +735,12 @@ int kvm_arch_vcpu_ioctl_get_sregs(struct kvm_vcpu *vcpu,
 	return 0;
 }
 
-int kvm_arch_vcpu_ioctl_set_sregs(struct kvm_vcpu *vcpu,
-				  struct kvm_sregs *sregs)
+static int kvm_arch_vcpu_ioctl_set_sregs_hv(struct kvm_vcpu *vcpu,
+					    struct kvm_sregs *sregs)
 {
 	int i, j;
 
-	kvmppc_set_pvr(vcpu, sregs->pvr);
+	kvmppc_set_pvr_hv(vcpu, sregs->pvr);
 
 	j = 0;
 	for (i = 0; i < vcpu->arch.slb_nr; i++) {
@@ -714,7 +755,23 @@ int kvm_arch_vcpu_ioctl_set_sregs(struct kvm_vcpu *vcpu,
 	return 0;
 }
 
-int kvmppc_get_one_reg(struct kvm_vcpu *vcpu, u64 id, union kvmppc_one_reg *val)
+static void kvmppc_set_lpcr(struct kvm_vcpu *vcpu, u64 new_lpcr)
+{
+	struct kvmppc_vcore *vc = vcpu->arch.vcore;
+	u64 mask;
+
+	spin_lock(&vc->lock);
+	/*
+	 * Userspace can only modify DPFD (default prefetch depth),
+	 * ILE (interrupt little-endian) and TC (translation control).
+	 */
+	mask = LPCR_DPFD | LPCR_ILE | LPCR_TC;
+	vc->lpcr = (vc->lpcr & ~mask) | (new_lpcr & mask);
+	spin_unlock(&vc->lock);
+}
+
+static int kvmppc_get_one_reg_hv(struct kvm_vcpu *vcpu, u64 id,
+				 union kvmppc_one_reg *val)
 {
 	int r = 0;
 	long int i;
@@ -749,6 +806,12 @@ int kvmppc_get_one_reg(struct kvm_vcpu *vcpu, u64 id, union kvmppc_one_reg *val)
 		i = id - KVM_REG_PPC_PMC1;
 		*val = get_reg_val(id, vcpu->arch.pmc[i]);
 		break;
+	case KVM_REG_PPC_SIAR:
+		*val = get_reg_val(id, vcpu->arch.siar);
+		break;
+	case KVM_REG_PPC_SDAR:
+		*val = get_reg_val(id, vcpu->arch.sdar);
+		break;
 #ifdef CONFIG_VSX
 	case KVM_REG_PPC_FPR0 ... KVM_REG_PPC_FPR31:
 		if (cpu_has_feature(CPU_FTR_VSX)) {
@@ -787,6 +850,18 @@ int kvmppc_get_one_reg(struct kvm_vcpu *vcpu, u64 id, union kvmppc_one_reg *val)
 		val->vpaval.length = vcpu->arch.dtl.len;
 		spin_unlock(&vcpu->arch.vpa_update_lock);
 		break;
+	case KVM_REG_PPC_TB_OFFSET:
+		*val = get_reg_val(id, vcpu->arch.vcore->tb_offset);
+		break;
+	case KVM_REG_PPC_LPCR:
+		*val = get_reg_val(id, vcpu->arch.vcore->lpcr);
+		break;
+	case KVM_REG_PPC_PPR:
+		*val = get_reg_val(id, vcpu->arch.ppr);
+		break;
+	case KVM_REG_PPC_ARCH_COMPAT:
+		*val = get_reg_val(id, vcpu->arch.vcore->arch_compat);
+		break;
 	default:
 		r = -EINVAL;
 		break;
@@ -795,7 +870,8 @@ int kvmppc_get_one_reg(struct kvm_vcpu *vcpu, u64 id, union kvmppc_one_reg *val)
 	return r;
 }
 
-int kvmppc_set_one_reg(struct kvm_vcpu *vcpu, u64 id, union kvmppc_one_reg *val)
+static int kvmppc_set_one_reg_hv(struct kvm_vcpu *vcpu, u64 id,
+				 union kvmppc_one_reg *val)
 {
 	int r = 0;
 	long int i;
@@ -833,6 +909,12 @@ int kvmppc_set_one_reg(struct kvm_vcpu *vcpu, u64 id, union kvmppc_one_reg *val)
 		i = id - KVM_REG_PPC_PMC1;
 		vcpu->arch.pmc[i] = set_reg_val(id, *val);
 		break;
+	case KVM_REG_PPC_SIAR:
+		vcpu->arch.siar = set_reg_val(id, *val);
+		break;
+	case KVM_REG_PPC_SDAR:
+		vcpu->arch.sdar = set_reg_val(id, *val);
+		break;
 #ifdef CONFIG_VSX
 	case KVM_REG_PPC_FPR0 ... KVM_REG_PPC_FPR31:
 		if (cpu_has_feature(CPU_FTR_VSX)) {
@@ -880,6 +962,20 @@ int kvmppc_set_one_reg(struct kvm_vcpu *vcpu, u64 id, union kvmppc_one_reg *val)
 		len -= len % sizeof(struct dtl_entry);
 		r = set_vpa(vcpu, &vcpu->arch.dtl, addr, len);
 		break;
+	case KVM_REG_PPC_TB_OFFSET:
+		/* round up to multiple of 2^24 */
+		vcpu->arch.vcore->tb_offset =
+			ALIGN(set_reg_val(id, *val), 1UL << 24);
+		break;
+	case KVM_REG_PPC_LPCR:
+		kvmppc_set_lpcr(vcpu, set_reg_val(id, *val));
+		break;
+	case KVM_REG_PPC_PPR:
+		vcpu->arch.ppr = set_reg_val(id, *val);
+		break;
+	case KVM_REG_PPC_ARCH_COMPAT:
+		r = kvmppc_set_arch_compat(vcpu, set_reg_val(id, *val));
+		break;
 	default:
 		r = -EINVAL;
 		break;
@@ -888,14 +984,8 @@ int kvmppc_set_one_reg(struct kvm_vcpu *vcpu, u64 id, union kvmppc_one_reg *val)
 	return r;
 }
 
-int kvmppc_core_check_processor_compat(void)
-{
-	if (cpu_has_feature(CPU_FTR_HVMODE))
-		return 0;
-	return -EIO;
-}
-
-struct kvm_vcpu *kvmppc_core_vcpu_create(struct kvm *kvm, unsigned int id)
+static struct kvm_vcpu *kvmppc_core_vcpu_create_hv(struct kvm *kvm,
+						   unsigned int id)
 {
 	struct kvm_vcpu *vcpu;
 	int err = -EINVAL;
@@ -919,8 +1009,7 @@ struct kvm_vcpu *kvmppc_core_vcpu_create(struct kvm *kvm, unsigned int id)
 	vcpu->arch.mmcr[0] = MMCR0_FC;
 	vcpu->arch.ctrl = CTRL_RUNLATCH;
 	/* default to host PVR, since we can't spoof it */
-	vcpu->arch.pvr = mfspr(SPRN_PVR);
-	kvmppc_set_pvr(vcpu, vcpu->arch.pvr);
+	kvmppc_set_pvr_hv(vcpu, mfspr(SPRN_PVR));
 	spin_lock_init(&vcpu->arch.vpa_update_lock);
 	spin_lock_init(&vcpu->arch.tbacct_lock);
 	vcpu->arch.busy_preempt = TB_NIL;
@@ -940,6 +1029,7 @@ struct kvm_vcpu *kvmppc_core_vcpu_create(struct kvm *kvm, unsigned int id)
 			spin_lock_init(&vcore->lock);
 			init_waitqueue_head(&vcore->wq);
 			vcore->preempt_tb = TB_NIL;
+			vcore->lpcr = kvm->arch.lpcr;
 		}
 		kvm->arch.vcores[core] = vcore;
 		kvm->arch.online_vcores++;
@@ -972,7 +1062,7 @@ static void unpin_vpa(struct kvm *kvm, struct kvmppc_vpa *vpa)
 					vpa->dirty);
 }
 
-void kvmppc_core_vcpu_free(struct kvm_vcpu *vcpu)
+static void kvmppc_core_vcpu_free_hv(struct kvm_vcpu *vcpu)
 {
 	spin_lock(&vcpu->arch.vpa_update_lock);
 	unpin_vpa(vcpu->kvm, &vcpu->arch.dtl);
@@ -983,6 +1073,12 @@ void kvmppc_core_vcpu_free(struct kvm_vcpu *vcpu)
 	kmem_cache_free(kvm_vcpu_cache, vcpu);
 }
 
+static int kvmppc_core_check_requests_hv(struct kvm_vcpu *vcpu)
+{
+	/* Indicate we want to get back into the guest */
+	return 1;
+}
+
 static void kvmppc_set_timer(struct kvm_vcpu *vcpu)
 {
 	unsigned long dec_nsec, now;
@@ -1264,8 +1360,8 @@ static void kvmppc_run_core(struct kvmppc_vcore *vc)
 
 		ret = RESUME_GUEST;
 		if (vcpu->arch.trap)
-			ret = kvmppc_handle_exit(vcpu->arch.kvm_run, vcpu,
-						 vcpu->arch.run_task);
+			ret = kvmppc_handle_exit_hv(vcpu->arch.kvm_run, vcpu,
+						    vcpu->arch.run_task);
 
 		vcpu->arch.ret = ret;
 		vcpu->arch.trap = 0;
@@ -1424,7 +1520,7 @@ static int kvmppc_run_vcpu(struct kvm_run *kvm_run, struct kvm_vcpu *vcpu)
 	return vcpu->arch.ret;
 }
 
-int kvmppc_vcpu_run(struct kvm_run *run, struct kvm_vcpu *vcpu)
+static int kvmppc_vcpu_run_hv(struct kvm_run *run, struct kvm_vcpu *vcpu)
 {
 	int r;
 	int srcu_idx;
@@ -1546,7 +1642,8 @@ static const struct file_operations kvm_rma_fops = {
 	.release	= kvm_rma_release,
 };
 
-long kvm_vm_ioctl_allocate_rma(struct kvm *kvm, struct kvm_allocate_rma *ret)
+static long kvm_vm_ioctl_allocate_rma(struct kvm *kvm,
+				      struct kvm_allocate_rma *ret)
 {
 	long fd;
 	struct kvm_rma_info *ri;
@@ -1592,7 +1689,8 @@ static void kvmppc_add_seg_page_size(struct kvm_ppc_one_seg_page_size **sps,
 	(*sps)++;
 }
 
-int kvm_vm_ioctl_get_smmu_info(struct kvm *kvm, struct kvm_ppc_smmu_info *info)
+static int kvm_vm_ioctl_get_smmu_info_hv(struct kvm *kvm,
+					 struct kvm_ppc_smmu_info *info)
 {
 	struct kvm_ppc_one_seg_page_size *sps;
 
@@ -1613,7 +1711,8 @@ int kvm_vm_ioctl_get_smmu_info(struct kvm *kvm, struct kvm_ppc_smmu_info *info)
 /*
  * Get (and clear) the dirty memory log for a memory slot.
  */
-int kvm_vm_ioctl_get_dirty_log(struct kvm *kvm, struct kvm_dirty_log *log)
+static int kvm_vm_ioctl_get_dirty_log_hv(struct kvm *kvm,
+					 struct kvm_dirty_log *log)
 {
 	struct kvm_memory_slot *memslot;
 	int r;
@@ -1667,8 +1766,8 @@ static void unpin_slot(struct kvm_memory_slot *memslot)
 	}
 }
 
-void kvmppc_core_free_memslot(struct kvm_memory_slot *free,
-			      struct kvm_memory_slot *dont)
+static void kvmppc_core_free_memslot_hv(struct kvm_memory_slot *free,
+					struct kvm_memory_slot *dont)
 {
 	if (!dont || free->arch.rmap != dont->arch.rmap) {
 		vfree(free->arch.rmap);
@@ -1681,8 +1780,8 @@ void kvmppc_core_free_memslot(struct kvm_memory_slot *free,
 	}
 }
 
-int kvmppc_core_create_memslot(struct kvm_memory_slot *slot,
-			       unsigned long npages)
+static int kvmppc_core_create_memslot_hv(struct kvm_memory_slot *slot,
+					 unsigned long npages)
 {
 	slot->arch.rmap = vzalloc(npages * sizeof(*slot->arch.rmap));
 	if (!slot->arch.rmap)
@@ -1692,9 +1791,9 @@ int kvmppc_core_create_memslot(struct kvm_memory_slot *slot,
 	return 0;
 }
 
-int kvmppc_core_prepare_memory_region(struct kvm *kvm,
-				      struct kvm_memory_slot *memslot,
-				      struct kvm_userspace_memory_region *mem)
+static int kvmppc_core_prepare_memory_region_hv(struct kvm *kvm,
+					struct kvm_memory_slot *memslot,
+					struct kvm_userspace_memory_region *mem)
 {
 	unsigned long *phys;
 
@@ -1710,9 +1809,9 @@ int kvmppc_core_prepare_memory_region(struct kvm *kvm,
 	return 0;
 }
 
-void kvmppc_core_commit_memory_region(struct kvm *kvm,
-				      struct kvm_userspace_memory_region *mem,
-				      const struct kvm_memory_slot *old)
+static void kvmppc_core_commit_memory_region_hv(struct kvm *kvm,
+				struct kvm_userspace_memory_region *mem,
+				const struct kvm_memory_slot *old)
 {
 	unsigned long npages = mem->memory_size >> PAGE_SHIFT;
 	struct kvm_memory_slot *memslot;
@@ -1729,6 +1828,37 @@ void kvmppc_core_commit_memory_region(struct kvm *kvm,
 	}
 }
 
+/*
+ * Update LPCR values in kvm->arch and in vcores.
+ * Caller must hold kvm->lock.
+ */
+void kvmppc_update_lpcr(struct kvm *kvm, unsigned long lpcr, unsigned long mask)
+{
+	long int i;
+	u32 cores_done = 0;
+
+	if ((kvm->arch.lpcr & mask) == lpcr)
+		return;
+
+	kvm->arch.lpcr = (kvm->arch.lpcr & ~mask) | lpcr;
+
+	for (i = 0; i < KVM_MAX_VCORES; ++i) {
+		struct kvmppc_vcore *vc = kvm->arch.vcores[i];
+		if (!vc)
+			continue;
+		spin_lock(&vc->lock);
+		vc->lpcr = (vc->lpcr & ~mask) | lpcr;
+		spin_unlock(&vc->lock);
+		if (++cores_done >= kvm->arch.online_vcores)
+			break;
+	}
+}
+
+static void kvmppc_mmu_destroy_hv(struct kvm_vcpu *vcpu)
+{
+	return;
+}
+
 static int kvmppc_hv_setup_htab_rma(struct kvm_vcpu *vcpu)
 {
 	int err = 0;
@@ -1737,7 +1867,8 @@ static int kvmppc_hv_setup_htab_rma(struct kvm_vcpu *vcpu)
 	unsigned long hva;
 	struct kvm_memory_slot *memslot;
 	struct vm_area_struct *vma;
-	unsigned long lpcr, senc;
+	unsigned long lpcr = 0, senc;
+	unsigned long lpcr_mask = 0;
 	unsigned long psize, porder;
 	unsigned long rma_size;
 	unsigned long rmls;
@@ -1802,9 +1933,9 @@ static int kvmppc_hv_setup_htab_rma(struct kvm_vcpu *vcpu)
 		senc = slb_pgsize_encoding(psize);
 		kvm->arch.vrma_slb_v = senc | SLB_VSID_B_1T |
 			(VRMA_VSID << SLB_VSID_SHIFT_1T);
-		lpcr = kvm->arch.lpcr & ~LPCR_VRMASD;
-		lpcr |= senc << (LPCR_VRMASD_SH - 4);
-		kvm->arch.lpcr = lpcr;
+		lpcr_mask = LPCR_VRMASD;
+		/* the -4 is to account for senc values starting at 0x10 */
+		lpcr = senc << (LPCR_VRMASD_SH - 4);
 
 		/* Create HPTEs in the hash page table for the VRMA */
 		kvmppc_map_vrma(vcpu, memslot, porder);
@@ -1825,23 +1956,21 @@ static int kvmppc_hv_setup_htab_rma(struct kvm_vcpu *vcpu)
 		kvm->arch.rma = ri;
 
 		/* Update LPCR and RMOR */
-		lpcr = kvm->arch.lpcr;
 		if (cpu_has_feature(CPU_FTR_ARCH_201)) {
 			/* PPC970; insert RMLS value (split field) in HID4 */
-			lpcr &= ~((1ul << HID4_RMLS0_SH) |
-				  (3ul << HID4_RMLS2_SH));
-			lpcr |= ((rmls >> 2) << HID4_RMLS0_SH) |
+			lpcr_mask = (1ul << HID4_RMLS0_SH) |
+				(3ul << HID4_RMLS2_SH) | HID4_RMOR;
+			lpcr = ((rmls >> 2) << HID4_RMLS0_SH) |
 				((rmls & 3) << HID4_RMLS2_SH);
 			/* RMOR is also in HID4 */
 			lpcr |= ((ri->base_pfn >> (26 - PAGE_SHIFT)) & 0xffff)
 				<< HID4_RMOR_SH;
 		} else {
 			/* POWER7 */
-			lpcr &= ~(LPCR_VPM0 | LPCR_VRMA_L);
-			lpcr |= rmls << LPCR_RMLS_SH;
+			lpcr_mask = LPCR_VPM0 | LPCR_VRMA_L | LPCR_RMLS;
+			lpcr = rmls << LPCR_RMLS_SH;
 			kvm->arch.rmor = ri->base_pfn << PAGE_SHIFT;
 		}
-		kvm->arch.lpcr = lpcr;
 		pr_info("KVM: Using RMO at %lx size %lx (LPCR = %lx)\n",
 			ri->base_pfn << PAGE_SHIFT, rma_size, lpcr);
 
@@ -1860,6 +1989,8 @@ static int kvmppc_hv_setup_htab_rma(struct kvm_vcpu *vcpu)
 		}
 	}
 
+	kvmppc_update_lpcr(kvm, lpcr, lpcr_mask);
+
 	/* Order updates to kvm->arch.lpcr etc. vs. rma_setup_done */
 	smp_wmb();
 	kvm->arch.rma_setup_done = 1;
@@ -1875,7 +2006,7 @@ static int kvmppc_hv_setup_htab_rma(struct kvm_vcpu *vcpu)
 	goto out_srcu;
 }
 
-int kvmppc_core_init_vm(struct kvm *kvm)
+static int kvmppc_core_init_vm_hv(struct kvm *kvm)
 {
 	unsigned long lpcr, lpid;
 
@@ -1893,9 +2024,6 @@ int kvmppc_core_init_vm(struct kvm *kvm)
 	 */
 	cpumask_setall(&kvm->arch.need_tlb_flush);
 
-	INIT_LIST_HEAD(&kvm->arch.spapr_tce_tables);
-	INIT_LIST_HEAD(&kvm->arch.rtas_tokens);
-
 	kvm->arch.rma = NULL;
 
 	kvm->arch.host_sdr1 = mfspr(SPRN_SDR1);
@@ -1931,61 +2059,162 @@ int kvmppc_core_init_vm(struct kvm *kvm)
 	return 0;
 }
 
-void kvmppc_core_destroy_vm(struct kvm *kvm)
+static void kvmppc_free_vcores(struct kvm *kvm)
+{
+	long int i;
+
+	for (i = 0; i < KVM_MAX_VCORES; ++i)
+		kfree(kvm->arch.vcores[i]);
+	kvm->arch.online_vcores = 0;
+}
+
+static void kvmppc_core_destroy_vm_hv(struct kvm *kvm)
 {
 	uninhibit_secondary_onlining();
 
+	kvmppc_free_vcores(kvm);
 	if (kvm->arch.rma) {
 		kvm_release_rma(kvm->arch.rma);
 		kvm->arch.rma = NULL;
 	}
 
-	kvmppc_rtas_tokens_free(kvm);
-
 	kvmppc_free_hpt(kvm);
-	WARN_ON(!list_empty(&kvm->arch.spapr_tce_tables));
 }
 
-/* These are stubs for now */
-void kvmppc_mmu_pte_pflush(struct kvm_vcpu *vcpu, ulong pa_start, ulong pa_end)
+/* We don't need to emulate any privileged instructions or dcbz */
+static int kvmppc_core_emulate_op_hv(struct kvm_run *run, struct kvm_vcpu *vcpu,
+				     unsigned int inst, int *advance)
 {
+	return EMULATE_FAIL;
 }
 
-/* We don't need to emulate any privileged instructions or dcbz */
-int kvmppc_core_emulate_op(struct kvm_run *run, struct kvm_vcpu *vcpu,
-                           unsigned int inst, int *advance)
+static int kvmppc_core_emulate_mtspr_hv(struct kvm_vcpu *vcpu, int sprn,
+					ulong spr_val)
 {
 	return EMULATE_FAIL;
 }
 
-int kvmppc_core_emulate_mtspr(struct kvm_vcpu *vcpu, int sprn, ulong spr_val)
+static int kvmppc_core_emulate_mfspr_hv(struct kvm_vcpu *vcpu, int sprn,
+					ulong *spr_val)
 {
 	return EMULATE_FAIL;
 }
 
-int kvmppc_core_emulate_mfspr(struct kvm_vcpu *vcpu, int sprn, ulong *spr_val)
+static int kvmppc_core_check_processor_compat_hv(void)
 {
-	return EMULATE_FAIL;
+	if (!cpu_has_feature(CPU_FTR_HVMODE))
+		return -EIO;
+	return 0;
 }
 
-static int kvmppc_book3s_hv_init(void)
+static long kvm_arch_vm_ioctl_hv(struct file *filp,
+				 unsigned int ioctl, unsigned long arg)
 {
-	int r;
+	struct kvm *kvm __maybe_unused = filp->private_data;
+	void __user *argp = (void __user *)arg;
+	long r;
 
-	r = kvm_init(NULL, sizeof(struct kvm_vcpu), 0, THIS_MODULE);
+	switch (ioctl) {
 
-	if (r)
+	case KVM_ALLOCATE_RMA: {
+		struct kvm_allocate_rma rma;
+		struct kvm *kvm = filp->private_data;
+
+		r = kvm_vm_ioctl_allocate_rma(kvm, &rma);
+		if (r >= 0 && copy_to_user(argp, &rma, sizeof(rma)))
+			r = -EFAULT;
+		break;
+	}
+
+	case KVM_PPC_ALLOCATE_HTAB: {
+		u32 htab_order;
+
+		r = -EFAULT;
+		if (get_user(htab_order, (u32 __user *)argp))
+			break;
+		r = kvmppc_alloc_reset_hpt(kvm, &htab_order);
+		if (r)
+			break;
+		r = -EFAULT;
+		if (put_user(htab_order, (u32 __user *)argp))
+			break;
+		r = 0;
+		break;
+	}
+
+	case KVM_PPC_GET_HTAB_FD: {
+		struct kvm_get_htab_fd ghf;
+
+		r = -EFAULT;
+		if (copy_from_user(&ghf, argp, sizeof(ghf)))
+			break;
+		r = kvm_vm_ioctl_get_htab_fd(kvm, &ghf);
+		break;
+	}
+
+	default:
+		r = -ENOTTY;
+	}
+
+	return r;
+}
+
+static struct kvmppc_ops kvm_ops_hv = {
+	.get_sregs = kvm_arch_vcpu_ioctl_get_sregs_hv,
+	.set_sregs = kvm_arch_vcpu_ioctl_set_sregs_hv,
+	.get_one_reg = kvmppc_get_one_reg_hv,
+	.set_one_reg = kvmppc_set_one_reg_hv,
+	.vcpu_load   = kvmppc_core_vcpu_load_hv,
+	.vcpu_put    = kvmppc_core_vcpu_put_hv,
+	.set_msr     = kvmppc_set_msr_hv,
+	.vcpu_run    = kvmppc_vcpu_run_hv,
+	.vcpu_create = kvmppc_core_vcpu_create_hv,
+	.vcpu_free   = kvmppc_core_vcpu_free_hv,
+	.check_requests = kvmppc_core_check_requests_hv,
+	.get_dirty_log  = kvm_vm_ioctl_get_dirty_log_hv,
+	.flush_memslot  = kvmppc_core_flush_memslot_hv,
+	.prepare_memory_region = kvmppc_core_prepare_memory_region_hv,
+	.commit_memory_region  = kvmppc_core_commit_memory_region_hv,
+	.unmap_hva = kvm_unmap_hva_hv,
+	.unmap_hva_range = kvm_unmap_hva_range_hv,
+	.age_hva  = kvm_age_hva_hv,
+	.test_age_hva = kvm_test_age_hva_hv,
+	.set_spte_hva = kvm_set_spte_hva_hv,
+	.mmu_destroy  = kvmppc_mmu_destroy_hv,
+	.free_memslot = kvmppc_core_free_memslot_hv,
+	.create_memslot = kvmppc_core_create_memslot_hv,
+	.init_vm =  kvmppc_core_init_vm_hv,
+	.destroy_vm = kvmppc_core_destroy_vm_hv,
+	.get_smmu_info = kvm_vm_ioctl_get_smmu_info_hv,
+	.emulate_op = kvmppc_core_emulate_op_hv,
+	.emulate_mtspr = kvmppc_core_emulate_mtspr_hv,
+	.emulate_mfspr = kvmppc_core_emulate_mfspr_hv,
+	.fast_vcpu_kick = kvmppc_fast_vcpu_kick_hv,
+	.arch_vm_ioctl  = kvm_arch_vm_ioctl_hv,
+};
+
+static int kvmppc_book3s_init_hv(void)
+{
+	int r;
+	/*
+	 * FIXME!! Do we need to check on all cpus ?
+	 */
+	r = kvmppc_core_check_processor_compat_hv();
+	if (r < 0)
 		return r;
 
-	r = kvmppc_mmu_hv_init();
+	kvm_ops_hv.owner = THIS_MODULE;
+	kvmppc_hv_ops = &kvm_ops_hv;
 
+	r = kvmppc_mmu_hv_init();
 	return r;
 }
 
-static void kvmppc_book3s_hv_exit(void)
+static void kvmppc_book3s_exit_hv(void)
 {
-	kvm_exit();
+	kvmppc_hv_ops = NULL;
 }
 
-module_init(kvmppc_book3s_hv_init);
-module_exit(kvmppc_book3s_hv_exit);
+module_init(kvmppc_book3s_init_hv);
+module_exit(kvmppc_book3s_exit_hv);
+MODULE_LICENSE("GPL");
diff --git a/arch/powerpc/kvm/book3s_hv_interrupts.S b/arch/powerpc/kvm/book3s_hv_interrupts.S
index 37f1cc417ca0..928142c64cb0 100644
--- a/arch/powerpc/kvm/book3s_hv_interrupts.S
+++ b/arch/powerpc/kvm/book3s_hv_interrupts.S
@@ -158,9 +158,6 @@ END_FTR_SECTION_IFSET(CPU_FTR_ARCH_201)
  * Interrupts are enabled again at this point.
  */
 
-.global kvmppc_handler_highmem
-kvmppc_handler_highmem:
-
 	/*
 	 * Register usage at this point:
 	 *
diff --git a/arch/powerpc/kvm/book3s_hv_rmhandlers.S b/arch/powerpc/kvm/book3s_hv_rmhandlers.S
index c71103b8a748..bc8de75b1925 100644
--- a/arch/powerpc/kvm/book3s_hv_rmhandlers.S
+++ b/arch/powerpc/kvm/book3s_hv_rmhandlers.S
@@ -33,30 +33,6 @@
 #error Need to fix lppaca and SLB shadow accesses in little endian mode
 #endif
 
-/*****************************************************************************
- *                                                                           *
- *        Real Mode handlers that need to be in the linear mapping           *
- *                                                                           *
- ****************************************************************************/
-
-	.globl	kvmppc_skip_interrupt
-kvmppc_skip_interrupt:
-	mfspr	r13,SPRN_SRR0
-	addi	r13,r13,4
-	mtspr	SPRN_SRR0,r13
-	GET_SCRATCH0(r13)
-	rfid
-	b	.
-
-	.globl	kvmppc_skip_Hinterrupt
-kvmppc_skip_Hinterrupt:
-	mfspr	r13,SPRN_HSRR0
-	addi	r13,r13,4
-	mtspr	SPRN_HSRR0,r13
-	GET_SCRATCH0(r13)
-	hrfid
-	b	.
-
 /*
  * Call kvmppc_hv_entry in real mode.
  * Must be called with interrupts hard-disabled.
@@ -66,8 +42,11 @@ kvmppc_skip_Hinterrupt:
  * LR = return address to continue at after eventually re-enabling MMU
  */
 _GLOBAL(kvmppc_hv_entry_trampoline)
+	mflr	r0
+	std	r0, PPC_LR_STKOFF(r1)
+	stdu	r1, -112(r1)
 	mfmsr	r10
-	LOAD_REG_ADDR(r5, kvmppc_hv_entry)
+	LOAD_REG_ADDR(r5, kvmppc_call_hv_entry)
 	li	r0,MSR_RI
 	andc	r0,r10,r0
 	li	r6,MSR_IR | MSR_DR
@@ -77,11 +56,103 @@ _GLOBAL(kvmppc_hv_entry_trampoline)
 	mtsrr1	r6
 	RFI
 
-/******************************************************************************
- *                                                                            *
- *                               Entry code                                   *
- *                                                                            *
- *****************************************************************************/
+kvmppc_call_hv_entry:
+	bl	kvmppc_hv_entry
+
+	/* Back from guest - restore host state and return to caller */
+
+	/* Restore host DABR and DABRX */
+	ld	r5,HSTATE_DABR(r13)
+	li	r6,7
+	mtspr	SPRN_DABR,r5
+	mtspr	SPRN_DABRX,r6
+
+	/* Restore SPRG3 */
+	ld	r3,PACA_SPRG3(r13)
+	mtspr	SPRN_SPRG3,r3
+
+	/*
+	 * Reload DEC.  HDEC interrupts were disabled when
+	 * we reloaded the host's LPCR value.
+	 */
+	ld	r3, HSTATE_DECEXP(r13)
+	mftb	r4
+	subf	r4, r4, r3
+	mtspr	SPRN_DEC, r4
+
+	/* Reload the host's PMU registers */
+	ld	r3, PACALPPACAPTR(r13)	/* is the host using the PMU? */
+	lbz	r4, LPPACA_PMCINUSE(r3)
+	cmpwi	r4, 0
+	beq	23f			/* skip if not */
+	lwz	r3, HSTATE_PMC(r13)
+	lwz	r4, HSTATE_PMC + 4(r13)
+	lwz	r5, HSTATE_PMC + 8(r13)
+	lwz	r6, HSTATE_PMC + 12(r13)
+	lwz	r8, HSTATE_PMC + 16(r13)
+	lwz	r9, HSTATE_PMC + 20(r13)
+BEGIN_FTR_SECTION
+	lwz	r10, HSTATE_PMC + 24(r13)
+	lwz	r11, HSTATE_PMC + 28(r13)
+END_FTR_SECTION_IFSET(CPU_FTR_ARCH_201)
+	mtspr	SPRN_PMC1, r3
+	mtspr	SPRN_PMC2, r4
+	mtspr	SPRN_PMC3, r5
+	mtspr	SPRN_PMC4, r6
+	mtspr	SPRN_PMC5, r8
+	mtspr	SPRN_PMC6, r9
+BEGIN_FTR_SECTION
+	mtspr	SPRN_PMC7, r10
+	mtspr	SPRN_PMC8, r11
+END_FTR_SECTION_IFSET(CPU_FTR_ARCH_201)
+	ld	r3, HSTATE_MMCR(r13)
+	ld	r4, HSTATE_MMCR + 8(r13)
+	ld	r5, HSTATE_MMCR + 16(r13)
+	mtspr	SPRN_MMCR1, r4
+	mtspr	SPRN_MMCRA, r5
+	mtspr	SPRN_MMCR0, r3
+	isync
+23:
+
+	/*
+	 * For external and machine check interrupts, we need
+	 * to call the Linux handler to process the interrupt.
+	 * We do that by jumping to absolute address 0x500 for
+	 * external interrupts, or the machine_check_fwnmi label
+	 * for machine checks (since firmware might have patched
+	 * the vector area at 0x200).  The [h]rfid at the end of the
+	 * handler will return to the book3s_hv_interrupts.S code.
+	 * For other interrupts we do the rfid to get back
+	 * to the book3s_hv_interrupts.S code here.
+	 */
+	ld	r8, 112+PPC_LR_STKOFF(r1)
+	addi	r1, r1, 112
+	ld	r7, HSTATE_HOST_MSR(r13)
+
+	cmpwi	cr1, r12, BOOK3S_INTERRUPT_MACHINE_CHECK
+	cmpwi	r12, BOOK3S_INTERRUPT_EXTERNAL
+BEGIN_FTR_SECTION
+	beq	11f
+END_FTR_SECTION_IFSET(CPU_FTR_ARCH_206)
+
+	/* RFI into the highmem handler, or branch to interrupt handler */
+	mfmsr	r6
+	li	r0, MSR_RI
+	andc	r6, r6, r0
+	mtmsrd	r6, 1			/* Clear RI in MSR */
+	mtsrr0	r8
+	mtsrr1	r7
+	beqa	0x500			/* external interrupt (PPC970) */
+	beq	cr1, 13f		/* machine check */
+	RFI
+
+	/* On POWER7, we have external interrupts set to use HSRR0/1 */
+11:	mtspr	SPRN_HSRR0, r8
+	mtspr	SPRN_HSRR1, r7
+	ba	0x500
+
+13:	b	machine_check_fwnmi
+
 
 /*
  * We come in here when wakened from nap mode on a secondary hw thread.
@@ -137,7 +208,7 @@ kvm_start_guest:
 	cmpdi	r4,0
 	/* if we have no vcpu to run, go back to sleep */
 	beq	kvm_no_guest
-	b	kvmppc_hv_entry
+	b	30f
 
 27:	/* XXX should handle hypervisor maintenance interrupts etc. here */
 	b	kvm_no_guest
@@ -147,6 +218,57 @@ kvm_start_guest:
 	stw	r8,HSTATE_SAVED_XIRR(r13)
 	b	kvm_no_guest
 
+30:	bl	kvmppc_hv_entry
+
+	/* Back from the guest, go back to nap */
+	/* Clear our vcpu pointer so we don't come back in early */
+	li	r0, 0
+	std	r0, HSTATE_KVM_VCPU(r13)
+	lwsync
+	/* Clear any pending IPI - we're an offline thread */
+	ld	r5, HSTATE_XICS_PHYS(r13)
+	li	r7, XICS_XIRR
+	lwzcix	r3, r5, r7		/* ack any pending interrupt */
+	rlwinm.	r0, r3, 0, 0xffffff	/* any pending? */
+	beq	37f
+	sync
+	li	r0, 0xff
+	li	r6, XICS_MFRR
+	stbcix	r0, r5, r6		/* clear the IPI */
+	stwcix	r3, r5, r7		/* EOI it */
+37:	sync
+
+	/* increment the nap count and then go to nap mode */
+	ld	r4, HSTATE_KVM_VCORE(r13)
+	addi	r4, r4, VCORE_NAP_COUNT
+	lwsync				/* make previous updates visible */
+51:	lwarx	r3, 0, r4
+	addi	r3, r3, 1
+	stwcx.	r3, 0, r4
+	bne	51b
+
+kvm_no_guest:
+	li	r0, KVM_HWTHREAD_IN_NAP
+	stb	r0, HSTATE_HWTHREAD_STATE(r13)
+	li	r3, LPCR_PECE0
+	mfspr	r4, SPRN_LPCR
+	rlwimi	r4, r3, 0, LPCR_PECE0 | LPCR_PECE1
+	mtspr	SPRN_LPCR, r4
+	isync
+	std	r0, HSTATE_SCRATCH0(r13)
+	ptesync
+	ld	r0, HSTATE_SCRATCH0(r13)
+1:	cmpd	r0, r0
+	bne	1b
+	nap
+	b	.
+
+/******************************************************************************
+ *                                                                            *
+ *                               Entry code                                   *
+ *                                                                            *
+ *****************************************************************************/
+
 .global kvmppc_hv_entry
 kvmppc_hv_entry:
 
@@ -159,7 +281,8 @@ kvmppc_hv_entry:
 	 * all other volatile GPRS = free
 	 */
 	mflr	r0
-	std	r0, HSTATE_VMHANDLER(r13)
+	std	r0, PPC_LR_STKOFF(r1)
+	stdu	r1, -112(r1)
 
 	/* Set partition DABR */
 	/* Do this before re-enabling PMU to avoid P7 DABR corruption bug */
@@ -200,8 +323,12 @@ END_FTR_SECTION_IFSET(CPU_FTR_ARCH_201)
 	ld	r3, VCPU_MMCR(r4)
 	ld	r5, VCPU_MMCR + 8(r4)
 	ld	r6, VCPU_MMCR + 16(r4)
+	ld	r7, VCPU_SIAR(r4)
+	ld	r8, VCPU_SDAR(r4)
 	mtspr	SPRN_MMCR1, r5
 	mtspr	SPRN_MMCRA, r6
+	mtspr	SPRN_SIAR, r7
+	mtspr	SPRN_SDAR, r8
 	mtspr	SPRN_MMCR0, r3
 	isync
 
@@ -254,22 +381,15 @@ END_FTR_SECTION_IFSET(CPU_FTR_ARCH_206)
 	/* Save R1 in the PACA */
 	std	r1, HSTATE_HOST_R1(r13)
 
-	/* Increment yield count if they have a VPA */
-	ld	r3, VCPU_VPA(r4)
-	cmpdi	r3, 0
-	beq	25f
-	lwz	r5, LPPACA_YIELDCOUNT(r3)
-	addi	r5, r5, 1
-	stw	r5, LPPACA_YIELDCOUNT(r3)
-	li	r6, 1
-	stb	r6, VCPU_VPA_DIRTY(r4)
-25:
 	/* Load up DAR and DSISR */
 	ld	r5, VCPU_DAR(r4)
 	lwz	r6, VCPU_DSISR(r4)
 	mtspr	SPRN_DAR, r5
 	mtspr	SPRN_DSISR, r6
 
+	li	r6, KVM_GUEST_MODE_HOST_HV
+	stb	r6, HSTATE_IN_GUEST(r13)
+
 BEGIN_FTR_SECTION
 	/* Restore AMR and UAMOR, set AMOR to all 1s */
 	ld	r5,VCPU_AMR(r4)
@@ -343,7 +463,28 @@ END_FTR_SECTION_IFSET(CPU_FTR_ARCH_201)
 	bdnz	28b
 	ptesync
 
-22:	li	r0,1
+	/* Add timebase offset onto timebase */
+22:	ld	r8,VCORE_TB_OFFSET(r5)
+	cmpdi	r8,0
+	beq	37f
+	mftb	r6		/* current host timebase */
+	add	r8,r8,r6
+	mtspr	SPRN_TBU40,r8	/* update upper 40 bits */
+	mftb	r7		/* check if lower 24 bits overflowed */
+	clrldi	r6,r6,40
+	clrldi	r7,r7,40
+	cmpld	r7,r6
+	bge	37f
+	addis	r8,r8,0x100	/* if so, increment upper 40 bits */
+	mtspr	SPRN_TBU40,r8
+
+	/* Load guest PCR value to select appropriate compat mode */
+37:	ld	r7, VCORE_PCR(r5)
+	cmpdi	r7, 0
+	beq	38f
+	mtspr	SPRN_PCR, r7
+38:
+	li	r0,1
 	stb	r0,VCORE_IN_GUEST(r5)	/* signal secondaries to continue */
 	b	10f
 
@@ -353,12 +494,22 @@ END_FTR_SECTION_IFSET(CPU_FTR_ARCH_201)
 	beq	20b
 
 	/* Set LPCR and RMOR. */
-10:	ld	r8,KVM_LPCR(r9)
+10:	ld	r8,VCORE_LPCR(r5)
 	mtspr	SPRN_LPCR,r8
 	ld	r8,KVM_RMOR(r9)
 	mtspr	SPRN_RMOR,r8
 	isync
 
+	/* Increment yield count if they have a VPA */
+	ld	r3, VCPU_VPA(r4)
+	cmpdi	r3, 0
+	beq	25f
+	lwz	r5, LPPACA_YIELDCOUNT(r3)
+	addi	r5, r5, 1
+	stw	r5, LPPACA_YIELDCOUNT(r3)
+	li	r6, 1
+	stb	r6, VCPU_VPA_DIRTY(r4)
+25:
 	/* Check if HDEC expires soon */
 	mfspr	r3,SPRN_HDEC
 	cmpwi	r3,10
@@ -405,7 +556,8 @@ toc_tlbie_lock:
 	bne	24b
 	isync
 
-	ld	r7,KVM_LPCR(r9)		/* use kvm->arch.lpcr to store HID4 */
+	ld	r5,HSTATE_KVM_VCORE(r13)
+	ld	r7,VCORE_LPCR(r5)	/* use vcore->lpcr to store HID4 */
 	li	r0,0x18f
 	rotldi	r0,r0,HID4_LPID5_SH	/* all lpid bits in HID4 = 1 */
 	or	r0,r7,r0
@@ -541,7 +693,7 @@ fast_guest_return:
 	mtspr	SPRN_HSRR1,r11
 
 	/* Activate guest mode, so faults get handled by KVM */
-	li	r9, KVM_GUEST_MODE_GUEST
+	li	r9, KVM_GUEST_MODE_GUEST_HV
 	stb	r9, HSTATE_IN_GUEST(r13)
 
 	/* Enter guest */
@@ -550,13 +702,15 @@ BEGIN_FTR_SECTION
 	ld	r5, VCPU_CFAR(r4)
 	mtspr	SPRN_CFAR, r5
 END_FTR_SECTION_IFSET(CPU_FTR_CFAR)
+BEGIN_FTR_SECTION
+	ld	r0, VCPU_PPR(r4)
+END_FTR_SECTION_IFSET(CPU_FTR_HAS_PPR)
 
 	ld	r5, VCPU_LR(r4)
 	lwz	r6, VCPU_CR(r4)
 	mtlr	r5
 	mtcr	r6
 
-	ld	r0, VCPU_GPR(R0)(r4)
 	ld	r1, VCPU_GPR(R1)(r4)
 	ld	r2, VCPU_GPR(R2)(r4)
 	ld	r3, VCPU_GPR(R3)(r4)
@@ -570,6 +724,10 @@ END_FTR_SECTION_IFSET(CPU_FTR_CFAR)
 	ld	r12, VCPU_GPR(R12)(r4)
 	ld	r13, VCPU_GPR(R13)(r4)
 
+BEGIN_FTR_SECTION
+	mtspr	SPRN_PPR, r0
+END_FTR_SECTION_IFSET(CPU_FTR_HAS_PPR)
+	ld	r0, VCPU_GPR(R0)(r4)
 	ld	r4, VCPU_GPR(R4)(r4)
 
 	hrfid
@@ -584,8 +742,8 @@ END_FTR_SECTION_IFSET(CPU_FTR_CFAR)
 /*
  * We come here from the first-level interrupt handlers.
  */
-	.globl	kvmppc_interrupt
-kvmppc_interrupt:
+	.globl	kvmppc_interrupt_hv
+kvmppc_interrupt_hv:
 	/*
 	 * Register contents:
 	 * R12		= interrupt vector
@@ -595,6 +753,19 @@ kvmppc_interrupt:
 	 */
 	/* abuse host_r2 as third scratch area; we get r2 from PACATOC(r13) */
 	std	r9, HSTATE_HOST_R2(r13)
+
+	lbz	r9, HSTATE_IN_GUEST(r13)
+	cmpwi	r9, KVM_GUEST_MODE_HOST_HV
+	beq	kvmppc_bad_host_intr
+#ifdef CONFIG_KVM_BOOK3S_PR_POSSIBLE
+	cmpwi	r9, KVM_GUEST_MODE_GUEST
+	ld	r9, HSTATE_HOST_R2(r13)
+	beq	kvmppc_interrupt_pr
+#endif
+	/* We're now back in the host but in guest MMU context */
+	li	r9, KVM_GUEST_MODE_HOST_HV
+	stb	r9, HSTATE_IN_GUEST(r13)
+
 	ld	r9, HSTATE_KVM_VCPU(r13)
 
 	/* Save registers */
@@ -620,6 +791,10 @@ BEGIN_FTR_SECTION
 	ld	r3, HSTATE_CFAR(r13)
 	std	r3, VCPU_CFAR(r9)
 END_FTR_SECTION_IFSET(CPU_FTR_CFAR)
+BEGIN_FTR_SECTION
+	ld	r4, HSTATE_PPR(r13)
+	std	r4, VCPU_PPR(r9)
+END_FTR_SECTION_IFSET(CPU_FTR_HAS_PPR)
 
 	/* Restore R1/R2 so we can handle faults */
 	ld	r1, HSTATE_HOST_R1(r13)
@@ -642,10 +817,6 @@ END_FTR_SECTION_IFSET(CPU_FTR_CFAR)
 	std	r3, VCPU_GPR(R13)(r9)
 	std	r4, VCPU_LR(r9)
 
-	/* Unset guest mode */
-	li	r0, KVM_GUEST_MODE_NONE
-	stb	r0, HSTATE_IN_GUEST(r13)
-
 	stw	r12,VCPU_TRAP(r9)
 
 	/* Save HEIR (HV emulation assist reg) in last_inst
@@ -696,46 +867,11 @@ END_FTR_SECTION_IFCLR(CPU_FTR_ARCH_206)
 	 * set, we know the host wants us out so let's do it now
 	 */
 do_ext_interrupt:
-	lbz	r0, HSTATE_HOST_IPI(r13)
-	cmpwi	r0, 0
-	bne	ext_interrupt_to_host
-
-	/* Now read the interrupt from the ICP */
-	ld	r5, HSTATE_XICS_PHYS(r13)
-	li	r7, XICS_XIRR
-	cmpdi	r5, 0
-	beq-	ext_interrupt_to_host
-	lwzcix	r3, r5, r7
-	rlwinm.	r0, r3, 0, 0xffffff
-	sync
-	beq	3f		/* if nothing pending in the ICP */
-
-	/* We found something in the ICP...
-	 *
-	 * If it's not an IPI, stash it in the PACA and return to
-	 * the host, we don't (yet) handle directing real external
-	 * interrupts directly to the guest
-	 */
-	cmpwi	r0, XICS_IPI
-	bne	ext_stash_for_host
-
-	/* It's an IPI, clear the MFRR and EOI it */
-	li	r0, 0xff
-	li	r6, XICS_MFRR
-	stbcix	r0, r5, r6		/* clear the IPI */
-	stwcix	r3, r5, r7		/* EOI it */
-	sync
-
-	/* We need to re-check host IPI now in case it got set in the
-	 * meantime. If it's clear, we bounce the interrupt to the
-	 * guest
-	 */
-	lbz	r0, HSTATE_HOST_IPI(r13)
-	cmpwi	r0, 0
-	bne-	1f
+	bl	kvmppc_read_intr
+	cmpdi	r3, 0
+	bgt	ext_interrupt_to_host
 
 	/* Allright, looks like an IPI for the guest, we need to set MER */
-3:
 	/* Check if any CPU is heading out to the host, if so head out too */
 	ld	r5, HSTATE_KVM_VCORE(r13)
 	lwz	r0, VCORE_ENTRY_EXIT(r5)
@@ -764,27 +900,9 @@ do_ext_interrupt:
 	mtspr	SPRN_LPCR, r8
 	b	fast_guest_return
 
-	/* We raced with the host, we need to resend that IPI, bummer */
-1:	li	r0, IPI_PRIORITY
-	stbcix	r0, r5, r6		/* set the IPI */
-	sync
-	b	ext_interrupt_to_host
-
-ext_stash_for_host:
-	/* It's not an IPI and it's for the host, stash it in the PACA
-	 * before exit, it will be picked up by the host ICP driver
-	 */
-	stw	r3, HSTATE_SAVED_XIRR(r13)
 ext_interrupt_to_host:
 
 guest_exit_cont:		/* r9 = vcpu, r12 = trap, r13 = paca */
-	/* Save DEC */
-	mfspr	r5,SPRN_DEC
-	mftb	r6
-	extsw	r5,r5
-	add	r5,r5,r6
-	std	r5,VCPU_DEC_EXPIRES(r9)
-
 	/* Save more register state  */
 	mfdar	r6
 	mfdsisr	r7
@@ -954,7 +1072,30 @@ END_FTR_SECTION_IFSET(CPU_FTR_ARCH_201)
 	mtspr	SPRN_SDR1,r6		/* switch to partition page table */
 	mtspr	SPRN_LPID,r7
 	isync
-	li	r0,0
+
+	/* Subtract timebase offset from timebase */
+	ld	r8,VCORE_TB_OFFSET(r5)
+	cmpdi	r8,0
+	beq	17f
+	mftb	r6			/* current host timebase */
+	subf	r8,r8,r6
+	mtspr	SPRN_TBU40,r8		/* update upper 40 bits */
+	mftb	r7			/* check if lower 24 bits overflowed */
+	clrldi	r6,r6,40
+	clrldi	r7,r7,40
+	cmpld	r7,r6
+	bge	17f
+	addis	r8,r8,0x100		/* if so, increment upper 40 bits */
+	mtspr	SPRN_TBU40,r8
+
+	/* Reset PCR */
+17:	ld	r0, VCORE_PCR(r5)
+	cmpdi	r0, 0
+	beq	18f
+	li	r0, 0
+	mtspr	SPRN_PCR, r0
+18:
+	/* Signal secondary CPUs to continue */
 	stb	r0,VCORE_IN_GUEST(r5)
 	lis	r8,0x7fff		/* MAX_INT@h */
 	mtspr	SPRN_HDEC,r8
@@ -1052,6 +1193,13 @@ END_FTR_SECTION_IFSET(CPU_FTR_ARCH_201)
 1:	addi	r8,r8,16
 	.endr
 
+	/* Save DEC */
+	mfspr	r5,SPRN_DEC
+	mftb	r6
+	extsw	r5,r5
+	add	r5,r5,r6
+	std	r5,VCPU_DEC_EXPIRES(r9)
+
 	/* Save and reset AMR and UAMOR before turning on the MMU */
 BEGIN_FTR_SECTION
 	mfspr	r5,SPRN_AMR
@@ -1062,6 +1210,10 @@ BEGIN_FTR_SECTION
 	mtspr	SPRN_AMR,r6
 END_FTR_SECTION_IFSET(CPU_FTR_ARCH_206)
 
+	/* Unset guest mode */
+	li	r0, KVM_GUEST_MODE_NONE
+	stb	r0, HSTATE_IN_GUEST(r13)
+
 	/* Switch DSCR back to host value */
 BEGIN_FTR_SECTION
 	mfspr	r8, SPRN_DSCR
@@ -1134,9 +1286,13 @@ END_FTR_SECTION_IFSET(CPU_FTR_ARCH_206)
 	std	r3, VCPU_MMCR(r9)	/* if not, set saved MMCR0 to FC */
 	b	22f
 21:	mfspr	r5, SPRN_MMCR1
+	mfspr	r7, SPRN_SIAR
+	mfspr	r8, SPRN_SDAR
 	std	r4, VCPU_MMCR(r9)
 	std	r5, VCPU_MMCR + 8(r9)
 	std	r6, VCPU_MMCR + 16(r9)
+	std	r7, VCPU_SIAR(r9)
+	std	r8, VCPU_SDAR(r9)
 	mfspr	r3, SPRN_PMC1
 	mfspr	r4, SPRN_PMC2
 	mfspr	r5, SPRN_PMC3
@@ -1158,103 +1314,30 @@ BEGIN_FTR_SECTION
 	stw	r11, VCPU_PMC + 28(r9)
 END_FTR_SECTION_IFSET(CPU_FTR_ARCH_201)
 22:
+	ld	r0, 112+PPC_LR_STKOFF(r1)
+	addi	r1, r1, 112
+	mtlr	r0
+	blr
+secondary_too_late:
+	ld	r5,HSTATE_KVM_VCORE(r13)
+	HMT_LOW
+13:	lbz	r3,VCORE_IN_GUEST(r5)
+	cmpwi	r3,0
+	bne	13b
+	HMT_MEDIUM
+	li	r0, KVM_GUEST_MODE_NONE
+	stb	r0, HSTATE_IN_GUEST(r13)
+	ld	r11,PACA_SLBSHADOWPTR(r13)
 
-	/* Secondary threads go off to take a nap on POWER7 */
-BEGIN_FTR_SECTION
-	lwz	r0,VCPU_PTID(r9)
-	cmpwi	r0,0
-	bne	secondary_nap
-END_FTR_SECTION_IFSET(CPU_FTR_ARCH_206)
-
-	/* Restore host DABR and DABRX */
-	ld	r5,HSTATE_DABR(r13)
-	li	r6,7
-	mtspr	SPRN_DABR,r5
-	mtspr	SPRN_DABRX,r6
-
-	/* Restore SPRG3 */
-	ld	r3,PACA_SPRG3(r13)
-	mtspr	SPRN_SPRG3,r3
-
-	/*
-	 * Reload DEC.  HDEC interrupts were disabled when
-	 * we reloaded the host's LPCR value.
-	 */
-	ld	r3, HSTATE_DECEXP(r13)
-	mftb	r4
-	subf	r4, r4, r3
-	mtspr	SPRN_DEC, r4
-
-	/* Reload the host's PMU registers */
-	ld	r3, PACALPPACAPTR(r13)	/* is the host using the PMU? */
-	lbz	r4, LPPACA_PMCINUSE(r3)
-	cmpwi	r4, 0
-	beq	23f			/* skip if not */
-	lwz	r3, HSTATE_PMC(r13)
-	lwz	r4, HSTATE_PMC + 4(r13)
-	lwz	r5, HSTATE_PMC + 8(r13)
-	lwz	r6, HSTATE_PMC + 12(r13)
-	lwz	r8, HSTATE_PMC + 16(r13)
-	lwz	r9, HSTATE_PMC + 20(r13)
-BEGIN_FTR_SECTION
-	lwz	r10, HSTATE_PMC + 24(r13)
-	lwz	r11, HSTATE_PMC + 28(r13)
-END_FTR_SECTION_IFSET(CPU_FTR_ARCH_201)
-	mtspr	SPRN_PMC1, r3
-	mtspr	SPRN_PMC2, r4
-	mtspr	SPRN_PMC3, r5
-	mtspr	SPRN_PMC4, r6
-	mtspr	SPRN_PMC5, r8
-	mtspr	SPRN_PMC6, r9
-BEGIN_FTR_SECTION
-	mtspr	SPRN_PMC7, r10
-	mtspr	SPRN_PMC8, r11
-END_FTR_SECTION_IFSET(CPU_FTR_ARCH_201)
-	ld	r3, HSTATE_MMCR(r13)
-	ld	r4, HSTATE_MMCR + 8(r13)
-	ld	r5, HSTATE_MMCR + 16(r13)
-	mtspr	SPRN_MMCR1, r4
-	mtspr	SPRN_MMCRA, r5
-	mtspr	SPRN_MMCR0, r3
-	isync
-23:
-	/*
-	 * For external and machine check interrupts, we need
-	 * to call the Linux handler to process the interrupt.
-	 * We do that by jumping to absolute address 0x500 for
-	 * external interrupts, or the machine_check_fwnmi label
-	 * for machine checks (since firmware might have patched
-	 * the vector area at 0x200).  The [h]rfid at the end of the
-	 * handler will return to the book3s_hv_interrupts.S code.
-	 * For other interrupts we do the rfid to get back
-	 * to the book3s_hv_interrupts.S code here.
-	 */
-	ld	r8, HSTATE_VMHANDLER(r13)
-	ld	r7, HSTATE_HOST_MSR(r13)
-
-	cmpwi	cr1, r12, BOOK3S_INTERRUPT_MACHINE_CHECK
-	cmpwi	r12, BOOK3S_INTERRUPT_EXTERNAL
-BEGIN_FTR_SECTION
-	beq	11f
-END_FTR_SECTION_IFSET(CPU_FTR_ARCH_206)
-
-	/* RFI into the highmem handler, or branch to interrupt handler */
-	mfmsr	r6
-	li	r0, MSR_RI
-	andc	r6, r6, r0
-	mtmsrd	r6, 1			/* Clear RI in MSR */
-	mtsrr0	r8
-	mtsrr1	r7
-	beqa	0x500			/* external interrupt (PPC970) */
-	beq	cr1, 13f		/* machine check */
-	RFI
-
-	/* On POWER7, we have external interrupts set to use HSRR0/1 */
-11:	mtspr	SPRN_HSRR0, r8
-	mtspr	SPRN_HSRR1, r7
-	ba	0x500
-
-13:	b	machine_check_fwnmi
+	.rept	SLB_NUM_BOLTED
+	ld	r5,SLBSHADOW_SAVEAREA(r11)
+	ld	r6,SLBSHADOW_SAVEAREA+8(r11)
+	andis.	r7,r5,SLB_ESID_V@h
+	beq	1f
+	slbmte	r6,r5
+1:	addi	r11,r11,16
+	.endr
+	b	22b
 
 /*
  * Check whether an HDSI is an HPTE not found fault or something else.
@@ -1333,7 +1416,7 @@ fast_interrupt_c_return:
 	stw	r8, VCPU_LAST_INST(r9)
 
 	/* Unset guest mode. */
-	li	r0, KVM_GUEST_MODE_NONE
+	li	r0, KVM_GUEST_MODE_HOST_HV
 	stb	r0, HSTATE_IN_GUEST(r13)
 	b	guest_exit_cont
 
@@ -1701,67 +1784,70 @@ machine_check_realmode:
 	rotldi	r11, r11, 63
 	b	fast_interrupt_c_return
 
-secondary_too_late:
-	ld	r5,HSTATE_KVM_VCORE(r13)
-	HMT_LOW
-13:	lbz	r3,VCORE_IN_GUEST(r5)
-	cmpwi	r3,0
-	bne	13b
-	HMT_MEDIUM
-	ld	r11,PACA_SLBSHADOWPTR(r13)
-
-	.rept	SLB_NUM_BOLTED
-	ld	r5,SLBSHADOW_SAVEAREA(r11)
-	ld	r6,SLBSHADOW_SAVEAREA+8(r11)
-	andis.	r7,r5,SLB_ESID_V@h
-	beq	1f
-	slbmte	r6,r5
-1:	addi	r11,r11,16
-	.endr
+/*
+ * Determine what sort of external interrupt is pending (if any).
+ * Returns:
+ *	0 if no interrupt is pending
+ *	1 if an interrupt is pending that needs to be handled by the host
+ *	-1 if there was a guest wakeup IPI (which has now been cleared)
+ */
+kvmppc_read_intr:
+	/* see if a host IPI is pending */
+	li	r3, 1
+	lbz	r0, HSTATE_HOST_IPI(r13)
+	cmpwi	r0, 0
+	bne	1f
 
-secondary_nap:
-	/* Clear our vcpu pointer so we don't come back in early */
-	li	r0, 0
-	std	r0, HSTATE_KVM_VCPU(r13)
-	lwsync
-	/* Clear any pending IPI - assume we're a secondary thread */
-	ld	r5, HSTATE_XICS_PHYS(r13)
+	/* Now read the interrupt from the ICP */
+	ld	r6, HSTATE_XICS_PHYS(r13)
 	li	r7, XICS_XIRR
-	lwzcix	r3, r5, r7		/* ack any pending interrupt */
-	rlwinm.	r0, r3, 0, 0xffffff	/* any pending? */
-	beq	37f
+	cmpdi	r6, 0
+	beq-	1f
+	lwzcix	r0, r6, r7
+	rlwinm.	r3, r0, 0, 0xffffff
 	sync
-	li	r0, 0xff
-	li	r6, XICS_MFRR
-	stbcix	r0, r5, r6		/* clear the IPI */
-	stwcix	r3, r5, r7		/* EOI it */
-37:	sync
+	beq	1f			/* if nothing pending in the ICP */
 
-	/* increment the nap count and then go to nap mode */
-	ld	r4, HSTATE_KVM_VCORE(r13)
-	addi	r4, r4, VCORE_NAP_COUNT
-	lwsync				/* make previous updates visible */
-51:	lwarx	r3, 0, r4
-	addi	r3, r3, 1
-	stwcx.	r3, 0, r4
-	bne	51b
+	/* We found something in the ICP...
+	 *
+	 * If it's not an IPI, stash it in the PACA and return to
+	 * the host, we don't (yet) handle directing real external
+	 * interrupts directly to the guest
+	 */
+	cmpwi	r3, XICS_IPI		/* if there is, is it an IPI? */
+	li	r3, 1
+	bne	42f
 
-kvm_no_guest:
-	li	r0, KVM_HWTHREAD_IN_NAP
-	stb	r0, HSTATE_HWTHREAD_STATE(r13)
+	/* It's an IPI, clear the MFRR and EOI it */
+	li	r3, 0xff
+	li	r8, XICS_MFRR
+	stbcix	r3, r6, r8		/* clear the IPI */
+	stwcix	r0, r6, r7		/* EOI it */
+	sync
 
-	li	r3, LPCR_PECE0
-	mfspr	r4, SPRN_LPCR
-	rlwimi	r4, r3, 0, LPCR_PECE0 | LPCR_PECE1
-	mtspr	SPRN_LPCR, r4
-	isync
-	std	r0, HSTATE_SCRATCH0(r13)
-	ptesync
-	ld	r0, HSTATE_SCRATCH0(r13)
-1:	cmpd	r0, r0
-	bne	1b
-	nap
-	b	.
+	/* We need to re-check host IPI now in case it got set in the
+	 * meantime. If it's clear, we bounce the interrupt to the
+	 * guest
+	 */
+	lbz	r0, HSTATE_HOST_IPI(r13)
+	cmpwi	r0, 0
+	bne-	43f
+
+	/* OK, it's an IPI for us */
+	li	r3, -1
+1:	blr
+
+42:	/* It's not an IPI and it's for the host, stash it in the PACA
+	 * before exit, it will be picked up by the host ICP driver
+	 */
+	stw	r0, HSTATE_SAVED_XIRR(r13)
+	b	1b
+
+43:	/* We raced with the host, we need to resend that IPI, bummer */
+	li	r0, IPI_PRIORITY
+	stbcix	r0, r6, r8		/* set the IPI */
+	sync
+	b	1b
 
 /*
  * Save away FP, VMX and VSX registers.
@@ -1879,3 +1965,11 @@ END_FTR_SECTION_IFSET(CPU_FTR_ALTIVEC)
 	lwz	r7,VCPU_VRSAVE(r4)
 	mtspr	SPRN_VRSAVE,r7
 	blr
+
+/*
+ * We come here if we get any exception or interrupt while we are
+ * executing host real mode code while in guest MMU context.
+ * For now just spin, but we should do something better.
+ */
+kvmppc_bad_host_intr:
+	b	.
diff --git a/arch/powerpc/kvm/book3s_interrupts.S b/arch/powerpc/kvm/book3s_interrupts.S
index 17cfae5497a3..f4dd041c14ea 100644
--- a/arch/powerpc/kvm/book3s_interrupts.S
+++ b/arch/powerpc/kvm/book3s_interrupts.S
@@ -26,8 +26,12 @@
 
 #if defined(CONFIG_PPC_BOOK3S_64)
 #define FUNC(name) 		GLUE(.,name)
+#define GET_SHADOW_VCPU(reg)    addi	reg, r13, PACA_SVCPU
+
 #elif defined(CONFIG_PPC_BOOK3S_32)
 #define FUNC(name)		name
+#define GET_SHADOW_VCPU(reg)	lwz     reg, (THREAD + THREAD_KVM_SVCPU)(r2)
+
 #endif /* CONFIG_PPC_BOOK3S_XX */
 
 #define VCPU_LOAD_NVGPRS(vcpu) \
@@ -87,8 +91,14 @@ kvm_start_entry:
 	VCPU_LOAD_NVGPRS(r4)
 
 kvm_start_lightweight:
+	/* Copy registers into shadow vcpu so we can access them in real mode */
+	GET_SHADOW_VCPU(r3)
+	bl	FUNC(kvmppc_copy_to_svcpu)
+	nop
+	REST_GPR(4, r1)
 
 #ifdef CONFIG_PPC_BOOK3S_64
+	/* Get the dcbz32 flag */
 	PPC_LL	r3, VCPU_HFLAGS(r4)
 	rldicl	r3, r3, 0, 63		/* r3 &= 1 */
 	stb	r3, HSTATE_RESTORE_HID5(r13)
@@ -111,9 +121,6 @@ kvm_start_lightweight:
  *
  */
 
-.global kvmppc_handler_highmem
-kvmppc_handler_highmem:
-
 	/*
 	 * Register usage at this point:
 	 *
@@ -125,18 +132,31 @@ kvmppc_handler_highmem:
 	 *
 	 */
 
-	/* R7 = vcpu */
-	PPC_LL	r7, GPR4(r1)
+	/* Transfer reg values from shadow vcpu back to vcpu struct */
+	/* On 64-bit, interrupts are still off at this point */
+	PPC_LL	r3, GPR4(r1)		/* vcpu pointer */
+	GET_SHADOW_VCPU(r4)
+	bl	FUNC(kvmppc_copy_from_svcpu)
+	nop
 
 #ifdef CONFIG_PPC_BOOK3S_64
+	/* Re-enable interrupts */
+	ld	r3, HSTATE_HOST_MSR(r13)
+	ori	r3, r3, MSR_EE
+	MTMSR_EERI(r3)
+
 	/*
 	 * Reload kernel SPRG3 value.
 	 * No need to save guest value as usermode can't modify SPRG3.
 	 */
 	ld	r3, PACA_SPRG3(r13)
 	mtspr	SPRN_SPRG3, r3
+
 #endif /* CONFIG_PPC_BOOK3S_64 */
 
+	/* R7 = vcpu */
+	PPC_LL	r7, GPR4(r1)
+
 	PPC_STL	r14, VCPU_GPR(R14)(r7)
 	PPC_STL	r15, VCPU_GPR(R15)(r7)
 	PPC_STL	r16, VCPU_GPR(R16)(r7)
@@ -161,7 +181,7 @@ kvmppc_handler_highmem:
 
 	/* Restore r3 (kvm_run) and r4 (vcpu) */
 	REST_2GPRS(3, r1)
-	bl	FUNC(kvmppc_handle_exit)
+	bl	FUNC(kvmppc_handle_exit_pr)
 
 	/* If RESUME_GUEST, get back in the loop */
 	cmpwi	r3, RESUME_GUEST
diff --git a/arch/powerpc/kvm/book3s_mmu_hpte.c b/arch/powerpc/kvm/book3s_mmu_hpte.c
index da8b13c4b776..5a1ab1250a05 100644
--- a/arch/powerpc/kvm/book3s_mmu_hpte.c
+++ b/arch/powerpc/kvm/book3s_mmu_hpte.c
@@ -28,7 +28,7 @@
 #include <asm/mmu_context.h>
 #include <asm/hw_irq.h>
 
-#include "trace.h"
+#include "trace_pr.h"
 
 #define PTE_SIZE	12
 
@@ -56,6 +56,14 @@ static inline u64 kvmppc_mmu_hash_vpte_long(u64 vpage)
 		       HPTEG_HASH_BITS_VPTE_LONG);
 }
 
+#ifdef CONFIG_PPC_BOOK3S_64
+static inline u64 kvmppc_mmu_hash_vpte_64k(u64 vpage)
+{
+	return hash_64((vpage & 0xffffffff0ULL) >> 4,
+		       HPTEG_HASH_BITS_VPTE_64K);
+}
+#endif
+
 void kvmppc_mmu_hpte_cache_map(struct kvm_vcpu *vcpu, struct hpte_cache *pte)
 {
 	u64 index;
@@ -83,6 +91,15 @@ void kvmppc_mmu_hpte_cache_map(struct kvm_vcpu *vcpu, struct hpte_cache *pte)
 	hlist_add_head_rcu(&pte->list_vpte_long,
 			   &vcpu3s->hpte_hash_vpte_long[index]);
 
+#ifdef CONFIG_PPC_BOOK3S_64
+	/* Add to vPTE_64k list */
+	index = kvmppc_mmu_hash_vpte_64k(pte->pte.vpage);
+	hlist_add_head_rcu(&pte->list_vpte_64k,
+			   &vcpu3s->hpte_hash_vpte_64k[index]);
+#endif
+
+	vcpu3s->hpte_cache_count++;
+
 	spin_unlock(&vcpu3s->mmu_lock);
 }
 
@@ -113,10 +130,13 @@ static void invalidate_pte(struct kvm_vcpu *vcpu, struct hpte_cache *pte)
 	hlist_del_init_rcu(&pte->list_pte_long);
 	hlist_del_init_rcu(&pte->list_vpte);
 	hlist_del_init_rcu(&pte->list_vpte_long);
+#ifdef CONFIG_PPC_BOOK3S_64
+	hlist_del_init_rcu(&pte->list_vpte_64k);
+#endif
+	vcpu3s->hpte_cache_count--;
 
 	spin_unlock(&vcpu3s->mmu_lock);
 
-	vcpu3s->hpte_cache_count--;
 	call_rcu(&pte->rcu_head, free_pte_rcu);
 }
 
@@ -219,6 +239,29 @@ static void kvmppc_mmu_pte_vflush_short(struct kvm_vcpu *vcpu, u64 guest_vp)
 	rcu_read_unlock();
 }
 
+#ifdef CONFIG_PPC_BOOK3S_64
+/* Flush with mask 0xffffffff0 */
+static void kvmppc_mmu_pte_vflush_64k(struct kvm_vcpu *vcpu, u64 guest_vp)
+{
+	struct kvmppc_vcpu_book3s *vcpu3s = to_book3s(vcpu);
+	struct hlist_head *list;
+	struct hpte_cache *pte;
+	u64 vp_mask = 0xffffffff0ULL;
+
+	list = &vcpu3s->hpte_hash_vpte_64k[
+		kvmppc_mmu_hash_vpte_64k(guest_vp)];
+
+	rcu_read_lock();
+
+	/* Check the list for matching entries and invalidate */
+	hlist_for_each_entry_rcu(pte, list, list_vpte_64k)
+		if ((pte->pte.vpage & vp_mask) == guest_vp)
+			invalidate_pte(vcpu, pte);
+
+	rcu_read_unlock();
+}
+#endif
+
 /* Flush with mask 0xffffff000 */
 static void kvmppc_mmu_pte_vflush_long(struct kvm_vcpu *vcpu, u64 guest_vp)
 {
@@ -249,6 +292,11 @@ void kvmppc_mmu_pte_vflush(struct kvm_vcpu *vcpu, u64 guest_vp, u64 vp_mask)
 	case 0xfffffffffULL:
 		kvmppc_mmu_pte_vflush_short(vcpu, guest_vp);
 		break;
+#ifdef CONFIG_PPC_BOOK3S_64
+	case 0xffffffff0ULL:
+		kvmppc_mmu_pte_vflush_64k(vcpu, guest_vp);
+		break;
+#endif
 	case 0xffffff000ULL:
 		kvmppc_mmu_pte_vflush_long(vcpu, guest_vp);
 		break;
@@ -285,15 +333,19 @@ struct hpte_cache *kvmppc_mmu_hpte_cache_next(struct kvm_vcpu *vcpu)
 	struct kvmppc_vcpu_book3s *vcpu3s = to_book3s(vcpu);
 	struct hpte_cache *pte;
 
-	pte = kmem_cache_zalloc(hpte_cache, GFP_KERNEL);
-	vcpu3s->hpte_cache_count++;
-
 	if (vcpu3s->hpte_cache_count == HPTEG_CACHE_NUM)
 		kvmppc_mmu_pte_flush_all(vcpu);
 
+	pte = kmem_cache_zalloc(hpte_cache, GFP_KERNEL);
+
 	return pte;
 }
 
+void kvmppc_mmu_hpte_cache_free(struct hpte_cache *pte)
+{
+	kmem_cache_free(hpte_cache, pte);
+}
+
 void kvmppc_mmu_hpte_destroy(struct kvm_vcpu *vcpu)
 {
 	kvmppc_mmu_pte_flush(vcpu, 0, 0);
@@ -320,6 +372,10 @@ int kvmppc_mmu_hpte_init(struct kvm_vcpu *vcpu)
 				  ARRAY_SIZE(vcpu3s->hpte_hash_vpte));
 	kvmppc_mmu_hpte_init_hash(vcpu3s->hpte_hash_vpte_long,
 				  ARRAY_SIZE(vcpu3s->hpte_hash_vpte_long));
+#ifdef CONFIG_PPC_BOOK3S_64
+	kvmppc_mmu_hpte_init_hash(vcpu3s->hpte_hash_vpte_64k,
+				  ARRAY_SIZE(vcpu3s->hpte_hash_vpte_64k));
+#endif
 
 	spin_lock_init(&vcpu3s->mmu_lock);
 
diff --git a/arch/powerpc/kvm/book3s_pr.c b/arch/powerpc/kvm/book3s_pr.c
index c0b48f96a91c..fe14ca3dd171 100644
--- a/arch/powerpc/kvm/book3s_pr.c
+++ b/arch/powerpc/kvm/book3s_pr.c
@@ -40,8 +40,12 @@
 #include <linux/sched.h>
 #include <linux/vmalloc.h>
 #include <linux/highmem.h>
+#include <linux/module.h>
 
-#include "trace.h"
+#include "book3s.h"
+
+#define CREATE_TRACE_POINTS
+#include "trace_pr.h"
 
 /* #define EXIT_DEBUG */
 /* #define DEBUG_EXT */
@@ -56,29 +60,25 @@ static int kvmppc_handle_ext(struct kvm_vcpu *vcpu, unsigned int exit_nr,
 #define HW_PAGE_SIZE PAGE_SIZE
 #endif
 
-void kvmppc_core_vcpu_load(struct kvm_vcpu *vcpu, int cpu)
+static void kvmppc_core_vcpu_load_pr(struct kvm_vcpu *vcpu, int cpu)
 {
 #ifdef CONFIG_PPC_BOOK3S_64
 	struct kvmppc_book3s_shadow_vcpu *svcpu = svcpu_get(vcpu);
 	memcpy(svcpu->slb, to_book3s(vcpu)->slb_shadow, sizeof(svcpu->slb));
-	memcpy(&get_paca()->shadow_vcpu, to_book3s(vcpu)->shadow_vcpu,
-	       sizeof(get_paca()->shadow_vcpu));
 	svcpu->slb_max = to_book3s(vcpu)->slb_shadow_max;
 	svcpu_put(svcpu);
 #endif
 	vcpu->cpu = smp_processor_id();
 #ifdef CONFIG_PPC_BOOK3S_32
-	current->thread.kvm_shadow_vcpu = to_book3s(vcpu)->shadow_vcpu;
+	current->thread.kvm_shadow_vcpu = vcpu->arch.shadow_vcpu;
 #endif
 }
 
-void kvmppc_core_vcpu_put(struct kvm_vcpu *vcpu)
+static void kvmppc_core_vcpu_put_pr(struct kvm_vcpu *vcpu)
 {
 #ifdef CONFIG_PPC_BOOK3S_64
 	struct kvmppc_book3s_shadow_vcpu *svcpu = svcpu_get(vcpu);
 	memcpy(to_book3s(vcpu)->slb_shadow, svcpu->slb, sizeof(svcpu->slb));
-	memcpy(to_book3s(vcpu)->shadow_vcpu, &get_paca()->shadow_vcpu,
-	       sizeof(get_paca()->shadow_vcpu));
 	to_book3s(vcpu)->slb_shadow_max = svcpu->slb_max;
 	svcpu_put(svcpu);
 #endif
@@ -87,7 +87,61 @@ void kvmppc_core_vcpu_put(struct kvm_vcpu *vcpu)
 	vcpu->cpu = -1;
 }
 
-int kvmppc_core_check_requests(struct kvm_vcpu *vcpu)
+/* Copy data needed by real-mode code from vcpu to shadow vcpu */
+void kvmppc_copy_to_svcpu(struct kvmppc_book3s_shadow_vcpu *svcpu,
+			  struct kvm_vcpu *vcpu)
+{
+	svcpu->gpr[0] = vcpu->arch.gpr[0];
+	svcpu->gpr[1] = vcpu->arch.gpr[1];
+	svcpu->gpr[2] = vcpu->arch.gpr[2];
+	svcpu->gpr[3] = vcpu->arch.gpr[3];
+	svcpu->gpr[4] = vcpu->arch.gpr[4];
+	svcpu->gpr[5] = vcpu->arch.gpr[5];
+	svcpu->gpr[6] = vcpu->arch.gpr[6];
+	svcpu->gpr[7] = vcpu->arch.gpr[7];
+	svcpu->gpr[8] = vcpu->arch.gpr[8];
+	svcpu->gpr[9] = vcpu->arch.gpr[9];
+	svcpu->gpr[10] = vcpu->arch.gpr[10];
+	svcpu->gpr[11] = vcpu->arch.gpr[11];
+	svcpu->gpr[12] = vcpu->arch.gpr[12];
+	svcpu->gpr[13] = vcpu->arch.gpr[13];
+	svcpu->cr  = vcpu->arch.cr;
+	svcpu->xer = vcpu->arch.xer;
+	svcpu->ctr = vcpu->arch.ctr;
+	svcpu->lr  = vcpu->arch.lr;
+	svcpu->pc  = vcpu->arch.pc;
+}
+
+/* Copy data touched by real-mode code from shadow vcpu back to vcpu */
+void kvmppc_copy_from_svcpu(struct kvm_vcpu *vcpu,
+			    struct kvmppc_book3s_shadow_vcpu *svcpu)
+{
+	vcpu->arch.gpr[0] = svcpu->gpr[0];
+	vcpu->arch.gpr[1] = svcpu->gpr[1];
+	vcpu->arch.gpr[2] = svcpu->gpr[2];
+	vcpu->arch.gpr[3] = svcpu->gpr[3];
+	vcpu->arch.gpr[4] = svcpu->gpr[4];
+	vcpu->arch.gpr[5] = svcpu->gpr[5];
+	vcpu->arch.gpr[6] = svcpu->gpr[6];
+	vcpu->arch.gpr[7] = svcpu->gpr[7];
+	vcpu->arch.gpr[8] = svcpu->gpr[8];
+	vcpu->arch.gpr[9] = svcpu->gpr[9];
+	vcpu->arch.gpr[10] = svcpu->gpr[10];
+	vcpu->arch.gpr[11] = svcpu->gpr[11];
+	vcpu->arch.gpr[12] = svcpu->gpr[12];
+	vcpu->arch.gpr[13] = svcpu->gpr[13];
+	vcpu->arch.cr  = svcpu->cr;
+	vcpu->arch.xer = svcpu->xer;
+	vcpu->arch.ctr = svcpu->ctr;
+	vcpu->arch.lr  = svcpu->lr;
+	vcpu->arch.pc  = svcpu->pc;
+	vcpu->arch.shadow_srr1 = svcpu->shadow_srr1;
+	vcpu->arch.fault_dar   = svcpu->fault_dar;
+	vcpu->arch.fault_dsisr = svcpu->fault_dsisr;
+	vcpu->arch.last_inst   = svcpu->last_inst;
+}
+
+static int kvmppc_core_check_requests_pr(struct kvm_vcpu *vcpu)
 {
 	int r = 1; /* Indicate we want to get back into the guest */
 
@@ -100,44 +154,69 @@ int kvmppc_core_check_requests(struct kvm_vcpu *vcpu)
 }
 
 /************* MMU Notifiers *************/
+static void do_kvm_unmap_hva(struct kvm *kvm, unsigned long start,
+			     unsigned long end)
+{
+	long i;
+	struct kvm_vcpu *vcpu;
+	struct kvm_memslots *slots;
+	struct kvm_memory_slot *memslot;
+
+	slots = kvm_memslots(kvm);
+	kvm_for_each_memslot(memslot, slots) {
+		unsigned long hva_start, hva_end;
+		gfn_t gfn, gfn_end;
+
+		hva_start = max(start, memslot->userspace_addr);
+		hva_end = min(end, memslot->userspace_addr +
+					(memslot->npages << PAGE_SHIFT));
+		if (hva_start >= hva_end)
+			continue;
+		/*
+		 * {gfn(page) | page intersects with [hva_start, hva_end)} =
+		 * {gfn, gfn+1, ..., gfn_end-1}.
+		 */
+		gfn = hva_to_gfn_memslot(hva_start, memslot);
+		gfn_end = hva_to_gfn_memslot(hva_end + PAGE_SIZE - 1, memslot);
+		kvm_for_each_vcpu(i, vcpu, kvm)
+			kvmppc_mmu_pte_pflush(vcpu, gfn << PAGE_SHIFT,
+					      gfn_end << PAGE_SHIFT);
+	}
+}
 
-int kvm_unmap_hva(struct kvm *kvm, unsigned long hva)
+static int kvm_unmap_hva_pr(struct kvm *kvm, unsigned long hva)
 {
 	trace_kvm_unmap_hva(hva);
 
-	/*
-	 * Flush all shadow tlb entries everywhere. This is slow, but
-	 * we are 100% sure that we catch the to be unmapped page
-	 */
-	kvm_flush_remote_tlbs(kvm);
+	do_kvm_unmap_hva(kvm, hva, hva + PAGE_SIZE);
 
 	return 0;
 }
 
-int kvm_unmap_hva_range(struct kvm *kvm, unsigned long start, unsigned long end)
+static int kvm_unmap_hva_range_pr(struct kvm *kvm, unsigned long start,
+				  unsigned long end)
 {
-	/* kvm_unmap_hva flushes everything anyways */
-	kvm_unmap_hva(kvm, start);
+	do_kvm_unmap_hva(kvm, start, end);
 
 	return 0;
 }
 
-int kvm_age_hva(struct kvm *kvm, unsigned long hva)
+static int kvm_age_hva_pr(struct kvm *kvm, unsigned long hva)
 {
 	/* XXX could be more clever ;) */
 	return 0;
 }
 
-int kvm_test_age_hva(struct kvm *kvm, unsigned long hva)
+static int kvm_test_age_hva_pr(struct kvm *kvm, unsigned long hva)
 {
 	/* XXX could be more clever ;) */
 	return 0;
 }
 
-void kvm_set_spte_hva(struct kvm *kvm, unsigned long hva, pte_t pte)
+static void kvm_set_spte_hva_pr(struct kvm *kvm, unsigned long hva, pte_t pte)
 {
 	/* The page will get remapped properly on its next fault */
-	kvm_unmap_hva(kvm, hva);
+	do_kvm_unmap_hva(kvm, hva, hva + PAGE_SIZE);
 }
 
 /*****************************************/
@@ -159,7 +238,7 @@ static void kvmppc_recalc_shadow_msr(struct kvm_vcpu *vcpu)
 	vcpu->arch.shadow_msr = smsr;
 }
 
-void kvmppc_set_msr(struct kvm_vcpu *vcpu, u64 msr)
+static void kvmppc_set_msr_pr(struct kvm_vcpu *vcpu, u64 msr)
 {
 	ulong old_msr = vcpu->arch.shared->msr;
 
@@ -219,7 +298,7 @@ void kvmppc_set_msr(struct kvm_vcpu *vcpu, u64 msr)
 		kvmppc_handle_ext(vcpu, BOOK3S_INTERRUPT_FP_UNAVAIL, MSR_FP);
 }
 
-void kvmppc_set_pvr(struct kvm_vcpu *vcpu, u32 pvr)
+void kvmppc_set_pvr_pr(struct kvm_vcpu *vcpu, u32 pvr)
 {
 	u32 host_pvr;
 
@@ -256,6 +335,23 @@ void kvmppc_set_pvr(struct kvm_vcpu *vcpu, u32 pvr)
 	if (!strcmp(cur_cpu_spec->platform, "ppc-cell-be"))
 		to_book3s(vcpu)->msr_mask &= ~(MSR_FE0 | MSR_FE1);
 
+	/*
+	 * If they're asking for POWER6 or later, set the flag
+	 * indicating that we can do multiple large page sizes
+	 * and 1TB segments.
+	 * Also set the flag that indicates that tlbie has the large
+	 * page bit in the RB operand instead of the instruction.
+	 */
+	switch (PVR_VER(pvr)) {
+	case PVR_POWER6:
+	case PVR_POWER7:
+	case PVR_POWER7p:
+	case PVR_POWER8:
+		vcpu->arch.hflags |= BOOK3S_HFLAG_MULTI_PGSIZE |
+			BOOK3S_HFLAG_NEW_TLBIE;
+		break;
+	}
+
 #ifdef CONFIG_PPC_BOOK3S_32
 	/* 32 bit Book3S always has 32 byte dcbz */
 	vcpu->arch.hflags |= BOOK3S_HFLAG_DCBZ32;
@@ -334,6 +430,7 @@ int kvmppc_handle_pagefault(struct kvm_run *run, struct kvm_vcpu *vcpu,
 			    ulong eaddr, int vec)
 {
 	bool data = (vec == BOOK3S_INTERRUPT_DATA_STORAGE);
+	bool iswrite = false;
 	int r = RESUME_GUEST;
 	int relocated;
 	int page_found = 0;
@@ -344,10 +441,12 @@ int kvmppc_handle_pagefault(struct kvm_run *run, struct kvm_vcpu *vcpu,
 	u64 vsid;
 
 	relocated = data ? dr : ir;
+	if (data && (vcpu->arch.fault_dsisr & DSISR_ISSTORE))
+		iswrite = true;
 
 	/* Resolve real address if translation turned on */
 	if (relocated) {
-		page_found = vcpu->arch.mmu.xlate(vcpu, eaddr, &pte, data);
+		page_found = vcpu->arch.mmu.xlate(vcpu, eaddr, &pte, data, iswrite);
 	} else {
 		pte.may_execute = true;
 		pte.may_read = true;
@@ -355,6 +454,7 @@ int kvmppc_handle_pagefault(struct kvm_run *run, struct kvm_vcpu *vcpu,
 		pte.raddr = eaddr & KVM_PAM;
 		pte.eaddr = eaddr;
 		pte.vpage = eaddr >> 12;
+		pte.page_size = MMU_PAGE_64K;
 	}
 
 	switch (vcpu->arch.shared->msr & (MSR_DR|MSR_IR)) {
@@ -388,22 +488,18 @@ int kvmppc_handle_pagefault(struct kvm_run *run, struct kvm_vcpu *vcpu,
 
 	if (page_found == -ENOENT) {
 		/* Page not found in guest PTE entries */
-		struct kvmppc_book3s_shadow_vcpu *svcpu = svcpu_get(vcpu);
 		vcpu->arch.shared->dar = kvmppc_get_fault_dar(vcpu);
-		vcpu->arch.shared->dsisr = svcpu->fault_dsisr;
+		vcpu->arch.shared->dsisr = vcpu->arch.fault_dsisr;
 		vcpu->arch.shared->msr |=
-			(svcpu->shadow_srr1 & 0x00000000f8000000ULL);
-		svcpu_put(svcpu);
+			vcpu->arch.shadow_srr1 & 0x00000000f8000000ULL;
 		kvmppc_book3s_queue_irqprio(vcpu, vec);
 	} else if (page_found == -EPERM) {
 		/* Storage protection */
-		struct kvmppc_book3s_shadow_vcpu *svcpu = svcpu_get(vcpu);
 		vcpu->arch.shared->dar = kvmppc_get_fault_dar(vcpu);
-		vcpu->arch.shared->dsisr = svcpu->fault_dsisr & ~DSISR_NOHPTE;
+		vcpu->arch.shared->dsisr = vcpu->arch.fault_dsisr & ~DSISR_NOHPTE;
 		vcpu->arch.shared->dsisr |= DSISR_PROTFAULT;
 		vcpu->arch.shared->msr |=
-			svcpu->shadow_srr1 & 0x00000000f8000000ULL;
-		svcpu_put(svcpu);
+			vcpu->arch.shadow_srr1 & 0x00000000f8000000ULL;
 		kvmppc_book3s_queue_irqprio(vcpu, vec);
 	} else if (page_found == -EINVAL) {
 		/* Page not found in guest SLB */
@@ -411,12 +507,20 @@ int kvmppc_handle_pagefault(struct kvm_run *run, struct kvm_vcpu *vcpu,
 		kvmppc_book3s_queue_irqprio(vcpu, vec + 0x80);
 	} else if (!is_mmio &&
 		   kvmppc_visible_gfn(vcpu, pte.raddr >> PAGE_SHIFT)) {
+		if (data && !(vcpu->arch.fault_dsisr & DSISR_NOHPTE)) {
+			/*
+			 * There is already a host HPTE there, presumably
+			 * a read-only one for a page the guest thinks
+			 * is writable, so get rid of it first.
+			 */
+			kvmppc_mmu_unmap_page(vcpu, &pte);
+		}
 		/* The guest's PTE is not mapped yet. Map on the host */
-		kvmppc_mmu_map_page(vcpu, &pte);
+		kvmppc_mmu_map_page(vcpu, &pte, iswrite);
 		if (data)
 			vcpu->stat.sp_storage++;
 		else if (vcpu->arch.mmu.is_dcbz32(vcpu) &&
-			(!(vcpu->arch.hflags & BOOK3S_HFLAG_DCBZ32)))
+			 (!(vcpu->arch.hflags & BOOK3S_HFLAG_DCBZ32)))
 			kvmppc_patch_dcbz(vcpu, &pte);
 	} else {
 		/* MMIO */
@@ -619,13 +723,15 @@ static void kvmppc_handle_lost_ext(struct kvm_vcpu *vcpu)
 
 	if (lost_ext & MSR_FP)
 		kvmppc_load_up_fpu();
+#ifdef CONFIG_ALTIVEC
 	if (lost_ext & MSR_VEC)
 		kvmppc_load_up_altivec();
+#endif
 	current->thread.regs->msr |= lost_ext;
 }
 
-int kvmppc_handle_exit(struct kvm_run *run, struct kvm_vcpu *vcpu,
-                       unsigned int exit_nr)
+int kvmppc_handle_exit_pr(struct kvm_run *run, struct kvm_vcpu *vcpu,
+			  unsigned int exit_nr)
 {
 	int r = RESUME_HOST;
 	int s;
@@ -643,25 +749,32 @@ int kvmppc_handle_exit(struct kvm_run *run, struct kvm_vcpu *vcpu,
 	switch (exit_nr) {
 	case BOOK3S_INTERRUPT_INST_STORAGE:
 	{
-		struct kvmppc_book3s_shadow_vcpu *svcpu = svcpu_get(vcpu);
-		ulong shadow_srr1 = svcpu->shadow_srr1;
+		ulong shadow_srr1 = vcpu->arch.shadow_srr1;
 		vcpu->stat.pf_instruc++;
 
 #ifdef CONFIG_PPC_BOOK3S_32
 		/* We set segments as unused segments when invalidating them. So
 		 * treat the respective fault as segment fault. */
-		if (svcpu->sr[kvmppc_get_pc(vcpu) >> SID_SHIFT] == SR_INVALID) {
-			kvmppc_mmu_map_segment(vcpu, kvmppc_get_pc(vcpu));
-			r = RESUME_GUEST;
+		{
+			struct kvmppc_book3s_shadow_vcpu *svcpu;
+			u32 sr;
+
+			svcpu = svcpu_get(vcpu);
+			sr = svcpu->sr[kvmppc_get_pc(vcpu) >> SID_SHIFT];
 			svcpu_put(svcpu);
-			break;
+			if (sr == SR_INVALID) {
+				kvmppc_mmu_map_segment(vcpu, kvmppc_get_pc(vcpu));
+				r = RESUME_GUEST;
+				break;
+			}
 		}
 #endif
-		svcpu_put(svcpu);
 
 		/* only care about PTEG not found errors, but leave NX alone */
 		if (shadow_srr1 & 0x40000000) {
+			int idx = srcu_read_lock(&vcpu->kvm->srcu);
 			r = kvmppc_handle_pagefault(run, vcpu, kvmppc_get_pc(vcpu), exit_nr);
+			srcu_read_unlock(&vcpu->kvm->srcu, idx);
 			vcpu->stat.sp_instruc++;
 		} else if (vcpu->arch.mmu.is_dcbz32(vcpu) &&
 			  (!(vcpu->arch.hflags & BOOK3S_HFLAG_DCBZ32))) {
@@ -682,25 +795,36 @@ int kvmppc_handle_exit(struct kvm_run *run, struct kvm_vcpu *vcpu,
 	case BOOK3S_INTERRUPT_DATA_STORAGE:
 	{
 		ulong dar = kvmppc_get_fault_dar(vcpu);
-		struct kvmppc_book3s_shadow_vcpu *svcpu = svcpu_get(vcpu);
-		u32 fault_dsisr = svcpu->fault_dsisr;
+		u32 fault_dsisr = vcpu->arch.fault_dsisr;
 		vcpu->stat.pf_storage++;
 
 #ifdef CONFIG_PPC_BOOK3S_32
 		/* We set segments as unused segments when invalidating them. So
 		 * treat the respective fault as segment fault. */
-		if ((svcpu->sr[dar >> SID_SHIFT]) == SR_INVALID) {
-			kvmppc_mmu_map_segment(vcpu, dar);
-			r = RESUME_GUEST;
+		{
+			struct kvmppc_book3s_shadow_vcpu *svcpu;
+			u32 sr;
+
+			svcpu = svcpu_get(vcpu);
+			sr = svcpu->sr[dar >> SID_SHIFT];
 			svcpu_put(svcpu);
-			break;
+			if (sr == SR_INVALID) {
+				kvmppc_mmu_map_segment(vcpu, dar);
+				r = RESUME_GUEST;
+				break;
+			}
 		}
 #endif
-		svcpu_put(svcpu);
 
-		/* The only case we need to handle is missing shadow PTEs */
-		if (fault_dsisr & DSISR_NOHPTE) {
+		/*
+		 * We need to handle missing shadow PTEs, and
+		 * protection faults due to us mapping a page read-only
+		 * when the guest thinks it is writable.
+		 */
+		if (fault_dsisr & (DSISR_NOHPTE | DSISR_PROTFAULT)) {
+			int idx = srcu_read_lock(&vcpu->kvm->srcu);
 			r = kvmppc_handle_pagefault(run, vcpu, dar, exit_nr);
+			srcu_read_unlock(&vcpu->kvm->srcu, idx);
 		} else {
 			vcpu->arch.shared->dar = dar;
 			vcpu->arch.shared->dsisr = fault_dsisr;
@@ -743,13 +867,10 @@ int kvmppc_handle_exit(struct kvm_run *run, struct kvm_vcpu *vcpu,
 	case BOOK3S_INTERRUPT_H_EMUL_ASSIST:
 	{
 		enum emulation_result er;
-		struct kvmppc_book3s_shadow_vcpu *svcpu;
 		ulong flags;
 
 program_interrupt:
-		svcpu = svcpu_get(vcpu);
-		flags = svcpu->shadow_srr1 & 0x1f0000ull;
-		svcpu_put(svcpu);
+		flags = vcpu->arch.shadow_srr1 & 0x1f0000ull;
 
 		if (vcpu->arch.shared->msr & MSR_PR) {
 #ifdef EXIT_DEBUG
@@ -798,7 +919,7 @@ program_interrupt:
 			ulong cmd = kvmppc_get_gpr(vcpu, 3);
 			int i;
 
-#ifdef CONFIG_KVM_BOOK3S_64_PR
+#ifdef CONFIG_PPC_BOOK3S_64
 			if (kvmppc_h_pr(vcpu, cmd) == EMULATE_DONE) {
 				r = RESUME_GUEST;
 				break;
@@ -881,9 +1002,7 @@ program_interrupt:
 		break;
 	default:
 	{
-		struct kvmppc_book3s_shadow_vcpu *svcpu = svcpu_get(vcpu);
-		ulong shadow_srr1 = svcpu->shadow_srr1;
-		svcpu_put(svcpu);
+		ulong shadow_srr1 = vcpu->arch.shadow_srr1;
 		/* Ugh - bork here! What did we get? */
 		printk(KERN_EMERG "exit_nr=0x%x | pc=0x%lx | msr=0x%lx\n",
 			exit_nr, kvmppc_get_pc(vcpu), shadow_srr1);
@@ -920,8 +1039,8 @@ program_interrupt:
 	return r;
 }
 
-int kvm_arch_vcpu_ioctl_get_sregs(struct kvm_vcpu *vcpu,
-                                  struct kvm_sregs *sregs)
+static int kvm_arch_vcpu_ioctl_get_sregs_pr(struct kvm_vcpu *vcpu,
+					    struct kvm_sregs *sregs)
 {
 	struct kvmppc_vcpu_book3s *vcpu3s = to_book3s(vcpu);
 	int i;
@@ -947,13 +1066,13 @@ int kvm_arch_vcpu_ioctl_get_sregs(struct kvm_vcpu *vcpu,
 	return 0;
 }
 
-int kvm_arch_vcpu_ioctl_set_sregs(struct kvm_vcpu *vcpu,
-                                  struct kvm_sregs *sregs)
+static int kvm_arch_vcpu_ioctl_set_sregs_pr(struct kvm_vcpu *vcpu,
+					    struct kvm_sregs *sregs)
 {
 	struct kvmppc_vcpu_book3s *vcpu3s = to_book3s(vcpu);
 	int i;
 
-	kvmppc_set_pvr(vcpu, sregs->pvr);
+	kvmppc_set_pvr_pr(vcpu, sregs->pvr);
 
 	vcpu3s->sdr1 = sregs->u.s.sdr1;
 	if (vcpu->arch.hflags & BOOK3S_HFLAG_SLB) {
@@ -983,7 +1102,8 @@ int kvm_arch_vcpu_ioctl_set_sregs(struct kvm_vcpu *vcpu,
 	return 0;
 }
 
-int kvmppc_get_one_reg(struct kvm_vcpu *vcpu, u64 id, union kvmppc_one_reg *val)
+static int kvmppc_get_one_reg_pr(struct kvm_vcpu *vcpu, u64 id,
+				 union kvmppc_one_reg *val)
 {
 	int r = 0;
 
@@ -1012,7 +1132,8 @@ int kvmppc_get_one_reg(struct kvm_vcpu *vcpu, u64 id, union kvmppc_one_reg *val)
 	return r;
 }
 
-int kvmppc_set_one_reg(struct kvm_vcpu *vcpu, u64 id, union kvmppc_one_reg *val)
+static int kvmppc_set_one_reg_pr(struct kvm_vcpu *vcpu, u64 id,
+				 union kvmppc_one_reg *val)
 {
 	int r = 0;
 
@@ -1042,28 +1163,30 @@ int kvmppc_set_one_reg(struct kvm_vcpu *vcpu, u64 id, union kvmppc_one_reg *val)
 	return r;
 }
 
-int kvmppc_core_check_processor_compat(void)
-{
-	return 0;
-}
-
-struct kvm_vcpu *kvmppc_core_vcpu_create(struct kvm *kvm, unsigned int id)
+static struct kvm_vcpu *kvmppc_core_vcpu_create_pr(struct kvm *kvm,
+						   unsigned int id)
 {
 	struct kvmppc_vcpu_book3s *vcpu_book3s;
 	struct kvm_vcpu *vcpu;
 	int err = -ENOMEM;
 	unsigned long p;
 
-	vcpu_book3s = vzalloc(sizeof(struct kvmppc_vcpu_book3s));
-	if (!vcpu_book3s)
+	vcpu = kmem_cache_zalloc(kvm_vcpu_cache, GFP_KERNEL);
+	if (!vcpu)
 		goto out;
 
-	vcpu_book3s->shadow_vcpu =
-		kzalloc(sizeof(*vcpu_book3s->shadow_vcpu), GFP_KERNEL);
-	if (!vcpu_book3s->shadow_vcpu)
+	vcpu_book3s = vzalloc(sizeof(struct kvmppc_vcpu_book3s));
+	if (!vcpu_book3s)
 		goto free_vcpu;
+	vcpu->arch.book3s = vcpu_book3s;
+
+#ifdef CONFIG_KVM_BOOK3S_32
+	vcpu->arch.shadow_vcpu =
+		kzalloc(sizeof(*vcpu->arch.shadow_vcpu), GFP_KERNEL);
+	if (!vcpu->arch.shadow_vcpu)
+		goto free_vcpu3s;
+#endif
 
-	vcpu = &vcpu_book3s->vcpu;
 	err = kvm_vcpu_init(vcpu, kvm, id);
 	if (err)
 		goto free_shadow_vcpu;
@@ -1076,13 +1199,19 @@ struct kvm_vcpu *kvmppc_core_vcpu_create(struct kvm *kvm, unsigned int id)
 	vcpu->arch.shared = (void *)(p + PAGE_SIZE - 4096);
 
 #ifdef CONFIG_PPC_BOOK3S_64
-	/* default to book3s_64 (970fx) */
+	/*
+	 * Default to the same as the host if we're on sufficiently
+	 * recent machine that we have 1TB segments;
+	 * otherwise default to PPC970FX.
+	 */
 	vcpu->arch.pvr = 0x3C0301;
+	if (mmu_has_feature(MMU_FTR_1T_SEGMENT))
+		vcpu->arch.pvr = mfspr(SPRN_PVR);
 #else
 	/* default to book3s_32 (750) */
 	vcpu->arch.pvr = 0x84202;
 #endif
-	kvmppc_set_pvr(vcpu, vcpu->arch.pvr);
+	kvmppc_set_pvr_pr(vcpu, vcpu->arch.pvr);
 	vcpu->arch.slb_nr = 64;
 
 	vcpu->arch.shadow_msr = MSR_USER64;
@@ -1096,24 +1225,31 @@ struct kvm_vcpu *kvmppc_core_vcpu_create(struct kvm *kvm, unsigned int id)
 uninit_vcpu:
 	kvm_vcpu_uninit(vcpu);
 free_shadow_vcpu:
-	kfree(vcpu_book3s->shadow_vcpu);
-free_vcpu:
+#ifdef CONFIG_KVM_BOOK3S_32
+	kfree(vcpu->arch.shadow_vcpu);
+free_vcpu3s:
+#endif
 	vfree(vcpu_book3s);
+free_vcpu:
+	kmem_cache_free(kvm_vcpu_cache, vcpu);
 out:
 	return ERR_PTR(err);
 }
 
-void kvmppc_core_vcpu_free(struct kvm_vcpu *vcpu)
+static void kvmppc_core_vcpu_free_pr(struct kvm_vcpu *vcpu)
 {
 	struct kvmppc_vcpu_book3s *vcpu_book3s = to_book3s(vcpu);
 
 	free_page((unsigned long)vcpu->arch.shared & PAGE_MASK);
 	kvm_vcpu_uninit(vcpu);
-	kfree(vcpu_book3s->shadow_vcpu);
+#ifdef CONFIG_KVM_BOOK3S_32
+	kfree(vcpu->arch.shadow_vcpu);
+#endif
 	vfree(vcpu_book3s);
+	kmem_cache_free(kvm_vcpu_cache, vcpu);
 }
 
-int kvmppc_vcpu_run(struct kvm_run *kvm_run, struct kvm_vcpu *vcpu)
+static int kvmppc_vcpu_run_pr(struct kvm_run *kvm_run, struct kvm_vcpu *vcpu)
 {
 	int ret;
 	struct thread_fp_state fp;
@@ -1216,8 +1352,8 @@ out:
 /*
  * Get (and clear) the dirty memory log for a memory slot.
  */
-int kvm_vm_ioctl_get_dirty_log(struct kvm *kvm,
-				      struct kvm_dirty_log *log)
+static int kvm_vm_ioctl_get_dirty_log_pr(struct kvm *kvm,
+					 struct kvm_dirty_log *log)
 {
 	struct kvm_memory_slot *memslot;
 	struct kvm_vcpu *vcpu;
@@ -1252,67 +1388,100 @@ out:
 	return r;
 }
 
-#ifdef CONFIG_PPC64
-int kvm_vm_ioctl_get_smmu_info(struct kvm *kvm, struct kvm_ppc_smmu_info *info)
+static void kvmppc_core_flush_memslot_pr(struct kvm *kvm,
+					 struct kvm_memory_slot *memslot)
 {
-	info->flags = KVM_PPC_1T_SEGMENTS;
-
-	/* SLB is always 64 entries */
-	info->slb_size = 64;
-
-	/* Standard 4k base page size segment */
-	info->sps[0].page_shift = 12;
-	info->sps[0].slb_enc = 0;
-	info->sps[0].enc[0].page_shift = 12;
-	info->sps[0].enc[0].pte_enc = 0;
-
-	/* Standard 16M large page size segment */
-	info->sps[1].page_shift = 24;
-	info->sps[1].slb_enc = SLB_VSID_L;
-	info->sps[1].enc[0].page_shift = 24;
-	info->sps[1].enc[0].pte_enc = 0;
+	return;
+}
 
+static int kvmppc_core_prepare_memory_region_pr(struct kvm *kvm,
+					struct kvm_memory_slot *memslot,
+					struct kvm_userspace_memory_region *mem)
+{
 	return 0;
 }
-#endif /* CONFIG_PPC64 */
 
-void kvmppc_core_free_memslot(struct kvm_memory_slot *free,
-			      struct kvm_memory_slot *dont)
+static void kvmppc_core_commit_memory_region_pr(struct kvm *kvm,
+				struct kvm_userspace_memory_region *mem,
+				const struct kvm_memory_slot *old)
 {
+	return;
 }
 
-int kvmppc_core_create_memslot(struct kvm_memory_slot *slot,
-			       unsigned long npages)
+static void kvmppc_core_free_memslot_pr(struct kvm_memory_slot *free,
+					struct kvm_memory_slot *dont)
 {
-	return 0;
+	return;
 }
 
-int kvmppc_core_prepare_memory_region(struct kvm *kvm,
-				      struct kvm_memory_slot *memslot,
-				      struct kvm_userspace_memory_region *mem)
+static int kvmppc_core_create_memslot_pr(struct kvm_memory_slot *slot,
+					 unsigned long npages)
 {
 	return 0;
 }
 
-void kvmppc_core_commit_memory_region(struct kvm *kvm,
-				struct kvm_userspace_memory_region *mem,
-				const struct kvm_memory_slot *old)
+
+#ifdef CONFIG_PPC64
+static int kvm_vm_ioctl_get_smmu_info_pr(struct kvm *kvm,
+					 struct kvm_ppc_smmu_info *info)
 {
-}
+	long int i;
+	struct kvm_vcpu *vcpu;
+
+	info->flags = 0;
+
+	/* SLB is always 64 entries */
+	info->slb_size = 64;
+
+	/* Standard 4k base page size segment */
+	info->sps[0].page_shift = 12;
+	info->sps[0].slb_enc = 0;
+	info->sps[0].enc[0].page_shift = 12;
+	info->sps[0].enc[0].pte_enc = 0;
+
+	/*
+	 * 64k large page size.
+	 * We only want to put this in if the CPUs we're emulating
+	 * support it, but unfortunately we don't have a vcpu easily
+	 * to hand here to test.  Just pick the first vcpu, and if
+	 * that doesn't exist yet, report the minimum capability,
+	 * i.e., no 64k pages.
+	 * 1T segment support goes along with 64k pages.
+	 */
+	i = 1;
+	vcpu = kvm_get_vcpu(kvm, 0);
+	if (vcpu && (vcpu->arch.hflags & BOOK3S_HFLAG_MULTI_PGSIZE)) {
+		info->flags = KVM_PPC_1T_SEGMENTS;
+		info->sps[i].page_shift = 16;
+		info->sps[i].slb_enc = SLB_VSID_L | SLB_VSID_LP_01;
+		info->sps[i].enc[0].page_shift = 16;
+		info->sps[i].enc[0].pte_enc = 1;
+		++i;
+	}
+
+	/* Standard 16M large page size segment */
+	info->sps[i].page_shift = 24;
+	info->sps[i].slb_enc = SLB_VSID_L;
+	info->sps[i].enc[0].page_shift = 24;
+	info->sps[i].enc[0].pte_enc = 0;
 
-void kvmppc_core_flush_memslot(struct kvm *kvm, struct kvm_memory_slot *memslot)
+	return 0;
+}
+#else
+static int kvm_vm_ioctl_get_smmu_info_pr(struct kvm *kvm,
+					 struct kvm_ppc_smmu_info *info)
 {
+	/* We should not get called */
+	BUG();
 }
+#endif /* CONFIG_PPC64 */
 
 static unsigned int kvm_global_user_count = 0;
 static DEFINE_SPINLOCK(kvm_global_user_count_lock);
 
-int kvmppc_core_init_vm(struct kvm *kvm)
+static int kvmppc_core_init_vm_pr(struct kvm *kvm)
 {
-#ifdef CONFIG_PPC64
-	INIT_LIST_HEAD(&kvm->arch.spapr_tce_tables);
-	INIT_LIST_HEAD(&kvm->arch.rtas_tokens);
-#endif
+	mutex_init(&kvm->arch.hpt_mutex);
 
 	if (firmware_has_feature(FW_FEATURE_SET_MODE)) {
 		spin_lock(&kvm_global_user_count_lock);
@@ -1323,7 +1492,7 @@ int kvmppc_core_init_vm(struct kvm *kvm)
 	return 0;
 }
 
-void kvmppc_core_destroy_vm(struct kvm *kvm)
+static void kvmppc_core_destroy_vm_pr(struct kvm *kvm)
 {
 #ifdef CONFIG_PPC64
 	WARN_ON(!list_empty(&kvm->arch.spapr_tce_tables));
@@ -1338,26 +1507,81 @@ void kvmppc_core_destroy_vm(struct kvm *kvm)
 	}
 }
 
-static int kvmppc_book3s_init(void)
+static int kvmppc_core_check_processor_compat_pr(void)
 {
-	int r;
+	/* we are always compatible */
+	return 0;
+}
 
-	r = kvm_init(NULL, sizeof(struct kvmppc_vcpu_book3s), 0,
-		     THIS_MODULE);
+static long kvm_arch_vm_ioctl_pr(struct file *filp,
+				 unsigned int ioctl, unsigned long arg)
+{
+	return -ENOTTY;
+}
 
-	if (r)
+static struct kvmppc_ops kvm_ops_pr = {
+	.get_sregs = kvm_arch_vcpu_ioctl_get_sregs_pr,
+	.set_sregs = kvm_arch_vcpu_ioctl_set_sregs_pr,
+	.get_one_reg = kvmppc_get_one_reg_pr,
+	.set_one_reg = kvmppc_set_one_reg_pr,
+	.vcpu_load   = kvmppc_core_vcpu_load_pr,
+	.vcpu_put    = kvmppc_core_vcpu_put_pr,
+	.set_msr     = kvmppc_set_msr_pr,
+	.vcpu_run    = kvmppc_vcpu_run_pr,
+	.vcpu_create = kvmppc_core_vcpu_create_pr,
+	.vcpu_free   = kvmppc_core_vcpu_free_pr,
+	.check_requests = kvmppc_core_check_requests_pr,
+	.get_dirty_log = kvm_vm_ioctl_get_dirty_log_pr,
+	.flush_memslot = kvmppc_core_flush_memslot_pr,
+	.prepare_memory_region = kvmppc_core_prepare_memory_region_pr,
+	.commit_memory_region = kvmppc_core_commit_memory_region_pr,
+	.unmap_hva = kvm_unmap_hva_pr,
+	.unmap_hva_range = kvm_unmap_hva_range_pr,
+	.age_hva  = kvm_age_hva_pr,
+	.test_age_hva = kvm_test_age_hva_pr,
+	.set_spte_hva = kvm_set_spte_hva_pr,
+	.mmu_destroy  = kvmppc_mmu_destroy_pr,
+	.free_memslot = kvmppc_core_free_memslot_pr,
+	.create_memslot = kvmppc_core_create_memslot_pr,
+	.init_vm = kvmppc_core_init_vm_pr,
+	.destroy_vm = kvmppc_core_destroy_vm_pr,
+	.get_smmu_info = kvm_vm_ioctl_get_smmu_info_pr,
+	.emulate_op = kvmppc_core_emulate_op_pr,
+	.emulate_mtspr = kvmppc_core_emulate_mtspr_pr,
+	.emulate_mfspr = kvmppc_core_emulate_mfspr_pr,
+	.fast_vcpu_kick = kvm_vcpu_kick,
+	.arch_vm_ioctl  = kvm_arch_vm_ioctl_pr,
+};
+
+
+int kvmppc_book3s_init_pr(void)
+{
+	int r;
+
+	r = kvmppc_core_check_processor_compat_pr();
+	if (r < 0)
 		return r;
 
-	r = kvmppc_mmu_hpte_sysinit();
+	kvm_ops_pr.owner = THIS_MODULE;
+	kvmppc_pr_ops = &kvm_ops_pr;
 
+	r = kvmppc_mmu_hpte_sysinit();
 	return r;
 }
 
-static void kvmppc_book3s_exit(void)
+void kvmppc_book3s_exit_pr(void)
 {
+	kvmppc_pr_ops = NULL;
 	kvmppc_mmu_hpte_sysexit();
-	kvm_exit();
 }
 
-module_init(kvmppc_book3s_init);
-module_exit(kvmppc_book3s_exit);
+/*
+ * We only support separate modules for book3s 64
+ */
+#ifdef CONFIG_PPC_BOOK3S_64
+
+module_init(kvmppc_book3s_init_pr);
+module_exit(kvmppc_book3s_exit_pr);
+
+MODULE_LICENSE("GPL");
+#endif
diff --git a/arch/powerpc/kvm/book3s_pr_papr.c b/arch/powerpc/kvm/book3s_pr_papr.c
index da0e0bc268bd..5efa97b993d8 100644
--- a/arch/powerpc/kvm/book3s_pr_papr.c
+++ b/arch/powerpc/kvm/book3s_pr_papr.c
@@ -21,6 +21,8 @@
 #include <asm/kvm_ppc.h>
 #include <asm/kvm_book3s.h>
 
+#define HPTE_SIZE	16		/* bytes per HPT entry */
+
 static unsigned long get_pteg_addr(struct kvm_vcpu *vcpu, long pte_index)
 {
 	struct kvmppc_vcpu_book3s *vcpu_book3s = to_book3s(vcpu);
@@ -40,32 +42,41 @@ static int kvmppc_h_pr_enter(struct kvm_vcpu *vcpu)
 	long pte_index = kvmppc_get_gpr(vcpu, 5);
 	unsigned long pteg[2 * 8];
 	unsigned long pteg_addr, i, *hpte;
+	long int ret;
 
+	i = pte_index & 7;
 	pte_index &= ~7UL;
 	pteg_addr = get_pteg_addr(vcpu, pte_index);
 
+	mutex_lock(&vcpu->kvm->arch.hpt_mutex);
 	copy_from_user(pteg, (void __user *)pteg_addr, sizeof(pteg));
 	hpte = pteg;
 
+	ret = H_PTEG_FULL;
 	if (likely((flags & H_EXACT) == 0)) {
-		pte_index &= ~7UL;
 		for (i = 0; ; ++i) {
 			if (i == 8)
-				return H_PTEG_FULL;
+				goto done;
 			if ((*hpte & HPTE_V_VALID) == 0)
 				break;
 			hpte += 2;
 		}
 	} else {
-		i = kvmppc_get_gpr(vcpu, 5) & 7UL;
 		hpte += i * 2;
+		if (*hpte & HPTE_V_VALID)
+			goto done;
 	}
 
 	hpte[0] = kvmppc_get_gpr(vcpu, 6);
 	hpte[1] = kvmppc_get_gpr(vcpu, 7);
-	copy_to_user((void __user *)pteg_addr, pteg, sizeof(pteg));
-	kvmppc_set_gpr(vcpu, 3, H_SUCCESS);
+	pteg_addr += i * HPTE_SIZE;
+	copy_to_user((void __user *)pteg_addr, hpte, HPTE_SIZE);
 	kvmppc_set_gpr(vcpu, 4, pte_index | i);
+	ret = H_SUCCESS;
+
+ done:
+	mutex_unlock(&vcpu->kvm->arch.hpt_mutex);
+	kvmppc_set_gpr(vcpu, 3, ret);
 
 	return EMULATE_DONE;
 }
@@ -77,26 +88,31 @@ static int kvmppc_h_pr_remove(struct kvm_vcpu *vcpu)
 	unsigned long avpn = kvmppc_get_gpr(vcpu, 6);
 	unsigned long v = 0, pteg, rb;
 	unsigned long pte[2];
+	long int ret;
 
 	pteg = get_pteg_addr(vcpu, pte_index);
+	mutex_lock(&vcpu->kvm->arch.hpt_mutex);
 	copy_from_user(pte, (void __user *)pteg, sizeof(pte));
 
+	ret = H_NOT_FOUND;
 	if ((pte[0] & HPTE_V_VALID) == 0 ||
 	    ((flags & H_AVPN) && (pte[0] & ~0x7fUL) != avpn) ||
-	    ((flags & H_ANDCOND) && (pte[0] & avpn) != 0)) {
-		kvmppc_set_gpr(vcpu, 3, H_NOT_FOUND);
-		return EMULATE_DONE;
-	}
+	    ((flags & H_ANDCOND) && (pte[0] & avpn) != 0))
+		goto done;
 
 	copy_to_user((void __user *)pteg, &v, sizeof(v));
 
 	rb = compute_tlbie_rb(pte[0], pte[1], pte_index);
 	vcpu->arch.mmu.tlbie(vcpu, rb, rb & 1 ? true : false);
 
-	kvmppc_set_gpr(vcpu, 3, H_SUCCESS);
+	ret = H_SUCCESS;
 	kvmppc_set_gpr(vcpu, 4, pte[0]);
 	kvmppc_set_gpr(vcpu, 5, pte[1]);
 
+ done:
+	mutex_unlock(&vcpu->kvm->arch.hpt_mutex);
+	kvmppc_set_gpr(vcpu, 3, ret);
+
 	return EMULATE_DONE;
 }
 
@@ -124,6 +140,7 @@ static int kvmppc_h_pr_bulk_remove(struct kvm_vcpu *vcpu)
 	int paramnr = 4;
 	int ret = H_SUCCESS;
 
+	mutex_lock(&vcpu->kvm->arch.hpt_mutex);
 	for (i = 0; i < H_BULK_REMOVE_MAX_BATCH; i++) {
 		unsigned long tsh = kvmppc_get_gpr(vcpu, paramnr+(2*i));
 		unsigned long tsl = kvmppc_get_gpr(vcpu, paramnr+(2*i)+1);
@@ -172,6 +189,7 @@ static int kvmppc_h_pr_bulk_remove(struct kvm_vcpu *vcpu)
 		}
 		kvmppc_set_gpr(vcpu, paramnr+(2*i), tsh);
 	}
+	mutex_unlock(&vcpu->kvm->arch.hpt_mutex);
 	kvmppc_set_gpr(vcpu, 3, ret);
 
 	return EMULATE_DONE;
@@ -184,15 +202,16 @@ static int kvmppc_h_pr_protect(struct kvm_vcpu *vcpu)
 	unsigned long avpn = kvmppc_get_gpr(vcpu, 6);
 	unsigned long rb, pteg, r, v;
 	unsigned long pte[2];
+	long int ret;
 
 	pteg = get_pteg_addr(vcpu, pte_index);
+	mutex_lock(&vcpu->kvm->arch.hpt_mutex);
 	copy_from_user(pte, (void __user *)pteg, sizeof(pte));
 
+	ret = H_NOT_FOUND;
 	if ((pte[0] & HPTE_V_VALID) == 0 ||
-	    ((flags & H_AVPN) && (pte[0] & ~0x7fUL) != avpn)) {
-		kvmppc_set_gpr(vcpu, 3, H_NOT_FOUND);
-		return EMULATE_DONE;
-	}
+	    ((flags & H_AVPN) && (pte[0] & ~0x7fUL) != avpn))
+		goto done;
 
 	v = pte[0];
 	r = pte[1];
@@ -207,8 +226,11 @@ static int kvmppc_h_pr_protect(struct kvm_vcpu *vcpu)
 	rb = compute_tlbie_rb(v, r, pte_index);
 	vcpu->arch.mmu.tlbie(vcpu, rb, rb & 1 ? true : false);
 	copy_to_user((void __user *)pteg, pte, sizeof(pte));
+	ret = H_SUCCESS;
 
-	kvmppc_set_gpr(vcpu, 3, H_SUCCESS);
+ done:
+	mutex_unlock(&vcpu->kvm->arch.hpt_mutex);
+	kvmppc_set_gpr(vcpu, 3, ret);
 
 	return EMULATE_DONE;
 }
diff --git a/arch/powerpc/kvm/book3s_rmhandlers.S b/arch/powerpc/kvm/book3s_rmhandlers.S
index 8f7633e3afb8..a38c4c9edab8 100644
--- a/arch/powerpc/kvm/book3s_rmhandlers.S
+++ b/arch/powerpc/kvm/book3s_rmhandlers.S
@@ -38,32 +38,6 @@
 
 #define FUNC(name) 		GLUE(.,name)
 
-	.globl	kvmppc_skip_interrupt
-kvmppc_skip_interrupt:
-	/*
-	 * Here all GPRs are unchanged from when the interrupt happened
-	 * except for r13, which is saved in SPRG_SCRATCH0.
-	 */
-	mfspr	r13, SPRN_SRR0
-	addi	r13, r13, 4
-	mtspr	SPRN_SRR0, r13
-	GET_SCRATCH0(r13)
-	rfid
-	b	.
-
-	.globl	kvmppc_skip_Hinterrupt
-kvmppc_skip_Hinterrupt:
-	/*
-	 * Here all GPRs are unchanged from when the interrupt happened
-	 * except for r13, which is saved in SPRG_SCRATCH0.
-	 */
-	mfspr	r13, SPRN_HSRR0
-	addi	r13, r13, 4
-	mtspr	SPRN_HSRR0, r13
-	GET_SCRATCH0(r13)
-	hrfid
-	b	.
-
 #elif defined(CONFIG_PPC_BOOK3S_32)
 
 #define FUNC(name)		name
@@ -179,11 +153,15 @@ _GLOBAL(kvmppc_entry_trampoline)
 
 	li	r6, MSR_IR | MSR_DR
 	andc	r6, r5, r6	/* Clear DR and IR in MSR value */
+#ifdef CONFIG_PPC_BOOK3S_32
 	/*
 	 * Set EE in HOST_MSR so that it's enabled when we get into our
-	 * C exit handler function
+	 * C exit handler function.  On 64-bit we delay enabling
+	 * interrupts until we have finished transferring stuff
+	 * to or from the PACA.
 	 */
 	ori	r5, r5, MSR_EE
+#endif
 	mtsrr0	r7
 	mtsrr1	r6
 	RFI
diff --git a/arch/powerpc/kvm/book3s_rtas.c b/arch/powerpc/kvm/book3s_rtas.c
index 3219ba895246..cf95cdef73c9 100644
--- a/arch/powerpc/kvm/book3s_rtas.c
+++ b/arch/powerpc/kvm/book3s_rtas.c
@@ -260,6 +260,7 @@ fail:
 	 */
 	return rc;
 }
+EXPORT_SYMBOL_GPL(kvmppc_rtas_hcall);
 
 void kvmppc_rtas_tokens_free(struct kvm *kvm)
 {
diff --git a/arch/powerpc/kvm/book3s_segment.S b/arch/powerpc/kvm/book3s_segment.S
index 1abe4788191a..bc50c97751d3 100644
--- a/arch/powerpc/kvm/book3s_segment.S
+++ b/arch/powerpc/kvm/book3s_segment.S
@@ -161,8 +161,8 @@ kvmppc_handler_trampoline_enter_end:
 .global kvmppc_handler_trampoline_exit
 kvmppc_handler_trampoline_exit:
 
-.global kvmppc_interrupt
-kvmppc_interrupt:
+.global kvmppc_interrupt_pr
+kvmppc_interrupt_pr:
 
 	/* Register usage at this point:
 	 *
diff --git a/arch/powerpc/kvm/book3s_xics.c b/arch/powerpc/kvm/book3s_xics.c
index a3a5cb8ee7ea..02a17dcf1610 100644
--- a/arch/powerpc/kvm/book3s_xics.c
+++ b/arch/powerpc/kvm/book3s_xics.c
@@ -818,7 +818,7 @@ int kvmppc_xics_hcall(struct kvm_vcpu *vcpu, u32 req)
 	}
 
 	/* Check for real mode returning too hard */
-	if (xics->real_mode)
+	if (xics->real_mode && is_kvmppc_hv_enabled(vcpu->kvm))
 		return kvmppc_xics_rm_complete(vcpu, req);
 
 	switch (req) {
@@ -840,6 +840,7 @@ int kvmppc_xics_hcall(struct kvm_vcpu *vcpu, u32 req)
 
 	return rc;
 }
+EXPORT_SYMBOL_GPL(kvmppc_xics_hcall);
 
 
 /* -- Initialisation code etc. -- */
@@ -1250,13 +1251,13 @@ static int kvmppc_xics_create(struct kvm_device *dev, u32 type)
 
 	xics_debugfs_init(xics);
 
-#ifdef CONFIG_KVM_BOOK3S_64_HV
+#ifdef CONFIG_KVM_BOOK3S_HV_POSSIBLE
 	if (cpu_has_feature(CPU_FTR_ARCH_206)) {
 		/* Enable real mode support */
 		xics->real_mode = ENABLE_REALMODE;
 		xics->real_mode_dbg = DEBUG_REALMODE;
 	}
-#endif /* CONFIG_KVM_BOOK3S_64_HV */
+#endif /* CONFIG_KVM_BOOK3S_HV_POSSIBLE */
 
 	return 0;
 }
diff --git a/arch/powerpc/kvm/booke.c b/arch/powerpc/kvm/booke.c
index 5133199f6cb7..53e65a210b9a 100644
--- a/arch/powerpc/kvm/booke.c
+++ b/arch/powerpc/kvm/booke.c
@@ -40,7 +40,9 @@
 
 #include "timing.h"
 #include "booke.h"
-#include "trace.h"
+
+#define CREATE_TRACE_POINTS
+#include "trace_booke.h"
 
 unsigned long kvmppc_booke_handlers;
 
@@ -133,6 +135,29 @@ static void kvmppc_vcpu_sync_fpu(struct kvm_vcpu *vcpu)
 #endif
 }
 
+static void kvmppc_vcpu_sync_debug(struct kvm_vcpu *vcpu)
+{
+	/* Synchronize guest's desire to get debug interrupts into shadow MSR */
+#ifndef CONFIG_KVM_BOOKE_HV
+	vcpu->arch.shadow_msr &= ~MSR_DE;
+	vcpu->arch.shadow_msr |= vcpu->arch.shared->msr & MSR_DE;
+#endif
+
+	/* Force enable debug interrupts when user space wants to debug */
+	if (vcpu->guest_debug) {
+#ifdef CONFIG_KVM_BOOKE_HV
+		/*
+		 * Since there is no shadow MSR, sync MSR_DE into the guest
+		 * visible MSR.
+		 */
+		vcpu->arch.shared->msr |= MSR_DE;
+#else
+		vcpu->arch.shadow_msr |= MSR_DE;
+		vcpu->arch.shared->msr &= ~MSR_DE;
+#endif
+	}
+}
+
 /*
  * Helper function for "full" MSR writes.  No need to call this if only
  * EE/CE/ME/DE/RI are changing.
@@ -150,6 +175,7 @@ void kvmppc_set_msr(struct kvm_vcpu *vcpu, u32 new_msr)
 	kvmppc_mmu_msr_notify(vcpu, old_msr);
 	kvmppc_vcpu_sync_spe(vcpu);
 	kvmppc_vcpu_sync_fpu(vcpu);
+	kvmppc_vcpu_sync_debug(vcpu);
 }
 
 static void kvmppc_booke_queue_irqprio(struct kvm_vcpu *vcpu,
@@ -655,6 +681,7 @@ int kvmppc_core_check_requests(struct kvm_vcpu *vcpu)
 int kvmppc_vcpu_run(struct kvm_run *kvm_run, struct kvm_vcpu *vcpu)
 {
 	int ret, s;
+	struct thread_struct thread;
 #ifdef CONFIG_PPC_FPU
 	struct thread_fp_state fp;
 	int fpexc_mode;
@@ -695,6 +722,12 @@ int kvmppc_vcpu_run(struct kvm_run *kvm_run, struct kvm_vcpu *vcpu)
 	kvmppc_load_guest_fp(vcpu);
 #endif
 
+	/* Switch to guest debug context */
+	thread.debug = vcpu->arch.shadow_dbg_reg;
+	switch_booke_debug_regs(&thread);
+	thread.debug = current->thread.debug;
+	current->thread.debug = vcpu->arch.shadow_dbg_reg;
+
 	kvmppc_fix_ee_before_entry();
 
 	ret = __kvmppc_vcpu_run(kvm_run, vcpu);
@@ -702,6 +735,10 @@ int kvmppc_vcpu_run(struct kvm_run *kvm_run, struct kvm_vcpu *vcpu)
 	/* No need for kvm_guest_exit. It's done in handle_exit.
 	   We also get here with interrupts enabled. */
 
+	/* Switch back to user space debug context */
+	switch_booke_debug_regs(&thread);
+	current->thread.debug = thread.debug;
+
 #ifdef CONFIG_PPC_FPU
 	kvmppc_save_guest_fp(vcpu);
 
@@ -757,6 +794,30 @@ static int emulation_exit(struct kvm_run *run, struct kvm_vcpu *vcpu)
 	}
 }
 
+static int kvmppc_handle_debug(struct kvm_run *run, struct kvm_vcpu *vcpu)
+{
+	struct debug_reg *dbg_reg = &(vcpu->arch.shadow_dbg_reg);
+	u32 dbsr = vcpu->arch.dbsr;
+
+	run->debug.arch.status = 0;
+	run->debug.arch.address = vcpu->arch.pc;
+
+	if (dbsr & (DBSR_IAC1 | DBSR_IAC2 | DBSR_IAC3 | DBSR_IAC4)) {
+		run->debug.arch.status |= KVMPPC_DEBUG_BREAKPOINT;
+	} else {
+		if (dbsr & (DBSR_DAC1W | DBSR_DAC2W))
+			run->debug.arch.status |= KVMPPC_DEBUG_WATCH_WRITE;
+		else if (dbsr & (DBSR_DAC1R | DBSR_DAC2R))
+			run->debug.arch.status |= KVMPPC_DEBUG_WATCH_READ;
+		if (dbsr & (DBSR_DAC1R | DBSR_DAC1W))
+			run->debug.arch.address = dbg_reg->dac1;
+		else if (dbsr & (DBSR_DAC2R | DBSR_DAC2W))
+			run->debug.arch.address = dbg_reg->dac2;
+	}
+
+	return RESUME_HOST;
+}
+
 static void kvmppc_fill_pt_regs(struct pt_regs *regs)
 {
 	ulong r1, ip, msr, lr;
@@ -817,6 +878,11 @@ static void kvmppc_restart_interrupt(struct kvm_vcpu *vcpu,
 	case BOOKE_INTERRUPT_CRITICAL:
 		unknown_exception(&regs);
 		break;
+	case BOOKE_INTERRUPT_DEBUG:
+		/* Save DBSR before preemption is enabled */
+		vcpu->arch.dbsr = mfspr(SPRN_DBSR);
+		kvmppc_clear_dbsr();
+		break;
 	}
 }
 
@@ -1134,18 +1200,10 @@ int kvmppc_handle_exit(struct kvm_run *run, struct kvm_vcpu *vcpu,
 	}
 
 	case BOOKE_INTERRUPT_DEBUG: {
-		u32 dbsr;
-
-		vcpu->arch.pc = mfspr(SPRN_CSRR0);
-
-		/* clear IAC events in DBSR register */
-		dbsr = mfspr(SPRN_DBSR);
-		dbsr &= DBSR_IAC1 | DBSR_IAC2 | DBSR_IAC3 | DBSR_IAC4;
-		mtspr(SPRN_DBSR, dbsr);
-
-		run->exit_reason = KVM_EXIT_DEBUG;
+		r = kvmppc_handle_debug(run, vcpu);
+		if (r == RESUME_HOST)
+			run->exit_reason = KVM_EXIT_DEBUG;
 		kvmppc_account_exit(vcpu, DEBUG_EXITS);
-		r = RESUME_HOST;
 		break;
 	}
 
@@ -1196,7 +1254,7 @@ int kvm_arch_vcpu_setup(struct kvm_vcpu *vcpu)
 	kvmppc_set_msr(vcpu, 0);
 
 #ifndef CONFIG_KVM_BOOKE_HV
-	vcpu->arch.shadow_msr = MSR_USER | MSR_DE | MSR_IS | MSR_DS;
+	vcpu->arch.shadow_msr = MSR_USER | MSR_IS | MSR_DS;
 	vcpu->arch.shadow_pid = 1;
 	vcpu->arch.shared->msr = 0;
 #endif
@@ -1358,7 +1416,7 @@ static int set_sregs_arch206(struct kvm_vcpu *vcpu,
 	return 0;
 }
 
-void kvmppc_get_sregs_ivor(struct kvm_vcpu *vcpu, struct kvm_sregs *sregs)
+int kvmppc_get_sregs_ivor(struct kvm_vcpu *vcpu, struct kvm_sregs *sregs)
 {
 	sregs->u.e.features |= KVM_SREGS_E_IVOR;
 
@@ -1378,6 +1436,7 @@ void kvmppc_get_sregs_ivor(struct kvm_vcpu *vcpu, struct kvm_sregs *sregs)
 	sregs->u.e.ivor_low[13] = vcpu->arch.ivor[BOOKE_IRQPRIO_DTLB_MISS];
 	sregs->u.e.ivor_low[14] = vcpu->arch.ivor[BOOKE_IRQPRIO_ITLB_MISS];
 	sregs->u.e.ivor_low[15] = vcpu->arch.ivor[BOOKE_IRQPRIO_DEBUG];
+	return 0;
 }
 
 int kvmppc_set_sregs_ivor(struct kvm_vcpu *vcpu, struct kvm_sregs *sregs)
@@ -1412,8 +1471,7 @@ int kvm_arch_vcpu_ioctl_get_sregs(struct kvm_vcpu *vcpu,
 
 	get_sregs_base(vcpu, sregs);
 	get_sregs_arch206(vcpu, sregs);
-	kvmppc_core_get_sregs(vcpu, sregs);
-	return 0;
+	return vcpu->kvm->arch.kvm_ops->get_sregs(vcpu, sregs);
 }
 
 int kvm_arch_vcpu_ioctl_set_sregs(struct kvm_vcpu *vcpu,
@@ -1432,7 +1490,7 @@ int kvm_arch_vcpu_ioctl_set_sregs(struct kvm_vcpu *vcpu,
 	if (ret < 0)
 		return ret;
 
-	return kvmppc_core_set_sregs(vcpu, sregs);
+	return vcpu->kvm->arch.kvm_ops->set_sregs(vcpu, sregs);
 }
 
 int kvm_vcpu_ioctl_get_one_reg(struct kvm_vcpu *vcpu, struct kvm_one_reg *reg)
@@ -1440,7 +1498,6 @@ int kvm_vcpu_ioctl_get_one_reg(struct kvm_vcpu *vcpu, struct kvm_one_reg *reg)
 	int r = 0;
 	union kvmppc_one_reg val;
 	int size;
-	long int i;
 
 	size = one_reg_size(reg->id);
 	if (size > sizeof(val))
@@ -1448,16 +1505,24 @@ int kvm_vcpu_ioctl_get_one_reg(struct kvm_vcpu *vcpu, struct kvm_one_reg *reg)
 
 	switch (reg->id) {
 	case KVM_REG_PPC_IAC1:
+		val = get_reg_val(reg->id, vcpu->arch.dbg_reg.iac1);
+		break;
 	case KVM_REG_PPC_IAC2:
+		val = get_reg_val(reg->id, vcpu->arch.dbg_reg.iac2);
+		break;
+#if CONFIG_PPC_ADV_DEBUG_IACS > 2
 	case KVM_REG_PPC_IAC3:
+		val = get_reg_val(reg->id, vcpu->arch.dbg_reg.iac3);
+		break;
 	case KVM_REG_PPC_IAC4:
-		i = reg->id - KVM_REG_PPC_IAC1;
-		val = get_reg_val(reg->id, vcpu->arch.dbg_reg.iac[i]);
+		val = get_reg_val(reg->id, vcpu->arch.dbg_reg.iac4);
 		break;
+#endif
 	case KVM_REG_PPC_DAC1:
+		val = get_reg_val(reg->id, vcpu->arch.dbg_reg.dac1);
+		break;
 	case KVM_REG_PPC_DAC2:
-		i = reg->id - KVM_REG_PPC_DAC1;
-		val = get_reg_val(reg->id, vcpu->arch.dbg_reg.dac[i]);
+		val = get_reg_val(reg->id, vcpu->arch.dbg_reg.dac2);
 		break;
 	case KVM_REG_PPC_EPR: {
 		u32 epr = get_guest_epr(vcpu);
@@ -1476,10 +1541,13 @@ int kvm_vcpu_ioctl_get_one_reg(struct kvm_vcpu *vcpu, struct kvm_one_reg *reg)
 		val = get_reg_val(reg->id, vcpu->arch.tsr);
 		break;
 	case KVM_REG_PPC_DEBUG_INST:
-		val = get_reg_val(reg->id, KVMPPC_INST_EHPRIV);
+		val = get_reg_val(reg->id, KVMPPC_INST_EHPRIV_DEBUG);
+		break;
+	case KVM_REG_PPC_VRSAVE:
+		val = get_reg_val(reg->id, vcpu->arch.vrsave);
 		break;
 	default:
-		r = kvmppc_get_one_reg(vcpu, reg->id, &val);
+		r = vcpu->kvm->arch.kvm_ops->get_one_reg(vcpu, reg->id, &val);
 		break;
 	}
 
@@ -1497,7 +1565,6 @@ int kvm_vcpu_ioctl_set_one_reg(struct kvm_vcpu *vcpu, struct kvm_one_reg *reg)
 	int r = 0;
 	union kvmppc_one_reg val;
 	int size;
-	long int i;
 
 	size = one_reg_size(reg->id);
 	if (size > sizeof(val))
@@ -1508,16 +1575,24 @@ int kvm_vcpu_ioctl_set_one_reg(struct kvm_vcpu *vcpu, struct kvm_one_reg *reg)
 
 	switch (reg->id) {
 	case KVM_REG_PPC_IAC1:
+		vcpu->arch.dbg_reg.iac1 = set_reg_val(reg->id, val);
+		break;
 	case KVM_REG_PPC_IAC2:
+		vcpu->arch.dbg_reg.iac2 = set_reg_val(reg->id, val);
+		break;
+#if CONFIG_PPC_ADV_DEBUG_IACS > 2
 	case KVM_REG_PPC_IAC3:
+		vcpu->arch.dbg_reg.iac3 = set_reg_val(reg->id, val);
+		break;
 	case KVM_REG_PPC_IAC4:
-		i = reg->id - KVM_REG_PPC_IAC1;
-		vcpu->arch.dbg_reg.iac[i] = set_reg_val(reg->id, val);
+		vcpu->arch.dbg_reg.iac4 = set_reg_val(reg->id, val);
 		break;
+#endif
 	case KVM_REG_PPC_DAC1:
+		vcpu->arch.dbg_reg.dac1 = set_reg_val(reg->id, val);
+		break;
 	case KVM_REG_PPC_DAC2:
-		i = reg->id - KVM_REG_PPC_DAC1;
-		vcpu->arch.dbg_reg.dac[i] = set_reg_val(reg->id, val);
+		vcpu->arch.dbg_reg.dac2 = set_reg_val(reg->id, val);
 		break;
 	case KVM_REG_PPC_EPR: {
 		u32 new_epr = set_reg_val(reg->id, val);
@@ -1551,20 +1626,17 @@ int kvm_vcpu_ioctl_set_one_reg(struct kvm_vcpu *vcpu, struct kvm_one_reg *reg)
 		kvmppc_set_tcr(vcpu, tcr);
 		break;
 	}
+	case KVM_REG_PPC_VRSAVE:
+		vcpu->arch.vrsave = set_reg_val(reg->id, val);
+		break;
 	default:
-		r = kvmppc_set_one_reg(vcpu, reg->id, &val);
+		r = vcpu->kvm->arch.kvm_ops->set_one_reg(vcpu, reg->id, &val);
 		break;
 	}
 
 	return r;
 }
 
-int kvm_arch_vcpu_ioctl_set_guest_debug(struct kvm_vcpu *vcpu,
-					 struct kvm_guest_debug *dbg)
-{
-	return -EINVAL;
-}
-
 int kvm_arch_vcpu_ioctl_get_fpu(struct kvm_vcpu *vcpu, struct kvm_fpu *fpu)
 {
 	return -ENOTSUPP;
@@ -1589,12 +1661,12 @@ int kvm_vm_ioctl_get_dirty_log(struct kvm *kvm, struct kvm_dirty_log *log)
 	return -ENOTSUPP;
 }
 
-void kvmppc_core_free_memslot(struct kvm_memory_slot *free,
+void kvmppc_core_free_memslot(struct kvm *kvm, struct kvm_memory_slot *free,
 			      struct kvm_memory_slot *dont)
 {
 }
 
-int kvmppc_core_create_memslot(struct kvm_memory_slot *slot,
+int kvmppc_core_create_memslot(struct kvm *kvm, struct kvm_memory_slot *slot,
 			       unsigned long npages)
 {
 	return 0;
@@ -1670,6 +1742,157 @@ void kvmppc_decrementer_func(unsigned long data)
 	kvmppc_set_tsr_bits(vcpu, TSR_DIS);
 }
 
+static int kvmppc_booke_add_breakpoint(struct debug_reg *dbg_reg,
+				       uint64_t addr, int index)
+{
+	switch (index) {
+	case 0:
+		dbg_reg->dbcr0 |= DBCR0_IAC1;
+		dbg_reg->iac1 = addr;
+		break;
+	case 1:
+		dbg_reg->dbcr0 |= DBCR0_IAC2;
+		dbg_reg->iac2 = addr;
+		break;
+#if CONFIG_PPC_ADV_DEBUG_IACS > 2
+	case 2:
+		dbg_reg->dbcr0 |= DBCR0_IAC3;
+		dbg_reg->iac3 = addr;
+		break;
+	case 3:
+		dbg_reg->dbcr0 |= DBCR0_IAC4;
+		dbg_reg->iac4 = addr;
+		break;
+#endif
+	default:
+		return -EINVAL;
+	}
+
+	dbg_reg->dbcr0 |= DBCR0_IDM;
+	return 0;
+}
+
+static int kvmppc_booke_add_watchpoint(struct debug_reg *dbg_reg, uint64_t addr,
+				       int type, int index)
+{
+	switch (index) {
+	case 0:
+		if (type & KVMPPC_DEBUG_WATCH_READ)
+			dbg_reg->dbcr0 |= DBCR0_DAC1R;
+		if (type & KVMPPC_DEBUG_WATCH_WRITE)
+			dbg_reg->dbcr0 |= DBCR0_DAC1W;
+		dbg_reg->dac1 = addr;
+		break;
+	case 1:
+		if (type & KVMPPC_DEBUG_WATCH_READ)
+			dbg_reg->dbcr0 |= DBCR0_DAC2R;
+		if (type & KVMPPC_DEBUG_WATCH_WRITE)
+			dbg_reg->dbcr0 |= DBCR0_DAC2W;
+		dbg_reg->dac2 = addr;
+		break;
+	default:
+		return -EINVAL;
+	}
+
+	dbg_reg->dbcr0 |= DBCR0_IDM;
+	return 0;
+}
+void kvm_guest_protect_msr(struct kvm_vcpu *vcpu, ulong prot_bitmap, bool set)
+{
+	/* XXX: Add similar MSR protection for BookE-PR */
+#ifdef CONFIG_KVM_BOOKE_HV
+	BUG_ON(prot_bitmap & ~(MSRP_UCLEP | MSRP_DEP | MSRP_PMMP));
+	if (set) {
+		if (prot_bitmap & MSR_UCLE)
+			vcpu->arch.shadow_msrp |= MSRP_UCLEP;
+		if (prot_bitmap & MSR_DE)
+			vcpu->arch.shadow_msrp |= MSRP_DEP;
+		if (prot_bitmap & MSR_PMM)
+			vcpu->arch.shadow_msrp |= MSRP_PMMP;
+	} else {
+		if (prot_bitmap & MSR_UCLE)
+			vcpu->arch.shadow_msrp &= ~MSRP_UCLEP;
+		if (prot_bitmap & MSR_DE)
+			vcpu->arch.shadow_msrp &= ~MSRP_DEP;
+		if (prot_bitmap & MSR_PMM)
+			vcpu->arch.shadow_msrp &= ~MSRP_PMMP;
+	}
+#endif
+}
+
+int kvm_arch_vcpu_ioctl_set_guest_debug(struct kvm_vcpu *vcpu,
+					 struct kvm_guest_debug *dbg)
+{
+	struct debug_reg *dbg_reg;
+	int n, b = 0, w = 0;
+
+	if (!(dbg->control & KVM_GUESTDBG_ENABLE)) {
+		vcpu->arch.shadow_dbg_reg.dbcr0 = 0;
+		vcpu->guest_debug = 0;
+		kvm_guest_protect_msr(vcpu, MSR_DE, false);
+		return 0;
+	}
+
+	kvm_guest_protect_msr(vcpu, MSR_DE, true);
+	vcpu->guest_debug = dbg->control;
+	vcpu->arch.shadow_dbg_reg.dbcr0 = 0;
+	/* Set DBCR0_EDM in guest visible DBCR0 register. */
+	vcpu->arch.dbg_reg.dbcr0 = DBCR0_EDM;
+
+	if (vcpu->guest_debug & KVM_GUESTDBG_SINGLESTEP)
+		vcpu->arch.shadow_dbg_reg.dbcr0 |= DBCR0_IDM | DBCR0_IC;
+
+	/* Code below handles only HW breakpoints */
+	dbg_reg = &(vcpu->arch.shadow_dbg_reg);
+
+#ifdef CONFIG_KVM_BOOKE_HV
+	/*
+	 * On BookE-HV (e500mc) the guest is always executed with MSR.GS=1
+	 * DBCR1 and DBCR2 are set to trigger debug events when MSR.PR is 0
+	 */
+	dbg_reg->dbcr1 = 0;
+	dbg_reg->dbcr2 = 0;
+#else
+	/*
+	 * On BookE-PR (e500v2) the guest is always executed with MSR.PR=1
+	 * We set DBCR1 and DBCR2 to only trigger debug events when MSR.PR
+	 * is set.
+	 */
+	dbg_reg->dbcr1 = DBCR1_IAC1US | DBCR1_IAC2US | DBCR1_IAC3US |
+			  DBCR1_IAC4US;
+	dbg_reg->dbcr2 = DBCR2_DAC1US | DBCR2_DAC2US;
+#endif
+
+	if (!(vcpu->guest_debug & KVM_GUESTDBG_USE_HW_BP))
+		return 0;
+
+	for (n = 0; n < (KVMPPC_BOOKE_IAC_NUM + KVMPPC_BOOKE_DAC_NUM); n++) {
+		uint64_t addr = dbg->arch.bp[n].addr;
+		uint32_t type = dbg->arch.bp[n].type;
+
+		if (type == KVMPPC_DEBUG_NONE)
+			continue;
+
+		if (type & !(KVMPPC_DEBUG_WATCH_READ |
+			     KVMPPC_DEBUG_WATCH_WRITE |
+			     KVMPPC_DEBUG_BREAKPOINT))
+			return -EINVAL;
+
+		if (type & KVMPPC_DEBUG_BREAKPOINT) {
+			/* Setting H/W breakpoint */
+			if (kvmppc_booke_add_breakpoint(dbg_reg, addr, b++))
+				return -EINVAL;
+		} else {
+			/* Setting H/W watchpoint */
+			if (kvmppc_booke_add_watchpoint(dbg_reg, addr,
+							type, w++))
+				return -EINVAL;
+		}
+	}
+
+	return 0;
+}
+
 void kvmppc_booke_vcpu_load(struct kvm_vcpu *vcpu, int cpu)
 {
 	vcpu->cpu = smp_processor_id();
@@ -1680,6 +1903,44 @@ void kvmppc_booke_vcpu_put(struct kvm_vcpu *vcpu)
 {
 	current->thread.kvm_vcpu = NULL;
 	vcpu->cpu = -1;
+
+	/* Clear pending debug event in DBSR */
+	kvmppc_clear_dbsr();
+}
+
+void kvmppc_mmu_destroy(struct kvm_vcpu *vcpu)
+{
+	vcpu->kvm->arch.kvm_ops->mmu_destroy(vcpu);
+}
+
+int kvmppc_core_init_vm(struct kvm *kvm)
+{
+	return kvm->arch.kvm_ops->init_vm(kvm);
+}
+
+struct kvm_vcpu *kvmppc_core_vcpu_create(struct kvm *kvm, unsigned int id)
+{
+	return kvm->arch.kvm_ops->vcpu_create(kvm, id);
+}
+
+void kvmppc_core_vcpu_free(struct kvm_vcpu *vcpu)
+{
+	vcpu->kvm->arch.kvm_ops->vcpu_free(vcpu);
+}
+
+void kvmppc_core_destroy_vm(struct kvm *kvm)
+{
+	kvm->arch.kvm_ops->destroy_vm(kvm);
+}
+
+void kvmppc_core_vcpu_load(struct kvm_vcpu *vcpu, int cpu)
+{
+	vcpu->kvm->arch.kvm_ops->vcpu_load(vcpu, cpu);
+}
+
+void kvmppc_core_vcpu_put(struct kvm_vcpu *vcpu)
+{
+	vcpu->kvm->arch.kvm_ops->vcpu_put(vcpu);
 }
 
 int __init kvmppc_booke_init(void)
diff --git a/arch/powerpc/kvm/booke.h b/arch/powerpc/kvm/booke.h
index 5fd1ba693579..09bfd9bc7cf8 100644
--- a/arch/powerpc/kvm/booke.h
+++ b/arch/powerpc/kvm/booke.h
@@ -99,6 +99,30 @@ enum int_class {
 
 void kvmppc_set_pending_interrupt(struct kvm_vcpu *vcpu, enum int_class type);
 
+extern void kvmppc_mmu_destroy_44x(struct kvm_vcpu *vcpu);
+extern int kvmppc_core_emulate_op_44x(struct kvm_run *run, struct kvm_vcpu *vcpu,
+				      unsigned int inst, int *advance);
+extern int kvmppc_core_emulate_mtspr_44x(struct kvm_vcpu *vcpu, int sprn,
+					 ulong spr_val);
+extern int kvmppc_core_emulate_mfspr_44x(struct kvm_vcpu *vcpu, int sprn,
+					 ulong *spr_val);
+extern void kvmppc_mmu_destroy_e500(struct kvm_vcpu *vcpu);
+extern int kvmppc_core_emulate_op_e500(struct kvm_run *run,
+				       struct kvm_vcpu *vcpu,
+				       unsigned int inst, int *advance);
+extern int kvmppc_core_emulate_mtspr_e500(struct kvm_vcpu *vcpu, int sprn,
+					  ulong spr_val);
+extern int kvmppc_core_emulate_mfspr_e500(struct kvm_vcpu *vcpu, int sprn,
+					  ulong *spr_val);
+extern void kvmppc_mmu_destroy_e500(struct kvm_vcpu *vcpu);
+extern int kvmppc_core_emulate_op_e500(struct kvm_run *run,
+				       struct kvm_vcpu *vcpu,
+				       unsigned int inst, int *advance);
+extern int kvmppc_core_emulate_mtspr_e500(struct kvm_vcpu *vcpu, int sprn,
+					  ulong spr_val);
+extern int kvmppc_core_emulate_mfspr_e500(struct kvm_vcpu *vcpu, int sprn,
+					  ulong *spr_val);
+
 /*
  * Load up guest vcpu FP state if it's needed.
  * It also set the MSR_FP in thread so that host know
@@ -129,4 +153,9 @@ static inline void kvmppc_save_guest_fp(struct kvm_vcpu *vcpu)
 		giveup_fpu(current);
 #endif
 }
+
+static inline void kvmppc_clear_dbsr(void)
+{
+	mtspr(SPRN_DBSR, mfspr(SPRN_DBSR));
+}
 #endif /* __KVM_BOOKE_H__ */
diff --git a/arch/powerpc/kvm/e500.c b/arch/powerpc/kvm/e500.c
index ce6b73c29612..497b142f651c 100644
--- a/arch/powerpc/kvm/e500.c
+++ b/arch/powerpc/kvm/e500.c
@@ -305,7 +305,7 @@ void kvmppc_core_load_guest_debugstate(struct kvm_vcpu *vcpu)
 {
 }
 
-void kvmppc_core_vcpu_load(struct kvm_vcpu *vcpu, int cpu)
+static void kvmppc_core_vcpu_load_e500(struct kvm_vcpu *vcpu, int cpu)
 {
 	kvmppc_booke_vcpu_load(vcpu, cpu);
 
@@ -313,7 +313,7 @@ void kvmppc_core_vcpu_load(struct kvm_vcpu *vcpu, int cpu)
 	kvmppc_e500_recalc_shadow_pid(to_e500(vcpu));
 }
 
-void kvmppc_core_vcpu_put(struct kvm_vcpu *vcpu)
+static void kvmppc_core_vcpu_put_e500(struct kvm_vcpu *vcpu)
 {
 #ifdef CONFIG_SPE
 	if (vcpu->arch.shadow_msr & MSR_SPE)
@@ -367,7 +367,8 @@ int kvmppc_core_vcpu_setup(struct kvm_vcpu *vcpu)
 	return 0;
 }
 
-void kvmppc_core_get_sregs(struct kvm_vcpu *vcpu, struct kvm_sregs *sregs)
+static int kvmppc_core_get_sregs_e500(struct kvm_vcpu *vcpu,
+				      struct kvm_sregs *sregs)
 {
 	struct kvmppc_vcpu_e500 *vcpu_e500 = to_e500(vcpu);
 
@@ -388,9 +389,11 @@ void kvmppc_core_get_sregs(struct kvm_vcpu *vcpu, struct kvm_sregs *sregs)
 
 	kvmppc_get_sregs_ivor(vcpu, sregs);
 	kvmppc_get_sregs_e500_tlb(vcpu, sregs);
+	return 0;
 }
 
-int kvmppc_core_set_sregs(struct kvm_vcpu *vcpu, struct kvm_sregs *sregs)
+static int kvmppc_core_set_sregs_e500(struct kvm_vcpu *vcpu,
+				      struct kvm_sregs *sregs)
 {
 	struct kvmppc_vcpu_e500 *vcpu_e500 = to_e500(vcpu);
 	int ret;
@@ -425,21 +428,22 @@ int kvmppc_core_set_sregs(struct kvm_vcpu *vcpu, struct kvm_sregs *sregs)
 	return kvmppc_set_sregs_ivor(vcpu, sregs);
 }
 
-int kvmppc_get_one_reg(struct kvm_vcpu *vcpu, u64 id,
-			union kvmppc_one_reg *val)
+static int kvmppc_get_one_reg_e500(struct kvm_vcpu *vcpu, u64 id,
+				   union kvmppc_one_reg *val)
 {
 	int r = kvmppc_get_one_reg_e500_tlb(vcpu, id, val);
 	return r;
 }
 
-int kvmppc_set_one_reg(struct kvm_vcpu *vcpu, u64 id,
-		       union kvmppc_one_reg *val)
+static int kvmppc_set_one_reg_e500(struct kvm_vcpu *vcpu, u64 id,
+				   union kvmppc_one_reg *val)
 {
 	int r = kvmppc_get_one_reg_e500_tlb(vcpu, id, val);
 	return r;
 }
 
-struct kvm_vcpu *kvmppc_core_vcpu_create(struct kvm *kvm, unsigned int id)
+static struct kvm_vcpu *kvmppc_core_vcpu_create_e500(struct kvm *kvm,
+						     unsigned int id)
 {
 	struct kvmppc_vcpu_e500 *vcpu_e500;
 	struct kvm_vcpu *vcpu;
@@ -481,7 +485,7 @@ out:
 	return ERR_PTR(err);
 }
 
-void kvmppc_core_vcpu_free(struct kvm_vcpu *vcpu)
+static void kvmppc_core_vcpu_free_e500(struct kvm_vcpu *vcpu)
 {
 	struct kvmppc_vcpu_e500 *vcpu_e500 = to_e500(vcpu);
 
@@ -492,15 +496,32 @@ void kvmppc_core_vcpu_free(struct kvm_vcpu *vcpu)
 	kmem_cache_free(kvm_vcpu_cache, vcpu_e500);
 }
 
-int kvmppc_core_init_vm(struct kvm *kvm)
+static int kvmppc_core_init_vm_e500(struct kvm *kvm)
 {
 	return 0;
 }
 
-void kvmppc_core_destroy_vm(struct kvm *kvm)
+static void kvmppc_core_destroy_vm_e500(struct kvm *kvm)
 {
 }
 
+static struct kvmppc_ops kvm_ops_e500 = {
+	.get_sregs = kvmppc_core_get_sregs_e500,
+	.set_sregs = kvmppc_core_set_sregs_e500,
+	.get_one_reg = kvmppc_get_one_reg_e500,
+	.set_one_reg = kvmppc_set_one_reg_e500,
+	.vcpu_load   = kvmppc_core_vcpu_load_e500,
+	.vcpu_put    = kvmppc_core_vcpu_put_e500,
+	.vcpu_create = kvmppc_core_vcpu_create_e500,
+	.vcpu_free   = kvmppc_core_vcpu_free_e500,
+	.mmu_destroy  = kvmppc_mmu_destroy_e500,
+	.init_vm = kvmppc_core_init_vm_e500,
+	.destroy_vm = kvmppc_core_destroy_vm_e500,
+	.emulate_op = kvmppc_core_emulate_op_e500,
+	.emulate_mtspr = kvmppc_core_emulate_mtspr_e500,
+	.emulate_mfspr = kvmppc_core_emulate_mfspr_e500,
+};
+
 static int __init kvmppc_e500_init(void)
 {
 	int r, i;
@@ -512,11 +533,11 @@ static int __init kvmppc_e500_init(void)
 
 	r = kvmppc_core_check_processor_compat();
 	if (r)
-		return r;
+		goto err_out;
 
 	r = kvmppc_booke_init();
 	if (r)
-		return r;
+		goto err_out;
 
 	/* copy extra E500 exception handlers */
 	ivor[0] = mfspr(SPRN_IVOR32);
@@ -534,11 +555,19 @@ static int __init kvmppc_e500_init(void)
 	flush_icache_range(kvmppc_booke_handlers, kvmppc_booke_handlers +
 			   ivor[max_ivor] + handler_len);
 
-	return kvm_init(NULL, sizeof(struct kvmppc_vcpu_e500), 0, THIS_MODULE);
+	r = kvm_init(NULL, sizeof(struct kvmppc_vcpu_e500), 0, THIS_MODULE);
+	if (r)
+		goto err_out;
+	kvm_ops_e500.owner = THIS_MODULE;
+	kvmppc_pr_ops = &kvm_ops_e500;
+
+err_out:
+	return r;
 }
 
 static void __exit kvmppc_e500_exit(void)
 {
+	kvmppc_pr_ops = NULL;
 	kvmppc_booke_exit();
 }
 
diff --git a/arch/powerpc/kvm/e500.h b/arch/powerpc/kvm/e500.h
index c2e5e98453a6..4fd9650eb018 100644
--- a/arch/powerpc/kvm/e500.h
+++ b/arch/powerpc/kvm/e500.h
@@ -117,7 +117,7 @@ static inline struct kvmppc_vcpu_e500 *to_e500(struct kvm_vcpu *vcpu)
 #define E500_TLB_USER_PERM_MASK (MAS3_UX|MAS3_UR|MAS3_UW)
 #define E500_TLB_SUPER_PERM_MASK (MAS3_SX|MAS3_SR|MAS3_SW)
 #define MAS2_ATTRIB_MASK \
-	  (MAS2_X0 | MAS2_X1)
+	  (MAS2_X0 | MAS2_X1 | MAS2_E | MAS2_G)
 #define MAS3_ATTRIB_MASK \
 	  (MAS3_U0 | MAS3_U1 | MAS3_U2 | MAS3_U3 \
 	   | E500_TLB_USER_PERM_MASK | E500_TLB_SUPER_PERM_MASK)
diff --git a/arch/powerpc/kvm/e500_emulate.c b/arch/powerpc/kvm/e500_emulate.c
index b10a01243abd..89b7f821f6c4 100644
--- a/arch/powerpc/kvm/e500_emulate.c
+++ b/arch/powerpc/kvm/e500_emulate.c
@@ -26,6 +26,7 @@
 #define XOP_TLBRE   946
 #define XOP_TLBWE   978
 #define XOP_TLBILX  18
+#define XOP_EHPRIV  270
 
 #ifdef CONFIG_KVM_E500MC
 static int dbell2prio(ulong param)
@@ -82,8 +83,28 @@ static int kvmppc_e500_emul_msgsnd(struct kvm_vcpu *vcpu, int rb)
 }
 #endif
 
-int kvmppc_core_emulate_op(struct kvm_run *run, struct kvm_vcpu *vcpu,
-                           unsigned int inst, int *advance)
+static int kvmppc_e500_emul_ehpriv(struct kvm_run *run, struct kvm_vcpu *vcpu,
+				   unsigned int inst, int *advance)
+{
+	int emulated = EMULATE_DONE;
+
+	switch (get_oc(inst)) {
+	case EHPRIV_OC_DEBUG:
+		run->exit_reason = KVM_EXIT_DEBUG;
+		run->debug.arch.address = vcpu->arch.pc;
+		run->debug.arch.status = 0;
+		kvmppc_account_exit(vcpu, DEBUG_EXITS);
+		emulated = EMULATE_EXIT_USER;
+		*advance = 0;
+		break;
+	default:
+		emulated = EMULATE_FAIL;
+	}
+	return emulated;
+}
+
+int kvmppc_core_emulate_op_e500(struct kvm_run *run, struct kvm_vcpu *vcpu,
+				unsigned int inst, int *advance)
 {
 	int emulated = EMULATE_DONE;
 	int ra = get_ra(inst);
@@ -130,6 +151,11 @@ int kvmppc_core_emulate_op(struct kvm_run *run, struct kvm_vcpu *vcpu,
 			emulated = kvmppc_e500_emul_tlbivax(vcpu, ea);
 			break;
 
+		case XOP_EHPRIV:
+			emulated = kvmppc_e500_emul_ehpriv(run, vcpu, inst,
+							   advance);
+			break;
+
 		default:
 			emulated = EMULATE_FAIL;
 		}
@@ -146,7 +172,7 @@ int kvmppc_core_emulate_op(struct kvm_run *run, struct kvm_vcpu *vcpu,
 	return emulated;
 }
 
-int kvmppc_core_emulate_mtspr(struct kvm_vcpu *vcpu, int sprn, ulong spr_val)
+int kvmppc_core_emulate_mtspr_e500(struct kvm_vcpu *vcpu, int sprn, ulong spr_val)
 {
 	struct kvmppc_vcpu_e500 *vcpu_e500 = to_e500(vcpu);
 	int emulated = EMULATE_DONE;
@@ -237,7 +263,7 @@ int kvmppc_core_emulate_mtspr(struct kvm_vcpu *vcpu, int sprn, ulong spr_val)
 	return emulated;
 }
 
-int kvmppc_core_emulate_mfspr(struct kvm_vcpu *vcpu, int sprn, ulong *spr_val)
+int kvmppc_core_emulate_mfspr_e500(struct kvm_vcpu *vcpu, int sprn, ulong *spr_val)
 {
 	struct kvmppc_vcpu_e500 *vcpu_e500 = to_e500(vcpu);
 	int emulated = EMULATE_DONE;
diff --git a/arch/powerpc/kvm/e500_mmu.c b/arch/powerpc/kvm/e500_mmu.c
index 6d6f153b6c1d..ebca6b88ea5e 100644
--- a/arch/powerpc/kvm/e500_mmu.c
+++ b/arch/powerpc/kvm/e500_mmu.c
@@ -32,7 +32,7 @@
 #include <asm/kvm_ppc.h>
 
 #include "e500.h"
-#include "trace.h"
+#include "trace_booke.h"
 #include "timing.h"
 #include "e500_mmu_host.h"
 
@@ -536,7 +536,7 @@ gpa_t kvmppc_mmu_xlate(struct kvm_vcpu *vcpu, unsigned int index,
 	return get_tlb_raddr(gtlbe) | (eaddr & pgmask);
 }
 
-void kvmppc_mmu_destroy(struct kvm_vcpu *vcpu)
+void kvmppc_mmu_destroy_e500(struct kvm_vcpu *vcpu)
 {
 }
 
diff --git a/arch/powerpc/kvm/e500_mmu_host.c b/arch/powerpc/kvm/e500_mmu_host.c
index c65593abae8e..ecf2247b13be 100644
--- a/arch/powerpc/kvm/e500_mmu_host.c
+++ b/arch/powerpc/kvm/e500_mmu_host.c
@@ -32,10 +32,11 @@
 #include <asm/kvm_ppc.h>
 
 #include "e500.h"
-#include "trace.h"
 #include "timing.h"
 #include "e500_mmu_host.h"
 
+#include "trace_booke.h"
+
 #define to_htlb1_esel(esel) (host_tlb_params[1].entries - (esel) - 1)
 
 static struct kvmppc_e500_tlb_params host_tlb_params[E500_TLB_NUM];
@@ -253,6 +254,9 @@ static inline void kvmppc_e500_ref_setup(struct tlbe_ref *ref,
 	ref->pfn = pfn;
 	ref->flags |= E500_TLB_VALID;
 
+	/* Mark the page accessed */
+	kvm_set_pfn_accessed(pfn);
+
 	if (tlbe_is_writable(gtlbe))
 		kvm_set_pfn_dirty(pfn);
 }
diff --git a/arch/powerpc/kvm/e500mc.c b/arch/powerpc/kvm/e500mc.c
index 19c8379575f7..4132cd2fc171 100644
--- a/arch/powerpc/kvm/e500mc.c
+++ b/arch/powerpc/kvm/e500mc.c
@@ -110,7 +110,7 @@ void kvmppc_mmu_msr_notify(struct kvm_vcpu *vcpu, u32 old_msr)
 
 static DEFINE_PER_CPU(struct kvm_vcpu *, last_vcpu_on_cpu);
 
-void kvmppc_core_vcpu_load(struct kvm_vcpu *vcpu, int cpu)
+static void kvmppc_core_vcpu_load_e500mc(struct kvm_vcpu *vcpu, int cpu)
 {
 	struct kvmppc_vcpu_e500 *vcpu_e500 = to_e500(vcpu);
 
@@ -147,7 +147,7 @@ void kvmppc_core_vcpu_load(struct kvm_vcpu *vcpu, int cpu)
 	kvmppc_load_guest_fp(vcpu);
 }
 
-void kvmppc_core_vcpu_put(struct kvm_vcpu *vcpu)
+static void kvmppc_core_vcpu_put_e500mc(struct kvm_vcpu *vcpu)
 {
 	vcpu->arch.eplc = mfspr(SPRN_EPLC);
 	vcpu->arch.epsc = mfspr(SPRN_EPSC);
@@ -204,7 +204,8 @@ int kvmppc_core_vcpu_setup(struct kvm_vcpu *vcpu)
 	return 0;
 }
 
-void kvmppc_core_get_sregs(struct kvm_vcpu *vcpu, struct kvm_sregs *sregs)
+static int kvmppc_core_get_sregs_e500mc(struct kvm_vcpu *vcpu,
+					struct kvm_sregs *sregs)
 {
 	struct kvmppc_vcpu_e500 *vcpu_e500 = to_e500(vcpu);
 
@@ -224,10 +225,11 @@ void kvmppc_core_get_sregs(struct kvm_vcpu *vcpu, struct kvm_sregs *sregs)
 	sregs->u.e.ivor_high[4] = vcpu->arch.ivor[BOOKE_IRQPRIO_DBELL];
 	sregs->u.e.ivor_high[5] = vcpu->arch.ivor[BOOKE_IRQPRIO_DBELL_CRIT];
 
-	kvmppc_get_sregs_ivor(vcpu, sregs);
+	return kvmppc_get_sregs_ivor(vcpu, sregs);
 }
 
-int kvmppc_core_set_sregs(struct kvm_vcpu *vcpu, struct kvm_sregs *sregs)
+static int kvmppc_core_set_sregs_e500mc(struct kvm_vcpu *vcpu,
+					struct kvm_sregs *sregs)
 {
 	struct kvmppc_vcpu_e500 *vcpu_e500 = to_e500(vcpu);
 	int ret;
@@ -260,21 +262,22 @@ int kvmppc_core_set_sregs(struct kvm_vcpu *vcpu, struct kvm_sregs *sregs)
 	return kvmppc_set_sregs_ivor(vcpu, sregs);
 }
 
-int kvmppc_get_one_reg(struct kvm_vcpu *vcpu, u64 id,
-			union kvmppc_one_reg *val)
+static int kvmppc_get_one_reg_e500mc(struct kvm_vcpu *vcpu, u64 id,
+			      union kvmppc_one_reg *val)
 {
 	int r = kvmppc_get_one_reg_e500_tlb(vcpu, id, val);
 	return r;
 }
 
-int kvmppc_set_one_reg(struct kvm_vcpu *vcpu, u64 id,
-		       union kvmppc_one_reg *val)
+static int kvmppc_set_one_reg_e500mc(struct kvm_vcpu *vcpu, u64 id,
+			      union kvmppc_one_reg *val)
 {
 	int r = kvmppc_set_one_reg_e500_tlb(vcpu, id, val);
 	return r;
 }
 
-struct kvm_vcpu *kvmppc_core_vcpu_create(struct kvm *kvm, unsigned int id)
+static struct kvm_vcpu *kvmppc_core_vcpu_create_e500mc(struct kvm *kvm,
+						       unsigned int id)
 {
 	struct kvmppc_vcpu_e500 *vcpu_e500;
 	struct kvm_vcpu *vcpu;
@@ -315,7 +318,7 @@ out:
 	return ERR_PTR(err);
 }
 
-void kvmppc_core_vcpu_free(struct kvm_vcpu *vcpu)
+static void kvmppc_core_vcpu_free_e500mc(struct kvm_vcpu *vcpu)
 {
 	struct kvmppc_vcpu_e500 *vcpu_e500 = to_e500(vcpu);
 
@@ -325,7 +328,7 @@ void kvmppc_core_vcpu_free(struct kvm_vcpu *vcpu)
 	kmem_cache_free(kvm_vcpu_cache, vcpu_e500);
 }
 
-int kvmppc_core_init_vm(struct kvm *kvm)
+static int kvmppc_core_init_vm_e500mc(struct kvm *kvm)
 {
 	int lpid;
 
@@ -337,27 +340,52 @@ int kvmppc_core_init_vm(struct kvm *kvm)
 	return 0;
 }
 
-void kvmppc_core_destroy_vm(struct kvm *kvm)
+static void kvmppc_core_destroy_vm_e500mc(struct kvm *kvm)
 {
 	kvmppc_free_lpid(kvm->arch.lpid);
 }
 
+static struct kvmppc_ops kvm_ops_e500mc = {
+	.get_sregs = kvmppc_core_get_sregs_e500mc,
+	.set_sregs = kvmppc_core_set_sregs_e500mc,
+	.get_one_reg = kvmppc_get_one_reg_e500mc,
+	.set_one_reg = kvmppc_set_one_reg_e500mc,
+	.vcpu_load   = kvmppc_core_vcpu_load_e500mc,
+	.vcpu_put    = kvmppc_core_vcpu_put_e500mc,
+	.vcpu_create = kvmppc_core_vcpu_create_e500mc,
+	.vcpu_free   = kvmppc_core_vcpu_free_e500mc,
+	.mmu_destroy  = kvmppc_mmu_destroy_e500,
+	.init_vm = kvmppc_core_init_vm_e500mc,
+	.destroy_vm = kvmppc_core_destroy_vm_e500mc,
+	.emulate_op = kvmppc_core_emulate_op_e500,
+	.emulate_mtspr = kvmppc_core_emulate_mtspr_e500,
+	.emulate_mfspr = kvmppc_core_emulate_mfspr_e500,
+};
+
 static int __init kvmppc_e500mc_init(void)
 {
 	int r;
 
 	r = kvmppc_booke_init();
 	if (r)
-		return r;
+		goto err_out;
 
 	kvmppc_init_lpid(64);
 	kvmppc_claim_lpid(0); /* host */
 
-	return kvm_init(NULL, sizeof(struct kvmppc_vcpu_e500), 0, THIS_MODULE);
+	r = kvm_init(NULL, sizeof(struct kvmppc_vcpu_e500), 0, THIS_MODULE);
+	if (r)
+		goto err_out;
+	kvm_ops_e500mc.owner = THIS_MODULE;
+	kvmppc_pr_ops = &kvm_ops_e500mc;
+
+err_out:
+	return r;
 }
 
 static void __exit kvmppc_e500mc_exit(void)
 {
+	kvmppc_pr_ops = NULL;
 	kvmppc_booke_exit();
 }
 
diff --git a/arch/powerpc/kvm/emulate.c b/arch/powerpc/kvm/emulate.c
index 751cd45f65a0..2f9a0873b44f 100644
--- a/arch/powerpc/kvm/emulate.c
+++ b/arch/powerpc/kvm/emulate.c
@@ -130,8 +130,8 @@ static int kvmppc_emulate_mtspr(struct kvm_vcpu *vcpu, int sprn, int rs)
 	case SPRN_PIR: break;
 
 	default:
-		emulated = kvmppc_core_emulate_mtspr(vcpu, sprn,
-						     spr_val);
+		emulated = vcpu->kvm->arch.kvm_ops->emulate_mtspr(vcpu, sprn,
+								  spr_val);
 		if (emulated == EMULATE_FAIL)
 			printk(KERN_INFO "mtspr: unknown spr "
 				"0x%x\n", sprn);
@@ -191,8 +191,8 @@ static int kvmppc_emulate_mfspr(struct kvm_vcpu *vcpu, int sprn, int rt)
 		spr_val = kvmppc_get_dec(vcpu, get_tb());
 		break;
 	default:
-		emulated = kvmppc_core_emulate_mfspr(vcpu, sprn,
-						     &spr_val);
+		emulated = vcpu->kvm->arch.kvm_ops->emulate_mfspr(vcpu, sprn,
+								  &spr_val);
 		if (unlikely(emulated == EMULATE_FAIL)) {
 			printk(KERN_INFO "mfspr: unknown spr "
 				"0x%x\n", sprn);
@@ -464,7 +464,8 @@ int kvmppc_emulate_instruction(struct kvm_run *run, struct kvm_vcpu *vcpu)
 	}
 
 	if (emulated == EMULATE_FAIL) {
-		emulated = kvmppc_core_emulate_op(run, vcpu, inst, &advance);
+		emulated = vcpu->kvm->arch.kvm_ops->emulate_op(run, vcpu, inst,
+							       &advance);
 		if (emulated == EMULATE_AGAIN) {
 			advance = 0;
 		} else if (emulated == EMULATE_FAIL) {
@@ -483,3 +484,4 @@ int kvmppc_emulate_instruction(struct kvm_run *run, struct kvm_vcpu *vcpu)
 
 	return emulated;
 }
+EXPORT_SYMBOL_GPL(kvmppc_emulate_instruction);
diff --git a/arch/powerpc/kvm/powerpc.c b/arch/powerpc/kvm/powerpc.c
index 07c0106fab76..9ae97686e9f4 100644
--- a/arch/powerpc/kvm/powerpc.c
+++ b/arch/powerpc/kvm/powerpc.c
@@ -26,6 +26,7 @@
 #include <linux/fs.h>
 #include <linux/slab.h>
 #include <linux/file.h>
+#include <linux/module.h>
 #include <asm/cputable.h>
 #include <asm/uaccess.h>
 #include <asm/kvm_ppc.h>
@@ -39,6 +40,12 @@
 #define CREATE_TRACE_POINTS
 #include "trace.h"
 
+struct kvmppc_ops *kvmppc_hv_ops;
+EXPORT_SYMBOL_GPL(kvmppc_hv_ops);
+struct kvmppc_ops *kvmppc_pr_ops;
+EXPORT_SYMBOL_GPL(kvmppc_pr_ops);
+
+
 int kvm_arch_vcpu_runnable(struct kvm_vcpu *v)
 {
 	return !!(v->arch.pending_exceptions) ||
@@ -50,7 +57,6 @@ int kvm_arch_vcpu_should_kick(struct kvm_vcpu *vcpu)
 	return 1;
 }
 
-#ifndef CONFIG_KVM_BOOK3S_64_HV
 /*
  * Common checks before entering the guest world.  Call with interrupts
  * disabled.
@@ -125,7 +131,7 @@ int kvmppc_prepare_to_enter(struct kvm_vcpu *vcpu)
 
 	return r;
 }
-#endif /* CONFIG_KVM_BOOK3S_64_HV */
+EXPORT_SYMBOL_GPL(kvmppc_prepare_to_enter);
 
 int kvmppc_kvm_pv(struct kvm_vcpu *vcpu)
 {
@@ -179,6 +185,7 @@ int kvmppc_kvm_pv(struct kvm_vcpu *vcpu)
 
 	return r;
 }
+EXPORT_SYMBOL_GPL(kvmppc_kvm_pv);
 
 int kvmppc_sanity_check(struct kvm_vcpu *vcpu)
 {
@@ -192,11 +199,9 @@ int kvmppc_sanity_check(struct kvm_vcpu *vcpu)
 	if ((vcpu->arch.cpu_type != KVM_CPU_3S_64) && vcpu->arch.papr_enabled)
 		goto out;
 
-#ifdef CONFIG_KVM_BOOK3S_64_HV
 	/* HV KVM can only do PAPR mode for now */
-	if (!vcpu->arch.papr_enabled)
+	if (!vcpu->arch.papr_enabled && is_kvmppc_hv_enabled(vcpu->kvm))
 		goto out;
-#endif
 
 #ifdef CONFIG_KVM_BOOKE_HV
 	if (!cpu_has_feature(CPU_FTR_EMB_HV))
@@ -209,6 +214,7 @@ out:
 	vcpu->arch.sane = r;
 	return r ? 0 : -EINVAL;
 }
+EXPORT_SYMBOL_GPL(kvmppc_sanity_check);
 
 int kvmppc_emulate_mmio(struct kvm_run *run, struct kvm_vcpu *vcpu)
 {
@@ -243,6 +249,7 @@ int kvmppc_emulate_mmio(struct kvm_run *run, struct kvm_vcpu *vcpu)
 
 	return r;
 }
+EXPORT_SYMBOL_GPL(kvmppc_emulate_mmio);
 
 int kvm_arch_hardware_enable(void *garbage)
 {
@@ -269,10 +276,35 @@ void kvm_arch_check_processor_compat(void *rtn)
 
 int kvm_arch_init_vm(struct kvm *kvm, unsigned long type)
 {
-	if (type)
-		return -EINVAL;
-
+	struct kvmppc_ops *kvm_ops = NULL;
+	/*
+	 * if we have both HV and PR enabled, default is HV
+	 */
+	if (type == 0) {
+		if (kvmppc_hv_ops)
+			kvm_ops = kvmppc_hv_ops;
+		else
+			kvm_ops = kvmppc_pr_ops;
+		if (!kvm_ops)
+			goto err_out;
+	} else	if (type == KVM_VM_PPC_HV) {
+		if (!kvmppc_hv_ops)
+			goto err_out;
+		kvm_ops = kvmppc_hv_ops;
+	} else if (type == KVM_VM_PPC_PR) {
+		if (!kvmppc_pr_ops)
+			goto err_out;
+		kvm_ops = kvmppc_pr_ops;
+	} else
+		goto err_out;
+
+	if (kvm_ops->owner && !try_module_get(kvm_ops->owner))
+		return -ENOENT;
+
+	kvm->arch.kvm_ops = kvm_ops;
 	return kvmppc_core_init_vm(kvm);
+err_out:
+	return -EINVAL;
 }
 
 void kvm_arch_destroy_vm(struct kvm *kvm)
@@ -292,6 +324,9 @@ void kvm_arch_destroy_vm(struct kvm *kvm)
 	kvmppc_core_destroy_vm(kvm);
 
 	mutex_unlock(&kvm->lock);
+
+	/* drop the module reference */
+	module_put(kvm->arch.kvm_ops->owner);
 }
 
 void kvm_arch_sync_events(struct kvm *kvm)
@@ -301,6 +336,10 @@ void kvm_arch_sync_events(struct kvm *kvm)
 int kvm_dev_ioctl_check_extension(long ext)
 {
 	int r;
+	/* FIXME!!
+	 * Should some of this be vm ioctl ? is it possible now ?
+	 */
+	int hv_enabled = kvmppc_hv_ops ? 1 : 0;
 
 	switch (ext) {
 #ifdef CONFIG_BOOKE
@@ -320,22 +359,26 @@ int kvm_dev_ioctl_check_extension(long ext)
 	case KVM_CAP_DEVICE_CTRL:
 		r = 1;
 		break;
-#ifndef CONFIG_KVM_BOOK3S_64_HV
 	case KVM_CAP_PPC_PAIRED_SINGLES:
 	case KVM_CAP_PPC_OSI:
 	case KVM_CAP_PPC_GET_PVINFO:
 #if defined(CONFIG_KVM_E500V2) || defined(CONFIG_KVM_E500MC)
 	case KVM_CAP_SW_TLB:
 #endif
-#ifdef CONFIG_KVM_MPIC
-	case KVM_CAP_IRQ_MPIC:
-#endif
-		r = 1;
+		/* We support this only for PR */
+		r = !hv_enabled;
 		break;
+#ifdef CONFIG_KVM_MMIO
 	case KVM_CAP_COALESCED_MMIO:
 		r = KVM_COALESCED_MMIO_PAGE_OFFSET;
 		break;
 #endif
+#ifdef CONFIG_KVM_MPIC
+	case KVM_CAP_IRQ_MPIC:
+		r = 1;
+		break;
+#endif
+
 #ifdef CONFIG_PPC_BOOK3S_64
 	case KVM_CAP_SPAPR_TCE:
 	case KVM_CAP_PPC_ALLOC_HTAB:
@@ -346,32 +389,37 @@ int kvm_dev_ioctl_check_extension(long ext)
 		r = 1;
 		break;
 #endif /* CONFIG_PPC_BOOK3S_64 */
-#ifdef CONFIG_KVM_BOOK3S_64_HV
+#ifdef CONFIG_KVM_BOOK3S_HV_POSSIBLE
 	case KVM_CAP_PPC_SMT:
-		r = threads_per_core;
+		if (hv_enabled)
+			r = threads_per_core;
+		else
+			r = 0;
 		break;
 	case KVM_CAP_PPC_RMA:
-		r = 1;
+		r = hv_enabled;
 		/* PPC970 requires an RMA */
-		if (cpu_has_feature(CPU_FTR_ARCH_201))
+		if (r && cpu_has_feature(CPU_FTR_ARCH_201))
 			r = 2;
 		break;
 #endif
 	case KVM_CAP_SYNC_MMU:
-#ifdef CONFIG_KVM_BOOK3S_64_HV
-		r = cpu_has_feature(CPU_FTR_ARCH_206) ? 1 : 0;
+#ifdef CONFIG_KVM_BOOK3S_HV_POSSIBLE
+		if (hv_enabled)
+			r = cpu_has_feature(CPU_FTR_ARCH_206) ? 1 : 0;
+		else
+			r = 0;
 #elif defined(KVM_ARCH_WANT_MMU_NOTIFIER)
 		r = 1;
 #else
 		r = 0;
-		break;
 #endif
-#ifdef CONFIG_KVM_BOOK3S_64_HV
+		break;
+#ifdef CONFIG_KVM_BOOK3S_HV_POSSIBLE
 	case KVM_CAP_PPC_HTAB_FD:
-		r = 1;
+		r = hv_enabled;
 		break;
 #endif
-		break;
 	case KVM_CAP_NR_VCPUS:
 		/*
 		 * Recommending a number of CPUs is somewhat arbitrary; we
@@ -379,11 +427,10 @@ int kvm_dev_ioctl_check_extension(long ext)
 		 * will have secondary threads "offline"), and for other KVM
 		 * implementations just count online CPUs.
 		 */
-#ifdef CONFIG_KVM_BOOK3S_64_HV
-		r = num_present_cpus();
-#else
-		r = num_online_cpus();
-#endif
+		if (hv_enabled)
+			r = num_present_cpus();
+		else
+			r = num_online_cpus();
 		break;
 	case KVM_CAP_MAX_VCPUS:
 		r = KVM_MAX_VCPUS;
@@ -407,15 +454,16 @@ long kvm_arch_dev_ioctl(struct file *filp,
 	return -EINVAL;
 }
 
-void kvm_arch_free_memslot(struct kvm_memory_slot *free,
+void kvm_arch_free_memslot(struct kvm *kvm, struct kvm_memory_slot *free,
 			   struct kvm_memory_slot *dont)
 {
-	kvmppc_core_free_memslot(free, dont);
+	kvmppc_core_free_memslot(kvm, free, dont);
 }
 
-int kvm_arch_create_memslot(struct kvm_memory_slot *slot, unsigned long npages)
+int kvm_arch_create_memslot(struct kvm *kvm, struct kvm_memory_slot *slot,
+			    unsigned long npages)
 {
-	return kvmppc_core_create_memslot(slot, npages);
+	return kvmppc_core_create_memslot(kvm, slot, npages);
 }
 
 void kvm_arch_memslots_updated(struct kvm *kvm)
@@ -659,6 +707,7 @@ int kvmppc_handle_load(struct kvm_run *run, struct kvm_vcpu *vcpu,
 
 	return EMULATE_DO_MMIO;
 }
+EXPORT_SYMBOL_GPL(kvmppc_handle_load);
 
 /* Same as above, but sign extends */
 int kvmppc_handle_loads(struct kvm_run *run, struct kvm_vcpu *vcpu,
@@ -720,6 +769,7 @@ int kvmppc_handle_store(struct kvm_run *run, struct kvm_vcpu *vcpu,
 
 	return EMULATE_DO_MMIO;
 }
+EXPORT_SYMBOL_GPL(kvmppc_handle_store);
 
 int kvm_arch_vcpu_ioctl_run(struct kvm_vcpu *vcpu, struct kvm_run *run)
 {
@@ -1024,52 +1074,12 @@ long kvm_arch_vm_ioctl(struct file *filp,
 		r = kvm_vm_ioctl_create_spapr_tce(kvm, &create_tce);
 		goto out;
 	}
-#endif /* CONFIG_PPC_BOOK3S_64 */
-
-#ifdef CONFIG_KVM_BOOK3S_64_HV
-	case KVM_ALLOCATE_RMA: {
-		struct kvm_allocate_rma rma;
-		struct kvm *kvm = filp->private_data;
-
-		r = kvm_vm_ioctl_allocate_rma(kvm, &rma);
-		if (r >= 0 && copy_to_user(argp, &rma, sizeof(rma)))
-			r = -EFAULT;
-		break;
-	}
-
-	case KVM_PPC_ALLOCATE_HTAB: {
-		u32 htab_order;
-
-		r = -EFAULT;
-		if (get_user(htab_order, (u32 __user *)argp))
-			break;
-		r = kvmppc_alloc_reset_hpt(kvm, &htab_order);
-		if (r)
-			break;
-		r = -EFAULT;
-		if (put_user(htab_order, (u32 __user *)argp))
-			break;
-		r = 0;
-		break;
-	}
-
-	case KVM_PPC_GET_HTAB_FD: {
-		struct kvm_get_htab_fd ghf;
-
-		r = -EFAULT;
-		if (copy_from_user(&ghf, argp, sizeof(ghf)))
-			break;
-		r = kvm_vm_ioctl_get_htab_fd(kvm, &ghf);
-		break;
-	}
-#endif /* CONFIG_KVM_BOOK3S_64_HV */
-
-#ifdef CONFIG_PPC_BOOK3S_64
 	case KVM_PPC_GET_SMMU_INFO: {
 		struct kvm_ppc_smmu_info info;
+		struct kvm *kvm = filp->private_data;
 
 		memset(&info, 0, sizeof(info));
-		r = kvm_vm_ioctl_get_smmu_info(kvm, &info);
+		r = kvm->arch.kvm_ops->get_smmu_info(kvm, &info);
 		if (r >= 0 && copy_to_user(argp, &info, sizeof(info)))
 			r = -EFAULT;
 		break;
@@ -1080,11 +1090,15 @@ long kvm_arch_vm_ioctl(struct file *filp,
 		r = kvm_vm_ioctl_rtas_define_token(kvm, argp);
 		break;
 	}
-#endif /* CONFIG_PPC_BOOK3S_64 */
+	default: {
+		struct kvm *kvm = filp->private_data;
+		r = kvm->arch.kvm_ops->arch_vm_ioctl(filp, ioctl, arg);
+	}
+#else /* CONFIG_PPC_BOOK3S_64 */
 	default:
 		r = -ENOTTY;
+#endif
 	}
-
 out:
 	return r;
 }
@@ -1106,22 +1120,26 @@ long kvmppc_alloc_lpid(void)
 
 	return lpid;
 }
+EXPORT_SYMBOL_GPL(kvmppc_alloc_lpid);
 
 void kvmppc_claim_lpid(long lpid)
 {
 	set_bit(lpid, lpid_inuse);
 }
+EXPORT_SYMBOL_GPL(kvmppc_claim_lpid);
 
 void kvmppc_free_lpid(long lpid)
 {
 	clear_bit(lpid, lpid_inuse);
 }
+EXPORT_SYMBOL_GPL(kvmppc_free_lpid);
 
 void kvmppc_init_lpid(unsigned long nr_lpids_param)
 {
 	nr_lpids = min_t(unsigned long, KVMPPC_NR_LPIDS, nr_lpids_param);
 	memset(lpid_inuse, 0, sizeof(lpid_inuse));
 }
+EXPORT_SYMBOL_GPL(kvmppc_init_lpid);
 
 int kvm_arch_init(void *opaque)
 {
@@ -1130,4 +1148,5 @@ int kvm_arch_init(void *opaque)
 
 void kvm_arch_exit(void)
 {
+
 }
diff --git a/arch/powerpc/kvm/trace.h b/arch/powerpc/kvm/trace.h
index e326489a5420..2e0e67ef3544 100644
--- a/arch/powerpc/kvm/trace.h
+++ b/arch/powerpc/kvm/trace.h
@@ -31,126 +31,6 @@ TRACE_EVENT(kvm_ppc_instr,
 		  __entry->inst, __entry->pc, __entry->emulate)
 );
 
-#ifdef CONFIG_PPC_BOOK3S
-#define kvm_trace_symbol_exit \
-	{0x100, "SYSTEM_RESET"}, \
-	{0x200, "MACHINE_CHECK"}, \
-	{0x300, "DATA_STORAGE"}, \
-	{0x380, "DATA_SEGMENT"}, \
-	{0x400, "INST_STORAGE"}, \
-	{0x480, "INST_SEGMENT"}, \
-	{0x500, "EXTERNAL"}, \
-	{0x501, "EXTERNAL_LEVEL"}, \
-	{0x502, "EXTERNAL_HV"}, \
-	{0x600, "ALIGNMENT"}, \
-	{0x700, "PROGRAM"}, \
-	{0x800, "FP_UNAVAIL"}, \
-	{0x900, "DECREMENTER"}, \
-	{0x980, "HV_DECREMENTER"}, \
-	{0xc00, "SYSCALL"}, \
-	{0xd00, "TRACE"}, \
-	{0xe00, "H_DATA_STORAGE"}, \
-	{0xe20, "H_INST_STORAGE"}, \
-	{0xe40, "H_EMUL_ASSIST"}, \
-	{0xf00, "PERFMON"}, \
-	{0xf20, "ALTIVEC"}, \
-	{0xf40, "VSX"}
-#else
-#define kvm_trace_symbol_exit \
-	{0, "CRITICAL"}, \
-	{1, "MACHINE_CHECK"}, \
-	{2, "DATA_STORAGE"}, \
-	{3, "INST_STORAGE"}, \
-	{4, "EXTERNAL"}, \
-	{5, "ALIGNMENT"}, \
-	{6, "PROGRAM"}, \
-	{7, "FP_UNAVAIL"}, \
-	{8, "SYSCALL"}, \
-	{9, "AP_UNAVAIL"}, \
-	{10, "DECREMENTER"}, \
-	{11, "FIT"}, \
-	{12, "WATCHDOG"}, \
-	{13, "DTLB_MISS"}, \
-	{14, "ITLB_MISS"}, \
-	{15, "DEBUG"}, \
-	{32, "SPE_UNAVAIL"}, \
-	{33, "SPE_FP_DATA"}, \
-	{34, "SPE_FP_ROUND"}, \
-	{35, "PERFORMANCE_MONITOR"}, \
-	{36, "DOORBELL"}, \
-	{37, "DOORBELL_CRITICAL"}, \
-	{38, "GUEST_DBELL"}, \
-	{39, "GUEST_DBELL_CRIT"}, \
-	{40, "HV_SYSCALL"}, \
-	{41, "HV_PRIV"}
-#endif
-
-TRACE_EVENT(kvm_exit,
-	TP_PROTO(unsigned int exit_nr, struct kvm_vcpu *vcpu),
-	TP_ARGS(exit_nr, vcpu),
-
-	TP_STRUCT__entry(
-		__field(	unsigned int,	exit_nr		)
-		__field(	unsigned long,	pc		)
-		__field(	unsigned long,	msr		)
-		__field(	unsigned long,	dar		)
-#ifdef CONFIG_KVM_BOOK3S_PR
-		__field(	unsigned long,	srr1		)
-#endif
-		__field(	unsigned long,	last_inst	)
-	),
-
-	TP_fast_assign(
-#ifdef CONFIG_KVM_BOOK3S_PR
-		struct kvmppc_book3s_shadow_vcpu *svcpu;
-#endif
-		__entry->exit_nr	= exit_nr;
-		__entry->pc		= kvmppc_get_pc(vcpu);
-		__entry->dar		= kvmppc_get_fault_dar(vcpu);
-		__entry->msr		= vcpu->arch.shared->msr;
-#ifdef CONFIG_KVM_BOOK3S_PR
-		svcpu = svcpu_get(vcpu);
-		__entry->srr1		= svcpu->shadow_srr1;
-		svcpu_put(svcpu);
-#endif
-		__entry->last_inst	= vcpu->arch.last_inst;
-	),
-
-	TP_printk("exit=%s"
-		" | pc=0x%lx"
-		" | msr=0x%lx"
-		" | dar=0x%lx"
-#ifdef CONFIG_KVM_BOOK3S_PR
-		" | srr1=0x%lx"
-#endif
-		" | last_inst=0x%lx"
-		,
-		__print_symbolic(__entry->exit_nr, kvm_trace_symbol_exit),
-		__entry->pc,
-		__entry->msr,
-		__entry->dar,
-#ifdef CONFIG_KVM_BOOK3S_PR
-		__entry->srr1,
-#endif
-		__entry->last_inst
-		)
-);
-
-TRACE_EVENT(kvm_unmap_hva,
-	TP_PROTO(unsigned long hva),
-	TP_ARGS(hva),
-
-	TP_STRUCT__entry(
-		__field(	unsigned long,	hva		)
-	),
-
-	TP_fast_assign(
-		__entry->hva		= hva;
-	),
-
-	TP_printk("unmap hva 0x%lx\n", __entry->hva)
-);
-
 TRACE_EVENT(kvm_stlb_inval,
 	TP_PROTO(unsigned int stlb_index),
 	TP_ARGS(stlb_index),
@@ -236,315 +116,6 @@ TRACE_EVENT(kvm_check_requests,
 		__entry->cpu_nr, __entry->requests)
 );
 
-
-/*************************************************************************
- *                         Book3S trace points                           *
- *************************************************************************/
-
-#ifdef CONFIG_KVM_BOOK3S_PR
-
-TRACE_EVENT(kvm_book3s_reenter,
-	TP_PROTO(int r, struct kvm_vcpu *vcpu),
-	TP_ARGS(r, vcpu),
-
-	TP_STRUCT__entry(
-		__field(	unsigned int,	r		)
-		__field(	unsigned long,	pc		)
-	),
-
-	TP_fast_assign(
-		__entry->r		= r;
-		__entry->pc		= kvmppc_get_pc(vcpu);
-	),
-
-	TP_printk("reentry r=%d | pc=0x%lx", __entry->r, __entry->pc)
-);
-
-#ifdef CONFIG_PPC_BOOK3S_64
-
-TRACE_EVENT(kvm_book3s_64_mmu_map,
-	TP_PROTO(int rflags, ulong hpteg, ulong va, pfn_t hpaddr,
-		 struct kvmppc_pte *orig_pte),
-	TP_ARGS(rflags, hpteg, va, hpaddr, orig_pte),
-
-	TP_STRUCT__entry(
-		__field(	unsigned char,		flag_w		)
-		__field(	unsigned char,		flag_x		)
-		__field(	unsigned long,		eaddr		)
-		__field(	unsigned long,		hpteg		)
-		__field(	unsigned long,		va		)
-		__field(	unsigned long long,	vpage		)
-		__field(	unsigned long,		hpaddr		)
-	),
-
-	TP_fast_assign(
-		__entry->flag_w	= ((rflags & HPTE_R_PP) == 3) ? '-' : 'w';
-		__entry->flag_x	= (rflags & HPTE_R_N) ? '-' : 'x';
-		__entry->eaddr	= orig_pte->eaddr;
-		__entry->hpteg	= hpteg;
-		__entry->va	= va;
-		__entry->vpage	= orig_pte->vpage;
-		__entry->hpaddr	= hpaddr;
-	),
-
-	TP_printk("KVM: %c%c Map 0x%lx: [%lx] 0x%lx (0x%llx) -> %lx",
-		  __entry->flag_w, __entry->flag_x, __entry->eaddr,
-		  __entry->hpteg, __entry->va, __entry->vpage, __entry->hpaddr)
-);
-
-#endif /* CONFIG_PPC_BOOK3S_64 */
-
-TRACE_EVENT(kvm_book3s_mmu_map,
-	TP_PROTO(struct hpte_cache *pte),
-	TP_ARGS(pte),
-
-	TP_STRUCT__entry(
-		__field(	u64,		host_vpn	)
-		__field(	u64,		pfn		)
-		__field(	ulong,		eaddr		)
-		__field(	u64,		vpage		)
-		__field(	ulong,		raddr		)
-		__field(	int,		flags		)
-	),
-
-	TP_fast_assign(
-		__entry->host_vpn	= pte->host_vpn;
-		__entry->pfn		= pte->pfn;
-		__entry->eaddr		= pte->pte.eaddr;
-		__entry->vpage		= pte->pte.vpage;
-		__entry->raddr		= pte->pte.raddr;
-		__entry->flags		= (pte->pte.may_read ? 0x4 : 0) |
-					  (pte->pte.may_write ? 0x2 : 0) |
-					  (pte->pte.may_execute ? 0x1 : 0);
-	),
-
-	TP_printk("Map: hvpn=%llx pfn=%llx ea=%lx vp=%llx ra=%lx [%x]",
-		  __entry->host_vpn, __entry->pfn, __entry->eaddr,
-		  __entry->vpage, __entry->raddr, __entry->flags)
-);
-
-TRACE_EVENT(kvm_book3s_mmu_invalidate,
-	TP_PROTO(struct hpte_cache *pte),
-	TP_ARGS(pte),
-
-	TP_STRUCT__entry(
-		__field(	u64,		host_vpn	)
-		__field(	u64,		pfn		)
-		__field(	ulong,		eaddr		)
-		__field(	u64,		vpage		)
-		__field(	ulong,		raddr		)
-		__field(	int,		flags		)
-	),
-
-	TP_fast_assign(
-		__entry->host_vpn	= pte->host_vpn;
-		__entry->pfn		= pte->pfn;
-		__entry->eaddr		= pte->pte.eaddr;
-		__entry->vpage		= pte->pte.vpage;
-		__entry->raddr		= pte->pte.raddr;
-		__entry->flags		= (pte->pte.may_read ? 0x4 : 0) |
-					  (pte->pte.may_write ? 0x2 : 0) |
-					  (pte->pte.may_execute ? 0x1 : 0);
-	),
-
-	TP_printk("Flush: hva=%llx pfn=%llx ea=%lx vp=%llx ra=%lx [%x]",
-		  __entry->host_vpn, __entry->pfn, __entry->eaddr,
-		  __entry->vpage, __entry->raddr, __entry->flags)
-);
-
-TRACE_EVENT(kvm_book3s_mmu_flush,
-	TP_PROTO(const char *type, struct kvm_vcpu *vcpu, unsigned long long p1,
-		 unsigned long long p2),
-	TP_ARGS(type, vcpu, p1, p2),
-
-	TP_STRUCT__entry(
-		__field(	int,			count		)
-		__field(	unsigned long long,	p1		)
-		__field(	unsigned long long,	p2		)
-		__field(	const char *,		type		)
-	),
-
-	TP_fast_assign(
-		__entry->count		= to_book3s(vcpu)->hpte_cache_count;
-		__entry->p1		= p1;
-		__entry->p2		= p2;
-		__entry->type		= type;
-	),
-
-	TP_printk("Flush %d %sPTEs: %llx - %llx",
-		  __entry->count, __entry->type, __entry->p1, __entry->p2)
-);
-
-TRACE_EVENT(kvm_book3s_slb_found,
-	TP_PROTO(unsigned long long gvsid, unsigned long long hvsid),
-	TP_ARGS(gvsid, hvsid),
-
-	TP_STRUCT__entry(
-		__field(	unsigned long long,	gvsid		)
-		__field(	unsigned long long,	hvsid		)
-	),
-
-	TP_fast_assign(
-		__entry->gvsid		= gvsid;
-		__entry->hvsid		= hvsid;
-	),
-
-	TP_printk("%llx -> %llx", __entry->gvsid, __entry->hvsid)
-);
-
-TRACE_EVENT(kvm_book3s_slb_fail,
-	TP_PROTO(u16 sid_map_mask, unsigned long long gvsid),
-	TP_ARGS(sid_map_mask, gvsid),
-
-	TP_STRUCT__entry(
-		__field(	unsigned short,		sid_map_mask	)
-		__field(	unsigned long long,	gvsid		)
-	),
-
-	TP_fast_assign(
-		__entry->sid_map_mask	= sid_map_mask;
-		__entry->gvsid		= gvsid;
-	),
-
-	TP_printk("%x/%x: %llx", __entry->sid_map_mask,
-		  SID_MAP_MASK - __entry->sid_map_mask, __entry->gvsid)
-);
-
-TRACE_EVENT(kvm_book3s_slb_map,
-	TP_PROTO(u16 sid_map_mask, unsigned long long gvsid,
-		 unsigned long long hvsid),
-	TP_ARGS(sid_map_mask, gvsid, hvsid),
-
-	TP_STRUCT__entry(
-		__field(	unsigned short,		sid_map_mask	)
-		__field(	unsigned long long,	guest_vsid	)
-		__field(	unsigned long long,	host_vsid	)
-	),
-
-	TP_fast_assign(
-		__entry->sid_map_mask	= sid_map_mask;
-		__entry->guest_vsid	= gvsid;
-		__entry->host_vsid	= hvsid;
-	),
-
-	TP_printk("%x: %llx -> %llx", __entry->sid_map_mask,
-		  __entry->guest_vsid, __entry->host_vsid)
-);
-
-TRACE_EVENT(kvm_book3s_slbmte,
-	TP_PROTO(u64 slb_vsid, u64 slb_esid),
-	TP_ARGS(slb_vsid, slb_esid),
-
-	TP_STRUCT__entry(
-		__field(	u64,	slb_vsid	)
-		__field(	u64,	slb_esid	)
-	),
-
-	TP_fast_assign(
-		__entry->slb_vsid	= slb_vsid;
-		__entry->slb_esid	= slb_esid;
-	),
-
-	TP_printk("%llx, %llx", __entry->slb_vsid, __entry->slb_esid)
-);
-
-#endif /* CONFIG_PPC_BOOK3S */
-
-
-/*************************************************************************
- *                         Book3E trace points                           *
- *************************************************************************/
-
-#ifdef CONFIG_BOOKE
-
-TRACE_EVENT(kvm_booke206_stlb_write,
-	TP_PROTO(__u32 mas0, __u32 mas8, __u32 mas1, __u64 mas2, __u64 mas7_3),
-	TP_ARGS(mas0, mas8, mas1, mas2, mas7_3),
-
-	TP_STRUCT__entry(
-		__field(	__u32,	mas0		)
-		__field(	__u32,	mas8		)
-		__field(	__u32,	mas1		)
-		__field(	__u64,	mas2		)
-		__field(	__u64,	mas7_3		)
-	),
-
-	TP_fast_assign(
-		__entry->mas0		= mas0;
-		__entry->mas8		= mas8;
-		__entry->mas1		= mas1;
-		__entry->mas2		= mas2;
-		__entry->mas7_3		= mas7_3;
-	),
-
-	TP_printk("mas0=%x mas8=%x mas1=%x mas2=%llx mas7_3=%llx",
-		__entry->mas0, __entry->mas8, __entry->mas1,
-		__entry->mas2, __entry->mas7_3)
-);
-
-TRACE_EVENT(kvm_booke206_gtlb_write,
-	TP_PROTO(__u32 mas0, __u32 mas1, __u64 mas2, __u64 mas7_3),
-	TP_ARGS(mas0, mas1, mas2, mas7_3),
-
-	TP_STRUCT__entry(
-		__field(	__u32,	mas0		)
-		__field(	__u32,	mas1		)
-		__field(	__u64,	mas2		)
-		__field(	__u64,	mas7_3		)
-	),
-
-	TP_fast_assign(
-		__entry->mas0		= mas0;
-		__entry->mas1		= mas1;
-		__entry->mas2		= mas2;
-		__entry->mas7_3		= mas7_3;
-	),
-
-	TP_printk("mas0=%x mas1=%x mas2=%llx mas7_3=%llx",
-		__entry->mas0, __entry->mas1,
-		__entry->mas2, __entry->mas7_3)
-);
-
-TRACE_EVENT(kvm_booke206_ref_release,
-	TP_PROTO(__u64 pfn, __u32 flags),
-	TP_ARGS(pfn, flags),
-
-	TP_STRUCT__entry(
-		__field(	__u64,	pfn		)
-		__field(	__u32,	flags		)
-	),
-
-	TP_fast_assign(
-		__entry->pfn		= pfn;
-		__entry->flags		= flags;
-	),
-
-	TP_printk("pfn=%llx flags=%x",
-		__entry->pfn, __entry->flags)
-);
-
-TRACE_EVENT(kvm_booke_queue_irqprio,
-	TP_PROTO(struct kvm_vcpu *vcpu, unsigned int priority),
-	TP_ARGS(vcpu, priority),
-
-	TP_STRUCT__entry(
-		__field(	__u32,	cpu_nr		)
-		__field(	__u32,	priority		)
-		__field(	unsigned long,	pending		)
-	),
-
-	TP_fast_assign(
-		__entry->cpu_nr		= vcpu->vcpu_id;
-		__entry->priority	= priority;
-		__entry->pending	= vcpu->arch.pending_exceptions;
-	),
-
-	TP_printk("vcpu=%x prio=%x pending=%lx",
-		__entry->cpu_nr, __entry->priority, __entry->pending)
-);
-
-#endif
-
 #endif /* _TRACE_KVM_H */
 
 /* This part must be outside protection */
diff --git a/arch/powerpc/kvm/trace_booke.h b/arch/powerpc/kvm/trace_booke.h
new file mode 100644
index 000000000000..f7537cf26ce7
--- /dev/null
+++ b/arch/powerpc/kvm/trace_booke.h
@@ -0,0 +1,177 @@
+#if !defined(_TRACE_KVM_BOOKE_H) || defined(TRACE_HEADER_MULTI_READ)
+#define _TRACE_KVM_BOOKE_H
+
+#include <linux/tracepoint.h>
+
+#undef TRACE_SYSTEM
+#define TRACE_SYSTEM kvm_booke
+#define TRACE_INCLUDE_PATH .
+#define TRACE_INCLUDE_FILE trace_booke
+
+#define kvm_trace_symbol_exit \
+	{0, "CRITICAL"}, \
+	{1, "MACHINE_CHECK"}, \
+	{2, "DATA_STORAGE"}, \
+	{3, "INST_STORAGE"}, \
+	{4, "EXTERNAL"}, \
+	{5, "ALIGNMENT"}, \
+	{6, "PROGRAM"}, \
+	{7, "FP_UNAVAIL"}, \
+	{8, "SYSCALL"}, \
+	{9, "AP_UNAVAIL"}, \
+	{10, "DECREMENTER"}, \
+	{11, "FIT"}, \
+	{12, "WATCHDOG"}, \
+	{13, "DTLB_MISS"}, \
+	{14, "ITLB_MISS"}, \
+	{15, "DEBUG"}, \
+	{32, "SPE_UNAVAIL"}, \
+	{33, "SPE_FP_DATA"}, \
+	{34, "SPE_FP_ROUND"}, \
+	{35, "PERFORMANCE_MONITOR"}, \
+	{36, "DOORBELL"}, \
+	{37, "DOORBELL_CRITICAL"}, \
+	{38, "GUEST_DBELL"}, \
+	{39, "GUEST_DBELL_CRIT"}, \
+	{40, "HV_SYSCALL"}, \
+	{41, "HV_PRIV"}
+
+TRACE_EVENT(kvm_exit,
+	TP_PROTO(unsigned int exit_nr, struct kvm_vcpu *vcpu),
+	TP_ARGS(exit_nr, vcpu),
+
+	TP_STRUCT__entry(
+		__field(	unsigned int,	exit_nr		)
+		__field(	unsigned long,	pc		)
+		__field(	unsigned long,	msr		)
+		__field(	unsigned long,	dar		)
+		__field(	unsigned long,	last_inst	)
+	),
+
+	TP_fast_assign(
+		__entry->exit_nr	= exit_nr;
+		__entry->pc		= kvmppc_get_pc(vcpu);
+		__entry->dar		= kvmppc_get_fault_dar(vcpu);
+		__entry->msr		= vcpu->arch.shared->msr;
+		__entry->last_inst	= vcpu->arch.last_inst;
+	),
+
+	TP_printk("exit=%s"
+		" | pc=0x%lx"
+		" | msr=0x%lx"
+		" | dar=0x%lx"
+		" | last_inst=0x%lx"
+		,
+		__print_symbolic(__entry->exit_nr, kvm_trace_symbol_exit),
+		__entry->pc,
+		__entry->msr,
+		__entry->dar,
+		__entry->last_inst
+		)
+);
+
+TRACE_EVENT(kvm_unmap_hva,
+	TP_PROTO(unsigned long hva),
+	TP_ARGS(hva),
+
+	TP_STRUCT__entry(
+		__field(	unsigned long,	hva		)
+	),
+
+	TP_fast_assign(
+		__entry->hva		= hva;
+	),
+
+	TP_printk("unmap hva 0x%lx\n", __entry->hva)
+);
+
+TRACE_EVENT(kvm_booke206_stlb_write,
+	TP_PROTO(__u32 mas0, __u32 mas8, __u32 mas1, __u64 mas2, __u64 mas7_3),
+	TP_ARGS(mas0, mas8, mas1, mas2, mas7_3),
+
+	TP_STRUCT__entry(
+		__field(	__u32,	mas0		)
+		__field(	__u32,	mas8		)
+		__field(	__u32,	mas1		)
+		__field(	__u64,	mas2		)
+		__field(	__u64,	mas7_3		)
+	),
+
+	TP_fast_assign(
+		__entry->mas0		= mas0;
+		__entry->mas8		= mas8;
+		__entry->mas1		= mas1;
+		__entry->mas2		= mas2;
+		__entry->mas7_3		= mas7_3;
+	),
+
+	TP_printk("mas0=%x mas8=%x mas1=%x mas2=%llx mas7_3=%llx",
+		__entry->mas0, __entry->mas8, __entry->mas1,
+		__entry->mas2, __entry->mas7_3)
+);
+
+TRACE_EVENT(kvm_booke206_gtlb_write,
+	TP_PROTO(__u32 mas0, __u32 mas1, __u64 mas2, __u64 mas7_3),
+	TP_ARGS(mas0, mas1, mas2, mas7_3),
+
+	TP_STRUCT__entry(
+		__field(	__u32,	mas0		)
+		__field(	__u32,	mas1		)
+		__field(	__u64,	mas2		)
+		__field(	__u64,	mas7_3		)
+	),
+
+	TP_fast_assign(
+		__entry->mas0		= mas0;
+		__entry->mas1		= mas1;
+		__entry->mas2		= mas2;
+		__entry->mas7_3		= mas7_3;
+	),
+
+	TP_printk("mas0=%x mas1=%x mas2=%llx mas7_3=%llx",
+		__entry->mas0, __entry->mas1,
+		__entry->mas2, __entry->mas7_3)
+);
+
+TRACE_EVENT(kvm_booke206_ref_release,
+	TP_PROTO(__u64 pfn, __u32 flags),
+	TP_ARGS(pfn, flags),
+
+	TP_STRUCT__entry(
+		__field(	__u64,	pfn		)
+		__field(	__u32,	flags		)
+	),
+
+	TP_fast_assign(
+		__entry->pfn		= pfn;
+		__entry->flags		= flags;
+	),
+
+	TP_printk("pfn=%llx flags=%x",
+		__entry->pfn, __entry->flags)
+);
+
+TRACE_EVENT(kvm_booke_queue_irqprio,
+	TP_PROTO(struct kvm_vcpu *vcpu, unsigned int priority),
+	TP_ARGS(vcpu, priority),
+
+	TP_STRUCT__entry(
+		__field(	__u32,	cpu_nr		)
+		__field(	__u32,	priority		)
+		__field(	unsigned long,	pending		)
+	),
+
+	TP_fast_assign(
+		__entry->cpu_nr		= vcpu->vcpu_id;
+		__entry->priority	= priority;
+		__entry->pending	= vcpu->arch.pending_exceptions;
+	),
+
+	TP_printk("vcpu=%x prio=%x pending=%lx",
+		__entry->cpu_nr, __entry->priority, __entry->pending)
+);
+
+#endif
+
+/* This part must be outside protection */
+#include <trace/define_trace.h>
diff --git a/arch/powerpc/kvm/trace_pr.h b/arch/powerpc/kvm/trace_pr.h
new file mode 100644
index 000000000000..8b22e4748344
--- /dev/null
+++ b/arch/powerpc/kvm/trace_pr.h
@@ -0,0 +1,297 @@
+
+#if !defined(_TRACE_KVM_PR_H) || defined(TRACE_HEADER_MULTI_READ)
+#define _TRACE_KVM_PR_H
+
+#include <linux/tracepoint.h>
+
+#undef TRACE_SYSTEM
+#define TRACE_SYSTEM kvm_pr
+#define TRACE_INCLUDE_PATH .
+#define TRACE_INCLUDE_FILE trace_pr
+
+#define kvm_trace_symbol_exit \
+	{0x100, "SYSTEM_RESET"}, \
+	{0x200, "MACHINE_CHECK"}, \
+	{0x300, "DATA_STORAGE"}, \
+	{0x380, "DATA_SEGMENT"}, \
+	{0x400, "INST_STORAGE"}, \
+	{0x480, "INST_SEGMENT"}, \
+	{0x500, "EXTERNAL"}, \
+	{0x501, "EXTERNAL_LEVEL"}, \
+	{0x502, "EXTERNAL_HV"}, \
+	{0x600, "ALIGNMENT"}, \
+	{0x700, "PROGRAM"}, \
+	{0x800, "FP_UNAVAIL"}, \
+	{0x900, "DECREMENTER"}, \
+	{0x980, "HV_DECREMENTER"}, \
+	{0xc00, "SYSCALL"}, \
+	{0xd00, "TRACE"}, \
+	{0xe00, "H_DATA_STORAGE"}, \
+	{0xe20, "H_INST_STORAGE"}, \
+	{0xe40, "H_EMUL_ASSIST"}, \
+	{0xf00, "PERFMON"}, \
+	{0xf20, "ALTIVEC"}, \
+	{0xf40, "VSX"}
+
+TRACE_EVENT(kvm_book3s_reenter,
+	TP_PROTO(int r, struct kvm_vcpu *vcpu),
+	TP_ARGS(r, vcpu),
+
+	TP_STRUCT__entry(
+		__field(	unsigned int,	r		)
+		__field(	unsigned long,	pc		)
+	),
+
+	TP_fast_assign(
+		__entry->r		= r;
+		__entry->pc		= kvmppc_get_pc(vcpu);
+	),
+
+	TP_printk("reentry r=%d | pc=0x%lx", __entry->r, __entry->pc)
+);
+
+#ifdef CONFIG_PPC_BOOK3S_64
+
+TRACE_EVENT(kvm_book3s_64_mmu_map,
+	TP_PROTO(int rflags, ulong hpteg, ulong va, pfn_t hpaddr,
+		 struct kvmppc_pte *orig_pte),
+	TP_ARGS(rflags, hpteg, va, hpaddr, orig_pte),
+
+	TP_STRUCT__entry(
+		__field(	unsigned char,		flag_w		)
+		__field(	unsigned char,		flag_x		)
+		__field(	unsigned long,		eaddr		)
+		__field(	unsigned long,		hpteg		)
+		__field(	unsigned long,		va		)
+		__field(	unsigned long long,	vpage		)
+		__field(	unsigned long,		hpaddr		)
+	),
+
+	TP_fast_assign(
+		__entry->flag_w	= ((rflags & HPTE_R_PP) == 3) ? '-' : 'w';
+		__entry->flag_x	= (rflags & HPTE_R_N) ? '-' : 'x';
+		__entry->eaddr	= orig_pte->eaddr;
+		__entry->hpteg	= hpteg;
+		__entry->va	= va;
+		__entry->vpage	= orig_pte->vpage;
+		__entry->hpaddr	= hpaddr;
+	),
+
+	TP_printk("KVM: %c%c Map 0x%lx: [%lx] 0x%lx (0x%llx) -> %lx",
+		  __entry->flag_w, __entry->flag_x, __entry->eaddr,
+		  __entry->hpteg, __entry->va, __entry->vpage, __entry->hpaddr)
+);
+
+#endif /* CONFIG_PPC_BOOK3S_64 */
+
+TRACE_EVENT(kvm_book3s_mmu_map,
+	TP_PROTO(struct hpte_cache *pte),
+	TP_ARGS(pte),
+
+	TP_STRUCT__entry(
+		__field(	u64,		host_vpn	)
+		__field(	u64,		pfn		)
+		__field(	ulong,		eaddr		)
+		__field(	u64,		vpage		)
+		__field(	ulong,		raddr		)
+		__field(	int,		flags		)
+	),
+
+	TP_fast_assign(
+		__entry->host_vpn	= pte->host_vpn;
+		__entry->pfn		= pte->pfn;
+		__entry->eaddr		= pte->pte.eaddr;
+		__entry->vpage		= pte->pte.vpage;
+		__entry->raddr		= pte->pte.raddr;
+		__entry->flags		= (pte->pte.may_read ? 0x4 : 0) |
+					  (pte->pte.may_write ? 0x2 : 0) |
+					  (pte->pte.may_execute ? 0x1 : 0);
+	),
+
+	TP_printk("Map: hvpn=%llx pfn=%llx ea=%lx vp=%llx ra=%lx [%x]",
+		  __entry->host_vpn, __entry->pfn, __entry->eaddr,
+		  __entry->vpage, __entry->raddr, __entry->flags)
+);
+
+TRACE_EVENT(kvm_book3s_mmu_invalidate,
+	TP_PROTO(struct hpte_cache *pte),
+	TP_ARGS(pte),
+
+	TP_STRUCT__entry(
+		__field(	u64,		host_vpn	)
+		__field(	u64,		pfn		)
+		__field(	ulong,		eaddr		)
+		__field(	u64,		vpage		)
+		__field(	ulong,		raddr		)
+		__field(	int,		flags		)
+	),
+
+	TP_fast_assign(
+		__entry->host_vpn	= pte->host_vpn;
+		__entry->pfn		= pte->pfn;
+		__entry->eaddr		= pte->pte.eaddr;
+		__entry->vpage		= pte->pte.vpage;
+		__entry->raddr		= pte->pte.raddr;
+		__entry->flags		= (pte->pte.may_read ? 0x4 : 0) |
+					  (pte->pte.may_write ? 0x2 : 0) |
+					  (pte->pte.may_execute ? 0x1 : 0);
+	),
+
+	TP_printk("Flush: hva=%llx pfn=%llx ea=%lx vp=%llx ra=%lx [%x]",
+		  __entry->host_vpn, __entry->pfn, __entry->eaddr,
+		  __entry->vpage, __entry->raddr, __entry->flags)
+);
+
+TRACE_EVENT(kvm_book3s_mmu_flush,
+	TP_PROTO(const char *type, struct kvm_vcpu *vcpu, unsigned long long p1,
+		 unsigned long long p2),
+	TP_ARGS(type, vcpu, p1, p2),
+
+	TP_STRUCT__entry(
+		__field(	int,			count		)
+		__field(	unsigned long long,	p1		)
+		__field(	unsigned long long,	p2		)
+		__field(	const char *,		type		)
+	),
+
+	TP_fast_assign(
+		__entry->count		= to_book3s(vcpu)->hpte_cache_count;
+		__entry->p1		= p1;
+		__entry->p2		= p2;
+		__entry->type		= type;
+	),
+
+	TP_printk("Flush %d %sPTEs: %llx - %llx",
+		  __entry->count, __entry->type, __entry->p1, __entry->p2)
+);
+
+TRACE_EVENT(kvm_book3s_slb_found,
+	TP_PROTO(unsigned long long gvsid, unsigned long long hvsid),
+	TP_ARGS(gvsid, hvsid),
+
+	TP_STRUCT__entry(
+		__field(	unsigned long long,	gvsid		)
+		__field(	unsigned long long,	hvsid		)
+	),
+
+	TP_fast_assign(
+		__entry->gvsid		= gvsid;
+		__entry->hvsid		= hvsid;
+	),
+
+	TP_printk("%llx -> %llx", __entry->gvsid, __entry->hvsid)
+);
+
+TRACE_EVENT(kvm_book3s_slb_fail,
+	TP_PROTO(u16 sid_map_mask, unsigned long long gvsid),
+	TP_ARGS(sid_map_mask, gvsid),
+
+	TP_STRUCT__entry(
+		__field(	unsigned short,		sid_map_mask	)
+		__field(	unsigned long long,	gvsid		)
+	),
+
+	TP_fast_assign(
+		__entry->sid_map_mask	= sid_map_mask;
+		__entry->gvsid		= gvsid;
+	),
+
+	TP_printk("%x/%x: %llx", __entry->sid_map_mask,
+		  SID_MAP_MASK - __entry->sid_map_mask, __entry->gvsid)
+);
+
+TRACE_EVENT(kvm_book3s_slb_map,
+	TP_PROTO(u16 sid_map_mask, unsigned long long gvsid,
+		 unsigned long long hvsid),
+	TP_ARGS(sid_map_mask, gvsid, hvsid),
+
+	TP_STRUCT__entry(
+		__field(	unsigned short,		sid_map_mask	)
+		__field(	unsigned long long,	guest_vsid	)
+		__field(	unsigned long long,	host_vsid	)
+	),
+
+	TP_fast_assign(
+		__entry->sid_map_mask	= sid_map_mask;
+		__entry->guest_vsid	= gvsid;
+		__entry->host_vsid	= hvsid;
+	),
+
+	TP_printk("%x: %llx -> %llx", __entry->sid_map_mask,
+		  __entry->guest_vsid, __entry->host_vsid)
+);
+
+TRACE_EVENT(kvm_book3s_slbmte,
+	TP_PROTO(u64 slb_vsid, u64 slb_esid),
+	TP_ARGS(slb_vsid, slb_esid),
+
+	TP_STRUCT__entry(
+		__field(	u64,	slb_vsid	)
+		__field(	u64,	slb_esid	)
+	),
+
+	TP_fast_assign(
+		__entry->slb_vsid	= slb_vsid;
+		__entry->slb_esid	= slb_esid;
+	),
+
+	TP_printk("%llx, %llx", __entry->slb_vsid, __entry->slb_esid)
+);
+
+TRACE_EVENT(kvm_exit,
+	TP_PROTO(unsigned int exit_nr, struct kvm_vcpu *vcpu),
+	TP_ARGS(exit_nr, vcpu),
+
+	TP_STRUCT__entry(
+		__field(	unsigned int,	exit_nr		)
+		__field(	unsigned long,	pc		)
+		__field(	unsigned long,	msr		)
+		__field(	unsigned long,	dar		)
+		__field(	unsigned long,	srr1		)
+		__field(	unsigned long,	last_inst	)
+	),
+
+	TP_fast_assign(
+		__entry->exit_nr	= exit_nr;
+		__entry->pc		= kvmppc_get_pc(vcpu);
+		__entry->dar		= kvmppc_get_fault_dar(vcpu);
+		__entry->msr		= vcpu->arch.shared->msr;
+		__entry->srr1		= vcpu->arch.shadow_srr1;
+		__entry->last_inst	= vcpu->arch.last_inst;
+	),
+
+	TP_printk("exit=%s"
+		" | pc=0x%lx"
+		" | msr=0x%lx"
+		" | dar=0x%lx"
+		" | srr1=0x%lx"
+		" | last_inst=0x%lx"
+		,
+		__print_symbolic(__entry->exit_nr, kvm_trace_symbol_exit),
+		__entry->pc,
+		__entry->msr,
+		__entry->dar,
+		__entry->srr1,
+		__entry->last_inst
+		)
+);
+
+TRACE_EVENT(kvm_unmap_hva,
+	TP_PROTO(unsigned long hva),
+	TP_ARGS(hva),
+
+	TP_STRUCT__entry(
+		__field(	unsigned long,	hva		)
+	),
+
+	TP_fast_assign(
+		__entry->hva		= hva;
+	),
+
+	TP_printk("unmap hva 0x%lx\n", __entry->hva)
+);
+
+#endif /* _TRACE_KVM_H */
+
+/* This part must be outside protection */
+#include <trace/define_trace.h>
diff --git a/arch/powerpc/mm/pgtable_32.c b/arch/powerpc/mm/pgtable_32.c
index 6c856fb8c15b..5b9601715289 100644
--- a/arch/powerpc/mm/pgtable_32.c
+++ b/arch/powerpc/mm/pgtable_32.c
@@ -121,7 +121,10 @@ pgtable_t pte_alloc_one(struct mm_struct *mm, unsigned long address)
 	ptepage = alloc_pages(flags, 0);
 	if (!ptepage)
 		return NULL;
-	pgtable_page_ctor(ptepage);
+	if (!pgtable_page_ctor(ptepage)) {
+		__free_page(ptepage);
+		return NULL;
+	}
 	return ptepage;
 }
 
diff --git a/arch/powerpc/mm/pgtable_64.c b/arch/powerpc/mm/pgtable_64.c
index 536eec72c0f7..9d95786aa80f 100644
--- a/arch/powerpc/mm/pgtable_64.c
+++ b/arch/powerpc/mm/pgtable_64.c
@@ -378,6 +378,10 @@ static pte_t *__alloc_for_cache(struct mm_struct *mm, int kernel)
 				       __GFP_REPEAT | __GFP_ZERO);
 	if (!page)
 		return NULL;
+	if (!kernel && !pgtable_page_ctor(page)) {
+		__free_page(page);
+		return NULL;
+	}
 
 	ret = page_address(page);
 	spin_lock(&mm->page_table_lock);
@@ -392,9 +396,6 @@ static pte_t *__alloc_for_cache(struct mm_struct *mm, int kernel)
 	}
 	spin_unlock(&mm->page_table_lock);
 
-	if (!kernel)
-		pgtable_page_ctor(page);
-
 	return (pte_t *)ret;
 }
 
diff --git a/arch/powerpc/platforms/powermac/low_i2c.c b/arch/powerpc/platforms/powermac/low_i2c.c
index fc536f2971c0..7553b6a77c64 100644
--- a/arch/powerpc/platforms/powermac/low_i2c.c
+++ b/arch/powerpc/platforms/powermac/low_i2c.c
@@ -452,7 +452,7 @@ static int kw_i2c_xfer(struct pmac_i2c_bus *bus, u8 addrdir, int subsize,
 	 */
 	if (use_irq) {
 		/* Clear completion */
-		INIT_COMPLETION(host->complete);
+		reinit_completion(&host->complete);
 		/* Ack stale interrupts */
 		kw_write_reg(reg_isr, kw_read_reg(reg_isr));
 		/* Arm timeout */
@@ -717,7 +717,7 @@ static int pmu_i2c_xfer(struct pmac_i2c_bus *bus, u8 addrdir, int subsize,
 			return -EINVAL;
 		}
 
-		INIT_COMPLETION(comp);
+		reinit_completion(&comp);
 		req->data[0] = PMU_I2C_CMD;
 		req->reply[0] = 0xff;
 		req->nbytes = sizeof(struct pmu_i2c_hdr) + 1;
@@ -748,7 +748,7 @@ static int pmu_i2c_xfer(struct pmac_i2c_bus *bus, u8 addrdir, int subsize,
 
 		hdr->bus = PMU_I2C_BUS_STATUS;
 
-		INIT_COMPLETION(comp);
+		reinit_completion(&comp);
 		req->data[0] = PMU_I2C_CMD;
 		req->reply[0] = 0xff;
 		req->nbytes = 2;
diff --git a/arch/powerpc/platforms/pseries/nvram.c b/arch/powerpc/platforms/pseries/nvram.c
index 057fc894be51..7bfaf58d4664 100644
--- a/arch/powerpc/platforms/pseries/nvram.c
+++ b/arch/powerpc/platforms/pseries/nvram.c
@@ -31,7 +31,7 @@
 #define NVRW_CNT 0x20
 
 /*
- * Set oops header version to distingush between old and new format header.
+ * Set oops header version to distinguish between old and new format header.
  * lnx,oops-log partition max size is 4000, header version > 4000 will
  * help in identifying new header.
  */
diff --git a/arch/powerpc/platforms/pseries/suspend.c b/arch/powerpc/platforms/pseries/suspend.c
index 5f997e79d570..16a255255d30 100644
--- a/arch/powerpc/platforms/pseries/suspend.c
+++ b/arch/powerpc/platforms/pseries/suspend.c
@@ -106,7 +106,7 @@ static int pseries_prepare_late(void)
 	atomic_set(&suspend_data.done, 0);
 	atomic_set(&suspend_data.error, 0);
 	suspend_data.complete = &suspend_work;
-	INIT_COMPLETION(suspend_work);
+	reinit_completion(&suspend_work);
 	return 0;
 }
 
diff --git a/arch/s390/Kconfig b/arch/s390/Kconfig
index f75d7e517927..314fced4fc14 100644
--- a/arch/s390/Kconfig
+++ b/arch/s390/Kconfig
@@ -141,7 +141,6 @@ config S390
 	select OLD_SIGACTION
 	select OLD_SIGSUSPEND3
 	select SYSCTL_EXCEPTION_TRACE
-	select USE_GENERIC_SMP_HELPERS if SMP
 	select VIRT_CPU_ACCOUNTING
 	select VIRT_TO_BUS
 
diff --git a/arch/s390/boot/Makefile b/arch/s390/boot/Makefile
index f2737a005afc..9a42ecec5647 100644
--- a/arch/s390/boot/Makefile
+++ b/arch/s390/boot/Makefile
@@ -21,6 +21,6 @@ $(obj)/bzImage: $(obj)/compressed/vmlinux FORCE
 $(obj)/compressed/vmlinux: FORCE
 	$(Q)$(MAKE) $(build)=$(obj)/compressed $@
 
-install: $(CONFIGURE) $(obj)/image
-	sh -x  $(srctree)/$(obj)/install.sh $(KERNELRELEASE) $(obj)/image \
+install: $(CONFIGURE) $(obj)/bzImage
+	sh -x  $(srctree)/$(obj)/install.sh $(KERNELRELEASE) $(obj)/bzImage \
 	      System.map "$(INSTALL_PATH)"
diff --git a/arch/s390/include/asm/ctl_reg.h b/arch/s390/include/asm/ctl_reg.h
index 9b69c0befdca..4e63f1a13600 100644
--- a/arch/s390/include/asm/ctl_reg.h
+++ b/arch/s390/include/asm/ctl_reg.h
@@ -7,6 +7,8 @@
 #ifndef __ASM_CTL_REG_H
 #define __ASM_CTL_REG_H
 
+#include <linux/bug.h>
+
 #ifdef CONFIG_64BIT
 # define __CTL_LOAD	"lctlg"
 # define __CTL_STORE	"stctg"
diff --git a/arch/s390/include/asm/eadm.h b/arch/s390/include/asm/eadm.h
index dc9200ca32ed..67026300c88e 100644
--- a/arch/s390/include/asm/eadm.h
+++ b/arch/s390/include/asm/eadm.h
@@ -111,18 +111,7 @@ struct scm_driver {
 int scm_driver_register(struct scm_driver *scmdrv);
 void scm_driver_unregister(struct scm_driver *scmdrv);
 
-int scm_start_aob(struct aob *aob);
+int eadm_start_aob(struct aob *aob);
 void scm_irq_handler(struct aob *aob, int error);
 
-struct eadm_ops {
-	int (*eadm_start) (struct aob *aob);
-	struct module *owner;
-};
-
-int scm_get_ref(void);
-void scm_put_ref(void);
-
-void register_eadm_ops(struct eadm_ops *ops);
-void unregister_eadm_ops(struct eadm_ops *ops);
-
 #endif /* _ASM_S390_EADM_H */
diff --git a/arch/s390/include/asm/hardirq.h b/arch/s390/include/asm/hardirq.h
index a908d2941c5d..b7eabaaeffbd 100644
--- a/arch/s390/include/asm/hardirq.h
+++ b/arch/s390/include/asm/hardirq.h
@@ -18,8 +18,6 @@
 #define __ARCH_HAS_DO_SOFTIRQ
 #define __ARCH_IRQ_EXIT_IRQS_DISABLED
 
-#define HARDIRQ_BITS	8
-
 static inline void ack_bad_irq(unsigned int irq)
 {
 	printk(KERN_CRIT "unexpected IRQ trap at vector %02x\n", irq);
diff --git a/arch/s390/include/asm/kvm_host.h b/arch/s390/include/asm/kvm_host.h
index e87ecaa2c569..d5bc3750616e 100644
--- a/arch/s390/include/asm/kvm_host.h
+++ b/arch/s390/include/asm/kvm_host.h
@@ -38,13 +38,6 @@ struct sca_block {
 	struct sca_entry cpu[64];
 } __attribute__((packed));
 
-#define KVM_NR_PAGE_SIZES 2
-#define KVM_HPAGE_GFN_SHIFT(x) (((x) - 1) * 8)
-#define KVM_HPAGE_SHIFT(x) (PAGE_SHIFT + KVM_HPAGE_GFN_SHIFT(x))
-#define KVM_HPAGE_SIZE(x) (1UL << KVM_HPAGE_SHIFT(x))
-#define KVM_HPAGE_MASK(x)	(~(KVM_HPAGE_SIZE(x) - 1))
-#define KVM_PAGES_PER_HPAGE(x)	(KVM_HPAGE_SIZE(x) / PAGE_SIZE)
-
 #define CPUSTAT_STOPPED    0x80000000
 #define CPUSTAT_WAIT       0x10000000
 #define CPUSTAT_ECALL_PEND 0x08000000
@@ -220,7 +213,6 @@ struct kvm_s390_interrupt_info {
 /* for local_interrupt.action_flags */
 #define ACTION_STORE_ON_STOP		(1<<0)
 #define ACTION_STOP_ON_STOP		(1<<1)
-#define ACTION_RELOADVCPU_ON_STOP	(1<<2)
 
 struct kvm_s390_local_interrupt {
 	spinlock_t lock;
diff --git a/arch/s390/include/asm/pci.h b/arch/s390/include/asm/pci.h
index 1cc185da9d38..c129ab2ac731 100644
--- a/arch/s390/include/asm/pci.h
+++ b/arch/s390/include/asm/pci.h
@@ -63,9 +63,10 @@ enum zpci_state {
 };
 
 struct zpci_bar_struct {
+	struct resource *res;		/* bus resource */
 	u32		val;		/* bar start & 3 flag bits */
-	u8		size;		/* order 2 exponent */
 	u16		map_idx;	/* index into bar mapping array */
+	u8		size;		/* order 2 exponent */
 };
 
 /* Private data per function */
@@ -97,6 +98,7 @@ struct zpci_dev {
 	unsigned long	iommu_pages;
 	unsigned int	next_bit;
 
+	char res_name[16];
 	struct zpci_bar_struct bars[PCI_BAR_COUNT];
 
 	u64		start_dma;	/* Start of available DMA addresses */
@@ -122,12 +124,10 @@ static inline bool zdev_enabled(struct zpci_dev *zdev)
   Prototypes
 ----------------------------------------------------------------------------- */
 /* Base stuff */
-struct zpci_dev *zpci_alloc_device(void);
 int zpci_create_device(struct zpci_dev *);
 int zpci_enable_device(struct zpci_dev *);
 int zpci_disable_device(struct zpci_dev *);
 void zpci_stop_device(struct zpci_dev *);
-void zpci_free_device(struct zpci_dev *);
 int zpci_register_ioat(struct zpci_dev *, u8, u64, u64, u64);
 int zpci_unregister_ioat(struct zpci_dev *, u8);
 
diff --git a/arch/s390/include/asm/sclp.h b/arch/s390/include/asm/sclp.h
index 7dc7f9c63b65..30ef748bc161 100644
--- a/arch/s390/include/asm/sclp.h
+++ b/arch/s390/include/asm/sclp.h
@@ -43,7 +43,6 @@ struct sclp_cpu_info {
 int sclp_get_cpu_info(struct sclp_cpu_info *info);
 int sclp_cpu_configure(u8 cpu);
 int sclp_cpu_deconfigure(u8 cpu);
-void sclp_facilities_detect(void);
 unsigned long long sclp_get_rnmax(void);
 unsigned long long sclp_get_rzm(void);
 int sclp_sdias_blk_count(void);
@@ -57,5 +56,7 @@ bool sclp_has_vt220(void);
 int sclp_pci_configure(u32 fid);
 int sclp_pci_deconfigure(u32 fid);
 int memcpy_hsa(void *dest, unsigned long src, size_t count, int mode);
+unsigned long sclp_get_hsa_size(void);
+void sclp_early_detect(void);
 
 #endif /* _ASM_S390_SCLP_H */
diff --git a/arch/s390/include/asm/setup.h b/arch/s390/include/asm/setup.h
index df802ee14af6..94cfbe442f12 100644
--- a/arch/s390/include/asm/setup.h
+++ b/arch/s390/include/asm/setup.h
@@ -107,9 +107,6 @@ void create_mem_hole(struct mem_chunk mem_chunk[], unsigned long addr,
 #define MACHINE_HAS_RRBM	(S390_lowcore.machine_flags & MACHINE_FLAG_RRBM)
 #endif /* CONFIG_64BIT */
 
-#define ZFCPDUMP_HSA_SIZE	(32UL<<20)
-#define ZFCPDUMP_HSA_SIZE_MAX	(64UL<<20)
-
 /*
  * Console mode. Override with conmode=
  */
diff --git a/arch/s390/include/asm/thread_info.h b/arch/s390/include/asm/thread_info.h
index eb5f64d26d06..10e0fcd3633d 100644
--- a/arch/s390/include/asm/thread_info.h
+++ b/arch/s390/include/asm/thread_info.h
@@ -111,6 +111,4 @@ static inline struct thread_info *current_thread_info(void)
 #define is_32bit_task()		(1)
 #endif
 
-#define PREEMPT_ACTIVE		0x4000000
-
 #endif /* _ASM_THREAD_INFO_H */
diff --git a/arch/s390/kernel/crash_dump.c b/arch/s390/kernel/crash_dump.c
index f45b2ab0cb81..d7658c4b2ed5 100644
--- a/arch/s390/kernel/crash_dump.c
+++ b/arch/s390/kernel/crash_dump.c
@@ -95,7 +95,7 @@ static void *elfcorehdr_newmem;
 /*
  * Copy one page from zfcpdump "oldmem"
  *
- * For pages below ZFCPDUMP_HSA_SIZE memory from the HSA is copied. Otherwise
+ * For pages below HSA size memory from the HSA is copied. Otherwise
  * real memory copy is used.
  */
 static ssize_t copy_oldmem_page_zfcpdump(char *buf, size_t csize,
@@ -103,7 +103,7 @@ static ssize_t copy_oldmem_page_zfcpdump(char *buf, size_t csize,
 {
 	int rc;
 
-	if (src < ZFCPDUMP_HSA_SIZE) {
+	if (src < sclp_get_hsa_size()) {
 		rc = memcpy_hsa(buf, src, csize, userbuf);
 	} else {
 		if (userbuf)
@@ -188,18 +188,19 @@ static int remap_oldmem_pfn_range_kdump(struct vm_area_struct *vma,
 /*
  * Remap "oldmem" for zfcpdump
  *
- * We only map available memory above ZFCPDUMP_HSA_SIZE. Memory below
- * ZFCPDUMP_HSA_SIZE is read on demand using the copy_oldmem_page() function.
+ * We only map available memory above HSA size. Memory below HSA size
+ * is read on demand using the copy_oldmem_page() function.
  */
 static int remap_oldmem_pfn_range_zfcpdump(struct vm_area_struct *vma,
 					   unsigned long from,
 					   unsigned long pfn,
 					   unsigned long size, pgprot_t prot)
 {
+	unsigned long hsa_end = sclp_get_hsa_size();
 	unsigned long size_hsa;
 
-	if (pfn < ZFCPDUMP_HSA_SIZE >> PAGE_SHIFT) {
-		size_hsa = min(size, ZFCPDUMP_HSA_SIZE - (pfn << PAGE_SHIFT));
+	if (pfn < hsa_end >> PAGE_SHIFT) {
+		size_hsa = min(size, hsa_end - (pfn << PAGE_SHIFT));
 		if (size == size_hsa)
 			return 0;
 		size -= size_hsa;
@@ -238,9 +239,9 @@ int copy_from_oldmem(void *dest, void *src, size_t count)
 				return rc;
 		}
 	} else {
-		if ((unsigned long) src < ZFCPDUMP_HSA_SIZE) {
-			copied = min(count,
-				     ZFCPDUMP_HSA_SIZE - (unsigned long) src);
+		unsigned long hsa_end = sclp_get_hsa_size();
+		if ((unsigned long) src < hsa_end) {
+			copied = min(count, hsa_end - (unsigned long) src);
 			rc = memcpy_hsa(dest, (unsigned long) src, copied, 0);
 			if (rc)
 				return rc;
@@ -580,6 +581,9 @@ int elfcorehdr_alloc(unsigned long long *addr, unsigned long long *size)
 	/* If elfcorehdr= has been passed via cmdline, we use that one */
 	if (elfcorehdr_addr != ELFCORE_ADDR_MAX)
 		return 0;
+	/* If we cannot get HSA size for zfcpdump return error */
+	if (ipl_info.type == IPL_TYPE_FCP_DUMP && !sclp_get_hsa_size())
+		return -ENODEV;
 	mem_chunk_cnt = get_mem_chunk_cnt();
 
 	alloc_size = 0x1000 + get_cpu_cnt() * 0x300 +
diff --git a/arch/s390/kernel/early.c b/arch/s390/kernel/early.c
index 96543ac400a7..fca20b5fe79e 100644
--- a/arch/s390/kernel/early.c
+++ b/arch/s390/kernel/early.c
@@ -483,7 +483,7 @@ void __init startup_init(void)
 	detect_diag44();
 	detect_machine_facilities();
 	setup_topology();
-	sclp_facilities_detect();
+	sclp_early_detect();
 #ifdef CONFIG_DYNAMIC_FTRACE
 	S390_lowcore.ftrace_func = (unsigned long)ftrace_caller;
 #endif
diff --git a/arch/s390/kernel/kprobes.c b/arch/s390/kernel/kprobes.c
index 59a9c35c4598..bc71a7b95af5 100644
--- a/arch/s390/kernel/kprobes.c
+++ b/arch/s390/kernel/kprobes.c
@@ -680,7 +680,7 @@ static int __kprobes kprobe_trap_handler(struct pt_regs *regs, int trapnr)
 	case KPROBE_HIT_SSDONE:
 		/*
 		 * We increment the nmissed count for accounting,
-		 * we can also use npre/npostfault count for accouting
+		 * we can also use npre/npostfault count for accounting
 		 * these specific fault cases.
 		 */
 		kprobes_inc_nmissed_count(p);
diff --git a/arch/s390/kernel/setup.c b/arch/s390/kernel/setup.c
index ffe1c53264a7..4444875266ee 100644
--- a/arch/s390/kernel/setup.c
+++ b/arch/s390/kernel/setup.c
@@ -471,8 +471,9 @@ static void __init setup_memory_end(void)
 
 
 #ifdef CONFIG_ZFCPDUMP
-	if (ipl_info.type == IPL_TYPE_FCP_DUMP && !OLDMEM_BASE) {
-		memory_end = ZFCPDUMP_HSA_SIZE;
+	if (ipl_info.type == IPL_TYPE_FCP_DUMP &&
+	    !OLDMEM_BASE && sclp_get_hsa_size()) {
+		memory_end = sclp_get_hsa_size();
 		memory_end_set = 1;
 	}
 #endif
@@ -586,7 +587,7 @@ static unsigned long __init find_crash_base(unsigned long crash_size,
 		crash_base = (chunk->addr + chunk->size) - crash_size;
 		if (crash_base < crash_size)
 			continue;
-		if (crash_base < ZFCPDUMP_HSA_SIZE_MAX)
+		if (crash_base < sclp_get_hsa_size())
 			continue;
 		if (crash_base < (unsigned long) INITRD_START + INITRD_SIZE)
 			continue;
diff --git a/arch/s390/kvm/diag.c b/arch/s390/kvm/diag.c
index 3a74d8af0d69..78d967f180f4 100644
--- a/arch/s390/kvm/diag.c
+++ b/arch/s390/kvm/diag.c
@@ -107,14 +107,13 @@ static int __diag_ipl_functions(struct kvm_vcpu *vcpu)
 
 static int __diag_virtio_hypercall(struct kvm_vcpu *vcpu)
 {
-	int ret, idx;
+	int ret;
 
 	/* No virtio-ccw notification? Get out quickly. */
 	if (!vcpu->kvm->arch.css_support ||
 	    (vcpu->run->s.regs.gprs[1] != KVM_S390_VIRTIO_CCW_NOTIFY))
 		return -EOPNOTSUPP;
 
-	idx = srcu_read_lock(&vcpu->kvm->srcu);
 	/*
 	 * The layout is as follows:
 	 * - gpr 2 contains the subchannel id (passed as addr)
@@ -125,7 +124,6 @@ static int __diag_virtio_hypercall(struct kvm_vcpu *vcpu)
 				      vcpu->run->s.regs.gprs[2],
 				      8, &vcpu->run->s.regs.gprs[3],
 				      vcpu->run->s.regs.gprs[4]);
-	srcu_read_unlock(&vcpu->kvm->srcu, idx);
 
 	/*
 	 * Return cookie in gpr 2, but don't overwrite the register if the
diff --git a/arch/s390/kvm/gaccess.h b/arch/s390/kvm/gaccess.h
index 99d789e8a018..374a439ccc60 100644
--- a/arch/s390/kvm/gaccess.h
+++ b/arch/s390/kvm/gaccess.h
@@ -18,20 +18,27 @@
 #include <asm/uaccess.h>
 #include "kvm-s390.h"
 
+/* Convert real to absolute address by applying the prefix of the CPU */
+static inline unsigned long kvm_s390_real_to_abs(struct kvm_vcpu *vcpu,
+						 unsigned long gaddr)
+{
+	unsigned long prefix  = vcpu->arch.sie_block->prefix;
+	if (gaddr < 2 * PAGE_SIZE)
+		gaddr += prefix;
+	else if (gaddr >= prefix && gaddr < prefix + 2 * PAGE_SIZE)
+		gaddr -= prefix;
+	return gaddr;
+}
+
 static inline void __user *__gptr_to_uptr(struct kvm_vcpu *vcpu,
 					  void __user *gptr,
 					  int prefixing)
 {
-	unsigned long prefix  = vcpu->arch.sie_block->prefix;
 	unsigned long gaddr = (unsigned long) gptr;
 	unsigned long uaddr;
 
-	if (prefixing) {
-		if (gaddr < 2 * PAGE_SIZE)
-			gaddr += prefix;
-		else if ((gaddr >= prefix) && (gaddr < prefix + 2 * PAGE_SIZE))
-			gaddr -= prefix;
-	}
+	if (prefixing)
+		gaddr = kvm_s390_real_to_abs(vcpu, gaddr);
 	uaddr = gmap_fault(gaddr, vcpu->arch.gmap);
 	if (IS_ERR_VALUE(uaddr))
 		uaddr = -EFAULT;
diff --git a/arch/s390/kvm/intercept.c b/arch/s390/kvm/intercept.c
index 5ee56e5acc23..5ddbbde6f65c 100644
--- a/arch/s390/kvm/intercept.c
+++ b/arch/s390/kvm/intercept.c
@@ -62,12 +62,6 @@ static int handle_stop(struct kvm_vcpu *vcpu)
 
 	trace_kvm_s390_stop_request(vcpu->arch.local_int.action_bits);
 
-	if (vcpu->arch.local_int.action_bits & ACTION_RELOADVCPU_ON_STOP) {
-		vcpu->arch.local_int.action_bits &= ~ACTION_RELOADVCPU_ON_STOP;
-		rc = SIE_INTERCEPT_RERUNVCPU;
-		vcpu->run->exit_reason = KVM_EXIT_INTR;
-	}
-
 	if (vcpu->arch.local_int.action_bits & ACTION_STOP_ON_STOP) {
 		atomic_set_mask(CPUSTAT_STOPPED,
 				&vcpu->arch.sie_block->cpuflags);
diff --git a/arch/s390/kvm/interrupt.c b/arch/s390/kvm/interrupt.c
index 7f1f7ac5cf7f..5f79d2d79ca7 100644
--- a/arch/s390/kvm/interrupt.c
+++ b/arch/s390/kvm/interrupt.c
@@ -436,6 +436,7 @@ int kvm_s390_handle_wait(struct kvm_vcpu *vcpu)
 	hrtimer_start(&vcpu->arch.ckc_timer, ktime_set (0, sltime) , HRTIMER_MODE_REL);
 	VCPU_EVENT(vcpu, 5, "enabled wait via clock comparator: %llx ns", sltime);
 no_timer:
+	srcu_read_unlock(&vcpu->kvm->srcu, vcpu->srcu_idx);
 	spin_lock(&vcpu->arch.local_int.float_int->lock);
 	spin_lock_bh(&vcpu->arch.local_int.lock);
 	add_wait_queue(&vcpu->wq, &wait);
@@ -455,6 +456,8 @@ no_timer:
 	remove_wait_queue(&vcpu->wq, &wait);
 	spin_unlock_bh(&vcpu->arch.local_int.lock);
 	spin_unlock(&vcpu->arch.local_int.float_int->lock);
+	vcpu->srcu_idx = srcu_read_lock(&vcpu->kvm->srcu);
+
 	hrtimer_try_to_cancel(&vcpu->arch.ckc_timer);
 	return 0;
 }
diff --git a/arch/s390/kvm/kvm-s390.c b/arch/s390/kvm/kvm-s390.c
index ed8064cb5c49..569494e01ec6 100644
--- a/arch/s390/kvm/kvm-s390.c
+++ b/arch/s390/kvm/kvm-s390.c
@@ -695,9 +695,9 @@ static int kvm_s390_handle_requests(struct kvm_vcpu *vcpu)
 	return 0;
 }
 
-static int __vcpu_run(struct kvm_vcpu *vcpu)
+static int vcpu_pre_run(struct kvm_vcpu *vcpu)
 {
-	int rc;
+	int rc, cpuflags;
 
 	memcpy(&vcpu->arch.sie_block->gg14, &vcpu->run->s.regs.gprs[14], 16);
 
@@ -715,28 +715,24 @@ static int __vcpu_run(struct kvm_vcpu *vcpu)
 		return rc;
 
 	vcpu->arch.sie_block->icptcode = 0;
-	VCPU_EVENT(vcpu, 6, "entering sie flags %x",
-		   atomic_read(&vcpu->arch.sie_block->cpuflags));
-	trace_kvm_s390_sie_enter(vcpu,
-				 atomic_read(&vcpu->arch.sie_block->cpuflags));
+	cpuflags = atomic_read(&vcpu->arch.sie_block->cpuflags);
+	VCPU_EVENT(vcpu, 6, "entering sie flags %x", cpuflags);
+	trace_kvm_s390_sie_enter(vcpu, cpuflags);
 
-	/*
-	 * As PF_VCPU will be used in fault handler, between guest_enter
-	 * and guest_exit should be no uaccess.
-	 */
-	preempt_disable();
-	kvm_guest_enter();
-	preempt_enable();
-	rc = sie64a(vcpu->arch.sie_block, vcpu->run->s.regs.gprs);
-	kvm_guest_exit();
+	return 0;
+}
+
+static int vcpu_post_run(struct kvm_vcpu *vcpu, int exit_reason)
+{
+	int rc;
 
 	VCPU_EVENT(vcpu, 6, "exit sie icptcode %d",
 		   vcpu->arch.sie_block->icptcode);
 	trace_kvm_s390_sie_exit(vcpu, vcpu->arch.sie_block->icptcode);
 
-	if (rc > 0)
+	if (exit_reason >= 0) {
 		rc = 0;
-	if (rc < 0) {
+	} else {
 		if (kvm_is_ucontrol(vcpu->kvm)) {
 			rc = SIE_INTERCEPT_UCONTROL;
 		} else {
@@ -747,6 +743,49 @@ static int __vcpu_run(struct kvm_vcpu *vcpu)
 	}
 
 	memcpy(&vcpu->run->s.regs.gprs[14], &vcpu->arch.sie_block->gg14, 16);
+
+	if (rc == 0) {
+		if (kvm_is_ucontrol(vcpu->kvm))
+			rc = -EOPNOTSUPP;
+		else
+			rc = kvm_handle_sie_intercept(vcpu);
+	}
+
+	return rc;
+}
+
+static int __vcpu_run(struct kvm_vcpu *vcpu)
+{
+	int rc, exit_reason;
+
+	/*
+	 * We try to hold kvm->srcu during most of vcpu_run (except when run-
+	 * ning the guest), so that memslots (and other stuff) are protected
+	 */
+	vcpu->srcu_idx = srcu_read_lock(&vcpu->kvm->srcu);
+
+	do {
+		rc = vcpu_pre_run(vcpu);
+		if (rc)
+			break;
+
+		srcu_read_unlock(&vcpu->kvm->srcu, vcpu->srcu_idx);
+		/*
+		 * As PF_VCPU will be used in fault handler, between
+		 * guest_enter and guest_exit should be no uaccess.
+		 */
+		preempt_disable();
+		kvm_guest_enter();
+		preempt_enable();
+		exit_reason = sie64a(vcpu->arch.sie_block,
+				     vcpu->run->s.regs.gprs);
+		kvm_guest_exit();
+		vcpu->srcu_idx = srcu_read_lock(&vcpu->kvm->srcu);
+
+		rc = vcpu_post_run(vcpu, exit_reason);
+	} while (!signal_pending(current) && !rc);
+
+	srcu_read_unlock(&vcpu->kvm->srcu, vcpu->srcu_idx);
 	return rc;
 }
 
@@ -755,7 +794,6 @@ int kvm_arch_vcpu_ioctl_run(struct kvm_vcpu *vcpu, struct kvm_run *kvm_run)
 	int rc;
 	sigset_t sigsaved;
 
-rerun_vcpu:
 	if (vcpu->sigset_active)
 		sigprocmask(SIG_SETMASK, &vcpu->sigset, &sigsaved);
 
@@ -788,19 +826,7 @@ rerun_vcpu:
 	}
 
 	might_fault();
-
-	do {
-		rc = __vcpu_run(vcpu);
-		if (rc)
-			break;
-		if (kvm_is_ucontrol(vcpu->kvm))
-			rc = -EOPNOTSUPP;
-		else
-			rc = kvm_handle_sie_intercept(vcpu);
-	} while (!signal_pending(current) && !rc);
-
-	if (rc == SIE_INTERCEPT_RERUNVCPU)
-		goto rerun_vcpu;
+	rc = __vcpu_run(vcpu);
 
 	if (signal_pending(current) && !rc) {
 		kvm_run->exit_reason = KVM_EXIT_INTR;
@@ -958,6 +984,7 @@ long kvm_arch_vcpu_ioctl(struct file *filp,
 {
 	struct kvm_vcpu *vcpu = filp->private_data;
 	void __user *argp = (void __user *)arg;
+	int idx;
 	long r;
 
 	switch (ioctl) {
@@ -971,7 +998,9 @@ long kvm_arch_vcpu_ioctl(struct file *filp,
 		break;
 	}
 	case KVM_S390_STORE_STATUS:
+		idx = srcu_read_lock(&vcpu->kvm->srcu);
 		r = kvm_s390_vcpu_store_status(vcpu, arg);
+		srcu_read_unlock(&vcpu->kvm->srcu, idx);
 		break;
 	case KVM_S390_SET_INITIAL_PSW: {
 		psw_t psw;
@@ -1067,12 +1096,13 @@ int kvm_arch_vcpu_fault(struct kvm_vcpu *vcpu, struct vm_fault *vmf)
 	return VM_FAULT_SIGBUS;
 }
 
-void kvm_arch_free_memslot(struct kvm_memory_slot *free,
+void kvm_arch_free_memslot(struct kvm *kvm, struct kvm_memory_slot *free,
 			   struct kvm_memory_slot *dont)
 {
 }
 
-int kvm_arch_create_memslot(struct kvm_memory_slot *slot, unsigned long npages)
+int kvm_arch_create_memslot(struct kvm *kvm, struct kvm_memory_slot *slot,
+			    unsigned long npages)
 {
 	return 0;
 }
diff --git a/arch/s390/kvm/kvm-s390.h b/arch/s390/kvm/kvm-s390.h
index dc99f1ca4267..b44912a32949 100644
--- a/arch/s390/kvm/kvm-s390.h
+++ b/arch/s390/kvm/kvm-s390.h
@@ -28,8 +28,7 @@ typedef int (*intercept_handler_t)(struct kvm_vcpu *vcpu);
 extern unsigned long *vfacilities;
 
 /* negativ values are error codes, positive values for internal conditions */
-#define SIE_INTERCEPT_RERUNVCPU		(1<<0)
-#define SIE_INTERCEPT_UCONTROL		(1<<1)
+#define SIE_INTERCEPT_UCONTROL		(1<<0)
 int kvm_handle_sie_intercept(struct kvm_vcpu *vcpu);
 
 #define VM_EVENT(d_kvm, d_loglevel, d_string, d_args...)\
@@ -91,8 +90,10 @@ static inline void kvm_s390_get_base_disp_sse(struct kvm_vcpu *vcpu,
 
 static inline void kvm_s390_get_regs_rre(struct kvm_vcpu *vcpu, int *r1, int *r2)
 {
-	*r1 = (vcpu->arch.sie_block->ipb & 0x00f00000) >> 20;
-	*r2 = (vcpu->arch.sie_block->ipb & 0x000f0000) >> 16;
+	if (r1)
+		*r1 = (vcpu->arch.sie_block->ipb & 0x00f00000) >> 20;
+	if (r2)
+		*r2 = (vcpu->arch.sie_block->ipb & 0x000f0000) >> 16;
 }
 
 static inline u64 kvm_s390_get_base_disp_rsy(struct kvm_vcpu *vcpu)
diff --git a/arch/s390/kvm/priv.c b/arch/s390/kvm/priv.c
index 59200ee275e5..2440602e6df1 100644
--- a/arch/s390/kvm/priv.c
+++ b/arch/s390/kvm/priv.c
@@ -30,6 +30,38 @@
 #include "kvm-s390.h"
 #include "trace.h"
 
+/* Handle SCK (SET CLOCK) interception */
+static int handle_set_clock(struct kvm_vcpu *vcpu)
+{
+	struct kvm_vcpu *cpup;
+	s64 hostclk, val;
+	u64 op2;
+	int i;
+
+	if (vcpu->arch.sie_block->gpsw.mask & PSW_MASK_PSTATE)
+		return kvm_s390_inject_program_int(vcpu, PGM_PRIVILEGED_OP);
+
+	op2 = kvm_s390_get_base_disp_s(vcpu);
+	if (op2 & 7)	/* Operand must be on a doubleword boundary */
+		return kvm_s390_inject_program_int(vcpu, PGM_SPECIFICATION);
+	if (get_guest(vcpu, val, (u64 __user *) op2))
+		return kvm_s390_inject_program_int(vcpu, PGM_ADDRESSING);
+
+	if (store_tod_clock(&hostclk)) {
+		kvm_s390_set_psw_cc(vcpu, 3);
+		return 0;
+	}
+	val = (val - hostclk) & ~0x3fUL;
+
+	mutex_lock(&vcpu->kvm->lock);
+	kvm_for_each_vcpu(i, cpup, vcpu->kvm)
+		cpup->arch.sie_block->epoch = val;
+	mutex_unlock(&vcpu->kvm->lock);
+
+	kvm_s390_set_psw_cc(vcpu, 0);
+	return 0;
+}
+
 static int handle_set_prefix(struct kvm_vcpu *vcpu)
 {
 	u64 operand2;
@@ -128,6 +160,33 @@ static int handle_skey(struct kvm_vcpu *vcpu)
 	return 0;
 }
 
+static int handle_test_block(struct kvm_vcpu *vcpu)
+{
+	unsigned long hva;
+	gpa_t addr;
+	int reg2;
+
+	if (vcpu->arch.sie_block->gpsw.mask & PSW_MASK_PSTATE)
+		return kvm_s390_inject_program_int(vcpu, PGM_PRIVILEGED_OP);
+
+	kvm_s390_get_regs_rre(vcpu, NULL, &reg2);
+	addr = vcpu->run->s.regs.gprs[reg2] & PAGE_MASK;
+	addr = kvm_s390_real_to_abs(vcpu, addr);
+
+	hva = gfn_to_hva(vcpu->kvm, gpa_to_gfn(addr));
+	if (kvm_is_error_hva(hva))
+		return kvm_s390_inject_program_int(vcpu, PGM_ADDRESSING);
+	/*
+	 * We don't expect errors on modern systems, and do not care
+	 * about storage keys (yet), so let's just clear the page.
+	 */
+	if (clear_user((void __user *)hva, PAGE_SIZE) != 0)
+		return -EFAULT;
+	kvm_s390_set_psw_cc(vcpu, 0);
+	vcpu->run->s.regs.gprs[0] = 0;
+	return 0;
+}
+
 static int handle_tpi(struct kvm_vcpu *vcpu)
 {
 	struct kvm_s390_interrupt_info *inti;
@@ -438,12 +497,14 @@ out_exception:
 
 static const intercept_handler_t b2_handlers[256] = {
 	[0x02] = handle_stidp,
+	[0x04] = handle_set_clock,
 	[0x10] = handle_set_prefix,
 	[0x11] = handle_store_prefix,
 	[0x12] = handle_store_cpu_address,
 	[0x29] = handle_skey,
 	[0x2a] = handle_skey,
 	[0x2b] = handle_skey,
+	[0x2c] = handle_test_block,
 	[0x30] = handle_io_inst,
 	[0x31] = handle_io_inst,
 	[0x32] = handle_io_inst,
diff --git a/arch/s390/mm/pgtable.c b/arch/s390/mm/pgtable.c
index 0a2e5e086749..e794c88f699a 100644
--- a/arch/s390/mm/pgtable.c
+++ b/arch/s390/mm/pgtable.c
@@ -772,7 +772,11 @@ static inline unsigned long *page_table_alloc_pgste(struct mm_struct *mm,
 		__free_page(page);
 		return NULL;
 	}
-	pgtable_page_ctor(page);
+	if (!pgtable_page_ctor(page)) {
+		kfree(mp);
+		__free_page(page);
+		return NULL;
+	}
 	mp->vmaddr = vmaddr & PMD_MASK;
 	INIT_LIST_HEAD(&mp->mapper);
 	page->index = (unsigned long) mp;
@@ -902,7 +906,10 @@ unsigned long *page_table_alloc(struct mm_struct *mm, unsigned long vmaddr)
 		page = alloc_page(GFP_KERNEL|__GFP_REPEAT);
 		if (!page)
 			return NULL;
-		pgtable_page_ctor(page);
+		if (!pgtable_page_ctor(page)) {
+			__free_page(page);
+			return NULL;
+		}
 		atomic_set(&page->_mapcount, 1);
 		table = (unsigned long *) page_to_phys(page);
 		clear_table(table, _PAGE_INVALID, PAGE_SIZE);
@@ -1244,11 +1251,11 @@ void pgtable_trans_huge_deposit(struct mm_struct *mm, pmd_t *pmdp,
 	assert_spin_locked(&mm->page_table_lock);
 
 	/* FIFO */
-	if (!mm->pmd_huge_pte)
+	if (!pmd_huge_pte(mm, pmdp))
 		INIT_LIST_HEAD(lh);
 	else
-		list_add(lh, (struct list_head *) mm->pmd_huge_pte);
-	mm->pmd_huge_pte = pgtable;
+		list_add(lh, (struct list_head *) pmd_huge_pte(mm, pmdp));
+	pmd_huge_pte(mm, pmdp) = pgtable;
 }
 
 pgtable_t pgtable_trans_huge_withdraw(struct mm_struct *mm, pmd_t *pmdp)
@@ -1260,12 +1267,12 @@ pgtable_t pgtable_trans_huge_withdraw(struct mm_struct *mm, pmd_t *pmdp)
 	assert_spin_locked(&mm->page_table_lock);
 
 	/* FIFO */
-	pgtable = mm->pmd_huge_pte;
+	pgtable = pmd_huge_pte(mm, pmdp);
 	lh = (struct list_head *) pgtable;
 	if (list_empty(lh))
-		mm->pmd_huge_pte = NULL;
+		pmd_huge_pte(mm, pmdp) = NULL;
 	else {
-		mm->pmd_huge_pte = (pgtable_t) lh->next;
+		pmd_huge_pte(mm, pmdp) = (pgtable_t) lh->next;
 		list_del(lh);
 	}
 	ptep = (pte_t *) pgtable;
diff --git a/arch/s390/pci/pci.c b/arch/s390/pci/pci.c
index 0c9a17780e4b..bf7c73d71eef 100644
--- a/arch/s390/pci/pci.c
+++ b/arch/s390/pci/pci.c
@@ -530,20 +530,6 @@ static void zpci_unmap_resources(struct zpci_dev *zdev)
 	}
 }
 
-struct zpci_dev *zpci_alloc_device(void)
-{
-	struct zpci_dev *zdev;
-
-	/* Alloc memory for our private pci device data */
-	zdev = kzalloc(sizeof(*zdev), GFP_KERNEL);
-	return zdev ? : ERR_PTR(-ENOMEM);
-}
-
-void zpci_free_device(struct zpci_dev *zdev)
-{
-	kfree(zdev);
-}
-
 int pcibios_add_platform_entries(struct pci_dev *pdev)
 {
 	return zpci_sysfs_add_device(&pdev->dev);
@@ -579,37 +565,6 @@ static void zpci_irq_exit(void)
 	unregister_adapter_interrupt(&zpci_airq);
 }
 
-static struct resource *zpci_alloc_bus_resource(unsigned long start, unsigned long size,
-						unsigned long flags, int domain)
-{
-	struct resource *r;
-	char *name;
-	int rc;
-
-	r = kzalloc(sizeof(*r), GFP_KERNEL);
-	if (!r)
-		return ERR_PTR(-ENOMEM);
-	r->start = start;
-	r->end = r->start + size - 1;
-	r->flags = flags;
-	r->parent = &iomem_resource;
-	name = kmalloc(18, GFP_KERNEL);
-	if (!name) {
-		kfree(r);
-		return ERR_PTR(-ENOMEM);
-	}
-	sprintf(name, "PCI Bus: %04x:%02x", domain, ZPCI_BUS_NR);
-	r->name = name;
-
-	rc = request_resource(&iomem_resource, r);
-	if (rc) {
-		kfree(r->name);
-		kfree(r);
-		return ERR_PTR(-ENOMEM);
-	}
-	return r;
-}
-
 static int zpci_alloc_iomap(struct zpci_dev *zdev)
 {
 	int entry;
@@ -633,6 +588,82 @@ static void zpci_free_iomap(struct zpci_dev *zdev, int entry)
 	spin_unlock(&zpci_iomap_lock);
 }
 
+static struct resource *__alloc_res(struct zpci_dev *zdev, unsigned long start,
+				    unsigned long size, unsigned long flags)
+{
+	struct resource *r;
+
+	r = kzalloc(sizeof(*r), GFP_KERNEL);
+	if (!r)
+		return NULL;
+
+	r->start = start;
+	r->end = r->start + size - 1;
+	r->flags = flags;
+	r->name = zdev->res_name;
+
+	if (request_resource(&iomem_resource, r)) {
+		kfree(r);
+		return NULL;
+	}
+	return r;
+}
+
+static int zpci_setup_bus_resources(struct zpci_dev *zdev,
+				    struct list_head *resources)
+{
+	unsigned long addr, size, flags;
+	struct resource *res;
+	int i, entry;
+
+	snprintf(zdev->res_name, sizeof(zdev->res_name),
+		 "PCI Bus %04x:%02x", zdev->domain, ZPCI_BUS_NR);
+
+	for (i = 0; i < PCI_BAR_COUNT; i++) {
+		if (!zdev->bars[i].size)
+			continue;
+		entry = zpci_alloc_iomap(zdev);
+		if (entry < 0)
+			return entry;
+		zdev->bars[i].map_idx = entry;
+
+		/* only MMIO is supported */
+		flags = IORESOURCE_MEM;
+		if (zdev->bars[i].val & 8)
+			flags |= IORESOURCE_PREFETCH;
+		if (zdev->bars[i].val & 4)
+			flags |= IORESOURCE_MEM_64;
+
+		addr = ZPCI_IOMAP_ADDR_BASE + ((u64) entry << 48);
+
+		size = 1UL << zdev->bars[i].size;
+
+		res = __alloc_res(zdev, addr, size, flags);
+		if (!res) {
+			zpci_free_iomap(zdev, entry);
+			return -ENOMEM;
+		}
+		zdev->bars[i].res = res;
+		pci_add_resource(resources, res);
+	}
+
+	return 0;
+}
+
+static void zpci_cleanup_bus_resources(struct zpci_dev *zdev)
+{
+	int i;
+
+	for (i = 0; i < PCI_BAR_COUNT; i++) {
+		if (!zdev->bars[i].size)
+			continue;
+
+		zpci_free_iomap(zdev, zdev->bars[i].map_idx);
+		release_resource(zdev->bars[i].res);
+		kfree(zdev->bars[i].res);
+	}
+}
+
 int pcibios_add_device(struct pci_dev *pdev)
 {
 	struct zpci_dev *zdev = get_zdev(pdev);
@@ -729,52 +760,6 @@ struct dev_pm_ops pcibios_pm_ops = {
 };
 #endif /* CONFIG_HIBERNATE_CALLBACKS */
 
-static int zpci_scan_bus(struct zpci_dev *zdev)
-{
-	struct resource *res;
-	LIST_HEAD(resources);
-	int i;
-
-	/* allocate mapping entry for each used bar */
-	for (i = 0; i < PCI_BAR_COUNT; i++) {
-		unsigned long addr, size, flags;
-		int entry;
-
-		if (!zdev->bars[i].size)
-			continue;
-		entry = zpci_alloc_iomap(zdev);
-		if (entry < 0)
-			return entry;
-		zdev->bars[i].map_idx = entry;
-
-		/* only MMIO is supported */
-		flags = IORESOURCE_MEM;
-		if (zdev->bars[i].val & 8)
-			flags |= IORESOURCE_PREFETCH;
-		if (zdev->bars[i].val & 4)
-			flags |= IORESOURCE_MEM_64;
-
-		addr = ZPCI_IOMAP_ADDR_BASE + ((u64) entry << 48);
-
-		size = 1UL << zdev->bars[i].size;
-
-		res = zpci_alloc_bus_resource(addr, size, flags, zdev->domain);
-		if (IS_ERR(res)) {
-			zpci_free_iomap(zdev, entry);
-			return PTR_ERR(res);
-		}
-		pci_add_resource(&resources, res);
-	}
-
-	zdev->bus = pci_scan_root_bus(NULL, ZPCI_BUS_NR, &pci_root_ops,
-				      zdev, &resources);
-	if (!zdev->bus)
-		return -EIO;
-
-	zdev->bus->max_bus_speed = zdev->max_bus_speed;
-	return 0;
-}
-
 static int zpci_alloc_domain(struct zpci_dev *zdev)
 {
 	spin_lock(&zpci_domain_lock);
@@ -795,6 +780,41 @@ static void zpci_free_domain(struct zpci_dev *zdev)
 	spin_unlock(&zpci_domain_lock);
 }
 
+void pcibios_remove_bus(struct pci_bus *bus)
+{
+	struct zpci_dev *zdev = get_zdev_by_bus(bus);
+
+	zpci_exit_slot(zdev);
+	zpci_cleanup_bus_resources(zdev);
+	zpci_free_domain(zdev);
+
+	spin_lock(&zpci_list_lock);
+	list_del(&zdev->entry);
+	spin_unlock(&zpci_list_lock);
+
+	kfree(zdev);
+}
+
+static int zpci_scan_bus(struct zpci_dev *zdev)
+{
+	LIST_HEAD(resources);
+	int ret;
+
+	ret = zpci_setup_bus_resources(zdev, &resources);
+	if (ret)
+		return ret;
+
+	zdev->bus = pci_scan_root_bus(NULL, ZPCI_BUS_NR, &pci_root_ops,
+				      zdev, &resources);
+	if (!zdev->bus) {
+		zpci_cleanup_bus_resources(zdev);
+		return -EIO;
+	}
+
+	zdev->bus->max_bus_speed = zdev->max_bus_speed;
+	return 0;
+}
+
 int zpci_enable_device(struct zpci_dev *zdev)
 {
 	int rc;
diff --git a/arch/s390/pci/pci_clp.c b/arch/s390/pci/pci_clp.c
index 84147984224a..c747394029ee 100644
--- a/arch/s390/pci/pci_clp.c
+++ b/arch/s390/pci/pci_clp.c
@@ -155,9 +155,9 @@ int clp_add_pci_device(u32 fid, u32 fh, int configured)
 	int rc;
 
 	zpci_dbg(3, "add fid:%x, fh:%x, c:%d\n", fid, fh, configured);
-	zdev = zpci_alloc_device();
-	if (IS_ERR(zdev))
-		return PTR_ERR(zdev);
+	zdev = kzalloc(sizeof(*zdev), GFP_KERNEL);
+	if (!zdev)
+		return -ENOMEM;
 
 	zdev->fh = fh;
 	zdev->fid = fid;
@@ -178,7 +178,7 @@ int clp_add_pci_device(u32 fid, u32 fh, int configured)
 	return 0;
 
 error:
-	zpci_free_device(zdev);
+	kfree(zdev);
 	return rc;
 }
 
diff --git a/arch/s390/pci/pci_event.c b/arch/s390/pci/pci_event.c
index 278e671ec9ac..800f064b0da7 100644
--- a/arch/s390/pci/pci_event.c
+++ b/arch/s390/pci/pci_event.c
@@ -11,6 +11,7 @@
 #include <linux/kernel.h>
 #include <linux/pci.h>
 #include <asm/pci_debug.h>
+#include <asm/sclp.h>
 
 /* Content Code Description for PCI Function Error */
 struct zpci_ccdf_err {
@@ -42,10 +43,27 @@ struct zpci_ccdf_avail {
 	u16 pec;			/* PCI event code */
 } __packed;
 
-static void zpci_event_log_avail(struct zpci_ccdf_avail *ccdf)
+void zpci_event_error(void *data)
 {
+	struct zpci_ccdf_err *ccdf = data;
+	struct zpci_dev *zdev = get_zdev_by_fid(ccdf->fid);
+
+	zpci_err("error CCDF:\n");
+	zpci_err_hex(ccdf, sizeof(*ccdf));
+
+	if (!zdev)
+		return;
+
+	pr_err("%s: Event 0x%x reports an error for PCI function 0x%x\n",
+	       pci_name(zdev->pdev), ccdf->pec, ccdf->fid);
+}
+
+void zpci_event_availability(void *data)
+{
+	struct zpci_ccdf_avail *ccdf = data;
 	struct zpci_dev *zdev = get_zdev_by_fid(ccdf->fid);
 	struct pci_dev *pdev = zdev ? zdev->pdev : NULL;
+	int ret;
 
 	pr_info("%s: Event 0x%x reconfigured PCI function 0x%x\n",
 		pdev ? pci_name(pdev) : "n/a", ccdf->pec, ccdf->fid);
@@ -53,36 +71,47 @@ static void zpci_event_log_avail(struct zpci_ccdf_avail *ccdf)
 	zpci_err_hex(ccdf, sizeof(*ccdf));
 
 	switch (ccdf->pec) {
-	case 0x0301:
-		zpci_enable_device(zdev);
+	case 0x0301: /* Standby -> Configured */
+		if (!zdev || zdev->state == ZPCI_FN_STATE_CONFIGURED)
+			break;
+		zdev->state = ZPCI_FN_STATE_CONFIGURED;
+		ret = zpci_enable_device(zdev);
+		if (ret)
+			break;
+		pci_rescan_bus(zdev->bus);
 		break;
-	case 0x0302:
+	case 0x0302: /* Reserved -> Standby */
 		clp_add_pci_device(ccdf->fid, ccdf->fh, 0);
 		break;
-	case 0x0306:
+	case 0x0303: /* Deconfiguration requested */
+		if (pdev)
+			pci_stop_and_remove_bus_device(pdev);
+
+		ret = zpci_disable_device(zdev);
+		if (ret)
+			break;
+
+		ret = sclp_pci_deconfigure(zdev->fid);
+		zpci_dbg(3, "deconf fid:%x, rc:%d\n", zdev->fid, ret);
+		if (!ret)
+			zdev->state = ZPCI_FN_STATE_STANDBY;
+
+		break;
+	case 0x0304: /* Configured -> Standby */
+		if (pdev)
+			pci_stop_and_remove_bus_device(pdev);
+
+		zpci_disable_device(zdev);
+		zdev->state = ZPCI_FN_STATE_STANDBY;
+		break;
+	case 0x0306: /* 0x308 or 0x302 for multiple devices */
 		clp_rescan_pci_devices();
 		break;
+	case 0x0308: /* Standby -> Reserved */
+		pci_stop_root_bus(zdev->bus);
+		pci_remove_root_bus(zdev->bus);
+		break;
 	default:
 		break;
 	}
 }
-
-void zpci_event_error(void *data)
-{
-	struct zpci_ccdf_err *ccdf = data;
-	struct zpci_dev *zdev = get_zdev_by_fid(ccdf->fid);
-
-	zpci_err("error CCDF:\n");
-	zpci_err_hex(ccdf, sizeof(*ccdf));
-
-	if (!zdev)
-		return;
-
-	pr_err("%s: Event 0x%x reports an error for PCI function 0x%x\n",
-	       pci_name(zdev->pdev), ccdf->pec, ccdf->fid);
-}
-
-void zpci_event_availability(void *data)
-{
-	zpci_event_log_avail(data);
-}
diff --git a/arch/score/include/asm/pgalloc.h b/arch/score/include/asm/pgalloc.h
index 716b3fd1d863..2e067657db98 100644
--- a/arch/score/include/asm/pgalloc.h
+++ b/arch/score/include/asm/pgalloc.h
@@ -54,9 +54,12 @@ static inline struct page *pte_alloc_one(struct mm_struct *mm,
 	struct page *pte;
 
 	pte = alloc_pages(GFP_KERNEL | __GFP_REPEAT, PTE_ORDER);
-	if (pte) {
-		clear_highpage(pte);
-		pgtable_page_ctor(pte);
+	if (!pte)
+		return NULL;
+	clear_highpage(pte);
+	if (!pgtable_page_ctor(pte)) {
+		__free_page(pte);
+		return NULL;
 	}
 	return pte;
 }
diff --git a/arch/score/include/asm/thread_info.h b/arch/score/include/asm/thread_info.h
index 1425cc034872..656b7ada9326 100644
--- a/arch/score/include/asm/thread_info.h
+++ b/arch/score/include/asm/thread_info.h
@@ -72,8 +72,6 @@ register struct thread_info *__current_thread_info __asm__("r28");
 
 #endif /* !__ASSEMBLY__ */
 
-#define PREEMPT_ACTIVE		0x10000000
-
 /*
  * thread information flags
  * - these are process state flags that various assembly files may need to
diff --git a/arch/sh/Kconfig b/arch/sh/Kconfig
index 224f4bc9925e..9b0979f4df7a 100644
--- a/arch/sh/Kconfig
+++ b/arch/sh/Kconfig
@@ -1,5 +1,6 @@
 config SUPERH
 	def_bool y
+	select ARCH_MIGHT_HAVE_PC_PARPORT
 	select EXPERT
 	select CLKDEV_LOOKUP
 	select HAVE_IDE if HAS_IOPORT
@@ -711,7 +712,6 @@ config CC_STACKPROTECTOR
 config SMP
 	bool "Symmetric multi-processing support"
 	depends on SYS_SUPPORTS_SMP
-	select USE_GENERIC_SMP_HELPERS
 	---help---
 	  This enables support for systems with more than one CPU. If you have
 	  a system with only one CPU, like most personal computers, say N. If
diff --git a/arch/sh/boards/mach-ecovec24/setup.c b/arch/sh/boards/mach-ecovec24/setup.c
index 1fa8be409771..122f737a901f 100644
--- a/arch/sh/boards/mach-ecovec24/setup.c
+++ b/arch/sh/boards/mach-ecovec24/setup.c
@@ -15,6 +15,7 @@
 #include <linux/mmc/sh_mmcif.h>
 #include <linux/mmc/sh_mobile_sdhi.h>
 #include <linux/mtd/physmap.h>
+#include <linux/mfd/tmio.h>
 #include <linux/gpio.h>
 #include <linux/interrupt.h>
 #include <linux/io.h>
diff --git a/arch/sh/include/asm/mmu_context.h b/arch/sh/include/asm/mmu_context.h
index 21c5088788da..b9d9489a5012 100644
--- a/arch/sh/include/asm/mmu_context.h
+++ b/arch/sh/include/asm/mmu_context.h
@@ -81,7 +81,7 @@ static inline void get_mmu_context(struct mm_struct *mm, unsigned int cpu)
 
 		/*
 		 * Fix version; Note that we avoid version #0
-		 * to distingush NO_CONTEXT.
+		 * to distinguish NO_CONTEXT.
 		 */
 		if (!asid)
 			asid = MMU_CONTEXT_FIRST_VERSION;
diff --git a/arch/sh/include/asm/pgalloc.h b/arch/sh/include/asm/pgalloc.h
index 8c00785c60d5..a33673b3687d 100644
--- a/arch/sh/include/asm/pgalloc.h
+++ b/arch/sh/include/asm/pgalloc.h
@@ -47,7 +47,10 @@ static inline pgtable_t pte_alloc_one(struct mm_struct *mm,
 	if (!pg)
 		return NULL;
 	page = virt_to_page(pg);
-	pgtable_page_ctor(page);
+	if (!pgtable_page_ctor(page)) {
+		quicklist_free(QUICK_PT, NULL, pg);
+		return NULL;
+	}
 	return page;
 }
 
diff --git a/arch/sh/include/asm/thread_info.h b/arch/sh/include/asm/thread_info.h
index 45a93669289d..ad27ffa65e2e 100644
--- a/arch/sh/include/asm/thread_info.h
+++ b/arch/sh/include/asm/thread_info.h
@@ -41,8 +41,6 @@ struct thread_info {
 
 #endif
 
-#define PREEMPT_ACTIVE		0x10000000
-
 #if defined(CONFIG_4KSTACKS)
 #define THREAD_SHIFT	12
 #else
diff --git a/arch/sh/kernel/entry-common.S b/arch/sh/kernel/entry-common.S
index 9b6e4beeb296..ca46834294b7 100644
--- a/arch/sh/kernel/entry-common.S
+++ b/arch/sh/kernel/entry-common.S
@@ -108,7 +108,7 @@ need_resched:
 	and	#(0xf0>>1), r0		! interrupts off (exception path)?
 	cmp/eq	#(0xf0>>1), r0
 	bt	noresched
-	mov.l	3f, r0
+	mov.l	1f, r0
 	jsr	@r0			! call preempt_schedule_irq
 	 nop
 	bra	need_resched
@@ -119,9 +119,7 @@ noresched:
 	 nop
 
 	.align 2
-1:	.long	PREEMPT_ACTIVE
-2:	.long	schedule
-3:	.long	preempt_schedule_irq
+1:	.long	preempt_schedule_irq
 #endif
 
 ENTRY(resume_userspace)
diff --git a/arch/sparc/Kconfig b/arch/sparc/Kconfig
index 78c4fdb91bc5..d4f7a6a163dc 100644
--- a/arch/sparc/Kconfig
+++ b/arch/sparc/Kconfig
@@ -12,6 +12,7 @@ config 64BIT
 config SPARC
 	bool
 	default y
+	select ARCH_MIGHT_HAVE_PC_PARPORT if SPARC64 && PCI
 	select OF
 	select OF_PROMTREE
 	select HAVE_IDE
@@ -28,7 +29,6 @@ config SPARC
 	select HAVE_ARCH_JUMP_LABEL
 	select GENERIC_IRQ_SHOW
 	select ARCH_WANT_IPC_PARSE_VERSION
-	select USE_GENERIC_SMP_HELPERS if SMP
 	select GENERIC_PCI_IOMAP
 	select HAVE_NMI_WATCHDOG if SPARC64
 	select HAVE_BPF_JIT
@@ -64,6 +64,7 @@ config SPARC64
 	select HAVE_DYNAMIC_FTRACE
 	select HAVE_FTRACE_MCOUNT_RECORD
 	select HAVE_SYSCALL_TRACEPOINTS
+	select HAVE_CONTEXT_TRACKING
 	select HAVE_DEBUG_KMEMLEAK
 	select RTC_DRV_CMOS
 	select RTC_DRV_BQ4802
diff --git a/arch/sparc/include/asm/hardirq_32.h b/arch/sparc/include/asm/hardirq_32.h
index 162007643cdc..ee93923b7f82 100644
--- a/arch/sparc/include/asm/hardirq_32.h
+++ b/arch/sparc/include/asm/hardirq_32.h
@@ -7,7 +7,6 @@
 #ifndef __SPARC_HARDIRQ_H
 #define __SPARC_HARDIRQ_H
 
-#define HARDIRQ_BITS    8
 #include <asm-generic/hardirq.h>
 
 #endif /* __SPARC_HARDIRQ_H */
diff --git a/arch/sparc/include/asm/hardirq_64.h b/arch/sparc/include/asm/hardirq_64.h
index 7c29fd1a87aa..f478ff1ddd02 100644
--- a/arch/sparc/include/asm/hardirq_64.h
+++ b/arch/sparc/include/asm/hardirq_64.h
@@ -14,6 +14,4 @@
 
 void ack_bad_irq(unsigned int irq);
 
-#define HARDIRQ_BITS	8
-
 #endif /* !(__SPARC64_HARDIRQ_H) */
diff --git a/arch/sparc/include/asm/mmu_64.h b/arch/sparc/include/asm/mmu_64.h
index 76092c4dd277..f668797ae234 100644
--- a/arch/sparc/include/asm/mmu_64.h
+++ b/arch/sparc/include/asm/mmu_64.h
@@ -93,7 +93,6 @@ typedef struct {
 	spinlock_t		lock;
 	unsigned long		sparc64_ctx_val;
 	unsigned long		huge_pte_count;
-	struct page		*pgtable_page;
 	struct tsb_config	tsb_block[MM_NUM_TSBS];
 	struct hv_tsb_descr	tsb_descr[MM_NUM_TSBS];
 } mm_context_t;
diff --git a/arch/sparc/include/asm/page_64.h b/arch/sparc/include/asm/page_64.h
index e15538899f3d..aac53fcea807 100644
--- a/arch/sparc/include/asm/page_64.h
+++ b/arch/sparc/include/asm/page_64.h
@@ -15,7 +15,10 @@
 #define DCACHE_ALIASING_POSSIBLE
 #endif
 
-#define HPAGE_SHIFT		22
+#define HPAGE_SHIFT		23
+#define REAL_HPAGE_SHIFT	22
+
+#define REAL_HPAGE_SIZE		(_AC(1,UL) << REAL_HPAGE_SHIFT)
 
 #if defined(CONFIG_HUGETLB_PAGE) || defined(CONFIG_TRANSPARENT_HUGEPAGE)
 #define HPAGE_SIZE		(_AC(1,UL) << HPAGE_SHIFT)
@@ -53,8 +56,8 @@ extern void copy_user_page(void *to, void *from, unsigned long vaddr, struct pag
 /* These are used to make use of C type-checking.. */
 typedef struct { unsigned long pte; } pte_t;
 typedef struct { unsigned long iopte; } iopte_t;
-typedef struct { unsigned int pmd; } pmd_t;
-typedef struct { unsigned int pgd; } pgd_t;
+typedef struct { unsigned long pmd; } pmd_t;
+typedef struct { unsigned long pgd; } pgd_t;
 typedef struct { unsigned long pgprot; } pgprot_t;
 
 #define pte_val(x)	((x).pte)
@@ -73,8 +76,8 @@ typedef struct { unsigned long pgprot; } pgprot_t;
 /* .. while these make it easier on the compiler */
 typedef unsigned long pte_t;
 typedef unsigned long iopte_t;
-typedef unsigned int pmd_t;
-typedef unsigned int pgd_t;
+typedef unsigned long pmd_t;
+typedef unsigned long pgd_t;
 typedef unsigned long pgprot_t;
 
 #define pte_val(x)	(x)
@@ -93,18 +96,44 @@ typedef unsigned long pgprot_t;
 
 typedef pte_t *pgtable_t;
 
+/* These two values define the virtual address space range in which we
+ * must forbid 64-bit user processes from making mappings.  It used to
+ * represent precisely the virtual address space hole present in most
+ * early sparc64 chips including UltraSPARC-I.  But now it also is
+ * further constrained by the limits of our page tables, which is
+ * 43-bits of virtual address.
+ */
+#define SPARC64_VA_HOLE_TOP	_AC(0xfffffc0000000000,UL)
+#define SPARC64_VA_HOLE_BOTTOM	_AC(0x0000040000000000,UL)
+
+/* The next two defines specify the actual exclusion region we
+ * enforce, wherein we use a 4GB red zone on each side of the VA hole.
+ */
+#define VA_EXCLUDE_START (SPARC64_VA_HOLE_BOTTOM - (1UL << 32UL))
+#define VA_EXCLUDE_END   (SPARC64_VA_HOLE_TOP + (1UL << 32UL))
+
 #define TASK_UNMAPPED_BASE	(test_thread_flag(TIF_32BIT) ? \
-				 (_AC(0x0000000070000000,UL)) : \
-				 (_AC(0xfffff80000000000,UL) + (1UL << 32UL)))
+				 _AC(0x0000000070000000,UL) : \
+				 VA_EXCLUDE_END)
 
 #include <asm-generic/memory_model.h>
 
+#define PAGE_OFFSET_BY_BITS(X)	(-(_AC(1,UL) << (X)))
+extern unsigned long PAGE_OFFSET;
+
 #endif /* !(__ASSEMBLY__) */
 
-/* We used to stick this into a hard-coded global register (%g4)
- * but that does not make sense anymore.
+/* The maximum number of physical memory address bits we support, this
+ * is used to size various tables used to manage kernel TLB misses and
+ * also the sparsemem code.
+ */
+#define MAX_PHYS_ADDRESS_BITS	47
+
+/* These two shift counts are used when indexing sparc64_valid_addr_bitmap
+ * and kpte_linear_bitmap.
  */
-#define PAGE_OFFSET		_AC(0xFFFFF80000000000,UL)
+#define ILOG2_4MB		22
+#define ILOG2_256MB		28
 
 #ifndef __ASSEMBLY__
 
diff --git a/arch/sparc/include/asm/pgtable_64.h b/arch/sparc/include/asm/pgtable_64.h
index 36760317814f..8358dc144959 100644
--- a/arch/sparc/include/asm/pgtable_64.h
+++ b/arch/sparc/include/asm/pgtable_64.h
@@ -48,18 +48,18 @@
 /* PMD_SHIFT determines the size of the area a second-level page
  * table can map
  */
-#define PMD_SHIFT	(PAGE_SHIFT + (PAGE_SHIFT-4))
+#define PMD_SHIFT	(PAGE_SHIFT + (PAGE_SHIFT-3))
 #define PMD_SIZE	(_AC(1,UL) << PMD_SHIFT)
 #define PMD_MASK	(~(PMD_SIZE-1))
-#define PMD_BITS	(PAGE_SHIFT - 2)
+#define PMD_BITS	(PAGE_SHIFT - 3)
 
 /* PGDIR_SHIFT determines what a third-level page table entry can map */
-#define PGDIR_SHIFT	(PAGE_SHIFT + (PAGE_SHIFT-4) + PMD_BITS)
+#define PGDIR_SHIFT	(PAGE_SHIFT + (PAGE_SHIFT-3) + PMD_BITS)
 #define PGDIR_SIZE	(_AC(1,UL) << PGDIR_SHIFT)
 #define PGDIR_MASK	(~(PGDIR_SIZE-1))
-#define PGDIR_BITS	(PAGE_SHIFT - 2)
+#define PGDIR_BITS	(PAGE_SHIFT - 3)
 
-#if (PGDIR_SHIFT + PGDIR_BITS) != 44
+#if (PGDIR_SHIFT + PGDIR_BITS) != 43
 #error Page table parameters do not cover virtual address space properly.
 #endif
 
@@ -67,35 +67,12 @@
 #error PMD_SHIFT must equal HPAGE_SHIFT for transparent huge pages.
 #endif
 
-/* PMDs point to PTE tables which are 4K aligned.  */
-#define PMD_PADDR	_AC(0xfffffffe,UL)
-#define PMD_PADDR_SHIFT	_AC(11,UL)
-
-#define PMD_ISHUGE	_AC(0x00000001,UL)
-
-/* This is the PMD layout when PMD_ISHUGE is set.  With 4MB huge
- * pages, this frees up a bunch of bits in the layout that we can
- * use for the protection settings and software metadata.
- */
-#define PMD_HUGE_PADDR		_AC(0xfffff800,UL)
-#define PMD_HUGE_PROTBITS	_AC(0x000007ff,UL)
-#define PMD_HUGE_PRESENT	_AC(0x00000400,UL)
-#define PMD_HUGE_WRITE		_AC(0x00000200,UL)
-#define PMD_HUGE_DIRTY		_AC(0x00000100,UL)
-#define PMD_HUGE_ACCESSED	_AC(0x00000080,UL)
-#define PMD_HUGE_EXEC		_AC(0x00000040,UL)
-#define PMD_HUGE_SPLITTING	_AC(0x00000020,UL)
-
-/* PGDs point to PMD tables which are 8K aligned.  */
-#define PGD_PADDR	_AC(0xfffffffc,UL)
-#define PGD_PADDR_SHIFT	_AC(11,UL)
-
 #ifndef __ASSEMBLY__
 
 #include <linux/sched.h>
 
 /* Entries per page directory level. */
-#define PTRS_PER_PTE	(1UL << (PAGE_SHIFT-4))
+#define PTRS_PER_PTE	(1UL << (PAGE_SHIFT-3))
 #define PTRS_PER_PMD	(1UL << PMD_BITS)
 #define PTRS_PER_PGD	(1UL << PGDIR_BITS)
 
@@ -112,6 +89,7 @@
 #define _PAGE_VALID	  _AC(0x8000000000000000,UL) /* Valid TTE            */
 #define _PAGE_R	  	  _AC(0x8000000000000000,UL) /* Keep ref bit uptodate*/
 #define _PAGE_SPECIAL     _AC(0x0200000000000000,UL) /* Special page         */
+#define _PAGE_PMD_HUGE    _AC(0x0100000000000000,UL) /* Huge page            */
 
 /* Advertise support for _PAGE_SPECIAL */
 #define __HAVE_ARCH_PTE_SPECIAL
@@ -125,6 +103,7 @@
 #define _PAGE_IE_4U	  _AC(0x0800000000000000,UL) /* Invert Endianness    */
 #define _PAGE_SOFT2_4U	  _AC(0x07FC000000000000,UL) /* Software bits, set 2 */
 #define _PAGE_SPECIAL_4U  _AC(0x0200000000000000,UL) /* Special page         */
+#define _PAGE_PMD_HUGE_4U _AC(0x0100000000000000,UL) /* Huge page            */
 #define _PAGE_RES1_4U	  _AC(0x0002000000000000,UL) /* Reserved             */
 #define _PAGE_SZ32MB_4U	  _AC(0x0001000000000000,UL) /* (Panther) 32MB page  */
 #define _PAGE_SZ256MB_4U  _AC(0x2001000000000000,UL) /* (Panther) 256MB page */
@@ -155,6 +134,7 @@
 #define _PAGE_READ_4V	  _AC(0x0800000000000000,UL) /* Readable SW Bit      */
 #define _PAGE_WRITE_4V	  _AC(0x0400000000000000,UL) /* Writable SW Bit      */
 #define _PAGE_SPECIAL_4V  _AC(0x0200000000000000,UL) /* Special page         */
+#define _PAGE_PMD_HUGE_4V _AC(0x0100000000000000,UL) /* Huge page            */
 #define _PAGE_PADDR_4V	  _AC(0x00FFFFFFFFFFE000,UL) /* paddr[55:13]         */
 #define _PAGE_IE_4V	  _AC(0x0000000000001000,UL) /* Invert Endianness    */
 #define _PAGE_E_4V	  _AC(0x0000000000000800,UL) /* side-Effect          */
@@ -180,6 +160,10 @@
 #define _PAGE_SZBITS_4U	_PAGE_SZ8K_4U
 #define _PAGE_SZBITS_4V	_PAGE_SZ8K_4V
 
+#if REAL_HPAGE_SHIFT != 22
+#error REAL_HPAGE_SHIFT and _PAGE_SZHUGE_foo must match up
+#endif
+
 #define _PAGE_SZHUGE_4U	_PAGE_SZ4MB_4U
 #define _PAGE_SZHUGE_4V	_PAGE_SZ4MB_4V
 
@@ -239,16 +223,13 @@ static inline pte_t pfn_pte(unsigned long pfn, pgprot_t prot)
 #define mk_pte(page, pgprot)	pfn_pte(page_to_pfn(page), (pgprot))
 
 #ifdef CONFIG_TRANSPARENT_HUGEPAGE
-extern pmd_t pfn_pmd(unsigned long page_nr, pgprot_t pgprot);
-#define mk_pmd(page, pgprot)	pfn_pmd(page_to_pfn(page), (pgprot))
-
-extern pmd_t pmd_modify(pmd_t pmd, pgprot_t newprot);
-
-static inline pmd_t pmd_mkhuge(pmd_t pmd)
+static inline pmd_t pfn_pmd(unsigned long page_nr, pgprot_t pgprot)
 {
-	/* Do nothing, mk_pmd() does this part.  */
-	return pmd;
+	pte_t pte = pfn_pte(page_nr, pgprot);
+
+	return __pmd(pte_val(pte));
 }
+#define mk_pmd(page, pgprot)	pfn_pmd(page_to_pfn(page), (pgprot))
 #endif
 
 /* This one can be done with two shifts.  */
@@ -309,14 +290,25 @@ static inline pte_t pte_modify(pte_t pte, pgprot_t prot)
 	: "=r" (mask), "=r" (tmp)
 	: "i" (_PAGE_PADDR_4U | _PAGE_MODIFIED_4U | _PAGE_ACCESSED_4U |
 	       _PAGE_CP_4U | _PAGE_CV_4U | _PAGE_E_4U | _PAGE_PRESENT_4U |
-	       _PAGE_SPECIAL),
+	       _PAGE_SPECIAL | _PAGE_PMD_HUGE | _PAGE_SZALL_4U),
 	  "i" (_PAGE_PADDR_4V | _PAGE_MODIFIED_4V | _PAGE_ACCESSED_4V |
 	       _PAGE_CP_4V | _PAGE_CV_4V | _PAGE_E_4V | _PAGE_PRESENT_4V |
-	       _PAGE_SPECIAL));
+	       _PAGE_SPECIAL | _PAGE_PMD_HUGE | _PAGE_SZALL_4V));
 
 	return __pte((pte_val(pte) & mask) | (pgprot_val(prot) & ~mask));
 }
 
+#ifdef CONFIG_TRANSPARENT_HUGEPAGE
+static inline pmd_t pmd_modify(pmd_t pmd, pgprot_t newprot)
+{
+	pte_t pte = __pte(pmd_val(pmd));
+
+	pte = pte_modify(pte, newprot);
+
+	return __pmd(pte_val(pte));
+}
+#endif
+
 static inline pte_t pgoff_to_pte(unsigned long off)
 {
 	off <<= PAGE_SHIFT;
@@ -357,7 +349,7 @@ static inline pgprot_t pgprot_noncached(pgprot_t prot)
  */
 #define pgprot_noncached pgprot_noncached
 
-#ifdef CONFIG_HUGETLB_PAGE
+#if defined(CONFIG_HUGETLB_PAGE) || defined(CONFIG_TRANSPARENT_HUGEPAGE)
 static inline pte_t pte_mkhuge(pte_t pte)
 {
 	unsigned long mask;
@@ -375,6 +367,17 @@ static inline pte_t pte_mkhuge(pte_t pte)
 
 	return __pte(pte_val(pte) | mask);
 }
+#ifdef CONFIG_TRANSPARENT_HUGEPAGE
+static inline pmd_t pmd_mkhuge(pmd_t pmd)
+{
+	pte_t pte = __pte(pmd_val(pmd));
+
+	pte = pte_mkhuge(pte);
+	pte_val(pte) |= _PAGE_PMD_HUGE;
+
+	return __pmd(pte_val(pte));
+}
+#endif
 #endif
 
 static inline pte_t pte_mkdirty(pte_t pte)
@@ -626,91 +629,130 @@ static inline unsigned long pte_special(pte_t pte)
 	return pte_val(pte) & _PAGE_SPECIAL;
 }
 
-static inline int pmd_large(pmd_t pmd)
+static inline unsigned long pmd_large(pmd_t pmd)
 {
-	return (pmd_val(pmd) & (PMD_ISHUGE | PMD_HUGE_PRESENT)) ==
-		(PMD_ISHUGE | PMD_HUGE_PRESENT);
+	pte_t pte = __pte(pmd_val(pmd));
+
+	return (pte_val(pte) & _PAGE_PMD_HUGE) && pte_present(pte);
 }
 
 #ifdef CONFIG_TRANSPARENT_HUGEPAGE
-static inline int pmd_young(pmd_t pmd)
+static inline unsigned long pmd_young(pmd_t pmd)
 {
-	return pmd_val(pmd) & PMD_HUGE_ACCESSED;
+	pte_t pte = __pte(pmd_val(pmd));
+
+	return pte_young(pte);
 }
 
-static inline int pmd_write(pmd_t pmd)
+static inline unsigned long pmd_write(pmd_t pmd)
 {
-	return pmd_val(pmd) & PMD_HUGE_WRITE;
+	pte_t pte = __pte(pmd_val(pmd));
+
+	return pte_write(pte);
 }
 
 static inline unsigned long pmd_pfn(pmd_t pmd)
 {
-	unsigned long val = pmd_val(pmd) & PMD_HUGE_PADDR;
+	pte_t pte = __pte(pmd_val(pmd));
 
-	return val >> (PAGE_SHIFT - PMD_PADDR_SHIFT);
+	return pte_pfn(pte);
 }
 
-static inline int pmd_trans_splitting(pmd_t pmd)
+static inline unsigned long pmd_trans_huge(pmd_t pmd)
 {
-	return (pmd_val(pmd) & (PMD_ISHUGE|PMD_HUGE_SPLITTING)) ==
-		(PMD_ISHUGE|PMD_HUGE_SPLITTING);
+	pte_t pte = __pte(pmd_val(pmd));
+
+	return pte_val(pte) & _PAGE_PMD_HUGE;
 }
 
-static inline int pmd_trans_huge(pmd_t pmd)
+static inline unsigned long pmd_trans_splitting(pmd_t pmd)
 {
-	return pmd_val(pmd) & PMD_ISHUGE;
+	pte_t pte = __pte(pmd_val(pmd));
+
+	return pmd_trans_huge(pmd) && pte_special(pte);
 }
 
 #define has_transparent_hugepage() 1
 
 static inline pmd_t pmd_mkold(pmd_t pmd)
 {
-	pmd_val(pmd) &= ~PMD_HUGE_ACCESSED;
-	return pmd;
+	pte_t pte = __pte(pmd_val(pmd));
+
+	pte = pte_mkold(pte);
+
+	return __pmd(pte_val(pte));
 }
 
 static inline pmd_t pmd_wrprotect(pmd_t pmd)
 {
-	pmd_val(pmd) &= ~PMD_HUGE_WRITE;
-	return pmd;
+	pte_t pte = __pte(pmd_val(pmd));
+
+	pte = pte_wrprotect(pte);
+
+	return __pmd(pte_val(pte));
 }
 
 static inline pmd_t pmd_mkdirty(pmd_t pmd)
 {
-	pmd_val(pmd) |= PMD_HUGE_DIRTY;
-	return pmd;
+	pte_t pte = __pte(pmd_val(pmd));
+
+	pte = pte_mkdirty(pte);
+
+	return __pmd(pte_val(pte));
 }
 
 static inline pmd_t pmd_mkyoung(pmd_t pmd)
 {
-	pmd_val(pmd) |= PMD_HUGE_ACCESSED;
-	return pmd;
+	pte_t pte = __pte(pmd_val(pmd));
+
+	pte = pte_mkyoung(pte);
+
+	return __pmd(pte_val(pte));
 }
 
 static inline pmd_t pmd_mkwrite(pmd_t pmd)
 {
-	pmd_val(pmd) |= PMD_HUGE_WRITE;
-	return pmd;
+	pte_t pte = __pte(pmd_val(pmd));
+
+	pte = pte_mkwrite(pte);
+
+	return __pmd(pte_val(pte));
 }
 
 static inline pmd_t pmd_mknotpresent(pmd_t pmd)
 {
-	pmd_val(pmd) &= ~PMD_HUGE_PRESENT;
+	unsigned long mask;
+
+	if (tlb_type == hypervisor)
+		mask = _PAGE_PRESENT_4V;
+	else
+		mask = _PAGE_PRESENT_4U;
+
+	pmd_val(pmd) &= ~mask;
+
 	return pmd;
 }
 
 static inline pmd_t pmd_mksplitting(pmd_t pmd)
 {
-	pmd_val(pmd) |= PMD_HUGE_SPLITTING;
-	return pmd;
+	pte_t pte = __pte(pmd_val(pmd));
+
+	pte = pte_mkspecial(pte);
+
+	return __pmd(pte_val(pte));
 }
 
-extern pgprot_t pmd_pgprot(pmd_t entry);
+static inline pgprot_t pmd_pgprot(pmd_t entry)
+{
+	unsigned long val = pmd_val(entry);
+
+	return __pgprot(val);
+}
 #endif
 
 static inline int pmd_present(pmd_t pmd)
 {
-	return pmd_val(pmd) != 0U;
+	return pmd_val(pmd) != 0UL;
 }
 
 #define pmd_none(pmd)			(!pmd_val(pmd))
@@ -728,33 +770,32 @@ static inline void set_pmd_at(struct mm_struct *mm, unsigned long addr,
 
 static inline void pmd_set(struct mm_struct *mm, pmd_t *pmdp, pte_t *ptep)
 {
-	unsigned long val = __pa((unsigned long) (ptep)) >> PMD_PADDR_SHIFT;
+	unsigned long val = __pa((unsigned long) (ptep));
 
 	pmd_val(*pmdp) = val;
 }
 
 #define pud_set(pudp, pmdp)	\
-	(pud_val(*(pudp)) = (__pa((unsigned long) (pmdp)) >> PGD_PADDR_SHIFT))
+	(pud_val(*(pudp)) = (__pa((unsigned long) (pmdp))))
 static inline unsigned long __pmd_page(pmd_t pmd)
 {
-	unsigned long paddr = (unsigned long) pmd_val(pmd);
-#ifdef CONFIG_TRANSPARENT_HUGEPAGE
-	if (pmd_val(pmd) & PMD_ISHUGE)
-		paddr &= PMD_HUGE_PADDR;
-#endif
-	paddr <<= PMD_PADDR_SHIFT;
-	return ((unsigned long) __va(paddr));
+	pte_t pte = __pte(pmd_val(pmd));
+	unsigned long pfn;
+
+	pfn = pte_pfn(pte);
+
+	return ((unsigned long) __va(pfn << PAGE_SHIFT));
 }
 #define pmd_page(pmd) 			virt_to_page((void *)__pmd_page(pmd))
 #define pud_page_vaddr(pud)		\
-	((unsigned long) __va((((unsigned long)pud_val(pud))<<PGD_PADDR_SHIFT)))
+	((unsigned long) __va(pud_val(pud)))
 #define pud_page(pud) 			virt_to_page((void *)pud_page_vaddr(pud))
 #define pmd_bad(pmd)			(0)
-#define pmd_clear(pmdp)			(pmd_val(*(pmdp)) = 0U)
+#define pmd_clear(pmdp)			(pmd_val(*(pmdp)) = 0UL)
 #define pud_none(pud)			(!pud_val(pud))
 #define pud_bad(pud)			(0)
 #define pud_present(pud)		(pud_val(pud) != 0U)
-#define pud_clear(pudp)			(pud_val(*(pudp)) = 0U)
+#define pud_clear(pudp)			(pud_val(*(pudp)) = 0UL)
 
 /* Same in both SUN4V and SUN4U.  */
 #define pte_none(pte) 			(!pte_val(pte))
@@ -789,7 +830,7 @@ static inline pmd_t pmdp_get_and_clear(struct mm_struct *mm,
 				       pmd_t *pmdp)
 {
 	pmd_t pmd = *pmdp;
-	set_pmd_at(mm, addr, pmdp, __pmd(0U));
+	set_pmd_at(mm, addr, pmdp, __pmd(0UL));
 	return pmd;
 }
 
@@ -837,8 +878,8 @@ static inline void __set_pte_at(struct mm_struct *mm, unsigned long addr,
 })
 #endif
 
-extern pgd_t swapper_pg_dir[2048];
-extern pmd_t swapper_low_pmd_dir[2048];
+extern pgd_t swapper_pg_dir[PTRS_PER_PGD];
+extern pmd_t swapper_low_pmd_dir[PTRS_PER_PMD];
 
 extern void paging_init(void);
 extern unsigned long find_ecache_flush_span(unsigned long size);
diff --git a/arch/sparc/include/asm/sparsemem.h b/arch/sparc/include/asm/sparsemem.h
index b99d4e4b6d28..e5e1752d5d78 100644
--- a/arch/sparc/include/asm/sparsemem.h
+++ b/arch/sparc/include/asm/sparsemem.h
@@ -3,9 +3,11 @@
 
 #ifdef __KERNEL__
 
+#include <asm/page.h>
+
 #define SECTION_SIZE_BITS       30
-#define MAX_PHYSADDR_BITS       42
-#define MAX_PHYSMEM_BITS        42
+#define MAX_PHYSADDR_BITS       MAX_PHYS_ADDRESS_BITS
+#define MAX_PHYSMEM_BITS        MAX_PHYS_ADDRESS_BITS
 
 #endif /* !(__KERNEL__) */
 
diff --git a/arch/sparc/include/asm/thread_info_32.h b/arch/sparc/include/asm/thread_info_32.h
index dd3807599bb9..96efa7adc223 100644
--- a/arch/sparc/include/asm/thread_info_32.h
+++ b/arch/sparc/include/asm/thread_info_32.h
@@ -105,8 +105,6 @@ register struct thread_info *current_thread_info_reg asm("g6");
 #define TI_W_SAVED	0x250
 /* #define TI_RESTART_BLOCK 0x25n */ /* Nobody cares */
 
-#define PREEMPT_ACTIVE		0x4000000
-
 /*
  * thread information flag bit numbers
  */
diff --git a/arch/sparc/include/asm/thread_info_64.h b/arch/sparc/include/asm/thread_info_64.h
index d5e504251079..a5f01ac6d0f1 100644
--- a/arch/sparc/include/asm/thread_info_64.h
+++ b/arch/sparc/include/asm/thread_info_64.h
@@ -111,8 +111,6 @@ struct thread_info {
 #define THREAD_SHIFT PAGE_SHIFT
 #endif /* PAGE_SHIFT == 13 */
 
-#define PREEMPT_ACTIVE		0x10000000
-
 /*
  * macros/functions for gaining access to the thread information structure
  */
@@ -192,7 +190,7 @@ register struct thread_info *current_thread_info_reg asm("g6");
 #define TIF_UNALIGNED		5	/* allowed to do unaligned accesses */
 /* flag bit 6 is available */
 #define TIF_32BIT		7	/* 32-bit binary */
-/* flag bit 8 is available */
+#define TIF_NOHZ		8	/* in adaptive nohz mode */
 #define TIF_SECCOMP		9	/* secure computing */
 #define TIF_SYSCALL_AUDIT	10	/* syscall auditing active */
 #define TIF_SYSCALL_TRACEPOINT	11	/* syscall tracepoint instrumentation */
@@ -210,6 +208,7 @@ register struct thread_info *current_thread_info_reg asm("g6");
 #define _TIF_NEED_RESCHED	(1<<TIF_NEED_RESCHED)
 #define _TIF_UNALIGNED		(1<<TIF_UNALIGNED)
 #define _TIF_32BIT		(1<<TIF_32BIT)
+#define _TIF_NOHZ		(1<<TIF_NOHZ)
 #define _TIF_SECCOMP		(1<<TIF_SECCOMP)
 #define _TIF_SYSCALL_AUDIT	(1<<TIF_SYSCALL_AUDIT)
 #define _TIF_SYSCALL_TRACEPOINT	(1<<TIF_SYSCALL_TRACEPOINT)
diff --git a/arch/sparc/include/asm/tlbflush_64.h b/arch/sparc/include/asm/tlbflush_64.h
index f0d6a9700f4c..3c3c89f52643 100644
--- a/arch/sparc/include/asm/tlbflush_64.h
+++ b/arch/sparc/include/asm/tlbflush_64.h
@@ -1,7 +1,6 @@
 #ifndef _SPARC64_TLBFLUSH_H
 #define _SPARC64_TLBFLUSH_H
 
-#include <linux/mm.h>
 #include <asm/mmu_context.h>
 
 /* TSB flush operations. */
diff --git a/arch/sparc/include/asm/tsb.h b/arch/sparc/include/asm/tsb.h
index e696432b950d..2230f80d9fe3 100644
--- a/arch/sparc/include/asm/tsb.h
+++ b/arch/sparc/include/asm/tsb.h
@@ -142,98 +142,39 @@ extern struct tsb_phys_patch_entry __tsb_phys_patch, __tsb_phys_patch_end;
 	or		REG1, %lo(swapper_pg_dir), REG1; \
 	sllx		VADDR, 64 - (PGDIR_SHIFT + PGDIR_BITS), REG2; \
 	srlx		REG2, 64 - PAGE_SHIFT, REG2; \
-	andn		REG2, 0x3, REG2; \
-	lduw		[REG1 + REG2], REG1; \
+	andn		REG2, 0x7, REG2; \
+	ldx		[REG1 + REG2], REG1; \
 	brz,pn		REG1, FAIL_LABEL; \
 	 sllx		VADDR, 64 - (PMD_SHIFT + PMD_BITS), REG2; \
 	srlx		REG2, 64 - PAGE_SHIFT, REG2; \
-	sllx		REG1, PGD_PADDR_SHIFT, REG1; \
-	andn		REG2, 0x3, REG2; \
-	lduwa		[REG1 + REG2] ASI_PHYS_USE_EC, REG1; \
+	andn		REG2, 0x7, REG2; \
+	ldxa		[REG1 + REG2] ASI_PHYS_USE_EC, REG1; \
 	brz,pn		REG1, FAIL_LABEL; \
 	 sllx		VADDR, 64 - PMD_SHIFT, REG2; \
-	srlx		REG2, 64 - (PAGE_SHIFT - 1), REG2; \
-	sllx		REG1, PMD_PADDR_SHIFT, REG1; \
+	srlx		REG2, 64 - PAGE_SHIFT, REG2; \
 	andn		REG2, 0x7, REG2; \
 	add		REG1, REG2, REG1;
 
-	/* These macros exists only to make the PMD translator below
-	 * easier to read.  It hides the ELF section switch for the
-	 * sun4v code patching.
-	 */
-#define OR_PTE_BIT_1INSN(REG, NAME)			\
-661:	or		REG, _PAGE_##NAME##_4U, REG;	\
-	.section	.sun4v_1insn_patch, "ax";	\
-	.word		661b;				\
-	or		REG, _PAGE_##NAME##_4V, REG;	\
-	.previous;
-
-#define OR_PTE_BIT_2INSN(REG, TMP, NAME)		\
-661:	sethi		%hi(_PAGE_##NAME##_4U), TMP;	\
-	or		REG, TMP, REG;			\
-	.section	.sun4v_2insn_patch, "ax";	\
-	.word		661b;				\
-	mov		-1, TMP;			\
-	or		REG, _PAGE_##NAME##_4V, REG;	\
-	.previous;
-
-	/* Load into REG the PTE value for VALID, CACHE, and SZHUGE.  */
-#define BUILD_PTE_VALID_SZHUGE_CACHE(REG)				   \
-661:	sethi		%uhi(_PAGE_VALID|_PAGE_SZHUGE_4U), REG;		   \
-	.section	.sun4v_1insn_patch, "ax";			   \
-	.word		661b;						   \
-	sethi		%uhi(_PAGE_VALID), REG;				   \
-	.previous;							   \
-	sllx		REG, 32, REG;					   \
-661:	or		REG, _PAGE_CP_4U|_PAGE_CV_4U, REG;		   \
-	.section	.sun4v_1insn_patch, "ax";			   \
-	.word		661b;						   \
-	or		REG, _PAGE_CP_4V|_PAGE_CV_4V|_PAGE_SZHUGE_4V, REG; \
-	.previous;
-
 	/* PMD has been loaded into REG1, interpret the value, seeing
 	 * if it is a HUGE PMD or a normal one.  If it is not valid
 	 * then jump to FAIL_LABEL.  If it is a HUGE PMD, and it
 	 * translates to a valid PTE, branch to PTE_LABEL.
 	 *
-	 * We translate the PMD by hand, one bit at a time,
-	 * constructing the huge PTE.
-	 *
-	 * So we construct the PTE in REG2 as follows:
-	 *
-	 * 1) Extract the PMD PFN from REG1 and place it into REG2.
-	 *
-	 * 2) Translate PMD protection bits in REG1 into REG2, one bit
-	 *    at a time using andcc tests on REG1 and OR's into REG2.
-	 *
-	 *    Only two bits to be concerned with here, EXEC and WRITE.
-	 *    Now REG1 is freed up and we can use it as a temporary.
-	 *
-	 * 3) Construct the VALID, CACHE, and page size PTE bits in
-	 *    REG1, OR with REG2 to form final PTE.
+	 * We have to propagate the 4MB bit of the virtual address
+	 * because we are fabricating 8MB pages using 4MB hw pages.
 	 */
 #ifdef CONFIG_TRANSPARENT_HUGEPAGE
 #define USER_PGTABLE_CHECK_PMD_HUGE(VADDR, REG1, REG2, FAIL_LABEL, PTE_LABEL) \
-	brz,pn		REG1, FAIL_LABEL;				      \
-	 andcc		REG1, PMD_ISHUGE, %g0;				      \
-	be,pt		%xcc, 700f;					      \
-	 and		REG1, PMD_HUGE_PRESENT|PMD_HUGE_ACCESSED, REG2;	      \
-	cmp		REG2, PMD_HUGE_PRESENT|PMD_HUGE_ACCESSED;	      \
-	bne,pn		%xcc, FAIL_LABEL;				      \
-	 andn		REG1, PMD_HUGE_PROTBITS, REG2;			      \
-	sllx		REG2, PMD_PADDR_SHIFT, REG2;			      \
-	/* REG2 now holds PFN << PAGE_SHIFT */				      \
-	andcc		REG1, PMD_HUGE_WRITE, %g0;			      \
-	bne,a,pt	%xcc, 1f;					      \
-	 OR_PTE_BIT_1INSN(REG2, W);					      \
-1:	andcc		REG1, PMD_HUGE_EXEC, %g0;			      \
-	be,pt		%xcc, 1f;					      \
-	 nop;								      \
-	OR_PTE_BIT_2INSN(REG2, REG1, EXEC);				      \
-	/* REG1 can now be clobbered, build final PTE */		      \
-1:	BUILD_PTE_VALID_SZHUGE_CACHE(REG1);				      \
-	ba,pt		%xcc, PTE_LABEL;				      \
-	 or		REG1, REG2, REG1;				      \
+	brz,pn		REG1, FAIL_LABEL;		\
+	 sethi		%uhi(_PAGE_PMD_HUGE), REG2;	\
+	sllx		REG2, 32, REG2;			\
+	andcc		REG1, REG2, %g0;		\
+	be,pt		%xcc, 700f;			\
+	 sethi		%hi(4 * 1024 * 1024), REG2;	\
+	andn		REG1, REG2, REG1;		\
+	and		VADDR, REG2, REG2;		\
+	brlz,pt		REG1, PTE_LABEL;		\
+	 or		REG1, REG2, REG1;		\
 700:
 #else
 #define USER_PGTABLE_CHECK_PMD_HUGE(VADDR, REG1, REG2, FAIL_LABEL, PTE_LABEL) \
@@ -253,18 +194,16 @@ extern struct tsb_phys_patch_entry __tsb_phys_patch, __tsb_phys_patch_end;
 #define USER_PGTABLE_WALK_TL1(VADDR, PHYS_PGD, REG1, REG2, FAIL_LABEL)	\
 	sllx		VADDR, 64 - (PGDIR_SHIFT + PGDIR_BITS), REG2; \
 	srlx		REG2, 64 - PAGE_SHIFT, REG2; \
-	andn		REG2, 0x3, REG2; \
-	lduwa		[PHYS_PGD + REG2] ASI_PHYS_USE_EC, REG1; \
+	andn		REG2, 0x7, REG2; \
+	ldxa		[PHYS_PGD + REG2] ASI_PHYS_USE_EC, REG1; \
 	brz,pn		REG1, FAIL_LABEL; \
 	 sllx		VADDR, 64 - (PMD_SHIFT + PMD_BITS), REG2; \
 	srlx		REG2, 64 - PAGE_SHIFT, REG2; \
-	sllx		REG1, PGD_PADDR_SHIFT, REG1; \
-	andn		REG2, 0x3, REG2; \
-	lduwa		[REG1 + REG2] ASI_PHYS_USE_EC, REG1; \
+	andn		REG2, 0x7, REG2; \
+	ldxa		[REG1 + REG2] ASI_PHYS_USE_EC, REG1; \
 	USER_PGTABLE_CHECK_PMD_HUGE(VADDR, REG1, REG2, FAIL_LABEL, 800f) \
 	sllx		VADDR, 64 - PMD_SHIFT, REG2; \
-	srlx		REG2, 64 - (PAGE_SHIFT - 1), REG2; \
-	sllx		REG1, PMD_PADDR_SHIFT, REG1; \
+	srlx		REG2, 64 - PAGE_SHIFT, REG2; \
 	andn		REG2, 0x7, REG2; \
 	add		REG1, REG2, REG1; \
 	ldxa		[REG1] ASI_PHYS_USE_EC, REG1; \
diff --git a/arch/sparc/kernel/entry.h b/arch/sparc/kernel/entry.h
index 9c179fbfb219..140966fbd303 100644
--- a/arch/sparc/kernel/entry.h
+++ b/arch/sparc/kernel/entry.h
@@ -88,7 +88,6 @@ extern asmlinkage void syscall_trace_leave(struct pt_regs *regs);
 
 extern void bad_trap_tl1(struct pt_regs *regs, long lvl);
 
-extern void do_fpe_common(struct pt_regs *regs);
 extern void do_fpieee(struct pt_regs *regs);
 extern void do_fpother(struct pt_regs *regs);
 extern void do_tof(struct pt_regs *regs);
diff --git a/arch/sparc/kernel/kgdb_64.c b/arch/sparc/kernel/kgdb_64.c
index 53c0a82e6030..60b19f50c80a 100644
--- a/arch/sparc/kernel/kgdb_64.c
+++ b/arch/sparc/kernel/kgdb_64.c
@@ -159,11 +159,12 @@ int kgdb_arch_handle_exception(int e_vector, int signo, int err_code,
 
 asmlinkage void kgdb_trap(unsigned long trap_level, struct pt_regs *regs)
 {
+	enum ctx_state prev_state = exception_enter();
 	unsigned long flags;
 
 	if (user_mode(regs)) {
 		bad_trap(regs, trap_level);
-		return;
+		goto out;
 	}
 
 	flushw_all();
@@ -171,6 +172,8 @@ asmlinkage void kgdb_trap(unsigned long trap_level, struct pt_regs *regs)
 	local_irq_save(flags);
 	kgdb_handle_exception(0x172, SIGTRAP, 0, regs);
 	local_irq_restore(flags);
+out:
+	exception_exit(prev_state);
 }
 
 int kgdb_arch_init(void)
diff --git a/arch/sparc/kernel/kprobes.c b/arch/sparc/kernel/kprobes.c
index e72212148d2a..1b0973503197 100644
--- a/arch/sparc/kernel/kprobes.c
+++ b/arch/sparc/kernel/kprobes.c
@@ -8,6 +8,7 @@
 #include <linux/module.h>
 #include <linux/kdebug.h>
 #include <linux/slab.h>
+#include <linux/context_tracking.h>
 #include <asm/signal.h>
 #include <asm/cacheflush.h>
 #include <asm/uaccess.h>
@@ -349,7 +350,7 @@ int __kprobes kprobe_fault_handler(struct pt_regs *regs, int trapnr)
 	case KPROBE_HIT_SSDONE:
 		/*
 		 * We increment the nmissed count for accounting,
-		 * we can also use npre/npostfault count for accouting
+		 * we can also use npre/npostfault count for accounting
 		 * these specific fault cases.
 		 */
 		kprobes_inc_nmissed_count(cur);
@@ -418,12 +419,14 @@ int __kprobes kprobe_exceptions_notify(struct notifier_block *self,
 asmlinkage void __kprobes kprobe_trap(unsigned long trap_level,
 				      struct pt_regs *regs)
 {
+	enum ctx_state prev_state = exception_enter();
+
 	BUG_ON(trap_level != 0x170 && trap_level != 0x171);
 
 	if (user_mode(regs)) {
 		local_irq_enable();
 		bad_trap(regs, trap_level);
-		return;
+		goto out;
 	}
 
 	/* trap_level == 0x170 --> ta 0x70
@@ -433,6 +436,8 @@ asmlinkage void __kprobes kprobe_trap(unsigned long trap_level,
 		       (trap_level == 0x170) ? "debug" : "debug_2",
 		       regs, 0, trap_level, SIGTRAP) != NOTIFY_STOP)
 		bad_trap(regs, trap_level);
+out:
+	exception_exit(prev_state);
 }
 
 /* Jprobes support.  */
diff --git a/arch/sparc/kernel/ktlb.S b/arch/sparc/kernel/ktlb.S
index fde5a419cf27..542e96ac4d39 100644
--- a/arch/sparc/kernel/ktlb.S
+++ b/arch/sparc/kernel/ktlb.S
@@ -153,12 +153,19 @@ kvmap_dtlb_tsb4m_miss:
 	/* Clear the PAGE_OFFSET top virtual bits, shift
 	 * down to get PFN, and make sure PFN is in range.
 	 */
-	sllx		%g4, 21, %g5
+661:	sllx		%g4, 0, %g5
+	.section	.page_offset_shift_patch, "ax"
+	.word		661b
+	.previous
 
 	/* Check to see if we know about valid memory at the 4MB
 	 * chunk this physical address will reside within.
 	 */
-	srlx		%g5, 21 + 41, %g2
+661:	srlx		%g5, MAX_PHYS_ADDRESS_BITS, %g2
+	.section	.page_offset_shift_patch, "ax"
+	.word		661b
+	.previous
+
 	brnz,pn		%g2, kvmap_dtlb_longpath
 	 nop
 
@@ -176,7 +183,11 @@ valid_addr_bitmap_patch:
 	or		%g7, %lo(sparc64_valid_addr_bitmap), %g7
 	.previous
 
-	srlx		%g5, 21 + 22, %g2
+661:	srlx		%g5, ILOG2_4MB, %g2
+	.section	.page_offset_shift_patch, "ax"
+	.word		661b
+	.previous
+
 	srlx		%g2, 6, %g5
 	and		%g2, 63, %g2
 	sllx		%g5, 3, %g5
@@ -189,9 +200,18 @@ valid_addr_bitmap_patch:
 2:	 sethi		%hi(kpte_linear_bitmap), %g2
 
 	/* Get the 256MB physical address index. */
-	sllx		%g4, 21, %g5
+661:	sllx		%g4, 0, %g5
+	.section	.page_offset_shift_patch, "ax"
+	.word		661b
+	.previous
+
 	or		%g2, %lo(kpte_linear_bitmap), %g2
-	srlx		%g5, 21 + 28, %g5
+
+661:	srlx		%g5, ILOG2_256MB, %g5
+	.section	.page_offset_shift_patch, "ax"
+	.word		661b
+	.previous
+
 	and		%g5, (32 - 1), %g7
 
 	/* Divide by 32 to get the offset into the bitmask.  */
diff --git a/arch/sparc/kernel/pci.c b/arch/sparc/kernel/pci.c
index bc4d3f5d2e5d..cb021453de2a 100644
--- a/arch/sparc/kernel/pci.c
+++ b/arch/sparc/kernel/pci.c
@@ -398,8 +398,8 @@ static void apb_fake_ranges(struct pci_dev *dev,
 	apb_calc_first_last(map, &first, &last);
 	res = bus->resource[1];
 	res->flags = IORESOURCE_MEM;
-	region.start = (first << 21);
-	region.end = (last << 21) + ((1 << 21) - 1);
+	region.start = (first << 29);
+	region.end = (last << 29) + ((1 << 29) - 1);
 	pcibios_bus_to_resource(dev, res, &region);
 }
 
diff --git a/arch/sparc/kernel/process_64.c b/arch/sparc/kernel/process_64.c
index baebab215492..32a280ec38c1 100644
--- a/arch/sparc/kernel/process_64.c
+++ b/arch/sparc/kernel/process_64.c
@@ -31,6 +31,7 @@
 #include <linux/elfcore.h>
 #include <linux/sysrq.h>
 #include <linux/nmi.h>
+#include <linux/context_tracking.h>
 
 #include <asm/uaccess.h>
 #include <asm/page.h>
@@ -557,6 +558,7 @@ void fault_in_user_windows(void)
 
 barf:
 	set_thread_wsaved(window + 1);
+	user_exit();
 	do_exit(SIGILL);
 }
 
diff --git a/arch/sparc/kernel/ptrace_64.c b/arch/sparc/kernel/ptrace_64.c
index 773c1f2983ce..c13c9f25d83a 100644
--- a/arch/sparc/kernel/ptrace_64.c
+++ b/arch/sparc/kernel/ptrace_64.c
@@ -27,6 +27,7 @@
 #include <trace/syscall.h>
 #include <linux/compat.h>
 #include <linux/elf.h>
+#include <linux/context_tracking.h>
 
 #include <asm/asi.h>
 #include <asm/pgtable.h>
@@ -1066,6 +1067,9 @@ asmlinkage int syscall_trace_enter(struct pt_regs *regs)
 	/* do the secure computing check first */
 	secure_computing_strict(regs->u_regs[UREG_G1]);
 
+	if (test_thread_flag(TIF_NOHZ))
+		user_exit();
+
 	if (test_thread_flag(TIF_SYSCALL_TRACE))
 		ret = tracehook_report_syscall_entry(regs);
 
@@ -1086,6 +1090,9 @@ asmlinkage int syscall_trace_enter(struct pt_regs *regs)
 
 asmlinkage void syscall_trace_leave(struct pt_regs *regs)
 {
+	if (test_thread_flag(TIF_NOHZ))
+		user_exit();
+
 	audit_syscall_exit(regs);
 
 	if (unlikely(test_thread_flag(TIF_SYSCALL_TRACEPOINT)))
@@ -1093,4 +1100,7 @@ asmlinkage void syscall_trace_leave(struct pt_regs *regs)
 
 	if (test_thread_flag(TIF_SYSCALL_TRACE))
 		tracehook_report_syscall_exit(regs, 0);
+
+	if (test_thread_flag(TIF_NOHZ))
+		user_enter();
 }
diff --git a/arch/sparc/kernel/rtrap_64.S b/arch/sparc/kernel/rtrap_64.S
index afa2a9e3d0a0..39f0c662f4c8 100644
--- a/arch/sparc/kernel/rtrap_64.S
+++ b/arch/sparc/kernel/rtrap_64.S
@@ -18,10 +18,16 @@
 #define		RTRAP_PSTATE_IRQOFF	(PSTATE_TSO|PSTATE_PEF|PSTATE_PRIV)
 #define		RTRAP_PSTATE_AG_IRQOFF	(PSTATE_TSO|PSTATE_PEF|PSTATE_PRIV|PSTATE_AG)
 
+#ifdef CONFIG_CONTEXT_TRACKING
+# define SCHEDULE_USER schedule_user
+#else
+# define SCHEDULE_USER schedule
+#endif
+
 		.text
 		.align			32
 __handle_preemption:
-		call			schedule
+		call			SCHEDULE_USER
 		 wrpr			%g0, RTRAP_PSTATE, %pstate
 		ba,pt			%xcc, __handle_preemption_continue
 		 wrpr			%g0, RTRAP_PSTATE_IRQOFF, %pstate
@@ -306,12 +312,10 @@ to_kernel:
 		 nop
 		cmp			%l4, 0
 		bne,pn			%xcc, kern_fpucheck
-		 sethi			%hi(PREEMPT_ACTIVE), %l6
-		stw			%l6, [%g6 + TI_PRE_COUNT]
-		call			schedule
+		 nop
+		call			preempt_schedule_irq
 		 nop
 		ba,pt			%xcc, rtrap
-		 stw			%g0, [%g6 + TI_PRE_COUNT]
 #endif
 kern_fpucheck:	ldub			[%g6 + TI_FPDEPTH], %l5
 		brz,pt			%l5, rt_continue
diff --git a/arch/sparc/kernel/signal_64.c b/arch/sparc/kernel/signal_64.c
index 35923e8abd82..cd91d010e6d3 100644
--- a/arch/sparc/kernel/signal_64.c
+++ b/arch/sparc/kernel/signal_64.c
@@ -23,6 +23,7 @@
 #include <linux/tty.h>
 #include <linux/binfmts.h>
 #include <linux/bitops.h>
+#include <linux/context_tracking.h>
 
 #include <asm/uaccess.h>
 #include <asm/ptrace.h>
@@ -43,6 +44,7 @@ asmlinkage void sparc64_set_context(struct pt_regs *regs)
 {
 	struct ucontext __user *ucp = (struct ucontext __user *)
 		regs->u_regs[UREG_I0];
+	enum ctx_state prev_state = exception_enter();
 	mc_gregset_t __user *grp;
 	unsigned long pc, npc, tstate;
 	unsigned long fp, i7;
@@ -129,16 +131,19 @@ asmlinkage void sparc64_set_context(struct pt_regs *regs)
 	}
 	if (err)
 		goto do_sigsegv;
-
+out:
+	exception_exit(prev_state);
 	return;
 do_sigsegv:
 	force_sig(SIGSEGV, current);
+	goto out;
 }
 
 asmlinkage void sparc64_get_context(struct pt_regs *regs)
 {
 	struct ucontext __user *ucp = (struct ucontext __user *)
 		regs->u_regs[UREG_I0];
+	enum ctx_state prev_state = exception_enter();
 	mc_gregset_t __user *grp;
 	mcontext_t __user *mcp;
 	unsigned long fp, i7;
@@ -220,10 +225,12 @@ asmlinkage void sparc64_get_context(struct pt_regs *regs)
 	}
 	if (err)
 		goto do_sigsegv;
-
+out:
+	exception_exit(prev_state);
 	return;
 do_sigsegv:
 	force_sig(SIGSEGV, current);
+	goto out;
 }
 
 struct rt_signal_frame {
@@ -528,11 +535,13 @@ static void do_signal(struct pt_regs *regs, unsigned long orig_i0)
 
 void do_notify_resume(struct pt_regs *regs, unsigned long orig_i0, unsigned long thread_info_flags)
 {
+	user_exit();
 	if (thread_info_flags & _TIF_SIGPENDING)
 		do_signal(regs, orig_i0);
 	if (thread_info_flags & _TIF_NOTIFY_RESUME) {
 		clear_thread_flag(TIF_NOTIFY_RESUME);
 		tracehook_notify_resume(regs);
 	}
+	user_enter();
 }
 
diff --git a/arch/sparc/kernel/smp_64.c b/arch/sparc/kernel/smp_64.c
index e142545244f2..b66a5338231e 100644
--- a/arch/sparc/kernel/smp_64.c
+++ b/arch/sparc/kernel/smp_64.c
@@ -1399,8 +1399,13 @@ void __init smp_cpus_done(unsigned int max_cpus)
 
 void smp_send_reschedule(int cpu)
 {
-	xcall_deliver((u64) &xcall_receive_signal, 0, 0,
-		      cpumask_of(cpu));
+	if (cpu == smp_processor_id()) {
+		WARN_ON_ONCE(preemptible());
+		set_softint(1 << PIL_SMP_RECEIVE_SIGNAL);
+	} else {
+		xcall_deliver((u64) &xcall_receive_signal,
+			      0, 0, cpumask_of(cpu));
+	}
 }
 
 void __irq_entry smp_receive_signal_client(int irq, struct pt_regs *regs)
diff --git a/arch/sparc/kernel/sun4v_tlb_miss.S b/arch/sparc/kernel/sun4v_tlb_miss.S
index bde867fd71e8..e0c09bf85610 100644
--- a/arch/sparc/kernel/sun4v_tlb_miss.S
+++ b/arch/sparc/kernel/sun4v_tlb_miss.S
@@ -182,7 +182,7 @@ sun4v_tsb_miss_common:
 	cmp	%g5, -1
 	be,pt	%xcc, 80f
 	 nop
-	COMPUTE_TSB_PTR(%g5, %g4, HPAGE_SHIFT, %g2, %g7)
+	COMPUTE_TSB_PTR(%g5, %g4, REAL_HPAGE_SHIFT, %g2, %g7)
 
 	/* That clobbered %g2, reload it.  */
 	ldxa	[%g0] ASI_SCRATCHPAD, %g2
diff --git a/arch/sparc/kernel/sys_sparc_64.c b/arch/sparc/kernel/sys_sparc_64.c
index 51561b8b15ba..beb0b5a5f21f 100644
--- a/arch/sparc/kernel/sys_sparc_64.c
+++ b/arch/sparc/kernel/sys_sparc_64.c
@@ -24,6 +24,7 @@
 #include <linux/personality.h>
 #include <linux/random.h>
 #include <linux/export.h>
+#include <linux/context_tracking.h>
 
 #include <asm/uaccess.h>
 #include <asm/utrap.h>
@@ -39,9 +40,6 @@ asmlinkage unsigned long sys_getpagesize(void)
 	return PAGE_SIZE;
 }
 
-#define VA_EXCLUDE_START (0x0000080000000000UL - (1UL << 32UL))
-#define VA_EXCLUDE_END   (0xfffff80000000000UL + (1UL << 32UL))
-
 /* Does addr --> addr+len fall within 4GB of the VA-space hole or
  * overflow past the end of the 64-bit address space?
  */
@@ -499,6 +497,7 @@ asmlinkage unsigned long c_sys_nis_syscall(struct pt_regs *regs)
 
 asmlinkage void sparc_breakpoint(struct pt_regs *regs)
 {
+	enum ctx_state prev_state = exception_enter();
 	siginfo_t info;
 
 	if (test_thread_flag(TIF_32BIT)) {
@@ -517,6 +516,7 @@ asmlinkage void sparc_breakpoint(struct pt_regs *regs)
 #ifdef DEBUG_SPARC_BREAKPOINT
 	printk ("TRAP: Returning to space: PC=%lx nPC=%lx\n", regs->tpc, regs->tnpc);
 #endif
+	exception_exit(prev_state);
 }
 
 extern void check_pending(int signum);
diff --git a/arch/sparc/kernel/syscalls.S b/arch/sparc/kernel/syscalls.S
index d950197a17e1..87729fff13b9 100644
--- a/arch/sparc/kernel/syscalls.S
+++ b/arch/sparc/kernel/syscalls.S
@@ -52,7 +52,7 @@ sys32_rt_sigreturn:
 #endif
 	.align	32
 1:	ldx	[%g6 + TI_FLAGS], %l5
-	andcc	%l5, (_TIF_SYSCALL_TRACE|_TIF_SECCOMP|_TIF_SYSCALL_AUDIT|_TIF_SYSCALL_TRACEPOINT), %g0
+	andcc	%l5, (_TIF_SYSCALL_TRACE|_TIF_SECCOMP|_TIF_SYSCALL_AUDIT|_TIF_SYSCALL_TRACEPOINT|_TIF_NOHZ), %g0
 	be,pt	%icc, rtrap
 	 nop
 	call	syscall_trace_leave
@@ -184,7 +184,7 @@ linux_sparc_syscall32:
 
 	srl	%i3, 0, %o3				! IEU0
 	srl	%i2, 0, %o2				! IEU0	Group
-	andcc	%l0, (_TIF_SYSCALL_TRACE|_TIF_SECCOMP|_TIF_SYSCALL_AUDIT|_TIF_SYSCALL_TRACEPOINT), %g0
+	andcc	%l0, (_TIF_SYSCALL_TRACE|_TIF_SECCOMP|_TIF_SYSCALL_AUDIT|_TIF_SYSCALL_TRACEPOINT|_TIF_NOHZ), %g0
 	bne,pn	%icc, linux_syscall_trace32		! CTI
 	 mov	%i0, %l5				! IEU1
 5:	call	%l7					! CTI	Group brk forced
@@ -207,7 +207,7 @@ linux_sparc_syscall:
 
 	mov	%i3, %o3				! IEU1
 	mov	%i4, %o4				! IEU0	Group
-	andcc	%l0, (_TIF_SYSCALL_TRACE|_TIF_SECCOMP|_TIF_SYSCALL_AUDIT|_TIF_SYSCALL_TRACEPOINT), %g0
+	andcc	%l0, (_TIF_SYSCALL_TRACE|_TIF_SECCOMP|_TIF_SYSCALL_AUDIT|_TIF_SYSCALL_TRACEPOINT|_TIF_NOHZ), %g0
 	bne,pn	%icc, linux_syscall_trace		! CTI	Group
 	 mov	%i0, %l5				! IEU0
 2:	call	%l7					! CTI	Group brk forced
@@ -223,7 +223,7 @@ ret_sys_call:
 
 	cmp	%o0, -ERESTART_RESTARTBLOCK
 	bgeu,pn	%xcc, 1f
-	 andcc	%l0, (_TIF_SYSCALL_TRACE|_TIF_SECCOMP|_TIF_SYSCALL_AUDIT|_TIF_SYSCALL_TRACEPOINT), %g0
+	 andcc	%l0, (_TIF_SYSCALL_TRACE|_TIF_SECCOMP|_TIF_SYSCALL_AUDIT|_TIF_SYSCALL_TRACEPOINT|_TIF_NOHZ), %g0
 	ldx	[%sp + PTREGS_OFF + PT_V9_TNPC], %l1 ! pc = npc
 
 2:
diff --git a/arch/sparc/kernel/traps_64.c b/arch/sparc/kernel/traps_64.c
index b3f833ab90eb..4ced92f05358 100644
--- a/arch/sparc/kernel/traps_64.c
+++ b/arch/sparc/kernel/traps_64.c
@@ -20,6 +20,7 @@
 #include <linux/ftrace.h>
 #include <linux/reboot.h>
 #include <linux/gfp.h>
+#include <linux/context_tracking.h>
 
 #include <asm/smp.h>
 #include <asm/delay.h>
@@ -186,11 +187,12 @@ EXPORT_SYMBOL_GPL(unregister_dimm_printer);
 
 void spitfire_insn_access_exception(struct pt_regs *regs, unsigned long sfsr, unsigned long sfar)
 {
+	enum ctx_state prev_state = exception_enter();
 	siginfo_t info;
 
 	if (notify_die(DIE_TRAP, "instruction access exception", regs,
 		       0, 0x8, SIGTRAP) == NOTIFY_STOP)
-		return;
+		goto out;
 
 	if (regs->tstate & TSTATE_PRIV) {
 		printk("spitfire_insn_access_exception: SFSR[%016lx] "
@@ -207,6 +209,8 @@ void spitfire_insn_access_exception(struct pt_regs *regs, unsigned long sfsr, un
 	info.si_addr = (void __user *)regs->tpc;
 	info.si_trapno = 0;
 	force_sig_info(SIGSEGV, &info, current);
+out:
+	exception_exit(prev_state);
 }
 
 void spitfire_insn_access_exception_tl1(struct pt_regs *regs, unsigned long sfsr, unsigned long sfar)
@@ -260,11 +264,12 @@ void sun4v_insn_access_exception_tl1(struct pt_regs *regs, unsigned long addr, u
 
 void spitfire_data_access_exception(struct pt_regs *regs, unsigned long sfsr, unsigned long sfar)
 {
+	enum ctx_state prev_state = exception_enter();
 	siginfo_t info;
 
 	if (notify_die(DIE_TRAP, "data access exception", regs,
 		       0, 0x30, SIGTRAP) == NOTIFY_STOP)
-		return;
+		goto out;
 
 	if (regs->tstate & TSTATE_PRIV) {
 		/* Test if this comes from uaccess places. */
@@ -280,7 +285,7 @@ void spitfire_data_access_exception(struct pt_regs *regs, unsigned long sfsr, un
 #endif
 			regs->tpc = entry->fixup;
 			regs->tnpc = regs->tpc + 4;
-			return;
+			goto out;
 		}
 		/* Shit... */
 		printk("spitfire_data_access_exception: SFSR[%016lx] "
@@ -294,6 +299,8 @@ void spitfire_data_access_exception(struct pt_regs *regs, unsigned long sfsr, un
 	info.si_addr = (void __user *)sfar;
 	info.si_trapno = 0;
 	force_sig_info(SIGSEGV, &info, current);
+out:
+	exception_exit(prev_state);
 }
 
 void spitfire_data_access_exception_tl1(struct pt_regs *regs, unsigned long sfsr, unsigned long sfar)
@@ -1994,6 +2001,7 @@ static void sun4v_log_error(struct pt_regs *regs, struct sun4v_error_entry *ent,
  */
 void sun4v_resum_error(struct pt_regs *regs, unsigned long offset)
 {
+	enum ctx_state prev_state = exception_enter();
 	struct sun4v_error_entry *ent, local_copy;
 	struct trap_per_cpu *tb;
 	unsigned long paddr;
@@ -2022,12 +2030,14 @@ void sun4v_resum_error(struct pt_regs *regs, unsigned long offset)
 		pr_info("Shutdown request, %u seconds...\n",
 			local_copy.err_secs);
 		orderly_poweroff(true);
-		return;
+		goto out;
 	}
 
 	sun4v_log_error(regs, &local_copy, cpu,
 			KERN_ERR "RESUMABLE ERROR",
 			&sun4v_resum_oflow_cnt);
+out:
+	exception_exit(prev_state);
 }
 
 /* If we try to printk() we'll probably make matters worse, by trying
@@ -2152,7 +2162,7 @@ void hypervisor_tlbop_error_xcall(unsigned long err, unsigned long op)
 	       err, op);
 }
 
-void do_fpe_common(struct pt_regs *regs)
+static void do_fpe_common(struct pt_regs *regs)
 {
 	if (regs->tstate & TSTATE_PRIV) {
 		regs->tpc = regs->tnpc;
@@ -2188,23 +2198,28 @@ void do_fpe_common(struct pt_regs *regs)
 
 void do_fpieee(struct pt_regs *regs)
 {
+	enum ctx_state prev_state = exception_enter();
+
 	if (notify_die(DIE_TRAP, "fpu exception ieee", regs,
 		       0, 0x24, SIGFPE) == NOTIFY_STOP)
-		return;
+		goto out;
 
 	do_fpe_common(regs);
+out:
+	exception_exit(prev_state);
 }
 
 extern int do_mathemu(struct pt_regs *, struct fpustate *, bool);
 
 void do_fpother(struct pt_regs *regs)
 {
+	enum ctx_state prev_state = exception_enter();
 	struct fpustate *f = FPUSTATE;
 	int ret = 0;
 
 	if (notify_die(DIE_TRAP, "fpu exception other", regs,
 		       0, 0x25, SIGFPE) == NOTIFY_STOP)
-		return;
+		goto out;
 
 	switch ((current_thread_info()->xfsr[0] & 0x1c000)) {
 	case (2 << 14): /* unfinished_FPop */
@@ -2213,17 +2228,20 @@ void do_fpother(struct pt_regs *regs)
 		break;
 	}
 	if (ret)
-		return;
+		goto out;
 	do_fpe_common(regs);
+out:
+	exception_exit(prev_state);
 }
 
 void do_tof(struct pt_regs *regs)
 {
+	enum ctx_state prev_state = exception_enter();
 	siginfo_t info;
 
 	if (notify_die(DIE_TRAP, "tagged arithmetic overflow", regs,
 		       0, 0x26, SIGEMT) == NOTIFY_STOP)
-		return;
+		goto out;
 
 	if (regs->tstate & TSTATE_PRIV)
 		die_if_kernel("Penguin overflow trap from kernel mode", regs);
@@ -2237,15 +2255,18 @@ void do_tof(struct pt_regs *regs)
 	info.si_addr = (void __user *)regs->tpc;
 	info.si_trapno = 0;
 	force_sig_info(SIGEMT, &info, current);
+out:
+	exception_exit(prev_state);
 }
 
 void do_div0(struct pt_regs *regs)
 {
+	enum ctx_state prev_state = exception_enter();
 	siginfo_t info;
 
 	if (notify_die(DIE_TRAP, "integer division by zero", regs,
 		       0, 0x28, SIGFPE) == NOTIFY_STOP)
-		return;
+		goto out;
 
 	if (regs->tstate & TSTATE_PRIV)
 		die_if_kernel("TL0: Kernel divide by zero.", regs);
@@ -2259,6 +2280,8 @@ void do_div0(struct pt_regs *regs)
 	info.si_addr = (void __user *)regs->tpc;
 	info.si_trapno = 0;
 	force_sig_info(SIGFPE, &info, current);
+out:
+	exception_exit(prev_state);
 }
 
 static void instruction_dump(unsigned int *pc)
@@ -2415,6 +2438,7 @@ extern int handle_ldf_stq(u32 insn, struct pt_regs *regs);
 
 void do_illegal_instruction(struct pt_regs *regs)
 {
+	enum ctx_state prev_state = exception_enter();
 	unsigned long pc = regs->tpc;
 	unsigned long tstate = regs->tstate;
 	u32 insn;
@@ -2422,7 +2446,7 @@ void do_illegal_instruction(struct pt_regs *regs)
 
 	if (notify_die(DIE_TRAP, "illegal instruction", regs,
 		       0, 0x10, SIGILL) == NOTIFY_STOP)
-		return;
+		goto out;
 
 	if (tstate & TSTATE_PRIV)
 		die_if_kernel("Kernel illegal instruction", regs);
@@ -2431,14 +2455,14 @@ void do_illegal_instruction(struct pt_regs *regs)
 	if (get_user(insn, (u32 __user *) pc) != -EFAULT) {
 		if ((insn & 0xc1ffc000) == 0x81700000) /* POPC */ {
 			if (handle_popc(insn, regs))
-				return;
+				goto out;
 		} else if ((insn & 0xc1580000) == 0xc1100000) /* LDQ/STQ */ {
 			if (handle_ldf_stq(insn, regs))
-				return;
+				goto out;
 		} else if (tlb_type == hypervisor) {
 			if ((insn & VIS_OPCODE_MASK) == VIS_OPCODE_VAL) {
 				if (!vis_emul(regs, insn))
-					return;
+					goto out;
 			} else {
 				struct fpustate *f = FPUSTATE;
 
@@ -2448,7 +2472,7 @@ void do_illegal_instruction(struct pt_regs *regs)
 				 * Trap in the %fsr to unimplemented_FPop.
 				 */
 				if (do_mathemu(regs, f, true))
-					return;
+					goto out;
 			}
 		}
 	}
@@ -2458,21 +2482,24 @@ void do_illegal_instruction(struct pt_regs *regs)
 	info.si_addr = (void __user *)pc;
 	info.si_trapno = 0;
 	force_sig_info(SIGILL, &info, current);
+out:
+	exception_exit(prev_state);
 }
 
 extern void kernel_unaligned_trap(struct pt_regs *regs, unsigned int insn);
 
 void mem_address_unaligned(struct pt_regs *regs, unsigned long sfar, unsigned long sfsr)
 {
+	enum ctx_state prev_state = exception_enter();
 	siginfo_t info;
 
 	if (notify_die(DIE_TRAP, "memory address unaligned", regs,
 		       0, 0x34, SIGSEGV) == NOTIFY_STOP)
-		return;
+		goto out;
 
 	if (regs->tstate & TSTATE_PRIV) {
 		kernel_unaligned_trap(regs, *((unsigned int *)regs->tpc));
-		return;
+		goto out;
 	}
 	info.si_signo = SIGBUS;
 	info.si_errno = 0;
@@ -2480,6 +2507,8 @@ void mem_address_unaligned(struct pt_regs *regs, unsigned long sfar, unsigned lo
 	info.si_addr = (void __user *)sfar;
 	info.si_trapno = 0;
 	force_sig_info(SIGBUS, &info, current);
+out:
+	exception_exit(prev_state);
 }
 
 void sun4v_do_mna(struct pt_regs *regs, unsigned long addr, unsigned long type_ctx)
@@ -2504,11 +2533,12 @@ void sun4v_do_mna(struct pt_regs *regs, unsigned long addr, unsigned long type_c
 
 void do_privop(struct pt_regs *regs)
 {
+	enum ctx_state prev_state = exception_enter();
 	siginfo_t info;
 
 	if (notify_die(DIE_TRAP, "privileged operation", regs,
 		       0, 0x11, SIGILL) == NOTIFY_STOP)
-		return;
+		goto out;
 
 	if (test_thread_flag(TIF_32BIT)) {
 		regs->tpc &= 0xffffffff;
@@ -2520,6 +2550,8 @@ void do_privop(struct pt_regs *regs)
 	info.si_addr = (void __user *)regs->tpc;
 	info.si_trapno = 0;
 	force_sig_info(SIGILL, &info, current);
+out:
+	exception_exit(prev_state);
 }
 
 void do_privact(struct pt_regs *regs)
@@ -2530,99 +2562,116 @@ void do_privact(struct pt_regs *regs)
 /* Trap level 1 stuff or other traps we should never see... */
 void do_cee(struct pt_regs *regs)
 {
+	exception_enter();
 	die_if_kernel("TL0: Cache Error Exception", regs);
 }
 
 void do_cee_tl1(struct pt_regs *regs)
 {
+	exception_enter();
 	dump_tl1_traplog((struct tl1_traplog *)(regs + 1));
 	die_if_kernel("TL1: Cache Error Exception", regs);
 }
 
 void do_dae_tl1(struct pt_regs *regs)
 {
+	exception_enter();
 	dump_tl1_traplog((struct tl1_traplog *)(regs + 1));
 	die_if_kernel("TL1: Data Access Exception", regs);
 }
 
 void do_iae_tl1(struct pt_regs *regs)
 {
+	exception_enter();
 	dump_tl1_traplog((struct tl1_traplog *)(regs + 1));
 	die_if_kernel("TL1: Instruction Access Exception", regs);
 }
 
 void do_div0_tl1(struct pt_regs *regs)
 {
+	exception_enter();
 	dump_tl1_traplog((struct tl1_traplog *)(regs + 1));
 	die_if_kernel("TL1: DIV0 Exception", regs);
 }
 
 void do_fpdis_tl1(struct pt_regs *regs)
 {
+	exception_enter();
 	dump_tl1_traplog((struct tl1_traplog *)(regs + 1));
 	die_if_kernel("TL1: FPU Disabled", regs);
 }
 
 void do_fpieee_tl1(struct pt_regs *regs)
 {
+	exception_enter();
 	dump_tl1_traplog((struct tl1_traplog *)(regs + 1));
 	die_if_kernel("TL1: FPU IEEE Exception", regs);
 }
 
 void do_fpother_tl1(struct pt_regs *regs)
 {
+	exception_enter();
 	dump_tl1_traplog((struct tl1_traplog *)(regs + 1));
 	die_if_kernel("TL1: FPU Other Exception", regs);
 }
 
 void do_ill_tl1(struct pt_regs *regs)
 {
+	exception_enter();
 	dump_tl1_traplog((struct tl1_traplog *)(regs + 1));
 	die_if_kernel("TL1: Illegal Instruction Exception", regs);
 }
 
 void do_irq_tl1(struct pt_regs *regs)
 {
+	exception_enter();
 	dump_tl1_traplog((struct tl1_traplog *)(regs + 1));
 	die_if_kernel("TL1: IRQ Exception", regs);
 }
 
 void do_lddfmna_tl1(struct pt_regs *regs)
 {
+	exception_enter();
 	dump_tl1_traplog((struct tl1_traplog *)(regs + 1));
 	die_if_kernel("TL1: LDDF Exception", regs);
 }
 
 void do_stdfmna_tl1(struct pt_regs *regs)
 {
+	exception_enter();
 	dump_tl1_traplog((struct tl1_traplog *)(regs + 1));
 	die_if_kernel("TL1: STDF Exception", regs);
 }
 
 void do_paw(struct pt_regs *regs)
 {
+	exception_enter();
 	die_if_kernel("TL0: Phys Watchpoint Exception", regs);
 }
 
 void do_paw_tl1(struct pt_regs *regs)
 {
+	exception_enter();
 	dump_tl1_traplog((struct tl1_traplog *)(regs + 1));
 	die_if_kernel("TL1: Phys Watchpoint Exception", regs);
 }
 
 void do_vaw(struct pt_regs *regs)
 {
+	exception_enter();
 	die_if_kernel("TL0: Virt Watchpoint Exception", regs);
 }
 
 void do_vaw_tl1(struct pt_regs *regs)
 {
+	exception_enter();
 	dump_tl1_traplog((struct tl1_traplog *)(regs + 1));
 	die_if_kernel("TL1: Virt Watchpoint Exception", regs);
 }
 
 void do_tof_tl1(struct pt_regs *regs)
 {
+	exception_enter();
 	dump_tl1_traplog((struct tl1_traplog *)(regs + 1));
 	die_if_kernel("TL1: Tag Overflow Exception", regs);
 }
diff --git a/arch/sparc/kernel/tsb.S b/arch/sparc/kernel/tsb.S
index a313e4a9399b..14158d40ba76 100644
--- a/arch/sparc/kernel/tsb.S
+++ b/arch/sparc/kernel/tsb.S
@@ -75,7 +75,7 @@ tsb_miss_page_table_walk:
 	mov		512, %g7
 	andn		%g5, 0x7, %g5
 	sllx		%g7, %g6, %g7
-	srlx		%g4, HPAGE_SHIFT, %g6
+	srlx		%g4, REAL_HPAGE_SHIFT, %g6
 	sub		%g7, 1, %g7
 	and		%g6, %g7, %g6
 	sllx		%g6, 4, %g6
diff --git a/arch/sparc/kernel/unaligned_64.c b/arch/sparc/kernel/unaligned_64.c
index 8201c25e7669..3c1a7cb31579 100644
--- a/arch/sparc/kernel/unaligned_64.c
+++ b/arch/sparc/kernel/unaligned_64.c
@@ -21,9 +21,12 @@
 #include <linux/bitops.h>
 #include <linux/perf_event.h>
 #include <linux/ratelimit.h>
+#include <linux/context_tracking.h>
 #include <asm/fpumacro.h>
 #include <asm/cacheflush.h>
 
+#include "entry.h"
+
 enum direction {
 	load,    /* ld, ldd, ldh, ldsh */
 	store,   /* st, std, sth, stsh */
@@ -418,9 +421,6 @@ int handle_popc(u32 insn, struct pt_regs *regs)
 
 extern void do_fpother(struct pt_regs *regs);
 extern void do_privact(struct pt_regs *regs);
-extern void spitfire_data_access_exception(struct pt_regs *regs,
-					   unsigned long sfsr,
-					   unsigned long sfar);
 extern void sun4v_data_access_exception(struct pt_regs *regs,
 					unsigned long addr,
 					unsigned long type_ctx);
@@ -578,6 +578,7 @@ void handle_ld_nf(u32 insn, struct pt_regs *regs)
 
 void handle_lddfmna(struct pt_regs *regs, unsigned long sfar, unsigned long sfsr)
 {
+	enum ctx_state prev_state = exception_enter();
 	unsigned long pc = regs->tpc;
 	unsigned long tstate = regs->tstate;
 	u32 insn;
@@ -632,13 +633,16 @@ daex:
 			sun4v_data_access_exception(regs, sfar, sfsr);
 		else
 			spitfire_data_access_exception(regs, sfsr, sfar);
-		return;
+		goto out;
 	}
 	advance(regs);
+out:
+	exception_exit(prev_state);
 }
 
 void handle_stdfmna(struct pt_regs *regs, unsigned long sfar, unsigned long sfsr)
 {
+	enum ctx_state prev_state = exception_enter();
 	unsigned long pc = regs->tpc;
 	unsigned long tstate = regs->tstate;
 	u32 insn;
@@ -680,7 +684,9 @@ daex:
 			sun4v_data_access_exception(regs, sfar, sfsr);
 		else
 			spitfire_data_access_exception(regs, sfsr, sfar);
-		return;
+		goto out;
 	}
 	advance(regs);
+out:
+	exception_exit(prev_state);
 }
diff --git a/arch/sparc/kernel/vmlinux.lds.S b/arch/sparc/kernel/vmlinux.lds.S
index 0bacceb19150..932ff90fd760 100644
--- a/arch/sparc/kernel/vmlinux.lds.S
+++ b/arch/sparc/kernel/vmlinux.lds.S
@@ -122,6 +122,11 @@ SECTIONS
 		*(.swapper_4m_tsb_phys_patch)
 		__swapper_4m_tsb_phys_patch_end = .;
 	}
+	.page_offset_shift_patch : {
+		__page_offset_shift_patch = .;
+		*(.page_offset_shift_patch)
+		__page_offset_shift_patch_end = .;
+	}
 	.popc_3insn_patch : {
 		__popc_3insn_patch = .;
 		*(.popc_3insn_patch)
diff --git a/arch/sparc/lib/clear_page.S b/arch/sparc/lib/clear_page.S
index 77e531f6c2a7..46272dfc26e8 100644
--- a/arch/sparc/lib/clear_page.S
+++ b/arch/sparc/lib/clear_page.S
@@ -37,10 +37,10 @@ _clear_page:		/* %o0=dest */
 	.globl		clear_user_page
 clear_user_page:	/* %o0=dest, %o1=vaddr */
 	lduw		[%g6 + TI_PRE_COUNT], %o2
-	sethi		%uhi(PAGE_OFFSET), %g2
+	sethi		%hi(PAGE_OFFSET), %g2
 	sethi		%hi(PAGE_SIZE), %o4
 
-	sllx		%g2, 32, %g2
+	ldx		[%g2 + %lo(PAGE_OFFSET)], %g2
 	sethi		%hi(PAGE_KERNEL_LOCKED), %g3
 
 	ldx		[%g3 + %lo(PAGE_KERNEL_LOCKED)], %g3
diff --git a/arch/sparc/lib/copy_page.S b/arch/sparc/lib/copy_page.S
index 4d2df328e514..dd16c61f3263 100644
--- a/arch/sparc/lib/copy_page.S
+++ b/arch/sparc/lib/copy_page.S
@@ -46,10 +46,10 @@
 	.type		copy_user_page,#function
 copy_user_page:		/* %o0=dest, %o1=src, %o2=vaddr */
 	lduw		[%g6 + TI_PRE_COUNT], %o4
-	sethi		%uhi(PAGE_OFFSET), %g2
+	sethi		%hi(PAGE_OFFSET), %g2
 	sethi		%hi(PAGE_SIZE), %o3
 
-	sllx		%g2, 32, %g2
+	ldx		[%g2 + %lo(PAGE_OFFSET)], %g2
 	sethi		%hi(PAGE_KERNEL_LOCKED), %g3
 
 	ldx		[%g3 + %lo(PAGE_KERNEL_LOCKED)], %g3
diff --git a/arch/sparc/mm/fault_64.c b/arch/sparc/mm/fault_64.c
index 2ebec263d685..69bb818fdd79 100644
--- a/arch/sparc/mm/fault_64.c
+++ b/arch/sparc/mm/fault_64.c
@@ -21,6 +21,7 @@
 #include <linux/kprobes.h>
 #include <linux/kdebug.h>
 #include <linux/percpu.h>
+#include <linux/context_tracking.h>
 
 #include <asm/page.h>
 #include <asm/pgtable.h>
@@ -272,6 +273,7 @@ static void noinline __kprobes bogus_32bit_fault_address(struct pt_regs *regs,
 
 asmlinkage void __kprobes do_sparc64_fault(struct pt_regs *regs)
 {
+	enum ctx_state prev_state = exception_enter();
 	struct mm_struct *mm = current->mm;
 	struct vm_area_struct *vma;
 	unsigned int insn = 0;
@@ -282,7 +284,7 @@ asmlinkage void __kprobes do_sparc64_fault(struct pt_regs *regs)
 	fault_code = get_thread_fault_code();
 
 	if (notify_page_fault(regs))
-		return;
+		goto exit_exception;
 
 	si_code = SEGV_MAPERR;
 	address = current_thread_info()->fault_address;
@@ -313,7 +315,7 @@ asmlinkage void __kprobes do_sparc64_fault(struct pt_regs *regs)
 			/* Valid, no problems... */
 		} else {
 			bad_kernel_pc(regs, address);
-			return;
+			goto exit_exception;
 		}
 	} else
 		flags |= FAULT_FLAG_USER;
@@ -430,7 +432,7 @@ good_area:
 	fault = handle_mm_fault(mm, vma, address, flags);
 
 	if ((fault & VM_FAULT_RETRY) && fatal_signal_pending(current))
-		return;
+		goto exit_exception;
 
 	if (unlikely(fault & VM_FAULT_ERROR)) {
 		if (fault & VM_FAULT_OOM)
@@ -482,6 +484,8 @@ good_area:
 
 	}
 #endif
+exit_exception:
+	exception_exit(prev_state);
 	return;
 
 	/*
@@ -494,7 +498,7 @@ bad_area:
 
 handle_kernel_fault:
 	do_kernel_fault(regs, si_code, fault_code, insn, address);
-	return;
+	goto exit_exception;
 
 /*
  * We ran out of memory, or some other thing happened to us that made
@@ -505,7 +509,7 @@ out_of_memory:
 	up_read(&mm->mmap_sem);
 	if (!(regs->tstate & TSTATE_PRIV)) {
 		pagefault_out_of_memory();
-		return;
+		goto exit_exception;
 	}
 	goto handle_kernel_fault;
 
diff --git a/arch/sparc/mm/gup.c b/arch/sparc/mm/gup.c
index 01ee23dd724d..c4d3da68b800 100644
--- a/arch/sparc/mm/gup.c
+++ b/arch/sparc/mm/gup.c
@@ -71,13 +71,12 @@ static int gup_huge_pmd(pmd_t *pmdp, pmd_t pmd, unsigned long addr,
 			int *nr)
 {
 	struct page *head, *page, *tail;
-	u32 mask;
 	int refs;
 
-	mask = PMD_HUGE_PRESENT;
-	if (write)
-		mask |= PMD_HUGE_WRITE;
-	if ((pmd_val(pmd) & mask) != mask)
+	if (!pmd_large(pmd))
+		return 0;
+
+	if (write && !pmd_write(pmd))
 		return 0;
 
 	refs = 0;
diff --git a/arch/sparc/mm/hugetlbpage.c b/arch/sparc/mm/hugetlbpage.c
index 96399646570a..30963178d7e9 100644
--- a/arch/sparc/mm/hugetlbpage.c
+++ b/arch/sparc/mm/hugetlbpage.c
@@ -21,8 +21,6 @@
 /* Slightly simplified from the non-hugepage variant because by
  * definition we don't have to worry about any page coloring stuff
  */
-#define VA_EXCLUDE_START (0x0000080000000000UL - (1UL << 32UL))
-#define VA_EXCLUDE_END   (0xfffff80000000000UL + (1UL << 32UL))
 
 static unsigned long hugetlb_get_unmapped_area_bottomup(struct file *filp,
 							unsigned long addr,
diff --git a/arch/sparc/mm/init_64.c b/arch/sparc/mm/init_64.c
index ed82edad1a39..5322e530d09c 100644
--- a/arch/sparc/mm/init_64.c
+++ b/arch/sparc/mm/init_64.c
@@ -354,7 +354,7 @@ void update_mmu_cache(struct vm_area_struct *vma, unsigned long address, pte_t *
 
 #if defined(CONFIG_HUGETLB_PAGE) || defined(CONFIG_TRANSPARENT_HUGEPAGE)
 	if (mm->context.huge_pte_count && is_hugetlb_pte(pte))
-		__update_mmu_tsb_insert(mm, MM_TSB_HUGE, HPAGE_SHIFT,
+		__update_mmu_tsb_insert(mm, MM_TSB_HUGE, REAL_HPAGE_SHIFT,
 					address, pte_val(pte));
 	else
 #endif
@@ -1557,6 +1557,96 @@ unsigned long __init find_ecache_flush_span(unsigned long size)
 	return ~0UL;
 }
 
+unsigned long PAGE_OFFSET;
+EXPORT_SYMBOL(PAGE_OFFSET);
+
+static void __init page_offset_shift_patch_one(unsigned int *insn, unsigned long phys_bits)
+{
+	unsigned long final_shift;
+	unsigned int val = *insn;
+	unsigned int cnt;
+
+	/* We are patching in ilog2(max_supported_phys_address), and
+	 * we are doing so in a manner similar to a relocation addend.
+	 * That is, we are adding the shift value to whatever value
+	 * is in the shift instruction count field already.
+	 */
+	cnt = (val & 0x3f);
+	val &= ~0x3f;
+
+	/* If we are trying to shift >= 64 bits, clear the destination
+	 * register.  This can happen when phys_bits ends up being equal
+	 * to MAX_PHYS_ADDRESS_BITS.
+	 */
+	final_shift = (cnt + (64 - phys_bits));
+	if (final_shift >= 64) {
+		unsigned int rd = (val >> 25) & 0x1f;
+
+		val = 0x80100000 | (rd << 25);
+	} else {
+		val |= final_shift;
+	}
+	*insn = val;
+
+	__asm__ __volatile__("flush	%0"
+			     : /* no outputs */
+			     : "r" (insn));
+}
+
+static void __init page_offset_shift_patch(unsigned long phys_bits)
+{
+	extern unsigned int __page_offset_shift_patch;
+	extern unsigned int __page_offset_shift_patch_end;
+	unsigned int *p;
+
+	p = &__page_offset_shift_patch;
+	while (p < &__page_offset_shift_patch_end) {
+		unsigned int *insn = (unsigned int *)(unsigned long)*p;
+
+		page_offset_shift_patch_one(insn, phys_bits);
+
+		p++;
+	}
+}
+
+static void __init setup_page_offset(void)
+{
+	unsigned long max_phys_bits = 40;
+
+	if (tlb_type == cheetah || tlb_type == cheetah_plus) {
+		max_phys_bits = 42;
+	} else if (tlb_type == hypervisor) {
+		switch (sun4v_chip_type) {
+		case SUN4V_CHIP_NIAGARA1:
+		case SUN4V_CHIP_NIAGARA2:
+			max_phys_bits = 39;
+			break;
+		case SUN4V_CHIP_NIAGARA3:
+			max_phys_bits = 43;
+			break;
+		case SUN4V_CHIP_NIAGARA4:
+		case SUN4V_CHIP_NIAGARA5:
+		case SUN4V_CHIP_SPARC64X:
+		default:
+			max_phys_bits = 47;
+			break;
+		}
+	}
+
+	if (max_phys_bits > MAX_PHYS_ADDRESS_BITS) {
+		prom_printf("MAX_PHYS_ADDRESS_BITS is too small, need %lu\n",
+			    max_phys_bits);
+		prom_halt();
+	}
+
+	PAGE_OFFSET = PAGE_OFFSET_BY_BITS(max_phys_bits);
+
+	pr_info("PAGE_OFFSET is 0x%016lx (max_phys_bits == %lu)\n",
+		PAGE_OFFSET, max_phys_bits);
+
+	page_offset_shift_patch(max_phys_bits);
+}
+
 static void __init tsb_phys_patch(void)
 {
 	struct tsb_ldquad_phys_patch_entry *pquad;
@@ -1722,7 +1812,7 @@ static void __init sun4v_linear_pte_xor_finalize(void)
 #ifndef CONFIG_DEBUG_PAGEALLOC
 	if (cpu_pgsz_mask & HV_PGSZ_MASK_256MB) {
 		kern_linear_pte_xor[1] = (_PAGE_VALID | _PAGE_SZ256MB_4V) ^
-			0xfffff80000000000UL;
+			PAGE_OFFSET;
 		kern_linear_pte_xor[1] |= (_PAGE_CP_4V | _PAGE_CV_4V |
 					   _PAGE_P_4V | _PAGE_W_4V);
 	} else {
@@ -1731,7 +1821,7 @@ static void __init sun4v_linear_pte_xor_finalize(void)
 
 	if (cpu_pgsz_mask & HV_PGSZ_MASK_2GB) {
 		kern_linear_pte_xor[2] = (_PAGE_VALID | _PAGE_SZ2GB_4V) ^
-			0xfffff80000000000UL;
+			PAGE_OFFSET;
 		kern_linear_pte_xor[2] |= (_PAGE_CP_4V | _PAGE_CV_4V |
 					   _PAGE_P_4V | _PAGE_W_4V);
 	} else {
@@ -1740,7 +1830,7 @@ static void __init sun4v_linear_pte_xor_finalize(void)
 
 	if (cpu_pgsz_mask & HV_PGSZ_MASK_16GB) {
 		kern_linear_pte_xor[3] = (_PAGE_VALID | _PAGE_SZ16GB_4V) ^
-			0xfffff80000000000UL;
+			PAGE_OFFSET;
 		kern_linear_pte_xor[3] |= (_PAGE_CP_4V | _PAGE_CV_4V |
 					   _PAGE_P_4V | _PAGE_W_4V);
 	} else {
@@ -1752,7 +1842,7 @@ static void __init sun4v_linear_pte_xor_finalize(void)
 /* paging_init() sets up the page tables */
 
 static unsigned long last_valid_pfn;
-pgd_t swapper_pg_dir[2048];
+pgd_t swapper_pg_dir[PTRS_PER_PGD];
 
 static void sun4u_pgprot_init(void);
 static void sun4v_pgprot_init(void);
@@ -1763,6 +1853,8 @@ void __init paging_init(void)
 	unsigned long real_end, i;
 	int node;
 
+	setup_page_offset();
+
 	/* These build time checkes make sure that the dcache_dirty_cpu()
 	 * page->flags usage will work.
 	 *
@@ -2261,10 +2353,10 @@ static void __init sun4u_pgprot_init(void)
 		     __ACCESS_BITS_4U | _PAGE_E_4U);
 
 #ifdef CONFIG_DEBUG_PAGEALLOC
-	kern_linear_pte_xor[0] = _PAGE_VALID ^ 0xfffff80000000000UL;
+	kern_linear_pte_xor[0] = _PAGE_VALID ^ PAGE_OFFSET;
 #else
 	kern_linear_pte_xor[0] = (_PAGE_VALID | _PAGE_SZ4MB_4U) ^
-		0xfffff80000000000UL;
+		PAGE_OFFSET;
 #endif
 	kern_linear_pte_xor[0] |= (_PAGE_CP_4U | _PAGE_CV_4U |
 				   _PAGE_P_4U | _PAGE_W_4U);
@@ -2308,10 +2400,10 @@ static void __init sun4v_pgprot_init(void)
 	_PAGE_CACHE = _PAGE_CACHE_4V;
 
 #ifdef CONFIG_DEBUG_PAGEALLOC
-	kern_linear_pte_xor[0] = _PAGE_VALID ^ 0xfffff80000000000UL;
+	kern_linear_pte_xor[0] = _PAGE_VALID ^ PAGE_OFFSET;
 #else
 	kern_linear_pte_xor[0] = (_PAGE_VALID | _PAGE_SZ4MB_4V) ^
-		0xfffff80000000000UL;
+		PAGE_OFFSET;
 #endif
 	kern_linear_pte_xor[0] |= (_PAGE_CP_4V | _PAGE_CV_4V |
 				   _PAGE_P_4V | _PAGE_W_4V);
@@ -2455,53 +2547,13 @@ void __flush_tlb_all(void)
 			     : : "r" (pstate));
 }
 
-static pte_t *get_from_cache(struct mm_struct *mm)
-{
-	struct page *page;
-	pte_t *ret;
-
-	spin_lock(&mm->page_table_lock);
-	page = mm->context.pgtable_page;
-	ret = NULL;
-	if (page) {
-		void *p = page_address(page);
-
-		mm->context.pgtable_page = NULL;
-
-		ret = (pte_t *) (p + (PAGE_SIZE / 2));
-	}
-	spin_unlock(&mm->page_table_lock);
-
-	return ret;
-}
-
-static struct page *__alloc_for_cache(struct mm_struct *mm)
-{
-	struct page *page = alloc_page(GFP_KERNEL | __GFP_NOTRACK |
-				       __GFP_REPEAT | __GFP_ZERO);
-
-	if (page) {
-		spin_lock(&mm->page_table_lock);
-		if (!mm->context.pgtable_page) {
-			atomic_set(&page->_count, 2);
-			mm->context.pgtable_page = page;
-		}
-		spin_unlock(&mm->page_table_lock);
-	}
-	return page;
-}
-
 pte_t *pte_alloc_one_kernel(struct mm_struct *mm,
 			    unsigned long address)
 {
-	struct page *page;
-	pte_t *pte;
-
-	pte = get_from_cache(mm);
-	if (pte)
-		return pte;
+	struct page *page = alloc_page(GFP_KERNEL | __GFP_NOTRACK |
+				       __GFP_REPEAT | __GFP_ZERO);
+	pte_t *pte = NULL;
 
-	page = __alloc_for_cache(mm);
 	if (page)
 		pte = (pte_t *) page_address(page);
 
@@ -2511,36 +2563,28 @@ pte_t *pte_alloc_one_kernel(struct mm_struct *mm,
 pgtable_t pte_alloc_one(struct mm_struct *mm,
 			unsigned long address)
 {
-	struct page *page;
-	pte_t *pte;
-
-	pte = get_from_cache(mm);
-	if (pte)
-		return pte;
-
-	page = __alloc_for_cache(mm);
-	if (page) {
-		pgtable_page_ctor(page);
-		pte = (pte_t *) page_address(page);
+	struct page *page = alloc_page(GFP_KERNEL | __GFP_NOTRACK |
+				       __GFP_REPEAT | __GFP_ZERO);
+	if (!page)
+		return NULL;
+	if (!pgtable_page_ctor(page)) {
+		free_hot_cold_page(page, 0);
+		return NULL;
 	}
-
-	return pte;
+	return (pte_t *) page_address(page);
 }
 
 void pte_free_kernel(struct mm_struct *mm, pte_t *pte)
 {
-	struct page *page = virt_to_page(pte);
-	if (put_page_testzero(page))
-		free_hot_cold_page(page, 0);
+	free_page((unsigned long)pte);
 }
 
 static void __pte_free(pgtable_t pte)
 {
 	struct page *page = virt_to_page(pte);
-	if (put_page_testzero(page)) {
-		pgtable_page_dtor(page);
-		free_hot_cold_page(page, 0);
-	}
+
+	pgtable_page_dtor(page);
+	__free_page(page);
 }
 
 void pte_free(struct mm_struct *mm, pgtable_t pte)
@@ -2557,124 +2601,27 @@ void pgtable_free(void *table, bool is_page)
 }
 
 #ifdef CONFIG_TRANSPARENT_HUGEPAGE
-static pmd_t pmd_set_protbits(pmd_t pmd, pgprot_t pgprot, bool for_modify)
-{
-	if (pgprot_val(pgprot) & _PAGE_VALID)
-		pmd_val(pmd) |= PMD_HUGE_PRESENT;
-	if (tlb_type == hypervisor) {
-		if (pgprot_val(pgprot) & _PAGE_WRITE_4V)
-			pmd_val(pmd) |= PMD_HUGE_WRITE;
-		if (pgprot_val(pgprot) & _PAGE_EXEC_4V)
-			pmd_val(pmd) |= PMD_HUGE_EXEC;
-
-		if (!for_modify) {
-			if (pgprot_val(pgprot) & _PAGE_ACCESSED_4V)
-				pmd_val(pmd) |= PMD_HUGE_ACCESSED;
-			if (pgprot_val(pgprot) & _PAGE_MODIFIED_4V)
-				pmd_val(pmd) |= PMD_HUGE_DIRTY;
-		}
-	} else {
-		if (pgprot_val(pgprot) & _PAGE_WRITE_4U)
-			pmd_val(pmd) |= PMD_HUGE_WRITE;
-		if (pgprot_val(pgprot) & _PAGE_EXEC_4U)
-			pmd_val(pmd) |= PMD_HUGE_EXEC;
-
-		if (!for_modify) {
-			if (pgprot_val(pgprot) & _PAGE_ACCESSED_4U)
-				pmd_val(pmd) |= PMD_HUGE_ACCESSED;
-			if (pgprot_val(pgprot) & _PAGE_MODIFIED_4U)
-				pmd_val(pmd) |= PMD_HUGE_DIRTY;
-		}
-	}
-
-	return pmd;
-}
-
-pmd_t pfn_pmd(unsigned long page_nr, pgprot_t pgprot)
-{
-	pmd_t pmd;
-
-	pmd_val(pmd) = (page_nr << ((PAGE_SHIFT - PMD_PADDR_SHIFT)));
-	pmd_val(pmd) |= PMD_ISHUGE;
-	pmd = pmd_set_protbits(pmd, pgprot, false);
-	return pmd;
-}
-
-pmd_t pmd_modify(pmd_t pmd, pgprot_t newprot)
-{
-	pmd_val(pmd) &= ~(PMD_HUGE_PRESENT |
-			  PMD_HUGE_WRITE |
-			  PMD_HUGE_EXEC);
-	pmd = pmd_set_protbits(pmd, newprot, true);
-	return pmd;
-}
-
-pgprot_t pmd_pgprot(pmd_t entry)
-{
-	unsigned long pte = 0;
-
-	if (pmd_val(entry) & PMD_HUGE_PRESENT)
-		pte |= _PAGE_VALID;
-
-	if (tlb_type == hypervisor) {
-		if (pmd_val(entry) & PMD_HUGE_PRESENT)
-			pte |= _PAGE_PRESENT_4V;
-		if (pmd_val(entry) & PMD_HUGE_EXEC)
-			pte |= _PAGE_EXEC_4V;
-		if (pmd_val(entry) & PMD_HUGE_WRITE)
-			pte |= _PAGE_W_4V;
-		if (pmd_val(entry) & PMD_HUGE_ACCESSED)
-			pte |= _PAGE_ACCESSED_4V;
-		if (pmd_val(entry) & PMD_HUGE_DIRTY)
-			pte |= _PAGE_MODIFIED_4V;
-		pte |= _PAGE_CP_4V|_PAGE_CV_4V;
-	} else {
-		if (pmd_val(entry) & PMD_HUGE_PRESENT)
-			pte |= _PAGE_PRESENT_4U;
-		if (pmd_val(entry) & PMD_HUGE_EXEC)
-			pte |= _PAGE_EXEC_4U;
-		if (pmd_val(entry) & PMD_HUGE_WRITE)
-			pte |= _PAGE_W_4U;
-		if (pmd_val(entry) & PMD_HUGE_ACCESSED)
-			pte |= _PAGE_ACCESSED_4U;
-		if (pmd_val(entry) & PMD_HUGE_DIRTY)
-			pte |= _PAGE_MODIFIED_4U;
-		pte |= _PAGE_CP_4U|_PAGE_CV_4U;
-	}
-
-	return __pgprot(pte);
-}
-
 void update_mmu_cache_pmd(struct vm_area_struct *vma, unsigned long addr,
 			  pmd_t *pmd)
 {
 	unsigned long pte, flags;
 	struct mm_struct *mm;
 	pmd_t entry = *pmd;
-	pgprot_t prot;
 
 	if (!pmd_large(entry) || !pmd_young(entry))
 		return;
 
-	pte = (pmd_val(entry) & ~PMD_HUGE_PROTBITS);
-	pte <<= PMD_PADDR_SHIFT;
-	pte |= _PAGE_VALID;
-
-	prot = pmd_pgprot(entry);
-
-	if (tlb_type == hypervisor)
-		pgprot_val(prot) |= _PAGE_SZHUGE_4V;
-	else
-		pgprot_val(prot) |= _PAGE_SZHUGE_4U;
+	pte = pmd_val(entry);
 
-	pte |= pgprot_val(prot);
+	/* We are fabricating 8MB pages using 4MB real hw pages.  */
+	pte |= (addr & (1UL << REAL_HPAGE_SHIFT));
 
 	mm = vma->vm_mm;
 
 	spin_lock_irqsave(&mm->context.lock, flags);
 
 	if (mm->context.tsb_block[MM_TSB_HUGE].tsb != NULL)
-		__update_mmu_tsb_insert(mm, MM_TSB_HUGE, HPAGE_SHIFT,
+		__update_mmu_tsb_insert(mm, MM_TSB_HUGE, REAL_HPAGE_SHIFT,
 					addr, pte);
 
 	spin_unlock_irqrestore(&mm->context.lock, flags);
diff --git a/arch/sparc/mm/init_64.h b/arch/sparc/mm/init_64.h
index 0661aa606dec..5d3782deb403 100644
--- a/arch/sparc/mm/init_64.h
+++ b/arch/sparc/mm/init_64.h
@@ -1,11 +1,13 @@
 #ifndef _SPARC64_MM_INIT_H
 #define _SPARC64_MM_INIT_H
 
+#include <asm/page.h>
+
 /* Most of the symbols in this file are defined in init.c and
  * marked non-static so that assembler code can get at them.
  */
 
-#define MAX_PHYS_ADDRESS	(1UL << 41UL)
+#define MAX_PHYS_ADDRESS	(1UL << MAX_PHYS_ADDRESS_BITS)
 #define KPTE_BITMAP_CHUNK_SZ		(256UL * 1024UL * 1024UL)
 #define KPTE_BITMAP_BYTES	\
 	((MAX_PHYS_ADDRESS / KPTE_BITMAP_CHUNK_SZ) / 4)
diff --git a/arch/sparc/mm/srmmu.c b/arch/sparc/mm/srmmu.c
index 5d721df48a72..869023abe5a4 100644
--- a/arch/sparc/mm/srmmu.c
+++ b/arch/sparc/mm/srmmu.c
@@ -345,7 +345,10 @@ pgtable_t pte_alloc_one(struct mm_struct *mm, unsigned long address)
 	if ((pte = (unsigned long)pte_alloc_one_kernel(mm, address)) == 0)
 		return NULL;
 	page = pfn_to_page(__nocache_pa(pte) >> PAGE_SHIFT);
-	pgtable_page_ctor(page);
+	if (!pgtable_page_ctor(page)) {
+		__free_page(page);
+		return NULL;
+	}
 	return page;
 }
 
diff --git a/arch/sparc/mm/tlb.c b/arch/sparc/mm/tlb.c
index 7a91f288c708..ad3bf4b4324d 100644
--- a/arch/sparc/mm/tlb.c
+++ b/arch/sparc/mm/tlb.c
@@ -161,8 +161,8 @@ void set_pmd_at(struct mm_struct *mm, unsigned long addr,
 	if (mm == &init_mm)
 		return;
 
-	if ((pmd_val(pmd) ^ pmd_val(orig)) & PMD_ISHUGE) {
-		if (pmd_val(pmd) & PMD_ISHUGE)
+	if ((pmd_val(pmd) ^ pmd_val(orig)) & _PAGE_PMD_HUGE) {
+		if (pmd_val(pmd) & _PAGE_PMD_HUGE)
 			mm->context.huge_pte_count++;
 		else
 			mm->context.huge_pte_count--;
@@ -178,13 +178,16 @@ void set_pmd_at(struct mm_struct *mm, unsigned long addr,
 	}
 
 	if (!pmd_none(orig)) {
-		bool exec = ((pmd_val(orig) & PMD_HUGE_EXEC) != 0);
+		pte_t orig_pte = __pte(pmd_val(orig));
+		bool exec = pte_exec(orig_pte);
 
 		addr &= HPAGE_MASK;
-		if (pmd_val(orig) & PMD_ISHUGE)
+		if (pmd_trans_huge(orig)) {
 			tlb_batch_add_one(mm, addr, exec);
-		else
+			tlb_batch_add_one(mm, addr + REAL_HPAGE_SIZE, exec);
+		} else {
 			tlb_batch_pmd_scan(mm, addr, orig, exec);
+		}
 	}
 }
 
@@ -196,11 +199,11 @@ void pgtable_trans_huge_deposit(struct mm_struct *mm, pmd_t *pmdp,
 	assert_spin_locked(&mm->page_table_lock);
 
 	/* FIFO */
-	if (!mm->pmd_huge_pte)
+	if (!pmd_huge_pte(mm, pmdp))
 		INIT_LIST_HEAD(lh);
 	else
-		list_add(lh, (struct list_head *) mm->pmd_huge_pte);
-	mm->pmd_huge_pte = pgtable;
+		list_add(lh, (struct list_head *) pmd_huge_pte(mm, pmdp));
+	pmd_huge_pte(mm, pmdp) = pgtable;
 }
 
 pgtable_t pgtable_trans_huge_withdraw(struct mm_struct *mm, pmd_t *pmdp)
@@ -211,12 +214,12 @@ pgtable_t pgtable_trans_huge_withdraw(struct mm_struct *mm, pmd_t *pmdp)
 	assert_spin_locked(&mm->page_table_lock);
 
 	/* FIFO */
-	pgtable = mm->pmd_huge_pte;
+	pgtable = pmd_huge_pte(mm, pmdp);
 	lh = (struct list_head *) pgtable;
 	if (list_empty(lh))
-		mm->pmd_huge_pte = NULL;
+		pmd_huge_pte(mm, pmdp) = NULL;
 	else {
-		mm->pmd_huge_pte = (pgtable_t) lh->next;
+		pmd_huge_pte(mm, pmdp) = (pgtable_t) lh->next;
 		list_del(lh);
 	}
 	pte_val(pgtable[0]) = 0;
diff --git a/arch/sparc/mm/tsb.c b/arch/sparc/mm/tsb.c
index 2cc3bce5ee91..3b3a360b429a 100644
--- a/arch/sparc/mm/tsb.c
+++ b/arch/sparc/mm/tsb.c
@@ -87,7 +87,7 @@ void flush_tsb_user(struct tlb_batch *tb)
 		nentries = mm->context.tsb_block[MM_TSB_HUGE].tsb_nentries;
 		if (tlb_type == cheetah_plus || tlb_type == hypervisor)
 			base = __pa(base);
-		__flush_tsb_one(tb, HPAGE_SHIFT, base, nentries);
+		__flush_tsb_one(tb, REAL_HPAGE_SHIFT, base, nentries);
 	}
 #endif
 	spin_unlock_irqrestore(&mm->context.lock, flags);
@@ -111,7 +111,7 @@ void flush_tsb_user_page(struct mm_struct *mm, unsigned long vaddr)
 		nentries = mm->context.tsb_block[MM_TSB_HUGE].tsb_nentries;
 		if (tlb_type == cheetah_plus || tlb_type == hypervisor)
 			base = __pa(base);
-		__flush_tsb_one_entry(base, vaddr, HPAGE_SHIFT, nentries);
+		__flush_tsb_one_entry(base, vaddr, REAL_HPAGE_SHIFT, nentries);
 	}
 #endif
 	spin_unlock_irqrestore(&mm->context.lock, flags);
@@ -472,8 +472,6 @@ int init_new_context(struct task_struct *tsk, struct mm_struct *mm)
 	mm->context.huge_pte_count = 0;
 #endif
 
-	mm->context.pgtable_page = NULL;
-
 	/* copy_mm() copies over the parent's mm_struct before calling
 	 * us, so we need to zero out the TSB pointer or else tsb_grow()
 	 * will be confused and think there is an older TSB to free up.
@@ -512,17 +510,10 @@ static void tsb_destroy_one(struct tsb_config *tp)
 void destroy_context(struct mm_struct *mm)
 {
 	unsigned long flags, i;
-	struct page *page;
 
 	for (i = 0; i < MM_NUM_TSBS; i++)
 		tsb_destroy_one(&mm->context.tsb_block[i]);
 
-	page = mm->context.pgtable_page;
-	if (page && put_page_testzero(page)) {
-		pgtable_page_dtor(page);
-		free_hot_cold_page(page, 0);
-	}
-
 	spin_lock_irqsave(&ctx_alloc_lock, flags);
 
 	if (CTX_VALID(mm->context)) {
diff --git a/arch/sparc/mm/ultra.S b/arch/sparc/mm/ultra.S
index 432aa0cb1b38..b4f4733abc6e 100644
--- a/arch/sparc/mm/ultra.S
+++ b/arch/sparc/mm/ultra.S
@@ -153,10 +153,10 @@ __spitfire_flush_tlb_mm_slow:
 	.globl		__flush_icache_page
 __flush_icache_page:	/* %o0 = phys_page */
 	srlx		%o0, PAGE_SHIFT, %o0
-	sethi		%uhi(PAGE_OFFSET), %g1
+	sethi		%hi(PAGE_OFFSET), %g1
 	sllx		%o0, PAGE_SHIFT, %o0
 	sethi		%hi(PAGE_SIZE), %g2
-	sllx		%g1, 32, %g1
+	ldx		[%g1 + %lo(PAGE_OFFSET)], %g1
 	add		%o0, %g1, %o0
 1:	subcc		%g2, 32, %g2
 	bne,pt		%icc, 1b
@@ -178,8 +178,8 @@ __flush_icache_page:	/* %o0 = phys_page */
 	.align		64
 	.globl		__flush_dcache_page
 __flush_dcache_page:	/* %o0=kaddr, %o1=flush_icache */
-	sethi		%uhi(PAGE_OFFSET), %g1
-	sllx		%g1, 32, %g1
+	sethi		%hi(PAGE_OFFSET), %g1
+	ldx		[%g1 + %lo(PAGE_OFFSET)], %g1
 	sub		%o0, %g1, %o0			! physical address
 	srlx		%o0, 11, %o0			! make D-cache TAG
 	sethi		%hi(1 << 14), %o2		! D-cache size
@@ -287,8 +287,8 @@ __cheetah_flush_tlb_pending:	/* 27 insns */
 
 #ifdef DCACHE_ALIASING_POSSIBLE
 __cheetah_flush_dcache_page: /* 11 insns */
-	sethi		%uhi(PAGE_OFFSET), %g1
-	sllx		%g1, 32, %g1
+	sethi		%hi(PAGE_OFFSET), %g1
+	ldx		[%g1 + %lo(PAGE_OFFSET)], %g1
 	sub		%o0, %g1, %o0
 	sethi		%hi(PAGE_SIZE), %o4
 1:	subcc		%o4, (1 << 5), %o4
diff --git a/arch/tile/Kconfig b/arch/tile/Kconfig
index d45a2c48f185..b3692ce78f90 100644
--- a/arch/tile/Kconfig
+++ b/arch/tile/Kconfig
@@ -8,7 +8,6 @@ config TILE
 	select HAVE_KVM if !TILEGX
 	select GENERIC_FIND_FIRST_BIT
 	select SYSCTL_EXCEPTION_TRACE
-	select USE_GENERIC_SMP_HELPERS
 	select CC_OPTIMIZE_FOR_SIZE
 	select HAVE_DEBUG_KMEMLEAK
 	select GENERIC_IRQ_PROBE
diff --git a/arch/tile/include/asm/hardirq.h b/arch/tile/include/asm/hardirq.h
index 822390f9a154..54110af23985 100644
--- a/arch/tile/include/asm/hardirq.h
+++ b/arch/tile/include/asm/hardirq.h
@@ -42,6 +42,4 @@ DECLARE_PER_CPU(irq_cpustat_t, irq_stat);
 
 #include <linux/irq_cpustat.h>	/* Standard mappings for irq_cpustat_t above */
 
-#define HARDIRQ_BITS	8
-
 #endif /* _ASM_TILE_HARDIRQ_H */
diff --git a/arch/tile/include/asm/thread_info.h b/arch/tile/include/asm/thread_info.h
index b8aa6df3e102..729aa107f64e 100644
--- a/arch/tile/include/asm/thread_info.h
+++ b/arch/tile/include/asm/thread_info.h
@@ -113,8 +113,6 @@ extern void _cpu_idle(void);
 
 #endif /* !__ASSEMBLY__ */
 
-#define PREEMPT_ACTIVE		0x10000000
-
 /*
  * Thread information flags that various assembly files may need to access.
  * Keep flags accessed frequently in low bits, particular since it makes
diff --git a/arch/tile/mm/pgtable.c b/arch/tile/mm/pgtable.c
index 4fd9ec0b58ed..5e86eac4bfae 100644
--- a/arch/tile/mm/pgtable.c
+++ b/arch/tile/mm/pgtable.c
@@ -241,6 +241,11 @@ struct page *pgtable_alloc_one(struct mm_struct *mm, unsigned long address,
 	if (p == NULL)
 		return NULL;
 
+	if (!pgtable_page_ctor(p)) {
+		__free_pages(p, L2_USER_PGTABLE_ORDER);
+		return NULL;
+	}
+
 	/*
 	 * Make every page have a page_count() of one, not just the first.
 	 * We don't use __GFP_COMP since it doesn't look like it works
@@ -251,7 +256,6 @@ struct page *pgtable_alloc_one(struct mm_struct *mm, unsigned long address,
 		inc_zone_page_state(p+i, NR_PAGETABLE);
 	}
 
-	pgtable_page_ctor(p);
 	return p;
 }
 
diff --git a/arch/um/Kconfig.char b/arch/um/Kconfig.char
index b9d7c4276682..f10738d68b2d 100644
--- a/arch/um/Kconfig.char
+++ b/arch/um/Kconfig.char
@@ -6,10 +6,6 @@ config STDERR_CONSOLE
 	help
 	  console driver which dumps all printk messages to stderr.
 
-config STDIO_CONSOLE
-	bool
-	default y
-
 config SSL
 	bool "Virtual serial line"
 	help
diff --git a/arch/um/Kconfig.common b/arch/um/Kconfig.common
index 8ddea1f8006a..21ca44c4f6d5 100644
--- a/arch/um/Kconfig.common
+++ b/arch/um/Kconfig.common
@@ -1,8 +1,3 @@
-config DEFCONFIG_LIST
-	string
-	option defconfig_list
-	default "arch/$ARCH/defconfig"
-
 config UML
 	bool
 	default y
diff --git a/arch/um/Makefile b/arch/um/Makefile
index 133f7de2a13d..48d92bbe62e9 100644
--- a/arch/um/Makefile
+++ b/arch/um/Makefile
@@ -6,6 +6,17 @@
 # Licensed under the GPL
 #
 
+# select defconfig based on actual architecture
+ifeq ($(SUBARCH),x86)
+  ifeq ($(shell uname -m),x86_64)
+        KBUILD_DEFCONFIG := x86_64_defconfig
+  else
+        KBUILD_DEFCONFIG := i386_defconfig
+  endif
+else
+        KBUILD_DEFCONFIG := $(SUBARCH)_defconfig
+endif
+
 ARCH_DIR := arch/um
 OS := $(shell uname -s)
 # We require bash because the vmlinux link and loader script cpp use bash
diff --git a/arch/um/configs/i386_defconfig b/arch/um/configs/i386_defconfig
new file mode 100644
index 000000000000..a12bf68c9f3a
--- /dev/null
+++ b/arch/um/configs/i386_defconfig
@@ -0,0 +1,76 @@
+CONFIG_3_LEVEL_PGTABLES=y
+# CONFIG_COMPACTION is not set
+CONFIG_BINFMT_MISC=m
+CONFIG_HOSTFS=y
+CONFIG_MAGIC_SYSRQ=y
+CONFIG_KERNEL_STACK_ORDER=1
+CONFIG_SYSVIPC=y
+CONFIG_POSIX_MQUEUE=y
+CONFIG_NO_HZ=y
+CONFIG_HIGH_RES_TIMERS=y
+CONFIG_BSD_PROCESS_ACCT=y
+CONFIG_IKCONFIG=y
+CONFIG_IKCONFIG_PROC=y
+CONFIG_LOG_BUF_SHIFT=14
+CONFIG_CGROUPS=y
+CONFIG_CGROUP_FREEZER=y
+CONFIG_CGROUP_DEVICE=y
+CONFIG_CPUSETS=y
+CONFIG_CGROUP_CPUACCT=y
+CONFIG_RESOURCE_COUNTERS=y
+CONFIG_CGROUP_SCHED=y
+CONFIG_BLK_CGROUP=y
+# CONFIG_PID_NS is not set
+CONFIG_SYSFS_DEPRECATED=y
+CONFIG_CC_OPTIMIZE_FOR_SIZE=y
+CONFIG_SLAB=y
+CONFIG_MODULES=y
+CONFIG_MODULE_UNLOAD=y
+# CONFIG_BLK_DEV_BSG is not set
+CONFIG_IOSCHED_CFQ=m
+CONFIG_SSL=y
+CONFIG_NULL_CHAN=y
+CONFIG_PORT_CHAN=y
+CONFIG_PTY_CHAN=y
+CONFIG_TTY_CHAN=y
+CONFIG_XTERM_CHAN=y
+CONFIG_CON_CHAN="pts"
+CONFIG_SSL_CHAN="pts"
+CONFIG_UML_SOUND=m
+CONFIG_UEVENT_HELPER_PATH="/sbin/hotplug"
+CONFIG_DEVTMPFS=y
+CONFIG_DEVTMPFS_MOUNT=y
+CONFIG_BLK_DEV_UBD=y
+CONFIG_BLK_DEV_LOOP=m
+CONFIG_BLK_DEV_NBD=m
+CONFIG_DUMMY=m
+CONFIG_TUN=m
+CONFIG_PPP=m
+CONFIG_SLIP=m
+CONFIG_LEGACY_PTY_COUNT=32
+# CONFIG_HW_RANDOM is not set
+CONFIG_UML_RANDOM=y
+CONFIG_NET=y
+CONFIG_PACKET=y
+CONFIG_UNIX=y
+CONFIG_INET=y
+# CONFIG_INET_LRO is not set
+# CONFIG_IPV6 is not set
+CONFIG_UML_NET=y
+CONFIG_UML_NET_ETHERTAP=y
+CONFIG_UML_NET_TUNTAP=y
+CONFIG_UML_NET_SLIP=y
+CONFIG_UML_NET_DAEMON=y
+CONFIG_UML_NET_MCAST=y
+CONFIG_UML_NET_SLIRP=y
+CONFIG_EXT4_FS=y
+CONFIG_REISERFS_FS=y
+CONFIG_QUOTA=y
+CONFIG_AUTOFS4_FS=m
+CONFIG_ISO9660_FS=m
+CONFIG_JOLIET=y
+CONFIG_PROC_KCORE=y
+CONFIG_TMPFS=y
+CONFIG_NLS=y
+CONFIG_DEBUG_INFO=y
+CONFIG_DEBUG_KERNEL=y
diff --git a/arch/um/configs/x86_64_defconfig b/arch/um/configs/x86_64_defconfig
new file mode 100644
index 000000000000..3aab117bd553
--- /dev/null
+++ b/arch/um/configs/x86_64_defconfig
@@ -0,0 +1,75 @@
+# CONFIG_COMPACTION is not set
+CONFIG_BINFMT_MISC=m
+CONFIG_HOSTFS=y
+CONFIG_MAGIC_SYSRQ=y
+CONFIG_SYSVIPC=y
+CONFIG_POSIX_MQUEUE=y
+CONFIG_NO_HZ=y
+CONFIG_HIGH_RES_TIMERS=y
+CONFIG_BSD_PROCESS_ACCT=y
+CONFIG_IKCONFIG=y
+CONFIG_IKCONFIG_PROC=y
+CONFIG_LOG_BUF_SHIFT=14
+CONFIG_CGROUPS=y
+CONFIG_CGROUP_FREEZER=y
+CONFIG_CGROUP_DEVICE=y
+CONFIG_CPUSETS=y
+CONFIG_CGROUP_CPUACCT=y
+CONFIG_RESOURCE_COUNTERS=y
+CONFIG_CGROUP_SCHED=y
+CONFIG_BLK_CGROUP=y
+# CONFIG_PID_NS is not set
+CONFIG_SYSFS_DEPRECATED=y
+CONFIG_CC_OPTIMIZE_FOR_SIZE=y
+CONFIG_SLAB=y
+CONFIG_MODULES=y
+CONFIG_MODULE_UNLOAD=y
+# CONFIG_BLK_DEV_BSG is not set
+CONFIG_IOSCHED_CFQ=m
+CONFIG_SSL=y
+CONFIG_NULL_CHAN=y
+CONFIG_PORT_CHAN=y
+CONFIG_PTY_CHAN=y
+CONFIG_TTY_CHAN=y
+CONFIG_XTERM_CHAN=y
+CONFIG_CON_CHAN="pts"
+CONFIG_SSL_CHAN="pts"
+CONFIG_UML_SOUND=m
+CONFIG_UEVENT_HELPER_PATH="/sbin/hotplug"
+CONFIG_DEVTMPFS=y
+CONFIG_DEVTMPFS_MOUNT=y
+CONFIG_BLK_DEV_UBD=y
+CONFIG_BLK_DEV_LOOP=m
+CONFIG_BLK_DEV_NBD=m
+CONFIG_DUMMY=m
+CONFIG_TUN=m
+CONFIG_PPP=m
+CONFIG_SLIP=m
+CONFIG_LEGACY_PTY_COUNT=32
+# CONFIG_HW_RANDOM is not set
+CONFIG_UML_RANDOM=y
+CONFIG_NET=y
+CONFIG_PACKET=y
+CONFIG_UNIX=y
+CONFIG_INET=y
+# CONFIG_INET_LRO is not set
+# CONFIG_IPV6 is not set
+CONFIG_UML_NET=y
+CONFIG_UML_NET_ETHERTAP=y
+CONFIG_UML_NET_TUNTAP=y
+CONFIG_UML_NET_SLIP=y
+CONFIG_UML_NET_DAEMON=y
+CONFIG_UML_NET_MCAST=y
+CONFIG_UML_NET_SLIRP=y
+CONFIG_EXT4_FS=y
+CONFIG_REISERFS_FS=y
+CONFIG_QUOTA=y
+CONFIG_AUTOFS4_FS=m
+CONFIG_ISO9660_FS=m
+CONFIG_JOLIET=y
+CONFIG_PROC_KCORE=y
+CONFIG_TMPFS=y
+CONFIG_NLS=y
+CONFIG_DEBUG_INFO=y
+CONFIG_FRAME_WARN=1024
+CONFIG_DEBUG_KERNEL=y
diff --git a/arch/um/defconfig b/arch/um/defconfig
deleted file mode 100644
index 2665e6b683f5..000000000000
--- a/arch/um/defconfig
+++ /dev/null
@@ -1,899 +0,0 @@
-#
-# Automatically generated file; DO NOT EDIT.
-# User Mode Linux/i386 3.3.0 Kernel Configuration
-#
-CONFIG_DEFCONFIG_LIST="arch/$ARCH/defconfig"
-CONFIG_UML=y
-CONFIG_MMU=y
-CONFIG_NO_IOMEM=y
-# CONFIG_TRACE_IRQFLAGS_SUPPORT is not set
-CONFIG_LOCKDEP_SUPPORT=y
-# CONFIG_STACKTRACE_SUPPORT is not set
-CONFIG_GENERIC_CALIBRATE_DELAY=y
-CONFIG_GENERIC_BUG=y
-CONFIG_GENERIC_CLOCKEVENTS=y
-CONFIG_HZ=100
-
-#
-# UML-specific options
-#
-
-#
-# Host processor type and features
-#
-# CONFIG_M486 is not set
-# CONFIG_M586 is not set
-# CONFIG_M586TSC is not set
-# CONFIG_M586MMX is not set
-CONFIG_M686=y
-# CONFIG_MPENTIUMII is not set
-# CONFIG_MPENTIUMIII is not set
-# CONFIG_MPENTIUMM is not set
-# CONFIG_MPENTIUM4 is not set
-# CONFIG_MK6 is not set
-# CONFIG_MK7 is not set
-# CONFIG_MK8 is not set
-# CONFIG_MCRUSOE is not set
-# CONFIG_MEFFICEON is not set
-# CONFIG_MWINCHIPC6 is not set
-# CONFIG_MWINCHIP3D is not set
-# CONFIG_MELAN is not set
-# CONFIG_MGEODEGX1 is not set
-# CONFIG_MGEODE_LX is not set
-# CONFIG_MCYRIXIII is not set
-# CONFIG_MVIAC3_2 is not set
-# CONFIG_MVIAC7 is not set
-# CONFIG_MCORE2 is not set
-# CONFIG_MATOM is not set
-# CONFIG_X86_GENERIC is not set
-CONFIG_X86_INTERNODE_CACHE_SHIFT=5
-CONFIG_X86_CMPXCHG=y
-CONFIG_X86_L1_CACHE_SHIFT=5
-CONFIG_X86_XADD=y
-CONFIG_X86_PPRO_FENCE=y
-CONFIG_X86_WP_WORKS_OK=y
-CONFIG_X86_INVLPG=y
-CONFIG_X86_BSWAP=y
-CONFIG_X86_POPAD_OK=y
-CONFIG_X86_USE_PPRO_CHECKSUM=y
-CONFIG_X86_TSC=y
-CONFIG_X86_CMPXCHG64=y
-CONFIG_X86_CMOV=y
-CONFIG_X86_MINIMUM_CPU_FAMILY=5
-CONFIG_CPU_SUP_INTEL=y
-CONFIG_CPU_SUP_CYRIX_32=y
-CONFIG_CPU_SUP_AMD=y
-CONFIG_CPU_SUP_CENTAUR=y
-CONFIG_CPU_SUP_TRANSMETA_32=y
-CONFIG_CPU_SUP_UMC_32=y
-CONFIG_UML_X86=y
-# CONFIG_64BIT is not set
-CONFIG_X86_32=y
-# CONFIG_X86_64 is not set
-# CONFIG_RWSEM_XCHGADD_ALGORITHM is not set
-CONFIG_RWSEM_GENERIC_SPINLOCK=y
-# CONFIG_3_LEVEL_PGTABLES is not set
-CONFIG_ARCH_HAS_SC_SIGNALS=y
-CONFIG_ARCH_REUSE_HOST_VSYSCALL_AREA=y
-CONFIG_GENERIC_HWEIGHT=y
-# CONFIG_STATIC_LINK is not set
-CONFIG_SELECT_MEMORY_MODEL=y
-CONFIG_FLATMEM_MANUAL=y
-CONFIG_FLATMEM=y
-CONFIG_FLAT_NODE_MEM_MAP=y
-CONFIG_PAGEFLAGS_EXTENDED=y
-CONFIG_SPLIT_PTLOCK_CPUS=4
-# CONFIG_COMPACTION is not set
-# CONFIG_PHYS_ADDR_T_64BIT is not set
-CONFIG_ZONE_DMA_FLAG=0
-CONFIG_VIRT_TO_BUS=y
-# CONFIG_KSM is not set
-CONFIG_DEFAULT_MMAP_MIN_ADDR=4096
-CONFIG_NEED_PER_CPU_KM=y
-# CONFIG_CLEANCACHE is not set
-CONFIG_TICK_ONESHOT=y
-CONFIG_NO_HZ=y
-CONFIG_HIGH_RES_TIMERS=y
-CONFIG_GENERIC_CLOCKEVENTS_BUILD=y
-CONFIG_LD_SCRIPT_DYN=y
-CONFIG_BINFMT_ELF=y
-CONFIG_CORE_DUMP_DEFAULT_ELF_HEADERS=y
-CONFIG_HAVE_AOUT=y
-# CONFIG_BINFMT_AOUT is not set
-CONFIG_BINFMT_MISC=m
-CONFIG_HOSTFS=y
-# CONFIG_HPPFS is not set
-CONFIG_MCONSOLE=y
-CONFIG_MAGIC_SYSRQ=y
-CONFIG_KERNEL_STACK_ORDER=0
-# CONFIG_MMAPPER is not set
-CONFIG_NO_DMA=y
-
-#
-# General setup
-#
-CONFIG_EXPERIMENTAL=y
-CONFIG_BROKEN_ON_SMP=y
-CONFIG_INIT_ENV_ARG_LIMIT=128
-CONFIG_CROSS_COMPILE=""
-CONFIG_LOCALVERSION=""
-CONFIG_LOCALVERSION_AUTO=y
-CONFIG_DEFAULT_HOSTNAME="(none)"
-CONFIG_SWAP=y
-CONFIG_SYSVIPC=y
-CONFIG_SYSVIPC_SYSCTL=y
-CONFIG_POSIX_MQUEUE=y
-CONFIG_POSIX_MQUEUE_SYSCTL=y
-CONFIG_BSD_PROCESS_ACCT=y
-# CONFIG_BSD_PROCESS_ACCT_V3 is not set
-# CONFIG_FHANDLE is not set
-# CONFIG_TASKSTATS is not set
-# CONFIG_AUDIT is not set
-
-#
-# IRQ subsystem
-#
-CONFIG_GENERIC_IRQ_SHOW=y
-
-#
-# RCU Subsystem
-#
-CONFIG_TINY_RCU=y
-# CONFIG_PREEMPT_RCU is not set
-# CONFIG_RCU_TRACE is not set
-# CONFIG_TREE_RCU_TRACE is not set
-CONFIG_IKCONFIG=y
-CONFIG_IKCONFIG_PROC=y
-CONFIG_LOG_BUF_SHIFT=14
-CONFIG_CGROUPS=y
-# CONFIG_CGROUP_DEBUG is not set
-CONFIG_CGROUP_FREEZER=y
-CONFIG_CGROUP_DEVICE=y
-CONFIG_CPUSETS=y
-CONFIG_PROC_PID_CPUSET=y
-CONFIG_CGROUP_CPUACCT=y
-CONFIG_RESOURCE_COUNTERS=y
-CONFIG_CGROUP_MEMCG=y
-CONFIG_CGROUP_MEMCG_SWAP=y
-# CONFIG_CGROUP_MEMCG_SWAP_ENABLED is not set
-# CONFIG_CGROUP_MEMCG_KMEM is not set
-CONFIG_CGROUP_SCHED=y
-CONFIG_FAIR_GROUP_SCHED=y
-# CONFIG_CFS_BANDWIDTH is not set
-# CONFIG_RT_GROUP_SCHED is not set
-CONFIG_BLK_CGROUP=y
-# CONFIG_DEBUG_BLK_CGROUP is not set
-# CONFIG_CHECKPOINT_RESTORE is not set
-CONFIG_NAMESPACES=y
-CONFIG_UTS_NS=y
-CONFIG_IPC_NS=y
-# CONFIG_USER_NS is not set
-# CONFIG_PID_NS is not set
-CONFIG_NET_NS=y
-# CONFIG_SCHED_AUTOGROUP is not set
-CONFIG_MM_OWNER=y
-CONFIG_SYSFS_DEPRECATED=y
-# CONFIG_SYSFS_DEPRECATED_V2 is not set
-# CONFIG_RELAY is not set
-# CONFIG_BLK_DEV_INITRD is not set
-CONFIG_CC_OPTIMIZE_FOR_SIZE=y
-CONFIG_SYSCTL=y
-CONFIG_ANON_INODES=y
-# CONFIG_EXPERT is not set
-CONFIG_UID16=y
-# CONFIG_SYSCTL_SYSCALL is not set
-CONFIG_KALLSYMS=y
-# CONFIG_KALLSYMS_ALL is not set
-CONFIG_HOTPLUG=y
-CONFIG_PRINTK=y
-CONFIG_BUG=y
-CONFIG_ELF_CORE=y
-CONFIG_BASE_FULL=y
-CONFIG_FUTEX=y
-CONFIG_EPOLL=y
-CONFIG_SIGNALFD=y
-CONFIG_TIMERFD=y
-CONFIG_EVENTFD=y
-CONFIG_SHMEM=y
-CONFIG_AIO=y
-# CONFIG_EMBEDDED is not set
-
-#
-# Kernel Performance Events And Counters
-#
-CONFIG_VM_EVENT_COUNTERS=y
-CONFIG_COMPAT_BRK=y
-CONFIG_SLAB=y
-# CONFIG_SLUB is not set
-# CONFIG_PROFILING is not set
-
-#
-# GCOV-based kernel profiling
-#
-# CONFIG_HAVE_GENERIC_DMA_COHERENT is not set
-CONFIG_SLABINFO=y
-CONFIG_RT_MUTEXES=y
-CONFIG_BASE_SMALL=0
-CONFIG_MODULES=y
-# CONFIG_MODULE_FORCE_LOAD is not set
-CONFIG_MODULE_UNLOAD=y
-# CONFIG_MODULE_FORCE_UNLOAD is not set
-# CONFIG_MODVERSIONS is not set
-# CONFIG_MODULE_SRCVERSION_ALL is not set
-CONFIG_BLOCK=y
-CONFIG_LBDAF=y
-# CONFIG_BLK_DEV_BSG is not set
-# CONFIG_BLK_DEV_BSGLIB is not set
-# CONFIG_BLK_DEV_INTEGRITY is not set
-
-#
-# Partition Types
-#
-# CONFIG_PARTITION_ADVANCED is not set
-CONFIG_MSDOS_PARTITION=y
-
-#
-# IO Schedulers
-#
-CONFIG_IOSCHED_NOOP=y
-CONFIG_IOSCHED_DEADLINE=y
-CONFIG_IOSCHED_CFQ=m
-# CONFIG_CFQ_GROUP_IOSCHED is not set
-CONFIG_DEFAULT_DEADLINE=y
-# CONFIG_DEFAULT_CFQ is not set
-# CONFIG_DEFAULT_NOOP is not set
-CONFIG_DEFAULT_IOSCHED="deadline"
-# CONFIG_INLINE_SPIN_TRYLOCK is not set
-# CONFIG_INLINE_SPIN_TRYLOCK_BH is not set
-# CONFIG_INLINE_SPIN_LOCK is not set
-# CONFIG_INLINE_SPIN_LOCK_BH is not set
-# CONFIG_INLINE_SPIN_LOCK_IRQ is not set
-# CONFIG_INLINE_SPIN_LOCK_IRQSAVE is not set
-CONFIG_INLINE_SPIN_UNLOCK=y
-# CONFIG_INLINE_SPIN_UNLOCK_BH is not set
-CONFIG_INLINE_SPIN_UNLOCK_IRQ=y
-# CONFIG_INLINE_SPIN_UNLOCK_IRQRESTORE is not set
-# CONFIG_INLINE_READ_TRYLOCK is not set
-# CONFIG_INLINE_READ_LOCK is not set
-# CONFIG_INLINE_READ_LOCK_BH is not set
-# CONFIG_INLINE_READ_LOCK_IRQ is not set
-# CONFIG_INLINE_READ_LOCK_IRQSAVE is not set
-CONFIG_INLINE_READ_UNLOCK=y
-# CONFIG_INLINE_READ_UNLOCK_BH is not set
-CONFIG_INLINE_READ_UNLOCK_IRQ=y
-# CONFIG_INLINE_READ_UNLOCK_IRQRESTORE is not set
-# CONFIG_INLINE_WRITE_TRYLOCK is not set
-# CONFIG_INLINE_WRITE_LOCK is not set
-# CONFIG_INLINE_WRITE_LOCK_BH is not set
-# CONFIG_INLINE_WRITE_LOCK_IRQ is not set
-# CONFIG_INLINE_WRITE_LOCK_IRQSAVE is not set
-CONFIG_INLINE_WRITE_UNLOCK=y
-# CONFIG_INLINE_WRITE_UNLOCK_BH is not set
-CONFIG_INLINE_WRITE_UNLOCK_IRQ=y
-# CONFIG_INLINE_WRITE_UNLOCK_IRQRESTORE is not set
-# CONFIG_MUTEX_SPIN_ON_OWNER is not set
-CONFIG_FREEZER=y
-
-#
-# UML Character Devices
-#
-CONFIG_STDERR_CONSOLE=y
-CONFIG_STDIO_CONSOLE=y
-CONFIG_SSL=y
-CONFIG_NULL_CHAN=y
-CONFIG_PORT_CHAN=y
-CONFIG_PTY_CHAN=y
-CONFIG_TTY_CHAN=y
-CONFIG_XTERM_CHAN=y
-# CONFIG_NOCONFIG_CHAN is not set
-CONFIG_CON_ZERO_CHAN="fd:0,fd:1"
-CONFIG_CON_CHAN="xterm"
-CONFIG_SSL_CHAN="pts"
-CONFIG_UML_SOUND=m
-CONFIG_SOUND=m
-CONFIG_SOUND_OSS_CORE=y
-CONFIG_HOSTAUDIO=m
-
-#
-# Device Drivers
-#
-
-#
-# Generic Driver Options
-#
-CONFIG_UEVENT_HELPER_PATH="/sbin/hotplug"
-CONFIG_DEVTMPFS=y
-CONFIG_DEVTMPFS_MOUNT=y
-CONFIG_STANDALONE=y
-CONFIG_PREVENT_FIRMWARE_BUILD=y
-CONFIG_FW_LOADER=y
-CONFIG_FIRMWARE_IN_KERNEL=y
-CONFIG_EXTRA_FIRMWARE=""
-# CONFIG_DEBUG_DRIVER is not set
-# CONFIG_DEBUG_DEVRES is not set
-# CONFIG_SYS_HYPERVISOR is not set
-CONFIG_GENERIC_CPU_DEVICES=y
-# CONFIG_DMA_SHARED_BUFFER is not set
-# CONFIG_CONNECTOR is not set
-# CONFIG_MTD is not set
-CONFIG_BLK_DEV=y
-CONFIG_BLK_DEV_UBD=y
-# CONFIG_BLK_DEV_UBD_SYNC is not set
-CONFIG_BLK_DEV_COW_COMMON=y
-CONFIG_BLK_DEV_LOOP=m
-CONFIG_BLK_DEV_LOOP_MIN_COUNT=8
-# CONFIG_BLK_DEV_CRYPTOLOOP is not set
-
-#
-# DRBD disabled because PROC_FS, INET or CONNECTOR not selected
-#
-CONFIG_BLK_DEV_NBD=m
-# CONFIG_BLK_DEV_RAM is not set
-# CONFIG_ATA_OVER_ETH is not set
-# CONFIG_BLK_DEV_RBD is not set
-
-#
-# Misc devices
-#
-# CONFIG_ENCLOSURE_SERVICES is not set
-# CONFIG_C2PORT is not set
-
-#
-# EEPROM support
-#
-# CONFIG_EEPROM_93CX6 is not set
-
-#
-# Texas Instruments shared transport line discipline
-#
-
-#
-# Altera FPGA firmware download module
-#
-
-#
-# SCSI device support
-#
-CONFIG_SCSI_MOD=y
-# CONFIG_RAID_ATTRS is not set
-# CONFIG_SCSI is not set
-# CONFIG_SCSI_DMA is not set
-# CONFIG_SCSI_NETLINK is not set
-# CONFIG_MD is not set
-CONFIG_NETDEVICES=y
-CONFIG_NET_CORE=y
-# CONFIG_BONDING is not set
-CONFIG_DUMMY=m
-# CONFIG_EQUALIZER is not set
-# CONFIG_MII is not set
-# CONFIG_NET_TEAM is not set
-# CONFIG_MACVLAN is not set
-# CONFIG_NETCONSOLE is not set
-# CONFIG_NETPOLL is not set
-# CONFIG_NET_POLL_CONTROLLER is not set
-CONFIG_TUN=m
-# CONFIG_VETH is not set
-
-#
-# CAIF transport drivers
-#
-CONFIG_ETHERNET=y
-CONFIG_NET_VENDOR_CHELSIO=y
-CONFIG_NET_VENDOR_INTEL=y
-CONFIG_NET_VENDOR_I825XX=y
-CONFIG_NET_VENDOR_MARVELL=y
-CONFIG_NET_VENDOR_NATSEMI=y
-CONFIG_NET_VENDOR_8390=y
-# CONFIG_PHYLIB is not set
-CONFIG_PPP=m
-# CONFIG_PPP_BSDCOMP is not set
-# CONFIG_PPP_DEFLATE is not set
-# CONFIG_PPP_FILTER is not set
-# CONFIG_PPP_MPPE is not set
-# CONFIG_PPP_MULTILINK is not set
-# CONFIG_PPPOE is not set
-# CONFIG_PPP_ASYNC is not set
-# CONFIG_PPP_SYNC_TTY is not set
-CONFIG_SLIP=m
-CONFIG_SLHC=m
-# CONFIG_SLIP_COMPRESSED is not set
-# CONFIG_SLIP_SMART is not set
-# CONFIG_SLIP_MODE_SLIP6 is not set
-CONFIG_WLAN=y
-# CONFIG_HOSTAP is not set
-
-#
-# Enable WiMAX (Networking options) to see the WiMAX drivers
-#
-# CONFIG_WAN is not set
-
-#
-# Character devices
-#
-CONFIG_UNIX98_PTYS=y
-# CONFIG_DEVPTS_MULTIPLE_INSTANCES is not set
-CONFIG_LEGACY_PTYS=y
-CONFIG_LEGACY_PTY_COUNT=32
-# CONFIG_N_GSM is not set
-# CONFIG_TRACE_SINK is not set
-CONFIG_DEVKMEM=y
-# CONFIG_HW_RANDOM is not set
-CONFIG_UML_RANDOM=y
-# CONFIG_R3964 is not set
-# CONFIG_NSC_GPIO is not set
-# CONFIG_RAW_DRIVER is not set
-
-#
-# PPS support
-#
-# CONFIG_PPS is not set
-
-#
-# PPS generators support
-#
-
-#
-# PTP clock support
-#
-
-#
-# Enable Device Drivers -> PPS to see the PTP clock options.
-#
-# CONFIG_POWER_SUPPLY is not set
-# CONFIG_THERMAL is not set
-# CONFIG_WATCHDOG is not set
-# CONFIG_REGULATOR is not set
-CONFIG_SOUND_OSS_CORE_PRECLAIM=y
-# CONFIG_MEMSTICK is not set
-# CONFIG_NEW_LEDS is not set
-# CONFIG_ACCESSIBILITY is not set
-# CONFIG_AUXDISPLAY is not set
-# CONFIG_UIO is not set
-
-#
-# Virtio drivers
-#
-# CONFIG_VIRTIO_BALLOON is not set
-
-#
-# Microsoft Hyper-V guest support
-#
-# CONFIG_STAGING is not set
-
-#
-# Hardware Spinlock drivers
-#
-CONFIG_IOMMU_SUPPORT=y
-# CONFIG_VIRT_DRIVERS is not set
-# CONFIG_PM_DEVFREQ is not set
-CONFIG_NET=y
-
-#
-# Networking options
-#
-CONFIG_PACKET=y
-CONFIG_UNIX=y
-# CONFIG_UNIX_DIAG is not set
-CONFIG_XFRM=y
-# CONFIG_XFRM_USER is not set
-# CONFIG_XFRM_SUB_POLICY is not set
-# CONFIG_XFRM_MIGRATE is not set
-# CONFIG_XFRM_STATISTICS is not set
-# CONFIG_NET_KEY is not set
-CONFIG_INET=y
-# CONFIG_IP_MULTICAST is not set
-# CONFIG_IP_ADVANCED_ROUTER is not set
-# CONFIG_IP_PNP is not set
-# CONFIG_NET_IPIP is not set
-# CONFIG_NET_IPGRE_DEMUX is not set
-# CONFIG_ARPD is not set
-# CONFIG_SYN_COOKIES is not set
-# CONFIG_INET_AH is not set
-# CONFIG_INET_ESP is not set
-# CONFIG_INET_IPCOMP is not set
-# CONFIG_INET_XFRM_TUNNEL is not set
-# CONFIG_INET_TUNNEL is not set
-CONFIG_INET_XFRM_MODE_TRANSPORT=y
-CONFIG_INET_XFRM_MODE_TUNNEL=y
-CONFIG_INET_XFRM_MODE_BEET=y
-# CONFIG_INET_LRO is not set
-CONFIG_INET_DIAG=y
-CONFIG_INET_TCP_DIAG=y
-# CONFIG_INET_UDP_DIAG is not set
-# CONFIG_TCP_CONG_ADVANCED is not set
-CONFIG_TCP_CONG_CUBIC=y
-CONFIG_DEFAULT_TCP_CONG="cubic"
-# CONFIG_TCP_MD5SIG is not set
-# CONFIG_IPV6 is not set
-# CONFIG_NETWORK_SECMARK is not set
-# CONFIG_NETWORK_PHY_TIMESTAMPING is not set
-# CONFIG_NETFILTER is not set
-# CONFIG_IP_DCCP is not set
-# CONFIG_IP_SCTP is not set
-# CONFIG_RDS is not set
-# CONFIG_TIPC is not set
-# CONFIG_ATM is not set
-# CONFIG_L2TP is not set
-# CONFIG_BRIDGE is not set
-# CONFIG_NET_DSA is not set
-# CONFIG_VLAN_8021Q is not set
-# CONFIG_DECNET is not set
-# CONFIG_LLC2 is not set
-# CONFIG_IPX is not set
-# CONFIG_ATALK is not set
-# CONFIG_X25 is not set
-# CONFIG_LAPB is not set
-# CONFIG_ECONET is not set
-# CONFIG_WAN_ROUTER is not set
-# CONFIG_PHONET is not set
-# CONFIG_IEEE802154 is not set
-# CONFIG_NET_SCHED is not set
-# CONFIG_DCB is not set
-# CONFIG_BATMAN_ADV is not set
-# CONFIG_OPENVSWITCH is not set
-# CONFIG_NETPRIO_CGROUP is not set
-CONFIG_BQL=y
-
-#
-# Network testing
-#
-# CONFIG_NET_PKTGEN is not set
-# CONFIG_HAMRADIO is not set
-# CONFIG_CAN is not set
-# CONFIG_IRDA is not set
-# CONFIG_BT is not set
-# CONFIG_AF_RXRPC is not set
-CONFIG_WIRELESS=y
-# CONFIG_CFG80211 is not set
-# CONFIG_LIB80211 is not set
-
-#
-# CFG80211 needs to be enabled for MAC80211
-#
-# CONFIG_WIMAX is not set
-# CONFIG_RFKILL is not set
-# CONFIG_NET_9P is not set
-# CONFIG_CAIF is not set
-# CONFIG_CEPH_LIB is not set
-# CONFIG_NFC is not set
-
-#
-# UML Network Devices
-#
-CONFIG_UML_NET=y
-CONFIG_UML_NET_ETHERTAP=y
-CONFIG_UML_NET_TUNTAP=y
-CONFIG_UML_NET_SLIP=y
-CONFIG_UML_NET_DAEMON=y
-# CONFIG_UML_NET_VDE is not set
-CONFIG_UML_NET_MCAST=y
-# CONFIG_UML_NET_PCAP is not set
-CONFIG_UML_NET_SLIRP=y
-
-#
-# File systems
-#
-# CONFIG_EXT2_FS is not set
-# CONFIG_EXT3_FS is not set
-CONFIG_EXT4_FS=y
-CONFIG_EXT4_USE_FOR_EXT23=y
-CONFIG_EXT4_FS_XATTR=y
-# CONFIG_EXT4_FS_POSIX_ACL is not set
-# CONFIG_EXT4_FS_SECURITY is not set
-# CONFIG_EXT4_DEBUG is not set
-CONFIG_JBD2=y
-CONFIG_FS_MBCACHE=y
-CONFIG_REISERFS_FS=y
-# CONFIG_REISERFS_CHECK is not set
-# CONFIG_REISERFS_PROC_INFO is not set
-# CONFIG_REISERFS_FS_XATTR is not set
-# CONFIG_JFS_FS is not set
-# CONFIG_XFS_FS is not set
-# CONFIG_GFS2_FS is not set
-# CONFIG_BTRFS_FS is not set
-# CONFIG_NILFS2_FS is not set
-# CONFIG_FS_POSIX_ACL is not set
-CONFIG_FILE_LOCKING=y
-CONFIG_FSNOTIFY=y
-CONFIG_DNOTIFY=y
-CONFIG_INOTIFY_USER=y
-# CONFIG_FANOTIFY is not set
-CONFIG_QUOTA=y
-# CONFIG_QUOTA_NETLINK_INTERFACE is not set
-CONFIG_PRINT_QUOTA_WARNING=y
-# CONFIG_QUOTA_DEBUG is not set
-# CONFIG_QFMT_V1 is not set
-# CONFIG_QFMT_V2 is not set
-CONFIG_QUOTACTL=y
-CONFIG_AUTOFS4_FS=m
-# CONFIG_FUSE_FS is not set
-
-#
-# Caches
-#
-# CONFIG_FSCACHE is not set
-
-#
-# CD-ROM/DVD Filesystems
-#
-CONFIG_ISO9660_FS=m
-CONFIG_JOLIET=y
-# CONFIG_ZISOFS is not set
-# CONFIG_UDF_FS is not set
-
-#
-# DOS/FAT/NT Filesystems
-#
-# CONFIG_MSDOS_FS is not set
-# CONFIG_VFAT_FS is not set
-# CONFIG_NTFS_FS is not set
-
-#
-# Pseudo filesystems
-#
-CONFIG_PROC_FS=y
-CONFIG_PROC_KCORE=y
-CONFIG_PROC_SYSCTL=y
-CONFIG_PROC_PAGE_MONITOR=y
-CONFIG_SYSFS=y
-CONFIG_TMPFS=y
-# CONFIG_TMPFS_POSIX_ACL is not set
-# CONFIG_TMPFS_XATTR is not set
-# CONFIG_HUGETLB_PAGE is not set
-# CONFIG_CONFIGFS_FS is not set
-CONFIG_MISC_FILESYSTEMS=y
-# CONFIG_ADFS_FS is not set
-# CONFIG_AFFS_FS is not set
-# CONFIG_HFS_FS is not set
-# CONFIG_HFSPLUS_FS is not set
-# CONFIG_BEFS_FS is not set
-# CONFIG_BFS_FS is not set
-# CONFIG_EFS_FS is not set
-# CONFIG_LOGFS is not set
-# CONFIG_CRAMFS is not set
-# CONFIG_SQUASHFS is not set
-# CONFIG_VXFS_FS is not set
-# CONFIG_MINIX_FS is not set
-# CONFIG_OMFS_FS is not set
-# CONFIG_HPFS_FS is not set
-# CONFIG_QNX4FS_FS is not set
-# CONFIG_ROMFS_FS is not set
-# CONFIG_PSTORE is not set
-# CONFIG_SYSV_FS is not set
-# CONFIG_UFS_FS is not set
-CONFIG_NETWORK_FILESYSTEMS=y
-# CONFIG_NFS_FS is not set
-# CONFIG_NFSD is not set
-# CONFIG_CEPH_FS is not set
-# CONFIG_CIFS is not set
-# CONFIG_NCP_FS is not set
-# CONFIG_CODA_FS is not set
-# CONFIG_AFS_FS is not set
-CONFIG_NLS=y
-CONFIG_NLS_DEFAULT="iso8859-1"
-# CONFIG_NLS_CODEPAGE_437 is not set
-# CONFIG_NLS_CODEPAGE_737 is not set
-# CONFIG_NLS_CODEPAGE_775 is not set
-# CONFIG_NLS_CODEPAGE_850 is not set
-# CONFIG_NLS_CODEPAGE_852 is not set
-# CONFIG_NLS_CODEPAGE_855 is not set
-# CONFIG_NLS_CODEPAGE_857 is not set
-# CONFIG_NLS_CODEPAGE_860 is not set
-# CONFIG_NLS_CODEPAGE_861 is not set
-# CONFIG_NLS_CODEPAGE_862 is not set
-# CONFIG_NLS_CODEPAGE_863 is not set
-# CONFIG_NLS_CODEPAGE_864 is not set
-# CONFIG_NLS_CODEPAGE_865 is not set
-# CONFIG_NLS_CODEPAGE_866 is not set
-# CONFIG_NLS_CODEPAGE_869 is not set
-# CONFIG_NLS_CODEPAGE_936 is not set
-# CONFIG_NLS_CODEPAGE_950 is not set
-# CONFIG_NLS_CODEPAGE_932 is not set
-# CONFIG_NLS_CODEPAGE_949 is not set
-# CONFIG_NLS_CODEPAGE_874 is not set
-# CONFIG_NLS_ISO8859_8 is not set
-# CONFIG_NLS_CODEPAGE_1250 is not set
-# CONFIG_NLS_CODEPAGE_1251 is not set
-# CONFIG_NLS_ASCII is not set
-# CONFIG_NLS_ISO8859_1 is not set
-# CONFIG_NLS_ISO8859_2 is not set
-# CONFIG_NLS_ISO8859_3 is not set
-# CONFIG_NLS_ISO8859_4 is not set
-# CONFIG_NLS_ISO8859_5 is not set
-# CONFIG_NLS_ISO8859_6 is not set
-# CONFIG_NLS_ISO8859_7 is not set
-# CONFIG_NLS_ISO8859_9 is not set
-# CONFIG_NLS_ISO8859_13 is not set
-# CONFIG_NLS_ISO8859_14 is not set
-# CONFIG_NLS_ISO8859_15 is not set
-# CONFIG_NLS_KOI8_R is not set
-# CONFIG_NLS_KOI8_U is not set
-# CONFIG_NLS_UTF8 is not set
-
-#
-# Security options
-#
-# CONFIG_KEYS is not set
-# CONFIG_SECURITY_DMESG_RESTRICT is not set
-# CONFIG_SECURITY is not set
-# CONFIG_SECURITYFS is not set
-CONFIG_DEFAULT_SECURITY_DAC=y
-CONFIG_DEFAULT_SECURITY=""
-CONFIG_CRYPTO=y
-
-#
-# Crypto core or helper
-#
-# CONFIG_CRYPTO_FIPS is not set
-CONFIG_CRYPTO_ALGAPI=m
-CONFIG_CRYPTO_ALGAPI2=m
-CONFIG_CRYPTO_RNG=m
-CONFIG_CRYPTO_RNG2=m
-# CONFIG_CRYPTO_MANAGER is not set
-# CONFIG_CRYPTO_MANAGER2 is not set
-# CONFIG_CRYPTO_USER is not set
-# CONFIG_CRYPTO_GF128MUL is not set
-# CONFIG_CRYPTO_NULL is not set
-# CONFIG_CRYPTO_CRYPTD is not set
-# CONFIG_CRYPTO_AUTHENC is not set
-# CONFIG_CRYPTO_TEST is not set
-
-#
-# Authenticated Encryption with Associated Data
-#
-# CONFIG_CRYPTO_CCM is not set
-# CONFIG_CRYPTO_GCM is not set
-# CONFIG_CRYPTO_SEQIV is not set
-
-#
-# Block modes
-#
-# CONFIG_CRYPTO_CBC is not set
-# CONFIG_CRYPTO_CTR is not set
-# CONFIG_CRYPTO_CTS is not set
-# CONFIG_CRYPTO_ECB is not set
-# CONFIG_CRYPTO_LRW is not set
-# CONFIG_CRYPTO_PCBC is not set
-# CONFIG_CRYPTO_XTS is not set
-
-#
-# Hash modes
-#
-# CONFIG_CRYPTO_HMAC is not set
-# CONFIG_CRYPTO_XCBC is not set
-# CONFIG_CRYPTO_VMAC is not set
-
-#
-# Digest
-#
-# CONFIG_CRYPTO_CRC32C is not set
-# CONFIG_CRYPTO_GHASH is not set
-# CONFIG_CRYPTO_MD4 is not set
-# CONFIG_CRYPTO_MD5 is not set
-# CONFIG_CRYPTO_MICHAEL_MIC is not set
-# CONFIG_CRYPTO_RMD128 is not set
-# CONFIG_CRYPTO_RMD160 is not set
-# CONFIG_CRYPTO_RMD256 is not set
-# CONFIG_CRYPTO_RMD320 is not set
-# CONFIG_CRYPTO_SHA1 is not set
-# CONFIG_CRYPTO_SHA256 is not set
-# CONFIG_CRYPTO_SHA512 is not set
-# CONFIG_CRYPTO_TGR192 is not set
-# CONFIG_CRYPTO_WP512 is not set
-
-#
-# Ciphers
-#
-CONFIG_CRYPTO_AES=m
-# CONFIG_CRYPTO_AES_586 is not set
-# CONFIG_CRYPTO_ANUBIS is not set
-# CONFIG_CRYPTO_ARC4 is not set
-# CONFIG_CRYPTO_BLOWFISH is not set
-# CONFIG_CRYPTO_CAMELLIA is not set
-# CONFIG_CRYPTO_CAST5 is not set
-# CONFIG_CRYPTO_CAST6 is not set
-# CONFIG_CRYPTO_DES is not set
-# CONFIG_CRYPTO_FCRYPT is not set
-# CONFIG_CRYPTO_KHAZAD is not set
-# CONFIG_CRYPTO_SALSA20 is not set
-# CONFIG_CRYPTO_SALSA20_586 is not set
-# CONFIG_CRYPTO_SEED is not set
-# CONFIG_CRYPTO_SERPENT is not set
-# CONFIG_CRYPTO_TEA is not set
-# CONFIG_CRYPTO_TWOFISH is not set
-# CONFIG_CRYPTO_TWOFISH_586 is not set
-
-#
-# Compression
-#
-# CONFIG_CRYPTO_DEFLATE is not set
-# CONFIG_CRYPTO_ZLIB is not set
-# CONFIG_CRYPTO_LZO is not set
-
-#
-# Random Number Generation
-#
-CONFIG_CRYPTO_ANSI_CPRNG=m
-# CONFIG_CRYPTO_USER_API_HASH is not set
-# CONFIG_CRYPTO_USER_API_SKCIPHER is not set
-CONFIG_CRYPTO_HW=y
-# CONFIG_BINARY_PRINTF is not set
-
-#
-# Library routines
-#
-CONFIG_BITREVERSE=y
-CONFIG_GENERIC_FIND_FIRST_BIT=y
-CONFIG_GENERIC_IO=y
-# CONFIG_CRC_CCITT is not set
-CONFIG_CRC16=y
-# CONFIG_CRC_T10DIF is not set
-# CONFIG_CRC_ITU_T is not set
-CONFIG_CRC32=y
-# CONFIG_CRC7 is not set
-# CONFIG_LIBCRC32C is not set
-# CONFIG_CRC8 is not set
-# CONFIG_XZ_DEC is not set
-# CONFIG_XZ_DEC_BCJ is not set
-CONFIG_DQL=y
-CONFIG_NLATTR=y
-# CONFIG_AVERAGE is not set
-# CONFIG_CORDIC is not set
-
-#
-# Kernel hacking
-#
-# CONFIG_PRINTK_TIME is not set
-CONFIG_DEFAULT_MESSAGE_LOGLEVEL=4
-CONFIG_ENABLE_WARN_DEPRECATED=y
-CONFIG_ENABLE_MUST_CHECK=y
-CONFIG_FRAME_WARN=1024
-# CONFIG_STRIP_ASM_SYMS is not set
-# CONFIG_UNUSED_SYMBOLS is not set
-# CONFIG_DEBUG_FS is not set
-# CONFIG_DEBUG_SECTION_MISMATCH is not set
-CONFIG_DEBUG_KERNEL=y
-# CONFIG_DEBUG_SHIRQ is not set
-# CONFIG_LOCKUP_DETECTOR is not set
-# CONFIG_HARDLOCKUP_DETECTOR is not set
-# CONFIG_DETECT_HUNG_TASK is not set
-CONFIG_SCHED_DEBUG=y
-# CONFIG_SCHEDSTATS is not set
-# CONFIG_TIMER_STATS is not set
-# CONFIG_DEBUG_OBJECTS is not set
-# CONFIG_DEBUG_SLAB is not set
-# CONFIG_DEBUG_RT_MUTEXES is not set
-# CONFIG_RT_MUTEX_TESTER is not set
-# CONFIG_DEBUG_SPINLOCK is not set
-# CONFIG_DEBUG_MUTEXES is not set
-# CONFIG_SPARSE_RCU_POINTER is not set
-# CONFIG_DEBUG_ATOMIC_SLEEP is not set
-# CONFIG_DEBUG_LOCKING_API_SELFTESTS is not set
-# CONFIG_DEBUG_STACK_USAGE is not set
-# CONFIG_DEBUG_KOBJECT is not set
-CONFIG_DEBUG_BUGVERBOSE=y
-CONFIG_DEBUG_INFO=y
-# CONFIG_DEBUG_INFO_REDUCED is not set
-# CONFIG_DEBUG_VM is not set
-# CONFIG_DEBUG_WRITECOUNT is not set
-CONFIG_DEBUG_MEMORY_INIT=y
-# CONFIG_DEBUG_LIST is not set
-# CONFIG_TEST_LIST_SORT is not set
-# CONFIG_DEBUG_SG is not set
-# CONFIG_DEBUG_NOTIFIERS is not set
-# CONFIG_DEBUG_CREDENTIALS is not set
-CONFIG_FRAME_POINTER=y
-# CONFIG_BOOT_PRINTK_DELAY is not set
-# CONFIG_RCU_TORTURE_TEST is not set
-# CONFIG_BACKTRACE_SELF_TEST is not set
-# CONFIG_DEBUG_BLOCK_EXT_DEVT is not set
-# CONFIG_DEBUG_FORCE_WEAK_PER_CPU is not set
-# CONFIG_FAULT_INJECTION is not set
-# CONFIG_SYSCTL_SYSCALL_CHECK is not set
-# CONFIG_DEBUG_PAGEALLOC is not set
-# CONFIG_ATOMIC64_SELFTEST is not set
-# CONFIG_SAMPLES is not set
-# CONFIG_TEST_KSTRTOX is not set
-# CONFIG_GPROF is not set
-# CONFIG_GCOV is not set
-CONFIG_EARLY_PRINTK=y
diff --git a/arch/um/drivers/mconsole_kern.c b/arch/um/drivers/mconsole_kern.c
index 3df3bd544492..29880c9b324e 100644
--- a/arch/um/drivers/mconsole_kern.c
+++ b/arch/um/drivers/mconsole_kern.c
@@ -645,11 +645,9 @@ void mconsole_sysrq(struct mc_request *req)
 
 static void stack_proc(void *arg)
 {
-	struct task_struct *from = current, *to = arg;
+	struct task_struct *task = arg;
 
-	to->thread.saved_task = from;
-	rcu_user_hooks_switch(from, to);
-	switch_to(from, to, from);
+	show_stack(task, NULL);
 }
 
 /*
diff --git a/arch/um/include/asm/processor-generic.h b/arch/um/include/asm/processor-generic.h
index c03cd5a02364..d89b02bb6262 100644
--- a/arch/um/include/asm/processor-generic.h
+++ b/arch/um/include/asm/processor-generic.h
@@ -19,8 +19,8 @@ struct task_struct;
 struct mm_struct;
 
 struct thread_struct {
-	struct task_struct *saved_task;
 	struct pt_regs regs;
+	struct pt_regs *segv_regs;
 	int singlestep_syscall;
 	void *fault_addr;
 	jmp_buf *fault_catcher;
diff --git a/arch/um/include/asm/thread_info.h b/arch/um/include/asm/thread_info.h
index 2c8eeb2df8b4..1c5b2a83046a 100644
--- a/arch/um/include/asm/thread_info.h
+++ b/arch/um/include/asm/thread_info.h
@@ -60,8 +60,6 @@ static inline struct thread_info *current_thread_info(void)
 
 #endif
 
-#define PREEMPT_ACTIVE		0x10000000
-
 #define TIF_SYSCALL_TRACE	0	/* syscall trace active */
 #define TIF_SIGPENDING		1	/* signal pending */
 #define TIF_NEED_RESCHED	2	/* rescheduling necessary */
diff --git a/arch/um/include/shared/as-layout.h b/arch/um/include/shared/as-layout.h
index 694c792bab4e..41c8c774ec10 100644
--- a/arch/um/include/shared/as-layout.h
+++ b/arch/um/include/shared/as-layout.h
@@ -44,7 +44,6 @@ struct cpu_task {
 
 extern struct cpu_task cpu_tasks[];
 
-extern unsigned long low_physmem;
 extern unsigned long high_physmem;
 extern unsigned long uml_physmem;
 extern unsigned long uml_reserved;
@@ -52,8 +51,6 @@ extern unsigned long end_vm;
 extern unsigned long start_vm;
 extern unsigned long long highmem;
 
-extern unsigned long _stext, _etext, _sdata, _edata, __bss_start, _end;
-extern unsigned long _unprotected_end;
 extern unsigned long brk_start;
 
 extern unsigned long host_task_size;
diff --git a/arch/um/include/shared/os.h b/arch/um/include/shared/os.h
index 021104d98cb3..75298d3358e7 100644
--- a/arch/um/include/shared/os.h
+++ b/arch/um/include/shared/os.h
@@ -227,6 +227,7 @@ extern void block_signals(void);
 extern void unblock_signals(void);
 extern int get_signals(void);
 extern int set_signals(int enable);
+extern int os_is_signal_stack(void);
 
 /* util.c */
 extern void stack_protections(unsigned long address);
diff --git a/arch/um/kernel/mem.c b/arch/um/kernel/mem.c
index 7ddb64baf327..8636e905426f 100644
--- a/arch/um/kernel/mem.c
+++ b/arch/um/kernel/mem.c
@@ -279,8 +279,12 @@ pgtable_t pte_alloc_one(struct mm_struct *mm, unsigned long address)
 	struct page *pte;
 
 	pte = alloc_page(GFP_KERNEL|__GFP_REPEAT|__GFP_ZERO);
-	if (pte)
-		pgtable_page_ctor(pte);
+	if (!pte)
+		return NULL;
+	if (!pgtable_page_ctor(pte)) {
+		__free_page(pte);
+		return NULL;
+	}
 	return pte;
 }
 
diff --git a/arch/um/kernel/process.c b/arch/um/kernel/process.c
index bbcef522bcb1..eecc4142764c 100644
--- a/arch/um/kernel/process.c
+++ b/arch/um/kernel/process.c
@@ -82,19 +82,8 @@ void *__switch_to(struct task_struct *from, struct task_struct *to)
 	to->thread.prev_sched = from;
 	set_current(to);
 
-	do {
-		current->thread.saved_task = NULL;
-
-		switch_threads(&from->thread.switch_buf,
-			       &to->thread.switch_buf);
-
-		arch_switch_to(current);
-
-		if (current->thread.saved_task)
-			show_regs(&(current->thread.regs));
-		to = current->thread.saved_task;
-		from = current;
-	} while (current->thread.saved_task);
+	switch_threads(&from->thread.switch_buf, &to->thread.switch_buf);
+	arch_switch_to(current);
 
 	return current->thread.prev_sched;
 }
diff --git a/arch/um/kernel/sysrq.c b/arch/um/kernel/sysrq.c
index 0dc4d1c6f98a..4d6fdf68edf3 100644
--- a/arch/um/kernel/sysrq.c
+++ b/arch/um/kernel/sysrq.c
@@ -1,6 +1,10 @@
 /*
  * Copyright (C) 2001 - 2007 Jeff Dike (jdike@{addtoit,linux.intel}.com)
- * Licensed under the GPL
+ * Copyright (C) 2013 Richard Weinberger <richrd@nod.at>
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
  */
 
 #include <linux/kallsyms.h>
@@ -8,59 +12,87 @@
 #include <linux/module.h>
 #include <linux/sched.h>
 #include <asm/sysrq.h>
+#include <os.h>
 
-/* Catch non-i386 SUBARCH's. */
-#if !defined(CONFIG_UML_X86) || defined(CONFIG_64BIT)
-void show_trace(struct task_struct *task, unsigned long * stack)
+struct stack_frame {
+	struct stack_frame *next_frame;
+	unsigned long return_address;
+};
+
+static void print_stack_trace(unsigned long *sp, unsigned long bp)
 {
+	int reliable;
 	unsigned long addr;
+	struct stack_frame *frame = (struct stack_frame *)bp;
 
-	if (!stack) {
-		stack = (unsigned long*) &stack;
-		WARN_ON(1);
-	}
-
-	printk(KERN_INFO "Call Trace: \n");
-	while (((long) stack & (THREAD_SIZE-1)) != 0) {
-		addr = *stack;
+	printk(KERN_INFO "Call Trace:\n");
+	while (((long) sp & (THREAD_SIZE-1)) != 0) {
+		addr = *sp;
 		if (__kernel_text_address(addr)) {
-			printk(KERN_INFO "%08lx:  [<%08lx>]",
-			       (unsigned long) stack, addr);
-			print_symbol(KERN_CONT " %s", addr);
+			reliable = 0;
+			if ((unsigned long) sp == bp + sizeof(long)) {
+				frame = frame ? frame->next_frame : NULL;
+				bp = (unsigned long)frame;
+				reliable = 1;
+			}
+
+			printk(KERN_INFO " [<%08lx>]", addr);
+			printk(KERN_CONT " %s", reliable ? "" : "? ");
+			print_symbol(KERN_CONT "%s", addr);
 			printk(KERN_CONT "\n");
 		}
-		stack++;
+		sp++;
 	}
 	printk(KERN_INFO "\n");
 }
-#endif
 
-/*Stolen from arch/i386/kernel/traps.c */
-static const int kstack_depth_to_print = 24;
+static unsigned long get_frame_pointer(struct task_struct *task,
+				       struct pt_regs *segv_regs)
+{
+	if (!task || task == current)
+		return segv_regs ? PT_REGS_BP(segv_regs) : current_bp();
+	else
+		return KSTK_EBP(task);
+}
 
-/* This recently started being used in arch-independent code too, as in
- * kernel/sched/core.c.*/
-void show_stack(struct task_struct *task, unsigned long *esp)
+static unsigned long *get_stack_pointer(struct task_struct *task,
+					struct pt_regs *segv_regs)
 {
-	unsigned long *stack;
+	if (!task || task == current)
+		return segv_regs ? (unsigned long *)PT_REGS_SP(segv_regs) : current_sp();
+	else
+		return (unsigned long *)KSTK_ESP(task);
+}
+
+void show_stack(struct task_struct *task, unsigned long *stack)
+{
+	unsigned long *sp = stack, bp = 0;
+	struct pt_regs *segv_regs = current->thread.segv_regs;
 	int i;
 
-	if (esp == NULL) {
-		if (task != current && task != NULL) {
-			esp = (unsigned long *) KSTK_ESP(task);
-		} else {
-			esp = (unsigned long *) &esp;
-		}
+	if (!segv_regs && os_is_signal_stack()) {
+		printk(KERN_ERR "Received SIGSEGV in SIGSEGV handler,"
+				" aborting stack trace!\n");
+		return;
 	}
 
-	stack = esp;
-	for (i = 0; i < kstack_depth_to_print; i++) {
+#ifdef CONFIG_FRAME_POINTER
+	bp = get_frame_pointer(task, segv_regs);
+#endif
+
+	if (!stack)
+		sp = get_stack_pointer(task, segv_regs);
+
+	printk(KERN_INFO "Stack:\n");
+	stack = sp;
+	for (i = 0; i < 3 * STACKSLOTS_PER_LINE; i++) {
 		if (kstack_end(stack))
 			break;
-		if (i && ((i % 8) == 0))
-			printk(KERN_INFO "       ");
-		printk(KERN_CONT "%08lx ", *stack++);
+		if (i && ((i % STACKSLOTS_PER_LINE) == 0))
+			printk(KERN_CONT "\n");
+		printk(KERN_CONT " %08lx", *stack++);
 	}
+	printk(KERN_CONT "\n");
 
-	show_trace(task, esp);
+	print_stack_trace(sp, bp);
 }
diff --git a/arch/um/kernel/trap.c b/arch/um/kernel/trap.c
index 5c3aef74237f..974b87474a99 100644
--- a/arch/um/kernel/trap.c
+++ b/arch/um/kernel/trap.c
@@ -206,9 +206,12 @@ unsigned long segv(struct faultinfo fi, unsigned long ip, int is_user,
 	int is_write = FAULT_WRITE(fi);
 	unsigned long address = FAULT_ADDRESS(fi);
 
+	if (regs)
+		current->thread.segv_regs = container_of(regs, struct pt_regs, regs);
+
 	if (!is_user && (address >= start_vm) && (address < end_vm)) {
 		flush_tlb_kernel_vm();
-		return 0;
+		goto out;
 	}
 	else if (current->mm == NULL) {
 		show_regs(container_of(regs, struct pt_regs, regs));
@@ -230,7 +233,7 @@ unsigned long segv(struct faultinfo fi, unsigned long ip, int is_user,
 
 	catcher = current->thread.fault_catcher;
 	if (!err)
-		return 0;
+		goto out;
 	else if (catcher != NULL) {
 		current->thread.fault_addr = (void *) address;
 		UML_LONGJMP(catcher, 1);
@@ -238,7 +241,7 @@ unsigned long segv(struct faultinfo fi, unsigned long ip, int is_user,
 	else if (current->thread.fault_addr != NULL)
 		panic("fault_addr set but no fault catcher");
 	else if (!is_user && arch_fixup(ip, regs))
-		return 0;
+		goto out;
 
 	if (!is_user) {
 		show_regs(container_of(regs, struct pt_regs, regs));
@@ -262,6 +265,11 @@ unsigned long segv(struct faultinfo fi, unsigned long ip, int is_user,
 		current->thread.arch.faultinfo = fi;
 		force_sig_info(SIGSEGV, &si, current);
 	}
+
+out:
+	if (regs)
+		current->thread.segv_regs = NULL;
+
 	return 0;
 }
 
diff --git a/arch/um/kernel/um_arch.c b/arch/um/kernel/um_arch.c
index 87df5e3acc26..016adf0985d5 100644
--- a/arch/um/kernel/um_arch.c
+++ b/arch/um/kernel/um_arch.c
@@ -13,6 +13,7 @@
 #include <linux/sched.h>
 #include <asm/pgtable.h>
 #include <asm/processor.h>
+#include <asm/sections.h>
 #include <asm/setup.h>
 #include <as-layout.h>
 #include <arch.h>
@@ -234,7 +235,6 @@ static int panic_exit(struct notifier_block *self, unsigned long unused1,
 		      void *unused2)
 {
 	bust_spinlocks(1);
-	show_regs(&(current->thread.regs));
 	bust_spinlocks(0);
 	uml_exitcode = 1;
 	os_dump_core();
diff --git a/arch/um/os-Linux/signal.c b/arch/um/os-Linux/signal.c
index 905924b773d3..7b605e4dfffa 100644
--- a/arch/um/os-Linux/signal.c
+++ b/arch/um/os-Linux/signal.c
@@ -304,3 +304,11 @@ int set_signals(int enable)
 
 	return ret;
 }
+
+int os_is_signal_stack(void)
+{
+	stack_t ss;
+	sigaltstack(NULL, &ss);
+
+	return ss.ss_flags & SS_ONSTACK;
+}
diff --git a/arch/unicore32/Kconfig b/arch/unicore32/Kconfig
index 82cdd8906f3d..a7ba27b2752b 100644
--- a/arch/unicore32/Kconfig
+++ b/arch/unicore32/Kconfig
@@ -1,5 +1,6 @@
 config UNICORE32
 	def_bool y
+	select ARCH_MIGHT_HAVE_PC_PARPORT
 	select HAVE_MEMBLOCK
 	select HAVE_GENERIC_DMA_COHERENT
 	select HAVE_DMA_ATTRS
diff --git a/arch/unicore32/include/asm/pgalloc.h b/arch/unicore32/include/asm/pgalloc.h
index 0213e373a895..2e02d1356fdf 100644
--- a/arch/unicore32/include/asm/pgalloc.h
+++ b/arch/unicore32/include/asm/pgalloc.h
@@ -51,12 +51,14 @@ pte_alloc_one(struct mm_struct *mm, unsigned long addr)
 	struct page *pte;
 
 	pte = alloc_pages(PGALLOC_GFP, 0);
-	if (pte) {
-		if (!PageHighMem(pte)) {
-			void *page = page_address(pte);
-			clean_dcache_area(page, PTRS_PER_PTE * sizeof(pte_t));
-		}
-		pgtable_page_ctor(pte);
+	if (!pte)
+		return NULL;
+	if (!PageHighMem(pte)) {
+		void *page = page_address(pte);
+		clean_dcache_area(page, PTRS_PER_PTE * sizeof(pte_t));
+	}
+	if (!pgtable_page_ctor(pte)) {
+		__free_page(pte);
 	}
 
 	return pte;
diff --git a/arch/unicore32/include/asm/thread_info.h b/arch/unicore32/include/asm/thread_info.h
index 818b4a1edb5b..af36d8eabdf1 100644
--- a/arch/unicore32/include/asm/thread_info.h
+++ b/arch/unicore32/include/asm/thread_info.h
@@ -118,12 +118,6 @@ static inline struct thread_info *current_thread_info(void)
 #endif
 
 /*
- * We use bit 30 of the preempt_count to indicate that kernel
- * preemption is occurring.  See <asm/hardirq.h>.
- */
-#define PREEMPT_ACTIVE	0x40000000
-
-/*
  * thread information flags:
  *  TIF_SYSCALL_TRACE	- syscall trace active
  *  TIF_SIGPENDING	- signal pending
diff --git a/arch/unicore32/kernel/puv3-nb0916.c b/arch/unicore32/kernel/puv3-nb0916.c
index 181108b8ecce..0c6618e71897 100644
--- a/arch/unicore32/kernel/puv3-nb0916.c
+++ b/arch/unicore32/kernel/puv3-nb0916.c
@@ -54,6 +54,7 @@ static struct platform_pwm_backlight_data nb0916_backlight_data = {
 	.max_brightness	= 100,
 	.dft_brightness	= 100,
 	.pwm_period_ns	= 70 * 1024,
+	.enable_gpio	= -1,
 };
 
 static struct gpio_keys_button nb0916_gpio_keys[] = {
diff --git a/arch/x86/Kconfig b/arch/x86/Kconfig
index 14dc9c797abb..e903c71f7e69 100644
--- a/arch/x86/Kconfig
+++ b/arch/x86/Kconfig
@@ -22,6 +22,7 @@ config X86_64
 config X86
 	def_bool y
 	select ARCH_HAS_DEBUG_STRICT_USER_COPY_CHECKS
+	select ARCH_MIGHT_HAVE_PC_PARPORT
 	select HAVE_AOUT if X86_32
 	select HAVE_UNSTABLE_SCHED_CLOCK
 	select ARCH_SUPPORTS_NUMA_BALANCING
@@ -90,7 +91,6 @@ config X86
 	select GENERIC_IRQ_SHOW
 	select GENERIC_CLOCKEVENTS_MIN_ADJUST
 	select IRQ_FORCED_THREADING
-	select USE_GENERIC_SMP_HELPERS if SMP
 	select HAVE_BPF_JIT if X86_64
 	select HAVE_ARCH_TRANSPARENT_HUGEPAGE
 	select CLKEVT_I8253
@@ -635,10 +635,10 @@ config PARAVIRT_SPINLOCKS
 	  spinlock implementation with something virtualization-friendly
 	  (for example, block the virtual CPU rather than spinning).
 
-	  Unfortunately the downside is an up to 5% performance hit on
-	  native kernels, with various workloads.
+	  It has a minimal impact on native kernels and gives a nice performance
+	  benefit on paravirtualized KVM / Xen kernels.
 
-	  If you are unsure how to answer this question, answer N.
+	  If you are unsure how to answer this question, answer Y.
 
 source "arch/x86/xen/Kconfig"
 
@@ -1885,6 +1885,10 @@ config USE_PERCPU_NUMA_NODE_ID
 	def_bool y
 	depends on NUMA
 
+config ARCH_ENABLE_SPLIT_PMD_PTLOCK
+	def_bool y
+	depends on X86_64 || X86_PAE
+
 menu "Power management and ACPI options"
 
 config ARCH_HIBERNATION_HEADER
diff --git a/arch/x86/include/asm/desc.h b/arch/x86/include/asm/desc.h
index b90e5dfeee46..50d033a8947d 100644
--- a/arch/x86/include/asm/desc.h
+++ b/arch/x86/include/asm/desc.h
@@ -327,10 +327,25 @@ static inline void write_trace_idt_entry(int entry, const gate_desc *gate)
 {
 	write_idt_entry(trace_idt_table, entry, gate);
 }
+
+static inline void _trace_set_gate(int gate, unsigned type, void *addr,
+				   unsigned dpl, unsigned ist, unsigned seg)
+{
+	gate_desc s;
+
+	pack_gate(&s, type, (unsigned long)addr, dpl, ist, seg);
+	/*
+	 * does not need to be atomic because it is only done once at
+	 * setup time
+	 */
+	write_trace_idt_entry(gate, &s);
+}
 #else
 static inline void write_trace_idt_entry(int entry, const gate_desc *gate)
 {
 }
+
+#define _trace_set_gate(gate, type, addr, dpl, ist, seg)
 #endif
 
 static inline void _set_gate(int gate, unsigned type, void *addr,
@@ -353,11 +368,14 @@ static inline void _set_gate(int gate, unsigned type, void *addr,
  * Pentium F0 0F bugfix can have resulted in the mapped
  * IDT being write-protected.
  */
-static inline void set_intr_gate(unsigned int n, void *addr)
-{
-	BUG_ON((unsigned)n > 0xFF);
-	_set_gate(n, GATE_INTERRUPT, addr, 0, 0, __KERNEL_CS);
-}
+#define set_intr_gate(n, addr)						\
+	do {								\
+		BUG_ON((unsigned)n > 0xFF);				\
+		_set_gate(n, GATE_INTERRUPT, (void *)addr, 0, 0,	\
+			  __KERNEL_CS);					\
+		_trace_set_gate(n, GATE_INTERRUPT, (void *)trace_##addr,\
+				0, 0, __KERNEL_CS);			\
+	} while (0)
 
 extern int first_system_vector;
 /* used_vectors is BITMAP for irq is not managed by percpu vector_irq */
@@ -374,37 +392,10 @@ static inline void alloc_system_vector(int vector)
 	}
 }
 
-#ifdef CONFIG_TRACING
-static inline void trace_set_intr_gate(unsigned int gate, void *addr)
-{
-	gate_desc s;
-
-	pack_gate(&s, GATE_INTERRUPT, (unsigned long)addr, 0, 0, __KERNEL_CS);
-	write_idt_entry(trace_idt_table, gate, &s);
-}
-
-static inline void __trace_alloc_intr_gate(unsigned int n, void *addr)
-{
-	trace_set_intr_gate(n, addr);
-}
-#else
-static inline void trace_set_intr_gate(unsigned int gate, void *addr)
-{
-}
-
-#define __trace_alloc_intr_gate(n, addr)
-#endif
-
-static inline void __alloc_intr_gate(unsigned int n, void *addr)
-{
-	set_intr_gate(n, addr);
-}
-
 #define alloc_intr_gate(n, addr)				\
 	do {							\
 		alloc_system_vector(n);				\
-		__alloc_intr_gate(n, addr);			\
-		__trace_alloc_intr_gate(n, trace_##addr);	\
+		set_intr_gate(n, addr);				\
 	} while (0)
 
 /*
diff --git a/arch/x86/include/asm/hw_irq.h b/arch/x86/include/asm/hw_irq.h
index 92b3bae08b74..cba45d99ac1a 100644
--- a/arch/x86/include/asm/hw_irq.h
+++ b/arch/x86/include/asm/hw_irq.h
@@ -187,6 +187,9 @@ extern __visible void smp_invalidate_interrupt(struct pt_regs *);
 #endif
 
 extern void (*__initconst interrupt[NR_VECTORS-FIRST_EXTERNAL_VECTOR])(void);
+#ifdef CONFIG_TRACING
+#define trace_interrupt interrupt
+#endif
 
 typedef int vector_irq_t[NR_VECTORS];
 DECLARE_PER_CPU(vector_irq_t, vector_irq);
diff --git a/arch/x86/include/asm/kdebug.h b/arch/x86/include/asm/kdebug.h
index 2c37aadcbc35..32ce71375b21 100644
--- a/arch/x86/include/asm/kdebug.h
+++ b/arch/x86/include/asm/kdebug.h
@@ -21,7 +21,7 @@ enum die_val {
 	DIE_NMIUNKNOWN,
 };
 
-extern void printk_address(unsigned long address, int reliable);
+extern void printk_address(unsigned long address);
 extern void die(const char *, struct pt_regs *,long);
 extern int __must_check __die(const char *, struct pt_regs *, long);
 extern void show_trace(struct task_struct *t, struct pt_regs *regs,
diff --git a/arch/x86/include/asm/kvm_emulate.h b/arch/x86/include/asm/kvm_emulate.h
index 15f960c06ff7..24ec1216596e 100644
--- a/arch/x86/include/asm/kvm_emulate.h
+++ b/arch/x86/include/asm/kvm_emulate.h
@@ -274,13 +274,17 @@ struct x86_emulate_ctxt {
 
 	bool guest_mode; /* guest running a nested guest */
 	bool perm_ok; /* do not check permissions if true */
-	bool only_vendor_specific_insn;
+	bool ud;	/* inject an #UD if host doesn't support insn */
 
 	bool have_exception;
 	struct x86_exception exception;
 
-	/* decode cache */
-	u8 twobyte;
+	/*
+	 * decode cache
+	 */
+
+	/* current opcode length in bytes */
+	u8 opcode_len;
 	u8 b;
 	u8 intercept;
 	u8 lock_prefix;
diff --git a/arch/x86/include/asm/kvm_host.h b/arch/x86/include/asm/kvm_host.h
index c76ff74a98f2..ae5d7830855c 100644
--- a/arch/x86/include/asm/kvm_host.h
+++ b/arch/x86/include/asm/kvm_host.h
@@ -79,6 +79,13 @@
 #define KVM_HPAGE_MASK(x)	(~(KVM_HPAGE_SIZE(x) - 1))
 #define KVM_PAGES_PER_HPAGE(x)	(KVM_HPAGE_SIZE(x) / PAGE_SIZE)
 
+static inline gfn_t gfn_to_index(gfn_t gfn, gfn_t base_gfn, int level)
+{
+	/* KVM_HPAGE_GFN_SHIFT(PT_PAGE_TABLE_LEVEL) must be 0. */
+	return (gfn >> KVM_HPAGE_GFN_SHIFT(level)) -
+		(base_gfn >> KVM_HPAGE_GFN_SHIFT(level));
+}
+
 #define SELECTOR_TI_MASK (1 << 2)
 #define SELECTOR_RPL_MASK 0x03
 
@@ -253,7 +260,6 @@ struct kvm_pio_request {
  * mode.
  */
 struct kvm_mmu {
-	void (*new_cr3)(struct kvm_vcpu *vcpu);
 	void (*set_cr3)(struct kvm_vcpu *vcpu, unsigned long root);
 	unsigned long (*get_cr3)(struct kvm_vcpu *vcpu);
 	u64 (*get_pdptr)(struct kvm_vcpu *vcpu, int index);
@@ -261,7 +267,6 @@ struct kvm_mmu {
 			  bool prefault);
 	void (*inject_page_fault)(struct kvm_vcpu *vcpu,
 				  struct x86_exception *fault);
-	void (*free)(struct kvm_vcpu *vcpu);
 	gpa_t (*gva_to_gpa)(struct kvm_vcpu *vcpu, gva_t gva, u32 access,
 			    struct x86_exception *exception);
 	gpa_t (*translate_gpa)(struct kvm_vcpu *vcpu, gpa_t gpa, u32 access);
@@ -389,6 +394,8 @@ struct kvm_vcpu_arch {
 
 	struct fpu guest_fpu;
 	u64 xcr0;
+	u64 guest_supported_xcr0;
+	u32 guest_xstate_size;
 
 	struct kvm_pio_request pio;
 	void *pio_data;
@@ -557,7 +564,9 @@ struct kvm_arch {
 
 	struct list_head assigned_dev_head;
 	struct iommu_domain *iommu_domain;
-	int iommu_flags;
+	bool iommu_noncoherent;
+#define __KVM_HAVE_ARCH_NONCOHERENT_DMA
+	atomic_t noncoherent_dma_count;
 	struct kvm_pic *vpic;
 	struct kvm_ioapic *vioapic;
 	struct kvm_pit *vpit;
@@ -780,11 +789,11 @@ void kvm_mmu_module_exit(void);
 
 void kvm_mmu_destroy(struct kvm_vcpu *vcpu);
 int kvm_mmu_create(struct kvm_vcpu *vcpu);
-int kvm_mmu_setup(struct kvm_vcpu *vcpu);
+void kvm_mmu_setup(struct kvm_vcpu *vcpu);
 void kvm_mmu_set_mask_ptes(u64 user_mask, u64 accessed_mask,
 		u64 dirty_mask, u64 nx_mask, u64 x_mask);
 
-int kvm_mmu_reset_context(struct kvm_vcpu *vcpu);
+void kvm_mmu_reset_context(struct kvm_vcpu *vcpu);
 void kvm_mmu_slot_remove_write_access(struct kvm *kvm, int slot);
 void kvm_mmu_write_protect_pt_masked(struct kvm *kvm,
 				     struct kvm_memory_slot *slot,
@@ -922,13 +931,11 @@ int kvm_emulate_hypercall(struct kvm_vcpu *vcpu);
 int kvm_mmu_page_fault(struct kvm_vcpu *vcpu, gva_t gva, u32 error_code,
 		       void *insn, int insn_len);
 void kvm_mmu_invlpg(struct kvm_vcpu *vcpu, gva_t gva);
+void kvm_mmu_new_cr3(struct kvm_vcpu *vcpu);
 
 void kvm_enable_tdp(void);
 void kvm_disable_tdp(void);
 
-int complete_pio(struct kvm_vcpu *vcpu);
-bool kvm_check_iopl(struct kvm_vcpu *vcpu);
-
 static inline gpa_t translate_gpa(struct kvm_vcpu *vcpu, gpa_t gpa, u32 access)
 {
 	return gpa;
diff --git a/arch/x86/include/asm/pgalloc.h b/arch/x86/include/asm/pgalloc.h
index b4389a468fb6..c4412e972bbd 100644
--- a/arch/x86/include/asm/pgalloc.h
+++ b/arch/x86/include/asm/pgalloc.h
@@ -80,12 +80,21 @@ static inline void pmd_populate(struct mm_struct *mm, pmd_t *pmd,
 #if PAGETABLE_LEVELS > 2
 static inline pmd_t *pmd_alloc_one(struct mm_struct *mm, unsigned long addr)
 {
-	return (pmd_t *)get_zeroed_page(GFP_KERNEL|__GFP_REPEAT);
+	struct page *page;
+	page = alloc_pages(GFP_KERNEL | __GFP_REPEAT | __GFP_ZERO, 0);
+	if (!page)
+		return NULL;
+	if (!pgtable_pmd_page_ctor(page)) {
+		__free_pages(page, 0);
+		return NULL;
+	}
+	return (pmd_t *)page_address(page);
 }
 
 static inline void pmd_free(struct mm_struct *mm, pmd_t *pmd)
 {
 	BUG_ON((unsigned long)pmd & (PAGE_SIZE-1));
+	pgtable_pmd_page_dtor(virt_to_page(pmd));
 	free_page((unsigned long)pmd);
 }
 
diff --git a/arch/x86/include/asm/pvclock.h b/arch/x86/include/asm/pvclock.h
index be8269b00e2a..d6b078e9fa28 100644
--- a/arch/x86/include/asm/pvclock.h
+++ b/arch/x86/include/asm/pvclock.h
@@ -14,6 +14,8 @@ void pvclock_read_wallclock(struct pvclock_wall_clock *wall,
 			    struct timespec *ts);
 void pvclock_resume(void);
 
+void pvclock_touch_watchdogs(void);
+
 /*
  * Scale a 64-bit delta by scaling and multiplying by a 32-bit fraction,
  * yielding a 64-bit result.
diff --git a/arch/x86/include/asm/segment.h b/arch/x86/include/asm/segment.h
index c48a95035a77..6f1c3a8a33ab 100644
--- a/arch/x86/include/asm/segment.h
+++ b/arch/x86/include/asm/segment.h
@@ -214,6 +214,9 @@
 #ifdef __KERNEL__
 #ifndef __ASSEMBLY__
 extern const char early_idt_handlers[NUM_EXCEPTION_VECTORS][2+2+5];
+#ifdef CONFIG_TRACING
+#define trace_early_idt_handlers early_idt_handlers
+#endif
 
 /*
  * Load a segment. Fall back on loading the zero
diff --git a/arch/x86/include/asm/thread_info.h b/arch/x86/include/asm/thread_info.h
index c46a46be1ec6..3ba3de457d05 100644
--- a/arch/x86/include/asm/thread_info.h
+++ b/arch/x86/include/asm/thread_info.h
@@ -153,8 +153,6 @@ struct thread_info {
 #define _TIF_WORK_CTXSW_PREV (_TIF_WORK_CTXSW|_TIF_USER_RETURN_NOTIFY)
 #define _TIF_WORK_CTXSW_NEXT (_TIF_WORK_CTXSW)
 
-#define PREEMPT_ACTIVE		0x10000000
-
 #ifdef CONFIG_X86_32
 
 #define STACK_WARN	(THREAD_SIZE/8)
diff --git a/arch/x86/include/asm/trace/exceptions.h b/arch/x86/include/asm/trace/exceptions.h
new file mode 100644
index 000000000000..2fbc66c7885b
--- /dev/null
+++ b/arch/x86/include/asm/trace/exceptions.h
@@ -0,0 +1,52 @@
+#undef TRACE_SYSTEM
+#define TRACE_SYSTEM exceptions
+
+#if !defined(_TRACE_PAGE_FAULT_H) || defined(TRACE_HEADER_MULTI_READ)
+#define _TRACE_PAGE_FAULT_H
+
+#include <linux/tracepoint.h>
+
+extern void trace_irq_vector_regfunc(void);
+extern void trace_irq_vector_unregfunc(void);
+
+DECLARE_EVENT_CLASS(x86_exceptions,
+
+	TP_PROTO(unsigned long address, struct pt_regs *regs,
+		 unsigned long error_code),
+
+	TP_ARGS(address, regs, error_code),
+
+	TP_STRUCT__entry(
+		__field(		unsigned long, address	)
+		__field(		unsigned long, ip	)
+		__field(		unsigned long, error_code )
+	),
+
+	TP_fast_assign(
+		__entry->address = address;
+		__entry->ip = regs->ip;
+		__entry->error_code = error_code;
+	),
+
+	TP_printk("address=%pf ip=%pf error_code=0x%lx",
+		  (void *)__entry->address, (void *)__entry->ip,
+		  __entry->error_code) );
+
+#define DEFINE_PAGE_FAULT_EVENT(name)				\
+DEFINE_EVENT_FN(x86_exceptions, name,				\
+	TP_PROTO(unsigned long address,	struct pt_regs *regs,	\
+		 unsigned long error_code),			\
+	TP_ARGS(address, regs, error_code),			\
+	trace_irq_vector_regfunc,				\
+	trace_irq_vector_unregfunc);
+
+DEFINE_PAGE_FAULT_EVENT(page_fault_user);
+DEFINE_PAGE_FAULT_EVENT(page_fault_kernel);
+
+#undef TRACE_INCLUDE_PATH
+#define TRACE_INCLUDE_PATH .
+#define TRACE_INCLUDE_FILE exceptions
+#endif /*  _TRACE_PAGE_FAULT_H */
+
+/* This part must be outside protection */
+#include <trace/define_trace.h>
diff --git a/arch/x86/include/asm/traps.h b/arch/x86/include/asm/traps.h
index 7036cb60cd87..58d66fe06b61 100644
--- a/arch/x86/include/asm/traps.h
+++ b/arch/x86/include/asm/traps.h
@@ -37,6 +37,23 @@ asmlinkage void machine_check(void);
 #endif /* CONFIG_X86_MCE */
 asmlinkage void simd_coprocessor_error(void);
 
+#ifdef CONFIG_TRACING
+asmlinkage void trace_page_fault(void);
+#define trace_divide_error divide_error
+#define trace_bounds bounds
+#define trace_invalid_op invalid_op
+#define trace_device_not_available device_not_available
+#define trace_coprocessor_segment_overrun coprocessor_segment_overrun
+#define trace_invalid_TSS invalid_TSS
+#define trace_segment_not_present segment_not_present
+#define trace_general_protection general_protection
+#define trace_spurious_interrupt_bug spurious_interrupt_bug
+#define trace_coprocessor_error coprocessor_error
+#define trace_alignment_check alignment_check
+#define trace_simd_coprocessor_error simd_coprocessor_error
+#define trace_async_page_fault async_page_fault
+#endif
+
 dotraplinkage void do_divide_error(struct pt_regs *, long);
 dotraplinkage void do_debug(struct pt_regs *, long);
 dotraplinkage void do_nmi(struct pt_regs *, long);
@@ -55,6 +72,9 @@ asmlinkage __kprobes struct pt_regs *sync_regs(struct pt_regs *);
 #endif
 dotraplinkage void do_general_protection(struct pt_regs *, long);
 dotraplinkage void do_page_fault(struct pt_regs *, unsigned long);
+#ifdef CONFIG_TRACING
+dotraplinkage void trace_do_page_fault(struct pt_regs *, unsigned long);
+#endif
 dotraplinkage void do_spurious_interrupt_bug(struct pt_regs *, long);
 dotraplinkage void do_coprocessor_error(struct pt_regs *, long);
 dotraplinkage void do_alignment_check(struct pt_regs *, long);
diff --git a/arch/x86/include/asm/xen/page-coherent.h b/arch/x86/include/asm/xen/page-coherent.h
new file mode 100644
index 000000000000..7f02fe4e2c7b
--- /dev/null
+++ b/arch/x86/include/asm/xen/page-coherent.h
@@ -0,0 +1,38 @@
+#ifndef _ASM_X86_XEN_PAGE_COHERENT_H
+#define _ASM_X86_XEN_PAGE_COHERENT_H
+
+#include <asm/page.h>
+#include <linux/dma-attrs.h>
+#include <linux/dma-mapping.h>
+
+static inline void *xen_alloc_coherent_pages(struct device *hwdev, size_t size,
+		dma_addr_t *dma_handle, gfp_t flags,
+		struct dma_attrs *attrs)
+{
+	void *vstart = (void*)__get_free_pages(flags, get_order(size));
+	*dma_handle = virt_to_phys(vstart);
+	return vstart;
+}
+
+static inline void xen_free_coherent_pages(struct device *hwdev, size_t size,
+		void *cpu_addr, dma_addr_t dma_handle,
+		struct dma_attrs *attrs)
+{
+	free_pages((unsigned long) cpu_addr, get_order(size));
+}
+
+static inline void xen_dma_map_page(struct device *hwdev, struct page *page,
+	     unsigned long offset, size_t size, enum dma_data_direction dir,
+	     struct dma_attrs *attrs) { }
+
+static inline void xen_dma_unmap_page(struct device *hwdev, dma_addr_t handle,
+		size_t size, enum dma_data_direction dir,
+		struct dma_attrs *attrs) { }
+
+static inline void xen_dma_sync_single_for_cpu(struct device *hwdev,
+		dma_addr_t handle, size_t size, enum dma_data_direction dir) { }
+
+static inline void xen_dma_sync_single_for_device(struct device *hwdev,
+		dma_addr_t handle, size_t size, enum dma_data_direction dir) { }
+
+#endif /* _ASM_X86_XEN_PAGE_COHERENT_H */
diff --git a/arch/x86/include/uapi/asm/kvm.h b/arch/x86/include/uapi/asm/kvm.h
index 5d9a3033b3d7..d3a87780c70b 100644
--- a/arch/x86/include/uapi/asm/kvm.h
+++ b/arch/x86/include/uapi/asm/kvm.h
@@ -211,9 +211,9 @@ struct kvm_cpuid_entry2 {
 	__u32 padding[3];
 };
 
-#define KVM_CPUID_FLAG_SIGNIFCANT_INDEX 1
-#define KVM_CPUID_FLAG_STATEFUL_FUNC    2
-#define KVM_CPUID_FLAG_STATE_READ_NEXT  4
+#define KVM_CPUID_FLAG_SIGNIFCANT_INDEX		BIT(0)
+#define KVM_CPUID_FLAG_STATEFUL_FUNC		BIT(1)
+#define KVM_CPUID_FLAG_STATE_READ_NEXT		BIT(2)
 
 /* for KVM_SET_CPUID2 */
 struct kvm_cpuid2 {
diff --git a/arch/x86/include/uapi/asm/msr-index.h b/arch/x86/include/uapi/asm/msr-index.h
index 940ed3fd889a..37813b5ddc37 100644
--- a/arch/x86/include/uapi/asm/msr-index.h
+++ b/arch/x86/include/uapi/asm/msr-index.h
@@ -538,6 +538,7 @@
 
 /* MSR_IA32_VMX_MISC bits */
 #define MSR_IA32_VMX_MISC_VMWRITE_SHADOW_RO_FIELDS (1ULL << 29)
+#define MSR_IA32_VMX_MISC_PREEMPTION_TIMER_SCALE   0x1F
 /* AMD-V MSRs */
 
 #define MSR_VM_CR                       0xc0010114
diff --git a/arch/x86/kernel/alternative.c b/arch/x86/kernel/alternative.c
index 15e8563e5c24..df94598ad05a 100644
--- a/arch/x86/kernel/alternative.c
+++ b/arch/x86/kernel/alternative.c
@@ -402,17 +402,6 @@ void alternatives_enable_smp(void)
 {
 	struct smp_alt_module *mod;
 
-#ifdef CONFIG_LOCKDEP
-	/*
-	 * Older binutils section handling bug prevented
-	 * alternatives-replacement from working reliably.
-	 *
-	 * If this still occurs then you should see a hang
-	 * or crash shortly after this line:
-	 */
-	pr_info("lockdep: fixing up alternatives\n");
-#endif
-
 	/* Why bother if there are no other CPUs? */
 	BUG_ON(num_possible_cpus() == 1);
 
diff --git a/arch/x86/kernel/apic/apic.c b/arch/x86/kernel/apic/apic.c
index ed165d657380..d278736bf774 100644
--- a/arch/x86/kernel/apic/apic.c
+++ b/arch/x86/kernel/apic/apic.c
@@ -62,6 +62,7 @@ unsigned disabled_cpus;
 
 /* Processor that is doing the boot up */
 unsigned int boot_cpu_physical_apicid = -1U;
+EXPORT_SYMBOL_GPL(boot_cpu_physical_apicid);
 
 /*
  * The highest APIC ID seen during enumeration.
diff --git a/arch/x86/kernel/cpu/amd.c b/arch/x86/kernel/cpu/amd.c
index 3daece79a142..bca023bdd6b2 100644
--- a/arch/x86/kernel/cpu/amd.c
+++ b/arch/x86/kernel/cpu/amd.c
@@ -339,7 +339,7 @@ static void amd_get_topology(struct cpuinfo_x86 *c)
 #endif
 
 /*
- * On a AMD dual core setup the lower bits of the APIC id distingush the cores.
+ * On a AMD dual core setup the lower bits of the APIC id distinguish the cores.
  * Assumes number of cores is a power of two.
  */
 static void amd_detect_cmp(struct cpuinfo_x86 *c)
diff --git a/arch/x86/kernel/cpu/intel_cacheinfo.c b/arch/x86/kernel/cpu/intel_cacheinfo.c
index 1414c90feaba..0641113e2965 100644
--- a/arch/x86/kernel/cpu/intel_cacheinfo.c
+++ b/arch/x86/kernel/cpu/intel_cacheinfo.c
@@ -1,5 +1,5 @@
 /*
- *	Routines to indentify caches on Intel CPU.
+ *	Routines to identify caches on Intel CPU.
  *
  *	Changes:
  *	Venkatesh Pallipadi	: Adding cache identification through cpuid(4)
diff --git a/arch/x86/kernel/cpu/scattered.c b/arch/x86/kernel/cpu/scattered.c
index f2cc63e9cf08..b6f794aa1693 100644
--- a/arch/x86/kernel/cpu/scattered.c
+++ b/arch/x86/kernel/cpu/scattered.c
@@ -1,5 +1,5 @@
 /*
- *	Routines to indentify additional cpu features that are scattered in
+ *	Routines to identify additional cpu features that are scattered in
  *	cpuid space.
  */
 #include <linux/cpu.h>
diff --git a/arch/x86/kernel/dumpstack.c b/arch/x86/kernel/dumpstack.c
index deb6421c9e69..d9c12d3022a7 100644
--- a/arch/x86/kernel/dumpstack.c
+++ b/arch/x86/kernel/dumpstack.c
@@ -25,12 +25,17 @@ unsigned int code_bytes = 64;
 int kstack_depth_to_print = 3 * STACKSLOTS_PER_LINE;
 static int die_counter;
 
-void printk_address(unsigned long address, int reliable)
+static void printk_stack_address(unsigned long address, int reliable)
 {
 	pr_cont(" [<%p>] %s%pB\n",
 		(void *)address, reliable ? "" : "? ", (void *)address);
 }
 
+void printk_address(unsigned long address)
+{
+	pr_cont(" [<%p>] %pS\n", (void *)address, (void *)address);
+}
+
 #ifdef CONFIG_FUNCTION_GRAPH_TRACER
 static void
 print_ftrace_graph_addr(unsigned long addr, void *data,
@@ -151,7 +156,7 @@ static void print_trace_address(void *data, unsigned long addr, int reliable)
 {
 	touch_nmi_watchdog();
 	printk(data);
-	printk_address(addr, reliable);
+	printk_stack_address(addr, reliable);
 }
 
 static const struct stacktrace_ops print_trace_ops = {
@@ -281,7 +286,7 @@ int __kprobes __die(const char *str, struct pt_regs *regs, long err)
 #else
 	/* Executive summary in case the oops scrolled away */
 	printk(KERN_ALERT "RIP ");
-	printk_address(regs->ip, 1);
+	printk_address(regs->ip);
 	printk(" RSP <%016lx>\n", regs->sp);
 #endif
 	return 0;
diff --git a/arch/x86/kernel/early-quirks.c b/arch/x86/kernel/early-quirks.c
index b3cd3ebae077..96f958d8cd45 100644
--- a/arch/x86/kernel/early-quirks.c
+++ b/arch/x86/kernel/early-quirks.c
@@ -313,6 +313,16 @@ static size_t __init gen6_stolen_size(int num, int slot, int func)
 	return gmch_ctrl << 25; /* 32 MB units */
 }
 
+static inline size_t gen8_stolen_size(int num, int slot, int func)
+{
+	u16 gmch_ctrl;
+
+	gmch_ctrl = read_pci_config_16(num, slot, func, SNB_GMCH_CTRL);
+	gmch_ctrl >>= BDW_GMCH_GMS_SHIFT;
+	gmch_ctrl &= BDW_GMCH_GMS_MASK;
+	return gmch_ctrl << 25; /* 32 MB units */
+}
+
 typedef size_t (*stolen_size_fn)(int num, int slot, int func);
 
 static struct pci_device_id intel_stolen_ids[] __initdata = {
@@ -336,6 +346,8 @@ static struct pci_device_id intel_stolen_ids[] __initdata = {
 	INTEL_IVB_D_IDS(gen6_stolen_size),
 	INTEL_HSW_D_IDS(gen6_stolen_size),
 	INTEL_HSW_M_IDS(gen6_stolen_size),
+	INTEL_BDW_M_IDS(gen8_stolen_size),
+	INTEL_BDW_D_IDS(gen8_stolen_size)
 };
 
 static void __init intel_graphics_stolen(int num, int slot, int func)
diff --git a/arch/x86/kernel/entry_32.S b/arch/x86/kernel/entry_32.S
index fd1bc1b15e6d..51e2988c5728 100644
--- a/arch/x86/kernel/entry_32.S
+++ b/arch/x86/kernel/entry_32.S
@@ -1244,6 +1244,16 @@ return_to_handler:
  */
 	.pushsection .kprobes.text, "ax"
 
+#ifdef CONFIG_TRACING
+ENTRY(trace_page_fault)
+	RING0_EC_FRAME
+	ASM_CLAC
+	pushl_cfi $trace_do_page_fault
+	jmp error_code
+	CFI_ENDPROC
+END(trace_page_fault)
+#endif
+
 ENTRY(page_fault)
 	RING0_EC_FRAME
 	ASM_CLAC
diff --git a/arch/x86/kernel/entry_64.S b/arch/x86/kernel/entry_64.S
index 603be7c70675..e21b0785a85b 100644
--- a/arch/x86/kernel/entry_64.S
+++ b/arch/x86/kernel/entry_64.S
@@ -1278,6 +1278,17 @@ ENTRY(\sym)
 END(\sym)
 .endm
 
+#ifdef CONFIG_TRACING
+.macro trace_errorentry sym do_sym
+errorentry trace(\sym) trace(\do_sym)
+errorentry \sym \do_sym
+.endm
+#else
+.macro trace_errorentry sym do_sym
+errorentry \sym \do_sym
+.endm
+#endif
+
 	/* error code is on the stack already */
 .macro paranoiderrorentry sym do_sym
 ENTRY(\sym)
@@ -1480,7 +1491,7 @@ zeroentry xen_int3 do_int3
 errorentry xen_stack_segment do_stack_segment
 #endif
 errorentry general_protection do_general_protection
-errorentry page_fault do_page_fault
+trace_errorentry page_fault do_page_fault
 #ifdef CONFIG_KVM_GUEST
 errorentry async_page_fault do_async_page_fault
 #endif
diff --git a/arch/x86/kernel/ftrace.c b/arch/x86/kernel/ftrace.c
index 42a392a9fd02..d4bdd253fea7 100644
--- a/arch/x86/kernel/ftrace.c
+++ b/arch/x86/kernel/ftrace.c
@@ -248,6 +248,15 @@ int ftrace_update_ftrace_func(ftrace_func_t func)
 	return ret;
 }
 
+static int is_ftrace_caller(unsigned long ip)
+{
+	if (ip == (unsigned long)(&ftrace_call) ||
+		ip == (unsigned long)(&ftrace_regs_call))
+		return 1;
+
+	return 0;
+}
+
 /*
  * A breakpoint was added to the code address we are about to
  * modify, and this is the handle that will just skip over it.
@@ -257,10 +266,13 @@ int ftrace_update_ftrace_func(ftrace_func_t func)
  */
 int ftrace_int3_handler(struct pt_regs *regs)
 {
+	unsigned long ip;
+
 	if (WARN_ON_ONCE(!regs))
 		return 0;
 
-	if (!ftrace_location(regs->ip - 1))
+	ip = regs->ip - 1;
+	if (!ftrace_location(ip) && !is_ftrace_caller(ip))
 		return 0;
 
 	regs->ip += MCOUNT_INSN_SIZE - 1;
diff --git a/arch/x86/kernel/head64.c b/arch/x86/kernel/head64.c
index 1be8e43b669e..85126ccbdf6b 100644
--- a/arch/x86/kernel/head64.c
+++ b/arch/x86/kernel/head64.c
@@ -162,7 +162,7 @@ asmlinkage void __init x86_64_start_kernel(char * real_mode_data)
 	clear_bss();
 
 	for (i = 0; i < NUM_EXCEPTION_VECTORS; i++)
-		set_intr_gate(i, &early_idt_handlers[i]);
+		set_intr_gate(i, early_idt_handlers[i]);
 	load_idt((const struct desc_ptr *)&idt_descr);
 
 	copy_bootdata(__va(real_mode_data));
diff --git a/arch/x86/kernel/kvm.c b/arch/x86/kernel/kvm.c
index b2046e4d0b59..6dd802c6d780 100644
--- a/arch/x86/kernel/kvm.c
+++ b/arch/x86/kernel/kvm.c
@@ -464,7 +464,7 @@ static struct notifier_block kvm_cpu_notifier = {
 
 static void __init kvm_apf_trap_init(void)
 {
-	set_intr_gate(14, &async_page_fault);
+	set_intr_gate(14, async_page_fault);
 }
 
 void __init kvm_guest_init(void)
diff --git a/arch/x86/kernel/kvmclock.c b/arch/x86/kernel/kvmclock.c
index 1570e0741344..e6041094ff26 100644
--- a/arch/x86/kernel/kvmclock.c
+++ b/arch/x86/kernel/kvmclock.c
@@ -139,6 +139,7 @@ bool kvm_check_and_clear_guest_paused(void)
 	src = &hv_clock[cpu].pvti;
 	if ((src->flags & PVCLOCK_GUEST_STOPPED) != 0) {
 		src->flags &= ~PVCLOCK_GUEST_STOPPED;
+		pvclock_touch_watchdogs();
 		ret = true;
 	}
 
diff --git a/arch/x86/kernel/microcode_amd.c b/arch/x86/kernel/microcode_amd.c
index af99f71aeb7f..c3d4cc972eca 100644
--- a/arch/x86/kernel/microcode_amd.c
+++ b/arch/x86/kernel/microcode_amd.c
@@ -431,7 +431,7 @@ static enum ucode_state request_microcode_amd(int cpu, struct device *device,
 		snprintf(fw_name, sizeof(fw_name), "amd-ucode/microcode_amd_fam%.2xh.bin", c->x86);
 
 	if (request_firmware(&fw, (const char *)fw_name, device)) {
-		pr_err("failed to load file %s\n", fw_name);
+		pr_debug("failed to load file %s\n", fw_name);
 		goto out;
 	}
 
diff --git a/arch/x86/kernel/process_64.c b/arch/x86/kernel/process_64.c
index 10fe4c189621..9c0280f93d05 100644
--- a/arch/x86/kernel/process_64.c
+++ b/arch/x86/kernel/process_64.c
@@ -63,7 +63,7 @@ void __show_regs(struct pt_regs *regs, int all)
 	unsigned int ds, cs, es;
 
 	printk(KERN_DEFAULT "RIP: %04lx:[<%016lx>] ", regs->cs & 0xffff, regs->ip);
-	printk_address(regs->ip, 1);
+	printk_address(regs->ip);
 	printk(KERN_DEFAULT "RSP: %04lx:%016lx  EFLAGS: %08lx\n", regs->ss,
 			regs->sp, regs->flags);
 	printk(KERN_DEFAULT "RAX: %016lx RBX: %016lx RCX: %016lx\n",
diff --git a/arch/x86/kernel/pvclock.c b/arch/x86/kernel/pvclock.c
index a16bae3f83b3..2f355d229a58 100644
--- a/arch/x86/kernel/pvclock.c
+++ b/arch/x86/kernel/pvclock.c
@@ -43,6 +43,14 @@ unsigned long pvclock_tsc_khz(struct pvclock_vcpu_time_info *src)
 	return pv_tsc_khz;
 }
 
+void pvclock_touch_watchdogs(void)
+{
+	touch_softlockup_watchdog_sync();
+	clocksource_touch_watchdog();
+	rcu_cpu_stall_reset();
+	reset_hung_task_detector();
+}
+
 static atomic64_t last_value = ATOMIC64_INIT(0);
 
 void pvclock_resume(void)
@@ -74,6 +82,11 @@ cycle_t pvclock_clocksource_read(struct pvclock_vcpu_time_info *src)
 		version = __pvclock_read_cycles(src, &ret, &flags);
 	} while ((src->version & 1) || version != src->version);
 
+	if (unlikely((flags & PVCLOCK_GUEST_STOPPED) != 0)) {
+		src->flags &= ~PVCLOCK_GUEST_STOPPED;
+		pvclock_touch_watchdogs();
+	}
+
 	if ((valid_flags & PVCLOCK_TSC_STABLE_BIT) &&
 		(flags & PVCLOCK_TSC_STABLE_BIT))
 		return ret;
diff --git a/arch/x86/kernel/traps.c b/arch/x86/kernel/traps.c
index 996ce2313ce6..b857ed890b4c 100644
--- a/arch/x86/kernel/traps.c
+++ b/arch/x86/kernel/traps.c
@@ -713,7 +713,7 @@ void __init early_trap_init(void)
 	/* int3 can be called from all */
 	set_system_intr_gate_ist(X86_TRAP_BP, &int3, DEBUG_STACK);
 #ifdef CONFIG_X86_32
-	set_intr_gate(X86_TRAP_PF, &page_fault);
+	set_intr_gate(X86_TRAP_PF, page_fault);
 #endif
 	load_idt(&idt_descr);
 }
@@ -721,7 +721,7 @@ void __init early_trap_init(void)
 void __init early_trap_pf_init(void)
 {
 #ifdef CONFIG_X86_64
-	set_intr_gate(X86_TRAP_PF, &page_fault);
+	set_intr_gate(X86_TRAP_PF, page_fault);
 #endif
 }
 
@@ -737,30 +737,30 @@ void __init trap_init(void)
 	early_iounmap(p, 4);
 #endif
 
-	set_intr_gate(X86_TRAP_DE, &divide_error);
+	set_intr_gate(X86_TRAP_DE, divide_error);
 	set_intr_gate_ist(X86_TRAP_NMI, &nmi, NMI_STACK);
 	/* int4 can be called from all */
 	set_system_intr_gate(X86_TRAP_OF, &overflow);
-	set_intr_gate(X86_TRAP_BR, &bounds);
-	set_intr_gate(X86_TRAP_UD, &invalid_op);
-	set_intr_gate(X86_TRAP_NM, &device_not_available);
+	set_intr_gate(X86_TRAP_BR, bounds);
+	set_intr_gate(X86_TRAP_UD, invalid_op);
+	set_intr_gate(X86_TRAP_NM, device_not_available);
 #ifdef CONFIG_X86_32
 	set_task_gate(X86_TRAP_DF, GDT_ENTRY_DOUBLEFAULT_TSS);
 #else
 	set_intr_gate_ist(X86_TRAP_DF, &double_fault, DOUBLEFAULT_STACK);
 #endif
-	set_intr_gate(X86_TRAP_OLD_MF, &coprocessor_segment_overrun);
-	set_intr_gate(X86_TRAP_TS, &invalid_TSS);
-	set_intr_gate(X86_TRAP_NP, &segment_not_present);
+	set_intr_gate(X86_TRAP_OLD_MF, coprocessor_segment_overrun);
+	set_intr_gate(X86_TRAP_TS, invalid_TSS);
+	set_intr_gate(X86_TRAP_NP, segment_not_present);
 	set_intr_gate_ist(X86_TRAP_SS, &stack_segment, STACKFAULT_STACK);
-	set_intr_gate(X86_TRAP_GP, &general_protection);
-	set_intr_gate(X86_TRAP_SPURIOUS, &spurious_interrupt_bug);
-	set_intr_gate(X86_TRAP_MF, &coprocessor_error);
-	set_intr_gate(X86_TRAP_AC, &alignment_check);
+	set_intr_gate(X86_TRAP_GP, general_protection);
+	set_intr_gate(X86_TRAP_SPURIOUS, spurious_interrupt_bug);
+	set_intr_gate(X86_TRAP_MF, coprocessor_error);
+	set_intr_gate(X86_TRAP_AC, alignment_check);
 #ifdef CONFIG_X86_MCE
 	set_intr_gate_ist(X86_TRAP_MC, &machine_check, MCE_STACK);
 #endif
-	set_intr_gate(X86_TRAP_XF, &simd_coprocessor_error);
+	set_intr_gate(X86_TRAP_XF, simd_coprocessor_error);
 
 	/* Reserve all the builtin and the syscall vector: */
 	for (i = 0; i < FIRST_EXTERNAL_VECTOR; i++)
diff --git a/arch/x86/kvm/Kconfig b/arch/x86/kvm/Kconfig
index a47a3e54b964..b89c5db2b832 100644
--- a/arch/x86/kvm/Kconfig
+++ b/arch/x86/kvm/Kconfig
@@ -38,6 +38,7 @@ config KVM
 	select PERF_EVENTS
 	select HAVE_KVM_MSI
 	select HAVE_KVM_CPU_RELAX_INTERCEPT
+	select KVM_VFIO
 	---help---
 	  Support hosting fully virtualized guest machines using hardware
 	  virtualization extensions.  You will need a fairly recent
diff --git a/arch/x86/kvm/Makefile b/arch/x86/kvm/Makefile
index bf4fb04d0112..25d22b2d6509 100644
--- a/arch/x86/kvm/Makefile
+++ b/arch/x86/kvm/Makefile
@@ -9,7 +9,7 @@ KVM := ../../../virt/kvm
 
 kvm-y			+= $(KVM)/kvm_main.o $(KVM)/ioapic.o \
 				$(KVM)/coalesced_mmio.o $(KVM)/irq_comm.o \
-				$(KVM)/eventfd.o $(KVM)/irqchip.o
+				$(KVM)/eventfd.o $(KVM)/irqchip.o $(KVM)/vfio.o
 kvm-$(CONFIG_KVM_DEVICE_ASSIGNMENT)	+= $(KVM)/assigned-dev.o $(KVM)/iommu.o
 kvm-$(CONFIG_KVM_ASYNC_PF)	+= $(KVM)/async_pf.o
 
diff --git a/arch/x86/kvm/cpuid.c b/arch/x86/kvm/cpuid.c
index b110fe6c03d4..c6976257eff5 100644
--- a/arch/x86/kvm/cpuid.c
+++ b/arch/x86/kvm/cpuid.c
@@ -23,6 +23,26 @@
 #include "mmu.h"
 #include "trace.h"
 
+static u32 xstate_required_size(u64 xstate_bv)
+{
+	int feature_bit = 0;
+	u32 ret = XSAVE_HDR_SIZE + XSAVE_HDR_OFFSET;
+
+	xstate_bv &= ~XSTATE_FPSSE;
+	while (xstate_bv) {
+		if (xstate_bv & 0x1) {
+		        u32 eax, ebx, ecx, edx;
+		        cpuid_count(0xD, feature_bit, &eax, &ebx, &ecx, &edx);
+			ret = max(ret, eax + ebx);
+		}
+
+		xstate_bv >>= 1;
+		feature_bit++;
+	}
+
+	return ret;
+}
+
 void kvm_update_cpuid(struct kvm_vcpu *vcpu)
 {
 	struct kvm_cpuid_entry2 *best;
@@ -46,6 +66,18 @@ void kvm_update_cpuid(struct kvm_vcpu *vcpu)
 			apic->lapic_timer.timer_mode_mask = 1 << 17;
 	}
 
+	best = kvm_find_cpuid_entry(vcpu, 0xD, 0);
+	if (!best) {
+		vcpu->arch.guest_supported_xcr0 = 0;
+		vcpu->arch.guest_xstate_size = XSAVE_HDR_SIZE + XSAVE_HDR_OFFSET;
+	} else {
+		vcpu->arch.guest_supported_xcr0 =
+			(best->eax | ((u64)best->edx << 32)) &
+			host_xcr0 & KVM_SUPPORTED_XCR0;
+		vcpu->arch.guest_xstate_size =
+			xstate_required_size(vcpu->arch.guest_supported_xcr0);
+	}
+
 	kvm_pmu_cpuid_update(vcpu);
 }
 
@@ -182,13 +214,35 @@ static bool supported_xcr0_bit(unsigned bit)
 {
 	u64 mask = ((u64)1 << bit);
 
-	return mask & (XSTATE_FP | XSTATE_SSE | XSTATE_YMM) & host_xcr0;
+	return mask & KVM_SUPPORTED_XCR0 & host_xcr0;
 }
 
 #define F(x) bit(X86_FEATURE_##x)
 
-static int do_cpuid_ent(struct kvm_cpuid_entry2 *entry, u32 function,
-			 u32 index, int *nent, int maxnent)
+static int __do_cpuid_ent_emulated(struct kvm_cpuid_entry2 *entry,
+				   u32 func, u32 index, int *nent, int maxnent)
+{
+	switch (func) {
+	case 0:
+		entry->eax = 1;		/* only one leaf currently */
+		++*nent;
+		break;
+	case 1:
+		entry->ecx = F(MOVBE);
+		++*nent;
+		break;
+	default:
+		break;
+	}
+
+	entry->function = func;
+	entry->index = index;
+
+	return 0;
+}
+
+static inline int __do_cpuid_ent(struct kvm_cpuid_entry2 *entry, u32 function,
+				 u32 index, int *nent, int maxnent)
 {
 	int r;
 	unsigned f_nx = is_efer_nx() ? F(NX) : 0;
@@ -383,6 +437,8 @@ static int do_cpuid_ent(struct kvm_cpuid_entry2 *entry, u32 function,
 	case 0xd: {
 		int idx, i;
 
+		entry->eax &= host_xcr0 & KVM_SUPPORTED_XCR0;
+		entry->edx &= (host_xcr0 & KVM_SUPPORTED_XCR0) >> 32;
 		entry->flags |= KVM_CPUID_FLAG_SIGNIFCANT_INDEX;
 		for (idx = 1, i = 1; idx < 64; ++idx) {
 			if (*nent >= maxnent)
@@ -481,6 +537,15 @@ out:
 	return r;
 }
 
+static int do_cpuid_ent(struct kvm_cpuid_entry2 *entry, u32 func,
+			u32 idx, int *nent, int maxnent, unsigned int type)
+{
+	if (type == KVM_GET_EMULATED_CPUID)
+		return __do_cpuid_ent_emulated(entry, func, idx, nent, maxnent);
+
+	return __do_cpuid_ent(entry, func, idx, nent, maxnent);
+}
+
 #undef F
 
 struct kvm_cpuid_param {
@@ -495,8 +560,36 @@ static bool is_centaur_cpu(const struct kvm_cpuid_param *param)
 	return boot_cpu_data.x86_vendor == X86_VENDOR_CENTAUR;
 }
 
-int kvm_dev_ioctl_get_supported_cpuid(struct kvm_cpuid2 *cpuid,
-				      struct kvm_cpuid_entry2 __user *entries)
+static bool sanity_check_entries(struct kvm_cpuid_entry2 __user *entries,
+				 __u32 num_entries, unsigned int ioctl_type)
+{
+	int i;
+	__u32 pad[3];
+
+	if (ioctl_type != KVM_GET_EMULATED_CPUID)
+		return false;
+
+	/*
+	 * We want to make sure that ->padding is being passed clean from
+	 * userspace in case we want to use it for something in the future.
+	 *
+	 * Sadly, this wasn't enforced for KVM_GET_SUPPORTED_CPUID and so we
+	 * have to give ourselves satisfied only with the emulated side. /me
+	 * sheds a tear.
+	 */
+	for (i = 0; i < num_entries; i++) {
+		if (copy_from_user(pad, entries[i].padding, sizeof(pad)))
+			return true;
+
+		if (pad[0] || pad[1] || pad[2])
+			return true;
+	}
+	return false;
+}
+
+int kvm_dev_ioctl_get_cpuid(struct kvm_cpuid2 *cpuid,
+			    struct kvm_cpuid_entry2 __user *entries,
+			    unsigned int type)
 {
 	struct kvm_cpuid_entry2 *cpuid_entries;
 	int limit, nent = 0, r = -E2BIG, i;
@@ -513,8 +606,12 @@ int kvm_dev_ioctl_get_supported_cpuid(struct kvm_cpuid2 *cpuid,
 		goto out;
 	if (cpuid->nent > KVM_MAX_CPUID_ENTRIES)
 		cpuid->nent = KVM_MAX_CPUID_ENTRIES;
+
+	if (sanity_check_entries(entries, cpuid->nent, type))
+		return -EINVAL;
+
 	r = -ENOMEM;
-	cpuid_entries = vmalloc(sizeof(struct kvm_cpuid_entry2) * cpuid->nent);
+	cpuid_entries = vzalloc(sizeof(struct kvm_cpuid_entry2) * cpuid->nent);
 	if (!cpuid_entries)
 		goto out;
 
@@ -526,7 +623,7 @@ int kvm_dev_ioctl_get_supported_cpuid(struct kvm_cpuid2 *cpuid,
 			continue;
 
 		r = do_cpuid_ent(&cpuid_entries[nent], ent->func, ent->idx,
-				&nent, cpuid->nent);
+				&nent, cpuid->nent, type);
 
 		if (r)
 			goto out_free;
@@ -537,7 +634,7 @@ int kvm_dev_ioctl_get_supported_cpuid(struct kvm_cpuid2 *cpuid,
 		limit = cpuid_entries[nent - 1].eax;
 		for (func = ent->func + 1; func <= limit && nent < cpuid->nent && r == 0; ++func)
 			r = do_cpuid_ent(&cpuid_entries[nent], func, ent->idx,
-				     &nent, cpuid->nent);
+				     &nent, cpuid->nent, type);
 
 		if (r)
 			goto out_free;
@@ -661,6 +758,7 @@ void kvm_cpuid(struct kvm_vcpu *vcpu, u32 *eax, u32 *ebx, u32 *ecx, u32 *edx)
 		*edx = best->edx;
 	} else
 		*eax = *ebx = *ecx = *edx = 0;
+	trace_kvm_cpuid(function, *eax, *ebx, *ecx, *edx);
 }
 EXPORT_SYMBOL_GPL(kvm_cpuid);
 
@@ -676,6 +774,5 @@ void kvm_emulate_cpuid(struct kvm_vcpu *vcpu)
 	kvm_register_write(vcpu, VCPU_REGS_RCX, ecx);
 	kvm_register_write(vcpu, VCPU_REGS_RDX, edx);
 	kvm_x86_ops->skip_emulated_instruction(vcpu);
-	trace_kvm_cpuid(function, eax, ebx, ecx, edx);
 }
 EXPORT_SYMBOL_GPL(kvm_emulate_cpuid);
diff --git a/arch/x86/kvm/cpuid.h b/arch/x86/kvm/cpuid.h
index b7fd07984888..f1e4895174b2 100644
--- a/arch/x86/kvm/cpuid.h
+++ b/arch/x86/kvm/cpuid.h
@@ -6,8 +6,9 @@
 void kvm_update_cpuid(struct kvm_vcpu *vcpu);
 struct kvm_cpuid_entry2 *kvm_find_cpuid_entry(struct kvm_vcpu *vcpu,
 					      u32 function, u32 index);
-int kvm_dev_ioctl_get_supported_cpuid(struct kvm_cpuid2 *cpuid,
-				      struct kvm_cpuid_entry2 __user *entries);
+int kvm_dev_ioctl_get_cpuid(struct kvm_cpuid2 *cpuid,
+			    struct kvm_cpuid_entry2 __user *entries,
+			    unsigned int type);
 int kvm_vcpu_ioctl_set_cpuid(struct kvm_vcpu *vcpu,
 			     struct kvm_cpuid *cpuid,
 			     struct kvm_cpuid_entry __user *entries);
diff --git a/arch/x86/kvm/emulate.c b/arch/x86/kvm/emulate.c
index ddc3f3d2afdb..07ffca0a89e9 100644
--- a/arch/x86/kvm/emulate.c
+++ b/arch/x86/kvm/emulate.c
@@ -130,7 +130,7 @@
 #define Mov         (1<<20)
 /* Misc flags */
 #define Prot        (1<<21) /* instruction generates #UD if not in prot-mode */
-#define VendorSpecific (1<<22) /* Vendor specific instruction */
+#define EmulateOnUD (1<<22) /* Emulate if unsupported by the host */
 #define NoAccess    (1<<23) /* Don't access memory (lea/invlpg/verr etc) */
 #define Op3264      (1<<24) /* Operand is 64b in long mode, 32b otherwise */
 #define Undefined   (1<<25) /* No Such Instruction */
@@ -785,9 +785,10 @@ static int do_insn_fetch(struct x86_emulate_ctxt *ctxt,
  * @highbyte_regs specifies whether to decode AH,CH,DH,BH.
  */
 static void *decode_register(struct x86_emulate_ctxt *ctxt, u8 modrm_reg,
-			     int highbyte_regs)
+			     int byteop)
 {
 	void *p;
+	int highbyte_regs = (ctxt->rex_prefix == 0) && byteop;
 
 	if (highbyte_regs && modrm_reg >= 4 && modrm_reg < 8)
 		p = (unsigned char *)reg_rmw(ctxt, modrm_reg & 3) + 1;
@@ -1024,7 +1025,6 @@ static void decode_register_operand(struct x86_emulate_ctxt *ctxt,
 				    struct operand *op)
 {
 	unsigned reg = ctxt->modrm_reg;
-	int highbyte_regs = ctxt->rex_prefix == 0;
 
 	if (!(ctxt->d & ModRM))
 		reg = (ctxt->b & 7) | ((ctxt->rex_prefix & 1) << 3);
@@ -1045,13 +1045,9 @@ static void decode_register_operand(struct x86_emulate_ctxt *ctxt,
 	}
 
 	op->type = OP_REG;
-	if (ctxt->d & ByteOp) {
-		op->addr.reg = decode_register(ctxt, reg, highbyte_regs);
-		op->bytes = 1;
-	} else {
-		op->addr.reg = decode_register(ctxt, reg, 0);
-		op->bytes = ctxt->op_bytes;
-	}
+	op->bytes = (ctxt->d & ByteOp) ? 1 : ctxt->op_bytes;
+	op->addr.reg = decode_register(ctxt, reg, ctxt->d & ByteOp);
+
 	fetch_register_operand(op);
 	op->orig_val = op->val;
 }
@@ -1082,12 +1078,10 @@ static int decode_modrm(struct x86_emulate_ctxt *ctxt,
 	ctxt->modrm_seg = VCPU_SREG_DS;
 
 	if (ctxt->modrm_mod == 3) {
-		int highbyte_regs = ctxt->rex_prefix == 0;
-
 		op->type = OP_REG;
 		op->bytes = (ctxt->d & ByteOp) ? 1 : ctxt->op_bytes;
 		op->addr.reg = decode_register(ctxt, ctxt->modrm_rm,
-					       highbyte_regs && (ctxt->d & ByteOp));
+				ctxt->d & ByteOp);
 		if (ctxt->d & Sse) {
 			op->type = OP_XMM;
 			op->bytes = 16;
@@ -2961,6 +2955,46 @@ static int em_mov(struct x86_emulate_ctxt *ctxt)
 	return X86EMUL_CONTINUE;
 }
 
+#define FFL(x) bit(X86_FEATURE_##x)
+
+static int em_movbe(struct x86_emulate_ctxt *ctxt)
+{
+	u32 ebx, ecx, edx, eax = 1;
+	u16 tmp;
+
+	/*
+	 * Check MOVBE is set in the guest-visible CPUID leaf.
+	 */
+	ctxt->ops->get_cpuid(ctxt, &eax, &ebx, &ecx, &edx);
+	if (!(ecx & FFL(MOVBE)))
+		return emulate_ud(ctxt);
+
+	switch (ctxt->op_bytes) {
+	case 2:
+		/*
+		 * From MOVBE definition: "...When the operand size is 16 bits,
+		 * the upper word of the destination register remains unchanged
+		 * ..."
+		 *
+		 * Both casting ->valptr and ->val to u16 breaks strict aliasing
+		 * rules so we have to do the operation almost per hand.
+		 */
+		tmp = (u16)ctxt->src.val;
+		ctxt->dst.val &= ~0xffffUL;
+		ctxt->dst.val |= (unsigned long)swab16(tmp);
+		break;
+	case 4:
+		ctxt->dst.val = swab32((u32)ctxt->src.val);
+		break;
+	case 8:
+		ctxt->dst.val = swab64(ctxt->src.val);
+		break;
+	default:
+		return X86EMUL_PROPAGATE_FAULT;
+	}
+	return X86EMUL_CONTINUE;
+}
+
 static int em_cr_write(struct x86_emulate_ctxt *ctxt)
 {
 	if (ctxt->ops->set_cr(ctxt, ctxt->modrm_reg, ctxt->src.val))
@@ -3256,6 +3290,18 @@ static int em_cpuid(struct x86_emulate_ctxt *ctxt)
 	return X86EMUL_CONTINUE;
 }
 
+static int em_sahf(struct x86_emulate_ctxt *ctxt)
+{
+	u32 flags;
+
+	flags = EFLG_CF | EFLG_PF | EFLG_AF | EFLG_ZF | EFLG_SF;
+	flags &= *reg_rmw(ctxt, VCPU_REGS_RAX) >> 8;
+
+	ctxt->eflags &= ~0xffUL;
+	ctxt->eflags |= flags | X86_EFLAGS_FIXED;
+	return X86EMUL_CONTINUE;
+}
+
 static int em_lahf(struct x86_emulate_ctxt *ctxt)
 {
 	*reg_rmw(ctxt, VCPU_REGS_RAX) &= ~0xff00UL;
@@ -3502,7 +3548,7 @@ static const struct opcode group7_rm1[] = {
 
 static const struct opcode group7_rm3[] = {
 	DIP(SrcNone | Prot | Priv,		vmrun,		check_svme_pa),
-	II(SrcNone  | Prot | VendorSpecific,	em_vmmcall,	vmmcall),
+	II(SrcNone  | Prot | EmulateOnUD,	em_vmmcall,	vmmcall),
 	DIP(SrcNone | Prot | Priv,		vmload,		check_svme_pa),
 	DIP(SrcNone | Prot | Priv,		vmsave,		check_svme_pa),
 	DIP(SrcNone | Prot | Priv,		stgi,		check_svme),
@@ -3587,7 +3633,7 @@ static const struct group_dual group7 = { {
 	II(SrcMem16 | Mov | Priv,		em_lmsw, lmsw),
 	II(SrcMem | ByteOp | Priv | NoAccess,	em_invlpg, invlpg),
 }, {
-	I(SrcNone | Priv | VendorSpecific,	em_vmcall),
+	I(SrcNone | Priv | EmulateOnUD,	em_vmcall),
 	EXT(0, group7_rm1),
 	N, EXT(0, group7_rm3),
 	II(SrcNone | DstMem | Mov,		em_smsw, smsw), N,
@@ -3750,7 +3796,8 @@ static const struct opcode opcode_table[256] = {
 	D(DstAcc | SrcNone), I(ImplicitOps | SrcAcc, em_cwd),
 	I(SrcImmFAddr | No64, em_call_far), N,
 	II(ImplicitOps | Stack, em_pushf, pushf),
-	II(ImplicitOps | Stack, em_popf, popf), N, I(ImplicitOps, em_lahf),
+	II(ImplicitOps | Stack, em_popf, popf),
+	I(ImplicitOps, em_sahf), I(ImplicitOps, em_lahf),
 	/* 0xA0 - 0xA7 */
 	I2bv(DstAcc | SrcMem | Mov | MemAbs, em_mov),
 	I2bv(DstMem | SrcAcc | Mov | MemAbs | PageTable, em_mov),
@@ -3810,7 +3857,7 @@ static const struct opcode opcode_table[256] = {
 static const struct opcode twobyte_table[256] = {
 	/* 0x00 - 0x0F */
 	G(0, group6), GD(0, &group7), N, N,
-	N, I(ImplicitOps | VendorSpecific, em_syscall),
+	N, I(ImplicitOps | EmulateOnUD, em_syscall),
 	II(ImplicitOps | Priv, em_clts, clts), N,
 	DI(ImplicitOps | Priv, invd), DI(ImplicitOps | Priv, wbinvd), N, N,
 	N, D(ImplicitOps | ModRM), N, N,
@@ -3830,8 +3877,8 @@ static const struct opcode twobyte_table[256] = {
 	IIP(ImplicitOps, em_rdtsc, rdtsc, check_rdtsc),
 	II(ImplicitOps | Priv, em_rdmsr, rdmsr),
 	IIP(ImplicitOps, em_rdpmc, rdpmc, check_rdpmc),
-	I(ImplicitOps | VendorSpecific, em_sysenter),
-	I(ImplicitOps | Priv | VendorSpecific, em_sysexit),
+	I(ImplicitOps | EmulateOnUD, em_sysenter),
+	I(ImplicitOps | Priv | EmulateOnUD, em_sysexit),
 	N, N,
 	N, N, N, N, N, N, N, N,
 	/* 0x40 - 0x4F */
@@ -3892,6 +3939,30 @@ static const struct opcode twobyte_table[256] = {
 	N, N, N, N, N, N, N, N, N, N, N, N, N, N, N, N
 };
 
+static const struct gprefix three_byte_0f_38_f0 = {
+	I(DstReg | SrcMem | Mov, em_movbe), N, N, N
+};
+
+static const struct gprefix three_byte_0f_38_f1 = {
+	I(DstMem | SrcReg | Mov, em_movbe), N, N, N
+};
+
+/*
+ * Insns below are selected by the prefix which indexed by the third opcode
+ * byte.
+ */
+static const struct opcode opcode_map_0f_38[256] = {
+	/* 0x00 - 0x7f */
+	X16(N), X16(N), X16(N), X16(N), X16(N), X16(N), X16(N), X16(N),
+	/* 0x80 - 0xef */
+	X16(N), X16(N), X16(N), X16(N), X16(N), X16(N), X16(N),
+	/* 0xf0 - 0xf1 */
+	GP(EmulateOnUD | ModRM | Prefix, &three_byte_0f_38_f0),
+	GP(EmulateOnUD | ModRM | Prefix, &three_byte_0f_38_f1),
+	/* 0xf2 - 0xff */
+	N, N, X4(N), X8(N)
+};
+
 #undef D
 #undef N
 #undef G
@@ -4040,7 +4111,8 @@ static int decode_operand(struct x86_emulate_ctxt *ctxt, struct operand *op,
 	case OpMem8:
 		ctxt->memop.bytes = 1;
 		if (ctxt->memop.type == OP_REG) {
-			ctxt->memop.addr.reg = decode_register(ctxt, ctxt->modrm_rm, 1);
+			ctxt->memop.addr.reg = decode_register(ctxt,
+					ctxt->modrm_rm, true);
 			fetch_register_operand(&ctxt->memop);
 		}
 		goto mem_common;
@@ -4126,6 +4198,7 @@ int x86_decode_insn(struct x86_emulate_ctxt *ctxt, void *insn, int insn_len)
 	ctxt->_eip = ctxt->eip;
 	ctxt->fetch.start = ctxt->_eip;
 	ctxt->fetch.end = ctxt->fetch.start + insn_len;
+	ctxt->opcode_len = 1;
 	if (insn_len > 0)
 		memcpy(ctxt->fetch.data, insn, insn_len);
 
@@ -4208,9 +4281,16 @@ done_prefixes:
 	opcode = opcode_table[ctxt->b];
 	/* Two-byte opcode? */
 	if (ctxt->b == 0x0f) {
-		ctxt->twobyte = 1;
+		ctxt->opcode_len = 2;
 		ctxt->b = insn_fetch(u8, ctxt);
 		opcode = twobyte_table[ctxt->b];
+
+		/* 0F_38 opcode map */
+		if (ctxt->b == 0x38) {
+			ctxt->opcode_len = 3;
+			ctxt->b = insn_fetch(u8, ctxt);
+			opcode = opcode_map_0f_38[ctxt->b];
+		}
 	}
 	ctxt->d = opcode.flags;
 
@@ -4267,7 +4347,7 @@ done_prefixes:
 	if (ctxt->d == 0 || (ctxt->d & NotImpl))
 		return EMULATION_FAILED;
 
-	if (!(ctxt->d & VendorSpecific) && ctxt->only_vendor_specific_insn)
+	if (!(ctxt->d & EmulateOnUD) && ctxt->ud)
 		return EMULATION_FAILED;
 
 	if (mode == X86EMUL_MODE_PROT64 && (ctxt->d & Stack))
@@ -4540,8 +4620,10 @@ special_insn:
 		goto writeback;
 	}
 
-	if (ctxt->twobyte)
+	if (ctxt->opcode_len == 2)
 		goto twobyte_insn;
+	else if (ctxt->opcode_len == 3)
+		goto threebyte_insn;
 
 	switch (ctxt->b) {
 	case 0x63:		/* movsxd */
@@ -4726,6 +4808,8 @@ twobyte_insn:
 		goto cannot_emulate;
 	}
 
+threebyte_insn:
+
 	if (rc != X86EMUL_CONTINUE)
 		goto done;
 
diff --git a/arch/x86/kvm/mmu.c b/arch/x86/kvm/mmu.c
index dce0df8150df..40772ef0f2b1 100644
--- a/arch/x86/kvm/mmu.c
+++ b/arch/x86/kvm/mmu.c
@@ -2570,11 +2570,6 @@ static void mmu_set_spte(struct kvm_vcpu *vcpu, u64 *sptep,
 	kvm_release_pfn_clean(pfn);
 }
 
-static void nonpaging_new_cr3(struct kvm_vcpu *vcpu)
-{
-	mmu_free_roots(vcpu);
-}
-
 static pfn_t pte_prefetch_gfn_to_pfn(struct kvm_vcpu *vcpu, gfn_t gfn,
 				     bool no_dirty_log)
 {
@@ -3424,18 +3419,11 @@ out_unlock:
 	return 0;
 }
 
-static void nonpaging_free(struct kvm_vcpu *vcpu)
-{
-	mmu_free_roots(vcpu);
-}
-
-static int nonpaging_init_context(struct kvm_vcpu *vcpu,
-				  struct kvm_mmu *context)
+static void nonpaging_init_context(struct kvm_vcpu *vcpu,
+				   struct kvm_mmu *context)
 {
-	context->new_cr3 = nonpaging_new_cr3;
 	context->page_fault = nonpaging_page_fault;
 	context->gva_to_gpa = nonpaging_gva_to_gpa;
-	context->free = nonpaging_free;
 	context->sync_page = nonpaging_sync_page;
 	context->invlpg = nonpaging_invlpg;
 	context->update_pte = nonpaging_update_pte;
@@ -3444,7 +3432,6 @@ static int nonpaging_init_context(struct kvm_vcpu *vcpu,
 	context->root_hpa = INVALID_PAGE;
 	context->direct_map = true;
 	context->nx = false;
-	return 0;
 }
 
 void kvm_mmu_flush_tlb(struct kvm_vcpu *vcpu)
@@ -3454,9 +3441,8 @@ void kvm_mmu_flush_tlb(struct kvm_vcpu *vcpu)
 }
 EXPORT_SYMBOL_GPL(kvm_mmu_flush_tlb);
 
-static void paging_new_cr3(struct kvm_vcpu *vcpu)
+void kvm_mmu_new_cr3(struct kvm_vcpu *vcpu)
 {
-	pgprintk("%s: cr3 %lx\n", __func__, kvm_read_cr3(vcpu));
 	mmu_free_roots(vcpu);
 }
 
@@ -3471,11 +3457,6 @@ static void inject_page_fault(struct kvm_vcpu *vcpu,
 	vcpu->arch.mmu.inject_page_fault(vcpu, fault);
 }
 
-static void paging_free(struct kvm_vcpu *vcpu)
-{
-	nonpaging_free(vcpu);
-}
-
 static bool sync_mmio_spte(struct kvm *kvm, u64 *sptep, gfn_t gfn,
 			   unsigned access, int *nr_present)
 {
@@ -3665,9 +3646,9 @@ static void update_last_pte_bitmap(struct kvm_vcpu *vcpu, struct kvm_mmu *mmu)
 	mmu->last_pte_bitmap = map;
 }
 
-static int paging64_init_context_common(struct kvm_vcpu *vcpu,
-					struct kvm_mmu *context,
-					int level)
+static void paging64_init_context_common(struct kvm_vcpu *vcpu,
+					 struct kvm_mmu *context,
+					 int level)
 {
 	context->nx = is_nx(vcpu);
 	context->root_level = level;
@@ -3677,27 +3658,24 @@ static int paging64_init_context_common(struct kvm_vcpu *vcpu,
 	update_last_pte_bitmap(vcpu, context);
 
 	ASSERT(is_pae(vcpu));
-	context->new_cr3 = paging_new_cr3;
 	context->page_fault = paging64_page_fault;
 	context->gva_to_gpa = paging64_gva_to_gpa;
 	context->sync_page = paging64_sync_page;
 	context->invlpg = paging64_invlpg;
 	context->update_pte = paging64_update_pte;
-	context->free = paging_free;
 	context->shadow_root_level = level;
 	context->root_hpa = INVALID_PAGE;
 	context->direct_map = false;
-	return 0;
 }
 
-static int paging64_init_context(struct kvm_vcpu *vcpu,
-				 struct kvm_mmu *context)
+static void paging64_init_context(struct kvm_vcpu *vcpu,
+				  struct kvm_mmu *context)
 {
-	return paging64_init_context_common(vcpu, context, PT64_ROOT_LEVEL);
+	paging64_init_context_common(vcpu, context, PT64_ROOT_LEVEL);
 }
 
-static int paging32_init_context(struct kvm_vcpu *vcpu,
-				 struct kvm_mmu *context)
+static void paging32_init_context(struct kvm_vcpu *vcpu,
+				  struct kvm_mmu *context)
 {
 	context->nx = false;
 	context->root_level = PT32_ROOT_LEVEL;
@@ -3706,33 +3684,28 @@ static int paging32_init_context(struct kvm_vcpu *vcpu,
 	update_permission_bitmask(vcpu, context, false);
 	update_last_pte_bitmap(vcpu, context);
 
-	context->new_cr3 = paging_new_cr3;
 	context->page_fault = paging32_page_fault;
 	context->gva_to_gpa = paging32_gva_to_gpa;
-	context->free = paging_free;
 	context->sync_page = paging32_sync_page;
 	context->invlpg = paging32_invlpg;
 	context->update_pte = paging32_update_pte;
 	context->shadow_root_level = PT32E_ROOT_LEVEL;
 	context->root_hpa = INVALID_PAGE;
 	context->direct_map = false;
-	return 0;
 }
 
-static int paging32E_init_context(struct kvm_vcpu *vcpu,
-				  struct kvm_mmu *context)
+static void paging32E_init_context(struct kvm_vcpu *vcpu,
+				   struct kvm_mmu *context)
 {
-	return paging64_init_context_common(vcpu, context, PT32E_ROOT_LEVEL);
+	paging64_init_context_common(vcpu, context, PT32E_ROOT_LEVEL);
 }
 
-static int init_kvm_tdp_mmu(struct kvm_vcpu *vcpu)
+static void init_kvm_tdp_mmu(struct kvm_vcpu *vcpu)
 {
 	struct kvm_mmu *context = vcpu->arch.walk_mmu;
 
 	context->base_role.word = 0;
-	context->new_cr3 = nonpaging_new_cr3;
 	context->page_fault = tdp_page_fault;
-	context->free = nonpaging_free;
 	context->sync_page = nonpaging_sync_page;
 	context->invlpg = nonpaging_invlpg;
 	context->update_pte = nonpaging_update_pte;
@@ -3767,37 +3740,32 @@ static int init_kvm_tdp_mmu(struct kvm_vcpu *vcpu)
 
 	update_permission_bitmask(vcpu, context, false);
 	update_last_pte_bitmap(vcpu, context);
-
-	return 0;
 }
 
-int kvm_init_shadow_mmu(struct kvm_vcpu *vcpu, struct kvm_mmu *context)
+void kvm_init_shadow_mmu(struct kvm_vcpu *vcpu, struct kvm_mmu *context)
 {
-	int r;
 	bool smep = kvm_read_cr4_bits(vcpu, X86_CR4_SMEP);
 	ASSERT(vcpu);
 	ASSERT(!VALID_PAGE(vcpu->arch.mmu.root_hpa));
 
 	if (!is_paging(vcpu))
-		r = nonpaging_init_context(vcpu, context);
+		nonpaging_init_context(vcpu, context);
 	else if (is_long_mode(vcpu))
-		r = paging64_init_context(vcpu, context);
+		paging64_init_context(vcpu, context);
 	else if (is_pae(vcpu))
-		r = paging32E_init_context(vcpu, context);
+		paging32E_init_context(vcpu, context);
 	else
-		r = paging32_init_context(vcpu, context);
+		paging32_init_context(vcpu, context);
 
 	vcpu->arch.mmu.base_role.nxe = is_nx(vcpu);
 	vcpu->arch.mmu.base_role.cr4_pae = !!is_pae(vcpu);
 	vcpu->arch.mmu.base_role.cr0_wp  = is_write_protection(vcpu);
 	vcpu->arch.mmu.base_role.smep_andnot_wp
 		= smep && !is_write_protection(vcpu);
-
-	return r;
 }
 EXPORT_SYMBOL_GPL(kvm_init_shadow_mmu);
 
-int kvm_init_shadow_ept_mmu(struct kvm_vcpu *vcpu, struct kvm_mmu *context,
+void kvm_init_shadow_ept_mmu(struct kvm_vcpu *vcpu, struct kvm_mmu *context,
 		bool execonly)
 {
 	ASSERT(vcpu);
@@ -3806,37 +3774,30 @@ int kvm_init_shadow_ept_mmu(struct kvm_vcpu *vcpu, struct kvm_mmu *context,
 	context->shadow_root_level = kvm_x86_ops->get_tdp_level();
 
 	context->nx = true;
-	context->new_cr3 = paging_new_cr3;
 	context->page_fault = ept_page_fault;
 	context->gva_to_gpa = ept_gva_to_gpa;
 	context->sync_page = ept_sync_page;
 	context->invlpg = ept_invlpg;
 	context->update_pte = ept_update_pte;
-	context->free = paging_free;
 	context->root_level = context->shadow_root_level;
 	context->root_hpa = INVALID_PAGE;
 	context->direct_map = false;
 
 	update_permission_bitmask(vcpu, context, true);
 	reset_rsvds_bits_mask_ept(vcpu, context, execonly);
-
-	return 0;
 }
 EXPORT_SYMBOL_GPL(kvm_init_shadow_ept_mmu);
 
-static int init_kvm_softmmu(struct kvm_vcpu *vcpu)
+static void init_kvm_softmmu(struct kvm_vcpu *vcpu)
 {
-	int r = kvm_init_shadow_mmu(vcpu, vcpu->arch.walk_mmu);
-
+	kvm_init_shadow_mmu(vcpu, vcpu->arch.walk_mmu);
 	vcpu->arch.walk_mmu->set_cr3           = kvm_x86_ops->set_cr3;
 	vcpu->arch.walk_mmu->get_cr3           = get_cr3;
 	vcpu->arch.walk_mmu->get_pdptr         = kvm_pdptr_read;
 	vcpu->arch.walk_mmu->inject_page_fault = kvm_inject_page_fault;
-
-	return r;
 }
 
-static int init_kvm_nested_mmu(struct kvm_vcpu *vcpu)
+static void init_kvm_nested_mmu(struct kvm_vcpu *vcpu)
 {
 	struct kvm_mmu *g_context = &vcpu->arch.nested_mmu;
 
@@ -3873,11 +3834,9 @@ static int init_kvm_nested_mmu(struct kvm_vcpu *vcpu)
 
 	update_permission_bitmask(vcpu, g_context, false);
 	update_last_pte_bitmap(vcpu, g_context);
-
-	return 0;
 }
 
-static int init_kvm_mmu(struct kvm_vcpu *vcpu)
+static void init_kvm_mmu(struct kvm_vcpu *vcpu)
 {
 	if (mmu_is_nested(vcpu))
 		return init_kvm_nested_mmu(vcpu);
@@ -3887,18 +3846,12 @@ static int init_kvm_mmu(struct kvm_vcpu *vcpu)
 		return init_kvm_softmmu(vcpu);
 }
 
-static void destroy_kvm_mmu(struct kvm_vcpu *vcpu)
+void kvm_mmu_reset_context(struct kvm_vcpu *vcpu)
 {
 	ASSERT(vcpu);
-	if (VALID_PAGE(vcpu->arch.mmu.root_hpa))
-		/* mmu.free() should set root_hpa = INVALID_PAGE */
-		vcpu->arch.mmu.free(vcpu);
-}
 
-int kvm_mmu_reset_context(struct kvm_vcpu *vcpu)
-{
-	destroy_kvm_mmu(vcpu);
-	return init_kvm_mmu(vcpu);
+	kvm_mmu_unload(vcpu);
+	init_kvm_mmu(vcpu);
 }
 EXPORT_SYMBOL_GPL(kvm_mmu_reset_context);
 
@@ -3923,6 +3876,7 @@ EXPORT_SYMBOL_GPL(kvm_mmu_load);
 void kvm_mmu_unload(struct kvm_vcpu *vcpu)
 {
 	mmu_free_roots(vcpu);
+	WARN_ON(VALID_PAGE(vcpu->arch.mmu.root_hpa));
 }
 EXPORT_SYMBOL_GPL(kvm_mmu_unload);
 
@@ -4281,12 +4235,12 @@ int kvm_mmu_create(struct kvm_vcpu *vcpu)
 	return alloc_mmu_pages(vcpu);
 }
 
-int kvm_mmu_setup(struct kvm_vcpu *vcpu)
+void kvm_mmu_setup(struct kvm_vcpu *vcpu)
 {
 	ASSERT(vcpu);
 	ASSERT(!VALID_PAGE(vcpu->arch.mmu.root_hpa));
 
-	return init_kvm_mmu(vcpu);
+	init_kvm_mmu(vcpu);
 }
 
 void kvm_mmu_slot_remove_write_access(struct kvm *kvm, int slot)
@@ -4428,7 +4382,7 @@ mmu_shrink_scan(struct shrinker *shrink, struct shrink_control *sc)
 	int nr_to_scan = sc->nr_to_scan;
 	unsigned long freed = 0;
 
-	raw_spin_lock(&kvm_lock);
+	spin_lock(&kvm_lock);
 
 	list_for_each_entry(kvm, &vm_list, vm_list) {
 		int idx;
@@ -4478,9 +4432,8 @@ unlock:
 		break;
 	}
 
-	raw_spin_unlock(&kvm_lock);
+	spin_unlock(&kvm_lock);
 	return freed;
-
 }
 
 static unsigned long
@@ -4574,7 +4527,7 @@ void kvm_mmu_destroy(struct kvm_vcpu *vcpu)
 {
 	ASSERT(vcpu);
 
-	destroy_kvm_mmu(vcpu);
+	kvm_mmu_unload(vcpu);
 	free_mmu_pages(vcpu);
 	mmu_free_memory_caches(vcpu);
 }
diff --git a/arch/x86/kvm/mmu.h b/arch/x86/kvm/mmu.h
index 77e044a0f5f7..292615274358 100644
--- a/arch/x86/kvm/mmu.h
+++ b/arch/x86/kvm/mmu.h
@@ -70,8 +70,8 @@ enum {
 };
 
 int handle_mmio_page_fault_common(struct kvm_vcpu *vcpu, u64 addr, bool direct);
-int kvm_init_shadow_mmu(struct kvm_vcpu *vcpu, struct kvm_mmu *context);
-int kvm_init_shadow_ept_mmu(struct kvm_vcpu *vcpu, struct kvm_mmu *context,
+void kvm_init_shadow_mmu(struct kvm_vcpu *vcpu, struct kvm_mmu *context);
+void kvm_init_shadow_ept_mmu(struct kvm_vcpu *vcpu, struct kvm_mmu *context,
 		bool execonly);
 
 static inline unsigned int kvm_mmu_available_pages(struct kvm *kvm)
diff --git a/arch/x86/kvm/svm.c b/arch/x86/kvm/svm.c
index c0bc80391e40..c7168a5cff1b 100644
--- a/arch/x86/kvm/svm.c
+++ b/arch/x86/kvm/svm.c
@@ -1959,11 +1959,9 @@ static void nested_svm_inject_npf_exit(struct kvm_vcpu *vcpu,
 	nested_svm_vmexit(svm);
 }
 
-static int nested_svm_init_mmu_context(struct kvm_vcpu *vcpu)
+static void nested_svm_init_mmu_context(struct kvm_vcpu *vcpu)
 {
-	int r;
-
-	r = kvm_init_shadow_mmu(vcpu, &vcpu->arch.mmu);
+	kvm_init_shadow_mmu(vcpu, &vcpu->arch.mmu);
 
 	vcpu->arch.mmu.set_cr3           = nested_svm_set_tdp_cr3;
 	vcpu->arch.mmu.get_cr3           = nested_svm_get_tdp_cr3;
@@ -1971,8 +1969,6 @@ static int nested_svm_init_mmu_context(struct kvm_vcpu *vcpu)
 	vcpu->arch.mmu.inject_page_fault = nested_svm_inject_npf_exit;
 	vcpu->arch.mmu.shadow_root_level = get_npt_level();
 	vcpu->arch.walk_mmu              = &vcpu->arch.nested_mmu;
-
-	return r;
 }
 
 static void nested_svm_uninit_mmu_context(struct kvm_vcpu *vcpu)
diff --git a/arch/x86/kvm/vmx.c b/arch/x86/kvm/vmx.c
index 2b2fce1b2009..b2fe1c252f35 100644
--- a/arch/x86/kvm/vmx.c
+++ b/arch/x86/kvm/vmx.c
@@ -1498,7 +1498,7 @@ static void add_atomic_switch_msr(struct vcpu_vmx *vmx, unsigned msr,
 			break;
 
 	if (i == NR_AUTOLOAD_MSRS) {
-		printk_once(KERN_WARNING"Not enough mst switch entries. "
+		printk_once(KERN_WARNING "Not enough msr switch entries. "
 				"Can't add msr %x\n", msr);
 		return;
 	} else if (i == m->nr) {
@@ -1898,16 +1898,12 @@ static void skip_emulated_instruction(struct kvm_vcpu *vcpu)
 /*
  * KVM wants to inject page-faults which it got to the guest. This function
  * checks whether in a nested guest, we need to inject them to L1 or L2.
- * This function assumes it is called with the exit reason in vmcs02 being
- * a #PF exception (this is the only case in which KVM injects a #PF when L2
- * is running).
  */
-static int nested_pf_handled(struct kvm_vcpu *vcpu)
+static int nested_vmx_check_exception(struct kvm_vcpu *vcpu, unsigned nr)
 {
 	struct vmcs12 *vmcs12 = get_vmcs12(vcpu);
 
-	/* TODO: also check PFEC_MATCH/MASK, not just EB.PF. */
-	if (!(vmcs12->exception_bitmap & (1u << PF_VECTOR)))
+	if (!(vmcs12->exception_bitmap & (1u << nr)))
 		return 0;
 
 	nested_vmx_vmexit(vcpu);
@@ -1921,8 +1917,8 @@ static void vmx_queue_exception(struct kvm_vcpu *vcpu, unsigned nr,
 	struct vcpu_vmx *vmx = to_vmx(vcpu);
 	u32 intr_info = nr | INTR_INFO_VALID_MASK;
 
-	if (nr == PF_VECTOR && is_guest_mode(vcpu) &&
-	    !vmx->nested.nested_run_pending && nested_pf_handled(vcpu))
+	if (!reinject && is_guest_mode(vcpu) &&
+	    nested_vmx_check_exception(vcpu, nr))
 		return;
 
 	if (has_error_code) {
@@ -2204,9 +2200,15 @@ static __init void nested_vmx_setup_ctls_msrs(void)
 #ifdef CONFIG_X86_64
 		VM_EXIT_HOST_ADDR_SPACE_SIZE |
 #endif
-		VM_EXIT_LOAD_IA32_PAT | VM_EXIT_SAVE_IA32_PAT;
+		VM_EXIT_LOAD_IA32_PAT | VM_EXIT_SAVE_IA32_PAT |
+		VM_EXIT_SAVE_VMX_PREEMPTION_TIMER;
+	if (!(nested_vmx_pinbased_ctls_high & PIN_BASED_VMX_PREEMPTION_TIMER) ||
+	    !(nested_vmx_exit_ctls_high & VM_EXIT_SAVE_VMX_PREEMPTION_TIMER)) {
+		nested_vmx_exit_ctls_high &= ~VM_EXIT_SAVE_VMX_PREEMPTION_TIMER;
+		nested_vmx_pinbased_ctls_high &= ~PIN_BASED_VMX_PREEMPTION_TIMER;
+	}
 	nested_vmx_exit_ctls_high |= (VM_EXIT_ALWAYSON_WITHOUT_TRUE_MSR |
-				      VM_EXIT_LOAD_IA32_EFER);
+		VM_EXIT_LOAD_IA32_EFER | VM_EXIT_SAVE_IA32_EFER);
 
 	/* entry controls */
 	rdmsr(MSR_IA32_VMX_ENTRY_CTLS,
@@ -2226,7 +2228,8 @@ static __init void nested_vmx_setup_ctls_msrs(void)
 		nested_vmx_procbased_ctls_low, nested_vmx_procbased_ctls_high);
 	nested_vmx_procbased_ctls_low = 0;
 	nested_vmx_procbased_ctls_high &=
-		CPU_BASED_VIRTUAL_INTR_PENDING | CPU_BASED_USE_TSC_OFFSETING |
+		CPU_BASED_VIRTUAL_INTR_PENDING |
+		CPU_BASED_VIRTUAL_NMI_PENDING | CPU_BASED_USE_TSC_OFFSETING |
 		CPU_BASED_HLT_EXITING | CPU_BASED_INVLPG_EXITING |
 		CPU_BASED_MWAIT_EXITING | CPU_BASED_CR3_LOAD_EXITING |
 		CPU_BASED_CR3_STORE_EXITING |
@@ -2252,13 +2255,15 @@ static __init void nested_vmx_setup_ctls_msrs(void)
 	nested_vmx_secondary_ctls_low = 0;
 	nested_vmx_secondary_ctls_high &=
 		SECONDARY_EXEC_VIRTUALIZE_APIC_ACCESSES |
+		SECONDARY_EXEC_UNRESTRICTED_GUEST |
 		SECONDARY_EXEC_WBINVD_EXITING;
 
 	if (enable_ept) {
 		/* nested EPT: emulate EPT also to L1 */
 		nested_vmx_secondary_ctls_high |= SECONDARY_EXEC_ENABLE_EPT;
 		nested_vmx_ept_caps = VMX_EPT_PAGE_WALK_4_BIT |
-			 VMX_EPTP_WB_BIT | VMX_EPT_INVEPT_BIT;
+			 VMX_EPTP_WB_BIT | VMX_EPT_2MB_PAGE_BIT |
+			 VMX_EPT_INVEPT_BIT;
 		nested_vmx_ept_caps &= vmx_capability.ept;
 		/*
 		 * Since invept is completely emulated we support both global
@@ -3380,8 +3385,10 @@ static void vmx_set_cr3(struct kvm_vcpu *vcpu, unsigned long cr3)
 	if (enable_ept) {
 		eptp = construct_eptp(cr3);
 		vmcs_write64(EPT_POINTER, eptp);
-		guest_cr3 = is_paging(vcpu) ? kvm_read_cr3(vcpu) :
-			vcpu->kvm->arch.ept_identity_map_addr;
+		if (is_paging(vcpu) || is_guest_mode(vcpu))
+			guest_cr3 = kvm_read_cr3(vcpu);
+		else
+			guest_cr3 = vcpu->kvm->arch.ept_identity_map_addr;
 		ept_load_pdptrs(vcpu);
 	}
 
@@ -4879,6 +4886,17 @@ vmx_patch_hypercall(struct kvm_vcpu *vcpu, unsigned char *hypercall)
 	hypercall[2] = 0xc1;
 }
 
+static bool nested_cr0_valid(struct vmcs12 *vmcs12, unsigned long val)
+{
+	unsigned long always_on = VMXON_CR0_ALWAYSON;
+
+	if (nested_vmx_secondary_ctls_high &
+		SECONDARY_EXEC_UNRESTRICTED_GUEST &&
+	    nested_cpu_has2(vmcs12, SECONDARY_EXEC_UNRESTRICTED_GUEST))
+		always_on &= ~(X86_CR0_PE | X86_CR0_PG);
+	return (val & always_on) == always_on;
+}
+
 /* called to set cr0 as appropriate for a mov-to-cr0 exit. */
 static int handle_set_cr0(struct kvm_vcpu *vcpu, unsigned long val)
 {
@@ -4897,9 +4915,7 @@ static int handle_set_cr0(struct kvm_vcpu *vcpu, unsigned long val)
 		val = (val & ~vmcs12->cr0_guest_host_mask) |
 			(vmcs12->guest_cr0 & vmcs12->cr0_guest_host_mask);
 
-		/* TODO: will have to take unrestricted guest mode into
-		 * account */
-		if ((val & VMXON_CR0_ALWAYSON) != VMXON_CR0_ALWAYSON)
+		if (!nested_cr0_valid(vmcs12, val))
 			return 1;
 
 		if (kvm_set_cr0(vcpu, val))
@@ -6627,6 +6643,9 @@ static bool nested_vmx_exit_handled(struct kvm_vcpu *vcpu)
 			return 0;
 		else if (is_page_fault(intr_info))
 			return enable_ept;
+		else if (is_no_device(intr_info) &&
+			 !(nested_read_cr0(vmcs12) & X86_CR0_TS))
+			return 0;
 		return vmcs12->exception_bitmap &
 				(1u << (intr_info & INTR_INFO_VECTOR_MASK));
 	case EXIT_REASON_EXTERNAL_INTERRUPT:
@@ -6722,6 +6741,27 @@ static void vmx_get_exit_info(struct kvm_vcpu *vcpu, u64 *info1, u64 *info2)
 	*info2 = vmcs_read32(VM_EXIT_INTR_INFO);
 }
 
+static void nested_adjust_preemption_timer(struct kvm_vcpu *vcpu)
+{
+	u64 delta_tsc_l1;
+	u32 preempt_val_l1, preempt_val_l2, preempt_scale;
+
+	if (!(get_vmcs12(vcpu)->pin_based_vm_exec_control &
+			PIN_BASED_VMX_PREEMPTION_TIMER))
+		return;
+	preempt_scale = native_read_msr(MSR_IA32_VMX_MISC) &
+			MSR_IA32_VMX_MISC_PREEMPTION_TIMER_SCALE;
+	preempt_val_l2 = vmcs_read32(VMX_PREEMPTION_TIMER_VALUE);
+	delta_tsc_l1 = vmx_read_l1_tsc(vcpu, native_read_tsc())
+		- vcpu->arch.last_guest_tsc;
+	preempt_val_l1 = delta_tsc_l1 >> preempt_scale;
+	if (preempt_val_l2 <= preempt_val_l1)
+		preempt_val_l2 = 0;
+	else
+		preempt_val_l2 -= preempt_val_l1;
+	vmcs_write32(VMX_PREEMPTION_TIMER_VALUE, preempt_val_l2);
+}
+
 /*
  * The guest has exited.  See if we can fix it or if we need userspace
  * assistance.
@@ -6736,20 +6776,6 @@ static int vmx_handle_exit(struct kvm_vcpu *vcpu)
 	if (vmx->emulation_required)
 		return handle_invalid_guest_state(vcpu);
 
-	/*
-	 * the KVM_REQ_EVENT optimization bit is only on for one entry, and if
-	 * we did not inject a still-pending event to L1 now because of
-	 * nested_run_pending, we need to re-enable this bit.
-	 */
-	if (vmx->nested.nested_run_pending)
-		kvm_make_request(KVM_REQ_EVENT, vcpu);
-
-	if (!is_guest_mode(vcpu) && (exit_reason == EXIT_REASON_VMLAUNCH ||
-	    exit_reason == EXIT_REASON_VMRESUME))
-		vmx->nested.nested_run_pending = 1;
-	else
-		vmx->nested.nested_run_pending = 0;
-
 	if (is_guest_mode(vcpu) && nested_vmx_exit_handled(vcpu)) {
 		nested_vmx_vmexit(vcpu);
 		return 1;
@@ -7061,9 +7087,9 @@ static void __vmx_complete_interrupts(struct kvm_vcpu *vcpu,
 	case INTR_TYPE_HARD_EXCEPTION:
 		if (idt_vectoring_info & VECTORING_INFO_DELIVER_CODE_MASK) {
 			u32 err = vmcs_read32(error_code_field);
-			kvm_queue_exception_e(vcpu, vector, err);
+			kvm_requeue_exception_e(vcpu, vector, err);
 		} else
-			kvm_queue_exception(vcpu, vector);
+			kvm_requeue_exception(vcpu, vector);
 		break;
 	case INTR_TYPE_SOFT_INTR:
 		vcpu->arch.event_exit_inst_len = vmcs_read32(instr_len_field);
@@ -7146,6 +7172,8 @@ static void __noclone vmx_vcpu_run(struct kvm_vcpu *vcpu)
 	atomic_switch_perf_msrs(vmx);
 	debugctlmsr = get_debugctlmsr();
 
+	if (is_guest_mode(vcpu) && !vmx->nested.nested_run_pending)
+		nested_adjust_preemption_timer(vcpu);
 	vmx->__launched = vmx->loaded_vmcs->launched;
 	asm(
 		/* Store host registers */
@@ -7284,6 +7312,16 @@ static void __noclone vmx_vcpu_run(struct kvm_vcpu *vcpu)
 	vmx->exit_reason = vmcs_read32(VM_EXIT_REASON);
 	trace_kvm_exit(vmx->exit_reason, vcpu, KVM_ISA_VMX);
 
+	/*
+	 * the KVM_REQ_EVENT optimization bit is only on for one entry, and if
+	 * we did not inject a still-pending event to L1 now because of
+	 * nested_run_pending, we need to re-enable this bit.
+	 */
+	if (vmx->nested.nested_run_pending)
+		kvm_make_request(KVM_REQ_EVENT, vcpu);
+
+	vmx->nested.nested_run_pending = 0;
+
 	vmx_complete_atomic_exit(vmx);
 	vmx_recover_nmi_blocking(vmx);
 	vmx_complete_interrupts(vmx);
@@ -7410,8 +7448,7 @@ static u64 vmx_get_mt_mask(struct kvm_vcpu *vcpu, gfn_t gfn, bool is_mmio)
 	 */
 	if (is_mmio)
 		ret = MTRR_TYPE_UNCACHABLE << VMX_EPT_MT_EPTE_SHIFT;
-	else if (vcpu->kvm->arch.iommu_domain &&
-		!(vcpu->kvm->arch.iommu_flags & KVM_IOMMU_CACHE_COHERENCY))
+	else if (kvm_arch_has_noncoherent_dma(vcpu->kvm))
 		ret = kvm_get_guest_memory_type(vcpu, gfn) <<
 		      VMX_EPT_MT_EPTE_SHIFT;
 	else
@@ -7501,9 +7538,9 @@ static unsigned long nested_ept_get_cr3(struct kvm_vcpu *vcpu)
 	return get_vmcs12(vcpu)->ept_pointer;
 }
 
-static int nested_ept_init_mmu_context(struct kvm_vcpu *vcpu)
+static void nested_ept_init_mmu_context(struct kvm_vcpu *vcpu)
 {
-	int r = kvm_init_shadow_ept_mmu(vcpu, &vcpu->arch.mmu,
+	kvm_init_shadow_ept_mmu(vcpu, &vcpu->arch.mmu,
 			nested_vmx_ept_caps & VMX_EPT_EXECUTE_ONLY_BIT);
 
 	vcpu->arch.mmu.set_cr3           = vmx_set_cr3;
@@ -7511,8 +7548,6 @@ static int nested_ept_init_mmu_context(struct kvm_vcpu *vcpu)
 	vcpu->arch.mmu.inject_page_fault = nested_ept_inject_page_fault;
 
 	vcpu->arch.walk_mmu              = &vcpu->arch.nested_mmu;
-
-	return r;
 }
 
 static void nested_ept_uninit_mmu_context(struct kvm_vcpu *vcpu)
@@ -7520,6 +7555,20 @@ static void nested_ept_uninit_mmu_context(struct kvm_vcpu *vcpu)
 	vcpu->arch.walk_mmu = &vcpu->arch.mmu;
 }
 
+static void vmx_inject_page_fault_nested(struct kvm_vcpu *vcpu,
+		struct x86_exception *fault)
+{
+	struct vmcs12 *vmcs12 = get_vmcs12(vcpu);
+
+	WARN_ON(!is_guest_mode(vcpu));
+
+	/* TODO: also check PFEC_MATCH/MASK, not just EB.PF. */
+	if (vmcs12->exception_bitmap & (1u << PF_VECTOR))
+		nested_vmx_vmexit(vcpu);
+	else
+		kvm_inject_page_fault(vcpu, fault);
+}
+
 /*
  * prepare_vmcs02 is called when the L1 guest hypervisor runs its nested
  * L2 guest. L1 has a vmcs for L2 (vmcs12), and this function "merges" it
@@ -7533,6 +7582,7 @@ static void prepare_vmcs02(struct kvm_vcpu *vcpu, struct vmcs12 *vmcs12)
 {
 	struct vcpu_vmx *vmx = to_vmx(vcpu);
 	u32 exec_control;
+	u32 exit_control;
 
 	vmcs_write16(GUEST_ES_SELECTOR, vmcs12->guest_es_selector);
 	vmcs_write16(GUEST_CS_SELECTOR, vmcs12->guest_cs_selector);
@@ -7706,7 +7756,10 @@ static void prepare_vmcs02(struct kvm_vcpu *vcpu, struct vmcs12 *vmcs12)
 	 * we should use its exit controls. Note that VM_EXIT_LOAD_IA32_EFER
 	 * bits are further modified by vmx_set_efer() below.
 	 */
-	vmcs_write32(VM_EXIT_CONTROLS, vmcs_config.vmexit_ctrl);
+	exit_control = vmcs_config.vmexit_ctrl;
+	if (vmcs12->pin_based_vm_exec_control & PIN_BASED_VMX_PREEMPTION_TIMER)
+		exit_control |= VM_EXIT_SAVE_VMX_PREEMPTION_TIMER;
+	vmcs_write32(VM_EXIT_CONTROLS, exit_control);
 
 	/* vmcs12's VM_ENTRY_LOAD_IA32_EFER and VM_ENTRY_IA32E_MODE are
 	 * emulated by vmx_set_efer(), below.
@@ -7773,6 +7826,9 @@ static void prepare_vmcs02(struct kvm_vcpu *vcpu, struct vmcs12 *vmcs12)
 	kvm_set_cr3(vcpu, vmcs12->guest_cr3);
 	kvm_mmu_reset_context(vcpu);
 
+	if (!enable_ept)
+		vcpu->arch.walk_mmu->inject_page_fault = vmx_inject_page_fault_nested;
+
 	/*
 	 * L1 may access the L2's PDPTR, so save them to construct vmcs12
 	 */
@@ -7876,7 +7932,7 @@ static int nested_vmx_run(struct kvm_vcpu *vcpu, bool launch)
 		return 1;
 	}
 
-	if (((vmcs12->guest_cr0 & VMXON_CR0_ALWAYSON) != VMXON_CR0_ALWAYSON) ||
+	if (!nested_cr0_valid(vmcs12, vmcs12->guest_cr0) ||
 	    ((vmcs12->guest_cr4 & VMXON_CR4_ALWAYSON) != VMXON_CR4_ALWAYSON)) {
 		nested_vmx_entry_failure(vcpu, vmcs12,
 			EXIT_REASON_INVALID_STATE, ENTRY_FAIL_DEFAULT);
@@ -7938,6 +7994,8 @@ static int nested_vmx_run(struct kvm_vcpu *vcpu, bool launch)
 
 	enter_guest_mode(vcpu);
 
+	vmx->nested.nested_run_pending = 1;
+
 	vmx->nested.vmcs01_tsc_offset = vmcs_read64(TSC_OFFSET);
 
 	cpu = get_cpu();
@@ -8005,7 +8063,7 @@ static void vmcs12_save_pending_event(struct kvm_vcpu *vcpu,
 	u32 idt_vectoring;
 	unsigned int nr;
 
-	if (vcpu->arch.exception.pending) {
+	if (vcpu->arch.exception.pending && vcpu->arch.exception.reinject) {
 		nr = vcpu->arch.exception.nr;
 		idt_vectoring = nr | VECTORING_INFO_VALID_MASK;
 
@@ -8023,7 +8081,7 @@ static void vmcs12_save_pending_event(struct kvm_vcpu *vcpu,
 		}
 
 		vmcs12->idt_vectoring_info_field = idt_vectoring;
-	} else if (vcpu->arch.nmi_pending) {
+	} else if (vcpu->arch.nmi_injected) {
 		vmcs12->idt_vectoring_info_field =
 			INTR_TYPE_NMI_INTR | INTR_INFO_VALID_MASK | NMI_VECTOR;
 	} else if (vcpu->arch.interrupt.pending) {
@@ -8105,6 +8163,11 @@ static void prepare_vmcs12(struct kvm_vcpu *vcpu, struct vmcs12 *vmcs12)
 	vmcs12->guest_pending_dbg_exceptions =
 		vmcs_readl(GUEST_PENDING_DBG_EXCEPTIONS);
 
+	if ((vmcs12->pin_based_vm_exec_control & PIN_BASED_VMX_PREEMPTION_TIMER) &&
+	    (vmcs12->vm_exit_controls & VM_EXIT_SAVE_VMX_PREEMPTION_TIMER))
+		vmcs12->vmx_preemption_timer_value =
+			vmcs_read32(VMX_PREEMPTION_TIMER_VALUE);
+
 	/*
 	 * In some cases (usually, nested EPT), L2 is allowed to change its
 	 * own CR3 without exiting. If it has changed it, we must keep it.
@@ -8130,6 +8193,8 @@ static void prepare_vmcs12(struct kvm_vcpu *vcpu, struct vmcs12 *vmcs12)
 	vmcs12->guest_ia32_debugctl = vmcs_read64(GUEST_IA32_DEBUGCTL);
 	if (vmcs12->vm_exit_controls & VM_EXIT_SAVE_IA32_PAT)
 		vmcs12->guest_ia32_pat = vmcs_read64(GUEST_IA32_PAT);
+	if (vmcs12->vm_exit_controls & VM_EXIT_SAVE_IA32_EFER)
+		vmcs12->guest_ia32_efer = vcpu->arch.efer;
 	vmcs12->guest_sysenter_cs = vmcs_read32(GUEST_SYSENTER_CS);
 	vmcs12->guest_sysenter_esp = vmcs_readl(GUEST_SYSENTER_ESP);
 	vmcs12->guest_sysenter_eip = vmcs_readl(GUEST_SYSENTER_EIP);
@@ -8201,7 +8266,7 @@ static void load_vmcs12_host_state(struct kvm_vcpu *vcpu,
 	 * fpu_active (which may have changed).
 	 * Note that vmx_set_cr0 refers to efer set above.
 	 */
-	kvm_set_cr0(vcpu, vmcs12->host_cr0);
+	vmx_set_cr0(vcpu, vmcs12->host_cr0);
 	/*
 	 * If we did fpu_activate()/fpu_deactivate() during L2's run, we need
 	 * to apply the same changes to L1's vmcs. We just set cr0 correctly,
@@ -8224,6 +8289,9 @@ static void load_vmcs12_host_state(struct kvm_vcpu *vcpu,
 	kvm_set_cr3(vcpu, vmcs12->host_cr3);
 	kvm_mmu_reset_context(vcpu);
 
+	if (!enable_ept)
+		vcpu->arch.walk_mmu->inject_page_fault = kvm_inject_page_fault;
+
 	if (enable_vpid) {
 		/*
 		 * Trivially support vpid by letting L2s share their parent
diff --git a/arch/x86/kvm/x86.c b/arch/x86/kvm/x86.c
index e5ca72a5cdb6..21ef1ba184ae 100644
--- a/arch/x86/kvm/x86.c
+++ b/arch/x86/kvm/x86.c
@@ -577,6 +577,7 @@ static void kvm_put_guest_xcr0(struct kvm_vcpu *vcpu)
 int __kvm_set_xcr(struct kvm_vcpu *vcpu, u32 index, u64 xcr)
 {
 	u64 xcr0;
+	u64 valid_bits;
 
 	/* Only support XCR_XFEATURE_ENABLED_MASK(xcr0) now  */
 	if (index != XCR_XFEATURE_ENABLED_MASK)
@@ -586,8 +587,16 @@ int __kvm_set_xcr(struct kvm_vcpu *vcpu, u32 index, u64 xcr)
 		return 1;
 	if ((xcr0 & XSTATE_YMM) && !(xcr0 & XSTATE_SSE))
 		return 1;
-	if (xcr0 & ~host_xcr0)
+
+	/*
+	 * Do not allow the guest to set bits that we do not support
+	 * saving.  However, xcr0 bit 0 is always set, even if the
+	 * emulated CPU does not support XSAVE (see fx_init).
+	 */
+	valid_bits = vcpu->arch.guest_supported_xcr0 | XSTATE_FP;
+	if (xcr0 & ~valid_bits)
 		return 1;
+
 	kvm_put_guest_xcr0(vcpu);
 	vcpu->arch.xcr0 = xcr0;
 	return 0;
@@ -684,7 +693,7 @@ int kvm_set_cr3(struct kvm_vcpu *vcpu, unsigned long cr3)
 
 	vcpu->arch.cr3 = cr3;
 	__set_bit(VCPU_EXREG_CR3, (ulong *)&vcpu->arch.regs_avail);
-	vcpu->arch.mmu.new_cr3(vcpu);
+	kvm_mmu_new_cr3(vcpu);
 	return 0;
 }
 EXPORT_SYMBOL_GPL(kvm_set_cr3);
@@ -2564,6 +2573,7 @@ int kvm_dev_ioctl_check_extension(long ext)
 	case KVM_CAP_MMU_SHADOW_CACHE_CONTROL:
 	case KVM_CAP_SET_TSS_ADDR:
 	case KVM_CAP_EXT_CPUID:
+	case KVM_CAP_EXT_EMUL_CPUID:
 	case KVM_CAP_CLOCKSOURCE:
 	case KVM_CAP_PIT:
 	case KVM_CAP_NOP_IO_DELAY:
@@ -2673,15 +2683,17 @@ long kvm_arch_dev_ioctl(struct file *filp,
 		r = 0;
 		break;
 	}
-	case KVM_GET_SUPPORTED_CPUID: {
+	case KVM_GET_SUPPORTED_CPUID:
+	case KVM_GET_EMULATED_CPUID: {
 		struct kvm_cpuid2 __user *cpuid_arg = argp;
 		struct kvm_cpuid2 cpuid;
 
 		r = -EFAULT;
 		if (copy_from_user(&cpuid, cpuid_arg, sizeof cpuid))
 			goto out;
-		r = kvm_dev_ioctl_get_supported_cpuid(&cpuid,
-						      cpuid_arg->entries);
+
+		r = kvm_dev_ioctl_get_cpuid(&cpuid, cpuid_arg->entries,
+					    ioctl);
 		if (r)
 			goto out;
 
@@ -2715,8 +2727,7 @@ static void wbinvd_ipi(void *garbage)
 
 static bool need_emulate_wbinvd(struct kvm_vcpu *vcpu)
 {
-	return vcpu->kvm->arch.iommu_domain &&
-		!(vcpu->kvm->arch.iommu_flags & KVM_IOMMU_CACHE_COHERENCY);
+	return kvm_arch_has_noncoherent_dma(vcpu->kvm);
 }
 
 void kvm_arch_vcpu_load(struct kvm_vcpu *vcpu, int cpu)
@@ -2984,11 +2995,13 @@ static int kvm_vcpu_ioctl_x86_set_debugregs(struct kvm_vcpu *vcpu,
 static void kvm_vcpu_ioctl_x86_get_xsave(struct kvm_vcpu *vcpu,
 					 struct kvm_xsave *guest_xsave)
 {
-	if (cpu_has_xsave)
+	if (cpu_has_xsave) {
 		memcpy(guest_xsave->region,
 			&vcpu->arch.guest_fpu.state->xsave,
-			xstate_size);
-	else {
+			vcpu->arch.guest_xstate_size);
+		*(u64 *)&guest_xsave->region[XSAVE_HDR_OFFSET / sizeof(u32)] &=
+			vcpu->arch.guest_supported_xcr0 | XSTATE_FPSSE;
+	} else {
 		memcpy(guest_xsave->region,
 			&vcpu->arch.guest_fpu.state->fxsave,
 			sizeof(struct i387_fxsave_struct));
@@ -3003,10 +3016,19 @@ static int kvm_vcpu_ioctl_x86_set_xsave(struct kvm_vcpu *vcpu,
 	u64 xstate_bv =
 		*(u64 *)&guest_xsave->region[XSAVE_HDR_OFFSET / sizeof(u32)];
 
-	if (cpu_has_xsave)
+	if (cpu_has_xsave) {
+		/*
+		 * Here we allow setting states that are not present in
+		 * CPUID leaf 0xD, index 0, EDX:EAX.  This is for compatibility
+		 * with old userspace.
+		 */
+		if (xstate_bv & ~KVM_SUPPORTED_XCR0)
+			return -EINVAL;
+		if (xstate_bv & ~host_xcr0)
+			return -EINVAL;
 		memcpy(&vcpu->arch.guest_fpu.state->xsave,
-			guest_xsave->region, xstate_size);
-	else {
+			guest_xsave->region, vcpu->arch.guest_xstate_size);
+	} else {
 		if (xstate_bv & ~XSTATE_FPSSE)
 			return -EINVAL;
 		memcpy(&vcpu->arch.guest_fpu.state->fxsave,
@@ -3042,9 +3064,9 @@ static int kvm_vcpu_ioctl_x86_set_xcrs(struct kvm_vcpu *vcpu,
 
 	for (i = 0; i < guest_xcrs->nr_xcrs; i++)
 		/* Only support XCR0 currently */
-		if (guest_xcrs->xcrs[0].xcr == XCR_XFEATURE_ENABLED_MASK) {
+		if (guest_xcrs->xcrs[i].xcr == XCR_XFEATURE_ENABLED_MASK) {
 			r = __kvm_set_xcr(vcpu, XCR_XFEATURE_ENABLED_MASK,
-				guest_xcrs->xcrs[0].value);
+				guest_xcrs->xcrs[i].value);
 			break;
 		}
 	if (r)
@@ -4775,8 +4797,8 @@ static void inject_emulated_exception(struct kvm_vcpu *vcpu)
 
 static void init_decode_cache(struct x86_emulate_ctxt *ctxt)
 {
-	memset(&ctxt->twobyte, 0,
-	       (void *)&ctxt->_regs - (void *)&ctxt->twobyte);
+	memset(&ctxt->opcode_len, 0,
+	       (void *)&ctxt->_regs - (void *)&ctxt->opcode_len);
 
 	ctxt->fetch.start = 0;
 	ctxt->fetch.end = 0;
@@ -5094,8 +5116,7 @@ int x86_emulate_instruction(struct kvm_vcpu *vcpu,
 		ctxt->have_exception = false;
 		ctxt->perm_ok = false;
 
-		ctxt->only_vendor_specific_insn
-			= emulation_type & EMULTYPE_TRAP_UD;
+		ctxt->ud = emulation_type & EMULTYPE_TRAP_UD;
 
 		r = x86_decode_insn(ctxt, insn, insn_len);
 
@@ -5263,7 +5284,7 @@ static int kvmclock_cpufreq_notifier(struct notifier_block *nb, unsigned long va
 
 	smp_call_function_single(freq->cpu, tsc_khz_changed, freq, 1);
 
-	raw_spin_lock(&kvm_lock);
+	spin_lock(&kvm_lock);
 	list_for_each_entry(kvm, &vm_list, vm_list) {
 		kvm_for_each_vcpu(i, vcpu, kvm) {
 			if (vcpu->cpu != freq->cpu)
@@ -5273,7 +5294,7 @@ static int kvmclock_cpufreq_notifier(struct notifier_block *nb, unsigned long va
 				send_ipi = 1;
 		}
 	}
-	raw_spin_unlock(&kvm_lock);
+	spin_unlock(&kvm_lock);
 
 	if (freq->old < freq->new && send_ipi) {
 		/*
@@ -5426,12 +5447,12 @@ static void pvclock_gtod_update_fn(struct work_struct *work)
 	struct kvm_vcpu *vcpu;
 	int i;
 
-	raw_spin_lock(&kvm_lock);
+	spin_lock(&kvm_lock);
 	list_for_each_entry(kvm, &vm_list, vm_list)
 		kvm_for_each_vcpu(i, vcpu, kvm)
 			set_bit(KVM_REQ_MASTERCLOCK_UPDATE, &vcpu->requests);
 	atomic_set(&kvm_guest_has_master_clock, 0);
-	raw_spin_unlock(&kvm_lock);
+	spin_unlock(&kvm_lock);
 }
 
 static DECLARE_WORK(pvclock_gtod_work, pvclock_gtod_update_fn);
@@ -5945,10 +5966,12 @@ static int vcpu_enter_guest(struct kvm_vcpu *vcpu)
 
 	vcpu->mode = IN_GUEST_MODE;
 
+	srcu_read_unlock(&vcpu->kvm->srcu, vcpu->srcu_idx);
+
 	/* We should set ->mode before check ->requests,
 	 * see the comment in make_all_cpus_request.
 	 */
-	smp_mb();
+	smp_mb__after_srcu_read_unlock();
 
 	local_irq_disable();
 
@@ -5958,12 +5981,11 @@ static int vcpu_enter_guest(struct kvm_vcpu *vcpu)
 		smp_wmb();
 		local_irq_enable();
 		preempt_enable();
+		vcpu->srcu_idx = srcu_read_lock(&vcpu->kvm->srcu);
 		r = 1;
 		goto cancel_injection;
 	}
 
-	srcu_read_unlock(&vcpu->kvm->srcu, vcpu->srcu_idx);
-
 	if (req_immediate_exit)
 		smp_send_reschedule(vcpu->cpu);
 
@@ -6688,7 +6710,7 @@ int kvm_arch_vcpu_setup(struct kvm_vcpu *vcpu)
 	if (r)
 		return r;
 	kvm_vcpu_reset(vcpu);
-	r = kvm_mmu_setup(vcpu);
+	kvm_mmu_setup(vcpu);
 	vcpu_put(vcpu);
 
 	return r;
@@ -6940,6 +6962,10 @@ int kvm_arch_vcpu_init(struct kvm_vcpu *vcpu)
 
 	vcpu->arch.ia32_tsc_adjust_msr = 0x0;
 	vcpu->arch.pv_time_enabled = false;
+
+	vcpu->arch.guest_supported_xcr0 = 0;
+	vcpu->arch.guest_xstate_size = XSAVE_HDR_SIZE + XSAVE_HDR_OFFSET;
+
 	kvm_async_pf_hash_reset(vcpu);
 	kvm_pmu_init(vcpu);
 
@@ -6981,6 +7007,7 @@ int kvm_arch_init_vm(struct kvm *kvm, unsigned long type)
 	INIT_LIST_HEAD(&kvm->arch.active_mmu_pages);
 	INIT_LIST_HEAD(&kvm->arch.zapped_obsolete_pages);
 	INIT_LIST_HEAD(&kvm->arch.assigned_dev_head);
+	atomic_set(&kvm->arch.noncoherent_dma_count, 0);
 
 	/* Reserve bit 0 of irq_sources_bitmap for userspace irq source */
 	set_bit(KVM_USERSPACE_IRQ_SOURCE_ID, &kvm->arch.irq_sources_bitmap);
@@ -7065,7 +7092,7 @@ void kvm_arch_destroy_vm(struct kvm *kvm)
 	kfree(rcu_dereference_check(kvm->arch.apic_map, 1));
 }
 
-void kvm_arch_free_memslot(struct kvm_memory_slot *free,
+void kvm_arch_free_memslot(struct kvm *kvm, struct kvm_memory_slot *free,
 			   struct kvm_memory_slot *dont)
 {
 	int i;
@@ -7086,7 +7113,8 @@ void kvm_arch_free_memslot(struct kvm_memory_slot *free,
 	}
 }
 
-int kvm_arch_create_memslot(struct kvm_memory_slot *slot, unsigned long npages)
+int kvm_arch_create_memslot(struct kvm *kvm, struct kvm_memory_slot *slot,
+			    unsigned long npages)
 {
 	int i;
 
@@ -7283,7 +7311,7 @@ void kvm_arch_async_page_ready(struct kvm_vcpu *vcpu, struct kvm_async_pf *work)
 	int r;
 
 	if ((vcpu->arch.mmu.direct_map != work->arch.direct_map) ||
-	      is_error_page(work->page))
+	      work->wakeup_all)
 		return;
 
 	r = kvm_mmu_reload(vcpu);
@@ -7393,7 +7421,7 @@ void kvm_arch_async_page_present(struct kvm_vcpu *vcpu,
 	struct x86_exception fault;
 
 	trace_kvm_async_pf_ready(work->arch.token, work->gva);
-	if (is_error_page(work->page))
+	if (work->wakeup_all)
 		work->arch.token = ~0; /* broadcast wakeup */
 	else
 		kvm_del_async_pf_gfn(vcpu, work->arch.gfn);
@@ -7420,6 +7448,24 @@ bool kvm_arch_can_inject_async_page_present(struct kvm_vcpu *vcpu)
 			kvm_x86_ops->interrupt_allowed(vcpu);
 }
 
+void kvm_arch_register_noncoherent_dma(struct kvm *kvm)
+{
+	atomic_inc(&kvm->arch.noncoherent_dma_count);
+}
+EXPORT_SYMBOL_GPL(kvm_arch_register_noncoherent_dma);
+
+void kvm_arch_unregister_noncoherent_dma(struct kvm *kvm)
+{
+	atomic_dec(&kvm->arch.noncoherent_dma_count);
+}
+EXPORT_SYMBOL_GPL(kvm_arch_unregister_noncoherent_dma);
+
+bool kvm_arch_has_noncoherent_dma(struct kvm *kvm)
+{
+	return atomic_read(&kvm->arch.noncoherent_dma_count);
+}
+EXPORT_SYMBOL_GPL(kvm_arch_has_noncoherent_dma);
+
 EXPORT_TRACEPOINT_SYMBOL_GPL(kvm_exit);
 EXPORT_TRACEPOINT_SYMBOL_GPL(kvm_inj_virq);
 EXPORT_TRACEPOINT_SYMBOL_GPL(kvm_page_fault);
diff --git a/arch/x86/kvm/x86.h b/arch/x86/kvm/x86.h
index e224f7a671b6..587fb9ede436 100644
--- a/arch/x86/kvm/x86.h
+++ b/arch/x86/kvm/x86.h
@@ -122,6 +122,7 @@ int kvm_write_guest_virt_system(struct x86_emulate_ctxt *ctxt,
 	gva_t addr, void *val, unsigned int bytes,
 	struct x86_exception *exception);
 
+#define KVM_SUPPORTED_XCR0	(XSTATE_FP | XSTATE_SSE | XSTATE_YMM)
 extern u64 host_xcr0;
 
 extern struct static_key kvm_no_apic_vcpu;
diff --git a/arch/x86/mm/Makefile b/arch/x86/mm/Makefile
index 23d8e5fecf76..6a19ad9f370d 100644
--- a/arch/x86/mm/Makefile
+++ b/arch/x86/mm/Makefile
@@ -6,6 +6,8 @@ nostackp := $(call cc-option, -fno-stack-protector)
 CFLAGS_physaddr.o		:= $(nostackp)
 CFLAGS_setup_nx.o		:= $(nostackp)
 
+CFLAGS_fault.o := -I$(src)/../include/asm/trace
+
 obj-$(CONFIG_X86_PAT)		+= pat_rbtree.o
 obj-$(CONFIG_SMP)		+= tlb.o
 
diff --git a/arch/x86/mm/fault.c b/arch/x86/mm/fault.c
index 7a517bb41060..9ff85bb8dd69 100644
--- a/arch/x86/mm/fault.c
+++ b/arch/x86/mm/fault.c
@@ -20,6 +20,9 @@
 #include <asm/kmemcheck.h>		/* kmemcheck_*(), ...		*/
 #include <asm/fixmap.h>			/* VSYSCALL_START		*/
 
+#define CREATE_TRACE_POINTS
+#include <asm/trace/exceptions.h>
+
 /*
  * Page fault error code bits:
  *
@@ -596,7 +599,7 @@ show_fault_oops(struct pt_regs *regs, unsigned long error_code,
 
 	printk(KERN_CONT " at %p\n", (void *) address);
 	printk(KERN_ALERT "IP:");
-	printk_address(regs->ip, 1);
+	printk_address(regs->ip);
 
 	dump_pagetable(address);
 }
@@ -1232,3 +1235,23 @@ do_page_fault(struct pt_regs *regs, unsigned long error_code)
 	__do_page_fault(regs, error_code);
 	exception_exit(prev_state);
 }
+
+static void trace_page_fault_entries(struct pt_regs *regs,
+				     unsigned long error_code)
+{
+	if (user_mode(regs))
+		trace_page_fault_user(read_cr2(), regs, error_code);
+	else
+		trace_page_fault_kernel(read_cr2(), regs, error_code);
+}
+
+dotraplinkage void __kprobes
+trace_do_page_fault(struct pt_regs *regs, unsigned long error_code)
+{
+	enum ctx_state prev_state;
+
+	prev_state = exception_enter();
+	trace_page_fault_entries(regs, error_code);
+	__do_page_fault(regs, error_code);
+	exception_exit(prev_state);
+}
diff --git a/arch/x86/mm/pgtable.c b/arch/x86/mm/pgtable.c
index dfa537a03be1..a7cccb6d7fec 100644
--- a/arch/x86/mm/pgtable.c
+++ b/arch/x86/mm/pgtable.c
@@ -25,8 +25,12 @@ pgtable_t pte_alloc_one(struct mm_struct *mm, unsigned long address)
 	struct page *pte;
 
 	pte = alloc_pages(__userpte_alloc_gfp, 0);
-	if (pte)
-		pgtable_page_ctor(pte);
+	if (!pte)
+		return NULL;
+	if (!pgtable_page_ctor(pte)) {
+		__free_page(pte);
+		return NULL;
+	}
 	return pte;
 }
 
@@ -189,8 +193,10 @@ static void free_pmds(pmd_t *pmds[])
 	int i;
 
 	for(i = 0; i < PREALLOCATED_PMDS; i++)
-		if (pmds[i])
+		if (pmds[i]) {
+			pgtable_pmd_page_dtor(virt_to_page(pmds[i]));
 			free_page((unsigned long)pmds[i]);
+		}
 }
 
 static int preallocate_pmds(pmd_t *pmds[])
@@ -200,8 +206,13 @@ static int preallocate_pmds(pmd_t *pmds[])
 
 	for(i = 0; i < PREALLOCATED_PMDS; i++) {
 		pmd_t *pmd = (pmd_t *)__get_free_page(PGALLOC_GFP);
-		if (pmd == NULL)
+		if (!pmd)
 			failed = true;
+		if (pmd && !pgtable_pmd_page_ctor(virt_to_page(pmd))) {
+			free_page((unsigned long)pmds[i]);
+			pmd = NULL;
+			failed = true;
+		}
 		pmds[i] = pmd;
 	}
 
diff --git a/arch/x86/platform/uv/uv_nmi.c b/arch/x86/platform/uv/uv_nmi.c
index 2e863ad4a772..8eeccba73130 100644
--- a/arch/x86/platform/uv/uv_nmi.c
+++ b/arch/x86/platform/uv/uv_nmi.c
@@ -399,7 +399,7 @@ static void uv_nmi_dump_cpu_ip(int cpu, struct pt_regs *regs)
 	printk(KERN_DEFAULT "UV: %4d %6d %-32.32s ",
 		cpu, current->pid, current->comm);
 
-	printk_address(regs->ip, 1);
+	printk_address(regs->ip);
 }
 
 /* Dump this cpu's state */
diff --git a/arch/x86/um/Kconfig b/arch/x86/um/Kconfig
index 14ef8d1dbc33..ed56a1c4ae73 100644
--- a/arch/x86/um/Kconfig
+++ b/arch/x86/um/Kconfig
@@ -31,6 +31,11 @@ config X86_64
 	def_bool 64BIT
 	select MODULES_USE_ELF_RELA
 
+config ARCH_DEFCONFIG
+	string
+	default "arch/um/configs/i386_defconfig" if X86_32
+	default "arch/um/configs/x86_64_defconfig" if X86_64
+
 config RWSEM_XCHGADD_ALGORITHM
 	def_bool 64BIT
 
diff --git a/arch/x86/um/asm/processor_32.h b/arch/x86/um/asm/processor_32.h
index 6c6689e574ce..c112de81c9e1 100644
--- a/arch/x86/um/asm/processor_32.h
+++ b/arch/x86/um/asm/processor_32.h
@@ -33,6 +33,8 @@ struct arch_thread {
 	.faultinfo		= { 0, 0, 0 } \
 }
 
+#define STACKSLOTS_PER_LINE 8
+
 static inline void arch_flush_thread(struct arch_thread *thread)
 {
 	/* Clear any TLS still hanging */
@@ -53,4 +55,7 @@ static inline void arch_copy_thread(struct arch_thread *from,
 #define current_text_addr() \
 	({ void *pc; __asm__("movl $1f,%0\n1:":"=g" (pc)); pc; })
 
+#define current_sp() ({ void *sp; __asm__("movl %%esp, %0" : "=r" (sp) : ); sp; })
+#define current_bp() ({ unsigned long bp; __asm__("movl %%ebp, %0" : "=r" (bp) : ); bp; })
+
 #endif
diff --git a/arch/x86/um/asm/processor_64.h b/arch/x86/um/asm/processor_64.h
index 4b02a8455bd1..c3be85205a65 100644
--- a/arch/x86/um/asm/processor_64.h
+++ b/arch/x86/um/asm/processor_64.h
@@ -19,6 +19,8 @@ struct arch_thread {
 			   .fs			= 0, \
 			   .faultinfo		= { 0, 0, 0 } }
 
+#define STACKSLOTS_PER_LINE 4
+
 static inline void arch_flush_thread(struct arch_thread *thread)
 {
 }
@@ -32,4 +34,7 @@ static inline void arch_copy_thread(struct arch_thread *from,
 #define current_text_addr() \
 	({ void *pc; __asm__("movq $1f,%0\n1:":"=g" (pc)); pc; })
 
+#define current_sp() ({ void *sp; __asm__("movq %%rsp, %0" : "=r" (sp) : ); sp; })
+#define current_bp() ({ unsigned long bp; __asm__("movq %%rbp, %0" : "=r" (bp) : ); bp; })
+
 #endif
diff --git a/arch/x86/um/sysrq_32.c b/arch/x86/um/sysrq_32.c
index c9bee5b8c0d3..16ee0e450e3e 100644
--- a/arch/x86/um/sysrq_32.c
+++ b/arch/x86/um/sysrq_32.c
@@ -30,70 +30,4 @@ void show_regs(struct pt_regs *regs)
         printk(" DS: %04lx ES: %04lx\n",
 	       0xffff & PT_REGS_DS(regs), 
 	       0xffff & PT_REGS_ES(regs));
-
-        show_trace(NULL, (unsigned long *) &regs);
 }
-
-/* Copied from i386. */
-static inline int valid_stack_ptr(struct thread_info *tinfo, void *p)
-{
-	return	p > (void *)tinfo &&
-		p < (void *)tinfo + THREAD_SIZE - 3;
-}
-
-/* Adapted from i386 (we also print the address we read from). */
-static inline unsigned long print_context_stack(struct thread_info *tinfo,
-				unsigned long *stack, unsigned long ebp)
-{
-	unsigned long addr;
-
-#ifdef CONFIG_FRAME_POINTER
-	while (valid_stack_ptr(tinfo, (void *)ebp)) {
-		addr = *(unsigned long *)(ebp + 4);
-		printk("%08lx:  [<%08lx>]", ebp + 4, addr);
-		print_symbol(" %s", addr);
-		printk("\n");
-		ebp = *(unsigned long *)ebp;
-	}
-#else
-	while (valid_stack_ptr(tinfo, stack)) {
-		addr = *stack;
-		if (__kernel_text_address(addr)) {
-			printk("%08lx:  [<%08lx>]", (unsigned long) stack, addr);
-			print_symbol(" %s", addr);
-			printk("\n");
-		}
-		stack++;
-	}
-#endif
-	return ebp;
-}
-
-void show_trace(struct task_struct* task, unsigned long * stack)
-{
-	unsigned long ebp;
-	struct thread_info *context;
-
-	/* Turn this into BUG_ON if possible. */
-	if (!stack) {
-		stack = (unsigned long*) &stack;
-		printk("show_trace: got NULL stack, implicit assumption task == current");
-		WARN_ON(1);
-	}
-
-	if (!task)
-		task = current;
-
-	if (task != current) {
-		ebp = (unsigned long) KSTK_EBP(task);
-	} else {
-		asm ("movl %%ebp, %0" : "=r" (ebp) : );
-	}
-
-	context = (struct thread_info *)
-		((unsigned long)stack & (~(THREAD_SIZE - 1)));
-	print_context_stack(context, stack, ebp);
-
-	printk("\n");
-}
-
diff --git a/arch/x86/um/sysrq_64.c b/arch/x86/um/sysrq_64.c
index a0e7fb1134a0..38b4e4abd0f8 100644
--- a/arch/x86/um/sysrq_64.c
+++ b/arch/x86/um/sysrq_64.c
@@ -12,7 +12,7 @@
 #include <asm/ptrace.h>
 #include <asm/sysrq.h>
 
-void __show_regs(struct pt_regs *regs)
+void show_regs(struct pt_regs *regs)
 {
 	printk("\n");
 	print_modules();
@@ -33,9 +33,3 @@ void __show_regs(struct pt_regs *regs)
 	printk(KERN_INFO "R13: %016lx R14: %016lx R15: %016lx\n",
 	       PT_REGS_R13(regs), PT_REGS_R14(regs), PT_REGS_R15(regs));
 }
-
-void show_regs(struct pt_regs *regs)
-{
-	__show_regs(regs);
-	show_trace(current, (unsigned long *) &regs);
-}
diff --git a/arch/x86/um/vdso/.gitignore b/arch/x86/um/vdso/.gitignore
new file mode 100644
index 000000000000..9cac6d072199
--- /dev/null
+++ b/arch/x86/um/vdso/.gitignore
@@ -0,0 +1,2 @@
+vdso-syms.lds
+vdso.lds
diff --git a/arch/x86/vdso/vclock_gettime.c b/arch/x86/vdso/vclock_gettime.c
index 72074d528400..2ada505067cc 100644
--- a/arch/x86/vdso/vclock_gettime.c
+++ b/arch/x86/vdso/vclock_gettime.c
@@ -178,7 +178,7 @@ notrace static int __always_inline do_realtime(struct timespec *ts)
 
 	ts->tv_nsec = 0;
 	do {
-		seq = read_seqcount_begin(&gtod->seq);
+		seq = read_seqcount_begin_no_lockdep(&gtod->seq);
 		mode = gtod->clock.vclock_mode;
 		ts->tv_sec = gtod->wall_time_sec;
 		ns = gtod->wall_time_snsec;
@@ -198,7 +198,7 @@ notrace static int do_monotonic(struct timespec *ts)
 
 	ts->tv_nsec = 0;
 	do {
-		seq = read_seqcount_begin(&gtod->seq);
+		seq = read_seqcount_begin_no_lockdep(&gtod->seq);
 		mode = gtod->clock.vclock_mode;
 		ts->tv_sec = gtod->monotonic_time_sec;
 		ns = gtod->monotonic_time_snsec;
@@ -214,7 +214,7 @@ notrace static int do_realtime_coarse(struct timespec *ts)
 {
 	unsigned long seq;
 	do {
-		seq = read_seqcount_begin(&gtod->seq);
+		seq = read_seqcount_begin_no_lockdep(&gtod->seq);
 		ts->tv_sec = gtod->wall_time_coarse.tv_sec;
 		ts->tv_nsec = gtod->wall_time_coarse.tv_nsec;
 	} while (unlikely(read_seqcount_retry(&gtod->seq, seq)));
@@ -225,7 +225,7 @@ notrace static int do_monotonic_coarse(struct timespec *ts)
 {
 	unsigned long seq;
 	do {
-		seq = read_seqcount_begin(&gtod->seq);
+		seq = read_seqcount_begin_no_lockdep(&gtod->seq);
 		ts->tv_sec = gtod->monotonic_time_coarse.tv_sec;
 		ts->tv_nsec = gtod->monotonic_time_coarse.tv_nsec;
 	} while (unlikely(read_seqcount_retry(&gtod->seq, seq)));
diff --git a/arch/x86/xen/mmu.c b/arch/x86/xen/mmu.c
index fdc3ba28ca38..ce563be09cc1 100644
--- a/arch/x86/xen/mmu.c
+++ b/arch/x86/xen/mmu.c
@@ -468,8 +468,8 @@ PV_CALLEE_SAVE_REGS_THUNK(xen_pgd_val);
  * 3        PCD PWT      UC       UC     UC
  * 4    PAT              WB       WC     WB
  * 5    PAT     PWT      WC       WP     WT
- * 6    PAT PCD          UC-      UC     UC-
- * 7    PAT PCD PWT      UC       UC     UC
+ * 6    PAT PCD          UC-      rsv    UC-
+ * 7    PAT PCD PWT      UC       rsv    UC
  */
 
 void xen_set_pat(u64 pat)
@@ -796,8 +796,8 @@ static spinlock_t *xen_pte_lock(struct page *page, struct mm_struct *mm)
 {
 	spinlock_t *ptl = NULL;
 
-#if USE_SPLIT_PTLOCKS
-	ptl = __pte_lockptr(page);
+#if USE_SPLIT_PTE_PTLOCKS
+	ptl = ptlock_ptr(page);
 	spin_lock_nest_lock(ptl, &mm->page_table_lock);
 #endif
 
@@ -1637,7 +1637,7 @@ static inline void xen_alloc_ptpage(struct mm_struct *mm, unsigned long pfn,
 
 			__set_pfn_prot(pfn, PAGE_KERNEL_RO);
 
-			if (level == PT_PTE && USE_SPLIT_PTLOCKS)
+			if (level == PT_PTE && USE_SPLIT_PTE_PTLOCKS)
 				__pin_pagetable_pfn(MMUEXT_PIN_L1_TABLE, pfn);
 
 			xen_mc_issue(PARAVIRT_LAZY_MMU);
@@ -1671,7 +1671,7 @@ static inline void xen_release_ptpage(unsigned long pfn, unsigned level)
 		if (!PageHighMem(page)) {
 			xen_mc_batch();
 
-			if (level == PT_PTE && USE_SPLIT_PTLOCKS)
+			if (level == PT_PTE && USE_SPLIT_PTE_PTLOCKS)
 				__pin_pagetable_pfn(MMUEXT_UNPIN_TABLE, pfn);
 
 			__set_pfn_prot(pfn, PAGE_KERNEL);
@@ -2328,12 +2328,14 @@ static int xen_exchange_memory(unsigned long extents_in, unsigned int order_in,
 	return success;
 }
 
-int xen_create_contiguous_region(unsigned long vstart, unsigned int order,
-				 unsigned int address_bits)
+int xen_create_contiguous_region(phys_addr_t pstart, unsigned int order,
+				 unsigned int address_bits,
+				 dma_addr_t *dma_handle)
 {
 	unsigned long *in_frames = discontig_frames, out_frame;
 	unsigned long  flags;
 	int            success;
+	unsigned long vstart = (unsigned long)phys_to_virt(pstart);
 
 	/*
 	 * Currently an auto-translated guest will not perform I/O, nor will
@@ -2368,15 +2370,17 @@ int xen_create_contiguous_region(unsigned long vstart, unsigned int order,
 
 	spin_unlock_irqrestore(&xen_reservation_lock, flags);
 
+	*dma_handle = virt_to_machine(vstart).maddr;
 	return success ? 0 : -ENOMEM;
 }
 EXPORT_SYMBOL_GPL(xen_create_contiguous_region);
 
-void xen_destroy_contiguous_region(unsigned long vstart, unsigned int order)
+void xen_destroy_contiguous_region(phys_addr_t pstart, unsigned int order)
 {
 	unsigned long *out_frames = discontig_frames, in_frame;
 	unsigned long  flags;
 	int success;
+	unsigned long vstart;
 
 	if (xen_feature(XENFEAT_auto_translated_physmap))
 		return;
@@ -2384,6 +2388,7 @@ void xen_destroy_contiguous_region(unsigned long vstart, unsigned int order)
 	if (unlikely(order > MAX_CONTIG_ORDER))
 		return;
 
+	vstart = (unsigned long)phys_to_virt(pstart);
 	memset((void *) vstart, 0, PAGE_SIZE << order);
 
 	spin_lock_irqsave(&xen_reservation_lock, flags);
diff --git a/arch/x86/xen/p2m.c b/arch/x86/xen/p2m.c
index a61c7d5811be..2ae8699e8767 100644
--- a/arch/x86/xen/p2m.c
+++ b/arch/x86/xen/p2m.c
@@ -799,10 +799,10 @@ bool __set_phys_to_machine(unsigned long pfn, unsigned long mfn)
 {
 	unsigned topidx, mididx, idx;
 
-	if (unlikely(xen_feature(XENFEAT_auto_translated_physmap))) {
-		BUG_ON(pfn != mfn && mfn != INVALID_P2M_ENTRY);
+	/* don't track P2M changes in autotranslate guests */
+	if (unlikely(xen_feature(XENFEAT_auto_translated_physmap)))
 		return true;
-	}
+
 	if (unlikely(pfn >= MAX_P2M_PFN)) {
 		BUG_ON(mfn != INVALID_P2M_ENTRY);
 		return true;
diff --git a/arch/x86/xen/pci-swiotlb-xen.c b/arch/x86/xen/pci-swiotlb-xen.c
index 969570491c39..0e98e5d241d0 100644
--- a/arch/x86/xen/pci-swiotlb-xen.c
+++ b/arch/x86/xen/pci-swiotlb-xen.c
@@ -75,8 +75,10 @@ void __init pci_xen_swiotlb_init(void)
 		xen_swiotlb_init(1, true /* early */);
 		dma_ops = &xen_swiotlb_dma_ops;
 
+#ifdef CONFIG_PCI
 		/* Make sure ACS will be enabled */
 		pci_request_acs();
+#endif
 	}
 }
 
@@ -92,8 +94,10 @@ int pci_xen_swiotlb_init_late(void)
 		return rc;
 
 	dma_ops = &xen_swiotlb_dma_ops;
+#ifdef CONFIG_PCI
 	/* Make sure ACS will be enabled */
 	pci_request_acs();
+#endif
 
 	return 0;
 }
diff --git a/arch/x86/xen/setup.c b/arch/x86/xen/setup.c
index 09f3059cb00b..68c054f59de6 100644
--- a/arch/x86/xen/setup.c
+++ b/arch/x86/xen/setup.c
@@ -556,7 +556,7 @@ void xen_enable_syscall(void)
 	}
 #endif /* CONFIG_X86_64 */
 }
-void __cpuinit xen_enable_nmi(void)
+void xen_enable_nmi(void)
 {
 #ifdef CONFIG_X86_64
 	if (register_callback(CALLBACKTYPE_nmi, nmi))
diff --git a/arch/x86/xen/smp.c b/arch/x86/xen/smp.c
index 31d04758b76f..c36b325abd83 100644
--- a/arch/x86/xen/smp.c
+++ b/arch/x86/xen/smp.c
@@ -149,7 +149,7 @@ static int xen_smp_intr_init(unsigned int cpu)
 	rc = bind_ipi_to_irqhandler(XEN_RESCHEDULE_VECTOR,
 				    cpu,
 				    xen_reschedule_interrupt,
-				    IRQF_DISABLED|IRQF_PERCPU|IRQF_NOBALANCING,
+				    IRQF_PERCPU|IRQF_NOBALANCING,
 				    resched_name,
 				    NULL);
 	if (rc < 0)
@@ -161,7 +161,7 @@ static int xen_smp_intr_init(unsigned int cpu)
 	rc = bind_ipi_to_irqhandler(XEN_CALL_FUNCTION_VECTOR,
 				    cpu,
 				    xen_call_function_interrupt,
-				    IRQF_DISABLED|IRQF_PERCPU|IRQF_NOBALANCING,
+				    IRQF_PERCPU|IRQF_NOBALANCING,
 				    callfunc_name,
 				    NULL);
 	if (rc < 0)
@@ -171,7 +171,7 @@ static int xen_smp_intr_init(unsigned int cpu)
 
 	debug_name = kasprintf(GFP_KERNEL, "debug%d", cpu);
 	rc = bind_virq_to_irqhandler(VIRQ_DEBUG, cpu, xen_debug_interrupt,
-				     IRQF_DISABLED | IRQF_PERCPU | IRQF_NOBALANCING,
+				     IRQF_PERCPU | IRQF_NOBALANCING,
 				     debug_name, NULL);
 	if (rc < 0)
 		goto fail;
@@ -182,7 +182,7 @@ static int xen_smp_intr_init(unsigned int cpu)
 	rc = bind_ipi_to_irqhandler(XEN_CALL_FUNCTION_SINGLE_VECTOR,
 				    cpu,
 				    xen_call_function_single_interrupt,
-				    IRQF_DISABLED|IRQF_PERCPU|IRQF_NOBALANCING,
+				    IRQF_PERCPU|IRQF_NOBALANCING,
 				    callfunc_name,
 				    NULL);
 	if (rc < 0)
@@ -201,7 +201,7 @@ static int xen_smp_intr_init(unsigned int cpu)
 	rc = bind_ipi_to_irqhandler(XEN_IRQ_WORK_VECTOR,
 				    cpu,
 				    xen_irq_work_interrupt,
-				    IRQF_DISABLED|IRQF_PERCPU|IRQF_NOBALANCING,
+				    IRQF_PERCPU|IRQF_NOBALANCING,
 				    callfunc_name,
 				    NULL);
 	if (rc < 0)
diff --git a/arch/x86/xen/spinlock.c b/arch/x86/xen/spinlock.c
index be6b86078957..0e36cde12f7e 100644
--- a/arch/x86/xen/spinlock.c
+++ b/arch/x86/xen/spinlock.c
@@ -234,7 +234,7 @@ void xen_init_lock_cpu(int cpu)
 	irq = bind_ipi_to_irqhandler(XEN_SPIN_UNLOCK_VECTOR,
 				     cpu,
 				     dummy_handler,
-				     IRQF_DISABLED|IRQF_PERCPU|IRQF_NOBALANCING,
+				     IRQF_PERCPU|IRQF_NOBALANCING,
 				     name,
 				     NULL);
 
diff --git a/arch/x86/xen/time.c b/arch/x86/xen/time.c
index ee365895b06b..12a1ca707b94 100644
--- a/arch/x86/xen/time.c
+++ b/arch/x86/xen/time.c
@@ -443,8 +443,7 @@ void xen_setup_timer(int cpu)
 		name = "<timer kasprintf failed>";
 
 	irq = bind_virq_to_irqhandler(VIRQ_TIMER, cpu, xen_timer_interrupt,
-				      IRQF_DISABLED|IRQF_PERCPU|
-				      IRQF_NOBALANCING|IRQF_TIMER|
+				      IRQF_PERCPU|IRQF_NOBALANCING|IRQF_TIMER|
 				      IRQF_FORCE_RESUME,
 				      name, NULL);
 
diff --git a/arch/xtensa/include/asm/pgalloc.h b/arch/xtensa/include/asm/pgalloc.h
index cf914c8c249a..d38eb9237e64 100644
--- a/arch/xtensa/include/asm/pgalloc.h
+++ b/arch/xtensa/include/asm/pgalloc.h
@@ -38,35 +38,46 @@ static inline void pgd_free(struct mm_struct *mm, pgd_t *pgd)
 	free_page((unsigned long)pgd);
 }
 
-/* Use a slab cache for the pte pages (see also sparc64 implementation) */
-
-extern struct kmem_cache *pgtable_cache;
-
 static inline pte_t *pte_alloc_one_kernel(struct mm_struct *mm,
 					 unsigned long address)
 {
-	return kmem_cache_alloc(pgtable_cache, GFP_KERNEL|__GFP_REPEAT);
+	pte_t *ptep;
+	int i;
+
+	ptep = (pte_t *)__get_free_page(GFP_KERNEL|__GFP_REPEAT);
+	if (!ptep)
+		return NULL;
+	for (i = 0; i < 1024; i++)
+		pte_clear(NULL, 0, ptep + i);
+	return ptep;
 }
 
 static inline pgtable_t pte_alloc_one(struct mm_struct *mm,
 					unsigned long addr)
 {
+	pte_t *pte;
 	struct page *page;
 
-	page = virt_to_page(pte_alloc_one_kernel(mm, addr));
-	pgtable_page_ctor(page);
+	pte = pte_alloc_one_kernel(mm, addr);
+	if (!pte)
+		return NULL;
+	page = virt_to_page(pte);
+	if (!pgtable_page_ctor(page)) {
+		__free_page(page);
+		return NULL;
+	}
 	return page;
 }
 
 static inline void pte_free_kernel(struct mm_struct *mm, pte_t *pte)
 {
-	kmem_cache_free(pgtable_cache, pte);
+	free_page((unsigned long)pte);
 }
 
 static inline void pte_free(struct mm_struct *mm, pgtable_t pte)
 {
 	pgtable_page_dtor(pte);
-	kmem_cache_free(pgtable_cache, page_address(pte));
+	__free_page(pte);
 }
 #define pmd_pgtable(pmd) pmd_page(pmd)
 
diff --git a/arch/xtensa/include/asm/pgtable.h b/arch/xtensa/include/asm/pgtable.h
index 0fdf5d043f82..216446295ada 100644
--- a/arch/xtensa/include/asm/pgtable.h
+++ b/arch/xtensa/include/asm/pgtable.h
@@ -220,12 +220,11 @@ extern unsigned long empty_zero_page[1024];
 #ifdef CONFIG_MMU
 extern pgd_t swapper_pg_dir[PAGE_SIZE/sizeof(pgd_t)];
 extern void paging_init(void);
-extern void pgtable_cache_init(void);
 #else
 # define swapper_pg_dir NULL
 static inline void paging_init(void) { }
-static inline void pgtable_cache_init(void) { }
 #endif
+static inline void pgtable_cache_init(void) { }
 
 /*
  * The pmd contains the kernel virtual address of the pte page.
diff --git a/arch/xtensa/include/asm/thread_info.h b/arch/xtensa/include/asm/thread_info.h
index 9481004ac119..470153e8547c 100644
--- a/arch/xtensa/include/asm/thread_info.h
+++ b/arch/xtensa/include/asm/thread_info.h
@@ -76,8 +76,6 @@ struct thread_info {
 
 #endif
 
-#define PREEMPT_ACTIVE		0x10000000
-
 /*
  * macros/functions for gaining access to the thread information structure
  */
diff --git a/arch/xtensa/mm/mmu.c b/arch/xtensa/mm/mmu.c
index a1077570e383..c43771c974be 100644
--- a/arch/xtensa/mm/mmu.c
+++ b/arch/xtensa/mm/mmu.c
@@ -50,23 +50,3 @@ void __init init_mmu(void)
 	 */
 	set_ptevaddr_register(PGTABLE_START);
 }
-
-struct kmem_cache *pgtable_cache __read_mostly;
-
-static void pgd_ctor(void *addr)
-{
-	pte_t *ptep = (pte_t *)addr;
-	int i;
-
-	for (i = 0; i < 1024; i++, ptep++)
-		pte_clear(NULL, 0, ptep);
-
-}
-
-void __init pgtable_cache_init(void)
-{
-	pgtable_cache = kmem_cache_create("pgd",
-			PAGE_SIZE, PAGE_SIZE,
-			SLAB_HWCACHE_ALIGN,
-			pgd_ctor);
-}
diff --git a/block/blk-cgroup.h b/block/blk-cgroup.h
index ae6969a7ffd4..1610b22edf09 100644
--- a/block/blk-cgroup.h
+++ b/block/blk-cgroup.h
@@ -402,6 +402,11 @@ struct request_list *__blk_queue_next_rl(struct request_list *rl,
 #define blk_queue_for_each_rl(rl, q)	\
 	for ((rl) = &(q)->root_rl; (rl); (rl) = __blk_queue_next_rl((rl), (q)))
 
+static inline void blkg_stat_init(struct blkg_stat *stat)
+{
+	u64_stats_init(&stat->syncp);
+}
+
 /**
  * blkg_stat_add - add a value to a blkg_stat
  * @stat: target blkg_stat
@@ -458,6 +463,11 @@ static inline void blkg_stat_merge(struct blkg_stat *to, struct blkg_stat *from)
 	blkg_stat_add(to, blkg_stat_read(from));
 }
 
+static inline void blkg_rwstat_init(struct blkg_rwstat *rwstat)
+{
+	u64_stats_init(&rwstat->syncp);
+}
+
 /**
  * blkg_rwstat_add - add a value to a blkg_rwstat
  * @rwstat: target blkg_rwstat
diff --git a/block/blk-ioc.c b/block/blk-ioc.c
index 46cd7bd18b34..242df01413f6 100644
--- a/block/blk-ioc.c
+++ b/block/blk-ioc.c
@@ -6,7 +6,6 @@
 #include <linux/init.h>
 #include <linux/bio.h>
 #include <linux/blkdev.h>
-#include <linux/bootmem.h>	/* for max_pfn/max_low_pfn */
 #include <linux/slab.h>
 
 #include "blk.h"
diff --git a/block/blk-mq-cpu.c b/block/blk-mq-cpu.c
index f8ea39d7ae54..0045ace9bdf0 100644
--- a/block/blk-mq-cpu.c
+++ b/block/blk-mq-cpu.c
@@ -13,8 +13,8 @@
 static LIST_HEAD(blk_mq_cpu_notify_list);
 static DEFINE_SPINLOCK(blk_mq_cpu_notify_lock);
 
-static int __cpuinit blk_mq_main_cpu_notify(struct notifier_block *self,
-					    unsigned long action, void *hcpu)
+static int blk_mq_main_cpu_notify(struct notifier_block *self,
+				  unsigned long action, void *hcpu)
 {
 	unsigned int cpu = (unsigned long) hcpu;
 	struct blk_mq_cpu_notifier *notify;
@@ -28,8 +28,8 @@ static int __cpuinit blk_mq_main_cpu_notify(struct notifier_block *self,
 	return NOTIFY_OK;
 }
 
-static void __cpuinit blk_mq_cpu_notify(void *data, unsigned long action,
-					unsigned int cpu)
+static void blk_mq_cpu_notify(void *data, unsigned long action,
+			      unsigned int cpu)
 {
 	if (action == CPU_DEAD || action == CPU_DEAD_FROZEN) {
 		/*
diff --git a/block/blk-mq.c b/block/blk-mq.c
index 88d4e864d4c0..cdc629cf075b 100644
--- a/block/blk-mq.c
+++ b/block/blk-mq.c
@@ -171,9 +171,12 @@ bool blk_mq_can_queue(struct blk_mq_hw_ctx *hctx)
 }
 EXPORT_SYMBOL(blk_mq_can_queue);
 
-static void blk_mq_rq_ctx_init(struct blk_mq_ctx *ctx, struct request *rq,
-			       unsigned int rw_flags)
+static void blk_mq_rq_ctx_init(struct request_queue *q, struct blk_mq_ctx *ctx,
+			       struct request *rq, unsigned int rw_flags)
 {
+	if (blk_queue_io_stat(q))
+		rw_flags |= REQ_IO_STAT;
+
 	rq->mq_ctx = ctx;
 	rq->cmd_flags = rw_flags;
 	ctx->rq_dispatched[rw_is_sync(rw_flags)]++;
@@ -197,7 +200,7 @@ static struct request *blk_mq_alloc_request_pinned(struct request_queue *q,
 
 		rq = __blk_mq_alloc_request(hctx, gfp & ~__GFP_WAIT, reserved);
 		if (rq) {
-			blk_mq_rq_ctx_init(ctx, rq, rw);
+			blk_mq_rq_ctx_init(q, ctx, rq, rw);
 			break;
 		} else if (!(gfp & __GFP_WAIT))
 			break;
@@ -319,7 +322,7 @@ void __blk_mq_end_io(struct request *rq, int error)
 		blk_mq_complete_request(rq, error);
 }
 
-#if defined(CONFIG_SMP) && defined(CONFIG_USE_GENERIC_SMP_HELPERS)
+#if defined(CONFIG_SMP)
 
 /*
  * Called with interrupts disabled.
@@ -361,7 +364,7 @@ static int ipi_remote_cpu(struct blk_mq_ctx *ctx, const int cpu,
 
 	return true;
 }
-#else /* CONFIG_SMP && CONFIG_USE_GENERIC_SMP_HELPERS */
+#else /* CONFIG_SMP */
 static int ipi_remote_cpu(struct blk_mq_ctx *ctx, const int cpu,
 			  struct request *rq, const int error)
 {
@@ -718,6 +721,8 @@ static void __blk_mq_insert_request(struct blk_mq_hw_ctx *hctx,
 {
 	struct blk_mq_ctx *ctx = rq->mq_ctx;
 
+	trace_block_rq_insert(hctx->queue, rq);
+
 	list_add_tail(&rq->queuelist, &ctx->rq_list);
 	blk_mq_hctx_mark_pending(hctx, ctx);
 
@@ -921,7 +926,7 @@ static void blk_mq_make_request(struct request_queue *q, struct bio *bio)
 	trace_block_getrq(q, bio, rw);
 	rq = __blk_mq_alloc_request(hctx, GFP_ATOMIC, false);
 	if (likely(rq))
-		blk_mq_rq_ctx_init(ctx, rq, rw);
+		blk_mq_rq_ctx_init(q, ctx, rq, rw);
 	else {
 		blk_mq_put_ctx(ctx);
 		trace_block_sleeprq(q, bio, rw);
@@ -1377,6 +1382,7 @@ struct request_queue *blk_mq_init_queue(struct blk_mq_reg *reg,
 	q->queue_hw_ctx = hctxs;
 
 	q->mq_ops = reg->ops;
+	q->queue_flags |= QUEUE_FLAG_MQ_DEFAULT;
 
 	blk_queue_make_request(q, blk_mq_make_request);
 	blk_queue_rq_timed_out(q, reg->ops->timeout);
@@ -1444,7 +1450,7 @@ void blk_mq_free_queue(struct request_queue *q)
 EXPORT_SYMBOL(blk_mq_free_queue);
 
 /* Basically redo blk_mq_init_queue with queue frozen */
-static void __cpuinit blk_mq_queue_reinit(struct request_queue *q)
+static void blk_mq_queue_reinit(struct request_queue *q)
 {
 	blk_mq_freeze_queue(q);
 
@@ -1461,8 +1467,8 @@ static void __cpuinit blk_mq_queue_reinit(struct request_queue *q)
 	blk_mq_unfreeze_queue(q);
 }
 
-static int __cpuinit blk_mq_queue_reinit_notify(struct notifier_block *nb,
-		unsigned long action, void *hcpu)
+static int blk_mq_queue_reinit_notify(struct notifier_block *nb,
+				      unsigned long action, void *hcpu)
 {
 	struct request_queue *q;
 
diff --git a/block/blk-softirq.c b/block/blk-softirq.c
index ce4b8bfd3d27..57790c1a97eb 100644
--- a/block/blk-softirq.c
+++ b/block/blk-softirq.c
@@ -36,7 +36,7 @@ static void blk_done_softirq(struct softirq_action *h)
 	}
 }
 
-#if defined(CONFIG_SMP) && defined(CONFIG_USE_GENERIC_SMP_HELPERS)
+#ifdef CONFIG_SMP
 static void trigger_softirq(void *data)
 {
 	struct request *rq = data;
@@ -71,7 +71,7 @@ static int raise_blk_irq(int cpu, struct request *rq)
 
 	return 1;
 }
-#else /* CONFIG_SMP && CONFIG_USE_GENERIC_SMP_HELPERS */
+#else /* CONFIG_SMP */
 static int raise_blk_irq(int cpu, struct request *rq)
 {
 	return 1;
diff --git a/block/blk-sysfs.c b/block/blk-sysfs.c
index 4f8c4d90ec73..97779522472f 100644
--- a/block/blk-sysfs.c
+++ b/block/blk-sysfs.c
@@ -288,7 +288,7 @@ static ssize_t
 queue_rq_affinity_store(struct request_queue *q, const char *page, size_t count)
 {
 	ssize_t ret = -EINVAL;
-#if defined(CONFIG_USE_GENERIC_SMP_HELPERS)
+#ifdef CONFIG_SMP
 	unsigned long val;
 
 	ret = queue_var_store(&val, page, count);
diff --git a/block/blk-throttle.c b/block/blk-throttle.c
index 8331aba9426f..06534049afba 100644
--- a/block/blk-throttle.c
+++ b/block/blk-throttle.c
@@ -256,6 +256,12 @@ static struct throtl_data *sq_to_td(struct throtl_service_queue *sq)
 	}								\
 } while (0)
 
+static void tg_stats_init(struct tg_stats_cpu *tg_stats)
+{
+	blkg_rwstat_init(&tg_stats->service_bytes);
+	blkg_rwstat_init(&tg_stats->serviced);
+}
+
 /*
  * Worker for allocating per cpu stat for tgs. This is scheduled on the
  * system_wq once there are some groups on the alloc_list waiting for
@@ -269,12 +275,16 @@ static void tg_stats_alloc_fn(struct work_struct *work)
 
 alloc_stats:
 	if (!stats_cpu) {
+		int cpu;
+
 		stats_cpu = alloc_percpu(struct tg_stats_cpu);
 		if (!stats_cpu) {
 			/* allocation failed, try again after some time */
 			schedule_delayed_work(dwork, msecs_to_jiffies(10));
 			return;
 		}
+		for_each_possible_cpu(cpu)
+			tg_stats_init(per_cpu_ptr(stats_cpu, cpu));
 	}
 
 	spin_lock_irq(&tg_stats_alloc_lock);
diff --git a/block/cfq-iosched.c b/block/cfq-iosched.c
index 434944cbd761..4d5cec1ad80d 100644
--- a/block/cfq-iosched.c
+++ b/block/cfq-iosched.c
@@ -1508,6 +1508,29 @@ static void cfq_init_cfqg_base(struct cfq_group *cfqg)
 }
 
 #ifdef CONFIG_CFQ_GROUP_IOSCHED
+static void cfqg_stats_init(struct cfqg_stats *stats)
+{
+	blkg_rwstat_init(&stats->service_bytes);
+	blkg_rwstat_init(&stats->serviced);
+	blkg_rwstat_init(&stats->merged);
+	blkg_rwstat_init(&stats->service_time);
+	blkg_rwstat_init(&stats->wait_time);
+	blkg_rwstat_init(&stats->queued);
+
+	blkg_stat_init(&stats->sectors);
+	blkg_stat_init(&stats->time);
+
+#ifdef CONFIG_DEBUG_BLK_CGROUP
+	blkg_stat_init(&stats->unaccounted_time);
+	blkg_stat_init(&stats->avg_queue_size_sum);
+	blkg_stat_init(&stats->avg_queue_size_samples);
+	blkg_stat_init(&stats->dequeue);
+	blkg_stat_init(&stats->group_wait_time);
+	blkg_stat_init(&stats->idle_time);
+	blkg_stat_init(&stats->empty_time);
+#endif
+}
+
 static void cfq_pd_init(struct blkcg_gq *blkg)
 {
 	struct cfq_group *cfqg = blkg_to_cfqg(blkg);
@@ -1515,6 +1538,8 @@ static void cfq_pd_init(struct blkcg_gq *blkg)
 	cfq_init_cfqg_base(cfqg);
 	cfqg->weight = blkg->blkcg->cfq_weight;
 	cfqg->leaf_weight = blkg->blkcg->cfq_leaf_weight;
+	cfqg_stats_init(&cfqg->stats);
+	cfqg_stats_init(&cfqg->dead_stats);
 }
 
 static void cfq_pd_offline(struct blkcg_gq *blkg)
diff --git a/crypto/af_alg.c b/crypto/af_alg.c
index ac33d5f30778..966f893711b3 100644
--- a/crypto/af_alg.c
+++ b/crypto/af_alg.c
@@ -434,7 +434,7 @@ int af_alg_wait_for_completion(int err, struct af_alg_completion *completion)
 	case -EINPROGRESS:
 	case -EBUSY:
 		wait_for_completion(&completion->completion);
-		INIT_COMPLETION(completion->completion);
+		reinit_completion(&completion->completion);
 		err = completion->err;
 		break;
 	};
diff --git a/crypto/async_tx/async_memcpy.c b/crypto/async_tx/async_memcpy.c
index 9e62feffb374..f8c0b8dbeb75 100644
--- a/crypto/async_tx/async_memcpy.c
+++ b/crypto/async_tx/async_memcpy.c
@@ -50,33 +50,36 @@ async_memcpy(struct page *dest, struct page *src, unsigned int dest_offset,
 						      &dest, 1, &src, 1, len);
 	struct dma_device *device = chan ? chan->device : NULL;
 	struct dma_async_tx_descriptor *tx = NULL;
+	struct dmaengine_unmap_data *unmap = NULL;
 
-	if (device && is_dma_copy_aligned(device, src_offset, dest_offset, len)) {
-		dma_addr_t dma_dest, dma_src;
+	if (device)
+		unmap = dmaengine_get_unmap_data(device->dev, 2, GFP_NOIO);
+
+	if (unmap && is_dma_copy_aligned(device, src_offset, dest_offset, len)) {
 		unsigned long dma_prep_flags = 0;
 
 		if (submit->cb_fn)
 			dma_prep_flags |= DMA_PREP_INTERRUPT;
 		if (submit->flags & ASYNC_TX_FENCE)
 			dma_prep_flags |= DMA_PREP_FENCE;
-		dma_dest = dma_map_page(device->dev, dest, dest_offset, len,
-					DMA_FROM_DEVICE);
-
-		dma_src = dma_map_page(device->dev, src, src_offset, len,
-				       DMA_TO_DEVICE);
-
-		tx = device->device_prep_dma_memcpy(chan, dma_dest, dma_src,
-						    len, dma_prep_flags);
-		if (!tx) {
-			dma_unmap_page(device->dev, dma_dest, len,
-				       DMA_FROM_DEVICE);
-			dma_unmap_page(device->dev, dma_src, len,
-				       DMA_TO_DEVICE);
-		}
+
+		unmap->to_cnt = 1;
+		unmap->addr[0] = dma_map_page(device->dev, src, src_offset, len,
+					      DMA_TO_DEVICE);
+		unmap->from_cnt = 1;
+		unmap->addr[1] = dma_map_page(device->dev, dest, dest_offset, len,
+					      DMA_FROM_DEVICE);
+		unmap->len = len;
+
+		tx = device->device_prep_dma_memcpy(chan, unmap->addr[1],
+						    unmap->addr[0], len,
+						    dma_prep_flags);
 	}
 
 	if (tx) {
 		pr_debug("%s: (async) len: %zu\n", __func__, len);
+
+		dma_set_unmap(tx, unmap);
 		async_tx_submit(chan, tx, submit);
 	} else {
 		void *dest_buf, *src_buf;
@@ -96,6 +99,8 @@ async_memcpy(struct page *dest, struct page *src, unsigned int dest_offset,
 		async_tx_sync_epilog(submit);
 	}
 
+	dmaengine_unmap_put(unmap);
+
 	return tx;
 }
 EXPORT_SYMBOL_GPL(async_memcpy);
diff --git a/crypto/async_tx/async_pq.c b/crypto/async_tx/async_pq.c
index 91d5d385899e..d05327caf69d 100644
--- a/crypto/async_tx/async_pq.c
+++ b/crypto/async_tx/async_pq.c
@@ -46,49 +46,24 @@ static struct page *pq_scribble_page;
  * do_async_gen_syndrome - asynchronously calculate P and/or Q
  */
 static __async_inline struct dma_async_tx_descriptor *
-do_async_gen_syndrome(struct dma_chan *chan, struct page **blocks,
-		      const unsigned char *scfs, unsigned int offset, int disks,
-		      size_t len, dma_addr_t *dma_src,
+do_async_gen_syndrome(struct dma_chan *chan,
+		      const unsigned char *scfs, int disks,
+		      struct dmaengine_unmap_data *unmap,
+		      enum dma_ctrl_flags dma_flags,
 		      struct async_submit_ctl *submit)
 {
 	struct dma_async_tx_descriptor *tx = NULL;
 	struct dma_device *dma = chan->device;
-	enum dma_ctrl_flags dma_flags = 0;
 	enum async_tx_flags flags_orig = submit->flags;
 	dma_async_tx_callback cb_fn_orig = submit->cb_fn;
 	dma_async_tx_callback cb_param_orig = submit->cb_param;
 	int src_cnt = disks - 2;
-	unsigned char coefs[src_cnt];
 	unsigned short pq_src_cnt;
 	dma_addr_t dma_dest[2];
 	int src_off = 0;
-	int idx;
-	int i;
 
-	/* DMAs use destinations as sources, so use BIDIRECTIONAL mapping */
-	if (P(blocks, disks))
-		dma_dest[0] = dma_map_page(dma->dev, P(blocks, disks), offset,
-					   len, DMA_BIDIRECTIONAL);
-	else
-		dma_flags |= DMA_PREP_PQ_DISABLE_P;
-	if (Q(blocks, disks))
-		dma_dest[1] = dma_map_page(dma->dev, Q(blocks, disks), offset,
-					   len, DMA_BIDIRECTIONAL);
-	else
-		dma_flags |= DMA_PREP_PQ_DISABLE_Q;
-
-	/* convert source addresses being careful to collapse 'empty'
-	 * sources and update the coefficients accordingly
-	 */
-	for (i = 0, idx = 0; i < src_cnt; i++) {
-		if (blocks[i] == NULL)
-			continue;
-		dma_src[idx] = dma_map_page(dma->dev, blocks[i], offset, len,
-					    DMA_TO_DEVICE);
-		coefs[idx] = scfs[i];
-		idx++;
-	}
-	src_cnt = idx;
+	if (submit->flags & ASYNC_TX_FENCE)
+		dma_flags |= DMA_PREP_FENCE;
 
 	while (src_cnt > 0) {
 		submit->flags = flags_orig;
@@ -100,28 +75,25 @@ do_async_gen_syndrome(struct dma_chan *chan, struct page **blocks,
 		if (src_cnt > pq_src_cnt) {
 			submit->flags &= ~ASYNC_TX_ACK;
 			submit->flags |= ASYNC_TX_FENCE;
-			dma_flags |= DMA_COMPL_SKIP_DEST_UNMAP;
 			submit->cb_fn = NULL;
 			submit->cb_param = NULL;
 		} else {
-			dma_flags &= ~DMA_COMPL_SKIP_DEST_UNMAP;
 			submit->cb_fn = cb_fn_orig;
 			submit->cb_param = cb_param_orig;
 			if (cb_fn_orig)
 				dma_flags |= DMA_PREP_INTERRUPT;
 		}
-		if (submit->flags & ASYNC_TX_FENCE)
-			dma_flags |= DMA_PREP_FENCE;
 
-		/* Since we have clobbered the src_list we are committed
-		 * to doing this asynchronously.  Drivers force forward
-		 * progress in case they can not provide a descriptor
+		/* Drivers force forward progress in case they can not provide
+		 * a descriptor
 		 */
 		for (;;) {
+			dma_dest[0] = unmap->addr[disks - 2];
+			dma_dest[1] = unmap->addr[disks - 1];
 			tx = dma->device_prep_dma_pq(chan, dma_dest,
-						     &dma_src[src_off],
+						     &unmap->addr[src_off],
 						     pq_src_cnt,
-						     &coefs[src_off], len,
+						     &scfs[src_off], unmap->len,
 						     dma_flags);
 			if (likely(tx))
 				break;
@@ -129,6 +101,7 @@ do_async_gen_syndrome(struct dma_chan *chan, struct page **blocks,
 			dma_async_issue_pending(chan);
 		}
 
+		dma_set_unmap(tx, unmap);
 		async_tx_submit(chan, tx, submit);
 		submit->depend_tx = tx;
 
@@ -188,10 +161,6 @@ do_sync_gen_syndrome(struct page **blocks, unsigned int offset, int disks,
  * set to NULL those buffers will be replaced with the raid6_zero_page
  * in the synchronous path and omitted in the hardware-asynchronous
  * path.
- *
- * 'blocks' note: if submit->scribble is NULL then the contents of
- * 'blocks' may be overwritten to perform address conversions
- * (dma_map_page() or page_address()).
  */
 struct dma_async_tx_descriptor *
 async_gen_syndrome(struct page **blocks, unsigned int offset, int disks,
@@ -202,26 +171,69 @@ async_gen_syndrome(struct page **blocks, unsigned int offset, int disks,
 						      &P(blocks, disks), 2,
 						      blocks, src_cnt, len);
 	struct dma_device *device = chan ? chan->device : NULL;
-	dma_addr_t *dma_src = NULL;
+	struct dmaengine_unmap_data *unmap = NULL;
 
 	BUG_ON(disks > 255 || !(P(blocks, disks) || Q(blocks, disks)));
 
-	if (submit->scribble)
-		dma_src = submit->scribble;
-	else if (sizeof(dma_addr_t) <= sizeof(struct page *))
-		dma_src = (dma_addr_t *) blocks;
+	if (device)
+		unmap = dmaengine_get_unmap_data(device->dev, disks, GFP_NOIO);
 
-	if (dma_src && device &&
+	if (unmap &&
 	    (src_cnt <= dma_maxpq(device, 0) ||
 	     dma_maxpq(device, DMA_PREP_CONTINUE) > 0) &&
 	    is_dma_pq_aligned(device, offset, 0, len)) {
+		struct dma_async_tx_descriptor *tx;
+		enum dma_ctrl_flags dma_flags = 0;
+		unsigned char coefs[src_cnt];
+		int i, j;
+
 		/* run the p+q asynchronously */
 		pr_debug("%s: (async) disks: %d len: %zu\n",
 			 __func__, disks, len);
-		return do_async_gen_syndrome(chan, blocks, raid6_gfexp, offset,
-					     disks, len, dma_src, submit);
+
+		/* convert source addresses being careful to collapse 'empty'
+		 * sources and update the coefficients accordingly
+		 */
+		unmap->len = len;
+		for (i = 0, j = 0; i < src_cnt; i++) {
+			if (blocks[i] == NULL)
+				continue;
+			unmap->addr[j] = dma_map_page(device->dev, blocks[i], offset,
+						      len, DMA_TO_DEVICE);
+			coefs[j] = raid6_gfexp[i];
+			unmap->to_cnt++;
+			j++;
+		}
+
+		/*
+		 * DMAs use destinations as sources,
+		 * so use BIDIRECTIONAL mapping
+		 */
+		unmap->bidi_cnt++;
+		if (P(blocks, disks))
+			unmap->addr[j++] = dma_map_page(device->dev, P(blocks, disks),
+							offset, len, DMA_BIDIRECTIONAL);
+		else {
+			unmap->addr[j++] = 0;
+			dma_flags |= DMA_PREP_PQ_DISABLE_P;
+		}
+
+		unmap->bidi_cnt++;
+		if (Q(blocks, disks))
+			unmap->addr[j++] = dma_map_page(device->dev, Q(blocks, disks),
+						       offset, len, DMA_BIDIRECTIONAL);
+		else {
+			unmap->addr[j++] = 0;
+			dma_flags |= DMA_PREP_PQ_DISABLE_Q;
+		}
+
+		tx = do_async_gen_syndrome(chan, coefs, j, unmap, dma_flags, submit);
+		dmaengine_unmap_put(unmap);
+		return tx;
 	}
 
+	dmaengine_unmap_put(unmap);
+
 	/* run the pq synchronously */
 	pr_debug("%s: (sync) disks: %d len: %zu\n", __func__, disks, len);
 
@@ -277,50 +289,60 @@ async_syndrome_val(struct page **blocks, unsigned int offset, int disks,
 	struct dma_async_tx_descriptor *tx;
 	unsigned char coefs[disks-2];
 	enum dma_ctrl_flags dma_flags = submit->cb_fn ? DMA_PREP_INTERRUPT : 0;
-	dma_addr_t *dma_src = NULL;
-	int src_cnt = 0;
+	struct dmaengine_unmap_data *unmap = NULL;
 
 	BUG_ON(disks < 4);
 
-	if (submit->scribble)
-		dma_src = submit->scribble;
-	else if (sizeof(dma_addr_t) <= sizeof(struct page *))
-		dma_src = (dma_addr_t *) blocks;
+	if (device)
+		unmap = dmaengine_get_unmap_data(device->dev, disks, GFP_NOIO);
 
-	if (dma_src && device && disks <= dma_maxpq(device, 0) &&
+	if (unmap && disks <= dma_maxpq(device, 0) &&
 	    is_dma_pq_aligned(device, offset, 0, len)) {
 		struct device *dev = device->dev;
-		dma_addr_t *pq = &dma_src[disks-2];
-		int i;
+		dma_addr_t pq[2];
+		int i, j = 0, src_cnt = 0;
 
 		pr_debug("%s: (async) disks: %d len: %zu\n",
 			 __func__, disks, len);
-		if (!P(blocks, disks))
+
+		unmap->len = len;
+		for (i = 0; i < disks-2; i++)
+			if (likely(blocks[i])) {
+				unmap->addr[j] = dma_map_page(dev, blocks[i],
+							      offset, len,
+							      DMA_TO_DEVICE);
+				coefs[j] = raid6_gfexp[i];
+				unmap->to_cnt++;
+				src_cnt++;
+				j++;
+			}
+
+		if (!P(blocks, disks)) {
+			pq[0] = 0;
 			dma_flags |= DMA_PREP_PQ_DISABLE_P;
-		else
+		} else {
 			pq[0] = dma_map_page(dev, P(blocks, disks),
 					     offset, len,
 					     DMA_TO_DEVICE);
-		if (!Q(blocks, disks))
+			unmap->addr[j++] = pq[0];
+			unmap->to_cnt++;
+		}
+		if (!Q(blocks, disks)) {
+			pq[1] = 0;
 			dma_flags |= DMA_PREP_PQ_DISABLE_Q;
-		else
+		} else {
 			pq[1] = dma_map_page(dev, Q(blocks, disks),
 					     offset, len,
 					     DMA_TO_DEVICE);
+			unmap->addr[j++] = pq[1];
+			unmap->to_cnt++;
+		}
 
 		if (submit->flags & ASYNC_TX_FENCE)
 			dma_flags |= DMA_PREP_FENCE;
-		for (i = 0; i < disks-2; i++)
-			if (likely(blocks[i])) {
-				dma_src[src_cnt] = dma_map_page(dev, blocks[i],
-								offset, len,
-								DMA_TO_DEVICE);
-				coefs[src_cnt] = raid6_gfexp[i];
-				src_cnt++;
-			}
-
 		for (;;) {
-			tx = device->device_prep_dma_pq_val(chan, pq, dma_src,
+			tx = device->device_prep_dma_pq_val(chan, pq,
+							    unmap->addr,
 							    src_cnt,
 							    coefs,
 							    len, pqres,
@@ -330,6 +352,8 @@ async_syndrome_val(struct page **blocks, unsigned int offset, int disks,
 			async_tx_quiesce(&submit->depend_tx);
 			dma_async_issue_pending(chan);
 		}
+
+		dma_set_unmap(tx, unmap);
 		async_tx_submit(chan, tx, submit);
 
 		return tx;
diff --git a/crypto/async_tx/async_raid6_recov.c b/crypto/async_tx/async_raid6_recov.c
index a9f08a6a582e..934a84981495 100644
--- a/crypto/async_tx/async_raid6_recov.c
+++ b/crypto/async_tx/async_raid6_recov.c
@@ -26,6 +26,7 @@
 #include <linux/dma-mapping.h>
 #include <linux/raid/pq.h>
 #include <linux/async_tx.h>
+#include <linux/dmaengine.h>
 
 static struct dma_async_tx_descriptor *
 async_sum_product(struct page *dest, struct page **srcs, unsigned char *coef,
@@ -34,35 +35,45 @@ async_sum_product(struct page *dest, struct page **srcs, unsigned char *coef,
 	struct dma_chan *chan = async_tx_find_channel(submit, DMA_PQ,
 						      &dest, 1, srcs, 2, len);
 	struct dma_device *dma = chan ? chan->device : NULL;
+	struct dmaengine_unmap_data *unmap = NULL;
 	const u8 *amul, *bmul;
 	u8 ax, bx;
 	u8 *a, *b, *c;
 
-	if (dma) {
-		dma_addr_t dma_dest[2];
-		dma_addr_t dma_src[2];
+	if (dma)
+		unmap = dmaengine_get_unmap_data(dma->dev, 3, GFP_NOIO);
+
+	if (unmap) {
 		struct device *dev = dma->dev;
+		dma_addr_t pq[2];
 		struct dma_async_tx_descriptor *tx;
 		enum dma_ctrl_flags dma_flags = DMA_PREP_PQ_DISABLE_P;
 
 		if (submit->flags & ASYNC_TX_FENCE)
 			dma_flags |= DMA_PREP_FENCE;
-		dma_dest[1] = dma_map_page(dev, dest, 0, len, DMA_BIDIRECTIONAL);
-		dma_src[0] = dma_map_page(dev, srcs[0], 0, len, DMA_TO_DEVICE);
-		dma_src[1] = dma_map_page(dev, srcs[1], 0, len, DMA_TO_DEVICE);
-		tx = dma->device_prep_dma_pq(chan, dma_dest, dma_src, 2, coef,
+		unmap->addr[0] = dma_map_page(dev, srcs[0], 0, len, DMA_TO_DEVICE);
+		unmap->addr[1] = dma_map_page(dev, srcs[1], 0, len, DMA_TO_DEVICE);
+		unmap->to_cnt = 2;
+
+		unmap->addr[2] = dma_map_page(dev, dest, 0, len, DMA_BIDIRECTIONAL);
+		unmap->bidi_cnt = 1;
+		/* engine only looks at Q, but expects it to follow P */
+		pq[1] = unmap->addr[2];
+
+		unmap->len = len;
+		tx = dma->device_prep_dma_pq(chan, pq, unmap->addr, 2, coef,
 					     len, dma_flags);
 		if (tx) {
+			dma_set_unmap(tx, unmap);
 			async_tx_submit(chan, tx, submit);
+			dmaengine_unmap_put(unmap);
 			return tx;
 		}
 
 		/* could not get a descriptor, unmap and fall through to
 		 * the synchronous path
 		 */
-		dma_unmap_page(dev, dma_dest[1], len, DMA_BIDIRECTIONAL);
-		dma_unmap_page(dev, dma_src[0], len, DMA_TO_DEVICE);
-		dma_unmap_page(dev, dma_src[1], len, DMA_TO_DEVICE);
+		dmaengine_unmap_put(unmap);
 	}
 
 	/* run the operation synchronously */
@@ -89,23 +100,38 @@ async_mult(struct page *dest, struct page *src, u8 coef, size_t len,
 	struct dma_chan *chan = async_tx_find_channel(submit, DMA_PQ,
 						      &dest, 1, &src, 1, len);
 	struct dma_device *dma = chan ? chan->device : NULL;
+	struct dmaengine_unmap_data *unmap = NULL;
 	const u8 *qmul; /* Q multiplier table */
 	u8 *d, *s;
 
-	if (dma) {
+	if (dma)
+		unmap = dmaengine_get_unmap_data(dma->dev, 3, GFP_NOIO);
+
+	if (unmap) {
 		dma_addr_t dma_dest[2];
-		dma_addr_t dma_src[1];
 		struct device *dev = dma->dev;
 		struct dma_async_tx_descriptor *tx;
 		enum dma_ctrl_flags dma_flags = DMA_PREP_PQ_DISABLE_P;
 
 		if (submit->flags & ASYNC_TX_FENCE)
 			dma_flags |= DMA_PREP_FENCE;
-		dma_dest[1] = dma_map_page(dev, dest, 0, len, DMA_BIDIRECTIONAL);
-		dma_src[0] = dma_map_page(dev, src, 0, len, DMA_TO_DEVICE);
-		tx = dma->device_prep_dma_pq(chan, dma_dest, dma_src, 1, &coef,
-					     len, dma_flags);
+		unmap->addr[0] = dma_map_page(dev, src, 0, len, DMA_TO_DEVICE);
+		unmap->to_cnt++;
+		unmap->addr[1] = dma_map_page(dev, dest, 0, len, DMA_BIDIRECTIONAL);
+		dma_dest[1] = unmap->addr[1];
+		unmap->bidi_cnt++;
+		unmap->len = len;
+
+		/* this looks funny, but the engine looks for Q at
+		 * dma_dest[1] and ignores dma_dest[0] as a dest
+		 * due to DMA_PREP_PQ_DISABLE_P
+		 */
+		tx = dma->device_prep_dma_pq(chan, dma_dest, unmap->addr,
+					     1, &coef, len, dma_flags);
+
 		if (tx) {
+			dma_set_unmap(tx, unmap);
+			dmaengine_unmap_put(unmap);
 			async_tx_submit(chan, tx, submit);
 			return tx;
 		}
@@ -113,8 +139,7 @@ async_mult(struct page *dest, struct page *src, u8 coef, size_t len,
 		/* could not get a descriptor, unmap and fall through to
 		 * the synchronous path
 		 */
-		dma_unmap_page(dev, dma_dest[1], len, DMA_BIDIRECTIONAL);
-		dma_unmap_page(dev, dma_src[0], len, DMA_TO_DEVICE);
+		dmaengine_unmap_put(unmap);
 	}
 
 	/* no channel available, or failed to allocate a descriptor, so
diff --git a/crypto/async_tx/async_tx.c b/crypto/async_tx/async_tx.c
index 7be34248b450..39ea4791a3c9 100644
--- a/crypto/async_tx/async_tx.c
+++ b/crypto/async_tx/async_tx.c
@@ -128,7 +128,7 @@ async_tx_channel_switch(struct dma_async_tx_descriptor *depend_tx,
 		}
 		device->device_issue_pending(chan);
 	} else {
-		if (dma_wait_for_async_tx(depend_tx) != DMA_SUCCESS)
+		if (dma_wait_for_async_tx(depend_tx) != DMA_COMPLETE)
 			panic("%s: DMA error waiting for depend_tx\n",
 			      __func__);
 		tx->tx_submit(tx);
@@ -280,7 +280,7 @@ void async_tx_quiesce(struct dma_async_tx_descriptor **tx)
 		 * we are referring to the correct operation
 		 */
 		BUG_ON(async_tx_test_ack(*tx));
-		if (dma_wait_for_async_tx(*tx) != DMA_SUCCESS)
+		if (dma_wait_for_async_tx(*tx) != DMA_COMPLETE)
 			panic("%s: DMA error waiting for transaction\n",
 			      __func__);
 		async_tx_ack(*tx);
diff --git a/crypto/async_tx/async_xor.c b/crypto/async_tx/async_xor.c
index 8ade0a0481c6..3c562f5a60bb 100644
--- a/crypto/async_tx/async_xor.c
+++ b/crypto/async_tx/async_xor.c
@@ -33,48 +33,31 @@
 
 /* do_async_xor - dma map the pages and perform the xor with an engine */
 static __async_inline struct dma_async_tx_descriptor *
-do_async_xor(struct dma_chan *chan, struct page *dest, struct page **src_list,
-	     unsigned int offset, int src_cnt, size_t len, dma_addr_t *dma_src,
+do_async_xor(struct dma_chan *chan, struct dmaengine_unmap_data *unmap,
 	     struct async_submit_ctl *submit)
 {
 	struct dma_device *dma = chan->device;
 	struct dma_async_tx_descriptor *tx = NULL;
-	int src_off = 0;
-	int i;
 	dma_async_tx_callback cb_fn_orig = submit->cb_fn;
 	void *cb_param_orig = submit->cb_param;
 	enum async_tx_flags flags_orig = submit->flags;
-	enum dma_ctrl_flags dma_flags;
-	int xor_src_cnt = 0;
-	dma_addr_t dma_dest;
-
-	/* map the dest bidrectional in case it is re-used as a source */
-	dma_dest = dma_map_page(dma->dev, dest, offset, len, DMA_BIDIRECTIONAL);
-	for (i = 0; i < src_cnt; i++) {
-		/* only map the dest once */
-		if (!src_list[i])
-			continue;
-		if (unlikely(src_list[i] == dest)) {
-			dma_src[xor_src_cnt++] = dma_dest;
-			continue;
-		}
-		dma_src[xor_src_cnt++] = dma_map_page(dma->dev, src_list[i], offset,
-						      len, DMA_TO_DEVICE);
-	}
-	src_cnt = xor_src_cnt;
+	enum dma_ctrl_flags dma_flags = 0;
+	int src_cnt = unmap->to_cnt;
+	int xor_src_cnt;
+	dma_addr_t dma_dest = unmap->addr[unmap->to_cnt];
+	dma_addr_t *src_list = unmap->addr;
 
 	while (src_cnt) {
+		dma_addr_t tmp;
+
 		submit->flags = flags_orig;
-		dma_flags = 0;
 		xor_src_cnt = min(src_cnt, (int)dma->max_xor);
-		/* if we are submitting additional xors, leave the chain open,
-		 * clear the callback parameters, and leave the destination
-		 * buffer mapped
+		/* if we are submitting additional xors, leave the chain open
+		 * and clear the callback parameters
 		 */
 		if (src_cnt > xor_src_cnt) {
 			submit->flags &= ~ASYNC_TX_ACK;
 			submit->flags |= ASYNC_TX_FENCE;
-			dma_flags = DMA_COMPL_SKIP_DEST_UNMAP;
 			submit->cb_fn = NULL;
 			submit->cb_param = NULL;
 		} else {
@@ -85,12 +68,18 @@ do_async_xor(struct dma_chan *chan, struct page *dest, struct page **src_list,
 			dma_flags |= DMA_PREP_INTERRUPT;
 		if (submit->flags & ASYNC_TX_FENCE)
 			dma_flags |= DMA_PREP_FENCE;
-		/* Since we have clobbered the src_list we are committed
-		 * to doing this asynchronously.  Drivers force forward progress
-		 * in case they can not provide a descriptor
+
+		/* Drivers force forward progress in case they can not provide a
+		 * descriptor
 		 */
-		tx = dma->device_prep_dma_xor(chan, dma_dest, &dma_src[src_off],
-					      xor_src_cnt, len, dma_flags);
+		tmp = src_list[0];
+		if (src_list > unmap->addr)
+			src_list[0] = dma_dest;
+		tx = dma->device_prep_dma_xor(chan, dma_dest, src_list,
+					      xor_src_cnt, unmap->len,
+					      dma_flags);
+		src_list[0] = tmp;
+
 
 		if (unlikely(!tx))
 			async_tx_quiesce(&submit->depend_tx);
@@ -99,22 +88,21 @@ do_async_xor(struct dma_chan *chan, struct page *dest, struct page **src_list,
 		while (unlikely(!tx)) {
 			dma_async_issue_pending(chan);
 			tx = dma->device_prep_dma_xor(chan, dma_dest,
-						      &dma_src[src_off],
-						      xor_src_cnt, len,
+						      src_list,
+						      xor_src_cnt, unmap->len,
 						      dma_flags);
 		}
 
+		dma_set_unmap(tx, unmap);
 		async_tx_submit(chan, tx, submit);
 		submit->depend_tx = tx;
 
 		if (src_cnt > xor_src_cnt) {
 			/* drop completed sources */
 			src_cnt -= xor_src_cnt;
-			src_off += xor_src_cnt;
-
 			/* use the intermediate result a source */
-			dma_src[--src_off] = dma_dest;
 			src_cnt++;
+			src_list += xor_src_cnt - 1;
 		} else
 			break;
 	}
@@ -189,22 +177,40 @@ async_xor(struct page *dest, struct page **src_list, unsigned int offset,
 	struct dma_chan *chan = async_tx_find_channel(submit, DMA_XOR,
 						      &dest, 1, src_list,
 						      src_cnt, len);
-	dma_addr_t *dma_src = NULL;
+	struct dma_device *device = chan ? chan->device : NULL;
+	struct dmaengine_unmap_data *unmap = NULL;
 
 	BUG_ON(src_cnt <= 1);
 
-	if (submit->scribble)
-		dma_src = submit->scribble;
-	else if (sizeof(dma_addr_t) <= sizeof(struct page *))
-		dma_src = (dma_addr_t *) src_list;
+	if (device)
+		unmap = dmaengine_get_unmap_data(device->dev, src_cnt+1, GFP_NOIO);
+
+	if (unmap && is_dma_xor_aligned(device, offset, 0, len)) {
+		struct dma_async_tx_descriptor *tx;
+		int i, j;
 
-	if (dma_src && chan && is_dma_xor_aligned(chan->device, offset, 0, len)) {
 		/* run the xor asynchronously */
 		pr_debug("%s (async): len: %zu\n", __func__, len);
 
-		return do_async_xor(chan, dest, src_list, offset, src_cnt, len,
-				    dma_src, submit);
+		unmap->len = len;
+		for (i = 0, j = 0; i < src_cnt; i++) {
+			if (!src_list[i])
+				continue;
+			unmap->to_cnt++;
+			unmap->addr[j++] = dma_map_page(device->dev, src_list[i],
+							offset, len, DMA_TO_DEVICE);
+		}
+
+		/* map it bidirectional as it may be re-used as a source */
+		unmap->addr[j] = dma_map_page(device->dev, dest, offset, len,
+					      DMA_BIDIRECTIONAL);
+		unmap->bidi_cnt = 1;
+
+		tx = do_async_xor(chan, unmap, submit);
+		dmaengine_unmap_put(unmap);
+		return tx;
 	} else {
+		dmaengine_unmap_put(unmap);
 		/* run the xor synchronously */
 		pr_debug("%s (sync): len: %zu\n", __func__, len);
 		WARN_ONCE(chan, "%s: no space for dma address conversion\n",
@@ -268,16 +274,14 @@ async_xor_val(struct page *dest, struct page **src_list, unsigned int offset,
 	struct dma_chan *chan = xor_val_chan(submit, dest, src_list, src_cnt, len);
 	struct dma_device *device = chan ? chan->device : NULL;
 	struct dma_async_tx_descriptor *tx = NULL;
-	dma_addr_t *dma_src = NULL;
+	struct dmaengine_unmap_data *unmap = NULL;
 
 	BUG_ON(src_cnt <= 1);
 
-	if (submit->scribble)
-		dma_src = submit->scribble;
-	else if (sizeof(dma_addr_t) <= sizeof(struct page *))
-		dma_src = (dma_addr_t *) src_list;
+	if (device)
+		unmap = dmaengine_get_unmap_data(device->dev, src_cnt, GFP_NOIO);
 
-	if (dma_src && device && src_cnt <= device->max_xor &&
+	if (unmap && src_cnt <= device->max_xor &&
 	    is_dma_xor_aligned(device, offset, 0, len)) {
 		unsigned long dma_prep_flags = 0;
 		int i;
@@ -288,11 +292,15 @@ async_xor_val(struct page *dest, struct page **src_list, unsigned int offset,
 			dma_prep_flags |= DMA_PREP_INTERRUPT;
 		if (submit->flags & ASYNC_TX_FENCE)
 			dma_prep_flags |= DMA_PREP_FENCE;
-		for (i = 0; i < src_cnt; i++)
-			dma_src[i] = dma_map_page(device->dev, src_list[i],
-						  offset, len, DMA_TO_DEVICE);
 
-		tx = device->device_prep_dma_xor_val(chan, dma_src, src_cnt,
+		for (i = 0; i < src_cnt; i++) {
+			unmap->addr[i] = dma_map_page(device->dev, src_list[i],
+						      offset, len, DMA_TO_DEVICE);
+			unmap->to_cnt++;
+		}
+		unmap->len = len;
+
+		tx = device->device_prep_dma_xor_val(chan, unmap->addr, src_cnt,
 						     len, result,
 						     dma_prep_flags);
 		if (unlikely(!tx)) {
@@ -301,11 +309,11 @@ async_xor_val(struct page *dest, struct page **src_list, unsigned int offset,
 			while (!tx) {
 				dma_async_issue_pending(chan);
 				tx = device->device_prep_dma_xor_val(chan,
-					dma_src, src_cnt, len, result,
+					unmap->addr, src_cnt, len, result,
 					dma_prep_flags);
 			}
 		}
-
+		dma_set_unmap(tx, unmap);
 		async_tx_submit(chan, tx, submit);
 	} else {
 		enum async_tx_flags flags_orig = submit->flags;
@@ -327,6 +335,7 @@ async_xor_val(struct page *dest, struct page **src_list, unsigned int offset,
 		async_tx_sync_epilog(submit);
 		submit->flags = flags_orig;
 	}
+	dmaengine_unmap_put(unmap);
 
 	return tx;
 }
diff --git a/crypto/async_tx/raid6test.c b/crypto/async_tx/raid6test.c
index 4a92bac744dc..dad95f45b88f 100644
--- a/crypto/async_tx/raid6test.c
+++ b/crypto/async_tx/raid6test.c
@@ -28,7 +28,7 @@
 #undef pr
 #define pr(fmt, args...) pr_info("raid6test: " fmt, ##args)
 
-#define NDISKS 16 /* Including P and Q */
+#define NDISKS 64 /* Including P and Q */
 
 static struct page *dataptrs[NDISKS];
 static addr_conv_t addr_conv[NDISKS];
@@ -219,6 +219,14 @@ static int raid6_test(void)
 		err += test(11, &tests);
 		err += test(12, &tests);
 	}
+
+	/* the 24 disk case is special for ioatdma as it is the boudary point
+	 * at which it needs to switch from 8-source ops to 16-source
+	 * ops for continuation (assumes DMA_HAS_PQ_CONTINUE is not set)
+	 */
+	if (NDISKS > 24)
+		err += test(24, &tests);
+
 	err += test(NDISKS, &tests);
 
 	pr("\n");
diff --git a/crypto/tcrypt.c b/crypto/tcrypt.c
index 25a5934f0e50..1ab8258fcf56 100644
--- a/crypto/tcrypt.c
+++ b/crypto/tcrypt.c
@@ -493,7 +493,7 @@ static inline int do_one_ahash_op(struct ahash_request *req, int ret)
 		ret = wait_for_completion_interruptible(&tr->completion);
 		if (!ret)
 			ret = tr->err;
-		INIT_COMPLETION(tr->completion);
+		reinit_completion(&tr->completion);
 	}
 	return ret;
 }
@@ -721,7 +721,7 @@ static inline int do_one_acipher_op(struct ablkcipher_request *req, int ret)
 		ret = wait_for_completion_interruptible(&tr->completion);
 		if (!ret)
 			ret = tr->err;
-		INIT_COMPLETION(tr->completion);
+		reinit_completion(&tr->completion);
 	}
 
 	return ret;
diff --git a/crypto/testmgr.c b/crypto/testmgr.c
index e091ef6e1791..432afc03e7c3 100644
--- a/crypto/testmgr.c
+++ b/crypto/testmgr.c
@@ -179,7 +179,7 @@ static int do_one_async_hash_op(struct ahash_request *req,
 		ret = wait_for_completion_interruptible(&tr->completion);
 		if (!ret)
 			ret = tr->err;
-		INIT_COMPLETION(tr->completion);
+		reinit_completion(&tr->completion);
 	}
 	return ret;
 }
@@ -336,7 +336,7 @@ static int __test_hash(struct crypto_ahash *tfm, struct hash_testvec *template,
 				ret = wait_for_completion_interruptible(
 					&tresult.completion);
 				if (!ret && !(ret = tresult.err)) {
-					INIT_COMPLETION(tresult.completion);
+					reinit_completion(&tresult.completion);
 					break;
 				}
 				/* fall through */
@@ -543,7 +543,7 @@ static int __test_aead(struct crypto_aead *tfm, int enc,
 				ret = wait_for_completion_interruptible(
 					&result.completion);
 				if (!ret && !(ret = result.err)) {
-					INIT_COMPLETION(result.completion);
+					reinit_completion(&result.completion);
 					break;
 				}
 			case -EBADMSG:
@@ -697,7 +697,7 @@ static int __test_aead(struct crypto_aead *tfm, int enc,
 				ret = wait_for_completion_interruptible(
 					&result.completion);
 				if (!ret && !(ret = result.err)) {
-					INIT_COMPLETION(result.completion);
+					reinit_completion(&result.completion);
 					break;
 				}
 			case -EBADMSG:
@@ -983,7 +983,7 @@ static int __test_skcipher(struct crypto_ablkcipher *tfm, int enc,
 				ret = wait_for_completion_interruptible(
 					&result.completion);
 				if (!ret && !((ret = result.err))) {
-					INIT_COMPLETION(result.completion);
+					reinit_completion(&result.completion);
 					break;
 				}
 				/* fall through */
@@ -1086,7 +1086,7 @@ static int __test_skcipher(struct crypto_ablkcipher *tfm, int enc,
 				ret = wait_for_completion_interruptible(
 					&result.completion);
 				if (!ret && !((ret = result.err))) {
-					INIT_COMPLETION(result.completion);
+					reinit_completion(&result.completion);
 					break;
 				}
 				/* fall through */
diff --git a/drivers/acpi/acpica/utobject.c b/drivers/acpi/acpica/utobject.c
index 13e045025c33..517af700399d 100644
--- a/drivers/acpi/acpica/utobject.c
+++ b/drivers/acpi/acpica/utobject.c
@@ -356,7 +356,7 @@ u8 acpi_ut_valid_internal_object(void *object)
 	default:
 
 		ACPI_DEBUG_PRINT((ACPI_DB_EXEC,
-				  "%p is not not an ACPI operand obj [%s]\n",
+				  "%p is not an ACPI operand obj [%s]\n",
 				  object, acpi_ut_get_descriptor_name(object)));
 		break;
 	}
diff --git a/drivers/acpi/event.c b/drivers/acpi/event.c
index fdef416c0ff6..cae3b387b867 100644
--- a/drivers/acpi/event.c
+++ b/drivers/acpi/event.c
@@ -78,15 +78,17 @@ enum {
 #define ACPI_GENL_VERSION		0x01
 #define ACPI_GENL_MCAST_GROUP_NAME 	"acpi_mc_group"
 
+static const struct genl_multicast_group acpi_event_mcgrps[] = {
+	{ .name = ACPI_GENL_MCAST_GROUP_NAME, },
+};
+
 static struct genl_family acpi_event_genl_family = {
 	.id = GENL_ID_GENERATE,
 	.name = ACPI_GENL_FAMILY_NAME,
 	.version = ACPI_GENL_VERSION,
 	.maxattr = ACPI_GENL_ATTR_MAX,
-};
-
-static struct genl_multicast_group acpi_event_mcgrp = {
-	.name = ACPI_GENL_MCAST_GROUP_NAME,
+	.mcgrps = acpi_event_mcgrps,
+	.n_mcgrps = ARRAY_SIZE(acpi_event_mcgrps),
 };
 
 int acpi_bus_generate_netlink_event(const char *device_class,
@@ -141,7 +143,7 @@ int acpi_bus_generate_netlink_event(const char *device_class,
 		return result;
 	}
 
-	genlmsg_multicast(skb, 0, acpi_event_mcgrp.id, GFP_ATOMIC);
+	genlmsg_multicast(&acpi_event_genl_family, skb, 0, 0, GFP_ATOMIC);
 	return 0;
 }
 
@@ -149,18 +151,7 @@ EXPORT_SYMBOL(acpi_bus_generate_netlink_event);
 
 static int acpi_event_genetlink_init(void)
 {
-	int result;
-
-	result = genl_register_family(&acpi_event_genl_family);
-	if (result)
-		return result;
-
-	result = genl_register_mc_group(&acpi_event_genl_family,
-					&acpi_event_mcgrp);
-	if (result)
-		genl_unregister_family(&acpi_event_genl_family);
-
-	return result;
+	return genl_register_family(&acpi_event_genl_family);
 }
 
 #else
diff --git a/drivers/acpi/thermal.c b/drivers/acpi/thermal.c
index e600b5dbfcb6..0d9f46b5ae6d 100644
--- a/drivers/acpi/thermal.c
+++ b/drivers/acpi/thermal.c
@@ -514,10 +514,9 @@ static void acpi_thermal_check(void *data)
 {
 	struct acpi_thermal *tz = data;
 
-	if (!tz->tz_enabled) {
-		pr_warn("thermal zone is disabled \n");
+	if (!tz->tz_enabled)
 		return;
-	}
+
 	thermal_zone_device_update(tz->thermal_zone);
 }
 
@@ -569,9 +568,10 @@ static int thermal_set_mode(struct thermal_zone_device *thermal,
 	 */
 	if (mode == THERMAL_DEVICE_ENABLED)
 		enable = 1;
-	else if (mode == THERMAL_DEVICE_DISABLED)
+	else if (mode == THERMAL_DEVICE_DISABLED) {
 		enable = 0;
-	else
+		pr_warn("thermal zone will be disabled\n");
+	} else
 		return -EINVAL;
 
 	if (enable != tz->tz_enabled) {
diff --git a/drivers/ata/libata-eh.c b/drivers/ata/libata-eh.c
index 77bbc8266883..92d7797223be 100644
--- a/drivers/ata/libata-eh.c
+++ b/drivers/ata/libata-eh.c
@@ -3017,7 +3017,7 @@ static inline void ata_eh_pull_park_action(struct ata_port *ap)
 	 * ourselves at the beginning of each pass over the loop.
 	 *
 	 * Additionally, all write accesses to &ap->park_req_pending
-	 * through INIT_COMPLETION() (see below) or complete_all()
+	 * through reinit_completion() (see below) or complete_all()
 	 * (see ata_scsi_park_store()) are protected by the host lock.
 	 * As a result we have that park_req_pending.done is zero on
 	 * exit from this function, i.e. when ATA_EH_PARK actions for
@@ -3031,7 +3031,7 @@ static inline void ata_eh_pull_park_action(struct ata_port *ap)
 	 */
 
 	spin_lock_irqsave(ap->lock, flags);
-	INIT_COMPLETION(ap->park_req_pending);
+	reinit_completion(&ap->park_req_pending);
 	ata_for_each_link(link, ap, EDGE) {
 		ata_for_each_dev(dev, link, ALL) {
 			struct ata_eh_info *ehi = &link->eh_info;
diff --git a/drivers/ata/pata_arasan_cf.c b/drivers/ata/pata_arasan_cf.c
index 853f610af28f..e88690ebfd82 100644
--- a/drivers/ata/pata_arasan_cf.c
+++ b/drivers/ata/pata_arasan_cf.c
@@ -396,8 +396,7 @@ dma_xfer(struct arasan_cf_dev *acdev, dma_addr_t src, dma_addr_t dest, u32 len)
 	struct dma_async_tx_descriptor *tx;
 	struct dma_chan *chan = acdev->dma_chan;
 	dma_cookie_t cookie;
-	unsigned long flags = DMA_PREP_INTERRUPT | DMA_COMPL_SKIP_SRC_UNMAP |
-		DMA_COMPL_SKIP_DEST_UNMAP;
+	unsigned long flags = DMA_PREP_INTERRUPT;
 	int ret = 0;
 
 	tx = chan->device->device_prep_dma_memcpy(chan, dest, src, len, flags);
diff --git a/drivers/atm/idt77252.c b/drivers/atm/idt77252.c
index 272f00927761..1bdf104e90bb 100644
--- a/drivers/atm/idt77252.c
+++ b/drivers/atm/idt77252.c
@@ -3511,7 +3511,7 @@ static int init_card(struct atm_dev *dev)
 	tmp = dev_get_by_name(&init_net, tname);	/* jhs: was "tmp = dev_get(tname);" */
 	if (tmp) {
 		memcpy(card->atmdev->esi, tmp->dev_addr, 6);
-
+		dev_put(tmp);
 		printk("%s: ESI %pM\n", card->name, card->atmdev->esi);
 	}
 	/*
diff --git a/drivers/auxdisplay/cfag12864bfb.c b/drivers/auxdisplay/cfag12864bfb.c
index d585735430dd..a3874034e2ce 100644
--- a/drivers/auxdisplay/cfag12864bfb.c
+++ b/drivers/auxdisplay/cfag12864bfb.c
@@ -102,8 +102,7 @@ static int cfag12864bfb_probe(struct platform_device *device)
 
 	platform_set_drvdata(device, info);
 
-	printk(KERN_INFO "fb%d: %s frame buffer device\n", info->node,
-		info->fix.id);
+	fb_info(info, "%s frame buffer device\n", info->fix.id);
 
 	return 0;
 
diff --git a/drivers/base/dma-contiguous.c b/drivers/base/dma-contiguous.c
index 99802d6f3c60..165c2c299e57 100644
--- a/drivers/base/dma-contiguous.c
+++ b/drivers/base/dma-contiguous.c
@@ -49,7 +49,7 @@ struct cma *dma_contiguous_default_area;
 
 /*
  * Default global CMA area size can be defined in kernel's .config.
- * This is usefull mainly for distro maintainers to create a kernel
+ * This is useful mainly for distro maintainers to create a kernel
  * that works correctly for most supported systems.
  * The size can be set in bytes or as a percentage of the total memory
  * in the system.
diff --git a/drivers/base/power/main.c b/drivers/base/power/main.c
index 2a1b06ae0bdc..1b41fca3d65a 100644
--- a/drivers/base/power/main.c
+++ b/drivers/base/power/main.c
@@ -757,7 +757,7 @@ void dpm_resume(pm_message_t state)
 	async_error = 0;
 
 	list_for_each_entry(dev, &dpm_suspended_list, power.entry) {
-		INIT_COMPLETION(dev->power.completion);
+		reinit_completion(&dev->power.completion);
 		if (is_async(dev)) {
 			get_device(dev);
 			async_schedule(async_resume, dev);
@@ -1237,7 +1237,7 @@ static void async_suspend(void *data, async_cookie_t cookie)
 
 static int device_suspend(struct device *dev)
 {
-	INIT_COMPLETION(dev->power.completion);
+	reinit_completion(&dev->power.completion);
 
 	if (pm_async_enabled && dev->power.async_suspend) {
 		get_device(dev);
diff --git a/drivers/block/amiflop.c b/drivers/block/amiflop.c
index 4ff85b8785ee..748dea4f34dc 100644
--- a/drivers/block/amiflop.c
+++ b/drivers/block/amiflop.c
@@ -343,7 +343,7 @@ static int fd_motor_on(int nr)
 		unit[nr].motor = 1;
 		fd_select(nr);
 
-		INIT_COMPLETION(motor_on_completion);
+		reinit_completion(&motor_on_completion);
 		motor_on_timer.data = nr;
 		mod_timer(&motor_on_timer, jiffies + HZ/2);
 
diff --git a/drivers/block/cciss.c b/drivers/block/cciss.c
index 0c004ac05811..b35fc4f5237c 100644
--- a/drivers/block/cciss.c
+++ b/drivers/block/cciss.c
@@ -2808,7 +2808,7 @@ resend_cmd2:
 		/* erase the old error information */
 		memset(c->err_info, 0, sizeof(ErrorInfo_struct));
 		return_status = IO_OK;
-		INIT_COMPLETION(wait);
+		reinit_completion(&wait);
 		goto resend_cmd2;
 	}
 
@@ -3669,7 +3669,7 @@ static int add_to_scan_list(struct ctlr_info *h)
 		}
 	}
 	if (!found && !h->busy_scanning) {
-		INIT_COMPLETION(h->scan_wait);
+		reinit_completion(&h->scan_wait);
 		list_add_tail(&h->scan_list, &scan_q);
 		ret = 1;
 	}
diff --git a/drivers/block/virtio_blk.c b/drivers/block/virtio_blk.c
index 5cdf88b7ad9e..6a680d4de7f1 100644
--- a/drivers/block/virtio_blk.c
+++ b/drivers/block/virtio_blk.c
@@ -11,12 +11,11 @@
 #include <linux/string_helpers.h>
 #include <scsi/scsi_cmnd.h>
 #include <linux/idr.h>
+#include <linux/blk-mq.h>
+#include <linux/numa.h>
 
 #define PART_BITS 4
 
-static bool use_bio;
-module_param(use_bio, bool, S_IRUGO);
-
 static int major;
 static DEFINE_IDA(vd_index_ida);
 
@@ -26,13 +25,11 @@ struct virtio_blk
 {
 	struct virtio_device *vdev;
 	struct virtqueue *vq;
-	wait_queue_head_t queue_wait;
+	spinlock_t vq_lock;
 
 	/* The disk structure for the kernel. */
 	struct gendisk *disk;
 
-	mempool_t *pool;
-
 	/* Process context for config space updates */
 	struct work_struct config_work;
 
@@ -47,31 +44,17 @@ struct virtio_blk
 
 	/* Ida index - used to track minor number allocations. */
 	int index;
-
-	/* Scatterlist: can be too big for stack. */
-	struct scatterlist sg[/*sg_elems*/];
 };
 
 struct virtblk_req
 {
 	struct request *req;
-	struct bio *bio;
 	struct virtio_blk_outhdr out_hdr;
 	struct virtio_scsi_inhdr in_hdr;
-	struct work_struct work;
-	struct virtio_blk *vblk;
-	int flags;
 	u8 status;
 	struct scatterlist sg[];
 };
 
-enum {
-	VBLK_IS_FLUSH		= 1,
-	VBLK_REQ_FLUSH		= 2,
-	VBLK_REQ_DATA		= 4,
-	VBLK_REQ_FUA		= 8,
-};
-
 static inline int virtblk_result(struct virtblk_req *vbr)
 {
 	switch (vbr->status) {
@@ -84,22 +67,6 @@ static inline int virtblk_result(struct virtblk_req *vbr)
 	}
 }
 
-static inline struct virtblk_req *virtblk_alloc_req(struct virtio_blk *vblk,
-						    gfp_t gfp_mask)
-{
-	struct virtblk_req *vbr;
-
-	vbr = mempool_alloc(vblk->pool, gfp_mask);
-	if (!vbr)
-		return NULL;
-
-	vbr->vblk = vblk;
-	if (use_bio)
-		sg_init_table(vbr->sg, vblk->sg_elems);
-
-	return vbr;
-}
-
 static int __virtblk_add_req(struct virtqueue *vq,
 			     struct virtblk_req *vbr,
 			     struct scatterlist *data_sg,
@@ -143,83 +110,8 @@ static int __virtblk_add_req(struct virtqueue *vq,
 	return virtqueue_add_sgs(vq, sgs, num_out, num_in, vbr, GFP_ATOMIC);
 }
 
-static void virtblk_add_req(struct virtblk_req *vbr, bool have_data)
-{
-	struct virtio_blk *vblk = vbr->vblk;
-	DEFINE_WAIT(wait);
-	int ret;
-
-	spin_lock_irq(vblk->disk->queue->queue_lock);
-	while (unlikely((ret = __virtblk_add_req(vblk->vq, vbr, vbr->sg,
-						 have_data)) < 0)) {
-		prepare_to_wait_exclusive(&vblk->queue_wait, &wait,
-					  TASK_UNINTERRUPTIBLE);
-
-		spin_unlock_irq(vblk->disk->queue->queue_lock);
-		io_schedule();
-		spin_lock_irq(vblk->disk->queue->queue_lock);
-
-		finish_wait(&vblk->queue_wait, &wait);
-	}
-
-	virtqueue_kick(vblk->vq);
-	spin_unlock_irq(vblk->disk->queue->queue_lock);
-}
-
-static void virtblk_bio_send_flush(struct virtblk_req *vbr)
-{
-	vbr->flags |= VBLK_IS_FLUSH;
-	vbr->out_hdr.type = VIRTIO_BLK_T_FLUSH;
-	vbr->out_hdr.sector = 0;
-	vbr->out_hdr.ioprio = 0;
-
-	virtblk_add_req(vbr, false);
-}
-
-static void virtblk_bio_send_data(struct virtblk_req *vbr)
-{
-	struct virtio_blk *vblk = vbr->vblk;
-	struct bio *bio = vbr->bio;
-	bool have_data;
-
-	vbr->flags &= ~VBLK_IS_FLUSH;
-	vbr->out_hdr.type = 0;
-	vbr->out_hdr.sector = bio->bi_sector;
-	vbr->out_hdr.ioprio = bio_prio(bio);
-
-	if (blk_bio_map_sg(vblk->disk->queue, bio, vbr->sg)) {
-		have_data = true;
-		if (bio->bi_rw & REQ_WRITE)
-			vbr->out_hdr.type |= VIRTIO_BLK_T_OUT;
-		else
-			vbr->out_hdr.type |= VIRTIO_BLK_T_IN;
-	} else
-		have_data = false;
-
-	virtblk_add_req(vbr, have_data);
-}
-
-static void virtblk_bio_send_data_work(struct work_struct *work)
-{
-	struct virtblk_req *vbr;
-
-	vbr = container_of(work, struct virtblk_req, work);
-
-	virtblk_bio_send_data(vbr);
-}
-
-static void virtblk_bio_send_flush_work(struct work_struct *work)
-{
-	struct virtblk_req *vbr;
-
-	vbr = container_of(work, struct virtblk_req, work);
-
-	virtblk_bio_send_flush(vbr);
-}
-
 static inline void virtblk_request_done(struct virtblk_req *vbr)
 {
-	struct virtio_blk *vblk = vbr->vblk;
 	struct request *req = vbr->req;
 	int error = virtblk_result(vbr);
 
@@ -231,90 +123,45 @@ static inline void virtblk_request_done(struct virtblk_req *vbr)
 		req->errors = (error != 0);
 	}
 
-	__blk_end_request_all(req, error);
-	mempool_free(vbr, vblk->pool);
-}
-
-static inline void virtblk_bio_flush_done(struct virtblk_req *vbr)
-{
-	struct virtio_blk *vblk = vbr->vblk;
-
-	if (vbr->flags & VBLK_REQ_DATA) {
-		/* Send out the actual write data */
-		INIT_WORK(&vbr->work, virtblk_bio_send_data_work);
-		queue_work(virtblk_wq, &vbr->work);
-	} else {
-		bio_endio(vbr->bio, virtblk_result(vbr));
-		mempool_free(vbr, vblk->pool);
-	}
-}
-
-static inline void virtblk_bio_data_done(struct virtblk_req *vbr)
-{
-	struct virtio_blk *vblk = vbr->vblk;
-
-	if (unlikely(vbr->flags & VBLK_REQ_FUA)) {
-		/* Send out a flush before end the bio */
-		vbr->flags &= ~VBLK_REQ_DATA;
-		INIT_WORK(&vbr->work, virtblk_bio_send_flush_work);
-		queue_work(virtblk_wq, &vbr->work);
-	} else {
-		bio_endio(vbr->bio, virtblk_result(vbr));
-		mempool_free(vbr, vblk->pool);
-	}
-}
-
-static inline void virtblk_bio_done(struct virtblk_req *vbr)
-{
-	if (unlikely(vbr->flags & VBLK_IS_FLUSH))
-		virtblk_bio_flush_done(vbr);
-	else
-		virtblk_bio_data_done(vbr);
+	blk_mq_end_io(req, error);
 }
 
 static void virtblk_done(struct virtqueue *vq)
 {
 	struct virtio_blk *vblk = vq->vdev->priv;
-	bool bio_done = false, req_done = false;
+	bool req_done = false;
 	struct virtblk_req *vbr;
 	unsigned long flags;
 	unsigned int len;
 
-	spin_lock_irqsave(vblk->disk->queue->queue_lock, flags);
+	spin_lock_irqsave(&vblk->vq_lock, flags);
 	do {
 		virtqueue_disable_cb(vq);
 		while ((vbr = virtqueue_get_buf(vblk->vq, &len)) != NULL) {
-			if (vbr->bio) {
-				virtblk_bio_done(vbr);
-				bio_done = true;
-			} else {
-				virtblk_request_done(vbr);
-				req_done = true;
-			}
+			virtblk_request_done(vbr);
+			req_done = true;
 		}
+		if (unlikely(virtqueue_is_broken(vq)))
+			break;
 	} while (!virtqueue_enable_cb(vq));
+	spin_unlock_irqrestore(&vblk->vq_lock, flags);
+
 	/* In case queue is stopped waiting for more buffers. */
 	if (req_done)
-		blk_start_queue(vblk->disk->queue);
-	spin_unlock_irqrestore(vblk->disk->queue->queue_lock, flags);
-
-	if (bio_done)
-		wake_up(&vblk->queue_wait);
+		blk_mq_start_stopped_hw_queues(vblk->disk->queue);
 }
 
-static bool do_req(struct request_queue *q, struct virtio_blk *vblk,
-		   struct request *req)
+static int virtio_queue_rq(struct blk_mq_hw_ctx *hctx, struct request *req)
 {
+	struct virtio_blk *vblk = hctx->queue->queuedata;
+	struct virtblk_req *vbr = req->special;
+	unsigned long flags;
 	unsigned int num;
-	struct virtblk_req *vbr;
+	const bool last = (req->cmd_flags & REQ_END) != 0;
 
-	vbr = virtblk_alloc_req(vblk, GFP_ATOMIC);
-	if (!vbr)
-		/* When another request finishes we'll try again. */
-		return false;
+	BUG_ON(req->nr_phys_segments + 2 > vblk->sg_elems);
 
 	vbr->req = req;
-	vbr->bio = NULL;
 	if (req->cmd_flags & REQ_FLUSH) {
 		vbr->out_hdr.type = VIRTIO_BLK_T_FLUSH;
 		vbr->out_hdr.sector = 0;
@@ -342,7 +189,7 @@ static bool do_req(struct request_queue *q, struct virtio_blk *vblk,
 		}
 	}
 
-	num = blk_rq_map_sg(q, vbr->req, vblk->sg);
+	num = blk_rq_map_sg(hctx->queue, vbr->req, vbr->sg);
 	if (num) {
 		if (rq_data_dir(vbr->req) == WRITE)
 			vbr->out_hdr.type |= VIRTIO_BLK_T_OUT;
@@ -350,63 +197,19 @@ static bool do_req(struct request_queue *q, struct virtio_blk *vblk,
 			vbr->out_hdr.type |= VIRTIO_BLK_T_IN;
 	}
 
-	if (__virtblk_add_req(vblk->vq, vbr, vblk->sg, num) < 0) {
-		mempool_free(vbr, vblk->pool);
-		return false;
-	}
-
-	return true;
-}
-
-static void virtblk_request(struct request_queue *q)
-{
-	struct virtio_blk *vblk = q->queuedata;
-	struct request *req;
-	unsigned int issued = 0;
-
-	while ((req = blk_peek_request(q)) != NULL) {
-		BUG_ON(req->nr_phys_segments + 2 > vblk->sg_elems);
-
-		/* If this request fails, stop queue and wait for something to
-		   finish to restart it. */
-		if (!do_req(q, vblk, req)) {
-			blk_stop_queue(q);
-			break;
-		}
-		blk_start_request(req);
-		issued++;
+	spin_lock_irqsave(&vblk->vq_lock, flags);
+	if (__virtblk_add_req(vblk->vq, vbr, vbr->sg, num) < 0) {
+		virtqueue_kick(vblk->vq);
+		spin_unlock_irqrestore(&vblk->vq_lock, flags);
+		blk_mq_stop_hw_queue(hctx);
+		return BLK_MQ_RQ_QUEUE_BUSY;
 	}
 
-	if (issued)
+	if (last)
 		virtqueue_kick(vblk->vq);
-}
 
-static void virtblk_make_request(struct request_queue *q, struct bio *bio)
-{
-	struct virtio_blk *vblk = q->queuedata;
-	struct virtblk_req *vbr;
-
-	BUG_ON(bio->bi_phys_segments + 2 > vblk->sg_elems);
-
-	vbr = virtblk_alloc_req(vblk, GFP_NOIO);
-	if (!vbr) {
-		bio_endio(bio, -ENOMEM);
-		return;
-	}
-
-	vbr->bio = bio;
-	vbr->flags = 0;
-	if (bio->bi_rw & REQ_FLUSH)
-		vbr->flags |= VBLK_REQ_FLUSH;
-	if (bio->bi_rw & REQ_FUA)
-		vbr->flags |= VBLK_REQ_FUA;
-	if (bio->bi_size)
-		vbr->flags |= VBLK_REQ_DATA;
-
-	if (unlikely(vbr->flags & VBLK_REQ_FLUSH))
-		virtblk_bio_send_flush(vbr);
-	else
-		virtblk_bio_send_data(vbr);
+	spin_unlock_irqrestore(&vblk->vq_lock, flags);
+	return BLK_MQ_RQ_QUEUE_OK;
 }
 
 /* return id (s/n) string for *disk to *id_str
@@ -456,18 +259,15 @@ static int virtblk_ioctl(struct block_device *bdev, fmode_t mode,
 static int virtblk_getgeo(struct block_device *bd, struct hd_geometry *geo)
 {
 	struct virtio_blk *vblk = bd->bd_disk->private_data;
-	struct virtio_blk_geometry vgeo;
-	int err;
 
 	/* see if the host passed in geometry config */
-	err = virtio_config_val(vblk->vdev, VIRTIO_BLK_F_GEOMETRY,
-				offsetof(struct virtio_blk_config, geometry),
-				&vgeo);
-
-	if (!err) {
-		geo->heads = vgeo.heads;
-		geo->sectors = vgeo.sectors;
-		geo->cylinders = vgeo.cylinders;
+	if (virtio_has_feature(vblk->vdev, VIRTIO_BLK_F_GEOMETRY)) {
+		virtio_cread(vblk->vdev, struct virtio_blk_config,
+			     geometry.cylinders, &geo->cylinders);
+		virtio_cread(vblk->vdev, struct virtio_blk_config,
+			     geometry.heads, &geo->heads);
+		virtio_cread(vblk->vdev, struct virtio_blk_config,
+			     geometry.sectors, &geo->sectors);
 	} else {
 		/* some standard values, similar to sd */
 		geo->heads = 1 << 6;
@@ -529,8 +329,7 @@ static void virtblk_config_changed_work(struct work_struct *work)
 		goto done;
 
 	/* Host must always specify the capacity. */
-	vdev->config->get(vdev, offsetof(struct virtio_blk_config, capacity),
-			  &capacity, sizeof(capacity));
+	virtio_cread(vdev, struct virtio_blk_config, capacity, &capacity);
 
 	/* If capacity is too big, truncate with warning. */
 	if ((sector_t)capacity != capacity) {
@@ -608,9 +407,9 @@ static int virtblk_get_cache_mode(struct virtio_device *vdev)
 	u8 writeback;
 	int err;
 
-	err = virtio_config_val(vdev, VIRTIO_BLK_F_CONFIG_WCE,
-				offsetof(struct virtio_blk_config, wce),
-				&writeback);
+	err = virtio_cread_feature(vdev, VIRTIO_BLK_F_CONFIG_WCE,
+				   struct virtio_blk_config, wce,
+				   &writeback);
 	if (err)
 		writeback = virtio_has_feature(vdev, VIRTIO_BLK_F_WCE);
 
@@ -642,7 +441,6 @@ virtblk_cache_type_store(struct device *dev, struct device_attribute *attr,
 	struct virtio_blk *vblk = disk->private_data;
 	struct virtio_device *vdev = vblk->vdev;
 	int i;
-	u8 writeback;
 
 	BUG_ON(!virtio_has_feature(vblk->vdev, VIRTIO_BLK_F_CONFIG_WCE));
 	for (i = ARRAY_SIZE(virtblk_cache_types); --i >= 0; )
@@ -652,11 +450,7 @@ virtblk_cache_type_store(struct device *dev, struct device_attribute *attr,
 	if (i < 0)
 		return -EINVAL;
 
-	writeback = i;
-	vdev->config->set(vdev,
-			  offsetof(struct virtio_blk_config, wce),
-			  &writeback, sizeof(writeback));
-
+	virtio_cwrite8(vdev, offsetof(struct virtio_blk_config, wce), i);
 	virtblk_update_cache_mode(vdev);
 	return count;
 }
@@ -680,12 +474,35 @@ static const struct device_attribute dev_attr_cache_type_rw =
 	__ATTR(cache_type, S_IRUGO|S_IWUSR,
 	       virtblk_cache_type_show, virtblk_cache_type_store);
 
+static struct blk_mq_ops virtio_mq_ops = {
+	.queue_rq	= virtio_queue_rq,
+	.map_queue	= blk_mq_map_queue,
+	.alloc_hctx	= blk_mq_alloc_single_hw_queue,
+	.free_hctx	= blk_mq_free_single_hw_queue,
+};
+
+static struct blk_mq_reg virtio_mq_reg = {
+	.ops		= &virtio_mq_ops,
+	.nr_hw_queues	= 1,
+	.queue_depth	= 64,
+	.numa_node	= NUMA_NO_NODE,
+	.flags		= BLK_MQ_F_SHOULD_MERGE,
+};
+
+static void virtblk_init_vbr(void *data, struct blk_mq_hw_ctx *hctx,
+			     struct request *rq, unsigned int nr)
+{
+	struct virtio_blk *vblk = data;
+	struct virtblk_req *vbr = rq->special;
+
+	sg_init_table(vbr->sg, vblk->sg_elems);
+}
+
 static int virtblk_probe(struct virtio_device *vdev)
 {
 	struct virtio_blk *vblk;
 	struct request_queue *q;
 	int err, index;
-	int pool_size;
 
 	u64 cap;
 	u32 v, blk_size, sg_elems, opt_io_size;
@@ -699,9 +516,9 @@ static int virtblk_probe(struct virtio_device *vdev)
 	index = err;
 
 	/* We need to know how many segments before we allocate. */
-	err = virtio_config_val(vdev, VIRTIO_BLK_F_SEG_MAX,
-				offsetof(struct virtio_blk_config, seg_max),
-				&sg_elems);
+	err = virtio_cread_feature(vdev, VIRTIO_BLK_F_SEG_MAX,
+				   struct virtio_blk_config, seg_max,
+				   &sg_elems);
 
 	/* We need at least one SG element, whatever they say. */
 	if (err || !sg_elems)
@@ -709,17 +526,14 @@ static int virtblk_probe(struct virtio_device *vdev)
 
 	/* We need an extra sg elements at head and tail. */
 	sg_elems += 2;
-	vdev->priv = vblk = kmalloc(sizeof(*vblk) +
-				    sizeof(vblk->sg[0]) * sg_elems, GFP_KERNEL);
+	vdev->priv = vblk = kmalloc(sizeof(*vblk), GFP_KERNEL);
 	if (!vblk) {
 		err = -ENOMEM;
 		goto out_free_index;
 	}
 
-	init_waitqueue_head(&vblk->queue_wait);
 	vblk->vdev = vdev;
 	vblk->sg_elems = sg_elems;
-	sg_init_table(vblk->sg, vblk->sg_elems);
 	mutex_init(&vblk->config_lock);
 
 	INIT_WORK(&vblk->config_work, virtblk_config_changed_work);
@@ -728,31 +542,27 @@ static int virtblk_probe(struct virtio_device *vdev)
 	err = init_vq(vblk);
 	if (err)
 		goto out_free_vblk;
-
-	pool_size = sizeof(struct virtblk_req);
-	if (use_bio)
-		pool_size += sizeof(struct scatterlist) * sg_elems;
-	vblk->pool = mempool_create_kmalloc_pool(1, pool_size);
-	if (!vblk->pool) {
-		err = -ENOMEM;
-		goto out_free_vq;
-	}
+	spin_lock_init(&vblk->vq_lock);
 
 	/* FIXME: How many partitions?  How long is a piece of string? */
 	vblk->disk = alloc_disk(1 << PART_BITS);
 	if (!vblk->disk) {
 		err = -ENOMEM;
-		goto out_mempool;
+		goto out_free_vq;
 	}
 
-	q = vblk->disk->queue = blk_init_queue(virtblk_request, NULL);
+	virtio_mq_reg.cmd_size =
+		sizeof(struct virtblk_req) +
+		sizeof(struct scatterlist) * sg_elems;
+
+	q = vblk->disk->queue = blk_mq_init_queue(&virtio_mq_reg, vblk);
 	if (!q) {
 		err = -ENOMEM;
 		goto out_put_disk;
 	}
 
-	if (use_bio)
-		blk_queue_make_request(q, virtblk_make_request);
+	blk_mq_init_commands(q, virtblk_init_vbr, vblk);
+
 	q->queuedata = vblk;
 
 	virtblk_name_format("vd", index, vblk->disk->disk_name, DISK_NAME_LEN);
@@ -772,8 +582,7 @@ static int virtblk_probe(struct virtio_device *vdev)
 		set_disk_ro(vblk->disk, 1);
 
 	/* Host must always specify the capacity. */
-	vdev->config->get(vdev, offsetof(struct virtio_blk_config, capacity),
-			  &cap, sizeof(cap));
+	virtio_cread(vdev, struct virtio_blk_config, capacity, &cap);
 
 	/* If capacity is too big, truncate with warning. */
 	if ((sector_t)cap != cap) {
@@ -794,46 +603,45 @@ static int virtblk_probe(struct virtio_device *vdev)
 
 	/* Host can optionally specify maximum segment size and number of
 	 * segments. */
-	err = virtio_config_val(vdev, VIRTIO_BLK_F_SIZE_MAX,
-				offsetof(struct virtio_blk_config, size_max),
-				&v);
+	err = virtio_cread_feature(vdev, VIRTIO_BLK_F_SIZE_MAX,
+				   struct virtio_blk_config, size_max, &v);
 	if (!err)
 		blk_queue_max_segment_size(q, v);
 	else
 		blk_queue_max_segment_size(q, -1U);
 
 	/* Host can optionally specify the block size of the device */
-	err = virtio_config_val(vdev, VIRTIO_BLK_F_BLK_SIZE,
-				offsetof(struct virtio_blk_config, blk_size),
-				&blk_size);
+	err = virtio_cread_feature(vdev, VIRTIO_BLK_F_BLK_SIZE,
+				   struct virtio_blk_config, blk_size,
+				   &blk_size);
 	if (!err)
 		blk_queue_logical_block_size(q, blk_size);
 	else
 		blk_size = queue_logical_block_size(q);
 
 	/* Use topology information if available */
-	err = virtio_config_val(vdev, VIRTIO_BLK_F_TOPOLOGY,
-			offsetof(struct virtio_blk_config, physical_block_exp),
-			&physical_block_exp);
+	err = virtio_cread_feature(vdev, VIRTIO_BLK_F_TOPOLOGY,
+				   struct virtio_blk_config, physical_block_exp,
+				   &physical_block_exp);
 	if (!err && physical_block_exp)
 		blk_queue_physical_block_size(q,
 				blk_size * (1 << physical_block_exp));
 
-	err = virtio_config_val(vdev, VIRTIO_BLK_F_TOPOLOGY,
-			offsetof(struct virtio_blk_config, alignment_offset),
-			&alignment_offset);
+	err = virtio_cread_feature(vdev, VIRTIO_BLK_F_TOPOLOGY,
+				   struct virtio_blk_config, alignment_offset,
+				   &alignment_offset);
 	if (!err && alignment_offset)
 		blk_queue_alignment_offset(q, blk_size * alignment_offset);
 
-	err = virtio_config_val(vdev, VIRTIO_BLK_F_TOPOLOGY,
-			offsetof(struct virtio_blk_config, min_io_size),
-			&min_io_size);
+	err = virtio_cread_feature(vdev, VIRTIO_BLK_F_TOPOLOGY,
+				   struct virtio_blk_config, min_io_size,
+				   &min_io_size);
 	if (!err && min_io_size)
 		blk_queue_io_min(q, blk_size * min_io_size);
 
-	err = virtio_config_val(vdev, VIRTIO_BLK_F_TOPOLOGY,
-			offsetof(struct virtio_blk_config, opt_io_size),
-			&opt_io_size);
+	err = virtio_cread_feature(vdev, VIRTIO_BLK_F_TOPOLOGY,
+				   struct virtio_blk_config, opt_io_size,
+				   &opt_io_size);
 	if (!err && opt_io_size)
 		blk_queue_io_opt(q, blk_size * opt_io_size);
 
@@ -857,8 +665,6 @@ out_del_disk:
 	blk_cleanup_queue(vblk->disk->queue);
 out_put_disk:
 	put_disk(vblk->disk);
-out_mempool:
-	mempool_destroy(vblk->pool);
 out_free_vq:
 	vdev->config->del_vqs(vdev);
 out_free_vblk:
@@ -890,7 +696,6 @@ static void virtblk_remove(struct virtio_device *vdev)
 
 	refc = atomic_read(&disk_to_dev(vblk->disk)->kobj.kref.refcount);
 	put_disk(vblk->disk);
-	mempool_destroy(vblk->pool);
 	vdev->config->del_vqs(vdev);
 	kfree(vblk);
 
@@ -899,7 +704,7 @@ static void virtblk_remove(struct virtio_device *vdev)
 		ida_simple_remove(&vd_index_ida, index);
 }
 
-#ifdef CONFIG_PM
+#ifdef CONFIG_PM_SLEEP
 static int virtblk_freeze(struct virtio_device *vdev)
 {
 	struct virtio_blk *vblk = vdev->priv;
@@ -914,10 +719,7 @@ static int virtblk_freeze(struct virtio_device *vdev)
 
 	flush_work(&vblk->config_work);
 
-	spin_lock_irq(vblk->disk->queue->queue_lock);
-	blk_stop_queue(vblk->disk->queue);
-	spin_unlock_irq(vblk->disk->queue->queue_lock);
-	blk_sync_queue(vblk->disk->queue);
+	blk_mq_stop_hw_queues(vblk->disk->queue);
 
 	vdev->config->del_vqs(vdev);
 	return 0;
@@ -930,11 +732,9 @@ static int virtblk_restore(struct virtio_device *vdev)
 
 	vblk->config_enable = true;
 	ret = init_vq(vdev->priv);
-	if (!ret) {
-		spin_lock_irq(vblk->disk->queue->queue_lock);
-		blk_start_queue(vblk->disk->queue);
-		spin_unlock_irq(vblk->disk->queue->queue_lock);
-	}
+	if (!ret)
+		blk_mq_start_stopped_hw_queues(vblk->disk->queue);
+
 	return ret;
 }
 #endif
@@ -959,7 +759,7 @@ static struct virtio_driver virtio_blk = {
 	.probe			= virtblk_probe,
 	.remove			= virtblk_remove,
 	.config_changed		= virtblk_config_changed,
-#ifdef CONFIG_PM
+#ifdef CONFIG_PM_SLEEP
 	.freeze			= virtblk_freeze,
 	.restore		= virtblk_restore,
 #endif
diff --git a/drivers/char/hw_random/timeriomem-rng.c b/drivers/char/hw_random/timeriomem-rng.c
index d2120ba8f3f9..73ce739f8e19 100644
--- a/drivers/char/hw_random/timeriomem-rng.c
+++ b/drivers/char/hw_random/timeriomem-rng.c
@@ -79,7 +79,7 @@ static int timeriomem_rng_data_read(struct hwrng *rng, u32 *data)
 	priv->expires = cur + delay;
 	priv->present = 0;
 
-	INIT_COMPLETION(priv->completion);
+	reinit_completion(&priv->completion);
 	mod_timer(&priv->timer, priv->expires);
 
 	return 4;
diff --git a/drivers/char/hw_random/virtio-rng.c b/drivers/char/hw_random/virtio-rng.c
index ef46a9cfd832..c12398d1517c 100644
--- a/drivers/char/hw_random/virtio-rng.c
+++ b/drivers/char/hw_random/virtio-rng.c
@@ -133,7 +133,7 @@ static void virtrng_remove(struct virtio_device *vdev)
 	remove_common(vdev);
 }
 
-#ifdef CONFIG_PM
+#ifdef CONFIG_PM_SLEEP
 static int virtrng_freeze(struct virtio_device *vdev)
 {
 	remove_common(vdev);
@@ -157,7 +157,7 @@ static struct virtio_driver virtio_rng_driver = {
 	.id_table =	id_table,
 	.probe =	virtrng_probe,
 	.remove =	virtrng_remove,
-#ifdef CONFIG_PM
+#ifdef CONFIG_PM_SLEEP
 	.freeze =	virtrng_freeze,
 	.restore =	virtrng_restore,
 #endif
diff --git a/drivers/char/random.c b/drivers/char/random.c
index 4fe5609eeb72..429b75bb60e8 100644
--- a/drivers/char/random.c
+++ b/drivers/char/random.c
@@ -255,6 +255,7 @@
 #include <linux/fips.h>
 #include <linux/ptrace.h>
 #include <linux/kmemcheck.h>
+#include <linux/workqueue.h>
 #include <linux/irq.h>
 
 #include <asm/processor.h>
@@ -269,14 +270,28 @@
 /*
  * Configuration information
  */
-#define INPUT_POOL_WORDS 128
-#define OUTPUT_POOL_WORDS 32
-#define SEC_XFER_SIZE 512
-#define EXTRACT_SIZE 10
+#define INPUT_POOL_SHIFT	12
+#define INPUT_POOL_WORDS	(1 << (INPUT_POOL_SHIFT-5))
+#define OUTPUT_POOL_SHIFT	10
+#define OUTPUT_POOL_WORDS	(1 << (OUTPUT_POOL_SHIFT-5))
+#define SEC_XFER_SIZE		512
+#define EXTRACT_SIZE		10
+
+#define DEBUG_RANDOM_BOOT 0
 
 #define LONGS(x) (((x) + sizeof(unsigned long) - 1)/sizeof(unsigned long))
 
 /*
+ * To allow fractional bits to be tracked, the entropy_count field is
+ * denominated in units of 1/8th bits.
+ *
+ * 2*(ENTROPY_SHIFT + log2(poolbits)) must <= 31, or the multiply in
+ * credit_entropy_bits() needs to be 64 bits wide.
+ */
+#define ENTROPY_SHIFT 3
+#define ENTROPY_BITS(r) ((r)->entropy_count >> ENTROPY_SHIFT)
+
+/*
  * The minimum number of bits of entropy before we wake up a read on
  * /dev/random.  Should be enough to do a significant reseed.
  */
@@ -287,108 +302,100 @@ static int random_read_wakeup_thresh = 64;
  * should wake up processes which are selecting or polling on write
  * access to /dev/random.
  */
-static int random_write_wakeup_thresh = 128;
+static int random_write_wakeup_thresh = 28 * OUTPUT_POOL_WORDS;
 
 /*
- * When the input pool goes over trickle_thresh, start dropping most
- * samples to avoid wasting CPU time and reduce lock contention.
+ * The minimum number of seconds between urandom pool resending.  We
+ * do this to limit the amount of entropy that can be drained from the
+ * input pool even if there are heavy demands on /dev/urandom.
  */
-
-static int trickle_thresh __read_mostly = INPUT_POOL_WORDS * 28;
-
-static DEFINE_PER_CPU(int, trickle_count);
+static int random_min_urandom_seed = 60;
 
 /*
- * A pool of size .poolwords is stirred with a primitive polynomial
- * of degree .poolwords over GF(2).  The taps for various sizes are
- * defined below.  They are chosen to be evenly spaced (minimum RMS
- * distance from evenly spaced; the numbers in the comments are a
- * scaled squared error sum) except for the last tap, which is 1 to
- * get the twisting happening as fast as possible.
+ * Originally, we used a primitive polynomial of degree .poolwords
+ * over GF(2).  The taps for various sizes are defined below.  They
+ * were chosen to be evenly spaced except for the last tap, which is 1
+ * to get the twisting happening as fast as possible.
+ *
+ * For the purposes of better mixing, we use the CRC-32 polynomial as
+ * well to make a (modified) twisted Generalized Feedback Shift
+ * Register.  (See M. Matsumoto & Y. Kurita, 1992.  Twisted GFSR
+ * generators.  ACM Transactions on Modeling and Computer Simulation
+ * 2(3):179-194.  Also see M. Matsumoto & Y. Kurita, 1994.  Twisted
+ * GFSR generators II.  ACM Transactions on Mdeling and Computer
+ * Simulation 4:254-266)
+ *
+ * Thanks to Colin Plumb for suggesting this.
+ *
+ * The mixing operation is much less sensitive than the output hash,
+ * where we use SHA-1.  All that we want of mixing operation is that
+ * it be a good non-cryptographic hash; i.e. it not produce collisions
+ * when fed "random" data of the sort we expect to see.  As long as
+ * the pool state differs for different inputs, we have preserved the
+ * input entropy and done a good job.  The fact that an intelligent
+ * attacker can construct inputs that will produce controlled
+ * alterations to the pool's state is not important because we don't
+ * consider such inputs to contribute any randomness.  The only
+ * property we need with respect to them is that the attacker can't
+ * increase his/her knowledge of the pool's state.  Since all
+ * additions are reversible (knowing the final state and the input,
+ * you can reconstruct the initial state), if an attacker has any
+ * uncertainty about the initial state, he/she can only shuffle that
+ * uncertainty about, but never cause any collisions (which would
+ * decrease the uncertainty).
+ *
+ * Our mixing functions were analyzed by Lacharme, Roeck, Strubel, and
+ * Videau in their paper, "The Linux Pseudorandom Number Generator
+ * Revisited" (see: http://eprint.iacr.org/2012/251.pdf).  In their
+ * paper, they point out that we are not using a true Twisted GFSR,
+ * since Matsumoto & Kurita used a trinomial feedback polynomial (that
+ * is, with only three taps, instead of the six that we are using).
+ * As a result, the resulting polynomial is neither primitive nor
+ * irreducible, and hence does not have a maximal period over
+ * GF(2**32).  They suggest a slight change to the generator
+ * polynomial which improves the resulting TGFSR polynomial to be
+ * irreducible, which we have made here.
  */
 static struct poolinfo {
-	int poolwords;
+	int poolbitshift, poolwords, poolbytes, poolbits, poolfracbits;
+#define S(x) ilog2(x)+5, (x), (x)*4, (x)*32, (x) << (ENTROPY_SHIFT+5)
 	int tap1, tap2, tap3, tap4, tap5;
 } poolinfo_table[] = {
-	/* x^128 + x^103 + x^76 + x^51 +x^25 + x + 1 -- 105 */
-	{ 128,	103,	76,	51,	25,	1 },
-	/* x^32 + x^26 + x^20 + x^14 + x^7 + x + 1 -- 15 */
-	{ 32,	26,	20,	14,	7,	1 },
+	/* was: x^128 + x^103 + x^76 + x^51 +x^25 + x + 1 */
+	/* x^128 + x^104 + x^76 + x^51 +x^25 + x + 1 */
+	{ S(128),	104,	76,	51,	25,	1 },
+	/* was: x^32 + x^26 + x^20 + x^14 + x^7 + x + 1 */
+	/* x^32 + x^26 + x^19 + x^14 + x^7 + x + 1 */
+	{ S(32),	26,	19,	14,	7,	1 },
 #if 0
 	/* x^2048 + x^1638 + x^1231 + x^819 + x^411 + x + 1  -- 115 */
-	{ 2048,	1638,	1231,	819,	411,	1 },
+	{ S(2048),	1638,	1231,	819,	411,	1 },
 
 	/* x^1024 + x^817 + x^615 + x^412 + x^204 + x + 1 -- 290 */
-	{ 1024,	817,	615,	412,	204,	1 },
+	{ S(1024),	817,	615,	412,	204,	1 },
 
 	/* x^1024 + x^819 + x^616 + x^410 + x^207 + x^2 + 1 -- 115 */
-	{ 1024,	819,	616,	410,	207,	2 },
+	{ S(1024),	819,	616,	410,	207,	2 },
 
 	/* x^512 + x^411 + x^308 + x^208 + x^104 + x + 1 -- 225 */
-	{ 512,	411,	308,	208,	104,	1 },
+	{ S(512),	411,	308,	208,	104,	1 },
 
 	/* x^512 + x^409 + x^307 + x^206 + x^102 + x^2 + 1 -- 95 */
-	{ 512,	409,	307,	206,	102,	2 },
+	{ S(512),	409,	307,	206,	102,	2 },
 	/* x^512 + x^409 + x^309 + x^205 + x^103 + x^2 + 1 -- 95 */
-	{ 512,	409,	309,	205,	103,	2 },
+	{ S(512),	409,	309,	205,	103,	2 },
 
 	/* x^256 + x^205 + x^155 + x^101 + x^52 + x + 1 -- 125 */
-	{ 256,	205,	155,	101,	52,	1 },
+	{ S(256),	205,	155,	101,	52,	1 },
 
 	/* x^128 + x^103 + x^78 + x^51 + x^27 + x^2 + 1 -- 70 */
-	{ 128,	103,	78,	51,	27,	2 },
+	{ S(128),	103,	78,	51,	27,	2 },
 
 	/* x^64 + x^52 + x^39 + x^26 + x^14 + x + 1 -- 15 */
-	{ 64,	52,	39,	26,	14,	1 },
+	{ S(64),	52,	39,	26,	14,	1 },
 #endif
 };
 
-#define POOLBITS	poolwords*32
-#define POOLBYTES	poolwords*4
-
-/*
- * For the purposes of better mixing, we use the CRC-32 polynomial as
- * well to make a twisted Generalized Feedback Shift Reigster
- *
- * (See M. Matsumoto & Y. Kurita, 1992.  Twisted GFSR generators.  ACM
- * Transactions on Modeling and Computer Simulation 2(3):179-194.
- * Also see M. Matsumoto & Y. Kurita, 1994.  Twisted GFSR generators
- * II.  ACM Transactions on Mdeling and Computer Simulation 4:254-266)
- *
- * Thanks to Colin Plumb for suggesting this.
- *
- * We have not analyzed the resultant polynomial to prove it primitive;
- * in fact it almost certainly isn't.  Nonetheless, the irreducible factors
- * of a random large-degree polynomial over GF(2) are more than large enough
- * that periodicity is not a concern.
- *
- * The input hash is much less sensitive than the output hash.  All
- * that we want of it is that it be a good non-cryptographic hash;
- * i.e. it not produce collisions when fed "random" data of the sort
- * we expect to see.  As long as the pool state differs for different
- * inputs, we have preserved the input entropy and done a good job.
- * The fact that an intelligent attacker can construct inputs that
- * will produce controlled alterations to the pool's state is not
- * important because we don't consider such inputs to contribute any
- * randomness.  The only property we need with respect to them is that
- * the attacker can't increase his/her knowledge of the pool's state.
- * Since all additions are reversible (knowing the final state and the
- * input, you can reconstruct the initial state), if an attacker has
- * any uncertainty about the initial state, he/she can only shuffle
- * that uncertainty about, but never cause any collisions (which would
- * decrease the uncertainty).
- *
- * The chosen system lets the state of the pool be (essentially) the input
- * modulo the generator polymnomial.  Now, for random primitive polynomials,
- * this is a universal class of hash functions, meaning that the chance
- * of a collision is limited by the attacker's knowledge of the generator
- * polynomail, so if it is chosen at random, an attacker can never force
- * a collision.  Here, we use a fixed polynomial, but we *can* assume that
- * ###--> it is unknown to the processes generating the input entropy. <-###
- * Because of this important property, this is a good, collision-resistant
- * hash; hash collisions will occur no more often than chance.
- */
-
 /*
  * Static global variables
  */
@@ -396,17 +403,6 @@ static DECLARE_WAIT_QUEUE_HEAD(random_read_wait);
 static DECLARE_WAIT_QUEUE_HEAD(random_write_wait);
 static struct fasync_struct *fasync;
 
-static bool debug;
-module_param(debug, bool, 0644);
-#define DEBUG_ENT(fmt, arg...) do { \
-	if (debug) \
-		printk(KERN_DEBUG "random %04d %04d %04d: " \
-		fmt,\
-		input_pool.entropy_count,\
-		blocking_pool.entropy_count,\
-		nonblocking_pool.entropy_count,\
-		## arg); } while (0)
-
 /**********************************************************************
  *
  * OS independent entropy store.   Here are the functions which handle
@@ -417,23 +413,26 @@ module_param(debug, bool, 0644);
 struct entropy_store;
 struct entropy_store {
 	/* read-only data: */
-	struct poolinfo *poolinfo;
+	const struct poolinfo *poolinfo;
 	__u32 *pool;
 	const char *name;
 	struct entropy_store *pull;
-	int limit;
+	struct work_struct push_work;
 
 	/* read-write data: */
+	unsigned long last_pulled;
 	spinlock_t lock;
-	unsigned add_ptr;
-	unsigned input_rotate;
+	unsigned short add_ptr;
+	unsigned short input_rotate;
 	int entropy_count;
 	int entropy_total;
 	unsigned int initialized:1;
-	bool last_data_init;
+	unsigned int limit:1;
+	unsigned int last_data_init:1;
 	__u8 last_data[EXTRACT_SIZE];
 };
 
+static void push_to_pool(struct work_struct *work);
 static __u32 input_pool_data[INPUT_POOL_WORDS];
 static __u32 blocking_pool_data[OUTPUT_POOL_WORDS];
 static __u32 nonblocking_pool_data[OUTPUT_POOL_WORDS];
@@ -452,7 +451,9 @@ static struct entropy_store blocking_pool = {
 	.limit = 1,
 	.pull = &input_pool,
 	.lock = __SPIN_LOCK_UNLOCKED(blocking_pool.lock),
-	.pool = blocking_pool_data
+	.pool = blocking_pool_data,
+	.push_work = __WORK_INITIALIZER(blocking_pool.push_work,
+					push_to_pool),
 };
 
 static struct entropy_store nonblocking_pool = {
@@ -460,7 +461,9 @@ static struct entropy_store nonblocking_pool = {
 	.name = "nonblocking",
 	.pull = &input_pool,
 	.lock = __SPIN_LOCK_UNLOCKED(nonblocking_pool.lock),
-	.pool = nonblocking_pool_data
+	.pool = nonblocking_pool_data,
+	.push_work = __WORK_INITIALIZER(nonblocking_pool.push_work,
+					push_to_pool),
 };
 
 static __u32 const twist_table[8] = {
@@ -498,7 +501,7 @@ static void _mix_pool_bytes(struct entropy_store *r, const void *in,
 
 	/* mix one byte at a time to simplify size handling and churn faster */
 	while (nbytes--) {
-		w = rol32(*bytes++, input_rotate & 31);
+		w = rol32(*bytes++, input_rotate);
 		i = (i - 1) & wordmask;
 
 		/* XOR in the various taps */
@@ -518,7 +521,7 @@ static void _mix_pool_bytes(struct entropy_store *r, const void *in,
 		 * rotation, so that successive passes spread the
 		 * input bits across the pool evenly.
 		 */
-		input_rotate += i ? 7 : 14;
+		input_rotate = (input_rotate + (i ? 7 : 14)) & 31;
 	}
 
 	ACCESS_ONCE(r->input_rotate) = input_rotate;
@@ -561,65 +564,151 @@ struct fast_pool {
  * collector.  It's hardcoded for an 128 bit pool and assumes that any
  * locks that might be needed are taken by the caller.
  */
-static void fast_mix(struct fast_pool *f, const void *in, int nbytes)
+static void fast_mix(struct fast_pool *f, __u32 input[4])
 {
-	const char	*bytes = in;
 	__u32		w;
-	unsigned	i = f->count;
 	unsigned	input_rotate = f->rotate;
 
-	while (nbytes--) {
-		w = rol32(*bytes++, input_rotate & 31) ^ f->pool[i & 3] ^
-			f->pool[(i + 1) & 3];
-		f->pool[i & 3] = (w >> 3) ^ twist_table[w & 7];
-		input_rotate += (i++ & 3) ? 7 : 14;
-	}
-	f->count = i;
+	w = rol32(input[0], input_rotate) ^ f->pool[0] ^ f->pool[3];
+	f->pool[0] = (w >> 3) ^ twist_table[w & 7];
+	input_rotate = (input_rotate + 14) & 31;
+	w = rol32(input[1], input_rotate) ^ f->pool[1] ^ f->pool[0];
+	f->pool[1] = (w >> 3) ^ twist_table[w & 7];
+	input_rotate = (input_rotate + 7) & 31;
+	w = rol32(input[2], input_rotate) ^ f->pool[2] ^ f->pool[1];
+	f->pool[2] = (w >> 3) ^ twist_table[w & 7];
+	input_rotate = (input_rotate + 7) & 31;
+	w = rol32(input[3], input_rotate) ^ f->pool[3] ^ f->pool[2];
+	f->pool[3] = (w >> 3) ^ twist_table[w & 7];
+	input_rotate = (input_rotate + 7) & 31;
+
 	f->rotate = input_rotate;
+	f->count++;
 }
 
 /*
- * Credit (or debit) the entropy store with n bits of entropy
+ * Credit (or debit) the entropy store with n bits of entropy.
+ * Use credit_entropy_bits_safe() if the value comes from userspace
+ * or otherwise should be checked for extreme values.
  */
 static void credit_entropy_bits(struct entropy_store *r, int nbits)
 {
 	int entropy_count, orig;
+	const int pool_size = r->poolinfo->poolfracbits;
+	int nfrac = nbits << ENTROPY_SHIFT;
 
 	if (!nbits)
 		return;
 
-	DEBUG_ENT("added %d entropy credits to %s\n", nbits, r->name);
 retry:
 	entropy_count = orig = ACCESS_ONCE(r->entropy_count);
-	entropy_count += nbits;
+	if (nfrac < 0) {
+		/* Debit */
+		entropy_count += nfrac;
+	} else {
+		/*
+		 * Credit: we have to account for the possibility of
+		 * overwriting already present entropy.	 Even in the
+		 * ideal case of pure Shannon entropy, new contributions
+		 * approach the full value asymptotically:
+		 *
+		 * entropy <- entropy + (pool_size - entropy) *
+		 *	(1 - exp(-add_entropy/pool_size))
+		 *
+		 * For add_entropy <= pool_size/2 then
+		 * (1 - exp(-add_entropy/pool_size)) >=
+		 *    (add_entropy/pool_size)*0.7869...
+		 * so we can approximate the exponential with
+		 * 3/4*add_entropy/pool_size and still be on the
+		 * safe side by adding at most pool_size/2 at a time.
+		 *
+		 * The use of pool_size-2 in the while statement is to
+		 * prevent rounding artifacts from making the loop
+		 * arbitrarily long; this limits the loop to log2(pool_size)*2
+		 * turns no matter how large nbits is.
+		 */
+		int pnfrac = nfrac;
+		const int s = r->poolinfo->poolbitshift + ENTROPY_SHIFT + 2;
+		/* The +2 corresponds to the /4 in the denominator */
+
+		do {
+			unsigned int anfrac = min(pnfrac, pool_size/2);
+			unsigned int add =
+				((pool_size - entropy_count)*anfrac*3) >> s;
+
+			entropy_count += add;
+			pnfrac -= anfrac;
+		} while (unlikely(entropy_count < pool_size-2 && pnfrac));
+	}
 
 	if (entropy_count < 0) {
-		DEBUG_ENT("negative entropy/overflow\n");
+		pr_warn("random: negative entropy/overflow: pool %s count %d\n",
+			r->name, entropy_count);
+		WARN_ON(1);
 		entropy_count = 0;
-	} else if (entropy_count > r->poolinfo->POOLBITS)
-		entropy_count = r->poolinfo->POOLBITS;
+	} else if (entropy_count > pool_size)
+		entropy_count = pool_size;
 	if (cmpxchg(&r->entropy_count, orig, entropy_count) != orig)
 		goto retry;
 
-	if (!r->initialized && nbits > 0) {
-		r->entropy_total += nbits;
-		if (r->entropy_total > 128) {
-			r->initialized = 1;
-			if (r == &nonblocking_pool)
-				prandom_reseed_late();
+	r->entropy_total += nbits;
+	if (!r->initialized && r->entropy_total > 128) {
+		r->initialized = 1;
+		r->entropy_total = 0;
+		if (r == &nonblocking_pool) {
+			prandom_reseed_late();
+			pr_notice("random: %s pool is initialized\n", r->name);
 		}
 	}
 
-	trace_credit_entropy_bits(r->name, nbits, entropy_count,
+	trace_credit_entropy_bits(r->name, nbits,
+				  entropy_count >> ENTROPY_SHIFT,
 				  r->entropy_total, _RET_IP_);
 
-	/* should we wake readers? */
-	if (r == &input_pool && entropy_count >= random_read_wakeup_thresh) {
-		wake_up_interruptible(&random_read_wait);
-		kill_fasync(&fasync, SIGIO, POLL_IN);
+	if (r == &input_pool) {
+		int entropy_bytes = entropy_count >> ENTROPY_SHIFT;
+
+		/* should we wake readers? */
+		if (entropy_bytes >= random_read_wakeup_thresh) {
+			wake_up_interruptible(&random_read_wait);
+			kill_fasync(&fasync, SIGIO, POLL_IN);
+		}
+		/* If the input pool is getting full, send some
+		 * entropy to the two output pools, flipping back and
+		 * forth between them, until the output pools are 75%
+		 * full.
+		 */
+		if (entropy_bytes > random_write_wakeup_thresh &&
+		    r->initialized &&
+		    r->entropy_total >= 2*random_read_wakeup_thresh) {
+			static struct entropy_store *last = &blocking_pool;
+			struct entropy_store *other = &blocking_pool;
+
+			if (last == &blocking_pool)
+				other = &nonblocking_pool;
+			if (other->entropy_count <=
+			    3 * other->poolinfo->poolfracbits / 4)
+				last = other;
+			if (last->entropy_count <=
+			    3 * last->poolinfo->poolfracbits / 4) {
+				schedule_work(&last->push_work);
+				r->entropy_total = 0;
+			}
+		}
 	}
 }
 
+static void credit_entropy_bits_safe(struct entropy_store *r, int nbits)
+{
+	const int nbits_max = (int)(~0U >> (ENTROPY_SHIFT + 1));
+
+	/* Cap the value to avoid overflows */
+	nbits = min(nbits,  nbits_max);
+	nbits = max(nbits, -nbits_max);
+
+	credit_entropy_bits(r, nbits);
+}
+
 /*********************************************************************
  *
  * Entropy input management
@@ -633,6 +722,8 @@ struct timer_rand_state {
 	unsigned dont_count_entropy:1;
 };
 
+#define INIT_TIMER_RAND_STATE { INITIAL_JIFFIES, };
+
 /*
  * Add device- or boot-specific data to the input and nonblocking
  * pools to help initialize them to unique values.
@@ -644,15 +735,22 @@ struct timer_rand_state {
 void add_device_randomness(const void *buf, unsigned int size)
 {
 	unsigned long time = random_get_entropy() ^ jiffies;
+	unsigned long flags;
 
-	mix_pool_bytes(&input_pool, buf, size, NULL);
-	mix_pool_bytes(&input_pool, &time, sizeof(time), NULL);
-	mix_pool_bytes(&nonblocking_pool, buf, size, NULL);
-	mix_pool_bytes(&nonblocking_pool, &time, sizeof(time), NULL);
+	trace_add_device_randomness(size, _RET_IP_);
+	spin_lock_irqsave(&input_pool.lock, flags);
+	_mix_pool_bytes(&input_pool, buf, size, NULL);
+	_mix_pool_bytes(&input_pool, &time, sizeof(time), NULL);
+	spin_unlock_irqrestore(&input_pool.lock, flags);
+
+	spin_lock_irqsave(&nonblocking_pool.lock, flags);
+	_mix_pool_bytes(&nonblocking_pool, buf, size, NULL);
+	_mix_pool_bytes(&nonblocking_pool, &time, sizeof(time), NULL);
+	spin_unlock_irqrestore(&nonblocking_pool.lock, flags);
 }
 EXPORT_SYMBOL(add_device_randomness);
 
-static struct timer_rand_state input_timer_state;
+static struct timer_rand_state input_timer_state = INIT_TIMER_RAND_STATE;
 
 /*
  * This function adds entropy to the entropy "pool" by using timing
@@ -666,6 +764,7 @@ static struct timer_rand_state input_timer_state;
  */
 static void add_timer_randomness(struct timer_rand_state *state, unsigned num)
 {
+	struct entropy_store	*r;
 	struct {
 		long jiffies;
 		unsigned cycles;
@@ -674,15 +773,12 @@ static void add_timer_randomness(struct timer_rand_state *state, unsigned num)
 	long delta, delta2, delta3;
 
 	preempt_disable();
-	/* if over the trickle threshold, use only 1 in 4096 samples */
-	if (input_pool.entropy_count > trickle_thresh &&
-	    ((__this_cpu_inc_return(trickle_count) - 1) & 0xfff))
-		goto out;
 
 	sample.jiffies = jiffies;
 	sample.cycles = random_get_entropy();
 	sample.num = num;
-	mix_pool_bytes(&input_pool, &sample, sizeof(sample), NULL);
+	r = nonblocking_pool.initialized ? &input_pool : &nonblocking_pool;
+	mix_pool_bytes(r, &sample, sizeof(sample), NULL);
 
 	/*
 	 * Calculate number of bits of randomness we probably added.
@@ -716,10 +812,8 @@ static void add_timer_randomness(struct timer_rand_state *state, unsigned num)
 		 * Round down by 1 bit on general principles,
 		 * and limit entropy entimate to 12 bits.
 		 */
-		credit_entropy_bits(&input_pool,
-				    min_t(int, fls(delta>>1), 11));
+		credit_entropy_bits(r, min_t(int, fls(delta>>1), 11));
 	}
-out:
 	preempt_enable();
 }
 
@@ -732,10 +826,10 @@ void add_input_randomness(unsigned int type, unsigned int code,
 	if (value == last_value)
 		return;
 
-	DEBUG_ENT("input event\n");
 	last_value = value;
 	add_timer_randomness(&input_timer_state,
 			     (type << 4) ^ code ^ (code >> 4) ^ value);
+	trace_add_input_randomness(ENTROPY_BITS(&input_pool));
 }
 EXPORT_SYMBOL_GPL(add_input_randomness);
 
@@ -747,20 +841,21 @@ void add_interrupt_randomness(int irq, int irq_flags)
 	struct fast_pool	*fast_pool = &__get_cpu_var(irq_randomness);
 	struct pt_regs		*regs = get_irq_regs();
 	unsigned long		now = jiffies;
-	__u32			input[4], cycles = random_get_entropy();
-
-	input[0] = cycles ^ jiffies;
-	input[1] = irq;
-	if (regs) {
-		__u64 ip = instruction_pointer(regs);
-		input[2] = ip;
-		input[3] = ip >> 32;
-	}
+	cycles_t		cycles = random_get_entropy();
+	__u32			input[4], c_high, j_high;
+	__u64			ip;
 
-	fast_mix(fast_pool, input, sizeof(input));
+	c_high = (sizeof(cycles) > 4) ? cycles >> 32 : 0;
+	j_high = (sizeof(now) > 4) ? now >> 32 : 0;
+	input[0] = cycles ^ j_high ^ irq;
+	input[1] = now ^ c_high;
+	ip = regs ? instruction_pointer(regs) : _RET_IP_;
+	input[2] = ip;
+	input[3] = ip >> 32;
 
-	if ((fast_pool->count & 1023) &&
-	    !time_after(now, fast_pool->last + HZ))
+	fast_mix(fast_pool, input);
+
+	if ((fast_pool->count & 63) && !time_after(now, fast_pool->last + HZ))
 		return;
 
 	fast_pool->last = now;
@@ -789,10 +884,8 @@ void add_disk_randomness(struct gendisk *disk)
 	if (!disk || !disk->random)
 		return;
 	/* first major is 1, so we get >= 0x200 here */
-	DEBUG_ENT("disk event %d:%d\n",
-		  MAJOR(disk_devt(disk)), MINOR(disk_devt(disk)));
-
 	add_timer_randomness(disk->random, 0x100 + disk_devt(disk));
+	trace_add_disk_randomness(disk_devt(disk), ENTROPY_BITS(&input_pool));
 }
 #endif
 
@@ -810,30 +903,58 @@ static ssize_t extract_entropy(struct entropy_store *r, void *buf,
  * from the primary pool to the secondary extraction pool. We make
  * sure we pull enough for a 'catastrophic reseed'.
  */
+static void _xfer_secondary_pool(struct entropy_store *r, size_t nbytes);
 static void xfer_secondary_pool(struct entropy_store *r, size_t nbytes)
 {
-	__u32	tmp[OUTPUT_POOL_WORDS];
+	if (r->limit == 0 && random_min_urandom_seed) {
+		unsigned long now = jiffies;
 
-	if (r->pull && r->entropy_count < nbytes * 8 &&
-	    r->entropy_count < r->poolinfo->POOLBITS) {
-		/* If we're limited, always leave two wakeup worth's BITS */
-		int rsvd = r->limit ? 0 : random_read_wakeup_thresh/4;
-		int bytes = nbytes;
-
-		/* pull at least as many as BYTES as wakeup BITS */
-		bytes = max_t(int, bytes, random_read_wakeup_thresh / 8);
-		/* but never more than the buffer size */
-		bytes = min_t(int, bytes, sizeof(tmp));
-
-		DEBUG_ENT("going to reseed %s with %d bits "
-			  "(%zu of %d requested)\n",
-			  r->name, bytes * 8, nbytes * 8, r->entropy_count);
-
-		bytes = extract_entropy(r->pull, tmp, bytes,
-					random_read_wakeup_thresh / 8, rsvd);
-		mix_pool_bytes(r, tmp, bytes, NULL);
-		credit_entropy_bits(r, bytes*8);
+		if (time_before(now,
+				r->last_pulled + random_min_urandom_seed * HZ))
+			return;
+		r->last_pulled = now;
 	}
+	if (r->pull &&
+	    r->entropy_count < (nbytes << (ENTROPY_SHIFT + 3)) &&
+	    r->entropy_count < r->poolinfo->poolfracbits)
+		_xfer_secondary_pool(r, nbytes);
+}
+
+static void _xfer_secondary_pool(struct entropy_store *r, size_t nbytes)
+{
+	__u32	tmp[OUTPUT_POOL_WORDS];
+
+	/* For /dev/random's pool, always leave two wakeup worth's BITS */
+	int rsvd = r->limit ? 0 : random_read_wakeup_thresh/4;
+	int bytes = nbytes;
+
+	/* pull at least as many as BYTES as wakeup BITS */
+	bytes = max_t(int, bytes, random_read_wakeup_thresh / 8);
+	/* but never more than the buffer size */
+	bytes = min_t(int, bytes, sizeof(tmp));
+
+	trace_xfer_secondary_pool(r->name, bytes * 8, nbytes * 8,
+				  ENTROPY_BITS(r), ENTROPY_BITS(r->pull));
+	bytes = extract_entropy(r->pull, tmp, bytes,
+				random_read_wakeup_thresh / 8, rsvd);
+	mix_pool_bytes(r, tmp, bytes, NULL);
+	credit_entropy_bits(r, bytes*8);
+}
+
+/*
+ * Used as a workqueue function so that when the input pool is getting
+ * full, we can "spill over" some entropy to the output pools.  That
+ * way the output pools can store some of the excess entropy instead
+ * of letting it go to waste.
+ */
+static void push_to_pool(struct work_struct *work)
+{
+	struct entropy_store *r = container_of(work, struct entropy_store,
+					      push_work);
+	BUG_ON(!r);
+	_xfer_secondary_pool(r, random_read_wakeup_thresh/8);
+	trace_push_to_pool(r->name, r->entropy_count >> ENTROPY_SHIFT,
+			   r->pull->entropy_count >> ENTROPY_SHIFT);
 }
 
 /*
@@ -853,50 +974,48 @@ static size_t account(struct entropy_store *r, size_t nbytes, int min,
 {
 	unsigned long flags;
 	int wakeup_write = 0;
+	int have_bytes;
+	int entropy_count, orig;
+	size_t ibytes;
 
 	/* Hold lock while accounting */
 	spin_lock_irqsave(&r->lock, flags);
 
-	BUG_ON(r->entropy_count > r->poolinfo->POOLBITS);
-	DEBUG_ENT("trying to extract %zu bits from %s\n",
-		  nbytes * 8, r->name);
+	BUG_ON(r->entropy_count > r->poolinfo->poolfracbits);
 
 	/* Can we pull enough? */
-	if (r->entropy_count / 8 < min + reserved) {
-		nbytes = 0;
-	} else {
-		int entropy_count, orig;
 retry:
-		entropy_count = orig = ACCESS_ONCE(r->entropy_count);
+	entropy_count = orig = ACCESS_ONCE(r->entropy_count);
+	have_bytes = entropy_count >> (ENTROPY_SHIFT + 3);
+	ibytes = nbytes;
+	if (have_bytes < min + reserved) {
+		ibytes = 0;
+	} else {
 		/* If limited, never pull more than available */
-		if (r->limit && nbytes + reserved >= entropy_count / 8)
-			nbytes = entropy_count/8 - reserved;
-
-		if (entropy_count / 8 >= nbytes + reserved) {
-			entropy_count -= nbytes*8;
-			if (cmpxchg(&r->entropy_count, orig, entropy_count) != orig)
-				goto retry;
-		} else {
-			entropy_count = reserved;
-			if (cmpxchg(&r->entropy_count, orig, entropy_count) != orig)
-				goto retry;
-		}
+		if (r->limit && ibytes + reserved >= have_bytes)
+			ibytes = have_bytes - reserved;
 
-		if (entropy_count < random_write_wakeup_thresh)
-			wakeup_write = 1;
-	}
+		if (have_bytes >= ibytes + reserved)
+			entropy_count -= ibytes << (ENTROPY_SHIFT + 3);
+		else
+			entropy_count = reserved << (ENTROPY_SHIFT + 3);
 
-	DEBUG_ENT("debiting %zu entropy credits from %s%s\n",
-		  nbytes * 8, r->name, r->limit ? "" : " (unlimited)");
+		if (cmpxchg(&r->entropy_count, orig, entropy_count) != orig)
+			goto retry;
 
+		if ((r->entropy_count >> ENTROPY_SHIFT)
+		    < random_write_wakeup_thresh)
+			wakeup_write = 1;
+	}
 	spin_unlock_irqrestore(&r->lock, flags);
 
+	trace_debit_entropy(r->name, 8 * ibytes);
 	if (wakeup_write) {
 		wake_up_interruptible(&random_write_wait);
 		kill_fasync(&fasync, SIGIO, POLL_OUT);
 	}
 
-	return nbytes;
+	return ibytes;
 }
 
 static void extract_buf(struct entropy_store *r, __u8 *out)
@@ -904,7 +1023,7 @@ static void extract_buf(struct entropy_store *r, __u8 *out)
 	int i;
 	union {
 		__u32 w[5];
-		unsigned long l[LONGS(EXTRACT_SIZE)];
+		unsigned long l[LONGS(20)];
 	} hash;
 	__u32 workspace[SHA_WORKSPACE_WORDS];
 	__u8 extract[64];
@@ -917,6 +1036,17 @@ static void extract_buf(struct entropy_store *r, __u8 *out)
 		sha_transform(hash.w, (__u8 *)(r->pool + i), workspace);
 
 	/*
+	 * If we have a architectural hardware random number
+	 * generator, mix that in, too.
+	 */
+	for (i = 0; i < LONGS(20); i++) {
+		unsigned long v;
+		if (!arch_get_random_long(&v))
+			break;
+		hash.l[i] ^= v;
+	}
+
+	/*
 	 * We mix the hash back into the pool to prevent backtracking
 	 * attacks (where the attacker knows the state of the pool
 	 * plus the current outputs, and attempts to find previous
@@ -945,17 +1075,6 @@ static void extract_buf(struct entropy_store *r, __u8 *out)
 	hash.w[1] ^= hash.w[4];
 	hash.w[2] ^= rol32(hash.w[2], 16);
 
-	/*
-	 * If we have a architectural hardware random number
-	 * generator, mix that in, too.
-	 */
-	for (i = 0; i < LONGS(EXTRACT_SIZE); i++) {
-		unsigned long v;
-		if (!arch_get_random_long(&v))
-			break;
-		hash.l[i] ^= v;
-	}
-
 	memcpy(out, &hash, EXTRACT_SIZE);
 	memset(&hash, 0, sizeof(hash));
 }
@@ -971,10 +1090,10 @@ static ssize_t extract_entropy(struct entropy_store *r, void *buf,
 	if (fips_enabled) {
 		spin_lock_irqsave(&r->lock, flags);
 		if (!r->last_data_init) {
-			r->last_data_init = true;
+			r->last_data_init = 1;
 			spin_unlock_irqrestore(&r->lock, flags);
 			trace_extract_entropy(r->name, EXTRACT_SIZE,
-					      r->entropy_count, _RET_IP_);
+					      ENTROPY_BITS(r), _RET_IP_);
 			xfer_secondary_pool(r, EXTRACT_SIZE);
 			extract_buf(r, tmp);
 			spin_lock_irqsave(&r->lock, flags);
@@ -983,7 +1102,7 @@ static ssize_t extract_entropy(struct entropy_store *r, void *buf,
 		spin_unlock_irqrestore(&r->lock, flags);
 	}
 
-	trace_extract_entropy(r->name, nbytes, r->entropy_count, _RET_IP_);
+	trace_extract_entropy(r->name, nbytes, ENTROPY_BITS(r), _RET_IP_);
 	xfer_secondary_pool(r, nbytes);
 	nbytes = account(r, nbytes, min, reserved);
 
@@ -1016,7 +1135,7 @@ static ssize_t extract_entropy_user(struct entropy_store *r, void __user *buf,
 	ssize_t ret = 0, i;
 	__u8 tmp[EXTRACT_SIZE];
 
-	trace_extract_entropy_user(r->name, nbytes, r->entropy_count, _RET_IP_);
+	trace_extract_entropy_user(r->name, nbytes, ENTROPY_BITS(r), _RET_IP_);
 	xfer_secondary_pool(r, nbytes);
 	nbytes = account(r, nbytes, 0, 0);
 
@@ -1056,6 +1175,14 @@ static ssize_t extract_entropy_user(struct entropy_store *r, void __user *buf,
  */
 void get_random_bytes(void *buf, int nbytes)
 {
+#if DEBUG_RANDOM_BOOT > 0
+	if (unlikely(nonblocking_pool.initialized == 0))
+		printk(KERN_NOTICE "random: %pF get_random_bytes called "
+		       "with %d bits of entropy available\n",
+		       (void *) _RET_IP_,
+		       nonblocking_pool.entropy_total);
+#endif
+	trace_get_random_bytes(nbytes, _RET_IP_);
 	extract_entropy(&nonblocking_pool, buf, nbytes, 0, 0);
 }
 EXPORT_SYMBOL(get_random_bytes);
@@ -1074,7 +1201,7 @@ void get_random_bytes_arch(void *buf, int nbytes)
 {
 	char *p = buf;
 
-	trace_get_random_bytes(nbytes, _RET_IP_);
+	trace_get_random_bytes_arch(nbytes, _RET_IP_);
 	while (nbytes) {
 		unsigned long v;
 		int chunk = min(nbytes, (int)sizeof(unsigned long));
@@ -1108,13 +1235,11 @@ static void init_std_data(struct entropy_store *r)
 	ktime_t now = ktime_get_real();
 	unsigned long rv;
 
-	r->entropy_count = 0;
-	r->entropy_total = 0;
-	r->last_data_init = false;
+	r->last_pulled = jiffies;
 	mix_pool_bytes(r, &now, sizeof(now), NULL);
-	for (i = r->poolinfo->POOLBYTES; i > 0; i -= sizeof(rv)) {
+	for (i = r->poolinfo->poolbytes; i > 0; i -= sizeof(rv)) {
 		if (!arch_get_random_long(&rv))
-			break;
+			rv = random_get_entropy();
 		mix_pool_bytes(r, &rv, sizeof(rv), NULL);
 	}
 	mix_pool_bytes(r, utsname(), sizeof(*(utsname())), NULL);
@@ -1137,7 +1262,7 @@ static int rand_initialize(void)
 	init_std_data(&nonblocking_pool);
 	return 0;
 }
-module_init(rand_initialize);
+early_initcall(rand_initialize);
 
 #ifdef CONFIG_BLOCK
 void rand_initialize_disk(struct gendisk *disk)
@@ -1149,8 +1274,10 @@ void rand_initialize_disk(struct gendisk *disk)
 	 * source.
 	 */
 	state = kzalloc(sizeof(struct timer_rand_state), GFP_KERNEL);
-	if (state)
+	if (state) {
+		state->last_time = INITIAL_JIFFIES;
 		disk->random = state;
+	}
 }
 #endif
 
@@ -1167,8 +1294,6 @@ random_read(struct file *file, char __user *buf, size_t nbytes, loff_t *ppos)
 		if (n > SEC_XFER_SIZE)
 			n = SEC_XFER_SIZE;
 
-		DEBUG_ENT("reading %zu bits\n", n*8);
-
 		n = extract_entropy_user(&blocking_pool, buf, n);
 
 		if (n < 0) {
@@ -1176,8 +1301,9 @@ random_read(struct file *file, char __user *buf, size_t nbytes, loff_t *ppos)
 			break;
 		}
 
-		DEBUG_ENT("read got %zd bits (%zd still needed)\n",
-			  n*8, (nbytes-n)*8);
+		trace_random_read(n*8, (nbytes-n)*8,
+				  ENTROPY_BITS(&blocking_pool),
+				  ENTROPY_BITS(&input_pool));
 
 		if (n == 0) {
 			if (file->f_flags & O_NONBLOCK) {
@@ -1185,13 +1311,9 @@ random_read(struct file *file, char __user *buf, size_t nbytes, loff_t *ppos)
 				break;
 			}
 
-			DEBUG_ENT("sleeping?\n");
-
 			wait_event_interruptible(random_read_wait,
-				input_pool.entropy_count >=
-						 random_read_wakeup_thresh);
-
-			DEBUG_ENT("awake\n");
+				ENTROPY_BITS(&input_pool) >=
+				random_read_wakeup_thresh);
 
 			if (signal_pending(current)) {
 				retval = -ERESTARTSYS;
@@ -1214,7 +1336,18 @@ random_read(struct file *file, char __user *buf, size_t nbytes, loff_t *ppos)
 static ssize_t
 urandom_read(struct file *file, char __user *buf, size_t nbytes, loff_t *ppos)
 {
-	return extract_entropy_user(&nonblocking_pool, buf, nbytes);
+	int ret;
+
+	if (unlikely(nonblocking_pool.initialized == 0))
+		printk_once(KERN_NOTICE "random: %s urandom read "
+			    "with %d bits of entropy available\n",
+			    current->comm, nonblocking_pool.entropy_total);
+
+	ret = extract_entropy_user(&nonblocking_pool, buf, nbytes);
+
+	trace_urandom_read(8 * nbytes, ENTROPY_BITS(&nonblocking_pool),
+			   ENTROPY_BITS(&input_pool));
+	return ret;
 }
 
 static unsigned int
@@ -1225,9 +1358,9 @@ random_poll(struct file *file, poll_table * wait)
 	poll_wait(file, &random_read_wait, wait);
 	poll_wait(file, &random_write_wait, wait);
 	mask = 0;
-	if (input_pool.entropy_count >= random_read_wakeup_thresh)
+	if (ENTROPY_BITS(&input_pool) >= random_read_wakeup_thresh)
 		mask |= POLLIN | POLLRDNORM;
-	if (input_pool.entropy_count < random_write_wakeup_thresh)
+	if (ENTROPY_BITS(&input_pool) < random_write_wakeup_thresh)
 		mask |= POLLOUT | POLLWRNORM;
 	return mask;
 }
@@ -1278,7 +1411,8 @@ static long random_ioctl(struct file *f, unsigned int cmd, unsigned long arg)
 	switch (cmd) {
 	case RNDGETENTCNT:
 		/* inherently racy, no point locking */
-		if (put_user(input_pool.entropy_count, p))
+		ent_count = ENTROPY_BITS(&input_pool);
+		if (put_user(ent_count, p))
 			return -EFAULT;
 		return 0;
 	case RNDADDTOENTCNT:
@@ -1286,7 +1420,7 @@ static long random_ioctl(struct file *f, unsigned int cmd, unsigned long arg)
 			return -EPERM;
 		if (get_user(ent_count, p))
 			return -EFAULT;
-		credit_entropy_bits(&input_pool, ent_count);
+		credit_entropy_bits_safe(&input_pool, ent_count);
 		return 0;
 	case RNDADDENTROPY:
 		if (!capable(CAP_SYS_ADMIN))
@@ -1301,14 +1435,19 @@ static long random_ioctl(struct file *f, unsigned int cmd, unsigned long arg)
 				    size);
 		if (retval < 0)
 			return retval;
-		credit_entropy_bits(&input_pool, ent_count);
+		credit_entropy_bits_safe(&input_pool, ent_count);
 		return 0;
 	case RNDZAPENTCNT:
 	case RNDCLEARPOOL:
-		/* Clear the entropy pool counters. */
+		/*
+		 * Clear the entropy pool counters. We no longer clear
+		 * the entropy pool, as that's silly.
+		 */
 		if (!capable(CAP_SYS_ADMIN))
 			return -EPERM;
-		rand_initialize();
+		input_pool.entropy_count = 0;
+		nonblocking_pool.entropy_count = 0;
+		blocking_pool.entropy_count = 0;
 		return 0;
 	default:
 		return -EINVAL;
@@ -1408,6 +1547,23 @@ static int proc_do_uuid(struct ctl_table *table, int write,
 	return proc_dostring(&fake_table, write, buffer, lenp, ppos);
 }
 
+/*
+ * Return entropy available scaled to integral bits
+ */
+static int proc_do_entropy(ctl_table *table, int write,
+			   void __user *buffer, size_t *lenp, loff_t *ppos)
+{
+	ctl_table fake_table;
+	int entropy_count;
+
+	entropy_count = *(int *)table->data >> ENTROPY_SHIFT;
+
+	fake_table.data = &entropy_count;
+	fake_table.maxlen = sizeof(entropy_count);
+
+	return proc_dointvec(&fake_table, write, buffer, lenp, ppos);
+}
+
 static int sysctl_poolsize = INPUT_POOL_WORDS * 32;
 extern struct ctl_table random_table[];
 struct ctl_table random_table[] = {
@@ -1422,7 +1578,7 @@ struct ctl_table random_table[] = {
 		.procname	= "entropy_avail",
 		.maxlen		= sizeof(int),
 		.mode		= 0444,
-		.proc_handler	= proc_dointvec,
+		.proc_handler	= proc_do_entropy,
 		.data		= &input_pool.entropy_count,
 	},
 	{
@@ -1444,6 +1600,13 @@ struct ctl_table random_table[] = {
 		.extra2		= &max_write_thresh,
 	},
 	{
+		.procname	= "urandom_min_reseed_secs",
+		.data		= &random_min_urandom_seed,
+		.maxlen		= sizeof(int),
+		.mode		= 0644,
+		.proc_handler	= proc_dointvec,
+	},
+	{
 		.procname	= "boot_id",
 		.data		= &sysctl_bootid,
 		.maxlen		= 16,
diff --git a/drivers/char/virtio_console.c b/drivers/char/virtio_console.c
index b79cf3e1b793..feea87cc6b8f 100644
--- a/drivers/char/virtio_console.c
+++ b/drivers/char/virtio_console.c
@@ -577,7 +577,8 @@ static ssize_t __send_control_msg(struct ports_device *portdev, u32 port_id,
 	spin_lock(&portdev->c_ovq_lock);
 	if (virtqueue_add_outbuf(vq, sg, 1, &cpkt, GFP_ATOMIC) == 0) {
 		virtqueue_kick(vq);
-		while (!virtqueue_get_buf(vq, &len))
+		while (!virtqueue_get_buf(vq, &len)
+			&& !virtqueue_is_broken(vq))
 			cpu_relax();
 	}
 	spin_unlock(&portdev->c_ovq_lock);
@@ -650,7 +651,8 @@ static ssize_t __send_to_port(struct port *port, struct scatterlist *sg,
 	 * we need to kmalloc a GFP_ATOMIC buffer each time the
 	 * console driver writes something out.
 	 */
-	while (!virtqueue_get_buf(out_vq, &len))
+	while (!virtqueue_get_buf(out_vq, &len)
+		&& !virtqueue_is_broken(out_vq))
 		cpu_relax();
 done:
 	spin_unlock_irqrestore(&port->outvq_lock, flags);
@@ -1837,12 +1839,8 @@ static void config_intr(struct virtio_device *vdev)
 		struct port *port;
 		u16 rows, cols;
 
-		vdev->config->get(vdev,
-				  offsetof(struct virtio_console_config, cols),
-				  &cols, sizeof(u16));
-		vdev->config->get(vdev,
-				  offsetof(struct virtio_console_config, rows),
-				  &rows, sizeof(u16));
+		virtio_cread(vdev, struct virtio_console_config, cols, &cols);
+		virtio_cread(vdev, struct virtio_console_config, rows, &rows);
 
 		port = find_port_by_id(portdev, 0);
 		set_console_size(port, rows, cols);
@@ -2014,10 +2012,9 @@ static int virtcons_probe(struct virtio_device *vdev)
 
 	/* Don't test MULTIPORT at all if we're rproc: not a valid feature! */
 	if (!is_rproc_serial(vdev) &&
-	    virtio_config_val(vdev, VIRTIO_CONSOLE_F_MULTIPORT,
-				  offsetof(struct virtio_console_config,
-					   max_nr_ports),
-				  &portdev->config.max_nr_ports) == 0) {
+	    virtio_cread_feature(vdev, VIRTIO_CONSOLE_F_MULTIPORT,
+				 struct virtio_console_config, max_nr_ports,
+				 &portdev->config.max_nr_ports) == 0) {
 		multiport = true;
 	}
 
@@ -2142,7 +2139,7 @@ static struct virtio_device_id rproc_serial_id_table[] = {
 static unsigned int rproc_serial_features[] = {
 };
 
-#ifdef CONFIG_PM
+#ifdef CONFIG_PM_SLEEP
 static int virtcons_freeze(struct virtio_device *vdev)
 {
 	struct ports_device *portdev;
@@ -2220,7 +2217,7 @@ static struct virtio_driver virtio_console = {
 	.probe =	virtcons_probe,
 	.remove =	virtcons_remove,
 	.config_changed = config_intr,
-#ifdef CONFIG_PM
+#ifdef CONFIG_PM_SLEEP
 	.freeze =	virtcons_freeze,
 	.restore =	virtcons_restore,
 #endif
diff --git a/drivers/clk/clk-fixed-factor.c b/drivers/clk/clk-fixed-factor.c
index 0e1d89b4321b..d9e3f671c2ea 100644
--- a/drivers/clk/clk-fixed-factor.c
+++ b/drivers/clk/clk-fixed-factor.c
@@ -117,7 +117,7 @@ void __init of_fixed_factor_clk_setup(struct device_node *node)
 	}
 
 	if (of_property_read_u32(node, "clock-mult", &mult)) {
-		pr_err("%s Fixed factor clock <%s> must have a clokc-mult property\n",
+		pr_err("%s Fixed factor clock <%s> must have a clock-mult property\n",
 			__func__, node->name);
 		return;
 	}
diff --git a/drivers/connector/cn_proc.c b/drivers/connector/cn_proc.c
index c73fc2b74de2..18c5b9b16645 100644
--- a/drivers/connector/cn_proc.c
+++ b/drivers/connector/cn_proc.c
@@ -32,11 +32,23 @@
 #include <linux/atomic.h>
 #include <linux/pid_namespace.h>
 
-#include <asm/unaligned.h>
-
 #include <linux/cn_proc.h>
 
-#define CN_PROC_MSG_SIZE (sizeof(struct cn_msg) + sizeof(struct proc_event))
+/*
+ * Size of a cn_msg followed by a proc_event structure.  Since the
+ * sizeof struct cn_msg is a multiple of 4 bytes, but not 8 bytes, we
+ * add one 4-byte word to the size here, and then start the actual
+ * cn_msg structure 4 bytes into the stack buffer.  The result is that
+ * the immediately following proc_event structure is aligned to 8 bytes.
+ */
+#define CN_PROC_MSG_SIZE (sizeof(struct cn_msg) + sizeof(struct proc_event) + 4)
+
+/* See comment above; we test our assumption about sizeof struct cn_msg here. */
+static inline struct cn_msg *buffer_to_cn_msg(__u8 *buffer)
+{
+	BUILD_BUG_ON(sizeof(struct cn_msg) != 20);
+	return (struct cn_msg *)(buffer + 4);
+}
 
 static atomic_t proc_event_num_listeners = ATOMIC_INIT(0);
 static struct cb_id cn_proc_event_id = { CN_IDX_PROC, CN_VAL_PROC };
@@ -56,19 +68,19 @@ void proc_fork_connector(struct task_struct *task)
 {
 	struct cn_msg *msg;
 	struct proc_event *ev;
-	__u8 buffer[CN_PROC_MSG_SIZE];
+	__u8 buffer[CN_PROC_MSG_SIZE] __aligned(8);
 	struct timespec ts;
 	struct task_struct *parent;
 
 	if (atomic_read(&proc_event_num_listeners) < 1)
 		return;
 
-	msg = (struct cn_msg *)buffer;
+	msg = buffer_to_cn_msg(buffer);
 	ev = (struct proc_event *)msg->data;
 	memset(&ev->event_data, 0, sizeof(ev->event_data));
 	get_seq(&msg->seq, &ev->cpu);
 	ktime_get_ts(&ts); /* get high res monotonic timestamp */
-	put_unaligned(timespec_to_ns(&ts), (__u64 *)&ev->timestamp_ns);
+	ev->timestamp_ns = timespec_to_ns(&ts);
 	ev->what = PROC_EVENT_FORK;
 	rcu_read_lock();
 	parent = rcu_dereference(task->real_parent);
@@ -91,17 +103,17 @@ void proc_exec_connector(struct task_struct *task)
 	struct cn_msg *msg;
 	struct proc_event *ev;
 	struct timespec ts;
-	__u8 buffer[CN_PROC_MSG_SIZE];
+	__u8 buffer[CN_PROC_MSG_SIZE] __aligned(8);
 
 	if (atomic_read(&proc_event_num_listeners) < 1)
 		return;
 
-	msg = (struct cn_msg *)buffer;
+	msg = buffer_to_cn_msg(buffer);
 	ev = (struct proc_event *)msg->data;
 	memset(&ev->event_data, 0, sizeof(ev->event_data));
 	get_seq(&msg->seq, &ev->cpu);
 	ktime_get_ts(&ts); /* get high res monotonic timestamp */
-	put_unaligned(timespec_to_ns(&ts), (__u64 *)&ev->timestamp_ns);
+	ev->timestamp_ns = timespec_to_ns(&ts);
 	ev->what = PROC_EVENT_EXEC;
 	ev->event_data.exec.process_pid = task->pid;
 	ev->event_data.exec.process_tgid = task->tgid;
@@ -117,14 +129,14 @@ void proc_id_connector(struct task_struct *task, int which_id)
 {
 	struct cn_msg *msg;
 	struct proc_event *ev;
-	__u8 buffer[CN_PROC_MSG_SIZE];
+	__u8 buffer[CN_PROC_MSG_SIZE] __aligned(8);
 	struct timespec ts;
 	const struct cred *cred;
 
 	if (atomic_read(&proc_event_num_listeners) < 1)
 		return;
 
-	msg = (struct cn_msg *)buffer;
+	msg = buffer_to_cn_msg(buffer);
 	ev = (struct proc_event *)msg->data;
 	memset(&ev->event_data, 0, sizeof(ev->event_data));
 	ev->what = which_id;
@@ -145,7 +157,7 @@ void proc_id_connector(struct task_struct *task, int which_id)
 	rcu_read_unlock();
 	get_seq(&msg->seq, &ev->cpu);
 	ktime_get_ts(&ts); /* get high res monotonic timestamp */
-	put_unaligned(timespec_to_ns(&ts), (__u64 *)&ev->timestamp_ns);
+	ev->timestamp_ns = timespec_to_ns(&ts);
 
 	memcpy(&msg->id, &cn_proc_event_id, sizeof(msg->id));
 	msg->ack = 0; /* not used */
@@ -159,17 +171,17 @@ void proc_sid_connector(struct task_struct *task)
 	struct cn_msg *msg;
 	struct proc_event *ev;
 	struct timespec ts;
-	__u8 buffer[CN_PROC_MSG_SIZE];
+	__u8 buffer[CN_PROC_MSG_SIZE] __aligned(8);
 
 	if (atomic_read(&proc_event_num_listeners) < 1)
 		return;
 
-	msg = (struct cn_msg *)buffer;
+	msg = buffer_to_cn_msg(buffer);
 	ev = (struct proc_event *)msg->data;
 	memset(&ev->event_data, 0, sizeof(ev->event_data));
 	get_seq(&msg->seq, &ev->cpu);
 	ktime_get_ts(&ts); /* get high res monotonic timestamp */
-	put_unaligned(timespec_to_ns(&ts), (__u64 *)&ev->timestamp_ns);
+	ev->timestamp_ns = timespec_to_ns(&ts);
 	ev->what = PROC_EVENT_SID;
 	ev->event_data.sid.process_pid = task->pid;
 	ev->event_data.sid.process_tgid = task->tgid;
@@ -186,17 +198,17 @@ void proc_ptrace_connector(struct task_struct *task, int ptrace_id)
 	struct cn_msg *msg;
 	struct proc_event *ev;
 	struct timespec ts;
-	__u8 buffer[CN_PROC_MSG_SIZE];
+	__u8 buffer[CN_PROC_MSG_SIZE] __aligned(8);
 
 	if (atomic_read(&proc_event_num_listeners) < 1)
 		return;
 
-	msg = (struct cn_msg *)buffer;
+	msg = buffer_to_cn_msg(buffer);
 	ev = (struct proc_event *)msg->data;
 	memset(&ev->event_data, 0, sizeof(ev->event_data));
 	get_seq(&msg->seq, &ev->cpu);
 	ktime_get_ts(&ts); /* get high res monotonic timestamp */
-	put_unaligned(timespec_to_ns(&ts), (__u64 *)&ev->timestamp_ns);
+	ev->timestamp_ns = timespec_to_ns(&ts);
 	ev->what = PROC_EVENT_PTRACE;
 	ev->event_data.ptrace.process_pid  = task->pid;
 	ev->event_data.ptrace.process_tgid = task->tgid;
@@ -221,17 +233,17 @@ void proc_comm_connector(struct task_struct *task)
 	struct cn_msg *msg;
 	struct proc_event *ev;
 	struct timespec ts;
-	__u8 buffer[CN_PROC_MSG_SIZE];
+	__u8 buffer[CN_PROC_MSG_SIZE] __aligned(8);
 
 	if (atomic_read(&proc_event_num_listeners) < 1)
 		return;
 
-	msg = (struct cn_msg *)buffer;
+	msg = buffer_to_cn_msg(buffer);
 	ev = (struct proc_event *)msg->data;
 	memset(&ev->event_data, 0, sizeof(ev->event_data));
 	get_seq(&msg->seq, &ev->cpu);
 	ktime_get_ts(&ts); /* get high res monotonic timestamp */
-	put_unaligned(timespec_to_ns(&ts), (__u64 *)&ev->timestamp_ns);
+	ev->timestamp_ns = timespec_to_ns(&ts);
 	ev->what = PROC_EVENT_COMM;
 	ev->event_data.comm.process_pid  = task->pid;
 	ev->event_data.comm.process_tgid = task->tgid;
@@ -248,18 +260,18 @@ void proc_coredump_connector(struct task_struct *task)
 {
 	struct cn_msg *msg;
 	struct proc_event *ev;
-	__u8 buffer[CN_PROC_MSG_SIZE];
+	__u8 buffer[CN_PROC_MSG_SIZE] __aligned(8);
 	struct timespec ts;
 
 	if (atomic_read(&proc_event_num_listeners) < 1)
 		return;
 
-	msg = (struct cn_msg *)buffer;
+	msg = buffer_to_cn_msg(buffer);
 	ev = (struct proc_event *)msg->data;
 	memset(&ev->event_data, 0, sizeof(ev->event_data));
 	get_seq(&msg->seq, &ev->cpu);
 	ktime_get_ts(&ts); /* get high res monotonic timestamp */
-	put_unaligned(timespec_to_ns(&ts), (__u64 *)&ev->timestamp_ns);
+	ev->timestamp_ns = timespec_to_ns(&ts);
 	ev->what = PROC_EVENT_COREDUMP;
 	ev->event_data.coredump.process_pid = task->pid;
 	ev->event_data.coredump.process_tgid = task->tgid;
@@ -275,18 +287,18 @@ void proc_exit_connector(struct task_struct *task)
 {
 	struct cn_msg *msg;
 	struct proc_event *ev;
-	__u8 buffer[CN_PROC_MSG_SIZE];
+	__u8 buffer[CN_PROC_MSG_SIZE] __aligned(8);
 	struct timespec ts;
 
 	if (atomic_read(&proc_event_num_listeners) < 1)
 		return;
 
-	msg = (struct cn_msg *)buffer;
+	msg = buffer_to_cn_msg(buffer);
 	ev = (struct proc_event *)msg->data;
 	memset(&ev->event_data, 0, sizeof(ev->event_data));
 	get_seq(&msg->seq, &ev->cpu);
 	ktime_get_ts(&ts); /* get high res monotonic timestamp */
-	put_unaligned(timespec_to_ns(&ts), (__u64 *)&ev->timestamp_ns);
+	ev->timestamp_ns = timespec_to_ns(&ts);
 	ev->what = PROC_EVENT_EXIT;
 	ev->event_data.exit.process_pid = task->pid;
 	ev->event_data.exit.process_tgid = task->tgid;
@@ -312,18 +324,18 @@ static void cn_proc_ack(int err, int rcvd_seq, int rcvd_ack)
 {
 	struct cn_msg *msg;
 	struct proc_event *ev;
-	__u8 buffer[CN_PROC_MSG_SIZE];
+	__u8 buffer[CN_PROC_MSG_SIZE] __aligned(8);
 	struct timespec ts;
 
 	if (atomic_read(&proc_event_num_listeners) < 1)
 		return;
 
-	msg = (struct cn_msg *)buffer;
+	msg = buffer_to_cn_msg(buffer);
 	ev = (struct proc_event *)msg->data;
 	memset(&ev->event_data, 0, sizeof(ev->event_data));
 	msg->seq = rcvd_seq;
 	ktime_get_ts(&ts); /* get high res monotonic timestamp */
-	put_unaligned(timespec_to_ns(&ts), (__u64 *)&ev->timestamp_ns);
+	ev->timestamp_ns = timespec_to_ns(&ts);
 	ev->cpu = -1;
 	ev->what = PROC_EVENT_NONE;
 	ev->event_data.ack.err = err;
diff --git a/drivers/cpufreq/Kconfig.x86 b/drivers/cpufreq/Kconfig.x86
index 6897ad85b046..d369349eeaab 100644
--- a/drivers/cpufreq/Kconfig.x86
+++ b/drivers/cpufreq/Kconfig.x86
@@ -129,7 +129,7 @@ config X86_AMD_FREQ_SENSITIVITY
 	help
 	  This adds AMD-specific powersave bias function to the ondemand
 	  governor, which allows it to make more power-conscious frequency
-	  change decisions based on feedback from hardware (availble on AMD
+	  change decisions based on feedback from hardware (available on AMD
 	  Family 16h and above).
 
 	  Hardware feedback tells software how "sensitive" to frequency changes
diff --git a/drivers/cpufreq/exynos-cpufreq.c b/drivers/cpufreq/exynos-cpufreq.c
index 7b6dc06b1bd4..f3c22874da75 100644
--- a/drivers/cpufreq/exynos-cpufreq.c
+++ b/drivers/cpufreq/exynos-cpufreq.c
@@ -67,7 +67,7 @@ static int exynos_cpufreq_scale(unsigned int target_freq)
 	/*
 	 * The policy max have been changed so that we cannot get proper
 	 * old_index with cpufreq_frequency_table_target(). Thus, ignore
-	 * policy and get the index from the raw freqeuncy table.
+	 * policy and get the index from the raw frequency table.
 	 */
 	old_index = exynos_cpufreq_get_index(old_freq);
 	if (old_index < 0) {
diff --git a/drivers/crypto/tegra-aes.c b/drivers/crypto/tegra-aes.c
index 2d58da972ae2..fa05e3c329bd 100644
--- a/drivers/crypto/tegra-aes.c
+++ b/drivers/crypto/tegra-aes.c
@@ -268,7 +268,7 @@ static int aes_start_crypt(struct tegra_aes_dev *dd, u32 in_addr, u32 out_addr,
 	aes_writel(dd, value, TEGRA_AES_SECURE_INPUT_SELECT);
 
 	aes_writel(dd, out_addr, TEGRA_AES_SECURE_DEST_ADDR);
-	INIT_COMPLETION(dd->op_complete);
+	reinit_completion(&dd->op_complete);
 
 	for (i = 0; i < AES_HW_MAX_ICQ_LENGTH - 1; i++) {
 		do {
diff --git a/drivers/dma/Kconfig b/drivers/dma/Kconfig
index c61a6eccf169..446687cc2334 100644
--- a/drivers/dma/Kconfig
+++ b/drivers/dma/Kconfig
@@ -89,14 +89,15 @@ config AT_HDMAC
 	  Support the Atmel AHB DMA controller.
 
 config FSL_DMA
-	tristate "Freescale Elo and Elo Plus DMA support"
+	tristate "Freescale Elo series DMA support"
 	depends on FSL_SOC
 	select DMA_ENGINE
 	select ASYNC_TX_ENABLE_CHANNEL_SWITCH
 	---help---
-	  Enable support for the Freescale Elo and Elo Plus DMA controllers.
-	  The Elo is the DMA controller on some 82xx and 83xx parts, and the
-	  Elo Plus is the DMA controller on 85xx and 86xx parts.
+	  Enable support for the Freescale Elo series DMA controllers.
+	  The Elo is the DMA controller on some mpc82xx and mpc83xx parts, the
+	  EloPlus is on mpc85xx and mpc86xx and Pxxx parts, and the Elo3 is on
+	  some Txxx and Bxxx parts.
 
 config MPC512X_DMA
 	tristate "Freescale MPC512x built-in DMA engine support"
@@ -313,7 +314,7 @@ config MMP_PDMA
 	depends on (ARCH_MMP || ARCH_PXA)
 	select DMA_ENGINE
 	help
-	  Support the MMP PDMA engine for PXA and MMP platfrom.
+	  Support the MMP PDMA engine for PXA and MMP platform.
 
 config DMA_JZ4740
 	tristate "JZ4740 DMA support"
diff --git a/drivers/dma/amba-pl08x.c b/drivers/dma/amba-pl08x.c
index e51a9832ef0d..16a2aa28f856 100644
--- a/drivers/dma/amba-pl08x.c
+++ b/drivers/dma/amba-pl08x.c
@@ -1164,42 +1164,12 @@ static void pl08x_free_txd(struct pl08x_driver_data *pl08x,
 	kfree(txd);
 }
 
-static void pl08x_unmap_buffers(struct pl08x_txd *txd)
-{
-	struct device *dev = txd->vd.tx.chan->device->dev;
-	struct pl08x_sg *dsg;
-
-	if (!(txd->vd.tx.flags & DMA_COMPL_SKIP_SRC_UNMAP)) {
-		if (txd->vd.tx.flags & DMA_COMPL_SRC_UNMAP_SINGLE)
-			list_for_each_entry(dsg, &txd->dsg_list, node)
-				dma_unmap_single(dev, dsg->src_addr, dsg->len,
-						DMA_TO_DEVICE);
-		else {
-			list_for_each_entry(dsg, &txd->dsg_list, node)
-				dma_unmap_page(dev, dsg->src_addr, dsg->len,
-						DMA_TO_DEVICE);
-		}
-	}
-	if (!(txd->vd.tx.flags & DMA_COMPL_SKIP_DEST_UNMAP)) {
-		if (txd->vd.tx.flags & DMA_COMPL_DEST_UNMAP_SINGLE)
-			list_for_each_entry(dsg, &txd->dsg_list, node)
-				dma_unmap_single(dev, dsg->dst_addr, dsg->len,
-						DMA_FROM_DEVICE);
-		else
-			list_for_each_entry(dsg, &txd->dsg_list, node)
-				dma_unmap_page(dev, dsg->dst_addr, dsg->len,
-						DMA_FROM_DEVICE);
-	}
-}
-
 static void pl08x_desc_free(struct virt_dma_desc *vd)
 {
 	struct pl08x_txd *txd = to_pl08x_txd(&vd->tx);
 	struct pl08x_dma_chan *plchan = to_pl08x_chan(vd->tx.chan);
 
-	if (!plchan->slave)
-		pl08x_unmap_buffers(txd);
-
+	dma_descriptor_unmap(txd);
 	if (!txd->done)
 		pl08x_release_mux(plchan);
 
@@ -1252,7 +1222,7 @@ static enum dma_status pl08x_dma_tx_status(struct dma_chan *chan,
 	size_t bytes = 0;
 
 	ret = dma_cookie_status(chan, cookie, txstate);
-	if (ret == DMA_SUCCESS)
+	if (ret == DMA_COMPLETE)
 		return ret;
 
 	/*
@@ -1267,7 +1237,7 @@ static enum dma_status pl08x_dma_tx_status(struct dma_chan *chan,
 
 	spin_lock_irqsave(&plchan->vc.lock, flags);
 	ret = dma_cookie_status(chan, cookie, txstate);
-	if (ret != DMA_SUCCESS) {
+	if (ret != DMA_COMPLETE) {
 		vd = vchan_find_desc(&plchan->vc, cookie);
 		if (vd) {
 			/* On the issued list, so hasn't been processed yet */
@@ -2138,8 +2108,7 @@ static int pl08x_probe(struct amba_device *adev, const struct amba_id *id)
 	writel(0x000000FF, pl08x->base + PL080_ERR_CLEAR);
 	writel(0x000000FF, pl08x->base + PL080_TC_CLEAR);
 
-	ret = request_irq(adev->irq[0], pl08x_irq, IRQF_DISABLED,
-			  DRIVER_NAME, pl08x);
+	ret = request_irq(adev->irq[0], pl08x_irq, 0, DRIVER_NAME, pl08x);
 	if (ret) {
 		dev_err(&adev->dev, "%s failed to request interrupt %d\n",
 			__func__, adev->irq[0]);
diff --git a/drivers/dma/at_hdmac.c b/drivers/dma/at_hdmac.c
index c787f38a186a..e2c04dc81e2a 100644
--- a/drivers/dma/at_hdmac.c
+++ b/drivers/dma/at_hdmac.c
@@ -344,31 +344,7 @@ atc_chain_complete(struct at_dma_chan *atchan, struct at_desc *desc)
 	/* move myself to free_list */
 	list_move(&desc->desc_node, &atchan->free_list);
 
-	/* unmap dma addresses (not on slave channels) */
-	if (!atchan->chan_common.private) {
-		struct device *parent = chan2parent(&atchan->chan_common);
-		if (!(txd->flags & DMA_COMPL_SKIP_DEST_UNMAP)) {
-			if (txd->flags & DMA_COMPL_DEST_UNMAP_SINGLE)
-				dma_unmap_single(parent,
-						desc->lli.daddr,
-						desc->len, DMA_FROM_DEVICE);
-			else
-				dma_unmap_page(parent,
-						desc->lli.daddr,
-						desc->len, DMA_FROM_DEVICE);
-		}
-		if (!(txd->flags & DMA_COMPL_SKIP_SRC_UNMAP)) {
-			if (txd->flags & DMA_COMPL_SRC_UNMAP_SINGLE)
-				dma_unmap_single(parent,
-						desc->lli.saddr,
-						desc->len, DMA_TO_DEVICE);
-			else
-				dma_unmap_page(parent,
-						desc->lli.saddr,
-						desc->len, DMA_TO_DEVICE);
-		}
-	}
-
+	dma_descriptor_unmap(txd);
 	/* for cyclic transfers,
 	 * no need to replay callback function while stopping */
 	if (!atc_chan_is_cyclic(atchan)) {
@@ -1102,7 +1078,7 @@ atc_tx_status(struct dma_chan *chan,
 	int bytes = 0;
 
 	ret = dma_cookie_status(chan, cookie, txstate);
-	if (ret == DMA_SUCCESS)
+	if (ret == DMA_COMPLETE)
 		return ret;
 	/*
 	 * There's no point calculating the residue if there's
diff --git a/drivers/dma/coh901318.c b/drivers/dma/coh901318.c
index 31011d2a26fc..3c6716e0b78e 100644
--- a/drivers/dma/coh901318.c
+++ b/drivers/dma/coh901318.c
@@ -2369,7 +2369,7 @@ coh901318_tx_status(struct dma_chan *chan, dma_cookie_t cookie,
 	enum dma_status ret;
 
 	ret = dma_cookie_status(chan, cookie, txstate);
-	if (ret == DMA_SUCCESS)
+	if (ret == DMA_COMPLETE)
 		return ret;
 
 	dma_set_residue(txstate, coh901318_get_bytes_left(chan));
@@ -2694,7 +2694,7 @@ static int __init coh901318_probe(struct platform_device *pdev)
 	if (irq < 0)
 		return irq;
 
-	err = devm_request_irq(&pdev->dev, irq, dma_irq_handler, IRQF_DISABLED,
+	err = devm_request_irq(&pdev->dev, irq, dma_irq_handler, 0,
 			       "coh901318", base);
 	if (err)
 		return err;
diff --git a/drivers/dma/cppi41.c b/drivers/dma/cppi41.c
index 7c82b92f9b16..c29dacff66fa 100644
--- a/drivers/dma/cppi41.c
+++ b/drivers/dma/cppi41.c
@@ -141,6 +141,9 @@ struct cppi41_dd {
 	const struct chan_queues *queues_rx;
 	const struct chan_queues *queues_tx;
 	struct chan_queues td_queue;
+
+	/* context for suspend/resume */
+	unsigned int dma_tdfdq;
 };
 
 #define FIST_COMPLETION_QUEUE	93
@@ -263,6 +266,15 @@ static u32 pd_trans_len(u32 val)
 	return val & ((1 << (DESC_LENGTH_BITS_NUM + 1)) - 1);
 }
 
+static u32 cppi41_pop_desc(struct cppi41_dd *cdd, unsigned queue_num)
+{
+	u32 desc;
+
+	desc = cppi_readl(cdd->qmgr_mem + QMGR_QUEUE_D(queue_num));
+	desc &= ~0x1f;
+	return desc;
+}
+
 static irqreturn_t cppi41_irq(int irq, void *data)
 {
 	struct cppi41_dd *cdd = data;
@@ -300,8 +312,7 @@ static irqreturn_t cppi41_irq(int irq, void *data)
 			q_num = __fls(val);
 			val &= ~(1 << q_num);
 			q_num += 32 * i;
-			desc = cppi_readl(cdd->qmgr_mem + QMGR_QUEUE_D(q_num));
-			desc &= ~0x1f;
+			desc = cppi41_pop_desc(cdd, q_num);
 			c = desc_to_chan(cdd, desc);
 			if (WARN_ON(!c)) {
 				pr_err("%s() q %d desc %08x\n", __func__,
@@ -353,7 +364,7 @@ static enum dma_status cppi41_dma_tx_status(struct dma_chan *chan,
 
 	/* lock */
 	ret = dma_cookie_status(chan, cookie, txstate);
-	if (txstate && ret == DMA_SUCCESS)
+	if (txstate && ret == DMA_COMPLETE)
 		txstate->residue = c->residue;
 	/* unlock */
 
@@ -517,15 +528,6 @@ static void cppi41_compute_td_desc(struct cppi41_desc *d)
 	d->pd0 = DESC_TYPE_TEARD << DESC_TYPE;
 }
 
-static u32 cppi41_pop_desc(struct cppi41_dd *cdd, unsigned queue_num)
-{
-	u32 desc;
-
-	desc = cppi_readl(cdd->qmgr_mem + QMGR_QUEUE_D(queue_num));
-	desc &= ~0x1f;
-	return desc;
-}
-
 static int cppi41_tear_down_chan(struct cppi41_channel *c)
 {
 	struct cppi41_dd *cdd = c->cdd;
@@ -561,36 +563,26 @@ static int cppi41_tear_down_chan(struct cppi41_channel *c)
 		c->td_retry = 100;
 	}
 
-	if (!c->td_seen) {
-		unsigned td_comp_queue;
+	if (!c->td_seen || !c->td_desc_seen) {
 
-		if (c->is_tx)
-			td_comp_queue =  cdd->td_queue.complete;
-		else
-			td_comp_queue =  c->q_comp_num;
+		desc_phys = cppi41_pop_desc(cdd, cdd->td_queue.complete);
+		if (!desc_phys)
+			desc_phys = cppi41_pop_desc(cdd, c->q_comp_num);
 
-		desc_phys = cppi41_pop_desc(cdd, td_comp_queue);
-		if (desc_phys) {
-			__iormb();
+		if (desc_phys == c->desc_phys) {
+			c->td_desc_seen = 1;
+
+		} else if (desc_phys == td_desc_phys) {
+			u32 pd0;
 
-			if (desc_phys == td_desc_phys) {
-				u32 pd0;
-				pd0 = td->pd0;
-				WARN_ON((pd0 >> DESC_TYPE) != DESC_TYPE_TEARD);
-				WARN_ON(!c->is_tx && !(pd0 & TD_DESC_IS_RX));
-				WARN_ON((pd0 & 0x1f) != c->port_num);
-			} else {
-				WARN_ON_ONCE(1);
-			}
-			c->td_seen = 1;
-		}
-	}
-	if (!c->td_desc_seen) {
-		desc_phys = cppi41_pop_desc(cdd, c->q_comp_num);
-		if (desc_phys) {
 			__iormb();
-			WARN_ON(c->desc_phys != desc_phys);
-			c->td_desc_seen = 1;
+			pd0 = td->pd0;
+			WARN_ON((pd0 >> DESC_TYPE) != DESC_TYPE_TEARD);
+			WARN_ON(!c->is_tx && !(pd0 & TD_DESC_IS_RX));
+			WARN_ON((pd0 & 0x1f) != c->port_num);
+			c->td_seen = 1;
+		} else if (desc_phys) {
+			WARN_ON_ONCE(1);
 		}
 	}
 	c->td_retry--;
@@ -609,7 +601,7 @@ static int cppi41_tear_down_chan(struct cppi41_channel *c)
 
 	WARN_ON(!c->td_retry);
 	if (!c->td_desc_seen) {
-		desc_phys = cppi_readl(cdd->qmgr_mem + QMGR_QUEUE_D(c->q_num));
+		desc_phys = cppi41_pop_desc(cdd, c->q_num);
 		WARN_ON(!desc_phys);
 	}
 
@@ -674,14 +666,14 @@ static void cleanup_chans(struct cppi41_dd *cdd)
 	}
 }
 
-static int cppi41_add_chans(struct platform_device *pdev, struct cppi41_dd *cdd)
+static int cppi41_add_chans(struct device *dev, struct cppi41_dd *cdd)
 {
 	struct cppi41_channel *cchan;
 	int i;
 	int ret;
 	u32 n_chans;
 
-	ret = of_property_read_u32(pdev->dev.of_node, "#dma-channels",
+	ret = of_property_read_u32(dev->of_node, "#dma-channels",
 			&n_chans);
 	if (ret)
 		return ret;
@@ -719,7 +711,7 @@ err:
 	return -ENOMEM;
 }
 
-static void purge_descs(struct platform_device *pdev, struct cppi41_dd *cdd)
+static void purge_descs(struct device *dev, struct cppi41_dd *cdd)
 {
 	unsigned int mem_decs;
 	int i;
@@ -731,7 +723,7 @@ static void purge_descs(struct platform_device *pdev, struct cppi41_dd *cdd)
 		cppi_writel(0, cdd->qmgr_mem + QMGR_MEMBASE(i));
 		cppi_writel(0, cdd->qmgr_mem + QMGR_MEMCTRL(i));
 
-		dma_free_coherent(&pdev->dev, mem_decs, cdd->cd,
+		dma_free_coherent(dev, mem_decs, cdd->cd,
 				cdd->descs_phys);
 	}
 }
@@ -741,19 +733,19 @@ static void disable_sched(struct cppi41_dd *cdd)
 	cppi_writel(0, cdd->sched_mem + DMA_SCHED_CTRL);
 }
 
-static void deinit_cpii41(struct platform_device *pdev, struct cppi41_dd *cdd)
+static void deinit_cppi41(struct device *dev, struct cppi41_dd *cdd)
 {
 	disable_sched(cdd);
 
-	purge_descs(pdev, cdd);
+	purge_descs(dev, cdd);
 
 	cppi_writel(0, cdd->qmgr_mem + QMGR_LRAM0_BASE);
 	cppi_writel(0, cdd->qmgr_mem + QMGR_LRAM0_BASE);
-	dma_free_coherent(&pdev->dev, QMGR_SCRATCH_SIZE, cdd->qmgr_scratch,
+	dma_free_coherent(dev, QMGR_SCRATCH_SIZE, cdd->qmgr_scratch,
 			cdd->scratch_phys);
 }
 
-static int init_descs(struct platform_device *pdev, struct cppi41_dd *cdd)
+static int init_descs(struct device *dev, struct cppi41_dd *cdd)
 {
 	unsigned int desc_size;
 	unsigned int mem_decs;
@@ -777,7 +769,7 @@ static int init_descs(struct platform_device *pdev, struct cppi41_dd *cdd)
 		reg |= ilog2(ALLOC_DECS_NUM) - 5;
 
 		BUILD_BUG_ON(DESCS_AREAS != 1);
-		cdd->cd = dma_alloc_coherent(&pdev->dev, mem_decs,
+		cdd->cd = dma_alloc_coherent(dev, mem_decs,
 				&cdd->descs_phys, GFP_KERNEL);
 		if (!cdd->cd)
 			return -ENOMEM;
@@ -813,12 +805,12 @@ static void init_sched(struct cppi41_dd *cdd)
 	cppi_writel(reg, cdd->sched_mem + DMA_SCHED_CTRL);
 }
 
-static int init_cppi41(struct platform_device *pdev, struct cppi41_dd *cdd)
+static int init_cppi41(struct device *dev, struct cppi41_dd *cdd)
 {
 	int ret;
 
 	BUILD_BUG_ON(QMGR_SCRATCH_SIZE > ((1 << 14) - 1));
-	cdd->qmgr_scratch = dma_alloc_coherent(&pdev->dev, QMGR_SCRATCH_SIZE,
+	cdd->qmgr_scratch = dma_alloc_coherent(dev, QMGR_SCRATCH_SIZE,
 			&cdd->scratch_phys, GFP_KERNEL);
 	if (!cdd->qmgr_scratch)
 		return -ENOMEM;
@@ -827,7 +819,7 @@ static int init_cppi41(struct platform_device *pdev, struct cppi41_dd *cdd)
 	cppi_writel(QMGR_SCRATCH_SIZE, cdd->qmgr_mem + QMGR_LRAM_SIZE);
 	cppi_writel(0, cdd->qmgr_mem + QMGR_LRAM1_BASE);
 
-	ret = init_descs(pdev, cdd);
+	ret = init_descs(dev, cdd);
 	if (ret)
 		goto err_td;
 
@@ -835,7 +827,7 @@ static int init_cppi41(struct platform_device *pdev, struct cppi41_dd *cdd)
 	init_sched(cdd);
 	return 0;
 err_td:
-	deinit_cpii41(pdev, cdd);
+	deinit_cppi41(dev, cdd);
 	return ret;
 }
 
@@ -914,11 +906,11 @@ static const struct of_device_id cppi41_dma_ids[] = {
 };
 MODULE_DEVICE_TABLE(of, cppi41_dma_ids);
 
-static const struct cppi_glue_infos *get_glue_info(struct platform_device *pdev)
+static const struct cppi_glue_infos *get_glue_info(struct device *dev)
 {
 	const struct of_device_id *of_id;
 
-	of_id = of_match_node(cppi41_dma_ids, pdev->dev.of_node);
+	of_id = of_match_node(cppi41_dma_ids, dev->of_node);
 	if (!of_id)
 		return NULL;
 	return of_id->data;
@@ -927,11 +919,12 @@ static const struct cppi_glue_infos *get_glue_info(struct platform_device *pdev)
 static int cppi41_dma_probe(struct platform_device *pdev)
 {
 	struct cppi41_dd *cdd;
+	struct device *dev = &pdev->dev;
 	const struct cppi_glue_infos *glue_info;
 	int irq;
 	int ret;
 
-	glue_info = get_glue_info(pdev);
+	glue_info = get_glue_info(dev);
 	if (!glue_info)
 		return -EINVAL;
 
@@ -946,14 +939,14 @@ static int cppi41_dma_probe(struct platform_device *pdev)
 	cdd->ddev.device_issue_pending = cppi41_dma_issue_pending;
 	cdd->ddev.device_prep_slave_sg = cppi41_dma_prep_slave_sg;
 	cdd->ddev.device_control = cppi41_dma_control;
-	cdd->ddev.dev = &pdev->dev;
+	cdd->ddev.dev = dev;
 	INIT_LIST_HEAD(&cdd->ddev.channels);
 	cpp41_dma_info.dma_cap = cdd->ddev.cap_mask;
 
-	cdd->usbss_mem = of_iomap(pdev->dev.of_node, 0);
-	cdd->ctrl_mem = of_iomap(pdev->dev.of_node, 1);
-	cdd->sched_mem = of_iomap(pdev->dev.of_node, 2);
-	cdd->qmgr_mem = of_iomap(pdev->dev.of_node, 3);
+	cdd->usbss_mem = of_iomap(dev->of_node, 0);
+	cdd->ctrl_mem = of_iomap(dev->of_node, 1);
+	cdd->sched_mem = of_iomap(dev->of_node, 2);
+	cdd->qmgr_mem = of_iomap(dev->of_node, 3);
 
 	if (!cdd->usbss_mem || !cdd->ctrl_mem || !cdd->sched_mem ||
 			!cdd->qmgr_mem) {
@@ -961,31 +954,31 @@ static int cppi41_dma_probe(struct platform_device *pdev)
 		goto err_remap;
 	}
 
-	pm_runtime_enable(&pdev->dev);
-	ret = pm_runtime_get_sync(&pdev->dev);
-	if (ret)
+	pm_runtime_enable(dev);
+	ret = pm_runtime_get_sync(dev);
+	if (ret < 0)
 		goto err_get_sync;
 
 	cdd->queues_rx = glue_info->queues_rx;
 	cdd->queues_tx = glue_info->queues_tx;
 	cdd->td_queue = glue_info->td_queue;
 
-	ret = init_cppi41(pdev, cdd);
+	ret = init_cppi41(dev, cdd);
 	if (ret)
 		goto err_init_cppi;
 
-	ret = cppi41_add_chans(pdev, cdd);
+	ret = cppi41_add_chans(dev, cdd);
 	if (ret)
 		goto err_chans;
 
-	irq = irq_of_parse_and_map(pdev->dev.of_node, 0);
+	irq = irq_of_parse_and_map(dev->of_node, 0);
 	if (!irq)
 		goto err_irq;
 
 	cppi_writel(USBSS_IRQ_PD_COMP, cdd->usbss_mem + USBSS_IRQ_ENABLER);
 
 	ret = request_irq(irq, glue_info->isr, IRQF_SHARED,
-			dev_name(&pdev->dev), cdd);
+			dev_name(dev), cdd);
 	if (ret)
 		goto err_irq;
 	cdd->irq = irq;
@@ -994,7 +987,7 @@ static int cppi41_dma_probe(struct platform_device *pdev)
 	if (ret)
 		goto err_dma_reg;
 
-	ret = of_dma_controller_register(pdev->dev.of_node,
+	ret = of_dma_controller_register(dev->of_node,
 			cppi41_dma_xlate, &cpp41_dma_info);
 	if (ret)
 		goto err_of;
@@ -1009,11 +1002,11 @@ err_irq:
 	cppi_writel(0, cdd->usbss_mem + USBSS_IRQ_CLEARR);
 	cleanup_chans(cdd);
 err_chans:
-	deinit_cpii41(pdev, cdd);
+	deinit_cppi41(dev, cdd);
 err_init_cppi:
-	pm_runtime_put(&pdev->dev);
+	pm_runtime_put(dev);
 err_get_sync:
-	pm_runtime_disable(&pdev->dev);
+	pm_runtime_disable(dev);
 	iounmap(cdd->usbss_mem);
 	iounmap(cdd->ctrl_mem);
 	iounmap(cdd->sched_mem);
@@ -1033,7 +1026,7 @@ static int cppi41_dma_remove(struct platform_device *pdev)
 	cppi_writel(0, cdd->usbss_mem + USBSS_IRQ_CLEARR);
 	free_irq(cdd->irq, cdd);
 	cleanup_chans(cdd);
-	deinit_cpii41(pdev, cdd);
+	deinit_cppi41(&pdev->dev, cdd);
 	iounmap(cdd->usbss_mem);
 	iounmap(cdd->ctrl_mem);
 	iounmap(cdd->sched_mem);
@@ -1044,12 +1037,53 @@ static int cppi41_dma_remove(struct platform_device *pdev)
 	return 0;
 }
 
+#ifdef CONFIG_PM_SLEEP
+static int cppi41_suspend(struct device *dev)
+{
+	struct cppi41_dd *cdd = dev_get_drvdata(dev);
+
+	cdd->dma_tdfdq = cppi_readl(cdd->ctrl_mem + DMA_TDFDQ);
+	cppi_writel(0, cdd->usbss_mem + USBSS_IRQ_CLEARR);
+	disable_sched(cdd);
+
+	return 0;
+}
+
+static int cppi41_resume(struct device *dev)
+{
+	struct cppi41_dd *cdd = dev_get_drvdata(dev);
+	struct cppi41_channel *c;
+	int i;
+
+	for (i = 0; i < DESCS_AREAS; i++)
+		cppi_writel(cdd->descs_phys, cdd->qmgr_mem + QMGR_MEMBASE(i));
+
+	list_for_each_entry(c, &cdd->ddev.channels, chan.device_node)
+		if (!c->is_tx)
+			cppi_writel(c->q_num, c->gcr_reg + RXHPCRA0);
+
+	init_sched(cdd);
+
+	cppi_writel(cdd->dma_tdfdq, cdd->ctrl_mem + DMA_TDFDQ);
+	cppi_writel(cdd->scratch_phys, cdd->qmgr_mem + QMGR_LRAM0_BASE);
+	cppi_writel(QMGR_SCRATCH_SIZE, cdd->qmgr_mem + QMGR_LRAM_SIZE);
+	cppi_writel(0, cdd->qmgr_mem + QMGR_LRAM1_BASE);
+
+	cppi_writel(USBSS_IRQ_PD_COMP, cdd->usbss_mem + USBSS_IRQ_ENABLER);
+
+	return 0;
+}
+#endif
+
+static SIMPLE_DEV_PM_OPS(cppi41_pm_ops, cppi41_suspend, cppi41_resume);
+
 static struct platform_driver cpp41_dma_driver = {
 	.probe  = cppi41_dma_probe,
 	.remove = cppi41_dma_remove,
 	.driver = {
 		.name = "cppi41-dma-engine",
 		.owner = THIS_MODULE,
+		.pm = &cppi41_pm_ops,
 		.of_match_table = of_match_ptr(cppi41_dma_ids),
 	},
 };
diff --git a/drivers/dma/dma-jz4740.c b/drivers/dma/dma-jz4740.c
index b0c0c8268d42..94c380f07538 100644
--- a/drivers/dma/dma-jz4740.c
+++ b/drivers/dma/dma-jz4740.c
@@ -491,7 +491,7 @@ static enum dma_status jz4740_dma_tx_status(struct dma_chan *c,
 	unsigned long flags;
 
 	status = dma_cookie_status(c, cookie, state);
-	if (status == DMA_SUCCESS || !state)
+	if (status == DMA_COMPLETE || !state)
 		return status;
 
 	spin_lock_irqsave(&chan->vchan.lock, flags);
diff --git a/drivers/dma/dmaengine.c b/drivers/dma/dmaengine.c
index 9162ac80c18f..ea806bdc12ef 100644
--- a/drivers/dma/dmaengine.c
+++ b/drivers/dma/dmaengine.c
@@ -65,6 +65,7 @@
 #include <linux/acpi.h>
 #include <linux/acpi_dma.h>
 #include <linux/of_dma.h>
+#include <linux/mempool.h>
 
 static DEFINE_MUTEX(dma_list_mutex);
 static DEFINE_IDR(dma_idr);
@@ -901,98 +902,132 @@ void dma_async_device_unregister(struct dma_device *device)
 }
 EXPORT_SYMBOL(dma_async_device_unregister);
 
-/**
- * dma_async_memcpy_buf_to_buf - offloaded copy between virtual addresses
- * @chan: DMA channel to offload copy to
- * @dest: destination address (virtual)
- * @src: source address (virtual)
- * @len: length
- *
- * Both @dest and @src must be mappable to a bus address according to the
- * DMA mapping API rules for streaming mappings.
- * Both @dest and @src must stay memory resident (kernel memory or locked
- * user space pages).
- */
-dma_cookie_t
-dma_async_memcpy_buf_to_buf(struct dma_chan *chan, void *dest,
-			void *src, size_t len)
-{
-	struct dma_device *dev = chan->device;
-	struct dma_async_tx_descriptor *tx;
-	dma_addr_t dma_dest, dma_src;
-	dma_cookie_t cookie;
-	unsigned long flags;
+struct dmaengine_unmap_pool {
+	struct kmem_cache *cache;
+	const char *name;
+	mempool_t *pool;
+	size_t size;
+};
 
-	dma_src = dma_map_single(dev->dev, src, len, DMA_TO_DEVICE);
-	dma_dest = dma_map_single(dev->dev, dest, len, DMA_FROM_DEVICE);
-	flags = DMA_CTRL_ACK |
-		DMA_COMPL_SRC_UNMAP_SINGLE |
-		DMA_COMPL_DEST_UNMAP_SINGLE;
-	tx = dev->device_prep_dma_memcpy(chan, dma_dest, dma_src, len, flags);
+#define __UNMAP_POOL(x) { .size = x, .name = "dmaengine-unmap-" __stringify(x) }
+static struct dmaengine_unmap_pool unmap_pool[] = {
+	__UNMAP_POOL(2),
+	#if IS_ENABLED(CONFIG_ASYNC_TX_DMA)
+	__UNMAP_POOL(16),
+	__UNMAP_POOL(128),
+	__UNMAP_POOL(256),
+	#endif
+};
 
-	if (!tx) {
-		dma_unmap_single(dev->dev, dma_src, len, DMA_TO_DEVICE);
-		dma_unmap_single(dev->dev, dma_dest, len, DMA_FROM_DEVICE);
-		return -ENOMEM;
+static struct dmaengine_unmap_pool *__get_unmap_pool(int nr)
+{
+	int order = get_count_order(nr);
+
+	switch (order) {
+	case 0 ... 1:
+		return &unmap_pool[0];
+	case 2 ... 4:
+		return &unmap_pool[1];
+	case 5 ... 7:
+		return &unmap_pool[2];
+	case 8:
+		return &unmap_pool[3];
+	default:
+		BUG();
+		return NULL;
 	}
+}
 
-	tx->callback = NULL;
-	cookie = tx->tx_submit(tx);
+static void dmaengine_unmap(struct kref *kref)
+{
+	struct dmaengine_unmap_data *unmap = container_of(kref, typeof(*unmap), kref);
+	struct device *dev = unmap->dev;
+	int cnt, i;
+
+	cnt = unmap->to_cnt;
+	for (i = 0; i < cnt; i++)
+		dma_unmap_page(dev, unmap->addr[i], unmap->len,
+			       DMA_TO_DEVICE);
+	cnt += unmap->from_cnt;
+	for (; i < cnt; i++)
+		dma_unmap_page(dev, unmap->addr[i], unmap->len,
+			       DMA_FROM_DEVICE);
+	cnt += unmap->bidi_cnt;
+	for (; i < cnt; i++) {
+		if (unmap->addr[i] == 0)
+			continue;
+		dma_unmap_page(dev, unmap->addr[i], unmap->len,
+			       DMA_BIDIRECTIONAL);
+	}
+	mempool_free(unmap, __get_unmap_pool(cnt)->pool);
+}
 
-	preempt_disable();
-	__this_cpu_add(chan->local->bytes_transferred, len);
-	__this_cpu_inc(chan->local->memcpy_count);
-	preempt_enable();
+void dmaengine_unmap_put(struct dmaengine_unmap_data *unmap)
+{
+	if (unmap)
+		kref_put(&unmap->kref, dmaengine_unmap);
+}
+EXPORT_SYMBOL_GPL(dmaengine_unmap_put);
 
-	return cookie;
+static void dmaengine_destroy_unmap_pool(void)
+{
+	int i;
+
+	for (i = 0; i < ARRAY_SIZE(unmap_pool); i++) {
+		struct dmaengine_unmap_pool *p = &unmap_pool[i];
+
+		if (p->pool)
+			mempool_destroy(p->pool);
+		p->pool = NULL;
+		if (p->cache)
+			kmem_cache_destroy(p->cache);
+		p->cache = NULL;
+	}
 }
-EXPORT_SYMBOL(dma_async_memcpy_buf_to_buf);
 
-/**
- * dma_async_memcpy_buf_to_pg - offloaded copy from address to page
- * @chan: DMA channel to offload copy to
- * @page: destination page
- * @offset: offset in page to copy to
- * @kdata: source address (virtual)
- * @len: length
- *
- * Both @page/@offset and @kdata must be mappable to a bus address according
- * to the DMA mapping API rules for streaming mappings.
- * Both @page/@offset and @kdata must stay memory resident (kernel memory or
- * locked user space pages)
- */
-dma_cookie_t
-dma_async_memcpy_buf_to_pg(struct dma_chan *chan, struct page *page,
-			unsigned int offset, void *kdata, size_t len)
+static int __init dmaengine_init_unmap_pool(void)
 {
-	struct dma_device *dev = chan->device;
-	struct dma_async_tx_descriptor *tx;
-	dma_addr_t dma_dest, dma_src;
-	dma_cookie_t cookie;
-	unsigned long flags;
+	int i;
 
-	dma_src = dma_map_single(dev->dev, kdata, len, DMA_TO_DEVICE);
-	dma_dest = dma_map_page(dev->dev, page, offset, len, DMA_FROM_DEVICE);
-	flags = DMA_CTRL_ACK | DMA_COMPL_SRC_UNMAP_SINGLE;
-	tx = dev->device_prep_dma_memcpy(chan, dma_dest, dma_src, len, flags);
+	for (i = 0; i < ARRAY_SIZE(unmap_pool); i++) {
+		struct dmaengine_unmap_pool *p = &unmap_pool[i];
+		size_t size;
 
-	if (!tx) {
-		dma_unmap_single(dev->dev, dma_src, len, DMA_TO_DEVICE);
-		dma_unmap_page(dev->dev, dma_dest, len, DMA_FROM_DEVICE);
-		return -ENOMEM;
+		size = sizeof(struct dmaengine_unmap_data) +
+		       sizeof(dma_addr_t) * p->size;
+
+		p->cache = kmem_cache_create(p->name, size, 0,
+					     SLAB_HWCACHE_ALIGN, NULL);
+		if (!p->cache)
+			break;
+		p->pool = mempool_create_slab_pool(1, p->cache);
+		if (!p->pool)
+			break;
 	}
 
-	tx->callback = NULL;
-	cookie = tx->tx_submit(tx);
+	if (i == ARRAY_SIZE(unmap_pool))
+		return 0;
 
-	preempt_disable();
-	__this_cpu_add(chan->local->bytes_transferred, len);
-	__this_cpu_inc(chan->local->memcpy_count);
-	preempt_enable();
+	dmaengine_destroy_unmap_pool();
+	return -ENOMEM;
+}
 
-	return cookie;
+struct dmaengine_unmap_data *
+dmaengine_get_unmap_data(struct device *dev, int nr, gfp_t flags)
+{
+	struct dmaengine_unmap_data *unmap;
+
+	unmap = mempool_alloc(__get_unmap_pool(nr)->pool, flags);
+	if (!unmap)
+		return NULL;
+
+	memset(unmap, 0, sizeof(*unmap));
+	kref_init(&unmap->kref);
+	unmap->dev = dev;
+
+	return unmap;
 }
-EXPORT_SYMBOL(dma_async_memcpy_buf_to_pg);
+EXPORT_SYMBOL(dmaengine_get_unmap_data);
 
 /**
  * dma_async_memcpy_pg_to_pg - offloaded copy from page to page
@@ -1015,24 +1050,33 @@ dma_async_memcpy_pg_to_pg(struct dma_chan *chan, struct page *dest_pg,
 {
 	struct dma_device *dev = chan->device;
 	struct dma_async_tx_descriptor *tx;
-	dma_addr_t dma_dest, dma_src;
+	struct dmaengine_unmap_data *unmap;
 	dma_cookie_t cookie;
 	unsigned long flags;
 
-	dma_src = dma_map_page(dev->dev, src_pg, src_off, len, DMA_TO_DEVICE);
-	dma_dest = dma_map_page(dev->dev, dest_pg, dest_off, len,
-				DMA_FROM_DEVICE);
+	unmap = dmaengine_get_unmap_data(dev->dev, 2, GFP_NOIO);
+	if (!unmap)
+		return -ENOMEM;
+
+	unmap->to_cnt = 1;
+	unmap->from_cnt = 1;
+	unmap->addr[0] = dma_map_page(dev->dev, src_pg, src_off, len,
+				      DMA_TO_DEVICE);
+	unmap->addr[1] = dma_map_page(dev->dev, dest_pg, dest_off, len,
+				      DMA_FROM_DEVICE);
+	unmap->len = len;
 	flags = DMA_CTRL_ACK;
-	tx = dev->device_prep_dma_memcpy(chan, dma_dest, dma_src, len, flags);
+	tx = dev->device_prep_dma_memcpy(chan, unmap->addr[1], unmap->addr[0],
+					 len, flags);
 
 	if (!tx) {
-		dma_unmap_page(dev->dev, dma_src, len, DMA_TO_DEVICE);
-		dma_unmap_page(dev->dev, dma_dest, len, DMA_FROM_DEVICE);
+		dmaengine_unmap_put(unmap);
 		return -ENOMEM;
 	}
 
-	tx->callback = NULL;
+	dma_set_unmap(tx, unmap);
 	cookie = tx->tx_submit(tx);
+	dmaengine_unmap_put(unmap);
 
 	preempt_disable();
 	__this_cpu_add(chan->local->bytes_transferred, len);
@@ -1043,6 +1087,52 @@ dma_async_memcpy_pg_to_pg(struct dma_chan *chan, struct page *dest_pg,
 }
 EXPORT_SYMBOL(dma_async_memcpy_pg_to_pg);
 
+/**
+ * dma_async_memcpy_buf_to_buf - offloaded copy between virtual addresses
+ * @chan: DMA channel to offload copy to
+ * @dest: destination address (virtual)
+ * @src: source address (virtual)
+ * @len: length
+ *
+ * Both @dest and @src must be mappable to a bus address according to the
+ * DMA mapping API rules for streaming mappings.
+ * Both @dest and @src must stay memory resident (kernel memory or locked
+ * user space pages).
+ */
+dma_cookie_t
+dma_async_memcpy_buf_to_buf(struct dma_chan *chan, void *dest,
+			    void *src, size_t len)
+{
+	return dma_async_memcpy_pg_to_pg(chan, virt_to_page(dest),
+					 (unsigned long) dest & ~PAGE_MASK,
+					 virt_to_page(src),
+					 (unsigned long) src & ~PAGE_MASK, len);
+}
+EXPORT_SYMBOL(dma_async_memcpy_buf_to_buf);
+
+/**
+ * dma_async_memcpy_buf_to_pg - offloaded copy from address to page
+ * @chan: DMA channel to offload copy to
+ * @page: destination page
+ * @offset: offset in page to copy to
+ * @kdata: source address (virtual)
+ * @len: length
+ *
+ * Both @page/@offset and @kdata must be mappable to a bus address according
+ * to the DMA mapping API rules for streaming mappings.
+ * Both @page/@offset and @kdata must stay memory resident (kernel memory or
+ * locked user space pages)
+ */
+dma_cookie_t
+dma_async_memcpy_buf_to_pg(struct dma_chan *chan, struct page *page,
+			   unsigned int offset, void *kdata, size_t len)
+{
+	return dma_async_memcpy_pg_to_pg(chan, page, offset,
+					 virt_to_page(kdata),
+					 (unsigned long) kdata & ~PAGE_MASK, len);
+}
+EXPORT_SYMBOL(dma_async_memcpy_buf_to_pg);
+
 void dma_async_tx_descriptor_init(struct dma_async_tx_descriptor *tx,
 	struct dma_chan *chan)
 {
@@ -1062,7 +1152,7 @@ dma_wait_for_async_tx(struct dma_async_tx_descriptor *tx)
 	unsigned long dma_sync_wait_timeout = jiffies + msecs_to_jiffies(5000);
 
 	if (!tx)
-		return DMA_SUCCESS;
+		return DMA_COMPLETE;
 
 	while (tx->cookie == -EBUSY) {
 		if (time_after_eq(jiffies, dma_sync_wait_timeout)) {
@@ -1116,6 +1206,10 @@ EXPORT_SYMBOL_GPL(dma_run_dependencies);
 
 static int __init dma_bus_init(void)
 {
+	int err = dmaengine_init_unmap_pool();
+
+	if (err)
+		return err;
 	return class_register(&dma_devclass);
 }
 arch_initcall(dma_bus_init);
diff --git a/drivers/dma/dmatest.c b/drivers/dma/dmatest.c
index 92f796cdc6ab..20f9a3aaf926 100644
--- a/drivers/dma/dmatest.c
+++ b/drivers/dma/dmatest.c
@@ -8,6 +8,8 @@
  * it under the terms of the GNU General Public License version 2 as
  * published by the Free Software Foundation.
  */
+#define pr_fmt(fmt) KBUILD_MODNAME ": " fmt
+
 #include <linux/delay.h>
 #include <linux/dma-mapping.h>
 #include <linux/dmaengine.h>
@@ -19,10 +21,6 @@
 #include <linux/random.h>
 #include <linux/slab.h>
 #include <linux/wait.h>
-#include <linux/ctype.h>
-#include <linux/debugfs.h>
-#include <linux/uaccess.h>
-#include <linux/seq_file.h>
 
 static unsigned int test_buf_size = 16384;
 module_param(test_buf_size, uint, S_IRUGO | S_IWUSR);
@@ -68,92 +66,13 @@ module_param(timeout, uint, S_IRUGO | S_IWUSR);
 MODULE_PARM_DESC(timeout, "Transfer Timeout in msec (default: 3000), "
 		 "Pass -1 for infinite timeout");
 
-/* Maximum amount of mismatched bytes in buffer to print */
-#define MAX_ERROR_COUNT		32
-
-/*
- * Initialization patterns. All bytes in the source buffer has bit 7
- * set, all bytes in the destination buffer has bit 7 cleared.
- *
- * Bit 6 is set for all bytes which are to be copied by the DMA
- * engine. Bit 5 is set for all bytes which are to be overwritten by
- * the DMA engine.
- *
- * The remaining bits are the inverse of a counter which increments by
- * one for each byte address.
- */
-#define PATTERN_SRC		0x80
-#define PATTERN_DST		0x00
-#define PATTERN_COPY		0x40
-#define PATTERN_OVERWRITE	0x20
-#define PATTERN_COUNT_MASK	0x1f
-
-enum dmatest_error_type {
-	DMATEST_ET_OK,
-	DMATEST_ET_MAP_SRC,
-	DMATEST_ET_MAP_DST,
-	DMATEST_ET_PREP,
-	DMATEST_ET_SUBMIT,
-	DMATEST_ET_TIMEOUT,
-	DMATEST_ET_DMA_ERROR,
-	DMATEST_ET_DMA_IN_PROGRESS,
-	DMATEST_ET_VERIFY,
-	DMATEST_ET_VERIFY_BUF,
-};
-
-struct dmatest_verify_buffer {
-	unsigned int	index;
-	u8		expected;
-	u8		actual;
-};
-
-struct dmatest_verify_result {
-	unsigned int			error_count;
-	struct dmatest_verify_buffer	data[MAX_ERROR_COUNT];
-	u8				pattern;
-	bool				is_srcbuf;
-};
-
-struct dmatest_thread_result {
-	struct list_head	node;
-	unsigned int		n;
-	unsigned int		src_off;
-	unsigned int		dst_off;
-	unsigned int		len;
-	enum dmatest_error_type	type;
-	union {
-		unsigned long			data;
-		dma_cookie_t			cookie;
-		enum dma_status			status;
-		int				error;
-		struct dmatest_verify_result	*vr;
-	};
-};
-
-struct dmatest_result {
-	struct list_head	node;
-	char			*name;
-	struct list_head	results;
-};
-
-struct dmatest_info;
-
-struct dmatest_thread {
-	struct list_head	node;
-	struct dmatest_info	*info;
-	struct task_struct	*task;
-	struct dma_chan		*chan;
-	u8			**srcs;
-	u8			**dsts;
-	enum dma_transaction_type type;
-	bool			done;
-};
+static bool noverify;
+module_param(noverify, bool, S_IRUGO | S_IWUSR);
+MODULE_PARM_DESC(noverify, "Disable random data setup and verification");
 
-struct dmatest_chan {
-	struct list_head	node;
-	struct dma_chan		*chan;
-	struct list_head	threads;
-};
+static bool verbose;
+module_param(verbose, bool, S_IRUGO | S_IWUSR);
+MODULE_PARM_DESC(verbose, "Enable \"success\" result messages (default: off)");
 
 /**
  * struct dmatest_params - test parameters.
@@ -177,6 +96,7 @@ struct dmatest_params {
 	unsigned int	xor_sources;
 	unsigned int	pq_sources;
 	int		timeout;
+	bool		noverify;
 };
 
 /**
@@ -184,7 +104,7 @@ struct dmatest_params {
  * @params:		test parameters
  * @lock:		access protection to the fields of this structure
  */
-struct dmatest_info {
+static struct dmatest_info {
 	/* Test parameters */
 	struct dmatest_params	params;
 
@@ -192,16 +112,95 @@ struct dmatest_info {
 	struct list_head	channels;
 	unsigned int		nr_channels;
 	struct mutex		lock;
+	bool			did_init;
+} test_info = {
+	.channels = LIST_HEAD_INIT(test_info.channels),
+	.lock = __MUTEX_INITIALIZER(test_info.lock),
+};
+
+static int dmatest_run_set(const char *val, const struct kernel_param *kp);
+static int dmatest_run_get(char *val, const struct kernel_param *kp);
+static struct kernel_param_ops run_ops = {
+	.set = dmatest_run_set,
+	.get = dmatest_run_get,
+};
+static bool dmatest_run;
+module_param_cb(run, &run_ops, &dmatest_run, S_IRUGO | S_IWUSR);
+MODULE_PARM_DESC(run, "Run the test (default: false)");
+
+/* Maximum amount of mismatched bytes in buffer to print */
+#define MAX_ERROR_COUNT		32
+
+/*
+ * Initialization patterns. All bytes in the source buffer has bit 7
+ * set, all bytes in the destination buffer has bit 7 cleared.
+ *
+ * Bit 6 is set for all bytes which are to be copied by the DMA
+ * engine. Bit 5 is set for all bytes which are to be overwritten by
+ * the DMA engine.
+ *
+ * The remaining bits are the inverse of a counter which increments by
+ * one for each byte address.
+ */
+#define PATTERN_SRC		0x80
+#define PATTERN_DST		0x00
+#define PATTERN_COPY		0x40
+#define PATTERN_OVERWRITE	0x20
+#define PATTERN_COUNT_MASK	0x1f
 
-	/* debugfs related stuff */
-	struct dentry		*root;
+struct dmatest_thread {
+	struct list_head	node;
+	struct dmatest_info	*info;
+	struct task_struct	*task;
+	struct dma_chan		*chan;
+	u8			**srcs;
+	u8			**dsts;
+	enum dma_transaction_type type;
+	bool			done;
+};
 
-	/* Test results */
-	struct list_head	results;
-	struct mutex		results_lock;
+struct dmatest_chan {
+	struct list_head	node;
+	struct dma_chan		*chan;
+	struct list_head	threads;
 };
 
-static struct dmatest_info test_info;
+static DECLARE_WAIT_QUEUE_HEAD(thread_wait);
+static bool wait;
+
+static bool is_threaded_test_run(struct dmatest_info *info)
+{
+	struct dmatest_chan *dtc;
+
+	list_for_each_entry(dtc, &info->channels, node) {
+		struct dmatest_thread *thread;
+
+		list_for_each_entry(thread, &dtc->threads, node) {
+			if (!thread->done)
+				return true;
+		}
+	}
+
+	return false;
+}
+
+static int dmatest_wait_get(char *val, const struct kernel_param *kp)
+{
+	struct dmatest_info *info = &test_info;
+	struct dmatest_params *params = &info->params;
+
+	if (params->iterations)
+		wait_event(thread_wait, !is_threaded_test_run(info));
+	wait = true;
+	return param_get_bool(val, kp);
+}
+
+static struct kernel_param_ops wait_ops = {
+	.get = dmatest_wait_get,
+	.set = param_set_bool,
+};
+module_param_cb(wait, &wait_ops, &wait, S_IRUGO);
+MODULE_PARM_DESC(wait, "Wait for tests to complete (default: false)");
 
 static bool dmatest_match_channel(struct dmatest_params *params,
 		struct dma_chan *chan)
@@ -223,7 +222,7 @@ static unsigned long dmatest_random(void)
 {
 	unsigned long buf;
 
-	get_random_bytes(&buf, sizeof(buf));
+	prandom_bytes(&buf, sizeof(buf));
 	return buf;
 }
 
@@ -262,9 +261,31 @@ static void dmatest_init_dsts(u8 **bufs, unsigned int start, unsigned int len,
 	}
 }
 
-static unsigned int dmatest_verify(struct dmatest_verify_result *vr, u8 **bufs,
-		unsigned int start, unsigned int end, unsigned int counter,
-		u8 pattern, bool is_srcbuf)
+static void dmatest_mismatch(u8 actual, u8 pattern, unsigned int index,
+		unsigned int counter, bool is_srcbuf)
+{
+	u8		diff = actual ^ pattern;
+	u8		expected = pattern | (~counter & PATTERN_COUNT_MASK);
+	const char	*thread_name = current->comm;
+
+	if (is_srcbuf)
+		pr_warn("%s: srcbuf[0x%x] overwritten! Expected %02x, got %02x\n",
+			thread_name, index, expected, actual);
+	else if ((pattern & PATTERN_COPY)
+			&& (diff & (PATTERN_COPY | PATTERN_OVERWRITE)))
+		pr_warn("%s: dstbuf[0x%x] not copied! Expected %02x, got %02x\n",
+			thread_name, index, expected, actual);
+	else if (diff & PATTERN_SRC)
+		pr_warn("%s: dstbuf[0x%x] was copied! Expected %02x, got %02x\n",
+			thread_name, index, expected, actual);
+	else
+		pr_warn("%s: dstbuf[0x%x] mismatch! Expected %02x, got %02x\n",
+			thread_name, index, expected, actual);
+}
+
+static unsigned int dmatest_verify(u8 **bufs, unsigned int start,
+		unsigned int end, unsigned int counter, u8 pattern,
+		bool is_srcbuf)
 {
 	unsigned int i;
 	unsigned int error_count = 0;
@@ -272,7 +293,6 @@ static unsigned int dmatest_verify(struct dmatest_verify_result *vr, u8 **bufs,
 	u8 expected;
 	u8 *buf;
 	unsigned int counter_orig = counter;
-	struct dmatest_verify_buffer *vb;
 
 	for (; (buf = *bufs); bufs++) {
 		counter = counter_orig;
@@ -280,12 +300,9 @@ static unsigned int dmatest_verify(struct dmatest_verify_result *vr, u8 **bufs,
 			actual = buf[i];
 			expected = pattern | (~counter & PATTERN_COUNT_MASK);
 			if (actual != expected) {
-				if (error_count < MAX_ERROR_COUNT && vr) {
-					vb = &vr->data[error_count];
-					vb->index = i;
-					vb->expected = expected;
-					vb->actual = actual;
-				}
+				if (error_count < MAX_ERROR_COUNT)
+					dmatest_mismatch(actual, pattern, i,
+							 counter, is_srcbuf);
 				error_count++;
 			}
 			counter++;
@@ -293,7 +310,7 @@ static unsigned int dmatest_verify(struct dmatest_verify_result *vr, u8 **bufs,
 	}
 
 	if (error_count > MAX_ERROR_COUNT)
-		pr_warning("%s: %u errors suppressed\n",
+		pr_warn("%s: %u errors suppressed\n",
 			current->comm, error_count - MAX_ERROR_COUNT);
 
 	return error_count;
@@ -313,20 +330,6 @@ static void dmatest_callback(void *arg)
 	wake_up_all(done->wait);
 }
 
-static inline void unmap_src(struct device *dev, dma_addr_t *addr, size_t len,
-			     unsigned int count)
-{
-	while (count--)
-		dma_unmap_single(dev, addr[count], len, DMA_TO_DEVICE);
-}
-
-static inline void unmap_dst(struct device *dev, dma_addr_t *addr, size_t len,
-			     unsigned int count)
-{
-	while (count--)
-		dma_unmap_single(dev, addr[count], len, DMA_BIDIRECTIONAL);
-}
-
 static unsigned int min_odd(unsigned int x, unsigned int y)
 {
 	unsigned int val = min(x, y);
@@ -334,172 +337,49 @@ static unsigned int min_odd(unsigned int x, unsigned int y)
 	return val % 2 ? val : val - 1;
 }
 
-static char *verify_result_get_one(struct dmatest_verify_result *vr,
-		unsigned int i)
+static void result(const char *err, unsigned int n, unsigned int src_off,
+		   unsigned int dst_off, unsigned int len, unsigned long data)
 {
-	struct dmatest_verify_buffer *vb = &vr->data[i];
-	u8 diff = vb->actual ^ vr->pattern;
-	static char buf[512];
-	char *msg;
-
-	if (vr->is_srcbuf)
-		msg = "srcbuf overwritten!";
-	else if ((vr->pattern & PATTERN_COPY)
-			&& (diff & (PATTERN_COPY | PATTERN_OVERWRITE)))
-		msg = "dstbuf not copied!";
-	else if (diff & PATTERN_SRC)
-		msg = "dstbuf was copied!";
-	else
-		msg = "dstbuf mismatch!";
-
-	snprintf(buf, sizeof(buf) - 1, "%s [0x%x] Expected %02x, got %02x", msg,
-		 vb->index, vb->expected, vb->actual);
-
-	return buf;
+	pr_info("%s: result #%u: '%s' with src_off=0x%x dst_off=0x%x len=0x%x (%lu)",
+		current->comm, n, err, src_off, dst_off, len, data);
 }
 
-static char *thread_result_get(const char *name,
-		struct dmatest_thread_result *tr)
+static void dbg_result(const char *err, unsigned int n, unsigned int src_off,
+		       unsigned int dst_off, unsigned int len,
+		       unsigned long data)
 {
-	static const char * const messages[] = {
-		[DMATEST_ET_OK]			= "No errors",
-		[DMATEST_ET_MAP_SRC]		= "src mapping error",
-		[DMATEST_ET_MAP_DST]		= "dst mapping error",
-		[DMATEST_ET_PREP]		= "prep error",
-		[DMATEST_ET_SUBMIT]		= "submit error",
-		[DMATEST_ET_TIMEOUT]		= "test timed out",
-		[DMATEST_ET_DMA_ERROR]		=
-			"got completion callback (DMA_ERROR)",
-		[DMATEST_ET_DMA_IN_PROGRESS]	=
-			"got completion callback (DMA_IN_PROGRESS)",
-		[DMATEST_ET_VERIFY]		= "errors",
-		[DMATEST_ET_VERIFY_BUF]		= "verify errors",
-	};
-	static char buf[512];
-
-	snprintf(buf, sizeof(buf) - 1,
-		 "%s: #%u: %s with src_off=0x%x ""dst_off=0x%x len=0x%x (%lu)",
-		 name, tr->n, messages[tr->type], tr->src_off, tr->dst_off,
-		 tr->len, tr->data);
-
-	return buf;
+	pr_debug("%s: result #%u: '%s' with src_off=0x%x dst_off=0x%x len=0x%x (%lu)",
+		   current->comm, n, err, src_off, dst_off, len, data);
 }
 
-static int thread_result_add(struct dmatest_info *info,
-		struct dmatest_result *r, enum dmatest_error_type type,
-		unsigned int n, unsigned int src_off, unsigned int dst_off,
-		unsigned int len, unsigned long data)
-{
-	struct dmatest_thread_result *tr;
-
-	tr = kzalloc(sizeof(*tr), GFP_KERNEL);
-	if (!tr)
-		return -ENOMEM;
-
-	tr->type = type;
-	tr->n = n;
-	tr->src_off = src_off;
-	tr->dst_off = dst_off;
-	tr->len = len;
-	tr->data = data;
+#define verbose_result(err, n, src_off, dst_off, len, data) ({ \
+	if (verbose) \
+		result(err, n, src_off, dst_off, len, data); \
+	else \
+		dbg_result(err, n, src_off, dst_off, len, data); \
+})
 
-	mutex_lock(&info->results_lock);
-	list_add_tail(&tr->node, &r->results);
-	mutex_unlock(&info->results_lock);
-
-	if (tr->type == DMATEST_ET_OK)
-		pr_debug("%s\n", thread_result_get(r->name, tr));
-	else
-		pr_warn("%s\n", thread_result_get(r->name, tr));
-
-	return 0;
-}
-
-static unsigned int verify_result_add(struct dmatest_info *info,
-		struct dmatest_result *r, unsigned int n,
-		unsigned int src_off, unsigned int dst_off, unsigned int len,
-		u8 **bufs, int whence, unsigned int counter, u8 pattern,
-		bool is_srcbuf)
+static unsigned long long dmatest_persec(s64 runtime, unsigned int val)
 {
-	struct dmatest_verify_result *vr;
-	unsigned int error_count;
-	unsigned int buf_off = is_srcbuf ? src_off : dst_off;
-	unsigned int start, end;
-
-	if (whence < 0) {
-		start = 0;
-		end = buf_off;
-	} else if (whence > 0) {
-		start = buf_off + len;
-		end = info->params.buf_size;
-	} else {
-		start = buf_off;
-		end = buf_off + len;
-	}
+	unsigned long long per_sec = 1000000;
 
-	vr = kmalloc(sizeof(*vr), GFP_KERNEL);
-	if (!vr) {
-		pr_warn("dmatest: No memory to store verify result\n");
-		return dmatest_verify(NULL, bufs, start, end, counter, pattern,
-				      is_srcbuf);
-	}
-
-	vr->pattern = pattern;
-	vr->is_srcbuf = is_srcbuf;
-
-	error_count = dmatest_verify(vr, bufs, start, end, counter, pattern,
-				     is_srcbuf);
-	if (error_count) {
-		vr->error_count = error_count;
-		thread_result_add(info, r, DMATEST_ET_VERIFY_BUF, n, src_off,
-				  dst_off, len, (unsigned long)vr);
-		return error_count;
-	}
-
-	kfree(vr);
-	return 0;
-}
-
-static void result_free(struct dmatest_info *info, const char *name)
-{
-	struct dmatest_result *r, *_r;
-
-	mutex_lock(&info->results_lock);
-	list_for_each_entry_safe(r, _r, &info->results, node) {
-		struct dmatest_thread_result *tr, *_tr;
-
-		if (name && strcmp(r->name, name))
-			continue;
-
-		list_for_each_entry_safe(tr, _tr, &r->results, node) {
-			if (tr->type == DMATEST_ET_VERIFY_BUF)
-				kfree(tr->vr);
-			list_del(&tr->node);
-			kfree(tr);
-		}
+	if (runtime <= 0)
+		return 0;
 
-		kfree(r->name);
-		list_del(&r->node);
-		kfree(r);
+	/* drop precision until runtime is 32-bits */
+	while (runtime > UINT_MAX) {
+		runtime >>= 1;
+		per_sec <<= 1;
 	}
 
-	mutex_unlock(&info->results_lock);
+	per_sec *= val;
+	do_div(per_sec, runtime);
+	return per_sec;
 }
 
-static struct dmatest_result *result_init(struct dmatest_info *info,
-		const char *name)
+static unsigned long long dmatest_KBs(s64 runtime, unsigned long long len)
 {
-	struct dmatest_result *r;
-
-	r = kzalloc(sizeof(*r), GFP_KERNEL);
-	if (r) {
-		r->name = kstrdup(name, GFP_KERNEL);
-		INIT_LIST_HEAD(&r->results);
-		mutex_lock(&info->results_lock);
-		list_add_tail(&r->node, &info->results);
-		mutex_unlock(&info->results_lock);
-	}
-	return r;
+	return dmatest_persec(runtime, len >> 10);
 }
 
 /*
@@ -525,7 +405,6 @@ static int dmatest_func(void *data)
 	struct dmatest_params	*params;
 	struct dma_chan		*chan;
 	struct dma_device	*dev;
-	const char		*thread_name;
 	unsigned int		src_off, dst_off, len;
 	unsigned int		error_count;
 	unsigned int		failed_tests = 0;
@@ -538,9 +417,10 @@ static int dmatest_func(void *data)
 	int			src_cnt;
 	int			dst_cnt;
 	int			i;
-	struct dmatest_result	*result;
+	ktime_t			ktime;
+	s64			runtime = 0;
+	unsigned long long	total_len = 0;
 
-	thread_name = current->comm;
 	set_freezable();
 
 	ret = -ENOMEM;
@@ -570,10 +450,6 @@ static int dmatest_func(void *data)
 	} else
 		goto err_thread_type;
 
-	result = result_init(info, thread_name);
-	if (!result)
-		goto err_srcs;
-
 	thread->srcs = kcalloc(src_cnt+1, sizeof(u8 *), GFP_KERNEL);
 	if (!thread->srcs)
 		goto err_srcs;
@@ -597,17 +473,17 @@ static int dmatest_func(void *data)
 	set_user_nice(current, 10);
 
 	/*
-	 * src buffers are freed by the DMAEngine code with dma_unmap_single()
-	 * dst buffers are freed by ourselves below
+	 * src and dst buffers are freed by ourselves below
 	 */
-	flags = DMA_CTRL_ACK | DMA_PREP_INTERRUPT
-	      | DMA_COMPL_SKIP_DEST_UNMAP | DMA_COMPL_SRC_UNMAP_SINGLE;
+	flags = DMA_CTRL_ACK | DMA_PREP_INTERRUPT;
 
+	ktime = ktime_get();
 	while (!kthread_should_stop()
 	       && !(params->iterations && total_tests >= params->iterations)) {
 		struct dma_async_tx_descriptor *tx = NULL;
-		dma_addr_t dma_srcs[src_cnt];
-		dma_addr_t dma_dsts[dst_cnt];
+		struct dmaengine_unmap_data *um;
+		dma_addr_t srcs[src_cnt];
+		dma_addr_t *dsts;
 		u8 align = 0;
 
 		total_tests++;
@@ -626,81 +502,103 @@ static int dmatest_func(void *data)
 			break;
 		}
 
-		len = dmatest_random() % params->buf_size + 1;
+		if (params->noverify) {
+			len = params->buf_size;
+			src_off = 0;
+			dst_off = 0;
+		} else {
+			len = dmatest_random() % params->buf_size + 1;
+			len = (len >> align) << align;
+			if (!len)
+				len = 1 << align;
+			src_off = dmatest_random() % (params->buf_size - len + 1);
+			dst_off = dmatest_random() % (params->buf_size - len + 1);
+
+			src_off = (src_off >> align) << align;
+			dst_off = (dst_off >> align) << align;
+
+			dmatest_init_srcs(thread->srcs, src_off, len,
+					  params->buf_size);
+			dmatest_init_dsts(thread->dsts, dst_off, len,
+					  params->buf_size);
+		}
+
 		len = (len >> align) << align;
 		if (!len)
 			len = 1 << align;
-		src_off = dmatest_random() % (params->buf_size - len + 1);
-		dst_off = dmatest_random() % (params->buf_size - len + 1);
+		total_len += len;
 
-		src_off = (src_off >> align) << align;
-		dst_off = (dst_off >> align) << align;
-
-		dmatest_init_srcs(thread->srcs, src_off, len, params->buf_size);
-		dmatest_init_dsts(thread->dsts, dst_off, len, params->buf_size);
+		um = dmaengine_get_unmap_data(dev->dev, src_cnt+dst_cnt,
+					      GFP_KERNEL);
+		if (!um) {
+			failed_tests++;
+			result("unmap data NULL", total_tests,
+			       src_off, dst_off, len, ret);
+			continue;
+		}
 
+		um->len = params->buf_size;
 		for (i = 0; i < src_cnt; i++) {
-			u8 *buf = thread->srcs[i] + src_off;
-
-			dma_srcs[i] = dma_map_single(dev->dev, buf, len,
-						     DMA_TO_DEVICE);
-			ret = dma_mapping_error(dev->dev, dma_srcs[i]);
+			unsigned long buf = (unsigned long) thread->srcs[i];
+			struct page *pg = virt_to_page(buf);
+			unsigned pg_off = buf & ~PAGE_MASK;
+
+			um->addr[i] = dma_map_page(dev->dev, pg, pg_off,
+						   um->len, DMA_TO_DEVICE);
+			srcs[i] = um->addr[i] + src_off;
+			ret = dma_mapping_error(dev->dev, um->addr[i]);
 			if (ret) {
-				unmap_src(dev->dev, dma_srcs, len, i);
-				thread_result_add(info, result,
-						  DMATEST_ET_MAP_SRC,
-						  total_tests, src_off, dst_off,
-						  len, ret);
+				dmaengine_unmap_put(um);
+				result("src mapping error", total_tests,
+				       src_off, dst_off, len, ret);
 				failed_tests++;
 				continue;
 			}
+			um->to_cnt++;
 		}
 		/* map with DMA_BIDIRECTIONAL to force writeback/invalidate */
+		dsts = &um->addr[src_cnt];
 		for (i = 0; i < dst_cnt; i++) {
-			dma_dsts[i] = dma_map_single(dev->dev, thread->dsts[i],
-						     params->buf_size,
-						     DMA_BIDIRECTIONAL);
-			ret = dma_mapping_error(dev->dev, dma_dsts[i]);
+			unsigned long buf = (unsigned long) thread->dsts[i];
+			struct page *pg = virt_to_page(buf);
+			unsigned pg_off = buf & ~PAGE_MASK;
+
+			dsts[i] = dma_map_page(dev->dev, pg, pg_off, um->len,
+					       DMA_BIDIRECTIONAL);
+			ret = dma_mapping_error(dev->dev, dsts[i]);
 			if (ret) {
-				unmap_src(dev->dev, dma_srcs, len, src_cnt);
-				unmap_dst(dev->dev, dma_dsts, params->buf_size,
-					  i);
-				thread_result_add(info, result,
-						  DMATEST_ET_MAP_DST,
-						  total_tests, src_off, dst_off,
-						  len, ret);
+				dmaengine_unmap_put(um);
+				result("dst mapping error", total_tests,
+				       src_off, dst_off, len, ret);
 				failed_tests++;
 				continue;
 			}
+			um->bidi_cnt++;
 		}
 
 		if (thread->type == DMA_MEMCPY)
 			tx = dev->device_prep_dma_memcpy(chan,
-							 dma_dsts[0] + dst_off,
-							 dma_srcs[0], len,
-							 flags);
+							 dsts[0] + dst_off,
+							 srcs[0], len, flags);
 		else if (thread->type == DMA_XOR)
 			tx = dev->device_prep_dma_xor(chan,
-						      dma_dsts[0] + dst_off,
-						      dma_srcs, src_cnt,
+						      dsts[0] + dst_off,
+						      srcs, src_cnt,
 						      len, flags);
 		else if (thread->type == DMA_PQ) {
 			dma_addr_t dma_pq[dst_cnt];
 
 			for (i = 0; i < dst_cnt; i++)
-				dma_pq[i] = dma_dsts[i] + dst_off;
-			tx = dev->device_prep_dma_pq(chan, dma_pq, dma_srcs,
+				dma_pq[i] = dsts[i] + dst_off;
+			tx = dev->device_prep_dma_pq(chan, dma_pq, srcs,
 						     src_cnt, pq_coefs,
 						     len, flags);
 		}
 
 		if (!tx) {
-			unmap_src(dev->dev, dma_srcs, len, src_cnt);
-			unmap_dst(dev->dev, dma_dsts, params->buf_size,
-				  dst_cnt);
-			thread_result_add(info, result, DMATEST_ET_PREP,
-					  total_tests, src_off, dst_off,
-					  len, 0);
+			dmaengine_unmap_put(um);
+			result("prep error", total_tests, src_off,
+			       dst_off, len, ret);
 			msleep(100);
 			failed_tests++;
 			continue;
@@ -712,9 +610,9 @@ static int dmatest_func(void *data)
 		cookie = tx->tx_submit(tx);
 
 		if (dma_submit_error(cookie)) {
-			thread_result_add(info, result, DMATEST_ET_SUBMIT,
-					  total_tests, src_off, dst_off,
-					  len, cookie);
+			dmaengine_unmap_put(um);
+			result("submit error", total_tests, src_off,
+			       dst_off, len, ret);
 			msleep(100);
 			failed_tests++;
 			continue;
@@ -735,59 +633,59 @@ static int dmatest_func(void *data)
 			 * free it this time?" dancing.  For now, just
 			 * leave it dangling.
 			 */
-			thread_result_add(info, result, DMATEST_ET_TIMEOUT,
-					  total_tests, src_off, dst_off,
-					  len, 0);
+			dmaengine_unmap_put(um);
+			result("test timed out", total_tests, src_off, dst_off,
+			       len, 0);
 			failed_tests++;
 			continue;
-		} else if (status != DMA_SUCCESS) {
-			enum dmatest_error_type type = (status == DMA_ERROR) ?
-				DMATEST_ET_DMA_ERROR : DMATEST_ET_DMA_IN_PROGRESS;
-			thread_result_add(info, result, type,
-					  total_tests, src_off, dst_off,
-					  len, status);
+		} else if (status != DMA_COMPLETE) {
+			dmaengine_unmap_put(um);
+			result(status == DMA_ERROR ?
+			       "completion error status" :
+			       "completion busy status", total_tests, src_off,
+			       dst_off, len, ret);
 			failed_tests++;
 			continue;
 		}
 
-		/* Unmap by myself (see DMA_COMPL_SKIP_DEST_UNMAP above) */
-		unmap_dst(dev->dev, dma_dsts, params->buf_size, dst_cnt);
+		dmaengine_unmap_put(um);
 
-		error_count = 0;
+		if (params->noverify) {
+			verbose_result("test passed", total_tests, src_off,
+				       dst_off, len, 0);
+			continue;
+		}
 
-		pr_debug("%s: verifying source buffer...\n", thread_name);
-		error_count += verify_result_add(info, result, total_tests,
-				src_off, dst_off, len, thread->srcs, -1,
+		pr_debug("%s: verifying source buffer...\n", current->comm);
+		error_count = dmatest_verify(thread->srcs, 0, src_off,
 				0, PATTERN_SRC, true);
-		error_count += verify_result_add(info, result, total_tests,
-				src_off, dst_off, len, thread->srcs, 0,
-				src_off, PATTERN_SRC | PATTERN_COPY, true);
-		error_count += verify_result_add(info, result, total_tests,
-				src_off, dst_off, len, thread->srcs, 1,
-				src_off + len, PATTERN_SRC, true);
-
-		pr_debug("%s: verifying dest buffer...\n", thread_name);
-		error_count += verify_result_add(info, result, total_tests,
-				src_off, dst_off, len, thread->dsts, -1,
+		error_count += dmatest_verify(thread->srcs, src_off,
+				src_off + len, src_off,
+				PATTERN_SRC | PATTERN_COPY, true);
+		error_count += dmatest_verify(thread->srcs, src_off + len,
+				params->buf_size, src_off + len,
+				PATTERN_SRC, true);
+
+		pr_debug("%s: verifying dest buffer...\n", current->comm);
+		error_count += dmatest_verify(thread->dsts, 0, dst_off,
 				0, PATTERN_DST, false);
-		error_count += verify_result_add(info, result, total_tests,
-				src_off, dst_off, len, thread->dsts, 0,
-				src_off, PATTERN_SRC | PATTERN_COPY, false);
-		error_count += verify_result_add(info, result, total_tests,
-				src_off, dst_off, len, thread->dsts, 1,
-				dst_off + len, PATTERN_DST, false);
+		error_count += dmatest_verify(thread->dsts, dst_off,
+				dst_off + len, src_off,
+				PATTERN_SRC | PATTERN_COPY, false);
+		error_count += dmatest_verify(thread->dsts, dst_off + len,
+				params->buf_size, dst_off + len,
+				PATTERN_DST, false);
 
 		if (error_count) {
-			thread_result_add(info, result, DMATEST_ET_VERIFY,
-					  total_tests, src_off, dst_off,
-					  len, error_count);
+			result("data error", total_tests, src_off, dst_off,
+			       len, error_count);
 			failed_tests++;
 		} else {
-			thread_result_add(info, result, DMATEST_ET_OK,
-					  total_tests, src_off, dst_off,
-					  len, 0);
+			verbose_result("test passed", total_tests, src_off,
+				       dst_off, len, 0);
 		}
 	}
+	runtime = ktime_us_delta(ktime_get(), ktime);
 
 	ret = 0;
 	for (i = 0; thread->dsts[i]; i++)
@@ -802,20 +700,17 @@ err_srcbuf:
 err_srcs:
 	kfree(pq_coefs);
 err_thread_type:
-	pr_notice("%s: terminating after %u tests, %u failures (status %d)\n",
-			thread_name, total_tests, failed_tests, ret);
+	pr_info("%s: summary %u tests, %u failures %llu iops %llu KB/s (%d)\n",
+		current->comm, total_tests, failed_tests,
+		dmatest_persec(runtime, total_tests),
+		dmatest_KBs(runtime, total_len), ret);
 
 	/* terminate all transfers on specified channels */
 	if (ret)
 		dmaengine_terminate_all(chan);
 
 	thread->done = true;
-
-	if (params->iterations > 0)
-		while (!kthread_should_stop()) {
-			DECLARE_WAIT_QUEUE_HEAD_ONSTACK(wait_dmatest_exit);
-			interruptible_sleep_on(&wait_dmatest_exit);
-		}
+	wake_up(&thread_wait);
 
 	return ret;
 }
@@ -828,9 +723,10 @@ static void dmatest_cleanup_channel(struct dmatest_chan *dtc)
 
 	list_for_each_entry_safe(thread, _thread, &dtc->threads, node) {
 		ret = kthread_stop(thread->task);
-		pr_debug("dmatest: thread %s exited with status %d\n",
-				thread->task->comm, ret);
+		pr_debug("thread %s exited with status %d\n",
+			 thread->task->comm, ret);
 		list_del(&thread->node);
+		put_task_struct(thread->task);
 		kfree(thread);
 	}
 
@@ -861,27 +757,27 @@ static int dmatest_add_threads(struct dmatest_info *info,
 	for (i = 0; i < params->threads_per_chan; i++) {
 		thread = kzalloc(sizeof(struct dmatest_thread), GFP_KERNEL);
 		if (!thread) {
-			pr_warning("dmatest: No memory for %s-%s%u\n",
-				   dma_chan_name(chan), op, i);
-
+			pr_warn("No memory for %s-%s%u\n",
+				dma_chan_name(chan), op, i);
 			break;
 		}
 		thread->info = info;
 		thread->chan = dtc->chan;
 		thread->type = type;
 		smp_wmb();
-		thread->task = kthread_run(dmatest_func, thread, "%s-%s%u",
+		thread->task = kthread_create(dmatest_func, thread, "%s-%s%u",
 				dma_chan_name(chan), op, i);
 		if (IS_ERR(thread->task)) {
-			pr_warning("dmatest: Failed to run thread %s-%s%u\n",
-					dma_chan_name(chan), op, i);
+			pr_warn("Failed to create thread %s-%s%u\n",
+				dma_chan_name(chan), op, i);
 			kfree(thread);
 			break;
 		}
 
 		/* srcbuf and dstbuf are allocated by the thread itself */
-
+		get_task_struct(thread->task);
 		list_add_tail(&thread->node, &dtc->threads);
+		wake_up_process(thread->task);
 	}
 
 	return i;
@@ -897,7 +793,7 @@ static int dmatest_add_channel(struct dmatest_info *info,
 
 	dtc = kmalloc(sizeof(struct dmatest_chan), GFP_KERNEL);
 	if (!dtc) {
-		pr_warning("dmatest: No memory for %s\n", dma_chan_name(chan));
+		pr_warn("No memory for %s\n", dma_chan_name(chan));
 		return -ENOMEM;
 	}
 
@@ -917,7 +813,7 @@ static int dmatest_add_channel(struct dmatest_info *info,
 		thread_count += cnt > 0 ? cnt : 0;
 	}
 
-	pr_info("dmatest: Started %u threads using %s\n",
+	pr_info("Started %u threads using %s\n",
 		thread_count, dma_chan_name(chan));
 
 	list_add_tail(&dtc->node, &info->channels);
@@ -937,20 +833,20 @@ static bool filter(struct dma_chan *chan, void *param)
 		return true;
 }
 
-static int __run_threaded_test(struct dmatest_info *info)
+static void request_channels(struct dmatest_info *info,
+			     enum dma_transaction_type type)
 {
 	dma_cap_mask_t mask;
-	struct dma_chan *chan;
-	struct dmatest_params *params = &info->params;
-	int err = 0;
 
 	dma_cap_zero(mask);
-	dma_cap_set(DMA_MEMCPY, mask);
+	dma_cap_set(type, mask);
 	for (;;) {
+		struct dmatest_params *params = &info->params;
+		struct dma_chan *chan;
+
 		chan = dma_request_channel(mask, filter, params);
 		if (chan) {
-			err = dmatest_add_channel(info, chan);
-			if (err) {
+			if (dmatest_add_channel(info, chan)) {
 				dma_release_channel(chan);
 				break; /* add_channel failed, punt */
 			}
@@ -960,22 +856,30 @@ static int __run_threaded_test(struct dmatest_info *info)
 		    info->nr_channels >= params->max_channels)
 			break; /* we have all we need */
 	}
-	return err;
 }
 
-#ifndef MODULE
-static int run_threaded_test(struct dmatest_info *info)
+static void run_threaded_test(struct dmatest_info *info)
 {
-	int ret;
+	struct dmatest_params *params = &info->params;
 
-	mutex_lock(&info->lock);
-	ret = __run_threaded_test(info);
-	mutex_unlock(&info->lock);
-	return ret;
+	/* Copy test parameters */
+	params->buf_size = test_buf_size;
+	strlcpy(params->channel, strim(test_channel), sizeof(params->channel));
+	strlcpy(params->device, strim(test_device), sizeof(params->device));
+	params->threads_per_chan = threads_per_chan;
+	params->max_channels = max_channels;
+	params->iterations = iterations;
+	params->xor_sources = xor_sources;
+	params->pq_sources = pq_sources;
+	params->timeout = timeout;
+	params->noverify = noverify;
+
+	request_channels(info, DMA_MEMCPY);
+	request_channels(info, DMA_XOR);
+	request_channels(info, DMA_PQ);
 }
-#endif
 
-static void __stop_threaded_test(struct dmatest_info *info)
+static void stop_threaded_test(struct dmatest_info *info)
 {
 	struct dmatest_chan *dtc, *_dtc;
 	struct dma_chan *chan;
@@ -984,203 +888,86 @@ static void __stop_threaded_test(struct dmatest_info *info)
 		list_del(&dtc->node);
 		chan = dtc->chan;
 		dmatest_cleanup_channel(dtc);
-		pr_debug("dmatest: dropped channel %s\n", dma_chan_name(chan));
+		pr_debug("dropped channel %s\n", dma_chan_name(chan));
 		dma_release_channel(chan);
 	}
 
 	info->nr_channels = 0;
 }
 
-static void stop_threaded_test(struct dmatest_info *info)
+static void restart_threaded_test(struct dmatest_info *info, bool run)
 {
-	mutex_lock(&info->lock);
-	__stop_threaded_test(info);
-	mutex_unlock(&info->lock);
-}
-
-static int __restart_threaded_test(struct dmatest_info *info, bool run)
-{
-	struct dmatest_params *params = &info->params;
+	/* we might be called early to set run=, defer running until all
+	 * parameters have been evaluated
+	 */
+	if (!info->did_init)
+		return;
 
 	/* Stop any running test first */
-	__stop_threaded_test(info);
-
-	if (run == false)
-		return 0;
-
-	/* Clear results from previous run */
-	result_free(info, NULL);
-
-	/* Copy test parameters */
-	params->buf_size = test_buf_size;
-	strlcpy(params->channel, strim(test_channel), sizeof(params->channel));
-	strlcpy(params->device, strim(test_device), sizeof(params->device));
-	params->threads_per_chan = threads_per_chan;
-	params->max_channels = max_channels;
-	params->iterations = iterations;
-	params->xor_sources = xor_sources;
-	params->pq_sources = pq_sources;
-	params->timeout = timeout;
+	stop_threaded_test(info);
 
 	/* Run test with new parameters */
-	return __run_threaded_test(info);
-}
-
-static bool __is_threaded_test_run(struct dmatest_info *info)
-{
-	struct dmatest_chan *dtc;
-
-	list_for_each_entry(dtc, &info->channels, node) {
-		struct dmatest_thread *thread;
-
-		list_for_each_entry(thread, &dtc->threads, node) {
-			if (!thread->done)
-				return true;
-		}
-	}
-
-	return false;
+	run_threaded_test(info);
 }
 
-static ssize_t dtf_read_run(struct file *file, char __user *user_buf,
-		size_t count, loff_t *ppos)
+static int dmatest_run_get(char *val, const struct kernel_param *kp)
 {
-	struct dmatest_info *info = file->private_data;
-	char buf[3];
+	struct dmatest_info *info = &test_info;
 
 	mutex_lock(&info->lock);
-
-	if (__is_threaded_test_run(info)) {
-		buf[0] = 'Y';
+	if (is_threaded_test_run(info)) {
+		dmatest_run = true;
 	} else {
-		__stop_threaded_test(info);
-		buf[0] = 'N';
+		stop_threaded_test(info);
+		dmatest_run = false;
 	}
-
 	mutex_unlock(&info->lock);
-	buf[1] = '\n';
-	buf[2] = 0x00;
-	return simple_read_from_buffer(user_buf, count, ppos, buf, 2);
-}
-
-static ssize_t dtf_write_run(struct file *file, const char __user *user_buf,
-		size_t count, loff_t *ppos)
-{
-	struct dmatest_info *info = file->private_data;
-	char buf[16];
-	bool bv;
-	int ret = 0;
 
-	if (copy_from_user(buf, user_buf, min(count, (sizeof(buf) - 1))))
-		return -EFAULT;
-
-	if (strtobool(buf, &bv) == 0) {
-		mutex_lock(&info->lock);
-
-		if (__is_threaded_test_run(info))
-			ret = -EBUSY;
-		else
-			ret = __restart_threaded_test(info, bv);
-
-		mutex_unlock(&info->lock);
-	}
-
-	return ret ? ret : count;
+	return param_get_bool(val, kp);
 }
 
-static const struct file_operations dtf_run_fops = {
-	.read	= dtf_read_run,
-	.write	= dtf_write_run,
-	.open	= simple_open,
-	.llseek	= default_llseek,
-};
-
-static int dtf_results_show(struct seq_file *sf, void *data)
+static int dmatest_run_set(const char *val, const struct kernel_param *kp)
 {
-	struct dmatest_info *info = sf->private;
-	struct dmatest_result *result;
-	struct dmatest_thread_result *tr;
-	unsigned int i;
+	struct dmatest_info *info = &test_info;
+	int ret;
 
-	mutex_lock(&info->results_lock);
-	list_for_each_entry(result, &info->results, node) {
-		list_for_each_entry(tr, &result->results, node) {
-			seq_printf(sf, "%s\n",
-				thread_result_get(result->name, tr));
-			if (tr->type == DMATEST_ET_VERIFY_BUF) {
-				for (i = 0; i < tr->vr->error_count; i++) {
-					seq_printf(sf, "\t%s\n",
-						verify_result_get_one(tr->vr, i));
-				}
-			}
-		}
+	mutex_lock(&info->lock);
+	ret = param_set_bool(val, kp);
+	if (ret) {
+		mutex_unlock(&info->lock);
+		return ret;
 	}
 
-	mutex_unlock(&info->results_lock);
-	return 0;
-}
-
-static int dtf_results_open(struct inode *inode, struct file *file)
-{
-	return single_open(file, dtf_results_show, inode->i_private);
-}
-
-static const struct file_operations dtf_results_fops = {
-	.open		= dtf_results_open,
-	.read		= seq_read,
-	.llseek		= seq_lseek,
-	.release	= single_release,
-};
-
-static int dmatest_register_dbgfs(struct dmatest_info *info)
-{
-	struct dentry *d;
-
-	d = debugfs_create_dir("dmatest", NULL);
-	if (IS_ERR(d))
-		return PTR_ERR(d);
-	if (!d)
-		goto err_root;
+	if (is_threaded_test_run(info))
+		ret = -EBUSY;
+	else if (dmatest_run)
+		restart_threaded_test(info, dmatest_run);
 
-	info->root = d;
-
-	/* Run or stop threaded test */
-	debugfs_create_file("run", S_IWUSR | S_IRUGO, info->root, info,
-			    &dtf_run_fops);
-
-	/* Results of test in progress */
-	debugfs_create_file("results", S_IRUGO, info->root, info,
-			    &dtf_results_fops);
-
-	return 0;
+	mutex_unlock(&info->lock);
 
-err_root:
-	pr_err("dmatest: Failed to initialize debugfs\n");
-	return -ENOMEM;
+	return ret;
 }
 
 static int __init dmatest_init(void)
 {
 	struct dmatest_info *info = &test_info;
-	int ret;
-
-	memset(info, 0, sizeof(*info));
+	struct dmatest_params *params = &info->params;
 
-	mutex_init(&info->lock);
-	INIT_LIST_HEAD(&info->channels);
+	if (dmatest_run) {
+		mutex_lock(&info->lock);
+		run_threaded_test(info);
+		mutex_unlock(&info->lock);
+	}
 
-	mutex_init(&info->results_lock);
-	INIT_LIST_HEAD(&info->results);
+	if (params->iterations && wait)
+		wait_event(thread_wait, !is_threaded_test_run(info));
 
-	ret = dmatest_register_dbgfs(info);
-	if (ret)
-		return ret;
+	/* module parameters are stable, inittime tests are started,
+	 * let userspace take over 'run' control
+	 */
+	info->did_init = true;
 
-#ifdef MODULE
 	return 0;
-#else
-	return run_threaded_test(info);
-#endif
 }
 /* when compiled-in wait for drivers to load first */
 late_initcall(dmatest_init);
@@ -1189,9 +976,9 @@ static void __exit dmatest_exit(void)
 {
 	struct dmatest_info *info = &test_info;
 
-	debugfs_remove_recursive(info->root);
+	mutex_lock(&info->lock);
 	stop_threaded_test(info);
-	result_free(info, NULL);
+	mutex_unlock(&info->lock);
 }
 module_exit(dmatest_exit);
 
diff --git a/drivers/dma/dw/core.c b/drivers/dma/dw/core.c
index 89eb89f22284..7516be4677cf 100644
--- a/drivers/dma/dw/core.c
+++ b/drivers/dma/dw/core.c
@@ -85,10 +85,6 @@ static struct device *chan2dev(struct dma_chan *chan)
 {
 	return &chan->dev->device;
 }
-static struct device *chan2parent(struct dma_chan *chan)
-{
-	return chan->dev->device.parent;
-}
 
 static struct dw_desc *dwc_first_active(struct dw_dma_chan *dwc)
 {
@@ -311,26 +307,7 @@ dwc_descriptor_complete(struct dw_dma_chan *dwc, struct dw_desc *desc,
 	list_splice_init(&desc->tx_list, &dwc->free_list);
 	list_move(&desc->desc_node, &dwc->free_list);
 
-	if (!is_slave_direction(dwc->direction)) {
-		struct device *parent = chan2parent(&dwc->chan);
-		if (!(txd->flags & DMA_COMPL_SKIP_DEST_UNMAP)) {
-			if (txd->flags & DMA_COMPL_DEST_UNMAP_SINGLE)
-				dma_unmap_single(parent, desc->lli.dar,
-					desc->total_len, DMA_FROM_DEVICE);
-			else
-				dma_unmap_page(parent, desc->lli.dar,
-					desc->total_len, DMA_FROM_DEVICE);
-		}
-		if (!(txd->flags & DMA_COMPL_SKIP_SRC_UNMAP)) {
-			if (txd->flags & DMA_COMPL_SRC_UNMAP_SINGLE)
-				dma_unmap_single(parent, desc->lli.sar,
-					desc->total_len, DMA_TO_DEVICE);
-			else
-				dma_unmap_page(parent, desc->lli.sar,
-					desc->total_len, DMA_TO_DEVICE);
-		}
-	}
-
+	dma_descriptor_unmap(txd);
 	spin_unlock_irqrestore(&dwc->lock, flags);
 
 	if (callback)
@@ -1098,13 +1075,13 @@ dwc_tx_status(struct dma_chan *chan,
 	enum dma_status		ret;
 
 	ret = dma_cookie_status(chan, cookie, txstate);
-	if (ret == DMA_SUCCESS)
+	if (ret == DMA_COMPLETE)
 		return ret;
 
 	dwc_scan_descriptors(to_dw_dma(chan->device), dwc);
 
 	ret = dma_cookie_status(chan, cookie, txstate);
-	if (ret != DMA_SUCCESS)
+	if (ret != DMA_COMPLETE)
 		dma_set_residue(txstate, dwc_get_residue(dwc));
 
 	if (dwc->paused && ret == DMA_IN_PROGRESS)
diff --git a/drivers/dma/edma.c b/drivers/dma/edma.c
index bef8a368c8dd..2539ea0cbc63 100644
--- a/drivers/dma/edma.c
+++ b/drivers/dma/edma.c
@@ -46,14 +46,21 @@
 #define EDMA_CHANS	64
 #endif /* CONFIG_ARCH_DAVINCI_DA8XX */
 
-/* Max of 16 segments per channel to conserve PaRAM slots */
-#define MAX_NR_SG		16
+/*
+ * Max of 20 segments per channel to conserve PaRAM slots
+ * Also note that MAX_NR_SG should be atleast the no.of periods
+ * that are required for ASoC, otherwise DMA prep calls will
+ * fail. Today davinci-pcm is the only user of this driver and
+ * requires atleast 17 slots, so we setup the default to 20.
+ */
+#define MAX_NR_SG		20
 #define EDMA_MAX_SLOTS		MAX_NR_SG
 #define EDMA_DESCRIPTORS	16
 
 struct edma_desc {
 	struct virt_dma_desc		vdesc;
 	struct list_head		node;
+	int				cyclic;
 	int				absync;
 	int				pset_nr;
 	int				processed;
@@ -167,8 +174,13 @@ static void edma_execute(struct edma_chan *echan)
 	 * then setup a link to the dummy slot, this results in all future
 	 * events being absorbed and that's OK because we're done
 	 */
-	if (edesc->processed == edesc->pset_nr)
-		edma_link(echan->slot[nslots-1], echan->ecc->dummy_slot);
+	if (edesc->processed == edesc->pset_nr) {
+		if (edesc->cyclic)
+			edma_link(echan->slot[nslots-1], echan->slot[1]);
+		else
+			edma_link(echan->slot[nslots-1],
+				  echan->ecc->dummy_slot);
+	}
 
 	edma_resume(echan->ch_num);
 
@@ -250,6 +262,117 @@ static int edma_control(struct dma_chan *chan, enum dma_ctrl_cmd cmd,
 	return ret;
 }
 
+/*
+ * A PaRAM set configuration abstraction used by other modes
+ * @chan: Channel who's PaRAM set we're configuring
+ * @pset: PaRAM set to initialize and setup.
+ * @src_addr: Source address of the DMA
+ * @dst_addr: Destination address of the DMA
+ * @burst: In units of dev_width, how much to send
+ * @dev_width: How much is the dev_width
+ * @dma_length: Total length of the DMA transfer
+ * @direction: Direction of the transfer
+ */
+static int edma_config_pset(struct dma_chan *chan, struct edmacc_param *pset,
+	dma_addr_t src_addr, dma_addr_t dst_addr, u32 burst,
+	enum dma_slave_buswidth dev_width, unsigned int dma_length,
+	enum dma_transfer_direction direction)
+{
+	struct edma_chan *echan = to_edma_chan(chan);
+	struct device *dev = chan->device->dev;
+	int acnt, bcnt, ccnt, cidx;
+	int src_bidx, dst_bidx, src_cidx, dst_cidx;
+	int absync;
+
+	acnt = dev_width;
+	/*
+	 * If the maxburst is equal to the fifo width, use
+	 * A-synced transfers. This allows for large contiguous
+	 * buffer transfers using only one PaRAM set.
+	 */
+	if (burst == 1) {
+		/*
+		 * For the A-sync case, bcnt and ccnt are the remainder
+		 * and quotient respectively of the division of:
+		 * (dma_length / acnt) by (SZ_64K -1). This is so
+		 * that in case bcnt over flows, we have ccnt to use.
+		 * Note: In A-sync tranfer only, bcntrld is used, but it
+		 * only applies for sg_dma_len(sg) >= SZ_64K.
+		 * In this case, the best way adopted is- bccnt for the
+		 * first frame will be the remainder below. Then for
+		 * every successive frame, bcnt will be SZ_64K-1. This
+		 * is assured as bcntrld = 0xffff in end of function.
+		 */
+		absync = false;
+		ccnt = dma_length / acnt / (SZ_64K - 1);
+		bcnt = dma_length / acnt - ccnt * (SZ_64K - 1);
+		/*
+		 * If bcnt is non-zero, we have a remainder and hence an
+		 * extra frame to transfer, so increment ccnt.
+		 */
+		if (bcnt)
+			ccnt++;
+		else
+			bcnt = SZ_64K - 1;
+		cidx = acnt;
+	} else {
+		/*
+		 * If maxburst is greater than the fifo address_width,
+		 * use AB-synced transfers where A count is the fifo
+		 * address_width and B count is the maxburst. In this
+		 * case, we are limited to transfers of C count frames
+		 * of (address_width * maxburst) where C count is limited
+		 * to SZ_64K-1. This places an upper bound on the length
+		 * of an SG segment that can be handled.
+		 */
+		absync = true;
+		bcnt = burst;
+		ccnt = dma_length / (acnt * bcnt);
+		if (ccnt > (SZ_64K - 1)) {
+			dev_err(dev, "Exceeded max SG segment size\n");
+			return -EINVAL;
+		}
+		cidx = acnt * bcnt;
+	}
+
+	if (direction == DMA_MEM_TO_DEV) {
+		src_bidx = acnt;
+		src_cidx = cidx;
+		dst_bidx = 0;
+		dst_cidx = 0;
+	} else if (direction == DMA_DEV_TO_MEM)  {
+		src_bidx = 0;
+		src_cidx = 0;
+		dst_bidx = acnt;
+		dst_cidx = cidx;
+	} else {
+		dev_err(dev, "%s: direction not implemented yet\n", __func__);
+		return -EINVAL;
+	}
+
+	pset->opt = EDMA_TCC(EDMA_CHAN_SLOT(echan->ch_num));
+	/* Configure A or AB synchronized transfers */
+	if (absync)
+		pset->opt |= SYNCDIM;
+
+	pset->src = src_addr;
+	pset->dst = dst_addr;
+
+	pset->src_dst_bidx = (dst_bidx << 16) | src_bidx;
+	pset->src_dst_cidx = (dst_cidx << 16) | src_cidx;
+
+	pset->a_b_cnt = bcnt << 16 | acnt;
+	pset->ccnt = ccnt;
+	/*
+	 * Only time when (bcntrld) auto reload is required is for
+	 * A-sync case, and in this case, a requirement of reload value
+	 * of SZ_64K-1 only is assured. 'link' is initially set to NULL
+	 * and then later will be populated by edma_execute.
+	 */
+	pset->link_bcntrld = 0xffffffff;
+	return absync;
+}
+
 static struct dma_async_tx_descriptor *edma_prep_slave_sg(
 	struct dma_chan *chan, struct scatterlist *sgl,
 	unsigned int sg_len, enum dma_transfer_direction direction,
@@ -258,23 +381,21 @@ static struct dma_async_tx_descriptor *edma_prep_slave_sg(
 	struct edma_chan *echan = to_edma_chan(chan);
 	struct device *dev = chan->device->dev;
 	struct edma_desc *edesc;
-	dma_addr_t dev_addr;
+	dma_addr_t src_addr = 0, dst_addr = 0;
 	enum dma_slave_buswidth dev_width;
 	u32 burst;
 	struct scatterlist *sg;
-	int acnt, bcnt, ccnt, src, dst, cidx;
-	int src_bidx, dst_bidx, src_cidx, dst_cidx;
-	int i, nslots;
+	int i, nslots, ret;
 
 	if (unlikely(!echan || !sgl || !sg_len))
 		return NULL;
 
 	if (direction == DMA_DEV_TO_MEM) {
-		dev_addr = echan->cfg.src_addr;
+		src_addr = echan->cfg.src_addr;
 		dev_width = echan->cfg.src_addr_width;
 		burst = echan->cfg.src_maxburst;
 	} else if (direction == DMA_MEM_TO_DEV) {
-		dev_addr = echan->cfg.dst_addr;
+		dst_addr = echan->cfg.dst_addr;
 		dev_width = echan->cfg.dst_addr_width;
 		burst = echan->cfg.dst_maxburst;
 	} else {
@@ -307,7 +428,6 @@ static struct dma_async_tx_descriptor *edma_prep_slave_sg(
 			if (echan->slot[i] < 0) {
 				kfree(edesc);
 				dev_err(dev, "Failed to allocate slot\n");
-				kfree(edesc);
 				return NULL;
 			}
 		}
@@ -315,64 +435,21 @@ static struct dma_async_tx_descriptor *edma_prep_slave_sg(
 
 	/* Configure PaRAM sets for each SG */
 	for_each_sg(sgl, sg, sg_len, i) {
-
-		acnt = dev_width;
-
-		/*
-		 * If the maxburst is equal to the fifo width, use
-		 * A-synced transfers. This allows for large contiguous
-		 * buffer transfers using only one PaRAM set.
-		 */
-		if (burst == 1) {
-			edesc->absync = false;
-			ccnt = sg_dma_len(sg) / acnt / (SZ_64K - 1);
-			bcnt = sg_dma_len(sg) / acnt - ccnt * (SZ_64K - 1);
-			if (bcnt)
-				ccnt++;
-			else
-				bcnt = SZ_64K - 1;
-			cidx = acnt;
-		/*
-		 * If maxburst is greater than the fifo address_width,
-		 * use AB-synced transfers where A count is the fifo
-		 * address_width and B count is the maxburst. In this
-		 * case, we are limited to transfers of C count frames
-		 * of (address_width * maxburst) where C count is limited
-		 * to SZ_64K-1. This places an upper bound on the length
-		 * of an SG segment that can be handled.
-		 */
-		} else {
-			edesc->absync = true;
-			bcnt = burst;
-			ccnt = sg_dma_len(sg) / (acnt * bcnt);
-			if (ccnt > (SZ_64K - 1)) {
-				dev_err(dev, "Exceeded max SG segment size\n");
-				kfree(edesc);
-				return NULL;
-			}
-			cidx = acnt * bcnt;
+		/* Get address for each SG */
+		if (direction == DMA_DEV_TO_MEM)
+			dst_addr = sg_dma_address(sg);
+		else
+			src_addr = sg_dma_address(sg);
+
+		ret = edma_config_pset(chan, &edesc->pset[i], src_addr,
+				       dst_addr, burst, dev_width,
+				       sg_dma_len(sg), direction);
+		if (ret < 0) {
+			kfree(edesc);
+			return NULL;
 		}
 
-		if (direction == DMA_MEM_TO_DEV) {
-			src = sg_dma_address(sg);
-			dst = dev_addr;
-			src_bidx = acnt;
-			src_cidx = cidx;
-			dst_bidx = 0;
-			dst_cidx = 0;
-		} else {
-			src = dev_addr;
-			dst = sg_dma_address(sg);
-			src_bidx = 0;
-			src_cidx = 0;
-			dst_bidx = acnt;
-			dst_cidx = cidx;
-		}
-
-		edesc->pset[i].opt = EDMA_TCC(EDMA_CHAN_SLOT(echan->ch_num));
-		/* Configure A or AB synchronized transfers */
-		if (edesc->absync)
-			edesc->pset[i].opt |= SYNCDIM;
+		edesc->absync = ret;
 
 		/* If this is the last in a current SG set of transactions,
 		   enable interrupts so that next set is processed */
@@ -382,17 +459,138 @@ static struct dma_async_tx_descriptor *edma_prep_slave_sg(
 		/* If this is the last set, enable completion interrupt flag */
 		if (i == sg_len - 1)
 			edesc->pset[i].opt |= TCINTEN;
+	}
 
-		edesc->pset[i].src = src;
-		edesc->pset[i].dst = dst;
+	return vchan_tx_prep(&echan->vchan, &edesc->vdesc, tx_flags);
+}
 
-		edesc->pset[i].src_dst_bidx = (dst_bidx << 16) | src_bidx;
-		edesc->pset[i].src_dst_cidx = (dst_cidx << 16) | src_cidx;
+static struct dma_async_tx_descriptor *edma_prep_dma_cyclic(
+	struct dma_chan *chan, dma_addr_t buf_addr, size_t buf_len,
+	size_t period_len, enum dma_transfer_direction direction,
+	unsigned long tx_flags, void *context)
+{
+	struct edma_chan *echan = to_edma_chan(chan);
+	struct device *dev = chan->device->dev;
+	struct edma_desc *edesc;
+	dma_addr_t src_addr, dst_addr;
+	enum dma_slave_buswidth dev_width;
+	u32 burst;
+	int i, ret, nslots;
+
+	if (unlikely(!echan || !buf_len || !period_len))
+		return NULL;
+
+	if (direction == DMA_DEV_TO_MEM) {
+		src_addr = echan->cfg.src_addr;
+		dst_addr = buf_addr;
+		dev_width = echan->cfg.src_addr_width;
+		burst = echan->cfg.src_maxburst;
+	} else if (direction == DMA_MEM_TO_DEV) {
+		src_addr = buf_addr;
+		dst_addr = echan->cfg.dst_addr;
+		dev_width = echan->cfg.dst_addr_width;
+		burst = echan->cfg.dst_maxburst;
+	} else {
+		dev_err(dev, "%s: bad direction?\n", __func__);
+		return NULL;
+	}
+
+	if (dev_width == DMA_SLAVE_BUSWIDTH_UNDEFINED) {
+		dev_err(dev, "Undefined slave buswidth\n");
+		return NULL;
+	}
+
+	if (unlikely(buf_len % period_len)) {
+		dev_err(dev, "Period should be multiple of Buffer length\n");
+		return NULL;
+	}
+
+	nslots = (buf_len / period_len) + 1;
+
+	/*
+	 * Cyclic DMA users such as audio cannot tolerate delays introduced
+	 * by cases where the number of periods is more than the maximum
+	 * number of SGs the EDMA driver can handle at a time. For DMA types
+	 * such as Slave SGs, such delays are tolerable and synchronized,
+	 * but the synchronization is difficult to achieve with Cyclic and
+	 * cannot be guaranteed, so we error out early.
+	 */
+	if (nslots > MAX_NR_SG)
+		return NULL;
+
+	edesc = kzalloc(sizeof(*edesc) + nslots *
+		sizeof(edesc->pset[0]), GFP_ATOMIC);
+	if (!edesc) {
+		dev_dbg(dev, "Failed to allocate a descriptor\n");
+		return NULL;
+	}
+
+	edesc->cyclic = 1;
+	edesc->pset_nr = nslots;
+
+	dev_dbg(dev, "%s: nslots=%d\n", __func__, nslots);
+	dev_dbg(dev, "%s: period_len=%d\n", __func__, period_len);
+	dev_dbg(dev, "%s: buf_len=%d\n", __func__, buf_len);
+
+	for (i = 0; i < nslots; i++) {
+		/* Allocate a PaRAM slot, if needed */
+		if (echan->slot[i] < 0) {
+			echan->slot[i] =
+				edma_alloc_slot(EDMA_CTLR(echan->ch_num),
+						EDMA_SLOT_ANY);
+			if (echan->slot[i] < 0) {
+				dev_err(dev, "Failed to allocate slot\n");
+				return NULL;
+			}
+		}
+
+		if (i == nslots - 1) {
+			memcpy(&edesc->pset[i], &edesc->pset[0],
+			       sizeof(edesc->pset[0]));
+			break;
+		}
+
+		ret = edma_config_pset(chan, &edesc->pset[i], src_addr,
+				       dst_addr, burst, dev_width, period_len,
+				       direction);
+		if (ret < 0)
+			return NULL;
 
-		edesc->pset[i].a_b_cnt = bcnt << 16 | acnt;
-		edesc->pset[i].ccnt = ccnt;
-		edesc->pset[i].link_bcntrld = 0xffffffff;
+		if (direction == DMA_DEV_TO_MEM)
+			dst_addr += period_len;
+		else
+			src_addr += period_len;
 
+		dev_dbg(dev, "%s: Configure period %d of buf:\n", __func__, i);
+		dev_dbg(dev,
+			"\n pset[%d]:\n"
+			"  chnum\t%d\n"
+			"  slot\t%d\n"
+			"  opt\t%08x\n"
+			"  src\t%08x\n"
+			"  dst\t%08x\n"
+			"  abcnt\t%08x\n"
+			"  ccnt\t%08x\n"
+			"  bidx\t%08x\n"
+			"  cidx\t%08x\n"
+			"  lkrld\t%08x\n",
+			i, echan->ch_num, echan->slot[i],
+			edesc->pset[i].opt,
+			edesc->pset[i].src,
+			edesc->pset[i].dst,
+			edesc->pset[i].a_b_cnt,
+			edesc->pset[i].ccnt,
+			edesc->pset[i].src_dst_bidx,
+			edesc->pset[i].src_dst_cidx,
+			edesc->pset[i].link_bcntrld);
+
+		edesc->absync = ret;
+
+		/*
+		 * Enable interrupts for every period because callback
+		 * has to be called for every period.
+		 */
+		edesc->pset[i].opt |= TCINTEN;
 	}
 
 	return vchan_tx_prep(&echan->vchan, &edesc->vdesc, tx_flags);
@@ -406,30 +604,34 @@ static void edma_callback(unsigned ch_num, u16 ch_status, void *data)
 	unsigned long flags;
 	struct edmacc_param p;
 
-	/* Pause the channel */
-	edma_pause(echan->ch_num);
+	edesc = echan->edesc;
+
+	/* Pause the channel for non-cyclic */
+	if (!edesc || (edesc && !edesc->cyclic))
+		edma_pause(echan->ch_num);
 
 	switch (ch_status) {
-	case DMA_COMPLETE:
+	case EDMA_DMA_COMPLETE:
 		spin_lock_irqsave(&echan->vchan.lock, flags);
 
-		edesc = echan->edesc;
 		if (edesc) {
-			if (edesc->processed == edesc->pset_nr) {
+			if (edesc->cyclic) {
+				vchan_cyclic_callback(&edesc->vdesc);
+			} else if (edesc->processed == edesc->pset_nr) {
 				dev_dbg(dev, "Transfer complete, stopping channel %d\n", ch_num);
 				edma_stop(echan->ch_num);
 				vchan_cookie_complete(&edesc->vdesc);
+				edma_execute(echan);
 			} else {
 				dev_dbg(dev, "Intermediate transfer complete on channel %d\n", ch_num);
+				edma_execute(echan);
 			}
-
-			edma_execute(echan);
 		}
 
 		spin_unlock_irqrestore(&echan->vchan.lock, flags);
 
 		break;
-	case DMA_CC_ERROR:
+	case EDMA_DMA_CC_ERROR:
 		spin_lock_irqsave(&echan->vchan.lock, flags);
 
 		edma_read_slot(EDMA_CHAN_SLOT(echan->slot[0]), &p);
@@ -579,7 +781,7 @@ static enum dma_status edma_tx_status(struct dma_chan *chan,
 	unsigned long flags;
 
 	ret = dma_cookie_status(chan, cookie, txstate);
-	if (ret == DMA_SUCCESS || !txstate)
+	if (ret == DMA_COMPLETE || !txstate)
 		return ret;
 
 	spin_lock_irqsave(&echan->vchan.lock, flags);
@@ -619,6 +821,7 @@ static void edma_dma_init(struct edma_cc *ecc, struct dma_device *dma,
 			  struct device *dev)
 {
 	dma->device_prep_slave_sg = edma_prep_slave_sg;
+	dma->device_prep_dma_cyclic = edma_prep_dma_cyclic;
 	dma->device_alloc_chan_resources = edma_alloc_chan_resources;
 	dma->device_free_chan_resources = edma_free_chan_resources;
 	dma->device_issue_pending = edma_issue_pending;
diff --git a/drivers/dma/ep93xx_dma.c b/drivers/dma/ep93xx_dma.c
index 591cd8c63abb..cb4bf682a708 100644
--- a/drivers/dma/ep93xx_dma.c
+++ b/drivers/dma/ep93xx_dma.c
@@ -733,28 +733,6 @@ static void ep93xx_dma_advance_work(struct ep93xx_dma_chan *edmac)
 	spin_unlock_irqrestore(&edmac->lock, flags);
 }
 
-static void ep93xx_dma_unmap_buffers(struct ep93xx_dma_desc *desc)
-{
-	struct device *dev = desc->txd.chan->device->dev;
-
-	if (!(desc->txd.flags & DMA_COMPL_SKIP_SRC_UNMAP)) {
-		if (desc->txd.flags & DMA_COMPL_SRC_UNMAP_SINGLE)
-			dma_unmap_single(dev, desc->src_addr, desc->size,
-					 DMA_TO_DEVICE);
-		else
-			dma_unmap_page(dev, desc->src_addr, desc->size,
-				       DMA_TO_DEVICE);
-	}
-	if (!(desc->txd.flags & DMA_COMPL_SKIP_DEST_UNMAP)) {
-		if (desc->txd.flags & DMA_COMPL_DEST_UNMAP_SINGLE)
-			dma_unmap_single(dev, desc->dst_addr, desc->size,
-					 DMA_FROM_DEVICE);
-		else
-			dma_unmap_page(dev, desc->dst_addr, desc->size,
-				       DMA_FROM_DEVICE);
-	}
-}
-
 static void ep93xx_dma_tasklet(unsigned long data)
 {
 	struct ep93xx_dma_chan *edmac = (struct ep93xx_dma_chan *)data;
@@ -787,13 +765,7 @@ static void ep93xx_dma_tasklet(unsigned long data)
 
 	/* Now we can release all the chained descriptors */
 	list_for_each_entry_safe(desc, d, &list, node) {
-		/*
-		 * For the memcpy channels the API requires us to unmap the
-		 * buffers unless requested otherwise.
-		 */
-		if (!edmac->chan.private)
-			ep93xx_dma_unmap_buffers(desc);
-
+		dma_descriptor_unmap(&desc->txd);
 		ep93xx_dma_desc_put(edmac, desc);
 	}
 
diff --git a/drivers/dma/fsldma.c b/drivers/dma/fsldma.c
index 61517dd0d0b7..7086a16a55f2 100644
--- a/drivers/dma/fsldma.c
+++ b/drivers/dma/fsldma.c
@@ -870,22 +870,7 @@ static void fsldma_cleanup_descriptor(struct fsldma_chan *chan,
 	/* Run any dependencies */
 	dma_run_dependencies(txd);
 
-	/* Unmap the dst buffer, if requested */
-	if (!(txd->flags & DMA_COMPL_SKIP_DEST_UNMAP)) {
-		if (txd->flags & DMA_COMPL_DEST_UNMAP_SINGLE)
-			dma_unmap_single(dev, dst, len, DMA_FROM_DEVICE);
-		else
-			dma_unmap_page(dev, dst, len, DMA_FROM_DEVICE);
-	}
-
-	/* Unmap the src buffer, if requested */
-	if (!(txd->flags & DMA_COMPL_SKIP_SRC_UNMAP)) {
-		if (txd->flags & DMA_COMPL_SRC_UNMAP_SINGLE)
-			dma_unmap_single(dev, src, len, DMA_TO_DEVICE);
-		else
-			dma_unmap_page(dev, src, len, DMA_TO_DEVICE);
-	}
-
+	dma_descriptor_unmap(txd);
 #ifdef FSL_DMA_LD_DEBUG
 	chan_dbg(chan, "LD %p free\n", desc);
 #endif
@@ -1255,7 +1240,9 @@ static int fsl_dma_chan_probe(struct fsldma_device *fdev,
 	WARN_ON(fdev->feature != chan->feature);
 
 	chan->dev = fdev->dev;
-	chan->id = ((res.start - 0x100) & 0xfff) >> 7;
+	chan->id = (res.start & 0xfff) < 0x300 ?
+		   ((res.start - 0x100) & 0xfff) >> 7 :
+		   ((res.start - 0x200) & 0xfff) >> 7;
 	if (chan->id >= FSL_DMA_MAX_CHANS_PER_DEVICE) {
 		dev_err(fdev->dev, "too many channels for device\n");
 		err = -EINVAL;
@@ -1428,6 +1415,7 @@ static int fsldma_of_remove(struct platform_device *op)
 }
 
 static const struct of_device_id fsldma_of_ids[] = {
+	{ .compatible = "fsl,elo3-dma", },
 	{ .compatible = "fsl,eloplus-dma", },
 	{ .compatible = "fsl,elo-dma", },
 	{}
@@ -1449,7 +1437,7 @@ static struct platform_driver fsldma_of_driver = {
 
 static __init int fsldma_init(void)
 {
-	pr_info("Freescale Elo / Elo Plus DMA driver\n");
+	pr_info("Freescale Elo series DMA driver\n");
 	return platform_driver_register(&fsldma_of_driver);
 }
 
@@ -1461,5 +1449,5 @@ static void __exit fsldma_exit(void)
 subsys_initcall(fsldma_init);
 module_exit(fsldma_exit);
 
-MODULE_DESCRIPTION("Freescale Elo / Elo Plus DMA driver");
+MODULE_DESCRIPTION("Freescale Elo series DMA driver");
 MODULE_LICENSE("GPL");
diff --git a/drivers/dma/fsldma.h b/drivers/dma/fsldma.h
index f5c38791fc74..1ffc24484d23 100644
--- a/drivers/dma/fsldma.h
+++ b/drivers/dma/fsldma.h
@@ -112,7 +112,7 @@ struct fsldma_chan_regs {
 };
 
 struct fsldma_chan;
-#define FSL_DMA_MAX_CHANS_PER_DEVICE 4
+#define FSL_DMA_MAX_CHANS_PER_DEVICE 8
 
 struct fsldma_device {
 	void __iomem *regs;	/* DGSR register base */
diff --git a/drivers/dma/imx-dma.c b/drivers/dma/imx-dma.c
index 55852c026791..6f9ac2022abd 100644
--- a/drivers/dma/imx-dma.c
+++ b/drivers/dma/imx-dma.c
@@ -572,9 +572,11 @@ static int imxdma_xfer_desc(struct imxdma_desc *d)
 
 		imx_dmav1_writel(imxdma, d->len, DMA_CNTR(imxdmac->channel));
 
-		dev_dbg(imxdma->dev, "%s channel: %d dest=0x%08x src=0x%08x "
-			"dma_length=%d\n", __func__, imxdmac->channel,
-			d->dest, d->src, d->len);
+		dev_dbg(imxdma->dev,
+			"%s channel: %d dest=0x%08llx src=0x%08llx dma_length=%zu\n",
+			__func__, imxdmac->channel,
+			(unsigned long long)d->dest,
+			(unsigned long long)d->src, d->len);
 
 		break;
 	/* Cyclic transfer is the same as slave_sg with special sg configuration. */
@@ -586,20 +588,22 @@ static int imxdma_xfer_desc(struct imxdma_desc *d)
 			imx_dmav1_writel(imxdma, imxdmac->ccr_from_device,
 					 DMA_CCR(imxdmac->channel));
 
-			dev_dbg(imxdma->dev, "%s channel: %d sg=%p sgcount=%d "
-				"total length=%d dev_addr=0x%08x (dev2mem)\n",
-				__func__, imxdmac->channel, d->sg, d->sgcount,
-				d->len, imxdmac->per_address);
+			dev_dbg(imxdma->dev,
+				"%s channel: %d sg=%p sgcount=%d total length=%zu dev_addr=0x%08llx (dev2mem)\n",
+				__func__, imxdmac->channel,
+				d->sg, d->sgcount, d->len,
+				(unsigned long long)imxdmac->per_address);
 		} else if (d->direction == DMA_MEM_TO_DEV) {
 			imx_dmav1_writel(imxdma, imxdmac->per_address,
 					 DMA_DAR(imxdmac->channel));
 			imx_dmav1_writel(imxdma, imxdmac->ccr_to_device,
 					 DMA_CCR(imxdmac->channel));
 
-			dev_dbg(imxdma->dev, "%s channel: %d sg=%p sgcount=%d "
-				"total length=%d dev_addr=0x%08x (mem2dev)\n",
-				__func__, imxdmac->channel, d->sg, d->sgcount,
-				d->len, imxdmac->per_address);
+			dev_dbg(imxdma->dev,
+				"%s channel: %d sg=%p sgcount=%d total length=%zu dev_addr=0x%08llx (mem2dev)\n",
+				__func__, imxdmac->channel,
+				d->sg, d->sgcount, d->len,
+				(unsigned long long)imxdmac->per_address);
 		} else {
 			dev_err(imxdma->dev, "%s channel: %d bad dma mode\n",
 				__func__, imxdmac->channel);
@@ -771,7 +775,7 @@ static int imxdma_alloc_chan_resources(struct dma_chan *chan)
 		desc->desc.tx_submit = imxdma_tx_submit;
 		/* txd.flags will be overwritten in prep funcs */
 		desc->desc.flags = DMA_CTRL_ACK;
-		desc->status = DMA_SUCCESS;
+		desc->status = DMA_COMPLETE;
 
 		list_add_tail(&desc->node, &imxdmac->ld_free);
 		imxdmac->descs_allocated++;
@@ -870,7 +874,7 @@ static struct dma_async_tx_descriptor *imxdma_prep_dma_cyclic(
 	int i;
 	unsigned int periods = buf_len / period_len;
 
-	dev_dbg(imxdma->dev, "%s channel: %d buf_len=%d period_len=%d\n",
+	dev_dbg(imxdma->dev, "%s channel: %d buf_len=%zu period_len=%zu\n",
 			__func__, imxdmac->channel, buf_len, period_len);
 
 	if (list_empty(&imxdmac->ld_free) ||
@@ -926,8 +930,9 @@ static struct dma_async_tx_descriptor *imxdma_prep_dma_memcpy(
 	struct imxdma_engine *imxdma = imxdmac->imxdma;
 	struct imxdma_desc *desc;
 
-	dev_dbg(imxdma->dev, "%s channel: %d src=0x%x dst=0x%x len=%d\n",
-			__func__, imxdmac->channel, src, dest, len);
+	dev_dbg(imxdma->dev, "%s channel: %d src=0x%llx dst=0x%llx len=%zu\n",
+		__func__, imxdmac->channel, (unsigned long long)src,
+		(unsigned long long)dest, len);
 
 	if (list_empty(&imxdmac->ld_free) ||
 	    imxdma_chan_is_doing_cyclic(imxdmac))
@@ -956,9 +961,10 @@ static struct dma_async_tx_descriptor *imxdma_prep_dma_interleaved(
 	struct imxdma_engine *imxdma = imxdmac->imxdma;
 	struct imxdma_desc *desc;
 
-	dev_dbg(imxdma->dev, "%s channel: %d src_start=0x%x dst_start=0x%x\n"
-		"   src_sgl=%s dst_sgl=%s numf=%d frame_size=%d\n", __func__,
-		imxdmac->channel, xt->src_start, xt->dst_start,
+	dev_dbg(imxdma->dev, "%s channel: %d src_start=0x%llx dst_start=0x%llx\n"
+		"   src_sgl=%s dst_sgl=%s numf=%zu frame_size=%zu\n", __func__,
+		imxdmac->channel, (unsigned long long)xt->src_start,
+		(unsigned long long) xt->dst_start,
 		xt->src_sgl ? "true" : "false", xt->dst_sgl ? "true" : "false",
 		xt->numf, xt->frame_size);
 
diff --git a/drivers/dma/imx-sdma.c b/drivers/dma/imx-sdma.c
index c1fd504cae28..c75679d42028 100644
--- a/drivers/dma/imx-sdma.c
+++ b/drivers/dma/imx-sdma.c
@@ -638,7 +638,7 @@ static void mxc_sdma_handle_channel_normal(struct sdma_channel *sdmac)
 	if (error)
 		sdmac->status = DMA_ERROR;
 	else
-		sdmac->status = DMA_SUCCESS;
+		sdmac->status = DMA_COMPLETE;
 
 	dma_cookie_complete(&sdmac->desc);
 	if (sdmac->desc.callback)
@@ -1089,8 +1089,8 @@ static struct dma_async_tx_descriptor *sdma_prep_slave_sg(
 			param &= ~BD_CONT;
 		}
 
-		dev_dbg(sdma->dev, "entry %d: count: %d dma: 0x%08x %s%s\n",
-				i, count, sg->dma_address,
+		dev_dbg(sdma->dev, "entry %d: count: %d dma: %#llx %s%s\n",
+				i, count, (u64)sg->dma_address,
 				param & BD_WRAP ? "wrap" : "",
 				param & BD_INTR ? " intr" : "");
 
@@ -1163,8 +1163,8 @@ static struct dma_async_tx_descriptor *sdma_prep_dma_cyclic(
 		if (i + 1 == num_periods)
 			param |= BD_WRAP;
 
-		dev_dbg(sdma->dev, "entry %d: count: %d dma: 0x%08x %s%s\n",
-				i, period_len, dma_addr,
+		dev_dbg(sdma->dev, "entry %d: count: %d dma: %#llx %s%s\n",
+				i, period_len, (u64)dma_addr,
 				param & BD_WRAP ? "wrap" : "",
 				param & BD_INTR ? " intr" : "");
 
diff --git a/drivers/dma/intel_mid_dma.c b/drivers/dma/intel_mid_dma.c
index a975ebebea8a..1aab8130efa1 100644
--- a/drivers/dma/intel_mid_dma.c
+++ b/drivers/dma/intel_mid_dma.c
@@ -309,7 +309,7 @@ static void midc_descriptor_complete(struct intel_mid_dma_chan *midc,
 		callback_txd(param_txd);
 	}
 	if (midc->raw_tfr) {
-		desc->status = DMA_SUCCESS;
+		desc->status = DMA_COMPLETE;
 		if (desc->lli != NULL) {
 			pci_pool_free(desc->lli_pool, desc->lli,
 						desc->lli_phys);
@@ -481,7 +481,7 @@ static enum dma_status intel_mid_dma_tx_status(struct dma_chan *chan,
 	enum dma_status ret;
 
 	ret = dma_cookie_status(chan, cookie, txstate);
-	if (ret != DMA_SUCCESS) {
+	if (ret != DMA_COMPLETE) {
 		spin_lock_bh(&midc->lock);
 		midc_scan_descriptors(to_middma_device(chan->device), midc);
 		spin_unlock_bh(&midc->lock);
diff --git a/drivers/dma/ioat/dma.c b/drivers/dma/ioat/dma.c
index 5ff6fc1819dc..1a49c777607c 100644
--- a/drivers/dma/ioat/dma.c
+++ b/drivers/dma/ioat/dma.c
@@ -531,21 +531,6 @@ static void ioat1_cleanup_event(unsigned long data)
 	writew(IOAT_CHANCTRL_RUN, ioat->base.reg_base + IOAT_CHANCTRL_OFFSET);
 }
 
-void ioat_dma_unmap(struct ioat_chan_common *chan, enum dma_ctrl_flags flags,
-		    size_t len, struct ioat_dma_descriptor *hw)
-{
-	struct pci_dev *pdev = chan->device->pdev;
-	size_t offset = len - hw->size;
-
-	if (!(flags & DMA_COMPL_SKIP_DEST_UNMAP))
-		ioat_unmap(pdev, hw->dst_addr - offset, len,
-			   PCI_DMA_FROMDEVICE, flags, 1);
-
-	if (!(flags & DMA_COMPL_SKIP_SRC_UNMAP))
-		ioat_unmap(pdev, hw->src_addr - offset, len,
-			   PCI_DMA_TODEVICE, flags, 0);
-}
-
 dma_addr_t ioat_get_current_completion(struct ioat_chan_common *chan)
 {
 	dma_addr_t phys_complete;
@@ -602,7 +587,7 @@ static void __cleanup(struct ioat_dma_chan *ioat, dma_addr_t phys_complete)
 		dump_desc_dbg(ioat, desc);
 		if (tx->cookie) {
 			dma_cookie_complete(tx);
-			ioat_dma_unmap(chan, tx->flags, desc->len, desc->hw);
+			dma_descriptor_unmap(tx);
 			ioat->active -= desc->hw->tx_cnt;
 			if (tx->callback) {
 				tx->callback(tx->callback_param);
@@ -733,7 +718,7 @@ ioat_dma_tx_status(struct dma_chan *c, dma_cookie_t cookie,
 	enum dma_status ret;
 
 	ret = dma_cookie_status(c, cookie, txstate);
-	if (ret == DMA_SUCCESS)
+	if (ret == DMA_COMPLETE)
 		return ret;
 
 	device->cleanup_fn((unsigned long) c);
@@ -833,8 +818,7 @@ int ioat_dma_self_test(struct ioatdma_device *device)
 
 	dma_src = dma_map_single(dev, src, IOAT_TEST_SIZE, DMA_TO_DEVICE);
 	dma_dest = dma_map_single(dev, dest, IOAT_TEST_SIZE, DMA_FROM_DEVICE);
-	flags = DMA_COMPL_SKIP_SRC_UNMAP | DMA_COMPL_SKIP_DEST_UNMAP |
-		DMA_PREP_INTERRUPT;
+	flags = DMA_PREP_INTERRUPT;
 	tx = device->common.device_prep_dma_memcpy(dma_chan, dma_dest, dma_src,
 						   IOAT_TEST_SIZE, flags);
 	if (!tx) {
@@ -859,7 +843,7 @@ int ioat_dma_self_test(struct ioatdma_device *device)
 
 	if (tmo == 0 ||
 	    dma->device_tx_status(dma_chan, cookie, NULL)
-					!= DMA_SUCCESS) {
+					!= DMA_COMPLETE) {
 		dev_err(dev, "Self-test copy timed out, disabling\n");
 		err = -ENODEV;
 		goto unmap_dma;
@@ -885,8 +869,7 @@ static char ioat_interrupt_style[32] = "msix";
 module_param_string(ioat_interrupt_style, ioat_interrupt_style,
 		    sizeof(ioat_interrupt_style), 0644);
 MODULE_PARM_DESC(ioat_interrupt_style,
-		 "set ioat interrupt style: msix (default), "
-		 "msix-single-vector, msi, intx)");
+		 "set ioat interrupt style: msix (default), msi, intx");
 
 /**
  * ioat_dma_setup_interrupts - setup interrupt handler
@@ -904,8 +887,6 @@ int ioat_dma_setup_interrupts(struct ioatdma_device *device)
 
 	if (!strcmp(ioat_interrupt_style, "msix"))
 		goto msix;
-	if (!strcmp(ioat_interrupt_style, "msix-single-vector"))
-		goto msix_single_vector;
 	if (!strcmp(ioat_interrupt_style, "msi"))
 		goto msi;
 	if (!strcmp(ioat_interrupt_style, "intx"))
@@ -920,10 +901,8 @@ msix:
 		device->msix_entries[i].entry = i;
 
 	err = pci_enable_msix(pdev, device->msix_entries, msixcnt);
-	if (err < 0)
+	if (err)
 		goto msi;
-	if (err > 0)
-		goto msix_single_vector;
 
 	for (i = 0; i < msixcnt; i++) {
 		msix = &device->msix_entries[i];
@@ -937,29 +916,13 @@ msix:
 				chan = ioat_chan_by_index(device, j);
 				devm_free_irq(dev, msix->vector, chan);
 			}
-			goto msix_single_vector;
+			goto msi;
 		}
 	}
 	intrctrl |= IOAT_INTRCTRL_MSIX_VECTOR_CONTROL;
 	device->irq_mode = IOAT_MSIX;
 	goto done;
 
-msix_single_vector:
-	msix = &device->msix_entries[0];
-	msix->entry = 0;
-	err = pci_enable_msix(pdev, device->msix_entries, 1);
-	if (err)
-		goto msi;
-
-	err = devm_request_irq(dev, msix->vector, ioat_dma_do_interrupt, 0,
-			       "ioat-msix", device);
-	if (err) {
-		pci_disable_msix(pdev);
-		goto msi;
-	}
-	device->irq_mode = IOAT_MSIX_SINGLE;
-	goto done;
-
 msi:
 	err = pci_enable_msi(pdev);
 	if (err)
@@ -971,7 +934,7 @@ msi:
 		pci_disable_msi(pdev);
 		goto intx;
 	}
-	device->irq_mode = IOAT_MSIX;
+	device->irq_mode = IOAT_MSI;
 	goto done;
 
 intx:
diff --git a/drivers/dma/ioat/dma.h b/drivers/dma/ioat/dma.h
index 54fb7b9ff9aa..11fb877ddca9 100644
--- a/drivers/dma/ioat/dma.h
+++ b/drivers/dma/ioat/dma.h
@@ -52,7 +52,6 @@
 enum ioat_irq_mode {
 	IOAT_NOIRQ = 0,
 	IOAT_MSIX,
-	IOAT_MSIX_SINGLE,
 	IOAT_MSI,
 	IOAT_INTX
 };
@@ -83,7 +82,6 @@ struct ioatdma_device {
 	struct pci_pool *completion_pool;
 #define MAX_SED_POOLS	5
 	struct dma_pool *sed_hw_pool[MAX_SED_POOLS];
-	struct kmem_cache *sed_pool;
 	struct dma_device common;
 	u8 version;
 	struct msix_entry msix_entries[4];
@@ -342,16 +340,6 @@ static inline bool is_ioat_bug(unsigned long err)
 	return !!err;
 }
 
-static inline void ioat_unmap(struct pci_dev *pdev, dma_addr_t addr, size_t len,
-			      int direction, enum dma_ctrl_flags flags, bool dst)
-{
-	if ((dst && (flags & DMA_COMPL_DEST_UNMAP_SINGLE)) ||
-	    (!dst && (flags & DMA_COMPL_SRC_UNMAP_SINGLE)))
-		pci_unmap_single(pdev, addr, len, direction);
-	else
-		pci_unmap_page(pdev, addr, len, direction);
-}
-
 int ioat_probe(struct ioatdma_device *device);
 int ioat_register(struct ioatdma_device *device);
 int ioat1_dma_probe(struct ioatdma_device *dev, int dca);
@@ -363,8 +351,6 @@ void ioat_init_channel(struct ioatdma_device *device,
 		       struct ioat_chan_common *chan, int idx);
 enum dma_status ioat_dma_tx_status(struct dma_chan *c, dma_cookie_t cookie,
 				   struct dma_tx_state *txstate);
-void ioat_dma_unmap(struct ioat_chan_common *chan, enum dma_ctrl_flags flags,
-		    size_t len, struct ioat_dma_descriptor *hw);
 bool ioat_cleanup_preamble(struct ioat_chan_common *chan,
 			   dma_addr_t *phys_complete);
 void ioat_kobject_add(struct ioatdma_device *device, struct kobj_type *type);
diff --git a/drivers/dma/ioat/dma_v2.c b/drivers/dma/ioat/dma_v2.c
index b925e1b1d139..5d3affe7e976 100644
--- a/drivers/dma/ioat/dma_v2.c
+++ b/drivers/dma/ioat/dma_v2.c
@@ -148,7 +148,7 @@ static void __cleanup(struct ioat2_dma_chan *ioat, dma_addr_t phys_complete)
 		tx = &desc->txd;
 		dump_desc_dbg(ioat, desc);
 		if (tx->cookie) {
-			ioat_dma_unmap(chan, tx->flags, desc->len, desc->hw);
+			dma_descriptor_unmap(tx);
 			dma_cookie_complete(tx);
 			if (tx->callback) {
 				tx->callback(tx->callback_param);
diff --git a/drivers/dma/ioat/dma_v2.h b/drivers/dma/ioat/dma_v2.h
index 212d584fe427..470292767e68 100644
--- a/drivers/dma/ioat/dma_v2.h
+++ b/drivers/dma/ioat/dma_v2.h
@@ -157,7 +157,6 @@ static inline void ioat2_set_chainaddr(struct ioat2_dma_chan *ioat, u64 addr)
 
 int ioat2_dma_probe(struct ioatdma_device *dev, int dca);
 int ioat3_dma_probe(struct ioatdma_device *dev, int dca);
-void ioat3_dma_remove(struct ioatdma_device *dev);
 struct dca_provider *ioat2_dca_init(struct pci_dev *pdev, void __iomem *iobase);
 struct dca_provider *ioat3_dca_init(struct pci_dev *pdev, void __iomem *iobase);
 int ioat2_check_space_lock(struct ioat2_dma_chan *ioat, int num_descs);
diff --git a/drivers/dma/ioat/dma_v3.c b/drivers/dma/ioat/dma_v3.c
index d8ececaf1b57..820817e97e62 100644
--- a/drivers/dma/ioat/dma_v3.c
+++ b/drivers/dma/ioat/dma_v3.c
@@ -67,6 +67,8 @@
 #include "dma.h"
 #include "dma_v2.h"
 
+extern struct kmem_cache *ioat3_sed_cache;
+
 /* ioat hardware assumes at least two sources for raid operations */
 #define src_cnt_to_sw(x) ((x) + 2)
 #define src_cnt_to_hw(x) ((x) - 2)
@@ -87,22 +89,8 @@ static const u8 pq_idx_to_field[] = { 1, 4, 5, 0, 1, 2, 4, 5 };
 static const u8 pq16_idx_to_field[] = { 1, 4, 1, 2, 3, 4, 5, 6, 7,
 					0, 1, 2, 3, 4, 5, 6 };
 
-/*
- * technically sources 1 and 2 do not require SED, but the op will have
- * at least 9 descriptors so that's irrelevant.
- */
-static const u8 pq16_idx_to_sed[] = { 0, 0, 0, 0, 0, 0, 0, 0, 0,
-				      1, 1, 1, 1, 1, 1, 1 };
-
 static void ioat3_eh(struct ioat2_dma_chan *ioat);
 
-static dma_addr_t xor_get_src(struct ioat_raw_descriptor *descs[2], int idx)
-{
-	struct ioat_raw_descriptor *raw = descs[xor_idx_to_desc >> idx & 1];
-
-	return raw->field[xor_idx_to_field[idx]];
-}
-
 static void xor_set_src(struct ioat_raw_descriptor *descs[2],
 			dma_addr_t addr, u32 offset, int idx)
 {
@@ -135,12 +123,6 @@ static void pq_set_src(struct ioat_raw_descriptor *descs[2],
 	pq->coef[idx] = coef;
 }
 
-static int sed_get_pq16_pool_idx(int src_cnt)
-{
-
-	return pq16_idx_to_sed[src_cnt];
-}
-
 static bool is_jf_ioat(struct pci_dev *pdev)
 {
 	switch (pdev->device) {
@@ -272,7 +254,7 @@ ioat3_alloc_sed(struct ioatdma_device *device, unsigned int hw_pool)
 	struct ioat_sed_ent *sed;
 	gfp_t flags = __GFP_ZERO | GFP_ATOMIC;
 
-	sed = kmem_cache_alloc(device->sed_pool, flags);
+	sed = kmem_cache_alloc(ioat3_sed_cache, flags);
 	if (!sed)
 		return NULL;
 
@@ -280,7 +262,7 @@ ioat3_alloc_sed(struct ioatdma_device *device, unsigned int hw_pool)
 	sed->hw = dma_pool_alloc(device->sed_hw_pool[hw_pool],
 				 flags, &sed->dma);
 	if (!sed->hw) {
-		kmem_cache_free(device->sed_pool, sed);
+		kmem_cache_free(ioat3_sed_cache, sed);
 		return NULL;
 	}
 
@@ -293,165 +275,7 @@ static void ioat3_free_sed(struct ioatdma_device *device, struct ioat_sed_ent *s
 		return;
 
 	dma_pool_free(device->sed_hw_pool[sed->hw_pool], sed->hw, sed->dma);
-	kmem_cache_free(device->sed_pool, sed);
-}
-
-static void ioat3_dma_unmap(struct ioat2_dma_chan *ioat,
-			    struct ioat_ring_ent *desc, int idx)
-{
-	struct ioat_chan_common *chan = &ioat->base;
-	struct pci_dev *pdev = chan->device->pdev;
-	size_t len = desc->len;
-	size_t offset = len - desc->hw->size;
-	struct dma_async_tx_descriptor *tx = &desc->txd;
-	enum dma_ctrl_flags flags = tx->flags;
-
-	switch (desc->hw->ctl_f.op) {
-	case IOAT_OP_COPY:
-		if (!desc->hw->ctl_f.null) /* skip 'interrupt' ops */
-			ioat_dma_unmap(chan, flags, len, desc->hw);
-		break;
-	case IOAT_OP_XOR_VAL:
-	case IOAT_OP_XOR: {
-		struct ioat_xor_descriptor *xor = desc->xor;
-		struct ioat_ring_ent *ext;
-		struct ioat_xor_ext_descriptor *xor_ex = NULL;
-		int src_cnt = src_cnt_to_sw(xor->ctl_f.src_cnt);
-		struct ioat_raw_descriptor *descs[2];
-		int i;
-
-		if (src_cnt > 5) {
-			ext = ioat2_get_ring_ent(ioat, idx + 1);
-			xor_ex = ext->xor_ex;
-		}
-
-		if (!(flags & DMA_COMPL_SKIP_SRC_UNMAP)) {
-			descs[0] = (struct ioat_raw_descriptor *) xor;
-			descs[1] = (struct ioat_raw_descriptor *) xor_ex;
-			for (i = 0; i < src_cnt; i++) {
-				dma_addr_t src = xor_get_src(descs, i);
-
-				ioat_unmap(pdev, src - offset, len,
-					   PCI_DMA_TODEVICE, flags, 0);
-			}
-
-			/* dest is a source in xor validate operations */
-			if (xor->ctl_f.op == IOAT_OP_XOR_VAL) {
-				ioat_unmap(pdev, xor->dst_addr - offset, len,
-					   PCI_DMA_TODEVICE, flags, 1);
-				break;
-			}
-		}
-
-		if (!(flags & DMA_COMPL_SKIP_DEST_UNMAP))
-			ioat_unmap(pdev, xor->dst_addr - offset, len,
-				   PCI_DMA_FROMDEVICE, flags, 1);
-		break;
-	}
-	case IOAT_OP_PQ_VAL:
-	case IOAT_OP_PQ: {
-		struct ioat_pq_descriptor *pq = desc->pq;
-		struct ioat_ring_ent *ext;
-		struct ioat_pq_ext_descriptor *pq_ex = NULL;
-		int src_cnt = src_cnt_to_sw(pq->ctl_f.src_cnt);
-		struct ioat_raw_descriptor *descs[2];
-		int i;
-
-		if (src_cnt > 3) {
-			ext = ioat2_get_ring_ent(ioat, idx + 1);
-			pq_ex = ext->pq_ex;
-		}
-
-		/* in the 'continue' case don't unmap the dests as sources */
-		if (dmaf_p_disabled_continue(flags))
-			src_cnt--;
-		else if (dmaf_continue(flags))
-			src_cnt -= 3;
-
-		if (!(flags & DMA_COMPL_SKIP_SRC_UNMAP)) {
-			descs[0] = (struct ioat_raw_descriptor *) pq;
-			descs[1] = (struct ioat_raw_descriptor *) pq_ex;
-			for (i = 0; i < src_cnt; i++) {
-				dma_addr_t src = pq_get_src(descs, i);
-
-				ioat_unmap(pdev, src - offset, len,
-					   PCI_DMA_TODEVICE, flags, 0);
-			}
-
-			/* the dests are sources in pq validate operations */
-			if (pq->ctl_f.op == IOAT_OP_XOR_VAL) {
-				if (!(flags & DMA_PREP_PQ_DISABLE_P))
-					ioat_unmap(pdev, pq->p_addr - offset,
-						   len, PCI_DMA_TODEVICE, flags, 0);
-				if (!(flags & DMA_PREP_PQ_DISABLE_Q))
-					ioat_unmap(pdev, pq->q_addr - offset,
-						   len, PCI_DMA_TODEVICE, flags, 0);
-				break;
-			}
-		}
-
-		if (!(flags & DMA_COMPL_SKIP_DEST_UNMAP)) {
-			if (!(flags & DMA_PREP_PQ_DISABLE_P))
-				ioat_unmap(pdev, pq->p_addr - offset, len,
-					   PCI_DMA_BIDIRECTIONAL, flags, 1);
-			if (!(flags & DMA_PREP_PQ_DISABLE_Q))
-				ioat_unmap(pdev, pq->q_addr - offset, len,
-					   PCI_DMA_BIDIRECTIONAL, flags, 1);
-		}
-		break;
-	}
-	case IOAT_OP_PQ_16S:
-	case IOAT_OP_PQ_VAL_16S: {
-		struct ioat_pq_descriptor *pq = desc->pq;
-		int src_cnt = src16_cnt_to_sw(pq->ctl_f.src_cnt);
-		struct ioat_raw_descriptor *descs[4];
-		int i;
-
-		/* in the 'continue' case don't unmap the dests as sources */
-		if (dmaf_p_disabled_continue(flags))
-			src_cnt--;
-		else if (dmaf_continue(flags))
-			src_cnt -= 3;
-
-		if (!(flags & DMA_COMPL_SKIP_SRC_UNMAP)) {
-			descs[0] = (struct ioat_raw_descriptor *)pq;
-			descs[1] = (struct ioat_raw_descriptor *)(desc->sed->hw);
-			descs[2] = (struct ioat_raw_descriptor *)(&desc->sed->hw->b[0]);
-			for (i = 0; i < src_cnt; i++) {
-				dma_addr_t src = pq16_get_src(descs, i);
-
-				ioat_unmap(pdev, src - offset, len,
-					   PCI_DMA_TODEVICE, flags, 0);
-			}
-
-			/* the dests are sources in pq validate operations */
-			if (pq->ctl_f.op == IOAT_OP_XOR_VAL) {
-				if (!(flags & DMA_PREP_PQ_DISABLE_P))
-					ioat_unmap(pdev, pq->p_addr - offset,
-						   len, PCI_DMA_TODEVICE,
-						   flags, 0);
-				if (!(flags & DMA_PREP_PQ_DISABLE_Q))
-					ioat_unmap(pdev, pq->q_addr - offset,
-						   len, PCI_DMA_TODEVICE,
-						   flags, 0);
-				break;
-			}
-		}
-
-		if (!(flags & DMA_COMPL_SKIP_DEST_UNMAP)) {
-			if (!(flags & DMA_PREP_PQ_DISABLE_P))
-				ioat_unmap(pdev, pq->p_addr - offset, len,
-					   PCI_DMA_BIDIRECTIONAL, flags, 1);
-			if (!(flags & DMA_PREP_PQ_DISABLE_Q))
-				ioat_unmap(pdev, pq->q_addr - offset, len,
-					   PCI_DMA_BIDIRECTIONAL, flags, 1);
-		}
-		break;
-	}
-	default:
-		dev_err(&pdev->dev, "%s: unknown op type: %#x\n",
-			__func__, desc->hw->ctl_f.op);
-	}
+	kmem_cache_free(ioat3_sed_cache, sed);
 }
 
 static bool desc_has_ext(struct ioat_ring_ent *desc)
@@ -577,7 +401,7 @@ static void __cleanup(struct ioat2_dma_chan *ioat, dma_addr_t phys_complete)
 		tx = &desc->txd;
 		if (tx->cookie) {
 			dma_cookie_complete(tx);
-			ioat3_dma_unmap(ioat, desc, idx + i);
+			dma_descriptor_unmap(tx);
 			if (tx->callback) {
 				tx->callback(tx->callback_param);
 				tx->callback = NULL;
@@ -807,7 +631,7 @@ ioat3_tx_status(struct dma_chan *c, dma_cookie_t cookie,
 	enum dma_status ret;
 
 	ret = dma_cookie_status(c, cookie, txstate);
-	if (ret == DMA_SUCCESS)
+	if (ret == DMA_COMPLETE)
 		return ret;
 
 	ioat3_cleanup(ioat);
@@ -1129,9 +953,6 @@ __ioat3_prep_pq16_lock(struct dma_chan *c, enum sum_check_flags *result,
 	u8 op;
 	int i, s, idx, num_descs;
 
-	/* this function only handles src_cnt 9 - 16 */
-	BUG_ON(src_cnt < 9);
-
 	/* this function is only called with 9-16 sources */
 	op = result ? IOAT_OP_PQ_VAL_16S : IOAT_OP_PQ_16S;
 
@@ -1159,8 +980,7 @@ __ioat3_prep_pq16_lock(struct dma_chan *c, enum sum_check_flags *result,
 
 		descs[0] = (struct ioat_raw_descriptor *) pq;
 
-		desc->sed = ioat3_alloc_sed(device,
-					    sed_get_pq16_pool_idx(src_cnt));
+		desc->sed = ioat3_alloc_sed(device, (src_cnt-2) >> 3);
 		if (!desc->sed) {
 			dev_err(to_dev(chan),
 				"%s: no free sed entries\n", __func__);
@@ -1218,13 +1038,21 @@ __ioat3_prep_pq16_lock(struct dma_chan *c, enum sum_check_flags *result,
 	return &desc->txd;
 }
 
+static int src_cnt_flags(unsigned int src_cnt, unsigned long flags)
+{
+	if (dmaf_p_disabled_continue(flags))
+		return src_cnt + 1;
+	else if (dmaf_continue(flags))
+		return src_cnt + 3;
+	else
+		return src_cnt;
+}
+
 static struct dma_async_tx_descriptor *
 ioat3_prep_pq(struct dma_chan *chan, dma_addr_t *dst, dma_addr_t *src,
 	      unsigned int src_cnt, const unsigned char *scf, size_t len,
 	      unsigned long flags)
 {
-	struct dma_device *dma = chan->device;
-
 	/* specify valid address for disabled result */
 	if (flags & DMA_PREP_PQ_DISABLE_P)
 		dst[0] = dst[1];
@@ -1244,7 +1072,7 @@ ioat3_prep_pq(struct dma_chan *chan, dma_addr_t *dst, dma_addr_t *src,
 		single_source_coef[0] = scf[0];
 		single_source_coef[1] = 0;
 
-		return (src_cnt > 8) && (dma->max_pq > 8) ?
+		return src_cnt_flags(src_cnt, flags) > 8 ?
 			__ioat3_prep_pq16_lock(chan, NULL, dst, single_source,
 					       2, single_source_coef, len,
 					       flags) :
@@ -1252,7 +1080,7 @@ ioat3_prep_pq(struct dma_chan *chan, dma_addr_t *dst, dma_addr_t *src,
 					     single_source_coef, len, flags);
 
 	} else {
-		return (src_cnt > 8) && (dma->max_pq > 8) ?
+		return src_cnt_flags(src_cnt, flags) > 8 ?
 			__ioat3_prep_pq16_lock(chan, NULL, dst, src, src_cnt,
 					       scf, len, flags) :
 			__ioat3_prep_pq_lock(chan, NULL, dst, src, src_cnt,
@@ -1265,8 +1093,6 @@ ioat3_prep_pq_val(struct dma_chan *chan, dma_addr_t *pq, dma_addr_t *src,
 		  unsigned int src_cnt, const unsigned char *scf, size_t len,
 		  enum sum_check_flags *pqres, unsigned long flags)
 {
-	struct dma_device *dma = chan->device;
-
 	/* specify valid address for disabled result */
 	if (flags & DMA_PREP_PQ_DISABLE_P)
 		pq[0] = pq[1];
@@ -1278,7 +1104,7 @@ ioat3_prep_pq_val(struct dma_chan *chan, dma_addr_t *pq, dma_addr_t *src,
 	 */
 	*pqres = 0;
 
-	return (src_cnt > 8) && (dma->max_pq > 8) ?
+	return src_cnt_flags(src_cnt, flags) > 8 ?
 		__ioat3_prep_pq16_lock(chan, pqres, pq, src, src_cnt, scf, len,
 				       flags) :
 		__ioat3_prep_pq_lock(chan, pqres, pq, src, src_cnt, scf, len,
@@ -1289,7 +1115,6 @@ static struct dma_async_tx_descriptor *
 ioat3_prep_pqxor(struct dma_chan *chan, dma_addr_t dst, dma_addr_t *src,
 		 unsigned int src_cnt, size_t len, unsigned long flags)
 {
-	struct dma_device *dma = chan->device;
 	unsigned char scf[src_cnt];
 	dma_addr_t pq[2];
 
@@ -1298,7 +1123,7 @@ ioat3_prep_pqxor(struct dma_chan *chan, dma_addr_t dst, dma_addr_t *src,
 	flags |= DMA_PREP_PQ_DISABLE_Q;
 	pq[1] = dst; /* specify valid address for disabled result */
 
-	return (src_cnt > 8) && (dma->max_pq > 8) ?
+	return src_cnt_flags(src_cnt, flags) > 8 ?
 		__ioat3_prep_pq16_lock(chan, NULL, pq, src, src_cnt, scf, len,
 				       flags) :
 		__ioat3_prep_pq_lock(chan, NULL, pq, src, src_cnt, scf, len,
@@ -1310,7 +1135,6 @@ ioat3_prep_pqxor_val(struct dma_chan *chan, dma_addr_t *src,
 		     unsigned int src_cnt, size_t len,
 		     enum sum_check_flags *result, unsigned long flags)
 {
-	struct dma_device *dma = chan->device;
 	unsigned char scf[src_cnt];
 	dma_addr_t pq[2];
 
@@ -1324,8 +1148,7 @@ ioat3_prep_pqxor_val(struct dma_chan *chan, dma_addr_t *src,
 	flags |= DMA_PREP_PQ_DISABLE_Q;
 	pq[1] = pq[0]; /* specify valid address for disabled result */
 
-
-	return (src_cnt > 8) && (dma->max_pq > 8) ?
+	return src_cnt_flags(src_cnt, flags) > 8 ?
 		__ioat3_prep_pq16_lock(chan, result, pq, &src[1], src_cnt - 1,
 				       scf, len, flags) :
 		__ioat3_prep_pq_lock(chan, result, pq, &src[1], src_cnt - 1,
@@ -1444,9 +1267,7 @@ static int ioat_xor_val_self_test(struct ioatdma_device *device)
 					   DMA_TO_DEVICE);
 	tx = dma->device_prep_dma_xor(dma_chan, dest_dma, dma_srcs,
 				      IOAT_NUM_SRC_TEST, PAGE_SIZE,
-				      DMA_PREP_INTERRUPT |
-				      DMA_COMPL_SKIP_SRC_UNMAP |
-				      DMA_COMPL_SKIP_DEST_UNMAP);
+				      DMA_PREP_INTERRUPT);
 
 	if (!tx) {
 		dev_err(dev, "Self-test xor prep failed\n");
@@ -1468,7 +1289,7 @@ static int ioat_xor_val_self_test(struct ioatdma_device *device)
 
 	tmo = wait_for_completion_timeout(&cmp, msecs_to_jiffies(3000));
 
-	if (dma->device_tx_status(dma_chan, cookie, NULL) != DMA_SUCCESS) {
+	if (dma->device_tx_status(dma_chan, cookie, NULL) != DMA_COMPLETE) {
 		dev_err(dev, "Self-test xor timed out\n");
 		err = -ENODEV;
 		goto dma_unmap;
@@ -1507,9 +1328,7 @@ static int ioat_xor_val_self_test(struct ioatdma_device *device)
 					   DMA_TO_DEVICE);
 	tx = dma->device_prep_dma_xor_val(dma_chan, dma_srcs,
 					  IOAT_NUM_SRC_TEST + 1, PAGE_SIZE,
-					  &xor_val_result, DMA_PREP_INTERRUPT |
-					  DMA_COMPL_SKIP_SRC_UNMAP |
-					  DMA_COMPL_SKIP_DEST_UNMAP);
+					  &xor_val_result, DMA_PREP_INTERRUPT);
 	if (!tx) {
 		dev_err(dev, "Self-test zero prep failed\n");
 		err = -ENODEV;
@@ -1530,7 +1349,7 @@ static int ioat_xor_val_self_test(struct ioatdma_device *device)
 
 	tmo = wait_for_completion_timeout(&cmp, msecs_to_jiffies(3000));
 
-	if (dma->device_tx_status(dma_chan, cookie, NULL) != DMA_SUCCESS) {
+	if (dma->device_tx_status(dma_chan, cookie, NULL) != DMA_COMPLETE) {
 		dev_err(dev, "Self-test validate timed out\n");
 		err = -ENODEV;
 		goto dma_unmap;
@@ -1545,6 +1364,8 @@ static int ioat_xor_val_self_test(struct ioatdma_device *device)
 		goto free_resources;
 	}
 
+	memset(page_address(dest), 0, PAGE_SIZE);
+
 	/* test for non-zero parity sum */
 	op = IOAT_OP_XOR_VAL;
 
@@ -1554,9 +1375,7 @@ static int ioat_xor_val_self_test(struct ioatdma_device *device)
 					   DMA_TO_DEVICE);
 	tx = dma->device_prep_dma_xor_val(dma_chan, dma_srcs,
 					  IOAT_NUM_SRC_TEST + 1, PAGE_SIZE,
-					  &xor_val_result, DMA_PREP_INTERRUPT |
-					  DMA_COMPL_SKIP_SRC_UNMAP |
-					  DMA_COMPL_SKIP_DEST_UNMAP);
+					  &xor_val_result, DMA_PREP_INTERRUPT);
 	if (!tx) {
 		dev_err(dev, "Self-test 2nd zero prep failed\n");
 		err = -ENODEV;
@@ -1577,7 +1396,7 @@ static int ioat_xor_val_self_test(struct ioatdma_device *device)
 
 	tmo = wait_for_completion_timeout(&cmp, msecs_to_jiffies(3000));
 
-	if (dma->device_tx_status(dma_chan, cookie, NULL) != DMA_SUCCESS) {
+	if (dma->device_tx_status(dma_chan, cookie, NULL) != DMA_COMPLETE) {
 		dev_err(dev, "Self-test 2nd validate timed out\n");
 		err = -ENODEV;
 		goto dma_unmap;
@@ -1630,52 +1449,36 @@ static int ioat3_dma_self_test(struct ioatdma_device *device)
 
 static int ioat3_irq_reinit(struct ioatdma_device *device)
 {
-	int msixcnt = device->common.chancnt;
 	struct pci_dev *pdev = device->pdev;
-	int i;
-	struct msix_entry *msix;
-	struct ioat_chan_common *chan;
-	int err = 0;
+	int irq = pdev->irq, i;
+
+	if (!is_bwd_ioat(pdev))
+		return 0;
 
 	switch (device->irq_mode) {
 	case IOAT_MSIX:
+		for (i = 0; i < device->common.chancnt; i++) {
+			struct msix_entry *msix = &device->msix_entries[i];
+			struct ioat_chan_common *chan;
 
-		for (i = 0; i < msixcnt; i++) {
-			msix = &device->msix_entries[i];
 			chan = ioat_chan_by_index(device, i);
 			devm_free_irq(&pdev->dev, msix->vector, chan);
 		}
 
 		pci_disable_msix(pdev);
 		break;
-
-	case IOAT_MSIX_SINGLE:
-		msix = &device->msix_entries[0];
-		chan = ioat_chan_by_index(device, 0);
-		devm_free_irq(&pdev->dev, msix->vector, chan);
-		pci_disable_msix(pdev);
-		break;
-
 	case IOAT_MSI:
-		chan = ioat_chan_by_index(device, 0);
-		devm_free_irq(&pdev->dev, pdev->irq, chan);
 		pci_disable_msi(pdev);
-		break;
-
+		/* fall through */
 	case IOAT_INTX:
-		chan = ioat_chan_by_index(device, 0);
-		devm_free_irq(&pdev->dev, pdev->irq, chan);
+		devm_free_irq(&pdev->dev, irq, device);
 		break;
-
 	default:
 		return 0;
 	}
-
 	device->irq_mode = IOAT_NOIRQ;
 
-	err = ioat_dma_setup_interrupts(device);
-
-	return err;
+	return ioat_dma_setup_interrupts(device);
 }
 
 static int ioat3_reset_hw(struct ioat_chan_common *chan)
@@ -1718,14 +1521,12 @@ static int ioat3_reset_hw(struct ioat_chan_common *chan)
 	}
 
 	err = ioat2_reset_sync(chan, msecs_to_jiffies(200));
-	if (err) {
-		dev_err(&pdev->dev, "Failed to reset!\n");
-		return err;
-	}
-
-	if (device->irq_mode != IOAT_NOIRQ && is_bwd_ioat(pdev))
+	if (!err)
 		err = ioat3_irq_reinit(device);
 
+	if (err)
+		dev_err(&pdev->dev, "Failed to reset: %d\n", err);
+
 	return err;
 }
 
@@ -1835,21 +1636,15 @@ int ioat3_dma_probe(struct ioatdma_device *device, int dca)
 		char pool_name[14];
 		int i;
 
-		/* allocate sw descriptor pool for SED */
-		device->sed_pool = kmem_cache_create("ioat_sed",
-				sizeof(struct ioat_sed_ent), 0, 0, NULL);
-		if (!device->sed_pool)
-			return -ENOMEM;
-
 		for (i = 0; i < MAX_SED_POOLS; i++) {
 			snprintf(pool_name, 14, "ioat_hw%d_sed", i);
 
 			/* allocate SED DMA pool */
-			device->sed_hw_pool[i] = dma_pool_create(pool_name,
+			device->sed_hw_pool[i] = dmam_pool_create(pool_name,
 					&pdev->dev,
 					SED_SIZE * (i + 1), 64, 0);
 			if (!device->sed_hw_pool[i])
-				goto sed_pool_cleanup;
+				return -ENOMEM;
 
 		}
 	}
@@ -1875,28 +1670,4 @@ int ioat3_dma_probe(struct ioatdma_device *device, int dca)
 		device->dca = ioat3_dca_init(pdev, device->reg_base);
 
 	return 0;
-
-sed_pool_cleanup:
-	if (device->sed_pool) {
-		int i;
-		kmem_cache_destroy(device->sed_pool);
-
-		for (i = 0; i < MAX_SED_POOLS; i++)
-			if (device->sed_hw_pool[i])
-				dma_pool_destroy(device->sed_hw_pool[i]);
-	}
-
-	return -ENOMEM;
-}
-
-void ioat3_dma_remove(struct ioatdma_device *device)
-{
-	if (device->sed_pool) {
-		int i;
-		kmem_cache_destroy(device->sed_pool);
-
-		for (i = 0; i < MAX_SED_POOLS; i++)
-			if (device->sed_hw_pool[i])
-				dma_pool_destroy(device->sed_hw_pool[i]);
-	}
 }
diff --git a/drivers/dma/ioat/pci.c b/drivers/dma/ioat/pci.c
index 2c8d560e6334..1d051cd045db 100644
--- a/drivers/dma/ioat/pci.c
+++ b/drivers/dma/ioat/pci.c
@@ -123,6 +123,7 @@ module_param(ioat_dca_enabled, int, 0644);
 MODULE_PARM_DESC(ioat_dca_enabled, "control support of dca service (default: 1)");
 
 struct kmem_cache *ioat2_cache;
+struct kmem_cache *ioat3_sed_cache;
 
 #define DRV_NAME "ioatdma"
 
@@ -207,9 +208,6 @@ static void ioat_remove(struct pci_dev *pdev)
 	if (!device)
 		return;
 
-	if (device->version >= IOAT_VER_3_0)
-		ioat3_dma_remove(device);
-
 	dev_err(&pdev->dev, "Removing dma and dca services\n");
 	if (device->dca) {
 		unregister_dca_provider(device->dca, &pdev->dev);
@@ -221,7 +219,7 @@ static void ioat_remove(struct pci_dev *pdev)
 
 static int __init ioat_init_module(void)
 {
-	int err;
+	int err = -ENOMEM;
 
 	pr_info("%s: Intel(R) QuickData Technology Driver %s\n",
 		DRV_NAME, IOAT_DMA_VERSION);
@@ -231,9 +229,21 @@ static int __init ioat_init_module(void)
 	if (!ioat2_cache)
 		return -ENOMEM;
 
+	ioat3_sed_cache = KMEM_CACHE(ioat_sed_ent, 0);
+	if (!ioat3_sed_cache)
+		goto err_ioat2_cache;
+
 	err = pci_register_driver(&ioat_pci_driver);
 	if (err)
-		kmem_cache_destroy(ioat2_cache);
+		goto err_ioat3_cache;
+
+	return 0;
+
+ err_ioat3_cache:
+	kmem_cache_destroy(ioat3_sed_cache);
+
+ err_ioat2_cache:
+	kmem_cache_destroy(ioat2_cache);
 
 	return err;
 }
diff --git a/drivers/dma/iop-adma.c b/drivers/dma/iop-adma.c
index dd8b44a56e5d..c56137bc3868 100644
--- a/drivers/dma/iop-adma.c
+++ b/drivers/dma/iop-adma.c
@@ -61,80 +61,6 @@ static void iop_adma_free_slots(struct iop_adma_desc_slot *slot)
 	}
 }
 
-static void
-iop_desc_unmap(struct iop_adma_chan *iop_chan, struct iop_adma_desc_slot *desc)
-{
-	struct dma_async_tx_descriptor *tx = &desc->async_tx;
-	struct iop_adma_desc_slot *unmap = desc->group_head;
-	struct device *dev = &iop_chan->device->pdev->dev;
-	u32 len = unmap->unmap_len;
-	enum dma_ctrl_flags flags = tx->flags;
-	u32 src_cnt;
-	dma_addr_t addr;
-	dma_addr_t dest;
-
-	src_cnt = unmap->unmap_src_cnt;
-	dest = iop_desc_get_dest_addr(unmap, iop_chan);
-	if (!(flags & DMA_COMPL_SKIP_DEST_UNMAP)) {
-		enum dma_data_direction dir;
-
-		if (src_cnt > 1) /* is xor? */
-			dir = DMA_BIDIRECTIONAL;
-		else
-			dir = DMA_FROM_DEVICE;
-
-		dma_unmap_page(dev, dest, len, dir);
-	}
-
-	if (!(flags & DMA_COMPL_SKIP_SRC_UNMAP)) {
-		while (src_cnt--) {
-			addr = iop_desc_get_src_addr(unmap, iop_chan, src_cnt);
-			if (addr == dest)
-				continue;
-			dma_unmap_page(dev, addr, len, DMA_TO_DEVICE);
-		}
-	}
-	desc->group_head = NULL;
-}
-
-static void
-iop_desc_unmap_pq(struct iop_adma_chan *iop_chan, struct iop_adma_desc_slot *desc)
-{
-	struct dma_async_tx_descriptor *tx = &desc->async_tx;
-	struct iop_adma_desc_slot *unmap = desc->group_head;
-	struct device *dev = &iop_chan->device->pdev->dev;
-	u32 len = unmap->unmap_len;
-	enum dma_ctrl_flags flags = tx->flags;
-	u32 src_cnt = unmap->unmap_src_cnt;
-	dma_addr_t pdest = iop_desc_get_dest_addr(unmap, iop_chan);
-	dma_addr_t qdest = iop_desc_get_qdest_addr(unmap, iop_chan);
-	int i;
-
-	if (tx->flags & DMA_PREP_CONTINUE)
-		src_cnt -= 3;
-
-	if (!(flags & DMA_COMPL_SKIP_DEST_UNMAP) && !desc->pq_check_result) {
-		dma_unmap_page(dev, pdest, len, DMA_BIDIRECTIONAL);
-		dma_unmap_page(dev, qdest, len, DMA_BIDIRECTIONAL);
-	}
-
-	if (!(flags & DMA_COMPL_SKIP_SRC_UNMAP)) {
-		dma_addr_t addr;
-
-		for (i = 0; i < src_cnt; i++) {
-			addr = iop_desc_get_src_addr(unmap, iop_chan, i);
-			dma_unmap_page(dev, addr, len, DMA_TO_DEVICE);
-		}
-		if (desc->pq_check_result) {
-			dma_unmap_page(dev, pdest, len, DMA_TO_DEVICE);
-			dma_unmap_page(dev, qdest, len, DMA_TO_DEVICE);
-		}
-	}
-
-	desc->group_head = NULL;
-}
-
-
 static dma_cookie_t
 iop_adma_run_tx_complete_actions(struct iop_adma_desc_slot *desc,
 	struct iop_adma_chan *iop_chan, dma_cookie_t cookie)
@@ -152,15 +78,9 @@ iop_adma_run_tx_complete_actions(struct iop_adma_desc_slot *desc,
 		if (tx->callback)
 			tx->callback(tx->callback_param);
 
-		/* unmap dma addresses
-		 * (unmap_single vs unmap_page?)
-		 */
-		if (desc->group_head && desc->unmap_len) {
-			if (iop_desc_is_pq(desc))
-				iop_desc_unmap_pq(iop_chan, desc);
-			else
-				iop_desc_unmap(iop_chan, desc);
-		}
+		dma_descriptor_unmap(tx);
+		if (desc->group_head)
+			desc->group_head = NULL;
 	}
 
 	/* run dependent operations */
@@ -591,7 +511,6 @@ iop_adma_prep_dma_interrupt(struct dma_chan *chan, unsigned long flags)
 	if (sw_desc) {
 		grp_start = sw_desc->group_head;
 		iop_desc_init_interrupt(grp_start, iop_chan);
-		grp_start->unmap_len = 0;
 		sw_desc->async_tx.flags = flags;
 	}
 	spin_unlock_bh(&iop_chan->lock);
@@ -623,8 +542,6 @@ iop_adma_prep_dma_memcpy(struct dma_chan *chan, dma_addr_t dma_dest,
 		iop_desc_set_byte_count(grp_start, iop_chan, len);
 		iop_desc_set_dest_addr(grp_start, iop_chan, dma_dest);
 		iop_desc_set_memcpy_src_addr(grp_start, dma_src);
-		sw_desc->unmap_src_cnt = 1;
-		sw_desc->unmap_len = len;
 		sw_desc->async_tx.flags = flags;
 	}
 	spin_unlock_bh(&iop_chan->lock);
@@ -657,8 +574,6 @@ iop_adma_prep_dma_xor(struct dma_chan *chan, dma_addr_t dma_dest,
 		iop_desc_init_xor(grp_start, src_cnt, flags);
 		iop_desc_set_byte_count(grp_start, iop_chan, len);
 		iop_desc_set_dest_addr(grp_start, iop_chan, dma_dest);
-		sw_desc->unmap_src_cnt = src_cnt;
-		sw_desc->unmap_len = len;
 		sw_desc->async_tx.flags = flags;
 		while (src_cnt--)
 			iop_desc_set_xor_src_addr(grp_start, src_cnt,
@@ -694,8 +609,6 @@ iop_adma_prep_dma_xor_val(struct dma_chan *chan, dma_addr_t *dma_src,
 		grp_start->xor_check_result = result;
 		pr_debug("\t%s: grp_start->xor_check_result: %p\n",
 			__func__, grp_start->xor_check_result);
-		sw_desc->unmap_src_cnt = src_cnt;
-		sw_desc->unmap_len = len;
 		sw_desc->async_tx.flags = flags;
 		while (src_cnt--)
 			iop_desc_set_zero_sum_src_addr(grp_start, src_cnt,
@@ -748,8 +661,6 @@ iop_adma_prep_dma_pq(struct dma_chan *chan, dma_addr_t *dst, dma_addr_t *src,
 			dst[0] = dst[1] & 0x7;
 
 		iop_desc_set_pq_addr(g, dst);
-		sw_desc->unmap_src_cnt = src_cnt;
-		sw_desc->unmap_len = len;
 		sw_desc->async_tx.flags = flags;
 		for (i = 0; i < src_cnt; i++)
 			iop_desc_set_pq_src_addr(g, i, src[i], scf[i]);
@@ -804,8 +715,6 @@ iop_adma_prep_dma_pq_val(struct dma_chan *chan, dma_addr_t *pq, dma_addr_t *src,
 		g->pq_check_result = pqres;
 		pr_debug("\t%s: g->pq_check_result: %p\n",
 			__func__, g->pq_check_result);
-		sw_desc->unmap_src_cnt = src_cnt+2;
-		sw_desc->unmap_len = len;
 		sw_desc->async_tx.flags = flags;
 		while (src_cnt--)
 			iop_desc_set_pq_zero_sum_src_addr(g, src_cnt,
@@ -864,7 +773,7 @@ static enum dma_status iop_adma_status(struct dma_chan *chan,
 	int ret;
 
 	ret = dma_cookie_status(chan, cookie, txstate);
-	if (ret == DMA_SUCCESS)
+	if (ret == DMA_COMPLETE)
 		return ret;
 
 	iop_adma_slot_cleanup(iop_chan);
@@ -983,7 +892,7 @@ static int iop_adma_memcpy_self_test(struct iop_adma_device *device)
 	msleep(1);
 
 	if (iop_adma_status(dma_chan, cookie, NULL) !=
-			DMA_SUCCESS) {
+			DMA_COMPLETE) {
 		dev_err(dma_chan->device->dev,
 			"Self-test copy timed out, disabling\n");
 		err = -ENODEV;
@@ -1083,7 +992,7 @@ iop_adma_xor_val_self_test(struct iop_adma_device *device)
 	msleep(8);
 
 	if (iop_adma_status(dma_chan, cookie, NULL) !=
-		DMA_SUCCESS) {
+		DMA_COMPLETE) {
 		dev_err(dma_chan->device->dev,
 			"Self-test xor timed out, disabling\n");
 		err = -ENODEV;
@@ -1129,7 +1038,7 @@ iop_adma_xor_val_self_test(struct iop_adma_device *device)
 	iop_adma_issue_pending(dma_chan);
 	msleep(8);
 
-	if (iop_adma_status(dma_chan, cookie, NULL) != DMA_SUCCESS) {
+	if (iop_adma_status(dma_chan, cookie, NULL) != DMA_COMPLETE) {
 		dev_err(dma_chan->device->dev,
 			"Self-test zero sum timed out, disabling\n");
 		err = -ENODEV;
@@ -1158,7 +1067,7 @@ iop_adma_xor_val_self_test(struct iop_adma_device *device)
 	iop_adma_issue_pending(dma_chan);
 	msleep(8);
 
-	if (iop_adma_status(dma_chan, cookie, NULL) != DMA_SUCCESS) {
+	if (iop_adma_status(dma_chan, cookie, NULL) != DMA_COMPLETE) {
 		dev_err(dma_chan->device->dev,
 			"Self-test non-zero sum timed out, disabling\n");
 		err = -ENODEV;
@@ -1254,7 +1163,7 @@ iop_adma_pq_zero_sum_self_test(struct iop_adma_device *device)
 	msleep(8);
 
 	if (iop_adma_status(dma_chan, cookie, NULL) !=
-		DMA_SUCCESS) {
+		DMA_COMPLETE) {
 		dev_err(dev, "Self-test pq timed out, disabling\n");
 		err = -ENODEV;
 		goto free_resources;
@@ -1291,7 +1200,7 @@ iop_adma_pq_zero_sum_self_test(struct iop_adma_device *device)
 	msleep(8);
 
 	if (iop_adma_status(dma_chan, cookie, NULL) !=
-		DMA_SUCCESS) {
+		DMA_COMPLETE) {
 		dev_err(dev, "Self-test pq-zero-sum timed out, disabling\n");
 		err = -ENODEV;
 		goto free_resources;
@@ -1323,7 +1232,7 @@ iop_adma_pq_zero_sum_self_test(struct iop_adma_device *device)
 	msleep(8);
 
 	if (iop_adma_status(dma_chan, cookie, NULL) !=
-		DMA_SUCCESS) {
+		DMA_COMPLETE) {
 		dev_err(dev, "Self-test !pq-zero-sum timed out, disabling\n");
 		err = -ENODEV;
 		goto free_resources;
diff --git a/drivers/dma/ipu/ipu_idmac.c b/drivers/dma/ipu/ipu_idmac.c
index cb9c0bc317e8..128ca143486d 100644
--- a/drivers/dma/ipu/ipu_idmac.c
+++ b/drivers/dma/ipu/ipu_idmac.c
@@ -1232,8 +1232,10 @@ static irqreturn_t idmac_interrupt(int irq, void *dev_id)
 	desc = list_entry(ichan->queue.next, struct idmac_tx_desc, list);
 	descnew = desc;
 
-	dev_dbg(dev, "IDMAC irq %d, dma 0x%08x, next dma 0x%08x, current %d, curbuf 0x%08x\n",
-		irq, sg_dma_address(*sg), sgnext ? sg_dma_address(sgnext) : 0, ichan->active_buffer, curbuf);
+	dev_dbg(dev, "IDMAC irq %d, dma %#llx, next dma %#llx, current %d, curbuf %#x\n",
+		irq, (u64)sg_dma_address(*sg),
+		sgnext ? (u64)sg_dma_address(sgnext) : 0,
+		ichan->active_buffer, curbuf);
 
 	/* Find the descriptor of sgnext */
 	sgnew = idmac_sg_next(ichan, &descnew, *sg);
diff --git a/drivers/dma/k3dma.c b/drivers/dma/k3dma.c
index a2c330f5f952..e26075408e9b 100644
--- a/drivers/dma/k3dma.c
+++ b/drivers/dma/k3dma.c
@@ -344,7 +344,7 @@ static enum dma_status k3_dma_tx_status(struct dma_chan *chan,
 	size_t bytes = 0;
 
 	ret = dma_cookie_status(&c->vc.chan, cookie, state);
-	if (ret == DMA_SUCCESS)
+	if (ret == DMA_COMPLETE)
 		return ret;
 
 	spin_lock_irqsave(&c->vc.lock, flags);
@@ -693,7 +693,7 @@ static int k3_dma_probe(struct platform_device *op)
 
 	irq = platform_get_irq(op, 0);
 	ret = devm_request_irq(&op->dev, irq,
-			k3_dma_int_handler, IRQF_DISABLED, DRIVER_NAME, d);
+			k3_dma_int_handler, 0, DRIVER_NAME, d);
 	if (ret)
 		return ret;
 
diff --git a/drivers/dma/mmp_pdma.c b/drivers/dma/mmp_pdma.c
index ff8d7827f8cb..dcb1e05149a7 100644
--- a/drivers/dma/mmp_pdma.c
+++ b/drivers/dma/mmp_pdma.c
@@ -798,8 +798,7 @@ static void dma_do_tasklet(unsigned long data)
 		 * move the descriptors to a temporary list so we can drop
 		 * the lock during the entire cleanup operation
 		 */
-		list_del(&desc->node);
-		list_add(&desc->node, &chain_cleanup);
+		list_move(&desc->node, &chain_cleanup);
 
 		/*
 		 * Look for the first list entry which has the ENDIRQEN flag
@@ -863,7 +862,7 @@ static int mmp_pdma_chan_init(struct mmp_pdma_device *pdev,
 
 	if (irq) {
 		ret = devm_request_irq(pdev->dev, irq,
-			mmp_pdma_chan_handler, IRQF_DISABLED, "pdma", phy);
+			mmp_pdma_chan_handler, 0, "pdma", phy);
 		if (ret) {
 			dev_err(pdev->dev, "channel request irq fail!\n");
 			return ret;
@@ -970,7 +969,7 @@ static int mmp_pdma_probe(struct platform_device *op)
 		/* all chan share one irq, demux inside */
 		irq = platform_get_irq(op, 0);
 		ret = devm_request_irq(pdev->dev, irq,
-			mmp_pdma_int_handler, IRQF_DISABLED, "pdma", pdev);
+			mmp_pdma_int_handler, 0, "pdma", pdev);
 		if (ret)
 			return ret;
 	}
diff --git a/drivers/dma/mmp_tdma.c b/drivers/dma/mmp_tdma.c
index d3b6358e5a27..3ddacc14a736 100644
--- a/drivers/dma/mmp_tdma.c
+++ b/drivers/dma/mmp_tdma.c
@@ -62,6 +62,11 @@
 #define TDCR_BURSTSZ_16B	(0x3 << 6)
 #define TDCR_BURSTSZ_32B	(0x6 << 6)
 #define TDCR_BURSTSZ_64B	(0x7 << 6)
+#define TDCR_BURSTSZ_SQU_1B		(0x5 << 6)
+#define TDCR_BURSTSZ_SQU_2B		(0x6 << 6)
+#define TDCR_BURSTSZ_SQU_4B		(0x0 << 6)
+#define TDCR_BURSTSZ_SQU_8B		(0x1 << 6)
+#define TDCR_BURSTSZ_SQU_16B	(0x3 << 6)
 #define TDCR_BURSTSZ_SQU_32B	(0x7 << 6)
 #define TDCR_BURSTSZ_128B	(0x5 << 6)
 #define TDCR_DSTDIR_MSK		(0x3 << 4)	/* Dst Direction */
@@ -158,7 +163,7 @@ static void mmp_tdma_disable_chan(struct mmp_tdma_chan *tdmac)
 	/* disable irq */
 	writel(0, tdmac->reg_base + TDIMR);
 
-	tdmac->status = DMA_SUCCESS;
+	tdmac->status = DMA_COMPLETE;
 }
 
 static void mmp_tdma_resume_chan(struct mmp_tdma_chan *tdmac)
@@ -228,8 +233,31 @@ static int mmp_tdma_config_chan(struct mmp_tdma_chan *tdmac)
 			return -EINVAL;
 		}
 	} else if (tdmac->type == PXA910_SQU) {
-		tdcr |= TDCR_BURSTSZ_SQU_32B;
 		tdcr |= TDCR_SSPMOD;
+
+		switch (tdmac->burst_sz) {
+		case 1:
+			tdcr |= TDCR_BURSTSZ_SQU_1B;
+			break;
+		case 2:
+			tdcr |= TDCR_BURSTSZ_SQU_2B;
+			break;
+		case 4:
+			tdcr |= TDCR_BURSTSZ_SQU_4B;
+			break;
+		case 8:
+			tdcr |= TDCR_BURSTSZ_SQU_8B;
+			break;
+		case 16:
+			tdcr |= TDCR_BURSTSZ_SQU_16B;
+			break;
+		case 32:
+			tdcr |= TDCR_BURSTSZ_SQU_32B;
+			break;
+		default:
+			dev_err(tdmac->dev, "mmp_tdma: unknown burst size.\n");
+			return -EINVAL;
+		}
 	}
 
 	writel(tdcr, tdmac->reg_base + TDCR);
@@ -324,7 +352,7 @@ static int mmp_tdma_alloc_chan_resources(struct dma_chan *chan)
 
 	if (tdmac->irq) {
 		ret = devm_request_irq(tdmac->dev, tdmac->irq,
-			mmp_tdma_chan_handler, IRQF_DISABLED, "tdma", tdmac);
+			mmp_tdma_chan_handler, 0, "tdma", tdmac);
 		if (ret)
 			return ret;
 	}
@@ -365,7 +393,7 @@ static struct dma_async_tx_descriptor *mmp_tdma_prep_dma_cyclic(
 	int num_periods = buf_len / period_len;
 	int i = 0, buf = 0;
 
-	if (tdmac->status != DMA_SUCCESS)
+	if (tdmac->status != DMA_COMPLETE)
 		return NULL;
 
 	if (period_len > TDMA_MAX_XFER_BYTES) {
@@ -499,7 +527,7 @@ static int mmp_tdma_chan_init(struct mmp_tdma_device *tdev,
 	tdmac->idx	   = idx;
 	tdmac->type	   = type;
 	tdmac->reg_base	   = (unsigned long)tdev->base + idx * 4;
-	tdmac->status = DMA_SUCCESS;
+	tdmac->status = DMA_COMPLETE;
 	tdev->tdmac[tdmac->idx] = tdmac;
 	tasklet_init(&tdmac->tasklet, dma_do_tasklet, (unsigned long)tdmac);
 
@@ -554,7 +582,7 @@ static int mmp_tdma_probe(struct platform_device *pdev)
 	if (irq_num != chan_num) {
 		irq = platform_get_irq(pdev, 0);
 		ret = devm_request_irq(&pdev->dev, irq,
-			mmp_tdma_int_handler, IRQF_DISABLED, "tdma", tdev);
+			mmp_tdma_int_handler, 0, "tdma", tdev);
 		if (ret)
 			return ret;
 	}
diff --git a/drivers/dma/mv_xor.c b/drivers/dma/mv_xor.c
index 536dcb8ba5fd..7807f0ef4e20 100644
--- a/drivers/dma/mv_xor.c
+++ b/drivers/dma/mv_xor.c
@@ -60,14 +60,6 @@ static u32 mv_desc_get_dest_addr(struct mv_xor_desc_slot *desc)
 	return hw_desc->phy_dest_addr;
 }
 
-static u32 mv_desc_get_src_addr(struct mv_xor_desc_slot *desc,
-				int src_idx)
-{
-	struct mv_xor_desc *hw_desc = desc->hw_desc;
-	return hw_desc->phy_src_addr[mv_phy_src_idx(src_idx)];
-}
-
-
 static void mv_desc_set_byte_count(struct mv_xor_desc_slot *desc,
 				   u32 byte_count)
 {
@@ -278,42 +270,9 @@ mv_xor_run_tx_complete_actions(struct mv_xor_desc_slot *desc,
 			desc->async_tx.callback(
 				desc->async_tx.callback_param);
 
-		/* unmap dma addresses
-		 * (unmap_single vs unmap_page?)
-		 */
-		if (desc->group_head && desc->unmap_len) {
-			struct mv_xor_desc_slot *unmap = desc->group_head;
-			struct device *dev = mv_chan_to_devp(mv_chan);
-			u32 len = unmap->unmap_len;
-			enum dma_ctrl_flags flags = desc->async_tx.flags;
-			u32 src_cnt;
-			dma_addr_t addr;
-			dma_addr_t dest;
-
-			src_cnt = unmap->unmap_src_cnt;
-			dest = mv_desc_get_dest_addr(unmap);
-			if (!(flags & DMA_COMPL_SKIP_DEST_UNMAP)) {
-				enum dma_data_direction dir;
-
-				if (src_cnt > 1) /* is xor ? */
-					dir = DMA_BIDIRECTIONAL;
-				else
-					dir = DMA_FROM_DEVICE;
-				dma_unmap_page(dev, dest, len, dir);
-			}
-
-			if (!(flags & DMA_COMPL_SKIP_SRC_UNMAP)) {
-				while (src_cnt--) {
-					addr = mv_desc_get_src_addr(unmap,
-								    src_cnt);
-					if (addr == dest)
-						continue;
-					dma_unmap_page(dev, addr, len,
-						       DMA_TO_DEVICE);
-				}
-			}
+		dma_descriptor_unmap(&desc->async_tx);
+		if (desc->group_head)
 			desc->group_head = NULL;
-		}
 	}
 
 	/* run dependent operations */
@@ -749,7 +708,7 @@ static enum dma_status mv_xor_status(struct dma_chan *chan,
 	enum dma_status ret;
 
 	ret = dma_cookie_status(chan, cookie, txstate);
-	if (ret == DMA_SUCCESS) {
+	if (ret == DMA_COMPLETE) {
 		mv_xor_clean_completed_slots(mv_chan);
 		return ret;
 	}
@@ -874,7 +833,7 @@ static int mv_xor_memcpy_self_test(struct mv_xor_chan *mv_chan)
 	msleep(1);
 
 	if (mv_xor_status(dma_chan, cookie, NULL) !=
-	    DMA_SUCCESS) {
+	    DMA_COMPLETE) {
 		dev_err(dma_chan->device->dev,
 			"Self-test copy timed out, disabling\n");
 		err = -ENODEV;
@@ -968,7 +927,7 @@ mv_xor_xor_self_test(struct mv_xor_chan *mv_chan)
 	msleep(8);
 
 	if (mv_xor_status(dma_chan, cookie, NULL) !=
-	    DMA_SUCCESS) {
+	    DMA_COMPLETE) {
 		dev_err(dma_chan->device->dev,
 			"Self-test xor timed out, disabling\n");
 		err = -ENODEV;
@@ -1076,10 +1035,7 @@ mv_xor_channel_add(struct mv_xor_device *xordev,
 	}
 
 	mv_chan->mmr_base = xordev->xor_base;
-	if (!mv_chan->mmr_base) {
-		ret = -ENOMEM;
-		goto err_free_dma;
-	}
+	mv_chan->mmr_high_base = xordev->xor_high_base;
 	tasklet_init(&mv_chan->irq_tasklet, mv_xor_tasklet, (unsigned long)
 		     mv_chan);
 
@@ -1138,7 +1094,7 @@ static void
 mv_xor_conf_mbus_windows(struct mv_xor_device *xordev,
 			 const struct mbus_dram_target_info *dram)
 {
-	void __iomem *base = xordev->xor_base;
+	void __iomem *base = xordev->xor_high_base;
 	u32 win_enable = 0;
 	int i;
 
diff --git a/drivers/dma/mv_xor.h b/drivers/dma/mv_xor.h
index 06b067f24c9b..d0749229c875 100644
--- a/drivers/dma/mv_xor.h
+++ b/drivers/dma/mv_xor.h
@@ -34,13 +34,13 @@
 #define XOR_OPERATION_MODE_MEMCPY	2
 #define XOR_DESCRIPTOR_SWAP		BIT(14)
 
-#define XOR_CURR_DESC(chan)	(chan->mmr_base + 0x210 + (chan->idx * 4))
-#define XOR_NEXT_DESC(chan)	(chan->mmr_base + 0x200 + (chan->idx * 4))
-#define XOR_BYTE_COUNT(chan)	(chan->mmr_base + 0x220 + (chan->idx * 4))
-#define XOR_DEST_POINTER(chan)	(chan->mmr_base + 0x2B0 + (chan->idx * 4))
-#define XOR_BLOCK_SIZE(chan)	(chan->mmr_base + 0x2C0 + (chan->idx * 4))
-#define XOR_INIT_VALUE_LOW(chan)	(chan->mmr_base + 0x2E0)
-#define XOR_INIT_VALUE_HIGH(chan)	(chan->mmr_base + 0x2E4)
+#define XOR_CURR_DESC(chan)	(chan->mmr_high_base + 0x10 + (chan->idx * 4))
+#define XOR_NEXT_DESC(chan)	(chan->mmr_high_base + 0x00 + (chan->idx * 4))
+#define XOR_BYTE_COUNT(chan)	(chan->mmr_high_base + 0x20 + (chan->idx * 4))
+#define XOR_DEST_POINTER(chan)	(chan->mmr_high_base + 0xB0 + (chan->idx * 4))
+#define XOR_BLOCK_SIZE(chan)	(chan->mmr_high_base + 0xC0 + (chan->idx * 4))
+#define XOR_INIT_VALUE_LOW(chan)	(chan->mmr_high_base + 0xE0)
+#define XOR_INIT_VALUE_HIGH(chan)	(chan->mmr_high_base + 0xE4)
 
 #define XOR_CONFIG(chan)	(chan->mmr_base + 0x10 + (chan->idx * 4))
 #define XOR_ACTIVATION(chan)	(chan->mmr_base + 0x20 + (chan->idx * 4))
@@ -50,11 +50,11 @@
 #define XOR_ERROR_ADDR(chan)	(chan->mmr_base + 0x60)
 #define XOR_INTR_MASK_VALUE	0x3F5
 
-#define WINDOW_BASE(w)		(0x250 + ((w) << 2))
-#define WINDOW_SIZE(w)		(0x270 + ((w) << 2))
-#define WINDOW_REMAP_HIGH(w)	(0x290 + ((w) << 2))
-#define WINDOW_BAR_ENABLE(chan)	(0x240 + ((chan) << 2))
-#define WINDOW_OVERRIDE_CTRL(chan)	(0x2A0 + ((chan) << 2))
+#define WINDOW_BASE(w)		(0x50 + ((w) << 2))
+#define WINDOW_SIZE(w)		(0x70 + ((w) << 2))
+#define WINDOW_REMAP_HIGH(w)	(0x90 + ((w) << 2))
+#define WINDOW_BAR_ENABLE(chan)	(0x40 + ((chan) << 2))
+#define WINDOW_OVERRIDE_CTRL(chan)	(0xA0 + ((chan) << 2))
 
 struct mv_xor_device {
 	void __iomem	     *xor_base;
@@ -82,6 +82,7 @@ struct mv_xor_chan {
 	int			pending;
 	spinlock_t		lock; /* protects the descriptor slot pool */
 	void __iomem		*mmr_base;
+	void __iomem		*mmr_high_base;
 	unsigned int		idx;
 	int                     irq;
 	enum dma_transaction_type	current_type;
diff --git a/drivers/dma/mxs-dma.c b/drivers/dma/mxs-dma.c
index ccd13df841db..ead491346da7 100644
--- a/drivers/dma/mxs-dma.c
+++ b/drivers/dma/mxs-dma.c
@@ -27,6 +27,7 @@
 #include <linux/of.h>
 #include <linux/of_device.h>
 #include <linux/of_dma.h>
+#include <linux/list.h>
 
 #include <asm/irq.h>
 
@@ -57,6 +58,9 @@
 	(((dma_is_apbh(d) && apbh_is_old(d)) ? 0x050 : 0x110) + (n) * 0x70)
 #define HW_APBHX_CHn_SEMA(d, n) \
 	(((dma_is_apbh(d) && apbh_is_old(d)) ? 0x080 : 0x140) + (n) * 0x70)
+#define HW_APBHX_CHn_BAR(d, n) \
+	(((dma_is_apbh(d) && apbh_is_old(d)) ? 0x070 : 0x130) + (n) * 0x70)
+#define HW_APBX_CHn_DEBUG1(d, n) (0x150 + (n) * 0x70)
 
 /*
  * ccw bits definitions
@@ -115,7 +119,9 @@ struct mxs_dma_chan {
 	int				desc_count;
 	enum dma_status			status;
 	unsigned int			flags;
+	bool				reset;
 #define MXS_DMA_SG_LOOP			(1 << 0)
+#define MXS_DMA_USE_SEMAPHORE		(1 << 1)
 };
 
 #define MXS_DMA_CHANNELS		16
@@ -201,12 +207,47 @@ static void mxs_dma_reset_chan(struct mxs_dma_chan *mxs_chan)
 	struct mxs_dma_engine *mxs_dma = mxs_chan->mxs_dma;
 	int chan_id = mxs_chan->chan.chan_id;
 
-	if (dma_is_apbh(mxs_dma) && apbh_is_old(mxs_dma))
+	/*
+	 * mxs dma channel resets can cause a channel stall. To recover from a
+	 * channel stall, we have to reset the whole DMA engine. To avoid this,
+	 * we use cyclic DMA with semaphores, that are enhanced in
+	 * mxs_dma_int_handler. To reset the channel, we can simply stop writing
+	 * into the semaphore counter.
+	 */
+	if (mxs_chan->flags & MXS_DMA_USE_SEMAPHORE &&
+			mxs_chan->flags & MXS_DMA_SG_LOOP) {
+		mxs_chan->reset = true;
+	} else if (dma_is_apbh(mxs_dma) && apbh_is_old(mxs_dma)) {
 		writel(1 << (chan_id + BP_APBH_CTRL0_RESET_CHANNEL),
 			mxs_dma->base + HW_APBHX_CTRL0 + STMP_OFFSET_REG_SET);
-	else
+	} else {
+		unsigned long elapsed = 0;
+		const unsigned long max_wait = 50000; /* 50ms */
+		void __iomem *reg_dbg1 = mxs_dma->base +
+				HW_APBX_CHn_DEBUG1(mxs_dma, chan_id);
+
+		/*
+		 * On i.MX28 APBX, the DMA channel can stop working if we reset
+		 * the channel while it is in READ_FLUSH (0x08) state.
+		 * We wait here until we leave the state. Then we trigger the
+		 * reset. Waiting a maximum of 50ms, the kernel shouldn't crash
+		 * because of this.
+		 */
+		while ((readl(reg_dbg1) & 0xf) == 0x8 && elapsed < max_wait) {
+			udelay(100);
+			elapsed += 100;
+		}
+
+		if (elapsed >= max_wait)
+			dev_err(&mxs_chan->mxs_dma->pdev->dev,
+					"Failed waiting for the DMA channel %d to leave state READ_FLUSH, trying to reset channel in READ_FLUSH state now\n",
+					chan_id);
+
 		writel(1 << (chan_id + BP_APBHX_CHANNEL_CTRL_RESET_CHANNEL),
 			mxs_dma->base + HW_APBHX_CHANNEL_CTRL + STMP_OFFSET_REG_SET);
+	}
+
+	mxs_chan->status = DMA_COMPLETE;
 }
 
 static void mxs_dma_enable_chan(struct mxs_dma_chan *mxs_chan)
@@ -219,12 +260,21 @@ static void mxs_dma_enable_chan(struct mxs_dma_chan *mxs_chan)
 		mxs_dma->base + HW_APBHX_CHn_NXTCMDAR(mxs_dma, chan_id));
 
 	/* write 1 to SEMA to kick off the channel */
-	writel(1, mxs_dma->base + HW_APBHX_CHn_SEMA(mxs_dma, chan_id));
+	if (mxs_chan->flags & MXS_DMA_USE_SEMAPHORE &&
+			mxs_chan->flags & MXS_DMA_SG_LOOP) {
+		/* A cyclic DMA consists of at least 2 segments, so initialize
+		 * the semaphore with 2 so we have enough time to add 1 to the
+		 * semaphore if we need to */
+		writel(2, mxs_dma->base + HW_APBHX_CHn_SEMA(mxs_dma, chan_id));
+	} else {
+		writel(1, mxs_dma->base + HW_APBHX_CHn_SEMA(mxs_dma, chan_id));
+	}
+	mxs_chan->reset = false;
 }
 
 static void mxs_dma_disable_chan(struct mxs_dma_chan *mxs_chan)
 {
-	mxs_chan->status = DMA_SUCCESS;
+	mxs_chan->status = DMA_COMPLETE;
 }
 
 static void mxs_dma_pause_chan(struct mxs_dma_chan *mxs_chan)
@@ -272,58 +322,88 @@ static void mxs_dma_tasklet(unsigned long data)
 		mxs_chan->desc.callback(mxs_chan->desc.callback_param);
 }
 
+static int mxs_dma_irq_to_chan(struct mxs_dma_engine *mxs_dma, int irq)
+{
+	int i;
+
+	for (i = 0; i != mxs_dma->nr_channels; ++i)
+		if (mxs_dma->mxs_chans[i].chan_irq == irq)
+			return i;
+
+	return -EINVAL;
+}
+
 static irqreturn_t mxs_dma_int_handler(int irq, void *dev_id)
 {
 	struct mxs_dma_engine *mxs_dma = dev_id;
-	u32 stat1, stat2;
+	struct mxs_dma_chan *mxs_chan;
+	u32 completed;
+	u32 err;
+	int chan = mxs_dma_irq_to_chan(mxs_dma, irq);
+
+	if (chan < 0)
+		return IRQ_NONE;
 
 	/* completion status */
-	stat1 = readl(mxs_dma->base + HW_APBHX_CTRL1);
-	stat1 &= MXS_DMA_CHANNELS_MASK;
-	writel(stat1, mxs_dma->base + HW_APBHX_CTRL1 + STMP_OFFSET_REG_CLR);
+	completed = readl(mxs_dma->base + HW_APBHX_CTRL1);
+	completed = (completed >> chan) & 0x1;
+
+	/* Clear interrupt */
+	writel((1 << chan),
+			mxs_dma->base + HW_APBHX_CTRL1 + STMP_OFFSET_REG_CLR);
 
 	/* error status */
-	stat2 = readl(mxs_dma->base + HW_APBHX_CTRL2);
-	writel(stat2, mxs_dma->base + HW_APBHX_CTRL2 + STMP_OFFSET_REG_CLR);
+	err = readl(mxs_dma->base + HW_APBHX_CTRL2);
+	err &= (1 << (MXS_DMA_CHANNELS + chan)) | (1 << chan);
+
+	/*
+	 * error status bit is in the upper 16 bits, error irq bit in the lower
+	 * 16 bits. We transform it into a simpler error code:
+	 * err: 0x00 = no error, 0x01 = TERMINATION, 0x02 = BUS_ERROR
+	 */
+	err = (err >> (MXS_DMA_CHANNELS + chan)) + (err >> chan);
+
+	/* Clear error irq */
+	writel((1 << chan),
+			mxs_dma->base + HW_APBHX_CTRL2 + STMP_OFFSET_REG_CLR);
 
 	/*
 	 * When both completion and error of termination bits set at the
 	 * same time, we do not take it as an error.  IOW, it only becomes
-	 * an error we need to handle here in case of either it's (1) a bus
-	 * error or (2) a termination error with no completion.
+	 * an error we need to handle here in case of either it's a bus
+	 * error or a termination error with no completion. 0x01 is termination
+	 * error, so we can subtract err & completed to get the real error case.
 	 */
-	stat2 = ((stat2 >> MXS_DMA_CHANNELS) & stat2) | /* (1) */
-		(~(stat2 >> MXS_DMA_CHANNELS) & stat2 & ~stat1); /* (2) */
-
-	/* combine error and completion status for checking */
-	stat1 = (stat2 << MXS_DMA_CHANNELS) | stat1;
-	while (stat1) {
-		int channel = fls(stat1) - 1;
-		struct mxs_dma_chan *mxs_chan =
-			&mxs_dma->mxs_chans[channel % MXS_DMA_CHANNELS];
-
-		if (channel >= MXS_DMA_CHANNELS) {
-			dev_dbg(mxs_dma->dma_device.dev,
-				"%s: error in channel %d\n", __func__,
-				channel - MXS_DMA_CHANNELS);
-			mxs_chan->status = DMA_ERROR;
-			mxs_dma_reset_chan(mxs_chan);
-		} else {
-			if (mxs_chan->flags & MXS_DMA_SG_LOOP)
-				mxs_chan->status = DMA_IN_PROGRESS;
-			else
-				mxs_chan->status = DMA_SUCCESS;
-		}
+	err -= err & completed;
 
-		stat1 &= ~(1 << channel);
+	mxs_chan = &mxs_dma->mxs_chans[chan];
 
-		if (mxs_chan->status == DMA_SUCCESS)
-			dma_cookie_complete(&mxs_chan->desc);
+	if (err) {
+		dev_dbg(mxs_dma->dma_device.dev,
+			"%s: error in channel %d\n", __func__,
+			chan);
+		mxs_chan->status = DMA_ERROR;
+		mxs_dma_reset_chan(mxs_chan);
+	} else if (mxs_chan->status != DMA_COMPLETE) {
+		if (mxs_chan->flags & MXS_DMA_SG_LOOP) {
+			mxs_chan->status = DMA_IN_PROGRESS;
+			if (mxs_chan->flags & MXS_DMA_USE_SEMAPHORE)
+				writel(1, mxs_dma->base +
+					HW_APBHX_CHn_SEMA(mxs_dma, chan));
+		} else {
+			mxs_chan->status = DMA_COMPLETE;
+		}
+	}
 
-		/* schedule tasklet on this channel */
-		tasklet_schedule(&mxs_chan->tasklet);
+	if (mxs_chan->status == DMA_COMPLETE) {
+		if (mxs_chan->reset)
+			return IRQ_HANDLED;
+		dma_cookie_complete(&mxs_chan->desc);
 	}
 
+	/* schedule tasklet on this channel */
+	tasklet_schedule(&mxs_chan->tasklet);
+
 	return IRQ_HANDLED;
 }
 
@@ -523,6 +603,7 @@ static struct dma_async_tx_descriptor *mxs_dma_prep_dma_cyclic(
 
 	mxs_chan->status = DMA_IN_PROGRESS;
 	mxs_chan->flags |= MXS_DMA_SG_LOOP;
+	mxs_chan->flags |= MXS_DMA_USE_SEMAPHORE;
 
 	if (num_periods > NUM_CCW) {
 		dev_err(mxs_dma->dma_device.dev,
@@ -554,6 +635,7 @@ static struct dma_async_tx_descriptor *mxs_dma_prep_dma_cyclic(
 		ccw->bits |= CCW_IRQ;
 		ccw->bits |= CCW_HALT_ON_TERM;
 		ccw->bits |= CCW_TERM_FLUSH;
+		ccw->bits |= CCW_DEC_SEM;
 		ccw->bits |= BF_CCW(direction == DMA_DEV_TO_MEM ?
 				MXS_DMA_CMD_WRITE : MXS_DMA_CMD_READ, COMMAND);
 
@@ -599,8 +681,24 @@ static enum dma_status mxs_dma_tx_status(struct dma_chan *chan,
 			dma_cookie_t cookie, struct dma_tx_state *txstate)
 {
 	struct mxs_dma_chan *mxs_chan = to_mxs_dma_chan(chan);
+	struct mxs_dma_engine *mxs_dma = mxs_chan->mxs_dma;
+	u32 residue = 0;
+
+	if (mxs_chan->status == DMA_IN_PROGRESS &&
+			mxs_chan->flags & MXS_DMA_SG_LOOP) {
+		struct mxs_dma_ccw *last_ccw;
+		u32 bar;
+
+		last_ccw = &mxs_chan->ccw[mxs_chan->desc_count - 1];
+		residue = last_ccw->xfer_bytes + last_ccw->bufaddr;
+
+		bar = readl(mxs_dma->base +
+				HW_APBHX_CHn_BAR(mxs_dma, chan->chan_id));
+		residue -= bar;
+	}
 
-	dma_set_tx_state(txstate, chan->completed_cookie, chan->cookie, 0);
+	dma_set_tx_state(txstate, chan->completed_cookie, chan->cookie,
+			residue);
 
 	return mxs_chan->status;
 }
diff --git a/drivers/dma/omap-dma.c b/drivers/dma/omap-dma.c
index ec3fc4fd9160..2f66cf4e54fe 100644
--- a/drivers/dma/omap-dma.c
+++ b/drivers/dma/omap-dma.c
@@ -248,7 +248,7 @@ static enum dma_status omap_dma_tx_status(struct dma_chan *chan,
 	unsigned long flags;
 
 	ret = dma_cookie_status(chan, cookie, txstate);
-	if (ret == DMA_SUCCESS || !txstate)
+	if (ret == DMA_COMPLETE || !txstate)
 		return ret;
 
 	spin_lock_irqsave(&c->vc.lock, flags);
diff --git a/drivers/dma/pl330.c b/drivers/dma/pl330.c
index df8b10fd1726..cdf0483b8f2d 100644
--- a/drivers/dma/pl330.c
+++ b/drivers/dma/pl330.c
@@ -2268,6 +2268,8 @@ static void pl330_tasklet(unsigned long data)
 			list_move_tail(&desc->node, &pch->dmac->desc_pool);
 		}
 
+		dma_descriptor_unmap(&desc->txd);
+
 		if (callback) {
 			spin_unlock_irqrestore(&pch->lock, flags);
 			callback(callback_param);
@@ -2314,7 +2316,7 @@ bool pl330_filter(struct dma_chan *chan, void *param)
 		return false;
 
 	peri_id = chan->private;
-	return *peri_id == (unsigned)param;
+	return *peri_id == (unsigned long)param;
 }
 EXPORT_SYMBOL(pl330_filter);
 
@@ -2926,16 +2928,23 @@ pl330_probe(struct amba_device *adev, const struct amba_id *id)
 
 	amba_set_drvdata(adev, pdmac);
 
-	irq = adev->irq[0];
-	ret = request_irq(irq, pl330_irq_handler, 0,
-			dev_name(&adev->dev), pi);
-	if (ret)
-		return ret;
+	for (i = 0; i < AMBA_NR_IRQS; i++) {
+		irq = adev->irq[i];
+		if (irq) {
+			ret = devm_request_irq(&adev->dev, irq,
+					       pl330_irq_handler, 0,
+					       dev_name(&adev->dev), pi);
+			if (ret)
+				return ret;
+		} else {
+			break;
+		}
+	}
 
 	pi->pcfg.periph_id = adev->periphid;
 	ret = pl330_add(pi);
 	if (ret)
-		goto probe_err1;
+		return ret;
 
 	INIT_LIST_HEAD(&pdmac->desc_pool);
 	spin_lock_init(&pdmac->pool_lock);
@@ -3033,8 +3042,6 @@ pl330_probe(struct amba_device *adev, const struct amba_id *id)
 
 	return 0;
 probe_err3:
-	amba_set_drvdata(adev, NULL);
-
 	/* Idle the DMAC */
 	list_for_each_entry_safe(pch, _p, &pdmac->ddma.channels,
 			chan.device_node) {
@@ -3048,8 +3055,6 @@ probe_err3:
 	}
 probe_err2:
 	pl330_del(pi);
-probe_err1:
-	free_irq(irq, pi);
 
 	return ret;
 }
@@ -3059,7 +3064,6 @@ static int pl330_remove(struct amba_device *adev)
 	struct dma_pl330_dmac *pdmac = amba_get_drvdata(adev);
 	struct dma_pl330_chan *pch, *_p;
 	struct pl330_info *pi;
-	int irq;
 
 	if (!pdmac)
 		return 0;
@@ -3068,7 +3072,6 @@ static int pl330_remove(struct amba_device *adev)
 		of_dma_controller_free(adev->dev.of_node);
 
 	dma_async_device_unregister(&pdmac->ddma);
-	amba_set_drvdata(adev, NULL);
 
 	/* Idle the DMAC */
 	list_for_each_entry_safe(pch, _p, &pdmac->ddma.channels,
@@ -3086,9 +3089,6 @@ static int pl330_remove(struct amba_device *adev)
 
 	pl330_del(pi);
 
-	irq = adev->irq[0];
-	free_irq(irq, pi);
-
 	return 0;
 }
 
diff --git a/drivers/dma/ppc4xx/adma.c b/drivers/dma/ppc4xx/adma.c
index e24b5ef486b5..8da48c6b2a38 100644
--- a/drivers/dma/ppc4xx/adma.c
+++ b/drivers/dma/ppc4xx/adma.c
@@ -804,218 +804,6 @@ static void ppc440spe_desc_set_link(struct ppc440spe_adma_chan *chan,
 }
 
 /**
- * ppc440spe_desc_get_src_addr - extract the source address from the descriptor
- */
-static u32 ppc440spe_desc_get_src_addr(struct ppc440spe_adma_desc_slot *desc,
-				struct ppc440spe_adma_chan *chan, int src_idx)
-{
-	struct dma_cdb *dma_hw_desc;
-	struct xor_cb *xor_hw_desc;
-
-	switch (chan->device->id) {
-	case PPC440SPE_DMA0_ID:
-	case PPC440SPE_DMA1_ID:
-		dma_hw_desc = desc->hw_desc;
-		/* May have 0, 1, 2, or 3 sources */
-		switch (dma_hw_desc->opc) {
-		case DMA_CDB_OPC_NO_OP:
-		case DMA_CDB_OPC_DFILL128:
-			return 0;
-		case DMA_CDB_OPC_DCHECK128:
-			if (unlikely(src_idx)) {
-				printk(KERN_ERR "%s: try to get %d source for"
-				    " DCHECK128\n", __func__, src_idx);
-				BUG();
-			}
-			return le32_to_cpu(dma_hw_desc->sg1l);
-		case DMA_CDB_OPC_MULTICAST:
-		case DMA_CDB_OPC_MV_SG1_SG2:
-			if (unlikely(src_idx > 2)) {
-				printk(KERN_ERR "%s: try to get %d source from"
-				    " DMA descr\n", __func__, src_idx);
-				BUG();
-			}
-			if (src_idx) {
-				if (le32_to_cpu(dma_hw_desc->sg1u) &
-				    DMA_CUED_XOR_WIN_MSK) {
-					u8 region;
-
-					if (src_idx == 1)
-						return le32_to_cpu(
-						    dma_hw_desc->sg1l) +
-							desc->unmap_len;
-
-					region = (le32_to_cpu(
-					    dma_hw_desc->sg1u)) >>
-						DMA_CUED_REGION_OFF;
-
-					region &= DMA_CUED_REGION_MSK;
-					switch (region) {
-					case DMA_RXOR123:
-						return le32_to_cpu(
-						    dma_hw_desc->sg1l) +
-							(desc->unmap_len << 1);
-					case DMA_RXOR124:
-						return le32_to_cpu(
-						    dma_hw_desc->sg1l) +
-							(desc->unmap_len * 3);
-					case DMA_RXOR125:
-						return le32_to_cpu(
-						    dma_hw_desc->sg1l) +
-							(desc->unmap_len << 2);
-					default:
-						printk(KERN_ERR
-						    "%s: try to"
-						    " get src3 for region %02x"
-						    "PPC440SPE_DESC_RXOR12?\n",
-						    __func__, region);
-						BUG();
-					}
-				} else {
-					printk(KERN_ERR
-						"%s: try to get %d"
-						" source for non-cued descr\n",
-						__func__, src_idx);
-					BUG();
-				}
-			}
-			return le32_to_cpu(dma_hw_desc->sg1l);
-		default:
-			printk(KERN_ERR "%s: unknown OPC 0x%02x\n",
-				__func__, dma_hw_desc->opc);
-			BUG();
-		}
-		return le32_to_cpu(dma_hw_desc->sg1l);
-	case PPC440SPE_XOR_ID:
-		/* May have up to 16 sources */
-		xor_hw_desc = desc->hw_desc;
-		return xor_hw_desc->ops[src_idx].l;
-	}
-	return 0;
-}
-
-/**
- * ppc440spe_desc_get_dest_addr - extract the destination address from the
- * descriptor
- */
-static u32 ppc440spe_desc_get_dest_addr(struct ppc440spe_adma_desc_slot *desc,
-				struct ppc440spe_adma_chan *chan, int idx)
-{
-	struct dma_cdb *dma_hw_desc;
-	struct xor_cb *xor_hw_desc;
-
-	switch (chan->device->id) {
-	case PPC440SPE_DMA0_ID:
-	case PPC440SPE_DMA1_ID:
-		dma_hw_desc = desc->hw_desc;
-
-		if (likely(!idx))
-			return le32_to_cpu(dma_hw_desc->sg2l);
-		return le32_to_cpu(dma_hw_desc->sg3l);
-	case PPC440SPE_XOR_ID:
-		xor_hw_desc = desc->hw_desc;
-		return xor_hw_desc->cbtal;
-	}
-	return 0;
-}
-
-/**
- * ppc440spe_desc_get_src_num - extract the number of source addresses from
- * the descriptor
- */
-static u32 ppc440spe_desc_get_src_num(struct ppc440spe_adma_desc_slot *desc,
-				struct ppc440spe_adma_chan *chan)
-{
-	struct dma_cdb *dma_hw_desc;
-	struct xor_cb *xor_hw_desc;
-
-	switch (chan->device->id) {
-	case PPC440SPE_DMA0_ID:
-	case PPC440SPE_DMA1_ID:
-		dma_hw_desc = desc->hw_desc;
-
-		switch (dma_hw_desc->opc) {
-		case DMA_CDB_OPC_NO_OP:
-		case DMA_CDB_OPC_DFILL128:
-			return 0;
-		case DMA_CDB_OPC_DCHECK128:
-			return 1;
-		case DMA_CDB_OPC_MV_SG1_SG2:
-		case DMA_CDB_OPC_MULTICAST:
-			/*
-			 * Only for RXOR operations we have more than
-			 * one source
-			 */
-			if (le32_to_cpu(dma_hw_desc->sg1u) &
-			    DMA_CUED_XOR_WIN_MSK) {
-				/* RXOR op, there are 2 or 3 sources */
-				if (((le32_to_cpu(dma_hw_desc->sg1u) >>
-				    DMA_CUED_REGION_OFF) &
-				      DMA_CUED_REGION_MSK) == DMA_RXOR12) {
-					/* RXOR 1-2 */
-					return 2;
-				} else {
-					/* RXOR 1-2-3/1-2-4/1-2-5 */
-					return 3;
-				}
-			}
-			return 1;
-		default:
-			printk(KERN_ERR "%s: unknown OPC 0x%02x\n",
-				__func__, dma_hw_desc->opc);
-			BUG();
-		}
-	case PPC440SPE_XOR_ID:
-		/* up to 16 sources */
-		xor_hw_desc = desc->hw_desc;
-		return xor_hw_desc->cbc & XOR_CDCR_OAC_MSK;
-	default:
-		BUG();
-	}
-	return 0;
-}
-
-/**
- * ppc440spe_desc_get_dst_num - get the number of destination addresses in
- * this descriptor
- */
-static u32 ppc440spe_desc_get_dst_num(struct ppc440spe_adma_desc_slot *desc,
-				struct ppc440spe_adma_chan *chan)
-{
-	struct dma_cdb *dma_hw_desc;
-
-	switch (chan->device->id) {
-	case PPC440SPE_DMA0_ID:
-	case PPC440SPE_DMA1_ID:
-		/* May be 1 or 2 destinations */
-		dma_hw_desc = desc->hw_desc;
-		switch (dma_hw_desc->opc) {
-		case DMA_CDB_OPC_NO_OP:
-		case DMA_CDB_OPC_DCHECK128:
-			return 0;
-		case DMA_CDB_OPC_MV_SG1_SG2:
-		case DMA_CDB_OPC_DFILL128:
-			return 1;
-		case DMA_CDB_OPC_MULTICAST:
-			if (desc->dst_cnt == 2)
-				return 2;
-			else
-				return 1;
-		default:
-			printk(KERN_ERR "%s: unknown OPC 0x%02x\n",
-				__func__, dma_hw_desc->opc);
-			BUG();
-		}
-	case PPC440SPE_XOR_ID:
-		/* Always only 1 destination */
-		return 1;
-	default:
-		BUG();
-	}
-	return 0;
-}
-
-/**
  * ppc440spe_desc_get_link - get the address of the descriptor that
  * follows this one
  */
@@ -1707,43 +1495,6 @@ static void ppc440spe_adma_free_slots(struct ppc440spe_adma_desc_slot *slot,
 	}
 }
 
-static void ppc440spe_adma_unmap(struct ppc440spe_adma_chan *chan,
-				 struct ppc440spe_adma_desc_slot *desc)
-{
-	u32 src_cnt, dst_cnt;
-	dma_addr_t addr;
-
-	/*
-	 * get the number of sources & destination
-	 * included in this descriptor and unmap
-	 * them all
-	 */
-	src_cnt = ppc440spe_desc_get_src_num(desc, chan);
-	dst_cnt = ppc440spe_desc_get_dst_num(desc, chan);
-
-	/* unmap destinations */
-	if (!(desc->async_tx.flags & DMA_COMPL_SKIP_DEST_UNMAP)) {
-		while (dst_cnt--) {
-			addr = ppc440spe_desc_get_dest_addr(
-				desc, chan, dst_cnt);
-			dma_unmap_page(chan->device->dev,
-					addr, desc->unmap_len,
-					DMA_FROM_DEVICE);
-		}
-	}
-
-	/* unmap sources */
-	if (!(desc->async_tx.flags & DMA_COMPL_SKIP_SRC_UNMAP)) {
-		while (src_cnt--) {
-			addr = ppc440spe_desc_get_src_addr(
-				desc, chan, src_cnt);
-			dma_unmap_page(chan->device->dev,
-					addr, desc->unmap_len,
-					DMA_TO_DEVICE);
-		}
-	}
-}
-
 /**
  * ppc440spe_adma_run_tx_complete_actions - call functions to be called
  * upon completion
@@ -1767,26 +1518,7 @@ static dma_cookie_t ppc440spe_adma_run_tx_complete_actions(
 			desc->async_tx.callback(
 				desc->async_tx.callback_param);
 
-		/* unmap dma addresses
-		 * (unmap_single vs unmap_page?)
-		 *
-		 * actually, ppc's dma_unmap_page() functions are empty, so
-		 * the following code is just for the sake of completeness
-		 */
-		if (chan && chan->needs_unmap && desc->group_head &&
-		     desc->unmap_len) {
-			struct ppc440spe_adma_desc_slot *unmap =
-							desc->group_head;
-			/* assume 1 slot per op always */
-			u32 slot_count = unmap->slot_cnt;
-
-			/* Run through the group list and unmap addresses */
-			for (i = 0; i < slot_count; i++) {
-				BUG_ON(!unmap);
-				ppc440spe_adma_unmap(chan, unmap);
-				unmap = unmap->hw_next;
-			}
-		}
+		dma_descriptor_unmap(&desc->async_tx);
 	}
 
 	/* run dependent operations */
@@ -3893,7 +3625,7 @@ static enum dma_status ppc440spe_adma_tx_status(struct dma_chan *chan,
 
 	ppc440spe_chan = to_ppc440spe_adma_chan(chan);
 	ret = dma_cookie_status(chan, cookie, txstate);
-	if (ret == DMA_SUCCESS)
+	if (ret == DMA_COMPLETE)
 		return ret;
 
 	ppc440spe_adma_slot_cleanup(ppc440spe_chan);
diff --git a/drivers/dma/sa11x0-dma.c b/drivers/dma/sa11x0-dma.c
index 461a91ab70bb..ab26d46bbe15 100644
--- a/drivers/dma/sa11x0-dma.c
+++ b/drivers/dma/sa11x0-dma.c
@@ -436,7 +436,7 @@ static enum dma_status sa11x0_dma_tx_status(struct dma_chan *chan,
 	enum dma_status ret;
 
 	ret = dma_cookie_status(&c->vc.chan, cookie, state);
-	if (ret == DMA_SUCCESS)
+	if (ret == DMA_COMPLETE)
 		return ret;
 
 	if (!state)
diff --git a/drivers/dma/sh/shdma-base.c b/drivers/dma/sh/shdma-base.c
index d94ab592cc1b..2e7b394def80 100644
--- a/drivers/dma/sh/shdma-base.c
+++ b/drivers/dma/sh/shdma-base.c
@@ -724,7 +724,7 @@ static enum dma_status shdma_tx_status(struct dma_chan *chan,
 	 * If we don't find cookie on the queue, it has been aborted and we have
 	 * to report error
 	 */
-	if (status != DMA_SUCCESS) {
+	if (status != DMA_COMPLETE) {
 		struct shdma_desc *sdesc;
 		status = DMA_ERROR;
 		list_for_each_entry(sdesc, &schan->ld_queue, node)
diff --git a/drivers/dma/sh/shdmac.c b/drivers/dma/sh/shdmac.c
index 1069e8869f20..0d765c0e21ec 100644
--- a/drivers/dma/sh/shdmac.c
+++ b/drivers/dma/sh/shdmac.c
@@ -685,7 +685,7 @@ MODULE_DEVICE_TABLE(of, sh_dmae_of_match);
 static int sh_dmae_probe(struct platform_device *pdev)
 {
 	const struct sh_dmae_pdata *pdata;
-	unsigned long irqflags = IRQF_DISABLED,
+	unsigned long irqflags = 0,
 		chan_flag[SH_DMAE_MAX_CHANNELS] = {};
 	int errirq, chan_irq[SH_DMAE_MAX_CHANNELS];
 	int err, i, irq_cnt = 0, irqres = 0, irq_cap = 0;
@@ -838,7 +838,7 @@ static int sh_dmae_probe(struct platform_device *pdev)
 				    IORESOURCE_IRQ_SHAREABLE)
 					chan_flag[irq_cnt] = IRQF_SHARED;
 				else
-					chan_flag[irq_cnt] = IRQF_DISABLED;
+					chan_flag[irq_cnt] = 0;
 				dev_dbg(&pdev->dev,
 					"Found IRQ %d for channel %d\n",
 					i, irq_cnt);
diff --git a/drivers/dma/ste_dma40.c b/drivers/dma/ste_dma40.c
index 82d2b97ad942..b8c031b7de4e 100644
--- a/drivers/dma/ste_dma40.c
+++ b/drivers/dma/ste_dma40.c
@@ -14,6 +14,7 @@
 #include <linux/platform_device.h>
 #include <linux/clk.h>
 #include <linux/delay.h>
+#include <linux/log2.h>
 #include <linux/pm.h>
 #include <linux/pm_runtime.h>
 #include <linux/err.h>
@@ -2626,7 +2627,7 @@ static enum dma_status d40_tx_status(struct dma_chan *chan,
 	}
 
 	ret = dma_cookie_status(chan, cookie, txstate);
-	if (ret != DMA_SUCCESS)
+	if (ret != DMA_COMPLETE)
 		dma_set_residue(txstate, stedma40_residue(chan));
 
 	if (d40_is_paused(d40c))
@@ -2796,8 +2797,8 @@ static int d40_set_runtime_config(struct dma_chan *chan,
 	    src_addr_width >  DMA_SLAVE_BUSWIDTH_8_BYTES   ||
 	    dst_addr_width <= DMA_SLAVE_BUSWIDTH_UNDEFINED ||
 	    dst_addr_width >  DMA_SLAVE_BUSWIDTH_8_BYTES   ||
-	    ((src_addr_width > 1) && (src_addr_width & 1)) ||
-	    ((dst_addr_width > 1) && (dst_addr_width & 1)))
+	    !is_power_of_2(src_addr_width) ||
+	    !is_power_of_2(dst_addr_width))
 		return -EINVAL;
 
 	cfg->src_info.data_width = src_addr_width;
diff --git a/drivers/dma/tegra20-apb-dma.c b/drivers/dma/tegra20-apb-dma.c
index 5d4986e5f5fa..73654e33f13b 100644
--- a/drivers/dma/tegra20-apb-dma.c
+++ b/drivers/dma/tegra20-apb-dma.c
@@ -570,7 +570,7 @@ static void handle_once_dma_done(struct tegra_dma_channel *tdc,
 
 	list_del(&sgreq->node);
 	if (sgreq->last_sg) {
-		dma_desc->dma_status = DMA_SUCCESS;
+		dma_desc->dma_status = DMA_COMPLETE;
 		dma_cookie_complete(&dma_desc->txd);
 		if (!dma_desc->cb_count)
 			list_add_tail(&dma_desc->cb_node, &tdc->cb_desc);
@@ -768,7 +768,7 @@ static enum dma_status tegra_dma_tx_status(struct dma_chan *dc,
 	unsigned int residual;
 
 	ret = dma_cookie_status(dc, cookie, txstate);
-	if (ret == DMA_SUCCESS)
+	if (ret == DMA_COMPLETE)
 		return ret;
 
 	spin_lock_irqsave(&tdc->lock, flags);
@@ -1018,7 +1018,7 @@ static struct dma_async_tx_descriptor *tegra_dma_prep_slave_sg(
 	return &dma_desc->txd;
 }
 
-struct dma_async_tx_descriptor *tegra_dma_prep_dma_cyclic(
+static struct dma_async_tx_descriptor *tegra_dma_prep_dma_cyclic(
 	struct dma_chan *dc, dma_addr_t buf_addr, size_t buf_len,
 	size_t period_len, enum dma_transfer_direction direction,
 	unsigned long flags, void *context)
diff --git a/drivers/dma/timb_dma.c b/drivers/dma/timb_dma.c
index 28af214fce04..4506a7b4f972 100644
--- a/drivers/dma/timb_dma.c
+++ b/drivers/dma/timb_dma.c
@@ -154,38 +154,6 @@ static bool __td_dma_done_ack(struct timb_dma_chan *td_chan)
 	return done;
 }
 
-static void __td_unmap_desc(struct timb_dma_chan *td_chan, const u8 *dma_desc,
-	bool single)
-{
-	dma_addr_t addr;
-	int len;
-
-	addr = (dma_desc[7] << 24) | (dma_desc[6] << 16) | (dma_desc[5] << 8) |
-		dma_desc[4];
-
-	len = (dma_desc[3] << 8) | dma_desc[2];
-
-	if (single)
-		dma_unmap_single(chan2dev(&td_chan->chan), addr, len,
-			DMA_TO_DEVICE);
-	else
-		dma_unmap_page(chan2dev(&td_chan->chan), addr, len,
-			DMA_TO_DEVICE);
-}
-
-static void __td_unmap_descs(struct timb_dma_desc *td_desc, bool single)
-{
-	struct timb_dma_chan *td_chan = container_of(td_desc->txd.chan,
-		struct timb_dma_chan, chan);
-	u8 *descs;
-
-	for (descs = td_desc->desc_list; ; descs += TIMB_DMA_DESC_SIZE) {
-		__td_unmap_desc(td_chan, descs, single);
-		if (descs[0] & 0x02)
-			break;
-	}
-}
-
 static int td_fill_desc(struct timb_dma_chan *td_chan, u8 *dma_desc,
 	struct scatterlist *sg, bool last)
 {
@@ -293,10 +261,7 @@ static void __td_finish(struct timb_dma_chan *td_chan)
 
 	list_move(&td_desc->desc_node, &td_chan->free_list);
 
-	if (!(txd->flags & DMA_COMPL_SKIP_SRC_UNMAP))
-		__td_unmap_descs(td_desc,
-			txd->flags & DMA_COMPL_SRC_UNMAP_SINGLE);
-
+	dma_descriptor_unmap(txd);
 	/*
 	 * The API requires that no submissions are done from a
 	 * callback, so we don't need to drop the lock here
diff --git a/drivers/dma/txx9dmac.c b/drivers/dma/txx9dmac.c
index 71e8e775189e..bae6c29f5502 100644
--- a/drivers/dma/txx9dmac.c
+++ b/drivers/dma/txx9dmac.c
@@ -419,30 +419,7 @@ txx9dmac_descriptor_complete(struct txx9dmac_chan *dc,
 	list_splice_init(&desc->tx_list, &dc->free_list);
 	list_move(&desc->desc_node, &dc->free_list);
 
-	if (!ds) {
-		dma_addr_t dmaaddr;
-		if (!(txd->flags & DMA_COMPL_SKIP_DEST_UNMAP)) {
-			dmaaddr = is_dmac64(dc) ?
-				desc->hwdesc.DAR : desc->hwdesc32.DAR;
-			if (txd->flags & DMA_COMPL_DEST_UNMAP_SINGLE)
-				dma_unmap_single(chan2parent(&dc->chan),
-					dmaaddr, desc->len, DMA_FROM_DEVICE);
-			else
-				dma_unmap_page(chan2parent(&dc->chan),
-					dmaaddr, desc->len, DMA_FROM_DEVICE);
-		}
-		if (!(txd->flags & DMA_COMPL_SKIP_SRC_UNMAP)) {
-			dmaaddr = is_dmac64(dc) ?
-				desc->hwdesc.SAR : desc->hwdesc32.SAR;
-			if (txd->flags & DMA_COMPL_SRC_UNMAP_SINGLE)
-				dma_unmap_single(chan2parent(&dc->chan),
-					dmaaddr, desc->len, DMA_TO_DEVICE);
-			else
-				dma_unmap_page(chan2parent(&dc->chan),
-					dmaaddr, desc->len, DMA_TO_DEVICE);
-		}
-	}
-
+	dma_descriptor_unmap(txd);
 	/*
 	 * The API requires that no submissions are done from a
 	 * callback, so we don't need to drop the lock here
@@ -962,8 +939,8 @@ txx9dmac_tx_status(struct dma_chan *chan, dma_cookie_t cookie,
 	enum dma_status ret;
 
 	ret = dma_cookie_status(chan, cookie, txstate);
-	if (ret == DMA_SUCCESS)
-		return DMA_SUCCESS;
+	if (ret == DMA_COMPLETE)
+		return DMA_COMPLETE;
 
 	spin_lock_bh(&dc->lock);
 	txx9dmac_scan_descriptors(dc);
diff --git a/drivers/edac/cell_edac.c b/drivers/edac/cell_edac.c
index 9ee1c76da7b9..374b57fc596d 100644
--- a/drivers/edac/cell_edac.c
+++ b/drivers/edac/cell_edac.c
@@ -163,6 +163,7 @@ static void cell_edac_init_csrows(struct mem_ctl_info *mci)
 			csrow->first_page, nr_pages);
 		break;
 	}
+	of_node_put(np);
 }
 
 static int cell_edac_probe(struct platform_device *pdev)
diff --git a/drivers/edac/edac_device.c b/drivers/edac/edac_device.c
index 211021dfec73..102674346035 100644
--- a/drivers/edac/edac_device.c
+++ b/drivers/edac/edac_device.c
@@ -530,12 +530,9 @@ int edac_device_add_device(struct edac_device_ctl_info *edac_dev)
 
 	/* Report action taken */
 	edac_device_printk(edac_dev, KERN_INFO,
-				"Giving out device to module '%s' controller "
-				"'%s': DEV '%s' (%s)\n",
-				edac_dev->mod_name,
-				edac_dev->ctl_name,
-				edac_dev_name(edac_dev),
-				edac_op_state_to_string(edac_dev->op_state));
+		"Giving out device to module %s controller %s: DEV %s (%s)\n",
+		edac_dev->mod_name, edac_dev->ctl_name, edac_dev->dev_name,
+		edac_op_state_to_string(edac_dev->op_state));
 
 	mutex_unlock(&device_ctls_mutex);
 	return 0;
diff --git a/drivers/edac/edac_mc.c b/drivers/edac/edac_mc.c
index 89e109022d78..e8c9ef03495b 100644
--- a/drivers/edac/edac_mc.c
+++ b/drivers/edac/edac_mc.c
@@ -788,8 +788,10 @@ int edac_mc_add_mc(struct mem_ctl_info *mci)
 	}
 
 	/* Report action taken */
-	edac_mc_printk(mci, KERN_INFO, "Giving out device to '%s' '%s':"
-		" DEV %s\n", mci->mod_name, mci->ctl_name, edac_dev_name(mci));
+	edac_mc_printk(mci, KERN_INFO,
+		"Giving out device to module %s controller %s: DEV %s (%s)\n",
+		mci->mod_name, mci->ctl_name, mci->dev_name,
+		edac_op_state_to_string(mci->op_state));
 
 	edac_mc_owner = mci->mod_name;
 
diff --git a/drivers/edac/edac_pci.c b/drivers/edac/edac_pci.c
index dd370f92ace3..2cf44b4db80c 100644
--- a/drivers/edac/edac_pci.c
+++ b/drivers/edac/edac_pci.c
@@ -358,11 +358,9 @@ int edac_pci_add_device(struct edac_pci_ctl_info *pci, int edac_idx)
 	}
 
 	edac_pci_printk(pci, KERN_INFO,
-			"Giving out device to module '%s' controller '%s':"
-			" DEV '%s' (%s)\n",
-			pci->mod_name,
-			pci->ctl_name,
-			edac_dev_name(pci), edac_op_state_to_string(pci->op_state));
+		"Giving out device to module %s controller %s: DEV %s (%s)\n",
+		pci->mod_name, pci->ctl_name, pci->dev_name,
+		edac_op_state_to_string(pci->op_state));
 
 	mutex_unlock(&edac_pci_ctls_mutex);
 	return 0;
diff --git a/drivers/edac/highbank_l2_edac.c b/drivers/edac/highbank_l2_edac.c
index c2bd8c6a4349..2f193668ebc7 100644
--- a/drivers/edac/highbank_l2_edac.c
+++ b/drivers/edac/highbank_l2_edac.c
@@ -50,8 +50,15 @@ static irqreturn_t highbank_l2_err_handler(int irq, void *dev_id)
 	return IRQ_HANDLED;
 }
 
+static const struct of_device_id hb_l2_err_of_match[] = {
+	{ .compatible = "calxeda,hb-sregs-l2-ecc", },
+	{},
+};
+MODULE_DEVICE_TABLE(of, hb_l2_err_of_match);
+
 static int highbank_l2_err_probe(struct platform_device *pdev)
 {
+	const struct of_device_id *id;
 	struct edac_device_ctl_info *dci;
 	struct hb_l2_drvdata *drvdata;
 	struct resource *r;
@@ -90,28 +97,32 @@ static int highbank_l2_err_probe(struct platform_device *pdev)
 		goto err;
 	}
 
+	id = of_match_device(hb_l2_err_of_match, &pdev->dev);
+	dci->mod_name = pdev->dev.driver->name;
+	dci->ctl_name = id ? id->compatible : "unknown";
+	dci->dev_name = dev_name(&pdev->dev);
+
+	if (edac_device_add_device(dci))
+		goto err;
+
 	drvdata->db_irq = platform_get_irq(pdev, 0);
 	res = devm_request_irq(&pdev->dev, drvdata->db_irq,
 			       highbank_l2_err_handler,
 			       0, dev_name(&pdev->dev), dci);
 	if (res < 0)
-		goto err;
+		goto err2;
 
 	drvdata->sb_irq = platform_get_irq(pdev, 1);
 	res = devm_request_irq(&pdev->dev, drvdata->sb_irq,
 			       highbank_l2_err_handler,
 			       0, dev_name(&pdev->dev), dci);
 	if (res < 0)
-		goto err;
-
-	dci->mod_name = dev_name(&pdev->dev);
-	dci->dev_name = dev_name(&pdev->dev);
-
-	if (edac_device_add_device(dci))
-		goto err;
+		goto err2;
 
 	devres_close_group(&pdev->dev, NULL);
 	return 0;
+err2:
+	edac_device_del_device(&pdev->dev);
 err:
 	devres_release_group(&pdev->dev, NULL);
 	edac_device_free_ctl_info(dci);
@@ -127,12 +138,6 @@ static int highbank_l2_err_remove(struct platform_device *pdev)
 	return 0;
 }
 
-static const struct of_device_id hb_l2_err_of_match[] = {
-	{ .compatible = "calxeda,hb-sregs-l2-ecc", },
-	{},
-};
-MODULE_DEVICE_TABLE(of, hb_l2_err_of_match);
-
 static struct platform_driver highbank_l2_edac_driver = {
 	.probe = highbank_l2_err_probe,
 	.remove = highbank_l2_err_remove,
diff --git a/drivers/edac/highbank_mc_edac.c b/drivers/edac/highbank_mc_edac.c
index 4695dd2d71fd..f784de1dc793 100644
--- a/drivers/edac/highbank_mc_edac.c
+++ b/drivers/edac/highbank_mc_edac.c
@@ -26,31 +26,40 @@
 #include "edac_module.h"
 
 /* DDR Ctrlr Error Registers */
-#define HB_DDR_ECC_OPT			0x128
-#define HB_DDR_ECC_U_ERR_ADDR		0x130
-#define HB_DDR_ECC_U_ERR_STAT		0x134
-#define HB_DDR_ECC_U_ERR_DATAL		0x138
-#define HB_DDR_ECC_U_ERR_DATAH		0x13c
-#define HB_DDR_ECC_C_ERR_ADDR		0x140
-#define HB_DDR_ECC_C_ERR_STAT		0x144
-#define HB_DDR_ECC_C_ERR_DATAL		0x148
-#define HB_DDR_ECC_C_ERR_DATAH		0x14c
-#define HB_DDR_ECC_INT_STATUS		0x180
-#define HB_DDR_ECC_INT_ACK		0x184
-#define HB_DDR_ECC_U_ERR_ID		0x424
-#define HB_DDR_ECC_C_ERR_ID		0x428
 
-#define HB_DDR_ECC_INT_STAT_CE		0x8
-#define HB_DDR_ECC_INT_STAT_DOUBLE_CE	0x10
-#define HB_DDR_ECC_INT_STAT_UE		0x20
-#define HB_DDR_ECC_INT_STAT_DOUBLE_UE	0x40
+#define HB_DDR_ECC_ERR_BASE		0x128
+#define MW_DDR_ECC_ERR_BASE		0x1b4
+
+#define HB_DDR_ECC_OPT			0x00
+#define HB_DDR_ECC_U_ERR_ADDR		0x08
+#define HB_DDR_ECC_U_ERR_STAT		0x0c
+#define HB_DDR_ECC_U_ERR_DATAL		0x10
+#define HB_DDR_ECC_U_ERR_DATAH		0x14
+#define HB_DDR_ECC_C_ERR_ADDR		0x18
+#define HB_DDR_ECC_C_ERR_STAT		0x1c
+#define HB_DDR_ECC_C_ERR_DATAL		0x20
+#define HB_DDR_ECC_C_ERR_DATAH		0x24
 
 #define HB_DDR_ECC_OPT_MODE_MASK	0x3
 #define HB_DDR_ECC_OPT_FWC		0x100
 #define HB_DDR_ECC_OPT_XOR_SHIFT	16
 
+/* DDR Ctrlr Interrupt Registers */
+
+#define HB_DDR_ECC_INT_BASE		0x180
+#define MW_DDR_ECC_INT_BASE		0x218
+
+#define HB_DDR_ECC_INT_STATUS		0x00
+#define HB_DDR_ECC_INT_ACK		0x04
+
+#define HB_DDR_ECC_INT_STAT_CE		0x8
+#define HB_DDR_ECC_INT_STAT_DOUBLE_CE	0x10
+#define HB_DDR_ECC_INT_STAT_UE		0x20
+#define HB_DDR_ECC_INT_STAT_DOUBLE_UE	0x40
+
 struct hb_mc_drvdata {
-	void __iomem *mc_vbase;
+	void __iomem *mc_err_base;
+	void __iomem *mc_int_base;
 };
 
 static irqreturn_t highbank_mc_err_handler(int irq, void *dev_id)
@@ -60,10 +69,10 @@ static irqreturn_t highbank_mc_err_handler(int irq, void *dev_id)
 	u32 status, err_addr;
 
 	/* Read the interrupt status register */
-	status = readl(drvdata->mc_vbase + HB_DDR_ECC_INT_STATUS);
+	status = readl(drvdata->mc_int_base + HB_DDR_ECC_INT_STATUS);
 
 	if (status & HB_DDR_ECC_INT_STAT_UE) {
-		err_addr = readl(drvdata->mc_vbase + HB_DDR_ECC_U_ERR_ADDR);
+		err_addr = readl(drvdata->mc_err_base + HB_DDR_ECC_U_ERR_ADDR);
 		edac_mc_handle_error(HW_EVENT_ERR_UNCORRECTED, mci, 1,
 				     err_addr >> PAGE_SHIFT,
 				     err_addr & ~PAGE_MASK, 0,
@@ -71,9 +80,9 @@ static irqreturn_t highbank_mc_err_handler(int irq, void *dev_id)
 				     mci->ctl_name, "");
 	}
 	if (status & HB_DDR_ECC_INT_STAT_CE) {
-		u32 syndrome = readl(drvdata->mc_vbase + HB_DDR_ECC_C_ERR_STAT);
+		u32 syndrome = readl(drvdata->mc_err_base + HB_DDR_ECC_C_ERR_STAT);
 		syndrome = (syndrome >> 8) & 0xff;
-		err_addr = readl(drvdata->mc_vbase + HB_DDR_ECC_C_ERR_ADDR);
+		err_addr = readl(drvdata->mc_err_base + HB_DDR_ECC_C_ERR_ADDR);
 		edac_mc_handle_error(HW_EVENT_ERR_CORRECTED, mci, 1,
 				     err_addr >> PAGE_SHIFT,
 				     err_addr & ~PAGE_MASK, syndrome,
@@ -82,66 +91,79 @@ static irqreturn_t highbank_mc_err_handler(int irq, void *dev_id)
 	}
 
 	/* clear the error, clears the interrupt */
-	writel(status, drvdata->mc_vbase + HB_DDR_ECC_INT_ACK);
+	writel(status, drvdata->mc_int_base + HB_DDR_ECC_INT_ACK);
 	return IRQ_HANDLED;
 }
 
-#ifdef CONFIG_EDAC_DEBUG
-static ssize_t highbank_mc_err_inject_write(struct file *file,
-				      const char __user *data,
-				      size_t count, loff_t *ppos)
+static void highbank_mc_err_inject(struct mem_ctl_info *mci, u8 synd)
 {
-	struct mem_ctl_info *mci = file->private_data;
 	struct hb_mc_drvdata *pdata = mci->pvt_info;
-	char buf[32];
-	size_t buf_size;
 	u32 reg;
+
+	reg = readl(pdata->mc_err_base + HB_DDR_ECC_OPT);
+	reg &= HB_DDR_ECC_OPT_MODE_MASK;
+	reg |= (synd << HB_DDR_ECC_OPT_XOR_SHIFT) | HB_DDR_ECC_OPT_FWC;
+	writel(reg, pdata->mc_err_base + HB_DDR_ECC_OPT);
+}
+
+#define to_mci(k) container_of(k, struct mem_ctl_info, dev)
+
+static ssize_t highbank_mc_inject_ctrl(struct device *dev,
+	struct device_attribute *attr, const char *buf, size_t count)
+{
+	struct mem_ctl_info *mci = to_mci(dev);
 	u8 synd;
 
-	buf_size = min(count, (sizeof(buf)-1));
-	if (copy_from_user(buf, data, buf_size))
-		return -EFAULT;
-	buf[buf_size] = 0;
+	if (kstrtou8(buf, 16, &synd))
+		return -EINVAL;
 
-	if (!kstrtou8(buf, 16, &synd)) {
-		reg = readl(pdata->mc_vbase + HB_DDR_ECC_OPT);
-		reg &= HB_DDR_ECC_OPT_MODE_MASK;
-		reg |= (synd << HB_DDR_ECC_OPT_XOR_SHIFT) | HB_DDR_ECC_OPT_FWC;
-		writel(reg, pdata->mc_vbase + HB_DDR_ECC_OPT);
-	}
+	highbank_mc_err_inject(mci, synd);
 
 	return count;
 }
 
-static const struct file_operations highbank_mc_debug_inject_fops = {
-	.open = simple_open,
-	.write = highbank_mc_err_inject_write,
-	.llseek = generic_file_llseek,
+static DEVICE_ATTR(inject_ctrl, S_IWUSR, NULL, highbank_mc_inject_ctrl);
+
+struct hb_mc_settings {
+	int	err_offset;
+	int	int_offset;
 };
 
-static void highbank_mc_create_debugfs_nodes(struct mem_ctl_info *mci)
-{
-	if (mci->debugfs)
-		debugfs_create_file("inject_ctrl", S_IWUSR, mci->debugfs, mci,
-				    &highbank_mc_debug_inject_fops);
-;
-}
-#else
-static void highbank_mc_create_debugfs_nodes(struct mem_ctl_info *mci)
-{}
-#endif
+static struct hb_mc_settings hb_settings = {
+	.err_offset = HB_DDR_ECC_ERR_BASE,
+	.int_offset = HB_DDR_ECC_INT_BASE,
+};
+
+static struct hb_mc_settings mw_settings = {
+	.err_offset = MW_DDR_ECC_ERR_BASE,
+	.int_offset = MW_DDR_ECC_INT_BASE,
+};
+
+static struct of_device_id hb_ddr_ctrl_of_match[] = {
+	{ .compatible = "calxeda,hb-ddr-ctrl",		.data = &hb_settings },
+	{ .compatible = "calxeda,ecx-2000-ddr-ctrl",	.data = &mw_settings },
+	{},
+};
+MODULE_DEVICE_TABLE(of, hb_ddr_ctrl_of_match);
 
 static int highbank_mc_probe(struct platform_device *pdev)
 {
+	const struct of_device_id *id;
+	const struct hb_mc_settings *settings;
 	struct edac_mc_layer layers[2];
 	struct mem_ctl_info *mci;
 	struct hb_mc_drvdata *drvdata;
 	struct dimm_info *dimm;
 	struct resource *r;
+	void __iomem *base;
 	u32 control;
 	int irq;
 	int res = 0;
 
+	id = of_match_device(hb_ddr_ctrl_of_match, &pdev->dev);
+	if (!id)
+		return -ENODEV;
+
 	layers[0].type = EDAC_MC_LAYER_CHIP_SELECT;
 	layers[0].size = 1;
 	layers[0].is_virt_csrow = true;
@@ -174,35 +196,31 @@ static int highbank_mc_probe(struct platform_device *pdev)
 		goto err;
 	}
 
-	drvdata->mc_vbase = devm_ioremap(&pdev->dev,
-					  r->start, resource_size(r));
-	if (!drvdata->mc_vbase) {
+	base = devm_ioremap(&pdev->dev, r->start, resource_size(r));
+	if (!base) {
 		dev_err(&pdev->dev, "Unable to map regs\n");
 		res = -ENOMEM;
 		goto err;
 	}
 
-	control = readl(drvdata->mc_vbase + HB_DDR_ECC_OPT) & 0x3;
+	settings = id->data;
+	drvdata->mc_err_base = base + settings->err_offset;
+	drvdata->mc_int_base = base + settings->int_offset;
+
+	control = readl(drvdata->mc_err_base + HB_DDR_ECC_OPT) & 0x3;
 	if (!control || (control == 0x2)) {
 		dev_err(&pdev->dev, "No ECC present, or ECC disabled\n");
 		res = -ENODEV;
 		goto err;
 	}
 
-	irq = platform_get_irq(pdev, 0);
-	res = devm_request_irq(&pdev->dev, irq, highbank_mc_err_handler,
-			       0, dev_name(&pdev->dev), mci);
-	if (res < 0) {
-		dev_err(&pdev->dev, "Unable to request irq %d\n", irq);
-		goto err;
-	}
-
 	mci->mtype_cap = MEM_FLAG_DDR3;
 	mci->edac_ctl_cap = EDAC_FLAG_NONE | EDAC_FLAG_SECDED;
 	mci->edac_cap = EDAC_FLAG_SECDED;
-	mci->mod_name = dev_name(&pdev->dev);
+	mci->mod_name = pdev->dev.driver->name;
 	mci->mod_ver = "1";
-	mci->ctl_name = dev_name(&pdev->dev);
+	mci->ctl_name = id->compatible;
+	mci->dev_name = dev_name(&pdev->dev);
 	mci->scrub_mode = SCRUB_SW_SRC;
 
 	/* Only a single 4GB DIMM is supported */
@@ -217,10 +235,20 @@ static int highbank_mc_probe(struct platform_device *pdev)
 	if (res < 0)
 		goto err;
 
-	highbank_mc_create_debugfs_nodes(mci);
+	irq = platform_get_irq(pdev, 0);
+	res = devm_request_irq(&pdev->dev, irq, highbank_mc_err_handler,
+			       0, dev_name(&pdev->dev), mci);
+	if (res < 0) {
+		dev_err(&pdev->dev, "Unable to request irq %d\n", irq);
+		goto err2;
+	}
+
+	device_create_file(&mci->dev, &dev_attr_inject_ctrl);
 
 	devres_close_group(&pdev->dev, NULL);
 	return 0;
+err2:
+	edac_mc_del_mc(&pdev->dev);
 err:
 	devres_release_group(&pdev->dev, NULL);
 	edac_mc_free(mci);
@@ -231,17 +259,12 @@ static int highbank_mc_remove(struct platform_device *pdev)
 {
 	struct mem_ctl_info *mci = platform_get_drvdata(pdev);
 
+	device_remove_file(&mci->dev, &dev_attr_inject_ctrl);
 	edac_mc_del_mc(&pdev->dev);
 	edac_mc_free(mci);
 	return 0;
 }
 
-static const struct of_device_id hb_ddr_ctrl_of_match[] = {
-	{ .compatible = "calxeda,hb-ddr-ctrl", },
-	{},
-};
-MODULE_DEVICE_TABLE(of, hb_ddr_ctrl_of_match);
-
 static struct platform_driver highbank_mc_edac_driver = {
 	.probe = highbank_mc_probe,
 	.remove = highbank_mc_remove,
diff --git a/drivers/edac/mpc85xx_edac.c b/drivers/edac/mpc85xx_edac.c
index 3eb32f62d72a..fd46b0bd5f2a 100644
--- a/drivers/edac/mpc85xx_edac.c
+++ b/drivers/edac/mpc85xx_edac.c
@@ -327,28 +327,6 @@ err:
 }
 EXPORT_SYMBOL(mpc85xx_pci_err_probe);
 
-static int mpc85xx_pci_err_remove(struct platform_device *op)
-{
-	struct edac_pci_ctl_info *pci = dev_get_drvdata(&op->dev);
-	struct mpc85xx_pci_pdata *pdata = pci->pvt_info;
-
-	edac_dbg(0, "\n");
-
-	out_be32(pdata->pci_vbase + MPC85XX_PCI_ERR_CAP_DR,
-		 orig_pci_err_cap_dr);
-
-	out_be32(pdata->pci_vbase + MPC85XX_PCI_ERR_EN, orig_pci_err_en);
-
-	edac_pci_del_device(pci->dev);
-
-	if (edac_op_state == EDAC_OPSTATE_INT)
-		irq_dispose_mapping(pdata->irq);
-
-	edac_pci_free_ctl_info(pci);
-
-	return 0;
-}
-
 #endif				/* CONFIG_PCI */
 
 /**************************** L2 Err device ***************************/
diff --git a/drivers/edac/sb_edac.c b/drivers/edac/sb_edac.c
index 88f60c5fecbc..8472405c5586 100644
--- a/drivers/edac/sb_edac.c
+++ b/drivers/edac/sb_edac.c
@@ -34,7 +34,7 @@ static int probed;
 /*
  * Alter this version for the module when modifications are made
  */
-#define SBRIDGE_REVISION    " Ver: 1.0.0 "
+#define SBRIDGE_REVISION    " Ver: 1.1.0 "
 #define EDAC_MOD_STR      "sbridge_edac"
 
 /*
@@ -83,11 +83,17 @@ static int probed;
 #define PCI_DEVICE_ID_INTEL_SBRIDGE_IMC_ERR3	0x3c77	/* 16.7 */
 
 /* Devices 12 Function 6, Offsets 0x80 to 0xcc */
-static const u32 dram_rule[] = {
+static const u32 sbridge_dram_rule[] = {
 	0x80, 0x88, 0x90, 0x98, 0xa0,
 	0xa8, 0xb0, 0xb8, 0xc0, 0xc8,
 };
-#define MAX_SAD		ARRAY_SIZE(dram_rule)
+
+static const u32 ibridge_dram_rule[] = {
+	0x60, 0x68, 0x70, 0x78, 0x80,
+	0x88, 0x90, 0x98, 0xa0,	0xa8,
+	0xb0, 0xb8, 0xc0, 0xc8, 0xd0,
+	0xd8, 0xe0, 0xe8, 0xf0, 0xf8,
+};
 
 #define SAD_LIMIT(reg)		((GET_BITFIELD(reg, 6, 25) << 26) | 0x3ffffff)
 #define DRAM_ATTR(reg)		GET_BITFIELD(reg, 2,  3)
@@ -108,43 +114,50 @@ static char *get_dram_attr(u32 reg)
 	}
 }
 
-static const u32 interleave_list[] = {
+static const u32 sbridge_interleave_list[] = {
 	0x84, 0x8c, 0x94, 0x9c, 0xa4,
 	0xac, 0xb4, 0xbc, 0xc4, 0xcc,
 };
-#define MAX_INTERLEAVE	ARRAY_SIZE(interleave_list)
-
-#define SAD_PKG0(reg)		GET_BITFIELD(reg, 0, 2)
-#define SAD_PKG1(reg)		GET_BITFIELD(reg, 3, 5)
-#define SAD_PKG2(reg)		GET_BITFIELD(reg, 8, 10)
-#define SAD_PKG3(reg)		GET_BITFIELD(reg, 11, 13)
-#define SAD_PKG4(reg)		GET_BITFIELD(reg, 16, 18)
-#define SAD_PKG5(reg)		GET_BITFIELD(reg, 19, 21)
-#define SAD_PKG6(reg)		GET_BITFIELD(reg, 24, 26)
-#define SAD_PKG7(reg)		GET_BITFIELD(reg, 27, 29)
-
-static inline int sad_pkg(u32 reg, int interleave)
+
+static const u32 ibridge_interleave_list[] = {
+	0x64, 0x6c, 0x74, 0x7c, 0x84,
+	0x8c, 0x94, 0x9c, 0xa4, 0xac,
+	0xb4, 0xbc, 0xc4, 0xcc, 0xd4,
+	0xdc, 0xe4, 0xec, 0xf4, 0xfc,
+};
+
+struct interleave_pkg {
+	unsigned char start;
+	unsigned char end;
+};
+
+static const struct interleave_pkg sbridge_interleave_pkg[] = {
+	{ 0, 2 },
+	{ 3, 5 },
+	{ 8, 10 },
+	{ 11, 13 },
+	{ 16, 18 },
+	{ 19, 21 },
+	{ 24, 26 },
+	{ 27, 29 },
+};
+
+static const struct interleave_pkg ibridge_interleave_pkg[] = {
+	{ 0, 3 },
+	{ 4, 7 },
+	{ 8, 11 },
+	{ 12, 15 },
+	{ 16, 19 },
+	{ 20, 23 },
+	{ 24, 27 },
+	{ 28, 31 },
+};
+
+static inline int sad_pkg(const struct interleave_pkg *table, u32 reg,
+			  int interleave)
 {
-	switch (interleave) {
-	case 0:
-		return SAD_PKG0(reg);
-	case 1:
-		return SAD_PKG1(reg);
-	case 2:
-		return SAD_PKG2(reg);
-	case 3:
-		return SAD_PKG3(reg);
-	case 4:
-		return SAD_PKG4(reg);
-	case 5:
-		return SAD_PKG5(reg);
-	case 6:
-		return SAD_PKG6(reg);
-	case 7:
-		return SAD_PKG7(reg);
-	default:
-		return -EINVAL;
-	}
+	return GET_BITFIELD(reg, table[interleave].start,
+			    table[interleave].end);
 }
 
 /* Devices 12 Function 7 */
@@ -262,7 +275,9 @@ static const u32 correrrthrsld[] = {
 
 /* Device 17, function 0 */
 
-#define RANK_CFG_A		0x0328
+#define SB_RANK_CFG_A		0x0328
+
+#define IB_RANK_CFG_A		0x0320
 
 #define IS_RDIMM_ENABLED(reg)		GET_BITFIELD(reg, 11, 11)
 
@@ -273,8 +288,23 @@ static const u32 correrrthrsld[] = {
 #define NUM_CHANNELS	4
 #define MAX_DIMMS	3		/* Max DIMMS per channel */
 
+enum type {
+	SANDY_BRIDGE,
+	IVY_BRIDGE,
+};
+
+struct sbridge_pvt;
 struct sbridge_info {
-	u32	mcmtr;
+	enum type	type;
+	u32		mcmtr;
+	u32		rankcfgr;
+	u64		(*get_tolm)(struct sbridge_pvt *pvt);
+	u64		(*get_tohm)(struct sbridge_pvt *pvt);
+	const u32	*dram_rule;
+	const u32	*interleave_list;
+	const struct interleave_pkg *interleave_pkg;
+	u8		max_sad;
+	u8		max_interleave;
 };
 
 struct sbridge_channel {
@@ -305,8 +335,9 @@ struct sbridge_dev {
 
 struct sbridge_pvt {
 	struct pci_dev		*pci_ta, *pci_ddrio, *pci_ras;
-	struct pci_dev		*pci_sad0, *pci_sad1, *pci_ha0;
-	struct pci_dev		*pci_br;
+	struct pci_dev		*pci_sad0, *pci_sad1;
+	struct pci_dev		*pci_ha0, *pci_ha1;
+	struct pci_dev		*pci_br0, *pci_br1;
 	struct pci_dev		*pci_tad[NUM_CHANNELS];
 
 	struct sbridge_dev	*sbridge_dev;
@@ -364,11 +395,75 @@ static const struct pci_id_table pci_dev_descr_sbridge_table[] = {
 	{0,}			/* 0 terminated list. */
 };
 
+/* This changes depending if 1HA or 2HA:
+ * 1HA:
+ *	0x0eb8 (17.0) is DDRIO0
+ * 2HA:
+ *	0x0ebc (17.4) is DDRIO0
+ */
+#define PCI_DEVICE_ID_INTEL_IBRIDGE_IMC_1HA_DDRIO0	0x0eb8
+#define PCI_DEVICE_ID_INTEL_IBRIDGE_IMC_2HA_DDRIO0	0x0ebc
+
+/* pci ids */
+#define PCI_DEVICE_ID_INTEL_IBRIDGE_IMC_HA0		0x0ea0
+#define PCI_DEVICE_ID_INTEL_IBRIDGE_IMC_HA0_TA		0x0ea8
+#define PCI_DEVICE_ID_INTEL_IBRIDGE_IMC_HA0_RAS		0x0e71
+#define PCI_DEVICE_ID_INTEL_IBRIDGE_IMC_HA0_TAD0	0x0eaa
+#define PCI_DEVICE_ID_INTEL_IBRIDGE_IMC_HA0_TAD1	0x0eab
+#define PCI_DEVICE_ID_INTEL_IBRIDGE_IMC_HA0_TAD2	0x0eac
+#define PCI_DEVICE_ID_INTEL_IBRIDGE_IMC_HA0_TAD3	0x0ead
+#define PCI_DEVICE_ID_INTEL_IBRIDGE_SAD			0x0ec8
+#define PCI_DEVICE_ID_INTEL_IBRIDGE_BR0			0x0ec9
+#define PCI_DEVICE_ID_INTEL_IBRIDGE_BR1			0x0eca
+#define PCI_DEVICE_ID_INTEL_IBRIDGE_IMC_HA1		0x0e60
+#define PCI_DEVICE_ID_INTEL_IBRIDGE_IMC_HA1_TA		0x0e68
+#define PCI_DEVICE_ID_INTEL_IBRIDGE_IMC_HA1_RAS		0x0e79
+#define PCI_DEVICE_ID_INTEL_IBRIDGE_IMC_HA1_TAD0	0x0e6a
+#define PCI_DEVICE_ID_INTEL_IBRIDGE_IMC_HA1_TAD1	0x0e6b
+
+static const struct pci_id_descr pci_dev_descr_ibridge[] = {
+		/* Processor Home Agent */
+	{ PCI_DESCR(14, 0, PCI_DEVICE_ID_INTEL_IBRIDGE_IMC_HA0, 0)	},
+
+		/* Memory controller */
+	{ PCI_DESCR(15, 0, PCI_DEVICE_ID_INTEL_IBRIDGE_IMC_HA0_TA, 0)	},
+	{ PCI_DESCR(15, 1, PCI_DEVICE_ID_INTEL_IBRIDGE_IMC_HA0_RAS, 0)	},
+	{ PCI_DESCR(15, 2, PCI_DEVICE_ID_INTEL_IBRIDGE_IMC_HA0_TAD0, 0)	},
+	{ PCI_DESCR(15, 3, PCI_DEVICE_ID_INTEL_IBRIDGE_IMC_HA0_TAD1, 0)	},
+	{ PCI_DESCR(15, 4, PCI_DEVICE_ID_INTEL_IBRIDGE_IMC_HA0_TAD2, 0)	},
+	{ PCI_DESCR(15, 5, PCI_DEVICE_ID_INTEL_IBRIDGE_IMC_HA0_TAD3, 0)	},
+
+		/* System Address Decoder */
+	{ PCI_DESCR(22, 0, PCI_DEVICE_ID_INTEL_IBRIDGE_SAD, 0)		},
+
+		/* Broadcast Registers */
+	{ PCI_DESCR(22, 1, PCI_DEVICE_ID_INTEL_IBRIDGE_BR0, 1)		},
+	{ PCI_DESCR(22, 2, PCI_DEVICE_ID_INTEL_IBRIDGE_BR1, 0)		},
+
+		/* Optional, mode 2HA */
+	{ PCI_DESCR(28, 0, PCI_DEVICE_ID_INTEL_IBRIDGE_IMC_HA1, 1)	},
+#if 0
+	{ PCI_DESCR(29, 0, PCI_DEVICE_ID_INTEL_IBRIDGE_IMC_HA1_TA, 1)	},
+	{ PCI_DESCR(29, 1, PCI_DEVICE_ID_INTEL_IBRIDGE_IMC_HA1_RAS, 1)	},
+#endif
+	{ PCI_DESCR(29, 2, PCI_DEVICE_ID_INTEL_IBRIDGE_IMC_HA1_TAD0, 1)	},
+	{ PCI_DESCR(29, 3, PCI_DEVICE_ID_INTEL_IBRIDGE_IMC_HA1_TAD1, 1)	},
+
+	{ PCI_DESCR(17, 0, PCI_DEVICE_ID_INTEL_IBRIDGE_IMC_1HA_DDRIO0, 1) },
+	{ PCI_DESCR(17, 4, PCI_DEVICE_ID_INTEL_IBRIDGE_IMC_2HA_DDRIO0, 1) },
+};
+
+static const struct pci_id_table pci_dev_descr_ibridge_table[] = {
+	PCI_ID_TABLE_ENTRY(pci_dev_descr_ibridge),
+	{0,}			/* 0 terminated list. */
+};
+
 /*
  *	pci_device_id	table for which devices we are looking for
  */
 static DEFINE_PCI_DEVICE_TABLE(sbridge_pci_tbl) = {
 	{PCI_DEVICE(PCI_VENDOR_ID_INTEL, PCI_DEVICE_ID_INTEL_SBRIDGE_IMC_TA)},
+	{PCI_DEVICE(PCI_VENDOR_ID_INTEL, PCI_DEVICE_ID_INTEL_IBRIDGE_IMC_HA0_TA)},
 	{0,}			/* 0 terminated list. */
 };
 
@@ -458,6 +553,52 @@ static void free_sbridge_dev(struct sbridge_dev *sbridge_dev)
 	kfree(sbridge_dev);
 }
 
+static u64 sbridge_get_tolm(struct sbridge_pvt *pvt)
+{
+	u32 reg;
+
+	/* Address range is 32:28 */
+	pci_read_config_dword(pvt->pci_sad1, TOLM, &reg);
+	return GET_TOLM(reg);
+}
+
+static u64 sbridge_get_tohm(struct sbridge_pvt *pvt)
+{
+	u32 reg;
+
+	pci_read_config_dword(pvt->pci_sad1, TOHM, &reg);
+	return GET_TOHM(reg);
+}
+
+static u64 ibridge_get_tolm(struct sbridge_pvt *pvt)
+{
+	u32 reg;
+
+	pci_read_config_dword(pvt->pci_br1, TOLM, &reg);
+
+	return GET_TOLM(reg);
+}
+
+static u64 ibridge_get_tohm(struct sbridge_pvt *pvt)
+{
+	u32 reg;
+
+	pci_read_config_dword(pvt->pci_br1, TOHM, &reg);
+
+	return GET_TOHM(reg);
+}
+
+static inline u8 sad_pkg_socket(u8 pkg)
+{
+	/* on Ivy Bridge, nodeID is SASS, where A is HA and S is node id */
+	return (pkg >> 3) | (pkg & 0x3);
+}
+
+static inline u8 sad_pkg_ha(u8 pkg)
+{
+	return (pkg >> 2) & 0x1;
+}
+
 /****************************************************************************
 			Memory check routines
  ****************************************************************************/
@@ -520,10 +661,10 @@ static int get_dimm_config(struct mem_ctl_info *mci)
 	enum edac_type mode;
 	enum mem_type mtype;
 
-	pci_read_config_dword(pvt->pci_br, SAD_TARGET, &reg);
+	pci_read_config_dword(pvt->pci_br0, SAD_TARGET, &reg);
 	pvt->sbridge_dev->source_id = SOURCE_ID(reg);
 
-	pci_read_config_dword(pvt->pci_br, SAD_CONTROL, &reg);
+	pci_read_config_dword(pvt->pci_br0, SAD_CONTROL, &reg);
 	pvt->sbridge_dev->node_id = NODE_ID(reg);
 	edac_dbg(0, "mc#%d: Node ID: %d, source ID: %d\n",
 		 pvt->sbridge_dev->mc,
@@ -558,7 +699,8 @@ static int get_dimm_config(struct mem_ctl_info *mci)
 	}
 
 	if (pvt->pci_ddrio) {
-		pci_read_config_dword(pvt->pci_ddrio, RANK_CFG_A, &reg);
+		pci_read_config_dword(pvt->pci_ddrio, pvt->info.rankcfgr,
+				      &reg);
 		if (IS_RDIMM_ENABLED(reg)) {
 			/* FIXME: Can also be LRDIMM */
 			edac_dbg(0, "Memory is registered\n");
@@ -629,19 +771,14 @@ static void get_memory_layout(const struct mem_ctl_info *mci)
 	 * Step 1) Get TOLM/TOHM ranges
 	 */
 
-	/* Address range is 32:28 */
-	pci_read_config_dword(pvt->pci_sad1, TOLM,
-			      &reg);
-	pvt->tolm = GET_TOLM(reg);
+	pvt->tolm = pvt->info.get_tolm(pvt);
 	tmp_mb = (1 + pvt->tolm) >> 20;
 
 	mb = div_u64_rem(tmp_mb, 1000, &kb);
 	edac_dbg(0, "TOLM: %u.%03u GB (0x%016Lx)\n", mb, kb, (u64)pvt->tolm);
 
 	/* Address range is already 45:25 */
-	pci_read_config_dword(pvt->pci_sad1, TOHM,
-			      &reg);
-	pvt->tohm = GET_TOHM(reg);
+	pvt->tohm = pvt->info.get_tohm(pvt);
 	tmp_mb = (1 + pvt->tohm) >> 20;
 
 	mb = div_u64_rem(tmp_mb, 1000, &kb);
@@ -654,9 +791,9 @@ static void get_memory_layout(const struct mem_ctl_info *mci)
 	 * algorithm bellow.
 	 */
 	prv = 0;
-	for (n_sads = 0; n_sads < MAX_SAD; n_sads++) {
+	for (n_sads = 0; n_sads < pvt->info.max_sad; n_sads++) {
 		/* SAD_LIMIT Address range is 45:26 */
-		pci_read_config_dword(pvt->pci_sad0, dram_rule[n_sads],
+		pci_read_config_dword(pvt->pci_sad0, pvt->info.dram_rule[n_sads],
 				      &reg);
 		limit = SAD_LIMIT(reg);
 
@@ -677,15 +814,16 @@ static void get_memory_layout(const struct mem_ctl_info *mci)
 			 reg);
 		prv = limit;
 
-		pci_read_config_dword(pvt->pci_sad0, interleave_list[n_sads],
+		pci_read_config_dword(pvt->pci_sad0, pvt->info.interleave_list[n_sads],
 				      &reg);
-		sad_interl = sad_pkg(reg, 0);
+		sad_interl = sad_pkg(pvt->info.interleave_pkg, reg, 0);
 		for (j = 0; j < 8; j++) {
-			if (j > 0 && sad_interl == sad_pkg(reg, j))
+			u32 pkg = sad_pkg(pvt->info.interleave_pkg, reg, j);
+			if (j > 0 && sad_interl == pkg)
 				break;
 
 			edac_dbg(0, "SAD#%d, interleave #%d: %d\n",
-				 n_sads, j, sad_pkg(reg, j));
+				 n_sads, j, pkg);
 		}
 	}
 
@@ -797,12 +935,13 @@ static int get_memory_error_data(struct mem_ctl_info *mci,
 {
 	struct mem_ctl_info	*new_mci;
 	struct sbridge_pvt *pvt = mci->pvt_info;
+	struct pci_dev		*pci_ha;
 	int 			n_rir, n_sads, n_tads, sad_way, sck_xch;
 	int			sad_interl, idx, base_ch;
 	int			interleave_mode;
-	unsigned		sad_interleave[MAX_INTERLEAVE];
+	unsigned		sad_interleave[pvt->info.max_interleave];
 	u32			reg;
-	u8			ch_way,sck_way;
+	u8			ch_way, sck_way, pkg, sad_ha = 0;
 	u32			tad_offset;
 	u32			rir_way;
 	u32			mb, kb;
@@ -828,8 +967,8 @@ static int get_memory_error_data(struct mem_ctl_info *mci,
 	/*
 	 * Step 1) Get socket
 	 */
-	for (n_sads = 0; n_sads < MAX_SAD; n_sads++) {
-		pci_read_config_dword(pvt->pci_sad0, dram_rule[n_sads],
+	for (n_sads = 0; n_sads < pvt->info.max_sad; n_sads++) {
+		pci_read_config_dword(pvt->pci_sad0, pvt->info.dram_rule[n_sads],
 				      &reg);
 
 		if (!DRAM_RULE_ENABLE(reg))
@@ -844,53 +983,65 @@ static int get_memory_error_data(struct mem_ctl_info *mci,
 			break;
 		prv = limit;
 	}
-	if (n_sads == MAX_SAD) {
+	if (n_sads == pvt->info.max_sad) {
 		sprintf(msg, "Can't discover the memory socket");
 		return -EINVAL;
 	}
 	*area_type = get_dram_attr(reg);
 	interleave_mode = INTERLEAVE_MODE(reg);
 
-	pci_read_config_dword(pvt->pci_sad0, interleave_list[n_sads],
+	pci_read_config_dword(pvt->pci_sad0, pvt->info.interleave_list[n_sads],
 			      &reg);
-	sad_interl = sad_pkg(reg, 0);
-	for (sad_way = 0; sad_way < 8; sad_way++) {
-		if (sad_way > 0 && sad_interl == sad_pkg(reg, sad_way))
+
+	if (pvt->info.type == SANDY_BRIDGE) {
+		sad_interl = sad_pkg(pvt->info.interleave_pkg, reg, 0);
+		for (sad_way = 0; sad_way < 8; sad_way++) {
+			u32 pkg = sad_pkg(pvt->info.interleave_pkg, reg, sad_way);
+			if (sad_way > 0 && sad_interl == pkg)
+				break;
+			sad_interleave[sad_way] = pkg;
+			edac_dbg(0, "SAD interleave #%d: %d\n",
+				 sad_way, sad_interleave[sad_way]);
+		}
+		edac_dbg(0, "mc#%d: Error detected on SAD#%d: address 0x%016Lx < 0x%016Lx, Interleave [%d:6]%s\n",
+			 pvt->sbridge_dev->mc,
+			 n_sads,
+			 addr,
+			 limit,
+			 sad_way + 7,
+			 !interleave_mode ? "" : "XOR[18:16]");
+		if (interleave_mode)
+			idx = ((addr >> 6) ^ (addr >> 16)) & 7;
+		else
+			idx = (addr >> 6) & 7;
+		switch (sad_way) {
+		case 1:
+			idx = 0;
 			break;
-		sad_interleave[sad_way] = sad_pkg(reg, sad_way);
-		edac_dbg(0, "SAD interleave #%d: %d\n",
-			 sad_way, sad_interleave[sad_way]);
-	}
-	edac_dbg(0, "mc#%d: Error detected on SAD#%d: address 0x%016Lx < 0x%016Lx, Interleave [%d:6]%s\n",
-		 pvt->sbridge_dev->mc,
-		 n_sads,
-		 addr,
-		 limit,
-		 sad_way + 7,
-		 interleave_mode ? "" : "XOR[18:16]");
-	if (interleave_mode)
-		idx = ((addr >> 6) ^ (addr >> 16)) & 7;
-	else
+		case 2:
+			idx = idx & 1;
+			break;
+		case 4:
+			idx = idx & 3;
+			break;
+		case 8:
+			break;
+		default:
+			sprintf(msg, "Can't discover socket interleave");
+			return -EINVAL;
+		}
+		*socket = sad_interleave[idx];
+		edac_dbg(0, "SAD interleave index: %d (wayness %d) = CPU socket %d\n",
+			 idx, sad_way, *socket);
+	} else {
+		/* Ivy Bridge's SAD mode doesn't support XOR interleave mode */
 		idx = (addr >> 6) & 7;
-	switch (sad_way) {
-	case 1:
-		idx = 0;
-		break;
-	case 2:
-		idx = idx & 1;
-		break;
-	case 4:
-		idx = idx & 3;
-		break;
-	case 8:
-		break;
-	default:
-		sprintf(msg, "Can't discover socket interleave");
-		return -EINVAL;
+		pkg = sad_pkg(pvt->info.interleave_pkg, reg, idx);
+		*socket = sad_pkg_socket(pkg);
+		sad_ha = sad_pkg_ha(pkg);
+		edac_dbg(0, "SAD interleave package: %d = CPU socket %d, HA %d\n",
+			 idx, *socket, sad_ha);
 	}
-	*socket = sad_interleave[idx];
-	edac_dbg(0, "SAD interleave index: %d (wayness %d) = CPU socket %d\n",
-		 idx, sad_way, *socket);
 
 	/*
 	 * Move to the proper node structure, in order to access the
@@ -909,9 +1060,16 @@ static int get_memory_error_data(struct mem_ctl_info *mci,
 	 * Step 2) Get memory channel
 	 */
 	prv = 0;
+	if (pvt->info.type == SANDY_BRIDGE)
+		pci_ha = pvt->pci_ha0;
+	else {
+		if (sad_ha)
+			pci_ha = pvt->pci_ha1;
+		else
+			pci_ha = pvt->pci_ha0;
+	}
 	for (n_tads = 0; n_tads < MAX_TAD; n_tads++) {
-		pci_read_config_dword(pvt->pci_ha0, tad_dram_rule[n_tads],
-				      &reg);
+		pci_read_config_dword(pci_ha, tad_dram_rule[n_tads], &reg);
 		limit = TAD_LIMIT(reg);
 		if (limit <= prv) {
 			sprintf(msg, "Can't discover the memory channel");
@@ -921,14 +1079,13 @@ static int get_memory_error_data(struct mem_ctl_info *mci,
 			break;
 		prv = limit;
 	}
+	if (n_tads == MAX_TAD) {
+		sprintf(msg, "Can't discover the memory channel");
+		return -EINVAL;
+	}
+
 	ch_way = TAD_CH(reg) + 1;
 	sck_way = TAD_SOCK(reg) + 1;
-	/*
-	 * FIXME: Is it right to always use channel 0 for offsets?
-	 */
-	pci_read_config_dword(pvt->pci_tad[0],
-				tad_ch_nilv_offset[n_tads],
-				&tad_offset);
 
 	if (ch_way == 3)
 		idx = addr >> 6;
@@ -958,6 +1115,10 @@ static int get_memory_error_data(struct mem_ctl_info *mci,
 	}
 	*channel_mask = 1 << base_ch;
 
+	pci_read_config_dword(pvt->pci_tad[base_ch],
+				tad_ch_nilv_offset[n_tads],
+				&tad_offset);
+
 	if (pvt->is_mirrored) {
 		*channel_mask |= 1 << ((base_ch + 2) % 4);
 		switch(ch_way) {
@@ -1091,12 +1252,6 @@ static void sbridge_put_all_devices(void)
 	}
 }
 
-/*
- *	sbridge_get_all_devices	Find and perform 'get' operation on the MCH's
- *			device/functions we want to reference for this driver
- *
- *			Need to 'get' device 16 func 1 and func 2
- */
 static int sbridge_get_onedevice(struct pci_dev **prev,
 				 u8 *num_mc,
 				 const struct pci_id_table *table,
@@ -1198,11 +1353,21 @@ static int sbridge_get_onedevice(struct pci_dev **prev,
 	return 0;
 }
 
-static int sbridge_get_all_devices(u8 *num_mc)
+/*
+ * sbridge_get_all_devices - Find and perform 'get' operation on the MCH's
+ *			     device/functions we want to reference for this driver.
+ *			     Need to 'get' device 16 func 1 and func 2.
+ * @num_mc: pointer to the memory controllers count, to be incremented in case
+ * 	    of success.
+ * @table: model specific table
+ *
+ * returns 0 in case of success or error code
+ */
+static int sbridge_get_all_devices(u8 *num_mc,
+				   const struct pci_id_table *table)
 {
 	int i, rc;
 	struct pci_dev *pdev = NULL;
-	const struct pci_id_table *table = pci_dev_descr_sbridge_table;
 
 	while (table && table->descr) {
 		for (i = 0; i < table->n_devs; i++) {
@@ -1226,8 +1391,8 @@ static int sbridge_get_all_devices(u8 *num_mc)
 	return 0;
 }
 
-static int mci_bind_devs(struct mem_ctl_info *mci,
-			 struct sbridge_dev *sbridge_dev)
+static int sbridge_mci_bind_devs(struct mem_ctl_info *mci,
+				 struct sbridge_dev *sbridge_dev)
 {
 	struct sbridge_pvt *pvt = mci->pvt_info;
 	struct pci_dev *pdev;
@@ -1255,7 +1420,7 @@ static int mci_bind_devs(struct mem_ctl_info *mci,
 		case 13:
 			switch (func) {
 			case 6:
-				pvt->pci_br = pdev;
+				pvt->pci_br0 = pdev;
 				break;
 			default:
 				goto error;
@@ -1329,6 +1494,131 @@ error:
 	return -EINVAL;
 }
 
+static int ibridge_mci_bind_devs(struct mem_ctl_info *mci,
+				 struct sbridge_dev *sbridge_dev)
+{
+	struct sbridge_pvt *pvt = mci->pvt_info;
+	struct pci_dev *pdev, *tmp;
+	int i, func, slot;
+	bool mode_2ha = false;
+
+	tmp = pci_get_device(PCI_VENDOR_ID_INTEL,
+			     PCI_DEVICE_ID_INTEL_IBRIDGE_IMC_HA1, NULL);
+	if (tmp) {
+		mode_2ha = true;
+		pci_dev_put(tmp);
+	}
+
+	for (i = 0; i < sbridge_dev->n_devs; i++) {
+		pdev = sbridge_dev->pdev[i];
+		if (!pdev)
+			continue;
+		slot = PCI_SLOT(pdev->devfn);
+		func = PCI_FUNC(pdev->devfn);
+
+		switch (slot) {
+		case 14:
+			if (func == 0) {
+				pvt->pci_ha0 = pdev;
+				break;
+			}
+			goto error;
+		case 15:
+			switch (func) {
+			case 0:
+				pvt->pci_ta = pdev;
+				break;
+			case 1:
+				pvt->pci_ras = pdev;
+				break;
+			case 4:
+			case 5:
+				/* if we have 2 HAs active, channels 2 and 3
+				 * are in other device */
+				if (mode_2ha)
+					break;
+				/* fall through */
+			case 2:
+			case 3:
+				pvt->pci_tad[func - 2] = pdev;
+				break;
+			default:
+				goto error;
+			}
+			break;
+		case 17:
+			if (func == 4) {
+				pvt->pci_ddrio = pdev;
+				break;
+			} else if (func == 0) {
+				if (!mode_2ha)
+					pvt->pci_ddrio = pdev;
+				break;
+			}
+			goto error;
+		case 22:
+			switch (func) {
+			case 0:
+				pvt->pci_sad0 = pdev;
+				break;
+			case 1:
+				pvt->pci_br0 = pdev;
+				break;
+			case 2:
+				pvt->pci_br1 = pdev;
+				break;
+			default:
+				goto error;
+			}
+			break;
+		case 28:
+			if (func == 0) {
+				pvt->pci_ha1 = pdev;
+				break;
+			}
+			goto error;
+		case 29:
+			/* we shouldn't have this device if we have just one
+			 * HA present */
+			WARN_ON(!mode_2ha);
+			if (func == 2 || func == 3) {
+				pvt->pci_tad[func] = pdev;
+				break;
+			}
+			goto error;
+		default:
+			goto error;
+		}
+
+		edac_dbg(0, "Associated PCI %02x.%02d.%d with dev = %p\n",
+			 sbridge_dev->bus,
+			 PCI_SLOT(pdev->devfn), PCI_FUNC(pdev->devfn),
+			 pdev);
+	}
+
+	/* Check if everything were registered */
+	if (!pvt->pci_sad0 || !pvt->pci_ha0 || !pvt->pci_br0 ||
+	    !pvt->pci_br1 || !pvt->pci_tad || !pvt->pci_ras  ||
+	    !pvt->pci_ta)
+		goto enodev;
+
+	for (i = 0; i < NUM_CHANNELS; i++) {
+		if (!pvt->pci_tad[i])
+			goto enodev;
+	}
+	return 0;
+
+enodev:
+	sbridge_printk(KERN_ERR, "Some needed devices are missing\n");
+	return -ENODEV;
+
+error:
+	sbridge_printk(KERN_ERR,
+		       "Device %d, function %d is out of the expected range\n",
+		       slot, func);
+	return -EINVAL;
+}
+
 /****************************************************************************
 			Error check routines
  ****************************************************************************/
@@ -1349,7 +1639,7 @@ static void sbridge_mce_output_error(struct mem_ctl_info *mci,
 	bool ripv = GET_BITFIELD(m->mcgstatus, 0, 0);
 	bool overflow = GET_BITFIELD(m->status, 62, 62);
 	bool uncorrected_error = GET_BITFIELD(m->status, 61, 61);
-	bool recoverable = GET_BITFIELD(m->status, 56, 56);
+	bool recoverable;
 	u32 core_err_cnt = GET_BITFIELD(m->status, 38, 52);
 	u32 mscod = GET_BITFIELD(m->status, 16, 31);
 	u32 errcode = GET_BITFIELD(m->status, 0, 15);
@@ -1360,6 +1650,11 @@ static void sbridge_mce_output_error(struct mem_ctl_info *mci,
 	int rc, dimm;
 	char *area_type = NULL;
 
+	if (pvt->info.type == IVY_BRIDGE)
+		recoverable = true;
+	else
+		recoverable = GET_BITFIELD(m->status, 56, 56);
+
 	if (uncorrected_error) {
 		if (ripv) {
 			type = "FATAL";
@@ -1409,6 +1704,10 @@ static void sbridge_mce_output_error(struct mem_ctl_info *mci,
 		}
 	}
 
+	/* Only decode errors with an valid address (ADDRV) */
+	if (!GET_BITFIELD(m->status, 58, 58))
+		return;
+
 	rc = get_memory_error_data(mci, m->addr, &socket,
 				   &channel_mask, &rank, &area_type, msg);
 	if (rc < 0)
@@ -1614,11 +1913,12 @@ static void sbridge_unregister_mci(struct sbridge_dev *sbridge_dev)
 	sbridge_dev->mci = NULL;
 }
 
-static int sbridge_register_mci(struct sbridge_dev *sbridge_dev)
+static int sbridge_register_mci(struct sbridge_dev *sbridge_dev, enum type type)
 {
 	struct mem_ctl_info *mci;
 	struct edac_mc_layer layers[2];
 	struct sbridge_pvt *pvt;
+	struct pci_dev *pdev = sbridge_dev->pdev[0];
 	int rc;
 
 	/* Check the number of active and not disabled channels */
@@ -1640,7 +1940,7 @@ static int sbridge_register_mci(struct sbridge_dev *sbridge_dev)
 		return -ENOMEM;
 
 	edac_dbg(0, "MC: mci = %p, dev = %p\n",
-		 mci, &sbridge_dev->pdev[0]->dev);
+		 mci, &pdev->dev);
 
 	pvt = mci->pvt_info;
 	memset(pvt, 0, sizeof(*pvt));
@@ -1654,24 +1954,52 @@ static int sbridge_register_mci(struct sbridge_dev *sbridge_dev)
 	mci->edac_cap = EDAC_FLAG_NONE;
 	mci->mod_name = "sbridge_edac.c";
 	mci->mod_ver = SBRIDGE_REVISION;
-	mci->ctl_name = kasprintf(GFP_KERNEL, "Sandy Bridge Socket#%d", mci->mc_idx);
-	mci->dev_name = pci_name(sbridge_dev->pdev[0]);
+	mci->dev_name = pci_name(pdev);
 	mci->ctl_page_to_phys = NULL;
 
 	/* Set the function pointer to an actual operation function */
 	mci->edac_check = sbridge_check_error;
 
-	/* Store pci devices at mci for faster access */
-	rc = mci_bind_devs(mci, sbridge_dev);
-	if (unlikely(rc < 0))
-		goto fail0;
+	pvt->info.type = type;
+	if (type == IVY_BRIDGE) {
+		pvt->info.rankcfgr = IB_RANK_CFG_A;
+		pvt->info.get_tolm = ibridge_get_tolm;
+		pvt->info.get_tohm = ibridge_get_tohm;
+		pvt->info.dram_rule = ibridge_dram_rule;
+		pvt->info.max_sad = ARRAY_SIZE(ibridge_dram_rule);
+		pvt->info.interleave_list = ibridge_interleave_list;
+		pvt->info.max_interleave = ARRAY_SIZE(ibridge_interleave_list);
+		pvt->info.interleave_pkg = ibridge_interleave_pkg;
+		mci->ctl_name = kasprintf(GFP_KERNEL, "Ivy Bridge Socket#%d", mci->mc_idx);
+
+		/* Store pci devices at mci for faster access */
+		rc = ibridge_mci_bind_devs(mci, sbridge_dev);
+		if (unlikely(rc < 0))
+			goto fail0;
+	} else {
+		pvt->info.rankcfgr = SB_RANK_CFG_A;
+		pvt->info.get_tolm = sbridge_get_tolm;
+		pvt->info.get_tohm = sbridge_get_tohm;
+		pvt->info.dram_rule = sbridge_dram_rule;
+		pvt->info.max_sad = ARRAY_SIZE(sbridge_dram_rule);
+		pvt->info.interleave_list = sbridge_interleave_list;
+		pvt->info.max_interleave = ARRAY_SIZE(sbridge_interleave_list);
+		pvt->info.interleave_pkg = sbridge_interleave_pkg;
+		mci->ctl_name = kasprintf(GFP_KERNEL, "Sandy Bridge Socket#%d", mci->mc_idx);
+
+		/* Store pci devices at mci for faster access */
+		rc = sbridge_mci_bind_devs(mci, sbridge_dev);
+		if (unlikely(rc < 0))
+			goto fail0;
+	}
+
 
 	/* Get dimm basic config and the memory layout */
 	get_dimm_config(mci);
 	get_memory_layout(mci);
 
 	/* record ptr to the generic device */
-	mci->pdev = &sbridge_dev->pdev[0]->dev;
+	mci->pdev = &pdev->dev;
 
 	/* add this new MC control structure to EDAC's list of MCs */
 	if (unlikely(edac_mc_add_mc(mci))) {
@@ -1702,6 +2030,7 @@ static int sbridge_probe(struct pci_dev *pdev, const struct pci_device_id *id)
 	int rc;
 	u8 mc, num_mc = 0;
 	struct sbridge_dev *sbridge_dev;
+	enum type type;
 
 	/* get the pci devices we want to reserve for our use */
 	mutex_lock(&sbridge_edac_lock);
@@ -1715,7 +2044,13 @@ static int sbridge_probe(struct pci_dev *pdev, const struct pci_device_id *id)
 	}
 	probed++;
 
-	rc = sbridge_get_all_devices(&num_mc);
+	if (pdev->device == PCI_DEVICE_ID_INTEL_IBRIDGE_IMC_HA0_TA) {
+		rc = sbridge_get_all_devices(&num_mc, pci_dev_descr_ibridge_table);
+		type = IVY_BRIDGE;
+	} else {
+		rc = sbridge_get_all_devices(&num_mc, pci_dev_descr_sbridge_table);
+		type = SANDY_BRIDGE;
+	}
 	if (unlikely(rc < 0))
 		goto fail0;
 	mc = 0;
@@ -1724,7 +2059,7 @@ static int sbridge_probe(struct pci_dev *pdev, const struct pci_device_id *id)
 		edac_dbg(0, "Registering MC#%d (%d of %d)\n",
 			 mc, mc + 1, num_mc);
 		sbridge_dev->mc = mc++;
-		rc = sbridge_register_mci(sbridge_dev);
+		rc = sbridge_register_mci(sbridge_dev, type);
 		if (unlikely(rc < 0))
 			goto fail1;
 	}
@@ -1839,5 +2174,5 @@ MODULE_PARM_DESC(edac_op_state, "EDAC Error Reporting state: 0=Poll,1=NMI");
 MODULE_LICENSE("GPL");
 MODULE_AUTHOR("Mauro Carvalho Chehab <mchehab@redhat.com>");
 MODULE_AUTHOR("Red Hat Inc. (http://www.redhat.com)");
-MODULE_DESCRIPTION("MC Driver for Intel Sandy Bridge memory controllers - "
+MODULE_DESCRIPTION("MC Driver for Intel Sandy Bridge and Ivy Bridge memory controllers - "
 		   SBRIDGE_REVISION);
diff --git a/drivers/firewire/core-transaction.c b/drivers/firewire/core-transaction.c
index e5af0e3a26ec..0e799516a2ab 100644
--- a/drivers/firewire/core-transaction.c
+++ b/drivers/firewire/core-transaction.c
@@ -477,7 +477,7 @@ void fw_send_phy_config(struct fw_card *card,
 	phy_config_packet.header[1] = data;
 	phy_config_packet.header[2] = ~data;
 	phy_config_packet.generation = generation;
-	INIT_COMPLETION(phy_config_done);
+	reinit_completion(&phy_config_done);
 
 	card->driver->send_request(card, &phy_config_packet);
 	wait_for_completion_timeout(&phy_config_done, timeout);
diff --git a/drivers/fmc/Kconfig b/drivers/fmc/Kconfig
index c01cf45bc3d8..3a75f4256d08 100644
--- a/drivers/fmc/Kconfig
+++ b/drivers/fmc/Kconfig
@@ -46,6 +46,6 @@ config FMC_CHARDEV
 	  This driver matches every mezzanine device and allows user
 	  space to read and write registers using a char device. It
 	  can be used to write user-space drivers, or just get
-	  aquainted with a mezzanine before writing its specific driver.
+	  acquainted with a mezzanine before writing its specific driver.
 
 endif # FMC
diff --git a/drivers/gpu/drm/Kconfig b/drivers/gpu/drm/Kconfig
index 955555d6ec88..f86427591167 100644
--- a/drivers/gpu/drm/Kconfig
+++ b/drivers/gpu/drm/Kconfig
@@ -29,11 +29,17 @@ config DRM_USB
 config DRM_KMS_HELPER
 	tristate
 	depends on DRM
+	help
+	  CRTC helpers for KMS drivers.
+
+config DRM_KMS_FB_HELPER
+	bool
+	depends on DRM_KMS_HELPER
 	select FB
 	select FRAMEBUFFER_CONSOLE if !EXPERT
 	select FRAMEBUFFER_CONSOLE_DETECT_PRIMARY if FRAMEBUFFER_CONSOLE
 	help
-	  FB and CRTC helpers for KMS drivers.
+	  FBDEV helpers for KMS drivers.
 
 config DRM_LOAD_EDID_FIRMWARE
 	bool "Allow to specify an EDID data set instead of probing for it"
@@ -64,6 +70,7 @@ config DRM_GEM_CMA_HELPER
 config DRM_KMS_CMA_HELPER
 	bool
 	select DRM_GEM_CMA_HELPER
+	select DRM_KMS_FB_HELPER
 	select FB_SYS_FILLRECT
 	select FB_SYS_COPYAREA
 	select FB_SYS_IMAGEBLIT
@@ -96,6 +103,7 @@ config DRM_RADEON
 	select FB_CFB_IMAGEBLIT
 	select FW_LOADER
         select DRM_KMS_HELPER
+	select DRM_KMS_FB_HELPER
         select DRM_TTM
 	select POWER_SUPPLY
 	select HWMON
@@ -120,64 +128,7 @@ config DRM_I810
 	  selected, the module will be called i810.  AGP support is required
 	  for this driver to work.
 
-config DRM_I915
-	tristate "Intel 8xx/9xx/G3x/G4x/HD Graphics"
-	depends on DRM
-	depends on AGP
-	depends on AGP_INTEL
-	# we need shmfs for the swappable backing store, and in particular
-	# the shmem_readpage() which depends upon tmpfs
-	select SHMEM
-	select TMPFS
-	select DRM_KMS_HELPER
-	select FB_CFB_FILLRECT
-	select FB_CFB_COPYAREA
-	select FB_CFB_IMAGEBLIT
-	# i915 depends on ACPI_VIDEO when ACPI is enabled
-	# but for select to work, need to select ACPI_VIDEO's dependencies, ick
-	select BACKLIGHT_LCD_SUPPORT if ACPI
-	select BACKLIGHT_CLASS_DEVICE if ACPI
-	select VIDEO_OUTPUT_CONTROL if ACPI
-	select INPUT if ACPI
-	select THERMAL if ACPI
-	select ACPI_VIDEO if ACPI
-	select ACPI_BUTTON if ACPI
-	help
-	  Choose this option if you have a system that has "Intel Graphics
-	  Media Accelerator" or "HD Graphics" integrated graphics,
-	  including 830M, 845G, 852GM, 855GM, 865G, 915G, 945G, 965G,
-	  G35, G41, G43, G45 chipsets and Celeron, Pentium, Core i3,
-	  Core i5, Core i7 as well as Atom CPUs with integrated graphics.
-	  If M is selected, the module will be called i915.  AGP support
-	  is required for this driver to work. This driver is used by
-	  the Intel driver in X.org 6.8 and XFree86 4.4 and above. It
-	  replaces the older i830 module that supported a subset of the
-	  hardware in older X.org releases.
-
-	  Note that the older i810/i815 chipsets require the use of the
-	  i810 driver instead, and the Atom z5xx series has an entirely
-	  different implementation.
-
-config DRM_I915_KMS
-	bool "Enable modesetting on intel by default"
-	depends on DRM_I915
-	help
-	  Choose this option if you want kernel modesetting enabled by default,
-	  and you have a new enough userspace to support this. Running old
-	  userspaces with this enabled will cause pain.  Note that this causes
-	  the driver to bind to PCI devices, which precludes loading things
-	  like intelfb.
-
-config DRM_I915_PRELIMINARY_HW_SUPPORT
-	bool "Enable preliminary support for prerelease Intel hardware by default"
-	depends on DRM_I915
-	help
-	  Choose this option if you have prerelease Intel hardware and want the
-	  i915 driver to support it by default.  You can enable such support at
-	  runtime with the module option i915.preliminary_hw_support=1; this
-	  option changes the default for that module option.
-
-	  If in doubt, say "N".
+source "drivers/gpu/drm/i915/Kconfig"
 
 config DRM_MGA
 	tristate "Matrox g200/g400"
@@ -225,6 +176,8 @@ source "drivers/gpu/drm/mgag200/Kconfig"
 
 source "drivers/gpu/drm/cirrus/Kconfig"
 
+source "drivers/gpu/drm/armada/Kconfig"
+
 source "drivers/gpu/drm/rcar-du/Kconfig"
 
 source "drivers/gpu/drm/shmobile/Kconfig"
@@ -236,3 +189,5 @@ source "drivers/gpu/drm/tilcdc/Kconfig"
 source "drivers/gpu/drm/qxl/Kconfig"
 
 source "drivers/gpu/drm/msm/Kconfig"
+
+source "drivers/gpu/drm/tegra/Kconfig"
diff --git a/drivers/gpu/drm/Makefile b/drivers/gpu/drm/Makefile
index f089adfe70ee..cc08b845f965 100644
--- a/drivers/gpu/drm/Makefile
+++ b/drivers/gpu/drm/Makefile
@@ -21,8 +21,9 @@ drm-$(CONFIG_PCI) += ati_pcigart.o
 
 drm-usb-y   := drm_usb.o
 
-drm_kms_helper-y := drm_fb_helper.o drm_crtc_helper.o drm_dp_helper.o
+drm_kms_helper-y := drm_crtc_helper.o drm_dp_helper.o
 drm_kms_helper-$(CONFIG_DRM_LOAD_EDID_FIRMWARE) += drm_edid_load.o
+drm_kms_helper-$(CONFIG_DRM_KMS_FB_HELPER) += drm_fb_helper.o
 drm_kms_helper-$(CONFIG_DRM_KMS_CMA_HELPER) += drm_fb_cma_helper.o
 
 obj-$(CONFIG_DRM_KMS_HELPER) += drm_kms_helper.o
@@ -49,10 +50,12 @@ obj-$(CONFIG_DRM_EXYNOS) +=exynos/
 obj-$(CONFIG_DRM_GMA500) += gma500/
 obj-$(CONFIG_DRM_UDL) += udl/
 obj-$(CONFIG_DRM_AST) += ast/
+obj-$(CONFIG_DRM_ARMADA) += armada/
 obj-$(CONFIG_DRM_RCAR_DU) += rcar-du/
 obj-$(CONFIG_DRM_SHMOBILE) +=shmobile/
 obj-$(CONFIG_DRM_OMAP)	+= omapdrm/
 obj-$(CONFIG_DRM_TILCDC)	+= tilcdc/
 obj-$(CONFIG_DRM_QXL) += qxl/
 obj-$(CONFIG_DRM_MSM) += msm/
+obj-$(CONFIG_DRM_TEGRA) += tegra/
 obj-y			+= i2c/
diff --git a/drivers/gpu/drm/armada/Kconfig b/drivers/gpu/drm/armada/Kconfig
new file mode 100644
index 000000000000..40d371521fe1
--- /dev/null
+++ b/drivers/gpu/drm/armada/Kconfig
@@ -0,0 +1,24 @@
+config DRM_ARMADA
+	tristate "DRM support for Marvell Armada SoCs"
+	depends on DRM && HAVE_CLK && ARM
+	select FB_CFB_FILLRECT
+	select FB_CFB_COPYAREA
+	select FB_CFB_IMAGEBLIT
+	select DRM_KMS_HELPER
+	help
+	  Support the "LCD" controllers found on the Marvell Armada 510
+	  devices.  There are two controllers on the device, each controller
+	  supports graphics and video overlays.
+
+	  This driver provides no built-in acceleration; acceleration is
+	  performed by other IP found on the SoC.  This driver provides
+	  kernel mode setting and buffer management to userspace.
+
+config DRM_ARMADA_TDA1998X
+	bool "Support TDA1998X HDMI output"
+	depends on DRM_ARMADA != n
+	depends on I2C && DRM_I2C_NXP_TDA998X = y
+	default y
+	help
+	  Support the TDA1998x HDMI output device found on the Solid-Run
+	  CuBox.
diff --git a/drivers/gpu/drm/armada/Makefile b/drivers/gpu/drm/armada/Makefile
new file mode 100644
index 000000000000..d6f43e06150a
--- /dev/null
+++ b/drivers/gpu/drm/armada/Makefile
@@ -0,0 +1,7 @@
+armada-y	:= armada_crtc.o armada_drv.o armada_fb.o armada_fbdev.o \
+		   armada_gem.o armada_output.o armada_overlay.o \
+		   armada_slave.o
+armada-y	+= armada_510.o
+armada-$(CONFIG_DEBUG_FS) += armada_debugfs.o
+
+obj-$(CONFIG_DRM_ARMADA) := armada.o
diff --git a/drivers/gpu/drm/armada/armada_510.c b/drivers/gpu/drm/armada/armada_510.c
new file mode 100644
index 000000000000..59948eff6095
--- /dev/null
+++ b/drivers/gpu/drm/armada/armada_510.c
@@ -0,0 +1,87 @@
+/*
+ * Copyright (C) 2012 Russell King
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ *
+ * Armada 510 (aka Dove) variant support
+ */
+#include <linux/clk.h>
+#include <linux/io.h>
+#include <drm/drmP.h>
+#include <drm/drm_crtc_helper.h>
+#include "armada_crtc.h"
+#include "armada_drm.h"
+#include "armada_hw.h"
+
+static int armada510_init(struct armada_private *priv, struct device *dev)
+{
+	priv->extclk[0] = devm_clk_get(dev, "ext_ref_clk_1");
+
+	if (IS_ERR(priv->extclk[0]) && PTR_ERR(priv->extclk[0]) == -ENOENT)
+		priv->extclk[0] = ERR_PTR(-EPROBE_DEFER);
+
+	return PTR_RET(priv->extclk[0]);
+}
+
+static int armada510_crtc_init(struct armada_crtc *dcrtc)
+{
+	/* Lower the watermark so to eliminate jitter at higher bandwidths */
+	armada_updatel(0x20, (1 << 11) | 0xff, dcrtc->base + LCD_CFG_RDREG4F);
+	return 0;
+}
+
+/*
+ * Armada510 specific SCLK register selection.
+ * This gets called with sclk = NULL to test whether the mode is
+ * supportable, and again with sclk != NULL to set the clocks up for
+ * that.  The former can return an error, but the latter is expected
+ * not to.
+ *
+ * We currently are pretty rudimentary here, always selecting
+ * EXT_REF_CLK_1 for LCD0 and erroring LCD1.  This needs improvement!
+ */
+static int armada510_crtc_compute_clock(struct armada_crtc *dcrtc,
+	const struct drm_display_mode *mode, uint32_t *sclk)
+{
+	struct armada_private *priv = dcrtc->crtc.dev->dev_private;
+	struct clk *clk = priv->extclk[0];
+	int ret;
+
+	if (dcrtc->num == 1)
+		return -EINVAL;
+
+	if (IS_ERR(clk))
+		return PTR_ERR(clk);
+
+	if (dcrtc->clk != clk) {
+		ret = clk_prepare_enable(clk);
+		if (ret)
+			return ret;
+		dcrtc->clk = clk;
+	}
+
+	if (sclk) {
+		uint32_t rate, ref, div;
+
+		rate = mode->clock * 1000;
+		ref = clk_round_rate(clk, rate);
+		div = DIV_ROUND_UP(ref, rate);
+		if (div < 1)
+			div = 1;
+
+		clk_set_rate(clk, ref);
+		*sclk = div | SCLK_510_EXTCLK1;
+	}
+
+	return 0;
+}
+
+const struct armada_variant armada510_ops = {
+	.has_spu_adv_reg = true,
+	.spu_adv_reg = ADV_HWC32ENABLE | ADV_HWC32ARGB | ADV_HWC32BLEND,
+	.init = armada510_init,
+	.crtc_init = armada510_crtc_init,
+	.crtc_compute_clock = armada510_crtc_compute_clock,
+};
diff --git a/drivers/gpu/drm/armada/armada_crtc.c b/drivers/gpu/drm/armada/armada_crtc.c
new file mode 100644
index 000000000000..d8e398275ca8
--- /dev/null
+++ b/drivers/gpu/drm/armada/armada_crtc.c
@@ -0,0 +1,1098 @@
+/*
+ * Copyright (C) 2012 Russell King
+ *  Rewritten from the dovefb driver, and Armada510 manuals.
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ */
+#include <linux/clk.h>
+#include <drm/drmP.h>
+#include <drm/drm_crtc_helper.h>
+#include "armada_crtc.h"
+#include "armada_drm.h"
+#include "armada_fb.h"
+#include "armada_gem.h"
+#include "armada_hw.h"
+
+struct armada_frame_work {
+	struct drm_pending_vblank_event *event;
+	struct armada_regs regs[4];
+	struct drm_framebuffer *old_fb;
+};
+
+enum csc_mode {
+	CSC_AUTO = 0,
+	CSC_YUV_CCIR601 = 1,
+	CSC_YUV_CCIR709 = 2,
+	CSC_RGB_COMPUTER = 1,
+	CSC_RGB_STUDIO = 2,
+};
+
+/*
+ * A note about interlacing.  Let's consider HDMI 1920x1080i.
+ * The timing parameters we have from X are:
+ *  Hact HsyA HsyI Htot  Vact VsyA VsyI Vtot
+ *  1920 2448 2492 2640  1080 1084 1094 1125
+ * Which get translated to:
+ *  Hact HsyA HsyI Htot  Vact VsyA VsyI Vtot
+ *  1920 2448 2492 2640   540  542  547  562
+ *
+ * This is how it is defined by CEA-861-D - line and pixel numbers are
+ * referenced to the rising edge of VSYNC and HSYNC.  Total clocks per
+ * line: 2640.  The odd frame, the first active line is at line 21, and
+ * the even frame, the first active line is 584.
+ *
+ * LN:    560     561     562     563             567     568    569
+ * DE:    ~~~|____________________________//__________________________
+ * HSYNC: ____|~|_____|~|_____|~|_____|~|_//__|~|_____|~|_____|~|_____
+ * VSYNC: _________________________|~~~~~~//~~~~~~~~~~~~~~~|__________
+ *  22 blanking lines.  VSYNC at 1320 (referenced to the HSYNC rising edge).
+ *
+ * LN:    1123   1124    1125      1               5       6      7
+ * DE:    ~~~|____________________________//__________________________
+ * HSYNC: ____|~|_____|~|_____|~|_____|~|_//__|~|_____|~|_____|~|_____
+ * VSYNC: ____________________|~~~~~~~~~~~//~~~~~~~~~~|_______________
+ *  23 blanking lines
+ *
+ * The Armada LCD Controller line and pixel numbers are, like X timings,
+ * referenced to the top left of the active frame.
+ *
+ * So, translating these to our LCD controller:
+ *  Odd frame, 563 total lines, VSYNC at line 543-548, pixel 1128.
+ *  Even frame, 562 total lines, VSYNC at line 542-547, pixel 2448.
+ * Note: Vsync front porch remains constant!
+ *
+ * if (odd_frame) {
+ *   vtotal = mode->crtc_vtotal + 1;
+ *   vbackporch = mode->crtc_vsync_start - mode->crtc_vdisplay + 1;
+ *   vhorizpos = mode->crtc_hsync_start - mode->crtc_htotal / 2
+ * } else {
+ *   vtotal = mode->crtc_vtotal;
+ *   vbackporch = mode->crtc_vsync_start - mode->crtc_vdisplay;
+ *   vhorizpos = mode->crtc_hsync_start;
+ * }
+ * vfrontporch = mode->crtc_vtotal - mode->crtc_vsync_end;
+ *
+ * So, we need to reprogram these registers on each vsync event:
+ *  LCD_SPU_V_PORCH, LCD_SPU_ADV_REG, LCD_SPUT_V_H_TOTAL
+ *
+ * Note: we do not use the frame done interrupts because these appear
+ * to happen too early, and lead to jitter on the display (presumably
+ * they occur at the end of the last active line, before the vsync back
+ * porch, which we're reprogramming.)
+ */
+
+void
+armada_drm_crtc_update_regs(struct armada_crtc *dcrtc, struct armada_regs *regs)
+{
+	while (regs->offset != ~0) {
+		void __iomem *reg = dcrtc->base + regs->offset;
+		uint32_t val;
+
+		val = regs->mask;
+		if (val != 0)
+			val &= readl_relaxed(reg);
+		writel_relaxed(val | regs->val, reg);
+		++regs;
+	}
+}
+
+#define dpms_blanked(dpms)	((dpms) != DRM_MODE_DPMS_ON)
+
+static void armada_drm_crtc_update(struct armada_crtc *dcrtc)
+{
+	uint32_t dumb_ctrl;
+
+	dumb_ctrl = dcrtc->cfg_dumb_ctrl;
+
+	if (!dpms_blanked(dcrtc->dpms))
+		dumb_ctrl |= CFG_DUMB_ENA;
+
+	/*
+	 * When the dumb interface isn't in DUMB24_RGB888_0 mode, it might
+	 * be using SPI or GPIO.  If we set this to DUMB_BLANK, we will
+	 * force LCD_D[23:0] to output blank color, overriding the GPIO or
+	 * SPI usage.  So leave it as-is unless in DUMB24_RGB888_0 mode.
+	 */
+	if (dpms_blanked(dcrtc->dpms) &&
+	    (dumb_ctrl & DUMB_MASK) == DUMB24_RGB888_0) {
+		dumb_ctrl &= ~DUMB_MASK;
+		dumb_ctrl |= DUMB_BLANK;
+	}
+
+	/*
+	 * The documentation doesn't indicate what the normal state of
+	 * the sync signals are.  Sebastian Hesselbart kindly probed
+	 * these signals on his board to determine their state.
+	 *
+	 * The non-inverted state of the sync signals is active high.
+	 * Setting these bits makes the appropriate signal active low.
+	 */
+	if (dcrtc->crtc.mode.flags & DRM_MODE_FLAG_NCSYNC)
+		dumb_ctrl |= CFG_INV_CSYNC;
+	if (dcrtc->crtc.mode.flags & DRM_MODE_FLAG_NHSYNC)
+		dumb_ctrl |= CFG_INV_HSYNC;
+	if (dcrtc->crtc.mode.flags & DRM_MODE_FLAG_NVSYNC)
+		dumb_ctrl |= CFG_INV_VSYNC;
+
+	if (dcrtc->dumb_ctrl != dumb_ctrl) {
+		dcrtc->dumb_ctrl = dumb_ctrl;
+		writel_relaxed(dumb_ctrl, dcrtc->base + LCD_SPU_DUMB_CTRL);
+	}
+}
+
+static unsigned armada_drm_crtc_calc_fb(struct drm_framebuffer *fb,
+	int x, int y, struct armada_regs *regs, bool interlaced)
+{
+	struct armada_gem_object *obj = drm_fb_obj(fb);
+	unsigned pitch = fb->pitches[0];
+	unsigned offset = y * pitch + x * fb->bits_per_pixel / 8;
+	uint32_t addr_odd, addr_even;
+	unsigned i = 0;
+
+	DRM_DEBUG_DRIVER("pitch %u x %d y %d bpp %d\n",
+		pitch, x, y, fb->bits_per_pixel);
+
+	addr_odd = addr_even = obj->dev_addr + offset;
+
+	if (interlaced) {
+		addr_even += pitch;
+		pitch *= 2;
+	}
+
+	/* write offset, base, and pitch */
+	armada_reg_queue_set(regs, i, addr_odd, LCD_CFG_GRA_START_ADDR0);
+	armada_reg_queue_set(regs, i, addr_even, LCD_CFG_GRA_START_ADDR1);
+	armada_reg_queue_mod(regs, i, pitch, 0xffff, LCD_CFG_GRA_PITCH);
+
+	return i;
+}
+
+static int armada_drm_crtc_queue_frame_work(struct armada_crtc *dcrtc,
+	struct armada_frame_work *work)
+{
+	struct drm_device *dev = dcrtc->crtc.dev;
+	unsigned long flags;
+	int ret;
+
+	ret = drm_vblank_get(dev, dcrtc->num);
+	if (ret) {
+		DRM_ERROR("failed to acquire vblank counter\n");
+		return ret;
+	}
+
+	spin_lock_irqsave(&dev->event_lock, flags);
+	if (!dcrtc->frame_work)
+		dcrtc->frame_work = work;
+	else
+		ret = -EBUSY;
+	spin_unlock_irqrestore(&dev->event_lock, flags);
+
+	if (ret)
+		drm_vblank_put(dev, dcrtc->num);
+
+	return ret;
+}
+
+static void armada_drm_crtc_complete_frame_work(struct armada_crtc *dcrtc)
+{
+	struct drm_device *dev = dcrtc->crtc.dev;
+	struct armada_frame_work *work = dcrtc->frame_work;
+
+	dcrtc->frame_work = NULL;
+
+	armada_drm_crtc_update_regs(dcrtc, work->regs);
+
+	if (work->event)
+		drm_send_vblank_event(dev, dcrtc->num, work->event);
+
+	drm_vblank_put(dev, dcrtc->num);
+
+	/* Finally, queue the process-half of the cleanup. */
+	__armada_drm_queue_unref_work(dcrtc->crtc.dev, work->old_fb);
+	kfree(work);
+}
+
+static void armada_drm_crtc_finish_fb(struct armada_crtc *dcrtc,
+	struct drm_framebuffer *fb, bool force)
+{
+	struct armada_frame_work *work;
+
+	if (!fb)
+		return;
+
+	if (force) {
+		/* Display is disabled, so just drop the old fb */
+		drm_framebuffer_unreference(fb);
+		return;
+	}
+
+	work = kmalloc(sizeof(*work), GFP_KERNEL);
+	if (work) {
+		int i = 0;
+		work->event = NULL;
+		work->old_fb = fb;
+		armada_reg_queue_end(work->regs, i);
+
+		if (armada_drm_crtc_queue_frame_work(dcrtc, work) == 0)
+			return;
+
+		kfree(work);
+	}
+
+	/*
+	 * Oops - just drop the reference immediately and hope for
+	 * the best.  The worst that will happen is the buffer gets
+	 * reused before it has finished being displayed.
+	 */
+	drm_framebuffer_unreference(fb);
+}
+
+static void armada_drm_vblank_off(struct armada_crtc *dcrtc)
+{
+	struct drm_device *dev = dcrtc->crtc.dev;
+
+	/*
+	 * Tell the DRM core that vblank IRQs aren't going to happen for
+	 * a while.  This cleans up any pending vblank events for us.
+	 */
+	drm_vblank_off(dev, dcrtc->num);
+
+	/* Handle any pending flip event. */
+	spin_lock_irq(&dev->event_lock);
+	if (dcrtc->frame_work)
+		armada_drm_crtc_complete_frame_work(dcrtc);
+	spin_unlock_irq(&dev->event_lock);
+}
+
+void armada_drm_crtc_gamma_set(struct drm_crtc *crtc, u16 r, u16 g, u16 b,
+	int idx)
+{
+}
+
+void armada_drm_crtc_gamma_get(struct drm_crtc *crtc, u16 *r, u16 *g, u16 *b,
+	int idx)
+{
+}
+
+/* The mode_config.mutex will be held for this call */
+static void armada_drm_crtc_dpms(struct drm_crtc *crtc, int dpms)
+{
+	struct armada_crtc *dcrtc = drm_to_armada_crtc(crtc);
+
+	if (dcrtc->dpms != dpms) {
+		dcrtc->dpms = dpms;
+		armada_drm_crtc_update(dcrtc);
+		if (dpms_blanked(dpms))
+			armada_drm_vblank_off(dcrtc);
+	}
+}
+
+/*
+ * Prepare for a mode set.  Turn off overlay to ensure that we don't end
+ * up with the overlay size being bigger than the active screen size.
+ * We rely upon X refreshing this state after the mode set has completed.
+ *
+ * The mode_config.mutex will be held for this call
+ */
+static void armada_drm_crtc_prepare(struct drm_crtc *crtc)
+{
+	struct armada_crtc *dcrtc = drm_to_armada_crtc(crtc);
+	struct drm_plane *plane;
+
+	/*
+	 * If we have an overlay plane associated with this CRTC, disable
+	 * it before the modeset to avoid its coordinates being outside
+	 * the new mode parameters.  DRM doesn't provide help with this.
+	 */
+	plane = dcrtc->plane;
+	if (plane) {
+		struct drm_framebuffer *fb = plane->fb;
+
+		plane->funcs->disable_plane(plane);
+		plane->fb = NULL;
+		plane->crtc = NULL;
+		drm_framebuffer_unreference(fb);
+	}
+}
+
+/* The mode_config.mutex will be held for this call */
+static void armada_drm_crtc_commit(struct drm_crtc *crtc)
+{
+	struct armada_crtc *dcrtc = drm_to_armada_crtc(crtc);
+
+	if (dcrtc->dpms != DRM_MODE_DPMS_ON) {
+		dcrtc->dpms = DRM_MODE_DPMS_ON;
+		armada_drm_crtc_update(dcrtc);
+	}
+}
+
+/* The mode_config.mutex will be held for this call */
+static bool armada_drm_crtc_mode_fixup(struct drm_crtc *crtc,
+	const struct drm_display_mode *mode, struct drm_display_mode *adj)
+{
+	struct armada_private *priv = crtc->dev->dev_private;
+	struct armada_crtc *dcrtc = drm_to_armada_crtc(crtc);
+	int ret;
+
+	/* We can't do interlaced modes if we don't have the SPU_ADV_REG */
+	if (!priv->variant->has_spu_adv_reg &&
+	    adj->flags & DRM_MODE_FLAG_INTERLACE)
+		return false;
+
+	/* Check whether the display mode is possible */
+	ret = priv->variant->crtc_compute_clock(dcrtc, adj, NULL);
+	if (ret)
+		return false;
+
+	return true;
+}
+
+void armada_drm_crtc_irq(struct armada_crtc *dcrtc, u32 stat)
+{
+	struct armada_vbl_event *e, *n;
+	void __iomem *base = dcrtc->base;
+
+	if (stat & DMA_FF_UNDERFLOW)
+		DRM_ERROR("video underflow on crtc %u\n", dcrtc->num);
+	if (stat & GRA_FF_UNDERFLOW)
+		DRM_ERROR("graphics underflow on crtc %u\n", dcrtc->num);
+
+	if (stat & VSYNC_IRQ)
+		drm_handle_vblank(dcrtc->crtc.dev, dcrtc->num);
+
+	spin_lock(&dcrtc->irq_lock);
+
+	list_for_each_entry_safe(e, n, &dcrtc->vbl_list, node) {
+		list_del_init(&e->node);
+		drm_vblank_put(dcrtc->crtc.dev, dcrtc->num);
+		e->fn(dcrtc, e->data);
+	}
+
+	if (stat & GRA_FRAME_IRQ && dcrtc->interlaced) {
+		int i = stat & GRA_FRAME_IRQ0 ? 0 : 1;
+		uint32_t val;
+
+		writel_relaxed(dcrtc->v[i].spu_v_porch, base + LCD_SPU_V_PORCH);
+		writel_relaxed(dcrtc->v[i].spu_v_h_total,
+			       base + LCD_SPUT_V_H_TOTAL);
+
+		val = readl_relaxed(base + LCD_SPU_ADV_REG);
+		val &= ~(ADV_VSYNC_L_OFF | ADV_VSYNC_H_OFF | ADV_VSYNCOFFEN);
+		val |= dcrtc->v[i].spu_adv_reg;
+		writel_relaxed(val, base + LCD_SPU_ADV_REG);
+	}
+
+	if (stat & DUMB_FRAMEDONE && dcrtc->cursor_update) {
+		writel_relaxed(dcrtc->cursor_hw_pos,
+			       base + LCD_SPU_HWC_OVSA_HPXL_VLN);
+		writel_relaxed(dcrtc->cursor_hw_sz,
+			       base + LCD_SPU_HWC_HPXL_VLN);
+		armada_updatel(CFG_HWC_ENA,
+			       CFG_HWC_ENA | CFG_HWC_1BITMOD | CFG_HWC_1BITENA,
+			       base + LCD_SPU_DMA_CTRL0);
+		dcrtc->cursor_update = false;
+		armada_drm_crtc_disable_irq(dcrtc, DUMB_FRAMEDONE_ENA);
+	}
+
+	spin_unlock(&dcrtc->irq_lock);
+
+	if (stat & GRA_FRAME_IRQ) {
+		struct drm_device *dev = dcrtc->crtc.dev;
+
+		spin_lock(&dev->event_lock);
+		if (dcrtc->frame_work)
+			armada_drm_crtc_complete_frame_work(dcrtc);
+		spin_unlock(&dev->event_lock);
+
+		wake_up(&dcrtc->frame_wait);
+	}
+}
+
+/* These are locked by dev->vbl_lock */
+void armada_drm_crtc_disable_irq(struct armada_crtc *dcrtc, u32 mask)
+{
+	if (dcrtc->irq_ena & mask) {
+		dcrtc->irq_ena &= ~mask;
+		writel(dcrtc->irq_ena, dcrtc->base + LCD_SPU_IRQ_ENA);
+	}
+}
+
+void armada_drm_crtc_enable_irq(struct armada_crtc *dcrtc, u32 mask)
+{
+	if ((dcrtc->irq_ena & mask) != mask) {
+		dcrtc->irq_ena |= mask;
+		writel(dcrtc->irq_ena, dcrtc->base + LCD_SPU_IRQ_ENA);
+		if (readl_relaxed(dcrtc->base + LCD_SPU_IRQ_ISR) & mask)
+			writel(0, dcrtc->base + LCD_SPU_IRQ_ISR);
+	}
+}
+
+static uint32_t armada_drm_crtc_calculate_csc(struct armada_crtc *dcrtc)
+{
+	struct drm_display_mode *adj = &dcrtc->crtc.mode;
+	uint32_t val = 0;
+
+	if (dcrtc->csc_yuv_mode == CSC_YUV_CCIR709)
+		val |= CFG_CSC_YUV_CCIR709;
+	if (dcrtc->csc_rgb_mode == CSC_RGB_STUDIO)
+		val |= CFG_CSC_RGB_STUDIO;
+
+	/*
+	 * In auto mode, set the colorimetry, based upon the HDMI spec.
+	 * 1280x720p, 1920x1080p and 1920x1080i use ITU709, others use
+	 * ITU601.  It may be more appropriate to set this depending on
+	 * the source - but what if the graphic frame is YUV and the
+	 * video frame is RGB?
+	 */
+	if ((adj->hdisplay == 1280 && adj->vdisplay == 720 &&
+	     !(adj->flags & DRM_MODE_FLAG_INTERLACE)) ||
+	    (adj->hdisplay == 1920 && adj->vdisplay == 1080)) {
+		if (dcrtc->csc_yuv_mode == CSC_AUTO)
+			val |= CFG_CSC_YUV_CCIR709;
+	}
+
+	/*
+	 * We assume we're connected to a TV-like device, so the YUV->RGB
+	 * conversion should produce a limited range.  We should set this
+	 * depending on the connectors attached to this CRTC, and what
+	 * kind of device they report being connected.
+	 */
+	if (dcrtc->csc_rgb_mode == CSC_AUTO)
+		val |= CFG_CSC_RGB_STUDIO;
+
+	return val;
+}
+
+/* The mode_config.mutex will be held for this call */
+static int armada_drm_crtc_mode_set(struct drm_crtc *crtc,
+	struct drm_display_mode *mode, struct drm_display_mode *adj,
+	int x, int y, struct drm_framebuffer *old_fb)
+{
+	struct armada_private *priv = crtc->dev->dev_private;
+	struct armada_crtc *dcrtc = drm_to_armada_crtc(crtc);
+	struct armada_regs regs[17];
+	uint32_t lm, rm, tm, bm, val, sclk;
+	unsigned long flags;
+	unsigned i;
+	bool interlaced;
+
+	drm_framebuffer_reference(crtc->fb);
+
+	interlaced = !!(adj->flags & DRM_MODE_FLAG_INTERLACE);
+
+	i = armada_drm_crtc_calc_fb(dcrtc->crtc.fb, x, y, regs, interlaced);
+
+	rm = adj->crtc_hsync_start - adj->crtc_hdisplay;
+	lm = adj->crtc_htotal - adj->crtc_hsync_end;
+	bm = adj->crtc_vsync_start - adj->crtc_vdisplay;
+	tm = adj->crtc_vtotal - adj->crtc_vsync_end;
+
+	DRM_DEBUG_DRIVER("H: %d %d %d %d lm %d rm %d\n",
+		adj->crtc_hdisplay,
+		adj->crtc_hsync_start,
+		adj->crtc_hsync_end,
+		adj->crtc_htotal, lm, rm);
+	DRM_DEBUG_DRIVER("V: %d %d %d %d tm %d bm %d\n",
+		adj->crtc_vdisplay,
+		adj->crtc_vsync_start,
+		adj->crtc_vsync_end,
+		adj->crtc_vtotal, tm, bm);
+
+	/* Wait for pending flips to complete */
+	wait_event(dcrtc->frame_wait, !dcrtc->frame_work);
+
+	drm_vblank_pre_modeset(crtc->dev, dcrtc->num);
+
+	crtc->mode = *adj;
+
+	val = dcrtc->dumb_ctrl & ~CFG_DUMB_ENA;
+	if (val != dcrtc->dumb_ctrl) {
+		dcrtc->dumb_ctrl = val;
+		writel_relaxed(val, dcrtc->base + LCD_SPU_DUMB_CTRL);
+	}
+
+	/* Now compute the divider for real */
+	priv->variant->crtc_compute_clock(dcrtc, adj, &sclk);
+
+	/* Ensure graphic fifo is enabled */
+	armada_reg_queue_mod(regs, i, 0, CFG_PDWN64x66, LCD_SPU_SRAM_PARA1);
+	armada_reg_queue_set(regs, i, sclk, LCD_CFG_SCLK_DIV);
+
+	if (interlaced ^ dcrtc->interlaced) {
+		if (adj->flags & DRM_MODE_FLAG_INTERLACE)
+			drm_vblank_get(dcrtc->crtc.dev, dcrtc->num);
+		else
+			drm_vblank_put(dcrtc->crtc.dev, dcrtc->num);
+		dcrtc->interlaced = interlaced;
+	}
+
+	spin_lock_irqsave(&dcrtc->irq_lock, flags);
+
+	/* Even interlaced/progressive frame */
+	dcrtc->v[1].spu_v_h_total = adj->crtc_vtotal << 16 |
+				    adj->crtc_htotal;
+	dcrtc->v[1].spu_v_porch = tm << 16 | bm;
+	val = adj->crtc_hsync_start;
+	dcrtc->v[1].spu_adv_reg = val << 20 | val | ADV_VSYNCOFFEN |
+		priv->variant->spu_adv_reg;
+
+	if (interlaced) {
+		/* Odd interlaced frame */
+		dcrtc->v[0].spu_v_h_total = dcrtc->v[1].spu_v_h_total +
+						(1 << 16);
+		dcrtc->v[0].spu_v_porch = dcrtc->v[1].spu_v_porch + 1;
+		val = adj->crtc_hsync_start - adj->crtc_htotal / 2;
+		dcrtc->v[0].spu_adv_reg = val << 20 | val | ADV_VSYNCOFFEN |
+			priv->variant->spu_adv_reg;
+	} else {
+		dcrtc->v[0] = dcrtc->v[1];
+	}
+
+	val = adj->crtc_vdisplay << 16 | adj->crtc_hdisplay;
+
+	armada_reg_queue_set(regs, i, val, LCD_SPU_V_H_ACTIVE);
+	armada_reg_queue_set(regs, i, val, LCD_SPU_GRA_HPXL_VLN);
+	armada_reg_queue_set(regs, i, val, LCD_SPU_GZM_HPXL_VLN);
+	armada_reg_queue_set(regs, i, (lm << 16) | rm, LCD_SPU_H_PORCH);
+	armada_reg_queue_set(regs, i, dcrtc->v[0].spu_v_porch, LCD_SPU_V_PORCH);
+	armada_reg_queue_set(regs, i, dcrtc->v[0].spu_v_h_total,
+			   LCD_SPUT_V_H_TOTAL);
+
+	if (priv->variant->has_spu_adv_reg) {
+		armada_reg_queue_mod(regs, i, dcrtc->v[0].spu_adv_reg,
+				     ADV_VSYNC_L_OFF | ADV_VSYNC_H_OFF |
+				     ADV_VSYNCOFFEN, LCD_SPU_ADV_REG);
+	}
+
+	val = CFG_GRA_ENA | CFG_GRA_HSMOOTH;
+	val |= CFG_GRA_FMT(drm_fb_to_armada_fb(dcrtc->crtc.fb)->fmt);
+	val |= CFG_GRA_MOD(drm_fb_to_armada_fb(dcrtc->crtc.fb)->mod);
+
+	if (drm_fb_to_armada_fb(dcrtc->crtc.fb)->fmt > CFG_420)
+		val |= CFG_PALETTE_ENA;
+
+	if (interlaced)
+		val |= CFG_GRA_FTOGGLE;
+
+	armada_reg_queue_mod(regs, i, val, CFG_GRAFORMAT |
+			     CFG_GRA_MOD(CFG_SWAPRB | CFG_SWAPUV |
+					 CFG_SWAPYU | CFG_YUV2RGB) |
+			     CFG_PALETTE_ENA | CFG_GRA_FTOGGLE,
+			     LCD_SPU_DMA_CTRL0);
+
+	val = adj->flags & DRM_MODE_FLAG_NVSYNC ? CFG_VSYNC_INV : 0;
+	armada_reg_queue_mod(regs, i, val, CFG_VSYNC_INV, LCD_SPU_DMA_CTRL1);
+
+	val = dcrtc->spu_iopad_ctrl | armada_drm_crtc_calculate_csc(dcrtc);
+	armada_reg_queue_set(regs, i, val, LCD_SPU_IOPAD_CONTROL);
+	armada_reg_queue_end(regs, i);
+
+	armada_drm_crtc_update_regs(dcrtc, regs);
+	spin_unlock_irqrestore(&dcrtc->irq_lock, flags);
+
+	armada_drm_crtc_update(dcrtc);
+
+	drm_vblank_post_modeset(crtc->dev, dcrtc->num);
+	armada_drm_crtc_finish_fb(dcrtc, old_fb, dpms_blanked(dcrtc->dpms));
+
+	return 0;
+}
+
+/* The mode_config.mutex will be held for this call */
+static int armada_drm_crtc_mode_set_base(struct drm_crtc *crtc, int x, int y,
+	struct drm_framebuffer *old_fb)
+{
+	struct armada_crtc *dcrtc = drm_to_armada_crtc(crtc);
+	struct armada_regs regs[4];
+	unsigned i;
+
+	i = armada_drm_crtc_calc_fb(crtc->fb, crtc->x, crtc->y, regs,
+				    dcrtc->interlaced);
+	armada_reg_queue_end(regs, i);
+
+	/* Wait for pending flips to complete */
+	wait_event(dcrtc->frame_wait, !dcrtc->frame_work);
+
+	/* Take a reference to the new fb as we're using it */
+	drm_framebuffer_reference(crtc->fb);
+
+	/* Update the base in the CRTC */
+	armada_drm_crtc_update_regs(dcrtc, regs);
+
+	/* Drop our previously held reference */
+	armada_drm_crtc_finish_fb(dcrtc, old_fb, dpms_blanked(dcrtc->dpms));
+
+	return 0;
+}
+
+static void armada_drm_crtc_load_lut(struct drm_crtc *crtc)
+{
+}
+
+/* The mode_config.mutex will be held for this call */
+static void armada_drm_crtc_disable(struct drm_crtc *crtc)
+{
+	struct armada_crtc *dcrtc = drm_to_armada_crtc(crtc);
+
+	armada_drm_crtc_dpms(crtc, DRM_MODE_DPMS_OFF);
+	armada_drm_crtc_finish_fb(dcrtc, crtc->fb, true);
+
+	/* Power down most RAMs and FIFOs */
+	writel_relaxed(CFG_PDWN256x32 | CFG_PDWN256x24 | CFG_PDWN256x8 |
+		       CFG_PDWN32x32 | CFG_PDWN16x66 | CFG_PDWN32x66 |
+		       CFG_PDWN64x66, dcrtc->base + LCD_SPU_SRAM_PARA1);
+}
+
+static const struct drm_crtc_helper_funcs armada_crtc_helper_funcs = {
+	.dpms		= armada_drm_crtc_dpms,
+	.prepare	= armada_drm_crtc_prepare,
+	.commit		= armada_drm_crtc_commit,
+	.mode_fixup	= armada_drm_crtc_mode_fixup,
+	.mode_set	= armada_drm_crtc_mode_set,
+	.mode_set_base	= armada_drm_crtc_mode_set_base,
+	.load_lut	= armada_drm_crtc_load_lut,
+	.disable	= armada_drm_crtc_disable,
+};
+
+static void armada_load_cursor_argb(void __iomem *base, uint32_t *pix,
+	unsigned stride, unsigned width, unsigned height)
+{
+	uint32_t addr;
+	unsigned y;
+
+	addr = SRAM_HWC32_RAM1;
+	for (y = 0; y < height; y++) {
+		uint32_t *p = &pix[y * stride];
+		unsigned x;
+
+		for (x = 0; x < width; x++, p++) {
+			uint32_t val = *p;
+
+			val = (val & 0xff00ff00) |
+			      (val & 0x000000ff) << 16 |
+			      (val & 0x00ff0000) >> 16;
+
+			writel_relaxed(val,
+				       base + LCD_SPU_SRAM_WRDAT);
+			writel_relaxed(addr | SRAM_WRITE,
+				       base + LCD_SPU_SRAM_CTRL);
+			addr += 1;
+			if ((addr & 0x00ff) == 0)
+				addr += 0xf00;
+			if ((addr & 0x30ff) == 0)
+				addr = SRAM_HWC32_RAM2;
+		}
+	}
+}
+
+static void armada_drm_crtc_cursor_tran(void __iomem *base)
+{
+	unsigned addr;
+
+	for (addr = 0; addr < 256; addr++) {
+		/* write the default value */
+		writel_relaxed(0x55555555, base + LCD_SPU_SRAM_WRDAT);
+		writel_relaxed(addr | SRAM_WRITE | SRAM_HWC32_TRAN,
+			       base + LCD_SPU_SRAM_CTRL);
+	}
+}
+
+static int armada_drm_crtc_cursor_update(struct armada_crtc *dcrtc, bool reload)
+{
+	uint32_t xoff, xscr, w = dcrtc->cursor_w, s;
+	uint32_t yoff, yscr, h = dcrtc->cursor_h;
+	uint32_t para1;
+
+	/*
+	 * Calculate the visible width and height of the cursor,
+	 * screen position, and the position in the cursor bitmap.
+	 */
+	if (dcrtc->cursor_x < 0) {
+		xoff = -dcrtc->cursor_x;
+		xscr = 0;
+		w -= min(xoff, w);
+	} else if (dcrtc->cursor_x + w > dcrtc->crtc.mode.hdisplay) {
+		xoff = 0;
+		xscr = dcrtc->cursor_x;
+		w = max_t(int, dcrtc->crtc.mode.hdisplay - dcrtc->cursor_x, 0);
+	} else {
+		xoff = 0;
+		xscr = dcrtc->cursor_x;
+	}
+
+	if (dcrtc->cursor_y < 0) {
+		yoff = -dcrtc->cursor_y;
+		yscr = 0;
+		h -= min(yoff, h);
+	} else if (dcrtc->cursor_y + h > dcrtc->crtc.mode.vdisplay) {
+		yoff = 0;
+		yscr = dcrtc->cursor_y;
+		h = max_t(int, dcrtc->crtc.mode.vdisplay - dcrtc->cursor_y, 0);
+	} else {
+		yoff = 0;
+		yscr = dcrtc->cursor_y;
+	}
+
+	/* On interlaced modes, the vertical cursor size must be halved */
+	s = dcrtc->cursor_w;
+	if (dcrtc->interlaced) {
+		s *= 2;
+		yscr /= 2;
+		h /= 2;
+	}
+
+	if (!dcrtc->cursor_obj || !h || !w) {
+		spin_lock_irq(&dcrtc->irq_lock);
+		armada_drm_crtc_disable_irq(dcrtc, DUMB_FRAMEDONE_ENA);
+		dcrtc->cursor_update = false;
+		armada_updatel(0, CFG_HWC_ENA, dcrtc->base + LCD_SPU_DMA_CTRL0);
+		spin_unlock_irq(&dcrtc->irq_lock);
+		return 0;
+	}
+
+	para1 = readl_relaxed(dcrtc->base + LCD_SPU_SRAM_PARA1);
+	armada_updatel(CFG_CSB_256x32, CFG_CSB_256x32 | CFG_PDWN256x32,
+		       dcrtc->base + LCD_SPU_SRAM_PARA1);
+
+	/*
+	 * Initialize the transparency if the SRAM was powered down.
+	 * We must also reload the cursor data as well.
+	 */
+	if (!(para1 & CFG_CSB_256x32)) {
+		armada_drm_crtc_cursor_tran(dcrtc->base);
+		reload = true;
+	}
+
+	if (dcrtc->cursor_hw_sz != (h << 16 | w)) {
+		spin_lock_irq(&dcrtc->irq_lock);
+		armada_drm_crtc_disable_irq(dcrtc, DUMB_FRAMEDONE_ENA);
+		dcrtc->cursor_update = false;
+		armada_updatel(0, CFG_HWC_ENA, dcrtc->base + LCD_SPU_DMA_CTRL0);
+		spin_unlock_irq(&dcrtc->irq_lock);
+		reload = true;
+	}
+	if (reload) {
+		struct armada_gem_object *obj = dcrtc->cursor_obj;
+		uint32_t *pix;
+		/* Set the top-left corner of the cursor image */
+		pix = obj->addr;
+		pix += yoff * s + xoff;
+		armada_load_cursor_argb(dcrtc->base, pix, s, w, h);
+	}
+
+	/* Reload the cursor position, size and enable in the IRQ handler */
+	spin_lock_irq(&dcrtc->irq_lock);
+	dcrtc->cursor_hw_pos = yscr << 16 | xscr;
+	dcrtc->cursor_hw_sz = h << 16 | w;
+	dcrtc->cursor_update = true;
+	armada_drm_crtc_enable_irq(dcrtc, DUMB_FRAMEDONE_ENA);
+	spin_unlock_irq(&dcrtc->irq_lock);
+
+	return 0;
+}
+
+static void cursor_update(void *data)
+{
+	armada_drm_crtc_cursor_update(data, true);
+}
+
+static int armada_drm_crtc_cursor_set(struct drm_crtc *crtc,
+	struct drm_file *file, uint32_t handle, uint32_t w, uint32_t h)
+{
+	struct drm_device *dev = crtc->dev;
+	struct armada_crtc *dcrtc = drm_to_armada_crtc(crtc);
+	struct armada_private *priv = crtc->dev->dev_private;
+	struct armada_gem_object *obj = NULL;
+	int ret;
+
+	/* If no cursor support, replicate drm's return value */
+	if (!priv->variant->has_spu_adv_reg)
+		return -ENXIO;
+
+	if (handle && w > 0 && h > 0) {
+		/* maximum size is 64x32 or 32x64 */
+		if (w > 64 || h > 64 || (w > 32 && h > 32))
+			return -ENOMEM;
+
+		obj = armada_gem_object_lookup(dev, file, handle);
+		if (!obj)
+			return -ENOENT;
+
+		/* Must be a kernel-mapped object */
+		if (!obj->addr) {
+			drm_gem_object_unreference_unlocked(&obj->obj);
+			return -EINVAL;
+		}
+
+		if (obj->obj.size < w * h * 4) {
+			DRM_ERROR("buffer is too small\n");
+			drm_gem_object_unreference_unlocked(&obj->obj);
+			return -ENOMEM;
+		}
+	}
+
+	mutex_lock(&dev->struct_mutex);
+	if (dcrtc->cursor_obj) {
+		dcrtc->cursor_obj->update = NULL;
+		dcrtc->cursor_obj->update_data = NULL;
+		drm_gem_object_unreference(&dcrtc->cursor_obj->obj);
+	}
+	dcrtc->cursor_obj = obj;
+	dcrtc->cursor_w = w;
+	dcrtc->cursor_h = h;
+	ret = armada_drm_crtc_cursor_update(dcrtc, true);
+	if (obj) {
+		obj->update_data = dcrtc;
+		obj->update = cursor_update;
+	}
+	mutex_unlock(&dev->struct_mutex);
+
+	return ret;
+}
+
+static int armada_drm_crtc_cursor_move(struct drm_crtc *crtc, int x, int y)
+{
+	struct drm_device *dev = crtc->dev;
+	struct armada_crtc *dcrtc = drm_to_armada_crtc(crtc);
+	struct armada_private *priv = crtc->dev->dev_private;
+	int ret;
+
+	/* If no cursor support, replicate drm's return value */
+	if (!priv->variant->has_spu_adv_reg)
+		return -EFAULT;
+
+	mutex_lock(&dev->struct_mutex);
+	dcrtc->cursor_x = x;
+	dcrtc->cursor_y = y;
+	ret = armada_drm_crtc_cursor_update(dcrtc, false);
+	mutex_unlock(&dev->struct_mutex);
+
+	return ret;
+}
+
+static void armada_drm_crtc_destroy(struct drm_crtc *crtc)
+{
+	struct armada_crtc *dcrtc = drm_to_armada_crtc(crtc);
+	struct armada_private *priv = crtc->dev->dev_private;
+
+	if (dcrtc->cursor_obj)
+		drm_gem_object_unreference(&dcrtc->cursor_obj->obj);
+
+	priv->dcrtc[dcrtc->num] = NULL;
+	drm_crtc_cleanup(&dcrtc->crtc);
+
+	if (!IS_ERR(dcrtc->clk))
+		clk_disable_unprepare(dcrtc->clk);
+
+	kfree(dcrtc);
+}
+
+/*
+ * The mode_config lock is held here, to prevent races between this
+ * and a mode_set.
+ */
+static int armada_drm_crtc_page_flip(struct drm_crtc *crtc,
+	struct drm_framebuffer *fb, struct drm_pending_vblank_event *event, uint32_t page_flip_flags)
+{
+	struct armada_crtc *dcrtc = drm_to_armada_crtc(crtc);
+	struct armada_frame_work *work;
+	struct drm_device *dev = crtc->dev;
+	unsigned long flags;
+	unsigned i;
+	int ret;
+
+	/* We don't support changing the pixel format */
+	if (fb->pixel_format != crtc->fb->pixel_format)
+		return -EINVAL;
+
+	work = kmalloc(sizeof(*work), GFP_KERNEL);
+	if (!work)
+		return -ENOMEM;
+
+	work->event = event;
+	work->old_fb = dcrtc->crtc.fb;
+
+	i = armada_drm_crtc_calc_fb(fb, crtc->x, crtc->y, work->regs,
+				    dcrtc->interlaced);
+	armada_reg_queue_end(work->regs, i);
+
+	/*
+	 * Hold the old framebuffer for the work - DRM appears to drop our
+	 * reference to the old framebuffer in drm_mode_page_flip_ioctl().
+	 */
+	drm_framebuffer_reference(work->old_fb);
+
+	ret = armada_drm_crtc_queue_frame_work(dcrtc, work);
+	if (ret) {
+		/*
+		 * Undo our reference above; DRM does not drop the reference
+		 * to this object on error, so that's okay.
+		 */
+		drm_framebuffer_unreference(work->old_fb);
+		kfree(work);
+		return ret;
+	}
+
+	/*
+	 * Don't take a reference on the new framebuffer;
+	 * drm_mode_page_flip_ioctl() has already grabbed a reference and
+	 * will _not_ drop that reference on successful return from this
+	 * function.  Simply mark this new framebuffer as the current one.
+	 */
+	dcrtc->crtc.fb = fb;
+
+	/*
+	 * Finally, if the display is blanked, we won't receive an
+	 * interrupt, so complete it now.
+	 */
+	if (dpms_blanked(dcrtc->dpms)) {
+		spin_lock_irqsave(&dev->event_lock, flags);
+		if (dcrtc->frame_work)
+			armada_drm_crtc_complete_frame_work(dcrtc);
+		spin_unlock_irqrestore(&dev->event_lock, flags);
+	}
+
+	return 0;
+}
+
+static int
+armada_drm_crtc_set_property(struct drm_crtc *crtc,
+	struct drm_property *property, uint64_t val)
+{
+	struct armada_private *priv = crtc->dev->dev_private;
+	struct armada_crtc *dcrtc = drm_to_armada_crtc(crtc);
+	bool update_csc = false;
+
+	if (property == priv->csc_yuv_prop) {
+		dcrtc->csc_yuv_mode = val;
+		update_csc = true;
+	} else if (property == priv->csc_rgb_prop) {
+		dcrtc->csc_rgb_mode = val;
+		update_csc = true;
+	}
+
+	if (update_csc) {
+		uint32_t val;
+
+		val = dcrtc->spu_iopad_ctrl |
+		      armada_drm_crtc_calculate_csc(dcrtc);
+		writel_relaxed(val, dcrtc->base + LCD_SPU_IOPAD_CONTROL);
+	}
+
+	return 0;
+}
+
+static struct drm_crtc_funcs armada_crtc_funcs = {
+	.cursor_set	= armada_drm_crtc_cursor_set,
+	.cursor_move	= armada_drm_crtc_cursor_move,
+	.destroy	= armada_drm_crtc_destroy,
+	.set_config	= drm_crtc_helper_set_config,
+	.page_flip	= armada_drm_crtc_page_flip,
+	.set_property	= armada_drm_crtc_set_property,
+};
+
+static struct drm_prop_enum_list armada_drm_csc_yuv_enum_list[] = {
+	{ CSC_AUTO,        "Auto" },
+	{ CSC_YUV_CCIR601, "CCIR601" },
+	{ CSC_YUV_CCIR709, "CCIR709" },
+};
+
+static struct drm_prop_enum_list armada_drm_csc_rgb_enum_list[] = {
+	{ CSC_AUTO,         "Auto" },
+	{ CSC_RGB_COMPUTER, "Computer system" },
+	{ CSC_RGB_STUDIO,   "Studio" },
+};
+
+static int armada_drm_crtc_create_properties(struct drm_device *dev)
+{
+	struct armada_private *priv = dev->dev_private;
+
+	if (priv->csc_yuv_prop)
+		return 0;
+
+	priv->csc_yuv_prop = drm_property_create_enum(dev, 0,
+				"CSC_YUV", armada_drm_csc_yuv_enum_list,
+				ARRAY_SIZE(armada_drm_csc_yuv_enum_list));
+	priv->csc_rgb_prop = drm_property_create_enum(dev, 0,
+				"CSC_RGB", armada_drm_csc_rgb_enum_list,
+				ARRAY_SIZE(armada_drm_csc_rgb_enum_list));
+
+	if (!priv->csc_yuv_prop || !priv->csc_rgb_prop)
+		return -ENOMEM;
+
+	return 0;
+}
+
+int armada_drm_crtc_create(struct drm_device *dev, unsigned num,
+	struct resource *res)
+{
+	struct armada_private *priv = dev->dev_private;
+	struct armada_crtc *dcrtc;
+	void __iomem *base;
+	int ret;
+
+	ret = armada_drm_crtc_create_properties(dev);
+	if (ret)
+		return ret;
+
+	base = devm_request_and_ioremap(dev->dev, res);
+	if (!base) {
+		DRM_ERROR("failed to ioremap register\n");
+		return -ENOMEM;
+	}
+
+	dcrtc = kzalloc(sizeof(*dcrtc), GFP_KERNEL);
+	if (!dcrtc) {
+		DRM_ERROR("failed to allocate Armada crtc\n");
+		return -ENOMEM;
+	}
+
+	dcrtc->base = base;
+	dcrtc->num = num;
+	dcrtc->clk = ERR_PTR(-EINVAL);
+	dcrtc->csc_yuv_mode = CSC_AUTO;
+	dcrtc->csc_rgb_mode = CSC_AUTO;
+	dcrtc->cfg_dumb_ctrl = DUMB24_RGB888_0;
+	dcrtc->spu_iopad_ctrl = CFG_VSCALE_LN_EN | CFG_IOPAD_DUMB24;
+	spin_lock_init(&dcrtc->irq_lock);
+	dcrtc->irq_ena = CLEAN_SPU_IRQ_ISR;
+	INIT_LIST_HEAD(&dcrtc->vbl_list);
+	init_waitqueue_head(&dcrtc->frame_wait);
+
+	/* Initialize some registers which we don't otherwise set */
+	writel_relaxed(0x00000001, dcrtc->base + LCD_CFG_SCLK_DIV);
+	writel_relaxed(0x00000000, dcrtc->base + LCD_SPU_BLANKCOLOR);
+	writel_relaxed(dcrtc->spu_iopad_ctrl,
+		       dcrtc->base + LCD_SPU_IOPAD_CONTROL);
+	writel_relaxed(0x00000000, dcrtc->base + LCD_SPU_SRAM_PARA0);
+	writel_relaxed(CFG_PDWN256x32 | CFG_PDWN256x24 | CFG_PDWN256x8 |
+		       CFG_PDWN32x32 | CFG_PDWN16x66 | CFG_PDWN32x66 |
+		       CFG_PDWN64x66, dcrtc->base + LCD_SPU_SRAM_PARA1);
+	writel_relaxed(0x2032ff81, dcrtc->base + LCD_SPU_DMA_CTRL1);
+	writel_relaxed(0x00000000, dcrtc->base + LCD_SPU_GRA_OVSA_HPXL_VLN);
+
+	if (priv->variant->crtc_init) {
+		ret = priv->variant->crtc_init(dcrtc);
+		if (ret) {
+			kfree(dcrtc);
+			return ret;
+		}
+	}
+
+	/* Ensure AXI pipeline is enabled */
+	armada_updatel(CFG_ARBFAST_ENA, 0, dcrtc->base + LCD_SPU_DMA_CTRL0);
+
+	priv->dcrtc[dcrtc->num] = dcrtc;
+
+	drm_crtc_init(dev, &dcrtc->crtc, &armada_crtc_funcs);
+	drm_crtc_helper_add(&dcrtc->crtc, &armada_crtc_helper_funcs);
+
+	drm_object_attach_property(&dcrtc->crtc.base, priv->csc_yuv_prop,
+				   dcrtc->csc_yuv_mode);
+	drm_object_attach_property(&dcrtc->crtc.base, priv->csc_rgb_prop,
+				   dcrtc->csc_rgb_mode);
+
+	return armada_overlay_plane_create(dev, 1 << dcrtc->num);
+}
diff --git a/drivers/gpu/drm/armada/armada_crtc.h b/drivers/gpu/drm/armada/armada_crtc.h
new file mode 100644
index 000000000000..9c10a07e7492
--- /dev/null
+++ b/drivers/gpu/drm/armada/armada_crtc.h
@@ -0,0 +1,83 @@
+/*
+ * Copyright (C) 2012 Russell King
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ */
+#ifndef ARMADA_CRTC_H
+#define ARMADA_CRTC_H
+
+struct armada_gem_object;
+
+struct armada_regs {
+	uint32_t offset;
+	uint32_t mask;
+	uint32_t val;
+};
+
+#define armada_reg_queue_mod(_r, _i, _v, _m, _o)	\
+	do {					\
+		struct armada_regs *__reg = _r;	\
+		__reg[_i].offset = _o;		\
+		__reg[_i].mask = ~(_m);		\
+		__reg[_i].val = _v;		\
+		_i++;				\
+	} while (0)
+
+#define armada_reg_queue_set(_r, _i, _v, _o)	\
+	armada_reg_queue_mod(_r, _i, _v, ~0, _o)
+
+#define armada_reg_queue_end(_r, _i)		\
+	armada_reg_queue_mod(_r, _i, 0, 0, ~0)
+
+struct armada_frame_work;
+
+struct armada_crtc {
+	struct drm_crtc		crtc;
+	unsigned		num;
+	void __iomem		*base;
+	struct clk		*clk;
+	struct {
+		uint32_t	spu_v_h_total;
+		uint32_t	spu_v_porch;
+		uint32_t	spu_adv_reg;
+	} v[2];
+	bool			interlaced;
+	bool			cursor_update;
+	uint8_t			csc_yuv_mode;
+	uint8_t			csc_rgb_mode;
+
+	struct drm_plane	*plane;
+
+	struct armada_gem_object	*cursor_obj;
+	int			cursor_x;
+	int			cursor_y;
+	uint32_t		cursor_hw_pos;
+	uint32_t		cursor_hw_sz;
+	uint32_t		cursor_w;
+	uint32_t		cursor_h;
+
+	int			dpms;
+	uint32_t		cfg_dumb_ctrl;
+	uint32_t		dumb_ctrl;
+	uint32_t		spu_iopad_ctrl;
+
+	wait_queue_head_t	frame_wait;
+	struct armada_frame_work *frame_work;
+
+	spinlock_t		irq_lock;
+	uint32_t		irq_ena;
+	struct list_head	vbl_list;
+};
+#define drm_to_armada_crtc(c) container_of(c, struct armada_crtc, crtc)
+
+int armada_drm_crtc_create(struct drm_device *, unsigned, struct resource *);
+void armada_drm_crtc_gamma_set(struct drm_crtc *, u16, u16, u16, int);
+void armada_drm_crtc_gamma_get(struct drm_crtc *, u16 *, u16 *, u16 *, int);
+void armada_drm_crtc_irq(struct armada_crtc *, u32);
+void armada_drm_crtc_disable_irq(struct armada_crtc *, u32);
+void armada_drm_crtc_enable_irq(struct armada_crtc *, u32);
+void armada_drm_crtc_update_regs(struct armada_crtc *, struct armada_regs *);
+
+#endif
diff --git a/drivers/gpu/drm/armada/armada_debugfs.c b/drivers/gpu/drm/armada/armada_debugfs.c
new file mode 100644
index 000000000000..471e45627f1e
--- /dev/null
+++ b/drivers/gpu/drm/armada/armada_debugfs.c
@@ -0,0 +1,177 @@
+/*
+ * Copyright (C) 2012 Russell King
+ *  Rewritten from the dovefb driver, and Armada510 manuals.
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ */
+#include <linux/ctype.h>
+#include <linux/debugfs.h>
+#include <linux/module.h>
+#include <linux/seq_file.h>
+#include <drm/drmP.h>
+#include "armada_crtc.h"
+#include "armada_drm.h"
+
+static int armada_debugfs_gem_linear_show(struct seq_file *m, void *data)
+{
+	struct drm_info_node *node = m->private;
+	struct drm_device *dev = node->minor->dev;
+	struct armada_private *priv = dev->dev_private;
+	int ret;
+
+	mutex_lock(&dev->struct_mutex);
+	ret = drm_mm_dump_table(m, &priv->linear);
+	mutex_unlock(&dev->struct_mutex);
+
+	return ret;
+}
+
+static int armada_debugfs_reg_show(struct seq_file *m, void *data)
+{
+	struct drm_device *dev = m->private;
+	struct armada_private *priv = dev->dev_private;
+	int n, i;
+
+	if (priv) {
+		for (n = 0; n < ARRAY_SIZE(priv->dcrtc); n++) {
+			struct armada_crtc *dcrtc = priv->dcrtc[n];
+			if (!dcrtc)
+				continue;
+
+			for (i = 0x84; i <= 0x1c4; i += 4) {
+				uint32_t v = readl_relaxed(dcrtc->base + i);
+				seq_printf(m, "%u: 0x%04x: 0x%08x\n", n, i, v);
+			}
+		}
+	}
+
+	return 0;
+}
+
+static int armada_debugfs_reg_r_open(struct inode *inode, struct file *file)
+{
+	return single_open(file, armada_debugfs_reg_show, inode->i_private);
+}
+
+static const struct file_operations fops_reg_r = {
+	.owner = THIS_MODULE,
+	.open = armada_debugfs_reg_r_open,
+	.read = seq_read,
+	.llseek = seq_lseek,
+	.release = single_release,
+};
+
+static int armada_debugfs_write(struct file *file, const char __user *ptr,
+	size_t len, loff_t *off)
+{
+	struct drm_device *dev = file->private_data;
+	struct armada_private *priv = dev->dev_private;
+	struct armada_crtc *dcrtc = priv->dcrtc[0];
+	char buf[32], *p;
+	uint32_t reg, val;
+	int ret;
+
+	if (*off != 0)
+		return 0;
+
+	if (len > sizeof(buf) - 1)
+		len = sizeof(buf) - 1;
+
+	ret = strncpy_from_user(buf, ptr, len);
+	if (ret < 0)
+		return ret;
+	buf[len] = '\0';
+
+	reg = simple_strtoul(buf, &p, 16);
+	if (!isspace(*p))
+		return -EINVAL;
+	val = simple_strtoul(p + 1, NULL, 16);
+
+	if (reg >= 0x84 && reg <= 0x1c4)
+		writel(val, dcrtc->base + reg);
+
+	return len;
+}
+
+static const struct file_operations fops_reg_w = {
+	.owner = THIS_MODULE,
+	.open = simple_open,
+	.write = armada_debugfs_write,
+	.llseek = noop_llseek,
+};
+
+static struct drm_info_list armada_debugfs_list[] = {
+	{ "gem_linear", armada_debugfs_gem_linear_show, 0 },
+};
+#define ARMADA_DEBUGFS_ENTRIES ARRAY_SIZE(armada_debugfs_list)
+
+static int drm_add_fake_info_node(struct drm_minor *minor, struct dentry *ent,
+	const void *key)
+{
+	struct drm_info_node *node;
+
+	node = kmalloc(sizeof(struct drm_info_node), GFP_KERNEL);
+	if (node == NULL) {
+		debugfs_remove(ent);
+		return -ENOMEM;
+	}
+
+	node->minor = minor;
+	node->dent = ent;
+	node->info_ent = (void *) key;
+
+	mutex_lock(&minor->debugfs_lock);
+	list_add(&node->list, &minor->debugfs_list);
+	mutex_unlock(&minor->debugfs_lock);
+
+	return 0;
+}
+
+static int armada_debugfs_create(struct dentry *root, struct drm_minor *minor,
+	const char *name, umode_t mode, const struct file_operations *fops)
+{
+	struct dentry *de;
+
+	de = debugfs_create_file(name, mode, root, minor->dev, fops);
+
+	return drm_add_fake_info_node(minor, de, fops);
+}
+
+int armada_drm_debugfs_init(struct drm_minor *minor)
+{
+	int ret;
+
+	ret = drm_debugfs_create_files(armada_debugfs_list,
+				       ARMADA_DEBUGFS_ENTRIES,
+				       minor->debugfs_root, minor);
+	if (ret)
+		return ret;
+
+	ret = armada_debugfs_create(minor->debugfs_root, minor,
+				   "reg", S_IFREG | S_IRUSR, &fops_reg_r);
+	if (ret)
+		goto err_1;
+
+	ret = armada_debugfs_create(minor->debugfs_root, minor,
+				"reg_wr", S_IFREG | S_IWUSR, &fops_reg_w);
+	if (ret)
+		goto err_2;
+	return ret;
+
+ err_2:
+	drm_debugfs_remove_files((struct drm_info_list *)&fops_reg_r, 1, minor);
+ err_1:
+	drm_debugfs_remove_files(armada_debugfs_list, ARMADA_DEBUGFS_ENTRIES,
+				 minor);
+	return ret;
+}
+
+void armada_drm_debugfs_cleanup(struct drm_minor *minor)
+{
+	drm_debugfs_remove_files((struct drm_info_list *)&fops_reg_w, 1, minor);
+	drm_debugfs_remove_files((struct drm_info_list *)&fops_reg_r, 1, minor);
+	drm_debugfs_remove_files(armada_debugfs_list, ARMADA_DEBUGFS_ENTRIES,
+				 minor);
+}
diff --git a/drivers/gpu/drm/armada/armada_drm.h b/drivers/gpu/drm/armada/armada_drm.h
new file mode 100644
index 000000000000..eef09ec9a5ff
--- /dev/null
+++ b/drivers/gpu/drm/armada/armada_drm.h
@@ -0,0 +1,113 @@
+/*
+ * Copyright (C) 2012 Russell King
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ */
+#ifndef ARMADA_DRM_H
+#define ARMADA_DRM_H
+
+#include <linux/kfifo.h>
+#include <linux/io.h>
+#include <linux/workqueue.h>
+#include <drm/drmP.h>
+
+struct armada_crtc;
+struct armada_gem_object;
+struct clk;
+struct drm_fb_helper;
+
+static inline void
+armada_updatel(uint32_t val, uint32_t mask, void __iomem *ptr)
+{
+	uint32_t ov, v;
+
+	ov = v = readl_relaxed(ptr);
+	v = (v & ~mask) | val;
+	if (ov != v)
+		writel_relaxed(v, ptr);
+}
+
+static inline uint32_t armada_pitch(uint32_t width, uint32_t bpp)
+{
+	uint32_t pitch = bpp != 4 ? width * ((bpp + 7) / 8) : width / 2;
+
+	/* 88AP510 spec recommends pitch be a multiple of 128 */
+	return ALIGN(pitch, 128);
+}
+
+struct armada_vbl_event {
+	struct list_head	node;
+	void			*data;
+	void			(*fn)(struct armada_crtc *, void *);
+};
+void armada_drm_vbl_event_add(struct armada_crtc *,
+	struct armada_vbl_event *);
+void armada_drm_vbl_event_remove(struct armada_crtc *,
+	struct armada_vbl_event *);
+void armada_drm_vbl_event_remove_unlocked(struct armada_crtc *,
+	struct armada_vbl_event *);
+#define armada_drm_vbl_event_init(_e, _f, _d) do {	\
+	struct armada_vbl_event *__e = _e;		\
+	INIT_LIST_HEAD(&__e->node);			\
+	__e->data = _d;					\
+	__e->fn = _f;					\
+} while (0)
+
+
+struct armada_private;
+
+struct armada_variant {
+	bool	has_spu_adv_reg;
+	uint32_t spu_adv_reg;
+	int (*init)(struct armada_private *, struct device *);
+	int (*crtc_init)(struct armada_crtc *);
+	int (*crtc_compute_clock)(struct armada_crtc *,
+				  const struct drm_display_mode *,
+				  uint32_t *);
+};
+
+/* Variant ops */
+extern const struct armada_variant armada510_ops;
+
+struct armada_private {
+	const struct armada_variant *variant;
+	struct work_struct	fb_unref_work;
+	DECLARE_KFIFO(fb_unref, struct drm_framebuffer *, 8);
+	struct drm_fb_helper	*fbdev;
+	struct armada_crtc	*dcrtc[2];
+	struct drm_mm		linear;
+	struct clk		*extclk[2];
+	struct drm_property	*csc_yuv_prop;
+	struct drm_property	*csc_rgb_prop;
+	struct drm_property	*colorkey_prop;
+	struct drm_property	*colorkey_min_prop;
+	struct drm_property	*colorkey_max_prop;
+	struct drm_property	*colorkey_val_prop;
+	struct drm_property	*colorkey_alpha_prop;
+	struct drm_property	*colorkey_mode_prop;
+	struct drm_property	*brightness_prop;
+	struct drm_property	*contrast_prop;
+	struct drm_property	*saturation_prop;
+#ifdef CONFIG_DEBUG_FS
+	struct dentry		*de;
+#endif
+};
+
+void __armada_drm_queue_unref_work(struct drm_device *,
+	struct drm_framebuffer *);
+void armada_drm_queue_unref_work(struct drm_device *,
+	struct drm_framebuffer *);
+
+extern const struct drm_mode_config_funcs armada_drm_mode_config_funcs;
+
+int armada_fbdev_init(struct drm_device *);
+void armada_fbdev_fini(struct drm_device *);
+
+int armada_overlay_plane_create(struct drm_device *, unsigned long);
+
+int armada_drm_debugfs_init(struct drm_minor *);
+void armada_drm_debugfs_cleanup(struct drm_minor *);
+
+#endif
diff --git a/drivers/gpu/drm/armada/armada_drv.c b/drivers/gpu/drm/armada/armada_drv.c
new file mode 100644
index 000000000000..4f2b28354915
--- /dev/null
+++ b/drivers/gpu/drm/armada/armada_drv.c
@@ -0,0 +1,421 @@
+/*
+ * Copyright (C) 2012 Russell King
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ */
+#include <linux/clk.h>
+#include <linux/module.h>
+#include <drm/drmP.h>
+#include <drm/drm_crtc_helper.h>
+#include "armada_crtc.h"
+#include "armada_drm.h"
+#include "armada_gem.h"
+#include "armada_hw.h"
+#include <drm/armada_drm.h>
+#include "armada_ioctlP.h"
+
+#ifdef CONFIG_DRM_ARMADA_TDA1998X
+#include <drm/i2c/tda998x.h>
+#include "armada_slave.h"
+
+static struct tda998x_encoder_params params = {
+	/* With 0x24, there is no translation between vp_out and int_vp
+	FB	LCD out	Pins	VIP	Int Vp
+	R:23:16	R:7:0	VPC7:0	7:0	7:0[R]
+	G:15:8	G:15:8	VPB7:0	23:16	23:16[G]
+	B:7:0	B:23:16	VPA7:0	15:8	15:8[B]
+	*/
+	.swap_a = 2,
+	.swap_b = 3,
+	.swap_c = 4,
+	.swap_d = 5,
+	.swap_e = 0,
+	.swap_f = 1,
+	.audio_cfg = BIT(2),
+	.audio_frame[1] = 1,
+	.audio_format = AFMT_SPDIF,
+	.audio_sample_rate = 44100,
+};
+
+static const struct armada_drm_slave_config tda19988_config = {
+	.i2c_adapter_id = 0,
+	.crtcs = 1 << 0, /* Only LCD0 at the moment */
+	.polled = DRM_CONNECTOR_POLL_CONNECT | DRM_CONNECTOR_POLL_DISCONNECT,
+	.interlace_allowed = true,
+	.info = {
+		.type = "tda998x",
+		.addr = 0x70,
+		.platform_data = &params,
+	},
+};
+#endif
+
+static void armada_drm_unref_work(struct work_struct *work)
+{
+	struct armada_private *priv =
+		container_of(work, struct armada_private, fb_unref_work);
+	struct drm_framebuffer *fb;
+
+	while (kfifo_get(&priv->fb_unref, &fb))
+		drm_framebuffer_unreference(fb);
+}
+
+/* Must be called with dev->event_lock held */
+void __armada_drm_queue_unref_work(struct drm_device *dev,
+	struct drm_framebuffer *fb)
+{
+	struct armada_private *priv = dev->dev_private;
+
+	/*
+	 * Yes, we really must jump through these hoops just to store a
+	 * _pointer_ to something into the kfifo.  This is utterly insane
+	 * and idiotic, because it kfifo requires the _data_ pointed to by
+	 * the pointer const, not the pointer itself.  Not only that, but
+	 * you have to pass a pointer _to_ the pointer you want stored.
+	 */
+	const struct drm_framebuffer *silly_api_alert = fb;
+	WARN_ON(!kfifo_put(&priv->fb_unref, &silly_api_alert));
+	schedule_work(&priv->fb_unref_work);
+}
+
+void armada_drm_queue_unref_work(struct drm_device *dev,
+	struct drm_framebuffer *fb)
+{
+	unsigned long flags;
+
+	spin_lock_irqsave(&dev->event_lock, flags);
+	__armada_drm_queue_unref_work(dev, fb);
+	spin_unlock_irqrestore(&dev->event_lock, flags);
+}
+
+static int armada_drm_load(struct drm_device *dev, unsigned long flags)
+{
+	const struct platform_device_id *id;
+	struct armada_private *priv;
+	struct resource *res[ARRAY_SIZE(priv->dcrtc)];
+	struct resource *mem = NULL;
+	int ret, n, i;
+
+	memset(res, 0, sizeof(res));
+
+	for (n = i = 0; ; n++) {
+		struct resource *r = platform_get_resource(dev->platformdev,
+							   IORESOURCE_MEM, n);
+		if (!r)
+			break;
+
+		/* Resources above 64K are graphics memory */
+		if (resource_size(r) > SZ_64K)
+			mem = r;
+		else if (i < ARRAY_SIZE(priv->dcrtc))
+			res[i++] = r;
+		else
+			return -EINVAL;
+	}
+
+	if (!res[0] || !mem)
+		return -ENXIO;
+
+	if (!devm_request_mem_region(dev->dev, mem->start,
+			resource_size(mem), "armada-drm"))
+		return -EBUSY;
+
+	priv = devm_kzalloc(dev->dev, sizeof(*priv), GFP_KERNEL);
+	if (!priv) {
+		DRM_ERROR("failed to allocate private\n");
+		return -ENOMEM;
+	}
+
+	dev->dev_private = priv;
+
+	/* Get the implementation specific driver data. */
+	id = platform_get_device_id(dev->platformdev);
+	if (!id)
+		return -ENXIO;
+
+	priv->variant = (struct armada_variant *)id->driver_data;
+
+	ret = priv->variant->init(priv, dev->dev);
+	if (ret)
+		return ret;
+
+	INIT_WORK(&priv->fb_unref_work, armada_drm_unref_work);
+	INIT_KFIFO(priv->fb_unref);
+
+	/* Mode setting support */
+	drm_mode_config_init(dev);
+	dev->mode_config.min_width = 320;
+	dev->mode_config.min_height = 200;
+
+	/*
+	 * With vscale enabled, the maximum width is 1920 due to the
+	 * 1920 by 3 lines RAM
+	 */
+	dev->mode_config.max_width = 1920;
+	dev->mode_config.max_height = 2048;
+
+	dev->mode_config.preferred_depth = 24;
+	dev->mode_config.funcs = &armada_drm_mode_config_funcs;
+	drm_mm_init(&priv->linear, mem->start, resource_size(mem));
+
+	/* Create all LCD controllers */
+	for (n = 0; n < ARRAY_SIZE(priv->dcrtc); n++) {
+		if (!res[n])
+			break;
+
+		ret = armada_drm_crtc_create(dev, n, res[n]);
+		if (ret)
+			goto err_kms;
+	}
+
+#ifdef CONFIG_DRM_ARMADA_TDA1998X
+	ret = armada_drm_connector_slave_create(dev, &tda19988_config);
+	if (ret)
+		goto err_kms;
+#endif
+
+	ret = drm_vblank_init(dev, n);
+	if (ret)
+		goto err_kms;
+
+	ret = drm_irq_install(dev);
+	if (ret)
+		goto err_kms;
+
+	dev->vblank_disable_allowed = 1;
+
+	ret = armada_fbdev_init(dev);
+	if (ret)
+		goto err_irq;
+
+	drm_kms_helper_poll_init(dev);
+
+	return 0;
+
+ err_irq:
+	drm_irq_uninstall(dev);
+ err_kms:
+	drm_mode_config_cleanup(dev);
+	drm_mm_takedown(&priv->linear);
+	flush_work(&priv->fb_unref_work);
+
+	return ret;
+}
+
+static int armada_drm_unload(struct drm_device *dev)
+{
+	struct armada_private *priv = dev->dev_private;
+
+	drm_kms_helper_poll_fini(dev);
+	armada_fbdev_fini(dev);
+	drm_irq_uninstall(dev);
+	drm_mode_config_cleanup(dev);
+	drm_mm_takedown(&priv->linear);
+	flush_work(&priv->fb_unref_work);
+	dev->dev_private = NULL;
+
+	return 0;
+}
+
+void armada_drm_vbl_event_add(struct armada_crtc *dcrtc,
+	struct armada_vbl_event *evt)
+{
+	unsigned long flags;
+
+	spin_lock_irqsave(&dcrtc->irq_lock, flags);
+	if (list_empty(&evt->node)) {
+		list_add_tail(&evt->node, &dcrtc->vbl_list);
+
+		drm_vblank_get(dcrtc->crtc.dev, dcrtc->num);
+	}
+	spin_unlock_irqrestore(&dcrtc->irq_lock, flags);
+}
+
+void armada_drm_vbl_event_remove(struct armada_crtc *dcrtc,
+	struct armada_vbl_event *evt)
+{
+	if (!list_empty(&evt->node)) {
+		list_del_init(&evt->node);
+		drm_vblank_put(dcrtc->crtc.dev, dcrtc->num);
+	}
+}
+
+void armada_drm_vbl_event_remove_unlocked(struct armada_crtc *dcrtc,
+	struct armada_vbl_event *evt)
+{
+	unsigned long flags;
+
+	spin_lock_irqsave(&dcrtc->irq_lock, flags);
+	armada_drm_vbl_event_remove(dcrtc, evt);
+	spin_unlock_irqrestore(&dcrtc->irq_lock, flags);
+}
+
+/* These are called under the vbl_lock. */
+static int armada_drm_enable_vblank(struct drm_device *dev, int crtc)
+{
+	struct armada_private *priv = dev->dev_private;
+	armada_drm_crtc_enable_irq(priv->dcrtc[crtc], VSYNC_IRQ_ENA);
+	return 0;
+}
+
+static void armada_drm_disable_vblank(struct drm_device *dev, int crtc)
+{
+	struct armada_private *priv = dev->dev_private;
+	armada_drm_crtc_disable_irq(priv->dcrtc[crtc], VSYNC_IRQ_ENA);
+}
+
+static irqreturn_t armada_drm_irq_handler(int irq, void *arg)
+{
+	struct drm_device *dev = arg;
+	struct armada_private *priv = dev->dev_private;
+	struct armada_crtc *dcrtc = priv->dcrtc[0];
+	uint32_t v, stat = readl_relaxed(dcrtc->base + LCD_SPU_IRQ_ISR);
+	irqreturn_t handled = IRQ_NONE;
+
+	/*
+	 * This is rediculous - rather than writing bits to clear, we
+	 * have to set the actual status register value.  This is racy.
+	 */
+	writel_relaxed(0, dcrtc->base + LCD_SPU_IRQ_ISR);
+
+	/* Mask out those interrupts we haven't enabled */
+	v = stat & dcrtc->irq_ena;
+
+	if (v & (VSYNC_IRQ|GRA_FRAME_IRQ|DUMB_FRAMEDONE)) {
+		armada_drm_crtc_irq(dcrtc, stat);
+		handled = IRQ_HANDLED;
+	}
+
+	return handled;
+}
+
+static int armada_drm_irq_postinstall(struct drm_device *dev)
+{
+	struct armada_private *priv = dev->dev_private;
+	struct armada_crtc *dcrtc = priv->dcrtc[0];
+
+	spin_lock_irq(&dev->vbl_lock);
+	writel_relaxed(dcrtc->irq_ena, dcrtc->base + LCD_SPU_IRQ_ENA);
+	writel(0, dcrtc->base + LCD_SPU_IRQ_ISR);
+	spin_unlock_irq(&dev->vbl_lock);
+
+	return 0;
+}
+
+static void armada_drm_irq_uninstall(struct drm_device *dev)
+{
+	struct armada_private *priv = dev->dev_private;
+	struct armada_crtc *dcrtc = priv->dcrtc[0];
+
+	writel(0, dcrtc->base + LCD_SPU_IRQ_ENA);
+}
+
+static struct drm_ioctl_desc armada_ioctls[] = {
+	DRM_IOCTL_DEF_DRV(ARMADA_GEM_CREATE, armada_gem_create_ioctl,
+		DRM_UNLOCKED),
+	DRM_IOCTL_DEF_DRV(ARMADA_GEM_MMAP, armada_gem_mmap_ioctl,
+		DRM_UNLOCKED),
+	DRM_IOCTL_DEF_DRV(ARMADA_GEM_PWRITE, armada_gem_pwrite_ioctl,
+		DRM_UNLOCKED),
+};
+
+static const struct file_operations armada_drm_fops = {
+	.owner			= THIS_MODULE,
+	.llseek			= no_llseek,
+	.read			= drm_read,
+	.poll			= drm_poll,
+	.unlocked_ioctl		= drm_ioctl,
+	.mmap			= drm_gem_mmap,
+	.open			= drm_open,
+	.release		= drm_release,
+};
+
+static struct drm_driver armada_drm_driver = {
+	.load			= armada_drm_load,
+	.open			= NULL,
+	.preclose		= NULL,
+	.postclose		= NULL,
+	.lastclose		= NULL,
+	.unload			= armada_drm_unload,
+	.get_vblank_counter	= drm_vblank_count,
+	.enable_vblank		= armada_drm_enable_vblank,
+	.disable_vblank		= armada_drm_disable_vblank,
+	.irq_handler		= armada_drm_irq_handler,
+	.irq_postinstall	= armada_drm_irq_postinstall,
+	.irq_uninstall		= armada_drm_irq_uninstall,
+#ifdef CONFIG_DEBUG_FS
+	.debugfs_init		= armada_drm_debugfs_init,
+	.debugfs_cleanup	= armada_drm_debugfs_cleanup,
+#endif
+	.gem_free_object	= armada_gem_free_object,
+	.prime_handle_to_fd	= drm_gem_prime_handle_to_fd,
+	.prime_fd_to_handle	= drm_gem_prime_fd_to_handle,
+	.gem_prime_export	= armada_gem_prime_export,
+	.gem_prime_import	= armada_gem_prime_import,
+	.dumb_create		= armada_gem_dumb_create,
+	.dumb_map_offset	= armada_gem_dumb_map_offset,
+	.dumb_destroy		= armada_gem_dumb_destroy,
+	.gem_vm_ops		= &armada_gem_vm_ops,
+	.major			= 1,
+	.minor			= 0,
+	.name			= "armada-drm",
+	.desc			= "Armada SoC DRM",
+	.date			= "20120730",
+	.driver_features	= DRIVER_GEM | DRIVER_MODESET |
+				  DRIVER_HAVE_IRQ | DRIVER_PRIME,
+	.ioctls			= armada_ioctls,
+	.fops			= &armada_drm_fops,
+};
+
+static int armada_drm_probe(struct platform_device *pdev)
+{
+	return drm_platform_init(&armada_drm_driver, pdev);
+}
+
+static int armada_drm_remove(struct platform_device *pdev)
+{
+	drm_platform_exit(&armada_drm_driver, pdev);
+	return 0;
+}
+
+static const struct platform_device_id armada_drm_platform_ids[] = {
+	{
+		.name		= "armada-drm",
+		.driver_data	= (unsigned long)&armada510_ops,
+	}, {
+		.name		= "armada-510-drm",
+		.driver_data	= (unsigned long)&armada510_ops,
+	},
+	{ },
+};
+MODULE_DEVICE_TABLE(platform, armada_drm_platform_ids);
+
+static struct platform_driver armada_drm_platform_driver = {
+	.probe	= armada_drm_probe,
+	.remove	= armada_drm_remove,
+	.driver	= {
+		.name	= "armada-drm",
+		.owner	= THIS_MODULE,
+	},
+	.id_table = armada_drm_platform_ids,
+};
+
+static int __init armada_drm_init(void)
+{
+	armada_drm_driver.num_ioctls = DRM_ARRAY_SIZE(armada_ioctls);
+	return platform_driver_register(&armada_drm_platform_driver);
+}
+module_init(armada_drm_init);
+
+static void __exit armada_drm_exit(void)
+{
+	platform_driver_unregister(&armada_drm_platform_driver);
+}
+module_exit(armada_drm_exit);
+
+MODULE_AUTHOR("Russell King <rmk+kernel@arm.linux.org.uk>");
+MODULE_DESCRIPTION("Armada DRM Driver");
+MODULE_LICENSE("GPL");
+MODULE_ALIAS("platform:armada-drm");
diff --git a/drivers/gpu/drm/armada/armada_fb.c b/drivers/gpu/drm/armada/armada_fb.c
new file mode 100644
index 000000000000..1c90969def3e
--- /dev/null
+++ b/drivers/gpu/drm/armada/armada_fb.c
@@ -0,0 +1,170 @@
+/*
+ * Copyright (C) 2012 Russell King
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ */
+#include <drm/drmP.h>
+#include <drm/drm_crtc_helper.h>
+#include <drm/drm_fb_helper.h>
+#include "armada_drm.h"
+#include "armada_fb.h"
+#include "armada_gem.h"
+#include "armada_hw.h"
+
+static void armada_fb_destroy(struct drm_framebuffer *fb)
+{
+	struct armada_framebuffer *dfb = drm_fb_to_armada_fb(fb);
+
+	drm_framebuffer_cleanup(&dfb->fb);
+	drm_gem_object_unreference_unlocked(&dfb->obj->obj);
+	kfree(dfb);
+}
+
+static int armada_fb_create_handle(struct drm_framebuffer *fb,
+	struct drm_file *dfile, unsigned int *handle)
+{
+	struct armada_framebuffer *dfb = drm_fb_to_armada_fb(fb);
+	return drm_gem_handle_create(dfile, &dfb->obj->obj, handle);
+}
+
+static const struct drm_framebuffer_funcs armada_fb_funcs = {
+	.destroy	= armada_fb_destroy,
+	.create_handle	= armada_fb_create_handle,
+};
+
+struct armada_framebuffer *armada_framebuffer_create(struct drm_device *dev,
+	struct drm_mode_fb_cmd2 *mode, struct armada_gem_object *obj)
+{
+	struct armada_framebuffer *dfb;
+	uint8_t format, config;
+	int ret;
+
+	switch (mode->pixel_format) {
+#define FMT(drm, fmt, mod)		\
+	case DRM_FORMAT_##drm:		\
+		format = CFG_##fmt;	\
+		config = mod;		\
+		break
+	FMT(RGB565,	565,		CFG_SWAPRB);
+	FMT(BGR565,	565,		0);
+	FMT(ARGB1555,	1555,		CFG_SWAPRB);
+	FMT(ABGR1555,	1555,		0);
+	FMT(RGB888,	888PACK,	CFG_SWAPRB);
+	FMT(BGR888,	888PACK,	0);
+	FMT(XRGB8888,	X888,		CFG_SWAPRB);
+	FMT(XBGR8888,	X888,		0);
+	FMT(ARGB8888,	8888,		CFG_SWAPRB);
+	FMT(ABGR8888,	8888,		0);
+	FMT(YUYV,	422PACK,	CFG_YUV2RGB | CFG_SWAPYU | CFG_SWAPUV);
+	FMT(UYVY,	422PACK,	CFG_YUV2RGB);
+	FMT(VYUY,	422PACK,	CFG_YUV2RGB | CFG_SWAPUV);
+	FMT(YVYU,	422PACK,	CFG_YUV2RGB | CFG_SWAPYU);
+	FMT(YUV422,	422,		CFG_YUV2RGB);
+	FMT(YVU422,	422,		CFG_YUV2RGB | CFG_SWAPUV);
+	FMT(YUV420,	420,		CFG_YUV2RGB);
+	FMT(YVU420,	420,		CFG_YUV2RGB | CFG_SWAPUV);
+	FMT(C8,		PSEUDO8,	0);
+#undef FMT
+	default:
+		return ERR_PTR(-EINVAL);
+	}
+
+	dfb = kzalloc(sizeof(*dfb), GFP_KERNEL);
+	if (!dfb) {
+		DRM_ERROR("failed to allocate Armada fb object\n");
+		return ERR_PTR(-ENOMEM);
+	}
+
+	dfb->fmt = format;
+	dfb->mod = config;
+	dfb->obj = obj;
+
+	drm_helper_mode_fill_fb_struct(&dfb->fb, mode);
+
+	ret = drm_framebuffer_init(dev, &dfb->fb, &armada_fb_funcs);
+	if (ret) {
+		kfree(dfb);
+		return ERR_PTR(ret);
+	}
+
+	/*
+	 * Take a reference on our object as we're successful - the
+	 * caller already holds a reference, which keeps us safe for
+	 * the above call, but the caller will drop their reference
+	 * to it.  Hence we need to take our own reference.
+	 */
+	drm_gem_object_reference(&obj->obj);
+
+	return dfb;
+}
+
+static struct drm_framebuffer *armada_fb_create(struct drm_device *dev,
+	struct drm_file *dfile, struct drm_mode_fb_cmd2 *mode)
+{
+	struct armada_gem_object *obj;
+	struct armada_framebuffer *dfb;
+	int ret;
+
+	DRM_DEBUG_DRIVER("w%u h%u pf%08x f%u p%u,%u,%u\n",
+		mode->width, mode->height, mode->pixel_format,
+		mode->flags, mode->pitches[0], mode->pitches[1],
+		mode->pitches[2]);
+
+	/* We can only handle a single plane at the moment */
+	if (drm_format_num_planes(mode->pixel_format) > 1 &&
+	    (mode->handles[0] != mode->handles[1] ||
+	     mode->handles[0] != mode->handles[2])) {
+		ret = -EINVAL;
+		goto err;
+	}
+
+	obj = armada_gem_object_lookup(dev, dfile, mode->handles[0]);
+	if (!obj) {
+		ret = -ENOENT;
+		goto err;
+	}
+
+	if (obj->obj.import_attach && !obj->sgt) {
+		ret = armada_gem_map_import(obj);
+		if (ret)
+			goto err_unref;
+	}
+
+	/* Framebuffer objects must have a valid device address for scanout */
+	if (obj->dev_addr == DMA_ERROR_CODE) {
+		ret = -EINVAL;
+		goto err_unref;
+	}
+
+	dfb = armada_framebuffer_create(dev, mode, obj);
+	if (IS_ERR(dfb)) {
+		ret = PTR_ERR(dfb);
+		goto err;
+	}
+
+	drm_gem_object_unreference_unlocked(&obj->obj);
+
+	return &dfb->fb;
+
+ err_unref:
+	drm_gem_object_unreference_unlocked(&obj->obj);
+ err:
+	DRM_ERROR("failed to initialize framebuffer: %d\n", ret);
+	return ERR_PTR(ret);
+}
+
+static void armada_output_poll_changed(struct drm_device *dev)
+{
+	struct armada_private *priv = dev->dev_private;
+	struct drm_fb_helper *fbh = priv->fbdev;
+
+	if (fbh)
+		drm_fb_helper_hotplug_event(fbh);
+}
+
+const struct drm_mode_config_funcs armada_drm_mode_config_funcs = {
+	.fb_create		= armada_fb_create,
+	.output_poll_changed	= armada_output_poll_changed,
+};
diff --git a/drivers/gpu/drm/armada/armada_fb.h b/drivers/gpu/drm/armada/armada_fb.h
new file mode 100644
index 000000000000..ce3f12ebfc53
--- /dev/null
+++ b/drivers/gpu/drm/armada/armada_fb.h
@@ -0,0 +1,24 @@
+/*
+ * Copyright (C) 2012 Russell King
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ */
+#ifndef ARMADA_FB_H
+#define ARMADA_FB_H
+
+struct armada_framebuffer {
+	struct drm_framebuffer	fb;
+	struct armada_gem_object *obj;
+	uint8_t			fmt;
+	uint8_t			mod;
+};
+#define drm_fb_to_armada_fb(dfb) \
+	container_of(dfb, struct armada_framebuffer, fb)
+#define drm_fb_obj(fb) drm_fb_to_armada_fb(fb)->obj
+
+struct armada_framebuffer *armada_framebuffer_create(struct drm_device *,
+	struct drm_mode_fb_cmd2 *, struct armada_gem_object *);
+
+#endif
diff --git a/drivers/gpu/drm/armada/armada_fbdev.c b/drivers/gpu/drm/armada/armada_fbdev.c
new file mode 100644
index 000000000000..dd5ea77dac96
--- /dev/null
+++ b/drivers/gpu/drm/armada/armada_fbdev.c
@@ -0,0 +1,202 @@
+/*
+ * Copyright (C) 2012 Russell King
+ *  Written from the i915 driver.
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ */
+#include <linux/errno.h>
+#include <linux/fb.h>
+#include <linux/kernel.h>
+#include <linux/module.h>
+
+#include <drm/drmP.h>
+#include <drm/drm_fb_helper.h>
+#include "armada_crtc.h"
+#include "armada_drm.h"
+#include "armada_fb.h"
+#include "armada_gem.h"
+
+static /*const*/ struct fb_ops armada_fb_ops = {
+	.owner		= THIS_MODULE,
+	.fb_check_var	= drm_fb_helper_check_var,
+	.fb_set_par	= drm_fb_helper_set_par,
+	.fb_fillrect	= cfb_fillrect,
+	.fb_copyarea	= cfb_copyarea,
+	.fb_imageblit	= cfb_imageblit,
+	.fb_pan_display	= drm_fb_helper_pan_display,
+	.fb_blank	= drm_fb_helper_blank,
+	.fb_setcmap	= drm_fb_helper_setcmap,
+	.fb_debug_enter	= drm_fb_helper_debug_enter,
+	.fb_debug_leave	= drm_fb_helper_debug_leave,
+};
+
+static int armada_fb_create(struct drm_fb_helper *fbh,
+	struct drm_fb_helper_surface_size *sizes)
+{
+	struct drm_device *dev = fbh->dev;
+	struct drm_mode_fb_cmd2 mode;
+	struct armada_framebuffer *dfb;
+	struct armada_gem_object *obj;
+	struct fb_info *info;
+	int size, ret;
+	void *ptr;
+
+	memset(&mode, 0, sizeof(mode));
+	mode.width = sizes->surface_width;
+	mode.height = sizes->surface_height;
+	mode.pitches[0] = armada_pitch(mode.width, sizes->surface_bpp);
+	mode.pixel_format = drm_mode_legacy_fb_format(sizes->surface_bpp,
+					sizes->surface_depth);
+
+	size = mode.pitches[0] * mode.height;
+	obj = armada_gem_alloc_private_object(dev, size);
+	if (!obj) {
+		DRM_ERROR("failed to allocate fb memory\n");
+		return -ENOMEM;
+	}
+
+	ret = armada_gem_linear_back(dev, obj);
+	if (ret) {
+		drm_gem_object_unreference_unlocked(&obj->obj);
+		return ret;
+	}
+
+	ptr = armada_gem_map_object(dev, obj);
+	if (!ptr) {
+		drm_gem_object_unreference_unlocked(&obj->obj);
+		return -ENOMEM;
+	}
+
+	dfb = armada_framebuffer_create(dev, &mode, obj);
+
+	/*
+	 * A reference is now held by the framebuffer object if
+	 * successful, otherwise this drops the ref for the error path.
+	 */
+	drm_gem_object_unreference_unlocked(&obj->obj);
+
+	if (IS_ERR(dfb))
+		return PTR_ERR(dfb);
+
+	info = framebuffer_alloc(0, dev->dev);
+	if (!info) {
+		ret = -ENOMEM;
+		goto err_fballoc;
+	}
+
+	ret = fb_alloc_cmap(&info->cmap, 256, 0);
+	if (ret) {
+		ret = -ENOMEM;
+		goto err_fbcmap;
+	}
+
+	strlcpy(info->fix.id, "armada-drmfb", sizeof(info->fix.id));
+	info->par = fbh;
+	info->flags = FBINFO_DEFAULT | FBINFO_CAN_FORCE_OUTPUT;
+	info->fbops = &armada_fb_ops;
+	info->fix.smem_start = obj->phys_addr;
+	info->fix.smem_len = obj->obj.size;
+	info->screen_size = obj->obj.size;
+	info->screen_base = ptr;
+	fbh->fb = &dfb->fb;
+	fbh->fbdev = info;
+	drm_fb_helper_fill_fix(info, dfb->fb.pitches[0], dfb->fb.depth);
+	drm_fb_helper_fill_var(info, fbh, sizes->fb_width, sizes->fb_height);
+
+	DRM_DEBUG_KMS("allocated %dx%d %dbpp fb: 0x%08x\n",
+		dfb->fb.width, dfb->fb.height,
+		dfb->fb.bits_per_pixel, obj->phys_addr);
+
+	return 0;
+
+ err_fbcmap:
+	framebuffer_release(info);
+ err_fballoc:
+	dfb->fb.funcs->destroy(&dfb->fb);
+	return ret;
+}
+
+static int armada_fb_probe(struct drm_fb_helper *fbh,
+	struct drm_fb_helper_surface_size *sizes)
+{
+	int ret = 0;
+
+	if (!fbh->fb) {
+		ret = armada_fb_create(fbh, sizes);
+		if (ret == 0)
+			ret = 1;
+	}
+	return ret;
+}
+
+static struct drm_fb_helper_funcs armada_fb_helper_funcs = {
+	.gamma_set	= armada_drm_crtc_gamma_set,
+	.gamma_get	= armada_drm_crtc_gamma_get,
+	.fb_probe	= armada_fb_probe,
+};
+
+int armada_fbdev_init(struct drm_device *dev)
+{
+	struct armada_private *priv = dev->dev_private;
+	struct drm_fb_helper *fbh;
+	int ret;
+
+	fbh = devm_kzalloc(dev->dev, sizeof(*fbh), GFP_KERNEL);
+	if (!fbh)
+		return -ENOMEM;
+
+	priv->fbdev = fbh;
+
+	fbh->funcs = &armada_fb_helper_funcs;
+
+	ret = drm_fb_helper_init(dev, fbh, 1, 1);
+	if (ret) {
+		DRM_ERROR("failed to initialize drm fb helper\n");
+		goto err_fb_helper;
+	}
+
+	ret = drm_fb_helper_single_add_all_connectors(fbh);
+	if (ret) {
+		DRM_ERROR("failed to add fb connectors\n");
+		goto err_fb_setup;
+	}
+
+	ret = drm_fb_helper_initial_config(fbh, 32);
+	if (ret) {
+		DRM_ERROR("failed to set initial config\n");
+		goto err_fb_setup;
+	}
+
+	return 0;
+ err_fb_setup:
+	drm_fb_helper_fini(fbh);
+ err_fb_helper:
+	priv->fbdev = NULL;
+	return ret;
+}
+
+void armada_fbdev_fini(struct drm_device *dev)
+{
+	struct armada_private *priv = dev->dev_private;
+	struct drm_fb_helper *fbh = priv->fbdev;
+
+	if (fbh) {
+		struct fb_info *info = fbh->fbdev;
+
+		if (info) {
+			unregister_framebuffer(info);
+			if (info->cmap.len)
+				fb_dealloc_cmap(&info->cmap);
+			framebuffer_release(info);
+		}
+
+		if (fbh->fb)
+			fbh->fb->funcs->destroy(fbh->fb);
+
+		drm_fb_helper_fini(fbh);
+
+		priv->fbdev = NULL;
+	}
+}
diff --git a/drivers/gpu/drm/armada/armada_gem.c b/drivers/gpu/drm/armada/armada_gem.c
new file mode 100644
index 000000000000..9f2356bae7fd
--- /dev/null
+++ b/drivers/gpu/drm/armada/armada_gem.c
@@ -0,0 +1,611 @@
+/*
+ * Copyright (C) 2012 Russell King
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ */
+#include <linux/dma-buf.h>
+#include <linux/dma-mapping.h>
+#include <linux/shmem_fs.h>
+#include <drm/drmP.h>
+#include "armada_drm.h"
+#include "armada_gem.h"
+#include <drm/armada_drm.h>
+#include "armada_ioctlP.h"
+
+static int armada_gem_vm_fault(struct vm_area_struct *vma, struct vm_fault *vmf)
+{
+	struct armada_gem_object *obj = drm_to_armada_gem(vma->vm_private_data);
+	unsigned long addr = (unsigned long)vmf->virtual_address;
+	unsigned long pfn = obj->phys_addr >> PAGE_SHIFT;
+	int ret;
+
+	pfn += (addr - vma->vm_start) >> PAGE_SHIFT;
+	ret = vm_insert_pfn(vma, addr, pfn);
+
+	switch (ret) {
+	case 0:
+	case -EBUSY:
+		return VM_FAULT_NOPAGE;
+	case -ENOMEM:
+		return VM_FAULT_OOM;
+	default:
+		return VM_FAULT_SIGBUS;
+	}
+}
+
+const struct vm_operations_struct armada_gem_vm_ops = {
+	.fault	= armada_gem_vm_fault,
+	.open	= drm_gem_vm_open,
+	.close	= drm_gem_vm_close,
+};
+
+static size_t roundup_gem_size(size_t size)
+{
+	return roundup(size, PAGE_SIZE);
+}
+
+/* dev->struct_mutex is held here */
+void armada_gem_free_object(struct drm_gem_object *obj)
+{
+	struct armada_gem_object *dobj = drm_to_armada_gem(obj);
+
+	DRM_DEBUG_DRIVER("release obj %p\n", dobj);
+
+	drm_gem_free_mmap_offset(&dobj->obj);
+
+	if (dobj->page) {
+		/* page backed memory */
+		unsigned int order = get_order(dobj->obj.size);
+		__free_pages(dobj->page, order);
+	} else if (dobj->linear) {
+		/* linear backed memory */
+		drm_mm_remove_node(dobj->linear);
+		kfree(dobj->linear);
+		if (dobj->addr)
+			iounmap(dobj->addr);
+	}
+
+	if (dobj->obj.import_attach) {
+		/* We only ever display imported data */
+		dma_buf_unmap_attachment(dobj->obj.import_attach, dobj->sgt,
+					 DMA_TO_DEVICE);
+		drm_prime_gem_destroy(&dobj->obj, NULL);
+	}
+
+	drm_gem_object_release(&dobj->obj);
+
+	kfree(dobj);
+}
+
+int
+armada_gem_linear_back(struct drm_device *dev, struct armada_gem_object *obj)
+{
+	struct armada_private *priv = dev->dev_private;
+	size_t size = obj->obj.size;
+
+	if (obj->page || obj->linear)
+		return 0;
+
+	/*
+	 * If it is a small allocation (typically cursor, which will
+	 * be 32x64 or 64x32 ARGB pixels) try to get it from the system.
+	 * Framebuffers will never be this small (our minimum size for
+	 * framebuffers is larger than this anyway.)  Such objects are
+	 * only accessed by the CPU so we don't need any special handing
+	 * here.
+	 */
+	if (size <= 8192) {
+		unsigned int order = get_order(size);
+		struct page *p = alloc_pages(GFP_KERNEL, order);
+
+		if (p) {
+			obj->addr = page_address(p);
+			obj->phys_addr = page_to_phys(p);
+			obj->page = p;
+
+			memset(obj->addr, 0, PAGE_ALIGN(size));
+		}
+	}
+
+	/*
+	 * We could grab something from CMA if it's enabled, but that
+	 * involves building in a problem:
+	 *
+	 * CMA's interface uses dma_alloc_coherent(), which provides us
+	 * with an CPU virtual address and a device address.
+	 *
+	 * The CPU virtual address may be either an address in the kernel
+	 * direct mapped region (for example, as it would be on x86) or
+	 * it may be remapped into another part of kernel memory space
+	 * (eg, as it would be on ARM.)  This means virt_to_phys() on the
+	 * returned virtual address is invalid depending on the architecture
+	 * implementation.
+	 *
+	 * The device address may also not be a physical address; it may
+	 * be that there is some kind of remapping between the device and
+	 * system RAM, which makes the use of the device address also
+	 * unsafe to re-use as a physical address.
+	 *
+	 * This makes DRM usage of dma_alloc_coherent() in a generic way
+	 * at best very questionable and unsafe.
+	 */
+
+	/* Otherwise, grab it from our linear allocation */
+	if (!obj->page) {
+		struct drm_mm_node *node;
+		unsigned align = min_t(unsigned, size, SZ_2M);
+		void __iomem *ptr;
+		int ret;
+
+		node = kzalloc(sizeof(*node), GFP_KERNEL);
+		if (!node)
+			return -ENOSPC;
+
+		mutex_lock(&dev->struct_mutex);
+		ret = drm_mm_insert_node(&priv->linear, node, size, align,
+					 DRM_MM_SEARCH_DEFAULT);
+		mutex_unlock(&dev->struct_mutex);
+		if (ret) {
+			kfree(node);
+			return ret;
+		}
+
+		obj->linear = node;
+
+		/* Ensure that the memory we're returning is cleared. */
+		ptr = ioremap_wc(obj->linear->start, size);
+		if (!ptr) {
+			mutex_lock(&dev->struct_mutex);
+			drm_mm_remove_node(obj->linear);
+			mutex_unlock(&dev->struct_mutex);
+			kfree(obj->linear);
+			obj->linear = NULL;
+			return -ENOMEM;
+		}
+
+		memset_io(ptr, 0, size);
+		iounmap(ptr);
+
+		obj->phys_addr = obj->linear->start;
+		obj->dev_addr = obj->linear->start;
+	}
+
+	DRM_DEBUG_DRIVER("obj %p phys %#x dev %#x\n",
+			 obj, obj->phys_addr, obj->dev_addr);
+
+	return 0;
+}
+
+void *
+armada_gem_map_object(struct drm_device *dev, struct armada_gem_object *dobj)
+{
+	/* only linear objects need to be ioremap'd */
+	if (!dobj->addr && dobj->linear)
+		dobj->addr = ioremap_wc(dobj->phys_addr, dobj->obj.size);
+	return dobj->addr;
+}
+
+struct armada_gem_object *
+armada_gem_alloc_private_object(struct drm_device *dev, size_t size)
+{
+	struct armada_gem_object *obj;
+
+	size = roundup_gem_size(size);
+
+	obj = kzalloc(sizeof(*obj), GFP_KERNEL);
+	if (!obj)
+		return NULL;
+
+	drm_gem_private_object_init(dev, &obj->obj, size);
+	obj->dev_addr = DMA_ERROR_CODE;
+
+	DRM_DEBUG_DRIVER("alloc private obj %p size %zu\n", obj, size);
+
+	return obj;
+}
+
+struct armada_gem_object *armada_gem_alloc_object(struct drm_device *dev,
+	size_t size)
+{
+	struct armada_gem_object *obj;
+	struct address_space *mapping;
+
+	size = roundup_gem_size(size);
+
+	obj = kzalloc(sizeof(*obj), GFP_KERNEL);
+	if (!obj)
+		return NULL;
+
+	if (drm_gem_object_init(dev, &obj->obj, size)) {
+		kfree(obj);
+		return NULL;
+	}
+
+	obj->dev_addr = DMA_ERROR_CODE;
+
+	mapping = obj->obj.filp->f_path.dentry->d_inode->i_mapping;
+	mapping_set_gfp_mask(mapping, GFP_HIGHUSER | __GFP_RECLAIMABLE);
+
+	DRM_DEBUG_DRIVER("alloc obj %p size %zu\n", obj, size);
+
+	return obj;
+}
+
+/* Dumb alloc support */
+int armada_gem_dumb_create(struct drm_file *file, struct drm_device *dev,
+	struct drm_mode_create_dumb *args)
+{
+	struct armada_gem_object *dobj;
+	u32 handle;
+	size_t size;
+	int ret;
+
+	args->pitch = armada_pitch(args->width, args->bpp);
+	args->size = size = args->pitch * args->height;
+
+	dobj = armada_gem_alloc_private_object(dev, size);
+	if (dobj == NULL)
+		return -ENOMEM;
+
+	ret = armada_gem_linear_back(dev, dobj);
+	if (ret)
+		goto err;
+
+	ret = drm_gem_handle_create(file, &dobj->obj, &handle);
+	if (ret)
+		goto err;
+
+	args->handle = handle;
+
+	/* drop reference from allocate - handle holds it now */
+	DRM_DEBUG_DRIVER("obj %p size %zu handle %#x\n", dobj, size, handle);
+ err:
+	drm_gem_object_unreference_unlocked(&dobj->obj);
+	return ret;
+}
+
+int armada_gem_dumb_map_offset(struct drm_file *file, struct drm_device *dev,
+	uint32_t handle, uint64_t *offset)
+{
+	struct armada_gem_object *obj;
+	int ret = 0;
+
+	mutex_lock(&dev->struct_mutex);
+	obj = armada_gem_object_lookup(dev, file, handle);
+	if (!obj) {
+		DRM_ERROR("failed to lookup gem object\n");
+		ret = -EINVAL;
+		goto err_unlock;
+	}
+
+	/* Don't allow imported objects to be mapped */
+	if (obj->obj.import_attach) {
+		ret = -EINVAL;
+		goto err_unlock;
+	}
+
+	ret = drm_gem_create_mmap_offset(&obj->obj);
+	if (ret == 0) {
+		*offset = drm_vma_node_offset_addr(&obj->obj.vma_node);
+		DRM_DEBUG_DRIVER("handle %#x offset %llx\n", handle, *offset);
+	}
+
+	drm_gem_object_unreference(&obj->obj);
+ err_unlock:
+	mutex_unlock(&dev->struct_mutex);
+
+	return ret;
+}
+
+int armada_gem_dumb_destroy(struct drm_file *file, struct drm_device *dev,
+	uint32_t handle)
+{
+	return drm_gem_handle_delete(file, handle);
+}
+
+/* Private driver gem ioctls */
+int armada_gem_create_ioctl(struct drm_device *dev, void *data,
+	struct drm_file *file)
+{
+	struct drm_armada_gem_create *args = data;
+	struct armada_gem_object *dobj;
+	size_t size;
+	u32 handle;
+	int ret;
+
+	if (args->size == 0)
+		return -ENOMEM;
+
+	size = args->size;
+
+	dobj = armada_gem_alloc_object(dev, size);
+	if (dobj == NULL)
+		return -ENOMEM;
+
+	ret = drm_gem_handle_create(file, &dobj->obj, &handle);
+	if (ret)
+		goto err;
+
+	args->handle = handle;
+
+	/* drop reference from allocate - handle holds it now */
+	DRM_DEBUG_DRIVER("obj %p size %zu handle %#x\n", dobj, size, handle);
+ err:
+	drm_gem_object_unreference_unlocked(&dobj->obj);
+	return ret;
+}
+
+/* Map a shmem-backed object into process memory space */
+int armada_gem_mmap_ioctl(struct drm_device *dev, void *data,
+	struct drm_file *file)
+{
+	struct drm_armada_gem_mmap *args = data;
+	struct armada_gem_object *dobj;
+	unsigned long addr;
+
+	dobj = armada_gem_object_lookup(dev, file, args->handle);
+	if (dobj == NULL)
+		return -ENOENT;
+
+	if (!dobj->obj.filp) {
+		drm_gem_object_unreference(&dobj->obj);
+		return -EINVAL;
+	}
+
+	addr = vm_mmap(dobj->obj.filp, 0, args->size, PROT_READ | PROT_WRITE,
+		       MAP_SHARED, args->offset);
+	drm_gem_object_unreference(&dobj->obj);
+	if (IS_ERR_VALUE(addr))
+		return addr;
+
+	args->addr = addr;
+
+	return 0;
+}
+
+int armada_gem_pwrite_ioctl(struct drm_device *dev, void *data,
+	struct drm_file *file)
+{
+	struct drm_armada_gem_pwrite *args = data;
+	struct armada_gem_object *dobj;
+	char __user *ptr;
+	int ret;
+
+	DRM_DEBUG_DRIVER("handle %u off %u size %u ptr 0x%llx\n",
+		args->handle, args->offset, args->size, args->ptr);
+
+	if (args->size == 0)
+		return 0;
+
+	ptr = (char __user *)(uintptr_t)args->ptr;
+
+	if (!access_ok(VERIFY_READ, ptr, args->size))
+		return -EFAULT;
+
+	ret = fault_in_multipages_readable(ptr, args->size);
+	if (ret)
+		return ret;
+
+	dobj = armada_gem_object_lookup(dev, file, args->handle);
+	if (dobj == NULL)
+		return -ENOENT;
+
+	/* Must be a kernel-mapped object */
+	if (!dobj->addr)
+		return -EINVAL;
+
+	if (args->offset > dobj->obj.size ||
+	    args->size > dobj->obj.size - args->offset) {
+		DRM_ERROR("invalid size: object size %u\n", dobj->obj.size);
+		ret = -EINVAL;
+		goto unref;
+	}
+
+	if (copy_from_user(dobj->addr + args->offset, ptr, args->size)) {
+		ret = -EFAULT;
+	} else if (dobj->update) {
+		dobj->update(dobj->update_data);
+		ret = 0;
+	}
+
+ unref:
+	drm_gem_object_unreference_unlocked(&dobj->obj);
+	return ret;
+}
+
+/* Prime support */
+struct sg_table *
+armada_gem_prime_map_dma_buf(struct dma_buf_attachment *attach,
+	enum dma_data_direction dir)
+{
+	struct drm_gem_object *obj = attach->dmabuf->priv;
+	struct armada_gem_object *dobj = drm_to_armada_gem(obj);
+	struct scatterlist *sg;
+	struct sg_table *sgt;
+	int i, num;
+
+	sgt = kmalloc(sizeof(*sgt), GFP_KERNEL);
+	if (!sgt)
+		return NULL;
+
+	if (dobj->obj.filp) {
+		struct address_space *mapping;
+		gfp_t gfp;
+		int count;
+
+		count = dobj->obj.size / PAGE_SIZE;
+		if (sg_alloc_table(sgt, count, GFP_KERNEL))
+			goto free_sgt;
+
+		mapping = file_inode(dobj->obj.filp)->i_mapping;
+		gfp = mapping_gfp_mask(mapping);
+
+		for_each_sg(sgt->sgl, sg, count, i) {
+			struct page *page;
+
+			page = shmem_read_mapping_page_gfp(mapping, i, gfp);
+			if (IS_ERR(page)) {
+				num = i;
+				goto release;
+			}
+
+			sg_set_page(sg, page, PAGE_SIZE, 0);
+		}
+
+		if (dma_map_sg(attach->dev, sgt->sgl, sgt->nents, dir) == 0) {
+			num = sgt->nents;
+			goto release;
+		}
+	} else if (dobj->page) {
+		/* Single contiguous page */
+		if (sg_alloc_table(sgt, 1, GFP_KERNEL))
+			goto free_sgt;
+
+		sg_set_page(sgt->sgl, dobj->page, dobj->obj.size, 0);
+
+		if (dma_map_sg(attach->dev, sgt->sgl, sgt->nents, dir) == 0)
+			goto free_table;
+	} else if (dobj->linear) {
+		/* Single contiguous physical region - no struct page */
+		if (sg_alloc_table(sgt, 1, GFP_KERNEL))
+			goto free_sgt;
+		sg_dma_address(sgt->sgl) = dobj->dev_addr;
+		sg_dma_len(sgt->sgl) = dobj->obj.size;
+	} else {
+		goto free_sgt;
+	}
+	return sgt;
+
+ release:
+	for_each_sg(sgt->sgl, sg, num, i)
+		page_cache_release(sg_page(sg));
+ free_table:
+	sg_free_table(sgt);
+ free_sgt:
+	kfree(sgt);
+	return NULL;
+}
+
+static void armada_gem_prime_unmap_dma_buf(struct dma_buf_attachment *attach,
+	struct sg_table *sgt, enum dma_data_direction dir)
+{
+	struct drm_gem_object *obj = attach->dmabuf->priv;
+	struct armada_gem_object *dobj = drm_to_armada_gem(obj);
+	int i;
+
+	if (!dobj->linear)
+		dma_unmap_sg(attach->dev, sgt->sgl, sgt->nents, dir);
+
+	if (dobj->obj.filp) {
+		struct scatterlist *sg;
+		for_each_sg(sgt->sgl, sg, sgt->nents, i)
+			page_cache_release(sg_page(sg));
+	}
+
+	sg_free_table(sgt);
+	kfree(sgt);
+}
+
+static void *armada_gem_dmabuf_no_kmap(struct dma_buf *buf, unsigned long n)
+{
+	return NULL;
+}
+
+static void
+armada_gem_dmabuf_no_kunmap(struct dma_buf *buf, unsigned long n, void *addr)
+{
+}
+
+static int
+armada_gem_dmabuf_mmap(struct dma_buf *buf, struct vm_area_struct *vma)
+{
+	return -EINVAL;
+}
+
+static const struct dma_buf_ops armada_gem_prime_dmabuf_ops = {
+	.map_dma_buf	= armada_gem_prime_map_dma_buf,
+	.unmap_dma_buf	= armada_gem_prime_unmap_dma_buf,
+	.release	= drm_gem_dmabuf_release,
+	.kmap_atomic	= armada_gem_dmabuf_no_kmap,
+	.kunmap_atomic	= armada_gem_dmabuf_no_kunmap,
+	.kmap		= armada_gem_dmabuf_no_kmap,
+	.kunmap		= armada_gem_dmabuf_no_kunmap,
+	.mmap		= armada_gem_dmabuf_mmap,
+};
+
+struct dma_buf *
+armada_gem_prime_export(struct drm_device *dev, struct drm_gem_object *obj,
+	int flags)
+{
+	return dma_buf_export(obj, &armada_gem_prime_dmabuf_ops, obj->size,
+			      O_RDWR);
+}
+
+struct drm_gem_object *
+armada_gem_prime_import(struct drm_device *dev, struct dma_buf *buf)
+{
+	struct dma_buf_attachment *attach;
+	struct armada_gem_object *dobj;
+
+	if (buf->ops == &armada_gem_prime_dmabuf_ops) {
+		struct drm_gem_object *obj = buf->priv;
+		if (obj->dev == dev) {
+			/*
+			 * Importing our own dmabuf(s) increases the
+			 * refcount on the gem object itself.
+			 */
+			drm_gem_object_reference(obj);
+			dma_buf_put(buf);
+			return obj;
+		}
+	}
+
+	attach = dma_buf_attach(buf, dev->dev);
+	if (IS_ERR(attach))
+		return ERR_CAST(attach);
+
+	dobj = armada_gem_alloc_private_object(dev, buf->size);
+	if (!dobj) {
+		dma_buf_detach(buf, attach);
+		return ERR_PTR(-ENOMEM);
+	}
+
+	dobj->obj.import_attach = attach;
+
+	/*
+	 * Don't call dma_buf_map_attachment() here - it maps the
+	 * scatterlist immediately for DMA, and this is not always
+	 * an appropriate thing to do.
+	 */
+	return &dobj->obj;
+}
+
+int armada_gem_map_import(struct armada_gem_object *dobj)
+{
+	int ret;
+
+	dobj->sgt = dma_buf_map_attachment(dobj->obj.import_attach,
+					  DMA_TO_DEVICE);
+	if (!dobj->sgt) {
+		DRM_ERROR("dma_buf_map_attachment() returned NULL\n");
+		return -EINVAL;
+	}
+	if (IS_ERR(dobj->sgt)) {
+		ret = PTR_ERR(dobj->sgt);
+		dobj->sgt = NULL;
+		DRM_ERROR("dma_buf_map_attachment() error: %d\n", ret);
+		return ret;
+	}
+	if (dobj->sgt->nents > 1) {
+		DRM_ERROR("dma_buf_map_attachment() returned an (unsupported) scattered list\n");
+		return -EINVAL;
+	}
+	if (sg_dma_len(dobj->sgt->sgl) < dobj->obj.size) {
+		DRM_ERROR("dma_buf_map_attachment() returned a small buffer\n");
+		return -EINVAL;
+	}
+	dobj->dev_addr = sg_dma_address(dobj->sgt->sgl);
+	return 0;
+}
diff --git a/drivers/gpu/drm/armada/armada_gem.h b/drivers/gpu/drm/armada/armada_gem.h
new file mode 100644
index 000000000000..00b6cd461a03
--- /dev/null
+++ b/drivers/gpu/drm/armada/armada_gem.h
@@ -0,0 +1,52 @@
+/*
+ * Copyright (C) 2012 Russell King
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ */
+#ifndef ARMADA_GEM_H
+#define ARMADA_GEM_H
+
+/* GEM */
+struct armada_gem_object {
+	struct drm_gem_object	obj;
+	void			*addr;
+	phys_addr_t		phys_addr;
+	resource_size_t		dev_addr;
+	struct drm_mm_node	*linear;	/* for linear backed */
+	struct page		*page;		/* for page backed */
+	struct sg_table		*sgt;		/* for imported */
+	void			(*update)(void *);
+	void			*update_data;
+};
+
+extern const struct vm_operations_struct armada_gem_vm_ops;
+
+#define drm_to_armada_gem(o) container_of(o, struct armada_gem_object, obj)
+
+void armada_gem_free_object(struct drm_gem_object *);
+int armada_gem_linear_back(struct drm_device *, struct armada_gem_object *);
+void *armada_gem_map_object(struct drm_device *, struct armada_gem_object *);
+struct armada_gem_object *armada_gem_alloc_private_object(struct drm_device *,
+	size_t);
+int armada_gem_dumb_create(struct drm_file *, struct drm_device *,
+	struct drm_mode_create_dumb *);
+int armada_gem_dumb_map_offset(struct drm_file *, struct drm_device *,
+	uint32_t, uint64_t *);
+int armada_gem_dumb_destroy(struct drm_file *, struct drm_device *,
+	uint32_t);
+struct dma_buf *armada_gem_prime_export(struct drm_device *dev,
+	struct drm_gem_object *obj, int flags);
+struct drm_gem_object *armada_gem_prime_import(struct drm_device *,
+	struct dma_buf *);
+int armada_gem_map_import(struct armada_gem_object *);
+
+static inline struct armada_gem_object *armada_gem_object_lookup(
+	struct drm_device *dev, struct drm_file *dfile, unsigned handle)
+{
+	struct drm_gem_object *obj = drm_gem_object_lookup(dev, dfile, handle);
+
+	return obj ? drm_to_armada_gem(obj) : NULL;
+}
+#endif
diff --git a/drivers/gpu/drm/armada/armada_hw.h b/drivers/gpu/drm/armada/armada_hw.h
new file mode 100644
index 000000000000..27319a8335e2
--- /dev/null
+++ b/drivers/gpu/drm/armada/armada_hw.h
@@ -0,0 +1,318 @@
+/*
+ * Copyright (C) 2012 Russell King
+ *  Rewritten from the dovefb driver, and Armada510 manuals.
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ */
+#ifndef ARMADA_HW_H
+#define ARMADA_HW_H
+
+/*
+ * Note: the following registers are written from IRQ context:
+ *  LCD_SPU_V_PORCH, LCD_SPU_ADV_REG, LCD_SPUT_V_H_TOTAL
+ *  LCD_SPU_DMA_START_ADDR_[YUV][01], LCD_SPU_DMA_PITCH_YC,
+ *  LCD_SPU_DMA_PITCH_UV, LCD_SPU_DMA_OVSA_HPXL_VLN,
+ *  LCD_SPU_DMA_HPXL_VLN, LCD_SPU_DZM_HPXL_VLN, LCD_SPU_DMA_CTRL0
+ */
+enum {
+	LCD_SPU_ADV_REG			= 0x0084,	/* Armada 510 */
+	LCD_SPU_DMA_START_ADDR_Y0	= 0x00c0,
+	LCD_SPU_DMA_START_ADDR_U0	= 0x00c4,
+	LCD_SPU_DMA_START_ADDR_V0	= 0x00c8,
+	LCD_CFG_DMA_START_ADDR_0	= 0x00cc,
+	LCD_SPU_DMA_START_ADDR_Y1	= 0x00d0,
+	LCD_SPU_DMA_START_ADDR_U1	= 0x00d4,
+	LCD_SPU_DMA_START_ADDR_V1	= 0x00d8,
+	LCD_CFG_DMA_START_ADDR_1	= 0x00dc,
+	LCD_SPU_DMA_PITCH_YC		= 0x00e0,
+	LCD_SPU_DMA_PITCH_UV		= 0x00e4,
+	LCD_SPU_DMA_OVSA_HPXL_VLN	= 0x00e8,
+	LCD_SPU_DMA_HPXL_VLN		= 0x00ec,
+	LCD_SPU_DZM_HPXL_VLN		= 0x00f0,
+	LCD_CFG_GRA_START_ADDR0		= 0x00f4,
+	LCD_CFG_GRA_START_ADDR1		= 0x00f8,
+	LCD_CFG_GRA_PITCH		= 0x00fc,
+	LCD_SPU_GRA_OVSA_HPXL_VLN	= 0x0100,
+	LCD_SPU_GRA_HPXL_VLN		= 0x0104,
+	LCD_SPU_GZM_HPXL_VLN		= 0x0108,
+	LCD_SPU_HWC_OVSA_HPXL_VLN	= 0x010c,
+	LCD_SPU_HWC_HPXL_VLN		= 0x0110,
+	LCD_SPUT_V_H_TOTAL		= 0x0114,
+	LCD_SPU_V_H_ACTIVE		= 0x0118,
+	LCD_SPU_H_PORCH			= 0x011c,
+	LCD_SPU_V_PORCH			= 0x0120,
+	LCD_SPU_BLANKCOLOR		= 0x0124,
+	LCD_SPU_ALPHA_COLOR1		= 0x0128,
+	LCD_SPU_ALPHA_COLOR2		= 0x012c,
+	LCD_SPU_COLORKEY_Y		= 0x0130,
+	LCD_SPU_COLORKEY_U		= 0x0134,
+	LCD_SPU_COLORKEY_V		= 0x0138,
+	LCD_CFG_RDREG4F			= 0x013c,	/* Armada 510 */
+	LCD_SPU_SPI_RXDATA		= 0x0140,
+	LCD_SPU_ISA_RXDATA		= 0x0144,
+	LCD_SPU_HWC_RDDAT		= 0x0158,
+	LCD_SPU_GAMMA_RDDAT		= 0x015c,
+	LCD_SPU_PALETTE_RDDAT		= 0x0160,
+	LCD_SPU_IOPAD_IN		= 0x0178,
+	LCD_CFG_RDREG5F			= 0x017c,
+	LCD_SPU_SPI_CTRL		= 0x0180,
+	LCD_SPU_SPI_TXDATA		= 0x0184,
+	LCD_SPU_SMPN_CTRL		= 0x0188,
+	LCD_SPU_DMA_CTRL0		= 0x0190,
+	LCD_SPU_DMA_CTRL1		= 0x0194,
+	LCD_SPU_SRAM_CTRL		= 0x0198,
+	LCD_SPU_SRAM_WRDAT		= 0x019c,
+	LCD_SPU_SRAM_PARA0		= 0x01a0,	/* Armada 510 */
+	LCD_SPU_SRAM_PARA1		= 0x01a4,
+	LCD_CFG_SCLK_DIV		= 0x01a8,
+	LCD_SPU_CONTRAST		= 0x01ac,
+	LCD_SPU_SATURATION		= 0x01b0,
+	LCD_SPU_CBSH_HUE		= 0x01b4,
+	LCD_SPU_DUMB_CTRL		= 0x01b8,
+	LCD_SPU_IOPAD_CONTROL		= 0x01bc,
+	LCD_SPU_IRQ_ENA			= 0x01c0,
+	LCD_SPU_IRQ_ISR			= 0x01c4,
+};
+
+/* For LCD_SPU_ADV_REG */
+enum {
+	ADV_VSYNC_L_OFF	= 0xfff << 20,
+	ADV_GRACOLORKEY	= 1 << 19,
+	ADV_VIDCOLORKEY	= 1 << 18,
+	ADV_HWC32BLEND	= 1 << 15,
+	ADV_HWC32ARGB	= 1 << 14,
+	ADV_HWC32ENABLE	= 1 << 13,
+	ADV_VSYNCOFFEN	= 1 << 12,
+	ADV_VSYNC_H_OFF	= 0xfff << 0,
+};
+
+enum {
+	CFG_565		= 0,
+	CFG_1555	= 1,
+	CFG_888PACK	= 2,
+	CFG_X888	= 3,
+	CFG_8888	= 4,
+	CFG_422PACK	= 5,
+	CFG_422		= 6,
+	CFG_420		= 7,
+	CFG_PSEUDO4	= 9,
+	CFG_PSEUDO8	= 10,
+	CFG_SWAPRB	= 1 << 4,
+	CFG_SWAPUV	= 1 << 3,
+	CFG_SWAPYU	= 1 << 2,
+	CFG_YUV2RGB	= 1 << 1,
+};
+
+/* For LCD_SPU_DMA_CTRL0 */
+enum {
+	CFG_NOBLENDING	= 1 << 31,
+	CFG_GAMMA_ENA	= 1 << 30,
+	CFG_CBSH_ENA	= 1 << 29,
+	CFG_PALETTE_ENA	= 1 << 28,
+	CFG_ARBFAST_ENA	= 1 << 27,
+	CFG_HWC_1BITMOD	= 1 << 26,
+	CFG_HWC_1BITENA	= 1 << 25,
+	CFG_HWC_ENA	= 1 << 24,
+	CFG_DMAFORMAT	= 0xf << 20,
+#define	CFG_DMA_FMT(x)	((x) << 20)
+	CFG_GRAFORMAT	= 0xf << 16,
+#define	CFG_GRA_FMT(x)	((x) << 16)
+#define CFG_GRA_MOD(x)	((x) << 8)
+	CFG_GRA_FTOGGLE	= 1 << 15,
+	CFG_GRA_HSMOOTH	= 1 << 14,
+	CFG_GRA_TSTMODE	= 1 << 13,
+	CFG_GRA_ENA	= 1 << 8,
+#define CFG_DMA_MOD(x)	((x) << 0)
+	CFG_DMA_FTOGGLE	= 1 << 7,
+	CFG_DMA_HSMOOTH	= 1 << 6,
+	CFG_DMA_TSTMODE	= 1 << 5,
+	CFG_DMA_ENA	= 1 << 0,
+};
+
+enum {
+	CKMODE_DISABLE	= 0,
+	CKMODE_Y	= 1,
+	CKMODE_U	= 2,
+	CKMODE_RGB	= 3,
+	CKMODE_V	= 4,
+	CKMODE_R	= 5,
+	CKMODE_G	= 6,
+	CKMODE_B	= 7,
+};
+
+/* For LCD_SPU_DMA_CTRL1 */
+enum {
+	CFG_FRAME_TRIG		= 1 << 31,
+	CFG_VSYNC_INV		= 1 << 27,
+	CFG_CKMODE_MASK		= 0x7 << 24,
+#define CFG_CKMODE(x)		((x) << 24)
+	CFG_CARRY		= 1 << 23,
+	CFG_GATED_CLK		= 1 << 21,
+	CFG_PWRDN_ENA		= 1 << 20,
+	CFG_DSCALE_MASK		= 0x3 << 18,
+	CFG_DSCALE_NONE		= 0x0 << 18,
+	CFG_DSCALE_HALF		= 0x1 << 18,
+	CFG_DSCALE_QUAR		= 0x2 << 18,
+	CFG_ALPHAM_MASK		= 0x3 << 16,
+	CFG_ALPHAM_VIDEO	= 0x0 << 16,
+	CFG_ALPHAM_GRA		= 0x1 << 16,
+	CFG_ALPHAM_CFG		= 0x2 << 16,
+	CFG_ALPHA_MASK		= 0xff << 8,
+	CFG_PIXCMD_MASK		= 0xff,
+};
+
+/* For LCD_SPU_SRAM_CTRL */
+enum {
+	SRAM_READ	= 0 << 14,
+	SRAM_WRITE	= 2 << 14,
+	SRAM_INIT	= 3 << 14,
+	SRAM_HWC32_RAM1	= 0xc << 8,
+	SRAM_HWC32_RAM2	= 0xd << 8,
+	SRAM_HWC32_RAMR	= SRAM_HWC32_RAM1,
+	SRAM_HWC32_RAMG	= SRAM_HWC32_RAM2,
+	SRAM_HWC32_RAMB	= 0xe << 8,
+	SRAM_HWC32_TRAN	= 0xf << 8,
+	SRAM_HWC	= 0xf << 8,
+};
+
+/* For LCD_SPU_SRAM_PARA1 */
+enum {
+	CFG_CSB_256x32	= 1 << 15,	/* cursor */
+	CFG_CSB_256x24	= 1 << 14,	/* palette */
+	CFG_CSB_256x8	= 1 << 13,	/* gamma */
+	CFG_PDWN1920x32	= 1 << 8,	/* Armada 510: power down vscale ram */
+	CFG_PDWN256x32	= 1 << 7,	/* power down cursor */
+	CFG_PDWN256x24	= 1 << 6,	/* power down palette */
+	CFG_PDWN256x8	= 1 << 5,	/* power down gamma */
+	CFG_PDWNHWC	= 1 << 4,	/* Armada 510: power down all hwc ram */
+	CFG_PDWN32x32	= 1 << 3,	/* power down slave->smart ram */
+	CFG_PDWN16x66	= 1 << 2,	/* power down UV fifo */
+	CFG_PDWN32x66	= 1 << 1,	/* power down Y fifo */
+	CFG_PDWN64x66	= 1 << 0,	/* power down graphic fifo */
+};
+
+/* For LCD_CFG_SCLK_DIV */
+enum {
+	/* Armada 510 */
+	SCLK_510_AXI		= 0x0 << 30,
+	SCLK_510_EXTCLK0	= 0x1 << 30,
+	SCLK_510_PLL		= 0x2 << 30,
+	SCLK_510_EXTCLK1	= 0x3 << 30,
+	SCLK_510_DIV_CHANGE	= 1 << 29,
+	SCLK_510_FRAC_DIV_MASK	= 0xfff << 16,
+	SCLK_510_INT_DIV_MASK	= 0xffff << 0,
+
+	/* Armada 16x */
+	SCLK_16X_AHB		= 0x0 << 28,
+	SCLK_16X_PCLK		= 0x1 << 28,
+	SCLK_16X_AXI		= 0x4 << 28,
+	SCLK_16X_PLL		= 0x8 << 28,
+	SCLK_16X_FRAC_DIV_MASK	= 0xfff << 16,
+	SCLK_16X_INT_DIV_MASK	= 0xffff << 0,
+};
+
+/* For LCD_SPU_DUMB_CTRL */
+enum {
+	DUMB16_RGB565_0	= 0x0 << 28,
+	DUMB16_RGB565_1	= 0x1 << 28,
+	DUMB18_RGB666_0	= 0x2 << 28,
+	DUMB18_RGB666_1	= 0x3 << 28,
+	DUMB12_RGB444_0	= 0x4 << 28,
+	DUMB12_RGB444_1	= 0x5 << 28,
+	DUMB24_RGB888_0	= 0x6 << 28,
+	DUMB_BLANK	= 0x7 << 28,
+	DUMB_MASK	= 0xf << 28,
+	CFG_BIAS_OUT	= 1 << 8,
+	CFG_REV_RGB	= 1 << 7,
+	CFG_INV_CBLANK	= 1 << 6,
+	CFG_INV_CSYNC	= 1 << 5,	/* Normally active high */
+	CFG_INV_HENA	= 1 << 4,
+	CFG_INV_VSYNC	= 1 << 3,	/* Normally active high */
+	CFG_INV_HSYNC	= 1 << 2,	/* Normally active high */
+	CFG_INV_PCLK	= 1 << 1,
+	CFG_DUMB_ENA	= 1 << 0,
+};
+
+/* For LCD_SPU_IOPAD_CONTROL */
+enum {
+	CFG_VSCALE_LN_EN	= 3 << 18,
+	CFG_GRA_VM_ENA		= 1 << 15,
+	CFG_DMA_VM_ENA		= 1 << 13,
+	CFG_CMD_VM_ENA		= 1 << 11,
+	CFG_CSC_MASK		= 3 << 8,
+	CFG_CSC_YUV_CCIR709	= 1 << 9,
+	CFG_CSC_YUV_CCIR601	= 0 << 9,
+	CFG_CSC_RGB_STUDIO	= 1 << 8,
+	CFG_CSC_RGB_COMPUTER	= 0 << 8,
+	CFG_IOPAD_MASK		= 0xf << 0,
+	CFG_IOPAD_DUMB24	= 0x0 << 0,
+	CFG_IOPAD_DUMB18SPI	= 0x1 << 0,
+	CFG_IOPAD_DUMB18GPIO	= 0x2 << 0,
+	CFG_IOPAD_DUMB16SPI	= 0x3 << 0,
+	CFG_IOPAD_DUMB16GPIO	= 0x4 << 0,
+	CFG_IOPAD_DUMB12GPIO	= 0x5 << 0,
+	CFG_IOPAD_SMART18	= 0x6 << 0,
+	CFG_IOPAD_SMART16	= 0x7 << 0,
+	CFG_IOPAD_SMART8	= 0x8 << 0,
+};
+
+#define IOPAD_DUMB24                0x0
+
+/* For LCD_SPU_IRQ_ENA */
+enum {
+	DMA_FRAME_IRQ0_ENA	= 1 << 31,
+	DMA_FRAME_IRQ1_ENA	= 1 << 30,
+	DMA_FRAME_IRQ_ENA	= DMA_FRAME_IRQ0_ENA | DMA_FRAME_IRQ1_ENA,
+	DMA_FF_UNDERFLOW_ENA	= 1 << 29,
+	GRA_FRAME_IRQ0_ENA	= 1 << 27,
+	GRA_FRAME_IRQ1_ENA	= 1 << 26,
+	GRA_FRAME_IRQ_ENA	= GRA_FRAME_IRQ0_ENA | GRA_FRAME_IRQ1_ENA,
+	GRA_FF_UNDERFLOW_ENA	= 1 << 25,
+	VSYNC_IRQ_ENA		= 1 << 23,
+	DUMB_FRAMEDONE_ENA	= 1 << 22,
+	TWC_FRAMEDONE_ENA	= 1 << 21,
+	HWC_FRAMEDONE_ENA	= 1 << 20,
+	SLV_IRQ_ENA		= 1 << 19,
+	SPI_IRQ_ENA		= 1 << 18,
+	PWRDN_IRQ_ENA		= 1 << 17,
+	ERR_IRQ_ENA		= 1 << 16,
+	CLEAN_SPU_IRQ_ISR	= 0xffff,
+};
+
+/* For LCD_SPU_IRQ_ISR */
+enum {
+	DMA_FRAME_IRQ0		= 1 << 31,
+	DMA_FRAME_IRQ1		= 1 << 30,
+	DMA_FRAME_IRQ		= DMA_FRAME_IRQ0 | DMA_FRAME_IRQ1,
+	DMA_FF_UNDERFLOW	= 1 << 29,
+	GRA_FRAME_IRQ0		= 1 << 27,
+	GRA_FRAME_IRQ1		= 1 << 26,
+	GRA_FRAME_IRQ		= GRA_FRAME_IRQ0 | GRA_FRAME_IRQ1,
+	GRA_FF_UNDERFLOW	= 1 << 25,
+	VSYNC_IRQ		= 1 << 23,
+	DUMB_FRAMEDONE		= 1 << 22,
+	TWC_FRAMEDONE		= 1 << 21,
+	HWC_FRAMEDONE		= 1 << 20,
+	SLV_IRQ			= 1 << 19,
+	SPI_IRQ			= 1 << 18,
+	PWRDN_IRQ		= 1 << 17,
+	ERR_IRQ			= 1 << 16,
+	DMA_FRAME_IRQ0_LEVEL	= 1 << 15,
+	DMA_FRAME_IRQ1_LEVEL	= 1 << 14,
+	DMA_FRAME_CNT_ISR	= 3 << 12,
+	GRA_FRAME_IRQ0_LEVEL	= 1 << 11,
+	GRA_FRAME_IRQ1_LEVEL	= 1 << 10,
+	GRA_FRAME_CNT_ISR	= 3 << 8,
+	VSYNC_IRQ_LEVEL		= 1 << 7,
+	DUMB_FRAMEDONE_LEVEL	= 1 << 6,
+	TWC_FRAMEDONE_LEVEL	= 1 << 5,
+	HWC_FRAMEDONE_LEVEL	= 1 << 4,
+	SLV_FF_EMPTY		= 1 << 3,
+	DMA_FF_ALLEMPTY		= 1 << 2,
+	GRA_FF_ALLEMPTY		= 1 << 1,
+	PWRDN_IRQ_LEVEL		= 1 << 0,
+};
+
+#endif
diff --git a/drivers/gpu/drm/armada/armada_ioctlP.h b/drivers/gpu/drm/armada/armada_ioctlP.h
new file mode 100644
index 000000000000..bd8c4562066c
--- /dev/null
+++ b/drivers/gpu/drm/armada/armada_ioctlP.h
@@ -0,0 +1,18 @@
+/*
+ * Copyright (C) 2012 Russell King
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ */
+#ifndef ARMADA_IOCTLP_H
+#define ARMADA_IOCTLP_H
+
+#define ARMADA_IOCTL_PROTO(name)\
+extern int armada_##name##_ioctl(struct drm_device *, void *, struct drm_file *)
+
+ARMADA_IOCTL_PROTO(gem_create);
+ARMADA_IOCTL_PROTO(gem_mmap);
+ARMADA_IOCTL_PROTO(gem_pwrite);
+
+#endif
diff --git a/drivers/gpu/drm/armada/armada_output.c b/drivers/gpu/drm/armada/armada_output.c
new file mode 100644
index 000000000000..d685a5421485
--- /dev/null
+++ b/drivers/gpu/drm/armada/armada_output.c
@@ -0,0 +1,158 @@
+/*
+ * Copyright (C) 2012 Russell King
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ */
+#include <drm/drmP.h>
+#include <drm/drm_crtc_helper.h>
+#include <drm/drm_edid.h>
+#include <drm/drm_encoder_slave.h>
+#include "armada_output.h"
+#include "armada_drm.h"
+
+struct armada_connector {
+	struct drm_connector conn;
+	const struct armada_output_type *type;
+};
+
+#define drm_to_armada_conn(c) container_of(c, struct armada_connector, conn)
+
+struct drm_encoder *armada_drm_connector_encoder(struct drm_connector *conn)
+{
+	struct drm_encoder *enc = conn->encoder;
+
+	return enc ? enc : drm_encoder_find(conn->dev, conn->encoder_ids[0]);
+}
+
+static enum drm_connector_status armada_drm_connector_detect(
+	struct drm_connector *conn, bool force)
+{
+	struct armada_connector *dconn = drm_to_armada_conn(conn);
+	enum drm_connector_status status = connector_status_disconnected;
+
+	if (dconn->type->detect) {
+		status = dconn->type->detect(conn, force);
+	} else {
+		struct drm_encoder *enc = armada_drm_connector_encoder(conn);
+
+		if (enc)
+			status = encoder_helper_funcs(enc)->detect(enc, conn);
+	}
+
+	return status;
+}
+
+static void armada_drm_connector_destroy(struct drm_connector *conn)
+{
+	struct armada_connector *dconn = drm_to_armada_conn(conn);
+
+	drm_sysfs_connector_remove(conn);
+	drm_connector_cleanup(conn);
+	kfree(dconn);
+}
+
+static int armada_drm_connector_set_property(struct drm_connector *conn,
+	struct drm_property *property, uint64_t value)
+{
+	struct armada_connector *dconn = drm_to_armada_conn(conn);
+
+	if (!dconn->type->set_property)
+		return -EINVAL;
+
+	return dconn->type->set_property(conn, property, value);
+}
+
+static const struct drm_connector_funcs armada_drm_conn_funcs = {
+	.dpms		= drm_helper_connector_dpms,
+	.fill_modes	= drm_helper_probe_single_connector_modes,
+	.detect		= armada_drm_connector_detect,
+	.destroy	= armada_drm_connector_destroy,
+	.set_property	= armada_drm_connector_set_property,
+};
+
+void armada_drm_encoder_prepare(struct drm_encoder *encoder)
+{
+	encoder_helper_funcs(encoder)->dpms(encoder, DRM_MODE_DPMS_OFF);
+}
+
+void armada_drm_encoder_commit(struct drm_encoder *encoder)
+{
+	encoder_helper_funcs(encoder)->dpms(encoder, DRM_MODE_DPMS_ON);
+}
+
+bool armada_drm_encoder_mode_fixup(struct drm_encoder *encoder,
+	const struct drm_display_mode *mode, struct drm_display_mode *adjusted)
+{
+	return true;
+}
+
+/* Shouldn't this be a generic helper function? */
+int armada_drm_slave_encoder_mode_valid(struct drm_connector *conn,
+	struct drm_display_mode *mode)
+{
+	struct drm_encoder *encoder = armada_drm_connector_encoder(conn);
+	int valid = MODE_BAD;
+
+	if (encoder) {
+		struct drm_encoder_slave *slave = to_encoder_slave(encoder);
+
+		valid = slave->slave_funcs->mode_valid(encoder, mode);
+	}
+	return valid;
+}
+
+int armada_drm_slave_encoder_set_property(struct drm_connector *conn,
+	struct drm_property *property, uint64_t value)
+{
+	struct drm_encoder *encoder = armada_drm_connector_encoder(conn);
+	int rc = -EINVAL;
+
+	if (encoder) {
+		struct drm_encoder_slave *slave = to_encoder_slave(encoder);
+
+		rc = slave->slave_funcs->set_property(encoder, conn, property,
+						      value);
+	}
+	return rc;
+}
+
+int armada_output_create(struct drm_device *dev,
+	const struct armada_output_type *type, const void *data)
+{
+	struct armada_connector *dconn;
+	int ret;
+
+	dconn = kzalloc(sizeof(*dconn), GFP_KERNEL);
+	if (!dconn)
+		return -ENOMEM;
+
+	dconn->type = type;
+
+	ret = drm_connector_init(dev, &dconn->conn, &armada_drm_conn_funcs,
+				 type->connector_type);
+	if (ret) {
+		DRM_ERROR("unable to init connector\n");
+		goto err_destroy_dconn;
+	}
+
+	ret = type->create(&dconn->conn, data);
+	if (ret)
+		goto err_conn;
+
+	ret = drm_sysfs_connector_add(&dconn->conn);
+	if (ret)
+		goto err_sysfs;
+
+	return 0;
+
+ err_sysfs:
+	if (dconn->conn.encoder)
+		dconn->conn.encoder->funcs->destroy(dconn->conn.encoder);
+ err_conn:
+	drm_connector_cleanup(&dconn->conn);
+ err_destroy_dconn:
+	kfree(dconn);
+	return ret;
+}
diff --git a/drivers/gpu/drm/armada/armada_output.h b/drivers/gpu/drm/armada/armada_output.h
new file mode 100644
index 000000000000..4126d43b5057
--- /dev/null
+++ b/drivers/gpu/drm/armada/armada_output.h
@@ -0,0 +1,39 @@
+/*
+ * Copyright (C) 2012 Russell King
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ */
+#ifndef ARMADA_CONNETOR_H
+#define ARMADA_CONNETOR_H
+
+#define encoder_helper_funcs(encoder) \
+	((struct drm_encoder_helper_funcs *)encoder->helper_private)
+
+struct armada_output_type {
+	int connector_type;
+	enum drm_connector_status (*detect)(struct drm_connector *, bool);
+	int (*create)(struct drm_connector *, const void *);
+	int (*set_property)(struct drm_connector *, struct drm_property *,
+			    uint64_t);
+};
+
+struct drm_encoder *armada_drm_connector_encoder(struct drm_connector *conn);
+
+void armada_drm_encoder_prepare(struct drm_encoder *encoder);
+void armada_drm_encoder_commit(struct drm_encoder *encoder);
+
+bool armada_drm_encoder_mode_fixup(struct drm_encoder *encoder,
+	const struct drm_display_mode *mode, struct drm_display_mode *adj);
+
+int armada_drm_slave_encoder_mode_valid(struct drm_connector *conn,
+	struct drm_display_mode *mode);
+
+int armada_drm_slave_encoder_set_property(struct drm_connector *conn,
+	struct drm_property *property, uint64_t value);
+
+int armada_output_create(struct drm_device *dev,
+	const struct armada_output_type *type, const void *data);
+
+#endif
diff --git a/drivers/gpu/drm/armada/armada_overlay.c b/drivers/gpu/drm/armada/armada_overlay.c
new file mode 100644
index 000000000000..c5b06fdb459c
--- /dev/null
+++ b/drivers/gpu/drm/armada/armada_overlay.c
@@ -0,0 +1,477 @@
+/*
+ * Copyright (C) 2012 Russell King
+ *  Rewritten from the dovefb driver, and Armada510 manuals.
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ */
+#include <drm/drmP.h>
+#include "armada_crtc.h"
+#include "armada_drm.h"
+#include "armada_fb.h"
+#include "armada_gem.h"
+#include "armada_hw.h"
+#include <drm/armada_drm.h>
+#include "armada_ioctlP.h"
+
+struct armada_plane_properties {
+	uint32_t colorkey_yr;
+	uint32_t colorkey_ug;
+	uint32_t colorkey_vb;
+#define K2R(val) (((val) >> 0) & 0xff)
+#define K2G(val) (((val) >> 8) & 0xff)
+#define K2B(val) (((val) >> 16) & 0xff)
+	int16_t  brightness;
+	uint16_t contrast;
+	uint16_t saturation;
+	uint32_t colorkey_mode;
+};
+
+struct armada_plane {
+	struct drm_plane base;
+	spinlock_t lock;
+	struct drm_framebuffer *old_fb;
+	uint32_t src_hw;
+	uint32_t dst_hw;
+	uint32_t dst_yx;
+	uint32_t ctrl0;
+	struct {
+		struct armada_vbl_event update;
+		struct armada_regs regs[13];
+		wait_queue_head_t wait;
+	} vbl;
+	struct armada_plane_properties prop;
+};
+#define drm_to_armada_plane(p) container_of(p, struct armada_plane, base)
+
+
+static void
+armada_ovl_update_attr(struct armada_plane_properties *prop,
+	struct armada_crtc *dcrtc)
+{
+	writel_relaxed(prop->colorkey_yr, dcrtc->base + LCD_SPU_COLORKEY_Y);
+	writel_relaxed(prop->colorkey_ug, dcrtc->base + LCD_SPU_COLORKEY_U);
+	writel_relaxed(prop->colorkey_vb, dcrtc->base + LCD_SPU_COLORKEY_V);
+
+	writel_relaxed(prop->brightness << 16 | prop->contrast,
+		       dcrtc->base + LCD_SPU_CONTRAST);
+	/* Docs say 15:0, but it seems to actually be 31:16 on Armada 510 */
+	writel_relaxed(prop->saturation << 16,
+		       dcrtc->base + LCD_SPU_SATURATION);
+	writel_relaxed(0x00002000, dcrtc->base + LCD_SPU_CBSH_HUE);
+
+	spin_lock_irq(&dcrtc->irq_lock);
+	armada_updatel(prop->colorkey_mode | CFG_ALPHAM_GRA,
+		     CFG_CKMODE_MASK | CFG_ALPHAM_MASK | CFG_ALPHA_MASK,
+		     dcrtc->base + LCD_SPU_DMA_CTRL1);
+
+	armada_updatel(ADV_GRACOLORKEY, 0, dcrtc->base + LCD_SPU_ADV_REG);
+	spin_unlock_irq(&dcrtc->irq_lock);
+}
+
+/* === Plane support === */
+static void armada_plane_vbl(struct armada_crtc *dcrtc, void *data)
+{
+	struct armada_plane *dplane = data;
+	struct drm_framebuffer *fb;
+
+	armada_drm_crtc_update_regs(dcrtc, dplane->vbl.regs);
+
+	spin_lock(&dplane->lock);
+	fb = dplane->old_fb;
+	dplane->old_fb = NULL;
+	spin_unlock(&dplane->lock);
+
+	if (fb)
+		armada_drm_queue_unref_work(dcrtc->crtc.dev, fb);
+}
+
+static unsigned armada_limit(int start, unsigned size, unsigned max)
+{
+	int end = start + size;
+	if (end < 0)
+		return 0;
+	if (start < 0)
+		start = 0;
+	return (unsigned)end > max ? max - start : end - start;
+}
+
+static int
+armada_plane_update(struct drm_plane *plane, struct drm_crtc *crtc,
+	struct drm_framebuffer *fb,
+	int crtc_x, int crtc_y, unsigned crtc_w, unsigned crtc_h,
+	uint32_t src_x, uint32_t src_y, uint32_t src_w, uint32_t src_h)
+{
+	struct armada_plane *dplane = drm_to_armada_plane(plane);
+	struct armada_crtc *dcrtc = drm_to_armada_crtc(crtc);
+	uint32_t val, ctrl0;
+	unsigned idx = 0;
+	int ret;
+
+	crtc_w = armada_limit(crtc_x, crtc_w, dcrtc->crtc.mode.hdisplay);
+	crtc_h = armada_limit(crtc_y, crtc_h, dcrtc->crtc.mode.vdisplay);
+	ctrl0 = CFG_DMA_FMT(drm_fb_to_armada_fb(fb)->fmt) |
+		CFG_DMA_MOD(drm_fb_to_armada_fb(fb)->mod) |
+		CFG_CBSH_ENA | CFG_DMA_HSMOOTH | CFG_DMA_ENA;
+
+	/* Does the position/size result in nothing to display? */
+	if (crtc_w == 0 || crtc_h == 0) {
+		ctrl0 &= ~CFG_DMA_ENA;
+	}
+
+	/*
+	 * FIXME: if the starting point is off screen, we need to
+	 * adjust src_x, src_y, src_w, src_h appropriately, and
+	 * according to the scale.
+	 */
+
+	if (!dcrtc->plane) {
+		dcrtc->plane = plane;
+		armada_ovl_update_attr(&dplane->prop, dcrtc);
+	}
+
+	/* FIXME: overlay on an interlaced display */
+	/* Just updating the position/size? */
+	if (plane->fb == fb && dplane->ctrl0 == ctrl0) {
+		val = (src_h & 0xffff0000) | src_w >> 16;
+		dplane->src_hw = val;
+		writel_relaxed(val, dcrtc->base + LCD_SPU_DMA_HPXL_VLN);
+		val = crtc_h << 16 | crtc_w;
+		dplane->dst_hw = val;
+		writel_relaxed(val, dcrtc->base + LCD_SPU_DZM_HPXL_VLN);
+		val = crtc_y << 16 | crtc_x;
+		dplane->dst_yx = val;
+		writel_relaxed(val, dcrtc->base + LCD_SPU_DMA_OVSA_HPXL_VLN);
+		return 0;
+	} else if (~dplane->ctrl0 & ctrl0 & CFG_DMA_ENA) {
+		/* Power up the Y/U/V FIFOs on ENA 0->1 transitions */
+		armada_updatel(0, CFG_PDWN16x66 | CFG_PDWN32x66,
+			       dcrtc->base + LCD_SPU_SRAM_PARA1);
+	}
+
+	ret = wait_event_timeout(dplane->vbl.wait,
+				 list_empty(&dplane->vbl.update.node),
+				 HZ/25);
+	if (ret < 0)
+		return ret;
+
+	if (plane->fb != fb) {
+		struct armada_gem_object *obj = drm_fb_obj(fb);
+		uint32_t sy, su, sv;
+
+		/*
+		 * Take a reference on the new framebuffer - we want to
+		 * hold on to it while the hardware is displaying it.
+		 */
+		drm_framebuffer_reference(fb);
+
+		if (plane->fb) {
+			struct drm_framebuffer *older_fb;
+
+			spin_lock_irq(&dplane->lock);
+			older_fb = dplane->old_fb;
+			dplane->old_fb = plane->fb;
+			spin_unlock_irq(&dplane->lock);
+			if (older_fb)
+				armada_drm_queue_unref_work(dcrtc->crtc.dev,
+							    older_fb);
+		}
+
+		src_y >>= 16;
+		src_x >>= 16;
+		sy = obj->dev_addr + fb->offsets[0] + src_y * fb->pitches[0] +
+			src_x * fb->bits_per_pixel / 8;
+		su = obj->dev_addr + fb->offsets[1] + src_y * fb->pitches[1] +
+			src_x;
+		sv = obj->dev_addr + fb->offsets[2] + src_y * fb->pitches[2] +
+			src_x;
+
+		armada_reg_queue_set(dplane->vbl.regs, idx, sy,
+				     LCD_SPU_DMA_START_ADDR_Y0);
+		armada_reg_queue_set(dplane->vbl.regs, idx, su,
+				     LCD_SPU_DMA_START_ADDR_U0);
+		armada_reg_queue_set(dplane->vbl.regs, idx, sv,
+				     LCD_SPU_DMA_START_ADDR_V0);
+		armada_reg_queue_set(dplane->vbl.regs, idx, sy,
+				     LCD_SPU_DMA_START_ADDR_Y1);
+		armada_reg_queue_set(dplane->vbl.regs, idx, su,
+				     LCD_SPU_DMA_START_ADDR_U1);
+		armada_reg_queue_set(dplane->vbl.regs, idx, sv,
+				     LCD_SPU_DMA_START_ADDR_V1);
+
+		val = fb->pitches[0] << 16 | fb->pitches[0];
+		armada_reg_queue_set(dplane->vbl.regs, idx, val,
+				     LCD_SPU_DMA_PITCH_YC);
+		val = fb->pitches[1] << 16 | fb->pitches[2];
+		armada_reg_queue_set(dplane->vbl.regs, idx, val,
+				     LCD_SPU_DMA_PITCH_UV);
+	}
+
+	val = (src_h & 0xffff0000) | src_w >> 16;
+	if (dplane->src_hw != val) {
+		dplane->src_hw = val;
+		armada_reg_queue_set(dplane->vbl.regs, idx, val,
+				     LCD_SPU_DMA_HPXL_VLN);
+	}
+	val = crtc_h << 16 | crtc_w;
+	if (dplane->dst_hw != val) {
+		dplane->dst_hw = val;
+		armada_reg_queue_set(dplane->vbl.regs, idx, val,
+				     LCD_SPU_DZM_HPXL_VLN);
+	}
+	val = crtc_y << 16 | crtc_x;
+	if (dplane->dst_yx != val) {
+		dplane->dst_yx = val;
+		armada_reg_queue_set(dplane->vbl.regs, idx, val,
+				     LCD_SPU_DMA_OVSA_HPXL_VLN);
+	}
+	if (dplane->ctrl0 != ctrl0) {
+		dplane->ctrl0 = ctrl0;
+		armada_reg_queue_mod(dplane->vbl.regs, idx, ctrl0,
+			CFG_CBSH_ENA | CFG_DMAFORMAT | CFG_DMA_FTOGGLE |
+			CFG_DMA_HSMOOTH | CFG_DMA_TSTMODE |
+			CFG_DMA_MOD(CFG_SWAPRB | CFG_SWAPUV | CFG_SWAPYU |
+			CFG_YUV2RGB) | CFG_DMA_ENA,
+			LCD_SPU_DMA_CTRL0);
+	}
+	if (idx) {
+		armada_reg_queue_end(dplane->vbl.regs, idx);
+		armada_drm_vbl_event_add(dcrtc, &dplane->vbl.update);
+	}
+	return 0;
+}
+
+static int armada_plane_disable(struct drm_plane *plane)
+{
+	struct armada_plane *dplane = drm_to_armada_plane(plane);
+	struct drm_framebuffer *fb;
+	struct armada_crtc *dcrtc;
+
+	if (!dplane->base.crtc)
+		return 0;
+
+	dcrtc = drm_to_armada_crtc(dplane->base.crtc);
+	dcrtc->plane = NULL;
+
+	spin_lock_irq(&dcrtc->irq_lock);
+	armada_drm_vbl_event_remove(dcrtc, &dplane->vbl.update);
+	armada_updatel(0, CFG_DMA_ENA, dcrtc->base + LCD_SPU_DMA_CTRL0);
+	dplane->ctrl0 = 0;
+	spin_unlock_irq(&dcrtc->irq_lock);
+
+	/* Power down the Y/U/V FIFOs */
+	armada_updatel(CFG_PDWN16x66 | CFG_PDWN32x66, 0,
+		       dcrtc->base + LCD_SPU_SRAM_PARA1);
+
+	if (plane->fb)
+		drm_framebuffer_unreference(plane->fb);
+
+	spin_lock_irq(&dplane->lock);
+	fb = dplane->old_fb;
+	dplane->old_fb = NULL;
+	spin_unlock_irq(&dplane->lock);
+	if (fb)
+		drm_framebuffer_unreference(fb);
+
+	return 0;
+}
+
+static void armada_plane_destroy(struct drm_plane *plane)
+{
+	kfree(plane);
+}
+
+static int armada_plane_set_property(struct drm_plane *plane,
+	struct drm_property *property, uint64_t val)
+{
+	struct armada_private *priv = plane->dev->dev_private;
+	struct armada_plane *dplane = drm_to_armada_plane(plane);
+	bool update_attr = false;
+
+	if (property == priv->colorkey_prop) {
+#define CCC(v) ((v) << 24 | (v) << 16 | (v) << 8)
+		dplane->prop.colorkey_yr = CCC(K2R(val));
+		dplane->prop.colorkey_ug = CCC(K2G(val));
+		dplane->prop.colorkey_vb = CCC(K2B(val));
+#undef CCC
+		update_attr = true;
+	} else if (property == priv->colorkey_min_prop) {
+		dplane->prop.colorkey_yr &= ~0x00ff0000;
+		dplane->prop.colorkey_yr |= K2R(val) << 16;
+		dplane->prop.colorkey_ug &= ~0x00ff0000;
+		dplane->prop.colorkey_ug |= K2G(val) << 16;
+		dplane->prop.colorkey_vb &= ~0x00ff0000;
+		dplane->prop.colorkey_vb |= K2B(val) << 16;
+		update_attr = true;
+	} else if (property == priv->colorkey_max_prop) {
+		dplane->prop.colorkey_yr &= ~0xff000000;
+		dplane->prop.colorkey_yr |= K2R(val) << 24;
+		dplane->prop.colorkey_ug &= ~0xff000000;
+		dplane->prop.colorkey_ug |= K2G(val) << 24;
+		dplane->prop.colorkey_vb &= ~0xff000000;
+		dplane->prop.colorkey_vb |= K2B(val) << 24;
+		update_attr = true;
+	} else if (property == priv->colorkey_val_prop) {
+		dplane->prop.colorkey_yr &= ~0x0000ff00;
+		dplane->prop.colorkey_yr |= K2R(val) << 8;
+		dplane->prop.colorkey_ug &= ~0x0000ff00;
+		dplane->prop.colorkey_ug |= K2G(val) << 8;
+		dplane->prop.colorkey_vb &= ~0x0000ff00;
+		dplane->prop.colorkey_vb |= K2B(val) << 8;
+		update_attr = true;
+	} else if (property == priv->colorkey_alpha_prop) {
+		dplane->prop.colorkey_yr &= ~0x000000ff;
+		dplane->prop.colorkey_yr |= K2R(val);
+		dplane->prop.colorkey_ug &= ~0x000000ff;
+		dplane->prop.colorkey_ug |= K2G(val);
+		dplane->prop.colorkey_vb &= ~0x000000ff;
+		dplane->prop.colorkey_vb |= K2B(val);
+		update_attr = true;
+	} else if (property == priv->colorkey_mode_prop) {
+		dplane->prop.colorkey_mode &= ~CFG_CKMODE_MASK;
+		dplane->prop.colorkey_mode |= CFG_CKMODE(val);
+		update_attr = true;
+	} else if (property == priv->brightness_prop) {
+		dplane->prop.brightness = val - 256;
+		update_attr = true;
+	} else if (property == priv->contrast_prop) {
+		dplane->prop.contrast = val;
+		update_attr = true;
+	} else if (property == priv->saturation_prop) {
+		dplane->prop.saturation = val;
+		update_attr = true;
+	}
+
+	if (update_attr && dplane->base.crtc)
+		armada_ovl_update_attr(&dplane->prop,
+				       drm_to_armada_crtc(dplane->base.crtc));
+
+	return 0;
+}
+
+static const struct drm_plane_funcs armada_plane_funcs = {
+	.update_plane	= armada_plane_update,
+	.disable_plane	= armada_plane_disable,
+	.destroy	= armada_plane_destroy,
+	.set_property	= armada_plane_set_property,
+};
+
+static const uint32_t armada_formats[] = {
+	DRM_FORMAT_UYVY,
+	DRM_FORMAT_YUYV,
+	DRM_FORMAT_YUV420,
+	DRM_FORMAT_YVU420,
+	DRM_FORMAT_YUV422,
+	DRM_FORMAT_YVU422,
+	DRM_FORMAT_VYUY,
+	DRM_FORMAT_YVYU,
+	DRM_FORMAT_ARGB8888,
+	DRM_FORMAT_ABGR8888,
+	DRM_FORMAT_XRGB8888,
+	DRM_FORMAT_XBGR8888,
+	DRM_FORMAT_RGB888,
+	DRM_FORMAT_BGR888,
+	DRM_FORMAT_ARGB1555,
+	DRM_FORMAT_ABGR1555,
+	DRM_FORMAT_RGB565,
+	DRM_FORMAT_BGR565,
+};
+
+static struct drm_prop_enum_list armada_drm_colorkey_enum_list[] = {
+	{ CKMODE_DISABLE, "disabled" },
+	{ CKMODE_Y,       "Y component" },
+	{ CKMODE_U,       "U component" },
+	{ CKMODE_V,       "V component" },
+	{ CKMODE_RGB,     "RGB" },
+	{ CKMODE_R,       "R component" },
+	{ CKMODE_G,       "G component" },
+	{ CKMODE_B,       "B component" },
+};
+
+static int armada_overlay_create_properties(struct drm_device *dev)
+{
+	struct armada_private *priv = dev->dev_private;
+
+	if (priv->colorkey_prop)
+		return 0;
+
+	priv->colorkey_prop = drm_property_create_range(dev, 0,
+				"colorkey", 0, 0xffffff);
+	priv->colorkey_min_prop = drm_property_create_range(dev, 0,
+				"colorkey_min", 0, 0xffffff);
+	priv->colorkey_max_prop = drm_property_create_range(dev, 0,
+				"colorkey_max", 0, 0xffffff);
+	priv->colorkey_val_prop = drm_property_create_range(dev, 0,
+				"colorkey_val", 0, 0xffffff);
+	priv->colorkey_alpha_prop = drm_property_create_range(dev, 0,
+				"colorkey_alpha", 0, 0xffffff);
+	priv->colorkey_mode_prop = drm_property_create_enum(dev, 0,
+				"colorkey_mode",
+				armada_drm_colorkey_enum_list,
+				ARRAY_SIZE(armada_drm_colorkey_enum_list));
+	priv->brightness_prop = drm_property_create_range(dev, 0,
+				"brightness", 0, 256 + 255);
+	priv->contrast_prop = drm_property_create_range(dev, 0,
+				"contrast", 0, 0x7fff);
+	priv->saturation_prop = drm_property_create_range(dev, 0,
+				"saturation", 0, 0x7fff);
+
+	if (!priv->colorkey_prop)
+		return -ENOMEM;
+
+	return 0;
+}
+
+int armada_overlay_plane_create(struct drm_device *dev, unsigned long crtcs)
+{
+	struct armada_private *priv = dev->dev_private;
+	struct drm_mode_object *mobj;
+	struct armada_plane *dplane;
+	int ret;
+
+	ret = armada_overlay_create_properties(dev);
+	if (ret)
+		return ret;
+
+	dplane = kzalloc(sizeof(*dplane), GFP_KERNEL);
+	if (!dplane)
+		return -ENOMEM;
+
+	spin_lock_init(&dplane->lock);
+	init_waitqueue_head(&dplane->vbl.wait);
+	armada_drm_vbl_event_init(&dplane->vbl.update, armada_plane_vbl,
+				  dplane);
+
+	drm_plane_init(dev, &dplane->base, crtcs, &armada_plane_funcs,
+		       armada_formats, ARRAY_SIZE(armada_formats), false);
+
+	dplane->prop.colorkey_yr = 0xfefefe00;
+	dplane->prop.colorkey_ug = 0x01010100;
+	dplane->prop.colorkey_vb = 0x01010100;
+	dplane->prop.colorkey_mode = CFG_CKMODE(CKMODE_RGB);
+	dplane->prop.brightness = 0;
+	dplane->prop.contrast = 0x4000;
+	dplane->prop.saturation = 0x4000;
+
+	mobj = &dplane->base.base;
+	drm_object_attach_property(mobj, priv->colorkey_prop,
+				   0x0101fe);
+	drm_object_attach_property(mobj, priv->colorkey_min_prop,
+				   0x0101fe);
+	drm_object_attach_property(mobj, priv->colorkey_max_prop,
+				   0x0101fe);
+	drm_object_attach_property(mobj, priv->colorkey_val_prop,
+				   0x0101fe);
+	drm_object_attach_property(mobj, priv->colorkey_alpha_prop,
+				   0x000000);
+	drm_object_attach_property(mobj, priv->colorkey_mode_prop,
+				   CKMODE_RGB);
+	drm_object_attach_property(mobj, priv->brightness_prop, 256);
+	drm_object_attach_property(mobj, priv->contrast_prop,
+				   dplane->prop.contrast);
+	drm_object_attach_property(mobj, priv->saturation_prop,
+				   dplane->prop.saturation);
+
+	return 0;
+}
diff --git a/drivers/gpu/drm/armada/armada_slave.c b/drivers/gpu/drm/armada/armada_slave.c
new file mode 100644
index 000000000000..00d0facb42f3
--- /dev/null
+++ b/drivers/gpu/drm/armada/armada_slave.c
@@ -0,0 +1,139 @@
+/*
+ * Copyright (C) 2012 Russell King
+ *  Rewritten from the dovefb driver, and Armada510 manuals.
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ */
+#include <drm/drmP.h>
+#include <drm/drm_crtc_helper.h>
+#include <drm/drm_edid.h>
+#include <drm/drm_encoder_slave.h>
+#include "armada_drm.h"
+#include "armada_output.h"
+#include "armada_slave.h"
+
+static int armada_drm_slave_get_modes(struct drm_connector *conn)
+{
+	struct drm_encoder *enc = armada_drm_connector_encoder(conn);
+	int count = 0;
+
+	if (enc) {
+		struct drm_encoder_slave *slave = to_encoder_slave(enc);
+
+		count = slave->slave_funcs->get_modes(enc, conn);
+	}
+
+	return count;
+}
+
+static void armada_drm_slave_destroy(struct drm_encoder *enc)
+{
+	struct drm_encoder_slave *slave = to_encoder_slave(enc);
+	struct i2c_client *client = drm_i2c_encoder_get_client(enc);
+
+	if (slave->slave_funcs)
+		slave->slave_funcs->destroy(enc);
+	if (client)
+		i2c_put_adapter(client->adapter);
+
+	drm_encoder_cleanup(&slave->base);
+	kfree(slave);
+}
+
+static const struct drm_encoder_funcs armada_drm_slave_encoder_funcs = {
+	.destroy	= armada_drm_slave_destroy,
+};
+
+static const struct drm_connector_helper_funcs armada_drm_slave_helper_funcs = {
+	.get_modes	= armada_drm_slave_get_modes,
+	.mode_valid	= armada_drm_slave_encoder_mode_valid,
+	.best_encoder	= armada_drm_connector_encoder,
+};
+
+static const struct drm_encoder_helper_funcs drm_slave_encoder_helpers = {
+	.dpms = drm_i2c_encoder_dpms,
+	.save = drm_i2c_encoder_save,
+	.restore = drm_i2c_encoder_restore,
+	.mode_fixup = drm_i2c_encoder_mode_fixup,
+	.prepare = drm_i2c_encoder_prepare,
+	.commit = drm_i2c_encoder_commit,
+	.mode_set = drm_i2c_encoder_mode_set,
+	.detect = drm_i2c_encoder_detect,
+};
+
+static int
+armada_drm_conn_slave_create(struct drm_connector *conn, const void *data)
+{
+	const struct armada_drm_slave_config *config = data;
+	struct drm_encoder_slave *slave;
+	struct i2c_adapter *adap;
+	int ret;
+
+	conn->interlace_allowed = config->interlace_allowed;
+	conn->doublescan_allowed = config->doublescan_allowed;
+	conn->polled = config->polled;
+
+	drm_connector_helper_add(conn, &armada_drm_slave_helper_funcs);
+
+	slave = kzalloc(sizeof(*slave), GFP_KERNEL);
+	if (!slave)
+		return -ENOMEM;
+
+	slave->base.possible_crtcs = config->crtcs;
+
+	adap = i2c_get_adapter(config->i2c_adapter_id);
+	if (!adap) {
+		kfree(slave);
+		return -EPROBE_DEFER;
+	}
+
+	ret = drm_encoder_init(conn->dev, &slave->base,
+			       &armada_drm_slave_encoder_funcs,
+			       DRM_MODE_ENCODER_TMDS);
+	if (ret) {
+		DRM_ERROR("unable to init encoder\n");
+		i2c_put_adapter(adap);
+		kfree(slave);
+		return ret;
+	}
+
+	ret = drm_i2c_encoder_init(conn->dev, slave, adap, &config->info);
+	i2c_put_adapter(adap);
+	if (ret) {
+		DRM_ERROR("unable to init encoder slave\n");
+		armada_drm_slave_destroy(&slave->base);
+		return ret;
+	}
+
+	drm_encoder_helper_add(&slave->base, &drm_slave_encoder_helpers);
+
+	ret = slave->slave_funcs->create_resources(&slave->base, conn);
+	if (ret) {
+		armada_drm_slave_destroy(&slave->base);
+		return ret;
+	}
+
+	ret = drm_mode_connector_attach_encoder(conn, &slave->base);
+	if (ret) {
+		armada_drm_slave_destroy(&slave->base);
+		return ret;
+	}
+
+	conn->encoder = &slave->base;
+
+	return ret;
+}
+
+static const struct armada_output_type armada_drm_conn_slave = {
+	.connector_type	= DRM_MODE_CONNECTOR_HDMIA,
+	.create		= armada_drm_conn_slave_create,
+	.set_property	= armada_drm_slave_encoder_set_property,
+};
+
+int armada_drm_connector_slave_create(struct drm_device *dev,
+	const struct armada_drm_slave_config *config)
+{
+	return armada_output_create(dev, &armada_drm_conn_slave, config);
+}
diff --git a/drivers/gpu/drm/armada/armada_slave.h b/drivers/gpu/drm/armada/armada_slave.h
new file mode 100644
index 000000000000..bf2374c96fc1
--- /dev/null
+++ b/drivers/gpu/drm/armada/armada_slave.h
@@ -0,0 +1,26 @@
+/*
+ * Copyright (C) 2012 Russell King
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ */
+#ifndef ARMADA_SLAVE_H
+#define ARMADA_SLAVE_H
+
+#include <linux/i2c.h>
+#include <drm/drmP.h>
+
+struct armada_drm_slave_config {
+	int i2c_adapter_id;
+	uint32_t crtcs;
+	uint8_t polled;
+	bool interlace_allowed;
+	bool doublescan_allowed;
+	struct i2c_board_info info;
+};
+
+int armada_drm_connector_slave_create(struct drm_device *dev,
+	const struct armada_drm_slave_config *);
+
+#endif
diff --git a/drivers/gpu/drm/ast/Kconfig b/drivers/gpu/drm/ast/Kconfig
index da4a51eae824..8a784c460c89 100644
--- a/drivers/gpu/drm/ast/Kconfig
+++ b/drivers/gpu/drm/ast/Kconfig
@@ -6,6 +6,7 @@ config DRM_AST
 	select FB_SYS_FILLRECT
 	select FB_SYS_IMAGEBLIT
 	select DRM_KMS_HELPER
+	select DRM_KMS_FB_HELPER
 	select DRM_TTM
 	help
 	 Say yes for experimental AST GPU driver. Do not enable
diff --git a/drivers/gpu/drm/ast/ast_drv.c b/drivers/gpu/drm/ast/ast_drv.c
index 32e270dc714e..5137f15dba19 100644
--- a/drivers/gpu/drm/ast/ast_drv.c
+++ b/drivers/gpu/drm/ast/ast_drv.c
@@ -211,7 +211,6 @@ static struct drm_driver driver = {
 	.minor = DRIVER_MINOR,
 	.patchlevel = DRIVER_PATCHLEVEL,
 
-	.gem_init_object = ast_gem_init_object,
 	.gem_free_object = ast_gem_free_object,
 	.dumb_create = ast_dumb_create,
 	.dumb_map_offset = ast_dumb_mmap_offset,
diff --git a/drivers/gpu/drm/ast/ast_drv.h b/drivers/gpu/drm/ast/ast_drv.h
index 8492b68e873c..9833a1b1acc1 100644
--- a/drivers/gpu/drm/ast/ast_drv.h
+++ b/drivers/gpu/drm/ast/ast_drv.h
@@ -323,7 +323,6 @@ extern int ast_dumb_create(struct drm_file *file,
 			   struct drm_device *dev,
 			   struct drm_mode_create_dumb *args);
 
-extern int ast_gem_init_object(struct drm_gem_object *obj);
 extern void ast_gem_free_object(struct drm_gem_object *obj);
 extern int ast_dumb_mmap_offset(struct drm_file *file,
 				struct drm_device *dev,
diff --git a/drivers/gpu/drm/ast/ast_main.c b/drivers/gpu/drm/ast/ast_main.c
index 7f6152d374ca..af0b868a9dfd 100644
--- a/drivers/gpu/drm/ast/ast_main.c
+++ b/drivers/gpu/drm/ast/ast_main.c
@@ -449,12 +449,6 @@ int ast_dumb_create(struct drm_file *file,
 	return 0;
 }
 
-int ast_gem_init_object(struct drm_gem_object *obj)
-{
-	BUG();
-	return 0;
-}
-
 void ast_bo_unref(struct ast_bo **bo)
 {
 	struct ttm_buffer_object *tbo;
diff --git a/drivers/gpu/drm/cirrus/Kconfig b/drivers/gpu/drm/cirrus/Kconfig
index bf67b22723f9..9864559e5fb9 100644
--- a/drivers/gpu/drm/cirrus/Kconfig
+++ b/drivers/gpu/drm/cirrus/Kconfig
@@ -5,6 +5,7 @@ config DRM_CIRRUS_QEMU
 	select FB_SYS_COPYAREA
 	select FB_SYS_IMAGEBLIT
 	select DRM_KMS_HELPER
+	select DRM_KMS_FB_HELPER
 	select DRM_TTM
 	help
 	 This is a KMS driver for emulated cirrus device in qemu.
diff --git a/drivers/gpu/drm/cirrus/cirrus_drv.c b/drivers/gpu/drm/cirrus/cirrus_drv.c
index 138364d91782..953fc8aea69c 100644
--- a/drivers/gpu/drm/cirrus/cirrus_drv.c
+++ b/drivers/gpu/drm/cirrus/cirrus_drv.c
@@ -97,7 +97,6 @@ static struct drm_driver driver = {
 	.major = DRIVER_MAJOR,
 	.minor = DRIVER_MINOR,
 	.patchlevel = DRIVER_PATCHLEVEL,
-	.gem_init_object = cirrus_gem_init_object,
 	.gem_free_object = cirrus_gem_free_object,
 	.dumb_create = cirrus_dumb_create,
 	.dumb_map_offset = cirrus_dumb_mmap_offset,
diff --git a/drivers/gpu/drm/cirrus/cirrus_drv.h b/drivers/gpu/drm/cirrus/cirrus_drv.h
index 9b0bb9184afd..b6aded73838b 100644
--- a/drivers/gpu/drm/cirrus/cirrus_drv.h
+++ b/drivers/gpu/drm/cirrus/cirrus_drv.h
@@ -191,7 +191,6 @@ int cirrus_device_init(struct cirrus_device *cdev,
 		      struct pci_dev *pdev,
 		      uint32_t flags);
 void cirrus_device_fini(struct cirrus_device *cdev);
-int cirrus_gem_init_object(struct drm_gem_object *obj);
 void cirrus_gem_free_object(struct drm_gem_object *obj);
 int cirrus_dumb_mmap_offset(struct drm_file *file,
 			    struct drm_device *dev,
diff --git a/drivers/gpu/drm/cirrus/cirrus_main.c b/drivers/gpu/drm/cirrus/cirrus_main.c
index f130a533a512..78e76f24343d 100644
--- a/drivers/gpu/drm/cirrus/cirrus_main.c
+++ b/drivers/gpu/drm/cirrus/cirrus_main.c
@@ -255,12 +255,6 @@ int cirrus_dumb_create(struct drm_file *file,
 	return 0;
 }
 
-int cirrus_gem_init_object(struct drm_gem_object *obj)
-{
-	BUG();
-	return 0;
-}
-
 void cirrus_bo_unref(struct cirrus_bo **bo)
 {
 	struct ttm_buffer_object *tbo;
diff --git a/drivers/gpu/drm/cirrus/cirrus_mode.c b/drivers/gpu/drm/cirrus/cirrus_mode.c
index 60685b21cc36..adabc3daaa5b 100644
--- a/drivers/gpu/drm/cirrus/cirrus_mode.c
+++ b/drivers/gpu/drm/cirrus/cirrus_mode.c
@@ -494,13 +494,12 @@ static struct drm_encoder *cirrus_encoder_init(struct drm_device *dev)
 
 int cirrus_vga_get_modes(struct drm_connector *connector)
 {
-	/* Just add a static list of modes */
-	drm_add_modes_noedid(connector, 640, 480);
-	drm_add_modes_noedid(connector, 800, 600);
-	drm_add_modes_noedid(connector, 1024, 768);
-	drm_add_modes_noedid(connector, 1280, 1024);
+	int count;
 
-	return 4;
+	/* Just add a static list of modes */
+	count = drm_add_modes_noedid(connector, 1280, 1024);
+	drm_set_preferred_mode(connector, 1024, 768);
+	return count;
 }
 
 static int cirrus_vga_mode_valid(struct drm_connector *connector,
diff --git a/drivers/gpu/drm/drm_context.c b/drivers/gpu/drm/drm_context.c
index 224ff965bcf7..a4b017b6849e 100644
--- a/drivers/gpu/drm/drm_context.c
+++ b/drivers/gpu/drm/drm_context.c
@@ -334,7 +334,6 @@ int drm_addctx(struct drm_device *dev, void *data,
 
 	mutex_lock(&dev->ctxlist_mutex);
 	list_add(&ctx_entry->head, &dev->ctxlist);
-	++dev->ctx_count;
 	mutex_unlock(&dev->ctxlist_mutex);
 
 	return 0;
@@ -432,7 +431,6 @@ int drm_rmctx(struct drm_device *dev, void *data,
 			if (pos->handle == ctx->handle) {
 				list_del(&pos->head);
 				kfree(pos);
-				--dev->ctx_count;
 			}
 		}
 	}
diff --git a/drivers/gpu/drm/drm_crtc.c b/drivers/gpu/drm/drm_crtc.c
index bff2fa941f60..d6cf77c472e7 100644
--- a/drivers/gpu/drm/drm_crtc.c
+++ b/drivers/gpu/drm/drm_crtc.c
@@ -202,6 +202,7 @@ static struct drm_conn_prop_enum_list drm_connector_enum_list[] =
 	{ DRM_MODE_CONNECTOR_TV, "TV" },
 	{ DRM_MODE_CONNECTOR_eDP, "eDP" },
 	{ DRM_MODE_CONNECTOR_VIRTUAL, "Virtual" },
+	{ DRM_MODE_CONNECTOR_DSI, "DSI" },
 };
 
 static const struct drm_prop_enum_list drm_encoder_enum_list[] =
@@ -211,6 +212,7 @@ static const struct drm_prop_enum_list drm_encoder_enum_list[] =
 	{ DRM_MODE_ENCODER_LVDS, "LVDS" },
 	{ DRM_MODE_ENCODER_TVDAC, "TV" },
 	{ DRM_MODE_ENCODER_VIRTUAL, "Virtual" },
+	{ DRM_MODE_ENCODER_DSI, "DSI" },
 };
 
 void drm_connector_ida_init(void)
@@ -1301,7 +1303,7 @@ static void drm_crtc_convert_to_umode(struct drm_mode_modeinfo *out,
 }
 
 /**
- * drm_crtc_convert_to_umode - convert a modeinfo into a drm_display_mode
+ * drm_crtc_convert_umode - convert a modeinfo into a drm_display_mode
  * @out: drm_display_mode to return to the user
  * @in: drm_mode_modeinfo to use
  *
@@ -1317,6 +1319,9 @@ static int drm_crtc_convert_umode(struct drm_display_mode *out,
 	if (in->clock > INT_MAX || in->vrefresh > INT_MAX)
 		return -ERANGE;
 
+	if ((in->flags & DRM_MODE_FLAG_3D_MASK) > DRM_MODE_FLAG_3D_MAX)
+		return -EINVAL;
+
 	out->clock = in->clock;
 	out->hdisplay = in->hdisplay;
 	out->hsync_start = in->hsync_start;
@@ -1552,7 +1557,7 @@ int drm_mode_getcrtc(struct drm_device *dev,
 	obj = drm_mode_object_find(dev, crtc_resp->crtc_id,
 				   DRM_MODE_OBJECT_CRTC);
 	if (!obj) {
-		ret = -EINVAL;
+		ret = -ENOENT;
 		goto out;
 	}
 	crtc = obj_to_crtc(obj);
@@ -1579,6 +1584,19 @@ out:
 	return ret;
 }
 
+static bool drm_mode_expose_to_userspace(const struct drm_display_mode *mode,
+					 const struct drm_file *file_priv)
+{
+	/*
+	 * If user-space hasn't configured the driver to expose the stereo 3D
+	 * modes, don't expose them.
+	 */
+	if (!file_priv->stereo_allowed && drm_mode_is_stereo(mode))
+		return false;
+
+	return true;
+}
+
 /**
  * drm_mode_getconnector - get connector configuration
  * @dev: drm device for the ioctl
@@ -1623,7 +1641,7 @@ int drm_mode_getconnector(struct drm_device *dev, void *data,
 	obj = drm_mode_object_find(dev, out_resp->connector_id,
 				   DRM_MODE_OBJECT_CONNECTOR);
 	if (!obj) {
-		ret = -EINVAL;
+		ret = -ENOENT;
 		goto out;
 	}
 	connector = obj_to_connector(obj);
@@ -1644,7 +1662,8 @@ int drm_mode_getconnector(struct drm_device *dev, void *data,
 
 	/* delayed so we get modes regardless of pre-fill_modes state */
 	list_for_each_entry(mode, &connector->modes, head)
-		mode_count++;
+		if (drm_mode_expose_to_userspace(mode, file_priv))
+			mode_count++;
 
 	out_resp->connector_id = connector->base.id;
 	out_resp->connector_type = connector->connector_type;
@@ -1666,6 +1685,9 @@ int drm_mode_getconnector(struct drm_device *dev, void *data,
 		copied = 0;
 		mode_ptr = (struct drm_mode_modeinfo __user *)(unsigned long)out_resp->modes_ptr;
 		list_for_each_entry(mode, &connector->modes, head) {
+			if (!drm_mode_expose_to_userspace(mode, file_priv))
+				continue;
+
 			drm_crtc_convert_to_umode(&u_mode, mode);
 			if (copy_to_user(mode_ptr + copied,
 					 &u_mode, sizeof(u_mode))) {
@@ -1735,7 +1757,7 @@ int drm_mode_getencoder(struct drm_device *dev, void *data,
 	obj = drm_mode_object_find(dev, enc_resp->encoder_id,
 				   DRM_MODE_OBJECT_ENCODER);
 	if (!obj) {
-		ret = -EINVAL;
+		ret = -ENOENT;
 		goto out;
 	}
 	encoder = obj_to_encoder(obj);
@@ -2040,6 +2062,45 @@ int drm_mode_set_config_internal(struct drm_mode_set *set)
 }
 EXPORT_SYMBOL(drm_mode_set_config_internal);
 
+/*
+ * Checks that the framebuffer is big enough for the CRTC viewport
+ * (x, y, hdisplay, vdisplay)
+ */
+static int drm_crtc_check_viewport(const struct drm_crtc *crtc,
+				   int x, int y,
+				   const struct drm_display_mode *mode,
+				   const struct drm_framebuffer *fb)
+
+{
+	int hdisplay, vdisplay;
+
+	hdisplay = mode->hdisplay;
+	vdisplay = mode->vdisplay;
+
+	if (drm_mode_is_stereo(mode)) {
+		struct drm_display_mode adjusted = *mode;
+
+		drm_mode_set_crtcinfo(&adjusted, CRTC_STEREO_DOUBLE);
+		hdisplay = adjusted.crtc_hdisplay;
+		vdisplay = adjusted.crtc_vdisplay;
+	}
+
+	if (crtc->invert_dimensions)
+		swap(hdisplay, vdisplay);
+
+	if (hdisplay > fb->width ||
+	    vdisplay > fb->height ||
+	    x > fb->width - hdisplay ||
+	    y > fb->height - vdisplay) {
+		DRM_DEBUG_KMS("Invalid fb size %ux%u for CRTC viewport %ux%u+%d+%d%s.\n",
+			      fb->width, fb->height, hdisplay, vdisplay, x, y,
+			      crtc->invert_dimensions ? " (inverted)" : "");
+		return -ENOSPC;
+	}
+
+	return 0;
+}
+
 /**
  * drm_mode_setcrtc - set CRTC configuration
  * @dev: drm device for the ioctl
@@ -2080,14 +2141,13 @@ int drm_mode_setcrtc(struct drm_device *dev, void *data,
 				   DRM_MODE_OBJECT_CRTC);
 	if (!obj) {
 		DRM_DEBUG_KMS("Unknown CRTC ID %d\n", crtc_req->crtc_id);
-		ret = -EINVAL;
+		ret = -ENOENT;
 		goto out;
 	}
 	crtc = obj_to_crtc(obj);
 	DRM_DEBUG_KMS("[CRTC:%d]\n", crtc->base.id);
 
 	if (crtc_req->mode_valid) {
-		int hdisplay, vdisplay;
 		/* If we have a mode we need a framebuffer. */
 		/* If we pass -1, set the mode with the currently bound fb */
 		if (crtc_req->fb_id == -1) {
@@ -2104,7 +2164,7 @@ int drm_mode_setcrtc(struct drm_device *dev, void *data,
 			if (!fb) {
 				DRM_DEBUG_KMS("Unknown FB ID%d\n",
 						crtc_req->fb_id);
-				ret = -EINVAL;
+				ret = -ENOENT;
 				goto out;
 			}
 		}
@@ -2123,23 +2183,11 @@ int drm_mode_setcrtc(struct drm_device *dev, void *data,
 
 		drm_mode_set_crtcinfo(mode, CRTC_INTERLACE_HALVE_V);
 
-		hdisplay = mode->hdisplay;
-		vdisplay = mode->vdisplay;
-
-		if (crtc->invert_dimensions)
-			swap(hdisplay, vdisplay);
-
-		if (hdisplay > fb->width ||
-		    vdisplay > fb->height ||
-		    crtc_req->x > fb->width - hdisplay ||
-		    crtc_req->y > fb->height - vdisplay) {
-			DRM_DEBUG_KMS("Invalid fb size %ux%u for CRTC viewport %ux%u+%d+%d%s.\n",
-				      fb->width, fb->height,
-				      hdisplay, vdisplay, crtc_req->x, crtc_req->y,
-				      crtc->invert_dimensions ? " (inverted)" : "");
-			ret = -ENOSPC;
+		ret = drm_crtc_check_viewport(crtc, crtc_req->x, crtc_req->y,
+					      mode, fb);
+		if (ret)
 			goto out;
-		}
+
 	}
 
 	if (crtc_req->count_connectors == 0 && mode) {
@@ -2184,7 +2232,7 @@ int drm_mode_setcrtc(struct drm_device *dev, void *data,
 			if (!obj) {
 				DRM_DEBUG_KMS("Connector id %d unknown\n",
 						out_id);
-				ret = -EINVAL;
+				ret = -ENOENT;
 				goto out;
 			}
 			connector = obj_to_connector(obj);
@@ -2232,7 +2280,7 @@ static int drm_mode_cursor_common(struct drm_device *dev,
 	obj = drm_mode_object_find(dev, req->crtc_id, DRM_MODE_OBJECT_CRTC);
 	if (!obj) {
 		DRM_DEBUG_KMS("Unknown CRTC ID %d\n", req->crtc_id);
-		return -EINVAL;
+		return -ENOENT;
 	}
 	crtc = obj_to_crtc(obj);
 
@@ -2441,6 +2489,8 @@ static int format_check(const struct drm_mode_fb_cmd2 *r)
 	case DRM_FORMAT_YVU444:
 		return 0;
 	default:
+		DRM_DEBUG_KMS("invalid pixel format %s\n",
+			      drm_get_format_name(r->pixel_format));
 		return -EINVAL;
 	}
 }
@@ -2606,7 +2656,7 @@ fail_lookup:
 	mutex_unlock(&dev->mode_config.fb_lock);
 	mutex_unlock(&file_priv->fbs_lock);
 
-	return -EINVAL;
+	return -ENOENT;
 }
 
 /**
@@ -2634,7 +2684,7 @@ int drm_mode_getfb(struct drm_device *dev,
 
 	fb = drm_framebuffer_lookup(dev, r->fb_id);
 	if (!fb)
-		return -EINVAL;
+		return -ENOENT;
 
 	r->height = fb->height;
 	r->width = fb->width;
@@ -2679,7 +2729,7 @@ int drm_mode_dirtyfb_ioctl(struct drm_device *dev,
 
 	fb = drm_framebuffer_lookup(dev, r->fb_id);
 	if (!fb)
-		return -EINVAL;
+		return -ENOENT;
 
 	num_clips = r->num_clips;
 	clips_ptr = (struct drm_clip_rect __user *)(unsigned long)r->clips_ptr;
@@ -3011,7 +3061,7 @@ int drm_mode_getproperty_ioctl(struct drm_device *dev,
 	drm_modeset_lock_all(dev);
 	obj = drm_mode_object_find(dev, out_resp->prop_id, DRM_MODE_OBJECT_PROPERTY);
 	if (!obj) {
-		ret = -EINVAL;
+		ret = -ENOENT;
 		goto done;
 	}
 	property = obj_to_property(obj);
@@ -3140,7 +3190,7 @@ int drm_mode_getblob_ioctl(struct drm_device *dev,
 	drm_modeset_lock_all(dev);
 	obj = drm_mode_object_find(dev, out_resp->blob_id, DRM_MODE_OBJECT_BLOB);
 	if (!obj) {
-		ret = -EINVAL;
+		ret = -ENOENT;
 		goto done;
 	}
 	blob = obj_to_blob(obj);
@@ -3301,7 +3351,7 @@ int drm_mode_obj_get_properties_ioctl(struct drm_device *dev, void *data,
 
 	obj = drm_mode_object_find(dev, arg->obj_id, arg->obj_type);
 	if (!obj) {
-		ret = -EINVAL;
+		ret = -ENOENT;
 		goto out;
 	}
 	if (!obj->properties) {
@@ -3354,8 +3404,10 @@ int drm_mode_obj_set_property_ioctl(struct drm_device *dev, void *data,
 	drm_modeset_lock_all(dev);
 
 	arg_obj = drm_mode_object_find(dev, arg->obj_id, arg->obj_type);
-	if (!arg_obj)
+	if (!arg_obj) {
+		ret = -ENOENT;
 		goto out;
+	}
 	if (!arg_obj->properties)
 		goto out;
 
@@ -3368,8 +3420,10 @@ int drm_mode_obj_set_property_ioctl(struct drm_device *dev, void *data,
 
 	prop_obj = drm_mode_object_find(dev, arg->prop_id,
 					DRM_MODE_OBJECT_PROPERTY);
-	if (!prop_obj)
+	if (!prop_obj) {
+		ret = -ENOENT;
 		goto out;
+	}
 	property = obj_to_property(prop_obj);
 
 	if (!drm_property_change_is_valid(property, arg->value))
@@ -3454,7 +3508,7 @@ int drm_mode_gamma_set_ioctl(struct drm_device *dev,
 	drm_modeset_lock_all(dev);
 	obj = drm_mode_object_find(dev, crtc_lut->crtc_id, DRM_MODE_OBJECT_CRTC);
 	if (!obj) {
-		ret = -EINVAL;
+		ret = -ENOENT;
 		goto out;
 	}
 	crtc = obj_to_crtc(obj);
@@ -3513,7 +3567,7 @@ int drm_mode_gamma_get_ioctl(struct drm_device *dev,
 	drm_modeset_lock_all(dev);
 	obj = drm_mode_object_find(dev, crtc_lut->crtc_id, DRM_MODE_OBJECT_CRTC);
 	if (!obj) {
-		ret = -EINVAL;
+		ret = -ENOENT;
 		goto out;
 	}
 	crtc = obj_to_crtc(obj);
@@ -3556,7 +3610,6 @@ int drm_mode_page_flip_ioctl(struct drm_device *dev,
 	struct drm_framebuffer *fb = NULL, *old_fb = NULL;
 	struct drm_pending_vblank_event *e = NULL;
 	unsigned long flags;
-	int hdisplay, vdisplay;
 	int ret = -EINVAL;
 
 	if (page_flip->flags & ~DRM_MODE_PAGE_FLIP_FLAGS ||
@@ -3568,7 +3621,7 @@ int drm_mode_page_flip_ioctl(struct drm_device *dev,
 
 	obj = drm_mode_object_find(dev, page_flip->crtc_id, DRM_MODE_OBJECT_CRTC);
 	if (!obj)
-		return -EINVAL;
+		return -ENOENT;
 	crtc = obj_to_crtc(obj);
 
 	mutex_lock(&crtc->mutex);
@@ -3585,25 +3638,14 @@ int drm_mode_page_flip_ioctl(struct drm_device *dev,
 		goto out;
 
 	fb = drm_framebuffer_lookup(dev, page_flip->fb_id);
-	if (!fb)
+	if (!fb) {
+		ret = -ENOENT;
 		goto out;
+	}
 
-	hdisplay = crtc->mode.hdisplay;
-	vdisplay = crtc->mode.vdisplay;
-
-	if (crtc->invert_dimensions)
-		swap(hdisplay, vdisplay);
-
-	if (hdisplay > fb->width ||
-	    vdisplay > fb->height ||
-	    crtc->x > fb->width - hdisplay ||
-	    crtc->y > fb->height - vdisplay) {
-		DRM_DEBUG_KMS("Invalid fb size %ux%u for CRTC viewport %ux%u+%d+%d%s.\n",
-			      fb->width, fb->height, hdisplay, vdisplay, crtc->x, crtc->y,
-			      crtc->invert_dimensions ? " (inverted)" : "");
-		ret = -ENOSPC;
+	ret = drm_crtc_check_viewport(crtc, crtc->x, crtc->y, &crtc->mode, fb);
+	if (ret)
 		goto out;
-	}
 
 	if (crtc->fb->pixel_format != fb->pixel_format) {
 		DRM_DEBUG_KMS("Page flip is not allowed to change frame buffer format.\n");
@@ -3788,7 +3830,8 @@ void drm_fb_get_bpp_depth(uint32_t format, unsigned int *depth,
 		*bpp = 32;
 		break;
 	default:
-		DRM_DEBUG_KMS("unsupported pixel format\n");
+		DRM_DEBUG_KMS("unsupported pixel format %s\n",
+			      drm_get_format_name(format));
 		*depth = 0;
 		*bpp = 0;
 		break;
diff --git a/drivers/gpu/drm/drm_crtc_helper.c b/drivers/gpu/drm/drm_crtc_helper.c
index c722c3b5404d..01361aba033b 100644
--- a/drivers/gpu/drm/drm_crtc_helper.c
+++ b/drivers/gpu/drm/drm_crtc_helper.c
@@ -39,6 +39,10 @@
 #include <drm/drm_fb_helper.h>
 #include <drm/drm_edid.h>
 
+MODULE_AUTHOR("David Airlie, Jesse Barnes");
+MODULE_DESCRIPTION("DRM KMS helper");
+MODULE_LICENSE("GPL and additional rights");
+
 /**
  * drm_helper_move_panel_connectors_to_head() - move panels to the front in the
  * 						connector list
@@ -76,7 +80,8 @@ static void drm_mode_validate_flag(struct drm_connector *connector,
 {
 	struct drm_display_mode *mode;
 
-	if (flags == (DRM_MODE_FLAG_DBLSCAN | DRM_MODE_FLAG_INTERLACE))
+	if (flags == (DRM_MODE_FLAG_DBLSCAN | DRM_MODE_FLAG_INTERLACE |
+		      DRM_MODE_FLAG_3D_MASK))
 		return;
 
 	list_for_each_entry(mode, &connector->modes, head) {
@@ -86,6 +91,9 @@ static void drm_mode_validate_flag(struct drm_connector *connector,
 		if ((mode->flags & DRM_MODE_FLAG_DBLSCAN) &&
 				!(flags & DRM_MODE_FLAG_DBLSCAN))
 			mode->status = MODE_NO_DBLESCAN;
+		if ((mode->flags & DRM_MODE_FLAG_3D_MASK) &&
+				!(flags & DRM_MODE_FLAG_3D_MASK))
+			mode->status = MODE_NO_STEREO;
 	}
 
 	return;
@@ -105,9 +113,9 @@ static void drm_mode_validate_flag(struct drm_connector *connector,
  * then culled (based on validity and the @maxX, @maxY parameters) and put into
  * the normal modes list.
  *
- * Intended to be use as a generic implementation of the ->probe() @connector
- * callback for drivers that use the crtc helpers for output mode filtering and
- * detection.
+ * Intended to be use as a generic implementation of the ->fill_modes()
+ * @connector vfunc for drivers that use the crtc helpers for output mode
+ * filtering and detection.
  *
  * RETURNS:
  * Number of modes found on @connector.
@@ -175,6 +183,8 @@ int drm_helper_probe_single_connector_modes(struct drm_connector *connector,
 		mode_flags |= DRM_MODE_FLAG_INTERLACE;
 	if (connector->doublescan_allowed)
 		mode_flags |= DRM_MODE_FLAG_DBLSCAN;
+	if (connector->stereo_allowed)
+		mode_flags |= DRM_MODE_FLAG_3D_MASK;
 	drm_mode_validate_flag(connector, mode_flags);
 
 	list_for_each_entry(mode, &connector->modes, head) {
@@ -395,22 +405,25 @@ bool drm_crtc_helper_set_mode(struct drm_crtc *crtc,
 			      struct drm_framebuffer *old_fb)
 {
 	struct drm_device *dev = crtc->dev;
-	struct drm_display_mode *adjusted_mode, saved_mode, saved_hwmode;
+	struct drm_display_mode *adjusted_mode, saved_mode;
 	struct drm_crtc_helper_funcs *crtc_funcs = crtc->helper_private;
 	struct drm_encoder_helper_funcs *encoder_funcs;
 	int saved_x, saved_y;
+	bool saved_enabled;
 	struct drm_encoder *encoder;
 	bool ret = true;
 
+	saved_enabled = crtc->enabled;
 	crtc->enabled = drm_helper_crtc_in_use(crtc);
 	if (!crtc->enabled)
 		return true;
 
 	adjusted_mode = drm_mode_duplicate(dev, mode);
-	if (!adjusted_mode)
+	if (!adjusted_mode) {
+		crtc->enabled = saved_enabled;
 		return false;
+	}
 
-	saved_hwmode = crtc->hwmode;
 	saved_mode = crtc->mode;
 	saved_x = crtc->x;
 	saved_y = crtc->y;
@@ -529,7 +542,7 @@ bool drm_crtc_helper_set_mode(struct drm_crtc *crtc,
 done:
 	drm_mode_destroy(dev, adjusted_mode);
 	if (!ret) {
-		crtc->hwmode = saved_hwmode;
+		crtc->enabled = saved_enabled;
 		crtc->mode = saved_mode;
 		crtc->x = saved_x;
 		crtc->y = saved_y;
@@ -557,6 +570,14 @@ drm_crtc_helper_disable(struct drm_crtc *crtc)
 				continue;
 
 			connector->encoder = NULL;
+
+			/*
+			 * drm_helper_disable_unused_functions() ought to be
+			 * doing this, but since we've decoupled the encoder
+			 * from the connector above, the required connection
+			 * between them is henceforth no longer available.
+			 */
+			connector->dpms = DRM_MODE_DPMS_OFF;
 		}
 	}
 
@@ -583,9 +604,8 @@ drm_crtc_helper_disable(struct drm_crtc *crtc)
 int drm_crtc_helper_set_config(struct drm_mode_set *set)
 {
 	struct drm_device *dev;
-	struct drm_crtc *save_crtcs, *new_crtc, *crtc;
+	struct drm_crtc *new_crtc;
 	struct drm_encoder *save_encoders, *new_encoder, *encoder;
-	struct drm_framebuffer *old_fb = NULL;
 	bool mode_changed = false; /* if true do a full mode set */
 	bool fb_changed = false; /* if true and !mode_changed just do a flip */
 	struct drm_connector *save_connectors, *connector;
@@ -621,38 +641,28 @@ int drm_crtc_helper_set_config(struct drm_mode_set *set)
 
 	dev = set->crtc->dev;
 
-	/* Allocate space for the backup of all (non-pointer) crtc, encoder and
-	 * connector data. */
-	save_crtcs = kzalloc(dev->mode_config.num_crtc *
-			     sizeof(struct drm_crtc), GFP_KERNEL);
-	if (!save_crtcs)
-		return -ENOMEM;
-
+	/*
+	 * Allocate space for the backup of all (non-pointer) encoder and
+	 * connector data.
+	 */
 	save_encoders = kzalloc(dev->mode_config.num_encoder *
 				sizeof(struct drm_encoder), GFP_KERNEL);
-	if (!save_encoders) {
-		kfree(save_crtcs);
+	if (!save_encoders)
 		return -ENOMEM;
-	}
 
 	save_connectors = kzalloc(dev->mode_config.num_connector *
 				sizeof(struct drm_connector), GFP_KERNEL);
 	if (!save_connectors) {
-		kfree(save_crtcs);
 		kfree(save_encoders);
 		return -ENOMEM;
 	}
 
-	/* Copy data. Note that driver private data is not affected.
+	/*
+	 * Copy data. Note that driver private data is not affected.
 	 * Should anything bad happen only the expected state is
 	 * restored, not the drivers personal bookkeeping.
 	 */
 	count = 0;
-	list_for_each_entry(crtc, &dev->mode_config.crtc_list, head) {
-		save_crtcs[count++] = *crtc;
-	}
-
-	count = 0;
 	list_for_each_entry(encoder, &dev->mode_config.encoder_list, head) {
 		save_encoders[count++] = *encoder;
 	}
@@ -775,19 +785,17 @@ int drm_crtc_helper_set_config(struct drm_mode_set *set)
 		mode_changed = true;
 
 	if (mode_changed) {
-		set->crtc->enabled = drm_helper_crtc_in_use(set->crtc);
-		if (set->crtc->enabled) {
+		if (drm_helper_crtc_in_use(set->crtc)) {
 			DRM_DEBUG_KMS("attempting to set mode from"
 					" userspace\n");
 			drm_mode_debug_printmodeline(set->mode);
-			old_fb = set->crtc->fb;
 			set->crtc->fb = set->fb;
 			if (!drm_crtc_helper_set_mode(set->crtc, set->mode,
 						      set->x, set->y,
-						      old_fb)) {
+						      save_set.fb)) {
 				DRM_ERROR("failed to set mode on [CRTC:%d]\n",
 					  set->crtc->base.id);
-				set->crtc->fb = old_fb;
+				set->crtc->fb = save_set.fb;
 				ret = -EINVAL;
 				goto fail;
 			}
@@ -802,31 +810,24 @@ int drm_crtc_helper_set_config(struct drm_mode_set *set)
 	} else if (fb_changed) {
 		set->crtc->x = set->x;
 		set->crtc->y = set->y;
-
-		old_fb = set->crtc->fb;
-		if (set->crtc->fb != set->fb)
-			set->crtc->fb = set->fb;
+		set->crtc->fb = set->fb;
 		ret = crtc_funcs->mode_set_base(set->crtc,
-						set->x, set->y, old_fb);
+						set->x, set->y, save_set.fb);
 		if (ret != 0) {
-			set->crtc->fb = old_fb;
+			set->crtc->x = save_set.x;
+			set->crtc->y = save_set.y;
+			set->crtc->fb = save_set.fb;
 			goto fail;
 		}
 	}
 
 	kfree(save_connectors);
 	kfree(save_encoders);
-	kfree(save_crtcs);
 	return 0;
 
 fail:
 	/* Restore all previous data. */
 	count = 0;
-	list_for_each_entry(crtc, &dev->mode_config.crtc_list, head) {
-		*crtc = save_crtcs[count++];
-	}
-
-	count = 0;
 	list_for_each_entry(encoder, &dev->mode_config.encoder_list, head) {
 		*encoder = save_encoders[count++];
 	}
@@ -844,7 +845,6 @@ fail:
 
 	kfree(save_connectors);
 	kfree(save_encoders);
-	kfree(save_crtcs);
 	return ret;
 }
 EXPORT_SYMBOL(drm_crtc_helper_set_config);
@@ -1125,14 +1125,14 @@ void drm_kms_helper_poll_fini(struct drm_device *dev)
 }
 EXPORT_SYMBOL(drm_kms_helper_poll_fini);
 
-void drm_helper_hpd_irq_event(struct drm_device *dev)
+bool drm_helper_hpd_irq_event(struct drm_device *dev)
 {
 	struct drm_connector *connector;
 	enum drm_connector_status old_status;
 	bool changed = false;
 
 	if (!dev->mode_config.poll_enabled)
-		return;
+		return false;
 
 	mutex_lock(&dev->mode_config.mutex);
 	list_for_each_entry(connector, &dev->mode_config.connector_list, head) {
@@ -1157,5 +1157,7 @@ void drm_helper_hpd_irq_event(struct drm_device *dev)
 
 	if (changed)
 		drm_kms_helper_hotplug_event(dev);
+
+	return changed;
 }
 EXPORT_SYMBOL(drm_helper_hpd_irq_event);
diff --git a/drivers/gpu/drm/drm_debugfs.c b/drivers/gpu/drm/drm_debugfs.c
index a05087cf846d..b4b51d46f339 100644
--- a/drivers/gpu/drm/drm_debugfs.c
+++ b/drivers/gpu/drm/drm_debugfs.c
@@ -42,7 +42,7 @@
  * Initialization, etc.
  **************************************************/
 
-static struct drm_info_list drm_debugfs_list[] = {
+static const struct drm_info_list drm_debugfs_list[] = {
 	{"name", drm_name_info, 0},
 	{"vm", drm_vm_info, 0},
 	{"clients", drm_clients_info, 0},
@@ -84,7 +84,7 @@ static const struct file_operations drm_debugfs_fops = {
  * Create a given set of debugfs files represented by an array of
  * gdm_debugfs_lists in the given root directory.
  */
-int drm_debugfs_create_files(struct drm_info_list *files, int count,
+int drm_debugfs_create_files(const struct drm_info_list *files, int count,
 			     struct dentry *root, struct drm_minor *minor)
 {
 	struct drm_device *dev = minor->dev;
@@ -188,7 +188,7 @@ int drm_debugfs_init(struct drm_minor *minor, int minor_id,
  *
  * Remove all debugfs entries created by debugfs_init().
  */
-int drm_debugfs_remove_files(struct drm_info_list *files, int count,
+int drm_debugfs_remove_files(const struct drm_info_list *files, int count,
 			     struct drm_minor *minor)
 {
 	struct list_head *pos, *q;
diff --git a/drivers/gpu/drm/drm_dp_helper.c b/drivers/gpu/drm/drm_dp_helper.c
index 89e196627160..9e978aae8972 100644
--- a/drivers/gpu/drm/drm_dp_helper.c
+++ b/drivers/gpu/drm/drm_dp_helper.c
@@ -228,12 +228,12 @@ i2c_dp_aux_add_bus(struct i2c_adapter *adapter)
 EXPORT_SYMBOL(i2c_dp_aux_add_bus);
 
 /* Helpers for DP link training */
-static u8 dp_link_status(u8 link_status[DP_LINK_STATUS_SIZE], int r)
+static u8 dp_link_status(const u8 link_status[DP_LINK_STATUS_SIZE], int r)
 {
 	return link_status[r - DP_LANE0_1_STATUS];
 }
 
-static u8 dp_get_lane_status(u8 link_status[DP_LINK_STATUS_SIZE],
+static u8 dp_get_lane_status(const u8 link_status[DP_LINK_STATUS_SIZE],
 			     int lane)
 {
 	int i = DP_LANE0_1_STATUS + (lane >> 1);
@@ -242,7 +242,7 @@ static u8 dp_get_lane_status(u8 link_status[DP_LINK_STATUS_SIZE],
 	return (l >> s) & 0xf;
 }
 
-bool drm_dp_channel_eq_ok(u8 link_status[DP_LINK_STATUS_SIZE],
+bool drm_dp_channel_eq_ok(const u8 link_status[DP_LINK_STATUS_SIZE],
 			  int lane_count)
 {
 	u8 lane_align;
@@ -262,7 +262,7 @@ bool drm_dp_channel_eq_ok(u8 link_status[DP_LINK_STATUS_SIZE],
 }
 EXPORT_SYMBOL(drm_dp_channel_eq_ok);
 
-bool drm_dp_clock_recovery_ok(u8 link_status[DP_LINK_STATUS_SIZE],
+bool drm_dp_clock_recovery_ok(const u8 link_status[DP_LINK_STATUS_SIZE],
 			      int lane_count)
 {
 	int lane;
@@ -277,7 +277,7 @@ bool drm_dp_clock_recovery_ok(u8 link_status[DP_LINK_STATUS_SIZE],
 }
 EXPORT_SYMBOL(drm_dp_clock_recovery_ok);
 
-u8 drm_dp_get_adjust_request_voltage(u8 link_status[DP_LINK_STATUS_SIZE],
+u8 drm_dp_get_adjust_request_voltage(const u8 link_status[DP_LINK_STATUS_SIZE],
 				     int lane)
 {
 	int i = DP_ADJUST_REQUEST_LANE0_1 + (lane >> 1);
@@ -290,7 +290,7 @@ u8 drm_dp_get_adjust_request_voltage(u8 link_status[DP_LINK_STATUS_SIZE],
 }
 EXPORT_SYMBOL(drm_dp_get_adjust_request_voltage);
 
-u8 drm_dp_get_adjust_request_pre_emphasis(u8 link_status[DP_LINK_STATUS_SIZE],
+u8 drm_dp_get_adjust_request_pre_emphasis(const u8 link_status[DP_LINK_STATUS_SIZE],
 					  int lane)
 {
 	int i = DP_ADJUST_REQUEST_LANE0_1 + (lane >> 1);
@@ -303,7 +303,7 @@ u8 drm_dp_get_adjust_request_pre_emphasis(u8 link_status[DP_LINK_STATUS_SIZE],
 }
 EXPORT_SYMBOL(drm_dp_get_adjust_request_pre_emphasis);
 
-void drm_dp_link_train_clock_recovery_delay(u8 dpcd[DP_RECEIVER_CAP_SIZE]) {
+void drm_dp_link_train_clock_recovery_delay(const u8 dpcd[DP_RECEIVER_CAP_SIZE]) {
 	if (dpcd[DP_TRAINING_AUX_RD_INTERVAL] == 0)
 		udelay(100);
 	else
@@ -311,7 +311,7 @@ void drm_dp_link_train_clock_recovery_delay(u8 dpcd[DP_RECEIVER_CAP_SIZE]) {
 }
 EXPORT_SYMBOL(drm_dp_link_train_clock_recovery_delay);
 
-void drm_dp_link_train_channel_eq_delay(u8 dpcd[DP_RECEIVER_CAP_SIZE]) {
+void drm_dp_link_train_channel_eq_delay(const u8 dpcd[DP_RECEIVER_CAP_SIZE]) {
 	if (dpcd[DP_TRAINING_AUX_RD_INTERVAL] == 0)
 		udelay(400);
 	else
diff --git a/drivers/gpu/drm/drm_drv.c b/drivers/gpu/drm/drm_drv.c
index fe58d0833a11..d9137e49c4e8 100644
--- a/drivers/gpu/drm/drm_drv.c
+++ b/drivers/gpu/drm/drm_drv.c
@@ -69,6 +69,7 @@ static const struct drm_ioctl_desc drm_ioctls[] = {
 	DRM_IOCTL_DEF(DRM_IOCTL_GET_CLIENT, drm_getclient, DRM_UNLOCKED),
 	DRM_IOCTL_DEF(DRM_IOCTL_GET_STATS, drm_getstats, DRM_UNLOCKED),
 	DRM_IOCTL_DEF(DRM_IOCTL_GET_CAP, drm_getcap, DRM_UNLOCKED|DRM_RENDER_ALLOW),
+	DRM_IOCTL_DEF(DRM_IOCTL_SET_CLIENT_CAP, drm_setclientcap, 0),
 	DRM_IOCTL_DEF(DRM_IOCTL_SET_VERSION, drm_setversion, DRM_MASTER),
 
 	DRM_IOCTL_DEF(DRM_IOCTL_SET_UNIQUE, drm_setunique, DRM_AUTH|DRM_MASTER|DRM_ROOT_ONLY),
@@ -170,76 +171,6 @@ static const struct drm_ioctl_desc drm_ioctls[] = {
 
 #define DRM_CORE_IOCTL_COUNT	ARRAY_SIZE( drm_ioctls )
 
-/**
- * drm_legacy_dev_reinit
- *
- * Reinitializes a legacy/ums drm device in it's lastclose function.
- */
-static void drm_legacy_dev_reinit(struct drm_device *dev)
-{
-	int i;
-
-	if (drm_core_check_feature(dev, DRIVER_MODESET))
-		return;
-
-	atomic_set(&dev->ioctl_count, 0);
-	atomic_set(&dev->vma_count, 0);
-
-	for (i = 0; i < ARRAY_SIZE(dev->counts); i++)
-		atomic_set(&dev->counts[i], 0);
-
-	dev->sigdata.lock = NULL;
-
-	dev->context_flag = 0;
-	dev->last_context = 0;
-	dev->if_version = 0;
-}
-
-/**
- * Take down the DRM device.
- *
- * \param dev DRM device structure.
- *
- * Frees every resource in \p dev.
- *
- * \sa drm_device
- */
-int drm_lastclose(struct drm_device * dev)
-{
-	struct drm_vma_entry *vma, *vma_temp;
-
-	DRM_DEBUG("\n");
-
-	if (dev->driver->lastclose)
-		dev->driver->lastclose(dev);
-	DRM_DEBUG("driver lastclose completed\n");
-
-	if (dev->irq_enabled && !drm_core_check_feature(dev, DRIVER_MODESET))
-		drm_irq_uninstall(dev);
-
-	mutex_lock(&dev->struct_mutex);
-
-	drm_agp_clear(dev);
-
-	drm_legacy_sg_cleanup(dev);
-
-	/* Clear vma list (only built for debugging) */
-	list_for_each_entry_safe(vma, vma_temp, &dev->vmalist, head) {
-		list_del(&vma->head);
-		kfree(vma);
-	}
-
-	drm_legacy_dma_takedown(dev);
-
-	dev->dev_mapping = NULL;
-	mutex_unlock(&dev->struct_mutex);
-
-	drm_legacy_dev_reinit(dev);
-
-	DRM_DEBUG("lastclose completed\n");
-	return 0;
-}
-
 /** File operations structure */
 static const struct file_operations drm_stub_fops = {
 	.owner = THIS_MODULE,
@@ -385,7 +316,6 @@ long drm_ioctl(struct file *filp,
 		return -ENODEV;
 
 	atomic_inc(&dev->ioctl_count);
-	atomic_inc(&dev->counts[_DRM_STAT_IOCTLS]);
 	++file_priv->ioctl_count;
 
 	if ((nr >= DRM_CORE_IOCTL_COUNT) &&
@@ -473,7 +403,7 @@ long drm_ioctl(struct file *filp,
 
       err_i1:
 	if (!ioctl)
-		DRM_DEBUG("invalid iotcl: pid=%d, dev=0x%lx, auth=%d, cmd=0x%02x, nr=0x%02x\n",
+		DRM_DEBUG("invalid ioctl: pid=%d, dev=0x%lx, auth=%d, cmd=0x%02x, nr=0x%02x\n",
 			  task_pid_nr(current),
 			  (long)old_encode_dev(file_priv->minor->device),
 			  file_priv->authenticated, cmd, nr);
diff --git a/drivers/gpu/drm/drm_edid.c b/drivers/gpu/drm/drm_edid.c
index 830f7501cb4d..fb7cf0e796f6 100644
--- a/drivers/gpu/drm/drm_edid.c
+++ b/drivers/gpu/drm/drm_edid.c
@@ -458,6 +458,15 @@ static const struct drm_display_mode drm_dmt_modes[] = {
 		   DRM_MODE_FLAG_PHSYNC | DRM_MODE_FLAG_NVSYNC) },
 };
 
+/*
+ * These more or less come from the DMT spec.  The 720x400 modes are
+ * inferred from historical 80x25 practice.  The 640x480@67 and 832x624@75
+ * modes are old-school Mac modes.  The EDID spec says the 1152x864@75 mode
+ * should be 1152x870, again for the Mac, but instead we use the x864 DMT
+ * mode.
+ *
+ * The DMT modes have been fact-checked; the rest are mild guesses.
+ */
 static const struct drm_display_mode edid_est_modes[] = {
 	{ DRM_MODE("800x600", DRM_MODE_TYPE_DRIVER, 40000, 800, 840,
 		   968, 1056, 0, 600, 601, 605, 628, 0,
@@ -560,7 +569,7 @@ static const struct minimode est3_modes[] = {
 	{ 1600, 1200, 75, 0 },
 	{ 1600, 1200, 85, 0 },
 	{ 1792, 1344, 60, 0 },
-	{ 1792, 1344, 85, 0 },
+	{ 1792, 1344, 75, 0 },
 	{ 1856, 1392, 60, 0 },
 	{ 1856, 1392, 75, 0 },
 	{ 1920, 1200, 60, 1 },
@@ -1264,6 +1273,18 @@ struct edid *drm_get_edid(struct drm_connector *connector,
 }
 EXPORT_SYMBOL(drm_get_edid);
 
+/**
+ * drm_edid_duplicate - duplicate an EDID and the extensions
+ * @edid: EDID to duplicate
+ *
+ * Return duplicate edid or NULL on allocation failure.
+ */
+struct edid *drm_edid_duplicate(const struct edid *edid)
+{
+	return kmemdup(edid, (edid->extensions + 1) * EDID_LENGTH, GFP_KERNEL);
+}
+EXPORT_SYMBOL(drm_edid_duplicate);
+
 /*** EDID parsing ***/
 
 /**
@@ -1308,7 +1329,7 @@ static u32 edid_get_quirks(struct edid *edid)
 }
 
 #define MODE_SIZE(m) ((m)->hdisplay * (m)->vdisplay)
-#define MODE_REFRESH_DIFF(m,r) (abs((m)->vrefresh - target_refresh))
+#define MODE_REFRESH_DIFF(c,t) (abs((c) - (t)))
 
 /**
  * edid_fixup_preferred - set preferred modes based on quirk list
@@ -1323,6 +1344,7 @@ static void edid_fixup_preferred(struct drm_connector *connector,
 {
 	struct drm_display_mode *t, *cur_mode, *preferred_mode;
 	int target_refresh = 0;
+	int cur_vrefresh, preferred_vrefresh;
 
 	if (list_empty(&connector->probed_modes))
 		return;
@@ -1345,10 +1367,14 @@ static void edid_fixup_preferred(struct drm_connector *connector,
 		if (MODE_SIZE(cur_mode) > MODE_SIZE(preferred_mode))
 			preferred_mode = cur_mode;
 
+		cur_vrefresh = cur_mode->vrefresh ?
+			cur_mode->vrefresh : drm_mode_vrefresh(cur_mode);
+		preferred_vrefresh = preferred_mode->vrefresh ?
+			preferred_mode->vrefresh : drm_mode_vrefresh(preferred_mode);
 		/* At a given size, try to get closest to target refresh */
 		if ((MODE_SIZE(cur_mode) == MODE_SIZE(preferred_mode)) &&
-		    MODE_REFRESH_DIFF(cur_mode, target_refresh) <
-		    MODE_REFRESH_DIFF(preferred_mode, target_refresh)) {
+		    MODE_REFRESH_DIFF(cur_vrefresh, target_refresh) <
+		    MODE_REFRESH_DIFF(preferred_vrefresh, target_refresh)) {
 			preferred_mode = cur_mode;
 		}
 	}
@@ -2068,7 +2094,7 @@ drm_est3_modes(struct drm_connector *connector, struct detailed_timing *timing)
 	u8 *est = ((u8 *)timing) + 5;
 
 	for (i = 0; i < 6; i++) {
-		for (j = 7; j > 0; j--) {
+		for (j = 7; j >= 0; j--) {
 			m = (i * 8) + (7 - j);
 			if (m >= ARRAY_SIZE(est3_modes))
 				break;
@@ -2404,7 +2430,7 @@ u8 drm_match_cea_mode(const struct drm_display_mode *to_match)
 
 		if ((KHZ2PICOS(to_match->clock) == KHZ2PICOS(clock1) ||
 		     KHZ2PICOS(to_match->clock) == KHZ2PICOS(clock2)) &&
-		    drm_mode_equal_no_clocks(to_match, cea_mode))
+		    drm_mode_equal_no_clocks_no_stereo(to_match, cea_mode))
 			return mode + 1;
 	}
 	return 0;
@@ -2453,7 +2479,7 @@ static u8 drm_match_hdmi_mode(const struct drm_display_mode *to_match)
 
 		if ((KHZ2PICOS(to_match->clock) == KHZ2PICOS(clock1) ||
 		     KHZ2PICOS(to_match->clock) == KHZ2PICOS(clock2)) &&
-		    drm_mode_equal_no_clocks(to_match, hdmi_mode))
+		    drm_mode_equal_no_clocks_no_stereo(to_match, hdmi_mode))
 			return mode + 1;
 	}
 	return 0;
@@ -2507,6 +2533,9 @@ add_alternate_cea_modes(struct drm_connector *connector, struct edid *edid)
 		if (!newmode)
 			continue;
 
+		/* Carry over the stereo flags */
+		newmode->flags |= mode->flags & DRM_MODE_FLAG_3D_MASK;
+
 		/*
 		 * The current mode could be either variant. Make
 		 * sure to pick the "other" clock for the new mode.
@@ -2553,20 +2582,151 @@ do_cea_modes(struct drm_connector *connector, const u8 *db, u8 len)
 	return modes;
 }
 
+struct stereo_mandatory_mode {
+	int width, height, vrefresh;
+	unsigned int flags;
+};
+
+static const struct stereo_mandatory_mode stereo_mandatory_modes[] = {
+	{ 1920, 1080, 24, DRM_MODE_FLAG_3D_TOP_AND_BOTTOM },
+	{ 1920, 1080, 24, DRM_MODE_FLAG_3D_FRAME_PACKING },
+	{ 1920, 1080, 50,
+	  DRM_MODE_FLAG_INTERLACE | DRM_MODE_FLAG_3D_SIDE_BY_SIDE_HALF },
+	{ 1920, 1080, 60,
+	  DRM_MODE_FLAG_INTERLACE | DRM_MODE_FLAG_3D_SIDE_BY_SIDE_HALF },
+	{ 1280, 720,  50, DRM_MODE_FLAG_3D_TOP_AND_BOTTOM },
+	{ 1280, 720,  50, DRM_MODE_FLAG_3D_FRAME_PACKING },
+	{ 1280, 720,  60, DRM_MODE_FLAG_3D_TOP_AND_BOTTOM },
+	{ 1280, 720,  60, DRM_MODE_FLAG_3D_FRAME_PACKING }
+};
+
+static bool
+stereo_match_mandatory(const struct drm_display_mode *mode,
+		       const struct stereo_mandatory_mode *stereo_mode)
+{
+	unsigned int interlaced = mode->flags & DRM_MODE_FLAG_INTERLACE;
+
+	return mode->hdisplay == stereo_mode->width &&
+	       mode->vdisplay == stereo_mode->height &&
+	       interlaced == (stereo_mode->flags & DRM_MODE_FLAG_INTERLACE) &&
+	       drm_mode_vrefresh(mode) == stereo_mode->vrefresh;
+}
+
+static int add_hdmi_mandatory_stereo_modes(struct drm_connector *connector)
+{
+	struct drm_device *dev = connector->dev;
+	const struct drm_display_mode *mode;
+	struct list_head stereo_modes;
+	int modes = 0, i;
+
+	INIT_LIST_HEAD(&stereo_modes);
+
+	list_for_each_entry(mode, &connector->probed_modes, head) {
+		for (i = 0; i < ARRAY_SIZE(stereo_mandatory_modes); i++) {
+			const struct stereo_mandatory_mode *mandatory;
+			struct drm_display_mode *new_mode;
+
+			if (!stereo_match_mandatory(mode,
+						    &stereo_mandatory_modes[i]))
+				continue;
+
+			mandatory = &stereo_mandatory_modes[i];
+			new_mode = drm_mode_duplicate(dev, mode);
+			if (!new_mode)
+				continue;
+
+			new_mode->flags |= mandatory->flags;
+			list_add_tail(&new_mode->head, &stereo_modes);
+			modes++;
+		}
+	}
+
+	list_splice_tail(&stereo_modes, &connector->probed_modes);
+
+	return modes;
+}
+
+static int add_hdmi_mode(struct drm_connector *connector, u8 vic)
+{
+	struct drm_device *dev = connector->dev;
+	struct drm_display_mode *newmode;
+
+	vic--; /* VICs start at 1 */
+	if (vic >= ARRAY_SIZE(edid_4k_modes)) {
+		DRM_ERROR("Unknown HDMI VIC: %d\n", vic);
+		return 0;
+	}
+
+	newmode = drm_mode_duplicate(dev, &edid_4k_modes[vic]);
+	if (!newmode)
+		return 0;
+
+	drm_mode_probed_add(connector, newmode);
+
+	return 1;
+}
+
+static int add_3d_struct_modes(struct drm_connector *connector, u16 structure,
+			       const u8 *video_db, u8 video_len, u8 video_index)
+{
+	struct drm_device *dev = connector->dev;
+	struct drm_display_mode *newmode;
+	int modes = 0;
+	u8 cea_mode;
+
+	if (video_db == NULL || video_index > video_len)
+		return 0;
+
+	/* CEA modes are numbered 1..127 */
+	cea_mode = (video_db[video_index] & 127) - 1;
+	if (cea_mode >= ARRAY_SIZE(edid_cea_modes))
+		return 0;
+
+	if (structure & (1 << 0)) {
+		newmode = drm_mode_duplicate(dev, &edid_cea_modes[cea_mode]);
+		if (newmode) {
+			newmode->flags |= DRM_MODE_FLAG_3D_FRAME_PACKING;
+			drm_mode_probed_add(connector, newmode);
+			modes++;
+		}
+	}
+	if (structure & (1 << 6)) {
+		newmode = drm_mode_duplicate(dev, &edid_cea_modes[cea_mode]);
+		if (newmode) {
+			newmode->flags |= DRM_MODE_FLAG_3D_TOP_AND_BOTTOM;
+			drm_mode_probed_add(connector, newmode);
+			modes++;
+		}
+	}
+	if (structure & (1 << 8)) {
+		newmode = drm_mode_duplicate(dev, &edid_cea_modes[cea_mode]);
+		if (newmode) {
+			newmode->flags = DRM_MODE_FLAG_3D_SIDE_BY_SIDE_HALF;
+			drm_mode_probed_add(connector, newmode);
+			modes++;
+		}
+	}
+
+	return modes;
+}
+
 /*
  * do_hdmi_vsdb_modes - Parse the HDMI Vendor Specific data block
  * @connector: connector corresponding to the HDMI sink
  * @db: start of the CEA vendor specific block
  * @len: length of the CEA block payload, ie. one can access up to db[len]
  *
- * Parses the HDMI VSDB looking for modes to add to @connector.
+ * Parses the HDMI VSDB looking for modes to add to @connector. This function
+ * also adds the stereo 3d modes when applicable.
  */
 static int
-do_hdmi_vsdb_modes(struct drm_connector *connector, const u8 *db, u8 len)
+do_hdmi_vsdb_modes(struct drm_connector *connector, const u8 *db, u8 len,
+		   const u8 *video_db, u8 video_len)
 {
-	struct drm_device *dev = connector->dev;
-	int modes = 0, offset = 0, i;
-	u8 vic_len;
+	int modes = 0, offset = 0, i, multi_present = 0;
+	u8 vic_len, hdmi_3d_len = 0;
+	u16 mask;
+	u16 structure_all;
 
 	if (len < 8)
 		goto out;
@@ -2585,30 +2745,56 @@ do_hdmi_vsdb_modes(struct drm_connector *connector, const u8 *db, u8 len)
 
 	/* the declared length is not long enough for the 2 first bytes
 	 * of additional video format capabilities */
-	offset += 2;
-	if (len < (8 + offset))
+	if (len < (8 + offset + 2))
 		goto out;
 
+	/* 3D_Present */
+	offset++;
+	if (db[8 + offset] & (1 << 7)) {
+		modes += add_hdmi_mandatory_stereo_modes(connector);
+
+		/* 3D_Multi_present */
+		multi_present = (db[8 + offset] & 0x60) >> 5;
+	}
+
+	offset++;
 	vic_len = db[8 + offset] >> 5;
+	hdmi_3d_len = db[8 + offset] & 0x1f;
 
 	for (i = 0; i < vic_len && len >= (9 + offset + i); i++) {
-		struct drm_display_mode *newmode;
 		u8 vic;
 
 		vic = db[9 + offset + i];
+		modes += add_hdmi_mode(connector, vic);
+	}
+	offset += 1 + vic_len;
 
-		vic--; /* VICs start at 1 */
-		if (vic >= ARRAY_SIZE(edid_4k_modes)) {
-			DRM_ERROR("Unknown HDMI VIC: %d\n", vic);
-			continue;
-		}
+	if (!(multi_present == 1 || multi_present == 2))
+		goto out;
 
-		newmode = drm_mode_duplicate(dev, &edid_4k_modes[vic]);
-		if (!newmode)
-			continue;
+	if ((multi_present == 1 && len < (9 + offset)) ||
+	    (multi_present == 2 && len < (11 + offset)))
+		goto out;
 
-		drm_mode_probed_add(connector, newmode);
-		modes++;
+	if ((multi_present == 1 && hdmi_3d_len < 2) ||
+	    (multi_present == 2 && hdmi_3d_len < 4))
+		goto out;
+
+	/* 3D_Structure_ALL */
+	structure_all = (db[8 + offset] << 8) | db[9 + offset];
+
+	/* check if 3D_MASK is present */
+	if (multi_present == 2)
+		mask = (db[10 + offset] << 8) | db[11 + offset];
+	else
+		mask = 0xffff;
+
+	for (i = 0; i < 16; i++) {
+		if (mask & (1 << i))
+			modes += add_3d_struct_modes(connector,
+						     structure_all,
+						     video_db,
+						     video_len, i);
 	}
 
 out:
@@ -2668,8 +2854,8 @@ static int
 add_cea_modes(struct drm_connector *connector, struct edid *edid)
 {
 	const u8 *cea = drm_find_cea_extension(edid);
-	const u8 *db;
-	u8 dbl;
+	const u8 *db, *hdmi = NULL, *video = NULL;
+	u8 dbl, hdmi_len, video_len = 0;
 	int modes = 0;
 
 	if (cea && cea_revision(cea) >= 3) {
@@ -2682,13 +2868,26 @@ add_cea_modes(struct drm_connector *connector, struct edid *edid)
 			db = &cea[i];
 			dbl = cea_db_payload_len(db);
 
-			if (cea_db_tag(db) == VIDEO_BLOCK)
-				modes += do_cea_modes(connector, db + 1, dbl);
-			else if (cea_db_is_hdmi_vsdb(db))
-				modes += do_hdmi_vsdb_modes(connector, db, dbl);
+			if (cea_db_tag(db) == VIDEO_BLOCK) {
+				video = db + 1;
+				video_len = dbl;
+				modes += do_cea_modes(connector, video, dbl);
+			}
+			else if (cea_db_is_hdmi_vsdb(db)) {
+				hdmi = db;
+				hdmi_len = dbl;
+			}
 		}
 	}
 
+	/*
+	 * We parse the HDMI VSDB after having added the cea modes as we will
+	 * be patching their flags when the sink supports stereo 3D.
+	 */
+	if (hdmi)
+		modes += do_hdmi_vsdb_modes(connector, hdmi, hdmi_len, video,
+					    video_len);
+
 	return modes;
 }
 
@@ -3288,6 +3487,19 @@ int drm_add_modes_noedid(struct drm_connector *connector,
 }
 EXPORT_SYMBOL(drm_add_modes_noedid);
 
+void drm_set_preferred_mode(struct drm_connector *connector,
+			   int hpref, int vpref)
+{
+	struct drm_display_mode *mode;
+
+	list_for_each_entry(mode, &connector->probed_modes, head) {
+		if (drm_mode_width(mode)  == hpref &&
+		    drm_mode_height(mode) == vpref)
+			mode->type |= DRM_MODE_TYPE_PREFERRED;
+	}
+}
+EXPORT_SYMBOL(drm_set_preferred_mode);
+
 /**
  * drm_hdmi_avi_infoframe_from_display_mode() - fill an HDMI AVI infoframe with
  *                                              data from a DRM display mode
@@ -3321,6 +3533,33 @@ drm_hdmi_avi_infoframe_from_display_mode(struct hdmi_avi_infoframe *frame,
 }
 EXPORT_SYMBOL(drm_hdmi_avi_infoframe_from_display_mode);
 
+static enum hdmi_3d_structure
+s3d_structure_from_display_mode(const struct drm_display_mode *mode)
+{
+	u32 layout = mode->flags & DRM_MODE_FLAG_3D_MASK;
+
+	switch (layout) {
+	case DRM_MODE_FLAG_3D_FRAME_PACKING:
+		return HDMI_3D_STRUCTURE_FRAME_PACKING;
+	case DRM_MODE_FLAG_3D_FIELD_ALTERNATIVE:
+		return HDMI_3D_STRUCTURE_FIELD_ALTERNATIVE;
+	case DRM_MODE_FLAG_3D_LINE_ALTERNATIVE:
+		return HDMI_3D_STRUCTURE_LINE_ALTERNATIVE;
+	case DRM_MODE_FLAG_3D_SIDE_BY_SIDE_FULL:
+		return HDMI_3D_STRUCTURE_SIDE_BY_SIDE_FULL;
+	case DRM_MODE_FLAG_3D_L_DEPTH:
+		return HDMI_3D_STRUCTURE_L_DEPTH;
+	case DRM_MODE_FLAG_3D_L_DEPTH_GFX_GFX_DEPTH:
+		return HDMI_3D_STRUCTURE_L_DEPTH_GFX_GFX_DEPTH;
+	case DRM_MODE_FLAG_3D_TOP_AND_BOTTOM:
+		return HDMI_3D_STRUCTURE_TOP_AND_BOTTOM;
+	case DRM_MODE_FLAG_3D_SIDE_BY_SIDE_HALF:
+		return HDMI_3D_STRUCTURE_SIDE_BY_SIDE_HALF;
+	default:
+		return HDMI_3D_STRUCTURE_INVALID;
+	}
+}
+
 /**
  * drm_hdmi_vendor_infoframe_from_display_mode() - fill an HDMI infoframe with
  * data from a DRM display mode
@@ -3338,20 +3577,29 @@ drm_hdmi_vendor_infoframe_from_display_mode(struct hdmi_vendor_infoframe *frame,
 					    const struct drm_display_mode *mode)
 {
 	int err;
+	u32 s3d_flags;
 	u8 vic;
 
 	if (!frame || !mode)
 		return -EINVAL;
 
 	vic = drm_match_hdmi_mode(mode);
-	if (!vic)
+	s3d_flags = mode->flags & DRM_MODE_FLAG_3D_MASK;
+
+	if (!vic && !s3d_flags)
+		return -EINVAL;
+
+	if (vic && s3d_flags)
 		return -EINVAL;
 
 	err = hdmi_vendor_infoframe_init(frame);
 	if (err < 0)
 		return err;
 
-	frame->vic = vic;
+	if (vic)
+		frame->vic = vic;
+	else
+		frame->s3d_struct = s3d_structure_from_display_mode(mode);
 
 	return 0;
 }
diff --git a/drivers/gpu/drm/drm_edid_load.c b/drivers/gpu/drm/drm_edid_load.c
index 271b42bbfb72..9081172ef057 100644
--- a/drivers/gpu/drm/drm_edid_load.c
+++ b/drivers/gpu/drm/drm_edid_load.c
@@ -32,7 +32,7 @@ MODULE_PARM_DESC(edid_firmware, "Do not probe monitor, use specified EDID blob "
 	"from built-in data or /lib/firmware instead. ");
 
 #define GENERIC_EDIDS 5
-static char *generic_edid_name[GENERIC_EDIDS] = {
+static const char *generic_edid_name[GENERIC_EDIDS] = {
 	"edid/1024x768.bin",
 	"edid/1280x1024.bin",
 	"edid/1600x1200.bin",
@@ -40,7 +40,7 @@ static char *generic_edid_name[GENERIC_EDIDS] = {
 	"edid/1920x1080.bin",
 };
 
-static u8 generic_edid[GENERIC_EDIDS][128] = {
+static const u8 generic_edid[GENERIC_EDIDS][128] = {
 	{
 	0x00, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0x00,
 	0x31, 0xd8, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00,
@@ -133,63 +133,68 @@ static u8 generic_edid[GENERIC_EDIDS][128] = {
 	},
 };
 
+static int edid_size(const u8 *edid, int data_size)
+{
+	if (data_size < EDID_LENGTH)
+		return 0;
+
+	return (edid[0x7e] + 1) * EDID_LENGTH;
+}
+
 static u8 *edid_load(struct drm_connector *connector, const char *name,
 			const char *connector_name)
 {
-	const struct firmware *fw;
-	struct platform_device *pdev;
-	u8 *fwdata = NULL, *edid, *new_edid;
-	int fwsize, expected;
-	int builtin = 0, err = 0;
+	const struct firmware *fw = NULL;
+	const u8 *fwdata;
+	u8 *edid;
+	int fwsize, builtin;
 	int i, valid_extensions = 0;
 	bool print_bad_edid = !connector->bad_edid_counter || (drm_debug & DRM_UT_KMS);
 
-	pdev = platform_device_register_simple(connector_name, -1, NULL, 0);
-	if (IS_ERR(pdev)) {
-		DRM_ERROR("Failed to register EDID firmware platform device "
-		    "for connector \"%s\"\n", connector_name);
-		err = -EINVAL;
-		goto out;
-	}
-
-	err = request_firmware(&fw, name, &pdev->dev);
-	platform_device_unregister(pdev);
-
-	if (err) {
-		i = 0;
-		while (i < GENERIC_EDIDS && strcmp(name, generic_edid_name[i]))
-			i++;
-		if (i < GENERIC_EDIDS) {
-			err = 0;
-			builtin = 1;
+	builtin = 0;
+	for (i = 0; i < GENERIC_EDIDS; i++) {
+		if (strcmp(name, generic_edid_name[i]) == 0) {
 			fwdata = generic_edid[i];
 			fwsize = sizeof(generic_edid[i]);
+			builtin = 1;
+			break;
 		}
 	}
+	if (!builtin) {
+		struct platform_device *pdev;
+		int err;
 
-	if (err) {
-		DRM_ERROR("Requesting EDID firmware \"%s\" failed (err=%d)\n",
-		    name, err);
-		goto out;
-	}
+		pdev = platform_device_register_simple(connector_name, -1, NULL, 0);
+		if (IS_ERR(pdev)) {
+			DRM_ERROR("Failed to register EDID firmware platform device "
+				  "for connector \"%s\"\n", connector_name);
+			return ERR_CAST(pdev);
+		}
+
+		err = request_firmware(&fw, name, &pdev->dev);
+		platform_device_unregister(pdev);
+		if (err) {
+			DRM_ERROR("Requesting EDID firmware \"%s\" failed (err=%d)\n",
+				  name, err);
+			return ERR_PTR(err);
+		}
 
-	if (fwdata == NULL) {
-		fwdata = (u8 *) fw->data;
+		fwdata = fw->data;
 		fwsize = fw->size;
 	}
 
-	expected = (fwdata[0x7e] + 1) * EDID_LENGTH;
-	if (expected != fwsize) {
+	if (edid_size(fwdata, fwsize) != fwsize) {
 		DRM_ERROR("Size of EDID firmware \"%s\" is invalid "
-		    "(expected %d, got %d)\n", name, expected, (int) fwsize);
-		err = -EINVAL;
-		goto relfw_out;
+			  "(expected %d, got %d\n", name,
+			  edid_size(fwdata, fwsize), (int)fwsize);
+		edid = ERR_PTR(-EINVAL);
+		goto out;
 	}
 
 	edid = kmemdup(fwdata, fwsize, GFP_KERNEL);
 	if (edid == NULL) {
-		err = -ENOMEM;
-		goto relfw_out;
+		edid = ERR_PTR(-ENOMEM);
+		goto out;
 	}
 
 	if (!drm_edid_block_valid(edid, 0, print_bad_edid)) {
@@ -197,8 +202,8 @@ static u8 *edid_load(struct drm_connector *connector, const char *name,
 		DRM_ERROR("Base block of EDID firmware \"%s\" is invalid ",
 		    name);
 		kfree(edid);
-		err = -EINVAL;
-		goto relfw_out;
+		edid = ERR_PTR(-EINVAL);
+		goto out;
 	}
 
 	for (i = 1; i <= edid[0x7e]; i++) {
@@ -210,19 +215,18 @@ static u8 *edid_load(struct drm_connector *connector, const char *name,
 	}
 
 	if (valid_extensions != edid[0x7e]) {
+		u8 *new_edid;
+
 		edid[EDID_LENGTH-1] += edid[0x7e] - valid_extensions;
 		DRM_INFO("Found %d valid extensions instead of %d in EDID data "
 		    "\"%s\" for connector \"%s\"\n", valid_extensions,
 		    edid[0x7e], name, connector_name);
 		edid[0x7e] = valid_extensions;
+
 		new_edid = krealloc(edid, (valid_extensions + 1) * EDID_LENGTH,
-		    GFP_KERNEL);
-		if (new_edid == NULL) {
-			err = -ENOMEM;
-			kfree(edid);
-			goto relfw_out;
-		}
-		edid = new_edid;
+				    GFP_KERNEL);
+		if (new_edid)
+			edid = new_edid;
 	}
 
 	DRM_INFO("Got %s EDID base block and %d extension%s from "
@@ -230,13 +234,9 @@ static u8 *edid_load(struct drm_connector *connector, const char *name,
 	    "external", valid_extensions, valid_extensions == 1 ? "" : "s",
 	    name, connector_name);
 
-relfw_out:
-	release_firmware(fw);
-
 out:
-	if (err)
-		return ERR_PTR(err);
-
+	if (fw)
+		release_firmware(fw);
 	return edid;
 }
 
diff --git a/drivers/gpu/drm/drm_encoder_slave.c b/drivers/gpu/drm/drm_encoder_slave.c
index 0cfb60f54766..d18b88b755c3 100644
--- a/drivers/gpu/drm/drm_encoder_slave.c
+++ b/drivers/gpu/drm/drm_encoder_slave.c
@@ -67,12 +67,12 @@ int drm_i2c_encoder_init(struct drm_device *dev,
 		goto fail;
 	}
 
-	if (!client->driver) {
+	if (!client->dev.driver) {
 		err = -ENODEV;
 		goto fail_unregister;
 	}
 
-	module = client->driver->driver.owner;
+	module = client->dev.driver->owner;
 	if (!try_module_get(module)) {
 		err = -ENODEV;
 		goto fail_unregister;
@@ -80,7 +80,7 @@ int drm_i2c_encoder_init(struct drm_device *dev,
 
 	encoder->bus_priv = client;
 
-	encoder_drv = to_drm_i2c_encoder_driver(client->driver);
+	encoder_drv = to_drm_i2c_encoder_driver(to_i2c_driver(client->dev.driver));
 
 	err = encoder_drv->encoder_init(client, dev, encoder);
 	if (err)
@@ -111,7 +111,7 @@ void drm_i2c_encoder_destroy(struct drm_encoder *drm_encoder)
 {
 	struct drm_encoder_slave *encoder = to_encoder_slave(drm_encoder);
 	struct i2c_client *client = drm_i2c_encoder_get_client(drm_encoder);
-	struct module *module = client->driver->driver.owner;
+	struct module *module = client->dev.driver->owner;
 
 	i2c_unregister_device(client);
 	encoder->bus_priv = NULL;
diff --git a/drivers/gpu/drm/drm_fb_helper.c b/drivers/gpu/drm/drm_fb_helper.c
index 3d13ca6e257f..0a19401aff80 100644
--- a/drivers/gpu/drm/drm_fb_helper.c
+++ b/drivers/gpu/drm/drm_fb_helper.c
@@ -39,10 +39,6 @@
 #include <drm/drm_fb_helper.h>
 #include <drm/drm_crtc_helper.h>
 
-MODULE_AUTHOR("David Airlie, Jesse Barnes");
-MODULE_DESCRIPTION("DRM KMS helper");
-MODULE_LICENSE("GPL and additional rights");
-
 static LIST_HEAD(kernel_fb_helper_list);
 
 /**
@@ -844,7 +840,6 @@ int drm_fb_helper_pan_display(struct fb_var_screeninfo *var,
 	struct drm_fb_helper *fb_helper = info->par;
 	struct drm_device *dev = fb_helper->dev;
 	struct drm_mode_set *modeset;
-	struct drm_crtc *crtc;
 	int ret = 0;
 	int i;
 
@@ -855,8 +850,6 @@ int drm_fb_helper_pan_display(struct fb_var_screeninfo *var,
 	}
 
 	for (i = 0; i < fb_helper->crtc_count; i++) {
-		crtc = fb_helper->crtc_info[i].mode_set.crtc;
-
 		modeset = &fb_helper->crtc_info[i].mode_set;
 
 		modeset->x = var->xoffset;
@@ -1352,7 +1345,6 @@ static int drm_pick_crtcs(struct drm_fb_helper *fb_helper,
 	struct drm_connector *connector;
 	struct drm_connector_helper_funcs *connector_funcs;
 	struct drm_encoder *encoder;
-	struct drm_fb_helper_crtc *best_crtc;
 	int my_score, best_score, score;
 	struct drm_fb_helper_crtc **crtcs, *crtc;
 	struct drm_fb_helper_connector *fb_helper_conn;
@@ -1364,7 +1356,6 @@ static int drm_pick_crtcs(struct drm_fb_helper *fb_helper,
 	connector = fb_helper_conn->connector;
 
 	best_crtcs[n] = NULL;
-	best_crtc = NULL;
 	best_score = drm_pick_crtcs(fb_helper, best_crtcs, modes, n+1, width, height);
 	if (modes[n] == NULL)
 		return best_score;
@@ -1413,7 +1404,6 @@ static int drm_pick_crtcs(struct drm_fb_helper *fb_helper,
 		score = my_score + drm_pick_crtcs(fb_helper, crtcs, modes, n + 1,
 						  width, height);
 		if (score > best_score) {
-			best_crtc = crtc;
 			best_score = score;
 			memcpy(best_crtcs, crtcs,
 			       dev->mode_config.num_connector *
@@ -1580,8 +1570,7 @@ EXPORT_SYMBOL(drm_fb_helper_initial_config);
 int drm_fb_helper_hotplug_event(struct drm_fb_helper *fb_helper)
 {
 	struct drm_device *dev = fb_helper->dev;
-	int count = 0;
-	u32 max_width, max_height, bpp_sel;
+	u32 max_width, max_height;
 
 	if (!fb_helper->fb)
 		return 0;
@@ -1596,10 +1585,8 @@ int drm_fb_helper_hotplug_event(struct drm_fb_helper *fb_helper)
 
 	max_width = fb_helper->fb->width;
 	max_height = fb_helper->fb->height;
-	bpp_sel = fb_helper->fb->bits_per_pixel;
 
-	count = drm_fb_helper_probe_connector_modes(fb_helper, max_width,
-						    max_height);
+	drm_fb_helper_probe_connector_modes(fb_helper, max_width, max_height);
 	mutex_unlock(&fb_helper->dev->mode_config.mutex);
 
 	drm_modeset_lock_all(dev);
diff --git a/drivers/gpu/drm/drm_flip_work.c b/drivers/gpu/drm/drm_flip_work.c
index e788882d9021..f9c7fa3d0012 100644
--- a/drivers/gpu/drm/drm_flip_work.c
+++ b/drivers/gpu/drm/drm_flip_work.c
@@ -34,7 +34,7 @@
  */
 void drm_flip_work_queue(struct drm_flip_work *work, void *val)
 {
-	if (kfifo_put(&work->fifo, (const void **)&val)) {
+	if (kfifo_put(&work->fifo, val)) {
 		atomic_inc(&work->pending);
 	} else {
 		DRM_ERROR("%s fifo full!\n", work->name);
diff --git a/drivers/gpu/drm/drm_fops.c b/drivers/gpu/drm/drm_fops.c
index 22d14ecbd3ec..c5b929c3f77a 100644
--- a/drivers/gpu/drm/drm_fops.c
+++ b/drivers/gpu/drm/drm_fops.c
@@ -113,7 +113,6 @@ int drm_open(struct inode *inode, struct file *filp)
 	retcode = drm_open_helper(inode, filp, dev);
 	if (retcode)
 		goto err_undo;
-	atomic_inc(&dev->counts[_DRM_STAT_OPENS]);
 	if (need_setup) {
 		retcode = drm_setup(dev);
 		if (retcode)
@@ -235,7 +234,8 @@ static int drm_open_helper(struct inode *inode, struct file *filp,
 
 	priv->ioctl_count = 0;
 	/* for compatibility root is always authenticated */
-	priv->authenticated = capable(CAP_SYS_ADMIN);
+	priv->always_authenticated = capable(CAP_SYS_ADMIN);
+	priv->authenticated = priv->always_authenticated;
 	priv->lock_count = 0;
 
 	INIT_LIST_HEAD(&priv->lhead);
@@ -374,13 +374,80 @@ static void drm_events_release(struct drm_file *file_priv)
 		}
 
 	/* Remove unconsumed events */
-	list_for_each_entry_safe(e, et, &file_priv->event_list, link)
+	list_for_each_entry_safe(e, et, &file_priv->event_list, link) {
+		list_del(&e->link);
 		e->destroy(e);
+	}
 
 	spin_unlock_irqrestore(&dev->event_lock, flags);
 }
 
 /**
+ * drm_legacy_dev_reinit
+ *
+ * Reinitializes a legacy/ums drm device in it's lastclose function.
+ */
+static void drm_legacy_dev_reinit(struct drm_device *dev)
+{
+	if (drm_core_check_feature(dev, DRIVER_MODESET))
+		return;
+
+	atomic_set(&dev->ioctl_count, 0);
+	atomic_set(&dev->vma_count, 0);
+
+	dev->sigdata.lock = NULL;
+
+	dev->context_flag = 0;
+	dev->last_context = 0;
+	dev->if_version = 0;
+}
+
+/**
+ * Take down the DRM device.
+ *
+ * \param dev DRM device structure.
+ *
+ * Frees every resource in \p dev.
+ *
+ * \sa drm_device
+ */
+int drm_lastclose(struct drm_device * dev)
+{
+	struct drm_vma_entry *vma, *vma_temp;
+
+	DRM_DEBUG("\n");
+
+	if (dev->driver->lastclose)
+		dev->driver->lastclose(dev);
+	DRM_DEBUG("driver lastclose completed\n");
+
+	if (dev->irq_enabled && !drm_core_check_feature(dev, DRIVER_MODESET))
+		drm_irq_uninstall(dev);
+
+	mutex_lock(&dev->struct_mutex);
+
+	drm_agp_clear(dev);
+
+	drm_legacy_sg_cleanup(dev);
+
+	/* Clear vma list (only built for debugging) */
+	list_for_each_entry_safe(vma, vma_temp, &dev->vmalist, head) {
+		list_del(&vma->head);
+		kfree(vma);
+	}
+
+	drm_legacy_dma_takedown(dev);
+
+	dev->dev_mapping = NULL;
+	mutex_unlock(&dev->struct_mutex);
+
+	drm_legacy_dev_reinit(dev);
+
+	DRM_DEBUG("lastclose completed\n");
+	return 0;
+}
+
+/**
  * Release file.
  *
  * \param inode device inode
@@ -449,7 +516,6 @@ int drm_release(struct inode *inode, struct file *filp)
 
 				list_del(&pos->head);
 				kfree(pos);
-				--dev->ctx_count;
 			}
 		}
 	}
@@ -463,7 +529,7 @@ int drm_release(struct inode *inode, struct file *filp)
 		list_for_each_entry(temp, &dev->filelist, lhead) {
 			if ((temp->master == file_priv->master) &&
 			    (temp != file_priv))
-				temp->authenticated = 0;
+				temp->authenticated = temp->always_authenticated;
 		}
 
 		/**
@@ -511,7 +577,6 @@ int drm_release(struct inode *inode, struct file *filp)
 	 * End inline drm_release
 	 */
 
-	atomic_inc(&dev->counts[_DRM_STAT_CLOSES]);
 	if (!--dev->open_count) {
 		if (atomic_read(&dev->ioctl_count)) {
 			DRM_ERROR("Device busy: %d\n",
diff --git a/drivers/gpu/drm/drm_gem.c b/drivers/gpu/drm/drm_gem.c
index 49293bdc972a..4761adedad2a 100644
--- a/drivers/gpu/drm/drm_gem.c
+++ b/drivers/gpu/drm/drm_gem.c
@@ -160,35 +160,6 @@ void drm_gem_private_object_init(struct drm_device *dev,
 }
 EXPORT_SYMBOL(drm_gem_private_object_init);
 
-/**
- * Allocate a GEM object of the specified size with shmfs backing store
- */
-struct drm_gem_object *
-drm_gem_object_alloc(struct drm_device *dev, size_t size)
-{
-	struct drm_gem_object *obj;
-
-	obj = kzalloc(sizeof(*obj), GFP_KERNEL);
-	if (!obj)
-		goto free;
-
-	if (drm_gem_object_init(dev, obj, size) != 0)
-		goto free;
-
-	if (dev->driver->gem_init_object != NULL &&
-	    dev->driver->gem_init_object(obj) != 0) {
-		goto fput;
-	}
-	return obj;
-fput:
-	/* Object_init mangles the global counters - readjust them. */
-	fput(obj->filp);
-free:
-	kfree(obj);
-	return NULL;
-}
-EXPORT_SYMBOL(drm_gem_object_alloc);
-
 static void
 drm_gem_remove_prime_handles(struct drm_gem_object *obj, struct drm_file *filp)
 {
diff --git a/drivers/gpu/drm/drm_global.c b/drivers/gpu/drm/drm_global.c
index f7311162a61d..3d2e91c4d78e 100644
--- a/drivers/gpu/drm/drm_global.c
+++ b/drivers/gpu/drm/drm_global.c
@@ -67,7 +67,6 @@ int drm_global_item_ref(struct drm_global_reference *ref)
 {
 	int ret;
 	struct drm_global_item *item = &glob[ref->global_type];
-	void *object;
 
 	mutex_lock(&item->mutex);
 	if (item->refcount == 0) {
@@ -85,7 +84,6 @@ int drm_global_item_ref(struct drm_global_reference *ref)
 	}
 	++item->refcount;
 	ref->object = item->object;
-	object = item->object;
 	mutex_unlock(&item->mutex);
 	return 0;
 out_err:
diff --git a/drivers/gpu/drm/drm_info.c b/drivers/gpu/drm/drm_info.c
index 53298320080b..7d5a152eeb02 100644
--- a/drivers/gpu/drm/drm_info.c
+++ b/drivers/gpu/drm/drm_info.c
@@ -163,13 +163,13 @@ int drm_vblank_info(struct seq_file *m, void *data)
 	mutex_lock(&dev->struct_mutex);
 	for (crtc = 0; crtc < dev->num_crtcs; crtc++) {
 		seq_printf(m, "CRTC %d enable:     %d\n",
-			   crtc, atomic_read(&dev->vblank_refcount[crtc]));
+			   crtc, atomic_read(&dev->vblank[crtc].refcount));
 		seq_printf(m, "CRTC %d counter:    %d\n",
 			   crtc, drm_vblank_count(dev, crtc));
 		seq_printf(m, "CRTC %d last wait:  %d\n",
-			   crtc, dev->last_vblank_wait[crtc]);
+			   crtc, dev->vblank[crtc].last_wait);
 		seq_printf(m, "CRTC %d in modeset: %d\n",
-			   crtc, dev->vblank_inmodeset[crtc]);
+			   crtc, dev->vblank[crtc].inmodeset);
 	}
 	mutex_unlock(&dev->struct_mutex);
 	return 0;
diff --git a/drivers/gpu/drm/drm_ioctl.c b/drivers/gpu/drm/drm_ioctl.c
index 07247e2855a2..dffc836144cc 100644
--- a/drivers/gpu/drm/drm_ioctl.c
+++ b/drivers/gpu/drm/drm_ioctl.c
@@ -303,6 +303,27 @@ int drm_getcap(struct drm_device *dev, void *data, struct drm_file *file_priv)
 }
 
 /**
+ * Set device/driver capabilities
+ */
+int
+drm_setclientcap(struct drm_device *dev, void *data, struct drm_file *file_priv)
+{
+	struct drm_set_client_cap *req = data;
+
+	switch (req->capability) {
+	case DRM_CLIENT_CAP_STEREO_3D:
+		if (req->value > 1)
+			return -EINVAL;
+		file_priv->stereo_allowed = req->value;
+		break;
+	default:
+		return -EINVAL;
+	}
+
+	return 0;
+}
+
+/**
  * Setversion ioctl.
  *
  * \param inode device inode.
diff --git a/drivers/gpu/drm/drm_irq.c b/drivers/gpu/drm/drm_irq.c
index f92da0a32f0d..64c34d5876ff 100644
--- a/drivers/gpu/drm/drm_irq.c
+++ b/drivers/gpu/drm/drm_irq.c
@@ -43,9 +43,8 @@
 #include <linux/export.h>
 
 /* Access macro for slots in vblank timestamp ringbuffer. */
-#define vblanktimestamp(dev, crtc, count) ( \
-	(dev)->_vblank_time[(crtc) * DRM_VBLANKTIME_RBSIZE + \
-	((count) % DRM_VBLANKTIME_RBSIZE)])
+#define vblanktimestamp(dev, crtc, count) \
+	((dev)->vblank[crtc].time[(count) % DRM_VBLANKTIME_RBSIZE])
 
 /* Retry timestamp calculation up to 3 times to satisfy
  * drm_timestamp_precision before giving up.
@@ -89,8 +88,7 @@ int drm_irq_by_busid(struct drm_device *dev, void *data,
  */
 static void clear_vblank_timestamps(struct drm_device *dev, int crtc)
 {
-	memset(&dev->_vblank_time[crtc * DRM_VBLANKTIME_RBSIZE], 0,
-		DRM_VBLANKTIME_RBSIZE * sizeof(struct timeval));
+	memset(dev->vblank[crtc].time, 0, sizeof(dev->vblank[crtc].time));
 }
 
 /*
@@ -115,7 +113,7 @@ static void vblank_disable_and_save(struct drm_device *dev, int crtc)
 	spin_lock_irqsave(&dev->vblank_time_lock, irqflags);
 
 	dev->driver->disable_vblank(dev, crtc);
-	dev->vblank_enabled[crtc] = 0;
+	dev->vblank[crtc].enabled = false;
 
 	/* No further vblank irq's will be processed after
 	 * this point. Get current hardware vblank count and
@@ -130,9 +128,9 @@ static void vblank_disable_and_save(struct drm_device *dev, int crtc)
 	 * delayed gpu counter increment.
 	 */
 	do {
-		dev->last_vblank[crtc] = dev->driver->get_vblank_counter(dev, crtc);
+		dev->vblank[crtc].last = dev->driver->get_vblank_counter(dev, crtc);
 		vblrc = drm_get_last_vbltimestamp(dev, crtc, &tvblank, 0);
-	} while (dev->last_vblank[crtc] != dev->driver->get_vblank_counter(dev, crtc) && (--count) && vblrc);
+	} while (dev->vblank[crtc].last != dev->driver->get_vblank_counter(dev, crtc) && (--count) && vblrc);
 
 	if (!count)
 		vblrc = 0;
@@ -140,7 +138,7 @@ static void vblank_disable_and_save(struct drm_device *dev, int crtc)
 	/* Compute time difference to stored timestamp of last vblank
 	 * as updated by last invocation of drm_handle_vblank() in vblank irq.
 	 */
-	vblcount = atomic_read(&dev->_vblank_count[crtc]);
+	vblcount = atomic_read(&dev->vblank[crtc].count);
 	diff_ns = timeval_to_ns(&tvblank) -
 		  timeval_to_ns(&vblanktimestamp(dev, crtc, vblcount));
 
@@ -157,7 +155,7 @@ static void vblank_disable_and_save(struct drm_device *dev, int crtc)
 	 * hope for the best.
 	 */
 	if ((vblrc > 0) && (abs64(diff_ns) > 1000000)) {
-		atomic_inc(&dev->_vblank_count[crtc]);
+		atomic_inc(&dev->vblank[crtc].count);
 		smp_mb__after_atomic_inc();
 	}
 
@@ -178,8 +176,8 @@ static void vblank_disable_fn(unsigned long arg)
 
 	for (i = 0; i < dev->num_crtcs; i++) {
 		spin_lock_irqsave(&dev->vbl_lock, irqflags);
-		if (atomic_read(&dev->vblank_refcount[i]) == 0 &&
-		    dev->vblank_enabled[i]) {
+		if (atomic_read(&dev->vblank[i].refcount) == 0 &&
+		    dev->vblank[i].enabled) {
 			DRM_DEBUG("disabling vblank on crtc %d\n", i);
 			vblank_disable_and_save(dev, i);
 		}
@@ -197,14 +195,7 @@ void drm_vblank_cleanup(struct drm_device *dev)
 
 	vblank_disable_fn((unsigned long)dev);
 
-	kfree(dev->vbl_queue);
-	kfree(dev->_vblank_count);
-	kfree(dev->vblank_refcount);
-	kfree(dev->vblank_enabled);
-	kfree(dev->last_vblank);
-	kfree(dev->last_vblank_wait);
-	kfree(dev->vblank_inmodeset);
-	kfree(dev->_vblank_time);
+	kfree(dev->vblank);
 
 	dev->num_crtcs = 0;
 }
@@ -221,42 +212,14 @@ int drm_vblank_init(struct drm_device *dev, int num_crtcs)
 
 	dev->num_crtcs = num_crtcs;
 
-	dev->vbl_queue = kmalloc(sizeof(wait_queue_head_t) * num_crtcs,
-				 GFP_KERNEL);
-	if (!dev->vbl_queue)
+	dev->vblank = kcalloc(num_crtcs, sizeof(*dev->vblank), GFP_KERNEL);
+	if (!dev->vblank)
 		goto err;
 
-	dev->_vblank_count = kmalloc(sizeof(atomic_t) * num_crtcs, GFP_KERNEL);
-	if (!dev->_vblank_count)
-		goto err;
-
-	dev->vblank_refcount = kmalloc(sizeof(atomic_t) * num_crtcs,
-				       GFP_KERNEL);
-	if (!dev->vblank_refcount)
-		goto err;
-
-	dev->vblank_enabled = kcalloc(num_crtcs, sizeof(int), GFP_KERNEL);
-	if (!dev->vblank_enabled)
-		goto err;
+	for (i = 0; i < num_crtcs; i++)
+		init_waitqueue_head(&dev->vblank[i].queue);
 
-	dev->last_vblank = kcalloc(num_crtcs, sizeof(u32), GFP_KERNEL);
-	if (!dev->last_vblank)
-		goto err;
-
-	dev->last_vblank_wait = kcalloc(num_crtcs, sizeof(u32), GFP_KERNEL);
-	if (!dev->last_vblank_wait)
-		goto err;
-
-	dev->vblank_inmodeset = kcalloc(num_crtcs, sizeof(int), GFP_KERNEL);
-	if (!dev->vblank_inmodeset)
-		goto err;
-
-	dev->_vblank_time = kcalloc(num_crtcs * DRM_VBLANKTIME_RBSIZE,
-				    sizeof(struct timeval), GFP_KERNEL);
-	if (!dev->_vblank_time)
-		goto err;
-
-	DRM_INFO("Supports vblank timestamp caching Rev 1 (10.10.2010).\n");
+	DRM_INFO("Supports vblank timestamp caching Rev 2 (21.10.2013).\n");
 
 	/* Driver specific high-precision vblank timestamping supported? */
 	if (dev->driver->get_vblank_timestamp)
@@ -264,14 +227,8 @@ int drm_vblank_init(struct drm_device *dev, int num_crtcs)
 	else
 		DRM_INFO("No driver support for vblank timestamp query.\n");
 
-	/* Zero per-crtc vblank stuff */
-	for (i = 0; i < num_crtcs; i++) {
-		init_waitqueue_head(&dev->vbl_queue[i]);
-		atomic_set(&dev->_vblank_count[i], 0);
-		atomic_set(&dev->vblank_refcount[i], 0);
-	}
+	dev->vblank_disable_allowed = false;
 
-	dev->vblank_disable_allowed = 0;
 	return 0;
 
 err:
@@ -336,7 +293,7 @@ int drm_irq_install(struct drm_device *dev)
 		mutex_unlock(&dev->struct_mutex);
 		return -EBUSY;
 	}
-	dev->irq_enabled = 1;
+	dev->irq_enabled = true;
 	mutex_unlock(&dev->struct_mutex);
 
 	DRM_DEBUG("irq=%d\n", drm_dev_to_irq(dev));
@@ -359,7 +316,7 @@ int drm_irq_install(struct drm_device *dev)
 
 	if (ret < 0) {
 		mutex_lock(&dev->struct_mutex);
-		dev->irq_enabled = 0;
+		dev->irq_enabled = false;
 		mutex_unlock(&dev->struct_mutex);
 		return ret;
 	}
@@ -373,7 +330,7 @@ int drm_irq_install(struct drm_device *dev)
 
 	if (ret < 0) {
 		mutex_lock(&dev->struct_mutex);
-		dev->irq_enabled = 0;
+		dev->irq_enabled = false;
 		mutex_unlock(&dev->struct_mutex);
 		if (!drm_core_check_feature(dev, DRIVER_MODESET))
 			vga_client_register(dev->pdev, NULL, NULL, NULL);
@@ -394,14 +351,15 @@ EXPORT_SYMBOL(drm_irq_install);
 int drm_irq_uninstall(struct drm_device *dev)
 {
 	unsigned long irqflags;
-	int irq_enabled, i;
+	bool irq_enabled;
+	int i;
 
 	if (!drm_core_check_feature(dev, DRIVER_HAVE_IRQ))
 		return -EINVAL;
 
 	mutex_lock(&dev->struct_mutex);
 	irq_enabled = dev->irq_enabled;
-	dev->irq_enabled = 0;
+	dev->irq_enabled = false;
 	mutex_unlock(&dev->struct_mutex);
 
 	/*
@@ -410,9 +368,9 @@ int drm_irq_uninstall(struct drm_device *dev)
 	if (dev->num_crtcs) {
 		spin_lock_irqsave(&dev->vbl_lock, irqflags);
 		for (i = 0; i < dev->num_crtcs; i++) {
-			DRM_WAKEUP(&dev->vbl_queue[i]);
-			dev->vblank_enabled[i] = 0;
-			dev->last_vblank[i] =
+			DRM_WAKEUP(&dev->vblank[i].queue);
+			dev->vblank[i].enabled = false;
+			dev->vblank[i].last =
 				dev->driver->get_vblank_counter(dev, i);
 		}
 		spin_unlock_irqrestore(&dev->vbl_lock, irqflags);
@@ -497,8 +455,8 @@ void drm_calc_timestamping_constants(struct drm_crtc *crtc)
 	/* Dot clock in Hz: */
 	dotclock = (u64) crtc->hwmode.clock * 1000;
 
-	/* Fields of interlaced scanout modes are only halve a frame duration.
-	 * Double the dotclock to get halve the frame-/line-/pixelduration.
+	/* Fields of interlaced scanout modes are only half a frame duration.
+	 * Double the dotclock to get half the frame-/line-/pixelduration.
 	 */
 	if (crtc->hwmode.flags & DRM_MODE_FLAG_INTERLACE)
 		dotclock *= 2;
@@ -628,24 +586,20 @@ int drm_calc_vbltimestamp_from_scanoutpos(struct drm_device *dev, int crtc,
 	 * code gets preempted or delayed for some reason.
 	 */
 	for (i = 0; i < DRM_TIMESTAMP_MAXRETRIES; i++) {
-		/* Disable preemption to make it very likely to
-		 * succeed in the first iteration even on PREEMPT_RT kernel.
+		/*
+		 * Get vertical and horizontal scanout position vpos, hpos,
+		 * and bounding timestamps stime, etime, pre/post query.
 		 */
-		preempt_disable();
+		vbl_status = dev->driver->get_scanout_position(dev, crtc, &vpos,
+							       &hpos, &stime, &etime);
 
-		/* Get system timestamp before query. */
-		stime = ktime_get();
-
-		/* Get vertical and horizontal scanout pos. vpos, hpos. */
-		vbl_status = dev->driver->get_scanout_position(dev, crtc, &vpos, &hpos);
-
-		/* Get system timestamp after query. */
-		etime = ktime_get();
+		/*
+		 * Get correction for CLOCK_MONOTONIC -> CLOCK_REALTIME if
+		 * CLOCK_REALTIME is requested.
+		 */
 		if (!drm_timestamp_monotonic)
 			mono_time_offset = ktime_get_monotonic_offset();
 
-		preempt_enable();
-
 		/* Return as no-op if scanout query unsupported or failed. */
 		if (!(vbl_status & DRM_SCANOUTPOS_VALID)) {
 			DRM_DEBUG("crtc %d : scanoutpos query failed [%d].\n",
@@ -653,6 +607,7 @@ int drm_calc_vbltimestamp_from_scanoutpos(struct drm_device *dev, int crtc,
 			return -EIO;
 		}
 
+		/* Compute uncertainty in timestamp of scanout position query. */
 		duration_ns = ktime_to_ns(etime) - ktime_to_ns(stime);
 
 		/* Accept result with <  max_error nsecs timing uncertainty. */
@@ -795,7 +750,7 @@ EXPORT_SYMBOL(drm_get_last_vbltimestamp);
  */
 u32 drm_vblank_count(struct drm_device *dev, int crtc)
 {
-	return atomic_read(&dev->_vblank_count[crtc]);
+	return atomic_read(&dev->vblank[crtc].count);
 }
 EXPORT_SYMBOL(drm_vblank_count);
 
@@ -824,10 +779,10 @@ u32 drm_vblank_count_and_time(struct drm_device *dev, int crtc,
 	 * a seqlock.
 	 */
 	do {
-		cur_vblank = atomic_read(&dev->_vblank_count[crtc]);
+		cur_vblank = atomic_read(&dev->vblank[crtc].count);
 		*vblanktime = vblanktimestamp(dev, crtc, cur_vblank);
 		smp_rmb();
-	} while (cur_vblank != atomic_read(&dev->_vblank_count[crtc]));
+	} while (cur_vblank != atomic_read(&dev->vblank[crtc].count));
 
 	return cur_vblank;
 }
@@ -914,12 +869,12 @@ static void drm_update_vblank_count(struct drm_device *dev, int crtc)
 	} while (cur_vblank != dev->driver->get_vblank_counter(dev, crtc));
 
 	/* Deal with counter wrap */
-	diff = cur_vblank - dev->last_vblank[crtc];
-	if (cur_vblank < dev->last_vblank[crtc]) {
+	diff = cur_vblank - dev->vblank[crtc].last;
+	if (cur_vblank < dev->vblank[crtc].last) {
 		diff += dev->max_vblank_count;
 
 		DRM_DEBUG("last_vblank[%d]=0x%x, cur_vblank=0x%x => diff=0x%x\n",
-			  crtc, dev->last_vblank[crtc], cur_vblank, diff);
+			  crtc, dev->vblank[crtc].last, cur_vblank, diff);
 	}
 
 	DRM_DEBUG("enabling vblank interrupts on crtc %d, missed %d\n",
@@ -930,12 +885,12 @@ static void drm_update_vblank_count(struct drm_device *dev, int crtc)
 	 * reinitialize delayed at next vblank interrupt in that case.
 	 */
 	if (rc) {
-		tslot = atomic_read(&dev->_vblank_count[crtc]) + diff;
+		tslot = atomic_read(&dev->vblank[crtc].count) + diff;
 		vblanktimestamp(dev, crtc, tslot) = t_vblank;
 	}
 
 	smp_mb__before_atomic_inc();
-	atomic_add(diff, &dev->_vblank_count[crtc]);
+	atomic_add(diff, &dev->vblank[crtc].count);
 	smp_mb__after_atomic_inc();
 }
 
@@ -957,9 +912,9 @@ int drm_vblank_get(struct drm_device *dev, int crtc)
 
 	spin_lock_irqsave(&dev->vbl_lock, irqflags);
 	/* Going from 0->1 means we have to enable interrupts again */
-	if (atomic_add_return(1, &dev->vblank_refcount[crtc]) == 1) {
+	if (atomic_add_return(1, &dev->vblank[crtc].refcount) == 1) {
 		spin_lock_irqsave(&dev->vblank_time_lock, irqflags2);
-		if (!dev->vblank_enabled[crtc]) {
+		if (!dev->vblank[crtc].enabled) {
 			/* Enable vblank irqs under vblank_time_lock protection.
 			 * All vblank count & timestamp updates are held off
 			 * until we are done reinitializing master counter and
@@ -970,16 +925,16 @@ int drm_vblank_get(struct drm_device *dev, int crtc)
 			DRM_DEBUG("enabling vblank on crtc %d, ret: %d\n",
 				  crtc, ret);
 			if (ret)
-				atomic_dec(&dev->vblank_refcount[crtc]);
+				atomic_dec(&dev->vblank[crtc].refcount);
 			else {
-				dev->vblank_enabled[crtc] = 1;
+				dev->vblank[crtc].enabled = true;
 				drm_update_vblank_count(dev, crtc);
 			}
 		}
 		spin_unlock_irqrestore(&dev->vblank_time_lock, irqflags2);
 	} else {
-		if (!dev->vblank_enabled[crtc]) {
-			atomic_dec(&dev->vblank_refcount[crtc]);
+		if (!dev->vblank[crtc].enabled) {
+			atomic_dec(&dev->vblank[crtc].refcount);
 			ret = -EINVAL;
 		}
 	}
@@ -999,10 +954,10 @@ EXPORT_SYMBOL(drm_vblank_get);
  */
 void drm_vblank_put(struct drm_device *dev, int crtc)
 {
-	BUG_ON(atomic_read(&dev->vblank_refcount[crtc]) == 0);
+	BUG_ON(atomic_read(&dev->vblank[crtc].refcount) == 0);
 
 	/* Last user schedules interrupt disable */
-	if (atomic_dec_and_test(&dev->vblank_refcount[crtc]) &&
+	if (atomic_dec_and_test(&dev->vblank[crtc].refcount) &&
 	    (drm_vblank_offdelay > 0))
 		mod_timer(&dev->vblank_disable_timer,
 			  jiffies + ((drm_vblank_offdelay * DRM_HZ)/1000));
@@ -1025,7 +980,7 @@ void drm_vblank_off(struct drm_device *dev, int crtc)
 
 	spin_lock_irqsave(&dev->vbl_lock, irqflags);
 	vblank_disable_and_save(dev, crtc);
-	DRM_WAKEUP(&dev->vbl_queue[crtc]);
+	DRM_WAKEUP(&dev->vblank[crtc].queue);
 
 	/* Send any queued vblank events, lest the natives grow disquiet */
 	seq = drm_vblank_count_and_time(dev, crtc, &now);
@@ -1067,10 +1022,10 @@ void drm_vblank_pre_modeset(struct drm_device *dev, int crtc)
 	 * to avoid corrupting the count if multiple, mismatch calls occur),
 	 * so that interrupts remain enabled in the interim.
 	 */
-	if (!dev->vblank_inmodeset[crtc]) {
-		dev->vblank_inmodeset[crtc] = 0x1;
+	if (!dev->vblank[crtc].inmodeset) {
+		dev->vblank[crtc].inmodeset = 0x1;
 		if (drm_vblank_get(dev, crtc) == 0)
-			dev->vblank_inmodeset[crtc] |= 0x2;
+			dev->vblank[crtc].inmodeset |= 0x2;
 	}
 }
 EXPORT_SYMBOL(drm_vblank_pre_modeset);
@@ -1083,15 +1038,15 @@ void drm_vblank_post_modeset(struct drm_device *dev, int crtc)
 	if (!dev->num_crtcs)
 		return;
 
-	if (dev->vblank_inmodeset[crtc]) {
+	if (dev->vblank[crtc].inmodeset) {
 		spin_lock_irqsave(&dev->vbl_lock, irqflags);
-		dev->vblank_disable_allowed = 1;
+		dev->vblank_disable_allowed = true;
 		spin_unlock_irqrestore(&dev->vbl_lock, irqflags);
 
-		if (dev->vblank_inmodeset[crtc] & 0x2)
+		if (dev->vblank[crtc].inmodeset & 0x2)
 			drm_vblank_put(dev, crtc);
 
-		dev->vblank_inmodeset[crtc] = 0;
+		dev->vblank[crtc].inmodeset = 0;
 	}
 }
 EXPORT_SYMBOL(drm_vblank_post_modeset);
@@ -1288,8 +1243,8 @@ int drm_wait_vblank(struct drm_device *dev, void *data,
 
 	DRM_DEBUG("waiting on vblank count %d, crtc %d\n",
 		  vblwait->request.sequence, crtc);
-	dev->last_vblank_wait[crtc] = vblwait->request.sequence;
-	DRM_WAIT_ON(ret, dev->vbl_queue[crtc], 3 * DRM_HZ,
+	dev->vblank[crtc].last_wait = vblwait->request.sequence;
+	DRM_WAIT_ON(ret, dev->vblank[crtc].queue, 3 * DRM_HZ,
 		    (((drm_vblank_count(dev, crtc) -
 		       vblwait->request.sequence) <= (1 << 23)) ||
 		     !dev->irq_enabled));
@@ -1367,7 +1322,7 @@ bool drm_handle_vblank(struct drm_device *dev, int crtc)
 	spin_lock_irqsave(&dev->vblank_time_lock, irqflags);
 
 	/* Vblank irq handling disabled. Nothing to do. */
-	if (!dev->vblank_enabled[crtc]) {
+	if (!dev->vblank[crtc].enabled) {
 		spin_unlock_irqrestore(&dev->vblank_time_lock, irqflags);
 		return false;
 	}
@@ -1377,7 +1332,7 @@ bool drm_handle_vblank(struct drm_device *dev, int crtc)
 	 */
 
 	/* Get current timestamp and count. */
-	vblcount = atomic_read(&dev->_vblank_count[crtc]);
+	vblcount = atomic_read(&dev->vblank[crtc].count);
 	drm_get_last_vbltimestamp(dev, crtc, &tvblank, DRM_CALLED_FROM_VBLIRQ);
 
 	/* Compute time difference to timestamp of last vblank */
@@ -1401,14 +1356,14 @@ bool drm_handle_vblank(struct drm_device *dev, int crtc)
 		 * the timestamp computed above.
 		 */
 		smp_mb__before_atomic_inc();
-		atomic_inc(&dev->_vblank_count[crtc]);
+		atomic_inc(&dev->vblank[crtc].count);
 		smp_mb__after_atomic_inc();
 	} else {
 		DRM_DEBUG("crtc %d: Redundant vblirq ignored. diff_ns = %d\n",
 			  crtc, (int) diff_ns);
 	}
 
-	DRM_WAKEUP(&dev->vbl_queue[crtc]);
+	DRM_WAKEUP(&dev->vblank[crtc].queue);
 	drm_handle_vblank_events(dev, crtc);
 
 	spin_unlock_irqrestore(&dev->vblank_time_lock, irqflags);
diff --git a/drivers/gpu/drm/drm_lock.c b/drivers/gpu/drm/drm_lock.c
index d752c96d6090..f6452682141b 100644
--- a/drivers/gpu/drm/drm_lock.c
+++ b/drivers/gpu/drm/drm_lock.c
@@ -86,7 +86,6 @@ int drm_lock(struct drm_device *dev, void *data, struct drm_file *file_priv)
 		if (drm_lock_take(&master->lock, lock->context)) {
 			master->lock.file_priv = file_priv;
 			master->lock.lock_time = jiffies;
-			atomic_inc(&dev->counts[_DRM_STAT_LOCKS]);
 			break;	/* Got lock */
 		}
 
@@ -157,8 +156,6 @@ int drm_unlock(struct drm_device *dev, void *data, struct drm_file *file_priv)
 		return -EINVAL;
 	}
 
-	atomic_inc(&dev->counts[_DRM_STAT_UNLOCKS]);
-
 	if (drm_lock_free(&master->lock, lock->context)) {
 		/* FIXME: Should really bail out here. */
 	}
diff --git a/drivers/gpu/drm/drm_modes.c b/drivers/gpu/drm/drm_modes.c
index fc2adb62b757..85071a1c4547 100644
--- a/drivers/gpu/drm/drm_modes.c
+++ b/drivers/gpu/drm/drm_modes.c
@@ -707,18 +707,25 @@ EXPORT_SYMBOL(drm_mode_vrefresh);
 /**
  * drm_mode_set_crtcinfo - set CRTC modesetting parameters
  * @p: mode
- * @adjust_flags: unused? (FIXME)
+ * @adjust_flags: a combination of adjustment flags
  *
  * LOCKING:
  * None.
  *
  * Setup the CRTC modesetting parameters for @p, adjusting if necessary.
+ *
+ * - The CRTC_INTERLACE_HALVE_V flag can be used to halve vertical timings of
+ *   interlaced modes.
+ * - The CRTC_STEREO_DOUBLE flag can be used to compute the timings for
+ *   buffers containing two eyes (only adjust the timings when needed, eg. for
+ *   "frame packing" or "side by side full").
  */
 void drm_mode_set_crtcinfo(struct drm_display_mode *p, int adjust_flags)
 {
 	if ((p == NULL) || ((p->type & DRM_MODE_TYPE_CRTC_C) == DRM_MODE_TYPE_BUILTIN))
 		return;
 
+	p->crtc_clock = p->clock;
 	p->crtc_hdisplay = p->hdisplay;
 	p->crtc_hsync_start = p->hsync_start;
 	p->crtc_hsync_end = p->hsync_end;
@@ -752,6 +759,20 @@ void drm_mode_set_crtcinfo(struct drm_display_mode *p, int adjust_flags)
 		p->crtc_vtotal *= p->vscan;
 	}
 
+	if (adjust_flags & CRTC_STEREO_DOUBLE) {
+		unsigned int layout = p->flags & DRM_MODE_FLAG_3D_MASK;
+
+		switch (layout) {
+		case DRM_MODE_FLAG_3D_FRAME_PACKING:
+			p->crtc_clock *= 2;
+			p->crtc_vdisplay += p->crtc_vtotal;
+			p->crtc_vsync_start += p->crtc_vtotal;
+			p->crtc_vsync_end += p->crtc_vtotal;
+			p->crtc_vtotal += p->crtc_vtotal;
+			break;
+		}
+	}
+
 	p->crtc_vblank_start = min(p->crtc_vsync_start, p->crtc_vdisplay);
 	p->crtc_vblank_end = max(p->crtc_vsync_end, p->crtc_vtotal);
 	p->crtc_hblank_start = min(p->crtc_hsync_start, p->crtc_hdisplay);
@@ -830,12 +851,16 @@ bool drm_mode_equal(const struct drm_display_mode *mode1, const struct drm_displ
 	} else if (mode1->clock != mode2->clock)
 		return false;
 
-	return drm_mode_equal_no_clocks(mode1, mode2);
+	if ((mode1->flags & DRM_MODE_FLAG_3D_MASK) !=
+	    (mode2->flags & DRM_MODE_FLAG_3D_MASK))
+		return false;
+
+	return drm_mode_equal_no_clocks_no_stereo(mode1, mode2);
 }
 EXPORT_SYMBOL(drm_mode_equal);
 
 /**
- * drm_mode_equal_no_clocks - test modes for equality
+ * drm_mode_equal_no_clocks_no_stereo - test modes for equality
  * @mode1: first mode
  * @mode2: second mode
  *
@@ -843,12 +868,13 @@ EXPORT_SYMBOL(drm_mode_equal);
  * None.
  *
  * Check to see if @mode1 and @mode2 are equivalent, but
- * don't check the pixel clocks.
+ * don't check the pixel clocks nor the stereo layout.
  *
  * RETURNS:
  * True if the modes are equal, false otherwise.
  */
-bool drm_mode_equal_no_clocks(const struct drm_display_mode *mode1, const struct drm_display_mode *mode2)
+bool drm_mode_equal_no_clocks_no_stereo(const struct drm_display_mode *mode1,
+					const struct drm_display_mode *mode2)
 {
 	if (mode1->hdisplay == mode2->hdisplay &&
 	    mode1->hsync_start == mode2->hsync_start &&
@@ -860,12 +886,13 @@ bool drm_mode_equal_no_clocks(const struct drm_display_mode *mode1, const struct
 	    mode1->vsync_end == mode2->vsync_end &&
 	    mode1->vtotal == mode2->vtotal &&
 	    mode1->vscan == mode2->vscan &&
-	    mode1->flags == mode2->flags)
+	    (mode1->flags & ~DRM_MODE_FLAG_3D_MASK) ==
+	     (mode2->flags & ~DRM_MODE_FLAG_3D_MASK))
 		return true;
 
 	return false;
 }
-EXPORT_SYMBOL(drm_mode_equal_no_clocks);
+EXPORT_SYMBOL(drm_mode_equal_no_clocks_no_stereo);
 
 /**
  * drm_mode_validate_size - make sure modes adhere to size constraints
@@ -1014,7 +1041,7 @@ void drm_mode_connector_list_update(struct drm_connector *connector)
 				/* if equal delete the probed mode */
 				mode->status = pmode->status;
 				/* Merge type bits together */
-				mode->type |= pmode->type;
+				mode->type = pmode->type;
 				list_del(&pmode->head);
 				drm_mode_destroy(connector->dev, pmode);
 				break;
diff --git a/drivers/gpu/drm/drm_pci.c b/drivers/gpu/drm/drm_pci.c
index 1f96cee6eee8..02679793c9e2 100644
--- a/drivers/gpu/drm/drm_pci.c
+++ b/drivers/gpu/drm/drm_pci.c
@@ -80,7 +80,7 @@ drm_dma_handle_t *drm_pci_alloc(struct drm_device * dev, size_t size, size_t ali
 	/* Reserve */
 	for (addr = (unsigned long)dmah->vaddr, sz = size;
 	     sz > 0; addr += PAGE_SIZE, sz -= PAGE_SIZE) {
-		SetPageReserved(virt_to_page(addr));
+		SetPageReserved(virt_to_page((void *)addr));
 	}
 
 	return dmah;
@@ -103,7 +103,7 @@ void __drm_pci_free(struct drm_device * dev, drm_dma_handle_t * dmah)
 		/* Unreserve */
 		for (addr = (unsigned long)dmah->vaddr, sz = dmah->size;
 		     sz > 0; addr += PAGE_SIZE, sz -= PAGE_SIZE) {
-			ClearPageReserved(virt_to_page(addr));
+			ClearPageReserved(virt_to_page((void *)addr));
 		}
 		dma_free_coherent(&dev->pdev->dev, dmah->size, dmah->vaddr,
 				  dmah->busaddr);
@@ -322,83 +322,36 @@ int drm_get_pci_dev(struct pci_dev *pdev, const struct pci_device_id *ent,
 
 	DRM_DEBUG("\n");
 
-	dev = kzalloc(sizeof(*dev), GFP_KERNEL);
+	dev = drm_dev_alloc(driver, &pdev->dev);
 	if (!dev)
 		return -ENOMEM;
 
 	ret = pci_enable_device(pdev);
 	if (ret)
-		goto err_g1;
+		goto err_free;
 
 	dev->pdev = pdev;
-	dev->dev = &pdev->dev;
-
-	dev->pci_device = pdev->device;
-	dev->pci_vendor = pdev->vendor;
-
 #ifdef __alpha__
 	dev->hose = pdev->sysdata;
 #endif
 
-	mutex_lock(&drm_global_mutex);
-
-	if ((ret = drm_fill_in_dev(dev, ent, driver))) {
-		printk(KERN_ERR "DRM: Fill_in_dev failed.\n");
-		goto err_g2;
-	}
-
-	if (drm_core_check_feature(dev, DRIVER_MODESET)) {
+	if (drm_core_check_feature(dev, DRIVER_MODESET))
 		pci_set_drvdata(pdev, dev);
-		ret = drm_get_minor(dev, &dev->control, DRM_MINOR_CONTROL);
-		if (ret)
-			goto err_g2;
-	}
-
-	if (drm_core_check_feature(dev, DRIVER_RENDER) && drm_rnodes) {
-		ret = drm_get_minor(dev, &dev->render, DRM_MINOR_RENDER);
-		if (ret)
-			goto err_g21;
-	}
-
-	if ((ret = drm_get_minor(dev, &dev->primary, DRM_MINOR_LEGACY)))
-		goto err_g3;
-
-	if (dev->driver->load) {
-		ret = dev->driver->load(dev, ent->driver_data);
-		if (ret)
-			goto err_g4;
-	}
 
-	/* setup the grouping for the legacy output */
-	if (drm_core_check_feature(dev, DRIVER_MODESET)) {
-		ret = drm_mode_group_init_legacy_group(dev,
-						&dev->primary->mode_group);
-		if (ret)
-			goto err_g4;
-	}
-
-	list_add_tail(&dev->driver_item, &driver->device_list);
+	ret = drm_dev_register(dev, ent->driver_data);
+	if (ret)
+		goto err_pci;
 
 	DRM_INFO("Initialized %s %d.%d.%d %s for %s on minor %d\n",
 		 driver->name, driver->major, driver->minor, driver->patchlevel,
 		 driver->date, pci_name(pdev), dev->primary->index);
 
-	mutex_unlock(&drm_global_mutex);
 	return 0;
 
-err_g4:
-	drm_put_minor(&dev->primary);
-err_g3:
-	if (dev->render)
-		drm_put_minor(&dev->render);
-err_g21:
-	if (drm_core_check_feature(dev, DRIVER_MODESET))
-		drm_put_minor(&dev->control);
-err_g2:
+err_pci:
 	pci_disable_device(pdev);
-err_g1:
-	kfree(dev);
-	mutex_unlock(&drm_global_mutex);
+err_free:
+	drm_dev_free(dev);
 	return ret;
 }
 EXPORT_SYMBOL(drm_get_pci_dev);
diff --git a/drivers/gpu/drm/drm_platform.c b/drivers/gpu/drm/drm_platform.c
index f7a18c6ba4c4..fc24fee8ec83 100644
--- a/drivers/gpu/drm/drm_platform.c
+++ b/drivers/gpu/drm/drm_platform.c
@@ -47,55 +47,15 @@ static int drm_get_platform_dev(struct platform_device *platdev,
 
 	DRM_DEBUG("\n");
 
-	dev = kzalloc(sizeof(*dev), GFP_KERNEL);
+	dev = drm_dev_alloc(driver, &platdev->dev);
 	if (!dev)
 		return -ENOMEM;
 
 	dev->platformdev = platdev;
-	dev->dev = &platdev->dev;
 
-	mutex_lock(&drm_global_mutex);
-
-	ret = drm_fill_in_dev(dev, NULL, driver);
-
-	if (ret) {
-		printk(KERN_ERR "DRM: Fill_in_dev failed.\n");
-		goto err_g1;
-	}
-
-	if (drm_core_check_feature(dev, DRIVER_MODESET)) {
-		ret = drm_get_minor(dev, &dev->control, DRM_MINOR_CONTROL);
-		if (ret)
-			goto err_g1;
-	}
-
-	if (drm_core_check_feature(dev, DRIVER_RENDER) && drm_rnodes) {
-		ret = drm_get_minor(dev, &dev->render, DRM_MINOR_RENDER);
-		if (ret)
-			goto err_g11;
-	}
-
-	ret = drm_get_minor(dev, &dev->primary, DRM_MINOR_LEGACY);
+	ret = drm_dev_register(dev, 0);
 	if (ret)
-		goto err_g2;
-
-	if (dev->driver->load) {
-		ret = dev->driver->load(dev, 0);
-		if (ret)
-			goto err_g3;
-	}
-
-	/* setup the grouping for the legacy output */
-	if (drm_core_check_feature(dev, DRIVER_MODESET)) {
-		ret = drm_mode_group_init_legacy_group(dev,
-				&dev->primary->mode_group);
-		if (ret)
-			goto err_g3;
-	}
-
-	list_add_tail(&dev->driver_item, &driver->device_list);
-
-	mutex_unlock(&drm_global_mutex);
+		goto err_free;
 
 	DRM_INFO("Initialized %s %d.%d.%d %s on minor %d\n",
 		 driver->name, driver->major, driver->minor, driver->patchlevel,
@@ -103,17 +63,8 @@ static int drm_get_platform_dev(struct platform_device *platdev,
 
 	return 0;
 
-err_g3:
-	drm_put_minor(&dev->primary);
-err_g2:
-	if (dev->render)
-		drm_put_minor(&dev->render);
-err_g11:
-	if (drm_core_check_feature(dev, DRIVER_MODESET))
-		drm_put_minor(&dev->control);
-err_g1:
-	kfree(dev);
-	mutex_unlock(&drm_global_mutex);
+err_free:
+	drm_dev_free(dev);
 	return ret;
 }
 
diff --git a/drivers/gpu/drm/drm_prime.c b/drivers/gpu/drm/drm_prime.c
index 276d470f7b3e..56805c39c906 100644
--- a/drivers/gpu/drm/drm_prime.c
+++ b/drivers/gpu/drm/drm_prime.c
@@ -637,14 +637,13 @@ int drm_prime_sg_to_page_addr_arrays(struct sg_table *sgt, struct page **pages,
 	unsigned count;
 	struct scatterlist *sg;
 	struct page *page;
-	u32 len, offset;
+	u32 len;
 	int pg_index;
 	dma_addr_t addr;
 
 	pg_index = 0;
 	for_each_sg(sgt->sgl, sg, sgt->nents, count) {
 		len = sg->length;
-		offset = sg->offset;
 		page = sg_page(sg);
 		addr = sg_dma_address(sg);
 
diff --git a/drivers/gpu/drm/drm_stub.c b/drivers/gpu/drm/drm_stub.c
index 39d864576be4..f53d5246979c 100644
--- a/drivers/gpu/drm/drm_stub.c
+++ b/drivers/gpu/drm/drm_stub.c
@@ -254,81 +254,21 @@ int drm_dropmaster_ioctl(struct drm_device *dev, void *data,
 	return 0;
 }
 
-int drm_fill_in_dev(struct drm_device *dev,
-			   const struct pci_device_id *ent,
-			   struct drm_driver *driver)
-{
-	int retcode;
-
-	INIT_LIST_HEAD(&dev->filelist);
-	INIT_LIST_HEAD(&dev->ctxlist);
-	INIT_LIST_HEAD(&dev->vmalist);
-	INIT_LIST_HEAD(&dev->maplist);
-	INIT_LIST_HEAD(&dev->vblank_event_list);
-
-	spin_lock_init(&dev->count_lock);
-	spin_lock_init(&dev->event_lock);
-	mutex_init(&dev->struct_mutex);
-	mutex_init(&dev->ctxlist_mutex);
-
-	if (drm_ht_create(&dev->map_hash, 12)) {
-		return -ENOMEM;
-	}
-
-	/* the DRM has 6 basic counters */
-	dev->counters = 6;
-	dev->types[0] = _DRM_STAT_LOCK;
-	dev->types[1] = _DRM_STAT_OPENS;
-	dev->types[2] = _DRM_STAT_CLOSES;
-	dev->types[3] = _DRM_STAT_IOCTLS;
-	dev->types[4] = _DRM_STAT_LOCKS;
-	dev->types[5] = _DRM_STAT_UNLOCKS;
-
-	dev->driver = driver;
-
-	if (dev->driver->bus->agp_init) {
-		retcode = dev->driver->bus->agp_init(dev);
-		if (retcode)
-			goto error_out_unreg;
-	}
-
-
-
-	retcode = drm_ctxbitmap_init(dev);
-	if (retcode) {
-		DRM_ERROR("Cannot allocate memory for context bitmap.\n");
-		goto error_out_unreg;
-	}
-
-	if (driver->driver_features & DRIVER_GEM) {
-		retcode = drm_gem_init(dev);
-		if (retcode) {
-			DRM_ERROR("Cannot initialize graphics execution "
-				  "manager (GEM)\n");
-			goto error_out_unreg;
-		}
-	}
-
-	return 0;
-
-      error_out_unreg:
-	drm_lastclose(dev);
-	return retcode;
-}
-EXPORT_SYMBOL(drm_fill_in_dev);
-
-
 /**
- * Get a secondary minor number.
+ * drm_get_minor - Allocate and register new DRM minor
+ * @dev: DRM device
+ * @minor: Pointer to where new minor is stored
+ * @type: Type of minor
  *
- * \param dev device data structure
- * \param sec-minor structure to hold the assigned minor
- * \return negative number on failure.
+ * Allocate a new minor of the given type and register it. A pointer to the new
+ * minor is returned in @minor.
+ * Caller must hold the global DRM mutex.
  *
- * Search an empty entry and initialize it to the given parameters. This
- * routines assigns minor numbers to secondary heads of multi-headed cards
+ * RETURNS:
+ * 0 on success, negative error code on failure.
  */
-int drm_get_minor(struct drm_device *dev, struct drm_minor **minor, int type)
+static int drm_get_minor(struct drm_device *dev, struct drm_minor **minor,
+			 int type)
 {
 	struct drm_minor *new_minor;
 	int ret;
@@ -385,37 +325,48 @@ err_idr:
 	*minor = NULL;
 	return ret;
 }
-EXPORT_SYMBOL(drm_get_minor);
 
 /**
- * Put a secondary minor number.
+ * drm_unplug_minor - Unplug DRM minor
+ * @minor: Minor to unplug
  *
- * \param sec_minor - structure to be released
- * \return always zero
+ * Unplugs the given DRM minor but keeps the object. So after this returns,
+ * minor->dev is still valid so existing open-files can still access it to get
+ * device information from their drm_file ojects.
+ * If the minor is already unplugged or if @minor is NULL, nothing is done.
+ * The global DRM mutex must be held by the caller.
  */
-int drm_put_minor(struct drm_minor **minor_p)
+static void drm_unplug_minor(struct drm_minor *minor)
 {
-	struct drm_minor *minor = *minor_p;
-
-	DRM_DEBUG("release secondary minor %d\n", minor->index);
+	if (!minor || !minor->kdev)
+		return;
 
 #if defined(CONFIG_DEBUG_FS)
 	drm_debugfs_cleanup(minor);
 #endif
 
 	drm_sysfs_device_remove(minor);
-
 	idr_remove(&drm_minors_idr, minor->index);
-
-	kfree(minor);
-	*minor_p = NULL;
-	return 0;
 }
-EXPORT_SYMBOL(drm_put_minor);
 
-static void drm_unplug_minor(struct drm_minor *minor)
+/**
+ * drm_put_minor - Destroy DRM minor
+ * @minor: Minor to destroy
+ *
+ * This calls drm_unplug_minor() on the given minor and then frees it. Nothing
+ * is done if @minor is NULL. It is fine to call this on already unplugged
+ * minors.
+ * The global DRM mutex must be held by the caller.
+ */
+static void drm_put_minor(struct drm_minor *minor)
 {
-	drm_sysfs_device_remove(minor);
+	if (!minor)
+		return;
+
+	DRM_DEBUG("release secondary minor %d\n", minor->index);
+
+	drm_unplug_minor(minor);
+	kfree(minor);
 }
 
 /**
@@ -427,66 +378,237 @@ static void drm_unplug_minor(struct drm_minor *minor)
  */
 void drm_put_dev(struct drm_device *dev)
 {
-	struct drm_driver *driver;
-	struct drm_map_list *r_list, *list_temp;
-
 	DRM_DEBUG("\n");
 
 	if (!dev) {
 		DRM_ERROR("cleanup called no dev\n");
 		return;
 	}
-	driver = dev->driver;
 
-	drm_lastclose(dev);
+	drm_dev_unregister(dev);
+	drm_dev_free(dev);
+}
+EXPORT_SYMBOL(drm_put_dev);
 
-	if (dev->driver->unload)
-		dev->driver->unload(dev);
+void drm_unplug_dev(struct drm_device *dev)
+{
+	/* for a USB device */
+	if (drm_core_check_feature(dev, DRIVER_MODESET))
+		drm_unplug_minor(dev->control);
+	if (dev->render)
+		drm_unplug_minor(dev->render);
+	drm_unplug_minor(dev->primary);
 
-	if (dev->driver->bus->agp_destroy)
-		dev->driver->bus->agp_destroy(dev);
+	mutex_lock(&drm_global_mutex);
 
-	drm_vblank_cleanup(dev);
+	drm_device_set_unplugged(dev);
 
-	list_for_each_entry_safe(r_list, list_temp, &dev->maplist, head)
-		drm_rmmap(dev, r_list->map);
-	drm_ht_remove(&dev->map_hash);
+	if (dev->open_count == 0) {
+		drm_put_dev(dev);
+	}
+	mutex_unlock(&drm_global_mutex);
+}
+EXPORT_SYMBOL(drm_unplug_dev);
 
-	drm_ctxbitmap_cleanup(dev);
+/**
+ * drm_dev_alloc - Allocate new drm device
+ * @driver: DRM driver to allocate device for
+ * @parent: Parent device object
+ *
+ * Allocate and initialize a new DRM device. No device registration is done.
+ * Call drm_dev_register() to advertice the device to user space and register it
+ * with other core subsystems.
+ *
+ * RETURNS:
+ * Pointer to new DRM device, or NULL if out of memory.
+ */
+struct drm_device *drm_dev_alloc(struct drm_driver *driver,
+				 struct device *parent)
+{
+	struct drm_device *dev;
+	int ret;
 
-	if (drm_core_check_feature(dev, DRIVER_MODESET))
-		drm_put_minor(&dev->control);
+	dev = kzalloc(sizeof(*dev), GFP_KERNEL);
+	if (!dev)
+		return NULL;
 
-	if (dev->render)
-		drm_put_minor(&dev->render);
+	dev->dev = parent;
+	dev->driver = driver;
 
-	if (driver->driver_features & DRIVER_GEM)
+	INIT_LIST_HEAD(&dev->filelist);
+	INIT_LIST_HEAD(&dev->ctxlist);
+	INIT_LIST_HEAD(&dev->vmalist);
+	INIT_LIST_HEAD(&dev->maplist);
+	INIT_LIST_HEAD(&dev->vblank_event_list);
+
+	spin_lock_init(&dev->count_lock);
+	spin_lock_init(&dev->event_lock);
+	mutex_init(&dev->struct_mutex);
+	mutex_init(&dev->ctxlist_mutex);
+
+	if (drm_ht_create(&dev->map_hash, 12))
+		goto err_free;
+
+	ret = drm_ctxbitmap_init(dev);
+	if (ret) {
+		DRM_ERROR("Cannot allocate memory for context bitmap.\n");
+		goto err_ht;
+	}
+
+	if (driver->driver_features & DRIVER_GEM) {
+		ret = drm_gem_init(dev);
+		if (ret) {
+			DRM_ERROR("Cannot initialize graphics execution manager (GEM)\n");
+			goto err_ctxbitmap;
+		}
+	}
+
+	return dev;
+
+err_ctxbitmap:
+	drm_ctxbitmap_cleanup(dev);
+err_ht:
+	drm_ht_remove(&dev->map_hash);
+err_free:
+	kfree(dev);
+	return NULL;
+}
+EXPORT_SYMBOL(drm_dev_alloc);
+
+/**
+ * drm_dev_free - Free DRM device
+ * @dev: DRM device to free
+ *
+ * Free a DRM device that has previously been allocated via drm_dev_alloc().
+ * You must not use kfree() instead or you will leak memory.
+ *
+ * This must not be called once the device got registered. Use drm_put_dev()
+ * instead, which then calls drm_dev_free().
+ */
+void drm_dev_free(struct drm_device *dev)
+{
+	drm_put_minor(dev->control);
+	drm_put_minor(dev->render);
+	drm_put_minor(dev->primary);
+
+	if (dev->driver->driver_features & DRIVER_GEM)
 		drm_gem_destroy(dev);
 
-	drm_put_minor(&dev->primary);
+	drm_ctxbitmap_cleanup(dev);
+	drm_ht_remove(&dev->map_hash);
 
-	list_del(&dev->driver_item);
 	kfree(dev->devname);
 	kfree(dev);
 }
-EXPORT_SYMBOL(drm_put_dev);
+EXPORT_SYMBOL(drm_dev_free);
 
-void drm_unplug_dev(struct drm_device *dev)
+/**
+ * drm_dev_register - Register DRM device
+ * @dev: Device to register
+ *
+ * Register the DRM device @dev with the system, advertise device to user-space
+ * and start normal device operation. @dev must be allocated via drm_dev_alloc()
+ * previously.
+ *
+ * Never call this twice on any device!
+ *
+ * RETURNS:
+ * 0 on success, negative error code on failure.
+ */
+int drm_dev_register(struct drm_device *dev, unsigned long flags)
 {
-	/* for a USB device */
-	if (drm_core_check_feature(dev, DRIVER_MODESET))
-		drm_unplug_minor(dev->control);
-	if (dev->render)
-		drm_unplug_minor(dev->render);
-	drm_unplug_minor(dev->primary);
+	int ret;
 
 	mutex_lock(&drm_global_mutex);
 
-	drm_device_set_unplugged(dev);
+	if (dev->driver->bus->agp_init) {
+		ret = dev->driver->bus->agp_init(dev);
+		if (ret)
+			goto out_unlock;
+	}
 
-	if (dev->open_count == 0) {
-		drm_put_dev(dev);
+	if (drm_core_check_feature(dev, DRIVER_MODESET)) {
+		ret = drm_get_minor(dev, &dev->control, DRM_MINOR_CONTROL);
+		if (ret)
+			goto err_agp;
+	}
+
+	if (drm_core_check_feature(dev, DRIVER_RENDER) && drm_rnodes) {
+		ret = drm_get_minor(dev, &dev->render, DRM_MINOR_RENDER);
+		if (ret)
+			goto err_control_node;
+	}
+
+	ret = drm_get_minor(dev, &dev->primary, DRM_MINOR_LEGACY);
+	if (ret)
+		goto err_render_node;
+
+	if (dev->driver->load) {
+		ret = dev->driver->load(dev, flags);
+		if (ret)
+			goto err_primary_node;
 	}
+
+	/* setup grouping for legacy outputs */
+	if (drm_core_check_feature(dev, DRIVER_MODESET)) {
+		ret = drm_mode_group_init_legacy_group(dev,
+				&dev->primary->mode_group);
+		if (ret)
+			goto err_unload;
+	}
+
+	list_add_tail(&dev->driver_item, &dev->driver->device_list);
+
+	ret = 0;
+	goto out_unlock;
+
+err_unload:
+	if (dev->driver->unload)
+		dev->driver->unload(dev);
+err_primary_node:
+	drm_put_minor(dev->primary);
+err_render_node:
+	drm_put_minor(dev->render);
+err_control_node:
+	drm_put_minor(dev->control);
+err_agp:
+	if (dev->driver->bus->agp_destroy)
+		dev->driver->bus->agp_destroy(dev);
+out_unlock:
 	mutex_unlock(&drm_global_mutex);
+	return ret;
 }
-EXPORT_SYMBOL(drm_unplug_dev);
+EXPORT_SYMBOL(drm_dev_register);
+
+/**
+ * drm_dev_unregister - Unregister DRM device
+ * @dev: Device to unregister
+ *
+ * Unregister the DRM device from the system. This does the reverse of
+ * drm_dev_register() but does not deallocate the device. The caller must call
+ * drm_dev_free() to free all resources.
+ */
+void drm_dev_unregister(struct drm_device *dev)
+{
+	struct drm_map_list *r_list, *list_temp;
+
+	drm_lastclose(dev);
+
+	if (dev->driver->unload)
+		dev->driver->unload(dev);
+
+	if (dev->driver->bus->agp_destroy)
+		dev->driver->bus->agp_destroy(dev);
+
+	drm_vblank_cleanup(dev);
+
+	list_for_each_entry_safe(r_list, list_temp, &dev->maplist, head)
+		drm_rmmap(dev, r_list->map);
+
+	drm_unplug_minor(dev->control);
+	drm_unplug_minor(dev->render);
+	drm_unplug_minor(dev->primary);
+
+	list_del(&dev->driver_item);
+}
+EXPORT_SYMBOL(drm_dev_unregister);
diff --git a/drivers/gpu/drm/drm_sysfs.c b/drivers/gpu/drm/drm_sysfs.c
index 2290b3b73832..1a35ea53106b 100644
--- a/drivers/gpu/drm/drm_sysfs.c
+++ b/drivers/gpu/drm/drm_sysfs.c
@@ -22,8 +22,8 @@
 #include <drm/drm_core.h>
 #include <drm/drmP.h>
 
-#define to_drm_minor(d) container_of(d, struct drm_minor, kdev)
-#define to_drm_connector(d) container_of(d, struct drm_connector, kdev)
+#define to_drm_minor(d) dev_get_drvdata(d)
+#define to_drm_connector(d) dev_get_drvdata(d)
 
 static struct device_type drm_sysfs_device_minor = {
 	.name = "drm_minor"
@@ -162,20 +162,6 @@ void drm_sysfs_destroy(void)
 	drm_class = NULL;
 }
 
-/**
- * drm_sysfs_device_release - do nothing
- * @dev: Linux device
- *
- * Normally, this would free the DRM device associated with @dev, along
- * with cleaning up any other stuff.  But we do that in the DRM core, so
- * this function can just return and hope that the core does its job.
- */
-static void drm_sysfs_device_release(struct device *dev)
-{
-	memset(dev, 0, sizeof(struct device));
-	return;
-}
-
 /*
  * Connector properties
  */
@@ -380,11 +366,6 @@ static struct bin_attribute edid_attr = {
  * properties (so far, connection status, dpms, mode list & edid) and
  * generate a hotplug event so userspace knows there's a new connector
  * available.
- *
- * Note:
- * This routine should only be called *once* for each registered connector.
- * A second call for an already registered connector will trigger the BUG_ON
- * below.
  */
 int drm_sysfs_connector_add(struct drm_connector *connector)
 {
@@ -394,29 +375,25 @@ int drm_sysfs_connector_add(struct drm_connector *connector)
 	int i;
 	int ret;
 
-	/* We shouldn't get called more than once for the same connector */
-	BUG_ON(device_is_registered(&connector->kdev));
-
-	connector->kdev.parent = &dev->primary->kdev;
-	connector->kdev.class = drm_class;
-	connector->kdev.release = drm_sysfs_device_release;
+	if (connector->kdev)
+		return 0;
 
+	connector->kdev = device_create(drm_class, dev->primary->kdev,
+					0, connector, "card%d-%s",
+					dev->primary->index, drm_get_connector_name(connector));
 	DRM_DEBUG("adding \"%s\" to sysfs\n",
 		  drm_get_connector_name(connector));
 
-	dev_set_name(&connector->kdev, "card%d-%s",
-		     dev->primary->index, drm_get_connector_name(connector));
-	ret = device_register(&connector->kdev);
-
-	if (ret) {
-		DRM_ERROR("failed to register connector device: %d\n", ret);
+	if (IS_ERR(connector->kdev)) {
+		DRM_ERROR("failed to register connector device: %ld\n", PTR_ERR(connector->kdev));
+		ret = PTR_ERR(connector->kdev);
 		goto out;
 	}
 
 	/* Standard attributes */
 
 	for (attr_cnt = 0; attr_cnt < ARRAY_SIZE(connector_attrs); attr_cnt++) {
-		ret = device_create_file(&connector->kdev, &connector_attrs[attr_cnt]);
+		ret = device_create_file(connector->kdev, &connector_attrs[attr_cnt]);
 		if (ret)
 			goto err_out_files;
 	}
@@ -433,7 +410,7 @@ int drm_sysfs_connector_add(struct drm_connector *connector)
 		case DRM_MODE_CONNECTOR_Component:
 		case DRM_MODE_CONNECTOR_TV:
 			for (opt_cnt = 0; opt_cnt < ARRAY_SIZE(connector_attrs_opt1); opt_cnt++) {
-				ret = device_create_file(&connector->kdev, &connector_attrs_opt1[opt_cnt]);
+				ret = device_create_file(connector->kdev, &connector_attrs_opt1[opt_cnt]);
 				if (ret)
 					goto err_out_files;
 			}
@@ -442,7 +419,7 @@ int drm_sysfs_connector_add(struct drm_connector *connector)
 			break;
 	}
 
-	ret = sysfs_create_bin_file(&connector->kdev.kobj, &edid_attr);
+	ret = sysfs_create_bin_file(&connector->kdev->kobj, &edid_attr);
 	if (ret)
 		goto err_out_files;
 
@@ -453,10 +430,10 @@ int drm_sysfs_connector_add(struct drm_connector *connector)
 
 err_out_files:
 	for (i = 0; i < opt_cnt; i++)
-		device_remove_file(&connector->kdev, &connector_attrs_opt1[i]);
+		device_remove_file(connector->kdev, &connector_attrs_opt1[i]);
 	for (i = 0; i < attr_cnt; i++)
-		device_remove_file(&connector->kdev, &connector_attrs[i]);
-	device_unregister(&connector->kdev);
+		device_remove_file(connector->kdev, &connector_attrs[i]);
+	device_unregister(connector->kdev);
 
 out:
 	return ret;
@@ -480,16 +457,16 @@ void drm_sysfs_connector_remove(struct drm_connector *connector)
 {
 	int i;
 
-	if (!connector->kdev.parent)
+	if (!connector->kdev)
 		return;
 	DRM_DEBUG("removing \"%s\" from sysfs\n",
 		  drm_get_connector_name(connector));
 
 	for (i = 0; i < ARRAY_SIZE(connector_attrs); i++)
-		device_remove_file(&connector->kdev, &connector_attrs[i]);
-	sysfs_remove_bin_file(&connector->kdev.kobj, &edid_attr);
-	device_unregister(&connector->kdev);
-	connector->kdev.parent = NULL;
+		device_remove_file(connector->kdev, &connector_attrs[i]);
+	sysfs_remove_bin_file(&connector->kdev->kobj, &edid_attr);
+	device_unregister(connector->kdev);
+	connector->kdev = NULL;
 }
 EXPORT_SYMBOL(drm_sysfs_connector_remove);
 
@@ -508,7 +485,7 @@ void drm_sysfs_hotplug_event(struct drm_device *dev)
 
 	DRM_DEBUG("generating hotplug event\n");
 
-	kobject_uevent_env(&dev->primary->kdev.kobj, KOBJ_CHANGE, envp);
+	kobject_uevent_env(&dev->primary->kdev->kobj, KOBJ_CHANGE, envp);
 }
 EXPORT_SYMBOL(drm_sysfs_hotplug_event);
 
@@ -523,15 +500,8 @@ EXPORT_SYMBOL(drm_sysfs_hotplug_event);
  */
 int drm_sysfs_device_add(struct drm_minor *minor)
 {
-	int err;
 	char *minor_str;
 
-	minor->kdev.parent = minor->dev->dev;
-
-	minor->kdev.class = drm_class;
-	minor->kdev.release = drm_sysfs_device_release;
-	minor->kdev.devt = minor->device;
-	minor->kdev.type = &drm_sysfs_device_minor;
 	if (minor->type == DRM_MINOR_CONTROL)
 		minor_str = "controlD%d";
         else if (minor->type == DRM_MINOR_RENDER)
@@ -539,18 +509,14 @@ int drm_sysfs_device_add(struct drm_minor *minor)
         else
                 minor_str = "card%d";
 
-	dev_set_name(&minor->kdev, minor_str, minor->index);
-
-	err = device_register(&minor->kdev);
-	if (err) {
-		DRM_ERROR("device add failed: %d\n", err);
-		goto err_out;
+	minor->kdev = device_create(drm_class, minor->dev->dev,
+				    MKDEV(DRM_MAJOR, minor->index),
+				    minor, minor_str, minor->index);
+	if (IS_ERR(minor->kdev)) {
+		DRM_ERROR("device create failed %ld\n", PTR_ERR(minor->kdev));
+		return PTR_ERR(minor->kdev);
 	}
-
 	return 0;
-
-err_out:
-	return err;
 }
 
 /**
@@ -562,9 +528,9 @@ err_out:
  */
 void drm_sysfs_device_remove(struct drm_minor *minor)
 {
-	if (minor->kdev.parent)
-		device_unregister(&minor->kdev);
-	minor->kdev.parent = NULL;
+	if (minor->kdev)
+		device_destroy(drm_class, MKDEV(DRM_MAJOR, minor->index));
+	minor->kdev = NULL;
 }
 
 
diff --git a/drivers/gpu/drm/drm_usb.c b/drivers/gpu/drm/drm_usb.c
index 87664723b9ce..b179b70e7853 100644
--- a/drivers/gpu/drm/drm_usb.c
+++ b/drivers/gpu/drm/drm_usb.c
@@ -7,57 +7,20 @@ int drm_get_usb_dev(struct usb_interface *interface,
 		    struct drm_driver *driver)
 {
 	struct drm_device *dev;
-	struct usb_device *usbdev;
 	int ret;
 
 	DRM_DEBUG("\n");
 
-	dev = kzalloc(sizeof(*dev), GFP_KERNEL);
+	dev = drm_dev_alloc(driver, &interface->dev);
 	if (!dev)
 		return -ENOMEM;
 
-	usbdev = interface_to_usbdev(interface);
-	dev->usbdev = usbdev;
-	dev->dev = &interface->dev;
-
-	mutex_lock(&drm_global_mutex);
-
-	ret = drm_fill_in_dev(dev, NULL, driver);
-	if (ret) {
-		printk(KERN_ERR "DRM: Fill_in_dev failed.\n");
-		goto err_g1;
-	}
-
+	dev->usbdev = interface_to_usbdev(interface);
 	usb_set_intfdata(interface, dev);
-	ret = drm_get_minor(dev, &dev->control, DRM_MINOR_CONTROL);
-	if (ret)
-		goto err_g1;
-
-	if (drm_core_check_feature(dev, DRIVER_RENDER) && drm_rnodes) {
-		ret = drm_get_minor(dev, &dev->render, DRM_MINOR_RENDER);
-		if (ret)
-			goto err_g11;
-	}
 
-	ret = drm_get_minor(dev, &dev->primary, DRM_MINOR_LEGACY);
+	ret = drm_dev_register(dev, 0);
 	if (ret)
-		goto err_g2;
-
-	if (dev->driver->load) {
-		ret = dev->driver->load(dev, 0);
-		if (ret)
-			goto err_g3;
-	}
-
-	/* setup the grouping for the legacy output */
-	ret = drm_mode_group_init_legacy_group(dev,
-					       &dev->primary->mode_group);
-	if (ret)
-		goto err_g3;
-
-	list_add_tail(&dev->driver_item, &driver->device_list);
-
-	mutex_unlock(&drm_global_mutex);
+		goto err_free;
 
 	DRM_INFO("Initialized %s %d.%d.%d %s on minor %d\n",
 		 driver->name, driver->major, driver->minor, driver->patchlevel,
@@ -65,16 +28,8 @@ int drm_get_usb_dev(struct usb_interface *interface,
 
 	return 0;
 
-err_g3:
-	drm_put_minor(&dev->primary);
-err_g2:
-	if (dev->render)
-		drm_put_minor(&dev->render);
-err_g11:
-	drm_put_minor(&dev->control);
-err_g1:
-	kfree(dev);
-	mutex_unlock(&drm_global_mutex);
+err_free:
+	drm_dev_free(dev);
 	return ret;
 
 }
diff --git a/drivers/gpu/drm/drm_vm.c b/drivers/gpu/drm/drm_vm.c
index b5c5af7328df..93e95d7efd57 100644
--- a/drivers/gpu/drm/drm_vm.c
+++ b/drivers/gpu/drm/drm_vm.c
@@ -301,7 +301,7 @@ static int drm_do_vm_dma_fault(struct vm_area_struct *vma, struct vm_fault *vmf)
 
 	offset = (unsigned long)vmf->virtual_address - vma->vm_start;	/* vm_[pg]off[set] should be 0 */
 	page_nr = offset >> PAGE_SHIFT; /* page_nr could just be vmf->pgoff */
-	page = virt_to_page((dma->pagelist[page_nr] + (offset & (~PAGE_MASK))));
+	page = virt_to_page((void *)dma->pagelist[page_nr]);
 
 	get_page(page);
 	vmf->page = page;
diff --git a/drivers/gpu/drm/exynos/Kconfig b/drivers/gpu/drm/exynos/Kconfig
index 45b6ef595965..f227f544aa36 100644
--- a/drivers/gpu/drm/exynos/Kconfig
+++ b/drivers/gpu/drm/exynos/Kconfig
@@ -2,6 +2,7 @@ config DRM_EXYNOS
 	tristate "DRM Support for Samsung SoC EXYNOS Series"
 	depends on OF && DRM && (PLAT_SAMSUNG || ARCH_MULTIPLATFORM)
 	select DRM_KMS_HELPER
+	select DRM_KMS_FB_HELPER
 	select FB_CFB_FILLRECT
 	select FB_CFB_COPYAREA
 	select FB_CFB_IMAGEBLIT
diff --git a/drivers/gpu/drm/exynos/exynos_drm_drv.c b/drivers/gpu/drm/exynos/exynos_drm_drv.c
index 81192d00b39e..b676006a95a0 100644
--- a/drivers/gpu/drm/exynos/exynos_drm_drv.c
+++ b/drivers/gpu/drm/exynos/exynos_drm_drv.c
@@ -264,7 +264,6 @@ static struct drm_driver exynos_drm_driver = {
 	.get_vblank_counter	= drm_vblank_count,
 	.enable_vblank		= exynos_drm_crtc_enable_vblank,
 	.disable_vblank		= exynos_drm_crtc_disable_vblank,
-	.gem_init_object	= exynos_drm_gem_init_object,
 	.gem_free_object	= exynos_drm_gem_free_object,
 	.gem_vm_ops		= &exynos_drm_gem_vm_ops,
 	.dumb_create		= exynos_drm_gem_dumb_create,
diff --git a/drivers/gpu/drm/exynos/exynos_drm_fimd.c b/drivers/gpu/drm/exynos/exynos_drm_fimd.c
index 868a14d52995..23da72b5eae9 100644
--- a/drivers/gpu/drm/exynos/exynos_drm_fimd.c
+++ b/drivers/gpu/drm/exynos/exynos_drm_fimd.c
@@ -716,20 +716,20 @@ static int fimd_subdrv_probe(struct drm_device *drm_dev, struct device *dev)
 {
 	/*
 	 * enable drm irq mode.
-	 * - with irq_enabled = 1, we can use the vblank feature.
+	 * - with irq_enabled = true, we can use the vblank feature.
 	 *
 	 * P.S. note that we wouldn't use drm irq handler but
 	 *	just specific driver own one instead because
 	 *	drm framework supports only one irq handler.
 	 */
-	drm_dev->irq_enabled = 1;
+	drm_dev->irq_enabled = true;
 
 	/*
-	 * with vblank_disable_allowed = 1, vblank interrupt will be disabled
+	 * with vblank_disable_allowed = true, vblank interrupt will be disabled
 	 * by drm timer once a current process gives up ownership of
 	 * vblank event.(after drm_vblank_put function is called)
 	 */
-	drm_dev->vblank_disable_allowed = 1;
+	drm_dev->vblank_disable_allowed = true;
 
 	/* attach this sub driver to iommu mapping if supported. */
 	if (is_drm_iommu_supported(drm_dev))
diff --git a/drivers/gpu/drm/exynos/exynos_drm_gem.c b/drivers/gpu/drm/exynos/exynos_drm_gem.c
index 49f9cd232757..1ade191d84f4 100644
--- a/drivers/gpu/drm/exynos/exynos_drm_gem.c
+++ b/drivers/gpu/drm/exynos/exynos_drm_gem.c
@@ -630,11 +630,6 @@ void exynos_gem_unmap_sgt_from_dma(struct drm_device *drm_dev,
 	dma_unmap_sg(drm_dev->dev, sgt->sgl, sgt->nents, dir);
 }
 
-int exynos_drm_gem_init_object(struct drm_gem_object *obj)
-{
-	return 0;
-}
-
 void exynos_drm_gem_free_object(struct drm_gem_object *obj)
 {
 	struct exynos_drm_gem_obj *exynos_gem_obj;
diff --git a/drivers/gpu/drm/exynos/exynos_drm_gem.h b/drivers/gpu/drm/exynos/exynos_drm_gem.h
index 09555afdfe9c..702ec3abe85c 100644
--- a/drivers/gpu/drm/exynos/exynos_drm_gem.h
+++ b/drivers/gpu/drm/exynos/exynos_drm_gem.h
@@ -135,9 +135,6 @@ unsigned long exynos_drm_gem_get_size(struct drm_device *dev,
 						unsigned int gem_handle,
 						struct drm_file *file_priv);
 
-/* initialize gem object. */
-int exynos_drm_gem_init_object(struct drm_gem_object *obj);
-
 /* free gem object. */
 void exynos_drm_gem_free_object(struct drm_gem_object *gem_obj);
 
diff --git a/drivers/gpu/drm/exynos/exynos_drm_vidi.c b/drivers/gpu/drm/exynos/exynos_drm_vidi.c
index 4400330e4449..ddaaedde173d 100644
--- a/drivers/gpu/drm/exynos/exynos_drm_vidi.c
+++ b/drivers/gpu/drm/exynos/exynos_drm_vidi.c
@@ -101,7 +101,6 @@ static struct edid *vidi_get_edid(struct device *dev,
 {
 	struct vidi_context *ctx = get_vidi_context(dev);
 	struct edid *edid;
-	int edid_len;
 
 	/*
 	 * the edid data comes from user side and it would be set
@@ -112,8 +111,7 @@ static struct edid *vidi_get_edid(struct device *dev,
 		return ERR_PTR(-EFAULT);
 	}
 
-	edid_len = (1 + ctx->raw_edid->extensions) * EDID_LENGTH;
-	edid = kmemdup(ctx->raw_edid, edid_len, GFP_KERNEL);
+	edid = drm_edid_duplicate(ctx->raw_edid);
 	if (!edid) {
 		DRM_DEBUG_KMS("failed to allocate edid\n");
 		return ERR_PTR(-ENOMEM);
@@ -385,20 +383,20 @@ static int vidi_subdrv_probe(struct drm_device *drm_dev, struct device *dev)
 {
 	/*
 	 * enable drm irq mode.
-	 * - with irq_enabled = 1, we can use the vblank feature.
+	 * - with irq_enabled = true, we can use the vblank feature.
 	 *
 	 * P.S. note that we wouldn't use drm irq handler but
 	 *	just specific driver own one instead because
 	 *	drm framework supports only one irq handler.
 	 */
-	drm_dev->irq_enabled = 1;
+	drm_dev->irq_enabled = true;
 
 	/*
-	 * with vblank_disable_allowed = 1, vblank interrupt will be disabled
+	 * with vblank_disable_allowed = true, vblank interrupt will be disabled
 	 * by drm timer once a current process gives up ownership of
 	 * vblank event.(after drm_vblank_put function is called)
 	 */
-	drm_dev->vblank_disable_allowed = 1;
+	drm_dev->vblank_disable_allowed = true;
 
 	return 0;
 }
@@ -485,7 +483,6 @@ int vidi_connection_ioctl(struct drm_device *drm_dev, void *data,
 	struct exynos_drm_manager *manager;
 	struct exynos_drm_display_ops *display_ops;
 	struct drm_exynos_vidi_connection *vidi = data;
-	int edid_len;
 
 	if (!vidi) {
 		DRM_DEBUG_KMS("user data for vidi is null.\n");
@@ -524,8 +521,7 @@ int vidi_connection_ioctl(struct drm_device *drm_dev, void *data,
 			DRM_DEBUG_KMS("edid data is invalid.\n");
 			return -EINVAL;
 		}
-		edid_len = (1 + raw_edid->extensions) * EDID_LENGTH;
-		ctx->raw_edid = kmemdup(raw_edid, edid_len, GFP_KERNEL);
+		ctx->raw_edid = drm_edid_duplicate(raw_edid);
 		if (!ctx->raw_edid) {
 			DRM_DEBUG_KMS("failed to allocate raw_edid.\n");
 			return -ENOMEM;
diff --git a/drivers/gpu/drm/gma500/Kconfig b/drivers/gpu/drm/gma500/Kconfig
index 1f6e2dfaaeae..508cf99a292d 100644
--- a/drivers/gpu/drm/gma500/Kconfig
+++ b/drivers/gpu/drm/gma500/Kconfig
@@ -5,6 +5,7 @@ config DRM_GMA500
 	select FB_CFB_FILLRECT
 	select FB_CFB_IMAGEBLIT
 	select DRM_KMS_HELPER
+	select DRM_KMS_FB_HELPER
 	select DRM_TTM
 	# GMA500 depends on ACPI_VIDEO when ACPI is enabled, just like i915
 	select ACPI_VIDEO if ACPI
diff --git a/drivers/gpu/drm/gma500/cdv_device.c b/drivers/gpu/drm/gma500/cdv_device.c
index 162f686c532d..5a9a6a3063a8 100644
--- a/drivers/gpu/drm/gma500/cdv_device.c
+++ b/drivers/gpu/drm/gma500/cdv_device.c
@@ -634,6 +634,7 @@ const struct psb_ops cdv_chip_ops = {
 	.crtcs = 2,
 	.hdmi_mask = (1 << 0) | (1 << 1),
 	.lvds_mask = (1 << 1),
+	.sdvo_mask = (1 << 0),
 	.cursor_needs_phys = 0,
 	.sgx_offset = MRST_SGX_OFFSET,
 	.chip_setup = cdv_chip_setup,
diff --git a/drivers/gpu/drm/gma500/cdv_intel_dp.c b/drivers/gpu/drm/gma500/cdv_intel_dp.c
index f4eb43573cad..f88a1815d87c 100644
--- a/drivers/gpu/drm/gma500/cdv_intel_dp.c
+++ b/drivers/gpu/drm/gma500/cdv_intel_dp.c
@@ -666,7 +666,7 @@ cdv_intel_dp_i2c_init(struct gma_connector *connector,
 	strncpy (intel_dp->adapter.name, name, sizeof(intel_dp->adapter.name) - 1);
 	intel_dp->adapter.name[sizeof(intel_dp->adapter.name) - 1] = '\0';
 	intel_dp->adapter.algo_data = &intel_dp->algo;
-	intel_dp->adapter.dev.parent = &connector->base.kdev;
+	intel_dp->adapter.dev.parent = connector->base.kdev;
 
 	if (is_edp(encoder))
 		cdv_intel_edp_panel_vdd_on(encoder);
diff --git a/drivers/gpu/drm/gma500/framebuffer.c b/drivers/gpu/drm/gma500/framebuffer.c
index 01dd7d225762..94b3fec22c28 100644
--- a/drivers/gpu/drm/gma500/framebuffer.c
+++ b/drivers/gpu/drm/gma500/framebuffer.c
@@ -714,7 +714,7 @@ static void psb_setup_outputs(struct drm_device *dev)
 			clone_mask = (1 << INTEL_OUTPUT_ANALOG);
 			break;
 		case INTEL_OUTPUT_SDVO:
-			crtc_mask = ((1 << 0) | (1 << 1));
+			crtc_mask = dev_priv->ops->sdvo_mask;
 			clone_mask = (1 << INTEL_OUTPUT_SDVO);
 			break;
 		case INTEL_OUTPUT_LVDS:
diff --git a/drivers/gpu/drm/gma500/gem.c b/drivers/gpu/drm/gma500/gem.c
index 10ae8c52d06f..e2db48a81ed0 100644
--- a/drivers/gpu/drm/gma500/gem.c
+++ b/drivers/gpu/drm/gma500/gem.c
@@ -29,11 +29,6 @@
 #include <drm/drm_vma_manager.h>
 #include "psb_drv.h"
 
-int psb_gem_init_object(struct drm_gem_object *obj)
-{
-	return -EINVAL;
-}
-
 void psb_gem_free_object(struct drm_gem_object *obj)
 {
 	struct gtt_range *gtt = container_of(obj, struct gtt_range, gem);
diff --git a/drivers/gpu/drm/gma500/intel_gmbus.c b/drivers/gpu/drm/gma500/intel_gmbus.c
index 62cd42e88f28..566d330aaeea 100644
--- a/drivers/gpu/drm/gma500/intel_gmbus.c
+++ b/drivers/gpu/drm/gma500/intel_gmbus.c
@@ -51,6 +51,9 @@
 #define wait_for(COND, MS) _wait_for(COND, MS, 1)
 #define wait_for_atomic(COND, MS) _wait_for(COND, MS, 0)
 
+#define GMBUS_REG_READ(reg) ioread32(dev_priv->gmbus_reg + (reg))
+#define GMBUS_REG_WRITE(reg, val) iowrite32((val), dev_priv->gmbus_reg + (reg))
+
 /* Intel GPIO access functions */
 
 #define I2C_RISEFALL_TIME 20
@@ -71,7 +74,8 @@ struct intel_gpio {
 void
 gma_intel_i2c_reset(struct drm_device *dev)
 {
-	REG_WRITE(GMBUS0, 0);
+	struct drm_psb_private *dev_priv = dev->dev_private;
+	GMBUS_REG_WRITE(GMBUS0, 0);
 }
 
 static void intel_i2c_quirk_set(struct drm_psb_private *dev_priv, bool enable)
@@ -98,11 +102,10 @@ static void intel_i2c_quirk_set(struct drm_psb_private *dev_priv, bool enable)
 static u32 get_reserved(struct intel_gpio *gpio)
 {
 	struct drm_psb_private *dev_priv = gpio->dev_priv;
-	struct drm_device *dev = dev_priv->dev;
 	u32 reserved = 0;
 
 	/* On most chips, these bits must be preserved in software. */
-	reserved = REG_READ(gpio->reg) &
+	reserved = GMBUS_REG_READ(gpio->reg) &
 				     (GPIO_DATA_PULLUP_DISABLE |
 				      GPIO_CLOCK_PULLUP_DISABLE);
 
@@ -113,29 +116,26 @@ static int get_clock(void *data)
 {
 	struct intel_gpio *gpio = data;
 	struct drm_psb_private *dev_priv = gpio->dev_priv;
-	struct drm_device *dev = dev_priv->dev;
 	u32 reserved = get_reserved(gpio);
-	REG_WRITE(gpio->reg, reserved | GPIO_CLOCK_DIR_MASK);
-	REG_WRITE(gpio->reg, reserved);
-	return (REG_READ(gpio->reg) & GPIO_CLOCK_VAL_IN) != 0;
+	GMBUS_REG_WRITE(gpio->reg, reserved | GPIO_CLOCK_DIR_MASK);
+	GMBUS_REG_WRITE(gpio->reg, reserved);
+	return (GMBUS_REG_READ(gpio->reg) & GPIO_CLOCK_VAL_IN) != 0;
 }
 
 static int get_data(void *data)
 {
 	struct intel_gpio *gpio = data;
 	struct drm_psb_private *dev_priv = gpio->dev_priv;
-	struct drm_device *dev = dev_priv->dev;
 	u32 reserved = get_reserved(gpio);
-	REG_WRITE(gpio->reg, reserved | GPIO_DATA_DIR_MASK);
-	REG_WRITE(gpio->reg, reserved);
-	return (REG_READ(gpio->reg) & GPIO_DATA_VAL_IN) != 0;
+	GMBUS_REG_WRITE(gpio->reg, reserved | GPIO_DATA_DIR_MASK);
+	GMBUS_REG_WRITE(gpio->reg, reserved);
+	return (GMBUS_REG_READ(gpio->reg) & GPIO_DATA_VAL_IN) != 0;
 }
 
 static void set_clock(void *data, int state_high)
 {
 	struct intel_gpio *gpio = data;
 	struct drm_psb_private *dev_priv = gpio->dev_priv;
-	struct drm_device *dev = dev_priv->dev;
 	u32 reserved = get_reserved(gpio);
 	u32 clock_bits;
 
@@ -145,15 +145,14 @@ static void set_clock(void *data, int state_high)
 		clock_bits = GPIO_CLOCK_DIR_OUT | GPIO_CLOCK_DIR_MASK |
 			GPIO_CLOCK_VAL_MASK;
 
-	REG_WRITE(gpio->reg, reserved | clock_bits);
-	REG_READ(gpio->reg); /* Posting */
+	GMBUS_REG_WRITE(gpio->reg, reserved | clock_bits);
+	GMBUS_REG_READ(gpio->reg); /* Posting */
 }
 
 static void set_data(void *data, int state_high)
 {
 	struct intel_gpio *gpio = data;
 	struct drm_psb_private *dev_priv = gpio->dev_priv;
-	struct drm_device *dev = dev_priv->dev;
 	u32 reserved = get_reserved(gpio);
 	u32 data_bits;
 
@@ -163,8 +162,8 @@ static void set_data(void *data, int state_high)
 		data_bits = GPIO_DATA_DIR_OUT | GPIO_DATA_DIR_MASK |
 			GPIO_DATA_VAL_MASK;
 
-	REG_WRITE(gpio->reg, reserved | data_bits);
-	REG_READ(gpio->reg);
+	GMBUS_REG_WRITE(gpio->reg, reserved | data_bits);
+	GMBUS_REG_READ(gpio->reg);
 }
 
 static struct i2c_adapter *
@@ -251,7 +250,6 @@ gmbus_xfer(struct i2c_adapter *adapter,
 					       struct intel_gmbus,
 					       adapter);
 	struct drm_psb_private *dev_priv = adapter->algo_data;
-	struct drm_device *dev = dev_priv->dev;
 	int i, reg_offset;
 
 	if (bus->force_bit)
@@ -260,28 +258,30 @@ gmbus_xfer(struct i2c_adapter *adapter,
 
 	reg_offset = 0;
 
-	REG_WRITE(GMBUS0 + reg_offset, bus->reg0);
+	GMBUS_REG_WRITE(GMBUS0 + reg_offset, bus->reg0);
 
 	for (i = 0; i < num; i++) {
 		u16 len = msgs[i].len;
 		u8 *buf = msgs[i].buf;
 
 		if (msgs[i].flags & I2C_M_RD) {
-			REG_WRITE(GMBUS1 + reg_offset,
-				   GMBUS_CYCLE_WAIT | (i + 1 == num ? GMBUS_CYCLE_STOP : 0) |
-				   (len << GMBUS_BYTE_COUNT_SHIFT) |
-				   (msgs[i].addr << GMBUS_SLAVE_ADDR_SHIFT) |
-				   GMBUS_SLAVE_READ | GMBUS_SW_RDY);
-			REG_READ(GMBUS2+reg_offset);
+			GMBUS_REG_WRITE(GMBUS1 + reg_offset,
+					GMBUS_CYCLE_WAIT |
+					(i + 1 == num ? GMBUS_CYCLE_STOP : 0) |
+					(len << GMBUS_BYTE_COUNT_SHIFT) |
+					(msgs[i].addr << GMBUS_SLAVE_ADDR_SHIFT) |
+					GMBUS_SLAVE_READ | GMBUS_SW_RDY);
+			GMBUS_REG_READ(GMBUS2+reg_offset);
 			do {
 				u32 val, loop = 0;
 
-				if (wait_for(REG_READ(GMBUS2 + reg_offset) & (GMBUS_SATOER | GMBUS_HW_RDY), 50))
+				if (wait_for(GMBUS_REG_READ(GMBUS2 + reg_offset) &
+					     (GMBUS_SATOER | GMBUS_HW_RDY), 50))
 					goto timeout;
-				if (REG_READ(GMBUS2 + reg_offset) & GMBUS_SATOER)
+				if (GMBUS_REG_READ(GMBUS2 + reg_offset) & GMBUS_SATOER)
 					goto clear_err;
 
-				val = REG_READ(GMBUS3 + reg_offset);
+				val = GMBUS_REG_READ(GMBUS3 + reg_offset);
 				do {
 					*buf++ = val & 0xff;
 					val >>= 8;
@@ -295,18 +295,20 @@ gmbus_xfer(struct i2c_adapter *adapter,
 				val |= *buf++ << (8 * loop);
 			} while (--len && ++loop < 4);
 
-			REG_WRITE(GMBUS3 + reg_offset, val);
-			REG_WRITE(GMBUS1 + reg_offset,
+			GMBUS_REG_WRITE(GMBUS3 + reg_offset, val);
+			GMBUS_REG_WRITE(GMBUS1 + reg_offset,
 				   (i + 1 == num ? GMBUS_CYCLE_STOP : GMBUS_CYCLE_WAIT) |
 				   (msgs[i].len << GMBUS_BYTE_COUNT_SHIFT) |
 				   (msgs[i].addr << GMBUS_SLAVE_ADDR_SHIFT) |
 				   GMBUS_SLAVE_WRITE | GMBUS_SW_RDY);
-			REG_READ(GMBUS2+reg_offset);
+			GMBUS_REG_READ(GMBUS2+reg_offset);
 
 			while (len) {
-				if (wait_for(REG_READ(GMBUS2 + reg_offset) & (GMBUS_SATOER | GMBUS_HW_RDY), 50))
+				if (wait_for(GMBUS_REG_READ(GMBUS2 + reg_offset) &
+					     (GMBUS_SATOER | GMBUS_HW_RDY), 50))
 					goto timeout;
-				if (REG_READ(GMBUS2 + reg_offset) & GMBUS_SATOER)
+				if (GMBUS_REG_READ(GMBUS2 + reg_offset) &
+				    GMBUS_SATOER)
 					goto clear_err;
 
 				val = loop = 0;
@@ -314,14 +316,14 @@ gmbus_xfer(struct i2c_adapter *adapter,
 					val |= *buf++ << (8 * loop);
 				} while (--len && ++loop < 4);
 
-				REG_WRITE(GMBUS3 + reg_offset, val);
-				REG_READ(GMBUS2+reg_offset);
+				GMBUS_REG_WRITE(GMBUS3 + reg_offset, val);
+				GMBUS_REG_READ(GMBUS2+reg_offset);
 			}
 		}
 
-		if (i + 1 < num && wait_for(REG_READ(GMBUS2 + reg_offset) & (GMBUS_SATOER | GMBUS_HW_WAIT_PHASE), 50))
+		if (i + 1 < num && wait_for(GMBUS_REG_READ(GMBUS2 + reg_offset) & (GMBUS_SATOER | GMBUS_HW_WAIT_PHASE), 50))
 			goto timeout;
-		if (REG_READ(GMBUS2 + reg_offset) & GMBUS_SATOER)
+		if (GMBUS_REG_READ(GMBUS2 + reg_offset) & GMBUS_SATOER)
 			goto clear_err;
 	}
 
@@ -332,20 +334,20 @@ clear_err:
 	 * of resetting the GMBUS controller and so clearing the
 	 * BUS_ERROR raised by the slave's NAK.
 	 */
-	REG_WRITE(GMBUS1 + reg_offset, GMBUS_SW_CLR_INT);
-	REG_WRITE(GMBUS1 + reg_offset, 0);
+	GMBUS_REG_WRITE(GMBUS1 + reg_offset, GMBUS_SW_CLR_INT);
+	GMBUS_REG_WRITE(GMBUS1 + reg_offset, 0);
 
 done:
 	/* Mark the GMBUS interface as disabled. We will re-enable it at the
 	 * start of the next xfer, till then let it sleep.
 	 */
-	REG_WRITE(GMBUS0 + reg_offset, 0);
+	GMBUS_REG_WRITE(GMBUS0 + reg_offset, 0);
 	return i;
 
 timeout:
 	DRM_INFO("GMBUS timed out, falling back to bit banging on pin %d [%s]\n",
 		 bus->reg0 & 0xff, bus->adapter.name);
-	REG_WRITE(GMBUS0 + reg_offset, 0);
+	GMBUS_REG_WRITE(GMBUS0 + reg_offset, 0);
 
 	/* Hardware may not support GMBUS over these pins? Try GPIO bitbanging instead. */
 	bus->force_bit = intel_gpio_create(dev_priv, bus->reg0 & 0xff);
@@ -399,6 +401,11 @@ int gma_intel_setup_gmbus(struct drm_device *dev)
 	if (dev_priv->gmbus == NULL)
 		return -ENOMEM;
 
+	if (IS_MRST(dev))
+		dev_priv->gmbus_reg = dev_priv->aux_reg;
+	else
+		dev_priv->gmbus_reg = dev_priv->vdc_reg;
+
 	for (i = 0; i < GMBUS_NUM_PORTS; i++) {
 		struct intel_gmbus *bus = &dev_priv->gmbus[i];
 
@@ -487,6 +494,7 @@ void gma_intel_teardown_gmbus(struct drm_device *dev)
 		i2c_del_adapter(&bus->adapter);
 	}
 
+	dev_priv->gmbus_reg = NULL; /* iounmap is done in driver_unload */
 	kfree(dev_priv->gmbus);
 	dev_priv->gmbus = NULL;
 }
diff --git a/drivers/gpu/drm/gma500/oaktrail_crtc.c b/drivers/gpu/drm/gma500/oaktrail_crtc.c
index 54c98962b73e..8195e8592107 100644
--- a/drivers/gpu/drm/gma500/oaktrail_crtc.c
+++ b/drivers/gpu/drm/gma500/oaktrail_crtc.c
@@ -26,24 +26,10 @@
 #include "gma_display.h"
 #include "power.h"
 
-struct psb_intel_range_t {
-	int min, max;
-};
-
-struct oaktrail_limit_t {
-	struct psb_intel_range_t dot, m, p1;
-};
-
-struct oaktrail_clock_t {
-	/* derived values */
-	int dot;
-	int m;
-	int p1;
-};
-
-#define MRST_LIMIT_LVDS_100L	    0
-#define MRST_LIMIT_LVDS_83	    1
-#define MRST_LIMIT_LVDS_100	    2
+#define MRST_LIMIT_LVDS_100L	0
+#define MRST_LIMIT_LVDS_83	1
+#define MRST_LIMIT_LVDS_100	2
+#define MRST_LIMIT_SDVO		3
 
 #define MRST_DOT_MIN		  19750
 #define MRST_DOT_MAX		  120000
@@ -57,21 +43,40 @@ struct oaktrail_clock_t {
 #define MRST_P1_MAX_0		    7
 #define MRST_P1_MAX_1		    8
 
-static const struct oaktrail_limit_t oaktrail_limits[] = {
+static bool mrst_lvds_find_best_pll(const struct gma_limit_t *limit,
+				    struct drm_crtc *crtc, int target,
+				    int refclk, struct gma_clock_t *best_clock);
+
+static bool mrst_sdvo_find_best_pll(const struct gma_limit_t *limit,
+				    struct drm_crtc *crtc, int target,
+				    int refclk, struct gma_clock_t *best_clock);
+
+static const struct gma_limit_t mrst_limits[] = {
 	{			/* MRST_LIMIT_LVDS_100L */
 	 .dot = {.min = MRST_DOT_MIN, .max = MRST_DOT_MAX},
 	 .m = {.min = MRST_M_MIN_100L, .max = MRST_M_MAX_100L},
 	 .p1 = {.min = MRST_P1_MIN, .max = MRST_P1_MAX_1},
+	 .find_pll = mrst_lvds_find_best_pll,
 	 },
 	{			/* MRST_LIMIT_LVDS_83L */
 	 .dot = {.min = MRST_DOT_MIN, .max = MRST_DOT_MAX},
 	 .m = {.min = MRST_M_MIN_83, .max = MRST_M_MAX_83},
 	 .p1 = {.min = MRST_P1_MIN, .max = MRST_P1_MAX_0},
+	 .find_pll = mrst_lvds_find_best_pll,
 	 },
 	{			/* MRST_LIMIT_LVDS_100 */
 	 .dot = {.min = MRST_DOT_MIN, .max = MRST_DOT_MAX},
 	 .m = {.min = MRST_M_MIN_100, .max = MRST_M_MAX_100},
 	 .p1 = {.min = MRST_P1_MIN, .max = MRST_P1_MAX_1},
+	 .find_pll = mrst_lvds_find_best_pll,
+	 },
+	{			/* MRST_LIMIT_SDVO */
+	 .vco = {.min = 1400000, .max = 2800000},
+	 .n = {.min = 3, .max = 7},
+	 .m = {.min = 80, .max = 137},
+	 .p1 = {.min = 1, .max = 2},
+	 .p2 = {.dot_limit = 200000, .p2_slow = 10, .p2_fast = 10},
+	 .find_pll = mrst_sdvo_find_best_pll,
 	 },
 };
 
@@ -82,9 +87,10 @@ static const u32 oaktrail_m_converts[] = {
 	0x12, 0x09, 0x24, 0x32, 0x39, 0x1c,
 };
 
-static const struct oaktrail_limit_t *oaktrail_limit(struct drm_crtc *crtc)
+static const struct gma_limit_t *mrst_limit(struct drm_crtc *crtc,
+					    int refclk)
 {
-	const struct oaktrail_limit_t *limit = NULL;
+	const struct gma_limit_t *limit = NULL;
 	struct drm_device *dev = crtc->dev;
 	struct drm_psb_private *dev_priv = dev->dev_private;
 
@@ -92,45 +98,100 @@ static const struct oaktrail_limit_t *oaktrail_limit(struct drm_crtc *crtc)
 	    || gma_pipe_has_type(crtc, INTEL_OUTPUT_MIPI)) {
 		switch (dev_priv->core_freq) {
 		case 100:
-			limit = &oaktrail_limits[MRST_LIMIT_LVDS_100L];
+			limit = &mrst_limits[MRST_LIMIT_LVDS_100L];
 			break;
 		case 166:
-			limit = &oaktrail_limits[MRST_LIMIT_LVDS_83];
+			limit = &mrst_limits[MRST_LIMIT_LVDS_83];
 			break;
 		case 200:
-			limit = &oaktrail_limits[MRST_LIMIT_LVDS_100];
+			limit = &mrst_limits[MRST_LIMIT_LVDS_100];
 			break;
 		}
+	} else if (gma_pipe_has_type(crtc, INTEL_OUTPUT_SDVO)) {
+		limit = &mrst_limits[MRST_LIMIT_SDVO];
 	} else {
 		limit = NULL;
-		dev_err(dev->dev, "oaktrail_limit Wrong display type.\n");
+		dev_err(dev->dev, "mrst_limit Wrong display type.\n");
 	}
 
 	return limit;
 }
 
 /** Derive the pixel clock for the given refclk and divisors for 8xx chips. */
-static void oaktrail_clock(int refclk, struct oaktrail_clock_t *clock)
+static void mrst_lvds_clock(int refclk, struct gma_clock_t *clock)
 {
 	clock->dot = (refclk * clock->m) / (14 * clock->p1);
 }
 
-static void mrstPrintPll(char *prefix, struct oaktrail_clock_t *clock)
+static void mrst_print_pll(struct gma_clock_t *clock)
 {
-	pr_debug("%s: dotclock = %d,  m = %d, p1 = %d.\n",
-	     prefix, clock->dot, clock->m, clock->p1);
+	DRM_DEBUG_DRIVER("dotclock=%d,  m=%d, m1=%d, m2=%d, n=%d, p1=%d, p2=%d\n",
+			 clock->dot, clock->m, clock->m1, clock->m2, clock->n,
+			 clock->p1, clock->p2);
+}
+
+static bool mrst_sdvo_find_best_pll(const struct gma_limit_t *limit,
+				    struct drm_crtc *crtc, int target,
+				    int refclk, struct gma_clock_t *best_clock)
+{
+	struct gma_clock_t clock;
+	u32 target_vco, actual_freq;
+	s32 freq_error, min_error = 100000;
+
+	memset(best_clock, 0, sizeof(*best_clock));
+
+	for (clock.m = limit->m.min; clock.m <= limit->m.max; clock.m++) {
+		for (clock.n = limit->n.min; clock.n <= limit->n.max;
+		     clock.n++) {
+			for (clock.p1 = limit->p1.min;
+			     clock.p1 <= limit->p1.max; clock.p1++) {
+				/* p2 value always stored in p2_slow on SDVO */
+				clock.p = clock.p1 * limit->p2.p2_slow;
+				target_vco = target * clock.p;
+
+				/* VCO will increase at this point so break */
+				if (target_vco > limit->vco.max)
+					break;
+
+				if (target_vco < limit->vco.min)
+					continue;
+
+				actual_freq = (refclk * clock.m) /
+					      (clock.n * clock.p);
+				freq_error = 10000 -
+					     ((target * 10000) / actual_freq);
+
+				if (freq_error < -min_error) {
+					/* freq_error will start to decrease at
+					   this point so break */
+					break;
+				}
+
+				if (freq_error < 0)
+					freq_error = -freq_error;
+
+				if (freq_error < min_error) {
+					min_error = freq_error;
+					*best_clock = clock;
+				}
+			}
+		}
+		if (min_error == 0)
+			break;
+	}
+
+	return min_error == 0;
 }
 
 /**
  * Returns a set of divisors for the desired target clock with the given refclk,
  * or FALSE.  Divisor values are the actual divisors for
  */
-static bool
-mrstFindBestPLL(struct drm_crtc *crtc, int target, int refclk,
-		struct oaktrail_clock_t *best_clock)
+static bool mrst_lvds_find_best_pll(const struct gma_limit_t *limit,
+				    struct drm_crtc *crtc, int target,
+				    int refclk, struct gma_clock_t *best_clock)
 {
-	struct oaktrail_clock_t clock;
-	const struct oaktrail_limit_t *limit = oaktrail_limit(crtc);
+	struct gma_clock_t clock;
 	int err = target;
 
 	memset(best_clock, 0, sizeof(*best_clock));
@@ -140,7 +201,7 @@ mrstFindBestPLL(struct drm_crtc *crtc, int target, int refclk,
 		     clock.p1++) {
 			int this_err;
 
-			oaktrail_clock(refclk, &clock);
+			mrst_lvds_clock(refclk, &clock);
 
 			this_err = abs(clock.dot - target);
 			if (this_err < err) {
@@ -149,7 +210,6 @@ mrstFindBestPLL(struct drm_crtc *crtc, int target, int refclk,
 			}
 		}
 	}
-	dev_dbg(crtc->dev->dev, "mrstFindBestPLL err = %d.\n", err);
 	return err != target;
 }
 
@@ -167,8 +227,10 @@ static void oaktrail_crtc_dpms(struct drm_crtc *crtc, int mode)
 	int pipe = gma_crtc->pipe;
 	const struct psb_offset *map = &dev_priv->regmap[pipe];
 	u32 temp;
+	int i;
+	int need_aux = gma_pipe_has_type(crtc, INTEL_OUTPUT_SDVO) ? 1 : 0;
 
-	if (pipe == 1) {
+	if (gma_pipe_has_type(crtc, INTEL_OUTPUT_HDMI)) {
 		oaktrail_crtc_hdmi_dpms(crtc, mode);
 		return;
 	}
@@ -183,35 +245,45 @@ static void oaktrail_crtc_dpms(struct drm_crtc *crtc, int mode)
 	case DRM_MODE_DPMS_ON:
 	case DRM_MODE_DPMS_STANDBY:
 	case DRM_MODE_DPMS_SUSPEND:
-		/* Enable the DPLL */
-		temp = REG_READ(map->dpll);
-		if ((temp & DPLL_VCO_ENABLE) == 0) {
-			REG_WRITE(map->dpll, temp);
-			REG_READ(map->dpll);
-			/* Wait for the clocks to stabilize. */
-			udelay(150);
-			REG_WRITE(map->dpll, temp | DPLL_VCO_ENABLE);
-			REG_READ(map->dpll);
-			/* Wait for the clocks to stabilize. */
-			udelay(150);
-			REG_WRITE(map->dpll, temp | DPLL_VCO_ENABLE);
-			REG_READ(map->dpll);
-			/* Wait for the clocks to stabilize. */
-			udelay(150);
-		}
-		/* Enable the pipe */
-		temp = REG_READ(map->conf);
-		if ((temp & PIPEACONF_ENABLE) == 0)
-			REG_WRITE(map->conf, temp | PIPEACONF_ENABLE);
-		/* Enable the plane */
-		temp = REG_READ(map->cntr);
-		if ((temp & DISPLAY_PLANE_ENABLE) == 0) {
-			REG_WRITE(map->cntr,
-				  temp | DISPLAY_PLANE_ENABLE);
-			/* Flush the plane changes */
-			REG_WRITE(map->base, REG_READ(map->base));
-		}
+		for (i = 0; i <= need_aux; i++) {
+			/* Enable the DPLL */
+			temp = REG_READ_WITH_AUX(map->dpll, i);
+			if ((temp & DPLL_VCO_ENABLE) == 0) {
+				REG_WRITE_WITH_AUX(map->dpll, temp, i);
+				REG_READ_WITH_AUX(map->dpll, i);
+				/* Wait for the clocks to stabilize. */
+				udelay(150);
+				REG_WRITE_WITH_AUX(map->dpll,
+						   temp | DPLL_VCO_ENABLE, i);
+				REG_READ_WITH_AUX(map->dpll, i);
+				/* Wait for the clocks to stabilize. */
+				udelay(150);
+				REG_WRITE_WITH_AUX(map->dpll,
+						   temp | DPLL_VCO_ENABLE, i);
+				REG_READ_WITH_AUX(map->dpll, i);
+				/* Wait for the clocks to stabilize. */
+				udelay(150);
+			}
+
+			/* Enable the pipe */
+			temp = REG_READ_WITH_AUX(map->conf, i);
+			if ((temp & PIPEACONF_ENABLE) == 0) {
+				REG_WRITE_WITH_AUX(map->conf,
+						   temp | PIPEACONF_ENABLE, i);
+			}
+
+			/* Enable the plane */
+			temp = REG_READ_WITH_AUX(map->cntr, i);
+			if ((temp & DISPLAY_PLANE_ENABLE) == 0) {
+				REG_WRITE_WITH_AUX(map->cntr,
+						   temp | DISPLAY_PLANE_ENABLE,
+						   i);
+				/* Flush the plane changes */
+				REG_WRITE_WITH_AUX(map->base,
+					REG_READ_WITH_AUX(map->base, i), i);
+			}
 
+		}
 		gma_crtc_load_lut(crtc);
 
 		/* Give the overlay scaler a chance to enable
@@ -223,48 +295,52 @@ static void oaktrail_crtc_dpms(struct drm_crtc *crtc, int mode)
 		 * if it's on this pipe */
 		/* psb_intel_crtc_dpms_video(crtc, FALSE); TODO */
 
-		/* Disable the VGA plane that we never use */
-		REG_WRITE(VGACNTRL, VGA_DISP_DISABLE);
-		/* Disable display plane */
-		temp = REG_READ(map->cntr);
-		if ((temp & DISPLAY_PLANE_ENABLE) != 0) {
-			REG_WRITE(map->cntr,
-				  temp & ~DISPLAY_PLANE_ENABLE);
-			/* Flush the plane changes */
-			REG_WRITE(map->base, REG_READ(map->base));
-			REG_READ(map->base);
-		}
+		for (i = 0; i <= need_aux; i++) {
+			/* Disable the VGA plane that we never use */
+			REG_WRITE_WITH_AUX(VGACNTRL, VGA_DISP_DISABLE, i);
+			/* Disable display plane */
+			temp = REG_READ_WITH_AUX(map->cntr, i);
+			if ((temp & DISPLAY_PLANE_ENABLE) != 0) {
+				REG_WRITE_WITH_AUX(map->cntr,
+					temp & ~DISPLAY_PLANE_ENABLE, i);
+				/* Flush the plane changes */
+				REG_WRITE_WITH_AUX(map->base,
+						   REG_READ(map->base), i);
+				REG_READ_WITH_AUX(map->base, i);
+			}
 
-		/* Next, disable display pipes */
-		temp = REG_READ(map->conf);
-		if ((temp & PIPEACONF_ENABLE) != 0) {
-			REG_WRITE(map->conf, temp & ~PIPEACONF_ENABLE);
-			REG_READ(map->conf);
-		}
-		/* Wait for for the pipe disable to take effect. */
-		gma_wait_for_vblank(dev);
+			/* Next, disable display pipes */
+			temp = REG_READ_WITH_AUX(map->conf, i);
+			if ((temp & PIPEACONF_ENABLE) != 0) {
+				REG_WRITE_WITH_AUX(map->conf,
+						   temp & ~PIPEACONF_ENABLE, i);
+				REG_READ_WITH_AUX(map->conf, i);
+			}
+			/* Wait for for the pipe disable to take effect. */
+			gma_wait_for_vblank(dev);
+
+			temp = REG_READ_WITH_AUX(map->dpll, i);
+			if ((temp & DPLL_VCO_ENABLE) != 0) {
+				REG_WRITE_WITH_AUX(map->dpll,
+						   temp & ~DPLL_VCO_ENABLE, i);
+				REG_READ_WITH_AUX(map->dpll, i);
+			}
 
-		temp = REG_READ(map->dpll);
-		if ((temp & DPLL_VCO_ENABLE) != 0) {
-			REG_WRITE(map->dpll, temp & ~DPLL_VCO_ENABLE);
-			REG_READ(map->dpll);
+			/* Wait for the clocks to turn off. */
+			udelay(150);
 		}
-
-		/* Wait for the clocks to turn off. */
-		udelay(150);
 		break;
 	}
 
-	/*Set FIFO Watermarks*/
-	REG_WRITE(DSPARB, 0x3FFF);
-	REG_WRITE(DSPFW1, 0x3F88080A);
-	REG_WRITE(DSPFW2, 0x0b060808);
+	/* Set FIFO Watermarks (values taken from EMGD) */
+	REG_WRITE(DSPARB, 0x3f80);
+	REG_WRITE(DSPFW1, 0x3f8f0404);
+	REG_WRITE(DSPFW2, 0x04040f04);
 	REG_WRITE(DSPFW3, 0x0);
-	REG_WRITE(DSPFW4, 0x08030404);
+	REG_WRITE(DSPFW4, 0x04040404);
 	REG_WRITE(DSPFW5, 0x04040404);
 	REG_WRITE(DSPFW6, 0x78);
-	REG_WRITE(0x70400, REG_READ(0x70400) | 0x4000);
-	/* Must write Bit 14 of the Chicken Bit Register */
+	REG_WRITE(DSPCHICKENBIT, REG_READ(DSPCHICKENBIT) | 0xc040);
 
 	gma_power_end(dev);
 }
@@ -297,7 +373,8 @@ static int oaktrail_crtc_mode_set(struct drm_crtc *crtc,
 	int pipe = gma_crtc->pipe;
 	const struct psb_offset *map = &dev_priv->regmap[pipe];
 	int refclk = 0;
-	struct oaktrail_clock_t clock;
+	struct gma_clock_t clock;
+	const struct gma_limit_t *limit;
 	u32 dpll = 0, fp = 0, dspcntr, pipeconf;
 	bool ok, is_sdvo = false;
 	bool is_lvds = false;
@@ -306,8 +383,10 @@ static int oaktrail_crtc_mode_set(struct drm_crtc *crtc,
 	struct gma_encoder *gma_encoder = NULL;
 	uint64_t scalingType = DRM_MODE_SCALE_FULLSCREEN;
 	struct drm_connector *connector;
+	int i;
+	int need_aux = gma_pipe_has_type(crtc, INTEL_OUTPUT_SDVO) ? 1 : 0;
 
-	if (pipe == 1)
+	if (gma_pipe_has_type(crtc, INTEL_OUTPUT_HDMI))
 		return oaktrail_crtc_hdmi_mode_set(crtc, mode, adjusted_mode, x, y, old_fb);
 
 	if (!gma_power_begin(dev, true))
@@ -340,15 +419,17 @@ static int oaktrail_crtc_mode_set(struct drm_crtc *crtc,
 	}
 
 	/* Disable the VGA plane that we never use */
-	REG_WRITE(VGACNTRL, VGA_DISP_DISABLE);
+	for (i = 0; i <= need_aux; i++)
+		REG_WRITE_WITH_AUX(VGACNTRL, VGA_DISP_DISABLE, i);
 
 	/* Disable the panel fitter if it was on our pipe */
 	if (oaktrail_panel_fitter_pipe(dev) == pipe)
 		REG_WRITE(PFIT_CONTROL, 0);
 
-	REG_WRITE(map->src,
-		  ((mode->crtc_hdisplay - 1) << 16) |
-		  (mode->crtc_vdisplay - 1));
+	for (i = 0; i <= need_aux; i++) {
+		REG_WRITE_WITH_AUX(map->src, ((mode->crtc_hdisplay - 1) << 16) |
+					     (mode->crtc_vdisplay - 1), i);
+	}
 
 	if (gma_encoder)
 		drm_object_property_get_value(&connector->base,
@@ -365,35 +446,39 @@ static int oaktrail_crtc_mode_set(struct drm_crtc *crtc,
 		offsetY = (adjusted_mode->crtc_vdisplay -
 			   mode->crtc_vdisplay) / 2;
 
-		REG_WRITE(map->htotal, (mode->crtc_hdisplay - 1) |
-			((adjusted_mode->crtc_htotal - 1) << 16));
-		REG_WRITE(map->vtotal, (mode->crtc_vdisplay - 1) |
-			((adjusted_mode->crtc_vtotal - 1) << 16));
-		REG_WRITE(map->hblank,
-			(adjusted_mode->crtc_hblank_start - offsetX - 1) |
-			((adjusted_mode->crtc_hblank_end - offsetX - 1) << 16));
-		REG_WRITE(map->hsync,
-			(adjusted_mode->crtc_hsync_start - offsetX - 1) |
-			((adjusted_mode->crtc_hsync_end - offsetX - 1) << 16));
-		REG_WRITE(map->vblank,
-			(adjusted_mode->crtc_vblank_start - offsetY - 1) |
-			((adjusted_mode->crtc_vblank_end - offsetY - 1) << 16));
-		REG_WRITE(map->vsync,
-			(adjusted_mode->crtc_vsync_start - offsetY - 1) |
-			((adjusted_mode->crtc_vsync_end - offsetY - 1) << 16));
+		for (i = 0; i <= need_aux; i++) {
+			REG_WRITE_WITH_AUX(map->htotal, (mode->crtc_hdisplay - 1) |
+				((adjusted_mode->crtc_htotal - 1) << 16), i);
+			REG_WRITE_WITH_AUX(map->vtotal, (mode->crtc_vdisplay - 1) |
+				((adjusted_mode->crtc_vtotal - 1) << 16), i);
+			REG_WRITE_WITH_AUX(map->hblank,
+				(adjusted_mode->crtc_hblank_start - offsetX - 1) |
+				((adjusted_mode->crtc_hblank_end - offsetX - 1) << 16), i);
+			REG_WRITE_WITH_AUX(map->hsync,
+				(adjusted_mode->crtc_hsync_start - offsetX - 1) |
+				((adjusted_mode->crtc_hsync_end - offsetX - 1) << 16), i);
+			REG_WRITE_WITH_AUX(map->vblank,
+				(adjusted_mode->crtc_vblank_start - offsetY - 1) |
+				((adjusted_mode->crtc_vblank_end - offsetY - 1) << 16), i);
+			REG_WRITE_WITH_AUX(map->vsync,
+				(adjusted_mode->crtc_vsync_start - offsetY - 1) |
+				((adjusted_mode->crtc_vsync_end - offsetY - 1) << 16), i);
+		}
 	} else {
-		REG_WRITE(map->htotal, (adjusted_mode->crtc_hdisplay - 1) |
-			((adjusted_mode->crtc_htotal - 1) << 16));
-		REG_WRITE(map->vtotal, (adjusted_mode->crtc_vdisplay - 1) |
-			((adjusted_mode->crtc_vtotal - 1) << 16));
-		REG_WRITE(map->hblank, (adjusted_mode->crtc_hblank_start - 1) |
-			((adjusted_mode->crtc_hblank_end - 1) << 16));
-		REG_WRITE(map->hsync, (adjusted_mode->crtc_hsync_start - 1) |
-			((adjusted_mode->crtc_hsync_end - 1) << 16));
-		REG_WRITE(map->vblank, (adjusted_mode->crtc_vblank_start - 1) |
-			((adjusted_mode->crtc_vblank_end - 1) << 16));
-		REG_WRITE(map->vsync, (adjusted_mode->crtc_vsync_start - 1) |
-			((adjusted_mode->crtc_vsync_end - 1) << 16));
+		for (i = 0; i <= need_aux; i++) {
+			REG_WRITE_WITH_AUX(map->htotal, (adjusted_mode->crtc_hdisplay - 1) |
+				((adjusted_mode->crtc_htotal - 1) << 16), i);
+			REG_WRITE_WITH_AUX(map->vtotal, (adjusted_mode->crtc_vdisplay - 1) |
+				((adjusted_mode->crtc_vtotal - 1) << 16), i);
+			REG_WRITE_WITH_AUX(map->hblank, (adjusted_mode->crtc_hblank_start - 1) |
+				((adjusted_mode->crtc_hblank_end - 1) << 16), i);
+			REG_WRITE_WITH_AUX(map->hsync, (adjusted_mode->crtc_hsync_start - 1) |
+				((adjusted_mode->crtc_hsync_end - 1) << 16), i);
+			REG_WRITE_WITH_AUX(map->vblank, (adjusted_mode->crtc_vblank_start - 1) |
+				((adjusted_mode->crtc_vblank_end - 1) << 16), i);
+			REG_WRITE_WITH_AUX(map->vsync, (adjusted_mode->crtc_vsync_start - 1) |
+				((adjusted_mode->crtc_vsync_end - 1) << 16), i);
+		}
 	}
 
 	/* Flush the plane changes */
@@ -418,21 +503,30 @@ static int oaktrail_crtc_mode_set(struct drm_crtc *crtc,
 	if (is_mipi)
 		goto oaktrail_crtc_mode_set_exit;
 
-	refclk = dev_priv->core_freq * 1000;
 
 	dpll = 0;		/*BIT16 = 0 for 100MHz reference */
 
-	ok = mrstFindBestPLL(crtc, adjusted_mode->clock, refclk, &clock);
+	refclk = is_sdvo ? 96000 : dev_priv->core_freq * 1000;
+	limit = mrst_limit(crtc, refclk);
+	ok = limit->find_pll(limit, crtc, adjusted_mode->clock,
+			     refclk, &clock);
 
-	if (!ok) {
-		dev_dbg(dev->dev, "mrstFindBestPLL fail in oaktrail_crtc_mode_set.\n");
-	} else {
-		dev_dbg(dev->dev, "oaktrail_crtc_mode_set pixel clock = %d,"
-			 "m = %x, p1 = %x.\n", clock.dot, clock.m,
-			 clock.p1);
+	if (is_sdvo) {
+		/* Convert calculated values to register values */
+		clock.p1 = (1L << (clock.p1 - 1));
+		clock.m -= 2;
+		clock.n = (1L << (clock.n - 1));
 	}
 
-	fp = oaktrail_m_converts[(clock.m - MRST_M_MIN)] << 8;
+	if (!ok)
+		DRM_ERROR("Failed to find proper PLL settings");
+
+	mrst_print_pll(&clock);
+
+	if (is_sdvo)
+		fp = clock.n << 16 | clock.m;
+	else
+		fp = oaktrail_m_converts[(clock.m - MRST_M_MIN)] << 8;
 
 	dpll |= DPLL_VGA_MODE_DIS;
 
@@ -456,38 +550,43 @@ static int oaktrail_crtc_mode_set(struct drm_crtc *crtc,
 
 
 	/* compute bitmask from p1 value */
-	dpll |= (1 << (clock.p1 - 2)) << 17;
+	if (is_sdvo)
+		dpll |= clock.p1 << 16; // dpll |= (1 << (clock.p1 - 1)) << 16;
+	else
+		dpll |= (1 << (clock.p1 - 2)) << 17;
 
 	dpll |= DPLL_VCO_ENABLE;
 
-	mrstPrintPll("chosen", &clock);
-
 	if (dpll & DPLL_VCO_ENABLE) {
-		REG_WRITE(map->fp0, fp);
-		REG_WRITE(map->dpll, dpll & ~DPLL_VCO_ENABLE);
-		REG_READ(map->dpll);
-		/* Check the DPLLA lock bit PIPEACONF[29] */
-		udelay(150);
+		for (i = 0; i <= need_aux; i++) {
+			REG_WRITE_WITH_AUX(map->fp0, fp, i);
+			REG_WRITE_WITH_AUX(map->dpll, dpll & ~DPLL_VCO_ENABLE, i);
+			REG_READ_WITH_AUX(map->dpll, i);
+			/* Check the DPLLA lock bit PIPEACONF[29] */
+			udelay(150);
+		}
 	}
 
-	REG_WRITE(map->fp0, fp);
-	REG_WRITE(map->dpll, dpll);
-	REG_READ(map->dpll);
-	/* Wait for the clocks to stabilize. */
-	udelay(150);
+	for (i = 0; i <= need_aux; i++) {
+		REG_WRITE_WITH_AUX(map->fp0, fp, i);
+		REG_WRITE_WITH_AUX(map->dpll, dpll, i);
+		REG_READ_WITH_AUX(map->dpll, i);
+		/* Wait for the clocks to stabilize. */
+		udelay(150);
 
-	/* write it again -- the BIOS does, after all */
-	REG_WRITE(map->dpll, dpll);
-	REG_READ(map->dpll);
-	/* Wait for the clocks to stabilize. */
-	udelay(150);
+		/* write it again -- the BIOS does, after all */
+		REG_WRITE_WITH_AUX(map->dpll, dpll, i);
+		REG_READ_WITH_AUX(map->dpll, i);
+		/* Wait for the clocks to stabilize. */
+		udelay(150);
 
-	REG_WRITE(map->conf, pipeconf);
-	REG_READ(map->conf);
-	gma_wait_for_vblank(dev);
+		REG_WRITE_WITH_AUX(map->conf, pipeconf, i);
+		REG_READ_WITH_AUX(map->conf, i);
+		gma_wait_for_vblank(dev);
 
-	REG_WRITE(map->cntr, dspcntr);
-	gma_wait_for_vblank(dev);
+		REG_WRITE_WITH_AUX(map->cntr, dspcntr, i);
+		gma_wait_for_vblank(dev);
+	}
 
 oaktrail_crtc_mode_set_exit:
 	gma_power_end(dev);
@@ -565,3 +664,9 @@ const struct drm_crtc_helper_funcs oaktrail_helper_funcs = {
 	.commit = gma_crtc_commit,
 };
 
+/* Not used yet */
+const struct gma_clock_funcs mrst_clock_funcs = {
+	.clock = mrst_lvds_clock,
+	.limit = mrst_limit,
+	.pll_is_valid = gma_pll_is_valid,
+};
diff --git a/drivers/gpu/drm/gma500/oaktrail_device.c b/drivers/gpu/drm/gma500/oaktrail_device.c
index 7a9ce000fd86..368a03ae3010 100644
--- a/drivers/gpu/drm/gma500/oaktrail_device.c
+++ b/drivers/gpu/drm/gma500/oaktrail_device.c
@@ -40,6 +40,9 @@ static int oaktrail_output_init(struct drm_device *dev)
 		dev_err(dev->dev, "DSI is not supported\n");
 	if (dev_priv->hdmi_priv)
 		oaktrail_hdmi_init(dev, &dev_priv->mode_dev);
+
+	psb_intel_sdvo_init(dev, SDVOB);
+
 	return 0;
 }
 
@@ -526,6 +529,7 @@ static int oaktrail_chip_setup(struct drm_device *dev)
 		psb_intel_opregion_init(dev);
 		psb_intel_init_bios(dev);
 	}
+	gma_intel_setup_gmbus(dev);
 	oaktrail_hdmi_setup(dev);
 	return 0;
 }
@@ -534,6 +538,7 @@ static void oaktrail_teardown(struct drm_device *dev)
 {
 	struct drm_psb_private *dev_priv = dev->dev_private;
 
+	gma_intel_teardown_gmbus(dev);
 	oaktrail_hdmi_teardown(dev);
 	if (!dev_priv->has_gct)
 		psb_intel_destroy_bios(dev);
@@ -546,6 +551,7 @@ const struct psb_ops oaktrail_chip_ops = {
 	.crtcs = 2,
 	.hdmi_mask = (1 << 1),
 	.lvds_mask = (1 << 0),
+	.sdvo_mask = (1 << 1),
 	.cursor_needs_phys = 0,
 	.sgx_offset = MRST_SGX_OFFSET,
 
diff --git a/drivers/gpu/drm/gma500/oaktrail_hdmi_i2c.c b/drivers/gpu/drm/gma500/oaktrail_hdmi_i2c.c
index 1eb86c79523e..e28107061148 100644
--- a/drivers/gpu/drm/gma500/oaktrail_hdmi_i2c.c
+++ b/drivers/gpu/drm/gma500/oaktrail_hdmi_i2c.c
@@ -99,7 +99,7 @@ static int xfer_read(struct i2c_adapter *adap, struct i2c_msg *pmsg)
 	i2c_dev->status = I2C_STAT_INIT;
 	i2c_dev->msg = pmsg;
 	i2c_dev->buf_offset = 0;
-	INIT_COMPLETION(i2c_dev->complete);
+	reinit_completion(&i2c_dev->complete);
 
 	/* Enable I2C transaction */
 	temp = ((pmsg->len) << 20) | HI2C_EDID_READ | HI2C_ENABLE_TRANSACTION;
diff --git a/drivers/gpu/drm/gma500/oaktrail_lvds.c b/drivers/gpu/drm/gma500/oaktrail_lvds.c
index 3ece553311fe..5e0697862736 100644
--- a/drivers/gpu/drm/gma500/oaktrail_lvds.c
+++ b/drivers/gpu/drm/gma500/oaktrail_lvds.c
@@ -218,30 +218,6 @@ static const struct drm_encoder_helper_funcs oaktrail_lvds_helper_funcs = {
 	.commit = oaktrail_lvds_commit,
 };
 
-static struct drm_display_mode lvds_configuration_modes[] = {
-	/* hard coded fixed mode for TPO LTPS LPJ040K001A */
-	{ DRM_MODE("800x480",  DRM_MODE_TYPE_DRIVER, 33264, 800, 836,
-		   846, 1056, 0, 480, 489, 491, 525, 0, 0) },
-	/* hard coded fixed mode for LVDS 800x480 */
-	{ DRM_MODE("800x480",  DRM_MODE_TYPE_DRIVER, 30994, 800, 801,
-		   802, 1024, 0, 480, 481, 482, 525, 0, 0) },
-	/* hard coded fixed mode for Samsung 480wsvga LVDS 1024x600@75 */
-	{ DRM_MODE("1024x600", DRM_MODE_TYPE_DRIVER, 53990, 1024, 1072,
-		   1104, 1184, 0, 600, 603, 604, 608, 0, 0) },
-	/* hard coded fixed mode for Samsung 480wsvga LVDS 1024x600@75 */
-	{ DRM_MODE("1024x600", DRM_MODE_TYPE_DRIVER, 53990, 1024, 1104,
-		   1136, 1184, 0, 600, 603, 604, 608, 0, 0) },
-	/* hard coded fixed mode for Sharp wsvga LVDS 1024x600 */
-	{ DRM_MODE("1024x600", DRM_MODE_TYPE_DRIVER, 48885, 1024, 1124,
-		   1204, 1312, 0, 600, 607, 610, 621, 0, 0) },
-	/* hard coded fixed mode for LVDS 1024x768 */
-	{ DRM_MODE("1024x768", DRM_MODE_TYPE_DRIVER, 65000, 1024, 1048,
-		   1184, 1344, 0, 768, 771, 777, 806, 0, 0) },
-	/* hard coded fixed mode for LVDS 1366x768 */
-	{ DRM_MODE("1366x768", DRM_MODE_TYPE_DRIVER, 77500, 1366, 1430,
-		   1558, 1664, 0, 768, 769, 770, 776, 0, 0) },
-};
-
 /* Returns the panel fixed mode from configuration. */
 
 static void oaktrail_lvds_get_configuration_mode(struct drm_device *dev,
@@ -303,10 +279,10 @@ static void oaktrail_lvds_get_configuration_mode(struct drm_device *dev,
 			mode_dev->panel_fixed_mode =
 				drm_mode_duplicate(dev,
 					dev_priv->lfp_lvds_vbt_mode);
-	/* Then guess */
+
+	/* If we still got no mode then bail */
 	if (mode_dev->panel_fixed_mode == NULL)
-		mode_dev->panel_fixed_mode
-			= drm_mode_duplicate(dev, &lvds_configuration_modes[2]);
+		return;
 
 	drm_mode_set_name(mode_dev->panel_fixed_mode);
 	drm_mode_set_crtcinfo(mode_dev->panel_fixed_mode, 0);
diff --git a/drivers/gpu/drm/gma500/psb_device.c b/drivers/gpu/drm/gma500/psb_device.c
index 697678619bd1..23fb33f1471b 100644
--- a/drivers/gpu/drm/gma500/psb_device.c
+++ b/drivers/gpu/drm/gma500/psb_device.c
@@ -373,6 +373,7 @@ const struct psb_ops psb_chip_ops = {
 	.crtcs = 2,
 	.hdmi_mask = (1 << 0),
 	.lvds_mask = (1 << 1),
+	.sdvo_mask = (1 << 0),
 	.cursor_needs_phys = 1,
 	.sgx_offset = PSB_SGX_OFFSET,
 	.chip_setup = psb_chip_setup,
diff --git a/drivers/gpu/drm/gma500/psb_drv.c b/drivers/gpu/drm/gma500/psb_drv.c
index fcb4e9ff1f20..1199180667c9 100644
--- a/drivers/gpu/drm/gma500/psb_drv.c
+++ b/drivers/gpu/drm/gma500/psb_drv.c
@@ -251,6 +251,12 @@ static int psb_driver_unload(struct drm_device *dev)
 			iounmap(dev_priv->sgx_reg);
 			dev_priv->sgx_reg = NULL;
 		}
+		if (dev_priv->aux_reg) {
+			iounmap(dev_priv->aux_reg);
+			dev_priv->aux_reg = NULL;
+		}
+		if (dev_priv->aux_pdev)
+			pci_dev_put(dev_priv->aux_pdev);
 
 		/* Destroy VBT data */
 		psb_intel_destroy_bios(dev);
@@ -266,7 +272,7 @@ static int psb_driver_unload(struct drm_device *dev)
 static int psb_driver_load(struct drm_device *dev, unsigned long chipset)
 {
 	struct drm_psb_private *dev_priv;
-	unsigned long resource_start;
+	unsigned long resource_start, resource_len;
 	unsigned long irqflags;
 	int ret = -ENOMEM;
 	struct drm_connector *connector;
@@ -296,6 +302,30 @@ static int psb_driver_load(struct drm_device *dev, unsigned long chipset)
 	if (!dev_priv->sgx_reg)
 		goto out_err;
 
+	if (IS_MRST(dev)) {
+		dev_priv->aux_pdev = pci_get_bus_and_slot(0, PCI_DEVFN(3, 0));
+
+		if (dev_priv->aux_pdev) {
+			resource_start = pci_resource_start(dev_priv->aux_pdev,
+							    PSB_AUX_RESOURCE);
+			resource_len = pci_resource_len(dev_priv->aux_pdev,
+							PSB_AUX_RESOURCE);
+			dev_priv->aux_reg = ioremap_nocache(resource_start,
+							    resource_len);
+			if (!dev_priv->aux_reg)
+				goto out_err;
+
+			DRM_DEBUG_KMS("Found aux vdc");
+		} else {
+			/* Couldn't find the aux vdc so map to primary vdc */
+			dev_priv->aux_reg = dev_priv->vdc_reg;
+			DRM_DEBUG_KMS("Couldn't find aux pci device");
+		}
+		dev_priv->gmbus_reg = dev_priv->aux_reg;
+	} else {
+		dev_priv->gmbus_reg = dev_priv->vdc_reg;
+	}
+
 	psb_intel_opregion_setup(dev);
 
 	ret = dev_priv->ops->chip_setup(dev);
@@ -359,7 +389,7 @@ static int psb_driver_load(struct drm_device *dev, unsigned long chipset)
 
 	drm_irq_install(dev);
 
-	dev->vblank_disable_allowed = 1;
+	dev->vblank_disable_allowed = true;
 
 	dev->max_vblank_count = 0xffffff; /* only 24 bits of frame count */
 
@@ -449,7 +479,7 @@ static int psb_gamma_ioctl(struct drm_device *dev, void *data,
 	obj = drm_mode_object_find(dev, obj_id, DRM_MODE_OBJECT_CONNECTOR);
 	if (!obj) {
 		dev_dbg(dev->dev, "Invalid Connector object.\n");
-		return -EINVAL;
+		return -ENOENT;
 	}
 
 	connector = obj_to_connector(obj);
@@ -491,7 +521,7 @@ static int psb_mode_operation_ioctl(struct drm_device *dev, void *data,
 		obj = drm_mode_object_find(dev, obj_id,
 					DRM_MODE_OBJECT_CONNECTOR);
 		if (!obj) {
-			ret = -EINVAL;
+			ret = -ENOENT;
 			goto mode_op_out;
 		}
 
@@ -646,7 +676,6 @@ static struct drm_driver driver = {
 	.preclose = psb_driver_preclose,
 	.postclose = psb_driver_close,
 
-	.gem_init_object = psb_gem_init_object,
 	.gem_free_object = psb_gem_free_object,
 	.gem_vm_ops = &psb_gem_vm_ops,
 	.dumb_create = psb_gem_dumb_create,
diff --git a/drivers/gpu/drm/gma500/psb_drv.h b/drivers/gpu/drm/gma500/psb_drv.h
index 4535ac7708f8..b59e6588c343 100644
--- a/drivers/gpu/drm/gma500/psb_drv.h
+++ b/drivers/gpu/drm/gma500/psb_drv.h
@@ -44,10 +44,10 @@ enum {
 	CHIP_MFLD_0130 = 3,		/* Medfield */
 };
 
-#define IS_PSB(dev) (((dev)->pci_device & 0xfffe) == 0x8108)
-#define IS_MRST(dev) (((dev)->pci_device & 0xfffc) == 0x4100)
-#define IS_MFLD(dev) (((dev)->pci_device & 0xfff8) == 0x0130)
-#define IS_CDV(dev) (((dev)->pci_device & 0xfff0) == 0x0be0)
+#define IS_PSB(dev) (((dev)->pdev->device & 0xfffe) == 0x8108)
+#define IS_MRST(dev) (((dev)->pdev->device & 0xfff0) == 0x4100)
+#define IS_MFLD(dev) (((dev)->pdev->device & 0xfff8) == 0x0130)
+#define IS_CDV(dev) (((dev)->pdev->device & 0xfff0) == 0x0be0)
 
 /*
  * Driver definitions
@@ -75,6 +75,7 @@ enum {
  *	PCI resource identifiers
  */
 #define PSB_MMIO_RESOURCE	 0
+#define PSB_AUX_RESOURCE	 0
 #define PSB_GATT_RESOURCE	 2
 #define PSB_GTT_RESOURCE	 3
 /*
@@ -455,6 +456,7 @@ struct psb_ops;
 
 struct drm_psb_private {
 	struct drm_device *dev;
+	struct pci_dev *aux_pdev; /* Currently only used by mrst */
 	const struct psb_ops *ops;
 	const struct psb_offset *regmap;
 	
@@ -486,6 +488,7 @@ struct drm_psb_private {
 
 	uint8_t __iomem *sgx_reg;
 	uint8_t __iomem *vdc_reg;
+	uint8_t __iomem *aux_reg; /* Auxillary vdc pipe regs */
 	uint32_t gatt_free_offset;
 
 	/*
@@ -532,6 +535,7 @@ struct drm_psb_private {
 
 	/* gmbus */
 	struct intel_gmbus *gmbus;
+	uint8_t __iomem *gmbus_reg;
 
 	/* Used by SDVO */
 	int crt_ddc_pin;
@@ -672,6 +676,7 @@ struct psb_ops {
 	int sgx_offset;		/* Base offset of SGX device */
 	int hdmi_mask;		/* Mask of HDMI CRTCs */
 	int lvds_mask;		/* Mask of LVDS CRTCs */
+	int sdvo_mask;		/* Mask of SDVO CRTCs */
 	int cursor_needs_phys;  /* If cursor base reg need physical address */
 
 	/* Sub functions */
@@ -837,7 +842,6 @@ extern const struct drm_connector_helper_funcs
 extern const struct drm_connector_funcs psb_intel_lvds_connector_funcs;
 
 /* gem.c */
-extern int psb_gem_init_object(struct drm_gem_object *obj);
 extern void psb_gem_free_object(struct drm_gem_object *obj);
 extern int psb_gem_get_aperture(struct drm_device *dev, void *data,
 			struct drm_file *file);
@@ -928,16 +932,58 @@ static inline uint32_t REGISTER_READ(struct drm_device *dev, uint32_t reg)
 	return ioread32(dev_priv->vdc_reg + reg);
 }
 
+static inline uint32_t REGISTER_READ_AUX(struct drm_device *dev, uint32_t reg)
+{
+	struct drm_psb_private *dev_priv = dev->dev_private;
+	return ioread32(dev_priv->aux_reg + reg);
+}
+
 #define REG_READ(reg)	       REGISTER_READ(dev, (reg))
+#define REG_READ_AUX(reg)      REGISTER_READ_AUX(dev, (reg))
+
+/* Useful for post reads */
+static inline uint32_t REGISTER_READ_WITH_AUX(struct drm_device *dev,
+					      uint32_t reg, int aux)
+{
+	uint32_t val;
+
+	if (aux)
+		val = REG_READ_AUX(reg);
+	else
+		val = REG_READ(reg);
+
+	return val;
+}
+
+#define REG_READ_WITH_AUX(reg, aux) REGISTER_READ_WITH_AUX(dev, (reg), (aux))
 
 static inline void REGISTER_WRITE(struct drm_device *dev, uint32_t reg,
-				      uint32_t val)
+				  uint32_t val)
 {
 	struct drm_psb_private *dev_priv = dev->dev_private;
 	iowrite32((val), dev_priv->vdc_reg + (reg));
 }
 
+static inline void REGISTER_WRITE_AUX(struct drm_device *dev, uint32_t reg,
+				      uint32_t val)
+{
+	struct drm_psb_private *dev_priv = dev->dev_private;
+	iowrite32((val), dev_priv->aux_reg + (reg));
+}
+
 #define REG_WRITE(reg, val)	REGISTER_WRITE(dev, (reg), (val))
+#define REG_WRITE_AUX(reg, val)	REGISTER_WRITE_AUX(dev, (reg), (val))
+
+static inline void REGISTER_WRITE_WITH_AUX(struct drm_device *dev, uint32_t reg,
+				      uint32_t val, int aux)
+{
+	if (aux)
+		REG_WRITE_AUX(reg, val);
+	else
+		REG_WRITE(reg, val);
+}
+
+#define REG_WRITE_WITH_AUX(reg, val, aux) REGISTER_WRITE_WITH_AUX(dev, (reg), (val), (aux))
 
 static inline void REGISTER_WRITE16(struct drm_device *dev,
 					uint32_t reg, uint32_t val)
diff --git a/drivers/gpu/drm/gma500/psb_intel_display.c b/drivers/gpu/drm/gma500/psb_intel_display.c
index 97f8a03fee43..c8841ac6c8f1 100644
--- a/drivers/gpu/drm/gma500/psb_intel_display.c
+++ b/drivers/gpu/drm/gma500/psb_intel_display.c
@@ -572,7 +572,7 @@ int psb_intel_get_pipe_from_crtc_id(struct drm_device *dev, void *data,
 
 	if (!drmmode_obj) {
 		dev_err(dev->dev, "no such CRTC id\n");
-		return -EINVAL;
+		return -ENOENT;
 	}
 
 	crtc = to_gma_crtc(obj_to_crtc(drmmode_obj));
diff --git a/drivers/gpu/drm/gma500/psb_intel_sdvo.c b/drivers/gpu/drm/gma500/psb_intel_sdvo.c
index 6f01cdf5e125..07d3a9e6d79b 100644
--- a/drivers/gpu/drm/gma500/psb_intel_sdvo.c
+++ b/drivers/gpu/drm/gma500/psb_intel_sdvo.c
@@ -228,24 +228,26 @@ static void psb_intel_sdvo_write_sdvox(struct psb_intel_sdvo *psb_intel_sdvo, u3
 {
 	struct drm_device *dev = psb_intel_sdvo->base.base.dev;
 	u32 bval = val, cval = val;
-	int i;
+	int i, j;
+	int need_aux = IS_MRST(dev) ? 1 : 0;
 
-	if (psb_intel_sdvo->sdvo_reg == SDVOB) {
-		cval = REG_READ(SDVOC);
-	} else {
-		bval = REG_READ(SDVOB);
-	}
-	/*
-	 * Write the registers twice for luck. Sometimes,
-	 * writing them only once doesn't appear to 'stick'.
-	 * The BIOS does this too. Yay, magic
-	 */
-	for (i = 0; i < 2; i++)
-	{
-		REG_WRITE(SDVOB, bval);
-		REG_READ(SDVOB);
-		REG_WRITE(SDVOC, cval);
-		REG_READ(SDVOC);
+	for (j = 0; j <= need_aux; j++) {
+		if (psb_intel_sdvo->sdvo_reg == SDVOB)
+			cval = REG_READ_WITH_AUX(SDVOC, j);
+		else
+			bval = REG_READ_WITH_AUX(SDVOB, j);
+
+		/*
+		* Write the registers twice for luck. Sometimes,
+		* writing them only once doesn't appear to 'stick'.
+		* The BIOS does this too. Yay, magic
+		*/
+		for (i = 0; i < 2; i++) {
+			REG_WRITE_WITH_AUX(SDVOB, bval, j);
+			REG_READ_WITH_AUX(SDVOB, j);
+			REG_WRITE_WITH_AUX(SDVOC, cval, j);
+			REG_READ_WITH_AUX(SDVOC, j);
+		}
 	}
 }
 
@@ -995,6 +997,7 @@ static void psb_intel_sdvo_mode_set(struct drm_encoder *encoder,
 	struct psb_intel_sdvo_dtd input_dtd;
 	int pixel_multiplier = psb_intel_mode_get_pixel_multiplier(adjusted_mode);
 	int rate;
+	int need_aux = IS_MRST(dev) ? 1 : 0;
 
 	if (!mode)
 		return;
@@ -1060,7 +1063,11 @@ static void psb_intel_sdvo_mode_set(struct drm_encoder *encoder,
 		return;
 
 	/* Set the SDVO control regs. */
-	sdvox = REG_READ(psb_intel_sdvo->sdvo_reg);
+	if (need_aux)
+		sdvox = REG_READ_AUX(psb_intel_sdvo->sdvo_reg);
+	else
+		sdvox = REG_READ(psb_intel_sdvo->sdvo_reg);
+
 	switch (psb_intel_sdvo->sdvo_reg) {
 	case SDVOB:
 		sdvox &= SDVOB_PRESERVE_MASK;
@@ -1090,6 +1097,8 @@ static void psb_intel_sdvo_dpms(struct drm_encoder *encoder, int mode)
 	struct drm_device *dev = encoder->dev;
 	struct psb_intel_sdvo *psb_intel_sdvo = to_psb_intel_sdvo(encoder);
 	u32 temp;
+	int i;
+	int need_aux = IS_MRST(dev) ? 1 : 0;
 
 	switch (mode) {
 	case DRM_MODE_DPMS_ON:
@@ -1108,19 +1117,27 @@ static void psb_intel_sdvo_dpms(struct drm_encoder *encoder, int mode)
 			psb_intel_sdvo_set_encoder_power_state(psb_intel_sdvo, mode);
 
 		if (mode == DRM_MODE_DPMS_OFF) {
-			temp = REG_READ(psb_intel_sdvo->sdvo_reg);
+			if (need_aux)
+				temp = REG_READ_AUX(psb_intel_sdvo->sdvo_reg);
+			else
+				temp = REG_READ(psb_intel_sdvo->sdvo_reg);
+
 			if ((temp & SDVO_ENABLE) != 0) {
 				psb_intel_sdvo_write_sdvox(psb_intel_sdvo, temp & ~SDVO_ENABLE);
 			}
 		}
 	} else {
 		bool input1, input2;
-		int i;
 		u8 status;
 
-		temp = REG_READ(psb_intel_sdvo->sdvo_reg);
+		if (need_aux)
+			temp = REG_READ_AUX(psb_intel_sdvo->sdvo_reg);
+		else
+			temp = REG_READ(psb_intel_sdvo->sdvo_reg);
+
 		if ((temp & SDVO_ENABLE) == 0)
 			psb_intel_sdvo_write_sdvox(psb_intel_sdvo, temp | SDVO_ENABLE);
+
 		for (i = 0; i < 2; i++)
 			gma_wait_for_vblank(dev);
 
diff --git a/drivers/gpu/drm/gma500/psb_irq.c b/drivers/gpu/drm/gma500/psb_irq.c
index 029eccf30137..ba4830342d34 100644
--- a/drivers/gpu/drm/gma500/psb_irq.c
+++ b/drivers/gpu/drm/gma500/psb_irq.c
@@ -271,15 +271,15 @@ void psb_irq_preinstall(struct drm_device *dev)
 
 	if (gma_power_is_on(dev))
 		PSB_WVDC32(0xFFFFFFFF, PSB_HWSTAM);
-	if (dev->vblank_enabled[0])
+	if (dev->vblank[0].enabled)
 		dev_priv->vdc_irq_mask |= _PSB_VSYNC_PIPEA_FLAG;
-	if (dev->vblank_enabled[1])
+	if (dev->vblank[1].enabled)
 		dev_priv->vdc_irq_mask |= _PSB_VSYNC_PIPEB_FLAG;
 
 	/* FIXME: Handle Medfield irq mask
-	if (dev->vblank_enabled[1])
+	if (dev->vblank[1].enabled)
 		dev_priv->vdc_irq_mask |= _MDFLD_PIPEB_EVENT_FLAG;
-	if (dev->vblank_enabled[2])
+	if (dev->vblank[2].enabled)
 		dev_priv->vdc_irq_mask |= _MDFLD_PIPEC_EVENT_FLAG;
 	*/
 
@@ -305,17 +305,17 @@ int psb_irq_postinstall(struct drm_device *dev)
 	PSB_WVDC32(dev_priv->vdc_irq_mask, PSB_INT_ENABLE_R);
 	PSB_WVDC32(0xFFFFFFFF, PSB_HWSTAM);
 
-	if (dev->vblank_enabled[0])
+	if (dev->vblank[0].enabled)
 		psb_enable_pipestat(dev_priv, 0, PIPE_VBLANK_INTERRUPT_ENABLE);
 	else
 		psb_disable_pipestat(dev_priv, 0, PIPE_VBLANK_INTERRUPT_ENABLE);
 
-	if (dev->vblank_enabled[1])
+	if (dev->vblank[1].enabled)
 		psb_enable_pipestat(dev_priv, 1, PIPE_VBLANK_INTERRUPT_ENABLE);
 	else
 		psb_disable_pipestat(dev_priv, 1, PIPE_VBLANK_INTERRUPT_ENABLE);
 
-	if (dev->vblank_enabled[2])
+	if (dev->vblank[2].enabled)
 		psb_enable_pipestat(dev_priv, 2, PIPE_VBLANK_INTERRUPT_ENABLE);
 	else
 		psb_disable_pipestat(dev_priv, 2, PIPE_VBLANK_INTERRUPT_ENABLE);
@@ -339,13 +339,13 @@ void psb_irq_uninstall(struct drm_device *dev)
 
 	PSB_WVDC32(0xFFFFFFFF, PSB_HWSTAM);
 
-	if (dev->vblank_enabled[0])
+	if (dev->vblank[0].enabled)
 		psb_disable_pipestat(dev_priv, 0, PIPE_VBLANK_INTERRUPT_ENABLE);
 
-	if (dev->vblank_enabled[1])
+	if (dev->vblank[1].enabled)
 		psb_disable_pipestat(dev_priv, 1, PIPE_VBLANK_INTERRUPT_ENABLE);
 
-	if (dev->vblank_enabled[2])
+	if (dev->vblank[2].enabled)
 		psb_disable_pipestat(dev_priv, 2, PIPE_VBLANK_INTERRUPT_ENABLE);
 
 	dev_priv->vdc_irq_mask &= _PSB_IRQ_SGX_FLAG |
@@ -456,7 +456,7 @@ static int psb_vblank_do_wait(struct drm_device *dev,
 {
 	unsigned int cur_vblank;
 	int ret = 0;
-	DRM_WAIT_ON(ret, dev->vbl_queue, 3 * DRM_HZ,
+	DRM_WAIT_ON(ret, dev->vblank.queue, 3 * DRM_HZ,
 		    (((cur_vblank = atomic_read(counter))
 		      - *sequence) <= (1 << 23)));
 	*sequence = cur_vblank;
diff --git a/drivers/gpu/drm/i2c/tda998x_drv.c b/drivers/gpu/drm/i2c/tda998x_drv.c
index 60e84043aa34..400b0c4a10fb 100644
--- a/drivers/gpu/drm/i2c/tda998x_drv.c
+++ b/drivers/gpu/drm/i2c/tda998x_drv.c
@@ -17,6 +17,7 @@
 
 
 
+#include <linux/hdmi.h>
 #include <linux/module.h>
 
 #include <drm/drmP.h>
@@ -549,6 +550,8 @@ tda998x_write_avi(struct drm_encoder *encoder, struct drm_display_mode *mode)
 	buf[HB(0)] = 0x82;
 	buf[HB(1)] = 0x02;
 	buf[HB(2)] = 13;
+	buf[PB(1)] = HDMI_SCAN_MODE_UNDERSCAN;
+	buf[PB(3)] = HDMI_QUANTIZATION_RANGE_FULL << 2;
 	buf[PB(4)] = drm_match_cea_mode(mode);
 
 	tda998x_write_if(encoder, DIP_IF_FLAGS_IF2, REG_IF2_HB0, buf,
diff --git a/drivers/gpu/drm/i810/i810_dma.c b/drivers/gpu/drm/i810/i810_dma.c
index ab1892eb1074..249fdff305c6 100644
--- a/drivers/gpu/drm/i810/i810_dma.c
+++ b/drivers/gpu/drm/i810/i810_dma.c
@@ -944,8 +944,6 @@ static int i810_dma_vertex(struct drm_device *dev, void *data,
 				 dma->buflist[vertex->idx],
 				 vertex->discard, vertex->used);
 
-	atomic_add(vertex->used, &dev->counts[_DRM_STAT_SECONDARY]);
-	atomic_inc(&dev->counts[_DRM_STAT_DMA]);
 	sarea_priv->last_enqueue = dev_priv->counter - 1;
 	sarea_priv->last_dispatch = (int)hw_status[5];
 
@@ -1105,8 +1103,6 @@ static int i810_dma_mc(struct drm_device *dev, void *data,
 	i810_dma_dispatch_mc(dev, dma->buflist[mc->idx], mc->used,
 			     mc->last_render);
 
-	atomic_add(mc->used, &dev->counts[_DRM_STAT_SECONDARY]);
-	atomic_inc(&dev->counts[_DRM_STAT_DMA]);
 	sarea_priv->last_enqueue = dev_priv->counter - 1;
 	sarea_priv->last_dispatch = (int)hw_status[5];
 
@@ -1197,13 +1193,6 @@ static int i810_flip_bufs(struct drm_device *dev, void *data,
 
 int i810_driver_load(struct drm_device *dev, unsigned long flags)
 {
-	/* i810 has 4 more counters */
-	dev->counters += 4;
-	dev->types[6] = _DRM_STAT_IRQ;
-	dev->types[7] = _DRM_STAT_PRIMARY;
-	dev->types[8] = _DRM_STAT_SECONDARY;
-	dev->types[9] = _DRM_STAT_DMA;
-
 	pci_set_master(dev->pdev);
 
 	return 0;
diff --git a/drivers/gpu/drm/i915/Kconfig b/drivers/gpu/drm/i915/Kconfig
new file mode 100644
index 000000000000..6199d0b5b958
--- /dev/null
+++ b/drivers/gpu/drm/i915/Kconfig
@@ -0,0 +1,67 @@
+config DRM_I915
+	tristate "Intel 8xx/9xx/G3x/G4x/HD Graphics"
+	depends on DRM
+	depends on AGP
+	depends on AGP_INTEL
+	# we need shmfs for the swappable backing store, and in particular
+	# the shmem_readpage() which depends upon tmpfs
+	select SHMEM
+	select TMPFS
+	select DRM_KMS_HELPER
+	# i915 depends on ACPI_VIDEO when ACPI is enabled
+	# but for select to work, need to select ACPI_VIDEO's dependencies, ick
+	select BACKLIGHT_LCD_SUPPORT if ACPI
+	select BACKLIGHT_CLASS_DEVICE if ACPI
+	select VIDEO_OUTPUT_CONTROL if ACPI
+	select INPUT if ACPI
+	select ACPI_VIDEO if ACPI
+	select ACPI_BUTTON if ACPI
+	help
+	  Choose this option if you have a system that has "Intel Graphics
+	  Media Accelerator" or "HD Graphics" integrated graphics,
+	  including 830M, 845G, 852GM, 855GM, 865G, 915G, 945G, 965G,
+	  G35, G41, G43, G45 chipsets and Celeron, Pentium, Core i3,
+	  Core i5, Core i7 as well as Atom CPUs with integrated graphics.
+	  If M is selected, the module will be called i915.  AGP support
+	  is required for this driver to work. This driver is used by
+	  the Intel driver in X.org 6.8 and XFree86 4.4 and above. It
+	  replaces the older i830 module that supported a subset of the
+	  hardware in older X.org releases.
+
+	  Note that the older i810/i815 chipsets require the use of the
+	  i810 driver instead, and the Atom z5xx series has an entirely
+	  different implementation.
+
+config DRM_I915_KMS
+	bool "Enable modesetting on intel by default"
+	depends on DRM_I915
+	help
+	  Choose this option if you want kernel modesetting enabled by default,
+	  and you have a new enough userspace to support this. Running old
+	  userspaces with this enabled will cause pain.  Note that this causes
+	  the driver to bind to PCI devices, which precludes loading things
+	  like intelfb.
+
+config DRM_I915_FBDEV
+	bool "Enable legacy fbdev support for the modesettting intel driver"
+	depends on DRM_I915
+	select DRM_KMS_FB_HELPER
+	select FB_CFB_FILLRECT
+	select FB_CFB_COPYAREA
+	select FB_CFB_IMAGEBLIT
+	default y
+	help
+	  Choose this option if you have a need for the legacy fbdev
+	  support. Note that this support also provide the linux console
+	  support on top of the intel modesetting driver.
+
+config DRM_I915_PRELIMINARY_HW_SUPPORT
+	bool "Enable preliminary support for prerelease Intel hardware by default"
+	depends on DRM_I915
+	help
+	  Choose this option if you have prerelease Intel hardware and want the
+	  i915 driver to support it by default.  You can enable such support at
+	  runtime with the module option i915.preliminary_hw_support=1; this
+	  option changes the default for that module option.
+
+	  If in doubt, say "N".
diff --git a/drivers/gpu/drm/i915/Makefile b/drivers/gpu/drm/i915/Makefile
index b8449a84a0dc..41838eaa799c 100644
--- a/drivers/gpu/drm/i915/Makefile
+++ b/drivers/gpu/drm/i915/Makefile
@@ -21,6 +21,9 @@ i915-y := i915_drv.o i915_dma.o i915_irq.o \
 	  intel_display.o \
 	  intel_crt.o \
 	  intel_lvds.o \
+	  intel_dsi.o \
+	  intel_dsi_cmd.o \
+	  intel_dsi_pll.o \
 	  intel_bios.o \
 	  intel_ddi.o \
 	  intel_dp.o \
@@ -30,7 +33,6 @@ i915-y := i915_drv.o i915_dma.o i915_irq.o \
 	  intel_panel.o \
 	  intel_pm.o \
 	  intel_i2c.o \
-	  intel_fb.o \
 	  intel_tv.o \
 	  intel_dvo.o \
 	  intel_ringbuffer.o \
@@ -51,6 +53,8 @@ i915-$(CONFIG_COMPAT)   += i915_ioc32.o
 
 i915-$(CONFIG_ACPI)	+= intel_acpi.o
 
+i915-$(CONFIG_DRM_I915_FBDEV) += intel_fbdev.o
+
 obj-$(CONFIG_DRM_I915)  += i915.o
 
 CFLAGS_i915_trace_points.o := -I$(src)
diff --git a/drivers/gpu/drm/i915/dvo.h b/drivers/gpu/drm/i915/dvo.h
index 33a62ad80100..312163379db9 100644
--- a/drivers/gpu/drm/i915/dvo.h
+++ b/drivers/gpu/drm/i915/dvo.h
@@ -77,17 +77,6 @@ struct intel_dvo_dev_ops {
 			  struct drm_display_mode *mode);
 
 	/*
-	 * Callback to adjust the mode to be set in the CRTC.
-	 *
-	 * This allows an output to adjust the clock or even the entire set of
-	 * timings, which is used for panels with fixed timings or for
-	 * buses with clock limitations.
-	 */
-	bool (*mode_fixup)(struct intel_dvo_device *dvo,
-			   const struct drm_display_mode *mode,
-			   struct drm_display_mode *adjusted_mode);
-
-	/*
 	 * Callback for preparing mode changes on an output
 	 */
 	void (*prepare)(struct intel_dvo_device *dvo);
diff --git a/drivers/gpu/drm/i915/i915_debugfs.c b/drivers/gpu/drm/i915/i915_debugfs.c
index a6f4cb5af185..6ed45a984230 100644
--- a/drivers/gpu/drm/i915/i915_debugfs.c
+++ b/drivers/gpu/drm/i915/i915_debugfs.c
@@ -27,6 +27,8 @@
  */
 
 #include <linux/seq_file.h>
+#include <linux/circ_buf.h>
+#include <linux/ctype.h>
 #include <linux/debugfs.h>
 #include <linux/slab.h>
 #include <linux/export.h>
@@ -38,9 +40,6 @@
 #include <drm/i915_drm.h>
 #include "i915_drv.h"
 
-#define DRM_I915_RING_DEBUG 1
-
-
 #if defined(CONFIG_DEBUG_FS)
 
 enum {
@@ -54,6 +53,32 @@ static const char *yesno(int v)
 	return v ? "yes" : "no";
 }
 
+/* As the drm_debugfs_init() routines are called before dev->dev_private is
+ * allocated we need to hook into the minor for release. */
+static int
+drm_add_fake_info_node(struct drm_minor *minor,
+		       struct dentry *ent,
+		       const void *key)
+{
+	struct drm_info_node *node;
+
+	node = kmalloc(sizeof(*node), GFP_KERNEL);
+	if (node == NULL) {
+		debugfs_remove(ent);
+		return -ENOMEM;
+	}
+
+	node->minor = minor;
+	node->dent = ent;
+	node->info_ent = (void *) key;
+
+	mutex_lock(&minor->debugfs_lock);
+	list_add(&node->list, &minor->debugfs_list);
+	mutex_unlock(&minor->debugfs_lock);
+
+	return 0;
+}
+
 static int i915_capabilities(struct seq_file *m, void *data)
 {
 	struct drm_info_node *node = (struct drm_info_node *) m->private;
@@ -145,6 +170,13 @@ describe_obj(struct seq_file *m, struct drm_i915_gem_object *obj)
 		seq_printf(m, " (%s)", obj->ring->name);
 }
 
+static void describe_ctx(struct seq_file *m, struct i915_hw_context *ctx)
+{
+	seq_putc(m, ctx->is_initialized ? 'I' : 'i');
+	seq_putc(m, ctx->remap_slice ? 'R' : 'r');
+	seq_putc(m, ' ');
+}
+
 static int i915_gem_object_list_info(struct seq_file *m, void *data)
 {
 	struct drm_info_node *node = (struct drm_info_node *) m->private;
@@ -554,7 +586,53 @@ static int i915_interrupt_info(struct seq_file *m, void *data)
 	if (ret)
 		return ret;
 
-	if (IS_VALLEYVIEW(dev)) {
+	if (INTEL_INFO(dev)->gen >= 8) {
+		int i;
+		seq_printf(m, "Master Interrupt Control:\t%08x\n",
+			   I915_READ(GEN8_MASTER_IRQ));
+
+		for (i = 0; i < 4; i++) {
+			seq_printf(m, "GT Interrupt IMR %d:\t%08x\n",
+				   i, I915_READ(GEN8_GT_IMR(i)));
+			seq_printf(m, "GT Interrupt IIR %d:\t%08x\n",
+				   i, I915_READ(GEN8_GT_IIR(i)));
+			seq_printf(m, "GT Interrupt IER %d:\t%08x\n",
+				   i, I915_READ(GEN8_GT_IER(i)));
+		}
+
+		for_each_pipe(i) {
+			seq_printf(m, "Pipe %c IMR:\t%08x\n",
+				   pipe_name(i),
+				   I915_READ(GEN8_DE_PIPE_IMR(i)));
+			seq_printf(m, "Pipe %c IIR:\t%08x\n",
+				   pipe_name(i),
+				   I915_READ(GEN8_DE_PIPE_IIR(i)));
+			seq_printf(m, "Pipe %c IER:\t%08x\n",
+				   pipe_name(i),
+				   I915_READ(GEN8_DE_PIPE_IER(i)));
+		}
+
+		seq_printf(m, "Display Engine port interrupt mask:\t%08x\n",
+			   I915_READ(GEN8_DE_PORT_IMR));
+		seq_printf(m, "Display Engine port interrupt identity:\t%08x\n",
+			   I915_READ(GEN8_DE_PORT_IIR));
+		seq_printf(m, "Display Engine port interrupt enable:\t%08x\n",
+			   I915_READ(GEN8_DE_PORT_IER));
+
+		seq_printf(m, "Display Engine misc interrupt mask:\t%08x\n",
+			   I915_READ(GEN8_DE_MISC_IMR));
+		seq_printf(m, "Display Engine misc interrupt identity:\t%08x\n",
+			   I915_READ(GEN8_DE_MISC_IIR));
+		seq_printf(m, "Display Engine misc interrupt enable:\t%08x\n",
+			   I915_READ(GEN8_DE_MISC_IER));
+
+		seq_printf(m, "PCU interrupt mask:\t%08x\n",
+			   I915_READ(GEN8_PCU_IMR));
+		seq_printf(m, "PCU interrupt identity:\t%08x\n",
+			   I915_READ(GEN8_PCU_IIR));
+		seq_printf(m, "PCU interrupt enable:\t%08x\n",
+			   I915_READ(GEN8_PCU_IER));
+	} else if (IS_VALLEYVIEW(dev)) {
 		seq_printf(m, "Display IER:\t%08x\n",
 			   I915_READ(VLV_IER));
 		seq_printf(m, "Display IIR:\t%08x\n",
@@ -626,7 +704,7 @@ static int i915_interrupt_info(struct seq_file *m, void *data)
 	seq_printf(m, "Interrupts received: %d\n",
 		   atomic_read(&dev_priv->irq_received));
 	for_each_ring(ring, dev_priv, i) {
-		if (IS_GEN6(dev) || IS_GEN7(dev)) {
+		if (INTEL_INFO(dev)->gen >= 6) {
 			seq_printf(m,
 				   "Graphics Interrupt mask (%s):	%08x\n",
 				   ring->name, I915_READ_IMR(ring));
@@ -843,6 +921,8 @@ static int i915_cur_delayinfo(struct seq_file *m, void *unused)
 	drm_i915_private_t *dev_priv = dev->dev_private;
 	int ret;
 
+	flush_delayed_work(&dev_priv->rps.delayed_resume_work);
+
 	if (IS_GEN5(dev)) {
 		u16 rgvswctl = I915_READ16(MEMSWCTL);
 		u16 rgvstat = I915_READ16(MEMSTAT_ILK);
@@ -1321,6 +1401,8 @@ static int i915_ring_freq_table(struct seq_file *m, void *unused)
 		return 0;
 	}
 
+	flush_delayed_work(&dev_priv->rps.delayed_resume_work);
+
 	ret = mutex_lock_interruptible(&dev_priv->rps.hw_lock);
 	if (ret)
 		return ret;
@@ -1395,12 +1477,12 @@ static int i915_gem_framebuffer_info(struct seq_file *m, void *data)
 {
 	struct drm_info_node *node = (struct drm_info_node *) m->private;
 	struct drm_device *dev = node->minor->dev;
-	drm_i915_private_t *dev_priv = dev->dev_private;
-	struct intel_fbdev *ifbdev;
+	struct intel_fbdev *ifbdev = NULL;
 	struct intel_framebuffer *fb;
-	int ret;
 
-	ret = mutex_lock_interruptible(&dev->mode_config.mutex);
+#ifdef CONFIG_DRM_I915_FBDEV
+	struct drm_i915_private *dev_priv = dev->dev_private;
+	int ret = mutex_lock_interruptible(&dev->mode_config.mutex);
 	if (ret)
 		return ret;
 
@@ -1416,10 +1498,11 @@ static int i915_gem_framebuffer_info(struct seq_file *m, void *data)
 	describe_obj(m, fb->obj);
 	seq_putc(m, '\n');
 	mutex_unlock(&dev->mode_config.mutex);
+#endif
 
 	mutex_lock(&dev->mode_config.fb_lock);
 	list_for_each_entry(fb, &dev->mode_config.fb_list, base.head) {
-		if (&fb->base == ifbdev->helper.fb)
+		if (ifbdev && &fb->base == ifbdev->helper.fb)
 			continue;
 
 		seq_printf(m, "user size: %d x %d, depth %d, %d bpp, refcount %d, obj ",
@@ -1442,6 +1525,7 @@ static int i915_context_status(struct seq_file *m, void *unused)
 	struct drm_device *dev = node->minor->dev;
 	drm_i915_private_t *dev_priv = dev->dev_private;
 	struct intel_ring_buffer *ring;
+	struct i915_hw_context *ctx;
 	int ret, i;
 
 	ret = mutex_lock_interruptible(&dev->mode_config.mutex);
@@ -1460,12 +1544,15 @@ static int i915_context_status(struct seq_file *m, void *unused)
 		seq_putc(m, '\n');
 	}
 
-	for_each_ring(ring, dev_priv, i) {
-		if (ring->default_context) {
-			seq_printf(m, "HW default context %s ring ", ring->name);
-			describe_obj(m, ring->default_context->obj);
-			seq_putc(m, '\n');
-		}
+	list_for_each_entry(ctx, &dev_priv->context_list, link) {
+		seq_puts(m, "HW context ");
+		describe_ctx(m, ctx);
+		for_each_ring(ring, dev_priv, i)
+			if (ring->default_context == ctx)
+				seq_printf(m, "(default context %s) ", ring->name);
+
+		describe_obj(m, ctx->obj);
+		seq_putc(m, '\n');
 	}
 
 	mutex_unlock(&dev->mode_config.mutex);
@@ -1536,7 +1623,7 @@ static int i915_swizzle_info(struct seq_file *m, void *data)
 			   I915_READ16(C0DRB3));
 		seq_printf(m, "C1DRB3 = 0x%04x\n",
 			   I915_READ16(C1DRB3));
-	} else if (IS_GEN6(dev) || IS_GEN7(dev)) {
+	} else if (INTEL_INFO(dev)->gen >= 6) {
 		seq_printf(m, "MAD_DIMM_C0 = 0x%08x\n",
 			   I915_READ(MAD_DIMM_C0));
 		seq_printf(m, "MAD_DIMM_C1 = 0x%08x\n",
@@ -1545,8 +1632,12 @@ static int i915_swizzle_info(struct seq_file *m, void *data)
 			   I915_READ(MAD_DIMM_C2));
 		seq_printf(m, "TILECTL = 0x%08x\n",
 			   I915_READ(TILECTL));
-		seq_printf(m, "ARB_MODE = 0x%08x\n",
-			   I915_READ(ARB_MODE));
+		if (IS_GEN8(dev))
+			seq_printf(m, "GAMTARBMODE = 0x%08x\n",
+				   I915_READ(GAMTARBMODE));
+		else
+			seq_printf(m, "ARB_MODE = 0x%08x\n",
+				   I915_READ(ARB_MODE));
 		seq_printf(m, "DISP_ARB_CTL = 0x%08x\n",
 			   I915_READ(DISP_ARB_CTL));
 	}
@@ -1555,18 +1646,37 @@ static int i915_swizzle_info(struct seq_file *m, void *data)
 	return 0;
 }
 
-static int i915_ppgtt_info(struct seq_file *m, void *data)
+static void gen8_ppgtt_info(struct seq_file *m, struct drm_device *dev)
 {
-	struct drm_info_node *node = (struct drm_info_node *) m->private;
-	struct drm_device *dev = node->minor->dev;
 	struct drm_i915_private *dev_priv = dev->dev_private;
 	struct intel_ring_buffer *ring;
-	int i, ret;
+	struct i915_hw_ppgtt *ppgtt = dev_priv->mm.aliasing_ppgtt;
+	int unused, i;
 
+	if (!ppgtt)
+		return;
+
+	seq_printf(m, "Page directories: %d\n", ppgtt->num_pd_pages);
+	seq_printf(m, "Page tables: %d\n", ppgtt->num_pt_pages);
+	for_each_ring(ring, dev_priv, unused) {
+		seq_printf(m, "%s\n", ring->name);
+		for (i = 0; i < 4; i++) {
+			u32 offset = 0x270 + i * 8;
+			u64 pdp = I915_READ(ring->mmio_base + offset + 4);
+			pdp <<= 32;
+			pdp |= I915_READ(ring->mmio_base + offset);
+			for (i = 0; i < 4; i++)
+				seq_printf(m, "\tPDP%d 0x%016llx\n", i, pdp);
+		}
+	}
+}
+
+static void gen6_ppgtt_info(struct seq_file *m, struct drm_device *dev)
+{
+	struct drm_i915_private *dev_priv = dev->dev_private;
+	struct intel_ring_buffer *ring;
+	int i;
 
-	ret = mutex_lock_interruptible(&dev->struct_mutex);
-	if (ret)
-		return ret;
 	if (INTEL_INFO(dev)->gen == 6)
 		seq_printf(m, "GFX_MODE: 0x%08x\n", I915_READ(GFX_MODE));
 
@@ -1585,6 +1695,22 @@ static int i915_ppgtt_info(struct seq_file *m, void *data)
 		seq_printf(m, "pd gtt offset: 0x%08x\n", ppgtt->pd_offset);
 	}
 	seq_printf(m, "ECOCHK: 0x%08x\n", I915_READ(GAM_ECOCHK));
+}
+
+static int i915_ppgtt_info(struct seq_file *m, void *data)
+{
+	struct drm_info_node *node = (struct drm_info_node *) m->private;
+	struct drm_device *dev = node->minor->dev;
+
+	int ret = mutex_lock_interruptible(&dev->struct_mutex);
+	if (ret)
+		return ret;
+
+	if (INTEL_INFO(dev)->gen >= 8)
+		gen8_ppgtt_info(m, dev);
+	else if (INTEL_INFO(dev)->gen >= 6)
+		gen6_ppgtt_info(m, dev);
+
 	mutex_unlock(&dev->struct_mutex);
 
 	return 0;
@@ -1610,27 +1736,27 @@ static int i915_dpio_info(struct seq_file *m, void *data)
 	seq_printf(m, "DPIO_CTL: 0x%08x\n", I915_READ(DPIO_CTL));
 
 	seq_printf(m, "DPIO_DIV_A: 0x%08x\n",
-		   vlv_dpio_read(dev_priv, _DPIO_DIV_A));
+		   vlv_dpio_read(dev_priv, PIPE_A, _DPIO_DIV_A));
 	seq_printf(m, "DPIO_DIV_B: 0x%08x\n",
-		   vlv_dpio_read(dev_priv, _DPIO_DIV_B));
+		   vlv_dpio_read(dev_priv, PIPE_A, _DPIO_DIV_B));
 
 	seq_printf(m, "DPIO_REFSFR_A: 0x%08x\n",
-		   vlv_dpio_read(dev_priv, _DPIO_REFSFR_A));
+		   vlv_dpio_read(dev_priv, PIPE_A, _DPIO_REFSFR_A));
 	seq_printf(m, "DPIO_REFSFR_B: 0x%08x\n",
-		   vlv_dpio_read(dev_priv, _DPIO_REFSFR_B));
+		   vlv_dpio_read(dev_priv, PIPE_A, _DPIO_REFSFR_B));
 
 	seq_printf(m, "DPIO_CORE_CLK_A: 0x%08x\n",
-		   vlv_dpio_read(dev_priv, _DPIO_CORE_CLK_A));
+		   vlv_dpio_read(dev_priv, PIPE_A, _DPIO_CORE_CLK_A));
 	seq_printf(m, "DPIO_CORE_CLK_B: 0x%08x\n",
-		   vlv_dpio_read(dev_priv, _DPIO_CORE_CLK_B));
+		   vlv_dpio_read(dev_priv, PIPE_A, _DPIO_CORE_CLK_B));
 
 	seq_printf(m, "DPIO_LPF_COEFF_A: 0x%08x\n",
-		   vlv_dpio_read(dev_priv, _DPIO_LPF_COEFF_A));
+		   vlv_dpio_read(dev_priv, PIPE_A, _DPIO_LPF_COEFF_A));
 	seq_printf(m, "DPIO_LPF_COEFF_B: 0x%08x\n",
-		   vlv_dpio_read(dev_priv, _DPIO_LPF_COEFF_B));
+		   vlv_dpio_read(dev_priv, PIPE_A, _DPIO_LPF_COEFF_B));
 
 	seq_printf(m, "DPIO_FASTCLK_DISABLE: 0x%08x\n",
-		   vlv_dpio_read(dev_priv, DPIO_FASTCLK_DISABLE));
+		   vlv_dpio_read(dev_priv, PIPE_A, DPIO_FASTCLK_DISABLE));
 
 	mutex_unlock(&dev_priv->dpio_lock);
 
@@ -1655,126 +1781,20 @@ static int i915_edp_psr_status(struct seq_file *m, void *data)
 	struct drm_info_node *node = m->private;
 	struct drm_device *dev = node->minor->dev;
 	struct drm_i915_private *dev_priv = dev->dev_private;
-	u32 psrstat, psrperf;
-
-	if (!IS_HASWELL(dev)) {
-		seq_puts(m, "PSR not supported on this platform\n");
-	} else if (IS_HASWELL(dev) && I915_READ(EDP_PSR_CTL) & EDP_PSR_ENABLE) {
-		seq_puts(m, "PSR enabled\n");
-	} else {
-		seq_puts(m, "PSR disabled: ");
-		switch (dev_priv->no_psr_reason) {
-		case PSR_NO_SOURCE:
-			seq_puts(m, "not supported on this platform");
-			break;
-		case PSR_NO_SINK:
-			seq_puts(m, "not supported by panel");
-			break;
-		case PSR_MODULE_PARAM:
-			seq_puts(m, "disabled by flag");
-			break;
-		case PSR_CRTC_NOT_ACTIVE:
-			seq_puts(m, "crtc not active");
-			break;
-		case PSR_PWR_WELL_ENABLED:
-			seq_puts(m, "power well enabled");
-			break;
-		case PSR_NOT_TILED:
-			seq_puts(m, "not tiled");
-			break;
-		case PSR_SPRITE_ENABLED:
-			seq_puts(m, "sprite enabled");
-			break;
-		case PSR_S3D_ENABLED:
-			seq_puts(m, "stereo 3d enabled");
-			break;
-		case PSR_INTERLACED_ENABLED:
-			seq_puts(m, "interlaced enabled");
-			break;
-		case PSR_HSW_NOT_DDIA:
-			seq_puts(m, "HSW ties PSR to DDI A (eDP)");
-			break;
-		default:
-			seq_puts(m, "unknown reason");
-		}
-		seq_puts(m, "\n");
-		return 0;
-	}
-
-	psrstat = I915_READ(EDP_PSR_STATUS_CTL);
-
-	seq_puts(m, "PSR Current State: ");
-	switch (psrstat & EDP_PSR_STATUS_STATE_MASK) {
-	case EDP_PSR_STATUS_STATE_IDLE:
-		seq_puts(m, "Reset state\n");
-		break;
-	case EDP_PSR_STATUS_STATE_SRDONACK:
-		seq_puts(m, "Wait for TG/Stream to send on frame of data after SRD conditions are met\n");
-		break;
-	case EDP_PSR_STATUS_STATE_SRDENT:
-		seq_puts(m, "SRD entry\n");
-		break;
-	case EDP_PSR_STATUS_STATE_BUFOFF:
-		seq_puts(m, "Wait for buffer turn off\n");
-		break;
-	case EDP_PSR_STATUS_STATE_BUFON:
-		seq_puts(m, "Wait for buffer turn on\n");
-		break;
-	case EDP_PSR_STATUS_STATE_AUXACK:
-		seq_puts(m, "Wait for AUX to acknowledge on SRD exit\n");
-		break;
-	case EDP_PSR_STATUS_STATE_SRDOFFACK:
-		seq_puts(m, "Wait for TG/Stream to acknowledge the SRD VDM exit\n");
-		break;
-	default:
-		seq_puts(m, "Unknown\n");
-		break;
-	}
-
-	seq_puts(m, "Link Status: ");
-	switch (psrstat & EDP_PSR_STATUS_LINK_MASK) {
-	case EDP_PSR_STATUS_LINK_FULL_OFF:
-		seq_puts(m, "Link is fully off\n");
-		break;
-	case EDP_PSR_STATUS_LINK_FULL_ON:
-		seq_puts(m, "Link is fully on\n");
-		break;
-	case EDP_PSR_STATUS_LINK_STANDBY:
-		seq_puts(m, "Link is in standby\n");
-		break;
-	default:
-		seq_puts(m, "Unknown\n");
-		break;
-	}
-
-	seq_printf(m, "PSR Entry Count: %u\n",
-		   psrstat >> EDP_PSR_STATUS_COUNT_SHIFT &
-		   EDP_PSR_STATUS_COUNT_MASK);
-
-	seq_printf(m, "Max Sleep Timer Counter: %u\n",
-		   psrstat >> EDP_PSR_STATUS_MAX_SLEEP_TIMER_SHIFT &
-		   EDP_PSR_STATUS_MAX_SLEEP_TIMER_MASK);
+	u32 psrperf = 0;
+	bool enabled = false;
 
-	seq_printf(m, "Had AUX error: %s\n",
-		   yesno(psrstat & EDP_PSR_STATUS_AUX_ERROR));
+	seq_printf(m, "Sink_Support: %s\n", yesno(dev_priv->psr.sink_support));
+	seq_printf(m, "Source_OK: %s\n", yesno(dev_priv->psr.source_ok));
 
-	seq_printf(m, "Sending AUX: %s\n",
-		   yesno(psrstat & EDP_PSR_STATUS_AUX_SENDING));
+	enabled = HAS_PSR(dev) &&
+		I915_READ(EDP_PSR_CTL(dev)) & EDP_PSR_ENABLE;
+	seq_printf(m, "Enabled: %s\n", yesno(enabled));
 
-	seq_printf(m, "Sending Idle: %s\n",
-		   yesno(psrstat & EDP_PSR_STATUS_SENDING_IDLE));
-
-	seq_printf(m, "Sending TP2 TP3: %s\n",
-		   yesno(psrstat & EDP_PSR_STATUS_SENDING_TP2_TP3));
-
-	seq_printf(m, "Sending TP1: %s\n",
-		   yesno(psrstat & EDP_PSR_STATUS_SENDING_TP1));
-
-	seq_printf(m, "Idle Count: %u\n",
-		   psrstat & EDP_PSR_STATUS_IDLE_MASK);
-
-	psrperf = (I915_READ(EDP_PSR_PERF_CNT)) & EDP_PSR_PERF_CNT_MASK;
-	seq_printf(m, "Performance Counter: %u\n", psrperf);
+	if (HAS_PSR(dev))
+		psrperf = I915_READ(EDP_PSR_PERF_CNT(dev)) &
+			EDP_PSR_PERF_CNT_MASK;
+	seq_printf(m, "Performance_Counter: %u\n", psrperf);
 
 	return 0;
 }
@@ -1825,6 +1845,751 @@ static int i915_pc8_status(struct seq_file *m, void *unused)
 	return 0;
 }
 
+struct pipe_crc_info {
+	const char *name;
+	struct drm_device *dev;
+	enum pipe pipe;
+};
+
+static int i915_pipe_crc_open(struct inode *inode, struct file *filep)
+{
+	struct pipe_crc_info *info = inode->i_private;
+	struct drm_i915_private *dev_priv = info->dev->dev_private;
+	struct intel_pipe_crc *pipe_crc = &dev_priv->pipe_crc[info->pipe];
+
+	spin_lock_irq(&pipe_crc->lock);
+
+	if (pipe_crc->opened) {
+		spin_unlock_irq(&pipe_crc->lock);
+		return -EBUSY; /* already open */
+	}
+
+	pipe_crc->opened = true;
+	filep->private_data = inode->i_private;
+
+	spin_unlock_irq(&pipe_crc->lock);
+
+	return 0;
+}
+
+static int i915_pipe_crc_release(struct inode *inode, struct file *filep)
+{
+	struct pipe_crc_info *info = inode->i_private;
+	struct drm_i915_private *dev_priv = info->dev->dev_private;
+	struct intel_pipe_crc *pipe_crc = &dev_priv->pipe_crc[info->pipe];
+
+	spin_lock_irq(&pipe_crc->lock);
+	pipe_crc->opened = false;
+	spin_unlock_irq(&pipe_crc->lock);
+
+	return 0;
+}
+
+/* (6 fields, 8 chars each, space separated (5) + '\n') */
+#define PIPE_CRC_LINE_LEN	(6 * 8 + 5 + 1)
+/* account for \'0' */
+#define PIPE_CRC_BUFFER_LEN	(PIPE_CRC_LINE_LEN + 1)
+
+static int pipe_crc_data_count(struct intel_pipe_crc *pipe_crc)
+{
+	assert_spin_locked(&pipe_crc->lock);
+	return CIRC_CNT(pipe_crc->head, pipe_crc->tail,
+			INTEL_PIPE_CRC_ENTRIES_NR);
+}
+
+static ssize_t
+i915_pipe_crc_read(struct file *filep, char __user *user_buf, size_t count,
+		   loff_t *pos)
+{
+	struct pipe_crc_info *info = filep->private_data;
+	struct drm_device *dev = info->dev;
+	struct drm_i915_private *dev_priv = dev->dev_private;
+	struct intel_pipe_crc *pipe_crc = &dev_priv->pipe_crc[info->pipe];
+	char buf[PIPE_CRC_BUFFER_LEN];
+	int head, tail, n_entries, n;
+	ssize_t bytes_read;
+
+	/*
+	 * Don't allow user space to provide buffers not big enough to hold
+	 * a line of data.
+	 */
+	if (count < PIPE_CRC_LINE_LEN)
+		return -EINVAL;
+
+	if (pipe_crc->source == INTEL_PIPE_CRC_SOURCE_NONE)
+		return 0;
+
+	/* nothing to read */
+	spin_lock_irq(&pipe_crc->lock);
+	while (pipe_crc_data_count(pipe_crc) == 0) {
+		int ret;
+
+		if (filep->f_flags & O_NONBLOCK) {
+			spin_unlock_irq(&pipe_crc->lock);
+			return -EAGAIN;
+		}
+
+		ret = wait_event_interruptible_lock_irq(pipe_crc->wq,
+				pipe_crc_data_count(pipe_crc), pipe_crc->lock);
+		if (ret) {
+			spin_unlock_irq(&pipe_crc->lock);
+			return ret;
+		}
+	}
+
+	/* We now have one or more entries to read */
+	head = pipe_crc->head;
+	tail = pipe_crc->tail;
+	n_entries = min((size_t)CIRC_CNT(head, tail, INTEL_PIPE_CRC_ENTRIES_NR),
+			count / PIPE_CRC_LINE_LEN);
+	spin_unlock_irq(&pipe_crc->lock);
+
+	bytes_read = 0;
+	n = 0;
+	do {
+		struct intel_pipe_crc_entry *entry = &pipe_crc->entries[tail];
+		int ret;
+
+		bytes_read += snprintf(buf, PIPE_CRC_BUFFER_LEN,
+				       "%8u %8x %8x %8x %8x %8x\n",
+				       entry->frame, entry->crc[0],
+				       entry->crc[1], entry->crc[2],
+				       entry->crc[3], entry->crc[4]);
+
+		ret = copy_to_user(user_buf + n * PIPE_CRC_LINE_LEN,
+				   buf, PIPE_CRC_LINE_LEN);
+		if (ret == PIPE_CRC_LINE_LEN)
+			return -EFAULT;
+
+		BUILD_BUG_ON_NOT_POWER_OF_2(INTEL_PIPE_CRC_ENTRIES_NR);
+		tail = (tail + 1) & (INTEL_PIPE_CRC_ENTRIES_NR - 1);
+		n++;
+	} while (--n_entries);
+
+	spin_lock_irq(&pipe_crc->lock);
+	pipe_crc->tail = tail;
+	spin_unlock_irq(&pipe_crc->lock);
+
+	return bytes_read;
+}
+
+static const struct file_operations i915_pipe_crc_fops = {
+	.owner = THIS_MODULE,
+	.open = i915_pipe_crc_open,
+	.read = i915_pipe_crc_read,
+	.release = i915_pipe_crc_release,
+};
+
+static struct pipe_crc_info i915_pipe_crc_data[I915_MAX_PIPES] = {
+	{
+		.name = "i915_pipe_A_crc",
+		.pipe = PIPE_A,
+	},
+	{
+		.name = "i915_pipe_B_crc",
+		.pipe = PIPE_B,
+	},
+	{
+		.name = "i915_pipe_C_crc",
+		.pipe = PIPE_C,
+	},
+};
+
+static int i915_pipe_crc_create(struct dentry *root, struct drm_minor *minor,
+				enum pipe pipe)
+{
+	struct drm_device *dev = minor->dev;
+	struct dentry *ent;
+	struct pipe_crc_info *info = &i915_pipe_crc_data[pipe];
+
+	info->dev = dev;
+	ent = debugfs_create_file(info->name, S_IRUGO, root, info,
+				  &i915_pipe_crc_fops);
+	if (IS_ERR(ent))
+		return PTR_ERR(ent);
+
+	return drm_add_fake_info_node(minor, ent, info);
+}
+
+static const char * const pipe_crc_sources[] = {
+	"none",
+	"plane1",
+	"plane2",
+	"pf",
+	"pipe",
+	"TV",
+	"DP-B",
+	"DP-C",
+	"DP-D",
+	"auto",
+};
+
+static const char *pipe_crc_source_name(enum intel_pipe_crc_source source)
+{
+	BUILD_BUG_ON(ARRAY_SIZE(pipe_crc_sources) != INTEL_PIPE_CRC_SOURCE_MAX);
+	return pipe_crc_sources[source];
+}
+
+static int display_crc_ctl_show(struct seq_file *m, void *data)
+{
+	struct drm_device *dev = m->private;
+	struct drm_i915_private *dev_priv = dev->dev_private;
+	int i;
+
+	for (i = 0; i < I915_MAX_PIPES; i++)
+		seq_printf(m, "%c %s\n", pipe_name(i),
+			   pipe_crc_source_name(dev_priv->pipe_crc[i].source));
+
+	return 0;
+}
+
+static int display_crc_ctl_open(struct inode *inode, struct file *file)
+{
+	struct drm_device *dev = inode->i_private;
+
+	return single_open(file, display_crc_ctl_show, dev);
+}
+
+static int i8xx_pipe_crc_ctl_reg(enum intel_pipe_crc_source *source,
+				 uint32_t *val)
+{
+	if (*source == INTEL_PIPE_CRC_SOURCE_AUTO)
+		*source = INTEL_PIPE_CRC_SOURCE_PIPE;
+
+	switch (*source) {
+	case INTEL_PIPE_CRC_SOURCE_PIPE:
+		*val = PIPE_CRC_ENABLE | PIPE_CRC_INCLUDE_BORDER_I8XX;
+		break;
+	case INTEL_PIPE_CRC_SOURCE_NONE:
+		*val = 0;
+		break;
+	default:
+		return -EINVAL;
+	}
+
+	return 0;
+}
+
+static int i9xx_pipe_crc_auto_source(struct drm_device *dev, enum pipe pipe,
+				     enum intel_pipe_crc_source *source)
+{
+	struct intel_encoder *encoder;
+	struct intel_crtc *crtc;
+	struct intel_digital_port *dig_port;
+	int ret = 0;
+
+	*source = INTEL_PIPE_CRC_SOURCE_PIPE;
+
+	mutex_lock(&dev->mode_config.mutex);
+	list_for_each_entry(encoder, &dev->mode_config.encoder_list,
+			    base.head) {
+		if (!encoder->base.crtc)
+			continue;
+
+		crtc = to_intel_crtc(encoder->base.crtc);
+
+		if (crtc->pipe != pipe)
+			continue;
+
+		switch (encoder->type) {
+		case INTEL_OUTPUT_TVOUT:
+			*source = INTEL_PIPE_CRC_SOURCE_TV;
+			break;
+		case INTEL_OUTPUT_DISPLAYPORT:
+		case INTEL_OUTPUT_EDP:
+			dig_port = enc_to_dig_port(&encoder->base);
+			switch (dig_port->port) {
+			case PORT_B:
+				*source = INTEL_PIPE_CRC_SOURCE_DP_B;
+				break;
+			case PORT_C:
+				*source = INTEL_PIPE_CRC_SOURCE_DP_C;
+				break;
+			case PORT_D:
+				*source = INTEL_PIPE_CRC_SOURCE_DP_D;
+				break;
+			default:
+				WARN(1, "nonexisting DP port %c\n",
+				     port_name(dig_port->port));
+				break;
+			}
+			break;
+		}
+	}
+	mutex_unlock(&dev->mode_config.mutex);
+
+	return ret;
+}
+
+static int vlv_pipe_crc_ctl_reg(struct drm_device *dev,
+				enum pipe pipe,
+				enum intel_pipe_crc_source *source,
+				uint32_t *val)
+{
+	struct drm_i915_private *dev_priv = dev->dev_private;
+	bool need_stable_symbols = false;
+
+	if (*source == INTEL_PIPE_CRC_SOURCE_AUTO) {
+		int ret = i9xx_pipe_crc_auto_source(dev, pipe, source);
+		if (ret)
+			return ret;
+	}
+
+	switch (*source) {
+	case INTEL_PIPE_CRC_SOURCE_PIPE:
+		*val = PIPE_CRC_ENABLE | PIPE_CRC_SOURCE_PIPE_VLV;
+		break;
+	case INTEL_PIPE_CRC_SOURCE_DP_B:
+		*val = PIPE_CRC_ENABLE | PIPE_CRC_SOURCE_DP_B_VLV;
+		need_stable_symbols = true;
+		break;
+	case INTEL_PIPE_CRC_SOURCE_DP_C:
+		*val = PIPE_CRC_ENABLE | PIPE_CRC_SOURCE_DP_C_VLV;
+		need_stable_symbols = true;
+		break;
+	case INTEL_PIPE_CRC_SOURCE_NONE:
+		*val = 0;
+		break;
+	default:
+		return -EINVAL;
+	}
+
+	/*
+	 * When the pipe CRC tap point is after the transcoders we need
+	 * to tweak symbol-level features to produce a deterministic series of
+	 * symbols for a given frame. We need to reset those features only once
+	 * a frame (instead of every nth symbol):
+	 *   - DC-balance: used to ensure a better clock recovery from the data
+	 *     link (SDVO)
+	 *   - DisplayPort scrambling: used for EMI reduction
+	 */
+	if (need_stable_symbols) {
+		uint32_t tmp = I915_READ(PORT_DFT2_G4X);
+
+		WARN_ON(!IS_G4X(dev));
+
+		tmp |= DC_BALANCE_RESET_VLV;
+		if (pipe == PIPE_A)
+			tmp |= PIPE_A_SCRAMBLE_RESET;
+		else
+			tmp |= PIPE_B_SCRAMBLE_RESET;
+
+		I915_WRITE(PORT_DFT2_G4X, tmp);
+	}
+
+	return 0;
+}
+
+static int i9xx_pipe_crc_ctl_reg(struct drm_device *dev,
+				 enum pipe pipe,
+				 enum intel_pipe_crc_source *source,
+				 uint32_t *val)
+{
+	struct drm_i915_private *dev_priv = dev->dev_private;
+	bool need_stable_symbols = false;
+
+	if (*source == INTEL_PIPE_CRC_SOURCE_AUTO) {
+		int ret = i9xx_pipe_crc_auto_source(dev, pipe, source);
+		if (ret)
+			return ret;
+	}
+
+	switch (*source) {
+	case INTEL_PIPE_CRC_SOURCE_PIPE:
+		*val = PIPE_CRC_ENABLE | PIPE_CRC_SOURCE_PIPE_I9XX;
+		break;
+	case INTEL_PIPE_CRC_SOURCE_TV:
+		if (!SUPPORTS_TV(dev))
+			return -EINVAL;
+		*val = PIPE_CRC_ENABLE | PIPE_CRC_SOURCE_TV_PRE;
+		break;
+	case INTEL_PIPE_CRC_SOURCE_DP_B:
+		if (!IS_G4X(dev))
+			return -EINVAL;
+		*val = PIPE_CRC_ENABLE | PIPE_CRC_SOURCE_DP_B_G4X;
+		need_stable_symbols = true;
+		break;
+	case INTEL_PIPE_CRC_SOURCE_DP_C:
+		if (!IS_G4X(dev))
+			return -EINVAL;
+		*val = PIPE_CRC_ENABLE | PIPE_CRC_SOURCE_DP_C_G4X;
+		need_stable_symbols = true;
+		break;
+	case INTEL_PIPE_CRC_SOURCE_DP_D:
+		if (!IS_G4X(dev))
+			return -EINVAL;
+		*val = PIPE_CRC_ENABLE | PIPE_CRC_SOURCE_DP_D_G4X;
+		need_stable_symbols = true;
+		break;
+	case INTEL_PIPE_CRC_SOURCE_NONE:
+		*val = 0;
+		break;
+	default:
+		return -EINVAL;
+	}
+
+	/*
+	 * When the pipe CRC tap point is after the transcoders we need
+	 * to tweak symbol-level features to produce a deterministic series of
+	 * symbols for a given frame. We need to reset those features only once
+	 * a frame (instead of every nth symbol):
+	 *   - DC-balance: used to ensure a better clock recovery from the data
+	 *     link (SDVO)
+	 *   - DisplayPort scrambling: used for EMI reduction
+	 */
+	if (need_stable_symbols) {
+		uint32_t tmp = I915_READ(PORT_DFT2_G4X);
+
+		WARN_ON(!IS_G4X(dev));
+
+		I915_WRITE(PORT_DFT_I9XX,
+			   I915_READ(PORT_DFT_I9XX) | DC_BALANCE_RESET);
+
+		if (pipe == PIPE_A)
+			tmp |= PIPE_A_SCRAMBLE_RESET;
+		else
+			tmp |= PIPE_B_SCRAMBLE_RESET;
+
+		I915_WRITE(PORT_DFT2_G4X, tmp);
+	}
+
+	return 0;
+}
+
+static void vlv_undo_pipe_scramble_reset(struct drm_device *dev,
+					 enum pipe pipe)
+{
+	struct drm_i915_private *dev_priv = dev->dev_private;
+	uint32_t tmp = I915_READ(PORT_DFT2_G4X);
+
+	if (pipe == PIPE_A)
+		tmp &= ~PIPE_A_SCRAMBLE_RESET;
+	else
+		tmp &= ~PIPE_B_SCRAMBLE_RESET;
+	if (!(tmp & PIPE_SCRAMBLE_RESET_MASK))
+		tmp &= ~DC_BALANCE_RESET_VLV;
+	I915_WRITE(PORT_DFT2_G4X, tmp);
+
+}
+
+static void g4x_undo_pipe_scramble_reset(struct drm_device *dev,
+					 enum pipe pipe)
+{
+	struct drm_i915_private *dev_priv = dev->dev_private;
+	uint32_t tmp = I915_READ(PORT_DFT2_G4X);
+
+	if (pipe == PIPE_A)
+		tmp &= ~PIPE_A_SCRAMBLE_RESET;
+	else
+		tmp &= ~PIPE_B_SCRAMBLE_RESET;
+	I915_WRITE(PORT_DFT2_G4X, tmp);
+
+	if (!(tmp & PIPE_SCRAMBLE_RESET_MASK)) {
+		I915_WRITE(PORT_DFT_I9XX,
+			   I915_READ(PORT_DFT_I9XX) & ~DC_BALANCE_RESET);
+	}
+}
+
+static int ilk_pipe_crc_ctl_reg(enum intel_pipe_crc_source *source,
+				uint32_t *val)
+{
+	if (*source == INTEL_PIPE_CRC_SOURCE_AUTO)
+		*source = INTEL_PIPE_CRC_SOURCE_PIPE;
+
+	switch (*source) {
+	case INTEL_PIPE_CRC_SOURCE_PLANE1:
+		*val = PIPE_CRC_ENABLE | PIPE_CRC_SOURCE_PRIMARY_ILK;
+		break;
+	case INTEL_PIPE_CRC_SOURCE_PLANE2:
+		*val = PIPE_CRC_ENABLE | PIPE_CRC_SOURCE_SPRITE_ILK;
+		break;
+	case INTEL_PIPE_CRC_SOURCE_PIPE:
+		*val = PIPE_CRC_ENABLE | PIPE_CRC_SOURCE_PIPE_ILK;
+		break;
+	case INTEL_PIPE_CRC_SOURCE_NONE:
+		*val = 0;
+		break;
+	default:
+		return -EINVAL;
+	}
+
+	return 0;
+}
+
+static int ivb_pipe_crc_ctl_reg(enum intel_pipe_crc_source *source,
+				uint32_t *val)
+{
+	if (*source == INTEL_PIPE_CRC_SOURCE_AUTO)
+		*source = INTEL_PIPE_CRC_SOURCE_PF;
+
+	switch (*source) {
+	case INTEL_PIPE_CRC_SOURCE_PLANE1:
+		*val = PIPE_CRC_ENABLE | PIPE_CRC_SOURCE_PRIMARY_IVB;
+		break;
+	case INTEL_PIPE_CRC_SOURCE_PLANE2:
+		*val = PIPE_CRC_ENABLE | PIPE_CRC_SOURCE_SPRITE_IVB;
+		break;
+	case INTEL_PIPE_CRC_SOURCE_PF:
+		*val = PIPE_CRC_ENABLE | PIPE_CRC_SOURCE_PF_IVB;
+		break;
+	case INTEL_PIPE_CRC_SOURCE_NONE:
+		*val = 0;
+		break;
+	default:
+		return -EINVAL;
+	}
+
+	return 0;
+}
+
+static int pipe_crc_set_source(struct drm_device *dev, enum pipe pipe,
+			       enum intel_pipe_crc_source source)
+{
+	struct drm_i915_private *dev_priv = dev->dev_private;
+	struct intel_pipe_crc *pipe_crc = &dev_priv->pipe_crc[pipe];
+	u32 val;
+	int ret;
+
+	if (pipe_crc->source == source)
+		return 0;
+
+	/* forbid changing the source without going back to 'none' */
+	if (pipe_crc->source && source)
+		return -EINVAL;
+
+	if (IS_GEN2(dev))
+		ret = i8xx_pipe_crc_ctl_reg(&source, &val);
+	else if (INTEL_INFO(dev)->gen < 5)
+		ret = i9xx_pipe_crc_ctl_reg(dev, pipe, &source, &val);
+	else if (IS_VALLEYVIEW(dev))
+		ret = vlv_pipe_crc_ctl_reg(dev,pipe, &source, &val);
+	else if (IS_GEN5(dev) || IS_GEN6(dev))
+		ret = ilk_pipe_crc_ctl_reg(&source, &val);
+	else
+		ret = ivb_pipe_crc_ctl_reg(&source, &val);
+
+	if (ret != 0)
+		return ret;
+
+	/* none -> real source transition */
+	if (source) {
+		DRM_DEBUG_DRIVER("collecting CRCs for pipe %c, %s\n",
+				 pipe_name(pipe), pipe_crc_source_name(source));
+
+		pipe_crc->entries = kzalloc(sizeof(*pipe_crc->entries) *
+					    INTEL_PIPE_CRC_ENTRIES_NR,
+					    GFP_KERNEL);
+		if (!pipe_crc->entries)
+			return -ENOMEM;
+
+		spin_lock_irq(&pipe_crc->lock);
+		pipe_crc->head = 0;
+		pipe_crc->tail = 0;
+		spin_unlock_irq(&pipe_crc->lock);
+	}
+
+	pipe_crc->source = source;
+
+	I915_WRITE(PIPE_CRC_CTL(pipe), val);
+	POSTING_READ(PIPE_CRC_CTL(pipe));
+
+	/* real source -> none transition */
+	if (source == INTEL_PIPE_CRC_SOURCE_NONE) {
+		struct intel_pipe_crc_entry *entries;
+
+		DRM_DEBUG_DRIVER("stopping CRCs for pipe %c\n",
+				 pipe_name(pipe));
+
+		intel_wait_for_vblank(dev, pipe);
+
+		spin_lock_irq(&pipe_crc->lock);
+		entries = pipe_crc->entries;
+		pipe_crc->entries = NULL;
+		spin_unlock_irq(&pipe_crc->lock);
+
+		kfree(entries);
+
+		if (IS_G4X(dev))
+			g4x_undo_pipe_scramble_reset(dev, pipe);
+		else if (IS_VALLEYVIEW(dev))
+			vlv_undo_pipe_scramble_reset(dev, pipe);
+	}
+
+	return 0;
+}
+
+/*
+ * Parse pipe CRC command strings:
+ *   command: wsp* object wsp+ name wsp+ source wsp*
+ *   object: 'pipe'
+ *   name: (A | B | C)
+ *   source: (none | plane1 | plane2 | pf)
+ *   wsp: (#0x20 | #0x9 | #0xA)+
+ *
+ * eg.:
+ *  "pipe A plane1"  ->  Start CRC computations on plane1 of pipe A
+ *  "pipe A none"    ->  Stop CRC
+ */
+static int display_crc_ctl_tokenize(char *buf, char *words[], int max_words)
+{
+	int n_words = 0;
+
+	while (*buf) {
+		char *end;
+
+		/* skip leading white space */
+		buf = skip_spaces(buf);
+		if (!*buf)
+			break;	/* end of buffer */
+
+		/* find end of word */
+		for (end = buf; *end && !isspace(*end); end++)
+			;
+
+		if (n_words == max_words) {
+			DRM_DEBUG_DRIVER("too many words, allowed <= %d\n",
+					 max_words);
+			return -EINVAL;	/* ran out of words[] before bytes */
+		}
+
+		if (*end)
+			*end++ = '\0';
+		words[n_words++] = buf;
+		buf = end;
+	}
+
+	return n_words;
+}
+
+enum intel_pipe_crc_object {
+	PIPE_CRC_OBJECT_PIPE,
+};
+
+static const char * const pipe_crc_objects[] = {
+	"pipe",
+};
+
+static int
+display_crc_ctl_parse_object(const char *buf, enum intel_pipe_crc_object *o)
+{
+	int i;
+
+	for (i = 0; i < ARRAY_SIZE(pipe_crc_objects); i++)
+		if (!strcmp(buf, pipe_crc_objects[i])) {
+			*o = i;
+			return 0;
+		    }
+
+	return -EINVAL;
+}
+
+static int display_crc_ctl_parse_pipe(const char *buf, enum pipe *pipe)
+{
+	const char name = buf[0];
+
+	if (name < 'A' || name >= pipe_name(I915_MAX_PIPES))
+		return -EINVAL;
+
+	*pipe = name - 'A';
+
+	return 0;
+}
+
+static int
+display_crc_ctl_parse_source(const char *buf, enum intel_pipe_crc_source *s)
+{
+	int i;
+
+	for (i = 0; i < ARRAY_SIZE(pipe_crc_sources); i++)
+		if (!strcmp(buf, pipe_crc_sources[i])) {
+			*s = i;
+			return 0;
+		    }
+
+	return -EINVAL;
+}
+
+static int display_crc_ctl_parse(struct drm_device *dev, char *buf, size_t len)
+{
+#define N_WORDS 3
+	int n_words;
+	char *words[N_WORDS];
+	enum pipe pipe;
+	enum intel_pipe_crc_object object;
+	enum intel_pipe_crc_source source;
+
+	n_words = display_crc_ctl_tokenize(buf, words, N_WORDS);
+	if (n_words != N_WORDS) {
+		DRM_DEBUG_DRIVER("tokenize failed, a command is %d words\n",
+				 N_WORDS);
+		return -EINVAL;
+	}
+
+	if (display_crc_ctl_parse_object(words[0], &object) < 0) {
+		DRM_DEBUG_DRIVER("unknown object %s\n", words[0]);
+		return -EINVAL;
+	}
+
+	if (display_crc_ctl_parse_pipe(words[1], &pipe) < 0) {
+		DRM_DEBUG_DRIVER("unknown pipe %s\n", words[1]);
+		return -EINVAL;
+	}
+
+	if (display_crc_ctl_parse_source(words[2], &source) < 0) {
+		DRM_DEBUG_DRIVER("unknown source %s\n", words[2]);
+		return -EINVAL;
+	}
+
+	return pipe_crc_set_source(dev, pipe, source);
+}
+
+static ssize_t display_crc_ctl_write(struct file *file, const char __user *ubuf,
+				     size_t len, loff_t *offp)
+{
+	struct seq_file *m = file->private_data;
+	struct drm_device *dev = m->private;
+	char *tmpbuf;
+	int ret;
+
+	if (len == 0)
+		return 0;
+
+	if (len > PAGE_SIZE - 1) {
+		DRM_DEBUG_DRIVER("expected <%lu bytes into pipe crc control\n",
+				 PAGE_SIZE);
+		return -E2BIG;
+	}
+
+	tmpbuf = kmalloc(len + 1, GFP_KERNEL);
+	if (!tmpbuf)
+		return -ENOMEM;
+
+	if (copy_from_user(tmpbuf, ubuf, len)) {
+		ret = -EFAULT;
+		goto out;
+	}
+	tmpbuf[len] = '\0';
+
+	ret = display_crc_ctl_parse(dev, tmpbuf, len);
+
+out:
+	kfree(tmpbuf);
+	if (ret < 0)
+		return ret;
+
+	*offp += len;
+	return len;
+}
+
+static const struct file_operations i915_display_crc_ctl_fops = {
+	.owner = THIS_MODULE,
+	.open = display_crc_ctl_open,
+	.read = seq_read,
+	.llseek = seq_lseek,
+	.release = single_release,
+	.write = display_crc_ctl_write
+};
+
 static int
 i915_wedged_get(void *data, u64 *val)
 {
@@ -1885,6 +2650,72 @@ DEFINE_SIMPLE_ATTRIBUTE(i915_ring_stop_fops,
 			i915_ring_stop_get, i915_ring_stop_set,
 			"0x%08llx\n");
 
+static int
+i915_ring_missed_irq_get(void *data, u64 *val)
+{
+	struct drm_device *dev = data;
+	struct drm_i915_private *dev_priv = dev->dev_private;
+
+	*val = dev_priv->gpu_error.missed_irq_rings;
+	return 0;
+}
+
+static int
+i915_ring_missed_irq_set(void *data, u64 val)
+{
+	struct drm_device *dev = data;
+	struct drm_i915_private *dev_priv = dev->dev_private;
+	int ret;
+
+	/* Lock against concurrent debugfs callers */
+	ret = mutex_lock_interruptible(&dev->struct_mutex);
+	if (ret)
+		return ret;
+	dev_priv->gpu_error.missed_irq_rings = val;
+	mutex_unlock(&dev->struct_mutex);
+
+	return 0;
+}
+
+DEFINE_SIMPLE_ATTRIBUTE(i915_ring_missed_irq_fops,
+			i915_ring_missed_irq_get, i915_ring_missed_irq_set,
+			"0x%08llx\n");
+
+static int
+i915_ring_test_irq_get(void *data, u64 *val)
+{
+	struct drm_device *dev = data;
+	struct drm_i915_private *dev_priv = dev->dev_private;
+
+	*val = dev_priv->gpu_error.test_irq_rings;
+
+	return 0;
+}
+
+static int
+i915_ring_test_irq_set(void *data, u64 val)
+{
+	struct drm_device *dev = data;
+	struct drm_i915_private *dev_priv = dev->dev_private;
+	int ret;
+
+	DRM_DEBUG_DRIVER("Masking interrupts on rings 0x%08llx\n", val);
+
+	/* Lock against concurrent debugfs callers */
+	ret = mutex_lock_interruptible(&dev->struct_mutex);
+	if (ret)
+		return ret;
+
+	dev_priv->gpu_error.test_irq_rings = val;
+	mutex_unlock(&dev->struct_mutex);
+
+	return 0;
+}
+
+DEFINE_SIMPLE_ATTRIBUTE(i915_ring_test_irq_fops,
+			i915_ring_test_irq_get, i915_ring_test_irq_set,
+			"0x%08llx\n");
+
 #define DROP_UNBOUND 0x1
 #define DROP_BOUND 0x2
 #define DROP_RETIRE 0x4
@@ -1972,6 +2803,8 @@ i915_max_freq_get(void *data, u64 *val)
 	if (!(IS_GEN6(dev) || IS_GEN7(dev)))
 		return -ENODEV;
 
+	flush_delayed_work(&dev_priv->rps.delayed_resume_work);
+
 	ret = mutex_lock_interruptible(&dev_priv->rps.hw_lock);
 	if (ret)
 		return ret;
@@ -1996,6 +2829,8 @@ i915_max_freq_set(void *data, u64 val)
 	if (!(IS_GEN6(dev) || IS_GEN7(dev)))
 		return -ENODEV;
 
+	flush_delayed_work(&dev_priv->rps.delayed_resume_work);
+
 	DRM_DEBUG_DRIVER("Manually setting max freq to %llu\n", val);
 
 	ret = mutex_lock_interruptible(&dev_priv->rps.hw_lock);
@@ -2034,6 +2869,8 @@ i915_min_freq_get(void *data, u64 *val)
 	if (!(IS_GEN6(dev) || IS_GEN7(dev)))
 		return -ENODEV;
 
+	flush_delayed_work(&dev_priv->rps.delayed_resume_work);
+
 	ret = mutex_lock_interruptible(&dev_priv->rps.hw_lock);
 	if (ret)
 		return ret;
@@ -2058,6 +2895,8 @@ i915_min_freq_set(void *data, u64 val)
 	if (!(IS_GEN6(dev) || IS_GEN7(dev)))
 		return -ENODEV;
 
+	flush_delayed_work(&dev_priv->rps.delayed_resume_work);
+
 	DRM_DEBUG_DRIVER("Manually setting min freq to %llu\n", val);
 
 	ret = mutex_lock_interruptible(&dev_priv->rps.hw_lock);
@@ -2136,32 +2975,6 @@ DEFINE_SIMPLE_ATTRIBUTE(i915_cache_sharing_fops,
 			i915_cache_sharing_get, i915_cache_sharing_set,
 			"%llu\n");
 
-/* As the drm_debugfs_init() routines are called before dev->dev_private is
- * allocated we need to hook into the minor for release. */
-static int
-drm_add_fake_info_node(struct drm_minor *minor,
-		       struct dentry *ent,
-		       const void *key)
-{
-	struct drm_info_node *node;
-
-	node = kmalloc(sizeof(struct drm_info_node), GFP_KERNEL);
-	if (node == NULL) {
-		debugfs_remove(ent);
-		return -ENOMEM;
-	}
-
-	node->minor = minor;
-	node->dent = ent;
-	node->info_ent = (void *) key;
-
-	mutex_lock(&minor->debugfs_lock);
-	list_add(&node->list, &minor->debugfs_list);
-	mutex_unlock(&minor->debugfs_lock);
-
-	return 0;
-}
-
 static int i915_forcewake_open(struct inode *inode, struct file *file)
 {
 	struct drm_device *dev = inode->i_private;
@@ -2227,7 +3040,7 @@ static int i915_debugfs_create(struct dentry *root,
 	return drm_add_fake_info_node(minor, ent, fops);
 }
 
-static struct drm_info_list i915_debugfs_list[] = {
+static const struct drm_info_list i915_debugfs_list[] = {
 	{"i915_capabilities", i915_capabilities, 0},
 	{"i915_gem_objects", i915_gem_object_info, 0},
 	{"i915_gem_gtt", i915_gem_gtt_info, 0},
@@ -2269,7 +3082,7 @@ static struct drm_info_list i915_debugfs_list[] = {
 };
 #define I915_DEBUGFS_ENTRIES ARRAY_SIZE(i915_debugfs_list)
 
-static struct i915_debugfs_files {
+static const struct i915_debugfs_files {
 	const char *name;
 	const struct file_operations *fops;
 } i915_debugfs_files[] = {
@@ -2278,11 +3091,28 @@ static struct i915_debugfs_files {
 	{"i915_min_freq", &i915_min_freq_fops},
 	{"i915_cache_sharing", &i915_cache_sharing_fops},
 	{"i915_ring_stop", &i915_ring_stop_fops},
+	{"i915_ring_missed_irq", &i915_ring_missed_irq_fops},
+	{"i915_ring_test_irq", &i915_ring_test_irq_fops},
 	{"i915_gem_drop_caches", &i915_drop_caches_fops},
 	{"i915_error_state", &i915_error_state_fops},
 	{"i915_next_seqno", &i915_next_seqno_fops},
+	{"i915_display_crc_ctl", &i915_display_crc_ctl_fops},
 };
 
+void intel_display_crc_init(struct drm_device *dev)
+{
+	struct drm_i915_private *dev_priv = dev->dev_private;
+	int i;
+
+	for (i = 0; i < INTEL_INFO(dev)->num_pipes; i++) {
+		struct intel_pipe_crc *pipe_crc = &dev_priv->pipe_crc[i];
+
+		pipe_crc->opened = false;
+		spin_lock_init(&pipe_crc->lock);
+		init_waitqueue_head(&pipe_crc->wq);
+	}
+}
+
 int i915_debugfs_init(struct drm_minor *minor)
 {
 	int ret, i;
@@ -2291,6 +3121,12 @@ int i915_debugfs_init(struct drm_minor *minor)
 	if (ret)
 		return ret;
 
+	for (i = 0; i < ARRAY_SIZE(i915_pipe_crc_data); i++) {
+		ret = i915_pipe_crc_create(minor->debugfs_root, minor, i);
+		if (ret)
+			return ret;
+	}
+
 	for (i = 0; i < ARRAY_SIZE(i915_debugfs_files); i++) {
 		ret = i915_debugfs_create(minor->debugfs_root, minor,
 					  i915_debugfs_files[i].name,
@@ -2310,8 +3146,17 @@ void i915_debugfs_cleanup(struct drm_minor *minor)
 
 	drm_debugfs_remove_files(i915_debugfs_list,
 				 I915_DEBUGFS_ENTRIES, minor);
+
 	drm_debugfs_remove_files((struct drm_info_list *) &i915_forcewake_fops,
 				 1, minor);
+
+	for (i = 0; i < ARRAY_SIZE(i915_pipe_crc_data); i++) {
+		struct drm_info_list *info_list =
+			(struct drm_info_list *)&i915_pipe_crc_data[i];
+
+		drm_debugfs_remove_files(info_list, 1, minor);
+	}
+
 	for (i = 0; i < ARRAY_SIZE(i915_debugfs_files); i++) {
 		struct drm_info_list *info_list =
 			(struct drm_info_list *) i915_debugfs_files[i].fops;
diff --git a/drivers/gpu/drm/i915/i915_dma.c b/drivers/gpu/drm/i915/i915_dma.c
index d5c784d48671..0cab2d045135 100644
--- a/drivers/gpu/drm/i915/i915_dma.c
+++ b/drivers/gpu/drm/i915/i915_dma.c
@@ -52,7 +52,7 @@
 	intel_ring_emit(LP_RING(dev_priv), x)
 
 #define ADVANCE_LP_RING() \
-	intel_ring_advance(LP_RING(dev_priv))
+	__intel_ring_advance(LP_RING(dev_priv))
 
 /**
  * Lock test for when it's just for synchronization of ring access.
@@ -641,7 +641,7 @@ static int i915_batchbuffer(struct drm_device *dev, void *data,
 
 	if (batch->num_cliprects) {
 		cliprects = kcalloc(batch->num_cliprects,
-				    sizeof(struct drm_clip_rect),
+				    sizeof(*cliprects),
 				    GFP_KERNEL);
 		if (cliprects == NULL)
 			return -ENOMEM;
@@ -703,7 +703,7 @@ static int i915_cmdbuffer(struct drm_device *dev, void *data,
 
 	if (cmdbuf->num_cliprects) {
 		cliprects = kcalloc(cmdbuf->num_cliprects,
-				    sizeof(struct drm_clip_rect), GFP_KERNEL);
+				    sizeof(*cliprects), GFP_KERNEL);
 		if (cliprects == NULL) {
 			ret = -ENOMEM;
 			goto fail_batch_free;
@@ -931,7 +931,7 @@ static int i915_getparam(struct drm_device *dev, void *data,
 		value = READ_BREADCRUMB(dev_priv);
 		break;
 	case I915_PARAM_CHIPSET_ID:
-		value = dev->pci_device;
+		value = dev->pdev->device;
 		break;
 	case I915_PARAM_HAS_GEM:
 		value = 1;
@@ -1311,13 +1311,15 @@ static int i915_load_modeset_init(struct drm_device *dev)
 	if (ret)
 		goto cleanup_gem_stolen;
 
+	intel_power_domains_init_hw(dev);
+
 	/* Important: The output setup functions called by modeset_init need
 	 * working irqs for e.g. gmbus and dp aux transfers. */
 	intel_modeset_init(dev);
 
 	ret = i915_gem_init(dev);
 	if (ret)
-		goto cleanup_irq;
+		goto cleanup_power;
 
 	INIT_WORK(&dev_priv->console_resume_work, intel_console_resume);
 
@@ -1325,9 +1327,11 @@ static int i915_load_modeset_init(struct drm_device *dev)
 
 	/* Always safe in the mode setting case. */
 	/* FIXME: do pre/post-mode set stuff in core KMS code */
-	dev->vblank_disable_allowed = 1;
-	if (INTEL_INFO(dev)->num_pipes == 0)
+	dev->vblank_disable_allowed = true;
+	if (INTEL_INFO(dev)->num_pipes == 0) {
+		intel_display_power_put(dev, POWER_DOMAIN_VGA);
 		return 0;
+	}
 
 	ret = intel_fbdev_init(dev);
 	if (ret)
@@ -1362,7 +1366,8 @@ cleanup_gem:
 	mutex_unlock(&dev->struct_mutex);
 	i915_gem_cleanup_aliasing_ppgtt(dev);
 	drm_mm_takedown(&dev_priv->gtt.base.mm);
-cleanup_irq:
+cleanup_power:
+	intel_display_power_put(dev, POWER_DOMAIN_VGA);
 	drm_irq_uninstall(dev);
 cleanup_gem_stolen:
 	i915_gem_cleanup_stolen(dev);
@@ -1398,6 +1403,7 @@ void i915_master_destroy(struct drm_device *dev, struct drm_master *master)
 	master->driver_priv = NULL;
 }
 
+#ifdef CONFIG_DRM_I915_FBDEV
 static void i915_kick_out_firmware_fb(struct drm_i915_private *dev_priv)
 {
 	struct apertures_struct *ap;
@@ -1418,6 +1424,11 @@ static void i915_kick_out_firmware_fb(struct drm_i915_private *dev_priv)
 
 	kfree(ap);
 }
+#else
+static void i915_kick_out_firmware_fb(struct drm_i915_private *dev_priv)
+{
+}
+#endif
 
 static void i915_dump_device_info(struct drm_i915_private *dev_priv)
 {
@@ -1459,17 +1470,13 @@ int i915_driver_load(struct drm_device *dev, unsigned long flags)
 	info = (struct intel_device_info *) flags;
 
 	/* Refuse to load on gen6+ without kms enabled. */
-	if (info->gen >= 6 && !drm_core_check_feature(dev, DRIVER_MODESET))
+	if (info->gen >= 6 && !drm_core_check_feature(dev, DRIVER_MODESET)) {
+		DRM_INFO("Your hardware requires kernel modesetting (KMS)\n");
+		DRM_INFO("See CONFIG_DRM_I915_KMS, nomodeset, and i915.modeset parameters\n");
 		return -ENODEV;
+	}
 
-	/* i915 has 4 more counters */
-	dev->counters += 4;
-	dev->types[6] = _DRM_STAT_IRQ;
-	dev->types[7] = _DRM_STAT_PRIMARY;
-	dev->types[8] = _DRM_STAT_SECONDARY;
-	dev->types[9] = _DRM_STAT_DMA;
-
-	dev_priv = kzalloc(sizeof(drm_i915_private_t), GFP_KERNEL);
+	dev_priv = kzalloc(sizeof(*dev_priv), GFP_KERNEL);
 	if (dev_priv == NULL)
 		return -ENOMEM;
 
@@ -1494,6 +1501,8 @@ int i915_driver_load(struct drm_device *dev, unsigned long flags)
 	dev_priv->pc8.disable_count = 2; /* requirements_met + gpu_idle */
 	INIT_DELAYED_WORK(&dev_priv->pc8.enable_work, hsw_enable_pc8_work);
 
+	intel_display_crc_init(dev);
+
 	i915_dump_device_info(dev_priv);
 
 	/* Not all pre-production machines fall into this category, only the
@@ -1531,19 +1540,14 @@ int i915_driver_load(struct drm_device *dev, unsigned long flags)
 
 	intel_uncore_early_sanitize(dev);
 
-	if (IS_HASWELL(dev) && (I915_READ(HSW_EDRAM_PRESENT) == 1)) {
-		/* The docs do not explain exactly how the calculation can be
-		 * made. It is somewhat guessable, but for now, it's always
-		 * 128MB.
-		 * NB: We can't write IDICR yet because we do not have gt funcs
-		 * set up */
-		dev_priv->ellc_size = 128;
-		DRM_INFO("Found %zuMB of eLLC\n", dev_priv->ellc_size);
-	}
+	/* This must be called before any calls to HAS_PCH_* */
+	intel_detect_pch(dev);
+
+	intel_uncore_init(dev);
 
 	ret = i915_gem_gtt_init(dev);
 	if (ret)
-		goto put_bridge;
+		goto out_regs;
 
 	if (drm_core_check_feature(dev, DRIVER_MODESET))
 		i915_kick_out_firmware_fb(dev_priv);
@@ -1572,7 +1576,7 @@ int i915_driver_load(struct drm_device *dev, unsigned long flags)
 				     aperture_size);
 	if (dev_priv->gtt.mappable == NULL) {
 		ret = -EIO;
-		goto out_rmmap;
+		goto out_gtt;
 	}
 
 	dev_priv->gtt.mtrr = arch_phys_wc_add(dev_priv->gtt.mappable_base,
@@ -1598,13 +1602,9 @@ int i915_driver_load(struct drm_device *dev, unsigned long flags)
 		goto out_mtrrfree;
 	}
 
-	/* This must be called before any calls to HAS_PCH_* */
-	intel_detect_pch(dev);
-
 	intel_irq_init(dev);
 	intel_pm_init(dev);
 	intel_uncore_sanitize(dev);
-	intel_uncore_init(dev);
 
 	/* Try to make sure MCHBAR is enabled before poking at it */
 	intel_setup_mchbar(dev);
@@ -1640,13 +1640,13 @@ int i915_driver_load(struct drm_device *dev, unsigned long flags)
 	}
 
 	if (HAS_POWER_WELL(dev))
-		i915_init_power_well(dev);
+		intel_power_domains_init(dev);
 
 	if (drm_core_check_feature(dev, DRIVER_MODESET)) {
 		ret = i915_load_modeset_init(dev);
 		if (ret < 0) {
 			DRM_ERROR("failed to init modeset\n");
-			goto out_gem_unload;
+			goto out_power_well;
 		}
 	} else {
 		/* Start out suspended in ums mode. */
@@ -1666,6 +1666,10 @@ int i915_driver_load(struct drm_device *dev, unsigned long flags)
 
 	return 0;
 
+out_power_well:
+	if (HAS_POWER_WELL(dev))
+		intel_power_domains_remove(dev);
+	drm_vblank_cleanup(dev);
 out_gem_unload:
 	if (dev_priv->mm.inactive_shrinker.scan_objects)
 		unregister_shrinker(&dev_priv->mm.inactive_shrinker);
@@ -1679,12 +1683,18 @@ out_gem_unload:
 out_mtrrfree:
 	arch_phys_wc_del(dev_priv->gtt.mtrr);
 	io_mapping_free(dev_priv->gtt.mappable);
+out_gtt:
+	list_del(&dev_priv->gtt.base.global_link);
+	drm_mm_takedown(&dev_priv->gtt.base.mm);
 	dev_priv->gtt.base.cleanup(&dev_priv->gtt.base);
-out_rmmap:
+out_regs:
+	intel_uncore_fini(dev);
 	pci_iounmap(dev->pdev, dev_priv->regs);
 put_bridge:
 	pci_dev_put(dev_priv->bridge_dev);
 free_priv:
+	if (dev_priv->slab)
+		kmem_cache_destroy(dev_priv->slab);
 	kfree(dev_priv);
 	return ret;
 }
@@ -1700,8 +1710,8 @@ int i915_driver_unload(struct drm_device *dev)
 		/* The i915.ko module is still not prepared to be loaded when
 		 * the power well is not enabled, so just enable it in case
 		 * we're going to unload/reload. */
-		intel_set_power_well(dev, true);
-		i915_remove_power_well(dev);
+		intel_display_set_init_power(dev, true);
+		intel_power_domains_remove(dev);
 	}
 
 	i915_teardown_sysfs(dev);
@@ -1709,15 +1719,9 @@ int i915_driver_unload(struct drm_device *dev)
 	if (dev_priv->mm.inactive_shrinker.scan_objects)
 		unregister_shrinker(&dev_priv->mm.inactive_shrinker);
 
-	mutex_lock(&dev->struct_mutex);
-	ret = i915_gpu_idle(dev);
+	ret = i915_gem_suspend(dev);
 	if (ret)
 		DRM_ERROR("failed to idle hardware: %d\n", ret);
-	i915_gem_retire_requests(dev);
-	mutex_unlock(&dev->struct_mutex);
-
-	/* Cancel the retire work handler, which should be idle now. */
-	cancel_delayed_work_sync(&dev_priv->mm.retire_work);
 
 	io_mapping_free(dev_priv->gtt.mappable);
 	arch_phys_wc_del(dev_priv->gtt.mtrr);
@@ -1774,8 +1778,8 @@ int i915_driver_unload(struct drm_device *dev)
 	list_del(&dev_priv->gtt.base.global_link);
 	WARN_ON(!list_empty(&dev_priv->vm_list));
 	drm_mm_takedown(&dev_priv->gtt.base.mm);
-	if (dev_priv->regs != NULL)
-		pci_iounmap(dev->pdev, dev_priv->regs);
+
+	drm_vblank_cleanup(dev);
 
 	intel_teardown_gmbus(dev);
 	intel_teardown_mchbar(dev);
@@ -1785,6 +1789,10 @@ int i915_driver_unload(struct drm_device *dev)
 
 	dev_priv->gtt.base.cleanup(&dev_priv->gtt.base);
 
+	intel_uncore_fini(dev);
+	if (dev_priv->regs != NULL)
+		pci_iounmap(dev->pdev, dev_priv->regs);
+
 	if (dev_priv->slab)
 		kmem_cache_destroy(dev_priv->slab);
 
@@ -1796,19 +1804,11 @@ int i915_driver_unload(struct drm_device *dev)
 
 int i915_driver_open(struct drm_device *dev, struct drm_file *file)
 {
-	struct drm_i915_file_private *file_priv;
-
-	DRM_DEBUG_DRIVER("\n");
-	file_priv = kzalloc(sizeof(*file_priv), GFP_KERNEL);
-	if (!file_priv)
-		return -ENOMEM;
-
-	file->driver_priv = file_priv;
-
-	spin_lock_init(&file_priv->mm.lock);
-	INIT_LIST_HEAD(&file_priv->mm.request_list);
+	int ret;
 
-	idr_init(&file_priv->context_idr);
+	ret = i915_gem_open(dev, file);
+	if (ret)
+		return ret;
 
 	return 0;
 }
@@ -1836,7 +1836,7 @@ void i915_driver_lastclose(struct drm_device * dev)
 		return;
 
 	if (drm_core_check_feature(dev, DRIVER_MODESET)) {
-		intel_fb_restore_mode(dev);
+		intel_fbdev_restore_mode(dev);
 		vga_switcheroo_process_delayed_switch();
 		return;
 	}
diff --git a/drivers/gpu/drm/i915/i915_drv.c b/drivers/gpu/drm/i915/i915_drv.c
index 2ad27880cd04..989be12cdd6e 100644
--- a/drivers/gpu/drm/i915/i915_drv.c
+++ b/drivers/gpu/drm/i915/i915_drv.c
@@ -160,49 +160,58 @@ extern int intel_agp_enabled;
 static const struct intel_device_info intel_i830_info = {
 	.gen = 2, .is_mobile = 1, .cursor_needs_physical = 1, .num_pipes = 2,
 	.has_overlay = 1, .overlay_needs_physical = 1,
+	.ring_mask = RENDER_RING,
 };
 
 static const struct intel_device_info intel_845g_info = {
 	.gen = 2, .num_pipes = 1,
 	.has_overlay = 1, .overlay_needs_physical = 1,
+	.ring_mask = RENDER_RING,
 };
 
 static const struct intel_device_info intel_i85x_info = {
 	.gen = 2, .is_i85x = 1, .is_mobile = 1, .num_pipes = 2,
 	.cursor_needs_physical = 1,
 	.has_overlay = 1, .overlay_needs_physical = 1,
+	.ring_mask = RENDER_RING,
 };
 
 static const struct intel_device_info intel_i865g_info = {
 	.gen = 2, .num_pipes = 1,
 	.has_overlay = 1, .overlay_needs_physical = 1,
+	.ring_mask = RENDER_RING,
 };
 
 static const struct intel_device_info intel_i915g_info = {
 	.gen = 3, .is_i915g = 1, .cursor_needs_physical = 1, .num_pipes = 2,
 	.has_overlay = 1, .overlay_needs_physical = 1,
+	.ring_mask = RENDER_RING,
 };
 static const struct intel_device_info intel_i915gm_info = {
 	.gen = 3, .is_mobile = 1, .num_pipes = 2,
 	.cursor_needs_physical = 1,
 	.has_overlay = 1, .overlay_needs_physical = 1,
 	.supports_tv = 1,
+	.ring_mask = RENDER_RING,
 };
 static const struct intel_device_info intel_i945g_info = {
 	.gen = 3, .has_hotplug = 1, .cursor_needs_physical = 1, .num_pipes = 2,
 	.has_overlay = 1, .overlay_needs_physical = 1,
+	.ring_mask = RENDER_RING,
 };
 static const struct intel_device_info intel_i945gm_info = {
 	.gen = 3, .is_i945gm = 1, .is_mobile = 1, .num_pipes = 2,
 	.has_hotplug = 1, .cursor_needs_physical = 1,
 	.has_overlay = 1, .overlay_needs_physical = 1,
 	.supports_tv = 1,
+	.ring_mask = RENDER_RING,
 };
 
 static const struct intel_device_info intel_i965g_info = {
 	.gen = 4, .is_broadwater = 1, .num_pipes = 2,
 	.has_hotplug = 1,
 	.has_overlay = 1,
+	.ring_mask = RENDER_RING,
 };
 
 static const struct intel_device_info intel_i965gm_info = {
@@ -210,18 +219,20 @@ static const struct intel_device_info intel_i965gm_info = {
 	.is_mobile = 1, .has_fbc = 1, .has_hotplug = 1,
 	.has_overlay = 1,
 	.supports_tv = 1,
+	.ring_mask = RENDER_RING,
 };
 
 static const struct intel_device_info intel_g33_info = {
 	.gen = 3, .is_g33 = 1, .num_pipes = 2,
 	.need_gfx_hws = 1, .has_hotplug = 1,
 	.has_overlay = 1,
+	.ring_mask = RENDER_RING,
 };
 
 static const struct intel_device_info intel_g45_info = {
 	.gen = 4, .is_g4x = 1, .need_gfx_hws = 1, .num_pipes = 2,
 	.has_pipe_cxsr = 1, .has_hotplug = 1,
-	.has_bsd_ring = 1,
+	.ring_mask = RENDER_RING | BSD_RING,
 };
 
 static const struct intel_device_info intel_gm45_info = {
@@ -229,7 +240,7 @@ static const struct intel_device_info intel_gm45_info = {
 	.is_mobile = 1, .need_gfx_hws = 1, .has_fbc = 1,
 	.has_pipe_cxsr = 1, .has_hotplug = 1,
 	.supports_tv = 1,
-	.has_bsd_ring = 1,
+	.ring_mask = RENDER_RING | BSD_RING,
 };
 
 static const struct intel_device_info intel_pineview_info = {
@@ -241,42 +252,36 @@ static const struct intel_device_info intel_pineview_info = {
 static const struct intel_device_info intel_ironlake_d_info = {
 	.gen = 5, .num_pipes = 2,
 	.need_gfx_hws = 1, .has_hotplug = 1,
-	.has_bsd_ring = 1,
+	.ring_mask = RENDER_RING | BSD_RING,
 };
 
 static const struct intel_device_info intel_ironlake_m_info = {
 	.gen = 5, .is_mobile = 1, .num_pipes = 2,
 	.need_gfx_hws = 1, .has_hotplug = 1,
 	.has_fbc = 1,
-	.has_bsd_ring = 1,
+	.ring_mask = RENDER_RING | BSD_RING,
 };
 
 static const struct intel_device_info intel_sandybridge_d_info = {
 	.gen = 6, .num_pipes = 2,
 	.need_gfx_hws = 1, .has_hotplug = 1,
-	.has_bsd_ring = 1,
-	.has_blt_ring = 1,
+	.ring_mask = RENDER_RING | BSD_RING | BLT_RING,
 	.has_llc = 1,
-	.has_force_wake = 1,
 };
 
 static const struct intel_device_info intel_sandybridge_m_info = {
 	.gen = 6, .is_mobile = 1, .num_pipes = 2,
 	.need_gfx_hws = 1, .has_hotplug = 1,
 	.has_fbc = 1,
-	.has_bsd_ring = 1,
-	.has_blt_ring = 1,
+	.ring_mask = RENDER_RING | BSD_RING | BLT_RING,
 	.has_llc = 1,
-	.has_force_wake = 1,
 };
 
 #define GEN7_FEATURES  \
 	.gen = 7, .num_pipes = 3, \
 	.need_gfx_hws = 1, .has_hotplug = 1, \
-	.has_bsd_ring = 1, \
-	.has_blt_ring = 1, \
-	.has_llc = 1, \
-	.has_force_wake = 1
+	.ring_mask = RENDER_RING | BSD_RING | BLT_RING, \
+	.has_llc = 1
 
 static const struct intel_device_info intel_ivybridge_d_info = {
 	GEN7_FEATURES,
@@ -318,7 +323,7 @@ static const struct intel_device_info intel_haswell_d_info = {
 	.is_haswell = 1,
 	.has_ddi = 1,
 	.has_fpga_dbg = 1,
-	.has_vebox_ring = 1,
+	.ring_mask = RENDER_RING | BSD_RING | BLT_RING | VEBOX_RING,
 };
 
 static const struct intel_device_info intel_haswell_m_info = {
@@ -328,7 +333,25 @@ static const struct intel_device_info intel_haswell_m_info = {
 	.has_ddi = 1,
 	.has_fpga_dbg = 1,
 	.has_fbc = 1,
-	.has_vebox_ring = 1,
+	.ring_mask = RENDER_RING | BSD_RING | BLT_RING | VEBOX_RING,
+};
+
+static const struct intel_device_info intel_broadwell_d_info = {
+	.is_preliminary = 1,
+	.gen = 8, .num_pipes = 3,
+	.need_gfx_hws = 1, .has_hotplug = 1,
+	.ring_mask = RENDER_RING | BSD_RING | BLT_RING | VEBOX_RING,
+	.has_llc = 1,
+	.has_ddi = 1,
+};
+
+static const struct intel_device_info intel_broadwell_m_info = {
+	.is_preliminary = 1,
+	.gen = 8, .is_mobile = 1, .num_pipes = 3,
+	.need_gfx_hws = 1, .has_hotplug = 1,
+	.ring_mask = RENDER_RING | BSD_RING | BLT_RING | VEBOX_RING,
+	.has_llc = 1,
+	.has_ddi = 1,
 };
 
 /*
@@ -362,7 +385,9 @@ static const struct intel_device_info intel_haswell_m_info = {
 	INTEL_HSW_D_IDS(&intel_haswell_d_info), \
 	INTEL_HSW_M_IDS(&intel_haswell_m_info), \
 	INTEL_VLV_M_IDS(&intel_valleyview_m_info),	\
-	INTEL_VLV_D_IDS(&intel_valleyview_d_info)
+	INTEL_VLV_D_IDS(&intel_valleyview_d_info),	\
+	INTEL_BDW_M_IDS(&intel_broadwell_m_info),	\
+	INTEL_BDW_D_IDS(&intel_broadwell_d_info)
 
 static const struct pci_device_id pciidlist[] = {		/* aka */
 	INTEL_PCI_IDS,
@@ -416,13 +441,19 @@ void intel_detect_pch(struct drm_device *dev)
 			} else if (id == INTEL_PCH_PPT_DEVICE_ID_TYPE) {
 				/* PantherPoint is CPT compatible */
 				dev_priv->pch_type = PCH_CPT;
-				DRM_DEBUG_KMS("Found PatherPoint PCH\n");
+				DRM_DEBUG_KMS("Found PantherPoint PCH\n");
 				WARN_ON(!(IS_GEN6(dev) || IS_IVYBRIDGE(dev)));
 			} else if (id == INTEL_PCH_LPT_DEVICE_ID_TYPE) {
 				dev_priv->pch_type = PCH_LPT;
 				DRM_DEBUG_KMS("Found LynxPoint PCH\n");
 				WARN_ON(!IS_HASWELL(dev));
 				WARN_ON(IS_ULT(dev));
+			} else if (IS_BROADWELL(dev)) {
+				dev_priv->pch_type = PCH_LPT;
+				dev_priv->pch_id =
+					INTEL_PCH_LPT_LP_DEVICE_ID_TYPE;
+				DRM_DEBUG_KMS("This is Broadwell, assuming "
+					      "LynxPoint LP PCH\n");
 			} else if (id == INTEL_PCH_LPT_LP_DEVICE_ID_TYPE) {
 				dev_priv->pch_type = PCH_LPT;
 				DRM_DEBUG_KMS("Found LynxPoint LP PCH\n");
@@ -447,6 +478,12 @@ bool i915_semaphore_is_enabled(struct drm_device *dev)
 	if (INTEL_INFO(dev)->gen < 6)
 		return 0;
 
+	/* Until we get further testing... */
+	if (IS_GEN8(dev)) {
+		WARN_ON(!i915_preliminary_hw_support);
+		return 0;
+	}
+
 	if (i915_semaphores >= 0)
 		return i915_semaphores;
 
@@ -472,7 +509,7 @@ static int i915_drm_freeze(struct drm_device *dev)
 	/* We do a lot of poking in a lot of registers, make sure they work
 	 * properly. */
 	hsw_disable_package_c8(dev_priv);
-	intel_set_power_well(dev, true);
+	intel_display_set_init_power(dev, true);
 
 	drm_kms_helper_poll_disable(dev);
 
@@ -482,9 +519,7 @@ static int i915_drm_freeze(struct drm_device *dev)
 	if (drm_core_check_feature(dev, DRIVER_MODESET)) {
 		int error;
 
-		mutex_lock(&dev->struct_mutex);
-		error = i915_gem_idle(dev);
-		mutex_unlock(&dev->struct_mutex);
+		error = i915_gem_suspend(dev);
 		if (error) {
 			dev_err(&dev->pdev->dev,
 				"GEM idle failed, resume might fail\n");
@@ -578,11 +613,24 @@ static void intel_resume_hotplug(struct drm_device *dev)
 	drm_helper_hpd_irq_event(dev);
 }
 
-static int __i915_drm_thaw(struct drm_device *dev)
+static int __i915_drm_thaw(struct drm_device *dev, bool restore_gtt_mappings)
 {
 	struct drm_i915_private *dev_priv = dev->dev_private;
 	int error = 0;
 
+	intel_uncore_early_sanitize(dev);
+
+	intel_uncore_sanitize(dev);
+
+	if (drm_core_check_feature(dev, DRIVER_MODESET) &&
+	    restore_gtt_mappings) {
+		mutex_lock(&dev->struct_mutex);
+		i915_gem_restore_gtt_mappings(dev);
+		mutex_unlock(&dev->struct_mutex);
+	}
+
+	intel_power_domains_init_hw(dev);
+
 	i915_restore_state(dev);
 	intel_opregion_setup(dev);
 
@@ -642,20 +690,10 @@ static int __i915_drm_thaw(struct drm_device *dev)
 
 static int i915_drm_thaw(struct drm_device *dev)
 {
-	int error = 0;
-
-	intel_uncore_sanitize(dev);
-
-	if (drm_core_check_feature(dev, DRIVER_MODESET)) {
-		mutex_lock(&dev->struct_mutex);
-		i915_gem_restore_gtt_mappings(dev);
-		mutex_unlock(&dev->struct_mutex);
-	} else if (drm_core_check_feature(dev, DRIVER_MODESET))
+	if (drm_core_check_feature(dev, DRIVER_MODESET))
 		i915_check_and_clear_faults(dev);
 
-	__i915_drm_thaw(dev);
-
-	return error;
+	return __i915_drm_thaw(dev, true);
 }
 
 int i915_resume(struct drm_device *dev)
@@ -671,20 +709,12 @@ int i915_resume(struct drm_device *dev)
 
 	pci_set_master(dev->pdev);
 
-	intel_uncore_sanitize(dev);
-
 	/*
 	 * Platforms with opregion should have sane BIOS, older ones (gen3 and
-	 * earlier) need this since the BIOS might clear all our scratch PTEs.
+	 * earlier) need to restore the GTT mappings since the BIOS might clear
+	 * all our scratch PTEs.
 	 */
-	if (drm_core_check_feature(dev, DRIVER_MODESET) &&
-	    !dev_priv->opregion.header) {
-		mutex_lock(&dev->struct_mutex);
-		i915_gem_restore_gtt_mappings(dev);
-		mutex_unlock(&dev->struct_mutex);
-	}
-
-	ret = __i915_drm_thaw(dev);
+	ret = __i915_drm_thaw(dev, !dev_priv->opregion.header);
 	if (ret)
 		return ret;
 
@@ -722,24 +752,19 @@ int i915_reset(struct drm_device *dev)
 
 	simulated = dev_priv->gpu_error.stop_rings != 0;
 
-	if (!simulated && get_seconds() - dev_priv->gpu_error.last_reset < 5) {
-		DRM_ERROR("GPU hanging too fast, declaring wedged!\n");
-		ret = -ENODEV;
-	} else {
-		ret = intel_gpu_reset(dev);
-
-		/* Also reset the gpu hangman. */
-		if (simulated) {
-			DRM_INFO("Simulated gpu hang, resetting stop_rings\n");
-			dev_priv->gpu_error.stop_rings = 0;
-			if (ret == -ENODEV) {
-				DRM_ERROR("Reset not implemented, but ignoring "
-					  "error for simulated gpu hangs\n");
-				ret = 0;
-			}
-		} else
-			dev_priv->gpu_error.last_reset = get_seconds();
+	ret = intel_gpu_reset(dev);
+
+	/* Also reset the gpu hangman. */
+	if (simulated) {
+		DRM_INFO("Simulated gpu hang, resetting stop_rings\n");
+		dev_priv->gpu_error.stop_rings = 0;
+		if (ret == -ENODEV) {
+			DRM_ERROR("Reset not implemented, but ignoring "
+				  "error for simulated gpu hangs\n");
+			ret = 0;
+		}
 	}
+
 	if (ret) {
 		DRM_ERROR("Failed to reset chip.\n");
 		mutex_unlock(&dev->struct_mutex);
@@ -762,30 +787,17 @@ int i915_reset(struct drm_device *dev)
 	 */
 	if (drm_core_check_feature(dev, DRIVER_MODESET) ||
 			!dev_priv->ums.mm_suspended) {
-		struct intel_ring_buffer *ring;
-		int i;
-
+		bool hw_contexts_disabled = dev_priv->hw_contexts_disabled;
 		dev_priv->ums.mm_suspended = 0;
 
-		i915_gem_init_swizzling(dev);
-
-		for_each_ring(ring, dev_priv, i)
-			ring->init(ring);
-
-		i915_gem_context_init(dev);
-		if (dev_priv->mm.aliasing_ppgtt) {
-			ret = dev_priv->mm.aliasing_ppgtt->enable(dev);
-			if (ret)
-				i915_gem_cleanup_aliasing_ppgtt(dev);
-		}
-
-		/*
-		 * It would make sense to re-init all the other hw state, at
-		 * least the rps/rc6/emon init done within modeset_init_hw. For
-		 * some unknown reason, this blows up my ilk, so don't.
-		 */
-
+		ret = i915_gem_init_hw(dev);
+		if (!hw_contexts_disabled && dev_priv->hw_contexts_disabled)
+			DRM_ERROR("HW contexts didn't survive reset\n");
 		mutex_unlock(&dev->struct_mutex);
+		if (ret) {
+			DRM_ERROR("Failed hw init on reset %d\n", ret);
+			return ret;
+		}
 
 		drm_irq_uninstall(dev);
 		drm_irq_install(dev);
@@ -802,6 +814,12 @@ static int i915_pci_probe(struct pci_dev *pdev, const struct pci_device_id *ent)
 	struct intel_device_info *intel_info =
 		(struct intel_device_info *) ent->driver_data;
 
+	if (IS_PRELIMINARY_HW(intel_info) && !i915_preliminary_hw_support) {
+		DRM_INFO("This hardware requires preliminary hardware support.\n"
+			 "See CONFIG_DRM_I915_PRELIMINARY_HW_SUPPORT, and/or modparam preliminary_hw_support\n");
+		return -ENODEV;
+	}
+
 	/* Only bind to function 0 of the device. Early generations
 	 * used function 1 as a placeholder for multi-head. This causes
 	 * us confusion instead, especially on the systems where both
@@ -949,7 +967,6 @@ static struct drm_driver driver = {
 	.debugfs_init = i915_debugfs_init,
 	.debugfs_cleanup = i915_debugfs_cleanup,
 #endif
-	.gem_init_object = i915_gem_init_object,
 	.gem_free_object = i915_gem_free_object,
 	.gem_vm_ops = &i915_gem_vm_ops,
 
diff --git a/drivers/gpu/drm/i915/i915_drv.h b/drivers/gpu/drm/i915/i915_drv.h
index ab0f2c0a440c..8600c315b4c4 100644
--- a/drivers/gpu/drm/i915/i915_drv.h
+++ b/drivers/gpu/drm/i915/i915_drv.h
@@ -54,6 +54,7 @@
 #define DRIVER_DATE		"20080730"
 
 enum pipe {
+	INVALID_PIPE = -1,
 	PIPE_A = 0,
 	PIPE_B,
 	PIPE_C,
@@ -98,13 +99,29 @@ enum intel_display_power_domain {
 	POWER_DOMAIN_TRANSCODER_A,
 	POWER_DOMAIN_TRANSCODER_B,
 	POWER_DOMAIN_TRANSCODER_C,
-	POWER_DOMAIN_TRANSCODER_EDP = POWER_DOMAIN_TRANSCODER_A + 0xF,
+	POWER_DOMAIN_TRANSCODER_EDP,
+	POWER_DOMAIN_VGA,
+	POWER_DOMAIN_INIT,
+
+	POWER_DOMAIN_NUM,
 };
 
+#define POWER_DOMAIN_MASK (BIT(POWER_DOMAIN_NUM) - 1)
+
 #define POWER_DOMAIN_PIPE(pipe) ((pipe) + POWER_DOMAIN_PIPE_A)
 #define POWER_DOMAIN_PIPE_PANEL_FITTER(pipe) \
 		((pipe) + POWER_DOMAIN_PIPE_A_PANEL_FITTER)
-#define POWER_DOMAIN_TRANSCODER(tran) ((tran) + POWER_DOMAIN_TRANSCODER_A)
+#define POWER_DOMAIN_TRANSCODER(tran) \
+	((tran) == TRANSCODER_EDP ? POWER_DOMAIN_TRANSCODER_EDP : \
+	 (tran) + POWER_DOMAIN_TRANSCODER_A)
+
+#define HSW_ALWAYS_ON_POWER_DOMAINS (		\
+	BIT(POWER_DOMAIN_PIPE_A) |		\
+	BIT(POWER_DOMAIN_TRANSCODER_EDP))
+#define BDW_ALWAYS_ON_POWER_DOMAINS (		\
+	BIT(POWER_DOMAIN_PIPE_A) |		\
+	BIT(POWER_DOMAIN_TRANSCODER_EDP) |	\
+	BIT(POWER_DOMAIN_PIPE_A_PANEL_FITTER))
 
 enum hpd_pin {
 	HPD_NONE = 0,
@@ -225,9 +242,12 @@ struct intel_opregion {
 	struct opregion_header __iomem *header;
 	struct opregion_acpi __iomem *acpi;
 	struct opregion_swsci __iomem *swsci;
+	u32 swsci_gbda_sub_functions;
+	u32 swsci_sbcb_sub_functions;
 	struct opregion_asle __iomem *asle;
 	void __iomem *vbt;
 	u32 __iomem *lid_state;
+	struct work_struct asle_work;
 };
 #define OPREGION_SIZE            (8*1024)
 
@@ -285,6 +305,7 @@ struct drm_i915_error_state {
 	u32 cpu_ring_tail[I915_NUM_RINGS];
 	u32 error; /* gen6+ */
 	u32 err_int; /* gen7 */
+	u32 bbstate[I915_NUM_RINGS];
 	u32 instpm[I915_NUM_RINGS];
 	u32 instps[I915_NUM_RINGS];
 	u32 extra_instdone[I915_NUM_INSTDONE_REG];
@@ -321,11 +342,13 @@ struct drm_i915_error_state {
 		u32 dirty:1;
 		u32 purgeable:1;
 		s32 ring:4;
-		u32 cache_level:2;
+		u32 cache_level:3;
 	} **active_bo, **pinned_bo;
 	u32 *active_bo_count, *pinned_bo_count;
 	struct intel_overlay_error_state *overlay;
 	struct intel_display_error_state *display;
+	int hangcheck_score[I915_NUM_RINGS];
+	enum intel_ring_hangcheck_action hangcheck_action[I915_NUM_RINGS];
 };
 
 struct intel_crtc_config;
@@ -357,7 +380,7 @@ struct drm_i915_display_funcs {
 			  int target, int refclk,
 			  struct dpll *match_clock,
 			  struct dpll *best_clock);
-	void (*update_wm)(struct drm_device *dev);
+	void (*update_wm)(struct drm_crtc *crtc);
 	void (*update_sprite_wm)(struct drm_plane *plane,
 				 struct drm_crtc *crtc,
 				 uint32_t sprite_width, int pixel_size,
@@ -367,7 +390,6 @@ struct drm_i915_display_funcs {
 	 * fills out the pipe-config with the hw state. */
 	bool (*get_pipe_config)(struct intel_crtc *,
 				struct intel_crtc_config *);
-	void (*get_clock)(struct intel_crtc *, struct intel_crtc_config *);
 	int (*crtc_mode_set)(struct drm_crtc *crtc,
 			     int x, int y,
 			     struct drm_framebuffer *old_fb);
@@ -375,7 +397,8 @@ struct drm_i915_display_funcs {
 	void (*crtc_disable)(struct drm_crtc *crtc);
 	void (*off)(struct drm_crtc *crtc);
 	void (*write_eld)(struct drm_connector *connector,
-			  struct drm_crtc *crtc);
+			  struct drm_crtc *crtc,
+			  struct drm_display_mode *mode);
 	void (*fdi_link_train)(struct drm_crtc *crtc);
 	void (*init_clock_gating)(struct drm_device *dev);
 	int (*queue_flip)(struct drm_device *dev, struct drm_crtc *crtc,
@@ -395,6 +418,20 @@ struct drm_i915_display_funcs {
 struct intel_uncore_funcs {
 	void (*force_wake_get)(struct drm_i915_private *dev_priv);
 	void (*force_wake_put)(struct drm_i915_private *dev_priv);
+
+	uint8_t  (*mmio_readb)(struct drm_i915_private *dev_priv, off_t offset, bool trace);
+	uint16_t (*mmio_readw)(struct drm_i915_private *dev_priv, off_t offset, bool trace);
+	uint32_t (*mmio_readl)(struct drm_i915_private *dev_priv, off_t offset, bool trace);
+	uint64_t (*mmio_readq)(struct drm_i915_private *dev_priv, off_t offset, bool trace);
+
+	void (*mmio_writeb)(struct drm_i915_private *dev_priv, off_t offset,
+				uint8_t val, bool trace);
+	void (*mmio_writew)(struct drm_i915_private *dev_priv, off_t offset,
+				uint16_t val, bool trace);
+	void (*mmio_writel)(struct drm_i915_private *dev_priv, off_t offset,
+				uint32_t val, bool trace);
+	void (*mmio_writeq)(struct drm_i915_private *dev_priv, off_t offset,
+				uint64_t val, bool trace);
 };
 
 struct intel_uncore {
@@ -404,6 +441,8 @@ struct intel_uncore {
 
 	unsigned fifo_count;
 	unsigned forcewake_count;
+
+	struct delayed_work force_wake_work;
 };
 
 #define DEV_INFO_FOR_EACH_FLAG(func, sep) \
@@ -420,7 +459,7 @@ struct intel_uncore {
 	func(is_ivybridge) sep \
 	func(is_valleyview) sep \
 	func(is_haswell) sep \
-	func(has_force_wake) sep \
+	func(is_preliminary) sep \
 	func(has_fbc) sep \
 	func(has_pipe_cxsr) sep \
 	func(has_hotplug) sep \
@@ -428,9 +467,6 @@ struct intel_uncore {
 	func(has_overlay) sep \
 	func(overlay_needs_physical) sep \
 	func(supports_tv) sep \
-	func(has_bsd_ring) sep \
-	func(has_blt_ring) sep \
-	func(has_vebox_ring) sep \
 	func(has_llc) sep \
 	func(has_ddi) sep \
 	func(has_fpga_dbg)
@@ -442,6 +478,7 @@ struct intel_device_info {
 	u32 display_mmio_offset;
 	u8 num_pipes:3;
 	u8 gen;
+	u8 ring_mask; /* Rings supported by the HW */
 	DEV_INFO_FOR_EACH_FLAG(DEFINE_FLAG, SEP_SEMICOLON);
 };
 
@@ -542,10 +579,21 @@ struct i915_gtt {
 struct i915_hw_ppgtt {
 	struct i915_address_space base;
 	unsigned num_pd_entries;
-	struct page **pt_pages;
-	uint32_t pd_offset;
-	dma_addr_t *pt_dma_addr;
-
+	union {
+		struct page **pt_pages;
+		struct page *gen8_pt_pages;
+	};
+	struct page *pd_pages;
+	int num_pd_pages;
+	int num_pt_pages;
+	union {
+		uint32_t pd_offset;
+		dma_addr_t pd_dma_addr[4];
+	};
+	union {
+		dma_addr_t *pt_dma_addr;
+		dma_addr_t *gen8_pt_dma_addr[4];
+	};
 	int (*enable)(struct drm_device *dev);
 };
 
@@ -570,6 +618,13 @@ struct i915_vma {
 	/** This vma's place in the batchbuffer or on the eviction list */
 	struct list_head exec_list;
 
+	/**
+	 * Used for performing relocations during execbuffer insertion.
+	 */
+	struct hlist_node exec_node;
+	unsigned long exec_handle;
+	struct drm_i915_gem_exec_object2 *exec_entry;
+
 };
 
 struct i915_ctx_hang_stats {
@@ -578,6 +633,12 @@ struct i915_ctx_hang_stats {
 
 	/* This context had batch active when hang was declared */
 	unsigned batch_active;
+
+	/* Time when this context was last blamed for a GPU reset */
+	unsigned long guilty_ts;
+
+	/* This context is banned to submit more work */
+	bool banned;
 };
 
 /* This must match up with the value previously used for execbuf2.rsvd1. */
@@ -586,10 +647,13 @@ struct i915_hw_context {
 	struct kref ref;
 	int id;
 	bool is_initialized;
+	uint8_t remap_slice;
 	struct drm_i915_file_private *file_priv;
 	struct intel_ring_buffer *ring;
 	struct drm_i915_gem_object *obj;
 	struct i915_ctx_hang_stats hang_stats;
+
+	struct list_head link;
 };
 
 struct i915_fbc {
@@ -623,17 +687,9 @@ struct i915_fbc {
 	} no_fbc_reason;
 };
 
-enum no_psr_reason {
-	PSR_NO_SOURCE, /* Not supported on platform */
-	PSR_NO_SINK, /* Not supported by panel */
-	PSR_MODULE_PARAM,
-	PSR_CRTC_NOT_ACTIVE,
-	PSR_PWR_WELL_ENABLED,
-	PSR_NOT_TILED,
-	PSR_SPRITE_ENABLED,
-	PSR_S3D_ENABLED,
-	PSR_INTERLACED_ENABLED,
-	PSR_HSW_NOT_DDIA,
+struct i915_psr {
+	bool sink_support;
+	bool source_ok;
 };
 
 enum intel_pch {
@@ -704,6 +760,9 @@ struct i915_suspend_saved_registers {
 	u32 saveBLC_HIST_CTL;
 	u32 saveBLC_PWM_CTL;
 	u32 saveBLC_PWM_CTL2;
+	u32 saveBLC_HIST_CTL_B;
+	u32 saveBLC_PWM_CTL_B;
+	u32 saveBLC_PWM_CTL2_B;
 	u32 saveBLC_CPU_PWM_CTL;
 	u32 saveBLC_CPU_PWM_CTL2;
 	u32 saveFPB0;
@@ -823,17 +882,20 @@ struct intel_gen6_power_mgmt {
 	struct work_struct work;
 	u32 pm_iir;
 
-	/* On vlv we need to manually drop to Vmin with a delayed work. */
-	struct delayed_work vlv_work;
-
 	/* The below variables an all the rps hw state are protected by
 	 * dev->struct mutext. */
 	u8 cur_delay;
 	u8 min_delay;
 	u8 max_delay;
 	u8 rpe_delay;
+	u8 rp1_delay;
+	u8 rp0_delay;
 	u8 hw_max;
 
+	int last_adj;
+	enum { LOW_POWER, BETWEEN, HIGH_POWER } power;
+
+	bool enabled;
 	struct delayed_work delayed_resume_work;
 
 	/*
@@ -870,11 +932,21 @@ struct intel_ilk_power_mgmt {
 
 /* Power well structure for haswell */
 struct i915_power_well {
-	struct drm_device *device;
-	spinlock_t lock;
 	/* power well enable/disable usage count */
 	int count;
-	int i915_request;
+};
+
+#define I915_MAX_POWER_WELLS 1
+
+struct i915_power_domains {
+	/*
+	 * Power wells needed for initialization at driver init and suspend
+	 * time are on. They are kept on until after the first modeset.
+	 */
+	bool init_power_on;
+
+	struct mutex lock;
+	struct i915_power_well power_wells[I915_MAX_POWER_WELLS];
 };
 
 struct i915_dri1_state {
@@ -902,9 +974,11 @@ struct i915_ums_state {
 	int mm_suspended;
 };
 
+#define MAX_L3_SLICES 2
 struct intel_l3_parity {
-	u32 *remap_info;
+	u32 *remap_info[MAX_L3_SLICES];
 	struct work_struct error_work;
+	int which_slice;
 };
 
 struct i915_gem_mm {
@@ -942,6 +1016,15 @@ struct i915_gem_mm {
 	struct delayed_work retire_work;
 
 	/**
+	 * When we detect an idle GPU, we want to turn on
+	 * powersaving features. So once we see that there
+	 * are no more requests outstanding and no more
+	 * arrive within a small period of time, we fire
+	 * off the idle_work.
+	 */
+	struct delayed_work idle_work;
+
+	/**
 	 * Are we in a non-interruptible section of code like
 	 * modesetting?
 	 */
@@ -979,6 +1062,9 @@ struct i915_gpu_error {
 	/* For hangcheck timer */
 #define DRM_I915_HANGCHECK_PERIOD 1500 /* in ms */
 #define DRM_I915_HANGCHECK_JIFFIES msecs_to_jiffies(DRM_I915_HANGCHECK_PERIOD)
+	/* Hang gpu twice in this window and your context gets banned */
+#define DRM_I915_CTX_BAN_PERIOD DIV_ROUND_UP(8*DRM_I915_HANGCHECK_PERIOD, 1000)
+
 	struct timer_list hangcheck_timer;
 
 	/* For reset and error_state handling. */
@@ -987,7 +1073,8 @@ struct i915_gpu_error {
 	struct drm_i915_error_state *first_error;
 	struct work_struct work;
 
-	unsigned long last_reset;
+
+	unsigned long missed_irq_rings;
 
 	/**
 	 * State variable and reset counter controlling the reset flow
@@ -1027,6 +1114,9 @@ struct i915_gpu_error {
 
 	/* For gpu hang simulation. */
 	unsigned int stop_rings;
+
+	/* For missed irq/seqno simulation. */
+	unsigned int test_irq_rings;
 };
 
 enum modeset_restore {
@@ -1035,6 +1125,14 @@ enum modeset_restore {
 	MODESET_SUSPENDED,
 };
 
+struct ddi_vbt_port_info {
+	uint8_t hdmi_level_shift;
+
+	uint8_t supports_dvi:1;
+	uint8_t supports_hdmi:1;
+	uint8_t supports_dp:1;
+};
+
 struct intel_vbt_data {
 	struct drm_display_mode *lfp_lvds_vbt_mode; /* if any */
 	struct drm_display_mode *sdvo_lvds_vbt_mode; /* if any */
@@ -1060,10 +1158,17 @@ struct intel_vbt_data {
 	int edp_bpp;
 	struct edp_power_seq edp_pps;
 
+	/* MIPI DSI */
+	struct {
+		u16 panel_id;
+	} dsi;
+
 	int crt_ddc_pin;
 
 	int child_dev_num;
-	struct child_device_config *child_dev;
+	union child_device_config *child_dev;
+
+	struct ddi_vbt_port_info ddi_port_info[I915_MAX_PORTS];
 };
 
 enum intel_ddb_partitioning {
@@ -1079,6 +1184,15 @@ struct intel_wm_level {
 	uint32_t fbc_val;
 };
 
+struct hsw_wm_values {
+	uint32_t wm_pipe[3];
+	uint32_t wm_lp[3];
+	uint32_t wm_lp_spr[3];
+	uint32_t wm_linetime[3];
+	bool enable_fbc_wm;
+	enum intel_ddb_partitioning partitioning;
+};
+
 /*
  * This struct tracks the state needed for the Package C8+ feature.
  *
@@ -1148,6 +1262,36 @@ struct i915_package_c8 {
 	} regsave;
 };
 
+enum intel_pipe_crc_source {
+	INTEL_PIPE_CRC_SOURCE_NONE,
+	INTEL_PIPE_CRC_SOURCE_PLANE1,
+	INTEL_PIPE_CRC_SOURCE_PLANE2,
+	INTEL_PIPE_CRC_SOURCE_PF,
+	INTEL_PIPE_CRC_SOURCE_PIPE,
+	/* TV/DP on pre-gen5/vlv can't use the pipe source. */
+	INTEL_PIPE_CRC_SOURCE_TV,
+	INTEL_PIPE_CRC_SOURCE_DP_B,
+	INTEL_PIPE_CRC_SOURCE_DP_C,
+	INTEL_PIPE_CRC_SOURCE_DP_D,
+	INTEL_PIPE_CRC_SOURCE_AUTO,
+	INTEL_PIPE_CRC_SOURCE_MAX,
+};
+
+struct intel_pipe_crc_entry {
+	uint32_t frame;
+	uint32_t crc[5];
+};
+
+#define INTEL_PIPE_CRC_ENTRIES_NR	128
+struct intel_pipe_crc {
+	spinlock_t lock;
+	bool opened;		/* exclusive access to the result file */
+	struct intel_pipe_crc_entry *entries;
+	enum intel_pipe_crc_source source;
+	int head, tail;
+	wait_queue_head_t wq;
+};
+
 typedef struct drm_i915_private {
 	struct drm_device *dev;
 	struct kmem_cache *slab;
@@ -1193,7 +1337,10 @@ typedef struct drm_i915_private {
 	struct mutex dpio_lock;
 
 	/** Cached value of IMR to avoid reads in updating the bitfield */
-	u32 irq_mask;
+	union {
+		u32 irq_mask;
+		u32 de_irq_mask[I915_MAX_PIPES];
+	};
 	u32 gt_irq_mask;
 	u32 pm_irq_mask;
 
@@ -1272,6 +1419,10 @@ typedef struct drm_i915_private {
 	struct drm_crtc *pipe_to_crtc_mapping[3];
 	wait_queue_head_t pending_flip_queue;
 
+#ifdef CONFIG_DEBUG_FS
+	struct intel_pipe_crc pipe_crc[I915_MAX_PIPES];
+#endif
+
 	int num_shared_dpll;
 	struct intel_shared_dpll shared_dplls[I915_NUM_PLLS];
 	struct intel_ddi_plls ddi_plls;
@@ -1297,17 +1448,18 @@ typedef struct drm_i915_private {
 	 * mchdev_lock in intel_pm.c */
 	struct intel_ilk_power_mgmt ips;
 
-	/* Haswell power well */
-	struct i915_power_well power_well;
+	struct i915_power_domains power_domains;
 
-	enum no_psr_reason no_psr_reason;
+	struct i915_psr psr;
 
 	struct i915_gpu_error gpu_error;
 
 	struct drm_i915_gem_object *vlv_pctx;
 
+#ifdef CONFIG_DRM_I915_FBDEV
 	/* list of fbdev register on this device */
 	struct intel_fbdev *fbdev;
+#endif
 
 	/*
 	 * The console may be contended at resume, but we don't
@@ -1320,6 +1472,7 @@ typedef struct drm_i915_private {
 
 	bool hw_contexts_disabled;
 	uint32_t hw_context_size;
+	struct list_head context_list;
 
 	u32 fdi_rx_config;
 
@@ -1337,6 +1490,9 @@ typedef struct drm_i915_private {
 		uint16_t spr_latency[5];
 		/* cursor */
 		uint16_t cur_latency[5];
+
+		/* current hardware state */
+		struct hsw_wm_values hw;
 	} wm;
 
 	struct i915_package_c8 pc8;
@@ -1400,8 +1556,6 @@ struct drm_i915_gem_object {
 	struct list_head ring_list;
 	/** Used in execbuf to temporarily hold a ref */
 	struct list_head obj_exec_link;
-	/** This object's place in the batchbuffer or on the eviction list */
-	struct list_head exec_list;
 
 	/**
 	 * This is set if the object is on the active lists (has pending
@@ -1487,13 +1641,6 @@ struct drm_i915_gem_object {
 	void *dma_buf_vmapping;
 	int vmapping_count;
 
-	/**
-	 * Used for performing relocations during execbuffer insertion.
-	 */
-	struct hlist_node exec_node;
-	unsigned long exec_handle;
-	struct drm_i915_gem_exec_object2 *exec_entry;
-
 	struct intel_ring_buffer *ring;
 
 	/** Breadcrumb of last rendering to the buffer. */
@@ -1505,11 +1652,14 @@ struct drm_i915_gem_object {
 	/** Current tiling stride for the object, if it's tiled. */
 	uint32_t stride;
 
+	/** References from framebuffers, locks out tiling changes. */
+	unsigned long framebuffer_references;
+
 	/** Record of address bit 17 of each page at last unbind. */
 	unsigned long *bit_17;
 
 	/** User space pin count and filp owning the pin */
-	uint32_t user_pin_count;
+	unsigned long user_pin_count;
 	struct drm_file *pin_filp;
 
 	/** for phy allocated objects */
@@ -1560,48 +1710,56 @@ struct drm_i915_gem_request {
 };
 
 struct drm_i915_file_private {
+	struct drm_i915_private *dev_priv;
+
 	struct {
 		spinlock_t lock;
 		struct list_head request_list;
+		struct delayed_work idle_work;
 	} mm;
 	struct idr context_idr;
 
 	struct i915_ctx_hang_stats hang_stats;
+	atomic_t rps_wait_boost;
 };
 
 #define INTEL_INFO(dev)	(to_i915(dev)->info)
 
-#define IS_I830(dev)		((dev)->pci_device == 0x3577)
-#define IS_845G(dev)		((dev)->pci_device == 0x2562)
+#define IS_I830(dev)		((dev)->pdev->device == 0x3577)
+#define IS_845G(dev)		((dev)->pdev->device == 0x2562)
 #define IS_I85X(dev)		(INTEL_INFO(dev)->is_i85x)
-#define IS_I865G(dev)		((dev)->pci_device == 0x2572)
+#define IS_I865G(dev)		((dev)->pdev->device == 0x2572)
 #define IS_I915G(dev)		(INTEL_INFO(dev)->is_i915g)
-#define IS_I915GM(dev)		((dev)->pci_device == 0x2592)
-#define IS_I945G(dev)		((dev)->pci_device == 0x2772)
+#define IS_I915GM(dev)		((dev)->pdev->device == 0x2592)
+#define IS_I945G(dev)		((dev)->pdev->device == 0x2772)
 #define IS_I945GM(dev)		(INTEL_INFO(dev)->is_i945gm)
 #define IS_BROADWATER(dev)	(INTEL_INFO(dev)->is_broadwater)
 #define IS_CRESTLINE(dev)	(INTEL_INFO(dev)->is_crestline)
-#define IS_GM45(dev)		((dev)->pci_device == 0x2A42)
+#define IS_GM45(dev)		((dev)->pdev->device == 0x2A42)
 #define IS_G4X(dev)		(INTEL_INFO(dev)->is_g4x)
-#define IS_PINEVIEW_G(dev)	((dev)->pci_device == 0xa001)
-#define IS_PINEVIEW_M(dev)	((dev)->pci_device == 0xa011)
+#define IS_PINEVIEW_G(dev)	((dev)->pdev->device == 0xa001)
+#define IS_PINEVIEW_M(dev)	((dev)->pdev->device == 0xa011)
 #define IS_PINEVIEW(dev)	(INTEL_INFO(dev)->is_pineview)
 #define IS_G33(dev)		(INTEL_INFO(dev)->is_g33)
-#define IS_IRONLAKE_M(dev)	((dev)->pci_device == 0x0046)
+#define IS_IRONLAKE_M(dev)	((dev)->pdev->device == 0x0046)
 #define IS_IVYBRIDGE(dev)	(INTEL_INFO(dev)->is_ivybridge)
-#define IS_IVB_GT1(dev)		((dev)->pci_device == 0x0156 || \
-				 (dev)->pci_device == 0x0152 ||	\
-				 (dev)->pci_device == 0x015a)
-#define IS_SNB_GT1(dev)		((dev)->pci_device == 0x0102 || \
-				 (dev)->pci_device == 0x0106 ||	\
-				 (dev)->pci_device == 0x010A)
+#define IS_IVB_GT1(dev)		((dev)->pdev->device == 0x0156 || \
+				 (dev)->pdev->device == 0x0152 || \
+				 (dev)->pdev->device == 0x015a)
+#define IS_SNB_GT1(dev)		((dev)->pdev->device == 0x0102 || \
+				 (dev)->pdev->device == 0x0106 || \
+				 (dev)->pdev->device == 0x010A)
 #define IS_VALLEYVIEW(dev)	(INTEL_INFO(dev)->is_valleyview)
 #define IS_HASWELL(dev)	(INTEL_INFO(dev)->is_haswell)
+#define IS_BROADWELL(dev)	(INTEL_INFO(dev)->gen == 8)
 #define IS_MOBILE(dev)		(INTEL_INFO(dev)->is_mobile)
 #define IS_HSW_EARLY_SDV(dev)	(IS_HASWELL(dev) && \
-				 ((dev)->pci_device & 0xFF00) == 0x0C00)
+				 ((dev)->pdev->device & 0xFF00) == 0x0C00)
 #define IS_ULT(dev)		(IS_HASWELL(dev) && \
-				 ((dev)->pci_device & 0xFF00) == 0x0A00)
+				 ((dev)->pdev->device & 0xFF00) == 0x0A00)
+#define IS_HSW_GT3(dev)		(IS_HASWELL(dev) && \
+				 ((dev)->pdev->device & 0x00F0) == 0x0020)
+#define IS_PRELIMINARY_HW(intel_info) ((intel_info)->is_preliminary)
 
 /*
  * The genX designation typically refers to the render engine, so render
@@ -1615,10 +1773,15 @@ struct drm_i915_file_private {
 #define IS_GEN5(dev)	(INTEL_INFO(dev)->gen == 5)
 #define IS_GEN6(dev)	(INTEL_INFO(dev)->gen == 6)
 #define IS_GEN7(dev)	(INTEL_INFO(dev)->gen == 7)
-
-#define HAS_BSD(dev)            (INTEL_INFO(dev)->has_bsd_ring)
-#define HAS_BLT(dev)            (INTEL_INFO(dev)->has_blt_ring)
-#define HAS_VEBOX(dev)          (INTEL_INFO(dev)->has_vebox_ring)
+#define IS_GEN8(dev)	(INTEL_INFO(dev)->gen == 8)
+
+#define RENDER_RING		(1<<RCS)
+#define BSD_RING		(1<<VCS)
+#define BLT_RING		(1<<BCS)
+#define VEBOX_RING		(1<<VECS)
+#define HAS_BSD(dev)            (INTEL_INFO(dev)->ring_mask & BSD_RING)
+#define HAS_BLT(dev)            (INTEL_INFO(dev)->ring_mask & BLT_RING)
+#define HAS_VEBOX(dev)            (INTEL_INFO(dev)->ring_mask & VEBOX_RING)
 #define HAS_LLC(dev)            (INTEL_INFO(dev)->has_llc)
 #define HAS_WT(dev)            (IS_HASWELL(dev) && to_i915(dev)->ellc_size)
 #define I915_NEED_GFX_HWS(dev)	(INTEL_INFO(dev)->need_gfx_hws)
@@ -1640,7 +1803,6 @@ struct drm_i915_file_private {
 #define SUPPORTS_DIGITAL_OUTPUTS(dev)	(!IS_GEN2(dev) && !IS_PINEVIEW(dev))
 #define SUPPORTS_INTEGRATED_HDMI(dev)	(IS_G4X(dev) || IS_GEN5(dev))
 #define SUPPORTS_INTEGRATED_DP(dev)	(IS_G4X(dev) || IS_GEN5(dev))
-#define SUPPORTS_EDP(dev)		(IS_IRONLAKE_M(dev))
 #define SUPPORTS_TV(dev)		(INTEL_INFO(dev)->supports_tv)
 #define I915_HAS_HOTPLUG(dev)		 (INTEL_INFO(dev)->has_hotplug)
 
@@ -1648,11 +1810,12 @@ struct drm_i915_file_private {
 #define HAS_PIPE_CXSR(dev) (INTEL_INFO(dev)->has_pipe_cxsr)
 #define I915_HAS_FBC(dev) (INTEL_INFO(dev)->has_fbc)
 
-#define HAS_IPS(dev)		(IS_ULT(dev))
+#define HAS_IPS(dev)		(IS_ULT(dev) || IS_BROADWELL(dev))
 
 #define HAS_DDI(dev)		(INTEL_INFO(dev)->has_ddi)
-#define HAS_POWER_WELL(dev)	(IS_HASWELL(dev))
+#define HAS_POWER_WELL(dev)	(IS_HASWELL(dev) || IS_BROADWELL(dev))
 #define HAS_FPGA_DBG_UNCLAIMED(dev)	(INTEL_INFO(dev)->has_fpga_dbg)
+#define HAS_PSR(dev)		(IS_HASWELL(dev) || IS_BROADWELL(dev))
 
 #define INTEL_PCH_DEVICE_ID_MASK		0xff00
 #define INTEL_PCH_IBX_DEVICE_ID_TYPE		0x3b00
@@ -1668,35 +1831,14 @@ struct drm_i915_file_private {
 #define HAS_PCH_NOP(dev) (INTEL_PCH_TYPE(dev) == PCH_NOP)
 #define HAS_PCH_SPLIT(dev) (INTEL_PCH_TYPE(dev) != PCH_NONE)
 
-#define HAS_FORCE_WAKE(dev) (INTEL_INFO(dev)->has_force_wake)
-
-#define HAS_L3_GPU_CACHE(dev) (IS_IVYBRIDGE(dev) || IS_HASWELL(dev))
+/* DPF == dynamic parity feature */
+#define HAS_L3_DPF(dev) (IS_IVYBRIDGE(dev) || IS_HASWELL(dev))
+#define NUM_L3_SLICES(dev) (IS_HSW_GT3(dev) ? 2 : HAS_L3_DPF(dev))
 
 #define GT_FREQUENCY_MULTIPLIER 50
 
 #include "i915_trace.h"
 
-/**
- * RC6 is a special power stage which allows the GPU to enter an very
- * low-voltage mode when idle, using down to 0V while at this stage.  This
- * stage is entered automatically when the GPU is idle when RC6 support is
- * enabled, and as soon as new workload arises GPU wakes up automatically as well.
- *
- * There are different RC6 modes available in Intel GPU, which differentiate
- * among each other with the latency required to enter and leave RC6 and
- * voltage consumed by the GPU in different states.
- *
- * The combination of the following flags define which states GPU is allowed
- * to enter, while RC6 is the normal RC6 state, RC6p is the deep RC6, and
- * RC6pp is deepest RC6. Their support by hardware varies according to the
- * GPU, BIOS, chipset and platform. RC6 is usually the safest one and the one
- * which brings the most power savings; deeper states save more power, but
- * require higher latency to switch to and wake up.
- */
-#define INTEL_RC6_ENABLE			(1<<0)
-#define INTEL_RC6p_ENABLE			(1<<1)
-#define INTEL_RC6pp_ENABLE			(1<<2)
-
 extern const struct drm_ioctl_desc i915_ioctls[];
 extern int i915_max_ioctl;
 extern unsigned int i915_fbpercrtc __always_unused;
@@ -1767,12 +1909,13 @@ extern void intel_uncore_early_sanitize(struct drm_device *dev);
 extern void intel_uncore_init(struct drm_device *dev);
 extern void intel_uncore_clear_errors(struct drm_device *dev);
 extern void intel_uncore_check_errors(struct drm_device *dev);
+extern void intel_uncore_fini(struct drm_device *dev);
 
 void
-i915_enable_pipestat(drm_i915_private_t *dev_priv, int pipe, u32 mask);
+i915_enable_pipestat(drm_i915_private_t *dev_priv, enum pipe pipe, u32 mask);
 
 void
-i915_disable_pipestat(drm_i915_private_t *dev_priv, int pipe, u32 mask);
+i915_disable_pipestat(drm_i915_private_t *dev_priv, enum pipe pipe, u32 mask);
 
 /* i915_gem.c */
 int i915_gem_init_ioctl(struct drm_device *dev, void *data,
@@ -1824,14 +1967,11 @@ int i915_gem_wait_ioctl(struct drm_device *dev, void *data,
 void i915_gem_load(struct drm_device *dev);
 void *i915_gem_object_alloc(struct drm_device *dev);
 void i915_gem_object_free(struct drm_i915_gem_object *obj);
-int i915_gem_init_object(struct drm_gem_object *obj);
 void i915_gem_object_init(struct drm_i915_gem_object *obj,
 			 const struct drm_i915_gem_object_ops *ops);
 struct drm_i915_gem_object *i915_gem_alloc_object(struct drm_device *dev,
 						  size_t size);
 void i915_gem_free_object(struct drm_gem_object *obj);
-struct i915_vma *i915_gem_vma_create(struct drm_i915_gem_object *obj,
-				     struct i915_address_space *vm);
 void i915_gem_vma_destroy(struct i915_vma *vma);
 
 int __must_check i915_gem_object_pin(struct drm_i915_gem_object *obj,
@@ -1870,9 +2010,8 @@ static inline void i915_gem_object_unpin_pages(struct drm_i915_gem_object *obj)
 int __must_check i915_mutex_lock_interruptible(struct drm_device *dev);
 int i915_gem_object_sync(struct drm_i915_gem_object *obj,
 			 struct intel_ring_buffer *to);
-void i915_gem_object_move_to_active(struct drm_i915_gem_object *obj,
-				    struct intel_ring_buffer *ring);
-
+void i915_vma_move_to_active(struct i915_vma *vma,
+			     struct intel_ring_buffer *ring);
 int i915_gem_dumb_create(struct drm_file *file_priv,
 			 struct drm_device *dev,
 			 struct drm_mode_create_dumb *args);
@@ -1913,7 +2052,7 @@ i915_gem_object_unpin_fence(struct drm_i915_gem_object *obj)
 	}
 }
 
-void i915_gem_retire_requests(struct drm_device *dev);
+bool i915_gem_retire_requests(struct drm_device *dev);
 void i915_gem_retire_requests_ring(struct intel_ring_buffer *ring);
 int __must_check i915_gem_check_wedge(struct i915_gpu_error *error,
 				      bool interruptible);
@@ -1933,11 +2072,11 @@ bool i915_gem_clflush_object(struct drm_i915_gem_object *obj, bool force);
 int __must_check i915_gem_object_finish_gpu(struct drm_i915_gem_object *obj);
 int __must_check i915_gem_init(struct drm_device *dev);
 int __must_check i915_gem_init_hw(struct drm_device *dev);
-void i915_gem_l3_remap(struct drm_device *dev);
+int i915_gem_l3_remap(struct intel_ring_buffer *ring, int slice);
 void i915_gem_init_swizzling(struct drm_device *dev);
 void i915_gem_cleanup_ringbuffer(struct drm_device *dev);
 int __must_check i915_gpu_idle(struct drm_device *dev);
-int __must_check i915_gem_idle(struct drm_device *dev);
+int __must_check i915_gem_suspend(struct drm_device *dev);
 int __i915_add_request(struct intel_ring_buffer *ring,
 		       struct drm_file *file,
 		       struct drm_i915_gem_object *batch_obj,
@@ -1964,6 +2103,7 @@ int i915_gem_attach_phys_object(struct drm_device *dev,
 void i915_gem_detach_phys_object(struct drm_device *dev,
 				 struct drm_i915_gem_object *obj);
 void i915_gem_free_all_phys_object(struct drm_device *dev);
+int i915_gem_open(struct drm_device *dev, struct drm_file *file);
 void i915_gem_release(struct drm_device *dev, struct drm_file *file);
 
 uint32_t
@@ -1995,6 +2135,9 @@ struct i915_vma *i915_gem_obj_to_vma(struct drm_i915_gem_object *obj,
 struct i915_vma *
 i915_gem_obj_lookup_or_create_vma(struct drm_i915_gem_object *obj,
 				  struct i915_address_space *vm);
+
+struct i915_vma *i915_gem_obj_to_ggtt(struct drm_i915_gem_object *obj);
+
 /* Some GGTT VM helpers */
 #define obj_to_ggtt(obj) \
 	(&((struct drm_i915_private *)(obj)->base.dev->dev_private)->gtt.base)
@@ -2031,7 +2174,6 @@ i915_gem_obj_ggtt_pin(struct drm_i915_gem_object *obj,
 	return i915_gem_object_pin(obj, obj_to_ggtt(obj), alignment,
 				   map_and_fenceable, nonblocking);
 }
-#undef obj_to_ggtt
 
 /* i915_gem_context.c */
 void i915_gem_context_init(struct drm_device *dev);
@@ -2094,6 +2236,7 @@ int __must_check i915_gem_evict_something(struct drm_device *dev,
 					  unsigned cache_level,
 					  bool mappable,
 					  bool nonblock);
+int i915_gem_evict_vm(struct i915_address_space *vm, bool do_idle);
 int i915_gem_evict_everything(struct drm_device *dev);
 
 /* i915_gem_stolen.c */
@@ -2133,6 +2276,11 @@ int i915_verify_lists(struct drm_device *dev);
 /* i915_debugfs.c */
 int i915_debugfs_init(struct drm_minor *minor);
 void i915_debugfs_cleanup(struct drm_minor *minor);
+#ifdef CONFIG_DEBUG_FS
+void intel_display_crc_init(struct drm_device *dev);
+#else
+static inline void intel_display_crc_init(struct drm_device *dev) {}
+#endif
 
 /* i915_gpu_error.c */
 __printf(2, 3)
@@ -2186,15 +2334,30 @@ static inline bool intel_gmbus_is_forced_bit(struct i2c_adapter *adapter)
 extern void intel_i2c_reset(struct drm_device *dev);
 
 /* intel_opregion.c */
+struct intel_encoder;
 extern int intel_opregion_setup(struct drm_device *dev);
 #ifdef CONFIG_ACPI
 extern void intel_opregion_init(struct drm_device *dev);
 extern void intel_opregion_fini(struct drm_device *dev);
 extern void intel_opregion_asle_intr(struct drm_device *dev);
+extern int intel_opregion_notify_encoder(struct intel_encoder *intel_encoder,
+					 bool enable);
+extern int intel_opregion_notify_adapter(struct drm_device *dev,
+					 pci_power_t state);
 #else
 static inline void intel_opregion_init(struct drm_device *dev) { return; }
 static inline void intel_opregion_fini(struct drm_device *dev) { return; }
 static inline void intel_opregion_asle_intr(struct drm_device *dev) { return; }
+static inline int
+intel_opregion_notify_encoder(struct intel_encoder *intel_encoder, bool enable)
+{
+	return 0;
+}
+static inline int
+intel_opregion_notify_adapter(struct drm_device *dev, pci_power_t state)
+{
+	return 0;
+}
 #endif
 
 /* intel_acpi.c */
@@ -2256,8 +2419,16 @@ int sandybridge_pcode_write(struct drm_i915_private *dev_priv, u8 mbox, u32 val)
 u32 vlv_punit_read(struct drm_i915_private *dev_priv, u8 addr);
 void vlv_punit_write(struct drm_i915_private *dev_priv, u8 addr, u32 val);
 u32 vlv_nc_read(struct drm_i915_private *dev_priv, u8 addr);
-u32 vlv_dpio_read(struct drm_i915_private *dev_priv, int reg);
-void vlv_dpio_write(struct drm_i915_private *dev_priv, int reg, u32 val);
+u32 vlv_gpio_nc_read(struct drm_i915_private *dev_priv, u32 reg);
+void vlv_gpio_nc_write(struct drm_i915_private *dev_priv, u32 reg, u32 val);
+u32 vlv_cck_read(struct drm_i915_private *dev_priv, u32 reg);
+void vlv_cck_write(struct drm_i915_private *dev_priv, u32 reg, u32 val);
+u32 vlv_ccu_read(struct drm_i915_private *dev_priv, u32 reg);
+void vlv_ccu_write(struct drm_i915_private *dev_priv, u32 reg, u32 val);
+u32 vlv_gps_core_read(struct drm_i915_private *dev_priv, u32 reg);
+void vlv_gps_core_write(struct drm_i915_private *dev_priv, u32 reg, u32 val);
+u32 vlv_dpio_read(struct drm_i915_private *dev_priv, enum pipe pipe, int reg);
+void vlv_dpio_write(struct drm_i915_private *dev_priv, enum pipe pipe, int reg, u32 val);
 u32 intel_sbi_read(struct drm_i915_private *dev_priv, u16 reg,
 		   enum intel_sbi_destination destination);
 void intel_sbi_write(struct drm_i915_private *dev_priv, u16 reg, u32 value,
@@ -2266,37 +2437,21 @@ void intel_sbi_write(struct drm_i915_private *dev_priv, u16 reg, u32 value,
 int vlv_gpu_freq(int ddr_freq, int val);
 int vlv_freq_opcode(int ddr_freq, int val);
 
-#define __i915_read(x) \
-	u##x i915_read##x(struct drm_i915_private *dev_priv, u32 reg, bool trace);
-__i915_read(8)
-__i915_read(16)
-__i915_read(32)
-__i915_read(64)
-#undef __i915_read
-
-#define __i915_write(x) \
-	void i915_write##x(struct drm_i915_private *dev_priv, u32 reg, u##x val, bool trace);
-__i915_write(8)
-__i915_write(16)
-__i915_write(32)
-__i915_write(64)
-#undef __i915_write
-
-#define I915_READ8(reg)		i915_read8(dev_priv, (reg), true)
-#define I915_WRITE8(reg, val)	i915_write8(dev_priv, (reg), (val), true)
-
-#define I915_READ16(reg)	i915_read16(dev_priv, (reg), true)
-#define I915_WRITE16(reg, val)	i915_write16(dev_priv, (reg), (val), true)
-#define I915_READ16_NOTRACE(reg)	i915_read16(dev_priv, (reg), false)
-#define I915_WRITE16_NOTRACE(reg, val)	i915_write16(dev_priv, (reg), (val), false)
-
-#define I915_READ(reg)		i915_read32(dev_priv, (reg), true)
-#define I915_WRITE(reg, val)	i915_write32(dev_priv, (reg), (val), true)
-#define I915_READ_NOTRACE(reg)		i915_read32(dev_priv, (reg), false)
-#define I915_WRITE_NOTRACE(reg, val)	i915_write32(dev_priv, (reg), (val), false)
-
-#define I915_WRITE64(reg, val)	i915_write64(dev_priv, (reg), (val), true)
-#define I915_READ64(reg)	i915_read64(dev_priv, (reg), true)
+#define I915_READ8(reg)		dev_priv->uncore.funcs.mmio_readb(dev_priv, (reg), true)
+#define I915_WRITE8(reg, val)	dev_priv->uncore.funcs.mmio_writeb(dev_priv, (reg), (val), true)
+
+#define I915_READ16(reg)	dev_priv->uncore.funcs.mmio_readw(dev_priv, (reg), true)
+#define I915_WRITE16(reg, val)	dev_priv->uncore.funcs.mmio_writew(dev_priv, (reg), (val), true)
+#define I915_READ16_NOTRACE(reg)	dev_priv->uncore.funcs.mmio_readw(dev_priv, (reg), false)
+#define I915_WRITE16_NOTRACE(reg, val)	dev_priv->uncore.funcs.mmio_writew(dev_priv, (reg), (val), false)
+
+#define I915_READ(reg)		dev_priv->uncore.funcs.mmio_readl(dev_priv, (reg), true)
+#define I915_WRITE(reg, val)	dev_priv->uncore.funcs.mmio_writel(dev_priv, (reg), (val), true)
+#define I915_READ_NOTRACE(reg)		dev_priv->uncore.funcs.mmio_readl(dev_priv, (reg), false)
+#define I915_WRITE_NOTRACE(reg, val)	dev_priv->uncore.funcs.mmio_writel(dev_priv, (reg), (val), false)
+
+#define I915_WRITE64(reg, val)	dev_priv->uncore.funcs.mmio_writeq(dev_priv, (reg), (val), true)
+#define I915_READ64(reg)	dev_priv->uncore.funcs.mmio_readq(dev_priv, (reg), true)
 
 #define POSTING_READ(reg)	(void)I915_READ_NOTRACE(reg)
 #define POSTING_READ16(reg)	(void)I915_READ16_NOTRACE(reg)
diff --git a/drivers/gpu/drm/i915/i915_gem.c b/drivers/gpu/drm/i915/i915_gem.c
index cdfb9da0e4ce..12bbd5eac70d 100644
--- a/drivers/gpu/drm/i915/i915_gem.c
+++ b/drivers/gpu/drm/i915/i915_gem.c
@@ -41,6 +41,9 @@ static void i915_gem_object_flush_gtt_write_domain(struct drm_i915_gem_object *o
 static void i915_gem_object_flush_cpu_write_domain(struct drm_i915_gem_object *obj,
 						   bool force);
 static __must_check int
+i915_gem_object_wait_rendering(struct drm_i915_gem_object *obj,
+			       bool readonly);
+static __must_check int
 i915_gem_object_bind_to_vm(struct drm_i915_gem_object *obj,
 			   struct i915_address_space *vm,
 			   unsigned alignment,
@@ -61,8 +64,8 @@ static unsigned long i915_gem_inactive_count(struct shrinker *shrinker,
 					     struct shrink_control *sc);
 static unsigned long i915_gem_inactive_scan(struct shrinker *shrinker,
 					    struct shrink_control *sc);
-static long i915_gem_purge(struct drm_i915_private *dev_priv, long target);
-static long i915_gem_shrink_all(struct drm_i915_private *dev_priv);
+static unsigned long i915_gem_purge(struct drm_i915_private *dev_priv, long target);
+static unsigned long i915_gem_shrink_all(struct drm_i915_private *dev_priv);
 static void i915_gem_object_truncate(struct drm_i915_gem_object *obj);
 
 static bool cpu_cache_is_coherent(struct drm_device *dev,
@@ -258,7 +261,7 @@ i915_gem_dumb_create(struct drm_file *file,
 		     struct drm_mode_create_dumb *args)
 {
 	/* have to work out size/pitch and return them */
-	args->pitch = ALIGN(args->width * ((args->bpp + 7) / 8), 64);
+	args->pitch = ALIGN(args->width * DIV_ROUND_UP(args->bpp, 8), 64);
 	args->size = args->pitch * args->height;
 	return i915_gem_create(file, dev,
 			       args->size, &args->handle);
@@ -432,11 +435,9 @@ i915_gem_shmem_pread(struct drm_device *dev,
 		 * optimizes for the case when the gpu will dirty the data
 		 * anyway again before the next pread happens. */
 		needs_clflush = !cpu_cache_is_coherent(dev, obj->cache_level);
-		if (i915_gem_obj_bound_any(obj)) {
-			ret = i915_gem_object_set_to_gtt_domain(obj, false);
-			if (ret)
-				return ret;
-		}
+		ret = i915_gem_object_wait_rendering(obj, true);
+		if (ret)
+			return ret;
 	}
 
 	ret = i915_gem_object_get_pages(obj);
@@ -748,11 +749,9 @@ i915_gem_shmem_pwrite(struct drm_device *dev,
 		 * optimizes for the case when the gpu will use the data
 		 * right away and we therefore have to clflush anyway. */
 		needs_clflush_after = cpu_write_needs_clflush(obj);
-		if (i915_gem_obj_bound_any(obj)) {
-			ret = i915_gem_object_set_to_gtt_domain(obj, true);
-			if (ret)
-				return ret;
-		}
+		ret = i915_gem_object_wait_rendering(obj, false);
+		if (ret)
+			return ret;
 	}
 	/* Same trick applies to invalidate partially written cachelines read
 	 * before writing. */
@@ -966,12 +965,31 @@ i915_gem_check_olr(struct intel_ring_buffer *ring, u32 seqno)
 	BUG_ON(!mutex_is_locked(&ring->dev->struct_mutex));
 
 	ret = 0;
-	if (seqno == ring->outstanding_lazy_request)
+	if (seqno == ring->outstanding_lazy_seqno)
 		ret = i915_add_request(ring, NULL);
 
 	return ret;
 }
 
+static void fake_irq(unsigned long data)
+{
+	wake_up_process((struct task_struct *)data);
+}
+
+static bool missed_irq(struct drm_i915_private *dev_priv,
+		       struct intel_ring_buffer *ring)
+{
+	return test_bit(ring->id, &dev_priv->gpu_error.missed_irq_rings);
+}
+
+static bool can_wait_boost(struct drm_i915_file_private *file_priv)
+{
+	if (file_priv == NULL)
+		return true;
+
+	return !atomic_xchg(&file_priv->rps_wait_boost, true);
+}
+
 /**
  * __wait_seqno - wait until execution of seqno has finished
  * @ring: the ring expected to report seqno
@@ -992,13 +1010,14 @@ i915_gem_check_olr(struct intel_ring_buffer *ring, u32 seqno)
  */
 static int __wait_seqno(struct intel_ring_buffer *ring, u32 seqno,
 			unsigned reset_counter,
-			bool interruptible, struct timespec *timeout)
+			bool interruptible,
+			struct timespec *timeout,
+			struct drm_i915_file_private *file_priv)
 {
 	drm_i915_private_t *dev_priv = ring->dev->dev_private;
-	struct timespec before, now, wait_time={1,0};
-	unsigned long timeout_jiffies;
-	long end;
-	bool wait_forever = true;
+	struct timespec before, now;
+	DEFINE_WAIT(wait);
+	long timeout_jiffies;
 	int ret;
 
 	WARN(dev_priv->pc8.irqs_disabled, "IRQs disabled\n");
@@ -1006,51 +1025,79 @@ static int __wait_seqno(struct intel_ring_buffer *ring, u32 seqno,
 	if (i915_seqno_passed(ring->get_seqno(ring, true), seqno))
 		return 0;
 
-	trace_i915_gem_request_wait_begin(ring, seqno);
+	timeout_jiffies = timeout ? timespec_to_jiffies_timeout(timeout) : 1;
 
-	if (timeout != NULL) {
-		wait_time = *timeout;
-		wait_forever = false;
+	if (dev_priv->info->gen >= 6 && can_wait_boost(file_priv)) {
+		gen6_rps_boost(dev_priv);
+		if (file_priv)
+			mod_delayed_work(dev_priv->wq,
+					 &file_priv->mm.idle_work,
+					 msecs_to_jiffies(100));
 	}
 
-	timeout_jiffies = timespec_to_jiffies_timeout(&wait_time);
-
-	if (WARN_ON(!ring->irq_get(ring)))
+	if (!(dev_priv->gpu_error.test_irq_rings & intel_ring_flag(ring)) &&
+	    WARN_ON(!ring->irq_get(ring)))
 		return -ENODEV;
 
-	/* Record current time in case interrupted by signal, or wedged * */
+	/* Record current time in case interrupted by signal, or wedged */
+	trace_i915_gem_request_wait_begin(ring, seqno);
 	getrawmonotonic(&before);
+	for (;;) {
+		struct timer_list timer;
+		unsigned long expire;
 
-#define EXIT_COND \
-	(i915_seqno_passed(ring->get_seqno(ring, false), seqno) || \
-	 i915_reset_in_progress(&dev_priv->gpu_error) || \
-	 reset_counter != atomic_read(&dev_priv->gpu_error.reset_counter))
-	do {
-		if (interruptible)
-			end = wait_event_interruptible_timeout(ring->irq_queue,
-							       EXIT_COND,
-							       timeout_jiffies);
-		else
-			end = wait_event_timeout(ring->irq_queue, EXIT_COND,
-						 timeout_jiffies);
+		prepare_to_wait(&ring->irq_queue, &wait,
+				interruptible ? TASK_INTERRUPTIBLE : TASK_UNINTERRUPTIBLE);
 
 		/* We need to check whether any gpu reset happened in between
 		 * the caller grabbing the seqno and now ... */
-		if (reset_counter != atomic_read(&dev_priv->gpu_error.reset_counter))
-			end = -EAGAIN;
+		if (reset_counter != atomic_read(&dev_priv->gpu_error.reset_counter)) {
+			/* ... but upgrade the -EAGAIN to an -EIO if the gpu
+			 * is truely gone. */
+			ret = i915_gem_check_wedge(&dev_priv->gpu_error, interruptible);
+			if (ret == 0)
+				ret = -EAGAIN;
+			break;
+		}
 
-		/* ... but upgrade the -EGAIN to an -EIO if the gpu is truely
-		 * gone. */
-		ret = i915_gem_check_wedge(&dev_priv->gpu_error, interruptible);
-		if (ret)
-			end = ret;
-	} while (end == 0 && wait_forever);
+		if (i915_seqno_passed(ring->get_seqno(ring, false), seqno)) {
+			ret = 0;
+			break;
+		}
 
+		if (interruptible && signal_pending(current)) {
+			ret = -ERESTARTSYS;
+			break;
+		}
+
+		if (timeout_jiffies <= 0) {
+			ret = -ETIME;
+			break;
+		}
+
+		timer.function = NULL;
+		if (timeout || missed_irq(dev_priv, ring)) {
+			setup_timer_on_stack(&timer, fake_irq, (unsigned long)current);
+			expire = jiffies + (missed_irq(dev_priv, ring) ? 1: timeout_jiffies);
+			mod_timer(&timer, expire);
+		}
+
+		io_schedule();
+
+		if (timeout)
+			timeout_jiffies = expire - jiffies;
+
+		if (timer.function) {
+			del_singleshot_timer_sync(&timer);
+			destroy_timer_on_stack(&timer);
+		}
+	}
 	getrawmonotonic(&now);
+	trace_i915_gem_request_wait_end(ring, seqno);
 
 	ring->irq_put(ring);
-	trace_i915_gem_request_wait_end(ring, seqno);
-#undef EXIT_COND
+
+	finish_wait(&ring->irq_queue, &wait);
 
 	if (timeout) {
 		struct timespec sleep_time = timespec_sub(now, before);
@@ -1059,17 +1106,7 @@ static int __wait_seqno(struct intel_ring_buffer *ring, u32 seqno,
 			set_normalized_timespec(timeout, 0, 0);
 	}
 
-	switch (end) {
-	case -EIO:
-	case -EAGAIN: /* Wedged */
-	case -ERESTARTSYS: /* Signal */
-		return (int)end;
-	case 0: /* Timeout */
-		return -ETIME;
-	default: /* Completed */
-		WARN_ON(end < 0); /* We're not aware of other errors */
-		return 0;
-	}
+	return ret;
 }
 
 /**
@@ -1097,7 +1134,7 @@ i915_wait_seqno(struct intel_ring_buffer *ring, uint32_t seqno)
 
 	return __wait_seqno(ring, seqno,
 			    atomic_read(&dev_priv->gpu_error.reset_counter),
-			    interruptible, NULL);
+			    interruptible, NULL, NULL);
 }
 
 static int
@@ -1147,6 +1184,7 @@ i915_gem_object_wait_rendering(struct drm_i915_gem_object *obj,
  */
 static __must_check int
 i915_gem_object_wait_rendering__nonblocking(struct drm_i915_gem_object *obj,
+					    struct drm_file *file,
 					    bool readonly)
 {
 	struct drm_device *dev = obj->base.dev;
@@ -1173,7 +1211,7 @@ i915_gem_object_wait_rendering__nonblocking(struct drm_i915_gem_object *obj,
 
 	reset_counter = atomic_read(&dev_priv->gpu_error.reset_counter);
 	mutex_unlock(&dev->struct_mutex);
-	ret = __wait_seqno(ring, seqno, reset_counter, true, NULL);
+	ret = __wait_seqno(ring, seqno, reset_counter, true, NULL, file->driver_priv);
 	mutex_lock(&dev->struct_mutex);
 	if (ret)
 		return ret;
@@ -1222,7 +1260,7 @@ i915_gem_set_domain_ioctl(struct drm_device *dev, void *data,
 	 * We will repeat the flush holding the lock in the normal manner
 	 * to catch cases where we are gazumped.
 	 */
-	ret = i915_gem_object_wait_rendering__nonblocking(obj, !write_domain);
+	ret = i915_gem_object_wait_rendering__nonblocking(obj, file, !write_domain);
 	if (ret)
 		goto unref;
 
@@ -1690,13 +1728,13 @@ i915_gem_object_put_pages(struct drm_i915_gem_object *obj)
 	return 0;
 }
 
-static long
+static unsigned long
 __i915_gem_shrink(struct drm_i915_private *dev_priv, long target,
 		  bool purgeable_only)
 {
 	struct list_head still_bound_list;
 	struct drm_i915_gem_object *obj, *next;
-	long count = 0;
+	unsigned long count = 0;
 
 	list_for_each_entry_safe(obj, next,
 				 &dev_priv->mm.unbound_list,
@@ -1762,13 +1800,13 @@ __i915_gem_shrink(struct drm_i915_private *dev_priv, long target,
 	return count;
 }
 
-static long
+static unsigned long
 i915_gem_purge(struct drm_i915_private *dev_priv, long target)
 {
 	return __i915_gem_shrink(dev_priv, target, true);
 }
 
-static long
+static unsigned long
 i915_gem_shrink_all(struct drm_i915_private *dev_priv)
 {
 	struct drm_i915_gem_object *obj, *next;
@@ -1778,9 +1816,8 @@ i915_gem_shrink_all(struct drm_i915_private *dev_priv)
 
 	list_for_each_entry_safe(obj, next, &dev_priv->mm.unbound_list,
 				 global_list) {
-		if (obj->pages_pin_count == 0)
+		if (i915_gem_object_put_pages(obj) == 0)
 			freed += obj->base.size >> PAGE_SHIFT;
-		i915_gem_object_put_pages(obj);
 	}
 	return freed;
 }
@@ -1865,6 +1902,9 @@ i915_gem_object_get_pages_gtt(struct drm_i915_gem_object *obj)
 			sg->length += PAGE_SIZE;
 		}
 		last_pfn = page_to_pfn(page);
+
+		/* Check that the i965g/gm workaround works. */
+		WARN_ON((gfp & __GFP_DMA32) && (last_pfn >= 0x00100000UL));
 	}
 #ifdef CONFIG_SWIOTLB
 	if (!swiotlb_nr_tbl())
@@ -1918,7 +1958,7 @@ i915_gem_object_get_pages(struct drm_i915_gem_object *obj)
 	return 0;
 }
 
-void
+static void
 i915_gem_object_move_to_active(struct drm_i915_gem_object *obj,
 			       struct intel_ring_buffer *ring)
 {
@@ -1957,6 +1997,13 @@ i915_gem_object_move_to_active(struct drm_i915_gem_object *obj,
 	}
 }
 
+void i915_vma_move_to_active(struct i915_vma *vma,
+			     struct intel_ring_buffer *ring)
+{
+	list_move_tail(&vma->mm_list, &vma->vm->active_list);
+	return i915_gem_object_move_to_active(vma->obj, ring);
+}
+
 static void
 i915_gem_object_move_to_inactive(struct drm_i915_gem_object *obj)
 {
@@ -2078,11 +2125,10 @@ int __i915_add_request(struct intel_ring_buffer *ring,
 	if (ret)
 		return ret;
 
-	request = kmalloc(sizeof(*request), GFP_KERNEL);
-	if (request == NULL)
+	request = ring->preallocated_lazy_request;
+	if (WARN_ON(request == NULL))
 		return -ENOMEM;
 
-
 	/* Record the position of the start of the request so that
 	 * should we detect the updated seqno part-way through the
 	 * GPU processing the request, we never over-estimate the
@@ -2091,17 +2137,13 @@ int __i915_add_request(struct intel_ring_buffer *ring,
 	request_ring_position = intel_ring_get_tail(ring);
 
 	ret = ring->add_request(ring);
-	if (ret) {
-		kfree(request);
+	if (ret)
 		return ret;
-	}
 
 	request->seqno = intel_ring_get_seqno(ring);
 	request->ring = ring;
 	request->head = request_start;
 	request->tail = request_ring_position;
-	request->ctx = ring->last_context;
-	request->batch_obj = obj;
 
 	/* Whilst this request exists, batch_obj will be on the
 	 * active_list, and so will hold the active reference. Only when this
@@ -2109,7 +2151,12 @@ int __i915_add_request(struct intel_ring_buffer *ring,
 	 * inactive_list and lose its active reference. Hence we do not need
 	 * to explicitly hold another reference here.
 	 */
+	request->batch_obj = obj;
 
+	/* Hold a reference to the current context so that we can inspect
+	 * it later in case a hangcheck error event fires.
+	 */
+	request->ctx = ring->last_context;
 	if (request->ctx)
 		i915_gem_context_reference(request->ctx);
 
@@ -2129,12 +2176,14 @@ int __i915_add_request(struct intel_ring_buffer *ring,
 	}
 
 	trace_i915_gem_request_add(ring, request->seqno);
-	ring->outstanding_lazy_request = 0;
+	ring->outstanding_lazy_seqno = 0;
+	ring->preallocated_lazy_request = NULL;
 
 	if (!dev_priv->ums.mm_suspended) {
 		i915_queue_hangcheck(ring->dev);
 
 		if (was_empty) {
+			cancel_delayed_work_sync(&dev_priv->mm.idle_work);
 			queue_delayed_work(dev_priv->wq,
 					   &dev_priv->mm.retire_work,
 					   round_jiffies_up_relative(HZ));
@@ -2156,10 +2205,8 @@ i915_gem_request_remove_from_client(struct drm_i915_gem_request *request)
 		return;
 
 	spin_lock(&file_priv->mm.lock);
-	if (request->file_priv) {
-		list_del(&request->client_list);
-		request->file_priv = NULL;
-	}
+	list_del(&request->client_list);
+	request->file_priv = NULL;
 	spin_unlock(&file_priv->mm.lock);
 }
 
@@ -2224,6 +2271,21 @@ static bool i915_request_guilty(struct drm_i915_gem_request *request,
 	return false;
 }
 
+static bool i915_context_is_banned(const struct i915_ctx_hang_stats *hs)
+{
+	const unsigned long elapsed = get_seconds() - hs->guilty_ts;
+
+	if (hs->banned)
+		return true;
+
+	if (elapsed <= DRM_I915_CTX_BAN_PERIOD) {
+		DRM_ERROR("context hanging too fast, declaring banned!\n");
+		return true;
+	}
+
+	return false;
+}
+
 static void i915_set_reset_status(struct intel_ring_buffer *ring,
 				  struct drm_i915_gem_request *request,
 				  u32 acthd)
@@ -2260,10 +2322,13 @@ static void i915_set_reset_status(struct intel_ring_buffer *ring,
 		hs = &request->file_priv->hang_stats;
 
 	if (hs) {
-		if (guilty)
+		if (guilty) {
+			hs->banned = i915_context_is_banned(hs);
 			hs->batch_active++;
-		else
+			hs->guilty_ts = get_seconds();
+		} else {
 			hs->batch_pending++;
+		}
 	}
 }
 
@@ -2341,6 +2406,8 @@ void i915_gem_reset(struct drm_device *dev)
 	for_each_ring(ring, dev_priv, i)
 		i915_gem_reset_ring_lists(dev_priv, ring);
 
+	i915_gem_cleanup_ringbuffer(dev);
+
 	i915_gem_restore_fences(dev);
 }
 
@@ -2405,57 +2472,53 @@ i915_gem_retire_requests_ring(struct intel_ring_buffer *ring)
 	WARN_ON(i915_verify_lists(ring->dev));
 }
 
-void
+bool
 i915_gem_retire_requests(struct drm_device *dev)
 {
 	drm_i915_private_t *dev_priv = dev->dev_private;
 	struct intel_ring_buffer *ring;
+	bool idle = true;
 	int i;
 
-	for_each_ring(ring, dev_priv, i)
+	for_each_ring(ring, dev_priv, i) {
 		i915_gem_retire_requests_ring(ring);
+		idle &= list_empty(&ring->request_list);
+	}
+
+	if (idle)
+		mod_delayed_work(dev_priv->wq,
+				   &dev_priv->mm.idle_work,
+				   msecs_to_jiffies(100));
+
+	return idle;
 }
 
 static void
 i915_gem_retire_work_handler(struct work_struct *work)
 {
-	drm_i915_private_t *dev_priv;
-	struct drm_device *dev;
-	struct intel_ring_buffer *ring;
+	struct drm_i915_private *dev_priv =
+		container_of(work, typeof(*dev_priv), mm.retire_work.work);
+	struct drm_device *dev = dev_priv->dev;
 	bool idle;
-	int i;
-
-	dev_priv = container_of(work, drm_i915_private_t,
-				mm.retire_work.work);
-	dev = dev_priv->dev;
 
 	/* Come back later if the device is busy... */
-	if (!mutex_trylock(&dev->struct_mutex)) {
-		queue_delayed_work(dev_priv->wq, &dev_priv->mm.retire_work,
-				   round_jiffies_up_relative(HZ));
-		return;
-	}
-
-	i915_gem_retire_requests(dev);
-
-	/* Send a periodic flush down the ring so we don't hold onto GEM
-	 * objects indefinitely.
-	 */
-	idle = true;
-	for_each_ring(ring, dev_priv, i) {
-		if (ring->gpu_caches_dirty)
-			i915_add_request(ring, NULL);
-
-		idle &= list_empty(&ring->request_list);
+	idle = false;
+	if (mutex_trylock(&dev->struct_mutex)) {
+		idle = i915_gem_retire_requests(dev);
+		mutex_unlock(&dev->struct_mutex);
 	}
-
-	if (!dev_priv->ums.mm_suspended && !idle)
+	if (!idle)
 		queue_delayed_work(dev_priv->wq, &dev_priv->mm.retire_work,
 				   round_jiffies_up_relative(HZ));
-	if (idle)
-		intel_mark_idle(dev);
+}
 
-	mutex_unlock(&dev->struct_mutex);
+static void
+i915_gem_idle_work_handler(struct work_struct *work)
+{
+	struct drm_i915_private *dev_priv =
+		container_of(work, typeof(*dev_priv), mm.idle_work.work);
+
+	intel_mark_idle(dev_priv->dev);
 }
 
 /**
@@ -2553,7 +2616,7 @@ i915_gem_wait_ioctl(struct drm_device *dev, void *data, struct drm_file *file)
 	reset_counter = atomic_read(&dev_priv->gpu_error.reset_counter);
 	mutex_unlock(&dev->struct_mutex);
 
-	ret = __wait_seqno(ring, seqno, reset_counter, true, timeout);
+	ret = __wait_seqno(ring, seqno, reset_counter, true, timeout, file->driver_priv);
 	if (timeout)
 		args->timeout_ns = timespec_to_ns(timeout);
 	return ret;
@@ -2600,6 +2663,7 @@ i915_gem_object_sync(struct drm_i915_gem_object *obj,
 	if (ret)
 		return ret;
 
+	trace_i915_gem_ring_sync_to(from, to, seqno);
 	ret = to->sync_to(to, from, seqno);
 	if (!ret)
 		/* We use last_read_seqno because sync_to()
@@ -2641,11 +2705,17 @@ int i915_vma_unbind(struct i915_vma *vma)
 	drm_i915_private_t *dev_priv = obj->base.dev->dev_private;
 	int ret;
 
+	/* For now we only ever use 1 vma per object */
+	WARN_ON(!list_is_singular(&obj->vma_list));
+
 	if (list_empty(&vma->vma_link))
 		return 0;
 
-	if (!drm_mm_node_allocated(&vma->node))
-		goto destroy;
+	if (!drm_mm_node_allocated(&vma->node)) {
+		i915_gem_vma_destroy(vma);
+
+		return 0;
+	}
 
 	if (obj->pin_count)
 		return -EBUSY;
@@ -2685,13 +2755,10 @@ int i915_vma_unbind(struct i915_vma *vma)
 
 	drm_mm_remove_node(&vma->node);
 
-destroy:
 	i915_gem_vma_destroy(vma);
 
 	/* Since the unbound list is global, only move to that list if
-	 * no more VMAs exist.
-	 * NB: Until we have real VMAs there will only ever be one */
-	WARN_ON(!list_empty(&obj->vma_list));
+	 * no more VMAs exist. */
 	if (list_empty(&obj->vma_list))
 		list_move_tail(&obj->global_list, &dev_priv->mm.unbound_list);
 
@@ -2887,6 +2954,7 @@ static void i915_gem_write_fence(struct drm_device *dev, int reg,
 	     obj->stride, obj->tiling_mode);
 
 	switch (INTEL_INFO(dev)->gen) {
+	case 8:
 	case 7:
 	case 6:
 	case 5:
@@ -3389,8 +3457,7 @@ i915_gem_object_set_to_gtt_domain(struct drm_i915_gem_object *obj, bool write)
 
 	/* And bump the LRU for this access */
 	if (i915_gem_object_is_inactive(obj)) {
-		struct i915_vma *vma = i915_gem_obj_to_vma(obj,
-							   &dev_priv->gtt.base);
+		struct i915_vma *vma = i915_gem_obj_to_ggtt(obj);
 		if (vma)
 			list_move_tail(&vma->mm_list,
 				       &dev_priv->gtt.base.inactive_list);
@@ -3761,7 +3828,7 @@ i915_gem_ring_throttle(struct drm_device *dev, struct drm_file *file)
 	if (seqno == 0)
 		return 0;
 
-	ret = __wait_seqno(ring, seqno, reset_counter, true, NULL);
+	ret = __wait_seqno(ring, seqno, reset_counter, true, NULL, NULL);
 	if (ret == 0)
 		queue_delayed_work(dev_priv->wq, &dev_priv->mm.retire_work, 0);
 
@@ -3865,6 +3932,11 @@ i915_gem_pin_ioctl(struct drm_device *dev, void *data,
 		goto out;
 	}
 
+	if (obj->user_pin_count == ULONG_MAX) {
+		ret = -EBUSY;
+		goto out;
+	}
+
 	if (obj->user_pin_count == 0) {
 		ret = i915_gem_obj_ggtt_pin(obj, args->alignment, true, false);
 		if (ret)
@@ -4015,7 +4087,6 @@ void i915_gem_object_init(struct drm_i915_gem_object *obj,
 {
 	INIT_LIST_HEAD(&obj->global_list);
 	INIT_LIST_HEAD(&obj->ring_list);
-	INIT_LIST_HEAD(&obj->exec_list);
 	INIT_LIST_HEAD(&obj->obj_exec_link);
 	INIT_LIST_HEAD(&obj->vma_list);
 
@@ -4087,13 +4158,6 @@ struct drm_i915_gem_object *i915_gem_alloc_object(struct drm_device *dev,
 	return obj;
 }
 
-int i915_gem_init_object(struct drm_gem_object *obj)
-{
-	BUG();
-
-	return 0;
-}
-
 void i915_gem_free_object(struct drm_gem_object *gem_obj)
 {
 	struct drm_i915_gem_object *obj = to_intel_bo(gem_obj);
@@ -4147,9 +4211,20 @@ void i915_gem_free_object(struct drm_gem_object *gem_obj)
 	i915_gem_object_free(obj);
 }
 
-struct i915_vma *i915_gem_vma_create(struct drm_i915_gem_object *obj,
+struct i915_vma *i915_gem_obj_to_vma(struct drm_i915_gem_object *obj,
 				     struct i915_address_space *vm)
 {
+	struct i915_vma *vma;
+	list_for_each_entry(vma, &obj->vma_list, vma_link)
+		if (vma->vm == vm)
+			return vma;
+
+	return NULL;
+}
+
+static struct i915_vma *__i915_gem_vma_create(struct drm_i915_gem_object *obj,
+					      struct i915_address_space *vm)
+{
 	struct i915_vma *vma = kzalloc(sizeof(*vma), GFP_KERNEL);
 	if (vma == NULL)
 		return ERR_PTR(-ENOMEM);
@@ -4169,76 +4244,103 @@ struct i915_vma *i915_gem_vma_create(struct drm_i915_gem_object *obj,
 	return vma;
 }
 
+struct i915_vma *
+i915_gem_obj_lookup_or_create_vma(struct drm_i915_gem_object *obj,
+				  struct i915_address_space *vm)
+{
+	struct i915_vma *vma;
+
+	vma = i915_gem_obj_to_vma(obj, vm);
+	if (!vma)
+		vma = __i915_gem_vma_create(obj, vm);
+
+	return vma;
+}
+
 void i915_gem_vma_destroy(struct i915_vma *vma)
 {
 	WARN_ON(vma->node.allocated);
+
+	/* Keep the vma as a placeholder in the execbuffer reservation lists */
+	if (!list_empty(&vma->exec_list))
+		return;
+
 	list_del(&vma->vma_link);
+
 	kfree(vma);
 }
 
 int
-i915_gem_idle(struct drm_device *dev)
+i915_gem_suspend(struct drm_device *dev)
 {
 	drm_i915_private_t *dev_priv = dev->dev_private;
-	int ret;
+	int ret = 0;
 
-	if (dev_priv->ums.mm_suspended) {
-		mutex_unlock(&dev->struct_mutex);
-		return 0;
-	}
+	mutex_lock(&dev->struct_mutex);
+	if (dev_priv->ums.mm_suspended)
+		goto err;
 
 	ret = i915_gpu_idle(dev);
-	if (ret) {
-		mutex_unlock(&dev->struct_mutex);
-		return ret;
-	}
+	if (ret)
+		goto err;
+
 	i915_gem_retire_requests(dev);
 
 	/* Under UMS, be paranoid and evict. */
 	if (!drm_core_check_feature(dev, DRIVER_MODESET))
 		i915_gem_evict_everything(dev);
 
-	del_timer_sync(&dev_priv->gpu_error.hangcheck_timer);
-
 	i915_kernel_lost_context(dev);
 	i915_gem_cleanup_ringbuffer(dev);
 
-	/* Cancel the retire work handler, which should be idle now. */
+	/* Hack!  Don't let anybody do execbuf while we don't control the chip.
+	 * We need to replace this with a semaphore, or something.
+	 * And not confound ums.mm_suspended!
+	 */
+	dev_priv->ums.mm_suspended = !drm_core_check_feature(dev,
+							     DRIVER_MODESET);
+	mutex_unlock(&dev->struct_mutex);
+
+	del_timer_sync(&dev_priv->gpu_error.hangcheck_timer);
 	cancel_delayed_work_sync(&dev_priv->mm.retire_work);
+	cancel_delayed_work_sync(&dev_priv->mm.idle_work);
 
 	return 0;
+
+err:
+	mutex_unlock(&dev->struct_mutex);
+	return ret;
 }
 
-void i915_gem_l3_remap(struct drm_device *dev)
+int i915_gem_l3_remap(struct intel_ring_buffer *ring, int slice)
 {
+	struct drm_device *dev = ring->dev;
 	drm_i915_private_t *dev_priv = dev->dev_private;
-	u32 misccpctl;
-	int i;
-
-	if (!HAS_L3_GPU_CACHE(dev))
-		return;
+	u32 reg_base = GEN7_L3LOG_BASE + (slice * 0x200);
+	u32 *remap_info = dev_priv->l3_parity.remap_info[slice];
+	int i, ret;
 
-	if (!dev_priv->l3_parity.remap_info)
-		return;
+	if (!HAS_L3_DPF(dev) || !remap_info)
+		return 0;
 
-	misccpctl = I915_READ(GEN7_MISCCPCTL);
-	I915_WRITE(GEN7_MISCCPCTL, misccpctl & ~GEN7_DOP_CLOCK_GATE_ENABLE);
-	POSTING_READ(GEN7_MISCCPCTL);
+	ret = intel_ring_begin(ring, GEN7_L3LOG_SIZE / 4 * 3);
+	if (ret)
+		return ret;
 
+	/*
+	 * Note: We do not worry about the concurrent register cacheline hang
+	 * here because no other code should access these registers other than
+	 * at initialization time.
+	 */
 	for (i = 0; i < GEN7_L3LOG_SIZE; i += 4) {
-		u32 remap = I915_READ(GEN7_L3LOG_BASE + i);
-		if (remap && remap != dev_priv->l3_parity.remap_info[i/4])
-			DRM_DEBUG("0x%x was already programmed to %x\n",
-				  GEN7_L3LOG_BASE + i, remap);
-		if (remap && !dev_priv->l3_parity.remap_info[i/4])
-			DRM_DEBUG_DRIVER("Clearing remapped register\n");
-		I915_WRITE(GEN7_L3LOG_BASE + i, dev_priv->l3_parity.remap_info[i/4]);
+		intel_ring_emit(ring, MI_LOAD_REGISTER_IMM(1));
+		intel_ring_emit(ring, reg_base + i);
+		intel_ring_emit(ring, remap_info[i/4]);
 	}
 
-	/* Make sure all the writes land before disabling dop clock gating */
-	POSTING_READ(GEN7_L3LOG_BASE);
+	intel_ring_advance(ring);
 
-	I915_WRITE(GEN7_MISCCPCTL, misccpctl);
+	return ret;
 }
 
 void i915_gem_init_swizzling(struct drm_device *dev)
@@ -4260,6 +4362,8 @@ void i915_gem_init_swizzling(struct drm_device *dev)
 		I915_WRITE(ARB_MODE, _MASKED_BIT_ENABLE(ARB_MODE_SWIZZLE_SNB));
 	else if (IS_GEN7(dev))
 		I915_WRITE(ARB_MODE, _MASKED_BIT_ENABLE(ARB_MODE_SWIZZLE_IVB));
+	else if (IS_GEN8(dev))
+		I915_WRITE(GAMTARBMODE, _MASKED_BIT_ENABLE(ARB_MODE_SWIZZLE_BDW));
 	else
 		BUG();
 }
@@ -4330,7 +4434,7 @@ int
 i915_gem_init_hw(struct drm_device *dev)
 {
 	drm_i915_private_t *dev_priv = dev->dev_private;
-	int ret;
+	int ret, i;
 
 	if (INTEL_INFO(dev)->gen < 6 && !intel_enable_gtt())
 		return -EIO;
@@ -4338,20 +4442,26 @@ i915_gem_init_hw(struct drm_device *dev)
 	if (dev_priv->ellc_size)
 		I915_WRITE(HSW_IDICR, I915_READ(HSW_IDICR) | IDIHASHMSK(0xf));
 
+	if (IS_HSW_GT3(dev))
+		I915_WRITE(MI_PREDICATE_RESULT_2, LOWER_SLICE_ENABLED);
+	else
+		I915_WRITE(MI_PREDICATE_RESULT_2, LOWER_SLICE_DISABLED);
+
 	if (HAS_PCH_NOP(dev)) {
 		u32 temp = I915_READ(GEN7_MSG_CTL);
 		temp &= ~(WAIT_FOR_PCH_FLR_ACK | WAIT_FOR_PCH_RESET_ACK);
 		I915_WRITE(GEN7_MSG_CTL, temp);
 	}
 
-	i915_gem_l3_remap(dev);
-
 	i915_gem_init_swizzling(dev);
 
 	ret = i915_gem_init_rings(dev);
 	if (ret)
 		return ret;
 
+	for (i = 0; i < NUM_L3_SLICES(dev); i++)
+		i915_gem_l3_remap(&dev_priv->ring[RCS], i);
+
 	/*
 	 * XXX: There was some w/a described somewhere suggesting loading
 	 * contexts before PPGTT.
@@ -4454,26 +4564,12 @@ int
 i915_gem_leavevt_ioctl(struct drm_device *dev, void *data,
 		       struct drm_file *file_priv)
 {
-	struct drm_i915_private *dev_priv = dev->dev_private;
-	int ret;
-
 	if (drm_core_check_feature(dev, DRIVER_MODESET))
 		return 0;
 
 	drm_irq_uninstall(dev);
 
-	mutex_lock(&dev->struct_mutex);
-	ret =  i915_gem_idle(dev);
-
-	/* Hack!  Don't let anybody do execbuf while we don't control the chip.
-	 * We need to replace this with a semaphore, or something.
-	 * And not confound ums.mm_suspended!
-	 */
-	if (ret != 0)
-		dev_priv->ums.mm_suspended = 1;
-	mutex_unlock(&dev->struct_mutex);
-
-	return ret;
+	return i915_gem_suspend(dev);
 }
 
 void
@@ -4484,11 +4580,9 @@ i915_gem_lastclose(struct drm_device *dev)
 	if (drm_core_check_feature(dev, DRIVER_MODESET))
 		return;
 
-	mutex_lock(&dev->struct_mutex);
-	ret = i915_gem_idle(dev);
+	ret = i915_gem_suspend(dev);
 	if (ret)
 		DRM_ERROR("failed to idle hardware: %d\n", ret);
-	mutex_unlock(&dev->struct_mutex);
 }
 
 static void
@@ -4523,6 +4617,7 @@ i915_gem_load(struct drm_device *dev)
 	INIT_LIST_HEAD(&dev_priv->vm_list);
 	i915_init_vm(dev_priv, &dev_priv->gtt.base);
 
+	INIT_LIST_HEAD(&dev_priv->context_list);
 	INIT_LIST_HEAD(&dev_priv->mm.unbound_list);
 	INIT_LIST_HEAD(&dev_priv->mm.bound_list);
 	INIT_LIST_HEAD(&dev_priv->mm.fence_list);
@@ -4532,6 +4627,8 @@ i915_gem_load(struct drm_device *dev)
 		INIT_LIST_HEAD(&dev_priv->fence_regs[i].lru_list);
 	INIT_DELAYED_WORK(&dev_priv->mm.retire_work,
 			  i915_gem_retire_work_handler);
+	INIT_DELAYED_WORK(&dev_priv->mm.idle_work,
+			  i915_gem_idle_work_handler);
 	init_waitqueue_head(&dev_priv->gpu_error.reset_queue);
 
 	/* On GEN3 we really need to make sure the ARB C3 LP bit is set */
@@ -4582,7 +4679,7 @@ static int i915_gem_init_phys_object(struct drm_device *dev,
 	if (dev_priv->mm.phys_objs[id - 1] || !size)
 		return 0;
 
-	phys_obj = kzalloc(sizeof(struct drm_i915_gem_phys_object), GFP_KERNEL);
+	phys_obj = kzalloc(sizeof(*phys_obj), GFP_KERNEL);
 	if (!phys_obj)
 		return -ENOMEM;
 
@@ -4756,6 +4853,8 @@ void i915_gem_release(struct drm_device *dev, struct drm_file *file)
 {
 	struct drm_i915_file_private *file_priv = file->driver_priv;
 
+	cancel_delayed_work_sync(&file_priv->mm.idle_work);
+
 	/* Clean up our request list when the client is going away, so that
 	 * later retire_requests won't dereference our soon-to-be-gone
 	 * file_priv.
@@ -4773,6 +4872,38 @@ void i915_gem_release(struct drm_device *dev, struct drm_file *file)
 	spin_unlock(&file_priv->mm.lock);
 }
 
+static void
+i915_gem_file_idle_work_handler(struct work_struct *work)
+{
+	struct drm_i915_file_private *file_priv =
+		container_of(work, typeof(*file_priv), mm.idle_work.work);
+
+	atomic_set(&file_priv->rps_wait_boost, false);
+}
+
+int i915_gem_open(struct drm_device *dev, struct drm_file *file)
+{
+	struct drm_i915_file_private *file_priv;
+
+	DRM_DEBUG_DRIVER("\n");
+
+	file_priv = kzalloc(sizeof(*file_priv), GFP_KERNEL);
+	if (!file_priv)
+		return -ENOMEM;
+
+	file->driver_priv = file_priv;
+	file_priv->dev_priv = dev->dev_private;
+
+	spin_lock_init(&file_priv->mm.lock);
+	INIT_LIST_HEAD(&file_priv->mm.request_list);
+	INIT_DELAYED_WORK(&file_priv->mm.idle_work,
+			  i915_gem_file_idle_work_handler);
+
+	idr_init(&file_priv->context_idr);
+
+	return 0;
+}
+
 static bool mutex_is_locked_by(struct mutex *mutex, struct task_struct *task)
 {
 	if (!mutex_is_locked(mutex))
@@ -4823,6 +4954,7 @@ i915_gem_inactive_count(struct shrinker *shrinker, struct shrink_control *sc)
 
 	if (unlock)
 		mutex_unlock(&dev->struct_mutex);
+
 	return count;
 }
 
@@ -4859,11 +4991,10 @@ bool i915_gem_obj_bound(struct drm_i915_gem_object *o,
 
 bool i915_gem_obj_bound_any(struct drm_i915_gem_object *o)
 {
-	struct drm_i915_private *dev_priv = o->base.dev->dev_private;
-	struct i915_address_space *vm;
+	struct i915_vma *vma;
 
-	list_for_each_entry(vm, &dev_priv->vm_list, global_link)
-		if (i915_gem_obj_bound(o, vm))
+	list_for_each_entry(vma, &o->vma_list, vma_link)
+		if (drm_mm_node_allocated(&vma->node))
 			return true;
 
 	return false;
@@ -4895,7 +5026,6 @@ i915_gem_inactive_scan(struct shrinker *shrinker, struct shrink_control *sc)
 			     struct drm_i915_private,
 			     mm.inactive_shrinker);
 	struct drm_device *dev = dev_priv->dev;
-	int nr_to_scan = sc->nr_to_scan;
 	unsigned long freed;
 	bool unlock = true;
 
@@ -4909,38 +5039,30 @@ i915_gem_inactive_scan(struct shrinker *shrinker, struct shrink_control *sc)
 		unlock = false;
 	}
 
-	freed = i915_gem_purge(dev_priv, nr_to_scan);
-	if (freed < nr_to_scan)
-		freed += __i915_gem_shrink(dev_priv, nr_to_scan,
-							false);
-	if (freed < nr_to_scan)
+	freed = i915_gem_purge(dev_priv, sc->nr_to_scan);
+	if (freed < sc->nr_to_scan)
+		freed += __i915_gem_shrink(dev_priv,
+					   sc->nr_to_scan - freed,
+					   false);
+	if (freed < sc->nr_to_scan)
 		freed += i915_gem_shrink_all(dev_priv);
 
 	if (unlock)
 		mutex_unlock(&dev->struct_mutex);
+
 	return freed;
 }
 
-struct i915_vma *i915_gem_obj_to_vma(struct drm_i915_gem_object *obj,
-				     struct i915_address_space *vm)
+struct i915_vma *i915_gem_obj_to_ggtt(struct drm_i915_gem_object *obj)
 {
 	struct i915_vma *vma;
-	list_for_each_entry(vma, &obj->vma_list, vma_link)
-		if (vma->vm == vm)
-			return vma;
 
-	return NULL;
-}
-
-struct i915_vma *
-i915_gem_obj_lookup_or_create_vma(struct drm_i915_gem_object *obj,
-				  struct i915_address_space *vm)
-{
-	struct i915_vma *vma;
+	if (WARN_ON(list_empty(&obj->vma_list)))
+		return NULL;
 
-	vma = i915_gem_obj_to_vma(obj, vm);
-	if (!vma)
-		vma = i915_gem_vma_create(obj, vm);
+	vma = list_first_entry(&obj->vma_list, typeof(*vma), vma_link);
+	if (WARN_ON(vma->vm != obj_to_ggtt(obj)))
+		return NULL;
 
 	return vma;
 }
diff --git a/drivers/gpu/drm/i915/i915_gem_context.c b/drivers/gpu/drm/i915/i915_gem_context.c
index 403309c2a7d6..72a3df32292f 100644
--- a/drivers/gpu/drm/i915/i915_gem_context.c
+++ b/drivers/gpu/drm/i915/i915_gem_context.c
@@ -73,7 +73,7 @@
  *
  * There are two confusing terms used above:
  *  The "current context" means the context which is currently running on the
- *  GPU. The GPU has loaded it's state already and has stored away the gtt
+ *  GPU. The GPU has loaded its state already and has stored away the gtt
  *  offset of the BO. The GPU is not actively referencing the data at this
  *  offset, but it will on the next context switch. The only way to avoid this
  *  is to do a GPU reset.
@@ -117,6 +117,9 @@ static int get_context_size(struct drm_device *dev)
 		else
 			ret = GEN7_CXT_TOTAL_SIZE(reg) * 64;
 		break;
+	case 8:
+		ret = GEN8_CXT_TOTAL_SIZE;
+		break;
 	default:
 		BUG();
 	}
@@ -129,6 +132,7 @@ void i915_gem_context_free(struct kref *ctx_ref)
 	struct i915_hw_context *ctx = container_of(ctx_ref,
 						   typeof(*ctx), ref);
 
+	list_del(&ctx->link);
 	drm_gem_object_unreference(&ctx->obj->base);
 	kfree(ctx);
 }
@@ -147,6 +151,7 @@ create_hw_context(struct drm_device *dev,
 
 	kref_init(&ctx->ref);
 	ctx->obj = i915_gem_alloc_object(dev, dev_priv->hw_context_size);
+	INIT_LIST_HEAD(&ctx->link);
 	if (ctx->obj == NULL) {
 		kfree(ctx);
 		DRM_DEBUG_DRIVER("Context object allocated failed\n");
@@ -166,6 +171,7 @@ create_hw_context(struct drm_device *dev,
 	 * assertion in the context switch code.
 	 */
 	ctx->ring = &dev_priv->ring[RCS];
+	list_add_tail(&ctx->link, &dev_priv->context_list);
 
 	/* Default context will never have a file_priv */
 	if (file_priv == NULL)
@@ -178,6 +184,10 @@ create_hw_context(struct drm_device *dev,
 
 	ctx->file_priv = file_priv;
 	ctx->id = ret;
+	/* NB: Mark all slices as needing a remap so that when the context first
+	 * loads it will restore whatever remap state already exists. If there
+	 * is no remap info, it will be a NOP. */
+	ctx->remap_slice = (1 << NUM_L3_SLICES(dev)) - 1;
 
 	return ctx;
 
@@ -213,7 +223,6 @@ static int create_default_context(struct drm_i915_private *dev_priv)
 	 * may not be available. To avoid this we always pin the
 	 * default context.
 	 */
-	dev_priv->ring[RCS].default_context = ctx;
 	ret = i915_gem_obj_ggtt_pin(ctx->obj, CONTEXT_ALIGN, false, false);
 	if (ret) {
 		DRM_DEBUG_DRIVER("Couldn't pin %d\n", ret);
@@ -226,6 +235,8 @@ static int create_default_context(struct drm_i915_private *dev_priv)
 		goto err_unpin;
 	}
 
+	dev_priv->ring[RCS].default_context = ctx;
+
 	DRM_DEBUG_DRIVER("Default HW context loaded\n");
 	return 0;
 
@@ -281,16 +292,24 @@ void i915_gem_context_fini(struct drm_device *dev)
 	 * other code, leading to spurious errors. */
 	intel_gpu_reset(dev);
 
-	i915_gem_object_unpin(dctx->obj);
-
 	/* When default context is created and switched to, base object refcount
 	 * will be 2 (+1 from object creation and +1 from do_switch()).
 	 * i915_gem_context_fini() will be called after gpu_idle() has switched
 	 * to default context. So we need to unreference the base object once
 	 * to offset the do_switch part, so that i915_gem_context_unreference()
 	 * can then free the base object correctly. */
-	drm_gem_object_unreference(&dctx->obj->base);
+	WARN_ON(!dev_priv->ring[RCS].last_context);
+	if (dev_priv->ring[RCS].last_context == dctx) {
+		/* Fake switch to NULL context */
+		WARN_ON(dctx->obj->active);
+		i915_gem_object_unpin(dctx->obj);
+		i915_gem_context_unreference(dctx);
+	}
+
+	i915_gem_object_unpin(dctx->obj);
 	i915_gem_context_unreference(dctx);
+	dev_priv->ring[RCS].default_context = NULL;
+	dev_priv->ring[RCS].last_context = NULL;
 }
 
 static int context_idr_cleanup(int id, void *p, void *data)
@@ -393,11 +412,11 @@ static int do_switch(struct i915_hw_context *to)
 	struct intel_ring_buffer *ring = to->ring;
 	struct i915_hw_context *from = ring->last_context;
 	u32 hw_flags = 0;
-	int ret;
+	int ret, i;
 
 	BUG_ON(from != NULL && from->obj != NULL && from->obj->pin_count == 0);
 
-	if (from == to)
+	if (from == to && !to->remap_slice)
 		return 0;
 
 	ret = i915_gem_obj_ggtt_pin(to->obj, CONTEXT_ALIGN, false, false);
@@ -420,8 +439,6 @@ static int do_switch(struct i915_hw_context *to)
 
 	if (!to->is_initialized || is_default_context(to))
 		hw_flags |= MI_RESTORE_INHIBIT;
-	else if (WARN_ON_ONCE(from == to)) /* not yet expected */
-		hw_flags |= MI_FORCE_RESTORE;
 
 	ret = mi_set_context(ring, to, hw_flags);
 	if (ret) {
@@ -429,6 +446,18 @@ static int do_switch(struct i915_hw_context *to)
 		return ret;
 	}
 
+	for (i = 0; i < MAX_L3_SLICES; i++) {
+		if (!(to->remap_slice & (1<<i)))
+			continue;
+
+		ret = i915_gem_l3_remap(ring, i);
+		/* If it failed, try again next round */
+		if (ret)
+			DRM_DEBUG_DRIVER("L3 remapping failed\n");
+		else
+			to->remap_slice &= ~(1<<i);
+	}
+
 	/* The backing object for the context is done after switching to the
 	 * *next* context. Therefore we cannot retire the previous context until
 	 * the next context has already started running. In fact, the below code
@@ -436,11 +465,8 @@ static int do_switch(struct i915_hw_context *to)
 	 * MI_SET_CONTEXT instead of when the next seqno has completed.
 	 */
 	if (from != NULL) {
-		struct drm_i915_private *dev_priv = from->obj->base.dev->dev_private;
-		struct i915_address_space *ggtt = &dev_priv->gtt.base;
 		from->obj->base.read_domains = I915_GEM_DOMAIN_INSTRUCTION;
-		list_move_tail(&i915_gem_obj_to_vma(from->obj, ggtt)->mm_list, &ggtt->active_list);
-		i915_gem_object_move_to_active(from->obj, ring);
+		i915_vma_move_to_active(i915_gem_obj_to_ggtt(from->obj), ring);
 		/* As long as MI_SET_CONTEXT is serializing, ie. it flushes the
 		 * whole damn pipeline, we don't need to explicitly mark the
 		 * object dirty. The only exception is that the context must be
@@ -451,17 +477,7 @@ static int do_switch(struct i915_hw_context *to)
 		from->obj->dirty = 1;
 		BUG_ON(from->obj->ring != ring);
 
-		ret = i915_add_request(ring, NULL);
-		if (ret) {
-			/* Too late, we've already scheduled a context switch.
-			 * Try to undo the change so that the hw state is
-			 * consistent with out tracking. In case of emergency,
-			 * scream.
-			 */
-			WARN_ON(mi_set_context(ring, from, MI_RESTORE_INHIBIT));
-			return ret;
-		}
-
+		/* obj is kept alive until the next request by its active ref */
 		i915_gem_object_unpin(from->obj);
 		i915_gem_context_unreference(from);
 	}
diff --git a/drivers/gpu/drm/i915/i915_gem_evict.c b/drivers/gpu/drm/i915/i915_gem_evict.c
index 91b700155850..b7376533633d 100644
--- a/drivers/gpu/drm/i915/i915_gem_evict.c
+++ b/drivers/gpu/drm/i915/i915_gem_evict.c
@@ -37,6 +37,9 @@ mark_free(struct i915_vma *vma, struct list_head *unwind)
 	if (vma->obj->pin_count)
 		return false;
 
+	if (WARN_ON(!list_empty(&vma->exec_list)))
+		return false;
+
 	list_add(&vma->exec_list, unwind);
 	return drm_mm_scan_add_block(&vma->node);
 }
@@ -113,7 +116,7 @@ none:
 	}
 
 	/* We expect the caller to unpin, evict all and try again, or give up.
-	 * So calling i915_gem_evict_everything() is unnecessary.
+	 * So calling i915_gem_evict_vm() is unnecessary.
 	 */
 	return -ENOSPC;
 
@@ -152,12 +155,48 @@ found:
 	return ret;
 }
 
+/**
+ * i915_gem_evict_vm - Try to free up VM space
+ *
+ * @vm: Address space to evict from
+ * @do_idle: Boolean directing whether to idle first.
+ *
+ * VM eviction is about freeing up virtual address space. If one wants fine
+ * grained eviction, they should see evict something for more details. In terms
+ * of freeing up actual system memory, this function may not accomplish the
+ * desired result. An object may be shared in multiple address space, and this
+ * function will not assert those objects be freed.
+ *
+ * Using do_idle will result in a more complete eviction because it retires, and
+ * inactivates current BOs.
+ */
+int i915_gem_evict_vm(struct i915_address_space *vm, bool do_idle)
+{
+	struct i915_vma *vma, *next;
+	int ret;
+
+	trace_i915_gem_evict_vm(vm);
+
+	if (do_idle) {
+		ret = i915_gpu_idle(vm->dev);
+		if (ret)
+			return ret;
+
+		i915_gem_retire_requests(vm->dev);
+	}
+
+	list_for_each_entry_safe(vma, next, &vm->inactive_list, mm_list)
+		if (vma->obj->pin_count == 0)
+			WARN_ON(i915_vma_unbind(vma));
+
+	return 0;
+}
+
 int
 i915_gem_evict_everything(struct drm_device *dev)
 {
 	drm_i915_private_t *dev_priv = dev->dev_private;
 	struct i915_address_space *vm;
-	struct i915_vma *vma, *next;
 	bool lists_empty = true;
 	int ret;
 
@@ -184,11 +223,8 @@ i915_gem_evict_everything(struct drm_device *dev)
 	i915_gem_retire_requests(dev);
 
 	/* Having flushed everything, unbind() should never raise an error */
-	list_for_each_entry(vm, &dev_priv->vm_list, global_link) {
-		list_for_each_entry_safe(vma, next, &vm->inactive_list, mm_list)
-			if (vma->obj->pin_count == 0)
-				WARN_ON(i915_vma_unbind(vma));
-	}
+	list_for_each_entry(vm, &dev_priv->vm_list, global_link)
+		WARN_ON(i915_gem_evict_vm(vm, false));
 
 	return 0;
 }
diff --git a/drivers/gpu/drm/i915/i915_gem_execbuffer.c b/drivers/gpu/drm/i915/i915_gem_execbuffer.c
index bf345777ae9f..885d595e0e02 100644
--- a/drivers/gpu/drm/i915/i915_gem_execbuffer.c
+++ b/drivers/gpu/drm/i915/i915_gem_execbuffer.c
@@ -33,35 +33,35 @@
 #include "intel_drv.h"
 #include <linux/dma_remapping.h>
 
-struct eb_objects {
-	struct list_head objects;
+struct eb_vmas {
+	struct list_head vmas;
 	int and;
 	union {
-		struct drm_i915_gem_object *lut[0];
+		struct i915_vma *lut[0];
 		struct hlist_head buckets[0];
 	};
 };
 
-static struct eb_objects *
-eb_create(struct drm_i915_gem_execbuffer2 *args)
+static struct eb_vmas *
+eb_create(struct drm_i915_gem_execbuffer2 *args, struct i915_address_space *vm)
 {
-	struct eb_objects *eb = NULL;
+	struct eb_vmas *eb = NULL;
 
 	if (args->flags & I915_EXEC_HANDLE_LUT) {
-		int size = args->buffer_count;
-		size *= sizeof(struct drm_i915_gem_object *);
-		size += sizeof(struct eb_objects);
+		unsigned size = args->buffer_count;
+		size *= sizeof(struct i915_vma *);
+		size += sizeof(struct eb_vmas);
 		eb = kmalloc(size, GFP_TEMPORARY | __GFP_NOWARN | __GFP_NORETRY);
 	}
 
 	if (eb == NULL) {
-		int size = args->buffer_count;
-		int count = PAGE_SIZE / sizeof(struct hlist_head) / 2;
+		unsigned size = args->buffer_count;
+		unsigned count = PAGE_SIZE / sizeof(struct hlist_head) / 2;
 		BUILD_BUG_ON_NOT_POWER_OF_2(PAGE_SIZE / sizeof(struct hlist_head));
 		while (count > 2*size)
 			count >>= 1;
 		eb = kzalloc(count*sizeof(struct hlist_head) +
-			     sizeof(struct eb_objects),
+			     sizeof(struct eb_vmas),
 			     GFP_TEMPORARY);
 		if (eb == NULL)
 			return eb;
@@ -70,64 +70,102 @@ eb_create(struct drm_i915_gem_execbuffer2 *args)
 	} else
 		eb->and = -args->buffer_count;
 
-	INIT_LIST_HEAD(&eb->objects);
+	INIT_LIST_HEAD(&eb->vmas);
 	return eb;
 }
 
 static void
-eb_reset(struct eb_objects *eb)
+eb_reset(struct eb_vmas *eb)
 {
 	if (eb->and >= 0)
 		memset(eb->buckets, 0, (eb->and+1)*sizeof(struct hlist_head));
 }
 
 static int
-eb_lookup_objects(struct eb_objects *eb,
-		  struct drm_i915_gem_exec_object2 *exec,
-		  const struct drm_i915_gem_execbuffer2 *args,
-		  struct drm_file *file)
+eb_lookup_vmas(struct eb_vmas *eb,
+	       struct drm_i915_gem_exec_object2 *exec,
+	       const struct drm_i915_gem_execbuffer2 *args,
+	       struct i915_address_space *vm,
+	       struct drm_file *file)
 {
-	int i;
+	struct drm_i915_gem_object *obj;
+	struct list_head objects;
+	int i, ret = 0;
 
+	INIT_LIST_HEAD(&objects);
 	spin_lock(&file->table_lock);
+	/* Grab a reference to the object and release the lock so we can lookup
+	 * or create the VMA without using GFP_ATOMIC */
 	for (i = 0; i < args->buffer_count; i++) {
-		struct drm_i915_gem_object *obj;
-
 		obj = to_intel_bo(idr_find(&file->object_idr, exec[i].handle));
 		if (obj == NULL) {
 			spin_unlock(&file->table_lock);
 			DRM_DEBUG("Invalid object handle %d at index %d\n",
 				   exec[i].handle, i);
-			return -ENOENT;
+			ret = -ENOENT;
+			goto out;
 		}
 
-		if (!list_empty(&obj->exec_list)) {
+		if (!list_empty(&obj->obj_exec_link)) {
 			spin_unlock(&file->table_lock);
 			DRM_DEBUG("Object %p [handle %d, index %d] appears more than once in object list\n",
 				   obj, exec[i].handle, i);
-			return -EINVAL;
+			ret = -EINVAL;
+			goto out;
 		}
 
 		drm_gem_object_reference(&obj->base);
-		list_add_tail(&obj->exec_list, &eb->objects);
+		list_add_tail(&obj->obj_exec_link, &objects);
+	}
+	spin_unlock(&file->table_lock);
 
-		obj->exec_entry = &exec[i];
+	i = 0;
+	list_for_each_entry(obj, &objects, obj_exec_link) {
+		struct i915_vma *vma;
+
+		/*
+		 * NOTE: We can leak any vmas created here when something fails
+		 * later on. But that's no issue since vma_unbind can deal with
+		 * vmas which are not actually bound. And since only
+		 * lookup_or_create exists as an interface to get at the vma
+		 * from the (obj, vm) we don't run the risk of creating
+		 * duplicated vmas for the same vm.
+		 */
+		vma = i915_gem_obj_lookup_or_create_vma(obj, vm);
+		if (IS_ERR(vma)) {
+			DRM_DEBUG("Failed to lookup VMA\n");
+			ret = PTR_ERR(vma);
+			goto out;
+		}
+
+		list_add_tail(&vma->exec_list, &eb->vmas);
+
+		vma->exec_entry = &exec[i];
 		if (eb->and < 0) {
-			eb->lut[i] = obj;
+			eb->lut[i] = vma;
 		} else {
 			uint32_t handle = args->flags & I915_EXEC_HANDLE_LUT ? i : exec[i].handle;
-			obj->exec_handle = handle;
-			hlist_add_head(&obj->exec_node,
+			vma->exec_handle = handle;
+			hlist_add_head(&vma->exec_node,
 				       &eb->buckets[handle & eb->and]);
 		}
+		++i;
 	}
-	spin_unlock(&file->table_lock);
 
-	return 0;
+
+out:
+	while (!list_empty(&objects)) {
+		obj = list_first_entry(&objects,
+				       struct drm_i915_gem_object,
+				       obj_exec_link);
+		list_del_init(&obj->obj_exec_link);
+		if (ret)
+			drm_gem_object_unreference(&obj->base);
+	}
+	return ret;
 }
 
-static struct drm_i915_gem_object *
-eb_get_object(struct eb_objects *eb, unsigned long handle)
+static struct i915_vma *eb_get_vma(struct eb_vmas *eb, unsigned long handle)
 {
 	if (eb->and < 0) {
 		if (handle >= -eb->and)
@@ -139,34 +177,33 @@ eb_get_object(struct eb_objects *eb, unsigned long handle)
 
 		head = &eb->buckets[handle & eb->and];
 		hlist_for_each(node, head) {
-			struct drm_i915_gem_object *obj;
+			struct i915_vma *vma;
 
-			obj = hlist_entry(node, struct drm_i915_gem_object, exec_node);
-			if (obj->exec_handle == handle)
-				return obj;
+			vma = hlist_entry(node, struct i915_vma, exec_node);
+			if (vma->exec_handle == handle)
+				return vma;
 		}
 		return NULL;
 	}
 }
 
-static void
-eb_destroy(struct eb_objects *eb)
-{
-	while (!list_empty(&eb->objects)) {
-		struct drm_i915_gem_object *obj;
+static void eb_destroy(struct eb_vmas *eb) {
+	while (!list_empty(&eb->vmas)) {
+		struct i915_vma *vma;
 
-		obj = list_first_entry(&eb->objects,
-				       struct drm_i915_gem_object,
+		vma = list_first_entry(&eb->vmas,
+				       struct i915_vma,
 				       exec_list);
-		list_del_init(&obj->exec_list);
-		drm_gem_object_unreference(&obj->base);
+		list_del_init(&vma->exec_list);
+		drm_gem_object_unreference(&vma->obj->base);
 	}
 	kfree(eb);
 }
 
 static inline int use_cpu_reloc(struct drm_i915_gem_object *obj)
 {
-	return (obj->base.write_domain == I915_GEM_DOMAIN_CPU ||
+	return (HAS_LLC(obj->base.dev) ||
+		obj->base.write_domain == I915_GEM_DOMAIN_CPU ||
 		!obj->map_and_fenceable ||
 		obj->cache_level != I915_CACHE_NONE);
 }
@@ -175,17 +212,31 @@ static int
 relocate_entry_cpu(struct drm_i915_gem_object *obj,
 		   struct drm_i915_gem_relocation_entry *reloc)
 {
+	struct drm_device *dev = obj->base.dev;
 	uint32_t page_offset = offset_in_page(reloc->offset);
 	char *vaddr;
 	int ret = -EINVAL;
 
-	ret = i915_gem_object_set_to_cpu_domain(obj, 1);
+	ret = i915_gem_object_set_to_cpu_domain(obj, true);
 	if (ret)
 		return ret;
 
 	vaddr = kmap_atomic(i915_gem_object_get_page(obj,
 				reloc->offset >> PAGE_SHIFT));
 	*(uint32_t *)(vaddr + page_offset) = reloc->delta;
+
+	if (INTEL_INFO(dev)->gen >= 8) {
+		page_offset = offset_in_page(page_offset + sizeof(uint32_t));
+
+		if (page_offset == 0) {
+			kunmap_atomic(vaddr);
+			vaddr = kmap_atomic(i915_gem_object_get_page(obj,
+			    (reloc->offset + sizeof(uint32_t)) >> PAGE_SHIFT));
+		}
+
+		*(uint32_t *)(vaddr + page_offset) = 0;
+	}
+
 	kunmap_atomic(vaddr);
 
 	return 0;
@@ -216,6 +267,21 @@ relocate_entry_gtt(struct drm_i915_gem_object *obj,
 	reloc_entry = (uint32_t __iomem *)
 		(reloc_page + offset_in_page(reloc->offset));
 	iowrite32(reloc->delta, reloc_entry);
+
+	if (INTEL_INFO(dev)->gen >= 8) {
+		reloc_entry += 1;
+
+		if (offset_in_page(reloc->offset + sizeof(uint32_t)) == 0) {
+			io_mapping_unmap_atomic(reloc_page);
+			reloc_page = io_mapping_map_atomic_wc(
+					dev_priv->gtt.mappable,
+					reloc->offset + sizeof(uint32_t));
+			reloc_entry = reloc_page;
+		}
+
+		iowrite32(0, reloc_entry);
+	}
+
 	io_mapping_unmap_atomic(reloc_page);
 
 	return 0;
@@ -223,22 +289,24 @@ relocate_entry_gtt(struct drm_i915_gem_object *obj,
 
 static int
 i915_gem_execbuffer_relocate_entry(struct drm_i915_gem_object *obj,
-				   struct eb_objects *eb,
+				   struct eb_vmas *eb,
 				   struct drm_i915_gem_relocation_entry *reloc,
 				   struct i915_address_space *vm)
 {
 	struct drm_device *dev = obj->base.dev;
 	struct drm_gem_object *target_obj;
 	struct drm_i915_gem_object *target_i915_obj;
+	struct i915_vma *target_vma;
 	uint32_t target_offset;
 	int ret = -EINVAL;
 
 	/* we've already hold a reference to all valid objects */
-	target_obj = &eb_get_object(eb, reloc->target_handle)->base;
-	if (unlikely(target_obj == NULL))
+	target_vma = eb_get_vma(eb, reloc->target_handle);
+	if (unlikely(target_vma == NULL))
 		return -ENOENT;
+	target_i915_obj = target_vma->obj;
+	target_obj = &target_vma->obj->base;
 
-	target_i915_obj = to_intel_bo(target_obj);
 	target_offset = i915_gem_obj_ggtt_offset(target_i915_obj);
 
 	/* Sandybridge PPGTT errata: We need a global gtt mapping for MI and
@@ -284,7 +352,8 @@ i915_gem_execbuffer_relocate_entry(struct drm_i915_gem_object *obj,
 		return 0;
 
 	/* Check that the relocation address is valid... */
-	if (unlikely(reloc->offset > obj->base.size - 4)) {
+	if (unlikely(reloc->offset >
+		obj->base.size - (INTEL_INFO(dev)->gen >= 8 ? 8 : 4))) {
 		DRM_DEBUG("Relocation beyond object bounds: "
 			  "obj %p target %d offset %d size %d.\n",
 			  obj, reloc->target_handle,
@@ -320,14 +389,13 @@ i915_gem_execbuffer_relocate_entry(struct drm_i915_gem_object *obj,
 }
 
 static int
-i915_gem_execbuffer_relocate_object(struct drm_i915_gem_object *obj,
-				    struct eb_objects *eb,
-				    struct i915_address_space *vm)
+i915_gem_execbuffer_relocate_vma(struct i915_vma *vma,
+				 struct eb_vmas *eb)
 {
 #define N_RELOC(x) ((x) / sizeof(struct drm_i915_gem_relocation_entry))
 	struct drm_i915_gem_relocation_entry stack_reloc[N_RELOC(512)];
 	struct drm_i915_gem_relocation_entry __user *user_relocs;
-	struct drm_i915_gem_exec_object2 *entry = obj->exec_entry;
+	struct drm_i915_gem_exec_object2 *entry = vma->exec_entry;
 	int remain, ret;
 
 	user_relocs = to_user_ptr(entry->relocs_ptr);
@@ -346,8 +414,8 @@ i915_gem_execbuffer_relocate_object(struct drm_i915_gem_object *obj,
 		do {
 			u64 offset = r->presumed_offset;
 
-			ret = i915_gem_execbuffer_relocate_entry(obj, eb, r,
-								 vm);
+			ret = i915_gem_execbuffer_relocate_entry(vma->obj, eb, r,
+								 vma->vm);
 			if (ret)
 				return ret;
 
@@ -368,17 +436,16 @@ i915_gem_execbuffer_relocate_object(struct drm_i915_gem_object *obj,
 }
 
 static int
-i915_gem_execbuffer_relocate_object_slow(struct drm_i915_gem_object *obj,
-					 struct eb_objects *eb,
-					 struct drm_i915_gem_relocation_entry *relocs,
-					 struct i915_address_space *vm)
+i915_gem_execbuffer_relocate_vma_slow(struct i915_vma *vma,
+				      struct eb_vmas *eb,
+				      struct drm_i915_gem_relocation_entry *relocs)
 {
-	const struct drm_i915_gem_exec_object2 *entry = obj->exec_entry;
+	const struct drm_i915_gem_exec_object2 *entry = vma->exec_entry;
 	int i, ret;
 
 	for (i = 0; i < entry->relocation_count; i++) {
-		ret = i915_gem_execbuffer_relocate_entry(obj, eb, &relocs[i],
-							 vm);
+		ret = i915_gem_execbuffer_relocate_entry(vma->obj, eb, &relocs[i],
+							 vma->vm);
 		if (ret)
 			return ret;
 	}
@@ -387,10 +454,10 @@ i915_gem_execbuffer_relocate_object_slow(struct drm_i915_gem_object *obj,
 }
 
 static int
-i915_gem_execbuffer_relocate(struct eb_objects *eb,
+i915_gem_execbuffer_relocate(struct eb_vmas *eb,
 			     struct i915_address_space *vm)
 {
-	struct drm_i915_gem_object *obj;
+	struct i915_vma *vma;
 	int ret = 0;
 
 	/* This is the fast path and we cannot handle a pagefault whilst
@@ -401,8 +468,8 @@ i915_gem_execbuffer_relocate(struct eb_objects *eb,
 	 * lockdep complains vehemently.
 	 */
 	pagefault_disable();
-	list_for_each_entry(obj, &eb->objects, exec_list) {
-		ret = i915_gem_execbuffer_relocate_object(obj, eb, vm);
+	list_for_each_entry(vma, &eb->vmas, exec_list) {
+		ret = i915_gem_execbuffer_relocate_vma(vma, eb);
 		if (ret)
 			break;
 	}
@@ -415,31 +482,32 @@ i915_gem_execbuffer_relocate(struct eb_objects *eb,
 #define  __EXEC_OBJECT_HAS_FENCE (1<<30)
 
 static int
-need_reloc_mappable(struct drm_i915_gem_object *obj)
+need_reloc_mappable(struct i915_vma *vma)
 {
-	struct drm_i915_gem_exec_object2 *entry = obj->exec_entry;
-	return entry->relocation_count && !use_cpu_reloc(obj);
+	struct drm_i915_gem_exec_object2 *entry = vma->exec_entry;
+	return entry->relocation_count && !use_cpu_reloc(vma->obj) &&
+		i915_is_ggtt(vma->vm);
 }
 
 static int
-i915_gem_execbuffer_reserve_object(struct drm_i915_gem_object *obj,
-				   struct intel_ring_buffer *ring,
-				   struct i915_address_space *vm,
-				   bool *need_reloc)
+i915_gem_execbuffer_reserve_vma(struct i915_vma *vma,
+				struct intel_ring_buffer *ring,
+				bool *need_reloc)
 {
-	struct drm_i915_private *dev_priv = obj->base.dev->dev_private;
-	struct drm_i915_gem_exec_object2 *entry = obj->exec_entry;
+	struct drm_i915_private *dev_priv = ring->dev->dev_private;
+	struct drm_i915_gem_exec_object2 *entry = vma->exec_entry;
 	bool has_fenced_gpu_access = INTEL_INFO(ring->dev)->gen < 4;
 	bool need_fence, need_mappable;
+	struct drm_i915_gem_object *obj = vma->obj;
 	int ret;
 
 	need_fence =
 		has_fenced_gpu_access &&
 		entry->flags & EXEC_OBJECT_NEEDS_FENCE &&
 		obj->tiling_mode != I915_TILING_NONE;
-	need_mappable = need_fence || need_reloc_mappable(obj);
+	need_mappable = need_fence || need_reloc_mappable(vma);
 
-	ret = i915_gem_object_pin(obj, vm, entry->alignment, need_mappable,
+	ret = i915_gem_object_pin(obj, vma->vm, entry->alignment, need_mappable,
 				  false);
 	if (ret)
 		return ret;
@@ -467,8 +535,8 @@ i915_gem_execbuffer_reserve_object(struct drm_i915_gem_object *obj,
 		obj->has_aliasing_ppgtt_mapping = 1;
 	}
 
-	if (entry->offset != i915_gem_obj_offset(obj, vm)) {
-		entry->offset = i915_gem_obj_offset(obj, vm);
+	if (entry->offset != vma->node.start) {
+		entry->offset = vma->node.start;
 		*need_reloc = true;
 	}
 
@@ -485,14 +553,15 @@ i915_gem_execbuffer_reserve_object(struct drm_i915_gem_object *obj,
 }
 
 static void
-i915_gem_execbuffer_unreserve_object(struct drm_i915_gem_object *obj)
+i915_gem_execbuffer_unreserve_vma(struct i915_vma *vma)
 {
 	struct drm_i915_gem_exec_object2 *entry;
+	struct drm_i915_gem_object *obj = vma->obj;
 
-	if (!i915_gem_obj_bound_any(obj))
+	if (!drm_mm_node_allocated(&vma->node))
 		return;
 
-	entry = obj->exec_entry;
+	entry = vma->exec_entry;
 
 	if (entry->flags & __EXEC_OBJECT_HAS_FENCE)
 		i915_gem_object_unpin_fence(obj);
@@ -505,41 +574,46 @@ i915_gem_execbuffer_unreserve_object(struct drm_i915_gem_object *obj)
 
 static int
 i915_gem_execbuffer_reserve(struct intel_ring_buffer *ring,
-			    struct list_head *objects,
-			    struct i915_address_space *vm,
+			    struct list_head *vmas,
 			    bool *need_relocs)
 {
 	struct drm_i915_gem_object *obj;
-	struct list_head ordered_objects;
+	struct i915_vma *vma;
+	struct i915_address_space *vm;
+	struct list_head ordered_vmas;
 	bool has_fenced_gpu_access = INTEL_INFO(ring->dev)->gen < 4;
 	int retry;
 
-	INIT_LIST_HEAD(&ordered_objects);
-	while (!list_empty(objects)) {
+	if (list_empty(vmas))
+		return 0;
+
+	vm = list_first_entry(vmas, struct i915_vma, exec_list)->vm;
+
+	INIT_LIST_HEAD(&ordered_vmas);
+	while (!list_empty(vmas)) {
 		struct drm_i915_gem_exec_object2 *entry;
 		bool need_fence, need_mappable;
 
-		obj = list_first_entry(objects,
-				       struct drm_i915_gem_object,
-				       exec_list);
-		entry = obj->exec_entry;
+		vma = list_first_entry(vmas, struct i915_vma, exec_list);
+		obj = vma->obj;
+		entry = vma->exec_entry;
 
 		need_fence =
 			has_fenced_gpu_access &&
 			entry->flags & EXEC_OBJECT_NEEDS_FENCE &&
 			obj->tiling_mode != I915_TILING_NONE;
-		need_mappable = need_fence || need_reloc_mappable(obj);
+		need_mappable = need_fence || need_reloc_mappable(vma);
 
 		if (need_mappable)
-			list_move(&obj->exec_list, &ordered_objects);
+			list_move(&vma->exec_list, &ordered_vmas);
 		else
-			list_move_tail(&obj->exec_list, &ordered_objects);
+			list_move_tail(&vma->exec_list, &ordered_vmas);
 
 		obj->base.pending_read_domains = I915_GEM_GPU_DOMAINS & ~I915_GEM_DOMAIN_COMMAND;
 		obj->base.pending_write_domain = 0;
 		obj->pending_fenced_gpu_access = false;
 	}
-	list_splice(&ordered_objects, objects);
+	list_splice(&ordered_vmas, vmas);
 
 	/* Attempt to pin all of the buffers into the GTT.
 	 * This is done in 3 phases:
@@ -558,52 +632,52 @@ i915_gem_execbuffer_reserve(struct intel_ring_buffer *ring,
 		int ret = 0;
 
 		/* Unbind any ill-fitting objects or pin. */
-		list_for_each_entry(obj, objects, exec_list) {
-			struct drm_i915_gem_exec_object2 *entry = obj->exec_entry;
+		list_for_each_entry(vma, vmas, exec_list) {
+			struct drm_i915_gem_exec_object2 *entry = vma->exec_entry;
 			bool need_fence, need_mappable;
-			u32 obj_offset;
 
-			if (!i915_gem_obj_bound(obj, vm))
+			obj = vma->obj;
+
+			if (!drm_mm_node_allocated(&vma->node))
 				continue;
 
-			obj_offset = i915_gem_obj_offset(obj, vm);
 			need_fence =
 				has_fenced_gpu_access &&
 				entry->flags & EXEC_OBJECT_NEEDS_FENCE &&
 				obj->tiling_mode != I915_TILING_NONE;
-			need_mappable = need_fence || need_reloc_mappable(obj);
+			need_mappable = need_fence || need_reloc_mappable(vma);
 
 			WARN_ON((need_mappable || need_fence) &&
-				!i915_is_ggtt(vm));
+			       !i915_is_ggtt(vma->vm));
 
 			if ((entry->alignment &&
-			     obj_offset & (entry->alignment - 1)) ||
+			     vma->node.start & (entry->alignment - 1)) ||
 			    (need_mappable && !obj->map_and_fenceable))
-				ret = i915_vma_unbind(i915_gem_obj_to_vma(obj, vm));
+				ret = i915_vma_unbind(vma);
 			else
-				ret = i915_gem_execbuffer_reserve_object(obj, ring, vm, need_relocs);
+				ret = i915_gem_execbuffer_reserve_vma(vma, ring, need_relocs);
 			if (ret)
 				goto err;
 		}
 
 		/* Bind fresh objects */
-		list_for_each_entry(obj, objects, exec_list) {
-			if (i915_gem_obj_bound(obj, vm))
+		list_for_each_entry(vma, vmas, exec_list) {
+			if (drm_mm_node_allocated(&vma->node))
 				continue;
 
-			ret = i915_gem_execbuffer_reserve_object(obj, ring, vm, need_relocs);
+			ret = i915_gem_execbuffer_reserve_vma(vma, ring, need_relocs);
 			if (ret)
 				goto err;
 		}
 
 err:		/* Decrement pin count for bound objects */
-		list_for_each_entry(obj, objects, exec_list)
-			i915_gem_execbuffer_unreserve_object(obj);
+		list_for_each_entry(vma, vmas, exec_list)
+			i915_gem_execbuffer_unreserve_vma(vma);
 
 		if (ret != -ENOSPC || retry++)
 			return ret;
 
-		ret = i915_gem_evict_everything(ring->dev);
+		ret = i915_gem_evict_vm(vm, true);
 		if (ret)
 			return ret;
 	} while (1);
@@ -614,24 +688,27 @@ i915_gem_execbuffer_relocate_slow(struct drm_device *dev,
 				  struct drm_i915_gem_execbuffer2 *args,
 				  struct drm_file *file,
 				  struct intel_ring_buffer *ring,
-				  struct eb_objects *eb,
-				  struct drm_i915_gem_exec_object2 *exec,
-				  struct i915_address_space *vm)
+				  struct eb_vmas *eb,
+				  struct drm_i915_gem_exec_object2 *exec)
 {
 	struct drm_i915_gem_relocation_entry *reloc;
-	struct drm_i915_gem_object *obj;
+	struct i915_address_space *vm;
+	struct i915_vma *vma;
 	bool need_relocs;
 	int *reloc_offset;
 	int i, total, ret;
-	int count = args->buffer_count;
+	unsigned count = args->buffer_count;
+
+	if (WARN_ON(list_empty(&eb->vmas)))
+		return 0;
+
+	vm = list_first_entry(&eb->vmas, struct i915_vma, exec_list)->vm;
 
 	/* We may process another execbuffer during the unlock... */
-	while (!list_empty(&eb->objects)) {
-		obj = list_first_entry(&eb->objects,
-				       struct drm_i915_gem_object,
-				       exec_list);
-		list_del_init(&obj->exec_list);
-		drm_gem_object_unreference(&obj->base);
+	while (!list_empty(&eb->vmas)) {
+		vma = list_first_entry(&eb->vmas, struct i915_vma, exec_list);
+		list_del_init(&vma->exec_list);
+		drm_gem_object_unreference(&vma->obj->base);
 	}
 
 	mutex_unlock(&dev->struct_mutex);
@@ -695,20 +772,19 @@ i915_gem_execbuffer_relocate_slow(struct drm_device *dev,
 
 	/* reacquire the objects */
 	eb_reset(eb);
-	ret = eb_lookup_objects(eb, exec, args, file);
+	ret = eb_lookup_vmas(eb, exec, args, vm, file);
 	if (ret)
 		goto err;
 
 	need_relocs = (args->flags & I915_EXEC_NO_RELOC) == 0;
-	ret = i915_gem_execbuffer_reserve(ring, &eb->objects, vm, &need_relocs);
+	ret = i915_gem_execbuffer_reserve(ring, &eb->vmas, &need_relocs);
 	if (ret)
 		goto err;
 
-	list_for_each_entry(obj, &eb->objects, exec_list) {
-		int offset = obj->exec_entry - exec;
-		ret = i915_gem_execbuffer_relocate_object_slow(obj, eb,
-							       reloc + reloc_offset[offset],
-							       vm);
+	list_for_each_entry(vma, &eb->vmas, exec_list) {
+		int offset = vma->exec_entry - exec;
+		ret = i915_gem_execbuffer_relocate_vma_slow(vma, eb,
+							    reloc + reloc_offset[offset]);
 		if (ret)
 			goto err;
 	}
@@ -727,14 +803,15 @@ err:
 
 static int
 i915_gem_execbuffer_move_to_gpu(struct intel_ring_buffer *ring,
-				struct list_head *objects)
+				struct list_head *vmas)
 {
-	struct drm_i915_gem_object *obj;
+	struct i915_vma *vma;
 	uint32_t flush_domains = 0;
 	bool flush_chipset = false;
 	int ret;
 
-	list_for_each_entry(obj, objects, exec_list) {
+	list_for_each_entry(vma, vmas, exec_list) {
+		struct drm_i915_gem_object *obj = vma->obj;
 		ret = i915_gem_object_sync(obj, ring);
 		if (ret)
 			return ret;
@@ -771,8 +848,8 @@ validate_exec_list(struct drm_i915_gem_exec_object2 *exec,
 		   int count)
 {
 	int i;
-	int relocs_total = 0;
-	int relocs_max = INT_MAX / sizeof(struct drm_i915_gem_relocation_entry);
+	unsigned relocs_total = 0;
+	unsigned relocs_max = UINT_MAX / sizeof(struct drm_i915_gem_relocation_entry);
 
 	for (i = 0; i < count; i++) {
 		char __user *ptr = to_user_ptr(exec[i].relocs_ptr);
@@ -809,13 +886,13 @@ validate_exec_list(struct drm_i915_gem_exec_object2 *exec,
 }
 
 static void
-i915_gem_execbuffer_move_to_active(struct list_head *objects,
-				   struct i915_address_space *vm,
+i915_gem_execbuffer_move_to_active(struct list_head *vmas,
 				   struct intel_ring_buffer *ring)
 {
-	struct drm_i915_gem_object *obj;
+	struct i915_vma *vma;
 
-	list_for_each_entry(obj, objects, exec_list) {
+	list_for_each_entry(vma, vmas, exec_list) {
+		struct drm_i915_gem_object *obj = vma->obj;
 		u32 old_read = obj->base.read_domains;
 		u32 old_write = obj->base.write_domain;
 
@@ -825,9 +902,7 @@ i915_gem_execbuffer_move_to_active(struct list_head *objects,
 		obj->base.read_domains = obj->base.pending_read_domains;
 		obj->fenced_gpu_access = obj->pending_fenced_gpu_access;
 
-		/* FIXME: This lookup gets fixed later <-- danvet */
-		list_move_tail(&i915_gem_obj_to_vma(obj, vm)->mm_list, &vm->active_list);
-		i915_gem_object_move_to_active(obj, ring);
+		i915_vma_move_to_active(vma, ring);
 		if (obj->base.write_domain) {
 			obj->dirty = 1;
 			obj->last_write_seqno = intel_ring_get_seqno(ring);
@@ -885,10 +960,11 @@ i915_gem_do_execbuffer(struct drm_device *dev, void *data,
 		       struct i915_address_space *vm)
 {
 	drm_i915_private_t *dev_priv = dev->dev_private;
-	struct eb_objects *eb;
+	struct eb_vmas *eb;
 	struct drm_i915_gem_object *batch_obj;
 	struct drm_clip_rect *cliprects = NULL;
 	struct intel_ring_buffer *ring;
+	struct i915_ctx_hang_stats *hs;
 	u32 ctx_id = i915_execbuffer2_get_context_id(*args);
 	u32 exec_start, exec_len;
 	u32 mask, flags;
@@ -1000,7 +1076,8 @@ i915_gem_do_execbuffer(struct drm_device *dev, void *data,
 			return -EINVAL;
 		}
 
-		cliprects = kmalloc(args->num_cliprects * sizeof(*cliprects),
+		cliprects = kcalloc(args->num_cliprects,
+				    sizeof(*cliprects),
 				    GFP_KERNEL);
 		if (cliprects == NULL) {
 			ret = -ENOMEM;
@@ -1025,7 +1102,7 @@ i915_gem_do_execbuffer(struct drm_device *dev, void *data,
 		goto pre_mutex_err;
 	}
 
-	eb = eb_create(args);
+	eb = eb_create(args, vm);
 	if (eb == NULL) {
 		mutex_unlock(&dev->struct_mutex);
 		ret = -ENOMEM;
@@ -1033,18 +1110,16 @@ i915_gem_do_execbuffer(struct drm_device *dev, void *data,
 	}
 
 	/* Look up object handles */
-	ret = eb_lookup_objects(eb, exec, args, file);
+	ret = eb_lookup_vmas(eb, exec, args, vm, file);
 	if (ret)
 		goto err;
 
 	/* take note of the batch buffer before we might reorder the lists */
-	batch_obj = list_entry(eb->objects.prev,
-			       struct drm_i915_gem_object,
-			       exec_list);
+	batch_obj = list_entry(eb->vmas.prev, struct i915_vma, exec_list)->obj;
 
 	/* Move the objects en-masse into the GTT, evicting if necessary. */
 	need_relocs = (args->flags & I915_EXEC_NO_RELOC) == 0;
-	ret = i915_gem_execbuffer_reserve(ring, &eb->objects, vm, &need_relocs);
+	ret = i915_gem_execbuffer_reserve(ring, &eb->vmas, &need_relocs);
 	if (ret)
 		goto err;
 
@@ -1054,7 +1129,7 @@ i915_gem_do_execbuffer(struct drm_device *dev, void *data,
 	if (ret) {
 		if (ret == -EFAULT) {
 			ret = i915_gem_execbuffer_relocate_slow(dev, args, file, ring,
-								eb, exec, vm);
+								eb, exec);
 			BUG_ON(!mutex_is_locked(&dev->struct_mutex));
 		}
 		if (ret)
@@ -1071,15 +1146,25 @@ i915_gem_do_execbuffer(struct drm_device *dev, void *data,
 
 	/* snb/ivb/vlv conflate the "batch in ppgtt" bit with the "non-secure
 	 * batch" bit. Hence we need to pin secure batches into the global gtt.
-	 * hsw should have this fixed, but let's be paranoid and do it
-	 * unconditionally for now. */
+	 * hsw should have this fixed, but bdw mucks it up again. */
 	if (flags & I915_DISPATCH_SECURE && !batch_obj->has_global_gtt_mapping)
 		i915_gem_gtt_bind_object(batch_obj, batch_obj->cache_level);
 
-	ret = i915_gem_execbuffer_move_to_gpu(ring, &eb->objects);
+	ret = i915_gem_execbuffer_move_to_gpu(ring, &eb->vmas);
 	if (ret)
 		goto err;
 
+	hs = i915_gem_context_get_hang_stats(dev, file, ctx_id);
+	if (IS_ERR(hs)) {
+		ret = PTR_ERR(hs);
+		goto err;
+	}
+
+	if (hs->banned) {
+		ret = -EIO;
+		goto err;
+	}
+
 	ret = i915_switch_context(ring, file, ctx_id);
 	if (ret)
 		goto err;
@@ -1131,7 +1216,7 @@ i915_gem_do_execbuffer(struct drm_device *dev, void *data,
 
 	trace_i915_gem_ring_dispatch(ring, intel_ring_get_seqno(ring), flags);
 
-	i915_gem_execbuffer_move_to_active(&eb->objects, vm, ring);
+	i915_gem_execbuffer_move_to_active(&eb->vmas, ring);
 	i915_gem_execbuffer_retire_commands(dev, file, ring, batch_obj);
 
 err:
diff --git a/drivers/gpu/drm/i915/i915_gem_gtt.c b/drivers/gpu/drm/i915/i915_gem_gtt.c
index 1f7b4caefb6e..3620a1b0a73c 100644
--- a/drivers/gpu/drm/i915/i915_gem_gtt.c
+++ b/drivers/gpu/drm/i915/i915_gem_gtt.c
@@ -30,6 +30,8 @@
 
 #define GEN6_PPGTT_PD_ENTRIES 512
 #define I915_PPGTT_PT_ENTRIES (PAGE_SIZE / sizeof(gen6_gtt_pte_t))
+typedef uint64_t gen8_gtt_pte_t;
+typedef gen8_gtt_pte_t gen8_ppgtt_pde_t;
 
 /* PPGTT stuff */
 #define GEN6_GTT_ADDR_ENCODE(addr)	((addr) | (((addr) >> 28) & 0xff0))
@@ -57,6 +59,41 @@
 #define HSW_WB_ELLC_LLC_AGE0		HSW_CACHEABILITY_CONTROL(0xb)
 #define HSW_WT_ELLC_LLC_AGE0		HSW_CACHEABILITY_CONTROL(0x6)
 
+#define GEN8_PTES_PER_PAGE		(PAGE_SIZE / sizeof(gen8_gtt_pte_t))
+#define GEN8_PDES_PER_PAGE		(PAGE_SIZE / sizeof(gen8_ppgtt_pde_t))
+#define GEN8_LEGACY_PDPS		4
+
+#define PPAT_UNCACHED_INDEX		(_PAGE_PWT | _PAGE_PCD)
+#define PPAT_CACHED_PDE_INDEX		0 /* WB LLC */
+#define PPAT_CACHED_INDEX		_PAGE_PAT /* WB LLCeLLC */
+#define PPAT_DISPLAY_ELLC_INDEX		_PAGE_PCD /* WT eLLC */
+
+static inline gen8_gtt_pte_t gen8_pte_encode(dma_addr_t addr,
+					     enum i915_cache_level level,
+					     bool valid)
+{
+	gen8_gtt_pte_t pte = valid ? _PAGE_PRESENT | _PAGE_RW : 0;
+	pte |= addr;
+	if (level != I915_CACHE_NONE)
+		pte |= PPAT_CACHED_INDEX;
+	else
+		pte |= PPAT_UNCACHED_INDEX;
+	return pte;
+}
+
+static inline gen8_ppgtt_pde_t gen8_pde_encode(struct drm_device *dev,
+					     dma_addr_t addr,
+					     enum i915_cache_level level)
+{
+	gen8_ppgtt_pde_t pde = _PAGE_PRESENT | _PAGE_RW;
+	pde |= addr;
+	if (level != I915_CACHE_NONE)
+		pde |= PPAT_CACHED_PDE_INDEX;
+	else
+		pde |= PPAT_UNCACHED_INDEX;
+	return pde;
+}
+
 static gen6_gtt_pte_t snb_pte_encode(dma_addr_t addr,
 				     enum i915_cache_level level,
 				     bool valid)
@@ -158,6 +195,257 @@ static gen6_gtt_pte_t iris_pte_encode(dma_addr_t addr,
 	return pte;
 }
 
+/* Broadwell Page Directory Pointer Descriptors */
+static int gen8_write_pdp(struct intel_ring_buffer *ring, unsigned entry,
+			   uint64_t val)
+{
+	int ret;
+
+	BUG_ON(entry >= 4);
+
+	ret = intel_ring_begin(ring, 6);
+	if (ret)
+		return ret;
+
+	intel_ring_emit(ring, MI_LOAD_REGISTER_IMM(1));
+	intel_ring_emit(ring, GEN8_RING_PDP_UDW(ring, entry));
+	intel_ring_emit(ring, (u32)(val >> 32));
+	intel_ring_emit(ring, MI_LOAD_REGISTER_IMM(1));
+	intel_ring_emit(ring, GEN8_RING_PDP_LDW(ring, entry));
+	intel_ring_emit(ring, (u32)(val));
+	intel_ring_advance(ring);
+
+	return 0;
+}
+
+static int gen8_ppgtt_enable(struct drm_device *dev)
+{
+	struct drm_i915_private *dev_priv = dev->dev_private;
+	struct intel_ring_buffer *ring;
+	struct i915_hw_ppgtt *ppgtt = dev_priv->mm.aliasing_ppgtt;
+	int i, j, ret;
+
+	/* bit of a hack to find the actual last used pd */
+	int used_pd = ppgtt->num_pd_entries / GEN8_PDES_PER_PAGE;
+
+	for_each_ring(ring, dev_priv, j) {
+		I915_WRITE(RING_MODE_GEN7(ring),
+			   _MASKED_BIT_ENABLE(GFX_PPGTT_ENABLE));
+	}
+
+	for (i = used_pd - 1; i >= 0; i--) {
+		dma_addr_t addr = ppgtt->pd_dma_addr[i];
+		for_each_ring(ring, dev_priv, j) {
+			ret = gen8_write_pdp(ring, i, addr);
+			if (ret)
+				return ret;
+		}
+	}
+	return 0;
+}
+
+static void gen8_ppgtt_clear_range(struct i915_address_space *vm,
+				   unsigned first_entry,
+				   unsigned num_entries,
+				   bool use_scratch)
+{
+	struct i915_hw_ppgtt *ppgtt =
+		container_of(vm, struct i915_hw_ppgtt, base);
+	gen8_gtt_pte_t *pt_vaddr, scratch_pte;
+	unsigned act_pt = first_entry / GEN8_PTES_PER_PAGE;
+	unsigned first_pte = first_entry % GEN8_PTES_PER_PAGE;
+	unsigned last_pte, i;
+
+	scratch_pte = gen8_pte_encode(ppgtt->base.scratch.addr,
+				      I915_CACHE_LLC, use_scratch);
+
+	while (num_entries) {
+		struct page *page_table = &ppgtt->gen8_pt_pages[act_pt];
+
+		last_pte = first_pte + num_entries;
+		if (last_pte > GEN8_PTES_PER_PAGE)
+			last_pte = GEN8_PTES_PER_PAGE;
+
+		pt_vaddr = kmap_atomic(page_table);
+
+		for (i = first_pte; i < last_pte; i++)
+			pt_vaddr[i] = scratch_pte;
+
+		kunmap_atomic(pt_vaddr);
+
+		num_entries -= last_pte - first_pte;
+		first_pte = 0;
+		act_pt++;
+	}
+}
+
+static void gen8_ppgtt_insert_entries(struct i915_address_space *vm,
+				      struct sg_table *pages,
+				      unsigned first_entry,
+				      enum i915_cache_level cache_level)
+{
+	struct i915_hw_ppgtt *ppgtt =
+		container_of(vm, struct i915_hw_ppgtt, base);
+	gen8_gtt_pte_t *pt_vaddr;
+	unsigned act_pt = first_entry / GEN8_PTES_PER_PAGE;
+	unsigned act_pte = first_entry % GEN8_PTES_PER_PAGE;
+	struct sg_page_iter sg_iter;
+
+	pt_vaddr = kmap_atomic(&ppgtt->gen8_pt_pages[act_pt]);
+	for_each_sg_page(pages->sgl, &sg_iter, pages->nents, 0) {
+		dma_addr_t page_addr;
+
+		page_addr = sg_dma_address(sg_iter.sg) +
+				(sg_iter.sg_pgoffset << PAGE_SHIFT);
+		pt_vaddr[act_pte] = gen8_pte_encode(page_addr, cache_level,
+						    true);
+		if (++act_pte == GEN8_PTES_PER_PAGE) {
+			kunmap_atomic(pt_vaddr);
+			act_pt++;
+			pt_vaddr = kmap_atomic(&ppgtt->gen8_pt_pages[act_pt]);
+			act_pte = 0;
+
+		}
+	}
+	kunmap_atomic(pt_vaddr);
+}
+
+static void gen8_ppgtt_cleanup(struct i915_address_space *vm)
+{
+	struct i915_hw_ppgtt *ppgtt =
+		container_of(vm, struct i915_hw_ppgtt, base);
+	int i, j;
+
+	for (i = 0; i < ppgtt->num_pd_pages ; i++) {
+		if (ppgtt->pd_dma_addr[i]) {
+			pci_unmap_page(ppgtt->base.dev->pdev,
+				       ppgtt->pd_dma_addr[i],
+				       PAGE_SIZE, PCI_DMA_BIDIRECTIONAL);
+
+			for (j = 0; j < GEN8_PDES_PER_PAGE; j++) {
+				dma_addr_t addr = ppgtt->gen8_pt_dma_addr[i][j];
+				if (addr)
+					pci_unmap_page(ppgtt->base.dev->pdev,
+						       addr,
+						       PAGE_SIZE,
+						       PCI_DMA_BIDIRECTIONAL);
+
+			}
+		}
+		kfree(ppgtt->gen8_pt_dma_addr[i]);
+	}
+
+	__free_pages(ppgtt->gen8_pt_pages, ppgtt->num_pt_pages << PAGE_SHIFT);
+	__free_pages(ppgtt->pd_pages, ppgtt->num_pd_pages << PAGE_SHIFT);
+}
+
+/**
+ * GEN8 legacy ppgtt programming is accomplished through 4 PDP registers with a
+ * net effect resembling a 2-level page table in normal x86 terms. Each PDP
+ * represents 1GB of memory
+ * 4 * 512 * 512 * 4096 = 4GB legacy 32b address space.
+ *
+ * TODO: Do something with the size parameter
+ **/
+static int gen8_ppgtt_init(struct i915_hw_ppgtt *ppgtt, uint64_t size)
+{
+	struct page *pt_pages;
+	int i, j, ret = -ENOMEM;
+	const int max_pdp = DIV_ROUND_UP(size, 1 << 30);
+	const int num_pt_pages = GEN8_PDES_PER_PAGE * max_pdp;
+
+	if (size % (1<<30))
+		DRM_INFO("Pages will be wasted unless GTT size (%llu) is divisible by 1GB\n", size);
+
+	/* FIXME: split allocation into smaller pieces. For now we only ever do
+	 * this once, but with full PPGTT, the multiple contiguous allocations
+	 * will be bad.
+	 */
+	ppgtt->pd_pages = alloc_pages(GFP_KERNEL, get_order(max_pdp << PAGE_SHIFT));
+	if (!ppgtt->pd_pages)
+		return -ENOMEM;
+
+	pt_pages = alloc_pages(GFP_KERNEL, get_order(num_pt_pages << PAGE_SHIFT));
+	if (!pt_pages) {
+		__free_pages(ppgtt->pd_pages, get_order(max_pdp << PAGE_SHIFT));
+		return -ENOMEM;
+	}
+
+	ppgtt->gen8_pt_pages = pt_pages;
+	ppgtt->num_pd_pages = 1 << get_order(max_pdp << PAGE_SHIFT);
+	ppgtt->num_pt_pages = 1 << get_order(num_pt_pages << PAGE_SHIFT);
+	ppgtt->num_pd_entries = max_pdp * GEN8_PDES_PER_PAGE;
+	ppgtt->enable = gen8_ppgtt_enable;
+	ppgtt->base.clear_range = gen8_ppgtt_clear_range;
+	ppgtt->base.insert_entries = gen8_ppgtt_insert_entries;
+	ppgtt->base.cleanup = gen8_ppgtt_cleanup;
+
+	BUG_ON(ppgtt->num_pd_pages > GEN8_LEGACY_PDPS);
+
+	/*
+	 * - Create a mapping for the page directories.
+	 * - For each page directory:
+	 *      allocate space for page table mappings.
+	 *      map each page table
+	 */
+	for (i = 0; i < max_pdp; i++) {
+		dma_addr_t temp;
+		temp = pci_map_page(ppgtt->base.dev->pdev,
+				    &ppgtt->pd_pages[i], 0,
+				    PAGE_SIZE, PCI_DMA_BIDIRECTIONAL);
+		if (pci_dma_mapping_error(ppgtt->base.dev->pdev, temp))
+			goto err_out;
+
+		ppgtt->pd_dma_addr[i] = temp;
+
+		ppgtt->gen8_pt_dma_addr[i] = kmalloc(sizeof(dma_addr_t) * GEN8_PDES_PER_PAGE, GFP_KERNEL);
+		if (!ppgtt->gen8_pt_dma_addr[i])
+			goto err_out;
+
+		for (j = 0; j < GEN8_PDES_PER_PAGE; j++) {
+			struct page *p = &pt_pages[i * GEN8_PDES_PER_PAGE + j];
+			temp = pci_map_page(ppgtt->base.dev->pdev,
+					    p, 0, PAGE_SIZE,
+					    PCI_DMA_BIDIRECTIONAL);
+
+			if (pci_dma_mapping_error(ppgtt->base.dev->pdev, temp))
+				goto err_out;
+
+			ppgtt->gen8_pt_dma_addr[i][j] = temp;
+		}
+	}
+
+	/* For now, the PPGTT helper functions all require that the PDEs are
+	 * plugged in correctly. So we do that now/here. For aliasing PPGTT, we
+	 * will never need to touch the PDEs again */
+	for (i = 0; i < max_pdp; i++) {
+		gen8_ppgtt_pde_t *pd_vaddr;
+		pd_vaddr = kmap_atomic(&ppgtt->pd_pages[i]);
+		for (j = 0; j < GEN8_PDES_PER_PAGE; j++) {
+			dma_addr_t addr = ppgtt->gen8_pt_dma_addr[i][j];
+			pd_vaddr[j] = gen8_pde_encode(ppgtt->base.dev, addr,
+						      I915_CACHE_LLC);
+		}
+		kunmap_atomic(pd_vaddr);
+	}
+
+	ppgtt->base.clear_range(&ppgtt->base, 0,
+				ppgtt->num_pd_entries * GEN8_PTES_PER_PAGE,
+				true);
+
+	DRM_DEBUG_DRIVER("Allocated %d pages for page directories (%d wasted)\n",
+			 ppgtt->num_pd_pages, ppgtt->num_pd_pages - max_pdp);
+	DRM_DEBUG_DRIVER("Allocated %d pages for page tables (%lld wasted)\n",
+			 ppgtt->num_pt_pages,
+			 (ppgtt->num_pt_pages - num_pt_pages) +
+			 size % (1<<30));
+	return 0;
+
+err_out:
+	ppgtt->base.cleanup(&ppgtt->base);
+	return ret;
+}
+
 static void gen6_write_pdes(struct i915_hw_ppgtt *ppgtt)
 {
 	struct drm_i915_private *dev_priv = ppgtt->base.dev->dev_private;
@@ -342,7 +630,7 @@ static int gen6_ppgtt_init(struct i915_hw_ppgtt *ppgtt)
 	ppgtt->base.insert_entries = gen6_ppgtt_insert_entries;
 	ppgtt->base.cleanup = gen6_ppgtt_cleanup;
 	ppgtt->base.scratch = dev_priv->gtt.base.scratch;
-	ppgtt->pt_pages = kzalloc(sizeof(struct page *)*ppgtt->num_pd_entries,
+	ppgtt->pt_pages = kcalloc(ppgtt->num_pd_entries, sizeof(struct page *),
 				  GFP_KERNEL);
 	if (!ppgtt->pt_pages)
 		return -ENOMEM;
@@ -353,7 +641,7 @@ static int gen6_ppgtt_init(struct i915_hw_ppgtt *ppgtt)
 			goto err_pt_alloc;
 	}
 
-	ppgtt->pt_dma_addr = kzalloc(sizeof(dma_addr_t) *ppgtt->num_pd_entries,
+	ppgtt->pt_dma_addr = kcalloc(ppgtt->num_pd_entries, sizeof(dma_addr_t),
 				     GFP_KERNEL);
 	if (!ppgtt->pt_dma_addr)
 		goto err_pt_alloc;
@@ -410,6 +698,8 @@ static int i915_gem_init_aliasing_ppgtt(struct drm_device *dev)
 
 	if (INTEL_INFO(dev)->gen < 8)
 		ret = gen6_ppgtt_init(ppgtt);
+	else if (IS_GEN8(dev))
+		ret = gen8_ppgtt_init(ppgtt, dev_priv->gtt.base.total);
 	else
 		BUG();
 
@@ -573,6 +863,57 @@ int i915_gem_gtt_prepare_object(struct drm_i915_gem_object *obj)
 	return 0;
 }
 
+static inline void gen8_set_pte(void __iomem *addr, gen8_gtt_pte_t pte)
+{
+#ifdef writeq
+	writeq(pte, addr);
+#else
+	iowrite32((u32)pte, addr);
+	iowrite32(pte >> 32, addr + 4);
+#endif
+}
+
+static void gen8_ggtt_insert_entries(struct i915_address_space *vm,
+				     struct sg_table *st,
+				     unsigned int first_entry,
+				     enum i915_cache_level level)
+{
+	struct drm_i915_private *dev_priv = vm->dev->dev_private;
+	gen8_gtt_pte_t __iomem *gtt_entries =
+		(gen8_gtt_pte_t __iomem *)dev_priv->gtt.gsm + first_entry;
+	int i = 0;
+	struct sg_page_iter sg_iter;
+	dma_addr_t addr;
+
+	for_each_sg_page(st->sgl, &sg_iter, st->nents, 0) {
+		addr = sg_dma_address(sg_iter.sg) +
+			(sg_iter.sg_pgoffset << PAGE_SHIFT);
+		gen8_set_pte(&gtt_entries[i],
+			     gen8_pte_encode(addr, level, true));
+		i++;
+	}
+
+	/*
+	 * XXX: This serves as a posting read to make sure that the PTE has
+	 * actually been updated. There is some concern that even though
+	 * registers and PTEs are within the same BAR that they are potentially
+	 * of NUMA access patterns. Therefore, even with the way we assume
+	 * hardware should work, we must keep this posting read for paranoia.
+	 */
+	if (i != 0)
+		WARN_ON(readq(&gtt_entries[i-1])
+			!= gen8_pte_encode(addr, level, true));
+
+#if 0 /* TODO: Still needed on GEN8? */
+	/* This next bit makes the above posting read even more important. We
+	 * want to flush the TLBs only after we're certain all the PTE updates
+	 * have finished.
+	 */
+	I915_WRITE(GFX_FLSH_CNTL_GEN6, GFX_FLSH_CNTL_EN);
+	POSTING_READ(GFX_FLSH_CNTL_GEN6);
+#endif
+}
+
 /*
  * Binds an object into the global gtt with the specified cache level. The object
  * will be accessible to the GPU via commands whose operands reference offsets
@@ -615,6 +956,30 @@ static void gen6_ggtt_insert_entries(struct i915_address_space *vm,
 	POSTING_READ(GFX_FLSH_CNTL_GEN6);
 }
 
+static void gen8_ggtt_clear_range(struct i915_address_space *vm,
+				  unsigned int first_entry,
+				  unsigned int num_entries,
+				  bool use_scratch)
+{
+	struct drm_i915_private *dev_priv = vm->dev->dev_private;
+	gen8_gtt_pte_t scratch_pte, __iomem *gtt_base =
+		(gen8_gtt_pte_t __iomem *) dev_priv->gtt.gsm + first_entry;
+	const int max_entries = gtt_total_entries(dev_priv->gtt) - first_entry;
+	int i;
+
+	if (WARN(num_entries > max_entries,
+		 "First entry = %d; Num entries = %d (max=%d)\n",
+		 first_entry, num_entries, max_entries))
+		num_entries = max_entries;
+
+	scratch_pte = gen8_pte_encode(vm->scratch.addr,
+				      I915_CACHE_LLC,
+				      use_scratch);
+	for (i = 0; i < num_entries; i++)
+		gen8_set_pte(&gtt_base[i], scratch_pte);
+	readl(gtt_base);
+}
+
 static void gen6_ggtt_clear_range(struct i915_address_space *vm,
 				  unsigned int first_entry,
 				  unsigned int num_entries,
@@ -638,7 +1003,6 @@ static void gen6_ggtt_clear_range(struct i915_address_space *vm,
 	readl(gtt_base);
 }
 
-
 static void i915_ggtt_insert_entries(struct i915_address_space *vm,
 				     struct sg_table *st,
 				     unsigned int pg_start,
@@ -720,6 +1084,7 @@ static void i915_gtt_color_adjust(struct drm_mm_node *node,
 			*end -= 4096;
 	}
 }
+
 void i915_gem_setup_global_gtt(struct drm_device *dev,
 			       unsigned long start,
 			       unsigned long mappable_end,
@@ -817,7 +1182,8 @@ void i915_gem_init_global_gtt(struct drm_device *dev)
 
 		DRM_ERROR("Aliased PPGTT setup failed %d\n", ret);
 		drm_mm_takedown(&dev_priv->gtt.base.mm);
-		gtt_size += GEN6_PPGTT_PD_ENTRIES * PAGE_SIZE;
+		if (INTEL_INFO(dev)->gen < 8)
+			gtt_size += GEN6_PPGTT_PD_ENTRIES*PAGE_SIZE;
 	}
 	i915_gem_setup_global_gtt(dev, 0, mappable_size, gtt_size);
 }
@@ -867,6 +1233,15 @@ static inline unsigned int gen6_get_total_gtt_size(u16 snb_gmch_ctl)
 	return snb_gmch_ctl << 20;
 }
 
+static inline unsigned int gen8_get_total_gtt_size(u16 bdw_gmch_ctl)
+{
+	bdw_gmch_ctl >>= BDW_GMCH_GGMS_SHIFT;
+	bdw_gmch_ctl &= BDW_GMCH_GGMS_MASK;
+	if (bdw_gmch_ctl)
+		bdw_gmch_ctl = 1 << bdw_gmch_ctl;
+	return bdw_gmch_ctl << 20;
+}
+
 static inline size_t gen6_get_stolen_size(u16 snb_gmch_ctl)
 {
 	snb_gmch_ctl >>= SNB_GMCH_GMS_SHIFT;
@@ -874,6 +1249,108 @@ static inline size_t gen6_get_stolen_size(u16 snb_gmch_ctl)
 	return snb_gmch_ctl << 25; /* 32 MB units */
 }
 
+static inline size_t gen8_get_stolen_size(u16 bdw_gmch_ctl)
+{
+	bdw_gmch_ctl >>= BDW_GMCH_GMS_SHIFT;
+	bdw_gmch_ctl &= BDW_GMCH_GMS_MASK;
+	return bdw_gmch_ctl << 25; /* 32 MB units */
+}
+
+static int ggtt_probe_common(struct drm_device *dev,
+			     size_t gtt_size)
+{
+	struct drm_i915_private *dev_priv = dev->dev_private;
+	phys_addr_t gtt_bus_addr;
+	int ret;
+
+	/* For Modern GENs the PTEs and register space are split in the BAR */
+	gtt_bus_addr = pci_resource_start(dev->pdev, 0) +
+		(pci_resource_len(dev->pdev, 0) / 2);
+
+	dev_priv->gtt.gsm = ioremap_wc(gtt_bus_addr, gtt_size);
+	if (!dev_priv->gtt.gsm) {
+		DRM_ERROR("Failed to map the gtt page table\n");
+		return -ENOMEM;
+	}
+
+	ret = setup_scratch_page(dev);
+	if (ret) {
+		DRM_ERROR("Scratch setup failed\n");
+		/* iounmap will also get called at remove, but meh */
+		iounmap(dev_priv->gtt.gsm);
+	}
+
+	return ret;
+}
+
+/* The GGTT and PPGTT need a private PPAT setup in order to handle cacheability
+ * bits. When using advanced contexts each context stores its own PAT, but
+ * writing this data shouldn't be harmful even in those cases. */
+static void gen8_setup_private_ppat(struct drm_i915_private *dev_priv)
+{
+#define GEN8_PPAT_UC		(0<<0)
+#define GEN8_PPAT_WC		(1<<0)
+#define GEN8_PPAT_WT		(2<<0)
+#define GEN8_PPAT_WB		(3<<0)
+#define GEN8_PPAT_ELLC_OVERRIDE	(0<<2)
+/* FIXME(BDW): Bspec is completely confused about cache control bits. */
+#define GEN8_PPAT_LLC		(1<<2)
+#define GEN8_PPAT_LLCELLC	(2<<2)
+#define GEN8_PPAT_LLCeLLC	(3<<2)
+#define GEN8_PPAT_AGE(x)	(x<<4)
+#define GEN8_PPAT(i, x) ((uint64_t) (x) << ((i) * 8))
+	uint64_t pat;
+
+	pat = GEN8_PPAT(0, GEN8_PPAT_WB | GEN8_PPAT_LLC)     | /* for normal objects, no eLLC */
+	      GEN8_PPAT(1, GEN8_PPAT_WC | GEN8_PPAT_LLCELLC) | /* for something pointing to ptes? */
+	      GEN8_PPAT(2, GEN8_PPAT_WT | GEN8_PPAT_LLCELLC) | /* for scanout with eLLC */
+	      GEN8_PPAT(3, GEN8_PPAT_UC)                     | /* Uncached objects, mostly for scanout */
+	      GEN8_PPAT(4, GEN8_PPAT_WB | GEN8_PPAT_LLCELLC | GEN8_PPAT_AGE(0)) |
+	      GEN8_PPAT(5, GEN8_PPAT_WB | GEN8_PPAT_LLCELLC | GEN8_PPAT_AGE(1)) |
+	      GEN8_PPAT(6, GEN8_PPAT_WB | GEN8_PPAT_LLCELLC | GEN8_PPAT_AGE(2)) |
+	      GEN8_PPAT(7, GEN8_PPAT_WB | GEN8_PPAT_LLCELLC | GEN8_PPAT_AGE(3));
+
+	/* XXX: spec defines this as 2 distinct registers. It's unclear if a 64b
+	 * write would work. */
+	I915_WRITE(GEN8_PRIVATE_PAT, pat);
+	I915_WRITE(GEN8_PRIVATE_PAT + 4, pat >> 32);
+}
+
+static int gen8_gmch_probe(struct drm_device *dev,
+			   size_t *gtt_total,
+			   size_t *stolen,
+			   phys_addr_t *mappable_base,
+			   unsigned long *mappable_end)
+{
+	struct drm_i915_private *dev_priv = dev->dev_private;
+	unsigned int gtt_size;
+	u16 snb_gmch_ctl;
+	int ret;
+
+	/* TODO: We're not aware of mappable constraints on gen8 yet */
+	*mappable_base = pci_resource_start(dev->pdev, 2);
+	*mappable_end = pci_resource_len(dev->pdev, 2);
+
+	if (!pci_set_dma_mask(dev->pdev, DMA_BIT_MASK(39)))
+		pci_set_consistent_dma_mask(dev->pdev, DMA_BIT_MASK(39));
+
+	pci_read_config_word(dev->pdev, SNB_GMCH_CTRL, &snb_gmch_ctl);
+
+	*stolen = gen8_get_stolen_size(snb_gmch_ctl);
+
+	gtt_size = gen8_get_total_gtt_size(snb_gmch_ctl);
+	*gtt_total = (gtt_size / sizeof(gen8_gtt_pte_t)) << PAGE_SHIFT;
+
+	gen8_setup_private_ppat(dev_priv);
+
+	ret = ggtt_probe_common(dev, gtt_size);
+
+	dev_priv->gtt.base.clear_range = gen8_ggtt_clear_range;
+	dev_priv->gtt.base.insert_entries = gen8_ggtt_insert_entries;
+
+	return ret;
+}
+
 static int gen6_gmch_probe(struct drm_device *dev,
 			   size_t *gtt_total,
 			   size_t *stolen,
@@ -881,7 +1358,6 @@ static int gen6_gmch_probe(struct drm_device *dev,
 			   unsigned long *mappable_end)
 {
 	struct drm_i915_private *dev_priv = dev->dev_private;
-	phys_addr_t gtt_bus_addr;
 	unsigned int gtt_size;
 	u16 snb_gmch_ctl;
 	int ret;
@@ -901,24 +1377,13 @@ static int gen6_gmch_probe(struct drm_device *dev,
 	if (!pci_set_dma_mask(dev->pdev, DMA_BIT_MASK(40)))
 		pci_set_consistent_dma_mask(dev->pdev, DMA_BIT_MASK(40));
 	pci_read_config_word(dev->pdev, SNB_GMCH_CTRL, &snb_gmch_ctl);
-	gtt_size = gen6_get_total_gtt_size(snb_gmch_ctl);
 
 	*stolen = gen6_get_stolen_size(snb_gmch_ctl);
-	*gtt_total = (gtt_size / sizeof(gen6_gtt_pte_t)) << PAGE_SHIFT;
-
-	/* For Modern GENs the PTEs and register space are split in the BAR */
-	gtt_bus_addr = pci_resource_start(dev->pdev, 0) +
-		(pci_resource_len(dev->pdev, 0) / 2);
 
-	dev_priv->gtt.gsm = ioremap_wc(gtt_bus_addr, gtt_size);
-	if (!dev_priv->gtt.gsm) {
-		DRM_ERROR("Failed to map the gtt page table\n");
-		return -ENOMEM;
-	}
+	gtt_size = gen6_get_total_gtt_size(snb_gmch_ctl);
+	*gtt_total = (gtt_size / sizeof(gen6_gtt_pte_t)) << PAGE_SHIFT;
 
-	ret = setup_scratch_page(dev);
-	if (ret)
-		DRM_ERROR("Scratch setup failed\n");
+	ret = ggtt_probe_common(dev, gtt_size);
 
 	dev_priv->gtt.base.clear_range = gen6_ggtt_clear_range;
 	dev_priv->gtt.base.insert_entries = gen6_ggtt_insert_entries;
@@ -972,7 +1437,7 @@ int i915_gem_gtt_init(struct drm_device *dev)
 	if (INTEL_INFO(dev)->gen <= 5) {
 		gtt->gtt_probe = i915_gmch_probe;
 		gtt->base.cleanup = i915_gmch_remove;
-	} else {
+	} else if (INTEL_INFO(dev)->gen < 8) {
 		gtt->gtt_probe = gen6_gmch_probe;
 		gtt->base.cleanup = gen6_gmch_remove;
 		if (IS_HASWELL(dev) && dev_priv->ellc_size)
@@ -985,6 +1450,9 @@ int i915_gem_gtt_init(struct drm_device *dev)
 			gtt->base.pte_encode = ivb_pte_encode;
 		else
 			gtt->base.pte_encode = snb_pte_encode;
+	} else {
+		dev_priv->gtt.gtt_probe = gen8_gmch_probe;
+		dev_priv->gtt.base.cleanup = gen6_gmch_remove;
 	}
 
 	ret = gtt->gtt_probe(dev, &gtt->base.total, &gtt->stolen_size,
diff --git a/drivers/gpu/drm/i915/i915_gem_stolen.c b/drivers/gpu/drm/i915/i915_gem_stolen.c
index e15a1d90037d..d284d892ed94 100644
--- a/drivers/gpu/drm/i915/i915_gem_stolen.c
+++ b/drivers/gpu/drm/i915/i915_gem_stolen.c
@@ -395,7 +395,7 @@ i915_gem_object_create_stolen_for_preallocated(struct drm_device *dev,
 	if (gtt_offset == I915_GTT_OFFSET_NONE)
 		return obj;
 
-	vma = i915_gem_vma_create(obj, ggtt);
+	vma = i915_gem_obj_lookup_or_create_vma(obj, ggtt);
 	if (IS_ERR(vma)) {
 		ret = PTR_ERR(vma);
 		goto err_out;
diff --git a/drivers/gpu/drm/i915/i915_gem_tiling.c b/drivers/gpu/drm/i915/i915_gem_tiling.c
index 032e9ef9c896..b13905348048 100644
--- a/drivers/gpu/drm/i915/i915_gem_tiling.c
+++ b/drivers/gpu/drm/i915/i915_gem_tiling.c
@@ -308,7 +308,7 @@ i915_gem_set_tiling(struct drm_device *dev, void *data,
 		return -EINVAL;
 	}
 
-	if (obj->pin_count) {
+	if (obj->pin_count || obj->framebuffer_references) {
 		drm_gem_object_unreference_unlocked(&obj->base);
 		return -EBUSY;
 	}
@@ -393,7 +393,7 @@ i915_gem_set_tiling(struct drm_device *dev, void *data,
 	/* Try to preallocate memory required to save swizzling on put-pages */
 	if (i915_gem_object_needs_bit17_swizzle(obj)) {
 		if (obj->bit_17 == NULL) {
-			obj->bit_17 = kmalloc(BITS_TO_LONGS(obj->base.size >> PAGE_SHIFT) *
+			obj->bit_17 = kcalloc(BITS_TO_LONGS(obj->base.size >> PAGE_SHIFT),
 					      sizeof(long), GFP_KERNEL);
 		}
 	} else {
@@ -504,8 +504,8 @@ i915_gem_object_save_bit_17_swizzle(struct drm_i915_gem_object *obj)
 	int i;
 
 	if (obj->bit_17 == NULL) {
-		obj->bit_17 = kmalloc(BITS_TO_LONGS(page_count) *
-					   sizeof(long), GFP_KERNEL);
+		obj->bit_17 = kcalloc(BITS_TO_LONGS(page_count),
+				      sizeof(long), GFP_KERNEL);
 		if (obj->bit_17 == NULL) {
 			DRM_ERROR("Failed to allocate memory for bit 17 "
 				  "record\n");
diff --git a/drivers/gpu/drm/i915/i915_gpu_error.c b/drivers/gpu/drm/i915/i915_gpu_error.c
index dae364f0028c..79dcb8f896c6 100644
--- a/drivers/gpu/drm/i915/i915_gpu_error.c
+++ b/drivers/gpu/drm/i915/i915_gpu_error.c
@@ -215,6 +215,24 @@ static void print_error_buffers(struct drm_i915_error_state_buf *m,
 	}
 }
 
+static const char *hangcheck_action_to_str(enum intel_ring_hangcheck_action a)
+{
+	switch (a) {
+	case HANGCHECK_IDLE:
+		return "idle";
+	case HANGCHECK_WAIT:
+		return "wait";
+	case HANGCHECK_ACTIVE:
+		return "active";
+	case HANGCHECK_KICK:
+		return "kick";
+	case HANGCHECK_HUNG:
+		return "hung";
+	}
+
+	return "unknown";
+}
+
 static void i915_ring_error_state(struct drm_i915_error_state_buf *m,
 				  struct drm_device *dev,
 				  struct drm_i915_error_state *error,
@@ -231,7 +249,8 @@ static void i915_ring_error_state(struct drm_i915_error_state_buf *m,
 	err_printf(m, "  INSTDONE: 0x%08x\n", error->instdone[ring]);
 	if (ring == RCS && INTEL_INFO(dev)->gen >= 4)
 		err_printf(m, "  BBADDR: 0x%08llx\n", error->bbaddr);
-
+	if (INTEL_INFO(dev)->gen >= 4)
+		err_printf(m, "  BB_STATE: 0x%08x\n", error->bbstate[ring]);
 	if (INTEL_INFO(dev)->gen >= 4)
 		err_printf(m, "  INSTPS: 0x%08x\n", error->instps[ring]);
 	err_printf(m, "  INSTPM: 0x%08x\n", error->instpm[ring]);
@@ -255,6 +274,9 @@ static void i915_ring_error_state(struct drm_i915_error_state_buf *m,
 	err_printf(m, "  waiting: %s\n", yesno(error->waiting[ring]));
 	err_printf(m, "  ring->head: 0x%08x\n", error->cpu_ring_head[ring]);
 	err_printf(m, "  ring->tail: 0x%08x\n", error->cpu_ring_tail[ring]);
+	err_printf(m, "  hangcheck: %s [%d]\n",
+		   hangcheck_action_to_str(error->hangcheck_action[ring]),
+		   error->hangcheck_score[ring]);
 }
 
 void i915_error_printf(struct drm_i915_error_state_buf *e, const char *f, ...)
@@ -283,13 +305,14 @@ int i915_error_state_to_str(struct drm_i915_error_state_buf *m,
 	err_printf(m, "Time: %ld s %ld us\n", error->time.tv_sec,
 		   error->time.tv_usec);
 	err_printf(m, "Kernel: " UTS_RELEASE "\n");
-	err_printf(m, "PCI ID: 0x%04x\n", dev->pci_device);
+	err_printf(m, "PCI ID: 0x%04x\n", dev->pdev->device);
 	err_printf(m, "EIR: 0x%08x\n", error->eir);
 	err_printf(m, "IER: 0x%08x\n", error->ier);
 	err_printf(m, "PGTBL_ER: 0x%08x\n", error->pgtbl_er);
 	err_printf(m, "FORCEWAKE: 0x%08x\n", error->forcewake);
 	err_printf(m, "DERRMR: 0x%08x\n", error->derrmr);
 	err_printf(m, "CCID: 0x%08x\n", error->ccid);
+	err_printf(m, "Missed interrupts: 0x%08lx\n", dev_priv->gpu_error.missed_irq_rings);
 
 	for (i = 0; i < dev_priv->num_fence_regs; i++)
 		err_printf(m, "  fence[%d] = %08llx\n", i, error->fence[i]);
@@ -601,6 +624,7 @@ static void i915_gem_record_fences(struct drm_device *dev,
 
 	/* Fences */
 	switch (INTEL_INFO(dev)->gen) {
+	case 8:
 	case 7:
 	case 6:
 		for (i = 0; i < dev_priv->num_fence_regs; i++)
@@ -703,6 +727,7 @@ static void i915_record_ring_state(struct drm_device *dev,
 		error->instps[ring->id] = I915_READ(RING_INSTPS(ring->mmio_base));
 		if (ring->id == RCS)
 			error->bbaddr = I915_READ64(BB_ADDR);
+		error->bbstate[ring->id] = I915_READ(RING_BBSTATE(ring->mmio_base));
 	} else {
 		error->faddr[ring->id] = I915_READ(DMA_FADD_I8XX);
 		error->ipeir[ring->id] = I915_READ(IPEIR);
@@ -720,6 +745,9 @@ static void i915_record_ring_state(struct drm_device *dev,
 
 	error->cpu_ring_head[ring->id] = ring->head;
 	error->cpu_ring_tail[ring->id] = ring->tail;
+
+	error->hangcheck_score[ring->id] = ring->hangcheck.score;
+	error->hangcheck_action[ring->id] = ring->hangcheck.action;
 }
 
 
@@ -769,7 +797,7 @@ static void i915_gem_record_rings(struct drm_device *dev,
 
 		error->ring[i].num_requests = count;
 		error->ring[i].requests =
-			kmalloc(count*sizeof(struct drm_i915_error_request),
+			kcalloc(count, sizeof(*error->ring[i].requests),
 				GFP_ATOMIC);
 		if (error->ring[i].requests == NULL) {
 			error->ring[i].num_requests = 0;
@@ -811,7 +839,7 @@ static void i915_gem_capture_vm(struct drm_i915_private *dev_priv,
 	error->pinned_bo_count[ndx] = i - error->active_bo_count[ndx];
 
 	if (i) {
-		active_bo = kmalloc(sizeof(*active_bo)*i, GFP_ATOMIC);
+		active_bo = kcalloc(i, sizeof(*active_bo), GFP_ATOMIC);
 		if (active_bo)
 			pinned_bo = active_bo + error->active_bo_count[ndx];
 	}
@@ -885,8 +913,12 @@ void i915_capture_error_state(struct drm_device *dev)
 		return;
 	}
 
-	DRM_INFO("capturing error event; look for more information in "
-		 "/sys/class/drm/card%d/error\n", dev->primary->index);
+	DRM_INFO("GPU crash dump saved to /sys/class/drm/card%d/error\n",
+		 dev->primary->index);
+	DRM_INFO("GPU hangs can indicate a bug anywhere in the entire gfx stack, including userspace.\n");
+	DRM_INFO("Please file a _new_ bug report on bugs.freedesktop.org against DRI -> DRM/Intel\n");
+	DRM_INFO("drm/i915 developers can then reassign to the right component if it's not a kernel issue.\n");
+	DRM_INFO("The gpu crash dump is required to analyze gpu hangs, so please always attach it.\n");
 
 	kref_init(&error->ref);
 	error->eir = I915_READ(EIR);
@@ -988,6 +1020,7 @@ const char *i915_cache_level_str(int type)
 	case I915_CACHE_NONE: return " uncached";
 	case I915_CACHE_LLC: return " snooped or LLC";
 	case I915_CACHE_L3_LLC: return " L3+LLC";
+	case I915_CACHE_WT: return " WT";
 	default: return "";
 	}
 }
@@ -1012,6 +1045,7 @@ void i915_get_extra_instdone(struct drm_device *dev, uint32_t *instdone)
 	default:
 		WARN_ONCE(1, "Unsupported platform\n");
 	case 7:
+	case 8:
 		instdone[0] = I915_READ(GEN7_INSTDONE_1);
 		instdone[1] = I915_READ(GEN7_SC_INSTDONE);
 		instdone[2] = I915_READ(GEN7_SAMPLER_INSTDONE);
diff --git a/drivers/gpu/drm/i915/i915_irq.c b/drivers/gpu/drm/i915/i915_irq.c
index 4b91228fd9bd..5d1dedc02f15 100644
--- a/drivers/gpu/drm/i915/i915_irq.c
+++ b/drivers/gpu/drm/i915/i915_irq.c
@@ -30,6 +30,7 @@
 
 #include <linux/sysrq.h>
 #include <linux/slab.h>
+#include <linux/circ_buf.h>
 #include <drm/drmP.h>
 #include <drm/i915_drm.h>
 #include "i915_drv.h"
@@ -269,6 +270,21 @@ static void ivybridge_set_fifo_underrun_reporting(struct drm_device *dev,
 	}
 }
 
+static void broadwell_set_fifo_underrun_reporting(struct drm_device *dev,
+						  enum pipe pipe, bool enable)
+{
+	struct drm_i915_private *dev_priv = dev->dev_private;
+
+	assert_spin_locked(&dev_priv->irq_lock);
+
+	if (enable)
+		dev_priv->de_irq_mask[pipe] &= ~GEN8_PIPE_FIFO_UNDERRUN;
+	else
+		dev_priv->de_irq_mask[pipe] |= GEN8_PIPE_FIFO_UNDERRUN;
+	I915_WRITE(GEN8_DE_PIPE_IMR(pipe), dev_priv->de_irq_mask[pipe]);
+	POSTING_READ(GEN8_DE_PIPE_IMR(pipe));
+}
+
 /**
  * ibx_display_interrupt_update - update SDEIMR
  * @dev_priv: driver private
@@ -381,6 +397,8 @@ bool intel_set_cpu_fifo_underrun_reporting(struct drm_device *dev,
 		ironlake_set_fifo_underrun_reporting(dev, pipe, enable);
 	else if (IS_GEN7(dev))
 		ivybridge_set_fifo_underrun_reporting(dev, pipe, enable);
+	else if (IS_GEN8(dev))
+		broadwell_set_fifo_underrun_reporting(dev, pipe, enable);
 
 done:
 	spin_unlock_irqrestore(&dev_priv->irq_lock, flags);
@@ -441,7 +459,7 @@ done:
 
 
 void
-i915_enable_pipestat(drm_i915_private_t *dev_priv, int pipe, u32 mask)
+i915_enable_pipestat(drm_i915_private_t *dev_priv, enum pipe pipe, u32 mask)
 {
 	u32 reg = PIPESTAT(pipe);
 	u32 pipestat = I915_READ(reg) & 0x7fff0000;
@@ -458,7 +476,7 @@ i915_enable_pipestat(drm_i915_private_t *dev_priv, int pipe, u32 mask)
 }
 
 void
-i915_disable_pipestat(drm_i915_private_t *dev_priv, int pipe, u32 mask)
+i915_disable_pipestat(drm_i915_private_t *dev_priv, enum pipe pipe, u32 mask)
 {
 	u32 reg = PIPESTAT(pipe);
 	u32 pipestat = I915_READ(reg) & 0x7fff0000;
@@ -486,9 +504,10 @@ static void i915_enable_asle_pipestat(struct drm_device *dev)
 
 	spin_lock_irqsave(&dev_priv->irq_lock, irqflags);
 
-	i915_enable_pipestat(dev_priv, 1, PIPE_LEGACY_BLC_EVENT_ENABLE);
+	i915_enable_pipestat(dev_priv, PIPE_B, PIPE_LEGACY_BLC_EVENT_ENABLE);
 	if (INTEL_INFO(dev)->gen >= 4)
-		i915_enable_pipestat(dev_priv, 0, PIPE_LEGACY_BLC_EVENT_ENABLE);
+		i915_enable_pipestat(dev_priv, PIPE_A,
+				     PIPE_LEGACY_BLC_EVENT_ENABLE);
 
 	spin_unlock_irqrestore(&dev_priv->irq_lock, irqflags);
 }
@@ -518,6 +537,12 @@ i915_pipe_enabled(struct drm_device *dev, int pipe)
 	}
 }
 
+static u32 i8xx_get_vblank_counter(struct drm_device *dev, int pipe)
+{
+	/* Gen2 doesn't have a hardware frame counter */
+	return 0;
+}
+
 /* Called from drm generic code, passed a 'crtc', which
  * we use as a pipe index
  */
@@ -526,7 +551,7 @@ static u32 i915_get_vblank_counter(struct drm_device *dev, int pipe)
 	drm_i915_private_t *dev_priv = (drm_i915_private_t *) dev->dev_private;
 	unsigned long high_frame;
 	unsigned long low_frame;
-	u32 high1, high2, low;
+	u32 high1, high2, low, pixel, vbl_start;
 
 	if (!i915_pipe_enabled(dev, pipe)) {
 		DRM_DEBUG_DRIVER("trying to get vblank count for disabled "
@@ -534,6 +559,24 @@ static u32 i915_get_vblank_counter(struct drm_device *dev, int pipe)
 		return 0;
 	}
 
+	if (drm_core_check_feature(dev, DRIVER_MODESET)) {
+		struct intel_crtc *intel_crtc =
+			to_intel_crtc(dev_priv->pipe_to_crtc_mapping[pipe]);
+		const struct drm_display_mode *mode =
+			&intel_crtc->config.adjusted_mode;
+
+		vbl_start = mode->crtc_vblank_start * mode->crtc_htotal;
+	} else {
+		enum transcoder cpu_transcoder =
+			intel_pipe_to_cpu_transcoder(dev_priv, pipe);
+		u32 htotal;
+
+		htotal = ((I915_READ(HTOTAL(cpu_transcoder)) >> 16) & 0x1fff) + 1;
+		vbl_start = (I915_READ(VBLANK(cpu_transcoder)) & 0x1fff) + 1;
+
+		vbl_start *= htotal;
+	}
+
 	high_frame = PIPEFRAME(pipe);
 	low_frame = PIPEFRAMEPIXEL(pipe);
 
@@ -544,13 +587,20 @@ static u32 i915_get_vblank_counter(struct drm_device *dev, int pipe)
 	 */
 	do {
 		high1 = I915_READ(high_frame) & PIPE_FRAME_HIGH_MASK;
-		low   = I915_READ(low_frame)  & PIPE_FRAME_LOW_MASK;
+		low   = I915_READ(low_frame);
 		high2 = I915_READ(high_frame) & PIPE_FRAME_HIGH_MASK;
 	} while (high1 != high2);
 
 	high1 >>= PIPE_FRAME_HIGH_SHIFT;
+	pixel = low & PIPE_PIXEL_MASK;
 	low >>= PIPE_FRAME_LOW_SHIFT;
-	return (high1 << 8) | low;
+
+	/*
+	 * The frame counter increments at beginning of active.
+	 * Cook up a vblank counter by also checking the pixel
+	 * counter against vblank start.
+	 */
+	return ((high1 << 8) | low) + (pixel >= vbl_start);
 }
 
 static u32 gm45_get_vblank_counter(struct drm_device *dev, int pipe)
@@ -567,66 +617,163 @@ static u32 gm45_get_vblank_counter(struct drm_device *dev, int pipe)
 	return I915_READ(reg);
 }
 
+/* raw reads, only for fast reads of display block, no need for forcewake etc. */
+#define __raw_i915_read32(dev_priv__, reg__) readl((dev_priv__)->regs + (reg__))
+#define __raw_i915_read16(dev_priv__, reg__) readw((dev_priv__)->regs + (reg__))
+
+static bool intel_pipe_in_vblank_locked(struct drm_device *dev, enum pipe pipe)
+{
+	struct drm_i915_private *dev_priv = dev->dev_private;
+	uint32_t status;
+	int reg;
+
+	if (IS_VALLEYVIEW(dev)) {
+		status = pipe == PIPE_A ?
+			I915_DISPLAY_PIPE_A_VBLANK_INTERRUPT :
+			I915_DISPLAY_PIPE_B_VBLANK_INTERRUPT;
+
+		reg = VLV_ISR;
+	} else if (IS_GEN2(dev)) {
+		status = pipe == PIPE_A ?
+			I915_DISPLAY_PIPE_A_VBLANK_INTERRUPT :
+			I915_DISPLAY_PIPE_B_VBLANK_INTERRUPT;
+
+		reg = ISR;
+	} else if (INTEL_INFO(dev)->gen < 5) {
+		status = pipe == PIPE_A ?
+			I915_DISPLAY_PIPE_A_VBLANK_INTERRUPT :
+			I915_DISPLAY_PIPE_B_VBLANK_INTERRUPT;
+
+		reg = ISR;
+	} else if (INTEL_INFO(dev)->gen < 7) {
+		status = pipe == PIPE_A ?
+			DE_PIPEA_VBLANK :
+			DE_PIPEB_VBLANK;
+
+		reg = DEISR;
+	} else {
+		switch (pipe) {
+		default:
+		case PIPE_A:
+			status = DE_PIPEA_VBLANK_IVB;
+			break;
+		case PIPE_B:
+			status = DE_PIPEB_VBLANK_IVB;
+			break;
+		case PIPE_C:
+			status = DE_PIPEC_VBLANK_IVB;
+			break;
+		}
+
+		reg = DEISR;
+	}
+
+	if (IS_GEN2(dev))
+		return __raw_i915_read16(dev_priv, reg) & status;
+	else
+		return __raw_i915_read32(dev_priv, reg) & status;
+}
+
 static int i915_get_crtc_scanoutpos(struct drm_device *dev, int pipe,
-			     int *vpos, int *hpos)
+			     int *vpos, int *hpos, ktime_t *stime, ktime_t *etime)
 {
-	drm_i915_private_t *dev_priv = (drm_i915_private_t *) dev->dev_private;
-	u32 vbl = 0, position = 0;
+	struct drm_i915_private *dev_priv = dev->dev_private;
+	struct drm_crtc *crtc = dev_priv->pipe_to_crtc_mapping[pipe];
+	struct intel_crtc *intel_crtc = to_intel_crtc(crtc);
+	const struct drm_display_mode *mode = &intel_crtc->config.adjusted_mode;
+	int position;
 	int vbl_start, vbl_end, htotal, vtotal;
 	bool in_vbl = true;
 	int ret = 0;
-	enum transcoder cpu_transcoder = intel_pipe_to_cpu_transcoder(dev_priv,
-								      pipe);
+	unsigned long irqflags;
 
-	if (!i915_pipe_enabled(dev, pipe)) {
+	if (!intel_crtc->active) {
 		DRM_DEBUG_DRIVER("trying to get scanoutpos for disabled "
 				 "pipe %c\n", pipe_name(pipe));
 		return 0;
 	}
 
-	/* Get vtotal. */
-	vtotal = 1 + ((I915_READ(VTOTAL(cpu_transcoder)) >> 16) & 0x1fff);
+	htotal = mode->crtc_htotal;
+	vtotal = mode->crtc_vtotal;
+	vbl_start = mode->crtc_vblank_start;
+	vbl_end = mode->crtc_vblank_end;
 
-	if (INTEL_INFO(dev)->gen >= 4) {
+	ret |= DRM_SCANOUTPOS_VALID | DRM_SCANOUTPOS_ACCURATE;
+
+	/*
+	 * Lock uncore.lock, as we will do multiple timing critical raw
+	 * register reads, potentially with preemption disabled, so the
+	 * following code must not block on uncore.lock.
+	 */
+	spin_lock_irqsave(&dev_priv->uncore.lock, irqflags);
+	
+	/* preempt_disable_rt() should go right here in PREEMPT_RT patchset. */
+
+	/* Get optional system timestamp before query. */
+	if (stime)
+		*stime = ktime_get();
+
+	if (IS_GEN2(dev) || IS_G4X(dev) || INTEL_INFO(dev)->gen >= 5) {
 		/* No obvious pixelcount register. Only query vertical
 		 * scanout position from Display scan line register.
 		 */
-		position = I915_READ(PIPEDSL(pipe));
+		if (IS_GEN2(dev))
+			position = __raw_i915_read32(dev_priv, PIPEDSL(pipe)) & DSL_LINEMASK_GEN2;
+		else
+			position = __raw_i915_read32(dev_priv, PIPEDSL(pipe)) & DSL_LINEMASK_GEN3;
 
-		/* Decode into vertical scanout position. Don't have
-		 * horizontal scanout position.
+		/*
+		 * The scanline counter increments at the leading edge
+		 * of hsync, ie. it completely misses the active portion
+		 * of the line. Fix up the counter at both edges of vblank
+		 * to get a more accurate picture whether we're in vblank
+		 * or not.
 		 */
-		*vpos = position & 0x1fff;
-		*hpos = 0;
+		in_vbl = intel_pipe_in_vblank_locked(dev, pipe);
+		if ((in_vbl && position == vbl_start - 1) ||
+		    (!in_vbl && position == vbl_end - 1))
+			position = (position + 1) % vtotal;
 	} else {
 		/* Have access to pixelcount since start of frame.
 		 * We can split this into vertical and horizontal
 		 * scanout position.
 		 */
-		position = (I915_READ(PIPEFRAMEPIXEL(pipe)) & PIPE_PIXEL_MASK) >> PIPE_PIXEL_SHIFT;
+		position = (__raw_i915_read32(dev_priv, PIPEFRAMEPIXEL(pipe)) & PIPE_PIXEL_MASK) >> PIPE_PIXEL_SHIFT;
 
-		htotal = 1 + ((I915_READ(HTOTAL(cpu_transcoder)) >> 16) & 0x1fff);
-		*vpos = position / htotal;
-		*hpos = position - (*vpos * htotal);
+		/* convert to pixel counts */
+		vbl_start *= htotal;
+		vbl_end *= htotal;
+		vtotal *= htotal;
 	}
 
-	/* Query vblank area. */
-	vbl = I915_READ(VBLANK(cpu_transcoder));
+	/* Get optional system timestamp after query. */
+	if (etime)
+		*etime = ktime_get();
 
-	/* Test position against vblank region. */
-	vbl_start = vbl & 0x1fff;
-	vbl_end = (vbl >> 16) & 0x1fff;
+	/* preempt_enable_rt() should go right here in PREEMPT_RT patchset. */
 
-	if ((*vpos < vbl_start) || (*vpos > vbl_end))
-		in_vbl = false;
+	spin_unlock_irqrestore(&dev_priv->uncore.lock, irqflags);
 
-	/* Inside "upper part" of vblank area? Apply corrective offset: */
-	if (in_vbl && (*vpos >= vbl_start))
-		*vpos = *vpos - vtotal;
+	in_vbl = position >= vbl_start && position < vbl_end;
 
-	/* Readouts valid? */
-	if (vbl > 0)
-		ret |= DRM_SCANOUTPOS_VALID | DRM_SCANOUTPOS_ACCURATE;
+	/*
+	 * While in vblank, position will be negative
+	 * counting up towards 0 at vbl_end. And outside
+	 * vblank, position will be positive counting
+	 * up since vbl_end.
+	 */
+	if (position >= vbl_start)
+		position -= vbl_end;
+	else
+		position += vtotal - vbl_end;
+
+	if (IS_GEN2(dev) || IS_G4X(dev) || INTEL_INFO(dev)->gen >= 5) {
+		*vpos = position;
+		*hpos = 0;
+	} else {
+		*vpos = position / htotal;
+		*hpos = position - (*vpos * htotal);
+	}
 
 	/* In vblank? */
 	if (in_vbl)
@@ -665,7 +812,8 @@ static int i915_get_vblank_timestamp(struct drm_device *dev, int pipe,
 						     crtc);
 }
 
-static int intel_hpd_irq_event(struct drm_device *dev, struct drm_connector *connector)
+static bool intel_hpd_irq_event(struct drm_device *dev,
+				struct drm_connector *connector)
 {
 	enum drm_connector_status old_status;
 
@@ -673,11 +821,16 @@ static int intel_hpd_irq_event(struct drm_device *dev, struct drm_connector *con
 	old_status = connector->status;
 
 	connector->status = connector->funcs->detect(connector, false);
-	DRM_DEBUG_KMS("[CONNECTOR:%d:%s] status updated from %d to %d\n",
+	if (old_status == connector->status)
+		return false;
+
+	DRM_DEBUG_KMS("[CONNECTOR:%d:%s] status updated from %s to %s\n",
 		      connector->base.id,
 		      drm_get_connector_name(connector),
-		      old_status, connector->status);
-	return (old_status != connector->status);
+		      drm_get_connector_status_name(old_status),
+		      drm_get_connector_status_name(connector->status));
+
+	return true;
 }
 
 /*
@@ -801,7 +954,7 @@ static void notify_ring(struct drm_device *dev,
 	if (ring->obj == NULL)
 		return;
 
-	trace_i915_gem_request_complete(ring, ring->get_seqno(ring, false));
+	trace_i915_gem_request_complete(ring);
 
 	wake_up_all(&ring->irq_queue);
 	i915_queue_hangcheck(dev);
@@ -812,7 +965,7 @@ static void gen6_pm_rps_work(struct work_struct *work)
 	drm_i915_private_t *dev_priv = container_of(work, drm_i915_private_t,
 						    rps.work);
 	u32 pm_iir;
-	u8 new_delay;
+	int new_delay, adj;
 
 	spin_lock_irq(&dev_priv->irq_lock);
 	pm_iir = dev_priv->rps.pm_iir;
@@ -829,40 +982,49 @@ static void gen6_pm_rps_work(struct work_struct *work)
 
 	mutex_lock(&dev_priv->rps.hw_lock);
 
+	adj = dev_priv->rps.last_adj;
 	if (pm_iir & GEN6_PM_RP_UP_THRESHOLD) {
-		new_delay = dev_priv->rps.cur_delay + 1;
+		if (adj > 0)
+			adj *= 2;
+		else
+			adj = 1;
+		new_delay = dev_priv->rps.cur_delay + adj;
 
 		/*
 		 * For better performance, jump directly
 		 * to RPe if we're below it.
 		 */
-		if (IS_VALLEYVIEW(dev_priv->dev) &&
-		    dev_priv->rps.cur_delay < dev_priv->rps.rpe_delay)
+		if (new_delay < dev_priv->rps.rpe_delay)
+			new_delay = dev_priv->rps.rpe_delay;
+	} else if (pm_iir & GEN6_PM_RP_DOWN_TIMEOUT) {
+		if (dev_priv->rps.cur_delay > dev_priv->rps.rpe_delay)
 			new_delay = dev_priv->rps.rpe_delay;
-	} else
-		new_delay = dev_priv->rps.cur_delay - 1;
+		else
+			new_delay = dev_priv->rps.min_delay;
+		adj = 0;
+	} else if (pm_iir & GEN6_PM_RP_DOWN_THRESHOLD) {
+		if (adj < 0)
+			adj *= 2;
+		else
+			adj = -1;
+		new_delay = dev_priv->rps.cur_delay + adj;
+	} else { /* unknown event */
+		new_delay = dev_priv->rps.cur_delay;
+	}
 
 	/* sysfs frequency interfaces may have snuck in while servicing the
 	 * interrupt
 	 */
-	if (new_delay >= dev_priv->rps.min_delay &&
-	    new_delay <= dev_priv->rps.max_delay) {
-		if (IS_VALLEYVIEW(dev_priv->dev))
-			valleyview_set_rps(dev_priv->dev, new_delay);
-		else
-			gen6_set_rps(dev_priv->dev, new_delay);
-	}
-
-	if (IS_VALLEYVIEW(dev_priv->dev)) {
-		/*
-		 * On VLV, when we enter RC6 we may not be at the minimum
-		 * voltage level, so arm a timer to check.  It should only
-		 * fire when there's activity or once after we've entered
-		 * RC6, and then won't be re-armed until the next RPS interrupt.
-		 */
-		mod_delayed_work(dev_priv->wq, &dev_priv->rps.vlv_work,
-				 msecs_to_jiffies(100));
-	}
+	if (new_delay < (int)dev_priv->rps.min_delay)
+		new_delay = dev_priv->rps.min_delay;
+	if (new_delay > (int)dev_priv->rps.max_delay)
+		new_delay = dev_priv->rps.max_delay;
+	dev_priv->rps.last_adj = new_delay - dev_priv->rps.cur_delay;
+
+	if (IS_VALLEYVIEW(dev_priv->dev))
+		valleyview_set_rps(dev_priv->dev, new_delay);
+	else
+		gen6_set_rps(dev_priv->dev, new_delay);
 
 	mutex_unlock(&dev_priv->rps.hw_lock);
 }
@@ -882,9 +1044,10 @@ static void ivybridge_parity_work(struct work_struct *work)
 	drm_i915_private_t *dev_priv = container_of(work, drm_i915_private_t,
 						    l3_parity.error_work);
 	u32 error_status, row, bank, subbank;
-	char *parity_event[5];
+	char *parity_event[6];
 	uint32_t misccpctl;
 	unsigned long flags;
+	uint8_t slice = 0;
 
 	/* We must turn off DOP level clock gating to access the L3 registers.
 	 * In order to prevent a get/put style interface, acquire struct mutex
@@ -892,55 +1055,81 @@ static void ivybridge_parity_work(struct work_struct *work)
 	 */
 	mutex_lock(&dev_priv->dev->struct_mutex);
 
+	/* If we've screwed up tracking, just let the interrupt fire again */
+	if (WARN_ON(!dev_priv->l3_parity.which_slice))
+		goto out;
+
 	misccpctl = I915_READ(GEN7_MISCCPCTL);
 	I915_WRITE(GEN7_MISCCPCTL, misccpctl & ~GEN7_DOP_CLOCK_GATE_ENABLE);
 	POSTING_READ(GEN7_MISCCPCTL);
 
-	error_status = I915_READ(GEN7_L3CDERRST1);
-	row = GEN7_PARITY_ERROR_ROW(error_status);
-	bank = GEN7_PARITY_ERROR_BANK(error_status);
-	subbank = GEN7_PARITY_ERROR_SUBBANK(error_status);
+	while ((slice = ffs(dev_priv->l3_parity.which_slice)) != 0) {
+		u32 reg;
 
-	I915_WRITE(GEN7_L3CDERRST1, GEN7_PARITY_ERROR_VALID |
-				    GEN7_L3CDERRST1_ENABLE);
-	POSTING_READ(GEN7_L3CDERRST1);
+		slice--;
+		if (WARN_ON_ONCE(slice >= NUM_L3_SLICES(dev_priv->dev)))
+			break;
 
-	I915_WRITE(GEN7_MISCCPCTL, misccpctl);
+		dev_priv->l3_parity.which_slice &= ~(1<<slice);
 
-	spin_lock_irqsave(&dev_priv->irq_lock, flags);
-	ilk_enable_gt_irq(dev_priv, GT_RENDER_L3_PARITY_ERROR_INTERRUPT);
-	spin_unlock_irqrestore(&dev_priv->irq_lock, flags);
+		reg = GEN7_L3CDERRST1 + (slice * 0x200);
 
-	mutex_unlock(&dev_priv->dev->struct_mutex);
+		error_status = I915_READ(reg);
+		row = GEN7_PARITY_ERROR_ROW(error_status);
+		bank = GEN7_PARITY_ERROR_BANK(error_status);
+		subbank = GEN7_PARITY_ERROR_SUBBANK(error_status);
 
-	parity_event[0] = I915_L3_PARITY_UEVENT "=1";
-	parity_event[1] = kasprintf(GFP_KERNEL, "ROW=%d", row);
-	parity_event[2] = kasprintf(GFP_KERNEL, "BANK=%d", bank);
-	parity_event[3] = kasprintf(GFP_KERNEL, "SUBBANK=%d", subbank);
-	parity_event[4] = NULL;
+		I915_WRITE(reg, GEN7_PARITY_ERROR_VALID | GEN7_L3CDERRST1_ENABLE);
+		POSTING_READ(reg);
+
+		parity_event[0] = I915_L3_PARITY_UEVENT "=1";
+		parity_event[1] = kasprintf(GFP_KERNEL, "ROW=%d", row);
+		parity_event[2] = kasprintf(GFP_KERNEL, "BANK=%d", bank);
+		parity_event[3] = kasprintf(GFP_KERNEL, "SUBBANK=%d", subbank);
+		parity_event[4] = kasprintf(GFP_KERNEL, "SLICE=%d", slice);
+		parity_event[5] = NULL;
+
+		kobject_uevent_env(&dev_priv->dev->primary->kdev->kobj,
+				   KOBJ_CHANGE, parity_event);
+
+		DRM_DEBUG("Parity error: Slice = %d, Row = %d, Bank = %d, Sub bank = %d.\n",
+			  slice, row, bank, subbank);
+
+		kfree(parity_event[4]);
+		kfree(parity_event[3]);
+		kfree(parity_event[2]);
+		kfree(parity_event[1]);
+	}
 
-	kobject_uevent_env(&dev_priv->dev->primary->kdev.kobj,
-			   KOBJ_CHANGE, parity_event);
+	I915_WRITE(GEN7_MISCCPCTL, misccpctl);
 
-	DRM_DEBUG("Parity error: Row = %d, Bank = %d, Sub bank = %d.\n",
-		  row, bank, subbank);
+out:
+	WARN_ON(dev_priv->l3_parity.which_slice);
+	spin_lock_irqsave(&dev_priv->irq_lock, flags);
+	ilk_enable_gt_irq(dev_priv, GT_PARITY_ERROR(dev_priv->dev));
+	spin_unlock_irqrestore(&dev_priv->irq_lock, flags);
 
-	kfree(parity_event[3]);
-	kfree(parity_event[2]);
-	kfree(parity_event[1]);
+	mutex_unlock(&dev_priv->dev->struct_mutex);
 }
 
-static void ivybridge_parity_error_irq_handler(struct drm_device *dev)
+static void ivybridge_parity_error_irq_handler(struct drm_device *dev, u32 iir)
 {
 	drm_i915_private_t *dev_priv = (drm_i915_private_t *) dev->dev_private;
 
-	if (!HAS_L3_GPU_CACHE(dev))
+	if (!HAS_L3_DPF(dev))
 		return;
 
 	spin_lock(&dev_priv->irq_lock);
-	ilk_disable_gt_irq(dev_priv, GT_RENDER_L3_PARITY_ERROR_INTERRUPT);
+	ilk_disable_gt_irq(dev_priv, GT_PARITY_ERROR(dev));
 	spin_unlock(&dev_priv->irq_lock);
 
+	iir &= GT_PARITY_ERROR(dev);
+	if (iir & GT_RENDER_L3_PARITY_ERROR_INTERRUPT_S1)
+		dev_priv->l3_parity.which_slice |= 1 << 1;
+
+	if (iir & GT_RENDER_L3_PARITY_ERROR_INTERRUPT)
+		dev_priv->l3_parity.which_slice |= 1 << 0;
+
 	queue_work(dev_priv->wq, &dev_priv->l3_parity.error_work);
 }
 
@@ -975,8 +1164,58 @@ static void snb_gt_irq_handler(struct drm_device *dev,
 		i915_handle_error(dev, false);
 	}
 
-	if (gt_iir & GT_RENDER_L3_PARITY_ERROR_INTERRUPT)
-		ivybridge_parity_error_irq_handler(dev);
+	if (gt_iir & GT_PARITY_ERROR(dev))
+		ivybridge_parity_error_irq_handler(dev, gt_iir);
+}
+
+static irqreturn_t gen8_gt_irq_handler(struct drm_device *dev,
+				       struct drm_i915_private *dev_priv,
+				       u32 master_ctl)
+{
+	u32 rcs, bcs, vcs;
+	uint32_t tmp = 0;
+	irqreturn_t ret = IRQ_NONE;
+
+	if (master_ctl & (GEN8_GT_RCS_IRQ | GEN8_GT_BCS_IRQ)) {
+		tmp = I915_READ(GEN8_GT_IIR(0));
+		if (tmp) {
+			ret = IRQ_HANDLED;
+			rcs = tmp >> GEN8_RCS_IRQ_SHIFT;
+			bcs = tmp >> GEN8_BCS_IRQ_SHIFT;
+			if (rcs & GT_RENDER_USER_INTERRUPT)
+				notify_ring(dev, &dev_priv->ring[RCS]);
+			if (bcs & GT_RENDER_USER_INTERRUPT)
+				notify_ring(dev, &dev_priv->ring[BCS]);
+			I915_WRITE(GEN8_GT_IIR(0), tmp);
+		} else
+			DRM_ERROR("The master control interrupt lied (GT0)!\n");
+	}
+
+	if (master_ctl & GEN8_GT_VCS1_IRQ) {
+		tmp = I915_READ(GEN8_GT_IIR(1));
+		if (tmp) {
+			ret = IRQ_HANDLED;
+			vcs = tmp >> GEN8_VCS1_IRQ_SHIFT;
+			if (vcs & GT_RENDER_USER_INTERRUPT)
+				notify_ring(dev, &dev_priv->ring[VCS]);
+			I915_WRITE(GEN8_GT_IIR(1), tmp);
+		} else
+			DRM_ERROR("The master control interrupt lied (GT1)!\n");
+	}
+
+	if (master_ctl & GEN8_GT_VECS_IRQ) {
+		tmp = I915_READ(GEN8_GT_IIR(3));
+		if (tmp) {
+			ret = IRQ_HANDLED;
+			vcs = tmp >> GEN8_VECS_IRQ_SHIFT;
+			if (vcs & GT_RENDER_USER_INTERRUPT)
+				notify_ring(dev, &dev_priv->ring[VECS]);
+			I915_WRITE(GEN8_GT_IIR(3), tmp);
+		} else
+			DRM_ERROR("The master control interrupt lied (GT3)!\n");
+	}
+
+	return ret;
 }
 
 #define HPD_STORM_DETECT_PERIOD 1000
@@ -1050,6 +1289,102 @@ static void dp_aux_irq_handler(struct drm_device *dev)
 	wake_up_all(&dev_priv->gmbus_wait_queue);
 }
 
+#if defined(CONFIG_DEBUG_FS)
+static void display_pipe_crc_irq_handler(struct drm_device *dev, enum pipe pipe,
+					 uint32_t crc0, uint32_t crc1,
+					 uint32_t crc2, uint32_t crc3,
+					 uint32_t crc4)
+{
+	struct drm_i915_private *dev_priv = dev->dev_private;
+	struct intel_pipe_crc *pipe_crc = &dev_priv->pipe_crc[pipe];
+	struct intel_pipe_crc_entry *entry;
+	int head, tail;
+
+	spin_lock(&pipe_crc->lock);
+
+	if (!pipe_crc->entries) {
+		spin_unlock(&pipe_crc->lock);
+		DRM_ERROR("spurious interrupt\n");
+		return;
+	}
+
+	head = pipe_crc->head;
+	tail = pipe_crc->tail;
+
+	if (CIRC_SPACE(head, tail, INTEL_PIPE_CRC_ENTRIES_NR) < 1) {
+		spin_unlock(&pipe_crc->lock);
+		DRM_ERROR("CRC buffer overflowing\n");
+		return;
+	}
+
+	entry = &pipe_crc->entries[head];
+
+	entry->frame = dev->driver->get_vblank_counter(dev, pipe);
+	entry->crc[0] = crc0;
+	entry->crc[1] = crc1;
+	entry->crc[2] = crc2;
+	entry->crc[3] = crc3;
+	entry->crc[4] = crc4;
+
+	head = (head + 1) & (INTEL_PIPE_CRC_ENTRIES_NR - 1);
+	pipe_crc->head = head;
+
+	spin_unlock(&pipe_crc->lock);
+
+	wake_up_interruptible(&pipe_crc->wq);
+}
+#else
+static inline void
+display_pipe_crc_irq_handler(struct drm_device *dev, enum pipe pipe,
+			     uint32_t crc0, uint32_t crc1,
+			     uint32_t crc2, uint32_t crc3,
+			     uint32_t crc4) {}
+#endif
+
+
+static void hsw_pipe_crc_irq_handler(struct drm_device *dev, enum pipe pipe)
+{
+	struct drm_i915_private *dev_priv = dev->dev_private;
+
+	display_pipe_crc_irq_handler(dev, pipe,
+				     I915_READ(PIPE_CRC_RES_1_IVB(pipe)),
+				     0, 0, 0, 0);
+}
+
+static void ivb_pipe_crc_irq_handler(struct drm_device *dev, enum pipe pipe)
+{
+	struct drm_i915_private *dev_priv = dev->dev_private;
+
+	display_pipe_crc_irq_handler(dev, pipe,
+				     I915_READ(PIPE_CRC_RES_1_IVB(pipe)),
+				     I915_READ(PIPE_CRC_RES_2_IVB(pipe)),
+				     I915_READ(PIPE_CRC_RES_3_IVB(pipe)),
+				     I915_READ(PIPE_CRC_RES_4_IVB(pipe)),
+				     I915_READ(PIPE_CRC_RES_5_IVB(pipe)));
+}
+
+static void i9xx_pipe_crc_irq_handler(struct drm_device *dev, enum pipe pipe)
+{
+	struct drm_i915_private *dev_priv = dev->dev_private;
+	uint32_t res1, res2;
+
+	if (INTEL_INFO(dev)->gen >= 3)
+		res1 = I915_READ(PIPE_CRC_RES_RES1_I915(pipe));
+	else
+		res1 = 0;
+
+	if (INTEL_INFO(dev)->gen >= 5 || IS_G4X(dev))
+		res2 = I915_READ(PIPE_CRC_RES_RES2_G4X(pipe));
+	else
+		res2 = 0;
+
+	display_pipe_crc_irq_handler(dev, pipe,
+				     I915_READ(PIPE_CRC_RES_RED(pipe)),
+				     I915_READ(PIPE_CRC_RES_GREEN(pipe)),
+				     I915_READ(PIPE_CRC_RES_BLUE(pipe)),
+				     res1, res2);
+}
+
 /* The RPS events need forcewake, so we add them to a work queue and mask their
  * IMR bits until the work is done. Other interrupts can be processed without
  * the work queue. */
@@ -1117,13 +1452,16 @@ static irqreturn_t valleyview_irq_handler(int irq, void *arg)
 		spin_unlock_irqrestore(&dev_priv->irq_lock, irqflags);
 
 		for_each_pipe(pipe) {
-			if (pipe_stats[pipe] & PIPE_VBLANK_INTERRUPT_STATUS)
+			if (pipe_stats[pipe] & PIPE_START_VBLANK_INTERRUPT_STATUS)
 				drm_handle_vblank(dev, pipe);
 
 			if (pipe_stats[pipe] & PLANE_FLIPDONE_INT_STATUS_VLV) {
 				intel_prepare_page_flip(dev, pipe);
 				intel_finish_page_flip(dev, pipe);
 			}
+
+			if (pipe_stats[pipe] & PIPE_CRC_DONE_INTERRUPT_STATUS)
+				i9xx_pipe_crc_irq_handler(dev, pipe);
 		}
 
 		/* Consume port.  Then clear IIR or we'll miss events */
@@ -1212,21 +1550,26 @@ static void ivb_err_int_handler(struct drm_device *dev)
 {
 	struct drm_i915_private *dev_priv = dev->dev_private;
 	u32 err_int = I915_READ(GEN7_ERR_INT);
+	enum pipe pipe;
 
 	if (err_int & ERR_INT_POISON)
 		DRM_ERROR("Poison interrupt\n");
 
-	if (err_int & ERR_INT_FIFO_UNDERRUN_A)
-		if (intel_set_cpu_fifo_underrun_reporting(dev, PIPE_A, false))
-			DRM_DEBUG_DRIVER("Pipe A FIFO underrun\n");
-
-	if (err_int & ERR_INT_FIFO_UNDERRUN_B)
-		if (intel_set_cpu_fifo_underrun_reporting(dev, PIPE_B, false))
-			DRM_DEBUG_DRIVER("Pipe B FIFO underrun\n");
+	for_each_pipe(pipe) {
+		if (err_int & ERR_INT_FIFO_UNDERRUN(pipe)) {
+			if (intel_set_cpu_fifo_underrun_reporting(dev, pipe,
+								  false))
+				DRM_DEBUG_DRIVER("Pipe %c FIFO underrun\n",
+						 pipe_name(pipe));
+		}
 
-	if (err_int & ERR_INT_FIFO_UNDERRUN_C)
-		if (intel_set_cpu_fifo_underrun_reporting(dev, PIPE_C, false))
-			DRM_DEBUG_DRIVER("Pipe C FIFO underrun\n");
+		if (err_int & ERR_INT_PIPE_CRC_DONE(pipe)) {
+			if (IS_IVYBRIDGE(dev))
+				ivb_pipe_crc_irq_handler(dev, pipe);
+			else
+				hsw_pipe_crc_irq_handler(dev, pipe);
+		}
+	}
 
 	I915_WRITE(GEN7_ERR_INT, err_int);
 }
@@ -1297,6 +1640,7 @@ static void cpt_irq_handler(struct drm_device *dev, u32 pch_iir)
 static void ilk_display_irq_handler(struct drm_device *dev, u32 de_iir)
 {
 	struct drm_i915_private *dev_priv = dev->dev_private;
+	enum pipe pipe;
 
 	if (de_iir & DE_AUX_CHANNEL_A)
 		dp_aux_irq_handler(dev);
@@ -1304,31 +1648,26 @@ static void ilk_display_irq_handler(struct drm_device *dev, u32 de_iir)
 	if (de_iir & DE_GSE)
 		intel_opregion_asle_intr(dev);
 
-	if (de_iir & DE_PIPEA_VBLANK)
-		drm_handle_vblank(dev, 0);
-
-	if (de_iir & DE_PIPEB_VBLANK)
-		drm_handle_vblank(dev, 1);
-
 	if (de_iir & DE_POISON)
 		DRM_ERROR("Poison interrupt\n");
 
-	if (de_iir & DE_PIPEA_FIFO_UNDERRUN)
-		if (intel_set_cpu_fifo_underrun_reporting(dev, PIPE_A, false))
-			DRM_DEBUG_DRIVER("Pipe A FIFO underrun\n");
+	for_each_pipe(pipe) {
+		if (de_iir & DE_PIPE_VBLANK(pipe))
+			drm_handle_vblank(dev, pipe);
 
-	if (de_iir & DE_PIPEB_FIFO_UNDERRUN)
-		if (intel_set_cpu_fifo_underrun_reporting(dev, PIPE_B, false))
-			DRM_DEBUG_DRIVER("Pipe B FIFO underrun\n");
+		if (de_iir & DE_PIPE_FIFO_UNDERRUN(pipe))
+			if (intel_set_cpu_fifo_underrun_reporting(dev, pipe, false))
+				DRM_DEBUG_DRIVER("Pipe %c FIFO underrun\n",
+						 pipe_name(pipe));
 
-	if (de_iir & DE_PLANEA_FLIP_DONE) {
-		intel_prepare_page_flip(dev, 0);
-		intel_finish_page_flip_plane(dev, 0);
-	}
+		if (de_iir & DE_PIPE_CRC_DONE(pipe))
+			i9xx_pipe_crc_irq_handler(dev, pipe);
 
-	if (de_iir & DE_PLANEB_FLIP_DONE) {
-		intel_prepare_page_flip(dev, 1);
-		intel_finish_page_flip_plane(dev, 1);
+		/* plane/pipes map 1:1 on ilk+ */
+		if (de_iir & DE_PLANE_FLIP_DONE(pipe)) {
+			intel_prepare_page_flip(dev, pipe);
+			intel_finish_page_flip_plane(dev, pipe);
+		}
 	}
 
 	/* check event from PCH */
@@ -1351,7 +1690,7 @@ static void ilk_display_irq_handler(struct drm_device *dev, u32 de_iir)
 static void ivb_display_irq_handler(struct drm_device *dev, u32 de_iir)
 {
 	struct drm_i915_private *dev_priv = dev->dev_private;
-	int i;
+	enum pipe i;
 
 	if (de_iir & DE_ERR_INT_IVB)
 		ivb_err_int_handler(dev);
@@ -1362,10 +1701,12 @@ static void ivb_display_irq_handler(struct drm_device *dev, u32 de_iir)
 	if (de_iir & DE_GSE_IVB)
 		intel_opregion_asle_intr(dev);
 
-	for (i = 0; i < 3; i++) {
-		if (de_iir & (DE_PIPEA_VBLANK_IVB << (5 * i)))
+	for_each_pipe(i) {
+		if (de_iir & (DE_PIPE_VBLANK_IVB(i)))
 			drm_handle_vblank(dev, i);
-		if (de_iir & (DE_PLANEA_FLIP_DONE_IVB << (5 * i))) {
+
+		/* plane/pipes map 1:1 on ilk+ */
+		if (de_iir & DE_PLANE_FLIP_DONE_IVB(i)) {
 			intel_prepare_page_flip(dev, i);
 			intel_finish_page_flip_plane(dev, i);
 		}
@@ -1388,7 +1729,6 @@ static irqreturn_t ironlake_irq_handler(int irq, void *arg)
 	drm_i915_private_t *dev_priv = (drm_i915_private_t *) dev->dev_private;
 	u32 de_iir, gt_iir, de_ier, sde_ier = 0;
 	irqreturn_t ret = IRQ_NONE;
-	bool err_int_reenable = false;
 
 	atomic_inc(&dev_priv->irq_received);
 
@@ -1412,17 +1752,6 @@ static irqreturn_t ironlake_irq_handler(int irq, void *arg)
 		POSTING_READ(SDEIER);
 	}
 
-	/* On Haswell, also mask ERR_INT because we don't want to risk
-	 * generating "unclaimed register" interrupts from inside the interrupt
-	 * handler. */
-	if (IS_HASWELL(dev)) {
-		spin_lock(&dev_priv->irq_lock);
-		err_int_reenable = ~dev_priv->irq_mask & DE_ERR_INT_IVB;
-		if (err_int_reenable)
-			ironlake_disable_display_irq(dev_priv, DE_ERR_INT_IVB);
-		spin_unlock(&dev_priv->irq_lock);
-	}
-
 	gt_iir = I915_READ(GTIIR);
 	if (gt_iir) {
 		if (INTEL_INFO(dev)->gen >= 6)
@@ -1452,13 +1781,6 @@ static irqreturn_t ironlake_irq_handler(int irq, void *arg)
 		}
 	}
 
-	if (err_int_reenable) {
-		spin_lock(&dev_priv->irq_lock);
-		if (ivb_can_enable_err_int(dev))
-			ironlake_enable_display_irq(dev_priv, DE_ERR_INT_IVB);
-		spin_unlock(&dev_priv->irq_lock);
-	}
-
 	I915_WRITE(DEIER, de_ier);
 	POSTING_READ(DEIER);
 	if (!HAS_PCH_NOP(dev)) {
@@ -1469,6 +1791,117 @@ static irqreturn_t ironlake_irq_handler(int irq, void *arg)
 	return ret;
 }
 
+static irqreturn_t gen8_irq_handler(int irq, void *arg)
+{
+	struct drm_device *dev = arg;
+	struct drm_i915_private *dev_priv = dev->dev_private;
+	u32 master_ctl;
+	irqreturn_t ret = IRQ_NONE;
+	uint32_t tmp = 0;
+	enum pipe pipe;
+
+	atomic_inc(&dev_priv->irq_received);
+
+	master_ctl = I915_READ(GEN8_MASTER_IRQ);
+	master_ctl &= ~GEN8_MASTER_IRQ_CONTROL;
+	if (!master_ctl)
+		return IRQ_NONE;
+
+	I915_WRITE(GEN8_MASTER_IRQ, 0);
+	POSTING_READ(GEN8_MASTER_IRQ);
+
+	ret = gen8_gt_irq_handler(dev, dev_priv, master_ctl);
+
+	if (master_ctl & GEN8_DE_MISC_IRQ) {
+		tmp = I915_READ(GEN8_DE_MISC_IIR);
+		if (tmp & GEN8_DE_MISC_GSE)
+			intel_opregion_asle_intr(dev);
+		else if (tmp)
+			DRM_ERROR("Unexpected DE Misc interrupt\n");
+		else
+			DRM_ERROR("The master control interrupt lied (DE MISC)!\n");
+
+		if (tmp) {
+			I915_WRITE(GEN8_DE_MISC_IIR, tmp);
+			ret = IRQ_HANDLED;
+		}
+	}
+
+	if (master_ctl & GEN8_DE_PORT_IRQ) {
+		tmp = I915_READ(GEN8_DE_PORT_IIR);
+		if (tmp & GEN8_AUX_CHANNEL_A)
+			dp_aux_irq_handler(dev);
+		else if (tmp)
+			DRM_ERROR("Unexpected DE Port interrupt\n");
+		else
+			DRM_ERROR("The master control interrupt lied (DE PORT)!\n");
+
+		if (tmp) {
+			I915_WRITE(GEN8_DE_PORT_IIR, tmp);
+			ret = IRQ_HANDLED;
+		}
+	}
+
+	for_each_pipe(pipe) {
+		uint32_t pipe_iir;
+
+		if (!(master_ctl & GEN8_DE_PIPE_IRQ(pipe)))
+			continue;
+
+		pipe_iir = I915_READ(GEN8_DE_PIPE_IIR(pipe));
+		if (pipe_iir & GEN8_PIPE_VBLANK)
+			drm_handle_vblank(dev, pipe);
+
+		if (pipe_iir & GEN8_PIPE_FLIP_DONE) {
+			intel_prepare_page_flip(dev, pipe);
+			intel_finish_page_flip_plane(dev, pipe);
+		}
+
+		if (pipe_iir & GEN8_PIPE_CDCLK_CRC_DONE)
+			hsw_pipe_crc_irq_handler(dev, pipe);
+
+		if (pipe_iir & GEN8_PIPE_FIFO_UNDERRUN) {
+			if (intel_set_cpu_fifo_underrun_reporting(dev, pipe,
+								  false))
+				DRM_DEBUG_DRIVER("Pipe %c FIFO underrun\n",
+						 pipe_name(pipe));
+		}
+
+		if (pipe_iir & GEN8_DE_PIPE_IRQ_FAULT_ERRORS) {
+			DRM_ERROR("Fault errors on pipe %c\n: 0x%08x",
+				  pipe_name(pipe),
+				  pipe_iir & GEN8_DE_PIPE_IRQ_FAULT_ERRORS);
+		}
+
+		if (pipe_iir) {
+			ret = IRQ_HANDLED;
+			I915_WRITE(GEN8_DE_PIPE_IIR(pipe), pipe_iir);
+		} else
+			DRM_ERROR("The master control interrupt lied (DE PIPE)!\n");
+	}
+
+	if (!HAS_PCH_NOP(dev) && master_ctl & GEN8_DE_PCH_IRQ) {
+		/*
+		 * FIXME(BDW): Assume for now that the new interrupt handling
+		 * scheme also closed the SDE interrupt handling race we've seen
+		 * on older pch-split platforms. But this needs testing.
+		 */
+		u32 pch_iir = I915_READ(SDEIIR);
+
+		cpt_irq_handler(dev, pch_iir);
+
+		if (pch_iir) {
+			I915_WRITE(SDEIIR, pch_iir);
+			ret = IRQ_HANDLED;
+		}
+	}
+
+	I915_WRITE(GEN8_MASTER_IRQ, GEN8_MASTER_IRQ_CONTROL);
+	POSTING_READ(GEN8_MASTER_IRQ);
+
+	return ret;
+}
+
 static void i915_error_wake_up(struct drm_i915_private *dev_priv,
 			       bool reset_completed)
 {
@@ -1516,7 +1949,7 @@ static void i915_error_work_func(struct work_struct *work)
 	char *reset_done_event[] = { I915_ERROR_UEVENT "=0", NULL };
 	int ret;
 
-	kobject_uevent_env(&dev->primary->kdev.kobj, KOBJ_CHANGE, error_event);
+	kobject_uevent_env(&dev->primary->kdev->kobj, KOBJ_CHANGE, error_event);
 
 	/*
 	 * Note that there's only one work item which does gpu resets, so we
@@ -1530,7 +1963,7 @@ static void i915_error_work_func(struct work_struct *work)
 	 */
 	if (i915_reset_in_progress(error) && !i915_terminally_wedged(error)) {
 		DRM_DEBUG_DRIVER("resetting chip\n");
-		kobject_uevent_env(&dev->primary->kdev.kobj, KOBJ_CHANGE,
+		kobject_uevent_env(&dev->primary->kdev->kobj, KOBJ_CHANGE,
 				   reset_event);
 
 		/*
@@ -1557,7 +1990,7 @@ static void i915_error_work_func(struct work_struct *work)
 			smp_mb__before_atomic_inc();
 			atomic_inc(&dev_priv->gpu_error.reset_counter);
 
-			kobject_uevent_env(&dev->primary->kdev.kobj,
+			kobject_uevent_env(&dev->primary->kdev->kobj,
 					   KOBJ_CHANGE, reset_done_event);
 		} else {
 			atomic_set(&error->reset_counter, I915_WEDGED);
@@ -1787,7 +2220,7 @@ static int ironlake_enable_vblank(struct drm_device *dev, int pipe)
 	drm_i915_private_t *dev_priv = (drm_i915_private_t *) dev->dev_private;
 	unsigned long irqflags;
 	uint32_t bit = (INTEL_INFO(dev)->gen >= 7) ? DE_PIPE_VBLANK_IVB(pipe) :
-						     DE_PIPE_VBLANK_ILK(pipe);
+						     DE_PIPE_VBLANK(pipe);
 
 	if (!i915_pipe_enabled(dev, pipe))
 		return -EINVAL;
@@ -1810,7 +2243,7 @@ static int valleyview_enable_vblank(struct drm_device *dev, int pipe)
 
 	spin_lock_irqsave(&dev_priv->irq_lock, irqflags);
 	imr = I915_READ(VLV_IMR);
-	if (pipe == 0)
+	if (pipe == PIPE_A)
 		imr &= ~I915_DISPLAY_PIPE_A_VBLANK_INTERRUPT;
 	else
 		imr &= ~I915_DISPLAY_PIPE_B_VBLANK_INTERRUPT;
@@ -1822,6 +2255,22 @@ static int valleyview_enable_vblank(struct drm_device *dev, int pipe)
 	return 0;
 }
 
+static int gen8_enable_vblank(struct drm_device *dev, int pipe)
+{
+	struct drm_i915_private *dev_priv = dev->dev_private;
+	unsigned long irqflags;
+
+	if (!i915_pipe_enabled(dev, pipe))
+		return -EINVAL;
+
+	spin_lock_irqsave(&dev_priv->irq_lock, irqflags);
+	dev_priv->de_irq_mask[pipe] &= ~GEN8_PIPE_VBLANK;
+	I915_WRITE(GEN8_DE_PIPE_IMR(pipe), dev_priv->de_irq_mask[pipe]);
+	POSTING_READ(GEN8_DE_PIPE_IMR(pipe));
+	spin_unlock_irqrestore(&dev_priv->irq_lock, irqflags);
+	return 0;
+}
+
 /* Called from drm generic code, passed 'crtc' which
  * we use as a pipe index
  */
@@ -1845,7 +2294,7 @@ static void ironlake_disable_vblank(struct drm_device *dev, int pipe)
 	drm_i915_private_t *dev_priv = (drm_i915_private_t *) dev->dev_private;
 	unsigned long irqflags;
 	uint32_t bit = (INTEL_INFO(dev)->gen >= 7) ? DE_PIPE_VBLANK_IVB(pipe) :
-						     DE_PIPE_VBLANK_ILK(pipe);
+						     DE_PIPE_VBLANK(pipe);
 
 	spin_lock_irqsave(&dev_priv->irq_lock, irqflags);
 	ironlake_disable_display_irq(dev_priv, bit);
@@ -1862,7 +2311,7 @@ static void valleyview_disable_vblank(struct drm_device *dev, int pipe)
 	i915_disable_pipestat(dev_priv, pipe,
 			      PIPE_START_VBLANK_INTERRUPT_ENABLE);
 	imr = I915_READ(VLV_IMR);
-	if (pipe == 0)
+	if (pipe == PIPE_A)
 		imr |= I915_DISPLAY_PIPE_A_VBLANK_INTERRUPT;
 	else
 		imr |= I915_DISPLAY_PIPE_B_VBLANK_INTERRUPT;
@@ -1870,6 +2319,21 @@ static void valleyview_disable_vblank(struct drm_device *dev, int pipe)
 	spin_unlock_irqrestore(&dev_priv->irq_lock, irqflags);
 }
 
+static void gen8_disable_vblank(struct drm_device *dev, int pipe)
+{
+	struct drm_i915_private *dev_priv = dev->dev_private;
+	unsigned long irqflags;
+
+	if (!i915_pipe_enabled(dev, pipe))
+		return;
+
+	spin_lock_irqsave(&dev_priv->irq_lock, irqflags);
+	dev_priv->de_irq_mask[pipe] |= GEN8_PIPE_VBLANK;
+	I915_WRITE(GEN8_DE_PIPE_IMR(pipe), dev_priv->de_irq_mask[pipe]);
+	POSTING_READ(GEN8_DE_PIPE_IMR(pipe));
+	spin_unlock_irqrestore(&dev_priv->irq_lock, irqflags);
+}
+
 static u32
 ring_last_seqno(struct intel_ring_buffer *ring)
 {
@@ -1965,6 +2429,7 @@ ring_stuck(struct intel_ring_buffer *ring, u32 acthd)
 	if (tmp & RING_WAIT) {
 		DRM_ERROR("Kicking stuck wait on %s\n",
 			  ring->name);
+		i915_handle_error(dev, false);
 		I915_WRITE_CTL(ring, tmp);
 		return HANGCHECK_KICK;
 	}
@@ -1976,6 +2441,7 @@ ring_stuck(struct intel_ring_buffer *ring, u32 acthd)
 		case 1:
 			DRM_ERROR("Kicking stuck semaphore on %s\n",
 				  ring->name);
+			i915_handle_error(dev, false);
 			I915_WRITE_CTL(ring, tmp);
 			return HANGCHECK_KICK;
 		case 0:
@@ -2021,12 +2487,21 @@ static void i915_hangcheck_elapsed(unsigned long data)
 
 		if (ring->hangcheck.seqno == seqno) {
 			if (ring_idle(ring, seqno)) {
+				ring->hangcheck.action = HANGCHECK_IDLE;
+
 				if (waitqueue_active(&ring->irq_queue)) {
 					/* Issue a wake-up to catch stuck h/w. */
-					DRM_ERROR("Hangcheck timer elapsed... %s idle\n",
-						  ring->name);
-					wake_up_all(&ring->irq_queue);
-					ring->hangcheck.score += HUNG;
+					if (!test_and_set_bit(ring->id, &dev_priv->gpu_error.missed_irq_rings)) {
+						if (!(dev_priv->gpu_error.test_irq_rings & intel_ring_flag(ring)))
+							DRM_ERROR("Hangcheck timer elapsed... %s idle\n",
+								  ring->name);
+						else
+							DRM_INFO("Fake missed irq on %s\n",
+								 ring->name);
+						wake_up_all(&ring->irq_queue);
+					}
+					/* Safeguard against driver failure */
+					ring->hangcheck.score += BUSY;
 				} else
 					busy = false;
 			} else {
@@ -2049,6 +2524,7 @@ static void i915_hangcheck_elapsed(unsigned long data)
 								    acthd);
 
 				switch (ring->hangcheck.action) {
+				case HANGCHECK_IDLE:
 				case HANGCHECK_WAIT:
 					break;
 				case HANGCHECK_ACTIVE:
@@ -2064,6 +2540,8 @@ static void i915_hangcheck_elapsed(unsigned long data)
 				}
 			}
 		} else {
+			ring->hangcheck.action = HANGCHECK_ACTIVE;
+
 			/* Gradually reduce the count so that we catch DoS
 			 * attempts across multiple batches.
 			 */
@@ -2190,6 +2668,53 @@ static void valleyview_irq_preinstall(struct drm_device *dev)
 	POSTING_READ(VLV_IER);
 }
 
+static void gen8_irq_preinstall(struct drm_device *dev)
+{
+	struct drm_i915_private *dev_priv = dev->dev_private;
+	int pipe;
+
+	atomic_set(&dev_priv->irq_received, 0);
+
+	I915_WRITE(GEN8_MASTER_IRQ, 0);
+	POSTING_READ(GEN8_MASTER_IRQ);
+
+	/* IIR can theoretically queue up two events. Be paranoid */
+#define GEN8_IRQ_INIT_NDX(type, which) do { \
+		I915_WRITE(GEN8_##type##_IMR(which), 0xffffffff); \
+		POSTING_READ(GEN8_##type##_IMR(which)); \
+		I915_WRITE(GEN8_##type##_IER(which), 0); \
+		I915_WRITE(GEN8_##type##_IIR(which), 0xffffffff); \
+		POSTING_READ(GEN8_##type##_IIR(which)); \
+		I915_WRITE(GEN8_##type##_IIR(which), 0xffffffff); \
+	} while (0)
+
+#define GEN8_IRQ_INIT(type) do { \
+		I915_WRITE(GEN8_##type##_IMR, 0xffffffff); \
+		POSTING_READ(GEN8_##type##_IMR); \
+		I915_WRITE(GEN8_##type##_IER, 0); \
+		I915_WRITE(GEN8_##type##_IIR, 0xffffffff); \
+		POSTING_READ(GEN8_##type##_IIR); \
+		I915_WRITE(GEN8_##type##_IIR, 0xffffffff); \
+	} while (0)
+
+	GEN8_IRQ_INIT_NDX(GT, 0);
+	GEN8_IRQ_INIT_NDX(GT, 1);
+	GEN8_IRQ_INIT_NDX(GT, 2);
+	GEN8_IRQ_INIT_NDX(GT, 3);
+
+	for_each_pipe(pipe) {
+		GEN8_IRQ_INIT_NDX(DE_PIPE, pipe);
+	}
+
+	GEN8_IRQ_INIT(DE_PORT);
+	GEN8_IRQ_INIT(DE_MISC);
+	GEN8_IRQ_INIT(PCU);
+#undef GEN8_IRQ_INIT
+#undef GEN8_IRQ_INIT_NDX
+
+	POSTING_READ(GEN8_PCU_IIR);
+}
+
 static void ibx_hpd_irq_setup(struct drm_device *dev)
 {
 	drm_i915_private_t *dev_priv = (drm_i915_private_t *) dev->dev_private;
@@ -2254,10 +2779,10 @@ static void gen5_gt_irq_postinstall(struct drm_device *dev)
 	pm_irqs = gt_irqs = 0;
 
 	dev_priv->gt_irq_mask = ~0;
-	if (HAS_L3_GPU_CACHE(dev)) {
+	if (HAS_L3_DPF(dev)) {
 		/* L3 parity interrupt is always unmasked. */
-		dev_priv->gt_irq_mask = ~GT_RENDER_L3_PARITY_ERROR_INTERRUPT;
-		gt_irqs |= GT_RENDER_L3_PARITY_ERROR_INTERRUPT;
+		dev_priv->gt_irq_mask = ~GT_PARITY_ERROR(dev);
+		gt_irqs |= GT_PARITY_ERROR(dev);
 	}
 
 	gt_irqs |= GT_RENDER_USER_INTERRUPT;
@@ -2306,8 +2831,10 @@ static int ironlake_irq_postinstall(struct drm_device *dev)
 	} else {
 		display_mask = (DE_MASTER_IRQ_CONTROL | DE_GSE | DE_PCH_EVENT |
 				DE_PLANEA_FLIP_DONE | DE_PLANEB_FLIP_DONE |
-				DE_AUX_CHANNEL_A | DE_PIPEB_FIFO_UNDERRUN |
-				DE_PIPEA_FIFO_UNDERRUN | DE_POISON);
+				DE_AUX_CHANNEL_A |
+				DE_PIPEB_FIFO_UNDERRUN | DE_PIPEA_FIFO_UNDERRUN |
+				DE_PIPEB_CRC_DONE | DE_PIPEA_CRC_DONE |
+				DE_POISON);
 		extra_mask = DE_PIPEA_VBLANK | DE_PIPEB_VBLANK | DE_PCU_EVENT;
 	}
 
@@ -2341,7 +2868,8 @@ static int valleyview_irq_postinstall(struct drm_device *dev)
 {
 	drm_i915_private_t *dev_priv = (drm_i915_private_t *) dev->dev_private;
 	u32 enable_mask;
-	u32 pipestat_enable = PLANE_FLIP_DONE_INT_EN_VLV;
+	u32 pipestat_enable = PLANE_FLIP_DONE_INT_EN_VLV |
+		PIPE_CRC_DONE_ENABLE;
 	unsigned long irqflags;
 
 	enable_mask = I915_DISPLAY_PORT_INTERRUPT;
@@ -2371,9 +2899,9 @@ static int valleyview_irq_postinstall(struct drm_device *dev)
 	/* Interrupt setup is already guaranteed to be single-threaded, this is
 	 * just to make the assert_spin_locked check happy. */
 	spin_lock_irqsave(&dev_priv->irq_lock, irqflags);
-	i915_enable_pipestat(dev_priv, 0, pipestat_enable);
-	i915_enable_pipestat(dev_priv, 0, PIPE_GMBUS_EVENT_ENABLE);
-	i915_enable_pipestat(dev_priv, 1, pipestat_enable);
+	i915_enable_pipestat(dev_priv, PIPE_A, pipestat_enable);
+	i915_enable_pipestat(dev_priv, PIPE_A, PIPE_GMBUS_EVENT_ENABLE);
+	i915_enable_pipestat(dev_priv, PIPE_B, pipestat_enable);
 	spin_unlock_irqrestore(&dev_priv->irq_lock, irqflags);
 
 	I915_WRITE(VLV_IIR, 0xffffffff);
@@ -2392,6 +2920,117 @@ static int valleyview_irq_postinstall(struct drm_device *dev)
 	return 0;
 }
 
+static void gen8_gt_irq_postinstall(struct drm_i915_private *dev_priv)
+{
+	int i;
+
+	/* These are interrupts we'll toggle with the ring mask register */
+	uint32_t gt_interrupts[] = {
+		GT_RENDER_USER_INTERRUPT << GEN8_RCS_IRQ_SHIFT |
+			GT_RENDER_L3_PARITY_ERROR_INTERRUPT |
+			GT_RENDER_USER_INTERRUPT << GEN8_BCS_IRQ_SHIFT,
+		GT_RENDER_USER_INTERRUPT << GEN8_VCS1_IRQ_SHIFT |
+			GT_RENDER_USER_INTERRUPT << GEN8_VCS2_IRQ_SHIFT,
+		0,
+		GT_RENDER_USER_INTERRUPT << GEN8_VECS_IRQ_SHIFT
+		};
+
+	for (i = 0; i < ARRAY_SIZE(gt_interrupts); i++) {
+		u32 tmp = I915_READ(GEN8_GT_IIR(i));
+		if (tmp)
+			DRM_ERROR("Interrupt (%d) should have been masked in pre-install 0x%08x\n",
+				  i, tmp);
+		I915_WRITE(GEN8_GT_IMR(i), ~gt_interrupts[i]);
+		I915_WRITE(GEN8_GT_IER(i), gt_interrupts[i]);
+	}
+	POSTING_READ(GEN8_GT_IER(0));
+}
+
+static void gen8_de_irq_postinstall(struct drm_i915_private *dev_priv)
+{
+	struct drm_device *dev = dev_priv->dev;
+	uint32_t de_pipe_masked = GEN8_PIPE_FLIP_DONE |
+		GEN8_PIPE_CDCLK_CRC_DONE |
+		GEN8_PIPE_FIFO_UNDERRUN |
+		GEN8_DE_PIPE_IRQ_FAULT_ERRORS;
+	uint32_t de_pipe_enables = de_pipe_masked | GEN8_PIPE_VBLANK;
+	int pipe;
+	dev_priv->de_irq_mask[PIPE_A] = ~de_pipe_masked;
+	dev_priv->de_irq_mask[PIPE_B] = ~de_pipe_masked;
+	dev_priv->de_irq_mask[PIPE_C] = ~de_pipe_masked;
+
+	for_each_pipe(pipe) {
+		u32 tmp = I915_READ(GEN8_DE_PIPE_IIR(pipe));
+		if (tmp)
+			DRM_ERROR("Interrupt (%d) should have been masked in pre-install 0x%08x\n",
+				  pipe, tmp);
+		I915_WRITE(GEN8_DE_PIPE_IMR(pipe), dev_priv->de_irq_mask[pipe]);
+		I915_WRITE(GEN8_DE_PIPE_IER(pipe), de_pipe_enables);
+	}
+	POSTING_READ(GEN8_DE_PIPE_ISR(0));
+
+	I915_WRITE(GEN8_DE_PORT_IMR, ~GEN8_AUX_CHANNEL_A);
+	I915_WRITE(GEN8_DE_PORT_IER, GEN8_AUX_CHANNEL_A);
+	POSTING_READ(GEN8_DE_PORT_IER);
+}
+
+static int gen8_irq_postinstall(struct drm_device *dev)
+{
+	struct drm_i915_private *dev_priv = dev->dev_private;
+
+	gen8_gt_irq_postinstall(dev_priv);
+	gen8_de_irq_postinstall(dev_priv);
+
+	ibx_irq_postinstall(dev);
+
+	I915_WRITE(GEN8_MASTER_IRQ, DE_MASTER_IRQ_CONTROL);
+	POSTING_READ(GEN8_MASTER_IRQ);
+
+	return 0;
+}
+
+static void gen8_irq_uninstall(struct drm_device *dev)
+{
+	struct drm_i915_private *dev_priv = dev->dev_private;
+	int pipe;
+
+	if (!dev_priv)
+		return;
+
+	atomic_set(&dev_priv->irq_received, 0);
+
+	I915_WRITE(GEN8_MASTER_IRQ, 0);
+
+#define GEN8_IRQ_FINI_NDX(type, which) do { \
+		I915_WRITE(GEN8_##type##_IMR(which), 0xffffffff); \
+		I915_WRITE(GEN8_##type##_IER(which), 0); \
+		I915_WRITE(GEN8_##type##_IIR(which), 0xffffffff); \
+	} while (0)
+
+#define GEN8_IRQ_FINI(type) do { \
+		I915_WRITE(GEN8_##type##_IMR, 0xffffffff); \
+		I915_WRITE(GEN8_##type##_IER, 0); \
+		I915_WRITE(GEN8_##type##_IIR, 0xffffffff); \
+	} while (0)
+
+	GEN8_IRQ_FINI_NDX(GT, 0);
+	GEN8_IRQ_FINI_NDX(GT, 1);
+	GEN8_IRQ_FINI_NDX(GT, 2);
+	GEN8_IRQ_FINI_NDX(GT, 3);
+
+	for_each_pipe(pipe) {
+		GEN8_IRQ_FINI_NDX(DE_PIPE, pipe);
+	}
+
+	GEN8_IRQ_FINI(DE_PORT);
+	GEN8_IRQ_FINI(DE_MISC);
+	GEN8_IRQ_FINI(PCU);
+#undef GEN8_IRQ_FINI
+#undef GEN8_IRQ_FINI_NDX
+
+	POSTING_READ(GEN8_PCU_IIR);
+}
+
 static void valleyview_irq_uninstall(struct drm_device *dev)
 {
 	drm_i915_private_t *dev_priv = (drm_i915_private_t *) dev->dev_private;
@@ -2464,6 +3103,7 @@ static void i8xx_irq_preinstall(struct drm_device * dev)
 static int i8xx_irq_postinstall(struct drm_device *dev)
 {
 	drm_i915_private_t *dev_priv = (drm_i915_private_t *) dev->dev_private;
+	unsigned long irqflags;
 
 	I915_WRITE16(EMR,
 		     ~(I915_ERROR_PAGE_TABLE | I915_ERROR_MEMORY_REFRESH));
@@ -2484,6 +3124,13 @@ static int i8xx_irq_postinstall(struct drm_device *dev)
 		     I915_USER_INTERRUPT);
 	POSTING_READ16(IER);
 
+	/* Interrupt setup is already guaranteed to be single-threaded, this is
+	 * just to make the assert_spin_locked check happy. */
+	spin_lock_irqsave(&dev_priv->irq_lock, irqflags);
+	i915_enable_pipestat(dev_priv, PIPE_A, PIPE_CRC_DONE_ENABLE);
+	i915_enable_pipestat(dev_priv, PIPE_B, PIPE_CRC_DONE_ENABLE);
+	spin_unlock_irqrestore(&dev_priv->irq_lock, irqflags);
+
 	return 0;
 }
 
@@ -2570,13 +3217,14 @@ static irqreturn_t i8xx_irq_handler(int irq, void *arg)
 		if (iir & I915_USER_INTERRUPT)
 			notify_ring(dev, &dev_priv->ring[RCS]);
 
-		if (pipe_stats[0] & PIPE_VBLANK_INTERRUPT_STATUS &&
-		    i8xx_handle_vblank(dev, 0, iir))
-			flip_mask &= ~DISPLAY_PLANE_FLIP_PENDING(0);
+		for_each_pipe(pipe) {
+			if (pipe_stats[pipe] & PIPE_VBLANK_INTERRUPT_STATUS &&
+			    i8xx_handle_vblank(dev, pipe, iir))
+				flip_mask &= ~DISPLAY_PLANE_FLIP_PENDING(pipe);
 
-		if (pipe_stats[1] & PIPE_VBLANK_INTERRUPT_STATUS &&
-		    i8xx_handle_vblank(dev, 1, iir))
-			flip_mask &= ~DISPLAY_PLANE_FLIP_PENDING(1);
+			if (pipe_stats[pipe] & PIPE_CRC_DONE_INTERRUPT_STATUS)
+				i9xx_pipe_crc_irq_handler(dev, pipe);
+		}
 
 		iir = new_iir;
 	}
@@ -2623,6 +3271,7 @@ static int i915_irq_postinstall(struct drm_device *dev)
 {
 	drm_i915_private_t *dev_priv = (drm_i915_private_t *) dev->dev_private;
 	u32 enable_mask;
+	unsigned long irqflags;
 
 	I915_WRITE(EMR, ~(I915_ERROR_PAGE_TABLE | I915_ERROR_MEMORY_REFRESH));
 
@@ -2658,6 +3307,13 @@ static int i915_irq_postinstall(struct drm_device *dev)
 
 	i915_enable_asle_pipestat(dev);
 
+	/* Interrupt setup is already guaranteed to be single-threaded, this is
+	 * just to make the assert_spin_locked check happy. */
+	spin_lock_irqsave(&dev_priv->irq_lock, irqflags);
+	i915_enable_pipestat(dev_priv, PIPE_A, PIPE_CRC_DONE_ENABLE);
+	i915_enable_pipestat(dev_priv, PIPE_B, PIPE_CRC_DONE_ENABLE);
+	spin_unlock_irqrestore(&dev_priv->irq_lock, irqflags);
+
 	return 0;
 }
 
@@ -2769,6 +3425,9 @@ static irqreturn_t i915_irq_handler(int irq, void *arg)
 
 			if (pipe_stats[pipe] & PIPE_LEGACY_BLC_EVENT_STATUS)
 				blc_event = true;
+
+			if (pipe_stats[pipe] & PIPE_CRC_DONE_INTERRUPT_STATUS)
+				i9xx_pipe_crc_irq_handler(dev, pipe);
 		}
 
 		if (blc_event || (iir & I915_ASLE_INTERRUPT))
@@ -2867,7 +3526,9 @@ static int i965_irq_postinstall(struct drm_device *dev)
 	/* Interrupt setup is already guaranteed to be single-threaded, this is
 	 * just to make the assert_spin_locked check happy. */
 	spin_lock_irqsave(&dev_priv->irq_lock, irqflags);
-	i915_enable_pipestat(dev_priv, 0, PIPE_GMBUS_EVENT_ENABLE);
+	i915_enable_pipestat(dev_priv, PIPE_A, PIPE_GMBUS_EVENT_ENABLE);
+	i915_enable_pipestat(dev_priv, PIPE_A, PIPE_CRC_DONE_ENABLE);
+	i915_enable_pipestat(dev_priv, PIPE_B, PIPE_CRC_DONE_ENABLE);
 	spin_unlock_irqrestore(&dev_priv->irq_lock, irqflags);
 
 	/*
@@ -3013,6 +3674,9 @@ static irqreturn_t i965_irq_handler(int irq, void *arg)
 
 			if (pipe_stats[pipe] & PIPE_LEGACY_BLC_EVENT_STATUS)
 				blc_event = true;
+
+			if (pipe_stats[pipe] & PIPE_CRC_DONE_INTERRUPT_STATUS)
+				i9xx_pipe_crc_irq_handler(dev, pipe);
 		}
 
 
@@ -3122,18 +3786,21 @@ void intel_irq_init(struct drm_device *dev)
 
 	pm_qos_add_request(&dev_priv->pm_qos, PM_QOS_CPU_DMA_LATENCY, PM_QOS_DEFAULT_VALUE);
 
-	dev->driver->get_vblank_counter = i915_get_vblank_counter;
-	dev->max_vblank_count = 0xffffff; /* only 24 bits of frame count */
-	if (IS_G4X(dev) || INTEL_INFO(dev)->gen >= 5) {
+	if (IS_GEN2(dev)) {
+		dev->max_vblank_count = 0;
+		dev->driver->get_vblank_counter = i8xx_get_vblank_counter;
+	} else if (IS_G4X(dev) || INTEL_INFO(dev)->gen >= 5) {
 		dev->max_vblank_count = 0xffffffff; /* full 32 bit counter */
 		dev->driver->get_vblank_counter = gm45_get_vblank_counter;
+	} else {
+		dev->driver->get_vblank_counter = i915_get_vblank_counter;
+		dev->max_vblank_count = 0xffffff; /* only 24 bits of frame count */
 	}
 
-	if (drm_core_check_feature(dev, DRIVER_MODESET))
+	if (drm_core_check_feature(dev, DRIVER_MODESET)) {
 		dev->driver->get_vblank_timestamp = i915_get_vblank_timestamp;
-	else
-		dev->driver->get_vblank_timestamp = NULL;
-	dev->driver->get_scanout_position = i915_get_crtc_scanoutpos;
+		dev->driver->get_scanout_position = i915_get_crtc_scanoutpos;
+	}
 
 	if (IS_VALLEYVIEW(dev)) {
 		dev->driver->irq_handler = valleyview_irq_handler;
@@ -3143,6 +3810,14 @@ void intel_irq_init(struct drm_device *dev)
 		dev->driver->enable_vblank = valleyview_enable_vblank;
 		dev->driver->disable_vblank = valleyview_disable_vblank;
 		dev_priv->display.hpd_irq_setup = i915_hpd_irq_setup;
+	} else if (IS_GEN8(dev)) {
+		dev->driver->irq_handler = gen8_irq_handler;
+		dev->driver->irq_preinstall = gen8_irq_preinstall;
+		dev->driver->irq_postinstall = gen8_irq_postinstall;
+		dev->driver->irq_uninstall = gen8_irq_uninstall;
+		dev->driver->enable_vblank = gen8_enable_vblank;
+		dev->driver->disable_vblank = gen8_disable_vblank;
+		dev_priv->display.hpd_irq_setup = ibx_hpd_irq_setup;
 	} else if (HAS_PCH_SPLIT(dev)) {
 		dev->driver->irq_handler = ironlake_irq_handler;
 		dev->driver->irq_preinstall = ironlake_irq_preinstall;
diff --git a/drivers/gpu/drm/i915/i915_reg.h b/drivers/gpu/drm/i915/i915_reg.h
index ef9b35479f01..f9eafb6ed523 100644
--- a/drivers/gpu/drm/i915/i915_reg.h
+++ b/drivers/gpu/drm/i915/i915_reg.h
@@ -26,6 +26,7 @@
 #define _I915_REG_H_
 
 #define _PIPE(pipe, a, b) ((a) + (pipe)*((b)-(a)))
+#define _PIPE_INC(pipe, base, inc) ((base) + (pipe)*(inc))
 #define _TRANSCODER(tran, a, b) ((a) + (tran)*((b)-(a)))
 
 #define _PORT(port, a, b) ((a) + (port)*((b)-(a)))
@@ -109,6 +110,9 @@
 #define RING_PP_DIR_DCLV(ring)		((ring)->mmio_base+0x220)
 #define   PP_DIR_DCLV_2G		0xffffffff
 
+#define GEN8_RING_PDP_UDW(ring, n)	((ring)->mmio_base+0x270 + ((n) * 8 + 4))
+#define GEN8_RING_PDP_LDW(ring, n)	((ring)->mmio_base+0x270 + (n) * 8)
+
 #define GAM_ECOCHK			0x4090
 #define   ECOCHK_SNB_BIT		(1<<10)
 #define   HSW_ECOCHK_ARB_PRIO_SOL	(1<<6)
@@ -246,6 +250,7 @@
 #define   MI_BATCH_NON_SECURE_HSW 	(1<<13)
 #define MI_BATCH_BUFFER_START	MI_INSTR(0x31, 0)
 #define   MI_BATCH_GTT		    (2<<6) /* aliased with (1<<7) on gen4 */
+#define MI_BATCH_BUFFER_START_GEN8	MI_INSTR(0x31, 1)
 #define MI_SEMAPHORE_MBOX	MI_INSTR(0x16, 1) /* gen6+ */
 #define  MI_SEMAPHORE_GLOBAL_GTT    (1<<22)
 #define  MI_SEMAPHORE_UPDATE	    (1<<21)
@@ -264,6 +269,11 @@
 #define  MI_SEMAPHORE_SYNC_VVE	    (1<<16) /* VECS wait for VCS  (VEVSYNC) */
 #define  MI_SEMAPHORE_SYNC_RVE	    (2<<16) /* VECS wait for RCS  (VERSYNC) */
 #define  MI_SEMAPHORE_SYNC_INVALID  (3<<16)
+
+#define MI_PREDICATE_RESULT_2	(0x2214)
+#define  LOWER_SLICE_ENABLED	(1<<0)
+#define  LOWER_SLICE_DISABLED	(0<<0)
+
 /*
  * 3D instructions used by the kernel
  */
@@ -346,12 +356,25 @@
 #define   IOSF_PORT_PUNIT			0x4
 #define   IOSF_PORT_NC				0x11
 #define   IOSF_PORT_DPIO			0x12
+#define   IOSF_PORT_GPIO_NC			0x13
+#define   IOSF_PORT_CCK				0x14
+#define   IOSF_PORT_CCU				0xA9
+#define   IOSF_PORT_GPS_CORE			0x48
 #define VLV_IOSF_DATA				(VLV_DISPLAY_BASE + 0x2104)
 #define VLV_IOSF_ADDR				(VLV_DISPLAY_BASE + 0x2108)
 
 #define PUNIT_OPCODE_REG_READ			6
 #define PUNIT_OPCODE_REG_WRITE			7
 
+#define PUNIT_REG_PWRGT_CTRL			0x60
+#define PUNIT_REG_PWRGT_STATUS			0x61
+#define	  PUNIT_CLK_GATE			1
+#define	  PUNIT_PWR_RESET			2
+#define	  PUNIT_PWR_GATE			3
+#define	  RENDER_PWRGT				(PUNIT_PWR_GATE << 0)
+#define	  MEDIA_PWRGT				(PUNIT_PWR_GATE << 2)
+#define	  DISP2D_PWRGT				(PUNIT_PWR_GATE << 6)
+
 #define PUNIT_REG_GPU_LFM			0xd3
 #define PUNIT_REG_GPU_FREQ_REQ			0xd4
 #define PUNIT_REG_GPU_FREQ_STS			0xd8
@@ -372,6 +395,40 @@
 #define   FB_FMAX_VMIN_FREQ_LO_SHIFT		27
 #define   FB_FMAX_VMIN_FREQ_LO_MASK		0xf8000000
 
+/* vlv2 north clock has */
+#define CCK_FUSE_REG				0x8
+#define  CCK_FUSE_HPLL_FREQ_MASK		0x3
+#define CCK_REG_DSI_PLL_FUSE			0x44
+#define CCK_REG_DSI_PLL_CONTROL			0x48
+#define  DSI_PLL_VCO_EN				(1 << 31)
+#define  DSI_PLL_LDO_GATE			(1 << 30)
+#define  DSI_PLL_P1_POST_DIV_SHIFT		17
+#define  DSI_PLL_P1_POST_DIV_MASK		(0x1ff << 17)
+#define  DSI_PLL_P2_MUX_DSI0_DIV2		(1 << 13)
+#define  DSI_PLL_P3_MUX_DSI1_DIV2		(1 << 12)
+#define  DSI_PLL_MUX_MASK			(3 << 9)
+#define  DSI_PLL_MUX_DSI0_DSIPLL		(0 << 10)
+#define  DSI_PLL_MUX_DSI0_CCK			(1 << 10)
+#define  DSI_PLL_MUX_DSI1_DSIPLL		(0 << 9)
+#define  DSI_PLL_MUX_DSI1_CCK			(1 << 9)
+#define  DSI_PLL_CLK_GATE_MASK			(0xf << 5)
+#define  DSI_PLL_CLK_GATE_DSI0_DSIPLL		(1 << 8)
+#define  DSI_PLL_CLK_GATE_DSI1_DSIPLL		(1 << 7)
+#define  DSI_PLL_CLK_GATE_DSI0_CCK		(1 << 6)
+#define  DSI_PLL_CLK_GATE_DSI1_CCK		(1 << 5)
+#define  DSI_PLL_LOCK				(1 << 0)
+#define CCK_REG_DSI_PLL_DIVIDER			0x4c
+#define  DSI_PLL_LFSR				(1 << 31)
+#define  DSI_PLL_FRACTION_EN			(1 << 30)
+#define  DSI_PLL_FRAC_COUNTER_SHIFT		27
+#define  DSI_PLL_FRAC_COUNTER_MASK		(7 << 27)
+#define  DSI_PLL_USYNC_CNT_SHIFT		18
+#define  DSI_PLL_USYNC_CNT_MASK			(0x1ff << 18)
+#define  DSI_PLL_N1_DIV_SHIFT			16
+#define  DSI_PLL_N1_DIV_MASK			(3 << 16)
+#define  DSI_PLL_M1_DIV_SHIFT			0
+#define  DSI_PLL_M1_DIV_MASK			(0x1ff << 0)
+
 /*
  * DPIO - a special bus for various display related registers to hide behind
  *
@@ -387,11 +444,11 @@
 #define  DPIO_MODSEL1			(1<<3) /* if ref clk b == 27 */
 #define  DPIO_MODSEL0			(1<<2) /* if ref clk a == 27 */
 #define  DPIO_SFR_BYPASS		(1<<1)
-#define  DPIO_RESET			(1<<0)
+#define  DPIO_CMNRST			(1<<0)
 
 #define _DPIO_TX3_SWING_CTL4_A		0x690
 #define _DPIO_TX3_SWING_CTL4_B		0x2a90
-#define DPIO_TX3_SWING_CTL4(pipe) _PIPE(pipe, _DPIO_TX_SWING_CTL4_A, \
+#define DPIO_TX3_SWING_CTL4(pipe) _PIPE(pipe, _DPIO_TX3_SWING_CTL4_A, \
 					_DPIO_TX3_SWING_CTL4_B)
 
 /*
@@ -602,6 +659,9 @@
 #define ARB_MODE		0x04030
 #define   ARB_MODE_SWIZZLE_SNB	(1<<4)
 #define   ARB_MODE_SWIZZLE_IVB	(1<<5)
+#define GAMTARBMODE		0x04a08
+#define   ARB_MODE_BWGTLB_DISABLE (1<<9)
+#define   ARB_MODE_SWIZZLE_BDW	(1<<1)
 #define RENDER_HWS_PGA_GEN7	(0x04080)
 #define RING_FAULT_REG(ring)	(0x4094 + 0x100*(ring)->id)
 #define   RING_FAULT_GTTSEL_MASK (1<<11)
@@ -609,6 +669,7 @@
 #define   RING_FAULT_FAULT_TYPE(x) ((x >> 1) & 0x3)
 #define   RING_FAULT_VALID	(1<<0)
 #define DONE_REG		0x40b0
+#define GEN8_PRIVATE_PAT	0x40e0
 #define BSD_HWS_PGA_GEN7	(0x04180)
 #define BLT_HWS_PGA_GEN7	(0x04280)
 #define VEBOX_HWS_PGA_GEN7	(0x04380)
@@ -669,13 +730,18 @@
 #define NOPID		0x02094
 #define HWSTAM		0x02098
 #define DMA_FADD_I8XX	0x020d0
+#define RING_BBSTATE(base)	((base)+0x110)
 
 #define ERROR_GEN6	0x040a0
 #define GEN7_ERR_INT	0x44040
 #define   ERR_INT_POISON		(1<<31)
 #define   ERR_INT_MMIO_UNCLAIMED	(1<<13)
+#define   ERR_INT_PIPE_CRC_DONE_C	(1<<8)
 #define   ERR_INT_FIFO_UNDERRUN_C	(1<<6)
+#define   ERR_INT_PIPE_CRC_DONE_B	(1<<5)
 #define   ERR_INT_FIFO_UNDERRUN_B	(1<<3)
+#define   ERR_INT_PIPE_CRC_DONE_A	(1<<2)
+#define   ERR_INT_PIPE_CRC_DONE(pipe)	(1<<(2 + pipe*3))
 #define   ERR_INT_FIFO_UNDERRUN_A	(1<<0)
 #define   ERR_INT_FIFO_UNDERRUN(pipe)	(1<<(pipe*3))
 
@@ -683,6 +749,7 @@
 #define   FPGA_DBG_RM_NOCLAIM	(1<<31)
 
 #define DERRMR		0x44050
+/* Note that HBLANK events are reserved on bdw+ */
 #define   DERRMR_PIPEA_SCANLINE		(1<<0)
 #define   DERRMR_PIPEA_PRI_FLIP_DONE	(1<<1)
 #define   DERRMR_PIPEA_SPR_FLIP_DONE	(1<<2)
@@ -716,6 +783,7 @@
 #define _3D_CHICKEN3	0x02090
 #define  _3D_CHICKEN_SF_DISABLE_OBJEND_CULL		(1 << 10)
 #define  _3D_CHICKEN3_SF_DISABLE_FASTCLIP_CULL		(1 << 5)
+#define  _3D_CHICKEN_SDE_LIMIT_FIFO_POLY_DEPTH(x)	((x)<<1)
 
 #define MI_MODE		0x0209c
 # define VS_TIMER_DISPATCH				(1 << 6)
@@ -890,6 +958,7 @@
 #define GT_BLT_USER_INTERRUPT			(1 << 22)
 #define GT_BSD_CS_ERROR_INTERRUPT		(1 << 15)
 #define GT_BSD_USER_INTERRUPT			(1 << 12)
+#define GT_RENDER_L3_PARITY_ERROR_INTERRUPT_S1	(1 << 11) /* hsw+; rsvd on snb, ivb, vlv */
 #define GT_RENDER_L3_PARITY_ERROR_INTERRUPT	(1 <<  5) /* !snb */
 #define GT_RENDER_PIPECTL_NOTIFY_INTERRUPT	(1 <<  4)
 #define GT_RENDER_CS_MASTER_ERROR_INTERRUPT	(1 <<  3)
@@ -900,6 +969,10 @@
 #define PM_VEBOX_CS_ERROR_INTERRUPT		(1 << 12) /* hsw+ */
 #define PM_VEBOX_USER_INTERRUPT			(1 << 10) /* hsw+ */
 
+#define GT_PARITY_ERROR(dev) \
+	(GT_RENDER_L3_PARITY_ERROR_INTERRUPT | \
+	 (IS_HASWELL(dev) ? GT_RENDER_L3_PARITY_ERROR_INTERRUPT_S1 : 0))
+
 /* These are all the "old" interrupts */
 #define ILK_BSD_USER_INTERRUPT				(1<<5)
 #define I915_PIPE_CONTROL_NOTIFY_INTERRUPT		(1<<18)
@@ -1048,9 +1121,6 @@
 					     _HSW_PIPE_SLICE_CHICKEN_1_A, + \
 					     _HSW_PIPE_SLICE_CHICKEN_1_B)
 
-#define HSW_CLKGATE_DISABLE_PART_1	0x46500
-#define   HSW_DPFC_GATING_DISABLE	(1<<23)
-
 /*
  * GPIO regs
  */
@@ -1387,6 +1457,12 @@
 
 #define MI_ARB_VLV		(VLV_DISPLAY_BASE + 0x6504)
 
+#define CZCLK_CDCLK_FREQ_RATIO	(VLV_DISPLAY_BASE + 0x6508)
+#define   CDCLK_FREQ_SHIFT	4
+#define   CDCLK_FREQ_MASK	(0x1f << CDCLK_FREQ_SHIFT)
+#define   CZCLK_FREQ_MASK	0xf
+#define GMBUSFREQ_VLV		(VLV_DISPLAY_BASE + 0x6510)
+
 /*
  * Palette regs
  */
@@ -1404,13 +1480,15 @@
  * device 0 function 0's pci config register 0x44 or 0x48 and matches it in
  * every way.  It is not accessible from the CP register read instructions.
  *
+ * Starting from Haswell, you can't write registers using the MCHBAR mirror,
+ * just read.
  */
 #define MCHBAR_MIRROR_BASE	0x10000
 
 #define MCHBAR_MIRROR_BASE_SNB	0x140000
 
 /* Memory controller frequency in MCHBAR for Haswell (possible SNB+) */
-#define DCLK 0x5e04
+#define DCLK (MCHBAR_MIRROR_BASE_SNB + 0x5e04)
 
 /** 915-945 and GM965 MCH register controlling DRAM channel access */
 #define DCC			0x10200
@@ -1705,9 +1783,9 @@
 #define GEN6_GT_THREAD_STATUS_CORE_MASK 0x7
 #define GEN6_GT_THREAD_STATUS_CORE_MASK_HSW (0x7 | (0x07 << 16))
 
-#define GEN6_GT_PERF_STATUS	0x145948
-#define GEN6_RP_STATE_LIMITS	0x145994
-#define GEN6_RP_STATE_CAP	0x145998
+#define GEN6_GT_PERF_STATUS	(MCHBAR_MIRROR_BASE_SNB + 0x5948)
+#define GEN6_RP_STATE_LIMITS	(MCHBAR_MIRROR_BASE_SNB + 0x5994)
+#define GEN6_RP_STATE_CAP	(MCHBAR_MIRROR_BASE_SNB + 0x5998)
 
 /*
  * Logical Context regs
@@ -1752,6 +1830,12 @@
  * on HSW) - so the final size is 66944 bytes, which rounds to 17 pages.
  */
 #define HSW_CXT_TOTAL_SIZE		(17 * PAGE_SIZE)
+/* Same as Haswell, but 72064 bytes now. */
+#define GEN8_CXT_TOTAL_SIZE		(18 * PAGE_SIZE)
+
+
+#define VLV_CLK_CTL2			0x101104
+#define   CLK_CTL2_CZCOUNT_30NS_SHIFT	28
 
 /*
  * Overlay regs
@@ -1771,6 +1855,83 @@
  * Display engine regs
  */
 
+/* Pipe A CRC regs */
+#define _PIPE_CRC_CTL_A		(dev_priv->info->display_mmio_offset + 0x60050)
+#define   PIPE_CRC_ENABLE		(1 << 31)
+/* ivb+ source selection */
+#define   PIPE_CRC_SOURCE_PRIMARY_IVB	(0 << 29)
+#define   PIPE_CRC_SOURCE_SPRITE_IVB	(1 << 29)
+#define   PIPE_CRC_SOURCE_PF_IVB	(2 << 29)
+/* ilk+ source selection */
+#define   PIPE_CRC_SOURCE_PRIMARY_ILK	(0 << 28)
+#define   PIPE_CRC_SOURCE_SPRITE_ILK	(1 << 28)
+#define   PIPE_CRC_SOURCE_PIPE_ILK	(2 << 28)
+/* embedded DP port on the north display block, reserved on ivb */
+#define   PIPE_CRC_SOURCE_PORT_A_ILK	(4 << 28)
+#define   PIPE_CRC_SOURCE_FDI_ILK	(5 << 28) /* reserved on ivb */
+/* vlv source selection */
+#define   PIPE_CRC_SOURCE_PIPE_VLV	(0 << 27)
+#define   PIPE_CRC_SOURCE_HDMIB_VLV	(1 << 27)
+#define   PIPE_CRC_SOURCE_HDMIC_VLV	(2 << 27)
+/* with DP port the pipe source is invalid */
+#define   PIPE_CRC_SOURCE_DP_D_VLV	(3 << 27)
+#define   PIPE_CRC_SOURCE_DP_B_VLV	(6 << 27)
+#define   PIPE_CRC_SOURCE_DP_C_VLV	(7 << 27)
+/* gen3+ source selection */
+#define   PIPE_CRC_SOURCE_PIPE_I9XX	(0 << 28)
+#define   PIPE_CRC_SOURCE_SDVOB_I9XX	(1 << 28)
+#define   PIPE_CRC_SOURCE_SDVOC_I9XX	(2 << 28)
+/* with DP/TV port the pipe source is invalid */
+#define   PIPE_CRC_SOURCE_DP_D_G4X	(3 << 28)
+#define   PIPE_CRC_SOURCE_TV_PRE	(4 << 28)
+#define   PIPE_CRC_SOURCE_TV_POST	(5 << 28)
+#define   PIPE_CRC_SOURCE_DP_B_G4X	(6 << 28)
+#define   PIPE_CRC_SOURCE_DP_C_G4X	(7 << 28)
+/* gen2 doesn't have source selection bits */
+#define   PIPE_CRC_INCLUDE_BORDER_I8XX	(1 << 30)
+
+#define _PIPE_CRC_RES_1_A_IVB		0x60064
+#define _PIPE_CRC_RES_2_A_IVB		0x60068
+#define _PIPE_CRC_RES_3_A_IVB		0x6006c
+#define _PIPE_CRC_RES_4_A_IVB		0x60070
+#define _PIPE_CRC_RES_5_A_IVB		0x60074
+
+#define _PIPE_CRC_RES_RED_A		(dev_priv->info->display_mmio_offset + 0x60060)
+#define _PIPE_CRC_RES_GREEN_A		(dev_priv->info->display_mmio_offset + 0x60064)
+#define _PIPE_CRC_RES_BLUE_A		(dev_priv->info->display_mmio_offset + 0x60068)
+#define _PIPE_CRC_RES_RES1_A_I915	(dev_priv->info->display_mmio_offset + 0x6006c)
+#define _PIPE_CRC_RES_RES2_A_G4X	(dev_priv->info->display_mmio_offset + 0x60080)
+
+/* Pipe B CRC regs */
+#define _PIPE_CRC_RES_1_B_IVB		0x61064
+#define _PIPE_CRC_RES_2_B_IVB		0x61068
+#define _PIPE_CRC_RES_3_B_IVB		0x6106c
+#define _PIPE_CRC_RES_4_B_IVB		0x61070
+#define _PIPE_CRC_RES_5_B_IVB		0x61074
+
+#define PIPE_CRC_CTL(pipe)	_PIPE_INC(pipe, _PIPE_CRC_CTL_A, 0x01000)
+#define PIPE_CRC_RES_1_IVB(pipe)	\
+	_PIPE(pipe, _PIPE_CRC_RES_1_A_IVB, _PIPE_CRC_RES_1_B_IVB)
+#define PIPE_CRC_RES_2_IVB(pipe)	\
+	_PIPE(pipe, _PIPE_CRC_RES_2_A_IVB, _PIPE_CRC_RES_2_B_IVB)
+#define PIPE_CRC_RES_3_IVB(pipe)	\
+	_PIPE(pipe, _PIPE_CRC_RES_3_A_IVB, _PIPE_CRC_RES_3_B_IVB)
+#define PIPE_CRC_RES_4_IVB(pipe)	\
+	_PIPE(pipe, _PIPE_CRC_RES_4_A_IVB, _PIPE_CRC_RES_4_B_IVB)
+#define PIPE_CRC_RES_5_IVB(pipe)	\
+	_PIPE(pipe, _PIPE_CRC_RES_5_A_IVB, _PIPE_CRC_RES_5_B_IVB)
+
+#define PIPE_CRC_RES_RED(pipe) \
+	_PIPE_INC(pipe, _PIPE_CRC_RES_RED_A, 0x01000)
+#define PIPE_CRC_RES_GREEN(pipe) \
+	_PIPE_INC(pipe, _PIPE_CRC_RES_GREEN_A, 0x01000)
+#define PIPE_CRC_RES_BLUE(pipe) \
+	_PIPE_INC(pipe, _PIPE_CRC_RES_BLUE_A, 0x01000)
+#define PIPE_CRC_RES_RES1_I915(pipe) \
+	_PIPE_INC(pipe, _PIPE_CRC_RES_RES1_A_I915, 0x01000)
+#define PIPE_CRC_RES_RES2_G4X(pipe) \
+	_PIPE_INC(pipe, _PIPE_CRC_RES_RES2_A_G4X, 0x01000)
+
 /* Pipe A timing regs */
 #define _HTOTAL_A	(dev_priv->info->display_mmio_offset + 0x60000)
 #define _HBLANK_A	(dev_priv->info->display_mmio_offset + 0x60004)
@@ -1793,7 +1954,6 @@
 #define _BCLRPAT_B	(dev_priv->info->display_mmio_offset + 0x61020)
 #define _VSYNCSHIFT_B	(dev_priv->info->display_mmio_offset + 0x61028)
 
-
 #define HTOTAL(trans) _TRANSCODER(trans, _HTOTAL_A, _HTOTAL_B)
 #define HBLANK(trans) _TRANSCODER(trans, _HBLANK_A, _HBLANK_B)
 #define HSYNC(trans) _TRANSCODER(trans, _HSYNC_A, _HSYNC_B)
@@ -1803,8 +1963,9 @@
 #define BCLRPAT(pipe) _PIPE(pipe, _BCLRPAT_A, _BCLRPAT_B)
 #define VSYNCSHIFT(trans) _TRANSCODER(trans, _VSYNCSHIFT_A, _VSYNCSHIFT_B)
 
-/* HSW eDP PSR registers */
-#define EDP_PSR_CTL				0x64800
+/* HSW+ eDP PSR registers */
+#define EDP_PSR_BASE(dev)                       (IS_HASWELL(dev) ? 0x64800 : 0x6f800)
+#define EDP_PSR_CTL(dev)			(EDP_PSR_BASE(dev) + 0)
 #define   EDP_PSR_ENABLE			(1<<31)
 #define   EDP_PSR_LINK_DISABLE			(0<<27)
 #define   EDP_PSR_LINK_STANDBY			(1<<27)
@@ -1827,16 +1988,16 @@
 #define   EDP_PSR_TP1_TIME_0us			(3<<4)
 #define   EDP_PSR_IDLE_FRAME_SHIFT		0
 
-#define EDP_PSR_AUX_CTL			0x64810
-#define EDP_PSR_AUX_DATA1		0x64814
+#define EDP_PSR_AUX_CTL(dev)			(EDP_PSR_BASE(dev) + 0x10)
+#define EDP_PSR_AUX_DATA1(dev)			(EDP_PSR_BASE(dev) + 0x14)
 #define   EDP_PSR_DPCD_COMMAND		0x80060000
-#define EDP_PSR_AUX_DATA2		0x64818
+#define EDP_PSR_AUX_DATA2(dev)			(EDP_PSR_BASE(dev) + 0x18)
 #define   EDP_PSR_DPCD_NORMAL_OPERATION	(1<<24)
-#define EDP_PSR_AUX_DATA3		0x6481c
-#define EDP_PSR_AUX_DATA4		0x64820
-#define EDP_PSR_AUX_DATA5		0x64824
+#define EDP_PSR_AUX_DATA3(dev)			(EDP_PSR_BASE(dev) + 0x1c)
+#define EDP_PSR_AUX_DATA4(dev)			(EDP_PSR_BASE(dev) + 0x20)
+#define EDP_PSR_AUX_DATA5(dev)			(EDP_PSR_BASE(dev) + 0x24)
 
-#define EDP_PSR_STATUS_CTL			0x64840
+#define EDP_PSR_STATUS_CTL(dev)			(EDP_PSR_BASE(dev) + 0x40)
 #define   EDP_PSR_STATUS_STATE_MASK		(7<<29)
 #define   EDP_PSR_STATUS_STATE_IDLE		(0<<29)
 #define   EDP_PSR_STATUS_STATE_SRDONACK		(1<<29)
@@ -1860,10 +2021,10 @@
 #define   EDP_PSR_STATUS_SENDING_TP1		(1<<4)
 #define   EDP_PSR_STATUS_IDLE_MASK		0xf
 
-#define EDP_PSR_PERF_CNT		0x64844
+#define EDP_PSR_PERF_CNT(dev)		(EDP_PSR_BASE(dev) + 0x44)
 #define   EDP_PSR_PERF_CNT_MASK		0xffffff
 
-#define EDP_PSR_DEBUG_CTL		0x64860
+#define EDP_PSR_DEBUG_CTL(dev)		(EDP_PSR_BASE(dev) + 0x60)
 #define   EDP_PSR_DEBUG_MASK_LPSP	(1<<27)
 #define   EDP_PSR_DEBUG_MASK_MEMUP	(1<<26)
 #define   EDP_PSR_DEBUG_MASK_HPD	(1<<25)
@@ -2006,6 +2167,14 @@
 #define PCH_HDMIC	0xe1150
 #define PCH_HDMID	0xe1160
 
+#define PORT_DFT_I9XX				0x61150
+#define   DC_BALANCE_RESET			(1 << 25)
+#define PORT_DFT2_G4X				0x61154
+#define   DC_BALANCE_RESET_VLV			(1 << 31)
+#define   PIPE_SCRAMBLE_RESET_MASK		(0x3 << 0)
+#define   PIPE_B_SCRAMBLE_RESET			(1 << 1)
+#define   PIPE_A_SCRAMBLE_RESET			(1 << 0)
+
 /* Gen 3 SDVO bits: */
 #define   SDVO_ENABLE				(1 << 31)
 #define   SDVO_PIPE_SEL(pipe)			((pipe) << 30)
@@ -2034,6 +2203,7 @@
 
 /* Gen 4 SDVO/HDMI bits: */
 #define   SDVO_COLOR_FORMAT_8bpc		(0 << 26)
+#define   SDVO_COLOR_FORMAT_MASK		(7 << 26)
 #define   SDVO_ENCODING_SDVO			(0 << 10)
 #define   SDVO_ENCODING_HDMI			(2 << 10)
 #define   HDMI_MODE_SELECT_HDMI			(1 << 9) /* HDMI only */
@@ -2238,6 +2408,21 @@
 
 #define PFIT_AUTO_RATIOS (dev_priv->info->display_mmio_offset + 0x61238)
 
+#define _VLV_BLC_PWM_CTL2_A (dev_priv->info->display_mmio_offset + 0x61250)
+#define _VLV_BLC_PWM_CTL2_B (dev_priv->info->display_mmio_offset + 0x61350)
+#define VLV_BLC_PWM_CTL2(pipe) _PIPE(pipe, _VLV_BLC_PWM_CTL2_A, \
+				     _VLV_BLC_PWM_CTL2_B)
+
+#define _VLV_BLC_PWM_CTL_A (dev_priv->info->display_mmio_offset + 0x61254)
+#define _VLV_BLC_PWM_CTL_B (dev_priv->info->display_mmio_offset + 0x61354)
+#define VLV_BLC_PWM_CTL(pipe) _PIPE(pipe, _VLV_BLC_PWM_CTL_A, \
+				    _VLV_BLC_PWM_CTL_B)
+
+#define _VLV_BLC_HIST_CTL_A (dev_priv->info->display_mmio_offset + 0x61260)
+#define _VLV_BLC_HIST_CTL_B (dev_priv->info->display_mmio_offset + 0x61360)
+#define VLV_BLC_HIST_CTL(pipe) _PIPE(pipe, _VLV_BLC_HIST_CTL_A, \
+				     _VLV_BLC_HIST_CTL_B)
+
 /* Backlight control */
 #define BLC_PWM_CTL2	(dev_priv->info->display_mmio_offset + 0x61250) /* 965+ only */
 #define   BLM_PWM_ENABLE		(1 << 31)
@@ -2986,6 +3171,7 @@
 #define   PIPECONF_DISABLE	0
 #define   PIPECONF_DOUBLE_WIDE	(1<<30)
 #define   I965_PIPECONF_ACTIVE	(1<<30)
+#define   PIPECONF_DSI_PLL_LOCKED	(1<<29) /* vlv & pipe A only */
 #define   PIPECONF_FRAME_START_DELAY_MASK (3<<27)
 #define   PIPECONF_SINGLE_WIDE	0
 #define   PIPECONF_PIPE_UNLOCKED 0
@@ -3068,6 +3254,18 @@
 #define PIPEFRAMEPIXEL(pipe)  _PIPE(pipe, _PIPEAFRAMEPIXEL, _PIPEBFRAMEPIXEL)
 #define PIPESTAT(pipe) _PIPE(pipe, _PIPEASTAT, _PIPEBSTAT)
 
+#define _PIPE_MISC_A			0x70030
+#define _PIPE_MISC_B			0x71030
+#define   PIPEMISC_DITHER_BPC_MASK	(7<<5)
+#define   PIPEMISC_DITHER_8_BPC		(0<<5)
+#define   PIPEMISC_DITHER_10_BPC	(1<<5)
+#define   PIPEMISC_DITHER_6_BPC		(2<<5)
+#define   PIPEMISC_DITHER_12_BPC	(3<<5)
+#define   PIPEMISC_DITHER_ENABLE	(1<<4)
+#define   PIPEMISC_DITHER_TYPE_MASK	(3<<2)
+#define   PIPEMISC_DITHER_TYPE_SP	(0<<2)
+#define PIPEMISC(pipe) _PIPE(pipe, _PIPE_MISC_A, _PIPE_MISC_B)
+
 #define VLV_DPFLIPSTAT				(VLV_DISPLAY_BASE + 0x70028)
 #define   PIPEB_LINE_COMPARE_INT_EN		(1<<29)
 #define   PIPEB_HLINE_INT_EN			(1<<28)
@@ -3184,11 +3382,11 @@
 
 /* define the Watermark register on Ironlake */
 #define WM0_PIPEA_ILK		0x45100
-#define  WM0_PIPE_PLANE_MASK	(0x7f<<16)
+#define  WM0_PIPE_PLANE_MASK	(0xffff<<16)
 #define  WM0_PIPE_PLANE_SHIFT	16
-#define  WM0_PIPE_SPRITE_MASK	(0x3f<<8)
+#define  WM0_PIPE_SPRITE_MASK	(0xff<<8)
 #define  WM0_PIPE_SPRITE_SHIFT	8
-#define  WM0_PIPE_CURSOR_MASK	(0x1f)
+#define  WM0_PIPE_CURSOR_MASK	(0xff)
 
 #define WM0_PIPEB_ILK		0x45104
 #define WM0_PIPEC_IVB		0x45200
@@ -3198,9 +3396,10 @@
 #define  WM1_LP_LATENCY_MASK	(0x7f<<24)
 #define  WM1_LP_FBC_MASK	(0xf<<20)
 #define  WM1_LP_FBC_SHIFT	20
-#define  WM1_LP_SR_MASK		(0x1ff<<8)
+#define  WM1_LP_FBC_SHIFT_BDW	19
+#define  WM1_LP_SR_MASK		(0x7ff<<8)
 #define  WM1_LP_SR_SHIFT	8
-#define  WM1_LP_CURSOR_MASK	(0x3f)
+#define  WM1_LP_CURSOR_MASK	(0xff)
 #define WM2_LP_ILK		0x4510c
 #define  WM2_LP_EN		(1<<31)
 #define WM3_LP_ILK		0x45110
@@ -3281,17 +3480,17 @@
  *  } while (high1 != high2);
  *  frame = (high1 << 8) | low1;
  */
-#define _PIPEAFRAMEHIGH          (dev_priv->info->display_mmio_offset + 0x70040)
+#define _PIPEAFRAMEHIGH          0x70040
 #define   PIPE_FRAME_HIGH_MASK    0x0000ffff
 #define   PIPE_FRAME_HIGH_SHIFT   0
-#define _PIPEAFRAMEPIXEL         (dev_priv->info->display_mmio_offset + 0x70044)
+#define _PIPEAFRAMEPIXEL         0x70044
 #define   PIPE_FRAME_LOW_MASK     0xff000000
 #define   PIPE_FRAME_LOW_SHIFT    24
 #define   PIPE_PIXEL_MASK         0x00ffffff
 #define   PIPE_PIXEL_SHIFT        0
 /* GM45+ just has to be different */
-#define _PIPEA_FRMCOUNT_GM45	0x70040
-#define _PIPEA_FLIPCOUNT_GM45	0x70044
+#define _PIPEA_FRMCOUNT_GM45	(dev_priv->info->display_mmio_offset + 0x70040)
+#define _PIPEA_FLIPCOUNT_GM45	(dev_priv->info->display_mmio_offset + 0x70044)
 #define PIPE_FRMCOUNT_GM45(pipe) _PIPE(pipe, _PIPEA_FRMCOUNT_GM45, _PIPEB_FRMCOUNT_GM45)
 
 /* Cursor A & B regs */
@@ -3422,10 +3621,10 @@
 #define _PIPEBDSL		(dev_priv->info->display_mmio_offset + 0x71000)
 #define _PIPEBCONF		(dev_priv->info->display_mmio_offset + 0x71008)
 #define _PIPEBSTAT		(dev_priv->info->display_mmio_offset + 0x71024)
-#define _PIPEBFRAMEHIGH		(dev_priv->info->display_mmio_offset + 0x71040)
-#define _PIPEBFRAMEPIXEL	(dev_priv->info->display_mmio_offset + 0x71044)
-#define _PIPEB_FRMCOUNT_GM45	0x71040
-#define _PIPEB_FLIPCOUNT_GM45	0x71044
+#define _PIPEBFRAMEHIGH		0x71040
+#define _PIPEBFRAMEPIXEL	0x71044
+#define _PIPEB_FRMCOUNT_GM45	(dev_priv->info->display_mmio_offset + 0x71040)
+#define _PIPEB_FLIPCOUNT_GM45	(dev_priv->info->display_mmio_offset + 0x71044)
 
 
 /* Display B control */
@@ -3780,6 +3979,7 @@
 #define DE_SPRITEA_FLIP_DONE    (1 << 28)
 #define DE_PLANEB_FLIP_DONE     (1 << 27)
 #define DE_PLANEA_FLIP_DONE     (1 << 26)
+#define DE_PLANE_FLIP_DONE(plane) (1 << (26 + (plane)))
 #define DE_PCU_EVENT            (1 << 25)
 #define DE_GTT_FAULT            (1 << 24)
 #define DE_POISON               (1 << 23)
@@ -3793,13 +3993,18 @@
 #define DE_PIPEB_ODD_FIELD      (1 << 13)
 #define DE_PIPEB_LINE_COMPARE   (1 << 12)
 #define DE_PIPEB_VSYNC          (1 << 11)
+#define DE_PIPEB_CRC_DONE	(1 << 10)
 #define DE_PIPEB_FIFO_UNDERRUN  (1 << 8)
 #define DE_PIPEA_VBLANK         (1 << 7)
+#define DE_PIPE_VBLANK(pipe)    (1 << (7 + 8*(pipe)))
 #define DE_PIPEA_EVEN_FIELD     (1 << 6)
 #define DE_PIPEA_ODD_FIELD      (1 << 5)
 #define DE_PIPEA_LINE_COMPARE   (1 << 4)
 #define DE_PIPEA_VSYNC          (1 << 3)
+#define DE_PIPEA_CRC_DONE	(1 << 2)
+#define DE_PIPE_CRC_DONE(pipe)	(1 << (2 + 8*(pipe)))
 #define DE_PIPEA_FIFO_UNDERRUN  (1 << 0)
+#define DE_PIPE_FIFO_UNDERRUN(pipe)  (1 << (8*(pipe)))
 
 /* More Ivybridge lolz */
 #define DE_ERR_INT_IVB			(1<<30)
@@ -3815,9 +4020,8 @@
 #define DE_PIPEB_VBLANK_IVB		(1<<5)
 #define DE_SPRITEA_FLIP_DONE_IVB	(1<<4)
 #define DE_PLANEA_FLIP_DONE_IVB		(1<<3)
+#define DE_PLANE_FLIP_DONE_IVB(plane)	(1<< (3 + 5*(plane)))
 #define DE_PIPEA_VBLANK_IVB		(1<<0)
-
-#define DE_PIPE_VBLANK_ILK(pipe)	(1 << ((pipe * 8) + 7))
 #define DE_PIPE_VBLANK_IVB(pipe)	(1 << (pipe * 5))
 
 #define VLV_MASTER_IER			0x4400c /* Gunit master IER */
@@ -3833,6 +4037,71 @@
 #define GTIIR   0x44018
 #define GTIER   0x4401c
 
+#define GEN8_MASTER_IRQ			0x44200
+#define  GEN8_MASTER_IRQ_CONTROL	(1<<31)
+#define  GEN8_PCU_IRQ			(1<<30)
+#define  GEN8_DE_PCH_IRQ		(1<<23)
+#define  GEN8_DE_MISC_IRQ		(1<<22)
+#define  GEN8_DE_PORT_IRQ		(1<<20)
+#define  GEN8_DE_PIPE_C_IRQ		(1<<18)
+#define  GEN8_DE_PIPE_B_IRQ		(1<<17)
+#define  GEN8_DE_PIPE_A_IRQ		(1<<16)
+#define  GEN8_DE_PIPE_IRQ(pipe)		(1<<(16+pipe))
+#define  GEN8_GT_VECS_IRQ		(1<<6)
+#define  GEN8_GT_VCS2_IRQ		(1<<3)
+#define  GEN8_GT_VCS1_IRQ		(1<<2)
+#define  GEN8_GT_BCS_IRQ		(1<<1)
+#define  GEN8_GT_RCS_IRQ		(1<<0)
+
+#define GEN8_GT_ISR(which) (0x44300 + (0x10 * (which)))
+#define GEN8_GT_IMR(which) (0x44304 + (0x10 * (which)))
+#define GEN8_GT_IIR(which) (0x44308 + (0x10 * (which)))
+#define GEN8_GT_IER(which) (0x4430c + (0x10 * (which)))
+
+#define GEN8_BCS_IRQ_SHIFT 16
+#define GEN8_RCS_IRQ_SHIFT 0
+#define GEN8_VCS2_IRQ_SHIFT 16
+#define GEN8_VCS1_IRQ_SHIFT 0
+#define GEN8_VECS_IRQ_SHIFT 0
+
+#define GEN8_DE_PIPE_ISR(pipe) (0x44400 + (0x10 * (pipe)))
+#define GEN8_DE_PIPE_IMR(pipe) (0x44404 + (0x10 * (pipe)))
+#define GEN8_DE_PIPE_IIR(pipe) (0x44408 + (0x10 * (pipe)))
+#define GEN8_DE_PIPE_IER(pipe) (0x4440c + (0x10 * (pipe)))
+#define  GEN8_PIPE_FIFO_UNDERRUN	(1 << 31)
+#define  GEN8_PIPE_CDCLK_CRC_ERROR	(1 << 29)
+#define  GEN8_PIPE_CDCLK_CRC_DONE	(1 << 28)
+#define  GEN8_PIPE_CURSOR_FAULT		(1 << 10)
+#define  GEN8_PIPE_SPRITE_FAULT		(1 << 9)
+#define  GEN8_PIPE_PRIMARY_FAULT	(1 << 8)
+#define  GEN8_PIPE_SPRITE_FLIP_DONE	(1 << 5)
+#define  GEN8_PIPE_FLIP_DONE		(1 << 4)
+#define  GEN8_PIPE_SCAN_LINE_EVENT	(1 << 2)
+#define  GEN8_PIPE_VSYNC		(1 << 1)
+#define  GEN8_PIPE_VBLANK		(1 << 0)
+#define GEN8_DE_PIPE_IRQ_FAULT_ERRORS \
+	(GEN8_PIPE_CURSOR_FAULT | \
+	 GEN8_PIPE_SPRITE_FAULT | \
+	 GEN8_PIPE_PRIMARY_FAULT)
+
+#define GEN8_DE_PORT_ISR 0x44440
+#define GEN8_DE_PORT_IMR 0x44444
+#define GEN8_DE_PORT_IIR 0x44448
+#define GEN8_DE_PORT_IER 0x4444c
+#define  GEN8_PORT_DP_A_HOTPLUG		(1 << 3)
+#define  GEN8_AUX_CHANNEL_A		(1 << 0)
+
+#define GEN8_DE_MISC_ISR 0x44460
+#define GEN8_DE_MISC_IMR 0x44464
+#define GEN8_DE_MISC_IIR 0x44468
+#define GEN8_DE_MISC_IER 0x4446c
+#define  GEN8_DE_MISC_GSE		(1 << 27)
+
+#define GEN8_PCU_ISR 0x444e0
+#define GEN8_PCU_IMR 0x444e4
+#define GEN8_PCU_IIR 0x444e8
+#define GEN8_PCU_IER 0x444ec
+
 #define ILK_DISPLAY_CHICKEN2	0x42004
 /* Required on all Ironlake and Sandybridge according to the B-Spec. */
 #define  ILK_ELPIN_409_SELECT	(1 << 25)
@@ -3858,8 +4127,14 @@
 # define CHICKEN3_DGMG_DONE_FIX_DISABLE		(1 << 2)
 
 #define CHICKEN_PAR1_1		0x42080
+#define  DPA_MASK_VBLANK_SRD	(1 << 15)
 #define  FORCE_ARB_IDLE_PLANES	(1 << 14)
 
+#define _CHICKEN_PIPESL_1_A	0x420b0
+#define _CHICKEN_PIPESL_1_B	0x420b4
+#define  DPRS_MASK_VBLANK_SRD	(1 << 0)
+#define CHICKEN_PIPESL_1(pipe) _PIPE(pipe, _CHICKEN_PIPESL_1_A, _CHICKEN_PIPESL_1_B)
+
 #define DISP_ARB_CTL	0x45000
 #define  DISP_TILE_SURFACE_SWIZZLING	(1<<13)
 #define  DISP_FBC_WM_DIS		(1<<15)
@@ -3870,6 +4145,8 @@
 /* GEN7 chicken */
 #define GEN7_COMMON_SLICE_CHICKEN1		0x7010
 # define GEN7_CSC1_RHWO_OPT_DISABLE_IN_RCC	((1<<10) | (1<<26))
+#define COMMON_SLICE_CHICKEN2			0x7014
+# define GEN8_CSC2_SBE_VUE_CACHE_CONSERVATIVE	(1<<0)
 
 #define GEN7_L3CNTLREG1				0xB01C
 #define  GEN7_WA_FOR_GEN7_L3_CONTROL			0x3C4FFF8C
@@ -4416,6 +4693,8 @@
 #define PIPEA_PP_STATUS         (VLV_DISPLAY_BASE + 0x61200)
 #define PIPEA_PP_CONTROL        (VLV_DISPLAY_BASE + 0x61204)
 #define PIPEA_PP_ON_DELAYS      (VLV_DISPLAY_BASE + 0x61208)
+#define  PANEL_PORT_SELECT_DPB_VLV	(1 << 30)
+#define  PANEL_PORT_SELECT_DPC_VLV	(2 << 30)
 #define PIPEA_PP_OFF_DELAYS     (VLV_DISPLAY_BASE + 0x6120c)
 #define PIPEA_PP_DIVISOR        (VLV_DISPLAY_BASE + 0x61210)
 
@@ -4447,7 +4726,6 @@
 #define  PANEL_PORT_SELECT_MASK	(3 << 30)
 #define  PANEL_PORT_SELECT_LVDS	(0 << 30)
 #define  PANEL_PORT_SELECT_DPA	(1 << 30)
-#define  EDP_PANEL		(1 << 30)
 #define  PANEL_PORT_SELECT_DPC	(2 << 30)
 #define  PANEL_PORT_SELECT_DPD	(3 << 30)
 #define  PANEL_POWER_UP_DELAY_MASK	(0x1fff0000)
@@ -4456,11 +4734,6 @@
 #define  PANEL_LIGHT_ON_DELAY_SHIFT	0
 
 #define PCH_PP_OFF_DELAYS	0xc720c
-#define  PANEL_POWER_PORT_SELECT_MASK	(0x3 << 30)
-#define  PANEL_POWER_PORT_LVDS		(0 << 30)
-#define  PANEL_POWER_PORT_DP_A		(1 << 30)
-#define  PANEL_POWER_PORT_DP_C		(2 << 30)
-#define  PANEL_POWER_PORT_DP_D		(3 << 30)
 #define  PANEL_POWER_DOWN_DELAY_MASK	(0x1fff0000)
 #define  PANEL_POWER_DOWN_DELAY_SHIFT	16
 #define  PANEL_LIGHT_OFF_DELAY_MASK	(0x1fff)
@@ -4638,7 +4911,7 @@
 #define   GEN6_RP_UP_IDLE_MIN			(0x1<<3)
 #define   GEN6_RP_UP_BUSY_AVG			(0x2<<3)
 #define   GEN6_RP_UP_BUSY_CONT			(0x4<<3)
-#define   GEN7_RP_DOWN_IDLE_AVG			(0x2<<0)
+#define   GEN6_RP_DOWN_IDLE_AVG			(0x2<<0)
 #define   GEN6_RP_DOWN_IDLE_CONT		(0x1<<0)
 #define GEN6_RP_UP_THRESHOLD			0xA02C
 #define GEN6_RP_DOWN_THRESHOLD			0xA030
@@ -4683,6 +4956,10 @@
 						 GEN6_PM_RP_DOWN_TIMEOUT)
 
 #define GEN6_GT_GFX_RC6_LOCKED			0x138104
+#define VLV_COUNTER_CONTROL			0x138104
+#define   VLV_COUNT_RANGE_HIGH			(1<<15)
+#define   VLV_MEDIA_RC6_COUNT_EN		(1<<1)
+#define   VLV_RENDER_RC6_COUNT_EN		(1<<0)
 #define GEN6_GT_GFX_RC6				0x138108
 #define GEN6_GT_GFX_RC6p			0x13810C
 #define GEN6_GT_GFX_RC6pp			0x138110
@@ -4694,8 +4971,11 @@
 #define   GEN6_PCODE_READ_MIN_FREQ_TABLE	0x9
 #define	  GEN6_PCODE_WRITE_RC6VIDS		0x4
 #define	  GEN6_PCODE_READ_RC6VIDS		0x5
+#define   GEN6_PCODE_READ_D_COMP		0x10
+#define   GEN6_PCODE_WRITE_D_COMP		0x11
 #define   GEN6_ENCODE_RC6_VID(mv)		(((mv) - 245) / 5)
 #define   GEN6_DECODE_RC6_VID(vids)		(((vids) * 5) + 245)
+#define   DISPLAY_IPS_CONTROL			0x19
 #define GEN6_PCODE_DATA				0x138128
 #define   GEN6_PCODE_FREQ_IA_RATIO_SHIFT	8
 #define   GEN6_PCODE_FREQ_RING_RATIO_SHIFT	16
@@ -4713,6 +4993,7 @@
 
 /* IVYBRIDGE DPF */
 #define GEN7_L3CDERRST1			0xB008 /* L3CD Error Status 1 */
+#define HSW_L3CDERRST11			0xB208 /* L3CD Error Status register 1 slice 1 */
 #define   GEN7_L3CDERRST1_ROW_MASK	(0x7ff<<14)
 #define   GEN7_PARITY_ERROR_VALID	(1<<13)
 #define   GEN7_L3CDERRST1_BANK_MASK	(3<<11)
@@ -4726,11 +5007,13 @@
 #define   GEN7_L3CDERRST1_ENABLE	(1<<7)
 
 #define GEN7_L3LOG_BASE			0xB070
+#define HSW_L3LOG_BASE_SLICE1		0xB270
 #define GEN7_L3LOG_SIZE			0x80
 
 #define GEN7_HALF_SLICE_CHICKEN1	0xe100 /* IVB GT1 + VLV */
 #define GEN7_HALF_SLICE_CHICKEN1_GT2	0xf100
 #define   GEN7_MAX_PS_THREAD_DEP		(8<<12)
+#define   GEN7_SINGLE_SUBSCAN_DISPATCH_ENABLE	(1<<10)
 #define   GEN7_PSD_SINGLE_PORT_DISPATCH_ENABLE	(1<<3)
 
 #define GEN7_ROW_CHICKEN2		0xe4f4
@@ -4740,6 +5023,10 @@
 #define HSW_ROW_CHICKEN3		0xe49c
 #define  HSW_ROW_CHICKEN3_L3_GLOBAL_ATOMICS_DISABLE    (1 << 6)
 
+#define HALF_SLICE_CHICKEN3		0xe184
+#define   GEN8_CENTROID_PIXEL_OPT_DIS	(1<<8)
+#define   GEN8_SAMPLER_POWER_BYPASS_DIS	(1<<1)
+
 #define G4X_AUD_VID_DID			(dev_priv->info->display_mmio_offset + 0x62020)
 #define INTEL_AUDIO_DEVCL		0x808629FB
 #define INTEL_AUDIO_DEVBLC		0x80862801
@@ -4781,6 +5068,18 @@
 					CPT_AUD_CNTL_ST_B)
 #define CPT_AUD_CNTRL_ST2		0xE50C0
 
+#define VLV_HDMIW_HDMIEDID_A		(VLV_DISPLAY_BASE + 0x62050)
+#define VLV_HDMIW_HDMIEDID_B		(VLV_DISPLAY_BASE + 0x62150)
+#define VLV_HDMIW_HDMIEDID(pipe) _PIPE(pipe, \
+					VLV_HDMIW_HDMIEDID_A, \
+					VLV_HDMIW_HDMIEDID_B)
+#define VLV_AUD_CNTL_ST_A		(VLV_DISPLAY_BASE + 0x620B4)
+#define VLV_AUD_CNTL_ST_B		(VLV_DISPLAY_BASE + 0x621B4)
+#define VLV_AUD_CNTL_ST(pipe) _PIPE(pipe, \
+					VLV_AUD_CNTL_ST_A, \
+					VLV_AUD_CNTL_ST_B)
+#define VLV_AUD_CNTL_ST2		(VLV_DISPLAY_BASE + 0x620C0)
+
 /* These are the 4 32-bit write offset registers for each stream
  * output buffer.  It determines the offset from the
  * 3DSTATE_SO_BUFFERs that the next streamed vertex output goes to.
@@ -4797,6 +5096,12 @@
 #define CPT_AUD_CFG(pipe) _PIPE(pipe, \
 					CPT_AUD_CONFIG_A, \
 					CPT_AUD_CONFIG_B)
+#define VLV_AUD_CONFIG_A		(VLV_DISPLAY_BASE + 0x62000)
+#define VLV_AUD_CONFIG_B		(VLV_DISPLAY_BASE + 0x62100)
+#define VLV_AUD_CFG(pipe) _PIPE(pipe, \
+					VLV_AUD_CONFIG_A, \
+					VLV_AUD_CONFIG_B)
+
 #define   AUD_CONFIG_N_VALUE_INDEX		(1 << 29)
 #define   AUD_CONFIG_N_PROG_ENABLE		(1 << 28)
 #define   AUD_CONFIG_UPPER_N_SHIFT		20
@@ -4804,7 +5109,17 @@
 #define   AUD_CONFIG_LOWER_N_SHIFT		4
 #define   AUD_CONFIG_LOWER_N_VALUE		(0xfff << 4)
 #define   AUD_CONFIG_PIXEL_CLOCK_HDMI_SHIFT	16
-#define   AUD_CONFIG_PIXEL_CLOCK_HDMI		(0xf << 16)
+#define   AUD_CONFIG_PIXEL_CLOCK_HDMI_MASK	(0xf << 16)
+#define   AUD_CONFIG_PIXEL_CLOCK_HDMI_25175	(0 << 16)
+#define   AUD_CONFIG_PIXEL_CLOCK_HDMI_25200	(1 << 16)
+#define   AUD_CONFIG_PIXEL_CLOCK_HDMI_27000	(2 << 16)
+#define   AUD_CONFIG_PIXEL_CLOCK_HDMI_27027	(3 << 16)
+#define   AUD_CONFIG_PIXEL_CLOCK_HDMI_54000	(4 << 16)
+#define   AUD_CONFIG_PIXEL_CLOCK_HDMI_54054	(5 << 16)
+#define   AUD_CONFIG_PIXEL_CLOCK_HDMI_74176	(6 << 16)
+#define   AUD_CONFIG_PIXEL_CLOCK_HDMI_74250	(7 << 16)
+#define   AUD_CONFIG_PIXEL_CLOCK_HDMI_148352	(8 << 16)
+#define   AUD_CONFIG_PIXEL_CLOCK_HDMI_148500	(9 << 16)
 #define   AUD_CONFIG_DISABLE_NCTS		(1 << 3)
 
 /* HSW Audio */
@@ -4929,6 +5244,7 @@
 #define DDI_BUF_CTL_B				0x64100
 #define DDI_BUF_CTL(port) _PORT(port, DDI_BUF_CTL_A, DDI_BUF_CTL_B)
 #define  DDI_BUF_CTL_ENABLE			(1<<31)
+/* Haswell */
 #define  DDI_BUF_EMP_400MV_0DB_HSW		(0<<24)   /* Sel0 */
 #define  DDI_BUF_EMP_400MV_3_5DB_HSW		(1<<24)   /* Sel1 */
 #define  DDI_BUF_EMP_400MV_6DB_HSW		(2<<24)   /* Sel2 */
@@ -4938,6 +5254,16 @@
 #define  DDI_BUF_EMP_600MV_6DB_HSW		(6<<24)   /* Sel6 */
 #define  DDI_BUF_EMP_800MV_0DB_HSW		(7<<24)   /* Sel7 */
 #define  DDI_BUF_EMP_800MV_3_5DB_HSW		(8<<24)   /* Sel8 */
+/* Broadwell */
+#define  DDI_BUF_EMP_400MV_0DB_BDW		(0<<24)   /* Sel0 */
+#define  DDI_BUF_EMP_400MV_3_5DB_BDW		(1<<24)   /* Sel1 */
+#define  DDI_BUF_EMP_400MV_6DB_BDW		(2<<24)   /* Sel2 */
+#define  DDI_BUF_EMP_600MV_0DB_BDW		(3<<24)   /* Sel3 */
+#define  DDI_BUF_EMP_600MV_3_5DB_BDW		(4<<24)   /* Sel4 */
+#define  DDI_BUF_EMP_600MV_6DB_BDW		(5<<24)   /* Sel5 */
+#define  DDI_BUF_EMP_800MV_0DB_BDW		(6<<24)   /* Sel6 */
+#define  DDI_BUF_EMP_800MV_3_5DB_BDW		(7<<24)   /* Sel7 */
+#define  DDI_BUF_EMP_1200MV_0DB_BDW		(8<<24)   /* Sel8 */
 #define  DDI_BUF_EMP_MASK			(0xf<<24)
 #define  DDI_BUF_PORT_REVERSAL			(1<<16)
 #define  DDI_BUF_IS_IDLE			(1<<7)
@@ -5047,6 +5373,9 @@
 #define  LCPLL_PLL_LOCK			(1<<30)
 #define  LCPLL_CLK_FREQ_MASK		(3<<26)
 #define  LCPLL_CLK_FREQ_450		(0<<26)
+#define  LCPLL_CLK_FREQ_54O_BDW		(1<<26)
+#define  LCPLL_CLK_FREQ_337_5_BDW	(2<<26)
+#define  LCPLL_CLK_FREQ_675_BDW		(3<<26)
 #define  LCPLL_CD_CLOCK_DISABLE		(1<<25)
 #define  LCPLL_CD2X_CLOCK_DISABLE	(1<<23)
 #define  LCPLL_POWER_DOWN_ALLOW		(1<<22)
@@ -5128,4 +5457,414 @@
 #define PIPE_CSC_POSTOFF_ME(pipe) _PIPE(pipe, _PIPE_A_CSC_POSTOFF_ME, _PIPE_B_CSC_POSTOFF_ME)
 #define PIPE_CSC_POSTOFF_LO(pipe) _PIPE(pipe, _PIPE_A_CSC_POSTOFF_LO, _PIPE_B_CSC_POSTOFF_LO)
 
+/* VLV MIPI registers */
+
+#define _MIPIA_PORT_CTRL			(VLV_DISPLAY_BASE + 0x61190)
+#define _MIPIB_PORT_CTRL			(VLV_DISPLAY_BASE + 0x61700)
+#define MIPI_PORT_CTRL(pipe)		_PIPE(pipe, _MIPIA_PORT_CTRL, _MIPIB_PORT_CTRL)
+#define  DPI_ENABLE					(1 << 31) /* A + B */
+#define  MIPIA_MIPI4DPHY_DELAY_COUNT_SHIFT		27
+#define  MIPIA_MIPI4DPHY_DELAY_COUNT_MASK		(0xf << 27)
+#define  DUAL_LINK_MODE_MASK				(1 << 26)
+#define  DUAL_LINK_MODE_FRONT_BACK			(0 << 26)
+#define  DUAL_LINK_MODE_PIXEL_ALTERNATIVE		(1 << 26)
+#define  DITHERING_ENABLE				(1 << 25) /* A + B */
+#define  FLOPPED_HSTX					(1 << 23)
+#define  DE_INVERT					(1 << 19) /* XXX */
+#define  MIPIA_FLISDSI_DELAY_COUNT_SHIFT		18
+#define  MIPIA_FLISDSI_DELAY_COUNT_MASK			(0xf << 18)
+#define  AFE_LATCHOUT					(1 << 17)
+#define  LP_OUTPUT_HOLD					(1 << 16)
+#define  MIPIB_FLISDSI_DELAY_COUNT_HIGH_SHIFT		15
+#define  MIPIB_FLISDSI_DELAY_COUNT_HIGH_MASK		(1 << 15)
+#define  MIPIB_MIPI4DPHY_DELAY_COUNT_SHIFT		11
+#define  MIPIB_MIPI4DPHY_DELAY_COUNT_MASK		(0xf << 11)
+#define  CSB_SHIFT					9
+#define  CSB_MASK					(3 << 9)
+#define  CSB_20MHZ					(0 << 9)
+#define  CSB_10MHZ					(1 << 9)
+#define  CSB_40MHZ					(2 << 9)
+#define  BANDGAP_MASK					(1 << 8)
+#define  BANDGAP_PNW_CIRCUIT				(0 << 8)
+#define  BANDGAP_LNC_CIRCUIT				(1 << 8)
+#define  MIPIB_FLISDSI_DELAY_COUNT_LOW_SHIFT		5
+#define  MIPIB_FLISDSI_DELAY_COUNT_LOW_MASK		(7 << 5)
+#define  TEARING_EFFECT_DELAY				(1 << 4) /* A + B */
+#define  TEARING_EFFECT_SHIFT				2 /* A + B */
+#define  TEARING_EFFECT_MASK				(3 << 2)
+#define  TEARING_EFFECT_OFF				(0 << 2)
+#define  TEARING_EFFECT_DSI				(1 << 2)
+#define  TEARING_EFFECT_GPIO				(2 << 2)
+#define  LANE_CONFIGURATION_SHIFT			0
+#define  LANE_CONFIGURATION_MASK			(3 << 0)
+#define  LANE_CONFIGURATION_4LANE			(0 << 0)
+#define  LANE_CONFIGURATION_DUAL_LINK_A			(1 << 0)
+#define  LANE_CONFIGURATION_DUAL_LINK_B			(2 << 0)
+
+#define _MIPIA_TEARING_CTRL			(VLV_DISPLAY_BASE + 0x61194)
+#define _MIPIB_TEARING_CTRL			(VLV_DISPLAY_BASE + 0x61704)
+#define MIPI_TEARING_CTRL(pipe)		_PIPE(pipe, _MIPIA_TEARING_CTRL, _MIPIB_TEARING_CTRL)
+#define  TEARING_EFFECT_DELAY_SHIFT			0
+#define  TEARING_EFFECT_DELAY_MASK			(0xffff << 0)
+
+/* XXX: all bits reserved */
+#define _MIPIA_AUTOPWG				(VLV_DISPLAY_BASE + 0x611a0)
+
+/* MIPI DSI Controller and D-PHY registers */
+
+#define _MIPIA_DEVICE_READY			(VLV_DISPLAY_BASE + 0xb000)
+#define _MIPIB_DEVICE_READY			(VLV_DISPLAY_BASE + 0xb800)
+#define MIPI_DEVICE_READY(pipe)		_PIPE(pipe, _MIPIA_DEVICE_READY, _MIPIB_DEVICE_READY)
+#define  BUS_POSSESSION					(1 << 3) /* set to give bus to receiver */
+#define  ULPS_STATE_MASK				(3 << 1)
+#define  ULPS_STATE_ENTER				(2 << 1)
+#define  ULPS_STATE_EXIT				(1 << 1)
+#define  ULPS_STATE_NORMAL_OPERATION			(0 << 1)
+#define  DEVICE_READY					(1 << 0)
+
+#define _MIPIA_INTR_STAT			(VLV_DISPLAY_BASE + 0xb004)
+#define _MIPIB_INTR_STAT			(VLV_DISPLAY_BASE + 0xb804)
+#define MIPI_INTR_STAT(pipe)		_PIPE(pipe, _MIPIA_INTR_STAT, _MIPIB_INTR_STAT)
+#define _MIPIA_INTR_EN				(VLV_DISPLAY_BASE + 0xb008)
+#define _MIPIB_INTR_EN				(VLV_DISPLAY_BASE + 0xb808)
+#define MIPI_INTR_EN(pipe)		_PIPE(pipe, _MIPIA_INTR_EN, _MIPIB_INTR_EN)
+#define  TEARING_EFFECT					(1 << 31)
+#define  SPL_PKT_SENT_INTERRUPT				(1 << 30)
+#define  GEN_READ_DATA_AVAIL				(1 << 29)
+#define  LP_GENERIC_WR_FIFO_FULL			(1 << 28)
+#define  HS_GENERIC_WR_FIFO_FULL			(1 << 27)
+#define  RX_PROT_VIOLATION				(1 << 26)
+#define  RX_INVALID_TX_LENGTH				(1 << 25)
+#define  ACK_WITH_NO_ERROR				(1 << 24)
+#define  TURN_AROUND_ACK_TIMEOUT			(1 << 23)
+#define  LP_RX_TIMEOUT					(1 << 22)
+#define  HS_TX_TIMEOUT					(1 << 21)
+#define  DPI_FIFO_UNDERRUN				(1 << 20)
+#define  LOW_CONTENTION					(1 << 19)
+#define  HIGH_CONTENTION				(1 << 18)
+#define  TXDSI_VC_ID_INVALID				(1 << 17)
+#define  TXDSI_DATA_TYPE_NOT_RECOGNISED			(1 << 16)
+#define  TXCHECKSUM_ERROR				(1 << 15)
+#define  TXECC_MULTIBIT_ERROR				(1 << 14)
+#define  TXECC_SINGLE_BIT_ERROR				(1 << 13)
+#define  TXFALSE_CONTROL_ERROR				(1 << 12)
+#define  RXDSI_VC_ID_INVALID				(1 << 11)
+#define  RXDSI_DATA_TYPE_NOT_REGOGNISED			(1 << 10)
+#define  RXCHECKSUM_ERROR				(1 << 9)
+#define  RXECC_MULTIBIT_ERROR				(1 << 8)
+#define  RXECC_SINGLE_BIT_ERROR				(1 << 7)
+#define  RXFALSE_CONTROL_ERROR				(1 << 6)
+#define  RXHS_RECEIVE_TIMEOUT_ERROR			(1 << 5)
+#define  RX_LP_TX_SYNC_ERROR				(1 << 4)
+#define  RXEXCAPE_MODE_ENTRY_ERROR			(1 << 3)
+#define  RXEOT_SYNC_ERROR				(1 << 2)
+#define  RXSOT_SYNC_ERROR				(1 << 1)
+#define  RXSOT_ERROR					(1 << 0)
+
+#define _MIPIA_DSI_FUNC_PRG			(VLV_DISPLAY_BASE + 0xb00c)
+#define _MIPIB_DSI_FUNC_PRG			(VLV_DISPLAY_BASE + 0xb80c)
+#define MIPI_DSI_FUNC_PRG(pipe)		_PIPE(pipe, _MIPIA_DSI_FUNC_PRG, _MIPIB_DSI_FUNC_PRG)
+#define  CMD_MODE_DATA_WIDTH_MASK			(7 << 13)
+#define  CMD_MODE_NOT_SUPPORTED				(0 << 13)
+#define  CMD_MODE_DATA_WIDTH_16_BIT			(1 << 13)
+#define  CMD_MODE_DATA_WIDTH_9_BIT			(2 << 13)
+#define  CMD_MODE_DATA_WIDTH_8_BIT			(3 << 13)
+#define  CMD_MODE_DATA_WIDTH_OPTION1			(4 << 13)
+#define  CMD_MODE_DATA_WIDTH_OPTION2			(5 << 13)
+#define  VID_MODE_FORMAT_MASK				(0xf << 7)
+#define  VID_MODE_NOT_SUPPORTED				(0 << 7)
+#define  VID_MODE_FORMAT_RGB565				(1 << 7)
+#define  VID_MODE_FORMAT_RGB666				(2 << 7)
+#define  VID_MODE_FORMAT_RGB666_LOOSE			(3 << 7)
+#define  VID_MODE_FORMAT_RGB888				(4 << 7)
+#define  CMD_MODE_CHANNEL_NUMBER_SHIFT			5
+#define  CMD_MODE_CHANNEL_NUMBER_MASK			(3 << 5)
+#define  VID_MODE_CHANNEL_NUMBER_SHIFT			3
+#define  VID_MODE_CHANNEL_NUMBER_MASK			(3 << 3)
+#define  DATA_LANES_PRG_REG_SHIFT			0
+#define  DATA_LANES_PRG_REG_MASK			(7 << 0)
+
+#define _MIPIA_HS_TX_TIMEOUT			(VLV_DISPLAY_BASE + 0xb010)
+#define _MIPIB_HS_TX_TIMEOUT			(VLV_DISPLAY_BASE + 0xb810)
+#define MIPI_HS_TX_TIMEOUT(pipe)	_PIPE(pipe, _MIPIA_HS_TX_TIMEOUT, _MIPIB_HS_TX_TIMEOUT)
+#define  HIGH_SPEED_TX_TIMEOUT_COUNTER_MASK		0xffffff
+
+#define _MIPIA_LP_RX_TIMEOUT			(VLV_DISPLAY_BASE + 0xb014)
+#define _MIPIB_LP_RX_TIMEOUT			(VLV_DISPLAY_BASE + 0xb814)
+#define MIPI_LP_RX_TIMEOUT(pipe)	_PIPE(pipe, _MIPIA_LP_RX_TIMEOUT, _MIPIB_LP_RX_TIMEOUT)
+#define  LOW_POWER_RX_TIMEOUT_COUNTER_MASK		0xffffff
+
+#define _MIPIA_TURN_AROUND_TIMEOUT		(VLV_DISPLAY_BASE + 0xb018)
+#define _MIPIB_TURN_AROUND_TIMEOUT		(VLV_DISPLAY_BASE + 0xb818)
+#define MIPI_TURN_AROUND_TIMEOUT(pipe)	_PIPE(pipe, _MIPIA_TURN_AROUND_TIMEOUT, _MIPIB_TURN_AROUND_TIMEOUT)
+#define  TURN_AROUND_TIMEOUT_MASK			0x3f
+
+#define _MIPIA_DEVICE_RESET_TIMER		(VLV_DISPLAY_BASE + 0xb01c)
+#define _MIPIB_DEVICE_RESET_TIMER		(VLV_DISPLAY_BASE + 0xb81c)
+#define MIPI_DEVICE_RESET_TIMER(pipe)	_PIPE(pipe, _MIPIA_DEVICE_RESET_TIMER, _MIPIB_DEVICE_RESET_TIMER)
+#define  DEVICE_RESET_TIMER_MASK			0xffff
+
+#define _MIPIA_DPI_RESOLUTION			(VLV_DISPLAY_BASE + 0xb020)
+#define _MIPIB_DPI_RESOLUTION			(VLV_DISPLAY_BASE + 0xb820)
+#define MIPI_DPI_RESOLUTION(pipe)	_PIPE(pipe, _MIPIA_DPI_RESOLUTION, _MIPIB_DPI_RESOLUTION)
+#define  VERTICAL_ADDRESS_SHIFT				16
+#define  VERTICAL_ADDRESS_MASK				(0xffff << 16)
+#define  HORIZONTAL_ADDRESS_SHIFT			0
+#define  HORIZONTAL_ADDRESS_MASK			0xffff
+
+#define _MIPIA_DBI_FIFO_THROTTLE		(VLV_DISPLAY_BASE + 0xb024)
+#define _MIPIB_DBI_FIFO_THROTTLE		(VLV_DISPLAY_BASE + 0xb824)
+#define MIPI_DBI_FIFO_THROTTLE(pipe)	_PIPE(pipe, _MIPIA_DBI_FIFO_THROTTLE, _MIPIB_DBI_FIFO_THROTTLE)
+#define  DBI_FIFO_EMPTY_HALF				(0 << 0)
+#define  DBI_FIFO_EMPTY_QUARTER				(1 << 0)
+#define  DBI_FIFO_EMPTY_7_LOCATIONS			(2 << 0)
+
+/* regs below are bits 15:0 */
+#define _MIPIA_HSYNC_PADDING_COUNT		(VLV_DISPLAY_BASE + 0xb028)
+#define _MIPIB_HSYNC_PADDING_COUNT		(VLV_DISPLAY_BASE + 0xb828)
+#define MIPI_HSYNC_PADDING_COUNT(pipe)	_PIPE(pipe, _MIPIA_HSYNC_PADDING_COUNT, _MIPIB_HSYNC_PADDING_COUNT)
+
+#define _MIPIA_HBP_COUNT			(VLV_DISPLAY_BASE + 0xb02c)
+#define _MIPIB_HBP_COUNT			(VLV_DISPLAY_BASE + 0xb82c)
+#define MIPI_HBP_COUNT(pipe)		_PIPE(pipe, _MIPIA_HBP_COUNT, _MIPIB_HBP_COUNT)
+
+#define _MIPIA_HFP_COUNT			(VLV_DISPLAY_BASE + 0xb030)
+#define _MIPIB_HFP_COUNT			(VLV_DISPLAY_BASE + 0xb830)
+#define MIPI_HFP_COUNT(pipe)		_PIPE(pipe, _MIPIA_HFP_COUNT, _MIPIB_HFP_COUNT)
+
+#define _MIPIA_HACTIVE_AREA_COUNT		(VLV_DISPLAY_BASE + 0xb034)
+#define _MIPIB_HACTIVE_AREA_COUNT		(VLV_DISPLAY_BASE + 0xb834)
+#define MIPI_HACTIVE_AREA_COUNT(pipe)	_PIPE(pipe, _MIPIA_HACTIVE_AREA_COUNT, _MIPIB_HACTIVE_AREA_COUNT)
+
+#define _MIPIA_VSYNC_PADDING_COUNT		(VLV_DISPLAY_BASE + 0xb038)
+#define _MIPIB_VSYNC_PADDING_COUNT		(VLV_DISPLAY_BASE + 0xb838)
+#define MIPI_VSYNC_PADDING_COUNT(pipe)	_PIPE(pipe, _MIPIA_VSYNC_PADDING_COUNT, _MIPIB_VSYNC_PADDING_COUNT)
+
+#define _MIPIA_VBP_COUNT			(VLV_DISPLAY_BASE + 0xb03c)
+#define _MIPIB_VBP_COUNT			(VLV_DISPLAY_BASE + 0xb83c)
+#define MIPI_VBP_COUNT(pipe)		_PIPE(pipe, _MIPIA_VBP_COUNT, _MIPIB_VBP_COUNT)
+
+#define _MIPIA_VFP_COUNT			(VLV_DISPLAY_BASE + 0xb040)
+#define _MIPIB_VFP_COUNT			(VLV_DISPLAY_BASE + 0xb840)
+#define MIPI_VFP_COUNT(pipe)		_PIPE(pipe, _MIPIA_VFP_COUNT, _MIPIB_VFP_COUNT)
+
+#define _MIPIA_HIGH_LOW_SWITCH_COUNT		(VLV_DISPLAY_BASE + 0xb044)
+#define _MIPIB_HIGH_LOW_SWITCH_COUNT		(VLV_DISPLAY_BASE + 0xb844)
+#define MIPI_HIGH_LOW_SWITCH_COUNT(pipe)	_PIPE(pipe, _MIPIA_HIGH_LOW_SWITCH_COUNT, _MIPIB_HIGH_LOW_SWITCH_COUNT)
+/* regs above are bits 15:0 */
+
+#define _MIPIA_DPI_CONTROL			(VLV_DISPLAY_BASE + 0xb048)
+#define _MIPIB_DPI_CONTROL			(VLV_DISPLAY_BASE + 0xb848)
+#define MIPI_DPI_CONTROL(pipe)		_PIPE(pipe, _MIPIA_DPI_CONTROL, _MIPIB_DPI_CONTROL)
+#define  DPI_LP_MODE					(1 << 6)
+#define  BACKLIGHT_OFF					(1 << 5)
+#define  BACKLIGHT_ON					(1 << 4)
+#define  COLOR_MODE_OFF					(1 << 3)
+#define  COLOR_MODE_ON					(1 << 2)
+#define  TURN_ON					(1 << 1)
+#define  SHUTDOWN					(1 << 0)
+
+#define _MIPIA_DPI_DATA				(VLV_DISPLAY_BASE + 0xb04c)
+#define _MIPIB_DPI_DATA				(VLV_DISPLAY_BASE + 0xb84c)
+#define MIPI_DPI_DATA(pipe)		_PIPE(pipe, _MIPIA_DPI_DATA, _MIPIB_DPI_DATA)
+#define  COMMAND_BYTE_SHIFT				0
+#define  COMMAND_BYTE_MASK				(0x3f << 0)
+
+#define _MIPIA_INIT_COUNT			(VLV_DISPLAY_BASE + 0xb050)
+#define _MIPIB_INIT_COUNT			(VLV_DISPLAY_BASE + 0xb850)
+#define MIPI_INIT_COUNT(pipe)		_PIPE(pipe, _MIPIA_INIT_COUNT, _MIPIB_INIT_COUNT)
+#define  MASTER_INIT_TIMER_SHIFT			0
+#define  MASTER_INIT_TIMER_MASK				(0xffff << 0)
+
+#define _MIPIA_MAX_RETURN_PKT_SIZE		(VLV_DISPLAY_BASE + 0xb054)
+#define _MIPIB_MAX_RETURN_PKT_SIZE		(VLV_DISPLAY_BASE + 0xb854)
+#define MIPI_MAX_RETURN_PKT_SIZE(pipe)	_PIPE(pipe, _MIPIA_MAX_RETURN_PKT_SIZE, _MIPIB_MAX_RETURN_PKT_SIZE)
+#define  MAX_RETURN_PKT_SIZE_SHIFT			0
+#define  MAX_RETURN_PKT_SIZE_MASK			(0x3ff << 0)
+
+#define _MIPIA_VIDEO_MODE_FORMAT		(VLV_DISPLAY_BASE + 0xb058)
+#define _MIPIB_VIDEO_MODE_FORMAT		(VLV_DISPLAY_BASE + 0xb858)
+#define MIPI_VIDEO_MODE_FORMAT(pipe)	_PIPE(pipe, _MIPIA_VIDEO_MODE_FORMAT, _MIPIB_VIDEO_MODE_FORMAT)
+#define  RANDOM_DPI_DISPLAY_RESOLUTION			(1 << 4)
+#define  DISABLE_VIDEO_BTA				(1 << 3)
+#define  IP_TG_CONFIG					(1 << 2)
+#define  VIDEO_MODE_NON_BURST_WITH_SYNC_PULSE		(1 << 0)
+#define  VIDEO_MODE_NON_BURST_WITH_SYNC_EVENTS		(2 << 0)
+#define  VIDEO_MODE_BURST				(3 << 0)
+
+#define _MIPIA_EOT_DISABLE			(VLV_DISPLAY_BASE + 0xb05c)
+#define _MIPIB_EOT_DISABLE			(VLV_DISPLAY_BASE + 0xb85c)
+#define MIPI_EOT_DISABLE(pipe)		_PIPE(pipe, _MIPIA_EOT_DISABLE, _MIPIB_EOT_DISABLE)
+#define  LP_RX_TIMEOUT_ERROR_RECOVERY_DISABLE		(1 << 7)
+#define  HS_RX_TIMEOUT_ERROR_RECOVERY_DISABLE		(1 << 6)
+#define  LOW_CONTENTION_RECOVERY_DISABLE		(1 << 5)
+#define  HIGH_CONTENTION_RECOVERY_DISABLE		(1 << 4)
+#define  TXDSI_TYPE_NOT_RECOGNISED_ERROR_RECOVERY_DISABLE (1 << 3)
+#define  TXECC_MULTIBIT_ERROR_RECOVERY_DISABLE		(1 << 2)
+#define  CLOCKSTOP					(1 << 1)
+#define  EOT_DISABLE					(1 << 0)
+
+#define _MIPIA_LP_BYTECLK			(VLV_DISPLAY_BASE + 0xb060)
+#define _MIPIB_LP_BYTECLK			(VLV_DISPLAY_BASE + 0xb860)
+#define MIPI_LP_BYTECLK(pipe)		_PIPE(pipe, _MIPIA_LP_BYTECLK, _MIPIB_LP_BYTECLK)
+#define  LP_BYTECLK_SHIFT				0
+#define  LP_BYTECLK_MASK				(0xffff << 0)
+
+/* bits 31:0 */
+#define _MIPIA_LP_GEN_DATA			(VLV_DISPLAY_BASE + 0xb064)
+#define _MIPIB_LP_GEN_DATA			(VLV_DISPLAY_BASE + 0xb864)
+#define MIPI_LP_GEN_DATA(pipe)		_PIPE(pipe, _MIPIA_LP_GEN_DATA, _MIPIB_LP_GEN_DATA)
+
+/* bits 31:0 */
+#define _MIPIA_HS_GEN_DATA			(VLV_DISPLAY_BASE + 0xb068)
+#define _MIPIB_HS_GEN_DATA			(VLV_DISPLAY_BASE + 0xb868)
+#define MIPI_HS_GEN_DATA(pipe)		_PIPE(pipe, _MIPIA_HS_GEN_DATA, _MIPIB_HS_GEN_DATA)
+
+#define _MIPIA_LP_GEN_CTRL			(VLV_DISPLAY_BASE + 0xb06c)
+#define _MIPIB_LP_GEN_CTRL			(VLV_DISPLAY_BASE + 0xb86c)
+#define MIPI_LP_GEN_CTRL(pipe)		_PIPE(pipe, _MIPIA_LP_GEN_CTRL, _MIPIB_LP_GEN_CTRL)
+#define _MIPIA_HS_GEN_CTRL			(VLV_DISPLAY_BASE + 0xb070)
+#define _MIPIB_HS_GEN_CTRL			(VLV_DISPLAY_BASE + 0xb870)
+#define MIPI_HS_GEN_CTRL(pipe)		_PIPE(pipe, _MIPIA_HS_GEN_CTRL, _MIPIB_HS_GEN_CTRL)
+#define  LONG_PACKET_WORD_COUNT_SHIFT			8
+#define  LONG_PACKET_WORD_COUNT_MASK			(0xffff << 8)
+#define  SHORT_PACKET_PARAM_SHIFT			8
+#define  SHORT_PACKET_PARAM_MASK			(0xffff << 8)
+#define  VIRTUAL_CHANNEL_SHIFT				6
+#define  VIRTUAL_CHANNEL_MASK				(3 << 6)
+#define  DATA_TYPE_SHIFT				0
+#define  DATA_TYPE_MASK					(3f << 0)
+/* data type values, see include/video/mipi_display.h */
+
+#define _MIPIA_GEN_FIFO_STAT			(VLV_DISPLAY_BASE + 0xb074)
+#define _MIPIB_GEN_FIFO_STAT			(VLV_DISPLAY_BASE + 0xb874)
+#define MIPI_GEN_FIFO_STAT(pipe)	_PIPE(pipe, _MIPIA_GEN_FIFO_STAT, _MIPIB_GEN_FIFO_STAT)
+#define  DPI_FIFO_EMPTY					(1 << 28)
+#define  DBI_FIFO_EMPTY					(1 << 27)
+#define  LP_CTRL_FIFO_EMPTY				(1 << 26)
+#define  LP_CTRL_FIFO_HALF_EMPTY			(1 << 25)
+#define  LP_CTRL_FIFO_FULL				(1 << 24)
+#define  HS_CTRL_FIFO_EMPTY				(1 << 18)
+#define  HS_CTRL_FIFO_HALF_EMPTY			(1 << 17)
+#define  HS_CTRL_FIFO_FULL				(1 << 16)
+#define  LP_DATA_FIFO_EMPTY				(1 << 10)
+#define  LP_DATA_FIFO_HALF_EMPTY			(1 << 9)
+#define  LP_DATA_FIFO_FULL				(1 << 8)
+#define  HS_DATA_FIFO_EMPTY				(1 << 2)
+#define  HS_DATA_FIFO_HALF_EMPTY			(1 << 1)
+#define  HS_DATA_FIFO_FULL				(1 << 0)
+
+#define _MIPIA_HS_LS_DBI_ENABLE			(VLV_DISPLAY_BASE + 0xb078)
+#define _MIPIB_HS_LS_DBI_ENABLE			(VLV_DISPLAY_BASE + 0xb878)
+#define MIPI_HS_LP_DBI_ENABLE(pipe)	_PIPE(pipe, _MIPIA_HS_LS_DBI_ENABLE, _MIPIB_HS_LS_DBI_ENABLE)
+#define  DBI_HS_LP_MODE_MASK				(1 << 0)
+#define  DBI_LP_MODE					(1 << 0)
+#define  DBI_HS_MODE					(0 << 0)
+
+#define _MIPIA_DPHY_PARAM			(VLV_DISPLAY_BASE + 0xb080)
+#define _MIPIB_DPHY_PARAM			(VLV_DISPLAY_BASE + 0xb880)
+#define MIPI_DPHY_PARAM(pipe)		_PIPE(pipe, _MIPIA_DPHY_PARAM, _MIPIB_DPHY_PARAM)
+#define  EXIT_ZERO_COUNT_SHIFT				24
+#define  EXIT_ZERO_COUNT_MASK				(0x3f << 24)
+#define  TRAIL_COUNT_SHIFT				16
+#define  TRAIL_COUNT_MASK				(0x1f << 16)
+#define  CLK_ZERO_COUNT_SHIFT				8
+#define  CLK_ZERO_COUNT_MASK				(0xff << 8)
+#define  PREPARE_COUNT_SHIFT				0
+#define  PREPARE_COUNT_MASK				(0x3f << 0)
+
+/* bits 31:0 */
+#define _MIPIA_DBI_BW_CTRL			(VLV_DISPLAY_BASE + 0xb084)
+#define _MIPIB_DBI_BW_CTRL			(VLV_DISPLAY_BASE + 0xb884)
+#define MIPI_DBI_BW_CTRL(pipe)		_PIPE(pipe, _MIPIA_DBI_BW_CTRL, _MIPIB_DBI_BW_CTRL)
+
+#define _MIPIA_CLK_LANE_SWITCH_TIME_CNT		(VLV_DISPLAY_BASE + 0xb088)
+#define _MIPIB_CLK_LANE_SWITCH_TIME_CNT		(VLV_DISPLAY_BASE + 0xb888)
+#define MIPI_CLK_LANE_SWITCH_TIME_CNT(pipe)	_PIPE(pipe, _MIPIA_CLK_LANE_SWITCH_TIME_CNT, _MIPIB_CLK_LANE_SWITCH_TIME_CNT)
+#define  LP_HS_SSW_CNT_SHIFT				16
+#define  LP_HS_SSW_CNT_MASK				(0xffff << 16)
+#define  HS_LP_PWR_SW_CNT_SHIFT				0
+#define  HS_LP_PWR_SW_CNT_MASK				(0xffff << 0)
+
+#define _MIPIA_STOP_STATE_STALL			(VLV_DISPLAY_BASE + 0xb08c)
+#define _MIPIB_STOP_STATE_STALL			(VLV_DISPLAY_BASE + 0xb88c)
+#define MIPI_STOP_STATE_STALL(pipe)	_PIPE(pipe, _MIPIA_STOP_STATE_STALL, _MIPIB_STOP_STATE_STALL)
+#define  STOP_STATE_STALL_COUNTER_SHIFT			0
+#define  STOP_STATE_STALL_COUNTER_MASK			(0xff << 0)
+
+#define _MIPIA_INTR_STAT_REG_1			(VLV_DISPLAY_BASE + 0xb090)
+#define _MIPIB_INTR_STAT_REG_1			(VLV_DISPLAY_BASE + 0xb890)
+#define MIPI_INTR_STAT_REG_1(pipe)	_PIPE(pipe, _MIPIA_INTR_STAT_REG_1, _MIPIB_INTR_STAT_REG_1)
+#define _MIPIA_INTR_EN_REG_1			(VLV_DISPLAY_BASE + 0xb094)
+#define _MIPIB_INTR_EN_REG_1			(VLV_DISPLAY_BASE + 0xb894)
+#define MIPI_INTR_EN_REG_1(pipe)	_PIPE(pipe, _MIPIA_INTR_EN_REG_1, _MIPIB_INTR_EN_REG_1)
+#define  RX_CONTENTION_DETECTED				(1 << 0)
+
+/* XXX: only pipe A ?!? */
+#define MIPIA_DBI_TYPEC_CTRL			(VLV_DISPLAY_BASE + 0xb100)
+#define  DBI_TYPEC_ENABLE				(1 << 31)
+#define  DBI_TYPEC_WIP					(1 << 30)
+#define  DBI_TYPEC_OPTION_SHIFT				28
+#define  DBI_TYPEC_OPTION_MASK				(3 << 28)
+#define  DBI_TYPEC_FREQ_SHIFT				24
+#define  DBI_TYPEC_FREQ_MASK				(0xf << 24)
+#define  DBI_TYPEC_OVERRIDE				(1 << 8)
+#define  DBI_TYPEC_OVERRIDE_COUNTER_SHIFT		0
+#define  DBI_TYPEC_OVERRIDE_COUNTER_MASK		(0xff << 0)
+
+
+/* MIPI adapter registers */
+
+#define _MIPIA_CTRL				(VLV_DISPLAY_BASE + 0xb104)
+#define _MIPIB_CTRL				(VLV_DISPLAY_BASE + 0xb904)
+#define MIPI_CTRL(pipe)			_PIPE(pipe, _MIPIA_CTRL, _MIPIB_CTRL)
+#define  ESCAPE_CLOCK_DIVIDER_SHIFT			5 /* A only */
+#define  ESCAPE_CLOCK_DIVIDER_MASK			(3 << 5)
+#define  ESCAPE_CLOCK_DIVIDER_1				(0 << 5)
+#define  ESCAPE_CLOCK_DIVIDER_2				(1 << 5)
+#define  ESCAPE_CLOCK_DIVIDER_4				(2 << 5)
+#define  READ_REQUEST_PRIORITY_SHIFT			3
+#define  READ_REQUEST_PRIORITY_MASK			(3 << 3)
+#define  READ_REQUEST_PRIORITY_LOW			(0 << 3)
+#define  READ_REQUEST_PRIORITY_HIGH			(3 << 3)
+#define  RGB_FLIP_TO_BGR				(1 << 2)
+
+#define _MIPIA_DATA_ADDRESS			(VLV_DISPLAY_BASE + 0xb108)
+#define _MIPIB_DATA_ADDRESS			(VLV_DISPLAY_BASE + 0xb908)
+#define MIPI_DATA_ADDRESS(pipe)		_PIPE(pipe, _MIPIA_DATA_ADDRESS, _MIPIB_DATA_ADDRESS)
+#define  DATA_MEM_ADDRESS_SHIFT				5
+#define  DATA_MEM_ADDRESS_MASK				(0x7ffffff << 5)
+#define  DATA_VALID					(1 << 0)
+
+#define _MIPIA_DATA_LENGTH			(VLV_DISPLAY_BASE + 0xb10c)
+#define _MIPIB_DATA_LENGTH			(VLV_DISPLAY_BASE + 0xb90c)
+#define MIPI_DATA_LENGTH(pipe)		_PIPE(pipe, _MIPIA_DATA_LENGTH, _MIPIB_DATA_LENGTH)
+#define  DATA_LENGTH_SHIFT				0
+#define  DATA_LENGTH_MASK				(0xfffff << 0)
+
+#define _MIPIA_COMMAND_ADDRESS			(VLV_DISPLAY_BASE + 0xb110)
+#define _MIPIB_COMMAND_ADDRESS			(VLV_DISPLAY_BASE + 0xb910)
+#define MIPI_COMMAND_ADDRESS(pipe)	_PIPE(pipe, _MIPIA_COMMAND_ADDRESS, _MIPIB_COMMAND_ADDRESS)
+#define  COMMAND_MEM_ADDRESS_SHIFT			5
+#define  COMMAND_MEM_ADDRESS_MASK			(0x7ffffff << 5)
+#define  AUTO_PWG_ENABLE				(1 << 2)
+#define  MEMORY_WRITE_DATA_FROM_PIPE_RENDERING		(1 << 1)
+#define  COMMAND_VALID					(1 << 0)
+
+#define _MIPIA_COMMAND_LENGTH			(VLV_DISPLAY_BASE + 0xb114)
+#define _MIPIB_COMMAND_LENGTH			(VLV_DISPLAY_BASE + 0xb914)
+#define MIPI_COMMAND_LENGTH(pipe)	_PIPE(pipe, _MIPIA_COMMAND_LENGTH, _MIPIB_COMMAND_LENGTH)
+#define  COMMAND_LENGTH_SHIFT(n)			(8 * (n)) /* n: 0...3 */
+#define  COMMAND_LENGTH_MASK(n)				(0xff << (8 * (n)))
+
+#define _MIPIA_READ_DATA_RETURN0		(VLV_DISPLAY_BASE + 0xb118)
+#define _MIPIB_READ_DATA_RETURN0		(VLV_DISPLAY_BASE + 0xb918)
+#define MIPI_READ_DATA_RETURN(pipe, n) \
+	(_PIPE(pipe, _MIPIA_READ_DATA_RETURN0, _MIPIB_READ_DATA_RETURN0) + 4 * (n)) /* n: 0...7 */
+
+#define _MIPIA_READ_DATA_VALID			(VLV_DISPLAY_BASE + 0xb138)
+#define _MIPIB_READ_DATA_VALID			(VLV_DISPLAY_BASE + 0xb938)
+#define MIPI_READ_DATA_VALID(pipe)	_PIPE(pipe, _MIPIA_READ_DATA_VALID, _MIPIB_READ_DATA_VALID)
+#define  READ_DATA_VALID(n)				(1 << (n))
+
 #endif /* _I915_REG_H_ */
diff --git a/drivers/gpu/drm/i915/i915_suspend.c b/drivers/gpu/drm/i915/i915_suspend.c
index 70db618989c4..98790c7cccb1 100644
--- a/drivers/gpu/drm/i915/i915_suspend.c
+++ b/drivers/gpu/drm/i915/i915_suspend.c
@@ -214,6 +214,22 @@ static void i915_save_display(struct drm_device *dev)
 		dev_priv->regfile.saveBLC_CPU_PWM_CTL2 = I915_READ(BLC_PWM_CPU_CTL2);
 		if (HAS_PCH_IBX(dev) || HAS_PCH_CPT(dev))
 			dev_priv->regfile.saveLVDS = I915_READ(PCH_LVDS);
+	} else if (IS_VALLEYVIEW(dev)) {
+		dev_priv->regfile.savePP_CONTROL = I915_READ(PP_CONTROL);
+		dev_priv->regfile.savePFIT_PGM_RATIOS = I915_READ(PFIT_PGM_RATIOS);
+
+		dev_priv->regfile.saveBLC_PWM_CTL =
+			I915_READ(VLV_BLC_PWM_CTL(PIPE_A));
+		dev_priv->regfile.saveBLC_HIST_CTL =
+			I915_READ(VLV_BLC_HIST_CTL(PIPE_A));
+		dev_priv->regfile.saveBLC_PWM_CTL2 =
+			I915_READ(VLV_BLC_PWM_CTL2(PIPE_A));
+		dev_priv->regfile.saveBLC_PWM_CTL_B =
+			I915_READ(VLV_BLC_PWM_CTL(PIPE_B));
+		dev_priv->regfile.saveBLC_HIST_CTL_B =
+			I915_READ(VLV_BLC_HIST_CTL(PIPE_B));
+		dev_priv->regfile.saveBLC_PWM_CTL2_B =
+			I915_READ(VLV_BLC_PWM_CTL2(PIPE_B));
 	} else {
 		dev_priv->regfile.savePP_CONTROL = I915_READ(PP_CONTROL);
 		dev_priv->regfile.savePFIT_PGM_RATIOS = I915_READ(PFIT_PGM_RATIOS);
@@ -302,6 +318,19 @@ static void i915_restore_display(struct drm_device *dev)
 		I915_WRITE(PCH_PP_CONTROL, dev_priv->regfile.savePP_CONTROL);
 		I915_WRITE(RSTDBYCTL,
 			   dev_priv->regfile.saveMCHBAR_RENDER_STANDBY);
+	} else if (IS_VALLEYVIEW(dev)) {
+		I915_WRITE(VLV_BLC_PWM_CTL(PIPE_A),
+			   dev_priv->regfile.saveBLC_PWM_CTL);
+		I915_WRITE(VLV_BLC_HIST_CTL(PIPE_A),
+			   dev_priv->regfile.saveBLC_HIST_CTL);
+		I915_WRITE(VLV_BLC_PWM_CTL2(PIPE_A),
+			   dev_priv->regfile.saveBLC_PWM_CTL2);
+		I915_WRITE(VLV_BLC_PWM_CTL(PIPE_B),
+			   dev_priv->regfile.saveBLC_PWM_CTL);
+		I915_WRITE(VLV_BLC_HIST_CTL(PIPE_B),
+			   dev_priv->regfile.saveBLC_HIST_CTL);
+		I915_WRITE(VLV_BLC_PWM_CTL2(PIPE_B),
+			   dev_priv->regfile.saveBLC_PWM_CTL2);
 	} else {
 		I915_WRITE(PFIT_PGM_RATIOS, dev_priv->regfile.savePFIT_PGM_RATIOS);
 		I915_WRITE(BLC_PWM_CTL, dev_priv->regfile.saveBLC_PWM_CTL);
@@ -340,7 +369,9 @@ int i915_save_state(struct drm_device *dev)
 	struct drm_i915_private *dev_priv = dev->dev_private;
 	int i;
 
-	pci_read_config_byte(dev->pdev, LBB, &dev_priv->regfile.saveLBB);
+	if (INTEL_INFO(dev)->gen <= 4)
+		pci_read_config_byte(dev->pdev, LBB,
+				     &dev_priv->regfile.saveLBB);
 
 	mutex_lock(&dev->struct_mutex);
 
@@ -367,7 +398,8 @@ int i915_save_state(struct drm_device *dev)
 	intel_disable_gt_powersave(dev);
 
 	/* Cache mode state */
-	dev_priv->regfile.saveCACHE_MODE_0 = I915_READ(CACHE_MODE_0);
+	if (INTEL_INFO(dev)->gen < 7)
+		dev_priv->regfile.saveCACHE_MODE_0 = I915_READ(CACHE_MODE_0);
 
 	/* Memory Arbitration state */
 	dev_priv->regfile.saveMI_ARB_STATE = I915_READ(MI_ARB_STATE);
@@ -390,7 +422,9 @@ int i915_restore_state(struct drm_device *dev)
 	struct drm_i915_private *dev_priv = dev->dev_private;
 	int i;
 
-	pci_write_config_byte(dev->pdev, LBB, dev_priv->regfile.saveLBB);
+	if (INTEL_INFO(dev)->gen <= 4)
+		pci_write_config_byte(dev->pdev, LBB,
+				      dev_priv->regfile.saveLBB);
 
 	mutex_lock(&dev->struct_mutex);
 
@@ -414,7 +448,9 @@ int i915_restore_state(struct drm_device *dev)
 	}
 
 	/* Cache mode state */
-	I915_WRITE(CACHE_MODE_0, dev_priv->regfile.saveCACHE_MODE_0 | 0xffff0000);
+	if (INTEL_INFO(dev)->gen < 7)
+		I915_WRITE(CACHE_MODE_0, dev_priv->regfile.saveCACHE_MODE_0 |
+			   0xffff0000);
 
 	/* Memory arbitration state */
 	I915_WRITE(MI_ARB_STATE, dev_priv->regfile.saveMI_ARB_STATE | 0xffff0000);
diff --git a/drivers/gpu/drm/i915/i915_sysfs.c b/drivers/gpu/drm/i915/i915_sysfs.c
index c8c4112de110..cef38fd320a7 100644
--- a/drivers/gpu/drm/i915/i915_sysfs.c
+++ b/drivers/gpu/drm/i915/i915_sysfs.c
@@ -32,30 +32,50 @@
 #include "intel_drv.h"
 #include "i915_drv.h"
 
+#define dev_to_drm_minor(d) dev_get_drvdata((d))
+
 #ifdef CONFIG_PM
 static u32 calc_residency(struct drm_device *dev, const u32 reg)
 {
 	struct drm_i915_private *dev_priv = dev->dev_private;
 	u64 raw_time; /* 32b value may overflow during fixed point math */
+	u64 units = 128ULL, div = 100000ULL, bias = 100ULL;
 
 	if (!intel_enable_rc6(dev))
 		return 0;
 
-	raw_time = I915_READ(reg) * 128ULL;
-	return DIV_ROUND_UP_ULL(raw_time, 100000);
+	/* On VLV, residency time is in CZ units rather than 1.28us */
+	if (IS_VALLEYVIEW(dev)) {
+		u32 clkctl2;
+
+		clkctl2 = I915_READ(VLV_CLK_CTL2) >>
+			CLK_CTL2_CZCOUNT_30NS_SHIFT;
+		if (!clkctl2) {
+			WARN(!clkctl2, "bogus CZ count value");
+			return 0;
+		}
+		units = DIV_ROUND_UP_ULL(30ULL * bias, (u64)clkctl2);
+		if (I915_READ(VLV_COUNTER_CONTROL) & VLV_COUNT_RANGE_HIGH)
+			units <<= 8;
+
+		div = 1000000ULL * bias;
+	}
+
+	raw_time = I915_READ(reg) * units;
+	return DIV_ROUND_UP_ULL(raw_time, div);
 }
 
 static ssize_t
 show_rc6_mask(struct device *kdev, struct device_attribute *attr, char *buf)
 {
-	struct drm_minor *dminor = container_of(kdev, struct drm_minor, kdev);
+	struct drm_minor *dminor = dev_to_drm_minor(kdev);
 	return snprintf(buf, PAGE_SIZE, "%x\n", intel_enable_rc6(dminor->dev));
 }
 
 static ssize_t
 show_rc6_ms(struct device *kdev, struct device_attribute *attr, char *buf)
 {
-	struct drm_minor *dminor = container_of(kdev, struct drm_minor, kdev);
+	struct drm_minor *dminor = dev_get_drvdata(kdev);
 	u32 rc6_residency = calc_residency(dminor->dev, GEN6_GT_GFX_RC6);
 	return snprintf(buf, PAGE_SIZE, "%u\n", rc6_residency);
 }
@@ -63,16 +83,20 @@ show_rc6_ms(struct device *kdev, struct device_attribute *attr, char *buf)
 static ssize_t
 show_rc6p_ms(struct device *kdev, struct device_attribute *attr, char *buf)
 {
-	struct drm_minor *dminor = container_of(kdev, struct drm_minor, kdev);
+	struct drm_minor *dminor = dev_to_drm_minor(kdev);
 	u32 rc6p_residency = calc_residency(dminor->dev, GEN6_GT_GFX_RC6p);
+	if (IS_VALLEYVIEW(dminor->dev))
+		rc6p_residency = 0;
 	return snprintf(buf, PAGE_SIZE, "%u\n", rc6p_residency);
 }
 
 static ssize_t
 show_rc6pp_ms(struct device *kdev, struct device_attribute *attr, char *buf)
 {
-	struct drm_minor *dminor = container_of(kdev, struct drm_minor, kdev);
+	struct drm_minor *dminor = dev_to_drm_minor(kdev);
 	u32 rc6pp_residency = calc_residency(dminor->dev, GEN6_GT_GFX_RC6pp);
+	if (IS_VALLEYVIEW(dminor->dev))
+		rc6pp_residency = 0;
 	return snprintf(buf, PAGE_SIZE, "%u\n", rc6pp_residency);
 }
 
@@ -97,7 +121,7 @@ static struct attribute_group rc6_attr_group = {
 
 static int l3_access_valid(struct drm_device *dev, loff_t offset)
 {
-	if (!HAS_L3_GPU_CACHE(dev))
+	if (!HAS_L3_DPF(dev))
 		return -EPERM;
 
 	if (offset % 4 != 0)
@@ -115,31 +139,34 @@ i915_l3_read(struct file *filp, struct kobject *kobj,
 	     loff_t offset, size_t count)
 {
 	struct device *dev = container_of(kobj, struct device, kobj);
-	struct drm_minor *dminor = container_of(dev, struct drm_minor, kdev);
+	struct drm_minor *dminor = dev_to_drm_minor(dev);
 	struct drm_device *drm_dev = dminor->dev;
 	struct drm_i915_private *dev_priv = drm_dev->dev_private;
-	uint32_t misccpctl;
-	int i, ret;
+	int slice = (int)(uintptr_t)attr->private;
+	int ret;
+
+	count = round_down(count, 4);
 
 	ret = l3_access_valid(drm_dev, offset);
 	if (ret)
 		return ret;
 
+	count = min_t(size_t, GEN7_L3LOG_SIZE - offset, count);
+
 	ret = i915_mutex_lock_interruptible(drm_dev);
 	if (ret)
 		return ret;
 
-	misccpctl = I915_READ(GEN7_MISCCPCTL);
-	I915_WRITE(GEN7_MISCCPCTL, misccpctl & ~GEN7_DOP_CLOCK_GATE_ENABLE);
-
-	for (i = offset; count >= 4 && i < GEN7_L3LOG_SIZE; i += 4, count -= 4)
-		*((uint32_t *)(&buf[i])) = I915_READ(GEN7_L3LOG_BASE + i);
-
-	I915_WRITE(GEN7_MISCCPCTL, misccpctl);
+	if (dev_priv->l3_parity.remap_info[slice])
+		memcpy(buf,
+		       dev_priv->l3_parity.remap_info[slice] + (offset/4),
+		       count);
+	else
+		memset(buf, 0, count);
 
 	mutex_unlock(&drm_dev->struct_mutex);
 
-	return i - offset;
+	return count;
 }
 
 static ssize_t
@@ -148,21 +175,26 @@ i915_l3_write(struct file *filp, struct kobject *kobj,
 	      loff_t offset, size_t count)
 {
 	struct device *dev = container_of(kobj, struct device, kobj);
-	struct drm_minor *dminor = container_of(dev, struct drm_minor, kdev);
+	struct drm_minor *dminor = dev_to_drm_minor(dev);
 	struct drm_device *drm_dev = dminor->dev;
 	struct drm_i915_private *dev_priv = drm_dev->dev_private;
+	struct i915_hw_context *ctx;
 	u32 *temp = NULL; /* Just here to make handling failures easy */
+	int slice = (int)(uintptr_t)attr->private;
 	int ret;
 
 	ret = l3_access_valid(drm_dev, offset);
 	if (ret)
 		return ret;
 
+	if (dev_priv->hw_contexts_disabled)
+		return -ENXIO;
+
 	ret = i915_mutex_lock_interruptible(drm_dev);
 	if (ret)
 		return ret;
 
-	if (!dev_priv->l3_parity.remap_info) {
+	if (!dev_priv->l3_parity.remap_info[slice]) {
 		temp = kzalloc(GEN7_L3LOG_SIZE, GFP_KERNEL);
 		if (!temp) {
 			mutex_unlock(&drm_dev->struct_mutex);
@@ -182,13 +214,13 @@ i915_l3_write(struct file *filp, struct kobject *kobj,
 	 * at this point it is left as a TODO.
 	*/
 	if (temp)
-		dev_priv->l3_parity.remap_info = temp;
+		dev_priv->l3_parity.remap_info[slice] = temp;
 
-	memcpy(dev_priv->l3_parity.remap_info + (offset/4),
-	       buf + (offset/4),
-	       count);
+	memcpy(dev_priv->l3_parity.remap_info[slice] + (offset/4), buf, count);
 
-	i915_gem_l3_remap(drm_dev);
+	/* NB: We defer the remapping until we switch to the context */
+	list_for_each_entry(ctx, &dev_priv->context_list, link)
+		ctx->remap_slice |= (1<<slice);
 
 	mutex_unlock(&drm_dev->struct_mutex);
 
@@ -200,17 +232,29 @@ static struct bin_attribute dpf_attrs = {
 	.size = GEN7_L3LOG_SIZE,
 	.read = i915_l3_read,
 	.write = i915_l3_write,
-	.mmap = NULL
+	.mmap = NULL,
+	.private = (void *)0
+};
+
+static struct bin_attribute dpf_attrs_1 = {
+	.attr = {.name = "l3_parity_slice_1", .mode = (S_IRUSR | S_IWUSR)},
+	.size = GEN7_L3LOG_SIZE,
+	.read = i915_l3_read,
+	.write = i915_l3_write,
+	.mmap = NULL,
+	.private = (void *)1
 };
 
 static ssize_t gt_cur_freq_mhz_show(struct device *kdev,
 				    struct device_attribute *attr, char *buf)
 {
-	struct drm_minor *minor = container_of(kdev, struct drm_minor, kdev);
+	struct drm_minor *minor = dev_to_drm_minor(kdev);
 	struct drm_device *dev = minor->dev;
 	struct drm_i915_private *dev_priv = dev->dev_private;
 	int ret;
 
+	flush_delayed_work(&dev_priv->rps.delayed_resume_work);
+
 	mutex_lock(&dev_priv->rps.hw_lock);
 	if (IS_VALLEYVIEW(dev_priv->dev)) {
 		u32 freq;
@@ -227,7 +271,7 @@ static ssize_t gt_cur_freq_mhz_show(struct device *kdev,
 static ssize_t vlv_rpe_freq_mhz_show(struct device *kdev,
 				     struct device_attribute *attr, char *buf)
 {
-	struct drm_minor *minor = container_of(kdev, struct drm_minor, kdev);
+	struct drm_minor *minor = dev_to_drm_minor(kdev);
 	struct drm_device *dev = minor->dev;
 	struct drm_i915_private *dev_priv = dev->dev_private;
 
@@ -238,11 +282,13 @@ static ssize_t vlv_rpe_freq_mhz_show(struct device *kdev,
 
 static ssize_t gt_max_freq_mhz_show(struct device *kdev, struct device_attribute *attr, char *buf)
 {
-	struct drm_minor *minor = container_of(kdev, struct drm_minor, kdev);
+	struct drm_minor *minor = dev_to_drm_minor(kdev);
 	struct drm_device *dev = minor->dev;
 	struct drm_i915_private *dev_priv = dev->dev_private;
 	int ret;
 
+	flush_delayed_work(&dev_priv->rps.delayed_resume_work);
+
 	mutex_lock(&dev_priv->rps.hw_lock);
 	if (IS_VALLEYVIEW(dev_priv->dev))
 		ret = vlv_gpu_freq(dev_priv->mem_freq, dev_priv->rps.max_delay);
@@ -257,7 +303,7 @@ static ssize_t gt_max_freq_mhz_store(struct device *kdev,
 				     struct device_attribute *attr,
 				     const char *buf, size_t count)
 {
-	struct drm_minor *minor = container_of(kdev, struct drm_minor, kdev);
+	struct drm_minor *minor = dev_to_drm_minor(kdev);
 	struct drm_device *dev = minor->dev;
 	struct drm_i915_private *dev_priv = dev->dev_private;
 	u32 val, rp_state_cap, hw_max, hw_min, non_oc_max;
@@ -267,6 +313,8 @@ static ssize_t gt_max_freq_mhz_store(struct device *kdev,
 	if (ret)
 		return ret;
 
+	flush_delayed_work(&dev_priv->rps.delayed_resume_work);
+
 	mutex_lock(&dev_priv->rps.hw_lock);
 
 	if (IS_VALLEYVIEW(dev_priv->dev)) {
@@ -310,11 +358,13 @@ static ssize_t gt_max_freq_mhz_store(struct device *kdev,
 
 static ssize_t gt_min_freq_mhz_show(struct device *kdev, struct device_attribute *attr, char *buf)
 {
-	struct drm_minor *minor = container_of(kdev, struct drm_minor, kdev);
+	struct drm_minor *minor = dev_to_drm_minor(kdev);
 	struct drm_device *dev = minor->dev;
 	struct drm_i915_private *dev_priv = dev->dev_private;
 	int ret;
 
+	flush_delayed_work(&dev_priv->rps.delayed_resume_work);
+
 	mutex_lock(&dev_priv->rps.hw_lock);
 	if (IS_VALLEYVIEW(dev_priv->dev))
 		ret = vlv_gpu_freq(dev_priv->mem_freq, dev_priv->rps.min_delay);
@@ -329,7 +379,7 @@ static ssize_t gt_min_freq_mhz_store(struct device *kdev,
 				     struct device_attribute *attr,
 				     const char *buf, size_t count)
 {
-	struct drm_minor *minor = container_of(kdev, struct drm_minor, kdev);
+	struct drm_minor *minor = dev_to_drm_minor(kdev);
 	struct drm_device *dev = minor->dev;
 	struct drm_i915_private *dev_priv = dev->dev_private;
 	u32 val, rp_state_cap, hw_max, hw_min;
@@ -339,6 +389,8 @@ static ssize_t gt_min_freq_mhz_store(struct device *kdev,
 	if (ret)
 		return ret;
 
+	flush_delayed_work(&dev_priv->rps.delayed_resume_work);
+
 	mutex_lock(&dev_priv->rps.hw_lock);
 
 	if (IS_VALLEYVIEW(dev)) {
@@ -388,7 +440,7 @@ static DEVICE_ATTR(gt_RPn_freq_mhz, S_IRUGO, gt_rp_mhz_show, NULL);
 /* For now we have a static number of RP states */
 static ssize_t gt_rp_mhz_show(struct device *kdev, struct device_attribute *attr, char *buf)
 {
-	struct drm_minor *minor = container_of(kdev, struct drm_minor, kdev);
+	struct drm_minor *minor = dev_to_drm_minor(kdev);
 	struct drm_device *dev = minor->dev;
 	struct drm_i915_private *dev_priv = dev->dev_private;
 	u32 val, rp_state_cap;
@@ -436,7 +488,7 @@ static ssize_t error_state_read(struct file *filp, struct kobject *kobj,
 {
 
 	struct device *kdev = container_of(kobj, struct device, kobj);
-	struct drm_minor *minor = container_of(kdev, struct drm_minor, kdev);
+	struct drm_minor *minor = dev_to_drm_minor(kdev);
 	struct drm_device *dev = minor->dev;
 	struct i915_error_state_file_priv error_priv;
 	struct drm_i915_error_state_buf error_str;
@@ -471,7 +523,7 @@ static ssize_t error_state_write(struct file *file, struct kobject *kobj,
 				 loff_t off, size_t count)
 {
 	struct device *kdev = container_of(kobj, struct device, kobj);
-	struct drm_minor *minor = container_of(kdev, struct drm_minor, kdev);
+	struct drm_minor *minor = dev_to_drm_minor(kdev);
 	struct drm_device *dev = minor->dev;
 	int ret;
 
@@ -501,27 +553,34 @@ void i915_setup_sysfs(struct drm_device *dev)
 
 #ifdef CONFIG_PM
 	if (INTEL_INFO(dev)->gen >= 6) {
-		ret = sysfs_merge_group(&dev->primary->kdev.kobj,
+		ret = sysfs_merge_group(&dev->primary->kdev->kobj,
 					&rc6_attr_group);
 		if (ret)
 			DRM_ERROR("RC6 residency sysfs setup failed\n");
 	}
 #endif
-	if (HAS_L3_GPU_CACHE(dev)) {
-		ret = device_create_bin_file(&dev->primary->kdev, &dpf_attrs);
+	if (HAS_L3_DPF(dev)) {
+		ret = device_create_bin_file(dev->primary->kdev, &dpf_attrs);
 		if (ret)
 			DRM_ERROR("l3 parity sysfs setup failed\n");
+
+		if (NUM_L3_SLICES(dev) > 1) {
+			ret = device_create_bin_file(dev->primary->kdev,
+						     &dpf_attrs_1);
+			if (ret)
+				DRM_ERROR("l3 parity slice 1 setup failed\n");
+		}
 	}
 
 	ret = 0;
 	if (IS_VALLEYVIEW(dev))
-		ret = sysfs_create_files(&dev->primary->kdev.kobj, vlv_attrs);
+		ret = sysfs_create_files(&dev->primary->kdev->kobj, vlv_attrs);
 	else if (INTEL_INFO(dev)->gen >= 6)
-		ret = sysfs_create_files(&dev->primary->kdev.kobj, gen6_attrs);
+		ret = sysfs_create_files(&dev->primary->kdev->kobj, gen6_attrs);
 	if (ret)
 		DRM_ERROR("RPS sysfs setup failed\n");
 
-	ret = sysfs_create_bin_file(&dev->primary->kdev.kobj,
+	ret = sysfs_create_bin_file(&dev->primary->kdev->kobj,
 				    &error_state_attr);
 	if (ret)
 		DRM_ERROR("error_state sysfs setup failed\n");
@@ -529,13 +588,14 @@ void i915_setup_sysfs(struct drm_device *dev)
 
 void i915_teardown_sysfs(struct drm_device *dev)
 {
-	sysfs_remove_bin_file(&dev->primary->kdev.kobj, &error_state_attr);
+	sysfs_remove_bin_file(&dev->primary->kdev->kobj, &error_state_attr);
 	if (IS_VALLEYVIEW(dev))
-		sysfs_remove_files(&dev->primary->kdev.kobj, vlv_attrs);
+		sysfs_remove_files(&dev->primary->kdev->kobj, vlv_attrs);
 	else
-		sysfs_remove_files(&dev->primary->kdev.kobj, gen6_attrs);
-	device_remove_bin_file(&dev->primary->kdev,  &dpf_attrs);
+		sysfs_remove_files(&dev->primary->kdev->kobj, gen6_attrs);
+	device_remove_bin_file(dev->primary->kdev,  &dpf_attrs_1);
+	device_remove_bin_file(dev->primary->kdev,  &dpf_attrs);
 #ifdef CONFIG_PM
-	sysfs_unmerge_group(&dev->primary->kdev.kobj, &rc6_attr_group);
+	sysfs_unmerge_group(&dev->primary->kdev->kobj, &rc6_attr_group);
 #endif
 }
diff --git a/drivers/gpu/drm/i915/i915_trace.h b/drivers/gpu/drm/i915/i915_trace.h
index e2c5ee6f6194..6e580c98dede 100644
--- a/drivers/gpu/drm/i915/i915_trace.h
+++ b/drivers/gpu/drm/i915/i915_trace.h
@@ -233,6 +233,47 @@ TRACE_EVENT(i915_gem_evict_everything,
 	    TP_printk("dev=%d", __entry->dev)
 );
 
+TRACE_EVENT(i915_gem_evict_vm,
+	    TP_PROTO(struct i915_address_space *vm),
+	    TP_ARGS(vm),
+
+	    TP_STRUCT__entry(
+			     __field(struct i915_address_space *, vm)
+			    ),
+
+	    TP_fast_assign(
+			   __entry->vm = vm;
+			  ),
+
+	    TP_printk("dev=%d, vm=%p", __entry->vm->dev->primary->index, __entry->vm)
+);
+
+TRACE_EVENT(i915_gem_ring_sync_to,
+	    TP_PROTO(struct intel_ring_buffer *from,
+		     struct intel_ring_buffer *to,
+		     u32 seqno),
+	    TP_ARGS(from, to, seqno),
+
+	    TP_STRUCT__entry(
+			     __field(u32, dev)
+			     __field(u32, sync_from)
+			     __field(u32, sync_to)
+			     __field(u32, seqno)
+			     ),
+
+	    TP_fast_assign(
+			   __entry->dev = from->dev->primary->index;
+			   __entry->sync_from = from->id;
+			   __entry->sync_to = to->id;
+			   __entry->seqno = seqno;
+			   ),
+
+	    TP_printk("dev=%u, sync-from=%u, sync-to=%u, seqno=%u",
+		      __entry->dev,
+		      __entry->sync_from, __entry->sync_to,
+		      __entry->seqno)
+);
+
 TRACE_EVENT(i915_gem_ring_dispatch,
 	    TP_PROTO(struct intel_ring_buffer *ring, u32 seqno, u32 flags),
 	    TP_ARGS(ring, seqno, flags),
@@ -304,9 +345,24 @@ DEFINE_EVENT(i915_gem_request, i915_gem_request_add,
 	    TP_ARGS(ring, seqno)
 );
 
-DEFINE_EVENT(i915_gem_request, i915_gem_request_complete,
-	    TP_PROTO(struct intel_ring_buffer *ring, u32 seqno),
-	    TP_ARGS(ring, seqno)
+TRACE_EVENT(i915_gem_request_complete,
+	    TP_PROTO(struct intel_ring_buffer *ring),
+	    TP_ARGS(ring),
+
+	    TP_STRUCT__entry(
+			     __field(u32, dev)
+			     __field(u32, ring)
+			     __field(u32, seqno)
+			     ),
+
+	    TP_fast_assign(
+			   __entry->dev = ring->dev->primary->index;
+			   __entry->ring = ring->id;
+			   __entry->seqno = ring->get_seqno(ring, false);
+			   ),
+
+	    TP_printk("dev=%u, ring=%u, seqno=%u",
+		      __entry->dev, __entry->ring, __entry->seqno)
 );
 
 DEFINE_EVENT(i915_gem_request, i915_gem_request_retire,
diff --git a/drivers/gpu/drm/i915/intel_bios.c b/drivers/gpu/drm/i915/intel_bios.c
index 53f2bed8bc5f..6dd622d733b9 100644
--- a/drivers/gpu/drm/i915/intel_bios.c
+++ b/drivers/gpu/drm/i915/intel_bios.c
@@ -389,7 +389,7 @@ parse_sdvo_device_mapping(struct drm_i915_private *dev_priv,
 {
 	struct sdvo_device_mapping *p_mapping;
 	struct bdb_general_definitions *p_defs;
-	struct child_device_config *p_child;
+	union child_device_config *p_child;
 	int i, child_device_num, count;
 	u16	block_size;
 
@@ -416,36 +416,36 @@ parse_sdvo_device_mapping(struct drm_i915_private *dev_priv,
 	count = 0;
 	for (i = 0; i < child_device_num; i++) {
 		p_child = &(p_defs->devices[i]);
-		if (!p_child->device_type) {
+		if (!p_child->old.device_type) {
 			/* skip the device block if device type is invalid */
 			continue;
 		}
-		if (p_child->slave_addr != SLAVE_ADDR1 &&
-			p_child->slave_addr != SLAVE_ADDR2) {
+		if (p_child->old.slave_addr != SLAVE_ADDR1 &&
+			p_child->old.slave_addr != SLAVE_ADDR2) {
 			/*
 			 * If the slave address is neither 0x70 nor 0x72,
 			 * it is not a SDVO device. Skip it.
 			 */
 			continue;
 		}
-		if (p_child->dvo_port != DEVICE_PORT_DVOB &&
-			p_child->dvo_port != DEVICE_PORT_DVOC) {
+		if (p_child->old.dvo_port != DEVICE_PORT_DVOB &&
+			p_child->old.dvo_port != DEVICE_PORT_DVOC) {
 			/* skip the incorrect SDVO port */
 			DRM_DEBUG_KMS("Incorrect SDVO port. Skip it\n");
 			continue;
 		}
 		DRM_DEBUG_KMS("the SDVO device with slave addr %2x is found on"
 				" %s port\n",
-				p_child->slave_addr,
-				(p_child->dvo_port == DEVICE_PORT_DVOB) ?
+				p_child->old.slave_addr,
+				(p_child->old.dvo_port == DEVICE_PORT_DVOB) ?
 					"SDVOB" : "SDVOC");
-		p_mapping = &(dev_priv->sdvo_mappings[p_child->dvo_port - 1]);
+		p_mapping = &(dev_priv->sdvo_mappings[p_child->old.dvo_port - 1]);
 		if (!p_mapping->initialized) {
-			p_mapping->dvo_port = p_child->dvo_port;
-			p_mapping->slave_addr = p_child->slave_addr;
-			p_mapping->dvo_wiring = p_child->dvo_wiring;
-			p_mapping->ddc_pin = p_child->ddc_pin;
-			p_mapping->i2c_pin = p_child->i2c_pin;
+			p_mapping->dvo_port = p_child->old.dvo_port;
+			p_mapping->slave_addr = p_child->old.slave_addr;
+			p_mapping->dvo_wiring = p_child->old.dvo_wiring;
+			p_mapping->ddc_pin = p_child->old.ddc_pin;
+			p_mapping->i2c_pin = p_child->old.i2c_pin;
 			p_mapping->initialized = 1;
 			DRM_DEBUG_KMS("SDVO device: dvo=%x, addr=%x, wiring=%d, ddc_pin=%d, i2c_pin=%d\n",
 				      p_mapping->dvo_port,
@@ -457,7 +457,7 @@ parse_sdvo_device_mapping(struct drm_i915_private *dev_priv,
 			DRM_DEBUG_KMS("Maybe one SDVO port is shared by "
 					 "two SDVO device.\n");
 		}
-		if (p_child->slave2_addr) {
+		if (p_child->old.slave2_addr) {
 			/* Maybe this is a SDVO device with multiple inputs */
 			/* And the mapping info is not added */
 			DRM_DEBUG_KMS("there exists the slave2_addr. Maybe this"
@@ -477,15 +477,13 @@ static void
 parse_driver_features(struct drm_i915_private *dev_priv,
 		       struct bdb_header *bdb)
 {
-	struct drm_device *dev = dev_priv->dev;
 	struct bdb_driver_features *driver;
 
 	driver = find_section(bdb, BDB_DRIVER_FEATURES);
 	if (!driver)
 		return;
 
-	if (SUPPORTS_EDP(dev) &&
-	    driver->lvds_config == BDB_DRIVER_FEATURE_EDP)
+	if (driver->lvds_config == BDB_DRIVER_FEATURE_EDP)
 		dev_priv->vbt.edp_support = 1;
 
 	if (driver->dual_frequency)
@@ -501,7 +499,7 @@ parse_edp(struct drm_i915_private *dev_priv, struct bdb_header *bdb)
 
 	edp = find_section(bdb, BDB_EDP);
 	if (!edp) {
-		if (SUPPORTS_EDP(dev_priv->dev) && dev_priv->vbt.edp_support)
+		if (dev_priv->vbt.edp_support)
 			DRM_DEBUG_KMS("No eDP BDB found but eDP panel supported.\n");
 		return;
 	}
@@ -569,11 +567,149 @@ parse_edp(struct drm_i915_private *dev_priv, struct bdb_header *bdb)
 }
 
 static void
+parse_mipi(struct drm_i915_private *dev_priv, struct bdb_header *bdb)
+{
+	struct bdb_mipi *mipi;
+
+	mipi = find_section(bdb, BDB_MIPI);
+	if (!mipi) {
+		DRM_DEBUG_KMS("No MIPI BDB found");
+		return;
+	}
+
+	/* XXX: add more info */
+	dev_priv->vbt.dsi.panel_id = mipi->panel_id;
+}
+
+static void parse_ddi_port(struct drm_i915_private *dev_priv, enum port port,
+			   struct bdb_header *bdb)
+{
+	union child_device_config *it, *child = NULL;
+	struct ddi_vbt_port_info *info = &dev_priv->vbt.ddi_port_info[port];
+	uint8_t hdmi_level_shift;
+	int i, j;
+	bool is_dvi, is_hdmi, is_dp, is_edp, is_crt;
+	uint8_t aux_channel;
+	/* Each DDI port can have more than one value on the "DVO Port" field,
+	 * so look for all the possible values for each port and abort if more
+	 * than one is found. */
+	int dvo_ports[][2] = {
+		{DVO_PORT_HDMIA, DVO_PORT_DPA},
+		{DVO_PORT_HDMIB, DVO_PORT_DPB},
+		{DVO_PORT_HDMIC, DVO_PORT_DPC},
+		{DVO_PORT_HDMID, DVO_PORT_DPD},
+		{DVO_PORT_CRT, -1 /* Port E can only be DVO_PORT_CRT */ },
+	};
+
+	/* Find the child device to use, abort if more than one found. */
+	for (i = 0; i < dev_priv->vbt.child_dev_num; i++) {
+		it = dev_priv->vbt.child_dev + i;
+
+		for (j = 0; j < 2; j++) {
+			if (dvo_ports[port][j] == -1)
+				break;
+
+			if (it->common.dvo_port == dvo_ports[port][j]) {
+				if (child) {
+					DRM_DEBUG_KMS("More than one child device for port %c in VBT.\n",
+						      port_name(port));
+					return;
+				}
+				child = it;
+			}
+		}
+	}
+	if (!child)
+		return;
+
+	aux_channel = child->raw[25];
+
+	is_dvi = child->common.device_type & DEVICE_TYPE_TMDS_DVI_SIGNALING;
+	is_dp = child->common.device_type & DEVICE_TYPE_DISPLAYPORT_OUTPUT;
+	is_crt = child->common.device_type & DEVICE_TYPE_ANALOG_OUTPUT;
+	is_hdmi = is_dvi && (child->common.device_type & DEVICE_TYPE_NOT_HDMI_OUTPUT) == 0;
+	is_edp = is_dp && (child->common.device_type & DEVICE_TYPE_INTERNAL_CONNECTOR);
+
+	info->supports_dvi = is_dvi;
+	info->supports_hdmi = is_hdmi;
+	info->supports_dp = is_dp;
+
+	DRM_DEBUG_KMS("Port %c VBT info: DP:%d HDMI:%d DVI:%d EDP:%d CRT:%d\n",
+		      port_name(port), is_dp, is_hdmi, is_dvi, is_edp, is_crt);
+
+	if (is_edp && is_dvi)
+		DRM_DEBUG_KMS("Internal DP port %c is TMDS compatible\n",
+			      port_name(port));
+	if (is_crt && port != PORT_E)
+		DRM_DEBUG_KMS("Port %c is analog\n", port_name(port));
+	if (is_crt && (is_dvi || is_dp))
+		DRM_DEBUG_KMS("Analog port %c is also DP or TMDS compatible\n",
+			      port_name(port));
+	if (is_dvi && (port == PORT_A || port == PORT_E))
+		DRM_DEBUG_KMS("Port %c is TMDS compabile\n", port_name(port));
+	if (!is_dvi && !is_dp && !is_crt)
+		DRM_DEBUG_KMS("Port %c is not DP/TMDS/CRT compatible\n",
+			      port_name(port));
+	if (is_edp && (port == PORT_B || port == PORT_C || port == PORT_E))
+		DRM_DEBUG_KMS("Port %c is internal DP\n", port_name(port));
+
+	if (is_dvi) {
+		if (child->common.ddc_pin == 0x05 && port != PORT_B)
+			DRM_DEBUG_KMS("Unexpected DDC pin for port B\n");
+		if (child->common.ddc_pin == 0x04 && port != PORT_C)
+			DRM_DEBUG_KMS("Unexpected DDC pin for port C\n");
+		if (child->common.ddc_pin == 0x06 && port != PORT_D)
+			DRM_DEBUG_KMS("Unexpected DDC pin for port D\n");
+	}
+
+	if (is_dp) {
+		if (aux_channel == 0x40 && port != PORT_A)
+			DRM_DEBUG_KMS("Unexpected AUX channel for port A\n");
+		if (aux_channel == 0x10 && port != PORT_B)
+			DRM_DEBUG_KMS("Unexpected AUX channel for port B\n");
+		if (aux_channel == 0x20 && port != PORT_C)
+			DRM_DEBUG_KMS("Unexpected AUX channel for port C\n");
+		if (aux_channel == 0x30 && port != PORT_D)
+			DRM_DEBUG_KMS("Unexpected AUX channel for port D\n");
+	}
+
+	if (bdb->version >= 158) {
+		/* The VBT HDMI level shift values match the table we have. */
+		hdmi_level_shift = child->raw[7] & 0xF;
+		if (hdmi_level_shift < 0xC) {
+			DRM_DEBUG_KMS("VBT HDMI level shift for port %c: %d\n",
+				      port_name(port),
+				      hdmi_level_shift);
+			info->hdmi_level_shift = hdmi_level_shift;
+		}
+	}
+}
+
+static void parse_ddi_ports(struct drm_i915_private *dev_priv,
+			    struct bdb_header *bdb)
+{
+	struct drm_device *dev = dev_priv->dev;
+	enum port port;
+
+	if (!HAS_DDI(dev))
+		return;
+
+	if (!dev_priv->vbt.child_dev_num)
+		return;
+
+	if (bdb->version < 155)
+		return;
+
+	for (port = PORT_A; port < I915_MAX_PORTS; port++)
+		parse_ddi_port(dev_priv, port, bdb);
+}
+
+static void
 parse_device_mapping(struct drm_i915_private *dev_priv,
 		       struct bdb_header *bdb)
 {
 	struct bdb_general_definitions *p_defs;
-	struct child_device_config *p_child, *child_dev_ptr;
+	union child_device_config *p_child, *child_dev_ptr;
 	int i, child_device_num, count;
 	u16	block_size;
 
@@ -601,7 +737,7 @@ parse_device_mapping(struct drm_i915_private *dev_priv,
 	/* get the number of child device that is present */
 	for (i = 0; i < child_device_num; i++) {
 		p_child = &(p_defs->devices[i]);
-		if (!p_child->device_type) {
+		if (!p_child->common.device_type) {
 			/* skip the device block if device type is invalid */
 			continue;
 		}
@@ -621,7 +757,7 @@ parse_device_mapping(struct drm_i915_private *dev_priv,
 	count = 0;
 	for (i = 0; i < child_device_num; i++) {
 		p_child = &(p_defs->devices[i]);
-		if (!p_child->device_type) {
+		if (!p_child->common.device_type) {
 			/* skip the device block if device type is invalid */
 			continue;
 		}
@@ -637,6 +773,7 @@ static void
 init_vbt_defaults(struct drm_i915_private *dev_priv)
 {
 	struct drm_device *dev = dev_priv->dev;
+	enum port port;
 
 	dev_priv->vbt.crt_ddc_pin = GMBUS_PORT_VGADDC;
 
@@ -655,6 +792,18 @@ init_vbt_defaults(struct drm_i915_private *dev_priv)
 	dev_priv->vbt.lvds_use_ssc = 1;
 	dev_priv->vbt.lvds_ssc_freq = intel_bios_ssc_frequency(dev, 1);
 	DRM_DEBUG_KMS("Set default to SSC at %dMHz\n", dev_priv->vbt.lvds_ssc_freq);
+
+	for (port = PORT_A; port < I915_MAX_PORTS; port++) {
+		struct ddi_vbt_port_info *info =
+			&dev_priv->vbt.ddi_port_info[port];
+
+		/* Recommended BSpec default: 800mV 0dB. */
+		info->hdmi_level_shift = 6;
+
+		info->supports_dvi = (port != PORT_A && port != PORT_E);
+		info->supports_hdmi = info->supports_dvi;
+		info->supports_dp = (port != PORT_E);
+	}
 }
 
 static int __init intel_no_opregion_vbt_callback(const struct dmi_system_id *id)
@@ -745,6 +894,8 @@ intel_parse_bios(struct drm_device *dev)
 	parse_device_mapping(dev_priv, bdb);
 	parse_driver_features(dev_priv, bdb);
 	parse_edp(dev_priv, bdb);
+	parse_mipi(dev_priv, bdb);
+	parse_ddi_ports(dev_priv, bdb);
 
 	if (bios)
 		pci_unmap_rom(pdev, bios);
diff --git a/drivers/gpu/drm/i915/intel_bios.h b/drivers/gpu/drm/i915/intel_bios.h
index e088d6f0956a..f580a2b0ddd3 100644
--- a/drivers/gpu/drm/i915/intel_bios.h
+++ b/drivers/gpu/drm/i915/intel_bios.h
@@ -104,6 +104,7 @@ struct vbios_data {
 #define BDB_LVDS_LFP_DATA	 42
 #define BDB_LVDS_BACKLIGHT	 43
 #define BDB_LVDS_POWER		 44
+#define BDB_MIPI		 50
 #define BDB_SKIP		254 /* VBIOS private block, ignore */
 
 struct bdb_general_features {
@@ -201,7 +202,10 @@ struct bdb_general_features {
 #define DEVICE_PORT_DVOB	0x01
 #define DEVICE_PORT_DVOC	0x02
 
-struct child_device_config {
+/* We used to keep this struct but without any version control. We should avoid
+ * using it in the future, but it should be safe to keep using it in the old
+ * code. */
+struct old_child_dev_config {
 	u16 handle;
 	u16 device_type;
 	u8  device_id[10]; /* ascii string */
@@ -223,6 +227,32 @@ struct child_device_config {
 	u8  dvo_function;
 } __attribute__((packed));
 
+/* This one contains field offsets that are known to be common for all BDB
+ * versions. Notice that the meaning of the contents contents may still change,
+ * but at least the offsets are consistent. */
+struct common_child_dev_config {
+	u16 handle;
+	u16 device_type;
+	u8 not_common1[12];
+	u8 dvo_port;
+	u8 not_common2[2];
+	u8 ddc_pin;
+	u16 edid_ptr;
+} __attribute__((packed));
+
+/* This field changes depending on the BDB version, so the most reliable way to
+ * read it is by checking the BDB version and reading the raw pointer. */
+union child_device_config {
+	/* This one is safe to be used anywhere, but the code should still check
+	 * the BDB version. */
+	u8 raw[33];
+	/* This one should only be kept for legacy code. */
+	struct old_child_dev_config old;
+	/* This one should also be safe to use anywhere, even without version
+	 * checks. */
+	struct common_child_dev_config common;
+};
+
 struct bdb_general_definitions {
 	/* DDC GPIO */
 	u8 crt_ddc_gmbus_pin;
@@ -248,7 +278,7 @@ struct bdb_general_definitions {
 	 * number = (block_size - sizeof(bdb_general_definitions))/
 	 *	     sizeof(child_device_config);
 	 */
-	struct child_device_config devices[0];
+	union child_device_config devices[0];
 } __attribute__((packed));
 
 struct bdb_lvds_options {
@@ -608,6 +638,40 @@ int intel_parse_bios(struct drm_device *dev);
 #define	 DEVICE_TYPE_DP		0x68C6
 #define	 DEVICE_TYPE_eDP	0x78C6
 
+#define  DEVICE_TYPE_CLASS_EXTENSION	(1 << 15)
+#define  DEVICE_TYPE_POWER_MANAGEMENT	(1 << 14)
+#define  DEVICE_TYPE_HOTPLUG_SIGNALING	(1 << 13)
+#define  DEVICE_TYPE_INTERNAL_CONNECTOR	(1 << 12)
+#define  DEVICE_TYPE_NOT_HDMI_OUTPUT	(1 << 11)
+#define  DEVICE_TYPE_MIPI_OUTPUT	(1 << 10)
+#define  DEVICE_TYPE_COMPOSITE_OUTPUT	(1 << 9)
+#define  DEVICE_TYPE_DUAL_CHANNEL	(1 << 8)
+#define  DEVICE_TYPE_HIGH_SPEED_LINK	(1 << 6)
+#define  DEVICE_TYPE_LVDS_SINGALING	(1 << 5)
+#define  DEVICE_TYPE_TMDS_DVI_SIGNALING	(1 << 4)
+#define  DEVICE_TYPE_VIDEO_SIGNALING	(1 << 3)
+#define  DEVICE_TYPE_DISPLAYPORT_OUTPUT	(1 << 2)
+#define  DEVICE_TYPE_DIGITAL_OUTPUT	(1 << 1)
+#define  DEVICE_TYPE_ANALOG_OUTPUT	(1 << 0)
+
+/*
+ * Bits we care about when checking for DEVICE_TYPE_eDP
+ * Depending on the system, the other bits may or may not
+ * be set for eDP outputs.
+ */
+#define DEVICE_TYPE_eDP_BITS \
+	(DEVICE_TYPE_INTERNAL_CONNECTOR | \
+	 DEVICE_TYPE_NOT_HDMI_OUTPUT | \
+	 DEVICE_TYPE_MIPI_OUTPUT | \
+	 DEVICE_TYPE_COMPOSITE_OUTPUT | \
+	 DEVICE_TYPE_DUAL_CHANNEL | \
+	 DEVICE_TYPE_LVDS_SINGALING | \
+	 DEVICE_TYPE_TMDS_DVI_SIGNALING | \
+	 DEVICE_TYPE_VIDEO_SIGNALING | \
+	 DEVICE_TYPE_DISPLAYPORT_OUTPUT | \
+	 DEVICE_TYPE_DIGITAL_OUTPUT | \
+	 DEVICE_TYPE_ANALOG_OUTPUT)
+
 /* define the DVO port for HDMI output type */
 #define		DVO_B		1
 #define		DVO_C		2
@@ -618,4 +682,57 @@ int intel_parse_bios(struct drm_device *dev);
 #define		PORT_IDPC	8
 #define		PORT_IDPD	9
 
+/* Possible values for the "DVO Port" field for versions >= 155: */
+#define DVO_PORT_HDMIA	0
+#define DVO_PORT_HDMIB	1
+#define DVO_PORT_HDMIC	2
+#define DVO_PORT_HDMID	3
+#define DVO_PORT_LVDS	4
+#define DVO_PORT_TV	5
+#define DVO_PORT_CRT	6
+#define DVO_PORT_DPB	7
+#define DVO_PORT_DPC	8
+#define DVO_PORT_DPD	9
+#define DVO_PORT_DPA	10
+
+/* MIPI DSI panel info */
+struct bdb_mipi {
+	u16 panel_id;
+	u16 bridge_revision;
+
+	/* General params */
+	u32 dithering:1;
+	u32 bpp_pixel_format:1;
+	u32 rsvd1:1;
+	u32 dphy_valid:1;
+	u32 resvd2:28;
+
+	u16 port_info;
+	u16 rsvd3:2;
+	u16 num_lanes:2;
+	u16 rsvd4:12;
+
+	/* DSI config */
+	u16 virt_ch_num:2;
+	u16 vtm:2;
+	u16 rsvd5:12;
+
+	u32 dsi_clock;
+	u32 bridge_ref_clk;
+	u16 rsvd_pwr;
+
+	/* Dphy Params */
+	u32 prepare_cnt:5;
+	u32 rsvd6:3;
+	u32 clk_zero_cnt:8;
+	u32 trail_cnt:5;
+	u32 rsvd7:3;
+	u32 exit_zero_cnt:6;
+	u32 rsvd8:2;
+
+	u32 hl_switch_cnt;
+	u32 lp_byte_clk;
+	u32 clk_lane_switch_cnt;
+} __attribute__((packed));
+
 #endif /* _I830_BIOS_H_ */
diff --git a/drivers/gpu/drm/i915/intel_crt.c b/drivers/gpu/drm/i915/intel_crt.c
index 10d1de5bce6f..b5b1b9b23adf 100644
--- a/drivers/gpu/drm/i915/intel_crt.c
+++ b/drivers/gpu/drm/i915/intel_crt.c
@@ -107,7 +107,17 @@ static unsigned int intel_crt_get_flags(struct intel_encoder *encoder)
 static void intel_crt_get_config(struct intel_encoder *encoder,
 				 struct intel_crtc_config *pipe_config)
 {
+	struct drm_device *dev = encoder->base.dev;
+	int dotclock;
+
 	pipe_config->adjusted_mode.flags |= intel_crt_get_flags(encoder);
+
+	dotclock = pipe_config->port_clock;
+
+	if (HAS_PCH_SPLIT(dev))
+		ironlake_check_encoder_dotclock(pipe_config, dotclock);
+
+	pipe_config->adjusted_mode.crtc_clock = dotclock;
 }
 
 static void hsw_crt_get_config(struct intel_encoder *encoder,
@@ -264,7 +274,7 @@ static void intel_crt_mode_set(struct intel_encoder *encoder)
 	struct drm_display_mode *adjusted_mode = &crtc->config.adjusted_mode;
 	u32 adpa;
 
-	if (HAS_PCH_SPLIT(dev))
+	if (INTEL_INFO(dev)->gen >= 5)
 		adpa = ADPA_HOTPLUG_BITS;
 	else
 		adpa = 0;
@@ -366,9 +376,6 @@ static bool valleyview_crt_detect_hotplug(struct drm_connector *connector)
 
 	DRM_DEBUG_KMS("valleyview hotplug adpa=0x%x, result %d\n", adpa, ret);
 
-	/* FIXME: debug force function and remove */
-	ret = true;
-
 	return ret;
 }
 
@@ -670,7 +677,6 @@ intel_crt_detect(struct drm_connector *connector, bool force)
 
 static void intel_crt_destroy(struct drm_connector *connector)
 {
-	drm_sysfs_connector_remove(connector);
 	drm_connector_cleanup(connector);
 	kfree(connector);
 }
@@ -776,7 +782,7 @@ void intel_crt_init(struct drm_device *dev)
 	if (!crt)
 		return;
 
-	intel_connector = kzalloc(sizeof(struct intel_connector), GFP_KERNEL);
+	intel_connector = kzalloc(sizeof(*intel_connector), GFP_KERNEL);
 	if (!intel_connector) {
 		kfree(crt);
 		return;
@@ -816,16 +822,15 @@ void intel_crt_init(struct drm_device *dev)
 	crt->base.mode_set = intel_crt_mode_set;
 	crt->base.disable = intel_disable_crt;
 	crt->base.enable = intel_enable_crt;
-	if (IS_HASWELL(dev))
-		crt->base.get_config = hsw_crt_get_config;
-	else
-		crt->base.get_config = intel_crt_get_config;
 	if (I915_HAS_HOTPLUG(dev))
 		crt->base.hpd_pin = HPD_CRT;
-	if (HAS_DDI(dev))
+	if (HAS_DDI(dev)) {
+		crt->base.get_config = hsw_crt_get_config;
 		crt->base.get_hw_state = intel_ddi_get_hw_state;
-	else
+	} else {
+		crt->base.get_config = intel_crt_get_config;
 		crt->base.get_hw_state = intel_crt_get_hw_state;
+	}
 	intel_connector->get_hw_state = intel_connector_get_hw_state;
 
 	drm_connector_helper_add(connector, &intel_crt_connector_helper_funcs);
diff --git a/drivers/gpu/drm/i915/intel_ddi.c b/drivers/gpu/drm/i915/intel_ddi.c
index b53fff84a7d5..1591576a6101 100644
--- a/drivers/gpu/drm/i915/intel_ddi.c
+++ b/drivers/gpu/drm/i915/intel_ddi.c
@@ -42,7 +42,6 @@ static const u32 hsw_ddi_translations_dp[] = {
 	0x80C30FFF, 0x000B0000,
 	0x00FFFFFF, 0x00040006,
 	0x80D75FFF, 0x000B0000,
-	0x00FFFFFF, 0x00040006		/* HDMI parameters */
 };
 
 static const u32 hsw_ddi_translations_fdi[] = {
@@ -55,10 +54,64 @@ static const u32 hsw_ddi_translations_fdi[] = {
 	0x00C30FFF, 0x001E0000,
 	0x00FFFFFF, 0x00060006,
 	0x00D75FFF, 0x001E0000,
-	0x00FFFFFF, 0x00040006		/* HDMI parameters */
 };
 
-static enum port intel_ddi_get_encoder_port(struct intel_encoder *intel_encoder)
+static const u32 hsw_ddi_translations_hdmi[] = {
+				/* Idx	NT mV diff	T mV diff	db  */
+	0x00FFFFFF, 0x0006000E, /* 0:	400		400		0   */
+	0x00E79FFF, 0x000E000C, /* 1:	400		500		2   */
+	0x00D75FFF, 0x0005000A, /* 2:	400		600		3.5 */
+	0x00FFFFFF, 0x0005000A, /* 3:	600		600		0   */
+	0x00E79FFF, 0x001D0007, /* 4:	600		750		2   */
+	0x00D75FFF, 0x000C0004, /* 5:	600		900		3.5 */
+	0x00FFFFFF, 0x00040006, /* 6:	800		800		0   */
+	0x80E79FFF, 0x00030002, /* 7:	800		1000		2   */
+	0x00FFFFFF, 0x00140005, /* 8:	850		850		0   */
+	0x00FFFFFF, 0x000C0004, /* 9:	900		900		0   */
+	0x00FFFFFF, 0x001C0003, /* 10:	950		950		0   */
+	0x80FFFFFF, 0x00030002, /* 11:	1000		1000		0   */
+};
+
+static const u32 bdw_ddi_translations_edp[] = {
+	0x00FFFFFF, 0x00000012,		/* DP parameters */
+	0x00EBAFFF, 0x00020011,
+	0x00C71FFF, 0x0006000F,
+	0x00FFFFFF, 0x00020011,
+	0x00DB6FFF, 0x0005000F,
+	0x00BEEFFF, 0x000A000C,
+	0x00FFFFFF, 0x0005000F,
+	0x00DB6FFF, 0x000A000C,
+	0x00FFFFFF, 0x000A000C,
+	0x00FFFFFF, 0x00140006		/* HDMI parameters 800mV 0dB*/
+};
+
+static const u32 bdw_ddi_translations_dp[] = {
+	0x00FFFFFF, 0x0007000E,		/* DP parameters */
+	0x00D75FFF, 0x000E000A,
+	0x00BEFFFF, 0x00140006,
+	0x00FFFFFF, 0x000E000A,
+	0x00D75FFF, 0x00180004,
+	0x80CB2FFF, 0x001B0002,
+	0x00F7DFFF, 0x00180004,
+	0x80D75FFF, 0x001B0002,
+	0x80FFFFFF, 0x001B0002,
+	0x00FFFFFF, 0x00140006		/* HDMI parameters 800mV 0dB*/
+};
+
+static const u32 bdw_ddi_translations_fdi[] = {
+	0x00FFFFFF, 0x0001000E,		/* FDI parameters */
+	0x00D75FFF, 0x0004000A,
+	0x00C30FFF, 0x00070006,
+	0x00AAAFFF, 0x000C0000,
+	0x00FFFFFF, 0x0004000A,
+	0x00D75FFF, 0x00090004,
+	0x00C30FFF, 0x000C0000,
+	0x00FFFFFF, 0x00070006,
+	0x00D75FFF, 0x000C0000,
+	0x00FFFFFF, 0x00140006		/* HDMI parameters 800mV 0dB*/
+};
+
+enum port intel_ddi_get_encoder_port(struct intel_encoder *intel_encoder)
 {
 	struct drm_encoder *encoder = &intel_encoder->base;
 	int type = intel_encoder->type;
@@ -78,8 +131,9 @@ static enum port intel_ddi_get_encoder_port(struct intel_encoder *intel_encoder)
 	}
 }
 
-/* On Haswell, DDI port buffers must be programmed with correct values
- * in advance. The buffer values are different for FDI and DP modes,
+/*
+ * Starting with Haswell, DDI port buffers must be programmed with correct
+ * values in advance. The buffer values are different for FDI and DP modes,
  * but the HDMI/DVI fields are shared among those. So we program the DDI
  * in either FDI or DP modes only, as HDMI connections will work with both
  * of those
@@ -89,15 +143,58 @@ static void intel_prepare_ddi_buffers(struct drm_device *dev, enum port port)
 	struct drm_i915_private *dev_priv = dev->dev_private;
 	u32 reg;
 	int i;
-	const u32 *ddi_translations = (port == PORT_E) ?
-		hsw_ddi_translations_fdi :
-		hsw_ddi_translations_dp;
+	int hdmi_level = dev_priv->vbt.ddi_port_info[port].hdmi_level_shift;
+	const u32 *ddi_translations_fdi;
+	const u32 *ddi_translations_dp;
+	const u32 *ddi_translations_edp;
+	const u32 *ddi_translations;
+
+	if (IS_BROADWELL(dev)) {
+		ddi_translations_fdi = bdw_ddi_translations_fdi;
+		ddi_translations_dp = bdw_ddi_translations_dp;
+		ddi_translations_edp = bdw_ddi_translations_edp;
+	} else if (IS_HASWELL(dev)) {
+		ddi_translations_fdi = hsw_ddi_translations_fdi;
+		ddi_translations_dp = hsw_ddi_translations_dp;
+		ddi_translations_edp = hsw_ddi_translations_dp;
+	} else {
+		WARN(1, "ddi translation table missing\n");
+		ddi_translations_edp = bdw_ddi_translations_dp;
+		ddi_translations_fdi = bdw_ddi_translations_fdi;
+		ddi_translations_dp = bdw_ddi_translations_dp;
+	}
+
+	switch (port) {
+	case PORT_A:
+		ddi_translations = ddi_translations_edp;
+		break;
+	case PORT_B:
+	case PORT_C:
+		ddi_translations = ddi_translations_dp;
+		break;
+	case PORT_D:
+		if (intel_dpd_is_edp(dev))
+			ddi_translations = ddi_translations_edp;
+		else
+			ddi_translations = ddi_translations_dp;
+		break;
+	case PORT_E:
+		ddi_translations = ddi_translations_fdi;
+		break;
+	default:
+		BUG();
+	}
 
 	for (i = 0, reg = DDI_BUF_TRANS(port);
 	     i < ARRAY_SIZE(hsw_ddi_translations_fdi); i++) {
 		I915_WRITE(reg, ddi_translations[i]);
 		reg += 4;
 	}
+	/* Entry 9 is for HDMI: */
+	for (i = 0; i < 2; i++) {
+		I915_WRITE(reg, hsw_ddi_translations_hdmi[hdmi_level * 2 + i]);
+		reg += 4;
+	}
 }
 
 /* Program DDI buffers translations for DP. By default, program ports A-D in DP
@@ -296,9 +393,6 @@ static void intel_ddi_mode_set(struct intel_encoder *encoder)
 			DRM_DEBUG_DRIVER("DP audio: write eld information\n");
 			intel_write_eld(&encoder->base, adjusted_mode);
 		}
-
-		intel_dp_init_link_config(intel_dp);
-
 	} else if (type == INTEL_OUTPUT_HDMI) {
 		struct intel_hdmi *intel_hdmi = enc_to_intel_hdmi(&encoder->base);
 
@@ -739,7 +833,8 @@ void intel_ddi_enable_transcoder_func(struct drm_crtc *crtc)
 	struct intel_crtc *intel_crtc = to_intel_crtc(crtc);
 	struct intel_encoder *intel_encoder = intel_ddi_get_crtc_encoder(crtc);
 	struct drm_encoder *encoder = &intel_encoder->base;
-	struct drm_i915_private *dev_priv = crtc->dev->dev_private;
+	struct drm_device *dev = crtc->dev;
+	struct drm_i915_private *dev_priv = dev->dev_private;
 	enum pipe pipe = intel_crtc->pipe;
 	enum transcoder cpu_transcoder = intel_crtc->config.cpu_transcoder;
 	enum port port = intel_ddi_get_encoder_port(intel_encoder);
@@ -767,18 +862,19 @@ void intel_ddi_enable_transcoder_func(struct drm_crtc *crtc)
 		BUG();
 	}
 
-	if (crtc->mode.flags & DRM_MODE_FLAG_PVSYNC)
+	if (intel_crtc->config.adjusted_mode.flags & DRM_MODE_FLAG_PVSYNC)
 		temp |= TRANS_DDI_PVSYNC;
-	if (crtc->mode.flags & DRM_MODE_FLAG_PHSYNC)
+	if (intel_crtc->config.adjusted_mode.flags & DRM_MODE_FLAG_PHSYNC)
 		temp |= TRANS_DDI_PHSYNC;
 
 	if (cpu_transcoder == TRANSCODER_EDP) {
 		switch (pipe) {
 		case PIPE_A:
-			/* Can only use the always-on power well for eDP when
-			 * not using the panel fitter, and when not using motion
-			  * blur mitigation (which we don't support). */
-			if (intel_crtc->config.pch_pfit.enabled)
+			/* On Haswell, can only use the always-on power well for
+			 * eDP when not using the panel fitter, and when not
+			 * using motion blur mitigation (which we don't
+			 * support). */
+			if (IS_HASWELL(dev) && intel_crtc->config.pch_pfit.enabled)
 				temp |= TRANS_DDI_EDP_INPUT_A_ONOFF;
 			else
 				temp |= TRANS_DDI_EDP_INPUT_A_ON;
@@ -1139,18 +1235,29 @@ static void intel_disable_ddi(struct intel_encoder *intel_encoder)
 
 int intel_ddi_get_cdclk_freq(struct drm_i915_private *dev_priv)
 {
+	struct drm_device *dev = dev_priv->dev;
 	uint32_t lcpll = I915_READ(LCPLL_CTL);
+	uint32_t freq = lcpll & LCPLL_CLK_FREQ_MASK;
 
-	if (lcpll & LCPLL_CD_SOURCE_FCLK)
+	if (lcpll & LCPLL_CD_SOURCE_FCLK) {
 		return 800000;
-	else if (I915_READ(HSW_FUSE_STRAP) & HSW_CDCLK_LIMIT)
+	} else if (I915_READ(HSW_FUSE_STRAP) & HSW_CDCLK_LIMIT) {
 		return 450000;
-	else if ((lcpll & LCPLL_CLK_FREQ_MASK) == LCPLL_CLK_FREQ_450)
+	} else if (freq == LCPLL_CLK_FREQ_450) {
 		return 450000;
-	else if (IS_ULT(dev_priv->dev))
-		return 337500;
-	else
-		return 540000;
+	} else if (IS_HASWELL(dev)) {
+		if (IS_ULT(dev))
+			return 337500;
+		else
+			return 540000;
+	} else {
+		if (freq == LCPLL_CLK_FREQ_54O_BDW)
+			return 540000;
+		else if (freq == LCPLL_CLK_FREQ_337_5_BDW)
+			return 337500;
+		else
+			return 675000;
+	}
 }
 
 void intel_ddi_pll_init(struct drm_device *dev)
@@ -1202,7 +1309,7 @@ void intel_ddi_prepare_link_retrain(struct drm_encoder *encoder)
 
 	val = DP_TP_CTL_ENABLE | DP_TP_CTL_MODE_SST |
 	      DP_TP_CTL_LINK_TRAIN_PAT1 | DP_TP_CTL_SCRAMBLE_DISABLE;
-	if (intel_dp->link_configuration[1] & DP_LANE_COUNT_ENHANCED_FRAME_EN)
+	if (drm_dp_enhanced_frame_cap(intel_dp->dpcd))
 		val |= DP_TP_CTL_ENHANCED_FRAME_ENABLE;
 	I915_WRITE(DP_TP_CTL(port), val);
 	POSTING_READ(DP_TP_CTL(port));
@@ -1285,6 +1392,20 @@ void intel_ddi_get_config(struct intel_encoder *encoder,
 	default:
 		break;
 	}
+
+	switch (temp & TRANS_DDI_MODE_SELECT_MASK) {
+	case TRANS_DDI_MODE_SELECT_HDMI:
+	case TRANS_DDI_MODE_SELECT_DVI:
+	case TRANS_DDI_MODE_SELECT_FDI:
+		break;
+	case TRANS_DDI_MODE_SELECT_DP_SST:
+	case TRANS_DDI_MODE_SELECT_DP_MST:
+		pipe_config->has_dp_encoder = true;
+		intel_dp_get_m_n(intel_crtc, pipe_config);
+		break;
+	default:
+		break;
+	}
 }
 
 static void intel_ddi_destroy(struct drm_encoder *encoder)
@@ -1314,6 +1435,41 @@ static const struct drm_encoder_funcs intel_ddi_funcs = {
 	.destroy = intel_ddi_destroy,
 };
 
+static struct intel_connector *
+intel_ddi_init_dp_connector(struct intel_digital_port *intel_dig_port)
+{
+	struct intel_connector *connector;
+	enum port port = intel_dig_port->port;
+
+	connector = kzalloc(sizeof(*connector), GFP_KERNEL);
+	if (!connector)
+		return NULL;
+
+	intel_dig_port->dp.output_reg = DDI_BUF_CTL(port);
+	if (!intel_dp_init_connector(intel_dig_port, connector)) {
+		kfree(connector);
+		return NULL;
+	}
+
+	return connector;
+}
+
+static struct intel_connector *
+intel_ddi_init_hdmi_connector(struct intel_digital_port *intel_dig_port)
+{
+	struct intel_connector *connector;
+	enum port port = intel_dig_port->port;
+
+	connector = kzalloc(sizeof(*connector), GFP_KERNEL);
+	if (!connector)
+		return NULL;
+
+	intel_dig_port->hdmi.hdmi_reg = DDI_BUF_CTL(port);
+	intel_hdmi_init_connector(intel_dig_port, connector);
+
+	return connector;
+}
+
 void intel_ddi_init(struct drm_device *dev, enum port port)
 {
 	struct drm_i915_private *dev_priv = dev->dev_private;
@@ -1322,17 +1478,22 @@ void intel_ddi_init(struct drm_device *dev, enum port port)
 	struct drm_encoder *encoder;
 	struct intel_connector *hdmi_connector = NULL;
 	struct intel_connector *dp_connector = NULL;
+	bool init_hdmi, init_dp;
+
+	init_hdmi = (dev_priv->vbt.ddi_port_info[port].supports_dvi ||
+		     dev_priv->vbt.ddi_port_info[port].supports_hdmi);
+	init_dp = dev_priv->vbt.ddi_port_info[port].supports_dp;
+	if (!init_dp && !init_hdmi) {
+		DRM_DEBUG_KMS("VBT says port %c is not DVI/HDMI/DP compatible\n",
+			      port_name(port));
+		init_hdmi = true;
+		init_dp = true;
+	}
 
-	intel_dig_port = kzalloc(sizeof(struct intel_digital_port), GFP_KERNEL);
+	intel_dig_port = kzalloc(sizeof(*intel_dig_port), GFP_KERNEL);
 	if (!intel_dig_port)
 		return;
 
-	dp_connector = kzalloc(sizeof(struct intel_connector), GFP_KERNEL);
-	if (!dp_connector) {
-		kfree(intel_dig_port);
-		return;
-	}
-
 	intel_encoder = &intel_dig_port->base;
 	encoder = &intel_encoder->base;
 
@@ -1352,28 +1513,22 @@ void intel_ddi_init(struct drm_device *dev, enum port port)
 	intel_dig_port->saved_port_bits = I915_READ(DDI_BUF_CTL(port)) &
 					  (DDI_BUF_PORT_REVERSAL |
 					   DDI_A_4_LANES);
-	intel_dig_port->dp.output_reg = DDI_BUF_CTL(port);
 
 	intel_encoder->type = INTEL_OUTPUT_UNKNOWN;
 	intel_encoder->crtc_mask =  (1 << 0) | (1 << 1) | (1 << 2);
 	intel_encoder->cloneable = false;
 	intel_encoder->hot_plug = intel_ddi_hot_plug;
 
-	if (!intel_dp_init_connector(intel_dig_port, dp_connector)) {
-		drm_encoder_cleanup(encoder);
-		kfree(intel_dig_port);
-		kfree(dp_connector);
-		return;
-	}
+	if (init_dp)
+		dp_connector = intel_ddi_init_dp_connector(intel_dig_port);
 
-	if (intel_encoder->type != INTEL_OUTPUT_EDP) {
-		hdmi_connector = kzalloc(sizeof(struct intel_connector),
-					 GFP_KERNEL);
-		if (!hdmi_connector) {
-			return;
-		}
+	/* In theory we don't need the encoder->type check, but leave it just in
+	 * case we have some really bad VBTs... */
+	if (intel_encoder->type != INTEL_OUTPUT_EDP && init_hdmi)
+		hdmi_connector = intel_ddi_init_hdmi_connector(intel_dig_port);
 
-		intel_dig_port->hdmi.hdmi_reg = DDI_BUF_CTL(port);
-		intel_hdmi_init_connector(intel_dig_port, hdmi_connector);
+	if (!dp_connector && !hdmi_connector) {
+		drm_encoder_cleanup(encoder);
+		kfree(intel_dig_port);
 	}
 }
diff --git a/drivers/gpu/drm/i915/intel_display.c b/drivers/gpu/drm/i915/intel_display.c
index d78d33f9337d..3cddd508d110 100644
--- a/drivers/gpu/drm/i915/intel_display.c
+++ b/drivers/gpu/drm/i915/intel_display.c
@@ -41,14 +41,13 @@
 #include <drm/drm_crtc_helper.h>
 #include <linux/dma_remapping.h>
 
-bool intel_pipe_has_type(struct drm_crtc *crtc, int type);
 static void intel_increase_pllclock(struct drm_crtc *crtc);
 static void intel_crtc_update_cursor(struct drm_crtc *crtc, bool on);
 
 static void i9xx_crtc_clock_get(struct intel_crtc *crtc,
 				struct intel_crtc_config *pipe_config);
-static void ironlake_crtc_clock_get(struct intel_crtc *crtc,
-				    struct intel_crtc_config *pipe_config);
+static void ironlake_pch_clock_get(struct intel_crtc *crtc,
+				   struct intel_crtc_config *pipe_config);
 
 static int intel_set_mode(struct drm_crtc *crtc, struct drm_display_mode *mode,
 			  int x, int y, struct drm_framebuffer *old_fb);
@@ -69,9 +68,6 @@ struct intel_limit {
 	intel_p2_t	    p2;
 };
 
-/* FDI */
-#define IRONLAKE_FDI_FREQ		2700000 /* in kHz for mode->clock */
-
 int
 intel_pch_rawclk(struct drm_device *dev)
 {
@@ -313,44 +309,44 @@ static const intel_limit_t intel_limits_ironlake_dual_lvds_100m = {
 		.p2_slow = 7, .p2_fast = 7 },
 };
 
-static const intel_limit_t intel_limits_vlv_dac = {
-	.dot = { .min = 25000, .max = 270000 },
-	.vco = { .min = 4000000, .max = 6000000 },
-	.n = { .min = 1, .max = 7 },
-	.m = { .min = 22, .max = 450 }, /* guess */
-	.m1 = { .min = 2, .max = 3 },
-	.m2 = { .min = 11, .max = 156 },
-	.p = { .min = 10, .max = 30 },
-	.p1 = { .min = 1, .max = 3 },
-	.p2 = { .dot_limit = 270000,
-		.p2_slow = 2, .p2_fast = 20 },
-};
-
-static const intel_limit_t intel_limits_vlv_hdmi = {
-	.dot = { .min = 25000, .max = 270000 },
+static const intel_limit_t intel_limits_vlv = {
+	 /*
+	  * These are the data rate limits (measured in fast clocks)
+	  * since those are the strictest limits we have. The fast
+	  * clock and actual rate limits are more relaxed, so checking
+	  * them would make no difference.
+	  */
+	.dot = { .min = 25000 * 5, .max = 270000 * 5 },
 	.vco = { .min = 4000000, .max = 6000000 },
 	.n = { .min = 1, .max = 7 },
-	.m = { .min = 60, .max = 300 }, /* guess */
 	.m1 = { .min = 2, .max = 3 },
 	.m2 = { .min = 11, .max = 156 },
-	.p = { .min = 10, .max = 30 },
 	.p1 = { .min = 2, .max = 3 },
-	.p2 = { .dot_limit = 270000,
-		.p2_slow = 2, .p2_fast = 20 },
+	.p2 = { .p2_slow = 2, .p2_fast = 20 }, /* slow=min, fast=max */
 };
 
-static const intel_limit_t intel_limits_vlv_dp = {
-	.dot = { .min = 25000, .max = 270000 },
-	.vco = { .min = 4000000, .max = 6000000 },
-	.n = { .min = 1, .max = 7 },
-	.m = { .min = 22, .max = 450 },
-	.m1 = { .min = 2, .max = 3 },
-	.m2 = { .min = 11, .max = 156 },
-	.p = { .min = 10, .max = 30 },
-	.p1 = { .min = 1, .max = 3 },
-	.p2 = { .dot_limit = 270000,
-		.p2_slow = 2, .p2_fast = 20 },
-};
+static void vlv_clock(int refclk, intel_clock_t *clock)
+{
+	clock->m = clock->m1 * clock->m2;
+	clock->p = clock->p1 * clock->p2;
+	clock->vco = DIV_ROUND_CLOSEST(refclk * clock->m, clock->n);
+	clock->dot = DIV_ROUND_CLOSEST(clock->vco, clock->p);
+}
+
+/**
+ * Returns whether any output on the specified pipe is of the specified type
+ */
+static bool intel_pipe_has_type(struct drm_crtc *crtc, int type)
+{
+	struct drm_device *dev = crtc->dev;
+	struct intel_encoder *encoder;
+
+	for_each_encoder_on_crtc(dev, crtc, encoder)
+		if (encoder->type == type)
+			return true;
+
+	return false;
+}
 
 static const intel_limit_t *intel_ironlake_limit(struct drm_crtc *crtc,
 						int refclk)
@@ -412,12 +408,7 @@ static const intel_limit_t *intel_limit(struct drm_crtc *crtc, int refclk)
 		else
 			limit = &intel_limits_pineview_sdvo;
 	} else if (IS_VALLEYVIEW(dev)) {
-		if (intel_pipe_has_type(crtc, INTEL_OUTPUT_ANALOG))
-			limit = &intel_limits_vlv_dac;
-		else if (intel_pipe_has_type(crtc, INTEL_OUTPUT_HDMI))
-			limit = &intel_limits_vlv_hdmi;
-		else
-			limit = &intel_limits_vlv_dp;
+		limit = &intel_limits_vlv;
 	} else if (!IS_GEN2(dev)) {
 		if (intel_pipe_has_type(crtc, INTEL_OUTPUT_LVDS))
 			limit = &intel_limits_i9xx_lvds;
@@ -439,8 +430,8 @@ static void pineview_clock(int refclk, intel_clock_t *clock)
 {
 	clock->m = clock->m2 + 2;
 	clock->p = clock->p1 * clock->p2;
-	clock->vco = refclk * clock->m / clock->n;
-	clock->dot = clock->vco / clock->p;
+	clock->vco = DIV_ROUND_CLOSEST(refclk * clock->m, clock->n);
+	clock->dot = DIV_ROUND_CLOSEST(clock->vco, clock->p);
 }
 
 static uint32_t i9xx_dpll_compute_m(struct dpll *dpll)
@@ -452,23 +443,8 @@ static void i9xx_clock(int refclk, intel_clock_t *clock)
 {
 	clock->m = i9xx_dpll_compute_m(clock);
 	clock->p = clock->p1 * clock->p2;
-	clock->vco = refclk * clock->m / (clock->n + 2);
-	clock->dot = clock->vco / clock->p;
-}
-
-/**
- * Returns whether any output on the specified pipe is of the specified type
- */
-bool intel_pipe_has_type(struct drm_crtc *crtc, int type)
-{
-	struct drm_device *dev = crtc->dev;
-	struct intel_encoder *encoder;
-
-	for_each_encoder_on_crtc(dev, crtc, encoder)
-		if (encoder->type == type)
-			return true;
-
-	return false;
+	clock->vco = DIV_ROUND_CLOSEST(refclk * clock->m, clock->n + 2);
+	clock->dot = DIV_ROUND_CLOSEST(clock->vco, clock->p);
 }
 
 #define INTELPllInvalid(s)   do { /* DRM_DEBUG(s); */ return false; } while (0)
@@ -481,20 +457,26 @@ static bool intel_PLL_is_valid(struct drm_device *dev,
 			       const intel_limit_t *limit,
 			       const intel_clock_t *clock)
 {
+	if (clock->n   < limit->n.min   || limit->n.max   < clock->n)
+		INTELPllInvalid("n out of range\n");
 	if (clock->p1  < limit->p1.min  || limit->p1.max  < clock->p1)
 		INTELPllInvalid("p1 out of range\n");
-	if (clock->p   < limit->p.min   || limit->p.max   < clock->p)
-		INTELPllInvalid("p out of range\n");
 	if (clock->m2  < limit->m2.min  || limit->m2.max  < clock->m2)
 		INTELPllInvalid("m2 out of range\n");
 	if (clock->m1  < limit->m1.min  || limit->m1.max  < clock->m1)
 		INTELPllInvalid("m1 out of range\n");
-	if (clock->m1 <= clock->m2 && !IS_PINEVIEW(dev))
-		INTELPllInvalid("m1 <= m2\n");
-	if (clock->m   < limit->m.min   || limit->m.max   < clock->m)
-		INTELPllInvalid("m out of range\n");
-	if (clock->n   < limit->n.min   || limit->n.max   < clock->n)
-		INTELPllInvalid("n out of range\n");
+
+	if (!IS_PINEVIEW(dev) && !IS_VALLEYVIEW(dev))
+		if (clock->m1 <= clock->m2)
+			INTELPllInvalid("m1 <= m2\n");
+
+	if (!IS_VALLEYVIEW(dev)) {
+		if (clock->p < limit->p.min || limit->p.max < clock->p)
+			INTELPllInvalid("p out of range\n");
+		if (clock->m < limit->m.min || limit->m.max < clock->m)
+			INTELPllInvalid("m out of range\n");
+	}
+
 	if (clock->vco < limit->vco.min || limit->vco.max < clock->vco)
 		INTELPllInvalid("vco out of range\n");
 	/* XXX: We may need to be checking "Dot clock" depending on the multiplier,
@@ -688,67 +670,73 @@ vlv_find_best_dpll(const intel_limit_t *limit, struct drm_crtc *crtc,
 		   int target, int refclk, intel_clock_t *match_clock,
 		   intel_clock_t *best_clock)
 {
-	u32 p1, p2, m1, m2, vco, bestn, bestm1, bestm2, bestp1, bestp2;
-	u32 m, n, fastclk;
-	u32 updrate, minupdate, p;
-	unsigned long bestppm, ppm, absppm;
-	int dotclk, flag;
-
-	flag = 0;
-	dotclk = target * 1000;
-	bestppm = 1000000;
-	ppm = absppm = 0;
-	fastclk = dotclk / (2*100);
-	updrate = 0;
-	minupdate = 19200;
-	n = p = p1 = p2 = m = m1 = m2 = vco = bestn = 0;
-	bestm1 = bestm2 = bestp1 = bestp2 = 0;
+	struct drm_device *dev = crtc->dev;
+	intel_clock_t clock;
+	unsigned int bestppm = 1000000;
+	/* min update 19.2 MHz */
+	int max_n = min(limit->n.max, refclk / 19200);
+	bool found = false;
+
+	target *= 5; /* fast clock */
+
+	memset(best_clock, 0, sizeof(*best_clock));
 
 	/* based on hardware requirement, prefer smaller n to precision */
-	for (n = limit->n.min; n <= ((refclk) / minupdate); n++) {
-		updrate = refclk / n;
-		for (p1 = limit->p1.max; p1 > limit->p1.min; p1--) {
-			for (p2 = limit->p2.p2_fast+1; p2 > 0; p2--) {
-				if (p2 > 10)
-					p2 = p2 - 1;
-				p = p1 * p2;
+	for (clock.n = limit->n.min; clock.n <= max_n; clock.n++) {
+		for (clock.p1 = limit->p1.max; clock.p1 >= limit->p1.min; clock.p1--) {
+			for (clock.p2 = limit->p2.p2_fast; clock.p2 >= limit->p2.p2_slow;
+			     clock.p2 -= clock.p2 > 10 ? 2 : 1) {
+				clock.p = clock.p1 * clock.p2;
 				/* based on hardware requirement, prefer bigger m1,m2 values */
-				for (m1 = limit->m1.min; m1 <= limit->m1.max; m1++) {
-					m2 = (((2*(fastclk * p * n / m1 )) +
-					       refclk) / (2*refclk));
-					m = m1 * m2;
-					vco = updrate * m;
-					if (vco >= limit->vco.min && vco < limit->vco.max) {
-						ppm = 1000000 * ((vco / p) - fastclk) / fastclk;
-						absppm = (ppm > 0) ? ppm : (-ppm);
-						if (absppm < 100 && ((p1 * p2) > (bestp1 * bestp2))) {
-							bestppm = 0;
-							flag = 1;
-						}
-						if (absppm < bestppm - 10) {
-							bestppm = absppm;
-							flag = 1;
-						}
-						if (flag) {
-							bestn = n;
-							bestm1 = m1;
-							bestm2 = m2;
-							bestp1 = p1;
-							bestp2 = p2;
-							flag = 0;
-						}
+				for (clock.m1 = limit->m1.min; clock.m1 <= limit->m1.max; clock.m1++) {
+					unsigned int ppm, diff;
+
+					clock.m2 = DIV_ROUND_CLOSEST(target * clock.p * clock.n,
+								     refclk * clock.m1);
+
+					vlv_clock(refclk, &clock);
+
+					if (!intel_PLL_is_valid(dev, limit,
+								&clock))
+						continue;
+
+					diff = abs(clock.dot - target);
+					ppm = div_u64(1000000ULL * diff, target);
+
+					if (ppm < 100 && clock.p > best_clock->p) {
+						bestppm = 0;
+						*best_clock = clock;
+						found = true;
+					}
+
+					if (bestppm >= 10 && ppm < bestppm - 10) {
+						bestppm = ppm;
+						*best_clock = clock;
+						found = true;
 					}
 				}
 			}
 		}
 	}
-	best_clock->n = bestn;
-	best_clock->m1 = bestm1;
-	best_clock->m2 = bestm2;
-	best_clock->p1 = bestp1;
-	best_clock->p2 = bestp2;
 
-	return true;
+	return found;
+}
+
+bool intel_crtc_active(struct drm_crtc *crtc)
+{
+	struct intel_crtc *intel_crtc = to_intel_crtc(crtc);
+
+	/* Be paranoid as we can arrive here with only partial
+	 * state retrieved from the hardware during setup.
+	 *
+	 * We can ditch the adjusted_mode.crtc_clock check as soon
+	 * as Haswell has gained clock readout/fastboot support.
+	 *
+	 * We can ditch the crtc->fb check as soon as we can
+	 * properly reconstruct framebuffers.
+	 */
+	return intel_crtc->active && crtc->fb &&
+		intel_crtc->config.adjusted_mode.crtc_clock;
 }
 
 enum transcoder intel_pipe_to_cpu_transcoder(struct drm_i915_private *dev_priv,
@@ -812,6 +800,25 @@ void intel_wait_for_vblank(struct drm_device *dev, int pipe)
 		DRM_DEBUG_KMS("vblank wait timed out\n");
 }
 
+static bool pipe_dsl_stopped(struct drm_device *dev, enum pipe pipe)
+{
+	struct drm_i915_private *dev_priv = dev->dev_private;
+	u32 reg = PIPEDSL(pipe);
+	u32 line1, line2;
+	u32 line_mask;
+
+	if (IS_GEN2(dev))
+		line_mask = DSL_LINEMASK_GEN2;
+	else
+		line_mask = DSL_LINEMASK_GEN3;
+
+	line1 = I915_READ(reg) & line_mask;
+	mdelay(5);
+	line2 = I915_READ(reg) & line_mask;
+
+	return line1 == line2;
+}
+
 /*
  * intel_wait_for_pipe_off - wait for pipe to turn off
  * @dev: drm device
@@ -843,22 +850,8 @@ void intel_wait_for_pipe_off(struct drm_device *dev, int pipe)
 			     100))
 			WARN(1, "pipe_off wait timed out\n");
 	} else {
-		u32 last_line, line_mask;
-		int reg = PIPEDSL(pipe);
-		unsigned long timeout = jiffies + msecs_to_jiffies(100);
-
-		if (IS_GEN2(dev))
-			line_mask = DSL_LINEMASK_GEN2;
-		else
-			line_mask = DSL_LINEMASK_GEN3;
-
 		/* Wait for the display line to settle */
-		do {
-			last_line = I915_READ(reg) & line_mask;
-			mdelay(5);
-		} while (((I915_READ(reg) & line_mask) != last_line) &&
-			 time_after(timeout, jiffies));
-		if (time_after(jiffies, timeout))
+		if (wait_for(pipe_dsl_stopped(dev, pipe), 100))
 			WARN(1, "pipe_off wait timed out\n");
 	}
 }
@@ -929,6 +922,24 @@ void assert_pll(struct drm_i915_private *dev_priv,
 	     state_string(state), state_string(cur_state));
 }
 
+/* XXX: the dsi pll is shared between MIPI DSI ports */
+static void assert_dsi_pll(struct drm_i915_private *dev_priv, bool state)
+{
+	u32 val;
+	bool cur_state;
+
+	mutex_lock(&dev_priv->dpio_lock);
+	val = vlv_cck_read(dev_priv, CCK_REG_DSI_PLL_CONTROL);
+	mutex_unlock(&dev_priv->dpio_lock);
+
+	cur_state = val & DSI_PLL_VCO_EN;
+	WARN(cur_state != state,
+	     "DSI PLL state assertion failure (expected %s, current %s)\n",
+	     state_string(state), state_string(cur_state));
+}
+#define assert_dsi_pll_enabled(d) assert_dsi_pll(d, true)
+#define assert_dsi_pll_disabled(d) assert_dsi_pll(d, false)
+
 struct intel_shared_dpll *
 intel_crtc_to_shared_dpll(struct intel_crtc *crtc)
 {
@@ -1069,6 +1080,26 @@ static void assert_panel_unlocked(struct drm_i915_private *dev_priv,
 	     pipe_name(pipe));
 }
 
+static void assert_cursor(struct drm_i915_private *dev_priv,
+			  enum pipe pipe, bool state)
+{
+	struct drm_device *dev = dev_priv->dev;
+	bool cur_state;
+
+	if (IS_IVYBRIDGE(dev) || IS_HASWELL(dev))
+		cur_state = I915_READ(CURCNTR_IVB(pipe)) & CURSOR_MODE;
+	else if (IS_845G(dev) || IS_I865G(dev))
+		cur_state = I915_READ(_CURACNTR) & CURSOR_ENABLE;
+	else
+		cur_state = I915_READ(CURCNTR(pipe)) & CURSOR_MODE;
+
+	WARN(cur_state != state,
+	     "cursor on pipe %c assertion failure (expected %s, current %s)\n",
+	     pipe_name(pipe), state_string(state), state_string(cur_state));
+}
+#define assert_cursor_enabled(d, p) assert_cursor(d, p, true)
+#define assert_cursor_disabled(d, p) assert_cursor(d, p, false)
+
 void assert_pipe(struct drm_i915_private *dev_priv,
 		 enum pipe pipe, bool state)
 {
@@ -1323,6 +1354,26 @@ static void assert_pch_ports_disabled(struct drm_i915_private *dev_priv,
 	assert_pch_hdmi_disabled(dev_priv, pipe, PCH_HDMID);
 }
 
+static void intel_init_dpio(struct drm_device *dev)
+{
+	struct drm_i915_private *dev_priv = dev->dev_private;
+
+	if (!IS_VALLEYVIEW(dev))
+		return;
+
+	/*
+	 * From VLV2A0_DP_eDP_DPIO_driver_vbios_notes_10.docx -
+	 *  6.	De-assert cmn_reset/side_reset. Same as VLV X0.
+	 *   a.	GUnit 0x2110 bit[0] set to 1 (def 0)
+	 *   b.	The other bits such as sfr settings / modesel may all be set
+	 *      to 0.
+	 *
+	 * This should only be done on init and resume from S3 with both
+	 * PLLs disabled, or we risk losing DPIO and PLL synchronization.
+	 */
+	I915_WRITE(DPIO_CTL, I915_READ(DPIO_CTL) | DPIO_CMNRST);
+}
+
 static void vlv_enable_pll(struct intel_crtc *crtc)
 {
 	struct drm_device *dev = crtc->base.dev;
@@ -1429,6 +1480,20 @@ static void i9xx_disable_pll(struct drm_i915_private *dev_priv, enum pipe pipe)
 	POSTING_READ(DPLL(pipe));
 }
 
+static void vlv_disable_pll(struct drm_i915_private *dev_priv, enum pipe pipe)
+{
+	u32 val = 0;
+
+	/* Make sure the pipe isn't still relying on us */
+	assert_pipe_disabled(dev_priv, pipe);
+
+	/* Leave integrated clock source enabled */
+	if (pipe == PIPE_B)
+		val = DPLL_INTEGRATED_CRI_CLK_VLV;
+	I915_WRITE(DPLL(pipe), val);
+	POSTING_READ(DPLL(pipe));
+}
+
 void vlv_wait_port_ready(struct drm_i915_private *dev_priv, int port)
 {
 	u32 port_mask;
@@ -1661,7 +1726,7 @@ static void lpt_disable_pch_transcoder(struct drm_i915_private *dev_priv)
  * returning.
  */
 static void intel_enable_pipe(struct drm_i915_private *dev_priv, enum pipe pipe,
-			      bool pch_port)
+			      bool pch_port, bool dsi)
 {
 	enum transcoder cpu_transcoder = intel_pipe_to_cpu_transcoder(dev_priv,
 								      pipe);
@@ -1670,6 +1735,7 @@ static void intel_enable_pipe(struct drm_i915_private *dev_priv, enum pipe pipe,
 	u32 val;
 
 	assert_planes_disabled(dev_priv, pipe);
+	assert_cursor_disabled(dev_priv, pipe);
 	assert_sprites_disabled(dev_priv, pipe);
 
 	if (HAS_PCH_LPT(dev_priv->dev))
@@ -1683,7 +1749,10 @@ static void intel_enable_pipe(struct drm_i915_private *dev_priv, enum pipe pipe,
 	 * need the check.
 	 */
 	if (!HAS_PCH_SPLIT(dev_priv->dev))
-		assert_pll_enabled(dev_priv, pipe);
+		if (dsi)
+			assert_dsi_pll_enabled(dev_priv);
+		else
+			assert_pll_enabled(dev_priv, pipe);
 	else {
 		if (pch_port) {
 			/* if driving the PCH, we need FDI enabled */
@@ -1728,6 +1797,7 @@ static void intel_disable_pipe(struct drm_i915_private *dev_priv,
 	 * or we might hang the display.
 	 */
 	assert_planes_disabled(dev_priv, pipe);
+	assert_cursor_disabled(dev_priv, pipe);
 	assert_sprites_disabled(dev_priv, pipe);
 
 	/* Don't disable pipe A or pipe A PLLs if needed */
@@ -1747,63 +1817,75 @@ static void intel_disable_pipe(struct drm_i915_private *dev_priv,
  * Plane regs are double buffered, going from enabled->disabled needs a
  * trigger in order to latch.  The display address reg provides this.
  */
-void intel_flush_display_plane(struct drm_i915_private *dev_priv,
-				      enum plane plane)
+void intel_flush_primary_plane(struct drm_i915_private *dev_priv,
+			       enum plane plane)
 {
-	if (dev_priv->info->gen >= 4)
-		I915_WRITE(DSPSURF(plane), I915_READ(DSPSURF(plane)));
-	else
-		I915_WRITE(DSPADDR(plane), I915_READ(DSPADDR(plane)));
+	u32 reg = dev_priv->info->gen >= 4 ? DSPSURF(plane) : DSPADDR(plane);
+
+	I915_WRITE(reg, I915_READ(reg));
+	POSTING_READ(reg);
 }
 
 /**
- * intel_enable_plane - enable a display plane on a given pipe
+ * intel_enable_primary_plane - enable the primary plane on a given pipe
  * @dev_priv: i915 private structure
  * @plane: plane to enable
  * @pipe: pipe being fed
  *
  * Enable @plane on @pipe, making sure that @pipe is running first.
  */
-static void intel_enable_plane(struct drm_i915_private *dev_priv,
-			       enum plane plane, enum pipe pipe)
+static void intel_enable_primary_plane(struct drm_i915_private *dev_priv,
+				       enum plane plane, enum pipe pipe)
 {
+	struct intel_crtc *intel_crtc =
+		to_intel_crtc(dev_priv->pipe_to_crtc_mapping[pipe]);
 	int reg;
 	u32 val;
 
 	/* If the pipe isn't enabled, we can't pump pixels and may hang */
 	assert_pipe_enabled(dev_priv, pipe);
 
+	WARN(intel_crtc->primary_enabled, "Primary plane already enabled\n");
+
+	intel_crtc->primary_enabled = true;
+
 	reg = DSPCNTR(plane);
 	val = I915_READ(reg);
 	if (val & DISPLAY_PLANE_ENABLE)
 		return;
 
 	I915_WRITE(reg, val | DISPLAY_PLANE_ENABLE);
-	intel_flush_display_plane(dev_priv, plane);
+	intel_flush_primary_plane(dev_priv, plane);
 	intel_wait_for_vblank(dev_priv->dev, pipe);
 }
 
 /**
- * intel_disable_plane - disable a display plane
+ * intel_disable_primary_plane - disable the primary plane
  * @dev_priv: i915 private structure
  * @plane: plane to disable
  * @pipe: pipe consuming the data
  *
  * Disable @plane; should be an independent operation.
  */
-static void intel_disable_plane(struct drm_i915_private *dev_priv,
-				enum plane plane, enum pipe pipe)
+static void intel_disable_primary_plane(struct drm_i915_private *dev_priv,
+					enum plane plane, enum pipe pipe)
 {
+	struct intel_crtc *intel_crtc =
+		to_intel_crtc(dev_priv->pipe_to_crtc_mapping[pipe]);
 	int reg;
 	u32 val;
 
+	WARN(!intel_crtc->primary_enabled, "Primary plane already disabled\n");
+
+	intel_crtc->primary_enabled = false;
+
 	reg = DSPCNTR(plane);
 	val = I915_READ(reg);
 	if ((val & DISPLAY_PLANE_ENABLE) == 0)
 		return;
 
 	I915_WRITE(reg, val & ~DISPLAY_PLANE_ENABLE);
-	intel_flush_display_plane(dev_priv, plane);
+	intel_flush_primary_plane(dev_priv, plane);
 	intel_wait_for_vblank(dev_priv->dev, pipe);
 }
 
@@ -1839,10 +1921,7 @@ intel_pin_and_fence_fb_obj(struct drm_device *dev,
 		alignment = 0;
 		break;
 	case I915_TILING_Y:
-		/* Despite that we check this in framebuffer_init userspace can
-		 * screw us over and change the tiling after the fact. Only
-		 * pinned buffers can't change their tiling. */
-		DRM_DEBUG_DRIVER("Y tiled not allowed for scan out buffers\n");
+		WARN(1, "Y tiled bo slipped through, driver bug!\n");
 		return -EINVAL;
 	default:
 		BUG();
@@ -2077,7 +2156,7 @@ static int ironlake_update_plane(struct drm_crtc *crtc,
 	else
 		dspcntr &= ~DISPPLANE_TILED;
 
-	if (IS_HASWELL(dev))
+	if (IS_HASWELL(dev) || IS_BROADWELL(dev))
 		dspcntr &= ~DISPPLANE_TRICKLE_FEED_DISABLE;
 	else
 		dspcntr |= DISPPLANE_TRICKLE_FEED_DISABLE;
@@ -2097,7 +2176,7 @@ static int ironlake_update_plane(struct drm_crtc *crtc,
 	I915_WRITE(DSPSTRIDE(plane), fb->pitches[0]);
 	I915_MODIFY_DISPBASE(DSPSURF(plane),
 			     i915_gem_obj_ggtt_offset(obj) + intel_crtc->dspaddr_offset);
-	if (IS_HASWELL(dev)) {
+	if (IS_HASWELL(dev) || IS_BROADWELL(dev)) {
 		I915_WRITE(DSPOFFSET(plane), (y << 16) | x);
 	} else {
 		I915_WRITE(DSPTILEOFF(plane), (y << 16) | x);
@@ -2244,11 +2323,26 @@ intel_pipe_set_base(struct drm_crtc *crtc, int x, int y,
 		return ret;
 	}
 
-	/* Update pipe size and adjust fitter if needed */
+	/*
+	 * Update pipe size and adjust fitter if needed: the reason for this is
+	 * that in compute_mode_changes we check the native mode (not the pfit
+	 * mode) to see if we can flip rather than do a full mode set. In the
+	 * fastboot case, we'll flip, but if we don't update the pipesrc and
+	 * pfit state, we'll end up with a big fb scanned out into the wrong
+	 * sized surface.
+	 *
+	 * To fix this properly, we need to hoist the checks up into
+	 * compute_mode_changes (or above), check the actual pfit state and
+	 * whether the platform allows pfit disable with pipe active, and only
+	 * then update the pipesrc and pfit state, even on the flip path.
+	 */
 	if (i915_fastboot) {
+		const struct drm_display_mode *adjusted_mode =
+			&intel_crtc->config.adjusted_mode;
+
 		I915_WRITE(PIPESRC(intel_crtc->pipe),
-			   ((crtc->mode.hdisplay - 1) << 16) |
-			   (crtc->mode.vdisplay - 1));
+			   ((adjusted_mode->crtc_hdisplay - 1) << 16) |
+			   (adjusted_mode->crtc_vdisplay - 1));
 		if (!intel_crtc->config.pch_pfit.enabled &&
 		    (intel_pipe_has_type(crtc, INTEL_OUTPUT_LVDS) ||
 		     intel_pipe_has_type(crtc, INTEL_OUTPUT_EDP))) {
@@ -2873,6 +2967,7 @@ static void lpt_program_iclkip(struct drm_crtc *crtc)
 {
 	struct drm_device *dev = crtc->dev;
 	struct drm_i915_private *dev_priv = dev->dev_private;
+	int clock = to_intel_crtc(crtc)->config.adjusted_mode.crtc_clock;
 	u32 divsel, phaseinc, auxdiv, phasedir = 0;
 	u32 temp;
 
@@ -2890,14 +2985,14 @@ static void lpt_program_iclkip(struct drm_crtc *crtc)
 			SBI_ICLK);
 
 	/* 20MHz is a corner case which is out of range for the 7-bit divisor */
-	if (crtc->mode.clock == 20000) {
+	if (clock == 20000) {
 		auxdiv = 1;
 		divsel = 0x41;
 		phaseinc = 0x20;
 	} else {
 		/* The iCLK virtual clock root frequency is in MHz,
-		 * but the crtc->mode.clock in in KHz. To get the divisors,
-		 * it is necessary to divide one by another, so we
+		 * but the adjusted_mode->crtc_clock in in KHz. To get the
+		 * divisors, it is necessary to divide one by another, so we
 		 * convert the virtual clock precision to KHz here for higher
 		 * precision.
 		 */
@@ -2905,7 +3000,7 @@ static void lpt_program_iclkip(struct drm_crtc *crtc)
 		u32 iclk_pi_range = 64;
 		u32 desired_divisor, msb_divisor_value, pi_value;
 
-		desired_divisor = (iclk_virtual_root_freq / crtc->mode.clock);
+		desired_divisor = (iclk_virtual_root_freq / clock);
 		msb_divisor_value = desired_divisor / iclk_pi_range;
 		pi_value = desired_divisor % iclk_pi_range;
 
@@ -2921,7 +3016,7 @@ static void lpt_program_iclkip(struct drm_crtc *crtc)
 		~SBI_SSCDIVINTPHASE_INCVAL_MASK);
 
 	DRM_DEBUG_KMS("iCLKIP clock: found settings for %dKHz refresh rate: auxdiv=%x, divsel=%x, phasedir=%x, phaseinc=%x\n",
-			crtc->mode.clock,
+			clock,
 			auxdiv,
 			divsel,
 			phasedir,
@@ -3286,6 +3381,108 @@ static void intel_disable_planes(struct drm_crtc *crtc)
 			intel_plane_disable(&intel_plane->base);
 }
 
+void hsw_enable_ips(struct intel_crtc *crtc)
+{
+	struct drm_i915_private *dev_priv = crtc->base.dev->dev_private;
+
+	if (!crtc->config.ips_enabled)
+		return;
+
+	/* We can only enable IPS after we enable a plane and wait for a vblank.
+	 * We guarantee that the plane is enabled by calling intel_enable_ips
+	 * only after intel_enable_plane. And intel_enable_plane already waits
+	 * for a vblank, so all we need to do here is to enable the IPS bit. */
+	assert_plane_enabled(dev_priv, crtc->plane);
+	if (IS_BROADWELL(crtc->base.dev)) {
+		mutex_lock(&dev_priv->rps.hw_lock);
+		WARN_ON(sandybridge_pcode_write(dev_priv, DISPLAY_IPS_CONTROL, 0xc0000000));
+		mutex_unlock(&dev_priv->rps.hw_lock);
+		/* Quoting Art Runyan: "its not safe to expect any particular
+		 * value in IPS_CTL bit 31 after enabling IPS through the
+		 * mailbox." Therefore we need to defer waiting on the state
+		 * change.
+		 * TODO: need to fix this for state checker
+		 */
+	} else {
+		I915_WRITE(IPS_CTL, IPS_ENABLE);
+		/* The bit only becomes 1 in the next vblank, so this wait here
+		 * is essentially intel_wait_for_vblank. If we don't have this
+		 * and don't wait for vblanks until the end of crtc_enable, then
+		 * the HW state readout code will complain that the expected
+		 * IPS_CTL value is not the one we read. */
+		if (wait_for(I915_READ_NOTRACE(IPS_CTL) & IPS_ENABLE, 50))
+			DRM_ERROR("Timed out waiting for IPS enable\n");
+	}
+}
+
+void hsw_disable_ips(struct intel_crtc *crtc)
+{
+	struct drm_device *dev = crtc->base.dev;
+	struct drm_i915_private *dev_priv = dev->dev_private;
+
+	if (!crtc->config.ips_enabled)
+		return;
+
+	assert_plane_enabled(dev_priv, crtc->plane);
+	if (IS_BROADWELL(crtc->base.dev)) {
+		mutex_lock(&dev_priv->rps.hw_lock);
+		WARN_ON(sandybridge_pcode_write(dev_priv, DISPLAY_IPS_CONTROL, 0));
+		mutex_unlock(&dev_priv->rps.hw_lock);
+	} else
+		I915_WRITE(IPS_CTL, 0);
+	POSTING_READ(IPS_CTL);
+
+	/* We need to wait for a vblank before we can disable the plane. */
+	intel_wait_for_vblank(dev, crtc->pipe);
+}
+
+/** Loads the palette/gamma unit for the CRTC with the prepared values */
+static void intel_crtc_load_lut(struct drm_crtc *crtc)
+{
+	struct drm_device *dev = crtc->dev;
+	struct drm_i915_private *dev_priv = dev->dev_private;
+	struct intel_crtc *intel_crtc = to_intel_crtc(crtc);
+	enum pipe pipe = intel_crtc->pipe;
+	int palreg = PALETTE(pipe);
+	int i;
+	bool reenable_ips = false;
+
+	/* The clocks have to be on to load the palette. */
+	if (!crtc->enabled || !intel_crtc->active)
+		return;
+
+	if (!HAS_PCH_SPLIT(dev_priv->dev)) {
+		if (intel_pipe_has_type(crtc, INTEL_OUTPUT_DSI))
+			assert_dsi_pll_enabled(dev_priv);
+		else
+			assert_pll_enabled(dev_priv, pipe);
+	}
+
+	/* use legacy palette for Ironlake */
+	if (HAS_PCH_SPLIT(dev))
+		palreg = LGC_PALETTE(pipe);
+
+	/* Workaround : Do not read or write the pipe palette/gamma data while
+	 * GAMMA_MODE is configured for split gamma and IPS_CTL has IPS enabled.
+	 */
+	if (intel_crtc->config.ips_enabled &&
+	    ((I915_READ(GAMMA_MODE(pipe)) & GAMMA_MODE_MODE_MASK) ==
+	     GAMMA_MODE_MODE_SPLIT)) {
+		hsw_disable_ips(intel_crtc);
+		reenable_ips = true;
+	}
+
+	for (i = 0; i < 256; i++) {
+		I915_WRITE(palreg + 4 * i,
+			   (intel_crtc->lut_r[i] << 16) |
+			   (intel_crtc->lut_g[i] << 8) |
+			   intel_crtc->lut_b[i]);
+	}
+
+	if (reenable_ips)
+		hsw_enable_ips(intel_crtc);
+}
+
 static void ironlake_crtc_enable(struct drm_crtc *crtc)
 {
 	struct drm_device *dev = crtc->dev;
@@ -3305,8 +3502,6 @@ static void ironlake_crtc_enable(struct drm_crtc *crtc)
 	intel_set_cpu_fifo_underrun_reporting(dev, pipe, true);
 	intel_set_pch_fifo_underrun_reporting(dev, pipe, true);
 
-	intel_update_watermarks(dev);
-
 	for_each_encoder_on_crtc(dev, crtc, encoder)
 		if (encoder->pre_enable)
 			encoder->pre_enable(encoder);
@@ -3329,9 +3524,10 @@ static void ironlake_crtc_enable(struct drm_crtc *crtc)
 	 */
 	intel_crtc_load_lut(crtc);
 
+	intel_update_watermarks(crtc);
 	intel_enable_pipe(dev_priv, pipe,
-			  intel_crtc->config.has_pch_encoder);
-	intel_enable_plane(dev_priv, plane, pipe);
+			  intel_crtc->config.has_pch_encoder, false);
+	intel_enable_primary_plane(dev_priv, plane, pipe);
 	intel_enable_planes(crtc);
 	intel_crtc_update_cursor(crtc, true);
 
@@ -3365,34 +3561,74 @@ static bool hsw_crtc_supports_ips(struct intel_crtc *crtc)
 	return HAS_IPS(crtc->base.dev) && crtc->pipe == PIPE_A;
 }
 
-static void hsw_enable_ips(struct intel_crtc *crtc)
+static void haswell_crtc_enable_planes(struct drm_crtc *crtc)
 {
-	struct drm_i915_private *dev_priv = crtc->base.dev->dev_private;
+	struct drm_device *dev = crtc->dev;
+	struct drm_i915_private *dev_priv = dev->dev_private;
+	struct intel_crtc *intel_crtc = to_intel_crtc(crtc);
+	int pipe = intel_crtc->pipe;
+	int plane = intel_crtc->plane;
 
-	if (!crtc->config.ips_enabled)
-		return;
+	intel_enable_primary_plane(dev_priv, plane, pipe);
+	intel_enable_planes(crtc);
+	intel_crtc_update_cursor(crtc, true);
 
-	/* We can only enable IPS after we enable a plane and wait for a vblank.
-	 * We guarantee that the plane is enabled by calling intel_enable_ips
-	 * only after intel_enable_plane. And intel_enable_plane already waits
-	 * for a vblank, so all we need to do here is to enable the IPS bit. */
-	assert_plane_enabled(dev_priv, crtc->plane);
-	I915_WRITE(IPS_CTL, IPS_ENABLE);
+	hsw_enable_ips(intel_crtc);
+
+	mutex_lock(&dev->struct_mutex);
+	intel_update_fbc(dev);
+	mutex_unlock(&dev->struct_mutex);
 }
 
-static void hsw_disable_ips(struct intel_crtc *crtc)
+static void haswell_crtc_disable_planes(struct drm_crtc *crtc)
 {
-	struct drm_device *dev = crtc->base.dev;
+	struct drm_device *dev = crtc->dev;
 	struct drm_i915_private *dev_priv = dev->dev_private;
+	struct intel_crtc *intel_crtc = to_intel_crtc(crtc);
+	int pipe = intel_crtc->pipe;
+	int plane = intel_crtc->plane;
 
-	if (!crtc->config.ips_enabled)
-		return;
+	intel_crtc_wait_for_pending_flips(crtc);
+	drm_vblank_off(dev, pipe);
 
-	assert_plane_enabled(dev_priv, crtc->plane);
-	I915_WRITE(IPS_CTL, 0);
+	/* FBC must be disabled before disabling the plane on HSW. */
+	if (dev_priv->fbc.plane == plane)
+		intel_disable_fbc(dev);
 
-	/* We need to wait for a vblank before we can disable the plane. */
-	intel_wait_for_vblank(dev, crtc->pipe);
+	hsw_disable_ips(intel_crtc);
+
+	intel_crtc_update_cursor(crtc, false);
+	intel_disable_planes(crtc);
+	intel_disable_primary_plane(dev_priv, plane, pipe);
+}
+
+/*
+ * This implements the workaround described in the "notes" section of the mode
+ * set sequence documentation. When going from no pipes or single pipe to
+ * multiple pipes, and planes are enabled after the pipe, we need to wait at
+ * least 2 vblanks on the first pipe before enabling planes on the second pipe.
+ */
+static void haswell_mode_set_planes_workaround(struct intel_crtc *crtc)
+{
+	struct drm_device *dev = crtc->base.dev;
+	struct intel_crtc *crtc_it, *other_active_crtc = NULL;
+
+	/* We want to get the other_active_crtc only if there's only 1 other
+	 * active crtc. */
+	list_for_each_entry(crtc_it, &dev->mode_config.crtc_list, base.head) {
+		if (!crtc_it->active || crtc_it == crtc)
+			continue;
+
+		if (other_active_crtc)
+			return;
+
+		other_active_crtc = crtc_it;
+	}
+	if (!other_active_crtc)
+		return;
+
+	intel_wait_for_vblank(dev, other_active_crtc->pipe);
+	intel_wait_for_vblank(dev, other_active_crtc->pipe);
 }
 
 static void haswell_crtc_enable(struct drm_crtc *crtc)
@@ -3402,7 +3638,6 @@ static void haswell_crtc_enable(struct drm_crtc *crtc)
 	struct intel_crtc *intel_crtc = to_intel_crtc(crtc);
 	struct intel_encoder *encoder;
 	int pipe = intel_crtc->pipe;
-	int plane = intel_crtc->plane;
 
 	WARN_ON(!crtc->enabled);
 
@@ -3415,8 +3650,6 @@ static void haswell_crtc_enable(struct drm_crtc *crtc)
 	if (intel_crtc->config.has_pch_encoder)
 		intel_set_pch_fifo_underrun_reporting(dev, TRANSCODER_A, true);
 
-	intel_update_watermarks(dev);
-
 	if (intel_crtc->config.has_pch_encoder)
 		dev_priv->display.fdi_link_train(crtc);
 
@@ -3437,23 +3670,22 @@ static void haswell_crtc_enable(struct drm_crtc *crtc)
 	intel_ddi_set_pipe_settings(crtc);
 	intel_ddi_enable_transcoder_func(crtc);
 
+	intel_update_watermarks(crtc);
 	intel_enable_pipe(dev_priv, pipe,
-			  intel_crtc->config.has_pch_encoder);
-	intel_enable_plane(dev_priv, plane, pipe);
-	intel_enable_planes(crtc);
-	intel_crtc_update_cursor(crtc, true);
-
-	hsw_enable_ips(intel_crtc);
+			  intel_crtc->config.has_pch_encoder, false);
 
 	if (intel_crtc->config.has_pch_encoder)
 		lpt_pch_enable(crtc);
 
-	mutex_lock(&dev->struct_mutex);
-	intel_update_fbc(dev);
-	mutex_unlock(&dev->struct_mutex);
-
-	for_each_encoder_on_crtc(dev, crtc, encoder)
+	for_each_encoder_on_crtc(dev, crtc, encoder) {
 		encoder->enable(encoder);
+		intel_opregion_notify_encoder(encoder, true);
+	}
+
+	/* If we change the relative order between pipe/planes enabling, we need
+	 * to change the workaround. */
+	haswell_mode_set_planes_workaround(intel_crtc);
+	haswell_crtc_enable_planes(crtc);
 
 	/*
 	 * There seems to be a race in PCH platform hw (at least on some
@@ -3506,7 +3738,7 @@ static void ironlake_crtc_disable(struct drm_crtc *crtc)
 
 	intel_crtc_update_cursor(crtc, false);
 	intel_disable_planes(crtc);
-	intel_disable_plane(dev_priv, plane, pipe);
+	intel_disable_primary_plane(dev_priv, plane, pipe);
 
 	if (intel_crtc->config.has_pch_encoder)
 		intel_set_pch_fifo_underrun_reporting(dev, pipe, false);
@@ -3547,7 +3779,7 @@ static void ironlake_crtc_disable(struct drm_crtc *crtc)
 	}
 
 	intel_crtc->active = false;
-	intel_update_watermarks(dev);
+	intel_update_watermarks(crtc);
 
 	mutex_lock(&dev->struct_mutex);
 	intel_update_fbc(dev);
@@ -3561,27 +3793,17 @@ static void haswell_crtc_disable(struct drm_crtc *crtc)
 	struct intel_crtc *intel_crtc = to_intel_crtc(crtc);
 	struct intel_encoder *encoder;
 	int pipe = intel_crtc->pipe;
-	int plane = intel_crtc->plane;
 	enum transcoder cpu_transcoder = intel_crtc->config.cpu_transcoder;
 
 	if (!intel_crtc->active)
 		return;
 
-	for_each_encoder_on_crtc(dev, crtc, encoder)
-		encoder->disable(encoder);
-
-	intel_crtc_wait_for_pending_flips(crtc);
-	drm_vblank_off(dev, pipe);
+	haswell_crtc_disable_planes(crtc);
 
-	/* FBC must be disabled before disabling the plane on HSW. */
-	if (dev_priv->fbc.plane == plane)
-		intel_disable_fbc(dev);
-
-	hsw_disable_ips(intel_crtc);
-
-	intel_crtc_update_cursor(crtc, false);
-	intel_disable_planes(crtc);
-	intel_disable_plane(dev_priv, plane, pipe);
+	for_each_encoder_on_crtc(dev, crtc, encoder) {
+		intel_opregion_notify_encoder(encoder, false);
+		encoder->disable(encoder);
+	}
 
 	if (intel_crtc->config.has_pch_encoder)
 		intel_set_pch_fifo_underrun_reporting(dev, TRANSCODER_A, false);
@@ -3604,7 +3826,7 @@ static void haswell_crtc_disable(struct drm_crtc *crtc)
 	}
 
 	intel_crtc->active = false;
-	intel_update_watermarks(dev);
+	intel_update_watermarks(crtc);
 
 	mutex_lock(&dev->struct_mutex);
 	intel_update_fbc(dev);
@@ -3696,6 +3918,7 @@ static void valleyview_crtc_enable(struct drm_crtc *crtc)
 	struct intel_encoder *encoder;
 	int pipe = intel_crtc->pipe;
 	int plane = intel_crtc->plane;
+	bool is_dsi;
 
 	WARN_ON(!crtc->enabled);
 
@@ -3703,13 +3926,15 @@ static void valleyview_crtc_enable(struct drm_crtc *crtc)
 		return;
 
 	intel_crtc->active = true;
-	intel_update_watermarks(dev);
 
 	for_each_encoder_on_crtc(dev, crtc, encoder)
 		if (encoder->pre_pll_enable)
 			encoder->pre_pll_enable(encoder);
 
-	vlv_enable_pll(intel_crtc);
+	is_dsi = intel_pipe_has_type(crtc, INTEL_OUTPUT_DSI);
+
+	if (!is_dsi)
+		vlv_enable_pll(intel_crtc);
 
 	for_each_encoder_on_crtc(dev, crtc, encoder)
 		if (encoder->pre_enable)
@@ -3719,8 +3944,9 @@ static void valleyview_crtc_enable(struct drm_crtc *crtc)
 
 	intel_crtc_load_lut(crtc);
 
-	intel_enable_pipe(dev_priv, pipe, false);
-	intel_enable_plane(dev_priv, plane, pipe);
+	intel_update_watermarks(crtc);
+	intel_enable_pipe(dev_priv, pipe, false, is_dsi);
+	intel_enable_primary_plane(dev_priv, plane, pipe);
 	intel_enable_planes(crtc);
 	intel_crtc_update_cursor(crtc, true);
 
@@ -3745,7 +3971,6 @@ static void i9xx_crtc_enable(struct drm_crtc *crtc)
 		return;
 
 	intel_crtc->active = true;
-	intel_update_watermarks(dev);
 
 	for_each_encoder_on_crtc(dev, crtc, encoder)
 		if (encoder->pre_enable)
@@ -3757,8 +3982,9 @@ static void i9xx_crtc_enable(struct drm_crtc *crtc)
 
 	intel_crtc_load_lut(crtc);
 
-	intel_enable_pipe(dev_priv, pipe, false);
-	intel_enable_plane(dev_priv, plane, pipe);
+	intel_update_watermarks(crtc);
+	intel_enable_pipe(dev_priv, pipe, false, false);
+	intel_enable_primary_plane(dev_priv, plane, pipe);
 	intel_enable_planes(crtc);
 	/* The fixup needs to happen before cursor is enabled */
 	if (IS_G4X(dev))
@@ -3814,7 +4040,7 @@ static void i9xx_crtc_disable(struct drm_crtc *crtc)
 	intel_crtc_dpms_overlay(intel_crtc, false);
 	intel_crtc_update_cursor(crtc, false);
 	intel_disable_planes(crtc);
-	intel_disable_plane(dev_priv, plane, pipe);
+	intel_disable_primary_plane(dev_priv, plane, pipe);
 
 	intel_disable_pipe(dev_priv, pipe);
 
@@ -3824,11 +4050,15 @@ static void i9xx_crtc_disable(struct drm_crtc *crtc)
 		if (encoder->post_disable)
 			encoder->post_disable(encoder);
 
-	i9xx_disable_pll(dev_priv, pipe);
+	if (IS_VALLEYVIEW(dev) && !intel_pipe_has_type(crtc, INTEL_OUTPUT_DSI))
+		vlv_disable_pll(dev_priv, pipe);
+	else if (!IS_VALLEYVIEW(dev))
+		i9xx_disable_pll(dev_priv, pipe);
 
 	intel_crtc->active = false;
+	intel_update_watermarks(crtc);
+
 	intel_update_fbc(dev);
-	intel_update_watermarks(dev);
 }
 
 static void i9xx_crtc_off(struct drm_crtc *crtc)
@@ -3902,6 +4132,7 @@ static void intel_crtc_disable(struct drm_crtc *crtc)
 	dev_priv->display.off(crtc);
 
 	assert_plane_disabled(dev->dev_private, to_intel_crtc(crtc)->plane);
+	assert_cursor_disabled(dev_priv, to_intel_crtc(crtc)->pipe);
 	assert_pipe_disabled(dev->dev_private, to_intel_crtc(crtc)->pipe);
 
 	if (crtc->fb) {
@@ -4029,7 +4260,7 @@ static bool ironlake_check_fdi_lanes(struct drm_device *dev, enum pipe pipe,
 		return false;
 	}
 
-	if (IS_HASWELL(dev)) {
+	if (IS_HASWELL(dev) || IS_BROADWELL(dev)) {
 		if (pipe_config->fdi_lanes > 2) {
 			DRM_DEBUG_KMS("only 2 lanes on haswell, required: %i lanes\n",
 				      pipe_config->fdi_lanes);
@@ -4091,8 +4322,7 @@ retry:
 	 */
 	link_bw = intel_fdi_link_freq(dev) * MHz(100)/KHz(1)/10;
 
-	fdi_dotclock = adjusted_mode->clock;
-	fdi_dotclock /= pipe_config->pixel_multiplier;
+	fdi_dotclock = adjusted_mode->crtc_clock;
 
 	lane = ironlake_get_lanes_required(fdi_dotclock, link_bw,
 					   pipe_config->pipe_bpp);
@@ -4134,13 +4364,39 @@ static int intel_crtc_compute_config(struct intel_crtc *crtc,
 	struct drm_device *dev = crtc->base.dev;
 	struct drm_display_mode *adjusted_mode = &pipe_config->adjusted_mode;
 
-	if (HAS_PCH_SPLIT(dev)) {
-		/* FDI link clock is fixed at 2.7G */
-		if (pipe_config->requested_mode.clock * 3
-		    > IRONLAKE_FDI_FREQ * 4)
+	/* FIXME should check pixel clock limits on all platforms */
+	if (INTEL_INFO(dev)->gen < 4) {
+		struct drm_i915_private *dev_priv = dev->dev_private;
+		int clock_limit =
+			dev_priv->display.get_display_clock_speed(dev);
+
+		/*
+		 * Enable pixel doubling when the dot clock
+		 * is > 90% of the (display) core speed.
+		 *
+		 * GDG double wide on either pipe,
+		 * otherwise pipe A only.
+		 */
+		if ((crtc->pipe == PIPE_A || IS_I915G(dev)) &&
+		    adjusted_mode->crtc_clock > clock_limit * 9 / 10) {
+			clock_limit *= 2;
+			pipe_config->double_wide = true;
+		}
+
+		if (adjusted_mode->crtc_clock > clock_limit * 9 / 10)
 			return -EINVAL;
 	}
 
+	/*
+	 * Pipe horizontal size must be even in:
+	 * - DVO ganged mode
+	 * - LVDS dual channel mode
+	 * - Double wide pipe
+	 */
+	if ((intel_pipe_has_type(&crtc->base, INTEL_OUTPUT_LVDS) &&
+	     intel_is_dual_link_lvds(dev)) || pipe_config->double_wide)
+		pipe_config->pipe_src_w &= ~1;
+
 	/* Cantiga+ cannot handle modes with a hsync front porch of 0.
 	 * WaPruneModeWithIncorrectHsyncOffset:ctg,elk,ilk,snb,ivb,vlv,hsw.
 	 */
@@ -4304,28 +4560,6 @@ static inline bool intel_panel_use_ssc(struct drm_i915_private *dev_priv)
 		&& !(dev_priv->quirks & QUIRK_LVDS_SSC_DISABLE);
 }
 
-static int vlv_get_refclk(struct drm_crtc *crtc)
-{
-	struct drm_device *dev = crtc->dev;
-	struct drm_i915_private *dev_priv = dev->dev_private;
-	int refclk = 27000; /* for DP & HDMI */
-
-	return 100000; /* only one validated so far */
-
-	if (intel_pipe_has_type(crtc, INTEL_OUTPUT_ANALOG)) {
-		refclk = 96000;
-	} else if (intel_pipe_has_type(crtc, INTEL_OUTPUT_LVDS)) {
-		if (intel_panel_use_ssc(dev_priv))
-			refclk = 100000;
-		else
-			refclk = 96000;
-	} else if (intel_pipe_has_type(crtc, INTEL_OUTPUT_EDP)) {
-		refclk = 100000;
-	}
-
-	return refclk;
-}
-
 static int i9xx_get_refclk(struct drm_crtc *crtc, int num_connectors)
 {
 	struct drm_device *dev = crtc->dev;
@@ -4333,7 +4567,7 @@ static int i9xx_get_refclk(struct drm_crtc *crtc, int num_connectors)
 	int refclk;
 
 	if (IS_VALLEYVIEW(dev)) {
-		refclk = vlv_get_refclk(crtc);
+		refclk = 100000;
 	} else if (intel_pipe_has_type(crtc, INTEL_OUTPUT_LVDS) &&
 	    intel_panel_use_ssc(dev_priv) && num_connectors < 2) {
 		refclk = dev_priv->vbt.lvds_ssc_freq * 1000;
@@ -4391,7 +4625,8 @@ static void i9xx_update_pll_dividers(struct intel_crtc *crtc,
 	}
 }
 
-static void vlv_pllb_recal_opamp(struct drm_i915_private *dev_priv)
+static void vlv_pllb_recal_opamp(struct drm_i915_private *dev_priv, enum pipe
+		pipe)
 {
 	u32 reg_val;
 
@@ -4399,24 +4634,24 @@ static void vlv_pllb_recal_opamp(struct drm_i915_private *dev_priv)
 	 * PLLB opamp always calibrates to max value of 0x3f, force enable it
 	 * and set it to a reasonable value instead.
 	 */
-	reg_val = vlv_dpio_read(dev_priv, DPIO_IREF(1));
+	reg_val = vlv_dpio_read(dev_priv, pipe, DPIO_IREF(1));
 	reg_val &= 0xffffff00;
 	reg_val |= 0x00000030;
-	vlv_dpio_write(dev_priv, DPIO_IREF(1), reg_val);
+	vlv_dpio_write(dev_priv, pipe, DPIO_IREF(1), reg_val);
 
-	reg_val = vlv_dpio_read(dev_priv, DPIO_CALIBRATION);
+	reg_val = vlv_dpio_read(dev_priv, pipe, DPIO_CALIBRATION);
 	reg_val &= 0x8cffffff;
 	reg_val = 0x8c000000;
-	vlv_dpio_write(dev_priv, DPIO_CALIBRATION, reg_val);
+	vlv_dpio_write(dev_priv, pipe, DPIO_CALIBRATION, reg_val);
 
-	reg_val = vlv_dpio_read(dev_priv, DPIO_IREF(1));
+	reg_val = vlv_dpio_read(dev_priv, pipe, DPIO_IREF(1));
 	reg_val &= 0xffffff00;
-	vlv_dpio_write(dev_priv, DPIO_IREF(1), reg_val);
+	vlv_dpio_write(dev_priv, pipe, DPIO_IREF(1), reg_val);
 
-	reg_val = vlv_dpio_read(dev_priv, DPIO_CALIBRATION);
+	reg_val = vlv_dpio_read(dev_priv, pipe, DPIO_CALIBRATION);
 	reg_val &= 0x00ffffff;
 	reg_val |= 0xb0000000;
-	vlv_dpio_write(dev_priv, DPIO_CALIBRATION, reg_val);
+	vlv_dpio_write(dev_priv, pipe, DPIO_CALIBRATION, reg_val);
 }
 
 static void intel_pch_transcoder_set_m_n(struct intel_crtc *crtc,
@@ -4482,18 +4717,18 @@ static void vlv_update_pll(struct intel_crtc *crtc)
 
 	/* PLL B needs special handling */
 	if (pipe)
-		vlv_pllb_recal_opamp(dev_priv);
+		vlv_pllb_recal_opamp(dev_priv, pipe);
 
 	/* Set up Tx target for periodic Rcomp update */
-	vlv_dpio_write(dev_priv, DPIO_IREF_BCAST, 0x0100000f);
+	vlv_dpio_write(dev_priv, pipe, DPIO_IREF_BCAST, 0x0100000f);
 
 	/* Disable target IRef on PLL */
-	reg_val = vlv_dpio_read(dev_priv, DPIO_IREF_CTL(pipe));
+	reg_val = vlv_dpio_read(dev_priv, pipe, DPIO_IREF_CTL(pipe));
 	reg_val &= 0x00ffffff;
-	vlv_dpio_write(dev_priv, DPIO_IREF_CTL(pipe), reg_val);
+	vlv_dpio_write(dev_priv, pipe, DPIO_IREF_CTL(pipe), reg_val);
 
 	/* Disable fast lock */
-	vlv_dpio_write(dev_priv, DPIO_FASTCLK_DISABLE, 0x610);
+	vlv_dpio_write(dev_priv, pipe, DPIO_FASTCLK_DISABLE, 0x610);
 
 	/* Set idtafcrecal before PLL is enabled */
 	mdiv = ((bestm1 << DPIO_M1DIV_SHIFT) | (bestm2 & DPIO_M2DIV_MASK));
@@ -4507,55 +4742,55 @@ static void vlv_update_pll(struct intel_crtc *crtc)
 	 * Note: don't use the DAC post divider as it seems unstable.
 	 */
 	mdiv |= (DPIO_POST_DIV_HDMIDP << DPIO_POST_DIV_SHIFT);
-	vlv_dpio_write(dev_priv, DPIO_DIV(pipe), mdiv);
+	vlv_dpio_write(dev_priv, pipe, DPIO_DIV(pipe), mdiv);
 
 	mdiv |= DPIO_ENABLE_CALIBRATION;
-	vlv_dpio_write(dev_priv, DPIO_DIV(pipe), mdiv);
+	vlv_dpio_write(dev_priv, pipe, DPIO_DIV(pipe), mdiv);
 
 	/* Set HBR and RBR LPF coefficients */
 	if (crtc->config.port_clock == 162000 ||
 	    intel_pipe_has_type(&crtc->base, INTEL_OUTPUT_ANALOG) ||
 	    intel_pipe_has_type(&crtc->base, INTEL_OUTPUT_HDMI))
-		vlv_dpio_write(dev_priv, DPIO_LPF_COEFF(pipe),
+		vlv_dpio_write(dev_priv, pipe, DPIO_LPF_COEFF(pipe),
 				 0x009f0003);
 	else
-		vlv_dpio_write(dev_priv, DPIO_LPF_COEFF(pipe),
+		vlv_dpio_write(dev_priv, pipe, DPIO_LPF_COEFF(pipe),
 				 0x00d0000f);
 
 	if (intel_pipe_has_type(&crtc->base, INTEL_OUTPUT_EDP) ||
 	    intel_pipe_has_type(&crtc->base, INTEL_OUTPUT_DISPLAYPORT)) {
 		/* Use SSC source */
 		if (!pipe)
-			vlv_dpio_write(dev_priv, DPIO_REFSFR(pipe),
+			vlv_dpio_write(dev_priv, pipe, DPIO_REFSFR(pipe),
 					 0x0df40000);
 		else
-			vlv_dpio_write(dev_priv, DPIO_REFSFR(pipe),
+			vlv_dpio_write(dev_priv, pipe, DPIO_REFSFR(pipe),
 					 0x0df70000);
 	} else { /* HDMI or VGA */
 		/* Use bend source */
 		if (!pipe)
-			vlv_dpio_write(dev_priv, DPIO_REFSFR(pipe),
+			vlv_dpio_write(dev_priv, pipe, DPIO_REFSFR(pipe),
 					 0x0df70000);
 		else
-			vlv_dpio_write(dev_priv, DPIO_REFSFR(pipe),
+			vlv_dpio_write(dev_priv, pipe, DPIO_REFSFR(pipe),
 					 0x0df40000);
 	}
 
-	coreclk = vlv_dpio_read(dev_priv, DPIO_CORE_CLK(pipe));
+	coreclk = vlv_dpio_read(dev_priv, pipe, DPIO_CORE_CLK(pipe));
 	coreclk = (coreclk & 0x0000ff00) | 0x01c00000;
 	if (intel_pipe_has_type(&crtc->base, INTEL_OUTPUT_DISPLAYPORT) ||
 	    intel_pipe_has_type(&crtc->base, INTEL_OUTPUT_EDP))
 		coreclk |= 0x01000000;
-	vlv_dpio_write(dev_priv, DPIO_CORE_CLK(pipe), coreclk);
+	vlv_dpio_write(dev_priv, pipe, DPIO_CORE_CLK(pipe), coreclk);
 
-	vlv_dpio_write(dev_priv, DPIO_PLL_CML(pipe), 0x87871000);
+	vlv_dpio_write(dev_priv, pipe, DPIO_PLL_CML(pipe), 0x87871000);
 
 	/* Enable DPIO clock input */
 	dpll = DPLL_EXT_BUFFER_ENABLE_VLV | DPLL_REFA_CLK_ENABLE_VLV |
 		DPLL_VGA_MODE_DIS | DPLL_INTEGRATED_CLOCK_VLV;
-	if (pipe)
+	/* We should never disable this, set it here for state tracking */
+	if (pipe == PIPE_B)
 		dpll |= DPLL_INTEGRATED_CRI_CLK_VLV;
-
 	dpll |= DPLL_VCO_ENABLE;
 	crtc->config.dpll_hw_state.dpll = dpll;
 
@@ -4693,7 +4928,6 @@ static void intel_set_pipe_timings(struct intel_crtc *intel_crtc)
 	enum transcoder cpu_transcoder = intel_crtc->config.cpu_transcoder;
 	struct drm_display_mode *adjusted_mode =
 		&intel_crtc->config.adjusted_mode;
-	struct drm_display_mode *mode = &intel_crtc->config.requested_mode;
 	uint32_t vsyncshift, crtc_vtotal, crtc_vblank_end;
 
 	/* We need to be careful not to changed the adjusted mode, for otherwise
@@ -4746,7 +4980,8 @@ static void intel_set_pipe_timings(struct intel_crtc *intel_crtc)
 	 * always be the user's requested size.
 	 */
 	I915_WRITE(PIPESRC(pipe),
-		   ((mode->hdisplay - 1) << 16) | (mode->vdisplay - 1));
+		   ((intel_crtc->config.pipe_src_w - 1) << 16) |
+		   (intel_crtc->config.pipe_src_h - 1));
 }
 
 static void intel_get_pipe_timings(struct intel_crtc *crtc,
@@ -4784,8 +5019,11 @@ static void intel_get_pipe_timings(struct intel_crtc *crtc,
 	}
 
 	tmp = I915_READ(PIPESRC(crtc->pipe));
-	pipe_config->requested_mode.vdisplay = (tmp & 0xffff) + 1;
-	pipe_config->requested_mode.hdisplay = ((tmp >> 16) & 0xffff) + 1;
+	pipe_config->pipe_src_h = (tmp & 0xffff) + 1;
+	pipe_config->pipe_src_w = ((tmp >> 16) & 0xffff) + 1;
+
+	pipe_config->requested_mode.vdisplay = pipe_config->pipe_src_h;
+	pipe_config->requested_mode.hdisplay = pipe_config->pipe_src_w;
 }
 
 static void intel_crtc_mode_from_pipe_config(struct intel_crtc *intel_crtc,
@@ -4805,7 +5043,7 @@ static void intel_crtc_mode_from_pipe_config(struct intel_crtc *intel_crtc,
 
 	crtc->mode.flags = pipe_config->adjusted_mode.flags;
 
-	crtc->mode.clock = pipe_config->adjusted_mode.clock;
+	crtc->mode.clock = pipe_config->adjusted_mode.crtc_clock;
 	crtc->mode.flags |= pipe_config->adjusted_mode.flags;
 }
 
@@ -4821,17 +5059,8 @@ static void i9xx_set_pipeconf(struct intel_crtc *intel_crtc)
 	    I915_READ(PIPECONF(intel_crtc->pipe)) & PIPECONF_ENABLE)
 		pipeconf |= PIPECONF_ENABLE;
 
-	if (intel_crtc->pipe == 0 && INTEL_INFO(dev)->gen < 4) {
-		/* Enable pixel doubling when the dot clock is > 90% of the (display)
-		 * core speed.
-		 *
-		 * XXX: No double-wide on 915GM pipe B. Is that the only reason for the
-		 * pipe == 0 check?
-		 */
-		if (intel_crtc->config.requested_mode.clock >
-		    dev_priv->display.get_display_clock_speed(dev) * 9 / 10)
-			pipeconf |= PIPECONF_DOUBLE_WIDE;
-	}
+	if (intel_crtc->config.double_wide)
+		pipeconf |= PIPECONF_DOUBLE_WIDE;
 
 	/* only g4x and later have fancy bpc/dither controls */
 	if (IS_G4X(dev) || IS_VALLEYVIEW(dev)) {
@@ -4885,14 +5114,13 @@ static int i9xx_crtc_mode_set(struct drm_crtc *crtc,
 	struct drm_device *dev = crtc->dev;
 	struct drm_i915_private *dev_priv = dev->dev_private;
 	struct intel_crtc *intel_crtc = to_intel_crtc(crtc);
-	struct drm_display_mode *mode = &intel_crtc->config.requested_mode;
 	int pipe = intel_crtc->pipe;
 	int plane = intel_crtc->plane;
 	int refclk, num_connectors = 0;
 	intel_clock_t clock, reduced_clock;
 	u32 dspcntr;
 	bool ok, has_reduced_clock = false;
-	bool is_lvds = false;
+	bool is_lvds = false, is_dsi = false;
 	struct intel_encoder *encoder;
 	const intel_limit_t *limit;
 	int ret;
@@ -4902,42 +5130,49 @@ static int i9xx_crtc_mode_set(struct drm_crtc *crtc,
 		case INTEL_OUTPUT_LVDS:
 			is_lvds = true;
 			break;
+		case INTEL_OUTPUT_DSI:
+			is_dsi = true;
+			break;
 		}
 
 		num_connectors++;
 	}
 
-	refclk = i9xx_get_refclk(crtc, num_connectors);
+	if (is_dsi)
+		goto skip_dpll;
 
-	/*
-	 * Returns a set of divisors for the desired target clock with the given
-	 * refclk, or FALSE.  The returned values represent the clock equation:
-	 * reflck * (5 * (m1 + 2) + (m2 + 2)) / (n + 2) / p1 / p2.
-	 */
-	limit = intel_limit(crtc, refclk);
-	ok = dev_priv->display.find_dpll(limit, crtc,
-					 intel_crtc->config.port_clock,
-					 refclk, NULL, &clock);
-	if (!ok && !intel_crtc->config.clock_set) {
-		DRM_ERROR("Couldn't find PLL settings for mode!\n");
-		return -EINVAL;
-	}
+	if (!intel_crtc->config.clock_set) {
+		refclk = i9xx_get_refclk(crtc, num_connectors);
 
-	if (is_lvds && dev_priv->lvds_downclock_avail) {
 		/*
-		 * Ensure we match the reduced clock's P to the target clock.
-		 * If the clocks don't match, we can't switch the display clock
-		 * by using the FP0/FP1. In such case we will disable the LVDS
-		 * downclock feature.
-		*/
-		has_reduced_clock =
-			dev_priv->display.find_dpll(limit, crtc,
-						    dev_priv->lvds_downclock,
-						    refclk, &clock,
-						    &reduced_clock);
-	}
-	/* Compat-code for transition, will disappear. */
-	if (!intel_crtc->config.clock_set) {
+		 * Returns a set of divisors for the desired target clock with
+		 * the given refclk, or FALSE.  The returned values represent
+		 * the clock equation: reflck * (5 * (m1 + 2) + (m2 + 2)) / (n +
+		 * 2) / p1 / p2.
+		 */
+		limit = intel_limit(crtc, refclk);
+		ok = dev_priv->display.find_dpll(limit, crtc,
+						 intel_crtc->config.port_clock,
+						 refclk, NULL, &clock);
+		if (!ok) {
+			DRM_ERROR("Couldn't find PLL settings for mode!\n");
+			return -EINVAL;
+		}
+
+		if (is_lvds && dev_priv->lvds_downclock_avail) {
+			/*
+			 * Ensure we match the reduced clock's P to the target
+			 * clock.  If the clocks don't match, we can't switch
+			 * the display clock by using the FP0/FP1. In such case
+			 * we will disable the LVDS downclock feature.
+			 */
+			has_reduced_clock =
+				dev_priv->display.find_dpll(limit, crtc,
+							    dev_priv->lvds_downclock,
+							    refclk, &clock,
+							    &reduced_clock);
+		}
+		/* Compat-code for transition, will disappear. */
 		intel_crtc->config.dpll.n = clock.n;
 		intel_crtc->config.dpll.m1 = clock.m1;
 		intel_crtc->config.dpll.m2 = clock.m2;
@@ -4945,17 +5180,19 @@ static int i9xx_crtc_mode_set(struct drm_crtc *crtc,
 		intel_crtc->config.dpll.p2 = clock.p2;
 	}
 
-	if (IS_GEN2(dev))
+	if (IS_GEN2(dev)) {
 		i8xx_update_pll(intel_crtc,
 				has_reduced_clock ? &reduced_clock : NULL,
 				num_connectors);
-	else if (IS_VALLEYVIEW(dev))
+	} else if (IS_VALLEYVIEW(dev)) {
 		vlv_update_pll(intel_crtc);
-	else
+	} else {
 		i9xx_update_pll(intel_crtc,
 				has_reduced_clock ? &reduced_clock : NULL,
                                 num_connectors);
+	}
 
+skip_dpll:
 	/* Set up the display plane register */
 	dspcntr = DISPPLANE_GAMMA_ENABLE;
 
@@ -4972,8 +5209,8 @@ static int i9xx_crtc_mode_set(struct drm_crtc *crtc,
 	 * which should always be the user's requested size.
 	 */
 	I915_WRITE(DSPSIZE(plane),
-		   ((mode->vdisplay - 1) << 16) |
-		   (mode->hdisplay - 1));
+		   ((intel_crtc->config.pipe_src_h - 1) << 16) |
+		   (intel_crtc->config.pipe_src_w - 1));
 	I915_WRITE(DSPPOS(plane), 0);
 
 	i9xx_set_pipeconf(intel_crtc);
@@ -4983,8 +5220,6 @@ static int i9xx_crtc_mode_set(struct drm_crtc *crtc,
 
 	ret = intel_pipe_set_base(crtc, x, y, fb);
 
-	intel_update_watermarks(dev);
-
 	return ret;
 }
 
@@ -5015,6 +5250,32 @@ static void i9xx_get_pfit_config(struct intel_crtc *crtc,
 			I915_READ(LVDS) & LVDS_BORDER_ENABLE;
 }
 
+static void vlv_crtc_clock_get(struct intel_crtc *crtc,
+			       struct intel_crtc_config *pipe_config)
+{
+	struct drm_device *dev = crtc->base.dev;
+	struct drm_i915_private *dev_priv = dev->dev_private;
+	int pipe = pipe_config->cpu_transcoder;
+	intel_clock_t clock;
+	u32 mdiv;
+	int refclk = 100000;
+
+	mutex_lock(&dev_priv->dpio_lock);
+	mdiv = vlv_dpio_read(dev_priv, pipe, DPIO_DIV(pipe));
+	mutex_unlock(&dev_priv->dpio_lock);
+
+	clock.m1 = (mdiv >> DPIO_M1DIV_SHIFT) & 7;
+	clock.m2 = mdiv & DPIO_M2DIV_MASK;
+	clock.n = (mdiv >> DPIO_N_SHIFT) & 0xf;
+	clock.p1 = (mdiv >> DPIO_P1_SHIFT) & 7;
+	clock.p2 = (mdiv >> DPIO_P2_SHIFT) & 0x1f;
+
+	vlv_clock(refclk, &clock);
+
+	/* clock.dot is the fast clock */
+	pipe_config->port_clock = clock.dot / 5;
+}
+
 static bool i9xx_get_pipe_config(struct intel_crtc *crtc,
 				 struct intel_crtc_config *pipe_config)
 {
@@ -5045,6 +5306,9 @@ static bool i9xx_get_pipe_config(struct intel_crtc *crtc,
 		}
 	}
 
+	if (INTEL_INFO(dev)->gen < 4)
+		pipe_config->double_wide = tmp & PIPECONF_DOUBLE_WIDE;
+
 	intel_get_pipe_timings(crtc, pipe_config);
 
 	i9xx_get_pfit_config(crtc, pipe_config);
@@ -5077,6 +5341,11 @@ static bool i9xx_get_pipe_config(struct intel_crtc *crtc,
 						     DPLL_PORTB_READY_MASK);
 	}
 
+	if (IS_VALLEYVIEW(dev))
+		vlv_crtc_clock_get(crtc, pipe_config);
+	else
+		i9xx_crtc_clock_get(crtc, pipe_config);
+
 	return true;
 }
 
@@ -5565,14 +5834,16 @@ static void intel_set_pipe_csc(struct drm_crtc *crtc)
 
 static void haswell_set_pipeconf(struct drm_crtc *crtc)
 {
-	struct drm_i915_private *dev_priv = crtc->dev->dev_private;
+	struct drm_device *dev = crtc->dev;
+	struct drm_i915_private *dev_priv = dev->dev_private;
 	struct intel_crtc *intel_crtc = to_intel_crtc(crtc);
+	enum pipe pipe = intel_crtc->pipe;
 	enum transcoder cpu_transcoder = intel_crtc->config.cpu_transcoder;
 	uint32_t val;
 
 	val = 0;
 
-	if (intel_crtc->config.dither)
+	if (IS_HASWELL(dev) && intel_crtc->config.dither)
 		val |= (PIPECONF_DITHER_EN | PIPECONF_DITHER_TYPE_SP);
 
 	if (intel_crtc->config.adjusted_mode.flags & DRM_MODE_FLAG_INTERLACE)
@@ -5585,6 +5856,33 @@ static void haswell_set_pipeconf(struct drm_crtc *crtc)
 
 	I915_WRITE(GAMMA_MODE(intel_crtc->pipe), GAMMA_MODE_MODE_8BIT);
 	POSTING_READ(GAMMA_MODE(intel_crtc->pipe));
+
+	if (IS_BROADWELL(dev)) {
+		val = 0;
+
+		switch (intel_crtc->config.pipe_bpp) {
+		case 18:
+			val |= PIPEMISC_DITHER_6_BPC;
+			break;
+		case 24:
+			val |= PIPEMISC_DITHER_8_BPC;
+			break;
+		case 30:
+			val |= PIPEMISC_DITHER_10_BPC;
+			break;
+		case 36:
+			val |= PIPEMISC_DITHER_12_BPC;
+			break;
+		default:
+			/* Case prevented by pipe_config_set_bpp. */
+			BUG();
+		}
+
+		if (intel_crtc->config.dither)
+			val |= PIPEMISC_DITHER_ENABLE | PIPEMISC_DITHER_TYPE_SP;
+
+		I915_WRITE(PIPEMISC(pipe), val);
+	}
 }
 
 static bool ironlake_compute_clocks(struct drm_crtc *crtc,
@@ -5819,11 +6117,6 @@ static int ironlake_crtc_mode_set(struct drm_crtc *crtc,
 	else
 		intel_crtc->lowfreq_avail = false;
 
-	if (intel_crtc->config.has_pch_encoder) {
-		pll = intel_crtc_to_shared_dpll(intel_crtc);
-
-	}
-
 	intel_set_pipe_timings(intel_crtc);
 
 	if (intel_crtc->config.has_pch_encoder) {
@@ -5839,25 +6132,67 @@ static int ironlake_crtc_mode_set(struct drm_crtc *crtc,
 
 	ret = intel_pipe_set_base(crtc, x, y, fb);
 
-	intel_update_watermarks(dev);
-
 	return ret;
 }
 
-static void ironlake_get_fdi_m_n_config(struct intel_crtc *crtc,
-					struct intel_crtc_config *pipe_config)
+static void intel_pch_transcoder_get_m_n(struct intel_crtc *crtc,
+					 struct intel_link_m_n *m_n)
 {
 	struct drm_device *dev = crtc->base.dev;
 	struct drm_i915_private *dev_priv = dev->dev_private;
-	enum transcoder transcoder = pipe_config->cpu_transcoder;
+	enum pipe pipe = crtc->pipe;
 
-	pipe_config->fdi_m_n.link_m = I915_READ(PIPE_LINK_M1(transcoder));
-	pipe_config->fdi_m_n.link_n = I915_READ(PIPE_LINK_N1(transcoder));
-	pipe_config->fdi_m_n.gmch_m = I915_READ(PIPE_DATA_M1(transcoder))
-					& ~TU_SIZE_MASK;
-	pipe_config->fdi_m_n.gmch_n = I915_READ(PIPE_DATA_N1(transcoder));
-	pipe_config->fdi_m_n.tu = ((I915_READ(PIPE_DATA_M1(transcoder))
-				   & TU_SIZE_MASK) >> TU_SIZE_SHIFT) + 1;
+	m_n->link_m = I915_READ(PCH_TRANS_LINK_M1(pipe));
+	m_n->link_n = I915_READ(PCH_TRANS_LINK_N1(pipe));
+	m_n->gmch_m = I915_READ(PCH_TRANS_DATA_M1(pipe))
+		& ~TU_SIZE_MASK;
+	m_n->gmch_n = I915_READ(PCH_TRANS_DATA_N1(pipe));
+	m_n->tu = ((I915_READ(PCH_TRANS_DATA_M1(pipe))
+		    & TU_SIZE_MASK) >> TU_SIZE_SHIFT) + 1;
+}
+
+static void intel_cpu_transcoder_get_m_n(struct intel_crtc *crtc,
+					 enum transcoder transcoder,
+					 struct intel_link_m_n *m_n)
+{
+	struct drm_device *dev = crtc->base.dev;
+	struct drm_i915_private *dev_priv = dev->dev_private;
+	enum pipe pipe = crtc->pipe;
+
+	if (INTEL_INFO(dev)->gen >= 5) {
+		m_n->link_m = I915_READ(PIPE_LINK_M1(transcoder));
+		m_n->link_n = I915_READ(PIPE_LINK_N1(transcoder));
+		m_n->gmch_m = I915_READ(PIPE_DATA_M1(transcoder))
+			& ~TU_SIZE_MASK;
+		m_n->gmch_n = I915_READ(PIPE_DATA_N1(transcoder));
+		m_n->tu = ((I915_READ(PIPE_DATA_M1(transcoder))
+			    & TU_SIZE_MASK) >> TU_SIZE_SHIFT) + 1;
+	} else {
+		m_n->link_m = I915_READ(PIPE_LINK_M_G4X(pipe));
+		m_n->link_n = I915_READ(PIPE_LINK_N_G4X(pipe));
+		m_n->gmch_m = I915_READ(PIPE_DATA_M_G4X(pipe))
+			& ~TU_SIZE_MASK;
+		m_n->gmch_n = I915_READ(PIPE_DATA_N_G4X(pipe));
+		m_n->tu = ((I915_READ(PIPE_DATA_M_G4X(pipe))
+			    & TU_SIZE_MASK) >> TU_SIZE_SHIFT) + 1;
+	}
+}
+
+void intel_dp_get_m_n(struct intel_crtc *crtc,
+		      struct intel_crtc_config *pipe_config)
+{
+	if (crtc->config.has_pch_encoder)
+		intel_pch_transcoder_get_m_n(crtc, &pipe_config->dp_m_n);
+	else
+		intel_cpu_transcoder_get_m_n(crtc, pipe_config->cpu_transcoder,
+					     &pipe_config->dp_m_n);
+}
+
+static void ironlake_get_fdi_m_n_config(struct intel_crtc *crtc,
+					struct intel_crtc_config *pipe_config)
+{
+	intel_cpu_transcoder_get_m_n(crtc, pipe_config->cpu_transcoder,
+				     &pipe_config->fdi_m_n);
 }
 
 static void ironlake_get_pfit_config(struct intel_crtc *crtc,
@@ -5946,6 +6281,8 @@ static bool ironlake_get_pipe_config(struct intel_crtc *crtc,
 		pipe_config->pixel_multiplier =
 			((tmp & PLL_REF_SDVO_HDMI_MULTIPLIER_MASK)
 			 >> PLL_REF_SDVO_HDMI_MULTIPLIER_SHIFT) + 1;
+
+		ironlake_pch_clock_get(crtc, pipe_config);
 	} else {
 		pipe_config->pixel_multiplier = 1;
 	}
@@ -6002,8 +6339,8 @@ static void assert_can_disable_lcpll(struct drm_i915_private *dev_priv)
  * register. Callers should take care of disabling all the display engine
  * functions, doing the mode unset, fixing interrupts, etc.
  */
-void hsw_disable_lcpll(struct drm_i915_private *dev_priv,
-		       bool switch_to_fclk, bool allow_power_down)
+static void hsw_disable_lcpll(struct drm_i915_private *dev_priv,
+			      bool switch_to_fclk, bool allow_power_down)
 {
 	uint32_t val;
 
@@ -6031,7 +6368,10 @@ void hsw_disable_lcpll(struct drm_i915_private *dev_priv,
 
 	val = I915_READ(D_COMP);
 	val |= D_COMP_COMP_DISABLE;
-	I915_WRITE(D_COMP, val);
+	mutex_lock(&dev_priv->rps.hw_lock);
+	if (sandybridge_pcode_write(dev_priv, GEN6_PCODE_WRITE_D_COMP, val))
+		DRM_ERROR("Failed to disable D_COMP\n");
+	mutex_unlock(&dev_priv->rps.hw_lock);
 	POSTING_READ(D_COMP);
 	ndelay(100);
 
@@ -6050,7 +6390,7 @@ void hsw_disable_lcpll(struct drm_i915_private *dev_priv,
  * Fully restores LCPLL, disallowing power down and switching back to LCPLL
  * source.
  */
-void hsw_restore_lcpll(struct drm_i915_private *dev_priv)
+static void hsw_restore_lcpll(struct drm_i915_private *dev_priv)
 {
 	uint32_t val;
 
@@ -6073,7 +6413,10 @@ void hsw_restore_lcpll(struct drm_i915_private *dev_priv)
 	val = I915_READ(D_COMP);
 	val |= D_COMP_COMP_FORCE;
 	val &= ~D_COMP_COMP_DISABLE;
-	I915_WRITE(D_COMP, val);
+	mutex_lock(&dev_priv->rps.hw_lock);
+	if (sandybridge_pcode_write(dev_priv, GEN6_PCODE_WRITE_D_COMP, val))
+		DRM_ERROR("Failed to enable D_COMP\n");
+	mutex_unlock(&dev_priv->rps.hw_lock);
 	POSTING_READ(D_COMP);
 
 	val = I915_READ(LCPLL_CTL);
@@ -6256,22 +6599,79 @@ static void hsw_package_c8_gpu_busy(struct drm_i915_private *dev_priv)
 	}
 }
 
-static void haswell_modeset_global_resources(struct drm_device *dev)
+#define for_each_power_domain(domain, mask)				\
+	for ((domain) = 0; (domain) < POWER_DOMAIN_NUM; (domain)++)	\
+		if ((1 << (domain)) & (mask))
+
+static unsigned long get_pipe_power_domains(struct drm_device *dev,
+					    enum pipe pipe, bool pfit_enabled)
 {
-	bool enable = false;
+	unsigned long mask;
+	enum transcoder transcoder;
+
+	transcoder = intel_pipe_to_cpu_transcoder(dev->dev_private, pipe);
+
+	mask = BIT(POWER_DOMAIN_PIPE(pipe));
+	mask |= BIT(POWER_DOMAIN_TRANSCODER(transcoder));
+	if (pfit_enabled)
+		mask |= BIT(POWER_DOMAIN_PIPE_PANEL_FITTER(pipe));
+
+	return mask;
+}
+
+void intel_display_set_init_power(struct drm_device *dev, bool enable)
+{
+	struct drm_i915_private *dev_priv = dev->dev_private;
+
+	if (dev_priv->power_domains.init_power_on == enable)
+		return;
+
+	if (enable)
+		intel_display_power_get(dev, POWER_DOMAIN_INIT);
+	else
+		intel_display_power_put(dev, POWER_DOMAIN_INIT);
+
+	dev_priv->power_domains.init_power_on = enable;
+}
+
+static void modeset_update_power_wells(struct drm_device *dev)
+{
+	unsigned long pipe_domains[I915_MAX_PIPES] = { 0, };
 	struct intel_crtc *crtc;
 
+	/*
+	 * First get all needed power domains, then put all unneeded, to avoid
+	 * any unnecessary toggling of the power wells.
+	 */
 	list_for_each_entry(crtc, &dev->mode_config.crtc_list, base.head) {
+		enum intel_display_power_domain domain;
+
 		if (!crtc->base.enabled)
 			continue;
 
-		if (crtc->pipe != PIPE_A || crtc->config.pch_pfit.enabled ||
-		    crtc->config.cpu_transcoder != TRANSCODER_EDP)
-			enable = true;
+		pipe_domains[crtc->pipe] = get_pipe_power_domains(dev,
+						crtc->pipe,
+						crtc->config.pch_pfit.enabled);
+
+		for_each_power_domain(domain, pipe_domains[crtc->pipe])
+			intel_display_power_get(dev, domain);
 	}
 
-	intel_set_power_well(dev, enable);
+	list_for_each_entry(crtc, &dev->mode_config.crtc_list, base.head) {
+		enum intel_display_power_domain domain;
+
+		for_each_power_domain(domain, crtc->enabled_power_domains)
+			intel_display_power_put(dev, domain);
 
+		crtc->enabled_power_domains = pipe_domains[crtc->pipe];
+	}
+
+	intel_display_set_init_power(dev, false);
+}
+
+static void haswell_modeset_global_resources(struct drm_device *dev)
+{
+	modeset_update_power_wells(dev);
 	hsw_update_package_c8(dev);
 }
 
@@ -6310,8 +6710,6 @@ static int haswell_crtc_mode_set(struct drm_crtc *crtc,
 
 	ret = intel_pipe_set_base(crtc, x, y, fb);
 
-	intel_update_watermarks(dev);
-
 	return ret;
 }
 
@@ -6419,6 +6817,44 @@ static int intel_crtc_mode_set(struct drm_crtc *crtc,
 	return 0;
 }
 
+static struct {
+	int clock;
+	u32 config;
+} hdmi_audio_clock[] = {
+	{ DIV_ROUND_UP(25200 * 1000, 1001), AUD_CONFIG_PIXEL_CLOCK_HDMI_25175 },
+	{ 25200, AUD_CONFIG_PIXEL_CLOCK_HDMI_25200 }, /* default per bspec */
+	{ 27000, AUD_CONFIG_PIXEL_CLOCK_HDMI_27000 },
+	{ 27000 * 1001 / 1000, AUD_CONFIG_PIXEL_CLOCK_HDMI_27027 },
+	{ 54000, AUD_CONFIG_PIXEL_CLOCK_HDMI_54000 },
+	{ 54000 * 1001 / 1000, AUD_CONFIG_PIXEL_CLOCK_HDMI_54054 },
+	{ DIV_ROUND_UP(74250 * 1000, 1001), AUD_CONFIG_PIXEL_CLOCK_HDMI_74176 },
+	{ 74250, AUD_CONFIG_PIXEL_CLOCK_HDMI_74250 },
+	{ DIV_ROUND_UP(148500 * 1000, 1001), AUD_CONFIG_PIXEL_CLOCK_HDMI_148352 },
+	{ 148500, AUD_CONFIG_PIXEL_CLOCK_HDMI_148500 },
+};
+
+/* get AUD_CONFIG_PIXEL_CLOCK_HDMI_* value for mode */
+static u32 audio_config_hdmi_pixel_clock(struct drm_display_mode *mode)
+{
+	int i;
+
+	for (i = 0; i < ARRAY_SIZE(hdmi_audio_clock); i++) {
+		if (mode->clock == hdmi_audio_clock[i].clock)
+			break;
+	}
+
+	if (i == ARRAY_SIZE(hdmi_audio_clock)) {
+		DRM_DEBUG_KMS("HDMI audio pixel clock setting for %d not found, falling back to defaults\n", mode->clock);
+		i = 1;
+	}
+
+	DRM_DEBUG_KMS("Configuring HDMI audio for pixel clock %d (0x%08x)\n",
+		      hdmi_audio_clock[i].clock,
+		      hdmi_audio_clock[i].config);
+
+	return hdmi_audio_clock[i].config;
+}
+
 static bool intel_eld_uptodate(struct drm_connector *connector,
 			       int reg_eldv, uint32_t bits_eldv,
 			       int reg_elda, uint32_t bits_elda,
@@ -6449,7 +6885,8 @@ static bool intel_eld_uptodate(struct drm_connector *connector,
 }
 
 static void g4x_write_eld(struct drm_connector *connector,
-			  struct drm_crtc *crtc)
+			  struct drm_crtc *crtc,
+			  struct drm_display_mode *mode)
 {
 	struct drm_i915_private *dev_priv = connector->dev->dev_private;
 	uint8_t *eld = connector->eld;
@@ -6489,7 +6926,8 @@ static void g4x_write_eld(struct drm_connector *connector,
 }
 
 static void haswell_write_eld(struct drm_connector *connector,
-				     struct drm_crtc *crtc)
+			      struct drm_crtc *crtc,
+			      struct drm_display_mode *mode)
 {
 	struct drm_i915_private *dev_priv = connector->dev->dev_private;
 	uint8_t *eld = connector->eld;
@@ -6542,8 +6980,9 @@ static void haswell_write_eld(struct drm_connector *connector,
 		DRM_DEBUG_DRIVER("ELD: DisplayPort detected\n");
 		eld[5] |= (1 << 2);	/* Conn_Type, 0x1 = DisplayPort */
 		I915_WRITE(aud_config, AUD_CONFIG_N_VALUE_INDEX); /* 0x1 = DP */
-	} else
-		I915_WRITE(aud_config, 0);
+	} else {
+		I915_WRITE(aud_config, audio_config_hdmi_pixel_clock(mode));
+	}
 
 	if (intel_eld_uptodate(connector,
 			       aud_cntrl_st2, eldv,
@@ -6576,7 +7015,8 @@ static void haswell_write_eld(struct drm_connector *connector,
 }
 
 static void ironlake_write_eld(struct drm_connector *connector,
-				     struct drm_crtc *crtc)
+			       struct drm_crtc *crtc,
+			       struct drm_display_mode *mode)
 {
 	struct drm_i915_private *dev_priv = connector->dev->dev_private;
 	uint8_t *eld = connector->eld;
@@ -6594,6 +7034,11 @@ static void ironlake_write_eld(struct drm_connector *connector,
 		aud_config = IBX_AUD_CFG(pipe);
 		aud_cntl_st = IBX_AUD_CNTL_ST(pipe);
 		aud_cntrl_st2 = IBX_AUD_CNTL_ST2;
+	} else if (IS_VALLEYVIEW(connector->dev)) {
+		hdmiw_hdmiedid = VLV_HDMIW_HDMIEDID(pipe);
+		aud_config = VLV_AUD_CFG(pipe);
+		aud_cntl_st = VLV_AUD_CNTL_ST(pipe);
+		aud_cntrl_st2 = VLV_AUD_CNTL_ST2;
 	} else {
 		hdmiw_hdmiedid = CPT_HDMIW_HDMIEDID(pipe);
 		aud_config = CPT_AUD_CFG(pipe);
@@ -6603,8 +7048,19 @@ static void ironlake_write_eld(struct drm_connector *connector,
 
 	DRM_DEBUG_DRIVER("ELD on pipe %c\n", pipe_name(pipe));
 
-	i = I915_READ(aud_cntl_st);
-	i = (i >> 29) & DIP_PORT_SEL_MASK;		/* DIP_Port_Select, 0x1 = PortB */
+	if (IS_VALLEYVIEW(connector->dev))  {
+		struct intel_encoder *intel_encoder;
+		struct intel_digital_port *intel_dig_port;
+
+		intel_encoder = intel_attached_encoder(connector);
+		intel_dig_port = enc_to_dig_port(&intel_encoder->base);
+		i = intel_dig_port->port;
+	} else {
+		i = I915_READ(aud_cntl_st);
+		i = (i >> 29) & DIP_PORT_SEL_MASK;
+		/* DIP_Port_Select, 0x1 = PortB */
+	}
+
 	if (!i) {
 		DRM_DEBUG_DRIVER("Audio directed to unknown port\n");
 		/* operate blindly on all ports */
@@ -6620,8 +7076,9 @@ static void ironlake_write_eld(struct drm_connector *connector,
 		DRM_DEBUG_DRIVER("ELD: DisplayPort detected\n");
 		eld[5] |= (1 << 2);	/* Conn_Type, 0x1 = DisplayPort */
 		I915_WRITE(aud_config, AUD_CONFIG_N_VALUE_INDEX); /* 0x1 = DP */
-	} else
-		I915_WRITE(aud_config, 0);
+	} else {
+		I915_WRITE(aud_config, audio_config_hdmi_pixel_clock(mode));
+	}
 
 	if (intel_eld_uptodate(connector,
 			       aud_cntrl_st2, eldv,
@@ -6671,50 +7128,7 @@ void intel_write_eld(struct drm_encoder *encoder,
 	connector->eld[6] = drm_av_sync_delay(connector, mode) / 2;
 
 	if (dev_priv->display.write_eld)
-		dev_priv->display.write_eld(connector, crtc);
-}
-
-/** Loads the palette/gamma unit for the CRTC with the prepared values */
-void intel_crtc_load_lut(struct drm_crtc *crtc)
-{
-	struct drm_device *dev = crtc->dev;
-	struct drm_i915_private *dev_priv = dev->dev_private;
-	struct intel_crtc *intel_crtc = to_intel_crtc(crtc);
-	enum pipe pipe = intel_crtc->pipe;
-	int palreg = PALETTE(pipe);
-	int i;
-	bool reenable_ips = false;
-
-	/* The clocks have to be on to load the palette. */
-	if (!crtc->enabled || !intel_crtc->active)
-		return;
-
-	if (!HAS_PCH_SPLIT(dev_priv->dev))
-		assert_pll_enabled(dev_priv, pipe);
-
-	/* use legacy palette for Ironlake */
-	if (HAS_PCH_SPLIT(dev))
-		palreg = LGC_PALETTE(pipe);
-
-	/* Workaround : Do not read or write the pipe palette/gamma data while
-	 * GAMMA_MODE is configured for split gamma and IPS_CTL has IPS enabled.
-	 */
-	if (intel_crtc->config.ips_enabled &&
-	    ((I915_READ(GAMMA_MODE(pipe)) & GAMMA_MODE_MODE_MASK) ==
-	     GAMMA_MODE_MODE_SPLIT)) {
-		hsw_disable_ips(intel_crtc);
-		reenable_ips = true;
-	}
-
-	for (i = 0; i < 256; i++) {
-		I915_WRITE(palreg + 4 * i,
-			   (intel_crtc->lut_r[i] << 16) |
-			   (intel_crtc->lut_g[i] << 8) |
-			   intel_crtc->lut_b[i]);
-	}
-
-	if (reenable_ips)
-		hsw_enable_ips(intel_crtc);
+		dev_priv->display.write_eld(connector, crtc, mode);
 }
 
 static void i845_update_cursor(struct drm_crtc *crtc, u32 base)
@@ -6790,7 +7204,7 @@ static void ivb_update_cursor(struct drm_crtc *crtc, u32 base)
 			cntl &= ~(CURSOR_MODE | MCURSOR_GAMMA_ENABLE);
 			cntl |= CURSOR_MODE_DISABLE;
 		}
-		if (IS_HASWELL(dev)) {
+		if (IS_HASWELL(dev) || IS_BROADWELL(dev)) {
 			cntl |= CURSOR_PIPE_CSC_ENABLE;
 			cntl &= ~CURSOR_TRICKLE_FEED_DISABLE;
 		}
@@ -6812,23 +7226,20 @@ static void intel_crtc_update_cursor(struct drm_crtc *crtc,
 	int pipe = intel_crtc->pipe;
 	int x = intel_crtc->cursor_x;
 	int y = intel_crtc->cursor_y;
-	u32 base, pos;
+	u32 base = 0, pos = 0;
 	bool visible;
 
-	pos = 0;
-
-	if (on && crtc->enabled && crtc->fb) {
+	if (on)
 		base = intel_crtc->cursor_addr;
-		if (x > (int) crtc->fb->width)
-			base = 0;
 
-		if (y > (int) crtc->fb->height)
-			base = 0;
-	} else
+	if (x >= intel_crtc->config.pipe_src_w)
+		base = 0;
+
+	if (y >= intel_crtc->config.pipe_src_h)
 		base = 0;
 
 	if (x < 0) {
-		if (x + intel_crtc->cursor_width < 0)
+		if (x + intel_crtc->cursor_width <= 0)
 			base = 0;
 
 		pos |= CURSOR_POS_SIGN << CURSOR_X_SHIFT;
@@ -6837,7 +7248,7 @@ static void intel_crtc_update_cursor(struct drm_crtc *crtc,
 	pos |= x << CURSOR_X_SHIFT;
 
 	if (y < 0) {
-		if (y + intel_crtc->cursor_height < 0)
+		if (y + intel_crtc->cursor_height <= 0)
 			base = 0;
 
 		pos |= CURSOR_POS_SIGN << CURSOR_Y_SHIFT;
@@ -6849,7 +7260,7 @@ static void intel_crtc_update_cursor(struct drm_crtc *crtc,
 	if (!visible && !intel_crtc->cursor_visible)
 		return;
 
-	if (IS_IVYBRIDGE(dev) || IS_HASWELL(dev)) {
+	if (IS_IVYBRIDGE(dev) || IS_HASWELL(dev) || IS_BROADWELL(dev)) {
 		I915_WRITE(CURPOS_IVB(pipe), pos);
 		ivb_update_cursor(crtc, base);
 	} else {
@@ -6980,8 +7391,8 @@ static int intel_crtc_cursor_move(struct drm_crtc *crtc, int x, int y)
 {
 	struct intel_crtc *intel_crtc = to_intel_crtc(crtc);
 
-	intel_crtc->cursor_x = x;
-	intel_crtc->cursor_y = y;
+	intel_crtc->cursor_x = clamp_t(int, x, SHRT_MIN, SHRT_MAX);
+	intel_crtc->cursor_y = clamp_t(int, y, SHRT_MIN, SHRT_MAX);
 
 	if (intel_crtc->active)
 		intel_crtc_update_cursor(crtc, intel_crtc->cursor_bo != NULL);
@@ -6989,27 +7400,6 @@ static int intel_crtc_cursor_move(struct drm_crtc *crtc, int x, int y)
 	return 0;
 }
 
-/** Sets the color ramps on behalf of RandR */
-void intel_crtc_fb_gamma_set(struct drm_crtc *crtc, u16 red, u16 green,
-				 u16 blue, int regno)
-{
-	struct intel_crtc *intel_crtc = to_intel_crtc(crtc);
-
-	intel_crtc->lut_r[regno] = red >> 8;
-	intel_crtc->lut_g[regno] = green >> 8;
-	intel_crtc->lut_b[regno] = blue >> 8;
-}
-
-void intel_crtc_fb_gamma_get(struct drm_crtc *crtc, u16 *red, u16 *green,
-			     u16 *blue, int regno)
-{
-	struct intel_crtc *intel_crtc = to_intel_crtc(crtc);
-
-	*red = intel_crtc->lut_r[regno] << 8;
-	*green = intel_crtc->lut_g[regno] << 8;
-	*blue = intel_crtc->lut_b[regno] << 8;
-}
-
 static void intel_crtc_gamma_set(struct drm_crtc *crtc, u16 *red, u16 *green,
 				 u16 *blue, uint32_t start, uint32_t size)
 {
@@ -7045,14 +7435,21 @@ intel_framebuffer_create(struct drm_device *dev,
 		return ERR_PTR(-ENOMEM);
 	}
 
+	ret = i915_mutex_lock_interruptible(dev);
+	if (ret)
+		goto err;
+
 	ret = intel_framebuffer_init(dev, intel_fb, mode_cmd, obj);
-	if (ret) {
-		drm_gem_object_unreference_unlocked(&obj->base);
-		kfree(intel_fb);
-		return ERR_PTR(ret);
-	}
+	mutex_unlock(&dev->struct_mutex);
+	if (ret)
+		goto err;
 
 	return &intel_fb->base;
+err:
+	drm_gem_object_unreference_unlocked(&obj->base);
+	kfree(intel_fb);
+
+	return ERR_PTR(ret);
 }
 
 static u32
@@ -7095,6 +7492,7 @@ static struct drm_framebuffer *
 mode_fits_in_fbdev(struct drm_device *dev,
 		   struct drm_display_mode *mode)
 {
+#ifdef CONFIG_DRM_I915_FBDEV
 	struct drm_i915_private *dev_priv = dev->dev_private;
 	struct drm_i915_gem_object *obj;
 	struct drm_framebuffer *fb;
@@ -7115,6 +7513,9 @@ mode_fits_in_fbdev(struct drm_device *dev,
 		return NULL;
 
 	return fb;
+#else
+	return NULL;
+#endif
 }
 
 bool intel_get_load_detect_pipe(struct drm_connector *connector,
@@ -7258,6 +7659,22 @@ void intel_release_load_detect_pipe(struct drm_connector *connector,
 	mutex_unlock(&crtc->mutex);
 }
 
+static int i9xx_pll_refclk(struct drm_device *dev,
+			   const struct intel_crtc_config *pipe_config)
+{
+	struct drm_i915_private *dev_priv = dev->dev_private;
+	u32 dpll = pipe_config->dpll_hw_state.dpll;
+
+	if ((dpll & PLL_REF_INPUT_MASK) == PLLB_REF_INPUT_SPREADSPECTRUMIN)
+		return dev_priv->vbt.lvds_ssc_freq * 1000;
+	else if (HAS_PCH_SPLIT(dev))
+		return 120000;
+	else if (!IS_GEN2(dev))
+		return 96000;
+	else
+		return 48000;
+}
+
 /* Returns the clock of the currently programmed mode of the given pipe. */
 static void i9xx_crtc_clock_get(struct intel_crtc *crtc,
 				struct intel_crtc_config *pipe_config)
@@ -7265,14 +7682,15 @@ static void i9xx_crtc_clock_get(struct intel_crtc *crtc,
 	struct drm_device *dev = crtc->base.dev;
 	struct drm_i915_private *dev_priv = dev->dev_private;
 	int pipe = pipe_config->cpu_transcoder;
-	u32 dpll = I915_READ(DPLL(pipe));
+	u32 dpll = pipe_config->dpll_hw_state.dpll;
 	u32 fp;
 	intel_clock_t clock;
+	int refclk = i9xx_pll_refclk(dev, pipe_config);
 
 	if ((dpll & DISPLAY_RATE_SELECT_FPA1) == 0)
-		fp = I915_READ(FP0(pipe));
+		fp = pipe_config->dpll_hw_state.fp0;
 	else
-		fp = I915_READ(FP1(pipe));
+		fp = pipe_config->dpll_hw_state.fp1;
 
 	clock.m1 = (fp & FP_M1_DIV_MASK) >> FP_M1_DIV_SHIFT;
 	if (IS_PINEVIEW(dev)) {
@@ -7303,14 +7721,13 @@ static void i9xx_crtc_clock_get(struct intel_crtc *crtc,
 		default:
 			DRM_DEBUG_KMS("Unknown DPLL mode %08x in programmed "
 				  "mode\n", (int)(dpll & DPLL_MODE_MASK));
-			pipe_config->adjusted_mode.clock = 0;
 			return;
 		}
 
 		if (IS_PINEVIEW(dev))
-			pineview_clock(96000, &clock);
+			pineview_clock(refclk, &clock);
 		else
-			i9xx_clock(96000, &clock);
+			i9xx_clock(refclk, &clock);
 	} else {
 		bool is_lvds = (pipe == 1) && (I915_READ(LVDS) & LVDS_PORT_EN);
 
@@ -7318,13 +7735,6 @@ static void i9xx_crtc_clock_get(struct intel_crtc *crtc,
 			clock.p1 = ffs((dpll & DPLL_FPA01_P1_POST_DIV_MASK_I830_LVDS) >>
 				       DPLL_FPA01_P1_POST_DIV_SHIFT);
 			clock.p2 = 14;
-
-			if ((dpll & PLL_REF_INPUT_MASK) ==
-			    PLLB_REF_INPUT_SPREADSPECTRUMIN) {
-				/* XXX: might not be 66MHz */
-				i9xx_clock(66000, &clock);
-			} else
-				i9xx_clock(48000, &clock);
 		} else {
 			if (dpll & PLL_P1_DIVIDE_BY_TWO)
 				clock.p1 = 2;
@@ -7336,59 +7746,55 @@ static void i9xx_crtc_clock_get(struct intel_crtc *crtc,
 				clock.p2 = 4;
 			else
 				clock.p2 = 2;
-
-			i9xx_clock(48000, &clock);
 		}
+
+		i9xx_clock(refclk, &clock);
 	}
 
-	pipe_config->adjusted_mode.clock = clock.dot;
+	/*
+	 * This value includes pixel_multiplier. We will use
+	 * port_clock to compute adjusted_mode.crtc_clock in the
+	 * encoder's get_config() function.
+	 */
+	pipe_config->port_clock = clock.dot;
 }
 
-static void ironlake_crtc_clock_get(struct intel_crtc *crtc,
-				    struct intel_crtc_config *pipe_config)
+int intel_dotclock_calculate(int link_freq,
+			     const struct intel_link_m_n *m_n)
 {
-	struct drm_device *dev = crtc->base.dev;
-	struct drm_i915_private *dev_priv = dev->dev_private;
-	enum transcoder cpu_transcoder = pipe_config->cpu_transcoder;
-	int link_freq, repeat;
-	u64 clock;
-	u32 link_m, link_n;
-
-	repeat = pipe_config->pixel_multiplier;
-
 	/*
 	 * The calculation for the data clock is:
-	 * pixel_clock = ((m/n)*(link_clock * nr_lanes * repeat))/bpp
+	 * pixel_clock = ((m/n)*(link_clock * nr_lanes))/bpp
 	 * But we want to avoid losing precison if possible, so:
-	 * pixel_clock = ((m * link_clock * nr_lanes * repeat)/(n*bpp))
+	 * pixel_clock = ((m * link_clock * nr_lanes)/(n*bpp))
 	 *
 	 * and the link clock is simpler:
-	 * link_clock = (m * link_clock * repeat) / n
+	 * link_clock = (m * link_clock) / n
 	 */
 
-	/*
-	 * We need to get the FDI or DP link clock here to derive
-	 * the M/N dividers.
-	 *
-	 * For FDI, we read it from the BIOS or use a fixed 2.7GHz.
-	 * For DP, it's either 1.62GHz or 2.7GHz.
-	 * We do our calculations in 10*MHz since we don't need much precison.
-	 */
-	if (pipe_config->has_pch_encoder)
-		link_freq = intel_fdi_link_freq(dev) * 10000;
-	else
-		link_freq = pipe_config->port_clock;
+	if (!m_n->link_n)
+		return 0;
 
-	link_m = I915_READ(PIPE_LINK_M1(cpu_transcoder));
-	link_n = I915_READ(PIPE_LINK_N1(cpu_transcoder));
+	return div_u64((u64)m_n->link_m * link_freq, m_n->link_n);
+}
 
-	if (!link_m || !link_n)
-		return;
+static void ironlake_pch_clock_get(struct intel_crtc *crtc,
+				   struct intel_crtc_config *pipe_config)
+{
+	struct drm_device *dev = crtc->base.dev;
 
-	clock = ((u64)link_m * (u64)link_freq * (u64)repeat);
-	do_div(clock, link_n);
+	/* read out port_clock from the DPLL */
+	i9xx_crtc_clock_get(crtc, pipe_config);
 
-	pipe_config->adjusted_mode.clock = clock;
+	/*
+	 * This value does not include pixel_multiplier.
+	 * We will check that port_clock and adjusted_mode.crtc_clock
+	 * agree once we know their relationship in the encoder's
+	 * get_config() function.
+	 */
+	pipe_config->adjusted_mode.crtc_clock =
+		intel_dotclock_calculate(intel_fdi_link_freq(dev) * 10000,
+					 &pipe_config->fdi_m_n);
 }
 
 /** Returns the currently programmed mode of the given pipe. */
@@ -7404,6 +7810,7 @@ struct drm_display_mode *intel_crtc_mode_get(struct drm_device *dev,
 	int hsync = I915_READ(HSYNC(cpu_transcoder));
 	int vtot = I915_READ(VTOTAL(cpu_transcoder));
 	int vsync = I915_READ(VSYNC(cpu_transcoder));
+	enum pipe pipe = intel_crtc->pipe;
 
 	mode = kzalloc(sizeof(*mode), GFP_KERNEL);
 	if (!mode)
@@ -7416,11 +7823,14 @@ struct drm_display_mode *intel_crtc_mode_get(struct drm_device *dev,
 	 * Note, if LVDS ever uses a non-1 pixel multiplier, we'll need
 	 * to use a real value here instead.
 	 */
-	pipe_config.cpu_transcoder = (enum transcoder) intel_crtc->pipe;
+	pipe_config.cpu_transcoder = (enum transcoder) pipe;
 	pipe_config.pixel_multiplier = 1;
+	pipe_config.dpll_hw_state.dpll = I915_READ(DPLL(pipe));
+	pipe_config.dpll_hw_state.fp0 = I915_READ(FP0(pipe));
+	pipe_config.dpll_hw_state.fp1 = I915_READ(FP1(pipe));
 	i9xx_crtc_clock_get(intel_crtc, &pipe_config);
 
-	mode->clock = pipe_config.adjusted_mode.clock;
+	mode->clock = pipe_config.port_clock / pipe_config.pixel_multiplier;
 	mode->hdisplay = (htot & 0xffff) + 1;
 	mode->htotal = ((htot & 0xffff0000) >> 16) + 1;
 	mode->hsync_start = (hsync & 0xffff) + 1;
@@ -7526,6 +7936,9 @@ void intel_mark_idle(struct drm_device *dev)
 
 		intel_decrease_pllclock(crtc);
 	}
+
+	if (dev_priv->info->gen >= 6)
+		gen6_rps_idle(dev->dev_private);
 }
 
 void intel_mark_fb_busy(struct drm_i915_gem_object *obj,
@@ -7714,7 +8127,7 @@ static int intel_gen2_queue_flip(struct drm_device *dev,
 	intel_ring_emit(ring, 0); /* aux display base address, unused */
 
 	intel_mark_page_flip_active(intel_crtc);
-	intel_ring_advance(ring);
+	__intel_ring_advance(ring);
 	return 0;
 
 err_unpin:
@@ -7756,7 +8169,7 @@ static int intel_gen3_queue_flip(struct drm_device *dev,
 	intel_ring_emit(ring, MI_NOOP);
 
 	intel_mark_page_flip_active(intel_crtc);
-	intel_ring_advance(ring);
+	__intel_ring_advance(ring);
 	return 0;
 
 err_unpin:
@@ -7805,7 +8218,7 @@ static int intel_gen4_queue_flip(struct drm_device *dev,
 	intel_ring_emit(ring, pf | pipesrc);
 
 	intel_mark_page_flip_active(intel_crtc);
-	intel_ring_advance(ring);
+	__intel_ring_advance(ring);
 	return 0;
 
 err_unpin:
@@ -7850,7 +8263,7 @@ static int intel_gen6_queue_flip(struct drm_device *dev,
 	intel_ring_emit(ring, pf | pipesrc);
 
 	intel_mark_page_flip_active(intel_crtc);
-	intel_ring_advance(ring);
+	__intel_ring_advance(ring);
 	return 0;
 
 err_unpin:
@@ -7929,7 +8342,7 @@ static int intel_gen7_queue_flip(struct drm_device *dev,
 	intel_ring_emit(ring, (MI_NOOP));
 
 	intel_mark_page_flip_active(intel_crtc);
-	intel_ring_advance(ring);
+	__intel_ring_advance(ring);
 	return 0;
 
 err_unpin:
@@ -7974,7 +8387,7 @@ static int intel_crtc_page_flip(struct drm_crtc *crtc,
 	     fb->pitches[0] != crtc->fb->pitches[0]))
 		return -EINVAL;
 
-	work = kzalloc(sizeof *work, GFP_KERNEL);
+	work = kzalloc(sizeof(*work), GFP_KERNEL);
 	if (work == NULL)
 		return -ENOMEM;
 
@@ -8209,6 +8622,17 @@ compute_baseline_pipe_bpp(struct intel_crtc *crtc,
 	return bpp;
 }
 
+static void intel_dump_crtc_timings(const struct drm_display_mode *mode)
+{
+	DRM_DEBUG_KMS("crtc timings: %d %d %d %d %d %d %d %d %d, "
+			"type: 0x%x flags: 0x%x\n",
+		mode->crtc_clock,
+		mode->crtc_hdisplay, mode->crtc_hsync_start,
+		mode->crtc_hsync_end, mode->crtc_htotal,
+		mode->crtc_vdisplay, mode->crtc_vsync_start,
+		mode->crtc_vsync_end, mode->crtc_vtotal, mode->type, mode->flags);
+}
+
 static void intel_dump_pipe_config(struct intel_crtc *crtc,
 				   struct intel_crtc_config *pipe_config,
 				   const char *context)
@@ -8225,10 +8649,19 @@ static void intel_dump_pipe_config(struct intel_crtc *crtc,
 		      pipe_config->fdi_m_n.gmch_m, pipe_config->fdi_m_n.gmch_n,
 		      pipe_config->fdi_m_n.link_m, pipe_config->fdi_m_n.link_n,
 		      pipe_config->fdi_m_n.tu);
+	DRM_DEBUG_KMS("dp: %i, gmch_m: %u, gmch_n: %u, link_m: %u, link_n: %u, tu: %u\n",
+		      pipe_config->has_dp_encoder,
+		      pipe_config->dp_m_n.gmch_m, pipe_config->dp_m_n.gmch_n,
+		      pipe_config->dp_m_n.link_m, pipe_config->dp_m_n.link_n,
+		      pipe_config->dp_m_n.tu);
 	DRM_DEBUG_KMS("requested mode:\n");
 	drm_mode_debug_printmodeline(&pipe_config->requested_mode);
 	DRM_DEBUG_KMS("adjusted mode:\n");
 	drm_mode_debug_printmodeline(&pipe_config->adjusted_mode);
+	intel_dump_crtc_timings(&pipe_config->adjusted_mode);
+	DRM_DEBUG_KMS("port clock: %d\n", pipe_config->port_clock);
+	DRM_DEBUG_KMS("pipe src size: %dx%d\n",
+		      pipe_config->pipe_src_w, pipe_config->pipe_src_h);
 	DRM_DEBUG_KMS("gmch pfit: control: 0x%08x, ratios: 0x%08x, lvds border: 0x%08x\n",
 		      pipe_config->gmch_pfit.control,
 		      pipe_config->gmch_pfit.pgm_ratios,
@@ -8238,6 +8671,7 @@ static void intel_dump_pipe_config(struct intel_crtc *crtc,
 		      pipe_config->pch_pfit.size,
 		      pipe_config->pch_pfit.enabled ? "enabled" : "disabled");
 	DRM_DEBUG_KMS("ips: %i\n", pipe_config->ips_enabled);
+	DRM_DEBUG_KMS("double wide: %i\n", pipe_config->double_wide);
 }
 
 static bool check_encoder_cloning(struct drm_crtc *crtc)
@@ -8281,6 +8715,7 @@ intel_modeset_pipe_config(struct drm_crtc *crtc,
 
 	drm_mode_copy(&pipe_config->adjusted_mode, mode);
 	drm_mode_copy(&pipe_config->requested_mode, mode);
+
 	pipe_config->cpu_transcoder =
 		(enum transcoder) to_intel_crtc(crtc)->pipe;
 	pipe_config->shared_dpll = DPLL_ID_PRIVATE;
@@ -8307,13 +8742,25 @@ intel_modeset_pipe_config(struct drm_crtc *crtc,
 	if (plane_bpp < 0)
 		goto fail;
 
+	/*
+	 * Determine the real pipe dimensions. Note that stereo modes can
+	 * increase the actual pipe size due to the frame doubling and
+	 * insertion of additional space for blanks between the frame. This
+	 * is stored in the crtc timings. We use the requested mode to do this
+	 * computation to clearly distinguish it from the adjusted mode, which
+	 * can be changed by the connectors in the below retry loop.
+	 */
+	drm_mode_set_crtcinfo(&pipe_config->requested_mode, CRTC_STEREO_DOUBLE);
+	pipe_config->pipe_src_w = pipe_config->requested_mode.crtc_hdisplay;
+	pipe_config->pipe_src_h = pipe_config->requested_mode.crtc_vdisplay;
+
 encoder_retry:
 	/* Ensure the port clock defaults are reset when retrying. */
 	pipe_config->port_clock = 0;
 	pipe_config->pixel_multiplier = 1;
 
 	/* Fill in default crtc timings, allow encoders to overwrite them. */
-	drm_mode_set_crtcinfo(&pipe_config->adjusted_mode, 0);
+	drm_mode_set_crtcinfo(&pipe_config->adjusted_mode, CRTC_STEREO_DOUBLE);
 
 	/* Pass our mode to the connectors and the CRTC to give them a chance to
 	 * adjust it according to limitations or connector properties, and also
@@ -8334,7 +8781,8 @@ encoder_retry:
 	/* Set default port clock if not overwritten by the encoder. Needs to be
 	 * done afterwards in case the encoder adjusts the mode. */
 	if (!pipe_config->port_clock)
-		pipe_config->port_clock = pipe_config->adjusted_mode.clock;
+		pipe_config->port_clock = pipe_config->adjusted_mode.crtc_clock
+			* pipe_config->pixel_multiplier;
 
 	ret = intel_crtc_compute_config(to_intel_crtc(crtc), pipe_config);
 	if (ret < 0) {
@@ -8521,13 +8969,9 @@ intel_modeset_update_state(struct drm_device *dev, unsigned prepare_pipes)
 
 }
 
-static bool intel_fuzzy_clock_check(struct intel_crtc_config *cur,
-				    struct intel_crtc_config *new)
+static bool intel_fuzzy_clock_check(int clock1, int clock2)
 {
-	int clock1, clock2, diff;
-
-	clock1 = cur->adjusted_mode.clock;
-	clock2 = new->adjusted_mode.clock;
+	int diff;
 
 	if (clock1 == clock2)
 		return true;
@@ -8581,6 +9025,15 @@ intel_pipe_config_compare(struct drm_device *dev,
 		return false; \
 	}
 
+#define PIPE_CONF_CHECK_CLOCK_FUZZY(name) \
+	if (!intel_fuzzy_clock_check(current_config->name, pipe_config->name)) { \
+		DRM_ERROR("mismatch in " #name " " \
+			  "(expected %i, found %i)\n", \
+			  current_config->name, \
+			  pipe_config->name); \
+		return false; \
+	}
+
 #define PIPE_CONF_QUIRK(quirk)	\
 	((current_config->quirks | pipe_config->quirks) & (quirk))
 
@@ -8594,6 +9047,13 @@ intel_pipe_config_compare(struct drm_device *dev,
 	PIPE_CONF_CHECK_I(fdi_m_n.link_n);
 	PIPE_CONF_CHECK_I(fdi_m_n.tu);
 
+	PIPE_CONF_CHECK_I(has_dp_encoder);
+	PIPE_CONF_CHECK_I(dp_m_n.gmch_m);
+	PIPE_CONF_CHECK_I(dp_m_n.gmch_n);
+	PIPE_CONF_CHECK_I(dp_m_n.link_m);
+	PIPE_CONF_CHECK_I(dp_m_n.link_n);
+	PIPE_CONF_CHECK_I(dp_m_n.tu);
+
 	PIPE_CONF_CHECK_I(adjusted_mode.crtc_hdisplay);
 	PIPE_CONF_CHECK_I(adjusted_mode.crtc_htotal);
 	PIPE_CONF_CHECK_I(adjusted_mode.crtc_hblank_start);
@@ -8624,8 +9084,8 @@ intel_pipe_config_compare(struct drm_device *dev,
 				      DRM_MODE_FLAG_NVSYNC);
 	}
 
-	PIPE_CONF_CHECK_I(requested_mode.hdisplay);
-	PIPE_CONF_CHECK_I(requested_mode.vdisplay);
+	PIPE_CONF_CHECK_I(pipe_src_w);
+	PIPE_CONF_CHECK_I(pipe_src_h);
 
 	PIPE_CONF_CHECK_I(gmch_pfit.control);
 	/* pfit ratios are autocomputed by the hw on gen4+ */
@@ -8640,6 +9100,8 @@ intel_pipe_config_compare(struct drm_device *dev,
 
 	PIPE_CONF_CHECK_I(ips_enabled);
 
+	PIPE_CONF_CHECK_I(double_wide);
+
 	PIPE_CONF_CHECK_I(shared_dpll);
 	PIPE_CONF_CHECK_X(dpll_hw_state.dpll);
 	PIPE_CONF_CHECK_X(dpll_hw_state.dpll_md);
@@ -8649,20 +9111,17 @@ intel_pipe_config_compare(struct drm_device *dev,
 	if (IS_G4X(dev) || INTEL_INFO(dev)->gen >= 5)
 		PIPE_CONF_CHECK_I(pipe_bpp);
 
+	if (!IS_HASWELL(dev)) {
+		PIPE_CONF_CHECK_CLOCK_FUZZY(adjusted_mode.crtc_clock);
+		PIPE_CONF_CHECK_CLOCK_FUZZY(port_clock);
+	}
+
 #undef PIPE_CONF_CHECK_X
 #undef PIPE_CONF_CHECK_I
 #undef PIPE_CONF_CHECK_FLAGS
+#undef PIPE_CONF_CHECK_CLOCK_FUZZY
 #undef PIPE_CONF_QUIRK
 
-	if (!IS_HASWELL(dev)) {
-		if (!intel_fuzzy_clock_check(current_config, pipe_config)) {
-			DRM_ERROR("mismatch in clock (expected %d, found %d)\n",
-				  current_config->adjusted_mode.clock,
-				  pipe_config->adjusted_mode.clock);
-			return false;
-		}
-	}
-
 	return true;
 }
 
@@ -8794,9 +9253,6 @@ check_crtc_state(struct drm_device *dev)
 				encoder->get_config(encoder, &pipe_config);
 		}
 
-		if (dev_priv->display.get_clock)
-			dev_priv->display.get_clock(crtc, &pipe_config);
-
 		WARN(crtc->active != active,
 		     "crtc active state doesn't match with hw state "
 		     "(expected %i, found %i)\n", crtc->active, active);
@@ -8871,6 +9327,18 @@ intel_modeset_check_state(struct drm_device *dev)
 	check_shared_dpll_state(dev);
 }
 
+void ironlake_check_encoder_dotclock(const struct intel_crtc_config *pipe_config,
+				     int dotclock)
+{
+	/*
+	 * FDI already provided one idea for the dotclock.
+	 * Yell if the encoder disagrees.
+	 */
+	WARN(!intel_fuzzy_clock_check(pipe_config->adjusted_mode.crtc_clock, dotclock),
+	     "FDI dotclock and encoder dotclock mismatch, fdi: %i, encoder: %i\n",
+	     pipe_config->adjusted_mode.crtc_clock, dotclock);
+}
+
 static int __intel_set_mode(struct drm_crtc *crtc,
 			    struct drm_display_mode *mode,
 			    int x, int y, struct drm_framebuffer *fb)
@@ -8883,7 +9351,7 @@ static int __intel_set_mode(struct drm_crtc *crtc,
 	unsigned disable_pipes, prepare_pipes, modeset_pipes;
 	int ret = 0;
 
-	saved_mode = kmalloc(2 * sizeof(*saved_mode), GFP_KERNEL);
+	saved_mode = kcalloc(2, sizeof(*saved_mode), GFP_KERNEL);
 	if (!saved_mode)
 		return -ENOMEM;
 	saved_hwmode = saved_mode + 1;
@@ -9422,7 +9890,7 @@ static void intel_crtc_init(struct drm_device *dev, int pipe)
 	struct intel_crtc *intel_crtc;
 	int i;
 
-	intel_crtc = kzalloc(sizeof(struct intel_crtc) + (INTELFB_CONN_LIMIT * sizeof(struct drm_connector *)), GFP_KERNEL);
+	intel_crtc = kzalloc(sizeof(*intel_crtc), GFP_KERNEL);
 	if (intel_crtc == NULL)
 		return;
 
@@ -9451,6 +9919,18 @@ static void intel_crtc_init(struct drm_device *dev, int pipe)
 	drm_crtc_helper_add(&intel_crtc->base, &intel_helper_funcs);
 }
 
+enum pipe intel_get_pipe_from_connector(struct intel_connector *connector)
+{
+	struct drm_encoder *encoder = connector->base.encoder;
+
+	WARN_ON(!mutex_is_locked(&connector->base.dev->mode_config.mutex));
+
+	if (!encoder)
+		return INVALID_PIPE;
+
+	return to_intel_crtc(encoder->crtc)->pipe;
+}
+
 int intel_get_pipe_from_crtc_id(struct drm_device *dev, void *data,
 				struct drm_file *file)
 {
@@ -9466,7 +9946,7 @@ int intel_get_pipe_from_crtc_id(struct drm_device *dev, void *data,
 
 	if (!drmmode_obj) {
 		DRM_ERROR("no such CRTC id\n");
-		return -EINVAL;
+		return -ENOENT;
 	}
 
 	crtc = to_intel_crtc(obj_to_crtc(drmmode_obj));
@@ -9573,7 +10053,13 @@ static void intel_setup_outputs(struct drm_device *dev)
 		if (I915_READ(PCH_DP_D) & DP_DETECTED)
 			intel_dp_init(dev, PCH_DP_D, PORT_D);
 	} else if (IS_VALLEYVIEW(dev)) {
-		/* Check for built-in panel first. Shares lanes with HDMI on SDVOC */
+		if (I915_READ(VLV_DISPLAY_BASE + GEN4_HDMIB) & SDVO_DETECTED) {
+			intel_hdmi_init(dev, VLV_DISPLAY_BASE + GEN4_HDMIB,
+					PORT_B);
+			if (I915_READ(VLV_DISPLAY_BASE + DP_B) & DP_DETECTED)
+				intel_dp_init(dev, VLV_DISPLAY_BASE + DP_B, PORT_B);
+		}
+
 		if (I915_READ(VLV_DISPLAY_BASE + GEN4_HDMIC) & SDVO_DETECTED) {
 			intel_hdmi_init(dev, VLV_DISPLAY_BASE + GEN4_HDMIC,
 					PORT_C);
@@ -9582,12 +10068,7 @@ static void intel_setup_outputs(struct drm_device *dev)
 					      PORT_C);
 		}
 
-		if (I915_READ(VLV_DISPLAY_BASE + GEN4_HDMIB) & SDVO_DETECTED) {
-			intel_hdmi_init(dev, VLV_DISPLAY_BASE + GEN4_HDMIB,
-					PORT_B);
-			if (I915_READ(VLV_DISPLAY_BASE + DP_B) & DP_DETECTED)
-				intel_dp_init(dev, VLV_DISPLAY_BASE + DP_B, PORT_B);
-		}
+		intel_dsi_init(dev);
 	} else if (SUPPORTS_DIGITAL_OUTPUTS(dev)) {
 		bool found = false;
 
@@ -9643,6 +10124,7 @@ static void intel_setup_outputs(struct drm_device *dev)
 void intel_framebuffer_fini(struct intel_framebuffer *fb)
 {
 	drm_framebuffer_cleanup(&fb->base);
+	WARN_ON(!fb->obj->framebuffer_references--);
 	drm_gem_object_unreference_unlocked(&fb->obj->base);
 }
 
@@ -9674,9 +10156,12 @@ int intel_framebuffer_init(struct drm_device *dev,
 			   struct drm_mode_fb_cmd2 *mode_cmd,
 			   struct drm_i915_gem_object *obj)
 {
+	int aligned_height, tile_height;
 	int pitch_limit;
 	int ret;
 
+	WARN_ON(!mutex_is_locked(&dev->struct_mutex));
+
 	if (obj->tiling_mode == I915_TILING_Y) {
 		DRM_DEBUG("hardware does not support tiling Y\n");
 		return -EINVAL;
@@ -9765,8 +10250,16 @@ int intel_framebuffer_init(struct drm_device *dev,
 	if (mode_cmd->offsets[0] != 0)
 		return -EINVAL;
 
+	tile_height = IS_GEN2(dev) ? 16 : 8;
+	aligned_height = ALIGN(mode_cmd->height,
+			       obj->tiling_mode ? tile_height : 1);
+	/* FIXME drm helper for size checks (especially planar formats)? */
+	if (obj->base.size < aligned_height * mode_cmd->pitches[0])
+		return -EINVAL;
+
 	drm_helper_mode_fill_fb_struct(&intel_fb->base, mode_cmd);
 	intel_fb->obj = obj;
+	intel_fb->obj->framebuffer_references++;
 
 	ret = drm_framebuffer_init(dev, &intel_fb->base, &intel_fb_funcs);
 	if (ret) {
@@ -9792,9 +10285,15 @@ intel_user_framebuffer_create(struct drm_device *dev,
 	return intel_framebuffer_create(dev, mode_cmd, obj);
 }
 
+#ifndef CONFIG_DRM_I915_FBDEV
+static inline void intel_fbdev_output_poll_changed(struct drm_device *dev)
+{
+}
+#endif
+
 static const struct drm_mode_config_funcs intel_mode_funcs = {
 	.fb_create = intel_user_framebuffer_create,
-	.output_poll_changed = intel_fb_output_poll_changed,
+	.output_poll_changed = intel_fbdev_output_poll_changed,
 };
 
 /* Set up chip specific display functions */
@@ -9820,7 +10319,6 @@ static void intel_init_display(struct drm_device *dev)
 		dev_priv->display.update_plane = ironlake_update_plane;
 	} else if (HAS_PCH_SPLIT(dev)) {
 		dev_priv->display.get_pipe_config = ironlake_get_pipe_config;
-		dev_priv->display.get_clock = ironlake_crtc_clock_get;
 		dev_priv->display.crtc_mode_set = ironlake_crtc_mode_set;
 		dev_priv->display.crtc_enable = ironlake_crtc_enable;
 		dev_priv->display.crtc_disable = ironlake_crtc_disable;
@@ -9828,7 +10326,6 @@ static void intel_init_display(struct drm_device *dev)
 		dev_priv->display.update_plane = ironlake_update_plane;
 	} else if (IS_VALLEYVIEW(dev)) {
 		dev_priv->display.get_pipe_config = i9xx_get_pipe_config;
-		dev_priv->display.get_clock = i9xx_crtc_clock_get;
 		dev_priv->display.crtc_mode_set = i9xx_crtc_mode_set;
 		dev_priv->display.crtc_enable = valleyview_crtc_enable;
 		dev_priv->display.crtc_disable = i9xx_crtc_disable;
@@ -9836,7 +10333,6 @@ static void intel_init_display(struct drm_device *dev)
 		dev_priv->display.update_plane = i9xx_update_plane;
 	} else {
 		dev_priv->display.get_pipe_config = i9xx_get_pipe_config;
-		dev_priv->display.get_clock = i9xx_crtc_clock_get;
 		dev_priv->display.crtc_mode_set = i9xx_crtc_mode_set;
 		dev_priv->display.crtc_enable = i9xx_crtc_enable;
 		dev_priv->display.crtc_disable = i9xx_crtc_disable;
@@ -9886,7 +10382,7 @@ static void intel_init_display(struct drm_device *dev)
 			dev_priv->display.write_eld = ironlake_write_eld;
 			dev_priv->display.modeset_global_resources =
 				ivb_modeset_global_resources;
-		} else if (IS_HASWELL(dev)) {
+		} else if (IS_HASWELL(dev) || IS_GEN8(dev)) {
 			dev_priv->display.fdi_link_train = hsw_fdi_link_train;
 			dev_priv->display.write_eld = haswell_write_eld;
 			dev_priv->display.modeset_global_resources =
@@ -9894,7 +10390,8 @@ static void intel_init_display(struct drm_device *dev)
 		}
 	} else if (IS_G4X(dev)) {
 		dev_priv->display.write_eld = g4x_write_eld;
-	}
+	} else if (IS_VALLEYVIEW(dev))
+		dev_priv->display.write_eld = ironlake_write_eld;
 
 	/* Default just returns -ENODEV to indicate unsupported */
 	dev_priv->display.queue_flip = intel_default_queue_flip;
@@ -9917,6 +10414,7 @@ static void intel_init_display(struct drm_device *dev)
 		dev_priv->display.queue_flip = intel_gen6_queue_flip;
 		break;
 	case 7:
+	case 8: /* FIXME(BDW): Check that the gen8 RCS flip works. */
 		dev_priv->display.queue_flip = intel_gen7_queue_flip;
 		break;
 	}
@@ -10012,8 +10510,7 @@ static struct intel_quirk intel_quirks[] = {
 	/* ThinkPad T60 needs pipe A force quirk (bug #16494) */
 	{ 0x2782, 0x17aa, 0x201a, quirk_pipea_force },
 
-	/* 830/845 need to leave pipe A & dpll A up */
-	{ 0x2562, PCI_ANY_ID, PCI_ANY_ID, quirk_pipea_force },
+	/* 830 needs to leave pipe A & dpll A up */
 	{ 0x3577, PCI_ANY_ID, PCI_ANY_ID, quirk_pipea_force },
 
 	/* Lenovo U160 cannot use SSC on LVDS */
@@ -10022,20 +10519,11 @@ static struct intel_quirk intel_quirks[] = {
 	/* Sony Vaio Y cannot use SSC on LVDS */
 	{ 0x0046, 0x104d, 0x9076, quirk_ssc_force_disable },
 
-	/* Acer Aspire 5734Z must invert backlight brightness */
-	{ 0x2a42, 0x1025, 0x0459, quirk_invert_brightness },
-
-	/* Acer/eMachines G725 */
-	{ 0x2a42, 0x1025, 0x0210, quirk_invert_brightness },
-
-	/* Acer/eMachines e725 */
-	{ 0x2a42, 0x1025, 0x0212, quirk_invert_brightness },
-
-	/* Acer/Packard Bell NCL20 */
-	{ 0x2a42, 0x1025, 0x034b, quirk_invert_brightness },
-
-	/* Acer Aspire 4736Z */
-	{ 0x2a42, 0x1025, 0x0260, quirk_invert_brightness },
+	/*
+	 * All GM45 Acer (and its brands eMachines and Packard Bell) laptops
+	 * seem to use inverted backlight PWM.
+	 */
+	{ 0x2a42, 0x1025, PCI_ANY_ID, quirk_invert_brightness },
 
 	/* Dell XPS13 HD Sandy Bridge */
 	{ 0x0116, 0x1028, 0x052e, quirk_no_pcm_pwm_enable },
@@ -10084,12 +10572,19 @@ static void i915_disable_vga(struct drm_device *dev)
 
 void intel_modeset_init_hw(struct drm_device *dev)
 {
-	intel_init_power_well(dev);
+	struct drm_i915_private *dev_priv = dev->dev_private;
 
 	intel_prepare_ddi(dev);
 
 	intel_init_clock_gating(dev);
 
+	/* Enable the CRI clock source so we can get at the display */
+	if (IS_VALLEYVIEW(dev))
+		I915_WRITE(DPLL(PIPE_B), I915_READ(DPLL(PIPE_B)) |
+			   DPLL_INTEGRATED_CRI_CLK_VLV);
+
+	intel_init_dpio(dev);
+
 	mutex_lock(&dev->struct_mutex);
 	intel_enable_gt_powersave(dev);
 	mutex_unlock(&dev->struct_mutex);
@@ -10357,7 +10852,7 @@ void i915_redisable_vga(struct drm_device *dev)
 	    (I915_READ(HSW_PWR_WELL_DRIVER) & HSW_PWR_WELL_STATE_ENABLED) == 0)
 		return;
 
-	if (I915_READ(vga_reg) != VGA_DISP_DISABLE) {
+	if (!(I915_READ(vga_reg) & VGA_DISP_DISABLE)) {
 		DRM_DEBUG_KMS("Something enabled VGA plane, disabling it\n");
 		i915_disable_vga(dev);
 	}
@@ -10380,6 +10875,7 @@ static void intel_modeset_readout_hw_state(struct drm_device *dev)
 								 &crtc->config);
 
 		crtc->base.enabled = crtc->active;
+		crtc->primary_enabled = crtc->active;
 
 		DRM_DEBUG_KMS("[CRTC:%d] hw state readout: %s\n",
 			      crtc->base.base.id,
@@ -10420,20 +10916,11 @@ static void intel_modeset_readout_hw_state(struct drm_device *dev)
 		}
 
 		encoder->connectors_active = false;
-		DRM_DEBUG_KMS("[ENCODER:%d:%s] hw state readout: %s, pipe=%i\n",
+		DRM_DEBUG_KMS("[ENCODER:%d:%s] hw state readout: %s, pipe %c\n",
 			      encoder->base.base.id,
 			      drm_get_encoder_name(&encoder->base),
 			      encoder->base.crtc ? "enabled" : "disabled",
-			      pipe);
-	}
-
-	list_for_each_entry(crtc, &dev->mode_config.crtc_list,
-			    base.head) {
-		if (!crtc->active)
-			continue;
-		if (dev_priv->display.get_clock)
-			dev_priv->display.get_clock(crtc,
-						    &crtc->config);
+			      pipe_name(pipe));
 	}
 
 	list_for_each_entry(connector, &dev->mode_config.connector_list,
@@ -10460,7 +10947,6 @@ void intel_modeset_setup_hw_state(struct drm_device *dev,
 {
 	struct drm_i915_private *dev_priv = dev->dev_private;
 	enum pipe pipe;
-	struct drm_plane *plane;
 	struct intel_crtc *crtc;
 	struct intel_encoder *encoder;
 	int i;
@@ -10507,7 +10993,12 @@ void intel_modeset_setup_hw_state(struct drm_device *dev,
 		pll->on = false;
 	}
 
+	if (IS_HASWELL(dev))
+		ilk_wm_get_hw_state(dev);
+
 	if (force_restore) {
+		i915_redisable_vga(dev);
+
 		/*
 		 * We need to use raw interfaces for restoring state to avoid
 		 * checking (bogus) intermediate states.
@@ -10519,10 +11010,6 @@ void intel_modeset_setup_hw_state(struct drm_device *dev,
 			__intel_set_mode(crtc, &crtc->mode, crtc->x, crtc->y,
 					 crtc->fb);
 		}
-		list_for_each_entry(plane, &dev->mode_config.plane_list, head)
-			intel_plane_restore(plane);
-
-		i915_redisable_vga(dev);
 	} else {
 		intel_modeset_update_staged_output_state(dev);
 	}
@@ -10545,6 +11032,7 @@ void intel_modeset_cleanup(struct drm_device *dev)
 {
 	struct drm_i915_private *dev_priv = dev->dev_private;
 	struct drm_crtc *crtc;
+	struct drm_connector *connector;
 
 	/*
 	 * Interrupts and polling as the first thing to avoid creating havoc.
@@ -10585,6 +11073,10 @@ void intel_modeset_cleanup(struct drm_device *dev)
 	/* destroy backlight, if any, before the connectors */
 	intel_panel_destroy_backlight(dev);
 
+	/* destroy the sysfs files before encoders/connectors */
+	list_for_each_entry(connector, &dev->mode_config.connector_list, head)
+		drm_sysfs_connector_remove(connector);
+
 	drm_mode_config_cleanup(dev);
 
 	intel_cleanup_overlay(dev);
@@ -10680,7 +11172,7 @@ intel_display_capture_error_state(struct drm_device *dev)
 	if (INTEL_INFO(dev)->num_pipes == 0)
 		return NULL;
 
-	error = kmalloc(sizeof(*error), GFP_ATOMIC);
+	error = kzalloc(sizeof(*error), GFP_ATOMIC);
 	if (error == NULL)
 		return NULL;
 
@@ -10688,6 +11180,9 @@ intel_display_capture_error_state(struct drm_device *dev)
 		error->power_well_driver = I915_READ(HSW_PWR_WELL_DRIVER);
 
 	for_each_pipe(i) {
+		if (!intel_display_power_enabled(dev, POWER_DOMAIN_PIPE(i)))
+			continue;
+
 		if (INTEL_INFO(dev)->gen <= 6 || IS_VALLEYVIEW(dev)) {
 			error->cursor[i].control = I915_READ(CURCNTR(i));
 			error->cursor[i].position = I915_READ(CURPOS(i));
@@ -10721,6 +11216,10 @@ intel_display_capture_error_state(struct drm_device *dev)
 	for (i = 0; i < error->num_transcoders; i++) {
 		enum transcoder cpu_transcoder = transcoders[i];
 
+		if (!intel_display_power_enabled(dev,
+				POWER_DOMAIN_TRANSCODER(cpu_transcoder)))
+			continue;
+
 		error->transcoder[i].cpu_transcoder = cpu_transcoder;
 
 		error->transcoder[i].conf = I915_READ(PIPECONF(cpu_transcoder));
@@ -10732,12 +11231,6 @@ intel_display_capture_error_state(struct drm_device *dev)
 		error->transcoder[i].vsync = I915_READ(VSYNC(cpu_transcoder));
 	}
 
-	/* In the code above we read the registers without checking if the power
-	 * well was on, so here we have to clear the FPGA_DBG_RM_NOCLAIM bit to
-	 * prevent the next I915_WRITE from detecting it and printing an error
-	 * message. */
-	intel_uncore_clear_errors(dev);
-
 	return error;
 }
 
@@ -10782,7 +11275,7 @@ intel_display_print_error_state(struct drm_i915_error_state_buf *m,
 	}
 
 	for (i = 0; i < error->num_transcoders; i++) {
-		err_printf(m, "  CPU transcoder: %c\n",
+		err_printf(m, "CPU transcoder: %c\n",
 			   transcoder_name(error->transcoder[i].cpu_transcoder));
 		err_printf(m, "  CONF: %08x\n", error->transcoder[i].conf);
 		err_printf(m, "  HTOTAL: %08x\n", error->transcoder[i].htotal);
diff --git a/drivers/gpu/drm/i915/intel_dp.c b/drivers/gpu/drm/i915/intel_dp.c
index 1a431377d83b..eb8139da9763 100644
--- a/drivers/gpu/drm/i915/intel_dp.c
+++ b/drivers/gpu/drm/i915/intel_dp.c
@@ -38,6 +38,32 @@
 
 #define DP_LINK_CHECK_TIMEOUT	(10 * 1000)
 
+struct dp_link_dpll {
+	int link_bw;
+	struct dpll dpll;
+};
+
+static const struct dp_link_dpll gen4_dpll[] = {
+	{ DP_LINK_BW_1_62,
+		{ .p1 = 2, .p2 = 10, .n = 2, .m1 = 23, .m2 = 8 } },
+	{ DP_LINK_BW_2_7,
+		{ .p1 = 1, .p2 = 10, .n = 1, .m1 = 14, .m2 = 2 } }
+};
+
+static const struct dp_link_dpll pch_dpll[] = {
+	{ DP_LINK_BW_1_62,
+		{ .p1 = 2, .p2 = 10, .n = 1, .m1 = 12, .m2 = 9 } },
+	{ DP_LINK_BW_2_7,
+		{ .p1 = 1, .p2 = 10, .n = 2, .m1 = 14, .m2 = 8 } }
+};
+
+static const struct dp_link_dpll vlv_dpll[] = {
+	{ DP_LINK_BW_1_62,
+		{ .p1 = 3, .p2 = 2, .n = 5, .m1 = 3, .m2 = 81 } },
+	{ DP_LINK_BW_2_7,
+		{ .p1 = 2, .p2 = 2, .n = 1, .m1 = 2, .m2 = 27 } }
+};
+
 /**
  * is_edp - is the given port attached to an eDP panel (either CPU or PCH)
  * @intel_dp: DP struct
@@ -211,24 +237,77 @@ intel_hrawclk(struct drm_device *dev)
 	}
 }
 
+static void
+intel_dp_init_panel_power_sequencer(struct drm_device *dev,
+				    struct intel_dp *intel_dp,
+				    struct edp_power_seq *out);
+static void
+intel_dp_init_panel_power_sequencer_registers(struct drm_device *dev,
+					      struct intel_dp *intel_dp,
+					      struct edp_power_seq *out);
+
+static enum pipe
+vlv_power_sequencer_pipe(struct intel_dp *intel_dp)
+{
+	struct intel_digital_port *intel_dig_port = dp_to_dig_port(intel_dp);
+	struct drm_crtc *crtc = intel_dig_port->base.base.crtc;
+	struct drm_device *dev = intel_dig_port->base.base.dev;
+	struct drm_i915_private *dev_priv = dev->dev_private;
+	enum port port = intel_dig_port->port;
+	enum pipe pipe;
+
+	/* modeset should have pipe */
+	if (crtc)
+		return to_intel_crtc(crtc)->pipe;
+
+	/* init time, try to find a pipe with this port selected */
+	for (pipe = PIPE_A; pipe <= PIPE_B; pipe++) {
+		u32 port_sel = I915_READ(VLV_PIPE_PP_ON_DELAYS(pipe)) &
+			PANEL_PORT_SELECT_MASK;
+		if (port_sel == PANEL_PORT_SELECT_DPB_VLV && port == PORT_B)
+			return pipe;
+		if (port_sel == PANEL_PORT_SELECT_DPC_VLV && port == PORT_C)
+			return pipe;
+	}
+
+	/* shrug */
+	return PIPE_A;
+}
+
+static u32 _pp_ctrl_reg(struct intel_dp *intel_dp)
+{
+	struct drm_device *dev = intel_dp_to_dev(intel_dp);
+
+	if (HAS_PCH_SPLIT(dev))
+		return PCH_PP_CONTROL;
+	else
+		return VLV_PIPE_PP_CONTROL(vlv_power_sequencer_pipe(intel_dp));
+}
+
+static u32 _pp_stat_reg(struct intel_dp *intel_dp)
+{
+	struct drm_device *dev = intel_dp_to_dev(intel_dp);
+
+	if (HAS_PCH_SPLIT(dev))
+		return PCH_PP_STATUS;
+	else
+		return VLV_PIPE_PP_STATUS(vlv_power_sequencer_pipe(intel_dp));
+}
+
 static bool ironlake_edp_have_panel_power(struct intel_dp *intel_dp)
 {
 	struct drm_device *dev = intel_dp_to_dev(intel_dp);
 	struct drm_i915_private *dev_priv = dev->dev_private;
-	u32 pp_stat_reg;
 
-	pp_stat_reg = IS_VALLEYVIEW(dev) ? PIPEA_PP_STATUS : PCH_PP_STATUS;
-	return (I915_READ(pp_stat_reg) & PP_ON) != 0;
+	return (I915_READ(_pp_stat_reg(intel_dp)) & PP_ON) != 0;
 }
 
 static bool ironlake_edp_have_panel_vdd(struct intel_dp *intel_dp)
 {
 	struct drm_device *dev = intel_dp_to_dev(intel_dp);
 	struct drm_i915_private *dev_priv = dev->dev_private;
-	u32 pp_ctrl_reg;
 
-	pp_ctrl_reg = IS_VALLEYVIEW(dev) ? PIPEA_PP_CONTROL : PCH_PP_CONTROL;
-	return (I915_READ(pp_ctrl_reg) & EDP_FORCE_VDD) != 0;
+	return (I915_READ(_pp_ctrl_reg(intel_dp)) & EDP_FORCE_VDD) != 0;
 }
 
 static void
@@ -236,19 +315,15 @@ intel_dp_check_edp(struct intel_dp *intel_dp)
 {
 	struct drm_device *dev = intel_dp_to_dev(intel_dp);
 	struct drm_i915_private *dev_priv = dev->dev_private;
-	u32 pp_stat_reg, pp_ctrl_reg;
 
 	if (!is_edp(intel_dp))
 		return;
 
-	pp_stat_reg = IS_VALLEYVIEW(dev) ? PIPEA_PP_STATUS : PCH_PP_STATUS;
-	pp_ctrl_reg = IS_VALLEYVIEW(dev) ? PIPEA_PP_CONTROL : PCH_PP_CONTROL;
-
 	if (!ironlake_edp_have_panel_power(intel_dp) && !ironlake_edp_have_panel_vdd(intel_dp)) {
 		WARN(1, "eDP powered off while attempting aux channel communication.\n");
 		DRM_DEBUG_KMS("Status 0x%08x Control 0x%08x\n",
-				I915_READ(pp_stat_reg),
-				I915_READ(pp_ctrl_reg));
+			      I915_READ(_pp_stat_reg(intel_dp)),
+			      I915_READ(_pp_ctrl_reg(intel_dp)));
 	}
 }
 
@@ -330,6 +405,7 @@ intel_dp_aux_ch(struct intel_dp *intel_dp,
 	uint32_t status;
 	int try, precharge, clock = 0;
 	bool has_aux_irq = INTEL_INFO(dev)->gen >= 5 && !IS_VALLEYVIEW(dev);
+	uint32_t timeout;
 
 	/* dp aux is extremely sensitive to irq latency, hence request the
 	 * lowest possible wakeup latency and so prevent the cpu from going into
@@ -344,6 +420,11 @@ intel_dp_aux_ch(struct intel_dp *intel_dp,
 	else
 		precharge = 5;
 
+	if (IS_BROADWELL(dev) && ch_ctl == DPA_AUX_CH_CTL)
+		timeout = DP_AUX_CH_CTL_TIME_OUT_600us;
+	else
+		timeout = DP_AUX_CH_CTL_TIME_OUT_400us;
+
 	intel_aux_display_runtime_get(dev_priv);
 
 	/* Try to wait for any previous AUX channel activity */
@@ -361,6 +442,12 @@ intel_dp_aux_ch(struct intel_dp *intel_dp,
 		goto out;
 	}
 
+	/* Only 5 data registers! */
+	if (WARN_ON(send_bytes > 20 || recv_size > 20)) {
+		ret = -E2BIG;
+		goto out;
+	}
+
 	while ((aux_clock_divider = get_aux_clock_divider(intel_dp, clock++))) {
 		/* Must try at least 3 times according to DP spec */
 		for (try = 0; try < 5; try++) {
@@ -373,7 +460,7 @@ intel_dp_aux_ch(struct intel_dp *intel_dp,
 			I915_WRITE(ch_ctl,
 				   DP_AUX_CH_CTL_SEND_BUSY |
 				   (has_aux_irq ? DP_AUX_CH_CTL_INTERRUPT : 0) |
-				   DP_AUX_CH_CTL_TIME_OUT_400us |
+				   timeout |
 				   (send_bytes << DP_AUX_CH_CTL_MESSAGE_SIZE_SHIFT) |
 				   (precharge << DP_AUX_CH_CTL_PRECHARGE_2US_SHIFT) |
 				   (aux_clock_divider << DP_AUX_CH_CTL_BIT_CLOCK_2X_SHIFT) |
@@ -451,9 +538,10 @@ intel_dp_aux_native_write(struct intel_dp *intel_dp,
 	int msg_bytes;
 	uint8_t	ack;
 
+	if (WARN_ON(send_bytes > 16))
+		return -E2BIG;
+
 	intel_dp_check_edp(intel_dp);
-	if (send_bytes > 16)
-		return -1;
 	msg[0] = AUX_NATIVE_WRITE << 4;
 	msg[1] = address >> 8;
 	msg[2] = address & 0xff;
@@ -494,6 +582,9 @@ intel_dp_aux_native_read(struct intel_dp *intel_dp,
 	uint8_t ack;
 	int ret;
 
+	if (WARN_ON(recv_bytes > 19))
+		return -E2BIG;
+
 	intel_dp_check_edp(intel_dp);
 	msg[0] = AUX_NATIVE_READ << 4;
 	msg[1] = address >> 8;
@@ -538,6 +629,7 @@ intel_dp_i2c_aux_ch(struct i2c_adapter *adapter, int mode,
 	int reply_bytes;
 	int ret;
 
+	ironlake_edp_panel_vdd_on(intel_dp);
 	intel_dp_check_edp(intel_dp);
 	/* Set up the command byte */
 	if (mode & MODE_I2C_READ)
@@ -569,13 +661,18 @@ intel_dp_i2c_aux_ch(struct i2c_adapter *adapter, int mode,
 		break;
 	}
 
-	for (retry = 0; retry < 5; retry++) {
+	/*
+	 * DP1.2 sections 2.7.7.1.5.6.1 and 2.7.7.1.6.6.1: A DP Source device is
+	 * required to retry at least seven times upon receiving AUX_DEFER
+	 * before giving up the AUX transaction.
+	 */
+	for (retry = 0; retry < 7; retry++) {
 		ret = intel_dp_aux_ch(intel_dp,
 				      msg, msg_bytes,
 				      reply, reply_bytes);
 		if (ret < 0) {
 			DRM_DEBUG_KMS("aux_ch failed %d\n", ret);
-			return ret;
+			goto out;
 		}
 
 		switch (reply[0] & AUX_NATIVE_REPLY_MASK) {
@@ -586,7 +683,8 @@ intel_dp_i2c_aux_ch(struct i2c_adapter *adapter, int mode,
 			break;
 		case AUX_NATIVE_REPLY_NACK:
 			DRM_DEBUG_KMS("aux_ch native nack\n");
-			return -EREMOTEIO;
+			ret = -EREMOTEIO;
+			goto out;
 		case AUX_NATIVE_REPLY_DEFER:
 			/*
 			 * For now, just give more slack to branch devices. We
@@ -604,7 +702,8 @@ intel_dp_i2c_aux_ch(struct i2c_adapter *adapter, int mode,
 		default:
 			DRM_ERROR("aux_ch invalid native reply 0x%02x\n",
 				  reply[0]);
-			return -EREMOTEIO;
+			ret = -EREMOTEIO;
+			goto out;
 		}
 
 		switch (reply[0] & AUX_I2C_REPLY_MASK) {
@@ -612,22 +711,29 @@ intel_dp_i2c_aux_ch(struct i2c_adapter *adapter, int mode,
 			if (mode == MODE_I2C_READ) {
 				*read_byte = reply[1];
 			}
-			return reply_bytes - 1;
+			ret = reply_bytes - 1;
+			goto out;
 		case AUX_I2C_REPLY_NACK:
 			DRM_DEBUG_KMS("aux_i2c nack\n");
-			return -EREMOTEIO;
+			ret = -EREMOTEIO;
+			goto out;
 		case AUX_I2C_REPLY_DEFER:
 			DRM_DEBUG_KMS("aux_i2c defer\n");
 			udelay(100);
 			break;
 		default:
 			DRM_ERROR("aux_i2c invalid reply 0x%02x\n", reply[0]);
-			return -EREMOTEIO;
+			ret = -EREMOTEIO;
+			goto out;
 		}
 	}
 
 	DRM_ERROR("too many retries, giving up\n");
-	return -EREMOTEIO;
+	ret = -EREMOTEIO;
+
+out:
+	ironlake_edp_panel_vdd_off(intel_dp, false);
+	return ret;
 }
 
 static int
@@ -647,11 +753,9 @@ intel_dp_i2c_init(struct intel_dp *intel_dp,
 	strncpy(intel_dp->adapter.name, name, sizeof(intel_dp->adapter.name) - 1);
 	intel_dp->adapter.name[sizeof(intel_dp->adapter.name) - 1] = '\0';
 	intel_dp->adapter.algo_data = &intel_dp->algo;
-	intel_dp->adapter.dev.parent = &intel_connector->base.kdev;
+	intel_dp->adapter.dev.parent = intel_connector->base.kdev;
 
-	ironlake_edp_panel_vdd_on(intel_dp);
 	ret = i2c_dp_aux_add_bus(&intel_dp->adapter);
-	ironlake_edp_panel_vdd_off(intel_dp, false);
 	return ret;
 }
 
@@ -660,41 +764,30 @@ intel_dp_set_clock(struct intel_encoder *encoder,
 		   struct intel_crtc_config *pipe_config, int link_bw)
 {
 	struct drm_device *dev = encoder->base.dev;
+	const struct dp_link_dpll *divisor = NULL;
+	int i, count = 0;
 
 	if (IS_G4X(dev)) {
-		if (link_bw == DP_LINK_BW_1_62) {
-			pipe_config->dpll.p1 = 2;
-			pipe_config->dpll.p2 = 10;
-			pipe_config->dpll.n = 2;
-			pipe_config->dpll.m1 = 23;
-			pipe_config->dpll.m2 = 8;
-		} else {
-			pipe_config->dpll.p1 = 1;
-			pipe_config->dpll.p2 = 10;
-			pipe_config->dpll.n = 1;
-			pipe_config->dpll.m1 = 14;
-			pipe_config->dpll.m2 = 2;
-		}
-		pipe_config->clock_set = true;
+		divisor = gen4_dpll;
+		count = ARRAY_SIZE(gen4_dpll);
 	} else if (IS_HASWELL(dev)) {
 		/* Haswell has special-purpose DP DDI clocks. */
 	} else if (HAS_PCH_SPLIT(dev)) {
-		if (link_bw == DP_LINK_BW_1_62) {
-			pipe_config->dpll.n = 1;
-			pipe_config->dpll.p1 = 2;
-			pipe_config->dpll.p2 = 10;
-			pipe_config->dpll.m1 = 12;
-			pipe_config->dpll.m2 = 9;
-		} else {
-			pipe_config->dpll.n = 2;
-			pipe_config->dpll.p1 = 1;
-			pipe_config->dpll.p2 = 10;
-			pipe_config->dpll.m1 = 14;
-			pipe_config->dpll.m2 = 8;
-		}
-		pipe_config->clock_set = true;
+		divisor = pch_dpll;
+		count = ARRAY_SIZE(pch_dpll);
 	} else if (IS_VALLEYVIEW(dev)) {
-		/* FIXME: Need to figure out optimized DP clocks for vlv. */
+		divisor = vlv_dpll;
+		count = ARRAY_SIZE(vlv_dpll);
+	}
+
+	if (divisor && count) {
+		for (i = 0; i < count; i++) {
+			if (link_bw == divisor[i].link_bw) {
+				pipe_config->dpll = divisor[i].dpll;
+				pipe_config->clock_set = true;
+				break;
+			}
+		}
 	}
 }
 
@@ -737,19 +830,22 @@ intel_dp_compute_config(struct intel_encoder *encoder,
 
 	DRM_DEBUG_KMS("DP link computation with max lane count %i "
 		      "max bw %02x pixel clock %iKHz\n",
-		      max_lane_count, bws[max_clock], adjusted_mode->clock);
+		      max_lane_count, bws[max_clock],
+		      adjusted_mode->crtc_clock);
 
 	/* Walk through all bpp values. Luckily they're all nicely spaced with 2
 	 * bpc in between. */
 	bpp = pipe_config->pipe_bpp;
-	if (is_edp(intel_dp) && dev_priv->vbt.edp_bpp) {
+	if (is_edp(intel_dp) && dev_priv->vbt.edp_bpp &&
+	    dev_priv->vbt.edp_bpp < bpp) {
 		DRM_DEBUG_KMS("clamping bpp for eDP panel to BIOS-provided %i\n",
 			      dev_priv->vbt.edp_bpp);
-		bpp = min_t(int, bpp, dev_priv->vbt.edp_bpp);
+		bpp = dev_priv->vbt.edp_bpp;
 	}
 
 	for (; bpp >= 6*3; bpp -= 2*3) {
-		mode_rate = intel_dp_link_required(adjusted_mode->clock, bpp);
+		mode_rate = intel_dp_link_required(adjusted_mode->crtc_clock,
+						   bpp);
 
 		for (clock = 0; clock <= max_clock; clock++) {
 			for (lane_count = 1; lane_count <= max_lane_count; lane_count <<= 1) {
@@ -794,7 +890,8 @@ found:
 		      mode_rate, link_avail);
 
 	intel_link_compute_m_n(bpp, lane_count,
-			       adjusted_mode->clock, pipe_config->port_clock,
+			       adjusted_mode->crtc_clock,
+			       pipe_config->port_clock,
 			       &pipe_config->dp_m_n);
 
 	intel_dp_set_clock(encoder, pipe_config, intel_dp->link_bw);
@@ -802,21 +899,6 @@ found:
 	return true;
 }
 
-void intel_dp_init_link_config(struct intel_dp *intel_dp)
-{
-	memset(intel_dp->link_configuration, 0, DP_LINK_CONFIGURATION_SIZE);
-	intel_dp->link_configuration[0] = intel_dp->link_bw;
-	intel_dp->link_configuration[1] = intel_dp->lane_count;
-	intel_dp->link_configuration[8] = DP_SET_ANSI_8B10B;
-	/*
-	 * Check for DPCD version > 1.1 and enhanced framing support
-	 */
-	if (intel_dp->dpcd[DP_DPCD_REV] >= 0x11 &&
-	    (intel_dp->dpcd[DP_MAX_LANE_COUNT] & DP_ENHANCED_FRAME_CAP)) {
-		intel_dp->link_configuration[1] |= DP_LANE_COUNT_ENHANCED_FRAME_EN;
-	}
-}
-
 static void ironlake_set_pll_cpu_edp(struct intel_dp *intel_dp)
 {
 	struct intel_digital_port *dig_port = dp_to_dig_port(intel_dp);
@@ -889,8 +971,6 @@ static void intel_dp_mode_set(struct intel_encoder *encoder)
 		intel_write_eld(&encoder->base, adjusted_mode);
 	}
 
-	intel_dp_init_link_config(intel_dp);
-
 	/* Split out the IBX/CPU vs CPT settings */
 
 	if (port == PORT_A && IS_GEN7(dev) && !IS_VALLEYVIEW(dev)) {
@@ -900,7 +980,7 @@ static void intel_dp_mode_set(struct intel_encoder *encoder)
 			intel_dp->DP |= DP_SYNC_VS_HIGH;
 		intel_dp->DP |= DP_LINK_TRAIN_OFF_CPT;
 
-		if (intel_dp->link_configuration[1] & DP_LANE_COUNT_ENHANCED_FRAME_EN)
+		if (drm_dp_enhanced_frame_cap(intel_dp->dpcd))
 			intel_dp->DP |= DP_ENHANCED_FRAMING;
 
 		intel_dp->DP |= crtc->pipe << 29;
@@ -914,7 +994,7 @@ static void intel_dp_mode_set(struct intel_encoder *encoder)
 			intel_dp->DP |= DP_SYNC_VS_HIGH;
 		intel_dp->DP |= DP_LINK_TRAIN_OFF;
 
-		if (intel_dp->link_configuration[1] & DP_LANE_COUNT_ENHANCED_FRAME_EN)
+		if (drm_dp_enhanced_frame_cap(intel_dp->dpcd))
 			intel_dp->DP |= DP_ENHANCED_FRAMING;
 
 		if (crtc->pipe == 1)
@@ -944,8 +1024,8 @@ static void ironlake_wait_panel_status(struct intel_dp *intel_dp,
 	struct drm_i915_private *dev_priv = dev->dev_private;
 	u32 pp_stat_reg, pp_ctrl_reg;
 
-	pp_stat_reg = IS_VALLEYVIEW(dev) ? PIPEA_PP_STATUS : PCH_PP_STATUS;
-	pp_ctrl_reg = IS_VALLEYVIEW(dev) ? PIPEA_PP_CONTROL : PCH_PP_CONTROL;
+	pp_stat_reg = _pp_stat_reg(intel_dp);
+	pp_ctrl_reg = _pp_ctrl_reg(intel_dp);
 
 	DRM_DEBUG_KMS("mask %08x value %08x status %08x control %08x\n",
 			mask, value,
@@ -987,11 +1067,8 @@ static  u32 ironlake_get_pp_control(struct intel_dp *intel_dp)
 	struct drm_device *dev = intel_dp_to_dev(intel_dp);
 	struct drm_i915_private *dev_priv = dev->dev_private;
 	u32 control;
-	u32 pp_ctrl_reg;
-
-	pp_ctrl_reg = IS_VALLEYVIEW(dev) ? PIPEA_PP_CONTROL : PCH_PP_CONTROL;
-	control = I915_READ(pp_ctrl_reg);
 
+	control = I915_READ(_pp_ctrl_reg(intel_dp));
 	control &= ~PANEL_UNLOCK_MASK;
 	control |= PANEL_UNLOCK_REGS;
 	return control;
@@ -1006,17 +1083,16 @@ void ironlake_edp_panel_vdd_on(struct intel_dp *intel_dp)
 
 	if (!is_edp(intel_dp))
 		return;
-	DRM_DEBUG_KMS("Turn eDP VDD on\n");
 
 	WARN(intel_dp->want_panel_vdd,
 	     "eDP VDD already requested on\n");
 
 	intel_dp->want_panel_vdd = true;
 
-	if (ironlake_edp_have_panel_vdd(intel_dp)) {
-		DRM_DEBUG_KMS("eDP VDD already on\n");
+	if (ironlake_edp_have_panel_vdd(intel_dp))
 		return;
-	}
+
+	DRM_DEBUG_KMS("Turning eDP VDD on\n");
 
 	if (!ironlake_edp_have_panel_power(intel_dp))
 		ironlake_wait_panel_power_cycle(intel_dp);
@@ -1024,8 +1100,8 @@ void ironlake_edp_panel_vdd_on(struct intel_dp *intel_dp)
 	pp = ironlake_get_pp_control(intel_dp);
 	pp |= EDP_FORCE_VDD;
 
-	pp_stat_reg = IS_VALLEYVIEW(dev) ? PIPEA_PP_STATUS : PCH_PP_STATUS;
-	pp_ctrl_reg = IS_VALLEYVIEW(dev) ? PIPEA_PP_CONTROL : PCH_PP_CONTROL;
+	pp_stat_reg = _pp_stat_reg(intel_dp);
+	pp_ctrl_reg = _pp_ctrl_reg(intel_dp);
 
 	I915_WRITE(pp_ctrl_reg, pp);
 	POSTING_READ(pp_ctrl_reg);
@@ -1050,11 +1126,13 @@ static void ironlake_panel_vdd_off_sync(struct intel_dp *intel_dp)
 	WARN_ON(!mutex_is_locked(&dev->mode_config.mutex));
 
 	if (!intel_dp->want_panel_vdd && ironlake_edp_have_panel_vdd(intel_dp)) {
+		DRM_DEBUG_KMS("Turning eDP VDD off\n");
+
 		pp = ironlake_get_pp_control(intel_dp);
 		pp &= ~EDP_FORCE_VDD;
 
-		pp_stat_reg = IS_VALLEYVIEW(dev) ? PIPEA_PP_STATUS : PCH_PP_STATUS;
-		pp_ctrl_reg = IS_VALLEYVIEW(dev) ? PIPEA_PP_CONTROL : PCH_PP_CONTROL;
+		pp_ctrl_reg = _pp_ctrl_reg(intel_dp);
+		pp_stat_reg = _pp_stat_reg(intel_dp);
 
 		I915_WRITE(pp_ctrl_reg, pp);
 		POSTING_READ(pp_ctrl_reg);
@@ -1082,7 +1160,6 @@ void ironlake_edp_panel_vdd_off(struct intel_dp *intel_dp, bool sync)
 	if (!is_edp(intel_dp))
 		return;
 
-	DRM_DEBUG_KMS("Turn eDP VDD off %d\n", intel_dp->want_panel_vdd);
 	WARN(!intel_dp->want_panel_vdd, "eDP VDD not forced on");
 
 	intel_dp->want_panel_vdd = false;
@@ -1119,20 +1196,19 @@ void ironlake_edp_panel_on(struct intel_dp *intel_dp)
 
 	ironlake_wait_panel_power_cycle(intel_dp);
 
+	pp_ctrl_reg = _pp_ctrl_reg(intel_dp);
 	pp = ironlake_get_pp_control(intel_dp);
 	if (IS_GEN5(dev)) {
 		/* ILK workaround: disable reset around power sequence */
 		pp &= ~PANEL_POWER_RESET;
-		I915_WRITE(PCH_PP_CONTROL, pp);
-		POSTING_READ(PCH_PP_CONTROL);
+		I915_WRITE(pp_ctrl_reg, pp);
+		POSTING_READ(pp_ctrl_reg);
 	}
 
 	pp |= POWER_TARGET_ON;
 	if (!IS_GEN5(dev))
 		pp |= PANEL_POWER_RESET;
 
-	pp_ctrl_reg = IS_VALLEYVIEW(dev) ? PIPEA_PP_CONTROL : PCH_PP_CONTROL;
-
 	I915_WRITE(pp_ctrl_reg, pp);
 	POSTING_READ(pp_ctrl_reg);
 
@@ -1140,8 +1216,8 @@ void ironlake_edp_panel_on(struct intel_dp *intel_dp)
 
 	if (IS_GEN5(dev)) {
 		pp |= PANEL_POWER_RESET; /* restore panel reset bit */
-		I915_WRITE(PCH_PP_CONTROL, pp);
-		POSTING_READ(PCH_PP_CONTROL);
+		I915_WRITE(pp_ctrl_reg, pp);
+		POSTING_READ(pp_ctrl_reg);
 	}
 }
 
@@ -1164,7 +1240,7 @@ void ironlake_edp_panel_off(struct intel_dp *intel_dp)
 	 * panels get very unhappy and cease to work. */
 	pp &= ~(POWER_TARGET_ON | EDP_FORCE_VDD | PANEL_POWER_RESET | EDP_BLC_ENABLE);
 
-	pp_ctrl_reg = IS_VALLEYVIEW(dev) ? PIPEA_PP_CONTROL : PCH_PP_CONTROL;
+	pp_ctrl_reg = _pp_ctrl_reg(intel_dp);
 
 	I915_WRITE(pp_ctrl_reg, pp);
 	POSTING_READ(pp_ctrl_reg);
@@ -1179,7 +1255,6 @@ void ironlake_edp_backlight_on(struct intel_dp *intel_dp)
 	struct intel_digital_port *intel_dig_port = dp_to_dig_port(intel_dp);
 	struct drm_device *dev = intel_dig_port->base.base.dev;
 	struct drm_i915_private *dev_priv = dev->dev_private;
-	int pipe = to_intel_crtc(intel_dig_port->base.base.crtc)->pipe;
 	u32 pp;
 	u32 pp_ctrl_reg;
 
@@ -1197,12 +1272,12 @@ void ironlake_edp_backlight_on(struct intel_dp *intel_dp)
 	pp = ironlake_get_pp_control(intel_dp);
 	pp |= EDP_BLC_ENABLE;
 
-	pp_ctrl_reg = IS_VALLEYVIEW(dev) ? PIPEA_PP_CONTROL : PCH_PP_CONTROL;
+	pp_ctrl_reg = _pp_ctrl_reg(intel_dp);
 
 	I915_WRITE(pp_ctrl_reg, pp);
 	POSTING_READ(pp_ctrl_reg);
 
-	intel_panel_enable_backlight(dev, pipe);
+	intel_panel_enable_backlight(intel_dp->attached_connector);
 }
 
 void ironlake_edp_backlight_off(struct intel_dp *intel_dp)
@@ -1215,13 +1290,13 @@ void ironlake_edp_backlight_off(struct intel_dp *intel_dp)
 	if (!is_edp(intel_dp))
 		return;
 
-	intel_panel_disable_backlight(dev);
+	intel_panel_disable_backlight(intel_dp->attached_connector);
 
 	DRM_DEBUG_KMS("\n");
 	pp = ironlake_get_pp_control(intel_dp);
 	pp &= ~EDP_BLC_ENABLE;
 
-	pp_ctrl_reg = IS_VALLEYVIEW(dev) ? PIPEA_PP_CONTROL : PCH_PP_CONTROL;
+	pp_ctrl_reg = _pp_ctrl_reg(intel_dp);
 
 	I915_WRITE(pp_ctrl_reg, pp);
 	POSTING_READ(pp_ctrl_reg);
@@ -1368,6 +1443,7 @@ static void intel_dp_get_config(struct intel_encoder *encoder,
 	struct drm_i915_private *dev_priv = dev->dev_private;
 	enum port port = dp_to_dig_port(intel_dp)->port;
 	struct intel_crtc *crtc = to_intel_crtc(encoder->base.crtc);
+	int dotclock;
 
 	if ((port == PORT_A) || !HAS_PCH_CPT(dev)) {
 		tmp = I915_READ(intel_dp->output_reg);
@@ -1395,13 +1471,25 @@ static void intel_dp_get_config(struct intel_encoder *encoder,
 
 	pipe_config->adjusted_mode.flags |= flags;
 
-	if (dp_to_dig_port(intel_dp)->port == PORT_A) {
+	pipe_config->has_dp_encoder = true;
+
+	intel_dp_get_m_n(crtc, pipe_config);
+
+	if (port == PORT_A) {
 		if ((I915_READ(DP_A) & DP_PLL_FREQ_MASK) == DP_PLL_FREQ_160MHZ)
 			pipe_config->port_clock = 162000;
 		else
 			pipe_config->port_clock = 270000;
 	}
 
+	dotclock = intel_dotclock_calculate(pipe_config->port_clock,
+					    &pipe_config->dp_m_n);
+
+	if (HAS_PCH_SPLIT(dev_priv->dev) && port != PORT_A)
+		ironlake_check_encoder_dotclock(pipe_config, dotclock);
+
+	pipe_config->adjusted_mode.crtc_clock = dotclock;
+
 	if (is_edp(intel_dp) && dev_priv->vbt.edp_bpp &&
 	    pipe_config->pipe_bpp > dev_priv->vbt.edp_bpp) {
 		/*
@@ -1423,20 +1511,21 @@ static void intel_dp_get_config(struct intel_encoder *encoder,
 	}
 }
 
-static bool is_edp_psr(struct intel_dp *intel_dp)
+static bool is_edp_psr(struct drm_device *dev)
 {
-	return is_edp(intel_dp) &&
-		intel_dp->psr_dpcd[0] & DP_PSR_IS_SUPPORTED;
+	struct drm_i915_private *dev_priv = dev->dev_private;
+
+	return dev_priv->psr.sink_support;
 }
 
 static bool intel_edp_is_psr_enabled(struct drm_device *dev)
 {
 	struct drm_i915_private *dev_priv = dev->dev_private;
 
-	if (!IS_HASWELL(dev))
+	if (!HAS_PSR(dev))
 		return false;
 
-	return I915_READ(EDP_PSR_CTL) & EDP_PSR_ENABLE;
+	return I915_READ(EDP_PSR_CTL(dev)) & EDP_PSR_ENABLE;
 }
 
 static void intel_edp_psr_write_vsc(struct intel_dp *intel_dp,
@@ -1486,7 +1575,7 @@ static void intel_edp_psr_setup(struct intel_dp *intel_dp)
 	intel_edp_psr_write_vsc(intel_dp, &psr_vsc);
 
 	/* Avoid continuous PSR exit by masking memup and hpd */
-	I915_WRITE(EDP_PSR_DEBUG_CTL, EDP_PSR_DEBUG_MASK_MEMUP |
+	I915_WRITE(EDP_PSR_DEBUG_CTL(dev), EDP_PSR_DEBUG_MASK_MEMUP |
 		   EDP_PSR_DEBUG_MASK_HPD | EDP_PSR_DEBUG_MASK_LPSP);
 
 	intel_dp->psr_setup_done = true;
@@ -1511,9 +1600,9 @@ static void intel_edp_psr_enable_sink(struct intel_dp *intel_dp)
 					    DP_PSR_MAIN_LINK_ACTIVE);
 
 	/* Setup AUX registers */
-	I915_WRITE(EDP_PSR_AUX_DATA1, EDP_PSR_DPCD_COMMAND);
-	I915_WRITE(EDP_PSR_AUX_DATA2, EDP_PSR_DPCD_NORMAL_OPERATION);
-	I915_WRITE(EDP_PSR_AUX_CTL,
+	I915_WRITE(EDP_PSR_AUX_DATA1(dev), EDP_PSR_DPCD_COMMAND);
+	I915_WRITE(EDP_PSR_AUX_DATA2(dev), EDP_PSR_DPCD_NORMAL_OPERATION);
+	I915_WRITE(EDP_PSR_AUX_CTL(dev),
 		   DP_AUX_CH_CTL_TIME_OUT_400us |
 		   (msg_size << DP_AUX_CH_CTL_MESSAGE_SIZE_SHIFT) |
 		   (precharge << DP_AUX_CH_CTL_PRECHARGE_2US_SHIFT) |
@@ -1527,6 +1616,7 @@ static void intel_edp_psr_enable_source(struct intel_dp *intel_dp)
 	uint32_t max_sleep_time = 0x1f;
 	uint32_t idle_frames = 1;
 	uint32_t val = 0x0;
+	const uint32_t link_entry_time = EDP_PSR_MIN_LINK_ENTRY_TIME_8_LINES;
 
 	if (intel_dp->psr_dpcd[1] & DP_PSR_NO_TRAIN_ON_EXIT) {
 		val |= EDP_PSR_LINK_STANDBY;
@@ -1536,8 +1626,8 @@ static void intel_edp_psr_enable_source(struct intel_dp *intel_dp)
 	} else
 		val |= EDP_PSR_LINK_DISABLE;
 
-	I915_WRITE(EDP_PSR_CTL, val |
-		   EDP_PSR_MIN_LINK_ENTRY_TIME_8_LINES |
+	I915_WRITE(EDP_PSR_CTL(dev), val |
+		   IS_BROADWELL(dev) ? 0 : link_entry_time |
 		   max_sleep_time << EDP_PSR_MAX_SLEEP_TIME_SHIFT |
 		   idle_frames << EDP_PSR_IDLE_FRAME_SHIFT |
 		   EDP_PSR_ENABLE);
@@ -1553,42 +1643,33 @@ static bool intel_edp_psr_match_conditions(struct intel_dp *intel_dp)
 	struct drm_i915_gem_object *obj = to_intel_framebuffer(crtc->fb)->obj;
 	struct intel_encoder *intel_encoder = &dp_to_dig_port(intel_dp)->base;
 
-	if (!IS_HASWELL(dev)) {
+	dev_priv->psr.source_ok = false;
+
+	if (!HAS_PSR(dev)) {
 		DRM_DEBUG_KMS("PSR not supported on this platform\n");
-		dev_priv->no_psr_reason = PSR_NO_SOURCE;
 		return false;
 	}
 
 	if ((intel_encoder->type != INTEL_OUTPUT_EDP) ||
 	    (dig_port->port != PORT_A)) {
 		DRM_DEBUG_KMS("HSW ties PSR to DDI A (eDP)\n");
-		dev_priv->no_psr_reason = PSR_HSW_NOT_DDIA;
-		return false;
-	}
-
-	if (!is_edp_psr(intel_dp)) {
-		DRM_DEBUG_KMS("PSR not supported by this panel\n");
-		dev_priv->no_psr_reason = PSR_NO_SINK;
 		return false;
 	}
 
 	if (!i915_enable_psr) {
 		DRM_DEBUG_KMS("PSR disable by flag\n");
-		dev_priv->no_psr_reason = PSR_MODULE_PARAM;
 		return false;
 	}
 
 	crtc = dig_port->base.base.crtc;
 	if (crtc == NULL) {
 		DRM_DEBUG_KMS("crtc not active for PSR\n");
-		dev_priv->no_psr_reason = PSR_CRTC_NOT_ACTIVE;
 		return false;
 	}
 
 	intel_crtc = to_intel_crtc(crtc);
-	if (!intel_crtc->active || !crtc->fb || !crtc->mode.clock) {
+	if (!intel_crtc_active(crtc)) {
 		DRM_DEBUG_KMS("crtc not active for PSR\n");
-		dev_priv->no_psr_reason = PSR_CRTC_NOT_ACTIVE;
 		return false;
 	}
 
@@ -1596,29 +1677,26 @@ static bool intel_edp_psr_match_conditions(struct intel_dp *intel_dp)
 	if (obj->tiling_mode != I915_TILING_X ||
 	    obj->fence_reg == I915_FENCE_REG_NONE) {
 		DRM_DEBUG_KMS("PSR condition failed: fb not tiled or fenced\n");
-		dev_priv->no_psr_reason = PSR_NOT_TILED;
 		return false;
 	}
 
 	if (I915_READ(SPRCTL(intel_crtc->pipe)) & SPRITE_ENABLE) {
 		DRM_DEBUG_KMS("PSR condition failed: Sprite is Enabled\n");
-		dev_priv->no_psr_reason = PSR_SPRITE_ENABLED;
 		return false;
 	}
 
 	if (I915_READ(HSW_STEREO_3D_CTL(intel_crtc->config.cpu_transcoder)) &
 	    S3D_ENABLE) {
 		DRM_DEBUG_KMS("PSR condition failed: Stereo 3D is Enabled\n");
-		dev_priv->no_psr_reason = PSR_S3D_ENABLED;
 		return false;
 	}
 
-	if (crtc->mode.flags & DRM_MODE_FLAG_INTERLACE) {
+	if (intel_crtc->config.adjusted_mode.flags & DRM_MODE_FLAG_INTERLACE) {
 		DRM_DEBUG_KMS("PSR condition failed: Interlaced is Enabled\n");
-		dev_priv->no_psr_reason = PSR_INTERLACED_ENABLED;
 		return false;
 	}
 
+	dev_priv->psr.source_ok = true;
 	return true;
 }
 
@@ -1657,10 +1735,11 @@ void intel_edp_psr_disable(struct intel_dp *intel_dp)
 	if (!intel_edp_is_psr_enabled(dev))
 		return;
 
-	I915_WRITE(EDP_PSR_CTL, I915_READ(EDP_PSR_CTL) & ~EDP_PSR_ENABLE);
+	I915_WRITE(EDP_PSR_CTL(dev),
+		   I915_READ(EDP_PSR_CTL(dev)) & ~EDP_PSR_ENABLE);
 
 	/* Wait till PSR is idle */
-	if (_wait_for((I915_READ(EDP_PSR_STATUS_CTL) &
+	if (_wait_for((I915_READ(EDP_PSR_STATUS_CTL(dev)) &
 		       EDP_PSR_STATUS_STATE_MASK) == 0, 2000, 10))
 		DRM_ERROR("Timed out waiting for PSR Idle State\n");
 }
@@ -1674,7 +1753,7 @@ void intel_edp_psr_update(struct drm_device *dev)
 		if (encoder->type == INTEL_OUTPUT_EDP) {
 			intel_dp = enc_to_intel_dp(&encoder->base);
 
-			if (!is_edp_psr(intel_dp))
+			if (!is_edp_psr(dev))
 				return;
 
 			if (!intel_edp_psr_match_conditions(intel_dp))
@@ -1733,14 +1812,24 @@ static void intel_enable_dp(struct intel_encoder *encoder)
 	ironlake_edp_panel_vdd_off(intel_dp, true);
 	intel_dp_complete_link_train(intel_dp);
 	intel_dp_stop_link_train(intel_dp);
+}
+
+static void g4x_enable_dp(struct intel_encoder *encoder)
+{
+	struct intel_dp *intel_dp = enc_to_intel_dp(&encoder->base);
+
+	intel_enable_dp(encoder);
 	ironlake_edp_backlight_on(intel_dp);
 }
 
 static void vlv_enable_dp(struct intel_encoder *encoder)
 {
+	struct intel_dp *intel_dp = enc_to_intel_dp(&encoder->base);
+
+	ironlake_edp_backlight_on(intel_dp);
 }
 
-static void intel_pre_enable_dp(struct intel_encoder *encoder)
+static void g4x_pre_enable_dp(struct intel_encoder *encoder)
 {
 	struct intel_dp *intel_dp = enc_to_intel_dp(&encoder->base);
 	struct intel_digital_port *dport = dp_to_dig_port(intel_dp);
@@ -1758,53 +1847,59 @@ static void vlv_pre_enable_dp(struct intel_encoder *encoder)
 	struct intel_crtc *intel_crtc = to_intel_crtc(encoder->base.crtc);
 	int port = vlv_dport_to_channel(dport);
 	int pipe = intel_crtc->pipe;
+	struct edp_power_seq power_seq;
 	u32 val;
 
 	mutex_lock(&dev_priv->dpio_lock);
 
-	val = vlv_dpio_read(dev_priv, DPIO_DATA_LANE_A(port));
+	val = vlv_dpio_read(dev_priv, pipe, DPIO_DATA_LANE_A(port));
 	val = 0;
 	if (pipe)
 		val |= (1<<21);
 	else
 		val &= ~(1<<21);
 	val |= 0x001000c4;
-	vlv_dpio_write(dev_priv, DPIO_DATA_CHANNEL(port), val);
-	vlv_dpio_write(dev_priv, DPIO_PCS_CLOCKBUF0(port), 0x00760018);
-	vlv_dpio_write(dev_priv, DPIO_PCS_CLOCKBUF8(port), 0x00400888);
+	vlv_dpio_write(dev_priv, pipe, DPIO_DATA_CHANNEL(port), val);
+	vlv_dpio_write(dev_priv, pipe, DPIO_PCS_CLOCKBUF0(port), 0x00760018);
+	vlv_dpio_write(dev_priv, pipe, DPIO_PCS_CLOCKBUF8(port), 0x00400888);
 
 	mutex_unlock(&dev_priv->dpio_lock);
 
+	/* init power sequencer on this pipe and port */
+	intel_dp_init_panel_power_sequencer(dev, intel_dp, &power_seq);
+	intel_dp_init_panel_power_sequencer_registers(dev, intel_dp,
+						      &power_seq);
+
 	intel_enable_dp(encoder);
 
 	vlv_wait_port_ready(dev_priv, port);
 }
 
-static void intel_dp_pre_pll_enable(struct intel_encoder *encoder)
+static void vlv_dp_pre_pll_enable(struct intel_encoder *encoder)
 {
 	struct intel_digital_port *dport = enc_to_dig_port(&encoder->base);
 	struct drm_device *dev = encoder->base.dev;
 	struct drm_i915_private *dev_priv = dev->dev_private;
+	struct intel_crtc *intel_crtc =
+		to_intel_crtc(encoder->base.crtc);
 	int port = vlv_dport_to_channel(dport);
-
-	if (!IS_VALLEYVIEW(dev))
-		return;
+	int pipe = intel_crtc->pipe;
 
 	/* Program Tx lane resets to default */
 	mutex_lock(&dev_priv->dpio_lock);
-	vlv_dpio_write(dev_priv, DPIO_PCS_TX(port),
+	vlv_dpio_write(dev_priv, pipe, DPIO_PCS_TX(port),
 			 DPIO_PCS_TX_LANE2_RESET |
 			 DPIO_PCS_TX_LANE1_RESET);
-	vlv_dpio_write(dev_priv, DPIO_PCS_CLK(port),
+	vlv_dpio_write(dev_priv, pipe, DPIO_PCS_CLK(port),
 			 DPIO_PCS_CLK_CRI_RXEB_EIOS_EN |
 			 DPIO_PCS_CLK_CRI_RXDIGFILTSG_EN |
 			 (1<<DPIO_PCS_CLK_DATAWIDTH_SHIFT) |
 				 DPIO_PCS_CLK_SOFT_RESET);
 
 	/* Fix up inter-pair skew failure */
-	vlv_dpio_write(dev_priv, DPIO_PCS_STAGGER1(port), 0x00750f00);
-	vlv_dpio_write(dev_priv, DPIO_TX_CTL(port), 0x00001500);
-	vlv_dpio_write(dev_priv, DPIO_TX_LANE(port), 0x40400000);
+	vlv_dpio_write(dev_priv, pipe, DPIO_PCS_STAGGER1(port), 0x00750f00);
+	vlv_dpio_write(dev_priv, pipe, DPIO_TX_CTL(port), 0x00001500);
+	vlv_dpio_write(dev_priv, pipe, DPIO_TX_LANE(port), 0x40400000);
 	mutex_unlock(&dev_priv->dpio_lock);
 }
 
@@ -1869,7 +1964,7 @@ intel_dp_voltage_max(struct intel_dp *intel_dp)
 	struct drm_device *dev = intel_dp_to_dev(intel_dp);
 	enum port port = dp_to_dig_port(intel_dp)->port;
 
-	if (IS_VALLEYVIEW(dev))
+	if (IS_VALLEYVIEW(dev) || IS_BROADWELL(dev))
 		return DP_TRAIN_VOLTAGE_SWING_1200;
 	else if (IS_GEN7(dev) && port == PORT_A)
 		return DP_TRAIN_VOLTAGE_SWING_800;
@@ -1885,7 +1980,18 @@ intel_dp_pre_emphasis_max(struct intel_dp *intel_dp, uint8_t voltage_swing)
 	struct drm_device *dev = intel_dp_to_dev(intel_dp);
 	enum port port = dp_to_dig_port(intel_dp)->port;
 
-	if (HAS_DDI(dev)) {
+	if (IS_BROADWELL(dev)) {
+		switch (voltage_swing & DP_TRAIN_VOLTAGE_SWING_MASK) {
+		case DP_TRAIN_VOLTAGE_SWING_400:
+		case DP_TRAIN_VOLTAGE_SWING_600:
+			return DP_TRAIN_PRE_EMPHASIS_6;
+		case DP_TRAIN_VOLTAGE_SWING_800:
+			return DP_TRAIN_PRE_EMPHASIS_3_5;
+		case DP_TRAIN_VOLTAGE_SWING_1200:
+		default:
+			return DP_TRAIN_PRE_EMPHASIS_0;
+		}
+	} else if (IS_HASWELL(dev)) {
 		switch (voltage_swing & DP_TRAIN_VOLTAGE_SWING_MASK) {
 		case DP_TRAIN_VOLTAGE_SWING_400:
 			return DP_TRAIN_PRE_EMPHASIS_9_5;
@@ -1939,10 +2045,13 @@ static uint32_t intel_vlv_signal_levels(struct intel_dp *intel_dp)
 	struct drm_device *dev = intel_dp_to_dev(intel_dp);
 	struct drm_i915_private *dev_priv = dev->dev_private;
 	struct intel_digital_port *dport = dp_to_dig_port(intel_dp);
+	struct intel_crtc *intel_crtc =
+		to_intel_crtc(dport->base.base.crtc);
 	unsigned long demph_reg_value, preemph_reg_value,
 		uniqtranscale_reg_value;
 	uint8_t train_set = intel_dp->train_set[0];
 	int port = vlv_dport_to_channel(dport);
+	int pipe = intel_crtc->pipe;
 
 	switch (train_set & DP_TRAIN_PRE_EMPHASIS_MASK) {
 	case DP_TRAIN_PRE_EMPHASIS_0:
@@ -2018,21 +2127,22 @@ static uint32_t intel_vlv_signal_levels(struct intel_dp *intel_dp)
 	}
 
 	mutex_lock(&dev_priv->dpio_lock);
-	vlv_dpio_write(dev_priv, DPIO_TX_OCALINIT(port), 0x00000000);
-	vlv_dpio_write(dev_priv, DPIO_TX_SWING_CTL4(port), demph_reg_value);
-	vlv_dpio_write(dev_priv, DPIO_TX_SWING_CTL2(port),
+	vlv_dpio_write(dev_priv, pipe, DPIO_TX_OCALINIT(port), 0x00000000);
+	vlv_dpio_write(dev_priv, pipe, DPIO_TX_SWING_CTL4(port), demph_reg_value);
+	vlv_dpio_write(dev_priv, pipe, DPIO_TX_SWING_CTL2(port),
 			 uniqtranscale_reg_value);
-	vlv_dpio_write(dev_priv, DPIO_TX_SWING_CTL3(port), 0x0C782040);
-	vlv_dpio_write(dev_priv, DPIO_PCS_STAGGER0(port), 0x00030000);
-	vlv_dpio_write(dev_priv, DPIO_PCS_CTL_OVER1(port), preemph_reg_value);
-	vlv_dpio_write(dev_priv, DPIO_TX_OCALINIT(port), 0x80000000);
+	vlv_dpio_write(dev_priv, pipe, DPIO_TX_SWING_CTL3(port), 0x0C782040);
+	vlv_dpio_write(dev_priv, pipe, DPIO_PCS_STAGGER0(port), 0x00030000);
+	vlv_dpio_write(dev_priv, pipe, DPIO_PCS_CTL_OVER1(port), preemph_reg_value);
+	vlv_dpio_write(dev_priv, pipe, DPIO_TX_OCALINIT(port), 0x80000000);
 	mutex_unlock(&dev_priv->dpio_lock);
 
 	return 0;
 }
 
 static void
-intel_get_adjust_train(struct intel_dp *intel_dp, uint8_t link_status[DP_LINK_STATUS_SIZE])
+intel_get_adjust_train(struct intel_dp *intel_dp,
+		       const uint8_t link_status[DP_LINK_STATUS_SIZE])
 {
 	uint8_t v = 0;
 	uint8_t p = 0;
@@ -2193,6 +2303,41 @@ intel_hsw_signal_levels(uint8_t train_set)
 	}
 }
 
+static uint32_t
+intel_bdw_signal_levels(uint8_t train_set)
+{
+	int signal_levels = train_set & (DP_TRAIN_VOLTAGE_SWING_MASK |
+					 DP_TRAIN_PRE_EMPHASIS_MASK);
+	switch (signal_levels) {
+	case DP_TRAIN_VOLTAGE_SWING_400 | DP_TRAIN_PRE_EMPHASIS_0:
+		return DDI_BUF_EMP_400MV_0DB_BDW;	/* Sel0 */
+	case DP_TRAIN_VOLTAGE_SWING_400 | DP_TRAIN_PRE_EMPHASIS_3_5:
+		return DDI_BUF_EMP_400MV_3_5DB_BDW;	/* Sel1 */
+	case DP_TRAIN_VOLTAGE_SWING_400 | DP_TRAIN_PRE_EMPHASIS_6:
+		return DDI_BUF_EMP_400MV_6DB_BDW;	/* Sel2 */
+
+	case DP_TRAIN_VOLTAGE_SWING_600 | DP_TRAIN_PRE_EMPHASIS_0:
+		return DDI_BUF_EMP_600MV_0DB_BDW;	/* Sel3 */
+	case DP_TRAIN_VOLTAGE_SWING_600 | DP_TRAIN_PRE_EMPHASIS_3_5:
+		return DDI_BUF_EMP_600MV_3_5DB_BDW;	/* Sel4 */
+	case DP_TRAIN_VOLTAGE_SWING_600 | DP_TRAIN_PRE_EMPHASIS_6:
+		return DDI_BUF_EMP_600MV_6DB_BDW;	/* Sel5 */
+
+	case DP_TRAIN_VOLTAGE_SWING_800 | DP_TRAIN_PRE_EMPHASIS_0:
+		return DDI_BUF_EMP_800MV_0DB_BDW;	/* Sel6 */
+	case DP_TRAIN_VOLTAGE_SWING_800 | DP_TRAIN_PRE_EMPHASIS_3_5:
+		return DDI_BUF_EMP_800MV_3_5DB_BDW;	/* Sel7 */
+
+	case DP_TRAIN_VOLTAGE_SWING_1200 | DP_TRAIN_PRE_EMPHASIS_0:
+		return DDI_BUF_EMP_1200MV_0DB_BDW;	/* Sel8 */
+
+	default:
+		DRM_DEBUG_KMS("Unsupported voltage swing/pre-emphasis level:"
+			      "0x%x\n", signal_levels);
+		return DDI_BUF_EMP_400MV_0DB_BDW;	/* Sel0 */
+	}
+}
+
 /* Properly updates "DP" with the correct signal levels. */
 static void
 intel_dp_set_signal_levels(struct intel_dp *intel_dp, uint32_t *DP)
@@ -2203,7 +2348,10 @@ intel_dp_set_signal_levels(struct intel_dp *intel_dp, uint32_t *DP)
 	uint32_t signal_levels, mask;
 	uint8_t train_set = intel_dp->train_set[0];
 
-	if (HAS_DDI(dev)) {
+	if (IS_BROADWELL(dev)) {
+		signal_levels = intel_bdw_signal_levels(train_set);
+		mask = DDI_BUF_EMP_MASK;
+	} else if (IS_HASWELL(dev)) {
 		signal_levels = intel_hsw_signal_levels(train_set);
 		mask = DDI_BUF_EMP_MASK;
 	} else if (IS_VALLEYVIEW(dev)) {
@@ -2227,14 +2375,15 @@ intel_dp_set_signal_levels(struct intel_dp *intel_dp, uint32_t *DP)
 
 static bool
 intel_dp_set_link_train(struct intel_dp *intel_dp,
-			uint32_t dp_reg_value,
+			uint32_t *DP,
 			uint8_t dp_train_pat)
 {
 	struct intel_digital_port *intel_dig_port = dp_to_dig_port(intel_dp);
 	struct drm_device *dev = intel_dig_port->base.base.dev;
 	struct drm_i915_private *dev_priv = dev->dev_private;
 	enum port port = intel_dig_port->port;
-	int ret;
+	uint8_t buf[sizeof(intel_dp->train_set) + 1];
+	int ret, len;
 
 	if (HAS_DDI(dev)) {
 		uint32_t temp = I915_READ(DP_TP_CTL(port));
@@ -2263,62 +2412,93 @@ intel_dp_set_link_train(struct intel_dp *intel_dp,
 		I915_WRITE(DP_TP_CTL(port), temp);
 
 	} else if (HAS_PCH_CPT(dev) && (IS_GEN7(dev) || port != PORT_A)) {
-		dp_reg_value &= ~DP_LINK_TRAIN_MASK_CPT;
+		*DP &= ~DP_LINK_TRAIN_MASK_CPT;
 
 		switch (dp_train_pat & DP_TRAINING_PATTERN_MASK) {
 		case DP_TRAINING_PATTERN_DISABLE:
-			dp_reg_value |= DP_LINK_TRAIN_OFF_CPT;
+			*DP |= DP_LINK_TRAIN_OFF_CPT;
 			break;
 		case DP_TRAINING_PATTERN_1:
-			dp_reg_value |= DP_LINK_TRAIN_PAT_1_CPT;
+			*DP |= DP_LINK_TRAIN_PAT_1_CPT;
 			break;
 		case DP_TRAINING_PATTERN_2:
-			dp_reg_value |= DP_LINK_TRAIN_PAT_2_CPT;
+			*DP |= DP_LINK_TRAIN_PAT_2_CPT;
 			break;
 		case DP_TRAINING_PATTERN_3:
 			DRM_ERROR("DP training pattern 3 not supported\n");
-			dp_reg_value |= DP_LINK_TRAIN_PAT_2_CPT;
+			*DP |= DP_LINK_TRAIN_PAT_2_CPT;
 			break;
 		}
 
 	} else {
-		dp_reg_value &= ~DP_LINK_TRAIN_MASK;
+		*DP &= ~DP_LINK_TRAIN_MASK;
 
 		switch (dp_train_pat & DP_TRAINING_PATTERN_MASK) {
 		case DP_TRAINING_PATTERN_DISABLE:
-			dp_reg_value |= DP_LINK_TRAIN_OFF;
+			*DP |= DP_LINK_TRAIN_OFF;
 			break;
 		case DP_TRAINING_PATTERN_1:
-			dp_reg_value |= DP_LINK_TRAIN_PAT_1;
+			*DP |= DP_LINK_TRAIN_PAT_1;
 			break;
 		case DP_TRAINING_PATTERN_2:
-			dp_reg_value |= DP_LINK_TRAIN_PAT_2;
+			*DP |= DP_LINK_TRAIN_PAT_2;
 			break;
 		case DP_TRAINING_PATTERN_3:
 			DRM_ERROR("DP training pattern 3 not supported\n");
-			dp_reg_value |= DP_LINK_TRAIN_PAT_2;
+			*DP |= DP_LINK_TRAIN_PAT_2;
 			break;
 		}
 	}
 
-	I915_WRITE(intel_dp->output_reg, dp_reg_value);
+	I915_WRITE(intel_dp->output_reg, *DP);
 	POSTING_READ(intel_dp->output_reg);
 
-	intel_dp_aux_native_write_1(intel_dp,
-				    DP_TRAINING_PATTERN_SET,
-				    dp_train_pat);
-
-	if ((dp_train_pat & DP_TRAINING_PATTERN_MASK) !=
+	buf[0] = dp_train_pat;
+	if ((dp_train_pat & DP_TRAINING_PATTERN_MASK) ==
 	    DP_TRAINING_PATTERN_DISABLE) {
-		ret = intel_dp_aux_native_write(intel_dp,
-						DP_TRAINING_LANE0_SET,
-						intel_dp->train_set,
-						intel_dp->lane_count);
-		if (ret != intel_dp->lane_count)
-			return false;
+		/* don't write DP_TRAINING_LANEx_SET on disable */
+		len = 1;
+	} else {
+		/* DP_TRAINING_LANEx_SET follow DP_TRAINING_PATTERN_SET */
+		memcpy(buf + 1, intel_dp->train_set, intel_dp->lane_count);
+		len = intel_dp->lane_count + 1;
 	}
 
-	return true;
+	ret = intel_dp_aux_native_write(intel_dp, DP_TRAINING_PATTERN_SET,
+					buf, len);
+
+	return ret == len;
+}
+
+static bool
+intel_dp_reset_link_train(struct intel_dp *intel_dp, uint32_t *DP,
+			uint8_t dp_train_pat)
+{
+	memset(intel_dp->train_set, 0, sizeof(intel_dp->train_set));
+	intel_dp_set_signal_levels(intel_dp, DP);
+	return intel_dp_set_link_train(intel_dp, DP, dp_train_pat);
+}
+
+static bool
+intel_dp_update_link_train(struct intel_dp *intel_dp, uint32_t *DP,
+			   const uint8_t link_status[DP_LINK_STATUS_SIZE])
+{
+	struct intel_digital_port *intel_dig_port = dp_to_dig_port(intel_dp);
+	struct drm_device *dev = intel_dig_port->base.base.dev;
+	struct drm_i915_private *dev_priv = dev->dev_private;
+	int ret;
+
+	intel_get_adjust_train(intel_dp, link_status);
+	intel_dp_set_signal_levels(intel_dp, DP);
+
+	I915_WRITE(intel_dp->output_reg, *DP);
+	POSTING_READ(intel_dp->output_reg);
+
+	ret = intel_dp_aux_native_write(intel_dp, DP_TRAINING_LANE0_SET,
+					intel_dp->train_set,
+					intel_dp->lane_count);
+
+	return ret == intel_dp->lane_count;
 }
 
 static void intel_dp_set_idle_link_train(struct intel_dp *intel_dp)
@@ -2362,32 +2542,37 @@ intel_dp_start_link_train(struct intel_dp *intel_dp)
 	uint8_t voltage;
 	int voltage_tries, loop_tries;
 	uint32_t DP = intel_dp->DP;
+	uint8_t link_config[2];
 
 	if (HAS_DDI(dev))
 		intel_ddi_prepare_link_retrain(encoder);
 
 	/* Write the link configuration data */
-	intel_dp_aux_native_write(intel_dp, DP_LINK_BW_SET,
-				  intel_dp->link_configuration,
-				  DP_LINK_CONFIGURATION_SIZE);
+	link_config[0] = intel_dp->link_bw;
+	link_config[1] = intel_dp->lane_count;
+	if (drm_dp_enhanced_frame_cap(intel_dp->dpcd))
+		link_config[1] |= DP_LANE_COUNT_ENHANCED_FRAME_EN;
+	intel_dp_aux_native_write(intel_dp, DP_LINK_BW_SET, link_config, 2);
+
+	link_config[0] = 0;
+	link_config[1] = DP_SET_ANSI_8B10B;
+	intel_dp_aux_native_write(intel_dp, DP_DOWNSPREAD_CTRL, link_config, 2);
 
 	DP |= DP_PORT_EN;
 
-	memset(intel_dp->train_set, 0, 4);
+	/* clock recovery */
+	if (!intel_dp_reset_link_train(intel_dp, &DP,
+				       DP_TRAINING_PATTERN_1 |
+				       DP_LINK_SCRAMBLING_DISABLE)) {
+		DRM_ERROR("failed to enable link training\n");
+		return;
+	}
+
 	voltage = 0xff;
 	voltage_tries = 0;
 	loop_tries = 0;
 	for (;;) {
-		/* Use intel_dp->train_set[0] to set the voltage and pre emphasis values */
-		uint8_t	    link_status[DP_LINK_STATUS_SIZE];
-
-		intel_dp_set_signal_levels(intel_dp, &DP);
-
-		/* Set training pattern 1 */
-		if (!intel_dp_set_link_train(intel_dp, DP,
-					     DP_TRAINING_PATTERN_1 |
-					     DP_LINK_SCRAMBLING_DISABLE))
-			break;
+		uint8_t link_status[DP_LINK_STATUS_SIZE];
 
 		drm_dp_link_train_clock_recovery_delay(intel_dp->dpcd);
 		if (!intel_dp_get_link_status(intel_dp, link_status)) {
@@ -2407,10 +2592,12 @@ intel_dp_start_link_train(struct intel_dp *intel_dp)
 		if (i == intel_dp->lane_count) {
 			++loop_tries;
 			if (loop_tries == 5) {
-				DRM_DEBUG_KMS("too many full retries, give up\n");
+				DRM_ERROR("too many full retries, give up\n");
 				break;
 			}
-			memset(intel_dp->train_set, 0, 4);
+			intel_dp_reset_link_train(intel_dp, &DP,
+						  DP_TRAINING_PATTERN_1 |
+						  DP_LINK_SCRAMBLING_DISABLE);
 			voltage_tries = 0;
 			continue;
 		}
@@ -2419,15 +2606,18 @@ intel_dp_start_link_train(struct intel_dp *intel_dp)
 		if ((intel_dp->train_set[0] & DP_TRAIN_VOLTAGE_SWING_MASK) == voltage) {
 			++voltage_tries;
 			if (voltage_tries == 5) {
-				DRM_DEBUG_KMS("too many voltage retries, give up\n");
+				DRM_ERROR("too many voltage retries, give up\n");
 				break;
 			}
 		} else
 			voltage_tries = 0;
 		voltage = intel_dp->train_set[0] & DP_TRAIN_VOLTAGE_SWING_MASK;
 
-		/* Compute new intel_dp->train_set as requested by target */
-		intel_get_adjust_train(intel_dp, link_status);
+		/* Update training set as requested by target */
+		if (!intel_dp_update_link_train(intel_dp, &DP, link_status)) {
+			DRM_ERROR("failed to update link training\n");
+			break;
+		}
 	}
 
 	intel_dp->DP = DP;
@@ -2441,11 +2631,18 @@ intel_dp_complete_link_train(struct intel_dp *intel_dp)
 	uint32_t DP = intel_dp->DP;
 
 	/* channel equalization */
+	if (!intel_dp_set_link_train(intel_dp, &DP,
+				     DP_TRAINING_PATTERN_2 |
+				     DP_LINK_SCRAMBLING_DISABLE)) {
+		DRM_ERROR("failed to start channel equalization\n");
+		return;
+	}
+
 	tries = 0;
 	cr_tries = 0;
 	channel_eq = false;
 	for (;;) {
-		uint8_t	    link_status[DP_LINK_STATUS_SIZE];
+		uint8_t link_status[DP_LINK_STATUS_SIZE];
 
 		if (cr_tries > 5) {
 			DRM_ERROR("failed to train DP, aborting\n");
@@ -2453,21 +2650,18 @@ intel_dp_complete_link_train(struct intel_dp *intel_dp)
 			break;
 		}
 
-		intel_dp_set_signal_levels(intel_dp, &DP);
-
-		/* channel eq pattern */
-		if (!intel_dp_set_link_train(intel_dp, DP,
-					     DP_TRAINING_PATTERN_2 |
-					     DP_LINK_SCRAMBLING_DISABLE))
-			break;
-
 		drm_dp_link_train_channel_eq_delay(intel_dp->dpcd);
-		if (!intel_dp_get_link_status(intel_dp, link_status))
+		if (!intel_dp_get_link_status(intel_dp, link_status)) {
+			DRM_ERROR("failed to get link status\n");
 			break;
+		}
 
 		/* Make sure clock is still ok */
 		if (!drm_dp_clock_recovery_ok(link_status, intel_dp->lane_count)) {
 			intel_dp_start_link_train(intel_dp);
+			intel_dp_set_link_train(intel_dp, &DP,
+						DP_TRAINING_PATTERN_2 |
+						DP_LINK_SCRAMBLING_DISABLE);
 			cr_tries++;
 			continue;
 		}
@@ -2481,13 +2675,19 @@ intel_dp_complete_link_train(struct intel_dp *intel_dp)
 		if (tries > 5) {
 			intel_dp_link_down(intel_dp);
 			intel_dp_start_link_train(intel_dp);
+			intel_dp_set_link_train(intel_dp, &DP,
+						DP_TRAINING_PATTERN_2 |
+						DP_LINK_SCRAMBLING_DISABLE);
 			tries = 0;
 			cr_tries++;
 			continue;
 		}
 
-		/* Compute new intel_dp->train_set as requested by target */
-		intel_get_adjust_train(intel_dp, link_status);
+		/* Update training set as requested by target */
+		if (!intel_dp_update_link_train(intel_dp, &DP, link_status)) {
+			DRM_ERROR("failed to update link training\n");
+			break;
+		}
 		++tries;
 	}
 
@@ -2502,7 +2702,7 @@ intel_dp_complete_link_train(struct intel_dp *intel_dp)
 
 void intel_dp_stop_link_train(struct intel_dp *intel_dp)
 {
-	intel_dp_set_link_train(intel_dp, intel_dp->DP,
+	intel_dp_set_link_train(intel_dp, &intel_dp->DP,
 				DP_TRAINING_PATTERN_DISABLE);
 }
 
@@ -2589,6 +2789,10 @@ intel_dp_link_down(struct intel_dp *intel_dp)
 static bool
 intel_dp_get_dpcd(struct intel_dp *intel_dp)
 {
+	struct intel_digital_port *dig_port = dp_to_dig_port(intel_dp);
+	struct drm_device *dev = dig_port->base.base.dev;
+	struct drm_i915_private *dev_priv = dev->dev_private;
+
 	char dpcd_hex_dump[sizeof(intel_dp->dpcd) * 3];
 
 	if (intel_dp_aux_native_read_retry(intel_dp, 0x000, intel_dp->dpcd,
@@ -2604,11 +2808,16 @@ intel_dp_get_dpcd(struct intel_dp *intel_dp)
 
 	/* Check if the panel supports PSR */
 	memset(intel_dp->psr_dpcd, 0, sizeof(intel_dp->psr_dpcd));
-	intel_dp_aux_native_read_retry(intel_dp, DP_PSR_SUPPORT,
-				       intel_dp->psr_dpcd,
-				       sizeof(intel_dp->psr_dpcd));
-	if (is_edp_psr(intel_dp))
-		DRM_DEBUG_KMS("Detected EDP PSR Panel.\n");
+	if (is_edp(intel_dp)) {
+		intel_dp_aux_native_read_retry(intel_dp, DP_PSR_SUPPORT,
+					       intel_dp->psr_dpcd,
+					       sizeof(intel_dp->psr_dpcd));
+		if (intel_dp->psr_dpcd[0] & DP_PSR_IS_SUPPORTED) {
+			dev_priv->psr.sink_support = true;
+			DRM_DEBUG_KMS("Detected EDP PSR Panel.\n");
+		}
+	}
+
 	if (!(intel_dp->dpcd[DP_DOWNSTREAMPORT_PRESENT] &
 	      DP_DWN_STRM_PORT_PRESENT))
 		return true; /* native DP sink */
@@ -2728,7 +2937,6 @@ static enum drm_connector_status
 intel_dp_detect_dpcd(struct intel_dp *intel_dp)
 {
 	uint8_t *dpcd = intel_dp->dpcd;
-	bool hpd;
 	uint8_t type;
 
 	if (!intel_dp_get_dpcd(intel_dp))
@@ -2739,8 +2947,8 @@ intel_dp_detect_dpcd(struct intel_dp *intel_dp)
 		return connector_status_connected;
 
 	/* If we're HPD-aware, SINK_COUNT changes dynamically */
-	hpd = !!(intel_dp->downstream_ports[0] & DP_DS_PORT_HPD);
-	if (hpd) {
+	if (intel_dp->dpcd[DP_DPCD_REV] >= 0x11 &&
+	    intel_dp->downstream_ports[0] & DP_DS_PORT_HPD) {
 		uint8_t reg;
 		if (!intel_dp_aux_native_read_retry(intel_dp, DP_SINK_COUNT,
 						    &reg, 1))
@@ -2754,9 +2962,18 @@ intel_dp_detect_dpcd(struct intel_dp *intel_dp)
 		return connector_status_connected;
 
 	/* Well we tried, say unknown for unreliable port types */
-	type = intel_dp->downstream_ports[0] & DP_DS_PORT_TYPE_MASK;
-	if (type == DP_DS_PORT_TYPE_VGA || type == DP_DS_PORT_TYPE_NON_EDID)
-		return connector_status_unknown;
+	if (intel_dp->dpcd[DP_DPCD_REV] >= 0x11) {
+		type = intel_dp->downstream_ports[0] & DP_DS_PORT_TYPE_MASK;
+		if (type == DP_DS_PORT_TYPE_VGA ||
+		    type == DP_DS_PORT_TYPE_NON_EDID)
+			return connector_status_unknown;
+	} else {
+		type = intel_dp->dpcd[DP_DOWNSTREAMPORT_PRESENT] &
+			DP_DWN_STRM_PORT_TYPE_MASK;
+		if (type == DP_DWN_STRM_PORT_TYPE_ANALOG ||
+		    type == DP_DWN_STRM_PORT_TYPE_OTHER)
+			return connector_status_unknown;
+	}
 
 	/* Anything else is out of spec, warn and ignore */
 	DRM_DEBUG_KMS("Broken DP branch device, ignoring\n");
@@ -2830,19 +3047,11 @@ intel_dp_get_edid(struct drm_connector *connector, struct i2c_adapter *adapter)
 
 	/* use cached edid if we have one */
 	if (intel_connector->edid) {
-		struct edid *edid;
-		int size;
-
 		/* invalid edid */
 		if (IS_ERR(intel_connector->edid))
 			return NULL;
 
-		size = (intel_connector->edid->extensions + 1) * EDID_LENGTH;
-		edid = kmemdup(intel_connector->edid, size, GFP_KERNEL);
-		if (!edid)
-			return NULL;
-
-		return edid;
+		return drm_edid_duplicate(intel_connector->edid);
 	}
 
 	return drm_get_edid(connector, adapter);
@@ -3050,7 +3259,6 @@ intel_dp_connector_destroy(struct drm_connector *connector)
 	if (connector->connector_type == DRM_MODE_CONNECTOR_eDP)
 		intel_panel_fini(&intel_connector->panel);
 
-	drm_sysfs_connector_remove(connector);
 	drm_connector_cleanup(connector);
 	kfree(connector);
 }
@@ -3121,7 +3329,7 @@ intel_trans_dp_port_sel(struct drm_crtc *crtc)
 bool intel_dpd_is_edp(struct drm_device *dev)
 {
 	struct drm_i915_private *dev_priv = dev->dev_private;
-	struct child_device_config *p_child;
+	union child_device_config *p_child;
 	int i;
 
 	if (!dev_priv->vbt.child_dev_num)
@@ -3130,8 +3338,9 @@ bool intel_dpd_is_edp(struct drm_device *dev)
 	for (i = 0; i < dev_priv->vbt.child_dev_num; i++) {
 		p_child = dev_priv->vbt.child_dev + i;
 
-		if (p_child->dvo_port == PORT_IDPD &&
-		    p_child->device_type == DEVICE_TYPE_eDP)
+		if (p_child->common.dvo_port == PORT_IDPD &&
+		    (p_child->common.device_type & DEVICE_TYPE_eDP_BITS) ==
+		    (DEVICE_TYPE_eDP & DEVICE_TYPE_eDP_BITS))
 			return true;
 	}
 	return false;
@@ -3164,24 +3373,26 @@ intel_dp_init_panel_power_sequencer(struct drm_device *dev,
 	struct drm_i915_private *dev_priv = dev->dev_private;
 	struct edp_power_seq cur, vbt, spec, final;
 	u32 pp_on, pp_off, pp_div, pp;
-	int pp_control_reg, pp_on_reg, pp_off_reg, pp_div_reg;
+	int pp_ctrl_reg, pp_on_reg, pp_off_reg, pp_div_reg;
 
 	if (HAS_PCH_SPLIT(dev)) {
-		pp_control_reg = PCH_PP_CONTROL;
+		pp_ctrl_reg = PCH_PP_CONTROL;
 		pp_on_reg = PCH_PP_ON_DELAYS;
 		pp_off_reg = PCH_PP_OFF_DELAYS;
 		pp_div_reg = PCH_PP_DIVISOR;
 	} else {
-		pp_control_reg = PIPEA_PP_CONTROL;
-		pp_on_reg = PIPEA_PP_ON_DELAYS;
-		pp_off_reg = PIPEA_PP_OFF_DELAYS;
-		pp_div_reg = PIPEA_PP_DIVISOR;
+		enum pipe pipe = vlv_power_sequencer_pipe(intel_dp);
+
+		pp_ctrl_reg = VLV_PIPE_PP_CONTROL(pipe);
+		pp_on_reg = VLV_PIPE_PP_ON_DELAYS(pipe);
+		pp_off_reg = VLV_PIPE_PP_OFF_DELAYS(pipe);
+		pp_div_reg = VLV_PIPE_PP_DIVISOR(pipe);
 	}
 
 	/* Workaround: Need to write PP_CONTROL with the unlock key as
 	 * the very first thing. */
 	pp = ironlake_get_pp_control(intel_dp);
-	I915_WRITE(pp_control_reg, pp);
+	I915_WRITE(pp_ctrl_reg, pp);
 
 	pp_on = I915_READ(pp_on_reg);
 	pp_off = I915_READ(pp_off_reg);
@@ -3269,9 +3480,11 @@ intel_dp_init_panel_power_sequencer_registers(struct drm_device *dev,
 		pp_off_reg = PCH_PP_OFF_DELAYS;
 		pp_div_reg = PCH_PP_DIVISOR;
 	} else {
-		pp_on_reg = PIPEA_PP_ON_DELAYS;
-		pp_off_reg = PIPEA_PP_OFF_DELAYS;
-		pp_div_reg = PIPEA_PP_DIVISOR;
+		enum pipe pipe = vlv_power_sequencer_pipe(intel_dp);
+
+		pp_on_reg = VLV_PIPE_PP_ON_DELAYS(pipe);
+		pp_off_reg = VLV_PIPE_PP_OFF_DELAYS(pipe);
+		pp_div_reg = VLV_PIPE_PP_DIVISOR(pipe);
 	}
 
 	/* And finally store the new values in the power sequencer. */
@@ -3288,12 +3501,15 @@ intel_dp_init_panel_power_sequencer_registers(struct drm_device *dev,
 	/* Haswell doesn't have any port selection bits for the panel
 	 * power sequencer any more. */
 	if (IS_VALLEYVIEW(dev)) {
-		port_sel = I915_READ(pp_on_reg) & 0xc0000000;
+		if (dp_to_dig_port(intel_dp)->port == PORT_B)
+			port_sel = PANEL_PORT_SELECT_DPB_VLV;
+		else
+			port_sel = PANEL_PORT_SELECT_DPC_VLV;
 	} else if (HAS_PCH_IBX(dev) || HAS_PCH_CPT(dev)) {
 		if (dp_to_dig_port(intel_dp)->port == PORT_A)
-			port_sel = PANEL_POWER_PORT_DP_A;
+			port_sel = PANEL_PORT_SELECT_DPA;
 		else
-			port_sel = PANEL_POWER_PORT_DP_D;
+			port_sel = PANEL_PORT_SELECT_DPD;
 	}
 
 	pp_on |= port_sel;
@@ -3346,7 +3562,6 @@ static bool intel_edp_init_connector(struct intel_dp *intel_dp,
 	intel_dp_init_panel_power_sequencer_registers(dev, intel_dp,
 						      &power_seq);
 
-	ironlake_edp_panel_vdd_on(intel_dp);
 	edid = drm_get_edid(connector, &intel_dp->adapter);
 	if (edid) {
 		if (drm_add_edid_modes(connector, edid)) {
@@ -3378,8 +3593,6 @@ static bool intel_edp_init_connector(struct intel_dp *intel_dp,
 			fixed_mode->type |= DRM_MODE_TYPE_PREFERRED;
 	}
 
-	ironlake_edp_panel_vdd_off(intel_dp, false);
-
 	intel_panel_init(&intel_connector->panel, fixed_mode);
 	intel_panel_setup_backlight(connector);
 
@@ -3536,11 +3749,11 @@ intel_dp_init(struct drm_device *dev, int output_reg, enum port port)
 	struct drm_encoder *encoder;
 	struct intel_connector *intel_connector;
 
-	intel_dig_port = kzalloc(sizeof(struct intel_digital_port), GFP_KERNEL);
+	intel_dig_port = kzalloc(sizeof(*intel_dig_port), GFP_KERNEL);
 	if (!intel_dig_port)
 		return;
 
-	intel_connector = kzalloc(sizeof(struct intel_connector), GFP_KERNEL);
+	intel_connector = kzalloc(sizeof(*intel_connector), GFP_KERNEL);
 	if (!intel_connector) {
 		kfree(intel_dig_port);
 		return;
@@ -3559,12 +3772,12 @@ intel_dp_init(struct drm_device *dev, int output_reg, enum port port)
 	intel_encoder->get_hw_state = intel_dp_get_hw_state;
 	intel_encoder->get_config = intel_dp_get_config;
 	if (IS_VALLEYVIEW(dev)) {
-		intel_encoder->pre_pll_enable = intel_dp_pre_pll_enable;
+		intel_encoder->pre_pll_enable = vlv_dp_pre_pll_enable;
 		intel_encoder->pre_enable = vlv_pre_enable_dp;
 		intel_encoder->enable = vlv_enable_dp;
 	} else {
-		intel_encoder->pre_enable = intel_pre_enable_dp;
-		intel_encoder->enable = intel_enable_dp;
+		intel_encoder->pre_enable = g4x_pre_enable_dp;
+		intel_encoder->enable = g4x_enable_dp;
 	}
 
 	intel_dig_port->port = port;
diff --git a/drivers/gpu/drm/i915/intel_drv.h b/drivers/gpu/drm/i915/intel_drv.h
index 7f2b384ac939..1e49aa8f5377 100644
--- a/drivers/gpu/drm/i915/intel_drv.h
+++ b/drivers/gpu/drm/i915/intel_drv.h
@@ -77,7 +77,6 @@
 /* the i915, i945 have a single sDVO i2c bus - which is different */
 #define MAX_OUTPUTS 6
 /* maximum connectors per crtcs in the mode set */
-#define INTELFB_CONN_LIMIT 4
 
 #define INTEL_I2C_BUS_DVO 1
 #define INTEL_I2C_BUS_SDVO 2
@@ -93,13 +92,17 @@
 #define INTEL_OUTPUT_HDMI 6
 #define INTEL_OUTPUT_DISPLAYPORT 7
 #define INTEL_OUTPUT_EDP 8
-#define INTEL_OUTPUT_UNKNOWN 9
+#define INTEL_OUTPUT_DSI 9
+#define INTEL_OUTPUT_UNKNOWN 10
 
 #define INTEL_DVO_CHIP_NONE 0
 #define INTEL_DVO_CHIP_LVDS 1
 #define INTEL_DVO_CHIP_TMDS 2
 #define INTEL_DVO_CHIP_TVOUT 4
 
+#define INTEL_DSI_COMMAND_MODE	0
+#define INTEL_DSI_VIDEO_MODE	1
+
 struct intel_framebuffer {
 	struct drm_framebuffer base;
 	struct drm_i915_gem_object *obj;
@@ -207,8 +210,21 @@ struct intel_crtc_config {
 #define PIPE_CONFIG_QUIRK_MODE_SYNC_FLAGS (1<<0) /* unreliable sync mode.flags */
 	unsigned long quirks;
 
+	/* User requested mode, only valid as a starting point to
+	 * compute adjusted_mode, except in the case of (S)DVO where
+	 * it's also for the output timings of the (S)DVO chip.
+	 * adjusted_mode will then correspond to the S(DVO) chip's
+	 * preferred input timings. */
 	struct drm_display_mode requested_mode;
+	/* Actual pipe timings ie. what we program into the pipe timing
+	 * registers. adjusted_mode.crtc_clock is the pipe pixel clock. */
 	struct drm_display_mode adjusted_mode;
+
+	/* Pipe source size (ie. panel fitter input size)
+	 * All planes will be positioned inside this space,
+	 * and get clipped at the edges. */
+	int pipe_src_w, pipe_src_h;
+
 	/* Whether to set up the PCH/FDI. Note that we never allow sharing
 	 * between pch encoders and cpu encoders. */
 	bool has_pch_encoder;
@@ -262,7 +278,8 @@ struct intel_crtc_config {
 
 	/*
 	 * Frequence the dpll for the port should run at. Differs from the
-	 * adjusted dotclock e.g. for DP or 12bpc hdmi mode.
+	 * adjusted dotclock e.g. for DP or 12bpc hdmi mode. This is also
+	 * already multiplied by pixel_multiplier.
 	 */
 	int port_clock;
 
@@ -288,6 +305,14 @@ struct intel_crtc_config {
 	struct intel_link_m_n fdi_m_n;
 
 	bool ips_enabled;
+
+	bool double_wide;
+};
+
+struct intel_pipe_wm {
+	struct intel_wm_level wm[5];
+	uint32_t linetime;
+	bool fbc_wm_enabled;
 };
 
 struct intel_crtc {
@@ -301,8 +326,9 @@ struct intel_crtc {
 	 * some outputs connected to this crtc.
 	 */
 	bool active;
+	unsigned long enabled_power_domains;
 	bool eld_vld;
-	bool primary_disabled; /* is the crtc obscured by a plane? */
+	bool primary_enabled; /* is the primary plane (partially) visible? */
 	bool lowfreq_avail;
 	struct intel_overlay *overlay;
 	struct intel_unpin_work *unpin_work;
@@ -330,6 +356,12 @@ struct intel_crtc {
 	/* Access to these should be protected by dev_priv->irq_lock. */
 	bool cpu_fifo_underrun_disabled;
 	bool pch_fifo_underrun_disabled;
+
+	/* per-pipe watermark state */
+	struct {
+		/* watermarks currently being used  */
+		struct intel_pipe_wm active;
+	} wm;
 };
 
 struct intel_plane_wm_parameters {
@@ -417,13 +449,11 @@ struct intel_hdmi {
 };
 
 #define DP_MAX_DOWNSTREAM_PORTS		0x10
-#define DP_LINK_CONFIGURATION_SIZE	9
 
 struct intel_dp {
 	uint32_t output_reg;
 	uint32_t aux_ch_ctl_reg;
 	uint32_t DP;
-	uint8_t  link_configuration[DP_LINK_CONFIGURATION_SIZE];
 	bool has_audio;
 	enum hdmi_force_audio force_audio;
 	uint32_t color_range;
@@ -495,80 +525,6 @@ struct intel_unpin_work {
 	bool enable_stall_check;
 };
 
-int intel_pch_rawclk(struct drm_device *dev);
-
-int intel_connector_update_modes(struct drm_connector *connector,
-				struct edid *edid);
-int intel_ddc_get_modes(struct drm_connector *c, struct i2c_adapter *adapter);
-
-extern void intel_attach_force_audio_property(struct drm_connector *connector);
-extern void intel_attach_broadcast_rgb_property(struct drm_connector *connector);
-
-extern bool intel_pipe_has_type(struct drm_crtc *crtc, int type);
-extern void intel_crt_init(struct drm_device *dev);
-extern void intel_hdmi_init(struct drm_device *dev,
-			    int hdmi_reg, enum port port);
-extern void intel_hdmi_init_connector(struct intel_digital_port *intel_dig_port,
-				      struct intel_connector *intel_connector);
-extern struct intel_hdmi *enc_to_intel_hdmi(struct drm_encoder *encoder);
-extern bool intel_hdmi_compute_config(struct intel_encoder *encoder,
-				      struct intel_crtc_config *pipe_config);
-extern bool intel_sdvo_init(struct drm_device *dev, uint32_t sdvo_reg,
-			    bool is_sdvob);
-extern void intel_dvo_init(struct drm_device *dev);
-extern void intel_tv_init(struct drm_device *dev);
-extern void intel_mark_busy(struct drm_device *dev);
-extern void intel_mark_fb_busy(struct drm_i915_gem_object *obj,
-			       struct intel_ring_buffer *ring);
-extern void intel_mark_idle(struct drm_device *dev);
-extern void intel_lvds_init(struct drm_device *dev);
-extern bool intel_is_dual_link_lvds(struct drm_device *dev);
-extern void intel_dp_init(struct drm_device *dev, int output_reg,
-			  enum port port);
-extern bool intel_dp_init_connector(struct intel_digital_port *intel_dig_port,
-				    struct intel_connector *intel_connector);
-extern void intel_dp_init_link_config(struct intel_dp *intel_dp);
-extern void intel_dp_start_link_train(struct intel_dp *intel_dp);
-extern void intel_dp_complete_link_train(struct intel_dp *intel_dp);
-extern void intel_dp_stop_link_train(struct intel_dp *intel_dp);
-extern void intel_dp_sink_dpms(struct intel_dp *intel_dp, int mode);
-extern void intel_dp_encoder_destroy(struct drm_encoder *encoder);
-extern void intel_dp_check_link_status(struct intel_dp *intel_dp);
-extern bool intel_dp_compute_config(struct intel_encoder *encoder,
-				    struct intel_crtc_config *pipe_config);
-extern bool intel_dpd_is_edp(struct drm_device *dev);
-extern void ironlake_edp_backlight_on(struct intel_dp *intel_dp);
-extern void ironlake_edp_backlight_off(struct intel_dp *intel_dp);
-extern void ironlake_edp_panel_on(struct intel_dp *intel_dp);
-extern void ironlake_edp_panel_off(struct intel_dp *intel_dp);
-extern void ironlake_edp_panel_vdd_on(struct intel_dp *intel_dp);
-extern void ironlake_edp_panel_vdd_off(struct intel_dp *intel_dp, bool sync);
-extern int intel_plane_init(struct drm_device *dev, enum pipe pipe, int plane);
-extern void intel_flush_display_plane(struct drm_i915_private *dev_priv,
-				      enum plane plane);
-
-/* intel_panel.c */
-extern int intel_panel_init(struct intel_panel *panel,
-			    struct drm_display_mode *fixed_mode);
-extern void intel_panel_fini(struct intel_panel *panel);
-
-extern void intel_fixed_panel_mode(const struct drm_display_mode *fixed_mode,
-				   struct drm_display_mode *adjusted_mode);
-extern void intel_pch_panel_fitting(struct intel_crtc *crtc,
-				    struct intel_crtc_config *pipe_config,
-				    int fitting_mode);
-extern void intel_gmch_panel_fitting(struct intel_crtc *crtc,
-				     struct intel_crtc_config *pipe_config,
-				     int fitting_mode);
-extern void intel_panel_set_backlight(struct drm_device *dev,
-				      u32 level, u32 max);
-extern int intel_panel_setup_backlight(struct drm_connector *connector);
-extern void intel_panel_enable_backlight(struct drm_device *dev,
-					 enum pipe pipe);
-extern void intel_panel_disable_backlight(struct drm_device *dev);
-extern void intel_panel_destroy_backlight(struct drm_device *dev);
-extern enum drm_connector_status intel_panel_detect(struct drm_device *dev);
-
 struct intel_set_config {
 	struct drm_encoder **save_connector_encoders;
 	struct drm_crtc **save_encoder_crtcs;
@@ -577,18 +533,14 @@ struct intel_set_config {
 	bool mode_changed;
 };
 
-extern void intel_crtc_restore_mode(struct drm_crtc *crtc);
-extern void intel_crtc_load_lut(struct drm_crtc *crtc);
-extern void intel_crtc_update_dpms(struct drm_crtc *crtc);
-extern void intel_encoder_destroy(struct drm_encoder *encoder);
-extern void intel_connector_dpms(struct drm_connector *, int mode);
-extern bool intel_connector_get_hw_state(struct intel_connector *connector);
-extern void intel_modeset_check_state(struct drm_device *dev);
-extern void intel_plane_restore(struct drm_plane *plane);
-extern void intel_plane_disable(struct drm_plane *plane);
-
+struct intel_load_detect_pipe {
+	struct drm_framebuffer *release_fb;
+	bool load_detect_temp;
+	int dpms_mode;
+};
 
-static inline struct intel_encoder *intel_attached_encoder(struct drm_connector *connector)
+static inline struct intel_encoder *
+intel_attached_encoder(struct drm_connector *connector)
 {
 	return to_intel_connector(connector)->encoder;
 }
@@ -616,73 +568,95 @@ hdmi_to_dig_port(struct intel_hdmi *intel_hdmi)
 	return container_of(intel_hdmi, struct intel_digital_port, hdmi);
 }
 
+
+/* i915_irq.c */
+bool intel_set_cpu_fifo_underrun_reporting(struct drm_device *dev,
+					   enum pipe pipe, bool enable);
+bool intel_set_pch_fifo_underrun_reporting(struct drm_device *dev,
+					   enum transcoder pch_transcoder,
+					   bool enable);
+void ilk_enable_gt_irq(struct drm_i915_private *dev_priv, uint32_t mask);
+void ilk_disable_gt_irq(struct drm_i915_private *dev_priv, uint32_t mask);
+void snb_enable_pm_irq(struct drm_i915_private *dev_priv, uint32_t mask);
+void snb_disable_pm_irq(struct drm_i915_private *dev_priv, uint32_t mask);
+void hsw_pc8_disable_interrupts(struct drm_device *dev);
+void hsw_pc8_restore_interrupts(struct drm_device *dev);
+
+
+/* intel_crt.c */
+void intel_crt_init(struct drm_device *dev);
+
+
+/* intel_ddi.c */
+void intel_prepare_ddi(struct drm_device *dev);
+void hsw_fdi_link_train(struct drm_crtc *crtc);
+void intel_ddi_init(struct drm_device *dev, enum port port);
+enum port intel_ddi_get_encoder_port(struct intel_encoder *intel_encoder);
+bool intel_ddi_get_hw_state(struct intel_encoder *encoder, enum pipe *pipe);
+int intel_ddi_get_cdclk_freq(struct drm_i915_private *dev_priv);
+void intel_ddi_pll_init(struct drm_device *dev);
+void intel_ddi_enable_transcoder_func(struct drm_crtc *crtc);
+void intel_ddi_disable_transcoder_func(struct drm_i915_private *dev_priv,
+				       enum transcoder cpu_transcoder);
+void intel_ddi_enable_pipe_clock(struct intel_crtc *intel_crtc);
+void intel_ddi_disable_pipe_clock(struct intel_crtc *intel_crtc);
+void intel_ddi_setup_hw_pll_state(struct drm_device *dev);
+bool intel_ddi_pll_mode_set(struct drm_crtc *crtc);
+void intel_ddi_put_crtc_pll(struct drm_crtc *crtc);
+void intel_ddi_set_pipe_settings(struct drm_crtc *crtc);
+void intel_ddi_prepare_link_retrain(struct drm_encoder *encoder);
+bool intel_ddi_connector_get_hw_state(struct intel_connector *intel_connector);
+void intel_ddi_fdi_disable(struct drm_crtc *crtc);
+void intel_ddi_get_config(struct intel_encoder *encoder,
+			  struct intel_crtc_config *pipe_config);
+
+
+/* intel_display.c */
+int intel_pch_rawclk(struct drm_device *dev);
+void intel_mark_busy(struct drm_device *dev);
+void intel_mark_fb_busy(struct drm_i915_gem_object *obj,
+			struct intel_ring_buffer *ring);
+void intel_mark_idle(struct drm_device *dev);
+void intel_crtc_restore_mode(struct drm_crtc *crtc);
+void intel_crtc_update_dpms(struct drm_crtc *crtc);
+void intel_encoder_destroy(struct drm_encoder *encoder);
+void intel_connector_dpms(struct drm_connector *, int mode);
+bool intel_connector_get_hw_state(struct intel_connector *connector);
+void intel_modeset_check_state(struct drm_device *dev);
 bool ibx_digital_port_connected(struct drm_i915_private *dev_priv,
 				struct intel_digital_port *port);
-
-extern void intel_connector_attach_encoder(struct intel_connector *connector,
-					   struct intel_encoder *encoder);
-extern struct drm_encoder *intel_best_encoder(struct drm_connector *connector);
-
-extern struct drm_display_mode *intel_crtc_mode_get(struct drm_device *dev,
-						    struct drm_crtc *crtc);
+void intel_connector_attach_encoder(struct intel_connector *connector,
+				    struct intel_encoder *encoder);
+struct drm_encoder *intel_best_encoder(struct drm_connector *connector);
+struct drm_display_mode *intel_crtc_mode_get(struct drm_device *dev,
+					     struct drm_crtc *crtc);
+enum pipe intel_get_pipe_from_connector(struct intel_connector *connector);
 int intel_get_pipe_from_crtc_id(struct drm_device *dev, void *data,
 				struct drm_file *file_priv);
-extern enum transcoder
-intel_pipe_to_cpu_transcoder(struct drm_i915_private *dev_priv,
-			     enum pipe pipe);
-extern void intel_wait_for_vblank(struct drm_device *dev, int pipe);
-extern void intel_wait_for_pipe_off(struct drm_device *dev, int pipe);
-extern int ironlake_get_lanes_required(int target_clock, int link_bw, int bpp);
-extern void vlv_wait_port_ready(struct drm_i915_private *dev_priv, int port);
-
-struct intel_load_detect_pipe {
-	struct drm_framebuffer *release_fb;
-	bool load_detect_temp;
-	int dpms_mode;
-};
-extern bool intel_get_load_detect_pipe(struct drm_connector *connector,
-				       struct drm_display_mode *mode,
-				       struct intel_load_detect_pipe *old);
-extern void intel_release_load_detect_pipe(struct drm_connector *connector,
-					   struct intel_load_detect_pipe *old);
-
-extern void intel_crtc_fb_gamma_set(struct drm_crtc *crtc, u16 red, u16 green,
-				    u16 blue, int regno);
-extern void intel_crtc_fb_gamma_get(struct drm_crtc *crtc, u16 *red, u16 *green,
-				    u16 *blue, int regno);
-
-extern int intel_pin_and_fence_fb_obj(struct drm_device *dev,
-				      struct drm_i915_gem_object *obj,
-				      struct intel_ring_buffer *pipelined);
-extern void intel_unpin_fb_obj(struct drm_i915_gem_object *obj);
-
-extern int intel_framebuffer_init(struct drm_device *dev,
-				  struct intel_framebuffer *ifb,
-				  struct drm_mode_fb_cmd2 *mode_cmd,
-				  struct drm_i915_gem_object *obj);
-extern void intel_framebuffer_fini(struct intel_framebuffer *fb);
-extern int intel_fbdev_init(struct drm_device *dev);
-extern void intel_fbdev_initial_config(struct drm_device *dev);
-extern void intel_fbdev_fini(struct drm_device *dev);
-extern void intel_fbdev_set_suspend(struct drm_device *dev, int state);
-extern void intel_prepare_page_flip(struct drm_device *dev, int plane);
-extern void intel_finish_page_flip(struct drm_device *dev, int pipe);
-extern void intel_finish_page_flip_plane(struct drm_device *dev, int plane);
-
-extern void intel_setup_overlay(struct drm_device *dev);
-extern void intel_cleanup_overlay(struct drm_device *dev);
-extern int intel_overlay_switch_off(struct intel_overlay *overlay);
-extern int intel_overlay_put_image(struct drm_device *dev, void *data,
-				   struct drm_file *file_priv);
-extern int intel_overlay_attrs(struct drm_device *dev, void *data,
-			       struct drm_file *file_priv);
-
-extern void intel_fb_output_poll_changed(struct drm_device *dev);
-extern void intel_fb_restore_mode(struct drm_device *dev);
-
-struct intel_shared_dpll *
-intel_crtc_to_shared_dpll(struct intel_crtc *crtc);
-
+enum transcoder intel_pipe_to_cpu_transcoder(struct drm_i915_private *dev_priv,
+					     enum pipe pipe);
+void intel_wait_for_vblank(struct drm_device *dev, int pipe);
+void intel_wait_for_pipe_off(struct drm_device *dev, int pipe);
+int ironlake_get_lanes_required(int target_clock, int link_bw, int bpp);
+void vlv_wait_port_ready(struct drm_i915_private *dev_priv, int port);
+bool intel_get_load_detect_pipe(struct drm_connector *connector,
+				struct drm_display_mode *mode,
+				struct intel_load_detect_pipe *old);
+void intel_release_load_detect_pipe(struct drm_connector *connector,
+				    struct intel_load_detect_pipe *old);
+int intel_pin_and_fence_fb_obj(struct drm_device *dev,
+			       struct drm_i915_gem_object *obj,
+			       struct intel_ring_buffer *pipelined);
+void intel_unpin_fb_obj(struct drm_i915_gem_object *obj);
+int intel_framebuffer_init(struct drm_device *dev,
+			   struct intel_framebuffer *ifb,
+			   struct drm_mode_fb_cmd2 *mode_cmd,
+			   struct drm_i915_gem_object *obj);
+void intel_framebuffer_fini(struct intel_framebuffer *fb);
+void intel_prepare_page_flip(struct drm_device *dev, int plane);
+void intel_finish_page_flip(struct drm_device *dev, int pipe);
+void intel_finish_page_flip_plane(struct drm_device *dev, int plane);
+struct intel_shared_dpll *intel_crtc_to_shared_dpll(struct intel_crtc *crtc);
 void assert_shared_dpll(struct drm_i915_private *dev_priv,
 			struct intel_shared_dpll *pll,
 			bool state);
@@ -696,104 +670,199 @@ void assert_fdi_rx_pll(struct drm_i915_private *dev_priv,
 		       enum pipe pipe, bool state);
 #define assert_fdi_rx_pll_enabled(d, p) assert_fdi_rx_pll(d, p, true)
 #define assert_fdi_rx_pll_disabled(d, p) assert_fdi_rx_pll(d, p, false)
-extern void assert_pipe(struct drm_i915_private *dev_priv, enum pipe pipe,
-			bool state);
+void assert_pipe(struct drm_i915_private *dev_priv, enum pipe pipe, bool state);
 #define assert_pipe_enabled(d, p) assert_pipe(d, p, true)
 #define assert_pipe_disabled(d, p) assert_pipe(d, p, false)
+void intel_write_eld(struct drm_encoder *encoder,
+		     struct drm_display_mode *mode);
+unsigned long intel_gen4_compute_page_offset(int *x, int *y,
+					     unsigned int tiling_mode,
+					     unsigned int bpp,
+					     unsigned int pitch);
+void intel_display_handle_reset(struct drm_device *dev);
+void hsw_enable_pc8_work(struct work_struct *__work);
+void hsw_enable_package_c8(struct drm_i915_private *dev_priv);
+void hsw_disable_package_c8(struct drm_i915_private *dev_priv);
+void intel_dp_get_m_n(struct intel_crtc *crtc,
+		      struct intel_crtc_config *pipe_config);
+int intel_dotclock_calculate(int link_freq, const struct intel_link_m_n *m_n);
+void
+ironlake_check_encoder_dotclock(const struct intel_crtc_config *pipe_config,
+				int dotclock);
+bool intel_crtc_active(struct drm_crtc *crtc);
+void i915_disable_vga_mem(struct drm_device *dev);
+void hsw_enable_ips(struct intel_crtc *crtc);
+void hsw_disable_ips(struct intel_crtc *crtc);
+void intel_display_set_init_power(struct drm_device *dev, bool enable);
+
+
+/* intel_dp.c */
+void intel_dp_init(struct drm_device *dev, int output_reg, enum port port);
+bool intel_dp_init_connector(struct intel_digital_port *intel_dig_port,
+			     struct intel_connector *intel_connector);
+void intel_dp_start_link_train(struct intel_dp *intel_dp);
+void intel_dp_complete_link_train(struct intel_dp *intel_dp);
+void intel_dp_stop_link_train(struct intel_dp *intel_dp);
+void intel_dp_sink_dpms(struct intel_dp *intel_dp, int mode);
+void intel_dp_encoder_destroy(struct drm_encoder *encoder);
+void intel_dp_check_link_status(struct intel_dp *intel_dp);
+bool intel_dp_compute_config(struct intel_encoder *encoder,
+			     struct intel_crtc_config *pipe_config);
+bool intel_dpd_is_edp(struct drm_device *dev);
+void ironlake_edp_backlight_on(struct intel_dp *intel_dp);
+void ironlake_edp_backlight_off(struct intel_dp *intel_dp);
+void ironlake_edp_panel_on(struct intel_dp *intel_dp);
+void ironlake_edp_panel_off(struct intel_dp *intel_dp);
+void ironlake_edp_panel_vdd_on(struct intel_dp *intel_dp);
+void ironlake_edp_panel_vdd_off(struct intel_dp *intel_dp, bool sync);
+void intel_edp_psr_enable(struct intel_dp *intel_dp);
+void intel_edp_psr_disable(struct intel_dp *intel_dp);
+void intel_edp_psr_update(struct drm_device *dev);
+
+
+/* intel_dsi.c */
+bool intel_dsi_init(struct drm_device *dev);
+
+
+/* intel_dvo.c */
+void intel_dvo_init(struct drm_device *dev);
+
+
+/* legacy fbdev emulation in intel_fbdev.c */
+#ifdef CONFIG_DRM_I915_FBDEV
+extern int intel_fbdev_init(struct drm_device *dev);
+extern void intel_fbdev_initial_config(struct drm_device *dev);
+extern void intel_fbdev_fini(struct drm_device *dev);
+extern void intel_fbdev_set_suspend(struct drm_device *dev, int state);
+extern void intel_fbdev_output_poll_changed(struct drm_device *dev);
+extern void intel_fbdev_restore_mode(struct drm_device *dev);
+#else
+static inline int intel_fbdev_init(struct drm_device *dev)
+{
+	return 0;
+}
 
-extern void intel_init_clock_gating(struct drm_device *dev);
-extern void intel_suspend_hw(struct drm_device *dev);
-extern void intel_write_eld(struct drm_encoder *encoder,
-			    struct drm_display_mode *mode);
-extern void intel_prepare_ddi(struct drm_device *dev);
-extern void hsw_fdi_link_train(struct drm_crtc *crtc);
-extern void intel_ddi_init(struct drm_device *dev, enum port port);
-
-/* For use by IVB LP watermark workaround in intel_sprite.c */
-extern void intel_update_watermarks(struct drm_device *dev);
-extern void intel_update_sprite_watermarks(struct drm_plane *plane,
-					   struct drm_crtc *crtc,
-					   uint32_t sprite_width, int pixel_size,
-					   bool enabled, bool scaled);
-
-extern unsigned long intel_gen4_compute_page_offset(int *x, int *y,
-						    unsigned int tiling_mode,
-						    unsigned int bpp,
-						    unsigned int pitch);
-
-extern int intel_sprite_set_colorkey(struct drm_device *dev, void *data,
-				     struct drm_file *file_priv);
-extern int intel_sprite_get_colorkey(struct drm_device *dev, void *data,
-				     struct drm_file *file_priv);
-
-/* Power-related functions, located in intel_pm.c */
-extern void intel_init_pm(struct drm_device *dev);
-/* FBC */
-extern bool intel_fbc_enabled(struct drm_device *dev);
-extern void intel_update_fbc(struct drm_device *dev);
-/* IPS */
-extern void intel_gpu_ips_init(struct drm_i915_private *dev_priv);
-extern void intel_gpu_ips_teardown(void);
-
-/* Power well */
-extern int i915_init_power_well(struct drm_device *dev);
-extern void i915_remove_power_well(struct drm_device *dev);
-
-extern bool intel_display_power_enabled(struct drm_device *dev,
-					enum intel_display_power_domain domain);
-extern void intel_init_power_well(struct drm_device *dev);
-extern void intel_set_power_well(struct drm_device *dev, bool enable);
-extern void intel_enable_gt_powersave(struct drm_device *dev);
-extern void intel_disable_gt_powersave(struct drm_device *dev);
-extern void ironlake_teardown_rc6(struct drm_device *dev);
+static inline void intel_fbdev_initial_config(struct drm_device *dev)
+{
+}
+
+static inline void intel_fbdev_fini(struct drm_device *dev)
+{
+}
+
+static inline void intel_fbdev_set_suspend(struct drm_device *dev, int state)
+{
+}
+
+static inline void intel_fbdev_restore_mode(struct drm_device *dev)
+{
+}
+#endif
+
+/* intel_hdmi.c */
+void intel_hdmi_init(struct drm_device *dev, int hdmi_reg, enum port port);
+void intel_hdmi_init_connector(struct intel_digital_port *intel_dig_port,
+			       struct intel_connector *intel_connector);
+struct intel_hdmi *enc_to_intel_hdmi(struct drm_encoder *encoder);
+bool intel_hdmi_compute_config(struct intel_encoder *encoder,
+			       struct intel_crtc_config *pipe_config);
+
+
+/* intel_lvds.c */
+void intel_lvds_init(struct drm_device *dev);
+bool intel_is_dual_link_lvds(struct drm_device *dev);
+
+
+/* intel_modes.c */
+int intel_connector_update_modes(struct drm_connector *connector,
+				 struct edid *edid);
+int intel_ddc_get_modes(struct drm_connector *c, struct i2c_adapter *adapter);
+void intel_attach_force_audio_property(struct drm_connector *connector);
+void intel_attach_broadcast_rgb_property(struct drm_connector *connector);
+
+
+/* intel_overlay.c */
+void intel_setup_overlay(struct drm_device *dev);
+void intel_cleanup_overlay(struct drm_device *dev);
+int intel_overlay_switch_off(struct intel_overlay *overlay);
+int intel_overlay_put_image(struct drm_device *dev, void *data,
+			    struct drm_file *file_priv);
+int intel_overlay_attrs(struct drm_device *dev, void *data,
+			struct drm_file *file_priv);
+
+
+/* intel_panel.c */
+int intel_panel_init(struct intel_panel *panel,
+		     struct drm_display_mode *fixed_mode);
+void intel_panel_fini(struct intel_panel *panel);
+void intel_fixed_panel_mode(const struct drm_display_mode *fixed_mode,
+			    struct drm_display_mode *adjusted_mode);
+void intel_pch_panel_fitting(struct intel_crtc *crtc,
+			     struct intel_crtc_config *pipe_config,
+			     int fitting_mode);
+void intel_gmch_panel_fitting(struct intel_crtc *crtc,
+			      struct intel_crtc_config *pipe_config,
+			      int fitting_mode);
+void intel_panel_set_backlight(struct intel_connector *connector, u32 level,
+			       u32 max);
+int intel_panel_setup_backlight(struct drm_connector *connector);
+void intel_panel_enable_backlight(struct intel_connector *connector);
+void intel_panel_disable_backlight(struct intel_connector *connector);
+void intel_panel_destroy_backlight(struct drm_device *dev);
+enum drm_connector_status intel_panel_detect(struct drm_device *dev);
+
+
+/* intel_pm.c */
+void intel_init_clock_gating(struct drm_device *dev);
+void intel_suspend_hw(struct drm_device *dev);
+void intel_update_watermarks(struct drm_crtc *crtc);
+void intel_update_sprite_watermarks(struct drm_plane *plane,
+				    struct drm_crtc *crtc,
+				    uint32_t sprite_width, int pixel_size,
+				    bool enabled, bool scaled);
+void intel_init_pm(struct drm_device *dev);
+bool intel_fbc_enabled(struct drm_device *dev);
+void intel_update_fbc(struct drm_device *dev);
+void intel_gpu_ips_init(struct drm_i915_private *dev_priv);
+void intel_gpu_ips_teardown(void);
+int intel_power_domains_init(struct drm_device *dev);
+void intel_power_domains_remove(struct drm_device *dev);
+bool intel_display_power_enabled(struct drm_device *dev,
+				 enum intel_display_power_domain domain);
+void intel_display_power_get(struct drm_device *dev,
+			     enum intel_display_power_domain domain);
+void intel_display_power_put(struct drm_device *dev,
+			     enum intel_display_power_domain domain);
+void intel_power_domains_init_hw(struct drm_device *dev);
+void intel_set_power_well(struct drm_device *dev, bool enable);
+void intel_enable_gt_powersave(struct drm_device *dev);
+void intel_disable_gt_powersave(struct drm_device *dev);
+void ironlake_teardown_rc6(struct drm_device *dev);
 void gen6_update_ring_freq(struct drm_device *dev);
+void gen6_rps_idle(struct drm_i915_private *dev_priv);
+void gen6_rps_boost(struct drm_i915_private *dev_priv);
+void intel_aux_display_runtime_get(struct drm_i915_private *dev_priv);
+void intel_aux_display_runtime_put(struct drm_i915_private *dev_priv);
+void ilk_wm_get_hw_state(struct drm_device *dev);
+
+
+/* intel_sdvo.c */
+bool intel_sdvo_init(struct drm_device *dev, uint32_t sdvo_reg, bool is_sdvob);
+
+
+/* intel_sprite.c */
+int intel_plane_init(struct drm_device *dev, enum pipe pipe, int plane);
+void intel_flush_primary_plane(struct drm_i915_private *dev_priv,
+			       enum plane plane);
+void intel_plane_restore(struct drm_plane *plane);
+void intel_plane_disable(struct drm_plane *plane);
+int intel_sprite_set_colorkey(struct drm_device *dev, void *data,
+			      struct drm_file *file_priv);
+int intel_sprite_get_colorkey(struct drm_device *dev, void *data,
+			      struct drm_file *file_priv);
+
 
-extern bool intel_ddi_get_hw_state(struct intel_encoder *encoder,
-				   enum pipe *pipe);
-extern int intel_ddi_get_cdclk_freq(struct drm_i915_private *dev_priv);
-extern void intel_ddi_pll_init(struct drm_device *dev);
-extern void intel_ddi_enable_transcoder_func(struct drm_crtc *crtc);
-extern void intel_ddi_disable_transcoder_func(struct drm_i915_private *dev_priv,
-					      enum transcoder cpu_transcoder);
-extern void intel_ddi_enable_pipe_clock(struct intel_crtc *intel_crtc);
-extern void intel_ddi_disable_pipe_clock(struct intel_crtc *intel_crtc);
-extern void intel_ddi_setup_hw_pll_state(struct drm_device *dev);
-extern bool intel_ddi_pll_mode_set(struct drm_crtc *crtc);
-extern void intel_ddi_put_crtc_pll(struct drm_crtc *crtc);
-extern void intel_ddi_set_pipe_settings(struct drm_crtc *crtc);
-extern void intel_ddi_prepare_link_retrain(struct drm_encoder *encoder);
-extern bool
-intel_ddi_connector_get_hw_state(struct intel_connector *intel_connector);
-extern void intel_ddi_fdi_disable(struct drm_crtc *crtc);
-extern void intel_ddi_get_config(struct intel_encoder *encoder,
-				 struct intel_crtc_config *pipe_config);
-
-extern void intel_display_handle_reset(struct drm_device *dev);
-extern bool intel_set_cpu_fifo_underrun_reporting(struct drm_device *dev,
-						  enum pipe pipe,
-						  bool enable);
-extern bool intel_set_pch_fifo_underrun_reporting(struct drm_device *dev,
-						 enum transcoder pch_transcoder,
-						 bool enable);
-
-extern void intel_edp_psr_enable(struct intel_dp *intel_dp);
-extern void intel_edp_psr_disable(struct intel_dp *intel_dp);
-extern void intel_edp_psr_update(struct drm_device *dev);
-extern void hsw_disable_lcpll(struct drm_i915_private *dev_priv,
-			      bool switch_to_fclk, bool allow_power_down);
-extern void hsw_restore_lcpll(struct drm_i915_private *dev_priv);
-extern void ilk_enable_gt_irq(struct drm_i915_private *dev_priv, uint32_t mask);
-extern void ilk_disable_gt_irq(struct drm_i915_private *dev_priv,
-			       uint32_t mask);
-extern void snb_enable_pm_irq(struct drm_i915_private *dev_priv, uint32_t mask);
-extern void snb_disable_pm_irq(struct drm_i915_private *dev_priv,
-			       uint32_t mask);
-extern void hsw_enable_pc8_work(struct work_struct *__work);
-extern void hsw_enable_package_c8(struct drm_i915_private *dev_priv);
-extern void hsw_disable_package_c8(struct drm_i915_private *dev_priv);
-extern void hsw_pc8_disable_interrupts(struct drm_device *dev);
-extern void hsw_pc8_restore_interrupts(struct drm_device *dev);
-extern void intel_aux_display_runtime_get(struct drm_i915_private *dev_priv);
-extern void intel_aux_display_runtime_put(struct drm_i915_private *dev_priv);
+/* intel_tv.c */
+void intel_tv_init(struct drm_device *dev);
 
 #endif /* __INTEL_DRV_H__ */
diff --git a/drivers/gpu/drm/i915/intel_dsi.c b/drivers/gpu/drm/i915/intel_dsi.c
new file mode 100644
index 000000000000..d257b093ca68
--- /dev/null
+++ b/drivers/gpu/drm/i915/intel_dsi.c
@@ -0,0 +1,620 @@
+/*
+ * Copyright © 2013 Intel Corporation
+ *
+ * Permission is hereby granted, free of charge, to any person obtaining a
+ * copy of this software and associated documentation files (the "Software"),
+ * to deal in the Software without restriction, including without limitation
+ * the rights to use, copy, modify, merge, publish, distribute, sublicense,
+ * and/or sell copies of the Software, and to permit persons to whom the
+ * Software is furnished to do so, subject to the following conditions:
+ *
+ * The above copyright notice and this permission notice (including the next
+ * paragraph) shall be included in all copies or substantial portions of the
+ * Software.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+ * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+ * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT.  IN NO EVENT SHALL
+ * THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
+ * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING
+ * FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER
+ * DEALINGS IN THE SOFTWARE.
+ *
+ * Author: Jani Nikula <jani.nikula@intel.com>
+ */
+
+#include <drm/drmP.h>
+#include <drm/drm_crtc.h>
+#include <drm/drm_edid.h>
+#include <drm/i915_drm.h>
+#include <linux/slab.h>
+#include "i915_drv.h"
+#include "intel_drv.h"
+#include "intel_dsi.h"
+#include "intel_dsi_cmd.h"
+
+/* the sub-encoders aka panel drivers */
+static const struct intel_dsi_device intel_dsi_devices[] = {
+};
+
+
+static void vlv_cck_modify(struct drm_i915_private *dev_priv, u32 reg, u32 val,
+			   u32 mask)
+{
+	u32 tmp = vlv_cck_read(dev_priv, reg);
+	tmp &= ~mask;
+	tmp |= val;
+	vlv_cck_write(dev_priv, reg, tmp);
+}
+
+static void band_gap_wa(struct drm_i915_private *dev_priv)
+{
+	mutex_lock(&dev_priv->dpio_lock);
+
+	/* Enable bandgap fix in GOP driver */
+	vlv_cck_modify(dev_priv, 0x6D, 0x00010000, 0x00030000);
+	msleep(20);
+	vlv_cck_modify(dev_priv, 0x6E, 0x00010000, 0x00030000);
+	msleep(20);
+	vlv_cck_modify(dev_priv, 0x6F, 0x00010000, 0x00030000);
+	msleep(20);
+	vlv_cck_modify(dev_priv, 0x00, 0x00008000, 0x00008000);
+	msleep(20);
+	vlv_cck_modify(dev_priv, 0x00, 0x00000000, 0x00008000);
+	msleep(20);
+
+	/* Turn Display Trunk on */
+	vlv_cck_modify(dev_priv, 0x6B, 0x00020000, 0x00030000);
+	msleep(20);
+
+	vlv_cck_modify(dev_priv, 0x6C, 0x00020000, 0x00030000);
+	msleep(20);
+
+	vlv_cck_modify(dev_priv, 0x6D, 0x00020000, 0x00030000);
+	msleep(20);
+	vlv_cck_modify(dev_priv, 0x6E, 0x00020000, 0x00030000);
+	msleep(20);
+	vlv_cck_modify(dev_priv, 0x6F, 0x00020000, 0x00030000);
+
+	mutex_unlock(&dev_priv->dpio_lock);
+
+	/* Need huge delay, otherwise clock is not stable */
+	msleep(100);
+}
+
+static struct intel_dsi *intel_attached_dsi(struct drm_connector *connector)
+{
+	return container_of(intel_attached_encoder(connector),
+			    struct intel_dsi, base);
+}
+
+static inline bool is_vid_mode(struct intel_dsi *intel_dsi)
+{
+	return intel_dsi->dev.type == INTEL_DSI_VIDEO_MODE;
+}
+
+static inline bool is_cmd_mode(struct intel_dsi *intel_dsi)
+{
+	return intel_dsi->dev.type == INTEL_DSI_COMMAND_MODE;
+}
+
+static void intel_dsi_hot_plug(struct intel_encoder *encoder)
+{
+	DRM_DEBUG_KMS("\n");
+}
+
+static bool intel_dsi_compute_config(struct intel_encoder *encoder,
+				     struct intel_crtc_config *config)
+{
+	struct intel_dsi *intel_dsi = container_of(encoder, struct intel_dsi,
+						   base);
+	struct intel_connector *intel_connector = intel_dsi->attached_connector;
+	struct drm_display_mode *fixed_mode = intel_connector->panel.fixed_mode;
+	struct drm_display_mode *adjusted_mode = &config->adjusted_mode;
+	struct drm_display_mode *mode = &config->requested_mode;
+
+	DRM_DEBUG_KMS("\n");
+
+	if (fixed_mode)
+		intel_fixed_panel_mode(fixed_mode, adjusted_mode);
+
+	if (intel_dsi->dev.dev_ops->mode_fixup)
+		return intel_dsi->dev.dev_ops->mode_fixup(&intel_dsi->dev,
+							  mode, adjusted_mode);
+
+	return true;
+}
+
+static void intel_dsi_pre_pll_enable(struct intel_encoder *encoder)
+{
+	DRM_DEBUG_KMS("\n");
+
+	vlv_enable_dsi_pll(encoder);
+}
+
+static void intel_dsi_pre_enable(struct intel_encoder *encoder)
+{
+	DRM_DEBUG_KMS("\n");
+}
+
+static void intel_dsi_enable(struct intel_encoder *encoder)
+{
+	struct drm_i915_private *dev_priv = encoder->base.dev->dev_private;
+	struct intel_crtc *intel_crtc = to_intel_crtc(encoder->base.crtc);
+	struct intel_dsi *intel_dsi = enc_to_intel_dsi(&encoder->base);
+	int pipe = intel_crtc->pipe;
+	u32 temp;
+
+	DRM_DEBUG_KMS("\n");
+
+	temp = I915_READ(MIPI_DEVICE_READY(pipe));
+	if ((temp & DEVICE_READY) == 0) {
+		temp &= ~ULPS_STATE_MASK;
+		I915_WRITE(MIPI_DEVICE_READY(pipe), temp | DEVICE_READY);
+	} else if (temp & ULPS_STATE_MASK) {
+		temp &= ~ULPS_STATE_MASK;
+		I915_WRITE(MIPI_DEVICE_READY(pipe), temp | ULPS_STATE_EXIT);
+		/*
+		 * We need to ensure that there is a minimum of 1 ms time
+		 * available before clearing the UPLS exit state.
+		 */
+		msleep(2);
+		I915_WRITE(MIPI_DEVICE_READY(pipe), temp);
+	}
+
+	if (is_cmd_mode(intel_dsi))
+		I915_WRITE(MIPI_MAX_RETURN_PKT_SIZE(pipe), 8 * 4);
+
+	if (is_vid_mode(intel_dsi)) {
+		msleep(20); /* XXX */
+		dpi_send_cmd(intel_dsi, TURN_ON);
+		msleep(100);
+
+		/* assert ip_tg_enable signal */
+		temp = I915_READ(MIPI_PORT_CTRL(pipe));
+		I915_WRITE(MIPI_PORT_CTRL(pipe), temp | DPI_ENABLE);
+		POSTING_READ(MIPI_PORT_CTRL(pipe));
+	}
+
+	intel_dsi->dev.dev_ops->enable(&intel_dsi->dev);
+}
+
+static void intel_dsi_disable(struct intel_encoder *encoder)
+{
+	struct drm_i915_private *dev_priv = encoder->base.dev->dev_private;
+	struct intel_crtc *intel_crtc = to_intel_crtc(encoder->base.crtc);
+	struct intel_dsi *intel_dsi = enc_to_intel_dsi(&encoder->base);
+	int pipe = intel_crtc->pipe;
+	u32 temp;
+
+	DRM_DEBUG_KMS("\n");
+
+	intel_dsi->dev.dev_ops->disable(&intel_dsi->dev);
+
+	if (is_vid_mode(intel_dsi)) {
+		dpi_send_cmd(intel_dsi, SHUTDOWN);
+		msleep(10);
+
+		/* de-assert ip_tg_enable signal */
+		temp = I915_READ(MIPI_PORT_CTRL(pipe));
+		I915_WRITE(MIPI_PORT_CTRL(pipe), temp & ~DPI_ENABLE);
+		POSTING_READ(MIPI_PORT_CTRL(pipe));
+
+		msleep(2);
+	}
+
+	temp = I915_READ(MIPI_DEVICE_READY(pipe));
+	if (temp & DEVICE_READY) {
+		temp &= ~DEVICE_READY;
+		temp &= ~ULPS_STATE_MASK;
+		I915_WRITE(MIPI_DEVICE_READY(pipe), temp);
+	}
+}
+
+static void intel_dsi_post_disable(struct intel_encoder *encoder)
+{
+	DRM_DEBUG_KMS("\n");
+
+	vlv_disable_dsi_pll(encoder);
+}
+
+static bool intel_dsi_get_hw_state(struct intel_encoder *encoder,
+				   enum pipe *pipe)
+{
+	struct drm_i915_private *dev_priv = encoder->base.dev->dev_private;
+	u32 port, func;
+	enum pipe p;
+
+	DRM_DEBUG_KMS("\n");
+
+	/* XXX: this only works for one DSI output */
+	for (p = PIPE_A; p <= PIPE_B; p++) {
+		port = I915_READ(MIPI_PORT_CTRL(p));
+		func = I915_READ(MIPI_DSI_FUNC_PRG(p));
+
+		if ((port & DPI_ENABLE) || (func & CMD_MODE_DATA_WIDTH_MASK)) {
+			if (I915_READ(MIPI_DEVICE_READY(p)) & DEVICE_READY) {
+				*pipe = p;
+				return true;
+			}
+		}
+	}
+
+	return false;
+}
+
+static void intel_dsi_get_config(struct intel_encoder *encoder,
+				 struct intel_crtc_config *pipe_config)
+{
+	DRM_DEBUG_KMS("\n");
+
+	/* XXX: read flags, set to adjusted_mode */
+}
+
+static int intel_dsi_mode_valid(struct drm_connector *connector,
+				struct drm_display_mode *mode)
+{
+	struct intel_connector *intel_connector = to_intel_connector(connector);
+	struct drm_display_mode *fixed_mode = intel_connector->panel.fixed_mode;
+	struct intel_dsi *intel_dsi = intel_attached_dsi(connector);
+
+	DRM_DEBUG_KMS("\n");
+
+	if (mode->flags & DRM_MODE_FLAG_DBLSCAN) {
+		DRM_DEBUG_KMS("MODE_NO_DBLESCAN\n");
+		return MODE_NO_DBLESCAN;
+	}
+
+	if (fixed_mode) {
+		if (mode->hdisplay > fixed_mode->hdisplay)
+			return MODE_PANEL;
+		if (mode->vdisplay > fixed_mode->vdisplay)
+			return MODE_PANEL;
+	}
+
+	return intel_dsi->dev.dev_ops->mode_valid(&intel_dsi->dev, mode);
+}
+
+/* return txclkesc cycles in terms of divider and duration in us */
+static u16 txclkesc(u32 divider, unsigned int us)
+{
+	switch (divider) {
+	case ESCAPE_CLOCK_DIVIDER_1:
+	default:
+		return 20 * us;
+	case ESCAPE_CLOCK_DIVIDER_2:
+		return 10 * us;
+	case ESCAPE_CLOCK_DIVIDER_4:
+		return 5 * us;
+	}
+}
+
+/* return pixels in terms of txbyteclkhs */
+static u16 txbyteclkhs(u16 pixels, int bpp, int lane_count)
+{
+	return DIV_ROUND_UP(DIV_ROUND_UP(pixels * bpp, 8), lane_count);
+}
+
+static void set_dsi_timings(struct drm_encoder *encoder,
+			    const struct drm_display_mode *mode)
+{
+	struct drm_device *dev = encoder->dev;
+	struct drm_i915_private *dev_priv = dev->dev_private;
+	struct intel_crtc *intel_crtc = to_intel_crtc(encoder->crtc);
+	struct intel_dsi *intel_dsi = enc_to_intel_dsi(encoder);
+	int pipe = intel_crtc->pipe;
+	unsigned int bpp = intel_crtc->config.pipe_bpp;
+	unsigned int lane_count = intel_dsi->lane_count;
+
+	u16 hactive, hfp, hsync, hbp, vfp, vsync, vbp;
+
+	hactive = mode->hdisplay;
+	hfp = mode->hsync_start - mode->hdisplay;
+	hsync = mode->hsync_end - mode->hsync_start;
+	hbp = mode->htotal - mode->hsync_end;
+
+	vfp = mode->vsync_start - mode->vdisplay;
+	vsync = mode->vsync_end - mode->vsync_start;
+	vbp = mode->vtotal - mode->vsync_end;
+
+	/* horizontal values are in terms of high speed byte clock */
+	hactive = txbyteclkhs(hactive, bpp, lane_count);
+	hfp = txbyteclkhs(hfp, bpp, lane_count);
+	hsync = txbyteclkhs(hsync, bpp, lane_count);
+	hbp = txbyteclkhs(hbp, bpp, lane_count);
+
+	I915_WRITE(MIPI_HACTIVE_AREA_COUNT(pipe), hactive);
+	I915_WRITE(MIPI_HFP_COUNT(pipe), hfp);
+
+	/* meaningful for video mode non-burst sync pulse mode only, can be zero
+	 * for non-burst sync events and burst modes */
+	I915_WRITE(MIPI_HSYNC_PADDING_COUNT(pipe), hsync);
+	I915_WRITE(MIPI_HBP_COUNT(pipe), hbp);
+
+	/* vertical values are in terms of lines */
+	I915_WRITE(MIPI_VFP_COUNT(pipe), vfp);
+	I915_WRITE(MIPI_VSYNC_PADDING_COUNT(pipe), vsync);
+	I915_WRITE(MIPI_VBP_COUNT(pipe), vbp);
+}
+
+static void intel_dsi_mode_set(struct intel_encoder *intel_encoder)
+{
+	struct drm_encoder *encoder = &intel_encoder->base;
+	struct drm_device *dev = encoder->dev;
+	struct drm_i915_private *dev_priv = dev->dev_private;
+	struct intel_crtc *intel_crtc = to_intel_crtc(encoder->crtc);
+	struct intel_dsi *intel_dsi = enc_to_intel_dsi(encoder);
+	struct drm_display_mode *adjusted_mode =
+		&intel_crtc->config.adjusted_mode;
+	int pipe = intel_crtc->pipe;
+	unsigned int bpp = intel_crtc->config.pipe_bpp;
+	u32 val, tmp;
+
+	DRM_DEBUG_KMS("pipe %c\n", pipe_name(pipe));
+
+	/* Update the DSI PLL */
+	vlv_enable_dsi_pll(intel_encoder);
+
+	/* XXX: Location of the call */
+	band_gap_wa(dev_priv);
+
+	/* escape clock divider, 20MHz, shared for A and C. device ready must be
+	 * off when doing this! txclkesc? */
+	tmp = I915_READ(MIPI_CTRL(0));
+	tmp &= ~ESCAPE_CLOCK_DIVIDER_MASK;
+	I915_WRITE(MIPI_CTRL(0), tmp | ESCAPE_CLOCK_DIVIDER_1);
+
+	/* read request priority is per pipe */
+	tmp = I915_READ(MIPI_CTRL(pipe));
+	tmp &= ~READ_REQUEST_PRIORITY_MASK;
+	I915_WRITE(MIPI_CTRL(pipe), tmp | READ_REQUEST_PRIORITY_HIGH);
+
+	/* XXX: why here, why like this? handling in irq handler?! */
+	I915_WRITE(MIPI_INTR_STAT(pipe), 0xffffffff);
+	I915_WRITE(MIPI_INTR_EN(pipe), 0xffffffff);
+
+	I915_WRITE(MIPI_DPHY_PARAM(pipe),
+		   0x3c << EXIT_ZERO_COUNT_SHIFT |
+		   0x1f << TRAIL_COUNT_SHIFT |
+		   0xc5 << CLK_ZERO_COUNT_SHIFT |
+		   0x1f << PREPARE_COUNT_SHIFT);
+
+	I915_WRITE(MIPI_DPI_RESOLUTION(pipe),
+		   adjusted_mode->vdisplay << VERTICAL_ADDRESS_SHIFT |
+		   adjusted_mode->hdisplay << HORIZONTAL_ADDRESS_SHIFT);
+
+	set_dsi_timings(encoder, adjusted_mode);
+
+	val = intel_dsi->lane_count << DATA_LANES_PRG_REG_SHIFT;
+	if (is_cmd_mode(intel_dsi)) {
+		val |= intel_dsi->channel << CMD_MODE_CHANNEL_NUMBER_SHIFT;
+		val |= CMD_MODE_DATA_WIDTH_8_BIT; /* XXX */
+	} else {
+		val |= intel_dsi->channel << VID_MODE_CHANNEL_NUMBER_SHIFT;
+
+		/* XXX: cross-check bpp vs. pixel format? */
+		val |= intel_dsi->pixel_format;
+	}
+	I915_WRITE(MIPI_DSI_FUNC_PRG(pipe), val);
+
+	/* timeouts for recovery. one frame IIUC. if counter expires, EOT and
+	 * stop state. */
+
+	/*
+	 * In burst mode, value greater than one DPI line Time in byte clock
+	 * (txbyteclkhs) To timeout this timer 1+ of the above said value is
+	 * recommended.
+	 *
+	 * In non-burst mode, Value greater than one DPI frame time in byte
+	 * clock(txbyteclkhs) To timeout this timer 1+ of the above said value
+	 * is recommended.
+	 *
+	 * In DBI only mode, value greater than one DBI frame time in byte
+	 * clock(txbyteclkhs) To timeout this timer 1+ of the above said value
+	 * is recommended.
+	 */
+
+	if (is_vid_mode(intel_dsi) &&
+	    intel_dsi->video_mode_format == VIDEO_MODE_BURST) {
+		I915_WRITE(MIPI_HS_TX_TIMEOUT(pipe),
+			   txbyteclkhs(adjusted_mode->htotal, bpp,
+				       intel_dsi->lane_count) + 1);
+	} else {
+		I915_WRITE(MIPI_HS_TX_TIMEOUT(pipe),
+			   txbyteclkhs(adjusted_mode->vtotal *
+				       adjusted_mode->htotal,
+				       bpp, intel_dsi->lane_count) + 1);
+	}
+	I915_WRITE(MIPI_LP_RX_TIMEOUT(pipe), 8309); /* max */
+	I915_WRITE(MIPI_TURN_AROUND_TIMEOUT(pipe), 0x14); /* max */
+	I915_WRITE(MIPI_DEVICE_RESET_TIMER(pipe), 0xffff); /* max */
+
+	/* dphy stuff */
+
+	/* in terms of low power clock */
+	I915_WRITE(MIPI_INIT_COUNT(pipe), txclkesc(ESCAPE_CLOCK_DIVIDER_1, 100));
+
+	/* recovery disables */
+	I915_WRITE(MIPI_EOT_DISABLE(pipe), intel_dsi->eot_disable);
+
+	/* in terms of txbyteclkhs. actual high to low switch +
+	 * MIPI_STOP_STATE_STALL * MIPI_LP_BYTECLK.
+	 *
+	 * XXX: write MIPI_STOP_STATE_STALL?
+	 */
+	I915_WRITE(MIPI_HIGH_LOW_SWITCH_COUNT(pipe), 0x46);
+
+	/* XXX: low power clock equivalence in terms of byte clock. the number
+	 * of byte clocks occupied in one low power clock. based on txbyteclkhs
+	 * and txclkesc. txclkesc time / txbyteclk time * (105 +
+	 * MIPI_STOP_STATE_STALL) / 105.???
+	 */
+	I915_WRITE(MIPI_LP_BYTECLK(pipe), 4);
+
+	/* the bw essential for transmitting 16 long packets containing 252
+	 * bytes meant for dcs write memory command is programmed in this
+	 * register in terms of byte clocks. based on dsi transfer rate and the
+	 * number of lanes configured the time taken to transmit 16 long packets
+	 * in a dsi stream varies. */
+	I915_WRITE(MIPI_DBI_BW_CTRL(pipe), 0x820);
+
+	I915_WRITE(MIPI_CLK_LANE_SWITCH_TIME_CNT(pipe),
+		   0xa << LP_HS_SSW_CNT_SHIFT |
+		   0x14 << HS_LP_PWR_SW_CNT_SHIFT);
+
+	if (is_vid_mode(intel_dsi))
+		I915_WRITE(MIPI_VIDEO_MODE_FORMAT(pipe),
+			   intel_dsi->video_mode_format);
+}
+
+static enum drm_connector_status
+intel_dsi_detect(struct drm_connector *connector, bool force)
+{
+	struct intel_dsi *intel_dsi = intel_attached_dsi(connector);
+	DRM_DEBUG_KMS("\n");
+	return intel_dsi->dev.dev_ops->detect(&intel_dsi->dev);
+}
+
+static int intel_dsi_get_modes(struct drm_connector *connector)
+{
+	struct intel_connector *intel_connector = to_intel_connector(connector);
+	struct drm_display_mode *mode;
+
+	DRM_DEBUG_KMS("\n");
+
+	if (!intel_connector->panel.fixed_mode) {
+		DRM_DEBUG_KMS("no fixed mode\n");
+		return 0;
+	}
+
+	mode = drm_mode_duplicate(connector->dev,
+				  intel_connector->panel.fixed_mode);
+	if (!mode) {
+		DRM_DEBUG_KMS("drm_mode_duplicate failed\n");
+		return 0;
+	}
+
+	drm_mode_probed_add(connector, mode);
+	return 1;
+}
+
+static void intel_dsi_destroy(struct drm_connector *connector)
+{
+	struct intel_connector *intel_connector = to_intel_connector(connector);
+
+	DRM_DEBUG_KMS("\n");
+	intel_panel_fini(&intel_connector->panel);
+	drm_connector_cleanup(connector);
+	kfree(connector);
+}
+
+static const struct drm_encoder_funcs intel_dsi_funcs = {
+	.destroy = intel_encoder_destroy,
+};
+
+static const struct drm_connector_helper_funcs intel_dsi_connector_helper_funcs = {
+	.get_modes = intel_dsi_get_modes,
+	.mode_valid = intel_dsi_mode_valid,
+	.best_encoder = intel_best_encoder,
+};
+
+static const struct drm_connector_funcs intel_dsi_connector_funcs = {
+	.dpms = intel_connector_dpms,
+	.detect = intel_dsi_detect,
+	.destroy = intel_dsi_destroy,
+	.fill_modes = drm_helper_probe_single_connector_modes,
+};
+
+bool intel_dsi_init(struct drm_device *dev)
+{
+	struct intel_dsi *intel_dsi;
+	struct intel_encoder *intel_encoder;
+	struct drm_encoder *encoder;
+	struct intel_connector *intel_connector;
+	struct drm_connector *connector;
+	struct drm_display_mode *fixed_mode = NULL;
+	const struct intel_dsi_device *dsi;
+	unsigned int i;
+
+	DRM_DEBUG_KMS("\n");
+
+	intel_dsi = kzalloc(sizeof(*intel_dsi), GFP_KERNEL);
+	if (!intel_dsi)
+		return false;
+
+	intel_connector = kzalloc(sizeof(*intel_connector), GFP_KERNEL);
+	if (!intel_connector) {
+		kfree(intel_dsi);
+		return false;
+	}
+
+	intel_encoder = &intel_dsi->base;
+	encoder = &intel_encoder->base;
+	intel_dsi->attached_connector = intel_connector;
+
+	connector = &intel_connector->base;
+
+	drm_encoder_init(dev, encoder, &intel_dsi_funcs, DRM_MODE_ENCODER_DSI);
+
+	/* XXX: very likely not all of these are needed */
+	intel_encoder->hot_plug = intel_dsi_hot_plug;
+	intel_encoder->compute_config = intel_dsi_compute_config;
+	intel_encoder->pre_pll_enable = intel_dsi_pre_pll_enable;
+	intel_encoder->pre_enable = intel_dsi_pre_enable;
+	intel_encoder->enable = intel_dsi_enable;
+	intel_encoder->mode_set = intel_dsi_mode_set;
+	intel_encoder->disable = intel_dsi_disable;
+	intel_encoder->post_disable = intel_dsi_post_disable;
+	intel_encoder->get_hw_state = intel_dsi_get_hw_state;
+	intel_encoder->get_config = intel_dsi_get_config;
+
+	intel_connector->get_hw_state = intel_connector_get_hw_state;
+
+	for (i = 0; i < ARRAY_SIZE(intel_dsi_devices); i++) {
+		dsi = &intel_dsi_devices[i];
+		intel_dsi->dev = *dsi;
+
+		if (dsi->dev_ops->init(&intel_dsi->dev))
+			break;
+	}
+
+	if (i == ARRAY_SIZE(intel_dsi_devices)) {
+		DRM_DEBUG_KMS("no device found\n");
+		goto err;
+	}
+
+	intel_encoder->type = INTEL_OUTPUT_DSI;
+	intel_encoder->crtc_mask = (1 << 0); /* XXX */
+
+	intel_encoder->cloneable = false;
+	drm_connector_init(dev, connector, &intel_dsi_connector_funcs,
+			   DRM_MODE_CONNECTOR_DSI);
+
+	drm_connector_helper_add(connector, &intel_dsi_connector_helper_funcs);
+
+	connector->display_info.subpixel_order = SubPixelHorizontalRGB; /*XXX*/
+	connector->interlace_allowed = false;
+	connector->doublescan_allowed = false;
+
+	intel_connector_attach_encoder(intel_connector, intel_encoder);
+
+	drm_sysfs_connector_add(connector);
+
+	fixed_mode = dsi->dev_ops->get_modes(&intel_dsi->dev);
+	if (!fixed_mode) {
+		DRM_DEBUG_KMS("no fixed mode\n");
+		goto err;
+	}
+
+	fixed_mode->type |= DRM_MODE_TYPE_PREFERRED;
+	intel_panel_init(&intel_connector->panel, fixed_mode);
+
+	return true;
+
+err:
+	drm_encoder_cleanup(&intel_encoder->base);
+	kfree(intel_dsi);
+	kfree(intel_connector);
+
+	return false;
+}
diff --git a/drivers/gpu/drm/i915/intel_dsi.h b/drivers/gpu/drm/i915/intel_dsi.h
new file mode 100644
index 000000000000..c7765f33d524
--- /dev/null
+++ b/drivers/gpu/drm/i915/intel_dsi.h
@@ -0,0 +1,102 @@
+/*
+ * Copyright © 2013 Intel Corporation
+ *
+ * Permission is hereby granted, free of charge, to any person obtaining a
+ * copy of this software and associated documentation files (the "Software"),
+ * to deal in the Software without restriction, including without limitation
+ * the rights to use, copy, modify, merge, publish, distribute, sublicense,
+ * and/or sell copies of the Software, and to permit persons to whom the
+ * Software is furnished to do so, subject to the following conditions:
+ *
+ * The above copyright notice and this permission notice (including the next
+ * paragraph) shall be included in all copies or substantial portions of the
+ * Software.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+ * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+ * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT.  IN NO EVENT SHALL
+ * THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
+ * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING
+ * FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER
+ * DEALINGS IN THE SOFTWARE.
+ */
+
+#ifndef _INTEL_DSI_H
+#define _INTEL_DSI_H
+
+#include <drm/drmP.h>
+#include <drm/drm_crtc.h>
+#include "intel_drv.h"
+
+struct intel_dsi_device {
+	unsigned int panel_id;
+	const char *name;
+	int type;
+	const struct intel_dsi_dev_ops *dev_ops;
+	void *dev_priv;
+};
+
+struct intel_dsi_dev_ops {
+	bool (*init)(struct intel_dsi_device *dsi);
+
+	/* This callback must be able to assume DSI commands can be sent */
+	void (*enable)(struct intel_dsi_device *dsi);
+
+	/* This callback must be able to assume DSI commands can be sent */
+	void (*disable)(struct intel_dsi_device *dsi);
+
+	int (*mode_valid)(struct intel_dsi_device *dsi,
+			  struct drm_display_mode *mode);
+
+	bool (*mode_fixup)(struct intel_dsi_device *dsi,
+			   const struct drm_display_mode *mode,
+			   struct drm_display_mode *adjusted_mode);
+
+	void (*mode_set)(struct intel_dsi_device *dsi,
+			 struct drm_display_mode *mode,
+			 struct drm_display_mode *adjusted_mode);
+
+	enum drm_connector_status (*detect)(struct intel_dsi_device *dsi);
+
+	bool (*get_hw_state)(struct intel_dsi_device *dev);
+
+	struct drm_display_mode *(*get_modes)(struct intel_dsi_device *dsi);
+
+	void (*destroy) (struct intel_dsi_device *dsi);
+};
+
+struct intel_dsi {
+	struct intel_encoder base;
+
+	struct intel_dsi_device dev;
+
+	struct intel_connector *attached_connector;
+
+	/* if true, use HS mode, otherwise LP */
+	bool hs;
+
+	/* virtual channel */
+	int channel;
+
+	/* number of DSI lanes */
+	unsigned int lane_count;
+
+	/* video mode pixel format for MIPI_DSI_FUNC_PRG register */
+	u32 pixel_format;
+
+	/* video mode format for MIPI_VIDEO_MODE_FORMAT register */
+	u32 video_mode_format;
+
+	/* eot for MIPI_EOT_DISABLE register */
+	u32 eot_disable;
+};
+
+static inline struct intel_dsi *enc_to_intel_dsi(struct drm_encoder *encoder)
+{
+	return container_of(encoder, struct intel_dsi, base.base);
+}
+
+extern void vlv_enable_dsi_pll(struct intel_encoder *encoder);
+extern void vlv_disable_dsi_pll(struct intel_encoder *encoder);
+
+#endif /* _INTEL_DSI_H */
diff --git a/drivers/gpu/drm/i915/intel_dsi_cmd.c b/drivers/gpu/drm/i915/intel_dsi_cmd.c
new file mode 100644
index 000000000000..7c40f981d2c7
--- /dev/null
+++ b/drivers/gpu/drm/i915/intel_dsi_cmd.c
@@ -0,0 +1,427 @@
+/*
+ * Copyright © 2013 Intel Corporation
+ *
+ * Permission is hereby granted, free of charge, to any person obtaining a
+ * copy of this software and associated documentation files (the "Software"),
+ * to deal in the Software without restriction, including without limitation
+ * the rights to use, copy, modify, merge, publish, distribute, sublicense,
+ * and/or sell copies of the Software, and to permit persons to whom the
+ * Software is furnished to do so, subject to the following conditions:
+ *
+ * The above copyright notice and this permission notice (including the next
+ * paragraph) shall be included in all copies or substantial portions of the
+ * Software.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+ * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+ * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT.  IN NO EVENT SHALL
+ * THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
+ * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING
+ * FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER
+ * DEALINGS IN THE SOFTWARE.
+ *
+ * Author: Jani Nikula <jani.nikula@intel.com>
+ */
+
+#include <linux/export.h>
+#include <drm/drmP.h>
+#include <drm/drm_crtc.h>
+#include <video/mipi_display.h>
+#include "i915_drv.h"
+#include "intel_drv.h"
+#include "intel_dsi.h"
+#include "intel_dsi_cmd.h"
+
+/*
+ * XXX: MIPI_DATA_ADDRESS, MIPI_DATA_LENGTH, MIPI_COMMAND_LENGTH, and
+ * MIPI_COMMAND_ADDRESS registers.
+ *
+ * Apparently these registers provide a MIPI adapter level way to send (lots of)
+ * commands and data to the receiver, without having to write the commands and
+ * data to MIPI_{HS,LP}_GEN_{CTRL,DATA} registers word by word.
+ *
+ * Presumably for anything other than MIPI_DCS_WRITE_MEMORY_START and
+ * MIPI_DCS_WRITE_MEMORY_CONTINUE (which are used to update the external
+ * framebuffer in command mode displays) these are just an optimization that can
+ * come later.
+ *
+ * For memory writes, these should probably be used for performance.
+ */
+
+static void print_stat(struct intel_dsi *intel_dsi)
+{
+	struct drm_encoder *encoder = &intel_dsi->base.base;
+	struct drm_device *dev = encoder->dev;
+	struct drm_i915_private *dev_priv = dev->dev_private;
+	struct intel_crtc *intel_crtc = to_intel_crtc(encoder->crtc);
+	enum pipe pipe = intel_crtc->pipe;
+	u32 val;
+
+	val = I915_READ(MIPI_INTR_STAT(pipe));
+
+#define STAT_BIT(val, bit) (val) & (bit) ? " " #bit : ""
+	DRM_DEBUG_KMS("MIPI_INTR_STAT(%d) = %08x"
+		      "%s%s%s%s%s%s%s%s%s%s%s%s%s%s%s%s%s%s%s%s%s%s%s%s%s%s%s%s%s%s%s%s"
+		      "\n", pipe, val,
+		      STAT_BIT(val, TEARING_EFFECT),
+		      STAT_BIT(val, SPL_PKT_SENT_INTERRUPT),
+		      STAT_BIT(val, GEN_READ_DATA_AVAIL),
+		      STAT_BIT(val, LP_GENERIC_WR_FIFO_FULL),
+		      STAT_BIT(val, HS_GENERIC_WR_FIFO_FULL),
+		      STAT_BIT(val, RX_PROT_VIOLATION),
+		      STAT_BIT(val, RX_INVALID_TX_LENGTH),
+		      STAT_BIT(val, ACK_WITH_NO_ERROR),
+		      STAT_BIT(val, TURN_AROUND_ACK_TIMEOUT),
+		      STAT_BIT(val, LP_RX_TIMEOUT),
+		      STAT_BIT(val, HS_TX_TIMEOUT),
+		      STAT_BIT(val, DPI_FIFO_UNDERRUN),
+		      STAT_BIT(val, LOW_CONTENTION),
+		      STAT_BIT(val, HIGH_CONTENTION),
+		      STAT_BIT(val, TXDSI_VC_ID_INVALID),
+		      STAT_BIT(val, TXDSI_DATA_TYPE_NOT_RECOGNISED),
+		      STAT_BIT(val, TXCHECKSUM_ERROR),
+		      STAT_BIT(val, TXECC_MULTIBIT_ERROR),
+		      STAT_BIT(val, TXECC_SINGLE_BIT_ERROR),
+		      STAT_BIT(val, TXFALSE_CONTROL_ERROR),
+		      STAT_BIT(val, RXDSI_VC_ID_INVALID),
+		      STAT_BIT(val, RXDSI_DATA_TYPE_NOT_REGOGNISED),
+		      STAT_BIT(val, RXCHECKSUM_ERROR),
+		      STAT_BIT(val, RXECC_MULTIBIT_ERROR),
+		      STAT_BIT(val, RXECC_SINGLE_BIT_ERROR),
+		      STAT_BIT(val, RXFALSE_CONTROL_ERROR),
+		      STAT_BIT(val, RXHS_RECEIVE_TIMEOUT_ERROR),
+		      STAT_BIT(val, RX_LP_TX_SYNC_ERROR),
+		      STAT_BIT(val, RXEXCAPE_MODE_ENTRY_ERROR),
+		      STAT_BIT(val, RXEOT_SYNC_ERROR),
+		      STAT_BIT(val, RXSOT_SYNC_ERROR),
+		      STAT_BIT(val, RXSOT_ERROR));
+#undef STAT_BIT
+}
+
+enum dsi_type {
+	DSI_DCS,
+	DSI_GENERIC,
+};
+
+/* enable or disable command mode hs transmissions */
+void dsi_hs_mode_enable(struct intel_dsi *intel_dsi, bool enable)
+{
+	struct drm_encoder *encoder = &intel_dsi->base.base;
+	struct drm_device *dev = encoder->dev;
+	struct drm_i915_private *dev_priv = dev->dev_private;
+	struct intel_crtc *intel_crtc = to_intel_crtc(encoder->crtc);
+	enum pipe pipe = intel_crtc->pipe;
+	u32 temp;
+	u32 mask = DBI_FIFO_EMPTY;
+
+	if (wait_for((I915_READ(MIPI_GEN_FIFO_STAT(pipe)) & mask) == mask, 50))
+		DRM_ERROR("Timeout waiting for DBI FIFO empty\n");
+
+	temp = I915_READ(MIPI_HS_LP_DBI_ENABLE(pipe));
+	temp &= DBI_HS_LP_MODE_MASK;
+	I915_WRITE(MIPI_HS_LP_DBI_ENABLE(pipe), enable ? DBI_HS_MODE : DBI_LP_MODE);
+
+	intel_dsi->hs = enable;
+}
+
+static int dsi_vc_send_short(struct intel_dsi *intel_dsi, int channel,
+			     u8 data_type, u16 data)
+{
+	struct drm_encoder *encoder = &intel_dsi->base.base;
+	struct drm_device *dev = encoder->dev;
+	struct drm_i915_private *dev_priv = dev->dev_private;
+	struct intel_crtc *intel_crtc = to_intel_crtc(encoder->crtc);
+	enum pipe pipe = intel_crtc->pipe;
+	u32 ctrl_reg;
+	u32 ctrl;
+	u32 mask;
+
+	DRM_DEBUG_KMS("channel %d, data_type %d, data %04x\n",
+		      channel, data_type, data);
+
+	if (intel_dsi->hs) {
+		ctrl_reg = MIPI_HS_GEN_CTRL(pipe);
+		mask = HS_CTRL_FIFO_FULL;
+	} else {
+		ctrl_reg = MIPI_LP_GEN_CTRL(pipe);
+		mask = LP_CTRL_FIFO_FULL;
+	}
+
+	if (wait_for((I915_READ(MIPI_GEN_FIFO_STAT(pipe)) & mask) == 0, 50)) {
+		DRM_ERROR("Timeout waiting for HS/LP CTRL FIFO !full\n");
+		print_stat(intel_dsi);
+	}
+
+	/*
+	 * Note: This function is also used for long packets, with length passed
+	 * as data, since SHORT_PACKET_PARAM_SHIFT ==
+	 * LONG_PACKET_WORD_COUNT_SHIFT.
+	 */
+	ctrl = data << SHORT_PACKET_PARAM_SHIFT |
+		channel << VIRTUAL_CHANNEL_SHIFT |
+		data_type << DATA_TYPE_SHIFT;
+
+	I915_WRITE(ctrl_reg, ctrl);
+
+	return 0;
+}
+
+static int dsi_vc_send_long(struct intel_dsi *intel_dsi, int channel,
+			    u8 data_type, const u8 *data, int len)
+{
+	struct drm_encoder *encoder = &intel_dsi->base.base;
+	struct drm_device *dev = encoder->dev;
+	struct drm_i915_private *dev_priv = dev->dev_private;
+	struct intel_crtc *intel_crtc = to_intel_crtc(encoder->crtc);
+	enum pipe pipe = intel_crtc->pipe;
+	u32 data_reg;
+	int i, j, n;
+	u32 mask;
+
+	DRM_DEBUG_KMS("channel %d, data_type %d, len %04x\n",
+		      channel, data_type, len);
+
+	if (intel_dsi->hs) {
+		data_reg = MIPI_HS_GEN_DATA(pipe);
+		mask = HS_DATA_FIFO_FULL;
+	} else {
+		data_reg = MIPI_LP_GEN_DATA(pipe);
+		mask = LP_DATA_FIFO_FULL;
+	}
+
+	if (wait_for((I915_READ(MIPI_GEN_FIFO_STAT(pipe)) & mask) == 0, 50))
+		DRM_ERROR("Timeout waiting for HS/LP DATA FIFO !full\n");
+
+	for (i = 0; i < len; i += n) {
+		u32 val = 0;
+		n = min_t(int, len - i, 4);
+
+		for (j = 0; j < n; j++)
+			val |= *data++ << 8 * j;
+
+		I915_WRITE(data_reg, val);
+		/* XXX: check for data fifo full, once that is set, write 4
+		 * dwords, then wait for not set, then continue. */
+	}
+
+	return dsi_vc_send_short(intel_dsi, channel, data_type, len);
+}
+
+static int dsi_vc_write_common(struct intel_dsi *intel_dsi,
+			       int channel, const u8 *data, int len,
+			       enum dsi_type type)
+{
+	int ret;
+
+	if (len == 0) {
+		BUG_ON(type == DSI_GENERIC);
+		ret = dsi_vc_send_short(intel_dsi, channel,
+					MIPI_DSI_GENERIC_SHORT_WRITE_0_PARAM,
+					0);
+	} else if (len == 1) {
+		ret = dsi_vc_send_short(intel_dsi, channel,
+					type == DSI_GENERIC ?
+					MIPI_DSI_GENERIC_SHORT_WRITE_1_PARAM :
+					MIPI_DSI_DCS_SHORT_WRITE, data[0]);
+	} else if (len == 2) {
+		ret = dsi_vc_send_short(intel_dsi, channel,
+					type == DSI_GENERIC ?
+					MIPI_DSI_GENERIC_SHORT_WRITE_2_PARAM :
+					MIPI_DSI_DCS_SHORT_WRITE_PARAM,
+					(data[1] << 8) | data[0]);
+	} else {
+		ret = dsi_vc_send_long(intel_dsi, channel,
+				       type == DSI_GENERIC ?
+				       MIPI_DSI_GENERIC_LONG_WRITE :
+				       MIPI_DSI_DCS_LONG_WRITE, data, len);
+	}
+
+	return ret;
+}
+
+int dsi_vc_dcs_write(struct intel_dsi *intel_dsi, int channel,
+		     const u8 *data, int len)
+{
+	return dsi_vc_write_common(intel_dsi, channel, data, len, DSI_DCS);
+}
+
+int dsi_vc_generic_write(struct intel_dsi *intel_dsi, int channel,
+			 const u8 *data, int len)
+{
+	return dsi_vc_write_common(intel_dsi, channel, data, len, DSI_GENERIC);
+}
+
+static int dsi_vc_dcs_send_read_request(struct intel_dsi *intel_dsi,
+					int channel, u8 dcs_cmd)
+{
+	return dsi_vc_send_short(intel_dsi, channel, MIPI_DSI_DCS_READ,
+				 dcs_cmd);
+}
+
+static int dsi_vc_generic_send_read_request(struct intel_dsi *intel_dsi,
+					    int channel, u8 *reqdata,
+					    int reqlen)
+{
+	u16 data;
+	u8 data_type;
+
+	switch (reqlen) {
+	case 0:
+		data_type = MIPI_DSI_GENERIC_READ_REQUEST_0_PARAM;
+		data = 0;
+		break;
+	case 1:
+		data_type = MIPI_DSI_GENERIC_READ_REQUEST_1_PARAM;
+		data = reqdata[0];
+		break;
+	case 2:
+		data_type = MIPI_DSI_GENERIC_READ_REQUEST_2_PARAM;
+		data = (reqdata[1] << 8) | reqdata[0];
+		break;
+	default:
+		BUG();
+	}
+
+	return dsi_vc_send_short(intel_dsi, channel, data_type, data);
+}
+
+static int dsi_read_data_return(struct intel_dsi *intel_dsi,
+				u8 *buf, int buflen)
+{
+	struct drm_encoder *encoder = &intel_dsi->base.base;
+	struct drm_device *dev = encoder->dev;
+	struct drm_i915_private *dev_priv = dev->dev_private;
+	struct intel_crtc *intel_crtc = to_intel_crtc(encoder->crtc);
+	enum pipe pipe = intel_crtc->pipe;
+	int i, len = 0;
+	u32 data_reg, val;
+
+	if (intel_dsi->hs) {
+		data_reg = MIPI_HS_GEN_DATA(pipe);
+	} else {
+		data_reg = MIPI_LP_GEN_DATA(pipe);
+	}
+
+	while (len < buflen) {
+		val = I915_READ(data_reg);
+		for (i = 0; i < 4 && len < buflen; i++, len++)
+			buf[len] = val >> 8 * i;
+	}
+
+	return len;
+}
+
+int dsi_vc_dcs_read(struct intel_dsi *intel_dsi, int channel, u8 dcs_cmd,
+		    u8 *buf, int buflen)
+{
+	struct drm_encoder *encoder = &intel_dsi->base.base;
+	struct drm_device *dev = encoder->dev;
+	struct drm_i915_private *dev_priv = dev->dev_private;
+	struct intel_crtc *intel_crtc = to_intel_crtc(encoder->crtc);
+	enum pipe pipe = intel_crtc->pipe;
+	u32 mask;
+	int ret;
+
+	/*
+	 * XXX: should issue multiple read requests and reads if request is
+	 * longer than MIPI_MAX_RETURN_PKT_SIZE
+	 */
+
+	I915_WRITE(MIPI_INTR_STAT(pipe), GEN_READ_DATA_AVAIL);
+
+	ret = dsi_vc_dcs_send_read_request(intel_dsi, channel, dcs_cmd);
+	if (ret)
+		return ret;
+
+	mask = GEN_READ_DATA_AVAIL;
+	if (wait_for((I915_READ(MIPI_INTR_STAT(pipe)) & mask) == mask, 50))
+		DRM_ERROR("Timeout waiting for read data.\n");
+
+	ret = dsi_read_data_return(intel_dsi, buf, buflen);
+	if (ret < 0)
+		return ret;
+
+	if (ret != buflen)
+		return -EIO;
+
+	return 0;
+}
+
+int dsi_vc_generic_read(struct intel_dsi *intel_dsi, int channel,
+			u8 *reqdata, int reqlen, u8 *buf, int buflen)
+{
+	struct drm_encoder *encoder = &intel_dsi->base.base;
+	struct drm_device *dev = encoder->dev;
+	struct drm_i915_private *dev_priv = dev->dev_private;
+	struct intel_crtc *intel_crtc = to_intel_crtc(encoder->crtc);
+	enum pipe pipe = intel_crtc->pipe;
+	u32 mask;
+	int ret;
+
+	/*
+	 * XXX: should issue multiple read requests and reads if request is
+	 * longer than MIPI_MAX_RETURN_PKT_SIZE
+	 */
+
+	I915_WRITE(MIPI_INTR_STAT(pipe), GEN_READ_DATA_AVAIL);
+
+	ret = dsi_vc_generic_send_read_request(intel_dsi, channel, reqdata,
+					       reqlen);
+	if (ret)
+		return ret;
+
+	mask = GEN_READ_DATA_AVAIL;
+	if (wait_for((I915_READ(MIPI_INTR_STAT(pipe)) & mask) == mask, 50))
+		DRM_ERROR("Timeout waiting for read data.\n");
+
+	ret = dsi_read_data_return(intel_dsi, buf, buflen);
+	if (ret < 0)
+		return ret;
+
+	if (ret != buflen)
+		return -EIO;
+
+	return 0;
+}
+
+/*
+ * send a video mode command
+ *
+ * XXX: commands with data in MIPI_DPI_DATA?
+ */
+int dpi_send_cmd(struct intel_dsi *intel_dsi, u32 cmd)
+{
+	struct drm_encoder *encoder = &intel_dsi->base.base;
+	struct drm_device *dev = encoder->dev;
+	struct drm_i915_private *dev_priv = dev->dev_private;
+	struct intel_crtc *intel_crtc = to_intel_crtc(encoder->crtc);
+	enum pipe pipe = intel_crtc->pipe;
+	u32 mask;
+
+	/* XXX: pipe, hs */
+	if (intel_dsi->hs)
+		cmd &= ~DPI_LP_MODE;
+	else
+		cmd |= DPI_LP_MODE;
+
+	/* DPI virtual channel?! */
+
+	mask = DPI_FIFO_EMPTY;
+	if (wait_for((I915_READ(MIPI_GEN_FIFO_STAT(pipe)) & mask) == mask, 50))
+		DRM_ERROR("Timeout waiting for DPI FIFO empty.\n");
+
+	/* clear bit */
+	I915_WRITE(MIPI_INTR_STAT(pipe), SPL_PKT_SENT_INTERRUPT);
+
+	/* XXX: old code skips write if control unchanged */
+	if (cmd == I915_READ(MIPI_DPI_CONTROL(pipe)))
+		DRM_ERROR("Same special packet %02x twice in a row.\n", cmd);
+
+	I915_WRITE(MIPI_DPI_CONTROL(pipe), cmd);
+
+	mask = SPL_PKT_SENT_INTERRUPT;
+	if (wait_for((I915_READ(MIPI_INTR_STAT(pipe)) & mask) == mask, 100))
+		DRM_ERROR("Video mode command 0x%08x send failed.\n", cmd);
+
+	return 0;
+}
diff --git a/drivers/gpu/drm/i915/intel_dsi_cmd.h b/drivers/gpu/drm/i915/intel_dsi_cmd.h
new file mode 100644
index 000000000000..54c8a234a2e0
--- /dev/null
+++ b/drivers/gpu/drm/i915/intel_dsi_cmd.h
@@ -0,0 +1,109 @@
+/*
+ * Copyright © 2013 Intel Corporation
+ *
+ * Permission is hereby granted, free of charge, to any person obtaining a
+ * copy of this software and associated documentation files (the "Software"),
+ * to deal in the Software without restriction, including without limitation
+ * the rights to use, copy, modify, merge, publish, distribute, sublicense,
+ * and/or sell copies of the Software, and to permit persons to whom the
+ * Software is furnished to do so, subject to the following conditions:
+ *
+ * The above copyright notice and this permission notice (including the next
+ * paragraph) shall be included in all copies or substantial portions of the
+ * Software.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+ * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+ * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT.  IN NO EVENT SHALL
+ * THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
+ * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING
+ * FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER
+ * DEALINGS IN THE SOFTWARE.
+ *
+ * Author: Jani Nikula <jani.nikula@intel.com>
+ */
+
+#ifndef _INTEL_DSI_DSI_H
+#define _INTEL_DSI_DSI_H
+
+#include <drm/drmP.h>
+#include <drm/drm_crtc.h>
+#include <video/mipi_display.h>
+#include "i915_drv.h"
+#include "intel_drv.h"
+#include "intel_dsi.h"
+
+void dsi_hs_mode_enable(struct intel_dsi *intel_dsi, bool enable);
+
+int dsi_vc_dcs_write(struct intel_dsi *intel_dsi, int channel,
+		     const u8 *data, int len);
+
+int dsi_vc_generic_write(struct intel_dsi *intel_dsi, int channel,
+			 const u8 *data, int len);
+
+int dsi_vc_dcs_read(struct intel_dsi *intel_dsi, int channel, u8 dcs_cmd,
+		    u8 *buf, int buflen);
+
+int dsi_vc_generic_read(struct intel_dsi *intel_dsi, int channel,
+			u8 *reqdata, int reqlen, u8 *buf, int buflen);
+
+int dpi_send_cmd(struct intel_dsi *intel_dsi, u32 cmd);
+
+/* XXX: questionable write helpers */
+static inline int dsi_vc_dcs_write_0(struct intel_dsi *intel_dsi,
+				     int channel, u8 dcs_cmd)
+{
+	return dsi_vc_dcs_write(intel_dsi, channel, &dcs_cmd, 1);
+}
+
+static inline int dsi_vc_dcs_write_1(struct intel_dsi *intel_dsi,
+				     int channel, u8 dcs_cmd, u8 param)
+{
+	u8 buf[2] = { dcs_cmd, param };
+	return dsi_vc_dcs_write(intel_dsi, channel, buf, 2);
+}
+
+static inline int dsi_vc_generic_write_0(struct intel_dsi *intel_dsi,
+					 int channel)
+{
+	return dsi_vc_generic_write(intel_dsi, channel, NULL, 0);
+}
+
+static inline int dsi_vc_generic_write_1(struct intel_dsi *intel_dsi,
+					 int channel, u8 param)
+{
+	return dsi_vc_generic_write(intel_dsi, channel, &param, 1);
+}
+
+static inline int dsi_vc_generic_write_2(struct intel_dsi *intel_dsi,
+					 int channel, u8 param1, u8 param2)
+{
+	u8 buf[2] = { param1, param2 };
+	return dsi_vc_generic_write(intel_dsi, channel, buf, 2);
+}
+
+/* XXX: questionable read helpers */
+static inline int dsi_vc_generic_read_0(struct intel_dsi *intel_dsi,
+					int channel, u8 *buf, int buflen)
+{
+	return dsi_vc_generic_read(intel_dsi, channel, NULL, 0, buf, buflen);
+}
+
+static inline int dsi_vc_generic_read_1(struct intel_dsi *intel_dsi,
+					int channel, u8 param, u8 *buf,
+					int buflen)
+{
+	return dsi_vc_generic_read(intel_dsi, channel, &param, 1, buf, buflen);
+}
+
+static inline int dsi_vc_generic_read_2(struct intel_dsi *intel_dsi,
+					int channel, u8 param1, u8 param2,
+					u8 *buf, int buflen)
+{
+	u8 req[2] = { param1, param2 };
+
+	return dsi_vc_generic_read(intel_dsi, channel, req, 2, buf, buflen);
+}
+
+
+#endif /* _INTEL_DSI_DSI_H */
diff --git a/drivers/gpu/drm/i915/intel_dsi_pll.c b/drivers/gpu/drm/i915/intel_dsi_pll.c
new file mode 100644
index 000000000000..44279b2ade88
--- /dev/null
+++ b/drivers/gpu/drm/i915/intel_dsi_pll.c
@@ -0,0 +1,317 @@
+/*
+ * Copyright © 2013 Intel Corporation
+ *
+ * Permission is hereby granted, free of charge, to any person obtaining a
+ * copy of this software and associated documentation files (the "Software"),
+ * to deal in the Software without restriction, including without limitation
+ * the rights to use, copy, modify, merge, publish, distribute, sublicense,
+ * and/or sell copies of the Software, and to permit persons to whom the
+ * Software is furnished to do so, subject to the following conditions:
+ *
+ * The above copyright notice and this permission notice (including the next
+ * paragraph) shall be included in all copies or substantial portions of the
+ * Software.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+ * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+ * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT.  IN NO EVENT SHALL
+ * THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
+ * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING
+ * FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER
+ * DEALINGS IN THE SOFTWARE.
+ *
+ * Authors:
+ *	Shobhit Kumar <shobhit.kumar@intel.com>
+ *	Yogesh Mohan Marimuthu <yogesh.mohan.marimuthu@intel.com>
+ */
+
+#include <linux/kernel.h>
+#include "intel_drv.h"
+#include "i915_drv.h"
+#include "intel_dsi.h"
+
+#define DSI_HSS_PACKET_SIZE		4
+#define DSI_HSE_PACKET_SIZE		4
+#define DSI_HSA_PACKET_EXTRA_SIZE	6
+#define DSI_HBP_PACKET_EXTRA_SIZE	6
+#define DSI_HACTIVE_PACKET_EXTRA_SIZE	6
+#define DSI_HFP_PACKET_EXTRA_SIZE	6
+#define DSI_EOTP_PACKET_SIZE		4
+
+struct dsi_mnp {
+	u32 dsi_pll_ctrl;
+	u32 dsi_pll_div;
+};
+
+static const u32 lfsr_converts[] = {
+	426, 469, 234, 373, 442, 221, 110, 311, 411,		/* 62 - 70 */
+	461, 486, 243, 377, 188, 350, 175, 343, 427, 213,	/* 71 - 80 */
+	106, 53, 282, 397, 354, 227, 113, 56, 284, 142,		/* 81 - 90 */
+	71, 35							/* 91 - 92 */
+};
+
+static u32 dsi_rr_formula(const struct drm_display_mode *mode,
+			  int pixel_format, int video_mode_format,
+			  int lane_count, bool eotp)
+{
+	u32 bpp;
+	u32 hactive, vactive, hfp, hsync, hbp, vfp, vsync, vbp;
+	u32 hsync_bytes, hbp_bytes, hactive_bytes, hfp_bytes;
+	u32 bytes_per_line, bytes_per_frame;
+	u32 num_frames;
+	u32 bytes_per_x_frames, bytes_per_x_frames_x_lanes;
+	u32 dsi_bit_clock_hz;
+	u32 dsi_clk;
+
+	switch (pixel_format) {
+	default:
+	case VID_MODE_FORMAT_RGB888:
+	case VID_MODE_FORMAT_RGB666_LOOSE:
+		bpp = 24;
+		break;
+	case VID_MODE_FORMAT_RGB666:
+		bpp = 18;
+		break;
+	case VID_MODE_FORMAT_RGB565:
+		bpp = 16;
+		break;
+	}
+
+	hactive = mode->hdisplay;
+	vactive = mode->vdisplay;
+	hfp = mode->hsync_start - mode->hdisplay;
+	hsync = mode->hsync_end - mode->hsync_start;
+	hbp = mode->htotal - mode->hsync_end;
+
+	vfp = mode->vsync_start - mode->vdisplay;
+	vsync = mode->vsync_end - mode->vsync_start;
+	vbp = mode->vtotal - mode->vsync_end;
+
+	hsync_bytes = DIV_ROUND_UP(hsync * bpp, 8);
+	hbp_bytes = DIV_ROUND_UP(hbp * bpp, 8);
+	hactive_bytes = DIV_ROUND_UP(hactive * bpp, 8);
+	hfp_bytes = DIV_ROUND_UP(hfp * bpp, 8);
+
+	bytes_per_line = DSI_HSS_PACKET_SIZE + hsync_bytes +
+		DSI_HSA_PACKET_EXTRA_SIZE + DSI_HSE_PACKET_SIZE +
+		hbp_bytes + DSI_HBP_PACKET_EXTRA_SIZE +
+		hactive_bytes + DSI_HACTIVE_PACKET_EXTRA_SIZE +
+		hfp_bytes + DSI_HFP_PACKET_EXTRA_SIZE;
+
+	/*
+	 * XXX: Need to accurately calculate LP to HS transition timeout and add
+	 * it to bytes_per_line/bytes_per_frame.
+	 */
+
+	if (eotp && video_mode_format == VIDEO_MODE_BURST)
+		bytes_per_line += DSI_EOTP_PACKET_SIZE;
+
+	bytes_per_frame = vsync * bytes_per_line + vbp * bytes_per_line +
+		vactive * bytes_per_line + vfp * bytes_per_line;
+
+	if (eotp &&
+	    (video_mode_format == VIDEO_MODE_NON_BURST_WITH_SYNC_PULSE ||
+	     video_mode_format == VIDEO_MODE_NON_BURST_WITH_SYNC_EVENTS))
+		bytes_per_frame += DSI_EOTP_PACKET_SIZE;
+
+	num_frames = drm_mode_vrefresh(mode);
+	bytes_per_x_frames = num_frames * bytes_per_frame;
+
+	bytes_per_x_frames_x_lanes = bytes_per_x_frames / lane_count;
+
+	/* the dsi clock is divided by 2 in the hardware to get dsi ddr clock */
+	dsi_bit_clock_hz = bytes_per_x_frames_x_lanes * 8;
+	dsi_clk = dsi_bit_clock_hz / (1000 * 1000);
+
+	if (eotp && video_mode_format == VIDEO_MODE_BURST)
+		dsi_clk *= 2;
+
+	return dsi_clk;
+}
+
+#ifdef MNP_FROM_TABLE
+
+struct dsi_clock_table {
+	u32 freq;
+	u8 m;
+	u8 p;
+};
+
+static const struct dsi_clock_table dsi_clk_tbl[] = {
+	{300, 72, 6}, {313, 75, 6}, {323, 78, 6}, {333, 80, 6},
+	{343, 82, 6}, {353, 85, 6}, {363, 87, 6}, {373, 90, 6},
+	{383, 92, 6}, {390, 78, 5}, {393, 79, 5}, {400, 80, 5},
+	{401, 80, 5}, {402, 80, 5}, {403, 81, 5}, {404, 81, 5},
+	{405, 81, 5}, {406, 81, 5}, {407, 81, 5}, {408, 82, 5},
+	{409, 82, 5}, {410, 82, 5}, {411, 82, 5}, {412, 82, 5},
+	{413, 83, 5}, {414, 83, 5}, {415, 83, 5}, {416, 83, 5},
+	{417, 83, 5}, {418, 84, 5}, {419, 84, 5}, {420, 84, 5},
+	{430, 86, 5}, {440, 88, 5}, {450, 90, 5}, {460, 92, 5},
+	{470, 75, 4}, {480, 77, 4}, {490, 78, 4}, {500, 80, 4},
+	{510, 82, 4}, {520, 83, 4}, {530, 85, 4}, {540, 86, 4},
+	{550, 88, 4}, {560, 90, 4}, {570, 91, 4}, {580, 70, 3},
+	{590, 71, 3}, {600, 72, 3}, {610, 73, 3}, {620, 74, 3},
+	{630, 76, 3}, {640, 77, 3}, {650, 78, 3}, {660, 79, 3},
+	{670, 80, 3}, {680, 82, 3}, {690, 83, 3}, {700, 84, 3},
+	{710, 85, 3}, {720, 86, 3}, {730, 88, 3}, {740, 89, 3},
+	{750, 90, 3}, {760, 91, 3}, {770, 92, 3}, {780, 62, 2},
+	{790, 63, 2}, {800, 64, 2}, {880, 70, 2}, {900, 72, 2},
+	{1000, 80, 2},		/* dsi clock frequency in Mhz*/
+};
+
+static int dsi_calc_mnp(u32 dsi_clk, struct dsi_mnp *dsi_mnp)
+{
+	unsigned int i;
+	u8 m;
+	u8 n;
+	u8 p;
+	u32 m_seed;
+
+	if (dsi_clk < 300 || dsi_clk > 1000)
+		return -ECHRNG;
+
+	for (i = 0; i <= ARRAY_SIZE(dsi_clk_tbl); i++) {
+		if (dsi_clk_tbl[i].freq > dsi_clk)
+			break;
+	}
+
+	m = dsi_clk_tbl[i].m;
+	p = dsi_clk_tbl[i].p;
+	m_seed = lfsr_converts[m - 62];
+	n = 1;
+	dsi_mnp->dsi_pll_ctrl = 1 << (DSI_PLL_P1_POST_DIV_SHIFT + p - 2);
+	dsi_mnp->dsi_pll_div = (n - 1) << DSI_PLL_N1_DIV_SHIFT |
+		m_seed << DSI_PLL_M1_DIV_SHIFT;
+
+	return 0;
+}
+
+#else
+
+static int dsi_calc_mnp(u32 dsi_clk, struct dsi_mnp *dsi_mnp)
+{
+	u32 m, n, p;
+	u32 ref_clk;
+	u32 error;
+	u32 tmp_error;
+	u32 target_dsi_clk;
+	u32 calc_dsi_clk;
+	u32 calc_m;
+	u32 calc_p;
+	u32 m_seed;
+
+	if (dsi_clk < 300 || dsi_clk > 1150) {
+		DRM_ERROR("DSI CLK Out of Range\n");
+		return -ECHRNG;
+	}
+
+	ref_clk = 25000;
+	target_dsi_clk = dsi_clk * 1000;
+	error = 0xFFFFFFFF;
+	calc_m = 0;
+	calc_p = 0;
+
+	for (m = 62; m <= 92; m++) {
+		for (p = 2; p <= 6; p++) {
+
+			calc_dsi_clk = (m * ref_clk) / p;
+			if (calc_dsi_clk >= target_dsi_clk) {
+				tmp_error = calc_dsi_clk - target_dsi_clk;
+				if (tmp_error < error) {
+					error = tmp_error;
+					calc_m = m;
+					calc_p = p;
+				}
+			}
+		}
+	}
+
+	m_seed = lfsr_converts[calc_m - 62];
+	n = 1;
+	dsi_mnp->dsi_pll_ctrl = 1 << (DSI_PLL_P1_POST_DIV_SHIFT + calc_p - 2);
+	dsi_mnp->dsi_pll_div = (n - 1) << DSI_PLL_N1_DIV_SHIFT |
+		m_seed << DSI_PLL_M1_DIV_SHIFT;
+
+	return 0;
+}
+
+#endif
+
+/*
+ * XXX: The muxing and gating is hard coded for now. Need to add support for
+ * sharing PLLs with two DSI outputs.
+ */
+static void vlv_configure_dsi_pll(struct intel_encoder *encoder)
+{
+	struct drm_i915_private *dev_priv = encoder->base.dev->dev_private;
+	struct intel_crtc *intel_crtc = to_intel_crtc(encoder->base.crtc);
+	const struct drm_display_mode *mode = &intel_crtc->config.adjusted_mode;
+	struct intel_dsi *intel_dsi = enc_to_intel_dsi(&encoder->base);
+	int ret;
+	struct dsi_mnp dsi_mnp;
+	u32 dsi_clk;
+
+	dsi_clk = dsi_rr_formula(mode, intel_dsi->pixel_format,
+				 intel_dsi->video_mode_format,
+				 intel_dsi->lane_count, !intel_dsi->eot_disable);
+
+	ret = dsi_calc_mnp(dsi_clk, &dsi_mnp);
+	if (ret) {
+		DRM_DEBUG_KMS("dsi_calc_mnp failed\n");
+		return;
+	}
+
+	dsi_mnp.dsi_pll_ctrl |= DSI_PLL_CLK_GATE_DSI0_DSIPLL;
+
+	DRM_DEBUG_KMS("dsi pll div %08x, ctrl %08x\n",
+		      dsi_mnp.dsi_pll_div, dsi_mnp.dsi_pll_ctrl);
+
+	vlv_cck_write(dev_priv, CCK_REG_DSI_PLL_CONTROL, 0);
+	vlv_cck_write(dev_priv, CCK_REG_DSI_PLL_DIVIDER, dsi_mnp.dsi_pll_div);
+	vlv_cck_write(dev_priv, CCK_REG_DSI_PLL_CONTROL, dsi_mnp.dsi_pll_ctrl);
+}
+
+void vlv_enable_dsi_pll(struct intel_encoder *encoder)
+{
+	struct drm_i915_private *dev_priv = encoder->base.dev->dev_private;
+	u32 tmp;
+
+	DRM_DEBUG_KMS("\n");
+
+	mutex_lock(&dev_priv->dpio_lock);
+
+	vlv_configure_dsi_pll(encoder);
+
+	/* wait at least 0.5 us after ungating before enabling VCO */
+	usleep_range(1, 10);
+
+	tmp = vlv_cck_read(dev_priv, CCK_REG_DSI_PLL_CONTROL);
+	tmp |= DSI_PLL_VCO_EN;
+	vlv_cck_write(dev_priv, CCK_REG_DSI_PLL_CONTROL, tmp);
+
+	mutex_unlock(&dev_priv->dpio_lock);
+
+	if (wait_for(I915_READ(PIPECONF(PIPE_A)) & PIPECONF_DSI_PLL_LOCKED, 20)) {
+		DRM_ERROR("DSI PLL lock failed\n");
+		return;
+	}
+
+	DRM_DEBUG_KMS("DSI PLL locked\n");
+}
+
+void vlv_disable_dsi_pll(struct intel_encoder *encoder)
+{
+	struct drm_i915_private *dev_priv = encoder->base.dev->dev_private;
+	u32 tmp;
+
+	DRM_DEBUG_KMS("\n");
+
+	mutex_lock(&dev_priv->dpio_lock);
+
+	tmp = vlv_cck_read(dev_priv, CCK_REG_DSI_PLL_CONTROL);
+	tmp &= ~DSI_PLL_VCO_EN;
+	tmp |= DSI_PLL_LDO_GATE;
+	vlv_cck_write(dev_priv, CCK_REG_DSI_PLL_CONTROL, tmp);
+
+	mutex_unlock(&dev_priv->dpio_lock);
+}
diff --git a/drivers/gpu/drm/i915/intel_dvo.c b/drivers/gpu/drm/i915/intel_dvo.c
index 7fa7df546c1e..3c7736546856 100644
--- a/drivers/gpu/drm/i915/intel_dvo.c
+++ b/drivers/gpu/drm/i915/intel_dvo.c
@@ -153,6 +153,8 @@ static void intel_dvo_get_config(struct intel_encoder *encoder,
 		flags |= DRM_MODE_FLAG_NVSYNC;
 
 	pipe_config->adjusted_mode.flags |= flags;
+
+	pipe_config->adjusted_mode.crtc_clock = pipe_config->port_clock;
 }
 
 static void intel_disable_dvo(struct intel_encoder *encoder)
@@ -171,11 +173,16 @@ static void intel_enable_dvo(struct intel_encoder *encoder)
 {
 	struct drm_i915_private *dev_priv = encoder->base.dev->dev_private;
 	struct intel_dvo *intel_dvo = enc_to_dvo(encoder);
+	struct intel_crtc *crtc = to_intel_crtc(encoder->base.crtc);
 	u32 dvo_reg = intel_dvo->dev.dvo_reg;
 	u32 temp = I915_READ(dvo_reg);
 
 	I915_WRITE(dvo_reg, temp | DVO_ENABLE);
 	I915_READ(dvo_reg);
+	intel_dvo->dev.dev_ops->mode_set(&intel_dvo->dev,
+					 &crtc->config.requested_mode,
+					 &crtc->config.adjusted_mode);
+
 	intel_dvo->dev.dev_ops->dpms(&intel_dvo->dev, true);
 }
 
@@ -184,6 +191,7 @@ static void intel_dvo_dpms(struct drm_connector *connector, int mode)
 {
 	struct intel_dvo *intel_dvo = intel_attached_dvo(connector);
 	struct drm_crtc *crtc;
+	struct intel_crtc_config *config;
 
 	/* dvo supports only 2 dpms states. */
 	if (mode != DRM_MODE_DPMS_ON)
@@ -204,10 +212,16 @@ static void intel_dvo_dpms(struct drm_connector *connector, int mode)
 	/* We call connector dpms manually below in case pipe dpms doesn't
 	 * change due to cloning. */
 	if (mode == DRM_MODE_DPMS_ON) {
+		config = &to_intel_crtc(crtc)->config;
+
 		intel_dvo->base.connectors_active = true;
 
 		intel_crtc_update_dpms(crtc);
 
+		intel_dvo->dev.dev_ops->mode_set(&intel_dvo->dev,
+						 &config->requested_mode,
+						 &config->adjusted_mode);
+
 		intel_dvo->dev.dev_ops->dpms(&intel_dvo->dev, true);
 	} else {
 		intel_dvo->dev.dev_ops->dpms(&intel_dvo->dev, false);
@@ -267,11 +281,6 @@ static bool intel_dvo_compute_config(struct intel_encoder *encoder,
 		drm_mode_set_crtcinfo(adjusted_mode, 0);
 	}
 
-	if (intel_dvo->dev.dev_ops->mode_fixup)
-		return intel_dvo->dev.dev_ops->mode_fixup(&intel_dvo->dev,
-							  &pipe_config->requested_mode,
-							  adjusted_mode);
-
 	return true;
 }
 
@@ -299,10 +308,6 @@ static void intel_dvo_mode_set(struct intel_encoder *encoder)
 		break;
 	}
 
-	intel_dvo->dev.dev_ops->mode_set(&intel_dvo->dev,
-					 &crtc->config.requested_mode,
-					 adjusted_mode);
-
 	/* Save the data order, since I don't know what it should be set to. */
 	dvo_val = I915_READ(dvo_reg) &
 		  (DVO_PRESERVE_MASK | DVO_DATA_ORDER_GBRG);
@@ -370,7 +375,6 @@ static int intel_dvo_get_modes(struct drm_connector *connector)
 
 static void intel_dvo_destroy(struct drm_connector *connector)
 {
-	drm_sysfs_connector_remove(connector);
 	drm_connector_cleanup(connector);
 	kfree(connector);
 }
@@ -451,11 +455,11 @@ void intel_dvo_init(struct drm_device *dev)
 	int i;
 	int encoder_type = DRM_MODE_ENCODER_NONE;
 
-	intel_dvo = kzalloc(sizeof(struct intel_dvo), GFP_KERNEL);
+	intel_dvo = kzalloc(sizeof(*intel_dvo), GFP_KERNEL);
 	if (!intel_dvo)
 		return;
 
-	intel_connector = kzalloc(sizeof(struct intel_connector), GFP_KERNEL);
+	intel_connector = kzalloc(sizeof(*intel_connector), GFP_KERNEL);
 	if (!intel_connector) {
 		kfree(intel_dvo);
 		return;
diff --git a/drivers/gpu/drm/i915/intel_fb.c b/drivers/gpu/drm/i915/intel_fbdev.c
index bc2100007b21..895fcb4fbd94 100644
--- a/drivers/gpu/drm/i915/intel_fb.c
+++ b/drivers/gpu/drm/i915/intel_fbdev.c
@@ -78,8 +78,8 @@ static int intelfb_create(struct drm_fb_helper *helper,
 	mode_cmd.width = sizes->surface_width;
 	mode_cmd.height = sizes->surface_height;
 
-	mode_cmd.pitches[0] = ALIGN(mode_cmd.width * ((sizes->surface_bpp + 7) /
-						      8), 64);
+	mode_cmd.pitches[0] = ALIGN(mode_cmd.width *
+				    DIV_ROUND_UP(sizes->surface_bpp, 8), 64);
 	mode_cmd.pixel_format = drm_mode_legacy_fb_format(sizes->surface_bpp,
 							  sizes->surface_depth);
 
@@ -184,6 +184,27 @@ out:
 	return ret;
 }
 
+/** Sets the color ramps on behalf of RandR */
+static void intel_crtc_fb_gamma_set(struct drm_crtc *crtc, u16 red, u16 green,
+				    u16 blue, int regno)
+{
+	struct intel_crtc *intel_crtc = to_intel_crtc(crtc);
+
+	intel_crtc->lut_r[regno] = red >> 8;
+	intel_crtc->lut_g[regno] = green >> 8;
+	intel_crtc->lut_b[regno] = blue >> 8;
+}
+
+static void intel_crtc_fb_gamma_get(struct drm_crtc *crtc, u16 *red, u16 *green,
+				    u16 *blue, int regno)
+{
+	struct intel_crtc *intel_crtc = to_intel_crtc(crtc);
+
+	*red = intel_crtc->lut_r[regno] << 8;
+	*green = intel_crtc->lut_g[regno] << 8;
+	*blue = intel_crtc->lut_b[regno] << 8;
+}
+
 static struct drm_fb_helper_funcs intel_fb_helper_funcs = {
 	.gamma_set = intel_crtc_fb_gamma_set,
 	.gamma_get = intel_crtc_fb_gamma_get,
@@ -216,7 +237,7 @@ int intel_fbdev_init(struct drm_device *dev)
 	struct drm_i915_private *dev_priv = dev->dev_private;
 	int ret;
 
-	ifbdev = kzalloc(sizeof(struct intel_fbdev), GFP_KERNEL);
+	ifbdev = kzalloc(sizeof(*ifbdev), GFP_KERNEL);
 	if (!ifbdev)
 		return -ENOMEM;
 
@@ -225,7 +246,7 @@ int intel_fbdev_init(struct drm_device *dev)
 
 	ret = drm_fb_helper_init(dev, &ifbdev->helper,
 				 INTEL_INFO(dev)->num_pipes,
-				 INTELFB_CONN_LIMIT);
+				 4);
 	if (ret) {
 		kfree(ifbdev);
 		return ret;
@@ -278,13 +299,13 @@ void intel_fbdev_set_suspend(struct drm_device *dev, int state)
 
 MODULE_LICENSE("GPL and additional rights");
 
-void intel_fb_output_poll_changed(struct drm_device *dev)
+void intel_fbdev_output_poll_changed(struct drm_device *dev)
 {
 	struct drm_i915_private *dev_priv = dev->dev_private;
 	drm_fb_helper_hotplug_event(&dev_priv->fbdev->helper);
 }
 
-void intel_fb_restore_mode(struct drm_device *dev)
+void intel_fbdev_restore_mode(struct drm_device *dev)
 {
 	int ret;
 	struct drm_i915_private *dev_priv = dev->dev_private;
diff --git a/drivers/gpu/drm/i915/intel_hdmi.c b/drivers/gpu/drm/i915/intel_hdmi.c
index 4148cc85bf7f..03f9ca70530c 100644
--- a/drivers/gpu/drm/i915/intel_hdmi.c
+++ b/drivers/gpu/drm/i915/intel_hdmi.c
@@ -713,6 +713,7 @@ static void intel_hdmi_get_config(struct intel_encoder *encoder,
 	struct intel_hdmi *intel_hdmi = enc_to_intel_hdmi(&encoder->base);
 	struct drm_i915_private *dev_priv = encoder->base.dev->dev_private;
 	u32 tmp, flags = 0;
+	int dotclock;
 
 	tmp = I915_READ(intel_hdmi->hdmi_reg);
 
@@ -727,6 +728,16 @@ static void intel_hdmi_get_config(struct intel_encoder *encoder,
 		flags |= DRM_MODE_FLAG_NVSYNC;
 
 	pipe_config->adjusted_mode.flags |= flags;
+
+	if ((tmp & SDVO_COLOR_FORMAT_MASK) == HDMI_COLOR_FORMAT_12bpc)
+		dotclock = pipe_config->port_clock * 2 / 3;
+	else
+		dotclock = pipe_config->port_clock;
+
+	if (HAS_PCH_SPLIT(dev_priv->dev))
+		ironlake_check_encoder_dotclock(pipe_config, dotclock);
+
+	pipe_config->adjusted_mode.crtc_clock = dotclock;
 }
 
 static void intel_enable_hdmi(struct intel_encoder *encoder)
@@ -836,7 +847,7 @@ static int hdmi_portclock_limit(struct intel_hdmi *hdmi)
 
 	if (IS_G4X(dev))
 		return 165000;
-	else if (IS_HASWELL(dev))
+	else if (IS_HASWELL(dev) || INTEL_INFO(dev)->gen >= 8)
 		return 300000;
 	else
 		return 225000;
@@ -862,7 +873,7 @@ bool intel_hdmi_compute_config(struct intel_encoder *encoder,
 	struct intel_hdmi *intel_hdmi = enc_to_intel_hdmi(&encoder->base);
 	struct drm_device *dev = encoder->base.dev;
 	struct drm_display_mode *adjusted_mode = &pipe_config->adjusted_mode;
-	int clock_12bpc = pipe_config->requested_mode.clock * 3 / 2;
+	int clock_12bpc = pipe_config->adjusted_mode.crtc_clock * 3 / 2;
 	int portclock_limit = hdmi_portclock_limit(intel_hdmi);
 	int desired_bpp;
 
@@ -904,7 +915,7 @@ bool intel_hdmi_compute_config(struct intel_encoder *encoder,
 		pipe_config->pipe_bpp = desired_bpp;
 	}
 
-	if (adjusted_mode->clock > portclock_limit) {
+	if (adjusted_mode->crtc_clock > portclock_limit) {
 		DRM_DEBUG_KMS("too high HDMI clock, rejecting mode\n");
 		return false;
 	}
@@ -1063,7 +1074,7 @@ done:
 	return 0;
 }
 
-static void intel_hdmi_pre_enable(struct intel_encoder *encoder)
+static void vlv_hdmi_pre_enable(struct intel_encoder *encoder)
 {
 	struct intel_digital_port *dport = enc_to_dig_port(&encoder->base);
 	struct drm_device *dev = encoder->base.dev;
@@ -1079,35 +1090,35 @@ static void intel_hdmi_pre_enable(struct intel_encoder *encoder)
 
 	/* Enable clock channels for this port */
 	mutex_lock(&dev_priv->dpio_lock);
-	val = vlv_dpio_read(dev_priv, DPIO_DATA_LANE_A(port));
+	val = vlv_dpio_read(dev_priv, pipe, DPIO_DATA_LANE_A(port));
 	val = 0;
 	if (pipe)
 		val |= (1<<21);
 	else
 		val &= ~(1<<21);
 	val |= 0x001000c4;
-	vlv_dpio_write(dev_priv, DPIO_DATA_CHANNEL(port), val);
+	vlv_dpio_write(dev_priv, pipe, DPIO_DATA_CHANNEL(port), val);
 
 	/* HDMI 1.0V-2dB */
-	vlv_dpio_write(dev_priv, DPIO_TX_OCALINIT(port), 0);
-	vlv_dpio_write(dev_priv, DPIO_TX_SWING_CTL4(port),
+	vlv_dpio_write(dev_priv, pipe, DPIO_TX_OCALINIT(port), 0);
+	vlv_dpio_write(dev_priv, pipe, DPIO_TX_SWING_CTL4(port),
 			 0x2b245f5f);
-	vlv_dpio_write(dev_priv, DPIO_TX_SWING_CTL2(port),
+	vlv_dpio_write(dev_priv, pipe, DPIO_TX_SWING_CTL2(port),
 			 0x5578b83a);
-	vlv_dpio_write(dev_priv, DPIO_TX_SWING_CTL3(port),
+	vlv_dpio_write(dev_priv, pipe, DPIO_TX_SWING_CTL3(port),
 			 0x0c782040);
-	vlv_dpio_write(dev_priv, DPIO_TX3_SWING_CTL4(port),
+	vlv_dpio_write(dev_priv, pipe, DPIO_TX3_SWING_CTL4(port),
 			 0x2b247878);
-	vlv_dpio_write(dev_priv, DPIO_PCS_STAGGER0(port), 0x00030000);
-	vlv_dpio_write(dev_priv, DPIO_PCS_CTL_OVER1(port),
+	vlv_dpio_write(dev_priv, pipe, DPIO_PCS_STAGGER0(port), 0x00030000);
+	vlv_dpio_write(dev_priv, pipe, DPIO_PCS_CTL_OVER1(port),
 			 0x00002000);
-	vlv_dpio_write(dev_priv, DPIO_TX_OCALINIT(port),
+	vlv_dpio_write(dev_priv, pipe, DPIO_TX_OCALINIT(port),
 			 DPIO_TX_OCALINIT_EN);
 
 	/* Program lane clock */
-	vlv_dpio_write(dev_priv, DPIO_PCS_CLOCKBUF0(port),
+	vlv_dpio_write(dev_priv, pipe, DPIO_PCS_CLOCKBUF0(port),
 			 0x00760018);
-	vlv_dpio_write(dev_priv, DPIO_PCS_CLOCKBUF8(port),
+	vlv_dpio_write(dev_priv, pipe, DPIO_PCS_CLOCKBUF8(port),
 			 0x00400888);
 	mutex_unlock(&dev_priv->dpio_lock);
 
@@ -1116,55 +1127,60 @@ static void intel_hdmi_pre_enable(struct intel_encoder *encoder)
 	vlv_wait_port_ready(dev_priv, port);
 }
 
-static void intel_hdmi_pre_pll_enable(struct intel_encoder *encoder)
+static void vlv_hdmi_pre_pll_enable(struct intel_encoder *encoder)
 {
 	struct intel_digital_port *dport = enc_to_dig_port(&encoder->base);
 	struct drm_device *dev = encoder->base.dev;
 	struct drm_i915_private *dev_priv = dev->dev_private;
+	struct intel_crtc *intel_crtc =
+		to_intel_crtc(encoder->base.crtc);
 	int port = vlv_dport_to_channel(dport);
+	int pipe = intel_crtc->pipe;
 
 	if (!IS_VALLEYVIEW(dev))
 		return;
 
 	/* Program Tx lane resets to default */
 	mutex_lock(&dev_priv->dpio_lock);
-	vlv_dpio_write(dev_priv, DPIO_PCS_TX(port),
+	vlv_dpio_write(dev_priv, pipe, DPIO_PCS_TX(port),
 			 DPIO_PCS_TX_LANE2_RESET |
 			 DPIO_PCS_TX_LANE1_RESET);
-	vlv_dpio_write(dev_priv, DPIO_PCS_CLK(port),
+	vlv_dpio_write(dev_priv, pipe, DPIO_PCS_CLK(port),
 			 DPIO_PCS_CLK_CRI_RXEB_EIOS_EN |
 			 DPIO_PCS_CLK_CRI_RXDIGFILTSG_EN |
 			 (1<<DPIO_PCS_CLK_DATAWIDTH_SHIFT) |
 			 DPIO_PCS_CLK_SOFT_RESET);
 
 	/* Fix up inter-pair skew failure */
-	vlv_dpio_write(dev_priv, DPIO_PCS_STAGGER1(port), 0x00750f00);
-	vlv_dpio_write(dev_priv, DPIO_TX_CTL(port), 0x00001500);
-	vlv_dpio_write(dev_priv, DPIO_TX_LANE(port), 0x40400000);
+	vlv_dpio_write(dev_priv, pipe, DPIO_PCS_STAGGER1(port), 0x00750f00);
+	vlv_dpio_write(dev_priv, pipe, DPIO_TX_CTL(port), 0x00001500);
+	vlv_dpio_write(dev_priv, pipe, DPIO_TX_LANE(port), 0x40400000);
 
-	vlv_dpio_write(dev_priv, DPIO_PCS_CTL_OVER1(port),
+	vlv_dpio_write(dev_priv, pipe, DPIO_PCS_CTL_OVER1(port),
 			 0x00002000);
-	vlv_dpio_write(dev_priv, DPIO_TX_OCALINIT(port),
+	vlv_dpio_write(dev_priv, pipe, DPIO_TX_OCALINIT(port),
 			 DPIO_TX_OCALINIT_EN);
 	mutex_unlock(&dev_priv->dpio_lock);
 }
 
-static void intel_hdmi_post_disable(struct intel_encoder *encoder)
+static void vlv_hdmi_post_disable(struct intel_encoder *encoder)
 {
 	struct intel_digital_port *dport = enc_to_dig_port(&encoder->base);
 	struct drm_i915_private *dev_priv = encoder->base.dev->dev_private;
+	struct intel_crtc *intel_crtc =
+		to_intel_crtc(encoder->base.crtc);
 	int port = vlv_dport_to_channel(dport);
+	int pipe = intel_crtc->pipe;
 
 	/* Reset lanes to avoid HDMI flicker (VLV w/a) */
 	mutex_lock(&dev_priv->dpio_lock);
-	vlv_dpio_write(dev_priv, DPIO_PCS_TX(port), 0x00000000);
-	vlv_dpio_write(dev_priv, DPIO_PCS_CLK(port), 0x00e00060);
+	vlv_dpio_write(dev_priv, pipe, DPIO_PCS_TX(port), 0x00000000);
+	vlv_dpio_write(dev_priv, pipe, DPIO_PCS_CLK(port), 0x00e00060);
 	mutex_unlock(&dev_priv->dpio_lock);
 }
 
 static void intel_hdmi_destroy(struct drm_connector *connector)
 {
-	drm_sysfs_connector_remove(connector);
 	drm_connector_cleanup(connector);
 	kfree(connector);
 }
@@ -1211,6 +1227,7 @@ void intel_hdmi_init_connector(struct intel_digital_port *intel_dig_port,
 
 	connector->interlace_allowed = 1;
 	connector->doublescan_allowed = 0;
+	connector->stereo_allowed = 1;
 
 	switch (port) {
 	case PORT_B:
@@ -1275,11 +1292,11 @@ void intel_hdmi_init(struct drm_device *dev, int hdmi_reg, enum port port)
 	struct intel_encoder *intel_encoder;
 	struct intel_connector *intel_connector;
 
-	intel_dig_port = kzalloc(sizeof(struct intel_digital_port), GFP_KERNEL);
+	intel_dig_port = kzalloc(sizeof(*intel_dig_port), GFP_KERNEL);
 	if (!intel_dig_port)
 		return;
 
-	intel_connector = kzalloc(sizeof(struct intel_connector), GFP_KERNEL);
+	intel_connector = kzalloc(sizeof(*intel_connector), GFP_KERNEL);
 	if (!intel_connector) {
 		kfree(intel_dig_port);
 		return;
@@ -1296,10 +1313,10 @@ void intel_hdmi_init(struct drm_device *dev, int hdmi_reg, enum port port)
 	intel_encoder->get_hw_state = intel_hdmi_get_hw_state;
 	intel_encoder->get_config = intel_hdmi_get_config;
 	if (IS_VALLEYVIEW(dev)) {
-		intel_encoder->pre_pll_enable = intel_hdmi_pre_pll_enable;
-		intel_encoder->pre_enable = intel_hdmi_pre_enable;
+		intel_encoder->pre_pll_enable = vlv_hdmi_pre_pll_enable;
+		intel_encoder->pre_enable = vlv_hdmi_pre_enable;
 		intel_encoder->enable = vlv_enable_hdmi;
-		intel_encoder->post_disable = intel_hdmi_post_disable;
+		intel_encoder->post_disable = vlv_hdmi_post_disable;
 	} else {
 		intel_encoder->enable = intel_enable_hdmi;
 	}
diff --git a/drivers/gpu/drm/i915/intel_i2c.c b/drivers/gpu/drm/i915/intel_i2c.c
index d1c1e0f7f262..2ca17b14b6c1 100644
--- a/drivers/gpu/drm/i915/intel_i2c.c
+++ b/drivers/gpu/drm/i915/intel_i2c.c
@@ -34,6 +34,11 @@
 #include <drm/i915_drm.h>
 #include "i915_drv.h"
 
+enum disp_clk {
+	CDCLK,
+	CZCLK
+};
+
 struct gmbus_port {
 	const char *name;
 	int reg;
@@ -58,10 +63,69 @@ to_intel_gmbus(struct i2c_adapter *i2c)
 	return container_of(i2c, struct intel_gmbus, adapter);
 }
 
+static int get_disp_clk_div(struct drm_i915_private *dev_priv,
+			    enum disp_clk clk)
+{
+	u32 reg_val;
+	int clk_ratio;
+
+	reg_val = I915_READ(CZCLK_CDCLK_FREQ_RATIO);
+
+	if (clk == CDCLK)
+		clk_ratio =
+			((reg_val & CDCLK_FREQ_MASK) >> CDCLK_FREQ_SHIFT) + 1;
+	else
+		clk_ratio = (reg_val & CZCLK_FREQ_MASK) + 1;
+
+	return clk_ratio;
+}
+
+static void gmbus_set_freq(struct drm_i915_private *dev_priv)
+{
+	int vco_freq[] = { 800, 1600, 2000, 2400 };
+	int gmbus_freq = 0, cdclk_div, hpll_freq;
+
+	BUG_ON(!IS_VALLEYVIEW(dev_priv->dev));
+
+	/* Skip setting the gmbus freq if BIOS has already programmed it */
+	if (I915_READ(GMBUSFREQ_VLV) != 0xA0)
+		return;
+
+	/* Obtain SKU information */
+	mutex_lock(&dev_priv->dpio_lock);
+	hpll_freq =
+		vlv_cck_read(dev_priv, CCK_FUSE_REG) & CCK_FUSE_HPLL_FREQ_MASK;
+	mutex_unlock(&dev_priv->dpio_lock);
+
+	/* Get the CDCLK divide ratio */
+	cdclk_div = get_disp_clk_div(dev_priv, CDCLK);
+
+	/*
+	 * Program the gmbus_freq based on the cdclk frequency.
+	 * BSpec erroneously claims we should aim for 4MHz, but
+	 * in fact 1MHz is the correct frequency.
+	 */
+	if (cdclk_div)
+		gmbus_freq = (vco_freq[hpll_freq] << 1) / cdclk_div;
+
+	if (WARN_ON(gmbus_freq == 0))
+		return;
+
+	I915_WRITE(GMBUSFREQ_VLV, gmbus_freq);
+}
+
 void
 intel_i2c_reset(struct drm_device *dev)
 {
 	struct drm_i915_private *dev_priv = dev->dev_private;
+
+	/*
+	 * In BIOS-less system, program the correct gmbus frequency
+	 * before reading edid.
+	 */
+	if (IS_VALLEYVIEW(dev))
+		gmbus_set_freq(dev_priv);
+
 	I915_WRITE(dev_priv->gpio_mmio_base + GMBUS0, 0);
 	I915_WRITE(dev_priv->gpio_mmio_base + GMBUS4, 0);
 }
diff --git a/drivers/gpu/drm/i915/intel_lvds.c b/drivers/gpu/drm/i915/intel_lvds.c
index b8af94a5be39..c3b4da7895ed 100644
--- a/drivers/gpu/drm/i915/intel_lvds.c
+++ b/drivers/gpu/drm/i915/intel_lvds.c
@@ -92,6 +92,7 @@ static void intel_lvds_get_config(struct intel_encoder *encoder,
 	struct drm_device *dev = encoder->base.dev;
 	struct drm_i915_private *dev_priv = dev->dev_private;
 	u32 lvds_reg, tmp, flags = 0;
+	int dotclock;
 
 	if (HAS_PCH_SPLIT(dev))
 		lvds_reg = PCH_LVDS;
@@ -116,6 +117,13 @@ static void intel_lvds_get_config(struct intel_encoder *encoder,
 
 		pipe_config->gmch_pfit.control |= tmp & PANEL_8TO6_DITHER_ENABLE;
 	}
+
+	dotclock = pipe_config->port_clock;
+
+	if (HAS_PCH_SPLIT(dev_priv->dev))
+		ironlake_check_encoder_dotclock(pipe_config, dotclock);
+
+	pipe_config->adjusted_mode.crtc_clock = dotclock;
 }
 
 /* The LVDS pin pair needs to be on before the DPLLs are enabled.
@@ -198,7 +206,8 @@ static void intel_enable_lvds(struct intel_encoder *encoder)
 {
 	struct drm_device *dev = encoder->base.dev;
 	struct intel_lvds_encoder *lvds_encoder = to_lvds_encoder(&encoder->base);
-	struct intel_crtc *intel_crtc = to_intel_crtc(encoder->base.crtc);
+	struct intel_connector *intel_connector =
+		&lvds_encoder->attached_connector->base;
 	struct drm_i915_private *dev_priv = dev->dev_private;
 	u32 ctl_reg, stat_reg;
 
@@ -217,13 +226,15 @@ static void intel_enable_lvds(struct intel_encoder *encoder)
 	if (wait_for((I915_READ(stat_reg) & PP_ON) != 0, 1000))
 		DRM_ERROR("timed out waiting for panel to power on\n");
 
-	intel_panel_enable_backlight(dev, intel_crtc->pipe);
+	intel_panel_enable_backlight(intel_connector);
 }
 
 static void intel_disable_lvds(struct intel_encoder *encoder)
 {
 	struct drm_device *dev = encoder->base.dev;
 	struct intel_lvds_encoder *lvds_encoder = to_lvds_encoder(&encoder->base);
+	struct intel_connector *intel_connector =
+		&lvds_encoder->attached_connector->base;
 	struct drm_i915_private *dev_priv = dev->dev_private;
 	u32 ctl_reg, stat_reg;
 
@@ -235,7 +246,7 @@ static void intel_disable_lvds(struct intel_encoder *encoder)
 		stat_reg = PP_STATUS;
 	}
 
-	intel_panel_disable_backlight(dev);
+	intel_panel_disable_backlight(intel_connector);
 
 	I915_WRITE(ctl_reg, I915_READ(ctl_reg) & ~POWER_TARGET_ON);
 	if (wait_for((I915_READ(stat_reg) & PP_ON) == 0, 1000))
@@ -466,7 +477,6 @@ static void intel_lvds_destroy(struct drm_connector *connector)
 
 	intel_panel_fini(&lvds_connector->base.panel);
 
-	drm_sysfs_connector_remove(connector);
 	drm_connector_cleanup(connector);
 	kfree(connector);
 }
@@ -802,7 +812,8 @@ static bool lvds_is_present_in_vbt(struct drm_device *dev,
 		return true;
 
 	for (i = 0; i < dev_priv->vbt.child_dev_num; i++) {
-		struct child_device_config *child = dev_priv->vbt.child_dev + i;
+		union child_device_config *uchild = dev_priv->vbt.child_dev + i;
+		struct old_child_dev_config *child = &uchild->old;
 
 		/* If the device type is not LFP, continue.
 		 * We have to check both the new identifiers as well as the
@@ -956,11 +967,11 @@ void intel_lvds_init(struct drm_device *dev)
 		}
 	}
 
-	lvds_encoder = kzalloc(sizeof(struct intel_lvds_encoder), GFP_KERNEL);
+	lvds_encoder = kzalloc(sizeof(*lvds_encoder), GFP_KERNEL);
 	if (!lvds_encoder)
 		return;
 
-	lvds_connector = kzalloc(sizeof(struct intel_lvds_connector), GFP_KERNEL);
+	lvds_connector = kzalloc(sizeof(*lvds_connector), GFP_KERNEL);
 	if (!lvds_connector) {
 		kfree(lvds_encoder);
 		return;
diff --git a/drivers/gpu/drm/i915/intel_opregion.c b/drivers/gpu/drm/i915/intel_opregion.c
index 8a94b8e15b27..6d69a9bad865 100644
--- a/drivers/gpu/drm/i915/intel_opregion.c
+++ b/drivers/gpu/drm/i915/intel_opregion.c
@@ -36,8 +36,11 @@
 #include "i915_drv.h"
 #include "intel_drv.h"
 
-#define PCI_ASLE 0xe4
-#define PCI_ASLS 0xfc
+#define PCI_ASLE		0xe4
+#define PCI_ASLS		0xfc
+#define PCI_SWSCI		0xe8
+#define PCI_SWSCI_SCISEL	(1 << 15)
+#define PCI_SWSCI_GSSCIE	(1 << 0)
 
 #define OPREGION_HEADER_OFFSET 0
 #define OPREGION_ACPI_OFFSET   0x100
@@ -107,25 +110,38 @@ struct opregion_asle {
 	u32 epfm;       /* enabled panel fitting modes */
 	u8 plut[74];    /* panel LUT and identifier */
 	u32 pfmb;       /* PWM freq and min brightness */
-	u8 rsvd[102];
+	u32 cddv;       /* color correction default values */
+	u32 pcft;       /* power conservation features */
+	u32 srot;       /* supported rotation angles */
+	u32 iuer;       /* IUER events */
+	u8 rsvd[86];
 } __attribute__((packed));
 
 /* Driver readiness indicator */
 #define ASLE_ARDY_READY		(1 << 0)
 #define ASLE_ARDY_NOT_READY	(0 << 0)
 
-/* ASLE irq request bits */
-#define ASLE_SET_ALS_ILLUM     (1 << 0)
-#define ASLE_SET_BACKLIGHT     (1 << 1)
-#define ASLE_SET_PFIT          (1 << 2)
-#define ASLE_SET_PWM_FREQ      (1 << 3)
-#define ASLE_REQ_MSK           0xf
-
-/* response bits of ASLE irq request */
-#define ASLE_ALS_ILLUM_FAILED	(1<<10)
-#define ASLE_BACKLIGHT_FAILED	(1<<12)
-#define ASLE_PFIT_FAILED	(1<<14)
-#define ASLE_PWM_FREQ_FAILED	(1<<16)
+/* ASLE Interrupt Command (ASLC) bits */
+#define ASLC_SET_ALS_ILLUM		(1 << 0)
+#define ASLC_SET_BACKLIGHT		(1 << 1)
+#define ASLC_SET_PFIT			(1 << 2)
+#define ASLC_SET_PWM_FREQ		(1 << 3)
+#define ASLC_SUPPORTED_ROTATION_ANGLES	(1 << 4)
+#define ASLC_BUTTON_ARRAY		(1 << 5)
+#define ASLC_CONVERTIBLE_INDICATOR	(1 << 6)
+#define ASLC_DOCKING_INDICATOR		(1 << 7)
+#define ASLC_ISCT_STATE_CHANGE		(1 << 8)
+#define ASLC_REQ_MSK			0x1ff
+/* response bits */
+#define ASLC_ALS_ILLUM_FAILED		(1 << 10)
+#define ASLC_BACKLIGHT_FAILED		(1 << 12)
+#define ASLC_PFIT_FAILED		(1 << 14)
+#define ASLC_PWM_FREQ_FAILED		(1 << 16)
+#define ASLC_ROTATION_ANGLES_FAILED	(1 << 18)
+#define ASLC_BUTTON_ARRAY_FAILED	(1 << 20)
+#define ASLC_CONVERTIBLE_FAILED		(1 << 22)
+#define ASLC_DOCKING_FAILED		(1 << 24)
+#define ASLC_ISCT_STATE_FAILED		(1 << 26)
 
 /* Technology enabled indicator */
 #define ASLE_TCHE_ALS_EN	(1 << 0)
@@ -151,6 +167,60 @@ struct opregion_asle {
 
 #define ASLE_CBLV_VALID         (1<<31)
 
+/* IUER */
+#define ASLE_IUER_DOCKING		(1 << 7)
+#define ASLE_IUER_CONVERTIBLE		(1 << 6)
+#define ASLE_IUER_ROTATION_LOCK_BTN	(1 << 4)
+#define ASLE_IUER_VOLUME_DOWN_BTN	(1 << 3)
+#define ASLE_IUER_VOLUME_UP_BTN		(1 << 2)
+#define ASLE_IUER_WINDOWS_BTN		(1 << 1)
+#define ASLE_IUER_POWER_BTN		(1 << 0)
+
+/* Software System Control Interrupt (SWSCI) */
+#define SWSCI_SCIC_INDICATOR		(1 << 0)
+#define SWSCI_SCIC_MAIN_FUNCTION_SHIFT	1
+#define SWSCI_SCIC_MAIN_FUNCTION_MASK	(0xf << 1)
+#define SWSCI_SCIC_SUB_FUNCTION_SHIFT	8
+#define SWSCI_SCIC_SUB_FUNCTION_MASK	(0xff << 8)
+#define SWSCI_SCIC_EXIT_PARAMETER_SHIFT	8
+#define SWSCI_SCIC_EXIT_PARAMETER_MASK	(0xff << 8)
+#define SWSCI_SCIC_EXIT_STATUS_SHIFT	5
+#define SWSCI_SCIC_EXIT_STATUS_MASK	(7 << 5)
+#define SWSCI_SCIC_EXIT_STATUS_SUCCESS	1
+
+#define SWSCI_FUNCTION_CODE(main, sub) \
+	((main) << SWSCI_SCIC_MAIN_FUNCTION_SHIFT | \
+	 (sub) << SWSCI_SCIC_SUB_FUNCTION_SHIFT)
+
+/* SWSCI: Get BIOS Data (GBDA) */
+#define SWSCI_GBDA			4
+#define SWSCI_GBDA_SUPPORTED_CALLS	SWSCI_FUNCTION_CODE(SWSCI_GBDA, 0)
+#define SWSCI_GBDA_REQUESTED_CALLBACKS	SWSCI_FUNCTION_CODE(SWSCI_GBDA, 1)
+#define SWSCI_GBDA_BOOT_DISPLAY_PREF	SWSCI_FUNCTION_CODE(SWSCI_GBDA, 4)
+#define SWSCI_GBDA_PANEL_DETAILS	SWSCI_FUNCTION_CODE(SWSCI_GBDA, 5)
+#define SWSCI_GBDA_TV_STANDARD		SWSCI_FUNCTION_CODE(SWSCI_GBDA, 6)
+#define SWSCI_GBDA_INTERNAL_GRAPHICS	SWSCI_FUNCTION_CODE(SWSCI_GBDA, 7)
+#define SWSCI_GBDA_SPREAD_SPECTRUM	SWSCI_FUNCTION_CODE(SWSCI_GBDA, 10)
+
+/* SWSCI: System BIOS Callbacks (SBCB) */
+#define SWSCI_SBCB			6
+#define SWSCI_SBCB_SUPPORTED_CALLBACKS	SWSCI_FUNCTION_CODE(SWSCI_SBCB, 0)
+#define SWSCI_SBCB_INIT_COMPLETION	SWSCI_FUNCTION_CODE(SWSCI_SBCB, 1)
+#define SWSCI_SBCB_PRE_HIRES_SET_MODE	SWSCI_FUNCTION_CODE(SWSCI_SBCB, 3)
+#define SWSCI_SBCB_POST_HIRES_SET_MODE	SWSCI_FUNCTION_CODE(SWSCI_SBCB, 4)
+#define SWSCI_SBCB_DISPLAY_SWITCH	SWSCI_FUNCTION_CODE(SWSCI_SBCB, 5)
+#define SWSCI_SBCB_SET_TV_FORMAT	SWSCI_FUNCTION_CODE(SWSCI_SBCB, 6)
+#define SWSCI_SBCB_ADAPTER_POWER_STATE	SWSCI_FUNCTION_CODE(SWSCI_SBCB, 7)
+#define SWSCI_SBCB_DISPLAY_POWER_STATE	SWSCI_FUNCTION_CODE(SWSCI_SBCB, 8)
+#define SWSCI_SBCB_SET_BOOT_DISPLAY	SWSCI_FUNCTION_CODE(SWSCI_SBCB, 9)
+#define SWSCI_SBCB_SET_PANEL_DETAILS	SWSCI_FUNCTION_CODE(SWSCI_SBCB, 10)
+#define SWSCI_SBCB_SET_INTERNAL_GFX	SWSCI_FUNCTION_CODE(SWSCI_SBCB, 11)
+#define SWSCI_SBCB_POST_HIRES_TO_DOS_FS	SWSCI_FUNCTION_CODE(SWSCI_SBCB, 16)
+#define SWSCI_SBCB_SUSPEND_RESUME	SWSCI_FUNCTION_CODE(SWSCI_SBCB, 17)
+#define SWSCI_SBCB_SET_SPREAD_SPECTRUM	SWSCI_FUNCTION_CODE(SWSCI_SBCB, 18)
+#define SWSCI_SBCB_POST_VBE_PM		SWSCI_FUNCTION_CODE(SWSCI_SBCB, 19)
+#define SWSCI_SBCB_ENABLE_DISABLE_AUDIO	SWSCI_FUNCTION_CODE(SWSCI_SBCB, 21)
+
 #define ACPI_OTHER_OUTPUT (0<<8)
 #define ACPI_VGA_OUTPUT (1<<8)
 #define ACPI_TV_OUTPUT (2<<8)
@@ -158,24 +228,224 @@ struct opregion_asle {
 #define ACPI_LVDS_OUTPUT (4<<8)
 
 #ifdef CONFIG_ACPI
+static int swsci(struct drm_device *dev, u32 function, u32 parm, u32 *parm_out)
+{
+	struct drm_i915_private *dev_priv = dev->dev_private;
+	struct opregion_swsci __iomem *swsci = dev_priv->opregion.swsci;
+	u32 main_function, sub_function, scic;
+	u16 pci_swsci;
+	u32 dslp;
+
+	if (!swsci)
+		return -ENODEV;
+
+	main_function = (function & SWSCI_SCIC_MAIN_FUNCTION_MASK) >>
+		SWSCI_SCIC_MAIN_FUNCTION_SHIFT;
+	sub_function = (function & SWSCI_SCIC_SUB_FUNCTION_MASK) >>
+		SWSCI_SCIC_SUB_FUNCTION_SHIFT;
+
+	/* Check if we can call the function. See swsci_setup for details. */
+	if (main_function == SWSCI_SBCB) {
+		if ((dev_priv->opregion.swsci_sbcb_sub_functions &
+		     (1 << sub_function)) == 0)
+			return -EINVAL;
+	} else if (main_function == SWSCI_GBDA) {
+		if ((dev_priv->opregion.swsci_gbda_sub_functions &
+		     (1 << sub_function)) == 0)
+			return -EINVAL;
+	}
+
+	/* Driver sleep timeout in ms. */
+	dslp = ioread32(&swsci->dslp);
+	if (!dslp) {
+		/* The spec says 2ms should be the default, but it's too small
+		 * for some machines. */
+		dslp = 50;
+	} else if (dslp > 500) {
+		/* Hey bios, trust must be earned. */
+		WARN_ONCE(1, "excessive driver sleep timeout (DSPL) %u\n", dslp);
+		dslp = 500;
+	}
+
+	/* The spec tells us to do this, but we are the only user... */
+	scic = ioread32(&swsci->scic);
+	if (scic & SWSCI_SCIC_INDICATOR) {
+		DRM_DEBUG_DRIVER("SWSCI request already in progress\n");
+		return -EBUSY;
+	}
+
+	scic = function | SWSCI_SCIC_INDICATOR;
+
+	iowrite32(parm, &swsci->parm);
+	iowrite32(scic, &swsci->scic);
+
+	/* Ensure SCI event is selected and event trigger is cleared. */
+	pci_read_config_word(dev->pdev, PCI_SWSCI, &pci_swsci);
+	if (!(pci_swsci & PCI_SWSCI_SCISEL) || (pci_swsci & PCI_SWSCI_GSSCIE)) {
+		pci_swsci |= PCI_SWSCI_SCISEL;
+		pci_swsci &= ~PCI_SWSCI_GSSCIE;
+		pci_write_config_word(dev->pdev, PCI_SWSCI, pci_swsci);
+	}
+
+	/* Use event trigger to tell bios to check the mail. */
+	pci_swsci |= PCI_SWSCI_GSSCIE;
+	pci_write_config_word(dev->pdev, PCI_SWSCI, pci_swsci);
+
+	/* Poll for the result. */
+#define C (((scic = ioread32(&swsci->scic)) & SWSCI_SCIC_INDICATOR) == 0)
+	if (wait_for(C, dslp)) {
+		DRM_DEBUG_DRIVER("SWSCI request timed out\n");
+		return -ETIMEDOUT;
+	}
+
+	scic = (scic & SWSCI_SCIC_EXIT_STATUS_MASK) >>
+		SWSCI_SCIC_EXIT_STATUS_SHIFT;
+
+	/* Note: scic == 0 is an error! */
+	if (scic != SWSCI_SCIC_EXIT_STATUS_SUCCESS) {
+		DRM_DEBUG_DRIVER("SWSCI request error %u\n", scic);
+		return -EIO;
+	}
+
+	if (parm_out)
+		*parm_out = ioread32(&swsci->parm);
+
+	return 0;
+
+#undef C
+}
+
+#define DISPLAY_TYPE_CRT			0
+#define DISPLAY_TYPE_TV				1
+#define DISPLAY_TYPE_EXTERNAL_FLAT_PANEL	2
+#define DISPLAY_TYPE_INTERNAL_FLAT_PANEL	3
+
+int intel_opregion_notify_encoder(struct intel_encoder *intel_encoder,
+				  bool enable)
+{
+	struct drm_device *dev = intel_encoder->base.dev;
+	u32 parm = 0;
+	u32 type = 0;
+	u32 port;
+
+	/* don't care about old stuff for now */
+	if (!HAS_DDI(dev))
+		return 0;
+
+	port = intel_ddi_get_encoder_port(intel_encoder);
+	if (port == PORT_E) {
+		port = 0;
+	} else {
+		parm |= 1 << port;
+		port++;
+	}
+
+	if (!enable)
+		parm |= 4 << 8;
+
+	switch (intel_encoder->type) {
+	case INTEL_OUTPUT_ANALOG:
+		type = DISPLAY_TYPE_CRT;
+		break;
+	case INTEL_OUTPUT_UNKNOWN:
+	case INTEL_OUTPUT_DISPLAYPORT:
+	case INTEL_OUTPUT_HDMI:
+		type = DISPLAY_TYPE_EXTERNAL_FLAT_PANEL;
+		break;
+	case INTEL_OUTPUT_EDP:
+		type = DISPLAY_TYPE_INTERNAL_FLAT_PANEL;
+		break;
+	default:
+		WARN_ONCE(1, "unsupported intel_encoder type %d\n",
+			  intel_encoder->type);
+		return -EINVAL;
+	}
+
+	parm |= type << (16 + port * 3);
+
+	return swsci(dev, SWSCI_SBCB_DISPLAY_POWER_STATE, parm, NULL);
+}
+
+static const struct {
+	pci_power_t pci_power_state;
+	u32 parm;
+} power_state_map[] = {
+	{ PCI_D0,	0x00 },
+	{ PCI_D1,	0x01 },
+	{ PCI_D2,	0x02 },
+	{ PCI_D3hot,	0x04 },
+	{ PCI_D3cold,	0x04 },
+};
+
+int intel_opregion_notify_adapter(struct drm_device *dev, pci_power_t state)
+{
+	int i;
+
+	if (!HAS_DDI(dev))
+		return 0;
+
+	for (i = 0; i < ARRAY_SIZE(power_state_map); i++) {
+		if (state == power_state_map[i].pci_power_state)
+			return swsci(dev, SWSCI_SBCB_ADAPTER_POWER_STATE,
+				     power_state_map[i].parm, NULL);
+	}
+
+	return -EINVAL;
+}
+
 static u32 asle_set_backlight(struct drm_device *dev, u32 bclp)
 {
 	struct drm_i915_private *dev_priv = dev->dev_private;
+	struct drm_encoder *encoder;
+	struct drm_connector *connector;
+	struct intel_connector *intel_connector = NULL;
+	struct drm_crtc *crtc = dev_priv->pipe_to_crtc_mapping[0];
 	struct opregion_asle __iomem *asle = dev_priv->opregion.asle;
+	u32 ret = 0;
+	bool found = false;
 
 	DRM_DEBUG_DRIVER("bclp = 0x%08x\n", bclp);
 
 	if (!(bclp & ASLE_BCLP_VALID))
-		return ASLE_BACKLIGHT_FAILED;
+		return ASLC_BACKLIGHT_FAILED;
 
 	bclp &= ASLE_BCLP_MSK;
 	if (bclp > 255)
-		return ASLE_BACKLIGHT_FAILED;
+		return ASLC_BACKLIGHT_FAILED;
+
+	mutex_lock(&dev->mode_config.mutex);
+	/*
+	 * Could match the OpRegion connector here instead, but we'd also need
+	 * to verify the connector could handle a backlight call.
+	 */
+	list_for_each_entry(encoder, &dev->mode_config.encoder_list, head)
+		if (encoder->crtc == crtc) {
+			found = true;
+			break;
+		}
+
+	if (!found) {
+		ret = ASLC_BACKLIGHT_FAILED;
+		goto out;
+	}
+
+	list_for_each_entry(connector, &dev->mode_config.connector_list, head)
+		if (connector->encoder == encoder)
+			intel_connector = to_intel_connector(connector);
+
+	if (!intel_connector) {
+		ret = ASLC_BACKLIGHT_FAILED;
+		goto out;
+	}
 
-	intel_panel_set_backlight(dev, bclp, 255);
+	DRM_DEBUG_KMS("updating opregion backlight %d/255\n", bclp);
+	intel_panel_set_backlight(intel_connector, bclp, 255);
 	iowrite32(DIV_ROUND_UP(bclp * 100, 255) | ASLE_CBLV_VALID, &asle->cblv);
 
-	return 0;
+out:
+	mutex_unlock(&dev->mode_config.mutex);
+
+	return ret;
 }
 
 static u32 asle_set_als_illum(struct drm_device *dev, u32 alsi)
@@ -183,13 +453,13 @@ static u32 asle_set_als_illum(struct drm_device *dev, u32 alsi)
 	/* alsi is the current ALS reading in lux. 0 indicates below sensor
 	   range, 0xffff indicates above sensor range. 1-0xfffe are valid */
 	DRM_DEBUG_DRIVER("Illum is not supported\n");
-	return ASLE_ALS_ILLUM_FAILED;
+	return ASLC_ALS_ILLUM_FAILED;
 }
 
 static u32 asle_set_pwm_freq(struct drm_device *dev, u32 pfmb)
 {
 	DRM_DEBUG_DRIVER("PWM freq is not supported\n");
-	return ASLE_PWM_FREQ_FAILED;
+	return ASLC_PWM_FREQ_FAILED;
 }
 
 static u32 asle_set_pfit(struct drm_device *dev, u32 pfit)
@@ -197,39 +467,118 @@ static u32 asle_set_pfit(struct drm_device *dev, u32 pfit)
 	/* Panel fitting is currently controlled by the X code, so this is a
 	   noop until modesetting support works fully */
 	DRM_DEBUG_DRIVER("Pfit is not supported\n");
-	return ASLE_PFIT_FAILED;
+	return ASLC_PFIT_FAILED;
 }
 
-void intel_opregion_asle_intr(struct drm_device *dev)
+static u32 asle_set_supported_rotation_angles(struct drm_device *dev, u32 srot)
 {
-	struct drm_i915_private *dev_priv = dev->dev_private;
+	DRM_DEBUG_DRIVER("SROT is not supported\n");
+	return ASLC_ROTATION_ANGLES_FAILED;
+}
+
+static u32 asle_set_button_array(struct drm_device *dev, u32 iuer)
+{
+	if (!iuer)
+		DRM_DEBUG_DRIVER("Button array event is not supported (nothing)\n");
+	if (iuer & ASLE_IUER_ROTATION_LOCK_BTN)
+		DRM_DEBUG_DRIVER("Button array event is not supported (rotation lock)\n");
+	if (iuer & ASLE_IUER_VOLUME_DOWN_BTN)
+		DRM_DEBUG_DRIVER("Button array event is not supported (volume down)\n");
+	if (iuer & ASLE_IUER_VOLUME_UP_BTN)
+		DRM_DEBUG_DRIVER("Button array event is not supported (volume up)\n");
+	if (iuer & ASLE_IUER_WINDOWS_BTN)
+		DRM_DEBUG_DRIVER("Button array event is not supported (windows)\n");
+	if (iuer & ASLE_IUER_POWER_BTN)
+		DRM_DEBUG_DRIVER("Button array event is not supported (power)\n");
+
+	return ASLC_BUTTON_ARRAY_FAILED;
+}
+
+static u32 asle_set_convertible(struct drm_device *dev, u32 iuer)
+{
+	if (iuer & ASLE_IUER_CONVERTIBLE)
+		DRM_DEBUG_DRIVER("Convertible is not supported (clamshell)\n");
+	else
+		DRM_DEBUG_DRIVER("Convertible is not supported (slate)\n");
+
+	return ASLC_CONVERTIBLE_FAILED;
+}
+
+static u32 asle_set_docking(struct drm_device *dev, u32 iuer)
+{
+	if (iuer & ASLE_IUER_DOCKING)
+		DRM_DEBUG_DRIVER("Docking is not supported (docked)\n");
+	else
+		DRM_DEBUG_DRIVER("Docking is not supported (undocked)\n");
+
+	return ASLC_DOCKING_FAILED;
+}
+
+static u32 asle_isct_state(struct drm_device *dev)
+{
+	DRM_DEBUG_DRIVER("ISCT is not supported\n");
+	return ASLC_ISCT_STATE_FAILED;
+}
+
+static void asle_work(struct work_struct *work)
+{
+	struct intel_opregion *opregion =
+		container_of(work, struct intel_opregion, asle_work);
+	struct drm_i915_private *dev_priv =
+		container_of(opregion, struct drm_i915_private, opregion);
+	struct drm_device *dev = dev_priv->dev;
 	struct opregion_asle __iomem *asle = dev_priv->opregion.asle;
-	u32 asle_stat = 0;
-	u32 asle_req;
+	u32 aslc_stat = 0;
+	u32 aslc_req;
 
 	if (!asle)
 		return;
 
-	asle_req = ioread32(&asle->aslc) & ASLE_REQ_MSK;
+	aslc_req = ioread32(&asle->aslc);
 
-	if (!asle_req) {
-		DRM_DEBUG_DRIVER("non asle set request??\n");
+	if (!(aslc_req & ASLC_REQ_MSK)) {
+		DRM_DEBUG_DRIVER("No request on ASLC interrupt 0x%08x\n",
+				 aslc_req);
 		return;
 	}
 
-	if (asle_req & ASLE_SET_ALS_ILLUM)
-		asle_stat |= asle_set_als_illum(dev, ioread32(&asle->alsi));
+	if (aslc_req & ASLC_SET_ALS_ILLUM)
+		aslc_stat |= asle_set_als_illum(dev, ioread32(&asle->alsi));
+
+	if (aslc_req & ASLC_SET_BACKLIGHT)
+		aslc_stat |= asle_set_backlight(dev, ioread32(&asle->bclp));
+
+	if (aslc_req & ASLC_SET_PFIT)
+		aslc_stat |= asle_set_pfit(dev, ioread32(&asle->pfit));
+
+	if (aslc_req & ASLC_SET_PWM_FREQ)
+		aslc_stat |= asle_set_pwm_freq(dev, ioread32(&asle->pfmb));
 
-	if (asle_req & ASLE_SET_BACKLIGHT)
-		asle_stat |= asle_set_backlight(dev, ioread32(&asle->bclp));
+	if (aslc_req & ASLC_SUPPORTED_ROTATION_ANGLES)
+		aslc_stat |= asle_set_supported_rotation_angles(dev,
+							ioread32(&asle->srot));
 
-	if (asle_req & ASLE_SET_PFIT)
-		asle_stat |= asle_set_pfit(dev, ioread32(&asle->pfit));
+	if (aslc_req & ASLC_BUTTON_ARRAY)
+		aslc_stat |= asle_set_button_array(dev, ioread32(&asle->iuer));
 
-	if (asle_req & ASLE_SET_PWM_FREQ)
-		asle_stat |= asle_set_pwm_freq(dev, ioread32(&asle->pfmb));
+	if (aslc_req & ASLC_CONVERTIBLE_INDICATOR)
+		aslc_stat |= asle_set_convertible(dev, ioread32(&asle->iuer));
 
-	iowrite32(asle_stat, &asle->aslc);
+	if (aslc_req & ASLC_DOCKING_INDICATOR)
+		aslc_stat |= asle_set_docking(dev, ioread32(&asle->iuer));
+
+	if (aslc_req & ASLC_ISCT_STATE_CHANGE)
+		aslc_stat |= asle_isct_state(dev);
+
+	iowrite32(aslc_stat, &asle->aslc);
+}
+
+void intel_opregion_asle_intr(struct drm_device *dev)
+{
+	struct drm_i915_private *dev_priv = dev->dev_private;
+
+	if (dev_priv->opregion.asle)
+		schedule_work(&dev_priv->opregion.asle_work);
 }
 
 #define ACPI_EV_DISPLAY_SWITCH (1<<0)
@@ -432,6 +781,8 @@ void intel_opregion_fini(struct drm_device *dev)
 	if (opregion->asle)
 		iowrite32(ASLE_ARDY_NOT_READY, &opregion->asle->ardy);
 
+	cancel_work_sync(&dev_priv->opregion.asle_work);
+
 	if (opregion->acpi) {
 		iowrite32(0, &opregion->acpi->drdy);
 
@@ -446,8 +797,68 @@ void intel_opregion_fini(struct drm_device *dev)
 	opregion->swsci = NULL;
 	opregion->asle = NULL;
 	opregion->vbt = NULL;
+	opregion->lid_state = NULL;
 }
-#endif
+
+static void swsci_setup(struct drm_device *dev)
+{
+	struct drm_i915_private *dev_priv = dev->dev_private;
+	struct intel_opregion *opregion = &dev_priv->opregion;
+	bool requested_callbacks = false;
+	u32 tmp;
+
+	/* Sub-function code 0 is okay, let's allow them. */
+	opregion->swsci_gbda_sub_functions = 1;
+	opregion->swsci_sbcb_sub_functions = 1;
+
+	/* We use GBDA to ask for supported GBDA calls. */
+	if (swsci(dev, SWSCI_GBDA_SUPPORTED_CALLS, 0, &tmp) == 0) {
+		/* make the bits match the sub-function codes */
+		tmp <<= 1;
+		opregion->swsci_gbda_sub_functions |= tmp;
+	}
+
+	/*
+	 * We also use GBDA to ask for _requested_ SBCB callbacks. The driver
+	 * must not call interfaces that are not specifically requested by the
+	 * bios.
+	 */
+	if (swsci(dev, SWSCI_GBDA_REQUESTED_CALLBACKS, 0, &tmp) == 0) {
+		/* here, the bits already match sub-function codes */
+		opregion->swsci_sbcb_sub_functions |= tmp;
+		requested_callbacks = true;
+	}
+
+	/*
+	 * But we use SBCB to ask for _supported_ SBCB calls. This does not mean
+	 * the callback is _requested_. But we still can't call interfaces that
+	 * are not requested.
+	 */
+	if (swsci(dev, SWSCI_SBCB_SUPPORTED_CALLBACKS, 0, &tmp) == 0) {
+		/* make the bits match the sub-function codes */
+		u32 low = tmp & 0x7ff;
+		u32 high = tmp & ~0xfff; /* bit 11 is reserved */
+		tmp = (high << 4) | (low << 1) | 1;
+
+		/* best guess what to do with supported wrt requested */
+		if (requested_callbacks) {
+			u32 req = opregion->swsci_sbcb_sub_functions;
+			if ((req & tmp) != req)
+				DRM_DEBUG_DRIVER("SWSCI BIOS requested (%08x) SBCB callbacks that are not supported (%08x)\n", req, tmp);
+			/* XXX: for now, trust the requested callbacks */
+			/* opregion->swsci_sbcb_sub_functions &= tmp; */
+		} else {
+			opregion->swsci_sbcb_sub_functions |= tmp;
+		}
+	}
+
+	DRM_DEBUG_DRIVER("SWSCI GBDA callbacks %08x, SBCB callbacks %08x\n",
+			 opregion->swsci_gbda_sub_functions,
+			 opregion->swsci_sbcb_sub_functions);
+}
+#else /* CONFIG_ACPI */
+static inline void swsci_setup(struct drm_device *dev) {}
+#endif  /* CONFIG_ACPI */
 
 int intel_opregion_setup(struct drm_device *dev)
 {
@@ -465,6 +876,10 @@ int intel_opregion_setup(struct drm_device *dev)
 		return -ENOTSUPP;
 	}
 
+#ifdef CONFIG_ACPI
+	INIT_WORK(&opregion->asle_work, asle_work);
+#endif
+
 	base = acpi_os_ioremap(asls, OPREGION_SIZE);
 	if (!base)
 		return -ENOMEM;
@@ -490,6 +905,7 @@ int intel_opregion_setup(struct drm_device *dev)
 	if (mboxes & MBOX_SWSCI) {
 		DRM_DEBUG_DRIVER("SWSCI supported\n");
 		opregion->swsci = base + OPREGION_SWSCI_OFFSET;
+		swsci_setup(dev);
 	}
 	if (mboxes & MBOX_ASLE) {
 		DRM_DEBUG_DRIVER("ASLE supported\n");
diff --git a/drivers/gpu/drm/i915/intel_overlay.c b/drivers/gpu/drm/i915/intel_overlay.c
index ddfd0aefe0c0..a98a990fbab3 100644
--- a/drivers/gpu/drm/i915/intel_overlay.c
+++ b/drivers/gpu/drm/i915/intel_overlay.c
@@ -821,14 +821,11 @@ int intel_overlay_switch_off(struct intel_overlay *overlay)
 static int check_overlay_possible_on_crtc(struct intel_overlay *overlay,
 					  struct intel_crtc *crtc)
 {
-	drm_i915_private_t *dev_priv = overlay->dev->dev_private;
-
 	if (!crtc->active)
 		return -EINVAL;
 
 	/* can't use the overlay with double wide pipe */
-	if (INTEL_INFO(overlay->dev)->gen < 4 &&
-	    (I915_READ(PIPECONF(crtc->pipe)) & (PIPECONF_DOUBLE_WIDE | PIPECONF_ENABLE)) != PIPECONF_ENABLE)
+	if (crtc->config.double_wide)
 		return -EINVAL;
 
 	return 0;
@@ -1056,7 +1053,7 @@ int intel_overlay_put_image(struct drm_device *dev, void *data,
 		return ret;
 	}
 
-	params = kmalloc(sizeof(struct put_image_params), GFP_KERNEL);
+	params = kmalloc(sizeof(*params), GFP_KERNEL);
 	if (!params)
 		return -ENOMEM;
 
@@ -1323,7 +1320,7 @@ void intel_setup_overlay(struct drm_device *dev)
 	if (!HAS_OVERLAY(dev))
 		return;
 
-	overlay = kzalloc(sizeof(struct intel_overlay), GFP_KERNEL);
+	overlay = kzalloc(sizeof(*overlay), GFP_KERNEL);
 	if (!overlay)
 		return;
 
diff --git a/drivers/gpu/drm/i915/intel_panel.c b/drivers/gpu/drm/i915/intel_panel.c
index 293564a2896a..f161ac02c4f6 100644
--- a/drivers/gpu/drm/i915/intel_panel.c
+++ b/drivers/gpu/drm/i915/intel_panel.c
@@ -50,23 +50,22 @@ intel_pch_panel_fitting(struct intel_crtc *intel_crtc,
 			struct intel_crtc_config *pipe_config,
 			int fitting_mode)
 {
-	struct drm_display_mode *mode, *adjusted_mode;
+	struct drm_display_mode *adjusted_mode;
 	int x, y, width, height;
 
-	mode = &pipe_config->requested_mode;
 	adjusted_mode = &pipe_config->adjusted_mode;
 
 	x = y = width = height = 0;
 
 	/* Native modes don't need fitting */
-	if (adjusted_mode->hdisplay == mode->hdisplay &&
-	    adjusted_mode->vdisplay == mode->vdisplay)
+	if (adjusted_mode->hdisplay == pipe_config->pipe_src_w &&
+	    adjusted_mode->vdisplay == pipe_config->pipe_src_h)
 		goto done;
 
 	switch (fitting_mode) {
 	case DRM_MODE_SCALE_CENTER:
-		width = mode->hdisplay;
-		height = mode->vdisplay;
+		width = pipe_config->pipe_src_w;
+		height = pipe_config->pipe_src_h;
 		x = (adjusted_mode->hdisplay - width + 1)/2;
 		y = (adjusted_mode->vdisplay - height + 1)/2;
 		break;
@@ -74,17 +73,19 @@ intel_pch_panel_fitting(struct intel_crtc *intel_crtc,
 	case DRM_MODE_SCALE_ASPECT:
 		/* Scale but preserve the aspect ratio */
 		{
-			u32 scaled_width = adjusted_mode->hdisplay * mode->vdisplay;
-			u32 scaled_height = mode->hdisplay * adjusted_mode->vdisplay;
+			u32 scaled_width = adjusted_mode->hdisplay
+				* pipe_config->pipe_src_h;
+			u32 scaled_height = pipe_config->pipe_src_w
+				* adjusted_mode->vdisplay;
 			if (scaled_width > scaled_height) { /* pillar */
-				width = scaled_height / mode->vdisplay;
+				width = scaled_height / pipe_config->pipe_src_h;
 				if (width & 1)
 					width++;
 				x = (adjusted_mode->hdisplay - width + 1) / 2;
 				y = 0;
 				height = adjusted_mode->vdisplay;
 			} else if (scaled_width < scaled_height) { /* letter */
-				height = scaled_width / mode->hdisplay;
+				height = scaled_width / pipe_config->pipe_src_w;
 				if (height & 1)
 				    height++;
 				y = (adjusted_mode->vdisplay - height + 1) / 2;
@@ -171,20 +172,96 @@ static inline u32 panel_fitter_scaling(u32 source, u32 target)
 	return (FACTOR * ratio + FACTOR/2) / FACTOR;
 }
 
+static void i965_scale_aspect(struct intel_crtc_config *pipe_config,
+			      u32 *pfit_control)
+{
+	struct drm_display_mode *adjusted_mode = &pipe_config->adjusted_mode;
+	u32 scaled_width = adjusted_mode->hdisplay *
+		pipe_config->pipe_src_h;
+	u32 scaled_height = pipe_config->pipe_src_w *
+		adjusted_mode->vdisplay;
+
+	/* 965+ is easy, it does everything in hw */
+	if (scaled_width > scaled_height)
+		*pfit_control |= PFIT_ENABLE |
+			PFIT_SCALING_PILLAR;
+	else if (scaled_width < scaled_height)
+		*pfit_control |= PFIT_ENABLE |
+			PFIT_SCALING_LETTER;
+	else if (adjusted_mode->hdisplay != pipe_config->pipe_src_w)
+		*pfit_control |= PFIT_ENABLE | PFIT_SCALING_AUTO;
+}
+
+static void i9xx_scale_aspect(struct intel_crtc_config *pipe_config,
+			      u32 *pfit_control, u32 *pfit_pgm_ratios,
+			      u32 *border)
+{
+	struct drm_display_mode *adjusted_mode = &pipe_config->adjusted_mode;
+	u32 scaled_width = adjusted_mode->hdisplay *
+		pipe_config->pipe_src_h;
+	u32 scaled_height = pipe_config->pipe_src_w *
+		adjusted_mode->vdisplay;
+	u32 bits;
+
+	/*
+	 * For earlier chips we have to calculate the scaling
+	 * ratio by hand and program it into the
+	 * PFIT_PGM_RATIO register
+	 */
+	if (scaled_width > scaled_height) { /* pillar */
+		centre_horizontally(adjusted_mode,
+				    scaled_height /
+				    pipe_config->pipe_src_h);
+
+		*border = LVDS_BORDER_ENABLE;
+		if (pipe_config->pipe_src_h != adjusted_mode->vdisplay) {
+			bits = panel_fitter_scaling(pipe_config->pipe_src_h,
+						    adjusted_mode->vdisplay);
+
+			*pfit_pgm_ratios |= (bits << PFIT_HORIZ_SCALE_SHIFT |
+					     bits << PFIT_VERT_SCALE_SHIFT);
+			*pfit_control |= (PFIT_ENABLE |
+					  VERT_INTERP_BILINEAR |
+					  HORIZ_INTERP_BILINEAR);
+		}
+	} else if (scaled_width < scaled_height) { /* letter */
+		centre_vertically(adjusted_mode,
+				  scaled_width /
+				  pipe_config->pipe_src_w);
+
+		*border = LVDS_BORDER_ENABLE;
+		if (pipe_config->pipe_src_w != adjusted_mode->hdisplay) {
+			bits = panel_fitter_scaling(pipe_config->pipe_src_w,
+						    adjusted_mode->hdisplay);
+
+			*pfit_pgm_ratios |= (bits << PFIT_HORIZ_SCALE_SHIFT |
+					     bits << PFIT_VERT_SCALE_SHIFT);
+			*pfit_control |= (PFIT_ENABLE |
+					  VERT_INTERP_BILINEAR |
+					  HORIZ_INTERP_BILINEAR);
+		}
+	} else {
+		/* Aspects match, Let hw scale both directions */
+		*pfit_control |= (PFIT_ENABLE |
+				  VERT_AUTO_SCALE | HORIZ_AUTO_SCALE |
+				  VERT_INTERP_BILINEAR |
+				  HORIZ_INTERP_BILINEAR);
+	}
+}
+
 void intel_gmch_panel_fitting(struct intel_crtc *intel_crtc,
 			      struct intel_crtc_config *pipe_config,
 			      int fitting_mode)
 {
 	struct drm_device *dev = intel_crtc->base.dev;
 	u32 pfit_control = 0, pfit_pgm_ratios = 0, border = 0;
-	struct drm_display_mode *mode, *adjusted_mode;
+	struct drm_display_mode *adjusted_mode;
 
-	mode = &pipe_config->requested_mode;
 	adjusted_mode = &pipe_config->adjusted_mode;
 
 	/* Native modes don't need fitting */
-	if (adjusted_mode->hdisplay == mode->hdisplay &&
-	    adjusted_mode->vdisplay == mode->vdisplay)
+	if (adjusted_mode->hdisplay == pipe_config->pipe_src_w &&
+	    adjusted_mode->vdisplay == pipe_config->pipe_src_h)
 		goto out;
 
 	switch (fitting_mode) {
@@ -193,81 +270,25 @@ void intel_gmch_panel_fitting(struct intel_crtc *intel_crtc,
 		 * For centered modes, we have to calculate border widths &
 		 * heights and modify the values programmed into the CRTC.
 		 */
-		centre_horizontally(adjusted_mode, mode->hdisplay);
-		centre_vertically(adjusted_mode, mode->vdisplay);
+		centre_horizontally(adjusted_mode, pipe_config->pipe_src_w);
+		centre_vertically(adjusted_mode, pipe_config->pipe_src_h);
 		border = LVDS_BORDER_ENABLE;
 		break;
 	case DRM_MODE_SCALE_ASPECT:
 		/* Scale but preserve the aspect ratio */
-		if (INTEL_INFO(dev)->gen >= 4) {
-			u32 scaled_width = adjusted_mode->hdisplay *
-				mode->vdisplay;
-			u32 scaled_height = mode->hdisplay *
-				adjusted_mode->vdisplay;
-
-			/* 965+ is easy, it does everything in hw */
-			if (scaled_width > scaled_height)
-				pfit_control |= PFIT_ENABLE |
-					PFIT_SCALING_PILLAR;
-			else if (scaled_width < scaled_height)
-				pfit_control |= PFIT_ENABLE |
-					PFIT_SCALING_LETTER;
-			else if (adjusted_mode->hdisplay != mode->hdisplay)
-				pfit_control |= PFIT_ENABLE | PFIT_SCALING_AUTO;
-		} else {
-			u32 scaled_width = adjusted_mode->hdisplay *
-				mode->vdisplay;
-			u32 scaled_height = mode->hdisplay *
-				adjusted_mode->vdisplay;
-			/*
-			 * For earlier chips we have to calculate the scaling
-			 * ratio by hand and program it into the
-			 * PFIT_PGM_RATIO register
-			 */
-			if (scaled_width > scaled_height) { /* pillar */
-				centre_horizontally(adjusted_mode,
-						    scaled_height /
-						    mode->vdisplay);
-
-				border = LVDS_BORDER_ENABLE;
-				if (mode->vdisplay != adjusted_mode->vdisplay) {
-					u32 bits = panel_fitter_scaling(mode->vdisplay, adjusted_mode->vdisplay);
-					pfit_pgm_ratios |= (bits << PFIT_HORIZ_SCALE_SHIFT |
-							    bits << PFIT_VERT_SCALE_SHIFT);
-					pfit_control |= (PFIT_ENABLE |
-							 VERT_INTERP_BILINEAR |
-							 HORIZ_INTERP_BILINEAR);
-				}
-			} else if (scaled_width < scaled_height) { /* letter */
-				centre_vertically(adjusted_mode,
-						  scaled_width /
-						  mode->hdisplay);
-
-				border = LVDS_BORDER_ENABLE;
-				if (mode->hdisplay != adjusted_mode->hdisplay) {
-					u32 bits = panel_fitter_scaling(mode->hdisplay, adjusted_mode->hdisplay);
-					pfit_pgm_ratios |= (bits << PFIT_HORIZ_SCALE_SHIFT |
-							    bits << PFIT_VERT_SCALE_SHIFT);
-					pfit_control |= (PFIT_ENABLE |
-							 VERT_INTERP_BILINEAR |
-							 HORIZ_INTERP_BILINEAR);
-				}
-			} else {
-				/* Aspects match, Let hw scale both directions */
-				pfit_control |= (PFIT_ENABLE |
-						 VERT_AUTO_SCALE | HORIZ_AUTO_SCALE |
-						 VERT_INTERP_BILINEAR |
-						 HORIZ_INTERP_BILINEAR);
-			}
-		}
+		if (INTEL_INFO(dev)->gen >= 4)
+			i965_scale_aspect(pipe_config, &pfit_control);
+		else
+			i9xx_scale_aspect(pipe_config, &pfit_control,
+					  &pfit_pgm_ratios, &border);
 		break;
 	case DRM_MODE_SCALE_FULLSCREEN:
 		/*
 		 * Full scaling, even if it changes the aspect ratio.
 		 * Fortunately this is all done for us in hw.
 		 */
-		if (mode->vdisplay != adjusted_mode->vdisplay ||
-		    mode->hdisplay != adjusted_mode->hdisplay) {
+		if (pipe_config->pipe_src_h != adjusted_mode->vdisplay ||
+		    pipe_config->pipe_src_w != adjusted_mode->hdisplay) {
 			pfit_control |= PFIT_ENABLE;
 			if (INTEL_INFO(dev)->gen >= 4)
 				pfit_control |= PFIT_SCALING_AUTO;
@@ -308,7 +329,7 @@ static int is_backlight_combination_mode(struct drm_device *dev)
 {
 	struct drm_i915_private *dev_priv = dev->dev_private;
 
-	if (INTEL_INFO(dev)->gen >= 4)
+	if (IS_GEN4(dev))
 		return I915_READ(BLC_PWM_CTL2) & BLM_COMBINATION_MODE;
 
 	if (IS_GEN2(dev))
@@ -320,7 +341,7 @@ static int is_backlight_combination_mode(struct drm_device *dev)
 /* XXX: query mode clock or hardware clock and program max PWM appropriately
  * when it's 0.
  */
-static u32 i915_read_blc_pwm_ctl(struct drm_device *dev)
+static u32 i915_read_blc_pwm_ctl(struct drm_device *dev, enum pipe pipe)
 {
 	struct drm_i915_private *dev_priv = dev->dev_private;
 	u32 val;
@@ -337,6 +358,21 @@ static u32 i915_read_blc_pwm_ctl(struct drm_device *dev)
 			val = dev_priv->regfile.saveBLC_PWM_CTL2;
 			I915_WRITE(BLC_PWM_PCH_CTL2, val);
 		}
+	} else if (IS_VALLEYVIEW(dev)) {
+		val = I915_READ(VLV_BLC_PWM_CTL(pipe));
+		if (dev_priv->regfile.saveBLC_PWM_CTL == 0) {
+			dev_priv->regfile.saveBLC_PWM_CTL = val;
+			dev_priv->regfile.saveBLC_PWM_CTL2 =
+				I915_READ(VLV_BLC_PWM_CTL2(pipe));
+		} else if (val == 0) {
+			val = dev_priv->regfile.saveBLC_PWM_CTL;
+			I915_WRITE(VLV_BLC_PWM_CTL(pipe), val);
+			I915_WRITE(VLV_BLC_PWM_CTL2(pipe),
+				   dev_priv->regfile.saveBLC_PWM_CTL2);
+		}
+
+		if (!val)
+			val = 0x0f42ffff;
 	} else {
 		val = I915_READ(BLC_PWM_CTL);
 		if (dev_priv->regfile.saveBLC_PWM_CTL == 0) {
@@ -356,11 +392,12 @@ static u32 i915_read_blc_pwm_ctl(struct drm_device *dev)
 	return val;
 }
 
-static u32 intel_panel_get_max_backlight(struct drm_device *dev)
+static u32 intel_panel_get_max_backlight(struct drm_device *dev,
+					 enum pipe pipe)
 {
 	u32 max;
 
-	max = i915_read_blc_pwm_ctl(dev);
+	max = i915_read_blc_pwm_ctl(dev, pipe);
 
 	if (HAS_PCH_SPLIT(dev)) {
 		max >>= 16;
@@ -386,7 +423,8 @@ MODULE_PARM_DESC(invert_brightness, "Invert backlight brightness "
 	"to dri-devel@lists.freedesktop.org, if your machine needs it. "
 	"It will then be included in an upcoming module version.");
 module_param_named(invert_brightness, i915_panel_invert_brightness, int, 0600);
-static u32 intel_panel_compute_brightness(struct drm_device *dev, u32 val)
+static u32 intel_panel_compute_brightness(struct drm_device *dev,
+					  enum pipe pipe, u32 val)
 {
 	struct drm_i915_private *dev_priv = dev->dev_private;
 
@@ -395,7 +433,7 @@ static u32 intel_panel_compute_brightness(struct drm_device *dev, u32 val)
 
 	if (i915_panel_invert_brightness > 0 ||
 	    dev_priv->quirks & QUIRK_INVERT_BRIGHTNESS) {
-		u32 max = intel_panel_get_max_backlight(dev);
+		u32 max = intel_panel_get_max_backlight(dev, pipe);
 		if (max)
 			return max - val;
 	}
@@ -403,18 +441,25 @@ static u32 intel_panel_compute_brightness(struct drm_device *dev, u32 val)
 	return val;
 }
 
-static u32 intel_panel_get_backlight(struct drm_device *dev)
+static u32 intel_panel_get_backlight(struct drm_device *dev,
+				     enum pipe pipe)
 {
 	struct drm_i915_private *dev_priv = dev->dev_private;
 	u32 val;
 	unsigned long flags;
+	int reg;
 
 	spin_lock_irqsave(&dev_priv->backlight.lock, flags);
 
 	if (HAS_PCH_SPLIT(dev)) {
 		val = I915_READ(BLC_PWM_CPU_CTL) & BACKLIGHT_DUTY_CYCLE_MASK;
 	} else {
-		val = I915_READ(BLC_PWM_CTL) & BACKLIGHT_DUTY_CYCLE_MASK;
+		if (IS_VALLEYVIEW(dev))
+			reg = VLV_BLC_PWM_CTL(pipe);
+		else
+			reg = BLC_PWM_CTL;
+
+		val = I915_READ(reg) & BACKLIGHT_DUTY_CYCLE_MASK;
 		if (INTEL_INFO(dev)->gen < 4)
 			val >>= 1;
 
@@ -426,7 +471,7 @@ static u32 intel_panel_get_backlight(struct drm_device *dev)
 		}
 	}
 
-	val = intel_panel_compute_brightness(dev, val);
+	val = intel_panel_compute_brightness(dev, pipe, val);
 
 	spin_unlock_irqrestore(&dev_priv->backlight.lock, flags);
 
@@ -441,19 +486,21 @@ static void intel_pch_panel_set_backlight(struct drm_device *dev, u32 level)
 	I915_WRITE(BLC_PWM_CPU_CTL, val | level);
 }
 
-static void intel_panel_actually_set_backlight(struct drm_device *dev, u32 level)
+static void intel_panel_actually_set_backlight(struct drm_device *dev,
+					       enum pipe pipe, u32 level)
 {
 	struct drm_i915_private *dev_priv = dev->dev_private;
 	u32 tmp;
+	int reg;
 
 	DRM_DEBUG_DRIVER("set backlight PWM = %d\n", level);
-	level = intel_panel_compute_brightness(dev, level);
+	level = intel_panel_compute_brightness(dev, pipe, level);
 
 	if (HAS_PCH_SPLIT(dev))
 		return intel_pch_panel_set_backlight(dev, level);
 
 	if (is_backlight_combination_mode(dev)) {
-		u32 max = intel_panel_get_max_backlight(dev);
+		u32 max = intel_panel_get_max_backlight(dev, pipe);
 		u8 lbpc;
 
 		/* we're screwed, but keep behaviour backwards compatible */
@@ -465,23 +512,34 @@ static void intel_panel_actually_set_backlight(struct drm_device *dev, u32 level
 		pci_write_config_byte(dev->pdev, PCI_LBPC, lbpc);
 	}
 
-	tmp = I915_READ(BLC_PWM_CTL);
+	if (IS_VALLEYVIEW(dev))
+		reg = VLV_BLC_PWM_CTL(pipe);
+	else
+		reg = BLC_PWM_CTL;
+
+	tmp = I915_READ(reg);
 	if (INTEL_INFO(dev)->gen < 4)
 		level <<= 1;
 	tmp &= ~BACKLIGHT_DUTY_CYCLE_MASK;
-	I915_WRITE(BLC_PWM_CTL, tmp | level);
+	I915_WRITE(reg, tmp | level);
 }
 
 /* set backlight brightness to level in range [0..max] */
-void intel_panel_set_backlight(struct drm_device *dev, u32 level, u32 max)
+void intel_panel_set_backlight(struct intel_connector *connector, u32 level,
+			       u32 max)
 {
+	struct drm_device *dev = connector->base.dev;
 	struct drm_i915_private *dev_priv = dev->dev_private;
+	enum pipe pipe = intel_get_pipe_from_connector(connector);
 	u32 freq;
 	unsigned long flags;
 
+	if (pipe == INVALID_PIPE)
+		return;
+
 	spin_lock_irqsave(&dev_priv->backlight.lock, flags);
 
-	freq = intel_panel_get_max_backlight(dev);
+	freq = intel_panel_get_max_backlight(dev, pipe);
 	if (!freq) {
 		/* we are screwed, bail out */
 		goto out;
@@ -498,16 +556,21 @@ void intel_panel_set_backlight(struct drm_device *dev, u32 level, u32 max)
 		dev_priv->backlight.device->props.brightness = level;
 
 	if (dev_priv->backlight.enabled)
-		intel_panel_actually_set_backlight(dev, level);
+		intel_panel_actually_set_backlight(dev, pipe, level);
 out:
 	spin_unlock_irqrestore(&dev_priv->backlight.lock, flags);
 }
 
-void intel_panel_disable_backlight(struct drm_device *dev)
+void intel_panel_disable_backlight(struct intel_connector *connector)
 {
+	struct drm_device *dev = connector->base.dev;
 	struct drm_i915_private *dev_priv = dev->dev_private;
+	enum pipe pipe = intel_get_pipe_from_connector(connector);
 	unsigned long flags;
 
+	if (pipe == INVALID_PIPE)
+		return;
+
 	/*
 	 * Do not disable backlight on the vgaswitcheroo path. When switching
 	 * away from i915, the other client may depend on i915 to handle the
@@ -522,12 +585,17 @@ void intel_panel_disable_backlight(struct drm_device *dev)
 	spin_lock_irqsave(&dev_priv->backlight.lock, flags);
 
 	dev_priv->backlight.enabled = false;
-	intel_panel_actually_set_backlight(dev, 0);
+	intel_panel_actually_set_backlight(dev, pipe, 0);
 
 	if (INTEL_INFO(dev)->gen >= 4) {
 		uint32_t reg, tmp;
 
-		reg = HAS_PCH_SPLIT(dev) ? BLC_PWM_CPU_CTL2 : BLC_PWM_CTL2;
+		if (HAS_PCH_SPLIT(dev))
+			reg = BLC_PWM_CPU_CTL2;
+		else if (IS_VALLEYVIEW(dev))
+			reg = VLV_BLC_PWM_CTL2(pipe);
+		else
+			reg = BLC_PWM_CTL2;
 
 		I915_WRITE(reg, I915_READ(reg) & ~BLM_PWM_ENABLE);
 
@@ -541,18 +609,25 @@ void intel_panel_disable_backlight(struct drm_device *dev)
 	spin_unlock_irqrestore(&dev_priv->backlight.lock, flags);
 }
 
-void intel_panel_enable_backlight(struct drm_device *dev,
-				  enum pipe pipe)
+void intel_panel_enable_backlight(struct intel_connector *connector)
 {
+	struct drm_device *dev = connector->base.dev;
 	struct drm_i915_private *dev_priv = dev->dev_private;
+	enum pipe pipe = intel_get_pipe_from_connector(connector);
 	enum transcoder cpu_transcoder =
 		intel_pipe_to_cpu_transcoder(dev_priv, pipe);
 	unsigned long flags;
 
+	if (pipe == INVALID_PIPE)
+		return;
+
+	DRM_DEBUG_KMS("pipe %c\n", pipe_name(pipe));
+
 	spin_lock_irqsave(&dev_priv->backlight.lock, flags);
 
 	if (dev_priv->backlight.level == 0) {
-		dev_priv->backlight.level = intel_panel_get_max_backlight(dev);
+		dev_priv->backlight.level = intel_panel_get_max_backlight(dev,
+									  pipe);
 		if (dev_priv->backlight.device)
 			dev_priv->backlight.device->props.brightness =
 				dev_priv->backlight.level;
@@ -561,8 +636,12 @@ void intel_panel_enable_backlight(struct drm_device *dev,
 	if (INTEL_INFO(dev)->gen >= 4) {
 		uint32_t reg, tmp;
 
-		reg = HAS_PCH_SPLIT(dev) ? BLC_PWM_CPU_CTL2 : BLC_PWM_CTL2;
-
+		if (HAS_PCH_SPLIT(dev))
+			reg = BLC_PWM_CPU_CTL2;
+		else if (IS_VALLEYVIEW(dev))
+			reg = VLV_BLC_PWM_CTL2(pipe);
+		else
+			reg = BLC_PWM_CTL2;
 
 		tmp = I915_READ(reg);
 
@@ -602,16 +681,41 @@ set_level:
 	 * registers are set.
 	 */
 	dev_priv->backlight.enabled = true;
-	intel_panel_actually_set_backlight(dev, dev_priv->backlight.level);
+	intel_panel_actually_set_backlight(dev, pipe,
+					   dev_priv->backlight.level);
 
 	spin_unlock_irqrestore(&dev_priv->backlight.lock, flags);
 }
 
+/* FIXME: use VBT vals to init PWM_CTL and PWM_CTL2 correctly */
+static void intel_panel_init_backlight_regs(struct drm_device *dev)
+{
+	struct drm_i915_private *dev_priv = dev->dev_private;
+
+	if (IS_VALLEYVIEW(dev)) {
+		enum pipe pipe;
+
+		for_each_pipe(pipe) {
+			u32 cur_val = I915_READ(VLV_BLC_PWM_CTL(pipe));
+
+			/* Skip if the modulation freq is already set */
+			if (cur_val & ~BACKLIGHT_DUTY_CYCLE_MASK)
+				continue;
+
+			cur_val &= BACKLIGHT_DUTY_CYCLE_MASK;
+			I915_WRITE(VLV_BLC_PWM_CTL(pipe), (0xf42 << 16) |
+				   cur_val);
+		}
+	}
+}
+
 static void intel_panel_init_backlight(struct drm_device *dev)
 {
 	struct drm_i915_private *dev_priv = dev->dev_private;
 
-	dev_priv->backlight.level = intel_panel_get_backlight(dev);
+	intel_panel_init_backlight_regs(dev);
+
+	dev_priv->backlight.level = intel_panel_get_backlight(dev, 0);
 	dev_priv->backlight.enabled = dev_priv->backlight.level != 0;
 }
 
@@ -637,19 +741,34 @@ intel_panel_detect(struct drm_device *dev)
 	}
 }
 
-#ifdef CONFIG_BACKLIGHT_CLASS_DEVICE
+#if IS_ENABLED(CONFIG_BACKLIGHT_CLASS_DEVICE)
 static int intel_panel_update_status(struct backlight_device *bd)
 {
-	struct drm_device *dev = bl_get_data(bd);
-	intel_panel_set_backlight(dev, bd->props.brightness,
+	struct intel_connector *connector = bl_get_data(bd);
+	struct drm_device *dev = connector->base.dev;
+
+	mutex_lock(&dev->mode_config.mutex);
+	DRM_DEBUG_KMS("updating intel_backlight, brightness=%d/%d\n",
+		      bd->props.brightness, bd->props.max_brightness);
+	intel_panel_set_backlight(connector, bd->props.brightness,
 				  bd->props.max_brightness);
+	mutex_unlock(&dev->mode_config.mutex);
 	return 0;
 }
 
 static int intel_panel_get_brightness(struct backlight_device *bd)
 {
-	struct drm_device *dev = bl_get_data(bd);
-	return intel_panel_get_backlight(dev);
+	struct intel_connector *connector = bl_get_data(bd);
+	struct drm_device *dev = connector->base.dev;
+	enum pipe pipe;
+
+	mutex_lock(&dev->mode_config.mutex);
+	pipe = intel_get_pipe_from_connector(connector);
+	mutex_unlock(&dev->mode_config.mutex);
+	if (pipe == INVALID_PIPE)
+		return 0;
+
+	return intel_panel_get_backlight(connector->base.dev, pipe);
 }
 
 static const struct backlight_ops intel_panel_bl_ops = {
@@ -674,7 +793,7 @@ int intel_panel_setup_backlight(struct drm_connector *connector)
 	props.brightness = dev_priv->backlight.level;
 
 	spin_lock_irqsave(&dev_priv->backlight.lock, flags);
-	props.max_brightness = intel_panel_get_max_backlight(dev);
+	props.max_brightness = intel_panel_get_max_backlight(dev, 0);
 	spin_unlock_irqrestore(&dev_priv->backlight.lock, flags);
 
 	if (props.max_brightness == 0) {
@@ -683,7 +802,8 @@ int intel_panel_setup_backlight(struct drm_connector *connector)
 	}
 	dev_priv->backlight.device =
 		backlight_device_register("intel_backlight",
-					  &connector->kdev, dev,
+					  connector->kdev,
+					  to_intel_connector(connector),
 					  &intel_panel_bl_ops, &props);
 
 	if (IS_ERR(dev_priv->backlight.device)) {
diff --git a/drivers/gpu/drm/i915/intel_pm.c b/drivers/gpu/drm/i915/intel_pm.c
index 26c2ea3e985c..0a07d7c9cafc 100644
--- a/drivers/gpu/drm/i915/intel_pm.c
+++ b/drivers/gpu/drm/i915/intel_pm.c
@@ -32,6 +32,27 @@
 #include <linux/module.h>
 #include <drm/i915_powerwell.h>
 
+/**
+ * RC6 is a special power stage which allows the GPU to enter an very
+ * low-voltage mode when idle, using down to 0V while at this stage.  This
+ * stage is entered automatically when the GPU is idle when RC6 support is
+ * enabled, and as soon as new workload arises GPU wakes up automatically as well.
+ *
+ * There are different RC6 modes available in Intel GPU, which differentiate
+ * among each other with the latency required to enter and leave RC6 and
+ * voltage consumed by the GPU in different states.
+ *
+ * The combination of the following flags define which states GPU is allowed
+ * to enter, while RC6 is the normal RC6 state, RC6p is the deep RC6, and
+ * RC6pp is deepest RC6. Their support by hardware varies according to the
+ * GPU, BIOS, chipset and platform. RC6 is usually the safest one and the one
+ * which brings the most power savings; deeper states save more power, but
+ * require higher latency to switch to and wake up.
+ */
+#define INTEL_RC6_ENABLE			(1<<0)
+#define INTEL_RC6p_ENABLE			(1<<1)
+#define INTEL_RC6pp_ENABLE			(1<<2)
+
 /* FBC, or Frame Buffer Compression, is a technique employed to compress the
  * framebuffer contents in-memory, aiming at reducing the required bandwidth
  * during in-memory transfers and, therefore, reduce the power packet.
@@ -43,14 +64,6 @@
  * i915.i915_enable_fbc parameter
  */
 
-static bool intel_crtc_active(struct drm_crtc *crtc)
-{
-	/* Be paranoid as we can arrive here with only partial
-	 * state retrieved from the hardware during setup.
-	 */
-	return to_intel_crtc(crtc)->active && crtc->fb && crtc->mode.clock;
-}
-
 static void i8xx_disable_fbc(struct drm_device *dev)
 {
 	struct drm_i915_private *dev_priv = dev->dev_private;
@@ -241,18 +254,6 @@ static void ironlake_disable_fbc(struct drm_device *dev)
 		dpfc_ctl &= ~DPFC_CTL_EN;
 		I915_WRITE(ILK_DPFC_CONTROL, dpfc_ctl);
 
-		if (IS_IVYBRIDGE(dev))
-			/* WaFbcDisableDpfcClockGating:ivb */
-			I915_WRITE(ILK_DSPCLK_GATE_D,
-				   I915_READ(ILK_DSPCLK_GATE_D) &
-				   ~ILK_DPFCUNIT_CLOCK_GATE_DISABLE);
-
-		if (IS_HASWELL(dev))
-			/* WaFbcDisableDpfcClockGating:hsw */
-			I915_WRITE(HSW_CLKGATE_DISABLE_PART_1,
-				   I915_READ(HSW_CLKGATE_DISABLE_PART_1) &
-				   ~HSW_DPFC_GATING_DISABLE);
-
 		DRM_DEBUG_KMS("disabled FBC\n");
 	}
 }
@@ -282,18 +283,10 @@ static void gen7_enable_fbc(struct drm_crtc *crtc, unsigned long interval)
 	if (IS_IVYBRIDGE(dev)) {
 		/* WaFbcAsynchFlipDisableFbcQueue:ivb */
 		I915_WRITE(ILK_DISPLAY_CHICKEN1, ILK_FBCQ_DIS);
-		/* WaFbcDisableDpfcClockGating:ivb */
-		I915_WRITE(ILK_DSPCLK_GATE_D,
-			   I915_READ(ILK_DSPCLK_GATE_D) |
-			   ILK_DPFCUNIT_CLOCK_GATE_DISABLE);
 	} else {
 		/* WaFbcAsynchFlipDisableFbcQueue:hsw */
 		I915_WRITE(HSW_PIPE_SLICE_CHICKEN_1(intel_crtc->pipe),
 			   HSW_BYPASS_FBC_QUEUE);
-		/* WaFbcDisableDpfcClockGating:hsw */
-		I915_WRITE(HSW_CLKGATE_DISABLE_PART_1,
-			   I915_READ(HSW_CLKGATE_DISABLE_PART_1) |
-			   HSW_DPFC_GATING_DISABLE);
 	}
 
 	I915_WRITE(SNB_DPFC_CTL_SA,
@@ -378,7 +371,7 @@ static void intel_enable_fbc(struct drm_crtc *crtc, unsigned long interval)
 
 	intel_cancel_fbc_work(dev_priv);
 
-	work = kzalloc(sizeof *work, GFP_KERNEL);
+	work = kzalloc(sizeof(*work), GFP_KERNEL);
 	if (work == NULL) {
 		DRM_ERROR("Failed to allocate FBC work structure\n");
 		dev_priv->display.enable_fbc(crtc, interval);
@@ -458,7 +451,8 @@ void intel_update_fbc(struct drm_device *dev)
 	struct drm_framebuffer *fb;
 	struct intel_framebuffer *intel_fb;
 	struct drm_i915_gem_object *obj;
-	unsigned int max_hdisplay, max_vdisplay;
+	const struct drm_display_mode *adjusted_mode;
+	unsigned int max_width, max_height;
 
 	if (!I915_HAS_FBC(dev)) {
 		set_no_fbc_reason(dev_priv, FBC_UNSUPPORTED);
@@ -482,7 +476,7 @@ void intel_update_fbc(struct drm_device *dev)
 	 */
 	list_for_each_entry(tmp_crtc, &dev->mode_config.crtc_list, head) {
 		if (intel_crtc_active(tmp_crtc) &&
-		    !to_intel_crtc(tmp_crtc)->primary_disabled) {
+		    to_intel_crtc(tmp_crtc)->primary_enabled) {
 			if (crtc) {
 				if (set_no_fbc_reason(dev_priv, FBC_MULTIPLE_PIPES))
 					DRM_DEBUG_KMS("more than one pipe active, disabling compression\n");
@@ -502,6 +496,7 @@ void intel_update_fbc(struct drm_device *dev)
 	fb = crtc->fb;
 	intel_fb = to_intel_framebuffer(fb);
 	obj = intel_fb->obj;
+	adjusted_mode = &intel_crtc->config.adjusted_mode;
 
 	if (i915_enable_fbc < 0 &&
 	    INTEL_INFO(dev)->gen <= 7 && !IS_HASWELL(dev)) {
@@ -514,8 +509,8 @@ void intel_update_fbc(struct drm_device *dev)
 			DRM_DEBUG_KMS("fbc disabled per module param\n");
 		goto out_disable;
 	}
-	if ((crtc->mode.flags & DRM_MODE_FLAG_INTERLACE) ||
-	    (crtc->mode.flags & DRM_MODE_FLAG_DBLSCAN)) {
+	if ((adjusted_mode->flags & DRM_MODE_FLAG_INTERLACE) ||
+	    (adjusted_mode->flags & DRM_MODE_FLAG_DBLSCAN)) {
 		if (set_no_fbc_reason(dev_priv, FBC_UNSUPPORTED_MODE))
 			DRM_DEBUG_KMS("mode incompatible with compression, "
 				      "disabling\n");
@@ -523,14 +518,14 @@ void intel_update_fbc(struct drm_device *dev)
 	}
 
 	if (IS_G4X(dev) || INTEL_INFO(dev)->gen >= 5) {
-		max_hdisplay = 4096;
-		max_vdisplay = 2048;
+		max_width = 4096;
+		max_height = 2048;
 	} else {
-		max_hdisplay = 2048;
-		max_vdisplay = 1536;
+		max_width = 2048;
+		max_height = 1536;
 	}
-	if ((crtc->mode.hdisplay > max_hdisplay) ||
-	    (crtc->mode.vdisplay > max_vdisplay)) {
+	if (intel_crtc->config.pipe_src_w > max_width ||
+	    intel_crtc->config.pipe_src_h > max_height) {
 		if (set_no_fbc_reason(dev_priv, FBC_MODE_TOO_LARGE))
 			DRM_DEBUG_KMS("mode too large for compression, disabling\n");
 		goto out_disable;
@@ -1087,8 +1082,9 @@ static struct drm_crtc *single_enabled_crtc(struct drm_device *dev)
 	return enabled;
 }
 
-static void pineview_update_wm(struct drm_device *dev)
+static void pineview_update_wm(struct drm_crtc *unused_crtc)
 {
+	struct drm_device *dev = unused_crtc->dev;
 	struct drm_i915_private *dev_priv = dev->dev_private;
 	struct drm_crtc *crtc;
 	const struct cxsr_latency *latency;
@@ -1105,8 +1101,12 @@ static void pineview_update_wm(struct drm_device *dev)
 
 	crtc = single_enabled_crtc(dev);
 	if (crtc) {
-		int clock = crtc->mode.clock;
+		const struct drm_display_mode *adjusted_mode;
 		int pixel_size = crtc->fb->bits_per_pixel / 8;
+		int clock;
+
+		adjusted_mode = &to_intel_crtc(crtc)->config.adjusted_mode;
+		clock = adjusted_mode->crtc_clock;
 
 		/* Display SR */
 		wm = intel_calculate_wm(clock, &pineview_display_wm,
@@ -1166,6 +1166,7 @@ static bool g4x_compute_wm0(struct drm_device *dev,
 			    int *cursor_wm)
 {
 	struct drm_crtc *crtc;
+	const struct drm_display_mode *adjusted_mode;
 	int htotal, hdisplay, clock, pixel_size;
 	int line_time_us, line_count;
 	int entries, tlb_miss;
@@ -1177,9 +1178,10 @@ static bool g4x_compute_wm0(struct drm_device *dev,
 		return false;
 	}
 
-	htotal = crtc->mode.htotal;
-	hdisplay = crtc->mode.hdisplay;
-	clock = crtc->mode.clock;
+	adjusted_mode = &to_intel_crtc(crtc)->config.adjusted_mode;
+	clock = adjusted_mode->crtc_clock;
+	htotal = adjusted_mode->htotal;
+	hdisplay = to_intel_crtc(crtc)->config.pipe_src_w;
 	pixel_size = crtc->fb->bits_per_pixel / 8;
 
 	/* Use the small buffer method to calculate plane watermark */
@@ -1250,6 +1252,7 @@ static bool g4x_compute_srwm(struct drm_device *dev,
 			     int *display_wm, int *cursor_wm)
 {
 	struct drm_crtc *crtc;
+	const struct drm_display_mode *adjusted_mode;
 	int hdisplay, htotal, pixel_size, clock;
 	unsigned long line_time_us;
 	int line_count, line_size;
@@ -1262,9 +1265,10 @@ static bool g4x_compute_srwm(struct drm_device *dev,
 	}
 
 	crtc = intel_get_crtc_for_plane(dev, plane);
-	hdisplay = crtc->mode.hdisplay;
-	htotal = crtc->mode.htotal;
-	clock = crtc->mode.clock;
+	adjusted_mode = &to_intel_crtc(crtc)->config.adjusted_mode;
+	clock = adjusted_mode->crtc_clock;
+	htotal = adjusted_mode->htotal;
+	hdisplay = to_intel_crtc(crtc)->config.pipe_src_w;
 	pixel_size = crtc->fb->bits_per_pixel / 8;
 
 	line_time_us = (htotal * 1000) / clock;
@@ -1303,7 +1307,7 @@ static bool vlv_compute_drain_latency(struct drm_device *dev,
 	if (!intel_crtc_active(crtc))
 		return false;
 
-	clock = crtc->mode.clock;	/* VESA DOT Clock */
+	clock = to_intel_crtc(crtc)->config.adjusted_mode.crtc_clock;
 	pixel_size = crtc->fb->bits_per_pixel / 8;	/* BPP */
 
 	entries = (clock / 1000) * pixel_size;
@@ -1365,8 +1369,9 @@ static void vlv_update_drain_latency(struct drm_device *dev)
 
 #define single_plane_enabled(mask) is_power_of_2(mask)
 
-static void valleyview_update_wm(struct drm_device *dev)
+static void valleyview_update_wm(struct drm_crtc *crtc)
 {
+	struct drm_device *dev = crtc->dev;
 	static const int sr_latency_ns = 12000;
 	struct drm_i915_private *dev_priv = dev->dev_private;
 	int planea_wm, planeb_wm, cursora_wm, cursorb_wm;
@@ -1424,8 +1429,9 @@ static void valleyview_update_wm(struct drm_device *dev)
 		   (cursor_sr << DSPFW_CURSOR_SR_SHIFT));
 }
 
-static void g4x_update_wm(struct drm_device *dev)
+static void g4x_update_wm(struct drm_crtc *crtc)
 {
+	struct drm_device *dev = crtc->dev;
 	static const int sr_latency_ns = 12000;
 	struct drm_i915_private *dev_priv = dev->dev_private;
 	int planea_wm, planeb_wm, cursora_wm, cursorb_wm;
@@ -1476,8 +1482,9 @@ static void g4x_update_wm(struct drm_device *dev)
 		   (cursor_sr << DSPFW_CURSOR_SR_SHIFT));
 }
 
-static void i965_update_wm(struct drm_device *dev)
+static void i965_update_wm(struct drm_crtc *unused_crtc)
 {
+	struct drm_device *dev = unused_crtc->dev;
 	struct drm_i915_private *dev_priv = dev->dev_private;
 	struct drm_crtc *crtc;
 	int srwm = 1;
@@ -1488,9 +1495,11 @@ static void i965_update_wm(struct drm_device *dev)
 	if (crtc) {
 		/* self-refresh has much higher latency */
 		static const int sr_latency_ns = 12000;
-		int clock = crtc->mode.clock;
-		int htotal = crtc->mode.htotal;
-		int hdisplay = crtc->mode.hdisplay;
+		const struct drm_display_mode *adjusted_mode =
+			&to_intel_crtc(crtc)->config.adjusted_mode;
+		int clock = adjusted_mode->crtc_clock;
+		int htotal = adjusted_mode->htotal;
+		int hdisplay = to_intel_crtc(crtc)->config.pipe_src_w;
 		int pixel_size = crtc->fb->bits_per_pixel / 8;
 		unsigned long line_time_us;
 		int entries;
@@ -1541,8 +1550,9 @@ static void i965_update_wm(struct drm_device *dev)
 	I915_WRITE(DSPFW3, (cursor_sr << DSPFW_CURSOR_SR_SHIFT));
 }
 
-static void i9xx_update_wm(struct drm_device *dev)
+static void i9xx_update_wm(struct drm_crtc *unused_crtc)
 {
+	struct drm_device *dev = unused_crtc->dev;
 	struct drm_i915_private *dev_priv = dev->dev_private;
 	const struct intel_watermark_params *wm_info;
 	uint32_t fwater_lo;
@@ -1562,11 +1572,13 @@ static void i9xx_update_wm(struct drm_device *dev)
 	fifo_size = dev_priv->display.get_fifo_size(dev, 0);
 	crtc = intel_get_crtc_for_plane(dev, 0);
 	if (intel_crtc_active(crtc)) {
+		const struct drm_display_mode *adjusted_mode;
 		int cpp = crtc->fb->bits_per_pixel / 8;
 		if (IS_GEN2(dev))
 			cpp = 4;
 
-		planea_wm = intel_calculate_wm(crtc->mode.clock,
+		adjusted_mode = &to_intel_crtc(crtc)->config.adjusted_mode;
+		planea_wm = intel_calculate_wm(adjusted_mode->crtc_clock,
 					       wm_info, fifo_size, cpp,
 					       latency_ns);
 		enabled = crtc;
@@ -1576,11 +1588,13 @@ static void i9xx_update_wm(struct drm_device *dev)
 	fifo_size = dev_priv->display.get_fifo_size(dev, 1);
 	crtc = intel_get_crtc_for_plane(dev, 1);
 	if (intel_crtc_active(crtc)) {
+		const struct drm_display_mode *adjusted_mode;
 		int cpp = crtc->fb->bits_per_pixel / 8;
 		if (IS_GEN2(dev))
 			cpp = 4;
 
-		planeb_wm = intel_calculate_wm(crtc->mode.clock,
+		adjusted_mode = &to_intel_crtc(crtc)->config.adjusted_mode;
+		planeb_wm = intel_calculate_wm(adjusted_mode->crtc_clock,
 					       wm_info, fifo_size, cpp,
 					       latency_ns);
 		if (enabled == NULL)
@@ -1607,9 +1621,11 @@ static void i9xx_update_wm(struct drm_device *dev)
 	if (HAS_FW_BLC(dev) && enabled) {
 		/* self-refresh has much higher latency */
 		static const int sr_latency_ns = 6000;
-		int clock = enabled->mode.clock;
-		int htotal = enabled->mode.htotal;
-		int hdisplay = enabled->mode.hdisplay;
+		const struct drm_display_mode *adjusted_mode =
+			&to_intel_crtc(enabled)->config.adjusted_mode;
+		int clock = adjusted_mode->crtc_clock;
+		int htotal = adjusted_mode->htotal;
+		int hdisplay = to_intel_crtc(crtc)->config.pipe_src_w;
 		int pixel_size = enabled->fb->bits_per_pixel / 8;
 		unsigned long line_time_us;
 		int entries;
@@ -1658,10 +1674,12 @@ static void i9xx_update_wm(struct drm_device *dev)
 	}
 }
 
-static void i830_update_wm(struct drm_device *dev)
+static void i830_update_wm(struct drm_crtc *unused_crtc)
 {
+	struct drm_device *dev = unused_crtc->dev;
 	struct drm_i915_private *dev_priv = dev->dev_private;
 	struct drm_crtc *crtc;
+	const struct drm_display_mode *adjusted_mode;
 	uint32_t fwater_lo;
 	int planea_wm;
 
@@ -1669,7 +1687,9 @@ static void i830_update_wm(struct drm_device *dev)
 	if (crtc == NULL)
 		return;
 
-	planea_wm = intel_calculate_wm(crtc->mode.clock, &i830_wm_info,
+	adjusted_mode = &to_intel_crtc(crtc)->config.adjusted_mode;
+	planea_wm = intel_calculate_wm(adjusted_mode->crtc_clock,
+				       &i830_wm_info,
 				       dev_priv->display.get_fifo_size(dev, 0),
 				       4, latency_ns);
 	fwater_lo = I915_READ(FW_BLC) & ~0xfff;
@@ -1741,6 +1761,7 @@ static bool ironlake_compute_srwm(struct drm_device *dev, int level, int plane,
 				  int *fbc_wm, int *display_wm, int *cursor_wm)
 {
 	struct drm_crtc *crtc;
+	const struct drm_display_mode *adjusted_mode;
 	unsigned long line_time_us;
 	int hdisplay, htotal, pixel_size, clock;
 	int line_count, line_size;
@@ -1753,9 +1774,10 @@ static bool ironlake_compute_srwm(struct drm_device *dev, int level, int plane,
 	}
 
 	crtc = intel_get_crtc_for_plane(dev, plane);
-	hdisplay = crtc->mode.hdisplay;
-	htotal = crtc->mode.htotal;
-	clock = crtc->mode.clock;
+	adjusted_mode = &to_intel_crtc(crtc)->config.adjusted_mode;
+	clock = adjusted_mode->crtc_clock;
+	htotal = adjusted_mode->htotal;
+	hdisplay = to_intel_crtc(crtc)->config.pipe_src_w;
 	pixel_size = crtc->fb->bits_per_pixel / 8;
 
 	line_time_us = (htotal * 1000) / clock;
@@ -1785,8 +1807,9 @@ static bool ironlake_compute_srwm(struct drm_device *dev, int level, int plane,
 				   display, cursor);
 }
 
-static void ironlake_update_wm(struct drm_device *dev)
+static void ironlake_update_wm(struct drm_crtc *crtc)
 {
+	struct drm_device *dev = crtc->dev;
 	struct drm_i915_private *dev_priv = dev->dev_private;
 	int fbc_wm, plane_wm, cursor_wm;
 	unsigned int enabled;
@@ -1868,8 +1891,9 @@ static void ironlake_update_wm(struct drm_device *dev)
 	 */
 }
 
-static void sandybridge_update_wm(struct drm_device *dev)
+static void sandybridge_update_wm(struct drm_crtc *crtc)
 {
+	struct drm_device *dev = crtc->dev;
 	struct drm_i915_private *dev_priv = dev->dev_private;
 	int latency = dev_priv->wm.pri_latency[0] * 100;	/* In unit 0.1us */
 	u32 val;
@@ -1970,8 +1994,9 @@ static void sandybridge_update_wm(struct drm_device *dev)
 		   cursor_wm);
 }
 
-static void ivybridge_update_wm(struct drm_device *dev)
+static void ivybridge_update_wm(struct drm_crtc *crtc)
 {
+	struct drm_device *dev = crtc->dev;
 	struct drm_i915_private *dev_priv = dev->dev_private;
 	int latency = dev_priv->wm.pri_latency[0] * 100;	/* In unit 0.1us */
 	u32 val;
@@ -2098,7 +2123,7 @@ static uint32_t ilk_pipe_pixel_rate(struct drm_device *dev,
 	struct intel_crtc *intel_crtc = to_intel_crtc(crtc);
 	uint32_t pixel_rate;
 
-	pixel_rate = intel_crtc->config.adjusted_mode.clock;
+	pixel_rate = intel_crtc->config.adjusted_mode.crtc_clock;
 
 	/* We only use IF-ID interlacing. If we ever use PF-ID we'll need to
 	 * adjust the pixel_rate here. */
@@ -2107,8 +2132,8 @@ static uint32_t ilk_pipe_pixel_rate(struct drm_device *dev,
 		uint64_t pipe_w, pipe_h, pfit_w, pfit_h;
 		uint32_t pfit_size = intel_crtc->config.pch_pfit.size;
 
-		pipe_w = intel_crtc->config.requested_mode.hdisplay;
-		pipe_h = intel_crtc->config.requested_mode.vdisplay;
+		pipe_w = intel_crtc->config.pipe_src_w;
+		pipe_h = intel_crtc->config.pipe_src_h;
 		pfit_w = (pfit_size >> 16) & 0xFFFF;
 		pfit_h = pfit_size & 0xFFFF;
 		if (pipe_w < pfit_w)
@@ -2176,27 +2201,18 @@ struct hsw_wm_maximums {
 	uint16_t fbc;
 };
 
-struct hsw_wm_values {
-	uint32_t wm_pipe[3];
-	uint32_t wm_lp[3];
-	uint32_t wm_lp_spr[3];
-	uint32_t wm_linetime[3];
-	bool enable_fbc_wm;
-};
-
 /* used in computing the new watermarks state */
 struct intel_wm_config {
 	unsigned int num_pipes_active;
 	bool sprites_enabled;
 	bool sprites_scaled;
-	bool fbc_wm_enabled;
 };
 
 /*
  * For both WM_PIPE and WM_LP.
  * mem_value must be in 0.1us units.
  */
-static uint32_t ilk_compute_pri_wm(struct hsw_pipe_wm_parameters *params,
+static uint32_t ilk_compute_pri_wm(const struct hsw_pipe_wm_parameters *params,
 				   uint32_t mem_value,
 				   bool is_lp)
 {
@@ -2225,7 +2241,7 @@ static uint32_t ilk_compute_pri_wm(struct hsw_pipe_wm_parameters *params,
  * For both WM_PIPE and WM_LP.
  * mem_value must be in 0.1us units.
  */
-static uint32_t ilk_compute_spr_wm(struct hsw_pipe_wm_parameters *params,
+static uint32_t ilk_compute_spr_wm(const struct hsw_pipe_wm_parameters *params,
 				   uint32_t mem_value)
 {
 	uint32_t method1, method2;
@@ -2248,7 +2264,7 @@ static uint32_t ilk_compute_spr_wm(struct hsw_pipe_wm_parameters *params,
  * For both WM_PIPE and WM_LP.
  * mem_value must be in 0.1us units.
  */
-static uint32_t ilk_compute_cur_wm(struct hsw_pipe_wm_parameters *params,
+static uint32_t ilk_compute_cur_wm(const struct hsw_pipe_wm_parameters *params,
 				   uint32_t mem_value)
 {
 	if (!params->active || !params->cur.enabled)
@@ -2262,7 +2278,7 @@ static uint32_t ilk_compute_cur_wm(struct hsw_pipe_wm_parameters *params,
 }
 
 /* Only for WM_LP. */
-static uint32_t ilk_compute_fbc_wm(struct hsw_pipe_wm_parameters *params,
+static uint32_t ilk_compute_fbc_wm(const struct hsw_pipe_wm_parameters *params,
 				   uint32_t pri_val)
 {
 	if (!params->active || !params->pri.enabled)
@@ -2275,7 +2291,9 @@ static uint32_t ilk_compute_fbc_wm(struct hsw_pipe_wm_parameters *params,
 
 static unsigned int ilk_display_fifo_size(const struct drm_device *dev)
 {
-	if (INTEL_INFO(dev)->gen >= 7)
+	if (INTEL_INFO(dev)->gen >= 8)
+		return 3072;
+	else if (INTEL_INFO(dev)->gen >= 7)
 		return 768;
 	else
 		return 512;
@@ -2320,7 +2338,9 @@ static unsigned int ilk_plane_wm_max(const struct drm_device *dev,
 	}
 
 	/* clamp to max that the registers can hold */
-	if (INTEL_INFO(dev)->gen >= 7)
+	if (INTEL_INFO(dev)->gen >= 8)
+		max = level == 0 ? 255 : 2047;
+	else if (INTEL_INFO(dev)->gen >= 7)
 		/* IVB/HSW primary/sprite plane watermarks */
 		max = level == 0 ? 127 : 1023;
 	else if (!is_sprite)
@@ -2350,27 +2370,30 @@ static unsigned int ilk_cursor_wm_max(const struct drm_device *dev,
 }
 
 /* Calculate the maximum FBC watermark */
-static unsigned int ilk_fbc_wm_max(void)
+static unsigned int ilk_fbc_wm_max(struct drm_device *dev)
 {
 	/* max that registers can hold */
-	return 15;
+	if (INTEL_INFO(dev)->gen >= 8)
+		return 31;
+	else
+		return 15;
 }
 
-static void ilk_wm_max(struct drm_device *dev,
-		       int level,
-		       const struct intel_wm_config *config,
-		       enum intel_ddb_partitioning ddb_partitioning,
-		       struct hsw_wm_maximums *max)
+static void ilk_compute_wm_maximums(struct drm_device *dev,
+				    int level,
+				    const struct intel_wm_config *config,
+				    enum intel_ddb_partitioning ddb_partitioning,
+				    struct hsw_wm_maximums *max)
 {
 	max->pri = ilk_plane_wm_max(dev, level, config, ddb_partitioning, false);
 	max->spr = ilk_plane_wm_max(dev, level, config, ddb_partitioning, true);
 	max->cur = ilk_cursor_wm_max(dev, level, config);
-	max->fbc = ilk_fbc_wm_max();
+	max->fbc = ilk_fbc_wm_max(dev);
 }
 
-static bool ilk_check_wm(int level,
-			 const struct hsw_wm_maximums *max,
-			 struct intel_wm_level *result)
+static bool ilk_validate_wm_level(int level,
+				  const struct hsw_wm_maximums *max,
+				  struct intel_wm_level *result)
 {
 	bool ret;
 
@@ -2406,14 +2429,12 @@ static bool ilk_check_wm(int level,
 		result->enable = true;
 	}
 
-	DRM_DEBUG_KMS("WM%d: %sabled\n", level, result->enable ? "en" : "dis");
-
 	return ret;
 }
 
 static void ilk_compute_wm_level(struct drm_i915_private *dev_priv,
 				 int level,
-				 struct hsw_pipe_wm_parameters *p,
+				 const struct hsw_pipe_wm_parameters *p,
 				 struct intel_wm_level *result)
 {
 	uint16_t pri_latency = dev_priv->wm.pri_latency[level];
@@ -2434,55 +2455,6 @@ static void ilk_compute_wm_level(struct drm_i915_private *dev_priv,
 	result->enable = true;
 }
 
-static bool hsw_compute_lp_wm(struct drm_i915_private *dev_priv,
-			      int level, struct hsw_wm_maximums *max,
-			      struct hsw_pipe_wm_parameters *params,
-			      struct intel_wm_level *result)
-{
-	enum pipe pipe;
-	struct intel_wm_level res[3];
-
-	for (pipe = PIPE_A; pipe <= PIPE_C; pipe++)
-		ilk_compute_wm_level(dev_priv, level, &params[pipe], &res[pipe]);
-
-	result->pri_val = max3(res[0].pri_val, res[1].pri_val, res[2].pri_val);
-	result->spr_val = max3(res[0].spr_val, res[1].spr_val, res[2].spr_val);
-	result->cur_val = max3(res[0].cur_val, res[1].cur_val, res[2].cur_val);
-	result->fbc_val = max3(res[0].fbc_val, res[1].fbc_val, res[2].fbc_val);
-	result->enable = true;
-
-	return ilk_check_wm(level, max, result);
-}
-
-static uint32_t hsw_compute_wm_pipe(struct drm_i915_private *dev_priv,
-				    enum pipe pipe,
-				    struct hsw_pipe_wm_parameters *params)
-{
-	uint32_t pri_val, cur_val, spr_val;
-	/* WM0 latency values stored in 0.1us units */
-	uint16_t pri_latency = dev_priv->wm.pri_latency[0];
-	uint16_t spr_latency = dev_priv->wm.spr_latency[0];
-	uint16_t cur_latency = dev_priv->wm.cur_latency[0];
-
-	pri_val = ilk_compute_pri_wm(params, pri_latency, false);
-	spr_val = ilk_compute_spr_wm(params, spr_latency);
-	cur_val = ilk_compute_cur_wm(params, cur_latency);
-
-	WARN(pri_val > 127,
-	     "Primary WM error, mode not supported for pipe %c\n",
-	     pipe_name(pipe));
-	WARN(spr_val > 127,
-	     "Sprite WM error, mode not supported for pipe %c\n",
-	     pipe_name(pipe));
-	WARN(cur_val > 63,
-	     "Cursor WM error, mode not supported for pipe %c\n",
-	     pipe_name(pipe));
-
-	return (pri_val << WM0_PIPE_PLANE_SHIFT) |
-	       (spr_val << WM0_PIPE_SPRITE_SHIFT) |
-	       cur_val;
-}
-
 static uint32_t
 hsw_compute_linetime_wm(struct drm_device *dev, struct drm_crtc *crtc)
 {
@@ -2554,19 +2526,22 @@ static void intel_fixup_cur_wm_latency(struct drm_device *dev, uint16_t wm[5])
 		wm[3] *= 2;
 }
 
-static void intel_print_wm_latency(struct drm_device *dev,
-				   const char *name,
-				   const uint16_t wm[5])
+static int ilk_wm_max_level(const struct drm_device *dev)
 {
-	int level, max_level;
-
 	/* how many WM levels are we expecting */
 	if (IS_HASWELL(dev))
-		max_level = 4;
+		return 4;
 	else if (INTEL_INFO(dev)->gen >= 6)
-		max_level = 3;
+		return 3;
 	else
-		max_level = 2;
+		return 2;
+}
+
+static void intel_print_wm_latency(struct drm_device *dev,
+				   const char *name,
+				   const uint16_t wm[5])
+{
+	int level, max_level = ilk_wm_max_level(dev);
 
 	for (level = 0; level <= max_level; level++) {
 		unsigned int latency = wm[level];
@@ -2606,218 +2581,321 @@ static void intel_setup_wm_latency(struct drm_device *dev)
 	intel_print_wm_latency(dev, "Cursor", dev_priv->wm.cur_latency);
 }
 
-static void hsw_compute_wm_parameters(struct drm_device *dev,
-				      struct hsw_pipe_wm_parameters *params,
-				      struct hsw_wm_maximums *lp_max_1_2,
-				      struct hsw_wm_maximums *lp_max_5_6)
+static void hsw_compute_wm_parameters(struct drm_crtc *crtc,
+				      struct hsw_pipe_wm_parameters *p,
+				      struct intel_wm_config *config)
 {
-	struct drm_crtc *crtc;
+	struct drm_device *dev = crtc->dev;
+	struct intel_crtc *intel_crtc = to_intel_crtc(crtc);
+	enum pipe pipe = intel_crtc->pipe;
 	struct drm_plane *plane;
-	enum pipe pipe;
-	struct intel_wm_config config = {};
-
-	list_for_each_entry(crtc, &dev->mode_config.crtc_list, head) {
-		struct intel_crtc *intel_crtc = to_intel_crtc(crtc);
-		struct hsw_pipe_wm_parameters *p;
-
-		pipe = intel_crtc->pipe;
-		p = &params[pipe];
-
-		p->active = intel_crtc_active(crtc);
-		if (!p->active)
-			continue;
-
-		config.num_pipes_active++;
 
+	p->active = intel_crtc_active(crtc);
+	if (p->active) {
 		p->pipe_htotal = intel_crtc->config.adjusted_mode.htotal;
 		p->pixel_rate = ilk_pipe_pixel_rate(dev, crtc);
 		p->pri.bytes_per_pixel = crtc->fb->bits_per_pixel / 8;
 		p->cur.bytes_per_pixel = 4;
-		p->pri.horiz_pixels =
-			intel_crtc->config.requested_mode.hdisplay;
+		p->pri.horiz_pixels = intel_crtc->config.pipe_src_w;
 		p->cur.horiz_pixels = 64;
 		/* TODO: for now, assume primary and cursor planes are always enabled. */
 		p->pri.enabled = true;
 		p->cur.enabled = true;
 	}
 
+	list_for_each_entry(crtc, &dev->mode_config.crtc_list, head)
+		config->num_pipes_active += intel_crtc_active(crtc);
+
 	list_for_each_entry(plane, &dev->mode_config.plane_list, head) {
 		struct intel_plane *intel_plane = to_intel_plane(plane);
-		struct hsw_pipe_wm_parameters *p;
 
-		pipe = intel_plane->pipe;
-		p = &params[pipe];
+		if (intel_plane->pipe == pipe)
+			p->spr = intel_plane->wm;
 
-		p->spr = intel_plane->wm;
-
-		config.sprites_enabled |= p->spr.enabled;
-		config.sprites_scaled |= p->spr.scaled;
+		config->sprites_enabled |= intel_plane->wm.enabled;
+		config->sprites_scaled |= intel_plane->wm.scaled;
 	}
+}
+
+/* Compute new watermarks for the pipe */
+static bool intel_compute_pipe_wm(struct drm_crtc *crtc,
+				  const struct hsw_pipe_wm_parameters *params,
+				  struct intel_pipe_wm *pipe_wm)
+{
+	struct drm_device *dev = crtc->dev;
+	struct drm_i915_private *dev_priv = dev->dev_private;
+	int level, max_level = ilk_wm_max_level(dev);
+	/* LP0 watermark maximums depend on this pipe alone */
+	struct intel_wm_config config = {
+		.num_pipes_active = 1,
+		.sprites_enabled = params->spr.enabled,
+		.sprites_scaled = params->spr.scaled,
+	};
+	struct hsw_wm_maximums max;
 
-	ilk_wm_max(dev, 1, &config, INTEL_DDB_PART_1_2, lp_max_1_2);
+	/* LP0 watermarks always use 1/2 DDB partitioning */
+	ilk_compute_wm_maximums(dev, 0, &config, INTEL_DDB_PART_1_2, &max);
 
-	/* 5/6 split only in single pipe config on IVB+ */
-	if (INTEL_INFO(dev)->gen >= 7 && config.num_pipes_active <= 1)
-		ilk_wm_max(dev, 1, &config, INTEL_DDB_PART_5_6, lp_max_5_6);
-	else
-		*lp_max_5_6 = *lp_max_1_2;
+	for (level = 0; level <= max_level; level++)
+		ilk_compute_wm_level(dev_priv, level, params,
+				     &pipe_wm->wm[level]);
+
+	pipe_wm->linetime = hsw_compute_linetime_wm(dev, crtc);
+
+	/* At least LP0 must be valid */
+	return ilk_validate_wm_level(0, &max, &pipe_wm->wm[0]);
 }
 
-static void hsw_compute_wm_results(struct drm_device *dev,
-				   struct hsw_pipe_wm_parameters *params,
-				   struct hsw_wm_maximums *lp_maximums,
-				   struct hsw_wm_values *results)
+/*
+ * Merge the watermarks from all active pipes for a specific level.
+ */
+static void ilk_merge_wm_level(struct drm_device *dev,
+			       int level,
+			       struct intel_wm_level *ret_wm)
 {
-	struct drm_i915_private *dev_priv = dev->dev_private;
-	struct drm_crtc *crtc;
-	struct intel_wm_level lp_results[4] = {};
-	enum pipe pipe;
-	int level, max_level, wm_lp;
+	const struct intel_crtc *intel_crtc;
 
-	for (level = 1; level <= 4; level++)
-		if (!hsw_compute_lp_wm(dev_priv, level,
-				       lp_maximums, params,
-				       &lp_results[level - 1]))
-			break;
-	max_level = level - 1;
+	list_for_each_entry(intel_crtc, &dev->mode_config.crtc_list, base.head) {
+		const struct intel_wm_level *wm =
+			&intel_crtc->wm.active.wm[level];
+
+		if (!wm->enable)
+			return;
+
+		ret_wm->pri_val = max(ret_wm->pri_val, wm->pri_val);
+		ret_wm->spr_val = max(ret_wm->spr_val, wm->spr_val);
+		ret_wm->cur_val = max(ret_wm->cur_val, wm->cur_val);
+		ret_wm->fbc_val = max(ret_wm->fbc_val, wm->fbc_val);
+	}
+
+	ret_wm->enable = true;
+}
 
-	memset(results, 0, sizeof(*results));
+/*
+ * Merge all low power watermarks for all active pipes.
+ */
+static void ilk_wm_merge(struct drm_device *dev,
+			 const struct hsw_wm_maximums *max,
+			 struct intel_pipe_wm *merged)
+{
+	int level, max_level = ilk_wm_max_level(dev);
+
+	merged->fbc_wm_enabled = true;
 
-	/* The spec says it is preferred to disable FBC WMs instead of disabling
-	 * a WM level. */
-	results->enable_fbc_wm = true;
+	/* merge each WM1+ level */
 	for (level = 1; level <= max_level; level++) {
-		if (lp_results[level - 1].fbc_val > lp_maximums->fbc) {
-			results->enable_fbc_wm = false;
-			lp_results[level - 1].fbc_val = 0;
+		struct intel_wm_level *wm = &merged->wm[level];
+
+		ilk_merge_wm_level(dev, level, wm);
+
+		if (!ilk_validate_wm_level(level, max, wm))
+			break;
+
+		/*
+		 * The spec says it is preferred to disable
+		 * FBC WMs instead of disabling a WM level.
+		 */
+		if (wm->fbc_val > max->fbc) {
+			merged->fbc_wm_enabled = false;
+			wm->fbc_val = 0;
 		}
 	}
+}
 
+static int ilk_wm_lp_to_level(int wm_lp, const struct intel_pipe_wm *pipe_wm)
+{
+	/* LP1,LP2,LP3 levels are either 1,2,3 or 1,3,4 */
+	return wm_lp + (wm_lp >= 2 && pipe_wm->wm[4].enable);
+}
+
+static void hsw_compute_wm_results(struct drm_device *dev,
+				   const struct intel_pipe_wm *merged,
+				   enum intel_ddb_partitioning partitioning,
+				   struct hsw_wm_values *results)
+{
+	struct intel_crtc *intel_crtc;
+	int level, wm_lp;
+
+	results->enable_fbc_wm = merged->fbc_wm_enabled;
+	results->partitioning = partitioning;
+
+	/* LP1+ register values */
 	for (wm_lp = 1; wm_lp <= 3; wm_lp++) {
 		const struct intel_wm_level *r;
 
-		level = (max_level == 4 && wm_lp > 1) ? wm_lp + 1 : wm_lp;
-		if (level > max_level)
+		level = ilk_wm_lp_to_level(wm_lp, merged);
+
+		r = &merged->wm[level];
+		if (!r->enable)
 			break;
 
-		r = &lp_results[level - 1];
-		results->wm_lp[wm_lp - 1] = HSW_WM_LP_VAL(level * 2,
-							  r->fbc_val,
-							  r->pri_val,
-							  r->cur_val);
+		results->wm_lp[wm_lp - 1] = WM3_LP_EN |
+			((level * 2) << WM1_LP_LATENCY_SHIFT) |
+			(r->pri_val << WM1_LP_SR_SHIFT) |
+			r->cur_val;
+
+		if (INTEL_INFO(dev)->gen >= 8)
+			results->wm_lp[wm_lp - 1] |=
+				r->fbc_val << WM1_LP_FBC_SHIFT_BDW;
+		else
+			results->wm_lp[wm_lp - 1] |=
+				r->fbc_val << WM1_LP_FBC_SHIFT;
+
 		results->wm_lp_spr[wm_lp - 1] = r->spr_val;
 	}
 
-	for_each_pipe(pipe)
-		results->wm_pipe[pipe] = hsw_compute_wm_pipe(dev_priv, pipe,
-							     &params[pipe]);
+	/* LP0 register values */
+	list_for_each_entry(intel_crtc, &dev->mode_config.crtc_list, base.head) {
+		enum pipe pipe = intel_crtc->pipe;
+		const struct intel_wm_level *r =
+			&intel_crtc->wm.active.wm[0];
 
-	for_each_pipe(pipe) {
-		crtc = dev_priv->pipe_to_crtc_mapping[pipe];
-		results->wm_linetime[pipe] = hsw_compute_linetime_wm(dev, crtc);
+		if (WARN_ON(!r->enable))
+			continue;
+
+		results->wm_linetime[pipe] = intel_crtc->wm.active.linetime;
+
+		results->wm_pipe[pipe] =
+			(r->pri_val << WM0_PIPE_PLANE_SHIFT) |
+			(r->spr_val << WM0_PIPE_SPRITE_SHIFT) |
+			r->cur_val;
 	}
 }
 
 /* Find the result with the highest level enabled. Check for enable_fbc_wm in
  * case both are at the same level. Prefer r1 in case they're the same. */
-static struct hsw_wm_values *hsw_find_best_result(struct hsw_wm_values *r1,
-						  struct hsw_wm_values *r2)
+static struct intel_pipe_wm *hsw_find_best_result(struct drm_device *dev,
+						  struct intel_pipe_wm *r1,
+						  struct intel_pipe_wm *r2)
 {
-	int i, val_r1 = 0, val_r2 = 0;
+	int level, max_level = ilk_wm_max_level(dev);
+	int level1 = 0, level2 = 0;
 
-	for (i = 0; i < 3; i++) {
-		if (r1->wm_lp[i] & WM3_LP_EN)
-			val_r1 = r1->wm_lp[i] & WM1_LP_LATENCY_MASK;
-		if (r2->wm_lp[i] & WM3_LP_EN)
-			val_r2 = r2->wm_lp[i] & WM1_LP_LATENCY_MASK;
+	for (level = 1; level <= max_level; level++) {
+		if (r1->wm[level].enable)
+			level1 = level;
+		if (r2->wm[level].enable)
+			level2 = level;
 	}
 
-	if (val_r1 == val_r2) {
-		if (r2->enable_fbc_wm && !r1->enable_fbc_wm)
+	if (level1 == level2) {
+		if (r2->fbc_wm_enabled && !r1->fbc_wm_enabled)
 			return r2;
 		else
 			return r1;
-	} else if (val_r1 > val_r2) {
+	} else if (level1 > level2) {
 		return r1;
 	} else {
 		return r2;
 	}
 }
 
+/* dirty bits used to track which watermarks need changes */
+#define WM_DIRTY_PIPE(pipe) (1 << (pipe))
+#define WM_DIRTY_LINETIME(pipe) (1 << (8 + (pipe)))
+#define WM_DIRTY_LP(wm_lp) (1 << (15 + (wm_lp)))
+#define WM_DIRTY_LP_ALL (WM_DIRTY_LP(1) | WM_DIRTY_LP(2) | WM_DIRTY_LP(3))
+#define WM_DIRTY_FBC (1 << 24)
+#define WM_DIRTY_DDB (1 << 25)
+
+static unsigned int ilk_compute_wm_dirty(struct drm_device *dev,
+					 const struct hsw_wm_values *old,
+					 const struct hsw_wm_values *new)
+{
+	unsigned int dirty = 0;
+	enum pipe pipe;
+	int wm_lp;
+
+	for_each_pipe(pipe) {
+		if (old->wm_linetime[pipe] != new->wm_linetime[pipe]) {
+			dirty |= WM_DIRTY_LINETIME(pipe);
+			/* Must disable LP1+ watermarks too */
+			dirty |= WM_DIRTY_LP_ALL;
+		}
+
+		if (old->wm_pipe[pipe] != new->wm_pipe[pipe]) {
+			dirty |= WM_DIRTY_PIPE(pipe);
+			/* Must disable LP1+ watermarks too */
+			dirty |= WM_DIRTY_LP_ALL;
+		}
+	}
+
+	if (old->enable_fbc_wm != new->enable_fbc_wm) {
+		dirty |= WM_DIRTY_FBC;
+		/* Must disable LP1+ watermarks too */
+		dirty |= WM_DIRTY_LP_ALL;
+	}
+
+	if (old->partitioning != new->partitioning) {
+		dirty |= WM_DIRTY_DDB;
+		/* Must disable LP1+ watermarks too */
+		dirty |= WM_DIRTY_LP_ALL;
+	}
+
+	/* LP1+ watermarks already deemed dirty, no need to continue */
+	if (dirty & WM_DIRTY_LP_ALL)
+		return dirty;
+
+	/* Find the lowest numbered LP1+ watermark in need of an update... */
+	for (wm_lp = 1; wm_lp <= 3; wm_lp++) {
+		if (old->wm_lp[wm_lp - 1] != new->wm_lp[wm_lp - 1] ||
+		    old->wm_lp_spr[wm_lp - 1] != new->wm_lp_spr[wm_lp - 1])
+			break;
+	}
+
+	/* ...and mark it and all higher numbered LP1+ watermarks as dirty */
+	for (; wm_lp <= 3; wm_lp++)
+		dirty |= WM_DIRTY_LP(wm_lp);
+
+	return dirty;
+}
+
 /*
  * The spec says we shouldn't write when we don't need, because every write
  * causes WMs to be re-evaluated, expending some power.
  */
 static void hsw_write_wm_values(struct drm_i915_private *dev_priv,
-				struct hsw_wm_values *results,
-				enum intel_ddb_partitioning partitioning)
+				struct hsw_wm_values *results)
 {
-	struct hsw_wm_values previous;
+	struct hsw_wm_values *previous = &dev_priv->wm.hw;
+	unsigned int dirty;
 	uint32_t val;
-	enum intel_ddb_partitioning prev_partitioning;
-	bool prev_enable_fbc_wm;
-
-	previous.wm_pipe[0] = I915_READ(WM0_PIPEA_ILK);
-	previous.wm_pipe[1] = I915_READ(WM0_PIPEB_ILK);
-	previous.wm_pipe[2] = I915_READ(WM0_PIPEC_IVB);
-	previous.wm_lp[0] = I915_READ(WM1_LP_ILK);
-	previous.wm_lp[1] = I915_READ(WM2_LP_ILK);
-	previous.wm_lp[2] = I915_READ(WM3_LP_ILK);
-	previous.wm_lp_spr[0] = I915_READ(WM1S_LP_ILK);
-	previous.wm_lp_spr[1] = I915_READ(WM2S_LP_IVB);
-	previous.wm_lp_spr[2] = I915_READ(WM3S_LP_IVB);
-	previous.wm_linetime[0] = I915_READ(PIPE_WM_LINETIME(PIPE_A));
-	previous.wm_linetime[1] = I915_READ(PIPE_WM_LINETIME(PIPE_B));
-	previous.wm_linetime[2] = I915_READ(PIPE_WM_LINETIME(PIPE_C));
-
-	prev_partitioning = (I915_READ(WM_MISC) & WM_MISC_DATA_PARTITION_5_6) ?
-				INTEL_DDB_PART_5_6 : INTEL_DDB_PART_1_2;
-
-	prev_enable_fbc_wm = !(I915_READ(DISP_ARB_CTL) & DISP_FBC_WM_DIS);
-
-	if (memcmp(results->wm_pipe, previous.wm_pipe,
-		   sizeof(results->wm_pipe)) == 0 &&
-	    memcmp(results->wm_lp, previous.wm_lp,
-		   sizeof(results->wm_lp)) == 0 &&
-	    memcmp(results->wm_lp_spr, previous.wm_lp_spr,
-		   sizeof(results->wm_lp_spr)) == 0 &&
-	    memcmp(results->wm_linetime, previous.wm_linetime,
-		   sizeof(results->wm_linetime)) == 0 &&
-	    partitioning == prev_partitioning &&
-	    results->enable_fbc_wm == prev_enable_fbc_wm)
+
+	dirty = ilk_compute_wm_dirty(dev_priv->dev, previous, results);
+	if (!dirty)
 		return;
 
-	if (previous.wm_lp[2] != 0)
+	if (dirty & WM_DIRTY_LP(3) && previous->wm_lp[2] != 0)
 		I915_WRITE(WM3_LP_ILK, 0);
-	if (previous.wm_lp[1] != 0)
+	if (dirty & WM_DIRTY_LP(2) && previous->wm_lp[1] != 0)
 		I915_WRITE(WM2_LP_ILK, 0);
-	if (previous.wm_lp[0] != 0)
+	if (dirty & WM_DIRTY_LP(1) && previous->wm_lp[0] != 0)
 		I915_WRITE(WM1_LP_ILK, 0);
 
-	if (previous.wm_pipe[0] != results->wm_pipe[0])
+	if (dirty & WM_DIRTY_PIPE(PIPE_A))
 		I915_WRITE(WM0_PIPEA_ILK, results->wm_pipe[0]);
-	if (previous.wm_pipe[1] != results->wm_pipe[1])
+	if (dirty & WM_DIRTY_PIPE(PIPE_B))
 		I915_WRITE(WM0_PIPEB_ILK, results->wm_pipe[1]);
-	if (previous.wm_pipe[2] != results->wm_pipe[2])
+	if (dirty & WM_DIRTY_PIPE(PIPE_C))
 		I915_WRITE(WM0_PIPEC_IVB, results->wm_pipe[2]);
 
-	if (previous.wm_linetime[0] != results->wm_linetime[0])
+	if (dirty & WM_DIRTY_LINETIME(PIPE_A))
 		I915_WRITE(PIPE_WM_LINETIME(PIPE_A), results->wm_linetime[0]);
-	if (previous.wm_linetime[1] != results->wm_linetime[1])
+	if (dirty & WM_DIRTY_LINETIME(PIPE_B))
 		I915_WRITE(PIPE_WM_LINETIME(PIPE_B), results->wm_linetime[1]);
-	if (previous.wm_linetime[2] != results->wm_linetime[2])
+	if (dirty & WM_DIRTY_LINETIME(PIPE_C))
 		I915_WRITE(PIPE_WM_LINETIME(PIPE_C), results->wm_linetime[2]);
 
-	if (prev_partitioning != partitioning) {
+	if (dirty & WM_DIRTY_DDB) {
 		val = I915_READ(WM_MISC);
-		if (partitioning == INTEL_DDB_PART_1_2)
+		if (results->partitioning == INTEL_DDB_PART_1_2)
 			val &= ~WM_MISC_DATA_PARTITION_5_6;
 		else
 			val |= WM_MISC_DATA_PARTITION_5_6;
 		I915_WRITE(WM_MISC, val);
 	}
 
-	if (prev_enable_fbc_wm != results->enable_fbc_wm) {
+	if (dirty & WM_DIRTY_FBC) {
 		val = I915_READ(DISP_ARB_CTL);
 		if (results->enable_fbc_wm)
 			val &= ~DISP_FBC_WM_DIS;
@@ -2826,45 +2904,65 @@ static void hsw_write_wm_values(struct drm_i915_private *dev_priv,
 		I915_WRITE(DISP_ARB_CTL, val);
 	}
 
-	if (previous.wm_lp_spr[0] != results->wm_lp_spr[0])
+	if (dirty & WM_DIRTY_LP(1) && previous->wm_lp_spr[0] != results->wm_lp_spr[0])
 		I915_WRITE(WM1S_LP_ILK, results->wm_lp_spr[0]);
-	if (previous.wm_lp_spr[1] != results->wm_lp_spr[1])
+	if (dirty & WM_DIRTY_LP(2) && previous->wm_lp_spr[1] != results->wm_lp_spr[1])
 		I915_WRITE(WM2S_LP_IVB, results->wm_lp_spr[1]);
-	if (previous.wm_lp_spr[2] != results->wm_lp_spr[2])
+	if (dirty & WM_DIRTY_LP(3) && previous->wm_lp_spr[2] != results->wm_lp_spr[2])
 		I915_WRITE(WM3S_LP_IVB, results->wm_lp_spr[2]);
 
-	if (results->wm_lp[0] != 0)
+	if (dirty & WM_DIRTY_LP(1) && results->wm_lp[0] != 0)
 		I915_WRITE(WM1_LP_ILK, results->wm_lp[0]);
-	if (results->wm_lp[1] != 0)
+	if (dirty & WM_DIRTY_LP(2) && results->wm_lp[1] != 0)
 		I915_WRITE(WM2_LP_ILK, results->wm_lp[1]);
-	if (results->wm_lp[2] != 0)
+	if (dirty & WM_DIRTY_LP(3) && results->wm_lp[2] != 0)
 		I915_WRITE(WM3_LP_ILK, results->wm_lp[2]);
+
+	dev_priv->wm.hw = *results;
 }
 
-static void haswell_update_wm(struct drm_device *dev)
+static void haswell_update_wm(struct drm_crtc *crtc)
 {
+	struct intel_crtc *intel_crtc = to_intel_crtc(crtc);
+	struct drm_device *dev = crtc->dev;
 	struct drm_i915_private *dev_priv = dev->dev_private;
-	struct hsw_wm_maximums lp_max_1_2, lp_max_5_6;
-	struct hsw_pipe_wm_parameters params[3];
-	struct hsw_wm_values results_1_2, results_5_6, *best_results;
+	struct hsw_wm_maximums max;
+	struct hsw_pipe_wm_parameters params = {};
+	struct hsw_wm_values results = {};
 	enum intel_ddb_partitioning partitioning;
+	struct intel_pipe_wm pipe_wm = {};
+	struct intel_pipe_wm lp_wm_1_2 = {}, lp_wm_5_6 = {}, *best_lp_wm;
+	struct intel_wm_config config = {};
 
-	hsw_compute_wm_parameters(dev, params, &lp_max_1_2, &lp_max_5_6);
+	hsw_compute_wm_parameters(crtc, &params, &config);
+
+	intel_compute_pipe_wm(crtc, &params, &pipe_wm);
+
+	if (!memcmp(&intel_crtc->wm.active, &pipe_wm, sizeof(pipe_wm)))
+		return;
 
-	hsw_compute_wm_results(dev, params,
-			       &lp_max_1_2, &results_1_2);
-	if (lp_max_1_2.pri != lp_max_5_6.pri) {
-		hsw_compute_wm_results(dev, params,
-				       &lp_max_5_6, &results_5_6);
-		best_results = hsw_find_best_result(&results_1_2, &results_5_6);
+	intel_crtc->wm.active = pipe_wm;
+
+	ilk_compute_wm_maximums(dev, 1, &config, INTEL_DDB_PART_1_2, &max);
+	ilk_wm_merge(dev, &max, &lp_wm_1_2);
+
+	/* 5/6 split only in single pipe config on IVB+ */
+	if (INTEL_INFO(dev)->gen >= 7 &&
+	    config.num_pipes_active == 1 && config.sprites_enabled) {
+		ilk_compute_wm_maximums(dev, 1, &config, INTEL_DDB_PART_5_6, &max);
+		ilk_wm_merge(dev, &max, &lp_wm_5_6);
+
+		best_lp_wm = hsw_find_best_result(dev, &lp_wm_1_2, &lp_wm_5_6);
 	} else {
-		best_results = &results_1_2;
+		best_lp_wm = &lp_wm_1_2;
 	}
 
-	partitioning = (best_results == &results_1_2) ?
+	partitioning = (best_lp_wm == &lp_wm_1_2) ?
 		       INTEL_DDB_PART_1_2 : INTEL_DDB_PART_5_6;
 
-	hsw_write_wm_values(dev_priv, best_results, partitioning);
+	hsw_compute_wm_results(dev, best_lp_wm, partitioning, &results);
+
+	hsw_write_wm_values(dev_priv, &results);
 }
 
 static void haswell_update_sprite_wm(struct drm_plane *plane,
@@ -2879,7 +2977,7 @@ static void haswell_update_sprite_wm(struct drm_plane *plane,
 	intel_plane->wm.horiz_pixels = sprite_width;
 	intel_plane->wm.bytes_per_pixel = pixel_size;
 
-	haswell_update_wm(plane->dev);
+	haswell_update_wm(crtc);
 }
 
 static bool
@@ -2898,7 +2996,7 @@ sandybridge_compute_sprite_wm(struct drm_device *dev, int plane,
 		return false;
 	}
 
-	clock = crtc->mode.clock;
+	clock = to_intel_crtc(crtc)->config.adjusted_mode.crtc_clock;
 
 	/* Use the small buffer method to calculate the sprite watermark */
 	entries = ((clock * pixel_size / 1000) * display_latency_ns) / 1000;
@@ -2933,7 +3031,7 @@ sandybridge_compute_sprite_srwm(struct drm_device *dev, int plane,
 	}
 
 	crtc = intel_get_crtc_for_plane(dev, plane);
-	clock = crtc->mode.clock;
+	clock = to_intel_crtc(crtc)->config.adjusted_mode.crtc_clock;
 	if (!clock) {
 		*sprite_wm = 0;
 		return false;
@@ -3044,6 +3142,74 @@ static void sandybridge_update_sprite_wm(struct drm_plane *plane,
 	I915_WRITE(WM3S_LP_IVB, sprite_wm);
 }
 
+static void ilk_pipe_wm_get_hw_state(struct drm_crtc *crtc)
+{
+	struct drm_device *dev = crtc->dev;
+	struct drm_i915_private *dev_priv = dev->dev_private;
+	struct hsw_wm_values *hw = &dev_priv->wm.hw;
+	struct intel_crtc *intel_crtc = to_intel_crtc(crtc);
+	struct intel_pipe_wm *active = &intel_crtc->wm.active;
+	enum pipe pipe = intel_crtc->pipe;
+	static const unsigned int wm0_pipe_reg[] = {
+		[PIPE_A] = WM0_PIPEA_ILK,
+		[PIPE_B] = WM0_PIPEB_ILK,
+		[PIPE_C] = WM0_PIPEC_IVB,
+	};
+
+	hw->wm_pipe[pipe] = I915_READ(wm0_pipe_reg[pipe]);
+	hw->wm_linetime[pipe] = I915_READ(PIPE_WM_LINETIME(pipe));
+
+	if (intel_crtc_active(crtc)) {
+		u32 tmp = hw->wm_pipe[pipe];
+
+		/*
+		 * For active pipes LP0 watermark is marked as
+		 * enabled, and LP1+ watermaks as disabled since
+		 * we can't really reverse compute them in case
+		 * multiple pipes are active.
+		 */
+		active->wm[0].enable = true;
+		active->wm[0].pri_val = (tmp & WM0_PIPE_PLANE_MASK) >> WM0_PIPE_PLANE_SHIFT;
+		active->wm[0].spr_val = (tmp & WM0_PIPE_SPRITE_MASK) >> WM0_PIPE_SPRITE_SHIFT;
+		active->wm[0].cur_val = tmp & WM0_PIPE_CURSOR_MASK;
+		active->linetime = hw->wm_linetime[pipe];
+	} else {
+		int level, max_level = ilk_wm_max_level(dev);
+
+		/*
+		 * For inactive pipes, all watermark levels
+		 * should be marked as enabled but zeroed,
+		 * which is what we'd compute them to.
+		 */
+		for (level = 0; level <= max_level; level++)
+			active->wm[level].enable = true;
+	}
+}
+
+void ilk_wm_get_hw_state(struct drm_device *dev)
+{
+	struct drm_i915_private *dev_priv = dev->dev_private;
+	struct hsw_wm_values *hw = &dev_priv->wm.hw;
+	struct drm_crtc *crtc;
+
+	list_for_each_entry(crtc, &dev->mode_config.crtc_list, head)
+		ilk_pipe_wm_get_hw_state(crtc);
+
+	hw->wm_lp[0] = I915_READ(WM1_LP_ILK);
+	hw->wm_lp[1] = I915_READ(WM2_LP_ILK);
+	hw->wm_lp[2] = I915_READ(WM3_LP_ILK);
+
+	hw->wm_lp_spr[0] = I915_READ(WM1S_LP_ILK);
+	hw->wm_lp_spr[1] = I915_READ(WM2S_LP_IVB);
+	hw->wm_lp_spr[2] = I915_READ(WM3S_LP_IVB);
+
+	hw->partitioning = (I915_READ(WM_MISC) & WM_MISC_DATA_PARTITION_5_6) ?
+		INTEL_DDB_PART_5_6 : INTEL_DDB_PART_1_2;
+
+	hw->enable_fbc_wm =
+		!(I915_READ(DISP_ARB_CTL) & DISP_FBC_WM_DIS);
+}
+
 /**
  * intel_update_watermarks - update FIFO watermark values based on current modes
  *
@@ -3076,12 +3242,12 @@ static void sandybridge_update_sprite_wm(struct drm_plane *plane,
  * We don't use the sprite, so we can ignore that.  And on Crestline we have
  * to set the non-SR watermarks to 8.
  */
-void intel_update_watermarks(struct drm_device *dev)
+void intel_update_watermarks(struct drm_crtc *crtc)
 {
-	struct drm_i915_private *dev_priv = dev->dev_private;
+	struct drm_i915_private *dev_priv = crtc->dev->dev_private;
 
 	if (dev_priv->display.update_wm)
-		dev_priv->display.update_wm(dev);
+		dev_priv->display.update_wm(crtc);
 }
 
 void intel_update_sprite_watermarks(struct drm_plane *plane,
@@ -3287,6 +3453,98 @@ static u32 gen6_rps_limits(struct drm_i915_private *dev_priv, u8 *val)
 	return limits;
 }
 
+static void gen6_set_rps_thresholds(struct drm_i915_private *dev_priv, u8 val)
+{
+	int new_power;
+
+	new_power = dev_priv->rps.power;
+	switch (dev_priv->rps.power) {
+	case LOW_POWER:
+		if (val > dev_priv->rps.rpe_delay + 1 && val > dev_priv->rps.cur_delay)
+			new_power = BETWEEN;
+		break;
+
+	case BETWEEN:
+		if (val <= dev_priv->rps.rpe_delay && val < dev_priv->rps.cur_delay)
+			new_power = LOW_POWER;
+		else if (val >= dev_priv->rps.rp0_delay && val > dev_priv->rps.cur_delay)
+			new_power = HIGH_POWER;
+		break;
+
+	case HIGH_POWER:
+		if (val < (dev_priv->rps.rp1_delay + dev_priv->rps.rp0_delay) >> 1 && val < dev_priv->rps.cur_delay)
+			new_power = BETWEEN;
+		break;
+	}
+	/* Max/min bins are special */
+	if (val == dev_priv->rps.min_delay)
+		new_power = LOW_POWER;
+	if (val == dev_priv->rps.max_delay)
+		new_power = HIGH_POWER;
+	if (new_power == dev_priv->rps.power)
+		return;
+
+	/* Note the units here are not exactly 1us, but 1280ns. */
+	switch (new_power) {
+	case LOW_POWER:
+		/* Upclock if more than 95% busy over 16ms */
+		I915_WRITE(GEN6_RP_UP_EI, 12500);
+		I915_WRITE(GEN6_RP_UP_THRESHOLD, 11800);
+
+		/* Downclock if less than 85% busy over 32ms */
+		I915_WRITE(GEN6_RP_DOWN_EI, 25000);
+		I915_WRITE(GEN6_RP_DOWN_THRESHOLD, 21250);
+
+		I915_WRITE(GEN6_RP_CONTROL,
+			   GEN6_RP_MEDIA_TURBO |
+			   GEN6_RP_MEDIA_HW_NORMAL_MODE |
+			   GEN6_RP_MEDIA_IS_GFX |
+			   GEN6_RP_ENABLE |
+			   GEN6_RP_UP_BUSY_AVG |
+			   GEN6_RP_DOWN_IDLE_AVG);
+		break;
+
+	case BETWEEN:
+		/* Upclock if more than 90% busy over 13ms */
+		I915_WRITE(GEN6_RP_UP_EI, 10250);
+		I915_WRITE(GEN6_RP_UP_THRESHOLD, 9225);
+
+		/* Downclock if less than 75% busy over 32ms */
+		I915_WRITE(GEN6_RP_DOWN_EI, 25000);
+		I915_WRITE(GEN6_RP_DOWN_THRESHOLD, 18750);
+
+		I915_WRITE(GEN6_RP_CONTROL,
+			   GEN6_RP_MEDIA_TURBO |
+			   GEN6_RP_MEDIA_HW_NORMAL_MODE |
+			   GEN6_RP_MEDIA_IS_GFX |
+			   GEN6_RP_ENABLE |
+			   GEN6_RP_UP_BUSY_AVG |
+			   GEN6_RP_DOWN_IDLE_AVG);
+		break;
+
+	case HIGH_POWER:
+		/* Upclock if more than 85% busy over 10ms */
+		I915_WRITE(GEN6_RP_UP_EI, 8000);
+		I915_WRITE(GEN6_RP_UP_THRESHOLD, 6800);
+
+		/* Downclock if less than 60% busy over 32ms */
+		I915_WRITE(GEN6_RP_DOWN_EI, 25000);
+		I915_WRITE(GEN6_RP_DOWN_THRESHOLD, 15000);
+
+		I915_WRITE(GEN6_RP_CONTROL,
+			   GEN6_RP_MEDIA_TURBO |
+			   GEN6_RP_MEDIA_HW_NORMAL_MODE |
+			   GEN6_RP_MEDIA_IS_GFX |
+			   GEN6_RP_ENABLE |
+			   GEN6_RP_UP_BUSY_AVG |
+			   GEN6_RP_DOWN_IDLE_AVG);
+		break;
+	}
+
+	dev_priv->rps.power = new_power;
+	dev_priv->rps.last_adj = 0;
+}
+
 void gen6_set_rps(struct drm_device *dev, u8 val)
 {
 	struct drm_i915_private *dev_priv = dev->dev_private;
@@ -3299,6 +3557,8 @@ void gen6_set_rps(struct drm_device *dev, u8 val)
 	if (val == dev_priv->rps.cur_delay)
 		return;
 
+	gen6_set_rps_thresholds(dev_priv, val);
+
 	if (IS_HASWELL(dev))
 		I915_WRITE(GEN6_RPNSWREQ,
 			   HSW_FREQUENCY(val));
@@ -3320,6 +3580,32 @@ void gen6_set_rps(struct drm_device *dev, u8 val)
 	trace_intel_gpu_freq_change(val * 50);
 }
 
+void gen6_rps_idle(struct drm_i915_private *dev_priv)
+{
+	mutex_lock(&dev_priv->rps.hw_lock);
+	if (dev_priv->rps.enabled) {
+		if (dev_priv->info->is_valleyview)
+			valleyview_set_rps(dev_priv->dev, dev_priv->rps.min_delay);
+		else
+			gen6_set_rps(dev_priv->dev, dev_priv->rps.min_delay);
+		dev_priv->rps.last_adj = 0;
+	}
+	mutex_unlock(&dev_priv->rps.hw_lock);
+}
+
+void gen6_rps_boost(struct drm_i915_private *dev_priv)
+{
+	mutex_lock(&dev_priv->rps.hw_lock);
+	if (dev_priv->rps.enabled) {
+		if (dev_priv->info->is_valleyview)
+			valleyview_set_rps(dev_priv->dev, dev_priv->rps.max_delay);
+		else
+			gen6_set_rps(dev_priv->dev, dev_priv->rps.max_delay);
+		dev_priv->rps.last_adj = 0;
+	}
+	mutex_unlock(&dev_priv->rps.hw_lock);
+}
+
 /*
  * Wait until the previous freq change has completed,
  * or the timeout elapsed, and then update our notion
@@ -3415,6 +3701,20 @@ static void valleyview_disable_rps(struct drm_device *dev)
 	}
 }
 
+static void intel_print_rc6_info(struct drm_device *dev, u32 mode)
+{
+	if (IS_GEN6(dev))
+		DRM_DEBUG_DRIVER("Sandybridge: deep RC6 disabled\n");
+
+	if (IS_HASWELL(dev))
+		DRM_DEBUG_DRIVER("Haswell: only RC6 available\n");
+
+	DRM_INFO("Enabling RC6 states: RC6 %s, RC6p %s, RC6pp %s\n",
+			(mode & GEN6_RC_CTL_RC6_ENABLE) ? "on" : "off",
+			(mode & GEN6_RC_CTL_RC6p_ENABLE) ? "on" : "off",
+			(mode & GEN6_RC_CTL_RC6pp_ENABLE) ? "on" : "off");
+}
+
 int intel_enable_rc6(const struct drm_device *dev)
 {
 	/* No RC6 before Ironlake */
@@ -3429,18 +3729,13 @@ int intel_enable_rc6(const struct drm_device *dev)
 	if (INTEL_INFO(dev)->gen == 5)
 		return 0;
 
-	if (IS_HASWELL(dev)) {
-		DRM_DEBUG_DRIVER("Haswell: only RC6 available\n");
+	if (IS_HASWELL(dev))
 		return INTEL_RC6_ENABLE;
-	}
 
 	/* snb/ivb have more than one rc6 state. */
-	if (INTEL_INFO(dev)->gen == 6) {
-		DRM_DEBUG_DRIVER("Sandybridge: deep RC6 disabled\n");
+	if (INTEL_INFO(dev)->gen == 6)
 		return INTEL_RC6_ENABLE;
-	}
 
-	DRM_DEBUG_DRIVER("RC6 and deep RC6 enabled\n");
 	return (INTEL_RC6_ENABLE | INTEL_RC6p_ENABLE);
 }
 
@@ -3467,6 +3762,78 @@ static void gen6_enable_rps_interrupts(struct drm_device *dev)
 	I915_WRITE(GEN6_PMINTRMSK, ~enabled_intrs);
 }
 
+static void gen8_enable_rps(struct drm_device *dev)
+{
+	struct drm_i915_private *dev_priv = dev->dev_private;
+	struct intel_ring_buffer *ring;
+	uint32_t rc6_mask = 0, rp_state_cap;
+	int unused;
+
+	/* 1a: Software RC state - RC0 */
+	I915_WRITE(GEN6_RC_STATE, 0);
+
+	/* 1c & 1d: Get forcewake during program sequence. Although the driver
+	 * hasn't enabled a state yet where we need forcewake, BIOS may have.*/
+	gen6_gt_force_wake_get(dev_priv);
+
+	/* 2a: Disable RC states. */
+	I915_WRITE(GEN6_RC_CONTROL, 0);
+
+	rp_state_cap = I915_READ(GEN6_RP_STATE_CAP);
+
+	/* 2b: Program RC6 thresholds.*/
+	I915_WRITE(GEN6_RC6_WAKE_RATE_LIMIT, 40 << 16);
+	I915_WRITE(GEN6_RC_EVALUATION_INTERVAL, 125000); /* 12500 * 1280ns */
+	I915_WRITE(GEN6_RC_IDLE_HYSTERSIS, 25); /* 25 * 1280ns */
+	for_each_ring(ring, dev_priv, unused)
+		I915_WRITE(RING_MAX_IDLE(ring->mmio_base), 10);
+	I915_WRITE(GEN6_RC_SLEEP, 0);
+	I915_WRITE(GEN6_RC6_THRESHOLD, 50000); /* 50/125ms per EI */
+
+	/* 3: Enable RC6 */
+	if (intel_enable_rc6(dev) & INTEL_RC6_ENABLE)
+		rc6_mask = GEN6_RC_CTL_RC6_ENABLE;
+	DRM_INFO("RC6 %s\n", (rc6_mask & GEN6_RC_CTL_RC6_ENABLE) ? "on" : "off");
+	I915_WRITE(GEN6_RC_CONTROL, GEN6_RC_CTL_HW_ENABLE |
+			GEN6_RC_CTL_EI_MODE(1) |
+			rc6_mask);
+
+	/* 4 Program defaults and thresholds for RPS*/
+	I915_WRITE(GEN6_RPNSWREQ, HSW_FREQUENCY(10)); /* Request 500 MHz */
+	I915_WRITE(GEN6_RC_VIDEO_FREQ, HSW_FREQUENCY(12)); /* Request 600 MHz */
+	/* NB: Docs say 1s, and 1000000 - which aren't equivalent */
+	I915_WRITE(GEN6_RP_DOWN_TIMEOUT, 100000000 / 128); /* 1 second timeout */
+
+	/* Docs recommend 900MHz, and 300 MHz respectively */
+	I915_WRITE(GEN6_RP_INTERRUPT_LIMITS,
+		   dev_priv->rps.max_delay << 24 |
+		   dev_priv->rps.min_delay << 16);
+
+	I915_WRITE(GEN6_RP_UP_THRESHOLD, 7600000 / 128); /* 76ms busyness per EI, 90% */
+	I915_WRITE(GEN6_RP_DOWN_THRESHOLD, 31300000 / 128); /* 313ms busyness per EI, 70%*/
+	I915_WRITE(GEN6_RP_UP_EI, 66000); /* 84.48ms, XXX: random? */
+	I915_WRITE(GEN6_RP_DOWN_EI, 350000); /* 448ms, XXX: random? */
+
+	I915_WRITE(GEN6_RP_IDLE_HYSTERSIS, 10);
+
+	/* 5: Enable RPS */
+	I915_WRITE(GEN6_RP_CONTROL,
+		   GEN6_RP_MEDIA_TURBO |
+		   GEN6_RP_MEDIA_HW_NORMAL_MODE |
+		   GEN6_RP_MEDIA_IS_GFX |
+		   GEN6_RP_ENABLE |
+		   GEN6_RP_UP_BUSY_AVG |
+		   GEN6_RP_DOWN_IDLE_AVG);
+
+	/* 6: Ring frequency + overclocking (our driver does this later */
+
+	gen6_set_rps(dev, (I915_READ(GEN6_GT_PERF_STATUS) & 0xff00) >> 8);
+
+	gen6_enable_rps_interrupts(dev);
+
+	gen6_gt_force_wake_put(dev_priv);
+}
+
 static void gen6_enable_rps(struct drm_device *dev)
 {
 	struct drm_i915_private *dev_priv = dev->dev_private;
@@ -3501,7 +3868,10 @@ static void gen6_enable_rps(struct drm_device *dev)
 
 	/* In units of 50MHz */
 	dev_priv->rps.hw_max = dev_priv->rps.max_delay = rp_state_cap & 0xff;
-	dev_priv->rps.min_delay = (rp_state_cap & 0xff0000) >> 16;
+	dev_priv->rps.min_delay = (rp_state_cap >> 16) & 0xff;
+	dev_priv->rps.rp1_delay = (rp_state_cap >>  8) & 0xff;
+	dev_priv->rps.rp0_delay = (rp_state_cap >>  0) & 0xff;
+	dev_priv->rps.rpe_delay = dev_priv->rps.rp1_delay;
 	dev_priv->rps.cur_delay = 0;
 
 	/* disable the counters and set deterministic thresholds */
@@ -3539,48 +3909,16 @@ static void gen6_enable_rps(struct drm_device *dev)
 			rc6_mask |= GEN6_RC_CTL_RC6pp_ENABLE;
 	}
 
-	DRM_INFO("Enabling RC6 states: RC6 %s, RC6p %s, RC6pp %s\n",
-			(rc6_mask & GEN6_RC_CTL_RC6_ENABLE) ? "on" : "off",
-			(rc6_mask & GEN6_RC_CTL_RC6p_ENABLE) ? "on" : "off",
-			(rc6_mask & GEN6_RC_CTL_RC6pp_ENABLE) ? "on" : "off");
+	intel_print_rc6_info(dev, rc6_mask);
 
 	I915_WRITE(GEN6_RC_CONTROL,
 		   rc6_mask |
 		   GEN6_RC_CTL_EI_MODE(1) |
 		   GEN6_RC_CTL_HW_ENABLE);
 
-	if (IS_HASWELL(dev)) {
-		I915_WRITE(GEN6_RPNSWREQ,
-			   HSW_FREQUENCY(10));
-		I915_WRITE(GEN6_RC_VIDEO_FREQ,
-			   HSW_FREQUENCY(12));
-	} else {
-		I915_WRITE(GEN6_RPNSWREQ,
-			   GEN6_FREQUENCY(10) |
-			   GEN6_OFFSET(0) |
-			   GEN6_AGGRESSIVE_TURBO);
-		I915_WRITE(GEN6_RC_VIDEO_FREQ,
-			   GEN6_FREQUENCY(12));
-	}
-
-	I915_WRITE(GEN6_RP_DOWN_TIMEOUT, 1000000);
-	I915_WRITE(GEN6_RP_INTERRUPT_LIMITS,
-		   dev_priv->rps.max_delay << 24 |
-		   dev_priv->rps.min_delay << 16);
-
-	I915_WRITE(GEN6_RP_UP_THRESHOLD, 59400);
-	I915_WRITE(GEN6_RP_DOWN_THRESHOLD, 245000);
-	I915_WRITE(GEN6_RP_UP_EI, 66000);
-	I915_WRITE(GEN6_RP_DOWN_EI, 350000);
-
+	/* Power down if completely idle for over 50ms */
+	I915_WRITE(GEN6_RP_DOWN_TIMEOUT, 50000);
 	I915_WRITE(GEN6_RP_IDLE_HYSTERSIS, 10);
-	I915_WRITE(GEN6_RP_CONTROL,
-		   GEN6_RP_MEDIA_TURBO |
-		   GEN6_RP_MEDIA_HW_NORMAL_MODE |
-		   GEN6_RP_MEDIA_IS_GFX |
-		   GEN6_RP_ENABLE |
-		   GEN6_RP_UP_BUSY_AVG |
-		   (IS_HASWELL(dev) ? GEN7_RP_DOWN_IDLE_AVG : GEN6_RP_DOWN_IDLE_CONT));
 
 	ret = sandybridge_pcode_write(dev_priv, GEN6_PCODE_WRITE_MIN_FREQ_TABLE, 0);
 	if (!ret) {
@@ -3596,7 +3934,8 @@ static void gen6_enable_rps(struct drm_device *dev)
 		DRM_DEBUG_DRIVER("Failed to set the min frequency\n");
 	}
 
-	gen6_set_rps(dev_priv->dev, (gt_perf_status & 0xff00) >> 8);
+	dev_priv->rps.power = HIGH_POWER; /* force a reset */
+	gen6_set_rps(dev_priv->dev, dev_priv->rps.min_delay);
 
 	gen6_enable_rps_interrupts(dev);
 
@@ -3624,23 +3963,28 @@ void gen6_update_ring_freq(struct drm_device *dev)
 	unsigned int gpu_freq;
 	unsigned int max_ia_freq, min_ring_freq;
 	int scaling_factor = 180;
+	struct cpufreq_policy *policy;
 
 	WARN_ON(!mutex_is_locked(&dev_priv->rps.hw_lock));
 
-	max_ia_freq = cpufreq_quick_get_max(0);
-	/*
-	 * Default to measured freq if none found, PCU will ensure we don't go
-	 * over
-	 */
-	if (!max_ia_freq)
+	policy = cpufreq_cpu_get(0);
+	if (policy) {
+		max_ia_freq = policy->cpuinfo.max_freq;
+		cpufreq_cpu_put(policy);
+	} else {
+		/*
+		 * Default to measured freq if none found, PCU will ensure we
+		 * don't go over
+		 */
 		max_ia_freq = tsc_khz;
+	}
 
 	/* Convert from kHz to MHz */
 	max_ia_freq /= 1000;
 
-	min_ring_freq = I915_READ(MCHBAR_MIRROR_BASE_SNB + DCLK);
-	/* convert DDR frequency from units of 133.3MHz to bandwidth */
-	min_ring_freq = (2 * 4 * min_ring_freq + 2) / 3;
+	min_ring_freq = I915_READ(DCLK) & 0xf;
+	/* convert DDR frequency from units of 266.6MHz to bandwidth */
+	min_ring_freq = mult_frac(min_ring_freq, 8, 3);
 
 	/*
 	 * For each potential GPU frequency, load a ring frequency we'd like
@@ -3652,8 +3996,11 @@ void gen6_update_ring_freq(struct drm_device *dev)
 		int diff = dev_priv->rps.max_delay - gpu_freq;
 		unsigned int ia_freq = 0, ring_freq = 0;
 
-		if (IS_HASWELL(dev)) {
-			ring_freq = (gpu_freq * 5 + 3) / 4;
+		if (INTEL_INFO(dev)->gen >= 8) {
+			/* max(2 * GT, DDR). NB: GT is 50MHz units */
+			ring_freq = max(min_ring_freq, gpu_freq);
+		} else if (IS_HASWELL(dev)) {
+			ring_freq = mult_frac(gpu_freq, 5, 4);
 			ring_freq = max(min_ring_freq, ring_freq);
 			/* leave ia_freq as the default, chosen by cpufreq */
 		} else {
@@ -3709,24 +4056,6 @@ int valleyview_rps_min_freq(struct drm_i915_private *dev_priv)
 	return vlv_punit_read(dev_priv, PUNIT_REG_GPU_LFM) & 0xff;
 }
 
-static void vlv_rps_timer_work(struct work_struct *work)
-{
-	drm_i915_private_t *dev_priv = container_of(work, drm_i915_private_t,
-						    rps.vlv_work.work);
-
-	/*
-	 * Timer fired, we must be idle.  Drop to min voltage state.
-	 * Note: we use RPe here since it should match the
-	 * Vmin we were shooting for.  That should give us better
-	 * perf when we come back out of RC6 than if we used the
-	 * min freq available.
-	 */
-	mutex_lock(&dev_priv->rps.hw_lock);
-	if (dev_priv->rps.cur_delay > dev_priv->rps.rpe_delay)
-		valleyview_set_rps(dev_priv->dev, dev_priv->rps.rpe_delay);
-	mutex_unlock(&dev_priv->rps.hw_lock);
-}
-
 static void valleyview_setup_pctx(struct drm_device *dev)
 {
 	struct drm_i915_private *dev_priv = dev->dev_private;
@@ -3773,13 +4102,14 @@ static void valleyview_enable_rps(struct drm_device *dev)
 {
 	struct drm_i915_private *dev_priv = dev->dev_private;
 	struct intel_ring_buffer *ring;
-	u32 gtfifodbg, val;
+	u32 gtfifodbg, val, rc6_mode = 0;
 	int i;
 
 	WARN_ON(!mutex_is_locked(&dev_priv->rps.hw_lock));
 
 	if ((gtfifodbg = I915_READ(GTFIFODBG))) {
-		DRM_ERROR("GT fifo had a previous error %x\n", gtfifodbg);
+		DRM_DEBUG_DRIVER("GT fifo had a previous error %x\n",
+				 gtfifodbg);
 		I915_WRITE(GTFIFODBG, gtfifodbg);
 	}
 
@@ -3812,9 +4142,16 @@ static void valleyview_enable_rps(struct drm_device *dev)
 	I915_WRITE(GEN6_RC6_THRESHOLD, 0xc350);
 
 	/* allows RC6 residency counter to work */
-	I915_WRITE(0x138104, _MASKED_BIT_ENABLE(0x3));
-	I915_WRITE(GEN6_RC_CONTROL,
-		   GEN7_RC_CTL_TO_MODE);
+	I915_WRITE(VLV_COUNTER_CONTROL,
+		   _MASKED_BIT_ENABLE(VLV_COUNT_RANGE_HIGH |
+				      VLV_MEDIA_RC6_COUNT_EN |
+				      VLV_RENDER_RC6_COUNT_EN));
+	if (intel_enable_rc6(dev) & INTEL_RC6_ENABLE)
+		rc6_mode = GEN7_RC_CTL_TO_MODE;
+
+	intel_print_rc6_info(dev, rc6_mode);
+
+	I915_WRITE(GEN6_RC_CONTROL, rc6_mode);
 
 	val = vlv_punit_read(dev_priv, PUNIT_REG_GPU_FREQ_STS);
 	switch ((val >> 6) & 3) {
@@ -3985,6 +4322,8 @@ static void ironlake_enable_rc6(struct drm_device *dev)
 
 	I915_WRITE(PWRCTXA, i915_gem_obj_ggtt_offset(dev_priv->ips.pwrctx) | PWRCTX_EN);
 	I915_WRITE(RSTDBYCTL, I915_READ(RSTDBYCTL) & ~RCX_SW_EXIT);
+
+	intel_print_rc6_info(dev, INTEL_RC6_ENABLE);
 }
 
 static unsigned long intel_pxfreq(u32 vidfreq)
@@ -4603,13 +4942,12 @@ void intel_disable_gt_powersave(struct drm_device *dev)
 	} else if (INTEL_INFO(dev)->gen >= 6) {
 		cancel_delayed_work_sync(&dev_priv->rps.delayed_resume_work);
 		cancel_work_sync(&dev_priv->rps.work);
-		if (IS_VALLEYVIEW(dev))
-			cancel_delayed_work_sync(&dev_priv->rps.vlv_work);
 		mutex_lock(&dev_priv->rps.hw_lock);
 		if (IS_VALLEYVIEW(dev))
 			valleyview_disable_rps(dev);
 		else
 			gen6_disable_rps(dev);
+		dev_priv->rps.enabled = false;
 		mutex_unlock(&dev_priv->rps.hw_lock);
 	}
 }
@@ -4625,10 +4963,14 @@ static void intel_gen6_powersave_work(struct work_struct *work)
 
 	if (IS_VALLEYVIEW(dev)) {
 		valleyview_enable_rps(dev);
+	} else if (IS_BROADWELL(dev)) {
+		gen8_enable_rps(dev);
+		gen6_update_ring_freq(dev);
 	} else {
 		gen6_enable_rps(dev);
 		gen6_update_ring_freq(dev);
 	}
+	dev_priv->rps.enabled = true;
 	mutex_unlock(&dev_priv->rps.hw_lock);
 }
 
@@ -4672,7 +5014,7 @@ static void g4x_disable_trickle_feed(struct drm_device *dev)
 		I915_WRITE(DSPCNTR(pipe),
 			   I915_READ(DSPCNTR(pipe)) |
 			   DISPPLANE_TRICKLE_FEED_DISABLE);
-		intel_flush_display_plane(dev_priv, pipe);
+		intel_flush_primary_plane(dev_priv, pipe);
 	}
 }
 
@@ -4932,6 +5274,50 @@ static void lpt_suspend_hw(struct drm_device *dev)
 	}
 }
 
+static void gen8_init_clock_gating(struct drm_device *dev)
+{
+	struct drm_i915_private *dev_priv = dev->dev_private;
+	enum pipe i;
+
+	I915_WRITE(WM3_LP_ILK, 0);
+	I915_WRITE(WM2_LP_ILK, 0);
+	I915_WRITE(WM1_LP_ILK, 0);
+
+	/* FIXME(BDW): Check all the w/a, some might only apply to
+	 * pre-production hw. */
+
+	WARN(!i915_preliminary_hw_support,
+	     "GEN8_CENTROID_PIXEL_OPT_DIS not be needed for production\n");
+	I915_WRITE(HALF_SLICE_CHICKEN3,
+		   _MASKED_BIT_ENABLE(GEN8_CENTROID_PIXEL_OPT_DIS));
+	I915_WRITE(HALF_SLICE_CHICKEN3,
+		   _MASKED_BIT_ENABLE(GEN8_SAMPLER_POWER_BYPASS_DIS));
+	I915_WRITE(GAMTARBMODE, _MASKED_BIT_ENABLE(ARB_MODE_BWGTLB_DISABLE));
+
+	I915_WRITE(_3D_CHICKEN3,
+		   _3D_CHICKEN_SDE_LIMIT_FIFO_POLY_DEPTH(2));
+
+	I915_WRITE(COMMON_SLICE_CHICKEN2,
+		   _MASKED_BIT_ENABLE(GEN8_CSC2_SBE_VUE_CACHE_CONSERVATIVE));
+
+	I915_WRITE(GEN7_HALF_SLICE_CHICKEN1,
+		   _MASKED_BIT_ENABLE(GEN7_SINGLE_SUBSCAN_DISPATCH_ENABLE));
+
+	/* WaSwitchSolVfFArbitrationPriority */
+	I915_WRITE(GAM_ECOCHK, I915_READ(GAM_ECOCHK) | HSW_ECOCHK_ARB_PRIO_SOL);
+
+	/* WaPsrDPAMaskVBlankInSRD */
+	I915_WRITE(CHICKEN_PAR1_1,
+		   I915_READ(CHICKEN_PAR1_1) | DPA_MASK_VBLANK_SRD);
+
+	/* WaPsrDPRSUnmaskVBlankInSRD */
+	for_each_pipe(i) {
+		I915_WRITE(CHICKEN_PIPESL_1(i),
+			   I915_READ(CHICKEN_PIPESL_1(i) |
+				     DPRS_MASK_VBLANK_SRD));
+	}
+}
+
 static void haswell_init_clock_gating(struct drm_device *dev)
 {
 	struct drm_i915_private *dev_priv = dev->dev_private;
@@ -5255,6 +5641,25 @@ void intel_suspend_hw(struct drm_device *dev)
 		lpt_suspend_hw(dev);
 }
 
+static bool is_always_on_power_domain(struct drm_device *dev,
+				      enum intel_display_power_domain domain)
+{
+	unsigned long always_on_domains;
+
+	BUG_ON(BIT(domain) & ~POWER_DOMAIN_MASK);
+
+	if (IS_BROADWELL(dev)) {
+		always_on_domains = BDW_ALWAYS_ON_POWER_DOMAINS;
+	} else if (IS_HASWELL(dev)) {
+		always_on_domains = HSW_ALWAYS_ON_POWER_DOMAINS;
+	} else {
+		WARN_ON(1);
+		return true;
+	}
+
+	return BIT(domain) & always_on_domains;
+}
+
 /**
  * We should only use the power well if we explicitly asked the hardware to
  * enable it, so check if it's enabled and also check if we've requested it to
@@ -5268,23 +5673,11 @@ bool intel_display_power_enabled(struct drm_device *dev,
 	if (!HAS_POWER_WELL(dev))
 		return true;
 
-	switch (domain) {
-	case POWER_DOMAIN_PIPE_A:
-	case POWER_DOMAIN_TRANSCODER_EDP:
+	if (is_always_on_power_domain(dev, domain))
 		return true;
-	case POWER_DOMAIN_PIPE_B:
-	case POWER_DOMAIN_PIPE_C:
-	case POWER_DOMAIN_PIPE_A_PANEL_FITTER:
-	case POWER_DOMAIN_PIPE_B_PANEL_FITTER:
-	case POWER_DOMAIN_PIPE_C_PANEL_FITTER:
-	case POWER_DOMAIN_TRANSCODER_A:
-	case POWER_DOMAIN_TRANSCODER_B:
-	case POWER_DOMAIN_TRANSCODER_C:
-		return I915_READ(HSW_PWR_WELL_DRIVER) ==
+
+	return I915_READ(HSW_PWR_WELL_DRIVER) ==
 		     (HSW_PWR_WELL_ENABLE_REQUEST | HSW_PWR_WELL_STATE_ENABLED);
-	default:
-		BUG();
-	}
 }
 
 static void __intel_set_power_well(struct drm_device *dev, bool enable)
@@ -5328,83 +5721,136 @@ static void __intel_set_power_well(struct drm_device *dev, bool enable)
 			spin_lock_irqsave(&dev->vbl_lock, irqflags);
 			for_each_pipe(p)
 				if (p != PIPE_A)
-					dev->last_vblank[p] = 0;
+					dev->vblank[p].last = 0;
 			spin_unlock_irqrestore(&dev->vbl_lock, irqflags);
 		}
 	}
 }
 
-static struct i915_power_well *hsw_pwr;
+static void __intel_power_well_get(struct drm_device *dev,
+				   struct i915_power_well *power_well)
+{
+	if (!power_well->count++)
+		__intel_set_power_well(dev, true);
+}
+
+static void __intel_power_well_put(struct drm_device *dev,
+				   struct i915_power_well *power_well)
+{
+	WARN_ON(!power_well->count);
+	if (!--power_well->count && i915_disable_power_well)
+		__intel_set_power_well(dev, false);
+}
+
+void intel_display_power_get(struct drm_device *dev,
+			     enum intel_display_power_domain domain)
+{
+	struct drm_i915_private *dev_priv = dev->dev_private;
+	struct i915_power_domains *power_domains;
+
+	if (!HAS_POWER_WELL(dev))
+		return;
+
+	if (is_always_on_power_domain(dev, domain))
+		return;
+
+	power_domains = &dev_priv->power_domains;
+
+	mutex_lock(&power_domains->lock);
+	__intel_power_well_get(dev, &power_domains->power_wells[0]);
+	mutex_unlock(&power_domains->lock);
+}
+
+void intel_display_power_put(struct drm_device *dev,
+			     enum intel_display_power_domain domain)
+{
+	struct drm_i915_private *dev_priv = dev->dev_private;
+	struct i915_power_domains *power_domains;
+
+	if (!HAS_POWER_WELL(dev))
+		return;
+
+	if (is_always_on_power_domain(dev, domain))
+		return;
+
+	power_domains = &dev_priv->power_domains;
+
+	mutex_lock(&power_domains->lock);
+	__intel_power_well_put(dev, &power_domains->power_wells[0]);
+	mutex_unlock(&power_domains->lock);
+}
+
+static struct i915_power_domains *hsw_pwr;
 
 /* Display audio driver power well request */
 void i915_request_power_well(void)
 {
+	struct drm_i915_private *dev_priv;
+
 	if (WARN_ON(!hsw_pwr))
 		return;
 
-	spin_lock_irq(&hsw_pwr->lock);
-	if (!hsw_pwr->count++ &&
-			!hsw_pwr->i915_request)
-		__intel_set_power_well(hsw_pwr->device, true);
-	spin_unlock_irq(&hsw_pwr->lock);
+	dev_priv = container_of(hsw_pwr, struct drm_i915_private,
+				power_domains);
+
+	mutex_lock(&hsw_pwr->lock);
+	__intel_power_well_get(dev_priv->dev, &hsw_pwr->power_wells[0]);
+	mutex_unlock(&hsw_pwr->lock);
 }
 EXPORT_SYMBOL_GPL(i915_request_power_well);
 
 /* Display audio driver power well release */
 void i915_release_power_well(void)
 {
+	struct drm_i915_private *dev_priv;
+
 	if (WARN_ON(!hsw_pwr))
 		return;
 
-	spin_lock_irq(&hsw_pwr->lock);
-	WARN_ON(!hsw_pwr->count);
-	if (!--hsw_pwr->count &&
-		       !hsw_pwr->i915_request)
-		__intel_set_power_well(hsw_pwr->device, false);
-	spin_unlock_irq(&hsw_pwr->lock);
+	dev_priv = container_of(hsw_pwr, struct drm_i915_private,
+				power_domains);
+
+	mutex_lock(&hsw_pwr->lock);
+	__intel_power_well_put(dev_priv->dev, &hsw_pwr->power_wells[0]);
+	mutex_unlock(&hsw_pwr->lock);
 }
 EXPORT_SYMBOL_GPL(i915_release_power_well);
 
-int i915_init_power_well(struct drm_device *dev)
+int intel_power_domains_init(struct drm_device *dev)
 {
 	struct drm_i915_private *dev_priv = dev->dev_private;
+	struct i915_power_domains *power_domains = &dev_priv->power_domains;
+	struct i915_power_well *power_well;
 
-	hsw_pwr = &dev_priv->power_well;
+	mutex_init(&power_domains->lock);
+	hsw_pwr = power_domains;
 
-	hsw_pwr->device = dev;
-	spin_lock_init(&hsw_pwr->lock);
-	hsw_pwr->count = 0;
+	power_well = &power_domains->power_wells[0];
+	power_well->count = 0;
 
 	return 0;
 }
 
-void i915_remove_power_well(struct drm_device *dev)
+void intel_power_domains_remove(struct drm_device *dev)
 {
 	hsw_pwr = NULL;
 }
 
-void intel_set_power_well(struct drm_device *dev, bool enable)
+static void intel_power_domains_resume(struct drm_device *dev)
 {
 	struct drm_i915_private *dev_priv = dev->dev_private;
-	struct i915_power_well *power_well = &dev_priv->power_well;
+	struct i915_power_domains *power_domains = &dev_priv->power_domains;
+	struct i915_power_well *power_well;
 
 	if (!HAS_POWER_WELL(dev))
 		return;
 
-	if (!i915_disable_power_well && !enable)
-		return;
+	mutex_lock(&power_domains->lock);
 
-	spin_lock_irq(&power_well->lock);
-	power_well->i915_request = enable;
+	power_well = &power_domains->power_wells[0];
+	__intel_set_power_well(dev, power_well->count > 0);
 
-	/* only reject "disable" power well request */
-	if (power_well->count && !enable) {
-		spin_unlock_irq(&power_well->lock);
-		return;
-	}
-
-	__intel_set_power_well(dev, enable);
-	spin_unlock_irq(&power_well->lock);
+	mutex_unlock(&power_domains->lock);
 }
 
 /*
@@ -5413,7 +5859,7 @@ void intel_set_power_well(struct drm_device *dev, bool enable)
  * to be enabled, and it will only be disabled if none of the registers is
  * requesting it to be enabled.
  */
-void intel_init_power_well(struct drm_device *dev)
+void intel_power_domains_init_hw(struct drm_device *dev)
 {
 	struct drm_i915_private *dev_priv = dev->dev_private;
 
@@ -5421,7 +5867,8 @@ void intel_init_power_well(struct drm_device *dev)
 		return;
 
 	/* For now, we need the power well to be always enabled. */
-	intel_set_power_well(dev, true);
+	intel_display_set_init_power(dev, true);
+	intel_power_domains_resume(dev);
 
 	/* We're taking over the BIOS, so clear any requests made by it since
 	 * the driver is in charge now. */
@@ -5525,6 +5972,8 @@ void intel_init_pm(struct drm_device *dev)
 				dev_priv->display.update_wm = NULL;
 			}
 			dev_priv->display.init_clock_gating = haswell_init_clock_gating;
+		} else if (INTEL_INFO(dev)->gen == 8) {
+			dev_priv->display.init_clock_gating = gen8_init_clock_gating;
 		} else
 			dev_priv->display.update_wm = NULL;
 	} else if (IS_VALLEYVIEW(dev)) {
@@ -5686,7 +6135,4 @@ void intel_pm_init(struct drm_device *dev)
 
 	INIT_DELAYED_WORK(&dev_priv->rps.delayed_resume_work,
 			  intel_gen6_powersave_work);
-
-	INIT_DELAYED_WORK(&dev_priv->rps.vlv_work, vlv_rps_timer_work);
 }
-
diff --git a/drivers/gpu/drm/i915/intel_ringbuffer.c b/drivers/gpu/drm/i915/intel_ringbuffer.c
index 460ee1026fca..b620337e6d67 100644
--- a/drivers/gpu/drm/i915/intel_ringbuffer.c
+++ b/drivers/gpu/drm/i915/intel_ringbuffer.c
@@ -41,6 +41,16 @@ static inline int ring_space(struct intel_ring_buffer *ring)
 	return space;
 }
 
+void __intel_ring_advance(struct intel_ring_buffer *ring)
+{
+	struct drm_i915_private *dev_priv = ring->dev->dev_private;
+
+	ring->tail &= ring->size - 1;
+	if (dev_priv->gpu_error.stop_rings & intel_ring_flag(ring))
+		return;
+	ring->write_tail(ring, ring->tail);
+}
+
 static int
 gen2_render_ring_flush(struct intel_ring_buffer *ring,
 		       u32	invalidate_domains,
@@ -350,6 +360,47 @@ gen7_render_ring_flush(struct intel_ring_buffer *ring,
 	return 0;
 }
 
+static int
+gen8_render_ring_flush(struct intel_ring_buffer *ring,
+		       u32 invalidate_domains, u32 flush_domains)
+{
+	u32 flags = 0;
+	u32 scratch_addr = ring->scratch.gtt_offset + 128;
+	int ret;
+
+	flags |= PIPE_CONTROL_CS_STALL;
+
+	if (flush_domains) {
+		flags |= PIPE_CONTROL_RENDER_TARGET_CACHE_FLUSH;
+		flags |= PIPE_CONTROL_DEPTH_CACHE_FLUSH;
+	}
+	if (invalidate_domains) {
+		flags |= PIPE_CONTROL_TLB_INVALIDATE;
+		flags |= PIPE_CONTROL_INSTRUCTION_CACHE_INVALIDATE;
+		flags |= PIPE_CONTROL_TEXTURE_CACHE_INVALIDATE;
+		flags |= PIPE_CONTROL_VF_CACHE_INVALIDATE;
+		flags |= PIPE_CONTROL_CONST_CACHE_INVALIDATE;
+		flags |= PIPE_CONTROL_STATE_CACHE_INVALIDATE;
+		flags |= PIPE_CONTROL_QW_WRITE;
+		flags |= PIPE_CONTROL_GLOBAL_GTT_IVB;
+	}
+
+	ret = intel_ring_begin(ring, 6);
+	if (ret)
+		return ret;
+
+	intel_ring_emit(ring, GFX_OP_PIPE_CONTROL(6));
+	intel_ring_emit(ring, flags);
+	intel_ring_emit(ring, scratch_addr);
+	intel_ring_emit(ring, 0);
+	intel_ring_emit(ring, 0);
+	intel_ring_emit(ring, 0);
+	intel_ring_advance(ring);
+
+	return 0;
+
+}
+
 static void ring_write_tail(struct intel_ring_buffer *ring,
 			    u32 value)
 {
@@ -385,8 +436,7 @@ static int init_ring_common(struct intel_ring_buffer *ring)
 	int ret = 0;
 	u32 head;
 
-	if (HAS_FORCE_WAKE(dev))
-		gen6_gt_force_wake_get(dev_priv);
+	gen6_gt_force_wake_get(dev_priv);
 
 	if (I915_NEED_GFX_HWS(dev))
 		intel_ring_setup_status_page(ring);
@@ -459,8 +509,7 @@ static int init_ring_common(struct intel_ring_buffer *ring)
 	memset(&ring->hangcheck, 0, sizeof(ring->hangcheck));
 
 out:
-	if (HAS_FORCE_WAKE(dev))
-		gen6_gt_force_wake_put(dev_priv);
+	gen6_gt_force_wake_put(dev_priv);
 
 	return ret;
 }
@@ -559,8 +608,8 @@ static int init_render_ring(struct intel_ring_buffer *ring)
 	if (INTEL_INFO(dev)->gen >= 6)
 		I915_WRITE(INSTPM, _MASKED_BIT_ENABLE(INSTPM_FORCE_ORDERING));
 
-	if (HAS_L3_GPU_CACHE(dev))
-		I915_WRITE_IMR(ring, ~GT_RENDER_L3_PARITY_ERROR_INTERRUPT);
+	if (HAS_L3_DPF(dev))
+		I915_WRITE_IMR(ring, ~GT_PARITY_ERROR(dev));
 
 	return ret;
 }
@@ -593,7 +642,7 @@ update_mboxes(struct intel_ring_buffer *ring,
 #define MBOX_UPDATE_DWORDS 4
 	intel_ring_emit(ring, MI_LOAD_REGISTER_IMM(1));
 	intel_ring_emit(ring, mmio_offset);
-	intel_ring_emit(ring, ring->outstanding_lazy_request);
+	intel_ring_emit(ring, ring->outstanding_lazy_seqno);
 	intel_ring_emit(ring, MI_NOOP);
 }
 
@@ -629,9 +678,9 @@ gen6_add_request(struct intel_ring_buffer *ring)
 
 	intel_ring_emit(ring, MI_STORE_DWORD_INDEX);
 	intel_ring_emit(ring, I915_GEM_HWS_INDEX << MI_STORE_DWORD_INDEX_SHIFT);
-	intel_ring_emit(ring, ring->outstanding_lazy_request);
+	intel_ring_emit(ring, ring->outstanding_lazy_seqno);
 	intel_ring_emit(ring, MI_USER_INTERRUPT);
-	intel_ring_advance(ring);
+	__intel_ring_advance(ring);
 
 	return 0;
 }
@@ -723,7 +772,7 @@ pc_render_add_request(struct intel_ring_buffer *ring)
 			PIPE_CONTROL_WRITE_FLUSH |
 			PIPE_CONTROL_TEXTURE_CACHE_INVALIDATE);
 	intel_ring_emit(ring, ring->scratch.gtt_offset | PIPE_CONTROL_GLOBAL_GTT);
-	intel_ring_emit(ring, ring->outstanding_lazy_request);
+	intel_ring_emit(ring, ring->outstanding_lazy_seqno);
 	intel_ring_emit(ring, 0);
 	PIPE_CONTROL_FLUSH(ring, scratch_addr);
 	scratch_addr += 128; /* write to separate cachelines */
@@ -742,9 +791,9 @@ pc_render_add_request(struct intel_ring_buffer *ring)
 			PIPE_CONTROL_TEXTURE_CACHE_INVALIDATE |
 			PIPE_CONTROL_NOTIFY);
 	intel_ring_emit(ring, ring->scratch.gtt_offset | PIPE_CONTROL_GLOBAL_GTT);
-	intel_ring_emit(ring, ring->outstanding_lazy_request);
+	intel_ring_emit(ring, ring->outstanding_lazy_seqno);
 	intel_ring_emit(ring, 0);
-	intel_ring_advance(ring);
+	__intel_ring_advance(ring);
 
 	return 0;
 }
@@ -963,9 +1012,9 @@ i9xx_add_request(struct intel_ring_buffer *ring)
 
 	intel_ring_emit(ring, MI_STORE_DWORD_INDEX);
 	intel_ring_emit(ring, I915_GEM_HWS_INDEX << MI_STORE_DWORD_INDEX_SHIFT);
-	intel_ring_emit(ring, ring->outstanding_lazy_request);
+	intel_ring_emit(ring, ring->outstanding_lazy_seqno);
 	intel_ring_emit(ring, MI_USER_INTERRUPT);
-	intel_ring_advance(ring);
+	__intel_ring_advance(ring);
 
 	return 0;
 }
@@ -987,10 +1036,10 @@ gen6_ring_get_irq(struct intel_ring_buffer *ring)
 
 	spin_lock_irqsave(&dev_priv->irq_lock, flags);
 	if (ring->irq_refcount++ == 0) {
-		if (HAS_L3_GPU_CACHE(dev) && ring->id == RCS)
+		if (HAS_L3_DPF(dev) && ring->id == RCS)
 			I915_WRITE_IMR(ring,
 				       ~(ring->irq_enable_mask |
-					 GT_RENDER_L3_PARITY_ERROR_INTERRUPT));
+					 GT_PARITY_ERROR(dev)));
 		else
 			I915_WRITE_IMR(ring, ~ring->irq_enable_mask);
 		ilk_enable_gt_irq(dev_priv, ring->irq_enable_mask);
@@ -1009,9 +1058,8 @@ gen6_ring_put_irq(struct intel_ring_buffer *ring)
 
 	spin_lock_irqsave(&dev_priv->irq_lock, flags);
 	if (--ring->irq_refcount == 0) {
-		if (HAS_L3_GPU_CACHE(dev) && ring->id == RCS)
-			I915_WRITE_IMR(ring,
-				       ~GT_RENDER_L3_PARITY_ERROR_INTERRUPT);
+		if (HAS_L3_DPF(dev) && ring->id == RCS)
+			I915_WRITE_IMR(ring, ~GT_PARITY_ERROR(dev));
 		else
 			I915_WRITE_IMR(ring, ~0);
 		ilk_disable_gt_irq(dev_priv, ring->irq_enable_mask);
@@ -1059,6 +1107,52 @@ hsw_vebox_put_irq(struct intel_ring_buffer *ring)
 	spin_unlock_irqrestore(&dev_priv->irq_lock, flags);
 }
 
+static bool
+gen8_ring_get_irq(struct intel_ring_buffer *ring)
+{
+	struct drm_device *dev = ring->dev;
+	struct drm_i915_private *dev_priv = dev->dev_private;
+	unsigned long flags;
+
+	if (!dev->irq_enabled)
+		return false;
+
+	spin_lock_irqsave(&dev_priv->irq_lock, flags);
+	if (ring->irq_refcount++ == 0) {
+		if (HAS_L3_DPF(dev) && ring->id == RCS) {
+			I915_WRITE_IMR(ring,
+				       ~(ring->irq_enable_mask |
+					 GT_RENDER_L3_PARITY_ERROR_INTERRUPT));
+		} else {
+			I915_WRITE_IMR(ring, ~ring->irq_enable_mask);
+		}
+		POSTING_READ(RING_IMR(ring->mmio_base));
+	}
+	spin_unlock_irqrestore(&dev_priv->irq_lock, flags);
+
+	return true;
+}
+
+static void
+gen8_ring_put_irq(struct intel_ring_buffer *ring)
+{
+	struct drm_device *dev = ring->dev;
+	struct drm_i915_private *dev_priv = dev->dev_private;
+	unsigned long flags;
+
+	spin_lock_irqsave(&dev_priv->irq_lock, flags);
+	if (--ring->irq_refcount == 0) {
+		if (HAS_L3_DPF(dev) && ring->id == RCS) {
+			I915_WRITE_IMR(ring,
+				       ~GT_RENDER_L3_PARITY_ERROR_INTERRUPT);
+		} else {
+			I915_WRITE_IMR(ring, ~0);
+		}
+		POSTING_READ(RING_IMR(ring->mmio_base));
+	}
+	spin_unlock_irqrestore(&dev_priv->irq_lock, flags);
+}
+
 static int
 i965_dispatch_execbuffer(struct intel_ring_buffer *ring,
 			 u32 offset, u32 length,
@@ -1317,7 +1411,7 @@ void intel_cleanup_ring_buffer(struct intel_ring_buffer *ring)
 	/* Disable the ring buffer. The ring must be idle at this point */
 	dev_priv = ring->dev->dev_private;
 	ret = intel_ring_idle(ring);
-	if (ret)
+	if (ret && !i915_reset_in_progress(&dev_priv->gpu_error))
 		DRM_ERROR("failed to quiesce %s whilst cleaning up: %d\n",
 			  ring->name, ret);
 
@@ -1328,6 +1422,8 @@ void intel_cleanup_ring_buffer(struct intel_ring_buffer *ring)
 	i915_gem_object_unpin(ring->obj);
 	drm_gem_object_unreference(&ring->obj->base);
 	ring->obj = NULL;
+	ring->preallocated_lazy_request = NULL;
+	ring->outstanding_lazy_seqno = 0;
 
 	if (ring->cleanup)
 		ring->cleanup(ring);
@@ -1414,6 +1510,9 @@ static int ring_wait_for_space(struct intel_ring_buffer *ring, int n)
 	if (ret != -ENOSPC)
 		return ret;
 
+	/* force the tail write in case we have been skipping them */
+	__intel_ring_advance(ring);
+
 	trace_i915_ring_wait_begin(ring);
 	/* With GEM the hangcheck timer should kick us out of the loop,
 	 * leaving it early runs the risk of corrupting GEM state (due
@@ -1475,7 +1574,7 @@ int intel_ring_idle(struct intel_ring_buffer *ring)
 	int ret;
 
 	/* We need to add any requests required to flush the objects and ring */
-	if (ring->outstanding_lazy_request) {
+	if (ring->outstanding_lazy_seqno) {
 		ret = i915_add_request(ring, NULL);
 		if (ret)
 			return ret;
@@ -1495,10 +1594,20 @@ int intel_ring_idle(struct intel_ring_buffer *ring)
 static int
 intel_ring_alloc_seqno(struct intel_ring_buffer *ring)
 {
-	if (ring->outstanding_lazy_request)
+	if (ring->outstanding_lazy_seqno)
 		return 0;
 
-	return i915_gem_get_seqno(ring->dev, &ring->outstanding_lazy_request);
+	if (ring->preallocated_lazy_request == NULL) {
+		struct drm_i915_gem_request *request;
+
+		request = kmalloc(sizeof(*request), GFP_KERNEL);
+		if (request == NULL)
+			return -ENOMEM;
+
+		ring->preallocated_lazy_request = request;
+	}
+
+	return i915_gem_get_seqno(ring->dev, &ring->outstanding_lazy_seqno);
 }
 
 static int __intel_ring_begin(struct intel_ring_buffer *ring,
@@ -1545,7 +1654,7 @@ void intel_ring_init_seqno(struct intel_ring_buffer *ring, u32 seqno)
 {
 	struct drm_i915_private *dev_priv = ring->dev->dev_private;
 
-	BUG_ON(ring->outstanding_lazy_request);
+	BUG_ON(ring->outstanding_lazy_seqno);
 
 	if (INTEL_INFO(ring->dev)->gen >= 6) {
 		I915_WRITE(RING_SYNC_0(ring->mmio_base), 0);
@@ -1558,17 +1667,6 @@ void intel_ring_init_seqno(struct intel_ring_buffer *ring, u32 seqno)
 	ring->hangcheck.seqno = seqno;
 }
 
-void intel_ring_advance(struct intel_ring_buffer *ring)
-{
-	struct drm_i915_private *dev_priv = ring->dev->dev_private;
-
-	ring->tail &= ring->size - 1;
-	if (dev_priv->gpu_error.stop_rings & intel_ring_flag(ring))
-		return;
-	ring->write_tail(ring, ring->tail);
-}
-
-
 static void gen6_bsd_ring_write_tail(struct intel_ring_buffer *ring,
 				     u32 value)
 {
@@ -1613,6 +1711,8 @@ static int gen6_bsd_ring_flush(struct intel_ring_buffer *ring,
 		return ret;
 
 	cmd = MI_FLUSH_DW;
+	if (INTEL_INFO(ring->dev)->gen >= 8)
+		cmd += 1;
 	/*
 	 * Bspec vol 1c.5 - video engine command streamer:
 	 * "If ENABLED, all TLBs will be invalidated once the flush
@@ -1624,9 +1724,38 @@ static int gen6_bsd_ring_flush(struct intel_ring_buffer *ring,
 			MI_FLUSH_DW_STORE_INDEX | MI_FLUSH_DW_OP_STOREDW;
 	intel_ring_emit(ring, cmd);
 	intel_ring_emit(ring, I915_GEM_HWS_SCRATCH_ADDR | MI_FLUSH_DW_USE_GTT);
+	if (INTEL_INFO(ring->dev)->gen >= 8) {
+		intel_ring_emit(ring, 0); /* upper addr */
+		intel_ring_emit(ring, 0); /* value */
+	} else  {
+		intel_ring_emit(ring, 0);
+		intel_ring_emit(ring, MI_NOOP);
+	}
+	intel_ring_advance(ring);
+	return 0;
+}
+
+static int
+gen8_ring_dispatch_execbuffer(struct intel_ring_buffer *ring,
+			      u32 offset, u32 len,
+			      unsigned flags)
+{
+	struct drm_i915_private *dev_priv = ring->dev->dev_private;
+	bool ppgtt = dev_priv->mm.aliasing_ppgtt != NULL &&
+		!(flags & I915_DISPATCH_SECURE);
+	int ret;
+
+	ret = intel_ring_begin(ring, 4);
+	if (ret)
+		return ret;
+
+	/* FIXME(BDW): Address space and security selectors. */
+	intel_ring_emit(ring, MI_BATCH_BUFFER_START_GEN8 | (ppgtt<<8));
+	intel_ring_emit(ring, offset);
 	intel_ring_emit(ring, 0);
 	intel_ring_emit(ring, MI_NOOP);
 	intel_ring_advance(ring);
+
 	return 0;
 }
 
@@ -1686,6 +1815,8 @@ static int gen6_ring_flush(struct intel_ring_buffer *ring,
 		return ret;
 
 	cmd = MI_FLUSH_DW;
+	if (INTEL_INFO(ring->dev)->gen >= 8)
+		cmd += 1;
 	/*
 	 * Bspec vol 1c.3 - blitter engine command streamer:
 	 * "If ENABLED, all TLBs will be invalidated once the flush
@@ -1697,8 +1828,13 @@ static int gen6_ring_flush(struct intel_ring_buffer *ring,
 			MI_FLUSH_DW_OP_STOREDW;
 	intel_ring_emit(ring, cmd);
 	intel_ring_emit(ring, I915_GEM_HWS_SCRATCH_ADDR | MI_FLUSH_DW_USE_GTT);
-	intel_ring_emit(ring, 0);
-	intel_ring_emit(ring, MI_NOOP);
+	if (INTEL_INFO(ring->dev)->gen >= 8) {
+		intel_ring_emit(ring, 0); /* upper addr */
+		intel_ring_emit(ring, 0); /* value */
+	} else  {
+		intel_ring_emit(ring, 0);
+		intel_ring_emit(ring, MI_NOOP);
+	}
 	intel_ring_advance(ring);
 
 	if (IS_GEN7(dev) && flush)
@@ -1721,8 +1857,14 @@ int intel_init_render_ring_buffer(struct drm_device *dev)
 		ring->flush = gen7_render_ring_flush;
 		if (INTEL_INFO(dev)->gen == 6)
 			ring->flush = gen6_render_ring_flush;
-		ring->irq_get = gen6_ring_get_irq;
-		ring->irq_put = gen6_ring_put_irq;
+		if (INTEL_INFO(dev)->gen >= 8) {
+			ring->flush = gen8_render_ring_flush;
+			ring->irq_get = gen8_ring_get_irq;
+			ring->irq_put = gen8_ring_put_irq;
+		} else {
+			ring->irq_get = gen6_ring_get_irq;
+			ring->irq_put = gen6_ring_put_irq;
+		}
 		ring->irq_enable_mask = GT_RENDER_USER_INTERRUPT;
 		ring->get_seqno = gen6_ring_get_seqno;
 		ring->set_seqno = ring_set_seqno;
@@ -1764,6 +1906,8 @@ int intel_init_render_ring_buffer(struct drm_device *dev)
 	ring->write_tail = ring_write_tail;
 	if (IS_HASWELL(dev))
 		ring->dispatch_execbuffer = hsw_ring_dispatch_execbuffer;
+	else if (IS_GEN8(dev))
+		ring->dispatch_execbuffer = gen8_ring_dispatch_execbuffer;
 	else if (INTEL_INFO(dev)->gen >= 6)
 		ring->dispatch_execbuffer = gen6_ring_dispatch_execbuffer;
 	else if (INTEL_INFO(dev)->gen >= 4)
@@ -1877,7 +2021,7 @@ int intel_init_bsd_ring_buffer(struct drm_device *dev)
 	ring->id = VCS;
 
 	ring->write_tail = ring_write_tail;
-	if (IS_GEN6(dev) || IS_GEN7(dev)) {
+	if (INTEL_INFO(dev)->gen >= 6) {
 		ring->mmio_base = GEN6_BSD_RING_BASE;
 		/* gen6 bsd needs a special wa for tail updates */
 		if (IS_GEN6(dev))
@@ -1886,10 +2030,20 @@ int intel_init_bsd_ring_buffer(struct drm_device *dev)
 		ring->add_request = gen6_add_request;
 		ring->get_seqno = gen6_ring_get_seqno;
 		ring->set_seqno = ring_set_seqno;
-		ring->irq_enable_mask = GT_BSD_USER_INTERRUPT;
-		ring->irq_get = gen6_ring_get_irq;
-		ring->irq_put = gen6_ring_put_irq;
-		ring->dispatch_execbuffer = gen6_ring_dispatch_execbuffer;
+		if (INTEL_INFO(dev)->gen >= 8) {
+			ring->irq_enable_mask =
+				GT_RENDER_USER_INTERRUPT << GEN8_VCS1_IRQ_SHIFT;
+			ring->irq_get = gen8_ring_get_irq;
+			ring->irq_put = gen8_ring_put_irq;
+			ring->dispatch_execbuffer =
+				gen8_ring_dispatch_execbuffer;
+		} else {
+			ring->irq_enable_mask = GT_BSD_USER_INTERRUPT;
+			ring->irq_get = gen6_ring_get_irq;
+			ring->irq_put = gen6_ring_put_irq;
+			ring->dispatch_execbuffer =
+				gen6_ring_dispatch_execbuffer;
+		}
 		ring->sync_to = gen6_ring_sync;
 		ring->semaphore_register[RCS] = MI_SEMAPHORE_SYNC_VR;
 		ring->semaphore_register[VCS] = MI_SEMAPHORE_SYNC_INVALID;
@@ -1935,10 +2089,18 @@ int intel_init_blt_ring_buffer(struct drm_device *dev)
 	ring->add_request = gen6_add_request;
 	ring->get_seqno = gen6_ring_get_seqno;
 	ring->set_seqno = ring_set_seqno;
-	ring->irq_enable_mask = GT_BLT_USER_INTERRUPT;
-	ring->irq_get = gen6_ring_get_irq;
-	ring->irq_put = gen6_ring_put_irq;
-	ring->dispatch_execbuffer = gen6_ring_dispatch_execbuffer;
+	if (INTEL_INFO(dev)->gen >= 8) {
+		ring->irq_enable_mask =
+			GT_RENDER_USER_INTERRUPT << GEN8_BCS_IRQ_SHIFT;
+		ring->irq_get = gen8_ring_get_irq;
+		ring->irq_put = gen8_ring_put_irq;
+		ring->dispatch_execbuffer = gen8_ring_dispatch_execbuffer;
+	} else {
+		ring->irq_enable_mask = GT_BLT_USER_INTERRUPT;
+		ring->irq_get = gen6_ring_get_irq;
+		ring->irq_put = gen6_ring_put_irq;
+		ring->dispatch_execbuffer = gen6_ring_dispatch_execbuffer;
+	}
 	ring->sync_to = gen6_ring_sync;
 	ring->semaphore_register[RCS] = MI_SEMAPHORE_SYNC_BR;
 	ring->semaphore_register[VCS] = MI_SEMAPHORE_SYNC_BV;
@@ -1967,10 +2129,19 @@ int intel_init_vebox_ring_buffer(struct drm_device *dev)
 	ring->add_request = gen6_add_request;
 	ring->get_seqno = gen6_ring_get_seqno;
 	ring->set_seqno = ring_set_seqno;
-	ring->irq_enable_mask = PM_VEBOX_USER_INTERRUPT;
-	ring->irq_get = hsw_vebox_get_irq;
-	ring->irq_put = hsw_vebox_put_irq;
-	ring->dispatch_execbuffer = gen6_ring_dispatch_execbuffer;
+
+	if (INTEL_INFO(dev)->gen >= 8) {
+		ring->irq_enable_mask =
+			GT_RENDER_USER_INTERRUPT << GEN8_VECS_IRQ_SHIFT;
+		ring->irq_get = gen8_ring_get_irq;
+		ring->irq_put = gen8_ring_put_irq;
+		ring->dispatch_execbuffer = gen8_ring_dispatch_execbuffer;
+	} else {
+		ring->irq_enable_mask = PM_VEBOX_USER_INTERRUPT;
+		ring->irq_get = hsw_vebox_get_irq;
+		ring->irq_put = hsw_vebox_put_irq;
+		ring->dispatch_execbuffer = gen6_ring_dispatch_execbuffer;
+	}
 	ring->sync_to = gen6_ring_sync;
 	ring->semaphore_register[RCS] = MI_SEMAPHORE_SYNC_VER;
 	ring->semaphore_register[VCS] = MI_SEMAPHORE_SYNC_VEV;
diff --git a/drivers/gpu/drm/i915/intel_ringbuffer.h b/drivers/gpu/drm/i915/intel_ringbuffer.h
index 68b1ca974d59..71a73f4fe252 100644
--- a/drivers/gpu/drm/i915/intel_ringbuffer.h
+++ b/drivers/gpu/drm/i915/intel_ringbuffer.h
@@ -34,6 +34,7 @@ struct  intel_hw_status_page {
 #define I915_WRITE_IMR(ring, val) I915_WRITE(RING_IMR((ring)->mmio_base), val)
 
 enum intel_ring_hangcheck_action {
+	HANGCHECK_IDLE = 0,
 	HANGCHECK_WAIT,
 	HANGCHECK_ACTIVE,
 	HANGCHECK_KICK,
@@ -140,7 +141,8 @@ struct  intel_ring_buffer {
 	/**
 	 * Do we have some not yet emitted requests outstanding?
 	 */
-	u32 outstanding_lazy_request;
+	struct drm_i915_gem_request *preallocated_lazy_request;
+	u32 outstanding_lazy_seqno;
 	bool gpu_caches_dirty;
 	bool fbc_dirty;
 
@@ -237,7 +239,12 @@ static inline void intel_ring_emit(struct intel_ring_buffer *ring,
 	iowrite32(data, ring->virtual_start + ring->tail);
 	ring->tail += 4;
 }
-void intel_ring_advance(struct intel_ring_buffer *ring);
+static inline void intel_ring_advance(struct intel_ring_buffer *ring)
+{
+	ring->tail &= ring->size - 1;
+}
+void __intel_ring_advance(struct intel_ring_buffer *ring);
+
 int __must_check intel_ring_idle(struct intel_ring_buffer *ring);
 void intel_ring_init_seqno(struct intel_ring_buffer *ring, u32 seqno);
 int intel_ring_flush_all_caches(struct intel_ring_buffer *ring);
@@ -258,8 +265,8 @@ static inline u32 intel_ring_get_tail(struct intel_ring_buffer *ring)
 
 static inline u32 intel_ring_get_seqno(struct intel_ring_buffer *ring)
 {
-	BUG_ON(ring->outstanding_lazy_request == 0);
-	return ring->outstanding_lazy_request;
+	BUG_ON(ring->outstanding_lazy_seqno == 0);
+	return ring->outstanding_lazy_seqno;
 }
 
 static inline void i915_trace_irq_get(struct intel_ring_buffer *ring, u32 seqno)
diff --git a/drivers/gpu/drm/i915/intel_sdvo.c b/drivers/gpu/drm/i915/intel_sdvo.c
index 49482fd5b76c..a583e8f718a7 100644
--- a/drivers/gpu/drm/i915/intel_sdvo.c
+++ b/drivers/gpu/drm/i915/intel_sdvo.c
@@ -539,7 +539,7 @@ static bool intel_sdvo_read_response(struct intel_sdvo *intel_sdvo,
 		goto log_fail;
 
 	while ((status == SDVO_CMD_STATUS_PENDING ||
-			status == SDVO_CMD_STATUS_TARGET_NOT_SPECIFIED) && --retry) {
+		status == SDVO_CMD_STATUS_TARGET_NOT_SPECIFIED) && --retry) {
 		if (retry < 10)
 			msleep(15);
 		else
@@ -1068,7 +1068,7 @@ intel_sdvo_get_preferred_input_mode(struct intel_sdvo *intel_sdvo,
 
 static void i9xx_adjust_sdvo_tv_clock(struct intel_crtc_config *pipe_config)
 {
-	unsigned dotclock = pipe_config->adjusted_mode.clock;
+	unsigned dotclock = pipe_config->port_clock;
 	struct dpll *clock = &pipe_config->dpll;
 
 	/* SDVO TV has fixed PLL values depend on its clock range,
@@ -1133,7 +1133,6 @@ static bool intel_sdvo_compute_config(struct intel_encoder *encoder,
 	 */
 	pipe_config->pixel_multiplier =
 		intel_sdvo_get_pixel_multiplier(adjusted_mode);
-	adjusted_mode->clock *= pipe_config->pixel_multiplier;
 
 	if (intel_sdvo->color_range_auto) {
 		/* See CEA-861-E - 5.1 Default Encoding Parameters */
@@ -1217,11 +1216,7 @@ static void intel_sdvo_mode_set(struct intel_encoder *intel_encoder)
 	    !intel_sdvo_set_tv_format(intel_sdvo))
 		return;
 
-	/* We have tried to get input timing in mode_fixup, and filled into
-	 * adjusted_mode.
-	 */
 	intel_sdvo_get_dtd_from_mode(&input_dtd, adjusted_mode);
-	input_dtd.part1.clock /= crtc->config.pixel_multiplier;
 
 	if (intel_sdvo->is_tv || intel_sdvo->is_lvds)
 		input_dtd.part2.sdvo_flags = intel_sdvo->dtd_sdvo_flags;
@@ -1330,6 +1325,7 @@ static void intel_sdvo_get_config(struct intel_encoder *encoder,
 	struct intel_sdvo *intel_sdvo = to_sdvo(encoder);
 	struct intel_sdvo_dtd dtd;
 	int encoder_pixel_multiplier = 0;
+	int dotclock;
 	u32 flags = 0, sdvox;
 	u8 val;
 	bool ret;
@@ -1368,6 +1364,13 @@ static void intel_sdvo_get_config(struct intel_encoder *encoder,
 			 >> SDVO_PORT_MULTIPLY_SHIFT) + 1;
 	}
 
+	dotclock = pipe_config->port_clock / pipe_config->pixel_multiplier;
+
+	if (HAS_PCH_SPLIT(dev))
+		ironlake_check_encoder_dotclock(pipe_config, dotclock);
+
+	pipe_config->adjusted_mode.crtc_clock = dotclock;
+
 	/* Cross check the port pixel multiplier with the sdvo encoder state. */
 	if (intel_sdvo_get_value(intel_sdvo, SDVO_CMD_GET_CLOCK_RATE_MULT,
 				 &val, 1)) {
@@ -1770,6 +1773,9 @@ static void intel_sdvo_get_ddc_modes(struct drm_connector *connector)
 {
 	struct edid *edid;
 
+	DRM_DEBUG_KMS("[CONNECTOR:%d:%s]\n",
+		      connector->base.id, drm_get_connector_name(connector));
+
 	/* set the bus switch and get the modes */
 	edid = intel_sdvo_get_edid(connector);
 
@@ -1865,6 +1871,9 @@ static void intel_sdvo_get_tv_modes(struct drm_connector *connector)
 	uint32_t reply = 0, format_map = 0;
 	int i;
 
+	DRM_DEBUG_KMS("[CONNECTOR:%d:%s]\n",
+		      connector->base.id, drm_get_connector_name(connector));
+
 	/* Read the list of supported input resolutions for the selected TV
 	 * format.
 	 */
@@ -1899,6 +1908,9 @@ static void intel_sdvo_get_lvds_modes(struct drm_connector *connector)
 	struct drm_i915_private *dev_priv = connector->dev->dev_private;
 	struct drm_display_mode *newmode;
 
+	DRM_DEBUG_KMS("[CONNECTOR:%d:%s]\n",
+		      connector->base.id, drm_get_connector_name(connector));
+
 	/*
 	 * Fetch modes from VBT. For SDVO prefer the VBT mode since some
 	 * SDVO->LVDS transcoders can't cope with the EDID mode.
@@ -1930,7 +1942,6 @@ static void intel_sdvo_get_lvds_modes(struct drm_connector *connector)
 			break;
 		}
 	}
-
 }
 
 static int intel_sdvo_get_modes(struct drm_connector *connector)
@@ -1998,7 +2009,6 @@ static void intel_sdvo_destroy(struct drm_connector *connector)
 				     intel_sdvo_connector->tv_format);
 
 	intel_sdvo_destroy_enhance_property(connector);
-	drm_sysfs_connector_remove(connector);
 	drm_connector_cleanup(connector);
 	kfree(intel_sdvo_connector);
 }
@@ -2394,7 +2404,9 @@ intel_sdvo_dvi_init(struct intel_sdvo *intel_sdvo, int device)
 	struct intel_connector *intel_connector;
 	struct intel_sdvo_connector *intel_sdvo_connector;
 
-	intel_sdvo_connector = kzalloc(sizeof(struct intel_sdvo_connector), GFP_KERNEL);
+	DRM_DEBUG_KMS("initialising DVI device %d\n", device);
+
+	intel_sdvo_connector = kzalloc(sizeof(*intel_sdvo_connector), GFP_KERNEL);
 	if (!intel_sdvo_connector)
 		return false;
 
@@ -2442,7 +2454,9 @@ intel_sdvo_tv_init(struct intel_sdvo *intel_sdvo, int type)
 	struct intel_connector *intel_connector;
 	struct intel_sdvo_connector *intel_sdvo_connector;
 
-	intel_sdvo_connector = kzalloc(sizeof(struct intel_sdvo_connector), GFP_KERNEL);
+	DRM_DEBUG_KMS("initialising TV type %d\n", type);
+
+	intel_sdvo_connector = kzalloc(sizeof(*intel_sdvo_connector), GFP_KERNEL);
 	if (!intel_sdvo_connector)
 		return false;
 
@@ -2467,6 +2481,7 @@ intel_sdvo_tv_init(struct intel_sdvo *intel_sdvo, int type)
 	return true;
 
 err:
+	drm_sysfs_connector_remove(connector);
 	intel_sdvo_destroy(connector);
 	return false;
 }
@@ -2479,7 +2494,9 @@ intel_sdvo_analog_init(struct intel_sdvo *intel_sdvo, int device)
 	struct intel_connector *intel_connector;
 	struct intel_sdvo_connector *intel_sdvo_connector;
 
-	intel_sdvo_connector = kzalloc(sizeof(struct intel_sdvo_connector), GFP_KERNEL);
+	DRM_DEBUG_KMS("initialising analog device %d\n", device);
+
+	intel_sdvo_connector = kzalloc(sizeof(*intel_sdvo_connector), GFP_KERNEL);
 	if (!intel_sdvo_connector)
 		return false;
 
@@ -2510,7 +2527,9 @@ intel_sdvo_lvds_init(struct intel_sdvo *intel_sdvo, int device)
 	struct intel_connector *intel_connector;
 	struct intel_sdvo_connector *intel_sdvo_connector;
 
-	intel_sdvo_connector = kzalloc(sizeof(struct intel_sdvo_connector), GFP_KERNEL);
+	DRM_DEBUG_KMS("initialising LVDS device %d\n", device);
+
+	intel_sdvo_connector = kzalloc(sizeof(*intel_sdvo_connector), GFP_KERNEL);
 	if (!intel_sdvo_connector)
 		return false;
 
@@ -2534,6 +2553,7 @@ intel_sdvo_lvds_init(struct intel_sdvo *intel_sdvo, int device)
 	return true;
 
 err:
+	drm_sysfs_connector_remove(connector);
 	intel_sdvo_destroy(connector);
 	return false;
 }
@@ -2605,8 +2625,10 @@ static void intel_sdvo_output_cleanup(struct intel_sdvo *intel_sdvo)
 
 	list_for_each_entry_safe(connector, tmp,
 				 &dev->mode_config.connector_list, head) {
-		if (intel_attached_encoder(connector) == &intel_sdvo->base)
+		if (intel_attached_encoder(connector) == &intel_sdvo->base) {
+			drm_sysfs_connector_remove(connector);
 			intel_sdvo_destroy(connector);
+		}
 	}
 }
 
@@ -2876,7 +2898,7 @@ bool intel_sdvo_init(struct drm_device *dev, uint32_t sdvo_reg, bool is_sdvob)
 	struct intel_encoder *intel_encoder;
 	struct intel_sdvo *intel_sdvo;
 	int i;
-	intel_sdvo = kzalloc(sizeof(struct intel_sdvo), GFP_KERNEL);
+	intel_sdvo = kzalloc(sizeof(*intel_sdvo), GFP_KERNEL);
 	if (!intel_sdvo)
 		return false;
 
diff --git a/drivers/gpu/drm/i915/intel_sideband.c b/drivers/gpu/drm/i915/intel_sideband.c
index 9a0e6c5ea540..9944d8135e87 100644
--- a/drivers/gpu/drm/i915/intel_sideband.c
+++ b/drivers/gpu/drm/i915/intel_sideband.c
@@ -25,7 +25,10 @@
 #include "i915_drv.h"
 #include "intel_drv.h"
 
-/* IOSF sideband */
+/*
+ * IOSF sideband, see VLV2_SidebandMsg_HAS.docx and
+ * VLV_VLV2_PUNIT_HAS_0.8.docx
+ */
 static int vlv_sideband_rw(struct drm_i915_private *dev_priv, u32 devfn,
 			   u32 port, u32 opcode, u32 addr, u32 *val)
 {
@@ -101,19 +104,83 @@ u32 vlv_nc_read(struct drm_i915_private *dev_priv, u8 addr)
 	return val;
 }
 
-u32 vlv_dpio_read(struct drm_i915_private *dev_priv, int reg)
+u32 vlv_gpio_nc_read(struct drm_i915_private *dev_priv, u32 reg)
 {
 	u32 val = 0;
+	vlv_sideband_rw(dev_priv, PCI_DEVFN(2, 0), IOSF_PORT_GPIO_NC,
+			PUNIT_OPCODE_REG_READ, reg, &val);
+	return val;
+}
 
-	vlv_sideband_rw(dev_priv, DPIO_DEVFN, IOSF_PORT_DPIO,
-			DPIO_OPCODE_REG_READ, reg, &val);
+void vlv_gpio_nc_write(struct drm_i915_private *dev_priv, u32 reg, u32 val)
+{
+	vlv_sideband_rw(dev_priv, PCI_DEVFN(2, 0), IOSF_PORT_GPIO_NC,
+			PUNIT_OPCODE_REG_WRITE, reg, &val);
+}
+
+u32 vlv_cck_read(struct drm_i915_private *dev_priv, u32 reg)
+{
+	u32 val = 0;
+	vlv_sideband_rw(dev_priv, PCI_DEVFN(2, 0), IOSF_PORT_CCK,
+			PUNIT_OPCODE_REG_READ, reg, &val);
+	return val;
+}
+
+void vlv_cck_write(struct drm_i915_private *dev_priv, u32 reg, u32 val)
+{
+	vlv_sideband_rw(dev_priv, PCI_DEVFN(2, 0), IOSF_PORT_CCK,
+			PUNIT_OPCODE_REG_WRITE, reg, &val);
+}
+
+u32 vlv_ccu_read(struct drm_i915_private *dev_priv, u32 reg)
+{
+	u32 val = 0;
+	vlv_sideband_rw(dev_priv, PCI_DEVFN(2, 0), IOSF_PORT_CCU,
+			PUNIT_OPCODE_REG_READ, reg, &val);
+	return val;
+}
 
+void vlv_ccu_write(struct drm_i915_private *dev_priv, u32 reg, u32 val)
+{
+	vlv_sideband_rw(dev_priv, PCI_DEVFN(2, 0), IOSF_PORT_CCU,
+			PUNIT_OPCODE_REG_WRITE, reg, &val);
+}
+
+u32 vlv_gps_core_read(struct drm_i915_private *dev_priv, u32 reg)
+{
+	u32 val = 0;
+	vlv_sideband_rw(dev_priv, PCI_DEVFN(2, 0), IOSF_PORT_GPS_CORE,
+			PUNIT_OPCODE_REG_READ, reg, &val);
+	return val;
+}
+
+void vlv_gps_core_write(struct drm_i915_private *dev_priv, u32 reg, u32 val)
+{
+	vlv_sideband_rw(dev_priv, PCI_DEVFN(2, 0), IOSF_PORT_GPS_CORE,
+			PUNIT_OPCODE_REG_WRITE, reg, &val);
+}
+
+static u32 vlv_get_phy_port(enum pipe pipe)
+{
+	u32 port = IOSF_PORT_DPIO;
+
+	WARN_ON ((pipe != PIPE_A) && (pipe != PIPE_B));
+
+	return port;
+}
+
+u32 vlv_dpio_read(struct drm_i915_private *dev_priv, enum pipe pipe, int reg)
+{
+	u32 val = 0;
+
+	vlv_sideband_rw(dev_priv, DPIO_DEVFN, vlv_get_phy_port(pipe),
+			DPIO_OPCODE_REG_READ, reg, &val);
 	return val;
 }
 
-void vlv_dpio_write(struct drm_i915_private *dev_priv, int reg, u32 val)
+void vlv_dpio_write(struct drm_i915_private *dev_priv, enum pipe pipe, int reg, u32 val)
 {
-	vlv_sideband_rw(dev_priv, DPIO_DEVFN, IOSF_PORT_DPIO,
+	vlv_sideband_rw(dev_priv, DPIO_DEVFN, vlv_get_phy_port(pipe),
 			DPIO_OPCODE_REG_WRITE, reg, &val);
 }
 
diff --git a/drivers/gpu/drm/i915/intel_sprite.c b/drivers/gpu/drm/i915/intel_sprite.c
index ad6ec4b39005..b9fabf826f7d 100644
--- a/drivers/gpu/drm/i915/intel_sprite.c
+++ b/drivers/gpu/drm/i915/intel_sprite.c
@@ -260,14 +260,14 @@ ivb_update_plane(struct drm_plane *plane, struct drm_crtc *crtc,
 	if (obj->tiling_mode != I915_TILING_NONE)
 		sprctl |= SPRITE_TILED;
 
-	if (IS_HASWELL(dev))
+	if (IS_HASWELL(dev) || IS_BROADWELL(dev))
 		sprctl &= ~SPRITE_TRICKLE_FEED_DISABLE;
 	else
 		sprctl |= SPRITE_TRICKLE_FEED_DISABLE;
 
 	sprctl |= SPRITE_ENABLE;
 
-	if (IS_HASWELL(dev))
+	if (IS_HASWELL(dev) || IS_BROADWELL(dev))
 		sprctl |= SPRITE_PIPE_CSC_ENABLE;
 
 	intel_update_sprite_watermarks(plane, crtc, src_w, pixel_size, true,
@@ -288,7 +288,7 @@ ivb_update_plane(struct drm_plane *plane, struct drm_crtc *crtc,
 		dev_priv->sprite_scaling_enabled |= 1 << pipe;
 
 		if (!scaling_was_enabled) {
-			intel_update_watermarks(dev);
+			intel_update_watermarks(crtc);
 			intel_wait_for_vblank(dev, pipe);
 		}
 		sprscale = SPRITE_SCALE_ENABLE | (src_w << 16) | src_h;
@@ -306,7 +306,7 @@ ivb_update_plane(struct drm_plane *plane, struct drm_crtc *crtc,
 
 	/* HSW consolidates SPRTILEOFF and SPRLINOFF into a single SPROFFSET
 	 * register */
-	if (IS_HASWELL(dev))
+	if (IS_HASWELL(dev) || IS_BROADWELL(dev))
 		I915_WRITE(SPROFFSET(pipe), (y << 16) | x);
 	else if (obj->tiling_mode != I915_TILING_NONE)
 		I915_WRITE(SPRTILEOFF(pipe), (y << 16) | x);
@@ -323,7 +323,7 @@ ivb_update_plane(struct drm_plane *plane, struct drm_crtc *crtc,
 
 	/* potentially re-enable LP watermarks */
 	if (scaling_was_enabled && !dev_priv->sprite_scaling_enabled)
-		intel_update_watermarks(dev);
+		intel_update_watermarks(crtc);
 }
 
 static void
@@ -349,7 +349,7 @@ ivb_disable_plane(struct drm_plane *plane, struct drm_crtc *crtc)
 
 	/* potentially re-enable LP watermarks */
 	if (scaling_was_enabled && !dev_priv->sprite_scaling_enabled)
-		intel_update_watermarks(dev);
+		intel_update_watermarks(crtc);
 }
 
 static int
@@ -521,13 +521,28 @@ intel_enable_primary(struct drm_crtc *crtc)
 	struct intel_crtc *intel_crtc = to_intel_crtc(crtc);
 	int reg = DSPCNTR(intel_crtc->plane);
 
-	if (!intel_crtc->primary_disabled)
+	if (intel_crtc->primary_enabled)
 		return;
 
-	intel_crtc->primary_disabled = false;
-	intel_update_fbc(dev);
+	intel_crtc->primary_enabled = true;
 
 	I915_WRITE(reg, I915_READ(reg) | DISPLAY_PLANE_ENABLE);
+	intel_flush_primary_plane(dev_priv, intel_crtc->plane);
+
+	/*
+	 * FIXME IPS should be fine as long as one plane is
+	 * enabled, but in practice it seems to have problems
+	 * when going from primary only to sprite only and vice
+	 * versa.
+	 */
+	if (intel_crtc->config.ips_enabled) {
+		intel_wait_for_vblank(dev, intel_crtc->pipe);
+		hsw_enable_ips(intel_crtc);
+	}
+
+	mutex_lock(&dev->struct_mutex);
+	intel_update_fbc(dev);
+	mutex_unlock(&dev->struct_mutex);
 }
 
 static void
@@ -538,13 +553,26 @@ intel_disable_primary(struct drm_crtc *crtc)
 	struct intel_crtc *intel_crtc = to_intel_crtc(crtc);
 	int reg = DSPCNTR(intel_crtc->plane);
 
-	if (intel_crtc->primary_disabled)
+	if (!intel_crtc->primary_enabled)
 		return;
 
-	I915_WRITE(reg, I915_READ(reg) & ~DISPLAY_PLANE_ENABLE);
+	intel_crtc->primary_enabled = false;
 
-	intel_crtc->primary_disabled = true;
-	intel_update_fbc(dev);
+	mutex_lock(&dev->struct_mutex);
+	if (dev_priv->fbc.plane == intel_crtc->plane)
+		intel_disable_fbc(dev);
+	mutex_unlock(&dev->struct_mutex);
+
+	/*
+	 * FIXME IPS should be fine as long as one plane is
+	 * enabled, but in practice it seems to have problems
+	 * when going from primary only to sprite only and vice
+	 * versa.
+	 */
+	hsw_disable_ips(intel_crtc);
+
+	I915_WRITE(reg, I915_READ(reg) & ~DISPLAY_PLANE_ENABLE);
+	intel_flush_primary_plane(dev_priv, intel_crtc->plane);
 }
 
 static int
@@ -623,15 +651,12 @@ intel_update_plane(struct drm_plane *plane, struct drm_crtc *crtc,
 		   uint32_t src_w, uint32_t src_h)
 {
 	struct drm_device *dev = plane->dev;
-	struct drm_i915_private *dev_priv = dev->dev_private;
 	struct intel_crtc *intel_crtc = to_intel_crtc(crtc);
 	struct intel_plane *intel_plane = to_intel_plane(plane);
-	struct intel_framebuffer *intel_fb;
-	struct drm_i915_gem_object *obj, *old_obj;
-	int pipe = intel_plane->pipe;
-	enum transcoder cpu_transcoder = intel_pipe_to_cpu_transcoder(dev_priv,
-								      pipe);
-	int ret = 0;
+	struct intel_framebuffer *intel_fb = to_intel_framebuffer(fb);
+	struct drm_i915_gem_object *obj = intel_fb->obj;
+	struct drm_i915_gem_object *old_obj = intel_plane->obj;
+	int ret;
 	bool disable_primary = false;
 	bool visible;
 	int hscale, vscale;
@@ -652,29 +677,23 @@ intel_update_plane(struct drm_plane *plane, struct drm_crtc *crtc,
 		.y2 = crtc_y + crtc_h,
 	};
 	const struct drm_rect clip = {
-		.x2 = crtc->mode.hdisplay,
-		.y2 = crtc->mode.vdisplay,
+		.x2 = intel_crtc->active ? intel_crtc->config.pipe_src_w : 0,
+		.y2 = intel_crtc->active ? intel_crtc->config.pipe_src_h : 0,
+	};
+	const struct {
+		int crtc_x, crtc_y;
+		unsigned int crtc_w, crtc_h;
+		uint32_t src_x, src_y, src_w, src_h;
+	} orig = {
+		.crtc_x = crtc_x,
+		.crtc_y = crtc_y,
+		.crtc_w = crtc_w,
+		.crtc_h = crtc_h,
+		.src_x = src_x,
+		.src_y = src_y,
+		.src_w = src_w,
+		.src_h = src_h,
 	};
-
-	intel_fb = to_intel_framebuffer(fb);
-	obj = intel_fb->obj;
-
-	old_obj = intel_plane->obj;
-
-	intel_plane->crtc_x = crtc_x;
-	intel_plane->crtc_y = crtc_y;
-	intel_plane->crtc_w = crtc_w;
-	intel_plane->crtc_h = crtc_h;
-	intel_plane->src_x = src_x;
-	intel_plane->src_y = src_y;
-	intel_plane->src_w = src_w;
-	intel_plane->src_h = src_h;
-
-	/* Pipe must be running... */
-	if (!(I915_READ(PIPECONF(cpu_transcoder)) & PIPECONF_ENABLE)) {
-		DRM_DEBUG_KMS("Pipe disabled\n");
-		return -EINVAL;
-	}
 
 	/* Don't modify another pipe's plane */
 	if (intel_plane->pipe != intel_crtc->pipe) {
@@ -810,7 +829,7 @@ intel_update_plane(struct drm_plane *plane, struct drm_crtc *crtc,
 	 * we can disable the primary and save power.
 	 */
 	disable_primary = drm_rect_equals(&dst, &clip);
-	WARN_ON(disable_primary && !visible);
+	WARN_ON(disable_primary && !visible && intel_crtc->active);
 
 	mutex_lock(&dev->struct_mutex);
 
@@ -820,27 +839,40 @@ intel_update_plane(struct drm_plane *plane, struct drm_crtc *crtc,
 	 * the sprite planes only require 128KiB alignment and 32 PTE padding.
 	 */
 	ret = intel_pin_and_fence_fb_obj(dev, obj, NULL);
-	if (ret)
-		goto out_unlock;
 
-	intel_plane->obj = obj;
-
-	/*
-	 * Be sure to re-enable the primary before the sprite is no longer
-	 * covering it fully.
-	 */
-	if (!disable_primary)
-		intel_enable_primary(crtc);
+	mutex_unlock(&dev->struct_mutex);
 
-	if (visible)
-		intel_plane->update_plane(plane, crtc, fb, obj,
-					  crtc_x, crtc_y, crtc_w, crtc_h,
-					  src_x, src_y, src_w, src_h);
-	else
-		intel_plane->disable_plane(plane, crtc);
+	if (ret)
+		return ret;
+
+	intel_plane->crtc_x = orig.crtc_x;
+	intel_plane->crtc_y = orig.crtc_y;
+	intel_plane->crtc_w = orig.crtc_w;
+	intel_plane->crtc_h = orig.crtc_h;
+	intel_plane->src_x = orig.src_x;
+	intel_plane->src_y = orig.src_y;
+	intel_plane->src_w = orig.src_w;
+	intel_plane->src_h = orig.src_h;
+	intel_plane->obj = obj;
 
-	if (disable_primary)
-		intel_disable_primary(crtc);
+	if (intel_crtc->active) {
+		/*
+		 * Be sure to re-enable the primary before the sprite is no longer
+		 * covering it fully.
+		 */
+		if (!disable_primary)
+			intel_enable_primary(crtc);
+
+		if (visible)
+			intel_plane->update_plane(plane, crtc, fb, obj,
+						  crtc_x, crtc_y, crtc_w, crtc_h,
+						  src_x, src_y, src_w, src_h);
+		else
+			intel_plane->disable_plane(plane, crtc);
+
+		if (disable_primary)
+			intel_disable_primary(crtc);
+	}
 
 	/* Unpin old obj after new one is active to avoid ugliness */
 	if (old_obj) {
@@ -850,17 +882,15 @@ intel_update_plane(struct drm_plane *plane, struct drm_crtc *crtc,
 		 * wait for vblank to avoid ugliness, we only need to
 		 * do the pin & ref bookkeeping.
 		 */
-		if (old_obj != obj) {
-			mutex_unlock(&dev->struct_mutex);
-			intel_wait_for_vblank(dev, to_intel_crtc(crtc)->pipe);
-			mutex_lock(&dev->struct_mutex);
-		}
+		if (old_obj != obj && intel_crtc->active)
+			intel_wait_for_vblank(dev, intel_crtc->pipe);
+
+		mutex_lock(&dev->struct_mutex);
 		intel_unpin_fb_obj(old_obj);
+		mutex_unlock(&dev->struct_mutex);
 	}
 
-out_unlock:
-	mutex_unlock(&dev->struct_mutex);
-	return ret;
+	return 0;
 }
 
 static int
@@ -868,7 +898,7 @@ intel_disable_plane(struct drm_plane *plane)
 {
 	struct drm_device *dev = plane->dev;
 	struct intel_plane *intel_plane = to_intel_plane(plane);
-	int ret = 0;
+	struct intel_crtc *intel_crtc;
 
 	if (!plane->fb)
 		return 0;
@@ -876,21 +906,25 @@ intel_disable_plane(struct drm_plane *plane)
 	if (WARN_ON(!plane->crtc))
 		return -EINVAL;
 
-	intel_enable_primary(plane->crtc);
-	intel_plane->disable_plane(plane, plane->crtc);
+	intel_crtc = to_intel_crtc(plane->crtc);
 
-	if (!intel_plane->obj)
-		goto out;
+	if (intel_crtc->active) {
+		intel_enable_primary(plane->crtc);
+		intel_plane->disable_plane(plane, plane->crtc);
+	}
 
-	intel_wait_for_vblank(dev, intel_plane->pipe);
+	if (intel_plane->obj) {
+		if (intel_crtc->active)
+			intel_wait_for_vblank(dev, intel_plane->pipe);
 
-	mutex_lock(&dev->struct_mutex);
-	intel_unpin_fb_obj(intel_plane->obj);
-	intel_plane->obj = NULL;
-	mutex_unlock(&dev->struct_mutex);
-out:
+		mutex_lock(&dev->struct_mutex);
+		intel_unpin_fb_obj(intel_plane->obj);
+		mutex_unlock(&dev->struct_mutex);
 
-	return ret;
+		intel_plane->obj = NULL;
+	}
+
+	return 0;
 }
 
 static void intel_destroy_plane(struct drm_plane *plane)
@@ -921,7 +955,7 @@ int intel_sprite_set_colorkey(struct drm_device *dev, void *data,
 
 	obj = drm_mode_object_find(dev, set->plane_id, DRM_MODE_OBJECT_PLANE);
 	if (!obj) {
-		ret = -EINVAL;
+		ret = -ENOENT;
 		goto out_unlock;
 	}
 
@@ -950,7 +984,7 @@ int intel_sprite_get_colorkey(struct drm_device *dev, void *data,
 
 	obj = drm_mode_object_find(dev, get->plane_id, DRM_MODE_OBJECT_PLANE);
 	if (!obj) {
-		ret = -EINVAL;
+		ret = -ENOENT;
 		goto out_unlock;
 	}
 
@@ -1034,7 +1068,7 @@ intel_plane_init(struct drm_device *dev, enum pipe pipe, int plane)
 	if (INTEL_INFO(dev)->gen < 5)
 		return -ENODEV;
 
-	intel_plane = kzalloc(sizeof(struct intel_plane), GFP_KERNEL);
+	intel_plane = kzalloc(sizeof(*intel_plane), GFP_KERNEL);
 	if (!intel_plane)
 		return -ENOMEM;
 
@@ -1058,6 +1092,7 @@ intel_plane_init(struct drm_device *dev, enum pipe pipe, int plane)
 		break;
 
 	case 7:
+	case 8:
 		if (IS_IVYBRIDGE(dev)) {
 			intel_plane->can_scale = true;
 			intel_plane->max_downscale = 2;
diff --git a/drivers/gpu/drm/i915/intel_tv.c b/drivers/gpu/drm/i915/intel_tv.c
index dd6f84bf6c22..18c406246a2d 100644
--- a/drivers/gpu/drm/i915/intel_tv.c
+++ b/drivers/gpu/drm/i915/intel_tv.c
@@ -912,7 +912,7 @@ intel_tv_compute_config(struct intel_encoder *encoder,
 	if (!tv_mode)
 		return false;
 
-	pipe_config->adjusted_mode.clock = tv_mode->clock;
+	pipe_config->adjusted_mode.crtc_clock = tv_mode->clock;
 	DRM_DEBUG_KMS("forcing bpc to 8 for TV\n");
 	pipe_config->pipe_bpp = 8*3;
 
@@ -1044,7 +1044,7 @@ static void intel_tv_mode_set(struct intel_encoder *encoder)
 		tv_mode->dda3_inc << TV_SCDDA3_INC_SHIFT;
 
 	/* Enable two fixes for the chips that need them. */
-	if (dev->pci_device < 0x2772)
+	if (dev->pdev->device < 0x2772)
 		tv_ctl |= TV_ENC_C0_FIX | TV_ENC_SDP_FIX;
 
 	I915_WRITE(TV_H_CTL_1, hctl1);
@@ -1094,7 +1094,7 @@ static void intel_tv_mode_set(struct intel_encoder *encoder)
 		unsigned int xsize, ysize;
 		/* Pipe must be off here */
 		I915_WRITE(dspcntr_reg, dspcntr & ~DISPLAY_PLANE_ENABLE);
-		intel_flush_display_plane(dev_priv, intel_crtc->plane);
+		intel_flush_primary_plane(dev_priv, intel_crtc->plane);
 
 		/* Wait for vblank for the disable to take effect */
 		if (IS_GEN2(dev))
@@ -1123,7 +1123,7 @@ static void intel_tv_mode_set(struct intel_encoder *encoder)
 
 		I915_WRITE(pipeconf_reg, pipeconf);
 		I915_WRITE(dspcntr_reg, dspcntr);
-		intel_flush_display_plane(dev_priv, intel_crtc->plane);
+		intel_flush_primary_plane(dev_priv, intel_crtc->plane);
 	}
 
 	j = 0;
@@ -1433,7 +1433,6 @@ intel_tv_get_modes(struct drm_connector *connector)
 static void
 intel_tv_destroy(struct drm_connector *connector)
 {
-	drm_sysfs_connector_remove(connector);
 	drm_connector_cleanup(connector);
 	kfree(connector);
 }
@@ -1518,7 +1517,7 @@ static const struct drm_encoder_funcs intel_tv_enc_funcs = {
 static int tv_is_present_in_vbt(struct drm_device *dev)
 {
 	struct drm_i915_private *dev_priv = dev->dev_private;
-	struct child_device_config *p_child;
+	union child_device_config *p_child;
 	int i, ret;
 
 	if (!dev_priv->vbt.child_dev_num)
@@ -1530,13 +1529,13 @@ static int tv_is_present_in_vbt(struct drm_device *dev)
 		/*
 		 * If the device type is not TV, continue.
 		 */
-		if (p_child->device_type != DEVICE_TYPE_INT_TV &&
-			p_child->device_type != DEVICE_TYPE_TV)
+		if (p_child->old.device_type != DEVICE_TYPE_INT_TV &&
+			p_child->old.device_type != DEVICE_TYPE_TV)
 			continue;
 		/* Only when the addin_offset is non-zero, it is regarded
 		 * as present.
 		 */
-		if (p_child->addin_offset) {
+		if (p_child->old.addin_offset) {
 			ret = 1;
 			break;
 		}
@@ -1590,12 +1589,12 @@ intel_tv_init(struct drm_device *dev)
 	    (tv_dac_off & TVDAC_STATE_CHG_EN) != 0)
 		return;
 
-	intel_tv = kzalloc(sizeof(struct intel_tv), GFP_KERNEL);
+	intel_tv = kzalloc(sizeof(*intel_tv), GFP_KERNEL);
 	if (!intel_tv) {
 		return;
 	}
 
-	intel_connector = kzalloc(sizeof(struct intel_connector), GFP_KERNEL);
+	intel_connector = kzalloc(sizeof(*intel_connector), GFP_KERNEL);
 	if (!intel_connector) {
 		kfree(intel_tv);
 		return;
diff --git a/drivers/gpu/drm/i915/intel_uncore.c b/drivers/gpu/drm/i915/intel_uncore.c
index 8649f1c36b00..f9883ceff946 100644
--- a/drivers/gpu/drm/i915/intel_uncore.c
+++ b/drivers/gpu/drm/i915/intel_uncore.c
@@ -93,7 +93,7 @@ static void __gen6_gt_force_wake_mt_get(struct drm_i915_private *dev_priv)
 {
 	u32 forcewake_ack;
 
-	if (IS_HASWELL(dev_priv->dev))
+	if (IS_HASWELL(dev_priv->dev) || IS_GEN8(dev_priv->dev))
 		forcewake_ack = FORCEWAKE_ACK_HSW;
 	else
 		forcewake_ack = FORCEWAKE_MT_ACK;
@@ -112,7 +112,8 @@ static void __gen6_gt_force_wake_mt_get(struct drm_i915_private *dev_priv)
 		DRM_ERROR("Timed out waiting for forcewake to ack request.\n");
 
 	/* WaRsForcewakeWaitTC0:ivb,hsw */
-	__gen6_gt_wait_for_thread_c0(dev_priv);
+	if (INTEL_INFO(dev_priv->dev)->gen < 8)
+		__gen6_gt_wait_for_thread_c0(dev_priv);
 }
 
 static void gen6_gt_check_fifodbg(struct drm_i915_private *dev_priv)
@@ -204,60 +205,34 @@ static void vlv_force_wake_put(struct drm_i915_private *dev_priv)
 	gen6_gt_check_fifodbg(dev_priv);
 }
 
-void intel_uncore_early_sanitize(struct drm_device *dev)
+static void gen6_force_wake_work(struct work_struct *work)
 {
-	struct drm_i915_private *dev_priv = dev->dev_private;
+	struct drm_i915_private *dev_priv =
+		container_of(work, typeof(*dev_priv), uncore.force_wake_work.work);
+	unsigned long irqflags;
 
-	if (HAS_FPGA_DBG_UNCLAIMED(dev))
-		__raw_i915_write32(dev_priv, FPGA_DBG, FPGA_DBG_RM_NOCLAIM);
+	spin_lock_irqsave(&dev_priv->uncore.lock, irqflags);
+	if (--dev_priv->uncore.forcewake_count == 0)
+		dev_priv->uncore.funcs.force_wake_put(dev_priv);
+	spin_unlock_irqrestore(&dev_priv->uncore.lock, irqflags);
 }
 
-void intel_uncore_init(struct drm_device *dev)
+void intel_uncore_early_sanitize(struct drm_device *dev)
 {
 	struct drm_i915_private *dev_priv = dev->dev_private;
 
-	if (IS_VALLEYVIEW(dev)) {
-		dev_priv->uncore.funcs.force_wake_get = vlv_force_wake_get;
-		dev_priv->uncore.funcs.force_wake_put = vlv_force_wake_put;
-	} else if (IS_HASWELL(dev)) {
-		dev_priv->uncore.funcs.force_wake_get = __gen6_gt_force_wake_mt_get;
-		dev_priv->uncore.funcs.force_wake_put = __gen6_gt_force_wake_mt_put;
-	} else if (IS_IVYBRIDGE(dev)) {
-		u32 ecobus;
-
-		/* IVB configs may use multi-threaded forcewake */
-
-		/* A small trick here - if the bios hasn't configured
-		 * MT forcewake, and if the device is in RC6, then
-		 * force_wake_mt_get will not wake the device and the
-		 * ECOBUS read will return zero. Which will be
-		 * (correctly) interpreted by the test below as MT
-		 * forcewake being disabled.
-		 */
-		mutex_lock(&dev->struct_mutex);
-		__gen6_gt_force_wake_mt_get(dev_priv);
-		ecobus = __raw_i915_read32(dev_priv, ECOBUS);
-		__gen6_gt_force_wake_mt_put(dev_priv);
-		mutex_unlock(&dev->struct_mutex);
+	if (HAS_FPGA_DBG_UNCLAIMED(dev))
+		__raw_i915_write32(dev_priv, FPGA_DBG, FPGA_DBG_RM_NOCLAIM);
 
-		if (ecobus & FORCEWAKE_MT_ENABLE) {
-			dev_priv->uncore.funcs.force_wake_get =
-				__gen6_gt_force_wake_mt_get;
-			dev_priv->uncore.funcs.force_wake_put =
-				__gen6_gt_force_wake_mt_put;
-		} else {
-			DRM_INFO("No MT forcewake available on Ivybridge, this can result in issues\n");
-			DRM_INFO("when using vblank-synced partial screen updates.\n");
-			dev_priv->uncore.funcs.force_wake_get =
-				__gen6_gt_force_wake_get;
-			dev_priv->uncore.funcs.force_wake_put =
-				__gen6_gt_force_wake_put;
-		}
-	} else if (IS_GEN6(dev)) {
-		dev_priv->uncore.funcs.force_wake_get =
-			__gen6_gt_force_wake_get;
-		dev_priv->uncore.funcs.force_wake_put =
-			__gen6_gt_force_wake_put;
+	if (IS_HASWELL(dev) &&
+	    (__raw_i915_read32(dev_priv, HSW_EDRAM_PRESENT) == 1)) {
+		/* The docs do not explain exactly how the calculation can be
+		 * made. It is somewhat guessable, but for now, it's always
+		 * 128MB.
+		 * NB: We can't write IDICR yet because we do not have gt funcs
+		 * set up */
+		dev_priv->ellc_size = 128;
+		DRM_INFO("Found %zuMB of eLLC\n", dev_priv->ellc_size);
 	}
 }
 
@@ -276,10 +251,26 @@ static void intel_uncore_forcewake_reset(struct drm_device *dev)
 
 void intel_uncore_sanitize(struct drm_device *dev)
 {
+	struct drm_i915_private *dev_priv = dev->dev_private;
+	u32 reg_val;
+
 	intel_uncore_forcewake_reset(dev);
 
 	/* BIOS often leaves RC6 enabled, but disable it for hw init */
 	intel_disable_gt_powersave(dev);
+
+	/* Turn off power gate, require especially for the BIOS less system */
+	if (IS_VALLEYVIEW(dev)) {
+
+		mutex_lock(&dev_priv->rps.hw_lock);
+		reg_val = vlv_punit_read(dev_priv, PUNIT_REG_PWRGT_STATUS);
+
+		if (reg_val & (RENDER_PWRGT | MEDIA_PWRGT | DISP2D_PWRGT))
+			vlv_punit_write(dev_priv, PUNIT_REG_PWRGT_CTRL, 0x0);
+
+		mutex_unlock(&dev_priv->rps.hw_lock);
+
+	}
 }
 
 /*
@@ -292,6 +283,9 @@ void gen6_gt_force_wake_get(struct drm_i915_private *dev_priv)
 {
 	unsigned long irqflags;
 
+	if (!dev_priv->uncore.funcs.force_wake_get)
+		return;
+
 	spin_lock_irqsave(&dev_priv->uncore.lock, irqflags);
 	if (dev_priv->uncore.forcewake_count++ == 0)
 		dev_priv->uncore.funcs.force_wake_get(dev_priv);
@@ -305,17 +299,22 @@ void gen6_gt_force_wake_put(struct drm_i915_private *dev_priv)
 {
 	unsigned long irqflags;
 
+	if (!dev_priv->uncore.funcs.force_wake_put)
+		return;
+
 	spin_lock_irqsave(&dev_priv->uncore.lock, irqflags);
-	if (--dev_priv->uncore.forcewake_count == 0)
-		dev_priv->uncore.funcs.force_wake_put(dev_priv);
+	if (--dev_priv->uncore.forcewake_count == 0) {
+		dev_priv->uncore.forcewake_count++;
+		mod_delayed_work(dev_priv->wq,
+				 &dev_priv->uncore.force_wake_work,
+				 1);
+	}
 	spin_unlock_irqrestore(&dev_priv->uncore.lock, irqflags);
 }
 
 /* We give fast paths for the really cool registers */
 #define NEEDS_FORCE_WAKE(dev_priv, reg) \
-	((HAS_FORCE_WAKE((dev_priv)->dev)) && \
-	 ((reg) < 0x40000) &&            \
-	 ((reg) != FORCEWAKE))
+	 ((reg) < 0x40000 && (reg) != FORCEWAKE)
 
 static void
 ilk_dummy_write(struct drm_i915_private *dev_priv)
@@ -329,8 +328,7 @@ ilk_dummy_write(struct drm_i915_private *dev_priv)
 static void
 hsw_unclaimed_reg_clear(struct drm_i915_private *dev_priv, u32 reg)
 {
-	if (HAS_FPGA_DBG_UNCLAIMED(dev_priv->dev) &&
-	    (__raw_i915_read32(dev_priv, FPGA_DBG) & FPGA_DBG_RM_NOCLAIM)) {
+	if (__raw_i915_read32(dev_priv, FPGA_DBG) & FPGA_DBG_RM_NOCLAIM) {
 		DRM_ERROR("Unknown unclaimed register before writing to %x\n",
 			  reg);
 		__raw_i915_write32(dev_priv, FPGA_DBG, FPGA_DBG_RM_NOCLAIM);
@@ -340,20 +338,43 @@ hsw_unclaimed_reg_clear(struct drm_i915_private *dev_priv, u32 reg)
 static void
 hsw_unclaimed_reg_check(struct drm_i915_private *dev_priv, u32 reg)
 {
-	if (HAS_FPGA_DBG_UNCLAIMED(dev_priv->dev) &&
-	    (__raw_i915_read32(dev_priv, FPGA_DBG) & FPGA_DBG_RM_NOCLAIM)) {
+	if (__raw_i915_read32(dev_priv, FPGA_DBG) & FPGA_DBG_RM_NOCLAIM) {
 		DRM_ERROR("Unclaimed write to %x\n", reg);
 		__raw_i915_write32(dev_priv, FPGA_DBG, FPGA_DBG_RM_NOCLAIM);
 	}
 }
 
-#define __i915_read(x) \
-u##x i915_read##x(struct drm_i915_private *dev_priv, u32 reg, bool trace) { \
+#define REG_READ_HEADER(x) \
 	unsigned long irqflags; \
 	u##x val = 0; \
-	spin_lock_irqsave(&dev_priv->uncore.lock, irqflags); \
-	if (dev_priv->info->gen == 5) \
-		ilk_dummy_write(dev_priv); \
+	spin_lock_irqsave(&dev_priv->uncore.lock, irqflags)
+
+#define REG_READ_FOOTER \
+	spin_unlock_irqrestore(&dev_priv->uncore.lock, irqflags); \
+	trace_i915_reg_rw(false, reg, val, sizeof(val), trace); \
+	return val
+
+#define __gen4_read(x) \
+static u##x \
+gen4_read##x(struct drm_i915_private *dev_priv, off_t reg, bool trace) { \
+	REG_READ_HEADER(x); \
+	val = __raw_i915_read##x(dev_priv, reg); \
+	REG_READ_FOOTER; \
+}
+
+#define __gen5_read(x) \
+static u##x \
+gen5_read##x(struct drm_i915_private *dev_priv, off_t reg, bool trace) { \
+	REG_READ_HEADER(x); \
+	ilk_dummy_write(dev_priv); \
+	val = __raw_i915_read##x(dev_priv, reg); \
+	REG_READ_FOOTER; \
+}
+
+#define __gen6_read(x) \
+static u##x \
+gen6_read##x(struct drm_i915_private *dev_priv, off_t reg, bool trace) { \
+	REG_READ_HEADER(x); \
 	if (NEEDS_FORCE_WAKE((dev_priv), (reg))) { \
 		if (dev_priv->uncore.forcewake_count == 0) \
 			dev_priv->uncore.funcs.force_wake_get(dev_priv); \
@@ -363,28 +384,73 @@ u##x i915_read##x(struct drm_i915_private *dev_priv, u32 reg, bool trace) { \
 	} else { \
 		val = __raw_i915_read##x(dev_priv, reg); \
 	} \
+	REG_READ_FOOTER; \
+}
+
+__gen6_read(8)
+__gen6_read(16)
+__gen6_read(32)
+__gen6_read(64)
+__gen5_read(8)
+__gen5_read(16)
+__gen5_read(32)
+__gen5_read(64)
+__gen4_read(8)
+__gen4_read(16)
+__gen4_read(32)
+__gen4_read(64)
+
+#undef __gen6_read
+#undef __gen5_read
+#undef __gen4_read
+#undef REG_READ_FOOTER
+#undef REG_READ_HEADER
+
+#define REG_WRITE_HEADER \
+	unsigned long irqflags; \
+	trace_i915_reg_rw(true, reg, val, sizeof(val), trace); \
+	spin_lock_irqsave(&dev_priv->uncore.lock, irqflags)
+
+#define __gen4_write(x) \
+static void \
+gen4_write##x(struct drm_i915_private *dev_priv, off_t reg, u##x val, bool trace) { \
+	REG_WRITE_HEADER; \
+	__raw_i915_write##x(dev_priv, reg, val); \
 	spin_unlock_irqrestore(&dev_priv->uncore.lock, irqflags); \
-	trace_i915_reg_rw(false, reg, val, sizeof(val), trace); \
-	return val; \
 }
 
-__i915_read(8)
-__i915_read(16)
-__i915_read(32)
-__i915_read(64)
-#undef __i915_read
+#define __gen5_write(x) \
+static void \
+gen5_write##x(struct drm_i915_private *dev_priv, off_t reg, u##x val, bool trace) { \
+	REG_WRITE_HEADER; \
+	ilk_dummy_write(dev_priv); \
+	__raw_i915_write##x(dev_priv, reg, val); \
+	spin_unlock_irqrestore(&dev_priv->uncore.lock, irqflags); \
+}
 
-#define __i915_write(x) \
-void i915_write##x(struct drm_i915_private *dev_priv, u32 reg, u##x val, bool trace) { \
-	unsigned long irqflags; \
+#define __gen6_write(x) \
+static void \
+gen6_write##x(struct drm_i915_private *dev_priv, off_t reg, u##x val, bool trace) { \
 	u32 __fifo_ret = 0; \
-	trace_i915_reg_rw(true, reg, val, sizeof(val), trace); \
-	spin_lock_irqsave(&dev_priv->uncore.lock, irqflags); \
+	REG_WRITE_HEADER; \
+	if (NEEDS_FORCE_WAKE((dev_priv), (reg))) { \
+		__fifo_ret = __gen6_gt_wait_for_fifo(dev_priv); \
+	} \
+	__raw_i915_write##x(dev_priv, reg, val); \
+	if (unlikely(__fifo_ret)) { \
+		gen6_gt_check_fifodbg(dev_priv); \
+	} \
+	spin_unlock_irqrestore(&dev_priv->uncore.lock, irqflags); \
+}
+
+#define __hsw_write(x) \
+static void \
+hsw_write##x(struct drm_i915_private *dev_priv, off_t reg, u##x val, bool trace) { \
+	u32 __fifo_ret = 0; \
+	REG_WRITE_HEADER; \
 	if (NEEDS_FORCE_WAKE((dev_priv), (reg))) { \
 		__fifo_ret = __gen6_gt_wait_for_fifo(dev_priv); \
 	} \
-	if (dev_priv->info->gen == 5) \
-		ilk_dummy_write(dev_priv); \
 	hsw_unclaimed_reg_clear(dev_priv, reg); \
 	__raw_i915_write##x(dev_priv, reg, val); \
 	if (unlikely(__fifo_ret)) { \
@@ -393,11 +459,185 @@ void i915_write##x(struct drm_i915_private *dev_priv, u32 reg, u##x val, bool tr
 	hsw_unclaimed_reg_check(dev_priv, reg); \
 	spin_unlock_irqrestore(&dev_priv->uncore.lock, irqflags); \
 }
-__i915_write(8)
-__i915_write(16)
-__i915_write(32)
-__i915_write(64)
-#undef __i915_write
+
+static const u32 gen8_shadowed_regs[] = {
+	FORCEWAKE_MT,
+	GEN6_RPNSWREQ,
+	GEN6_RC_VIDEO_FREQ,
+	RING_TAIL(RENDER_RING_BASE),
+	RING_TAIL(GEN6_BSD_RING_BASE),
+	RING_TAIL(VEBOX_RING_BASE),
+	RING_TAIL(BLT_RING_BASE),
+	/* TODO: Other registers are not yet used */
+};
+
+static bool is_gen8_shadowed(struct drm_i915_private *dev_priv, u32 reg)
+{
+	int i;
+	for (i = 0; i < ARRAY_SIZE(gen8_shadowed_regs); i++)
+		if (reg == gen8_shadowed_regs[i])
+			return true;
+
+	return false;
+}
+
+#define __gen8_write(x) \
+static void \
+gen8_write##x(struct drm_i915_private *dev_priv, off_t reg, u##x val, bool trace) { \
+	bool __needs_put = !is_gen8_shadowed(dev_priv, reg); \
+	REG_WRITE_HEADER; \
+	if (__needs_put) { \
+		dev_priv->uncore.funcs.force_wake_get(dev_priv); \
+	} \
+	__raw_i915_write##x(dev_priv, reg, val); \
+	if (__needs_put) { \
+		dev_priv->uncore.funcs.force_wake_put(dev_priv); \
+	} \
+	spin_unlock_irqrestore(&dev_priv->uncore.lock, irqflags); \
+}
+
+__gen8_write(8)
+__gen8_write(16)
+__gen8_write(32)
+__gen8_write(64)
+__hsw_write(8)
+__hsw_write(16)
+__hsw_write(32)
+__hsw_write(64)
+__gen6_write(8)
+__gen6_write(16)
+__gen6_write(32)
+__gen6_write(64)
+__gen5_write(8)
+__gen5_write(16)
+__gen5_write(32)
+__gen5_write(64)
+__gen4_write(8)
+__gen4_write(16)
+__gen4_write(32)
+__gen4_write(64)
+
+#undef __gen8_write
+#undef __hsw_write
+#undef __gen6_write
+#undef __gen5_write
+#undef __gen4_write
+#undef REG_WRITE_HEADER
+
+void intel_uncore_init(struct drm_device *dev)
+{
+	struct drm_i915_private *dev_priv = dev->dev_private;
+
+	INIT_DELAYED_WORK(&dev_priv->uncore.force_wake_work,
+			  gen6_force_wake_work);
+
+	if (IS_VALLEYVIEW(dev)) {
+		dev_priv->uncore.funcs.force_wake_get = vlv_force_wake_get;
+		dev_priv->uncore.funcs.force_wake_put = vlv_force_wake_put;
+	} else if (IS_HASWELL(dev) || IS_GEN8(dev)) {
+		dev_priv->uncore.funcs.force_wake_get = __gen6_gt_force_wake_mt_get;
+		dev_priv->uncore.funcs.force_wake_put = __gen6_gt_force_wake_mt_put;
+	} else if (IS_IVYBRIDGE(dev)) {
+		u32 ecobus;
+
+		/* IVB configs may use multi-threaded forcewake */
+
+		/* A small trick here - if the bios hasn't configured
+		 * MT forcewake, and if the device is in RC6, then
+		 * force_wake_mt_get will not wake the device and the
+		 * ECOBUS read will return zero. Which will be
+		 * (correctly) interpreted by the test below as MT
+		 * forcewake being disabled.
+		 */
+		mutex_lock(&dev->struct_mutex);
+		__gen6_gt_force_wake_mt_get(dev_priv);
+		ecobus = __raw_i915_read32(dev_priv, ECOBUS);
+		__gen6_gt_force_wake_mt_put(dev_priv);
+		mutex_unlock(&dev->struct_mutex);
+
+		if (ecobus & FORCEWAKE_MT_ENABLE) {
+			dev_priv->uncore.funcs.force_wake_get =
+				__gen6_gt_force_wake_mt_get;
+			dev_priv->uncore.funcs.force_wake_put =
+				__gen6_gt_force_wake_mt_put;
+		} else {
+			DRM_INFO("No MT forcewake available on Ivybridge, this can result in issues\n");
+			DRM_INFO("when using vblank-synced partial screen updates.\n");
+			dev_priv->uncore.funcs.force_wake_get =
+				__gen6_gt_force_wake_get;
+			dev_priv->uncore.funcs.force_wake_put =
+				__gen6_gt_force_wake_put;
+		}
+	} else if (IS_GEN6(dev)) {
+		dev_priv->uncore.funcs.force_wake_get =
+			__gen6_gt_force_wake_get;
+		dev_priv->uncore.funcs.force_wake_put =
+			__gen6_gt_force_wake_put;
+	}
+
+	switch (INTEL_INFO(dev)->gen) {
+	default:
+		dev_priv->uncore.funcs.mmio_writeb  = gen8_write8;
+		dev_priv->uncore.funcs.mmio_writew  = gen8_write16;
+		dev_priv->uncore.funcs.mmio_writel  = gen8_write32;
+		dev_priv->uncore.funcs.mmio_writeq  = gen8_write64;
+		dev_priv->uncore.funcs.mmio_readb  = gen6_read8;
+		dev_priv->uncore.funcs.mmio_readw  = gen6_read16;
+		dev_priv->uncore.funcs.mmio_readl  = gen6_read32;
+		dev_priv->uncore.funcs.mmio_readq  = gen6_read64;
+		break;
+	case 7:
+	case 6:
+		if (IS_HASWELL(dev)) {
+			dev_priv->uncore.funcs.mmio_writeb  = hsw_write8;
+			dev_priv->uncore.funcs.mmio_writew  = hsw_write16;
+			dev_priv->uncore.funcs.mmio_writel  = hsw_write32;
+			dev_priv->uncore.funcs.mmio_writeq  = hsw_write64;
+		} else {
+			dev_priv->uncore.funcs.mmio_writeb  = gen6_write8;
+			dev_priv->uncore.funcs.mmio_writew  = gen6_write16;
+			dev_priv->uncore.funcs.mmio_writel  = gen6_write32;
+			dev_priv->uncore.funcs.mmio_writeq  = gen6_write64;
+		}
+		dev_priv->uncore.funcs.mmio_readb  = gen6_read8;
+		dev_priv->uncore.funcs.mmio_readw  = gen6_read16;
+		dev_priv->uncore.funcs.mmio_readl  = gen6_read32;
+		dev_priv->uncore.funcs.mmio_readq  = gen6_read64;
+		break;
+	case 5:
+		dev_priv->uncore.funcs.mmio_writeb  = gen5_write8;
+		dev_priv->uncore.funcs.mmio_writew  = gen5_write16;
+		dev_priv->uncore.funcs.mmio_writel  = gen5_write32;
+		dev_priv->uncore.funcs.mmio_writeq  = gen5_write64;
+		dev_priv->uncore.funcs.mmio_readb  = gen5_read8;
+		dev_priv->uncore.funcs.mmio_readw  = gen5_read16;
+		dev_priv->uncore.funcs.mmio_readl  = gen5_read32;
+		dev_priv->uncore.funcs.mmio_readq  = gen5_read64;
+		break;
+	case 4:
+	case 3:
+	case 2:
+		dev_priv->uncore.funcs.mmio_writeb  = gen4_write8;
+		dev_priv->uncore.funcs.mmio_writew  = gen4_write16;
+		dev_priv->uncore.funcs.mmio_writel  = gen4_write32;
+		dev_priv->uncore.funcs.mmio_writeq  = gen4_write64;
+		dev_priv->uncore.funcs.mmio_readb  = gen4_read8;
+		dev_priv->uncore.funcs.mmio_readw  = gen4_read16;
+		dev_priv->uncore.funcs.mmio_readl  = gen4_read32;
+		dev_priv->uncore.funcs.mmio_readq  = gen4_read64;
+		break;
+	}
+}
+
+void intel_uncore_fini(struct drm_device *dev)
+{
+	struct drm_i915_private *dev_priv = dev->dev_private;
+
+	flush_delayed_work(&dev_priv->uncore.force_wake_work);
+
+	/* Paranoia: make sure we have disabled everything before we exit. */
+	intel_uncore_sanitize(dev);
+}
 
 static const struct register_whitelist {
 	uint64_t offset;
@@ -445,36 +685,6 @@ int i915_reg_read_ioctl(struct drm_device *dev,
 	return 0;
 }
 
-static int i8xx_do_reset(struct drm_device *dev)
-{
-	struct drm_i915_private *dev_priv = dev->dev_private;
-
-	if (IS_I85X(dev))
-		return -ENODEV;
-
-	I915_WRITE(D_STATE, I915_READ(D_STATE) | DSTATE_GFX_RESET_I830);
-	POSTING_READ(D_STATE);
-
-	if (IS_I830(dev) || IS_845G(dev)) {
-		I915_WRITE(DEBUG_RESET_I830,
-			   DEBUG_RESET_DISPLAY |
-			   DEBUG_RESET_RENDER |
-			   DEBUG_RESET_FULL);
-		POSTING_READ(DEBUG_RESET_I830);
-		msleep(1);
-
-		I915_WRITE(DEBUG_RESET_I830, 0);
-		POSTING_READ(DEBUG_RESET_I830);
-	}
-
-	msleep(1);
-
-	I915_WRITE(D_STATE, I915_READ(D_STATE) & ~DSTATE_GFX_RESET_I830);
-	POSTING_READ(D_STATE);
-
-	return 0;
-}
-
 static int i965_reset_complete(struct drm_device *dev)
 {
 	u8 gdrst;
@@ -576,7 +786,6 @@ int intel_gpu_reset(struct drm_device *dev)
 	case 6: return gen6_do_reset(dev);
 	case 5: return ironlake_do_reset(dev);
 	case 4: return i965_do_reset(dev);
-	case 2: return i8xx_do_reset(dev);
 	default: return -ENODEV;
 	}
 }
diff --git a/drivers/gpu/drm/mga/mga_dma.c b/drivers/gpu/drm/mga/mga_dma.c
index cc3166dd445a..087db33f6cff 100644
--- a/drivers/gpu/drm/mga/mga_dma.c
+++ b/drivers/gpu/drm/mga/mga_dma.c
@@ -406,11 +406,6 @@ int mga_driver_load(struct drm_device *dev, unsigned long flags)
 	dev_priv->mmio_base = pci_resource_start(dev->pdev, 1);
 	dev_priv->mmio_size = pci_resource_len(dev->pdev, 1);
 
-	dev->counters += 3;
-	dev->types[6] = _DRM_STAT_IRQ;
-	dev->types[7] = _DRM_STAT_PRIMARY;
-	dev->types[8] = _DRM_STAT_SECONDARY;
-
 	ret = drm_vblank_init(dev, 1);
 
 	if (ret) {
diff --git a/drivers/gpu/drm/mga/mga_irq.c b/drivers/gpu/drm/mga/mga_irq.c
index 598c281def0a..2b0ceb8dc11b 100644
--- a/drivers/gpu/drm/mga/mga_irq.c
+++ b/drivers/gpu/drm/mga/mga_irq.c
@@ -169,5 +169,5 @@ void mga_driver_irq_uninstall(struct drm_device *dev)
 	/* Disable *all* interrupts */
 	MGA_WRITE(MGA_IEN, 0);
 
-	dev->irq_enabled = 0;
+	dev->irq_enabled = false;
 }
diff --git a/drivers/gpu/drm/mgag200/Kconfig b/drivers/gpu/drm/mgag200/Kconfig
index b487cdec5ee7..3a1c5fbae54a 100644
--- a/drivers/gpu/drm/mgag200/Kconfig
+++ b/drivers/gpu/drm/mgag200/Kconfig
@@ -5,6 +5,7 @@ config DRM_MGAG200
 	select FB_SYS_COPYAREA
 	select FB_SYS_IMAGEBLIT
 	select DRM_KMS_HELPER
+	select DRM_KMS_FB_HELPER
 	select DRM_TTM
 	help
 	 This is a KMS driver for the MGA G200 server chips, it
diff --git a/drivers/gpu/drm/mgag200/mgag200_drv.c b/drivers/gpu/drm/mgag200/mgag200_drv.c
index fcce7b2f8011..f15ea3c4a90a 100644
--- a/drivers/gpu/drm/mgag200/mgag200_drv.c
+++ b/drivers/gpu/drm/mgag200/mgag200_drv.c
@@ -99,7 +99,6 @@ static struct drm_driver driver = {
 	.minor = DRIVER_MINOR,
 	.patchlevel = DRIVER_PATCHLEVEL,
 
-	.gem_init_object = mgag200_gem_init_object,
 	.gem_free_object = mgag200_gem_free_object,
 	.dumb_create = mgag200_dumb_create,
 	.dumb_map_offset = mgag200_dumb_mmap_offset,
diff --git a/drivers/gpu/drm/mgag200/mgag200_drv.h b/drivers/gpu/drm/mgag200/mgag200_drv.h
index baaae19332e2..cf11ee68a6d9 100644
--- a/drivers/gpu/drm/mgag200/mgag200_drv.h
+++ b/drivers/gpu/drm/mgag200/mgag200_drv.h
@@ -260,7 +260,6 @@ int mgag200_driver_unload(struct drm_device *dev);
 int mgag200_gem_create(struct drm_device *dev,
 		   u32 size, bool iskernel,
 		       struct drm_gem_object **obj);
-int mgag200_gem_init_object(struct drm_gem_object *obj);
 int mgag200_dumb_create(struct drm_file *file,
 			struct drm_device *dev,
 			struct drm_mode_create_dumb *args);
diff --git a/drivers/gpu/drm/mgag200/mgag200_main.c b/drivers/gpu/drm/mgag200/mgag200_main.c
index 0f8b861b10b3..b1120cb1db6d 100644
--- a/drivers/gpu/drm/mgag200/mgag200_main.c
+++ b/drivers/gpu/drm/mgag200/mgag200_main.c
@@ -310,12 +310,6 @@ int mgag200_dumb_create(struct drm_file *file,
 	return 0;
 }
 
-int mgag200_gem_init_object(struct drm_gem_object *obj)
-{
-	BUG();
-	return 0;
-}
-
 void mgag200_bo_unref(struct mgag200_bo **bo)
 {
 	struct ttm_buffer_object *tbo;
diff --git a/drivers/gpu/drm/mgag200/mgag200_mode.c b/drivers/gpu/drm/mgag200/mgag200_mode.c
index 503a414cbdad..ee6ed633b7b1 100644
--- a/drivers/gpu/drm/mgag200/mgag200_mode.c
+++ b/drivers/gpu/drm/mgag200/mgag200_mode.c
@@ -765,8 +765,6 @@ static int mga_crtc_do_set_base(struct drm_crtc *crtc,
 	}
 	mgag200_bo_unreserve(bo);
 
-	DRM_INFO("mga base %llx\n", gpu_addr);
-
 	mga_set_start_address(crtc, (u32)gpu_addr);
 
 	return 0;
diff --git a/drivers/gpu/drm/msm/Kconfig b/drivers/gpu/drm/msm/Kconfig
index a06c19cc56f8..f39ab7554fc9 100644
--- a/drivers/gpu/drm/msm/Kconfig
+++ b/drivers/gpu/drm/msm/Kconfig
@@ -14,6 +14,7 @@ config DRM_MSM
 config DRM_MSM_FBDEV
 	bool "Enable legacy fbdev support for MSM modesetting driver"
 	depends on DRM_MSM
+	select DRM_KMS_FB_HELPER
 	select FB_SYS_FILLRECT
 	select FB_SYS_COPYAREA
 	select FB_SYS_IMAGEBLIT
diff --git a/drivers/gpu/drm/msm/Makefile b/drivers/gpu/drm/msm/Makefile
index e17914889e54..e5fa12b0d21e 100644
--- a/drivers/gpu/drm/msm/Makefile
+++ b/drivers/gpu/drm/msm/Makefile
@@ -21,6 +21,7 @@ msm-y := \
 	msm_drv.o \
 	msm_fb.o \
 	msm_gem.o \
+	msm_gem_prime.o \
 	msm_gem_submit.o \
 	msm_gpu.o \
 	msm_ringbuffer.o
diff --git a/drivers/gpu/drm/msm/adreno/a2xx.xml.h b/drivers/gpu/drm/msm/adreno/a2xx.xml.h
index 35463864b959..9588098741b5 100644
--- a/drivers/gpu/drm/msm/adreno/a2xx.xml.h
+++ b/drivers/gpu/drm/msm/adreno/a2xx.xml.h
@@ -4,16 +4,16 @@
 /* Autogenerated file, DO NOT EDIT manually!
 
 This file was generated by the rules-ng-ng headergen tool in this git repository:
-http://0x04.net/cgit/index.cgi/rules-ng-ng
-git clone git://0x04.net/rules-ng-ng
+http://github.com/freedreno/envytools/
+git clone https://github.com/freedreno/envytools.git
 
 The rules-ng-ng source files this header was generated from are:
 - /home/robclark/src/freedreno/envytools/rnndb/adreno.xml              (    327 bytes, from 2013-07-05 19:21:12)
 - /home/robclark/src/freedreno/envytools/rnndb/freedreno_copyright.xml (   1453 bytes, from 2013-03-31 16:51:27)
-- /home/robclark/src/freedreno/envytools/rnndb/a2xx/a2xx.xml           (  30005 bytes, from 2013-07-19 21:30:48)
+- /home/robclark/src/freedreno/envytools/rnndb/a2xx/a2xx.xml           (  31003 bytes, from 2013-09-19 18:50:16)
 - /home/robclark/src/freedreno/envytools/rnndb/adreno_common.xml       (   8983 bytes, from 2013-07-24 01:38:36)
-- /home/robclark/src/freedreno/envytools/rnndb/adreno_pm4.xml          (   9712 bytes, from 2013-05-26 15:22:37)
-- /home/robclark/src/freedreno/envytools/rnndb/a3xx/a3xx.xml           (  51415 bytes, from 2013-08-03 14:26:05)
+- /home/robclark/src/freedreno/envytools/rnndb/adreno_pm4.xml          (   9759 bytes, from 2013-09-10 00:52:33)
+- /home/robclark/src/freedreno/envytools/rnndb/a3xx/a3xx.xml           (  51983 bytes, from 2013-09-10 00:52:32)
 
 Copyright (C) 2013 by the following authors:
 - Rob Clark <robdclark@gmail.com> (robclark)
@@ -317,6 +317,38 @@ static inline uint32_t A2XX_RBBM_STATUS_CMDFIFO_AVAIL(uint32_t val)
 #define A2XX_RBBM_STATUS_RB_CNTX_BUSY				0x40000000
 #define A2XX_RBBM_STATUS_GUI_ACTIVE				0x80000000
 
+#define REG_A2XX_MH_ARBITER_CONFIG				0x00000a40
+#define A2XX_MH_ARBITER_CONFIG_SAME_PAGE_LIMIT__MASK		0x0000003f
+#define A2XX_MH_ARBITER_CONFIG_SAME_PAGE_LIMIT__SHIFT		0
+static inline uint32_t A2XX_MH_ARBITER_CONFIG_SAME_PAGE_LIMIT(uint32_t val)
+{
+	return ((val) << A2XX_MH_ARBITER_CONFIG_SAME_PAGE_LIMIT__SHIFT) & A2XX_MH_ARBITER_CONFIG_SAME_PAGE_LIMIT__MASK;
+}
+#define A2XX_MH_ARBITER_CONFIG_SAME_PAGE_GRANULARITY		0x00000040
+#define A2XX_MH_ARBITER_CONFIG_L1_ARB_ENABLE			0x00000080
+#define A2XX_MH_ARBITER_CONFIG_L1_ARB_HOLD_ENABLE		0x00000100
+#define A2XX_MH_ARBITER_CONFIG_L2_ARB_CONTROL			0x00000200
+#define A2XX_MH_ARBITER_CONFIG_PAGE_SIZE__MASK			0x00001c00
+#define A2XX_MH_ARBITER_CONFIG_PAGE_SIZE__SHIFT			10
+static inline uint32_t A2XX_MH_ARBITER_CONFIG_PAGE_SIZE(uint32_t val)
+{
+	return ((val) << A2XX_MH_ARBITER_CONFIG_PAGE_SIZE__SHIFT) & A2XX_MH_ARBITER_CONFIG_PAGE_SIZE__MASK;
+}
+#define A2XX_MH_ARBITER_CONFIG_TC_REORDER_ENABLE		0x00002000
+#define A2XX_MH_ARBITER_CONFIG_TC_ARB_HOLD_ENABLE		0x00004000
+#define A2XX_MH_ARBITER_CONFIG_IN_FLIGHT_LIMIT_ENABLE		0x00008000
+#define A2XX_MH_ARBITER_CONFIG_IN_FLIGHT_LIMIT__MASK		0x003f0000
+#define A2XX_MH_ARBITER_CONFIG_IN_FLIGHT_LIMIT__SHIFT		16
+static inline uint32_t A2XX_MH_ARBITER_CONFIG_IN_FLIGHT_LIMIT(uint32_t val)
+{
+	return ((val) << A2XX_MH_ARBITER_CONFIG_IN_FLIGHT_LIMIT__SHIFT) & A2XX_MH_ARBITER_CONFIG_IN_FLIGHT_LIMIT__MASK;
+}
+#define A2XX_MH_ARBITER_CONFIG_CP_CLNT_ENABLE			0x00400000
+#define A2XX_MH_ARBITER_CONFIG_VGT_CLNT_ENABLE			0x00800000
+#define A2XX_MH_ARBITER_CONFIG_TC_CLNT_ENABLE			0x01000000
+#define A2XX_MH_ARBITER_CONFIG_RB_CLNT_ENABLE			0x02000000
+#define A2XX_MH_ARBITER_CONFIG_PA_CLNT_ENABLE			0x04000000
+
 #define REG_A2XX_A220_VSC_BIN_SIZE				0x00000c01
 #define A2XX_A220_VSC_BIN_SIZE_WIDTH__MASK			0x0000001f
 #define A2XX_A220_VSC_BIN_SIZE_WIDTH__SHIFT			0
diff --git a/drivers/gpu/drm/msm/adreno/a3xx.xml.h b/drivers/gpu/drm/msm/adreno/a3xx.xml.h
index d183516067b4..d4afdf657559 100644
--- a/drivers/gpu/drm/msm/adreno/a3xx.xml.h
+++ b/drivers/gpu/drm/msm/adreno/a3xx.xml.h
@@ -4,16 +4,16 @@
 /* Autogenerated file, DO NOT EDIT manually!
 
 This file was generated by the rules-ng-ng headergen tool in this git repository:
-http://0x04.net/cgit/index.cgi/rules-ng-ng
-git clone git://0x04.net/rules-ng-ng
+http://github.com/freedreno/envytools/
+git clone https://github.com/freedreno/envytools.git
 
 The rules-ng-ng source files this header was generated from are:
 - /home/robclark/src/freedreno/envytools/rnndb/adreno.xml              (    327 bytes, from 2013-07-05 19:21:12)
 - /home/robclark/src/freedreno/envytools/rnndb/freedreno_copyright.xml (   1453 bytes, from 2013-03-31 16:51:27)
-- /home/robclark/src/freedreno/envytools/rnndb/a2xx/a2xx.xml           (  30005 bytes, from 2013-07-19 21:30:48)
+- /home/robclark/src/freedreno/envytools/rnndb/a2xx/a2xx.xml           (  31003 bytes, from 2013-09-19 18:50:16)
 - /home/robclark/src/freedreno/envytools/rnndb/adreno_common.xml       (   8983 bytes, from 2013-07-24 01:38:36)
-- /home/robclark/src/freedreno/envytools/rnndb/adreno_pm4.xml          (   9712 bytes, from 2013-05-26 15:22:37)
-- /home/robclark/src/freedreno/envytools/rnndb/a3xx/a3xx.xml           (  51415 bytes, from 2013-08-03 14:26:05)
+- /home/robclark/src/freedreno/envytools/rnndb/adreno_pm4.xml          (   9759 bytes, from 2013-09-10 00:52:33)
+- /home/robclark/src/freedreno/envytools/rnndb/a3xx/a3xx.xml           (  51983 bytes, from 2013-09-10 00:52:32)
 
 Copyright (C) 2013 by the following authors:
 - Rob Clark <robdclark@gmail.com> (robclark)
@@ -637,11 +637,12 @@ static inline uint32_t A3XX_GRAS_SU_POLY_OFFSET_OFFSET(float val)
 #define REG_A3XX_GRAS_SU_MODE_CONTROL				0x00002070
 #define A3XX_GRAS_SU_MODE_CONTROL_CULL_FRONT			0x00000001
 #define A3XX_GRAS_SU_MODE_CONTROL_CULL_BACK			0x00000002
-#define A3XX_GRAS_SU_MODE_CONTROL_LINEHALFWIDTH__MASK		0x000007fc
-#define A3XX_GRAS_SU_MODE_CONTROL_LINEHALFWIDTH__SHIFT		2
-static inline uint32_t A3XX_GRAS_SU_MODE_CONTROL_LINEHALFWIDTH(uint32_t val)
+#define A3XX_GRAS_SU_MODE_CONTROL_FRONT_CW			0x00000004
+#define A3XX_GRAS_SU_MODE_CONTROL_LINEHALFWIDTH__MASK		0x000007f8
+#define A3XX_GRAS_SU_MODE_CONTROL_LINEHALFWIDTH__SHIFT		3
+static inline uint32_t A3XX_GRAS_SU_MODE_CONTROL_LINEHALFWIDTH(float val)
 {
-	return ((val) << A3XX_GRAS_SU_MODE_CONTROL_LINEHALFWIDTH__SHIFT) & A3XX_GRAS_SU_MODE_CONTROL_LINEHALFWIDTH__MASK;
+	return ((((uint32_t)(val * 4.0))) << A3XX_GRAS_SU_MODE_CONTROL_LINEHALFWIDTH__SHIFT) & A3XX_GRAS_SU_MODE_CONTROL_LINEHALFWIDTH__MASK;
 }
 #define A3XX_GRAS_SU_MODE_CONTROL_POLY_OFFSET			0x00000800
 
@@ -745,6 +746,7 @@ static inline uint32_t A3XX_RB_RENDER_CONTROL_BIN_WIDTH(uint32_t val)
 }
 #define A3XX_RB_RENDER_CONTROL_DISABLE_COLOR_PIPE		0x00001000
 #define A3XX_RB_RENDER_CONTROL_ENABLE_GMEM			0x00002000
+#define A3XX_RB_RENDER_CONTROL_ALPHA_TEST			0x00400000
 #define A3XX_RB_RENDER_CONTROL_ALPHA_TEST_FUNC__MASK		0x07000000
 #define A3XX_RB_RENDER_CONTROL_ALPHA_TEST_FUNC__SHIFT		24
 static inline uint32_t A3XX_RB_RENDER_CONTROL_ALPHA_TEST_FUNC(enum adreno_compare_func val)
@@ -767,7 +769,19 @@ static inline uint32_t A3XX_RB_MSAA_CONTROL_SAMPLE_MASK(uint32_t val)
 	return ((val) << A3XX_RB_MSAA_CONTROL_SAMPLE_MASK__SHIFT) & A3XX_RB_MSAA_CONTROL_SAMPLE_MASK__MASK;
 }
 
-#define REG_A3XX_UNKNOWN_20C3					0x000020c3
+#define REG_A3XX_RB_ALPHA_REF					0x000020c3
+#define A3XX_RB_ALPHA_REF_UINT__MASK				0x0000ff00
+#define A3XX_RB_ALPHA_REF_UINT__SHIFT				8
+static inline uint32_t A3XX_RB_ALPHA_REF_UINT(uint32_t val)
+{
+	return ((val) << A3XX_RB_ALPHA_REF_UINT__SHIFT) & A3XX_RB_ALPHA_REF_UINT__MASK;
+}
+#define A3XX_RB_ALPHA_REF_FLOAT__MASK				0xffff0000
+#define A3XX_RB_ALPHA_REF_FLOAT__SHIFT				16
+static inline uint32_t A3XX_RB_ALPHA_REF_FLOAT(float val)
+{
+	return ((util_float_to_half(val)) << A3XX_RB_ALPHA_REF_FLOAT__SHIFT) & A3XX_RB_ALPHA_REF_FLOAT__MASK;
+}
 
 static inline uint32_t REG_A3XX_RB_MRT(uint32_t i0) { return 0x000020c4 + 0x4*i0; }
 
@@ -1002,7 +1016,7 @@ static inline uint32_t A3XX_RB_COPY_DEST_INFO_ENDIAN(enum adreno_rb_surface_endi
 #define REG_A3XX_RB_DEPTH_CONTROL				0x00002100
 #define A3XX_RB_DEPTH_CONTROL_Z_ENABLE				0x00000002
 #define A3XX_RB_DEPTH_CONTROL_Z_WRITE_ENABLE			0x00000004
-#define A3XX_RB_DEPTH_CONTROL_EARLY_Z_ENABLE			0x00000008
+#define A3XX_RB_DEPTH_CONTROL_EARLY_Z_DISABLE			0x00000008
 #define A3XX_RB_DEPTH_CONTROL_ZFUNC__MASK			0x00000070
 #define A3XX_RB_DEPTH_CONTROL_ZFUNC__SHIFT			4
 static inline uint32_t A3XX_RB_DEPTH_CONTROL_ZFUNC(enum adreno_compare_func val)
@@ -1038,7 +1052,8 @@ static inline uint32_t A3XX_RB_DEPTH_PITCH(uint32_t val)
 
 #define REG_A3XX_RB_STENCIL_CONTROL				0x00002104
 #define A3XX_RB_STENCIL_CONTROL_STENCIL_ENABLE			0x00000001
-#define A3XX_RB_STENCIL_CONTROL_STENCIL_ENABLE_BF		0x00000004
+#define A3XX_RB_STENCIL_CONTROL_STENCIL_ENABLE_BF		0x00000002
+#define A3XX_RB_STENCIL_CONTROL_STENCIL_READ			0x00000004
 #define A3XX_RB_STENCIL_CONTROL_FUNC__MASK			0x00000700
 #define A3XX_RB_STENCIL_CONTROL_FUNC__SHIFT			8
 static inline uint32_t A3XX_RB_STENCIL_CONTROL_FUNC(enum adreno_compare_func val)
@@ -2074,6 +2089,7 @@ static inline uint32_t A3XX_UCHE_CACHE_INVALIDATE1_REG_OPCODE(enum a3xx_cache_op
 #define REG_A3XX_TP_PERFCOUNTER5_SELECT				0x00000f09
 
 #define REG_A3XX_TEX_SAMP_0					0x00000000
+#define A3XX_TEX_SAMP_0_MIPFILTER_LINEAR			0x00000002
 #define A3XX_TEX_SAMP_0_XY_MAG__MASK				0x0000000c
 #define A3XX_TEX_SAMP_0_XY_MAG__SHIFT				2
 static inline uint32_t A3XX_TEX_SAMP_0_XY_MAG(enum a3xx_tex_filter val)
@@ -2134,6 +2150,12 @@ static inline uint32_t A3XX_TEX_CONST_0_SWIZ_W(enum a3xx_tex_swiz val)
 {
 	return ((val) << A3XX_TEX_CONST_0_SWIZ_W__SHIFT) & A3XX_TEX_CONST_0_SWIZ_W__MASK;
 }
+#define A3XX_TEX_CONST_0_MIPLVLS__MASK				0x000f0000
+#define A3XX_TEX_CONST_0_MIPLVLS__SHIFT				16
+static inline uint32_t A3XX_TEX_CONST_0_MIPLVLS(uint32_t val)
+{
+	return ((val) << A3XX_TEX_CONST_0_MIPLVLS__SHIFT) & A3XX_TEX_CONST_0_MIPLVLS__MASK;
+}
 #define A3XX_TEX_CONST_0_FMT__MASK				0x1fc00000
 #define A3XX_TEX_CONST_0_FMT__SHIFT				22
 static inline uint32_t A3XX_TEX_CONST_0_FMT(enum a3xx_tex_fmt val)
diff --git a/drivers/gpu/drm/msm/adreno/adreno_common.xml.h b/drivers/gpu/drm/msm/adreno/adreno_common.xml.h
index 61979d458ac0..33dcc606c7c5 100644
--- a/drivers/gpu/drm/msm/adreno/adreno_common.xml.h
+++ b/drivers/gpu/drm/msm/adreno/adreno_common.xml.h
@@ -4,16 +4,16 @@
 /* Autogenerated file, DO NOT EDIT manually!
 
 This file was generated by the rules-ng-ng headergen tool in this git repository:
-http://0x04.net/cgit/index.cgi/rules-ng-ng
-git clone git://0x04.net/rules-ng-ng
+http://github.com/freedreno/envytools/
+git clone https://github.com/freedreno/envytools.git
 
 The rules-ng-ng source files this header was generated from are:
 - /home/robclark/src/freedreno/envytools/rnndb/adreno.xml              (    327 bytes, from 2013-07-05 19:21:12)
 - /home/robclark/src/freedreno/envytools/rnndb/freedreno_copyright.xml (   1453 bytes, from 2013-03-31 16:51:27)
-- /home/robclark/src/freedreno/envytools/rnndb/a2xx/a2xx.xml           (  30005 bytes, from 2013-07-19 21:30:48)
+- /home/robclark/src/freedreno/envytools/rnndb/a2xx/a2xx.xml           (  31003 bytes, from 2013-09-19 18:50:16)
 - /home/robclark/src/freedreno/envytools/rnndb/adreno_common.xml       (   8983 bytes, from 2013-07-24 01:38:36)
-- /home/robclark/src/freedreno/envytools/rnndb/adreno_pm4.xml          (   9712 bytes, from 2013-05-26 15:22:37)
-- /home/robclark/src/freedreno/envytools/rnndb/a3xx/a3xx.xml           (  51415 bytes, from 2013-08-03 14:26:05)
+- /home/robclark/src/freedreno/envytools/rnndb/adreno_pm4.xml          (   9759 bytes, from 2013-09-10 00:52:33)
+- /home/robclark/src/freedreno/envytools/rnndb/a3xx/a3xx.xml           (  51983 bytes, from 2013-09-10 00:52:32)
 
 Copyright (C) 2013 by the following authors:
 - Rob Clark <robdclark@gmail.com> (robclark)
diff --git a/drivers/gpu/drm/msm/adreno/adreno_pm4.xml.h b/drivers/gpu/drm/msm/adreno/adreno_pm4.xml.h
index 94c13f418e75..259ad709b0cc 100644
--- a/drivers/gpu/drm/msm/adreno/adreno_pm4.xml.h
+++ b/drivers/gpu/drm/msm/adreno/adreno_pm4.xml.h
@@ -4,16 +4,16 @@
 /* Autogenerated file, DO NOT EDIT manually!
 
 This file was generated by the rules-ng-ng headergen tool in this git repository:
-http://0x04.net/cgit/index.cgi/rules-ng-ng
-git clone git://0x04.net/rules-ng-ng
+http://github.com/freedreno/envytools/
+git clone https://github.com/freedreno/envytools.git
 
 The rules-ng-ng source files this header was generated from are:
 - /home/robclark/src/freedreno/envytools/rnndb/adreno.xml              (    327 bytes, from 2013-07-05 19:21:12)
 - /home/robclark/src/freedreno/envytools/rnndb/freedreno_copyright.xml (   1453 bytes, from 2013-03-31 16:51:27)
-- /home/robclark/src/freedreno/envytools/rnndb/a2xx/a2xx.xml           (  30005 bytes, from 2013-07-19 21:30:48)
+- /home/robclark/src/freedreno/envytools/rnndb/a2xx/a2xx.xml           (  31003 bytes, from 2013-09-19 18:50:16)
 - /home/robclark/src/freedreno/envytools/rnndb/adreno_common.xml       (   8983 bytes, from 2013-07-24 01:38:36)
-- /home/robclark/src/freedreno/envytools/rnndb/adreno_pm4.xml          (   9712 bytes, from 2013-05-26 15:22:37)
-- /home/robclark/src/freedreno/envytools/rnndb/a3xx/a3xx.xml           (  51415 bytes, from 2013-08-03 14:26:05)
+- /home/robclark/src/freedreno/envytools/rnndb/adreno_pm4.xml          (   9759 bytes, from 2013-09-10 00:52:33)
+- /home/robclark/src/freedreno/envytools/rnndb/a3xx/a3xx.xml           (  51983 bytes, from 2013-09-10 00:52:32)
 
 Copyright (C) 2013 by the following authors:
 - Rob Clark <robdclark@gmail.com> (robclark)
diff --git a/drivers/gpu/drm/msm/dsi/dsi.xml.h b/drivers/gpu/drm/msm/dsi/dsi.xml.h
index 6f8396be431d..6d4c62bf70dc 100644
--- a/drivers/gpu/drm/msm/dsi/dsi.xml.h
+++ b/drivers/gpu/drm/msm/dsi/dsi.xml.h
@@ -4,13 +4,13 @@
 /* Autogenerated file, DO NOT EDIT manually!
 
 This file was generated by the rules-ng-ng headergen tool in this git repository:
-http://0x04.net/cgit/index.cgi/rules-ng-ng
-git clone git://0x04.net/rules-ng-ng
+http://github.com/freedreno/envytools/
+git clone https://github.com/freedreno/envytools.git
 
 The rules-ng-ng source files this header was generated from are:
 - /home/robclark/src/freedreno/envytools/rnndb/msm.xml                 (    595 bytes, from 2013-07-05 19:21:12)
 - /home/robclark/src/freedreno/envytools/rnndb/freedreno_copyright.xml (   1453 bytes, from 2013-03-31 16:51:27)
-- /home/robclark/src/freedreno/envytools/rnndb/mdp4/mdp4.xml           (  19332 bytes, from 2013-08-16 22:16:36)
+- /home/robclark/src/freedreno/envytools/rnndb/mdp4/mdp4.xml           (  19332 bytes, from 2013-10-07 16:36:48)
 - /home/robclark/src/freedreno/envytools/rnndb/dsi/dsi.xml             (  11712 bytes, from 2013-08-17 17:13:43)
 - /home/robclark/src/freedreno/envytools/rnndb/dsi/sfpb.xml            (    344 bytes, from 2013-08-11 19:26:32)
 - /home/robclark/src/freedreno/envytools/rnndb/dsi/mmss_cc.xml         (   1544 bytes, from 2013-08-16 19:17:05)
diff --git a/drivers/gpu/drm/msm/dsi/mmss_cc.xml.h b/drivers/gpu/drm/msm/dsi/mmss_cc.xml.h
index aefc1b8feae9..d1df38bf5747 100644
--- a/drivers/gpu/drm/msm/dsi/mmss_cc.xml.h
+++ b/drivers/gpu/drm/msm/dsi/mmss_cc.xml.h
@@ -4,13 +4,13 @@
 /* Autogenerated file, DO NOT EDIT manually!
 
 This file was generated by the rules-ng-ng headergen tool in this git repository:
-http://0x04.net/cgit/index.cgi/rules-ng-ng
-git clone git://0x04.net/rules-ng-ng
+http://github.com/freedreno/envytools/
+git clone https://github.com/freedreno/envytools.git
 
 The rules-ng-ng source files this header was generated from are:
 - /home/robclark/src/freedreno/envytools/rnndb/msm.xml                 (    595 bytes, from 2013-07-05 19:21:12)
 - /home/robclark/src/freedreno/envytools/rnndb/freedreno_copyright.xml (   1453 bytes, from 2013-03-31 16:51:27)
-- /home/robclark/src/freedreno/envytools/rnndb/mdp4/mdp4.xml           (  19332 bytes, from 2013-08-16 22:16:36)
+- /home/robclark/src/freedreno/envytools/rnndb/mdp4/mdp4.xml           (  19332 bytes, from 2013-10-07 16:36:48)
 - /home/robclark/src/freedreno/envytools/rnndb/dsi/dsi.xml             (  11712 bytes, from 2013-08-17 17:13:43)
 - /home/robclark/src/freedreno/envytools/rnndb/dsi/sfpb.xml            (    344 bytes, from 2013-08-11 19:26:32)
 - /home/robclark/src/freedreno/envytools/rnndb/dsi/mmss_cc.xml         (   1544 bytes, from 2013-08-16 19:17:05)
diff --git a/drivers/gpu/drm/msm/dsi/sfpb.xml.h b/drivers/gpu/drm/msm/dsi/sfpb.xml.h
index a225e8170b2a..0030a111302d 100644
--- a/drivers/gpu/drm/msm/dsi/sfpb.xml.h
+++ b/drivers/gpu/drm/msm/dsi/sfpb.xml.h
@@ -4,13 +4,13 @@
 /* Autogenerated file, DO NOT EDIT manually!
 
 This file was generated by the rules-ng-ng headergen tool in this git repository:
-http://0x04.net/cgit/index.cgi/rules-ng-ng
-git clone git://0x04.net/rules-ng-ng
+http://github.com/freedreno/envytools/
+git clone https://github.com/freedreno/envytools.git
 
 The rules-ng-ng source files this header was generated from are:
 - /home/robclark/src/freedreno/envytools/rnndb/msm.xml                 (    595 bytes, from 2013-07-05 19:21:12)
 - /home/robclark/src/freedreno/envytools/rnndb/freedreno_copyright.xml (   1453 bytes, from 2013-03-31 16:51:27)
-- /home/robclark/src/freedreno/envytools/rnndb/mdp4/mdp4.xml           (  19332 bytes, from 2013-08-16 22:16:36)
+- /home/robclark/src/freedreno/envytools/rnndb/mdp4/mdp4.xml           (  19332 bytes, from 2013-10-07 16:36:48)
 - /home/robclark/src/freedreno/envytools/rnndb/dsi/dsi.xml             (  11712 bytes, from 2013-08-17 17:13:43)
 - /home/robclark/src/freedreno/envytools/rnndb/dsi/sfpb.xml            (    344 bytes, from 2013-08-11 19:26:32)
 - /home/robclark/src/freedreno/envytools/rnndb/dsi/mmss_cc.xml         (   1544 bytes, from 2013-08-16 19:17:05)
diff --git a/drivers/gpu/drm/msm/hdmi/hdmi.xml.h b/drivers/gpu/drm/msm/hdmi/hdmi.xml.h
index f5fa4865e059..4e939f82918c 100644
--- a/drivers/gpu/drm/msm/hdmi/hdmi.xml.h
+++ b/drivers/gpu/drm/msm/hdmi/hdmi.xml.h
@@ -4,13 +4,13 @@
 /* Autogenerated file, DO NOT EDIT manually!
 
 This file was generated by the rules-ng-ng headergen tool in this git repository:
-http://0x04.net/cgit/index.cgi/rules-ng-ng
-git clone git://0x04.net/rules-ng-ng
+http://github.com/freedreno/envytools/
+git clone https://github.com/freedreno/envytools.git
 
 The rules-ng-ng source files this header was generated from are:
 - /home/robclark/src/freedreno/envytools/rnndb/msm.xml                 (    595 bytes, from 2013-07-05 19:21:12)
 - /home/robclark/src/freedreno/envytools/rnndb/freedreno_copyright.xml (   1453 bytes, from 2013-03-31 16:51:27)
-- /home/robclark/src/freedreno/envytools/rnndb/mdp4/mdp4.xml           (  19332 bytes, from 2013-08-16 22:16:36)
+- /home/robclark/src/freedreno/envytools/rnndb/mdp4/mdp4.xml           (  19332 bytes, from 2013-10-07 16:36:48)
 - /home/robclark/src/freedreno/envytools/rnndb/dsi/dsi.xml             (  11712 bytes, from 2013-08-17 17:13:43)
 - /home/robclark/src/freedreno/envytools/rnndb/dsi/sfpb.xml            (    344 bytes, from 2013-08-11 19:26:32)
 - /home/robclark/src/freedreno/envytools/rnndb/dsi/mmss_cc.xml         (   1544 bytes, from 2013-08-16 19:17:05)
diff --git a/drivers/gpu/drm/msm/hdmi/qfprom.xml.h b/drivers/gpu/drm/msm/hdmi/qfprom.xml.h
index bee36363bcd0..dbde4f6339b9 100644
--- a/drivers/gpu/drm/msm/hdmi/qfprom.xml.h
+++ b/drivers/gpu/drm/msm/hdmi/qfprom.xml.h
@@ -4,13 +4,13 @@
 /* Autogenerated file, DO NOT EDIT manually!
 
 This file was generated by the rules-ng-ng headergen tool in this git repository:
-http://0x04.net/cgit/index.cgi/rules-ng-ng
-git clone git://0x04.net/rules-ng-ng
+http://github.com/freedreno/envytools/
+git clone https://github.com/freedreno/envytools.git
 
 The rules-ng-ng source files this header was generated from are:
 - /home/robclark/src/freedreno/envytools/rnndb/msm.xml                 (    595 bytes, from 2013-07-05 19:21:12)
 - /home/robclark/src/freedreno/envytools/rnndb/freedreno_copyright.xml (   1453 bytes, from 2013-03-31 16:51:27)
-- /home/robclark/src/freedreno/envytools/rnndb/mdp4/mdp4.xml           (  19332 bytes, from 2013-08-16 22:16:36)
+- /home/robclark/src/freedreno/envytools/rnndb/mdp4/mdp4.xml           (  19332 bytes, from 2013-10-07 16:36:48)
 - /home/robclark/src/freedreno/envytools/rnndb/dsi/dsi.xml             (  11712 bytes, from 2013-08-17 17:13:43)
 - /home/robclark/src/freedreno/envytools/rnndb/dsi/sfpb.xml            (    344 bytes, from 2013-08-11 19:26:32)
 - /home/robclark/src/freedreno/envytools/rnndb/dsi/mmss_cc.xml         (   1544 bytes, from 2013-08-16 19:17:05)
diff --git a/drivers/gpu/drm/msm/mdp4/mdp4.xml.h b/drivers/gpu/drm/msm/mdp4/mdp4.xml.h
index bbeeebe2db55..9908ffe1c3ad 100644
--- a/drivers/gpu/drm/msm/mdp4/mdp4.xml.h
+++ b/drivers/gpu/drm/msm/mdp4/mdp4.xml.h
@@ -4,13 +4,13 @@
 /* Autogenerated file, DO NOT EDIT manually!
 
 This file was generated by the rules-ng-ng headergen tool in this git repository:
-http://0x04.net/cgit/index.cgi/rules-ng-ng
-git clone git://0x04.net/rules-ng-ng
+http://github.com/freedreno/envytools/
+git clone https://github.com/freedreno/envytools.git
 
 The rules-ng-ng source files this header was generated from are:
 - /home/robclark/src/freedreno/envytools/rnndb/msm.xml                 (    595 bytes, from 2013-07-05 19:21:12)
 - /home/robclark/src/freedreno/envytools/rnndb/freedreno_copyright.xml (   1453 bytes, from 2013-03-31 16:51:27)
-- /home/robclark/src/freedreno/envytools/rnndb/mdp4/mdp4.xml           (  19332 bytes, from 2013-08-16 22:16:36)
+- /home/robclark/src/freedreno/envytools/rnndb/mdp4/mdp4.xml           (  19332 bytes, from 2013-10-07 16:36:48)
 - /home/robclark/src/freedreno/envytools/rnndb/dsi/dsi.xml             (  11712 bytes, from 2013-08-17 17:13:43)
 - /home/robclark/src/freedreno/envytools/rnndb/dsi/sfpb.xml            (    344 bytes, from 2013-08-11 19:26:32)
 - /home/robclark/src/freedreno/envytools/rnndb/dsi/mmss_cc.xml         (   1544 bytes, from 2013-08-16 19:17:05)
@@ -42,28 +42,28 @@ WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE SOFTWARE.
 */
 
 
-enum mpd4_bpc {
+enum mdp4_bpc {
 	BPC1 = 0,
 	BPC5 = 1,
 	BPC6 = 2,
 	BPC8 = 3,
 };
 
-enum mpd4_bpc_alpha {
+enum mdp4_bpc_alpha {
 	BPC1A = 0,
 	BPC4A = 1,
 	BPC6A = 2,
 	BPC8A = 3,
 };
 
-enum mpd4_alpha_type {
+enum mdp4_alpha_type {
 	FG_CONST = 0,
 	BG_CONST = 1,
 	FG_PIXEL = 2,
 	BG_PIXEL = 3,
 };
 
-enum mpd4_pipe {
+enum mdp4_pipe {
 	VG1 = 0,
 	VG2 = 1,
 	RGB1 = 2,
@@ -73,13 +73,13 @@ enum mpd4_pipe {
 	VG4 = 6,
 };
 
-enum mpd4_mixer {
+enum mdp4_mixer {
 	MIXER0 = 0,
 	MIXER1 = 1,
 	MIXER2 = 2,
 };
 
-enum mpd4_mixer_stage_id {
+enum mdp4_mixer_stage_id {
 	STAGE_UNUSED = 0,
 	STAGE_BASE = 1,
 	STAGE0 = 2,
@@ -194,56 +194,56 @@ static inline uint32_t MDP4_DISP_INTF_SEL_EXT(enum mdp4_intf val)
 #define REG_MDP4_LAYERMIXER2_IN_CFG				0x000100f0
 #define MDP4_LAYERMIXER2_IN_CFG_PIPE0__MASK			0x00000007
 #define MDP4_LAYERMIXER2_IN_CFG_PIPE0__SHIFT			0
-static inline uint32_t MDP4_LAYERMIXER2_IN_CFG_PIPE0(enum mpd4_mixer_stage_id val)
+static inline uint32_t MDP4_LAYERMIXER2_IN_CFG_PIPE0(enum mdp4_mixer_stage_id val)
 {
 	return ((val) << MDP4_LAYERMIXER2_IN_CFG_PIPE0__SHIFT) & MDP4_LAYERMIXER2_IN_CFG_PIPE0__MASK;
 }
 #define MDP4_LAYERMIXER2_IN_CFG_PIPE0_MIXER1			0x00000008
 #define MDP4_LAYERMIXER2_IN_CFG_PIPE1__MASK			0x00000070
 #define MDP4_LAYERMIXER2_IN_CFG_PIPE1__SHIFT			4
-static inline uint32_t MDP4_LAYERMIXER2_IN_CFG_PIPE1(enum mpd4_mixer_stage_id val)
+static inline uint32_t MDP4_LAYERMIXER2_IN_CFG_PIPE1(enum mdp4_mixer_stage_id val)
 {
 	return ((val) << MDP4_LAYERMIXER2_IN_CFG_PIPE1__SHIFT) & MDP4_LAYERMIXER2_IN_CFG_PIPE1__MASK;
 }
 #define MDP4_LAYERMIXER2_IN_CFG_PIPE1_MIXER1			0x00000080
 #define MDP4_LAYERMIXER2_IN_CFG_PIPE2__MASK			0x00000700
 #define MDP4_LAYERMIXER2_IN_CFG_PIPE2__SHIFT			8
-static inline uint32_t MDP4_LAYERMIXER2_IN_CFG_PIPE2(enum mpd4_mixer_stage_id val)
+static inline uint32_t MDP4_LAYERMIXER2_IN_CFG_PIPE2(enum mdp4_mixer_stage_id val)
 {
 	return ((val) << MDP4_LAYERMIXER2_IN_CFG_PIPE2__SHIFT) & MDP4_LAYERMIXER2_IN_CFG_PIPE2__MASK;
 }
 #define MDP4_LAYERMIXER2_IN_CFG_PIPE2_MIXER1			0x00000800
 #define MDP4_LAYERMIXER2_IN_CFG_PIPE3__MASK			0x00007000
 #define MDP4_LAYERMIXER2_IN_CFG_PIPE3__SHIFT			12
-static inline uint32_t MDP4_LAYERMIXER2_IN_CFG_PIPE3(enum mpd4_mixer_stage_id val)
+static inline uint32_t MDP4_LAYERMIXER2_IN_CFG_PIPE3(enum mdp4_mixer_stage_id val)
 {
 	return ((val) << MDP4_LAYERMIXER2_IN_CFG_PIPE3__SHIFT) & MDP4_LAYERMIXER2_IN_CFG_PIPE3__MASK;
 }
 #define MDP4_LAYERMIXER2_IN_CFG_PIPE3_MIXER1			0x00008000
 #define MDP4_LAYERMIXER2_IN_CFG_PIPE4__MASK			0x00070000
 #define MDP4_LAYERMIXER2_IN_CFG_PIPE4__SHIFT			16
-static inline uint32_t MDP4_LAYERMIXER2_IN_CFG_PIPE4(enum mpd4_mixer_stage_id val)
+static inline uint32_t MDP4_LAYERMIXER2_IN_CFG_PIPE4(enum mdp4_mixer_stage_id val)
 {
 	return ((val) << MDP4_LAYERMIXER2_IN_CFG_PIPE4__SHIFT) & MDP4_LAYERMIXER2_IN_CFG_PIPE4__MASK;
 }
 #define MDP4_LAYERMIXER2_IN_CFG_PIPE4_MIXER1			0x00080000
 #define MDP4_LAYERMIXER2_IN_CFG_PIPE5__MASK			0x00700000
 #define MDP4_LAYERMIXER2_IN_CFG_PIPE5__SHIFT			20
-static inline uint32_t MDP4_LAYERMIXER2_IN_CFG_PIPE5(enum mpd4_mixer_stage_id val)
+static inline uint32_t MDP4_LAYERMIXER2_IN_CFG_PIPE5(enum mdp4_mixer_stage_id val)
 {
 	return ((val) << MDP4_LAYERMIXER2_IN_CFG_PIPE5__SHIFT) & MDP4_LAYERMIXER2_IN_CFG_PIPE5__MASK;
 }
 #define MDP4_LAYERMIXER2_IN_CFG_PIPE5_MIXER1			0x00800000
 #define MDP4_LAYERMIXER2_IN_CFG_PIPE6__MASK			0x07000000
 #define MDP4_LAYERMIXER2_IN_CFG_PIPE6__SHIFT			24
-static inline uint32_t MDP4_LAYERMIXER2_IN_CFG_PIPE6(enum mpd4_mixer_stage_id val)
+static inline uint32_t MDP4_LAYERMIXER2_IN_CFG_PIPE6(enum mdp4_mixer_stage_id val)
 {
 	return ((val) << MDP4_LAYERMIXER2_IN_CFG_PIPE6__SHIFT) & MDP4_LAYERMIXER2_IN_CFG_PIPE6__MASK;
 }
 #define MDP4_LAYERMIXER2_IN_CFG_PIPE6_MIXER1			0x08000000
 #define MDP4_LAYERMIXER2_IN_CFG_PIPE7__MASK			0x70000000
 #define MDP4_LAYERMIXER2_IN_CFG_PIPE7__SHIFT			28
-static inline uint32_t MDP4_LAYERMIXER2_IN_CFG_PIPE7(enum mpd4_mixer_stage_id val)
+static inline uint32_t MDP4_LAYERMIXER2_IN_CFG_PIPE7(enum mdp4_mixer_stage_id val)
 {
 	return ((val) << MDP4_LAYERMIXER2_IN_CFG_PIPE7__SHIFT) & MDP4_LAYERMIXER2_IN_CFG_PIPE7__MASK;
 }
@@ -254,56 +254,56 @@ static inline uint32_t MDP4_LAYERMIXER2_IN_CFG_PIPE7(enum mpd4_mixer_stage_id va
 #define REG_MDP4_LAYERMIXER_IN_CFG				0x00010100
 #define MDP4_LAYERMIXER_IN_CFG_PIPE0__MASK			0x00000007
 #define MDP4_LAYERMIXER_IN_CFG_PIPE0__SHIFT			0
-static inline uint32_t MDP4_LAYERMIXER_IN_CFG_PIPE0(enum mpd4_mixer_stage_id val)
+static inline uint32_t MDP4_LAYERMIXER_IN_CFG_PIPE0(enum mdp4_mixer_stage_id val)
 {
 	return ((val) << MDP4_LAYERMIXER_IN_CFG_PIPE0__SHIFT) & MDP4_LAYERMIXER_IN_CFG_PIPE0__MASK;
 }
 #define MDP4_LAYERMIXER_IN_CFG_PIPE0_MIXER1			0x00000008
 #define MDP4_LAYERMIXER_IN_CFG_PIPE1__MASK			0x00000070
 #define MDP4_LAYERMIXER_IN_CFG_PIPE1__SHIFT			4
-static inline uint32_t MDP4_LAYERMIXER_IN_CFG_PIPE1(enum mpd4_mixer_stage_id val)
+static inline uint32_t MDP4_LAYERMIXER_IN_CFG_PIPE1(enum mdp4_mixer_stage_id val)
 {
 	return ((val) << MDP4_LAYERMIXER_IN_CFG_PIPE1__SHIFT) & MDP4_LAYERMIXER_IN_CFG_PIPE1__MASK;
 }
 #define MDP4_LAYERMIXER_IN_CFG_PIPE1_MIXER1			0x00000080
 #define MDP4_LAYERMIXER_IN_CFG_PIPE2__MASK			0x00000700
 #define MDP4_LAYERMIXER_IN_CFG_PIPE2__SHIFT			8
-static inline uint32_t MDP4_LAYERMIXER_IN_CFG_PIPE2(enum mpd4_mixer_stage_id val)
+static inline uint32_t MDP4_LAYERMIXER_IN_CFG_PIPE2(enum mdp4_mixer_stage_id val)
 {
 	return ((val) << MDP4_LAYERMIXER_IN_CFG_PIPE2__SHIFT) & MDP4_LAYERMIXER_IN_CFG_PIPE2__MASK;
 }
 #define MDP4_LAYERMIXER_IN_CFG_PIPE2_MIXER1			0x00000800
 #define MDP4_LAYERMIXER_IN_CFG_PIPE3__MASK			0x00007000
 #define MDP4_LAYERMIXER_IN_CFG_PIPE3__SHIFT			12
-static inline uint32_t MDP4_LAYERMIXER_IN_CFG_PIPE3(enum mpd4_mixer_stage_id val)
+static inline uint32_t MDP4_LAYERMIXER_IN_CFG_PIPE3(enum mdp4_mixer_stage_id val)
 {
 	return ((val) << MDP4_LAYERMIXER_IN_CFG_PIPE3__SHIFT) & MDP4_LAYERMIXER_IN_CFG_PIPE3__MASK;
 }
 #define MDP4_LAYERMIXER_IN_CFG_PIPE3_MIXER1			0x00008000
 #define MDP4_LAYERMIXER_IN_CFG_PIPE4__MASK			0x00070000
 #define MDP4_LAYERMIXER_IN_CFG_PIPE4__SHIFT			16
-static inline uint32_t MDP4_LAYERMIXER_IN_CFG_PIPE4(enum mpd4_mixer_stage_id val)
+static inline uint32_t MDP4_LAYERMIXER_IN_CFG_PIPE4(enum mdp4_mixer_stage_id val)
 {
 	return ((val) << MDP4_LAYERMIXER_IN_CFG_PIPE4__SHIFT) & MDP4_LAYERMIXER_IN_CFG_PIPE4__MASK;
 }
 #define MDP4_LAYERMIXER_IN_CFG_PIPE4_MIXER1			0x00080000
 #define MDP4_LAYERMIXER_IN_CFG_PIPE5__MASK			0x00700000
 #define MDP4_LAYERMIXER_IN_CFG_PIPE5__SHIFT			20
-static inline uint32_t MDP4_LAYERMIXER_IN_CFG_PIPE5(enum mpd4_mixer_stage_id val)
+static inline uint32_t MDP4_LAYERMIXER_IN_CFG_PIPE5(enum mdp4_mixer_stage_id val)
 {
 	return ((val) << MDP4_LAYERMIXER_IN_CFG_PIPE5__SHIFT) & MDP4_LAYERMIXER_IN_CFG_PIPE5__MASK;
 }
 #define MDP4_LAYERMIXER_IN_CFG_PIPE5_MIXER1			0x00800000
 #define MDP4_LAYERMIXER_IN_CFG_PIPE6__MASK			0x07000000
 #define MDP4_LAYERMIXER_IN_CFG_PIPE6__SHIFT			24
-static inline uint32_t MDP4_LAYERMIXER_IN_CFG_PIPE6(enum mpd4_mixer_stage_id val)
+static inline uint32_t MDP4_LAYERMIXER_IN_CFG_PIPE6(enum mdp4_mixer_stage_id val)
 {
 	return ((val) << MDP4_LAYERMIXER_IN_CFG_PIPE6__SHIFT) & MDP4_LAYERMIXER_IN_CFG_PIPE6__MASK;
 }
 #define MDP4_LAYERMIXER_IN_CFG_PIPE6_MIXER1			0x08000000
 #define MDP4_LAYERMIXER_IN_CFG_PIPE7__MASK			0x70000000
 #define MDP4_LAYERMIXER_IN_CFG_PIPE7__SHIFT			28
-static inline uint32_t MDP4_LAYERMIXER_IN_CFG_PIPE7(enum mpd4_mixer_stage_id val)
+static inline uint32_t MDP4_LAYERMIXER_IN_CFG_PIPE7(enum mdp4_mixer_stage_id val)
 {
 	return ((val) << MDP4_LAYERMIXER_IN_CFG_PIPE7__SHIFT) & MDP4_LAYERMIXER_IN_CFG_PIPE7__MASK;
 }
@@ -369,7 +369,7 @@ static inline uint32_t REG_MDP4_OVLP_STAGE(uint32_t i0, uint32_t i1) { return 0x
 static inline uint32_t REG_MDP4_OVLP_STAGE_OP(uint32_t i0, uint32_t i1) { return 0x00000000 + __offset_OVLP(i0) + __offset_STAGE(i1); }
 #define MDP4_OVLP_STAGE_OP_FG_ALPHA__MASK			0x00000003
 #define MDP4_OVLP_STAGE_OP_FG_ALPHA__SHIFT			0
-static inline uint32_t MDP4_OVLP_STAGE_OP_FG_ALPHA(enum mpd4_alpha_type val)
+static inline uint32_t MDP4_OVLP_STAGE_OP_FG_ALPHA(enum mdp4_alpha_type val)
 {
 	return ((val) << MDP4_OVLP_STAGE_OP_FG_ALPHA__SHIFT) & MDP4_OVLP_STAGE_OP_FG_ALPHA__MASK;
 }
@@ -377,7 +377,7 @@ static inline uint32_t MDP4_OVLP_STAGE_OP_FG_ALPHA(enum mpd4_alpha_type val)
 #define MDP4_OVLP_STAGE_OP_FG_MOD_ALPHA				0x00000008
 #define MDP4_OVLP_STAGE_OP_BG_ALPHA__MASK			0x00000030
 #define MDP4_OVLP_STAGE_OP_BG_ALPHA__SHIFT			4
-static inline uint32_t MDP4_OVLP_STAGE_OP_BG_ALPHA(enum mpd4_alpha_type val)
+static inline uint32_t MDP4_OVLP_STAGE_OP_BG_ALPHA(enum mdp4_alpha_type val)
 {
 	return ((val) << MDP4_OVLP_STAGE_OP_BG_ALPHA__SHIFT) & MDP4_OVLP_STAGE_OP_BG_ALPHA__MASK;
 }
@@ -472,19 +472,19 @@ static inline uint32_t REG_MDP4_DMA(enum mdp4_dma i0) { return 0x00000000 + __of
 static inline uint32_t REG_MDP4_DMA_CONFIG(enum mdp4_dma i0) { return 0x00000000 + __offset_DMA(i0); }
 #define MDP4_DMA_CONFIG_G_BPC__MASK				0x00000003
 #define MDP4_DMA_CONFIG_G_BPC__SHIFT				0
-static inline uint32_t MDP4_DMA_CONFIG_G_BPC(enum mpd4_bpc val)
+static inline uint32_t MDP4_DMA_CONFIG_G_BPC(enum mdp4_bpc val)
 {
 	return ((val) << MDP4_DMA_CONFIG_G_BPC__SHIFT) & MDP4_DMA_CONFIG_G_BPC__MASK;
 }
 #define MDP4_DMA_CONFIG_B_BPC__MASK				0x0000000c
 #define MDP4_DMA_CONFIG_B_BPC__SHIFT				2
-static inline uint32_t MDP4_DMA_CONFIG_B_BPC(enum mpd4_bpc val)
+static inline uint32_t MDP4_DMA_CONFIG_B_BPC(enum mdp4_bpc val)
 {
 	return ((val) << MDP4_DMA_CONFIG_B_BPC__SHIFT) & MDP4_DMA_CONFIG_B_BPC__MASK;
 }
 #define MDP4_DMA_CONFIG_R_BPC__MASK				0x00000030
 #define MDP4_DMA_CONFIG_R_BPC__SHIFT				4
-static inline uint32_t MDP4_DMA_CONFIG_R_BPC(enum mpd4_bpc val)
+static inline uint32_t MDP4_DMA_CONFIG_R_BPC(enum mdp4_bpc val)
 {
 	return ((val) << MDP4_DMA_CONFIG_R_BPC__SHIFT) & MDP4_DMA_CONFIG_R_BPC__MASK;
 }
@@ -601,9 +601,9 @@ static inline uint32_t REG_MDP4_DMA_CSC_POST_LV(enum mdp4_dma i0, uint32_t i1) {
 
 static inline uint32_t REG_MDP4_DMA_CSC_POST_LV_VAL(enum mdp4_dma i0, uint32_t i1) { return 0x00003680 + __offset_DMA(i0) + 0x4*i1; }
 
-static inline uint32_t REG_MDP4_PIPE(enum mpd4_pipe i0) { return 0x00020000 + 0x10000*i0; }
+static inline uint32_t REG_MDP4_PIPE(enum mdp4_pipe i0) { return 0x00020000 + 0x10000*i0; }
 
-static inline uint32_t REG_MDP4_PIPE_SRC_SIZE(enum mpd4_pipe i0) { return 0x00020000 + 0x10000*i0; }
+static inline uint32_t REG_MDP4_PIPE_SRC_SIZE(enum mdp4_pipe i0) { return 0x00020000 + 0x10000*i0; }
 #define MDP4_PIPE_SRC_SIZE_HEIGHT__MASK				0xffff0000
 #define MDP4_PIPE_SRC_SIZE_HEIGHT__SHIFT			16
 static inline uint32_t MDP4_PIPE_SRC_SIZE_HEIGHT(uint32_t val)
@@ -617,7 +617,7 @@ static inline uint32_t MDP4_PIPE_SRC_SIZE_WIDTH(uint32_t val)
 	return ((val) << MDP4_PIPE_SRC_SIZE_WIDTH__SHIFT) & MDP4_PIPE_SRC_SIZE_WIDTH__MASK;
 }
 
-static inline uint32_t REG_MDP4_PIPE_SRC_XY(enum mpd4_pipe i0) { return 0x00020004 + 0x10000*i0; }
+static inline uint32_t REG_MDP4_PIPE_SRC_XY(enum mdp4_pipe i0) { return 0x00020004 + 0x10000*i0; }
 #define MDP4_PIPE_SRC_XY_Y__MASK				0xffff0000
 #define MDP4_PIPE_SRC_XY_Y__SHIFT				16
 static inline uint32_t MDP4_PIPE_SRC_XY_Y(uint32_t val)
@@ -631,7 +631,7 @@ static inline uint32_t MDP4_PIPE_SRC_XY_X(uint32_t val)
 	return ((val) << MDP4_PIPE_SRC_XY_X__SHIFT) & MDP4_PIPE_SRC_XY_X__MASK;
 }
 
-static inline uint32_t REG_MDP4_PIPE_DST_SIZE(enum mpd4_pipe i0) { return 0x00020008 + 0x10000*i0; }
+static inline uint32_t REG_MDP4_PIPE_DST_SIZE(enum mdp4_pipe i0) { return 0x00020008 + 0x10000*i0; }
 #define MDP4_PIPE_DST_SIZE_HEIGHT__MASK				0xffff0000
 #define MDP4_PIPE_DST_SIZE_HEIGHT__SHIFT			16
 static inline uint32_t MDP4_PIPE_DST_SIZE_HEIGHT(uint32_t val)
@@ -645,7 +645,7 @@ static inline uint32_t MDP4_PIPE_DST_SIZE_WIDTH(uint32_t val)
 	return ((val) << MDP4_PIPE_DST_SIZE_WIDTH__SHIFT) & MDP4_PIPE_DST_SIZE_WIDTH__MASK;
 }
 
-static inline uint32_t REG_MDP4_PIPE_DST_XY(enum mpd4_pipe i0) { return 0x0002000c + 0x10000*i0; }
+static inline uint32_t REG_MDP4_PIPE_DST_XY(enum mdp4_pipe i0) { return 0x0002000c + 0x10000*i0; }
 #define MDP4_PIPE_DST_XY_Y__MASK				0xffff0000
 #define MDP4_PIPE_DST_XY_Y__SHIFT				16
 static inline uint32_t MDP4_PIPE_DST_XY_Y(uint32_t val)
@@ -659,13 +659,13 @@ static inline uint32_t MDP4_PIPE_DST_XY_X(uint32_t val)
 	return ((val) << MDP4_PIPE_DST_XY_X__SHIFT) & MDP4_PIPE_DST_XY_X__MASK;
 }
 
-static inline uint32_t REG_MDP4_PIPE_SRCP0_BASE(enum mpd4_pipe i0) { return 0x00020010 + 0x10000*i0; }
+static inline uint32_t REG_MDP4_PIPE_SRCP0_BASE(enum mdp4_pipe i0) { return 0x00020010 + 0x10000*i0; }
 
-static inline uint32_t REG_MDP4_PIPE_SRCP1_BASE(enum mpd4_pipe i0) { return 0x00020014 + 0x10000*i0; }
+static inline uint32_t REG_MDP4_PIPE_SRCP1_BASE(enum mdp4_pipe i0) { return 0x00020014 + 0x10000*i0; }
 
-static inline uint32_t REG_MDP4_PIPE_SRCP2_BASE(enum mpd4_pipe i0) { return 0x00020018 + 0x10000*i0; }
+static inline uint32_t REG_MDP4_PIPE_SRCP2_BASE(enum mdp4_pipe i0) { return 0x00020018 + 0x10000*i0; }
 
-static inline uint32_t REG_MDP4_PIPE_SRC_STRIDE_A(enum mpd4_pipe i0) { return 0x00020040 + 0x10000*i0; }
+static inline uint32_t REG_MDP4_PIPE_SRC_STRIDE_A(enum mdp4_pipe i0) { return 0x00020040 + 0x10000*i0; }
 #define MDP4_PIPE_SRC_STRIDE_A_P0__MASK				0x0000ffff
 #define MDP4_PIPE_SRC_STRIDE_A_P0__SHIFT			0
 static inline uint32_t MDP4_PIPE_SRC_STRIDE_A_P0(uint32_t val)
@@ -679,7 +679,7 @@ static inline uint32_t MDP4_PIPE_SRC_STRIDE_A_P1(uint32_t val)
 	return ((val) << MDP4_PIPE_SRC_STRIDE_A_P1__SHIFT) & MDP4_PIPE_SRC_STRIDE_A_P1__MASK;
 }
 
-static inline uint32_t REG_MDP4_PIPE_SRC_STRIDE_B(enum mpd4_pipe i0) { return 0x00020044 + 0x10000*i0; }
+static inline uint32_t REG_MDP4_PIPE_SRC_STRIDE_B(enum mdp4_pipe i0) { return 0x00020044 + 0x10000*i0; }
 #define MDP4_PIPE_SRC_STRIDE_B_P2__MASK				0x0000ffff
 #define MDP4_PIPE_SRC_STRIDE_B_P2__SHIFT			0
 static inline uint32_t MDP4_PIPE_SRC_STRIDE_B_P2(uint32_t val)
@@ -693,7 +693,7 @@ static inline uint32_t MDP4_PIPE_SRC_STRIDE_B_P3(uint32_t val)
 	return ((val) << MDP4_PIPE_SRC_STRIDE_B_P3__SHIFT) & MDP4_PIPE_SRC_STRIDE_B_P3__MASK;
 }
 
-static inline uint32_t REG_MDP4_PIPE_FRAME_SIZE(enum mpd4_pipe i0) { return 0x00020048 + 0x10000*i0; }
+static inline uint32_t REG_MDP4_PIPE_FRAME_SIZE(enum mdp4_pipe i0) { return 0x00020048 + 0x10000*i0; }
 #define MDP4_PIPE_FRAME_SIZE_HEIGHT__MASK			0xffff0000
 #define MDP4_PIPE_FRAME_SIZE_HEIGHT__SHIFT			16
 static inline uint32_t MDP4_PIPE_FRAME_SIZE_HEIGHT(uint32_t val)
@@ -707,28 +707,28 @@ static inline uint32_t MDP4_PIPE_FRAME_SIZE_WIDTH(uint32_t val)
 	return ((val) << MDP4_PIPE_FRAME_SIZE_WIDTH__SHIFT) & MDP4_PIPE_FRAME_SIZE_WIDTH__MASK;
 }
 
-static inline uint32_t REG_MDP4_PIPE_SRC_FORMAT(enum mpd4_pipe i0) { return 0x00020050 + 0x10000*i0; }
+static inline uint32_t REG_MDP4_PIPE_SRC_FORMAT(enum mdp4_pipe i0) { return 0x00020050 + 0x10000*i0; }
 #define MDP4_PIPE_SRC_FORMAT_G_BPC__MASK			0x00000003
 #define MDP4_PIPE_SRC_FORMAT_G_BPC__SHIFT			0
-static inline uint32_t MDP4_PIPE_SRC_FORMAT_G_BPC(enum mpd4_bpc val)
+static inline uint32_t MDP4_PIPE_SRC_FORMAT_G_BPC(enum mdp4_bpc val)
 {
 	return ((val) << MDP4_PIPE_SRC_FORMAT_G_BPC__SHIFT) & MDP4_PIPE_SRC_FORMAT_G_BPC__MASK;
 }
 #define MDP4_PIPE_SRC_FORMAT_B_BPC__MASK			0x0000000c
 #define MDP4_PIPE_SRC_FORMAT_B_BPC__SHIFT			2
-static inline uint32_t MDP4_PIPE_SRC_FORMAT_B_BPC(enum mpd4_bpc val)
+static inline uint32_t MDP4_PIPE_SRC_FORMAT_B_BPC(enum mdp4_bpc val)
 {
 	return ((val) << MDP4_PIPE_SRC_FORMAT_B_BPC__SHIFT) & MDP4_PIPE_SRC_FORMAT_B_BPC__MASK;
 }
 #define MDP4_PIPE_SRC_FORMAT_R_BPC__MASK			0x00000030
 #define MDP4_PIPE_SRC_FORMAT_R_BPC__SHIFT			4
-static inline uint32_t MDP4_PIPE_SRC_FORMAT_R_BPC(enum mpd4_bpc val)
+static inline uint32_t MDP4_PIPE_SRC_FORMAT_R_BPC(enum mdp4_bpc val)
 {
 	return ((val) << MDP4_PIPE_SRC_FORMAT_R_BPC__SHIFT) & MDP4_PIPE_SRC_FORMAT_R_BPC__MASK;
 }
 #define MDP4_PIPE_SRC_FORMAT_A_BPC__MASK			0x000000c0
 #define MDP4_PIPE_SRC_FORMAT_A_BPC__SHIFT			6
-static inline uint32_t MDP4_PIPE_SRC_FORMAT_A_BPC(enum mpd4_bpc_alpha val)
+static inline uint32_t MDP4_PIPE_SRC_FORMAT_A_BPC(enum mdp4_bpc_alpha val)
 {
 	return ((val) << MDP4_PIPE_SRC_FORMAT_A_BPC__SHIFT) & MDP4_PIPE_SRC_FORMAT_A_BPC__MASK;
 }
@@ -750,7 +750,7 @@ static inline uint32_t MDP4_PIPE_SRC_FORMAT_UNPACK_COUNT(uint32_t val)
 #define MDP4_PIPE_SRC_FORMAT_UNPACK_ALIGN_MSB			0x00040000
 #define MDP4_PIPE_SRC_FORMAT_SOLID_FILL				0x00400000
 
-static inline uint32_t REG_MDP4_PIPE_SRC_UNPACK(enum mpd4_pipe i0) { return 0x00020054 + 0x10000*i0; }
+static inline uint32_t REG_MDP4_PIPE_SRC_UNPACK(enum mdp4_pipe i0) { return 0x00020054 + 0x10000*i0; }
 #define MDP4_PIPE_SRC_UNPACK_ELEM0__MASK			0x000000ff
 #define MDP4_PIPE_SRC_UNPACK_ELEM0__SHIFT			0
 static inline uint32_t MDP4_PIPE_SRC_UNPACK_ELEM0(uint32_t val)
@@ -776,7 +776,7 @@ static inline uint32_t MDP4_PIPE_SRC_UNPACK_ELEM3(uint32_t val)
 	return ((val) << MDP4_PIPE_SRC_UNPACK_ELEM3__SHIFT) & MDP4_PIPE_SRC_UNPACK_ELEM3__MASK;
 }
 
-static inline uint32_t REG_MDP4_PIPE_OP_MODE(enum mpd4_pipe i0) { return 0x00020058 + 0x10000*i0; }
+static inline uint32_t REG_MDP4_PIPE_OP_MODE(enum mdp4_pipe i0) { return 0x00020058 + 0x10000*i0; }
 #define MDP4_PIPE_OP_MODE_SCALEX_EN				0x00000001
 #define MDP4_PIPE_OP_MODE_SCALEY_EN				0x00000002
 #define MDP4_PIPE_OP_MODE_SRC_YCBCR				0x00000200
@@ -789,36 +789,36 @@ static inline uint32_t REG_MDP4_PIPE_OP_MODE(enum mpd4_pipe i0) { return 0x00020
 #define MDP4_PIPE_OP_MODE_DEINT_EN				0x00040000
 #define MDP4_PIPE_OP_MODE_DEINT_ODD_REF				0x00080000
 
-static inline uint32_t REG_MDP4_PIPE_PHASEX_STEP(enum mpd4_pipe i0) { return 0x0002005c + 0x10000*i0; }
+static inline uint32_t REG_MDP4_PIPE_PHASEX_STEP(enum mdp4_pipe i0) { return 0x0002005c + 0x10000*i0; }
 
-static inline uint32_t REG_MDP4_PIPE_PHASEY_STEP(enum mpd4_pipe i0) { return 0x00020060 + 0x10000*i0; }
+static inline uint32_t REG_MDP4_PIPE_PHASEY_STEP(enum mdp4_pipe i0) { return 0x00020060 + 0x10000*i0; }
 
-static inline uint32_t REG_MDP4_PIPE_FETCH_CONFIG(enum mpd4_pipe i0) { return 0x00021004 + 0x10000*i0; }
+static inline uint32_t REG_MDP4_PIPE_FETCH_CONFIG(enum mdp4_pipe i0) { return 0x00021004 + 0x10000*i0; }
 
-static inline uint32_t REG_MDP4_PIPE_SOLID_COLOR(enum mpd4_pipe i0) { return 0x00021008 + 0x10000*i0; }
+static inline uint32_t REG_MDP4_PIPE_SOLID_COLOR(enum mdp4_pipe i0) { return 0x00021008 + 0x10000*i0; }
 
-static inline uint32_t REG_MDP4_PIPE_CSC(enum mpd4_pipe i0) { return 0x00024000 + 0x10000*i0; }
+static inline uint32_t REG_MDP4_PIPE_CSC(enum mdp4_pipe i0) { return 0x00024000 + 0x10000*i0; }
 
 
-static inline uint32_t REG_MDP4_PIPE_CSC_MV(enum mpd4_pipe i0, uint32_t i1) { return 0x00024400 + 0x10000*i0 + 0x4*i1; }
+static inline uint32_t REG_MDP4_PIPE_CSC_MV(enum mdp4_pipe i0, uint32_t i1) { return 0x00024400 + 0x10000*i0 + 0x4*i1; }
 
-static inline uint32_t REG_MDP4_PIPE_CSC_MV_VAL(enum mpd4_pipe i0, uint32_t i1) { return 0x00024400 + 0x10000*i0 + 0x4*i1; }
+static inline uint32_t REG_MDP4_PIPE_CSC_MV_VAL(enum mdp4_pipe i0, uint32_t i1) { return 0x00024400 + 0x10000*i0 + 0x4*i1; }
 
-static inline uint32_t REG_MDP4_PIPE_CSC_PRE_BV(enum mpd4_pipe i0, uint32_t i1) { return 0x00024500 + 0x10000*i0 + 0x4*i1; }
+static inline uint32_t REG_MDP4_PIPE_CSC_PRE_BV(enum mdp4_pipe i0, uint32_t i1) { return 0x00024500 + 0x10000*i0 + 0x4*i1; }
 
-static inline uint32_t REG_MDP4_PIPE_CSC_PRE_BV_VAL(enum mpd4_pipe i0, uint32_t i1) { return 0x00024500 + 0x10000*i0 + 0x4*i1; }
+static inline uint32_t REG_MDP4_PIPE_CSC_PRE_BV_VAL(enum mdp4_pipe i0, uint32_t i1) { return 0x00024500 + 0x10000*i0 + 0x4*i1; }
 
-static inline uint32_t REG_MDP4_PIPE_CSC_POST_BV(enum mpd4_pipe i0, uint32_t i1) { return 0x00024580 + 0x10000*i0 + 0x4*i1; }
+static inline uint32_t REG_MDP4_PIPE_CSC_POST_BV(enum mdp4_pipe i0, uint32_t i1) { return 0x00024580 + 0x10000*i0 + 0x4*i1; }
 
-static inline uint32_t REG_MDP4_PIPE_CSC_POST_BV_VAL(enum mpd4_pipe i0, uint32_t i1) { return 0x00024580 + 0x10000*i0 + 0x4*i1; }
+static inline uint32_t REG_MDP4_PIPE_CSC_POST_BV_VAL(enum mdp4_pipe i0, uint32_t i1) { return 0x00024580 + 0x10000*i0 + 0x4*i1; }
 
-static inline uint32_t REG_MDP4_PIPE_CSC_PRE_LV(enum mpd4_pipe i0, uint32_t i1) { return 0x00024600 + 0x10000*i0 + 0x4*i1; }
+static inline uint32_t REG_MDP4_PIPE_CSC_PRE_LV(enum mdp4_pipe i0, uint32_t i1) { return 0x00024600 + 0x10000*i0 + 0x4*i1; }
 
-static inline uint32_t REG_MDP4_PIPE_CSC_PRE_LV_VAL(enum mpd4_pipe i0, uint32_t i1) { return 0x00024600 + 0x10000*i0 + 0x4*i1; }
+static inline uint32_t REG_MDP4_PIPE_CSC_PRE_LV_VAL(enum mdp4_pipe i0, uint32_t i1) { return 0x00024600 + 0x10000*i0 + 0x4*i1; }
 
-static inline uint32_t REG_MDP4_PIPE_CSC_POST_LV(enum mpd4_pipe i0, uint32_t i1) { return 0x00024680 + 0x10000*i0 + 0x4*i1; }
+static inline uint32_t REG_MDP4_PIPE_CSC_POST_LV(enum mdp4_pipe i0, uint32_t i1) { return 0x00024680 + 0x10000*i0 + 0x4*i1; }
 
-static inline uint32_t REG_MDP4_PIPE_CSC_POST_LV_VAL(enum mpd4_pipe i0, uint32_t i1) { return 0x00024680 + 0x10000*i0 + 0x4*i1; }
+static inline uint32_t REG_MDP4_PIPE_CSC_POST_LV_VAL(enum mdp4_pipe i0, uint32_t i1) { return 0x00024680 + 0x10000*i0 + 0x4*i1; }
 
 #define REG_MDP4_LCDC						0x000c0000
 
diff --git a/drivers/gpu/drm/msm/mdp4/mdp4_crtc.c b/drivers/gpu/drm/msm/mdp4/mdp4_crtc.c
index de6bea297cda..019d530187ff 100644
--- a/drivers/gpu/drm/msm/mdp4/mdp4_crtc.c
+++ b/drivers/gpu/drm/msm/mdp4/mdp4_crtc.c
@@ -26,6 +26,7 @@ struct mdp4_crtc {
 	struct drm_crtc base;
 	char name[8];
 	struct drm_plane *plane;
+	struct drm_plane *planes[8];
 	int id;
 	int ovlp;
 	enum mdp4_dma dma;
@@ -50,7 +51,11 @@ struct mdp4_crtc {
 
 	/* if there is a pending flip, these will be non-null: */
 	struct drm_pending_vblank_event *event;
-	struct work_struct pageflip_work;
+	struct msm_fence_cb pageflip_cb;
+
+#define PENDING_CURSOR 0x1
+#define PENDING_FLIP   0x2
+	atomic_t pending;
 
 	/* the fb that we currently hold a scanout ref to: */
 	struct drm_framebuffer *fb;
@@ -92,7 +97,8 @@ static void update_fb(struct drm_crtc *crtc, bool async,
 	}
 }
 
-static void complete_flip(struct drm_crtc *crtc, bool canceled)
+/* if file!=NULL, this is preclose potential cancel-flip path */
+static void complete_flip(struct drm_crtc *crtc, struct drm_file *file)
 {
 	struct mdp4_crtc *mdp4_crtc = to_mdp4_crtc(crtc);
 	struct drm_device *dev = crtc->dev;
@@ -102,11 +108,14 @@ static void complete_flip(struct drm_crtc *crtc, bool canceled)
 	spin_lock_irqsave(&dev->event_lock, flags);
 	event = mdp4_crtc->event;
 	if (event) {
-		mdp4_crtc->event = NULL;
-		if (canceled)
-			event->base.destroy(&event->base);
-		else
+		/* if regular vblank case (!file) or if cancel-flip from
+		 * preclose on file that requested flip, then send the
+		 * event:
+		 */
+		if (!file || (event->base.file_priv == file)) {
+			mdp4_crtc->event = NULL;
 			drm_send_vblank_event(dev, mdp4_crtc->id, event);
+		}
 	}
 	spin_unlock_irqrestore(&dev->event_lock, flags);
 }
@@ -115,9 +124,15 @@ static void crtc_flush(struct drm_crtc *crtc)
 {
 	struct mdp4_crtc *mdp4_crtc = to_mdp4_crtc(crtc);
 	struct mdp4_kms *mdp4_kms = get_kms(crtc);
-	uint32_t flush = 0;
+	uint32_t i, flush = 0;
 
-	flush |= pipe2flush(mdp4_plane_pipe(mdp4_crtc->plane));
+	for (i = 0; i < ARRAY_SIZE(mdp4_crtc->planes); i++) {
+		struct drm_plane *plane = mdp4_crtc->planes[i];
+		if (plane) {
+			enum mdp4_pipe pipe_id = mdp4_plane_pipe(plane);
+			flush |= pipe2flush(pipe_id);
+		}
+	}
 	flush |= ovlp2flush(mdp4_crtc->ovlp);
 
 	DBG("%s: flush=%08x", mdp4_crtc->name, flush);
@@ -125,17 +140,29 @@ static void crtc_flush(struct drm_crtc *crtc)
 	mdp4_write(mdp4_kms, REG_MDP4_OVERLAY_FLUSH, flush);
 }
 
-static void pageflip_worker(struct work_struct *work)
+static void request_pending(struct drm_crtc *crtc, uint32_t pending)
+{
+	struct mdp4_crtc *mdp4_crtc = to_mdp4_crtc(crtc);
+
+	atomic_or(pending, &mdp4_crtc->pending);
+	mdp4_irq_register(get_kms(crtc), &mdp4_crtc->vblank);
+}
+
+static void pageflip_cb(struct msm_fence_cb *cb)
 {
 	struct mdp4_crtc *mdp4_crtc =
-		container_of(work, struct mdp4_crtc, pageflip_work);
+		container_of(cb, struct mdp4_crtc, pageflip_cb);
 	struct drm_crtc *crtc = &mdp4_crtc->base;
+	struct drm_framebuffer *fb = crtc->fb;
 
-	mdp4_plane_set_scanout(mdp4_crtc->plane, crtc->fb);
+	if (!fb)
+		return;
+
+	mdp4_plane_set_scanout(mdp4_crtc->plane, fb);
 	crtc_flush(crtc);
 
 	/* enable vblank to complete flip: */
-	mdp4_irq_register(get_kms(crtc), &mdp4_crtc->vblank);
+	request_pending(crtc, PENDING_FLIP);
 }
 
 static void unref_fb_worker(struct drm_flip_work *work, void *val)
@@ -205,67 +232,69 @@ static void blend_setup(struct drm_crtc *crtc)
 	struct mdp4_kms *mdp4_kms = get_kms(crtc);
 	int i, ovlp = mdp4_crtc->ovlp;
 	uint32_t mixer_cfg = 0;
-
-	/*
-	 * This probably would also need to be triggered by any attached
-	 * plane when it changes.. for now since we are only using a single
-	 * private plane, the configuration is hard-coded:
-	 */
+	static const enum mdp4_mixer_stage_id stages[] = {
+			STAGE_BASE, STAGE0, STAGE1, STAGE2, STAGE3,
+	};
+	/* statically (for now) map planes to mixer stage (z-order): */
+	static const int idxs[] = {
+			[VG1]  = 1,
+			[VG2]  = 2,
+			[RGB1] = 0,
+			[RGB2] = 0,
+			[RGB3] = 0,
+			[VG3]  = 3,
+			[VG4]  = 4,
+
+	};
+	bool alpha[4]= { false, false, false, false };
 
 	mdp4_write(mdp4_kms, REG_MDP4_OVLP_TRANSP_LOW0(ovlp), 0);
 	mdp4_write(mdp4_kms, REG_MDP4_OVLP_TRANSP_LOW1(ovlp), 0);
 	mdp4_write(mdp4_kms, REG_MDP4_OVLP_TRANSP_HIGH0(ovlp), 0);
 	mdp4_write(mdp4_kms, REG_MDP4_OVLP_TRANSP_HIGH1(ovlp), 0);
 
+	/* TODO single register for all CRTCs, so this won't work properly
+	 * when multiple CRTCs are active..
+	 */
+	for (i = 0; i < ARRAY_SIZE(mdp4_crtc->planes); i++) {
+		struct drm_plane *plane = mdp4_crtc->planes[i];
+		if (plane) {
+			enum mdp4_pipe pipe_id = mdp4_plane_pipe(plane);
+			int idx = idxs[pipe_id];
+			if (idx > 0) {
+				const struct mdp4_format *format =
+					to_mdp4_format(msm_framebuffer_format(plane->fb));
+				alpha[idx-1] = format->alpha_enable;
+			}
+			mixer_cfg |= mixercfg(mdp4_crtc->mixer, pipe_id, stages[idx]);
+		}
+	}
+
+	/* this shouldn't happen.. and seems to cause underflow: */
+	WARN_ON(!mixer_cfg);
+
 	for (i = 0; i < 4; i++) {
-		mdp4_write(mdp4_kms, REG_MDP4_OVLP_STAGE_FG_ALPHA(ovlp, i), 0);
-		mdp4_write(mdp4_kms, REG_MDP4_OVLP_STAGE_BG_ALPHA(ovlp, i), 0);
-		mdp4_write(mdp4_kms, REG_MDP4_OVLP_STAGE_OP(ovlp, i),
-				MDP4_OVLP_STAGE_OP_FG_ALPHA(FG_CONST) |
-				MDP4_OVLP_STAGE_OP_BG_ALPHA(BG_CONST));
-		mdp4_write(mdp4_kms, REG_MDP4_OVLP_STAGE_CO3(ovlp, i), 0);
+		uint32_t op;
+
+		if (alpha[i]) {
+			op = MDP4_OVLP_STAGE_OP_FG_ALPHA(FG_PIXEL) |
+					MDP4_OVLP_STAGE_OP_BG_ALPHA(FG_PIXEL) |
+					MDP4_OVLP_STAGE_OP_BG_INV_ALPHA;
+		} else {
+			op = MDP4_OVLP_STAGE_OP_FG_ALPHA(FG_CONST) |
+					MDP4_OVLP_STAGE_OP_BG_ALPHA(BG_CONST);
+		}
+
+		mdp4_write(mdp4_kms, REG_MDP4_OVLP_STAGE_FG_ALPHA(ovlp, i), 0xff);
+		mdp4_write(mdp4_kms, REG_MDP4_OVLP_STAGE_BG_ALPHA(ovlp, i), 0x00);
+		mdp4_write(mdp4_kms, REG_MDP4_OVLP_STAGE_OP(ovlp, i), op);
+		mdp4_write(mdp4_kms, REG_MDP4_OVLP_STAGE_CO3(ovlp, i), 1);
 		mdp4_write(mdp4_kms, REG_MDP4_OVLP_STAGE_TRANSP_LOW0(ovlp, i), 0);
 		mdp4_write(mdp4_kms, REG_MDP4_OVLP_STAGE_TRANSP_LOW1(ovlp, i), 0);
 		mdp4_write(mdp4_kms, REG_MDP4_OVLP_STAGE_TRANSP_HIGH0(ovlp, i), 0);
 		mdp4_write(mdp4_kms, REG_MDP4_OVLP_STAGE_TRANSP_HIGH1(ovlp, i), 0);
 	}
 
-	/* TODO single register for all CRTCs, so this won't work properly
-	 * when multiple CRTCs are active..
-	 */
-	switch (mdp4_plane_pipe(mdp4_crtc->plane)) {
-	case VG1:
-		mixer_cfg = MDP4_LAYERMIXER_IN_CFG_PIPE0(STAGE_BASE) |
-			COND(mdp4_crtc->mixer == 1, MDP4_LAYERMIXER_IN_CFG_PIPE0_MIXER1);
-		break;
-	case VG2:
-		mixer_cfg = MDP4_LAYERMIXER_IN_CFG_PIPE1(STAGE_BASE) |
-			COND(mdp4_crtc->mixer == 1, MDP4_LAYERMIXER_IN_CFG_PIPE1_MIXER1);
-		break;
-	case RGB1:
-		mixer_cfg = MDP4_LAYERMIXER_IN_CFG_PIPE2(STAGE_BASE) |
-			COND(mdp4_crtc->mixer == 1, MDP4_LAYERMIXER_IN_CFG_PIPE2_MIXER1);
-		break;
-	case RGB2:
-		mixer_cfg = MDP4_LAYERMIXER_IN_CFG_PIPE3(STAGE_BASE) |
-			COND(mdp4_crtc->mixer == 1, MDP4_LAYERMIXER_IN_CFG_PIPE3_MIXER1);
-		break;
-	case RGB3:
-		mixer_cfg = MDP4_LAYERMIXER_IN_CFG_PIPE4(STAGE_BASE) |
-			COND(mdp4_crtc->mixer == 1, MDP4_LAYERMIXER_IN_CFG_PIPE4_MIXER1);
-		break;
-	case VG3:
-		mixer_cfg = MDP4_LAYERMIXER_IN_CFG_PIPE5(STAGE_BASE) |
-			COND(mdp4_crtc->mixer == 1, MDP4_LAYERMIXER_IN_CFG_PIPE5_MIXER1);
-		break;
-	case VG4:
-		mixer_cfg = MDP4_LAYERMIXER_IN_CFG_PIPE6(STAGE_BASE) |
-			COND(mdp4_crtc->mixer == 1, MDP4_LAYERMIXER_IN_CFG_PIPE6_MIXER1);
-		break;
-	default:
-		WARN_ON("invalid pipe");
-		break;
-	}
 	mdp4_write(mdp4_kms, REG_MDP4_LAYERMIXER_IN_CFG, mixer_cfg);
 }
 
@@ -377,6 +406,7 @@ static int mdp4_crtc_page_flip(struct drm_crtc *crtc,
 	struct mdp4_crtc *mdp4_crtc = to_mdp4_crtc(crtc);
 	struct drm_device *dev = crtc->dev;
 	struct drm_gem_object *obj;
+	unsigned long flags;
 
 	if (mdp4_crtc->event) {
 		dev_err(dev->dev, "already pending flip!\n");
@@ -385,11 +415,13 @@ static int mdp4_crtc_page_flip(struct drm_crtc *crtc,
 
 	obj = msm_framebuffer_bo(new_fb, 0);
 
+	spin_lock_irqsave(&dev->event_lock, flags);
 	mdp4_crtc->event = event;
+	spin_unlock_irqrestore(&dev->event_lock, flags);
+
 	update_fb(crtc, true, new_fb);
 
-	return msm_gem_queue_inactive_work(obj,
-			&mdp4_crtc->pageflip_work);
+	return msm_gem_queue_inactive_cb(obj, &mdp4_crtc->pageflip_cb);
 }
 
 static int mdp4_crtc_set_property(struct drm_crtc *crtc,
@@ -498,6 +530,8 @@ static int mdp4_crtc_cursor_set(struct drm_crtc *crtc,
 		drm_gem_object_unreference_unlocked(old_bo);
 	}
 
+	request_pending(crtc, PENDING_CURSOR);
+
 	return 0;
 
 fail:
@@ -542,13 +576,21 @@ static void mdp4_crtc_vblank_irq(struct mdp4_irq *irq, uint32_t irqstatus)
 	struct mdp4_crtc *mdp4_crtc = container_of(irq, struct mdp4_crtc, vblank);
 	struct drm_crtc *crtc = &mdp4_crtc->base;
 	struct msm_drm_private *priv = crtc->dev->dev_private;
+	unsigned pending;
 
-	update_cursor(crtc);
-	complete_flip(crtc, false);
 	mdp4_irq_unregister(get_kms(crtc), &mdp4_crtc->vblank);
 
-	drm_flip_work_commit(&mdp4_crtc->unref_fb_work, priv->wq);
-	drm_flip_work_commit(&mdp4_crtc->unref_cursor_work, priv->wq);
+	pending = atomic_xchg(&mdp4_crtc->pending, 0);
+
+	if (pending & PENDING_FLIP) {
+		complete_flip(crtc, NULL);
+		drm_flip_work_commit(&mdp4_crtc->unref_fb_work, priv->wq);
+	}
+
+	if (pending & PENDING_CURSOR) {
+		update_cursor(crtc);
+		drm_flip_work_commit(&mdp4_crtc->unref_cursor_work, priv->wq);
+	}
 }
 
 static void mdp4_crtc_err_irq(struct mdp4_irq *irq, uint32_t irqstatus)
@@ -565,9 +607,10 @@ uint32_t mdp4_crtc_vblank(struct drm_crtc *crtc)
 	return mdp4_crtc->vblank.irqmask;
 }
 
-void mdp4_crtc_cancel_pending_flip(struct drm_crtc *crtc)
+void mdp4_crtc_cancel_pending_flip(struct drm_crtc *crtc, struct drm_file *file)
 {
-	complete_flip(crtc, true);
+	DBG("cancel: %p", file);
+	complete_flip(crtc, file);
 }
 
 /* set dma config, ie. the format the encoder wants. */
@@ -622,6 +665,32 @@ void mdp4_crtc_set_intf(struct drm_crtc *crtc, enum mdp4_intf intf)
 	mdp4_write(mdp4_kms, REG_MDP4_DISP_INTF_SEL, intf_sel);
 }
 
+static void set_attach(struct drm_crtc *crtc, enum mdp4_pipe pipe_id,
+		struct drm_plane *plane)
+{
+	struct mdp4_crtc *mdp4_crtc = to_mdp4_crtc(crtc);
+
+	BUG_ON(pipe_id >= ARRAY_SIZE(mdp4_crtc->planes));
+
+	if (mdp4_crtc->planes[pipe_id] == plane)
+		return;
+
+	mdp4_crtc->planes[pipe_id] = plane;
+	blend_setup(crtc);
+	if (mdp4_crtc->enabled && (plane != mdp4_crtc->plane))
+		crtc_flush(crtc);
+}
+
+void mdp4_crtc_attach(struct drm_crtc *crtc, struct drm_plane *plane)
+{
+	set_attach(crtc, mdp4_plane_pipe(plane), plane);
+}
+
+void mdp4_crtc_detach(struct drm_crtc *crtc, struct drm_plane *plane)
+{
+	set_attach(crtc, mdp4_plane_pipe(plane), NULL);
+}
+
 static const char *dma_names[] = {
 		"DMA_P", "DMA_S", "DMA_E",
 };
@@ -644,7 +713,6 @@ struct drm_crtc *mdp4_crtc_init(struct drm_device *dev,
 	crtc = &mdp4_crtc->base;
 
 	mdp4_crtc->plane = plane;
-	mdp4_crtc->plane->crtc = crtc;
 
 	mdp4_crtc->ovlp = ovlp_id;
 	mdp4_crtc->dma = dma_id;
@@ -668,7 +736,7 @@ struct drm_crtc *mdp4_crtc_init(struct drm_device *dev,
 	ret = drm_flip_work_init(&mdp4_crtc->unref_cursor_work, 64,
 			"unref cursor", unref_cursor_worker);
 
-	INIT_WORK(&mdp4_crtc->pageflip_work, pageflip_worker);
+	INIT_FENCE_CB(&mdp4_crtc->pageflip_cb, pageflip_cb);
 
 	drm_crtc_init(dev, crtc, &mdp4_crtc_funcs);
 	drm_crtc_helper_add(crtc, &mdp4_crtc_helper_funcs);
diff --git a/drivers/gpu/drm/msm/mdp4/mdp4_format.c b/drivers/gpu/drm/msm/mdp4/mdp4_format.c
index 7b645f2e837a..17330b0927b2 100644
--- a/drivers/gpu/drm/msm/mdp4/mdp4_format.c
+++ b/drivers/gpu/drm/msm/mdp4/mdp4_format.c
@@ -44,6 +44,22 @@ static const struct mdp4_format formats[] = {
 	FMT(BGR565,   0, 5, 6, 5,  2, 0, 1, 0,  false,  true,  2,  3),
 };
 
+uint32_t mdp4_get_formats(enum mdp4_pipe pipe_id, uint32_t *pixel_formats,
+		uint32_t max_formats)
+{
+	uint32_t i;
+	for (i = 0; i < ARRAY_SIZE(formats); i++) {
+		const struct mdp4_format *f = &formats[i];
+
+		if (i == max_formats)
+			break;
+
+		pixel_formats[i] = f->base.pixel_format;
+	}
+
+	return i;
+}
+
 const struct msm_format *mdp4_get_format(struct msm_kms *kms, uint32_t format)
 {
 	int i;
diff --git a/drivers/gpu/drm/msm/mdp4/mdp4_kms.c b/drivers/gpu/drm/msm/mdp4/mdp4_kms.c
index bc7fd11ad8be..8972ac35a43d 100644
--- a/drivers/gpu/drm/msm/mdp4/mdp4_kms.c
+++ b/drivers/gpu/drm/msm/mdp4/mdp4_kms.c
@@ -135,7 +135,7 @@ static void mdp4_preclose(struct msm_kms *kms, struct drm_file *file)
 	unsigned i;
 
 	for (i = 0; i < priv->num_crtcs; i++)
-		mdp4_crtc_cancel_pending_flip(priv->crtcs[i]);
+		mdp4_crtc_cancel_pending_flip(priv->crtcs[i], file);
 }
 
 static void mdp4_destroy(struct msm_kms *kms)
@@ -196,6 +196,23 @@ static int modeset_init(struct mdp4_kms *mdp4_kms)
 	 * for more than just RGB1->DMA_E->DTV->HDMI
 	 */
 
+	/* construct non-private planes: */
+	plane = mdp4_plane_init(dev, VG1, false);
+	if (IS_ERR(plane)) {
+		dev_err(dev->dev, "failed to construct plane for VG1\n");
+		ret = PTR_ERR(plane);
+		goto fail;
+	}
+	priv->planes[priv->num_planes++] = plane;
+
+	plane = mdp4_plane_init(dev, VG2, false);
+	if (IS_ERR(plane)) {
+		dev_err(dev->dev, "failed to construct plane for VG2\n");
+		ret = PTR_ERR(plane);
+		goto fail;
+	}
+	priv->planes[priv->num_planes++] = plane;
+
 	/* the CRTCs get constructed with a private plane: */
 	plane = mdp4_plane_init(dev, RGB1, true);
 	if (IS_ERR(plane)) {
diff --git a/drivers/gpu/drm/msm/mdp4/mdp4_kms.h b/drivers/gpu/drm/msm/mdp4/mdp4_kms.h
index 1e83554955f3..eb015c834087 100644
--- a/drivers/gpu/drm/msm/mdp4/mdp4_kms.h
+++ b/drivers/gpu/drm/msm/mdp4/mdp4_kms.h
@@ -75,8 +75,8 @@ struct mdp4_platform_config {
 
 struct mdp4_format {
 	struct msm_format base;
-	enum mpd4_bpc bpc_r, bpc_g, bpc_b;
-	enum mpd4_bpc_alpha bpc_a;
+	enum mdp4_bpc bpc_r, bpc_g, bpc_b;
+	enum mdp4_bpc_alpha bpc_a;
 	uint8_t unpack[4];
 	bool alpha_enable, unpack_tight;
 	uint8_t cpp, unpack_count;
@@ -93,7 +93,7 @@ static inline u32 mdp4_read(struct mdp4_kms *mdp4_kms, u32 reg)
 	return msm_readl(mdp4_kms->mmio + reg);
 }
 
-static inline uint32_t pipe2flush(enum mpd4_pipe pipe)
+static inline uint32_t pipe2flush(enum mdp4_pipe pipe)
 {
 	switch (pipe) {
 	case VG1:      return MDP4_OVERLAY_FLUSH_VG1;
@@ -133,6 +133,48 @@ static inline uint32_t dma2err(enum mdp4_dma dma)
 	}
 }
 
+static inline uint32_t mixercfg(int mixer, enum mdp4_pipe pipe,
+		enum mdp4_mixer_stage_id stage)
+{
+	uint32_t mixer_cfg = 0;
+
+	switch (pipe) {
+	case VG1:
+		mixer_cfg = MDP4_LAYERMIXER_IN_CFG_PIPE0(stage) |
+			COND(mixer == 1, MDP4_LAYERMIXER_IN_CFG_PIPE0_MIXER1);
+		break;
+	case VG2:
+		mixer_cfg = MDP4_LAYERMIXER_IN_CFG_PIPE1(stage) |
+			COND(mixer == 1, MDP4_LAYERMIXER_IN_CFG_PIPE1_MIXER1);
+		break;
+	case RGB1:
+		mixer_cfg = MDP4_LAYERMIXER_IN_CFG_PIPE2(stage) |
+			COND(mixer == 1, MDP4_LAYERMIXER_IN_CFG_PIPE2_MIXER1);
+		break;
+	case RGB2:
+		mixer_cfg = MDP4_LAYERMIXER_IN_CFG_PIPE3(stage) |
+			COND(mixer == 1, MDP4_LAYERMIXER_IN_CFG_PIPE3_MIXER1);
+		break;
+	case RGB3:
+		mixer_cfg = MDP4_LAYERMIXER_IN_CFG_PIPE4(stage) |
+			COND(mixer == 1, MDP4_LAYERMIXER_IN_CFG_PIPE4_MIXER1);
+		break;
+	case VG3:
+		mixer_cfg = MDP4_LAYERMIXER_IN_CFG_PIPE5(stage) |
+			COND(mixer == 1, MDP4_LAYERMIXER_IN_CFG_PIPE5_MIXER1);
+		break;
+	case VG4:
+		mixer_cfg = MDP4_LAYERMIXER_IN_CFG_PIPE6(stage) |
+			COND(mixer == 1, MDP4_LAYERMIXER_IN_CFG_PIPE6_MIXER1);
+		break;
+	default:
+		WARN_ON("invalid pipe");
+		break;
+	}
+
+	return mixer_cfg;
+}
+
 int mdp4_disable(struct mdp4_kms *mdp4_kms);
 int mdp4_enable(struct mdp4_kms *mdp4_kms);
 
@@ -146,6 +188,8 @@ void mdp4_irq_unregister(struct mdp4_kms *mdp4_kms, struct mdp4_irq *irq);
 int mdp4_enable_vblank(struct msm_kms *kms, struct drm_crtc *crtc);
 void mdp4_disable_vblank(struct msm_kms *kms, struct drm_crtc *crtc);
 
+uint32_t mdp4_get_formats(enum mdp4_pipe pipe_id, uint32_t *formats,
+		uint32_t max_formats);
 const struct msm_format *mdp4_get_format(struct msm_kms *kms, uint32_t format);
 
 void mdp4_plane_install_properties(struct drm_plane *plane,
@@ -158,14 +202,16 @@ int mdp4_plane_mode_set(struct drm_plane *plane,
 		unsigned int crtc_w, unsigned int crtc_h,
 		uint32_t src_x, uint32_t src_y,
 		uint32_t src_w, uint32_t src_h);
-enum mpd4_pipe mdp4_plane_pipe(struct drm_plane *plane);
+enum mdp4_pipe mdp4_plane_pipe(struct drm_plane *plane);
 struct drm_plane *mdp4_plane_init(struct drm_device *dev,
-		enum mpd4_pipe pipe_id, bool private_plane);
+		enum mdp4_pipe pipe_id, bool private_plane);
 
 uint32_t mdp4_crtc_vblank(struct drm_crtc *crtc);
-void mdp4_crtc_cancel_pending_flip(struct drm_crtc *crtc);
+void mdp4_crtc_cancel_pending_flip(struct drm_crtc *crtc, struct drm_file *file);
 void mdp4_crtc_set_config(struct drm_crtc *crtc, uint32_t config);
 void mdp4_crtc_set_intf(struct drm_crtc *crtc, enum mdp4_intf intf);
+void mdp4_crtc_attach(struct drm_crtc *crtc, struct drm_plane *plane);
+void mdp4_crtc_detach(struct drm_crtc *crtc, struct drm_plane *plane);
 struct drm_crtc *mdp4_crtc_init(struct drm_device *dev,
 		struct drm_plane *plane, int id, int ovlp_id,
 		enum mdp4_dma dma_id);
diff --git a/drivers/gpu/drm/msm/mdp4/mdp4_plane.c b/drivers/gpu/drm/msm/mdp4/mdp4_plane.c
index 3468229d58b3..0f0af243f6fc 100644
--- a/drivers/gpu/drm/msm/mdp4/mdp4_plane.c
+++ b/drivers/gpu/drm/msm/mdp4/mdp4_plane.c
@@ -22,7 +22,7 @@ struct mdp4_plane {
 	struct drm_plane base;
 	const char *name;
 
-	enum mpd4_pipe pipe;
+	enum mdp4_pipe pipe;
 
 	uint32_t nformats;
 	uint32_t formats[32];
@@ -61,7 +61,9 @@ static int mdp4_plane_update(struct drm_plane *plane,
 static int mdp4_plane_disable(struct drm_plane *plane)
 {
 	struct mdp4_plane *mdp4_plane = to_mdp4_plane(plane);
-	DBG("%s: TODO", mdp4_plane->name); // XXX
+	DBG("%s: disable", mdp4_plane->name);
+	if (plane->crtc)
+		mdp4_crtc_detach(plane->crtc, plane);
 	return 0;
 }
 
@@ -101,7 +103,7 @@ void mdp4_plane_set_scanout(struct drm_plane *plane,
 {
 	struct mdp4_plane *mdp4_plane = to_mdp4_plane(plane);
 	struct mdp4_kms *mdp4_kms = get_kms(plane);
-	enum mpd4_pipe pipe = mdp4_plane->pipe;
+	enum mdp4_pipe pipe = mdp4_plane->pipe;
 	uint32_t iova;
 
 	mdp4_write(mdp4_kms, REG_MDP4_PIPE_SRC_STRIDE_A(pipe),
@@ -129,7 +131,7 @@ int mdp4_plane_mode_set(struct drm_plane *plane,
 {
 	struct mdp4_plane *mdp4_plane = to_mdp4_plane(plane);
 	struct mdp4_kms *mdp4_kms = get_kms(plane);
-	enum mpd4_pipe pipe = mdp4_plane->pipe;
+	enum mdp4_pipe pipe = mdp4_plane->pipe;
 	const struct mdp4_format *format;
 	uint32_t op_mode = 0;
 	uint32_t phasex_step = MDP4_VG_PHASE_STEP_DEFAULT;
@@ -141,6 +143,10 @@ int mdp4_plane_mode_set(struct drm_plane *plane,
 	src_w = src_w >> 16;
 	src_h = src_h >> 16;
 
+	DBG("%s: FB[%u] %u,%u,%u,%u -> CRTC[%u] %d,%d,%u,%u", mdp4_plane->name,
+			fb->base.id, src_x, src_y, src_w, src_h,
+			crtc->base.id, crtc_x, crtc_y, crtc_w, crtc_h);
+
 	if (src_w != crtc_w) {
 		op_mode |= MDP4_PIPE_OP_MODE_SCALEX_EN;
 		/* TODO calc phasex_step */
@@ -191,7 +197,8 @@ int mdp4_plane_mode_set(struct drm_plane *plane,
 	mdp4_write(mdp4_kms, REG_MDP4_PIPE_PHASEX_STEP(pipe), phasex_step);
 	mdp4_write(mdp4_kms, REG_MDP4_PIPE_PHASEY_STEP(pipe), phasey_step);
 
-	plane->crtc = crtc;
+	/* TODO detach from old crtc (if we had more than one) */
+	mdp4_crtc_attach(crtc, plane);
 
 	return 0;
 }
@@ -202,7 +209,7 @@ static const char *pipe_names[] = {
 		"VG3", "VG4",
 };
 
-enum mpd4_pipe mdp4_plane_pipe(struct drm_plane *plane)
+enum mdp4_pipe mdp4_plane_pipe(struct drm_plane *plane)
 {
 	struct mdp4_plane *mdp4_plane = to_mdp4_plane(plane);
 	return mdp4_plane->pipe;
@@ -210,9 +217,8 @@ enum mpd4_pipe mdp4_plane_pipe(struct drm_plane *plane)
 
 /* initialize plane */
 struct drm_plane *mdp4_plane_init(struct drm_device *dev,
-		enum mpd4_pipe pipe_id, bool private_plane)
+		enum mdp4_pipe pipe_id, bool private_plane)
 {
-	struct msm_drm_private *priv = dev->dev_private;
 	struct drm_plane *plane = NULL;
 	struct mdp4_plane *mdp4_plane;
 	int ret;
@@ -228,8 +234,12 @@ struct drm_plane *mdp4_plane_init(struct drm_device *dev,
 	mdp4_plane->pipe = pipe_id;
 	mdp4_plane->name = pipe_names[pipe_id];
 
-	drm_plane_init(dev, plane, (1 << priv->num_crtcs) - 1, &mdp4_plane_funcs,
-			mdp4_plane->formats, mdp4_plane->nformats, private_plane);
+	mdp4_plane->nformats = mdp4_get_formats(pipe_id, mdp4_plane->formats,
+			ARRAY_SIZE(mdp4_plane->formats));
+
+	drm_plane_init(dev, plane, 0xff, &mdp4_plane_funcs,
+			mdp4_plane->formats, mdp4_plane->nformats,
+			private_plane);
 
 	mdp4_plane_install_properties(plane, &plane->base);
 
diff --git a/drivers/gpu/drm/msm/msm_drv.c b/drivers/gpu/drm/msm/msm_drv.c
index b3a2f1629041..86537692e45c 100644
--- a/drivers/gpu/drm/msm/msm_drv.c
+++ b/drivers/gpu/drm/msm/msm_drv.c
@@ -187,6 +187,7 @@ static int msm_load(struct drm_device *dev, unsigned long flags)
 	init_waitqueue_head(&priv->fence_event);
 
 	INIT_LIST_HEAD(&priv->inactive_list);
+	INIT_LIST_HEAD(&priv->fence_cbs);
 
 	drm_mode_config_init(dev);
 
@@ -539,15 +540,36 @@ int msm_wait_fence_interruptable(struct drm_device *dev, uint32_t fence,
 	return ret;
 }
 
-/* call under struct_mutex */
+/* called from workqueue */
 void msm_update_fence(struct drm_device *dev, uint32_t fence)
 {
 	struct msm_drm_private *priv = dev->dev_private;
 
-	if (fence > priv->completed_fence) {
-		priv->completed_fence = fence;
-		wake_up_all(&priv->fence_event);
+	mutex_lock(&dev->struct_mutex);
+	priv->completed_fence = max(fence, priv->completed_fence);
+
+	while (!list_empty(&priv->fence_cbs)) {
+		struct msm_fence_cb *cb;
+
+		cb = list_first_entry(&priv->fence_cbs,
+				struct msm_fence_cb, work.entry);
+
+		if (cb->fence > priv->completed_fence)
+			break;
+
+		list_del_init(&cb->work.entry);
+		queue_work(priv->wq, &cb->work);
 	}
+
+	mutex_unlock(&dev->struct_mutex);
+
+	wake_up_all(&priv->fence_event);
+}
+
+void __msm_fence_worker(struct work_struct *work)
+{
+	struct msm_fence_cb *cb = container_of(work, struct msm_fence_cb, work);
+	cb->func(cb);
 }
 
 /*
@@ -650,13 +672,13 @@ static int msm_ioctl_wait_fence(struct drm_device *dev, void *data,
 }
 
 static const struct drm_ioctl_desc msm_ioctls[] = {
-	DRM_IOCTL_DEF_DRV(MSM_GET_PARAM,    msm_ioctl_get_param,    DRM_UNLOCKED|DRM_AUTH),
-	DRM_IOCTL_DEF_DRV(MSM_GEM_NEW,      msm_ioctl_gem_new,      DRM_UNLOCKED|DRM_AUTH),
-	DRM_IOCTL_DEF_DRV(MSM_GEM_INFO,     msm_ioctl_gem_info,     DRM_UNLOCKED|DRM_AUTH),
-	DRM_IOCTL_DEF_DRV(MSM_GEM_CPU_PREP, msm_ioctl_gem_cpu_prep, DRM_UNLOCKED|DRM_AUTH),
-	DRM_IOCTL_DEF_DRV(MSM_GEM_CPU_FINI, msm_ioctl_gem_cpu_fini, DRM_UNLOCKED|DRM_AUTH),
-	DRM_IOCTL_DEF_DRV(MSM_GEM_SUBMIT,   msm_ioctl_gem_submit,   DRM_UNLOCKED|DRM_AUTH),
-	DRM_IOCTL_DEF_DRV(MSM_WAIT_FENCE,   msm_ioctl_wait_fence,   DRM_UNLOCKED|DRM_AUTH),
+	DRM_IOCTL_DEF_DRV(MSM_GET_PARAM,    msm_ioctl_get_param,    DRM_UNLOCKED|DRM_AUTH|DRM_RENDER_ALLOW),
+	DRM_IOCTL_DEF_DRV(MSM_GEM_NEW,      msm_ioctl_gem_new,      DRM_UNLOCKED|DRM_AUTH|DRM_RENDER_ALLOW),
+	DRM_IOCTL_DEF_DRV(MSM_GEM_INFO,     msm_ioctl_gem_info,     DRM_UNLOCKED|DRM_AUTH|DRM_RENDER_ALLOW),
+	DRM_IOCTL_DEF_DRV(MSM_GEM_CPU_PREP, msm_ioctl_gem_cpu_prep, DRM_UNLOCKED|DRM_AUTH|DRM_RENDER_ALLOW),
+	DRM_IOCTL_DEF_DRV(MSM_GEM_CPU_FINI, msm_ioctl_gem_cpu_fini, DRM_UNLOCKED|DRM_AUTH|DRM_RENDER_ALLOW),
+	DRM_IOCTL_DEF_DRV(MSM_GEM_SUBMIT,   msm_ioctl_gem_submit,   DRM_UNLOCKED|DRM_AUTH|DRM_RENDER_ALLOW),
+	DRM_IOCTL_DEF_DRV(MSM_WAIT_FENCE,   msm_ioctl_wait_fence,   DRM_UNLOCKED|DRM_AUTH|DRM_RENDER_ALLOW),
 };
 
 static const struct vm_operations_struct vm_ops = {
@@ -680,7 +702,11 @@ static const struct file_operations fops = {
 };
 
 static struct drm_driver msm_driver = {
-	.driver_features    = DRIVER_HAVE_IRQ | DRIVER_GEM | DRIVER_MODESET,
+	.driver_features    = DRIVER_HAVE_IRQ |
+				DRIVER_GEM |
+				DRIVER_PRIME |
+				DRIVER_RENDER |
+				DRIVER_MODESET,
 	.load               = msm_load,
 	.unload             = msm_unload,
 	.open               = msm_open,
@@ -698,6 +724,16 @@ static struct drm_driver msm_driver = {
 	.dumb_create        = msm_gem_dumb_create,
 	.dumb_map_offset    = msm_gem_dumb_map_offset,
 	.dumb_destroy       = drm_gem_dumb_destroy,
+	.prime_handle_to_fd = drm_gem_prime_handle_to_fd,
+	.prime_fd_to_handle = drm_gem_prime_fd_to_handle,
+	.gem_prime_export   = drm_gem_prime_export,
+	.gem_prime_import   = drm_gem_prime_import,
+	.gem_prime_pin      = msm_gem_prime_pin,
+	.gem_prime_unpin    = msm_gem_prime_unpin,
+	.gem_prime_get_sg_table = msm_gem_prime_get_sg_table,
+	.gem_prime_import_sg_table = msm_gem_prime_import_sg_table,
+	.gem_prime_vmap     = msm_gem_prime_vmap,
+	.gem_prime_vunmap   = msm_gem_prime_vunmap,
 #ifdef CONFIG_DEBUG_FS
 	.debugfs_init       = msm_debugfs_init,
 	.debugfs_cleanup    = msm_debugfs_cleanup,
diff --git a/drivers/gpu/drm/msm/msm_drv.h b/drivers/gpu/drm/msm/msm_drv.h
index df8f1d084bc1..d39f0862b19e 100644
--- a/drivers/gpu/drm/msm/msm_drv.h
+++ b/drivers/gpu/drm/msm/msm_drv.h
@@ -73,10 +73,16 @@ struct msm_drm_private {
 
 	struct workqueue_struct *wq;
 
+	/* callbacks deferred until bo is inactive: */
+	struct list_head fence_cbs;
+
 	/* registered IOMMU domains: */
 	unsigned int num_iommus;
 	struct iommu_domain *iommus[NUM_DOMAINS];
 
+	unsigned int num_planes;
+	struct drm_plane *planes[8];
+
 	unsigned int num_crtcs;
 	struct drm_crtc *crtcs[8];
 
@@ -94,6 +100,20 @@ struct msm_format {
 	uint32_t pixel_format;
 };
 
+/* callback from wq once fence has passed: */
+struct msm_fence_cb {
+	struct work_struct work;
+	uint32_t fence;
+	void (*func)(struct msm_fence_cb *cb);
+};
+
+void __msm_fence_worker(struct work_struct *work);
+
+#define INIT_FENCE_CB(_cb, _func)  do {                     \
+		INIT_WORK(&(_cb)->work, __msm_fence_worker); \
+		(_cb)->func = _func;                         \
+	} while (0)
+
 /* As there are different display controller blocks depending on the
  * snapdragon version, the kms support is split out and the appropriate
  * implementation is loaded at runtime.  The kms module is responsible
@@ -141,17 +161,24 @@ uint64_t msm_gem_mmap_offset(struct drm_gem_object *obj);
 int msm_gem_get_iova_locked(struct drm_gem_object *obj, int id,
 		uint32_t *iova);
 int msm_gem_get_iova(struct drm_gem_object *obj, int id, uint32_t *iova);
+struct page **msm_gem_get_pages(struct drm_gem_object *obj);
+void msm_gem_put_pages(struct drm_gem_object *obj);
 void msm_gem_put_iova(struct drm_gem_object *obj, int id);
 int msm_gem_dumb_create(struct drm_file *file, struct drm_device *dev,
 		struct drm_mode_create_dumb *args);
-int msm_gem_dumb_destroy(struct drm_file *file, struct drm_device *dev,
-		uint32_t handle);
 int msm_gem_dumb_map_offset(struct drm_file *file, struct drm_device *dev,
 		uint32_t handle, uint64_t *offset);
+struct sg_table *msm_gem_prime_get_sg_table(struct drm_gem_object *obj);
+void *msm_gem_prime_vmap(struct drm_gem_object *obj);
+void msm_gem_prime_vunmap(struct drm_gem_object *obj, void *vaddr);
+struct drm_gem_object *msm_gem_prime_import_sg_table(struct drm_device *dev,
+		size_t size, struct sg_table *sg);
+int msm_gem_prime_pin(struct drm_gem_object *obj);
+void msm_gem_prime_unpin(struct drm_gem_object *obj);
 void *msm_gem_vaddr_locked(struct drm_gem_object *obj);
 void *msm_gem_vaddr(struct drm_gem_object *obj);
-int msm_gem_queue_inactive_work(struct drm_gem_object *obj,
-		struct work_struct *work);
+int msm_gem_queue_inactive_cb(struct drm_gem_object *obj,
+		struct msm_fence_cb *cb);
 void msm_gem_move_to_active(struct drm_gem_object *obj,
 		struct msm_gpu *gpu, bool write, uint32_t fence);
 void msm_gem_move_to_inactive(struct drm_gem_object *obj);
@@ -163,6 +190,8 @@ int msm_gem_new_handle(struct drm_device *dev, struct drm_file *file,
 		uint32_t size, uint32_t flags, uint32_t *handle);
 struct drm_gem_object *msm_gem_new(struct drm_device *dev,
 		uint32_t size, uint32_t flags);
+struct drm_gem_object *msm_gem_import(struct drm_device *dev,
+		uint32_t size, struct sg_table *sgt);
 
 struct drm_gem_object *msm_framebuffer_bo(struct drm_framebuffer *fb, int plane);
 const struct msm_format *msm_framebuffer_format(struct drm_framebuffer *fb);
diff --git a/drivers/gpu/drm/msm/msm_gem.c b/drivers/gpu/drm/msm/msm_gem.c
index 2bae46c66a30..e587d251c590 100644
--- a/drivers/gpu/drm/msm/msm_gem.c
+++ b/drivers/gpu/drm/msm/msm_gem.c
@@ -17,6 +17,7 @@
 
 #include <linux/spinlock.h>
 #include <linux/shmem_fs.h>
+#include <linux/dma-buf.h>
 
 #include "msm_drv.h"
 #include "msm_gem.h"
@@ -77,6 +78,21 @@ static void put_pages(struct drm_gem_object *obj)
 	}
 }
 
+struct page **msm_gem_get_pages(struct drm_gem_object *obj)
+{
+	struct drm_device *dev = obj->dev;
+	struct page **p;
+	mutex_lock(&dev->struct_mutex);
+	p = get_pages(obj);
+	mutex_unlock(&dev->struct_mutex);
+	return p;
+}
+
+void msm_gem_put_pages(struct drm_gem_object *obj)
+{
+	/* when we start tracking the pin count, then do something here */
+}
+
 int msm_gem_mmap_obj(struct drm_gem_object *obj,
 		struct vm_area_struct *vma)
 {
@@ -162,6 +178,11 @@ out:
 	case 0:
 	case -ERESTARTSYS:
 	case -EINTR:
+	case -EBUSY:
+		/*
+		 * EBUSY is ok: this just means that another thread
+		 * already did the job.
+		 */
 		return VM_FAULT_NOPAGE;
 	case -ENOMEM:
 		return VM_FAULT_OOM;
@@ -293,7 +314,17 @@ int msm_gem_get_iova_locked(struct drm_gem_object *obj, int id,
 
 int msm_gem_get_iova(struct drm_gem_object *obj, int id, uint32_t *iova)
 {
+	struct msm_gem_object *msm_obj = to_msm_bo(obj);
 	int ret;
+
+	/* this is safe right now because we don't unmap until the
+	 * bo is deleted:
+	 */
+	if (msm_obj->domain[id].iova) {
+		*iova = msm_obj->domain[id].iova;
+		return 0;
+	}
+
 	mutex_lock(&obj->dev->struct_mutex);
 	ret = msm_gem_get_iova_locked(obj, id, iova);
 	mutex_unlock(&obj->dev->struct_mutex);
@@ -363,8 +394,11 @@ void *msm_gem_vaddr(struct drm_gem_object *obj)
 	return ret;
 }
 
-int msm_gem_queue_inactive_work(struct drm_gem_object *obj,
-		struct work_struct *work)
+/* setup callback for when bo is no longer busy..
+ * TODO probably want to differentiate read vs write..
+ */
+int msm_gem_queue_inactive_cb(struct drm_gem_object *obj,
+		struct msm_fence_cb *cb)
 {
 	struct drm_device *dev = obj->dev;
 	struct msm_drm_private *priv = dev->dev_private;
@@ -372,12 +406,13 @@ int msm_gem_queue_inactive_work(struct drm_gem_object *obj,
 	int ret = 0;
 
 	mutex_lock(&dev->struct_mutex);
-	if (!list_empty(&work->entry)) {
+	if (!list_empty(&cb->work.entry)) {
 		ret = -EINVAL;
 	} else if (is_active(msm_obj)) {
-		list_add_tail(&work->entry, &msm_obj->inactive_work);
+		cb->fence = max(msm_obj->read_fence, msm_obj->write_fence);
+		list_add_tail(&cb->work.entry, &priv->fence_cbs);
 	} else {
-		queue_work(priv->wq, work);
+		queue_work(priv->wq, &cb->work);
 	}
 	mutex_unlock(&dev->struct_mutex);
 
@@ -410,16 +445,6 @@ void msm_gem_move_to_inactive(struct drm_gem_object *obj)
 	msm_obj->write_fence = 0;
 	list_del_init(&msm_obj->mm_list);
 	list_add_tail(&msm_obj->mm_list, &priv->inactive_list);
-
-	while (!list_empty(&msm_obj->inactive_work)) {
-		struct work_struct *work;
-
-		work = list_first_entry(&msm_obj->inactive_work,
-				struct work_struct, entry);
-
-		list_del_init(&work->entry);
-		queue_work(priv->wq, work);
-	}
 }
 
 int msm_gem_cpu_prep(struct drm_gem_object *obj, uint32_t op,
@@ -510,10 +535,21 @@ void msm_gem_free_object(struct drm_gem_object *obj)
 
 	drm_gem_free_mmap_offset(obj);
 
-	if (msm_obj->vaddr)
-		vunmap(msm_obj->vaddr);
+	if (obj->import_attach) {
+		if (msm_obj->vaddr)
+			dma_buf_vunmap(obj->import_attach->dmabuf, msm_obj->vaddr);
 
-	put_pages(obj);
+		/* Don't drop the pages for imported dmabuf, as they are not
+		 * ours, just free the array we allocated:
+		 */
+		if (msm_obj->pages)
+			drm_free_large(msm_obj->pages);
+
+	} else {
+		if (msm_obj->vaddr)
+			vunmap(msm_obj->vaddr);
+		put_pages(obj);
+	}
 
 	if (msm_obj->resv == &msm_obj->_resv)
 		reservation_object_fini(msm_obj->resv);
@@ -549,17 +585,12 @@ int msm_gem_new_handle(struct drm_device *dev, struct drm_file *file,
 	return ret;
 }
 
-struct drm_gem_object *msm_gem_new(struct drm_device *dev,
-		uint32_t size, uint32_t flags)
+static int msm_gem_new_impl(struct drm_device *dev,
+		uint32_t size, uint32_t flags,
+		struct drm_gem_object **obj)
 {
 	struct msm_drm_private *priv = dev->dev_private;
 	struct msm_gem_object *msm_obj;
-	struct drm_gem_object *obj = NULL;
-	int ret;
-
-	WARN_ON(!mutex_is_locked(&dev->struct_mutex));
-
-	size = PAGE_ALIGN(size);
 
 	switch (flags & MSM_BO_CACHE_MASK) {
 	case MSM_BO_UNCACHED:
@@ -569,21 +600,12 @@ struct drm_gem_object *msm_gem_new(struct drm_device *dev,
 	default:
 		dev_err(dev->dev, "invalid cache flag: %x\n",
 				(flags & MSM_BO_CACHE_MASK));
-		ret = -EINVAL;
-		goto fail;
+		return -EINVAL;
 	}
 
 	msm_obj = kzalloc(sizeof(*msm_obj), GFP_KERNEL);
-	if (!msm_obj) {
-		ret = -ENOMEM;
-		goto fail;
-	}
-
-	obj = &msm_obj->base;
-
-	ret = drm_gem_object_init(dev, obj, size);
-	if (ret)
-		goto fail;
+	if (!msm_obj)
+		return -ENOMEM;
 
 	msm_obj->flags = flags;
 
@@ -591,9 +613,69 @@ struct drm_gem_object *msm_gem_new(struct drm_device *dev,
 	reservation_object_init(msm_obj->resv);
 
 	INIT_LIST_HEAD(&msm_obj->submit_entry);
-	INIT_LIST_HEAD(&msm_obj->inactive_work);
 	list_add_tail(&msm_obj->mm_list, &priv->inactive_list);
 
+	*obj = &msm_obj->base;
+
+	return 0;
+}
+
+struct drm_gem_object *msm_gem_new(struct drm_device *dev,
+		uint32_t size, uint32_t flags)
+{
+	struct drm_gem_object *obj;
+	int ret;
+
+	WARN_ON(!mutex_is_locked(&dev->struct_mutex));
+
+	size = PAGE_ALIGN(size);
+
+	ret = msm_gem_new_impl(dev, size, flags, &obj);
+	if (ret)
+		goto fail;
+
+	ret = drm_gem_object_init(dev, obj, size);
+	if (ret)
+		goto fail;
+
+	return obj;
+
+fail:
+	if (obj)
+		drm_gem_object_unreference_unlocked(obj);
+
+	return ERR_PTR(ret);
+}
+
+struct drm_gem_object *msm_gem_import(struct drm_device *dev,
+		uint32_t size, struct sg_table *sgt)
+{
+	struct msm_gem_object *msm_obj;
+	struct drm_gem_object *obj;
+	int ret, npages;
+
+	size = PAGE_ALIGN(size);
+
+	ret = msm_gem_new_impl(dev, size, MSM_BO_WC, &obj);
+	if (ret)
+		goto fail;
+
+	drm_gem_private_object_init(dev, obj, size);
+
+	npages = size / PAGE_SIZE;
+
+	msm_obj = to_msm_bo(obj);
+	msm_obj->sgt = sgt;
+	msm_obj->pages = drm_malloc_ab(npages, sizeof(struct page *));
+	if (!msm_obj->pages) {
+		ret = -ENOMEM;
+		goto fail;
+	}
+
+	ret = drm_prime_sg_to_page_addr_arrays(sgt, msm_obj->pages, NULL, npages);
+	if (ret)
+		goto fail;
+
 	return obj;
 
 fail:
diff --git a/drivers/gpu/drm/msm/msm_gem.h b/drivers/gpu/drm/msm/msm_gem.h
index 0676f32e2c6a..f4f23a578d9d 100644
--- a/drivers/gpu/drm/msm/msm_gem.h
+++ b/drivers/gpu/drm/msm/msm_gem.h
@@ -45,9 +45,6 @@ struct msm_gem_object {
 	 */
 	struct list_head submit_entry;
 
-	/* work defered until bo is inactive: */
-	struct list_head inactive_work;
-
 	struct page **pages;
 	struct sg_table *sgt;
 	void *vaddr;
diff --git a/drivers/gpu/drm/msm/msm_gem_prime.c b/drivers/gpu/drm/msm/msm_gem_prime.c
new file mode 100644
index 000000000000..d48f9fc5129b
--- /dev/null
+++ b/drivers/gpu/drm/msm/msm_gem_prime.c
@@ -0,0 +1,56 @@
+/*
+ * Copyright (C) 2013 Red Hat
+ * Author: Rob Clark <robdclark@gmail.com>
+ *
+ * This program is free software; you can redistribute it and/or modify it
+ * under the terms of the GNU General Public License version 2 as published by
+ * the Free Software Foundation.
+ *
+ * This program is distributed in the hope that it will be useful, but WITHOUT
+ * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or
+ * FITNESS FOR A PARTICULAR PURPOSE.  See the GNU General Public License for
+ * more details.
+ *
+ * You should have received a copy of the GNU General Public License along with
+ * this program.  If not, see <http://www.gnu.org/licenses/>.
+ */
+
+#include "msm_drv.h"
+#include "msm_gem.h"
+
+
+struct sg_table *msm_gem_prime_get_sg_table(struct drm_gem_object *obj)
+{
+	struct msm_gem_object *msm_obj = to_msm_bo(obj);
+	BUG_ON(!msm_obj->sgt);  /* should have already pinned! */
+	return msm_obj->sgt;
+}
+
+void *msm_gem_prime_vmap(struct drm_gem_object *obj)
+{
+	return msm_gem_vaddr(obj);
+}
+
+void msm_gem_prime_vunmap(struct drm_gem_object *obj, void *vaddr)
+{
+	/* TODO msm_gem_vunmap() */
+}
+
+struct drm_gem_object *msm_gem_prime_import_sg_table(struct drm_device *dev,
+		size_t size, struct sg_table *sg)
+{
+	return msm_gem_import(dev, size, sg);
+}
+
+int msm_gem_prime_pin(struct drm_gem_object *obj)
+{
+	if (!obj->import_attach)
+		msm_gem_get_pages(obj);
+	return 0;
+}
+
+void msm_gem_prime_unpin(struct drm_gem_object *obj)
+{
+	if (!obj->import_attach)
+		msm_gem_put_pages(obj);
+}
diff --git a/drivers/gpu/drm/msm/msm_gpu.c b/drivers/gpu/drm/msm/msm_gpu.c
index 3bab937965d1..4583d61556f5 100644
--- a/drivers/gpu/drm/msm/msm_gpu.c
+++ b/drivers/gpu/drm/msm/msm_gpu.c
@@ -268,6 +268,8 @@ static void retire_worker(struct work_struct *work)
 	struct drm_device *dev = gpu->dev;
 	uint32_t fence = gpu->funcs->last_fence(gpu);
 
+	msm_update_fence(gpu->dev, fence);
+
 	mutex_lock(&dev->struct_mutex);
 
 	while (!list_empty(&gpu->active_list)) {
@@ -287,8 +289,6 @@ static void retire_worker(struct work_struct *work)
 		}
 	}
 
-	msm_update_fence(gpu->dev, fence);
-
 	mutex_unlock(&dev->struct_mutex);
 }
 
diff --git a/drivers/gpu/drm/nouveau/Kconfig b/drivers/gpu/drm/nouveau/Kconfig
index ff80f12480ea..7cf787d697b1 100644
--- a/drivers/gpu/drm/nouveau/Kconfig
+++ b/drivers/gpu/drm/nouveau/Kconfig
@@ -3,6 +3,7 @@ config DRM_NOUVEAU
 	depends on DRM && PCI
         select FW_LOADER
 	select DRM_KMS_HELPER
+	select DRM_KMS_FB_HELPER
 	select DRM_TTM
 	select FB_CFB_FILLRECT
 	select FB_CFB_COPYAREA
diff --git a/drivers/gpu/drm/nouveau/Makefile b/drivers/gpu/drm/nouveau/Makefile
index d939a1da3203..edcf801613e6 100644
--- a/drivers/gpu/drm/nouveau/Makefile
+++ b/drivers/gpu/drm/nouveau/Makefile
@@ -28,7 +28,9 @@ nouveau-y += core/subdev/bar/nv50.o
 nouveau-y += core/subdev/bar/nvc0.o
 nouveau-y += core/subdev/bios/base.o
 nouveau-y += core/subdev/bios/bit.o
+nouveau-y += core/subdev/bios/boost.o
 nouveau-y += core/subdev/bios/conn.o
+nouveau-y += core/subdev/bios/cstep.o
 nouveau-y += core/subdev/bios/dcb.o
 nouveau-y += core/subdev/bios/disp.o
 nouveau-y += core/subdev/bios/dp.o
@@ -39,17 +41,26 @@ nouveau-y += core/subdev/bios/init.o
 nouveau-y += core/subdev/bios/mxm.o
 nouveau-y += core/subdev/bios/perf.o
 nouveau-y += core/subdev/bios/pll.o
+nouveau-y += core/subdev/bios/rammap.o
+nouveau-y += core/subdev/bios/timing.o
 nouveau-y += core/subdev/bios/therm.o
+nouveau-y += core/subdev/bios/vmap.o
+nouveau-y += core/subdev/bios/volt.o
 nouveau-y += core/subdev/bios/xpio.o
+nouveau-y += core/subdev/bus/hwsq.o
 nouveau-y += core/subdev/bus/nv04.o
 nouveau-y += core/subdev/bus/nv31.o
 nouveau-y += core/subdev/bus/nv50.o
+nouveau-y += core/subdev/bus/nv94.o
 nouveau-y += core/subdev/bus/nvc0.o
+nouveau-y += core/subdev/clock/base.o
 nouveau-y += core/subdev/clock/nv04.o
 nouveau-y += core/subdev/clock/nv40.o
 nouveau-y += core/subdev/clock/nv50.o
+nouveau-y += core/subdev/clock/nv84.o
 nouveau-y += core/subdev/clock/nva3.o
 nouveau-y += core/subdev/clock/nvc0.o
+nouveau-y += core/subdev/clock/nve0.o
 nouveau-y += core/subdev/clock/pllnv04.o
 nouveau-y += core/subdev/clock/pllnva3.o
 nouveau-y += core/subdev/devinit/base.o
@@ -78,7 +89,12 @@ nouveau-y += core/subdev/fb/nv47.o
 nouveau-y += core/subdev/fb/nv49.o
 nouveau-y += core/subdev/fb/nv4e.o
 nouveau-y += core/subdev/fb/nv50.o
+nouveau-y += core/subdev/fb/nv84.o
+nouveau-y += core/subdev/fb/nva3.o
+nouveau-y += core/subdev/fb/nvaa.o
+nouveau-y += core/subdev/fb/nvaf.o
 nouveau-y += core/subdev/fb/nvc0.o
+nouveau-y += core/subdev/fb/nve0.o
 nouveau-y += core/subdev/fb/ramnv04.o
 nouveau-y += core/subdev/fb/ramnv10.o
 nouveau-y += core/subdev/fb/ramnv1a.o
@@ -89,7 +105,12 @@ nouveau-y += core/subdev/fb/ramnv44.o
 nouveau-y += core/subdev/fb/ramnv49.o
 nouveau-y += core/subdev/fb/ramnv4e.o
 nouveau-y += core/subdev/fb/ramnv50.o
+nouveau-y += core/subdev/fb/ramnva3.o
+nouveau-y += core/subdev/fb/ramnvaa.o
 nouveau-y += core/subdev/fb/ramnvc0.o
+nouveau-y += core/subdev/fb/ramnve0.o
+nouveau-y += core/subdev/fb/sddr3.o
+nouveau-y += core/subdev/fb/gddr5.o
 nouveau-y += core/subdev/gpio/base.o
 nouveau-y += core/subdev/gpio/nv10.o
 nouveau-y += core/subdev/gpio/nv50.o
@@ -113,13 +134,22 @@ nouveau-y += core/subdev/instmem/nv50.o
 nouveau-y += core/subdev/ltcg/nvc0.o
 nouveau-y += core/subdev/mc/base.o
 nouveau-y += core/subdev/mc/nv04.o
+nouveau-y += core/subdev/mc/nv40.o
 nouveau-y += core/subdev/mc/nv44.o
 nouveau-y += core/subdev/mc/nv50.o
+nouveau-y += core/subdev/mc/nv94.o
 nouveau-y += core/subdev/mc/nv98.o
 nouveau-y += core/subdev/mc/nvc0.o
+nouveau-y += core/subdev/mc/nvc3.o
 nouveau-y += core/subdev/mxm/base.o
 nouveau-y += core/subdev/mxm/mxms.o
 nouveau-y += core/subdev/mxm/nv50.o
+nouveau-y += core/subdev/pwr/base.o
+nouveau-y += core/subdev/pwr/memx.o
+nouveau-y += core/subdev/pwr/nva3.o
+nouveau-y += core/subdev/pwr/nvc0.o
+nouveau-y += core/subdev/pwr/nvd0.o
+nouveau-y += core/subdev/pwr/nv108.o
 nouveau-y += core/subdev/therm/base.o
 nouveau-y += core/subdev/therm/fan.o
 nouveau-y += core/subdev/therm/fannil.o
@@ -140,6 +170,9 @@ nouveau-y += core/subdev/vm/nv41.o
 nouveau-y += core/subdev/vm/nv44.o
 nouveau-y += core/subdev/vm/nv50.o
 nouveau-y += core/subdev/vm/nvc0.o
+nouveau-y += core/subdev/volt/base.o
+nouveau-y += core/subdev/volt/gpio.o
+nouveau-y += core/subdev/volt/nv40.o
 
 nouveau-y += core/engine/falcon.o
 nouveau-y += core/engine/xtensa.o
@@ -158,6 +191,7 @@ nouveau-y += core/engine/copy/nve0.o
 nouveau-y += core/engine/crypt/nv84.o
 nouveau-y += core/engine/crypt/nv98.o
 nouveau-y += core/engine/device/base.o
+nouveau-y += core/engine/device/ctrl.o
 nouveau-y += core/engine/device/nv04.o
 nouveau-y += core/engine/device/nv10.o
 nouveau-y += core/engine/device/nv20.o
@@ -227,8 +261,18 @@ nouveau-y += core/engine/graph/nve4.o
 nouveau-y += core/engine/graph/nvf0.o
 nouveau-y += core/engine/mpeg/nv31.o
 nouveau-y += core/engine/mpeg/nv40.o
+nouveau-y += core/engine/mpeg/nv44.o
 nouveau-y += core/engine/mpeg/nv50.o
 nouveau-y += core/engine/mpeg/nv84.o
+nouveau-y += core/engine/perfmon/base.o
+nouveau-y += core/engine/perfmon/daemon.o
+nouveau-y += core/engine/perfmon/nv40.o
+nouveau-y += core/engine/perfmon/nv50.o
+nouveau-y += core/engine/perfmon/nv84.o
+nouveau-y += core/engine/perfmon/nva3.o
+nouveau-y += core/engine/perfmon/nvc0.o
+nouveau-y += core/engine/perfmon/nve0.o
+nouveau-y += core/engine/perfmon/nvf0.o
 nouveau-y += core/engine/ppp/nv98.o
 nouveau-y += core/engine/ppp/nvc0.o
 nouveau-y += core/engine/software/nv04.o
@@ -260,9 +304,7 @@ include $(src)/dispnv04/Makefile
 nouveau-y += nv50_display.o
 
 # drm/pm
-nouveau-y += nouveau_pm.o nouveau_volt.o nouveau_perf.o
-nouveau-y += nv04_pm.o nv40_pm.o nv50_pm.o nva3_pm.o nvc0_pm.o
-nouveau-y += nouveau_mem.o
+nouveau-y += nouveau_hwmon.o nouveau_sysfs.o
 
 # other random bits
 nouveau-$(CONFIG_COMPAT) += nouveau_ioc32.o
diff --git a/drivers/gpu/drm/nouveau/core/core/event.c b/drivers/gpu/drm/nouveau/core/core/event.c
index 7eb81c1b6fab..3f3c76581a9e 100644
--- a/drivers/gpu/drm/nouveau/core/core/event.c
+++ b/drivers/gpu/drm/nouveau/core/core/event.c
@@ -23,62 +23,114 @@
 #include <core/os.h>
 #include <core/event.h>
 
-static void
-nouveau_event_put_locked(struct nouveau_event *event, int index,
-			 struct nouveau_eventh *handler)
+void
+nouveau_event_put(struct nouveau_eventh *handler)
 {
-	if (!--event->index[index].refs) {
-		if (event->disable)
-			event->disable(event, index);
+	struct nouveau_event *event = handler->event;
+	unsigned long flags;
+	if (__test_and_clear_bit(NVKM_EVENT_ENABLE, &handler->flags)) {
+		spin_lock_irqsave(&event->refs_lock, flags);
+		if (!--event->index[handler->index].refs) {
+			if (event->disable)
+				event->disable(event, handler->index);
+		}
+		spin_unlock_irqrestore(&event->refs_lock, flags);
 	}
-	list_del(&handler->head);
 }
 
 void
-nouveau_event_put(struct nouveau_event *event, int index,
-		  struct nouveau_eventh *handler)
+nouveau_event_get(struct nouveau_eventh *handler)
 {
+	struct nouveau_event *event = handler->event;
 	unsigned long flags;
+	if (!__test_and_set_bit(NVKM_EVENT_ENABLE, &handler->flags)) {
+		spin_lock_irqsave(&event->refs_lock, flags);
+		if (!event->index[handler->index].refs++) {
+			if (event->enable)
+				event->enable(event, handler->index);
+		}
+		spin_unlock_irqrestore(&event->refs_lock, flags);
+	}
+}
 
-	spin_lock_irqsave(&event->lock, flags);
-	if (index < event->index_nr)
-		nouveau_event_put_locked(event, index, handler);
-	spin_unlock_irqrestore(&event->lock, flags);
+static void
+nouveau_event_fini(struct nouveau_eventh *handler)
+{
+	struct nouveau_event *event = handler->event;
+	unsigned long flags;
+	nouveau_event_put(handler);
+	spin_lock_irqsave(&event->list_lock, flags);
+	list_del(&handler->head);
+	spin_unlock_irqrestore(&event->list_lock, flags);
 }
 
-void
-nouveau_event_get(struct nouveau_event *event, int index,
-		  struct nouveau_eventh *handler)
+static int
+nouveau_event_init(struct nouveau_event *event, int index,
+		   int (*func)(void *, int), void *priv,
+		   struct nouveau_eventh *handler)
 {
 	unsigned long flags;
 
-	spin_lock_irqsave(&event->lock, flags);
-	if (index < event->index_nr) {
-		list_add(&handler->head, &event->index[index].list);
-		if (!event->index[index].refs++) {
-			if (event->enable)
-				event->enable(event, index);
-		}
+	if (index >= event->index_nr)
+		return -EINVAL;
+
+	handler->event = event;
+	handler->flags = 0;
+	handler->index = index;
+	handler->func = func;
+	handler->priv = priv;
+
+	spin_lock_irqsave(&event->list_lock, flags);
+	list_add_tail(&handler->head, &event->index[index].list);
+	spin_unlock_irqrestore(&event->list_lock, flags);
+	return 0;
+}
+
+int
+nouveau_event_new(struct nouveau_event *event, int index,
+		  int (*func)(void *, int), void *priv,
+		  struct nouveau_eventh **phandler)
+{
+	struct nouveau_eventh *handler;
+	int ret = -ENOMEM;
+
+	handler = *phandler = kmalloc(sizeof(*handler), GFP_KERNEL);
+	if (handler) {
+		ret = nouveau_event_init(event, index, func, priv, handler);
+		if (ret)
+			kfree(handler);
 	}
-	spin_unlock_irqrestore(&event->lock, flags);
+
+	return ret;
+}
+
+void
+nouveau_event_ref(struct nouveau_eventh *handler, struct nouveau_eventh **ref)
+{
+	BUG_ON(handler != NULL);
+	if (*ref) {
+		nouveau_event_fini(*ref);
+		kfree(*ref);
+	}
+	*ref = handler;
 }
 
 void
 nouveau_event_trigger(struct nouveau_event *event, int index)
 {
-	struct nouveau_eventh *handler, *temp;
+	struct nouveau_eventh *handler;
 	unsigned long flags;
 
-	if (index >= event->index_nr)
+	if (WARN_ON(index >= event->index_nr))
 		return;
 
-	spin_lock_irqsave(&event->lock, flags);
-	list_for_each_entry_safe(handler, temp, &event->index[index].list, head) {
-		if (handler->func(handler, index) == NVKM_EVENT_DROP) {
-			nouveau_event_put_locked(event, index, handler);
-		}
+	spin_lock_irqsave(&event->list_lock, flags);
+	list_for_each_entry(handler, &event->index[index].list, head) {
+		if (test_bit(NVKM_EVENT_ENABLE, &handler->flags) &&
+		    handler->func(handler->priv, index) == NVKM_EVENT_DROP)
+			nouveau_event_put(handler);
 	}
-	spin_unlock_irqrestore(&event->lock, flags);
+	spin_unlock_irqrestore(&event->list_lock, flags);
 }
 
 void
@@ -102,7 +154,8 @@ nouveau_event_create(int index_nr, struct nouveau_event **pevent)
 	if (!event)
 		return -ENOMEM;
 
-	spin_lock_init(&event->lock);
+	spin_lock_init(&event->list_lock);
+	spin_lock_init(&event->refs_lock);
 	for (i = 0; i < index_nr; i++)
 		INIT_LIST_HEAD(&event->index[i].list);
 	event->index_nr = index_nr;
diff --git a/drivers/gpu/drm/nouveau/core/core/option.c b/drivers/gpu/drm/nouveau/core/core/option.c
index 62a432ea39e5..9f6fcc5f66c2 100644
--- a/drivers/gpu/drm/nouveau/core/core/option.c
+++ b/drivers/gpu/drm/nouveau/core/core/option.c
@@ -25,15 +25,6 @@
 #include <core/option.h>
 #include <core/debug.h>
 
-/* compares unterminated string 'str' with zero-terminated string 'cmp' */
-static inline int
-strncasecmpz(const char *str, const char *cmp, size_t len)
-{
-	if (strlen(cmp) != len)
-		return len;
-	return strncasecmp(str, cmp, len);
-}
-
 const char *
 nouveau_stropt(const char *optstr, const char *opt, int *arglen)
 {
@@ -105,7 +96,7 @@ nouveau_dbgopt(const char *optstr, const char *sub)
 				else if (!strncasecmpz(optstr, "warn", len))
 					level = NV_DBG_WARN;
 				else if (!strncasecmpz(optstr, "info", len))
-					level = NV_DBG_INFO;
+					level = NV_DBG_INFO_NORMAL;
 				else if (!strncasecmpz(optstr, "debug", len))
 					level = NV_DBG_DEBUG;
 				else if (!strncasecmpz(optstr, "trace", len))
diff --git a/drivers/gpu/drm/nouveau/core/core/printk.c b/drivers/gpu/drm/nouveau/core/core/printk.c
index 52fb2aa129e8..03e0060b13da 100644
--- a/drivers/gpu/drm/nouveau/core/core/printk.c
+++ b/drivers/gpu/drm/nouveau/core/core/printk.c
@@ -27,16 +27,38 @@
 #include <core/subdev.h>
 #include <core/printk.h>
 
-int nv_printk_suspend_level = NV_DBG_DEBUG;
+int nv_info_debug_level = NV_DBG_INFO_NORMAL;
 
 void
-nv_printk_(struct nouveau_object *object, const char *pfx, int level,
-	   const char *fmt, ...)
+nv_printk_(struct nouveau_object *object, int level, const char *fmt, ...)
 {
 	static const char name[] = { '!', 'E', 'W', ' ', 'D', 'T', 'P', 'S' };
+	const char *pfx;
 	char mfmt[256];
 	va_list args;
 
+	switch (level) {
+	case NV_DBG_FATAL:
+		pfx = KERN_CRIT;
+		break;
+	case NV_DBG_ERROR:
+		pfx = KERN_ERR;
+		break;
+	case NV_DBG_WARN:
+		pfx = KERN_WARNING;
+		break;
+	case NV_DBG_INFO_NORMAL:
+		pfx = KERN_INFO;
+		break;
+	case NV_DBG_DEBUG:
+	case NV_DBG_PARANOIA:
+	case NV_DBG_TRACE:
+	case NV_DBG_SPAM:
+	default:
+		pfx = KERN_DEBUG;
+		break;
+	}
+
 	if (object && !nv_iclass(object, NV_CLIENT_CLASS)) {
 		struct nouveau_object *device = object;
 		struct nouveau_object *subdev = object;
@@ -74,20 +96,3 @@ nv_printk_(struct nouveau_object *object, const char *pfx, int level,
 	vprintk(mfmt, args);
 	va_end(args);
 }
-
-#define CONV_LEVEL(x) case NV_DBG_##x: return NV_PRINTK_##x
-
-const char *nv_printk_level_to_pfx(int level)
-{
-	switch (level) {
-	CONV_LEVEL(FATAL);
-	CONV_LEVEL(ERROR);
-	CONV_LEVEL(WARN);
-	CONV_LEVEL(INFO);
-	CONV_LEVEL(DEBUG);
-	CONV_LEVEL(PARANOIA);
-	CONV_LEVEL(TRACE);
-	CONV_LEVEL(SPAM);
-	}
-	return NV_PRINTK_DEBUG;
-}
diff --git a/drivers/gpu/drm/nouveau/core/engine/device/base.c b/drivers/gpu/drm/nouveau/core/engine/device/base.c
index 4c72571655ad..9135b25a29d0 100644
--- a/drivers/gpu/drm/nouveau/core/engine/device/base.c
+++ b/drivers/gpu/drm/nouveau/core/engine/device/base.c
@@ -29,7 +29,7 @@
 
 #include <core/class.h>
 
-#include <engine/device.h>
+#include "priv.h"
 
 static DEFINE_MUTEX(nv_devices_mutex);
 static LIST_HEAD(nv_devices);
@@ -75,7 +75,9 @@ static const u64 disable_map[] = {
 	[NVDEV_SUBDEV_BAR]	= NV_DEVICE_DISABLE_CORE,
 	[NVDEV_SUBDEV_VOLT]	= NV_DEVICE_DISABLE_CORE,
 	[NVDEV_SUBDEV_THERM]	= NV_DEVICE_DISABLE_CORE,
+	[NVDEV_SUBDEV_PWR]	= NV_DEVICE_DISABLE_CORE,
 	[NVDEV_ENGINE_DMAOBJ]	= NV_DEVICE_DISABLE_CORE,
+	[NVDEV_ENGINE_PERFMON]  = NV_DEVICE_DISABLE_CORE,
 	[NVDEV_ENGINE_FIFO]	= NV_DEVICE_DISABLE_FIFO,
 	[NVDEV_ENGINE_SW]	= NV_DEVICE_DISABLE_FIFO,
 	[NVDEV_ENGINE_GR]	= NV_DEVICE_DISABLE_GRAPH,
@@ -87,7 +89,7 @@ static const u64 disable_map[] = {
 	[NVDEV_ENGINE_PPP]	= NV_DEVICE_DISABLE_PPP,
 	[NVDEV_ENGINE_COPY0]	= NV_DEVICE_DISABLE_COPY0,
 	[NVDEV_ENGINE_COPY1]	= NV_DEVICE_DISABLE_COPY1,
-	[NVDEV_ENGINE_UNK1C1]	= NV_DEVICE_DISABLE_UNK1C1,
+	[NVDEV_ENGINE_VIC]	= NV_DEVICE_DISABLE_VIC,
 	[NVDEV_ENGINE_VENC]	= NV_DEVICE_DISABLE_VENC,
 	[NVDEV_ENGINE_DISP]	= NV_DEVICE_DISABLE_DISP,
 	[NVDEV_SUBDEV_NR]	= 0,
@@ -119,10 +121,12 @@ nouveau_devobj_ctor(struct nouveau_object *parent,
 			return -ENODEV;
 	}
 
-	ret = nouveau_parent_create(parent, nv_object(device), oclass, 0, NULL,
+	ret = nouveau_parent_create(parent, nv_object(device), oclass, 0,
+				    nouveau_control_oclass,
 				    (1ULL << NVDEV_ENGINE_DMAOBJ) |
 				    (1ULL << NVDEV_ENGINE_FIFO) |
-				    (1ULL << NVDEV_ENGINE_DISP), &devobj);
+				    (1ULL << NVDEV_ENGINE_DISP) |
+				    (1ULL << NVDEV_ENGINE_PERFMON), &devobj);
 	*pobject = nv_object(devobj);
 	if (ret)
 		return ret;
@@ -158,22 +162,29 @@ nouveau_devobj_ctor(struct nouveau_object *parent,
 		iounmap(map);
 
 		/* determine chipset and derive architecture from it */
-		if ((boot0 & 0x0f000000) > 0) {
-			device->chipset = (boot0 & 0xff00000) >> 20;
-			switch (device->chipset & 0xf0) {
-			case 0x10: device->card_type = NV_10; break;
-			case 0x20: device->card_type = NV_20; break;
-			case 0x30: device->card_type = NV_30; break;
-			case 0x40:
-			case 0x60: device->card_type = NV_40; break;
-			case 0x50:
-			case 0x80:
-			case 0x90:
-			case 0xa0: device->card_type = NV_50; break;
-			case 0xc0: device->card_type = NV_C0; break;
-			case 0xd0: device->card_type = NV_D0; break;
-			case 0xe0:
-			case 0xf0: device->card_type = NV_E0; break;
+		if ((boot0 & 0x1f000000) > 0) {
+			device->chipset = (boot0 & 0x1ff00000) >> 20;
+			switch (device->chipset & 0x1f0) {
+			case 0x010: {
+				if (0x461 & (1 << (device->chipset & 0xf)))
+					device->card_type = NV_10;
+				else
+					device->card_type = NV_11;
+				break;
+			}
+			case 0x020: device->card_type = NV_20; break;
+			case 0x030: device->card_type = NV_30; break;
+			case 0x040:
+			case 0x060: device->card_type = NV_40; break;
+			case 0x050:
+			case 0x080:
+			case 0x090:
+			case 0x0a0: device->card_type = NV_50; break;
+			case 0x0c0: device->card_type = NV_C0; break;
+			case 0x0d0: device->card_type = NV_D0; break;
+			case 0x0e0:
+			case 0x0f0:
+			case 0x100: device->card_type = NV_E0; break;
 			default:
 				break;
 			}
@@ -188,7 +199,8 @@ nouveau_devobj_ctor(struct nouveau_object *parent,
 
 		switch (device->card_type) {
 		case NV_04: ret = nv04_identify(device); break;
-		case NV_10: ret = nv10_identify(device); break;
+		case NV_10:
+		case NV_11: ret = nv10_identify(device); break;
 		case NV_20: ret = nv20_identify(device); break;
 		case NV_30: ret = nv30_identify(device); break;
 		case NV_40: ret = nv40_identify(device); break;
@@ -212,7 +224,7 @@ nouveau_devobj_ctor(struct nouveau_object *parent,
 		nv_info(device, "Family : NV%02X\n", device->card_type);
 
 		/* determine frequency of timing crystal */
-		if ( device->chipset < 0x17 ||
+		if ( device->card_type <= NV_10 || device->chipset < 0x17 ||
 		    (device->chipset >= 0x20 && device->chipset < 0x25))
 			strap &= 0x00000040;
 		else
diff --git a/drivers/gpu/drm/nouveau/core/engine/device/ctrl.c b/drivers/gpu/drm/nouveau/core/engine/device/ctrl.c
new file mode 100644
index 000000000000..4b69bf56ed01
--- /dev/null
+++ b/drivers/gpu/drm/nouveau/core/engine/device/ctrl.c
@@ -0,0 +1,144 @@
+/*
+ * Copyright 2013 Red Hat Inc.
+ *
+ * Permission is hereby granted, free of charge, to any person obtaining a
+ * copy of this software and associated documentation files (the "Software"),
+ * to deal in the Software without restriction, including without limitation
+ * the rights to use, copy, modify, merge, publish, distribute, sublicense,
+ * and/or sell copies of the Software, and to permit persons to whom the
+ * Software is furnished to do so, subject to the following conditions:
+ *
+ * The above copyright notice and this permission notice shall be included in
+ * all copies or substantial portions of the Software.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+ * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+ * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT.  IN NO EVENT SHALL
+ * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR
+ * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE,
+ * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR
+ * OTHER DEALINGS IN THE SOFTWARE.
+ *
+ * Authors: Ben Skeggs <bskeggs@redhat.com>
+ */
+
+#include <core/object.h>
+#include <core/class.h>
+
+#include <subdev/clock.h>
+
+#include "priv.h"
+
+static int
+nouveau_control_mthd_pstate_info(struct nouveau_object *object, u32 mthd,
+				void *data, u32 size)
+{
+	struct nouveau_clock *clk = nouveau_clock(object);
+	struct nv_control_pstate_info *args = data;
+
+	if (size < sizeof(*args))
+		return -EINVAL;
+
+	if (clk) {
+		args->count  = clk->state_nr;
+		args->ustate = clk->ustate;
+		args->pstate = clk->pstate;
+	} else {
+		args->count  = 0;
+		args->ustate = NV_CONTROL_PSTATE_INFO_USTATE_DISABLE;
+		args->pstate = NV_CONTROL_PSTATE_INFO_PSTATE_UNKNOWN;
+	}
+
+	return 0;
+}
+
+static int
+nouveau_control_mthd_pstate_attr(struct nouveau_object *object, u32 mthd,
+				void *data, u32 size)
+{
+	struct nouveau_clock *clk = nouveau_clock(object);
+	struct nv_control_pstate_attr *args = data;
+	struct nouveau_clocks *domain;
+	struct nouveau_pstate *pstate;
+	struct nouveau_cstate *cstate;
+	int i = 0, j = -1;
+	u32 lo, hi;
+
+	if ((size < sizeof(*args)) || !clk ||
+	    (args->state >= 0 && args->state >= clk->state_nr))
+		return -EINVAL;
+	domain = clk->domains;
+
+	while (domain->name != nv_clk_src_max) {
+		if (domain->mname && ++j == args->index)
+			break;
+		domain++;
+	}
+
+	if (domain->name == nv_clk_src_max)
+		return -EINVAL;
+
+	if (args->state != NV_CONTROL_PSTATE_ATTR_STATE_CURRENT) {
+		list_for_each_entry(pstate, &clk->states, head) {
+			if (i++ == args->state)
+				break;
+		}
+
+		lo = pstate->base.domain[domain->name];
+		hi = lo;
+		list_for_each_entry(cstate, &pstate->list, head) {
+			lo = min(lo, cstate->domain[domain->name]);
+			hi = max(hi, cstate->domain[domain->name]);
+		}
+
+		args->state = pstate->pstate;
+	} else {
+		lo = max(clk->read(clk, domain->name), 0);
+		hi = lo;
+	}
+
+	snprintf(args->name, sizeof(args->name), "%s", domain->mname);
+	snprintf(args->unit, sizeof(args->unit), "MHz");
+	args->min = lo / domain->mdiv;
+	args->max = hi / domain->mdiv;
+
+	args->index = 0;
+	while ((++domain)->name != nv_clk_src_max) {
+		if (domain->mname) {
+			args->index = ++j;
+			break;
+		}
+	}
+
+	return 0;
+}
+
+static int
+nouveau_control_mthd_pstate_user(struct nouveau_object *object, u32 mthd,
+				void *data, u32 size)
+{
+	struct nouveau_clock *clk = nouveau_clock(object);
+	struct nv_control_pstate_user *args = data;
+
+	if (size < sizeof(*args) || !clk)
+		return -EINVAL;
+
+	return nouveau_clock_ustate(clk, args->state);
+}
+
+struct nouveau_oclass
+nouveau_control_oclass[] = {
+	{ .handle = NV_CONTROL_CLASS,
+	  .ofuncs = &nouveau_object_ofuncs,
+	  .omthds = (struct nouveau_omthds[]) {
+		  { NV_CONTROL_PSTATE_INFO,
+		    NV_CONTROL_PSTATE_INFO, nouveau_control_mthd_pstate_info },
+		  { NV_CONTROL_PSTATE_ATTR,
+		    NV_CONTROL_PSTATE_ATTR, nouveau_control_mthd_pstate_attr },
+		  { NV_CONTROL_PSTATE_USER,
+		    NV_CONTROL_PSTATE_USER, nouveau_control_mthd_pstate_user },
+		  {},
+	  },
+	},
+	{}
+};
diff --git a/drivers/gpu/drm/nouveau/core/engine/device/nv04.c b/drivers/gpu/drm/nouveau/core/engine/device/nv04.c
index a0284cf09c0f..dbd2dde7b7e7 100644
--- a/drivers/gpu/drm/nouveau/core/engine/device/nv04.c
+++ b/drivers/gpu/drm/nouveau/core/engine/device/nv04.c
@@ -50,15 +50,15 @@ nv04_identify(struct nouveau_device *device)
 		device->oclass[NVDEV_SUBDEV_I2C    ] = &nv04_i2c_oclass;
 		device->oclass[NVDEV_SUBDEV_CLOCK  ] = &nv04_clock_oclass;
 		device->oclass[NVDEV_SUBDEV_DEVINIT] = &nv04_devinit_oclass;
-		device->oclass[NVDEV_SUBDEV_MC     ] = &nv04_mc_oclass;
-		device->oclass[NVDEV_SUBDEV_BUS    ] = &nv04_bus_oclass;
+		device->oclass[NVDEV_SUBDEV_MC     ] =  nv04_mc_oclass;
+		device->oclass[NVDEV_SUBDEV_BUS    ] =  nv04_bus_oclass;
 		device->oclass[NVDEV_SUBDEV_TIMER  ] = &nv04_timer_oclass;
-		device->oclass[NVDEV_SUBDEV_FB     ] = &nv04_fb_oclass;
+		device->oclass[NVDEV_SUBDEV_FB     ] =  nv04_fb_oclass;
 		device->oclass[NVDEV_SUBDEV_INSTMEM] = &nv04_instmem_oclass;
 		device->oclass[NVDEV_SUBDEV_VM     ] = &nv04_vmmgr_oclass;
 		device->oclass[NVDEV_ENGINE_DMAOBJ ] = &nv04_dmaeng_oclass;
-		device->oclass[NVDEV_ENGINE_FIFO   ] = &nv04_fifo_oclass;
-		device->oclass[NVDEV_ENGINE_SW     ] = &nv04_software_oclass;
+		device->oclass[NVDEV_ENGINE_FIFO   ] =  nv04_fifo_oclass;
+		device->oclass[NVDEV_ENGINE_SW     ] =  nv04_software_oclass;
 		device->oclass[NVDEV_ENGINE_GR     ] = &nv04_graph_oclass;
 		device->oclass[NVDEV_ENGINE_DISP   ] = &nv04_disp_oclass;
 		break;
@@ -68,15 +68,15 @@ nv04_identify(struct nouveau_device *device)
 		device->oclass[NVDEV_SUBDEV_I2C    ] = &nv04_i2c_oclass;
 		device->oclass[NVDEV_SUBDEV_CLOCK  ] = &nv04_clock_oclass;
 		device->oclass[NVDEV_SUBDEV_DEVINIT] = &nv05_devinit_oclass;
-		device->oclass[NVDEV_SUBDEV_MC     ] = &nv04_mc_oclass;
-		device->oclass[NVDEV_SUBDEV_BUS    ] = &nv04_bus_oclass;
+		device->oclass[NVDEV_SUBDEV_MC     ] =  nv04_mc_oclass;
+		device->oclass[NVDEV_SUBDEV_BUS    ] =  nv04_bus_oclass;
 		device->oclass[NVDEV_SUBDEV_TIMER  ] = &nv04_timer_oclass;
-		device->oclass[NVDEV_SUBDEV_FB     ] = &nv04_fb_oclass;
+		device->oclass[NVDEV_SUBDEV_FB     ] =  nv04_fb_oclass;
 		device->oclass[NVDEV_SUBDEV_INSTMEM] = &nv04_instmem_oclass;
 		device->oclass[NVDEV_SUBDEV_VM     ] = &nv04_vmmgr_oclass;
 		device->oclass[NVDEV_ENGINE_DMAOBJ ] = &nv04_dmaeng_oclass;
-		device->oclass[NVDEV_ENGINE_FIFO   ] = &nv04_fifo_oclass;
-		device->oclass[NVDEV_ENGINE_SW     ] = &nv04_software_oclass;
+		device->oclass[NVDEV_ENGINE_FIFO   ] =  nv04_fifo_oclass;
+		device->oclass[NVDEV_ENGINE_SW     ] =  nv04_software_oclass;
 		device->oclass[NVDEV_ENGINE_GR     ] = &nv04_graph_oclass;
 		device->oclass[NVDEV_ENGINE_DISP   ] = &nv04_disp_oclass;
 		break;
diff --git a/drivers/gpu/drm/nouveau/core/engine/device/nv10.c b/drivers/gpu/drm/nouveau/core/engine/device/nv10.c
index 1b7809a095c3..6e03dd6abeea 100644
--- a/drivers/gpu/drm/nouveau/core/engine/device/nv10.c
+++ b/drivers/gpu/drm/nouveau/core/engine/device/nv10.c
@@ -52,10 +52,10 @@ nv10_identify(struct nouveau_device *device)
 		device->oclass[NVDEV_SUBDEV_I2C    ] = &nv04_i2c_oclass;
 		device->oclass[NVDEV_SUBDEV_CLOCK  ] = &nv04_clock_oclass;
 		device->oclass[NVDEV_SUBDEV_DEVINIT] = &nv10_devinit_oclass;
-		device->oclass[NVDEV_SUBDEV_MC     ] = &nv04_mc_oclass;
-		device->oclass[NVDEV_SUBDEV_BUS    ] = &nv04_bus_oclass;
+		device->oclass[NVDEV_SUBDEV_MC     ] =  nv04_mc_oclass;
+		device->oclass[NVDEV_SUBDEV_BUS    ] =  nv04_bus_oclass;
 		device->oclass[NVDEV_SUBDEV_TIMER  ] = &nv04_timer_oclass;
-		device->oclass[NVDEV_SUBDEV_FB     ] = &nv10_fb_oclass;
+		device->oclass[NVDEV_SUBDEV_FB     ] =  nv10_fb_oclass;
 		device->oclass[NVDEV_SUBDEV_INSTMEM] = &nv04_instmem_oclass;
 		device->oclass[NVDEV_SUBDEV_VM     ] = &nv04_vmmgr_oclass;
 		device->oclass[NVDEV_ENGINE_DMAOBJ ] = &nv04_dmaeng_oclass;
@@ -69,15 +69,15 @@ nv10_identify(struct nouveau_device *device)
 		device->oclass[NVDEV_SUBDEV_I2C    ] = &nv04_i2c_oclass;
 		device->oclass[NVDEV_SUBDEV_CLOCK  ] = &nv04_clock_oclass;
 		device->oclass[NVDEV_SUBDEV_DEVINIT] = &nv10_devinit_oclass;
-		device->oclass[NVDEV_SUBDEV_MC     ] = &nv04_mc_oclass;
-		device->oclass[NVDEV_SUBDEV_BUS    ] = &nv04_bus_oclass;
+		device->oclass[NVDEV_SUBDEV_MC     ] =  nv04_mc_oclass;
+		device->oclass[NVDEV_SUBDEV_BUS    ] =  nv04_bus_oclass;
 		device->oclass[NVDEV_SUBDEV_TIMER  ] = &nv04_timer_oclass;
-		device->oclass[NVDEV_SUBDEV_FB     ] = &nv10_fb_oclass;
+		device->oclass[NVDEV_SUBDEV_FB     ] =  nv10_fb_oclass;
 		device->oclass[NVDEV_SUBDEV_INSTMEM] = &nv04_instmem_oclass;
 		device->oclass[NVDEV_SUBDEV_VM     ] = &nv04_vmmgr_oclass;
 		device->oclass[NVDEV_ENGINE_DMAOBJ ] = &nv04_dmaeng_oclass;
-		device->oclass[NVDEV_ENGINE_FIFO   ] = &nv10_fifo_oclass;
-		device->oclass[NVDEV_ENGINE_SW     ] = &nv10_software_oclass;
+		device->oclass[NVDEV_ENGINE_FIFO   ] =  nv10_fifo_oclass;
+		device->oclass[NVDEV_ENGINE_SW     ] =  nv10_software_oclass;
 		device->oclass[NVDEV_ENGINE_GR     ] = &nv10_graph_oclass;
 		device->oclass[NVDEV_ENGINE_DISP   ] = &nv04_disp_oclass;
 		break;
@@ -88,15 +88,15 @@ nv10_identify(struct nouveau_device *device)
 		device->oclass[NVDEV_SUBDEV_I2C    ] = &nv04_i2c_oclass;
 		device->oclass[NVDEV_SUBDEV_CLOCK  ] = &nv04_clock_oclass;
 		device->oclass[NVDEV_SUBDEV_DEVINIT] = &nv10_devinit_oclass;
-		device->oclass[NVDEV_SUBDEV_MC     ] = &nv04_mc_oclass;
-		device->oclass[NVDEV_SUBDEV_BUS    ] = &nv04_bus_oclass;
+		device->oclass[NVDEV_SUBDEV_MC     ] =  nv04_mc_oclass;
+		device->oclass[NVDEV_SUBDEV_BUS    ] =  nv04_bus_oclass;
 		device->oclass[NVDEV_SUBDEV_TIMER  ] = &nv04_timer_oclass;
-		device->oclass[NVDEV_SUBDEV_FB     ] = &nv10_fb_oclass;
+		device->oclass[NVDEV_SUBDEV_FB     ] =  nv10_fb_oclass;
 		device->oclass[NVDEV_SUBDEV_INSTMEM] = &nv04_instmem_oclass;
 		device->oclass[NVDEV_SUBDEV_VM     ] = &nv04_vmmgr_oclass;
 		device->oclass[NVDEV_ENGINE_DMAOBJ ] = &nv04_dmaeng_oclass;
-		device->oclass[NVDEV_ENGINE_FIFO   ] = &nv10_fifo_oclass;
-		device->oclass[NVDEV_ENGINE_SW     ] = &nv10_software_oclass;
+		device->oclass[NVDEV_ENGINE_FIFO   ] =  nv10_fifo_oclass;
+		device->oclass[NVDEV_ENGINE_SW     ] =  nv10_software_oclass;
 		device->oclass[NVDEV_ENGINE_GR     ] = &nv10_graph_oclass;
 		device->oclass[NVDEV_ENGINE_DISP   ] = &nv04_disp_oclass;
 		break;
@@ -107,15 +107,15 @@ nv10_identify(struct nouveau_device *device)
 		device->oclass[NVDEV_SUBDEV_I2C    ] = &nv04_i2c_oclass;
 		device->oclass[NVDEV_SUBDEV_CLOCK  ] = &nv04_clock_oclass;
 		device->oclass[NVDEV_SUBDEV_DEVINIT] = &nv1a_devinit_oclass;
-		device->oclass[NVDEV_SUBDEV_MC     ] = &nv04_mc_oclass;
-		device->oclass[NVDEV_SUBDEV_BUS    ] = &nv04_bus_oclass;
+		device->oclass[NVDEV_SUBDEV_MC     ] =  nv04_mc_oclass;
+		device->oclass[NVDEV_SUBDEV_BUS    ] =  nv04_bus_oclass;
 		device->oclass[NVDEV_SUBDEV_TIMER  ] = &nv04_timer_oclass;
-		device->oclass[NVDEV_SUBDEV_FB     ] = &nv1a_fb_oclass;
+		device->oclass[NVDEV_SUBDEV_FB     ] =  nv1a_fb_oclass;
 		device->oclass[NVDEV_SUBDEV_INSTMEM] = &nv04_instmem_oclass;
 		device->oclass[NVDEV_SUBDEV_VM     ] = &nv04_vmmgr_oclass;
 		device->oclass[NVDEV_ENGINE_DMAOBJ ] = &nv04_dmaeng_oclass;
-		device->oclass[NVDEV_ENGINE_FIFO   ] = &nv10_fifo_oclass;
-		device->oclass[NVDEV_ENGINE_SW     ] = &nv10_software_oclass;
+		device->oclass[NVDEV_ENGINE_FIFO   ] =  nv10_fifo_oclass;
+		device->oclass[NVDEV_ENGINE_SW     ] =  nv10_software_oclass;
 		device->oclass[NVDEV_ENGINE_GR     ] = &nv10_graph_oclass;
 		device->oclass[NVDEV_ENGINE_DISP   ] = &nv04_disp_oclass;
 		break;
@@ -126,15 +126,15 @@ nv10_identify(struct nouveau_device *device)
 		device->oclass[NVDEV_SUBDEV_I2C    ] = &nv04_i2c_oclass;
 		device->oclass[NVDEV_SUBDEV_CLOCK  ] = &nv04_clock_oclass;
 		device->oclass[NVDEV_SUBDEV_DEVINIT] = &nv10_devinit_oclass;
-		device->oclass[NVDEV_SUBDEV_MC     ] = &nv04_mc_oclass;
-		device->oclass[NVDEV_SUBDEV_BUS    ] = &nv04_bus_oclass;
+		device->oclass[NVDEV_SUBDEV_MC     ] =  nv04_mc_oclass;
+		device->oclass[NVDEV_SUBDEV_BUS    ] =  nv04_bus_oclass;
 		device->oclass[NVDEV_SUBDEV_TIMER  ] = &nv04_timer_oclass;
-		device->oclass[NVDEV_SUBDEV_FB     ] = &nv10_fb_oclass;
+		device->oclass[NVDEV_SUBDEV_FB     ] =  nv10_fb_oclass;
 		device->oclass[NVDEV_SUBDEV_INSTMEM] = &nv04_instmem_oclass;
 		device->oclass[NVDEV_SUBDEV_VM     ] = &nv04_vmmgr_oclass;
 		device->oclass[NVDEV_ENGINE_DMAOBJ ] = &nv04_dmaeng_oclass;
-		device->oclass[NVDEV_ENGINE_FIFO   ] = &nv10_fifo_oclass;
-		device->oclass[NVDEV_ENGINE_SW     ] = &nv10_software_oclass;
+		device->oclass[NVDEV_ENGINE_FIFO   ] =  nv10_fifo_oclass;
+		device->oclass[NVDEV_ENGINE_SW     ] =  nv10_software_oclass;
 		device->oclass[NVDEV_ENGINE_GR     ] = &nv10_graph_oclass;
 		device->oclass[NVDEV_ENGINE_DISP   ] = &nv04_disp_oclass;
 		break;
@@ -145,15 +145,15 @@ nv10_identify(struct nouveau_device *device)
 		device->oclass[NVDEV_SUBDEV_I2C    ] = &nv04_i2c_oclass;
 		device->oclass[NVDEV_SUBDEV_CLOCK  ] = &nv04_clock_oclass;
 		device->oclass[NVDEV_SUBDEV_DEVINIT] = &nv10_devinit_oclass;
-		device->oclass[NVDEV_SUBDEV_MC     ] = &nv04_mc_oclass;
-		device->oclass[NVDEV_SUBDEV_BUS    ] = &nv04_bus_oclass;
+		device->oclass[NVDEV_SUBDEV_MC     ] =  nv04_mc_oclass;
+		device->oclass[NVDEV_SUBDEV_BUS    ] =  nv04_bus_oclass;
 		device->oclass[NVDEV_SUBDEV_TIMER  ] = &nv04_timer_oclass;
-		device->oclass[NVDEV_SUBDEV_FB     ] = &nv10_fb_oclass;
+		device->oclass[NVDEV_SUBDEV_FB     ] =  nv10_fb_oclass;
 		device->oclass[NVDEV_SUBDEV_INSTMEM] = &nv04_instmem_oclass;
 		device->oclass[NVDEV_SUBDEV_VM     ] = &nv04_vmmgr_oclass;
 		device->oclass[NVDEV_ENGINE_DMAOBJ ] = &nv04_dmaeng_oclass;
-		device->oclass[NVDEV_ENGINE_FIFO   ] = &nv17_fifo_oclass;
-		device->oclass[NVDEV_ENGINE_SW     ] = &nv10_software_oclass;
+		device->oclass[NVDEV_ENGINE_FIFO   ] =  nv17_fifo_oclass;
+		device->oclass[NVDEV_ENGINE_SW     ] =  nv10_software_oclass;
 		device->oclass[NVDEV_ENGINE_GR     ] = &nv10_graph_oclass;
 		device->oclass[NVDEV_ENGINE_DISP   ] = &nv04_disp_oclass;
 		break;
@@ -164,15 +164,15 @@ nv10_identify(struct nouveau_device *device)
 		device->oclass[NVDEV_SUBDEV_I2C    ] = &nv04_i2c_oclass;
 		device->oclass[NVDEV_SUBDEV_CLOCK  ] = &nv04_clock_oclass;
 		device->oclass[NVDEV_SUBDEV_DEVINIT] = &nv1a_devinit_oclass;
-		device->oclass[NVDEV_SUBDEV_MC     ] = &nv04_mc_oclass;
-		device->oclass[NVDEV_SUBDEV_BUS    ] = &nv04_bus_oclass;
+		device->oclass[NVDEV_SUBDEV_MC     ] =  nv04_mc_oclass;
+		device->oclass[NVDEV_SUBDEV_BUS    ] =  nv04_bus_oclass;
 		device->oclass[NVDEV_SUBDEV_TIMER  ] = &nv04_timer_oclass;
-		device->oclass[NVDEV_SUBDEV_FB     ] = &nv1a_fb_oclass;
+		device->oclass[NVDEV_SUBDEV_FB     ] =  nv1a_fb_oclass;
 		device->oclass[NVDEV_SUBDEV_INSTMEM] = &nv04_instmem_oclass;
 		device->oclass[NVDEV_SUBDEV_VM     ] = &nv04_vmmgr_oclass;
 		device->oclass[NVDEV_ENGINE_DMAOBJ ] = &nv04_dmaeng_oclass;
-		device->oclass[NVDEV_ENGINE_FIFO   ] = &nv17_fifo_oclass;
-		device->oclass[NVDEV_ENGINE_SW     ] = &nv10_software_oclass;
+		device->oclass[NVDEV_ENGINE_FIFO   ] =  nv17_fifo_oclass;
+		device->oclass[NVDEV_ENGINE_SW     ] =  nv10_software_oclass;
 		device->oclass[NVDEV_ENGINE_GR     ] = &nv10_graph_oclass;
 		device->oclass[NVDEV_ENGINE_DISP   ] = &nv04_disp_oclass;
 		break;
@@ -183,15 +183,15 @@ nv10_identify(struct nouveau_device *device)
 		device->oclass[NVDEV_SUBDEV_I2C    ] = &nv04_i2c_oclass;
 		device->oclass[NVDEV_SUBDEV_CLOCK  ] = &nv04_clock_oclass;
 		device->oclass[NVDEV_SUBDEV_DEVINIT] = &nv10_devinit_oclass;
-		device->oclass[NVDEV_SUBDEV_MC     ] = &nv04_mc_oclass;
-		device->oclass[NVDEV_SUBDEV_BUS    ] = &nv04_bus_oclass;
+		device->oclass[NVDEV_SUBDEV_MC     ] =  nv04_mc_oclass;
+		device->oclass[NVDEV_SUBDEV_BUS    ] =  nv04_bus_oclass;
 		device->oclass[NVDEV_SUBDEV_TIMER  ] = &nv04_timer_oclass;
-		device->oclass[NVDEV_SUBDEV_FB     ] = &nv10_fb_oclass;
+		device->oclass[NVDEV_SUBDEV_FB     ] =  nv10_fb_oclass;
 		device->oclass[NVDEV_SUBDEV_INSTMEM] = &nv04_instmem_oclass;
 		device->oclass[NVDEV_SUBDEV_VM     ] = &nv04_vmmgr_oclass;
 		device->oclass[NVDEV_ENGINE_DMAOBJ ] = &nv04_dmaeng_oclass;
-		device->oclass[NVDEV_ENGINE_FIFO   ] = &nv17_fifo_oclass;
-		device->oclass[NVDEV_ENGINE_SW     ] = &nv10_software_oclass;
+		device->oclass[NVDEV_ENGINE_FIFO   ] =  nv17_fifo_oclass;
+		device->oclass[NVDEV_ENGINE_SW     ] =  nv10_software_oclass;
 		device->oclass[NVDEV_ENGINE_GR     ] = &nv10_graph_oclass;
 		device->oclass[NVDEV_ENGINE_DISP   ] = &nv04_disp_oclass;
 		break;
diff --git a/drivers/gpu/drm/nouveau/core/engine/device/nv20.c b/drivers/gpu/drm/nouveau/core/engine/device/nv20.c
index 12a4005fa619..dcde53b9f07f 100644
--- a/drivers/gpu/drm/nouveau/core/engine/device/nv20.c
+++ b/drivers/gpu/drm/nouveau/core/engine/device/nv20.c
@@ -53,15 +53,15 @@ nv20_identify(struct nouveau_device *device)
 		device->oclass[NVDEV_SUBDEV_I2C    ] = &nv04_i2c_oclass;
 		device->oclass[NVDEV_SUBDEV_CLOCK  ] = &nv04_clock_oclass;
 		device->oclass[NVDEV_SUBDEV_DEVINIT] = &nv20_devinit_oclass;
-		device->oclass[NVDEV_SUBDEV_MC     ] = &nv04_mc_oclass;
-		device->oclass[NVDEV_SUBDEV_BUS    ] = &nv04_bus_oclass;
+		device->oclass[NVDEV_SUBDEV_MC     ] =  nv04_mc_oclass;
+		device->oclass[NVDEV_SUBDEV_BUS    ] =  nv04_bus_oclass;
 		device->oclass[NVDEV_SUBDEV_TIMER  ] = &nv04_timer_oclass;
-		device->oclass[NVDEV_SUBDEV_FB     ] = &nv20_fb_oclass;
+		device->oclass[NVDEV_SUBDEV_FB     ] =  nv20_fb_oclass;
 		device->oclass[NVDEV_SUBDEV_INSTMEM] = &nv04_instmem_oclass;
 		device->oclass[NVDEV_SUBDEV_VM     ] = &nv04_vmmgr_oclass;
 		device->oclass[NVDEV_ENGINE_DMAOBJ ] = &nv04_dmaeng_oclass;
-		device->oclass[NVDEV_ENGINE_FIFO   ] = &nv17_fifo_oclass;
-		device->oclass[NVDEV_ENGINE_SW     ] = &nv10_software_oclass;
+		device->oclass[NVDEV_ENGINE_FIFO   ] =  nv17_fifo_oclass;
+		device->oclass[NVDEV_ENGINE_SW     ] =  nv10_software_oclass;
 		device->oclass[NVDEV_ENGINE_GR     ] = &nv20_graph_oclass;
 		device->oclass[NVDEV_ENGINE_DISP   ] = &nv04_disp_oclass;
 		break;
@@ -72,15 +72,15 @@ nv20_identify(struct nouveau_device *device)
 		device->oclass[NVDEV_SUBDEV_I2C    ] = &nv04_i2c_oclass;
 		device->oclass[NVDEV_SUBDEV_CLOCK  ] = &nv04_clock_oclass;
 		device->oclass[NVDEV_SUBDEV_DEVINIT] = &nv20_devinit_oclass;
-		device->oclass[NVDEV_SUBDEV_MC     ] = &nv04_mc_oclass;
-		device->oclass[NVDEV_SUBDEV_BUS    ] = &nv04_bus_oclass;
+		device->oclass[NVDEV_SUBDEV_MC     ] =  nv04_mc_oclass;
+		device->oclass[NVDEV_SUBDEV_BUS    ] =  nv04_bus_oclass;
 		device->oclass[NVDEV_SUBDEV_TIMER  ] = &nv04_timer_oclass;
-		device->oclass[NVDEV_SUBDEV_FB     ] = &nv25_fb_oclass;
+		device->oclass[NVDEV_SUBDEV_FB     ] =  nv25_fb_oclass;
 		device->oclass[NVDEV_SUBDEV_INSTMEM] = &nv04_instmem_oclass;
 		device->oclass[NVDEV_SUBDEV_VM     ] = &nv04_vmmgr_oclass;
 		device->oclass[NVDEV_ENGINE_DMAOBJ ] = &nv04_dmaeng_oclass;
-		device->oclass[NVDEV_ENGINE_FIFO   ] = &nv17_fifo_oclass;
-		device->oclass[NVDEV_ENGINE_SW     ] = &nv10_software_oclass;
+		device->oclass[NVDEV_ENGINE_FIFO   ] =  nv17_fifo_oclass;
+		device->oclass[NVDEV_ENGINE_SW     ] =  nv10_software_oclass;
 		device->oclass[NVDEV_ENGINE_GR     ] = &nv25_graph_oclass;
 		device->oclass[NVDEV_ENGINE_DISP   ] = &nv04_disp_oclass;
 		break;
@@ -91,15 +91,15 @@ nv20_identify(struct nouveau_device *device)
 		device->oclass[NVDEV_SUBDEV_I2C    ] = &nv04_i2c_oclass;
 		device->oclass[NVDEV_SUBDEV_CLOCK  ] = &nv04_clock_oclass;
 		device->oclass[NVDEV_SUBDEV_DEVINIT] = &nv20_devinit_oclass;
-		device->oclass[NVDEV_SUBDEV_MC     ] = &nv04_mc_oclass;
-		device->oclass[NVDEV_SUBDEV_BUS    ] = &nv04_bus_oclass;
+		device->oclass[NVDEV_SUBDEV_MC     ] =  nv04_mc_oclass;
+		device->oclass[NVDEV_SUBDEV_BUS    ] =  nv04_bus_oclass;
 		device->oclass[NVDEV_SUBDEV_TIMER  ] = &nv04_timer_oclass;
-		device->oclass[NVDEV_SUBDEV_FB     ] = &nv25_fb_oclass;
+		device->oclass[NVDEV_SUBDEV_FB     ] =  nv25_fb_oclass;
 		device->oclass[NVDEV_SUBDEV_INSTMEM] = &nv04_instmem_oclass;
 		device->oclass[NVDEV_SUBDEV_VM     ] = &nv04_vmmgr_oclass;
 		device->oclass[NVDEV_ENGINE_DMAOBJ ] = &nv04_dmaeng_oclass;
-		device->oclass[NVDEV_ENGINE_FIFO   ] = &nv17_fifo_oclass;
-		device->oclass[NVDEV_ENGINE_SW     ] = &nv10_software_oclass;
+		device->oclass[NVDEV_ENGINE_FIFO   ] =  nv17_fifo_oclass;
+		device->oclass[NVDEV_ENGINE_SW     ] =  nv10_software_oclass;
 		device->oclass[NVDEV_ENGINE_GR     ] = &nv25_graph_oclass;
 		device->oclass[NVDEV_ENGINE_DISP   ] = &nv04_disp_oclass;
 		break;
@@ -110,15 +110,15 @@ nv20_identify(struct nouveau_device *device)
 		device->oclass[NVDEV_SUBDEV_I2C    ] = &nv04_i2c_oclass;
 		device->oclass[NVDEV_SUBDEV_CLOCK  ] = &nv04_clock_oclass;
 		device->oclass[NVDEV_SUBDEV_DEVINIT] = &nv20_devinit_oclass;
-		device->oclass[NVDEV_SUBDEV_MC     ] = &nv04_mc_oclass;
-		device->oclass[NVDEV_SUBDEV_BUS    ] = &nv04_bus_oclass;
+		device->oclass[NVDEV_SUBDEV_MC     ] =  nv04_mc_oclass;
+		device->oclass[NVDEV_SUBDEV_BUS    ] =  nv04_bus_oclass;
 		device->oclass[NVDEV_SUBDEV_TIMER  ] = &nv04_timer_oclass;
-		device->oclass[NVDEV_SUBDEV_FB     ] = &nv25_fb_oclass;
+		device->oclass[NVDEV_SUBDEV_FB     ] =  nv25_fb_oclass;
 		device->oclass[NVDEV_SUBDEV_INSTMEM] = &nv04_instmem_oclass;
 		device->oclass[NVDEV_SUBDEV_VM     ] = &nv04_vmmgr_oclass;
 		device->oclass[NVDEV_ENGINE_DMAOBJ ] = &nv04_dmaeng_oclass;
-		device->oclass[NVDEV_ENGINE_FIFO   ] = &nv17_fifo_oclass;
-		device->oclass[NVDEV_ENGINE_SW     ] = &nv10_software_oclass;
+		device->oclass[NVDEV_ENGINE_FIFO   ] =  nv17_fifo_oclass;
+		device->oclass[NVDEV_ENGINE_SW     ] =  nv10_software_oclass;
 		device->oclass[NVDEV_ENGINE_GR     ] = &nv2a_graph_oclass;
 		device->oclass[NVDEV_ENGINE_DISP   ] = &nv04_disp_oclass;
 		break;
diff --git a/drivers/gpu/drm/nouveau/core/engine/device/nv30.c b/drivers/gpu/drm/nouveau/core/engine/device/nv30.c
index cef0f1ea4c21..7b8662ef4f59 100644
--- a/drivers/gpu/drm/nouveau/core/engine/device/nv30.c
+++ b/drivers/gpu/drm/nouveau/core/engine/device/nv30.c
@@ -53,15 +53,15 @@ nv30_identify(struct nouveau_device *device)
 		device->oclass[NVDEV_SUBDEV_I2C    ] = &nv04_i2c_oclass;
 		device->oclass[NVDEV_SUBDEV_CLOCK  ] = &nv04_clock_oclass;
 		device->oclass[NVDEV_SUBDEV_DEVINIT] = &nv20_devinit_oclass;
-		device->oclass[NVDEV_SUBDEV_MC     ] = &nv04_mc_oclass;
-		device->oclass[NVDEV_SUBDEV_BUS    ] = &nv04_bus_oclass;
+		device->oclass[NVDEV_SUBDEV_MC     ] =  nv04_mc_oclass;
+		device->oclass[NVDEV_SUBDEV_BUS    ] =  nv04_bus_oclass;
 		device->oclass[NVDEV_SUBDEV_TIMER  ] = &nv04_timer_oclass;
-		device->oclass[NVDEV_SUBDEV_FB     ] = &nv30_fb_oclass;
+		device->oclass[NVDEV_SUBDEV_FB     ] =  nv30_fb_oclass;
 		device->oclass[NVDEV_SUBDEV_INSTMEM] = &nv04_instmem_oclass;
 		device->oclass[NVDEV_SUBDEV_VM     ] = &nv04_vmmgr_oclass;
 		device->oclass[NVDEV_ENGINE_DMAOBJ ] = &nv04_dmaeng_oclass;
-		device->oclass[NVDEV_ENGINE_FIFO   ] = &nv17_fifo_oclass;
-		device->oclass[NVDEV_ENGINE_SW     ] = &nv10_software_oclass;
+		device->oclass[NVDEV_ENGINE_FIFO   ] =  nv17_fifo_oclass;
+		device->oclass[NVDEV_ENGINE_SW     ] =  nv10_software_oclass;
 		device->oclass[NVDEV_ENGINE_GR     ] = &nv30_graph_oclass;
 		device->oclass[NVDEV_ENGINE_DISP   ] = &nv04_disp_oclass;
 		break;
@@ -72,15 +72,15 @@ nv30_identify(struct nouveau_device *device)
 		device->oclass[NVDEV_SUBDEV_I2C    ] = &nv04_i2c_oclass;
 		device->oclass[NVDEV_SUBDEV_CLOCK  ] = &nv04_clock_oclass;
 		device->oclass[NVDEV_SUBDEV_DEVINIT] = &nv20_devinit_oclass;
-		device->oclass[NVDEV_SUBDEV_MC     ] = &nv04_mc_oclass;
-		device->oclass[NVDEV_SUBDEV_BUS    ] = &nv04_bus_oclass;
+		device->oclass[NVDEV_SUBDEV_MC     ] =  nv04_mc_oclass;
+		device->oclass[NVDEV_SUBDEV_BUS    ] =  nv04_bus_oclass;
 		device->oclass[NVDEV_SUBDEV_TIMER  ] = &nv04_timer_oclass;
-		device->oclass[NVDEV_SUBDEV_FB     ] = &nv35_fb_oclass;
+		device->oclass[NVDEV_SUBDEV_FB     ] =  nv35_fb_oclass;
 		device->oclass[NVDEV_SUBDEV_INSTMEM] = &nv04_instmem_oclass;
 		device->oclass[NVDEV_SUBDEV_VM     ] = &nv04_vmmgr_oclass;
 		device->oclass[NVDEV_ENGINE_DMAOBJ ] = &nv04_dmaeng_oclass;
-		device->oclass[NVDEV_ENGINE_FIFO   ] = &nv17_fifo_oclass;
-		device->oclass[NVDEV_ENGINE_SW     ] = &nv10_software_oclass;
+		device->oclass[NVDEV_ENGINE_FIFO   ] =  nv17_fifo_oclass;
+		device->oclass[NVDEV_ENGINE_SW     ] =  nv10_software_oclass;
 		device->oclass[NVDEV_ENGINE_GR     ] = &nv35_graph_oclass;
 		device->oclass[NVDEV_ENGINE_DISP   ] = &nv04_disp_oclass;
 		break;
@@ -91,15 +91,15 @@ nv30_identify(struct nouveau_device *device)
 		device->oclass[NVDEV_SUBDEV_I2C    ] = &nv04_i2c_oclass;
 		device->oclass[NVDEV_SUBDEV_CLOCK  ] = &nv04_clock_oclass;
 		device->oclass[NVDEV_SUBDEV_DEVINIT] = &nv20_devinit_oclass;
-		device->oclass[NVDEV_SUBDEV_MC     ] = &nv04_mc_oclass;
-		device->oclass[NVDEV_SUBDEV_BUS    ] = &nv31_bus_oclass;
+		device->oclass[NVDEV_SUBDEV_MC     ] =  nv04_mc_oclass;
+		device->oclass[NVDEV_SUBDEV_BUS    ] =  nv31_bus_oclass;
 		device->oclass[NVDEV_SUBDEV_TIMER  ] = &nv04_timer_oclass;
-		device->oclass[NVDEV_SUBDEV_FB     ] = &nv30_fb_oclass;
+		device->oclass[NVDEV_SUBDEV_FB     ] =  nv30_fb_oclass;
 		device->oclass[NVDEV_SUBDEV_INSTMEM] = &nv04_instmem_oclass;
 		device->oclass[NVDEV_SUBDEV_VM     ] = &nv04_vmmgr_oclass;
 		device->oclass[NVDEV_ENGINE_DMAOBJ ] = &nv04_dmaeng_oclass;
-		device->oclass[NVDEV_ENGINE_FIFO   ] = &nv17_fifo_oclass;
-		device->oclass[NVDEV_ENGINE_SW     ] = &nv10_software_oclass;
+		device->oclass[NVDEV_ENGINE_FIFO   ] =  nv17_fifo_oclass;
+		device->oclass[NVDEV_ENGINE_SW     ] =  nv10_software_oclass;
 		device->oclass[NVDEV_ENGINE_GR     ] = &nv30_graph_oclass;
 		device->oclass[NVDEV_ENGINE_MPEG   ] = &nv31_mpeg_oclass;
 		device->oclass[NVDEV_ENGINE_DISP   ] = &nv04_disp_oclass;
@@ -111,15 +111,15 @@ nv30_identify(struct nouveau_device *device)
 		device->oclass[NVDEV_SUBDEV_I2C    ] = &nv04_i2c_oclass;
 		device->oclass[NVDEV_SUBDEV_CLOCK  ] = &nv04_clock_oclass;
 		device->oclass[NVDEV_SUBDEV_DEVINIT] = &nv20_devinit_oclass;
-		device->oclass[NVDEV_SUBDEV_MC     ] = &nv04_mc_oclass;
-		device->oclass[NVDEV_SUBDEV_BUS    ] = &nv31_bus_oclass;
+		device->oclass[NVDEV_SUBDEV_MC     ] =  nv04_mc_oclass;
+		device->oclass[NVDEV_SUBDEV_BUS    ] =  nv31_bus_oclass;
 		device->oclass[NVDEV_SUBDEV_TIMER  ] = &nv04_timer_oclass;
-		device->oclass[NVDEV_SUBDEV_FB     ] = &nv36_fb_oclass;
+		device->oclass[NVDEV_SUBDEV_FB     ] =  nv36_fb_oclass;
 		device->oclass[NVDEV_SUBDEV_INSTMEM] = &nv04_instmem_oclass;
 		device->oclass[NVDEV_SUBDEV_VM     ] = &nv04_vmmgr_oclass;
 		device->oclass[NVDEV_ENGINE_DMAOBJ ] = &nv04_dmaeng_oclass;
-		device->oclass[NVDEV_ENGINE_FIFO   ] = &nv17_fifo_oclass;
-		device->oclass[NVDEV_ENGINE_SW     ] = &nv10_software_oclass;
+		device->oclass[NVDEV_ENGINE_FIFO   ] =  nv17_fifo_oclass;
+		device->oclass[NVDEV_ENGINE_SW     ] =  nv10_software_oclass;
 		device->oclass[NVDEV_ENGINE_GR     ] = &nv35_graph_oclass;
 		device->oclass[NVDEV_ENGINE_MPEG   ] = &nv31_mpeg_oclass;
 		device->oclass[NVDEV_ENGINE_DISP   ] = &nv04_disp_oclass;
@@ -131,15 +131,15 @@ nv30_identify(struct nouveau_device *device)
 		device->oclass[NVDEV_SUBDEV_I2C    ] = &nv04_i2c_oclass;
 		device->oclass[NVDEV_SUBDEV_CLOCK  ] = &nv04_clock_oclass;
 		device->oclass[NVDEV_SUBDEV_DEVINIT] = &nv10_devinit_oclass;
-		device->oclass[NVDEV_SUBDEV_MC     ] = &nv04_mc_oclass;
-		device->oclass[NVDEV_SUBDEV_BUS    ] = &nv31_bus_oclass;
+		device->oclass[NVDEV_SUBDEV_MC     ] =  nv04_mc_oclass;
+		device->oclass[NVDEV_SUBDEV_BUS    ] =  nv31_bus_oclass;
 		device->oclass[NVDEV_SUBDEV_TIMER  ] = &nv04_timer_oclass;
-		device->oclass[NVDEV_SUBDEV_FB     ] = &nv10_fb_oclass;
+		device->oclass[NVDEV_SUBDEV_FB     ] =  nv10_fb_oclass;
 		device->oclass[NVDEV_SUBDEV_INSTMEM] = &nv04_instmem_oclass;
 		device->oclass[NVDEV_SUBDEV_VM     ] = &nv04_vmmgr_oclass;
 		device->oclass[NVDEV_ENGINE_DMAOBJ ] = &nv04_dmaeng_oclass;
-		device->oclass[NVDEV_ENGINE_FIFO   ] = &nv17_fifo_oclass;
-		device->oclass[NVDEV_ENGINE_SW     ] = &nv10_software_oclass;
+		device->oclass[NVDEV_ENGINE_FIFO   ] =  nv17_fifo_oclass;
+		device->oclass[NVDEV_ENGINE_SW     ] =  nv10_software_oclass;
 		device->oclass[NVDEV_ENGINE_GR     ] = &nv34_graph_oclass;
 		device->oclass[NVDEV_ENGINE_MPEG   ] = &nv31_mpeg_oclass;
 		device->oclass[NVDEV_ENGINE_DISP   ] = &nv04_disp_oclass;
diff --git a/drivers/gpu/drm/nouveau/core/engine/device/nv40.c b/drivers/gpu/drm/nouveau/core/engine/device/nv40.c
index 1719cb0ee595..c8c41e93695e 100644
--- a/drivers/gpu/drm/nouveau/core/engine/device/nv40.c
+++ b/drivers/gpu/drm/nouveau/core/engine/device/nv40.c
@@ -35,6 +35,7 @@
 #include <subdev/fb.h>
 #include <subdev/instmem.h>
 #include <subdev/vm.h>
+#include <subdev/volt.h>
 
 #include <engine/device.h>
 #include <engine/dmaobj.h>
@@ -43,6 +44,7 @@
 #include <engine/graph.h>
 #include <engine/mpeg.h>
 #include <engine/disp.h>
+#include <engine/perfmon.h>
 
 int
 nv40_identify(struct nouveau_device *device)
@@ -56,18 +58,20 @@ nv40_identify(struct nouveau_device *device)
 		device->oclass[NVDEV_SUBDEV_CLOCK  ] = &nv40_clock_oclass;
 		device->oclass[NVDEV_SUBDEV_THERM  ] = &nv40_therm_oclass;
 		device->oclass[NVDEV_SUBDEV_DEVINIT] = &nv1a_devinit_oclass;
-		device->oclass[NVDEV_SUBDEV_MC     ] = &nv04_mc_oclass;
-		device->oclass[NVDEV_SUBDEV_BUS    ] = &nv31_bus_oclass;
+		device->oclass[NVDEV_SUBDEV_MC     ] =  nv40_mc_oclass;
+		device->oclass[NVDEV_SUBDEV_BUS    ] =  nv31_bus_oclass;
 		device->oclass[NVDEV_SUBDEV_TIMER  ] = &nv04_timer_oclass;
-		device->oclass[NVDEV_SUBDEV_FB     ] = &nv40_fb_oclass;
+		device->oclass[NVDEV_SUBDEV_FB     ] =  nv40_fb_oclass;
 		device->oclass[NVDEV_SUBDEV_INSTMEM] = &nv40_instmem_oclass;
 		device->oclass[NVDEV_SUBDEV_VM     ] = &nv04_vmmgr_oclass;
+		device->oclass[NVDEV_SUBDEV_VOLT   ] = &nv40_volt_oclass;
 		device->oclass[NVDEV_ENGINE_DMAOBJ ] = &nv04_dmaeng_oclass;
-		device->oclass[NVDEV_ENGINE_FIFO   ] = &nv40_fifo_oclass;
-		device->oclass[NVDEV_ENGINE_SW     ] = &nv10_software_oclass;
+		device->oclass[NVDEV_ENGINE_FIFO   ] =  nv40_fifo_oclass;
+		device->oclass[NVDEV_ENGINE_SW     ] =  nv10_software_oclass;
 		device->oclass[NVDEV_ENGINE_GR     ] = &nv40_graph_oclass;
 		device->oclass[NVDEV_ENGINE_MPEG   ] = &nv40_mpeg_oclass;
 		device->oclass[NVDEV_ENGINE_DISP   ] = &nv04_disp_oclass;
+		device->oclass[NVDEV_ENGINE_PERFMON] =  nv40_perfmon_oclass;
 		break;
 	case 0x41:
 		device->cname = "NV41";
@@ -77,18 +81,20 @@ nv40_identify(struct nouveau_device *device)
 		device->oclass[NVDEV_SUBDEV_CLOCK  ] = &nv40_clock_oclass;
 		device->oclass[NVDEV_SUBDEV_THERM  ] = &nv40_therm_oclass;
 		device->oclass[NVDEV_SUBDEV_DEVINIT] = &nv1a_devinit_oclass;
-		device->oclass[NVDEV_SUBDEV_MC     ] = &nv04_mc_oclass;
-		device->oclass[NVDEV_SUBDEV_BUS    ] = &nv31_bus_oclass;
+		device->oclass[NVDEV_SUBDEV_MC     ] =  nv40_mc_oclass;
+		device->oclass[NVDEV_SUBDEV_BUS    ] =  nv31_bus_oclass;
 		device->oclass[NVDEV_SUBDEV_TIMER  ] = &nv04_timer_oclass;
-		device->oclass[NVDEV_SUBDEV_FB     ] = &nv41_fb_oclass;
+		device->oclass[NVDEV_SUBDEV_FB     ] =  nv41_fb_oclass;
 		device->oclass[NVDEV_SUBDEV_INSTMEM] = &nv40_instmem_oclass;
 		device->oclass[NVDEV_SUBDEV_VM     ] = &nv41_vmmgr_oclass;
+		device->oclass[NVDEV_SUBDEV_VOLT   ] = &nv40_volt_oclass;
 		device->oclass[NVDEV_ENGINE_DMAOBJ ] = &nv04_dmaeng_oclass;
-		device->oclass[NVDEV_ENGINE_FIFO   ] = &nv40_fifo_oclass;
-		device->oclass[NVDEV_ENGINE_SW     ] = &nv10_software_oclass;
+		device->oclass[NVDEV_ENGINE_FIFO   ] =  nv40_fifo_oclass;
+		device->oclass[NVDEV_ENGINE_SW     ] =  nv10_software_oclass;
 		device->oclass[NVDEV_ENGINE_GR     ] = &nv40_graph_oclass;
 		device->oclass[NVDEV_ENGINE_MPEG   ] = &nv40_mpeg_oclass;
 		device->oclass[NVDEV_ENGINE_DISP   ] = &nv04_disp_oclass;
+		device->oclass[NVDEV_ENGINE_PERFMON] =  nv40_perfmon_oclass;
 		break;
 	case 0x42:
 		device->cname = "NV42";
@@ -98,18 +104,20 @@ nv40_identify(struct nouveau_device *device)
 		device->oclass[NVDEV_SUBDEV_CLOCK  ] = &nv40_clock_oclass;
 		device->oclass[NVDEV_SUBDEV_THERM  ] = &nv40_therm_oclass;
 		device->oclass[NVDEV_SUBDEV_DEVINIT] = &nv1a_devinit_oclass;
-		device->oclass[NVDEV_SUBDEV_MC     ] = &nv04_mc_oclass;
-		device->oclass[NVDEV_SUBDEV_BUS    ] = &nv31_bus_oclass;
+		device->oclass[NVDEV_SUBDEV_MC     ] =  nv40_mc_oclass;
+		device->oclass[NVDEV_SUBDEV_BUS    ] =  nv31_bus_oclass;
 		device->oclass[NVDEV_SUBDEV_TIMER  ] = &nv04_timer_oclass;
-		device->oclass[NVDEV_SUBDEV_FB     ] = &nv41_fb_oclass;
+		device->oclass[NVDEV_SUBDEV_FB     ] =  nv41_fb_oclass;
 		device->oclass[NVDEV_SUBDEV_INSTMEM] = &nv40_instmem_oclass;
 		device->oclass[NVDEV_SUBDEV_VM     ] = &nv41_vmmgr_oclass;
+		device->oclass[NVDEV_SUBDEV_VOLT   ] = &nv40_volt_oclass;
 		device->oclass[NVDEV_ENGINE_DMAOBJ ] = &nv04_dmaeng_oclass;
-		device->oclass[NVDEV_ENGINE_FIFO   ] = &nv40_fifo_oclass;
-		device->oclass[NVDEV_ENGINE_SW     ] = &nv10_software_oclass;
+		device->oclass[NVDEV_ENGINE_FIFO   ] =  nv40_fifo_oclass;
+		device->oclass[NVDEV_ENGINE_SW     ] =  nv10_software_oclass;
 		device->oclass[NVDEV_ENGINE_GR     ] = &nv40_graph_oclass;
 		device->oclass[NVDEV_ENGINE_MPEG   ] = &nv40_mpeg_oclass;
 		device->oclass[NVDEV_ENGINE_DISP   ] = &nv04_disp_oclass;
+		device->oclass[NVDEV_ENGINE_PERFMON] =  nv40_perfmon_oclass;
 		break;
 	case 0x43:
 		device->cname = "NV43";
@@ -119,18 +127,20 @@ nv40_identify(struct nouveau_device *device)
 		device->oclass[NVDEV_SUBDEV_CLOCK  ] = &nv40_clock_oclass;
 		device->oclass[NVDEV_SUBDEV_THERM  ] = &nv40_therm_oclass;
 		device->oclass[NVDEV_SUBDEV_DEVINIT] = &nv1a_devinit_oclass;
-		device->oclass[NVDEV_SUBDEV_MC     ] = &nv04_mc_oclass;
-		device->oclass[NVDEV_SUBDEV_BUS    ] = &nv31_bus_oclass;
+		device->oclass[NVDEV_SUBDEV_MC     ] =  nv40_mc_oclass;
+		device->oclass[NVDEV_SUBDEV_BUS    ] =  nv31_bus_oclass;
 		device->oclass[NVDEV_SUBDEV_TIMER  ] = &nv04_timer_oclass;
-		device->oclass[NVDEV_SUBDEV_FB     ] = &nv41_fb_oclass;
+		device->oclass[NVDEV_SUBDEV_FB     ] =  nv41_fb_oclass;
 		device->oclass[NVDEV_SUBDEV_INSTMEM] = &nv40_instmem_oclass;
 		device->oclass[NVDEV_SUBDEV_VM     ] = &nv41_vmmgr_oclass;
+		device->oclass[NVDEV_SUBDEV_VOLT   ] = &nv40_volt_oclass;
 		device->oclass[NVDEV_ENGINE_DMAOBJ ] = &nv04_dmaeng_oclass;
-		device->oclass[NVDEV_ENGINE_FIFO   ] = &nv40_fifo_oclass;
-		device->oclass[NVDEV_ENGINE_SW     ] = &nv10_software_oclass;
+		device->oclass[NVDEV_ENGINE_FIFO   ] =  nv40_fifo_oclass;
+		device->oclass[NVDEV_ENGINE_SW     ] =  nv10_software_oclass;
 		device->oclass[NVDEV_ENGINE_GR     ] = &nv40_graph_oclass;
 		device->oclass[NVDEV_ENGINE_MPEG   ] = &nv40_mpeg_oclass;
 		device->oclass[NVDEV_ENGINE_DISP   ] = &nv04_disp_oclass;
+		device->oclass[NVDEV_ENGINE_PERFMON] =  nv40_perfmon_oclass;
 		break;
 	case 0x45:
 		device->cname = "NV45";
@@ -140,18 +150,20 @@ nv40_identify(struct nouveau_device *device)
 		device->oclass[NVDEV_SUBDEV_CLOCK  ] = &nv40_clock_oclass;
 		device->oclass[NVDEV_SUBDEV_THERM  ] = &nv40_therm_oclass;
 		device->oclass[NVDEV_SUBDEV_DEVINIT] = &nv1a_devinit_oclass;
-		device->oclass[NVDEV_SUBDEV_MC     ] = &nv04_mc_oclass;
-		device->oclass[NVDEV_SUBDEV_BUS    ] = &nv31_bus_oclass;
+		device->oclass[NVDEV_SUBDEV_MC     ] =  nv40_mc_oclass;
+		device->oclass[NVDEV_SUBDEV_BUS    ] =  nv31_bus_oclass;
 		device->oclass[NVDEV_SUBDEV_TIMER  ] = &nv04_timer_oclass;
-		device->oclass[NVDEV_SUBDEV_FB     ] = &nv40_fb_oclass;
+		device->oclass[NVDEV_SUBDEV_FB     ] =  nv40_fb_oclass;
 		device->oclass[NVDEV_SUBDEV_INSTMEM] = &nv40_instmem_oclass;
 		device->oclass[NVDEV_SUBDEV_VM     ] = &nv04_vmmgr_oclass;
+		device->oclass[NVDEV_SUBDEV_VOLT   ] = &nv40_volt_oclass;
 		device->oclass[NVDEV_ENGINE_DMAOBJ ] = &nv04_dmaeng_oclass;
-		device->oclass[NVDEV_ENGINE_FIFO   ] = &nv40_fifo_oclass;
-		device->oclass[NVDEV_ENGINE_SW     ] = &nv10_software_oclass;
+		device->oclass[NVDEV_ENGINE_FIFO   ] =  nv40_fifo_oclass;
+		device->oclass[NVDEV_ENGINE_SW     ] =  nv10_software_oclass;
 		device->oclass[NVDEV_ENGINE_GR     ] = &nv40_graph_oclass;
-		device->oclass[NVDEV_ENGINE_MPEG   ] = &nv40_mpeg_oclass;
+		device->oclass[NVDEV_ENGINE_MPEG   ] = &nv44_mpeg_oclass;
 		device->oclass[NVDEV_ENGINE_DISP   ] = &nv04_disp_oclass;
+		device->oclass[NVDEV_ENGINE_PERFMON] =  nv40_perfmon_oclass;
 		break;
 	case 0x47:
 		device->cname = "G70";
@@ -161,18 +173,20 @@ nv40_identify(struct nouveau_device *device)
 		device->oclass[NVDEV_SUBDEV_CLOCK  ] = &nv40_clock_oclass;
 		device->oclass[NVDEV_SUBDEV_THERM  ] = &nv40_therm_oclass;
 		device->oclass[NVDEV_SUBDEV_DEVINIT] = &nv1a_devinit_oclass;
-		device->oclass[NVDEV_SUBDEV_MC     ] = &nv04_mc_oclass;
-		device->oclass[NVDEV_SUBDEV_BUS    ] = &nv31_bus_oclass;
+		device->oclass[NVDEV_SUBDEV_MC     ] =  nv40_mc_oclass;
+		device->oclass[NVDEV_SUBDEV_BUS    ] =  nv31_bus_oclass;
 		device->oclass[NVDEV_SUBDEV_TIMER  ] = &nv04_timer_oclass;
-		device->oclass[NVDEV_SUBDEV_FB     ] = &nv47_fb_oclass;
+		device->oclass[NVDEV_SUBDEV_FB     ] =  nv47_fb_oclass;
 		device->oclass[NVDEV_SUBDEV_INSTMEM] = &nv40_instmem_oclass;
 		device->oclass[NVDEV_SUBDEV_VM     ] = &nv41_vmmgr_oclass;
+		device->oclass[NVDEV_SUBDEV_VOLT   ] = &nv40_volt_oclass;
 		device->oclass[NVDEV_ENGINE_DMAOBJ ] = &nv04_dmaeng_oclass;
-		device->oclass[NVDEV_ENGINE_FIFO   ] = &nv40_fifo_oclass;
-		device->oclass[NVDEV_ENGINE_SW     ] = &nv10_software_oclass;
+		device->oclass[NVDEV_ENGINE_FIFO   ] =  nv40_fifo_oclass;
+		device->oclass[NVDEV_ENGINE_SW     ] =  nv10_software_oclass;
 		device->oclass[NVDEV_ENGINE_GR     ] = &nv40_graph_oclass;
-		device->oclass[NVDEV_ENGINE_MPEG   ] = &nv40_mpeg_oclass;
+		device->oclass[NVDEV_ENGINE_MPEG   ] = &nv44_mpeg_oclass;
 		device->oclass[NVDEV_ENGINE_DISP   ] = &nv04_disp_oclass;
+		device->oclass[NVDEV_ENGINE_PERFMON] =  nv40_perfmon_oclass;
 		break;
 	case 0x49:
 		device->cname = "G71";
@@ -182,18 +196,20 @@ nv40_identify(struct nouveau_device *device)
 		device->oclass[NVDEV_SUBDEV_CLOCK  ] = &nv40_clock_oclass;
 		device->oclass[NVDEV_SUBDEV_THERM  ] = &nv40_therm_oclass;
 		device->oclass[NVDEV_SUBDEV_DEVINIT] = &nv1a_devinit_oclass;
-		device->oclass[NVDEV_SUBDEV_MC     ] = &nv04_mc_oclass;
-		device->oclass[NVDEV_SUBDEV_BUS    ] = &nv31_bus_oclass;
+		device->oclass[NVDEV_SUBDEV_MC     ] =  nv40_mc_oclass;
+		device->oclass[NVDEV_SUBDEV_BUS    ] =  nv31_bus_oclass;
 		device->oclass[NVDEV_SUBDEV_TIMER  ] = &nv04_timer_oclass;
-		device->oclass[NVDEV_SUBDEV_FB     ] = &nv49_fb_oclass;
+		device->oclass[NVDEV_SUBDEV_FB     ] =  nv49_fb_oclass;
 		device->oclass[NVDEV_SUBDEV_INSTMEM] = &nv40_instmem_oclass;
 		device->oclass[NVDEV_SUBDEV_VM     ] = &nv41_vmmgr_oclass;
+		device->oclass[NVDEV_SUBDEV_VOLT   ] = &nv40_volt_oclass;
 		device->oclass[NVDEV_ENGINE_DMAOBJ ] = &nv04_dmaeng_oclass;
-		device->oclass[NVDEV_ENGINE_FIFO   ] = &nv40_fifo_oclass;
-		device->oclass[NVDEV_ENGINE_SW     ] = &nv10_software_oclass;
+		device->oclass[NVDEV_ENGINE_FIFO   ] =  nv40_fifo_oclass;
+		device->oclass[NVDEV_ENGINE_SW     ] =  nv10_software_oclass;
 		device->oclass[NVDEV_ENGINE_GR     ] = &nv40_graph_oclass;
-		device->oclass[NVDEV_ENGINE_MPEG   ] = &nv40_mpeg_oclass;
+		device->oclass[NVDEV_ENGINE_MPEG   ] = &nv44_mpeg_oclass;
 		device->oclass[NVDEV_ENGINE_DISP   ] = &nv04_disp_oclass;
+		device->oclass[NVDEV_ENGINE_PERFMON] =  nv40_perfmon_oclass;
 		break;
 	case 0x4b:
 		device->cname = "G73";
@@ -203,18 +219,20 @@ nv40_identify(struct nouveau_device *device)
 		device->oclass[NVDEV_SUBDEV_CLOCK  ] = &nv40_clock_oclass;
 		device->oclass[NVDEV_SUBDEV_THERM  ] = &nv40_therm_oclass;
 		device->oclass[NVDEV_SUBDEV_DEVINIT] = &nv1a_devinit_oclass;
-		device->oclass[NVDEV_SUBDEV_MC     ] = &nv04_mc_oclass;
-		device->oclass[NVDEV_SUBDEV_BUS    ] = &nv31_bus_oclass;
+		device->oclass[NVDEV_SUBDEV_MC     ] =  nv40_mc_oclass;
+		device->oclass[NVDEV_SUBDEV_BUS    ] =  nv31_bus_oclass;
 		device->oclass[NVDEV_SUBDEV_TIMER  ] = &nv04_timer_oclass;
-		device->oclass[NVDEV_SUBDEV_FB     ] = &nv49_fb_oclass;
+		device->oclass[NVDEV_SUBDEV_FB     ] =  nv49_fb_oclass;
 		device->oclass[NVDEV_SUBDEV_INSTMEM] = &nv40_instmem_oclass;
 		device->oclass[NVDEV_SUBDEV_VM     ] = &nv41_vmmgr_oclass;
+		device->oclass[NVDEV_SUBDEV_VOLT   ] = &nv40_volt_oclass;
 		device->oclass[NVDEV_ENGINE_DMAOBJ ] = &nv04_dmaeng_oclass;
-		device->oclass[NVDEV_ENGINE_FIFO   ] = &nv40_fifo_oclass;
-		device->oclass[NVDEV_ENGINE_SW     ] = &nv10_software_oclass;
+		device->oclass[NVDEV_ENGINE_FIFO   ] =  nv40_fifo_oclass;
+		device->oclass[NVDEV_ENGINE_SW     ] =  nv10_software_oclass;
 		device->oclass[NVDEV_ENGINE_GR     ] = &nv40_graph_oclass;
-		device->oclass[NVDEV_ENGINE_MPEG   ] = &nv40_mpeg_oclass;
+		device->oclass[NVDEV_ENGINE_MPEG   ] = &nv44_mpeg_oclass;
 		device->oclass[NVDEV_ENGINE_DISP   ] = &nv04_disp_oclass;
+		device->oclass[NVDEV_ENGINE_PERFMON] =  nv40_perfmon_oclass;
 		break;
 	case 0x44:
 		device->cname = "NV44";
@@ -224,18 +242,20 @@ nv40_identify(struct nouveau_device *device)
 		device->oclass[NVDEV_SUBDEV_CLOCK  ] = &nv40_clock_oclass;
 		device->oclass[NVDEV_SUBDEV_THERM  ] = &nv40_therm_oclass;
 		device->oclass[NVDEV_SUBDEV_DEVINIT] = &nv1a_devinit_oclass;
-		device->oclass[NVDEV_SUBDEV_MC     ] = &nv44_mc_oclass;
-		device->oclass[NVDEV_SUBDEV_BUS    ] = &nv31_bus_oclass;
+		device->oclass[NVDEV_SUBDEV_MC     ] =  nv44_mc_oclass;
+		device->oclass[NVDEV_SUBDEV_BUS    ] =  nv31_bus_oclass;
 		device->oclass[NVDEV_SUBDEV_TIMER  ] = &nv04_timer_oclass;
-		device->oclass[NVDEV_SUBDEV_FB     ] = &nv44_fb_oclass;
+		device->oclass[NVDEV_SUBDEV_FB     ] =  nv44_fb_oclass;
 		device->oclass[NVDEV_SUBDEV_INSTMEM] = &nv40_instmem_oclass;
 		device->oclass[NVDEV_SUBDEV_VM     ] = &nv44_vmmgr_oclass;
+		device->oclass[NVDEV_SUBDEV_VOLT   ] = &nv40_volt_oclass;
 		device->oclass[NVDEV_ENGINE_DMAOBJ ] = &nv04_dmaeng_oclass;
-		device->oclass[NVDEV_ENGINE_FIFO   ] = &nv40_fifo_oclass;
-		device->oclass[NVDEV_ENGINE_SW     ] = &nv10_software_oclass;
+		device->oclass[NVDEV_ENGINE_FIFO   ] =  nv40_fifo_oclass;
+		device->oclass[NVDEV_ENGINE_SW     ] =  nv10_software_oclass;
 		device->oclass[NVDEV_ENGINE_GR     ] = &nv40_graph_oclass;
-		device->oclass[NVDEV_ENGINE_MPEG   ] = &nv40_mpeg_oclass;
+		device->oclass[NVDEV_ENGINE_MPEG   ] = &nv44_mpeg_oclass;
 		device->oclass[NVDEV_ENGINE_DISP   ] = &nv04_disp_oclass;
+		device->oclass[NVDEV_ENGINE_PERFMON] =  nv40_perfmon_oclass;
 		break;
 	case 0x46:
 		device->cname = "G72";
@@ -245,18 +265,20 @@ nv40_identify(struct nouveau_device *device)
 		device->oclass[NVDEV_SUBDEV_CLOCK  ] = &nv40_clock_oclass;
 		device->oclass[NVDEV_SUBDEV_THERM  ] = &nv40_therm_oclass;
 		device->oclass[NVDEV_SUBDEV_DEVINIT] = &nv1a_devinit_oclass;
-		device->oclass[NVDEV_SUBDEV_MC     ] = &nv44_mc_oclass;
-		device->oclass[NVDEV_SUBDEV_BUS    ] = &nv31_bus_oclass;
+		device->oclass[NVDEV_SUBDEV_MC     ] =  nv44_mc_oclass;
+		device->oclass[NVDEV_SUBDEV_BUS    ] =  nv31_bus_oclass;
 		device->oclass[NVDEV_SUBDEV_TIMER  ] = &nv04_timer_oclass;
-		device->oclass[NVDEV_SUBDEV_FB     ] = &nv46_fb_oclass;
+		device->oclass[NVDEV_SUBDEV_FB     ] =  nv46_fb_oclass;
 		device->oclass[NVDEV_SUBDEV_INSTMEM] = &nv40_instmem_oclass;
 		device->oclass[NVDEV_SUBDEV_VM     ] = &nv44_vmmgr_oclass;
+		device->oclass[NVDEV_SUBDEV_VOLT   ] = &nv40_volt_oclass;
 		device->oclass[NVDEV_ENGINE_DMAOBJ ] = &nv04_dmaeng_oclass;
-		device->oclass[NVDEV_ENGINE_FIFO   ] = &nv40_fifo_oclass;
-		device->oclass[NVDEV_ENGINE_SW     ] = &nv10_software_oclass;
+		device->oclass[NVDEV_ENGINE_FIFO   ] =  nv40_fifo_oclass;
+		device->oclass[NVDEV_ENGINE_SW     ] =  nv10_software_oclass;
 		device->oclass[NVDEV_ENGINE_GR     ] = &nv40_graph_oclass;
-		device->oclass[NVDEV_ENGINE_MPEG   ] = &nv40_mpeg_oclass;
+		device->oclass[NVDEV_ENGINE_MPEG   ] = &nv44_mpeg_oclass;
 		device->oclass[NVDEV_ENGINE_DISP   ] = &nv04_disp_oclass;
+		device->oclass[NVDEV_ENGINE_PERFMON] =  nv40_perfmon_oclass;
 		break;
 	case 0x4a:
 		device->cname = "NV44A";
@@ -266,18 +288,20 @@ nv40_identify(struct nouveau_device *device)
 		device->oclass[NVDEV_SUBDEV_CLOCK  ] = &nv40_clock_oclass;
 		device->oclass[NVDEV_SUBDEV_THERM  ] = &nv40_therm_oclass;
 		device->oclass[NVDEV_SUBDEV_DEVINIT] = &nv1a_devinit_oclass;
-		device->oclass[NVDEV_SUBDEV_MC     ] = &nv44_mc_oclass;
-		device->oclass[NVDEV_SUBDEV_BUS    ] = &nv31_bus_oclass;
+		device->oclass[NVDEV_SUBDEV_MC     ] =  nv44_mc_oclass;
+		device->oclass[NVDEV_SUBDEV_BUS    ] =  nv31_bus_oclass;
 		device->oclass[NVDEV_SUBDEV_TIMER  ] = &nv04_timer_oclass;
-		device->oclass[NVDEV_SUBDEV_FB     ] = &nv44_fb_oclass;
+		device->oclass[NVDEV_SUBDEV_FB     ] =  nv44_fb_oclass;
 		device->oclass[NVDEV_SUBDEV_INSTMEM] = &nv40_instmem_oclass;
 		device->oclass[NVDEV_SUBDEV_VM     ] = &nv44_vmmgr_oclass;
+		device->oclass[NVDEV_SUBDEV_VOLT   ] = &nv40_volt_oclass;
 		device->oclass[NVDEV_ENGINE_DMAOBJ ] = &nv04_dmaeng_oclass;
-		device->oclass[NVDEV_ENGINE_FIFO   ] = &nv40_fifo_oclass;
-		device->oclass[NVDEV_ENGINE_SW     ] = &nv10_software_oclass;
+		device->oclass[NVDEV_ENGINE_FIFO   ] =  nv40_fifo_oclass;
+		device->oclass[NVDEV_ENGINE_SW     ] =  nv10_software_oclass;
 		device->oclass[NVDEV_ENGINE_GR     ] = &nv40_graph_oclass;
-		device->oclass[NVDEV_ENGINE_MPEG   ] = &nv40_mpeg_oclass;
+		device->oclass[NVDEV_ENGINE_MPEG   ] = &nv44_mpeg_oclass;
 		device->oclass[NVDEV_ENGINE_DISP   ] = &nv04_disp_oclass;
+		device->oclass[NVDEV_ENGINE_PERFMON] =  nv40_perfmon_oclass;
 		break;
 	case 0x4c:
 		device->cname = "C61";
@@ -287,18 +311,20 @@ nv40_identify(struct nouveau_device *device)
 		device->oclass[NVDEV_SUBDEV_CLOCK  ] = &nv40_clock_oclass;
 		device->oclass[NVDEV_SUBDEV_THERM  ] = &nv40_therm_oclass;
 		device->oclass[NVDEV_SUBDEV_DEVINIT] = &nv1a_devinit_oclass;
-		device->oclass[NVDEV_SUBDEV_MC     ] = &nv44_mc_oclass;
-		device->oclass[NVDEV_SUBDEV_BUS    ] = &nv31_bus_oclass;
+		device->oclass[NVDEV_SUBDEV_MC     ] =  nv44_mc_oclass;
+		device->oclass[NVDEV_SUBDEV_BUS    ] =  nv31_bus_oclass;
 		device->oclass[NVDEV_SUBDEV_TIMER  ] = &nv04_timer_oclass;
-		device->oclass[NVDEV_SUBDEV_FB     ] = &nv46_fb_oclass;
+		device->oclass[NVDEV_SUBDEV_FB     ] =  nv46_fb_oclass;
 		device->oclass[NVDEV_SUBDEV_INSTMEM] = &nv40_instmem_oclass;
 		device->oclass[NVDEV_SUBDEV_VM     ] = &nv44_vmmgr_oclass;
+		device->oclass[NVDEV_SUBDEV_VOLT   ] = &nv40_volt_oclass;
 		device->oclass[NVDEV_ENGINE_DMAOBJ ] = &nv04_dmaeng_oclass;
-		device->oclass[NVDEV_ENGINE_FIFO   ] = &nv40_fifo_oclass;
-		device->oclass[NVDEV_ENGINE_SW     ] = &nv10_software_oclass;
+		device->oclass[NVDEV_ENGINE_FIFO   ] =  nv40_fifo_oclass;
+		device->oclass[NVDEV_ENGINE_SW     ] =  nv10_software_oclass;
 		device->oclass[NVDEV_ENGINE_GR     ] = &nv40_graph_oclass;
-		device->oclass[NVDEV_ENGINE_MPEG   ] = &nv40_mpeg_oclass;
+		device->oclass[NVDEV_ENGINE_MPEG   ] = &nv44_mpeg_oclass;
 		device->oclass[NVDEV_ENGINE_DISP   ] = &nv04_disp_oclass;
+		device->oclass[NVDEV_ENGINE_PERFMON] =  nv40_perfmon_oclass;
 		break;
 	case 0x4e:
 		device->cname = "C51";
@@ -308,18 +334,20 @@ nv40_identify(struct nouveau_device *device)
 		device->oclass[NVDEV_SUBDEV_CLOCK  ] = &nv40_clock_oclass;
 		device->oclass[NVDEV_SUBDEV_THERM  ] = &nv40_therm_oclass;
 		device->oclass[NVDEV_SUBDEV_DEVINIT] = &nv1a_devinit_oclass;
-		device->oclass[NVDEV_SUBDEV_MC     ] = &nv44_mc_oclass;
-		device->oclass[NVDEV_SUBDEV_BUS    ] = &nv31_bus_oclass;
+		device->oclass[NVDEV_SUBDEV_MC     ] =  nv44_mc_oclass;
+		device->oclass[NVDEV_SUBDEV_BUS    ] =  nv31_bus_oclass;
 		device->oclass[NVDEV_SUBDEV_TIMER  ] = &nv04_timer_oclass;
-		device->oclass[NVDEV_SUBDEV_FB     ] = &nv4e_fb_oclass;
+		device->oclass[NVDEV_SUBDEV_FB     ] =  nv4e_fb_oclass;
 		device->oclass[NVDEV_SUBDEV_INSTMEM] = &nv40_instmem_oclass;
 		device->oclass[NVDEV_SUBDEV_VM     ] = &nv44_vmmgr_oclass;
+		device->oclass[NVDEV_SUBDEV_VOLT   ] = &nv40_volt_oclass;
 		device->oclass[NVDEV_ENGINE_DMAOBJ ] = &nv04_dmaeng_oclass;
-		device->oclass[NVDEV_ENGINE_FIFO   ] = &nv40_fifo_oclass;
-		device->oclass[NVDEV_ENGINE_SW     ] = &nv10_software_oclass;
+		device->oclass[NVDEV_ENGINE_FIFO   ] =  nv40_fifo_oclass;
+		device->oclass[NVDEV_ENGINE_SW     ] =  nv10_software_oclass;
 		device->oclass[NVDEV_ENGINE_GR     ] = &nv40_graph_oclass;
-		device->oclass[NVDEV_ENGINE_MPEG   ] = &nv40_mpeg_oclass;
+		device->oclass[NVDEV_ENGINE_MPEG   ] = &nv44_mpeg_oclass;
 		device->oclass[NVDEV_ENGINE_DISP   ] = &nv04_disp_oclass;
+		device->oclass[NVDEV_ENGINE_PERFMON] =  nv40_perfmon_oclass;
 		break;
 	case 0x63:
 		device->cname = "C73";
@@ -329,18 +357,20 @@ nv40_identify(struct nouveau_device *device)
 		device->oclass[NVDEV_SUBDEV_CLOCK  ] = &nv40_clock_oclass;
 		device->oclass[NVDEV_SUBDEV_THERM  ] = &nv40_therm_oclass;
 		device->oclass[NVDEV_SUBDEV_DEVINIT] = &nv1a_devinit_oclass;
-		device->oclass[NVDEV_SUBDEV_MC     ] = &nv44_mc_oclass;
-		device->oclass[NVDEV_SUBDEV_BUS    ] = &nv31_bus_oclass;
+		device->oclass[NVDEV_SUBDEV_MC     ] =  nv44_mc_oclass;
+		device->oclass[NVDEV_SUBDEV_BUS    ] =  nv31_bus_oclass;
 		device->oclass[NVDEV_SUBDEV_TIMER  ] = &nv04_timer_oclass;
-		device->oclass[NVDEV_SUBDEV_FB     ] = &nv46_fb_oclass;
+		device->oclass[NVDEV_SUBDEV_FB     ] =  nv46_fb_oclass;
 		device->oclass[NVDEV_SUBDEV_INSTMEM] = &nv40_instmem_oclass;
 		device->oclass[NVDEV_SUBDEV_VM     ] = &nv44_vmmgr_oclass;
+		device->oclass[NVDEV_SUBDEV_VOLT   ] = &nv40_volt_oclass;
 		device->oclass[NVDEV_ENGINE_DMAOBJ ] = &nv04_dmaeng_oclass;
-		device->oclass[NVDEV_ENGINE_FIFO   ] = &nv40_fifo_oclass;
-		device->oclass[NVDEV_ENGINE_SW     ] = &nv10_software_oclass;
+		device->oclass[NVDEV_ENGINE_FIFO   ] =  nv40_fifo_oclass;
+		device->oclass[NVDEV_ENGINE_SW     ] =  nv10_software_oclass;
 		device->oclass[NVDEV_ENGINE_GR     ] = &nv40_graph_oclass;
-		device->oclass[NVDEV_ENGINE_MPEG   ] = &nv40_mpeg_oclass;
+		device->oclass[NVDEV_ENGINE_MPEG   ] = &nv44_mpeg_oclass;
 		device->oclass[NVDEV_ENGINE_DISP   ] = &nv04_disp_oclass;
+		device->oclass[NVDEV_ENGINE_PERFMON] =  nv40_perfmon_oclass;
 		break;
 	case 0x67:
 		device->cname = "C67";
@@ -350,18 +380,20 @@ nv40_identify(struct nouveau_device *device)
 		device->oclass[NVDEV_SUBDEV_CLOCK  ] = &nv40_clock_oclass;
 		device->oclass[NVDEV_SUBDEV_THERM  ] = &nv40_therm_oclass;
 		device->oclass[NVDEV_SUBDEV_DEVINIT] = &nv1a_devinit_oclass;
-		device->oclass[NVDEV_SUBDEV_MC     ] = &nv44_mc_oclass;
-		device->oclass[NVDEV_SUBDEV_BUS    ] = &nv31_bus_oclass;
+		device->oclass[NVDEV_SUBDEV_MC     ] =  nv44_mc_oclass;
+		device->oclass[NVDEV_SUBDEV_BUS    ] =  nv31_bus_oclass;
 		device->oclass[NVDEV_SUBDEV_TIMER  ] = &nv04_timer_oclass;
-		device->oclass[NVDEV_SUBDEV_FB     ] = &nv46_fb_oclass;
+		device->oclass[NVDEV_SUBDEV_FB     ] =  nv46_fb_oclass;
 		device->oclass[NVDEV_SUBDEV_INSTMEM] = &nv40_instmem_oclass;
 		device->oclass[NVDEV_SUBDEV_VM     ] = &nv44_vmmgr_oclass;
+		device->oclass[NVDEV_SUBDEV_VOLT   ] = &nv40_volt_oclass;
 		device->oclass[NVDEV_ENGINE_DMAOBJ ] = &nv04_dmaeng_oclass;
-		device->oclass[NVDEV_ENGINE_FIFO   ] = &nv40_fifo_oclass;
-		device->oclass[NVDEV_ENGINE_SW     ] = &nv10_software_oclass;
+		device->oclass[NVDEV_ENGINE_FIFO   ] =  nv40_fifo_oclass;
+		device->oclass[NVDEV_ENGINE_SW     ] =  nv10_software_oclass;
 		device->oclass[NVDEV_ENGINE_GR     ] = &nv40_graph_oclass;
-		device->oclass[NVDEV_ENGINE_MPEG   ] = &nv40_mpeg_oclass;
+		device->oclass[NVDEV_ENGINE_MPEG   ] = &nv44_mpeg_oclass;
 		device->oclass[NVDEV_ENGINE_DISP   ] = &nv04_disp_oclass;
+		device->oclass[NVDEV_ENGINE_PERFMON] =  nv40_perfmon_oclass;
 		break;
 	case 0x68:
 		device->cname = "C68";
@@ -371,18 +403,20 @@ nv40_identify(struct nouveau_device *device)
 		device->oclass[NVDEV_SUBDEV_CLOCK  ] = &nv40_clock_oclass;
 		device->oclass[NVDEV_SUBDEV_THERM  ] = &nv40_therm_oclass;
 		device->oclass[NVDEV_SUBDEV_DEVINIT] = &nv1a_devinit_oclass;
-		device->oclass[NVDEV_SUBDEV_MC     ] = &nv44_mc_oclass;
-		device->oclass[NVDEV_SUBDEV_BUS    ] = &nv31_bus_oclass;
+		device->oclass[NVDEV_SUBDEV_MC     ] =  nv44_mc_oclass;
+		device->oclass[NVDEV_SUBDEV_BUS    ] =  nv31_bus_oclass;
 		device->oclass[NVDEV_SUBDEV_TIMER  ] = &nv04_timer_oclass;
-		device->oclass[NVDEV_SUBDEV_FB     ] = &nv46_fb_oclass;
+		device->oclass[NVDEV_SUBDEV_FB     ] =  nv46_fb_oclass;
 		device->oclass[NVDEV_SUBDEV_INSTMEM] = &nv40_instmem_oclass;
 		device->oclass[NVDEV_SUBDEV_VM     ] = &nv44_vmmgr_oclass;
+		device->oclass[NVDEV_SUBDEV_VOLT   ] = &nv40_volt_oclass;
 		device->oclass[NVDEV_ENGINE_DMAOBJ ] = &nv04_dmaeng_oclass;
-		device->oclass[NVDEV_ENGINE_FIFO   ] = &nv40_fifo_oclass;
-		device->oclass[NVDEV_ENGINE_SW     ] = &nv10_software_oclass;
+		device->oclass[NVDEV_ENGINE_FIFO   ] =  nv40_fifo_oclass;
+		device->oclass[NVDEV_ENGINE_SW     ] =  nv10_software_oclass;
 		device->oclass[NVDEV_ENGINE_GR     ] = &nv40_graph_oclass;
-		device->oclass[NVDEV_ENGINE_MPEG   ] = &nv40_mpeg_oclass;
+		device->oclass[NVDEV_ENGINE_MPEG   ] = &nv44_mpeg_oclass;
 		device->oclass[NVDEV_ENGINE_DISP   ] = &nv04_disp_oclass;
+		device->oclass[NVDEV_ENGINE_PERFMON] =  nv40_perfmon_oclass;
 		break;
 	default:
 		nv_fatal(device, "unknown Curie chipset\n");
diff --git a/drivers/gpu/drm/nouveau/core/engine/device/nv50.c b/drivers/gpu/drm/nouveau/core/engine/device/nv50.c
index ffc18b80c5d9..db139827047c 100644
--- a/drivers/gpu/drm/nouveau/core/engine/device/nv50.c
+++ b/drivers/gpu/drm/nouveau/core/engine/device/nv50.c
@@ -36,6 +36,8 @@
 #include <subdev/instmem.h>
 #include <subdev/vm.h>
 #include <subdev/bar.h>
+#include <subdev/pwr.h>
+#include <subdev/volt.h>
 
 #include <engine/device.h>
 #include <engine/dmaobj.h>
@@ -49,6 +51,7 @@
 #include <engine/ppp.h>
 #include <engine/copy.h>
 #include <engine/disp.h>
+#include <engine/perfmon.h>
 
 int
 nv50_identify(struct nouveau_device *device)
@@ -59,257 +62,277 @@ nv50_identify(struct nouveau_device *device)
 		device->oclass[NVDEV_SUBDEV_VBIOS  ] = &nouveau_bios_oclass;
 		device->oclass[NVDEV_SUBDEV_GPIO   ] = &nv50_gpio_oclass;
 		device->oclass[NVDEV_SUBDEV_I2C    ] = &nv50_i2c_oclass;
-		device->oclass[NVDEV_SUBDEV_CLOCK  ] = &nv50_clock_oclass;
+		device->oclass[NVDEV_SUBDEV_CLOCK  ] =  nv50_clock_oclass;
 		device->oclass[NVDEV_SUBDEV_THERM  ] = &nv50_therm_oclass;
 		device->oclass[NVDEV_SUBDEV_MXM    ] = &nv50_mxm_oclass;
 		device->oclass[NVDEV_SUBDEV_DEVINIT] = &nv50_devinit_oclass;
-		device->oclass[NVDEV_SUBDEV_MC     ] = &nv50_mc_oclass;
-		device->oclass[NVDEV_SUBDEV_BUS    ] = &nv50_bus_oclass;
+		device->oclass[NVDEV_SUBDEV_MC     ] =  nv50_mc_oclass;
+		device->oclass[NVDEV_SUBDEV_BUS    ] =  nv50_bus_oclass;
 		device->oclass[NVDEV_SUBDEV_TIMER  ] = &nv04_timer_oclass;
-		device->oclass[NVDEV_SUBDEV_FB     ] = &nv50_fb_oclass;
+		device->oclass[NVDEV_SUBDEV_FB     ] =  nv50_fb_oclass;
 		device->oclass[NVDEV_SUBDEV_INSTMEM] = &nv50_instmem_oclass;
 		device->oclass[NVDEV_SUBDEV_VM     ] = &nv50_vmmgr_oclass;
 		device->oclass[NVDEV_SUBDEV_BAR    ] = &nv50_bar_oclass;
+		device->oclass[NVDEV_SUBDEV_VOLT   ] = &nv40_volt_oclass;
 		device->oclass[NVDEV_ENGINE_DMAOBJ ] = &nv50_dmaeng_oclass;
-		device->oclass[NVDEV_ENGINE_FIFO   ] = &nv50_fifo_oclass;
-		device->oclass[NVDEV_ENGINE_SW     ] = &nv50_software_oclass;
+		device->oclass[NVDEV_ENGINE_FIFO   ] =  nv50_fifo_oclass;
+		device->oclass[NVDEV_ENGINE_SW     ] =  nv50_software_oclass;
 		device->oclass[NVDEV_ENGINE_GR     ] = &nv50_graph_oclass;
 		device->oclass[NVDEV_ENGINE_MPEG   ] = &nv50_mpeg_oclass;
 		device->oclass[NVDEV_ENGINE_DISP   ] = &nv50_disp_oclass;
+		device->oclass[NVDEV_ENGINE_PERFMON] =  nv50_perfmon_oclass;
 		break;
 	case 0x84:
 		device->cname = "G84";
 		device->oclass[NVDEV_SUBDEV_VBIOS  ] = &nouveau_bios_oclass;
 		device->oclass[NVDEV_SUBDEV_GPIO   ] = &nv50_gpio_oclass;
 		device->oclass[NVDEV_SUBDEV_I2C    ] = &nv50_i2c_oclass;
-		device->oclass[NVDEV_SUBDEV_CLOCK  ] = &nv50_clock_oclass;
+		device->oclass[NVDEV_SUBDEV_CLOCK  ] =  nv84_clock_oclass;
 		device->oclass[NVDEV_SUBDEV_THERM  ] = &nv84_therm_oclass;
 		device->oclass[NVDEV_SUBDEV_MXM    ] = &nv50_mxm_oclass;
 		device->oclass[NVDEV_SUBDEV_DEVINIT] = &nv50_devinit_oclass;
-		device->oclass[NVDEV_SUBDEV_MC     ] = &nv50_mc_oclass;
-		device->oclass[NVDEV_SUBDEV_BUS    ] = &nv50_bus_oclass;
+		device->oclass[NVDEV_SUBDEV_MC     ] =  nv50_mc_oclass;
+		device->oclass[NVDEV_SUBDEV_BUS    ] =  nv50_bus_oclass;
 		device->oclass[NVDEV_SUBDEV_TIMER  ] = &nv04_timer_oclass;
-		device->oclass[NVDEV_SUBDEV_FB     ] = &nv50_fb_oclass;
+		device->oclass[NVDEV_SUBDEV_FB     ] =  nv84_fb_oclass;
 		device->oclass[NVDEV_SUBDEV_INSTMEM] = &nv50_instmem_oclass;
 		device->oclass[NVDEV_SUBDEV_VM     ] = &nv50_vmmgr_oclass;
 		device->oclass[NVDEV_SUBDEV_BAR    ] = &nv50_bar_oclass;
+		device->oclass[NVDEV_SUBDEV_VOLT   ] = &nv40_volt_oclass;
 		device->oclass[NVDEV_ENGINE_DMAOBJ ] = &nv50_dmaeng_oclass;
-		device->oclass[NVDEV_ENGINE_FIFO   ] = &nv84_fifo_oclass;
-		device->oclass[NVDEV_ENGINE_SW     ] = &nv50_software_oclass;
+		device->oclass[NVDEV_ENGINE_FIFO   ] =  nv84_fifo_oclass;
+		device->oclass[NVDEV_ENGINE_SW     ] =  nv50_software_oclass;
 		device->oclass[NVDEV_ENGINE_GR     ] = &nv50_graph_oclass;
 		device->oclass[NVDEV_ENGINE_MPEG   ] = &nv84_mpeg_oclass;
 		device->oclass[NVDEV_ENGINE_VP     ] = &nv84_vp_oclass;
 		device->oclass[NVDEV_ENGINE_CRYPT  ] = &nv84_crypt_oclass;
 		device->oclass[NVDEV_ENGINE_BSP    ] = &nv84_bsp_oclass;
 		device->oclass[NVDEV_ENGINE_DISP   ] = &nv84_disp_oclass;
+		device->oclass[NVDEV_ENGINE_PERFMON] =  nv84_perfmon_oclass;
 		break;
 	case 0x86:
 		device->cname = "G86";
 		device->oclass[NVDEV_SUBDEV_VBIOS  ] = &nouveau_bios_oclass;
 		device->oclass[NVDEV_SUBDEV_GPIO   ] = &nv50_gpio_oclass;
 		device->oclass[NVDEV_SUBDEV_I2C    ] = &nv50_i2c_oclass;
-		device->oclass[NVDEV_SUBDEV_CLOCK  ] = &nv50_clock_oclass;
+		device->oclass[NVDEV_SUBDEV_CLOCK  ] =  nv84_clock_oclass;
 		device->oclass[NVDEV_SUBDEV_THERM  ] = &nv84_therm_oclass;
 		device->oclass[NVDEV_SUBDEV_MXM    ] = &nv50_mxm_oclass;
 		device->oclass[NVDEV_SUBDEV_DEVINIT] = &nv50_devinit_oclass;
-		device->oclass[NVDEV_SUBDEV_MC     ] = &nv50_mc_oclass;
-		device->oclass[NVDEV_SUBDEV_BUS    ] = &nv50_bus_oclass;
+		device->oclass[NVDEV_SUBDEV_MC     ] =  nv50_mc_oclass;
+		device->oclass[NVDEV_SUBDEV_BUS    ] =  nv50_bus_oclass;
 		device->oclass[NVDEV_SUBDEV_TIMER  ] = &nv04_timer_oclass;
-		device->oclass[NVDEV_SUBDEV_FB     ] = &nv50_fb_oclass;
+		device->oclass[NVDEV_SUBDEV_FB     ] =  nv84_fb_oclass;
 		device->oclass[NVDEV_SUBDEV_INSTMEM] = &nv50_instmem_oclass;
 		device->oclass[NVDEV_SUBDEV_VM     ] = &nv50_vmmgr_oclass;
 		device->oclass[NVDEV_SUBDEV_BAR    ] = &nv50_bar_oclass;
+		device->oclass[NVDEV_SUBDEV_VOLT   ] = &nv40_volt_oclass;
 		device->oclass[NVDEV_ENGINE_DMAOBJ ] = &nv50_dmaeng_oclass;
-		device->oclass[NVDEV_ENGINE_FIFO   ] = &nv84_fifo_oclass;
-		device->oclass[NVDEV_ENGINE_SW     ] = &nv50_software_oclass;
+		device->oclass[NVDEV_ENGINE_FIFO   ] =  nv84_fifo_oclass;
+		device->oclass[NVDEV_ENGINE_SW     ] =  nv50_software_oclass;
 		device->oclass[NVDEV_ENGINE_GR     ] = &nv50_graph_oclass;
 		device->oclass[NVDEV_ENGINE_MPEG   ] = &nv84_mpeg_oclass;
 		device->oclass[NVDEV_ENGINE_VP     ] = &nv84_vp_oclass;
 		device->oclass[NVDEV_ENGINE_CRYPT  ] = &nv84_crypt_oclass;
 		device->oclass[NVDEV_ENGINE_BSP    ] = &nv84_bsp_oclass;
 		device->oclass[NVDEV_ENGINE_DISP   ] = &nv84_disp_oclass;
+		device->oclass[NVDEV_ENGINE_PERFMON] =  nv84_perfmon_oclass;
 		break;
 	case 0x92:
 		device->cname = "G92";
 		device->oclass[NVDEV_SUBDEV_VBIOS  ] = &nouveau_bios_oclass;
 		device->oclass[NVDEV_SUBDEV_GPIO   ] = &nv50_gpio_oclass;
 		device->oclass[NVDEV_SUBDEV_I2C    ] = &nv50_i2c_oclass;
-		device->oclass[NVDEV_SUBDEV_CLOCK  ] = &nv50_clock_oclass;
+		device->oclass[NVDEV_SUBDEV_CLOCK  ] =  nv84_clock_oclass;
 		device->oclass[NVDEV_SUBDEV_THERM  ] = &nv84_therm_oclass;
 		device->oclass[NVDEV_SUBDEV_MXM    ] = &nv50_mxm_oclass;
 		device->oclass[NVDEV_SUBDEV_DEVINIT] = &nv50_devinit_oclass;
-		device->oclass[NVDEV_SUBDEV_MC     ] = &nv50_mc_oclass;
-		device->oclass[NVDEV_SUBDEV_BUS    ] = &nv50_bus_oclass;
+		device->oclass[NVDEV_SUBDEV_MC     ] =  nv50_mc_oclass;
+		device->oclass[NVDEV_SUBDEV_BUS    ] =  nv50_bus_oclass;
 		device->oclass[NVDEV_SUBDEV_TIMER  ] = &nv04_timer_oclass;
-		device->oclass[NVDEV_SUBDEV_FB     ] = &nv50_fb_oclass;
+		device->oclass[NVDEV_SUBDEV_FB     ] =  nv84_fb_oclass;
 		device->oclass[NVDEV_SUBDEV_INSTMEM] = &nv50_instmem_oclass;
 		device->oclass[NVDEV_SUBDEV_VM     ] = &nv50_vmmgr_oclass;
 		device->oclass[NVDEV_SUBDEV_BAR    ] = &nv50_bar_oclass;
+		device->oclass[NVDEV_SUBDEV_VOLT   ] = &nv40_volt_oclass;
 		device->oclass[NVDEV_ENGINE_DMAOBJ ] = &nv50_dmaeng_oclass;
-		device->oclass[NVDEV_ENGINE_FIFO   ] = &nv84_fifo_oclass;
-		device->oclass[NVDEV_ENGINE_SW     ] = &nv50_software_oclass;
+		device->oclass[NVDEV_ENGINE_FIFO   ] =  nv84_fifo_oclass;
+		device->oclass[NVDEV_ENGINE_SW     ] =  nv50_software_oclass;
 		device->oclass[NVDEV_ENGINE_GR     ] = &nv50_graph_oclass;
 		device->oclass[NVDEV_ENGINE_MPEG   ] = &nv84_mpeg_oclass;
 		device->oclass[NVDEV_ENGINE_VP     ] = &nv84_vp_oclass;
 		device->oclass[NVDEV_ENGINE_CRYPT  ] = &nv84_crypt_oclass;
 		device->oclass[NVDEV_ENGINE_BSP    ] = &nv84_bsp_oclass;
 		device->oclass[NVDEV_ENGINE_DISP   ] = &nv84_disp_oclass;
+		device->oclass[NVDEV_ENGINE_PERFMON] =  nv84_perfmon_oclass;
 		break;
 	case 0x94:
 		device->cname = "G94";
 		device->oclass[NVDEV_SUBDEV_VBIOS  ] = &nouveau_bios_oclass;
 		device->oclass[NVDEV_SUBDEV_GPIO   ] = &nv50_gpio_oclass;
 		device->oclass[NVDEV_SUBDEV_I2C    ] = &nv94_i2c_oclass;
-		device->oclass[NVDEV_SUBDEV_CLOCK  ] = &nv50_clock_oclass;
+		device->oclass[NVDEV_SUBDEV_CLOCK  ] =  nv84_clock_oclass;
 		device->oclass[NVDEV_SUBDEV_THERM  ] = &nv84_therm_oclass;
 		device->oclass[NVDEV_SUBDEV_MXM    ] = &nv50_mxm_oclass;
 		device->oclass[NVDEV_SUBDEV_DEVINIT] = &nv50_devinit_oclass;
-		device->oclass[NVDEV_SUBDEV_MC     ] = &nv50_mc_oclass;
-		device->oclass[NVDEV_SUBDEV_BUS    ] = &nv50_bus_oclass;
+		device->oclass[NVDEV_SUBDEV_MC     ] =  nv94_mc_oclass;
+		device->oclass[NVDEV_SUBDEV_BUS    ] =  nv94_bus_oclass;
 		device->oclass[NVDEV_SUBDEV_TIMER  ] = &nv04_timer_oclass;
-		device->oclass[NVDEV_SUBDEV_FB     ] = &nv50_fb_oclass;
+		device->oclass[NVDEV_SUBDEV_FB     ] =  nv84_fb_oclass;
 		device->oclass[NVDEV_SUBDEV_INSTMEM] = &nv50_instmem_oclass;
 		device->oclass[NVDEV_SUBDEV_VM     ] = &nv50_vmmgr_oclass;
 		device->oclass[NVDEV_SUBDEV_BAR    ] = &nv50_bar_oclass;
+		device->oclass[NVDEV_SUBDEV_VOLT   ] = &nv40_volt_oclass;
 		device->oclass[NVDEV_ENGINE_DMAOBJ ] = &nv50_dmaeng_oclass;
-		device->oclass[NVDEV_ENGINE_FIFO   ] = &nv84_fifo_oclass;
-		device->oclass[NVDEV_ENGINE_SW     ] = &nv50_software_oclass;
+		device->oclass[NVDEV_ENGINE_FIFO   ] =  nv84_fifo_oclass;
+		device->oclass[NVDEV_ENGINE_SW     ] =  nv50_software_oclass;
 		device->oclass[NVDEV_ENGINE_GR     ] = &nv50_graph_oclass;
 		device->oclass[NVDEV_ENGINE_MPEG   ] = &nv84_mpeg_oclass;
 		device->oclass[NVDEV_ENGINE_VP     ] = &nv84_vp_oclass;
 		device->oclass[NVDEV_ENGINE_CRYPT  ] = &nv84_crypt_oclass;
 		device->oclass[NVDEV_ENGINE_BSP    ] = &nv84_bsp_oclass;
 		device->oclass[NVDEV_ENGINE_DISP   ] = &nv94_disp_oclass;
+		device->oclass[NVDEV_ENGINE_PERFMON] =  nv84_perfmon_oclass;
 		break;
 	case 0x96:
 		device->cname = "G96";
 		device->oclass[NVDEV_SUBDEV_VBIOS  ] = &nouveau_bios_oclass;
 		device->oclass[NVDEV_SUBDEV_GPIO   ] = &nv50_gpio_oclass;
 		device->oclass[NVDEV_SUBDEV_I2C    ] = &nv94_i2c_oclass;
-		device->oclass[NVDEV_SUBDEV_CLOCK  ] = &nv50_clock_oclass;
+		device->oclass[NVDEV_SUBDEV_CLOCK  ] =  nv84_clock_oclass;
 		device->oclass[NVDEV_SUBDEV_THERM  ] = &nv84_therm_oclass;
 		device->oclass[NVDEV_SUBDEV_MXM    ] = &nv50_mxm_oclass;
 		device->oclass[NVDEV_SUBDEV_DEVINIT] = &nv50_devinit_oclass;
-		device->oclass[NVDEV_SUBDEV_MC     ] = &nv50_mc_oclass;
-		device->oclass[NVDEV_SUBDEV_BUS    ] = &nv50_bus_oclass;
+		device->oclass[NVDEV_SUBDEV_MC     ] =  nv94_mc_oclass;
+		device->oclass[NVDEV_SUBDEV_BUS    ] =  nv94_bus_oclass;
 		device->oclass[NVDEV_SUBDEV_TIMER  ] = &nv04_timer_oclass;
-		device->oclass[NVDEV_SUBDEV_FB     ] = &nv50_fb_oclass;
+		device->oclass[NVDEV_SUBDEV_FB     ] =  nv84_fb_oclass;
 		device->oclass[NVDEV_SUBDEV_INSTMEM] = &nv50_instmem_oclass;
 		device->oclass[NVDEV_SUBDEV_VM     ] = &nv50_vmmgr_oclass;
 		device->oclass[NVDEV_SUBDEV_BAR    ] = &nv50_bar_oclass;
+		device->oclass[NVDEV_SUBDEV_VOLT   ] = &nv40_volt_oclass;
 		device->oclass[NVDEV_ENGINE_DMAOBJ ] = &nv50_dmaeng_oclass;
-		device->oclass[NVDEV_ENGINE_FIFO   ] = &nv84_fifo_oclass;
-		device->oclass[NVDEV_ENGINE_SW     ] = &nv50_software_oclass;
+		device->oclass[NVDEV_ENGINE_FIFO   ] =  nv84_fifo_oclass;
+		device->oclass[NVDEV_ENGINE_SW     ] =  nv50_software_oclass;
 		device->oclass[NVDEV_ENGINE_GR     ] = &nv50_graph_oclass;
 		device->oclass[NVDEV_ENGINE_MPEG   ] = &nv84_mpeg_oclass;
 		device->oclass[NVDEV_ENGINE_VP     ] = &nv84_vp_oclass;
 		device->oclass[NVDEV_ENGINE_CRYPT  ] = &nv84_crypt_oclass;
 		device->oclass[NVDEV_ENGINE_BSP    ] = &nv84_bsp_oclass;
 		device->oclass[NVDEV_ENGINE_DISP   ] = &nv94_disp_oclass;
+		device->oclass[NVDEV_ENGINE_PERFMON] =  nv84_perfmon_oclass;
 		break;
 	case 0x98:
 		device->cname = "G98";
 		device->oclass[NVDEV_SUBDEV_VBIOS  ] = &nouveau_bios_oclass;
 		device->oclass[NVDEV_SUBDEV_GPIO   ] = &nv50_gpio_oclass;
 		device->oclass[NVDEV_SUBDEV_I2C    ] = &nv94_i2c_oclass;
-		device->oclass[NVDEV_SUBDEV_CLOCK  ] = &nv50_clock_oclass;
+		device->oclass[NVDEV_SUBDEV_CLOCK  ] =  nv84_clock_oclass;
 		device->oclass[NVDEV_SUBDEV_THERM  ] = &nv84_therm_oclass;
 		device->oclass[NVDEV_SUBDEV_MXM    ] = &nv50_mxm_oclass;
 		device->oclass[NVDEV_SUBDEV_DEVINIT] = &nv50_devinit_oclass;
-		device->oclass[NVDEV_SUBDEV_MC     ] = &nv98_mc_oclass;
-		device->oclass[NVDEV_SUBDEV_BUS    ] = &nv50_bus_oclass;
+		device->oclass[NVDEV_SUBDEV_MC     ] =  nv98_mc_oclass;
+		device->oclass[NVDEV_SUBDEV_BUS    ] =  nv94_bus_oclass;
 		device->oclass[NVDEV_SUBDEV_TIMER  ] = &nv04_timer_oclass;
-		device->oclass[NVDEV_SUBDEV_FB     ] = &nv50_fb_oclass;
+		device->oclass[NVDEV_SUBDEV_FB     ] =  nv84_fb_oclass;
 		device->oclass[NVDEV_SUBDEV_INSTMEM] = &nv50_instmem_oclass;
 		device->oclass[NVDEV_SUBDEV_VM     ] = &nv50_vmmgr_oclass;
 		device->oclass[NVDEV_SUBDEV_BAR    ] = &nv50_bar_oclass;
+		device->oclass[NVDEV_SUBDEV_VOLT   ] = &nv40_volt_oclass;
 		device->oclass[NVDEV_ENGINE_DMAOBJ ] = &nv50_dmaeng_oclass;
-		device->oclass[NVDEV_ENGINE_FIFO   ] = &nv84_fifo_oclass;
-		device->oclass[NVDEV_ENGINE_SW     ] = &nv50_software_oclass;
+		device->oclass[NVDEV_ENGINE_FIFO   ] =  nv84_fifo_oclass;
+		device->oclass[NVDEV_ENGINE_SW     ] =  nv50_software_oclass;
 		device->oclass[NVDEV_ENGINE_GR     ] = &nv50_graph_oclass;
 		device->oclass[NVDEV_ENGINE_VP     ] = &nv98_vp_oclass;
 		device->oclass[NVDEV_ENGINE_CRYPT  ] = &nv98_crypt_oclass;
 		device->oclass[NVDEV_ENGINE_BSP    ] = &nv98_bsp_oclass;
 		device->oclass[NVDEV_ENGINE_PPP    ] = &nv98_ppp_oclass;
 		device->oclass[NVDEV_ENGINE_DISP   ] = &nv94_disp_oclass;
+		device->oclass[NVDEV_ENGINE_PERFMON] =  nv84_perfmon_oclass;
 		break;
 	case 0xa0:
 		device->cname = "G200";
 		device->oclass[NVDEV_SUBDEV_VBIOS  ] = &nouveau_bios_oclass;
 		device->oclass[NVDEV_SUBDEV_GPIO   ] = &nv50_gpio_oclass;
 		device->oclass[NVDEV_SUBDEV_I2C    ] = &nv50_i2c_oclass;
-		device->oclass[NVDEV_SUBDEV_CLOCK  ] = &nv50_clock_oclass;
+		device->oclass[NVDEV_SUBDEV_CLOCK  ] =  nv84_clock_oclass;
 		device->oclass[NVDEV_SUBDEV_THERM  ] = &nv84_therm_oclass;
 		device->oclass[NVDEV_SUBDEV_MXM    ] = &nv50_mxm_oclass;
 		device->oclass[NVDEV_SUBDEV_DEVINIT] = &nv50_devinit_oclass;
-		device->oclass[NVDEV_SUBDEV_MC     ] = &nv98_mc_oclass;
-		device->oclass[NVDEV_SUBDEV_BUS    ] = &nv50_bus_oclass;
+		device->oclass[NVDEV_SUBDEV_MC     ] =  nv98_mc_oclass;
+		device->oclass[NVDEV_SUBDEV_BUS    ] =  nv94_bus_oclass;
 		device->oclass[NVDEV_SUBDEV_TIMER  ] = &nv04_timer_oclass;
-		device->oclass[NVDEV_SUBDEV_FB     ] = &nv50_fb_oclass;
+		device->oclass[NVDEV_SUBDEV_FB     ] =  nv84_fb_oclass;
 		device->oclass[NVDEV_SUBDEV_INSTMEM] = &nv50_instmem_oclass;
 		device->oclass[NVDEV_SUBDEV_VM     ] = &nv50_vmmgr_oclass;
 		device->oclass[NVDEV_SUBDEV_BAR    ] = &nv50_bar_oclass;
+		device->oclass[NVDEV_SUBDEV_VOLT   ] = &nv40_volt_oclass;
 		device->oclass[NVDEV_ENGINE_DMAOBJ ] = &nv50_dmaeng_oclass;
-		device->oclass[NVDEV_ENGINE_FIFO   ] = &nv84_fifo_oclass;
-		device->oclass[NVDEV_ENGINE_SW     ] = &nv50_software_oclass;
+		device->oclass[NVDEV_ENGINE_FIFO   ] =  nv84_fifo_oclass;
+		device->oclass[NVDEV_ENGINE_SW     ] =  nv50_software_oclass;
 		device->oclass[NVDEV_ENGINE_GR     ] = &nv50_graph_oclass;
 		device->oclass[NVDEV_ENGINE_MPEG   ] = &nv84_mpeg_oclass;
 		device->oclass[NVDEV_ENGINE_VP     ] = &nv84_vp_oclass;
 		device->oclass[NVDEV_ENGINE_CRYPT  ] = &nv84_crypt_oclass;
 		device->oclass[NVDEV_ENGINE_BSP    ] = &nv84_bsp_oclass;
 		device->oclass[NVDEV_ENGINE_DISP   ] = &nva0_disp_oclass;
+		device->oclass[NVDEV_ENGINE_PERFMON] =  nv84_perfmon_oclass;
 		break;
 	case 0xaa:
 		device->cname = "MCP77/MCP78";
 		device->oclass[NVDEV_SUBDEV_VBIOS  ] = &nouveau_bios_oclass;
 		device->oclass[NVDEV_SUBDEV_GPIO   ] = &nv50_gpio_oclass;
 		device->oclass[NVDEV_SUBDEV_I2C    ] = &nv94_i2c_oclass;
-		device->oclass[NVDEV_SUBDEV_CLOCK  ] = &nv50_clock_oclass;
+		device->oclass[NVDEV_SUBDEV_CLOCK  ] =  nv84_clock_oclass;
 		device->oclass[NVDEV_SUBDEV_THERM  ] = &nv84_therm_oclass;
 		device->oclass[NVDEV_SUBDEV_MXM    ] = &nv50_mxm_oclass;
 		device->oclass[NVDEV_SUBDEV_DEVINIT] = &nv50_devinit_oclass;
-		device->oclass[NVDEV_SUBDEV_MC     ] = &nv98_mc_oclass;
-		device->oclass[NVDEV_SUBDEV_BUS    ] = &nv50_bus_oclass;
+		device->oclass[NVDEV_SUBDEV_MC     ] =  nv98_mc_oclass;
+		device->oclass[NVDEV_SUBDEV_BUS    ] =  nv94_bus_oclass;
 		device->oclass[NVDEV_SUBDEV_TIMER  ] = &nv04_timer_oclass;
-		device->oclass[NVDEV_SUBDEV_FB     ] = &nv50_fb_oclass;
+		device->oclass[NVDEV_SUBDEV_FB     ] =  nvaa_fb_oclass;
 		device->oclass[NVDEV_SUBDEV_INSTMEM] = &nv50_instmem_oclass;
 		device->oclass[NVDEV_SUBDEV_VM     ] = &nv50_vmmgr_oclass;
 		device->oclass[NVDEV_SUBDEV_BAR    ] = &nv50_bar_oclass;
+		device->oclass[NVDEV_SUBDEV_VOLT   ] = &nv40_volt_oclass;
 		device->oclass[NVDEV_ENGINE_DMAOBJ ] = &nv50_dmaeng_oclass;
-		device->oclass[NVDEV_ENGINE_FIFO   ] = &nv84_fifo_oclass;
-		device->oclass[NVDEV_ENGINE_SW     ] = &nv50_software_oclass;
+		device->oclass[NVDEV_ENGINE_FIFO   ] =  nv84_fifo_oclass;
+		device->oclass[NVDEV_ENGINE_SW     ] =  nv50_software_oclass;
 		device->oclass[NVDEV_ENGINE_GR     ] = &nv50_graph_oclass;
 		device->oclass[NVDEV_ENGINE_VP     ] = &nv98_vp_oclass;
 		device->oclass[NVDEV_ENGINE_CRYPT  ] = &nv98_crypt_oclass;
 		device->oclass[NVDEV_ENGINE_BSP    ] = &nv98_bsp_oclass;
 		device->oclass[NVDEV_ENGINE_PPP    ] = &nv98_ppp_oclass;
 		device->oclass[NVDEV_ENGINE_DISP   ] = &nv94_disp_oclass;
+		device->oclass[NVDEV_ENGINE_PERFMON] =  nv84_perfmon_oclass;
 		break;
 	case 0xac:
 		device->cname = "MCP79/MCP7A";
 		device->oclass[NVDEV_SUBDEV_VBIOS  ] = &nouveau_bios_oclass;
 		device->oclass[NVDEV_SUBDEV_GPIO   ] = &nv50_gpio_oclass;
 		device->oclass[NVDEV_SUBDEV_I2C    ] = &nv94_i2c_oclass;
-		device->oclass[NVDEV_SUBDEV_CLOCK  ] = &nv50_clock_oclass;
+		device->oclass[NVDEV_SUBDEV_CLOCK  ] =  nv84_clock_oclass;
 		device->oclass[NVDEV_SUBDEV_THERM  ] = &nv84_therm_oclass;
 		device->oclass[NVDEV_SUBDEV_MXM    ] = &nv50_mxm_oclass;
 		device->oclass[NVDEV_SUBDEV_DEVINIT] = &nv50_devinit_oclass;
-		device->oclass[NVDEV_SUBDEV_MC     ] = &nv98_mc_oclass;
-		device->oclass[NVDEV_SUBDEV_BUS    ] = &nv50_bus_oclass;
+		device->oclass[NVDEV_SUBDEV_MC     ] =  nv98_mc_oclass;
+		device->oclass[NVDEV_SUBDEV_BUS    ] =  nv94_bus_oclass;
 		device->oclass[NVDEV_SUBDEV_TIMER  ] = &nv04_timer_oclass;
-		device->oclass[NVDEV_SUBDEV_FB     ] = &nv50_fb_oclass;
+		device->oclass[NVDEV_SUBDEV_FB     ] =  nvaa_fb_oclass;
 		device->oclass[NVDEV_SUBDEV_INSTMEM] = &nv50_instmem_oclass;
 		device->oclass[NVDEV_SUBDEV_VM     ] = &nv50_vmmgr_oclass;
 		device->oclass[NVDEV_SUBDEV_BAR    ] = &nv50_bar_oclass;
+		device->oclass[NVDEV_SUBDEV_VOLT   ] = &nv40_volt_oclass;
 		device->oclass[NVDEV_ENGINE_DMAOBJ ] = &nv50_dmaeng_oclass;
-		device->oclass[NVDEV_ENGINE_FIFO   ] = &nv84_fifo_oclass;
-		device->oclass[NVDEV_ENGINE_SW     ] = &nv50_software_oclass;
+		device->oclass[NVDEV_ENGINE_FIFO   ] =  nv84_fifo_oclass;
+		device->oclass[NVDEV_ENGINE_SW     ] =  nv50_software_oclass;
 		device->oclass[NVDEV_ENGINE_GR     ] = &nv50_graph_oclass;
 		device->oclass[NVDEV_ENGINE_VP     ] = &nv98_vp_oclass;
 		device->oclass[NVDEV_ENGINE_CRYPT  ] = &nv98_crypt_oclass;
 		device->oclass[NVDEV_ENGINE_BSP    ] = &nv98_bsp_oclass;
 		device->oclass[NVDEV_ENGINE_PPP    ] = &nv98_ppp_oclass;
 		device->oclass[NVDEV_ENGINE_DISP   ] = &nv94_disp_oclass;
+		device->oclass[NVDEV_ENGINE_PERFMON] =  nv84_perfmon_oclass;
 		break;
 	case 0xa3:
 		device->cname = "GT215";
@@ -320,16 +343,18 @@ nv50_identify(struct nouveau_device *device)
 		device->oclass[NVDEV_SUBDEV_THERM  ] = &nva3_therm_oclass;
 		device->oclass[NVDEV_SUBDEV_MXM    ] = &nv50_mxm_oclass;
 		device->oclass[NVDEV_SUBDEV_DEVINIT] = &nva3_devinit_oclass;
-		device->oclass[NVDEV_SUBDEV_MC     ] = &nv98_mc_oclass;
-		device->oclass[NVDEV_SUBDEV_BUS    ] = &nv50_bus_oclass;
+		device->oclass[NVDEV_SUBDEV_MC     ] =  nv98_mc_oclass;
+		device->oclass[NVDEV_SUBDEV_BUS    ] =  nv94_bus_oclass;
 		device->oclass[NVDEV_SUBDEV_TIMER  ] = &nv04_timer_oclass;
-		device->oclass[NVDEV_SUBDEV_FB     ] = &nv50_fb_oclass;
+		device->oclass[NVDEV_SUBDEV_FB     ] =  nva3_fb_oclass;
 		device->oclass[NVDEV_SUBDEV_INSTMEM] = &nv50_instmem_oclass;
 		device->oclass[NVDEV_SUBDEV_VM     ] = &nv50_vmmgr_oclass;
 		device->oclass[NVDEV_SUBDEV_BAR    ] = &nv50_bar_oclass;
+		device->oclass[NVDEV_SUBDEV_PWR    ] = &nva3_pwr_oclass;
+		device->oclass[NVDEV_SUBDEV_VOLT   ] = &nv40_volt_oclass;
 		device->oclass[NVDEV_ENGINE_DMAOBJ ] = &nv50_dmaeng_oclass;
-		device->oclass[NVDEV_ENGINE_FIFO   ] = &nv84_fifo_oclass;
-		device->oclass[NVDEV_ENGINE_SW     ] = &nv50_software_oclass;
+		device->oclass[NVDEV_ENGINE_FIFO   ] =  nv84_fifo_oclass;
+		device->oclass[NVDEV_ENGINE_SW     ] =  nv50_software_oclass;
 		device->oclass[NVDEV_ENGINE_GR     ] = &nv50_graph_oclass;
 		device->oclass[NVDEV_ENGINE_MPEG   ] = &nv84_mpeg_oclass;
 		device->oclass[NVDEV_ENGINE_VP     ] = &nv98_vp_oclass;
@@ -337,6 +362,7 @@ nv50_identify(struct nouveau_device *device)
 		device->oclass[NVDEV_ENGINE_PPP    ] = &nv98_ppp_oclass;
 		device->oclass[NVDEV_ENGINE_COPY0  ] = &nva3_copy_oclass;
 		device->oclass[NVDEV_ENGINE_DISP   ] = &nva3_disp_oclass;
+		device->oclass[NVDEV_ENGINE_PERFMON] =  nva3_perfmon_oclass;
 		break;
 	case 0xa5:
 		device->cname = "GT216";
@@ -347,22 +373,25 @@ nv50_identify(struct nouveau_device *device)
 		device->oclass[NVDEV_SUBDEV_THERM  ] = &nva3_therm_oclass;
 		device->oclass[NVDEV_SUBDEV_MXM    ] = &nv50_mxm_oclass;
 		device->oclass[NVDEV_SUBDEV_DEVINIT] = &nva3_devinit_oclass;
-		device->oclass[NVDEV_SUBDEV_MC     ] = &nv98_mc_oclass;
-		device->oclass[NVDEV_SUBDEV_BUS    ] = &nv50_bus_oclass;
+		device->oclass[NVDEV_SUBDEV_MC     ] =  nv98_mc_oclass;
+		device->oclass[NVDEV_SUBDEV_BUS    ] =  nv94_bus_oclass;
 		device->oclass[NVDEV_SUBDEV_TIMER  ] = &nv04_timer_oclass;
-		device->oclass[NVDEV_SUBDEV_FB     ] = &nv50_fb_oclass;
+		device->oclass[NVDEV_SUBDEV_FB     ] =  nva3_fb_oclass;
 		device->oclass[NVDEV_SUBDEV_INSTMEM] = &nv50_instmem_oclass;
 		device->oclass[NVDEV_SUBDEV_VM     ] = &nv50_vmmgr_oclass;
 		device->oclass[NVDEV_SUBDEV_BAR    ] = &nv50_bar_oclass;
+		device->oclass[NVDEV_SUBDEV_PWR    ] = &nva3_pwr_oclass;
+		device->oclass[NVDEV_SUBDEV_VOLT   ] = &nv40_volt_oclass;
 		device->oclass[NVDEV_ENGINE_DMAOBJ ] = &nv50_dmaeng_oclass;
-		device->oclass[NVDEV_ENGINE_FIFO   ] = &nv84_fifo_oclass;
-		device->oclass[NVDEV_ENGINE_SW     ] = &nv50_software_oclass;
+		device->oclass[NVDEV_ENGINE_FIFO   ] =  nv84_fifo_oclass;
+		device->oclass[NVDEV_ENGINE_SW     ] =  nv50_software_oclass;
 		device->oclass[NVDEV_ENGINE_GR     ] = &nv50_graph_oclass;
 		device->oclass[NVDEV_ENGINE_VP     ] = &nv98_vp_oclass;
 		device->oclass[NVDEV_ENGINE_BSP    ] = &nv98_bsp_oclass;
 		device->oclass[NVDEV_ENGINE_PPP    ] = &nv98_ppp_oclass;
 		device->oclass[NVDEV_ENGINE_COPY0  ] = &nva3_copy_oclass;
 		device->oclass[NVDEV_ENGINE_DISP   ] = &nva3_disp_oclass;
+		device->oclass[NVDEV_ENGINE_PERFMON] =  nva3_perfmon_oclass;
 		break;
 	case 0xa8:
 		device->cname = "GT218";
@@ -373,22 +402,25 @@ nv50_identify(struct nouveau_device *device)
 		device->oclass[NVDEV_SUBDEV_THERM  ] = &nva3_therm_oclass;
 		device->oclass[NVDEV_SUBDEV_MXM    ] = &nv50_mxm_oclass;
 		device->oclass[NVDEV_SUBDEV_DEVINIT] = &nva3_devinit_oclass;
-		device->oclass[NVDEV_SUBDEV_MC     ] = &nv98_mc_oclass;
-		device->oclass[NVDEV_SUBDEV_BUS    ] = &nv50_bus_oclass;
+		device->oclass[NVDEV_SUBDEV_MC     ] =  nv98_mc_oclass;
+		device->oclass[NVDEV_SUBDEV_BUS    ] =  nv94_bus_oclass;
 		device->oclass[NVDEV_SUBDEV_TIMER  ] = &nv04_timer_oclass;
-		device->oclass[NVDEV_SUBDEV_FB     ] = &nv50_fb_oclass;
+		device->oclass[NVDEV_SUBDEV_FB     ] =  nva3_fb_oclass;
 		device->oclass[NVDEV_SUBDEV_INSTMEM] = &nv50_instmem_oclass;
 		device->oclass[NVDEV_SUBDEV_VM     ] = &nv50_vmmgr_oclass;
 		device->oclass[NVDEV_SUBDEV_BAR    ] = &nv50_bar_oclass;
+		device->oclass[NVDEV_SUBDEV_PWR    ] = &nva3_pwr_oclass;
+		device->oclass[NVDEV_SUBDEV_VOLT   ] = &nv40_volt_oclass;
 		device->oclass[NVDEV_ENGINE_DMAOBJ ] = &nv50_dmaeng_oclass;
-		device->oclass[NVDEV_ENGINE_FIFO   ] = &nv84_fifo_oclass;
-		device->oclass[NVDEV_ENGINE_SW     ] = &nv50_software_oclass;
+		device->oclass[NVDEV_ENGINE_FIFO   ] =  nv84_fifo_oclass;
+		device->oclass[NVDEV_ENGINE_SW     ] =  nv50_software_oclass;
 		device->oclass[NVDEV_ENGINE_GR     ] = &nv50_graph_oclass;
 		device->oclass[NVDEV_ENGINE_VP     ] = &nv98_vp_oclass;
 		device->oclass[NVDEV_ENGINE_BSP    ] = &nv98_bsp_oclass;
 		device->oclass[NVDEV_ENGINE_PPP    ] = &nv98_ppp_oclass;
 		device->oclass[NVDEV_ENGINE_COPY0  ] = &nva3_copy_oclass;
 		device->oclass[NVDEV_ENGINE_DISP   ] = &nva3_disp_oclass;
+		device->oclass[NVDEV_ENGINE_PERFMON] =  nva3_perfmon_oclass;
 		break;
 	case 0xaf:
 		device->cname = "MCP89";
@@ -399,22 +431,25 @@ nv50_identify(struct nouveau_device *device)
 		device->oclass[NVDEV_SUBDEV_THERM  ] = &nva3_therm_oclass;
 		device->oclass[NVDEV_SUBDEV_MXM    ] = &nv50_mxm_oclass;
 		device->oclass[NVDEV_SUBDEV_DEVINIT] = &nva3_devinit_oclass;
-		device->oclass[NVDEV_SUBDEV_MC     ] = &nv98_mc_oclass;
-		device->oclass[NVDEV_SUBDEV_BUS    ] = &nv50_bus_oclass;
+		device->oclass[NVDEV_SUBDEV_MC     ] =  nv98_mc_oclass;
+		device->oclass[NVDEV_SUBDEV_BUS    ] =  nv94_bus_oclass;
 		device->oclass[NVDEV_SUBDEV_TIMER  ] = &nv04_timer_oclass;
-		device->oclass[NVDEV_SUBDEV_FB     ] = &nv50_fb_oclass;
+		device->oclass[NVDEV_SUBDEV_FB     ] =  nvaf_fb_oclass;
 		device->oclass[NVDEV_SUBDEV_INSTMEM] = &nv50_instmem_oclass;
 		device->oclass[NVDEV_SUBDEV_VM     ] = &nv50_vmmgr_oclass;
 		device->oclass[NVDEV_SUBDEV_BAR    ] = &nv50_bar_oclass;
+		device->oclass[NVDEV_SUBDEV_PWR    ] = &nva3_pwr_oclass;
+		device->oclass[NVDEV_SUBDEV_VOLT   ] = &nv40_volt_oclass;
 		device->oclass[NVDEV_ENGINE_DMAOBJ ] = &nv50_dmaeng_oclass;
-		device->oclass[NVDEV_ENGINE_FIFO   ] = &nv84_fifo_oclass;
-		device->oclass[NVDEV_ENGINE_SW     ] = &nv50_software_oclass;
+		device->oclass[NVDEV_ENGINE_FIFO   ] =  nv84_fifo_oclass;
+		device->oclass[NVDEV_ENGINE_SW     ] =  nv50_software_oclass;
 		device->oclass[NVDEV_ENGINE_GR     ] = &nv50_graph_oclass;
 		device->oclass[NVDEV_ENGINE_VP     ] = &nv98_vp_oclass;
 		device->oclass[NVDEV_ENGINE_BSP    ] = &nv98_bsp_oclass;
 		device->oclass[NVDEV_ENGINE_PPP    ] = &nv98_ppp_oclass;
 		device->oclass[NVDEV_ENGINE_COPY0  ] = &nva3_copy_oclass;
 		device->oclass[NVDEV_ENGINE_DISP   ] = &nva3_disp_oclass;
+		device->oclass[NVDEV_ENGINE_PERFMON] =  nva3_perfmon_oclass;
 		break;
 	default:
 		nv_fatal(device, "unknown Tesla chipset\n");
diff --git a/drivers/gpu/drm/nouveau/core/engine/device/nvc0.c b/drivers/gpu/drm/nouveau/core/engine/device/nvc0.c
index 418f51f50d7a..8d06eef2b9ee 100644
--- a/drivers/gpu/drm/nouveau/core/engine/device/nvc0.c
+++ b/drivers/gpu/drm/nouveau/core/engine/device/nvc0.c
@@ -38,6 +38,8 @@
 #include <subdev/instmem.h>
 #include <subdev/vm.h>
 #include <subdev/bar.h>
+#include <subdev/pwr.h>
+#include <subdev/volt.h>
 
 #include <engine/device.h>
 #include <engine/dmaobj.h>
@@ -49,6 +51,7 @@
 #include <engine/ppp.h>
 #include <engine/copy.h>
 #include <engine/disp.h>
+#include <engine/perfmon.h>
 
 int
 nvc0_identify(struct nouveau_device *device)
@@ -63,18 +66,20 @@ nvc0_identify(struct nouveau_device *device)
 		device->oclass[NVDEV_SUBDEV_THERM  ] = &nva3_therm_oclass;
 		device->oclass[NVDEV_SUBDEV_MXM    ] = &nv50_mxm_oclass;
 		device->oclass[NVDEV_SUBDEV_DEVINIT] = &nvc0_devinit_oclass;
-		device->oclass[NVDEV_SUBDEV_MC     ] = &nvc0_mc_oclass;
-		device->oclass[NVDEV_SUBDEV_BUS    ] = &nvc0_bus_oclass;
+		device->oclass[NVDEV_SUBDEV_MC     ] =  nvc0_mc_oclass;
+		device->oclass[NVDEV_SUBDEV_BUS    ] =  nvc0_bus_oclass;
 		device->oclass[NVDEV_SUBDEV_TIMER  ] = &nv04_timer_oclass;
-		device->oclass[NVDEV_SUBDEV_FB     ] = &nvc0_fb_oclass;
+		device->oclass[NVDEV_SUBDEV_FB     ] =  nvc0_fb_oclass;
 		device->oclass[NVDEV_SUBDEV_LTCG   ] = &nvc0_ltcg_oclass;
 		device->oclass[NVDEV_SUBDEV_IBUS   ] = &nvc0_ibus_oclass;
 		device->oclass[NVDEV_SUBDEV_INSTMEM] = &nv50_instmem_oclass;
 		device->oclass[NVDEV_SUBDEV_VM     ] = &nvc0_vmmgr_oclass;
 		device->oclass[NVDEV_SUBDEV_BAR    ] = &nvc0_bar_oclass;
+		device->oclass[NVDEV_SUBDEV_PWR    ] = &nvc0_pwr_oclass;
+		device->oclass[NVDEV_SUBDEV_VOLT   ] = &nv40_volt_oclass;
 		device->oclass[NVDEV_ENGINE_DMAOBJ ] = &nvc0_dmaeng_oclass;
-		device->oclass[NVDEV_ENGINE_FIFO   ] = &nvc0_fifo_oclass;
-		device->oclass[NVDEV_ENGINE_SW     ] = &nvc0_software_oclass;
+		device->oclass[NVDEV_ENGINE_FIFO   ] =  nvc0_fifo_oclass;
+		device->oclass[NVDEV_ENGINE_SW     ] =  nvc0_software_oclass;
 		device->oclass[NVDEV_ENGINE_GR     ] =  nvc0_graph_oclass;
 		device->oclass[NVDEV_ENGINE_VP     ] = &nvc0_vp_oclass;
 		device->oclass[NVDEV_ENGINE_BSP    ] = &nvc0_bsp_oclass;
@@ -82,6 +87,7 @@ nvc0_identify(struct nouveau_device *device)
 		device->oclass[NVDEV_ENGINE_COPY0  ] = &nvc0_copy0_oclass;
 		device->oclass[NVDEV_ENGINE_COPY1  ] = &nvc0_copy1_oclass;
 		device->oclass[NVDEV_ENGINE_DISP   ] = &nva3_disp_oclass;
+		device->oclass[NVDEV_ENGINE_PERFMON] = &nvc0_perfmon_oclass;
 		break;
 	case 0xc4:
 		device->cname = "GF104";
@@ -92,18 +98,20 @@ nvc0_identify(struct nouveau_device *device)
 		device->oclass[NVDEV_SUBDEV_THERM  ] = &nva3_therm_oclass;
 		device->oclass[NVDEV_SUBDEV_MXM    ] = &nv50_mxm_oclass;
 		device->oclass[NVDEV_SUBDEV_DEVINIT] = &nvc0_devinit_oclass;
-		device->oclass[NVDEV_SUBDEV_MC     ] = &nvc0_mc_oclass;
-		device->oclass[NVDEV_SUBDEV_BUS    ] = &nvc0_bus_oclass;
+		device->oclass[NVDEV_SUBDEV_MC     ] =  nvc0_mc_oclass;
+		device->oclass[NVDEV_SUBDEV_BUS    ] =  nvc0_bus_oclass;
 		device->oclass[NVDEV_SUBDEV_TIMER  ] = &nv04_timer_oclass;
-		device->oclass[NVDEV_SUBDEV_FB     ] = &nvc0_fb_oclass;
+		device->oclass[NVDEV_SUBDEV_FB     ] =  nvc0_fb_oclass;
 		device->oclass[NVDEV_SUBDEV_LTCG   ] = &nvc0_ltcg_oclass;
 		device->oclass[NVDEV_SUBDEV_IBUS   ] = &nvc0_ibus_oclass;
 		device->oclass[NVDEV_SUBDEV_INSTMEM] = &nv50_instmem_oclass;
 		device->oclass[NVDEV_SUBDEV_VM     ] = &nvc0_vmmgr_oclass;
 		device->oclass[NVDEV_SUBDEV_BAR    ] = &nvc0_bar_oclass;
+		device->oclass[NVDEV_SUBDEV_PWR    ] = &nvc0_pwr_oclass;
+		device->oclass[NVDEV_SUBDEV_VOLT   ] = &nv40_volt_oclass;
 		device->oclass[NVDEV_ENGINE_DMAOBJ ] = &nvc0_dmaeng_oclass;
-		device->oclass[NVDEV_ENGINE_FIFO   ] = &nvc0_fifo_oclass;
-		device->oclass[NVDEV_ENGINE_SW     ] = &nvc0_software_oclass;
+		device->oclass[NVDEV_ENGINE_FIFO   ] =  nvc0_fifo_oclass;
+		device->oclass[NVDEV_ENGINE_SW     ] =  nvc0_software_oclass;
 		device->oclass[NVDEV_ENGINE_GR     ] =  nvc3_graph_oclass;
 		device->oclass[NVDEV_ENGINE_VP     ] = &nvc0_vp_oclass;
 		device->oclass[NVDEV_ENGINE_BSP    ] = &nvc0_bsp_oclass;
@@ -111,6 +119,7 @@ nvc0_identify(struct nouveau_device *device)
 		device->oclass[NVDEV_ENGINE_COPY0  ] = &nvc0_copy0_oclass;
 		device->oclass[NVDEV_ENGINE_COPY1  ] = &nvc0_copy1_oclass;
 		device->oclass[NVDEV_ENGINE_DISP   ] = &nva3_disp_oclass;
+		device->oclass[NVDEV_ENGINE_PERFMON] = &nvc0_perfmon_oclass;
 		break;
 	case 0xc3:
 		device->cname = "GF106";
@@ -121,24 +130,27 @@ nvc0_identify(struct nouveau_device *device)
 		device->oclass[NVDEV_SUBDEV_THERM  ] = &nva3_therm_oclass;
 		device->oclass[NVDEV_SUBDEV_MXM    ] = &nv50_mxm_oclass;
 		device->oclass[NVDEV_SUBDEV_DEVINIT] = &nvc0_devinit_oclass;
-		device->oclass[NVDEV_SUBDEV_MC     ] = &nvc0_mc_oclass;
-		device->oclass[NVDEV_SUBDEV_BUS    ] = &nvc0_bus_oclass;
+		device->oclass[NVDEV_SUBDEV_MC     ] =  nvc3_mc_oclass;
+		device->oclass[NVDEV_SUBDEV_BUS    ] =  nvc0_bus_oclass;
 		device->oclass[NVDEV_SUBDEV_TIMER  ] = &nv04_timer_oclass;
-		device->oclass[NVDEV_SUBDEV_FB     ] = &nvc0_fb_oclass;
+		device->oclass[NVDEV_SUBDEV_FB     ] =  nvc0_fb_oclass;
 		device->oclass[NVDEV_SUBDEV_LTCG   ] = &nvc0_ltcg_oclass;
 		device->oclass[NVDEV_SUBDEV_IBUS   ] = &nvc0_ibus_oclass;
 		device->oclass[NVDEV_SUBDEV_INSTMEM] = &nv50_instmem_oclass;
 		device->oclass[NVDEV_SUBDEV_VM     ] = &nvc0_vmmgr_oclass;
 		device->oclass[NVDEV_SUBDEV_BAR    ] = &nvc0_bar_oclass;
+		device->oclass[NVDEV_SUBDEV_PWR    ] = &nvc0_pwr_oclass;
+		device->oclass[NVDEV_SUBDEV_VOLT   ] = &nv40_volt_oclass;
 		device->oclass[NVDEV_ENGINE_DMAOBJ ] = &nvc0_dmaeng_oclass;
-		device->oclass[NVDEV_ENGINE_FIFO   ] = &nvc0_fifo_oclass;
-		device->oclass[NVDEV_ENGINE_SW     ] = &nvc0_software_oclass;
+		device->oclass[NVDEV_ENGINE_FIFO   ] =  nvc0_fifo_oclass;
+		device->oclass[NVDEV_ENGINE_SW     ] =  nvc0_software_oclass;
 		device->oclass[NVDEV_ENGINE_GR     ] =  nvc3_graph_oclass;
 		device->oclass[NVDEV_ENGINE_VP     ] = &nvc0_vp_oclass;
 		device->oclass[NVDEV_ENGINE_BSP    ] = &nvc0_bsp_oclass;
 		device->oclass[NVDEV_ENGINE_PPP    ] = &nvc0_ppp_oclass;
 		device->oclass[NVDEV_ENGINE_COPY0  ] = &nvc0_copy0_oclass;
 		device->oclass[NVDEV_ENGINE_DISP   ] = &nva3_disp_oclass;
+		device->oclass[NVDEV_ENGINE_PERFMON] = &nvc0_perfmon_oclass;
 		break;
 	case 0xce:
 		device->cname = "GF114";
@@ -149,18 +161,20 @@ nvc0_identify(struct nouveau_device *device)
 		device->oclass[NVDEV_SUBDEV_THERM  ] = &nva3_therm_oclass;
 		device->oclass[NVDEV_SUBDEV_MXM    ] = &nv50_mxm_oclass;
 		device->oclass[NVDEV_SUBDEV_DEVINIT] = &nvc0_devinit_oclass;
-		device->oclass[NVDEV_SUBDEV_MC     ] = &nvc0_mc_oclass;
-		device->oclass[NVDEV_SUBDEV_BUS    ] = &nvc0_bus_oclass;
+		device->oclass[NVDEV_SUBDEV_MC     ] =  nvc3_mc_oclass;
+		device->oclass[NVDEV_SUBDEV_BUS    ] =  nvc0_bus_oclass;
 		device->oclass[NVDEV_SUBDEV_TIMER  ] = &nv04_timer_oclass;
-		device->oclass[NVDEV_SUBDEV_FB     ] = &nvc0_fb_oclass;
+		device->oclass[NVDEV_SUBDEV_FB     ] =  nvc0_fb_oclass;
 		device->oclass[NVDEV_SUBDEV_LTCG   ] = &nvc0_ltcg_oclass;
 		device->oclass[NVDEV_SUBDEV_IBUS   ] = &nvc0_ibus_oclass;
 		device->oclass[NVDEV_SUBDEV_INSTMEM] = &nv50_instmem_oclass;
 		device->oclass[NVDEV_SUBDEV_VM     ] = &nvc0_vmmgr_oclass;
 		device->oclass[NVDEV_SUBDEV_BAR    ] = &nvc0_bar_oclass;
+		device->oclass[NVDEV_SUBDEV_PWR    ] = &nvc0_pwr_oclass;
+		device->oclass[NVDEV_SUBDEV_VOLT   ] = &nv40_volt_oclass;
 		device->oclass[NVDEV_ENGINE_DMAOBJ ] = &nvc0_dmaeng_oclass;
-		device->oclass[NVDEV_ENGINE_FIFO   ] = &nvc0_fifo_oclass;
-		device->oclass[NVDEV_ENGINE_SW     ] = &nvc0_software_oclass;
+		device->oclass[NVDEV_ENGINE_FIFO   ] =  nvc0_fifo_oclass;
+		device->oclass[NVDEV_ENGINE_SW     ] =  nvc0_software_oclass;
 		device->oclass[NVDEV_ENGINE_GR     ] =  nvc3_graph_oclass;
 		device->oclass[NVDEV_ENGINE_VP     ] = &nvc0_vp_oclass;
 		device->oclass[NVDEV_ENGINE_BSP    ] = &nvc0_bsp_oclass;
@@ -168,6 +182,7 @@ nvc0_identify(struct nouveau_device *device)
 		device->oclass[NVDEV_ENGINE_COPY0  ] = &nvc0_copy0_oclass;
 		device->oclass[NVDEV_ENGINE_COPY1  ] = &nvc0_copy1_oclass;
 		device->oclass[NVDEV_ENGINE_DISP   ] = &nva3_disp_oclass;
+		device->oclass[NVDEV_ENGINE_PERFMON] = &nvc0_perfmon_oclass;
 		break;
 	case 0xcf:
 		device->cname = "GF116";
@@ -178,18 +193,20 @@ nvc0_identify(struct nouveau_device *device)
 		device->oclass[NVDEV_SUBDEV_THERM  ] = &nva3_therm_oclass;
 		device->oclass[NVDEV_SUBDEV_MXM    ] = &nv50_mxm_oclass;
 		device->oclass[NVDEV_SUBDEV_DEVINIT] = &nvc0_devinit_oclass;
-		device->oclass[NVDEV_SUBDEV_MC     ] = &nvc0_mc_oclass;
-		device->oclass[NVDEV_SUBDEV_BUS    ] = &nvc0_bus_oclass;
+		device->oclass[NVDEV_SUBDEV_MC     ] =  nvc3_mc_oclass;
+		device->oclass[NVDEV_SUBDEV_BUS    ] =  nvc0_bus_oclass;
 		device->oclass[NVDEV_SUBDEV_TIMER  ] = &nv04_timer_oclass;
-		device->oclass[NVDEV_SUBDEV_FB     ] = &nvc0_fb_oclass;
+		device->oclass[NVDEV_SUBDEV_FB     ] =  nvc0_fb_oclass;
 		device->oclass[NVDEV_SUBDEV_LTCG   ] = &nvc0_ltcg_oclass;
 		device->oclass[NVDEV_SUBDEV_IBUS   ] = &nvc0_ibus_oclass;
 		device->oclass[NVDEV_SUBDEV_INSTMEM] = &nv50_instmem_oclass;
 		device->oclass[NVDEV_SUBDEV_VM     ] = &nvc0_vmmgr_oclass;
 		device->oclass[NVDEV_SUBDEV_BAR    ] = &nvc0_bar_oclass;
+		device->oclass[NVDEV_SUBDEV_PWR    ] = &nvc0_pwr_oclass;
+		device->oclass[NVDEV_SUBDEV_VOLT   ] = &nv40_volt_oclass;
 		device->oclass[NVDEV_ENGINE_DMAOBJ ] = &nvc0_dmaeng_oclass;
-		device->oclass[NVDEV_ENGINE_FIFO   ] = &nvc0_fifo_oclass;
-		device->oclass[NVDEV_ENGINE_SW     ] = &nvc0_software_oclass;
+		device->oclass[NVDEV_ENGINE_FIFO   ] =  nvc0_fifo_oclass;
+		device->oclass[NVDEV_ENGINE_SW     ] =  nvc0_software_oclass;
 		device->oclass[NVDEV_ENGINE_GR     ] =  nvc3_graph_oclass;
 		device->oclass[NVDEV_ENGINE_VP     ] = &nvc0_vp_oclass;
 		device->oclass[NVDEV_ENGINE_BSP    ] = &nvc0_bsp_oclass;
@@ -197,6 +214,7 @@ nvc0_identify(struct nouveau_device *device)
 		device->oclass[NVDEV_ENGINE_COPY0  ] = &nvc0_copy0_oclass;
 		device->oclass[NVDEV_ENGINE_COPY1  ] = &nvc0_copy1_oclass;
 		device->oclass[NVDEV_ENGINE_DISP   ] = &nva3_disp_oclass;
+		device->oclass[NVDEV_ENGINE_PERFMON] = &nvc0_perfmon_oclass;
 		break;
 	case 0xc1:
 		device->cname = "GF108";
@@ -207,24 +225,27 @@ nvc0_identify(struct nouveau_device *device)
 		device->oclass[NVDEV_SUBDEV_THERM  ] = &nva3_therm_oclass;
 		device->oclass[NVDEV_SUBDEV_MXM    ] = &nv50_mxm_oclass;
 		device->oclass[NVDEV_SUBDEV_DEVINIT] = &nvc0_devinit_oclass;
-		device->oclass[NVDEV_SUBDEV_MC     ] = &nvc0_mc_oclass;
-		device->oclass[NVDEV_SUBDEV_BUS    ] = &nvc0_bus_oclass;
+		device->oclass[NVDEV_SUBDEV_MC     ] =  nvc3_mc_oclass;
+		device->oclass[NVDEV_SUBDEV_BUS    ] =  nvc0_bus_oclass;
 		device->oclass[NVDEV_SUBDEV_TIMER  ] = &nv04_timer_oclass;
-		device->oclass[NVDEV_SUBDEV_FB     ] = &nvc0_fb_oclass;
+		device->oclass[NVDEV_SUBDEV_FB     ] =  nvc0_fb_oclass;
 		device->oclass[NVDEV_SUBDEV_LTCG   ] = &nvc0_ltcg_oclass;
 		device->oclass[NVDEV_SUBDEV_IBUS   ] = &nvc0_ibus_oclass;
 		device->oclass[NVDEV_SUBDEV_INSTMEM] = &nv50_instmem_oclass;
 		device->oclass[NVDEV_SUBDEV_VM     ] = &nvc0_vmmgr_oclass;
 		device->oclass[NVDEV_SUBDEV_BAR    ] = &nvc0_bar_oclass;
+		device->oclass[NVDEV_SUBDEV_PWR    ] = &nvc0_pwr_oclass;
+		device->oclass[NVDEV_SUBDEV_VOLT   ] = &nv40_volt_oclass;
 		device->oclass[NVDEV_ENGINE_DMAOBJ ] = &nvc0_dmaeng_oclass;
-		device->oclass[NVDEV_ENGINE_FIFO   ] = &nvc0_fifo_oclass;
-		device->oclass[NVDEV_ENGINE_SW     ] = &nvc0_software_oclass;
+		device->oclass[NVDEV_ENGINE_FIFO   ] =  nvc0_fifo_oclass;
+		device->oclass[NVDEV_ENGINE_SW     ] =  nvc0_software_oclass;
 		device->oclass[NVDEV_ENGINE_GR     ] =  nvc1_graph_oclass;
 		device->oclass[NVDEV_ENGINE_VP     ] = &nvc0_vp_oclass;
 		device->oclass[NVDEV_ENGINE_BSP    ] = &nvc0_bsp_oclass;
 		device->oclass[NVDEV_ENGINE_PPP    ] = &nvc0_ppp_oclass;
 		device->oclass[NVDEV_ENGINE_COPY0  ] = &nvc0_copy0_oclass;
 		device->oclass[NVDEV_ENGINE_DISP   ] = &nva3_disp_oclass;
+		device->oclass[NVDEV_ENGINE_PERFMON] = &nvc0_perfmon_oclass;
 		break;
 	case 0xc8:
 		device->cname = "GF110";
@@ -235,18 +256,20 @@ nvc0_identify(struct nouveau_device *device)
 		device->oclass[NVDEV_SUBDEV_THERM  ] = &nva3_therm_oclass;
 		device->oclass[NVDEV_SUBDEV_MXM    ] = &nv50_mxm_oclass;
 		device->oclass[NVDEV_SUBDEV_DEVINIT] = &nvc0_devinit_oclass;
-		device->oclass[NVDEV_SUBDEV_MC     ] = &nvc0_mc_oclass;
-		device->oclass[NVDEV_SUBDEV_BUS    ] = &nvc0_bus_oclass;
+		device->oclass[NVDEV_SUBDEV_MC     ] =  nvc0_mc_oclass;
+		device->oclass[NVDEV_SUBDEV_BUS    ] =  nvc0_bus_oclass;
 		device->oclass[NVDEV_SUBDEV_TIMER  ] = &nv04_timer_oclass;
-		device->oclass[NVDEV_SUBDEV_FB     ] = &nvc0_fb_oclass;
+		device->oclass[NVDEV_SUBDEV_FB     ] =  nvc0_fb_oclass;
 		device->oclass[NVDEV_SUBDEV_LTCG   ] = &nvc0_ltcg_oclass;
 		device->oclass[NVDEV_SUBDEV_IBUS   ] = &nvc0_ibus_oclass;
 		device->oclass[NVDEV_SUBDEV_INSTMEM] = &nv50_instmem_oclass;
 		device->oclass[NVDEV_SUBDEV_VM     ] = &nvc0_vmmgr_oclass;
 		device->oclass[NVDEV_SUBDEV_BAR    ] = &nvc0_bar_oclass;
+		device->oclass[NVDEV_SUBDEV_PWR    ] = &nvc0_pwr_oclass;
+		device->oclass[NVDEV_SUBDEV_VOLT   ] = &nv40_volt_oclass;
 		device->oclass[NVDEV_ENGINE_DMAOBJ ] = &nvc0_dmaeng_oclass;
-		device->oclass[NVDEV_ENGINE_FIFO   ] = &nvc0_fifo_oclass;
-		device->oclass[NVDEV_ENGINE_SW     ] = &nvc0_software_oclass;
+		device->oclass[NVDEV_ENGINE_FIFO   ] =  nvc0_fifo_oclass;
+		device->oclass[NVDEV_ENGINE_SW     ] =  nvc0_software_oclass;
 		device->oclass[NVDEV_ENGINE_GR     ] =  nvc8_graph_oclass;
 		device->oclass[NVDEV_ENGINE_VP     ] = &nvc0_vp_oclass;
 		device->oclass[NVDEV_ENGINE_BSP    ] = &nvc0_bsp_oclass;
@@ -254,6 +277,7 @@ nvc0_identify(struct nouveau_device *device)
 		device->oclass[NVDEV_ENGINE_COPY0  ] = &nvc0_copy0_oclass;
 		device->oclass[NVDEV_ENGINE_COPY1  ] = &nvc0_copy1_oclass;
 		device->oclass[NVDEV_ENGINE_DISP   ] = &nva3_disp_oclass;
+		device->oclass[NVDEV_ENGINE_PERFMON] = &nvc0_perfmon_oclass;
 		break;
 	case 0xd9:
 		device->cname = "GF119";
@@ -264,24 +288,27 @@ nvc0_identify(struct nouveau_device *device)
 		device->oclass[NVDEV_SUBDEV_THERM  ] = &nvd0_therm_oclass;
 		device->oclass[NVDEV_SUBDEV_MXM    ] = &nv50_mxm_oclass;
 		device->oclass[NVDEV_SUBDEV_DEVINIT] = &nvc0_devinit_oclass;
-		device->oclass[NVDEV_SUBDEV_MC     ] = &nvc0_mc_oclass;
-		device->oclass[NVDEV_SUBDEV_BUS    ] = &nvc0_bus_oclass;
+		device->oclass[NVDEV_SUBDEV_MC     ] =  nvc3_mc_oclass;
+		device->oclass[NVDEV_SUBDEV_BUS    ] =  nvc0_bus_oclass;
 		device->oclass[NVDEV_SUBDEV_TIMER  ] = &nv04_timer_oclass;
-		device->oclass[NVDEV_SUBDEV_FB     ] = &nvc0_fb_oclass;
+		device->oclass[NVDEV_SUBDEV_FB     ] =  nvc0_fb_oclass;
 		device->oclass[NVDEV_SUBDEV_LTCG   ] = &nvc0_ltcg_oclass;
 		device->oclass[NVDEV_SUBDEV_IBUS   ] = &nvc0_ibus_oclass;
 		device->oclass[NVDEV_SUBDEV_INSTMEM] = &nv50_instmem_oclass;
 		device->oclass[NVDEV_SUBDEV_VM     ] = &nvc0_vmmgr_oclass;
 		device->oclass[NVDEV_SUBDEV_BAR    ] = &nvc0_bar_oclass;
+		device->oclass[NVDEV_SUBDEV_PWR    ] = &nvd0_pwr_oclass;
+		device->oclass[NVDEV_SUBDEV_VOLT   ] = &nv40_volt_oclass;
 		device->oclass[NVDEV_ENGINE_DMAOBJ ] = &nvd0_dmaeng_oclass;
-		device->oclass[NVDEV_ENGINE_FIFO   ] = &nvc0_fifo_oclass;
-		device->oclass[NVDEV_ENGINE_SW     ] = &nvc0_software_oclass;
+		device->oclass[NVDEV_ENGINE_FIFO   ] =  nvc0_fifo_oclass;
+		device->oclass[NVDEV_ENGINE_SW     ] =  nvc0_software_oclass;
 		device->oclass[NVDEV_ENGINE_GR     ] =  nvd9_graph_oclass;
 		device->oclass[NVDEV_ENGINE_VP     ] = &nvc0_vp_oclass;
 		device->oclass[NVDEV_ENGINE_BSP    ] = &nvc0_bsp_oclass;
 		device->oclass[NVDEV_ENGINE_PPP    ] = &nvc0_ppp_oclass;
 		device->oclass[NVDEV_ENGINE_COPY0  ] = &nvc0_copy0_oclass;
 		device->oclass[NVDEV_ENGINE_DISP   ] = &nvd0_disp_oclass;
+		device->oclass[NVDEV_ENGINE_PERFMON] = &nvc0_perfmon_oclass;
 		break;
 	case 0xd7:
 		device->cname = "GF117";
@@ -292,24 +319,25 @@ nvc0_identify(struct nouveau_device *device)
 		device->oclass[NVDEV_SUBDEV_THERM  ] = &nvd0_therm_oclass;
 		device->oclass[NVDEV_SUBDEV_MXM    ] = &nv50_mxm_oclass;
 		device->oclass[NVDEV_SUBDEV_DEVINIT] = &nvc0_devinit_oclass;
-		device->oclass[NVDEV_SUBDEV_MC     ] = &nvc0_mc_oclass;
-		device->oclass[NVDEV_SUBDEV_BUS    ] = &nvc0_bus_oclass;
+		device->oclass[NVDEV_SUBDEV_MC     ] =  nvc3_mc_oclass;
+		device->oclass[NVDEV_SUBDEV_BUS    ] =  nvc0_bus_oclass;
 		device->oclass[NVDEV_SUBDEV_TIMER  ] = &nv04_timer_oclass;
-		device->oclass[NVDEV_SUBDEV_FB     ] = &nvc0_fb_oclass;
+		device->oclass[NVDEV_SUBDEV_FB     ] =  nvc0_fb_oclass;
 		device->oclass[NVDEV_SUBDEV_LTCG   ] = &nvc0_ltcg_oclass;
 		device->oclass[NVDEV_SUBDEV_IBUS   ] = &nvc0_ibus_oclass;
 		device->oclass[NVDEV_SUBDEV_INSTMEM] = &nv50_instmem_oclass;
 		device->oclass[NVDEV_SUBDEV_VM     ] = &nvc0_vmmgr_oclass;
 		device->oclass[NVDEV_SUBDEV_BAR    ] = &nvc0_bar_oclass;
 		device->oclass[NVDEV_ENGINE_DMAOBJ ] = &nvd0_dmaeng_oclass;
-		device->oclass[NVDEV_ENGINE_FIFO   ] = &nvc0_fifo_oclass;
-		device->oclass[NVDEV_ENGINE_SW     ] = &nvc0_software_oclass;
+		device->oclass[NVDEV_ENGINE_FIFO   ] =  nvc0_fifo_oclass;
+		device->oclass[NVDEV_ENGINE_SW     ] =  nvc0_software_oclass;
 		device->oclass[NVDEV_ENGINE_GR     ] =  nvd7_graph_oclass;
 		device->oclass[NVDEV_ENGINE_VP     ] = &nvc0_vp_oclass;
 		device->oclass[NVDEV_ENGINE_BSP    ] = &nvc0_bsp_oclass;
 		device->oclass[NVDEV_ENGINE_PPP    ] = &nvc0_ppp_oclass;
 		device->oclass[NVDEV_ENGINE_COPY0  ] = &nvc0_copy0_oclass;
 		device->oclass[NVDEV_ENGINE_DISP   ] = &nvd0_disp_oclass;
+		device->oclass[NVDEV_ENGINE_PERFMON] = &nvc0_perfmon_oclass;
 		break;
 	default:
 		nv_fatal(device, "unknown Fermi chipset\n");
diff --git a/drivers/gpu/drm/nouveau/core/engine/device/nve0.c b/drivers/gpu/drm/nouveau/core/engine/device/nve0.c
index 7aca1877add4..3900104976fc 100644
--- a/drivers/gpu/drm/nouveau/core/engine/device/nve0.c
+++ b/drivers/gpu/drm/nouveau/core/engine/device/nve0.c
@@ -38,6 +38,8 @@
 #include <subdev/instmem.h>
 #include <subdev/vm.h>
 #include <subdev/bar.h>
+#include <subdev/pwr.h>
+#include <subdev/volt.h>
 
 #include <engine/device.h>
 #include <engine/dmaobj.h>
@@ -49,6 +51,7 @@
 #include <engine/bsp.h>
 #include <engine/vp.h>
 #include <engine/ppp.h>
+#include <engine/perfmon.h>
 
 int
 nve0_identify(struct nouveau_device *device)
@@ -59,22 +62,24 @@ nve0_identify(struct nouveau_device *device)
 		device->oclass[NVDEV_SUBDEV_VBIOS  ] = &nouveau_bios_oclass;
 		device->oclass[NVDEV_SUBDEV_GPIO   ] = &nve0_gpio_oclass;
 		device->oclass[NVDEV_SUBDEV_I2C    ] = &nvd0_i2c_oclass;
-		device->oclass[NVDEV_SUBDEV_CLOCK  ] = &nvc0_clock_oclass;
+		device->oclass[NVDEV_SUBDEV_CLOCK  ] = &nve0_clock_oclass;
 		device->oclass[NVDEV_SUBDEV_THERM  ] = &nvd0_therm_oclass;
 		device->oclass[NVDEV_SUBDEV_MXM    ] = &nv50_mxm_oclass;
 		device->oclass[NVDEV_SUBDEV_DEVINIT] = &nvc0_devinit_oclass;
-		device->oclass[NVDEV_SUBDEV_MC     ] = &nvc0_mc_oclass;
-		device->oclass[NVDEV_SUBDEV_BUS    ] = &nvc0_bus_oclass;
+		device->oclass[NVDEV_SUBDEV_MC     ] =  nvc3_mc_oclass;
+		device->oclass[NVDEV_SUBDEV_BUS    ] =  nvc0_bus_oclass;
 		device->oclass[NVDEV_SUBDEV_TIMER  ] = &nv04_timer_oclass;
-		device->oclass[NVDEV_SUBDEV_FB     ] = &nvc0_fb_oclass;
+		device->oclass[NVDEV_SUBDEV_FB     ] =  nve0_fb_oclass;
 		device->oclass[NVDEV_SUBDEV_LTCG   ] = &nvc0_ltcg_oclass;
 		device->oclass[NVDEV_SUBDEV_IBUS   ] = &nve0_ibus_oclass;
 		device->oclass[NVDEV_SUBDEV_INSTMEM] = &nv50_instmem_oclass;
 		device->oclass[NVDEV_SUBDEV_VM     ] = &nvc0_vmmgr_oclass;
 		device->oclass[NVDEV_SUBDEV_BAR    ] = &nvc0_bar_oclass;
+		device->oclass[NVDEV_SUBDEV_PWR    ] = &nvd0_pwr_oclass;
+		device->oclass[NVDEV_SUBDEV_VOLT   ] = &nv40_volt_oclass;
 		device->oclass[NVDEV_ENGINE_DMAOBJ ] = &nvd0_dmaeng_oclass;
-		device->oclass[NVDEV_ENGINE_FIFO   ] = &nve0_fifo_oclass;
-		device->oclass[NVDEV_ENGINE_SW     ] = &nvc0_software_oclass;
+		device->oclass[NVDEV_ENGINE_FIFO   ] =  nve0_fifo_oclass;
+		device->oclass[NVDEV_ENGINE_SW     ] =  nvc0_software_oclass;
 		device->oclass[NVDEV_ENGINE_GR     ] =  nve4_graph_oclass;
 		device->oclass[NVDEV_ENGINE_DISP   ] = &nve0_disp_oclass;
 		device->oclass[NVDEV_ENGINE_COPY0  ] = &nve0_copy0_oclass;
@@ -83,28 +88,31 @@ nve0_identify(struct nouveau_device *device)
 		device->oclass[NVDEV_ENGINE_BSP    ] = &nve0_bsp_oclass;
 		device->oclass[NVDEV_ENGINE_VP     ] = &nve0_vp_oclass;
 		device->oclass[NVDEV_ENGINE_PPP    ] = &nvc0_ppp_oclass;
+		device->oclass[NVDEV_ENGINE_PERFMON] = &nve0_perfmon_oclass;
 		break;
 	case 0xe7:
 		device->cname = "GK107";
 		device->oclass[NVDEV_SUBDEV_VBIOS  ] = &nouveau_bios_oclass;
 		device->oclass[NVDEV_SUBDEV_GPIO   ] = &nve0_gpio_oclass;
 		device->oclass[NVDEV_SUBDEV_I2C    ] = &nvd0_i2c_oclass;
-		device->oclass[NVDEV_SUBDEV_CLOCK  ] = &nvc0_clock_oclass;
+		device->oclass[NVDEV_SUBDEV_CLOCK  ] = &nve0_clock_oclass;
 		device->oclass[NVDEV_SUBDEV_THERM  ] = &nvd0_therm_oclass;
 		device->oclass[NVDEV_SUBDEV_MXM    ] = &nv50_mxm_oclass;
 		device->oclass[NVDEV_SUBDEV_DEVINIT] = &nvc0_devinit_oclass;
-		device->oclass[NVDEV_SUBDEV_MC     ] = &nvc0_mc_oclass;
-		device->oclass[NVDEV_SUBDEV_BUS    ] = &nvc0_bus_oclass;
+		device->oclass[NVDEV_SUBDEV_MC     ] =  nvc3_mc_oclass;
+		device->oclass[NVDEV_SUBDEV_BUS    ] =  nvc0_bus_oclass;
 		device->oclass[NVDEV_SUBDEV_TIMER  ] = &nv04_timer_oclass;
-		device->oclass[NVDEV_SUBDEV_FB     ] = &nvc0_fb_oclass;
+		device->oclass[NVDEV_SUBDEV_FB     ] =  nve0_fb_oclass;
 		device->oclass[NVDEV_SUBDEV_LTCG   ] = &nvc0_ltcg_oclass;
 		device->oclass[NVDEV_SUBDEV_IBUS   ] = &nve0_ibus_oclass;
 		device->oclass[NVDEV_SUBDEV_INSTMEM] = &nv50_instmem_oclass;
 		device->oclass[NVDEV_SUBDEV_VM     ] = &nvc0_vmmgr_oclass;
 		device->oclass[NVDEV_SUBDEV_BAR    ] = &nvc0_bar_oclass;
+		device->oclass[NVDEV_SUBDEV_PWR    ] = &nvd0_pwr_oclass;
+		device->oclass[NVDEV_SUBDEV_VOLT   ] = &nv40_volt_oclass;
 		device->oclass[NVDEV_ENGINE_DMAOBJ ] = &nvd0_dmaeng_oclass;
-		device->oclass[NVDEV_ENGINE_FIFO   ] = &nve0_fifo_oclass;
-		device->oclass[NVDEV_ENGINE_SW     ] = &nvc0_software_oclass;
+		device->oclass[NVDEV_ENGINE_FIFO   ] =  nve0_fifo_oclass;
+		device->oclass[NVDEV_ENGINE_SW     ] =  nvc0_software_oclass;
 		device->oclass[NVDEV_ENGINE_GR     ] =  nve4_graph_oclass;
 		device->oclass[NVDEV_ENGINE_DISP   ] = &nve0_disp_oclass;
 		device->oclass[NVDEV_ENGINE_COPY0  ] = &nve0_copy0_oclass;
@@ -113,28 +121,31 @@ nve0_identify(struct nouveau_device *device)
 		device->oclass[NVDEV_ENGINE_BSP    ] = &nve0_bsp_oclass;
 		device->oclass[NVDEV_ENGINE_VP     ] = &nve0_vp_oclass;
 		device->oclass[NVDEV_ENGINE_PPP    ] = &nvc0_ppp_oclass;
+		device->oclass[NVDEV_ENGINE_PERFMON] = &nve0_perfmon_oclass;
 		break;
 	case 0xe6:
 		device->cname = "GK106";
 		device->oclass[NVDEV_SUBDEV_VBIOS  ] = &nouveau_bios_oclass;
 		device->oclass[NVDEV_SUBDEV_GPIO   ] = &nve0_gpio_oclass;
 		device->oclass[NVDEV_SUBDEV_I2C    ] = &nvd0_i2c_oclass;
-		device->oclass[NVDEV_SUBDEV_CLOCK  ] = &nvc0_clock_oclass;
+		device->oclass[NVDEV_SUBDEV_CLOCK  ] = &nve0_clock_oclass;
 		device->oclass[NVDEV_SUBDEV_THERM  ] = &nvd0_therm_oclass;
 		device->oclass[NVDEV_SUBDEV_MXM    ] = &nv50_mxm_oclass;
 		device->oclass[NVDEV_SUBDEV_DEVINIT] = &nvc0_devinit_oclass;
-		device->oclass[NVDEV_SUBDEV_MC     ] = &nvc0_mc_oclass;
-		device->oclass[NVDEV_SUBDEV_BUS    ] = &nvc0_bus_oclass;
+		device->oclass[NVDEV_SUBDEV_MC     ] =  nvc3_mc_oclass;
+		device->oclass[NVDEV_SUBDEV_BUS    ] =  nvc0_bus_oclass;
 		device->oclass[NVDEV_SUBDEV_TIMER  ] = &nv04_timer_oclass;
-		device->oclass[NVDEV_SUBDEV_FB     ] = &nvc0_fb_oclass;
+		device->oclass[NVDEV_SUBDEV_FB     ] =  nve0_fb_oclass;
 		device->oclass[NVDEV_SUBDEV_LTCG   ] = &nvc0_ltcg_oclass;
 		device->oclass[NVDEV_SUBDEV_IBUS   ] = &nve0_ibus_oclass;
 		device->oclass[NVDEV_SUBDEV_INSTMEM] = &nv50_instmem_oclass;
 		device->oclass[NVDEV_SUBDEV_VM     ] = &nvc0_vmmgr_oclass;
 		device->oclass[NVDEV_SUBDEV_BAR    ] = &nvc0_bar_oclass;
+		device->oclass[NVDEV_SUBDEV_PWR    ] = &nvd0_pwr_oclass;
+		device->oclass[NVDEV_SUBDEV_VOLT   ] = &nv40_volt_oclass;
 		device->oclass[NVDEV_ENGINE_DMAOBJ ] = &nvd0_dmaeng_oclass;
-		device->oclass[NVDEV_ENGINE_FIFO   ] = &nve0_fifo_oclass;
-		device->oclass[NVDEV_ENGINE_SW     ] = &nvc0_software_oclass;
+		device->oclass[NVDEV_ENGINE_FIFO   ] =  nve0_fifo_oclass;
+		device->oclass[NVDEV_ENGINE_SW     ] =  nvc0_software_oclass;
 		device->oclass[NVDEV_ENGINE_GR     ] =  nve4_graph_oclass;
 		device->oclass[NVDEV_ENGINE_DISP   ] = &nve0_disp_oclass;
 		device->oclass[NVDEV_ENGINE_COPY0  ] = &nve0_copy0_oclass;
@@ -143,28 +154,31 @@ nve0_identify(struct nouveau_device *device)
 		device->oclass[NVDEV_ENGINE_BSP    ] = &nve0_bsp_oclass;
 		device->oclass[NVDEV_ENGINE_VP     ] = &nve0_vp_oclass;
 		device->oclass[NVDEV_ENGINE_PPP    ] = &nvc0_ppp_oclass;
+		device->oclass[NVDEV_ENGINE_PERFMON] = &nve0_perfmon_oclass;
 		break;
 	case 0xf0:
 		device->cname = "GK110";
 		device->oclass[NVDEV_SUBDEV_VBIOS  ] = &nouveau_bios_oclass;
 		device->oclass[NVDEV_SUBDEV_GPIO   ] = &nve0_gpio_oclass;
 		device->oclass[NVDEV_SUBDEV_I2C    ] = &nvd0_i2c_oclass;
-		device->oclass[NVDEV_SUBDEV_CLOCK  ] = &nvc0_clock_oclass;
+		device->oclass[NVDEV_SUBDEV_CLOCK  ] = &nve0_clock_oclass;
 		device->oclass[NVDEV_SUBDEV_THERM  ] = &nvd0_therm_oclass;
 		device->oclass[NVDEV_SUBDEV_MXM    ] = &nv50_mxm_oclass;
 		device->oclass[NVDEV_SUBDEV_DEVINIT] = &nvc0_devinit_oclass;
-		device->oclass[NVDEV_SUBDEV_MC     ] = &nvc0_mc_oclass;
-		device->oclass[NVDEV_SUBDEV_BUS    ] = &nvc0_bus_oclass;
+		device->oclass[NVDEV_SUBDEV_MC     ] =  nvc3_mc_oclass;
+		device->oclass[NVDEV_SUBDEV_BUS    ] =  nvc0_bus_oclass;
 		device->oclass[NVDEV_SUBDEV_TIMER  ] = &nv04_timer_oclass;
-		device->oclass[NVDEV_SUBDEV_FB     ] = &nvc0_fb_oclass;
+		device->oclass[NVDEV_SUBDEV_FB     ] =  nve0_fb_oclass;
 		device->oclass[NVDEV_SUBDEV_LTCG   ] = &nvc0_ltcg_oclass;
 		device->oclass[NVDEV_SUBDEV_IBUS   ] = &nve0_ibus_oclass;
 		device->oclass[NVDEV_SUBDEV_INSTMEM] = &nv50_instmem_oclass;
 		device->oclass[NVDEV_SUBDEV_VM     ] = &nvc0_vmmgr_oclass;
 		device->oclass[NVDEV_SUBDEV_BAR    ] = &nvc0_bar_oclass;
+		device->oclass[NVDEV_SUBDEV_PWR    ] = &nvd0_pwr_oclass;
+		device->oclass[NVDEV_SUBDEV_VOLT   ] = &nv40_volt_oclass;
 		device->oclass[NVDEV_ENGINE_DMAOBJ ] = &nvd0_dmaeng_oclass;
-		device->oclass[NVDEV_ENGINE_FIFO   ] = &nve0_fifo_oclass;
-		device->oclass[NVDEV_ENGINE_SW     ] = &nvc0_software_oclass;
+		device->oclass[NVDEV_ENGINE_FIFO   ] =  nve0_fifo_oclass;
+		device->oclass[NVDEV_ENGINE_SW     ] =  nvc0_software_oclass;
 		device->oclass[NVDEV_ENGINE_GR     ] =  nvf0_graph_oclass;
 		device->oclass[NVDEV_ENGINE_DISP   ] = &nvf0_disp_oclass;
 		device->oclass[NVDEV_ENGINE_COPY0  ] = &nve0_copy0_oclass;
@@ -175,6 +189,43 @@ nve0_identify(struct nouveau_device *device)
 		device->oclass[NVDEV_ENGINE_VP     ] = &nve0_vp_oclass;
 		device->oclass[NVDEV_ENGINE_PPP    ] = &nvc0_ppp_oclass;
 #endif
+		device->oclass[NVDEV_ENGINE_PERFMON] = &nvf0_perfmon_oclass;
+		break;
+	case 0x108:
+		device->cname = "GK208";
+		device->oclass[NVDEV_SUBDEV_VBIOS  ] = &nouveau_bios_oclass;
+		device->oclass[NVDEV_SUBDEV_GPIO   ] = &nve0_gpio_oclass;
+		device->oclass[NVDEV_SUBDEV_I2C    ] = &nvd0_i2c_oclass;
+		device->oclass[NVDEV_SUBDEV_CLOCK  ] = &nve0_clock_oclass;
+		device->oclass[NVDEV_SUBDEV_THERM  ] = &nvd0_therm_oclass;
+		device->oclass[NVDEV_SUBDEV_MXM    ] = &nv50_mxm_oclass;
+		device->oclass[NVDEV_SUBDEV_DEVINIT] = &nvc0_devinit_oclass;
+		device->oclass[NVDEV_SUBDEV_MC     ] =  nvc3_mc_oclass;
+		device->oclass[NVDEV_SUBDEV_BUS    ] =  nvc0_bus_oclass;
+		device->oclass[NVDEV_SUBDEV_TIMER  ] = &nv04_timer_oclass;
+		device->oclass[NVDEV_SUBDEV_FB     ] =  nve0_fb_oclass;
+		device->oclass[NVDEV_SUBDEV_LTCG   ] = &nvc0_ltcg_oclass;
+		device->oclass[NVDEV_SUBDEV_IBUS   ] = &nve0_ibus_oclass;
+		device->oclass[NVDEV_SUBDEV_INSTMEM] = &nv50_instmem_oclass;
+		device->oclass[NVDEV_SUBDEV_VM     ] = &nvc0_vmmgr_oclass;
+		device->oclass[NVDEV_SUBDEV_BAR    ] = &nvc0_bar_oclass;
+		device->oclass[NVDEV_SUBDEV_PWR    ] = &nv108_pwr_oclass;
+		device->oclass[NVDEV_SUBDEV_VOLT   ] = &nv40_volt_oclass;
+		device->oclass[NVDEV_ENGINE_DMAOBJ ] = &nvd0_dmaeng_oclass;
+#if 0
+		device->oclass[NVDEV_ENGINE_FIFO   ] =  nve0_fifo_oclass;
+		device->oclass[NVDEV_ENGINE_SW     ] =  nvc0_software_oclass;
+		device->oclass[NVDEV_ENGINE_GR     ] =  nvf0_graph_oclass;
+#endif
+		device->oclass[NVDEV_ENGINE_DISP   ] = &nvf0_disp_oclass;
+#if 0
+		device->oclass[NVDEV_ENGINE_COPY0  ] = &nve0_copy0_oclass;
+		device->oclass[NVDEV_ENGINE_COPY1  ] = &nve0_copy1_oclass;
+		device->oclass[NVDEV_ENGINE_COPY2  ] = &nve0_copy2_oclass;
+		device->oclass[NVDEV_ENGINE_BSP    ] = &nve0_bsp_oclass;
+		device->oclass[NVDEV_ENGINE_VP     ] = &nve0_vp_oclass;
+		device->oclass[NVDEV_ENGINE_PPP    ] = &nvc0_ppp_oclass;
+#endif
 		break;
 	default:
 		nv_fatal(device, "unknown Kepler chipset\n");
diff --git a/drivers/gpu/drm/nouveau/core/engine/device/priv.h b/drivers/gpu/drm/nouveau/core/engine/device/priv.h
new file mode 100644
index 000000000000..035fd5b9cfc3
--- /dev/null
+++ b/drivers/gpu/drm/nouveau/core/engine/device/priv.h
@@ -0,0 +1,8 @@
+#ifndef __NVKM_DEVICE_PRIV_H__
+#define __NVKM_DEVICE_PRIV_H__
+
+#include <engine/device.h>
+
+extern struct nouveau_oclass nouveau_control_oclass[];
+
+#endif
diff --git a/drivers/gpu/drm/nouveau/core/engine/disp/dport.c b/drivers/gpu/drm/nouveau/core/engine/disp/dport.c
index 054d9cff4f53..1bd4c63369c1 100644
--- a/drivers/gpu/drm/nouveau/core/engine/disp/dport.c
+++ b/drivers/gpu/drm/nouveau/core/engine/disp/dport.c
@@ -70,17 +70,10 @@ dp_set_link_config(struct dp_state *dp)
 	};
 	u32 lnkcmp;
 	u8 sink[2];
+	int ret;
 
 	DBG("%d lanes at %d KB/s\n", dp->link_nr, dp->link_bw);
 
-	/* set desired link configuration on the sink */
-	sink[0] = dp->link_bw / 27000;
-	sink[1] = dp->link_nr;
-	if (dp->dpcd[DPCD_RC02] & DPCD_RC02_ENHANCED_FRAME_CAP)
-		sink[1] |= DPCD_LC01_ENHANCED_FRAME_EN;
-
-	nv_wraux(dp->aux, DPCD_LC00, sink, 2);
-
 	/* set desired link configuration on the source */
 	if ((lnkcmp = dp->info.lnkcmp)) {
 		if (dp->version < 0x30) {
@@ -96,10 +89,22 @@ dp_set_link_config(struct dp_state *dp)
 		nvbios_exec(&init);
 	}
 
-	return dp->func->lnk_ctl(dp->disp, dp->outp, dp->head,
-				 dp->link_nr, dp->link_bw / 27000,
-				 dp->dpcd[DPCD_RC02] &
-					  DPCD_RC02_ENHANCED_FRAME_CAP);
+	ret = dp->func->lnk_ctl(dp->disp, dp->outp, dp->head,
+				dp->link_nr, dp->link_bw / 27000,
+				dp->dpcd[DPCD_RC02] &
+					 DPCD_RC02_ENHANCED_FRAME_CAP);
+	if (ret) {
+		ERR("lnk_ctl failed with %d\n", ret);
+		return ret;
+	}
+
+	/* set desired link configuration on the sink */
+	sink[0] = dp->link_bw / 27000;
+	sink[1] = dp->link_nr;
+	if (dp->dpcd[DPCD_RC02] & DPCD_RC02_ENHANCED_FRAME_CAP)
+		sink[1] |= DPCD_LC01_ENHANCED_FRAME_EN;
+
+	return nv_wraux(dp->aux, DPCD_LC00, sink, 2);
 }
 
 static void
@@ -294,8 +299,17 @@ nouveau_dp_train(struct nouveau_disp *disp, const struct nouveau_dp_func *func,
 
 	ret = nv_rdaux(dp->aux, 0x00000, dp->dpcd, sizeof(dp->dpcd));
 	if (ret) {
+		/* it's possible the display has been unplugged before we
+		 * get here.  we still need to execute the full set of
+		 * vbios scripts, and program the OR at a high enough
+		 * frequency to satisfy the target mode.  failure to do
+		 * so results at best in an UPDATE hanging, and at worst
+		 * with PDISP running away to join the circus.
+		 */
+		dp->dpcd[1] = link_bw[0] / 27000;
+		dp->dpcd[2] = 4;
+		dp->dpcd[3] = 0x00;
 		ERR("failed to read DPCD\n");
-		return ret;
 	}
 
 	/* adjust required bandwidth for 8B/10B coding overhead */
@@ -308,7 +322,7 @@ nouveau_dp_train(struct nouveau_disp *disp, const struct nouveau_dp_func *func,
 	while (*link_bw > (dp->dpcd[1] * 27000))
 		link_bw++;
 
-	while (link_bw[0]) {
+	while ((ret = -EIO) && link_bw[0]) {
 		/* find minimum required lane count at this link rate */
 		dp->link_nr = dp->dpcd[2] & DPCD_RC02_MAX_LANE_COUNT;
 		while ((dp->link_nr >> 1) * link_bw[0] > datarate)
@@ -328,8 +342,10 @@ nouveau_dp_train(struct nouveau_disp *disp, const struct nouveau_dp_func *func,
 			    !dp_link_train_eq(dp))
 				break;
 		} else
-		if (ret >= 1) {
-			/* dp_set_link_config() handled training */
+		if (ret) {
+			/* dp_set_link_config() handled training, or
+			 * we failed to communicate with the sink.
+			 */
 			break;
 		}
 
@@ -339,8 +355,10 @@ nouveau_dp_train(struct nouveau_disp *disp, const struct nouveau_dp_func *func,
 
 	/* finish link training */
 	dp_set_training_pattern(dp, 0);
+	if (ret < 0)
+		ERR("link training failed\n");
 
 	/* execute post-train script from vbios */
 	dp_link_train_fini(dp);
-	return true;
+	return (ret < 0) ? false : true;
 }
diff --git a/drivers/gpu/drm/nouveau/core/engine/disp/nv04.c b/drivers/gpu/drm/nouveau/core/engine/disp/nv04.c
index 05e903f08a36..a0bc8a89b699 100644
--- a/drivers/gpu/drm/nouveau/core/engine/disp/nv04.c
+++ b/drivers/gpu/drm/nouveau/core/engine/disp/nv04.c
@@ -59,6 +59,7 @@ nv04_disp_intr(struct nouveau_subdev *subdev)
 	struct nv04_disp_priv *priv = (void *)subdev;
 	u32 crtc0 = nv_rd32(priv, 0x600100);
 	u32 crtc1 = nv_rd32(priv, 0x602100);
+	u32 pvideo;
 
 	if (crtc0 & 0x00000001) {
 		nouveau_event_trigger(priv->base.vblank, 0);
@@ -69,6 +70,14 @@ nv04_disp_intr(struct nouveau_subdev *subdev)
 		nouveau_event_trigger(priv->base.vblank, 1);
 		nv_wr32(priv, 0x602100, 0x00000001);
 	}
+
+	if (nv_device(priv)->chipset >= 0x10 &&
+	    nv_device(priv)->chipset <= 0x40) {
+		pvideo = nv_rd32(priv, 0x8100);
+		if (pvideo & ~0x11)
+			nv_info(priv, "PVIDEO intr: %08x\n", pvideo);
+		nv_wr32(priv, 0x8100, pvideo);
+	}
 }
 
 static int
diff --git a/drivers/gpu/drm/nouveau/core/engine/disp/nvd0.c b/drivers/gpu/drm/nouveau/core/engine/disp/nvd0.c
index 52dd7a1db729..378a015091d2 100644
--- a/drivers/gpu/drm/nouveau/core/engine/disp/nvd0.c
+++ b/drivers/gpu/drm/nouveau/core/engine/disp/nvd0.c
@@ -541,6 +541,15 @@ nvd0_disp_base_init(struct nouveau_object *object)
 	nv_wr32(priv, 0x6100a0, 0x00000000);
 	nv_wr32(priv, 0x6100b0, 0x00000307);
 
+	/* disable underflow reporting, preventing an intermittent issue
+	 * on some nve4 boards where the production vbios left this
+	 * setting enabled by default.
+	 *
+	 * ftp://download.nvidia.com/open-gpu-doc/gk104-disable-underflow-reporting/1/gk104-disable-underflow-reporting.txt
+	 */
+	for (i = 0; i < priv->head.nr; i++)
+		nv_mask(priv, 0x616308 + (i * 0x800), 0x00000111, 0x00000010);
+
 	return 0;
 }
 
diff --git a/drivers/gpu/drm/nouveau/core/engine/disp/sornv94.c b/drivers/gpu/drm/nouveau/core/engine/disp/sornv94.c
index 7ec4ee83fb64..eea3ef59693d 100644
--- a/drivers/gpu/drm/nouveau/core/engine/disp/sornv94.c
+++ b/drivers/gpu/drm/nouveau/core/engine/disp/sornv94.c
@@ -97,8 +97,9 @@ nv94_sor_dp_drv_ctl(struct nouveau_disp *disp, struct dcb_output *outp,
 {
 	struct nouveau_bios *bios = nouveau_bios(disp);
 	struct nv50_disp_priv *priv = (void *)disp;
+	const u32 shift = nv94_sor_dp_lane_map(priv, lane);
 	const u32 loff = nv94_sor_loff(outp);
-	u32 addr, shift = nv94_sor_dp_lane_map(priv, lane);
+	u32 addr, data[3];
 	u8  ver, hdr, cnt, len;
 	struct nvbios_dpout info;
 	struct nvbios_dpcfg ocfg;
@@ -113,9 +114,12 @@ nv94_sor_dp_drv_ctl(struct nouveau_disp *disp, struct dcb_output *outp,
 	if (!addr)
 		return -EINVAL;
 
-	nv_mask(priv, 0x61c118 + loff, 0x000000ff << shift, ocfg.drv << shift);
-	nv_mask(priv, 0x61c120 + loff, 0x000000ff << shift, ocfg.pre << shift);
-	nv_mask(priv, 0x61c130 + loff, 0x0000ff00, ocfg.unk << 8);
+	data[0] = nv_rd32(priv, 0x61c118 + loff) & ~(0x000000ff << shift);
+	data[1] = nv_rd32(priv, 0x61c120 + loff) & ~(0x000000ff << shift);
+	data[2] = nv_rd32(priv, 0x61c130 + loff) & ~(0x0000ff00);
+	nv_wr32(priv, 0x61c118 + loff, data[0] | (ocfg.drv << shift));
+	nv_wr32(priv, 0x61c120 + loff, data[1] | (ocfg.pre << shift));
+	nv_wr32(priv, 0x61c130 + loff, data[2] | (ocfg.unk << 8));
 	return 0;
 }
 
diff --git a/drivers/gpu/drm/nouveau/core/engine/disp/sornvd0.c b/drivers/gpu/drm/nouveau/core/engine/disp/sornvd0.c
index 9e1d435d7282..d2df572f16a3 100644
--- a/drivers/gpu/drm/nouveau/core/engine/disp/sornvd0.c
+++ b/drivers/gpu/drm/nouveau/core/engine/disp/sornvd0.c
@@ -93,8 +93,9 @@ nvd0_sor_dp_drv_ctl(struct nouveau_disp *disp, struct dcb_output *outp,
 {
 	struct nouveau_bios *bios = nouveau_bios(disp);
 	struct nv50_disp_priv *priv = (void *)disp;
+	const u32 shift = nvd0_sor_dp_lane_map(priv, lane);
 	const u32 loff = nvd0_sor_loff(outp);
-	u32 addr, shift = nvd0_sor_dp_lane_map(priv, lane);
+	u32 addr, data[3];
 	u8  ver, hdr, cnt, len;
 	struct nvbios_dpout info;
 	struct nvbios_dpcfg ocfg;
@@ -109,9 +110,12 @@ nvd0_sor_dp_drv_ctl(struct nouveau_disp *disp, struct dcb_output *outp,
 	if (!addr)
 		return -EINVAL;
 
-	nv_mask(priv, 0x61c118 + loff, 0x000000ff << shift, ocfg.drv << shift);
-	nv_mask(priv, 0x61c120 + loff, 0x000000ff << shift, ocfg.pre << shift);
-	nv_mask(priv, 0x61c130 + loff, 0x0000ff00, ocfg.unk << 8);
+	data[0] = nv_rd32(priv, 0x61c118 + loff) & ~(0x000000ff << shift);
+	data[1] = nv_rd32(priv, 0x61c120 + loff) & ~(0x000000ff << shift);
+	data[2] = nv_rd32(priv, 0x61c130 + loff) & ~(0x0000ff00);
+	nv_wr32(priv, 0x61c118 + loff, data[0] | (ocfg.drv << shift));
+	nv_wr32(priv, 0x61c120 + loff, data[1] | (ocfg.pre << shift));
+	nv_wr32(priv, 0x61c130 + loff, data[2] | (ocfg.unk << 8));
 	nv_mask(priv, 0x61c13c + loff, 0x00000000, 0x00000000);
 	return 0;
 }
diff --git a/drivers/gpu/drm/nouveau/core/engine/fifo/nv04.c b/drivers/gpu/drm/nouveau/core/engine/fifo/nv04.c
index f877bd524a92..54f26cc801c7 100644
--- a/drivers/gpu/drm/nouveau/core/engine/fifo/nv04.c
+++ b/drivers/gpu/drm/nouveau/core/engine/fifo/nv04.c
@@ -632,8 +632,8 @@ nv04_fifo_init(struct nouveau_object *object)
 	return 0;
 }
 
-struct nouveau_oclass
-nv04_fifo_oclass = {
+struct nouveau_oclass *
+nv04_fifo_oclass = &(struct nouveau_oclass) {
 	.handle = NV_ENGINE(FIFO, 0x04),
 	.ofuncs = &(struct nouveau_ofuncs) {
 		.ctor = nv04_fifo_ctor,
diff --git a/drivers/gpu/drm/nouveau/core/engine/fifo/nv10.c b/drivers/gpu/drm/nouveau/core/engine/fifo/nv10.c
index 2c927c1d173b..571a22aa1ae5 100644
--- a/drivers/gpu/drm/nouveau/core/engine/fifo/nv10.c
+++ b/drivers/gpu/drm/nouveau/core/engine/fifo/nv10.c
@@ -159,8 +159,8 @@ nv10_fifo_ctor(struct nouveau_object *parent, struct nouveau_object *engine,
 	return 0;
 }
 
-struct nouveau_oclass
-nv10_fifo_oclass = {
+struct nouveau_oclass *
+nv10_fifo_oclass = &(struct nouveau_oclass) {
 	.handle = NV_ENGINE(FIFO, 0x10),
 	.ofuncs = &(struct nouveau_ofuncs) {
 		.ctor = nv10_fifo_ctor,
diff --git a/drivers/gpu/drm/nouveau/core/engine/fifo/nv17.c b/drivers/gpu/drm/nouveau/core/engine/fifo/nv17.c
index a9cb51d38c57..f25760209316 100644
--- a/drivers/gpu/drm/nouveau/core/engine/fifo/nv17.c
+++ b/drivers/gpu/drm/nouveau/core/engine/fifo/nv17.c
@@ -196,8 +196,8 @@ nv17_fifo_init(struct nouveau_object *object)
 	return 0;
 }
 
-struct nouveau_oclass
-nv17_fifo_oclass = {
+struct nouveau_oclass *
+nv17_fifo_oclass = &(struct nouveau_oclass) {
 	.handle = NV_ENGINE(FIFO, 0x17),
 	.ofuncs = &(struct nouveau_ofuncs) {
 		.ctor = nv17_fifo_ctor,
diff --git a/drivers/gpu/drm/nouveau/core/engine/fifo/nv40.c b/drivers/gpu/drm/nouveau/core/engine/fifo/nv40.c
index 5c7433d5069f..343487ed2238 100644
--- a/drivers/gpu/drm/nouveau/core/engine/fifo/nv40.c
+++ b/drivers/gpu/drm/nouveau/core/engine/fifo/nv40.c
@@ -337,8 +337,8 @@ nv40_fifo_init(struct nouveau_object *object)
 	return 0;
 }
 
-struct nouveau_oclass
-nv40_fifo_oclass = {
+struct nouveau_oclass *
+nv40_fifo_oclass = &(struct nouveau_oclass) {
 	.handle = NV_ENGINE(FIFO, 0x40),
 	.ofuncs = &(struct nouveau_ofuncs) {
 		.ctor = nv40_fifo_ctor,
diff --git a/drivers/gpu/drm/nouveau/core/engine/fifo/nv50.c b/drivers/gpu/drm/nouveau/core/engine/fifo/nv50.c
index 7e5dff51d3c5..5f555788121c 100644
--- a/drivers/gpu/drm/nouveau/core/engine/fifo/nv50.c
+++ b/drivers/gpu/drm/nouveau/core/engine/fifo/nv50.c
@@ -502,8 +502,8 @@ nv50_fifo_init(struct nouveau_object *object)
 	return 0;
 }
 
-struct nouveau_oclass
-nv50_fifo_oclass = {
+struct nouveau_oclass *
+nv50_fifo_oclass = &(struct nouveau_oclass) {
 	.handle = NV_ENGINE(FIFO, 0x50),
 	.ofuncs = &(struct nouveau_ofuncs) {
 		.ctor = nv50_fifo_ctor,
diff --git a/drivers/gpu/drm/nouveau/core/engine/fifo/nv84.c b/drivers/gpu/drm/nouveau/core/engine/fifo/nv84.c
index 91a87cd7195a..0908dc834c84 100644
--- a/drivers/gpu/drm/nouveau/core/engine/fifo/nv84.c
+++ b/drivers/gpu/drm/nouveau/core/engine/fifo/nv84.c
@@ -144,7 +144,7 @@ nv84_fifo_object_attach(struct nouveau_object *parent,
 	case NVDEV_ENGINE_COPY0 : context |= 0x00300000; break;
 	case NVDEV_ENGINE_VP    : context |= 0x00400000; break;
 	case NVDEV_ENGINE_CRYPT :
-	case NVDEV_ENGINE_UNK1C1: context |= 0x00500000; break;
+	case NVDEV_ENGINE_VIC   : context |= 0x00500000; break;
 	case NVDEV_ENGINE_BSP   : context |= 0x00600000; break;
 	default:
 		return -EINVAL;
@@ -180,7 +180,7 @@ nv84_fifo_chan_ctor_dma(struct nouveau_object *parent,
 					  (1ULL << NVDEV_ENGINE_BSP) |
 					  (1ULL << NVDEV_ENGINE_PPP) |
 					  (1ULL << NVDEV_ENGINE_COPY0) |
-					  (1ULL << NVDEV_ENGINE_UNK1C1), &chan);
+					  (1ULL << NVDEV_ENGINE_VIC), &chan);
 	*pobject = nv_object(chan);
 	if (ret)
 		return ret;
@@ -243,7 +243,7 @@ nv84_fifo_chan_ctor_ind(struct nouveau_object *parent,
 					  (1ULL << NVDEV_ENGINE_BSP) |
 					  (1ULL << NVDEV_ENGINE_PPP) |
 					  (1ULL << NVDEV_ENGINE_COPY0) |
-					  (1ULL << NVDEV_ENGINE_UNK1C1), &chan);
+					  (1ULL << NVDEV_ENGINE_VIC), &chan);
 	*pobject = nv_object(chan);
 	if (ret)
 		return ret;
@@ -435,8 +435,8 @@ nv84_fifo_ctor(struct nouveau_object *parent, struct nouveau_object *engine,
 	return 0;
 }
 
-struct nouveau_oclass
-nv84_fifo_oclass = {
+struct nouveau_oclass *
+nv84_fifo_oclass = &(struct nouveau_oclass) {
 	.handle = NV_ENGINE(FIFO, 0x84),
 	.ofuncs = &(struct nouveau_ofuncs) {
 		.ctor = nv84_fifo_ctor,
diff --git a/drivers/gpu/drm/nouveau/core/engine/fifo/nvc0.c b/drivers/gpu/drm/nouveau/core/engine/fifo/nvc0.c
index ce92f289e751..9ac94d4e5646 100644
--- a/drivers/gpu/drm/nouveau/core/engine/fifo/nvc0.c
+++ b/drivers/gpu/drm/nouveau/core/engine/fifo/nvc0.c
@@ -494,13 +494,6 @@ nvc0_fifo_isr_subfifo_intr(struct nvc0_fifo_priv *priv, int unit)
 	u32 mthd = (addr & 0x00003ffc);
 	u32 show = stat;
 
-	if (stat & 0x00200000) {
-		if (mthd == 0x0054) {
-			if (!nvc0_fifo_swmthd(priv, chid, 0x0500, 0x00000000))
-				show &= ~0x00200000;
-		}
-	}
-
 	if (stat & 0x00800000) {
 		if (!nvc0_fifo_swmthd(priv, chid, mthd, data))
 			show &= ~0x00800000;
@@ -720,8 +713,8 @@ nvc0_fifo_init(struct nouveau_object *object)
 	return 0;
 }
 
-struct nouveau_oclass
-nvc0_fifo_oclass = {
+struct nouveau_oclass *
+nvc0_fifo_oclass = &(struct nouveau_oclass) {
 	.handle = NV_ENGINE(FIFO, 0xc0),
 	.ofuncs = &(struct nouveau_ofuncs) {
 		.ctor = nvc0_fifo_ctor,
diff --git a/drivers/gpu/drm/nouveau/core/engine/fifo/nve0.c b/drivers/gpu/drm/nouveau/core/engine/fifo/nve0.c
index 8e8121abe31b..04f412922d2d 100644
--- a/drivers/gpu/drm/nouveau/core/engine/fifo/nve0.c
+++ b/drivers/gpu/drm/nouveau/core/engine/fifo/nve0.c
@@ -481,13 +481,6 @@ nve0_fifo_isr_subfifo_intr(struct nve0_fifo_priv *priv, int unit)
 	u32 mthd = (addr & 0x00003ffc);
 	u32 show = stat;
 
-	if (stat & 0x00200000) {
-		if (mthd == 0x0054) {
-			if (!nve0_fifo_swmthd(priv, chid, 0x0500, 0x00000000))
-				show &= ~0x00200000;
-		}
-	}
-
 	if (stat & 0x00800000) {
 		if (!nve0_fifo_swmthd(priv, chid, mthd, data))
 			show &= ~0x00800000;
@@ -675,8 +668,8 @@ nve0_fifo_init(struct nouveau_object *object)
 	return 0;
 }
 
-struct nouveau_oclass
-nve0_fifo_oclass = {
+struct nouveau_oclass *
+nve0_fifo_oclass = &(struct nouveau_oclass) {
 	.handle = NV_ENGINE(FIFO, 0xe0),
 	.ofuncs = &(struct nouveau_ofuncs) {
 		.ctor = nve0_fifo_ctor,
diff --git a/drivers/gpu/drm/nouveau/core/engine/graph/ctxnvc0.c b/drivers/gpu/drm/nouveau/core/engine/graph/ctxnvc0.c
index 64dca260912f..fe67415c3e17 100644
--- a/drivers/gpu/drm/nouveau/core/engine/graph/ctxnvc0.c
+++ b/drivers/gpu/drm/nouveau/core/engine/graph/ctxnvc0.c
@@ -1039,7 +1039,7 @@ nvc0_grctx_generate_r406800(struct nvc0_graph_priv *priv)
 			} while (!tpcnr[gpc]);
 			tpc = priv->tpc_nr[gpc] - tpcnr[gpc]--;
 
-			tpc_set |= 1 << ((gpc * 8) + tpc);
+			tpc_set |= 1ULL << ((gpc * 8) + tpc);
 		}
 
 		nv_wr32(priv, 0x406800 + (i * 0x20), lower_32_bits(tpc_set));
diff --git a/drivers/gpu/drm/nouveau/core/engine/graph/ctxnvc1.c b/drivers/gpu/drm/nouveau/core/engine/graph/ctxnvc1.c
index e5be3ee7f172..71b4283f7fad 100644
--- a/drivers/gpu/drm/nouveau/core/engine/graph/ctxnvc1.c
+++ b/drivers/gpu/drm/nouveau/core/engine/graph/ctxnvc1.c
@@ -587,6 +587,7 @@ nvc1_grctx_init_unk58xx[] = {
 	{ 0x405870,   4, 0x04, 0x00000001 },
 	{ 0x405a00,   2, 0x04, 0x00000000 },
 	{ 0x405a18,   1, 0x04, 0x00000000 },
+	{}
 };
 
 static struct nvc0_graph_init
@@ -598,6 +599,7 @@ nvc1_grctx_init_rop[] = {
 	{ 0x408904,   1, 0x04, 0x62000001 },
 	{ 0x408908,   1, 0x04, 0x00c80929 },
 	{ 0x408980,   1, 0x04, 0x0000011d },
+	{}
 };
 
 static struct nvc0_graph_init
@@ -671,6 +673,7 @@ nvc1_grctx_init_gpc_0[] = {
 	{ 0x419000,   1, 0x04, 0x00000780 },
 	{ 0x419004,   2, 0x04, 0x00000000 },
 	{ 0x419014,   1, 0x04, 0x00000004 },
+	{}
 };
 
 static struct nvc0_graph_init
@@ -717,6 +720,7 @@ nvc1_grctx_init_tpc[] = {
 	{ 0x419e98,   1, 0x04, 0x00000000 },
 	{ 0x419ee0,   1, 0x04, 0x00011110 },
 	{ 0x419f30,  11, 0x04, 0x00000000 },
+	{}
 };
 
 void
diff --git a/drivers/gpu/drm/nouveau/core/engine/graph/ctxnvd7.c b/drivers/gpu/drm/nouveau/core/engine/graph/ctxnvd7.c
index 438e78410808..c4740d528532 100644
--- a/drivers/gpu/drm/nouveau/core/engine/graph/ctxnvd7.c
+++ b/drivers/gpu/drm/nouveau/core/engine/graph/ctxnvd7.c
@@ -258,6 +258,7 @@ nvd7_grctx_init_hub[] = {
 	nvc0_grctx_init_unk78xx,
 	nvc0_grctx_init_unk80xx,
 	nvd9_grctx_init_rop,
+	NULL
 };
 
 struct nvc0_graph_init *
diff --git a/drivers/gpu/drm/nouveau/core/engine/graph/ctxnvd9.c b/drivers/gpu/drm/nouveau/core/engine/graph/ctxnvd9.c
index 818a4751df46..a1102cbf2fdc 100644
--- a/drivers/gpu/drm/nouveau/core/engine/graph/ctxnvd9.c
+++ b/drivers/gpu/drm/nouveau/core/engine/graph/ctxnvd9.c
@@ -466,6 +466,7 @@ nvd9_grctx_init_hub[] = {
 	nvc0_grctx_init_unk78xx,
 	nvc0_grctx_init_unk80xx,
 	nvd9_grctx_init_rop,
+	NULL
 };
 
 struct nvc0_graph_init *
diff --git a/drivers/gpu/drm/nouveau/core/engine/graph/nv10.c b/drivers/gpu/drm/nouveau/core/engine/graph/nv10.c
index 23c143aaa556..4532f7e5618c 100644
--- a/drivers/gpu/drm/nouveau/core/engine/graph/nv10.c
+++ b/drivers/gpu/drm/nouveau/core/engine/graph/nv10.c
@@ -945,7 +945,8 @@ nv10_graph_load_context(struct nv10_graph_chan *chan, int chid)
 	for (i = 0; i < ARRAY_SIZE(nv10_graph_ctx_regs); i++)
 		nv_wr32(priv, nv10_graph_ctx_regs[i], chan->nv10[i]);
 
-	if (nv_device(priv)->chipset >= 0x17) {
+	if (nv_device(priv)->card_type >= NV_11 &&
+	    nv_device(priv)->chipset >= 0x17) {
 		for (i = 0; i < ARRAY_SIZE(nv17_graph_ctx_regs); i++)
 			nv_wr32(priv, nv17_graph_ctx_regs[i], chan->nv17[i]);
 	}
@@ -970,7 +971,8 @@ nv10_graph_unload_context(struct nv10_graph_chan *chan)
 	for (i = 0; i < ARRAY_SIZE(nv10_graph_ctx_regs); i++)
 		chan->nv10[i] = nv_rd32(priv, nv10_graph_ctx_regs[i]);
 
-	if (nv_device(priv)->chipset >= 0x17) {
+	if (nv_device(priv)->card_type >= NV_11 &&
+	    nv_device(priv)->chipset >= 0x17) {
 		for (i = 0; i < ARRAY_SIZE(nv17_graph_ctx_regs); i++)
 			chan->nv17[i] = nv_rd32(priv, nv17_graph_ctx_regs[i]);
 	}
@@ -1052,7 +1054,8 @@ nv10_graph_context_ctor(struct nouveau_object *parent,
 	NV_WRITE_CTX(0x00400e14, 0x00001000);
 	NV_WRITE_CTX(0x00400e30, 0x00080008);
 	NV_WRITE_CTX(0x00400e34, 0x00080008);
-	if (nv_device(priv)->chipset >= 0x17) {
+	if (nv_device(priv)->card_type >= NV_11 &&
+	    nv_device(priv)->chipset >= 0x17) {
 		/* is it really needed ??? */
 		NV17_WRITE_CTX(NV10_PGRAPH_DEBUG_4,
 					nv_rd32(priv, NV10_PGRAPH_DEBUG_4));
@@ -1231,7 +1234,7 @@ nv10_graph_ctor(struct nouveau_object *parent, struct nouveau_object *engine,
 		nv_engine(priv)->sclass = nv10_graph_sclass;
 	else
 	if (nv_device(priv)->chipset <  0x17 ||
-	    nv_device(priv)->chipset == 0x1a)
+	    nv_device(priv)->card_type < NV_11)
 		nv_engine(priv)->sclass = nv15_graph_sclass;
 	else
 		nv_engine(priv)->sclass = nv17_graph_sclass;
@@ -1270,7 +1273,8 @@ nv10_graph_init(struct nouveau_object *object)
 	nv_wr32(priv, NV04_PGRAPH_DEBUG_2, 0x25f92ad9);
 	nv_wr32(priv, NV04_PGRAPH_DEBUG_3, 0x55DE0830 | (1 << 29) | (1 << 31));
 
-	if (nv_device(priv)->chipset >= 0x17) {
+	if (nv_device(priv)->card_type >= NV_11 &&
+	    nv_device(priv)->chipset >= 0x17) {
 		nv_wr32(priv, NV10_PGRAPH_DEBUG_4, 0x1f000000);
 		nv_wr32(priv, 0x400a10, 0x03ff3fb6);
 		nv_wr32(priv, 0x400838, 0x002f8684);
diff --git a/drivers/gpu/drm/nouveau/core/engine/graph/nvc0.c b/drivers/gpu/drm/nouveau/core/engine/graph/nvc0.c
index 3f4f35cc3848..434bb4b0fa2e 100644
--- a/drivers/gpu/drm/nouveau/core/engine/graph/nvc0.c
+++ b/drivers/gpu/drm/nouveau/core/engine/graph/nvc0.c
@@ -1138,7 +1138,7 @@ nvc0_graph_ctor(struct nouveau_object *parent, struct nouveau_object *engine,
 	if (ret)
 		return ret;
 
-	nv_subdev(priv)->unit = 0x18001000;
+	nv_subdev(priv)->unit = 0x08001000;
 	nv_subdev(priv)->intr = nvc0_graph_intr;
 
 	priv->base.units = nvc0_graph_units;
diff --git a/drivers/gpu/drm/nouveau/core/engine/mpeg/nv31.c b/drivers/gpu/drm/nouveau/core/engine/mpeg/nv31.c
index c19004301309..7eb6d94c84e2 100644
--- a/drivers/gpu/drm/nouveau/core/engine/mpeg/nv31.c
+++ b/drivers/gpu/drm/nouveau/core/engine/mpeg/nv31.c
@@ -34,16 +34,7 @@
 
 #include <engine/fifo.h>
 #include <engine/mpeg.h>
-#include <engine/graph/nv40.h>
-
-struct nv31_mpeg_priv {
-	struct nouveau_mpeg base;
-	atomic_t refcount;
-};
-
-struct nv31_mpeg_chan {
-	struct nouveau_object base;
-};
+#include <engine/mpeg/nv31.h>
 
 /*******************************************************************************
  * MPEG object classes
@@ -89,18 +80,18 @@ nv31_mpeg_mthd_dma(struct nouveau_object *object, u32 mthd, void *arg, u32 len)
 
 	if (mthd == 0x0190) {
 		/* DMA_CMD */
-		nv_mask(priv, 0x00b300, 0x00030000, (dma0 & 0x00030000));
+		nv_mask(priv, 0x00b300, 0x00010000, (dma0 & 0x00030000) ? 0x00010000 : 0);
 		nv_wr32(priv, 0x00b334, base);
 		nv_wr32(priv, 0x00b324, size);
 	} else
 	if (mthd == 0x01a0) {
 		/* DMA_DATA */
-		nv_mask(priv, 0x00b300, 0x000c0000, (dma0 & 0x00030000) << 2);
+		nv_mask(priv, 0x00b300, 0x00020000, (dma0 & 0x00030000) ? 0x00020000 : 0);
 		nv_wr32(priv, 0x00b360, base);
 		nv_wr32(priv, 0x00b364, size);
 	} else {
 		/* DMA_IMAGE, VRAM only */
-		if (dma0 & 0x000c0000)
+		if (dma0 & 0x00030000)
 			return -EINVAL;
 
 		nv_wr32(priv, 0x00b370, base);
@@ -110,7 +101,7 @@ nv31_mpeg_mthd_dma(struct nouveau_object *object, u32 mthd, void *arg, u32 len)
 	return 0;
 }
 
-static struct nouveau_ofuncs
+struct nouveau_ofuncs
 nv31_mpeg_ofuncs = {
 	.ctor = nv31_mpeg_object_ctor,
 	.dtor = _nouveau_gpuobj_dtor,
@@ -146,16 +137,23 @@ nv31_mpeg_context_ctor(struct nouveau_object *parent,
 {
 	struct nv31_mpeg_priv *priv = (void *)engine;
 	struct nv31_mpeg_chan *chan;
+	unsigned long flags;
 	int ret;
 
-	if (!atomic_add_unless(&priv->refcount, 1, 1))
-		return -EBUSY;
-
 	ret = nouveau_object_create(parent, engine, oclass, 0, &chan);
 	*pobject = nv_object(chan);
 	if (ret)
 		return ret;
 
+	spin_lock_irqsave(&nv_engine(priv)->lock, flags);
+	if (priv->chan) {
+		spin_unlock_irqrestore(&nv_engine(priv)->lock, flags);
+		nouveau_object_destroy(&chan->base);
+		*pobject = NULL;
+		return -EBUSY;
+	}
+	priv->chan = chan;
+	spin_unlock_irqrestore(&nv_engine(priv)->lock, flags);
 	return 0;
 }
 
@@ -164,11 +162,15 @@ nv31_mpeg_context_dtor(struct nouveau_object *object)
 {
 	struct nv31_mpeg_priv *priv = (void *)object->engine;
 	struct nv31_mpeg_chan *chan = (void *)object;
-	atomic_dec(&priv->refcount);
+	unsigned long flags;
+
+	spin_lock_irqsave(&nv_engine(priv)->lock, flags);
+	priv->chan = NULL;
+	spin_unlock_irqrestore(&nv_engine(priv)->lock, flags);
 	nouveau_object_destroy(&chan->base);
 }
 
-static struct nouveau_oclass
+struct nouveau_oclass
 nv31_mpeg_cclass = {
 	.handle = NV_ENGCTX(MPEG, 0x31),
 	.ofuncs = &(struct nouveau_ofuncs) {
@@ -197,21 +199,19 @@ nv31_mpeg_tile_prog(struct nouveau_engine *engine, int i)
 void
 nv31_mpeg_intr(struct nouveau_subdev *subdev)
 {
+	struct nv31_mpeg_priv *priv = (void *)subdev;
 	struct nouveau_fifo *pfifo = nouveau_fifo(subdev);
-	struct nouveau_engine *engine = nv_engine(subdev);
-	struct nouveau_object *engctx;
 	struct nouveau_handle *handle;
-	struct nv31_mpeg_priv *priv = (void *)subdev;
-	u32 inst = nv_rd32(priv, 0x00b318) & 0x000fffff;
+	struct nouveau_object *engctx;
 	u32 stat = nv_rd32(priv, 0x00b100);
 	u32 type = nv_rd32(priv, 0x00b230);
 	u32 mthd = nv_rd32(priv, 0x00b234);
 	u32 data = nv_rd32(priv, 0x00b238);
 	u32 show = stat;
-	int chid;
+	unsigned long flags;
 
-	engctx = nouveau_engctx_get(engine, inst);
-	chid   = pfifo->chid(pfifo, engctx);
+	spin_lock_irqsave(&nv_engine(priv)->lock, flags);
+	engctx = nv_object(priv->chan);
 
 	if (stat & 0x01000000) {
 		/* happens on initial binding of the object */
@@ -220,7 +220,7 @@ nv31_mpeg_intr(struct nouveau_subdev *subdev)
 			show &= ~0x01000000;
 		}
 
-		if (type == 0x00000010) {
+		if (type == 0x00000010 && engctx) {
 			handle = nouveau_handle_get_class(engctx, 0x3174);
 			if (handle && !nv_call(handle->object, mthd, data))
 				show &= ~0x01000000;
@@ -232,13 +232,12 @@ nv31_mpeg_intr(struct nouveau_subdev *subdev)
 	nv_wr32(priv, 0x00b230, 0x00000001);
 
 	if (show) {
-		nv_error(priv,
-			 "ch %d [0x%08x %s] 0x%08x 0x%08x 0x%08x 0x%08x\n",
-			 chid, inst << 4, nouveau_client_name(engctx), stat,
-			 type, mthd, data);
+		nv_error(priv, "ch %d [%s] 0x%08x 0x%08x 0x%08x 0x%08x\n",
+			 pfifo->chid(pfifo, engctx),
+			 nouveau_client_name(engctx), stat, type, mthd, data);
 	}
 
-	nouveau_engctx_put(engctx);
+	spin_unlock_irqrestore(&nv_engine(priv)->lock, flags);
 }
 
 static int
@@ -284,10 +283,7 @@ nv31_mpeg_init(struct nouveau_object *object)
 	/* PMPEG init */
 	nv_wr32(priv, 0x00b32c, 0x00000000);
 	nv_wr32(priv, 0x00b314, 0x00000100);
-	if (nv_device(priv)->chipset >= 0x40 && nv44_graph_class(priv))
-		nv_wr32(priv, 0x00b220, 0x00000044);
-	else
-		nv_wr32(priv, 0x00b220, 0x00000031);
+	nv_wr32(priv, 0x00b220, 0x00000031);
 	nv_wr32(priv, 0x00b300, 0x02001ec1);
 	nv_mask(priv, 0x00b32c, 0x00000001, 0x00000001);
 
diff --git a/drivers/gpu/drm/nouveau/core/engine/mpeg/nv31.h b/drivers/gpu/drm/nouveau/core/engine/mpeg/nv31.h
new file mode 100644
index 000000000000..d08629d0b6ad
--- /dev/null
+++ b/drivers/gpu/drm/nouveau/core/engine/mpeg/nv31.h
@@ -0,0 +1,15 @@
+#ifndef __NV31_MPEG_H__
+#define __NV31_MPEG_H__
+
+#include <engine/mpeg.h>
+
+struct nv31_mpeg_chan {
+	struct nouveau_object base;
+};
+
+struct nv31_mpeg_priv {
+	struct nouveau_mpeg base;
+	struct nv31_mpeg_chan *chan;
+};
+
+#endif
diff --git a/drivers/gpu/drm/nouveau/core/engine/mpeg/nv40.c b/drivers/gpu/drm/nouveau/core/engine/mpeg/nv40.c
index dd6196072e9c..d4e7ec0ba68c 100644
--- a/drivers/gpu/drm/nouveau/core/engine/mpeg/nv40.c
+++ b/drivers/gpu/drm/nouveau/core/engine/mpeg/nv40.c
@@ -31,66 +31,63 @@
 #include <subdev/instmem.h>
 
 #include <engine/mpeg.h>
-#include <engine/graph/nv40.h>
-
-struct nv40_mpeg_priv {
-	struct nouveau_mpeg base;
-};
-
-struct nv40_mpeg_chan {
-	struct nouveau_mpeg_chan base;
-};
+#include <engine/mpeg/nv31.h>
 
 /*******************************************************************************
- * PMPEG context
+ * MPEG object classes
  ******************************************************************************/
 
 static int
-nv40_mpeg_context_ctor(struct nouveau_object *parent,
-		       struct nouveau_object *engine,
-		       struct nouveau_oclass *oclass, void *data, u32 size,
-		       struct nouveau_object **pobject)
+nv40_mpeg_mthd_dma(struct nouveau_object *object, u32 mthd, void *arg, u32 len)
 {
-	struct nv40_mpeg_chan *chan;
-	int ret;
-
-	ret = nouveau_mpeg_context_create(parent, engine, oclass, NULL,
-					  264 * 4, 16,
-					  NVOBJ_FLAG_ZERO_ALLOC, &chan);
-	*pobject = nv_object(chan);
-	if (ret)
-		return ret;
+	struct nouveau_instmem *imem = nouveau_instmem(object);
+	struct nv31_mpeg_priv *priv = (void *)object->engine;
+	u32 inst = *(u32 *)arg << 4;
+	u32 dma0 = nv_ro32(imem, inst + 0);
+	u32 dma1 = nv_ro32(imem, inst + 4);
+	u32 dma2 = nv_ro32(imem, inst + 8);
+	u32 base = (dma2 & 0xfffff000) | (dma0 >> 20);
+	u32 size = dma1 + 1;
+
+	/* only allow linear DMA objects */
+	if (!(dma0 & 0x00002000))
+		return -EINVAL;
+
+	if (mthd == 0x0190) {
+		/* DMA_CMD */
+		nv_mask(priv, 0x00b300, 0x00030000, (dma0 & 0x00030000));
+		nv_wr32(priv, 0x00b334, base);
+		nv_wr32(priv, 0x00b324, size);
+	} else
+	if (mthd == 0x01a0) {
+		/* DMA_DATA */
+		nv_mask(priv, 0x00b300, 0x000c0000, (dma0 & 0x00030000) << 2);
+		nv_wr32(priv, 0x00b360, base);
+		nv_wr32(priv, 0x00b364, size);
+	} else {
+		/* DMA_IMAGE, VRAM only */
+		if (dma0 & 0x00030000)
+			return -EINVAL;
+
+		nv_wr32(priv, 0x00b370, base);
+		nv_wr32(priv, 0x00b374, size);
+	}
 
-	nv_wo32(&chan->base.base, 0x78, 0x02001ec1);
 	return 0;
 }
 
-static int
-nv40_mpeg_context_fini(struct nouveau_object *object, bool suspend)
-{
-
-	struct nv40_mpeg_priv *priv = (void *)object->engine;
-	struct nv40_mpeg_chan *chan = (void *)object;
-	u32 inst = 0x80000000 | nv_gpuobj(chan)->addr >> 4;
-
-	nv_mask(priv, 0x00b32c, 0x00000001, 0x00000000);
-	if (nv_rd32(priv, 0x00b318) == inst)
-		nv_mask(priv, 0x00b318, 0x80000000, 0x00000000);
-	nv_mask(priv, 0x00b32c, 0x00000001, 0x00000001);
-	return 0;
-}
+static struct nouveau_omthds
+nv40_mpeg_omthds[] = {
+	{ 0x0190, 0x0190, nv40_mpeg_mthd_dma },
+	{ 0x01a0, 0x01a0, nv40_mpeg_mthd_dma },
+	{ 0x01b0, 0x01b0, nv40_mpeg_mthd_dma },
+	{}
+};
 
-static struct nouveau_oclass
-nv40_mpeg_cclass = {
-	.handle = NV_ENGCTX(MPEG, 0x40),
-	.ofuncs = &(struct nouveau_ofuncs) {
-		.ctor = nv40_mpeg_context_ctor,
-		.dtor = _nouveau_mpeg_context_dtor,
-		.init = _nouveau_mpeg_context_init,
-		.fini = nv40_mpeg_context_fini,
-		.rd32 = _nouveau_mpeg_context_rd32,
-		.wr32 = _nouveau_mpeg_context_wr32,
-	},
+struct nouveau_oclass
+nv40_mpeg_sclass[] = {
+	{ 0x3174, &nv31_mpeg_ofuncs, nv40_mpeg_omthds },
+	{}
 };
 
 /*******************************************************************************
@@ -100,7 +97,7 @@ nv40_mpeg_cclass = {
 static void
 nv40_mpeg_intr(struct nouveau_subdev *subdev)
 {
-	struct nv40_mpeg_priv *priv = (void *)subdev;
+	struct nv31_mpeg_priv *priv = (void *)subdev;
 	u32 stat;
 
 	if ((stat = nv_rd32(priv, 0x00b100)))
@@ -117,7 +114,7 @@ nv40_mpeg_ctor(struct nouveau_object *parent, struct nouveau_object *engine,
 	       struct nouveau_oclass *oclass, void *data, u32 size,
 	       struct nouveau_object **pobject)
 {
-	struct nv40_mpeg_priv *priv;
+	struct nv31_mpeg_priv *priv;
 	int ret;
 
 	ret = nouveau_mpeg_create(parent, engine, oclass, &priv);
@@ -127,8 +124,8 @@ nv40_mpeg_ctor(struct nouveau_object *parent, struct nouveau_object *engine,
 
 	nv_subdev(priv)->unit = 0x00000002;
 	nv_subdev(priv)->intr = nv40_mpeg_intr;
-	nv_engine(priv)->cclass = &nv40_mpeg_cclass;
-	nv_engine(priv)->sclass = nv31_mpeg_sclass;
+	nv_engine(priv)->cclass = &nv31_mpeg_cclass;
+	nv_engine(priv)->sclass = nv40_mpeg_sclass;
 	nv_engine(priv)->tile_prog = nv31_mpeg_tile_prog;
 	return 0;
 }
diff --git a/drivers/gpu/drm/nouveau/core/engine/mpeg/nv44.c b/drivers/gpu/drm/nouveau/core/engine/mpeg/nv44.c
new file mode 100644
index 000000000000..3d8c2133e0e8
--- /dev/null
+++ b/drivers/gpu/drm/nouveau/core/engine/mpeg/nv44.c
@@ -0,0 +1,194 @@
+/*
+ * Copyright 2012 Red Hat Inc.
+ *
+ * Permission is hereby granted, free of charge, to any person obtaining a
+ * copy of this software and associated documentation files (the "Software"),
+ * to deal in the Software without restriction, including without limitation
+ * the rights to use, copy, modify, merge, publish, distribute, sublicense,
+ * and/or sell copies of the Software, and to permit persons to whom the
+ * Software is furnished to do so, subject to the following conditions:
+ *
+ * The above copyright notice and this permission notice shall be included in
+ * all copies or substantial portions of the Software.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+ * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+ * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT.  IN NO EVENT SHALL
+ * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR
+ * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE,
+ * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR
+ * OTHER DEALINGS IN THE SOFTWARE.
+ *
+ * Authors: Ben Skeggs
+ */
+
+#include <core/os.h>
+#include <core/class.h>
+#include <core/client.h>
+#include <core/engctx.h>
+#include <core/handle.h>
+
+#include <subdev/fb.h>
+#include <subdev/timer.h>
+#include <subdev/instmem.h>
+
+#include <engine/fifo.h>
+#include <engine/mpeg.h>
+
+struct nv44_mpeg_priv {
+	struct nouveau_mpeg base;
+};
+
+struct nv44_mpeg_chan {
+	struct nouveau_mpeg_chan base;
+};
+
+/*******************************************************************************
+ * PMPEG context
+ ******************************************************************************/
+
+static int
+nv44_mpeg_context_ctor(struct nouveau_object *parent,
+		       struct nouveau_object *engine,
+		       struct nouveau_oclass *oclass, void *data, u32 size,
+		       struct nouveau_object **pobject)
+{
+	struct nv44_mpeg_chan *chan;
+	int ret;
+
+	ret = nouveau_mpeg_context_create(parent, engine, oclass, NULL,
+					  264 * 4, 16,
+					  NVOBJ_FLAG_ZERO_ALLOC, &chan);
+	*pobject = nv_object(chan);
+	if (ret)
+		return ret;
+
+	nv_wo32(&chan->base.base, 0x78, 0x02001ec1);
+	return 0;
+}
+
+static int
+nv44_mpeg_context_fini(struct nouveau_object *object, bool suspend)
+{
+
+	struct nv44_mpeg_priv *priv = (void *)object->engine;
+	struct nv44_mpeg_chan *chan = (void *)object;
+	u32 inst = 0x80000000 | nv_gpuobj(chan)->addr >> 4;
+
+	nv_mask(priv, 0x00b32c, 0x00000001, 0x00000000);
+	if (nv_rd32(priv, 0x00b318) == inst)
+		nv_mask(priv, 0x00b318, 0x80000000, 0x00000000);
+	nv_mask(priv, 0x00b32c, 0x00000001, 0x00000001);
+	return 0;
+}
+
+static struct nouveau_oclass
+nv44_mpeg_cclass = {
+	.handle = NV_ENGCTX(MPEG, 0x44),
+	.ofuncs = &(struct nouveau_ofuncs) {
+		.ctor = nv44_mpeg_context_ctor,
+		.dtor = _nouveau_mpeg_context_dtor,
+		.init = _nouveau_mpeg_context_init,
+		.fini = nv44_mpeg_context_fini,
+		.rd32 = _nouveau_mpeg_context_rd32,
+		.wr32 = _nouveau_mpeg_context_wr32,
+	},
+};
+
+/*******************************************************************************
+ * PMPEG engine/subdev functions
+ ******************************************************************************/
+
+static void
+nv44_mpeg_intr(struct nouveau_subdev *subdev)
+{
+	struct nouveau_fifo *pfifo = nouveau_fifo(subdev);
+	struct nouveau_engine *engine = nv_engine(subdev);
+	struct nouveau_object *engctx;
+	struct nouveau_handle *handle;
+	struct nv44_mpeg_priv *priv = (void *)subdev;
+	u32 inst = nv_rd32(priv, 0x00b318) & 0x000fffff;
+	u32 stat = nv_rd32(priv, 0x00b100);
+	u32 type = nv_rd32(priv, 0x00b230);
+	u32 mthd = nv_rd32(priv, 0x00b234);
+	u32 data = nv_rd32(priv, 0x00b238);
+	u32 show = stat;
+	int chid;
+
+	engctx = nouveau_engctx_get(engine, inst);
+	chid   = pfifo->chid(pfifo, engctx);
+
+	if (stat & 0x01000000) {
+		/* happens on initial binding of the object */
+		if (type == 0x00000020 && mthd == 0x0000) {
+			nv_mask(priv, 0x00b308, 0x00000000, 0x00000000);
+			show &= ~0x01000000;
+		}
+
+		if (type == 0x00000010) {
+			handle = nouveau_handle_get_class(engctx, 0x3174);
+			if (handle && !nv_call(handle->object, mthd, data))
+				show &= ~0x01000000;
+			nouveau_handle_put(handle);
+		}
+	}
+
+	nv_wr32(priv, 0x00b100, stat);
+	nv_wr32(priv, 0x00b230, 0x00000001);
+
+	if (show) {
+		nv_error(priv,
+			 "ch %d [0x%08x %s] 0x%08x 0x%08x 0x%08x 0x%08x\n",
+			 chid, inst << 4, nouveau_client_name(engctx), stat,
+			 type, mthd, data);
+	}
+
+	nouveau_engctx_put(engctx);
+}
+
+static void
+nv44_mpeg_me_intr(struct nouveau_subdev *subdev)
+{
+	struct nv44_mpeg_priv *priv = (void *)subdev;
+	u32 stat;
+
+	if ((stat = nv_rd32(priv, 0x00b100)))
+		nv44_mpeg_intr(subdev);
+
+	if ((stat = nv_rd32(priv, 0x00b800))) {
+		nv_error(priv, "PMSRCH 0x%08x\n", stat);
+		nv_wr32(priv, 0x00b800, stat);
+	}
+}
+
+static int
+nv44_mpeg_ctor(struct nouveau_object *parent, struct nouveau_object *engine,
+	       struct nouveau_oclass *oclass, void *data, u32 size,
+	       struct nouveau_object **pobject)
+{
+	struct nv44_mpeg_priv *priv;
+	int ret;
+
+	ret = nouveau_mpeg_create(parent, engine, oclass, &priv);
+	*pobject = nv_object(priv);
+	if (ret)
+		return ret;
+
+	nv_subdev(priv)->unit = 0x00000002;
+	nv_subdev(priv)->intr = nv44_mpeg_me_intr;
+	nv_engine(priv)->cclass = &nv44_mpeg_cclass;
+	nv_engine(priv)->sclass = nv40_mpeg_sclass;
+	nv_engine(priv)->tile_prog = nv31_mpeg_tile_prog;
+	return 0;
+}
+
+struct nouveau_oclass
+nv44_mpeg_oclass = {
+	.handle = NV_ENGINE(MPEG, 0x44),
+	.ofuncs = &(struct nouveau_ofuncs) {
+		.ctor = nv44_mpeg_ctor,
+		.dtor = _nouveau_mpeg_dtor,
+		.init = nv31_mpeg_init,
+		.fini = _nouveau_mpeg_fini,
+	},
+};
diff --git a/drivers/gpu/drm/nouveau/core/engine/perfmon/base.c b/drivers/gpu/drm/nouveau/core/engine/perfmon/base.c
new file mode 100644
index 000000000000..e9c5e51943ef
--- /dev/null
+++ b/drivers/gpu/drm/nouveau/core/engine/perfmon/base.c
@@ -0,0 +1,449 @@
+/*
+ * Copyright 2013 Red Hat Inc.
+ *
+ * Permission is hereby granted, free of charge, to any person obtaining a
+ * copy of this software and associated documentation files (the "Software"),
+ * to deal in the Software without restriction, including without limitation
+ * the rights to use, copy, modify, merge, publish, distribute, sublicense,
+ * and/or sell copies of the Software, and to permit persons to whom the
+ * Software is furnished to do so, subject to the following conditions:
+ *
+ * The above copyright notice and this permission notice shall be included in
+ * all copies or substantial portions of the Software.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+ * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+ * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT.  IN NO EVENT SHALL
+ * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR
+ * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE,
+ * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR
+ * OTHER DEALINGS IN THE SOFTWARE.
+ *
+ * Authors: Ben Skeggs
+ */
+
+#include <core/option.h>
+#include <core/class.h>
+
+#include <subdev/clock.h>
+
+#include "priv.h"
+
+#define QUAD_MASK 0x0f
+#define QUAD_FREE 0x01
+
+static struct nouveau_perfsig *
+nouveau_perfsig_find_(struct nouveau_perfdom *dom, const char *name, u32 size)
+{
+	char path[64];
+	int i;
+
+	if (name[0] != '/') {
+		for (i = 0; i < dom->signal_nr; i++) {
+			if ( dom->signal[i].name &&
+			    !strncmp(name, dom->signal[i].name, size))
+				return &dom->signal[i];
+		}
+	} else {
+		for (i = 0; i < dom->signal_nr; i++) {
+			snprintf(path, sizeof(path), "/%s/%02x", dom->name, i);
+			if (!strncmp(name, path, size))
+				return &dom->signal[i];
+		}
+	}
+
+	return NULL;
+}
+
+struct nouveau_perfsig *
+nouveau_perfsig_find(struct nouveau_perfmon *ppm, const char *name, u32 size,
+		     struct nouveau_perfdom **pdom)
+{
+	struct nouveau_perfdom *dom = *pdom;
+	struct nouveau_perfsig *sig;
+
+	if (dom == NULL) {
+		list_for_each_entry(dom, &ppm->domains, head) {
+			sig = nouveau_perfsig_find_(dom, name, size);
+			if (sig) {
+				*pdom = dom;
+				return sig;
+			}
+		}
+
+		return NULL;
+	}
+
+	return nouveau_perfsig_find_(dom, name, size);
+}
+
+struct nouveau_perfctr *
+nouveau_perfsig_wrap(struct nouveau_perfmon *ppm, const char *name,
+		     struct nouveau_perfdom **pdom)
+{
+	struct nouveau_perfsig *sig;
+	struct nouveau_perfctr *ctr;
+
+	sig = nouveau_perfsig_find(ppm, name, strlen(name), pdom);
+	if (!sig)
+		return NULL;
+
+	ctr = kzalloc(sizeof(*ctr), GFP_KERNEL);
+	if (ctr) {
+		ctr->signal[0] = sig;
+		ctr->logic_op = 0xaaaa;
+	}
+
+	return ctr;
+}
+
+/*******************************************************************************
+ * Perfmon object classes
+ ******************************************************************************/
+static int
+nouveau_perfctr_query(struct nouveau_object *object, u32 mthd,
+		      void *data, u32 size)
+{
+	struct nouveau_device *device = nv_device(object);
+	struct nouveau_perfmon *ppm = (void *)object->engine;
+	struct nouveau_perfdom *dom = NULL, *chk;
+	struct nv_perfctr_query *args = data;
+	const bool all = nouveau_boolopt(device->cfgopt, "NvPmShowAll", false);
+	const bool raw = nouveau_boolopt(device->cfgopt, "NvPmUnnamed", all);
+	const char *name;
+	int tmp = 0, di, si;
+	char path[64];
+
+	if (size < sizeof(*args))
+		return -EINVAL;
+
+	di = (args->iter & 0xff000000) >> 24;
+	si = (args->iter & 0x00ffffff) - 1;
+
+	list_for_each_entry(chk, &ppm->domains, head) {
+		if (tmp++ == di) {
+			dom = chk;
+			break;
+		}
+	}
+
+	if (dom == NULL || si >= (int)dom->signal_nr)
+		return -EINVAL;
+
+	if (si >= 0) {
+		if (raw || !(name = dom->signal[si].name)) {
+			snprintf(path, sizeof(path), "/%s/%02x", dom->name, si);
+			name = path;
+		}
+
+		if (args->name)
+			strncpy(args->name, name, args->size);
+		args->size = strlen(name) + 1;
+	}
+
+	do {
+		while (++si < dom->signal_nr) {
+			if (all || dom->signal[si].name) {
+				args->iter = (di << 24) | ++si;
+				return 0;
+			}
+		}
+		si = -1;
+		di = di + 1;
+		dom = list_entry(dom->head.next, typeof(*dom), head);
+	} while (&dom->head != &ppm->domains);
+
+	args->iter = 0xffffffff;
+	return 0;
+}
+
+static int
+nouveau_perfctr_sample(struct nouveau_object *object, u32 mthd,
+		       void *data, u32 size)
+{
+	struct nouveau_perfmon *ppm = (void *)object->engine;
+	struct nouveau_perfctr *ctr, *tmp;
+	struct nouveau_perfdom *dom;
+	struct nv_perfctr_sample *args = data;
+
+	if (size < sizeof(*args))
+		return -EINVAL;
+	ppm->sequence++;
+
+	list_for_each_entry(dom, &ppm->domains, head) {
+		/* sample previous batch of counters */
+		if (dom->quad != QUAD_MASK) {
+			dom->func->next(ppm, dom);
+			tmp = NULL;
+			while (!list_empty(&dom->list)) {
+				ctr = list_first_entry(&dom->list,
+							typeof(*ctr), head);
+				if (ctr->slot < 0) break;
+				if ( tmp && tmp == ctr) break;
+				if (!tmp) tmp = ctr;
+				dom->func->read(ppm, dom, ctr);
+				ctr->slot  = -1;
+				list_move_tail(&ctr->head, &dom->list);
+			}
+		}
+
+		dom->quad = QUAD_MASK;
+
+		/* setup next batch of counters for sampling */
+		list_for_each_entry(ctr, &dom->list, head) {
+			ctr->slot = ffs(dom->quad) - 1;
+			if (ctr->slot < 0)
+				break;
+			dom->quad &= ~(QUAD_FREE << ctr->slot);
+			dom->func->init(ppm, dom, ctr);
+		}
+
+		if (dom->quad != QUAD_MASK)
+			dom->func->next(ppm, dom);
+	}
+
+	return 0;
+}
+
+static int
+nouveau_perfctr_read(struct nouveau_object *object, u32 mthd,
+		     void *data, u32 size)
+{
+	struct nouveau_perfctr *ctr = (void *)object;
+	struct nv_perfctr_read *args = data;
+
+	if (size < sizeof(*args))
+		return -EINVAL;
+	if (!ctr->clk)
+		return -EAGAIN;
+
+	args->clk = ctr->clk;
+	args->ctr = ctr->ctr;
+	return 0;
+}
+
+static void
+nouveau_perfctr_dtor(struct nouveau_object *object)
+{
+	struct nouveau_perfctr *ctr = (void *)object;
+	if (ctr->head.next)
+		list_del(&ctr->head);
+	nouveau_object_destroy(&ctr->base);
+}
+
+static int
+nouveau_perfctr_ctor(struct nouveau_object *parent,
+		     struct nouveau_object *engine,
+		     struct nouveau_oclass *oclass, void *data, u32 size,
+		     struct nouveau_object **pobject)
+{
+	struct nouveau_perfmon *ppm = (void *)engine;
+	struct nouveau_perfdom *dom = NULL;
+	struct nouveau_perfsig *sig[4] = {};
+	struct nouveau_perfctr *ctr;
+	struct nv_perfctr_class *args = data;
+	int ret, i;
+
+	if (size < sizeof(*args))
+		return -EINVAL;
+
+	for (i = 0; i < ARRAY_SIZE(args->signal) && args->signal[i].name; i++) {
+		sig[i] = nouveau_perfsig_find(ppm, args->signal[i].name,
+					      args->signal[i].size, &dom);
+		if (!sig[i])
+			return -EINVAL;
+	}
+
+	ret = nouveau_object_create(parent, engine, oclass, 0, &ctr);
+	*pobject = nv_object(ctr);
+	if (ret)
+		return ret;
+
+	ctr->slot = -1;
+	ctr->logic_op = args->logic_op;
+	ctr->signal[0] = sig[0];
+	ctr->signal[1] = sig[1];
+	ctr->signal[2] = sig[2];
+	ctr->signal[3] = sig[3];
+	if (dom)
+		list_add_tail(&ctr->head, &dom->list);
+	return 0;
+}
+
+static struct nouveau_ofuncs
+nouveau_perfctr_ofuncs = {
+	.ctor = nouveau_perfctr_ctor,
+	.dtor = nouveau_perfctr_dtor,
+	.init = nouveau_object_init,
+	.fini = nouveau_object_fini,
+};
+
+static struct nouveau_omthds
+nouveau_perfctr_omthds[] = {
+	{ NV_PERFCTR_QUERY, NV_PERFCTR_QUERY, nouveau_perfctr_query },
+	{ NV_PERFCTR_SAMPLE, NV_PERFCTR_SAMPLE, nouveau_perfctr_sample },
+	{ NV_PERFCTR_READ, NV_PERFCTR_READ, nouveau_perfctr_read },
+	{}
+};
+
+struct nouveau_oclass
+nouveau_perfmon_sclass[] = {
+	{ .handle = NV_PERFCTR_CLASS,
+	  .ofuncs = &nouveau_perfctr_ofuncs,
+	  .omthds =  nouveau_perfctr_omthds,
+	},
+	{},
+};
+
+/*******************************************************************************
+ * PPM context
+ ******************************************************************************/
+static void
+nouveau_perfctx_dtor(struct nouveau_object *object)
+{
+	struct nouveau_perfmon *ppm = (void *)object->engine;
+	mutex_lock(&nv_subdev(ppm)->mutex);
+	ppm->context = NULL;
+	mutex_unlock(&nv_subdev(ppm)->mutex);
+}
+
+static int
+nouveau_perfctx_ctor(struct nouveau_object *parent,
+		     struct nouveau_object *engine,
+		     struct nouveau_oclass *oclass, void *data, u32 size,
+		     struct nouveau_object **pobject)
+{
+	struct nouveau_perfmon *ppm = (void *)engine;
+	struct nouveau_perfctx *ctx;
+	int ret;
+
+	ret = nouveau_engctx_create(parent, engine, oclass, NULL,
+				    0, 0, 0, &ctx);
+	*pobject = nv_object(ctx);
+	if (ret)
+		return ret;
+
+	mutex_lock(&nv_subdev(ppm)->mutex);
+	if (ppm->context == NULL)
+		ppm->context = ctx;
+	mutex_unlock(&nv_subdev(ppm)->mutex);
+
+	if (ctx != ppm->context)
+		return -EBUSY;
+
+	return 0;
+}
+
+struct nouveau_oclass
+nouveau_perfmon_cclass = {
+	.handle = NV_ENGCTX(PERFMON, 0x00),
+	.ofuncs = &(struct nouveau_ofuncs) {
+		.ctor = nouveau_perfctx_ctor,
+		.dtor = nouveau_perfctx_dtor,
+		.init = _nouveau_engctx_init,
+		.fini = _nouveau_engctx_fini,
+	},
+};
+
+/*******************************************************************************
+ * PPM engine/subdev functions
+ ******************************************************************************/
+int
+nouveau_perfdom_new(struct nouveau_perfmon *ppm, const char *name, u32 mask,
+		    u32 base, u32 size_unit, u32 size_domain,
+		    const struct nouveau_specdom *spec)
+{
+	const struct nouveau_specdom *sdom;
+	const struct nouveau_specsig *ssig;
+	struct nouveau_perfdom *dom;
+	int i;
+
+	for (i = 0; i == 0 || mask; i++) {
+		u32 addr = base + (i * size_unit);
+		if (i && !(mask & (1 << i)))
+			continue;
+
+		sdom = spec;
+		while (sdom->signal_nr) {
+			dom = kzalloc(sizeof(*dom) + sdom->signal_nr *
+				      sizeof(*dom->signal), GFP_KERNEL);
+			if (!dom)
+				return -ENOMEM;
+
+			if (mask) {
+				snprintf(dom->name, sizeof(dom->name),
+					 "%s/%02x/%02x", name, i,
+					 (int)(sdom - spec));
+			} else {
+				snprintf(dom->name, sizeof(dom->name),
+					 "%s/%02x", name, (int)(sdom - spec));
+			}
+
+			list_add_tail(&dom->head, &ppm->domains);
+			INIT_LIST_HEAD(&dom->list);
+			dom->func = sdom->func;
+			dom->addr = addr;
+			dom->quad = QUAD_MASK;
+			dom->signal_nr = sdom->signal_nr;
+
+			ssig = (sdom++)->signal;
+			while (ssig->name) {
+				dom->signal[ssig->signal].name = ssig->name;
+				ssig++;
+			}
+
+			addr += size_domain;
+		}
+
+		mask &= ~(1 << i);
+	}
+
+	return 0;
+}
+
+int
+_nouveau_perfmon_fini(struct nouveau_object *object, bool suspend)
+{
+	struct nouveau_perfmon *ppm = (void *)object;
+	return nouveau_engine_fini(&ppm->base, suspend);
+}
+
+int
+_nouveau_perfmon_init(struct nouveau_object *object)
+{
+	struct nouveau_perfmon *ppm = (void *)object;
+	return nouveau_engine_init(&ppm->base);
+}
+
+void
+_nouveau_perfmon_dtor(struct nouveau_object *object)
+{
+	struct nouveau_perfmon *ppm = (void *)object;
+	struct nouveau_perfdom *dom, *tmp;
+
+	list_for_each_entry_safe(dom, tmp, &ppm->domains, head) {
+		list_del(&dom->head);
+		kfree(dom);
+	}
+
+	nouveau_engine_destroy(&ppm->base);
+}
+
+int
+nouveau_perfmon_create_(struct nouveau_object *parent,
+			struct nouveau_object *engine,
+			struct nouveau_oclass *oclass,
+			int length, void **pobject)
+{
+	struct nouveau_perfmon *ppm;
+	int ret;
+
+	ret = nouveau_engine_create_(parent, engine, oclass, true, "PPM",
+				     "perfmon", length, pobject);
+	ppm = *pobject;
+	if (ret)
+		return ret;
+
+	INIT_LIST_HEAD(&ppm->domains);
+	return 0;
+}
diff --git a/drivers/gpu/drm/nouveau/core/engine/perfmon/daemon.c b/drivers/gpu/drm/nouveau/core/engine/perfmon/daemon.c
new file mode 100644
index 000000000000..50696cc7b7d7
--- /dev/null
+++ b/drivers/gpu/drm/nouveau/core/engine/perfmon/daemon.c
@@ -0,0 +1,109 @@
+/*
+ * Copyright 2013 Red Hat Inc.
+ *
+ * Permission is hereby granted, free of charge, to any person obtaining a
+ * copy of this software and associated documentation files (the "Software"),
+ * to deal in the Software without restriction, including without limitation
+ * the rights to use, copy, modify, merge, publish, distribute, sublicense,
+ * and/or sell copies of the Software, and to permit persons to whom the
+ * Software is furnished to do so, subject to the following conditions:
+ *
+ * The above copyright notice and this permission notice shall be included in
+ * all copies or substantial portions of the Software.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+ * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+ * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT.  IN NO EVENT SHALL
+ * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR
+ * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE,
+ * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR
+ * OTHER DEALINGS IN THE SOFTWARE.
+ *
+ * Authors: Ben Skeggs
+ */
+
+#include "priv.h"
+
+static void
+pwr_perfctr_init(struct nouveau_perfmon *ppm, struct nouveau_perfdom *dom,
+		 struct nouveau_perfctr *ctr)
+{
+	u32 mask = 0x00000000;
+	u32 ctrl = 0x00000001;
+	int i;
+
+	for (i = 0; i < ARRAY_SIZE(ctr->signal) && ctr->signal[i]; i++)
+		mask |= 1 << (ctr->signal[i] - dom->signal);
+
+	nv_wr32(ppm, 0x10a504 + (ctr->slot * 0x10), mask);
+	nv_wr32(ppm, 0x10a50c + (ctr->slot * 0x10), ctrl);
+	nv_wr32(ppm, 0x10a50c + (ppm->last * 0x10), 0x00000003);
+}
+
+static void
+pwr_perfctr_read(struct nouveau_perfmon *ppm, struct nouveau_perfdom *dom,
+		 struct nouveau_perfctr *ctr)
+{
+	ctr->ctr = ppm->pwr[ctr->slot];
+	ctr->clk = ppm->pwr[ppm->last];
+}
+
+static void
+pwr_perfctr_next(struct nouveau_perfmon *ppm, struct nouveau_perfdom *dom)
+{
+	int i;
+
+	for (i = 0; i <= ppm->last; i++) {
+		ppm->pwr[i] = nv_rd32(ppm, 0x10a508 + (i * 0x10));
+		nv_wr32(ppm, 0x10a508 + (i * 0x10), 0x80000000);
+	}
+}
+
+static const struct nouveau_funcdom
+pwr_perfctr_func = {
+	.init = pwr_perfctr_init,
+	.read = pwr_perfctr_read,
+	.next = pwr_perfctr_next,
+};
+
+const struct nouveau_specdom
+nva3_perfmon_pwr[] = {
+	{ 0x20, (const struct nouveau_specsig[]) {
+			{ 0x00, "pwr_gr_idle" },
+			{ 0x04, "pwr_bsp_idle" },
+			{ 0x05, "pwr_vp_idle" },
+			{ 0x06, "pwr_ppp_idle" },
+			{ 0x13, "pwr_ce0_idle" },
+			{}
+		}, &pwr_perfctr_func },
+	{}
+};
+
+const struct nouveau_specdom
+nvc0_perfmon_pwr[] = {
+	{ 0x20, (const struct nouveau_specsig[]) {
+			{ 0x00, "pwr_gr_idle" },
+			{ 0x04, "pwr_bsp_idle" },
+			{ 0x05, "pwr_vp_idle" },
+			{ 0x06, "pwr_ppp_idle" },
+			{ 0x13, "pwr_ce0_idle" },
+			{ 0x14, "pwr_ce1_idle" },
+			{}
+		}, &pwr_perfctr_func },
+	{}
+};
+
+const struct nouveau_specdom
+nve0_perfmon_pwr[] = {
+	{ 0x20, (const struct nouveau_specsig[]) {
+			{ 0x00, "pwr_gr_idle" },
+			{ 0x04, "pwr_bsp_idle" },
+			{ 0x05, "pwr_vp_idle" },
+			{ 0x06, "pwr_ppp_idle" },
+			{ 0x13, "pwr_ce0_idle" },
+			{ 0x14, "pwr_ce1_idle" },
+			{ 0x15, "pwr_ce2_idle" },
+			{}
+		}, &pwr_perfctr_func },
+	{}
+};
diff --git a/drivers/gpu/drm/nouveau/core/engine/perfmon/nv40.c b/drivers/gpu/drm/nouveau/core/engine/perfmon/nv40.c
new file mode 100644
index 000000000000..b2a10785adb1
--- /dev/null
+++ b/drivers/gpu/drm/nouveau/core/engine/perfmon/nv40.c
@@ -0,0 +1,143 @@
+/*
+ * Copyright 2013 Red Hat Inc.
+ *
+ * Permission is hereby granted, free of charge, to any person obtaining a
+ * copy of this software and associated documentation files (the "Software"),
+ * to deal in the Software without restriction, including without limitation
+ * the rights to use, copy, modify, merge, publish, distribute, sublicense,
+ * and/or sell copies of the Software, and to permit persons to whom the
+ * Software is furnished to do so, subject to the following conditions:
+ *
+ * The above copyright notice and this permission notice shall be included in
+ * all copies or substantial portions of the Software.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+ * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+ * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT.  IN NO EVENT SHALL
+ * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR
+ * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE,
+ * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR
+ * OTHER DEALINGS IN THE SOFTWARE.
+ *
+ * Authors: Ben Skeggs
+ */
+
+#include "nv40.h"
+
+/*******************************************************************************
+ * Perfmon object classes
+ ******************************************************************************/
+
+/*******************************************************************************
+ * PPM context
+ ******************************************************************************/
+
+/*******************************************************************************
+ * PPM engine/subdev functions
+ ******************************************************************************/
+
+static void
+nv40_perfctr_init(struct nouveau_perfmon *ppm, struct nouveau_perfdom *dom,
+		  struct nouveau_perfctr *ctr)
+{
+	struct nv40_perfmon_priv *priv = (void *)ppm;
+	struct nv40_perfmon_cntr *cntr = (void *)ctr;
+	u32 log = ctr->logic_op;
+	u32 src = 0x00000000;
+	int i;
+
+	for (i = 0; i < 4 && ctr->signal[i]; i++)
+		src |= (ctr->signal[i] - dom->signal) << (i * 8);
+
+	nv_wr32(priv, 0x00a7c0 + dom->addr, 0x00000001);
+	nv_wr32(priv, 0x00a400 + dom->addr + (cntr->base.slot * 0x40), src);
+	nv_wr32(priv, 0x00a420 + dom->addr + (cntr->base.slot * 0x40), log);
+}
+
+static void
+nv40_perfctr_read(struct nouveau_perfmon *ppm, struct nouveau_perfdom *dom,
+		  struct nouveau_perfctr *ctr)
+{
+	struct nv40_perfmon_priv *priv = (void *)ppm;
+	struct nv40_perfmon_cntr *cntr = (void *)ctr;
+
+	switch (cntr->base.slot) {
+	case 0: cntr->base.ctr = nv_rd32(priv, 0x00a700 + dom->addr); break;
+	case 1: cntr->base.ctr = nv_rd32(priv, 0x00a6c0 + dom->addr); break;
+	case 2: cntr->base.ctr = nv_rd32(priv, 0x00a680 + dom->addr); break;
+	case 3: cntr->base.ctr = nv_rd32(priv, 0x00a740 + dom->addr); break;
+	}
+	cntr->base.clk = nv_rd32(priv, 0x00a600 + dom->addr);
+}
+
+static void
+nv40_perfctr_next(struct nouveau_perfmon *ppm, struct nouveau_perfdom *dom)
+{
+	struct nv40_perfmon_priv *priv = (void *)ppm;
+	if (priv->sequence != ppm->sequence) {
+		nv_wr32(priv, 0x400084, 0x00000020);
+		priv->sequence = ppm->sequence;
+	}
+}
+
+const struct nouveau_funcdom
+nv40_perfctr_func = {
+	.init = nv40_perfctr_init,
+	.read = nv40_perfctr_read,
+	.next = nv40_perfctr_next,
+};
+
+static const struct nouveau_specdom
+nv40_perfmon[] = {
+	{ 0x20, (const struct nouveau_specsig[]) {
+			{}
+		}, &nv40_perfctr_func },
+	{ 0x20, (const struct nouveau_specsig[]) {
+			{}
+		}, &nv40_perfctr_func },
+	{ 0x20, (const struct nouveau_specsig[]) {
+			{}
+		}, &nv40_perfctr_func },
+	{ 0x20, (const struct nouveau_specsig[]) {
+			{}
+		}, &nv40_perfctr_func },
+	{ 0x20, (const struct nouveau_specsig[]) {
+			{}
+		}, &nv40_perfctr_func },
+	{}
+};
+
+int
+nv40_perfmon_ctor(struct nouveau_object *parent, struct nouveau_object *engine,
+		  struct nouveau_oclass *oclass, void *data, u32 size,
+		  struct nouveau_object **pobject)
+{
+	struct nv40_perfmon_oclass *mclass = (void *)oclass;
+	struct nv40_perfmon_priv *priv;
+	int ret;
+
+	ret = nouveau_perfmon_create(parent, engine, oclass, &priv);
+	*pobject = nv_object(priv);
+	if (ret)
+		return ret;
+
+	ret = nouveau_perfdom_new(&priv->base, "pm", 0, 0, 0, 4, mclass->doms);
+	if (ret)
+		return ret;
+
+	nv_engine(priv)->cclass = &nouveau_perfmon_cclass;
+	nv_engine(priv)->sclass =  nouveau_perfmon_sclass;
+	return 0;
+}
+
+struct nouveau_oclass *
+nv40_perfmon_oclass = &(struct nv40_perfmon_oclass) {
+	.base.handle = NV_ENGINE(PERFMON, 0x40),
+	.base.ofuncs = &(struct nouveau_ofuncs) {
+		.ctor = nv40_perfmon_ctor,
+		.dtor = _nouveau_perfmon_dtor,
+		.init = _nouveau_perfmon_init,
+		.fini = _nouveau_perfmon_fini,
+	},
+	.doms = nv40_perfmon,
+}.base;
diff --git a/drivers/gpu/drm/nouveau/core/engine/perfmon/nv40.h b/drivers/gpu/drm/nouveau/core/engine/perfmon/nv40.h
new file mode 100644
index 000000000000..1b5792d1df14
--- /dev/null
+++ b/drivers/gpu/drm/nouveau/core/engine/perfmon/nv40.h
@@ -0,0 +1,26 @@
+#ifndef __NVKM_PM_NV40_H__
+#define __NVKM_PM_NV40_H__
+
+#include "priv.h"
+
+struct nv40_perfmon_oclass {
+	struct nouveau_oclass base;
+	const struct nouveau_specdom *doms;
+};
+
+struct nv40_perfmon_priv {
+	struct nouveau_perfmon base;
+	u32 sequence;
+};
+
+int nv40_perfmon_ctor(struct nouveau_object *, struct nouveau_object *,
+		      struct nouveau_oclass *, void *data, u32 size,
+		      struct nouveau_object **pobject);
+
+struct nv40_perfmon_cntr {
+	struct nouveau_perfctr base;
+};
+
+extern const struct nouveau_funcdom nv40_perfctr_func;
+
+#endif
diff --git a/drivers/gpu/drm/nouveau/core/engine/perfmon/nv50.c b/drivers/gpu/drm/nouveau/core/engine/perfmon/nv50.c
new file mode 100644
index 000000000000..94217691fe67
--- /dev/null
+++ b/drivers/gpu/drm/nouveau/core/engine/perfmon/nv50.c
@@ -0,0 +1,70 @@
+/*
+ * Copyright 2013 Red Hat Inc.
+ *
+ * Permission is hereby granted, free of charge, to any person obtaining a
+ * copy of this software and associated documentation files (the "Software"),
+ * to deal in the Software without restriction, including without limitation
+ * the rights to use, copy, modify, merge, publish, distribute, sublicense,
+ * and/or sell copies of the Software, and to permit persons to whom the
+ * Software is furnished to do so, subject to the following conditions:
+ *
+ * The above copyright notice and this permission notice shall be included in
+ * all copies or substantial portions of the Software.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+ * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+ * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT.  IN NO EVENT SHALL
+ * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR
+ * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE,
+ * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR
+ * OTHER DEALINGS IN THE SOFTWARE.
+ *
+ * Authors: Ben Skeggs
+ */
+
+#include "nv40.h"
+
+/*******************************************************************************
+ * Perfmon object classes
+ ******************************************************************************/
+
+/*******************************************************************************
+ * PPM context
+ ******************************************************************************/
+
+/*******************************************************************************
+ * PPM engine/subdev functions
+ ******************************************************************************/
+
+static const struct nouveau_specdom
+nv50_perfmon[] = {
+	{ 0x040, (const struct nouveau_specsig[]) {
+			{}
+		}, &nv40_perfctr_func },
+	{ 0x100, (const struct nouveau_specsig[]) {
+			{ 0xc8, "gr_idle" },
+			{}
+		}, &nv40_perfctr_func },
+	{ 0x100, (const struct nouveau_specsig[]) {
+			{}
+		}, &nv40_perfctr_func },
+	{ 0x020, (const struct nouveau_specsig[]) {
+			{}
+		}, &nv40_perfctr_func },
+	{ 0x040, (const struct nouveau_specsig[]) {
+			{}
+		}, &nv40_perfctr_func },
+	{}
+};
+
+struct nouveau_oclass *
+nv50_perfmon_oclass = &(struct nv40_perfmon_oclass) {
+	.base.handle = NV_ENGINE(PERFMON, 0x50),
+	.base.ofuncs = &(struct nouveau_ofuncs) {
+		.ctor = nv40_perfmon_ctor,
+		.dtor = _nouveau_perfmon_dtor,
+		.init = _nouveau_perfmon_init,
+		.fini = _nouveau_perfmon_fini,
+	},
+	.doms = nv50_perfmon,
+}.base;
diff --git a/drivers/gpu/drm/nouveau/core/engine/perfmon/nv84.c b/drivers/gpu/drm/nouveau/core/engine/perfmon/nv84.c
new file mode 100644
index 000000000000..9232c7fc6253
--- /dev/null
+++ b/drivers/gpu/drm/nouveau/core/engine/perfmon/nv84.c
@@ -0,0 +1,78 @@
+/*
+ * Copyright 2013 Red Hat Inc.
+ *
+ * Permission is hereby granted, free of charge, to any person obtaining a
+ * copy of this software and associated documentation files (the "Software"),
+ * to deal in the Software without restriction, including without limitation
+ * the rights to use, copy, modify, merge, publish, distribute, sublicense,
+ * and/or sell copies of the Software, and to permit persons to whom the
+ * Software is furnished to do so, subject to the following conditions:
+ *
+ * The above copyright notice and this permission notice shall be included in
+ * all copies or substantial portions of the Software.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+ * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+ * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT.  IN NO EVENT SHALL
+ * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR
+ * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE,
+ * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR
+ * OTHER DEALINGS IN THE SOFTWARE.
+ *
+ * Authors: Ben Skeggs
+ */
+
+#include "nv40.h"
+
+/*******************************************************************************
+ * Perfmon object classes
+ ******************************************************************************/
+
+/*******************************************************************************
+ * PPM context
+ ******************************************************************************/
+
+/*******************************************************************************
+ * PPM engine/subdev functions
+ ******************************************************************************/
+
+static const struct nouveau_specdom
+nv84_perfmon[] = {
+	{ 0x20, (const struct nouveau_specsig[]) {
+			{}
+		}, &nv40_perfctr_func },
+	{ 0x20, (const struct nouveau_specsig[]) {
+			{}
+		}, &nv40_perfctr_func },
+	{ 0x20, (const struct nouveau_specsig[]) {
+			{}
+		}, &nv40_perfctr_func },
+	{ 0x20, (const struct nouveau_specsig[]) {
+			{}
+		}, &nv40_perfctr_func },
+	{ 0x20, (const struct nouveau_specsig[]) {
+			{}
+		}, &nv40_perfctr_func },
+	{ 0x20, (const struct nouveau_specsig[]) {
+			{}
+		}, &nv40_perfctr_func },
+	{ 0x20, (const struct nouveau_specsig[]) {
+			{}
+		}, &nv40_perfctr_func },
+	{ 0x20, (const struct nouveau_specsig[]) {
+			{}
+		}, &nv40_perfctr_func },
+	{}
+};
+
+struct nouveau_oclass *
+nv84_perfmon_oclass = &(struct nv40_perfmon_oclass) {
+	.base.handle = NV_ENGINE(PERFMON, 0x84),
+	.base.ofuncs = &(struct nouveau_ofuncs) {
+		.ctor = nv40_perfmon_ctor,
+		.dtor = _nouveau_perfmon_dtor,
+		.init = _nouveau_perfmon_init,
+		.fini = _nouveau_perfmon_fini,
+	},
+	.doms = nv84_perfmon,
+}.base;
diff --git a/drivers/gpu/drm/nouveau/core/engine/perfmon/nva3.c b/drivers/gpu/drm/nouveau/core/engine/perfmon/nva3.c
new file mode 100644
index 000000000000..6197ebdeb648
--- /dev/null
+++ b/drivers/gpu/drm/nouveau/core/engine/perfmon/nva3.c
@@ -0,0 +1,96 @@
+/*
+ * Copyright 2013 Red Hat Inc.
+ *
+ * Permission is hereby granted, free of charge, to any person obtaining a
+ * copy of this software and associated documentation files (the "Software"),
+ * to deal in the Software without restriction, including without limitation
+ * the rights to use, copy, modify, merge, publish, distribute, sublicense,
+ * and/or sell copies of the Software, and to permit persons to whom the
+ * Software is furnished to do so, subject to the following conditions:
+ *
+ * The above copyright notice and this permission notice shall be included in
+ * all copies or substantial portions of the Software.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+ * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+ * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT.  IN NO EVENT SHALL
+ * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR
+ * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE,
+ * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR
+ * OTHER DEALINGS IN THE SOFTWARE.
+ *
+ * Authors: Ben Skeggs
+ */
+
+#include "nv40.h"
+
+/*******************************************************************************
+ * Perfmon object classes
+ ******************************************************************************/
+
+/*******************************************************************************
+ * PPM context
+ ******************************************************************************/
+
+/*******************************************************************************
+ * PPM engine/subdev functions
+ ******************************************************************************/
+
+static const struct nouveau_specdom
+nva3_perfmon[] = {
+	{ 0x20, (const struct nouveau_specsig[]) {
+			{}
+		}, &nv40_perfctr_func },
+	{ 0x20, (const struct nouveau_specsig[]) {
+			{}
+		}, &nv40_perfctr_func },
+	{ 0x20, (const struct nouveau_specsig[]) {
+			{}
+		}, &nv40_perfctr_func },
+	{ 0x20, (const struct nouveau_specsig[]) {
+			{}
+		}, &nv40_perfctr_func },
+	{ 0x20, (const struct nouveau_specsig[]) {
+			{}
+		}, &nv40_perfctr_func },
+	{ 0x20, (const struct nouveau_specsig[]) {
+			{}
+		}, &nv40_perfctr_func },
+	{ 0x20, (const struct nouveau_specsig[]) {
+			{}
+		}, &nv40_perfctr_func },
+	{ 0x20, (const struct nouveau_specsig[]) {
+			{}
+		}, &nv40_perfctr_func },
+	{}
+};
+
+static int
+nva3_perfmon_ctor(struct nouveau_object *parent, struct nouveau_object *engine,
+		  struct nouveau_oclass *oclass, void *data, u32 size,
+		  struct nouveau_object **object)
+{
+	int ret = nv40_perfmon_ctor(parent, engine, oclass, data, size, object);
+	if (ret == 0) {
+		struct nv40_perfmon_priv *priv = (void *)*object;
+		ret = nouveau_perfdom_new(&priv->base, "pwr", 0, 0, 0, 0,
+					   nva3_perfmon_pwr);
+		if (ret)
+			return ret;
+
+		priv->base.last = 3;
+	}
+	return ret;
+}
+
+struct nouveau_oclass *
+nva3_perfmon_oclass = &(struct nv40_perfmon_oclass) {
+	.base.handle = NV_ENGINE(PERFMON, 0xa3),
+	.base.ofuncs = &(struct nouveau_ofuncs) {
+		.ctor = nva3_perfmon_ctor,
+		.dtor = _nouveau_perfmon_dtor,
+		.init = _nouveau_perfmon_init,
+		.fini = _nouveau_perfmon_fini,
+	},
+	.doms = nva3_perfmon,
+}.base;
diff --git a/drivers/gpu/drm/nouveau/core/engine/perfmon/nvc0.c b/drivers/gpu/drm/nouveau/core/engine/perfmon/nvc0.c
new file mode 100644
index 000000000000..74b241042502
--- /dev/null
+++ b/drivers/gpu/drm/nouveau/core/engine/perfmon/nvc0.c
@@ -0,0 +1,173 @@
+/*
+ * Copyright 2013 Red Hat Inc.
+ *
+ * Permission is hereby granted, free of charge, to any person obtaining a
+ * copy of this software and associated documentation files (the "Software"),
+ * to deal in the Software without restriction, including without limitation
+ * the rights to use, copy, modify, merge, publish, distribute, sublicense,
+ * and/or sell copies of the Software, and to permit persons to whom the
+ * Software is furnished to do so, subject to the following conditions:
+ *
+ * The above copyright notice and this permission notice shall be included in
+ * all copies or substantial portions of the Software.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+ * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+ * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT.  IN NO EVENT SHALL
+ * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR
+ * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE,
+ * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR
+ * OTHER DEALINGS IN THE SOFTWARE.
+ *
+ * Authors: Ben Skeggs
+ */
+
+#include "nvc0.h"
+
+/*******************************************************************************
+ * Perfmon object classes
+ ******************************************************************************/
+
+/*******************************************************************************
+ * PPM context
+ ******************************************************************************/
+
+/*******************************************************************************
+ * PPM engine/subdev functions
+ ******************************************************************************/
+
+static const struct nouveau_specdom
+nvc0_perfmon_hub[] = {
+	{}
+};
+
+static const struct nouveau_specdom
+nvc0_perfmon_gpc[] = {
+	{}
+};
+
+static const struct nouveau_specdom
+nvc0_perfmon_part[] = {
+	{}
+};
+
+static void
+nvc0_perfctr_init(struct nouveau_perfmon *ppm, struct nouveau_perfdom *dom,
+		  struct nouveau_perfctr *ctr)
+{
+	struct nvc0_perfmon_priv *priv = (void *)ppm;
+	struct nvc0_perfmon_cntr *cntr = (void *)ctr;
+	u32 log = ctr->logic_op;
+	u32 src = 0x00000000;
+	int i;
+
+	for (i = 0; i < 4 && ctr->signal[i]; i++)
+		src |= (ctr->signal[i] - dom->signal) << (i * 8);
+
+	nv_wr32(priv, dom->addr + 0x09c, 0x00040002);
+	nv_wr32(priv, dom->addr + 0x100, 0x00000000);
+	nv_wr32(priv, dom->addr + 0x040 + (cntr->base.slot * 0x08), src);
+	nv_wr32(priv, dom->addr + 0x044 + (cntr->base.slot * 0x08), log);
+}
+
+static void
+nvc0_perfctr_read(struct nouveau_perfmon *ppm, struct nouveau_perfdom *dom,
+		  struct nouveau_perfctr *ctr)
+{
+	struct nvc0_perfmon_priv *priv = (void *)ppm;
+	struct nvc0_perfmon_cntr *cntr = (void *)ctr;
+
+	switch (cntr->base.slot) {
+	case 0: cntr->base.ctr = nv_rd32(priv, dom->addr + 0x08c); break;
+	case 1: cntr->base.ctr = nv_rd32(priv, dom->addr + 0x088); break;
+	case 2: cntr->base.ctr = nv_rd32(priv, dom->addr + 0x080); break;
+	case 3: cntr->base.ctr = nv_rd32(priv, dom->addr + 0x090); break;
+	}
+	cntr->base.clk = nv_rd32(priv, dom->addr + 0x070);
+}
+
+static void
+nvc0_perfctr_next(struct nouveau_perfmon *ppm, struct nouveau_perfdom *dom)
+{
+	struct nvc0_perfmon_priv *priv = (void *)ppm;
+	nv_wr32(priv, dom->addr + 0x06c, dom->signal_nr - 0x40 + 0x27);
+	nv_wr32(priv, dom->addr + 0x0ec, 0x00000011);
+}
+
+const struct nouveau_funcdom
+nvc0_perfctr_func = {
+	.init = nvc0_perfctr_init,
+	.read = nvc0_perfctr_read,
+	.next = nvc0_perfctr_next,
+};
+
+int
+nvc0_perfmon_fini(struct nouveau_object *object, bool suspend)
+{
+	struct nvc0_perfmon_priv *priv = (void *)object;
+	nv_mask(priv, 0x000200, 0x10000000, 0x00000000);
+	nv_mask(priv, 0x000200, 0x10000000, 0x10000000);
+	return nouveau_perfmon_fini(&priv->base, suspend);
+}
+
+static int
+nvc0_perfmon_ctor(struct nouveau_object *parent, struct nouveau_object *engine,
+		  struct nouveau_oclass *oclass, void *data, u32 size,
+		  struct nouveau_object **pobject)
+{
+	struct nvc0_perfmon_priv *priv;
+	u32 mask;
+	int ret;
+
+	ret = nouveau_perfmon_create(parent, engine, oclass, &priv);
+	*pobject = nv_object(priv);
+	if (ret)
+		return ret;
+
+	ret = nouveau_perfdom_new(&priv->base, "pwr", 0, 0, 0, 0,
+				   nvc0_perfmon_pwr);
+	if (ret)
+		return ret;
+
+	/* HUB */
+	ret = nouveau_perfdom_new(&priv->base, "hub", 0, 0x1b0000, 0, 0x200,
+				   nvc0_perfmon_hub);
+	if (ret)
+		return ret;
+
+	/* GPC */
+	mask  = (1 << nv_rd32(priv, 0x022430)) - 1;
+	mask &= ~nv_rd32(priv, 0x022504);
+	mask &= ~nv_rd32(priv, 0x022584);
+
+	ret = nouveau_perfdom_new(&priv->base, "gpc", mask, 0x180000,
+				  0x1000, 0x200, nvc0_perfmon_gpc);
+	if (ret)
+		return ret;
+
+	/* PART */
+	mask  = (1 << nv_rd32(priv, 0x022438)) - 1;
+	mask &= ~nv_rd32(priv, 0x022548);
+	mask &= ~nv_rd32(priv, 0x0225c8);
+
+	ret = nouveau_perfdom_new(&priv->base, "part", mask, 0x1a0000,
+				  0x1000, 0x200, nvc0_perfmon_part);
+	if (ret)
+		return ret;
+
+	nv_engine(priv)->cclass = &nouveau_perfmon_cclass;
+	nv_engine(priv)->sclass =  nouveau_perfmon_sclass;
+	priv->base.last = 7;
+	return 0;
+}
+
+struct nouveau_oclass
+nvc0_perfmon_oclass = {
+	.handle = NV_ENGINE(PERFMON, 0xc0),
+	.ofuncs = &(struct nouveau_ofuncs) {
+		.ctor = nvc0_perfmon_ctor,
+		.dtor = _nouveau_perfmon_dtor,
+		.init = _nouveau_perfmon_init,
+		.fini = nvc0_perfmon_fini,
+	},
+};
diff --git a/drivers/gpu/drm/nouveau/core/engine/perfmon/nvc0.h b/drivers/gpu/drm/nouveau/core/engine/perfmon/nvc0.h
new file mode 100644
index 000000000000..f66bca484263
--- /dev/null
+++ b/drivers/gpu/drm/nouveau/core/engine/perfmon/nvc0.h
@@ -0,0 +1,17 @@
+#ifndef __NVKM_PM_NVC0_H__
+#define __NVKM_PM_NVC0_H__
+
+#include "priv.h"
+
+struct nvc0_perfmon_priv {
+	struct nouveau_perfmon base;
+};
+
+struct nvc0_perfmon_cntr {
+	struct nouveau_perfctr base;
+};
+
+extern const struct nouveau_funcdom nvc0_perfctr_func;
+int nvc0_perfmon_fini(struct nouveau_object *, bool);
+
+#endif
diff --git a/drivers/gpu/drm/nouveau/core/engine/perfmon/nve0.c b/drivers/gpu/drm/nouveau/core/engine/perfmon/nve0.c
new file mode 100644
index 000000000000..71d718c12075
--- /dev/null
+++ b/drivers/gpu/drm/nouveau/core/engine/perfmon/nve0.c
@@ -0,0 +1,162 @@
+/*
+ * Copyright 2013 Red Hat Inc.
+ *
+ * Permission is hereby granted, free of charge, to any person obtaining a
+ * copy of this software and associated documentation files (the "Software"),
+ * to deal in the Software without restriction, including without limitation
+ * the rights to use, copy, modify, merge, publish, distribute, sublicense,
+ * and/or sell copies of the Software, and to permit persons to whom the
+ * Software is furnished to do so, subject to the following conditions:
+ *
+ * The above copyright notice and this permission notice shall be included in
+ * all copies or substantial portions of the Software.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+ * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+ * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT.  IN NO EVENT SHALL
+ * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR
+ * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE,
+ * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR
+ * OTHER DEALINGS IN THE SOFTWARE.
+ *
+ * Authors: Ben Skeggs
+ */
+
+#include "nvc0.h"
+
+/*******************************************************************************
+ * Perfmon object classes
+ ******************************************************************************/
+
+/*******************************************************************************
+ * PPM context
+ ******************************************************************************/
+
+/*******************************************************************************
+ * PPM engine/subdev functions
+ ******************************************************************************/
+
+static const struct nouveau_specdom
+nve0_perfmon_hub[] = {
+	{ 0x60, (const struct nouveau_specsig[]) {
+			{ 0x47, "hub00_user_0" },
+			{}
+		}, &nvc0_perfctr_func },
+	{ 0x40, (const struct nouveau_specsig[]) {
+			{ 0x27, "hub01_user_0" },
+			{}
+		}, &nvc0_perfctr_func },
+	{ 0x60, (const struct nouveau_specsig[]) {
+			{ 0x47, "hub02_user_0" },
+			{}
+		}, &nvc0_perfctr_func },
+	{ 0x60, (const struct nouveau_specsig[]) {
+			{ 0x47, "hub03_user_0" },
+			{}
+		}, &nvc0_perfctr_func },
+	{ 0x40, (const struct nouveau_specsig[]) {
+			{ 0x03, "host_mmio_rd" },
+			{ 0x27, "hub04_user_0" },
+			{}
+		}, &nvc0_perfctr_func },
+	{ 0x60, (const struct nouveau_specsig[]) {
+			{ 0x47, "hub05_user_0" },
+			{}
+		}, &nvc0_perfctr_func },
+	{ 0xc0, (const struct nouveau_specsig[]) {
+			{ 0x74, "host_fb_rd3x" },
+			{ 0x75, "host_fb_rd3x_2" },
+			{ 0xa7, "hub06_user_0" },
+			{}
+		}, &nvc0_perfctr_func },
+	{ 0x60, (const struct nouveau_specsig[]) {
+			{ 0x47, "hub07_user_0" },
+			{}
+		}, &nvc0_perfctr_func },
+	{}
+};
+
+static const struct nouveau_specdom
+nve0_perfmon_gpc[] = {
+	{ 0xe0, (const struct nouveau_specsig[]) {
+			{ 0xc7, "gpc00_user_0" },
+			{}
+		}, &nvc0_perfctr_func },
+	{}
+};
+
+static const struct nouveau_specdom
+nve0_perfmon_part[] = {
+	{ 0x60, (const struct nouveau_specsig[]) {
+			{ 0x47, "part00_user_0" },
+			{}
+		}, &nvc0_perfctr_func },
+	{ 0x60, (const struct nouveau_specsig[]) {
+			{ 0x47, "part01_user_0" },
+			{}
+		}, &nvc0_perfctr_func },
+	{}
+};
+
+static int
+nve0_perfmon_ctor(struct nouveau_object *parent, struct nouveau_object *engine,
+		  struct nouveau_oclass *oclass, void *data, u32 size,
+		  struct nouveau_object **pobject)
+{
+	struct nvc0_perfmon_priv *priv;
+	u32 mask;
+	int ret;
+
+	ret = nouveau_perfmon_create(parent, engine, oclass, &priv);
+	*pobject = nv_object(priv);
+	if (ret)
+		return ret;
+
+	/* PDAEMON */
+	ret = nouveau_perfdom_new(&priv->base, "pwr", 0, 0, 0, 0,
+				   nve0_perfmon_pwr);
+	if (ret)
+		return ret;
+
+	/* HUB */
+	ret = nouveau_perfdom_new(&priv->base, "hub", 0, 0x1b0000, 0, 0x200,
+				   nve0_perfmon_hub);
+	if (ret)
+		return ret;
+
+	/* GPC */
+	mask  = (1 << nv_rd32(priv, 0x022430)) - 1;
+	mask &= ~nv_rd32(priv, 0x022504);
+	mask &= ~nv_rd32(priv, 0x022584);
+
+	ret = nouveau_perfdom_new(&priv->base, "gpc", mask, 0x180000,
+				  0x1000, 0x200, nve0_perfmon_gpc);
+	if (ret)
+		return ret;
+
+	/* PART */
+	mask  = (1 << nv_rd32(priv, 0x022438)) - 1;
+	mask &= ~nv_rd32(priv, 0x022548);
+	mask &= ~nv_rd32(priv, 0x0225c8);
+
+	ret = nouveau_perfdom_new(&priv->base, "part", mask, 0x1a0000,
+				  0x1000, 0x200, nve0_perfmon_part);
+	if (ret)
+		return ret;
+
+	nv_engine(priv)->cclass = &nouveau_perfmon_cclass;
+	nv_engine(priv)->sclass =  nouveau_perfmon_sclass;
+	priv->base.last = 7;
+	return 0;
+}
+
+struct nouveau_oclass
+nve0_perfmon_oclass = {
+	.handle = NV_ENGINE(PERFMON, 0xe0),
+	.ofuncs = &(struct nouveau_ofuncs) {
+		.ctor = nve0_perfmon_ctor,
+		.dtor = _nouveau_perfmon_dtor,
+		.init = _nouveau_perfmon_init,
+		.fini = nvc0_perfmon_fini,
+	},
+};
diff --git a/drivers/gpu/drm/nouveau/core/engine/perfmon/nvf0.c b/drivers/gpu/drm/nouveau/core/engine/perfmon/nvf0.c
new file mode 100644
index 000000000000..47256f78a895
--- /dev/null
+++ b/drivers/gpu/drm/nouveau/core/engine/perfmon/nvf0.c
@@ -0,0 +1,71 @@
+/*
+ * Copyright 2013 Red Hat Inc.
+ *
+ * Permission is hereby granted, free of charge, to any person obtaining a
+ * copy of this software and associated documentation files (the "Software"),
+ * to deal in the Software without restriction, including without limitation
+ * the rights to use, copy, modify, merge, publish, distribute, sublicense,
+ * and/or sell copies of the Software, and to permit persons to whom the
+ * Software is furnished to do so, subject to the following conditions:
+ *
+ * The above copyright notice and this permission notice shall be included in
+ * all copies or substantial portions of the Software.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+ * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+ * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT.  IN NO EVENT SHALL
+ * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR
+ * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE,
+ * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR
+ * OTHER DEALINGS IN THE SOFTWARE.
+ *
+ * Authors: Ben Skeggs
+ */
+
+#include "nvc0.h"
+
+/*******************************************************************************
+ * Perfmon object classes
+ ******************************************************************************/
+
+/*******************************************************************************
+ * PPM context
+ ******************************************************************************/
+
+/*******************************************************************************
+ * PPM engine/subdev functions
+ ******************************************************************************/
+
+static int
+nvf0_perfmon_ctor(struct nouveau_object *parent, struct nouveau_object *engine,
+		  struct nouveau_oclass *oclass, void *data, u32 size,
+		  struct nouveau_object **pobject)
+{
+	struct nvc0_perfmon_priv *priv;
+	int ret;
+
+	ret = nouveau_perfmon_create(parent, engine, oclass, &priv);
+	*pobject = nv_object(priv);
+	if (ret)
+		return ret;
+
+	ret = nouveau_perfdom_new(&priv->base, "pwr", 0, 0, 0, 0,
+				   nve0_perfmon_pwr);
+	if (ret)
+		return ret;
+
+	nv_engine(priv)->cclass = &nouveau_perfmon_cclass;
+	nv_engine(priv)->sclass =  nouveau_perfmon_sclass;
+	return 0;
+}
+
+struct nouveau_oclass
+nvf0_perfmon_oclass = {
+	.handle = NV_ENGINE(PERFMON, 0xf0),
+	.ofuncs = &(struct nouveau_ofuncs) {
+		.ctor = nvf0_perfmon_ctor,
+		.dtor = _nouveau_perfmon_dtor,
+		.init = _nouveau_perfmon_init,
+		.fini = nvc0_perfmon_fini,
+	},
+};
diff --git a/drivers/gpu/drm/nouveau/core/engine/perfmon/priv.h b/drivers/gpu/drm/nouveau/core/engine/perfmon/priv.h
new file mode 100644
index 000000000000..0ac8714fe0ba
--- /dev/null
+++ b/drivers/gpu/drm/nouveau/core/engine/perfmon/priv.h
@@ -0,0 +1,91 @@
+#ifndef __NVKM_PERFMON_PRIV_H__
+#define __NVKM_PERFMON_PRIV_H__
+
+#include <engine/perfmon.h>
+
+struct nouveau_perfctr {
+	struct nouveau_object base;
+	struct list_head head;
+	struct nouveau_perfsig *signal[4];
+	int slot;
+	u32 logic_op;
+	u32 clk;
+	u32 ctr;
+};
+
+extern struct nouveau_oclass nouveau_perfmon_sclass[];
+
+struct nouveau_perfctx {
+	struct nouveau_engctx base;
+};
+
+extern struct nouveau_oclass nouveau_perfmon_cclass;
+
+struct nouveau_specsig {
+	u8 signal;
+	const char *name;
+};
+
+struct nouveau_perfsig {
+	const char *name;
+};
+
+struct nouveau_perfdom;
+struct nouveau_perfctr *
+nouveau_perfsig_wrap(struct nouveau_perfmon *, const char *,
+		     struct nouveau_perfdom **);
+
+struct nouveau_specdom {
+	u16 signal_nr;
+	const struct nouveau_specsig *signal;
+	const struct nouveau_funcdom *func;
+};
+
+extern const struct nouveau_specdom nva3_perfmon_pwr[];
+extern const struct nouveau_specdom nvc0_perfmon_pwr[];
+extern const struct nouveau_specdom nve0_perfmon_pwr[];
+
+struct nouveau_perfdom {
+	struct list_head head;
+	struct list_head list;
+	const struct nouveau_funcdom *func;
+	char name[32];
+	u32 addr;
+	u8  quad;
+	u32 signal_nr;
+	struct nouveau_perfsig signal[];
+};
+
+struct nouveau_funcdom {
+	void (*init)(struct nouveau_perfmon *, struct nouveau_perfdom *,
+		     struct nouveau_perfctr *);
+	void (*read)(struct nouveau_perfmon *, struct nouveau_perfdom *,
+		     struct nouveau_perfctr *);
+	void (*next)(struct nouveau_perfmon *, struct nouveau_perfdom *);
+};
+
+int nouveau_perfdom_new(struct nouveau_perfmon *, const char *, u32,
+			u32, u32, u32, const struct nouveau_specdom *);
+
+#define nouveau_perfmon_create(p,e,o,d)                                        \
+	nouveau_perfmon_create_((p), (e), (o), sizeof(**d), (void **)d)
+#define nouveau_perfmon_dtor(p) ({                                             \
+	struct nouveau_perfmon *c = (p);                                       \
+	_nouveau_perfmon_dtor(nv_object(c));                                   \
+})
+#define nouveau_perfmon_init(p) ({                                             \
+	struct nouveau_perfmon *c = (p);                                       \
+	_nouveau_perfmon_init(nv_object(c));                                   \
+})
+#define nouveau_perfmon_fini(p,s) ({                                           \
+	struct nouveau_perfmon *c = (p);                                       \
+	_nouveau_perfmon_fini(nv_object(c), (s));                              \
+})
+
+int nouveau_perfmon_create_(struct nouveau_object *, struct nouveau_object *,
+			    struct nouveau_oclass *, int, void **);
+void _nouveau_perfmon_dtor(struct nouveau_object *);
+int  _nouveau_perfmon_init(struct nouveau_object *);
+int  _nouveau_perfmon_fini(struct nouveau_object *, bool);
+
+#endif
diff --git a/drivers/gpu/drm/nouveau/core/engine/software/nv04.c b/drivers/gpu/drm/nouveau/core/engine/software/nv04.c
index 2a859a31c30d..c571758e4a27 100644
--- a/drivers/gpu/drm/nouveau/core/engine/software/nv04.c
+++ b/drivers/gpu/drm/nouveau/core/engine/software/nv04.c
@@ -135,8 +135,8 @@ nv04_software_ctor(struct nouveau_object *parent, struct nouveau_object *engine,
 	return 0;
 }
 
-struct nouveau_oclass
-nv04_software_oclass = {
+struct nouveau_oclass *
+nv04_software_oclass = &(struct nouveau_oclass) {
 	.handle = NV_ENGINE(SW, 0x04),
 	.ofuncs = &(struct nouveau_ofuncs) {
 		.ctor = nv04_software_ctor,
diff --git a/drivers/gpu/drm/nouveau/core/engine/software/nv10.c b/drivers/gpu/drm/nouveau/core/engine/software/nv10.c
index a019364b1e13..a62f11a78430 100644
--- a/drivers/gpu/drm/nouveau/core/engine/software/nv10.c
+++ b/drivers/gpu/drm/nouveau/core/engine/software/nv10.c
@@ -117,8 +117,8 @@ nv10_software_ctor(struct nouveau_object *parent, struct nouveau_object *engine,
 	return 0;
 }
 
-struct nouveau_oclass
-nv10_software_oclass = {
+struct nouveau_oclass *
+nv10_software_oclass = &(struct nouveau_oclass) {
 	.handle = NV_ENGINE(SW, 0x10),
 	.ofuncs = &(struct nouveau_ofuncs) {
 		.ctor = nv10_software_ctor,
diff --git a/drivers/gpu/drm/nouveau/core/engine/software/nv50.c b/drivers/gpu/drm/nouveau/core/engine/software/nv50.c
index c48e74953771..b574dd4bb828 100644
--- a/drivers/gpu/drm/nouveau/core/engine/software/nv50.c
+++ b/drivers/gpu/drm/nouveau/core/engine/software/nv50.c
@@ -32,16 +32,9 @@
 
 #include <subdev/bar.h>
 
-#include <engine/software.h>
 #include <engine/disp.h>
 
-struct nv50_software_priv {
-	struct nouveau_software base;
-};
-
-struct nv50_software_chan {
-	struct nouveau_software_chan base;
-};
+#include "nv50.h"
 
 /*******************************************************************************
  * software object classes
@@ -62,7 +55,7 @@ nv50_software_mthd_dma_vblsem(struct nouveau_object *object, u32 mthd,
 
 	if (nv_iclass(handle->object, NV_GPUOBJ_CLASS)) {
 		struct nouveau_gpuobj *gpuobj = nv_gpuobj(handle->object);
-		chan->base.vblank.ctxdma = gpuobj->node->offset >> 4;
+		chan->vblank.ctxdma = gpuobj->node->offset >> 4;
 		ret = 0;
 	}
 	nouveau_namedb_put(handle);
@@ -74,34 +67,33 @@ nv50_software_mthd_vblsem_offset(struct nouveau_object *object, u32 mthd,
 				 void *args, u32 size)
 {
 	struct nv50_software_chan *chan = (void *)nv_engctx(object->parent);
-	chan->base.vblank.offset = *(u32 *)args;
+	chan->vblank.offset = *(u32 *)args;
 	return 0;
 }
 
-static int
+int
 nv50_software_mthd_vblsem_value(struct nouveau_object *object, u32 mthd,
 				void *args, u32 size)
 {
 	struct nv50_software_chan *chan = (void *)nv_engctx(object->parent);
-	chan->base.vblank.value = *(u32 *)args;
+	chan->vblank.value = *(u32 *)args;
 	return 0;
 }
 
-static int
+int
 nv50_software_mthd_vblsem_release(struct nouveau_object *object, u32 mthd,
 				  void *args, u32 size)
 {
 	struct nv50_software_chan *chan = (void *)nv_engctx(object->parent);
-	struct nouveau_disp *disp = nouveau_disp(object);
-	u32 crtc = *(u32 *)args;
-	if (crtc > 1)
+	u32 head = *(u32 *)args;
+	if (head >= chan->vblank.nr_event)
 		return -EINVAL;
 
-	nouveau_event_get(disp->vblank, crtc, &chan->base.vblank.event);
+	nouveau_event_get(chan->vblank.event[head]);
 	return 0;
 }
 
-static int
+int
 nv50_software_mthd_flip(struct nouveau_object *object, u32 mthd,
 			void *args, u32 size)
 {
@@ -132,10 +124,9 @@ nv50_software_sclass[] = {
  ******************************************************************************/
 
 static int
-nv50_software_vblsem_release(struct nouveau_eventh *event, int head)
+nv50_software_vblsem_release(void *data, int head)
 {
-	struct nouveau_software_chan *chan =
-		container_of(event, struct nouveau_software_chan, vblank.event);
+	struct nv50_software_chan *chan = data;
 	struct nv50_software_priv *priv = (void *)nv_object(chan)->engine;
 	struct nouveau_bar *bar = nouveau_bar(priv);
 
@@ -154,45 +145,76 @@ nv50_software_vblsem_release(struct nouveau_eventh *event, int head)
 	return NVKM_EVENT_DROP;
 }
 
-static int
+void
+nv50_software_context_dtor(struct nouveau_object *object)
+{
+	struct nv50_software_chan *chan = (void *)object;
+	int i;
+
+	if (chan->vblank.event) {
+		for (i = 0; i < chan->vblank.nr_event; i++)
+			nouveau_event_ref(NULL, &chan->vblank.event[i]);
+		kfree(chan->vblank.event);
+	}
+
+	nouveau_software_context_destroy(&chan->base);
+}
+
+int
 nv50_software_context_ctor(struct nouveau_object *parent,
 			   struct nouveau_object *engine,
 			   struct nouveau_oclass *oclass, void *data, u32 size,
 			   struct nouveau_object **pobject)
 {
+	struct nouveau_disp *pdisp = nouveau_disp(parent);
+	struct nv50_software_cclass *pclass = (void *)oclass;
 	struct nv50_software_chan *chan;
-	int ret;
+	int ret, i;
 
 	ret = nouveau_software_context_create(parent, engine, oclass, &chan);
 	*pobject = nv_object(chan);
 	if (ret)
 		return ret;
 
-	chan->base.vblank.channel = nv_gpuobj(parent->parent)->addr >> 12;
-	chan->base.vblank.event.func = nv50_software_vblsem_release;
+	chan->vblank.nr_event = pdisp->vblank->index_nr;
+	chan->vblank.event = kzalloc(chan->vblank.nr_event *
+				     sizeof(*chan->vblank.event), GFP_KERNEL);
+	if (!chan->vblank.event)
+		return -ENOMEM;
+
+	for (i = 0; i < chan->vblank.nr_event; i++) {
+		ret = nouveau_event_new(pdisp->vblank, i, pclass->vblank,
+					chan, &chan->vblank.event[i]);
+		if (ret)
+			return ret;
+	}
+
+	chan->vblank.channel = nv_gpuobj(parent->parent)->addr >> 12;
 	return 0;
 }
 
-static struct nouveau_oclass
+static struct nv50_software_cclass
 nv50_software_cclass = {
-	.handle = NV_ENGCTX(SW, 0x50),
-	.ofuncs = &(struct nouveau_ofuncs) {
+	.base.handle = NV_ENGCTX(SW, 0x50),
+	.base.ofuncs = &(struct nouveau_ofuncs) {
 		.ctor = nv50_software_context_ctor,
 		.dtor = _nouveau_software_context_dtor,
 		.init = _nouveau_software_context_init,
 		.fini = _nouveau_software_context_fini,
 	},
+	.vblank = nv50_software_vblsem_release,
 };
 
 /*******************************************************************************
  * software engine/subdev functions
  ******************************************************************************/
 
-static int
+int
 nv50_software_ctor(struct nouveau_object *parent, struct nouveau_object *engine,
 		   struct nouveau_oclass *oclass, void *data, u32 size,
 		   struct nouveau_object **pobject)
 {
+	struct nv50_software_oclass *pclass = (void *)oclass;
 	struct nv50_software_priv *priv;
 	int ret;
 
@@ -201,19 +223,21 @@ nv50_software_ctor(struct nouveau_object *parent, struct nouveau_object *engine,
 	if (ret)
 		return ret;
 
-	nv_engine(priv)->cclass = &nv50_software_cclass;
-	nv_engine(priv)->sclass = nv50_software_sclass;
+	nv_engine(priv)->cclass = pclass->cclass;
+	nv_engine(priv)->sclass = pclass->sclass;
 	nv_subdev(priv)->intr = nv04_software_intr;
 	return 0;
 }
 
-struct nouveau_oclass
-nv50_software_oclass = {
-	.handle = NV_ENGINE(SW, 0x50),
-	.ofuncs = &(struct nouveau_ofuncs) {
+struct nouveau_oclass *
+nv50_software_oclass = &(struct nv50_software_oclass) {
+	.base.handle = NV_ENGINE(SW, 0x50),
+	.base.ofuncs = &(struct nouveau_ofuncs) {
 		.ctor = nv50_software_ctor,
 		.dtor = _nouveau_software_dtor,
 		.init = _nouveau_software_init,
 		.fini = _nouveau_software_fini,
 	},
-};
+	.cclass = &nv50_software_cclass.base,
+	.sclass =  nv50_software_sclass,
+}.base;
diff --git a/drivers/gpu/drm/nouveau/core/engine/software/nv50.h b/drivers/gpu/drm/nouveau/core/engine/software/nv50.h
new file mode 100644
index 000000000000..2de370c21279
--- /dev/null
+++ b/drivers/gpu/drm/nouveau/core/engine/software/nv50.h
@@ -0,0 +1,47 @@
+#ifndef __NVKM_SW_NV50_H__
+#define __NVKM_SW_NV50_H__
+
+#include <engine/software.h>
+
+struct nv50_software_oclass {
+	struct nouveau_oclass base;
+	struct nouveau_oclass *cclass;
+	struct nouveau_oclass *sclass;
+};
+
+struct nv50_software_priv {
+	struct nouveau_software base;
+};
+
+int  nv50_software_ctor(struct nouveau_object *, struct nouveau_object *,
+			struct nouveau_oclass *, void *, u32,
+			struct nouveau_object **);
+
+struct nv50_software_cclass {
+	struct nouveau_oclass base;
+	int (*vblank)(void *, int);
+};
+
+struct nv50_software_chan {
+	struct nouveau_software_chan base;
+	struct {
+		struct nouveau_eventh **event;
+		int nr_event;
+		u32 channel;
+		u32 ctxdma;
+		u64 offset;
+		u32 value;
+	} vblank;
+};
+
+int  nv50_software_context_ctor(struct nouveau_object *,
+				struct nouveau_object *,
+				struct nouveau_oclass *, void *, u32,
+				struct nouveau_object **);
+void nv50_software_context_dtor(struct nouveau_object *);
+
+int nv50_software_mthd_vblsem_value(struct nouveau_object *, u32, void *, u32);
+int nv50_software_mthd_vblsem_release(struct nouveau_object *, u32, void *, u32);
+int nv50_software_mthd_flip(struct nouveau_object *, u32, void *, u32);
+
+#endif
diff --git a/drivers/gpu/drm/nouveau/core/engine/software/nvc0.c b/drivers/gpu/drm/nouveau/core/engine/software/nvc0.c
index d698e710ddd4..f9430c1bf3e5 100644
--- a/drivers/gpu/drm/nouveau/core/engine/software/nvc0.c
+++ b/drivers/gpu/drm/nouveau/core/engine/software/nvc0.c
@@ -32,13 +32,7 @@
 #include <engine/software.h>
 #include <engine/disp.h>
 
-struct nvc0_software_priv {
-	struct nouveau_software base;
-};
-
-struct nvc0_software_chan {
-	struct nouveau_software_chan base;
-};
+#include "nv50.h"
 
 /*******************************************************************************
  * software object classes
@@ -48,58 +42,24 @@ static int
 nvc0_software_mthd_vblsem_offset(struct nouveau_object *object, u32 mthd,
 				 void *args, u32 size)
 {
-	struct nvc0_software_chan *chan = (void *)nv_engctx(object->parent);
+	struct nv50_software_chan *chan = (void *)nv_engctx(object->parent);
 	u64 data = *(u32 *)args;
 	if (mthd == 0x0400) {
-		chan->base.vblank.offset &= 0x00ffffffffULL;
-		chan->base.vblank.offset |= data << 32;
+		chan->vblank.offset &= 0x00ffffffffULL;
+		chan->vblank.offset |= data << 32;
 	} else {
-		chan->base.vblank.offset &= 0xff00000000ULL;
-		chan->base.vblank.offset |= data;
+		chan->vblank.offset &= 0xff00000000ULL;
+		chan->vblank.offset |= data;
 	}
 	return 0;
 }
 
 static int
-nvc0_software_mthd_vblsem_value(struct nouveau_object *object, u32 mthd,
-				void *args, u32 size)
-{
-	struct nvc0_software_chan *chan = (void *)nv_engctx(object->parent);
-	chan->base.vblank.value = *(u32 *)args;
-	return 0;
-}
-
-static int
-nvc0_software_mthd_vblsem_release(struct nouveau_object *object, u32 mthd,
-				  void *args, u32 size)
-{
-	struct nvc0_software_chan *chan = (void *)nv_engctx(object->parent);
-	struct nouveau_disp *disp = nouveau_disp(object);
-	u32 crtc = *(u32 *)args;
-
-	if ((nv_device(object)->card_type < NV_E0 && crtc > 1) || crtc > 3)
-		return -EINVAL;
-
-	nouveau_event_get(disp->vblank, crtc, &chan->base.vblank.event);
-	return 0;
-}
-
-static int
-nvc0_software_mthd_flip(struct nouveau_object *object, u32 mthd,
-			void *args, u32 size)
-{
-	struct nvc0_software_chan *chan = (void *)nv_engctx(object->parent);
-	if (chan->base.flip)
-		return chan->base.flip(chan->base.flip_data);
-	return -EINVAL;
-}
-
-static int
 nvc0_software_mthd_mp_control(struct nouveau_object *object, u32 mthd,
                               void *args, u32 size)
 {
-	struct nvc0_software_chan *chan = (void *)nv_engctx(object->parent);
-	struct nvc0_software_priv *priv = (void *)nv_object(chan)->engine;
+	struct nv50_software_chan *chan = (void *)nv_engctx(object->parent);
+	struct nv50_software_priv *priv = (void *)nv_object(chan)->engine;
 	u32 data = *(u32 *)args;
 
 	switch (mthd) {
@@ -124,9 +84,9 @@ static struct nouveau_omthds
 nvc0_software_omthds[] = {
 	{ 0x0400, 0x0400, nvc0_software_mthd_vblsem_offset },
 	{ 0x0404, 0x0404, nvc0_software_mthd_vblsem_offset },
-	{ 0x0408, 0x0408, nvc0_software_mthd_vblsem_value },
-	{ 0x040c, 0x040c, nvc0_software_mthd_vblsem_release },
-	{ 0x0500, 0x0500, nvc0_software_mthd_flip },
+	{ 0x0408, 0x0408, nv50_software_mthd_vblsem_value },
+	{ 0x040c, 0x040c, nv50_software_mthd_vblsem_release },
+	{ 0x0500, 0x0500, nv50_software_mthd_flip },
 	{ 0x0600, 0x0600, nvc0_software_mthd_mp_control },
 	{ 0x0644, 0x0644, nvc0_software_mthd_mp_control },
 	{ 0x06ac, 0x06ac, nvc0_software_mthd_mp_control },
@@ -144,11 +104,10 @@ nvc0_software_sclass[] = {
  ******************************************************************************/
 
 static int
-nvc0_software_vblsem_release(struct nouveau_eventh *event, int head)
+nvc0_software_vblsem_release(void *data, int head)
 {
-	struct nouveau_software_chan *chan =
-		container_of(event, struct nouveau_software_chan, vblank.event);
-	struct nvc0_software_priv *priv = (void *)nv_object(chan)->engine;
+	struct nv50_software_chan *chan = data;
+	struct nv50_software_priv *priv = (void *)nv_object(chan)->engine;
 	struct nouveau_bar *bar = nouveau_bar(priv);
 
 	nv_wr32(priv, 0x001718, 0x80000000 | chan->vblank.channel);
@@ -160,66 +119,31 @@ nvc0_software_vblsem_release(struct nouveau_eventh *event, int head)
 	return NVKM_EVENT_DROP;
 }
 
-static int
-nvc0_software_context_ctor(struct nouveau_object *parent,
-			   struct nouveau_object *engine,
-			   struct nouveau_oclass *oclass, void *data, u32 size,
-			   struct nouveau_object **pobject)
-{
-	struct nvc0_software_chan *chan;
-	int ret;
-
-	ret = nouveau_software_context_create(parent, engine, oclass, &chan);
-	*pobject = nv_object(chan);
-	if (ret)
-		return ret;
-
-	chan->base.vblank.channel = nv_gpuobj(parent->parent)->addr >> 12;
-	chan->base.vblank.event.func = nvc0_software_vblsem_release;
-	return 0;
-}
-
-static struct nouveau_oclass
+static struct nv50_software_cclass
 nvc0_software_cclass = {
-	.handle = NV_ENGCTX(SW, 0xc0),
-	.ofuncs = &(struct nouveau_ofuncs) {
-		.ctor = nvc0_software_context_ctor,
+	.base.handle = NV_ENGCTX(SW, 0xc0),
+	.base.ofuncs = &(struct nouveau_ofuncs) {
+		.ctor = nv50_software_context_ctor,
 		.dtor = _nouveau_software_context_dtor,
 		.init = _nouveau_software_context_init,
 		.fini = _nouveau_software_context_fini,
 	},
+	.vblank = nvc0_software_vblsem_release,
 };
 
 /*******************************************************************************
  * software engine/subdev functions
  ******************************************************************************/
 
-static int
-nvc0_software_ctor(struct nouveau_object *parent, struct nouveau_object *engine,
-		   struct nouveau_oclass *oclass, void *data, u32 size,
-		   struct nouveau_object **pobject)
-{
-	struct nvc0_software_priv *priv;
-	int ret;
-
-	ret = nouveau_software_create(parent, engine, oclass, &priv);
-	*pobject = nv_object(priv);
-	if (ret)
-		return ret;
-
-	nv_engine(priv)->cclass = &nvc0_software_cclass;
-	nv_engine(priv)->sclass = nvc0_software_sclass;
-	nv_subdev(priv)->intr = nv04_software_intr;
-	return 0;
-}
-
-struct nouveau_oclass
-nvc0_software_oclass = {
-	.handle = NV_ENGINE(SW, 0xc0),
-	.ofuncs = &(struct nouveau_ofuncs) {
-		.ctor = nvc0_software_ctor,
+struct nouveau_oclass *
+nvc0_software_oclass = &(struct nv50_software_oclass) {
+	.base.handle = NV_ENGINE(SW, 0xc0),
+	.base.ofuncs = &(struct nouveau_ofuncs) {
+		.ctor = nv50_software_ctor,
 		.dtor = _nouveau_software_dtor,
 		.init = _nouveau_software_init,
 		.fini = _nouveau_software_fini,
 	},
-};
+	.cclass = &nvc0_software_cclass.base,
+	.sclass =  nvc0_software_sclass,
+}.base;
diff --git a/drivers/gpu/drm/nouveau/core/include/core/class.h b/drivers/gpu/drm/nouveau/core/include/core/class.h
index 5a5961b6a6a3..560c3593dae7 100644
--- a/drivers/gpu/drm/nouveau/core/include/core/class.h
+++ b/drivers/gpu/drm/nouveau/core/include/core/class.h
@@ -22,7 +22,7 @@
 #define NV_DEVICE_DISABLE_PPP                             0x0000004000000000ULL
 #define NV_DEVICE_DISABLE_COPY0                           0x0000008000000000ULL
 #define NV_DEVICE_DISABLE_COPY1                           0x0000010000000000ULL
-#define NV_DEVICE_DISABLE_UNK1C1                          0x0000020000000000ULL
+#define NV_DEVICE_DISABLE_VIC                             0x0000020000000000ULL
 #define NV_DEVICE_DISABLE_VENC                            0x0000040000000000ULL
 
 struct nv_device_class {
@@ -98,6 +98,77 @@ struct nv_dma_class {
 	u32 conf0;
 };
 
+/* Perfmon counter class
+ *
+ * XXXX: NV_PERFCTR
+ */
+#define NV_PERFCTR_CLASS                                             0x0000ffff
+#define NV_PERFCTR_QUERY                                             0x00000000
+#define NV_PERFCTR_SAMPLE                                            0x00000001
+#define NV_PERFCTR_READ                                              0x00000002
+
+struct nv_perfctr_class {
+	u16 logic_op;
+	struct {
+		char __user *name; /*XXX: use cfu when exposed to userspace */
+		u32 size;
+	} signal[4];
+};
+
+struct nv_perfctr_query {
+	u32 iter;
+	u32 size;
+	char __user *name; /*XXX: use ctu when exposed to userspace */
+};
+
+struct nv_perfctr_sample {
+};
+
+struct nv_perfctr_read {
+	u32 ctr;
+	u32 clk;
+};
+
+/* Device control class
+ *
+ * XXXX: NV_CONTROL
+ */
+#define NV_CONTROL_CLASS                                             0x0000fffe
+
+#define NV_CONTROL_PSTATE_INFO                                       0x00000000
+#define NV_CONTROL_PSTATE_INFO_USTATE_DISABLE                              (-1)
+#define NV_CONTROL_PSTATE_INFO_USTATE_PERFMON                              (-2)
+#define NV_CONTROL_PSTATE_INFO_PSTATE_UNKNOWN                              (-1)
+#define NV_CONTROL_PSTATE_INFO_PSTATE_PERFMON                              (-2)
+#define NV_CONTROL_PSTATE_ATTR                                       0x00000001
+#define NV_CONTROL_PSTATE_ATTR_STATE_CURRENT                               (-1)
+#define NV_CONTROL_PSTATE_USER                                       0x00000002
+#define NV_CONTROL_PSTATE_USER_STATE_UNKNOWN                               (-1)
+#define NV_CONTROL_PSTATE_USER_STATE_PERFMON                               (-2)
+
+struct nv_control_pstate_info {
+	u32 count; /* out: number of power states */
+	s32 ustate; /* out: current target pstate index */
+	u32 pstate; /* out: current pstate index */
+};
+
+struct nv_control_pstate_attr {
+	s32 state; /*  in: index of pstate to query
+		    * out: pstate identifier
+		    */
+	u32 index; /*  in: index of attribute to query
+		    * out: index of next attribute, or 0 if no more
+		    */
+	char name[32];
+	char unit[16];
+	u32 min;
+	u32 max;
+};
+
+struct nv_control_pstate_user {
+	s32 state; /*  in: pstate identifier */
+};
+
 /* DMA FIFO channel classes
  *
  * 006b: NV03_CHANNEL_DMA
diff --git a/drivers/gpu/drm/nouveau/core/include/core/debug.h b/drivers/gpu/drm/nouveau/core/include/core/debug.h
index 9ea18dfcb4d0..8092e2e90323 100644
--- a/drivers/gpu/drm/nouveau/core/include/core/debug.h
+++ b/drivers/gpu/drm/nouveau/core/include/core/debug.h
@@ -1,13 +1,20 @@
 #ifndef __NOUVEAU_DEBUG_H__
 #define __NOUVEAU_DEBUG_H__
 
+extern int nv_info_debug_level;
+
 #define NV_DBG_FATAL    0
 #define NV_DBG_ERROR    1
 #define NV_DBG_WARN     2
-#define NV_DBG_INFO     3
+#define NV_DBG_INFO     nv_info_debug_level
 #define NV_DBG_DEBUG    4
 #define NV_DBG_TRACE    5
 #define NV_DBG_PARANOIA 6
 #define NV_DBG_SPAM     7
 
+#define NV_DBG_INFO_NORMAL 3
+#define NV_DBG_INFO_SILENT NV_DBG_DEBUG
+
+#define nv_debug_level(a) nv_info_debug_level = NV_DBG_INFO_##a
+
 #endif
diff --git a/drivers/gpu/drm/nouveau/core/include/core/device.h b/drivers/gpu/drm/nouveau/core/include/core/device.h
index 99b6600fe80a..ac2881d1776a 100644
--- a/drivers/gpu/drm/nouveau/core/include/core/device.h
+++ b/drivers/gpu/drm/nouveau/core/include/core/device.h
@@ -33,9 +33,10 @@ enum nv_subdev_type {
 	NVDEV_SUBDEV_INSTMEM,
 	NVDEV_SUBDEV_VM,
 	NVDEV_SUBDEV_BAR,
+	NVDEV_SUBDEV_PWR,
 	NVDEV_SUBDEV_VOLT,
-	NVDEV_SUBDEV_CLOCK,
 	NVDEV_SUBDEV_THERM,
+	NVDEV_SUBDEV_CLOCK,
 
 	NVDEV_ENGINE_DMAOBJ,
 	NVDEV_ENGINE_FIFO,
@@ -50,9 +51,10 @@ enum nv_subdev_type {
 	NVDEV_ENGINE_COPY0,
 	NVDEV_ENGINE_COPY1,
 	NVDEV_ENGINE_COPY2,
-	NVDEV_ENGINE_UNK1C1,
+	NVDEV_ENGINE_VIC,
 	NVDEV_ENGINE_VENC,
 	NVDEV_ENGINE_DISP,
+	NVDEV_ENGINE_PERFMON,
 
 	NVDEV_SUBDEV_NR,
 };
@@ -72,6 +74,7 @@ struct nouveau_device {
 	enum {
 		NV_04    = 0x04,
 		NV_10    = 0x10,
+		NV_11    = 0x11,
 		NV_20    = 0x20,
 		NV_30    = 0x30,
 		NV_40    = 0x40,
diff --git a/drivers/gpu/drm/nouveau/core/include/core/event.h b/drivers/gpu/drm/nouveau/core/include/core/event.h
index 9e094408f14e..5d539ebff3ed 100644
--- a/drivers/gpu/drm/nouveau/core/include/core/event.h
+++ b/drivers/gpu/drm/nouveau/core/include/core/event.h
@@ -5,13 +5,21 @@
 #define NVKM_EVENT_DROP 0
 #define NVKM_EVENT_KEEP 1
 
+/* nouveau_eventh.flags bit #s */
+#define NVKM_EVENT_ENABLE 0
+
 struct nouveau_eventh {
+	struct nouveau_event *event;
 	struct list_head head;
-	int (*func)(struct nouveau_eventh *, int index);
+	unsigned long flags;
+	int index;
+	int (*func)(void *, int);
+	void *priv;
 };
 
 struct nouveau_event {
-	spinlock_t lock;
+	spinlock_t list_lock;
+	spinlock_t refs_lock;
 
 	void *priv;
 	void (*enable)(struct nouveau_event *, int index);
@@ -28,9 +36,11 @@ int  nouveau_event_create(int index_nr, struct nouveau_event **);
 void nouveau_event_destroy(struct nouveau_event **);
 void nouveau_event_trigger(struct nouveau_event *, int index);
 
-void nouveau_event_get(struct nouveau_event *, int index,
-		       struct nouveau_eventh *);
-void nouveau_event_put(struct nouveau_event *, int index,
-		       struct nouveau_eventh *);
+int  nouveau_event_new(struct nouveau_event *, int index,
+		       int (*func)(void *, int), void *,
+		       struct nouveau_eventh **);
+void nouveau_event_ref(struct nouveau_eventh *, struct nouveau_eventh **);
+void nouveau_event_get(struct nouveau_eventh *);
+void nouveau_event_put(struct nouveau_eventh *);
 
 #endif
diff --git a/drivers/gpu/drm/nouveau/core/include/core/option.h b/drivers/gpu/drm/nouveau/core/include/core/option.h
index 27074957fd21..ed055847887e 100644
--- a/drivers/gpu/drm/nouveau/core/include/core/option.h
+++ b/drivers/gpu/drm/nouveau/core/include/core/option.h
@@ -8,4 +8,13 @@ bool nouveau_boolopt(const char *optstr, const char *opt, bool value);
 
 int nouveau_dbgopt(const char *optstr, const char *sub);
 
+/* compares unterminated string 'str' with zero-terminated string 'cmp' */
+static inline int
+strncasecmpz(const char *str, const char *cmp, size_t len)
+{
+	if (strlen(cmp) != len)
+		return len;
+	return strncasecmp(str, cmp, len);
+}
+
 #endif
diff --git a/drivers/gpu/drm/nouveau/core/include/core/printk.h b/drivers/gpu/drm/nouveau/core/include/core/printk.h
index d87836e3a704..0f9a37bd32b0 100644
--- a/drivers/gpu/drm/nouveau/core/include/core/printk.h
+++ b/drivers/gpu/drm/nouveau/core/include/core/printk.h
@@ -6,27 +6,12 @@
 
 struct nouveau_object;
 
-#define NV_PRINTK_FATAL    KERN_CRIT
-#define NV_PRINTK_ERROR    KERN_ERR
-#define NV_PRINTK_WARN     KERN_WARNING
-#define NV_PRINTK_INFO     KERN_INFO
-#define NV_PRINTK_DEBUG    KERN_DEBUG
-#define NV_PRINTK_PARANOIA KERN_DEBUG
-#define NV_PRINTK_TRACE    KERN_DEBUG
-#define NV_PRINTK_SPAM     KERN_DEBUG
-
-extern int nv_printk_suspend_level;
-
-#define NV_DBG_SUSPEND (nv_printk_suspend_level)
-#define NV_PRINTK_SUSPEND  (nv_printk_level_to_pfx(nv_printk_suspend_level))
-
-const char *nv_printk_level_to_pfx(int level);
-void __printf(4, 5)
-nv_printk_(struct nouveau_object *, const char *, int, const char *, ...);
+void __printf(3, 4)
+nv_printk_(struct nouveau_object *, int, const char *, ...);
 
 #define nv_printk(o,l,f,a...) do {                                             \
 	if (NV_DBG_##l <= CONFIG_NOUVEAU_DEBUG)                                \
-		nv_printk_(nv_object(o), NV_PRINTK_##l, NV_DBG_##l, f, ##a);   \
+		nv_printk_(nv_object(o), NV_DBG_##l, f, ##a);                  \
 } while(0)
 
 #define nv_fatal(o,f,a...) nv_printk((o), FATAL, f, ##a)
@@ -37,16 +22,9 @@ nv_printk_(struct nouveau_object *, const char *, int, const char *, ...);
 #define nv_trace(o,f,a...) nv_printk((o), TRACE, f, ##a)
 #define nv_spam(o,f,a...) nv_printk((o), SPAM, f, ##a)
 
-#define nv_suspend(o,f,a...) nv_printk((o), SUSPEND, f, ##a)
-
-static inline void nv_suspend_set_printk_level(int level)
-{
-	nv_printk_suspend_level = level;
-}
-
 #define nv_assert(f,a...) do {                                                 \
 	if (NV_DBG_FATAL <= CONFIG_NOUVEAU_DEBUG)                              \
-		nv_printk_(NULL, NV_PRINTK_FATAL, NV_DBG_FATAL, f "\n", ##a);  \
+		nv_printk_(NULL, NV_DBG_FATAL, f "\n", ##a);                   \
 	BUG_ON(1);                                                             \
 } while(0)
 
diff --git a/drivers/gpu/drm/nouveau/core/include/engine/fifo.h b/drivers/gpu/drm/nouveau/core/include/engine/fifo.h
index 633c2f806482..8c32cf4d83c7 100644
--- a/drivers/gpu/drm/nouveau/core/include/engine/fifo.h
+++ b/drivers/gpu/drm/nouveau/core/include/engine/fifo.h
@@ -101,14 +101,14 @@ nouveau_client_name_for_fifo_chid(struct nouveau_fifo *fifo, u32 chid);
 #define _nouveau_fifo_init _nouveau_engine_init
 #define _nouveau_fifo_fini _nouveau_engine_fini
 
-extern struct nouveau_oclass nv04_fifo_oclass;
-extern struct nouveau_oclass nv10_fifo_oclass;
-extern struct nouveau_oclass nv17_fifo_oclass;
-extern struct nouveau_oclass nv40_fifo_oclass;
-extern struct nouveau_oclass nv50_fifo_oclass;
-extern struct nouveau_oclass nv84_fifo_oclass;
-extern struct nouveau_oclass nvc0_fifo_oclass;
-extern struct nouveau_oclass nve0_fifo_oclass;
+extern struct nouveau_oclass *nv04_fifo_oclass;
+extern struct nouveau_oclass *nv10_fifo_oclass;
+extern struct nouveau_oclass *nv17_fifo_oclass;
+extern struct nouveau_oclass *nv40_fifo_oclass;
+extern struct nouveau_oclass *nv50_fifo_oclass;
+extern struct nouveau_oclass *nv84_fifo_oclass;
+extern struct nouveau_oclass *nvc0_fifo_oclass;
+extern struct nouveau_oclass *nve0_fifo_oclass;
 
 void nv04_fifo_intr(struct nouveau_subdev *);
 int  nv04_fifo_context_attach(struct nouveau_object *, struct nouveau_object *);
diff --git a/drivers/gpu/drm/nouveau/core/include/engine/mpeg.h b/drivers/gpu/drm/nouveau/core/include/engine/mpeg.h
index 1d1a89a06ee4..9b0d938199f6 100644
--- a/drivers/gpu/drm/nouveau/core/include/engine/mpeg.h
+++ b/drivers/gpu/drm/nouveau/core/include/engine/mpeg.h
@@ -42,10 +42,13 @@ struct nouveau_mpeg {
 
 extern struct nouveau_oclass nv31_mpeg_oclass;
 extern struct nouveau_oclass nv40_mpeg_oclass;
+extern struct nouveau_oclass nv44_mpeg_oclass;
 extern struct nouveau_oclass nv50_mpeg_oclass;
 extern struct nouveau_oclass nv84_mpeg_oclass;
-
+extern struct nouveau_ofuncs nv31_mpeg_ofuncs;
+extern struct nouveau_oclass nv31_mpeg_cclass;
 extern struct nouveau_oclass nv31_mpeg_sclass[];
+extern struct nouveau_oclass nv40_mpeg_sclass[];
 void nv31_mpeg_intr(struct nouveau_subdev *);
 void nv31_mpeg_tile_prog(struct nouveau_engine *, int);
 int  nv31_mpeg_init(struct nouveau_object *);
diff --git a/drivers/gpu/drm/nouveau/core/include/engine/perfmon.h b/drivers/gpu/drm/nouveau/core/include/engine/perfmon.h
new file mode 100644
index 000000000000..49b0024910fe
--- /dev/null
+++ b/drivers/gpu/drm/nouveau/core/include/engine/perfmon.h
@@ -0,0 +1,39 @@
+#ifndef __NVKM_PERFMON_H__
+#define __NVKM_PERFMON_H__
+
+#include <core/device.h>
+#include <core/engine.h>
+#include <core/engctx.h>
+#include <core/class.h>
+
+struct nouveau_perfdom;
+struct nouveau_perfctr;
+struct nouveau_perfmon {
+	struct nouveau_engine base;
+
+	struct nouveau_perfctx *context;
+	void *profile_data;
+
+	struct list_head domains;
+	u32 sequence;
+
+	/*XXX: temp for daemon backend */
+	u32 pwr[8];
+	u32 last;
+};
+
+static inline struct nouveau_perfmon *
+nouveau_perfmon(void *obj)
+{
+	return (void *)nv_device(obj)->subdev[NVDEV_ENGINE_PERFMON];
+}
+
+extern struct nouveau_oclass *nv40_perfmon_oclass;
+extern struct nouveau_oclass *nv50_perfmon_oclass;
+extern struct nouveau_oclass *nv84_perfmon_oclass;
+extern struct nouveau_oclass *nva3_perfmon_oclass;
+extern struct nouveau_oclass nvc0_perfmon_oclass;
+extern struct nouveau_oclass nve0_perfmon_oclass;
+extern struct nouveau_oclass nvf0_perfmon_oclass;
+
+#endif
diff --git a/drivers/gpu/drm/nouveau/core/include/engine/software.h b/drivers/gpu/drm/nouveau/core/include/engine/software.h
index 45799487e573..23a462b50d03 100644
--- a/drivers/gpu/drm/nouveau/core/include/engine/software.h
+++ b/drivers/gpu/drm/nouveau/core/include/engine/software.h
@@ -3,19 +3,10 @@
 
 #include <core/engine.h>
 #include <core/engctx.h>
-#include <core/event.h>
 
 struct nouveau_software_chan {
 	struct nouveau_engctx base;
 
-	struct {
-		struct nouveau_eventh event;
-		u32 channel;
-		u32 ctxdma;
-		u64 offset;
-		u32 value;
-	} vblank;
-
 	int (*flip)(void *);
 	void *flip_data;
 };
@@ -50,10 +41,10 @@ struct nouveau_software {
 #define _nouveau_software_init _nouveau_engine_init
 #define _nouveau_software_fini _nouveau_engine_fini
 
-extern struct nouveau_oclass nv04_software_oclass;
-extern struct nouveau_oclass nv10_software_oclass;
-extern struct nouveau_oclass nv50_software_oclass;
-extern struct nouveau_oclass nvc0_software_oclass;
+extern struct nouveau_oclass *nv04_software_oclass;
+extern struct nouveau_oclass *nv10_software_oclass;
+extern struct nouveau_oclass *nv50_software_oclass;
+extern struct nouveau_oclass *nvc0_software_oclass;
 
 void nv04_software_intr(struct nouveau_subdev *);
 
diff --git a/drivers/gpu/drm/nouveau/core/include/subdev/bios/boost.h b/drivers/gpu/drm/nouveau/core/include/subdev/bios/boost.h
new file mode 100644
index 000000000000..662b20726851
--- /dev/null
+++ b/drivers/gpu/drm/nouveau/core/include/subdev/bios/boost.h
@@ -0,0 +1,29 @@
+#ifndef __NVBIOS_BOOST_H__
+#define __NVBIOS_BOOST_H__
+
+u16 nvbios_boostTe(struct nouveau_bios *, u8 *, u8 *, u8 *, u8 *, u8 *, u8 *);
+
+struct nvbios_boostE {
+	u8  pstate;
+	u32 min;
+	u32 max;
+};
+
+u16 nvbios_boostEe(struct nouveau_bios *, int idx, u8 *, u8 *, u8 *, u8 *);
+u16 nvbios_boostEp(struct nouveau_bios *, int idx, u8 *, u8 *, u8 *, u8 *,
+		   struct nvbios_boostE *);
+u16 nvbios_boostEm(struct nouveau_bios *, u8, u8 *, u8 *, u8 *, u8 *,
+		   struct nvbios_boostE *);
+
+struct nvbios_boostS {
+	u8  domain;
+	u8  percent;
+	u32 min;
+	u32 max;
+};
+
+u16 nvbios_boostSe(struct nouveau_bios *, int, u16, u8 *, u8 *, u8, u8);
+u16 nvbios_boostSp(struct nouveau_bios *, int, u16, u8 *, u8 *, u8, u8,
+		   struct nvbios_boostS *);
+
+#endif
diff --git a/drivers/gpu/drm/nouveau/core/include/subdev/bios/cstep.h b/drivers/gpu/drm/nouveau/core/include/subdev/bios/cstep.h
new file mode 100644
index 000000000000..a80a43809883
--- /dev/null
+++ b/drivers/gpu/drm/nouveau/core/include/subdev/bios/cstep.h
@@ -0,0 +1,28 @@
+#ifndef __NVBIOS_CSTEP_H__
+#define __NVBIOS_CSTEP_H__
+
+u16 nvbios_cstepTe(struct nouveau_bios *,
+		   u8 *ver, u8 *hdr, u8 *cnt, u8 *len, u8 *xnr, u8 *xsz);
+
+struct nvbios_cstepE {
+	u8  pstate;
+	u8  index;
+};
+
+u16 nvbios_cstepEe(struct nouveau_bios *, int idx, u8 *ver, u8 *hdr);
+u16 nvbios_cstepEp(struct nouveau_bios *, int idx, u8 *ver, u8 *hdr,
+		   struct nvbios_cstepE *);
+u16 nvbios_cstepEm(struct nouveau_bios *, u8 pstate, u8 *ver, u8 *hdr,
+		   struct nvbios_cstepE *);
+
+struct nvbios_cstepX {
+	u32 freq;
+	u8  unkn[2];
+	u8  voltage;
+};
+
+u16 nvbios_cstepXe(struct nouveau_bios *, int idx, u8 *ver, u8 *hdr);
+u16 nvbios_cstepXp(struct nouveau_bios *, int idx, u8 *ver, u8 *hdr,
+		   struct nvbios_cstepX *);
+
+#endif
diff --git a/drivers/gpu/drm/nouveau/core/include/subdev/bios/gpio.h b/drivers/gpu/drm/nouveau/core/include/subdev/bios/gpio.h
index 96d3364f6db3..c7b2e586be0b 100644
--- a/drivers/gpu/drm/nouveau/core/include/subdev/bios/gpio.h
+++ b/drivers/gpu/drm/nouveau/core/include/subdev/bios/gpio.h
@@ -7,7 +7,15 @@ enum dcb_gpio_func_name {
 	DCB_GPIO_TVDAC1 = 0x2d,
 	DCB_GPIO_FAN = 0x09,
 	DCB_GPIO_FAN_SENSE = 0x3d,
-	DCB_GPIO_UNUSED = 0xff
+	DCB_GPIO_UNUSED = 0xff,
+	DCB_GPIO_VID0 = 0x04,
+	DCB_GPIO_VID1 = 0x05,
+	DCB_GPIO_VID2 = 0x06,
+	DCB_GPIO_VID3 = 0x1a,
+	DCB_GPIO_VID4 = 0x73,
+	DCB_GPIO_VID5 = 0x74,
+	DCB_GPIO_VID6 = 0x75,
+	DCB_GPIO_VID7 = 0x76,
 };
 
 #define DCB_GPIO_LOG_DIR     0x02
diff --git a/drivers/gpu/drm/nouveau/core/include/subdev/bios/perf.h b/drivers/gpu/drm/nouveau/core/include/subdev/bios/perf.h
index 0b285e99be5a..16ff06ec2a88 100644
--- a/drivers/gpu/drm/nouveau/core/include/subdev/bios/perf.h
+++ b/drivers/gpu/drm/nouveau/core/include/subdev/bios/perf.h
@@ -3,6 +3,39 @@
 
 struct nouveau_bios;
 
+u16 nvbios_perf_table(struct nouveau_bios *, u8 *ver, u8 *hdr,
+		      u8 *cnt, u8 *len, u8 *snr, u8 *ssz);
+
+struct nvbios_perfE {
+	u8  pstate;
+	u8  fanspeed;
+	u8  voltage;
+	u32 core;
+	u32 shader;
+	u32 memory;
+	u32 vdec;
+	u32 disp;
+	u32 script;
+};
+
+u16 nvbios_perf_entry(struct nouveau_bios *, int idx,
+		      u8 *ver, u8 *hdr, u8 *cnt, u8 *len);
+u16 nvbios_perfEp(struct nouveau_bios *, int idx,
+		  u8 *ver, u8 *hdr, u8 *cnt, u8 *len, struct nvbios_perfE *);
+
+struct nvbios_perfS {
+	union {
+		struct {
+			u32 freq;
+		} v40;
+	};
+};
+
+u32 nvbios_perfSe(struct nouveau_bios *, u32 data, int idx,
+		  u8 *ver, u8 *hdr, u8 cnt, u8 len);
+u32 nvbios_perfSp(struct nouveau_bios *, u32 data, int idx,
+		  u8 *ver, u8 *hdr, u8 cnt, u8 len, struct nvbios_perfS *);
+
 struct nvbios_perf_fan {
 	u32 pwm_divisor;
 };
diff --git a/drivers/gpu/drm/nouveau/core/include/subdev/bios/rammap.h b/drivers/gpu/drm/nouveau/core/include/subdev/bios/rammap.h
new file mode 100644
index 000000000000..bc15e0320877
--- /dev/null
+++ b/drivers/gpu/drm/nouveau/core/include/subdev/bios/rammap.h
@@ -0,0 +1,11 @@
+#ifndef __NVBIOS_RAMMAP_H__
+#define __NVBIOS_RAMMAP_H__
+
+u16 nvbios_rammap_table(struct nouveau_bios *, u8 *ver, u8 *hdr,
+			u8 *cnt, u8 *len, u8 *snr, u8 *ssz);
+u16 nvbios_rammap_entry(struct nouveau_bios *, int idx,
+			u8 *ver, u8 *hdr, u8 *cnt, u8 *len);
+u16 nvbios_rammap_match(struct nouveau_bios *, u16 khz,
+			u8 *ver, u8 *hdr, u8 *cnt, u8 *len);
+
+#endif
diff --git a/drivers/gpu/drm/nouveau/core/include/subdev/bios/timing.h b/drivers/gpu/drm/nouveau/core/include/subdev/bios/timing.h
new file mode 100644
index 000000000000..963694b54224
--- /dev/null
+++ b/drivers/gpu/drm/nouveau/core/include/subdev/bios/timing.h
@@ -0,0 +1,8 @@
+#ifndef __NVBIOS_TIMING_H__
+#define __NVBIOS_TIMING_H__
+
+u16 nvbios_timing_table(struct nouveau_bios *,
+			u8 *ver, u8 *hdr, u8 *cnt, u8 *len);
+u16 nvbios_timing_entry(struct nouveau_bios *, int idx, u8 *ver, u8 *hdr);
+
+#endif
diff --git a/drivers/gpu/drm/nouveau/core/include/subdev/bios/vmap.h b/drivers/gpu/drm/nouveau/core/include/subdev/bios/vmap.h
new file mode 100644
index 000000000000..ad5a8f20e113
--- /dev/null
+++ b/drivers/gpu/drm/nouveau/core/include/subdev/bios/vmap.h
@@ -0,0 +1,25 @@
+#ifndef __NVBIOS_VMAP_H__
+#define __NVBIOS_VMAP_H__
+
+struct nouveau_bios;
+
+struct nvbios_vmap {
+};
+
+u16 nvbios_vmap_table(struct nouveau_bios *, u8 *ver, u8 *hdr, u8 *cnt, u8 *len);
+u16 nvbios_vmap_parse(struct nouveau_bios *, u8 *ver, u8 *hdr, u8 *cnt, u8 *len,
+		      struct nvbios_vmap *);
+
+struct nvbios_vmap_entry {
+	u8  unk0;
+	u8  link;
+	u32 min;
+	u32 max;
+	s32 arg[6];
+};
+
+u16 nvbios_vmap_entry(struct nouveau_bios *, int idx, u8 *ver, u8 *len);
+u16 nvbios_vmap_entry_parse(struct nouveau_bios *, int idx, u8 *ver, u8 *len,
+			    struct nvbios_vmap_entry *);
+
+#endif
diff --git a/drivers/gpu/drm/nouveau/core/include/subdev/bios/volt.h b/drivers/gpu/drm/nouveau/core/include/subdev/bios/volt.h
new file mode 100644
index 000000000000..6a11dcd59770
--- /dev/null
+++ b/drivers/gpu/drm/nouveau/core/include/subdev/bios/volt.h
@@ -0,0 +1,27 @@
+#ifndef __NVBIOS_VOLT_H__
+#define __NVBIOS_VOLT_H__
+
+struct nouveau_bios;
+
+struct nvbios_volt {
+	u8  vidmask;
+	u32 min;
+	u32 max;
+	u32 base;
+	s16 step;
+};
+
+u16 nvbios_volt_table(struct nouveau_bios *, u8 *ver, u8 *hdr, u8 *cnt, u8 *len);
+u16 nvbios_volt_parse(struct nouveau_bios *, u8 *ver, u8 *hdr, u8 *cnt, u8 *len,
+		      struct nvbios_volt *);
+
+struct nvbios_volt_entry {
+	u32 voltage;
+	u8  vid;
+};
+
+u16 nvbios_volt_entry(struct nouveau_bios *, int idx, u8 *ver, u8 *len);
+u16 nvbios_volt_entry_parse(struct nouveau_bios *, int idx, u8 *ver, u8 *len,
+			    struct nvbios_volt_entry *);
+
+#endif
diff --git a/drivers/gpu/drm/nouveau/core/include/subdev/bus.h b/drivers/gpu/drm/nouveau/core/include/subdev/bus.h
index 7d88ec4a6d06..697f7ce70aab 100644
--- a/drivers/gpu/drm/nouveau/core/include/subdev/bus.h
+++ b/drivers/gpu/drm/nouveau/core/include/subdev/bus.h
@@ -11,6 +11,8 @@ struct nouveau_bus_intr {
 
 struct nouveau_bus {
 	struct nouveau_subdev base;
+	int (*hwsq_exec)(struct nouveau_bus *, u32 *, u32);
+	u32 hwsq_size;
 };
 
 static inline struct nouveau_bus *
@@ -33,9 +35,19 @@ nouveau_bus(void *obj)
 #define _nouveau_bus_init _nouveau_subdev_init
 #define _nouveau_bus_fini _nouveau_subdev_fini
 
-extern struct nouveau_oclass nv04_bus_oclass;
-extern struct nouveau_oclass nv31_bus_oclass;
-extern struct nouveau_oclass nv50_bus_oclass;
-extern struct nouveau_oclass nvc0_bus_oclass;
+extern struct nouveau_oclass *nv04_bus_oclass;
+extern struct nouveau_oclass *nv31_bus_oclass;
+extern struct nouveau_oclass *nv50_bus_oclass;
+extern struct nouveau_oclass *nv94_bus_oclass;
+extern struct nouveau_oclass *nvc0_bus_oclass;
+
+/* interface to sequencer */
+struct nouveau_hwsq;
+int  nouveau_hwsq_init(struct nouveau_bus *, struct nouveau_hwsq **);
+int  nouveau_hwsq_fini(struct nouveau_hwsq **, bool exec);
+void nouveau_hwsq_wr32(struct nouveau_hwsq *, u32 addr, u32 data);
+void nouveau_hwsq_setf(struct nouveau_hwsq *, u8 flag, int data);
+void nouveau_hwsq_wait(struct nouveau_hwsq *, u8 flag, u8 data);
+void nouveau_hwsq_nsec(struct nouveau_hwsq *, u32 nsec);
 
 #endif
diff --git a/drivers/gpu/drm/nouveau/core/include/subdev/clock.h b/drivers/gpu/drm/nouveau/core/include/subdev/clock.h
index 89ee289097a6..e2675bc0edba 100644
--- a/drivers/gpu/drm/nouveau/core/include/subdev/clock.h
+++ b/drivers/gpu/drm/nouveau/core/include/subdev/clock.h
@@ -7,9 +7,78 @@
 struct nouveau_pll_vals;
 struct nvbios_pll;
 
+enum nv_clk_src {
+	nv_clk_src_crystal,
+	nv_clk_src_href,
+
+	nv_clk_src_hclk,
+	nv_clk_src_hclkm3,
+	nv_clk_src_hclkm3d2,
+
+	nv_clk_src_host,
+
+	nv_clk_src_sppll0,
+	nv_clk_src_sppll1,
+
+	nv_clk_src_mpllsrcref,
+	nv_clk_src_mpllsrc,
+	nv_clk_src_mpll,
+	nv_clk_src_mdiv,
+
+	nv_clk_src_core,
+	nv_clk_src_shader,
+
+	nv_clk_src_mem,
+
+	nv_clk_src_gpc,
+	nv_clk_src_rop,
+	nv_clk_src_hubk01,
+	nv_clk_src_hubk06,
+	nv_clk_src_hubk07,
+	nv_clk_src_copy,
+	nv_clk_src_daemon,
+	nv_clk_src_disp,
+	nv_clk_src_vdec,
+
+	nv_clk_src_dom6,
+
+	nv_clk_src_max,
+};
+
+struct nouveau_cstate {
+	struct list_head head;
+	u8  voltage;
+	u32 domain[nv_clk_src_max];
+};
+
+struct nouveau_pstate {
+	struct list_head head;
+	struct list_head list; /* c-states */
+	struct nouveau_cstate base;
+	u8 pstate;
+	u8 fanspeed;
+};
+
 struct nouveau_clock {
 	struct nouveau_subdev base;
 
+	struct nouveau_clocks *domains;
+	struct nouveau_pstate bstate;
+
+	struct list_head states;
+	int state_nr;
+
+	int pstate; /* current */
+	int ustate; /* user-requested (-1 disabled, -2 perfmon) */
+	int astate; /* perfmon adjustment (base) */
+	int tstate; /* thermal adjustment (max-) */
+	int dstate; /* display adjustment (min+) */
+
+	int  (*read)(struct nouveau_clock *, enum nv_clk_src);
+	int  (*calc)(struct nouveau_clock *, struct nouveau_cstate *);
+	int  (*prog)(struct nouveau_clock *);
+	void (*tidy)(struct nouveau_clock *);
+
 	/*XXX: die, these are here *only* to support the completely
 	 *     bat-shit insane what-was-nouveau_hw.c code
 	 */
@@ -25,27 +94,42 @@ nouveau_clock(void *obj)
 	return (void *)nv_device(obj)->subdev[NVDEV_SUBDEV_CLOCK];
 }
 
-#define nouveau_clock_create(p,e,o,d)                                          \
-	nouveau_subdev_create((p), (e), (o), 0, "CLOCK", "clock", d)
-#define nouveau_clock_destroy(p)                                               \
-	nouveau_subdev_destroy(&(p)->base)
-#define nouveau_clock_init(p)                                                  \
-	nouveau_subdev_init(&(p)->base)
+struct nouveau_clocks {
+	enum nv_clk_src name;
+	u8 bios; /* 0xff for none */
+#define NVKM_CLK_DOM_FLAG_CORE 0x01
+	u8 flags;
+	const char *mname;
+	int mdiv;
+};
+
+#define nouveau_clock_create(p,e,o,i,d)                                        \
+	nouveau_clock_create_((p), (e), (o), (i), sizeof(**d), (void **)d)
+#define nouveau_clock_destroy(p) ({                                            \
+	struct nouveau_clock *clk = (p);                                       \
+	_nouveau_clock_dtor(nv_object(clk));                                   \
+})
+#define nouveau_clock_init(p) ({                                               \
+	struct nouveau_clock *clk = (p);                                       \
+	_nouveau_clock_init(nv_object(clk));                                   \
+})
 #define nouveau_clock_fini(p,s)                                                \
 	nouveau_subdev_fini(&(p)->base, (s))
 
 int  nouveau_clock_create_(struct nouveau_object *, struct nouveau_object *,
-			   struct nouveau_oclass *, void *, u32, int, void **);
-
-#define _nouveau_clock_dtor _nouveau_subdev_dtor
-#define _nouveau_clock_init _nouveau_subdev_init
+			   struct nouveau_oclass *,
+			   struct nouveau_clocks *, int, void **);
+void _nouveau_clock_dtor(struct nouveau_object *);
+int _nouveau_clock_init(struct nouveau_object *);
 #define _nouveau_clock_fini _nouveau_subdev_fini
 
 extern struct nouveau_oclass nv04_clock_oclass;
 extern struct nouveau_oclass nv40_clock_oclass;
-extern struct nouveau_oclass nv50_clock_oclass;
+extern struct nouveau_oclass *nv50_clock_oclass;
+extern struct nouveau_oclass *nv84_clock_oclass;
 extern struct nouveau_oclass nva3_clock_oclass;
 extern struct nouveau_oclass nvc0_clock_oclass;
+extern struct nouveau_oclass nve0_clock_oclass;
 
 int nv04_clock_pll_set(struct nouveau_clock *, u32 type, u32 freq);
 int nv04_clock_pll_calc(struct nouveau_clock *, struct nvbios_pll *,
@@ -55,4 +139,9 @@ int nv04_clock_pll_prog(struct nouveau_clock *, u32 reg1,
 int nva3_clock_pll_calc(struct nouveau_clock *, struct nvbios_pll *,
 			int clk, struct nouveau_pll_vals *);
 
+int nouveau_clock_ustate(struct nouveau_clock *, int req);
+int nouveau_clock_astate(struct nouveau_clock *, int req, int rel);
+int nouveau_clock_dstate(struct nouveau_clock *, int req, int rel);
+int nouveau_clock_tstate(struct nouveau_clock *, int req, int rel);
+
 #endif
diff --git a/drivers/gpu/drm/nouveau/core/include/subdev/fb.h b/drivers/gpu/drm/nouveau/core/include/subdev/fb.h
index 2e7405084261..8541aa382ff2 100644
--- a/drivers/gpu/drm/nouveau/core/include/subdev/fb.h
+++ b/drivers/gpu/drm/nouveau/core/include/subdev/fb.h
@@ -78,23 +78,28 @@ nouveau_fb(void *obj)
 	return (void *)nv_device(obj)->subdev[NVDEV_SUBDEV_FB];
 }
 
-extern struct nouveau_oclass nv04_fb_oclass;
-extern struct nouveau_oclass nv10_fb_oclass;
-extern struct nouveau_oclass nv1a_fb_oclass;
-extern struct nouveau_oclass nv20_fb_oclass;
-extern struct nouveau_oclass nv25_fb_oclass;
-extern struct nouveau_oclass nv30_fb_oclass;
-extern struct nouveau_oclass nv35_fb_oclass;
-extern struct nouveau_oclass nv36_fb_oclass;
-extern struct nouveau_oclass nv40_fb_oclass;
-extern struct nouveau_oclass nv41_fb_oclass;
-extern struct nouveau_oclass nv44_fb_oclass;
-extern struct nouveau_oclass nv46_fb_oclass;
-extern struct nouveau_oclass nv47_fb_oclass;
-extern struct nouveau_oclass nv49_fb_oclass;
-extern struct nouveau_oclass nv4e_fb_oclass;
-extern struct nouveau_oclass nv50_fb_oclass;
-extern struct nouveau_oclass nvc0_fb_oclass;
+extern struct nouveau_oclass *nv04_fb_oclass;
+extern struct nouveau_oclass *nv10_fb_oclass;
+extern struct nouveau_oclass *nv1a_fb_oclass;
+extern struct nouveau_oclass *nv20_fb_oclass;
+extern struct nouveau_oclass *nv25_fb_oclass;
+extern struct nouveau_oclass *nv30_fb_oclass;
+extern struct nouveau_oclass *nv35_fb_oclass;
+extern struct nouveau_oclass *nv36_fb_oclass;
+extern struct nouveau_oclass *nv40_fb_oclass;
+extern struct nouveau_oclass *nv41_fb_oclass;
+extern struct nouveau_oclass *nv44_fb_oclass;
+extern struct nouveau_oclass *nv46_fb_oclass;
+extern struct nouveau_oclass *nv47_fb_oclass;
+extern struct nouveau_oclass *nv49_fb_oclass;
+extern struct nouveau_oclass *nv4e_fb_oclass;
+extern struct nouveau_oclass *nv50_fb_oclass;
+extern struct nouveau_oclass *nv84_fb_oclass;
+extern struct nouveau_oclass *nva3_fb_oclass;
+extern struct nouveau_oclass *nvaa_fb_oclass;
+extern struct nouveau_oclass *nvaf_fb_oclass;
+extern struct nouveau_oclass *nvc0_fb_oclass;
+extern struct nouveau_oclass *nve0_fb_oclass;
 
 struct nouveau_ram {
 	struct nouveau_object base;
@@ -121,6 +126,17 @@ struct nouveau_ram {
 	int  (*get)(struct nouveau_fb *, u64 size, u32 align,
 		    u32 size_nc, u32 type, struct nouveau_mem **);
 	void (*put)(struct nouveau_fb *, struct nouveau_mem **);
+
+	int  (*calc)(struct nouveau_fb *, u32 freq);
+	int  (*prog)(struct nouveau_fb *);
+	void (*tidy)(struct nouveau_fb *);
+	struct {
+		u8  version;
+		u32 data;
+		u8  size;
+	} rammap, ramcfg, timing;
+	u32 freq;
+	u32 mr[16];
 };
 
 #endif
diff --git a/drivers/gpu/drm/nouveau/core/include/subdev/i2c.h b/drivers/gpu/drm/nouveau/core/include/subdev/i2c.h
index 7e4e2775f249..9fa5da723871 100644
--- a/drivers/gpu/drm/nouveau/core/include/subdev/i2c.h
+++ b/drivers/gpu/drm/nouveau/core/include/subdev/i2c.h
@@ -60,13 +60,18 @@ void _nouveau_i2c_port_dtor(struct nouveau_object *);
 #define _nouveau_i2c_port_init nouveau_object_init
 #define _nouveau_i2c_port_fini nouveau_object_fini
 
+struct nouveau_i2c_board_info {
+	struct i2c_board_info dev;
+	u8 udelay; /* set to 0 to use the standard delay */
+};
+
 struct nouveau_i2c {
 	struct nouveau_subdev base;
 
 	struct nouveau_i2c_port *(*find)(struct nouveau_i2c *, u8 index);
 	struct nouveau_i2c_port *(*find_type)(struct nouveau_i2c *, u16 type);
 	int (*identify)(struct nouveau_i2c *, int index,
-			const char *what, struct i2c_board_info *,
+			const char *what, struct nouveau_i2c_board_info *,
 			bool (*match)(struct nouveau_i2c_port *,
 				      struct i2c_board_info *));
 	struct list_head ports;
diff --git a/drivers/gpu/drm/nouveau/core/include/subdev/mc.h b/drivers/gpu/drm/nouveau/core/include/subdev/mc.h
index ce6569f365a7..adc88b73d911 100644
--- a/drivers/gpu/drm/nouveau/core/include/subdev/mc.h
+++ b/drivers/gpu/drm/nouveau/core/include/subdev/mc.h
@@ -11,7 +11,6 @@ struct nouveau_mc_intr {
 
 struct nouveau_mc {
 	struct nouveau_subdev base;
-	const struct nouveau_mc_intr *intr_map;
 	bool use_msi;
 };
 
@@ -21,8 +20,8 @@ nouveau_mc(void *obj)
 	return (void *)nv_device(obj)->subdev[NVDEV_SUBDEV_MC];
 }
 
-#define nouveau_mc_create(p,e,o,m,d)                                           \
-	nouveau_mc_create_((p), (e), (o), (m), sizeof(**d), (void **)d)
+#define nouveau_mc_create(p,e,o,d)                                             \
+	nouveau_mc_create_((p), (e), (o), sizeof(**d), (void **)d)
 #define nouveau_mc_destroy(p) ({                                               \
 	struct nouveau_mc *pmc = (p); _nouveau_mc_dtor(nv_object(pmc));        \
 })
@@ -34,20 +33,24 @@ nouveau_mc(void *obj)
 })
 
 int  nouveau_mc_create_(struct nouveau_object *, struct nouveau_object *,
-			struct nouveau_oclass *, const struct nouveau_mc_intr *,
-			int, void **);
+			struct nouveau_oclass *, int, void **);
 void _nouveau_mc_dtor(struct nouveau_object *);
 int  _nouveau_mc_init(struct nouveau_object *);
 int  _nouveau_mc_fini(struct nouveau_object *, bool);
 
-extern struct nouveau_oclass nv04_mc_oclass;
-extern struct nouveau_oclass nv44_mc_oclass;
-extern struct nouveau_oclass nv50_mc_oclass;
-extern struct nouveau_oclass nv98_mc_oclass;
-extern struct nouveau_oclass nvc0_mc_oclass;
+struct nouveau_mc_oclass {
+	struct nouveau_oclass base;
+	const struct nouveau_mc_intr *intr;
+	void (*msi_rearm)(struct nouveau_mc *);
+};
 
-extern const struct nouveau_mc_intr nv04_mc_intr[];
-int nv04_mc_init(struct nouveau_object *);
-int nv50_mc_init(struct nouveau_object *);
+extern struct nouveau_oclass *nv04_mc_oclass;
+extern struct nouveau_oclass *nv40_mc_oclass;
+extern struct nouveau_oclass *nv44_mc_oclass;
+extern struct nouveau_oclass *nv50_mc_oclass;
+extern struct nouveau_oclass *nv94_mc_oclass;
+extern struct nouveau_oclass *nv98_mc_oclass;
+extern struct nouveau_oclass *nvc0_mc_oclass;
+extern struct nouveau_oclass *nvc3_mc_oclass;
 
 #endif
diff --git a/drivers/gpu/drm/nouveau/core/include/subdev/pwr.h b/drivers/gpu/drm/nouveau/core/include/subdev/pwr.h
new file mode 100644
index 000000000000..c5c92cbed33f
--- /dev/null
+++ b/drivers/gpu/drm/nouveau/core/include/subdev/pwr.h
@@ -0,0 +1,80 @@
+#ifndef __NOUVEAU_PWR_H__
+#define __NOUVEAU_PWR_H__
+
+#include <core/subdev.h>
+#include <core/device.h>
+
+struct nouveau_pwr {
+	struct nouveau_subdev base;
+
+	struct {
+		u32 limit;
+		u32 *data;
+		u32  size;
+	} code;
+
+	struct {
+		u32 limit;
+		u32 *data;
+		u32  size;
+	} data;
+
+	struct {
+		u32 base;
+		u32 size;
+	} send;
+
+	struct {
+		u32 base;
+		u32 size;
+
+		struct work_struct work;
+		wait_queue_head_t wait;
+		u32 process;
+		u32 message;
+		u32 data[2];
+	} recv;
+
+	int (*message)(struct nouveau_pwr *, u32[2], u32, u32, u32, u32);
+};
+
+static inline struct nouveau_pwr *
+nouveau_pwr(void *obj)
+{
+	return (void *)nv_device(obj)->subdev[NVDEV_SUBDEV_PWR];
+}
+
+#define nouveau_pwr_create(p, e, o, d)                                         \
+	nouveau_pwr_create_((p), (e), (o), sizeof(**d), (void **)d)
+#define nouveau_pwr_destroy(p)                                                 \
+	nouveau_subdev_destroy(&(p)->base)
+#define nouveau_pwr_init(p) ({                                                 \
+	struct nouveau_pwr *ppwr = (p);                                        \
+	_nouveau_pwr_init(nv_object(ppwr));                                    \
+})
+#define nouveau_pwr_fini(p,s) ({                                               \
+	struct nouveau_pwr *ppwr = (p);                                        \
+	_nouveau_pwr_fini(nv_object(ppwr), (s));                               \
+})
+
+int nouveau_pwr_create_(struct nouveau_object *, struct nouveau_object *,
+			   struct nouveau_oclass *, int, void **);
+#define _nouveau_pwr_dtor _nouveau_subdev_dtor
+int _nouveau_pwr_init(struct nouveau_object *);
+int _nouveau_pwr_fini(struct nouveau_object *, bool);
+
+extern struct nouveau_oclass nva3_pwr_oclass;
+extern struct nouveau_oclass nvc0_pwr_oclass;
+extern struct nouveau_oclass nvd0_pwr_oclass;
+extern struct nouveau_oclass nv108_pwr_oclass;
+
+/* interface to MEMX process running on PPWR */
+struct nouveau_memx;
+int  nouveau_memx_init(struct nouveau_pwr *, struct nouveau_memx **);
+int  nouveau_memx_fini(struct nouveau_memx **, bool exec);
+void nouveau_memx_wr32(struct nouveau_memx *, u32 addr, u32 data);
+void nouveau_memx_wait(struct nouveau_memx *,
+		       u32 addr, u32 mask, u32 data, u32 nsec);
+void nouveau_memx_nsec(struct nouveau_memx *, u32 nsec);
+
+#endif
diff --git a/drivers/gpu/drm/nouveau/core/include/subdev/therm.h b/drivers/gpu/drm/nouveau/core/include/subdev/therm.h
index c075998d82e6..69891d4a3fe7 100644
--- a/drivers/gpu/drm/nouveau/core/include/subdev/therm.h
+++ b/drivers/gpu/drm/nouveau/core/include/subdev/therm.h
@@ -71,6 +71,8 @@ void _nouveau_therm_dtor(struct nouveau_object *);
 int  _nouveau_therm_init(struct nouveau_object *);
 int  _nouveau_therm_fini(struct nouveau_object *, bool);
 
+int  nouveau_therm_cstate(struct nouveau_therm *, int, int);
+
 extern struct nouveau_oclass nv40_therm_oclass;
 extern struct nouveau_oclass nv50_therm_oclass;
 extern struct nouveau_oclass nv84_therm_oclass;
diff --git a/drivers/gpu/drm/nouveau/core/include/subdev/volt.h b/drivers/gpu/drm/nouveau/core/include/subdev/volt.h
new file mode 100644
index 000000000000..820b62ffd75b
--- /dev/null
+++ b/drivers/gpu/drm/nouveau/core/include/subdev/volt.h
@@ -0,0 +1,60 @@
+#ifndef __NOUVEAU_VOLT_H__
+#define __NOUVEAU_VOLT_H__
+
+#include <core/subdev.h>
+#include <core/device.h>
+
+struct nouveau_voltage {
+	u32 uv;
+	u8  id;
+};
+
+struct nouveau_volt {
+	struct nouveau_subdev base;
+
+	int (*vid_get)(struct nouveau_volt *);
+	int (*get)(struct nouveau_volt *);
+	int (*vid_set)(struct nouveau_volt *, u8 vid);
+	int (*set)(struct nouveau_volt *, u32 uv);
+	int (*set_id)(struct nouveau_volt *, u8 id, int condition);
+
+	u8 vid_mask;
+	u8 vid_nr;
+	struct {
+		u32 uv;
+		u8 vid;
+	} vid[256];
+};
+
+static inline struct nouveau_volt *
+nouveau_volt(void *obj)
+{
+	return (void *)nv_device(obj)->subdev[NVDEV_SUBDEV_VOLT];
+}
+
+#define nouveau_volt_create(p, e, o, d)                                        \
+	nouveau_volt_create_((p), (e), (o), sizeof(**d), (void **)d)
+#define nouveau_volt_destroy(p) ({                                             \
+	struct nouveau_volt *v = (p);                                          \
+	_nouveau_volt_dtor(nv_object(v));                                      \
+})
+#define nouveau_volt_init(p) ({                                                \
+	struct nouveau_volt *v = (p);                                          \
+	_nouveau_volt_init(nv_object(v));                                      \
+})
+#define nouveau_volt_fini(p,s)                                                 \
+	nouveau_subdev_fini((p), (s))
+
+int  nouveau_volt_create_(struct nouveau_object *, struct nouveau_object *,
+			  struct nouveau_oclass *, int, void **);
+void _nouveau_volt_dtor(struct nouveau_object *);
+int  _nouveau_volt_init(struct nouveau_object *);
+#define _nouveau_volt_fini _nouveau_subdev_fini
+
+extern struct nouveau_oclass nv40_volt_oclass;
+
+int nouveau_voltgpio_init(struct nouveau_volt *);
+int nouveau_voltgpio_get(struct nouveau_volt *);
+int nouveau_voltgpio_set(struct nouveau_volt *, u8);
+
+#endif
diff --git a/drivers/gpu/drm/nouveau/core/subdev/bios/boost.c b/drivers/gpu/drm/nouveau/core/subdev/bios/boost.c
new file mode 100644
index 000000000000..c1835e591c44
--- /dev/null
+++ b/drivers/gpu/drm/nouveau/core/subdev/bios/boost.c
@@ -0,0 +1,127 @@
+/*
+ * Copyright 2013 Red Hat Inc.
+ *
+ * Permission is hereby granted, free of charge, to any person obtaining a
+ * copy of this software and associated documentation files (the "Software"),
+ * to deal in the Software without restriction, including without limitation
+ * the rights to use, copy, modify, merge, publish, distribute, sublicense,
+ * and/or sell copies of the Software, and to permit persons to whom the
+ * Software is furnished to do so, subject to the following conditions:
+ *
+ * The above copyright notice and this permission notice shall be included in
+ * all copies or substantial portions of the Software.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+ * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+ * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT.  IN NO EVENT SHALL
+ * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR
+ * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE,
+ * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR
+ * OTHER DEALINGS IN THE SOFTWARE.
+ *
+ * Authors: Ben Skeggs
+ */
+
+#include <subdev/bios.h>
+#include <subdev/bios/bit.h>
+#include <subdev/bios/boost.h>
+
+u16
+nvbios_boostTe(struct nouveau_bios *bios,
+	       u8 *ver, u8 *hdr, u8 *cnt, u8 *len, u8 *snr, u8 *ssz)
+{
+	struct bit_entry bit_P;
+	u16 boost = 0x0000;
+
+	if (!bit_entry(bios, 'P', &bit_P)) {
+		if (bit_P.version == 2)
+			boost = nv_ro16(bios, bit_P.offset + 0x30);
+
+		if (boost) {
+			*ver = nv_ro08(bios, boost + 0);
+			switch (*ver) {
+			case 0x11:
+				*hdr = nv_ro08(bios, boost + 1);
+				*cnt = nv_ro08(bios, boost + 5);
+				*len = nv_ro08(bios, boost + 2);
+				*snr = nv_ro08(bios, boost + 4);
+				*ssz = nv_ro08(bios, boost + 3);
+				return boost;
+			default:
+				break;
+			}
+		}
+	}
+
+	return 0x0000;
+}
+
+u16
+nvbios_boostEe(struct nouveau_bios *bios, int idx,
+	       u8 *ver, u8 *hdr, u8 *cnt, u8 *len)
+{
+	u8  snr, ssz;
+	u16 data = nvbios_boostTe(bios, ver, hdr, cnt, len, &snr, &ssz);
+	if (data && idx < *cnt) {
+		data = data + *hdr + (idx * (*len + (snr * ssz)));
+		*hdr = *len;
+		*cnt = snr;
+		*len = ssz;
+		return data;
+	}
+	return 0x0000;
+}
+
+u16
+nvbios_boostEp(struct nouveau_bios *bios, int idx,
+	       u8 *ver, u8 *hdr, u8 *cnt, u8 *len, struct nvbios_boostE *info)
+{
+	u16 data = nvbios_boostEe(bios, idx, ver, hdr, cnt, len);
+	memset(info, 0x00, sizeof(*info));
+	if (data) {
+		info->pstate = (nv_ro16(bios, data + 0x00) & 0x01e0) >> 5;
+		info->min    =  nv_ro16(bios, data + 0x02) * 1000;
+		info->max    =  nv_ro16(bios, data + 0x04) * 1000;
+	}
+	return data;
+}
+
+u16
+nvbios_boostEm(struct nouveau_bios *bios, u8 pstate,
+	       u8 *ver, u8 *hdr, u8 *cnt, u8 *len, struct nvbios_boostE *info)
+{
+	u32 data, idx = 0;
+	while ((data = nvbios_boostEp(bios, idx++, ver, hdr, cnt, len, info))) {
+		if (info->pstate == pstate)
+			break;
+	}
+	return data;
+}
+
+u16
+nvbios_boostSe(struct nouveau_bios *bios, int idx,
+	       u16 data, u8 *ver, u8 *hdr, u8 cnt, u8 len)
+{
+	if (data && idx < cnt) {
+		data = data + *hdr + (idx * len);
+		*hdr = len;
+		return data;
+	}
+	return 0x0000;
+}
+
+u16
+nvbios_boostSp(struct nouveau_bios *bios, int idx,
+	       u16 data, u8 *ver, u8 *hdr, u8 cnt, u8 len,
+	       struct nvbios_boostS *info)
+{
+	data = nvbios_boostSe(bios, idx, data, ver, hdr, cnt, len);
+	memset(info, 0x00, sizeof(*info));
+	if (data) {
+		info->domain  = nv_ro08(bios, data + 0x00);
+		info->percent = nv_ro08(bios, data + 0x01);
+		info->min     = nv_ro16(bios, data + 0x02) * 1000;
+		info->max     = nv_ro16(bios, data + 0x04) * 1000;
+	}
+	return data;
+}
diff --git a/drivers/gpu/drm/nouveau/core/subdev/bios/cstep.c b/drivers/gpu/drm/nouveau/core/subdev/bios/cstep.c
new file mode 100644
index 000000000000..d3b15327fbfd
--- /dev/null
+++ b/drivers/gpu/drm/nouveau/core/subdev/bios/cstep.c
@@ -0,0 +1,123 @@
+/*
+ * Copyright 2013 Red Hat Inc.
+ *
+ * Permission is hereby granted, free of charge, to any person obtaining a
+ * copy of this software and associated documentation files (the "Software"),
+ * to deal in the Software without restriction, including without limitation
+ * the rights to use, copy, modify, merge, publish, distribute, sublicense,
+ * and/or sell copies of the Software, and to permit persons to whom the
+ * Software is furnished to do so, subject to the following conditions:
+ *
+ * The above copyright notice and this permission notice shall be included in
+ * all copies or substantial portions of the Software.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+ * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+ * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT.  IN NO EVENT SHALL
+ * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR
+ * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE,
+ * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR
+ * OTHER DEALINGS IN THE SOFTWARE.
+ *
+ * Authors: Ben Skeggs
+ */
+
+#include <subdev/bios.h>
+#include <subdev/bios/bit.h>
+#include <subdev/bios/cstep.h>
+
+u16
+nvbios_cstepTe(struct nouveau_bios *bios,
+	       u8 *ver, u8 *hdr, u8 *cnt, u8 *len, u8 *xnr, u8 *xsz)
+{
+	struct bit_entry bit_P;
+	u16 cstep = 0x0000;
+
+	if (!bit_entry(bios, 'P', &bit_P)) {
+		if (bit_P.version == 2)
+			cstep = nv_ro16(bios, bit_P.offset + 0x34);
+
+		if (cstep) {
+			*ver = nv_ro08(bios, cstep + 0);
+			switch (*ver) {
+			case 0x10:
+				*hdr = nv_ro08(bios, cstep + 1);
+				*cnt = nv_ro08(bios, cstep + 3);
+				*len = nv_ro08(bios, cstep + 2);
+				*xnr = nv_ro08(bios, cstep + 5);
+				*xsz = nv_ro08(bios, cstep + 4);
+				return cstep;
+			default:
+				break;
+			}
+		}
+	}
+
+	return 0x0000;
+}
+
+u16
+nvbios_cstepEe(struct nouveau_bios *bios, int idx, u8 *ver, u8 *hdr)
+{
+	u8  cnt, len, xnr, xsz;
+	u16 data = nvbios_cstepTe(bios, ver, hdr, &cnt, &len, &xnr, &xsz);
+	if (data && idx < cnt) {
+		data = data + *hdr + (idx * len);
+		*hdr = len;
+		return data;
+	}
+	return 0x0000;
+}
+
+u16
+nvbios_cstepEp(struct nouveau_bios *bios, int idx, u8 *ver, u8 *hdr,
+	       struct nvbios_cstepE *info)
+{
+	u16 data = nvbios_cstepEe(bios, idx, ver, hdr);
+	memset(info, 0x00, sizeof(*info));
+	if (data) {
+		info->pstate = (nv_ro16(bios, data + 0x00) & 0x01e0) >> 5;
+		info->index   = nv_ro08(bios, data + 0x03);
+	}
+	return data;
+}
+
+u16
+nvbios_cstepEm(struct nouveau_bios *bios, u8 pstate, u8 *ver, u8 *hdr,
+	       struct nvbios_cstepE *info)
+{
+	u32 data, idx = 0;
+	while ((data = nvbios_cstepEp(bios, idx++, ver, hdr, info))) {
+		if (info->pstate == pstate)
+			break;
+	}
+	return data;
+}
+
+u16
+nvbios_cstepXe(struct nouveau_bios *bios, int idx, u8 *ver, u8 *hdr)
+{
+	u8  cnt, len, xnr, xsz;
+	u16 data = nvbios_cstepTe(bios, ver, hdr, &cnt, &len, &xnr, &xsz);
+	if (data && idx < xnr) {
+		data = data + *hdr + (cnt * len) + (idx * xsz);
+		*hdr = xsz;
+		return data;
+	}
+	return 0x0000;
+}
+
+u16
+nvbios_cstepXp(struct nouveau_bios *bios, int idx, u8 *ver, u8 *hdr,
+	       struct nvbios_cstepX *info)
+{
+	u16 data = nvbios_cstepXe(bios, idx, ver, hdr);
+	memset(info, 0x00, sizeof(*info));
+	if (data) {
+		info->freq    = nv_ro16(bios, data + 0x00) * 1000;
+		info->unkn[0] = nv_ro08(bios, data + 0x02);
+		info->unkn[1] = nv_ro08(bios, data + 0x03);
+		info->voltage = nv_ro08(bios, data + 0x04);
+	}
+	return data;
+}
diff --git a/drivers/gpu/drm/nouveau/core/subdev/bios/dp.c b/drivers/gpu/drm/nouveau/core/subdev/bios/dp.c
index 663853bcca82..7628fe759220 100644
--- a/drivers/gpu/drm/nouveau/core/subdev/bios/dp.c
+++ b/drivers/gpu/drm/nouveau/core/subdev/bios/dp.c
@@ -89,6 +89,7 @@ nvbios_dpout_parse(struct nouveau_bios *bios, u8 idx,
 		   struct nvbios_dpout *info)
 {
 	u16 data = nvbios_dpout_entry(bios, idx, ver, hdr, cnt, len);
+	memset(info, 0x00, sizeof(*info));
 	if (data && *ver) {
 		info->type = nv_ro16(bios, data + 0x00);
 		info->mask = nv_ro16(bios, data + 0x02);
@@ -99,9 +100,12 @@ nvbios_dpout_parse(struct nouveau_bios *bios, u8 idx,
 			info->script[0] = nv_ro16(bios, data + 0x06);
 			info->script[1] = nv_ro16(bios, data + 0x08);
 			info->lnkcmp    = nv_ro16(bios, data + 0x0a);
-			info->script[2] = nv_ro16(bios, data + 0x0c);
-			info->script[3] = nv_ro16(bios, data + 0x0e);
-			info->script[4] = nv_ro16(bios, data + 0x10);
+			if (*len >= 0x0f) {
+				info->script[2] = nv_ro16(bios, data + 0x0c);
+				info->script[3] = nv_ro16(bios, data + 0x0e);
+			}
+			if (*len >= 0x11)
+				info->script[4] = nv_ro16(bios, data + 0x10);
 			break;
 		case 0x40:
 			info->flags     = nv_ro08(bios, data + 0x04);
diff --git a/drivers/gpu/drm/nouveau/core/subdev/bios/init.c b/drivers/gpu/drm/nouveau/core/subdev/bios/init.c
index 57cda2a1437b..420908cb82b6 100644
--- a/drivers/gpu/drm/nouveau/core/subdev/bios/init.c
+++ b/drivers/gpu/drm/nouveau/core/subdev/bios/init.c
@@ -2180,7 +2180,7 @@ nvbios_init(struct nouveau_subdev *subdev, bool execute)
 	u16 data;
 
 	if (execute)
-		nv_suspend(bios, "running init tables\n");
+		nv_info(bios, "running init tables\n");
 	while (!ret && (data = (init_script(bios, ++i)))) {
 		struct nvbios_init init = {
 			.subdev = subdev,
@@ -2210,5 +2210,5 @@ nvbios_init(struct nouveau_subdev *subdev, bool execute)
 		ret = nvbios_exec(&init);
 	}
 
-	return 0;
+	return ret;
 }
diff --git a/drivers/gpu/drm/nouveau/core/subdev/bios/perf.c b/drivers/gpu/drm/nouveau/core/subdev/bios/perf.c
index bcbb056c2887..675e221680aa 100644
--- a/drivers/gpu/drm/nouveau/core/subdev/bios/perf.c
+++ b/drivers/gpu/drm/nouveau/core/subdev/bios/perf.c
@@ -26,8 +26,9 @@
 #include <subdev/bios/bit.h>
 #include <subdev/bios/perf.h>
 
-static u16
-perf_table(struct nouveau_bios *bios, u8 *ver, u8 *hdr, u8 *cnt, u8 *len)
+u16
+nvbios_perf_table(struct nouveau_bios *bios, u8 *ver, u8 *hdr,
+		  u8 *cnt, u8 *len, u8 *snr, u8 *ssz)
 {
 	struct bit_entry bit_P;
 	u16 perf = 0x0000;
@@ -38,10 +39,22 @@ perf_table(struct nouveau_bios *bios, u8 *ver, u8 *hdr, u8 *cnt, u8 *len)
 			if (perf) {
 				*ver = nv_ro08(bios, perf + 0);
 				*hdr = nv_ro08(bios, perf + 1);
+				if (*ver >= 0x40 && *ver < 0x41) {
+					*cnt = nv_ro08(bios, perf + 5);
+					*len = nv_ro08(bios, perf + 2);
+					*snr = nv_ro08(bios, perf + 4);
+					*ssz = nv_ro08(bios, perf + 3);
+					return perf;
+				} else
+				if (*ver >= 0x20 && *ver < 0x40) {
+					*cnt = nv_ro08(bios, perf + 2);
+					*len = nv_ro08(bios, perf + 3);
+					*snr = nv_ro08(bios, perf + 4);
+					*ssz = nv_ro08(bios, perf + 5);
+					return perf;
+				}
 			}
-		} else
-			nv_error(bios, "unknown offset for perf in BIT P %d\n",
-				bit_P.version);
+		}
 	}
 
 	if (bios->bmp_offset) {
@@ -50,19 +63,132 @@ perf_table(struct nouveau_bios *bios, u8 *ver, u8 *hdr, u8 *cnt, u8 *len)
 			if (perf) {
 				*hdr = nv_ro08(bios, perf + 0);
 				*ver = nv_ro08(bios, perf + 1);
+				*cnt = nv_ro08(bios, perf + 2);
+				*len = nv_ro08(bios, perf + 3);
+				*snr = 0;
+				*ssz = 0;
+				return perf;
 			}
 		}
 	}
 
+	return 0x0000;
+}
+
+u16
+nvbios_perf_entry(struct nouveau_bios *bios, int idx,
+		  u8 *ver, u8 *hdr, u8 *cnt, u8 *len)
+{
+	u8  snr, ssz;
+	u16 perf = nvbios_perf_table(bios, ver, hdr, cnt, len, &snr, &ssz);
+	if (perf && idx < *cnt) {
+		perf = perf + *hdr + (idx * (*len + (snr * ssz)));
+		*hdr = *len;
+		*cnt = snr;
+		*len = ssz;
+		return perf;
+	}
+	return 0x0000;
+}
+
+u16
+nvbios_perfEp(struct nouveau_bios *bios, int idx,
+	      u8 *ver, u8 *hdr, u8 *cnt, u8 *len,
+	      struct nvbios_perfE *info)
+{
+	u16 perf = nvbios_perf_entry(bios, idx, ver, hdr, cnt, len);
+	memset(info, 0x00, sizeof(*info));
+	info->pstate = nv_ro08(bios, perf + 0x00);
+	switch (!!perf * *ver) {
+	case 0x12:
+	case 0x13:
+	case 0x14:
+		info->core     = nv_ro32(bios, perf + 0x01) * 10;
+		info->memory   = nv_ro32(bios, perf + 0x05) * 20;
+		info->fanspeed = nv_ro08(bios, perf + 0x37);
+		if (*hdr > 0x38)
+			info->voltage = nv_ro08(bios, perf + 0x38);
+		break;
+	case 0x21:
+	case 0x23:
+	case 0x24:
+		info->fanspeed = nv_ro08(bios, perf + 0x04);
+		info->voltage  = nv_ro08(bios, perf + 0x05);
+		info->shader   = nv_ro16(bios, perf + 0x06) * 1000;
+		info->core     = info->shader + (signed char)
+				 nv_ro08(bios, perf + 0x08) * 1000;
+		switch (nv_device(bios)->chipset) {
+		case 0x49:
+		case 0x4b:
+			info->memory = nv_ro16(bios, perf + 0x0b) * 1000;
+			break;
+		default:
+			info->memory = nv_ro16(bios, perf + 0x0b) * 2000;
+			break;
+		}
+		break;
+	case 0x25:
+		info->fanspeed = nv_ro08(bios, perf + 0x04);
+		info->voltage  = nv_ro08(bios, perf + 0x05);
+		info->core     = nv_ro16(bios, perf + 0x06) * 1000;
+		info->shader   = nv_ro16(bios, perf + 0x0a) * 1000;
+		info->memory   = nv_ro16(bios, perf + 0x0c) * 1000;
+		break;
+	case 0x30:
+		info->script   = nv_ro16(bios, perf + 0x02);
+	case 0x35:
+		info->fanspeed = nv_ro08(bios, perf + 0x06);
+		info->voltage  = nv_ro08(bios, perf + 0x07);
+		info->core     = nv_ro16(bios, perf + 0x08) * 1000;
+		info->shader   = nv_ro16(bios, perf + 0x0a) * 1000;
+		info->memory   = nv_ro16(bios, perf + 0x0c) * 1000;
+		info->vdec     = nv_ro16(bios, perf + 0x10) * 1000;
+		info->disp     = nv_ro16(bios, perf + 0x14) * 1000;
+		break;
+	case 0x40:
+		info->voltage  = nv_ro08(bios, perf + 0x02);
+		break;
+	default:
+		return 0x0000;
+	}
 	return perf;
 }
 
+u32
+nvbios_perfSe(struct nouveau_bios *bios, u32 perfE, int idx,
+	      u8 *ver, u8 *hdr, u8 cnt, u8 len)
+{
+	u32 data = 0x00000000;
+	if (idx < cnt) {
+		data = perfE + *hdr + (idx * len);
+		*hdr = len;
+	}
+	return data;
+}
+
+u32
+nvbios_perfSp(struct nouveau_bios *bios, u32 perfE, int idx,
+	      u8 *ver, u8 *hdr, u8 cnt, u8 len,
+	      struct nvbios_perfS *info)
+{
+	u32 data = nvbios_perfSe(bios, perfE, idx, ver, hdr, cnt, len);
+	memset(info, 0x00, sizeof(*info));
+	switch (!!data * *ver) {
+	case 0x40:
+		info->v40.freq = (nv_ro16(bios, data + 0x00) & 0x3fff) * 1000;
+		break;
+	default:
+		break;
+	}
+	return data;
+}
+
 int
 nvbios_perf_fan_parse(struct nouveau_bios *bios,
 		      struct nvbios_perf_fan *fan)
 {
-	u8 ver = 0, hdr = 0, cnt = 0, len = 0;
-	u16 perf = perf_table(bios, &ver, &hdr, &cnt, &len);
+	u8  ver, hdr, cnt, len, snr, ssz;
+	u16 perf = nvbios_perf_table(bios, &ver, &hdr, &cnt, &len, &snr, &ssz);
 	if (!perf)
 		return -ENODEV;
 
diff --git a/drivers/gpu/drm/nouveau/core/subdev/bios/pll.c b/drivers/gpu/drm/nouveau/core/subdev/bios/pll.c
index f835501203e5..1f76de597d4b 100644
--- a/drivers/gpu/drm/nouveau/core/subdev/bios/pll.c
+++ b/drivers/gpu/drm/nouveau/core/subdev/bios/pll.c
@@ -114,6 +114,7 @@ pll_map(struct nouveau_bios *bios)
 	switch (nv_device(bios)->card_type) {
 	case NV_04:
 	case NV_10:
+	case NV_11:
 	case NV_20:
 	case NV_30:
 		return nv04_pll_mapping;
diff --git a/drivers/gpu/drm/nouveau/core/subdev/bios/rammap.c b/drivers/gpu/drm/nouveau/core/subdev/bios/rammap.c
new file mode 100644
index 000000000000..916fa9d302b7
--- /dev/null
+++ b/drivers/gpu/drm/nouveau/core/subdev/bios/rammap.c
@@ -0,0 +1,88 @@
+/*
+ * Copyright 2013 Red Hat Inc.
+ *
+ * Permission is hereby granted, free of charge, to any person obtaining a
+ * copy of this software and associated documentation files (the "Software"),
+ * to deal in the Software without restriction, including without limitation
+ * the rights to use, copy, modify, merge, publish, distribute, sublicense,
+ * and/or sell copies of the Software, and to permit persons to whom the
+ * Software is furnished to do so, subject to the following conditions:
+ *
+ * The above copyright notice and this permission notice shall be included in
+ * all copies or substantial portions of the Software.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+ * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+ * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT.  IN NO EVENT SHALL
+ * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR
+ * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE,
+ * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR
+ * OTHER DEALINGS IN THE SOFTWARE.
+ *
+ * Authors: Ben Skeggs
+ */
+
+#include <subdev/bios.h>
+#include <subdev/bios/bit.h>
+#include <subdev/bios/rammap.h>
+
+u16
+nvbios_rammap_table(struct nouveau_bios *bios, u8 *ver, u8 *hdr,
+		    u8 *cnt, u8 *len, u8 *snr, u8 *ssz)
+{
+	struct bit_entry bit_P;
+	u16 rammap = 0x0000;
+
+	if (!bit_entry(bios, 'P', &bit_P)) {
+		if (bit_P.version == 2)
+			rammap = nv_ro16(bios, bit_P.offset + 4);
+
+		if (rammap) {
+			*ver = nv_ro08(bios, rammap + 0);
+			switch (*ver) {
+			case 0x10:
+			case 0x11:
+				*hdr = nv_ro08(bios, rammap + 1);
+				*cnt = nv_ro08(bios, rammap + 5);
+				*len = nv_ro08(bios, rammap + 2);
+				*snr = nv_ro08(bios, rammap + 4);
+				*ssz = nv_ro08(bios, rammap + 3);
+				return rammap;
+			default:
+				break;
+			}
+		}
+	}
+
+	return 0x0000;
+}
+
+u16
+nvbios_rammap_entry(struct nouveau_bios *bios, int idx,
+		    u8 *ver, u8 *hdr, u8 *cnt, u8 *len)
+{
+	u8  snr, ssz;
+	u16 rammap = nvbios_rammap_table(bios, ver, hdr, cnt, len, &snr, &ssz);
+	if (rammap && idx < *cnt) {
+		rammap = rammap + *hdr + (idx * (*len + (snr * ssz)));
+		*hdr = *len;
+		*cnt = snr;
+		*len = ssz;
+		return rammap;
+	}
+	return 0x0000;
+}
+
+u16
+nvbios_rammap_match(struct nouveau_bios *bios, u16 khz,
+		    u8 *ver, u8 *hdr, u8 *cnt, u8 *len)
+{
+	int idx = 0;
+	u32 data;
+	while ((data = nvbios_rammap_entry(bios, idx++, ver, hdr, cnt, len))) {
+		if (khz >= nv_ro16(bios, data + 0x00) &&
+		    khz <= nv_ro16(bios, data + 0x02))
+			break;
+	}
+	return data;
+}
diff --git a/drivers/gpu/drm/nouveau/core/subdev/bios/timing.c b/drivers/gpu/drm/nouveau/core/subdev/bios/timing.c
new file mode 100644
index 000000000000..151c2d6aaee8
--- /dev/null
+++ b/drivers/gpu/drm/nouveau/core/subdev/bios/timing.c
@@ -0,0 +1,73 @@
+/*
+ * Copyright 2013 Red Hat Inc.
+ *
+ * Permission is hereby granted, free of charge, to any person obtaining a
+ * copy of this software and associated documentation files (the "Software"),
+ * to deal in the Software without restriction, including without limitation
+ * the rights to use, copy, modify, merge, publish, distribute, sublicense,
+ * and/or sell copies of the Software, and to permit persons to whom the
+ * Software is furnished to do so, subject to the following conditions:
+ *
+ * The above copyright notice and this permission notice shall be included in
+ * all copies or substantial portions of the Software.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+ * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+ * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT.  IN NO EVENT SHALL
+ * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR
+ * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE,
+ * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR
+ * OTHER DEALINGS IN THE SOFTWARE.
+ *
+ * Authors: Ben Skeggs
+ */
+
+#include <subdev/bios.h>
+#include <subdev/bios/bit.h>
+#include <subdev/bios/timing.h>
+
+u16
+nvbios_timing_table(struct nouveau_bios *bios,
+		    u8 *ver, u8 *hdr, u8 *cnt, u8 *len)
+{
+	struct bit_entry bit_P;
+	u16 timing = 0x0000;
+
+	if (!bit_entry(bios, 'P', &bit_P)) {
+		if (bit_P.version == 1)
+			timing = nv_ro16(bios, bit_P.offset + 4);
+		else
+		if (bit_P.version == 2)
+			timing = nv_ro16(bios, bit_P.offset + 8);
+
+		if (timing) {
+			*ver = nv_ro08(bios, timing + 0);
+			switch (*ver) {
+			case 0x10:
+				*hdr = nv_ro08(bios, timing + 1);
+				*cnt = nv_ro08(bios, timing + 2);
+				*len = nv_ro08(bios, timing + 3);
+				return timing;
+			case 0x20:
+				*hdr = nv_ro08(bios, timing + 1);
+				*cnt = nv_ro08(bios, timing + 3);
+				*len = nv_ro08(bios, timing + 2);
+				return timing;
+			default:
+				break;
+			}
+		}
+	}
+
+	return 0x0000;
+}
+
+u16
+nvbios_timing_entry(struct nouveau_bios *bios, int idx, u8 *ver, u8 *len)
+{
+	u8  hdr, cnt;
+	u16 timing = nvbios_timing_table(bios, ver, &hdr, &cnt, len);
+	if (timing && idx < cnt)
+		return timing + hdr + (idx * *len);
+	return 0x0000;
+}
diff --git a/drivers/gpu/drm/nouveau/core/subdev/bios/vmap.c b/drivers/gpu/drm/nouveau/core/subdev/bios/vmap.c
new file mode 100644
index 000000000000..f343a1b060e8
--- /dev/null
+++ b/drivers/gpu/drm/nouveau/core/subdev/bios/vmap.c
@@ -0,0 +1,112 @@
+/*
+ * Copyright 2012 Nouveau Community
+ *
+ * Permission is hereby granted, free of charge, to any person obtaining a
+ * copy of this software and associated documentation files (the "Software"),
+ * to deal in the Software without restriction, including without limitation
+ * the rights to use, copy, modify, merge, publish, distribute, sublicense,
+ * and/or sell copies of the Software, and to permit persons to whom the
+ * Software is furnished to do so, subject to the following conditions:
+ *
+ * The above copyright notice and this permission notice shall be included in
+ * all copies or substantial portions of the Software.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+ * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+ * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT.  IN NO EVENT SHALL
+ * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR
+ * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE,
+ * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR
+ * OTHER DEALINGS IN THE SOFTWARE.
+ *
+ * Authors: Martin Peres
+ */
+
+#include <subdev/bios.h>
+#include <subdev/bios/bit.h>
+#include <subdev/bios/vmap.h>
+
+u16
+nvbios_vmap_table(struct nouveau_bios *bios, u8 *ver, u8 *hdr, u8 *cnt, u8 *len)
+{
+	struct bit_entry bit_P;
+	u16 vmap = 0x0000;
+
+	if (!bit_entry(bios, 'P', &bit_P)) {
+		if (bit_P.version == 2) {
+			vmap = nv_ro16(bios, bit_P.offset + 0x20);
+			if (vmap) {
+				*ver = nv_ro08(bios, vmap + 0);
+				switch (*ver) {
+				case 0x10:
+				case 0x20:
+					*hdr = nv_ro08(bios, vmap + 1);
+					*cnt = nv_ro08(bios, vmap + 3);
+					*len = nv_ro08(bios, vmap + 2);
+					return vmap;
+				default:
+					break;
+				}
+			}
+		}
+	}
+
+	return 0x0000;
+}
+
+u16
+nvbios_vmap_parse(struct nouveau_bios *bios, u8 *ver, u8 *hdr, u8 *cnt, u8 *len,
+		  struct nvbios_vmap *info)
+{
+	u16 vmap = nvbios_vmap_table(bios, ver, hdr, cnt, len);
+	memset(info, 0x00, sizeof(*info));
+	switch (!!vmap * *ver) {
+	case 0x10:
+	case 0x20:
+		break;
+	}
+	return vmap;
+}
+
+u16
+nvbios_vmap_entry(struct nouveau_bios *bios, int idx, u8 *ver, u8 *len)
+{
+	u8  hdr, cnt;
+	u16 vmap = nvbios_vmap_table(bios, ver, &hdr, &cnt, len);
+	if (vmap && idx < cnt) {
+		vmap = vmap + hdr + (idx * *len);
+		return vmap;
+	}
+	return 0x0000;
+}
+
+u16
+nvbios_vmap_entry_parse(struct nouveau_bios *bios, int idx, u8 *ver, u8 *len,
+			struct nvbios_vmap_entry *info)
+{
+	u16 vmap = nvbios_vmap_entry(bios, idx, ver, len);
+	memset(info, 0x00, sizeof(*info));
+	switch (!!vmap * *ver) {
+	case 0x10:
+		info->link   = 0xff;
+		info->min    = nv_ro32(bios, vmap + 0x00);
+		info->max    = nv_ro32(bios, vmap + 0x04);
+		info->arg[0] = nv_ro32(bios, vmap + 0x08);
+		info->arg[1] = nv_ro32(bios, vmap + 0x0c);
+		info->arg[2] = nv_ro32(bios, vmap + 0x10);
+		break;
+	case 0x20:
+		info->unk0   = nv_ro08(bios, vmap + 0x00);
+		info->link   = nv_ro08(bios, vmap + 0x01);
+		info->min    = nv_ro32(bios, vmap + 0x02);
+		info->max    = nv_ro32(bios, vmap + 0x06);
+		info->arg[0] = nv_ro32(bios, vmap + 0x0a);
+		info->arg[1] = nv_ro32(bios, vmap + 0x0e);
+		info->arg[2] = nv_ro32(bios, vmap + 0x12);
+		info->arg[3] = nv_ro32(bios, vmap + 0x16);
+		info->arg[4] = nv_ro32(bios, vmap + 0x1a);
+		info->arg[5] = nv_ro32(bios, vmap + 0x1e);
+		break;
+	}
+	return vmap;
+}
diff --git a/drivers/gpu/drm/nouveau/core/subdev/bios/volt.c b/drivers/gpu/drm/nouveau/core/subdev/bios/volt.c
new file mode 100644
index 000000000000..bb590de4ecb2
--- /dev/null
+++ b/drivers/gpu/drm/nouveau/core/subdev/bios/volt.c
@@ -0,0 +1,137 @@
+/*
+ * Copyright 2012 Nouveau Community
+ *
+ * Permission is hereby granted, free of charge, to any person obtaining a
+ * copy of this software and associated documentation files (the "Software"),
+ * to deal in the Software without restriction, including without limitation
+ * the rights to use, copy, modify, merge, publish, distribute, sublicense,
+ * and/or sell copies of the Software, and to permit persons to whom the
+ * Software is furnished to do so, subject to the following conditions:
+ *
+ * The above copyright notice and this permission notice shall be included in
+ * all copies or substantial portions of the Software.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+ * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+ * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT.  IN NO EVENT SHALL
+ * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR
+ * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE,
+ * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR
+ * OTHER DEALINGS IN THE SOFTWARE.
+ *
+ * Authors: Martin Peres
+ */
+
+#include <subdev/bios.h>
+#include <subdev/bios/bit.h>
+#include <subdev/bios/volt.h>
+
+u16
+nvbios_volt_table(struct nouveau_bios *bios, u8 *ver, u8 *hdr, u8 *cnt, u8 *len)
+{
+	struct bit_entry bit_P;
+	u16 volt = 0x0000;
+
+	if (!bit_entry(bios, 'P', &bit_P)) {
+		if (bit_P.version == 2)
+			volt = nv_ro16(bios, bit_P.offset + 0x0c);
+		else
+		if (bit_P.version == 1)
+			volt = nv_ro16(bios, bit_P.offset + 0x10);
+
+		if (volt) {
+			*ver = nv_ro08(bios, volt + 0);
+			switch (*ver) {
+			case 0x12:
+				*hdr = 5;
+				*cnt = nv_ro08(bios, volt + 2);
+				*len = nv_ro08(bios, volt + 1);
+				return volt;
+			case 0x20:
+				*hdr = nv_ro08(bios, volt + 1);
+				*cnt = nv_ro08(bios, volt + 2);
+				*len = nv_ro08(bios, volt + 3);
+				return volt;
+			case 0x30:
+			case 0x40:
+			case 0x50:
+				*hdr = nv_ro08(bios, volt + 1);
+				*cnt = nv_ro08(bios, volt + 3);
+				*len = nv_ro08(bios, volt + 2);
+				return volt;
+			}
+		}
+	}
+
+	return 0x0000;
+}
+
+u16
+nvbios_volt_parse(struct nouveau_bios *bios, u8 *ver, u8 *hdr, u8 *cnt, u8 *len,
+		  struct nvbios_volt *info)
+{
+	u16 volt = nvbios_volt_table(bios, ver, hdr, cnt, len);
+	memset(info, 0x00, sizeof(*info));
+	switch (!!volt * *ver) {
+	case 0x12:
+		info->vidmask = nv_ro08(bios, volt + 0x04);
+		break;
+	case 0x20:
+		info->vidmask = nv_ro08(bios, volt + 0x05);
+		break;
+	case 0x30:
+		info->vidmask = nv_ro08(bios, volt + 0x04);
+		break;
+	case 0x40:
+		info->base    = nv_ro32(bios, volt + 0x04);
+		info->step    = nv_ro16(bios, volt + 0x08);
+		info->vidmask = nv_ro08(bios, volt + 0x0b);
+		/*XXX*/
+		info->min     = 0;
+		info->max     = info->base;
+		break;
+	case 0x50:
+		info->vidmask = nv_ro08(bios, volt + 0x06);
+		info->min     = nv_ro32(bios, volt + 0x0a);
+		info->max     = nv_ro32(bios, volt + 0x0e);
+		info->base    = nv_ro32(bios, volt + 0x12) & 0x00ffffff;
+		info->step    = nv_ro16(bios, volt + 0x16);
+		break;
+	}
+	return volt;
+}
+
+u16
+nvbios_volt_entry(struct nouveau_bios *bios, int idx, u8 *ver, u8 *len)
+{
+	u8  hdr, cnt;
+	u16 volt = nvbios_volt_table(bios, ver, &hdr, &cnt, len);
+	if (volt && idx < cnt) {
+		volt = volt + hdr + (idx * *len);
+		return volt;
+	}
+	return 0x0000;
+}
+
+u16
+nvbios_volt_entry_parse(struct nouveau_bios *bios, int idx, u8 *ver, u8 *len,
+			struct nvbios_volt_entry *info)
+{
+	u16 volt = nvbios_volt_entry(bios, idx, ver, len);
+	memset(info, 0x00, sizeof(*info));
+	switch (!!volt * *ver) {
+	case 0x12:
+	case 0x20:
+		info->voltage = nv_ro08(bios, volt + 0x00) * 10000;
+		info->vid     = nv_ro08(bios, volt + 0x01);
+		break;
+	case 0x30:
+		info->voltage = nv_ro08(bios, volt + 0x00) * 10000;
+		info->vid     = nv_ro08(bios, volt + 0x01) >> 2;
+		break;
+	case 0x40:
+	case 0x50:
+		break;
+	}
+	return volt;
+}
diff --git a/drivers/gpu/drm/nouveau/core/subdev/bus/hwsq.c b/drivers/gpu/drm/nouveau/core/subdev/bus/hwsq.c
new file mode 100644
index 000000000000..f757470e2284
--- /dev/null
+++ b/drivers/gpu/drm/nouveau/core/subdev/bus/hwsq.c
@@ -0,0 +1,145 @@
+/*
+ * Copyright 2013 Red Hat Inc.
+ *
+ * Permission is hereby granted, free of charge, to any person obtaining a
+ * copy of this software and associated documentation files (the "Software"),
+ * to deal in the Software without restriction, including without limitation
+ * the rights to use, copy, modify, merge, publish, distribute, sublicense,
+ * and/or sell copies of the Software, and to permit persons to whom the
+ * Software is furnished to do so, subject to the following conditions:
+ *
+ * The above copyright notice and this permission notice shall be included in
+ * all copies or substantial portions of the Software.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+ * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+ * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT.  IN NO EVENT SHALL
+ * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR
+ * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE,
+ * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR
+ * OTHER DEALINGS IN THE SOFTWARE.
+ *
+ * Authors: Ben Skeggs <bskeggs@redhat.com>
+ */
+
+#include <subdev/timer.h>
+#include <subdev/bus.h>
+
+struct nouveau_hwsq {
+	struct nouveau_bus *pbus;
+	u32 addr;
+	u32 data;
+	struct {
+		u8 data[512];
+		u8 size;
+	} c;
+};
+
+static void
+hwsq_cmd(struct nouveau_hwsq *hwsq, int size, u8 data[])
+{
+	memcpy(&hwsq->c.data[hwsq->c.size], data, size * sizeof(data[0]));
+	hwsq->c.size += size;
+}
+
+int
+nouveau_hwsq_init(struct nouveau_bus *pbus, struct nouveau_hwsq **phwsq)
+{
+	struct nouveau_hwsq *hwsq;
+
+	hwsq = *phwsq = kmalloc(sizeof(*hwsq), GFP_KERNEL);
+	if (hwsq) {
+		hwsq->pbus = pbus;
+		hwsq->addr = ~0;
+		hwsq->data = ~0;
+		memset(hwsq->c.data, 0x7f, sizeof(hwsq->c.data));
+		hwsq->c.size = 0;
+	}
+
+	return hwsq ? 0 : -ENOMEM;
+}
+
+int
+nouveau_hwsq_fini(struct nouveau_hwsq **phwsq, bool exec)
+{
+	struct nouveau_hwsq *hwsq = *phwsq;
+	int ret = 0, i;
+	if (hwsq) {
+		struct nouveau_bus *pbus = hwsq->pbus;
+		hwsq->c.size = (hwsq->c.size + 4) / 4;
+		if (hwsq->c.size <= pbus->hwsq_size) {
+			if (exec)
+				ret = pbus->hwsq_exec(pbus, (u32 *)hwsq->c.data,
+						      hwsq->c.size);
+			if (ret)
+				nv_error(pbus, "hwsq exec failed: %d\n", ret);
+		} else {
+			nv_error(pbus, "hwsq ucode too large\n");
+			ret = -ENOSPC;
+		}
+
+		for (i = 0; ret && i < hwsq->c.size; i++)
+			nv_error(pbus, "\t0x%08x\n", ((u32 *)hwsq->c.data)[i]);
+
+		*phwsq = NULL;
+		kfree(hwsq);
+	}
+	return ret;
+}
+
+void
+nouveau_hwsq_wr32(struct nouveau_hwsq *hwsq, u32 addr, u32 data)
+{
+	nv_debug(hwsq->pbus, "R[%06x] = 0x%08x\n", addr, data);
+
+	if (hwsq->data != data) {
+		if ((data & 0xffff0000) != (hwsq->data & 0xffff0000)) {
+			hwsq_cmd(hwsq, 5, (u8[]){ 0xe2, data, data >> 8,
+						  data >> 16, data >> 24 });
+		} else {
+			hwsq_cmd(hwsq, 3, (u8[]){ 0x42, data, data >> 8 });
+		}
+	}
+
+	if ((addr & 0xffff0000) != (hwsq->addr & 0xffff0000)) {
+		hwsq_cmd(hwsq, 5, (u8[]){ 0xe0, addr, addr >> 8,
+					  addr >> 16, addr >> 24 });
+	} else {
+		hwsq_cmd(hwsq, 3, (u8[]){ 0x40, addr, addr >> 8 });
+	}
+
+	hwsq->addr = addr;
+	hwsq->data = data;
+}
+
+void
+nouveau_hwsq_setf(struct nouveau_hwsq *hwsq, u8 flag, int data)
+{
+	nv_debug(hwsq->pbus, " FLAG[%02x] = %d\n", flag, data);
+	flag += 0x80;
+	if (data >= 0)
+		flag += 0x20;
+	if (data >= 1)
+		flag += 0x20;
+	hwsq_cmd(hwsq, 1, (u8[]){ flag });
+}
+
+void
+nouveau_hwsq_wait(struct nouveau_hwsq *hwsq, u8 flag, u8 data)
+{
+	nv_debug(hwsq->pbus, " WAIT[%02x] = %d\n", flag, data);
+	hwsq_cmd(hwsq, 3, (u8[]){ 0x5f, flag, data });
+}
+
+void
+nouveau_hwsq_nsec(struct nouveau_hwsq *hwsq, u32 nsec)
+{
+	u8 shift = 0, usec = nsec / 1000;
+	while (usec & ~3) {
+		usec >>= 2;
+		shift++;
+	}
+
+	nv_debug(hwsq->pbus, "    DELAY = %d ns\n", nsec);
+	hwsq_cmd(hwsq, 1, (u8[]){ 0x00 | (shift << 2) | usec });
+}
diff --git a/drivers/gpu/drm/nouveau/core/subdev/bus/hwsq.h b/drivers/gpu/drm/nouveau/core/subdev/bus/hwsq.h
new file mode 100644
index 000000000000..12176f9c1bc6
--- /dev/null
+++ b/drivers/gpu/drm/nouveau/core/subdev/bus/hwsq.h
@@ -0,0 +1,113 @@
+#ifndef __NVKM_BUS_HWSQ_H__
+#define __NVKM_BUS_HWSQ_H__
+
+#include <subdev/bus.h>
+
+struct hwsq {
+	struct nouveau_subdev *subdev;
+	struct nouveau_hwsq *hwsq;
+	int sequence;
+};
+
+struct hwsq_reg {
+	int sequence;
+	bool force;
+	u32 addr[2];
+	u32 data;
+};
+
+static inline struct hwsq_reg
+hwsq_reg2(u32 addr1, u32 addr2)
+{
+	return (struct hwsq_reg) {
+		.sequence = 0,
+		.force = 0,
+		.addr = { addr1, addr2 },
+		.data = 0xdeadbeef,
+	};
+}
+
+static inline struct hwsq_reg
+hwsq_reg(u32 addr)
+{
+	return hwsq_reg2(addr, addr);
+}
+
+static inline int
+hwsq_init(struct hwsq *ram, struct nouveau_subdev *subdev)
+{
+	struct nouveau_bus *pbus = nouveau_bus(subdev);
+	int ret;
+
+	ret = nouveau_hwsq_init(pbus, &ram->hwsq);
+	if (ret)
+		return ret;
+
+	ram->sequence++;
+	ram->subdev = subdev;
+	return 0;
+}
+
+static inline int
+hwsq_exec(struct hwsq *ram, bool exec)
+{
+	int ret = 0;
+	if (ram->subdev) {
+		ret = nouveau_hwsq_fini(&ram->hwsq, exec);
+		ram->subdev = NULL;
+	}
+	return ret;
+}
+
+static inline u32
+hwsq_rd32(struct hwsq *ram, struct hwsq_reg *reg)
+{
+	if (reg->sequence != ram->sequence)
+		reg->data = nv_rd32(ram->subdev, reg->addr[0]);
+	return reg->data;
+}
+
+static inline void
+hwsq_wr32(struct hwsq *ram, struct hwsq_reg *reg, u32 data)
+{
+	reg->sequence = ram->sequence;
+	reg->data = data;
+	if (reg->addr[0] != reg->addr[1])
+		nouveau_hwsq_wr32(ram->hwsq, reg->addr[1], reg->data);
+	nouveau_hwsq_wr32(ram->hwsq, reg->addr[0], reg->data);
+}
+
+static inline void
+hwsq_nuke(struct hwsq *ram, struct hwsq_reg *reg)
+{
+	reg->force = true;
+}
+
+static inline u32
+hwsq_mask(struct hwsq *ram, struct hwsq_reg *reg, u32 mask, u32 data)
+{
+	u32 temp = hwsq_rd32(ram, reg);
+	if (temp != ((temp & ~mask) | data) || reg->force)
+		hwsq_wr32(ram, reg, (temp & ~mask) | data);
+	return temp;
+}
+
+static inline void
+hwsq_setf(struct hwsq *ram, u8 flag, int data)
+{
+	nouveau_hwsq_setf(ram->hwsq, flag, data);
+}
+
+static inline void
+hwsq_wait(struct hwsq *ram, u8 flag, u8 data)
+{
+	nouveau_hwsq_wait(ram->hwsq, flag, data);
+}
+
+static inline void
+hwsq_nsec(struct hwsq *ram, u32 nsec)
+{
+	nouveau_hwsq_nsec(ram->hwsq, nsec);
+}
+
+#endif
diff --git a/drivers/gpu/drm/nouveau/core/subdev/bus/nv04.c b/drivers/gpu/drm/nouveau/core/subdev/bus/nv04.c
index 8c7f8057a185..23921b5351db 100644
--- a/drivers/gpu/drm/nouveau/core/subdev/bus/nv04.c
+++ b/drivers/gpu/drm/nouveau/core/subdev/bus/nv04.c
@@ -23,11 +23,7 @@
  *          Ben Skeggs
  */
 
-#include <subdev/bus.h>
-
-struct nv04_bus_priv {
-	struct nouveau_bus base;
-};
+#include "nv04.h"
 
 static void
 nv04_bus_intr(struct nouveau_subdev *subdev)
@@ -56,10 +52,22 @@ nv04_bus_intr(struct nouveau_subdev *subdev)
 }
 
 static int
+nv04_bus_init(struct nouveau_object *object)
+{
+	struct nv04_bus_priv *priv = (void *)object;
+
+	nv_wr32(priv, 0x001100, 0xffffffff);
+	nv_wr32(priv, 0x001140, 0x00000111);
+
+	return nouveau_bus_init(&priv->base);
+}
+
+int
 nv04_bus_ctor(struct nouveau_object *parent, struct nouveau_object *engine,
 	      struct nouveau_oclass *oclass, void *data, u32 size,
 	      struct nouveau_object **pobject)
 {
+	struct nv04_bus_impl *impl = (void *)oclass;
 	struct nv04_bus_priv *priv;
 	int ret;
 
@@ -68,28 +76,20 @@ nv04_bus_ctor(struct nouveau_object *parent, struct nouveau_object *engine,
 	if (ret)
 		return ret;
 
-	nv_subdev(priv)->intr = nv04_bus_intr;
+	nv_subdev(priv)->intr = impl->intr;
+	priv->base.hwsq_exec = impl->hwsq_exec;
+	priv->base.hwsq_size = impl->hwsq_size;
 	return 0;
 }
 
-static int
-nv04_bus_init(struct nouveau_object *object)
-{
-	struct nv04_bus_priv *priv = (void *)object;
-
-	nv_wr32(priv, 0x001100, 0xffffffff);
-	nv_wr32(priv, 0x001140, 0x00000111);
-
-	return nouveau_bus_init(&priv->base);
-}
-
-struct nouveau_oclass
-nv04_bus_oclass = {
-	.handle = NV_SUBDEV(BUS, 0x04),
-	.ofuncs = &(struct nouveau_ofuncs) {
+struct nouveau_oclass *
+nv04_bus_oclass = &(struct nv04_bus_impl) {
+	.base.handle = NV_SUBDEV(BUS, 0x04),
+	.base.ofuncs = &(struct nouveau_ofuncs) {
 		.ctor = nv04_bus_ctor,
 		.dtor = _nouveau_bus_dtor,
 		.init = nv04_bus_init,
 		.fini = _nouveau_bus_fini,
 	},
-};
+	.intr = nv04_bus_intr,
+}.base;
diff --git a/drivers/gpu/drm/nouveau/core/subdev/bus/nv04.h b/drivers/gpu/drm/nouveau/core/subdev/bus/nv04.h
new file mode 100644
index 000000000000..4d7602450a20
--- /dev/null
+++ b/drivers/gpu/drm/nouveau/core/subdev/bus/nv04.h
@@ -0,0 +1,23 @@
+#ifndef __NVKM_BUS_NV04_H__
+#define __NVKM_BUS_NV04_H__
+
+#include <subdev/bus.h>
+
+struct nv04_bus_priv {
+	struct nouveau_bus base;
+};
+
+int  nv04_bus_ctor(struct nouveau_object *, struct nouveau_object *,
+		   struct nouveau_oclass *, void *, u32,
+		   struct nouveau_object **);
+int  nv50_bus_init(struct nouveau_object *);
+void nv50_bus_intr(struct nouveau_subdev *);
+
+struct nv04_bus_impl {
+	struct nouveau_oclass base;
+	void (*intr)(struct nouveau_subdev *);
+	int  (*hwsq_exec)(struct nouveau_bus *, u32 *, u32);
+	u32  hwsq_size;
+};
+
+#endif
diff --git a/drivers/gpu/drm/nouveau/core/subdev/bus/nv31.c b/drivers/gpu/drm/nouveau/core/subdev/bus/nv31.c
index 34132aef34e1..94da46f61627 100644
--- a/drivers/gpu/drm/nouveau/core/subdev/bus/nv31.c
+++ b/drivers/gpu/drm/nouveau/core/subdev/bus/nv31.c
@@ -23,11 +23,7 @@
  *          Ben Skeggs
  */
 
-#include <subdev/bus.h>
-
-struct nv31_bus_priv {
-	struct nouveau_bus base;
-};
+#include "nv04.h"
 
 static void
 nv31_bus_intr(struct nouveau_subdev *subdev)
@@ -71,7 +67,7 @@ nv31_bus_intr(struct nouveau_subdev *subdev)
 static int
 nv31_bus_init(struct nouveau_object *object)
 {
-	struct nv31_bus_priv *priv = (void *)object;
+	struct nv04_bus_priv *priv = (void *)object;
 	int ret;
 
 	ret = nouveau_bus_init(&priv->base);
@@ -83,30 +79,14 @@ nv31_bus_init(struct nouveau_object *object)
 	return 0;
 }
 
-static int
-nv31_bus_ctor(struct nouveau_object *parent, struct nouveau_object *engine,
-	      struct nouveau_oclass *oclass, void *data, u32 size,
-	      struct nouveau_object **pobject)
-{
-	struct nv31_bus_priv *priv;
-	int ret;
-
-	ret = nouveau_bus_create(parent, engine, oclass, &priv);
-	*pobject = nv_object(priv);
-	if (ret)
-		return ret;
-
-	nv_subdev(priv)->intr = nv31_bus_intr;
-	return 0;
-}
-
-struct nouveau_oclass
-nv31_bus_oclass = {
-	.handle = NV_SUBDEV(BUS, 0x31),
-	.ofuncs = &(struct nouveau_ofuncs) {
-		.ctor = nv31_bus_ctor,
+struct nouveau_oclass *
+nv31_bus_oclass = &(struct nv04_bus_impl) {
+	.base.handle = NV_SUBDEV(BUS, 0x31),
+	.base.ofuncs = &(struct nouveau_ofuncs) {
+		.ctor = nv04_bus_ctor,
 		.dtor = _nouveau_bus_dtor,
 		.init = nv31_bus_init,
 		.fini = _nouveau_bus_fini,
 	},
-};
+	.intr = nv31_bus_intr,
+}.base;
diff --git a/drivers/gpu/drm/nouveau/core/subdev/bus/nv50.c b/drivers/gpu/drm/nouveau/core/subdev/bus/nv50.c
index f5b2117fa8c6..11918f7e2aca 100644
--- a/drivers/gpu/drm/nouveau/core/subdev/bus/nv50.c
+++ b/drivers/gpu/drm/nouveau/core/subdev/bus/nv50.c
@@ -23,13 +23,27 @@
  *          Ben Skeggs
  */
 
-#include <subdev/bus.h>
+#include <subdev/timer.h>
 
-struct nv50_bus_priv {
-	struct nouveau_bus base;
-};
+#include "nv04.h"
 
-static void
+static int
+nv50_bus_hwsq_exec(struct nouveau_bus *pbus, u32 *data, u32 size)
+{
+	struct nv50_bus_priv *priv = (void *)pbus;
+	int i;
+
+	nv_mask(pbus, 0x001098, 0x00000008, 0x00000000);
+	nv_wr32(pbus, 0x001304, 0x00000000);
+	for (i = 0; i < size; i++)
+		nv_wr32(priv, 0x001400 + (i * 4), data[i]);
+	nv_mask(pbus, 0x001098, 0x00000018, 0x00000018);
+	nv_wr32(pbus, 0x00130c, 0x00000003);
+
+	return nv_wait(pbus, 0x001308, 0x00000100, 0x00000000) ? 0 : -ETIMEDOUT;
+}
+
+void
 nv50_bus_intr(struct nouveau_subdev *subdev)
 {
 	struct nouveau_bus *pbus = nouveau_bus(subdev);
@@ -61,10 +75,10 @@ nv50_bus_intr(struct nouveau_subdev *subdev)
 	}
 }
 
-static int
+int
 nv50_bus_init(struct nouveau_object *object)
 {
-	struct nv50_bus_priv *priv = (void *)object;
+	struct nv04_bus_priv *priv = (void *)object;
 	int ret;
 
 	ret = nouveau_bus_init(&priv->base);
@@ -76,30 +90,16 @@ nv50_bus_init(struct nouveau_object *object)
 	return 0;
 }
 
-static int
-nv50_bus_ctor(struct nouveau_object *parent, struct nouveau_object *engine,
-	      struct nouveau_oclass *oclass, void *data, u32 size,
-	      struct nouveau_object **pobject)
-{
-	struct nv50_bus_priv *priv;
-	int ret;
-
-	ret = nouveau_bus_create(parent, engine, oclass, &priv);
-	*pobject = nv_object(priv);
-	if (ret)
-		return ret;
-
-	nv_subdev(priv)->intr = nv50_bus_intr;
-	return 0;
-}
-
-struct nouveau_oclass
-nv50_bus_oclass = {
-	.handle = NV_SUBDEV(BUS, 0x50),
-	.ofuncs = &(struct nouveau_ofuncs) {
-		.ctor = nv50_bus_ctor,
+struct nouveau_oclass *
+nv50_bus_oclass = &(struct nv04_bus_impl) {
+	.base.handle = NV_SUBDEV(BUS, 0x50),
+	.base.ofuncs = &(struct nouveau_ofuncs) {
+		.ctor = nv04_bus_ctor,
 		.dtor = _nouveau_bus_dtor,
 		.init = nv50_bus_init,
 		.fini = _nouveau_bus_fini,
 	},
-};
+	.intr = nv50_bus_intr,
+	.hwsq_exec = nv50_bus_hwsq_exec,
+	.hwsq_size = 64,
+}.base;
diff --git a/drivers/gpu/drm/nouveau/core/subdev/bus/nv94.c b/drivers/gpu/drm/nouveau/core/subdev/bus/nv94.c
new file mode 100644
index 000000000000..d3659055fa4b
--- /dev/null
+++ b/drivers/gpu/drm/nouveau/core/subdev/bus/nv94.c
@@ -0,0 +1,59 @@
+/*
+ * Copyright 2012 Nouveau Community
+ *
+ * Permission is hereby granted, free of charge, to any person obtaining a
+ * copy of this software and associated documentation files (the "Software"),
+ * to deal in the Software without restriction, including without limitation
+ * the rights to use, copy, modify, merge, publish, distribute, sublicense,
+ * and/or sell copies of the Software, and to permit persons to whom the
+ * Software is furnished to do so, subject to the following conditions:
+ *
+ * The above copyright notice and this permission notice shall be included in
+ * all copies or substantial portions of the Software.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+ * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+ * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT.  IN NO EVENT SHALL
+ * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR
+ * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE,
+ * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR
+ * OTHER DEALINGS IN THE SOFTWARE.
+ *
+ * Authors: Martin Peres <martin.peres@labri.fr>
+ *          Ben Skeggs
+ */
+
+#include <subdev/timer.h>
+
+#include "nv04.h"
+
+static int
+nv94_bus_hwsq_exec(struct nouveau_bus *pbus, u32 *data, u32 size)
+{
+	struct nv50_bus_priv *priv = (void *)pbus;
+	int i;
+
+	nv_mask(pbus, 0x001098, 0x00000008, 0x00000000);
+	nv_wr32(pbus, 0x001304, 0x00000000);
+	nv_wr32(pbus, 0x001318, 0x00000000);
+	for (i = 0; i < size; i++)
+		nv_wr32(priv, 0x080000 + (i * 4), data[i]);
+	nv_mask(pbus, 0x001098, 0x00000018, 0x00000018);
+	nv_wr32(pbus, 0x00130c, 0x00000001);
+
+	return nv_wait(pbus, 0x001308, 0x00000100, 0x00000000) ? 0 : -ETIMEDOUT;
+}
+
+struct nouveau_oclass *
+nv94_bus_oclass = &(struct nv04_bus_impl) {
+	.base.handle = NV_SUBDEV(BUS, 0x94),
+	.base.ofuncs = &(struct nouveau_ofuncs) {
+		.ctor = nv04_bus_ctor,
+		.dtor = _nouveau_bus_dtor,
+		.init = nv50_bus_init,
+		.fini = _nouveau_bus_fini,
+	},
+	.intr = nv50_bus_intr,
+	.hwsq_exec = nv94_bus_hwsq_exec,
+	.hwsq_size = 128,
+}.base;
diff --git a/drivers/gpu/drm/nouveau/core/subdev/bus/nvc0.c b/drivers/gpu/drm/nouveau/core/subdev/bus/nvc0.c
index b192d6246363..73839d7151a7 100644
--- a/drivers/gpu/drm/nouveau/core/subdev/bus/nvc0.c
+++ b/drivers/gpu/drm/nouveau/core/subdev/bus/nvc0.c
@@ -23,11 +23,7 @@
  *          Ben Skeggs
  */
 
-#include <subdev/bus.h>
-
-struct nvc0_bus_priv {
-	struct nouveau_bus base;
-};
+#include "nv04.h"
 
 static void
 nvc0_bus_intr(struct nouveau_subdev *subdev)
@@ -60,7 +56,7 @@ nvc0_bus_intr(struct nouveau_subdev *subdev)
 static int
 nvc0_bus_init(struct nouveau_object *object)
 {
-	struct nvc0_bus_priv *priv = (void *)object;
+	struct nv04_bus_priv *priv = (void *)object;
 	int ret;
 
 	ret = nouveau_bus_init(&priv->base);
@@ -72,30 +68,14 @@ nvc0_bus_init(struct nouveau_object *object)
 	return 0;
 }
 
-static int
-nvc0_bus_ctor(struct nouveau_object *parent, struct nouveau_object *engine,
-	      struct nouveau_oclass *oclass, void *data, u32 size,
-	      struct nouveau_object **pobject)
-{
-	struct nvc0_bus_priv *priv;
-	int ret;
-
-	ret = nouveau_bus_create(parent, engine, oclass, &priv);
-	*pobject = nv_object(priv);
-	if (ret)
-		return ret;
-
-	nv_subdev(priv)->intr = nvc0_bus_intr;
-	return 0;
-}
-
-struct nouveau_oclass
-nvc0_bus_oclass = {
-	.handle = NV_SUBDEV(BUS, 0xc0),
-	.ofuncs = &(struct nouveau_ofuncs) {
-		.ctor = nvc0_bus_ctor,
+struct nouveau_oclass *
+nvc0_bus_oclass = &(struct nv04_bus_impl) {
+	.base.handle = NV_SUBDEV(BUS, 0xc0),
+	.base.ofuncs = &(struct nouveau_ofuncs) {
+		.ctor = nv04_bus_ctor,
 		.dtor = _nouveau_bus_dtor,
 		.init = nvc0_bus_init,
 		.fini = _nouveau_bus_fini,
 	},
-};
+	.intr = nvc0_bus_intr,
+}.base;
diff --git a/drivers/gpu/drm/nouveau/core/subdev/clock/base.c b/drivers/gpu/drm/nouveau/core/subdev/clock/base.c
new file mode 100644
index 000000000000..e2938a21b06f
--- /dev/null
+++ b/drivers/gpu/drm/nouveau/core/subdev/clock/base.c
@@ -0,0 +1,494 @@
+/*
+ * Copyright 2013 Red Hat Inc.
+ *
+ * Permission is hereby granted, free of charge, to any person obtaining a
+ * copy of this software and associated documentation files (the "Software"),
+ * to deal in the Software without restriction, including without limitation
+ * the rights to use, copy, modify, merge, publish, distribute, sublicense,
+ * and/or sell copies of the Software, and to permit persons to whom the
+ * Software is furnished to do so, subject to the following conditions:
+ *
+ * The above copyright notice and this permission notice shall be included in
+ * all copies or substantial portions of the Software.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+ * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+ * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT.  IN NO EVENT SHALL
+ * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR
+ * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE,
+ * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR
+ * OTHER DEALINGS IN THE SOFTWARE.
+ *
+ * Authors: Ben Skeggs
+ */
+
+#include <core/option.h>
+
+#include <subdev/clock.h>
+#include <subdev/therm.h>
+#include <subdev/volt.h>
+#include <subdev/fb.h>
+
+#include <subdev/bios.h>
+#include <subdev/bios/boost.h>
+#include <subdev/bios/cstep.h>
+#include <subdev/bios/perf.h>
+
+/******************************************************************************
+ * misc
+ *****************************************************************************/
+static u32
+nouveau_clock_adjust(struct nouveau_clock *clk, bool adjust,
+		     u8 pstate, u8 domain, u32 input)
+{
+	struct nouveau_bios *bios = nouveau_bios(clk);
+	struct nvbios_boostE boostE;
+	u8  ver, hdr, cnt, len;
+	u16 data;
+
+	data = nvbios_boostEm(bios, pstate, &ver, &hdr, &cnt, &len, &boostE);
+	if (data) {
+		struct nvbios_boostS boostS;
+		u8  idx = 0, sver, shdr;
+		u16 subd;
+
+		input = max(boostE.min, input);
+		input = min(boostE.max, input);
+		do {
+			sver = ver;
+			shdr = hdr;
+			subd = nvbios_boostSp(bios, idx++, data, &sver, &shdr,
+					      cnt, len, &boostS);
+			if (subd && boostS.domain == domain) {
+				if (adjust)
+					input = input * boostS.percent / 100;
+				input = max(boostS.min, input);
+				input = min(boostS.max, input);
+				break;
+			}
+		} while (subd);
+	}
+
+	return input;
+}
+
+/******************************************************************************
+ * C-States
+ *****************************************************************************/
+static int
+nouveau_cstate_prog(struct nouveau_clock *clk,
+		    struct nouveau_pstate *pstate, int cstatei)
+{
+	struct nouveau_therm *ptherm = nouveau_therm(clk);
+	struct nouveau_volt *volt = nouveau_volt(clk);
+	struct nouveau_cstate *cstate;
+	int ret;
+
+	if (!list_empty(&pstate->list)) {
+		cstate = list_entry(pstate->list.prev, typeof(*cstate), head);
+	} else {
+		cstate = &pstate->base;
+	}
+
+	ret = nouveau_therm_cstate(ptherm, pstate->fanspeed, +1);
+	if (ret && ret != -ENODEV) {
+		nv_error(clk, "failed to raise fan speed: %d\n", ret);
+		return ret;
+	}
+
+	ret = volt->set_id(volt, cstate->voltage, +1);
+	if (ret && ret != -ENODEV) {
+		nv_error(clk, "failed to raise voltage: %d\n", ret);
+		return ret;
+	}
+
+	ret = clk->calc(clk, cstate);
+	if (ret == 0) {
+		ret = clk->prog(clk);
+		clk->tidy(clk);
+	}
+
+	ret = volt->set_id(volt, cstate->voltage, -1);
+	if (ret && ret != -ENODEV)
+		nv_error(clk, "failed to lower voltage: %d\n", ret);
+
+	ret = nouveau_therm_cstate(ptherm, pstate->fanspeed, -1);
+	if (ret && ret != -ENODEV)
+		nv_error(clk, "failed to lower fan speed: %d\n", ret);
+
+	return 0;
+}
+
+static void
+nouveau_cstate_del(struct nouveau_cstate *cstate)
+{
+	list_del(&cstate->head);
+	kfree(cstate);
+}
+
+static int
+nouveau_cstate_new(struct nouveau_clock *clk, int idx,
+		   struct nouveau_pstate *pstate)
+{
+	struct nouveau_bios *bios = nouveau_bios(clk);
+	struct nouveau_clocks *domain = clk->domains;
+	struct nouveau_cstate *cstate = NULL;
+	struct nvbios_cstepX cstepX;
+	u8  ver, hdr;
+	u16 data;
+
+	data = nvbios_cstepXp(bios, idx, &ver, &hdr, &cstepX);
+	if (!data)
+		return -ENOENT;
+
+	cstate = kzalloc(sizeof(*cstate), GFP_KERNEL);
+	if (!cstate)
+		return -ENOMEM;
+
+	*cstate = pstate->base;
+	cstate->voltage = cstepX.voltage;
+
+	while (domain && domain->name != nv_clk_src_max) {
+		if (domain->flags & NVKM_CLK_DOM_FLAG_CORE) {
+			u32 freq = nouveau_clock_adjust(clk, true,
+							pstate->pstate,
+							domain->bios,
+							cstepX.freq);
+			cstate->domain[domain->name] = freq;
+		}
+		domain++;
+	}
+
+	list_add(&cstate->head, &pstate->list);
+	return 0;
+}
+
+/******************************************************************************
+ * P-States
+ *****************************************************************************/
+static int
+nouveau_pstate_prog(struct nouveau_clock *clk, int pstatei)
+{
+	struct nouveau_fb *pfb = nouveau_fb(clk);
+	struct nouveau_pstate *pstate;
+	int ret, idx = 0;
+
+	list_for_each_entry(pstate, &clk->states, head) {
+		if (idx++ == pstatei)
+			break;
+	}
+
+	nv_debug(clk, "setting performance state %d\n", pstatei);
+	clk->pstate = pstatei;
+
+	if (pfb->ram->calc) {
+		ret = pfb->ram->calc(pfb, pstate->base.domain[nv_clk_src_mem]);
+		if (ret == 0)
+			ret = pfb->ram->prog(pfb);
+		pfb->ram->tidy(pfb);
+	}
+
+	return nouveau_cstate_prog(clk, pstate, 0);
+}
+
+static int
+nouveau_pstate_calc(struct nouveau_clock *clk)
+{
+	int pstate, ret = 0;
+
+	nv_trace(clk, "P %d U %d A %d T %d D %d\n", clk->pstate,
+		 clk->ustate, clk->astate, clk->tstate, clk->dstate);
+
+	if (clk->state_nr && clk->ustate != -1) {
+		pstate = (clk->ustate < 0) ? clk->astate : clk->ustate;
+		pstate = min(pstate, clk->state_nr - 1 - clk->tstate);
+		pstate = max(pstate, clk->dstate);
+	} else {
+		pstate = clk->pstate = -1;
+	}
+
+	nv_trace(clk, "-> %d\n", pstate);
+	if (pstate != clk->pstate)
+		ret = nouveau_pstate_prog(clk, pstate);
+	return ret;
+}
+
+static void
+nouveau_pstate_info(struct nouveau_clock *clk, struct nouveau_pstate *pstate)
+{
+	struct nouveau_clocks *clock = clk->domains - 1;
+	struct nouveau_cstate *cstate;
+	char info[3][32] = { "", "", "" };
+	char name[4] = "--";
+	int i = -1;
+
+	if (pstate->pstate != 0xff)
+		snprintf(name, sizeof(name), "%02x", pstate->pstate);
+
+	while ((++clock)->name != nv_clk_src_max) {
+		u32 lo = pstate->base.domain[clock->name];
+		u32 hi = lo;
+		if (hi == 0)
+			continue;
+
+		nv_debug(clk, "%02x: %10d KHz\n", clock->name, lo);
+		list_for_each_entry(cstate, &pstate->list, head) {
+			u32 freq = cstate->domain[clock->name];
+			lo = min(lo, freq);
+			hi = max(hi, freq);
+			nv_debug(clk, "%10d KHz\n", freq);
+		}
+
+		if (clock->mname && ++i < ARRAY_SIZE(info)) {
+			lo /= clock->mdiv;
+			hi /= clock->mdiv;
+			if (lo == hi) {
+				snprintf(info[i], sizeof(info[i]), "%s %d MHz",
+					 clock->mname, lo);
+			} else {
+				snprintf(info[i], sizeof(info[i]),
+					 "%s %d-%d MHz", clock->mname, lo, hi);
+			}
+		}
+	}
+
+	nv_info(clk, "%s: %s %s %s\n", name, info[0], info[1], info[2]);
+}
+
+static void
+nouveau_pstate_del(struct nouveau_pstate *pstate)
+{
+	struct nouveau_cstate *cstate, *temp;
+
+	list_for_each_entry_safe(cstate, temp, &pstate->list, head) {
+		nouveau_cstate_del(cstate);
+	}
+
+	list_del(&pstate->head);
+	kfree(pstate);
+}
+
+static int
+nouveau_pstate_new(struct nouveau_clock *clk, int idx)
+{
+	struct nouveau_bios *bios = nouveau_bios(clk);
+	struct nouveau_clocks *domain = clk->domains - 1;
+	struct nouveau_pstate *pstate;
+	struct nouveau_cstate *cstate;
+	struct nvbios_cstepE cstepE;
+	struct nvbios_perfE perfE;
+	u8  ver, hdr, cnt, len;
+	u16 data;
+
+	data = nvbios_perfEp(bios, idx, &ver, &hdr, &cnt, &len, &perfE);
+	if (!data)
+		return -EINVAL;
+	if (perfE.pstate == 0xff)
+		return 0;
+
+	pstate = kzalloc(sizeof(*pstate), GFP_KERNEL);
+	cstate = &pstate->base;
+	if (!pstate)
+		return -ENOMEM;
+
+	INIT_LIST_HEAD(&pstate->list);
+
+	pstate->pstate = perfE.pstate;
+	pstate->fanspeed = perfE.fanspeed;
+	cstate->voltage = perfE.voltage;
+	cstate->domain[nv_clk_src_core] = perfE.core;
+	cstate->domain[nv_clk_src_shader] = perfE.shader;
+	cstate->domain[nv_clk_src_mem] = perfE.memory;
+	cstate->domain[nv_clk_src_vdec] = perfE.vdec;
+	cstate->domain[nv_clk_src_dom6] = perfE.disp;
+
+	while (ver >= 0x40 && (++domain)->name != nv_clk_src_max) {
+		struct nvbios_perfS perfS;
+		u8  sver = ver, shdr = hdr;
+		u32 perfSe = nvbios_perfSp(bios, data, domain->bios,
+					  &sver, &shdr, cnt, len, &perfS);
+		if (perfSe == 0 || sver != 0x40)
+			continue;
+
+		if (domain->flags & NVKM_CLK_DOM_FLAG_CORE) {
+			perfS.v40.freq = nouveau_clock_adjust(clk, false,
+							      pstate->pstate,
+							      domain->bios,
+							      perfS.v40.freq);
+		}
+
+		cstate->domain[domain->name] = perfS.v40.freq;
+	}
+
+	data = nvbios_cstepEm(bios, pstate->pstate, &ver, &hdr, &cstepE);
+	if (data) {
+		int idx = cstepE.index;
+		do {
+			nouveau_cstate_new(clk, idx, pstate);
+		} while(idx--);
+	}
+
+	nouveau_pstate_info(clk, pstate);
+	list_add_tail(&pstate->head, &clk->states);
+	clk->state_nr++;
+	return 0;
+}
+
+/******************************************************************************
+ * Adjustment triggers
+ *****************************************************************************/
+static int
+nouveau_clock_ustate_update(struct nouveau_clock *clk, int req)
+{
+	struct nouveau_pstate *pstate;
+	int i = 0;
+
+	/* YKW repellant */
+	return -ENOSYS;
+
+	if (req != -1 && req != -2) {
+		list_for_each_entry(pstate, &clk->states, head) {
+			if (pstate->pstate == req)
+				break;
+			i++;
+		}
+
+		if (pstate->pstate != req)
+			return -EINVAL;
+		req = i;
+	}
+
+	clk->ustate = req;
+	return 0;
+}
+
+int
+nouveau_clock_ustate(struct nouveau_clock *clk, int req)
+{
+	int ret = nouveau_clock_ustate_update(clk, req);
+	if (ret)
+		return ret;
+	return nouveau_pstate_calc(clk);
+}
+
+int
+nouveau_clock_astate(struct nouveau_clock *clk, int req, int rel)
+{
+	if (!rel) clk->astate  = req;
+	if ( rel) clk->astate += rel;
+	clk->astate = min(clk->astate, clk->state_nr - 1);
+	clk->astate = max(clk->astate, 0);
+	return nouveau_pstate_calc(clk);
+}
+
+int
+nouveau_clock_tstate(struct nouveau_clock *clk, int req, int rel)
+{
+	if (!rel) clk->tstate  = req;
+	if ( rel) clk->tstate += rel;
+	clk->tstate = min(clk->tstate, 0);
+	clk->tstate = max(clk->tstate, -(clk->state_nr - 1));
+	return nouveau_pstate_calc(clk);
+}
+
+int
+nouveau_clock_dstate(struct nouveau_clock *clk, int req, int rel)
+{
+	if (!rel) clk->dstate  = req;
+	if ( rel) clk->dstate += rel;
+	clk->dstate = min(clk->dstate, clk->state_nr - 1);
+	clk->dstate = max(clk->dstate, 0);
+	return nouveau_pstate_calc(clk);
+}
+
+/******************************************************************************
+ * subdev base class implementation
+ *****************************************************************************/
+int
+_nouveau_clock_init(struct nouveau_object *object)
+{
+	struct nouveau_clock *clk = (void *)object;
+	struct nouveau_clocks *clock = clk->domains;
+	int ret;
+
+	memset(&clk->bstate, 0x00, sizeof(clk->bstate));
+	INIT_LIST_HEAD(&clk->bstate.list);
+	clk->bstate.pstate = 0xff;
+
+	while (clock->name != nv_clk_src_max) {
+		ret = clk->read(clk, clock->name);
+		if (ret < 0) {
+			nv_error(clk, "%02x freq unknown\n", clock->name);
+			return ret;
+		}
+		clk->bstate.base.domain[clock->name] = ret;
+		clock++;
+	}
+
+	nouveau_pstate_info(clk, &clk->bstate);
+
+	clk->astate = clk->state_nr - 1;
+	clk->tstate = 0;
+	clk->dstate = 0;
+	clk->pstate = -1;
+	nouveau_pstate_calc(clk);
+	return 0;
+}
+
+void
+_nouveau_clock_dtor(struct nouveau_object *object)
+{
+	struct nouveau_clock *clk = (void *)object;
+	struct nouveau_pstate *pstate, *temp;
+
+	list_for_each_entry_safe(pstate, temp, &clk->states, head) {
+		nouveau_pstate_del(pstate);
+	}
+
+	nouveau_subdev_destroy(&clk->base);
+}
+
+int
+nouveau_clock_create_(struct nouveau_object *parent,
+		      struct nouveau_object *engine,
+		      struct nouveau_oclass *oclass,
+		      struct nouveau_clocks *clocks,
+		      int length, void **object)
+{
+	struct nouveau_device *device = nv_device(parent);
+	struct nouveau_clock *clk;
+	int ret, idx, arglen;
+	const char *mode;
+
+	ret = nouveau_subdev_create_(parent, engine, oclass, 0, "CLK",
+				     "clock", length, object);
+	clk = *object;
+	if (ret)
+		return ret;
+
+	INIT_LIST_HEAD(&clk->states);
+	clk->domains = clocks;
+	clk->ustate = -1;
+
+	idx = 0;
+	do {
+		ret = nouveau_pstate_new(clk, idx++);
+	} while (ret == 0);
+
+	mode = nouveau_stropt(device->cfgopt, "NvClkMode", &arglen);
+	if (mode) {
+		if (!strncasecmpz(mode, "disabled", arglen)) {
+			clk->ustate = -1;
+		} else {
+			char save = mode[arglen];
+			long v;
+
+			((char *)mode)[arglen] = '\0';
+			if (!kstrtol(mode, 0, &v))
+				nouveau_clock_ustate_update(clk, v);
+			((char *)mode)[arglen] = save;
+		}
+	}
+
+	return 0;
+}
diff --git a/drivers/gpu/drm/nouveau/core/subdev/clock/nv04.c b/drivers/gpu/drm/nouveau/core/subdev/clock/nv04.c
index a14277586595..da50c1b12928 100644
--- a/drivers/gpu/drm/nouveau/core/subdev/clock/nv04.c
+++ b/drivers/gpu/drm/nouveau/core/subdev/clock/nv04.c
@@ -77,7 +77,7 @@ nv04_clock_ctor(struct nouveau_object *parent, struct nouveau_object *engine,
 	struct nv04_clock_priv *priv;
 	int ret;
 
-	ret = nouveau_clock_create(parent, engine, oclass, &priv);
+	ret = nouveau_clock_create(parent, engine, oclass, NULL, &priv);
 	*pobject = nv_object(priv);
 	if (ret)
 		return ret;
diff --git a/drivers/gpu/drm/nouveau/core/subdev/clock/nv40.c b/drivers/gpu/drm/nouveau/core/subdev/clock/nv40.c
index 0db5dbfd91b5..db7346f79080 100644
--- a/drivers/gpu/drm/nouveau/core/subdev/clock/nv40.c
+++ b/drivers/gpu/drm/nouveau/core/subdev/clock/nv40.c
@@ -23,11 +23,188 @@
  */
 
 #include <subdev/clock.h>
+#include <subdev/bios.h>
+#include <subdev/bios/pll.h>
+
+#include "pll.h"
 
 struct nv40_clock_priv {
 	struct nouveau_clock base;
+	u32 ctrl;
+	u32 npll_ctrl;
+	u32 npll_coef;
+	u32 spll;
+};
+
+static struct nouveau_clocks
+nv40_domain[] = {
+	{ nv_clk_src_crystal, 0xff },
+	{ nv_clk_src_href   , 0xff },
+	{ nv_clk_src_core   , 0xff, 0, "core", 1000 },
+	{ nv_clk_src_shader , 0xff, 0, "shader", 1000 },
+	{ nv_clk_src_mem    , 0xff, 0, "memory", 1000 },
+	{ nv_clk_src_max }
 };
 
+static u32
+read_pll_1(struct nv40_clock_priv *priv, u32 reg)
+{
+	u32 ctrl = nv_rd32(priv, reg + 0x00);
+	int P = (ctrl & 0x00070000) >> 16;
+	int N = (ctrl & 0x0000ff00) >> 8;
+	int M = (ctrl & 0x000000ff) >> 0;
+	u32 ref = 27000, clk = 0;
+
+	if (ctrl & 0x80000000)
+		clk = ref * N / M;
+
+	return clk >> P;
+}
+
+static u32
+read_pll_2(struct nv40_clock_priv *priv, u32 reg)
+{
+	u32 ctrl = nv_rd32(priv, reg + 0x00);
+	u32 coef = nv_rd32(priv, reg + 0x04);
+	int N2 = (coef & 0xff000000) >> 24;
+	int M2 = (coef & 0x00ff0000) >> 16;
+	int N1 = (coef & 0x0000ff00) >> 8;
+	int M1 = (coef & 0x000000ff) >> 0;
+	int P = (ctrl & 0x00070000) >> 16;
+	u32 ref = 27000, clk = 0;
+
+	if ((ctrl & 0x80000000) && M1) {
+		clk = ref * N1 / M1;
+		if ((ctrl & 0x40000100) == 0x40000000) {
+			if (M2)
+				clk = clk * N2 / M2;
+			else
+				clk = 0;
+		}
+	}
+
+	return clk >> P;
+}
+
+static u32
+read_clk(struct nv40_clock_priv *priv, u32 src)
+{
+	switch (src) {
+	case 3:
+		return read_pll_2(priv, 0x004000);
+	case 2:
+		return read_pll_1(priv, 0x004008);
+	default:
+		break;
+	}
+
+	return 0;
+}
+
+static int
+nv40_clock_read(struct nouveau_clock *clk, enum nv_clk_src src)
+{
+	struct nv40_clock_priv *priv = (void *)clk;
+	u32 mast = nv_rd32(priv, 0x00c040);
+
+	switch (src) {
+	case nv_clk_src_crystal:
+		return nv_device(priv)->crystal;
+	case nv_clk_src_href:
+		return 100000; /*XXX: PCIE/AGP differ*/
+	case nv_clk_src_core:
+		return read_clk(priv, (mast & 0x00000003) >> 0);
+	case nv_clk_src_shader:
+		return read_clk(priv, (mast & 0x00000030) >> 4);
+	case nv_clk_src_mem:
+		return read_pll_2(priv, 0x4020);
+	default:
+		break;
+	}
+
+	nv_debug(priv, "unknown clock source %d 0x%08x\n", src, mast);
+	return -EINVAL;
+}
+
+static int
+nv40_clock_calc_pll(struct nv40_clock_priv *priv, u32 reg, u32 clk,
+		    int *N1, int *M1, int *N2, int *M2, int *log2P)
+{
+	struct nouveau_bios *bios = nouveau_bios(priv);
+	struct nvbios_pll pll;
+	int ret;
+
+	ret = nvbios_pll_parse(bios, reg, &pll);
+	if (ret)
+		return ret;
+
+	if (clk < pll.vco1.max_freq)
+		pll.vco2.max_freq = 0;
+
+	ret = nv04_pll_calc(nv_subdev(priv), &pll, clk, N1, M1, N2, M2, log2P);
+	if (ret == 0)
+		return -ERANGE;
+	return ret;
+}
+
+static int
+nv40_clock_calc(struct nouveau_clock *clk, struct nouveau_cstate *cstate)
+{
+	struct nv40_clock_priv *priv = (void *)clk;
+	int gclk = cstate->domain[nv_clk_src_core];
+	int sclk = cstate->domain[nv_clk_src_shader];
+	int N1, M1, N2, M2, log2P;
+	int ret;
+
+	/* core/geometric clock */
+	ret = nv40_clock_calc_pll(priv, 0x004000, gclk,
+				 &N1, &M1, &N2, &M2, &log2P);
+	if (ret < 0)
+		return ret;
+
+	if (N2 == M2) {
+		priv->npll_ctrl = 0x80000100 | (log2P << 16);
+		priv->npll_coef = (N1 << 8) | M1;
+	} else {
+		priv->npll_ctrl = 0xc0000000 | (log2P << 16);
+		priv->npll_coef = (N2 << 24) | (M2 << 16) | (N1 << 8) | M1;
+	}
+
+	/* use the second pll for shader/rop clock, if it differs from core */
+	if (sclk && sclk != gclk) {
+		ret = nv40_clock_calc_pll(priv, 0x004008, sclk,
+					 &N1, &M1, NULL, NULL, &log2P);
+		if (ret < 0)
+			return ret;
+
+		priv->spll = 0xc0000000 | (log2P << 16) | (N1 << 8) | M1;
+		priv->ctrl = 0x00000223;
+	} else {
+		priv->spll = 0x00000000;
+		priv->ctrl = 0x00000333;
+	}
+
+	return 0;
+}
+
+static int
+nv40_clock_prog(struct nouveau_clock *clk)
+{
+	struct nv40_clock_priv *priv = (void *)clk;
+	nv_mask(priv, 0x00c040, 0x00000333, 0x00000000);
+	nv_wr32(priv, 0x004004, priv->npll_coef);
+	nv_mask(priv, 0x004000, 0xc0070100, priv->npll_ctrl);
+	nv_mask(priv, 0x004008, 0xc007ffff, priv->spll);
+	mdelay(5);
+	nv_mask(priv, 0x00c040, 0x00000333, priv->ctrl);
+	return 0;
+}
+
+static void
+nv40_clock_tidy(struct nouveau_clock *clk)
+{
+}
+
 static int
 nv40_clock_ctor(struct nouveau_object *parent, struct nouveau_object *engine,
 		struct nouveau_oclass *oclass, void *data, u32 size,
@@ -36,13 +213,17 @@ nv40_clock_ctor(struct nouveau_object *parent, struct nouveau_object *engine,
 	struct nv40_clock_priv *priv;
 	int ret;
 
-	ret = nouveau_clock_create(parent, engine, oclass, &priv);
+	ret = nouveau_clock_create(parent, engine, oclass, nv40_domain, &priv);
 	*pobject = nv_object(priv);
 	if (ret)
 		return ret;
 
 	priv->base.pll_calc = nv04_clock_pll_calc;
 	priv->base.pll_prog = nv04_clock_pll_prog;
+	priv->base.read = nv40_clock_read;
+	priv->base.calc = nv40_clock_calc;
+	priv->base.prog = nv40_clock_prog;
+	priv->base.tidy = nv40_clock_tidy;
 	return 0;
 }
 
diff --git a/drivers/gpu/drm/nouveau/core/subdev/clock/nv50.c b/drivers/gpu/drm/nouveau/core/subdev/clock/nv50.c
index d09d3e78040c..250a6d96016b 100644
--- a/drivers/gpu/drm/nouveau/core/subdev/clock/nv50.c
+++ b/drivers/gpu/drm/nouveau/core/subdev/clock/nv50.c
@@ -22,40 +22,538 @@
  * Authors: Ben Skeggs
  */
 
-#include <subdev/clock.h>
 #include <subdev/bios.h>
 #include <subdev/bios/pll.h>
 
+#include "nv50.h"
 #include "pll.h"
+#include "seq.h"
 
-struct nv50_clock_priv {
-	struct nouveau_clock base;
-};
+static u32
+read_div(struct nv50_clock_priv *priv)
+{
+	switch (nv_device(priv)->chipset) {
+	case 0x50: /* it exists, but only has bit 31, not the dividers.. */
+	case 0x84:
+	case 0x86:
+	case 0x98:
+	case 0xa0:
+		return nv_rd32(priv, 0x004700);
+	case 0x92:
+	case 0x94:
+	case 0x96:
+		return nv_rd32(priv, 0x004800);
+	default:
+		return 0x00000000;
+	}
+}
+
+static u32
+read_pll_src(struct nv50_clock_priv *priv, u32 base)
+{
+	struct nouveau_clock *clk = &priv->base;
+	u32 coef, ref = clk->read(clk, nv_clk_src_crystal);
+	u32 rsel = nv_rd32(priv, 0x00e18c);
+	int P, N, M, id;
+
+	switch (nv_device(priv)->chipset) {
+	case 0x50:
+	case 0xa0:
+		switch (base) {
+		case 0x4020:
+		case 0x4028: id = !!(rsel & 0x00000004); break;
+		case 0x4008: id = !!(rsel & 0x00000008); break;
+		case 0x4030: id = 0; break;
+		default:
+			nv_error(priv, "ref: bad pll 0x%06x\n", base);
+			return 0;
+		}
+
+		coef = nv_rd32(priv, 0x00e81c + (id * 0x0c));
+		ref *=  (coef & 0x01000000) ? 2 : 4;
+		P    =  (coef & 0x00070000) >> 16;
+		N    = ((coef & 0x0000ff00) >> 8) + 1;
+		M    = ((coef & 0x000000ff) >> 0) + 1;
+		break;
+	case 0x84:
+	case 0x86:
+	case 0x92:
+		coef = nv_rd32(priv, 0x00e81c);
+		P    = (coef & 0x00070000) >> 16;
+		N    = (coef & 0x0000ff00) >> 8;
+		M    = (coef & 0x000000ff) >> 0;
+		break;
+	case 0x94:
+	case 0x96:
+	case 0x98:
+		rsel = nv_rd32(priv, 0x00c050);
+		switch (base) {
+		case 0x4020: rsel = (rsel & 0x00000003) >> 0; break;
+		case 0x4008: rsel = (rsel & 0x0000000c) >> 2; break;
+		case 0x4028: rsel = (rsel & 0x00001800) >> 11; break;
+		case 0x4030: rsel = 3; break;
+		default:
+			nv_error(priv, "ref: bad pll 0x%06x\n", base);
+			return 0;
+		}
+
+		switch (rsel) {
+		case 0: id = 1; break;
+		case 1: return clk->read(clk, nv_clk_src_crystal);
+		case 2: return clk->read(clk, nv_clk_src_href);
+		case 3: id = 0; break;
+		}
+
+		coef =  nv_rd32(priv, 0x00e81c + (id * 0x28));
+		P    = (nv_rd32(priv, 0x00e824 + (id * 0x28)) >> 16) & 7;
+		P   += (coef & 0x00070000) >> 16;
+		N    = (coef & 0x0000ff00) >> 8;
+		M    = (coef & 0x000000ff) >> 0;
+		break;
+	default:
+		BUG_ON(1);
+	}
+
+	if (M)
+		return (ref * N / M) >> P;
+	return 0;
+}
+
+static u32
+read_pll_ref(struct nv50_clock_priv *priv, u32 base)
+{
+	struct nouveau_clock *clk = &priv->base;
+	u32 src, mast = nv_rd32(priv, 0x00c040);
+
+	switch (base) {
+	case 0x004028:
+		src = !!(mast & 0x00200000);
+		break;
+	case 0x004020:
+		src = !!(mast & 0x00400000);
+		break;
+	case 0x004008:
+		src = !!(mast & 0x00010000);
+		break;
+	case 0x004030:
+		src = !!(mast & 0x02000000);
+		break;
+	case 0x00e810:
+		return clk->read(clk, nv_clk_src_crystal);
+	default:
+		nv_error(priv, "bad pll 0x%06x\n", base);
+		return 0;
+	}
+
+	if (src)
+		return clk->read(clk, nv_clk_src_href);
+	return read_pll_src(priv, base);
+}
+
+static u32
+read_pll(struct nv50_clock_priv *priv, u32 base)
+{
+	struct nouveau_clock *clk = &priv->base;
+	u32 mast = nv_rd32(priv, 0x00c040);
+	u32 ctrl = nv_rd32(priv, base + 0);
+	u32 coef = nv_rd32(priv, base + 4);
+	u32 ref = read_pll_ref(priv, base);
+	u32 freq = 0;
+	int N1, N2, M1, M2;
+
+	if (base == 0x004028 && (mast & 0x00100000)) {
+		/* wtf, appears to only disable post-divider on nva0 */
+		if (nv_device(priv)->chipset != 0xa0)
+			return clk->read(clk, nv_clk_src_dom6);
+	}
+
+	N2 = (coef & 0xff000000) >> 24;
+	M2 = (coef & 0x00ff0000) >> 16;
+	N1 = (coef & 0x0000ff00) >> 8;
+	M1 = (coef & 0x000000ff);
+	if ((ctrl & 0x80000000) && M1) {
+		freq = ref * N1 / M1;
+		if ((ctrl & 0x40000100) == 0x40000000) {
+			if (M2)
+				freq = freq * N2 / M2;
+			else
+				freq = 0;
+		}
+	}
+
+	return freq;
+}
 
 static int
+nv50_clock_read(struct nouveau_clock *clk, enum nv_clk_src src)
+{
+	struct nv50_clock_priv *priv = (void *)clk;
+	u32 mast = nv_rd32(priv, 0x00c040);
+	u32 P = 0;
+
+	switch (src) {
+	case nv_clk_src_crystal:
+		return nv_device(priv)->crystal;
+	case nv_clk_src_href:
+		return 100000; /* PCIE reference clock */
+	case nv_clk_src_hclk:
+		return div_u64((u64)clk->read(clk, nv_clk_src_href) * 27778, 10000);
+	case nv_clk_src_hclkm3:
+		return clk->read(clk, nv_clk_src_hclk) * 3;
+	case nv_clk_src_hclkm3d2:
+		return clk->read(clk, nv_clk_src_hclk) * 3 / 2;
+	case nv_clk_src_host:
+		switch (mast & 0x30000000) {
+		case 0x00000000: return clk->read(clk, nv_clk_src_href);
+		case 0x10000000: break;
+		case 0x20000000: /* !0x50 */
+		case 0x30000000: return clk->read(clk, nv_clk_src_hclk);
+		}
+		break;
+	case nv_clk_src_core:
+		if (!(mast & 0x00100000))
+			P = (nv_rd32(priv, 0x004028) & 0x00070000) >> 16;
+		switch (mast & 0x00000003) {
+		case 0x00000000: return clk->read(clk, nv_clk_src_crystal) >> P;
+		case 0x00000001: return clk->read(clk, nv_clk_src_dom6);
+		case 0x00000002: return read_pll(priv, 0x004020) >> P;
+		case 0x00000003: return read_pll(priv, 0x004028) >> P;
+		}
+		break;
+	case nv_clk_src_shader:
+		P = (nv_rd32(priv, 0x004020) & 0x00070000) >> 16;
+		switch (mast & 0x00000030) {
+		case 0x00000000:
+			if (mast & 0x00000080)
+				return clk->read(clk, nv_clk_src_host) >> P;
+			return clk->read(clk, nv_clk_src_crystal) >> P;
+		case 0x00000010: break;
+		case 0x00000020: return read_pll(priv, 0x004028) >> P;
+		case 0x00000030: return read_pll(priv, 0x004020) >> P;
+		}
+		break;
+	case nv_clk_src_mem:
+		P = (nv_rd32(priv, 0x004008) & 0x00070000) >> 16;
+		if (nv_rd32(priv, 0x004008) & 0x00000200) {
+			switch (mast & 0x0000c000) {
+			case 0x00000000:
+				return clk->read(clk, nv_clk_src_crystal) >> P;
+			case 0x00008000:
+			case 0x0000c000:
+				return clk->read(clk, nv_clk_src_href) >> P;
+			}
+		} else {
+			return read_pll(priv, 0x004008) >> P;
+		}
+		break;
+	case nv_clk_src_vdec:
+		P = (read_div(priv) & 0x00000700) >> 8;
+		switch (nv_device(priv)->chipset) {
+		case 0x84:
+		case 0x86:
+		case 0x92:
+		case 0x94:
+		case 0x96:
+		case 0xa0:
+			switch (mast & 0x00000c00) {
+			case 0x00000000:
+				if (nv_device(priv)->chipset == 0xa0) /* wtf?? */
+					return clk->read(clk, nv_clk_src_core) >> P;
+				return clk->read(clk, nv_clk_src_crystal) >> P;
+			case 0x00000400:
+				return 0;
+			case 0x00000800:
+				if (mast & 0x01000000)
+					return read_pll(priv, 0x004028) >> P;
+				return read_pll(priv, 0x004030) >> P;
+			case 0x00000c00:
+				return clk->read(clk, nv_clk_src_core) >> P;
+			}
+			break;
+		case 0x98:
+			switch (mast & 0x00000c00) {
+			case 0x00000000:
+				return clk->read(clk, nv_clk_src_core) >> P;
+			case 0x00000400:
+				return 0;
+			case 0x00000800:
+				return clk->read(clk, nv_clk_src_hclkm3d2) >> P;
+			case 0x00000c00:
+				return clk->read(clk, nv_clk_src_mem) >> P;
+			}
+			break;
+		}
+		break;
+	case nv_clk_src_dom6:
+		switch (nv_device(priv)->chipset) {
+		case 0x50:
+		case 0xa0:
+			return read_pll(priv, 0x00e810) >> 2;
+		case 0x84:
+		case 0x86:
+		case 0x92:
+		case 0x94:
+		case 0x96:
+		case 0x98:
+			P = (read_div(priv) & 0x00000007) >> 0;
+			switch (mast & 0x0c000000) {
+			case 0x00000000: return clk->read(clk, nv_clk_src_href);
+			case 0x04000000: break;
+			case 0x08000000: return clk->read(clk, nv_clk_src_hclk);
+			case 0x0c000000:
+				return clk->read(clk, nv_clk_src_hclkm3) >> P;
+			}
+			break;
+		default:
+			break;
+		}
+	default:
+		break;
+	}
+
+	nv_debug(priv, "unknown clock source %d 0x%08x\n", src, mast);
+	return -EINVAL;
+}
+
+static u32
+calc_pll(struct nv50_clock_priv *priv, u32 reg, u32 clk, int *N, int *M, int *P)
+{
+	struct nouveau_bios *bios = nouveau_bios(priv);
+	struct nvbios_pll pll;
+	int ret;
+
+	ret = nvbios_pll_parse(bios, reg, &pll);
+	if (ret)
+		return 0;
+
+	pll.vco2.max_freq = 0;
+	pll.refclk = read_pll_ref(priv, reg);
+	if (!pll.refclk)
+		return 0;
+
+	return nv04_pll_calc(nv_subdev(priv), &pll, clk, N, M, NULL, NULL, P);
+}
+
+static inline u32
+calc_div(u32 src, u32 target, int *div)
+{
+	u32 clk0 = src, clk1 = src;
+	for (*div = 0; *div <= 7; (*div)++) {
+		if (clk0 <= target) {
+			clk1 = clk0 << (*div ? 1 : 0);
+			break;
+		}
+		clk0 >>= 1;
+	}
+
+	if (target - clk0 <= clk1 - target)
+		return clk0;
+	(*div)--;
+	return clk1;
+}
+
+static inline u32
+clk_same(u32 a, u32 b)
+{
+	return ((a / 1000) == (b / 1000));
+}
+
+static int
+nv50_clock_calc(struct nouveau_clock *clk, struct nouveau_cstate *cstate)
+{
+	struct nv50_clock_priv *priv = (void *)clk;
+	struct nv50_clock_hwsq *hwsq = &priv->hwsq;
+	const int shader = cstate->domain[nv_clk_src_shader];
+	const int core = cstate->domain[nv_clk_src_core];
+	const int vdec = cstate->domain[nv_clk_src_vdec];
+	const int dom6 = cstate->domain[nv_clk_src_dom6];
+	u32 mastm = 0, mastv = 0;
+	u32 divsm = 0, divsv = 0;
+	int N, M, P1, P2;
+	int freq, out;
+
+	/* prepare a hwsq script from which we'll perform the reclock */
+	out = clk_init(hwsq, nv_subdev(clk));
+	if (out)
+		return out;
+
+	clk_wr32(hwsq, fifo, 0x00000001); /* block fifo */
+	clk_nsec(hwsq, 8000);
+	clk_setf(hwsq, 0x10, 0x00); /* disable fb */
+	clk_wait(hwsq, 0x00, 0x01); /* wait for fb disabled */
+
+	/* vdec: avoid modifying xpll until we know exactly how the other
+	 * clock domains work, i suspect at least some of them can also be
+	 * tied to xpll...
+	 */
+	if (vdec) {
+		/* see how close we can get using nvclk as a source */
+		freq = calc_div(core, vdec, &P1);
+
+		/* see how close we can get using xpll/hclk as a source */
+		if (nv_device(priv)->chipset != 0x98)
+			out = read_pll(priv, 0x004030);
+		else
+			out = clk->read(clk, nv_clk_src_hclkm3d2);
+		out = calc_div(out, vdec, &P2);
+
+		/* select whichever gets us closest */
+		if (abs(vdec - freq) <= abs(vdec - out)) {
+			if (nv_device(priv)->chipset != 0x98)
+				mastv |= 0x00000c00;
+			divsv |= P1 << 8;
+		} else {
+			mastv |= 0x00000800;
+			divsv |= P2 << 8;
+		}
+
+		mastm |= 0x00000c00;
+		divsm |= 0x00000700;
+	}
+
+	/* dom6: nfi what this is, but we're limited to various combinations
+	 * of the host clock frequency
+	 */
+	if (dom6) {
+		if (clk_same(dom6, clk->read(clk, nv_clk_src_href))) {
+			mastv |= 0x00000000;
+		} else
+		if (clk_same(dom6, clk->read(clk, nv_clk_src_hclk))) {
+			mastv |= 0x08000000;
+		} else {
+			freq = clk->read(clk, nv_clk_src_hclk) * 3;
+			freq = calc_div(freq, dom6, &P1);
+
+			mastv |= 0x0c000000;
+			divsv |= P1;
+		}
+
+		mastm |= 0x0c000000;
+		divsm |= 0x00000007;
+	}
+
+	/* vdec/dom6: switch to "safe" clocks temporarily, update dividers
+	 * and then switch to target clocks
+	 */
+	clk_mask(hwsq, mast, mastm, 0x00000000);
+	clk_mask(hwsq, divs, divsm, divsv);
+	clk_mask(hwsq, mast, mastm, mastv);
+
+	/* core/shader: disconnect nvclk/sclk from their PLLs (nvclk to dom6,
+	 * sclk to hclk) before reprogramming
+	 */
+	if (nv_device(priv)->chipset < 0x92)
+		clk_mask(hwsq, mast, 0x001000b0, 0x00100080);
+	else
+		clk_mask(hwsq, mast, 0x000000b3, 0x00000081);
+
+	/* core: for the moment at least, always use nvpll */
+	freq = calc_pll(priv, 0x4028, core, &N, &M, &P1);
+	if (freq == 0)
+		return -ERANGE;
+
+	clk_mask(hwsq, nvpll[0], 0xc03f0100,
+				 0x80000000 | (P1 << 19) | (P1 << 16));
+	clk_mask(hwsq, nvpll[1], 0x0000ffff, (N << 8) | M);
+
+	/* shader: tie to nvclk if possible, otherwise use spll.  have to be
+	 * very careful that the shader clock is at least twice the core, or
+	 * some chipsets will be very unhappy.  i expect most or all of these
+	 * cases will be handled by tying to nvclk, but it's possible there's
+	 * corners
+	 */
+	if (P1-- && shader == (core << 1)) {
+		clk_mask(hwsq, spll[0], 0xc03f0100, (P1 << 19) | (P1 << 16));
+		clk_mask(hwsq, mast, 0x00100033, 0x00000023);
+	} else {
+		freq = calc_pll(priv, 0x4020, shader, &N, &M, &P1);
+		if (freq == 0)
+			return -ERANGE;
+
+		clk_mask(hwsq, spll[0], 0xc03f0100,
+					0x80000000 | (P1 << 19) | (P1 << 16));
+		clk_mask(hwsq, spll[1], 0x0000ffff, (N << 8) | M);
+		clk_mask(hwsq, mast, 0x00100033, 0x00000033);
+	}
+
+	/* restore normal operation */
+	clk_setf(hwsq, 0x10, 0x01); /* enable fb */
+	clk_wait(hwsq, 0x00, 0x00); /* wait for fb enabled */
+	clk_wr32(hwsq, fifo, 0x00000000); /* un-block fifo */
+	return 0;
+}
+
+static int
+nv50_clock_prog(struct nouveau_clock *clk)
+{
+	struct nv50_clock_priv *priv = (void *)clk;
+	return clk_exec(&priv->hwsq, true);
+}
+
+static void
+nv50_clock_tidy(struct nouveau_clock *clk)
+{
+	struct nv50_clock_priv *priv = (void *)clk;
+	clk_exec(&priv->hwsq, false);
+}
+
+int
 nv50_clock_ctor(struct nouveau_object *parent, struct nouveau_object *engine,
 		struct nouveau_oclass *oclass, void *data, u32 size,
 		struct nouveau_object **pobject)
 {
+	struct nv50_clock_oclass *pclass = (void *)oclass;
 	struct nv50_clock_priv *priv;
 	int ret;
 
-	ret = nouveau_clock_create(parent, engine, oclass, &priv);
+	ret = nouveau_clock_create(parent, engine, oclass, pclass->domains,
+				  &priv);
 	*pobject = nv_object(priv);
 	if (ret)
 		return ret;
 
-	priv->base.pll_calc = nv04_clock_pll_calc;
+	priv->hwsq.r_fifo = hwsq_reg(0x002504);
+	priv->hwsq.r_spll[0] = hwsq_reg(0x004020);
+	priv->hwsq.r_spll[1] = hwsq_reg(0x004024);
+	priv->hwsq.r_nvpll[0] = hwsq_reg(0x004028);
+	priv->hwsq.r_nvpll[1] = hwsq_reg(0x00402c);
+	switch (nv_device(priv)->chipset) {
+	case 0x92:
+	case 0x94:
+	case 0x96:
+		priv->hwsq.r_divs = hwsq_reg(0x004800);
+		break;
+	default:
+		priv->hwsq.r_divs = hwsq_reg(0x004700);
+		break;
+	}
+	priv->hwsq.r_mast = hwsq_reg(0x00c040);
+
+	priv->base.read = nv50_clock_read;
+	priv->base.calc = nv50_clock_calc;
+	priv->base.prog = nv50_clock_prog;
+	priv->base.tidy = nv50_clock_tidy;
 	return 0;
 }
 
-struct nouveau_oclass
-nv50_clock_oclass = {
-	.handle = NV_SUBDEV(CLOCK, 0x50),
-	.ofuncs = &(struct nouveau_ofuncs) {
+static struct nouveau_clocks
+nv50_domains[] = {
+	{ nv_clk_src_crystal, 0xff },
+	{ nv_clk_src_href   , 0xff },
+	{ nv_clk_src_core   , 0xff, 0, "core", 1000 },
+	{ nv_clk_src_shader , 0xff, 0, "shader", 1000 },
+	{ nv_clk_src_mem    , 0xff, 0, "memory", 1000 },
+	{ nv_clk_src_max }
+};
+
+struct nouveau_oclass *
+nv50_clock_oclass = &(struct nv50_clock_oclass) {
+	.base.handle = NV_SUBDEV(CLOCK, 0x50),
+	.base.ofuncs = &(struct nouveau_ofuncs) {
 		.ctor = nv50_clock_ctor,
 		.dtor = _nouveau_clock_dtor,
 		.init = _nouveau_clock_init,
 		.fini = _nouveau_clock_fini,
 	},
-};
+	.domains = nv50_domains,
+}.base;
diff --git a/drivers/gpu/drm/nouveau/core/subdev/clock/nv50.h b/drivers/gpu/drm/nouveau/core/subdev/clock/nv50.h
new file mode 100644
index 000000000000..f10917d789e8
--- /dev/null
+++ b/drivers/gpu/drm/nouveau/core/subdev/clock/nv50.h
@@ -0,0 +1,31 @@
+#ifndef __NVKM_CLK_NV50_H__
+#define __NVKM_CLK_NV50_H__
+
+#include <subdev/bus.h>
+#include <subdev/bus/hwsq.h>
+#include <subdev/clock.h>
+
+struct nv50_clock_hwsq {
+	struct hwsq base;
+	struct hwsq_reg r_fifo;
+	struct hwsq_reg r_spll[2];
+	struct hwsq_reg r_nvpll[2];
+	struct hwsq_reg r_divs;
+	struct hwsq_reg r_mast;
+};
+
+struct nv50_clock_priv {
+	struct nouveau_clock base;
+	struct nv50_clock_hwsq hwsq;
+};
+
+int  nv50_clock_ctor(struct nouveau_object *, struct nouveau_object *,
+		     struct nouveau_oclass *, void *, u32,
+		     struct nouveau_object **);
+
+struct nv50_clock_oclass {
+	struct nouveau_oclass base;
+	struct nouveau_clocks *domains;
+};
+
+#endif
diff --git a/drivers/gpu/drm/nouveau/core/subdev/clock/nv84.c b/drivers/gpu/drm/nouveau/core/subdev/clock/nv84.c
new file mode 100644
index 000000000000..b0b7c1437f10
--- /dev/null
+++ b/drivers/gpu/drm/nouveau/core/subdev/clock/nv84.c
@@ -0,0 +1,48 @@
+/*
+ * Copyright 2013 Red Hat Inc.
+ *
+ * Permission is hereby granted, free of charge, to any person obtaining a
+ * copy of this software and associated documentation files (the "Software"),
+ * to deal in the Software without restriction, including without limitation
+ * the rights to use, copy, modify, merge, publish, distribute, sublicense,
+ * and/or sell copies of the Software, and to permit persons to whom the
+ * Software is furnished to do so, subject to the following conditions:
+ *
+ * The above copyright notice and this permission notice shall be included in
+ * all copies or substantial portions of the Software.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+ * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+ * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT.  IN NO EVENT SHALL
+ * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR
+ * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE,
+ * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR
+ * OTHER DEALINGS IN THE SOFTWARE.
+ *
+ * Authors: Ben Skeggs <bskeggs@redhat.com>
+ */
+
+#include "nv50.h"
+
+static struct nouveau_clocks
+nv84_domains[] = {
+	{ nv_clk_src_crystal, 0xff },
+	{ nv_clk_src_href   , 0xff },
+	{ nv_clk_src_core   , 0xff, 0, "core", 1000 },
+	{ nv_clk_src_shader , 0xff, 0, "shader", 1000 },
+	{ nv_clk_src_mem    , 0xff, 0, "memory", 1000 },
+	{ nv_clk_src_vdec   , 0xff },
+	{ nv_clk_src_max }
+};
+
+struct nouveau_oclass *
+nv84_clock_oclass = &(struct nv50_clock_oclass) {
+	.base.handle = NV_SUBDEV(CLOCK, 0x84),
+	.base.ofuncs = &(struct nouveau_ofuncs) {
+		.ctor = nv50_clock_ctor,
+		.dtor = _nouveau_clock_dtor,
+		.init = _nouveau_clock_init,
+		.fini = _nouveau_clock_fini,
+	},
+	.domains = nv84_domains,
+}.base;
diff --git a/drivers/gpu/drm/nouveau/core/subdev/clock/nva3.c b/drivers/gpu/drm/nouveau/core/subdev/clock/nva3.c
index f074cd20bc9c..4f5a1373f002 100644
--- a/drivers/gpu/drm/nouveau/core/subdev/clock/nva3.c
+++ b/drivers/gpu/drm/nouveau/core/subdev/clock/nva3.c
@@ -22,33 +22,277 @@
  * Authors: Ben Skeggs
  */
 
-#include <subdev/clock.h>
 #include <subdev/bios.h>
 #include <subdev/bios/pll.h>
+#include <subdev/timer.h>
 
 #include "pll.h"
 
+#include "nva3.h"
+
 struct nva3_clock_priv {
 	struct nouveau_clock base;
+	struct nva3_clock_info eng[nv_clk_src_max];
 };
 
+static u32 read_clk(struct nva3_clock_priv *, int, bool);
+static u32 read_pll(struct nva3_clock_priv *, int, u32);
+
+static u32
+read_vco(struct nva3_clock_priv *priv, int clk)
+{
+	u32 sctl = nv_rd32(priv, 0x4120 + (clk * 4));
+	if ((sctl & 0x00000030) != 0x00000030)
+		return read_pll(priv, 0x41, 0x00e820);
+	return read_pll(priv, 0x42, 0x00e8a0);
+}
+
+static u32
+read_clk(struct nva3_clock_priv *priv, int clk, bool ignore_en)
+{
+	u32 sctl, sdiv, sclk;
+
+	/* refclk for the 0xe8xx plls is a fixed frequency */
+	if (clk >= 0x40) {
+		if (nv_device(priv)->chipset == 0xaf) {
+			/* no joke.. seriously.. sigh.. */
+			return nv_rd32(priv, 0x00471c) * 1000;
+		}
+
+		return nv_device(priv)->crystal;
+	}
+
+	sctl = nv_rd32(priv, 0x4120 + (clk * 4));
+	if (!ignore_en && !(sctl & 0x00000100))
+		return 0;
+
+	switch (sctl & 0x00003000) {
+	case 0x00000000:
+		return nv_device(priv)->crystal;
+	case 0x00002000:
+		if (sctl & 0x00000040)
+			return 108000;
+		return 100000;
+	case 0x00003000:
+		sclk = read_vco(priv, clk);
+		sdiv = ((sctl & 0x003f0000) >> 16) + 2;
+		return (sclk * 2) / sdiv;
+	default:
+		return 0;
+	}
+}
+
+static u32
+read_pll(struct nva3_clock_priv *priv, int clk, u32 pll)
+{
+	u32 ctrl = nv_rd32(priv, pll + 0);
+	u32 sclk = 0, P = 1, N = 1, M = 1;
+
+	if (!(ctrl & 0x00000008)) {
+		if (ctrl & 0x00000001) {
+			u32 coef = nv_rd32(priv, pll + 4);
+			M = (coef & 0x000000ff) >> 0;
+			N = (coef & 0x0000ff00) >> 8;
+			P = (coef & 0x003f0000) >> 16;
+
+			/* no post-divider on these.. */
+			if ((pll & 0x00ff00) == 0x00e800)
+				P = 1;
+
+			sclk = read_clk(priv, 0x00 + clk, false);
+		}
+	} else {
+		sclk = read_clk(priv, 0x10 + clk, false);
+	}
+
+	if (M * P)
+		return sclk * N / (M * P);
+	return 0;
+}
+
+static int
+nva3_clock_read(struct nouveau_clock *clk, enum nv_clk_src src)
+{
+	struct nva3_clock_priv *priv = (void *)clk;
+
+	switch (src) {
+	case nv_clk_src_crystal:
+		return nv_device(priv)->crystal;
+	case nv_clk_src_href:
+		return 100000;
+	case nv_clk_src_core:
+		return read_pll(priv, 0x00, 0x4200);
+	case nv_clk_src_shader:
+		return read_pll(priv, 0x01, 0x4220);
+	case nv_clk_src_mem:
+		return read_pll(priv, 0x02, 0x4000);
+	case nv_clk_src_disp:
+		return read_clk(priv, 0x20, false);
+	case nv_clk_src_vdec:
+		return read_clk(priv, 0x21, false);
+	case nv_clk_src_daemon:
+		return read_clk(priv, 0x25, false);
+	default:
+		nv_error(clk, "invalid clock source %d\n", src);
+		return -EINVAL;
+	}
+}
+
 int
-nva3_clock_pll_calc(struct nouveau_clock *clock, struct nvbios_pll *info,
-		    int clk, struct nouveau_pll_vals *pv)
+nva3_clock_info(struct nouveau_clock *clock, int clk, u32 pll, u32 khz,
+		struct nva3_clock_info *info)
 {
-	int ret, N, M, P;
+	struct nouveau_bios *bios = nouveau_bios(clock);
+	struct nva3_clock_priv *priv = (void *)clock;
+	struct nvbios_pll limits;
+	u32 oclk, sclk, sdiv;
+	int P, N, M, diff;
+	int ret;
+
+	info->pll = 0;
+	info->clk = 0;
+
+	switch (khz) {
+	case 27000:
+		info->clk = 0x00000100;
+		return khz;
+	case 100000:
+		info->clk = 0x00002100;
+		return khz;
+	case 108000:
+		info->clk = 0x00002140;
+		return khz;
+	default:
+		sclk = read_vco(priv, clk);
+		sdiv = min((sclk * 2) / (khz - 2999), (u32)65);
+		/* if the clock has a PLL attached, and we can get a within
+		 * [-2, 3) MHz of a divider, we'll disable the PLL and use
+		 * the divider instead.
+		 *
+		 * divider can go as low as 2, limited here because NVIDIA
+		 * and the VBIOS on my NVA8 seem to prefer using the PLL
+		 * for 810MHz - is there a good reason?
+		 */
+		if (sdiv > 4) {
+			oclk = (sclk * 2) / sdiv;
+			diff = khz - oclk;
+			if (!pll || (diff >= -2000 && diff < 3000)) {
+				info->clk = (((sdiv - 2) << 16) | 0x00003100);
+				return oclk;
+			}
+		}
+
+		if (!pll)
+			return -ERANGE;
+		break;
+	}
 
-	ret = nva3_pll_calc(nv_subdev(clock), info, clk, &N, NULL, &M, &P);
+	ret = nvbios_pll_parse(bios, pll, &limits);
+	if (ret)
+		return ret;
+
+	limits.refclk = read_clk(priv, clk - 0x10, true);
+	if (!limits.refclk)
+		return -EINVAL;
 
-	if (ret > 0) {
-		pv->refclk = info->refclk;
-		pv->N1 = N;
-		pv->M1 = M;
-		pv->log2P = P;
+	ret = nva3_pll_calc(nv_subdev(priv), &limits, khz, &N, NULL, &M, &P);
+	if (ret >= 0) {
+		info->clk = nv_rd32(priv, 0x4120 + (clk * 4));
+		info->pll = (P << 16) | (N << 8) | M;
 	}
+
+	return ret ? ret : -ERANGE;
+}
+
+static int
+calc_clk(struct nva3_clock_priv *priv, struct nouveau_cstate *cstate,
+	 int clk, u32 pll, int idx)
+{
+	int ret = nva3_clock_info(&priv->base, clk, pll, cstate->domain[idx],
+				  &priv->eng[idx]);
+	if (ret >= 0)
+		return 0;
 	return ret;
 }
 
+static void
+prog_pll(struct nva3_clock_priv *priv, int clk, u32 pll, int idx)
+{
+	struct nva3_clock_info *info = &priv->eng[idx];
+	const u32 src0 = 0x004120 + (clk * 4);
+	const u32 src1 = 0x004160 + (clk * 4);
+	const u32 ctrl = pll + 0;
+	const u32 coef = pll + 4;
+
+	if (info->pll) {
+		nv_mask(priv, src0, 0x00000101, 0x00000101);
+		nv_wr32(priv, coef, info->pll);
+		nv_mask(priv, ctrl, 0x00000015, 0x00000015);
+		nv_mask(priv, ctrl, 0x00000010, 0x00000000);
+		nv_wait(priv, ctrl, 0x00020000, 0x00020000);
+		nv_mask(priv, ctrl, 0x00000010, 0x00000010);
+		nv_mask(priv, ctrl, 0x00000008, 0x00000000);
+		nv_mask(priv, src1, 0x00000100, 0x00000000);
+		nv_mask(priv, src1, 0x00000001, 0x00000000);
+	} else {
+		nv_mask(priv, src1, 0x003f3141, 0x00000101 | info->clk);
+		nv_mask(priv, ctrl, 0x00000018, 0x00000018);
+		udelay(20);
+		nv_mask(priv, ctrl, 0x00000001, 0x00000000);
+		nv_mask(priv, src0, 0x00000100, 0x00000000);
+		nv_mask(priv, src0, 0x00000001, 0x00000000);
+	}
+}
+
+static void
+prog_clk(struct nva3_clock_priv *priv, int clk, int idx)
+{
+	struct nva3_clock_info *info = &priv->eng[idx];
+	nv_mask(priv, 0x004120 + (clk * 4), 0x003f3141, 0x00000101 | info->clk);
+}
+
+static int
+nva3_clock_calc(struct nouveau_clock *clk, struct nouveau_cstate *cstate)
+{
+	struct nva3_clock_priv *priv = (void *)clk;
+	int ret;
+
+	if ((ret = calc_clk(priv, cstate, 0x10, 0x4200, nv_clk_src_core)) ||
+	    (ret = calc_clk(priv, cstate, 0x11, 0x4220, nv_clk_src_shader)) ||
+	    (ret = calc_clk(priv, cstate, 0x20, 0x0000, nv_clk_src_disp)) ||
+	    (ret = calc_clk(priv, cstate, 0x21, 0x0000, nv_clk_src_vdec)))
+		return ret;
+
+	return 0;
+}
+
+static int
+nva3_clock_prog(struct nouveau_clock *clk)
+{
+	struct nva3_clock_priv *priv = (void *)clk;
+	prog_pll(priv, 0x00, 0x004200, nv_clk_src_core);
+	prog_pll(priv, 0x01, 0x004220, nv_clk_src_shader);
+	prog_clk(priv, 0x20, nv_clk_src_disp);
+	prog_clk(priv, 0x21, nv_clk_src_vdec);
+	return 0;
+}
+
+static void
+nva3_clock_tidy(struct nouveau_clock *clk)
+{
+}
+
+static struct nouveau_clocks
+nva3_domain[] = {
+	{ nv_clk_src_crystal, 0xff },
+	{ nv_clk_src_href   , 0xff },
+	{ nv_clk_src_core   , 0x00, 0, "core", 1000 },
+	{ nv_clk_src_shader , 0x01, 0, "shader", 1000 },
+	{ nv_clk_src_mem    , 0x02, 0, "memory", 1000 },
+	{ nv_clk_src_vdec   , 0x03 },
+	{ nv_clk_src_disp   , 0x04 },
+	{ nv_clk_src_max }
+};
 
 static int
 nva3_clock_ctor(struct nouveau_object *parent, struct nouveau_object *engine,
@@ -58,12 +302,15 @@ nva3_clock_ctor(struct nouveau_object *parent, struct nouveau_object *engine,
 	struct nva3_clock_priv *priv;
 	int ret;
 
-	ret = nouveau_clock_create(parent, engine, oclass, &priv);
+	ret = nouveau_clock_create(parent, engine, oclass, nva3_domain, &priv);
 	*pobject = nv_object(priv);
 	if (ret)
 		return ret;
 
-	priv->base.pll_calc = nva3_clock_pll_calc;
+	priv->base.read = nva3_clock_read;
+	priv->base.calc = nva3_clock_calc;
+	priv->base.prog = nva3_clock_prog;
+	priv->base.tidy = nva3_clock_tidy;
 	return 0;
 }
 
diff --git a/drivers/gpu/drm/nouveau/core/subdev/clock/nva3.h b/drivers/gpu/drm/nouveau/core/subdev/clock/nva3.h
new file mode 100644
index 000000000000..6229a509b42e
--- /dev/null
+++ b/drivers/gpu/drm/nouveau/core/subdev/clock/nva3.h
@@ -0,0 +1,14 @@
+#ifndef __NVKM_CLK_NVA3_H__
+#define __NVKM_CLK_NVA3_H__
+
+#include <subdev/clock.h>
+
+struct nva3_clock_info {
+	u32 clk;
+	u32 pll;
+};
+
+int nva3_clock_info(struct nouveau_clock *, int, u32, u32,
+		    struct nva3_clock_info *);
+
+#endif
diff --git a/drivers/gpu/drm/nouveau/core/subdev/clock/nvc0.c b/drivers/gpu/drm/nouveau/core/subdev/clock/nvc0.c
index 439d81c26130..c3105720ed24 100644
--- a/drivers/gpu/drm/nouveau/core/subdev/clock/nvc0.c
+++ b/drivers/gpu/drm/nouveau/core/subdev/clock/nvc0.c
@@ -25,11 +25,408 @@
 #include <subdev/clock.h>
 #include <subdev/bios.h>
 #include <subdev/bios/pll.h>
+#include <subdev/timer.h>
 
 #include "pll.h"
 
+struct nvc0_clock_info {
+	u32 freq;
+	u32 ssel;
+	u32 mdiv;
+	u32 dsrc;
+	u32 ddiv;
+	u32 coef;
+};
+
 struct nvc0_clock_priv {
 	struct nouveau_clock base;
+	struct nvc0_clock_info eng[16];
+};
+
+static u32 read_div(struct nvc0_clock_priv *, int, u32, u32);
+
+static u32
+read_vco(struct nvc0_clock_priv *priv, u32 dsrc)
+{
+	struct nouveau_clock *clk = &priv->base;
+	u32 ssrc = nv_rd32(priv, dsrc);
+	if (!(ssrc & 0x00000100))
+		return clk->read(clk, nv_clk_src_sppll0);
+	return clk->read(clk, nv_clk_src_sppll1);
+}
+
+static u32
+read_pll(struct nvc0_clock_priv *priv, u32 pll)
+{
+	struct nouveau_clock *clk = &priv->base;
+	u32 ctrl = nv_rd32(priv, pll + 0x00);
+	u32 coef = nv_rd32(priv, pll + 0x04);
+	u32 P = (coef & 0x003f0000) >> 16;
+	u32 N = (coef & 0x0000ff00) >> 8;
+	u32 M = (coef & 0x000000ff) >> 0;
+	u32 sclk;
+
+	if (!(ctrl & 0x00000001))
+		return 0;
+
+	switch (pll) {
+	case 0x00e800:
+	case 0x00e820:
+		sclk = nv_device(priv)->crystal;
+		P = 1;
+		break;
+	case 0x132000:
+		sclk = clk->read(clk, nv_clk_src_mpllsrc);
+		break;
+	case 0x132020:
+		sclk = clk->read(clk, nv_clk_src_mpllsrcref);
+		break;
+	case 0x137000:
+	case 0x137020:
+	case 0x137040:
+	case 0x1370e0:
+		sclk = read_div(priv, (pll & 0xff) / 0x20, 0x137120, 0x137140);
+		break;
+	default:
+		return 0;
+	}
+
+	return sclk * N / M / P;
+}
+
+static u32
+read_div(struct nvc0_clock_priv *priv, int doff, u32 dsrc, u32 dctl)
+{
+	u32 ssrc = nv_rd32(priv, dsrc + (doff * 4));
+	u32 sctl = nv_rd32(priv, dctl + (doff * 4));
+
+	switch (ssrc & 0x00000003) {
+	case 0:
+		if ((ssrc & 0x00030000) != 0x00030000)
+			return nv_device(priv)->crystal;
+		return 108000;
+	case 2:
+		return 100000;
+	case 3:
+		if (sctl & 0x80000000) {
+			u32 sclk = read_vco(priv, dsrc + (doff * 4));
+			u32 sdiv = (sctl & 0x0000003f) + 2;
+			return (sclk * 2) / sdiv;
+		}
+
+		return read_vco(priv, dsrc + (doff * 4));
+	default:
+		return 0;
+	}
+}
+
+static u32
+read_clk(struct nvc0_clock_priv *priv, int clk)
+{
+	u32 sctl = nv_rd32(priv, 0x137250 + (clk * 4));
+	u32 ssel = nv_rd32(priv, 0x137100);
+	u32 sclk, sdiv;
+
+	if (ssel & (1 << clk)) {
+		if (clk < 7)
+			sclk = read_pll(priv, 0x137000 + (clk * 0x20));
+		else
+			sclk = read_pll(priv, 0x1370e0);
+		sdiv = ((sctl & 0x00003f00) >> 8) + 2;
+	} else {
+		sclk = read_div(priv, clk, 0x137160, 0x1371d0);
+		sdiv = ((sctl & 0x0000003f) >> 0) + 2;
+	}
+
+	if (sctl & 0x80000000)
+		return (sclk * 2) / sdiv;
+
+	return sclk;
+}
+
+static int
+nvc0_clock_read(struct nouveau_clock *clk, enum nv_clk_src src)
+{
+	struct nouveau_device *device = nv_device(clk);
+	struct nvc0_clock_priv *priv = (void *)clk;
+
+	switch (src) {
+	case nv_clk_src_crystal:
+		return device->crystal;
+	case nv_clk_src_href:
+		return 100000;
+	case nv_clk_src_sppll0:
+		return read_pll(priv, 0x00e800);
+	case nv_clk_src_sppll1:
+		return read_pll(priv, 0x00e820);
+
+	case nv_clk_src_mpllsrcref:
+		return read_div(priv, 0, 0x137320, 0x137330);
+	case nv_clk_src_mpllsrc:
+		return read_pll(priv, 0x132020);
+	case nv_clk_src_mpll:
+		return read_pll(priv, 0x132000);
+	case nv_clk_src_mdiv:
+		return read_div(priv, 0, 0x137300, 0x137310);
+	case nv_clk_src_mem:
+		if (nv_rd32(priv, 0x1373f0) & 0x00000002)
+			return clk->read(clk, nv_clk_src_mpll);
+		return clk->read(clk, nv_clk_src_mdiv);
+
+	case nv_clk_src_gpc:
+		return read_clk(priv, 0x00);
+	case nv_clk_src_rop:
+		return read_clk(priv, 0x01);
+	case nv_clk_src_hubk07:
+		return read_clk(priv, 0x02);
+	case nv_clk_src_hubk06:
+		return read_clk(priv, 0x07);
+	case nv_clk_src_hubk01:
+		return read_clk(priv, 0x08);
+	case nv_clk_src_copy:
+		return read_clk(priv, 0x09);
+	case nv_clk_src_daemon:
+		return read_clk(priv, 0x0c);
+	case nv_clk_src_vdec:
+		return read_clk(priv, 0x0e);
+	default:
+		nv_error(clk, "invalid clock source %d\n", src);
+		return -EINVAL;
+	}
+}
+
+static u32
+calc_div(struct nvc0_clock_priv *priv, int clk, u32 ref, u32 freq, u32 *ddiv)
+{
+	u32 div = min((ref * 2) / freq, (u32)65);
+	if (div < 2)
+		div = 2;
+
+	*ddiv = div - 2;
+	return (ref * 2) / div;
+}
+
+static u32
+calc_src(struct nvc0_clock_priv *priv, int clk, u32 freq, u32 *dsrc, u32 *ddiv)
+{
+	u32 sclk;
+
+	/* use one of the fixed frequencies if possible */
+	*ddiv = 0x00000000;
+	switch (freq) {
+	case  27000:
+	case 108000:
+		*dsrc = 0x00000000;
+		if (freq == 108000)
+			*dsrc |= 0x00030000;
+		return freq;
+	case 100000:
+		*dsrc = 0x00000002;
+		return freq;
+	default:
+		*dsrc = 0x00000003;
+		break;
+	}
+
+	/* otherwise, calculate the closest divider */
+	sclk = read_vco(priv, 0x137160 + (clk * 4));
+	if (clk < 7)
+		sclk = calc_div(priv, clk, sclk, freq, ddiv);
+	return sclk;
+}
+
+static u32
+calc_pll(struct nvc0_clock_priv *priv, int clk, u32 freq, u32 *coef)
+{
+	struct nouveau_bios *bios = nouveau_bios(priv);
+	struct nvbios_pll limits;
+	int N, M, P, ret;
+
+	ret = nvbios_pll_parse(bios, 0x137000 + (clk * 0x20), &limits);
+	if (ret)
+		return 0;
+
+	limits.refclk = read_div(priv, clk, 0x137120, 0x137140);
+	if (!limits.refclk)
+		return 0;
+
+	ret = nva3_pll_calc(nv_subdev(priv), &limits, freq, &N, NULL, &M, &P);
+	if (ret <= 0)
+		return 0;
+
+	*coef = (P << 16) | (N << 8) | M;
+	return ret;
+}
+
+static int
+calc_clk(struct nvc0_clock_priv *priv,
+	 struct nouveau_cstate *cstate, int clk, int dom)
+{
+	struct nvc0_clock_info *info = &priv->eng[clk];
+	u32 freq = cstate->domain[dom];
+	u32 src0, div0, div1D, div1P = 0;
+	u32 clk0, clk1 = 0;
+
+	/* invalid clock domain */
+	if (!freq)
+		return 0;
+
+	/* first possible path, using only dividers */
+	clk0 = calc_src(priv, clk, freq, &src0, &div0);
+	clk0 = calc_div(priv, clk, clk0, freq, &div1D);
+
+	/* see if we can get any closer using PLLs */
+	if (clk0 != freq && (0x00004387 & (1 << clk))) {
+		if (clk <= 7)
+			clk1 = calc_pll(priv, clk, freq, &info->coef);
+		else
+			clk1 = cstate->domain[nv_clk_src_hubk06];
+		clk1 = calc_div(priv, clk, clk1, freq, &div1P);
+	}
+
+	/* select the method which gets closest to target freq */
+	if (abs((int)freq - clk0) <= abs((int)freq - clk1)) {
+		info->dsrc = src0;
+		if (div0) {
+			info->ddiv |= 0x80000000;
+			info->ddiv |= div0 << 8;
+			info->ddiv |= div0;
+		}
+		if (div1D) {
+			info->mdiv |= 0x80000000;
+			info->mdiv |= div1D;
+		}
+		info->ssel = info->coef = 0;
+		info->freq = clk0;
+	} else {
+		if (div1P) {
+			info->mdiv |= 0x80000000;
+			info->mdiv |= div1P << 8;
+		}
+		info->ssel = (1 << clk);
+		info->freq = clk1;
+	}
+
+	return 0;
+}
+
+static int
+nvc0_clock_calc(struct nouveau_clock *clk, struct nouveau_cstate *cstate)
+{
+	struct nvc0_clock_priv *priv = (void *)clk;
+	int ret;
+
+	if ((ret = calc_clk(priv, cstate, 0x00, nv_clk_src_gpc)) ||
+	    (ret = calc_clk(priv, cstate, 0x01, nv_clk_src_rop)) ||
+	    (ret = calc_clk(priv, cstate, 0x02, nv_clk_src_hubk07)) ||
+	    (ret = calc_clk(priv, cstate, 0x07, nv_clk_src_hubk06)) ||
+	    (ret = calc_clk(priv, cstate, 0x08, nv_clk_src_hubk01)) ||
+	    (ret = calc_clk(priv, cstate, 0x09, nv_clk_src_copy)) ||
+	    (ret = calc_clk(priv, cstate, 0x0c, nv_clk_src_daemon)) ||
+	    (ret = calc_clk(priv, cstate, 0x0e, nv_clk_src_vdec)))
+		return ret;
+
+	return 0;
+}
+
+static void
+nvc0_clock_prog_0(struct nvc0_clock_priv *priv, int clk)
+{
+	struct nvc0_clock_info *info = &priv->eng[clk];
+	if (clk < 7 && !info->ssel) {
+		nv_mask(priv, 0x1371d0 + (clk * 0x04), 0x80003f3f, info->ddiv);
+		nv_wr32(priv, 0x137160 + (clk * 0x04), info->dsrc);
+	}
+}
+
+static void
+nvc0_clock_prog_1(struct nvc0_clock_priv *priv, int clk)
+{
+	nv_mask(priv, 0x137100, (1 << clk), 0x00000000);
+	nv_wait(priv, 0x137100, (1 << clk), 0x00000000);
+}
+
+static void
+nvc0_clock_prog_2(struct nvc0_clock_priv *priv, int clk)
+{
+	struct nvc0_clock_info *info = &priv->eng[clk];
+	const u32 addr = 0x137000 + (clk * 0x20);
+	if (clk <= 7) {
+		nv_mask(priv, addr + 0x00, 0x00000004, 0x00000000);
+		nv_mask(priv, addr + 0x00, 0x00000001, 0x00000000);
+		if (info->coef) {
+			nv_wr32(priv, addr + 0x04, info->coef);
+			nv_mask(priv, addr + 0x00, 0x00000001, 0x00000001);
+			nv_wait(priv, addr + 0x00, 0x00020000, 0x00020000);
+			nv_mask(priv, addr + 0x00, 0x00020004, 0x00000004);
+		}
+	}
+}
+
+static void
+nvc0_clock_prog_3(struct nvc0_clock_priv *priv, int clk)
+{
+	struct nvc0_clock_info *info = &priv->eng[clk];
+	if (info->ssel) {
+		nv_mask(priv, 0x137100, (1 << clk), info->ssel);
+		nv_wait(priv, 0x137100, (1 << clk), info->ssel);
+	}
+}
+
+static void
+nvc0_clock_prog_4(struct nvc0_clock_priv *priv, int clk)
+{
+	struct nvc0_clock_info *info = &priv->eng[clk];
+	nv_mask(priv, 0x137250 + (clk * 0x04), 0x00003f3f, info->mdiv);
+}
+
+static int
+nvc0_clock_prog(struct nouveau_clock *clk)
+{
+	struct nvc0_clock_priv *priv = (void *)clk;
+	struct {
+		void (*exec)(struct nvc0_clock_priv *, int);
+	} stage[] = {
+		{ nvc0_clock_prog_0 }, /* div programming */
+		{ nvc0_clock_prog_1 }, /* select div mode */
+		{ nvc0_clock_prog_2 }, /* (maybe) program pll */
+		{ nvc0_clock_prog_3 }, /* (maybe) select pll mode */
+		{ nvc0_clock_prog_4 }, /* final divider */
+	};
+	int i, j;
+
+	for (i = 0; i < ARRAY_SIZE(stage); i++) {
+		for (j = 0; j < ARRAY_SIZE(priv->eng); j++) {
+			if (!priv->eng[j].freq)
+				continue;
+			stage[i].exec(priv, j);
+		}
+	}
+
+	return 0;
+}
+
+static void
+nvc0_clock_tidy(struct nouveau_clock *clk)
+{
+	struct nvc0_clock_priv *priv = (void *)clk;
+	memset(priv->eng, 0x00, sizeof(priv->eng));
+}
+
+static struct nouveau_clocks
+nvc0_domain[] = {
+	{ nv_clk_src_crystal, 0xff },
+	{ nv_clk_src_href   , 0xff },
+	{ nv_clk_src_hubk06 , 0x00 },
+	{ nv_clk_src_hubk01 , 0x01 },
+	{ nv_clk_src_copy   , 0x02 },
+	{ nv_clk_src_gpc    , 0x03, 0, "core", 2000 },
+	{ nv_clk_src_rop    , 0x04 },
+	{ nv_clk_src_mem    , 0x05, 0, "memory", 1000 },
+	{ nv_clk_src_vdec   , 0x06 },
+	{ nv_clk_src_daemon , 0x0a },
+	{ nv_clk_src_hubk07 , 0x0b },
+	{ nv_clk_src_max }
 };
 
 static int
@@ -40,12 +437,15 @@ nvc0_clock_ctor(struct nouveau_object *parent, struct nouveau_object *engine,
 	struct nvc0_clock_priv *priv;
 	int ret;
 
-	ret = nouveau_clock_create(parent, engine, oclass, &priv);
+	ret = nouveau_clock_create(parent, engine, oclass, nvc0_domain, &priv);
 	*pobject = nv_object(priv);
 	if (ret)
 		return ret;
 
-	priv->base.pll_calc = nva3_clock_pll_calc;
+	priv->base.read = nvc0_clock_read;
+	priv->base.calc = nvc0_clock_calc;
+	priv->base.prog = nvc0_clock_prog;
+	priv->base.tidy = nvc0_clock_tidy;
 	return 0;
 }
 
diff --git a/drivers/gpu/drm/nouveau/core/subdev/clock/nve0.c b/drivers/gpu/drm/nouveau/core/subdev/clock/nve0.c
new file mode 100644
index 000000000000..4c62e84b96f5
--- /dev/null
+++ b/drivers/gpu/drm/nouveau/core/subdev/clock/nve0.c
@@ -0,0 +1,497 @@
+/*
+ * Copyright 2013 Red Hat Inc.
+ *
+ * Permission is hereby granted, free of charge, to any person obtaining a
+ * copy of this software and associated documentation files (the "Software"),
+ * to deal in the Software without restriction, including without limitation
+ * the rights to use, copy, modify, merge, publish, distribute, sublicense,
+ * and/or sell copies of the Software, and to permit persons to whom the
+ * Software is furnished to do so, subject to the following conditions:
+ *
+ * The above copyright notice and this permission notice shall be included in
+ * all copies or substantial portions of the Software.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+ * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+ * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT.  IN NO EVENT SHALL
+ * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR
+ * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE,
+ * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR
+ * OTHER DEALINGS IN THE SOFTWARE.
+ *
+ * Authors: Ben Skeggs
+ */
+
+#include <subdev/clock.h>
+#include <subdev/timer.h>
+#include <subdev/bios.h>
+#include <subdev/bios/pll.h>
+
+#include "pll.h"
+
+struct nve0_clock_info {
+	u32 freq;
+	u32 ssel;
+	u32 mdiv;
+	u32 dsrc;
+	u32 ddiv;
+	u32 coef;
+};
+
+struct nve0_clock_priv {
+	struct nouveau_clock base;
+	struct nve0_clock_info eng[16];
+};
+
+static u32 read_div(struct nve0_clock_priv *, int, u32, u32);
+static u32 read_pll(struct nve0_clock_priv *, u32);
+
+static u32
+read_vco(struct nve0_clock_priv *priv, u32 dsrc)
+{
+	u32 ssrc = nv_rd32(priv, dsrc);
+	if (!(ssrc & 0x00000100))
+		return read_pll(priv, 0x00e800);
+	return read_pll(priv, 0x00e820);
+}
+
+static u32
+read_pll(struct nve0_clock_priv *priv, u32 pll)
+{
+	u32 ctrl = nv_rd32(priv, pll + 0x00);
+	u32 coef = nv_rd32(priv, pll + 0x04);
+	u32 P = (coef & 0x003f0000) >> 16;
+	u32 N = (coef & 0x0000ff00) >> 8;
+	u32 M = (coef & 0x000000ff) >> 0;
+	u32 sclk;
+	u16 fN = 0xf000;
+
+	if (!(ctrl & 0x00000001))
+		return 0;
+
+	switch (pll) {
+	case 0x00e800:
+	case 0x00e820:
+		sclk = nv_device(priv)->crystal;
+		P = 1;
+		break;
+	case 0x132000:
+		sclk = read_pll(priv, 0x132020);
+		P = (coef & 0x10000000) ? 2 : 1;
+		break;
+	case 0x132020:
+		sclk = read_div(priv, 0, 0x137320, 0x137330);
+		fN   = nv_rd32(priv, pll + 0x10) >> 16;
+		break;
+	case 0x137000:
+	case 0x137020:
+	case 0x137040:
+	case 0x1370e0:
+		sclk = read_div(priv, (pll & 0xff) / 0x20, 0x137120, 0x137140);
+		break;
+	default:
+		return 0;
+	}
+
+	if (P == 0)
+		P = 1;
+
+	sclk = (sclk * N) + (((u16)(fN + 4096) * sclk) >> 13);
+	return sclk / (M * P);
+}
+
+static u32
+read_div(struct nve0_clock_priv *priv, int doff, u32 dsrc, u32 dctl)
+{
+	u32 ssrc = nv_rd32(priv, dsrc + (doff * 4));
+	u32 sctl = nv_rd32(priv, dctl + (doff * 4));
+
+	switch (ssrc & 0x00000003) {
+	case 0:
+		if ((ssrc & 0x00030000) != 0x00030000)
+			return nv_device(priv)->crystal;
+		return 108000;
+	case 2:
+		return 100000;
+	case 3:
+		if (sctl & 0x80000000) {
+			u32 sclk = read_vco(priv, dsrc + (doff * 4));
+			u32 sdiv = (sctl & 0x0000003f) + 2;
+			return (sclk * 2) / sdiv;
+		}
+
+		return read_vco(priv, dsrc + (doff * 4));
+	default:
+		return 0;
+	}
+}
+
+static u32
+read_mem(struct nve0_clock_priv *priv)
+{
+	switch (nv_rd32(priv, 0x1373f4) & 0x0000000f) {
+	case 1: return read_pll(priv, 0x132020);
+	case 2: return read_pll(priv, 0x132000);
+	default:
+		return 0;
+	}
+}
+
+static u32
+read_clk(struct nve0_clock_priv *priv, int clk)
+{
+	u32 sctl = nv_rd32(priv, 0x137250 + (clk * 4));
+	u32 sclk, sdiv;
+
+	if (clk < 7) {
+		u32 ssel = nv_rd32(priv, 0x137100);
+		if (ssel & (1 << clk)) {
+			sclk = read_pll(priv, 0x137000 + (clk * 0x20));
+			sdiv = 1;
+		} else {
+			sclk = read_div(priv, clk, 0x137160, 0x1371d0);
+			sdiv = 0;
+		}
+	} else {
+		u32 ssrc = nv_rd32(priv, 0x137160 + (clk * 0x04));
+		if ((ssrc & 0x00000003) == 0x00000003) {
+			sclk = read_div(priv, clk, 0x137160, 0x1371d0);
+			if (ssrc & 0x00000100) {
+				if (ssrc & 0x40000000)
+					sclk = read_pll(priv, 0x1370e0);
+				sdiv = 1;
+			} else {
+				sdiv = 0;
+			}
+		} else {
+			sclk = read_div(priv, clk, 0x137160, 0x1371d0);
+			sdiv = 0;
+		}
+	}
+
+	if (sctl & 0x80000000) {
+		if (sdiv)
+			sdiv = ((sctl & 0x00003f00) >> 8) + 2;
+		else
+			sdiv = ((sctl & 0x0000003f) >> 0) + 2;
+		return (sclk * 2) / sdiv;
+	}
+
+	return sclk;
+}
+
+static int
+nve0_clock_read(struct nouveau_clock *clk, enum nv_clk_src src)
+{
+	struct nouveau_device *device = nv_device(clk);
+	struct nve0_clock_priv *priv = (void *)clk;
+
+	switch (src) {
+	case nv_clk_src_crystal:
+		return device->crystal;
+	case nv_clk_src_href:
+		return 100000;
+	case nv_clk_src_mem:
+		return read_mem(priv);
+	case nv_clk_src_gpc:
+		return read_clk(priv, 0x00);
+	case nv_clk_src_rop:
+		return read_clk(priv, 0x01);
+	case nv_clk_src_hubk07:
+		return read_clk(priv, 0x02);
+	case nv_clk_src_hubk06:
+		return read_clk(priv, 0x07);
+	case nv_clk_src_hubk01:
+		return read_clk(priv, 0x08);
+	case nv_clk_src_daemon:
+		return read_clk(priv, 0x0c);
+	case nv_clk_src_vdec:
+		return read_clk(priv, 0x0e);
+	default:
+		nv_error(clk, "invalid clock source %d\n", src);
+		return -EINVAL;
+	}
+}
+
+static u32
+calc_div(struct nve0_clock_priv *priv, int clk, u32 ref, u32 freq, u32 *ddiv)
+{
+	u32 div = min((ref * 2) / freq, (u32)65);
+	if (div < 2)
+		div = 2;
+
+	*ddiv = div - 2;
+	return (ref * 2) / div;
+}
+
+static u32
+calc_src(struct nve0_clock_priv *priv, int clk, u32 freq, u32 *dsrc, u32 *ddiv)
+{
+	u32 sclk;
+
+	/* use one of the fixed frequencies if possible */
+	*ddiv = 0x00000000;
+	switch (freq) {
+	case  27000:
+	case 108000:
+		*dsrc = 0x00000000;
+		if (freq == 108000)
+			*dsrc |= 0x00030000;
+		return freq;
+	case 100000:
+		*dsrc = 0x00000002;
+		return freq;
+	default:
+		*dsrc = 0x00000003;
+		break;
+	}
+
+	/* otherwise, calculate the closest divider */
+	sclk = read_vco(priv, 0x137160 + (clk * 4));
+	if (clk < 7)
+		sclk = calc_div(priv, clk, sclk, freq, ddiv);
+	return sclk;
+}
+
+static u32
+calc_pll(struct nve0_clock_priv *priv, int clk, u32 freq, u32 *coef)
+{
+	struct nouveau_bios *bios = nouveau_bios(priv);
+	struct nvbios_pll limits;
+	int N, M, P, ret;
+
+	ret = nvbios_pll_parse(bios, 0x137000 + (clk * 0x20), &limits);
+	if (ret)
+		return 0;
+
+	limits.refclk = read_div(priv, clk, 0x137120, 0x137140);
+	if (!limits.refclk)
+		return 0;
+
+	ret = nva3_pll_calc(nv_subdev(priv), &limits, freq, &N, NULL, &M, &P);
+	if (ret <= 0)
+		return 0;
+
+	*coef = (P << 16) | (N << 8) | M;
+	return ret;
+}
+
+static int
+calc_clk(struct nve0_clock_priv *priv,
+	 struct nouveau_cstate *cstate, int clk, int dom)
+{
+	struct nve0_clock_info *info = &priv->eng[clk];
+	u32 freq = cstate->domain[dom];
+	u32 src0, div0, div1D, div1P = 0;
+	u32 clk0, clk1 = 0;
+
+	/* invalid clock domain */
+	if (!freq)
+		return 0;
+
+	/* first possible path, using only dividers */
+	clk0 = calc_src(priv, clk, freq, &src0, &div0);
+	clk0 = calc_div(priv, clk, clk0, freq, &div1D);
+
+	/* see if we can get any closer using PLLs */
+	if (clk0 != freq && (0x0000ff87 & (1 << clk))) {
+		if (clk <= 7)
+			clk1 = calc_pll(priv, clk, freq, &info->coef);
+		else
+			clk1 = cstate->domain[nv_clk_src_hubk06];
+		clk1 = calc_div(priv, clk, clk1, freq, &div1P);
+	}
+
+	/* select the method which gets closest to target freq */
+	if (abs((int)freq - clk0) <= abs((int)freq - clk1)) {
+		info->dsrc = src0;
+		if (div0) {
+			info->ddiv |= 0x80000000;
+			info->ddiv |= div0 << 8;
+			info->ddiv |= div0;
+		}
+		if (div1D) {
+			info->mdiv |= 0x80000000;
+			info->mdiv |= div1D;
+		}
+		info->ssel = 0;
+		info->freq = clk0;
+	} else {
+		if (div1P) {
+			info->mdiv |= 0x80000000;
+			info->mdiv |= div1P << 8;
+		}
+		info->ssel = (1 << clk);
+		info->dsrc = 0x40000100;
+		info->freq = clk1;
+	}
+
+	return 0;
+}
+
+static int
+nve0_clock_calc(struct nouveau_clock *clk, struct nouveau_cstate *cstate)
+{
+	struct nve0_clock_priv *priv = (void *)clk;
+	int ret;
+
+	if ((ret = calc_clk(priv, cstate, 0x00, nv_clk_src_gpc)) ||
+	    (ret = calc_clk(priv, cstate, 0x01, nv_clk_src_rop)) ||
+	    (ret = calc_clk(priv, cstate, 0x02, nv_clk_src_hubk07)) ||
+	    (ret = calc_clk(priv, cstate, 0x07, nv_clk_src_hubk06)) ||
+	    (ret = calc_clk(priv, cstate, 0x08, nv_clk_src_hubk01)) ||
+	    (ret = calc_clk(priv, cstate, 0x0c, nv_clk_src_daemon)) ||
+	    (ret = calc_clk(priv, cstate, 0x0e, nv_clk_src_vdec)))
+		return ret;
+
+	return 0;
+}
+
+static void
+nve0_clock_prog_0(struct nve0_clock_priv *priv, int clk)
+{
+	struct nve0_clock_info *info = &priv->eng[clk];
+	if (!info->ssel) {
+		nv_mask(priv, 0x1371d0 + (clk * 0x04), 0x80003f3f, info->ddiv);
+		nv_wr32(priv, 0x137160 + (clk * 0x04), info->dsrc);
+	}
+}
+
+static void
+nve0_clock_prog_1_0(struct nve0_clock_priv *priv, int clk)
+{
+	nv_mask(priv, 0x137100, (1 << clk), 0x00000000);
+	nv_wait(priv, 0x137100, (1 << clk), 0x00000000);
+}
+
+static void
+nve0_clock_prog_1_1(struct nve0_clock_priv *priv, int clk)
+{
+	nv_mask(priv, 0x137160 + (clk * 0x04), 0x00000100, 0x00000000);
+}
+
+static void
+nve0_clock_prog_2(struct nve0_clock_priv *priv, int clk)
+{
+	struct nve0_clock_info *info = &priv->eng[clk];
+	const u32 addr = 0x137000 + (clk * 0x20);
+	nv_mask(priv, addr + 0x00, 0x00000004, 0x00000000);
+	nv_mask(priv, addr + 0x00, 0x00000001, 0x00000000);
+	if (info->coef) {
+		nv_wr32(priv, addr + 0x04, info->coef);
+		nv_mask(priv, addr + 0x00, 0x00000001, 0x00000001);
+		nv_wait(priv, addr + 0x00, 0x00020000, 0x00020000);
+		nv_mask(priv, addr + 0x00, 0x00020004, 0x00000004);
+	}
+}
+
+static void
+nve0_clock_prog_3(struct nve0_clock_priv *priv, int clk)
+{
+	struct nve0_clock_info *info = &priv->eng[clk];
+	nv_mask(priv, 0x137250 + (clk * 0x04), 0x00003f3f, info->mdiv);
+}
+
+static void
+nve0_clock_prog_4_0(struct nve0_clock_priv *priv, int clk)
+{
+	struct nve0_clock_info *info = &priv->eng[clk];
+	if (info->ssel) {
+		nv_mask(priv, 0x137100, (1 << clk), info->ssel);
+		nv_wait(priv, 0x137100, (1 << clk), info->ssel);
+	}
+}
+
+static void
+nve0_clock_prog_4_1(struct nve0_clock_priv *priv, int clk)
+{
+	struct nve0_clock_info *info = &priv->eng[clk];
+	if (info->ssel) {
+		nv_mask(priv, 0x137160 + (clk * 0x04), 0x40000000, 0x40000000);
+		nv_mask(priv, 0x137160 + (clk * 0x04), 0x00000100, 0x00000100);
+	}
+}
+
+static int
+nve0_clock_prog(struct nouveau_clock *clk)
+{
+	struct nve0_clock_priv *priv = (void *)clk;
+	struct {
+		u32 mask;
+		void (*exec)(struct nve0_clock_priv *, int);
+	} stage[] = {
+		{ 0x007f, nve0_clock_prog_0   }, /* div programming */
+		{ 0x007f, nve0_clock_prog_1_0 }, /* select div mode */
+		{ 0xff80, nve0_clock_prog_1_1 },
+		{ 0x00ff, nve0_clock_prog_2   }, /* (maybe) program pll */
+		{ 0xff80, nve0_clock_prog_3   }, /* final divider */
+		{ 0x007f, nve0_clock_prog_4_0 }, /* (maybe) select pll mode */
+		{ 0xff80, nve0_clock_prog_4_1 },
+	};
+	int i, j;
+
+	for (i = 0; i < ARRAY_SIZE(stage); i++) {
+		for (j = 0; j < ARRAY_SIZE(priv->eng); j++) {
+			if (!(stage[i].mask & (1 << j)))
+				continue;
+			if (!priv->eng[j].freq)
+				continue;
+			stage[i].exec(priv, j);
+		}
+	}
+
+	return 0;
+}
+
+static void
+nve0_clock_tidy(struct nouveau_clock *clk)
+{
+	struct nve0_clock_priv *priv = (void *)clk;
+	memset(priv->eng, 0x00, sizeof(priv->eng));
+}
+
+static struct nouveau_clocks
+nve0_domain[] = {
+	{ nv_clk_src_crystal, 0xff },
+	{ nv_clk_src_href   , 0xff },
+	{ nv_clk_src_gpc    , 0x00, NVKM_CLK_DOM_FLAG_CORE, "core", 2000 },
+	{ nv_clk_src_hubk07 , 0x01, NVKM_CLK_DOM_FLAG_CORE },
+	{ nv_clk_src_rop    , 0x02, NVKM_CLK_DOM_FLAG_CORE },
+	{ nv_clk_src_mem    , 0x03, 0, "memory", 1000 },
+	{ nv_clk_src_hubk06 , 0x04, NVKM_CLK_DOM_FLAG_CORE },
+	{ nv_clk_src_hubk01 , 0x05 },
+	{ nv_clk_src_vdec   , 0x06 },
+	{ nv_clk_src_daemon , 0x07 },
+	{ nv_clk_src_max }
+};
+
+static int
+nve0_clock_ctor(struct nouveau_object *parent, struct nouveau_object *engine,
+		struct nouveau_oclass *oclass, void *data, u32 size,
+		struct nouveau_object **pobject)
+{
+	struct nve0_clock_priv *priv;
+	int ret;
+
+	ret = nouveau_clock_create(parent, engine, oclass, nve0_domain, &priv);
+	*pobject = nv_object(priv);
+	if (ret)
+		return ret;
+
+	priv->base.read = nve0_clock_read;
+	priv->base.calc = nve0_clock_calc;
+	priv->base.prog = nve0_clock_prog;
+	priv->base.tidy = nve0_clock_tidy;
+	return 0;
+}
+
+struct nouveau_oclass
+nve0_clock_oclass = {
+	.handle = NV_SUBDEV(CLOCK, 0xe0),
+	.ofuncs = &(struct nouveau_ofuncs) {
+		.ctor = nve0_clock_ctor,
+		.dtor = _nouveau_clock_dtor,
+		.init = _nouveau_clock_init,
+		.fini = _nouveau_clock_fini,
+	},
+};
diff --git a/drivers/gpu/drm/nouveau/core/subdev/clock/pllnv04.c b/drivers/gpu/drm/nouveau/core/subdev/clock/pllnv04.c
index cf1ed0dc9bc9..b47d543ab2e3 100644
--- a/drivers/gpu/drm/nouveau/core/subdev/clock/pllnv04.c
+++ b/drivers/gpu/drm/nouveau/core/subdev/clock/pllnv04.c
@@ -38,7 +38,7 @@ getMNP_single(struct nouveau_subdev *subdev, struct nvbios_pll *info, int clk,
 	 * "clk" parameter in kHz
 	 * returns calculated clock
 	 */
-	int cv = nouveau_bios(subdev)->version.chip;
+	struct nouveau_bios *bios = nouveau_bios(subdev);
 	int minvco = info->vco1.min_freq, maxvco = info->vco1.max_freq;
 	int minM = info->vco1.min_m, maxM = info->vco1.max_m;
 	int minN = info->vco1.min_n, maxN = info->vco1.max_n;
@@ -54,18 +54,21 @@ getMNP_single(struct nouveau_subdev *subdev, struct nvbios_pll *info, int clk,
 
 	/* this division verified for nv20, nv18, nv28 (Haiku), and nv34 */
 	/* possibly correlated with introduction of 27MHz crystal */
-	if (cv < 0x17 || cv == 0x1a || cv == 0x20) {
-		if (clk > 250000)
-			maxM = 6;
-		if (clk > 340000)
-			maxM = 2;
-	} else if (cv < 0x40) {
-		if (clk > 150000)
-			maxM = 6;
-		if (clk > 200000)
-			maxM = 4;
-		if (clk > 340000)
-			maxM = 2;
+	if (bios->version.major < 0x60) {
+		int cv = bios->version.chip;
+		if (cv < 0x17 || cv == 0x1a || cv == 0x20) {
+			if (clk > 250000)
+				maxM = 6;
+			if (clk > 340000)
+				maxM = 2;
+		} else if (cv < 0x40) {
+			if (clk > 150000)
+				maxM = 6;
+			if (clk > 200000)
+				maxM = 4;
+			if (clk > 340000)
+				maxM = 2;
+		}
 	}
 
 	P = 1 << maxP;
@@ -227,10 +230,12 @@ nv04_pll_calc(struct nouveau_subdev *subdev, struct nvbios_pll *info, u32 freq,
 {
 	int ret;
 
-	if (!info->vco2.max_freq) {
+	if (!info->vco2.max_freq || !N2) {
 		ret = getMNP_single(subdev, info, freq, N1, M1, P);
-		*N2 = 1;
-		*M2 = 1;
+		if (N2) {
+			*N2 = 1;
+			*M2 = 1;
+		}
 	} else {
 		ret = getMNP_double(subdev, info, freq, N1, M1, N2, M2, P);
 	}
diff --git a/drivers/gpu/drm/nouveau/core/subdev/clock/pllnva3.c b/drivers/gpu/drm/nouveau/core/subdev/clock/pllnva3.c
index 2fe1f712eefa..8eca457c2814 100644
--- a/drivers/gpu/drm/nouveau/core/subdev/clock/pllnva3.c
+++ b/drivers/gpu/drm/nouveau/core/subdev/clock/pllnva3.c
@@ -45,6 +45,7 @@ nva3_pll_calc(struct nouveau_subdev *subdev, struct nvbios_pll *info,
 	lM = max(lM, (int)info->vco1.min_m);
 	hM = (info->refclk + info->vco1.min_inputfreq) / info->vco1.min_inputfreq;
 	hM = min(hM, (int)info->vco1.max_m);
+	lM = min(lM, hM);
 
 	for (M = lM; M <= hM; M++) {
 		u32 tmp = freq * *P * M;
diff --git a/drivers/gpu/drm/nouveau/core/subdev/clock/seq.h b/drivers/gpu/drm/nouveau/core/subdev/clock/seq.h
new file mode 100644
index 000000000000..fb33f06ebd59
--- /dev/null
+++ b/drivers/gpu/drm/nouveau/core/subdev/clock/seq.h
@@ -0,0 +1,17 @@
+#ifndef __NVKM_CLK_SEQ_H__
+#define __NVKM_CLK_SEQ_H__
+
+#include <subdev/bus.h>
+#include <subdev/bus/hwsq.h>
+
+#define clk_init(s,p)       hwsq_init(&(s)->base, (p))
+#define clk_exec(s,e)       hwsq_exec(&(s)->base, (e))
+#define clk_have(s,r)       ((s)->r_##r.addr != 0x000000)
+#define clk_rd32(s,r)       hwsq_rd32(&(s)->base, &(s)->r_##r)
+#define clk_wr32(s,r,d)     hwsq_wr32(&(s)->base, &(s)->r_##r, (d))
+#define clk_mask(s,r,m,d)   hwsq_mask(&(s)->base, &(s)->r_##r, (m), (d))
+#define clk_setf(s,f,d)     hwsq_setf(&(s)->base, (f), (d))
+#define clk_wait(s,f,d)     hwsq_wait(&(s)->base, (f), (d))
+#define clk_nsec(s,n)       hwsq_nsec(&(s)->base, (n))
+
+#endif
diff --git a/drivers/gpu/drm/nouveau/core/subdev/devinit/nv04.c b/drivers/gpu/drm/nouveau/core/subdev/devinit/nv04.c
index b22357d9b821..27c8235f1a85 100644
--- a/drivers/gpu/drm/nouveau/core/subdev/devinit/nv04.c
+++ b/drivers/gpu/drm/nouveau/core/subdev/devinit/nv04.c
@@ -168,7 +168,8 @@ setPLL_single(struct nouveau_devinit *devinit, u32 reg,
 		/* downclock -- write new NM first */
 		nv_wr32(devinit, reg, (oldpll & 0xffff0000) | pv->NM1);
 
-	if (chip_version < 0x17 && chip_version != 0x11)
+	if ((chip_version < 0x17 || chip_version == 0x1a) &&
+	    chip_version != 0x11)
 		/* wait a bit on older chips */
 		msleep(64);
 	nv_rd32(devinit, reg);
diff --git a/drivers/gpu/drm/nouveau/core/subdev/devinit/nv10.c b/drivers/gpu/drm/nouveau/core/subdev/devinit/nv10.c
index 463b08fa0968..8d274dba1ef1 100644
--- a/drivers/gpu/drm/nouveau/core/subdev/devinit/nv10.c
+++ b/drivers/gpu/drm/nouveau/core/subdev/devinit/nv10.c
@@ -38,12 +38,18 @@ static void
 nv10_devinit_meminit(struct nouveau_devinit *devinit)
 {
 	struct nv10_devinit_priv *priv = (void *)devinit;
-	const int mem_width[] = { 0x10, 0x00, 0x20 };
-	const int mem_width_count = nv_device(priv)->chipset >= 0x17 ? 3 : 2;
+	static const int mem_width[] = { 0x10, 0x00, 0x20 };
+	int mem_width_count;
 	uint32_t patt = 0xdeadbeef;
 	struct io_mapping *fb;
 	int i, j, k;
 
+	if (nv_device(priv)->card_type >= NV_11 &&
+	    nv_device(priv)->chipset >= 0x17)
+		mem_width_count = 3;
+	else
+		mem_width_count = 2;
+
 	/* Map the framebuffer aperture */
 	fb = fbmem_init(nv_device(priv)->pdev);
 	if (!fb) {
diff --git a/drivers/gpu/drm/nouveau/core/subdev/fb/base.c b/drivers/gpu/drm/nouveau/core/subdev/fb/base.c
index 821cd75b86a3..f009d8a39d9d 100644
--- a/drivers/gpu/drm/nouveau/core/subdev/fb/base.c
+++ b/drivers/gpu/drm/nouveau/core/subdev/fb/base.c
@@ -22,9 +22,10 @@
  * Authors: Ben Skeggs
  */
 
-#include "subdev/fb.h"
-#include "subdev/bios.h"
-#include "subdev/bios/bit.h"
+#include <subdev/bios.h>
+#include <subdev/bios/bit.h>
+
+#include "priv.h"
 
 int
 nouveau_fb_bios_memtype(struct nouveau_bios *bios)
@@ -106,9 +107,9 @@ _nouveau_fb_dtor(struct nouveau_object *object)
 
 int
 nouveau_fb_create_(struct nouveau_object *parent, struct nouveau_object *engine,
-		   struct nouveau_oclass *oclass, struct nouveau_oclass *ramcls,
-		   int length, void **pobject)
+		   struct nouveau_oclass *oclass, int length, void **pobject)
 {
+	struct nouveau_fb_impl *impl = (void *)oclass;
 	static const char *name[] = {
 		[NV_MEM_TYPE_UNKNOWN] = "unknown",
 		[NV_MEM_TYPE_STOLEN ] = "stolen system memory",
@@ -132,8 +133,10 @@ nouveau_fb_create_(struct nouveau_object *parent, struct nouveau_object *engine,
 	if (ret)
 		return ret;
 
+	pfb->memtype_valid = impl->memtype;
+
 	ret = nouveau_object_ctor(nv_object(pfb), nv_object(pfb),
-				  ramcls, NULL, 0, &ram);
+				  impl->ram, NULL, 0, &ram);
 	if (ret) {
 		nv_fatal(pfb, "error detecting memory configuration!!\n");
 		return ret;
diff --git a/drivers/gpu/drm/nouveau/core/subdev/fb/gddr5.c b/drivers/gpu/drm/nouveau/core/subdev/fb/gddr5.c
new file mode 100644
index 000000000000..34f9605ffee6
--- /dev/null
+++ b/drivers/gpu/drm/nouveau/core/subdev/fb/gddr5.c
@@ -0,0 +1,96 @@
+/*
+ * Copyright 2013 Red Hat Inc.
+ *
+ * Permission is hereby granted, free of charge, to any person obtaining a
+ * copy of this software and associated documentation files (the "Software"),
+ * to deal in the Software without restriction, including without limitation
+ * the rights to use, copy, modify, merge, publish, distribute, sublicense,
+ * and/or sell copies of the Software, and to permit persons to whom the
+ * Software is furnished to do so, subject to the following conditions:
+ *
+ * The above copyright notice and this permission notice shall be included in
+ * all copies or substantial portions of the Software.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+ * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+ * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT.  IN NO EVENT SHALL
+ * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR
+ * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE,
+ * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR
+ * OTHER DEALINGS IN THE SOFTWARE.
+ *
+ * Authors: Ben Skeggs <bskeggs@redhat.com>
+ */
+
+#include <subdev/bios.h>
+#include "priv.h"
+
+int
+nouveau_gddr5_calc(struct nouveau_ram *ram)
+{
+	struct nouveau_bios *bios = nouveau_bios(ram);
+	int pd, lf, xd, vh, vr, vo;
+	int WL, CL, WR, at, dt, ds;
+	int rq = ram->freq < 1000000; /* XXX */
+
+	switch (!!ram->ramcfg.data * ram->ramcfg.version) {
+	case 0x11:
+		pd =  (nv_ro08(bios, ram->ramcfg.data + 0x01) & 0x80) >> 7;
+		lf =  (nv_ro08(bios, ram->ramcfg.data + 0x01) & 0x40) >> 6;
+		xd = !(nv_ro08(bios, ram->ramcfg.data + 0x01) & 0x20);
+		vh =  (nv_ro08(bios, ram->ramcfg.data + 0x02) & 0x10) >> 4;
+		vr =  (nv_ro08(bios, ram->ramcfg.data + 0x02) & 0x04) >> 2;
+		vo =   nv_ro08(bios, ram->ramcfg.data + 0x06) & 0xff;
+		break;
+	default:
+		return -ENOSYS;
+	}
+
+	switch (!!ram->timing.data * ram->timing.version) {
+	case 0x20:
+		WL = (nv_ro16(bios, ram->timing.data + 0x04) & 0x0f80) >> 7;
+		CL =  nv_ro08(bios, ram->timing.data + 0x04) & 0x1f;
+		WR =  nv_ro08(bios, ram->timing.data + 0x0a) & 0x7f;
+		at = (nv_ro08(bios, ram->timing.data + 0x2e) & 0xc0) >> 6;
+		dt =  nv_ro08(bios, ram->timing.data + 0x2e) & 0x03;
+		ds =  nv_ro08(bios, ram->timing.data + 0x2f) & 0x03;
+		break;
+	default:
+		return -ENOSYS;
+	}
+
+	if (WL < 1 || WL > 7 || CL < 5 || CL > 36 || WR < 4 || WR > 35)
+		return -EINVAL;
+	CL -= 5;
+	WR -= 4;
+
+	ram->mr[0] &= ~0xf7f;
+	ram->mr[0] |= (WR & 0x0f) << 8;
+	ram->mr[0] |= (CL & 0x0f) << 3;
+	ram->mr[0] |= (WL & 0x07) << 0;
+
+	ram->mr[1] &= ~0x0bf;
+	ram->mr[1] |= (xd & 0x01) << 7;
+	ram->mr[1] |= (at & 0x03) << 4;
+	ram->mr[1] |= (dt & 0x03) << 2;
+	ram->mr[1] |= (ds & 0x03) << 0;
+
+	ram->mr[3] &= ~0x020;
+	ram->mr[3] |= (rq & 0x01) << 5;
+
+	if (!vo)
+		vo = (ram->mr[6] & 0xff0) >> 4;
+	if (ram->mr[6] & 0x001)
+		pd = 1; /* binary driver does this.. bug? */
+	ram->mr[6] &= ~0xff1;
+	ram->mr[6] |= (vo & 0xff) << 4;
+	ram->mr[6] |= (pd & 0x01) << 0;
+
+	if (!(ram->mr[7] & 0x100))
+		vr = 0; /* binary driver does this.. bug? */
+	ram->mr[7] &= ~0x188;
+	ram->mr[7] |= (vr & 0x01) << 8;
+	ram->mr[7] |= (vh & 0x01) << 7;
+	ram->mr[7] |= (lf & 0x01) << 3;
+	return 0;
+}
diff --git a/drivers/gpu/drm/nouveau/core/subdev/fb/nv04.c b/drivers/gpu/drm/nouveau/core/subdev/fb/nv04.c
index 1f103c7b89fa..8309fe33fe84 100644
--- a/drivers/gpu/drm/nouveau/core/subdev/fb/nv04.c
+++ b/drivers/gpu/drm/nouveau/core/subdev/fb/nv04.c
@@ -22,14 +22,10 @@
  * Authors: Ben Skeggs
  */
 
-#include "priv.h"
+#include "nv04.h"
 
 #define NV04_PFB_CFG0						0x00100200
 
-struct nv04_fb_priv {
-	struct nouveau_fb base;
-};
-
 bool
 nv04_fb_memtype_valid(struct nouveau_fb *pfb, u32 tile_flags)
 {
@@ -57,30 +53,37 @@ nv04_fb_init(struct nouveau_object *object)
 	return 0;
 }
 
-static int
+int
 nv04_fb_ctor(struct nouveau_object *parent, struct nouveau_object *engine,
 	     struct nouveau_oclass *oclass, void *data, u32 size,
 	     struct nouveau_object **pobject)
 {
+	struct nv04_fb_impl *impl = (void *)oclass;
 	struct nv04_fb_priv *priv;
 	int ret;
 
-	ret = nouveau_fb_create(parent, engine, oclass, &nv04_ram_oclass, &priv);
+	ret = nouveau_fb_create(parent, engine, oclass, &priv);
 	*pobject = nv_object(priv);
 	if (ret)
 		return ret;
 
-	priv->base.memtype_valid = nv04_fb_memtype_valid;
+	priv->base.tile.regions = impl->tile.regions;
+	priv->base.tile.init = impl->tile.init;
+	priv->base.tile.comp = impl->tile.comp;
+	priv->base.tile.fini = impl->tile.fini;
+	priv->base.tile.prog = impl->tile.prog;
 	return 0;
 }
 
-struct nouveau_oclass
-nv04_fb_oclass = {
-	.handle = NV_SUBDEV(FB, 0x04),
-	.ofuncs = &(struct nouveau_ofuncs) {
+struct nouveau_oclass *
+nv04_fb_oclass = &(struct nv04_fb_impl) {
+	.base.base.handle = NV_SUBDEV(FB, 0x04),
+	.base.base.ofuncs = &(struct nouveau_ofuncs) {
 		.ctor = nv04_fb_ctor,
 		.dtor = _nouveau_fb_dtor,
 		.init = nv04_fb_init,
 		.fini = _nouveau_fb_fini,
 	},
-};
+	.base.memtype = nv04_fb_memtype_valid,
+	.base.ram = &nv04_ram_oclass,
+}.base.base;
diff --git a/drivers/gpu/drm/nouveau/core/subdev/fb/nv04.h b/drivers/gpu/drm/nouveau/core/subdev/fb/nv04.h
new file mode 100644
index 000000000000..06ce71f87a74
--- /dev/null
+++ b/drivers/gpu/drm/nouveau/core/subdev/fb/nv04.h
@@ -0,0 +1,55 @@
+#ifndef __NVKM_FB_NV04_H__
+#define __NVKM_FB_NV04_H__
+
+#include "priv.h"
+
+struct nv04_fb_priv {
+	struct nouveau_fb base;
+};
+
+int  nv04_fb_ctor(struct nouveau_object *, struct nouveau_object *,
+		  struct nouveau_oclass *, void *, u32,
+		  struct nouveau_object **);
+
+struct nv04_fb_impl {
+	struct nouveau_fb_impl base;
+	struct {
+		int regions;
+		void (*init)(struct nouveau_fb *, int i, u32 addr, u32 size,
+			     u32 pitch, u32 flags, struct nouveau_fb_tile *);
+		void (*comp)(struct nouveau_fb *, int i, u32 size, u32 flags,
+			     struct nouveau_fb_tile *);
+		void (*fini)(struct nouveau_fb *, int i,
+			     struct nouveau_fb_tile *);
+		void (*prog)(struct nouveau_fb *, int i,
+			     struct nouveau_fb_tile *);
+	} tile;
+};
+
+void nv10_fb_tile_init(struct nouveau_fb *, int i, u32 addr, u32 size,
+		       u32 pitch, u32 flags, struct nouveau_fb_tile *);
+void nv10_fb_tile_fini(struct nouveau_fb *, int i, struct nouveau_fb_tile *);
+void nv10_fb_tile_prog(struct nouveau_fb *, int, struct nouveau_fb_tile *);
+
+void nv20_fb_tile_init(struct nouveau_fb *, int i, u32 addr, u32 size,
+		       u32 pitch, u32 flags, struct nouveau_fb_tile *);
+void nv20_fb_tile_fini(struct nouveau_fb *, int i, struct nouveau_fb_tile *);
+void nv20_fb_tile_prog(struct nouveau_fb *, int, struct nouveau_fb_tile *);
+
+int  nv30_fb_init(struct nouveau_object *);
+void nv30_fb_tile_init(struct nouveau_fb *, int i, u32 addr, u32 size,
+		       u32 pitch, u32 flags, struct nouveau_fb_tile *);
+
+void nv40_fb_tile_comp(struct nouveau_fb *, int i, u32 size, u32 flags,
+		       struct nouveau_fb_tile *);
+
+int  nv41_fb_init(struct nouveau_object *);
+void nv41_fb_tile_prog(struct nouveau_fb *, int, struct nouveau_fb_tile *);
+
+int  nv44_fb_init(struct nouveau_object *);
+void nv44_fb_tile_prog(struct nouveau_fb *, int, struct nouveau_fb_tile *);
+
+void nv46_fb_tile_init(struct nouveau_fb *, int i, u32 addr, u32 size,
+		       u32 pitch, u32 flags, struct nouveau_fb_tile *);
+
+#endif
diff --git a/drivers/gpu/drm/nouveau/core/subdev/fb/nv10.c b/drivers/gpu/drm/nouveau/core/subdev/fb/nv10.c
index be069b5306b6..ffb7ec6d97aa 100644
--- a/drivers/gpu/drm/nouveau/core/subdev/fb/nv10.c
+++ b/drivers/gpu/drm/nouveau/core/subdev/fb/nv10.c
@@ -24,11 +24,7 @@
  *
  */
 
-#include "priv.h"
-
-struct nv10_fb_priv {
-	struct nouveau_fb base;
-};
+#include "nv04.h"
 
 void
 nv10_fb_tile_init(struct nouveau_fb *pfb, int i, u32 addr, u32 size, u32 pitch,
@@ -57,34 +53,19 @@ nv10_fb_tile_prog(struct nouveau_fb *pfb, int i, struct nouveau_fb_tile *tile)
 	nv_rd32(pfb, 0x100240 + (i * 0x10));
 }
 
-static int
-nv10_fb_ctor(struct nouveau_object *parent, struct nouveau_object *engine,
-	     struct nouveau_oclass *oclass, void *data, u32 size,
-	     struct nouveau_object **pobject)
-{
-	struct nv10_fb_priv *priv;
-	int ret;
-
-	ret = nouveau_fb_create(parent, engine, oclass, &nv10_ram_oclass, &priv);
-	*pobject = nv_object(priv);
-	if (ret)
-		return ret;
-
-	priv->base.memtype_valid = nv04_fb_memtype_valid;
-	priv->base.tile.regions = 8;
-	priv->base.tile.init = nv10_fb_tile_init;
-	priv->base.tile.fini = nv10_fb_tile_fini;
-	priv->base.tile.prog = nv10_fb_tile_prog;
-	return 0;
-}
-
-struct nouveau_oclass
-nv10_fb_oclass = {
-	.handle = NV_SUBDEV(FB, 0x10),
-	.ofuncs = &(struct nouveau_ofuncs) {
-		.ctor = nv10_fb_ctor,
+struct nouveau_oclass *
+nv10_fb_oclass = &(struct nv04_fb_impl) {
+	.base.base.handle = NV_SUBDEV(FB, 0x10),
+	.base.base.ofuncs = &(struct nouveau_ofuncs) {
+		.ctor = nv04_fb_ctor,
 		.dtor = _nouveau_fb_dtor,
 		.init = _nouveau_fb_init,
 		.fini = _nouveau_fb_fini,
 	},
-};
+	.base.memtype = nv04_fb_memtype_valid,
+	.base.ram = &nv10_ram_oclass,
+	.tile.regions = 8,
+	.tile.init = nv10_fb_tile_init,
+	.tile.fini = nv10_fb_tile_fini,
+	.tile.prog = nv10_fb_tile_prog,
+}.base.base;
diff --git a/drivers/gpu/drm/nouveau/core/subdev/fb/nv1a.c b/drivers/gpu/drm/nouveau/core/subdev/fb/nv1a.c
index 57a2af0079b3..9159a5ccee93 100644
--- a/drivers/gpu/drm/nouveau/core/subdev/fb/nv1a.c
+++ b/drivers/gpu/drm/nouveau/core/subdev/fb/nv1a.c
@@ -24,40 +24,21 @@
  *
  */
 
-#include "priv.h"
+#include "nv04.h"
 
-struct nv1a_fb_priv {
-	struct nouveau_fb base;
-};
-
-static int
-nv1a_fb_ctor(struct nouveau_object *parent, struct nouveau_object *engine,
-	     struct nouveau_oclass *oclass, void *data, u32 size,
-	     struct nouveau_object **pobject)
-{
-	struct nv1a_fb_priv *priv;
-	int ret;
-
-	ret = nouveau_fb_create(parent, engine, oclass, &nv1a_ram_oclass, &priv);
-	*pobject = nv_object(priv);
-	if (ret)
-		return ret;
-
-	priv->base.memtype_valid = nv04_fb_memtype_valid;
-	priv->base.tile.regions = 8;
-	priv->base.tile.init = nv10_fb_tile_init;
-	priv->base.tile.fini = nv10_fb_tile_fini;
-	priv->base.tile.prog = nv10_fb_tile_prog;
-	return 0;
-}
-
-struct nouveau_oclass
-nv1a_fb_oclass = {
-	.handle = NV_SUBDEV(FB, 0x1a),
-	.ofuncs = &(struct nouveau_ofuncs) {
-		.ctor = nv1a_fb_ctor,
+struct nouveau_oclass *
+nv1a_fb_oclass = &(struct nv04_fb_impl) {
+	.base.base.handle = NV_SUBDEV(FB, 0x1a),
+	.base.base.ofuncs = &(struct nouveau_ofuncs) {
+		.ctor = nv04_fb_ctor,
 		.dtor = _nouveau_fb_dtor,
 		.init = _nouveau_fb_init,
 		.fini = _nouveau_fb_fini,
 	},
-};
+	.base.memtype = nv04_fb_memtype_valid,
+	.base.ram = &nv10_ram_oclass,
+	.tile.regions = 8,
+	.tile.init = nv10_fb_tile_init,
+	.tile.fini = nv10_fb_tile_fini,
+	.tile.prog = nv10_fb_tile_prog,
+}.base.base;
diff --git a/drivers/gpu/drm/nouveau/core/subdev/fb/nv20.c b/drivers/gpu/drm/nouveau/core/subdev/fb/nv20.c
index b18c4e63bb47..f003c1b1893f 100644
--- a/drivers/gpu/drm/nouveau/core/subdev/fb/nv20.c
+++ b/drivers/gpu/drm/nouveau/core/subdev/fb/nv20.c
@@ -24,11 +24,7 @@
  *
  */
 
-#include "priv.h"
-
-struct nv20_fb_priv {
-	struct nouveau_fb base;
-};
+#include "nv04.h"
 
 void
 nv20_fb_tile_init(struct nouveau_fb *pfb, int i, u32 addr, u32 size, u32 pitch,
@@ -80,35 +76,20 @@ nv20_fb_tile_prog(struct nouveau_fb *pfb, int i, struct nouveau_fb_tile *tile)
 	nv_wr32(pfb, 0x100300 + (i * 0x04), tile->zcomp);
 }
 
-static int
-nv20_fb_ctor(struct nouveau_object *parent, struct nouveau_object *engine,
-	     struct nouveau_oclass *oclass, void *data, u32 size,
-	     struct nouveau_object **pobject)
-{
-	struct nv20_fb_priv *priv;
-	int ret;
-
-	ret = nouveau_fb_create(parent, engine, oclass, &nv20_ram_oclass, &priv);
-	*pobject = nv_object(priv);
-	if (ret)
-		return ret;
-
-	priv->base.memtype_valid = nv04_fb_memtype_valid;
-	priv->base.tile.regions = 8;
-	priv->base.tile.init = nv20_fb_tile_init;
-	priv->base.tile.comp = nv20_fb_tile_comp;
-	priv->base.tile.fini = nv20_fb_tile_fini;
-	priv->base.tile.prog = nv20_fb_tile_prog;
-	return 0;
-}
-
-struct nouveau_oclass
-nv20_fb_oclass = {
-	.handle = NV_SUBDEV(FB, 0x20),
-	.ofuncs = &(struct nouveau_ofuncs) {
-		.ctor = nv20_fb_ctor,
+struct nouveau_oclass *
+nv20_fb_oclass = &(struct nv04_fb_impl) {
+	.base.base.handle = NV_SUBDEV(FB, 0x20),
+	.base.base.ofuncs = &(struct nouveau_ofuncs) {
+		.ctor = nv04_fb_ctor,
 		.dtor = _nouveau_fb_dtor,
 		.init = _nouveau_fb_init,
 		.fini = _nouveau_fb_fini,
 	},
-};
+	.base.memtype = nv04_fb_memtype_valid,
+	.base.ram = &nv20_ram_oclass,
+	.tile.regions = 8,
+	.tile.init = nv20_fb_tile_init,
+	.tile.comp = nv20_fb_tile_comp,
+	.tile.fini = nv20_fb_tile_fini,
+	.tile.prog = nv20_fb_tile_prog,
+}.base.base;
diff --git a/drivers/gpu/drm/nouveau/core/subdev/fb/nv25.c b/drivers/gpu/drm/nouveau/core/subdev/fb/nv25.c
index 32ccabf10c45..f34f4223210b 100644
--- a/drivers/gpu/drm/nouveau/core/subdev/fb/nv25.c
+++ b/drivers/gpu/drm/nouveau/core/subdev/fb/nv25.c
@@ -24,11 +24,7 @@
  *
  */
 
-#include "priv.h"
-
-struct nv25_fb_priv {
-	struct nouveau_fb base;
-};
+#include "nv04.h"
 
 static void
 nv25_fb_tile_comp(struct nouveau_fb *pfb, int i, u32 size, u32 flags,
@@ -46,35 +42,20 @@ nv25_fb_tile_comp(struct nouveau_fb *pfb, int i, u32 size, u32 flags,
 	}
 }
 
-static int
-nv25_fb_ctor(struct nouveau_object *parent, struct nouveau_object *engine,
-	     struct nouveau_oclass *oclass, void *data, u32 size,
-	     struct nouveau_object **pobject)
-{
-	struct nv25_fb_priv *priv;
-	int ret;
-
-	ret = nouveau_fb_create(parent, engine, oclass, &nv20_ram_oclass, &priv);
-	*pobject = nv_object(priv);
-	if (ret)
-		return ret;
-
-	priv->base.memtype_valid = nv04_fb_memtype_valid;
-	priv->base.tile.regions = 8;
-	priv->base.tile.init = nv20_fb_tile_init;
-	priv->base.tile.comp = nv25_fb_tile_comp;
-	priv->base.tile.fini = nv20_fb_tile_fini;
-	priv->base.tile.prog = nv20_fb_tile_prog;
-	return 0;
-}
-
-struct nouveau_oclass
-nv25_fb_oclass = {
-	.handle = NV_SUBDEV(FB, 0x25),
-	.ofuncs = &(struct nouveau_ofuncs) {
-		.ctor = nv25_fb_ctor,
+struct nouveau_oclass *
+nv25_fb_oclass = &(struct nv04_fb_impl) {
+	.base.base.handle = NV_SUBDEV(FB, 0x25),
+	.base.base.ofuncs = &(struct nouveau_ofuncs) {
+		.ctor = nv04_fb_ctor,
 		.dtor = _nouveau_fb_dtor,
 		.init = _nouveau_fb_init,
 		.fini = _nouveau_fb_fini,
 	},
-};
+	.base.memtype = nv04_fb_memtype_valid,
+	.base.ram = &nv20_ram_oclass,
+	.tile.regions = 8,
+	.tile.init = nv20_fb_tile_init,
+	.tile.comp = nv25_fb_tile_comp,
+	.tile.fini = nv20_fb_tile_fini,
+	.tile.prog = nv20_fb_tile_prog,
+}.base.base;
diff --git a/drivers/gpu/drm/nouveau/core/subdev/fb/nv30.c b/drivers/gpu/drm/nouveau/core/subdev/fb/nv30.c
index bef756d43d33..69093f7151f0 100644
--- a/drivers/gpu/drm/nouveau/core/subdev/fb/nv30.c
+++ b/drivers/gpu/drm/nouveau/core/subdev/fb/nv30.c
@@ -24,11 +24,7 @@
  *
  */
 
-#include "priv.h"
-
-struct nv30_fb_priv {
-	struct nouveau_fb base;
-};
+#include "nv04.h"
 
 void
 nv30_fb_tile_init(struct nouveau_fb *pfb, int i, u32 addr, u32 size, u32 pitch,
@@ -67,7 +63,7 @@ nv30_fb_tile_comp(struct nouveau_fb *pfb, int i, u32 size, u32 flags,
 }
 
 static int
-calc_bias(struct nv30_fb_priv *priv, int k, int i, int j)
+calc_bias(struct nv04_fb_priv *priv, int k, int i, int j)
 {
 	struct nouveau_device *device = nv_device(priv);
 	int b = (device->chipset > 0x30 ?
@@ -78,7 +74,7 @@ calc_bias(struct nv30_fb_priv *priv, int k, int i, int j)
 }
 
 static int
-calc_ref(struct nv30_fb_priv *priv, int l, int k, int i)
+calc_ref(struct nv04_fb_priv *priv, int l, int k, int i)
 {
 	int j, x = 0;
 
@@ -95,7 +91,7 @@ int
 nv30_fb_init(struct nouveau_object *object)
 {
 	struct nouveau_device *device = nv_device(object);
-	struct nv30_fb_priv *priv = (void *)object;
+	struct nv04_fb_priv *priv = (void *)object;
 	int ret, i, j;
 
 	ret = nouveau_fb_init(&priv->base);
@@ -124,35 +120,20 @@ nv30_fb_init(struct nouveau_object *object)
 	return 0;
 }
 
-static int
-nv30_fb_ctor(struct nouveau_object *parent, struct nouveau_object *engine,
-	     struct nouveau_oclass *oclass, void *data, u32 size,
-	     struct nouveau_object **pobject)
-{
-	struct nv30_fb_priv *priv;
-	int ret;
-
-	ret = nouveau_fb_create(parent, engine, oclass, &nv20_ram_oclass, &priv);
-	*pobject = nv_object(priv);
-	if (ret)
-		return ret;
-
-	priv->base.memtype_valid = nv04_fb_memtype_valid;
-	priv->base.tile.regions = 8;
-	priv->base.tile.init = nv30_fb_tile_init;
-	priv->base.tile.comp = nv30_fb_tile_comp;
-	priv->base.tile.fini = nv20_fb_tile_fini;
-	priv->base.tile.prog = nv20_fb_tile_prog;
-	return 0;
-}
-
-struct nouveau_oclass
-nv30_fb_oclass = {
-	.handle = NV_SUBDEV(FB, 0x30),
-	.ofuncs = &(struct nouveau_ofuncs) {
-		.ctor = nv30_fb_ctor,
+struct nouveau_oclass *
+nv30_fb_oclass = &(struct nv04_fb_impl) {
+	.base.base.handle = NV_SUBDEV(FB, 0x30),
+	.base.base.ofuncs = &(struct nouveau_ofuncs) {
+		.ctor = nv04_fb_ctor,
 		.dtor = _nouveau_fb_dtor,
 		.init = nv30_fb_init,
 		.fini = _nouveau_fb_fini,
 	},
-};
+	.base.memtype = nv04_fb_memtype_valid,
+	.base.ram = &nv20_ram_oclass,
+	.tile.regions = 8,
+	.tile.init = nv30_fb_tile_init,
+	.tile.comp = nv30_fb_tile_comp,
+	.tile.fini = nv20_fb_tile_fini,
+	.tile.prog = nv20_fb_tile_prog,
+}.base.base;
diff --git a/drivers/gpu/drm/nouveau/core/subdev/fb/nv35.c b/drivers/gpu/drm/nouveau/core/subdev/fb/nv35.c
index 097d8e3824f2..161b06e8fc3f 100644
--- a/drivers/gpu/drm/nouveau/core/subdev/fb/nv35.c
+++ b/drivers/gpu/drm/nouveau/core/subdev/fb/nv35.c
@@ -24,11 +24,7 @@
  *
  */
 
-#include "priv.h"
-
-struct nv35_fb_priv {
-	struct nouveau_fb base;
-};
+#include "nv04.h"
 
 static void
 nv35_fb_tile_comp(struct nouveau_fb *pfb, int i, u32 size, u32 flags,
@@ -47,35 +43,20 @@ nv35_fb_tile_comp(struct nouveau_fb *pfb, int i, u32 size, u32 flags,
 	}
 }
 
-static int
-nv35_fb_ctor(struct nouveau_object *parent, struct nouveau_object *engine,
-	     struct nouveau_oclass *oclass, void *data, u32 size,
-	     struct nouveau_object **pobject)
-{
-	struct nv35_fb_priv *priv;
-	int ret;
-
-	ret = nouveau_fb_create(parent, engine, oclass, &nv20_ram_oclass, &priv);
-	*pobject = nv_object(priv);
-	if (ret)
-		return ret;
-
-	priv->base.memtype_valid = nv04_fb_memtype_valid;
-	priv->base.tile.regions = 8;
-	priv->base.tile.init = nv30_fb_tile_init;
-	priv->base.tile.comp = nv35_fb_tile_comp;
-	priv->base.tile.fini = nv20_fb_tile_fini;
-	priv->base.tile.prog = nv20_fb_tile_prog;
-	return 0;
-}
-
-struct nouveau_oclass
-nv35_fb_oclass = {
-	.handle = NV_SUBDEV(FB, 0x35),
-	.ofuncs = &(struct nouveau_ofuncs) {
-		.ctor = nv35_fb_ctor,
+struct nouveau_oclass *
+nv35_fb_oclass = &(struct nv04_fb_impl) {
+	.base.base.handle = NV_SUBDEV(FB, 0x35),
+	.base.base.ofuncs = &(struct nouveau_ofuncs) {
+		.ctor = nv04_fb_ctor,
 		.dtor = _nouveau_fb_dtor,
 		.init = nv30_fb_init,
 		.fini = _nouveau_fb_fini,
 	},
-};
+	.base.memtype = nv04_fb_memtype_valid,
+	.base.ram = &nv20_ram_oclass,
+	.tile.regions = 8,
+	.tile.init = nv30_fb_tile_init,
+	.tile.comp = nv35_fb_tile_comp,
+	.tile.fini = nv20_fb_tile_fini,
+	.tile.prog = nv20_fb_tile_prog,
+}.base.base;
diff --git a/drivers/gpu/drm/nouveau/core/subdev/fb/nv36.c b/drivers/gpu/drm/nouveau/core/subdev/fb/nv36.c
index 9d6d9df896d9..2dd3d0aab6bb 100644
--- a/drivers/gpu/drm/nouveau/core/subdev/fb/nv36.c
+++ b/drivers/gpu/drm/nouveau/core/subdev/fb/nv36.c
@@ -24,11 +24,7 @@
  *
  */
 
-#include "priv.h"
-
-struct nv36_fb_priv {
-	struct nouveau_fb base;
-};
+#include "nv04.h"
 
 static void
 nv36_fb_tile_comp(struct nouveau_fb *pfb, int i, u32 size, u32 flags,
@@ -47,35 +43,20 @@ nv36_fb_tile_comp(struct nouveau_fb *pfb, int i, u32 size, u32 flags,
 	}
 }
 
-static int
-nv36_fb_ctor(struct nouveau_object *parent, struct nouveau_object *engine,
-	     struct nouveau_oclass *oclass, void *data, u32 size,
-	     struct nouveau_object **pobject)
-{
-	struct nv36_fb_priv *priv;
-	int ret;
-
-	ret = nouveau_fb_create(parent, engine, oclass, &nv20_ram_oclass, &priv);
-	*pobject = nv_object(priv);
-	if (ret)
-		return ret;
-
-	priv->base.memtype_valid = nv04_fb_memtype_valid;
-	priv->base.tile.regions = 8;
-	priv->base.tile.init = nv30_fb_tile_init;
-	priv->base.tile.comp = nv36_fb_tile_comp;
-	priv->base.tile.fini = nv20_fb_tile_fini;
-	priv->base.tile.prog = nv20_fb_tile_prog;
-	return 0;
-}
-
-struct nouveau_oclass
-nv36_fb_oclass = {
-	.handle = NV_SUBDEV(FB, 0x36),
-	.ofuncs = &(struct nouveau_ofuncs) {
-		.ctor = nv36_fb_ctor,
+struct nouveau_oclass *
+nv36_fb_oclass = &(struct nv04_fb_impl) {
+	.base.base.handle = NV_SUBDEV(FB, 0x36),
+	.base.base.ofuncs = &(struct nouveau_ofuncs) {
+		.ctor = nv04_fb_ctor,
 		.dtor = _nouveau_fb_dtor,
 		.init = nv30_fb_init,
 		.fini = _nouveau_fb_fini,
 	},
-};
+	.base.memtype = nv04_fb_memtype_valid,
+	.base.ram = &nv20_ram_oclass,
+	.tile.regions = 8,
+	.tile.init = nv30_fb_tile_init,
+	.tile.comp = nv36_fb_tile_comp,
+	.tile.fini = nv20_fb_tile_fini,
+	.tile.prog = nv20_fb_tile_prog,
+}.base.base;
diff --git a/drivers/gpu/drm/nouveau/core/subdev/fb/nv40.c b/drivers/gpu/drm/nouveau/core/subdev/fb/nv40.c
index 33b4393a7829..95a115ab0c86 100644
--- a/drivers/gpu/drm/nouveau/core/subdev/fb/nv40.c
+++ b/drivers/gpu/drm/nouveau/core/subdev/fb/nv40.c
@@ -24,11 +24,7 @@
  *
  */
 
-#include "priv.h"
-
-struct nv40_fb_priv {
-	struct nouveau_fb base;
-};
+#include "nv04.h"
 
 void
 nv40_fb_tile_comp(struct nouveau_fb *pfb, int i, u32 size, u32 flags,
@@ -50,7 +46,7 @@ nv40_fb_tile_comp(struct nouveau_fb *pfb, int i, u32 size, u32 flags,
 static int
 nv40_fb_init(struct nouveau_object *object)
 {
-	struct nv40_fb_priv *priv = (void *)object;
+	struct nv04_fb_priv *priv = (void *)object;
 	int ret;
 
 	ret = nouveau_fb_init(&priv->base);
@@ -61,36 +57,20 @@ nv40_fb_init(struct nouveau_object *object)
 	return 0;
 }
 
-static int
-nv40_fb_ctor(struct nouveau_object *parent, struct nouveau_object *engine,
-	     struct nouveau_oclass *oclass, void *data, u32 size,
-	     struct nouveau_object **pobject)
-{
-	struct nv40_fb_priv *priv;
-	int ret;
-
-	ret = nouveau_fb_create(parent, engine, oclass, &nv40_ram_oclass, &priv);
-	*pobject = nv_object(priv);
-	if (ret)
-		return ret;
-
-	priv->base.memtype_valid = nv04_fb_memtype_valid;
-	priv->base.tile.regions = 8;
-	priv->base.tile.init = nv30_fb_tile_init;
-	priv->base.tile.comp = nv40_fb_tile_comp;
-	priv->base.tile.fini = nv20_fb_tile_fini;
-	priv->base.tile.prog = nv20_fb_tile_prog;
-	return 0;
-}
-
-
-struct nouveau_oclass
-nv40_fb_oclass = {
-	.handle = NV_SUBDEV(FB, 0x40),
-	.ofuncs = &(struct nouveau_ofuncs) {
-		.ctor = nv40_fb_ctor,
+struct nouveau_oclass *
+nv40_fb_oclass = &(struct nv04_fb_impl) {
+	.base.base.handle = NV_SUBDEV(FB, 0x40),
+	.base.base.ofuncs = &(struct nouveau_ofuncs) {
+		.ctor = nv04_fb_ctor,
 		.dtor = _nouveau_fb_dtor,
 		.init = nv40_fb_init,
 		.fini = _nouveau_fb_fini,
 	},
-};
+	.base.memtype = nv04_fb_memtype_valid,
+	.base.ram = &nv40_ram_oclass,
+	.tile.regions = 8,
+	.tile.init = nv30_fb_tile_init,
+	.tile.comp = nv40_fb_tile_comp,
+	.tile.fini = nv20_fb_tile_fini,
+	.tile.prog = nv20_fb_tile_prog,
+}.base.base;
diff --git a/drivers/gpu/drm/nouveau/core/subdev/fb/nv40.h b/drivers/gpu/drm/nouveau/core/subdev/fb/nv40.h
new file mode 100644
index 000000000000..581f808527f2
--- /dev/null
+++ b/drivers/gpu/drm/nouveau/core/subdev/fb/nv40.h
@@ -0,0 +1,17 @@
+#ifndef __NVKM_FB_NV40_H__
+#define __NVKM_FB_NV40_H__
+
+#include "priv.h"
+
+struct nv40_ram {
+	struct nouveau_ram base;
+	u32 ctrl;
+	u32 coef;
+};
+
+
+int  nv40_ram_calc(struct nouveau_fb *, u32);
+int  nv40_ram_prog(struct nouveau_fb *);
+void nv40_ram_tidy(struct nouveau_fb *);
+
+#endif
diff --git a/drivers/gpu/drm/nouveau/core/subdev/fb/nv41.c b/drivers/gpu/drm/nouveau/core/subdev/fb/nv41.c
index 02cd83789cd4..b239a8615599 100644
--- a/drivers/gpu/drm/nouveau/core/subdev/fb/nv41.c
+++ b/drivers/gpu/drm/nouveau/core/subdev/fb/nv41.c
@@ -24,11 +24,7 @@
  *
  */
 
-#include "priv.h"
-
-struct nv41_fb_priv {
-	struct nouveau_fb base;
-};
+#include "nv04.h"
 
 void
 nv41_fb_tile_prog(struct nouveau_fb *pfb, int i, struct nouveau_fb_tile *tile)
@@ -43,7 +39,7 @@ nv41_fb_tile_prog(struct nouveau_fb *pfb, int i, struct nouveau_fb_tile *tile)
 int
 nv41_fb_init(struct nouveau_object *object)
 {
-	struct nv41_fb_priv *priv = (void *)object;
+	struct nv04_fb_priv *priv = (void *)object;
 	int ret;
 
 	ret = nouveau_fb_init(&priv->base);
@@ -54,36 +50,20 @@ nv41_fb_init(struct nouveau_object *object)
 	return 0;
 }
 
-static int
-nv41_fb_ctor(struct nouveau_object *parent, struct nouveau_object *engine,
-	     struct nouveau_oclass *oclass, void *data, u32 size,
-	     struct nouveau_object **pobject)
-{
-	struct nv41_fb_priv *priv;
-	int ret;
-
-	ret = nouveau_fb_create(parent, engine, oclass, &nv41_ram_oclass, &priv);
-	*pobject = nv_object(priv);
-	if (ret)
-		return ret;
-
-	priv->base.memtype_valid = nv04_fb_memtype_valid;
-	priv->base.tile.regions = 12;
-	priv->base.tile.init = nv30_fb_tile_init;
-	priv->base.tile.comp = nv40_fb_tile_comp;
-	priv->base.tile.fini = nv20_fb_tile_fini;
-	priv->base.tile.prog = nv41_fb_tile_prog;
-	return 0;
-}
-
-
-struct nouveau_oclass
-nv41_fb_oclass = {
-	.handle = NV_SUBDEV(FB, 0x41),
-	.ofuncs = &(struct nouveau_ofuncs) {
-		.ctor = nv41_fb_ctor,
+struct nouveau_oclass *
+nv41_fb_oclass = &(struct nv04_fb_impl) {
+	.base.base.handle = NV_SUBDEV(FB, 0x41),
+	.base.base.ofuncs = &(struct nouveau_ofuncs) {
+		.ctor = nv04_fb_ctor,
 		.dtor = _nouveau_fb_dtor,
 		.init = nv41_fb_init,
 		.fini = _nouveau_fb_fini,
 	},
-};
+	.base.memtype = nv04_fb_memtype_valid,
+	.base.ram = &nv41_ram_oclass,
+	.tile.regions = 12,
+	.tile.init = nv30_fb_tile_init,
+	.tile.comp = nv40_fb_tile_comp,
+	.tile.fini = nv20_fb_tile_fini,
+	.tile.prog = nv41_fb_tile_prog,
+}.base.base;
diff --git a/drivers/gpu/drm/nouveau/core/subdev/fb/nv44.c b/drivers/gpu/drm/nouveau/core/subdev/fb/nv44.c
index c5246c29f293..d8478208a681 100644
--- a/drivers/gpu/drm/nouveau/core/subdev/fb/nv44.c
+++ b/drivers/gpu/drm/nouveau/core/subdev/fb/nv44.c
@@ -24,11 +24,7 @@
  *
  */
 
-#include "priv.h"
-
-struct nv44_fb_priv {
-	struct nouveau_fb base;
-};
+#include "nv04.h"
 
 static void
 nv44_fb_tile_init(struct nouveau_fb *pfb, int i, u32 addr, u32 size, u32 pitch,
@@ -52,7 +48,7 @@ nv44_fb_tile_prog(struct nouveau_fb *pfb, int i, struct nouveau_fb_tile *tile)
 int
 nv44_fb_init(struct nouveau_object *object)
 {
-	struct nv44_fb_priv *priv = (void *)object;
+	struct nv04_fb_priv *priv = (void *)object;
 	int ret;
 
 	ret = nouveau_fb_init(&priv->base);
@@ -64,35 +60,19 @@ nv44_fb_init(struct nouveau_object *object)
 	return 0;
 }
 
-static int
-nv44_fb_ctor(struct nouveau_object *parent, struct nouveau_object *engine,
-	     struct nouveau_oclass *oclass, void *data, u32 size,
-	     struct nouveau_object **pobject)
-{
-	struct nv44_fb_priv *priv;
-	int ret;
-
-	ret = nouveau_fb_create(parent, engine, oclass, &nv44_ram_oclass, &priv);
-	*pobject = nv_object(priv);
-	if (ret)
-		return ret;
-
-	priv->base.memtype_valid = nv04_fb_memtype_valid;
-	priv->base.tile.regions = 12;
-	priv->base.tile.init = nv44_fb_tile_init;
-	priv->base.tile.fini = nv20_fb_tile_fini;
-	priv->base.tile.prog = nv44_fb_tile_prog;
-	return 0;
-}
-
-
-struct nouveau_oclass
-nv44_fb_oclass = {
-	.handle = NV_SUBDEV(FB, 0x44),
-	.ofuncs = &(struct nouveau_ofuncs) {
-		.ctor = nv44_fb_ctor,
+struct nouveau_oclass *
+nv44_fb_oclass = &(struct nv04_fb_impl) {
+	.base.base.handle = NV_SUBDEV(FB, 0x44),
+	.base.base.ofuncs = &(struct nouveau_ofuncs) {
+		.ctor = nv04_fb_ctor,
 		.dtor = _nouveau_fb_dtor,
 		.init = nv44_fb_init,
 		.fini = _nouveau_fb_fini,
 	},
-};
+	.base.memtype = nv04_fb_memtype_valid,
+	.base.ram = &nv44_ram_oclass,
+	.tile.regions = 12,
+	.tile.init = nv44_fb_tile_init,
+	.tile.fini = nv20_fb_tile_fini,
+	.tile.prog = nv44_fb_tile_prog,
+}.base.base;
diff --git a/drivers/gpu/drm/nouveau/core/subdev/fb/nv46.c b/drivers/gpu/drm/nouveau/core/subdev/fb/nv46.c
index e2b57909bfca..a5b77514d35b 100644
--- a/drivers/gpu/drm/nouveau/core/subdev/fb/nv46.c
+++ b/drivers/gpu/drm/nouveau/core/subdev/fb/nv46.c
@@ -24,11 +24,7 @@
  *
  */
 
-#include "priv.h"
-
-struct nv46_fb_priv {
-	struct nouveau_fb base;
-};
+#include "nv04.h"
 
 void
 nv46_fb_tile_init(struct nouveau_fb *pfb, int i, u32 addr, u32 size, u32 pitch,
@@ -44,35 +40,19 @@ nv46_fb_tile_init(struct nouveau_fb *pfb, int i, u32 addr, u32 size, u32 pitch,
 	tile->pitch = pitch;
 }
 
-static int
-nv46_fb_ctor(struct nouveau_object *parent, struct nouveau_object *engine,
-	     struct nouveau_oclass *oclass, void *data, u32 size,
-	     struct nouveau_object **pobject)
-{
-	struct nv46_fb_priv *priv;
-	int ret;
-
-	ret = nouveau_fb_create(parent, engine, oclass, &nv44_ram_oclass, &priv);
-	*pobject = nv_object(priv);
-	if (ret)
-		return ret;
-
-	priv->base.memtype_valid = nv04_fb_memtype_valid;
-	priv->base.tile.regions = 15;
-	priv->base.tile.init = nv46_fb_tile_init;
-	priv->base.tile.fini = nv20_fb_tile_fini;
-	priv->base.tile.prog = nv44_fb_tile_prog;
-	return 0;
-}
-
-
-struct nouveau_oclass
-nv46_fb_oclass = {
-	.handle = NV_SUBDEV(FB, 0x46),
-	.ofuncs = &(struct nouveau_ofuncs) {
-		.ctor = nv46_fb_ctor,
+struct nouveau_oclass *
+nv46_fb_oclass = &(struct nv04_fb_impl) {
+	.base.base.handle = NV_SUBDEV(FB, 0x46),
+	.base.base.ofuncs = &(struct nouveau_ofuncs) {
+		.ctor = nv04_fb_ctor,
 		.dtor = _nouveau_fb_dtor,
 		.init = nv44_fb_init,
 		.fini = _nouveau_fb_fini,
 	},
-};
+	.base.memtype = nv04_fb_memtype_valid,
+	.base.ram = &nv44_ram_oclass,
+	.tile.regions = 15,
+	.tile.init = nv46_fb_tile_init,
+	.tile.fini = nv20_fb_tile_fini,
+	.tile.prog = nv44_fb_tile_prog,
+}.base.base;
diff --git a/drivers/gpu/drm/nouveau/core/subdev/fb/nv47.c b/drivers/gpu/drm/nouveau/core/subdev/fb/nv47.c
index fe6a2278621d..3bea142376bc 100644
--- a/drivers/gpu/drm/nouveau/core/subdev/fb/nv47.c
+++ b/drivers/gpu/drm/nouveau/core/subdev/fb/nv47.c
@@ -24,42 +24,22 @@
  *
  */
 
-#include "priv.h"
+#include "nv04.h"
 
-struct nv47_fb_priv {
-	struct nouveau_fb base;
-};
-
-static int
-nv47_fb_ctor(struct nouveau_object *parent, struct nouveau_object *engine,
-	     struct nouveau_oclass *oclass, void *data, u32 size,
-	     struct nouveau_object **pobject)
-{
-	struct nv47_fb_priv *priv;
-	int ret;
-
-	ret = nouveau_fb_create(parent, engine, oclass, &nv41_ram_oclass, &priv);
-	*pobject = nv_object(priv);
-	if (ret)
-		return ret;
-
-	priv->base.memtype_valid = nv04_fb_memtype_valid;
-	priv->base.tile.regions = 15;
-	priv->base.tile.init = nv30_fb_tile_init;
-	priv->base.tile.comp = nv40_fb_tile_comp;
-	priv->base.tile.fini = nv20_fb_tile_fini;
-	priv->base.tile.prog = nv41_fb_tile_prog;
-	return 0;
-}
-
-
-struct nouveau_oclass
-nv47_fb_oclass = {
-	.handle = NV_SUBDEV(FB, 0x47),
-	.ofuncs = &(struct nouveau_ofuncs) {
-		.ctor = nv47_fb_ctor,
+struct nouveau_oclass *
+nv47_fb_oclass = &(struct nv04_fb_impl) {
+	.base.base.handle = NV_SUBDEV(FB, 0x47),
+	.base.base.ofuncs = &(struct nouveau_ofuncs) {
+		.ctor = nv04_fb_ctor,
 		.dtor = _nouveau_fb_dtor,
 		.init = nv41_fb_init,
 		.fini = _nouveau_fb_fini,
 	},
-};
+	.base.memtype = nv04_fb_memtype_valid,
+	.base.ram = &nv41_ram_oclass,
+	.tile.regions = 15,
+	.tile.init = nv30_fb_tile_init,
+	.tile.comp = nv40_fb_tile_comp,
+	.tile.fini = nv20_fb_tile_fini,
+	.tile.prog = nv41_fb_tile_prog,
+}.base.base;
diff --git a/drivers/gpu/drm/nouveau/core/subdev/fb/nv49.c b/drivers/gpu/drm/nouveau/core/subdev/fb/nv49.c
index 5eca99b8c7e2..666cbd5d47f5 100644
--- a/drivers/gpu/drm/nouveau/core/subdev/fb/nv49.c
+++ b/drivers/gpu/drm/nouveau/core/subdev/fb/nv49.c
@@ -24,42 +24,22 @@
  *
  */
 
-#include "priv.h"
+#include "nv04.h"
 
-struct nv49_fb_priv {
-	struct nouveau_fb base;
-};
-
-static int
-nv49_fb_ctor(struct nouveau_object *parent, struct nouveau_object *engine,
-	     struct nouveau_oclass *oclass, void *data, u32 size,
-	     struct nouveau_object **pobject)
-{
-	struct nv49_fb_priv *priv;
-	int ret;
-
-	ret = nouveau_fb_create(parent, engine, oclass, &nv49_ram_oclass, &priv);
-	*pobject = nv_object(priv);
-	if (ret)
-		return ret;
-
-	priv->base.memtype_valid = nv04_fb_memtype_valid;
-	priv->base.tile.regions = 15;
-	priv->base.tile.init = nv30_fb_tile_init;
-	priv->base.tile.comp = nv40_fb_tile_comp;
-	priv->base.tile.fini = nv20_fb_tile_fini;
-	priv->base.tile.prog = nv41_fb_tile_prog;
-	return 0;
-}
-
-
-struct nouveau_oclass
-nv49_fb_oclass = {
-	.handle = NV_SUBDEV(FB, 0x49),
-	.ofuncs = &(struct nouveau_ofuncs) {
-		.ctor = nv49_fb_ctor,
+struct nouveau_oclass *
+nv49_fb_oclass = &(struct nv04_fb_impl) {
+	.base.base.handle = NV_SUBDEV(FB, 0x49),
+	.base.base.ofuncs = &(struct nouveau_ofuncs) {
+		.ctor = nv04_fb_ctor,
 		.dtor = _nouveau_fb_dtor,
 		.init = nv41_fb_init,
 		.fini = _nouveau_fb_fini,
 	},
-};
+	.base.memtype = nv04_fb_memtype_valid,
+	.base.ram = &nv49_ram_oclass,
+	.tile.regions = 15,
+	.tile.init = nv30_fb_tile_init,
+	.tile.comp = nv40_fb_tile_comp,
+	.tile.fini = nv20_fb_tile_fini,
+	.tile.prog = nv41_fb_tile_prog,
+}.base.base;
diff --git a/drivers/gpu/drm/nouveau/core/subdev/fb/nv4e.c b/drivers/gpu/drm/nouveau/core/subdev/fb/nv4e.c
index 1190b78a1e91..42e64f364ec1 100644
--- a/drivers/gpu/drm/nouveau/core/subdev/fb/nv4e.c
+++ b/drivers/gpu/drm/nouveau/core/subdev/fb/nv4e.c
@@ -24,40 +24,21 @@
  *
  */
 
-#include "priv.h"
+#include "nv04.h"
 
-struct nv4e_fb_priv {
-	struct nouveau_fb base;
-};
-
-static int
-nv4e_fb_ctor(struct nouveau_object *parent, struct nouveau_object *engine,
-	     struct nouveau_oclass *oclass, void *data, u32 size,
-	     struct nouveau_object **pobject)
-{
-	struct nv4e_fb_priv *priv;
-	int ret;
-
-	ret = nouveau_fb_create(parent, engine, oclass, &nv4e_ram_oclass, &priv);
-	*pobject = nv_object(priv);
-	if (ret)
-		return ret;
-
-	priv->base.memtype_valid = nv04_fb_memtype_valid;
-	priv->base.tile.regions = 12;
-	priv->base.tile.init = nv46_fb_tile_init;
-	priv->base.tile.fini = nv20_fb_tile_fini;
-	priv->base.tile.prog = nv44_fb_tile_prog;
-	return 0;
-}
-
-struct nouveau_oclass
-nv4e_fb_oclass = {
-	.handle = NV_SUBDEV(FB, 0x4e),
-	.ofuncs = &(struct nouveau_ofuncs) {
-		.ctor = nv4e_fb_ctor,
+struct nouveau_oclass *
+nv4e_fb_oclass = &(struct nv04_fb_impl) {
+	.base.base.handle = NV_SUBDEV(FB, 0x4e),
+	.base.base.ofuncs = &(struct nouveau_ofuncs) {
+		.ctor = nv04_fb_ctor,
 		.dtor = _nouveau_fb_dtor,
 		.init = nv44_fb_init,
 		.fini = _nouveau_fb_fini,
 	},
-};
+	.base.memtype = nv04_fb_memtype_valid,
+	.base.ram = &nv4e_ram_oclass,
+	.tile.regions = 12,
+	.tile.init = nv46_fb_tile_init,
+	.tile.fini = nv20_fb_tile_fini,
+	.tile.prog = nv44_fb_tile_prog,
+}.base.base;
diff --git a/drivers/gpu/drm/nouveau/core/subdev/fb/nv50.c b/drivers/gpu/drm/nouveau/core/subdev/fb/nv50.c
index da614ec5564b..cbc7f00c1278 100644
--- a/drivers/gpu/drm/nouveau/core/subdev/fb/nv50.c
+++ b/drivers/gpu/drm/nouveau/core/subdev/fb/nv50.c
@@ -27,14 +27,9 @@
 #include <core/engctx.h>
 #include <core/object.h>
 
-#include "priv.h"
 #include <subdev/bios.h>
 
-struct nv50_fb_priv {
-	struct nouveau_fb base;
-	struct page *r100c08_page;
-	dma_addr_t r100c08;
-};
+#include "nv50.h"
 
 int
 nv50_fb_memtype[0x80] = {
@@ -48,7 +43,7 @@ nv50_fb_memtype[0x80] = {
 	1, 0, 2, 0, 1, 0, 2, 0, 1, 1, 2, 2, 1, 1, 0, 0
 };
 
-static bool
+bool
 nv50_fb_memtype_valid(struct nouveau_fb *pfb, u32 memtype)
 {
 	return nv50_fb_memtype[(memtype & 0xff00) >> 8] != 0;
@@ -239,7 +234,7 @@ nv50_fb_intr(struct nouveau_subdev *subdev)
 		pr_cont("0x%08x\n", st1);
 }
 
-static int
+int
 nv50_fb_ctor(struct nouveau_object *parent, struct nouveau_object *engine,
 	     struct nouveau_oclass *oclass, void *data, u32 size,
 	     struct nouveau_object **pobject)
@@ -248,7 +243,7 @@ nv50_fb_ctor(struct nouveau_object *parent, struct nouveau_object *engine,
 	struct nv50_fb_priv *priv;
 	int ret;
 
-	ret = nouveau_fb_create(parent, engine, oclass, &nv50_ram_oclass, &priv);
+	ret = nouveau_fb_create(parent, engine, oclass, &priv);
 	*pobject = nv_object(priv);
 	if (ret)
 		return ret;
@@ -264,12 +259,11 @@ nv50_fb_ctor(struct nouveau_object *parent, struct nouveau_object *engine,
 		nv_warn(priv, "failed 0x100c08 page alloc\n");
 	}
 
-	priv->base.memtype_valid = nv50_fb_memtype_valid;
 	nv_subdev(priv)->intr = nv50_fb_intr;
 	return 0;
 }
 
-static void
+void
 nv50_fb_dtor(struct nouveau_object *object)
 {
 	struct nouveau_device *device = nv_device(object);
@@ -284,10 +278,10 @@ nv50_fb_dtor(struct nouveau_object *object)
 	nouveau_fb_destroy(&priv->base);
 }
 
-static int
+int
 nv50_fb_init(struct nouveau_object *object)
 {
-	struct nouveau_device *device = nv_device(object);
+	struct nv50_fb_impl *impl = (void *)object->oclass;
 	struct nv50_fb_priv *priv = (void *)object;
 	int ret;
 
@@ -303,33 +297,20 @@ nv50_fb_init(struct nouveau_object *object)
 
 	/* This is needed to get meaningful information from 100c90
 	 * on traps. No idea what these values mean exactly. */
-	switch (device->chipset) {
-	case 0x50:
-		nv_wr32(priv, 0x100c90, 0x000707ff);
-		break;
-	case 0xa3:
-	case 0xa5:
-	case 0xa8:
-		nv_wr32(priv, 0x100c90, 0x000d0fff);
-		break;
-	case 0xaf:
-		nv_wr32(priv, 0x100c90, 0x089d1fff);
-		break;
-	default:
-		nv_wr32(priv, 0x100c90, 0x001d07ff);
-		break;
-	}
-
+	nv_wr32(priv, 0x100c90, impl->trap);
 	return 0;
 }
 
-struct nouveau_oclass
-nv50_fb_oclass = {
-	.handle = NV_SUBDEV(FB, 0x50),
-	.ofuncs = &(struct nouveau_ofuncs) {
+struct nouveau_oclass *
+nv50_fb_oclass = &(struct nv50_fb_impl) {
+	.base.base.handle = NV_SUBDEV(FB, 0x50),
+	.base.base.ofuncs = &(struct nouveau_ofuncs) {
 		.ctor = nv50_fb_ctor,
 		.dtor = nv50_fb_dtor,
 		.init = nv50_fb_init,
 		.fini = _nouveau_fb_fini,
 	},
-};
+	.base.memtype = nv50_fb_memtype_valid,
+	.base.ram = &nv50_ram_oclass,
+	.trap = 0x000707ff,
+}.base.base;
diff --git a/drivers/gpu/drm/nouveau/core/subdev/fb/nv50.h b/drivers/gpu/drm/nouveau/core/subdev/fb/nv50.h
new file mode 100644
index 000000000000..c5e5a888c607
--- /dev/null
+++ b/drivers/gpu/drm/nouveau/core/subdev/fb/nv50.h
@@ -0,0 +1,33 @@
+#ifndef __NVKM_FB_NV50_H__
+#define __NVKM_FB_NV50_H__
+
+#include "priv.h"
+
+struct nv50_fb_priv {
+	struct nouveau_fb base;
+	struct page *r100c08_page;
+	dma_addr_t r100c08;
+};
+
+int  nv50_fb_ctor(struct nouveau_object *, struct nouveau_object *,
+		  struct nouveau_oclass *, void *, u32,
+		  struct nouveau_object **);
+void nv50_fb_dtor(struct nouveau_object *);
+int  nv50_fb_init(struct nouveau_object *);
+
+struct nv50_fb_impl {
+	struct nouveau_fb_impl base;
+	u32 trap;
+};
+
+#define nv50_ram_create(p,e,o,d)                                               \
+	nv50_ram_create_((p), (e), (o), sizeof(**d), (void **)d)
+int  nv50_ram_create_(struct nouveau_object *, struct nouveau_object *,
+		      struct nouveau_oclass *, int, void **);
+int  nv50_ram_get(struct nouveau_fb *, u64 size, u32 align, u32 ncmin,
+		  u32 memtype, struct nouveau_mem **);
+void nv50_ram_put(struct nouveau_fb *, struct nouveau_mem **);
+void __nv50_ram_put(struct nouveau_fb *, struct nouveau_mem *);
+extern int nv50_fb_memtype[0x80];
+
+#endif
diff --git a/drivers/gpu/drm/nouveau/core/subdev/fb/nv84.c b/drivers/gpu/drm/nouveau/core/subdev/fb/nv84.c
new file mode 100644
index 000000000000..cf0e767d3833
--- /dev/null
+++ b/drivers/gpu/drm/nouveau/core/subdev/fb/nv84.c
@@ -0,0 +1,39 @@
+/*
+ * Copyright 2012 Red Hat Inc.
+ *
+ * Permission is hereby granted, free of charge, to any person obtaining a
+ * copy of this software and associated documentation files (the "Software"),
+ * to deal in the Software without restriction, including without limitation
+ * the rights to use, copy, modify, merge, publish, distribute, sublicense,
+ * and/or sell copies of the Software, and to permit persons to whom the
+ * Software is furnished to do so, subject to the following conditions:
+ *
+ * The above copyright notice and this permission notice shall be included in
+ * all copies or substantial portions of the Software.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+ * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+ * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT.  IN NO EVENT SHALL
+ * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR
+ * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE,
+ * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR
+ * OTHER DEALINGS IN THE SOFTWARE.
+ *
+ * Authors: Ben Skeggs
+ */
+
+#include "nv50.h"
+
+struct nouveau_oclass *
+nv84_fb_oclass = &(struct nv50_fb_impl) {
+	.base.base.handle = NV_SUBDEV(FB, 0x84),
+	.base.base.ofuncs = &(struct nouveau_ofuncs) {
+		.ctor = nv50_fb_ctor,
+		.dtor = nv50_fb_dtor,
+		.init = nv50_fb_init,
+		.fini = _nouveau_fb_fini,
+	},
+	.base.memtype = nv50_fb_memtype_valid,
+	.base.ram = &nv50_ram_oclass,
+	.trap = 0x001d07ff,
+}.base.base;
diff --git a/drivers/gpu/drm/nouveau/core/subdev/fb/nva3.c b/drivers/gpu/drm/nouveau/core/subdev/fb/nva3.c
new file mode 100644
index 000000000000..dab6e1c63d48
--- /dev/null
+++ b/drivers/gpu/drm/nouveau/core/subdev/fb/nva3.c
@@ -0,0 +1,39 @@
+/*
+ * Copyright 2012 Red Hat Inc.
+ *
+ * Permission is hereby granted, free of charge, to any person obtaining a
+ * copy of this software and associated documentation files (the "Software"),
+ * to deal in the Software without restriction, including without limitation
+ * the rights to use, copy, modify, merge, publish, distribute, sublicense,
+ * and/or sell copies of the Software, and to permit persons to whom the
+ * Software is furnished to do so, subject to the following conditions:
+ *
+ * The above copyright notice and this permission notice shall be included in
+ * all copies or substantial portions of the Software.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+ * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+ * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT.  IN NO EVENT SHALL
+ * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR
+ * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE,
+ * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR
+ * OTHER DEALINGS IN THE SOFTWARE.
+ *
+ * Authors: Ben Skeggs
+ */
+
+#include "nv50.h"
+
+struct nouveau_oclass *
+nva3_fb_oclass = &(struct nv50_fb_impl) {
+	.base.base.handle = NV_SUBDEV(FB, 0xa3),
+	.base.base.ofuncs = &(struct nouveau_ofuncs) {
+		.ctor = nv50_fb_ctor,
+		.dtor = nv50_fb_dtor,
+		.init = nv50_fb_init,
+		.fini = _nouveau_fb_fini,
+	},
+	.base.memtype = nv50_fb_memtype_valid,
+	.base.ram = &nva3_ram_oclass,
+	.trap = 0x000d0fff,
+}.base.base;
diff --git a/drivers/gpu/drm/nouveau/core/subdev/fb/nvaa.c b/drivers/gpu/drm/nouveau/core/subdev/fb/nvaa.c
new file mode 100644
index 000000000000..cba8e6818035
--- /dev/null
+++ b/drivers/gpu/drm/nouveau/core/subdev/fb/nvaa.c
@@ -0,0 +1,39 @@
+/*
+ * Copyright 2012 Red Hat Inc.
+ *
+ * Permission is hereby granted, free of charge, to any person obtaining a
+ * copy of this software and associated documentation files (the "Software"),
+ * to deal in the Software without restriction, including without limitation
+ * the rights to use, copy, modify, merge, publish, distribute, sublicense,
+ * and/or sell copies of the Software, and to permit persons to whom the
+ * Software is furnished to do so, subject to the following conditions:
+ *
+ * The above copyright notice and this permission notice shall be included in
+ * all copies or substantial portions of the Software.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+ * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+ * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT.  IN NO EVENT SHALL
+ * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR
+ * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE,
+ * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR
+ * OTHER DEALINGS IN THE SOFTWARE.
+ *
+ * Authors: Ben Skeggs
+ */
+
+#include "nv50.h"
+
+struct nouveau_oclass *
+nvaa_fb_oclass = &(struct nv50_fb_impl) {
+	.base.base.handle = NV_SUBDEV(FB, 0xaa),
+	.base.base.ofuncs = &(struct nouveau_ofuncs) {
+		.ctor = nv50_fb_ctor,
+		.dtor = nv50_fb_dtor,
+		.init = nv50_fb_init,
+		.fini = _nouveau_fb_fini,
+	},
+	.base.memtype = nv50_fb_memtype_valid,
+	.base.ram = &nvaa_ram_oclass,
+	.trap = 0x001d07ff,
+}.base.base;
diff --git a/drivers/gpu/drm/nouveau/core/subdev/fb/nvaf.c b/drivers/gpu/drm/nouveau/core/subdev/fb/nvaf.c
new file mode 100644
index 000000000000..5423faa2c09b
--- /dev/null
+++ b/drivers/gpu/drm/nouveau/core/subdev/fb/nvaf.c
@@ -0,0 +1,39 @@
+/*
+ * Copyright 2012 Red Hat Inc.
+ *
+ * Permission is hereby granted, free of charge, to any person obtaining a
+ * copy of this software and associated documentation files (the "Software"),
+ * to deal in the Software without restriction, including without limitation
+ * the rights to use, copy, modify, merge, publish, distribute, sublicense,
+ * and/or sell copies of the Software, and to permit persons to whom the
+ * Software is furnished to do so, subject to the following conditions:
+ *
+ * The above copyright notice and this permission notice shall be included in
+ * all copies or substantial portions of the Software.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+ * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+ * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT.  IN NO EVENT SHALL
+ * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR
+ * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE,
+ * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR
+ * OTHER DEALINGS IN THE SOFTWARE.
+ *
+ * Authors: Ben Skeggs
+ */
+
+#include "nv50.h"
+
+struct nouveau_oclass *
+nvaf_fb_oclass = &(struct nv50_fb_impl) {
+	.base.base.handle = NV_SUBDEV(FB, 0xaf),
+	.base.base.ofuncs = &(struct nouveau_ofuncs) {
+		.ctor = nv50_fb_ctor,
+		.dtor = nv50_fb_dtor,
+		.init = nv50_fb_init,
+		.fini = _nouveau_fb_fini,
+	},
+	.base.memtype = nv50_fb_memtype_valid,
+	.base.ram = &nvaa_ram_oclass,
+	.trap = 0x089d1fff,
+}.base.base;
diff --git a/drivers/gpu/drm/nouveau/core/subdev/fb/nvc0.c b/drivers/gpu/drm/nouveau/core/subdev/fb/nvc0.c
index f35d76fd746d..e5fc37c4caac 100644
--- a/drivers/gpu/drm/nouveau/core/subdev/fb/nvc0.c
+++ b/drivers/gpu/drm/nouveau/core/subdev/fb/nvc0.c
@@ -22,24 +22,18 @@
  * Authors: Ben Skeggs
  */
 
-#include "priv.h"
-
-struct nvc0_fb_priv {
-	struct nouveau_fb base;
-	struct page *r100c10_page;
-	dma_addr_t r100c10;
-};
+#include "nvc0.h"
 
 extern const u8 nvc0_pte_storage_type_map[256];
 
-static bool
+bool
 nvc0_fb_memtype_valid(struct nouveau_fb *pfb, u32 tile_flags)
 {
 	u8 memtype = (tile_flags & 0x0000ff00) >> 8;
 	return likely((nvc0_pte_storage_type_map[memtype] != 0xff));
 }
 
-static int
+int
 nvc0_fb_init(struct nouveau_object *object)
 {
 	struct nvc0_fb_priv *priv = (void *)object;
@@ -54,7 +48,7 @@ nvc0_fb_init(struct nouveau_object *object)
 	return 0;
 }
 
-static void
+void
 nvc0_fb_dtor(struct nouveau_object *object)
 {
 	struct nouveau_device *device = nv_device(object);
@@ -69,7 +63,7 @@ nvc0_fb_dtor(struct nouveau_object *object)
 	nouveau_fb_destroy(&priv->base);
 }
 
-static int
+int
 nvc0_fb_ctor(struct nouveau_object *parent, struct nouveau_object *engine,
 	     struct nouveau_oclass *oclass, void *data, u32 size,
 	     struct nouveau_object **pobject)
@@ -78,13 +72,11 @@ nvc0_fb_ctor(struct nouveau_object *parent, struct nouveau_object *engine,
 	struct nvc0_fb_priv *priv;
 	int ret;
 
-	ret = nouveau_fb_create(parent, engine, oclass, &nvc0_ram_oclass, &priv);
+	ret = nouveau_fb_create(parent, engine, oclass, &priv);
 	*pobject = nv_object(priv);
 	if (ret)
 		return ret;
 
-	priv->base.memtype_valid = nvc0_fb_memtype_valid;
-
 	priv->r100c10_page = alloc_page(GFP_KERNEL | __GFP_ZERO);
 	if (priv->r100c10_page) {
 		priv->r100c10 = pci_map_page(device->pdev, priv->r100c10_page,
@@ -97,14 +89,15 @@ nvc0_fb_ctor(struct nouveau_object *parent, struct nouveau_object *engine,
 	return 0;
 }
 
-
-struct nouveau_oclass
-nvc0_fb_oclass = {
-	.handle = NV_SUBDEV(FB, 0xc0),
-	.ofuncs = &(struct nouveau_ofuncs) {
+struct nouveau_oclass *
+nvc0_fb_oclass = &(struct nouveau_fb_impl) {
+	.base.handle = NV_SUBDEV(FB, 0xc0),
+	.base.ofuncs = &(struct nouveau_ofuncs) {
 		.ctor = nvc0_fb_ctor,
 		.dtor = nvc0_fb_dtor,
 		.init = nvc0_fb_init,
 		.fini = _nouveau_fb_fini,
 	},
-};
+	.memtype = nvc0_fb_memtype_valid,
+	.ram = &nvc0_ram_oclass,
+}.base;
diff --git a/drivers/gpu/drm/nouveau/core/subdev/fb/nvc0.h b/drivers/gpu/drm/nouveau/core/subdev/fb/nvc0.h
new file mode 100644
index 000000000000..9e1931eb746f
--- /dev/null
+++ b/drivers/gpu/drm/nouveau/core/subdev/fb/nvc0.h
@@ -0,0 +1,29 @@
+#ifndef __NVKM_RAM_NVC0_H__
+#define __NVKM_RAM_NVC0_H__
+
+#include "priv.h"
+#include "nv50.h"
+
+struct nvc0_fb_priv {
+	struct nouveau_fb base;
+	struct page *r100c10_page;
+	dma_addr_t r100c10;
+};
+
+int  nvc0_fb_ctor(struct nouveau_object *, struct nouveau_object *,
+		  struct nouveau_oclass *, void *, u32,
+		  struct nouveau_object **);
+void nvc0_fb_dtor(struct nouveau_object *);
+int  nvc0_fb_init(struct nouveau_object *);
+bool nvc0_fb_memtype_valid(struct nouveau_fb *, u32);
+
+
+#define nvc0_ram_create(p,e,o,d)                                               \
+	nvc0_ram_create_((p), (e), (o), sizeof(**d), (void **)d)
+int  nvc0_ram_create_(struct nouveau_object *, struct nouveau_object *,
+		      struct nouveau_oclass *, int, void **);
+int  nvc0_ram_get(struct nouveau_fb *, u64, u32, u32, u32,
+		  struct nouveau_mem **);
+void nvc0_ram_put(struct nouveau_fb *, struct nouveau_mem **);
+
+#endif
diff --git a/drivers/gpu/drm/nouveau/core/subdev/fb/nve0.c b/drivers/gpu/drm/nouveau/core/subdev/fb/nve0.c
new file mode 100644
index 000000000000..595db50cfef3
--- /dev/null
+++ b/drivers/gpu/drm/nouveau/core/subdev/fb/nve0.c
@@ -0,0 +1,38 @@
+/*
+ * Copyright 2012 Red Hat Inc.
+ *
+ * Permission is hereby granted, free of charge, to any person obtaining a
+ * copy of this software and associated documentation files (the "Software"),
+ * to deal in the Software without restriction, including without limitation
+ * the rights to use, copy, modify, merge, publish, distribute, sublicense,
+ * and/or sell copies of the Software, and to permit persons to whom the
+ * Software is furnished to do so, subject to the following conditions:
+ *
+ * The above copyright notice and this permission notice shall be included in
+ * all copies or substantial portions of the Software.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+ * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+ * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT.  IN NO EVENT SHALL
+ * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR
+ * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE,
+ * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR
+ * OTHER DEALINGS IN THE SOFTWARE.
+ *
+ * Authors: Ben Skeggs
+ */
+
+#include "nvc0.h"
+
+struct nouveau_oclass *
+nve0_fb_oclass = &(struct nouveau_fb_impl) {
+	.base.handle = NV_SUBDEV(FB, 0xe0),
+	.base.ofuncs = &(struct nouveau_ofuncs) {
+		.ctor = nvc0_fb_ctor,
+		.dtor = nvc0_fb_dtor,
+		.init = nvc0_fb_init,
+		.fini = _nouveau_fb_fini,
+	},
+	.memtype = nvc0_fb_memtype_valid,
+	.ram = &nve0_ram_oclass,
+}.base;
diff --git a/drivers/gpu/drm/nouveau/core/subdev/fb/priv.h b/drivers/gpu/drm/nouveau/core/subdev/fb/priv.h
index db9d6ddde52c..493125214e88 100644
--- a/drivers/gpu/drm/nouveau/core/subdev/fb/priv.h
+++ b/drivers/gpu/drm/nouveau/core/subdev/fb/priv.h
@@ -12,6 +12,8 @@
 #define nouveau_ram_fini(p,s)                                                  \
 	nouveau_object_fini(&(p)->base, (s))
 
+#define nouveau_ram_create_(p,e,o,s,d)                                         \
+	nouveau_object_create_((p), (e), (o), 0, (s), (void **)d)
 #define _nouveau_ram_dtor nouveau_object_destroy
 #define _nouveau_ram_init nouveau_object_init
 #define _nouveau_ram_fini nouveau_object_fini
@@ -26,10 +28,16 @@ extern struct nouveau_oclass nv44_ram_oclass;
 extern struct nouveau_oclass nv49_ram_oclass;
 extern struct nouveau_oclass nv4e_ram_oclass;
 extern struct nouveau_oclass nv50_ram_oclass;
+extern struct nouveau_oclass nva3_ram_oclass;
+extern struct nouveau_oclass nvaa_ram_oclass;
 extern struct nouveau_oclass nvc0_ram_oclass;
+extern struct nouveau_oclass nve0_ram_oclass;
 
-#define nouveau_fb_create(p,e,c,r,d)                                           \
-	nouveau_fb_create_((p), (e), (c), (r), sizeof(**d), (void **)d)
+int nouveau_sddr3_calc(struct nouveau_ram *ram);
+int nouveau_gddr5_calc(struct nouveau_ram *ram);
+
+#define nouveau_fb_create(p,e,c,d)                                             \
+	nouveau_fb_create_((p), (e), (c), sizeof(**d), (void **)d)
 #define nouveau_fb_destroy(p) ({                                               \
 	struct nouveau_fb *pfb = (p);                                          \
 	_nouveau_fb_dtor(nv_object(pfb));                                      \
@@ -44,44 +52,21 @@ extern struct nouveau_oclass nvc0_ram_oclass;
 })
 
 int nouveau_fb_create_(struct nouveau_object *, struct nouveau_object *,
-		       struct nouveau_oclass *, struct nouveau_oclass *,
-		       int length, void **pobject);
+		       struct nouveau_oclass *, int, void **);
 void _nouveau_fb_dtor(struct nouveau_object *);
 int  _nouveau_fb_init(struct nouveau_object *);
 int  _nouveau_fb_fini(struct nouveau_object *, bool);
 
-struct nouveau_bios;
-int  nouveau_fb_bios_memtype(struct nouveau_bios *);
+struct nouveau_fb_impl {
+	struct nouveau_oclass base;
+	struct nouveau_oclass *ram;
+	bool (*memtype)(struct nouveau_fb *, u32);
+};
 
 bool nv04_fb_memtype_valid(struct nouveau_fb *, u32 memtype);
+bool nv50_fb_memtype_valid(struct nouveau_fb *, u32 memtype);
 
-void nv10_fb_tile_init(struct nouveau_fb *, int i, u32 addr, u32 size,
-		       u32 pitch, u32 flags, struct nouveau_fb_tile *);
-void nv10_fb_tile_fini(struct nouveau_fb *, int i, struct nouveau_fb_tile *);
-void nv10_fb_tile_prog(struct nouveau_fb *, int, struct nouveau_fb_tile *);
-
-void nv20_fb_tile_init(struct nouveau_fb *, int i, u32 addr, u32 size,
-		       u32 pitch, u32 flags, struct nouveau_fb_tile *);
-void nv20_fb_tile_fini(struct nouveau_fb *, int i, struct nouveau_fb_tile *);
-void nv20_fb_tile_prog(struct nouveau_fb *, int, struct nouveau_fb_tile *);
-
-int  nv30_fb_init(struct nouveau_object *);
-void nv30_fb_tile_init(struct nouveau_fb *, int i, u32 addr, u32 size,
-		       u32 pitch, u32 flags, struct nouveau_fb_tile *);
-
-void nv40_fb_tile_comp(struct nouveau_fb *, int i, u32 size, u32 flags,
-		       struct nouveau_fb_tile *);
-
-int  nv41_fb_init(struct nouveau_object *);
-void nv41_fb_tile_prog(struct nouveau_fb *, int, struct nouveau_fb_tile *);
-
-int  nv44_fb_init(struct nouveau_object *);
-void nv44_fb_tile_prog(struct nouveau_fb *, int, struct nouveau_fb_tile *);
-
-void nv46_fb_tile_init(struct nouveau_fb *, int i, u32 addr, u32 size,
-		       u32 pitch, u32 flags, struct nouveau_fb_tile *);
-
-void __nv50_ram_put(struct nouveau_fb *, struct nouveau_mem *);
-extern int nv50_fb_memtype[0x80];
+struct nouveau_bios;
+int  nouveau_fb_bios_memtype(struct nouveau_bios *);
 
 #endif
diff --git a/drivers/gpu/drm/nouveau/core/subdev/fb/ramfuc.h b/drivers/gpu/drm/nouveau/core/subdev/fb/ramfuc.h
new file mode 100644
index 000000000000..0f57fcfe0bbf
--- /dev/null
+++ b/drivers/gpu/drm/nouveau/core/subdev/fb/ramfuc.h
@@ -0,0 +1,118 @@
+#ifndef __NVKM_FBRAM_FUC_H__
+#define __NVKM_FBRAM_FUC_H__
+
+#include <subdev/pwr.h>
+
+struct ramfuc {
+	struct nouveau_memx *memx;
+	struct nouveau_fb *pfb;
+	int sequence;
+};
+
+struct ramfuc_reg {
+	int sequence;
+	bool force;
+	u32 addr[2];
+	u32 data;
+};
+
+static inline struct ramfuc_reg
+ramfuc_reg2(u32 addr1, u32 addr2)
+{
+	return (struct ramfuc_reg) {
+		.sequence = 0,
+		.addr = { addr1, addr2 },
+		.data = 0xdeadbeef,
+	};
+}
+
+static inline struct ramfuc_reg
+ramfuc_reg(u32 addr)
+{
+	return ramfuc_reg2(addr, addr);
+}
+
+static inline int
+ramfuc_init(struct ramfuc *ram, struct nouveau_fb *pfb)
+{
+	struct nouveau_pwr *ppwr = nouveau_pwr(pfb);
+	int ret;
+
+	ret = nouveau_memx_init(ppwr, &ram->memx);
+	if (ret)
+		return ret;
+
+	ram->sequence++;
+	ram->pfb = pfb;
+	return 0;
+}
+
+static inline int
+ramfuc_exec(struct ramfuc *ram, bool exec)
+{
+	int ret = 0;
+	if (ram->pfb) {
+		ret = nouveau_memx_fini(&ram->memx, exec);
+		ram->pfb = NULL;
+	}
+	return ret;
+}
+
+static inline u32
+ramfuc_rd32(struct ramfuc *ram, struct ramfuc_reg *reg)
+{
+	if (reg->sequence != ram->sequence)
+		reg->data = nv_rd32(ram->pfb, reg->addr[0]);
+	return reg->data;
+}
+
+static inline void
+ramfuc_wr32(struct ramfuc *ram, struct ramfuc_reg *reg, u32 data)
+{
+	reg->sequence = ram->sequence;
+	reg->data = data;
+	if (reg->addr[0] != reg->addr[1])
+		nouveau_memx_wr32(ram->memx, reg->addr[1], reg->data);
+	nouveau_memx_wr32(ram->memx, reg->addr[0], reg->data);
+}
+
+static inline void
+ramfuc_nuke(struct ramfuc *ram, struct ramfuc_reg *reg)
+{
+	reg->force = true;
+}
+
+static inline u32
+ramfuc_mask(struct ramfuc *ram, struct ramfuc_reg *reg, u32 mask, u32 data)
+{
+	u32 temp = ramfuc_rd32(ram, reg);
+	if (temp != ((temp & ~mask) | data) || reg->force) {
+		ramfuc_wr32(ram, reg, (temp & ~mask) | data);
+		reg->force = false;
+	}
+	return temp;
+}
+
+static inline void
+ramfuc_wait(struct ramfuc *ram, u32 addr, u32 mask, u32 data, u32 nsec)
+{
+	nouveau_memx_wait(ram->memx, addr, mask, data, nsec);
+}
+
+static inline void
+ramfuc_nsec(struct ramfuc *ram, u32 nsec)
+{
+	nouveau_memx_nsec(ram->memx, nsec);
+}
+
+#define ram_init(s,p)       ramfuc_init(&(s)->base, (p))
+#define ram_exec(s,e)       ramfuc_exec(&(s)->base, (e))
+#define ram_have(s,r)       ((s)->r_##r.addr != 0x000000)
+#define ram_rd32(s,r)       ramfuc_rd32(&(s)->base, &(s)->r_##r)
+#define ram_wr32(s,r,d)     ramfuc_wr32(&(s)->base, &(s)->r_##r, (d))
+#define ram_nuke(s,r)       ramfuc_nuke(&(s)->base, &(s)->r_##r)
+#define ram_mask(s,r,m,d)   ramfuc_mask(&(s)->base, &(s)->r_##r, (m), (d))
+#define ram_wait(s,r,m,d,n) ramfuc_wait(&(s)->base, (r), (m), (d), (n))
+#define ram_nsec(s,n)       ramfuc_nsec(&(s)->base, (n))
+
+#endif
diff --git a/drivers/gpu/drm/nouveau/core/subdev/fb/ramnv40.c b/drivers/gpu/drm/nouveau/core/subdev/fb/ramnv40.c
index ee49ac4dbdb6..7648beb11199 100644
--- a/drivers/gpu/drm/nouveau/core/subdev/fb/ramnv40.c
+++ b/drivers/gpu/drm/nouveau/core/subdev/fb/ramnv40.c
@@ -22,7 +22,154 @@
  * Authors: Ben Skeggs
  */
 
-#include "priv.h"
+#include <subdev/bios.h>
+#include <subdev/bios/bit.h>
+#include <subdev/bios/pll.h>
+#include <subdev/bios/init.h>
+#include <subdev/clock.h>
+#include <subdev/clock/pll.h>
+#include <subdev/timer.h>
+
+#include <engine/fifo.h>
+
+#include "nv40.h"
+
+int
+nv40_ram_calc(struct nouveau_fb *pfb, u32 freq)
+{
+	struct nouveau_bios *bios = nouveau_bios(pfb);
+	struct nv40_ram *ram = (void *)pfb->ram;
+	struct nvbios_pll pll;
+	int N1, M1, N2, M2;
+	int log2P, ret;
+
+	ret = nvbios_pll_parse(bios, 0x04, &pll);
+	if (ret) {
+		nv_error(pfb, "mclk pll data not found\n");
+		return ret;
+	}
+
+	ret = nv04_pll_calc(nv_subdev(pfb), &pll, freq,
+			    &N1, &M1, &N2, &M2, &log2P);
+	if (ret < 0)
+		return ret;
+
+	ram->ctrl  = 0x80000000 | (log2P << 16);
+	ram->ctrl |= min(pll.bias_p + log2P, (int)pll.max_p) << 20;
+	if (N2 == M2) {
+		ram->ctrl |= 0x00000100;
+		ram->coef  = (N1 << 8) | M1;
+	} else {
+		ram->ctrl |= 0x40000000;
+		ram->coef  = (N2 << 24) | (M2 << 16) | (N1 << 8) | M1;
+	}
+
+	return 0;
+}
+
+int
+nv40_ram_prog(struct nouveau_fb *pfb)
+{
+	struct nouveau_bios *bios = nouveau_bios(pfb);
+	struct nv40_ram *ram = (void *)pfb->ram;
+	struct bit_entry M;
+	u32 crtc_mask = 0;
+	u8  sr1[2];
+	int i;
+
+	/* determine which CRTCs are active, fetch VGA_SR1 for each */
+	for (i = 0; i < 2; i++) {
+		u32 vbl = nv_rd32(pfb, 0x600808 + (i * 0x2000));
+		u32 cnt = 0;
+		do {
+			if (vbl != nv_rd32(pfb, 0x600808 + (i * 0x2000))) {
+				nv_wr08(pfb, 0x0c03c4 + (i * 0x2000), 0x01);
+				sr1[i] = nv_rd08(pfb, 0x0c03c5 + (i * 0x2000));
+				if (!(sr1[i] & 0x20))
+					crtc_mask |= (1 << i);
+				break;
+			}
+			udelay(1);
+		} while (cnt++ < 32);
+	}
+
+	/* wait for vblank start on active crtcs, disable memory access */
+	for (i = 0; i < 2; i++) {
+		if (!(crtc_mask & (1 << i)))
+			continue;
+		nv_wait(pfb, 0x600808 + (i * 0x2000), 0x00010000, 0x00000000);
+		nv_wait(pfb, 0x600808 + (i * 0x2000), 0x00010000, 0x00010000);
+		nv_wr08(pfb, 0x0c03c4 + (i * 0x2000), 0x01);
+		nv_wr08(pfb, 0x0c03c5 + (i * 0x2000), sr1[i] | 0x20);
+	}
+
+	/* prepare ram for reclocking */
+	nv_wr32(pfb, 0x1002d4, 0x00000001); /* precharge */
+	nv_wr32(pfb, 0x1002d0, 0x00000001); /* refresh */
+	nv_wr32(pfb, 0x1002d0, 0x00000001); /* refresh */
+	nv_mask(pfb, 0x100210, 0x80000000, 0x00000000); /* no auto refresh */
+	nv_wr32(pfb, 0x1002dc, 0x00000001); /* enable self-refresh */
+
+	/* change the PLL of each memory partition */
+	nv_mask(pfb, 0x00c040, 0x0000c000, 0x00000000);
+	switch (nv_device(pfb)->chipset) {
+	case 0x40:
+	case 0x45:
+	case 0x41:
+	case 0x42:
+	case 0x47:
+		nv_mask(pfb, 0x004044, 0xc0771100, ram->ctrl);
+		nv_mask(pfb, 0x00402c, 0xc0771100, ram->ctrl);
+		nv_wr32(pfb, 0x004048, ram->coef);
+		nv_wr32(pfb, 0x004030, ram->coef);
+	case 0x43:
+	case 0x49:
+	case 0x4b:
+		nv_mask(pfb, 0x004038, 0xc0771100, ram->ctrl);
+		nv_wr32(pfb, 0x00403c, ram->coef);
+	default:
+		nv_mask(pfb, 0x004020, 0xc0771100, ram->ctrl);
+		nv_wr32(pfb, 0x004024, ram->coef);
+		break;
+	}
+	udelay(100);
+	nv_mask(pfb, 0x00c040, 0x0000c000, 0x0000c000);
+
+	/* re-enable normal operation of memory controller */
+	nv_wr32(pfb, 0x1002dc, 0x00000000);
+	nv_mask(pfb, 0x100210, 0x80000000, 0x80000000);
+	udelay(100);
+
+	/* execute memory reset script from vbios */
+	if (!bit_entry(bios, 'M', &M)) {
+		struct nvbios_init init = {
+			.subdev = nv_subdev(pfb),
+			.bios = bios,
+			.offset = nv_ro16(bios, M.offset + 0x00),
+			.execute = 1,
+		};
+
+		nvbios_exec(&init);
+	}
+
+	/* make sure we're in vblank (hopefully the same one as before), and
+	 * then re-enable crtc memory access
+	 */
+	for (i = 0; i < 2; i++) {
+		if (!(crtc_mask & (1 << i)))
+			continue;
+		nv_wait(pfb, 0x600808 + (i * 0x2000), 0x00010000, 0x00010000);
+		nv_wr08(pfb, 0x0c03c4 + (i * 0x2000), 0x01);
+		nv_wr08(pfb, 0x0c03c5 + (i * 0x2000), sr1[i]);
+	}
+
+	return 0;
+}
+
+void
+nv40_ram_tidy(struct nouveau_fb *pfb)
+{
+}
 
 static int
 nv40_ram_create(struct nouveau_object *parent, struct nouveau_object *engine,
@@ -30,7 +177,7 @@ nv40_ram_create(struct nouveau_object *parent, struct nouveau_object *engine,
 		struct nouveau_object **pobject)
 {
 	struct nouveau_fb *pfb = nouveau_fb(parent);
-	struct nouveau_ram *ram;
+	struct nv40_ram *ram;
 	u32 pbus1218 = nv_rd32(pfb, 0x001218);
 	int ret;
 
@@ -40,15 +187,18 @@ nv40_ram_create(struct nouveau_object *parent, struct nouveau_object *engine,
 		return ret;
 
 	switch (pbus1218 & 0x00000300) {
-	case 0x00000000: ram->type = NV_MEM_TYPE_SDRAM; break;
-	case 0x00000100: ram->type = NV_MEM_TYPE_DDR1; break;
-	case 0x00000200: ram->type = NV_MEM_TYPE_GDDR3; break;
-	case 0x00000300: ram->type = NV_MEM_TYPE_DDR2; break;
+	case 0x00000000: ram->base.type = NV_MEM_TYPE_SDRAM; break;
+	case 0x00000100: ram->base.type = NV_MEM_TYPE_DDR1; break;
+	case 0x00000200: ram->base.type = NV_MEM_TYPE_GDDR3; break;
+	case 0x00000300: ram->base.type = NV_MEM_TYPE_DDR2; break;
 	}
 
-	ram->size  =  nv_rd32(pfb, 0x10020c) & 0xff000000;
-	ram->parts = (nv_rd32(pfb, 0x100200) & 0x00000003) + 1;
-	ram->tags  =  nv_rd32(pfb, 0x100320);
+	ram->base.size  =  nv_rd32(pfb, 0x10020c) & 0xff000000;
+	ram->base.parts = (nv_rd32(pfb, 0x100200) & 0x00000003) + 1;
+	ram->base.tags  =  nv_rd32(pfb, 0x100320);
+	ram->base.calc = nv40_ram_calc;
+	ram->base.prog = nv40_ram_prog;
+	ram->base.tidy = nv40_ram_tidy;
 	return 0;
 }
 
diff --git a/drivers/gpu/drm/nouveau/core/subdev/fb/ramnv41.c b/drivers/gpu/drm/nouveau/core/subdev/fb/ramnv41.c
index 1dab7e12abab..d64498a4d9ee 100644
--- a/drivers/gpu/drm/nouveau/core/subdev/fb/ramnv41.c
+++ b/drivers/gpu/drm/nouveau/core/subdev/fb/ramnv41.c
@@ -22,7 +22,7 @@
  * Authors: Ben Skeggs
  */
 
-#include "priv.h"
+#include "nv40.h"
 
 static int
 nv41_ram_create(struct nouveau_object *parent, struct nouveau_object *engine,
@@ -30,7 +30,7 @@ nv41_ram_create(struct nouveau_object *parent, struct nouveau_object *engine,
 		struct nouveau_object **pobject)
 {
 	struct nouveau_fb *pfb = nouveau_fb(parent);
-	struct nouveau_ram *ram;
+	struct nv40_ram *ram;
 	u32 pfb474 = nv_rd32(pfb, 0x100474);
 	int ret;
 
@@ -40,15 +40,18 @@ nv41_ram_create(struct nouveau_object *parent, struct nouveau_object *engine,
 		return ret;
 
 	if (pfb474 & 0x00000004)
-		ram->type = NV_MEM_TYPE_GDDR3;
+		ram->base.type = NV_MEM_TYPE_GDDR3;
 	if (pfb474 & 0x00000002)
-		ram->type = NV_MEM_TYPE_DDR2;
+		ram->base.type = NV_MEM_TYPE_DDR2;
 	if (pfb474 & 0x00000001)
-		ram->type = NV_MEM_TYPE_DDR1;
+		ram->base.type = NV_MEM_TYPE_DDR1;
 
-	ram->size  =  nv_rd32(pfb, 0x10020c) & 0xff000000;
-	ram->parts = (nv_rd32(pfb, 0x100200) & 0x00000003) + 1;
-	ram->tags  =  nv_rd32(pfb, 0x100320);
+	ram->base.size  =  nv_rd32(pfb, 0x10020c) & 0xff000000;
+	ram->base.parts = (nv_rd32(pfb, 0x100200) & 0x00000003) + 1;
+	ram->base.tags  =  nv_rd32(pfb, 0x100320);
+	ram->base.calc = nv40_ram_calc;
+	ram->base.prog = nv40_ram_prog;
+	ram->base.tidy = nv40_ram_tidy;
 	return 0;
 }
 
diff --git a/drivers/gpu/drm/nouveau/core/subdev/fb/ramnv44.c b/drivers/gpu/drm/nouveau/core/subdev/fb/ramnv44.c
index 25fff842e5c1..089acac810c5 100644
--- a/drivers/gpu/drm/nouveau/core/subdev/fb/ramnv44.c
+++ b/drivers/gpu/drm/nouveau/core/subdev/fb/ramnv44.c
@@ -22,7 +22,7 @@
  * Authors: Ben Skeggs
  */
 
-#include "priv.h"
+#include "nv40.h"
 
 static int
 nv44_ram_create(struct nouveau_object *parent, struct nouveau_object *engine,
@@ -30,7 +30,7 @@ nv44_ram_create(struct nouveau_object *parent, struct nouveau_object *engine,
 		struct nouveau_object **pobject)
 {
 	struct nouveau_fb *pfb = nouveau_fb(parent);
-	struct nouveau_ram *ram;
+	struct nv40_ram *ram;
 	u32 pfb474 = nv_rd32(pfb, 0x100474);
 	int ret;
 
@@ -40,13 +40,16 @@ nv44_ram_create(struct nouveau_object *parent, struct nouveau_object *engine,
 		return ret;
 
 	if (pfb474 & 0x00000004)
-		ram->type = NV_MEM_TYPE_GDDR3;
+		ram->base.type = NV_MEM_TYPE_GDDR3;
 	if (pfb474 & 0x00000002)
-		ram->type = NV_MEM_TYPE_DDR2;
+		ram->base.type = NV_MEM_TYPE_DDR2;
 	if (pfb474 & 0x00000001)
-		ram->type = NV_MEM_TYPE_DDR1;
+		ram->base.type = NV_MEM_TYPE_DDR1;
 
-	ram->size = nv_rd32(pfb, 0x10020c) & 0xff000000;
+	ram->base.size = nv_rd32(pfb, 0x10020c) & 0xff000000;
+	ram->base.calc = nv40_ram_calc;
+	ram->base.prog = nv40_ram_prog;
+	ram->base.tidy = nv40_ram_tidy;
 	return 0;
 }
 
diff --git a/drivers/gpu/drm/nouveau/core/subdev/fb/ramnv49.c b/drivers/gpu/drm/nouveau/core/subdev/fb/ramnv49.c
index ab7ef0ac9e34..baa013afa57b 100644
--- a/drivers/gpu/drm/nouveau/core/subdev/fb/ramnv49.c
+++ b/drivers/gpu/drm/nouveau/core/subdev/fb/ramnv49.c
@@ -22,7 +22,7 @@
  * Authors: Ben Skeggs
  */
 
-#include "priv.h"
+#include "nv40.h"
 
 static int
 nv49_ram_create(struct nouveau_object *parent, struct nouveau_object *engine,
@@ -30,7 +30,7 @@ nv49_ram_create(struct nouveau_object *parent, struct nouveau_object *engine,
 		struct nouveau_object **pobject)
 {
 	struct nouveau_fb *pfb = nouveau_fb(parent);
-	struct nouveau_ram *ram;
+	struct nv40_ram *ram;
 	u32 pfb914 = nv_rd32(pfb, 0x100914);
 	int ret;
 
@@ -40,15 +40,18 @@ nv49_ram_create(struct nouveau_object *parent, struct nouveau_object *engine,
 		return ret;
 
 	switch (pfb914 & 0x00000003) {
-	case 0x00000000: ram->type = NV_MEM_TYPE_DDR1; break;
-	case 0x00000001: ram->type = NV_MEM_TYPE_DDR2; break;
-	case 0x00000002: ram->type = NV_MEM_TYPE_GDDR3; break;
+	case 0x00000000: ram->base.type = NV_MEM_TYPE_DDR1; break;
+	case 0x00000001: ram->base.type = NV_MEM_TYPE_DDR2; break;
+	case 0x00000002: ram->base.type = NV_MEM_TYPE_GDDR3; break;
 	case 0x00000003: break;
 	}
 
-	ram->size  =  nv_rd32(pfb, 0x10020c) & 0xff000000;
-	ram->parts = (nv_rd32(pfb, 0x100200) & 0x00000003) + 1;
-	ram->tags  =  nv_rd32(pfb, 0x100320);
+	ram->base.size  =  nv_rd32(pfb, 0x10020c) & 0xff000000;
+	ram->base.parts = (nv_rd32(pfb, 0x100200) & 0x00000003) + 1;
+	ram->base.tags  =  nv_rd32(pfb, 0x100320);
+	ram->base.calc = nv40_ram_calc;
+	ram->base.prog = nv40_ram_prog;
+	ram->base.tidy = nv40_ram_tidy;
 	return 0;
 }
 
diff --git a/drivers/gpu/drm/nouveau/core/subdev/fb/ramnv50.c b/drivers/gpu/drm/nouveau/core/subdev/fb/ramnv50.c
index 903baff77fdd..76762a17d89c 100644
--- a/drivers/gpu/drm/nouveau/core/subdev/fb/ramnv50.c
+++ b/drivers/gpu/drm/nouveau/core/subdev/fb/ramnv50.c
@@ -23,8 +23,215 @@
  */
 
 #include <subdev/bios.h>
+#include <subdev/bios/bit.h>
+#include <subdev/bios/pll.h>
+#include <subdev/bios/perf.h>
+#include <subdev/bios/timing.h>
+#include <subdev/clock/pll.h>
+#include <subdev/fb.h>
+
+#include <core/option.h>
 #include <core/mm.h>
-#include "priv.h"
+
+#include "ramseq.h"
+
+#include "nv50.h"
+
+struct nv50_ramseq {
+	struct hwsq base;
+	struct hwsq_reg r_0x002504;
+	struct hwsq_reg r_0x004008;
+	struct hwsq_reg r_0x00400c;
+	struct hwsq_reg r_0x00c040;
+	struct hwsq_reg r_0x100210;
+	struct hwsq_reg r_0x1002d0;
+	struct hwsq_reg r_0x1002d4;
+	struct hwsq_reg r_0x1002dc;
+	struct hwsq_reg r_0x100da0[8];
+	struct hwsq_reg r_0x100e20;
+	struct hwsq_reg r_0x100e24;
+	struct hwsq_reg r_0x611200;
+	struct hwsq_reg r_timing[9];
+	struct hwsq_reg r_mr[4];
+};
+
+struct nv50_ram {
+	struct nouveau_ram base;
+	struct nv50_ramseq hwsq;
+};
+
+#define QFX5800NVA0 1
+
+static int
+nv50_ram_calc(struct nouveau_fb *pfb, u32 freq)
+{
+	struct nouveau_bios *bios = nouveau_bios(pfb);
+	struct nv50_ram *ram = (void *)pfb->ram;
+	struct nv50_ramseq *hwsq = &ram->hwsq;
+	struct nvbios_perfE perfE;
+	struct nvbios_pll mpll;
+	struct bit_entry M;
+	struct {
+		u32 data;
+		u8  size;
+	} ramcfg, timing;
+	u8  ver, hdr, cnt, strap;
+	u32 data;
+	int N1, M1, N2, M2, P;
+	int ret, i;
+
+	/* lookup closest matching performance table entry for frequency */
+	i = 0;
+	do {
+		ramcfg.data = nvbios_perfEp(bios, i++, &ver, &hdr, &cnt,
+					   &ramcfg.size, &perfE);
+		if (!ramcfg.data || (ver < 0x25 || ver >= 0x40) ||
+		    (ramcfg.size < 2)) {
+			nv_error(pfb, "invalid/missing perftab entry\n");
+			return -EINVAL;
+		}
+	} while (perfE.memory < freq);
+
+	/* locate specific data set for the attached memory */
+	if (bit_entry(bios, 'M', &M) || M.version != 1 || M.length < 5) {
+		nv_error(pfb, "invalid/missing memory table\n");
+		return -EINVAL;
+	}
+
+	strap = (nv_rd32(pfb, 0x101000) & 0x0000003c) >> 2;
+	data = nv_ro16(bios, M.offset + 3);
+	if (data)
+		strap = nv_ro08(bios, data + strap);
+
+	if (strap >= cnt) {
+		nv_error(pfb, "invalid ramcfg strap\n");
+		return -EINVAL;
+	}
+
+	ramcfg.data += hdr + (strap * ramcfg.size);
+
+	/* lookup memory timings, if bios says they're present */
+	strap = nv_ro08(bios, ramcfg.data + 0x01);
+	if (strap != 0xff) {
+		timing.data = nvbios_timing_entry(bios, strap, &ver, &hdr);
+		if (!timing.data || ver != 0x10 || hdr < 0x12) {
+			nv_error(pfb, "invalid/missing timing entry "
+				 "%02x %04x %02x %02x\n",
+				 strap, timing.data, ver, hdr);
+			return -EINVAL;
+		}
+	} else {
+		timing.data = 0;
+	}
+
+	ret = ram_init(hwsq, nv_subdev(pfb));
+	if (ret)
+		return ret;
+
+	ram_wait(hwsq, 0x01, 0x00); /* wait for !vblank */
+	ram_wait(hwsq, 0x01, 0x01); /* wait for vblank */
+	ram_wr32(hwsq, 0x611200, 0x00003300);
+	ram_wr32(hwsq, 0x002504, 0x00000001); /* block fifo */
+	ram_nsec(hwsq, 8000);
+	ram_setf(hwsq, 0x10, 0x00); /* disable fb */
+	ram_wait(hwsq, 0x00, 0x01); /* wait for fb disabled */
+
+	ram_wr32(hwsq, 0x1002d4, 0x00000001); /* precharge */
+	ram_wr32(hwsq, 0x1002d0, 0x00000001); /* refresh */
+	ram_wr32(hwsq, 0x1002d0, 0x00000001); /* refresh */
+	ram_wr32(hwsq, 0x100210, 0x00000000); /* disable auto-refresh */
+	ram_wr32(hwsq, 0x1002dc, 0x00000001); /* enable self-refresh */
+
+	ret = nvbios_pll_parse(bios, 0x004008, &mpll);
+	mpll.vco2.max_freq = 0;
+	if (ret == 0) {
+		ret = nv04_pll_calc(nv_subdev(pfb), &mpll, freq,
+				   &N1, &M1, &N2, &M2, &P);
+		if (ret == 0)
+			ret = -EINVAL;
+	}
+
+	if (ret < 0)
+		return ret;
+
+	ram_mask(hwsq, 0x00c040, 0xc000c000, 0x0000c000);
+	ram_mask(hwsq, 0x004008, 0x00000200, 0x00000200);
+	ram_mask(hwsq, 0x00400c, 0x0000ffff, (N1 << 8) | M1);
+	ram_mask(hwsq, 0x004008, 0x81ff0000, 0x80000000 | (mpll.bias_p << 19) |
+					     (P << 22) | (P << 16));
+#if QFX5800NVA0
+	for (i = 0; i < 8; i++)
+		ram_mask(hwsq, 0x100da0[i], 0x00000000, 0x00000000); /*XXX*/
+#endif
+	ram_nsec(hwsq, 96000); /*XXX*/
+	ram_mask(hwsq, 0x004008, 0x00002200, 0x00002000);
+
+	ram_wr32(hwsq, 0x1002dc, 0x00000000); /* disable self-refresh */
+	ram_wr32(hwsq, 0x100210, 0x80000000); /* enable auto-refresh */
+
+	ram_nsec(hwsq, 12000);
+
+	switch (ram->base.type) {
+	case NV_MEM_TYPE_DDR2:
+		ram_nuke(hwsq, mr[0]); /* force update */
+		ram_mask(hwsq, mr[0], 0x000, 0x000);
+		break;
+	case NV_MEM_TYPE_GDDR3:
+		ram_mask(hwsq, mr[2], 0x000, 0x000);
+		ram_nuke(hwsq, mr[0]); /* force update */
+		ram_mask(hwsq, mr[0], 0x000, 0x000);
+		break;
+	default:
+		break;
+	}
+
+	ram_mask(hwsq, timing[3], 0x00000000, 0x00000000); /*XXX*/
+	ram_mask(hwsq, timing[1], 0x00000000, 0x00000000); /*XXX*/
+	ram_mask(hwsq, timing[6], 0x00000000, 0x00000000); /*XXX*/
+	ram_mask(hwsq, timing[7], 0x00000000, 0x00000000); /*XXX*/
+	ram_mask(hwsq, timing[8], 0x00000000, 0x00000000); /*XXX*/
+	ram_mask(hwsq, timing[0], 0x00000000, 0x00000000); /*XXX*/
+	ram_mask(hwsq, timing[2], 0x00000000, 0x00000000); /*XXX*/
+	ram_mask(hwsq, timing[4], 0x00000000, 0x00000000); /*XXX*/
+	ram_mask(hwsq, timing[5], 0x00000000, 0x00000000); /*XXX*/
+
+	ram_mask(hwsq, timing[0], 0x00000000, 0x00000000); /*XXX*/
+
+#if QFX5800NVA0
+	ram_nuke(hwsq, 0x100e24);
+	ram_mask(hwsq, 0x100e24, 0x00000000, 0x00000000);
+	ram_nuke(hwsq, 0x100e20);
+	ram_mask(hwsq, 0x100e20, 0x00000000, 0x00000000);
+#endif
+
+	ram_mask(hwsq, mr[0], 0x100, 0x100);
+	ram_mask(hwsq, mr[0], 0x100, 0x000);
+
+	ram_setf(hwsq, 0x10, 0x01); /* enable fb */
+	ram_wait(hwsq, 0x00, 0x00); /* wait for fb enabled */
+	ram_wr32(hwsq, 0x611200, 0x00003330);
+	ram_wr32(hwsq, 0x002504, 0x00000000); /* un-block fifo */
+	return 0;
+}
+
+static int
+nv50_ram_prog(struct nouveau_fb *pfb)
+{
+	struct nouveau_device *device = nv_device(pfb);
+	struct nv50_ram *ram = (void *)pfb->ram;
+	struct nv50_ramseq *hwsq = &ram->hwsq;
+
+	ram_exec(hwsq, nouveau_boolopt(device->cfgopt, "NvMemExec", false));
+	return 0;
+}
+
+static void
+nv50_ram_tidy(struct nouveau_fb *pfb)
+{
+	struct nv50_ram *ram = (void *)pfb->ram;
+	struct nv50_ramseq *hwsq = &ram->hwsq;
+	ram_exec(hwsq, false);
+}
 
 void
 __nv50_ram_put(struct nouveau_fb *pfb, struct nouveau_mem *mem)
@@ -57,7 +264,7 @@ nv50_ram_put(struct nouveau_fb *pfb, struct nouveau_mem **pmem)
 	kfree(mem);
 }
 
-static int
+int
 nv50_ram_get(struct nouveau_fb *pfb, u64 size, u32 align, u32 ncmin,
 	     u32 memtype, struct nouveau_mem **pmem)
 {
@@ -160,77 +367,114 @@ nv50_fb_vram_rblock(struct nouveau_fb *pfb, struct nouveau_ram *ram)
 	return rblock_size;
 }
 
-static int
-nv50_ram_create(struct nouveau_object *parent, struct nouveau_object *engine,
-		struct nouveau_oclass *oclass, void *data, u32 datasize,
-		struct nouveau_object **pobject)
+int
+nv50_ram_create_(struct nouveau_object *parent, struct nouveau_object *engine,
+		 struct nouveau_oclass *oclass, int length, void **pobject)
 {
-	struct nouveau_fb *pfb = nouveau_fb(parent);
-	struct nouveau_device *device = nv_device(pfb);
-	struct nouveau_bios *bios = nouveau_bios(device);
-	struct nouveau_ram *ram;
 	const u32 rsvd_head = ( 256 * 1024) >> 12; /* vga memory */
 	const u32 rsvd_tail = (1024 * 1024) >> 12; /* vbios etc */
-	u32 size;
+	struct nouveau_bios *bios = nouveau_bios(parent);
+	struct nouveau_fb *pfb = nouveau_fb(parent);
+	struct nouveau_ram *ram;
 	int ret;
 
-	ret = nouveau_ram_create(parent, engine, oclass, &ram);
-	*pobject = nv_object(ram);
+	ret = nouveau_ram_create_(parent, engine, oclass, length, pobject);
+	ram = *pobject;
 	if (ret)
 		return ret;
 
 	ram->size = nv_rd32(pfb, 0x10020c);
-	ram->size = (ram->size & 0xffffff00) |
-		       ((ram->size & 0x000000ff) << 32);
-
-	size = (ram->size >> 12) - rsvd_head - rsvd_tail;
-	switch (device->chipset) {
-	case 0xaa:
-	case 0xac:
-	case 0xaf: /* IGPs, no reordering, no real VRAM */
-		ret = nouveau_mm_init(&pfb->vram, rsvd_head, size, 1);
-		if (ret)
-			return ret;
+	ram->size = (ram->size & 0xffffff00) | ((ram->size & 0x000000ff) << 32);
 
-		ram->type   = NV_MEM_TYPE_STOLEN;
-		ram->stolen = (u64)nv_rd32(pfb, 0x100e10) << 12;
+	switch (nv_rd32(pfb, 0x100714) & 0x00000007) {
+	case 0: ram->type = NV_MEM_TYPE_DDR1; break;
+	case 1:
+		if (nouveau_fb_bios_memtype(bios) == NV_MEM_TYPE_DDR3)
+			ram->type = NV_MEM_TYPE_DDR3;
+		else
+			ram->type = NV_MEM_TYPE_DDR2;
 		break;
+	case 2: ram->type = NV_MEM_TYPE_GDDR3; break;
+	case 3: ram->type = NV_MEM_TYPE_GDDR4; break;
+	case 4: ram->type = NV_MEM_TYPE_GDDR5; break;
 	default:
-		switch (nv_rd32(pfb, 0x100714) & 0x00000007) {
-		case 0: ram->type = NV_MEM_TYPE_DDR1; break;
-		case 1:
-			if (nouveau_fb_bios_memtype(bios) == NV_MEM_TYPE_DDR3)
-				ram->type = NV_MEM_TYPE_DDR3;
-			else
-				ram->type = NV_MEM_TYPE_DDR2;
-			break;
-		case 2: ram->type = NV_MEM_TYPE_GDDR3; break;
-		case 3: ram->type = NV_MEM_TYPE_GDDR4; break;
-		case 4: ram->type = NV_MEM_TYPE_GDDR5; break;
-		default:
-			break;
-		}
-
-		ret = nouveau_mm_init(&pfb->vram, rsvd_head, size,
-				      nv50_fb_vram_rblock(pfb, ram) >> 12);
-		if (ret)
-			return ret;
-
-		ram->ranks = (nv_rd32(pfb, 0x100200) & 0x4) ? 2 : 1;
-		ram->tags  =  nv_rd32(pfb, 0x100320);
 		break;
 	}
 
+	ret = nouveau_mm_init(&pfb->vram, rsvd_head, (ram->size >> 12) -
+			      (rsvd_head + rsvd_tail),
+			      nv50_fb_vram_rblock(pfb, ram) >> 12);
+	if (ret)
+		return ret;
+
+	ram->ranks = (nv_rd32(pfb, 0x100200) & 0x4) ? 2 : 1;
+	ram->tags  =  nv_rd32(pfb, 0x100320);
 	ram->get = nv50_ram_get;
 	ram->put = nv50_ram_put;
 	return 0;
 }
 
+static int
+nv50_ram_ctor(struct nouveau_object *parent, struct nouveau_object *engine,
+	      struct nouveau_oclass *oclass, void *data, u32 datasize,
+	      struct nouveau_object **pobject)
+{
+	struct nv50_ram *ram;
+	int ret, i;
+
+	ret = nv50_ram_create(parent, engine, oclass, &ram);
+	*pobject = nv_object(ram);
+	if (ret)
+		return ret;
+
+	switch (ram->base.type) {
+	case NV_MEM_TYPE_DDR2:
+	case NV_MEM_TYPE_GDDR3:
+		ram->base.calc = nv50_ram_calc;
+		ram->base.prog = nv50_ram_prog;
+		ram->base.tidy = nv50_ram_tidy;
+		break;
+	default:
+		nv_warn(ram, "reclocking of this ram type unsupported\n");
+		return 0;
+	}
+
+	ram->hwsq.r_0x002504 = hwsq_reg(0x002504);
+	ram->hwsq.r_0x00c040 = hwsq_reg(0x00c040);
+	ram->hwsq.r_0x004008 = hwsq_reg(0x004008);
+	ram->hwsq.r_0x00400c = hwsq_reg(0x00400c);
+	ram->hwsq.r_0x100210 = hwsq_reg(0x100210);
+	ram->hwsq.r_0x1002d0 = hwsq_reg(0x1002d0);
+	ram->hwsq.r_0x1002d4 = hwsq_reg(0x1002d4);
+	ram->hwsq.r_0x1002dc = hwsq_reg(0x1002dc);
+	for (i = 0; i < 8; i++)
+		ram->hwsq.r_0x100da0[i] = hwsq_reg(0x100da0 + (i * 0x04));
+	ram->hwsq.r_0x100e20 = hwsq_reg(0x100e20);
+	ram->hwsq.r_0x100e24 = hwsq_reg(0x100e24);
+	ram->hwsq.r_0x611200 = hwsq_reg(0x611200);
+
+	for (i = 0; i < 9; i++)
+		ram->hwsq.r_timing[i] = hwsq_reg(0x100220 + (i * 0x04));
+
+	if (ram->base.ranks > 1) {
+		ram->hwsq.r_mr[0] = hwsq_reg2(0x1002c0, 0x1002c8);
+		ram->hwsq.r_mr[1] = hwsq_reg2(0x1002c4, 0x1002cc);
+		ram->hwsq.r_mr[2] = hwsq_reg2(0x1002e0, 0x1002e8);
+		ram->hwsq.r_mr[3] = hwsq_reg2(0x1002e4, 0x1002ec);
+	} else {
+		ram->hwsq.r_mr[0] = hwsq_reg(0x1002c0);
+		ram->hwsq.r_mr[1] = hwsq_reg(0x1002c4);
+		ram->hwsq.r_mr[2] = hwsq_reg(0x1002e0);
+		ram->hwsq.r_mr[3] = hwsq_reg(0x1002e4);
+	}
+
+	return 0;
+}
+
 struct nouveau_oclass
 nv50_ram_oclass = {
-	.handle = 0,
 	.ofuncs = &(struct nouveau_ofuncs) {
-		.ctor = nv50_ram_create,
+		.ctor = nv50_ram_ctor,
 		.dtor = _nouveau_ram_dtor,
 		.init = _nouveau_ram_init,
 		.fini = _nouveau_ram_fini,
diff --git a/drivers/gpu/drm/nouveau/core/subdev/fb/ramnva3.c b/drivers/gpu/drm/nouveau/core/subdev/fb/ramnva3.c
new file mode 100644
index 000000000000..f6292cd9207c
--- /dev/null
+++ b/drivers/gpu/drm/nouveau/core/subdev/fb/ramnva3.c
@@ -0,0 +1,447 @@
+/*
+ * Copyright 2013 Red Hat Inc.
+ *
+ * Permission is hereby granted, free of charge, to any person obtaining a
+ * copy of this software and associated documentation files (the "Software"),
+ * to deal in the Software without restriction, including without limitation
+ * the rights to use, copy, modify, merge, publish, distribute, sublicense,
+ * and/or sell copies of the Software, and to permit persons to whom the
+ * Software is furnished to do so, subject to the following conditions:
+ *
+ * The above copyright notice and this permission notice shall be included in
+ * all copies or substantial portions of the Software.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+ * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+ * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT.  IN NO EVENT SHALL
+ * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR
+ * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE,
+ * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR
+ * OTHER DEALINGS IN THE SOFTWARE.
+ *
+ * Authors: Ben Skeggs
+ */
+
+#include <subdev/bios.h>
+#include <subdev/bios/bit.h>
+#include <subdev/bios/pll.h>
+#include <subdev/bios/rammap.h>
+#include <subdev/bios/timing.h>
+
+#include <subdev/clock/nva3.h>
+#include <subdev/clock/pll.h>
+
+#include <core/option.h>
+
+#include "ramfuc.h"
+
+#include "nv50.h"
+
+struct nva3_ramfuc {
+	struct ramfuc base;
+	struct ramfuc_reg r_0x004000;
+	struct ramfuc_reg r_0x004004;
+	struct ramfuc_reg r_0x004018;
+	struct ramfuc_reg r_0x004128;
+	struct ramfuc_reg r_0x004168;
+	struct ramfuc_reg r_0x100200;
+	struct ramfuc_reg r_0x100210;
+	struct ramfuc_reg r_0x100220[9];
+	struct ramfuc_reg r_0x1002d0;
+	struct ramfuc_reg r_0x1002d4;
+	struct ramfuc_reg r_0x1002dc;
+	struct ramfuc_reg r_0x10053c;
+	struct ramfuc_reg r_0x1005a0;
+	struct ramfuc_reg r_0x1005a4;
+	struct ramfuc_reg r_0x100714;
+	struct ramfuc_reg r_0x100718;
+	struct ramfuc_reg r_0x10071c;
+	struct ramfuc_reg r_0x100760;
+	struct ramfuc_reg r_0x1007a0;
+	struct ramfuc_reg r_0x1007e0;
+	struct ramfuc_reg r_0x10f804;
+	struct ramfuc_reg r_0x1110e0;
+	struct ramfuc_reg r_0x111100;
+	struct ramfuc_reg r_0x111104;
+	struct ramfuc_reg r_0x611200;
+	struct ramfuc_reg r_mr[4];
+};
+
+struct nva3_ram {
+	struct nouveau_ram base;
+	struct nva3_ramfuc fuc;
+};
+
+static int
+nva3_ram_calc(struct nouveau_fb *pfb, u32 freq)
+{
+	struct nouveau_bios *bios = nouveau_bios(pfb);
+	struct nva3_ram *ram = (void *)pfb->ram;
+	struct nva3_ramfuc *fuc = &ram->fuc;
+	struct nva3_clock_info mclk;
+	struct bit_entry M;
+	u8  ver, cnt, strap;
+	u32 data;
+	struct {
+		u32 data;
+		u8  size;
+	} rammap, ramcfg, timing;
+	u32 r004018, r100760, ctrl;
+	u32 unk714, unk718, unk71c;
+	int ret;
+
+	/* lookup memory config data relevant to the target frequency */
+	rammap.data = nvbios_rammap_match(bios, freq / 1000, &ver, &rammap.size,
+					 &cnt, &ramcfg.size);
+	if (!rammap.data || ver != 0x10 || rammap.size < 0x0e) {
+		nv_error(pfb, "invalid/missing rammap entry\n");
+		return -EINVAL;
+	}
+
+	/* locate specific data set for the attached memory */
+	if (bit_entry(bios, 'M', &M) || M.version != 2 || M.length < 3) {
+		nv_error(pfb, "invalid/missing memory table\n");
+		return -EINVAL;
+	}
+
+	strap = (nv_rd32(pfb, 0x101000) & 0x0000003c) >> 2;
+	data = nv_ro16(bios, M.offset + 1);
+	if (data)
+		strap = nv_ro08(bios, data + strap);
+
+	if (strap >= cnt) {
+		nv_error(pfb, "invalid ramcfg strap\n");
+		return -EINVAL;
+	}
+
+	ramcfg.data = rammap.data + rammap.size + (strap * ramcfg.size);
+	if (!ramcfg.data || ver != 0x10 || ramcfg.size < 0x0e) {
+		nv_error(pfb, "invalid/missing ramcfg entry\n");
+		return -EINVAL;
+	}
+
+	/* lookup memory timings, if bios says they're present */
+	strap = nv_ro08(bios, ramcfg.data + 0x01);
+	if (strap != 0xff) {
+		timing.data = nvbios_timing_entry(bios, strap, &ver,
+						 &timing.size);
+		if (!timing.data || ver != 0x10 || timing.size < 0x19) {
+			nv_error(pfb, "invalid/missing timing entry\n");
+			return -EINVAL;
+		}
+	} else {
+		timing.data = 0;
+	}
+
+	ret = nva3_clock_info(nouveau_clock(pfb), 0x12, 0x4000, freq, &mclk);
+	if (ret < 0) {
+		nv_error(pfb, "failed mclk calculation\n");
+		return ret;
+	}
+
+	ret = ram_init(fuc, pfb);
+	if (ret)
+		return ret;
+
+	/* XXX: where the fuck does 750MHz come from? */
+	if (freq <= 750000) {
+		r004018 = 0x10000000;
+		r100760 = 0x22222222;
+	} else {
+		r004018 = 0x00000000;
+		r100760 = 0x00000000;
+	}
+
+	ctrl = ram_rd32(fuc, 0x004000);
+	if (ctrl & 0x00000008) {
+		if (mclk.pll) {
+			ram_mask(fuc, 0x004128, 0x00000101, 0x00000101);
+			ram_wr32(fuc, 0x004004, mclk.pll);
+			ram_wr32(fuc, 0x004000, (ctrl |= 0x00000001));
+			ram_wr32(fuc, 0x004000, (ctrl &= 0xffffffef));
+			ram_wait(fuc, 0x004000, 0x00020000, 0x00020000, 64000);
+			ram_wr32(fuc, 0x004000, (ctrl |= 0x00000010));
+			ram_wr32(fuc, 0x004018, 0x00005000 | r004018);
+			ram_wr32(fuc, 0x004000, (ctrl |= 0x00000004));
+		}
+	} else {
+		u32 ssel = 0x00000101;
+		if (mclk.clk)
+			ssel |= mclk.clk;
+		else
+			ssel |= 0x00080000; /* 324MHz, shouldn't matter... */
+		ram_mask(fuc, 0x004168, 0x003f3141, ctrl);
+	}
+
+	if ( (nv_ro08(bios, ramcfg.data + 0x02) & 0x10)) {
+		ram_mask(fuc, 0x111104, 0x00000600, 0x00000000);
+	} else {
+		ram_mask(fuc, 0x111100, 0x40000000, 0x40000000);
+		ram_mask(fuc, 0x111104, 0x00000180, 0x00000000);
+	}
+
+	if (!(nv_ro08(bios, rammap.data + 0x04) & 0x02))
+		ram_mask(fuc, 0x100200, 0x00000800, 0x00000000);
+	ram_wr32(fuc, 0x611200, 0x00003300);
+	if (!(nv_ro08(bios, ramcfg.data + 0x02) & 0x10))
+		ram_wr32(fuc, 0x111100, 0x4c020000); /*XXX*/
+
+	ram_wr32(fuc, 0x1002d4, 0x00000001);
+	ram_wr32(fuc, 0x1002d0, 0x00000001);
+	ram_wr32(fuc, 0x1002d0, 0x00000001);
+	ram_wr32(fuc, 0x100210, 0x00000000);
+	ram_wr32(fuc, 0x1002dc, 0x00000001);
+	ram_nsec(fuc, 2000);
+
+	ctrl = ram_rd32(fuc, 0x004000);
+	if (!(ctrl & 0x00000008) && mclk.pll) {
+		ram_wr32(fuc, 0x004000, (ctrl |=  0x00000008));
+		ram_mask(fuc, 0x1110e0, 0x00088000, 0x00088000);
+		ram_wr32(fuc, 0x004018, 0x00001000);
+		ram_wr32(fuc, 0x004000, (ctrl &= ~0x00000001));
+		ram_wr32(fuc, 0x004004, mclk.pll);
+		ram_wr32(fuc, 0x004000, (ctrl |=  0x00000001));
+		udelay(64);
+		ram_wr32(fuc, 0x004018, 0x00005000 | r004018);
+		udelay(20);
+	} else
+	if (!mclk.pll) {
+		ram_mask(fuc, 0x004168, 0x003f3040, mclk.clk);
+		ram_wr32(fuc, 0x004000, (ctrl |= 0x00000008));
+		ram_mask(fuc, 0x1110e0, 0x00088000, 0x00088000);
+		ram_wr32(fuc, 0x004018, 0x0000d000 | r004018);
+	}
+
+	if ( (nv_ro08(bios, rammap.data + 0x04) & 0x08)) {
+		u32 unk5a0 = (nv_ro16(bios, ramcfg.data + 0x05) << 8) |
+			      nv_ro08(bios, ramcfg.data + 0x05);
+		u32 unk5a4 = (nv_ro16(bios, ramcfg.data + 0x07));
+		u32 unk804 = (nv_ro08(bios, ramcfg.data + 0x09) & 0xf0) << 16 |
+			     (nv_ro08(bios, ramcfg.data + 0x03) & 0x0f) << 16 |
+			     (nv_ro08(bios, ramcfg.data + 0x09) & 0x0f) |
+			     0x80000000;
+		ram_wr32(fuc, 0x1005a0, unk5a0);
+		ram_wr32(fuc, 0x1005a4, unk5a4);
+		ram_wr32(fuc, 0x10f804, unk804);
+		ram_mask(fuc, 0x10053c, 0x00001000, 0x00000000);
+	} else {
+		ram_mask(fuc, 0x10053c, 0x00001000, 0x00001000);
+		ram_mask(fuc, 0x10f804, 0x80000000, 0x00000000);
+		ram_mask(fuc, 0x100760, 0x22222222, r100760);
+		ram_mask(fuc, 0x1007a0, 0x22222222, r100760);
+		ram_mask(fuc, 0x1007e0, 0x22222222, r100760);
+	}
+
+	if (mclk.pll) {
+		ram_mask(fuc, 0x1110e0, 0x00088000, 0x00011000);
+		ram_wr32(fuc, 0x004000, (ctrl &= ~0x00000008));
+	}
+
+	/*XXX: LEAVE */
+	ram_wr32(fuc, 0x1002dc, 0x00000000);
+	ram_wr32(fuc, 0x1002d4, 0x00000001);
+	ram_wr32(fuc, 0x100210, 0x80000000);
+	ram_nsec(fuc, 1000);
+	ram_nsec(fuc, 1000);
+
+	ram_mask(fuc, mr[2], 0x00000000, 0x00000000);
+	ram_nsec(fuc, 1000);
+	ram_nuke(fuc, mr[0]);
+	ram_mask(fuc, mr[0], 0x00000000, 0x00000000);
+	ram_nsec(fuc, 1000);
+
+	ram_mask(fuc, 0x100220[3], 0x00000000, 0x00000000);
+	ram_mask(fuc, 0x100220[1], 0x00000000, 0x00000000);
+	ram_mask(fuc, 0x100220[6], 0x00000000, 0x00000000);
+	ram_mask(fuc, 0x100220[7], 0x00000000, 0x00000000);
+	ram_mask(fuc, 0x100220[2], 0x00000000, 0x00000000);
+	ram_mask(fuc, 0x100220[4], 0x00000000, 0x00000000);
+	ram_mask(fuc, 0x100220[5], 0x00000000, 0x00000000);
+	ram_mask(fuc, 0x100220[0], 0x00000000, 0x00000000);
+	ram_mask(fuc, 0x100220[8], 0x00000000, 0x00000000);
+
+	data = (nv_ro08(bios, ramcfg.data + 0x02) & 0x08) ? 0x00000000 : 0x00001000;
+	ram_mask(fuc, 0x100200, 0x00001000, data);
+
+	unk714 = ram_rd32(fuc, 0x100714) & ~0xf0000010;
+	unk718 = ram_rd32(fuc, 0x100718) & ~0x00000100;
+	unk71c = ram_rd32(fuc, 0x10071c) & ~0x00000100;
+	if ( (nv_ro08(bios, ramcfg.data + 0x02) & 0x20))
+		unk714 |= 0xf0000000;
+	if (!(nv_ro08(bios, ramcfg.data + 0x02) & 0x04))
+		unk714 |= 0x00000010;
+	ram_wr32(fuc, 0x100714, unk714);
+
+	if (nv_ro08(bios, ramcfg.data + 0x02) & 0x01)
+		unk71c |= 0x00000100;
+	ram_wr32(fuc, 0x10071c, unk71c);
+
+	if (nv_ro08(bios, ramcfg.data + 0x02) & 0x02)
+		unk718 |= 0x00000100;
+	ram_wr32(fuc, 0x100718, unk718);
+
+	if (nv_ro08(bios, ramcfg.data + 0x02) & 0x10)
+		ram_wr32(fuc, 0x111100, 0x48000000); /*XXX*/
+
+	ram_mask(fuc, mr[0], 0x100, 0x100);
+	ram_nsec(fuc, 1000);
+	ram_mask(fuc, mr[0], 0x100, 0x000);
+	ram_nsec(fuc, 1000);
+
+	ram_nsec(fuc, 2000);
+	ram_nsec(fuc, 12000);
+
+	ram_wr32(fuc, 0x611200, 0x00003330);
+	if ( (nv_ro08(bios, rammap.data + 0x04) & 0x02))
+		ram_mask(fuc, 0x100200, 0x00000800, 0x00000800);
+	if ( (nv_ro08(bios, ramcfg.data + 0x02) & 0x10)) {
+		ram_mask(fuc, 0x111104, 0x00000180, 0x00000180);
+		ram_mask(fuc, 0x111100, 0x40000000, 0x00000000);
+	} else {
+		ram_mask(fuc, 0x111104, 0x00000600, 0x00000600);
+	}
+
+	if (mclk.pll) {
+		ram_mask(fuc, 0x004168, 0x00000001, 0x00000000);
+		ram_mask(fuc, 0x004168, 0x00000100, 0x00000000);
+	} else {
+		ram_mask(fuc, 0x004000, 0x00000001, 0x00000000);
+		ram_mask(fuc, 0x004128, 0x00000001, 0x00000000);
+		ram_mask(fuc, 0x004128, 0x00000100, 0x00000000);
+	}
+
+	return 0;
+}
+
+static int
+nva3_ram_prog(struct nouveau_fb *pfb)
+{
+	struct nouveau_device *device = nv_device(pfb);
+	struct nva3_ram *ram = (void *)pfb->ram;
+	struct nva3_ramfuc *fuc = &ram->fuc;
+	ram_exec(fuc, nouveau_boolopt(device->cfgopt, "NvMemExec", false));
+	return 0;
+}
+
+static void
+nva3_ram_tidy(struct nouveau_fb *pfb)
+{
+	struct nva3_ram *ram = (void *)pfb->ram;
+	struct nva3_ramfuc *fuc = &ram->fuc;
+	ram_exec(fuc, false);
+}
+
+static int
+nva3_ram_init(struct nouveau_object *object)
+{
+	struct nouveau_fb *pfb = (void *)object->parent;
+	struct nva3_ram   *ram = (void *)object;
+	int ret, i;
+
+	ret = nouveau_ram_init(&ram->base);
+	if (ret)
+		return ret;
+
+	/* prepare for ddr link training, and load training patterns */
+	switch (ram->base.type) {
+	case NV_MEM_TYPE_DDR3: {
+		static const u32 pattern[16] = {
+			0xaaaaaaaa, 0xcccccccc, 0xdddddddd, 0xeeeeeeee,
+			0x00000000, 0x11111111, 0x44444444, 0xdddddddd,
+			0x33333333, 0x55555555, 0x77777777, 0x66666666,
+			0x99999999, 0x88888888, 0xeeeeeeee, 0xbbbbbbbb,
+		};
+
+		nv_wr32(pfb, 0x100538, 0x10001ff6); /*XXX*/
+		nv_wr32(pfb, 0x1005a8, 0x0000ffff);
+		nv_mask(pfb, 0x10f800, 0x00000001, 0x00000001);
+		for (i = 0; i < 0x30; i++) {
+			nv_wr32(pfb, 0x10f8c0, (i << 8) | i);
+			nv_wr32(pfb, 0x10f8e0, (i << 8) | i);
+			nv_wr32(pfb, 0x10f900, pattern[i % 16]);
+			nv_wr32(pfb, 0x10f920, pattern[i % 16]);
+		}
+	}
+		break;
+	default:
+		break;
+	}
+
+	return 0;
+}
+
+static int
+nva3_ram_ctor(struct nouveau_object *parent, struct nouveau_object *engine,
+	      struct nouveau_oclass *oclass, void *data, u32 datasize,
+	      struct nouveau_object **pobject)
+{
+	struct nva3_ram *ram;
+	int ret, i;
+
+	ret = nv50_ram_create(parent, engine, oclass, &ram);
+	*pobject = nv_object(ram);
+	if (ret)
+		return ret;
+
+	switch (ram->base.type) {
+	case NV_MEM_TYPE_DDR3:
+		ram->base.calc = nva3_ram_calc;
+		ram->base.prog = nva3_ram_prog;
+		ram->base.tidy = nva3_ram_tidy;
+		break;
+	default:
+		nv_warn(ram, "reclocking of this ram type unsupported\n");
+		return 0;
+	}
+
+	ram->fuc.r_0x004000 = ramfuc_reg(0x004000);
+	ram->fuc.r_0x004004 = ramfuc_reg(0x004004);
+	ram->fuc.r_0x004018 = ramfuc_reg(0x004018);
+	ram->fuc.r_0x004128 = ramfuc_reg(0x004128);
+	ram->fuc.r_0x004168 = ramfuc_reg(0x004168);
+	ram->fuc.r_0x100200 = ramfuc_reg(0x100200);
+	ram->fuc.r_0x100210 = ramfuc_reg(0x100210);
+	for (i = 0; i < 9; i++)
+		ram->fuc.r_0x100220[i] = ramfuc_reg(0x100220 + (i * 4));
+	ram->fuc.r_0x1002d0 = ramfuc_reg(0x1002d0);
+	ram->fuc.r_0x1002d4 = ramfuc_reg(0x1002d4);
+	ram->fuc.r_0x1002dc = ramfuc_reg(0x1002dc);
+	ram->fuc.r_0x10053c = ramfuc_reg(0x10053c);
+	ram->fuc.r_0x1005a0 = ramfuc_reg(0x1005a0);
+	ram->fuc.r_0x1005a4 = ramfuc_reg(0x1005a4);
+	ram->fuc.r_0x100714 = ramfuc_reg(0x100714);
+	ram->fuc.r_0x100718 = ramfuc_reg(0x100718);
+	ram->fuc.r_0x10071c = ramfuc_reg(0x10071c);
+	ram->fuc.r_0x100760 = ramfuc_reg(0x100760);
+	ram->fuc.r_0x1007a0 = ramfuc_reg(0x1007a0);
+	ram->fuc.r_0x1007e0 = ramfuc_reg(0x1007e0);
+	ram->fuc.r_0x10f804 = ramfuc_reg(0x10f804);
+	ram->fuc.r_0x1110e0 = ramfuc_reg(0x1110e0);
+	ram->fuc.r_0x111100 = ramfuc_reg(0x111100);
+	ram->fuc.r_0x111104 = ramfuc_reg(0x111104);
+	ram->fuc.r_0x611200 = ramfuc_reg(0x611200);
+
+	if (ram->base.ranks > 1) {
+		ram->fuc.r_mr[0] = ramfuc_reg2(0x1002c0, 0x1002c8);
+		ram->fuc.r_mr[1] = ramfuc_reg2(0x1002c4, 0x1002cc);
+		ram->fuc.r_mr[2] = ramfuc_reg2(0x1002e0, 0x1002e8);
+		ram->fuc.r_mr[3] = ramfuc_reg2(0x1002e4, 0x1002ec);
+	} else {
+		ram->fuc.r_mr[0] = ramfuc_reg(0x1002c0);
+		ram->fuc.r_mr[1] = ramfuc_reg(0x1002c4);
+		ram->fuc.r_mr[2] = ramfuc_reg(0x1002e0);
+		ram->fuc.r_mr[3] = ramfuc_reg(0x1002e4);
+	}
+
+	return 0;
+}
+
+struct nouveau_oclass
+nva3_ram_oclass = {
+	.ofuncs = &(struct nouveau_ofuncs) {
+		.ctor = nva3_ram_ctor,
+		.dtor = _nouveau_ram_dtor,
+		.init = nva3_ram_init,
+		.fini = _nouveau_ram_fini,
+	},
+};
diff --git a/drivers/gpu/drm/nouveau/core/subdev/fb/ramnvaa.c b/drivers/gpu/drm/nouveau/core/subdev/fb/ramnvaa.c
new file mode 100644
index 000000000000..00f2ca7e44a5
--- /dev/null
+++ b/drivers/gpu/drm/nouveau/core/subdev/fb/ramnvaa.c
@@ -0,0 +1,66 @@
+/*
+ * Copyright 2013 Red Hat Inc.
+ *
+ * Permission is hereby granted, free of charge, to any person obtaining a
+ * copy of this software and associated documentation files (the "Software"),
+ * to deal in the Software without restriction, including without limitation
+ * the rights to use, copy, modify, merge, publish, distribute, sublicense,
+ * and/or sell copies of the Software, and to permit persons to whom the
+ * Software is furnished to do so, subject to the following conditions:
+ *
+ * The above copyright notice and this permission notice shall be included in
+ * all copies or substantial portions of the Software.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+ * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+ * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT.  IN NO EVENT SHALL
+ * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR
+ * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE,
+ * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR
+ * OTHER DEALINGS IN THE SOFTWARE.
+ *
+ * Authors: Ben Skeggs
+ */
+
+#include "nv50.h"
+
+static int
+nvaa_ram_ctor(struct nouveau_object *parent, struct nouveau_object *engine,
+	      struct nouveau_oclass *oclass, void *data, u32 datasize,
+	      struct nouveau_object **pobject)
+{
+	const u32 rsvd_head = ( 256 * 1024) >> 12; /* vga memory */
+	const u32 rsvd_tail = (1024 * 1024) >> 12; /* vbios etc */
+	struct nouveau_fb *pfb = nouveau_fb(parent);
+	struct nouveau_ram *ram;
+	int ret;
+
+	ret = nouveau_ram_create(parent, engine, oclass, &ram);
+	*pobject = nv_object(ram);
+	if (ret)
+		return ret;
+
+	ram->size = nv_rd32(pfb, 0x10020c);
+	ram->size = (ram->size & 0xffffff00) | ((ram->size & 0x000000ff) << 32);
+
+	ret = nouveau_mm_init(&pfb->vram, rsvd_head, (ram->size >> 12) -
+			      (rsvd_head + rsvd_tail), 1);
+	if (ret)
+		return ret;
+
+	ram->type   = NV_MEM_TYPE_STOLEN;
+	ram->stolen = (u64)nv_rd32(pfb, 0x100e10) << 12;
+	ram->get = nv50_ram_get;
+	ram->put = nv50_ram_put;
+	return 0;
+}
+
+struct nouveau_oclass
+nvaa_ram_oclass = {
+	.ofuncs = &(struct nouveau_ofuncs) {
+		.ctor = nvaa_ram_ctor,
+		.dtor = _nouveau_ram_dtor,
+		.init = _nouveau_ram_init,
+		.fini = _nouveau_ram_fini,
+	},
+};
diff --git a/drivers/gpu/drm/nouveau/core/subdev/fb/ramnvc0.c b/drivers/gpu/drm/nouveau/core/subdev/fb/ramnvc0.c
index cf97c4de4a6b..f464547c6bab 100644
--- a/drivers/gpu/drm/nouveau/core/subdev/fb/ramnvc0.c
+++ b/drivers/gpu/drm/nouveau/core/subdev/fb/ramnvc0.c
@@ -23,9 +23,414 @@
  */
 
 #include <subdev/bios.h>
+#include <subdev/bios/bit.h>
+#include <subdev/bios/pll.h>
+#include <subdev/bios/rammap.h>
+#include <subdev/bios/timing.h>
 #include <subdev/ltcg.h>
 
-#include "priv.h"
+#include <subdev/clock.h>
+#include <subdev/clock/pll.h>
+
+#include <core/option.h>
+
+#include "ramfuc.h"
+
+#include "nvc0.h"
+
+struct nvc0_ramfuc {
+	struct ramfuc base;
+
+	struct ramfuc_reg r_0x10fe20;
+	struct ramfuc_reg r_0x10fe24;
+	struct ramfuc_reg r_0x137320;
+	struct ramfuc_reg r_0x137330;
+
+	struct ramfuc_reg r_0x132000;
+	struct ramfuc_reg r_0x132004;
+	struct ramfuc_reg r_0x132100;
+
+	struct ramfuc_reg r_0x137390;
+
+	struct ramfuc_reg r_0x10f290;
+	struct ramfuc_reg r_0x10f294;
+	struct ramfuc_reg r_0x10f298;
+	struct ramfuc_reg r_0x10f29c;
+	struct ramfuc_reg r_0x10f2a0;
+
+	struct ramfuc_reg r_0x10f300;
+	struct ramfuc_reg r_0x10f338;
+	struct ramfuc_reg r_0x10f340;
+	struct ramfuc_reg r_0x10f344;
+	struct ramfuc_reg r_0x10f348;
+
+	struct ramfuc_reg r_0x10f910;
+	struct ramfuc_reg r_0x10f914;
+
+	struct ramfuc_reg r_0x100b0c;
+	struct ramfuc_reg r_0x10f050;
+	struct ramfuc_reg r_0x10f090;
+	struct ramfuc_reg r_0x10f200;
+	struct ramfuc_reg r_0x10f210;
+	struct ramfuc_reg r_0x10f310;
+	struct ramfuc_reg r_0x10f314;
+	struct ramfuc_reg r_0x10f610;
+	struct ramfuc_reg r_0x10f614;
+	struct ramfuc_reg r_0x10f800;
+	struct ramfuc_reg r_0x10f808;
+	struct ramfuc_reg r_0x10f824;
+	struct ramfuc_reg r_0x10f830;
+	struct ramfuc_reg r_0x10f988;
+	struct ramfuc_reg r_0x10f98c;
+	struct ramfuc_reg r_0x10f990;
+	struct ramfuc_reg r_0x10f998;
+	struct ramfuc_reg r_0x10f9b0;
+	struct ramfuc_reg r_0x10f9b4;
+	struct ramfuc_reg r_0x10fb04;
+	struct ramfuc_reg r_0x10fb08;
+	struct ramfuc_reg r_0x137300;
+	struct ramfuc_reg r_0x137310;
+	struct ramfuc_reg r_0x137360;
+	struct ramfuc_reg r_0x1373ec;
+	struct ramfuc_reg r_0x1373f0;
+	struct ramfuc_reg r_0x1373f8;
+
+	struct ramfuc_reg r_0x61c140;
+	struct ramfuc_reg r_0x611200;
+
+	struct ramfuc_reg r_0x13d8f4;
+};
+
+struct nvc0_ram {
+	struct nouveau_ram base;
+	struct nvc0_ramfuc fuc;
+	struct nvbios_pll refpll;
+	struct nvbios_pll mempll;
+};
+
+static void
+nvc0_ram_train(struct nvc0_ramfuc *fuc, u32 magic)
+{
+	struct nvc0_ram *ram = container_of(fuc, typeof(*ram), fuc);
+	struct nouveau_fb *pfb = nouveau_fb(ram);
+	u32 part = nv_rd32(pfb, 0x022438), i;
+	u32 mask = nv_rd32(pfb, 0x022554);
+	u32 addr = 0x110974;
+
+	ram_wr32(fuc, 0x10f910, magic);
+	ram_wr32(fuc, 0x10f914, magic);
+
+	for (i = 0; (magic & 0x80000000) && i < part; addr += 0x1000, i++) {
+		if (mask & (1 << i))
+			continue;
+		ram_wait(fuc, addr, 0x0000000f, 0x00000000, 500000);
+	}
+}
+
+static int
+nvc0_ram_calc(struct nouveau_fb *pfb, u32 freq)
+{
+	struct nouveau_clock *clk = nouveau_clock(pfb);
+	struct nouveau_bios *bios = nouveau_bios(pfb);
+	struct nvc0_ram *ram = (void *)pfb->ram;
+	struct nvc0_ramfuc *fuc = &ram->fuc;
+	struct bit_entry M;
+	u8  ver, cnt, strap;
+	u32 data;
+	struct {
+		u32 data;
+		u8  size;
+	} rammap, ramcfg, timing;
+	int ref, div, out;
+	int from, mode;
+	int N1, M1, P;
+	int ret;
+
+	/* lookup memory config data relevant to the target frequency */
+	rammap.data = nvbios_rammap_match(bios, freq / 1000, &ver, &rammap.size,
+					 &cnt, &ramcfg.size);
+	if (!rammap.data || ver != 0x10 || rammap.size < 0x0e) {
+		nv_error(pfb, "invalid/missing rammap entry\n");
+		return -EINVAL;
+	}
+
+	/* locate specific data set for the attached memory */
+	if (bit_entry(bios, 'M', &M) || M.version != 2 || M.length < 3) {
+		nv_error(pfb, "invalid/missing memory table\n");
+		return -EINVAL;
+	}
+
+	strap = (nv_rd32(pfb, 0x101000) & 0x0000003c) >> 2;
+	data = nv_ro16(bios, M.offset + 1);
+	if (data)
+		strap = nv_ro08(bios, data + strap);
+
+	if (strap >= cnt) {
+		nv_error(pfb, "invalid ramcfg strap\n");
+		return -EINVAL;
+	}
+
+	ramcfg.data = rammap.data + rammap.size + (strap * ramcfg.size);
+	if (!ramcfg.data || ver != 0x10 || ramcfg.size < 0x0e) {
+		nv_error(pfb, "invalid/missing ramcfg entry\n");
+		return -EINVAL;
+	}
+
+	/* lookup memory timings, if bios says they're present */
+	strap = nv_ro08(bios, ramcfg.data + 0x01);
+	if (strap != 0xff) {
+		timing.data = nvbios_timing_entry(bios, strap, &ver,
+						 &timing.size);
+		if (!timing.data || ver != 0x10 || timing.size < 0x19) {
+			nv_error(pfb, "invalid/missing timing entry\n");
+			return -EINVAL;
+		}
+	} else {
+		timing.data = 0;
+	}
+
+	ret = ram_init(fuc, pfb);
+	if (ret)
+		return ret;
+
+	/* determine current mclk configuration */
+	from = !!(ram_rd32(fuc, 0x1373f0) & 0x00000002); /*XXX: ok? */
+
+	/* determine target mclk configuration */
+	if (!(ram_rd32(fuc, 0x137300) & 0x00000100))
+		ref = clk->read(clk, nv_clk_src_sppll0);
+	else
+		ref = clk->read(clk, nv_clk_src_sppll1);
+	div = max(min((ref * 2) / freq, (u32)65), (u32)2) - 2;
+	out = (ref * 2) / (div + 2);
+	mode = freq != out;
+
+	ram_mask(fuc, 0x137360, 0x00000002, 0x00000000);
+
+	if ((ram_rd32(fuc, 0x132000) & 0x00000002) || 0 /*XXX*/) {
+		ram_nuke(fuc, 0x132000);
+		ram_mask(fuc, 0x132000, 0x00000002, 0x00000002);
+		ram_mask(fuc, 0x132000, 0x00000002, 0x00000000);
+	}
+
+	if (mode == 1) {
+		ram_nuke(fuc, 0x10fe20);
+		ram_mask(fuc, 0x10fe20, 0x00000002, 0x00000002);
+		ram_mask(fuc, 0x10fe20, 0x00000002, 0x00000000);
+	}
+
+// 0x00020034 // 0x0000000a
+	ram_wr32(fuc, 0x132100, 0x00000001);
+
+	if (mode == 1 && from == 0) {
+		/* calculate refpll */
+		ret = nva3_pll_calc(nv_subdev(pfb), &ram->refpll,
+				    ram->mempll.refclk, &N1, NULL, &M1, &P);
+		if (ret <= 0) {
+			nv_error(pfb, "unable to calc refpll\n");
+			return ret ? ret : -ERANGE;
+		}
+
+		ram_wr32(fuc, 0x10fe20, 0x20010000);
+		ram_wr32(fuc, 0x137320, 0x00000003);
+		ram_wr32(fuc, 0x137330, 0x81200006);
+		ram_wr32(fuc, 0x10fe24, (P << 16) | (N1 << 8) | M1);
+		ram_wr32(fuc, 0x10fe20, 0x20010001);
+		ram_wait(fuc, 0x137390, 0x00020000, 0x00020000, 64000);
+
+		/* calculate mempll */
+		ret = nva3_pll_calc(nv_subdev(pfb), &ram->mempll, freq,
+				   &N1, NULL, &M1, &P);
+		if (ret <= 0) {
+			nv_error(pfb, "unable to calc refpll\n");
+			return ret ? ret : -ERANGE;
+		}
+
+		ram_wr32(fuc, 0x10fe20, 0x20010005);
+		ram_wr32(fuc, 0x132004, (P << 16) | (N1 << 8) | M1);
+		ram_wr32(fuc, 0x132000, 0x18010101);
+		ram_wait(fuc, 0x137390, 0x00000002, 0x00000002, 64000);
+	} else
+	if (mode == 0) {
+		ram_wr32(fuc, 0x137300, 0x00000003);
+	}
+
+	if (from == 0) {
+		ram_nuke(fuc, 0x10fb04);
+		ram_mask(fuc, 0x10fb04, 0x0000ffff, 0x00000000);
+		ram_nuke(fuc, 0x10fb08);
+		ram_mask(fuc, 0x10fb08, 0x0000ffff, 0x00000000);
+		ram_wr32(fuc, 0x10f988, 0x2004ff00);
+		ram_wr32(fuc, 0x10f98c, 0x003fc040);
+		ram_wr32(fuc, 0x10f990, 0x20012001);
+		ram_wr32(fuc, 0x10f998, 0x00011a00);
+		ram_wr32(fuc, 0x13d8f4, 0x00000000);
+	} else {
+		ram_wr32(fuc, 0x10f988, 0x20010000);
+		ram_wr32(fuc, 0x10f98c, 0x00000000);
+		ram_wr32(fuc, 0x10f990, 0x20012001);
+		ram_wr32(fuc, 0x10f998, 0x00010a00);
+	}
+
+	if (from == 0) {
+// 0x00020039 // 0x000000ba
+	}
+
+// 0x0002003a // 0x00000002
+	ram_wr32(fuc, 0x100b0c, 0x00080012);
+// 0x00030014 // 0x00000000 // 0x02b5f070
+// 0x00030014 // 0x00010000 // 0x02b5f070
+	ram_wr32(fuc, 0x611200, 0x00003300);
+// 0x00020034 // 0x0000000a
+// 0x00030020 // 0x00000001 // 0x00000000
+
+	ram_mask(fuc, 0x10f200, 0x00000800, 0x00000000);
+	ram_wr32(fuc, 0x10f210, 0x00000000);
+	ram_nsec(fuc, 1000);
+	if (mode == 0)
+		nvc0_ram_train(fuc, 0x000c1001);
+	ram_wr32(fuc, 0x10f310, 0x00000001);
+	ram_nsec(fuc, 1000);
+	ram_wr32(fuc, 0x10f090, 0x00000061);
+	ram_wr32(fuc, 0x10f090, 0xc000007f);
+	ram_nsec(fuc, 1000);
+
+	if (from == 0) {
+		ram_wr32(fuc, 0x10f824, 0x00007fd4);
+	} else {
+		ram_wr32(fuc, 0x1373ec, 0x00020404);
+	}
+
+	if (mode == 0) {
+		ram_mask(fuc, 0x10f808, 0x00080000, 0x00000000);
+		ram_mask(fuc, 0x10f200, 0x00008000, 0x00008000);
+		ram_wr32(fuc, 0x10f830, 0x41500010);
+		ram_mask(fuc, 0x10f830, 0x01000000, 0x00000000);
+		ram_mask(fuc, 0x132100, 0x00000100, 0x00000100);
+		ram_wr32(fuc, 0x10f050, 0xff000090);
+		ram_wr32(fuc, 0x1373ec, 0x00020f0f);
+		ram_wr32(fuc, 0x1373f0, 0x00000003);
+		ram_wr32(fuc, 0x137310, 0x81201616);
+		ram_wr32(fuc, 0x132100, 0x00000001);
+// 0x00020039 // 0x000000ba
+		ram_wr32(fuc, 0x10f830, 0x00300017);
+		ram_wr32(fuc, 0x1373f0, 0x00000001);
+		ram_wr32(fuc, 0x10f824, 0x00007e77);
+		ram_wr32(fuc, 0x132000, 0x18030001);
+		ram_wr32(fuc, 0x10f090, 0x4000007e);
+		ram_nsec(fuc, 2000);
+		ram_wr32(fuc, 0x10f314, 0x00000001);
+		ram_wr32(fuc, 0x10f210, 0x80000000);
+		ram_wr32(fuc, 0x10f338, 0x00300220);
+		ram_wr32(fuc, 0x10f300, 0x0000011d);
+		ram_nsec(fuc, 1000);
+		ram_wr32(fuc, 0x10f290, 0x02060505);
+		ram_wr32(fuc, 0x10f294, 0x34208288);
+		ram_wr32(fuc, 0x10f298, 0x44050411);
+		ram_wr32(fuc, 0x10f29c, 0x0000114c);
+		ram_wr32(fuc, 0x10f2a0, 0x42e10069);
+		ram_wr32(fuc, 0x10f614, 0x40044f77);
+		ram_wr32(fuc, 0x10f610, 0x40044f77);
+		ram_wr32(fuc, 0x10f344, 0x00600009);
+		ram_nsec(fuc, 1000);
+		ram_wr32(fuc, 0x10f348, 0x00700008);
+		ram_wr32(fuc, 0x61c140, 0x19240000);
+		ram_wr32(fuc, 0x10f830, 0x00300017);
+		nvc0_ram_train(fuc, 0x80021001);
+		nvc0_ram_train(fuc, 0x80081001);
+		ram_wr32(fuc, 0x10f340, 0x00500004);
+		ram_nsec(fuc, 1000);
+		ram_wr32(fuc, 0x10f830, 0x01300017);
+		ram_wr32(fuc, 0x10f830, 0x00300017);
+// 0x00030020 // 0x00000000 // 0x00000000
+// 0x00020034 // 0x0000000b
+		ram_wr32(fuc, 0x100b0c, 0x00080028);
+		ram_wr32(fuc, 0x611200, 0x00003330);
+	} else {
+		ram_wr32(fuc, 0x10f800, 0x00001800);
+		ram_wr32(fuc, 0x13d8f4, 0x00000000);
+		ram_wr32(fuc, 0x1373ec, 0x00020404);
+		ram_wr32(fuc, 0x1373f0, 0x00000003);
+		ram_wr32(fuc, 0x10f830, 0x40700010);
+		ram_wr32(fuc, 0x10f830, 0x40500010);
+		ram_wr32(fuc, 0x13d8f4, 0x00000000);
+		ram_wr32(fuc, 0x1373f8, 0x00000000);
+		ram_wr32(fuc, 0x132100, 0x00000101);
+		ram_wr32(fuc, 0x137310, 0x89201616);
+		ram_wr32(fuc, 0x10f050, 0xff000090);
+		ram_wr32(fuc, 0x1373ec, 0x00030404);
+		ram_wr32(fuc, 0x1373f0, 0x00000002);
+	// 0x00020039 // 0x00000011
+		ram_wr32(fuc, 0x132100, 0x00000001);
+		ram_wr32(fuc, 0x1373f8, 0x00002000);
+		ram_nsec(fuc, 2000);
+		ram_wr32(fuc, 0x10f808, 0x7aaa0050);
+		ram_wr32(fuc, 0x10f830, 0x00500010);
+		ram_wr32(fuc, 0x10f200, 0x00ce1000);
+		ram_wr32(fuc, 0x10f090, 0x4000007e);
+		ram_nsec(fuc, 2000);
+		ram_wr32(fuc, 0x10f314, 0x00000001);
+		ram_wr32(fuc, 0x10f210, 0x80000000);
+		ram_wr32(fuc, 0x10f338, 0x00300200);
+		ram_wr32(fuc, 0x10f300, 0x0000084d);
+		ram_nsec(fuc, 1000);
+		ram_wr32(fuc, 0x10f290, 0x0b343825);
+		ram_wr32(fuc, 0x10f294, 0x3483028e);
+		ram_wr32(fuc, 0x10f298, 0x440c0600);
+		ram_wr32(fuc, 0x10f29c, 0x0000214c);
+		ram_wr32(fuc, 0x10f2a0, 0x42e20069);
+		ram_wr32(fuc, 0x10f200, 0x00ce0000);
+		ram_wr32(fuc, 0x10f614, 0x60044e77);
+		ram_wr32(fuc, 0x10f610, 0x60044e77);
+		ram_wr32(fuc, 0x10f340, 0x00500000);
+		ram_nsec(fuc, 1000);
+		ram_wr32(fuc, 0x10f344, 0x00600228);
+		ram_nsec(fuc, 1000);
+		ram_wr32(fuc, 0x10f348, 0x00700000);
+		ram_wr32(fuc, 0x13d8f4, 0x00000000);
+		ram_wr32(fuc, 0x61c140, 0x09a40000);
+
+		nvc0_ram_train(fuc, 0x800e1008);
+
+		ram_nsec(fuc, 1000);
+		ram_wr32(fuc, 0x10f800, 0x00001804);
+	// 0x00030020 // 0x00000000 // 0x00000000
+	// 0x00020034 // 0x0000000b
+		ram_wr32(fuc, 0x13d8f4, 0x00000000);
+		ram_wr32(fuc, 0x100b0c, 0x00080028);
+		ram_wr32(fuc, 0x611200, 0x00003330);
+		ram_nsec(fuc, 100000);
+		ram_wr32(fuc, 0x10f9b0, 0x05313f41);
+		ram_wr32(fuc, 0x10f9b4, 0x00002f50);
+
+		nvc0_ram_train(fuc, 0x010c1001);
+	}
+
+	ram_mask(fuc, 0x10f200, 0x00000800, 0x00000800);
+// 0x00020016 // 0x00000000
+
+	if (mode == 0)
+		ram_mask(fuc, 0x132000, 0x00000001, 0x00000000);
+	return 0;
+}
+
+static int
+nvc0_ram_prog(struct nouveau_fb *pfb)
+{
+	struct nouveau_device *device = nv_device(pfb);
+	struct nvc0_ram *ram = (void *)pfb->ram;
+	struct nvc0_ramfuc *fuc = &ram->fuc;
+	ram_exec(fuc, nouveau_boolopt(device->cfgopt, "NvMemExec", false));
+	return 0;
+}
+
+static void
+nvc0_ram_tidy(struct nouveau_fb *pfb)
+{
+	struct nvc0_ram *ram = (void *)pfb->ram;
+	struct nvc0_ramfuc *fuc = &ram->fuc;
+	ram_exec(fuc, false);
+}
 
 extern const u8 nvc0_pte_storage_type_map[256];
 
@@ -110,10 +515,9 @@ nvc0_ram_get(struct nouveau_fb *pfb, u64 size, u32 align, u32 ncmin,
 	return 0;
 }
 
-static int
-nvc0_ram_create(struct nouveau_object *parent, struct nouveau_object *engine,
-		struct nouveau_oclass *oclass, void *data, u32 size,
-		struct nouveau_object **pobject)
+int
+nvc0_ram_create_(struct nouveau_object *parent, struct nouveau_object *engine,
+		 struct nouveau_oclass *oclass, int size, void **pobject)
 {
 	struct nouveau_fb *pfb = nouveau_fb(parent);
 	struct nouveau_bios *bios = nouveau_bios(pfb);
@@ -127,8 +531,8 @@ nvc0_ram_create(struct nouveau_object *parent, struct nouveau_object *engine,
 	bool uniform = true;
 	int ret, part;
 
-	ret = nouveau_ram_create(parent, engine, oclass, &ram);
-	*pobject = nv_object(ram);
+	ret = nouveau_ram_create_(parent, engine, oclass, size, pobject);
+	ram = *pobject;
 	if (ret)
 		return ret;
 
@@ -182,13 +586,158 @@ nvc0_ram_create(struct nouveau_object *parent, struct nouveau_object *engine,
 	return 0;
 }
 
+static int
+nvc0_ram_init(struct nouveau_object *object)
+{
+	struct nouveau_fb *pfb = (void *)object->parent;
+	struct nvc0_ram   *ram = (void *)object;
+	int ret, i;
+
+	ret = nouveau_ram_init(&ram->base);
+	if (ret)
+		return ret;
+
+	/* prepare for ddr link training, and load training patterns */
+	switch (ram->base.type) {
+	case NV_MEM_TYPE_GDDR5: {
+		static const u8  train0[] = {
+			0x00, 0xff, 0x55, 0xaa, 0x33, 0xcc,
+			0x00, 0xff, 0xff, 0x00, 0xff, 0x00,
+		};
+		static const u32 train1[] = {
+			0x00000000, 0xffffffff,
+			0x55555555, 0xaaaaaaaa,
+			0x33333333, 0xcccccccc,
+			0xf0f0f0f0, 0x0f0f0f0f,
+			0x00ff00ff, 0xff00ff00,
+			0x0000ffff, 0xffff0000,
+		};
+
+		for (i = 0; i < 0x30; i++) {
+			nv_wr32(pfb, 0x10f968, 0x00000000 | (i << 8));
+			nv_wr32(pfb, 0x10f96c, 0x00000000 | (i << 8));
+			nv_wr32(pfb, 0x10f920, 0x00000100 | train0[i % 12]);
+			nv_wr32(pfb, 0x10f924, 0x00000100 | train0[i % 12]);
+			nv_wr32(pfb, 0x10f918,              train1[i % 12]);
+			nv_wr32(pfb, 0x10f91c,              train1[i % 12]);
+			nv_wr32(pfb, 0x10f920, 0x00000000 | train0[i % 12]);
+			nv_wr32(pfb, 0x10f924, 0x00000000 | train0[i % 12]);
+			nv_wr32(pfb, 0x10f918,              train1[i % 12]);
+			nv_wr32(pfb, 0x10f91c,              train1[i % 12]);
+		}
+	}	break;
+	default:
+		break;
+	}
+
+	return 0;
+}
+
+static int
+nvc0_ram_ctor(struct nouveau_object *parent, struct nouveau_object *engine,
+	      struct nouveau_oclass *oclass, void *data, u32 size,
+	      struct nouveau_object **pobject)
+{
+	struct nouveau_bios *bios = nouveau_bios(parent);
+	struct nvc0_ram *ram;
+	int ret;
+
+	ret = nvc0_ram_create(parent, engine, oclass, &ram);
+	*pobject = nv_object(ram);
+	if (ret)
+		return ret;
+
+	ret = nvbios_pll_parse(bios, 0x0c, &ram->refpll);
+	if (ret) {
+		nv_error(ram, "mclk refpll data not found\n");
+		return ret;
+	}
+
+	ret = nvbios_pll_parse(bios, 0x04, &ram->mempll);
+	if (ret) {
+		nv_error(ram, "mclk pll data not found\n");
+		return ret;
+	}
+
+	switch (ram->base.type) {
+	case NV_MEM_TYPE_GDDR5:
+		ram->base.calc = nvc0_ram_calc;
+		ram->base.prog = nvc0_ram_prog;
+		ram->base.tidy = nvc0_ram_tidy;
+		break;
+	default:
+		nv_warn(ram, "reclocking of this ram type unsupported\n");
+		return 0;
+	}
+
+	ram->fuc.r_0x10fe20 = ramfuc_reg(0x10fe20);
+	ram->fuc.r_0x10fe24 = ramfuc_reg(0x10fe24);
+	ram->fuc.r_0x137320 = ramfuc_reg(0x137320);
+	ram->fuc.r_0x137330 = ramfuc_reg(0x137330);
+
+	ram->fuc.r_0x132000 = ramfuc_reg(0x132000);
+	ram->fuc.r_0x132004 = ramfuc_reg(0x132004);
+	ram->fuc.r_0x132100 = ramfuc_reg(0x132100);
+
+	ram->fuc.r_0x137390 = ramfuc_reg(0x137390);
+
+	ram->fuc.r_0x10f290 = ramfuc_reg(0x10f290);
+	ram->fuc.r_0x10f294 = ramfuc_reg(0x10f294);
+	ram->fuc.r_0x10f298 = ramfuc_reg(0x10f298);
+	ram->fuc.r_0x10f29c = ramfuc_reg(0x10f29c);
+	ram->fuc.r_0x10f2a0 = ramfuc_reg(0x10f2a0);
+
+	ram->fuc.r_0x10f300 = ramfuc_reg(0x10f300);
+	ram->fuc.r_0x10f338 = ramfuc_reg(0x10f338);
+	ram->fuc.r_0x10f340 = ramfuc_reg(0x10f340);
+	ram->fuc.r_0x10f344 = ramfuc_reg(0x10f344);
+	ram->fuc.r_0x10f348 = ramfuc_reg(0x10f348);
+
+	ram->fuc.r_0x10f910 = ramfuc_reg(0x10f910);
+	ram->fuc.r_0x10f914 = ramfuc_reg(0x10f914);
+
+	ram->fuc.r_0x100b0c = ramfuc_reg(0x100b0c);
+	ram->fuc.r_0x10f050 = ramfuc_reg(0x10f050);
+	ram->fuc.r_0x10f090 = ramfuc_reg(0x10f090);
+	ram->fuc.r_0x10f200 = ramfuc_reg(0x10f200);
+	ram->fuc.r_0x10f210 = ramfuc_reg(0x10f210);
+	ram->fuc.r_0x10f310 = ramfuc_reg(0x10f310);
+	ram->fuc.r_0x10f314 = ramfuc_reg(0x10f314);
+	ram->fuc.r_0x10f610 = ramfuc_reg(0x10f610);
+	ram->fuc.r_0x10f614 = ramfuc_reg(0x10f614);
+	ram->fuc.r_0x10f800 = ramfuc_reg(0x10f800);
+	ram->fuc.r_0x10f808 = ramfuc_reg(0x10f808);
+	ram->fuc.r_0x10f824 = ramfuc_reg(0x10f824);
+	ram->fuc.r_0x10f830 = ramfuc_reg(0x10f830);
+	ram->fuc.r_0x10f988 = ramfuc_reg(0x10f988);
+	ram->fuc.r_0x10f98c = ramfuc_reg(0x10f98c);
+	ram->fuc.r_0x10f990 = ramfuc_reg(0x10f990);
+	ram->fuc.r_0x10f998 = ramfuc_reg(0x10f998);
+	ram->fuc.r_0x10f9b0 = ramfuc_reg(0x10f9b0);
+	ram->fuc.r_0x10f9b4 = ramfuc_reg(0x10f9b4);
+	ram->fuc.r_0x10fb04 = ramfuc_reg(0x10fb04);
+	ram->fuc.r_0x10fb08 = ramfuc_reg(0x10fb08);
+	ram->fuc.r_0x137310 = ramfuc_reg(0x137300);
+	ram->fuc.r_0x137310 = ramfuc_reg(0x137310);
+	ram->fuc.r_0x137360 = ramfuc_reg(0x137360);
+	ram->fuc.r_0x1373ec = ramfuc_reg(0x1373ec);
+	ram->fuc.r_0x1373f0 = ramfuc_reg(0x1373f0);
+	ram->fuc.r_0x1373f8 = ramfuc_reg(0x1373f8);
+
+	ram->fuc.r_0x61c140 = ramfuc_reg(0x61c140);
+	ram->fuc.r_0x611200 = ramfuc_reg(0x611200);
+
+	ram->fuc.r_0x13d8f4 = ramfuc_reg(0x13d8f4);
+	return 0;
+}
+
 struct nouveau_oclass
 nvc0_ram_oclass = {
 	.handle = 0,
 	.ofuncs = &(struct nouveau_ofuncs) {
-		.ctor = nvc0_ram_create,
+		.ctor = nvc0_ram_ctor,
 		.dtor = _nouveau_ram_dtor,
-		.init = _nouveau_ram_init,
+		.init = nvc0_ram_init,
 		.fini = _nouveau_ram_fini,
 	}
 };
diff --git a/drivers/gpu/drm/nouveau/core/subdev/fb/ramnve0.c b/drivers/gpu/drm/nouveau/core/subdev/fb/ramnve0.c
new file mode 100644
index 000000000000..bc86cfd084f6
--- /dev/null
+++ b/drivers/gpu/drm/nouveau/core/subdev/fb/ramnve0.c
@@ -0,0 +1,1264 @@
+/*
+ * Copyright 2013 Red Hat Inc.
+ *
+ * Permission is hereby granted, free of charge, to any person obtaining a
+ * copy of this software and associated documentation files (the "Software"),
+ * to deal in the Software without restriction, including without limitation
+ * the rights to use, copy, modify, merge, publish, distribute, sublicense,
+ * and/or sell copies of the Software, and to permit persons to whom the
+ * Software is furnished to do so, subject to the following conditions:
+ *
+ * The above copyright notice and this permission notice shall be included in
+ * all copies or substantial portions of the Software.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+ * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+ * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT.  IN NO EVENT SHALL
+ * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR
+ * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE,
+ * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR
+ * OTHER DEALINGS IN THE SOFTWARE.
+ *
+ * Authors: Ben Skeggs
+ */
+
+#include <subdev/gpio.h>
+
+#include <subdev/bios.h>
+#include <subdev/bios/bit.h>
+#include <subdev/bios/pll.h>
+#include <subdev/bios/init.h>
+#include <subdev/bios/rammap.h>
+#include <subdev/bios/timing.h>
+
+#include <subdev/clock.h>
+#include <subdev/clock/pll.h>
+
+#include <subdev/timer.h>
+
+#include <core/option.h>
+
+#include "nvc0.h"
+
+#include "ramfuc.h"
+
+struct nve0_ramfuc {
+	struct ramfuc base;
+
+	struct nvbios_pll refpll;
+	struct nvbios_pll mempll;
+
+	struct ramfuc_reg r_gpioMV;
+	u32 r_funcMV[2];
+	struct ramfuc_reg r_gpio2E;
+	u32 r_func2E[2];
+	struct ramfuc_reg r_gpiotrig;
+
+	struct ramfuc_reg r_0x132020;
+	struct ramfuc_reg r_0x132028;
+	struct ramfuc_reg r_0x132024;
+	struct ramfuc_reg r_0x132030;
+	struct ramfuc_reg r_0x132034;
+	struct ramfuc_reg r_0x132000;
+	struct ramfuc_reg r_0x132004;
+	struct ramfuc_reg r_0x132040;
+
+	struct ramfuc_reg r_0x10f248;
+	struct ramfuc_reg r_0x10f290;
+	struct ramfuc_reg r_0x10f294;
+	struct ramfuc_reg r_0x10f298;
+	struct ramfuc_reg r_0x10f29c;
+	struct ramfuc_reg r_0x10f2a0;
+	struct ramfuc_reg r_0x10f2a4;
+	struct ramfuc_reg r_0x10f2a8;
+	struct ramfuc_reg r_0x10f2ac;
+	struct ramfuc_reg r_0x10f2cc;
+	struct ramfuc_reg r_0x10f2e8;
+	struct ramfuc_reg r_0x10f250;
+	struct ramfuc_reg r_0x10f24c;
+	struct ramfuc_reg r_0x10fec4;
+	struct ramfuc_reg r_0x10fec8;
+	struct ramfuc_reg r_0x10f604;
+	struct ramfuc_reg r_0x10f614;
+	struct ramfuc_reg r_0x10f610;
+	struct ramfuc_reg r_0x100770;
+	struct ramfuc_reg r_0x100778;
+	struct ramfuc_reg r_0x10f224;
+
+	struct ramfuc_reg r_0x10f870;
+	struct ramfuc_reg r_0x10f698;
+	struct ramfuc_reg r_0x10f694;
+	struct ramfuc_reg r_0x10f6b8;
+	struct ramfuc_reg r_0x10f808;
+	struct ramfuc_reg r_0x10f670;
+	struct ramfuc_reg r_0x10f60c;
+	struct ramfuc_reg r_0x10f830;
+	struct ramfuc_reg r_0x1373ec;
+	struct ramfuc_reg r_0x10f800;
+	struct ramfuc_reg r_0x10f82c;
+
+	struct ramfuc_reg r_0x10f978;
+	struct ramfuc_reg r_0x10f910;
+	struct ramfuc_reg r_0x10f914;
+
+	struct ramfuc_reg r_mr[16]; /* MR0 - MR8, MR15 */
+
+	struct ramfuc_reg r_0x62c000;
+	struct ramfuc_reg r_0x10f200;
+	struct ramfuc_reg r_0x10f210;
+	struct ramfuc_reg r_0x10f310;
+	struct ramfuc_reg r_0x10f314;
+	struct ramfuc_reg r_0x10f318;
+	struct ramfuc_reg r_0x10f090;
+	struct ramfuc_reg r_0x10f69c;
+	struct ramfuc_reg r_0x10f824;
+	struct ramfuc_reg r_0x1373f0;
+	struct ramfuc_reg r_0x1373f4;
+	struct ramfuc_reg r_0x137320;
+	struct ramfuc_reg r_0x10f65c;
+	struct ramfuc_reg r_0x10f6bc;
+	struct ramfuc_reg r_0x100710;
+	struct ramfuc_reg r_0x10f750;
+};
+
+struct nve0_ram {
+	struct nouveau_ram base;
+	struct nve0_ramfuc fuc;
+	int from;
+	int mode;
+	int N1, fN1, M1, P1;
+	int N2, M2, P2;
+};
+
+/*******************************************************************************
+ * GDDR5
+ ******************************************************************************/
+static void
+train(struct nve0_ramfuc *fuc, u32 magic)
+{
+	struct nve0_ram *ram = container_of(fuc, typeof(*ram), fuc);
+	struct nouveau_fb *pfb = nouveau_fb(ram);
+	const int mc = nv_rd32(pfb, 0x02243c);
+	int i;
+
+	ram_mask(fuc, 0x10f910, 0xbc0e0000, magic);
+	ram_mask(fuc, 0x10f914, 0xbc0e0000, magic);
+	for (i = 0; i < mc; i++) {
+		const u32 addr = 0x110974 + (i * 0x1000);
+		ram_wait(fuc, addr, 0x0000000f, 0x00000000, 500000);
+	}
+}
+
+static void
+r1373f4_init(struct nve0_ramfuc *fuc)
+{
+	struct nve0_ram *ram = container_of(fuc, typeof(*ram), fuc);
+	const u32 mcoef = ((--ram->P2 << 28) | (ram->N2 << 8) | ram->M2);
+	const u32 rcoef = ((  ram->P1 << 16) | (ram->N1 << 8) | ram->M1);
+	const u32 runk0 = ram->fN1 << 16;
+	const u32 runk1 = ram->fN1;
+
+	if (ram->from == 2) {
+		ram_mask(fuc, 0x1373f4, 0x00000000, 0x00001100);
+		ram_mask(fuc, 0x1373f4, 0x00000000, 0x00000010);
+	} else {
+		ram_mask(fuc, 0x1373f4, 0x00000000, 0x00010010);
+	}
+
+	ram_mask(fuc, 0x1373f4, 0x00000003, 0x00000000);
+	ram_mask(fuc, 0x1373f4, 0x00000010, 0x00000000);
+
+	/* (re)program refpll, if required */
+	if ((ram_rd32(fuc, 0x132024) & 0xffffffff) != rcoef ||
+	    (ram_rd32(fuc, 0x132034) & 0x0000ffff) != runk1) {
+		ram_mask(fuc, 0x132000, 0x00000001, 0x00000000);
+		ram_mask(fuc, 0x132020, 0x00000001, 0x00000000);
+		ram_wr32(fuc, 0x137320, 0x00000000);
+		ram_mask(fuc, 0x132030, 0xffff0000, runk0);
+		ram_mask(fuc, 0x132034, 0x0000ffff, runk1);
+		ram_wr32(fuc, 0x132024, rcoef);
+		ram_mask(fuc, 0x132028, 0x00080000, 0x00080000);
+		ram_mask(fuc, 0x132020, 0x00000001, 0x00000001);
+		ram_wait(fuc, 0x137390, 0x00020000, 0x00020000, 64000);
+		ram_mask(fuc, 0x132028, 0x00080000, 0x00000000);
+	}
+
+	/* (re)program mempll, if required */
+	if (ram->mode == 2) {
+		ram_mask(fuc, 0x1373f4, 0x00010000, 0x00000000);
+		ram_mask(fuc, 0x132000, 0x00000001, 0x00000000);
+		ram_mask(fuc, 0x132004, 0x103fffff, mcoef);
+		ram_mask(fuc, 0x132000, 0x00000001, 0x00000001);
+		ram_wait(fuc, 0x137390, 0x00000002, 0x00000002, 64000);
+		ram_mask(fuc, 0x1373f4, 0x00000000, 0x00001100);
+	} else {
+		ram_mask(fuc, 0x1373f4, 0x00000000, 0x00010100);
+	}
+
+	ram_mask(fuc, 0x1373f4, 0x00000000, 0x00000010);
+}
+
+static void
+r1373f4_fini(struct nve0_ramfuc *fuc, u32 ramcfg)
+{
+	struct nve0_ram *ram = container_of(fuc, typeof(*ram), fuc);
+	struct nouveau_bios *bios = nouveau_bios(ram);
+	u8 v0 = (nv_ro08(bios, ramcfg + 0x03) & 0xc0) >> 6;
+	u8 v1 = (nv_ro08(bios, ramcfg + 0x03) & 0x30) >> 4;
+	u32 tmp;
+
+	tmp = ram_rd32(fuc, 0x1373ec) & ~0x00030000;
+	ram_wr32(fuc, 0x1373ec, tmp | (v1 << 16));
+	ram_mask(fuc, 0x1373f0, (~ram->mode & 3), 0x00000000);
+	if (ram->mode == 2) {
+		ram_mask(fuc, 0x1373f4, 0x00000003, 0x000000002);
+		ram_mask(fuc, 0x1373f4, 0x00001100, 0x000000000);
+	} else {
+		ram_mask(fuc, 0x1373f4, 0x00000003, 0x000000001);
+		ram_mask(fuc, 0x1373f4, 0x00010000, 0x000000000);
+	}
+	ram_mask(fuc, 0x10f800, 0x00000030, (v0 ^ v1) << 4);
+}
+
+static int
+nve0_ram_calc_gddr5(struct nouveau_fb *pfb, u32 freq)
+{
+	struct nouveau_bios *bios = nouveau_bios(pfb);
+	struct nve0_ram *ram = (void *)pfb->ram;
+	struct nve0_ramfuc *fuc = &ram->fuc;
+	const u32 rammap = ram->base.rammap.data;
+	const u32 ramcfg = ram->base.ramcfg.data;
+	const u32 timing = ram->base.timing.data;
+	int vc = !(nv_ro08(bios, ramcfg + 0x02) & 0x08);
+	int mv = 1; /*XXX*/
+	u32 mask, data;
+
+	ram_mask(fuc, 0x10f808, 0x40000000, 0x40000000);
+	ram_wr32(fuc, 0x62c000, 0x0f0f0000);
+
+	/* MR1: turn termination on early, for some reason.. */
+	if ((ram->base.mr[1] & 0x03c) != 0x030)
+		ram_mask(fuc, mr[1], 0x03c, ram->base.mr[1] & 0x03c);
+
+	if (vc == 1 && ram_have(fuc, gpio2E)) {
+		u32 temp  = ram_mask(fuc, gpio2E, 0x3000, fuc->r_func2E[1]);
+		if (temp != ram_rd32(fuc, gpio2E)) {
+			ram_wr32(fuc, gpiotrig, 1);
+			ram_nsec(fuc, 20000);
+		}
+	}
+
+	ram_mask(fuc, 0x10f200, 0x00000800, 0x00000000);
+
+	ram_mask(fuc, 0x10f914, 0x01020000, 0x000c0000);
+	ram_mask(fuc, 0x10f910, 0x01020000, 0x000c0000);
+
+	ram_wr32(fuc, 0x10f210, 0x00000000); /* REFRESH_AUTO = 0 */
+	ram_nsec(fuc, 1000);
+	ram_wr32(fuc, 0x10f310, 0x00000001); /* REFRESH */
+	ram_nsec(fuc, 1000);
+
+	ram_mask(fuc, 0x10f200, 0x80000000, 0x80000000);
+	ram_wr32(fuc, 0x10f314, 0x00000001); /* PRECHARGE */
+	ram_mask(fuc, 0x10f200, 0x80000000, 0x00000000);
+	ram_wr32(fuc, 0x10f090, 0x00000061);
+	ram_wr32(fuc, 0x10f090, 0xc000007f);
+	ram_nsec(fuc, 1000);
+
+	ram_wr32(fuc, 0x10f698, 0x00000000);
+	ram_wr32(fuc, 0x10f69c, 0x00000000);
+
+	/*XXX: there does appear to be some kind of condition here, simply
+	 *     modifying these bits in the vbios from the default pl0
+	 *     entries shows no change.  however, the data does appear to
+	 *     be correct and may be required for the transition back
+	 */
+	mask = 0x800f07e0;
+	data = 0x00030000;
+	if (ram_rd32(fuc, 0x10f978) & 0x00800000)
+		data |= 0x00040000;
+
+	if (1) {
+		data |= 0x800807e0;
+		switch (nv_ro08(bios, ramcfg + 0x03) & 0xc0) {
+		case 0xc0: data &= ~0x00000040; break;
+		case 0x80: data &= ~0x00000100; break;
+		case 0x40: data &= ~0x80000000; break;
+		case 0x00: data &= ~0x00000400; break;
+		}
+
+		switch (nv_ro08(bios, ramcfg + 0x03) & 0x30) {
+		case 0x30: data &= ~0x00000020; break;
+		case 0x20: data &= ~0x00000080; break;
+		case 0x10: data &= ~0x00080000; break;
+		case 0x00: data &= ~0x00000200; break;
+		}
+	}
+
+	if (nv_ro08(bios, ramcfg + 0x02) & 0x80)
+		mask |= 0x03000000;
+	if (nv_ro08(bios, ramcfg + 0x02) & 0x40)
+		mask |= 0x00002000;
+	if (nv_ro08(bios, ramcfg + 0x07) & 0x10)
+		mask |= 0x00004000;
+	if (nv_ro08(bios, ramcfg + 0x07) & 0x08)
+		mask |= 0x00000003;
+	else {
+		mask |= 0x34000000;
+		if (ram_rd32(fuc, 0x10f978) & 0x00800000)
+			mask |= 0x40000000;
+	}
+	ram_mask(fuc, 0x10f824, mask, data);
+
+	ram_mask(fuc, 0x132040, 0x00010000, 0x00000000);
+
+	if (ram->from == 2 && ram->mode != 2) {
+		ram_mask(fuc, 0x10f808, 0x00080000, 0x00000000);
+		ram_mask(fuc, 0x10f200, 0x00008000, 0x00008000);
+		ram_mask(fuc, 0x10f800, 0x00000000, 0x00000004);
+		ram_mask(fuc, 0x10f830, 0x00008000, 0x01040010);
+		ram_mask(fuc, 0x10f830, 0x01000000, 0x00000000);
+		r1373f4_init(fuc);
+		ram_mask(fuc, 0x1373f0, 0x00000002, 0x00000001);
+		r1373f4_fini(fuc, ramcfg);
+		ram_mask(fuc, 0x10f830, 0x00c00000, 0x00240001);
+	} else
+	if (ram->from != 2 && ram->mode != 2) {
+		r1373f4_init(fuc);
+		r1373f4_fini(fuc, ramcfg);
+	}
+
+	if (ram_have(fuc, gpioMV)) {
+		u32 temp  = ram_mask(fuc, gpioMV, 0x3000, fuc->r_funcMV[mv]);
+		if (temp != ram_rd32(fuc, gpioMV)) {
+			ram_wr32(fuc, gpiotrig, 1);
+			ram_nsec(fuc, 64000);
+		}
+	}
+
+	if ( (nv_ro08(bios, ramcfg + 0x02) & 0x40) ||
+	     (nv_ro08(bios, ramcfg + 0x07) & 0x10)) {
+		ram_mask(fuc, 0x132040, 0x00010000, 0x00010000);
+		ram_nsec(fuc, 20000);
+	}
+
+	if (ram->from != 2 && ram->mode == 2) {
+		ram_mask(fuc, 0x10f800, 0x00000004, 0x00000000);
+		ram_mask(fuc, 0x1373f0, 0x00000000, 0x00000002);
+		ram_mask(fuc, 0x10f830, 0x00800001, 0x00408010);
+		r1373f4_init(fuc);
+		r1373f4_fini(fuc, ramcfg);
+		ram_mask(fuc, 0x10f808, 0x00000000, 0x00080000);
+		ram_mask(fuc, 0x10f200, 0x00808000, 0x00800000);
+	} else
+	if (ram->from == 2 && ram->mode == 2) {
+		ram_mask(fuc, 0x10f800, 0x00000004, 0x00000000);
+		r1373f4_init(fuc);
+		r1373f4_fini(fuc, ramcfg);
+	}
+
+	if (ram->mode != 2) /*XXX*/ {
+		if (nv_ro08(bios, ramcfg + 0x07) & 0x40)
+			ram_mask(fuc, 0x10f670, 0x80000000, 0x80000000);
+	}
+
+	data = (nv_ro08(bios, rammap + 0x11) & 0x0c) >> 2;
+	ram_wr32(fuc, 0x10f65c, 0x00000011 * data);
+	ram_wr32(fuc, 0x10f6b8, 0x01010101 * nv_ro08(bios, ramcfg + 0x09));
+	ram_wr32(fuc, 0x10f6bc, 0x01010101 * nv_ro08(bios, ramcfg + 0x09));
+
+	data = nv_ro08(bios, ramcfg + 0x04);
+	if (!(nv_ro08(bios, ramcfg + 0x07) & 0x08)) {
+		ram_wr32(fuc, 0x10f698, 0x01010101 * data);
+		ram_wr32(fuc, 0x10f69c, 0x01010101 * data);
+	}
+
+	if (ram->mode != 2) {
+		u32 temp = ram_rd32(fuc, 0x10f694) & ~0xff00ff00;
+		ram_wr32(fuc, 0x10f694, temp | (0x01000100 * data));
+	}
+
+	if (ram->mode == 2 && (nv_ro08(bios, ramcfg + 0x08) & 0x10))
+		data = 0x00000080;
+	else
+		data = 0x00000000;
+	ram_mask(fuc, 0x10f60c, 0x00000080, data);
+
+	mask = 0x00070000;
+	data = 0x00000000;
+	if (!(nv_ro08(bios, ramcfg + 0x02) & 0x80))
+		data |= 0x03000000;
+	if (!(nv_ro08(bios, ramcfg + 0x02) & 0x40))
+		data |= 0x00002000;
+	if (!(nv_ro08(bios, ramcfg + 0x07) & 0x10))
+		data |= 0x00004000;
+	if (!(nv_ro08(bios, ramcfg + 0x07) & 0x08))
+		data |= 0x00000003;
+	else
+		data |= 0x74000000;
+	ram_mask(fuc, 0x10f824, mask, data);
+
+	if (nv_ro08(bios, ramcfg + 0x01) & 0x08)
+		data = 0x00000000;
+	else
+		data = 0x00001000;
+	ram_mask(fuc, 0x10f200, 0x00001000, data);
+
+	if (ram_rd32(fuc, 0x10f670) & 0x80000000) {
+		ram_nsec(fuc, 10000);
+		ram_mask(fuc, 0x10f670, 0x80000000, 0x00000000);
+	}
+
+	if (nv_ro08(bios, ramcfg + 0x08) & 0x01)
+		data = 0x00100000;
+	else
+		data = 0x00000000;
+	ram_mask(fuc, 0x10f82c, 0x00100000, data);
+
+	data = 0x00000000;
+	if (nv_ro08(bios, ramcfg + 0x08) & 0x08)
+		data |= 0x00002000;
+	if (nv_ro08(bios, ramcfg + 0x08) & 0x04)
+		data |= 0x00001000;
+	if (nv_ro08(bios, ramcfg + 0x08) & 0x02)
+		data |= 0x00004000;
+	ram_mask(fuc, 0x10f830, 0x00007000, data);
+
+	/* PFB timing */
+	ram_mask(fuc, 0x10f248, 0xffffffff, nv_ro32(bios, timing + 0x28));
+	ram_mask(fuc, 0x10f290, 0xffffffff, nv_ro32(bios, timing + 0x00));
+	ram_mask(fuc, 0x10f294, 0xffffffff, nv_ro32(bios, timing + 0x04));
+	ram_mask(fuc, 0x10f298, 0xffffffff, nv_ro32(bios, timing + 0x08));
+	ram_mask(fuc, 0x10f29c, 0xffffffff, nv_ro32(bios, timing + 0x0c));
+	ram_mask(fuc, 0x10f2a0, 0xffffffff, nv_ro32(bios, timing + 0x10));
+	ram_mask(fuc, 0x10f2a4, 0xffffffff, nv_ro32(bios, timing + 0x14));
+	ram_mask(fuc, 0x10f2a8, 0xffffffff, nv_ro32(bios, timing + 0x18));
+	ram_mask(fuc, 0x10f2ac, 0xffffffff, nv_ro32(bios, timing + 0x1c));
+	ram_mask(fuc, 0x10f2cc, 0xffffffff, nv_ro32(bios, timing + 0x20));
+	ram_mask(fuc, 0x10f2e8, 0xffffffff, nv_ro32(bios, timing + 0x24));
+
+	data = (nv_ro08(bios, ramcfg + 0x02) & 0x03) << 8;
+	if (nv_ro08(bios, ramcfg + 0x01) & 0x10)
+		data |= 0x70000000;
+	ram_mask(fuc, 0x10f604, 0x70000300, data);
+
+	data = (nv_ro08(bios, timing + 0x30) & 0x07) << 28;
+	if (nv_ro08(bios, ramcfg + 0x01) & 0x01)
+		data |= 0x00000100;
+	ram_mask(fuc, 0x10f614, 0x70000000, data);
+
+	data = (nv_ro08(bios, timing + 0x30) & 0x07) << 28;
+	if (nv_ro08(bios, ramcfg + 0x01) & 0x02)
+		data |= 0x00000100;
+	ram_mask(fuc, 0x10f610, 0x70000000, data);
+
+	mask = 0x33f00000;
+	data = 0x00000000;
+	if (!(nv_ro08(bios, ramcfg + 0x01) & 0x04))
+		data |= 0x20200000;
+	if (!(nv_ro08(bios, ramcfg + 0x07) & 0x80))
+		data |= 0x12800000;
+	/*XXX: see note above about there probably being some condition
+	 *     for the 10f824 stuff that uses ramcfg 3...
+	 */
+	if ( (nv_ro08(bios, ramcfg + 0x03) & 0xf0)) {
+		if (nv_ro08(bios, rammap + 0x08) & 0x0c) {
+			if (!(nv_ro08(bios, ramcfg + 0x07) & 0x80))
+				mask |= 0x00000020;
+			else
+				data |= 0x00000020;
+			mask |= 0x00000004;
+		}
+	} else {
+		mask |= 0x40000020;
+		data |= 0x00000004;
+	}
+
+	ram_mask(fuc, 0x10f808, mask, data);
+
+	data = nv_ro08(bios, ramcfg + 0x03) & 0x0f;
+	ram_wr32(fuc, 0x10f870, 0x11111111 * data);
+
+	data = nv_ro08(bios, ramcfg + 0x02) & 0x03;
+	if (nv_ro08(bios, ramcfg + 0x01) & 0x10)
+		data |= 0x00000004;
+	if ((nv_rd32(bios, 0x100770) & 0x00000004) != (data & 0x00000004)) {
+		ram_wr32(fuc, 0x10f750, 0x04000009);
+		ram_wr32(fuc, 0x100710, 0x00000000);
+		ram_wait(fuc, 0x100710, 0x80000000, 0x80000000, 200000);
+	}
+	ram_mask(fuc, 0x100770, 0x00000007, data);
+
+	data = (nv_ro08(bios, timing + 0x30) & 0x07) << 8;
+	if (nv_ro08(bios, ramcfg + 0x01) & 0x01)
+		data |= 0x80000000;
+	ram_mask(fuc, 0x100778, 0x00000700, data);
+
+	data = nv_ro16(bios, timing + 0x2c);
+	ram_mask(fuc, 0x10f250, 0x000003f0, (data & 0x003f) <<  4);
+	ram_mask(fuc, 0x10f24c, 0x7f000000, (data & 0x1fc0) << 18);
+
+	data = nv_ro08(bios, timing + 0x30);
+	ram_mask(fuc, 0x10f224, 0x001f0000, (data & 0xf8) << 13);
+
+	data = nv_ro16(bios, timing + 0x31);
+	ram_mask(fuc, 0x10fec4, 0x041e0f07, (data & 0x0800) << 15 |
+					    (data & 0x0780) << 10 |
+					    (data & 0x0078) <<  5 |
+					    (data & 0x0007));
+	ram_mask(fuc, 0x10fec8, 0x00000027, (data & 0x8000) >> 10 |
+					    (data & 0x7000) >> 12);
+
+	ram_wr32(fuc, 0x10f090, 0x4000007e);
+	ram_nsec(fuc, 1000);
+	ram_wr32(fuc, 0x10f314, 0x00000001); /* PRECHARGE */
+	ram_wr32(fuc, 0x10f310, 0x00000001); /* REFRESH */
+	ram_nsec(fuc, 2000);
+	ram_wr32(fuc, 0x10f210, 0x80000000); /* REFRESH_AUTO = 1 */
+
+	if ((nv_ro08(bios, ramcfg + 0x08) & 0x10) && (ram->mode == 2) /*XXX*/) {
+		u32 temp = ram_mask(fuc, 0x10f294, 0xff000000, 0x24000000);
+		train(fuc, 0xa4010000); /*XXX*/
+		ram_nsec(fuc, 1000);
+		ram_wr32(fuc, 0x10f294, temp);
+	}
+
+	ram_mask(fuc, mr[3], 0xfff, ram->base.mr[3]);
+	ram_wr32(fuc, mr[0], ram->base.mr[0]);
+	ram_mask(fuc, mr[8], 0xfff, ram->base.mr[8]);
+	ram_nsec(fuc, 1000);
+	ram_mask(fuc, mr[1], 0xfff, ram->base.mr[1]);
+	ram_mask(fuc, mr[5], 0xfff, ram->base.mr[5]);
+	ram_mask(fuc, mr[6], 0xfff, ram->base.mr[6]);
+	ram_mask(fuc, mr[7], 0xfff, ram->base.mr[7]);
+
+	if (vc == 0 && ram_have(fuc, gpio2E)) {
+		u32 temp  = ram_mask(fuc, gpio2E, 0x3000, fuc->r_func2E[0]);
+		if (temp != ram_rd32(fuc, gpio2E)) {
+			ram_wr32(fuc, gpiotrig, 1);
+			ram_nsec(fuc, 20000);
+		}
+	}
+
+	ram_mask(fuc, 0x10f200, 0x80000000, 0x80000000);
+	ram_wr32(fuc, 0x10f318, 0x00000001); /* NOP? */
+	ram_mask(fuc, 0x10f200, 0x80000000, 0x00000000);
+	ram_nsec(fuc, 1000);
+
+	data  = ram_rd32(fuc, 0x10f978);
+	data &= ~0x00046144;
+	data |=  0x0000000b;
+	if (!(nv_ro08(bios, ramcfg + 0x07) & 0x08)) {
+		if (!(nv_ro08(bios, ramcfg + 0x07) & 0x04))
+			data |= 0x0000200c;
+		else
+			data |= 0x00000000;
+	} else {
+		data |= 0x00040044;
+	}
+	ram_wr32(fuc, 0x10f978, data);
+
+	if (ram->mode == 1) {
+		data = ram_rd32(fuc, 0x10f830) | 0x00000001;
+		ram_wr32(fuc, 0x10f830, data);
+	}
+
+	if (!(nv_ro08(bios, ramcfg + 0x07) & 0x08)) {
+		data = 0x88020000;
+		if ( (nv_ro08(bios, ramcfg + 0x07) & 0x04))
+			data |= 0x10000000;
+		if (!(nv_ro08(bios, rammap + 0x08) & 0x10))
+			data |= 0x00080000;
+	} else {
+		data = 0xa40e0000;
+	}
+	train(fuc, data);
+	ram_nsec(fuc, 1000);
+
+	if (ram->mode == 2) { /*XXX*/
+		ram_mask(fuc, 0x10f800, 0x00000004, 0x00000004);
+	}
+
+	/* MR5: (re)enable LP3 if necessary
+	 * XXX: need to find the switch, keeping off for now
+	 */
+	ram_mask(fuc, mr[5], 0x00000004, 0x00000000);
+
+	if (ram->mode != 2) {
+		ram_mask(fuc, 0x10f830, 0x01000000, 0x01000000);
+		ram_mask(fuc, 0x10f830, 0x01000000, 0x00000000);
+	}
+
+	if (nv_ro08(bios, ramcfg + 0x07) & 0x02) {
+		ram_mask(fuc, 0x10f910, 0x80020000, 0x01000000);
+		ram_mask(fuc, 0x10f914, 0x80020000, 0x01000000);
+	}
+
+	ram_wr32(fuc, 0x62c000, 0x0f0f0f00);
+
+	if (nv_ro08(bios, rammap + 0x08) & 0x01)
+		data = 0x00000800;
+	else
+		data = 0x00000000;
+	ram_mask(fuc, 0x10f200, 0x00000800, data);
+	return 0;
+}
+
+/*******************************************************************************
+ * DDR3
+ ******************************************************************************/
+
+static int
+nve0_ram_calc_sddr3(struct nouveau_fb *pfb, u32 freq)
+{
+	struct nouveau_bios *bios = nouveau_bios(pfb);
+	struct nve0_ram *ram = (void *)pfb->ram;
+	struct nve0_ramfuc *fuc = &ram->fuc;
+	const u32 rcoef = ((  ram->P1 << 16) | (ram->N1 << 8) | ram->M1);
+	const u32 runk0 = ram->fN1 << 16;
+	const u32 runk1 = ram->fN1;
+	const u32 rammap = ram->base.rammap.data;
+	const u32 ramcfg = ram->base.ramcfg.data;
+	const u32 timing = ram->base.timing.data;
+	int vc = !(nv_ro08(bios, ramcfg + 0x02) & 0x08);
+	int mv = 1; /*XXX*/
+	u32 mask, data;
+
+	ram_mask(fuc, 0x10f808, 0x40000000, 0x40000000);
+	ram_wr32(fuc, 0x62c000, 0x0f0f0000);
+
+	if (vc == 1 && ram_have(fuc, gpio2E)) {
+		u32 temp  = ram_mask(fuc, gpio2E, 0x3000, fuc->r_func2E[1]);
+		if (temp != ram_rd32(fuc, gpio2E)) {
+			ram_wr32(fuc, gpiotrig, 1);
+			ram_nsec(fuc, 20000);
+		}
+	}
+
+	ram_mask(fuc, 0x10f200, 0x00000800, 0x00000000);
+	if ((nv_ro08(bios, ramcfg + 0x03) & 0xf0))
+		ram_mask(fuc, 0x10f808, 0x04000000, 0x04000000);
+
+	ram_wr32(fuc, 0x10f314, 0x00000001); /* PRECHARGE */
+	ram_wr32(fuc, 0x10f210, 0x00000000); /* REFRESH_AUTO = 0 */
+	ram_wr32(fuc, 0x10f310, 0x00000001); /* REFRESH */
+	ram_mask(fuc, 0x10f200, 0x80000000, 0x80000000);
+	ram_wr32(fuc, 0x10f310, 0x00000001); /* REFRESH */
+	ram_mask(fuc, 0x10f200, 0x80000000, 0x00000000);
+	ram_nsec(fuc, 1000);
+
+	ram_wr32(fuc, 0x10f090, 0x00000060);
+	ram_wr32(fuc, 0x10f090, 0xc000007e);
+
+	/*XXX: there does appear to be some kind of condition here, simply
+	 *     modifying these bits in the vbios from the default pl0
+	 *     entries shows no change.  however, the data does appear to
+	 *     be correct and may be required for the transition back
+	 */
+	mask = 0x00010000;
+	data = 0x00010000;
+
+	if (1) {
+		mask |= 0x800807e0;
+		data |= 0x800807e0;
+		switch (nv_ro08(bios, ramcfg + 0x03) & 0xc0) {
+		case 0xc0: data &= ~0x00000040; break;
+		case 0x80: data &= ~0x00000100; break;
+		case 0x40: data &= ~0x80000000; break;
+		case 0x00: data &= ~0x00000400; break;
+		}
+
+		switch (nv_ro08(bios, ramcfg + 0x03) & 0x30) {
+		case 0x30: data &= ~0x00000020; break;
+		case 0x20: data &= ~0x00000080; break;
+		case 0x10: data &= ~0x00080000; break;
+		case 0x00: data &= ~0x00000200; break;
+		}
+	}
+
+	if (nv_ro08(bios, ramcfg + 0x02) & 0x80)
+		mask |= 0x03000000;
+	if (nv_ro08(bios, ramcfg + 0x02) & 0x40)
+		mask |= 0x00002000;
+	if (nv_ro08(bios, ramcfg + 0x07) & 0x10)
+		mask |= 0x00004000;
+	if (nv_ro08(bios, ramcfg + 0x07) & 0x08)
+		mask |= 0x00000003;
+	else
+		mask |= 0x14000000;
+	ram_mask(fuc, 0x10f824, mask, data);
+
+	ram_mask(fuc, 0x132040, 0x00010000, 0x00000000);
+
+	ram_mask(fuc, 0x1373f4, 0x00000000, 0x00010010);
+	data  = ram_rd32(fuc, 0x1373ec) & ~0x00030000;
+	data |= (nv_ro08(bios, ramcfg + 0x03) & 0x30) << 12;
+	ram_wr32(fuc, 0x1373ec, data);
+	ram_mask(fuc, 0x1373f4, 0x00000003, 0x00000000);
+	ram_mask(fuc, 0x1373f4, 0x00000010, 0x00000000);
+
+	/* (re)program refpll, if required */
+	if ((ram_rd32(fuc, 0x132024) & 0xffffffff) != rcoef ||
+	    (ram_rd32(fuc, 0x132034) & 0x0000ffff) != runk1) {
+		ram_mask(fuc, 0x132000, 0x00000001, 0x00000000);
+		ram_mask(fuc, 0x132020, 0x00000001, 0x00000000);
+		ram_wr32(fuc, 0x137320, 0x00000000);
+		ram_mask(fuc, 0x132030, 0xffff0000, runk0);
+		ram_mask(fuc, 0x132034, 0x0000ffff, runk1);
+		ram_wr32(fuc, 0x132024, rcoef);
+		ram_mask(fuc, 0x132028, 0x00080000, 0x00080000);
+		ram_mask(fuc, 0x132020, 0x00000001, 0x00000001);
+		ram_wait(fuc, 0x137390, 0x00020000, 0x00020000, 64000);
+		ram_mask(fuc, 0x132028, 0x00080000, 0x00000000);
+	}
+
+	ram_mask(fuc, 0x1373f4, 0x00000010, 0x00000010);
+	ram_mask(fuc, 0x1373f4, 0x00000003, 0x00000001);
+	ram_mask(fuc, 0x1373f4, 0x00010000, 0x00000000);
+
+	if (ram_have(fuc, gpioMV)) {
+		u32 temp  = ram_mask(fuc, gpioMV, 0x3000, fuc->r_funcMV[mv]);
+		if (temp != ram_rd32(fuc, gpioMV)) {
+			ram_wr32(fuc, gpiotrig, 1);
+			ram_nsec(fuc, 64000);
+		}
+	}
+
+	if ( (nv_ro08(bios, ramcfg + 0x02) & 0x40) ||
+	     (nv_ro08(bios, ramcfg + 0x07) & 0x10)) {
+		ram_mask(fuc, 0x132040, 0x00010000, 0x00010000);
+		ram_nsec(fuc, 20000);
+	}
+
+	if (ram->mode != 2) /*XXX*/ {
+		if (nv_ro08(bios, ramcfg + 0x07) & 0x40)
+			ram_mask(fuc, 0x10f670, 0x80000000, 0x80000000);
+	}
+
+	data = (nv_ro08(bios, rammap + 0x11) & 0x0c) >> 2;
+	ram_wr32(fuc, 0x10f65c, 0x00000011 * data);
+	ram_wr32(fuc, 0x10f6b8, 0x01010101 * nv_ro08(bios, ramcfg + 0x09));
+	ram_wr32(fuc, 0x10f6bc, 0x01010101 * nv_ro08(bios, ramcfg + 0x09));
+
+	mask = 0x00010000;
+	data = 0x00000000;
+	if (!(nv_ro08(bios, ramcfg + 0x02) & 0x80))
+		data |= 0x03000000;
+	if (!(nv_ro08(bios, ramcfg + 0x02) & 0x40))
+		data |= 0x00002000;
+	if (!(nv_ro08(bios, ramcfg + 0x07) & 0x10))
+		data |= 0x00004000;
+	if (!(nv_ro08(bios, ramcfg + 0x07) & 0x08))
+		data |= 0x00000003;
+	else
+		data |= 0x14000000;
+	ram_mask(fuc, 0x10f824, mask, data);
+	ram_nsec(fuc, 1000);
+
+	if (nv_ro08(bios, ramcfg + 0x08) & 0x01)
+		data = 0x00100000;
+	else
+		data = 0x00000000;
+	ram_mask(fuc, 0x10f82c, 0x00100000, data);
+
+	/* PFB timing */
+	ram_mask(fuc, 0x10f248, 0xffffffff, nv_ro32(bios, timing + 0x28));
+	ram_mask(fuc, 0x10f290, 0xffffffff, nv_ro32(bios, timing + 0x00));
+	ram_mask(fuc, 0x10f294, 0xffffffff, nv_ro32(bios, timing + 0x04));
+	ram_mask(fuc, 0x10f298, 0xffffffff, nv_ro32(bios, timing + 0x08));
+	ram_mask(fuc, 0x10f29c, 0xffffffff, nv_ro32(bios, timing + 0x0c));
+	ram_mask(fuc, 0x10f2a0, 0xffffffff, nv_ro32(bios, timing + 0x10));
+	ram_mask(fuc, 0x10f2a4, 0xffffffff, nv_ro32(bios, timing + 0x14));
+	ram_mask(fuc, 0x10f2a8, 0xffffffff, nv_ro32(bios, timing + 0x18));
+	ram_mask(fuc, 0x10f2ac, 0xffffffff, nv_ro32(bios, timing + 0x1c));
+	ram_mask(fuc, 0x10f2cc, 0xffffffff, nv_ro32(bios, timing + 0x20));
+	ram_mask(fuc, 0x10f2e8, 0xffffffff, nv_ro32(bios, timing + 0x24));
+
+	mask = 0x33f00000;
+	data = 0x00000000;
+	if (!(nv_ro08(bios, ramcfg + 0x01) & 0x04))
+		data |= 0x20200000;
+	if (!(nv_ro08(bios, ramcfg + 0x07) & 0x80))
+		data |= 0x12800000;
+	/*XXX: see note above about there probably being some condition
+	 *     for the 10f824 stuff that uses ramcfg 3...
+	 */
+	if ( (nv_ro08(bios, ramcfg + 0x03) & 0xf0)) {
+		if (nv_ro08(bios, rammap + 0x08) & 0x0c) {
+			if (!(nv_ro08(bios, ramcfg + 0x07) & 0x80))
+				mask |= 0x00000020;
+			else
+				data |= 0x00000020;
+			mask |= 0x08000004;
+		}
+		data |= 0x04000000;
+	} else {
+		mask |= 0x44000020;
+		data |= 0x08000004;
+	}
+
+	ram_mask(fuc, 0x10f808, mask, data);
+
+	data = nv_ro08(bios, ramcfg + 0x03) & 0x0f;
+	ram_wr32(fuc, 0x10f870, 0x11111111 * data);
+
+	data = nv_ro16(bios, timing + 0x2c);
+	ram_mask(fuc, 0x10f250, 0x000003f0, (data & 0x003f) <<  4);
+
+	if (((nv_ro32(bios, timing + 0x2c) & 0x00001fc0) >>  6) >
+	    ((nv_ro32(bios, timing + 0x28) & 0x7f000000) >> 24))
+		data = (nv_ro32(bios, timing + 0x2c) & 0x00001fc0) >>  6;
+	else
+		data = (nv_ro32(bios, timing + 0x28) & 0x1f000000) >> 24;
+	ram_mask(fuc, 0x10f24c, 0x7f000000, data << 24);
+
+	data = nv_ro08(bios, timing + 0x30);
+	ram_mask(fuc, 0x10f224, 0x001f0000, (data & 0xf8) << 13);
+
+	ram_wr32(fuc, 0x10f090, 0x4000007f);
+	ram_nsec(fuc, 1000);
+
+	ram_wr32(fuc, 0x10f314, 0x00000001); /* PRECHARGE */
+	ram_wr32(fuc, 0x10f310, 0x00000001); /* REFRESH */
+	ram_wr32(fuc, 0x10f210, 0x80000000); /* REFRESH_AUTO = 1 */
+	ram_nsec(fuc, 1000);
+
+	ram_nuke(fuc, mr[0]);
+	ram_mask(fuc, mr[0], 0x100, 0x100);
+	ram_mask(fuc, mr[0], 0x100, 0x000);
+
+	ram_mask(fuc, mr[2], 0xfff, ram->base.mr[2]);
+	ram_wr32(fuc, mr[0], ram->base.mr[0]);
+	ram_nsec(fuc, 1000);
+
+	ram_nuke(fuc, mr[0]);
+	ram_mask(fuc, mr[0], 0x100, 0x100);
+	ram_mask(fuc, mr[0], 0x100, 0x000);
+
+	if (vc == 0 && ram_have(fuc, gpio2E)) {
+		u32 temp  = ram_mask(fuc, gpio2E, 0x3000, fuc->r_func2E[0]);
+		if (temp != ram_rd32(fuc, gpio2E)) {
+			ram_wr32(fuc, gpiotrig, 1);
+			ram_nsec(fuc, 20000);
+		}
+	}
+
+	if (ram->mode != 2) {
+		ram_mask(fuc, 0x10f830, 0x01000000, 0x01000000);
+		ram_mask(fuc, 0x10f830, 0x01000000, 0x00000000);
+	}
+
+	ram_mask(fuc, 0x10f200, 0x80000000, 0x80000000);
+	ram_wr32(fuc, 0x10f318, 0x00000001); /* NOP? */
+	ram_mask(fuc, 0x10f200, 0x80000000, 0x00000000);
+	ram_nsec(fuc, 1000);
+
+	ram_wr32(fuc, 0x62c000, 0x0f0f0f00);
+
+	if (nv_ro08(bios, rammap + 0x08) & 0x01)
+		data = 0x00000800;
+	else
+		data = 0x00000000;
+	ram_mask(fuc, 0x10f200, 0x00000800, data);
+	return 0;
+}
+
+/*******************************************************************************
+ * main hooks
+ ******************************************************************************/
+
+static int
+nve0_ram_calc(struct nouveau_fb *pfb, u32 freq)
+{
+	struct nouveau_bios *bios = nouveau_bios(pfb);
+	struct nve0_ram *ram = (void *)pfb->ram;
+	struct nve0_ramfuc *fuc = &ram->fuc;
+	struct bit_entry M;
+	int ret, refclk, strap, i;
+	u32 data;
+	u8  cnt;
+
+	/* lookup memory config data relevant to the target frequency */
+	ram->base.rammap.data = nvbios_rammap_match(bios, freq / 1000,
+						   &ram->base.rammap.version,
+						   &ram->base.rammap.size, &cnt,
+						   &ram->base.ramcfg.size);
+	if (!ram->base.rammap.data || ram->base.rammap.version != 0x11 ||
+	     ram->base.rammap.size < 0x09) {
+		nv_error(pfb, "invalid/missing rammap entry\n");
+		return -EINVAL;
+	}
+
+	/* locate specific data set for the attached memory */
+	if (bit_entry(bios, 'M', &M) || M.version != 2 || M.length < 3) {
+		nv_error(pfb, "invalid/missing memory table\n");
+		return -EINVAL;
+	}
+
+	strap = (nv_rd32(pfb, 0x101000) & 0x0000003c) >> 2;
+	data = nv_ro16(bios, M.offset + 1);
+	if (data)
+		strap = nv_ro08(bios, data + strap);
+
+	if (strap >= cnt) {
+		nv_error(pfb, "invalid ramcfg strap\n");
+		return -EINVAL;
+	}
+
+	ram->base.ramcfg.version = ram->base.rammap.version;
+	ram->base.ramcfg.data = ram->base.rammap.data + ram->base.rammap.size +
+			       (ram->base.ramcfg.size * strap);
+	if (!ram->base.ramcfg.data || ram->base.ramcfg.version != 0x11 ||
+	     ram->base.ramcfg.size < 0x08) {
+		nv_error(pfb, "invalid/missing ramcfg entry\n");
+		return -EINVAL;
+	}
+
+	/* lookup memory timings, if bios says they're present */
+	strap = nv_ro08(bios, ram->base.ramcfg.data + 0x00);
+	if (strap != 0xff) {
+		ram->base.timing.data =
+			nvbios_timing_entry(bios, strap,
+					   &ram->base.timing.version,
+					   &ram->base.timing.size);
+		if (!ram->base.timing.data ||
+		     ram->base.timing.version != 0x20 ||
+		     ram->base.timing.size < 0x33) {
+			nv_error(pfb, "invalid/missing timing entry\n");
+			return -EINVAL;
+		}
+	} else {
+		ram->base.timing.data = 0;
+	}
+
+	ret = ram_init(fuc, pfb);
+	if (ret)
+		return ret;
+
+	ram->mode = (freq > fuc->refpll.vco1.max_freq) ? 2 : 1;
+	ram->from = ram_rd32(fuc, 0x1373f4) & 0x0000000f;
+
+	/* XXX: this is *not* what nvidia do.  on fermi nvidia generally
+	 * select, based on some unknown condition, one of the two possible
+	 * reference frequencies listed in the vbios table for mempll and
+	 * program refpll to that frequency.
+	 *
+	 * so far, i've seen very weird values being chosen by nvidia on
+	 * kepler boards, no idea how/why they're chosen.
+	 */
+	refclk = freq;
+	if (ram->mode == 2)
+		refclk = fuc->mempll.refclk;
+
+	/* calculate refpll coefficients */
+	ret = nva3_pll_calc(nv_subdev(pfb), &fuc->refpll, refclk, &ram->N1,
+			   &ram->fN1, &ram->M1, &ram->P1);
+	fuc->mempll.refclk = ret;
+	if (ret <= 0) {
+		nv_error(pfb, "unable to calc refpll\n");
+		return -EINVAL;
+	}
+
+	/* calculate mempll coefficients, if we're using it */
+	if (ram->mode == 2) {
+		/* post-divider doesn't work... the reg takes the values but
+		 * appears to completely ignore it.  there *is* a bit at
+		 * bit 28 that appears to divide the clock by 2 if set.
+		 */
+		fuc->mempll.min_p = 1;
+		fuc->mempll.max_p = 2;
+
+		ret = nva3_pll_calc(nv_subdev(pfb), &fuc->mempll, freq,
+				   &ram->N2, NULL, &ram->M2, &ram->P2);
+		if (ret <= 0) {
+			nv_error(pfb, "unable to calc mempll\n");
+			return -EINVAL;
+		}
+	}
+
+	for (i = 0; i < ARRAY_SIZE(fuc->r_mr); i++) {
+		if (ram_have(fuc, mr[i]))
+			ram->base.mr[i] = ram_rd32(fuc, mr[i]);
+	}
+
+	switch (ram->base.type) {
+	case NV_MEM_TYPE_DDR3:
+		ret = nouveau_sddr3_calc(&ram->base);
+		if (ret == 0)
+			ret = nve0_ram_calc_sddr3(pfb, freq);
+		break;
+	case NV_MEM_TYPE_GDDR5:
+		ret = nouveau_gddr5_calc(&ram->base);
+		if (ret == 0)
+			ret = nve0_ram_calc_gddr5(pfb, freq);
+		break;
+	default:
+		ret = -ENOSYS;
+		break;
+	}
+
+	return ret;
+}
+
+static int
+nve0_ram_prog(struct nouveau_fb *pfb)
+{
+	struct nouveau_device *device = nv_device(pfb);
+	struct nve0_ram *ram = (void *)pfb->ram;
+	struct nve0_ramfuc *fuc = &ram->fuc;
+	ram_exec(fuc, nouveau_boolopt(device->cfgopt, "NvMemExec", false));
+	return 0;
+}
+
+static void
+nve0_ram_tidy(struct nouveau_fb *pfb)
+{
+	struct nve0_ram *ram = (void *)pfb->ram;
+	struct nve0_ramfuc *fuc = &ram->fuc;
+	ram_exec(fuc, false);
+}
+
+static int
+nve0_ram_init(struct nouveau_object *object)
+{
+	struct nouveau_fb *pfb = (void *)object->parent;
+	struct nve0_ram *ram   = (void *)object;
+	struct nouveau_bios *bios = nouveau_bios(pfb);
+	static const u8  train0[] = {
+		0x00, 0xff, 0xff, 0x00, 0xff, 0x00,
+		0x00, 0xff, 0xff, 0x00, 0xff, 0x00,
+	};
+	static const u32 train1[] = {
+		0x00000000, 0xffffffff,
+		0x55555555, 0xaaaaaaaa,
+		0x33333333, 0xcccccccc,
+		0xf0f0f0f0, 0x0f0f0f0f,
+		0x00ff00ff, 0xff00ff00,
+		0x0000ffff, 0xffff0000,
+	};
+	u8  ver, hdr, cnt, len, snr, ssz;
+	u32 data, save;
+	int ret, i;
+
+	ret = nouveau_ram_init(&ram->base);
+	if (ret)
+		return ret;
+
+	/* run a bunch of tables from rammap table.  there's actually
+	 * individual pointers for each rammap entry too, but, nvidia
+	 * seem to just run the last two entries' scripts early on in
+	 * their init, and never again.. we'll just run 'em all once
+	 * for now.
+	 *
+	 * i strongly suspect that each script is for a separate mode
+	 * (likely selected by 0x10f65c's lower bits?), and the
+	 * binary driver skips the one that's already been setup by
+	 * the init tables.
+	 */
+	data = nvbios_rammap_table(bios, &ver, &hdr, &cnt, &len, &snr, &ssz);
+	if (!data || hdr < 0x15)
+		return -EINVAL;
+
+	cnt  = nv_ro08(bios, data + 0x14); /* guess at count */
+	data = nv_ro32(bios, data + 0x10); /* guess u32... */
+	save = nv_rd32(pfb, 0x10f65c);
+	for (i = 0; i < cnt; i++) {
+		nv_mask(pfb, 0x10f65c, 0x000000f0, i << 4);
+		nvbios_exec(&(struct nvbios_init) {
+				.subdev = nv_subdev(pfb),
+				.bios = bios,
+				.offset = nv_ro32(bios, data), /* guess u32 */
+				.execute = 1,
+			    });
+		data += 4;
+	}
+	nv_wr32(pfb, 0x10f65c, save);
+
+	switch (ram->base.type) {
+	case NV_MEM_TYPE_GDDR5:
+		for (i = 0; i < 0x30; i++) {
+			nv_wr32(pfb, 0x10f968, 0x00000000 | (i << 8));
+			nv_wr32(pfb, 0x10f920, 0x00000000 | train0[i % 12]);
+			nv_wr32(pfb, 0x10f918,              train1[i % 12]);
+			nv_wr32(pfb, 0x10f920, 0x00000100 | train0[i % 12]);
+			nv_wr32(pfb, 0x10f918,              train1[i % 12]);
+
+			nv_wr32(pfb, 0x10f96c, 0x00000000 | (i << 8));
+			nv_wr32(pfb, 0x10f924, 0x00000000 | train0[i % 12]);
+			nv_wr32(pfb, 0x10f91c,              train1[i % 12]);
+			nv_wr32(pfb, 0x10f924, 0x00000100 | train0[i % 12]);
+			nv_wr32(pfb, 0x10f91c,              train1[i % 12]);
+		}
+
+		for (i = 0; i < 0x100; i++) {
+			nv_wr32(pfb, 0x10f968, i);
+			nv_wr32(pfb, 0x10f900, train1[2 + (i & 1)]);
+		}
+
+		for (i = 0; i < 0x100; i++) {
+			nv_wr32(pfb, 0x10f96c, i);
+			nv_wr32(pfb, 0x10f900, train1[2 + (i & 1)]);
+		}
+		break;
+	default:
+		break;
+	}
+
+	return 0;
+}
+
+static int
+nve0_ram_ctor(struct nouveau_object *parent, struct nouveau_object *engine,
+	      struct nouveau_oclass *oclass, void *data, u32 size,
+	      struct nouveau_object **pobject)
+{
+	struct nouveau_fb *pfb = nouveau_fb(parent);
+	struct nouveau_bios *bios = nouveau_bios(pfb);
+	struct nouveau_gpio *gpio = nouveau_gpio(pfb);
+	struct dcb_gpio_func func;
+	struct nve0_ram *ram;
+	int ret;
+
+	ret = nvc0_ram_create(parent, engine, oclass, &ram);
+	*pobject = nv_object(ram);
+	if (ret)
+		return ret;
+
+	switch (ram->base.type) {
+	case NV_MEM_TYPE_DDR3:
+	case NV_MEM_TYPE_GDDR5:
+		ram->base.calc = nve0_ram_calc;
+		ram->base.prog = nve0_ram_prog;
+		ram->base.tidy = nve0_ram_tidy;
+		break;
+	default:
+		nv_warn(pfb, "reclocking of this RAM type is unsupported\n");
+		break;
+	}
+
+	// parse bios data for both pll's
+	ret = nvbios_pll_parse(bios, 0x0c, &ram->fuc.refpll);
+	if (ret) {
+		nv_error(pfb, "mclk refpll data not found\n");
+		return ret;
+	}
+
+	ret = nvbios_pll_parse(bios, 0x04, &ram->fuc.mempll);
+	if (ret) {
+		nv_error(pfb, "mclk pll data not found\n");
+		return ret;
+	}
+
+	ret = gpio->find(gpio, 0, 0x18, DCB_GPIO_UNUSED, &func);
+	if (ret == 0) {
+		ram->fuc.r_gpioMV = ramfuc_reg(0x00d610 + (func.line * 0x04));
+		ram->fuc.r_funcMV[0] = (func.log[0] ^ 2) << 12;
+		ram->fuc.r_funcMV[1] = (func.log[1] ^ 2) << 12;
+	}
+
+	ret = gpio->find(gpio, 0, 0x2e, DCB_GPIO_UNUSED, &func);
+	if (ret == 0) {
+		ram->fuc.r_gpio2E = ramfuc_reg(0x00d610 + (func.line * 0x04));
+		ram->fuc.r_func2E[0] = (func.log[0] ^ 2) << 12;
+		ram->fuc.r_func2E[1] = (func.log[1] ^ 2) << 12;
+	}
+
+	ram->fuc.r_gpiotrig = ramfuc_reg(0x00d604);
+
+	ram->fuc.r_0x132020 = ramfuc_reg(0x132020);
+	ram->fuc.r_0x132028 = ramfuc_reg(0x132028);
+	ram->fuc.r_0x132024 = ramfuc_reg(0x132024);
+	ram->fuc.r_0x132030 = ramfuc_reg(0x132030);
+	ram->fuc.r_0x132034 = ramfuc_reg(0x132034);
+	ram->fuc.r_0x132000 = ramfuc_reg(0x132000);
+	ram->fuc.r_0x132004 = ramfuc_reg(0x132004);
+	ram->fuc.r_0x132040 = ramfuc_reg(0x132040);
+
+	ram->fuc.r_0x10f248 = ramfuc_reg(0x10f248);
+	ram->fuc.r_0x10f290 = ramfuc_reg(0x10f290);
+	ram->fuc.r_0x10f294 = ramfuc_reg(0x10f294);
+	ram->fuc.r_0x10f298 = ramfuc_reg(0x10f298);
+	ram->fuc.r_0x10f29c = ramfuc_reg(0x10f29c);
+	ram->fuc.r_0x10f2a0 = ramfuc_reg(0x10f2a0);
+	ram->fuc.r_0x10f2a4 = ramfuc_reg(0x10f2a4);
+	ram->fuc.r_0x10f2a8 = ramfuc_reg(0x10f2a8);
+	ram->fuc.r_0x10f2ac = ramfuc_reg(0x10f2ac);
+	ram->fuc.r_0x10f2cc = ramfuc_reg(0x10f2cc);
+	ram->fuc.r_0x10f2e8 = ramfuc_reg(0x10f2e8);
+	ram->fuc.r_0x10f250 = ramfuc_reg(0x10f250);
+	ram->fuc.r_0x10f24c = ramfuc_reg(0x10f24c);
+	ram->fuc.r_0x10fec4 = ramfuc_reg(0x10fec4);
+	ram->fuc.r_0x10fec8 = ramfuc_reg(0x10fec8);
+	ram->fuc.r_0x10f604 = ramfuc_reg(0x10f604);
+	ram->fuc.r_0x10f614 = ramfuc_reg(0x10f614);
+	ram->fuc.r_0x10f610 = ramfuc_reg(0x10f610);
+	ram->fuc.r_0x100770 = ramfuc_reg(0x100770);
+	ram->fuc.r_0x100778 = ramfuc_reg(0x100778);
+	ram->fuc.r_0x10f224 = ramfuc_reg(0x10f224);
+
+	ram->fuc.r_0x10f870 = ramfuc_reg(0x10f870);
+	ram->fuc.r_0x10f698 = ramfuc_reg(0x10f698);
+	ram->fuc.r_0x10f694 = ramfuc_reg(0x10f694);
+	ram->fuc.r_0x10f6b8 = ramfuc_reg(0x10f6b8);
+	ram->fuc.r_0x10f808 = ramfuc_reg(0x10f808);
+	ram->fuc.r_0x10f670 = ramfuc_reg(0x10f670);
+	ram->fuc.r_0x10f60c = ramfuc_reg(0x10f60c);
+	ram->fuc.r_0x10f830 = ramfuc_reg(0x10f830);
+	ram->fuc.r_0x1373ec = ramfuc_reg(0x1373ec);
+	ram->fuc.r_0x10f800 = ramfuc_reg(0x10f800);
+	ram->fuc.r_0x10f82c = ramfuc_reg(0x10f82c);
+
+	ram->fuc.r_0x10f978 = ramfuc_reg(0x10f978);
+	ram->fuc.r_0x10f910 = ramfuc_reg(0x10f910);
+	ram->fuc.r_0x10f914 = ramfuc_reg(0x10f914);
+
+	switch (ram->base.type) {
+	case NV_MEM_TYPE_GDDR5:
+		ram->fuc.r_mr[0] = ramfuc_reg(0x10f300);
+		ram->fuc.r_mr[1] = ramfuc_reg(0x10f330);
+		ram->fuc.r_mr[2] = ramfuc_reg(0x10f334);
+		ram->fuc.r_mr[3] = ramfuc_reg(0x10f338);
+		ram->fuc.r_mr[4] = ramfuc_reg(0x10f33c);
+		ram->fuc.r_mr[5] = ramfuc_reg(0x10f340);
+		ram->fuc.r_mr[6] = ramfuc_reg(0x10f344);
+		ram->fuc.r_mr[7] = ramfuc_reg(0x10f348);
+		ram->fuc.r_mr[8] = ramfuc_reg(0x10f354);
+		ram->fuc.r_mr[15] = ramfuc_reg(0x10f34c);
+		break;
+	case NV_MEM_TYPE_DDR3:
+		ram->fuc.r_mr[0] = ramfuc_reg(0x10f300);
+		ram->fuc.r_mr[2] = ramfuc_reg(0x10f320);
+		break;
+	default:
+		break;
+	}
+
+	ram->fuc.r_0x62c000 = ramfuc_reg(0x62c000);
+	ram->fuc.r_0x10f200 = ramfuc_reg(0x10f200);
+	ram->fuc.r_0x10f210 = ramfuc_reg(0x10f210);
+	ram->fuc.r_0x10f310 = ramfuc_reg(0x10f310);
+	ram->fuc.r_0x10f314 = ramfuc_reg(0x10f314);
+	ram->fuc.r_0x10f318 = ramfuc_reg(0x10f318);
+	ram->fuc.r_0x10f090 = ramfuc_reg(0x10f090);
+	ram->fuc.r_0x10f69c = ramfuc_reg(0x10f69c);
+	ram->fuc.r_0x10f824 = ramfuc_reg(0x10f824);
+	ram->fuc.r_0x1373f0 = ramfuc_reg(0x1373f0);
+	ram->fuc.r_0x1373f4 = ramfuc_reg(0x1373f4);
+	ram->fuc.r_0x137320 = ramfuc_reg(0x137320);
+	ram->fuc.r_0x10f65c = ramfuc_reg(0x10f65c);
+	ram->fuc.r_0x10f6bc = ramfuc_reg(0x10f6bc);
+	ram->fuc.r_0x100710 = ramfuc_reg(0x100710);
+	ram->fuc.r_0x10f750 = ramfuc_reg(0x10f750);
+	return 0;
+}
+
+struct nouveau_oclass
+nve0_ram_oclass = {
+	.handle = 0,
+	.ofuncs = &(struct nouveau_ofuncs) {
+		.ctor = nve0_ram_ctor,
+		.dtor = _nouveau_ram_dtor,
+		.init = nve0_ram_init,
+		.fini = _nouveau_ram_fini,
+	}
+};
diff --git a/drivers/gpu/drm/nouveau/core/subdev/fb/ramseq.h b/drivers/gpu/drm/nouveau/core/subdev/fb/ramseq.h
new file mode 100644
index 000000000000..571077e39071
--- /dev/null
+++ b/drivers/gpu/drm/nouveau/core/subdev/fb/ramseq.h
@@ -0,0 +1,18 @@
+#ifndef __NVKM_FBRAM_SEQ_H__
+#define __NVKM_FBRAM_SEQ_H__
+
+#include <subdev/bus.h>
+#include <subdev/bus/hwsq.h>
+
+#define ram_init(s,p)       hwsq_init(&(s)->base, (p))
+#define ram_exec(s,e)       hwsq_exec(&(s)->base, (e))
+#define ram_have(s,r)       ((s)->r_##r.addr != 0x000000)
+#define ram_rd32(s,r)       hwsq_rd32(&(s)->base, &(s)->r_##r)
+#define ram_wr32(s,r,d)     hwsq_wr32(&(s)->base, &(s)->r_##r, (d))
+#define ram_nuke(s,r)       hwsq_nuke(&(s)->base, &(s)->r_##r)
+#define ram_mask(s,r,m,d)   hwsq_mask(&(s)->base, &(s)->r_##r, (m), (d))
+#define ram_setf(s,f,d)     hwsq_setf(&(s)->base, (f), (d))
+#define ram_wait(s,f,d)     hwsq_wait(&(s)->base, (f), (d))
+#define ram_nsec(s,n)       hwsq_nsec(&(s)->base, (n))
+
+#endif
diff --git a/drivers/gpu/drm/nouveau/core/subdev/fb/sddr3.c b/drivers/gpu/drm/nouveau/core/subdev/fb/sddr3.c
new file mode 100644
index 000000000000..ebd4cd9c35d9
--- /dev/null
+++ b/drivers/gpu/drm/nouveau/core/subdev/fb/sddr3.c
@@ -0,0 +1,99 @@
+/*
+ * Copyright 2013 Red Hat Inc.
+ *
+ * Permission is hereby granted, free of charge, to any person obtaining a
+ * copy of this software and associated documentation files (the "Software"),
+ * to deal in the Software without restriction, including without limitation
+ * the rights to use, copy, modify, merge, publish, distribute, sublicense,
+ * and/or sell copies of the Software, and to permit persons to whom the
+ * Software is furnished to do so, subject to the following conditions:
+ *
+ * The above copyright notice and this permission notice shall be included in
+ * all copies or substantial portions of the Software.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+ * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+ * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT.  IN NO EVENT SHALL
+ * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR
+ * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE,
+ * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR
+ * OTHER DEALINGS IN THE SOFTWARE.
+ *
+ * Authors: Ben Skeggs <bskeggs@redhat.com>
+ */
+
+#include <subdev/bios.h>
+#include "priv.h"
+
+struct ramxlat {
+	int id;
+	u8 enc;
+};
+
+static inline int
+ramxlat(const struct ramxlat *xlat, int id)
+{
+	while (xlat->id >= 0) {
+		if (xlat->id == id)
+			return xlat->enc;
+		xlat++;
+	}
+	return -EINVAL;
+}
+
+static const struct ramxlat
+ramddr3_cl[] = {
+	{ 5, 2 }, { 6, 4 }, { 7, 6 }, { 8, 8 }, { 9, 10 }, { 10, 12 },
+	{ 11, 14 },
+	/* the below are mentioned in some, but not all, ddr3 docs */
+	{ 12, 1 }, { 13, 3 }, { 14, 5 },
+	{ -1 }
+};
+
+static const struct ramxlat
+ramddr3_wr[] = {
+	{ 5, 1 }, { 6, 2 }, { 7, 3 }, { 8, 4 }, { 10, 5 }, { 12, 6 },
+	/* the below are mentioned in some, but not all, ddr3 docs */
+	{ 14, 7 }, { 16, 0 },
+	{ -1 }
+};
+
+static const struct ramxlat
+ramddr3_cwl[] = {
+	{ 5, 0 }, { 6, 1 }, { 7, 2 }, { 8, 3 },
+	/* the below are mentioned in some, but not all, ddr3 docs */
+	{ 9, 4 },
+	{ -1 }
+};
+
+int
+nouveau_sddr3_calc(struct nouveau_ram *ram)
+{
+	struct nouveau_bios *bios = nouveau_bios(ram);
+	int WL, CL, WR;
+
+	switch (!!ram->timing.data * ram->timing.version) {
+	case 0x20:
+		WL = (nv_ro16(bios, ram->timing.data + 0x04) & 0x0f80) >> 7;
+		CL =  nv_ro08(bios, ram->timing.data + 0x04) & 0x1f;
+		WR =  nv_ro08(bios, ram->timing.data + 0x0a) & 0x7f;
+		break;
+	default:
+		return -ENOSYS;
+	}
+
+	WL = ramxlat(ramddr3_cwl, WL);
+	CL = ramxlat(ramddr3_cl, CL);
+	WR = ramxlat(ramddr3_wr, WR);
+	if (WL < 0 || CL < 0 || WR < 0)
+		return -EINVAL;
+
+	ram->mr[0] &= ~0xe74;
+	ram->mr[0] |= (WR & 0x07) << 9;
+	ram->mr[0] |= (CL & 0x0e) << 3;
+	ram->mr[0] |= (CL & 0x01) << 2;
+
+	ram->mr[2] &= ~0x038;
+	ram->mr[2] |= (WL & 0x07) << 3;
+	return 0;
+}
diff --git a/drivers/gpu/drm/nouveau/core/subdev/gpio/base.c b/drivers/gpu/drm/nouveau/core/subdev/gpio/base.c
index d422acc9af15..f572c2804c32 100644
--- a/drivers/gpu/drm/nouveau/core/subdev/gpio/base.c
+++ b/drivers/gpu/drm/nouveau/core/subdev/gpio/base.c
@@ -67,7 +67,7 @@ nouveau_gpio_find(struct nouveau_gpio *gpio, int idx, u8 tag, u8 line,
 		}
 	}
 
-	return -EINVAL;
+	return -ENOENT;
 }
 
 static int
diff --git a/drivers/gpu/drm/nouveau/core/subdev/i2c/base.c b/drivers/gpu/drm/nouveau/core/subdev/i2c/base.c
index 2895c19bb152..041fd5edaebf 100644
--- a/drivers/gpu/drm/nouveau/core/subdev/i2c/base.c
+++ b/drivers/gpu/drm/nouveau/core/subdev/i2c/base.c
@@ -195,7 +195,7 @@ nouveau_i2c_find_type(struct nouveau_i2c *i2c, u16 type)
 
 static int
 nouveau_i2c_identify(struct nouveau_i2c *i2c, int index, const char *what,
-		     struct i2c_board_info *info,
+		     struct nouveau_i2c_board_info *info,
 		     bool (*match)(struct nouveau_i2c_port *,
 				   struct i2c_board_info *))
 {
@@ -208,12 +208,29 @@ nouveau_i2c_identify(struct nouveau_i2c *i2c, int index, const char *what,
 	}
 
 	nv_debug(i2c, "probing %ss on bus: %d\n", what, port->index);
-	for (i = 0; info[i].addr; i++) {
-		if (nv_probe_i2c(port, info[i].addr) &&
-		    (!match || match(port, &info[i]))) {
-			nv_info(i2c, "detected %s: %s\n", what, info[i].type);
+	for (i = 0; info[i].dev.addr; i++) {
+		u8 orig_udelay = 0;
+
+		if ((port->adapter.algo == &i2c_bit_algo) &&
+		    (info[i].udelay != 0)) {
+			struct i2c_algo_bit_data *algo = port->adapter.algo_data;
+			nv_debug(i2c, "using custom udelay %d instead of %d\n",
+			         info[i].udelay, algo->udelay);
+			orig_udelay = algo->udelay;
+			algo->udelay = info[i].udelay;
+		}
+
+		if (nv_probe_i2c(port, info[i].dev.addr) &&
+		    (!match || match(port, &info[i].dev))) {
+			nv_info(i2c, "detected %s: %s\n", what,
+				info[i].dev.type);
 			return i;
 		}
+
+		if (orig_udelay) {
+			struct i2c_algo_bit_data *algo = port->adapter.algo_data;
+			algo->udelay = orig_udelay;
+		}
 	}
 
 	nv_debug(i2c, "no devices found.\n");
diff --git a/drivers/gpu/drm/nouveau/core/subdev/mc/base.c b/drivers/gpu/drm/nouveau/core/subdev/mc/base.c
index e290cfa4acee..b4b9943773bc 100644
--- a/drivers/gpu/drm/nouveau/core/subdev/mc/base.c
+++ b/drivers/gpu/drm/nouveau/core/subdev/mc/base.c
@@ -25,38 +25,48 @@
 #include <subdev/mc.h>
 #include <core/option.h>
 
+static inline u32
+nouveau_mc_intr_mask(struct nouveau_mc *pmc)
+{
+	u32 intr = nv_rd32(pmc, 0x000100);
+	if (intr == 0xffffffff) /* likely fallen off the bus */
+		intr = 0x00000000;
+	return intr;
+}
+
 static irqreturn_t
 nouveau_mc_intr(int irq, void *arg)
 {
 	struct nouveau_mc *pmc = arg;
-	const struct nouveau_mc_intr *map = pmc->intr_map;
-	struct nouveau_device *device = nv_device(pmc);
+	const struct nouveau_mc_oclass *oclass = (void *)nv_object(pmc)->oclass;
+	const struct nouveau_mc_intr *map = oclass->intr;
 	struct nouveau_subdev *unit;
-	u32 stat, intr;
-
-	intr = stat = nv_rd32(pmc, 0x000100);
-	if (intr == 0xffffffff)
-		return IRQ_NONE;
-	while (stat && map->stat) {
-		if (stat & map->stat) {
-			unit = nouveau_subdev(pmc, map->unit);
-			if (unit && unit->intr)
-				unit->intr(unit);
-			intr &= ~map->stat;
-		}
-		map++;
-	}
+	u32 intr;
 
+	nv_wr32(pmc, 0x000140, 0x00000000);
+	nv_rd32(pmc, 0x000140);
+	intr = nouveau_mc_intr_mask(pmc);
 	if (pmc->use_msi)
-		nv_wr08(pmc->base.base.parent, 0x00088068, 0xff);
+		oclass->msi_rearm(pmc);
 
 	if (intr) {
-		nv_error(pmc, "unknown intr 0x%08x\n", stat);
+		u32 stat = intr = nouveau_mc_intr_mask(pmc);
+		while (map->stat) {
+			if (intr & map->stat) {
+				unit = nouveau_subdev(pmc, map->unit);
+				if (unit && unit->intr)
+					unit->intr(unit);
+				stat &= ~map->stat;
+			}
+			map++;
+		}
+
+		if (stat)
+			nv_error(pmc, "unknown intr 0x%08x\n", stat);
 	}
 
-	if (stat == IRQ_HANDLED)
-		pm_runtime_mark_last_busy(&device->pdev->dev);
-	return stat ? IRQ_HANDLED : IRQ_NONE;
+	nv_wr32(pmc, 0x000140, 0x00000001);
+	return intr ? IRQ_HANDLED : IRQ_NONE;
 }
 
 int
@@ -91,37 +101,42 @@ _nouveau_mc_dtor(struct nouveau_object *object)
 
 int
 nouveau_mc_create_(struct nouveau_object *parent, struct nouveau_object *engine,
-		   struct nouveau_oclass *oclass,
-		   const struct nouveau_mc_intr *intr_map,
-		   int length, void **pobject)
+		   struct nouveau_oclass *bclass, int length, void **pobject)
 {
+	const struct nouveau_mc_oclass *oclass = (void *)bclass;
 	struct nouveau_device *device = nv_device(parent);
 	struct nouveau_mc *pmc;
 	int ret;
 
-	ret = nouveau_subdev_create_(parent, engine, oclass, 0, "PMC",
+	ret = nouveau_subdev_create_(parent, engine, bclass, 0, "PMC",
 				     "master", length, pobject);
 	pmc = *pobject;
 	if (ret)
 		return ret;
 
-	pmc->intr_map = intr_map;
-
 	switch (device->pdev->device & 0x0ff0) {
-	case 0x00f0: /* BR02? */
-	case 0x02e0: /* BR02? */
-		pmc->use_msi = false;
+	case 0x00f0:
+	case 0x02e0:
+		/* BR02? NFI how these would be handled yet exactly */
 		break;
 	default:
-		pmc->use_msi = nouveau_boolopt(device->cfgopt, "NvMSI", false);
+		switch (device->chipset) {
+		case 0xaa: break; /* reported broken, nv also disable it */
+		default:
+			pmc->use_msi = true;
+			break;
+		}
+	}
+
+	pmc->use_msi = nouveau_boolopt(device->cfgopt, "NvMSI", pmc->use_msi);
+	if (pmc->use_msi && oclass->msi_rearm) {
+		pmc->use_msi = pci_enable_msi(device->pdev) == 0;
 		if (pmc->use_msi) {
-			pmc->use_msi = pci_enable_msi(device->pdev) == 0;
-			if (pmc->use_msi) {
-				nv_info(pmc, "MSI interrupts enabled\n");
-				nv_wr08(device, 0x00088068, 0xff);
-			}
+			nv_info(pmc, "MSI interrupts enabled\n");
+			oclass->msi_rearm(pmc);
 		}
-		break;
+	} else {
+		pmc->use_msi = false;
 	}
 
 	ret = request_irq(device->pdev->irq, nouveau_mc_intr,
diff --git a/drivers/gpu/drm/nouveau/core/subdev/mc/nv04.c b/drivers/gpu/drm/nouveau/core/subdev/mc/nv04.c
index 64aa4edb0d9d..2d787e4dfefa 100644
--- a/drivers/gpu/drm/nouveau/core/subdev/mc/nv04.c
+++ b/drivers/gpu/drm/nouveau/core/subdev/mc/nv04.c
@@ -22,17 +22,14 @@
  * Authors: Ben Skeggs
  */
 
-#include <subdev/mc.h>
-
-struct nv04_mc_priv {
-	struct nouveau_mc base;
-};
+#include "nv04.h"
 
 const struct nouveau_mc_intr
 nv04_mc_intr[] = {
 	{ 0x00000001, NVDEV_ENGINE_MPEG },	/* NV17- MPEG/ME */
 	{ 0x00000100, NVDEV_ENGINE_FIFO },
 	{ 0x00001000, NVDEV_ENGINE_GR },
+	{ 0x00010000, NVDEV_ENGINE_DISP },
 	{ 0x00020000, NVDEV_ENGINE_VP },	/* NV40- */
 	{ 0x00100000, NVDEV_SUBDEV_TIMER },
 	{ 0x01000000, NVDEV_ENGINE_DISP },	/* NV04- PCRTC0 */
@@ -42,7 +39,18 @@ nv04_mc_intr[] = {
 	{}
 };
 
-static int
+int
+nv04_mc_init(struct nouveau_object *object)
+{
+	struct nv04_mc_priv *priv = (void *)object;
+
+	nv_wr32(priv, 0x000200, 0xffffffff); /* everything enabled */
+	nv_wr32(priv, 0x001850, 0x00000001); /* disable rom access */
+
+	return nouveau_mc_init(&priv->base);
+}
+
+int
 nv04_mc_ctor(struct nouveau_object *parent, struct nouveau_object *engine,
 	     struct nouveau_oclass *oclass, void *data, u32 size,
 	     struct nouveau_object **pobject)
@@ -50,7 +58,7 @@ nv04_mc_ctor(struct nouveau_object *parent, struct nouveau_object *engine,
 	struct nv04_mc_priv *priv;
 	int ret;
 
-	ret = nouveau_mc_create(parent, engine, oclass, nv04_mc_intr, &priv);
+	ret = nouveau_mc_create(parent, engine, oclass, &priv);
 	*pobject = nv_object(priv);
 	if (ret)
 		return ret;
@@ -58,24 +66,14 @@ nv04_mc_ctor(struct nouveau_object *parent, struct nouveau_object *engine,
 	return 0;
 }
 
-int
-nv04_mc_init(struct nouveau_object *object)
-{
-	struct nv04_mc_priv *priv = (void *)object;
-
-	nv_wr32(priv, 0x000200, 0xffffffff); /* everything enabled */
-	nv_wr32(priv, 0x001850, 0x00000001); /* disable rom access */
-
-	return nouveau_mc_init(&priv->base);
-}
-
-struct nouveau_oclass
-nv04_mc_oclass = {
-	.handle = NV_SUBDEV(MC, 0x04),
-	.ofuncs = &(struct nouveau_ofuncs) {
+struct nouveau_oclass *
+nv04_mc_oclass = &(struct nouveau_mc_oclass) {
+	.base.handle = NV_SUBDEV(MC, 0x04),
+	.base.ofuncs = &(struct nouveau_ofuncs) {
 		.ctor = nv04_mc_ctor,
 		.dtor = _nouveau_mc_dtor,
 		.init = nv04_mc_init,
 		.fini = _nouveau_mc_fini,
 	},
-};
+	.intr = nv04_mc_intr,
+}.base;
diff --git a/drivers/gpu/drm/nouveau/core/subdev/mc/nv04.h b/drivers/gpu/drm/nouveau/core/subdev/mc/nv04.h
new file mode 100644
index 000000000000..b0d5c31606c1
--- /dev/null
+++ b/drivers/gpu/drm/nouveau/core/subdev/mc/nv04.h
@@ -0,0 +1,21 @@
+#ifndef __NVKM_MC_NV04_H__
+#define __NVKM_MC_NV04_H__
+
+#include <subdev/mc.h>
+
+struct nv04_mc_priv {
+	struct nouveau_mc base;
+};
+
+int  nv04_mc_ctor(struct nouveau_object *, struct nouveau_object *,
+		  struct nouveau_oclass *, void *, u32,
+		  struct nouveau_object **);
+
+extern const struct nouveau_mc_intr nv04_mc_intr[];
+int  nv04_mc_init(struct nouveau_object *);
+void nv40_mc_msi_rearm(struct nouveau_mc *);
+int  nv50_mc_init(struct nouveau_object *);
+extern const struct nouveau_mc_intr nv50_mc_intr[];
+extern const struct nouveau_mc_intr nvc0_mc_intr[];
+
+#endif
diff --git a/drivers/gpu/drm/nouveau/core/subdev/mc/nv40.c b/drivers/gpu/drm/nouveau/core/subdev/mc/nv40.c
new file mode 100644
index 000000000000..5b1faecfed2d
--- /dev/null
+++ b/drivers/gpu/drm/nouveau/core/subdev/mc/nv40.c
@@ -0,0 +1,45 @@
+/*
+ * Copyright 2012 Red Hat Inc.
+ *
+ * Permission is hereby granted, free of charge, to any person obtaining a
+ * copy of this software and associated documentation files (the "Software"),
+ * to deal in the Software without restriction, including without limitation
+ * the rights to use, copy, modify, merge, publish, distribute, sublicense,
+ * and/or sell copies of the Software, and to permit persons to whom the
+ * Software is furnished to do so, subject to the following conditions:
+ *
+ * The above copyright notice and this permission notice shall be included in
+ * all copies or substantial portions of the Software.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+ * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+ * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT.  IN NO EVENT SHALL
+ * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR
+ * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE,
+ * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR
+ * OTHER DEALINGS IN THE SOFTWARE.
+ *
+ * Authors: Ben Skeggs
+ */
+
+#include "nv04.h"
+
+void
+nv40_mc_msi_rearm(struct nouveau_mc *pmc)
+{
+	struct nv04_mc_priv *priv = (void *)pmc;
+	nv_wr08(priv, 0x088068, 0xff);
+}
+
+struct nouveau_oclass *
+nv40_mc_oclass = &(struct nouveau_mc_oclass) {
+	.base.handle = NV_SUBDEV(MC, 0x40),
+	.base.ofuncs = &(struct nouveau_ofuncs) {
+		.ctor = nv04_mc_ctor,
+		.dtor = _nouveau_mc_dtor,
+		.init = nv04_mc_init,
+		.fini = _nouveau_mc_fini,
+	},
+	.intr = nv04_mc_intr,
+	.msi_rearm = nv40_mc_msi_rearm,
+}.base;
diff --git a/drivers/gpu/drm/nouveau/core/subdev/mc/nv44.c b/drivers/gpu/drm/nouveau/core/subdev/mc/nv44.c
index d9891782bf28..3bfee5c6c4f2 100644
--- a/drivers/gpu/drm/nouveau/core/subdev/mc/nv44.c
+++ b/drivers/gpu/drm/nouveau/core/subdev/mc/nv44.c
@@ -22,32 +22,12 @@
  * Authors: Ben Skeggs
  */
 
-#include <subdev/mc.h>
-
-struct nv44_mc_priv {
-	struct nouveau_mc base;
-};
-
-static int
-nv44_mc_ctor(struct nouveau_object *parent, struct nouveau_object *engine,
-	     struct nouveau_oclass *oclass, void *data, u32 size,
-	     struct nouveau_object **pobject)
-{
-	struct nv44_mc_priv *priv;
-	int ret;
-
-	ret = nouveau_mc_create(parent, engine, oclass, nv04_mc_intr, &priv);
-	*pobject = nv_object(priv);
-	if (ret)
-		return ret;
-
-	return 0;
-}
+#include "nv04.h"
 
 static int
 nv44_mc_init(struct nouveau_object *object)
 {
-	struct nv44_mc_priv *priv = (void *)object;
+	struct nv04_mc_priv *priv = (void *)object;
 	u32 tmp = nv_rd32(priv, 0x10020c);
 
 	nv_wr32(priv, 0x000200, 0xffffffff); /* everything enabled */
@@ -60,13 +40,15 @@ nv44_mc_init(struct nouveau_object *object)
 	return nouveau_mc_init(&priv->base);
 }
 
-struct nouveau_oclass
-nv44_mc_oclass = {
-	.handle = NV_SUBDEV(MC, 0x44),
-	.ofuncs = &(struct nouveau_ofuncs) {
-		.ctor = nv44_mc_ctor,
+struct nouveau_oclass *
+nv44_mc_oclass = &(struct nouveau_mc_oclass) {
+	.base.handle = NV_SUBDEV(MC, 0x44),
+	.base.ofuncs = &(struct nouveau_ofuncs) {
+		.ctor = nv04_mc_ctor,
 		.dtor = _nouveau_mc_dtor,
 		.init = nv44_mc_init,
 		.fini = _nouveau_mc_fini,
 	},
-};
+	.intr = nv04_mc_intr,
+	.msi_rearm = nv40_mc_msi_rearm,
+}.base;
diff --git a/drivers/gpu/drm/nouveau/core/subdev/mc/nv50.c b/drivers/gpu/drm/nouveau/core/subdev/mc/nv50.c
index 2b1afe225db8..e8822a934c48 100644
--- a/drivers/gpu/drm/nouveau/core/subdev/mc/nv50.c
+++ b/drivers/gpu/drm/nouveau/core/subdev/mc/nv50.c
@@ -22,13 +22,9 @@
  * Authors: Ben Skeggs
  */
 
-#include <subdev/mc.h>
+#include "nv04.h"
 
-struct nv50_mc_priv {
-	struct nouveau_mc base;
-};
-
-static const struct nouveau_mc_intr
+const struct nouveau_mc_intr
 nv50_mc_intr[] = {
 	{ 0x00000001, NVDEV_ENGINE_MPEG },
 	{ 0x00000100, NVDEV_ENGINE_FIFO },
@@ -45,37 +41,30 @@ nv50_mc_intr[] = {
 	{},
 };
 
-static int
-nv50_mc_ctor(struct nouveau_object *parent, struct nouveau_object *engine,
-	     struct nouveau_oclass *oclass, void *data, u32 size,
-	     struct nouveau_object **pobject)
+static void
+nv50_mc_msi_rearm(struct nouveau_mc *pmc)
 {
-	struct nv50_mc_priv *priv;
-	int ret;
-
-	ret = nouveau_mc_create(parent, engine, oclass, nv50_mc_intr, &priv);
-	*pobject = nv_object(priv);
-	if (ret)
-		return ret;
-
-	return 0;
+	struct nouveau_device *device = nv_device(pmc);
+	pci_write_config_byte(device->pdev, 0x68, 0xff);
 }
 
 int
 nv50_mc_init(struct nouveau_object *object)
 {
-	struct nv50_mc_priv *priv = (void *)object;
+	struct nv04_mc_priv *priv = (void *)object;
 	nv_wr32(priv, 0x000200, 0xffffffff); /* everything on */
 	return nouveau_mc_init(&priv->base);
 }
 
-struct nouveau_oclass
-nv50_mc_oclass = {
-	.handle = NV_SUBDEV(MC, 0x50),
-	.ofuncs = &(struct nouveau_ofuncs) {
-		.ctor = nv50_mc_ctor,
+struct nouveau_oclass *
+nv50_mc_oclass = &(struct nouveau_mc_oclass) {
+	.base.handle = NV_SUBDEV(MC, 0x50),
+	.base.ofuncs = &(struct nouveau_ofuncs) {
+		.ctor = nv04_mc_ctor,
 		.dtor = _nouveau_mc_dtor,
 		.init = nv50_mc_init,
 		.fini = _nouveau_mc_fini,
 	},
-};
+	.intr = nv50_mc_intr,
+	.msi_rearm = nv50_mc_msi_rearm,
+}.base;
diff --git a/drivers/gpu/drm/nouveau/core/subdev/mc/nv94.c b/drivers/gpu/drm/nouveau/core/subdev/mc/nv94.c
new file mode 100644
index 000000000000..5f4541105e73
--- /dev/null
+++ b/drivers/gpu/drm/nouveau/core/subdev/mc/nv94.c
@@ -0,0 +1,38 @@
+/*
+ * Copyright 2012 Red Hat Inc.
+ *
+ * Permission is hereby granted, free of charge, to any person obtaining a
+ * copy of this software and associated documentation files (the "Software"),
+ * to deal in the Software without restriction, including without limitation
+ * the rights to use, copy, modify, merge, publish, distribute, sublicense,
+ * and/or sell copies of the Software, and to permit persons to whom the
+ * Software is furnished to do so, subject to the following conditions:
+ *
+ * The above copyright notice and this permission notice shall be included in
+ * all copies or substantial portions of the Software.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+ * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+ * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT.  IN NO EVENT SHALL
+ * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR
+ * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE,
+ * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR
+ * OTHER DEALINGS IN THE SOFTWARE.
+ *
+ * Authors: Ben Skeggs
+ */
+
+#include "nv04.h"
+
+struct nouveau_oclass *
+nv94_mc_oclass = &(struct nouveau_mc_oclass) {
+	.base.handle = NV_SUBDEV(MC, 0x94),
+	.base.ofuncs = &(struct nouveau_ofuncs) {
+		.ctor = nv04_mc_ctor,
+		.dtor = _nouveau_mc_dtor,
+		.init = nv50_mc_init,
+		.fini = _nouveau_mc_fini,
+	},
+	.intr = nv50_mc_intr,
+	.msi_rearm = nv40_mc_msi_rearm,
+}.base;
diff --git a/drivers/gpu/drm/nouveau/core/subdev/mc/nv98.c b/drivers/gpu/drm/nouveau/core/subdev/mc/nv98.c
index 06710419a59b..f8a6f18e2d34 100644
--- a/drivers/gpu/drm/nouveau/core/subdev/mc/nv98.c
+++ b/drivers/gpu/drm/nouveau/core/subdev/mc/nv98.c
@@ -22,11 +22,7 @@
  * Authors: Ben Skeggs
  */
 
-#include <subdev/mc.h>
-
-struct nv98_mc_priv {
-	struct nouveau_mc base;
-};
+#include "nv04.h"
 
 static const struct nouveau_mc_intr
 nv98_mc_intr[] = {
@@ -36,6 +32,7 @@ nv98_mc_intr[] = {
 	{ 0x00004000, NVDEV_ENGINE_CRYPT },	/* NV84:NVA3 */
 	{ 0x00008000, NVDEV_ENGINE_BSP },
 	{ 0x00020000, NVDEV_ENGINE_VP },
+	{ 0x00040000, NVDEV_SUBDEV_PWR },	/* NVA3:NVC0 */
 	{ 0x00080000, NVDEV_SUBDEV_THERM },	/* NVA3:NVC0 */
 	{ 0x00100000, NVDEV_SUBDEV_TIMER },
 	{ 0x00200000, NVDEV_SUBDEV_GPIO },
@@ -47,29 +44,15 @@ nv98_mc_intr[] = {
 	{},
 };
 
-static int
-nv98_mc_ctor(struct nouveau_object *parent, struct nouveau_object *engine,
-	     struct nouveau_oclass *oclass, void *data, u32 size,
-	     struct nouveau_object **pobject)
-{
-	struct nv98_mc_priv *priv;
-	int ret;
-
-	ret = nouveau_mc_create(parent, engine, oclass, nv98_mc_intr, &priv);
-	*pobject = nv_object(priv);
-	if (ret)
-		return ret;
-
-	return 0;
-}
-
-struct nouveau_oclass
-nv98_mc_oclass = {
-	.handle = NV_SUBDEV(MC, 0x98),
-	.ofuncs = &(struct nouveau_ofuncs) {
-		.ctor = nv98_mc_ctor,
+struct nouveau_oclass *
+nv98_mc_oclass = &(struct nouveau_mc_oclass) {
+	.base.handle = NV_SUBDEV(MC, 0x98),
+	.base.ofuncs = &(struct nouveau_ofuncs) {
+		.ctor = nv04_mc_ctor,
 		.dtor = _nouveau_mc_dtor,
 		.init = nv50_mc_init,
 		.fini = _nouveau_mc_fini,
 	},
-};
+	.intr = nv98_mc_intr,
+	.msi_rearm = nv40_mc_msi_rearm,
+}.base;
diff --git a/drivers/gpu/drm/nouveau/core/subdev/mc/nvc0.c b/drivers/gpu/drm/nouveau/core/subdev/mc/nvc0.c
index 104175c5a2dd..c02b4763a2d5 100644
--- a/drivers/gpu/drm/nouveau/core/subdev/mc/nvc0.c
+++ b/drivers/gpu/drm/nouveau/core/subdev/mc/nvc0.c
@@ -22,13 +22,9 @@
  * Authors: Ben Skeggs
  */
 
-#include <subdev/mc.h>
+#include "nv04.h"
 
-struct nvc0_mc_priv {
-	struct nouveau_mc base;
-};
-
-static const struct nouveau_mc_intr
+const struct nouveau_mc_intr
 nvc0_mc_intr[] = {
 	{ 0x00000001, NVDEV_ENGINE_PPP },
 	{ 0x00000020, NVDEV_ENGINE_COPY0 },
@@ -41,6 +37,7 @@ nvc0_mc_intr[] = {
 	{ 0x00020000, NVDEV_ENGINE_VP },
 	{ 0x00100000, NVDEV_SUBDEV_TIMER },
 	{ 0x00200000, NVDEV_SUBDEV_GPIO },
+	{ 0x01000000, NVDEV_SUBDEV_PWR },
 	{ 0x02000000, NVDEV_SUBDEV_LTCG },
 	{ 0x04000000, NVDEV_ENGINE_DISP },
 	{ 0x10000000, NVDEV_SUBDEV_BUS },
@@ -49,29 +46,22 @@ nvc0_mc_intr[] = {
 	{},
 };
 
-static int
-nvc0_mc_ctor(struct nouveau_object *parent, struct nouveau_object *engine,
-	     struct nouveau_oclass *oclass, void *data, u32 size,
-	     struct nouveau_object **pobject)
+static void
+nvc0_mc_msi_rearm(struct nouveau_mc *pmc)
 {
-	struct nvc0_mc_priv *priv;
-	int ret;
-
-	ret = nouveau_mc_create(parent, engine, oclass, nvc0_mc_intr, &priv);
-	*pobject = nv_object(priv);
-	if (ret)
-		return ret;
-
-	return 0;
+	struct nv04_mc_priv *priv = (void *)pmc;
+	nv_wr32(priv, 0x088704, 0x00000000);
 }
 
-struct nouveau_oclass
-nvc0_mc_oclass = {
-	.handle = NV_SUBDEV(MC, 0xc0),
-	.ofuncs = &(struct nouveau_ofuncs) {
-		.ctor = nvc0_mc_ctor,
+struct nouveau_oclass *
+nvc0_mc_oclass = &(struct nouveau_mc_oclass) {
+	.base.handle = NV_SUBDEV(MC, 0xc0),
+	.base.ofuncs = &(struct nouveau_ofuncs) {
+		.ctor = nv04_mc_ctor,
 		.dtor = _nouveau_mc_dtor,
 		.init = nv50_mc_init,
 		.fini = _nouveau_mc_fini,
 	},
-};
+	.intr = nvc0_mc_intr,
+	.msi_rearm = nvc0_mc_msi_rearm,
+}.base;
diff --git a/drivers/gpu/drm/nouveau/core/subdev/mc/nvc3.c b/drivers/gpu/drm/nouveau/core/subdev/mc/nvc3.c
new file mode 100644
index 000000000000..837e545aeb9f
--- /dev/null
+++ b/drivers/gpu/drm/nouveau/core/subdev/mc/nvc3.c
@@ -0,0 +1,38 @@
+/*
+ * Copyright 2012 Red Hat Inc.
+ *
+ * Permission is hereby granted, free of charge, to any person obtaining a
+ * copy of this software and associated documentation files (the "Software"),
+ * to deal in the Software without restriction, including without limitation
+ * the rights to use, copy, modify, merge, publish, distribute, sublicense,
+ * and/or sell copies of the Software, and to permit persons to whom the
+ * Software is furnished to do so, subject to the following conditions:
+ *
+ * The above copyright notice and this permission notice shall be included in
+ * all copies or substantial portions of the Software.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+ * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+ * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT.  IN NO EVENT SHALL
+ * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR
+ * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE,
+ * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR
+ * OTHER DEALINGS IN THE SOFTWARE.
+ *
+ * Authors: Ben Skeggs
+ */
+
+#include "nv04.h"
+
+struct nouveau_oclass *
+nvc3_mc_oclass = &(struct nouveau_mc_oclass) {
+	.base.handle = NV_SUBDEV(MC, 0xc3),
+	.base.ofuncs = &(struct nouveau_ofuncs) {
+		.ctor = nv04_mc_ctor,
+		.dtor = _nouveau_mc_dtor,
+		.init = nv50_mc_init,
+		.fini = _nouveau_mc_fini,
+	},
+	.intr = nvc0_mc_intr,
+	.msi_rearm = nv40_mc_msi_rearm,
+}.base;
diff --git a/drivers/gpu/drm/nouveau/core/subdev/pwr/base.c b/drivers/gpu/drm/nouveau/core/subdev/pwr/base.c
new file mode 100644
index 000000000000..d4fd3bc9c66f
--- /dev/null
+++ b/drivers/gpu/drm/nouveau/core/subdev/pwr/base.c
@@ -0,0 +1,247 @@
+/*
+ * Copyright 2013 Red Hat Inc.
+ *
+ * Permission is hereby granted, free of charge, to any person obtaining a
+ * copy of this software and associated documentation files (the "Software"),
+ * to deal in the Software without restriction, including without limitation
+ * the rights to use, copy, modify, merge, publish, distribute, sublicense,
+ * and/or sell copies of the Software, and to permit persons to whom the
+ * Software is furnished to do so, subject to the following conditions:
+ *
+ * The above copyright notice and this permission notice shall be included in
+ * all copies or substantial portions of the Software.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+ * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+ * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT.  IN NO EVENT SHALL
+ * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR
+ * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE,
+ * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR
+ * OTHER DEALINGS IN THE SOFTWARE.
+ *
+ * Authors: Ben Skeggs
+ */
+
+#include <subdev/pwr.h>
+#include <subdev/timer.h>
+
+static int
+nouveau_pwr_send(struct nouveau_pwr *ppwr, u32 reply[2],
+		 u32 process, u32 message, u32 data0, u32 data1)
+{
+	struct nouveau_subdev *subdev = nv_subdev(ppwr);
+	u32 addr;
+
+	/* wait for a free slot in the fifo */
+	addr  = nv_rd32(ppwr, 0x10a4a0);
+	if (!nv_wait_ne(ppwr, 0x10a4b0, 0xffffffff, addr ^ 8))
+		return -EBUSY;
+
+	/* we currently only support a single process at a time waiting
+	 * on a synchronous reply, take the PPWR mutex and tell the
+	 * receive handler what we're waiting for
+	 */
+	if (reply) {
+		mutex_lock(&subdev->mutex);
+		ppwr->recv.message = message;
+		ppwr->recv.process = process;
+	}
+
+	/* acquire data segment access */
+	do {
+		nv_wr32(ppwr, 0x10a580, 0x00000001);
+	} while (nv_rd32(ppwr, 0x10a580) != 0x00000001);
+
+	/* write the packet */
+	nv_wr32(ppwr, 0x10a1c0, 0x01000000 | (((addr & 0x07) << 4) +
+				ppwr->send.base));
+	nv_wr32(ppwr, 0x10a1c4, process);
+	nv_wr32(ppwr, 0x10a1c4, message);
+	nv_wr32(ppwr, 0x10a1c4, data0);
+	nv_wr32(ppwr, 0x10a1c4, data1);
+	nv_wr32(ppwr, 0x10a4a0, (addr + 1) & 0x0f);
+
+	/* release data segment access */
+	nv_wr32(ppwr, 0x10a580, 0x00000000);
+
+	/* wait for reply, if requested */
+	if (reply) {
+		wait_event(ppwr->recv.wait, (ppwr->recv.process == 0));
+		reply[0] = ppwr->recv.data[0];
+		reply[1] = ppwr->recv.data[1];
+		mutex_unlock(&subdev->mutex);
+	}
+
+	return 0;
+}
+
+static void
+nouveau_pwr_recv(struct work_struct *work)
+{
+	struct nouveau_pwr *ppwr =
+		container_of(work, struct nouveau_pwr, recv.work);
+	u32 process, message, data0, data1;
+
+	/* nothing to do if GET == PUT */
+	u32 addr =  nv_rd32(ppwr, 0x10a4cc);
+	if (addr == nv_rd32(ppwr, 0x10a4c8))
+		return;
+
+	/* acquire data segment access */
+	do {
+		nv_wr32(ppwr, 0x10a580, 0x00000002);
+	} while (nv_rd32(ppwr, 0x10a580) != 0x00000002);
+
+	/* read the packet */
+	nv_wr32(ppwr, 0x10a1c0, 0x02000000 | (((addr & 0x07) << 4) +
+				ppwr->recv.base));
+	process = nv_rd32(ppwr, 0x10a1c4);
+	message = nv_rd32(ppwr, 0x10a1c4);
+	data0   = nv_rd32(ppwr, 0x10a1c4);
+	data1   = nv_rd32(ppwr, 0x10a1c4);
+	nv_wr32(ppwr, 0x10a4cc, (addr + 1) & 0x0f);
+
+	/* release data segment access */
+	nv_wr32(ppwr, 0x10a580, 0x00000000);
+
+	/* wake process if it's waiting on a synchronous reply */
+	if (ppwr->recv.process) {
+		if (process == ppwr->recv.process &&
+		    message == ppwr->recv.message) {
+			ppwr->recv.data[0] = data0;
+			ppwr->recv.data[1] = data1;
+			ppwr->recv.process = 0;
+			wake_up(&ppwr->recv.wait);
+			return;
+		}
+	}
+
+	/* right now there's no other expected responses from the engine,
+	 * so assume that any unexpected message is an error.
+	 */
+	nv_warn(ppwr, "%c%c%c%c 0x%08x 0x%08x 0x%08x 0x%08x\n",
+		(char)((process & 0x000000ff) >>  0),
+		(char)((process & 0x0000ff00) >>  8),
+		(char)((process & 0x00ff0000) >> 16),
+		(char)((process & 0xff000000) >> 24),
+		process, message, data0, data1);
+}
+
+static void
+nouveau_pwr_intr(struct nouveau_subdev *subdev)
+{
+	struct nouveau_pwr *ppwr = (void *)subdev;
+	u32 disp = nv_rd32(ppwr, 0x10a01c);
+	u32 intr = nv_rd32(ppwr, 0x10a008) & disp & ~(disp >> 16);
+
+	if (intr & 0x00000020) {
+		u32 stat = nv_rd32(ppwr, 0x10a16c);
+		if (stat & 0x80000000) {
+			nv_error(ppwr, "UAS fault at 0x%06x addr 0x%08x\n",
+				 stat & 0x00ffffff, nv_rd32(ppwr, 0x10a168));
+			nv_wr32(ppwr, 0x10a16c, 0x00000000);
+			intr &= ~0x00000020;
+		}
+	}
+
+	if (intr & 0x00000040) {
+		schedule_work(&ppwr->recv.work);
+		nv_wr32(ppwr, 0x10a004, 0x00000040);
+		intr &= ~0x00000040;
+	}
+
+	if (intr & 0x00000080) {
+		nv_info(ppwr, "wr32 0x%06x 0x%08x\n", nv_rd32(ppwr, 0x10a7a0),
+						      nv_rd32(ppwr, 0x10a7a4));
+		nv_wr32(ppwr, 0x10a004, 0x00000080);
+		intr &= ~0x00000080;
+	}
+
+	if (intr) {
+		nv_error(ppwr, "intr 0x%08x\n", intr);
+		nv_wr32(ppwr, 0x10a004, intr);
+	}
+}
+
+int
+_nouveau_pwr_fini(struct nouveau_object *object, bool suspend)
+{
+	struct nouveau_pwr *ppwr = (void *)object;
+
+	nv_wr32(ppwr, 0x10a014, 0x00000060);
+	flush_work(&ppwr->recv.work);
+
+	return nouveau_subdev_fini(&ppwr->base, suspend);
+}
+
+int
+_nouveau_pwr_init(struct nouveau_object *object)
+{
+	struct nouveau_pwr *ppwr = (void *)object;
+	int ret, i;
+
+	ret = nouveau_subdev_init(&ppwr->base);
+	if (ret)
+		return ret;
+
+	nv_subdev(ppwr)->intr = nouveau_pwr_intr;
+	ppwr->message = nouveau_pwr_send;
+
+	/* prevent previous ucode from running, wait for idle, reset */
+	nv_wr32(ppwr, 0x10a014, 0x0000ffff); /* INTR_EN_CLR = ALL */
+	nv_wait(ppwr, 0x10a04c, 0xffffffff, 0x00000000);
+	nv_mask(ppwr, 0x000200, 0x00002000, 0x00000000);
+	nv_mask(ppwr, 0x000200, 0x00002000, 0x00002000);
+
+	/* upload data segment */
+	nv_wr32(ppwr, 0x10a1c0, 0x01000000);
+	for (i = 0; i < ppwr->data.size / 4; i++)
+		nv_wr32(ppwr, 0x10a1c4, ppwr->data.data[i]);
+
+	/* upload code segment */
+	nv_wr32(ppwr, 0x10a180, 0x01000000);
+	for (i = 0; i < ppwr->code.size / 4; i++) {
+		if ((i & 0x3f) == 0)
+			nv_wr32(ppwr, 0x10a188, i >> 6);
+		nv_wr32(ppwr, 0x10a184, ppwr->code.data[i]);
+	}
+
+	/* start it running */
+	nv_wr32(ppwr, 0x10a10c, 0x00000000);
+	nv_wr32(ppwr, 0x10a104, 0x00000000);
+	nv_wr32(ppwr, 0x10a100, 0x00000002);
+
+	/* wait for valid host->pwr ring configuration */
+	if (!nv_wait_ne(ppwr, 0x10a4d0, 0xffffffff, 0x00000000))
+		return -EBUSY;
+	ppwr->send.base = nv_rd32(ppwr, 0x10a4d0) & 0x0000ffff;
+	ppwr->send.size = nv_rd32(ppwr, 0x10a4d0) >> 16;
+
+	/* wait for valid pwr->host ring configuration */
+	if (!nv_wait_ne(ppwr, 0x10a4dc, 0xffffffff, 0x00000000))
+		return -EBUSY;
+	ppwr->recv.base = nv_rd32(ppwr, 0x10a4dc) & 0x0000ffff;
+	ppwr->recv.size = nv_rd32(ppwr, 0x10a4dc) >> 16;
+
+	nv_wr32(ppwr, 0x10a010, 0x000000e0);
+	return 0;
+}
+
+int
+nouveau_pwr_create_(struct nouveau_object *parent,
+		    struct nouveau_object *engine,
+		    struct nouveau_oclass *oclass, int length, void **pobject)
+{
+	struct nouveau_pwr *ppwr;
+	int ret;
+
+	ret = nouveau_subdev_create_(parent, engine, oclass, 0, "PPWR",
+				     "pwr", length, pobject);
+	ppwr = *pobject;
+	if (ret)
+		return ret;
+
+	INIT_WORK(&ppwr->recv.work, nouveau_pwr_recv);
+	init_waitqueue_head(&ppwr->recv.wait);
+	return 0;
+}
diff --git a/drivers/gpu/drm/nouveau/core/subdev/pwr/fuc/host.fuc b/drivers/gpu/drm/nouveau/core/subdev/pwr/fuc/host.fuc
new file mode 100644
index 000000000000..2284ecb1c9b8
--- /dev/null
+++ b/drivers/gpu/drm/nouveau/core/subdev/pwr/fuc/host.fuc
@@ -0,0 +1,151 @@
+/*
+ * Copyright 2013 Red Hat Inc.
+ *
+ * Permission is hereby granted, free of charge, to any person obtaining a
+ * copy of this software and associated documentation files (the "Software"),
+ * to deal in the Software without restriction, including without limitation
+ * the rights to use, copy, modify, merge, publish, distribute, sublicense,
+ * and/or sell copies of the Software, and to permit persons to whom the
+ * Software is furnished to do so, subject to the following conditions:
+ *
+ * The above copyright notice and this permission notice shall be included in
+ * all copies or substantial portions of the Software.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+ * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+ * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT.  IN NO EVENT SHALL
+ * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR
+ * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE,
+ * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR
+ * OTHER DEALINGS IN THE SOFTWARE.
+ *
+ * Authors: Ben Skeggs
+ */
+
+#ifdef INCLUDE_PROC
+process(PROC_HOST, #host_init, #host_recv)
+#endif
+
+/******************************************************************************
+ * HOST data segment
+ *****************************************************************************/
+#ifdef INCLUDE_DATA
+// HOST (R)FIFO packet format
+.equ #fifo_process 0x00
+.equ #fifo_message 0x04
+.equ #fifo_data0   0x08
+.equ #fifo_data1   0x0c
+
+// HOST HOST->PWR queue description
+.equ #fifo_qlen 4 // log2(size of queue entry in bytes)
+.equ #fifo_qnum 3 // log2(max number of entries in queue)
+.equ #fifo_qmaskb (1 << #fifo_qnum) // max number of entries in queue
+.equ #fifo_qmaskp (#fifo_qmaskb - 1)
+.equ #fifo_qmaskf ((#fifo_qmaskb << 1) - 1)
+.equ #fifo_qsize  (1 << (#fifo_qlen + #fifo_qnum))
+fifo_queue: .skip 128 // #fifo_qsize
+
+// HOST PWR->HOST queue description
+.equ #rfifo_qlen 4 // log2(size of queue entry in bytes)
+.equ #rfifo_qnum 3 // log2(max number of entries in queue)
+.equ #rfifo_qmaskb (1 << #rfifo_qnum) // max number of entries in queue
+.equ #rfifo_qmaskp (#rfifo_qmaskb - 1)
+.equ #rfifo_qmaskf ((#rfifo_qmaskb << 1) - 1)
+.equ #rfifo_qsize  (1 << (#rfifo_qlen + #rfifo_qnum))
+rfifo_queue: .skip 128 // #rfifo_qsize
+#endif
+
+/******************************************************************************
+ * HOST code segment
+ *****************************************************************************/
+#ifdef INCLUDE_CODE
+// HOST->PWR comms - dequeue message(s) for process(es) from FIFO
+//
+// $r15 - current (host)
+// $r0  - zero
+host_send:
+	nv_iord($r1, NV_PPWR_FIFO_GET(0))
+	nv_iord($r2, NV_PPWR_FIFO_PUT(0))
+	cmp b32 $r1 $r2
+	bra e #host_send_done
+		// calculate address of message
+		and $r14 $r1 #fifo_qmaskp
+		shl b32 $r14 $r14 #fifo_qlen
+		add b32 $r14 #fifo_queue
+
+		// read message data, and pass to appropriate process
+		ld b32 $r11 D[$r14 + #fifo_data1]
+		ld b32 $r12 D[$r14 + #fifo_data0]
+		ld b32 $r13 D[$r14 + #fifo_message]
+		ld b32 $r14 D[$r14 + #fifo_process]
+		call(send)
+
+		// increment GET
+		add b32 $r1 0x1
+		and $r14 $r1 #fifo_qmaskf
+		nv_iowr(NV_PPWR_FIFO_GET(0), $r1)
+		bra #host_send
+	host_send_done:
+	ret
+
+// PWR->HOST comms - enqueue message for HOST to RFIFO
+//
+// $r15 - current (host)
+// $r14 - process
+// $r13 - message
+// $r12 - message data 0
+// $r11 - message data 1
+// $r0  - zero
+host_recv:
+	// message from intr handler == HOST->PWR comms pending
+	mov $r1 (PROC_KERN & 0x0000ffff)
+	sethi $r1 (PROC_KERN & 0xffff0000)
+	cmp b32 $r14 $r1
+	bra e #host_send
+
+	// wait for space in RFIFO
+	host_recv_wait:
+	nv_iord($r1, NV_PPWR_RFIFO_GET)
+	nv_iord($r2, NV_PPWR_RFIFO_PUT)
+	xor $r1 #rfifo_qmaskb
+	cmp b32 $r1 $r2
+	bra e #host_recv_wait
+
+	and $r3 $r2 #rfifo_qmaskp
+	shl b32 $r3 #rfifo_qlen
+	add b32 $r3 #rfifo_queue
+
+	// enqueue message
+	st b32 D[$r3 + #fifo_data1] $r11
+	st b32 D[$r3 + #fifo_data0] $r12
+	st b32 D[$r3 + #fifo_message] $r13
+	st b32 D[$r3 + #fifo_process] $r14
+
+	add b32 $r2 0x1
+	and $r2 #rfifo_qmaskf
+	nv_iowr(NV_PPWR_RFIFO_PUT, $r2)
+
+	// notify host of pending message
+	mov $r2 NV_PPWR_INTR_TRIGGER_USER0
+	nv_iowr(NV_PPWR_INTR_TRIGGER, $r2)
+	ret
+
+// $r15 - current (host)
+// $r0  - zero
+host_init:
+	// store each fifo's base/size in H2D/D2H scratch regs
+	mov $r1 #fifo_qsize
+	shl b32 $r1 16
+	or $r1 #fifo_queue
+	nv_iowr(NV_PPWR_H2D, $r1);
+
+	mov $r1 #rfifo_qsize
+	shl b32 $r1 16
+	or $r1 #rfifo_queue
+	nv_iowr(NV_PPWR_D2H, $r1);
+
+	// enable fifo subintr for first fifo
+	mov $r1 1
+	nv_iowr(NV_PPWR_FIFO_INTR_EN, $r1)
+	ret
+#endif
diff --git a/drivers/gpu/drm/nouveau/core/subdev/pwr/fuc/idle.fuc b/drivers/gpu/drm/nouveau/core/subdev/pwr/fuc/idle.fuc
new file mode 100644
index 000000000000..98f1c3738b42
--- /dev/null
+++ b/drivers/gpu/drm/nouveau/core/subdev/pwr/fuc/idle.fuc
@@ -0,0 +1,84 @@
+/*
+ * Copyright 2013 Red Hat Inc.
+ *
+ * Permission is hereby granted, free of charge, to any person obtaining a
+ * copy of this software and associated documentation files (the "Software"),
+ * to deal in the Software without restriction, including without limitation
+ * the rights to use, copy, modify, merge, publish, distribute, sublicense,
+ * and/or sell copies of the Software, and to permit persons to whom the
+ * Software is furnished to do so, subject to the following conditions:
+ *
+ * The above copyright notice and this permission notice shall be included in
+ * all copies or substantial portions of the Software.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+ * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+ * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT.  IN NO EVENT SHALL
+ * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR
+ * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE,
+ * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR
+ * OTHER DEALINGS IN THE SOFTWARE.
+ *
+ * Authors: Ben Skeggs
+ */
+
+#ifdef INCLUDE_PROC
+process(PROC_IDLE, #idle, #idle_recv)
+#endif
+
+/******************************************************************************
+ * IDLE data segment
+ *****************************************************************************/
+#ifdef INCLUDE_DATA
+#endif
+
+/******************************************************************************
+ * IDLE code segment
+ *****************************************************************************/
+#ifdef INCLUDE_CODE
+// description
+//
+// $r15 - current (idle)
+// $r14 - message
+// $r0  - zero
+idle_recv:
+	ret
+
+// description
+//
+// $r15 - current (idle)
+// $r0  - zero
+idle:
+	// set our "no interrupt has occurred during our execution" flag
+	bset $flags $p0
+
+	// count IDLE invocations for debugging purposes
+	nv_iord($r1, NV_PPWR_DSCRATCH(1))
+	add b32 $r1 1
+	nv_iowr(NV_PPWR_DSCRATCH(1), $r1)
+
+	// keep looping while there's pending messages for any process
+	idle_loop:
+	mov $r1 #proc_list_head
+	bclr $flags $p2
+	idle_proc:
+		// process the process' messages until there's none left
+		idle_proc_exec:
+			push $r1
+			mov b32 $r14 $r1
+			call(recv)
+			pop $r1
+			bra not $p1 #idle_proc_next
+			bset $flags $p2
+			bra #idle_proc_exec
+		// next process!
+		idle_proc_next:
+		add b32 $r1 #proc_size
+		cmp b32 $r1 $r15
+		bra ne #idle_proc
+	bra $p2 #idle_loop
+
+	// sleep if no interrupts have occurred
+	sleep $p0
+	bra #idle
+#endif
diff --git a/drivers/gpu/drm/nouveau/core/subdev/pwr/fuc/kernel.fuc b/drivers/gpu/drm/nouveau/core/subdev/pwr/fuc/kernel.fuc
new file mode 100644
index 000000000000..0a7b05fa5c11
--- /dev/null
+++ b/drivers/gpu/drm/nouveau/core/subdev/pwr/fuc/kernel.fuc
@@ -0,0 +1,452 @@
+/*
+ * Copyright 2013 Red Hat Inc.
+ *
+ * Permission is hereby granted, free of charge, to any person obtaining a
+ * copy of this software and associated documentation files (the "Software"),
+ * to deal in the Software without restriction, including without limitation
+ * the rights to use, copy, modify, merge, publish, distribute, sublicense,
+ * and/or sell copies of the Software, and to permit persons to whom the
+ * Software is furnished to do so, subject to the following conditions:
+ *
+ * The above copyright notice and this permission notice shall be included in
+ * all copies or substantial portions of the Software.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+ * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+ * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT.  IN NO EVENT SHALL
+ * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR
+ * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE,
+ * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR
+ * OTHER DEALINGS IN THE SOFTWARE.
+ *
+ * Authors: Ben Skeggs
+ */
+
+/******************************************************************************
+ * kernel data segment
+ *****************************************************************************/
+#ifdef INCLUDE_PROC
+proc_kern:
+process(PROC_KERN, 0, 0)
+proc_list_head:
+#endif
+
+#ifdef INCLUDE_DATA
+proc_list_tail:
+time_prev: .b32 0
+time_next: .b32 0
+#endif
+
+/******************************************************************************
+ * kernel code segment
+ *****************************************************************************/
+#ifdef INCLUDE_CODE
+	bra #init
+
+// read nv register
+//
+// $r15 - current
+// $r14 - addr
+// $r13 - data (return)
+// $r0  - zero
+rd32:
+	nv_iowr(NV_PPWR_MMIO_ADDR, $r14)
+	mov $r14 NV_PPWR_MMIO_CTRL_OP_RD
+	sethi $r14 NV_PPWR_MMIO_CTRL_TRIGGER
+	nv_iowr(NV_PPWR_MMIO_CTRL, $r14)
+	rd32_wait:
+		nv_iord($r14, NV_PPWR_MMIO_CTRL)
+		and $r14 NV_PPWR_MMIO_CTRL_STATUS
+		bra nz #rd32_wait
+	nv_iord($r13, NV_PPWR_MMIO_DATA)
+	ret
+
+// write nv register
+//
+// $r15 - current
+// $r14 - addr
+// $r13 - data
+// $r0  - zero
+wr32:
+	nv_iowr(NV_PPWR_MMIO_ADDR, $r14)
+	nv_iowr(NV_PPWR_MMIO_DATA, $r13)
+	mov $r14 NV_PPWR_MMIO_CTRL_OP_WR
+	or $r14 NV_PPWR_MMIO_CTRL_MASK_B32_0
+	sethi $r14 NV_PPWR_MMIO_CTRL_TRIGGER
+
+#ifdef NVKM_FALCON_MMIO_TRAP
+	mov $r8 NV_PPWR_INTR_TRIGGER_USER1
+	nv_iowr(NV_PPWR_INTR_TRIGGER, $r8)
+	wr32_host:
+		nv_iord($r8, NV_PPWR_INTR)
+		and $r8 NV_PPWR_INTR_USER1
+		bra nz #wr32_host
+#endif
+
+	nv_iowr(NV_PPWR_MMIO_CTRL, $r14)
+	wr32_wait:
+		nv_iord($r14, NV_PPWR_MMIO_CTRL)
+		and $r14 NV_PPWR_MMIO_CTRL_STATUS
+		bra nz #wr32_wait
+	ret
+
+// busy-wait for a period of time
+//
+// $r15 - current
+// $r14 - ns
+// $r0  - zero
+nsec:
+	nv_iord($r8, NV_PPWR_TIMER_LOW)
+	nsec_loop:
+		nv_iord($r9, NV_PPWR_TIMER_LOW)
+		sub b32 $r9 $r8
+		cmp b32 $r9 $r14
+		bra l #nsec_loop
+	ret
+
+// busy-wait for a period of time
+//
+// $r15 - current
+// $r14 - addr
+// $r13 - mask
+// $r12 - data
+// $r11 - timeout (ns)
+// $r0  - zero
+wait:
+	nv_iord($r8, NV_PPWR_TIMER_LOW)
+	wait_loop:
+		nv_rd32($r10, $r14)
+		and $r10 $r13
+		cmp b32 $r10 $r12
+		bra e #wait_done
+		nv_iord($r9, NV_PPWR_TIMER_LOW)
+		sub b32 $r9 $r8
+		cmp b32 $r9 $r11
+		bra l #wait_loop
+	wait_done:
+	ret
+
+// $r15 - current (kern)
+// $r14 - process
+// $r8  - NV_PPWR_INTR
+intr_watchdog:
+	// read process' timer status, skip if not enabled
+	ld b32 $r9 D[$r14 + #proc_time]
+	cmp b32 $r9 0
+	bra z #intr_watchdog_next_proc
+
+	// subtract last timer's value from process' timer,
+	// if it's <= 0 then the timer has expired
+	ld b32 $r10 D[$r0 + #time_prev]
+	sub b32 $r9 $r10
+	bra g #intr_watchdog_next_time
+		mov $r13 KMSG_ALARM
+		call(send_proc)
+		clear b32 $r9
+		bra #intr_watchdog_next_proc
+
+	// otherwise, update the next timer's value if this
+	// process' timer is the soonest
+	intr_watchdog_next_time:
+		// ... or if there's no next timer yet
+		ld b32 $r10 D[$r0 + #time_next]
+		cmp b32 $r10 0
+		bra z #intr_watchdog_next_time_set
+
+		cmp b32 $r9 $r10
+		bra g #intr_watchdog_next_proc
+		intr_watchdog_next_time_set:
+		st b32 D[$r0 + #time_next] $r9
+
+	// update process' timer status, and advance
+	intr_watchdog_next_proc:
+	st b32 D[$r14 + #proc_time] $r9
+	add b32 $r14 #proc_size
+	cmp b32 $r14 #proc_list_tail
+	bra ne #intr_watchdog
+	ret
+
+intr:
+	push $r0
+	clear b32 $r0
+	push $r8
+	push $r9
+	push $r10
+	push $r11
+	push $r12
+	push $r13
+	push $r14
+	push $r15
+	mov $r15 #proc_kern
+	mov $r8 $flags
+	push $r8
+
+	nv_iord($r8, NV_PPWR_DSCRATCH(0))
+	add b32 $r8 1
+	nv_iowr(NV_PPWR_DSCRATCH(0), $r8)
+
+	nv_iord($r8, NV_PPWR_INTR)
+	and $r9 $r8 NV_PPWR_INTR_WATCHDOG
+	bra z #intr_skip_watchdog
+		st b32 D[$r0 + #time_next] $r0
+		mov $r14 #proc_list_head
+		call(intr_watchdog)
+		ld b32 $r9 D[$r0 + #time_next]
+		cmp b32 $r9 0
+		bra z #intr_skip_watchdog
+			nv_iowr(NV_PPWR_WATCHDOG_TIME, $r9)
+			st b32 D[$r0 + #time_prev] $r9
+
+	intr_skip_watchdog:
+	and $r9 $r8 NV_PPWR_INTR_SUBINTR
+	bra z #intr_skip_subintr
+		nv_iord($r9, NV_PPWR_SUBINTR)
+		and $r10 $r9 NV_PPWR_SUBINTR_FIFO
+		bra z #intr_subintr_skip_fifo
+			nv_iord($r12, NV_PPWR_FIFO_INTR)
+			push $r12
+			mov $r14 (PROC_HOST & 0x0000ffff)
+			sethi $r14 (PROC_HOST & 0xffff0000)
+			mov $r13 KMSG_FIFO
+			call(send)
+			pop $r12
+			nv_iowr(NV_PPWR_FIFO_INTR, $r12)
+		intr_subintr_skip_fifo:
+		nv_iowr(NV_PPWR_SUBINTR, $r9)
+
+	intr_skip_subintr:
+	and $r9 $r8 NV_PPWR_INTR_PAUSE
+	bra z #intr_skip_pause
+		and $r10 0xffbf
+
+	intr_skip_pause:
+	and $r9 $r8 NV_PPWR_INTR_USER0
+	bra z #intr_skip_user0
+		and $r10 0xffbf
+
+	intr_skip_user0:
+	nv_iowr(NV_PPWR_INTR_ACK, $r8)
+	pop $r8
+	mov $flags $r8
+	pop $r15
+	pop $r14
+	pop $r13
+	pop $r12
+	pop $r11
+	pop $r10
+	pop $r9
+	pop $r8
+	pop $r0
+	bclr $flags $p0
+	iret
+
+// request the current process be sent a message after a timeout expires
+//
+// $r15 - current
+// $r14 - ticks
+// $r0  - zero
+timer:
+	// interrupts off to prevent racing with timer isr
+	bclr $flags ie0
+
+	// if current process already has a timer set, bail
+	ld b32 $r8 D[$r15 + #proc_time]
+	cmp b32 $r8 0
+	bra g #timer_done
+	st b32 D[$r15 + #proc_time] $r14
+
+	// halt watchdog timer temporarily and check for a pending
+	// interrupt.  if there's one already pending, we can just
+	// bail since the timer isr will queue the next soonest
+	// right after it's done
+	nv_iowr(NV_PPWR_WATCHDOG_ENABLE, $r8)
+	nv_iord($r8, NV_PPWR_INTR)
+	and $r8 NV_PPWR_INTR_WATCHDOG
+	bra nz #timer_enable
+
+	// update the watchdog if this timer should expire first,
+	// or if there's no timeout already set
+	nv_iord($r8, NV_PPWR_WATCHDOG_TIME)
+	cmp b32 $r14 $r0
+	bra e #timer_reset
+	cmp b32 $r14 $r8
+	bra l #timer_done
+	timer_reset:
+	nv_iowr(NV_PPWR_WATCHDOG_TIME, $r14)
+	st b32 D[$r0 + #time_prev] $r14
+
+	// re-enable the watchdog timer
+	timer_enable:
+	mov $r8 1
+	nv_iowr(NV_PPWR_WATCHDOG_ENABLE, $r8)
+
+	// interrupts back on
+	timer_done:
+	bset $flags ie0
+	ret
+
+// send message to another process
+//
+// $r15 - current
+// $r14 - process
+// $r13 - message
+// $r12 - message data 0
+// $r11 - message data 1
+// $r0  - zero
+send_proc:
+	push $r8
+	push $r9
+	// check for space in queue
+	ld b32 $r8 D[$r14 + #proc_qget]
+	ld b32 $r9 D[$r14 + #proc_qput]
+	xor $r8 #proc_qmaskb
+	cmp b32 $r8 $r9
+	bra e #send_done
+
+	// enqueue message
+	and $r8 $r9 #proc_qmaskp
+	shl b32 $r8 $r8 #proc_qlen
+	add b32 $r8 #proc_queue
+	add b32 $r8 $r14
+
+	ld b32 $r10 D[$r15 + #proc_id]
+	st b32 D[$r8 + #msg_process] $r10
+	st b32 D[$r8 + #msg_message] $r13
+	st b32 D[$r8 + #msg_data0] $r12
+	st b32 D[$r8 + #msg_data1] $r11
+
+	// increment PUT
+	add b32 $r9 1
+	and $r9 #proc_qmaskf
+	st b32 D[$r14 + #proc_qput] $r9
+	bset $flags $p2
+	send_done:
+	pop $r9
+	pop $r8
+	ret
+
+// lookup process structure by its name
+//
+// $r15 - current
+// $r14 - process name
+// $r0  - zero
+//
+// $r14 - process
+// $p1  - success
+find:
+	push $r8
+	mov $r8 #proc_list_head
+	bset $flags $p1
+	find_loop:
+		ld b32 $r10 D[$r8 + #proc_id]
+		cmp b32 $r10 $r14
+		bra e #find_done
+		add b32 $r8 #proc_size
+		cmp b32 $r8 #proc_list_tail
+		bra ne #find_loop
+		bclr $flags $p1
+	find_done:
+	mov b32 $r14 $r8
+	pop $r8
+	ret
+
+// send message to another process
+//
+// $r15 - current
+// $r14 - process id
+// $r13 - message
+// $r12 - message data 0
+// $r11 - message data 1
+// $r0  - zero
+send:
+	call(find)
+	bra $p1 #send_proc
+	ret
+
+// process single message for a given process
+//
+// $r15 - current
+// $r14 - process
+// $r0  - zero
+recv:
+	ld b32 $r8 D[$r14 + #proc_qget]
+	ld b32 $r9 D[$r14 + #proc_qput]
+	bclr $flags $p1
+	cmp b32 $r8 $r9
+	bra e #recv_done
+		// dequeue message
+		and $r9 $r8 #proc_qmaskp
+		add b32 $r8 1
+		and $r8 #proc_qmaskf
+		st b32 D[$r14 + #proc_qget] $r8
+		ld b32 $r10 D[$r14 + #proc_recv]
+
+		push $r15
+		mov $r15 $flags
+		push $r15
+		mov b32 $r15 $r14
+
+		shl b32 $r9 $r9 #proc_qlen
+		add b32 $r14 $r9
+		add b32 $r14 #proc_queue
+		ld b32 $r11 D[$r14 + #msg_data1]
+		ld b32 $r12 D[$r14 + #msg_data0]
+		ld b32 $r13 D[$r14 + #msg_message]
+		ld b32 $r14 D[$r14 + #msg_process]
+
+		// process it
+		call $r10
+		pop $r15
+		mov $flags $r15
+		bset $flags $p1
+		pop $r15
+	recv_done:
+	ret
+
+init:
+	// setup stack
+	nv_iord($r1, NV_PPWR_CAPS)
+	extr $r1 $r1 9:17
+	shl b32 $r1 8
+	mov $sp $r1
+
+#ifdef NVKM_FALCON_MMIO_UAS
+	// somehow allows the magic "access mmio via D[]" stuff that's
+	// used by the nv_rd32/nv_wr32 macros to work
+	mov $r1 0x0010
+	sethi $r1 NV_PPWR_UAS_CONFIG_ENABLE
+	nv_iowrs(NV_PPWR_UAS_CONFIG, $r1)
+#endif
+
+	// route all interrupts except user0/1 and pause to fuc
+	mov $r1 0x00e0
+	sethi $r1 0x00000000
+	nv_iowr(NV_PPWR_INTR_ROUTE, $r1)
+
+	// enable watchdog and subintr intrs
+	mov $r1 NV_PPWR_INTR_EN_CLR_MASK
+	nv_iowr(NV_PPWR_INTR_EN_CLR, $r1)
+	mov $r1 NV_PPWR_INTR_EN_SET_WATCHDOG
+	or $r1 NV_PPWR_INTR_EN_SET_SUBINTR
+	nv_iowr(NV_PPWR_INTR_EN_SET, $r1)
+
+	// enable interrupts globally
+	mov $r1 #intr
+	sethi $r1 0x00000000
+	mov $iv0 $r1
+	bset $flags ie0
+
+	// enable watchdog timer
+	mov $r1 1
+	nv_iowr(NV_PPWR_WATCHDOG_ENABLE, $r1)
+
+	// bootstrap processes, idle process will be last, and not return
+	mov $r15 #proc_list_head
+	init_proc:
+		ld b32 $r1 D[$r15 + #proc_init]
+		cmp b32 $r1 0
+		bra z #init_proc
+		call $r1
+		add b32 $r15 #proc_size
+		bra #init_proc
+#endif
diff --git a/drivers/gpu/drm/nouveau/core/subdev/pwr/fuc/macros.fuc b/drivers/gpu/drm/nouveau/core/subdev/pwr/fuc/macros.fuc
new file mode 100644
index 000000000000..2a74ea907604
--- /dev/null
+++ b/drivers/gpu/drm/nouveau/core/subdev/pwr/fuc/macros.fuc
@@ -0,0 +1,199 @@
+/*
+ * Copyright 2013 Red Hat Inc.
+ *
+ * Permission is hereby granted, free of charge, to any person obtaining a
+ * copy of this software and associated documentation files (the "Software"),
+ * to deal in the Software without restriction, including without limitation
+ * the rights to use, copy, modify, merge, publish, distribute, sublicense,
+ * and/or sell copies of the Software, and to permit persons to whom the
+ * Software is furnished to do so, subject to the following conditions:
+ *
+ * The above copyright notice and this permission notice shall be included in
+ * all copies or substantial portions of the Software.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+ * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+ * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT.  IN NO EVENT SHALL
+ * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR
+ * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE,
+ * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR
+ * OTHER DEALINGS IN THE SOFTWARE.
+ *
+ * Authors: Ben Skeggs
+ */
+
+#define GT215 0xa3
+#define GF100 0xc0
+#define GF119 0xd9
+#define GK208 0x108
+
+#include "os.h"
+
+// IO addresses
+#define NV_PPWR_INTR_TRIGGER                                             0x0000
+#define NV_PPWR_INTR_TRIGGER_USER1                                   0x00000080
+#define NV_PPWR_INTR_TRIGGER_USER0                                   0x00000040
+#define NV_PPWR_INTR_ACK                                                 0x0004
+#define NV_PPWR_INTR_ACK_SUBINTR                                     0x00000800
+#define NV_PPWR_INTR_ACK_WATCHDOG                                    0x00000002
+#define NV_PPWR_INTR                                                     0x0008
+#define NV_PPWR_INTR_SUBINTR                                         0x00000800
+#define NV_PPWR_INTR_USER1                                           0x00000080
+#define NV_PPWR_INTR_USER0                                           0x00000040
+#define NV_PPWR_INTR_PAUSE                                           0x00000020
+#define NV_PPWR_INTR_WATCHDOG                                        0x00000002
+#define NV_PPWR_INTR_EN_SET                                              0x0010
+#define NV_PPWR_INTR_EN_SET_SUBINTR                                  0x00000800
+#define NV_PPWR_INTR_EN_SET_WATCHDOG                                 0x00000002
+#define NV_PPWR_INTR_EN_CLR                                              0x0014
+#define NV_PPWR_INTR_EN_CLR_MASK                    /* fuck i hate envyas */ -1
+#define NV_PPWR_INTR_ROUTE                                               0x001c
+#define NV_PPWR_TIMER_LOW                                                0x002c
+#define NV_PPWR_WATCHDOG_TIME                                            0x0034
+#define NV_PPWR_WATCHDOG_ENABLE                                          0x0038
+#define NV_PPWR_CAPS                                                     0x0108
+#define NV_PPWR_UAS_CONFIG                                               0x0164
+#define NV_PPWR_UAS_CONFIG_ENABLE                                    0x00010000
+#if NVKM_PPWR_CHIPSET >= GK208
+#define NV_PPWR_DSCRATCH(i)                                   (4 * (i) + 0x0450)
+#endif
+#define NV_PPWR_FIFO_PUT(i)                                   (4 * (i) + 0x04a0)
+#define NV_PPWR_FIFO_GET(i)                                   (4 * (i) + 0x04b0)
+#define NV_PPWR_FIFO_INTR                                                0x04c0
+#define NV_PPWR_FIFO_INTR_EN                                             0x04c4
+#define NV_PPWR_RFIFO_PUT                                                0x04c8
+#define NV_PPWR_RFIFO_GET                                                0x04cc
+#define NV_PPWR_H2D                                                      0x04d0
+#define NV_PPWR_D2H                                                      0x04dc
+#if NVKM_PPWR_CHIPSET < GK208
+#define NV_PPWR_DSCRATCH(i)                                   (4 * (i) + 0x05d0)
+#endif
+#define NV_PPWR_SUBINTR                                                  0x0688
+#define NV_PPWR_SUBINTR_FIFO                                         0x00000002
+#define NV_PPWR_MMIO_ADDR                                                0x07a0
+#define NV_PPWR_MMIO_DATA                                                0x07a4
+#define NV_PPWR_MMIO_CTRL                                                0x07ac
+#define NV_PPWR_MMIO_CTRL_TRIGGER                                    0x00010000
+#define NV_PPWR_MMIO_CTRL_STATUS                                     0x00007000
+#define NV_PPWR_MMIO_CTRL_STATUS_IDLE                                0x00000000
+#define NV_PPWR_MMIO_CTRL_MASK                                       0x000000f0
+#define NV_PPWR_MMIO_CTRL_MASK_B32_0                                 0x000000f0
+#define NV_PPWR_MMIO_CTRL_OP                                         0x00000003
+#define NV_PPWR_MMIO_CTRL_OP_RD                                      0x00000001
+#define NV_PPWR_MMIO_CTRL_OP_WR                                      0x00000002
+#define NV_PPWR_OUTPUT                                                   0x07c0
+#define NV_PPWR_OUTPUT_FB_PAUSE                                      0x00000004
+#define NV_PPWR_OUTPUT_SET                                               0x07e0
+#define NV_PPWR_OUTPUT_SET_FB_PAUSE                                  0x00000004
+#define NV_PPWR_OUTPUT_CLR                                               0x07e4
+#define NV_PPWR_OUTPUT_CLR_FB_PAUSE                                  0x00000004
+
+// Inter-process message format
+.equ #msg_process 0x00 /* send() target, recv() sender */
+.equ #msg_message 0x04
+.equ #msg_data0   0x08
+.equ #msg_data1   0x0c
+
+// Kernel message IDs
+#define KMSG_FIFO  0x00000000
+#define KMSG_ALARM 0x00000001
+
+// Process message queue description
+.equ #proc_qlen 4 // log2(size of queue entry in bytes)
+.equ #proc_qnum 2 // log2(max number of entries in queue)
+.equ #proc_qmaskb (1 << #proc_qnum) // max number of entries in queue
+.equ #proc_qmaskp (#proc_qmaskb - 1)
+.equ #proc_qmaskf ((#proc_qmaskb << 1) - 1)
+.equ #proc_qsize  (1 << (#proc_qlen + #proc_qnum))
+
+// Process table entry
+.equ #proc_id    0x00
+.equ #proc_init  0x04
+.equ #proc_recv  0x08
+.equ #proc_time  0x0c
+.equ #proc_qput  0x10
+.equ #proc_qget  0x14
+.equ #proc_queue 0x18
+.equ #proc_size (0x18 + #proc_qsize)
+
+#define process(id,init,recv) /*
+*/	.b32 id /*
+*/	.b32 init /*
+*/	.b32 recv /*
+*/	.b32 0 /*
+*/	.b32 0 /*
+*/	.b32 0 /*
+*/	.skip 64
+
+#ifndef NVKM_FALCON_UNSHIFTED_IO
+#define nv_iord(reg,ior) /*
+*/	mov reg ior /*
+*/ 	shl b32 reg 6 /*
+*/ 	iord reg I[reg + 0x000]
+#else
+#define nv_iord(reg,ior) /*
+*/	mov reg ior /*
+*/ 	iord reg I[reg + 0x000]
+#endif
+
+#ifndef NVKM_FALCON_UNSHIFTED_IO
+#define nv_iowr(ior,reg) /*
+*/	mov $r0 ior /*
+*/ 	shl b32 $r0 6 /*
+*/ 	iowr I[$r0 + 0x000] reg /*
+*/	clear b32 $r0
+#else
+#define nv_iowr(ior,reg) /*
+*/	mov $r0 ior /*
+*/ 	iowr I[$r0 + 0x000] reg /*
+*/	clear b32 $r0
+#endif
+
+#ifndef NVKM_FALCON_UNSHIFTED_IO
+#define nv_iowrs(ior,reg) /*
+*/	mov $r0 ior /*
+*/ 	shl b32 $r0 6 /*
+*/ 	iowrs I[$r0 + 0x000] reg /*
+*/	clear b32 $r0
+#else
+#define nv_iowrs(ior,reg) /*
+*/	mov $r0 ior /*
+*/ 	iowrs I[$r0 + 0x000] reg /*
+*/	clear b32 $r0
+#endif
+
+#define hash #
+#define fn(a) a
+#ifndef NVKM_FALCON_PC24
+#define call(a) call fn(hash)a
+#else
+#define call(a) lcall fn(hash)a
+#endif
+
+#ifndef NVKM_FALCON_MMIO_UAS
+#define nv_rd32(reg,addr) /*
+*/	mov b32 $r14 addr /*
+*/	call(rd32) /*
+*/	mov b32 reg $r13
+#else
+#define nv_rd32(reg,addr) /*
+*/ 	sethi $r0 0x14000000 /*
+*/	or $r0 addr /*
+*/	ld b32 reg D[$r0] /*
+*/	clear b32 $r0
+#endif
+
+#if !defined(NVKM_FALCON_MMIO_UAS) || defined(NVKM_FALCON_MMIO_TRAP)
+#define nv_wr32(addr,reg) /*
+*/	push addr /*
+*/	push reg /*
+*/	pop $r13 /*
+*/	pop $r14 /*
+*/	call(wr32) /*
+#else
+#define nv_wr32(addr,reg) /*
+*/ 	sethi $r0 0x14000000 /*
+*/	or $r0 addr /*
+*/	st b32 D[$r0] reg /*
+*/	clear b32 $r0
+#endif
diff --git a/drivers/gpu/drm/nouveau/core/subdev/pwr/fuc/memx.fuc b/drivers/gpu/drm/nouveau/core/subdev/pwr/fuc/memx.fuc
new file mode 100644
index 000000000000..d43741eccb11
--- /dev/null
+++ b/drivers/gpu/drm/nouveau/core/subdev/pwr/fuc/memx.fuc
@@ -0,0 +1,219 @@
+/*
+ * Copyright 2013 Red Hat Inc.
+ *
+ * Permission is hereby granted, free of charge, to any person obtaining a
+ * copy of this software and associated documentation files (the "Software"),
+ * to deal in the Software without restriction, including without limitation
+ * the rights to use, copy, modify, merge, publish, distribute, sublicense,
+ * and/or sell copies of the Software, and to permit persons to whom the
+ * Software is furnished to do so, subject to the following conditions:
+ *
+ * The above copyright notice and this permission notice shall be included in
+ * all copies or substantial portions of the Software.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+ * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+ * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT.  IN NO EVENT SHALL
+ * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR
+ * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE,
+ * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR
+ * OTHER DEALINGS IN THE SOFTWARE.
+ *
+ * Authors: Ben Skeggs
+ */
+
+#ifdef INCLUDE_PROC
+process(PROC_MEMX, #memx_init, #memx_recv)
+#endif
+
+/******************************************************************************
+ * MEMX data segment
+ *****************************************************************************/
+#ifdef INCLUDE_DATA
+.equ #memx_opcode 0
+.equ #memx_header 2
+.equ #memx_length 4
+.equ #memx_func   8
+
+#define handler(cmd,hdr,len,func) /*
+*/	.b16 MEMX_##cmd /*
+*/	.b16 hdr /*
+*/	.b16 len /*
+*/      .b16 0 /*
+*/	.b32 func
+
+memx_func_head:
+handler(ENTER , 0x0001, 0x0000, #memx_func_enter)
+memx_func_next:
+handler(LEAVE , 0x0000, 0x0000, #memx_func_leave)
+handler(WR32  , 0x0000, 0x0002, #memx_func_wr32)
+handler(WAIT  , 0x0004, 0x0000, #memx_func_wait)
+handler(DELAY , 0x0001, 0x0000, #memx_func_delay)
+memx_func_tail:
+
+.equ #memx_func_size #memx_func_next - #memx_func_head
+.equ #memx_func_num (#memx_func_tail - #memx_func_head) / #memx_func_size
+
+memx_data_head:
+.skip 0x0800
+memx_data_tail:
+#endif
+
+/******************************************************************************
+ * MEMX code segment
+ *****************************************************************************/
+#ifdef INCLUDE_CODE
+// description
+//
+// $r15 - current (memx)
+// $r4  - packet length
+//	+00: bitmask of heads to wait for vblank on
+// $r3  - opcode desciption
+// $r0  - zero
+memx_func_enter:
+	mov $r6 NV_PPWR_OUTPUT_SET_FB_PAUSE
+	nv_iowr(NV_PPWR_OUTPUT_SET, $r6)
+	memx_func_enter_wait:
+		nv_iord($r6, NV_PPWR_OUTPUT)
+		and $r6 NV_PPWR_OUTPUT_FB_PAUSE
+		bra z #memx_func_enter_wait
+	//XXX: TODO
+	ld b32 $r6 D[$r1 + 0x00]
+	add b32 $r1 0x04
+	ret
+
+// description
+//
+// $r15 - current (memx)
+// $r4  - packet length
+// $r3  - opcode desciption
+// $r0  - zero
+memx_func_leave:
+	mov $r6 NV_PPWR_OUTPUT_CLR_FB_PAUSE
+	nv_iowr(NV_PPWR_OUTPUT_CLR, $r6)
+	memx_func_leave_wait:
+		nv_iord($r6, NV_PPWR_OUTPUT)
+		and $r6 NV_PPWR_OUTPUT_FB_PAUSE
+		bra nz #memx_func_leave_wait
+	ret
+
+// description
+//
+// $r15 - current (memx)
+// $r4  - packet length
+//	+00*n: addr
+//	+04*n: data
+// $r3  - opcode desciption
+// $r0  - zero
+memx_func_wr32:
+	ld b32 $r6 D[$r1 + 0x00]
+	ld b32 $r5 D[$r1 + 0x04]
+	add b32 $r1 0x08
+	nv_wr32($r6, $r5)
+	sub b32 $r4 0x02
+	bra nz #memx_func_wr32
+	ret
+
+// description
+//
+// $r15 - current (memx)
+// $r4  - packet length
+//	+00: addr
+//	+04: mask
+//	+08: data
+//	+0c: timeout (ns)
+// $r3  - opcode desciption
+// $r0  - zero
+memx_func_wait:
+	nv_iord($r8, NV_PPWR_TIMER_LOW)
+	ld b32 $r14 D[$r1 + 0x00]
+	ld b32 $r13 D[$r1 + 0x04]
+	ld b32 $r12 D[$r1 + 0x08]
+	ld b32 $r11 D[$r1 + 0x0c]
+	add b32 $r1 0x10
+	call(wait)
+	ret
+
+// description
+//
+// $r15 - current (memx)
+// $r4  - packet length
+//	+00: time (ns)
+// $r3  - opcode desciption
+// $r0  - zero
+memx_func_delay:
+	ld b32 $r14 D[$r1 + 0x00]
+	add b32 $r1 0x04
+	call(nsec)
+	ret
+
+// description
+//
+// $r15 - current (memx)
+// $r14 - sender process name
+// $r13 - message (exec)
+// $r12 - head of script
+// $r11 - tail of script
+// $r0  - zero
+memx_exec:
+	push $r14
+	push $r13
+	mov b32 $r1 $r12
+	mov b32 $r2 $r11
+	memx_exec_next:
+		// fetch the packet header, and locate opcode info
+		ld b32 $r3 D[$r1]
+		add b32 $r1 4
+		shr b32 $r4 $r3 16
+		mulu $r3 #memx_func_size
+
+		// execute the opcode handler
+		ld b32 $r5 D[$r3 + #memx_func_head + #memx_func]
+		call $r5
+
+		// keep going, if we haven't reached the end
+		cmp b32 $r1 $r2
+		bra l #memx_exec_next
+
+	// send completion reply
+	pop $r13
+	pop $r14
+	call(send)
+	ret
+
+// description
+//
+// $r15 - current (memx)
+// $r14 - sender process name
+// $r13 - message
+// $r12 - data0
+// $r11 - data1
+// $r0  - zero
+memx_info:
+	mov $r12 #memx_data_head
+	mov $r11 #memx_data_tail - #memx_data_head
+	call(send)
+	ret
+
+// description
+//
+// $r15 - current (memx)
+// $r14 - sender process name
+// $r13 - message
+// $r12 - data0
+// $r11 - data1
+// $r0  - zero
+memx_recv:
+	cmp b32 $r13 MEMX_MSG_EXEC
+	bra e #memx_exec
+	cmp b32 $r13 MEMX_MSG_INFO
+	bra e #memx_info
+	ret
+
+// description
+//
+// $r15 - current (memx)
+// $r0  - zero
+memx_init:
+	ret
+#endif
diff --git a/drivers/gpu/drm/nouveau/core/subdev/pwr/fuc/nv108.fuc b/drivers/gpu/drm/nouveau/core/subdev/pwr/fuc/nv108.fuc
new file mode 100644
index 000000000000..947be536daef
--- /dev/null
+++ b/drivers/gpu/drm/nouveau/core/subdev/pwr/fuc/nv108.fuc
@@ -0,0 +1,63 @@
+/*
+ * Copyright 2013 Red Hat Inc.
+ *
+ * Permission is hereby granted, free of charge, to any person obtaining a
+ * copy of this software and associated documentation files (the "Software"),
+ * to deal in the Software without restriction, including without limitation
+ * the rights to use, copy, modify, merge, publish, distribute, sublicense,
+ * and/or sell copies of the Software, and to permit persons to whom the
+ * Software is furnished to do so, subject to the following conditions:
+ *
+ * The above copyright notice and this permission notice shall be included in
+ * all copies or substantial portions of the Software.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+ * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+ * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT.  IN NO EVENT SHALL
+ * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR
+ * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE,
+ * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR
+ * OTHER DEALINGS IN THE SOFTWARE.
+ *
+ * Authors: Ben Skeggs
+ */
+
+#define NVKM_PPWR_CHIPSET GK208
+
+#define NVKM_FALCON_PC24
+#define NVKM_FALCON_UNSHIFTED_IO
+//#define NVKM_FALCON_MMIO_UAS
+//#define NVKM_FALCON_MMIO_TRAP
+
+#include "macros.fuc"
+
+.section #nv108_pwr_data
+#define INCLUDE_PROC
+#include "kernel.fuc"
+#include "host.fuc"
+#include "memx.fuc"
+#include "perf.fuc"
+#include "test.fuc"
+#include "idle.fuc"
+#undef INCLUDE_PROC
+
+#define INCLUDE_DATA
+#include "kernel.fuc"
+#include "host.fuc"
+#include "memx.fuc"
+#include "perf.fuc"
+#include "test.fuc"
+#include "idle.fuc"
+#undef INCLUDE_DATA
+.align 256
+
+.section #nv108_pwr_code
+#define INCLUDE_CODE
+#include "kernel.fuc"
+#include "host.fuc"
+#include "memx.fuc"
+#include "perf.fuc"
+#include "test.fuc"
+#include "idle.fuc"
+#undef INCLUDE_CODE
+.align 256
diff --git a/drivers/gpu/drm/nouveau/core/subdev/pwr/fuc/nv108.fuc.h b/drivers/gpu/drm/nouveau/core/subdev/pwr/fuc/nv108.fuc.h
new file mode 100644
index 000000000000..9342e2d7d3b7
--- /dev/null
+++ b/drivers/gpu/drm/nouveau/core/subdev/pwr/fuc/nv108.fuc.h
@@ -0,0 +1,1165 @@
+uint32_t nv108_pwr_data[] = {
+/* 0x0000: proc_kern */
+	0x52544e49,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+/* 0x0058: proc_list_head */
+	0x54534f48,
+	0x00000379,
+	0x0000032a,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x584d454d,
+	0x0000046f,
+	0x00000461,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x46524550,
+	0x00000473,
+	0x00000471,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x54534554,
+	0x00000494,
+	0x00000475,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x454c4449,
+	0x0000049f,
+	0x0000049d,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+/* 0x0210: proc_list_tail */
+/* 0x0210: time_prev */
+	0x00000000,
+/* 0x0214: time_next */
+	0x00000000,
+/* 0x0218: fifo_queue */
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+/* 0x0298: rfifo_queue */
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+/* 0x0318: memx_func_head */
+	0x00010000,
+	0x00000000,
+	0x000003a9,
+/* 0x0324: memx_func_next */
+	0x00000001,
+	0x00000000,
+	0x000003c7,
+	0x00000002,
+	0x00000002,
+	0x000003df,
+	0x00040003,
+	0x00000000,
+	0x00000407,
+	0x00010004,
+	0x00000000,
+	0x00000421,
+/* 0x0354: memx_func_tail */
+/* 0x0354: memx_data_head */
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+/* 0x0b54: memx_data_tail */
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+};
+
+uint32_t nv108_pwr_code[] = {
+	0x02910ef5,
+/* 0x0004: rd32 */
+	0xf607a040,
+	0x04bd000e,
+	0xe3f0010e,
+	0x07ac4001,
+	0xbd000ef6,
+/* 0x0019: rd32_wait */
+	0x07ac4e04,
+	0xf100eecf,
+	0xf47000e4,
+	0xa44df61b,
+	0x00ddcf07,
+/* 0x002e: wr32 */
+	0xa04000f8,
+	0x000ef607,
+	0xa44004bd,
+	0x000df607,
+	0x020e04bd,
+	0xf0f0e5f0,
+	0xac4001e3,
+	0x000ef607,
+/* 0x004e: wr32_wait */
+	0xac4e04bd,
+	0x00eecf07,
+	0x7000e4f1,
+	0xf8f61bf4,
+/* 0x005d: nsec */
+	0xcf2c0800,
+/* 0x0062: nsec_loop */
+	0x2c090088,
+	0xbb0099cf,
+	0x9ea60298,
+	0xf8f61ef4,
+/* 0x0071: wait */
+	0xcf2c0800,
+/* 0x0076: wait_loop */
+	0xeeb20088,
+	0x0000047e,
+	0xadfddab2,
+	0xf4aca604,
+	0x2c09100b,
+	0xbb0099cf,
+	0x9ba60298,
+/* 0x0093: wait_done */
+	0xf8e61ef4,
+/* 0x0095: intr_watchdog */
+	0x03e99800,
+	0xf40096b0,
+	0x0a98280b,
+	0x029abb84,
+	0x0d0e1cf4,
+	0x01de7e01,
+	0xf494bd00,
+/* 0x00b2: intr_watchdog_next_time */
+	0x0a98140e,
+	0x00a6b085,
+	0xa6080bf4,
+	0x061cf49a,
+/* 0x00c0: intr_watchdog_next_time_set */
+/* 0x00c3: intr_watchdog_next_proc */
+	0xb58509b5,
+	0xe0b603e9,
+	0x10e6b158,
+	0xc81bf402,
+/* 0x00d2: intr */
+	0x00f900f8,
+	0x80f904bd,
+	0xa0f990f9,
+	0xc0f9b0f9,
+	0xe0f9d0f9,
+	0x000ff0f9,
+	0xf90188fe,
+	0x04504880,
+	0xb60088cf,
+	0x50400180,
+	0x0008f604,
+	0x080804bd,
+	0xc40088cf,
+	0x0bf40289,
+	0x8500b51f,
+	0x957e580e,
+	0x09980000,
+	0x0096b085,
+	0x000d0bf4,
+	0x0009f634,
+	0x09b504bd,
+/* 0x0125: intr_skip_watchdog */
+	0x0089e484,
+	0x360bf408,
+	0xcf068849,
+	0x9ac40099,
+	0x220bf402,
+	0xcf04c04c,
+	0xc0f900cc,
+	0xf14f484e,
+	0x0d5453e3,
+	0x023f7e00,
+	0x40c0fc00,
+	0x0cf604c0,
+/* 0x0157: intr_subintr_skip_fifo */
+	0x4004bd00,
+	0x09f60688,
+/* 0x015f: intr_skip_subintr */
+	0xc404bd00,
+	0x0bf42089,
+	0xbfa4f107,
+/* 0x0169: intr_skip_pause */
+	0x4089c4ff,
+	0xf1070bf4,
+/* 0x0173: intr_skip_user0 */
+	0x00ffbfa4,
+	0x0008f604,
+	0x80fc04bd,
+	0xfc0088fe,
+	0xfce0fcf0,
+	0xfcc0fcd0,
+	0xfca0fcb0,
+	0xfc80fc90,
+	0x0032f400,
+/* 0x0196: timer */
+	0x32f401f8,
+	0x03f89810,
+	0xf40086b0,
+	0xfeb53a1c,
+	0xf6380003,
+	0x04bd0008,
+	0x88cf0808,
+	0x0284f000,
+	0x081c1bf4,
+	0x0088cf34,
+	0x0bf4e0a6,
+	0xf4e8a608,
+/* 0x01c6: timer_reset */
+	0x3400161e,
+	0xbd000ef6,
+	0x840eb504,
+/* 0x01d0: timer_enable */
+	0x38000108,
+	0xbd0008f6,
+/* 0x01d9: timer_done */
+	0x1031f404,
+/* 0x01de: send_proc */
+	0x80f900f8,
+	0xe89890f9,
+	0x04e99805,
+	0xa60486f0,
+	0x2a0bf489,
+	0x940398c4,
+	0x80b60488,
+	0x008ebb18,
+	0xb500fa98,
+	0x8db5008a,
+	0x028cb501,
+	0xb6038bb5,
+	0x94f00190,
+	0x04e9b507,
+/* 0x0217: send_done */
+	0xfc0231f4,
+	0xf880fc90,
+/* 0x021d: find */
+	0x0880f900,
+	0x0131f458,
+/* 0x0224: find_loop */
+	0xa6008a98,
+	0x100bf4ae,
+	0xb15880b6,
+	0xf4021086,
+	0x32f4f11b,
+/* 0x0239: find_done */
+	0xfc8eb201,
+/* 0x023f: send */
+	0x7e00f880,
+	0xf400021d,
+	0x00f89b01,
+/* 0x0248: recv */
+	0x9805e898,
+	0x32f404e9,
+	0xf489a601,
+	0x89c43c0b,
+	0x0180b603,
+	0xb50784f0,
+	0xea9805e8,
+	0xfef0f902,
+	0xf0f9018f,
+	0x9994efb2,
+	0x00e9bb04,
+	0x9818e0b6,
+	0xec9803eb,
+	0x01ed9802,
+	0xf900ee98,
+	0xfef0fca5,
+	0x31f400f8,
+/* 0x028f: recv_done */
+	0xf8f0fc01,
+/* 0x0291: init */
+	0x01084100,
+	0xe70011cf,
+	0xb6010911,
+	0x14fe0814,
+	0x00e04100,
+	0x000013f0,
+	0x0001f61c,
+	0xff0104bd,
+	0x01f61400,
+	0x0104bd00,
+	0x0015f102,
+	0xf6100008,
+	0x04bd0001,
+	0xf000d241,
+	0x10fe0013,
+	0x1031f400,
+	0x38000101,
+	0xbd0001f6,
+/* 0x02db: init_proc */
+	0x98580f04,
+	0x16b001f1,
+	0xfa0bf400,
+	0xf0b615f9,
+	0xf20ef458,
+/* 0x02ec: host_send */
+	0xcf04b041,
+	0xa0420011,
+	0x0022cf04,
+	0x0bf412a6,
+	0x071ec42e,
+	0xb704ee94,
+	0x980218e0,
+	0xec9803eb,
+	0x01ed9802,
+	0x7e00ee98,
+	0xb600023f,
+	0x1ec40110,
+	0x04b0400f,
+	0xbd0001f6,
+	0xc70ef404,
+/* 0x0328: host_send_done */
+/* 0x032a: host_recv */
+	0x494100f8,
+	0x5413f14e,
+	0xf4e1a652,
+/* 0x0336: host_recv_wait */
+	0xcc41b90b,
+	0x0011cf04,
+	0xcf04c842,
+	0x16f00022,
+	0xf412a608,
+	0x23c4ef0b,
+	0x0434b607,
+	0x029830b7,
+	0xb5033bb5,
+	0x3db5023c,
+	0x003eb501,
+	0xf00120b6,
+	0xc8400f24,
+	0x0002f604,
+	0x400204bd,
+	0x02f60000,
+	0xf804bd00,
+/* 0x0379: host_init */
+	0x00804100,
+	0xf11014b6,
+	0x40021815,
+	0x01f604d0,
+	0x4104bd00,
+	0x14b60080,
+	0x9815f110,
+	0x04dc4002,
+	0xbd0001f6,
+	0x40010104,
+	0x01f604c4,
+	0xf804bd00,
+/* 0x03a9: memx_func_enter */
+	0x40040600,
+	0x06f607e0,
+/* 0x03b3: memx_func_enter_wait */
+	0x4604bd00,
+	0x66cf07c0,
+	0x0464f000,
+	0x98f70bf4,
+	0x10b60016,
+/* 0x03c7: memx_func_leave */
+	0x0600f804,
+	0x07e44004,
+	0xbd0006f6,
+/* 0x03d1: memx_func_leave_wait */
+	0x07c04604,
+	0xf00066cf,
+	0x1bf40464,
+/* 0x03df: memx_func_wr32 */
+	0x9800f8f7,
+	0x15980016,
+	0x0810b601,
+	0x50f960f9,
+	0xe0fcd0fc,
+	0x00002e7e,
+	0x140003f1,
+	0xa00506fd,
+	0xb604bd05,
+	0x1bf40242,
+/* 0x0407: memx_func_wait */
+	0x0800f8dd,
+	0x0088cf2c,
+	0x98001e98,
+	0x1c98011d,
+	0x031b9802,
+	0x7e1010b6,
+	0xf8000071,
+/* 0x0421: memx_func_delay */
+	0x001e9800,
+	0x7e0410b6,
+	0xf800005d,
+/* 0x042d: memx_exec */
+	0xf9e0f900,
+	0xb2c1b2d0,
+/* 0x0435: memx_exec_next */
+	0x001398b2,
+	0x950410b6,
+	0x30f01034,
+	0xc835980c,
+	0x12a655f9,
+	0xfced1ef4,
+	0x7ee0fcd0,
+	0xf800023f,
+/* 0x0455: memx_info */
+	0x03544c00,
+	0x7e08004b,
+	0xf800023f,
+/* 0x0461: memx_recv */
+	0x01d6b000,
+	0xb0c90bf4,
+	0x0bf400d6,
+/* 0x046f: memx_init */
+	0xf800f8eb,
+/* 0x0471: perf_recv */
+/* 0x0473: perf_init */
+	0xf800f800,
+/* 0x0475: test_recv */
+	0x04584100,
+	0xb60011cf,
+	0x58400110,
+	0x0001f604,
+	0xe7f104bd,
+	0xe3f1d900,
+	0x967e134f,
+	0x00f80001,
+/* 0x0494: test_init */
+	0x7e08004e,
+	0xf8000196,
+/* 0x049d: idle_recv */
+/* 0x049f: idle */
+	0xf400f800,
+	0x54410031,
+	0x0011cf04,
+	0x400110b6,
+	0x01f60454,
+/* 0x04b3: idle_loop */
+	0x0104bd00,
+	0x0232f458,
+/* 0x04b8: idle_proc */
+/* 0x04b8: idle_proc_exec */
+	0x1eb210f9,
+	0x0002487e,
+	0x11f410fc,
+	0x0231f409,
+/* 0x04cb: idle_proc_next */
+	0xb6f00ef4,
+	0x1fa65810,
+	0xf4e81bf4,
+	0x28f4e002,
+	0xc60ef400,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+};
diff --git a/drivers/gpu/drm/nouveau/core/subdev/pwr/fuc/nva3.fuc b/drivers/gpu/drm/nouveau/core/subdev/pwr/fuc/nva3.fuc
new file mode 100644
index 000000000000..6fde0b89e5aa
--- /dev/null
+++ b/drivers/gpu/drm/nouveau/core/subdev/pwr/fuc/nva3.fuc
@@ -0,0 +1,63 @@
+/*
+ * Copyright 2013 Red Hat Inc.
+ *
+ * Permission is hereby granted, free of charge, to any person obtaining a
+ * copy of this software and associated documentation files (the "Software"),
+ * to deal in the Software without restriction, including without limitation
+ * the rights to use, copy, modify, merge, publish, distribute, sublicense,
+ * and/or sell copies of the Software, and to permit persons to whom the
+ * Software is furnished to do so, subject to the following conditions:
+ *
+ * The above copyright notice and this permission notice shall be included in
+ * all copies or substantial portions of the Software.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+ * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+ * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT.  IN NO EVENT SHALL
+ * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR
+ * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE,
+ * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR
+ * OTHER DEALINGS IN THE SOFTWARE.
+ *
+ * Authors: Ben Skeggs
+ */
+
+#define NVKM_PPWR_CHIPSET GT215
+
+//#define NVKM_FALCON_PC24
+//#define NVKM_FALCON_UNSHIFTED_IO
+//#define NVKM_FALCON_MMIO_UAS
+//#define NVKM_FALCON_MMIO_TRAP
+
+#include "macros.fuc"
+
+.section #nva3_pwr_data
+#define INCLUDE_PROC
+#include "kernel.fuc"
+#include "host.fuc"
+#include "memx.fuc"
+#include "perf.fuc"
+#include "test.fuc"
+#include "idle.fuc"
+#undef INCLUDE_PROC
+
+#define INCLUDE_DATA
+#include "kernel.fuc"
+#include "host.fuc"
+#include "memx.fuc"
+#include "perf.fuc"
+#include "test.fuc"
+#include "idle.fuc"
+#undef INCLUDE_DATA
+.align 256
+
+.section #nva3_pwr_code
+#define INCLUDE_CODE
+#include "kernel.fuc"
+#include "host.fuc"
+#include "memx.fuc"
+#include "perf.fuc"
+#include "test.fuc"
+#include "idle.fuc"
+#undef INCLUDE_CODE
+.align 256
diff --git a/drivers/gpu/drm/nouveau/core/subdev/pwr/fuc/nva3.fuc.h b/drivers/gpu/drm/nouveau/core/subdev/pwr/fuc/nva3.fuc.h
new file mode 100644
index 000000000000..0fa4d7dcd407
--- /dev/null
+++ b/drivers/gpu/drm/nouveau/core/subdev/pwr/fuc/nva3.fuc.h
@@ -0,0 +1,1229 @@
+uint32_t nva3_pwr_data[] = {
+/* 0x0000: proc_kern */
+	0x52544e49,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+/* 0x0058: proc_list_head */
+	0x54534f48,
+	0x00000430,
+	0x000003cd,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x584d454d,
+	0x0000054e,
+	0x00000540,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x46524550,
+	0x00000552,
+	0x00000550,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x54534554,
+	0x0000057b,
+	0x00000554,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x454c4449,
+	0x00000587,
+	0x00000585,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+/* 0x0210: proc_list_tail */
+/* 0x0210: time_prev */
+	0x00000000,
+/* 0x0214: time_next */
+	0x00000000,
+/* 0x0218: fifo_queue */
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+/* 0x0298: rfifo_queue */
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+/* 0x0318: memx_func_head */
+	0x00010000,
+	0x00000000,
+	0x0000046f,
+/* 0x0324: memx_func_next */
+	0x00000001,
+	0x00000000,
+	0x00000496,
+	0x00000002,
+	0x00000002,
+	0x000004b7,
+	0x00040003,
+	0x00000000,
+	0x000004df,
+	0x00010004,
+	0x00000000,
+	0x000004fc,
+/* 0x0354: memx_func_tail */
+/* 0x0354: memx_data_head */
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+/* 0x0b54: memx_data_tail */
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+};
+
+uint32_t nva3_pwr_code[] = {
+	0x030d0ef5,
+/* 0x0004: rd32 */
+	0x07a007f1,
+	0xd00604b6,
+	0x04bd000e,
+	0xf001e7f0,
+	0x07f101e3,
+	0x04b607ac,
+	0x000ed006,
+/* 0x0022: rd32_wait */
+	0xe7f104bd,
+	0xe4b607ac,
+	0x00eecf06,
+	0x7000e4f1,
+	0xf1f21bf4,
+	0xb607a4d7,
+	0xddcf06d4,
+/* 0x003f: wr32 */
+	0xf100f800,
+	0xb607a007,
+	0x0ed00604,
+	0xf104bd00,
+	0xb607a407,
+	0x0dd00604,
+	0xf004bd00,
+	0xe5f002e7,
+	0x01e3f0f0,
+	0x07ac07f1,
+	0xd00604b6,
+	0x04bd000e,
+/* 0x006c: wr32_wait */
+	0x07ace7f1,
+	0xcf06e4b6,
+	0xe4f100ee,
+	0x1bf47000,
+/* 0x007f: nsec */
+	0xf000f8f2,
+	0x84b62c87,
+	0x0088cf06,
+/* 0x0088: nsec_loop */
+	0xb62c97f0,
+	0x99cf0694,
+	0x0298bb00,
+	0xf4069eb8,
+	0x00f8f11e,
+/* 0x009c: wait */
+	0xb62c87f0,
+	0x88cf0684,
+/* 0x00a5: wait_loop */
+	0x02eeb900,
+	0xb90421f4,
+	0xadfd02da,
+	0x06acb804,
+	0xf0150bf4,
+	0x94b62c97,
+	0x0099cf06,
+	0xb80298bb,
+	0x1ef4069b,
+/* 0x00c9: wait_done */
+/* 0x00cb: intr_watchdog */
+	0x9800f8df,
+	0x96b003e9,
+	0x2a0bf400,
+	0xbb840a98,
+	0x1cf4029a,
+	0x01d7f00f,
+	0x025421f5,
+	0x0ef494bd,
+/* 0x00e9: intr_watchdog_next_time */
+	0x850a9815,
+	0xf400a6b0,
+	0x9ab8090b,
+	0x061cf406,
+/* 0x00f8: intr_watchdog_next_time_set */
+/* 0x00fb: intr_watchdog_next_proc */
+	0x80850980,
+	0xe0b603e9,
+	0x10e6b158,
+	0xc61bf402,
+/* 0x010a: intr */
+	0x00f900f8,
+	0x80f904bd,
+	0xa0f990f9,
+	0xc0f9b0f9,
+	0xe0f9d0f9,
+	0xf7f0f0f9,
+	0x0188fe00,
+	0x87f180f9,
+	0x84b605d0,
+	0x0088cf06,
+	0xf10180b6,
+	0xb605d007,
+	0x08d00604,
+	0xf004bd00,
+	0x84b60887,
+	0x0088cf06,
+	0xf40289c4,
+	0x0080230b,
+	0x58e7f085,
+	0x98cb21f4,
+	0x96b08509,
+	0x110bf400,
+	0xb63407f0,
+	0x09d00604,
+	0x8004bd00,
+/* 0x016e: intr_skip_watchdog */
+	0x89e48409,
+	0x0bf40800,
+	0x8897f148,
+	0x0694b606,
+	0xc40099cf,
+	0x0bf4029a,
+	0xc0c7f12c,
+	0x06c4b604,
+	0xf900cccf,
+	0x48e7f1c0,
+	0x53e3f14f,
+	0x00d7f054,
+	0x02b921f5,
+	0x07f1c0fc,
+	0x04b604c0,
+	0x000cd006,
+/* 0x01ae: intr_subintr_skip_fifo */
+	0x07f104bd,
+	0x04b60688,
+	0x0009d006,
+/* 0x01ba: intr_skip_subintr */
+	0x89c404bd,
+	0x070bf420,
+	0xffbfa4f1,
+/* 0x01c4: intr_skip_pause */
+	0xf44089c4,
+	0xa4f1070b,
+/* 0x01ce: intr_skip_user0 */
+	0x07f0ffbf,
+	0x0604b604,
+	0xbd0008d0,
+	0xfe80fc04,
+	0xf0fc0088,
+	0xd0fce0fc,
+	0xb0fcc0fc,
+	0x90fca0fc,
+	0x00fc80fc,
+	0xf80032f4,
+/* 0x01f5: timer */
+	0x1032f401,
+	0xb003f898,
+	0x1cf40086,
+	0x03fe8051,
+	0xb63807f0,
+	0x08d00604,
+	0xf004bd00,
+	0x84b60887,
+	0x0088cf06,
+	0xf40284f0,
+	0x87f0261b,
+	0x0684b634,
+	0xb80088cf,
+	0x0bf406e0,
+	0x06e8b809,
+/* 0x0233: timer_reset */
+	0xf01f1ef4,
+	0x04b63407,
+	0x000ed006,
+	0x0e8004bd,
+/* 0x0241: timer_enable */
+	0x0187f084,
+	0xb63807f0,
+	0x08d00604,
+/* 0x024f: timer_done */
+	0xf404bd00,
+	0x00f81031,
+/* 0x0254: send_proc */
+	0x90f980f9,
+	0x9805e898,
+	0x86f004e9,
+	0x0689b804,
+	0xc42a0bf4,
+	0x88940398,
+	0x1880b604,
+	0x98008ebb,
+	0x8a8000fa,
+	0x018d8000,
+	0x80028c80,
+	0x90b6038b,
+	0x0794f001,
+	0xf404e980,
+/* 0x028e: send_done */
+	0x90fc0231,
+	0x00f880fc,
+/* 0x0294: find */
+	0x87f080f9,
+	0x0131f458,
+/* 0x029c: find_loop */
+	0xb8008a98,
+	0x0bf406ae,
+	0x5880b610,
+	0x021086b1,
+	0xf4f01bf4,
+/* 0x02b2: find_done */
+	0x8eb90132,
+	0xf880fc02,
+/* 0x02b9: send */
+	0x9421f500,
+	0x9701f402,
+/* 0x02c2: recv */
+	0xe89800f8,
+	0x04e99805,
+	0xb80132f4,
+	0x0bf40689,
+	0x0389c43d,
+	0xf00180b6,
+	0xe8800784,
+	0x02ea9805,
+	0x8ffef0f9,
+	0xb9f0f901,
+	0x999402ef,
+	0x00e9bb04,
+	0x9818e0b6,
+	0xec9803eb,
+	0x01ed9802,
+	0xf900ee98,
+	0xfef0fca5,
+	0x31f400f8,
+/* 0x030b: recv_done */
+	0xf8f0fc01,
+/* 0x030d: init */
+	0x0817f100,
+	0x0614b601,
+	0xe70011cf,
+	0xb6010911,
+	0x14fe0814,
+	0xe017f100,
+	0x0013f000,
+	0xb61c07f0,
+	0x01d00604,
+	0xf004bd00,
+	0x07f0ff17,
+	0x0604b614,
+	0xbd0001d0,
+	0x0217f004,
+	0x080015f1,
+	0xb61007f0,
+	0x01d00604,
+	0xf104bd00,
+	0xf0010a17,
+	0x10fe0013,
+	0x1031f400,
+	0xf00117f0,
+	0x04b63807,
+	0x0001d006,
+	0xf7f004bd,
+/* 0x0371: init_proc */
+	0x01f19858,
+	0xf40016b0,
+	0x15f9fa0b,
+	0xf458f0b6,
+/* 0x0382: host_send */
+	0x17f1f20e,
+	0x14b604b0,
+	0x0011cf06,
+	0x04a027f1,
+	0xcf0624b6,
+	0x12b80022,
+	0x320bf406,
+	0x94071ec4,
+	0xe0b704ee,
+	0xeb980218,
+	0x02ec9803,
+	0x9801ed98,
+	0x21f500ee,
+	0x10b602b9,
+	0x0f1ec401,
+	0x04b007f1,
+	0xd00604b6,
+	0x04bd0001,
+/* 0x03cb: host_send_done */
+	0xf8ba0ef4,
+/* 0x03cd: host_recv */
+	0x4917f100,
+	0x5413f14e,
+	0x06e1b852,
+/* 0x03db: host_recv_wait */
+	0xf1aa0bf4,
+	0xb604cc17,
+	0x11cf0614,
+	0xc827f100,
+	0x0624b604,
+	0xf00022cf,
+	0x12b80816,
+	0xe60bf406,
+	0xb60723c4,
+	0x30b70434,
+	0x3b800298,
+	0x023c8003,
+	0x80013d80,
+	0x20b6003e,
+	0x0f24f001,
+	0x04c807f1,
+	0xd00604b6,
+	0x04bd0002,
+	0xf04027f0,
+	0x04b60007,
+	0x0002d006,
+	0x00f804bd,
+/* 0x0430: host_init */
+	0x008017f1,
+	0xf11014b6,
+	0xf1021815,
+	0xb604d007,
+	0x01d00604,
+	0xf104bd00,
+	0xb6008017,
+	0x15f11014,
+	0x07f10298,
+	0x04b604dc,
+	0x0001d006,
+	0x17f004bd,
+	0xc407f101,
+	0x0604b604,
+	0xbd0001d0,
+/* 0x046f: memx_func_enter */
+	0xf000f804,
+	0x07f10467,
+	0x04b607e0,
+	0x0006d006,
+/* 0x047e: memx_func_enter_wait */
+	0x67f104bd,
+	0x64b607c0,
+	0x0066cf06,
+	0xf40464f0,
+	0x1698f30b,
+	0x0410b600,
+/* 0x0496: memx_func_leave */
+	0x67f000f8,
+	0xe407f104,
+	0x0604b607,
+	0xbd0006d0,
+/* 0x04a5: memx_func_leave_wait */
+	0xc067f104,
+	0x0664b607,
+	0xf00066cf,
+	0x1bf40464,
+/* 0x04b7: memx_func_wr32 */
+	0x9800f8f3,
+	0x15980016,
+	0x0810b601,
+	0x50f960f9,
+	0xe0fcd0fc,
+	0xf13f21f4,
+	0xfd140003,
+	0x05800506,
+	0xb604bd00,
+	0x1bf40242,
+/* 0x04df: memx_func_wait */
+	0xf000f8dd,
+	0x84b62c87,
+	0x0088cf06,
+	0x98001e98,
+	0x1c98011d,
+	0x031b9802,
+	0xf41010b6,
+	0x00f89c21,
+/* 0x04fc: memx_func_delay */
+	0xb6001e98,
+	0x21f40410,
+/* 0x0507: memx_exec */
+	0xf900f87f,
+	0xb9d0f9e0,
+	0xb2b902c1,
+/* 0x0511: memx_exec_next */
+	0x00139802,
+	0x950410b6,
+	0x30f01034,
+	0xc835980c,
+	0x12b855f9,
+	0xec1ef406,
+	0xe0fcd0fc,
+	0x02b921f5,
+/* 0x0532: memx_info */
+	0xc7f100f8,
+	0xb7f10354,
+	0x21f50800,
+	0x00f802b9,
+/* 0x0540: memx_recv */
+	0xf401d6b0,
+	0xd6b0c40b,
+	0xe90bf400,
+/* 0x054e: memx_init */
+	0x00f800f8,
+/* 0x0550: perf_recv */
+/* 0x0552: perf_init */
+	0x00f800f8,
+/* 0x0554: test_recv */
+	0x05d817f1,
+	0xcf0614b6,
+	0x10b60011,
+	0xd807f101,
+	0x0604b605,
+	0xbd0001d0,
+	0x00e7f104,
+	0x4fe3f1d9,
+	0xf521f513,
+/* 0x057b: test_init */
+	0xf100f801,
+	0xf50800e7,
+	0xf801f521,
+/* 0x0585: idle_recv */
+/* 0x0587: idle */
+	0xf400f800,
+	0x17f10031,
+	0x14b605d4,
+	0x0011cf06,
+	0xf10110b6,
+	0xb605d407,
+	0x01d00604,
+/* 0x05a3: idle_loop */
+	0xf004bd00,
+	0x32f45817,
+/* 0x05a9: idle_proc */
+/* 0x05a9: idle_proc_exec */
+	0xb910f902,
+	0x21f5021e,
+	0x10fc02c2,
+	0xf40911f4,
+	0x0ef40231,
+/* 0x05bd: idle_proc_next */
+	0x5810b6ef,
+	0xf4061fb8,
+	0x02f4e61b,
+	0x0028f4dd,
+	0x00bb0ef4,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+};
diff --git a/drivers/gpu/drm/nouveau/core/subdev/pwr/fuc/nvc0.fuc b/drivers/gpu/drm/nouveau/core/subdev/pwr/fuc/nvc0.fuc
new file mode 100644
index 000000000000..eaa64da68e36
--- /dev/null
+++ b/drivers/gpu/drm/nouveau/core/subdev/pwr/fuc/nvc0.fuc
@@ -0,0 +1,63 @@
+/*
+ * Copyright 2013 Red Hat Inc.
+ *
+ * Permission is hereby granted, free of charge, to any person obtaining a
+ * copy of this software and associated documentation files (the "Software"),
+ * to deal in the Software without restriction, including without limitation
+ * the rights to use, copy, modify, merge, publish, distribute, sublicense,
+ * and/or sell copies of the Software, and to permit persons to whom the
+ * Software is furnished to do so, subject to the following conditions:
+ *
+ * The above copyright notice and this permission notice shall be included in
+ * all copies or substantial portions of the Software.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+ * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+ * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT.  IN NO EVENT SHALL
+ * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR
+ * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE,
+ * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR
+ * OTHER DEALINGS IN THE SOFTWARE.
+ *
+ * Authors: Ben Skeggs
+ */
+
+#define NVKM_PPWR_CHIPSET GF100
+
+//#define NVKM_FALCON_PC24
+//#define NVKM_FALCON_UNSHIFTED_IO
+//#define NVKM_FALCON_MMIO_UAS
+//#define NVKM_FALCON_MMIO_TRAP
+
+#include "macros.fuc"
+
+.section #nvc0_pwr_data
+#define INCLUDE_PROC
+#include "kernel.fuc"
+#include "host.fuc"
+#include "memx.fuc"
+#include "perf.fuc"
+#include "test.fuc"
+#include "idle.fuc"
+#undef INCLUDE_PROC
+
+#define INCLUDE_DATA
+#include "kernel.fuc"
+#include "host.fuc"
+#include "memx.fuc"
+#include "perf.fuc"
+#include "test.fuc"
+#include "idle.fuc"
+#undef INCLUDE_DATA
+.align 256
+
+.section #nvc0_pwr_code
+#define INCLUDE_CODE
+#include "kernel.fuc"
+#include "host.fuc"
+#include "memx.fuc"
+#include "perf.fuc"
+#include "test.fuc"
+#include "idle.fuc"
+#undef INCLUDE_CODE
+.align 256
diff --git a/drivers/gpu/drm/nouveau/core/subdev/pwr/fuc/nvc0.fuc.h b/drivers/gpu/drm/nouveau/core/subdev/pwr/fuc/nvc0.fuc.h
new file mode 100644
index 000000000000..82c8e8b88917
--- /dev/null
+++ b/drivers/gpu/drm/nouveau/core/subdev/pwr/fuc/nvc0.fuc.h
@@ -0,0 +1,1229 @@
+uint32_t nvc0_pwr_data[] = {
+/* 0x0000: proc_kern */
+	0x52544e49,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+/* 0x0058: proc_list_head */
+	0x54534f48,
+	0x00000430,
+	0x000003cd,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x584d454d,
+	0x0000054e,
+	0x00000540,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x46524550,
+	0x00000552,
+	0x00000550,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x54534554,
+	0x0000057b,
+	0x00000554,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x454c4449,
+	0x00000587,
+	0x00000585,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+/* 0x0210: proc_list_tail */
+/* 0x0210: time_prev */
+	0x00000000,
+/* 0x0214: time_next */
+	0x00000000,
+/* 0x0218: fifo_queue */
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+/* 0x0298: rfifo_queue */
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+/* 0x0318: memx_func_head */
+	0x00010000,
+	0x00000000,
+	0x0000046f,
+/* 0x0324: memx_func_next */
+	0x00000001,
+	0x00000000,
+	0x00000496,
+	0x00000002,
+	0x00000002,
+	0x000004b7,
+	0x00040003,
+	0x00000000,
+	0x000004df,
+	0x00010004,
+	0x00000000,
+	0x000004fc,
+/* 0x0354: memx_func_tail */
+/* 0x0354: memx_data_head */
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+/* 0x0b54: memx_data_tail */
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+};
+
+uint32_t nvc0_pwr_code[] = {
+	0x030d0ef5,
+/* 0x0004: rd32 */
+	0x07a007f1,
+	0xd00604b6,
+	0x04bd000e,
+	0xf001e7f0,
+	0x07f101e3,
+	0x04b607ac,
+	0x000ed006,
+/* 0x0022: rd32_wait */
+	0xe7f104bd,
+	0xe4b607ac,
+	0x00eecf06,
+	0x7000e4f1,
+	0xf1f21bf4,
+	0xb607a4d7,
+	0xddcf06d4,
+/* 0x003f: wr32 */
+	0xf100f800,
+	0xb607a007,
+	0x0ed00604,
+	0xf104bd00,
+	0xb607a407,
+	0x0dd00604,
+	0xf004bd00,
+	0xe5f002e7,
+	0x01e3f0f0,
+	0x07ac07f1,
+	0xd00604b6,
+	0x04bd000e,
+/* 0x006c: wr32_wait */
+	0x07ace7f1,
+	0xcf06e4b6,
+	0xe4f100ee,
+	0x1bf47000,
+/* 0x007f: nsec */
+	0xf000f8f2,
+	0x84b62c87,
+	0x0088cf06,
+/* 0x0088: nsec_loop */
+	0xb62c97f0,
+	0x99cf0694,
+	0x0298bb00,
+	0xf4069eb8,
+	0x00f8f11e,
+/* 0x009c: wait */
+	0xb62c87f0,
+	0x88cf0684,
+/* 0x00a5: wait_loop */
+	0x02eeb900,
+	0xb90421f4,
+	0xadfd02da,
+	0x06acb804,
+	0xf0150bf4,
+	0x94b62c97,
+	0x0099cf06,
+	0xb80298bb,
+	0x1ef4069b,
+/* 0x00c9: wait_done */
+/* 0x00cb: intr_watchdog */
+	0x9800f8df,
+	0x96b003e9,
+	0x2a0bf400,
+	0xbb840a98,
+	0x1cf4029a,
+	0x01d7f00f,
+	0x025421f5,
+	0x0ef494bd,
+/* 0x00e9: intr_watchdog_next_time */
+	0x850a9815,
+	0xf400a6b0,
+	0x9ab8090b,
+	0x061cf406,
+/* 0x00f8: intr_watchdog_next_time_set */
+/* 0x00fb: intr_watchdog_next_proc */
+	0x80850980,
+	0xe0b603e9,
+	0x10e6b158,
+	0xc61bf402,
+/* 0x010a: intr */
+	0x00f900f8,
+	0x80f904bd,
+	0xa0f990f9,
+	0xc0f9b0f9,
+	0xe0f9d0f9,
+	0xf7f0f0f9,
+	0x0188fe00,
+	0x87f180f9,
+	0x84b605d0,
+	0x0088cf06,
+	0xf10180b6,
+	0xb605d007,
+	0x08d00604,
+	0xf004bd00,
+	0x84b60887,
+	0x0088cf06,
+	0xf40289c4,
+	0x0080230b,
+	0x58e7f085,
+	0x98cb21f4,
+	0x96b08509,
+	0x110bf400,
+	0xb63407f0,
+	0x09d00604,
+	0x8004bd00,
+/* 0x016e: intr_skip_watchdog */
+	0x89e48409,
+	0x0bf40800,
+	0x8897f148,
+	0x0694b606,
+	0xc40099cf,
+	0x0bf4029a,
+	0xc0c7f12c,
+	0x06c4b604,
+	0xf900cccf,
+	0x48e7f1c0,
+	0x53e3f14f,
+	0x00d7f054,
+	0x02b921f5,
+	0x07f1c0fc,
+	0x04b604c0,
+	0x000cd006,
+/* 0x01ae: intr_subintr_skip_fifo */
+	0x07f104bd,
+	0x04b60688,
+	0x0009d006,
+/* 0x01ba: intr_skip_subintr */
+	0x89c404bd,
+	0x070bf420,
+	0xffbfa4f1,
+/* 0x01c4: intr_skip_pause */
+	0xf44089c4,
+	0xa4f1070b,
+/* 0x01ce: intr_skip_user0 */
+	0x07f0ffbf,
+	0x0604b604,
+	0xbd0008d0,
+	0xfe80fc04,
+	0xf0fc0088,
+	0xd0fce0fc,
+	0xb0fcc0fc,
+	0x90fca0fc,
+	0x00fc80fc,
+	0xf80032f4,
+/* 0x01f5: timer */
+	0x1032f401,
+	0xb003f898,
+	0x1cf40086,
+	0x03fe8051,
+	0xb63807f0,
+	0x08d00604,
+	0xf004bd00,
+	0x84b60887,
+	0x0088cf06,
+	0xf40284f0,
+	0x87f0261b,
+	0x0684b634,
+	0xb80088cf,
+	0x0bf406e0,
+	0x06e8b809,
+/* 0x0233: timer_reset */
+	0xf01f1ef4,
+	0x04b63407,
+	0x000ed006,
+	0x0e8004bd,
+/* 0x0241: timer_enable */
+	0x0187f084,
+	0xb63807f0,
+	0x08d00604,
+/* 0x024f: timer_done */
+	0xf404bd00,
+	0x00f81031,
+/* 0x0254: send_proc */
+	0x90f980f9,
+	0x9805e898,
+	0x86f004e9,
+	0x0689b804,
+	0xc42a0bf4,
+	0x88940398,
+	0x1880b604,
+	0x98008ebb,
+	0x8a8000fa,
+	0x018d8000,
+	0x80028c80,
+	0x90b6038b,
+	0x0794f001,
+	0xf404e980,
+/* 0x028e: send_done */
+	0x90fc0231,
+	0x00f880fc,
+/* 0x0294: find */
+	0x87f080f9,
+	0x0131f458,
+/* 0x029c: find_loop */
+	0xb8008a98,
+	0x0bf406ae,
+	0x5880b610,
+	0x021086b1,
+	0xf4f01bf4,
+/* 0x02b2: find_done */
+	0x8eb90132,
+	0xf880fc02,
+/* 0x02b9: send */
+	0x9421f500,
+	0x9701f402,
+/* 0x02c2: recv */
+	0xe89800f8,
+	0x04e99805,
+	0xb80132f4,
+	0x0bf40689,
+	0x0389c43d,
+	0xf00180b6,
+	0xe8800784,
+	0x02ea9805,
+	0x8ffef0f9,
+	0xb9f0f901,
+	0x999402ef,
+	0x00e9bb04,
+	0x9818e0b6,
+	0xec9803eb,
+	0x01ed9802,
+	0xf900ee98,
+	0xfef0fca5,
+	0x31f400f8,
+/* 0x030b: recv_done */
+	0xf8f0fc01,
+/* 0x030d: init */
+	0x0817f100,
+	0x0614b601,
+	0xe70011cf,
+	0xb6010911,
+	0x14fe0814,
+	0xe017f100,
+	0x0013f000,
+	0xb61c07f0,
+	0x01d00604,
+	0xf004bd00,
+	0x07f0ff17,
+	0x0604b614,
+	0xbd0001d0,
+	0x0217f004,
+	0x080015f1,
+	0xb61007f0,
+	0x01d00604,
+	0xf104bd00,
+	0xf0010a17,
+	0x10fe0013,
+	0x1031f400,
+	0xf00117f0,
+	0x04b63807,
+	0x0001d006,
+	0xf7f004bd,
+/* 0x0371: init_proc */
+	0x01f19858,
+	0xf40016b0,
+	0x15f9fa0b,
+	0xf458f0b6,
+/* 0x0382: host_send */
+	0x17f1f20e,
+	0x14b604b0,
+	0x0011cf06,
+	0x04a027f1,
+	0xcf0624b6,
+	0x12b80022,
+	0x320bf406,
+	0x94071ec4,
+	0xe0b704ee,
+	0xeb980218,
+	0x02ec9803,
+	0x9801ed98,
+	0x21f500ee,
+	0x10b602b9,
+	0x0f1ec401,
+	0x04b007f1,
+	0xd00604b6,
+	0x04bd0001,
+/* 0x03cb: host_send_done */
+	0xf8ba0ef4,
+/* 0x03cd: host_recv */
+	0x4917f100,
+	0x5413f14e,
+	0x06e1b852,
+/* 0x03db: host_recv_wait */
+	0xf1aa0bf4,
+	0xb604cc17,
+	0x11cf0614,
+	0xc827f100,
+	0x0624b604,
+	0xf00022cf,
+	0x12b80816,
+	0xe60bf406,
+	0xb60723c4,
+	0x30b70434,
+	0x3b800298,
+	0x023c8003,
+	0x80013d80,
+	0x20b6003e,
+	0x0f24f001,
+	0x04c807f1,
+	0xd00604b6,
+	0x04bd0002,
+	0xf04027f0,
+	0x04b60007,
+	0x0002d006,
+	0x00f804bd,
+/* 0x0430: host_init */
+	0x008017f1,
+	0xf11014b6,
+	0xf1021815,
+	0xb604d007,
+	0x01d00604,
+	0xf104bd00,
+	0xb6008017,
+	0x15f11014,
+	0x07f10298,
+	0x04b604dc,
+	0x0001d006,
+	0x17f004bd,
+	0xc407f101,
+	0x0604b604,
+	0xbd0001d0,
+/* 0x046f: memx_func_enter */
+	0xf000f804,
+	0x07f10467,
+	0x04b607e0,
+	0x0006d006,
+/* 0x047e: memx_func_enter_wait */
+	0x67f104bd,
+	0x64b607c0,
+	0x0066cf06,
+	0xf40464f0,
+	0x1698f30b,
+	0x0410b600,
+/* 0x0496: memx_func_leave */
+	0x67f000f8,
+	0xe407f104,
+	0x0604b607,
+	0xbd0006d0,
+/* 0x04a5: memx_func_leave_wait */
+	0xc067f104,
+	0x0664b607,
+	0xf00066cf,
+	0x1bf40464,
+/* 0x04b7: memx_func_wr32 */
+	0x9800f8f3,
+	0x15980016,
+	0x0810b601,
+	0x50f960f9,
+	0xe0fcd0fc,
+	0xf13f21f4,
+	0xfd140003,
+	0x05800506,
+	0xb604bd00,
+	0x1bf40242,
+/* 0x04df: memx_func_wait */
+	0xf000f8dd,
+	0x84b62c87,
+	0x0088cf06,
+	0x98001e98,
+	0x1c98011d,
+	0x031b9802,
+	0xf41010b6,
+	0x00f89c21,
+/* 0x04fc: memx_func_delay */
+	0xb6001e98,
+	0x21f40410,
+/* 0x0507: memx_exec */
+	0xf900f87f,
+	0xb9d0f9e0,
+	0xb2b902c1,
+/* 0x0511: memx_exec_next */
+	0x00139802,
+	0x950410b6,
+	0x30f01034,
+	0xc835980c,
+	0x12b855f9,
+	0xec1ef406,
+	0xe0fcd0fc,
+	0x02b921f5,
+/* 0x0532: memx_info */
+	0xc7f100f8,
+	0xb7f10354,
+	0x21f50800,
+	0x00f802b9,
+/* 0x0540: memx_recv */
+	0xf401d6b0,
+	0xd6b0c40b,
+	0xe90bf400,
+/* 0x054e: memx_init */
+	0x00f800f8,
+/* 0x0550: perf_recv */
+/* 0x0552: perf_init */
+	0x00f800f8,
+/* 0x0554: test_recv */
+	0x05d817f1,
+	0xcf0614b6,
+	0x10b60011,
+	0xd807f101,
+	0x0604b605,
+	0xbd0001d0,
+	0x00e7f104,
+	0x4fe3f1d9,
+	0xf521f513,
+/* 0x057b: test_init */
+	0xf100f801,
+	0xf50800e7,
+	0xf801f521,
+/* 0x0585: idle_recv */
+/* 0x0587: idle */
+	0xf400f800,
+	0x17f10031,
+	0x14b605d4,
+	0x0011cf06,
+	0xf10110b6,
+	0xb605d407,
+	0x01d00604,
+/* 0x05a3: idle_loop */
+	0xf004bd00,
+	0x32f45817,
+/* 0x05a9: idle_proc */
+/* 0x05a9: idle_proc_exec */
+	0xb910f902,
+	0x21f5021e,
+	0x10fc02c2,
+	0xf40911f4,
+	0x0ef40231,
+/* 0x05bd: idle_proc_next */
+	0x5810b6ef,
+	0xf4061fb8,
+	0x02f4e61b,
+	0x0028f4dd,
+	0x00bb0ef4,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+};
diff --git a/drivers/gpu/drm/nouveau/core/subdev/pwr/fuc/nvd0.fuc b/drivers/gpu/drm/nouveau/core/subdev/pwr/fuc/nvd0.fuc
new file mode 100644
index 000000000000..32d65ea254dd
--- /dev/null
+++ b/drivers/gpu/drm/nouveau/core/subdev/pwr/fuc/nvd0.fuc
@@ -0,0 +1,63 @@
+/*
+ * Copyright 2013 Red Hat Inc.
+ *
+ * Permission is hereby granted, free of charge, to any person obtaining a
+ * copy of this software and associated documentation files (the "Software"),
+ * to deal in the Software without restriction, including without limitation
+ * the rights to use, copy, modify, merge, publish, distribute, sublicense,
+ * and/or sell copies of the Software, and to permit persons to whom the
+ * Software is furnished to do so, subject to the following conditions:
+ *
+ * The above copyright notice and this permission notice shall be included in
+ * all copies or substantial portions of the Software.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+ * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+ * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT.  IN NO EVENT SHALL
+ * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR
+ * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE,
+ * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR
+ * OTHER DEALINGS IN THE SOFTWARE.
+ *
+ * Authors: Ben Skeggs
+ */
+
+#define NVKM_PPWR_CHIPSET GF119
+
+//#define NVKM_FALCON_PC24
+#define NVKM_FALCON_UNSHIFTED_IO
+//#define NVKM_FALCON_MMIO_UAS
+//#define NVKM_FALCON_MMIO_TRAP
+
+#include "macros.fuc"
+
+.section #nvd0_pwr_data
+#define INCLUDE_PROC
+#include "kernel.fuc"
+#include "host.fuc"
+#include "memx.fuc"
+#include "perf.fuc"
+#include "test.fuc"
+#include "idle.fuc"
+#undef INCLUDE_PROC
+
+#define INCLUDE_DATA
+#include "kernel.fuc"
+#include "host.fuc"
+#include "memx.fuc"
+#include "perf.fuc"
+#include "test.fuc"
+#include "idle.fuc"
+#undef INCLUDE_DATA
+.align 256
+
+.section #nvd0_pwr_code
+#define INCLUDE_CODE
+#include "kernel.fuc"
+#include "host.fuc"
+#include "memx.fuc"
+#include "perf.fuc"
+#include "test.fuc"
+#include "idle.fuc"
+#undef INCLUDE_CODE
+.align 256
diff --git a/drivers/gpu/drm/nouveau/core/subdev/pwr/fuc/nvd0.fuc.h b/drivers/gpu/drm/nouveau/core/subdev/pwr/fuc/nvd0.fuc.h
new file mode 100644
index 000000000000..ce65e2a4b789
--- /dev/null
+++ b/drivers/gpu/drm/nouveau/core/subdev/pwr/fuc/nvd0.fuc.h
@@ -0,0 +1,1229 @@
+uint32_t nvd0_pwr_data[] = {
+/* 0x0000: proc_kern */
+	0x52544e49,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+/* 0x0058: proc_list_head */
+	0x54534f48,
+	0x000003be,
+	0x00000367,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x584d454d,
+	0x000004c4,
+	0x000004b6,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x46524550,
+	0x000004c8,
+	0x000004c6,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x54534554,
+	0x000004eb,
+	0x000004ca,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x454c4449,
+	0x000004f7,
+	0x000004f5,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+/* 0x0210: proc_list_tail */
+/* 0x0210: time_prev */
+	0x00000000,
+/* 0x0214: time_next */
+	0x00000000,
+/* 0x0218: fifo_queue */
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+/* 0x0298: rfifo_queue */
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+/* 0x0318: memx_func_head */
+	0x00010000,
+	0x00000000,
+	0x000003f4,
+/* 0x0324: memx_func_next */
+	0x00000001,
+	0x00000000,
+	0x00000415,
+	0x00000002,
+	0x00000002,
+	0x00000430,
+	0x00040003,
+	0x00000000,
+	0x00000458,
+	0x00010004,
+	0x00000000,
+	0x00000472,
+/* 0x0354: memx_func_tail */
+/* 0x0354: memx_data_head */
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+/* 0x0b54: memx_data_tail */
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+};
+
+uint32_t nvd0_pwr_code[] = {
+	0x02bf0ef5,
+/* 0x0004: rd32 */
+	0x07a007f1,
+	0xbd000ed0,
+	0x01e7f004,
+	0xf101e3f0,
+	0xd007ac07,
+	0x04bd000e,
+/* 0x001c: rd32_wait */
+	0x07ace7f1,
+	0xf100eecf,
+	0xf47000e4,
+	0xd7f1f51b,
+	0xddcf07a4,
+/* 0x0033: wr32 */
+	0xf100f800,
+	0xd007a007,
+	0x04bd000e,
+	0x07a407f1,
+	0xbd000dd0,
+	0x02e7f004,
+	0xf0f0e5f0,
+	0x07f101e3,
+	0x0ed007ac,
+/* 0x0057: wr32_wait */
+	0xf104bd00,
+	0xcf07ace7,
+	0xe4f100ee,
+	0x1bf47000,
+/* 0x0067: nsec */
+	0xf000f8f5,
+	0x88cf2c87,
+/* 0x006d: nsec_loop */
+	0x2c97f000,
+	0xbb0099cf,
+	0x9eb80298,
+	0xf41ef406,
+/* 0x007e: wait */
+	0x87f000f8,
+	0x0088cf2c,
+/* 0x0084: wait_loop */
+	0xf402eeb9,
+	0xdab90421,
+	0x04adfd02,
+	0xf406acb8,
+	0x97f0120b,
+	0x0099cf2c,
+	0xb80298bb,
+	0x1ef4069b,
+/* 0x00a5: wait_done */
+/* 0x00a7: intr_watchdog */
+	0x9800f8e2,
+	0x96b003e9,
+	0x2a0bf400,
+	0xbb840a98,
+	0x1cf4029a,
+	0x01d7f00f,
+	0x020621f5,
+	0x0ef494bd,
+/* 0x00c5: intr_watchdog_next_time */
+	0x850a9815,
+	0xf400a6b0,
+	0x9ab8090b,
+	0x061cf406,
+/* 0x00d4: intr_watchdog_next_time_set */
+/* 0x00d7: intr_watchdog_next_proc */
+	0x80850980,
+	0xe0b603e9,
+	0x10e6b158,
+	0xc61bf402,
+/* 0x00e6: intr */
+	0x00f900f8,
+	0x80f904bd,
+	0xa0f990f9,
+	0xc0f9b0f9,
+	0xe0f9d0f9,
+	0xf7f0f0f9,
+	0x0188fe00,
+	0x87f180f9,
+	0x88cf05d0,
+	0x0180b600,
+	0x05d007f1,
+	0xbd0008d0,
+	0x0887f004,
+	0xc40088cf,
+	0x0bf40289,
+	0x85008020,
+	0xf458e7f0,
+	0x0998a721,
+	0x0096b085,
+	0xf00e0bf4,
+	0x09d03407,
+	0x8004bd00,
+/* 0x013e: intr_skip_watchdog */
+	0x89e48409,
+	0x0bf40800,
+	0x8897f13c,
+	0x0099cf06,
+	0xf4029ac4,
+	0xc7f1260b,
+	0xcccf04c0,
+	0xf1c0f900,
+	0xf14f48e7,
+	0xf05453e3,
+	0x21f500d7,
+	0xc0fc026b,
+	0x04c007f1,
+	0xbd000cd0,
+/* 0x0175: intr_subintr_skip_fifo */
+	0x8807f104,
+	0x0009d006,
+/* 0x017e: intr_skip_subintr */
+	0x89c404bd,
+	0x070bf420,
+	0xffbfa4f1,
+/* 0x0188: intr_skip_pause */
+	0xf44089c4,
+	0xa4f1070b,
+/* 0x0192: intr_skip_user0 */
+	0x07f0ffbf,
+	0x0008d004,
+	0x80fc04bd,
+	0xfc0088fe,
+	0xfce0fcf0,
+	0xfcc0fcd0,
+	0xfca0fcb0,
+	0xfc80fc90,
+	0x0032f400,
+/* 0x01b6: timer */
+	0x32f401f8,
+	0x03f89810,
+	0xf40086b0,
+	0xfe80421c,
+	0x3807f003,
+	0xbd0008d0,
+	0x0887f004,
+	0xf00088cf,
+	0x1bf40284,
+	0x3487f020,
+	0xb80088cf,
+	0x0bf406e0,
+	0x06e8b809,
+/* 0x01eb: timer_reset */
+	0xf0191ef4,
+	0x0ed03407,
+	0x8004bd00,
+/* 0x01f6: timer_enable */
+	0x87f0840e,
+	0x3807f001,
+	0xbd0008d0,
+/* 0x0201: timer_done */
+	0x1031f404,
+/* 0x0206: send_proc */
+	0x80f900f8,
+	0xe89890f9,
+	0x04e99805,
+	0xb80486f0,
+	0x0bf40689,
+	0x0398c42a,
+	0xb6048894,
+	0x8ebb1880,
+	0x00fa9800,
+	0x80008a80,
+	0x8c80018d,
+	0x038b8002,
+	0xf00190b6,
+	0xe9800794,
+	0x0231f404,
+/* 0x0240: send_done */
+	0x80fc90fc,
+/* 0x0246: find */
+	0x80f900f8,
+	0xf45887f0,
+/* 0x024e: find_loop */
+	0x8a980131,
+	0x06aeb800,
+	0xb6100bf4,
+	0x86b15880,
+	0x1bf40210,
+	0x0132f4f0,
+/* 0x0264: find_done */
+	0xfc028eb9,
+/* 0x026b: send */
+	0xf500f880,
+	0xf4024621,
+	0x00f89701,
+/* 0x0274: recv */
+	0x9805e898,
+	0x32f404e9,
+	0x0689b801,
+	0xc43d0bf4,
+	0x80b60389,
+	0x0784f001,
+	0x9805e880,
+	0xf0f902ea,
+	0xf9018ffe,
+	0x02efb9f0,
+	0xbb049994,
+	0xe0b600e9,
+	0x03eb9818,
+	0x9802ec98,
+	0xee9801ed,
+	0xfca5f900,
+	0x00f8fef0,
+	0xfc0131f4,
+/* 0x02bd: recv_done */
+/* 0x02bf: init */
+	0xf100f8f0,
+	0xcf010817,
+	0x11e70011,
+	0x14b60109,
+	0x0014fe08,
+	0x00e017f1,
+	0xf00013f0,
+	0x01d01c07,
+	0xf004bd00,
+	0x07f0ff17,
+	0x0001d014,
+	0x17f004bd,
+	0x0015f102,
+	0x1007f008,
+	0xbd0001d0,
+	0xe617f104,
+	0x0013f000,
+	0xf40010fe,
+	0x17f01031,
+	0x3807f001,
+	0xbd0001d0,
+	0x58f7f004,
+/* 0x0314: init_proc */
+	0xb001f198,
+	0x0bf40016,
+	0xb615f9fa,
+	0x0ef458f0,
+/* 0x0325: host_send */
+	0xb017f1f2,
+	0x0011cf04,
+	0x04a027f1,
+	0xb80022cf,
+	0x0bf40612,
+	0x071ec42f,
+	0xb704ee94,
+	0x980218e0,
+	0xec9803eb,
+	0x01ed9802,
+	0xf500ee98,
+	0xb6026b21,
+	0x1ec40110,
+	0xb007f10f,
+	0x0001d004,
+	0x0ef404bd,
+/* 0x0365: host_send_done */
+/* 0x0367: host_recv */
+	0xf100f8c3,
+	0xf14e4917,
+	0xb8525413,
+	0x0bf406e1,
+/* 0x0375: host_recv_wait */
+	0xcc17f1b3,
+	0x0011cf04,
+	0x04c827f1,
+	0xf00022cf,
+	0x12b80816,
+	0xec0bf406,
+	0xb60723c4,
+	0x30b70434,
+	0x3b800298,
+	0x023c8003,
+	0x80013d80,
+	0x20b6003e,
+	0x0f24f001,
+	0x04c807f1,
+	0xbd0002d0,
+	0x4027f004,
+	0xd00007f0,
+	0x04bd0002,
+/* 0x03be: host_init */
+	0x17f100f8,
+	0x14b60080,
+	0x1815f110,
+	0xd007f102,
+	0x0001d004,
+	0x17f104bd,
+	0x14b60080,
+	0x9815f110,
+	0xdc07f102,
+	0x0001d004,
+	0x17f004bd,
+	0xc407f101,
+	0x0001d004,
+	0x00f804bd,
+/* 0x03f4: memx_func_enter */
+	0xf10467f0,
+	0xd007e007,
+	0x04bd0006,
+/* 0x0400: memx_func_enter_wait */
+	0x07c067f1,
+	0xf00066cf,
+	0x0bf40464,
+	0x001698f6,
+	0xf80410b6,
+/* 0x0415: memx_func_leave */
+	0x0467f000,
+	0x07e407f1,
+	0xbd0006d0,
+/* 0x0421: memx_func_leave_wait */
+	0xc067f104,
+	0x0066cf07,
+	0xf40464f0,
+	0x00f8f61b,
+/* 0x0430: memx_func_wr32 */
+	0x98001698,
+	0x10b60115,
+	0xf960f908,
+	0xfcd0fc50,
+	0x3321f4e0,
+	0x140003f1,
+	0x800506fd,
+	0x04bd0005,
+	0xf40242b6,
+	0x00f8dd1b,
+/* 0x0458: memx_func_wait */
+	0xcf2c87f0,
+	0x1e980088,
+	0x011d9800,
+	0x98021c98,
+	0x10b6031b,
+	0x7e21f410,
+/* 0x0472: memx_func_delay */
+	0x1e9800f8,
+	0x0410b600,
+	0xf86721f4,
+/* 0x047d: memx_exec */
+	0xf9e0f900,
+	0x02c1b9d0,
+/* 0x0487: memx_exec_next */
+	0x9802b2b9,
+	0x10b60013,
+	0x10349504,
+	0x980c30f0,
+	0x55f9c835,
+	0xf40612b8,
+	0xd0fcec1e,
+	0x21f5e0fc,
+	0x00f8026b,
+/* 0x04a8: memx_info */
+	0x0354c7f1,
+	0x0800b7f1,
+	0x026b21f5,
+/* 0x04b6: memx_recv */
+	0xd6b000f8,
+	0xc40bf401,
+	0xf400d6b0,
+	0x00f8e90b,
+/* 0x04c4: memx_init */
+/* 0x04c6: perf_recv */
+	0x00f800f8,
+/* 0x04c8: perf_init */
+/* 0x04ca: test_recv */
+	0x17f100f8,
+	0x11cf05d8,
+	0x0110b600,
+	0x05d807f1,
+	0xbd0001d0,
+	0x00e7f104,
+	0x4fe3f1d9,
+	0xb621f513,
+/* 0x04eb: test_init */
+	0xf100f801,
+	0xf50800e7,
+	0xf801b621,
+/* 0x04f5: idle_recv */
+/* 0x04f7: idle */
+	0xf400f800,
+	0x17f10031,
+	0x11cf05d4,
+	0x0110b600,
+	0x05d407f1,
+	0xbd0001d0,
+/* 0x050d: idle_loop */
+	0x5817f004,
+/* 0x0513: idle_proc */
+/* 0x0513: idle_proc_exec */
+	0xf90232f4,
+	0x021eb910,
+	0x027421f5,
+	0x11f410fc,
+	0x0231f409,
+/* 0x0527: idle_proc_next */
+	0xb6ef0ef4,
+	0x1fb85810,
+	0xe61bf406,
+	0xf4dd02f4,
+	0x0ef40028,
+	0x000000c1,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+};
diff --git a/drivers/gpu/drm/nouveau/core/subdev/pwr/fuc/os.h b/drivers/gpu/drm/nouveau/core/subdev/pwr/fuc/os.h
new file mode 100644
index 000000000000..5fb0cccc6c64
--- /dev/null
+++ b/drivers/gpu/drm/nouveau/core/subdev/pwr/fuc/os.h
@@ -0,0 +1,27 @@
+#ifndef __NVKM_PWR_OS_H__
+#define __NVKM_PWR_OS_H__
+
+/* Process names */
+#define PROC_KERN 0x52544e49
+#define PROC_IDLE 0x454c4449
+#define PROC_HOST 0x54534f48
+#define PROC_MEMX 0x584d454d
+#define PROC_PERF 0x46524550
+#define PROC_TEST 0x54534554
+
+/* KERN: message identifiers */
+#define KMSG_FIFO   0x00000000
+#define KMSG_ALARM  0x00000001
+
+/* MEMX: message identifiers */
+#define MEMX_MSG_INFO 0
+#define MEMX_MSG_EXEC 1
+
+/* MEMX: script opcode definitions */
+#define MEMX_ENTER  0
+#define MEMX_LEAVE  1
+#define MEMX_WR32   2
+#define MEMX_WAIT   3
+#define MEMX_DELAY  4
+
+#endif
diff --git a/drivers/gpu/drm/nouveau/core/subdev/pwr/fuc/perf.fuc b/drivers/gpu/drm/nouveau/core/subdev/pwr/fuc/perf.fuc
new file mode 100644
index 000000000000..38eadf705cbf
--- /dev/null
+++ b/drivers/gpu/drm/nouveau/core/subdev/pwr/fuc/perf.fuc
@@ -0,0 +1,57 @@
+/*
+ * Copyright 2013 Red Hat Inc.
+ *
+ * Permission is hereby granted, free of charge, to any person obtaining a
+ * copy of this software and associated documentation files (the "Software"),
+ * to deal in the Software without restriction, including without limitation
+ * the rights to use, copy, modify, merge, publish, distribute, sublicense,
+ * and/or sell copies of the Software, and to permit persons to whom the
+ * Software is furnished to do so, subject to the following conditions:
+ *
+ * The above copyright notice and this permission notice shall be included in
+ * all copies or substantial portions of the Software.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+ * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+ * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT.  IN NO EVENT SHALL
+ * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR
+ * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE,
+ * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR
+ * OTHER DEALINGS IN THE SOFTWARE.
+ *
+ * Authors: Ben Skeggs
+ */
+
+#ifdef INCLUDE_PROC
+process(PROC_PERF, #perf_init, #perf_recv)
+#endif
+
+/******************************************************************************
+ * PERF data segment
+ *****************************************************************************/
+#ifdef INCLUDE_DATA
+#endif
+
+/******************************************************************************
+ * PERF code segment
+ *****************************************************************************/
+#ifdef INCLUDE_CODE
+
+// description
+//
+// $r15 - current (perf)
+// $r14 - sender process name
+// $r13 - message
+// $r12 - data0
+// $r11 - data1
+// $r0  - zero
+perf_recv:
+	ret
+
+// description
+//
+// $r15 - current (perf)
+// $r0  - zero
+perf_init:
+	ret
+#endif
diff --git a/drivers/gpu/drm/nouveau/core/subdev/pwr/fuc/test.fuc b/drivers/gpu/drm/nouveau/core/subdev/pwr/fuc/test.fuc
new file mode 100644
index 000000000000..0c3a71bf5459
--- /dev/null
+++ b/drivers/gpu/drm/nouveau/core/subdev/pwr/fuc/test.fuc
@@ -0,0 +1,64 @@
+/*
+ * Copyright 2013 Red Hat Inc.
+ *
+ * Permission is hereby granted, free of charge, to any person obtaining a
+ * copy of this software and associated documentation files (the "Software"),
+ * to deal in the Software without restriction, including without limitation
+ * the rights to use, copy, modify, merge, publish, distribute, sublicense,
+ * and/or sell copies of the Software, and to permit persons to whom the
+ * Software is furnished to do so, subject to the following conditions:
+ *
+ * The above copyright notice and this permission notice shall be included in
+ * all copies or substantial portions of the Software.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+ * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+ * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT.  IN NO EVENT SHALL
+ * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR
+ * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE,
+ * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR
+ * OTHER DEALINGS IN THE SOFTWARE.
+ *
+ * Authors: Ben Skeggs
+ */
+
+#ifdef INCLUDE_PROC
+process(PROC_TEST, #test_init, #test_recv)
+#endif
+
+/******************************************************************************
+ * TEST data segment
+ *****************************************************************************/
+#ifdef INCLUDE_DATA
+#endif
+
+/******************************************************************************
+ * TEST code segment
+ *****************************************************************************/
+#ifdef INCLUDE_CODE
+// description
+//
+// $r15 - current (test)
+// $r14 - sender process name
+// $r13 - message
+// $r12 - data0
+// $r11 - data1
+// $r0  - zero
+test_recv:
+	nv_iord($r1, NV_PPWR_DSCRATCH(2))
+	add b32 $r1 1
+	nv_iowr(NV_PPWR_DSCRATCH(2), $r1)
+	mov $r14 -0x2700 /* 0xd900, envyas grrr! */
+	sethi $r14 0x134f0000
+	call(timer)
+	ret
+
+// description
+//
+// $r15 - current (test)
+// $r0  - zero
+test_init:
+	mov $r14 0x800
+	call(timer)
+	ret
+#endif
diff --git a/drivers/gpu/drm/nouveau/core/subdev/pwr/memx.c b/drivers/gpu/drm/nouveau/core/subdev/pwr/memx.c
new file mode 100644
index 000000000000..03de3107d29f
--- /dev/null
+++ b/drivers/gpu/drm/nouveau/core/subdev/pwr/memx.c
@@ -0,0 +1,121 @@
+#ifndef __NVKM_PWR_MEMX_H__
+#define __NVKM_PWR_MEMX_H__
+
+#include <subdev/pwr.h>
+#include <subdev/pwr/fuc/os.h>
+
+struct nouveau_memx {
+	struct nouveau_pwr *ppwr;
+	u32 base;
+	u32 size;
+	struct {
+		u32 mthd;
+		u32 size;
+		u32 data[64];
+	} c;
+};
+
+static void
+memx_out(struct nouveau_memx *memx)
+{
+	struct nouveau_pwr *ppwr = memx->ppwr;
+	int i;
+
+	if (memx->c.size) {
+		nv_wr32(ppwr, 0x10a1c4, (memx->c.size << 16) | memx->c.mthd);
+		for (i = 0; i < memx->c.size; i++)
+			nv_wr32(ppwr, 0x10a1c4, memx->c.data[i]);
+		memx->c.size = 0;
+	}
+}
+
+static void
+memx_cmd(struct nouveau_memx *memx, u32 mthd, u32 size, u32 data[])
+{
+	if ((memx->c.size + size >= ARRAY_SIZE(memx->c.data)) ||
+	    (memx->c.size && memx->c.mthd != mthd))
+		memx_out(memx);
+	memcpy(&memx->c.data[memx->c.size], data, size * sizeof(data[0]));
+	memx->c.size += size;
+	memx->c.mthd  = mthd;
+}
+
+int
+nouveau_memx_init(struct nouveau_pwr *ppwr, struct nouveau_memx **pmemx)
+{
+	struct nouveau_memx *memx;
+	u32 reply[2];
+	int ret;
+
+	ret = ppwr->message(ppwr, reply, PROC_MEMX, MEMX_MSG_INFO, 0, 0);
+	if (ret)
+		return ret;
+
+	memx = *pmemx = kzalloc(sizeof(*memx), GFP_KERNEL);
+	if (!memx)
+		return -ENOMEM;
+	memx->ppwr = ppwr;
+	memx->base = reply[0];
+	memx->size = reply[1];
+
+	/* acquire data segment access */
+	do {
+		nv_wr32(ppwr, 0x10a580, 0x00000003);
+	} while (nv_rd32(ppwr, 0x10a580) != 0x00000003);
+	nv_wr32(ppwr, 0x10a1c0, 0x01000000 | memx->base);
+	nv_wr32(ppwr, 0x10a1c4, 0x00010000 | MEMX_ENTER);
+	nv_wr32(ppwr, 0x10a1c4, 0x00000000);
+	return 0;
+}
+
+int
+nouveau_memx_fini(struct nouveau_memx **pmemx, bool exec)
+{
+	struct nouveau_memx *memx = *pmemx;
+	struct nouveau_pwr *ppwr = memx->ppwr;
+	u32 finish, reply[2];
+
+	/* flush the cache... */
+	memx_out(memx);
+
+	/* release data segment access */
+	nv_wr32(ppwr, 0x10a1c4, 0x00000000 | MEMX_LEAVE);
+	finish = nv_rd32(ppwr, 0x10a1c0) & 0x00ffffff;
+	nv_wr32(ppwr, 0x10a580, 0x00000000);
+
+	/* call MEMX process to execute the script, and wait for reply */
+	if (exec) {
+		ppwr->message(ppwr, reply, PROC_MEMX, MEMX_MSG_EXEC,
+				 memx->base, finish);
+	}
+
+	kfree(memx);
+	return 0;
+}
+
+void
+nouveau_memx_wr32(struct nouveau_memx *memx, u32 addr, u32 data)
+{
+	nv_debug(memx->ppwr, "R[%06x] = 0x%08x\n", addr, data);
+	memx_cmd(memx, MEMX_WR32, 2, (u32[]){ addr, data });
+}
+
+void
+nouveau_memx_wait(struct nouveau_memx *memx,
+		  u32 addr, u32 mask, u32 data, u32 nsec)
+{
+	nv_debug(memx->ppwr, "R[%06x] & 0x%08x == 0x%08x, %d us\n",
+				addr, mask, data, nsec);
+	memx_cmd(memx, MEMX_WAIT, 4, (u32[]){ addr, ~mask, data, nsec });
+	memx_out(memx); /* fuc can't handle multiple */
+}
+
+void
+nouveau_memx_nsec(struct nouveau_memx *memx, u32 nsec)
+{
+	nv_debug(memx->ppwr, "    DELAY = %d ns\n", nsec);
+	memx_cmd(memx, MEMX_DELAY, 1, (u32[]){ nsec });
+	memx_out(memx); /* fuc can't handle multiple */
+}
+
+#endif
diff --git a/drivers/gpu/drm/nouveau/core/subdev/pwr/nv108.c b/drivers/gpu/drm/nouveau/core/subdev/pwr/nv108.c
new file mode 100644
index 000000000000..52c85414866a
--- /dev/null
+++ b/drivers/gpu/drm/nouveau/core/subdev/pwr/nv108.c
@@ -0,0 +1,62 @@
+/*
+ * Copyright 2013 Red Hat Inc.
+ *
+ * Permission is hereby granted, free of charge, to any person obtaining a
+ * copy of this software and associated documentation files (the "Software"),
+ * to deal in the Software without restriction, including without limitation
+ * the rights to use, copy, modify, merge, publish, distribute, sublicense,
+ * and/or sell copies of the Software, and to permit persons to whom the
+ * Software is furnished to do so, subject to the following conditions:
+ *
+ * The above copyright notice and this permission notice shall be included in
+ * all copies or substantial portions of the Software.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+ * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+ * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT.  IN NO EVENT SHALL
+ * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR
+ * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE,
+ * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR
+ * OTHER DEALINGS IN THE SOFTWARE.
+ *
+ * Authors: Ben Skeggs
+ */
+
+#include <subdev/pwr.h>
+
+#include "fuc/nv108.fuc.h"
+
+struct nv108_pwr_priv {
+	struct nouveau_pwr base;
+};
+
+static int
+nv108_pwr_ctor(struct nouveau_object *parent, struct nouveau_object *engine,
+	       struct nouveau_oclass *oclass, void *data, u32 size,
+	       struct nouveau_object **pobject)
+{
+	struct nv108_pwr_priv *priv;
+	int ret;
+
+	ret = nouveau_pwr_create(parent, engine, oclass, &priv);
+	*pobject = nv_object(priv);
+	if (ret)
+		return ret;
+
+	priv->base.code.data = nv108_pwr_code;
+	priv->base.code.size = sizeof(nv108_pwr_code);
+	priv->base.data.data = nv108_pwr_data;
+	priv->base.data.size = sizeof(nv108_pwr_data);
+	return 0;
+}
+
+struct nouveau_oclass
+nv108_pwr_oclass = {
+	.handle = NV_SUBDEV(PWR, 0x00),
+	.ofuncs = &(struct nouveau_ofuncs) {
+		.ctor = nv108_pwr_ctor,
+		.dtor = _nouveau_pwr_dtor,
+		.init = _nouveau_pwr_init,
+		.fini = _nouveau_pwr_fini,
+	},
+};
diff --git a/drivers/gpu/drm/nouveau/core/subdev/pwr/nva3.c b/drivers/gpu/drm/nouveau/core/subdev/pwr/nva3.c
new file mode 100644
index 000000000000..c132b7ca9747
--- /dev/null
+++ b/drivers/gpu/drm/nouveau/core/subdev/pwr/nva3.c
@@ -0,0 +1,71 @@
+/*
+ * Copyright 2013 Red Hat Inc.
+ *
+ * Permission is hereby granted, free of charge, to any person obtaining a
+ * copy of this software and associated documentation files (the "Software"),
+ * to deal in the Software without restriction, including without limitation
+ * the rights to use, copy, modify, merge, publish, distribute, sublicense,
+ * and/or sell copies of the Software, and to permit persons to whom the
+ * Software is furnished to do so, subject to the following conditions:
+ *
+ * The above copyright notice and this permission notice shall be included in
+ * all copies or substantial portions of the Software.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+ * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+ * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT.  IN NO EVENT SHALL
+ * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR
+ * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE,
+ * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR
+ * OTHER DEALINGS IN THE SOFTWARE.
+ *
+ * Authors: Ben Skeggs
+ */
+
+#include <subdev/pwr.h>
+
+#include "fuc/nva3.fuc.h"
+
+struct nva3_pwr_priv {
+	struct nouveau_pwr base;
+};
+
+static int
+nva3_pwr_init(struct nouveau_object *object)
+{
+	struct nva3_pwr_priv *priv = (void *)object;
+	nv_mask(priv, 0x022210, 0x00000001, 0x00000000);
+	nv_mask(priv, 0x022210, 0x00000001, 0x00000001);
+	return nouveau_pwr_init(&priv->base);
+}
+
+static int
+nva3_pwr_ctor(struct nouveau_object *parent, struct nouveau_object *engine,
+	      struct nouveau_oclass *oclass, void *data, u32 size,
+	      struct nouveau_object **pobject)
+{
+	struct nva3_pwr_priv *priv;
+	int ret;
+
+	ret = nouveau_pwr_create(parent, engine, oclass, &priv);
+	*pobject = nv_object(priv);
+	if (ret)
+		return ret;
+
+	priv->base.code.data = nva3_pwr_code;
+	priv->base.code.size = sizeof(nva3_pwr_code);
+	priv->base.data.data = nva3_pwr_data;
+	priv->base.data.size = sizeof(nva3_pwr_data);
+	return 0;
+}
+
+struct nouveau_oclass
+nva3_pwr_oclass = {
+	.handle = NV_SUBDEV(PWR, 0xa3),
+	.ofuncs = &(struct nouveau_ofuncs) {
+		.ctor = nva3_pwr_ctor,
+		.dtor = _nouveau_pwr_dtor,
+		.init = nva3_pwr_init,
+		.fini = _nouveau_pwr_fini,
+	},
+};
diff --git a/drivers/gpu/drm/nouveau/core/subdev/pwr/nvc0.c b/drivers/gpu/drm/nouveau/core/subdev/pwr/nvc0.c
new file mode 100644
index 000000000000..495f6857428d
--- /dev/null
+++ b/drivers/gpu/drm/nouveau/core/subdev/pwr/nvc0.c
@@ -0,0 +1,62 @@
+/*
+ * Copyright 2013 Red Hat Inc.
+ *
+ * Permission is hereby granted, free of charge, to any person obtaining a
+ * copy of this software and associated documentation files (the "Software"),
+ * to deal in the Software without restriction, including without limitation
+ * the rights to use, copy, modify, merge, publish, distribute, sublicense,
+ * and/or sell copies of the Software, and to permit persons to whom the
+ * Software is furnished to do so, subject to the following conditions:
+ *
+ * The above copyright notice and this permission notice shall be included in
+ * all copies or substantial portions of the Software.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+ * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+ * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT.  IN NO EVENT SHALL
+ * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR
+ * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE,
+ * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR
+ * OTHER DEALINGS IN THE SOFTWARE.
+ *
+ * Authors: Ben Skeggs
+ */
+
+#include <subdev/pwr.h>
+
+#include "fuc/nvc0.fuc.h"
+
+struct nvc0_pwr_priv {
+	struct nouveau_pwr base;
+};
+
+static int
+nvc0_pwr_ctor(struct nouveau_object *parent, struct nouveau_object *engine,
+	      struct nouveau_oclass *oclass, void *data, u32 size,
+	      struct nouveau_object **pobject)
+{
+	struct nvc0_pwr_priv *priv;
+	int ret;
+
+	ret = nouveau_pwr_create(parent, engine, oclass, &priv);
+	*pobject = nv_object(priv);
+	if (ret)
+		return ret;
+
+	priv->base.code.data = nvc0_pwr_code;
+	priv->base.code.size = sizeof(nvc0_pwr_code);
+	priv->base.data.data = nvc0_pwr_data;
+	priv->base.data.size = sizeof(nvc0_pwr_data);
+	return 0;
+}
+
+struct nouveau_oclass
+nvc0_pwr_oclass = {
+	.handle = NV_SUBDEV(PWR, 0xc0),
+	.ofuncs = &(struct nouveau_ofuncs) {
+		.ctor = nvc0_pwr_ctor,
+		.dtor = _nouveau_pwr_dtor,
+		.init = _nouveau_pwr_init,
+		.fini = _nouveau_pwr_fini,
+	},
+};
diff --git a/drivers/gpu/drm/nouveau/core/subdev/pwr/nvd0.c b/drivers/gpu/drm/nouveau/core/subdev/pwr/nvd0.c
new file mode 100644
index 000000000000..043aa142fe82
--- /dev/null
+++ b/drivers/gpu/drm/nouveau/core/subdev/pwr/nvd0.c
@@ -0,0 +1,62 @@
+/*
+ * Copyright 2013 Red Hat Inc.
+ *
+ * Permission is hereby granted, free of charge, to any person obtaining a
+ * copy of this software and associated documentation files (the "Software"),
+ * to deal in the Software without restriction, including without limitation
+ * the rights to use, copy, modify, merge, publish, distribute, sublicense,
+ * and/or sell copies of the Software, and to permit persons to whom the
+ * Software is furnished to do so, subject to the following conditions:
+ *
+ * The above copyright notice and this permission notice shall be included in
+ * all copies or substantial portions of the Software.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+ * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+ * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT.  IN NO EVENT SHALL
+ * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR
+ * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE,
+ * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR
+ * OTHER DEALINGS IN THE SOFTWARE.
+ *
+ * Authors: Ben Skeggs
+ */
+
+#include <subdev/pwr.h>
+
+#include "fuc/nvd0.fuc.h"
+
+struct nvd0_pwr_priv {
+	struct nouveau_pwr base;
+};
+
+static int
+nvd0_pwr_ctor(struct nouveau_object *parent, struct nouveau_object *engine,
+	      struct nouveau_oclass *oclass, void *data, u32 size,
+	      struct nouveau_object **pobject)
+{
+	struct nvd0_pwr_priv *priv;
+	int ret;
+
+	ret = nouveau_pwr_create(parent, engine, oclass, &priv);
+	*pobject = nv_object(priv);
+	if (ret)
+		return ret;
+
+	priv->base.code.data = nvd0_pwr_code;
+	priv->base.code.size = sizeof(nvd0_pwr_code);
+	priv->base.data.data = nvd0_pwr_data;
+	priv->base.data.size = sizeof(nvd0_pwr_data);
+	return 0;
+}
+
+struct nouveau_oclass
+nvd0_pwr_oclass = {
+	.handle = NV_SUBDEV(PWR, 0xd0),
+	.ofuncs = &(struct nouveau_ofuncs) {
+		.ctor = nvd0_pwr_ctor,
+		.dtor = _nouveau_pwr_dtor,
+		.init = _nouveau_pwr_init,
+		.fini = _nouveau_pwr_fini,
+	},
+};
diff --git a/drivers/gpu/drm/nouveau/core/subdev/therm/base.c b/drivers/gpu/drm/nouveau/core/subdev/therm/base.c
index f1de7a9c572b..80e584a1bd1c 100644
--- a/drivers/gpu/drm/nouveau/core/subdev/therm/base.c
+++ b/drivers/gpu/drm/nouveau/core/subdev/therm/base.c
@@ -92,10 +92,11 @@ nouveau_therm_update(struct nouveau_therm *therm, int mode)
 	struct nouveau_timer *ptimer = nouveau_timer(therm);
 	struct nouveau_therm_priv *priv = (void *)therm;
 	unsigned long flags;
-	int duty;
+	bool immd = true;
+	bool poll = true;
+	int duty = -1;
 
 	spin_lock_irqsave(&priv->lock, flags);
-	nv_debug(therm, "FAN speed check\n");
 	if (mode < 0)
 		mode = priv->mode;
 	priv->mode = mode;
@@ -106,28 +107,49 @@ nouveau_therm_update(struct nouveau_therm *therm, int mode)
 		duty = nouveau_therm_fan_get(therm);
 		if (duty < 0)
 			duty = 100;
+		poll = false;
 		break;
 	case NOUVEAU_THERM_CTRL_AUTO:
-		if (priv->fan->bios.nr_fan_trip)
+		if (priv->fan->bios.nr_fan_trip) {
 			duty = nouveau_therm_update_trip(therm);
-		else
+		} else
+		if (priv->fan->bios.linear_min_temp ||
+		    priv->fan->bios.linear_max_temp) {
 			duty = nouveau_therm_update_linear(therm);
+		} else {
+			if (priv->cstate)
+				duty = priv->cstate;
+			poll = false;
+		}
+		immd = false;
 		break;
 	case NOUVEAU_THERM_CTRL_NONE:
 	default:
 		ptimer->alarm_cancel(ptimer, &priv->alarm);
-		goto done;
+		poll = false;
 	}
 
-	nv_debug(therm, "FAN target request: %d%%\n", duty);
-	nouveau_therm_fan_set(therm, (mode != NOUVEAU_THERM_CTRL_AUTO), duty);
-
-done:
-	if (list_empty(&priv->alarm.head) && (mode == NOUVEAU_THERM_CTRL_AUTO))
+	if (list_empty(&priv->alarm.head) && poll)
 		ptimer->alarm(ptimer, 1000000000ULL, &priv->alarm);
-	else if (!list_empty(&priv->alarm.head))
-		nv_debug(therm, "therm fan alarm list is not empty\n");
 	spin_unlock_irqrestore(&priv->lock, flags);
+
+	if (duty >= 0) {
+		nv_debug(therm, "FAN target request: %d%%\n", duty);
+		nouveau_therm_fan_set(therm, immd, duty);
+	}
+}
+
+int
+nouveau_therm_cstate(struct nouveau_therm *ptherm, int fan, int dir)
+{
+	struct nouveau_therm_priv *priv = (void *)ptherm;
+	if (!dir || (dir < 0 && fan < priv->cstate) ||
+		    (dir > 0 && fan > priv->cstate)) {
+		nv_debug(ptherm, "default fan speed -> %d%%\n", fan);
+		priv->cstate = fan;
+		nouveau_therm_update(ptherm, -1);
+	}
+	return 0;
 }
 
 static void
@@ -149,14 +171,15 @@ nouveau_therm_fan_mode(struct nouveau_therm *therm, int mode)
 		"automatic"
 	};
 
-	/* The default PDAEMON ucode interferes with fan management */
+	/* The default PPWR ucode on fermi interferes with fan management */
 	if ((mode >= ARRAY_SIZE(name)) ||
-	    (mode != NOUVEAU_THERM_CTRL_NONE && device->card_type >= NV_C0))
+	    (mode != NOUVEAU_THERM_CTRL_NONE && device->card_type >= NV_C0 &&
+	     !nouveau_subdev(device, NVDEV_SUBDEV_PWR)))
 		return -EINVAL;
 
 	/* do not allow automatic fan management if the thermal sensor is
 	 * not available */
-	if (priv->mode == 2 && therm->temp_get(therm) < 0)
+	if (priv->mode == NOUVEAU_THERM_CTRL_AUTO && therm->temp_get(therm) < 0)
 		return -EINVAL;
 
 	if (priv->mode == mode)
@@ -335,7 +358,7 @@ nouveau_therm_preinit(struct nouveau_therm *therm)
 	nouveau_therm_ic_ctor(therm);
 	nouveau_therm_fan_ctor(therm);
 
-	nouveau_therm_fan_mode(therm, NOUVEAU_THERM_CTRL_NONE);
+	nouveau_therm_fan_mode(therm, NOUVEAU_THERM_CTRL_AUTO);
 	nouveau_therm_sensor_preinit(therm);
 	return 0;
 }
diff --git a/drivers/gpu/drm/nouveau/core/subdev/therm/fan.c b/drivers/gpu/drm/nouveau/core/subdev/therm/fan.c
index 39f47b950ad1..95f6129eeede 100644
--- a/drivers/gpu/drm/nouveau/core/subdev/therm/fan.c
+++ b/drivers/gpu/drm/nouveau/core/subdev/therm/fan.c
@@ -185,8 +185,11 @@ nouveau_therm_fan_set_defaults(struct nouveau_therm *therm)
 	priv->fan->bios.max_duty = 100;
 	priv->fan->bios.bump_period = 500;
 	priv->fan->bios.slow_down_period = 2000;
+/*XXX: talk to mupuf */
+#if 0
 	priv->fan->bios.linear_min_temp = 40;
 	priv->fan->bios.linear_max_temp = 85;
+#endif
 }
 
 static void
diff --git a/drivers/gpu/drm/nouveau/core/subdev/therm/fantog.c b/drivers/gpu/drm/nouveau/core/subdev/therm/fantog.c
index e601773ee475..f69dab11f720 100644
--- a/drivers/gpu/drm/nouveau/core/subdev/therm/fantog.c
+++ b/drivers/gpu/drm/nouveau/core/subdev/therm/fantog.c
@@ -97,6 +97,13 @@ nouveau_fantog_create(struct nouveau_therm *therm, struct dcb_gpio_func *func)
 {
 	struct nouveau_therm_priv *tpriv = (void *)therm;
 	struct nouveau_fantog_priv *priv;
+	int ret;
+
+	if (therm->pwm_ctrl) {
+		ret = therm->pwm_ctrl(therm, func->line, false);
+		if (ret)
+			return ret;
+	}
 
 	priv = kzalloc(sizeof(*priv), GFP_KERNEL);
 	tpriv->fan = &priv->base;
diff --git a/drivers/gpu/drm/nouveau/core/subdev/therm/ic.c b/drivers/gpu/drm/nouveau/core/subdev/therm/ic.c
index 8b3adec5fbb1..e44ed7b93c6d 100644
--- a/drivers/gpu/drm/nouveau/core/subdev/therm/ic.c
+++ b/drivers/gpu/drm/nouveau/core/subdev/therm/ic.c
@@ -41,7 +41,8 @@ probe_monitoring_device(struct nouveau_i2c_port *i2c,
 	if (!client)
 		return false;
 
-	if (!client->driver || client->driver->detect(client, info)) {
+	if (!client->dev.driver ||
+	    to_i2c_driver(client->dev.driver)->detect(client, info)) {
 		i2c_unregister_device(client);
 		return false;
 	}
@@ -55,28 +56,28 @@ probe_monitoring_device(struct nouveau_i2c_port *i2c,
 	return true;
 }
 
-static struct i2c_board_info
+static struct nouveau_i2c_board_info
 nv_board_infos[] = {
-	{ I2C_BOARD_INFO("w83l785ts", 0x2d) },
-	{ I2C_BOARD_INFO("w83781d", 0x2d) },
-	{ I2C_BOARD_INFO("adt7473", 0x2e) },
-	{ I2C_BOARD_INFO("adt7473", 0x2d) },
-	{ I2C_BOARD_INFO("adt7473", 0x2c) },
-	{ I2C_BOARD_INFO("f75375", 0x2e) },
-	{ I2C_BOARD_INFO("lm99", 0x4c) },
-	{ I2C_BOARD_INFO("lm90", 0x4c) },
-	{ I2C_BOARD_INFO("lm90", 0x4d) },
-	{ I2C_BOARD_INFO("adm1021", 0x18) },
-	{ I2C_BOARD_INFO("adm1021", 0x19) },
-	{ I2C_BOARD_INFO("adm1021", 0x1a) },
-	{ I2C_BOARD_INFO("adm1021", 0x29) },
-	{ I2C_BOARD_INFO("adm1021", 0x2a) },
-	{ I2C_BOARD_INFO("adm1021", 0x2b) },
-	{ I2C_BOARD_INFO("adm1021", 0x4c) },
-	{ I2C_BOARD_INFO("adm1021", 0x4d) },
-	{ I2C_BOARD_INFO("adm1021", 0x4e) },
-	{ I2C_BOARD_INFO("lm63", 0x18) },
-	{ I2C_BOARD_INFO("lm63", 0x4e) },
+	{ { I2C_BOARD_INFO("w83l785ts", 0x2d) }, 0 },
+	{ { I2C_BOARD_INFO("w83781d", 0x2d) }, 0  },
+	{ { I2C_BOARD_INFO("adt7473", 0x2e) }, 20  },
+	{ { I2C_BOARD_INFO("adt7473", 0x2d) }, 20  },
+	{ { I2C_BOARD_INFO("adt7473", 0x2c) }, 20  },
+	{ { I2C_BOARD_INFO("f75375", 0x2e) }, 0  },
+	{ { I2C_BOARD_INFO("lm99", 0x4c) }, 0  },
+	{ { I2C_BOARD_INFO("lm90", 0x4c) }, 0  },
+	{ { I2C_BOARD_INFO("lm90", 0x4d) }, 0  },
+	{ { I2C_BOARD_INFO("adm1021", 0x18) }, 0  },
+	{ { I2C_BOARD_INFO("adm1021", 0x19) }, 0  },
+	{ { I2C_BOARD_INFO("adm1021", 0x1a) }, 0  },
+	{ { I2C_BOARD_INFO("adm1021", 0x29) }, 0  },
+	{ { I2C_BOARD_INFO("adm1021", 0x2a) }, 0  },
+	{ { I2C_BOARD_INFO("adm1021", 0x2b) }, 0  },
+	{ { I2C_BOARD_INFO("adm1021", 0x4c) }, 0  },
+	{ { I2C_BOARD_INFO("adm1021", 0x4d) }, 0  },
+	{ { I2C_BOARD_INFO("adm1021", 0x4e) }, 0  },
+	{ { I2C_BOARD_INFO("lm63", 0x18) }, 0  },
+	{ { I2C_BOARD_INFO("lm63", 0x4e) }, 0  },
 	{ }
 };
 
@@ -89,9 +90,9 @@ nouveau_therm_ic_ctor(struct nouveau_therm *therm)
 	struct nvbios_extdev_func extdev_entry;
 
 	if (!nvbios_extdev_find(bios, NVBIOS_EXTDEV_LM89, &extdev_entry)) {
-		struct i2c_board_info board[] = {
-			{ I2C_BOARD_INFO("lm90", extdev_entry.addr >> 1) },
-			{ }
+		struct nouveau_i2c_board_info board[] = {
+		  { { I2C_BOARD_INFO("lm90", extdev_entry.addr >> 1) }, 0},
+		  { }
 		};
 
 		i2c->identify(i2c, NV_I2C_DEFAULT(0), "monitoring device",
@@ -101,9 +102,9 @@ nouveau_therm_ic_ctor(struct nouveau_therm *therm)
 	}
 
 	if (!nvbios_extdev_find(bios, NVBIOS_EXTDEV_ADT7473, &extdev_entry)) {
-		struct i2c_board_info board[] = {
-			{ I2C_BOARD_INFO("adt7473", extdev_entry.addr >> 1) },
-			{ }
+		struct nouveau_i2c_board_info board[] = {
+		  { { I2C_BOARD_INFO("adt7473", extdev_entry.addr >> 1) }, 20 },
+		  { }
 		};
 
 		i2c->identify(i2c, NV_I2C_DEFAULT(0), "monitoring device",
diff --git a/drivers/gpu/drm/nouveau/core/subdev/therm/nv84.c b/drivers/gpu/drm/nouveau/core/subdev/therm/nv84.c
index 42ba633ccff7..1d15c52fad0c 100644
--- a/drivers/gpu/drm/nouveau/core/subdev/therm/nv84.c
+++ b/drivers/gpu/drm/nouveau/core/subdev/therm/nv84.c
@@ -126,7 +126,7 @@ nv84_therm_intr(struct nouveau_subdev *subdev)
 
 	spin_lock_irqsave(&priv->sensor.alarm_program_lock, flags);
 
-	intr = nv_rd32(therm, 0x20100);
+	intr = nv_rd32(therm, 0x20100) & 0x3ff;
 
 	/* THRS_4: downclock */
 	if (intr & 0x002) {
@@ -209,6 +209,19 @@ nv84_therm_ctor(struct nouveau_object *parent,
 	return nouveau_therm_preinit(&priv->base.base);
 }
 
+int
+nv84_therm_fini(struct nouveau_object *object, bool suspend)
+{
+	/* Disable PTherm IRQs */
+	nv_wr32(object, 0x20000, 0x00000000);
+
+	/* ACK all PTherm IRQs */
+	nv_wr32(object, 0x20100, 0xffffffff);
+	nv_wr32(object, 0x1100, 0x10000); /* PBUS */
+
+	return _nouveau_therm_fini(object, suspend);
+}
+
 struct nouveau_oclass
 nv84_therm_oclass = {
 	.handle = NV_SUBDEV(THERM, 0x84),
@@ -216,6 +229,6 @@ nv84_therm_oclass = {
 		.ctor = nv84_therm_ctor,
 		.dtor = _nouveau_therm_dtor,
 		.init = _nouveau_therm_init,
-		.fini = _nouveau_therm_fini,
+		.fini = nv84_therm_fini,
 	},
 };
diff --git a/drivers/gpu/drm/nouveau/core/subdev/therm/nva3.c b/drivers/gpu/drm/nouveau/core/subdev/therm/nva3.c
index d11a7c400813..3b2c4580098b 100644
--- a/drivers/gpu/drm/nouveau/core/subdev/therm/nva3.c
+++ b/drivers/gpu/drm/nouveau/core/subdev/therm/nva3.c
@@ -94,6 +94,6 @@ nva3_therm_oclass = {
 		.ctor = nva3_therm_ctor,
 		.dtor = _nouveau_therm_dtor,
 		.init = nva3_therm_init,
-		.fini = _nouveau_therm_fini,
+		.fini = nv84_therm_fini,
 	},
 };
diff --git a/drivers/gpu/drm/nouveau/core/subdev/therm/nvd0.c b/drivers/gpu/drm/nouveau/core/subdev/therm/nvd0.c
index 54c28bdc4204..4dd4f81ae873 100644
--- a/drivers/gpu/drm/nouveau/core/subdev/therm/nvd0.c
+++ b/drivers/gpu/drm/nouveau/core/subdev/therm/nvd0.c
@@ -148,6 +148,6 @@ nvd0_therm_oclass = {
 		.ctor = nvd0_therm_ctor,
 		.dtor = _nouveau_therm_dtor,
 		.init = nvd0_therm_init,
-		.fini = _nouveau_therm_fini,
+		.fini = nv84_therm_fini,
 	},
 };
diff --git a/drivers/gpu/drm/nouveau/core/subdev/therm/priv.h b/drivers/gpu/drm/nouveau/core/subdev/therm/priv.h
index dd38529262fb..96f8f95693ce 100644
--- a/drivers/gpu/drm/nouveau/core/subdev/therm/priv.h
+++ b/drivers/gpu/drm/nouveau/core/subdev/therm/priv.h
@@ -76,6 +76,7 @@ struct nouveau_therm_priv {
 	spinlock_t lock;
 	struct nouveau_therm_trip_point *last_trip;
 	int mode;
+	int cstate;
 	int suspend;
 
 	/* bios */
@@ -144,6 +145,7 @@ int nv50_fan_pwm_get(struct nouveau_therm *, int, u32 *, u32 *);
 int nv50_fan_pwm_set(struct nouveau_therm *, int, u32, u32);
 int nv50_fan_pwm_clock(struct nouveau_therm *);
 int nv84_temp_get(struct nouveau_therm *therm);
+int nv84_therm_fini(struct nouveau_object *object, bool suspend);
 
 int nva3_therm_fan_sense(struct nouveau_therm *);
 
diff --git a/drivers/gpu/drm/nouveau/core/subdev/therm/temp.c b/drivers/gpu/drm/nouveau/core/subdev/therm/temp.c
index b80a33011b93..cfde9eb44ad0 100644
--- a/drivers/gpu/drm/nouveau/core/subdev/therm/temp.c
+++ b/drivers/gpu/drm/nouveau/core/subdev/therm/temp.c
@@ -180,8 +180,6 @@ alarm_timer_callback(struct nouveau_alarm *alarm)
 
 	spin_lock_irqsave(&priv->sensor.alarm_program_lock, flags);
 
-	nv_debug(therm, "polling the internal temperature\n");
-
 	nouveau_therm_threshold_hyst_polling(therm, &sensor->thrs_fan_boost,
 					     NOUVEAU_THERM_THRS_FANBOOST);
 
diff --git a/drivers/gpu/drm/nouveau/core/subdev/timer/nv04.c b/drivers/gpu/drm/nouveau/core/subdev/timer/nv04.c
index 57711ecb566c..c0bdd10358d7 100644
--- a/drivers/gpu/drm/nouveau/core/subdev/timer/nv04.c
+++ b/drivers/gpu/drm/nouveau/core/subdev/timer/nv04.c
@@ -119,16 +119,8 @@ nv04_timer_alarm_cancel(struct nouveau_timer *ptimer,
 {
 	struct nv04_timer_priv *priv = (void *)ptimer;
 	unsigned long flags;
-
-	/* avoid deleting an entry while the alarm intr is running */
 	spin_lock_irqsave(&priv->lock, flags);
-
-	/* delete the alarm from the list */
-	list_del(&alarm->head);
-
-	/* reset the head so as list_empty returns 1 */
-	INIT_LIST_HEAD(&alarm->head);
-
+	list_del_init(&alarm->head);
 	spin_unlock_irqrestore(&priv->lock, flags);
 }
 
diff --git a/drivers/gpu/drm/nouveau/core/subdev/volt/base.c b/drivers/gpu/drm/nouveau/core/subdev/volt/base.c
new file mode 100644
index 000000000000..32794a999106
--- /dev/null
+++ b/drivers/gpu/drm/nouveau/core/subdev/volt/base.c
@@ -0,0 +1,198 @@
+/*
+ * Copyright 2013 Red Hat Inc.
+ *
+ * Permission is hereby granted, free of charge, to any person obtaining a
+ * copy of this software and associated documentation files (the "Software"),
+ * to deal in the Software without restriction, including without limitation
+ * the rights to use, copy, modify, merge, publish, distribute, sublicense,
+ * and/or sell copies of the Software, and to permit persons to whom the
+ * Software is furnished to do so, subject to the following conditions:
+ *
+ * The above copyright notice and this permission notice shall be included in
+ * all copies or substantial portions of the Software.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+ * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+ * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT.  IN NO EVENT SHALL
+ * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR
+ * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE,
+ * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR
+ * OTHER DEALINGS IN THE SOFTWARE.
+ *
+ * Authors: Ben Skeggs
+ */
+
+#include <subdev/volt.h>
+
+#include <subdev/bios.h>
+#include <subdev/bios/vmap.h>
+#include <subdev/bios/volt.h>
+
+static int
+nouveau_volt_get(struct nouveau_volt *volt)
+{
+	if (volt->vid_get) {
+		int ret = volt->vid_get(volt), i;
+		if (ret >= 0) {
+			for (i = 0; i < volt->vid_nr; i++) {
+				if (volt->vid[i].vid == ret)
+					return volt->vid[i].uv;
+			}
+			ret = -EINVAL;
+		}
+		return ret;
+	}
+	return -ENODEV;
+}
+
+static int
+nouveau_volt_set(struct nouveau_volt *volt, u32 uv)
+{
+	if (volt->vid_set) {
+		int i, ret = -EINVAL;
+		for (i = 0; i < volt->vid_nr; i++) {
+			if (volt->vid[i].uv == uv) {
+				ret = volt->vid_set(volt, volt->vid[i].vid);
+				nv_debug(volt, "set %duv: %d\n", uv, ret);
+				break;
+			}
+		}
+		return ret;
+	}
+	return -ENODEV;
+}
+
+static int
+nouveau_volt_map(struct nouveau_volt *volt, u8 id)
+{
+	struct nouveau_bios *bios = nouveau_bios(volt);
+	struct nvbios_vmap_entry info;
+	u8  ver, len;
+	u16 vmap;
+
+	vmap = nvbios_vmap_entry_parse(bios, id, &ver, &len, &info);
+	if (vmap) {
+		if (info.link != 0xff) {
+			int ret = nouveau_volt_map(volt, info.link);
+			if (ret < 0)
+				return ret;
+			info.min += ret;
+		}
+		return info.min;
+	}
+
+	return id ? id * 10000 : -ENODEV;
+}
+
+static int
+nouveau_volt_set_id(struct nouveau_volt *volt, u8 id, int condition)
+{
+	int ret = nouveau_volt_map(volt, id);
+	if (ret >= 0) {
+		int prev = nouveau_volt_get(volt);
+		if (!condition || prev < 0 ||
+		    (condition < 0 && ret < prev) ||
+		    (condition > 0 && ret > prev)) {
+			ret = nouveau_volt_set(volt, ret);
+		} else {
+			ret = 0;
+		}
+	}
+	return ret;
+}
+
+int
+_nouveau_volt_init(struct nouveau_object *object)
+{
+	struct nouveau_volt *volt = (void *)object;
+	int ret;
+
+	ret = nouveau_subdev_init(&volt->base);
+	if (ret)
+		return ret;
+
+	ret = volt->get(volt);
+	if (ret < 0) {
+		if (ret != -ENODEV)
+			nv_debug(volt, "current voltage unknown\n");
+		return 0;
+	}
+
+	nv_info(volt, "GPU voltage: %duv\n", ret);
+	return 0;
+}
+
+void
+_nouveau_volt_dtor(struct nouveau_object *object)
+{
+	struct nouveau_volt *volt = (void *)object;
+	nouveau_subdev_destroy(&volt->base);
+}
+
+int
+nouveau_volt_create_(struct nouveau_object *parent,
+		     struct nouveau_object *engine,
+		     struct nouveau_oclass *oclass, int length, void **pobject)
+{
+	struct nouveau_bios *bios = nouveau_bios(parent);
+	struct nouveau_volt *volt;
+	struct nvbios_volt_entry ivid;
+	struct nvbios_volt info;
+	u8  ver, hdr, cnt, len;
+	u16 data;
+	int ret, i;
+
+	ret = nouveau_subdev_create_(parent, engine, oclass, 0, "VOLT",
+				     "voltage", length, pobject);
+	volt = *pobject;
+	if (ret)
+		return ret;
+
+	volt->get = nouveau_volt_get;
+	volt->set = nouveau_volt_set;
+	volt->set_id = nouveau_volt_set_id;
+
+	data = nvbios_volt_parse(bios, &ver, &hdr, &cnt, &len, &info);
+	if (data && info.vidmask && info.base && info.step) {
+		for (i = 0; i < info.vidmask + 1; i++) {
+			if (info.base >= info.min &&
+			    info.base <= info.max) {
+				volt->vid[volt->vid_nr].uv = info.base;
+				volt->vid[volt->vid_nr].vid = i;
+				volt->vid_nr++;
+			}
+			info.base += info.step;
+		}
+		volt->vid_mask = info.vidmask;
+	} else
+	if (data && info.vidmask) {
+		for (i = 0; i < cnt; i++) {
+			data = nvbios_volt_entry_parse(bios, i, &ver, &hdr,
+						      &ivid);
+			if (data) {
+				volt->vid[volt->vid_nr].uv = ivid.voltage;
+				volt->vid[volt->vid_nr].vid = ivid.vid;
+				volt->vid_nr++;
+			}
+		}
+		volt->vid_mask = info.vidmask;
+	}
+
+	if (volt->vid_nr) {
+		for (i = 0; i < volt->vid_nr; i++) {
+			nv_debug(volt, "VID %02x: %duv\n",
+				 volt->vid[i].vid, volt->vid[i].uv);
+		}
+
+		/*XXX: this is an assumption.. there probably exists boards
+		 * out there with i2c-connected voltage controllers too..
+		 */
+		ret = nouveau_voltgpio_init(volt);
+		if (ret == 0) {
+			volt->vid_get = nouveau_voltgpio_get;
+			volt->vid_set = nouveau_voltgpio_set;
+		}
+	}
+
+	return ret;
+}
diff --git a/drivers/gpu/drm/nouveau/core/subdev/volt/gpio.c b/drivers/gpu/drm/nouveau/core/subdev/volt/gpio.c
new file mode 100644
index 000000000000..755fa91bcd09
--- /dev/null
+++ b/drivers/gpu/drm/nouveau/core/subdev/volt/gpio.c
@@ -0,0 +1,96 @@
+/*
+ * Copyright 2013 Red Hat Inc.
+ *
+ * Permission is hereby granted, free of charge, to any person obtaining a
+ * copy of this software and associated documentation files (the "Software"),
+ * to deal in the Software without restriction, including without limitation
+ * the rights to use, copy, modify, merge, publish, distribute, sublicense,
+ * and/or sell copies of the Software, and to permit persons to whom the
+ * Software is furnished to do so, subject to the following conditions:
+ *
+ * The above copyright notice and this permission notice shall be included in
+ * all copies or substantial portions of the Software.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+ * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+ * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT.  IN NO EVENT SHALL
+ * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR
+ * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE,
+ * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR
+ * OTHER DEALINGS IN THE SOFTWARE.
+ *
+ * Authors: Ben Skeggs
+ */
+
+#include <subdev/volt.h>
+#include <subdev/gpio.h>
+#include <subdev/bios/gpio.h>
+
+static const u8 tags[] = {
+	DCB_GPIO_VID0, DCB_GPIO_VID1, DCB_GPIO_VID2, DCB_GPIO_VID3,
+	DCB_GPIO_VID4, DCB_GPIO_VID5, DCB_GPIO_VID6, DCB_GPIO_VID7,
+};
+
+int
+nouveau_voltgpio_get(struct nouveau_volt *volt)
+{
+	struct nouveau_gpio *gpio = nouveau_gpio(volt);
+	u8 vid = 0;
+	int i;
+
+	for (i = 0; i < ARRAY_SIZE(tags); i++) {
+		if (volt->vid_mask & (1 << i)) {
+			int ret = gpio->get(gpio, 0, tags[i], 0xff);
+			if (ret < 0)
+				return ret;
+			vid |= ret << i;
+		}
+	}
+
+	return vid;
+}
+
+int
+nouveau_voltgpio_set(struct nouveau_volt *volt, u8 vid)
+{
+	struct nouveau_gpio *gpio = nouveau_gpio(volt);
+	int i;
+
+	for (i = 0; i < ARRAY_SIZE(tags); i++, vid >>= 1) {
+		if (volt->vid_mask & (1 << i)) {
+			int ret = gpio->set(gpio, 0, tags[i], 0xff, vid & 1);
+			if (ret < 0)
+				return ret;
+		}
+	}
+
+	return 0;
+}
+
+int
+nouveau_voltgpio_init(struct nouveau_volt *volt)
+{
+	struct nouveau_gpio *gpio = nouveau_gpio(volt);
+	struct dcb_gpio_func func;
+	int i;
+
+	/* check we have gpio function info for each vid bit.  on some
+	 * boards (ie. nvs295) the vid mask has more bits than there
+	 * are valid gpio functions... from traces, nvidia appear to
+	 * just touch the existing ones, so let's mask off the invalid
+	 * bits and continue with life
+	 */
+	for (i = 0; i < ARRAY_SIZE(tags); i++) {
+		if (volt->vid_mask & (1 << i)) {
+			int ret = gpio->find(gpio, 0, tags[i], 0xff, &func);
+			if (ret) {
+				if (ret != -ENOENT)
+					return ret;
+				nv_debug(volt, "VID bit %d has no GPIO\n", i);
+				volt->vid_mask &= ~(1 << i);
+			}
+		}
+	}
+
+	return 0;
+}
diff --git a/drivers/gpu/drm/nouveau/core/subdev/volt/nv40.c b/drivers/gpu/drm/nouveau/core/subdev/volt/nv40.c
new file mode 100644
index 000000000000..87d5358376a6
--- /dev/null
+++ b/drivers/gpu/drm/nouveau/core/subdev/volt/nv40.c
@@ -0,0 +1,56 @@
+/*
+ * Copyright 2013 Red Hat Inc.
+ *
+ * Permission is hereby granted, free of charge, to any person obtaining a
+ * copy of this software and associated documentation files (the "Software"),
+ * to deal in the Software without restriction, including without limitation
+ * the rights to use, copy, modify, merge, publish, distribute, sublicense,
+ * and/or sell copies of the Software, and to permit persons to whom the
+ * Software is furnished to do so, subject to the following conditions:
+ *
+ * The above copyright notice and this permission notice shall be included in
+ * all copies or substantial portions of the Software.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+ * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+ * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT.  IN NO EVENT SHALL
+ * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR
+ * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE,
+ * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR
+ * OTHER DEALINGS IN THE SOFTWARE.
+ *
+ * Authors: Ben Skeggs
+ */
+
+#include <subdev/volt.h>
+
+struct nv40_volt_priv {
+	struct nouveau_volt base;
+};
+
+static int
+nv40_volt_ctor(struct nouveau_object *parent, struct nouveau_object *engine,
+	       struct nouveau_oclass *oclass, void *data, u32 size,
+	       struct nouveau_object **pobject)
+{
+	struct nv40_volt_priv *priv;
+	int ret;
+
+	ret = nouveau_volt_create(parent, engine, oclass, &priv);
+	*pobject = nv_object(priv);
+	if (ret)
+		return ret;
+
+	return 0;
+}
+
+struct nouveau_oclass
+nv40_volt_oclass = {
+	.handle = NV_SUBDEV(VOLT, 0x40),
+	.ofuncs = &(struct nouveau_ofuncs) {
+		.ctor = nv40_volt_ctor,
+		.dtor = _nouveau_volt_dtor,
+		.init = _nouveau_volt_init,
+		.fini = _nouveau_volt_fini,
+	},
+};
diff --git a/drivers/gpu/drm/nouveau/dispnv04/Makefile b/drivers/gpu/drm/nouveau/dispnv04/Makefile
index ea3f5b8a0f95..424a489d0f03 100644
--- a/drivers/gpu/drm/nouveau/dispnv04/Makefile
+++ b/drivers/gpu/drm/nouveau/dispnv04/Makefile
@@ -5,6 +5,7 @@ nouveau-y += dispnv04/dac.o
 nouveau-y += dispnv04/dfp.o
 nouveau-y += dispnv04/disp.o
 nouveau-y += dispnv04/hw.o
+nouveau-y += dispnv04/overlay.o
 nouveau-y += dispnv04/tvmodesnv17.o
 nouveau-y += dispnv04/tvnv04.o
 nouveau-y += dispnv04/tvnv17.o
diff --git a/drivers/gpu/drm/nouveau/dispnv04/arb.c b/drivers/gpu/drm/nouveau/dispnv04/arb.c
index 2e70462883e8..2a15b98b4d2b 100644
--- a/drivers/gpu/drm/nouveau/dispnv04/arb.c
+++ b/drivers/gpu/drm/nouveau/dispnv04/arb.c
@@ -210,8 +210,8 @@ nv04_update_arb(struct drm_device *dev, int VClk, int bpp,
 	sim_data.nvclk_khz = NVClk;
 	sim_data.bpp = bpp;
 	sim_data.two_heads = nv_two_heads(dev);
-	if ((dev->pci_device & 0xffff) == 0x01a0 /*CHIPSET_NFORCE*/ ||
-	    (dev->pci_device & 0xffff) == 0x01f0 /*CHIPSET_NFORCE2*/) {
+	if ((dev->pdev->device & 0xffff) == 0x01a0 /*CHIPSET_NFORCE*/ ||
+	    (dev->pdev->device & 0xffff) == 0x01f0 /*CHIPSET_NFORCE2*/) {
 		uint32_t type;
 
 		pci_read_config_dword(pci_get_bus_and_slot(0, 1), 0x7c, &type);
@@ -256,8 +256,8 @@ nouveau_calc_arb(struct drm_device *dev, int vclk, int bpp, int *burst, int *lwm
 
 	if (nv_device(drm->device)->card_type < NV_20)
 		nv04_update_arb(dev, vclk, bpp, burst, lwm);
-	else if ((dev->pci_device & 0xfff0) == 0x0240 /*CHIPSET_C51*/ ||
-		 (dev->pci_device & 0xfff0) == 0x03d0 /*CHIPSET_C512*/) {
+	else if ((dev->pdev->device & 0xfff0) == 0x0240 /*CHIPSET_C51*/ ||
+		 (dev->pdev->device & 0xfff0) == 0x03d0 /*CHIPSET_C512*/) {
 		*burst = 128;
 		*lwm = 0x0480;
 	} else
diff --git a/drivers/gpu/drm/nouveau/dispnv04/crtc.c b/drivers/gpu/drm/nouveau/dispnv04/crtc.c
index d4fbf11360fe..0e3270c3ffd2 100644
--- a/drivers/gpu/drm/nouveau/dispnv04/crtc.c
+++ b/drivers/gpu/drm/nouveau/dispnv04/crtc.c
@@ -326,8 +326,6 @@ nv_crtc_mode_set_vga(struct drm_crtc *crtc, struct drm_display_mode *mode)
 			regp->MiscOutReg = 0x23;	/* +hsync +vsync */
 	}
 
-	regp->MiscOutReg |= (mode->clock_index & 0x03) << 2;
-
 	/*
 	 * Time Sequencer
 	 */
diff --git a/drivers/gpu/drm/nouveau/dispnv04/dfp.c b/drivers/gpu/drm/nouveau/dispnv04/dfp.c
index 93dd23ff0093..936a71c59080 100644
--- a/drivers/gpu/drm/nouveau/dispnv04/dfp.c
+++ b/drivers/gpu/drm/nouveau/dispnv04/dfp.c
@@ -490,10 +490,10 @@ static void nv04_dfp_update_backlight(struct drm_encoder *encoder, int mode)
 	/* BIOS scripts usually take care of the backlight, thanks
 	 * Apple for your consistency.
 	 */
-	if (dev->pci_device == 0x0174 || dev->pci_device == 0x0179 ||
-	    dev->pci_device == 0x0189 || dev->pci_device == 0x0329) {
+	if (dev->pdev->device == 0x0174 || dev->pdev->device == 0x0179 ||
+	    dev->pdev->device == 0x0189 || dev->pdev->device == 0x0329) {
 		if (mode == DRM_MODE_DPMS_ON) {
-			nv_mask(device, NV_PBUS_DEBUG_DUALHEAD_CTL, 0, 1 << 31);
+			nv_mask(device, NV_PBUS_DEBUG_DUALHEAD_CTL, 1 << 31, 1 << 31);
 			nv_mask(device, NV_PCRTC_GPIO_EXT, 3, 1);
 		} else {
 			nv_mask(device, NV_PBUS_DEBUG_DUALHEAD_CTL, 1 << 31, 0);
@@ -625,13 +625,15 @@ static void nv04_tmds_slave_init(struct drm_encoder *encoder)
 	struct nouveau_drm *drm = nouveau_drm(dev);
 	struct nouveau_i2c *i2c = nouveau_i2c(drm->device);
 	struct nouveau_i2c_port *port = i2c->find(i2c, 2);
-	struct i2c_board_info info[] = {
+	struct nouveau_i2c_board_info info[] = {
 		{
-			.type = "sil164",
-			.addr = (dcb->tmdsconf.slave_addr == 0x7 ? 0x3a : 0x38),
-			.platform_data = &(struct sil164_encoder_params) {
-				SIL164_INPUT_EDGE_RISING
-			}
+		    {
+		        .type = "sil164",
+		        .addr = (dcb->tmdsconf.slave_addr == 0x7 ? 0x3a : 0x38),
+		        .platform_data = &(struct sil164_encoder_params) {
+		            SIL164_INPUT_EDGE_RISING
+		         }
+		    }, 0
 		},
 		{ }
 	};
@@ -646,7 +648,7 @@ static void nv04_tmds_slave_init(struct drm_encoder *encoder)
 		return;
 
 	drm_i2c_encoder_init(dev, to_encoder_slave(encoder),
-			     &port->adapter, &info[type]);
+			     &port->adapter, &info[type].dev);
 }
 
 static const struct drm_encoder_helper_funcs nv04_lvds_helper_funcs = {
diff --git a/drivers/gpu/drm/nouveau/dispnv04/disp.c b/drivers/gpu/drm/nouveau/dispnv04/disp.c
index 4908d3fd0486..b13ff0fc42de 100644
--- a/drivers/gpu/drm/nouveau/dispnv04/disp.c
+++ b/drivers/gpu/drm/nouveau/dispnv04/disp.c
@@ -140,6 +140,8 @@ nv04_display_create(struct drm_device *dev)
 		func->save(encoder);
 	}
 
+	nouveau_overlay_init(dev);
+
 	return 0;
 }
 
diff --git a/drivers/gpu/drm/nouveau/dispnv04/disp.h b/drivers/gpu/drm/nouveau/dispnv04/disp.h
index 9928187f0a7d..56a28db04000 100644
--- a/drivers/gpu/drm/nouveau/dispnv04/disp.h
+++ b/drivers/gpu/drm/nouveau/dispnv04/disp.h
@@ -123,11 +123,14 @@ int nv04_tv_create(struct drm_connector *, struct dcb_output *);
 /* nv17_tv.c */
 int nv17_tv_create(struct drm_connector *, struct dcb_output *);
 
+/* overlay.c */
+void nouveau_overlay_init(struct drm_device *dev);
+
 static inline bool
 nv_two_heads(struct drm_device *dev)
 {
 	struct nouveau_drm *drm = nouveau_drm(dev);
-	const int impl = dev->pci_device & 0x0ff0;
+	const int impl = dev->pdev->device & 0x0ff0;
 
 	if (nv_device(drm->device)->card_type >= NV_10 && impl != 0x0100 &&
 	    impl != 0x0150 && impl != 0x01a0 && impl != 0x0200)
@@ -139,14 +142,14 @@ nv_two_heads(struct drm_device *dev)
 static inline bool
 nv_gf4_disp_arch(struct drm_device *dev)
 {
-	return nv_two_heads(dev) && (dev->pci_device & 0x0ff0) != 0x0110;
+	return nv_two_heads(dev) && (dev->pdev->device & 0x0ff0) != 0x0110;
 }
 
 static inline bool
 nv_two_reg_pll(struct drm_device *dev)
 {
 	struct nouveau_drm *drm = nouveau_drm(dev);
-	const int impl = dev->pci_device & 0x0ff0;
+	const int impl = dev->pdev->device & 0x0ff0;
 
 	if (impl == 0x0310 || impl == 0x0340 || nv_device(drm->device)->card_type >= NV_40)
 		return true;
diff --git a/drivers/gpu/drm/nouveau/dispnv04/hw.c b/drivers/gpu/drm/nouveau/dispnv04/hw.c
index 973056b86207..aca76af115b3 100644
--- a/drivers/gpu/drm/nouveau/dispnv04/hw.c
+++ b/drivers/gpu/drm/nouveau/dispnv04/hw.c
@@ -27,6 +27,7 @@
 #include "hw.h"
 
 #include <subdev/bios/pll.h>
+#include <subdev/fb.h>
 #include <subdev/clock.h>
 #include <subdev/timer.h>
 
@@ -220,7 +221,7 @@ nouveau_hw_get_clock(struct drm_device *dev, enum nvbios_pll_type plltype)
 	int ret;
 
 	if (plltype == PLL_MEMORY &&
-	    (dev->pci_device & 0x0ff0) == CHIPSET_NFORCE) {
+	    (dev->pdev->device & 0x0ff0) == CHIPSET_NFORCE) {
 		uint32_t mpllP;
 
 		pci_read_config_dword(pci_get_bus_and_slot(0, 3), 0x6c, &mpllP);
@@ -230,7 +231,7 @@ nouveau_hw_get_clock(struct drm_device *dev, enum nvbios_pll_type plltype)
 		return 400000 / mpllP;
 	} else
 	if (plltype == PLL_MEMORY &&
-	    (dev->pci_device & 0xff0) == CHIPSET_NFORCE2) {
+	    (dev->pdev->device & 0xff0) == CHIPSET_NFORCE2) {
 		uint32_t clock;
 
 		pci_read_config_dword(pci_get_bus_and_slot(0, 5), 0x4c, &clock);
@@ -664,6 +665,7 @@ nv_load_state_ext(struct drm_device *dev, int head,
 	struct nouveau_drm *drm = nouveau_drm(dev);
 	struct nouveau_device *device = nv_device(drm->device);
 	struct nouveau_timer *ptimer = nouveau_timer(device);
+	struct nouveau_fb *pfb = nouveau_fb(device);
 	struct nv04_crtc_reg *regp = &state->crtc_reg[head];
 	uint32_t reg900;
 	int i;
@@ -680,10 +682,10 @@ nv_load_state_ext(struct drm_device *dev, int head,
 		nv_wr32(device, NV_PVIDEO_INTR_EN, 0);
 		nv_wr32(device, NV_PVIDEO_OFFSET_BUFF(0), 0);
 		nv_wr32(device, NV_PVIDEO_OFFSET_BUFF(1), 0);
-		nv_wr32(device, NV_PVIDEO_LIMIT(0), 0); //drm->fb_available_size - 1);
-		nv_wr32(device, NV_PVIDEO_LIMIT(1), 0); //drm->fb_available_size - 1);
-		nv_wr32(device, NV_PVIDEO_UVPLANE_LIMIT(0), 0); //drm->fb_available_size - 1);
-		nv_wr32(device, NV_PVIDEO_UVPLANE_LIMIT(1), 0); //drm->fb_available_size - 1);
+		nv_wr32(device, NV_PVIDEO_LIMIT(0), pfb->ram->size - 1);
+		nv_wr32(device, NV_PVIDEO_LIMIT(1), pfb->ram->size - 1);
+		nv_wr32(device, NV_PVIDEO_UVPLANE_LIMIT(0), pfb->ram->size - 1);
+		nv_wr32(device, NV_PVIDEO_UVPLANE_LIMIT(1), pfb->ram->size - 1);
 		nv_wr32(device, NV_PBUS_POWERCTRL_2, 0);
 
 		NVWriteCRTC(dev, head, NV_PCRTC_CURSOR_CONFIG, regp->cursor_cfg);
@@ -740,7 +742,7 @@ nv_load_state_ext(struct drm_device *dev, int head,
 	}
 	/* NV11 and NV20 stop at 0x52. */
 	if (nv_gf4_disp_arch(dev)) {
-		if (nv_device(drm->device)->card_type == NV_10) {
+		if (nv_device(drm->device)->card_type < NV_20) {
 			/* Not waiting for vertical retrace before modifying
 			   CRE_53/CRE_54 causes lockups. */
 			nouveau_timer_wait_eq(ptimer, 650000000, NV_PRMCIO_INP0__COLOR, 0x8, 0x8);
diff --git a/drivers/gpu/drm/nouveau/dispnv04/overlay.c b/drivers/gpu/drm/nouveau/dispnv04/overlay.c
new file mode 100644
index 000000000000..3618ac6b6316
--- /dev/null
+++ b/drivers/gpu/drm/nouveau/dispnv04/overlay.c
@@ -0,0 +1,320 @@
+/*
+ * Copyright 2013 Ilia Mirkin
+ *
+ * Permission is hereby granted, free of charge, to any person obtaining a
+ * copy of this software and associated documentation files (the "Software"),
+ * to deal in the Software without restriction, including without limitation
+ * the rights to use, copy, modify, merge, publish, distribute, sublicense,
+ * and/or sell copies of the Software, and to permit persons to whom the
+ * Software is furnished to do so, subject to the following conditions:
+ *
+ * The above copyright notice and this permission notice shall be included in
+ * all copies or substantial portions of the Software.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+ * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+ * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT.  IN NO EVENT SHALL
+ * THE AUTHORS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY,
+ * WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF
+ * OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
+ * SOFTWARE.
+ *
+ * Implementation based on the pre-KMS implementation in xf86-video-nouveau,
+ * written by Arthur Huillet.
+ */
+
+#include <drm/drmP.h>
+#include <drm/drm_crtc.h>
+#include <drm/drm_fourcc.h>
+
+#include "nouveau_drm.h"
+
+#include "nouveau_bo.h"
+#include "nouveau_connector.h"
+#include "nouveau_display.h"
+#include "nvreg.h"
+
+
+struct nouveau_plane {
+	struct drm_plane base;
+	bool flip;
+	struct nouveau_bo *cur;
+
+	struct {
+		struct drm_property *colorkey;
+		struct drm_property *contrast;
+		struct drm_property *brightness;
+		struct drm_property *hue;
+		struct drm_property *saturation;
+		struct drm_property *iturbt_709;
+	} props;
+
+	int colorkey;
+	int contrast;
+	int brightness;
+	int hue;
+	int saturation;
+	int iturbt_709;
+};
+
+static uint32_t formats[] = {
+	DRM_FORMAT_NV12,
+	DRM_FORMAT_UYVY,
+};
+
+/* Sine can be approximated with
+ * http://en.wikipedia.org/wiki/Bhaskara_I's_sine_approximation_formula
+ * sin(x degrees) ~= 4 x (180 - x) / (40500 - x (180 - x) )
+ * Note that this only works for the range [0, 180].
+ * Also note that sin(x) == -sin(x - 180)
+ */
+static inline int
+sin_mul(int degrees, int factor)
+{
+	if (degrees > 180) {
+		degrees -= 180;
+		factor *= -1;
+	}
+	return factor * 4 * degrees * (180 - degrees) /
+		(40500 - degrees * (180 - degrees));
+}
+
+/* cos(x) = sin(x + 90) */
+static inline int
+cos_mul(int degrees, int factor)
+{
+	return sin_mul((degrees + 90) % 360, factor);
+}
+
+static int
+nv10_update_plane(struct drm_plane *plane, struct drm_crtc *crtc,
+		  struct drm_framebuffer *fb, int crtc_x, int crtc_y,
+		  unsigned int crtc_w, unsigned int crtc_h,
+		  uint32_t src_x, uint32_t src_y,
+		  uint32_t src_w, uint32_t src_h)
+{
+	struct nouveau_device *dev = nouveau_dev(plane->dev);
+	struct nouveau_plane *nv_plane = (struct nouveau_plane *)plane;
+	struct nouveau_framebuffer *nv_fb = nouveau_framebuffer(fb);
+	struct nouveau_crtc *nv_crtc = nouveau_crtc(crtc);
+	struct nouveau_bo *cur = nv_plane->cur;
+	bool flip = nv_plane->flip;
+	int format = ALIGN(src_w * 4, 0x100);
+	int soff = NV_PCRTC0_SIZE * nv_crtc->index;
+	int soff2 = NV_PCRTC0_SIZE * !nv_crtc->index;
+	int ret;
+
+	if (format > 0xffff)
+		return -EINVAL;
+
+	ret = nouveau_bo_pin(nv_fb->nvbo, TTM_PL_FLAG_VRAM);
+	if (ret)
+		return ret;
+
+	nv_plane->cur = nv_fb->nvbo;
+
+	/* Source parameters given in 16.16 fixed point, ignore fractional. */
+	src_x = src_x >> 16;
+	src_y = src_y >> 16;
+	src_w = src_w >> 16;
+	src_h = src_h >> 16;
+
+	nv_mask(dev, NV_PCRTC_ENGINE_CTRL + soff, NV_CRTC_FSEL_OVERLAY, NV_CRTC_FSEL_OVERLAY);
+	nv_mask(dev, NV_PCRTC_ENGINE_CTRL + soff2, NV_CRTC_FSEL_OVERLAY, 0);
+
+	nv_wr32(dev, NV_PVIDEO_BASE(flip), 0);
+	nv_wr32(dev, NV_PVIDEO_OFFSET_BUFF(flip), nv_fb->nvbo->bo.offset);
+	nv_wr32(dev, NV_PVIDEO_SIZE_IN(flip), src_h << 16 | src_w);
+	nv_wr32(dev, NV_PVIDEO_POINT_IN(flip), src_y << 16 | src_x);
+	nv_wr32(dev, NV_PVIDEO_DS_DX(flip), (src_w << 20) / crtc_w);
+	nv_wr32(dev, NV_PVIDEO_DT_DY(flip), (src_h << 20) / crtc_h);
+	nv_wr32(dev, NV_PVIDEO_POINT_OUT(flip), crtc_y << 16 | crtc_x);
+	nv_wr32(dev, NV_PVIDEO_SIZE_OUT(flip), crtc_h << 16 | crtc_w);
+
+	if (fb->pixel_format == DRM_FORMAT_NV12) {
+		format |= NV_PVIDEO_FORMAT_COLOR_LE_CR8YB8CB8YA8;
+		format |= NV_PVIDEO_FORMAT_PLANAR;
+	}
+	if (nv_plane->iturbt_709)
+		format |= NV_PVIDEO_FORMAT_MATRIX_ITURBT709;
+	if (nv_plane->colorkey & (1 << 24))
+		format |= NV_PVIDEO_FORMAT_DISPLAY_COLOR_KEY;
+
+	if (fb->pixel_format == DRM_FORMAT_NV12) {
+		nv_wr32(dev, NV_PVIDEO_UVPLANE_BASE(flip), 0);
+		nv_wr32(dev, NV_PVIDEO_UVPLANE_OFFSET_BUFF(flip),
+			nv_fb->nvbo->bo.offset + fb->offsets[1]);
+	}
+	nv_wr32(dev, NV_PVIDEO_FORMAT(flip), format);
+	nv_wr32(dev, NV_PVIDEO_STOP, 0);
+	/* TODO: wait for vblank? */
+	nv_wr32(dev, NV_PVIDEO_BUFFER, flip ? 0x10 : 0x1);
+	nv_plane->flip = !flip;
+
+	if (cur)
+		nouveau_bo_unpin(cur);
+
+	return 0;
+}
+
+static int
+nv10_disable_plane(struct drm_plane *plane)
+{
+	struct nouveau_device *dev = nouveau_dev(plane->dev);
+	struct nouveau_plane *nv_plane = (struct nouveau_plane *)plane;
+
+	nv_wr32(dev, NV_PVIDEO_STOP, 1);
+	if (nv_plane->cur) {
+		nouveau_bo_unpin(nv_plane->cur);
+		nv_plane->cur = NULL;
+	}
+
+	return 0;
+}
+
+static void
+nv10_destroy_plane(struct drm_plane *plane)
+{
+	nv10_disable_plane(plane);
+	drm_plane_cleanup(plane);
+	kfree(plane);
+}
+
+static void
+nv10_set_params(struct nouveau_plane *plane)
+{
+	struct nouveau_device *dev = nouveau_dev(plane->base.dev);
+	u32 luma = (plane->brightness - 512) << 16 | plane->contrast;
+	u32 chroma = ((sin_mul(plane->hue, plane->saturation) & 0xffff) << 16) |
+		(cos_mul(plane->hue, plane->saturation) & 0xffff);
+	u32 format = 0;
+
+	nv_wr32(dev, NV_PVIDEO_LUMINANCE(0), luma);
+	nv_wr32(dev, NV_PVIDEO_LUMINANCE(1), luma);
+	nv_wr32(dev, NV_PVIDEO_CHROMINANCE(0), chroma);
+	nv_wr32(dev, NV_PVIDEO_CHROMINANCE(1), chroma);
+	nv_wr32(dev, NV_PVIDEO_COLOR_KEY, plane->colorkey & 0xffffff);
+
+	if (plane->cur) {
+		if (plane->iturbt_709)
+			format |= NV_PVIDEO_FORMAT_MATRIX_ITURBT709;
+		if (plane->colorkey & (1 << 24))
+			format |= NV_PVIDEO_FORMAT_DISPLAY_COLOR_KEY;
+		nv_mask(dev, NV_PVIDEO_FORMAT(plane->flip),
+			NV_PVIDEO_FORMAT_MATRIX_ITURBT709 |
+			NV_PVIDEO_FORMAT_DISPLAY_COLOR_KEY,
+			format);
+	}
+}
+
+static int
+nv10_set_property(struct drm_plane *plane,
+		  struct drm_property *property,
+		  uint64_t value)
+{
+	struct nouveau_plane *nv_plane = (struct nouveau_plane *)plane;
+
+	if (property == nv_plane->props.colorkey)
+		nv_plane->colorkey = value;
+	else if (property == nv_plane->props.contrast)
+		nv_plane->contrast = value;
+	else if (property == nv_plane->props.brightness)
+		nv_plane->brightness = value;
+	else if (property == nv_plane->props.hue)
+		nv_plane->hue = value;
+	else if (property == nv_plane->props.saturation)
+		nv_plane->saturation = value;
+	else if (property == nv_plane->props.iturbt_709)
+		nv_plane->iturbt_709 = value;
+	else
+		return -EINVAL;
+
+	nv10_set_params(nv_plane);
+	return 0;
+}
+
+static const struct drm_plane_funcs nv10_plane_funcs = {
+	.update_plane = nv10_update_plane,
+	.disable_plane = nv10_disable_plane,
+	.set_property = nv10_set_property,
+	.destroy = nv10_destroy_plane,
+};
+
+static void
+nv10_overlay_init(struct drm_device *device)
+{
+	struct nouveau_device *dev = nouveau_dev(device);
+	struct nouveau_plane *plane = kzalloc(sizeof(struct nouveau_plane), GFP_KERNEL);
+	int ret;
+
+	if (!plane)
+		return;
+
+	ret = drm_plane_init(device, &plane->base, 3 /* both crtc's */,
+			     &nv10_plane_funcs,
+			     formats, ARRAY_SIZE(formats), false);
+	if (ret)
+		goto err;
+
+	/* Set up the plane properties */
+	plane->props.colorkey = drm_property_create_range(
+			device, 0, "colorkey", 0, 0x01ffffff);
+	plane->props.contrast = drm_property_create_range(
+			device, 0, "contrast", 0, 8192 - 1);
+	plane->props.brightness = drm_property_create_range(
+			device, 0, "brightness", 0, 1024);
+	plane->props.hue = drm_property_create_range(
+			device, 0, "hue", 0, 359);
+	plane->props.saturation = drm_property_create_range(
+			device, 0, "saturation", 0, 8192 - 1);
+	plane->props.iturbt_709 = drm_property_create_range(
+			device, 0, "iturbt_709", 0, 1);
+	if (!plane->props.colorkey ||
+	    !plane->props.contrast ||
+	    !plane->props.brightness ||
+	    !plane->props.hue ||
+	    !plane->props.saturation ||
+	    !plane->props.iturbt_709)
+		goto cleanup;
+
+	plane->colorkey = 0;
+	drm_object_attach_property(&plane->base.base,
+				   plane->props.colorkey, plane->colorkey);
+
+	plane->contrast = 0x1000;
+	drm_object_attach_property(&plane->base.base,
+				   plane->props.contrast, plane->contrast);
+
+	plane->brightness = 512;
+	drm_object_attach_property(&plane->base.base,
+				   plane->props.brightness, plane->brightness);
+
+	plane->hue = 0;
+	drm_object_attach_property(&plane->base.base,
+				   plane->props.hue, plane->hue);
+
+	plane->saturation = 0x1000;
+	drm_object_attach_property(&plane->base.base,
+				   plane->props.saturation, plane->saturation);
+
+	plane->iturbt_709 = 0;
+	drm_object_attach_property(&plane->base.base,
+				   plane->props.iturbt_709, plane->iturbt_709);
+
+	nv10_set_params(plane);
+	nv_wr32(dev, NV_PVIDEO_STOP, 1);
+	return;
+cleanup:
+	drm_plane_cleanup(&plane->base);
+err:
+	kfree(plane);
+	nv_error(dev, "Failed to create plane\n");
+}
+
+void
+nouveau_overlay_init(struct drm_device *device)
+{
+	struct nouveau_device *dev = nouveau_dev(device);
+	if (dev->chipset >= 0x10 && dev->chipset <= 0x40)
+		nv10_overlay_init(device);
+}
diff --git a/drivers/gpu/drm/nouveau/dispnv04/tvnv04.c b/drivers/gpu/drm/nouveau/dispnv04/tvnv04.c
index bf13db4e8631..cc4b208ce546 100644
--- a/drivers/gpu/drm/nouveau/dispnv04/tvnv04.c
+++ b/drivers/gpu/drm/nouveau/dispnv04/tvnv04.c
@@ -37,15 +37,18 @@
 
 #include <subdev/i2c.h>
 
-static struct i2c_board_info nv04_tv_encoder_info[] = {
+static struct nouveau_i2c_board_info nv04_tv_encoder_info[] = {
 	{
-		I2C_BOARD_INFO("ch7006", 0x75),
-		.platform_data = &(struct ch7006_encoder_params) {
-			CH7006_FORMAT_RGB24m12I, CH7006_CLOCK_MASTER,
-			0, 0, 0,
-			CH7006_SYNC_SLAVE, CH7006_SYNC_SEPARATED,
-			CH7006_POUT_3_3V, CH7006_ACTIVE_HSYNC
-		}
+		{
+			I2C_BOARD_INFO("ch7006", 0x75),
+			.platform_data = &(struct ch7006_encoder_params) {
+				CH7006_FORMAT_RGB24m12I, CH7006_CLOCK_MASTER,
+				0, 0, 0,
+				CH7006_SYNC_SLAVE, CH7006_SYNC_SEPARATED,
+				CH7006_POUT_3_3V, CH7006_ACTIVE_HSYNC
+			}
+		},
+		0
 	},
 	{ }
 };
@@ -229,7 +232,8 @@ nv04_tv_create(struct drm_connector *connector, struct dcb_output *entry)
 
 	/* Run the slave-specific initialization */
 	ret = drm_i2c_encoder_init(dev, to_encoder_slave(encoder),
-				   &port->adapter, &nv04_tv_encoder_info[type]);
+				   &port->adapter,
+				   &nv04_tv_encoder_info[type].dev);
 	if (ret < 0)
 		goto fail_cleanup;
 
diff --git a/drivers/gpu/drm/nouveau/nouveau_abi16.c b/drivers/gpu/drm/nouveau/nouveau_abi16.c
index 8f467e7bfd19..6828d81ed7b9 100644
--- a/drivers/gpu/drm/nouveau/nouveau_abi16.c
+++ b/drivers/gpu/drm/nouveau/nouveau_abi16.c
@@ -87,6 +87,7 @@ nouveau_abi16_swclass(struct nouveau_drm *drm)
 	case NV_04:
 		return 0x006e;
 	case NV_10:
+	case NV_11:
 	case NV_20:
 	case NV_30:
 	case NV_40:
@@ -130,7 +131,7 @@ nouveau_abi16_chan_fini(struct nouveau_abi16 *abi16,
 	if (chan->ntfy) {
 		nouveau_bo_vma_del(chan->ntfy, &chan->ntfy_vma);
 		nouveau_bo_unpin(chan->ntfy);
-		drm_gem_object_unreference_unlocked(chan->ntfy->gem);
+		drm_gem_object_unreference_unlocked(&chan->ntfy->gem);
 	}
 
 	if (chan->heap.block_size)
@@ -178,10 +179,10 @@ nouveau_abi16_ioctl_getparam(ABI16_IOCTL_ARGS)
 		getparam->value = device->chipset;
 		break;
 	case NOUVEAU_GETPARAM_PCI_VENDOR:
-		getparam->value = dev->pci_vendor;
+		getparam->value = dev->pdev->vendor;
 		break;
 	case NOUVEAU_GETPARAM_PCI_DEVICE:
-		getparam->value = dev->pci_device;
+		getparam->value = dev->pdev->device;
 		break;
 	case NOUVEAU_GETPARAM_BUS_TYPE:
 		if (drm_pci_device_is_agp(dev))
@@ -297,7 +298,7 @@ nouveau_abi16_ioctl_channel_alloc(ABI16_IOCTL_ARGS)
 	else
 		init->pushbuf_domains = NOUVEAU_GEM_DOMAIN_GART;
 
-	if (device->card_type < NV_C0) {
+	if (device->card_type < NV_10) {
 		init->subchan[0].handle = 0x00000000;
 		init->subchan[0].grclass = 0x0000;
 		init->subchan[1].handle = NvSw;
@@ -320,7 +321,7 @@ nouveau_abi16_ioctl_channel_alloc(ABI16_IOCTL_ARGS)
 			goto done;
 	}
 
-	ret = drm_gem_handle_create(file_priv, chan->ntfy->gem,
+	ret = drm_gem_handle_create(file_priv, &chan->ntfy->gem,
 				    &init->notifier_handle);
 	if (ret)
 		goto done;
diff --git a/drivers/gpu/drm/nouveau/nouveau_acpi.c b/drivers/gpu/drm/nouveau/nouveau_acpi.c
index d9eb65dae1f7..95c740454049 100644
--- a/drivers/gpu/drm/nouveau/nouveau_acpi.c
+++ b/drivers/gpu/drm/nouveau/nouveau_acpi.c
@@ -314,6 +314,16 @@ static bool nouveau_dsm_detect(void)
 			has_optimus = 1;
 	}
 
+	while ((pdev = pci_get_class(PCI_CLASS_DISPLAY_3D << 8, pdev)) != NULL) {
+		vga_count++;
+
+		retval = nouveau_dsm_pci_probe(pdev);
+		if (retval & NOUVEAU_DSM_HAS_MUX)
+			has_dsm |= 1;
+		if (retval & NOUVEAU_DSM_HAS_OPT)
+			has_optimus = 1;
+	}
+
 	/* find the optimus DSM or the old v1 DSM */
 	if (has_optimus == 1) {
 		acpi_get_name(nouveau_dsm_priv.dhandle, ACPI_FULL_PATHNAME,
diff --git a/drivers/gpu/drm/nouveau/nouveau_agp.c b/drivers/gpu/drm/nouveau/nouveau_agp.c
index 6e7a55f93a85..2953c4e91e1a 100644
--- a/drivers/gpu/drm/nouveau/nouveau_agp.c
+++ b/drivers/gpu/drm/nouveau/nouveau_agp.c
@@ -11,10 +11,28 @@ MODULE_PARM_DESC(agpmode, "AGP mode (0 to disable AGP)");
 static int nouveau_agpmode = -1;
 module_param_named(agpmode, nouveau_agpmode, int, 0400);
 
+struct nouveau_agpmode_quirk {
+	u16 hostbridge_vendor;
+	u16 hostbridge_device;
+	u16 chip_vendor;
+	u16 chip_device;
+	int mode;
+};
+
+static struct nouveau_agpmode_quirk nouveau_agpmode_quirk_list[] = {
+	/* VIA Apollo PRO133x / GeForce FX 5600 Ultra, max agpmode 2, fdo #20341 */
+	{ PCI_VENDOR_ID_VIA, 0x0691, PCI_VENDOR_ID_NVIDIA, 0x0311, 2 },
+
+	{},
+};
+
 static unsigned long
-get_agp_mode(struct nouveau_drm *drm, unsigned long mode)
+get_agp_mode(struct nouveau_drm *drm, const struct drm_agp_info *info)
 {
 	struct nouveau_device *device = nv_device(drm->device);
+	struct nouveau_agpmode_quirk *quirk = nouveau_agpmode_quirk_list;
+	int agpmode = nouveau_agpmode;
+	unsigned long mode = info->mode;
 
 	/*
 	 * FW seems to be broken on nv18, it makes the card lock up
@@ -24,11 +42,27 @@ get_agp_mode(struct nouveau_drm *drm, unsigned long mode)
 		mode &= ~PCI_AGP_COMMAND_FW;
 
 	/*
+	 * Go through the quirks list and adjust the agpmode accordingly.
+	 */
+	while (agpmode == -1 && quirk->hostbridge_vendor) {
+		if (info->id_vendor == quirk->hostbridge_vendor &&
+		    info->id_device == quirk->hostbridge_device &&
+		    device->pdev->vendor == quirk->chip_vendor &&
+		    device->pdev->device == quirk->chip_device) {
+			agpmode = quirk->mode;
+			nv_info(device, "Forcing agp mode to %dX. Use agpmode to override.\n",
+				agpmode);
+			break;
+		}
+		++quirk;
+	}
+
+	/*
 	 * AGP mode set in the command line.
 	 */
-	if (nouveau_agpmode > 0) {
+	if (agpmode > 0) {
 		bool agpv3 = mode & 0x8;
-		int rate = agpv3 ? nouveau_agpmode / 4 : nouveau_agpmode;
+		int rate = agpv3 ? agpmode / 4 : agpmode;
 
 		mode = (mode & ~0x7) | (rate & 0x7);
 	}
@@ -90,7 +124,7 @@ nouveau_agp_reset(struct nouveau_drm *drm)
 		if (ret)
 			return;
 
-		mode.mode  = get_agp_mode(drm, info.mode);
+		mode.mode  = get_agp_mode(drm, &info);
 		mode.mode &= ~PCI_AGP_COMMAND_FW;
 
 		ret = drm_agp_enable(dev, mode);
@@ -139,7 +173,7 @@ nouveau_agp_init(struct nouveau_drm *drm)
 	}
 
 	/* see agp.h for the AGPSTAT_* modes available */
-	mode.mode = get_agp_mode(drm, info.mode);
+	mode.mode = get_agp_mode(drm, &info);
 
 	ret = drm_agp_enable(dev, mode);
 	if (ret) {
diff --git a/drivers/gpu/drm/nouveau/nouveau_backlight.c b/drivers/gpu/drm/nouveau/nouveau_backlight.c
index 2ffad2176b7f..630f6e84fc01 100644
--- a/drivers/gpu/drm/nouveau/nouveau_backlight.c
+++ b/drivers/gpu/drm/nouveau/nouveau_backlight.c
@@ -82,7 +82,7 @@ nv40_backlight_init(struct drm_connector *connector)
 	memset(&props, 0, sizeof(struct backlight_properties));
 	props.type = BACKLIGHT_RAW;
 	props.max_brightness = 31;
-	bd = backlight_device_register("nv_backlight", &connector->kdev, drm,
+	bd = backlight_device_register("nv_backlight", connector->kdev, drm,
 				       &nv40_bl_ops, &props);
 	if (IS_ERR(bd))
 		return PTR_ERR(bd);
@@ -204,7 +204,7 @@ nv50_backlight_init(struct drm_connector *connector)
 	memset(&props, 0, sizeof(struct backlight_properties));
 	props.type = BACKLIGHT_RAW;
 	props.max_brightness = 100;
-	bd = backlight_device_register("nv_backlight", &connector->kdev,
+	bd = backlight_device_register("nv_backlight", connector->kdev,
 				       nv_encoder, ops, &props);
 	if (IS_ERR(bd))
 		return PTR_ERR(bd);
diff --git a/drivers/gpu/drm/nouveau/nouveau_bios.c b/drivers/gpu/drm/nouveau/nouveau_bios.c
index 3e7287675ecf..4c3feaaa1037 100644
--- a/drivers/gpu/drm/nouveau/nouveau_bios.c
+++ b/drivers/gpu/drm/nouveau/nouveau_bios.c
@@ -127,8 +127,8 @@ static int call_lvds_manufacturer_script(struct drm_device *dev, struct dcb_outp
 #ifdef __powerpc__
 	/* Powerbook specific quirks */
 	if (script == LVDS_RESET &&
-	    (dev->pci_device == 0x0179 || dev->pci_device == 0x0189 ||
-	     dev->pci_device == 0x0329))
+	    (dev->pdev->device == 0x0179 || dev->pdev->device == 0x0189 ||
+	     dev->pdev->device == 0x0329))
 		nv_write_tmds(dev, dcbent->or, 0, 0x02, 0x72);
 #endif
 
diff --git a/drivers/gpu/drm/nouveau/nouveau_bo.c b/drivers/gpu/drm/nouveau/nouveau_bo.c
index 755c38d06271..c0fde6b9393c 100644
--- a/drivers/gpu/drm/nouveau/nouveau_bo.c
+++ b/drivers/gpu/drm/nouveau/nouveau_bo.c
@@ -98,12 +98,7 @@ nv10_bo_put_tile_region(struct drm_device *dev, struct nouveau_drm_tile *tile,
 
 	if (tile) {
 		spin_lock(&drm->tile.lock);
-		if (fence) {
-			/* Mark it as pending. */
-			tile->fence = fence;
-			nouveau_fence_ref(fence);
-		}
-
+		tile->fence = nouveau_fence_ref(fence);
 		tile->used = false;
 		spin_unlock(&drm->tile.lock);
 	}
@@ -146,7 +141,7 @@ nouveau_bo_del_ttm(struct ttm_buffer_object *bo)
 	struct drm_device *dev = drm->dev;
 	struct nouveau_bo *nvbo = nouveau_bo(bo);
 
-	if (unlikely(nvbo->gem))
+	if (unlikely(nvbo->gem.filp))
 		DRM_ERROR("bo %p still attached to GEM object\n", bo);
 	WARN_ON(nvbo->pin_refcnt > 0);
 	nv10_bo_put_tile_region(dev, nvbo->tile, NULL);
@@ -269,7 +264,8 @@ set_placement_range(struct nouveau_bo *nvbo, uint32_t type)
 	struct nouveau_fb *pfb = nouveau_fb(drm->device);
 	u32 vram_pages = pfb->ram->size >> PAGE_SHIFT;
 
-	if (nv_device(drm->device)->card_type == NV_10 &&
+	if ((nv_device(drm->device)->card_type == NV_10 ||
+	     nv_device(drm->device)->card_type == NV_11) &&
 	    nvbo->tile_mode && (type & TTM_PL_FLAG_VRAM) &&
 	    nvbo->bo.mem.num_pages < vram_pages / 4) {
 		/*
@@ -982,7 +978,7 @@ nouveau_bo_move_m2mf(struct ttm_buffer_object *bo, int evict, bool intr,
 		     bool no_wait_gpu, struct ttm_mem_reg *new_mem)
 {
 	struct nouveau_drm *drm = nouveau_bdev(bo->bdev);
-	struct nouveau_channel *chan = chan = drm->ttm.chan;
+	struct nouveau_channel *chan = drm->ttm.chan;
 	struct nouveau_bo *nvbo = nouveau_bo(bo);
 	struct ttm_mem_reg *old_mem = &bo->mem;
 	int ret;
@@ -1267,7 +1263,7 @@ nouveau_bo_verify_access(struct ttm_buffer_object *bo, struct file *filp)
 {
 	struct nouveau_bo *nvbo = nouveau_bo(bo);
 
-	return drm_vma_node_verify_access(&nvbo->gem->vma_node, filp);
+	return drm_vma_node_verify_access(&nvbo->gem.vma_node, filp);
 }
 
 static int
@@ -1461,14 +1457,12 @@ nouveau_ttm_tt_unpopulate(struct ttm_tt *ttm)
 void
 nouveau_bo_fence(struct nouveau_bo *nvbo, struct nouveau_fence *fence)
 {
+	struct nouveau_fence *new_fence = nouveau_fence_ref(fence);
 	struct nouveau_fence *old_fence = NULL;
 
-	if (likely(fence))
-		nouveau_fence_ref(fence);
-
 	spin_lock(&nvbo->bo.bdev->fence_lock);
 	old_fence = nvbo->bo.sync_obj;
-	nvbo->bo.sync_obj = fence;
+	nvbo->bo.sync_obj = new_fence;
 	spin_unlock(&nvbo->bo.bdev->fence_lock);
 
 	nouveau_fence_unref(&old_fence);
@@ -1551,7 +1545,8 @@ nouveau_bo_vma_add(struct nouveau_bo *nvbo, struct nouveau_vm *vm,
 
 	if (nvbo->bo.mem.mem_type == TTM_PL_VRAM)
 		nouveau_vm_map(vma, nvbo->bo.mem.mm_node);
-	else if (nvbo->bo.mem.mem_type == TTM_PL_TT) {
+	else if (nvbo->bo.mem.mem_type == TTM_PL_TT &&
+		 nvbo->page_shift == vma->vm->vmm->spg_shift) {
 		if (node->sg)
 			nouveau_vm_map_sg_table(vma, 0, size, node);
 		else
diff --git a/drivers/gpu/drm/nouveau/nouveau_bo.h b/drivers/gpu/drm/nouveau/nouveau_bo.h
index 653dbbbd4fa1..ff17c1f432fc 100644
--- a/drivers/gpu/drm/nouveau/nouveau_bo.h
+++ b/drivers/gpu/drm/nouveau/nouveau_bo.h
@@ -27,7 +27,10 @@ struct nouveau_bo {
 	u32 tile_flags;
 	struct nouveau_drm_tile *tile;
 
-	struct drm_gem_object *gem;
+	/* Only valid if allocated via nouveau_gem_new() and iff you hold a
+	 * gem reference to it! For debugging, use gem.filp != NULL to test
+	 * whether it is valid. */
+	struct drm_gem_object gem;
 
 	/* protect by the ttm reservation lock */
 	int pin_refcnt;
diff --git a/drivers/gpu/drm/nouveau/nouveau_chan.c b/drivers/gpu/drm/nouveau/nouveau_chan.c
index e84f4c32331b..cc5152be2cf1 100644
--- a/drivers/gpu/drm/nouveau/nouveau_chan.c
+++ b/drivers/gpu/drm/nouveau/nouveau_chan.c
@@ -346,22 +346,17 @@ nouveau_channel_init(struct nouveau_channel *chan, u32 vram, u32 gart)
 	for (i = 0; i < NOUVEAU_DMA_SKIPS; i++)
 		OUT_RING(chan, 0x00000000);
 
-	/* allocate software object class (used for fences on <= nv05, and
-	 * to signal flip completion), bind it to a subchannel.
-	 */
-	if ((device->card_type < NV_E0) || gart /* nve0: want_nvsw */) {
+	/* allocate software object class (used for fences on <= nv05) */
+	if (device->card_type < NV_10) {
 		ret = nouveau_object_new(nv_object(client), chan->handle,
-					 NvSw, nouveau_abi16_swclass(chan->drm),
-					 NULL, 0, &object);
+					 NvSw, 0x006e, NULL, 0, &object);
 		if (ret)
 			return ret;
 
 		swch = (void *)object->parent;
 		swch->flip = nouveau_flip_complete;
 		swch->flip_data = chan;
-	}
 
-	if (device->card_type < NV_C0) {
 		ret = RING_SPACE(chan, 2);
 		if (ret)
 			return ret;
diff --git a/drivers/gpu/drm/nouveau/nouveau_connector.c b/drivers/gpu/drm/nouveau/nouveau_connector.c
index c5b36f9e9a10..1674882d60d5 100644
--- a/drivers/gpu/drm/nouveau/nouveau_connector.c
+++ b/drivers/gpu/drm/nouveau/nouveau_connector.c
@@ -100,6 +100,7 @@ static void
 nouveau_connector_destroy(struct drm_connector *connector)
 {
 	struct nouveau_connector *nv_connector = nouveau_connector(connector);
+	nouveau_event_ref(NULL, &nv_connector->hpd_func);
 	kfree(nv_connector->edid);
 	drm_sysfs_connector_remove(connector);
 	drm_connector_cleanup(connector);
@@ -214,9 +215,10 @@ nouveau_connector_set_encoder(struct drm_connector *connector,
 	} else {
 		connector->doublescan_allowed = true;
 		if (nv_device(drm->device)->card_type == NV_20 ||
-		   (nv_device(drm->device)->card_type == NV_10 &&
-		    (dev->pci_device & 0x0ff0) != 0x0100 &&
-		    (dev->pci_device & 0x0ff0) != 0x0150))
+		    ((nv_device(drm->device)->card_type == NV_10 ||
+		      nv_device(drm->device)->card_type == NV_11) &&
+		     (dev->pdev->device & 0x0ff0) != 0x0100 &&
+		     (dev->pdev->device & 0x0ff0) != 0x0150))
 			/* HW is broken */
 			connector->interlace_allowed = false;
 		else
@@ -932,10 +934,9 @@ nouveau_connector_hotplug_work(struct work_struct *work)
 }
 
 static int
-nouveau_connector_hotplug(struct nouveau_eventh *event, int index)
+nouveau_connector_hotplug(void *data, int index)
 {
-	struct nouveau_connector *nv_connector =
-		container_of(event, struct nouveau_connector, hpd_func);
+	struct nouveau_connector *nv_connector = data;
 	schedule_work(&nv_connector->hpd_work);
 	return NVKM_EVENT_KEEP;
 }
@@ -1007,10 +1008,16 @@ nouveau_connector_create(struct drm_device *dev, int index)
 
 		ret = gpio->find(gpio, 0, hpd[ffs((entry & 0x07033000) >> 12)],
 				 DCB_GPIO_UNUSED, &nv_connector->hpd);
-		nv_connector->hpd_func.func = nouveau_connector_hotplug;
 		if (ret)
 			nv_connector->hpd.func = DCB_GPIO_UNUSED;
 
+		if (nv_connector->hpd.func != DCB_GPIO_UNUSED) {
+			nouveau_event_new(gpio->events, nv_connector->hpd.line,
+					  nouveau_connector_hotplug,
+					  nv_connector,
+					 &nv_connector->hpd_func);
+		}
+
 		nv_connector->type = nv_connector->dcb[0];
 		if (drm_conntype_from_dcb(nv_connector->type) ==
 					  DRM_MODE_CONNECTOR_Unknown) {
diff --git a/drivers/gpu/drm/nouveau/nouveau_connector.h b/drivers/gpu/drm/nouveau/nouveau_connector.h
index 6e399aad491a..264a778f473b 100644
--- a/drivers/gpu/drm/nouveau/nouveau_connector.h
+++ b/drivers/gpu/drm/nouveau/nouveau_connector.h
@@ -69,7 +69,7 @@ struct nouveau_connector {
 
 	struct dcb_gpio_func hpd;
 	struct work_struct hpd_work;
-	struct nouveau_eventh hpd_func;
+	struct nouveau_eventh *hpd_func;
 
 	int dithering_mode;
 	int dithering_depth;
@@ -107,7 +107,4 @@ nouveau_crtc_connector_get(struct nouveau_crtc *nv_crtc)
 struct drm_connector *
 nouveau_connector_create(struct drm_device *, int index);
 
-int
-nouveau_connector_bpp(struct drm_connector *);
-
 #endif /* __NOUVEAU_CONNECTOR_H__ */
diff --git a/drivers/gpu/drm/nouveau/nouveau_display.c b/drivers/gpu/drm/nouveau/nouveau_display.c
index 7848590f5568..7809d92183c4 100644
--- a/drivers/gpu/drm/nouveau/nouveau_display.c
+++ b/drivers/gpu/drm/nouveau/nouveau_display.c
@@ -26,7 +26,6 @@
 
 #include <drm/drmP.h>
 #include <drm/drm_crtc_helper.h>
-#include <drm/ttm/ttm_execbuf_util.h>
 
 #include "nouveau_fbcon.h"
 #include "dispnv04/hw.h"
@@ -38,19 +37,92 @@
 
 #include "nouveau_fence.h"
 
-#include <subdev/bios/gpio.h>
-#include <subdev/gpio.h>
 #include <engine/disp.h>
 
 #include <core/class.h>
 
+static int
+nouveau_display_vblank_handler(void *data, int head)
+{
+	struct nouveau_drm *drm = data;
+	drm_handle_vblank(drm->dev, head);
+	return NVKM_EVENT_KEEP;
+}
+
+int
+nouveau_display_vblank_enable(struct drm_device *dev, int head)
+{
+	struct nouveau_display *disp = nouveau_display(dev);
+	if (disp) {
+		nouveau_event_get(disp->vblank[head]);
+		return 0;
+	}
+	return -EIO;
+}
+
+void
+nouveau_display_vblank_disable(struct drm_device *dev, int head)
+{
+	struct nouveau_display *disp = nouveau_display(dev);
+	if (disp)
+		nouveau_event_put(disp->vblank[head]);
+}
+
+static void
+nouveau_display_vblank_fini(struct drm_device *dev)
+{
+	struct nouveau_display *disp = nouveau_display(dev);
+	int i;
+
+	if (disp->vblank) {
+		for (i = 0; i < dev->mode_config.num_crtc; i++)
+			nouveau_event_ref(NULL, &disp->vblank[i]);
+		kfree(disp->vblank);
+		disp->vblank = NULL;
+	}
+
+	drm_vblank_cleanup(dev);
+}
+
+static int
+nouveau_display_vblank_init(struct drm_device *dev)
+{
+	struct nouveau_display *disp = nouveau_display(dev);
+	struct nouveau_drm *drm = nouveau_drm(dev);
+	struct nouveau_disp *pdisp = nouveau_disp(drm->device);
+	int ret, i;
+
+	disp->vblank = kzalloc(dev->mode_config.num_crtc *
+			       sizeof(*disp->vblank), GFP_KERNEL);
+	if (!disp->vblank)
+		return -ENOMEM;
+
+	for (i = 0; i < dev->mode_config.num_crtc; i++) {
+		ret = nouveau_event_new(pdisp->vblank, i,
+					nouveau_display_vblank_handler,
+					drm, &disp->vblank[i]);
+		if (ret) {
+			nouveau_display_vblank_fini(dev);
+			return ret;
+		}
+	}
+
+	ret = drm_vblank_init(dev, dev->mode_config.num_crtc);
+	if (ret) {
+		nouveau_display_vblank_fini(dev);
+		return ret;
+	}
+
+	return 0;
+}
+
 static void
 nouveau_user_framebuffer_destroy(struct drm_framebuffer *drm_fb)
 {
 	struct nouveau_framebuffer *fb = nouveau_framebuffer(drm_fb);
 
 	if (fb->nvbo)
-		drm_gem_object_unreference_unlocked(fb->nvbo->gem);
+		drm_gem_object_unreference_unlocked(&fb->nvbo->gem);
 
 	drm_framebuffer_cleanup(drm_fb);
 	kfree(fb);
@@ -63,7 +135,7 @@ nouveau_user_framebuffer_create_handle(struct drm_framebuffer *drm_fb,
 {
 	struct nouveau_framebuffer *fb = nouveau_framebuffer(drm_fb);
 
-	return drm_gem_handle_create(file_priv, fb->nvbo->gem, handle);
+	return drm_gem_handle_create(file_priv, &fb->nvbo->gem, handle);
 }
 
 static const struct drm_framebuffer_funcs nouveau_framebuffer_funcs = {
@@ -227,9 +299,7 @@ static struct nouveau_drm_prop_enum_list dither_depth[] = {
 int
 nouveau_display_init(struct drm_device *dev)
 {
-	struct nouveau_drm *drm = nouveau_drm(dev);
 	struct nouveau_display *disp = nouveau_display(dev);
-	struct nouveau_gpio *gpio = nouveau_gpio(drm->device);
 	struct drm_connector *connector;
 	int ret;
 
@@ -243,10 +313,7 @@ nouveau_display_init(struct drm_device *dev)
 	/* enable hotplug interrupts */
 	list_for_each_entry(connector, &dev->mode_config.connector_list, head) {
 		struct nouveau_connector *conn = nouveau_connector(connector);
-		if (gpio && conn->hpd.func != DCB_GPIO_UNUSED) {
-			nouveau_event_get(gpio->events, conn->hpd.line,
-					 &conn->hpd_func);
-		}
+		if (conn->hpd_func) nouveau_event_get(conn->hpd_func);
 	}
 
 	return ret;
@@ -255,18 +322,13 @@ nouveau_display_init(struct drm_device *dev)
 void
 nouveau_display_fini(struct drm_device *dev)
 {
-	struct nouveau_drm *drm = nouveau_drm(dev);
 	struct nouveau_display *disp = nouveau_display(dev);
-	struct nouveau_gpio *gpio = nouveau_gpio(drm->device);
 	struct drm_connector *connector;
 
 	/* disable hotplug interrupts */
 	list_for_each_entry(connector, &dev->mode_config.connector_list, head) {
 		struct nouveau_connector *conn = nouveau_connector(connector);
-		if (gpio && conn->hpd.func != DCB_GPIO_UNUSED) {
-			nouveau_event_put(gpio->events, conn->hpd.line,
-					 &conn->hpd_func);
-		}
+		if (conn->hpd_func) nouveau_event_put(conn->hpd_func);
 	}
 
 	drm_kms_helper_poll_disable(dev);
@@ -336,6 +398,11 @@ nouveau_display_create(struct drm_device *dev)
 	dev->mode_config.preferred_depth = 24;
 	dev->mode_config.prefer_shadow = 1;
 
+	if (nv_device(drm->device)->chipset < 0x11)
+		dev->mode_config.async_page_flip = false;
+	else
+		dev->mode_config.async_page_flip = true;
+
 	drm_kms_helper_poll_init(dev);
 	drm_kms_helper_poll_disable(dev);
 
@@ -352,7 +419,7 @@ nouveau_display_create(struct drm_device *dev)
 		goto disp_create_err;
 
 	if (dev->mode_config.num_crtc) {
-		ret = drm_vblank_init(dev, dev->mode_config.num_crtc);
+		ret = nouveau_display_vblank_init(dev);
 		if (ret)
 			goto vblank_err;
 	}
@@ -374,7 +441,7 @@ nouveau_display_destroy(struct drm_device *dev)
 	struct nouveau_display *disp = nouveau_display(dev);
 
 	nouveau_backlight_exit(dev);
-	drm_vblank_cleanup(dev);
+	nouveau_display_vblank_fini(dev);
 
 	drm_kms_helper_poll_fini(dev);
 	drm_mode_config_cleanup(dev);
@@ -394,7 +461,7 @@ nouveau_display_suspend(struct drm_device *dev)
 
 	nouveau_display_fini(dev);
 
-	NV_SUSPEND(drm, "unpinning framebuffer(s)...\n");
+	NV_INFO(drm, "unpinning framebuffer(s)...\n");
 	list_for_each_entry(crtc, &dev->mode_config.crtc_list, head) {
 		struct nouveau_framebuffer *nouveau_fb;
 
@@ -492,19 +559,15 @@ nouveau_page_flip_emit(struct nouveau_channel *chan,
 		goto fail;
 
 	/* Emit the pageflip */
-	ret = RING_SPACE(chan, 3);
+	ret = RING_SPACE(chan, 2);
 	if (ret)
 		goto fail;
 
-	if (nv_device(drm->device)->card_type < NV_C0) {
+	if (nv_device(drm->device)->card_type < NV_C0)
 		BEGIN_NV04(chan, NvSubSw, NV_SW_PAGE_FLIP, 1);
-		OUT_RING  (chan, 0x00000000);
-		OUT_RING  (chan, 0x00000000);
-	} else {
-		BEGIN_NVC0(chan, 0, NV10_SUBCHAN_REF_CNT, 1);
-		OUT_RING  (chan, 0);
-		BEGIN_IMC0(chan, 0, NVSW_SUBCHAN_PAGE_FLIP, 0x0000);
-	}
+	else
+		BEGIN_NVC0(chan, FermiSw, NV_SW_PAGE_FLIP, 1);
+	OUT_RING  (chan, 0x00000000);
 	FIRE_RING (chan);
 
 	ret = nouveau_fence_new(chan, false, pfence);
@@ -521,22 +584,16 @@ fail:
 
 int
 nouveau_crtc_page_flip(struct drm_crtc *crtc, struct drm_framebuffer *fb,
-		       struct drm_pending_vblank_event *event,
-		       uint32_t page_flip_flags)
+		       struct drm_pending_vblank_event *event, u32 flags)
 {
+	const int swap_interval = (flags & DRM_MODE_PAGE_FLIP_ASYNC) ? 0 : 1;
 	struct drm_device *dev = crtc->dev;
 	struct nouveau_drm *drm = nouveau_drm(dev);
 	struct nouveau_bo *old_bo = nouveau_framebuffer(crtc->fb)->nvbo;
 	struct nouveau_bo *new_bo = nouveau_framebuffer(fb)->nvbo;
 	struct nouveau_page_flip_state *s;
-	struct nouveau_channel *chan = NULL;
+	struct nouveau_channel *chan = drm->channel;
 	struct nouveau_fence *fence;
-	struct ttm_validate_buffer resv[2] = {
-		{ .bo = &old_bo->bo },
-		{ .bo = &new_bo->bo },
-	};
-	struct ww_acquire_ctx ticket;
-	LIST_HEAD(res);
 	int ret;
 
 	if (!drm->channel)
@@ -546,26 +603,22 @@ nouveau_crtc_page_flip(struct drm_crtc *crtc, struct drm_framebuffer *fb,
 	if (!s)
 		return -ENOMEM;
 
-	/* Choose the channel the flip will be handled in */
-	spin_lock(&old_bo->bo.bdev->fence_lock);
-	fence = new_bo->bo.sync_obj;
-	if (fence)
-		chan = fence->channel;
-	if (!chan)
-		chan = drm->channel;
-	spin_unlock(&old_bo->bo.bdev->fence_lock);
+	/* synchronise rendering channel with the kernel's channel */
+	spin_lock(&new_bo->bo.bdev->fence_lock);
+	fence = nouveau_fence_ref(new_bo->bo.sync_obj);
+	spin_unlock(&new_bo->bo.bdev->fence_lock);
+	ret = nouveau_fence_sync(fence, chan);
+	if (ret)
+		return ret;
 
 	if (new_bo != old_bo) {
 		ret = nouveau_bo_pin(new_bo, TTM_PL_FLAG_VRAM);
 		if (ret)
 			goto fail_free;
-
-		list_add(&resv[1].head, &res);
 	}
-	list_add(&resv[0].head, &res);
 
 	mutex_lock(&chan->cli->mutex);
-	ret = ttm_eu_reserve_buffers(&ticket, &res);
+	ret = ttm_bo_reserve(&old_bo->bo, true, false, false, NULL);
 	if (ret)
 		goto fail_unpin;
 
@@ -577,12 +630,29 @@ nouveau_crtc_page_flip(struct drm_crtc *crtc, struct drm_framebuffer *fb,
 
 	/* Emit a page flip */
 	if (nv_device(drm->device)->card_type >= NV_50) {
-		ret = nv50_display_flip_next(crtc, fb, chan, 0);
+		ret = nv50_display_flip_next(crtc, fb, chan, swap_interval);
 		if (ret)
 			goto fail_unreserve;
 	} else {
 		struct nv04_display *dispnv04 = nv04_display(dev);
-		nouveau_bo_ref(new_bo, &dispnv04->image[nouveau_crtc(crtc)->index]);
+		int head = nouveau_crtc(crtc)->index;
+
+		if (swap_interval) {
+			ret = RING_SPACE(chan, 8);
+			if (ret)
+				goto fail_unreserve;
+
+			BEGIN_NV04(chan, NvSubImageBlit, 0x012c, 1);
+			OUT_RING  (chan, 0);
+			BEGIN_NV04(chan, NvSubImageBlit, 0x0134, 1);
+			OUT_RING  (chan, head);
+			BEGIN_NV04(chan, NvSubImageBlit, 0x0100, 1);
+			OUT_RING  (chan, 0);
+			BEGIN_NV04(chan, NvSubImageBlit, 0x0130, 1);
+			OUT_RING  (chan, 0);
+		}
+
+		nouveau_bo_ref(new_bo, &dispnv04->image[head]);
 	}
 
 	ret = nouveau_page_flip_emit(chan, old_bo, new_bo, s, &fence);
@@ -593,14 +663,15 @@ nouveau_crtc_page_flip(struct drm_crtc *crtc, struct drm_framebuffer *fb,
 	/* Update the crtc struct and cleanup */
 	crtc->fb = fb;
 
-	ttm_eu_fence_buffer_objects(&ticket, &res, fence);
+	nouveau_bo_fence(old_bo, fence);
+	ttm_bo_unreserve(&old_bo->bo);
 	if (old_bo != new_bo)
 		nouveau_bo_unpin(old_bo);
 	nouveau_fence_unref(&fence);
 	return 0;
 
 fail_unreserve:
-	ttm_eu_backoff_reservation(&ticket, &res);
+	ttm_bo_unreserve(&old_bo->bo);
 fail_unpin:
 	mutex_unlock(&chan->cli->mutex);
 	if (old_bo != new_bo)
@@ -674,8 +745,8 @@ nouveau_display_dumb_create(struct drm_file *file_priv, struct drm_device *dev,
 	if (ret)
 		return ret;
 
-	ret = drm_gem_handle_create(file_priv, bo->gem, &args->handle);
-	drm_gem_object_unreference_unlocked(bo->gem);
+	ret = drm_gem_handle_create(file_priv, &bo->gem, &args->handle);
+	drm_gem_object_unreference_unlocked(&bo->gem);
 	return ret;
 }
 
@@ -688,7 +759,7 @@ nouveau_display_dumb_map_offset(struct drm_file *file_priv,
 
 	gem = drm_gem_object_lookup(dev, file_priv, handle);
 	if (gem) {
-		struct nouveau_bo *bo = gem->driver_private;
+		struct nouveau_bo *bo = nouveau_gem_object(gem);
 		*poffset = drm_vma_node_offset_addr(&bo->bo.vma_node);
 		drm_gem_object_unreference_unlocked(gem);
 		return 0;
diff --git a/drivers/gpu/drm/nouveau/nouveau_display.h b/drivers/gpu/drm/nouveau/nouveau_display.h
index 025c66f8e0ed..8bc8bab90e8d 100644
--- a/drivers/gpu/drm/nouveau/nouveau_display.h
+++ b/drivers/gpu/drm/nouveau/nouveau_display.h
@@ -36,6 +36,8 @@ struct nouveau_display {
 	int  (*init)(struct drm_device *);
 	void (*fini)(struct drm_device *);
 
+	struct nouveau_eventh **vblank;
+
 	struct drm_property *dithering_mode;
 	struct drm_property *dithering_depth;
 	struct drm_property *underscan_property;
@@ -59,6 +61,8 @@ void nouveau_display_fini(struct drm_device *dev);
 int  nouveau_display_suspend(struct drm_device *dev);
 void nouveau_display_repin(struct drm_device *dev);
 void nouveau_display_resume(struct drm_device *dev);
+int  nouveau_display_vblank_enable(struct drm_device *, int);
+void nouveau_display_vblank_disable(struct drm_device *, int);
 
 int  nouveau_crtc_page_flip(struct drm_crtc *crtc, struct drm_framebuffer *fb,
 			    struct drm_pending_vblank_event *event,
diff --git a/drivers/gpu/drm/nouveau/nouveau_dma.h b/drivers/gpu/drm/nouveau/nouveau_dma.h
index 690d5930ce32..984004d66a6d 100644
--- a/drivers/gpu/drm/nouveau/nouveau_dma.h
+++ b/drivers/gpu/drm/nouveau/nouveau_dma.h
@@ -51,9 +51,11 @@ enum {
 	NvSubCtxSurf2D  = 0,
 	NvSubSw		= 1,
 	NvSubImageBlit  = 2,
-	NvSub2D		= 3,
 	NvSubGdiRect    = 3,
-	NvSubCopy	= 4,
+
+	NvSub2D		= 3, /* DO NOT CHANGE - hardcoded for kepler gr fifo */
+	NvSubCopy	= 4, /* DO NOT CHANGE - hardcoded for kepler gr fifo */
+	FermiSw		= 5, /* DO NOT CHANGE (well.. 6/7 will work...) */
 };
 
 /* Object handles. */
@@ -194,7 +196,6 @@ WIND_RING(struct nouveau_channel *chan)
 #define NV84_SUBCHAN_UEVENT                                          0x00000020
 #define NV84_SUBCHAN_WRCACHE_FLUSH                                   0x00000024
 #define NV10_SUBCHAN_REF_CNT                                         0x00000050
-#define NVSW_SUBCHAN_PAGE_FLIP                                       0x00000054
 #define NV11_SUBCHAN_DMA_SEMAPHORE                                   0x00000060
 #define NV11_SUBCHAN_SEMAPHORE_OFFSET                                0x00000064
 #define NV11_SUBCHAN_SEMAPHORE_ACQUIRE                               0x00000068
diff --git a/drivers/gpu/drm/nouveau/nouveau_drm.c b/drivers/gpu/drm/nouveau/nouveau_drm.c
index e893c5362402..7a3759f1c41a 100644
--- a/drivers/gpu/drm/nouveau/nouveau_drm.c
+++ b/drivers/gpu/drm/nouveau/nouveau_drm.c
@@ -37,6 +37,7 @@
 #include <engine/device.h>
 #include <engine/disp.h>
 #include <engine/fifo.h>
+#include <engine/software.h>
 
 #include <subdev/vm.h>
 
@@ -46,7 +47,8 @@
 #include "nouveau_gem.h"
 #include "nouveau_agp.h"
 #include "nouveau_vga.h"
-#include "nouveau_pm.h"
+#include "nouveau_sysfs.h"
+#include "nouveau_hwmon.h"
 #include "nouveau_acpi.h"
 #include "nouveau_bios.h"
 #include "nouveau_ioctl.h"
@@ -78,41 +80,6 @@ module_param_named(runpm, nouveau_runtime_pm, int, 0400);
 
 static struct drm_driver driver;
 
-static int
-nouveau_drm_vblank_handler(struct nouveau_eventh *event, int head)
-{
-	struct nouveau_drm *drm =
-		container_of(event, struct nouveau_drm, vblank[head]);
-	drm_handle_vblank(drm->dev, head);
-	return NVKM_EVENT_KEEP;
-}
-
-static int
-nouveau_drm_vblank_enable(struct drm_device *dev, int head)
-{
-	struct nouveau_drm *drm = nouveau_drm(dev);
-	struct nouveau_disp *pdisp = nouveau_disp(drm->device);
-
-	if (WARN_ON_ONCE(head > ARRAY_SIZE(drm->vblank)))
-		return -EIO;
-	WARN_ON_ONCE(drm->vblank[head].func);
-	drm->vblank[head].func = nouveau_drm_vblank_handler;
-	nouveau_event_get(pdisp->vblank, head, &drm->vblank[head]);
-	return 0;
-}
-
-static void
-nouveau_drm_vblank_disable(struct drm_device *dev, int head)
-{
-	struct nouveau_drm *drm = nouveau_drm(dev);
-	struct nouveau_disp *pdisp = nouveau_disp(drm->device);
-	if (drm->vblank[head].func)
-		nouveau_event_put(pdisp->vblank, head, &drm->vblank[head]);
-	else
-		WARN_ON_ONCE(1);
-	drm->vblank[head].func = NULL;
-}
-
 static u64
 nouveau_name(struct pci_dev *pdev)
 {
@@ -177,7 +144,8 @@ nouveau_accel_init(struct nouveau_drm *drm)
 
 	/* initialise synchronisation routines */
 	if      (device->card_type < NV_10) ret = nv04_fence_create(drm);
-	else if (device->chipset   <  0x17) ret = nv10_fence_create(drm);
+	else if (device->card_type < NV_11 ||
+		 device->chipset   <  0x17) ret = nv10_fence_create(drm);
 	else if (device->card_type < NV_50) ret = nv17_fence_create(drm);
 	else if (device->chipset   <  0x84) ret = nv50_fence_create(drm);
 	else if (device->card_type < NV_C0) ret = nv84_fence_create(drm);
@@ -224,6 +192,32 @@ nouveau_accel_init(struct nouveau_drm *drm)
 		return;
 	}
 
+	ret = nouveau_object_new(nv_object(drm), NVDRM_CHAN, NVDRM_NVSW,
+				 nouveau_abi16_swclass(drm), NULL, 0, &object);
+	if (ret == 0) {
+		struct nouveau_software_chan *swch = (void *)object->parent;
+		ret = RING_SPACE(drm->channel, 2);
+		if (ret == 0) {
+			if (device->card_type < NV_C0) {
+				BEGIN_NV04(drm->channel, NvSubSw, 0, 1);
+				OUT_RING  (drm->channel, NVDRM_NVSW);
+			} else
+			if (device->card_type < NV_E0) {
+				BEGIN_NVC0(drm->channel, FermiSw, 0, 1);
+				OUT_RING  (drm->channel, 0x001f0000);
+			}
+		}
+		swch = (void *)object->parent;
+		swch->flip = nouveau_flip_complete;
+		swch->flip_data = drm->channel;
+	}
+
+	if (ret) {
+		NV_ERROR(drm, "failed to allocate software object, %d\n", ret);
+		nouveau_accel_fini(drm);
+		return;
+	}
+
 	if (device->card_type < NV_C0) {
 		ret = nouveau_gpuobj_new(drm->device, NULL, 32, 0, 0,
 					&drm->notify);
@@ -418,8 +412,8 @@ nouveau_drm_load(struct drm_device *dev, unsigned long flags)
 			goto fail_dispinit;
 	}
 
-	nouveau_pm_init(dev);
-
+	nouveau_sysfs_init(dev);
+	nouveau_hwmon_init(dev);
 	nouveau_accel_init(drm);
 	nouveau_fbcon_init(dev);
 
@@ -455,8 +449,8 @@ nouveau_drm_unload(struct drm_device *dev)
 	pm_runtime_get_sync(dev->dev);
 	nouveau_fbcon_fini(dev);
 	nouveau_accel_fini(drm);
-
-	nouveau_pm_fini(dev);
+	nouveau_hwmon_fini(dev);
+	nouveau_sysfs_fini(dev);
 
 	if (dev->mode_config.num_crtc)
 		nouveau_display_fini(dev);
@@ -496,16 +490,16 @@ nouveau_do_suspend(struct drm_device *dev)
 	int ret;
 
 	if (dev->mode_config.num_crtc) {
-		NV_SUSPEND(drm, "suspending display...\n");
+		NV_INFO(drm, "suspending display...\n");
 		ret = nouveau_display_suspend(dev);
 		if (ret)
 			return ret;
 	}
 
-	NV_SUSPEND(drm, "evicting buffers...\n");
+	NV_INFO(drm, "evicting buffers...\n");
 	ttm_bo_evict_mm(&drm->ttm.bdev, TTM_PL_VRAM);
 
-	NV_SUSPEND(drm, "waiting for kernel channels to go idle...\n");
+	NV_INFO(drm, "waiting for kernel channels to go idle...\n");
 	if (drm->cechan) {
 		ret = nouveau_channel_idle(drm->cechan);
 		if (ret)
@@ -518,7 +512,7 @@ nouveau_do_suspend(struct drm_device *dev)
 			return ret;
 	}
 
-	NV_SUSPEND(drm, "suspending client object trees...\n");
+	NV_INFO(drm, "suspending client object trees...\n");
 	if (drm->fence && nouveau_fence(drm)->suspend) {
 		if (!nouveau_fence(drm)->suspend(drm))
 			return -ENOMEM;
@@ -530,7 +524,7 @@ nouveau_do_suspend(struct drm_device *dev)
 			goto fail_client;
 	}
 
-	NV_SUSPEND(drm, "suspending kernel object tree...\n");
+	NV_INFO(drm, "suspending kernel object tree...\n");
 	ret = nouveau_client_fini(&drm->client.base, true);
 	if (ret)
 		goto fail_client;
@@ -544,7 +538,7 @@ fail_client:
 	}
 
 	if (dev->mode_config.num_crtc) {
-		NV_SUSPEND(drm, "resuming display...\n");
+		NV_INFO(drm, "resuming display...\n");
 		nouveau_display_resume(dev);
 	}
 	return ret;
@@ -563,7 +557,6 @@ int nouveau_pmops_suspend(struct device *dev)
 	if (drm_dev->mode_config.num_crtc)
 		nouveau_fbcon_set_suspend(drm_dev, 1);
 
-	nv_suspend_set_printk_level(NV_DBG_INFO);
 	ret = nouveau_do_suspend(drm_dev);
 	if (ret)
 		return ret;
@@ -571,8 +564,6 @@ int nouveau_pmops_suspend(struct device *dev)
 	pci_save_state(pdev);
 	pci_disable_device(pdev);
 	pci_set_power_state(pdev, PCI_D3hot);
-	nv_suspend_set_printk_level(NV_DBG_DEBUG);
-
 	return 0;
 }
 
@@ -582,15 +573,15 @@ nouveau_do_resume(struct drm_device *dev)
 	struct nouveau_drm *drm = nouveau_drm(dev);
 	struct nouveau_cli *cli;
 
-	NV_SUSPEND(drm, "re-enabling device...\n");
+	NV_INFO(drm, "re-enabling device...\n");
 
 	nouveau_agp_reset(drm);
 
-	NV_SUSPEND(drm, "resuming kernel object tree...\n");
+	NV_INFO(drm, "resuming kernel object tree...\n");
 	nouveau_client_init(&drm->client.base);
 	nouveau_agp_init(drm);
 
-	NV_SUSPEND(drm, "resuming client object trees...\n");
+	NV_INFO(drm, "resuming client object trees...\n");
 	if (drm->fence && nouveau_fence(drm)->resume)
 		nouveau_fence(drm)->resume(drm);
 
@@ -599,10 +590,9 @@ nouveau_do_resume(struct drm_device *dev)
 	}
 
 	nouveau_run_vbios_init(dev);
-	nouveau_pm_resume(dev);
 
 	if (dev->mode_config.num_crtc) {
-		NV_SUSPEND(drm, "resuming display...\n");
+		NV_INFO(drm, "resuming display...\n");
 		nouveau_display_repin(dev);
 	}
 
@@ -626,19 +616,15 @@ int nouveau_pmops_resume(struct device *dev)
 		return ret;
 	pci_set_master(pdev);
 
-	nv_suspend_set_printk_level(NV_DBG_INFO);
 	ret = nouveau_do_resume(drm_dev);
-	if (ret) {
-		nv_suspend_set_printk_level(NV_DBG_DEBUG);
+	if (ret)
 		return ret;
-	}
 	if (drm_dev->mode_config.num_crtc)
 		nouveau_fbcon_set_suspend(drm_dev, 0);
 
 	nouveau_fbcon_zfill_all(drm_dev);
 	if (drm_dev->mode_config.num_crtc)
 		nouveau_display_resume(drm_dev);
-	nv_suspend_set_printk_level(NV_DBG_DEBUG);
 	return 0;
 }
 
@@ -648,12 +634,10 @@ static int nouveau_pmops_freeze(struct device *dev)
 	struct drm_device *drm_dev = pci_get_drvdata(pdev);
 	int ret;
 
-	nv_suspend_set_printk_level(NV_DBG_INFO);
 	if (drm_dev->mode_config.num_crtc)
 		nouveau_fbcon_set_suspend(drm_dev, 1);
 
 	ret = nouveau_do_suspend(drm_dev);
-	nv_suspend_set_printk_level(NV_DBG_DEBUG);
 	return ret;
 }
 
@@ -663,18 +647,14 @@ static int nouveau_pmops_thaw(struct device *dev)
 	struct drm_device *drm_dev = pci_get_drvdata(pdev);
 	int ret;
 
-	nv_suspend_set_printk_level(NV_DBG_INFO);
 	ret = nouveau_do_resume(drm_dev);
-	if (ret) {
-		nv_suspend_set_printk_level(NV_DBG_DEBUG);
+	if (ret)
 		return ret;
-	}
 	if (drm_dev->mode_config.num_crtc)
 		nouveau_fbcon_set_suspend(drm_dev, 0);
 	nouveau_fbcon_zfill_all(drm_dev);
 	if (drm_dev->mode_config.num_crtc)
 		nouveau_display_resume(drm_dev);
-	nv_suspend_set_printk_level(NV_DBG_DEBUG);
 	return 0;
 }
 
@@ -816,8 +796,8 @@ driver = {
 #endif
 
 	.get_vblank_counter = drm_vblank_count,
-	.enable_vblank = nouveau_drm_vblank_enable,
-	.disable_vblank = nouveau_drm_vblank_disable,
+	.enable_vblank = nouveau_display_vblank_enable,
+	.disable_vblank = nouveau_display_vblank_disable,
 
 	.ioctls = nouveau_ioctls,
 	.num_ioctls = ARRAY_SIZE(nouveau_ioctls),
@@ -834,7 +814,6 @@ driver = {
 	.gem_prime_vmap = nouveau_gem_prime_vmap,
 	.gem_prime_vunmap = nouveau_gem_prime_vunmap,
 
-	.gem_init_object = nouveau_gem_object_new,
 	.gem_free_object = nouveau_gem_object_del,
 	.gem_open_object = nouveau_gem_object_open,
 	.gem_close_object = nouveau_gem_object_close,
@@ -879,6 +858,7 @@ static int nouveau_pmops_runtime_suspend(struct device *dev)
 	if (nouveau_runtime_pm == 0)
 		return -EINVAL;
 
+	nv_debug_level(SILENT);
 	drm_kms_helper_poll_disable(drm_dev);
 	vga_switcheroo_set_dynamic_switch(pdev, VGA_SWITCHEROO_OFF);
 	nouveau_switcheroo_optimus_dsm();
@@ -915,6 +895,7 @@ static int nouveau_pmops_runtime_resume(struct device *dev)
 	nv_mask(device, 0x88488, (1 << 25), (1 << 25));
 	vga_switcheroo_set_dynamic_switch(pdev, VGA_SWITCHEROO_ON);
 	drm_dev->switch_power_state = DRM_SWITCH_POWER_ON;
+	nv_debug_level(NORMAL);
 	return ret;
 }
 
diff --git a/drivers/gpu/drm/nouveau/nouveau_drm.h b/drivers/gpu/drm/nouveau/nouveau_drm.h
index 994fd6ec373b..4b0fb6c66be9 100644
--- a/drivers/gpu/drm/nouveau/nouveau_drm.h
+++ b/drivers/gpu/drm/nouveau/nouveau_drm.h
@@ -51,10 +51,12 @@ struct nouveau_drm_tile {
 };
 
 enum nouveau_drm_handle {
-	NVDRM_CLIENT = 0xffffffff,
-	NVDRM_DEVICE = 0xdddddddd,
-	NVDRM_PUSH   = 0xbbbb0000, /* |= client chid */
-	NVDRM_CHAN   = 0xcccc0000, /* |= client chid */
+	NVDRM_CLIENT  = 0xffffffff,
+	NVDRM_DEVICE  = 0xdddddddd,
+	NVDRM_CONTROL = 0xdddddddc,
+	NVDRM_PUSH    = 0xbbbb0000, /* |= client chid */
+	NVDRM_CHAN    = 0xcccc0000, /* |= client chid */
+	NVDRM_NVSW    = 0x55550000,
 };
 
 struct nouveau_cli {
@@ -127,10 +129,10 @@ struct nouveau_drm {
 	struct nvbios vbios;
 	struct nouveau_display *display;
 	struct backlight_device *backlight;
-	struct nouveau_eventh vblank[4];
 
 	/* power management */
-	struct nouveau_pm *pm;
+	struct nouveau_hwmon *hwmon;
+	struct nouveau_sysfs *sysfs;
 
 	/* display power reference */
 	bool have_disp_power_ref;
@@ -154,7 +156,6 @@ nouveau_dev(struct drm_device *dev)
 int nouveau_pmops_suspend(struct device *);
 int nouveau_pmops_resume(struct device *);
 
-#define NV_SUSPEND(cli, fmt, args...) nv_suspend((cli), fmt, ##args)
 #define NV_FATAL(cli, fmt, args...) nv_fatal((cli), fmt, ##args)
 #define NV_ERROR(cli, fmt, args...) nv_error((cli), fmt, ##args)
 #define NV_WARN(cli, fmt, args...) nv_warn((cli), fmt, ##args)
diff --git a/drivers/gpu/drm/nouveau/nouveau_fbcon.c b/drivers/gpu/drm/nouveau/nouveau_fbcon.c
index a86ecf65c164..7903e0ed3c75 100644
--- a/drivers/gpu/drm/nouveau/nouveau_fbcon.c
+++ b/drivers/gpu/drm/nouveau/nouveau_fbcon.c
@@ -420,7 +420,7 @@ nouveau_fbcon_destroy(struct drm_device *dev, struct nouveau_fbdev *fbcon)
 		nouveau_bo_unmap(nouveau_fb->nvbo);
 		nouveau_bo_vma_del(nouveau_fb->nvbo, &nouveau_fb->vma);
 		nouveau_bo_unpin(nouveau_fb->nvbo);
-		drm_gem_object_unreference_unlocked(nouveau_fb->nvbo->gem);
+		drm_gem_object_unreference_unlocked(&nouveau_fb->nvbo->gem);
 		nouveau_fb->nvbo = NULL;
 	}
 	drm_fb_helper_fini(&fbcon->helper);
@@ -503,34 +503,45 @@ nouveau_fbcon_fini(struct drm_device *dev)
 	drm->fbcon = NULL;
 }
 
-void nouveau_fbcon_save_disable_accel(struct drm_device *dev)
+void
+nouveau_fbcon_save_disable_accel(struct drm_device *dev)
 {
 	struct nouveau_drm *drm = nouveau_drm(dev);
-
-	drm->fbcon->saved_flags = drm->fbcon->helper.fbdev->flags;
-	drm->fbcon->helper.fbdev->flags |= FBINFO_HWACCEL_DISABLED;
+	if (drm->fbcon) {
+		drm->fbcon->saved_flags = drm->fbcon->helper.fbdev->flags;
+		drm->fbcon->helper.fbdev->flags |= FBINFO_HWACCEL_DISABLED;
+	}
 }
 
-void nouveau_fbcon_restore_accel(struct drm_device *dev)
+void
+nouveau_fbcon_restore_accel(struct drm_device *dev)
 {
 	struct nouveau_drm *drm = nouveau_drm(dev);
-	drm->fbcon->helper.fbdev->flags = drm->fbcon->saved_flags;
+	if (drm->fbcon) {
+		drm->fbcon->helper.fbdev->flags = drm->fbcon->saved_flags;
+	}
 }
 
-void nouveau_fbcon_set_suspend(struct drm_device *dev, int state)
+void
+nouveau_fbcon_set_suspend(struct drm_device *dev, int state)
 {
 	struct nouveau_drm *drm = nouveau_drm(dev);
-	console_lock();
-	if (state == 0)
-		nouveau_fbcon_save_disable_accel(dev);
-	fb_set_suspend(drm->fbcon->helper.fbdev, state);
-	if (state == 1)
-		nouveau_fbcon_restore_accel(dev);
-	console_unlock();
+	if (drm->fbcon) {
+		console_lock();
+		if (state == 0)
+			nouveau_fbcon_save_disable_accel(dev);
+		fb_set_suspend(drm->fbcon->helper.fbdev, state);
+		if (state == 1)
+			nouveau_fbcon_restore_accel(dev);
+		console_unlock();
+	}
 }
 
-void nouveau_fbcon_zfill_all(struct drm_device *dev)
+void
+nouveau_fbcon_zfill_all(struct drm_device *dev)
 {
 	struct nouveau_drm *drm = nouveau_drm(dev);
-	nouveau_fbcon_zfill(dev, drm->fbcon);
+	if (drm->fbcon) {
+		nouveau_fbcon_zfill(dev, drm->fbcon);
+	}
 }
diff --git a/drivers/gpu/drm/nouveau/nouveau_fence.c b/drivers/gpu/drm/nouveau/nouveau_fence.c
index be3149932c2d..40cf52e6d6d2 100644
--- a/drivers/gpu/drm/nouveau/nouveau_fence.c
+++ b/drivers/gpu/drm/nouveau/nouveau_fence.c
@@ -165,17 +165,11 @@ nouveau_fence_done(struct nouveau_fence *fence)
 	return !fence->channel;
 }
 
-struct nouveau_fence_uevent {
-	struct nouveau_eventh handler;
-	struct nouveau_fence_priv *priv;
-};
-
 static int
-nouveau_fence_wait_uevent_handler(struct nouveau_eventh *event, int index)
+nouveau_fence_wait_uevent_handler(void *data, int index)
 {
-	struct nouveau_fence_uevent *uevent =
-		container_of(event, struct nouveau_fence_uevent, handler);
-	wake_up_all(&uevent->priv->waiting);
+	struct nouveau_fence_priv *priv = data;
+	wake_up_all(&priv->waiting);
 	return NVKM_EVENT_KEEP;
 }
 
@@ -186,13 +180,16 @@ nouveau_fence_wait_uevent(struct nouveau_fence *fence, bool intr)
 	struct nouveau_channel *chan = fence->channel;
 	struct nouveau_fifo *pfifo = nouveau_fifo(chan->drm->device);
 	struct nouveau_fence_priv *priv = chan->drm->fence;
-	struct nouveau_fence_uevent uevent = {
-		.handler.func = nouveau_fence_wait_uevent_handler,
-		.priv = priv,
-	};
+	struct nouveau_eventh *handler;
 	int ret = 0;
 
-	nouveau_event_get(pfifo->uevent, 0, &uevent.handler);
+	ret = nouveau_event_new(pfifo->uevent, 0,
+				nouveau_fence_wait_uevent_handler,
+				priv, &handler);
+	if (ret)
+		return ret;
+
+	nouveau_event_get(handler);
 
 	if (fence->timeout) {
 		unsigned long timeout = fence->timeout - jiffies;
@@ -224,7 +221,7 @@ nouveau_fence_wait_uevent(struct nouveau_fence *fence, bool intr)
 		}
 	}
 
-	nouveau_event_put(pfifo->uevent, 0, &uevent.handler);
+	nouveau_event_ref(NULL, &handler);
 	if (unlikely(ret < 0))
 		return ret;
 
@@ -309,7 +306,8 @@ nouveau_fence_unref(struct nouveau_fence **pfence)
 struct nouveau_fence *
 nouveau_fence_ref(struct nouveau_fence *fence)
 {
-	kref_get(&fence->kref);
+	if (fence)
+		kref_get(&fence->kref);
 	return fence;
 }
 
diff --git a/drivers/gpu/drm/nouveau/nouveau_gem.c b/drivers/gpu/drm/nouveau/nouveau_gem.c
index f32b71238c03..78a27f8ad7d9 100644
--- a/drivers/gpu/drm/nouveau/nouveau_gem.c
+++ b/drivers/gpu/drm/nouveau/nouveau_gem.c
@@ -34,29 +34,20 @@
 #include "nouveau_ttm.h"
 #include "nouveau_gem.h"
 
-int
-nouveau_gem_object_new(struct drm_gem_object *gem)
-{
-	return 0;
-}
-
 void
 nouveau_gem_object_del(struct drm_gem_object *gem)
 {
-	struct nouveau_bo *nvbo = gem->driver_private;
+	struct nouveau_bo *nvbo = nouveau_gem_object(gem);
 	struct ttm_buffer_object *bo = &nvbo->bo;
 
-	if (!nvbo)
-		return;
-	nvbo->gem = NULL;
-
 	if (gem->import_attach)
 		drm_prime_gem_destroy(gem, nvbo->bo.sg);
 
-	ttm_bo_unref(&bo);
-
 	drm_gem_object_release(gem);
-	kfree(gem);
+
+	/* reset filp so nouveau_bo_del_ttm() can test for it */
+	gem->filp = NULL;
+	ttm_bo_unref(&bo);
 }
 
 int
@@ -115,8 +106,7 @@ nouveau_gem_object_unmap(struct nouveau_bo *nvbo, struct nouveau_vma *vma)
 
 	if (mapped) {
 		spin_lock(&nvbo->bo.bdev->fence_lock);
-		if (nvbo->bo.sync_obj)
-			fence = nouveau_fence_ref(nvbo->bo.sync_obj);
+		fence = nouveau_fence_ref(nvbo->bo.sync_obj);
 		spin_unlock(&nvbo->bo.bdev->fence_lock);
 	}
 
@@ -186,14 +176,15 @@ nouveau_gem_new(struct drm_device *dev, int size, int align, uint32_t domain,
 	if (nv_device(drm->device)->card_type >= NV_50)
 		nvbo->valid_domains &= domain;
 
-	nvbo->gem = drm_gem_object_alloc(dev, nvbo->bo.mem.size);
-	if (!nvbo->gem) {
+	/* Initialize the embedded gem-object. We return a single gem-reference
+	 * to the caller, instead of a normal nouveau_bo ttm reference. */
+	ret = drm_gem_object_init(dev, &nvbo->gem, nvbo->bo.mem.size);
+	if (ret) {
 		nouveau_bo_ref(NULL, pnvbo);
 		return -ENOMEM;
 	}
 
-	nvbo->bo.persistent_swap_storage = nvbo->gem->filp;
-	nvbo->gem->driver_private = nvbo;
+	nvbo->bo.persistent_swap_storage = nvbo->gem.filp;
 	return 0;
 }
 
@@ -250,15 +241,15 @@ nouveau_gem_ioctl_new(struct drm_device *dev, void *data,
 	if (ret)
 		return ret;
 
-	ret = drm_gem_handle_create(file_priv, nvbo->gem, &req->info.handle);
+	ret = drm_gem_handle_create(file_priv, &nvbo->gem, &req->info.handle);
 	if (ret == 0) {
-		ret = nouveau_gem_info(file_priv, nvbo->gem, &req->info);
+		ret = nouveau_gem_info(file_priv, &nvbo->gem, &req->info);
 		if (ret)
 			drm_gem_handle_delete(file_priv, req->info.handle);
 	}
 
 	/* drop reference from allocate - handle holds it now */
-	drm_gem_object_unreference_unlocked(nvbo->gem);
+	drm_gem_object_unreference_unlocked(&nvbo->gem);
 	return ret;
 }
 
@@ -266,7 +257,7 @@ static int
 nouveau_gem_set_domain(struct drm_gem_object *gem, uint32_t read_domains,
 		       uint32_t write_domains, uint32_t valid_domains)
 {
-	struct nouveau_bo *nvbo = gem->driver_private;
+	struct nouveau_bo *nvbo = nouveau_gem_object(gem);
 	struct ttm_buffer_object *bo = &nvbo->bo;
 	uint32_t domains = valid_domains & nvbo->valid_domains &
 		(write_domains ? write_domains : read_domains);
@@ -317,7 +308,8 @@ validate_fini_list(struct list_head *list, struct nouveau_fence *fence,
 	list_for_each_safe(entry, tmp, list) {
 		nvbo = list_entry(entry, struct nouveau_bo, entry);
 
-		nouveau_bo_fence(nvbo, fence);
+		if (likely(fence))
+			nouveau_bo_fence(nvbo, fence);
 
 		if (unlikely(nvbo->validate_mapped)) {
 			ttm_bo_kunmap(&nvbo->kmap);
@@ -327,7 +319,7 @@ validate_fini_list(struct list_head *list, struct nouveau_fence *fence,
 		list_del(&nvbo->entry);
 		nvbo->reserved_by = NULL;
 		ttm_bo_unreserve_ticket(&nvbo->bo, ticket);
-		drm_gem_object_unreference_unlocked(nvbo->gem);
+		drm_gem_object_unreference_unlocked(&nvbo->gem);
 	}
 }
 
@@ -376,7 +368,7 @@ retry:
 			validate_fini(op, NULL);
 			return -ENOENT;
 		}
-		nvbo = gem->driver_private;
+		nvbo = nouveau_gem_object(gem);
 		if (nvbo == res_bo) {
 			res_bo = NULL;
 			drm_gem_object_unreference_unlocked(gem);
@@ -446,8 +438,7 @@ validate_sync(struct nouveau_channel *chan, struct nouveau_bo *nvbo)
 	int ret = 0;
 
 	spin_lock(&nvbo->bo.bdev->fence_lock);
-	if (nvbo->bo.sync_obj)
-		fence = nouveau_fence_ref(nvbo->bo.sync_obj);
+	fence = nouveau_fence_ref(nvbo->bo.sync_obj);
 	spin_unlock(&nvbo->bo.bdev->fence_lock);
 
 	if (fence) {
@@ -478,7 +469,7 @@ validate_list(struct nouveau_channel *chan, struct nouveau_cli *cli,
 			return ret;
 		}
 
-		ret = nouveau_gem_set_domain(nvbo->gem, b->read_domains,
+		ret = nouveau_gem_set_domain(&nvbo->gem, b->read_domains,
 					     b->write_domains,
 					     b->valid_domains);
 		if (unlikely(ret)) {
diff --git a/drivers/gpu/drm/nouveau/nouveau_gem.h b/drivers/gpu/drm/nouveau/nouveau_gem.h
index 502e4290aa8f..7caca057bc38 100644
--- a/drivers/gpu/drm/nouveau/nouveau_gem.h
+++ b/drivers/gpu/drm/nouveau/nouveau_gem.h
@@ -12,14 +12,13 @@
 static inline struct nouveau_bo *
 nouveau_gem_object(struct drm_gem_object *gem)
 {
-	return gem ? gem->driver_private : NULL;
+	return gem ? container_of(gem, struct nouveau_bo, gem) : NULL;
 }
 
 /* nouveau_gem.c */
 extern int nouveau_gem_new(struct drm_device *, int size, int align,
 			   uint32_t domain, uint32_t tile_mode,
 			   uint32_t tile_flags, struct nouveau_bo **);
-extern int nouveau_gem_object_new(struct drm_gem_object *);
 extern void nouveau_gem_object_del(struct drm_gem_object *);
 extern int nouveau_gem_object_open(struct drm_gem_object *, struct drm_file *);
 extern void nouveau_gem_object_close(struct drm_gem_object *,
diff --git a/drivers/gpu/drm/nouveau/nouveau_pm.c b/drivers/gpu/drm/nouveau/nouveau_hwmon.c
index 936b442a6ab7..38a4db5bfe21 100644
--- a/drivers/gpu/drm/nouveau/nouveau_pm.c
+++ b/drivers/gpu/drm/nouveau/nouveau_hwmon.c
@@ -32,369 +32,12 @@
 #include <drm/drmP.h>
 
 #include "nouveau_drm.h"
-#include "nouveau_pm.h"
+#include "nouveau_hwmon.h"
 
 #include <subdev/gpio.h>
 #include <subdev/timer.h>
 #include <subdev/therm.h>
 
-MODULE_PARM_DESC(perflvl, "Performance level (default: boot)");
-static char *nouveau_perflvl;
-module_param_named(perflvl, nouveau_perflvl, charp, 0400);
-
-MODULE_PARM_DESC(perflvl_wr, "Allow perflvl changes (warning: dangerous!)");
-static int nouveau_perflvl_wr;
-module_param_named(perflvl_wr, nouveau_perflvl_wr, int, 0400);
-
-static int
-nouveau_pm_perflvl_aux(struct drm_device *dev, struct nouveau_pm_level *perflvl,
-		       struct nouveau_pm_level *a, struct nouveau_pm_level *b)
-{
-	struct nouveau_drm *drm = nouveau_drm(dev);
-	struct nouveau_pm *pm = nouveau_pm(dev);
-	struct nouveau_therm *therm = nouveau_therm(drm->device);
-	int ret;
-
-	/*XXX: not on all boards, we should control based on temperature
-	 *     on recent boards..  or maybe on some other factor we don't
-	 *     know about?
-	 */
-	if (therm && therm->fan_set &&
-		a->fanspeed && b->fanspeed && b->fanspeed > a->fanspeed) {
-		ret = therm->fan_set(therm, perflvl->fanspeed);
-		if (ret && ret != -ENODEV) {
-			NV_ERROR(drm, "fanspeed set failed: %d\n", ret);
-		}
-	}
-
-	if (pm->voltage.supported && pm->voltage_set) {
-		if (perflvl->volt_min && b->volt_min > a->volt_min) {
-			ret = pm->voltage_set(dev, perflvl->volt_min);
-			if (ret) {
-				NV_ERROR(drm, "voltage set failed: %d\n", ret);
-				return ret;
-			}
-		}
-	}
-
-	return 0;
-}
-
-static int
-nouveau_pm_perflvl_set(struct drm_device *dev, struct nouveau_pm_level *perflvl)
-{
-	struct nouveau_pm *pm = nouveau_pm(dev);
-	void *state;
-	int ret;
-
-	if (perflvl == pm->cur)
-		return 0;
-
-	ret = nouveau_pm_perflvl_aux(dev, perflvl, pm->cur, perflvl);
-	if (ret)
-		return ret;
-
-	state = pm->clocks_pre(dev, perflvl);
-	if (IS_ERR(state)) {
-		ret = PTR_ERR(state);
-		goto error;
-	}
-	ret = pm->clocks_set(dev, state);
-	if (ret)
-		goto error;
-
-	ret = nouveau_pm_perflvl_aux(dev, perflvl, perflvl, pm->cur);
-	if (ret)
-		return ret;
-
-	pm->cur = perflvl;
-	return 0;
-
-error:
-	/* restore the fan speed and voltage before leaving */
-	nouveau_pm_perflvl_aux(dev, perflvl, perflvl, pm->cur);
-	return ret;
-}
-
-void
-nouveau_pm_trigger(struct drm_device *dev)
-{
-	struct nouveau_drm *drm = nouveau_drm(dev);
-	struct nouveau_timer *ptimer = nouveau_timer(drm->device);
-	struct nouveau_pm *pm = nouveau_pm(dev);
-	struct nouveau_pm_profile *profile = NULL;
-	struct nouveau_pm_level *perflvl = NULL;
-	int ret;
-
-	/* select power profile based on current power source */
-	if (power_supply_is_system_supplied())
-		profile = pm->profile_ac;
-	else
-		profile = pm->profile_dc;
-
-	if (profile != pm->profile) {
-		pm->profile->func->fini(pm->profile);
-		pm->profile = profile;
-		pm->profile->func->init(pm->profile);
-	}
-
-	/* select performance level based on profile */
-	perflvl = profile->func->select(profile);
-
-	/* change perflvl, if necessary */
-	if (perflvl != pm->cur) {
-		u64 time0 = ptimer->read(ptimer);
-
-		NV_INFO(drm, "setting performance level: %d", perflvl->id);
-		ret = nouveau_pm_perflvl_set(dev, perflvl);
-		if (ret)
-			NV_INFO(drm, "> reclocking failed: %d\n\n", ret);
-
-		NV_INFO(drm, "> reclocking took %lluns\n\n",
-			     ptimer->read(ptimer) - time0);
-	}
-}
-
-static struct nouveau_pm_profile *
-profile_find(struct drm_device *dev, const char *string)
-{
-	struct nouveau_pm *pm = nouveau_pm(dev);
-	struct nouveau_pm_profile *profile;
-
-	list_for_each_entry(profile, &pm->profiles, head) {
-		if (!strncmp(profile->name, string, sizeof(profile->name)))
-			return profile;
-	}
-
-	return NULL;
-}
-
-static int
-nouveau_pm_profile_set(struct drm_device *dev, const char *profile)
-{
-	struct nouveau_pm *pm = nouveau_pm(dev);
-	struct nouveau_pm_profile *ac = NULL, *dc = NULL;
-	char string[16], *cur = string, *ptr;
-
-	/* safety precaution, for now */
-	if (nouveau_perflvl_wr != 7777)
-		return -EPERM;
-
-	strncpy(string, profile, sizeof(string));
-	string[sizeof(string) - 1] = 0;
-	if ((ptr = strchr(string, '\n')))
-		*ptr = '\0';
-
-	ptr = strsep(&cur, ",");
-	if (ptr)
-		ac = profile_find(dev, ptr);
-
-	ptr = strsep(&cur, ",");
-	if (ptr)
-		dc = profile_find(dev, ptr);
-	else
-		dc = ac;
-
-	if (ac == NULL || dc == NULL)
-		return -EINVAL;
-
-	pm->profile_ac = ac;
-	pm->profile_dc = dc;
-	nouveau_pm_trigger(dev);
-	return 0;
-}
-
-static void
-nouveau_pm_static_dummy(struct nouveau_pm_profile *profile)
-{
-}
-
-static struct nouveau_pm_level *
-nouveau_pm_static_select(struct nouveau_pm_profile *profile)
-{
-	return container_of(profile, struct nouveau_pm_level, profile);
-}
-
-const struct nouveau_pm_profile_func nouveau_pm_static_profile_func = {
-	.destroy = nouveau_pm_static_dummy,
-	.init = nouveau_pm_static_dummy,
-	.fini = nouveau_pm_static_dummy,
-	.select = nouveau_pm_static_select,
-};
-
-static int
-nouveau_pm_perflvl_get(struct drm_device *dev, struct nouveau_pm_level *perflvl)
-{
-	struct nouveau_drm *drm = nouveau_drm(dev);
-	struct nouveau_pm *pm = nouveau_pm(dev);
-	struct nouveau_therm *therm = nouveau_therm(drm->device);
-	int ret;
-
-	memset(perflvl, 0, sizeof(*perflvl));
-
-	if (pm->clocks_get) {
-		ret = pm->clocks_get(dev, perflvl);
-		if (ret)
-			return ret;
-	}
-
-	if (pm->voltage.supported && pm->voltage_get) {
-		ret = pm->voltage_get(dev);
-		if (ret > 0) {
-			perflvl->volt_min = ret;
-			perflvl->volt_max = ret;
-		}
-	}
-
-	if (therm && therm->fan_get) {
-		ret = therm->fan_get(therm);
-		if (ret >= 0)
-			perflvl->fanspeed = ret;
-	}
-
-	nouveau_mem_timing_read(dev, &perflvl->timing);
-	return 0;
-}
-
-static void
-nouveau_pm_perflvl_info(struct nouveau_pm_level *perflvl, char *ptr, int len)
-{
-	char c[16], s[16], v[32], f[16], m[16];
-
-	c[0] = '\0';
-	if (perflvl->core)
-		snprintf(c, sizeof(c), " core %dMHz", perflvl->core / 1000);
-
-	s[0] = '\0';
-	if (perflvl->shader)
-		snprintf(s, sizeof(s), " shader %dMHz", perflvl->shader / 1000);
-
-	m[0] = '\0';
-	if (perflvl->memory)
-		snprintf(m, sizeof(m), " memory %dMHz", perflvl->memory / 1000);
-
-	v[0] = '\0';
-	if (perflvl->volt_min && perflvl->volt_min != perflvl->volt_max) {
-		snprintf(v, sizeof(v), " voltage %dmV-%dmV",
-			 perflvl->volt_min / 1000, perflvl->volt_max / 1000);
-	} else
-	if (perflvl->volt_min) {
-		snprintf(v, sizeof(v), " voltage %dmV",
-			 perflvl->volt_min / 1000);
-	}
-
-	f[0] = '\0';
-	if (perflvl->fanspeed)
-		snprintf(f, sizeof(f), " fanspeed %d%%", perflvl->fanspeed);
-
-	snprintf(ptr, len, "%s%s%s%s%s\n", c, s, m, v, f);
-}
-
-static ssize_t
-nouveau_pm_get_perflvl_info(struct device *d,
-			    struct device_attribute *a, char *buf)
-{
-	struct nouveau_pm_level *perflvl =
-		container_of(a, struct nouveau_pm_level, dev_attr);
-	char *ptr = buf;
-	int len = PAGE_SIZE;
-
-	snprintf(ptr, len, "%d:", perflvl->id);
-	ptr += strlen(buf);
-	len -= strlen(buf);
-
-	nouveau_pm_perflvl_info(perflvl, ptr, len);
-	return strlen(buf);
-}
-
-static ssize_t
-nouveau_pm_get_perflvl(struct device *d, struct device_attribute *a, char *buf)
-{
-	struct drm_device *dev = pci_get_drvdata(to_pci_dev(d));
-	struct nouveau_pm *pm = nouveau_pm(dev);
-	struct nouveau_pm_level cur;
-	int len = PAGE_SIZE, ret;
-	char *ptr = buf;
-
-	snprintf(ptr, len, "profile: %s, %s\nc:",
-		 pm->profile_ac->name, pm->profile_dc->name);
-	ptr += strlen(buf);
-	len -= strlen(buf);
-
-	ret = nouveau_pm_perflvl_get(dev, &cur);
-	if (ret == 0)
-		nouveau_pm_perflvl_info(&cur, ptr, len);
-	return strlen(buf);
-}
-
-static ssize_t
-nouveau_pm_set_perflvl(struct device *d, struct device_attribute *a,
-		       const char *buf, size_t count)
-{
-	struct drm_device *dev = pci_get_drvdata(to_pci_dev(d));
-	int ret;
-
-	ret = nouveau_pm_profile_set(dev, buf);
-	if (ret)
-		return ret;
-	return strlen(buf);
-}
-
-static DEVICE_ATTR(performance_level, S_IRUGO | S_IWUSR,
-		   nouveau_pm_get_perflvl, nouveau_pm_set_perflvl);
-
-static int
-nouveau_sysfs_init(struct drm_device *dev)
-{
-	struct nouveau_drm *drm = nouveau_drm(dev);
-	struct nouveau_pm *pm = nouveau_pm(dev);
-	struct device *d = &dev->pdev->dev;
-	int ret, i;
-
-	ret = device_create_file(d, &dev_attr_performance_level);
-	if (ret)
-		return ret;
-
-	for (i = 0; i < pm->nr_perflvl; i++) {
-		struct nouveau_pm_level *perflvl = &pm->perflvl[i];
-
-		perflvl->dev_attr.attr.name = perflvl->name;
-		perflvl->dev_attr.attr.mode = S_IRUGO;
-		perflvl->dev_attr.show = nouveau_pm_get_perflvl_info;
-		perflvl->dev_attr.store = NULL;
-		sysfs_attr_init(&perflvl->dev_attr.attr);
-
-		ret = device_create_file(d, &perflvl->dev_attr);
-		if (ret) {
-			NV_ERROR(drm, "failed pervlvl %d sysfs: %d\n",
-				 perflvl->id, i);
-			perflvl->dev_attr.attr.name = NULL;
-			nouveau_pm_fini(dev);
-			return ret;
-		}
-	}
-
-	return 0;
-}
-
-static void
-nouveau_sysfs_fini(struct drm_device *dev)
-{
-	struct nouveau_pm *pm = nouveau_pm(dev);
-	struct device *d = &dev->pdev->dev;
-	int i;
-
-	device_remove_file(d, &dev_attr_performance_level);
-	for (i = 0; i < pm->nr_perflvl; i++) {
-		struct nouveau_pm_level *pl = &pm->perflvl[i];
-
-		if (!pl->dev_attr.attr.name)
-			break;
-
-		device_remove_file(d, &pl->dev_attr);
-	}
-}
-
 #if defined(CONFIG_HWMON) || (defined(MODULE) && defined(CONFIG_HWMON_MODULE))
 static ssize_t
 nouveau_hwmon_show_temp(struct device *d, struct device_attribute *a, char *buf)
@@ -778,9 +421,6 @@ nouveau_hwmon_set_pwm1(struct device *d, struct device_attribute *a,
 	int ret = -ENODEV;
 	long value;
 
-	if (nouveau_perflvl_wr != 7777)
-		return -EPERM;
-
 	if (kstrtol(buf, 10, &value) == -EINVAL)
 		return -EINVAL;
 
@@ -919,17 +559,21 @@ static const struct attribute_group hwmon_pwm_fan_attrgroup = {
 };
 #endif
 
-static int
+int
 nouveau_hwmon_init(struct drm_device *dev)
 {
-	struct nouveau_pm *pm = nouveau_pm(dev);
-
 #if defined(CONFIG_HWMON) || (defined(MODULE) && defined(CONFIG_HWMON_MODULE))
 	struct nouveau_drm *drm = nouveau_drm(dev);
 	struct nouveau_therm *therm = nouveau_therm(drm->device);
+	struct nouveau_hwmon *hwmon;
 	struct device *hwmon_dev;
 	int ret = 0;
 
+	hwmon = drm->hwmon = kzalloc(sizeof(*hwmon), GFP_KERNEL);
+	if (!hwmon)
+		return -ENOMEM;
+	hwmon->dev = dev;
+
 	if (!therm || !therm->temp_get || !therm->attr_get || !therm->attr_set)
 		return -ENODEV;
 
@@ -976,199 +620,37 @@ nouveau_hwmon_init(struct drm_device *dev)
 			goto error;
 	}
 
-	pm->hwmon = hwmon_dev;
+	hwmon->hwmon = hwmon_dev;
 
 	return 0;
 
 error:
 	NV_ERROR(drm, "Unable to create some hwmon sysfs files: %d\n", ret);
 	hwmon_device_unregister(hwmon_dev);
-	pm->hwmon = NULL;
+	hwmon->hwmon = NULL;
 	return ret;
 #else
-	pm->hwmon = NULL;
+	hwmon->hwmon = NULL;
 	return 0;
 #endif
 }
 
-static void
+void
 nouveau_hwmon_fini(struct drm_device *dev)
 {
 #if defined(CONFIG_HWMON) || (defined(MODULE) && defined(CONFIG_HWMON_MODULE))
-	struct nouveau_pm *pm = nouveau_pm(dev);
+	struct nouveau_hwmon *hwmon = nouveau_hwmon(dev);
 
-	if (pm->hwmon) {
-		sysfs_remove_group(&pm->hwmon->kobj, &hwmon_default_attrgroup);
-		sysfs_remove_group(&pm->hwmon->kobj, &hwmon_temp_attrgroup);
-		sysfs_remove_group(&pm->hwmon->kobj, &hwmon_pwm_fan_attrgroup);
-		sysfs_remove_group(&pm->hwmon->kobj, &hwmon_fan_rpm_attrgroup);
+	if (hwmon->hwmon) {
+		sysfs_remove_group(&hwmon->hwmon->kobj, &hwmon_default_attrgroup);
+		sysfs_remove_group(&hwmon->hwmon->kobj, &hwmon_temp_attrgroup);
+		sysfs_remove_group(&hwmon->hwmon->kobj, &hwmon_pwm_fan_attrgroup);
+		sysfs_remove_group(&hwmon->hwmon->kobj, &hwmon_fan_rpm_attrgroup);
 
-		hwmon_device_unregister(pm->hwmon);
+		hwmon_device_unregister(hwmon->hwmon);
 	}
-#endif
-}
-
-#if defined(CONFIG_ACPI) && defined(CONFIG_POWER_SUPPLY)
-static int
-nouveau_pm_acpi_event(struct notifier_block *nb, unsigned long val, void *data)
-{
-	struct nouveau_pm *pm = container_of(nb, struct nouveau_pm, acpi_nb);
-	struct nouveau_drm *drm = nouveau_drm(pm->dev);
-	struct acpi_bus_event *entry = (struct acpi_bus_event *)data;
-
-	if (strcmp(entry->device_class, "ac_adapter") == 0) {
-		bool ac = power_supply_is_system_supplied();
 
-		NV_DEBUG(drm, "power supply changed: %s\n", ac ? "AC" : "DC");
-		nouveau_pm_trigger(pm->dev);
-	}
-
-	return NOTIFY_OK;
-}
+	nouveau_drm(dev)->hwmon = NULL;
+	kfree(hwmon);
 #endif
-
-int
-nouveau_pm_init(struct drm_device *dev)
-{
-	struct nouveau_device *device = nouveau_dev(dev);
-	struct nouveau_drm *drm = nouveau_drm(dev);
-	struct nouveau_pm *pm;
-	char info[256];
-	int ret, i;
-
-	pm = drm->pm = kzalloc(sizeof(*pm), GFP_KERNEL);
-	if (!pm)
-		return -ENOMEM;
-
-	pm->dev = dev;
-
-	if (device->card_type < NV_40) {
-		pm->clocks_get = nv04_pm_clocks_get;
-		pm->clocks_pre = nv04_pm_clocks_pre;
-		pm->clocks_set = nv04_pm_clocks_set;
-		if (nouveau_gpio(drm->device)) {
-			pm->voltage_get = nouveau_voltage_gpio_get;
-			pm->voltage_set = nouveau_voltage_gpio_set;
-		}
-	} else
-	if (device->card_type < NV_50) {
-		pm->clocks_get = nv40_pm_clocks_get;
-		pm->clocks_pre = nv40_pm_clocks_pre;
-		pm->clocks_set = nv40_pm_clocks_set;
-		pm->voltage_get = nouveau_voltage_gpio_get;
-		pm->voltage_set = nouveau_voltage_gpio_set;
-	} else
-	if (device->card_type < NV_C0) {
-		if (device->chipset <  0xa3 ||
-		    device->chipset == 0xaa ||
-		    device->chipset == 0xac) {
-			pm->clocks_get = nv50_pm_clocks_get;
-			pm->clocks_pre = nv50_pm_clocks_pre;
-			pm->clocks_set = nv50_pm_clocks_set;
-		} else {
-			pm->clocks_get = nva3_pm_clocks_get;
-			pm->clocks_pre = nva3_pm_clocks_pre;
-			pm->clocks_set = nva3_pm_clocks_set;
-		}
-		pm->voltage_get = nouveau_voltage_gpio_get;
-		pm->voltage_set = nouveau_voltage_gpio_set;
-	} else
-	if (device->card_type < NV_E0) {
-		pm->clocks_get = nvc0_pm_clocks_get;
-		pm->clocks_pre = nvc0_pm_clocks_pre;
-		pm->clocks_set = nvc0_pm_clocks_set;
-		pm->voltage_get = nouveau_voltage_gpio_get;
-		pm->voltage_set = nouveau_voltage_gpio_set;
-	}
-
-
-	/* parse aux tables from vbios */
-	nouveau_volt_init(dev);
-
-	INIT_LIST_HEAD(&pm->profiles);
-
-	/* determine current ("boot") performance level */
-	ret = nouveau_pm_perflvl_get(dev, &pm->boot);
-	if (ret) {
-		NV_ERROR(drm, "failed to determine boot perflvl\n");
-		return ret;
-	}
-
-	strncpy(pm->boot.name, "boot", 4);
-	strncpy(pm->boot.profile.name, "boot", 4);
-	pm->boot.profile.func = &nouveau_pm_static_profile_func;
-
-	list_add(&pm->boot.profile.head, &pm->profiles);
-
-	pm->profile_ac = &pm->boot.profile;
-	pm->profile_dc = &pm->boot.profile;
-	pm->profile = &pm->boot.profile;
-	pm->cur = &pm->boot;
-
-	/* add performance levels from vbios */
-	nouveau_perf_init(dev);
-
-	/* display available performance levels */
-	NV_INFO(drm, "%d available performance level(s)\n", pm->nr_perflvl);
-	for (i = 0; i < pm->nr_perflvl; i++) {
-		nouveau_pm_perflvl_info(&pm->perflvl[i], info, sizeof(info));
-		NV_INFO(drm, "%d:%s", pm->perflvl[i].id, info);
-	}
-
-	nouveau_pm_perflvl_info(&pm->boot, info, sizeof(info));
-	NV_INFO(drm, "c:%s", info);
-
-	/* switch performance levels now if requested */
-	if (nouveau_perflvl != NULL)
-		nouveau_pm_profile_set(dev, nouveau_perflvl);
-
-	nouveau_sysfs_init(dev);
-	nouveau_hwmon_init(dev);
-#if defined(CONFIG_ACPI) && defined(CONFIG_POWER_SUPPLY)
-	pm->acpi_nb.notifier_call = nouveau_pm_acpi_event;
-	register_acpi_notifier(&pm->acpi_nb);
-#endif
-
-	return 0;
-}
-
-void
-nouveau_pm_fini(struct drm_device *dev)
-{
-	struct nouveau_pm *pm = nouveau_pm(dev);
-	struct nouveau_pm_profile *profile, *tmp;
-
-	list_for_each_entry_safe(profile, tmp, &pm->profiles, head) {
-		list_del(&profile->head);
-		profile->func->destroy(profile);
-	}
-
-	if (pm->cur != &pm->boot)
-		nouveau_pm_perflvl_set(dev, &pm->boot);
-
-	nouveau_perf_fini(dev);
-	nouveau_volt_fini(dev);
-
-#if defined(CONFIG_ACPI) && defined(CONFIG_POWER_SUPPLY)
-	unregister_acpi_notifier(&pm->acpi_nb);
-#endif
-	nouveau_hwmon_fini(dev);
-	nouveau_sysfs_fini(dev);
-
-	nouveau_drm(dev)->pm = NULL;
-	kfree(pm);
-}
-
-void
-nouveau_pm_resume(struct drm_device *dev)
-{
-	struct nouveau_pm *pm = nouveau_pm(dev);
-	struct nouveau_pm_level *perflvl;
-
-	if (!pm->cur || pm->cur == &pm->boot)
-		return;
-
-	perflvl = pm->cur;
-	pm->cur = &pm->boot;
-	nouveau_pm_perflvl_set(dev, perflvl);
 }
diff --git a/drivers/gpu/drm/nouveau/nouveau_hwmon.h b/drivers/gpu/drm/nouveau/nouveau_hwmon.h
new file mode 100644
index 000000000000..62ccbb39863c
--- /dev/null
+++ b/drivers/gpu/drm/nouveau/nouveau_hwmon.h
@@ -0,0 +1,43 @@
+/*
+ * Copyright 2010 Red Hat Inc.
+ *
+ * Permission is hereby granted, free of charge, to any person obtaining a
+ * copy of this software and associated documentation files (the "Software"),
+ * to deal in the Software without restriction, including without limitation
+ * the rights to use, copy, modify, merge, publish, distribute, sublicense,
+ * and/or sell copies of the Software, and to permit persons to whom the
+ * Software is furnished to do so, subject to the following conditions:
+ *
+ * The above copyright notice and this permission notice shall be included in
+ * all copies or substantial portions of the Software.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+ * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+ * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT.  IN NO EVENT SHALL
+ * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR
+ * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE,
+ * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR
+ * OTHER DEALINGS IN THE SOFTWARE.
+ *
+ * Authors: Ben Skeggs
+ */
+
+#ifndef __NOUVEAU_PM_H__
+#define __NOUVEAU_PM_H__
+
+struct nouveau_hwmon {
+	struct drm_device *dev;
+	struct device *hwmon;
+};
+
+static inline struct nouveau_hwmon *
+nouveau_hwmon(struct drm_device *dev)
+{
+	return nouveau_drm(dev)->hwmon;
+}
+
+/* nouveau_hwmon.c */
+int  nouveau_hwmon_init(struct drm_device *dev);
+void nouveau_hwmon_fini(struct drm_device *dev);
+
+#endif
diff --git a/drivers/gpu/drm/nouveau/nouveau_hwsq.h b/drivers/gpu/drm/nouveau/nouveau_hwsq.h
deleted file mode 100644
index 697687593a81..000000000000
--- a/drivers/gpu/drm/nouveau/nouveau_hwsq.h
+++ /dev/null
@@ -1,115 +0,0 @@
-/*
- * Copyright 2010 Red Hat Inc.
- *
- * Permission is hereby granted, free of charge, to any person obtaining a
- * copy of this software and associated documentation files (the "Software"),
- * to deal in the Software without restriction, including without limitation
- * the rights to use, copy, modify, merge, publish, distribute, sublicense,
- * and/or sell copies of the Software, and to permit persons to whom the
- * Software is furnished to do so, subject to the following conditions:
- *
- * The above copyright notice and this permission notice shall be included in
- * all copies or substantial portions of the Software.
- *
- * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
- * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
- * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT.  IN NO EVENT SHALL
- * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR
- * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE,
- * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR
- * OTHER DEALINGS IN THE SOFTWARE.
- *
- * Authors: Ben Skeggs
- */
-
-#ifndef __NOUVEAU_HWSQ_H__
-#define __NOUVEAU_HWSQ_H__
-
-struct hwsq_ucode {
-	u8 data[0x200];
-	union {
-		u8  *u08;
-		u16 *u16;
-		u32 *u32;
-	} ptr;
-	u16 len;
-
-	u32 reg;
-	u32 val;
-};
-
-static inline void
-hwsq_init(struct hwsq_ucode *hwsq)
-{
-	hwsq->ptr.u08 = hwsq->data;
-	hwsq->reg = 0xffffffff;
-	hwsq->val = 0xffffffff;
-}
-
-static inline void
-hwsq_fini(struct hwsq_ucode *hwsq)
-{
-	do {
-		*hwsq->ptr.u08++ = 0x7f;
-		hwsq->len = hwsq->ptr.u08 - hwsq->data;
-	} while (hwsq->len & 3);
-	hwsq->ptr.u08 = hwsq->data;
-}
-
-static inline void
-hwsq_usec(struct hwsq_ucode *hwsq, u8 usec)
-{
-	u32 shift = 0;
-	while (usec & ~3) {
-		usec >>= 2;
-		shift++;
-	}
-
-	*hwsq->ptr.u08++ = (shift << 2) | usec;
-}
-
-static inline void
-hwsq_setf(struct hwsq_ucode *hwsq, u8 flag, int val)
-{
-	flag += 0x80;
-	if (val >= 0)
-		flag += 0x20;
-	if (val >= 1)
-		flag += 0x20;
-	*hwsq->ptr.u08++ = flag;
-}
-
-static inline void
-hwsq_op5f(struct hwsq_ucode *hwsq, u8 v0, u8 v1)
-{
-	*hwsq->ptr.u08++ = 0x5f;
-	*hwsq->ptr.u08++ = v0;
-	*hwsq->ptr.u08++ = v1;
-}
-
-static inline void
-hwsq_wr32(struct hwsq_ucode *hwsq, u32 reg, u32 val)
-{
-	if (val != hwsq->val) {
-		if ((val & 0xffff0000) == (hwsq->val & 0xffff0000)) {
-			*hwsq->ptr.u08++ = 0x42;
-			*hwsq->ptr.u16++ = (val & 0x0000ffff);
-		} else {
-			*hwsq->ptr.u08++ = 0xe2;
-			*hwsq->ptr.u32++ = val;
-		}
-
-		hwsq->val = val;
-	}
-
-	if ((reg & 0xffff0000) == (hwsq->reg & 0xffff0000)) {
-		*hwsq->ptr.u08++ = 0x40;
-		*hwsq->ptr.u16++ = (reg & 0x0000ffff);
-	} else {
-		*hwsq->ptr.u08++ = 0xe0;
-		*hwsq->ptr.u32++ = reg;
-	}
-	hwsq->reg = reg;
-}
-
-#endif
diff --git a/drivers/gpu/drm/nouveau/nouveau_mem.c b/drivers/gpu/drm/nouveau/nouveau_mem.c
deleted file mode 100644
index 4f6a572f2258..000000000000
--- a/drivers/gpu/drm/nouveau/nouveau_mem.c
+++ /dev/null
@@ -1,647 +0,0 @@
-/*
- * Copyright (C) The Weather Channel, Inc.  2002.  All Rights Reserved.
- * Copyright 2005 Stephane Marchesin
- *
- * The Weather Channel (TM) funded Tungsten Graphics to develop the
- * initial release of the Radeon 8500 driver under the XFree86 license.
- * This notice must be preserved.
- *
- * Permission is hereby granted, free of charge, to any person obtaining a
- * copy of this software and associated documentation files (the "Software"),
- * to deal in the Software without restriction, including without limitation
- * the rights to use, copy, modify, merge, publish, distribute, sublicense,
- * and/or sell copies of the Software, and to permit persons to whom the
- * Software is furnished to do so, subject to the following conditions:
- *
- * The above copyright notice and this permission notice (including the next
- * paragraph) shall be included in all copies or substantial portions of the
- * Software.
- *
- * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
- * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
- * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT.  IN NO EVENT SHALL
- * THE AUTHORS AND/OR THEIR SUPPLIERS BE LIABLE FOR ANY CLAIM, DAMAGES OR
- * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE,
- * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER
- * DEALINGS IN THE SOFTWARE.
- *
- * Authors:
- *    Ben Skeggs <bskeggs@redhat.com>
- *    Roy Spliet <r.spliet@student.tudelft.nl>
- */
-
-#include "nouveau_drm.h"
-#include "nouveau_pm.h"
-
-#include <subdev/fb.h>
-
-static int
-nv40_mem_timing_calc(struct drm_device *dev, u32 freq,
-		     struct nouveau_pm_tbl_entry *e, u8 len,
-		     struct nouveau_pm_memtiming *boot,
-		     struct nouveau_pm_memtiming *t)
-{
-	struct nouveau_drm *drm = nouveau_drm(dev);
-
-	t->reg[0] = (e->tRP << 24 | e->tRAS << 16 | e->tRFC << 8 | e->tRC);
-
-	/* XXX: I don't trust the -1's and +1's... they must come
-	 *      from somewhere! */
-	t->reg[1] = (e->tWR + 2 + (t->tCWL - 1)) << 24 |
-		    1 << 16 |
-		    (e->tWTR + 2 + (t->tCWL - 1)) << 8 |
-		    (e->tCL + 2 - (t->tCWL - 1));
-
-	t->reg[2] = 0x20200000 |
-		    ((t->tCWL - 1) << 24 |
-		     e->tRRD << 16 |
-		     e->tRCDWR << 8 |
-		     e->tRCDRD);
-
-	NV_DEBUG(drm, "Entry %d: 220: %08x %08x %08x\n", t->id,
-		 t->reg[0], t->reg[1], t->reg[2]);
-	return 0;
-}
-
-static int
-nv50_mem_timing_calc(struct drm_device *dev, u32 freq,
-		     struct nouveau_pm_tbl_entry *e, u8 len,
-		     struct nouveau_pm_memtiming *boot,
-		     struct nouveau_pm_memtiming *t)
-{
-	struct nouveau_device *device = nouveau_dev(dev);
-	struct nouveau_fb *pfb = nouveau_fb(device);
-	struct nouveau_drm *drm = nouveau_drm(dev);
-	struct bit_entry P;
-	uint8_t unk18 = 1, unk20 = 0, unk21 = 0, tmp7_3;
-
-	if (bit_table(dev, 'P', &P))
-		return -EINVAL;
-
-	switch (min(len, (u8) 22)) {
-	case 22:
-		unk21 = e->tUNK_21;
-	case 21:
-		unk20 = e->tUNK_20;
-	case 20:
-		if (e->tCWL > 0)
-			t->tCWL = e->tCWL;
-	case 19:
-		unk18 = e->tUNK_18;
-		break;
-	}
-
-	t->reg[0] = (e->tRP << 24 | e->tRAS << 16 | e->tRFC << 8 | e->tRC);
-
-	t->reg[1] = (e->tWR + 2 + (t->tCWL - 1)) << 24 |
-				max(unk18, (u8) 1) << 16 |
-				(e->tWTR + 2 + (t->tCWL - 1)) << 8;
-
-	t->reg[2] = ((t->tCWL - 1) << 24 |
-		    e->tRRD << 16 |
-		    e->tRCDWR << 8 |
-		    e->tRCDRD);
-
-	t->reg[4] = e->tUNK_13 << 8  | e->tUNK_13;
-
-	t->reg[5] = (e->tRFC << 24 | max(e->tRCDRD, e->tRCDWR) << 16 | e->tRP);
-
-	t->reg[8] = boot->reg[8] & 0xffffff00;
-
-	if (P.version == 1) {
-		t->reg[1] |= (e->tCL + 2 - (t->tCWL - 1));
-
-		t->reg[3] = (0x14 + e->tCL) << 24 |
-			    0x16 << 16 |
-			    (e->tCL - 1) << 8 |
-			    (e->tCL - 1);
-
-		t->reg[4] |= boot->reg[4] & 0xffff0000;
-
-		t->reg[6] = (0x33 - t->tCWL) << 16 |
-			    t->tCWL << 8 |
-			    (0x2e + e->tCL - t->tCWL);
-
-		t->reg[7] = 0x4000202 | (e->tCL - 1) << 16;
-
-		/* XXX: P.version == 1 only has DDR2 and GDDR3? */
-		if (pfb->ram->type == NV_MEM_TYPE_DDR2) {
-			t->reg[5] |= (e->tCL + 3) << 8;
-			t->reg[6] |= (t->tCWL - 2) << 8;
-			t->reg[8] |= (e->tCL - 4);
-		} else {
-			t->reg[5] |= (e->tCL + 2) << 8;
-			t->reg[6] |= t->tCWL << 8;
-			t->reg[8] |= (e->tCL - 2);
-		}
-	} else {
-		t->reg[1] |= (5 + e->tCL - (t->tCWL));
-
-		/* XXX: 0xb? 0x30? */
-		t->reg[3] = (0x30 + e->tCL) << 24 |
-			    (boot->reg[3] & 0x00ff0000)|
-			    (0xb + e->tCL) << 8 |
-			    (e->tCL - 1);
-
-		t->reg[4] |= (unk20 << 24 | unk21 << 16);
-
-		/* XXX: +6? */
-		t->reg[5] |= (t->tCWL + 6) << 8;
-
-		t->reg[6] = (0x5a + e->tCL) << 16 |
-			    (6 - e->tCL + t->tCWL) << 8 |
-			    (0x50 + e->tCL - t->tCWL);
-
-		tmp7_3 = (boot->reg[7] & 0xff000000) >> 24;
-		t->reg[7] = (tmp7_3 << 24) |
-			    ((tmp7_3 - 6 + e->tCL) << 16) |
-			    0x202;
-	}
-
-	NV_DEBUG(drm, "Entry %d: 220: %08x %08x %08x %08x\n", t->id,
-		 t->reg[0], t->reg[1], t->reg[2], t->reg[3]);
-	NV_DEBUG(drm, "         230: %08x %08x %08x %08x\n",
-		 t->reg[4], t->reg[5], t->reg[6], t->reg[7]);
-	NV_DEBUG(drm, "         240: %08x\n", t->reg[8]);
-	return 0;
-}
-
-static int
-nvc0_mem_timing_calc(struct drm_device *dev, u32 freq,
-		     struct nouveau_pm_tbl_entry *e, u8 len,
-		     struct nouveau_pm_memtiming *boot,
-		     struct nouveau_pm_memtiming *t)
-{
-	struct nouveau_drm *drm = nouveau_drm(dev);
-
-	if (e->tCWL > 0)
-		t->tCWL = e->tCWL;
-
-	t->reg[0] = (e->tRP << 24 | (e->tRAS & 0x7f) << 17 |
-		     e->tRFC << 8 | e->tRC);
-
-	t->reg[1] = (boot->reg[1] & 0xff000000) |
-		    (e->tRCDWR & 0x0f) << 20 |
-		    (e->tRCDRD & 0x0f) << 14 |
-		    (t->tCWL << 7) |
-		    (e->tCL & 0x0f);
-
-	t->reg[2] = (boot->reg[2] & 0xff0000ff) |
-		    e->tWR << 16 | e->tWTR << 8;
-
-	t->reg[3] = (e->tUNK_20 & 0x1f) << 9 |
-		    (e->tUNK_21 & 0xf) << 5 |
-		    (e->tUNK_13 & 0x1f);
-
-	t->reg[4] = (boot->reg[4] & 0xfff00fff) |
-		    (e->tRRD&0x1f) << 15;
-
-	NV_DEBUG(drm, "Entry %d: 290: %08x %08x %08x %08x\n", t->id,
-		 t->reg[0], t->reg[1], t->reg[2], t->reg[3]);
-	NV_DEBUG(drm, "         2a0: %08x\n", t->reg[4]);
-	return 0;
-}
-
-/**
- * MR generation methods
- */
-
-static int
-nouveau_mem_ddr2_mr(struct drm_device *dev, u32 freq,
-		    struct nouveau_pm_tbl_entry *e, u8 len,
-		    struct nouveau_pm_memtiming *boot,
-		    struct nouveau_pm_memtiming *t)
-{
-	struct nouveau_drm *drm = nouveau_drm(dev);
-
-	t->drive_strength = 0;
-	if (len < 15) {
-		t->odt = boot->odt;
-	} else {
-		t->odt = e->RAM_FT1 & 0x07;
-	}
-
-	if (e->tCL >= NV_MEM_CL_DDR2_MAX) {
-		NV_WARN(drm, "(%u) Invalid tCL: %u", t->id, e->tCL);
-		return -ERANGE;
-	}
-
-	if (e->tWR >= NV_MEM_WR_DDR2_MAX) {
-		NV_WARN(drm, "(%u) Invalid tWR: %u", t->id, e->tWR);
-		return -ERANGE;
-	}
-
-	if (t->odt > 3) {
-		NV_WARN(drm, "(%u) Invalid odt value, assuming disabled: %x",
-			t->id, t->odt);
-		t->odt = 0;
-	}
-
-	t->mr[0] = (boot->mr[0] & 0x100f) |
-		   (e->tCL) << 4 |
-		   (e->tWR - 1) << 9;
-	t->mr[1] = (boot->mr[1] & 0x101fbb) |
-		   (t->odt & 0x1) << 2 |
-		   (t->odt & 0x2) << 5;
-
-	NV_DEBUG(drm, "(%u) MR: %08x", t->id, t->mr[0]);
-	return 0;
-}
-
-static const uint8_t nv_mem_wr_lut_ddr3[NV_MEM_WR_DDR3_MAX] = {
-	0, 0, 0, 0, 0, 1, 2, 3, 4, 5, 5, 6, 6, 7, 7, 0, 0};
-
-static int
-nouveau_mem_ddr3_mr(struct drm_device *dev, u32 freq,
-		    struct nouveau_pm_tbl_entry *e, u8 len,
-		    struct nouveau_pm_memtiming *boot,
-		    struct nouveau_pm_memtiming *t)
-{
-	struct nouveau_drm *drm = nouveau_drm(dev);
-	u8 cl = e->tCL - 4;
-
-	t->drive_strength = 0;
-	if (len < 15) {
-		t->odt = boot->odt;
-	} else {
-		t->odt = e->RAM_FT1 & 0x07;
-	}
-
-	if (e->tCL >= NV_MEM_CL_DDR3_MAX || e->tCL < 4) {
-		NV_WARN(drm, "(%u) Invalid tCL: %u", t->id, e->tCL);
-		return -ERANGE;
-	}
-
-	if (e->tWR >= NV_MEM_WR_DDR3_MAX || e->tWR < 4) {
-		NV_WARN(drm, "(%u) Invalid tWR: %u", t->id, e->tWR);
-		return -ERANGE;
-	}
-
-	if (e->tCWL < 5) {
-		NV_WARN(drm, "(%u) Invalid tCWL: %u", t->id, e->tCWL);
-		return -ERANGE;
-	}
-
-	t->mr[0] = (boot->mr[0] & 0x180b) |
-		   /* CAS */
-		   (cl & 0x7) << 4 |
-		   (cl & 0x8) >> 1 |
-		   (nv_mem_wr_lut_ddr3[e->tWR]) << 9;
-	t->mr[1] = (boot->mr[1] & 0x101dbb) |
-		   (t->odt & 0x1) << 2 |
-		   (t->odt & 0x2) << 5 |
-		   (t->odt & 0x4) << 7;
-	t->mr[2] = (boot->mr[2] & 0x20ffb7) | (e->tCWL - 5) << 3;
-
-	NV_DEBUG(drm, "(%u) MR: %08x %08x", t->id, t->mr[0], t->mr[2]);
-	return 0;
-}
-
-static const uint8_t nv_mem_cl_lut_gddr3[NV_MEM_CL_GDDR3_MAX] = {
-	0, 0, 0, 0, 4, 5, 6, 7, 0, 1, 2, 3, 8, 9, 10, 11};
-static const uint8_t nv_mem_wr_lut_gddr3[NV_MEM_WR_GDDR3_MAX] = {
-	0, 0, 0, 0, 0, 2, 3, 8, 9, 10, 11, 0, 0, 1, 1, 0, 3};
-
-static int
-nouveau_mem_gddr3_mr(struct drm_device *dev, u32 freq,
-		     struct nouveau_pm_tbl_entry *e, u8 len,
-		     struct nouveau_pm_memtiming *boot,
-		     struct nouveau_pm_memtiming *t)
-{
-	struct nouveau_drm *drm = nouveau_drm(dev);
-
-	if (len < 15) {
-		t->drive_strength = boot->drive_strength;
-		t->odt = boot->odt;
-	} else {
-		t->drive_strength = (e->RAM_FT1 & 0x30) >> 4;
-		t->odt = e->RAM_FT1 & 0x07;
-	}
-
-	if (e->tCL >= NV_MEM_CL_GDDR3_MAX) {
-		NV_WARN(drm, "(%u) Invalid tCL: %u", t->id, e->tCL);
-		return -ERANGE;
-	}
-
-	if (e->tWR >= NV_MEM_WR_GDDR3_MAX) {
-		NV_WARN(drm, "(%u) Invalid tWR: %u", t->id, e->tWR);
-		return -ERANGE;
-	}
-
-	if (t->odt > 3) {
-		NV_WARN(drm, "(%u) Invalid odt value, assuming autocal: %x",
-			t->id, t->odt);
-		t->odt = 0;
-	}
-
-	t->mr[0] = (boot->mr[0] & 0xe0b) |
-		   /* CAS */
-		   ((nv_mem_cl_lut_gddr3[e->tCL] & 0x7) << 4) |
-		   ((nv_mem_cl_lut_gddr3[e->tCL] & 0x8) >> 2);
-	t->mr[1] = (boot->mr[1] & 0x100f40) | t->drive_strength |
-		   (t->odt << 2) |
-		   (nv_mem_wr_lut_gddr3[e->tWR] & 0xf) << 4;
-	t->mr[2] = boot->mr[2];
-
-	NV_DEBUG(drm, "(%u) MR: %08x %08x %08x", t->id,
-		      t->mr[0], t->mr[1], t->mr[2]);
-	return 0;
-}
-
-static int
-nouveau_mem_gddr5_mr(struct drm_device *dev, u32 freq,
-		     struct nouveau_pm_tbl_entry *e, u8 len,
-		     struct nouveau_pm_memtiming *boot,
-		     struct nouveau_pm_memtiming *t)
-{
-	struct nouveau_drm *drm = nouveau_drm(dev);
-
-	if (len < 15) {
-		t->drive_strength = boot->drive_strength;
-		t->odt = boot->odt;
-	} else {
-		t->drive_strength = (e->RAM_FT1 & 0x30) >> 4;
-		t->odt = e->RAM_FT1 & 0x03;
-	}
-
-	if (e->tCL >= NV_MEM_CL_GDDR5_MAX) {
-		NV_WARN(drm, "(%u) Invalid tCL: %u", t->id, e->tCL);
-		return -ERANGE;
-	}
-
-	if (e->tWR >= NV_MEM_WR_GDDR5_MAX) {
-		NV_WARN(drm, "(%u) Invalid tWR: %u", t->id, e->tWR);
-		return -ERANGE;
-	}
-
-	if (t->odt > 3) {
-		NV_WARN(drm, "(%u) Invalid odt value, assuming autocal: %x",
-			t->id, t->odt);
-		t->odt = 0;
-	}
-
-	t->mr[0] = (boot->mr[0] & 0x007) |
-		   ((e->tCL - 5) << 3) |
-		   ((e->tWR - 4) << 8);
-	t->mr[1] = (boot->mr[1] & 0x1007f0) |
-		   t->drive_strength |
-		   (t->odt << 2);
-
-	NV_DEBUG(drm, "(%u) MR: %08x %08x", t->id, t->mr[0], t->mr[1]);
-	return 0;
-}
-
-int
-nouveau_mem_timing_calc(struct drm_device *dev, u32 freq,
-			struct nouveau_pm_memtiming *t)
-{
-	struct nouveau_device *device = nouveau_dev(dev);
-	struct nouveau_fb *pfb = nouveau_fb(device);
-	struct nouveau_pm *pm = nouveau_pm(dev);
-	struct nouveau_pm_memtiming *boot = &pm->boot.timing;
-	struct nouveau_pm_tbl_entry *e;
-	u8 ver, len, *ptr, *ramcfg;
-	int ret;
-
-	ptr = nouveau_perf_timing(dev, freq, &ver, &len);
-	if (!ptr || ptr[0] == 0x00) {
-		*t = *boot;
-		return 0;
-	}
-	e = (struct nouveau_pm_tbl_entry *)ptr;
-
-	t->tCWL = boot->tCWL;
-
-	switch (device->card_type) {
-	case NV_40:
-		ret = nv40_mem_timing_calc(dev, freq, e, len, boot, t);
-		break;
-	case NV_50:
-		ret = nv50_mem_timing_calc(dev, freq, e, len, boot, t);
-		break;
-	case NV_C0:
-	case NV_D0:
-		ret = nvc0_mem_timing_calc(dev, freq, e, len, boot, t);
-		break;
-	default:
-		ret = -ENODEV;
-		break;
-	}
-
-	switch (pfb->ram->type * !ret) {
-	case NV_MEM_TYPE_GDDR3:
-		ret = nouveau_mem_gddr3_mr(dev, freq, e, len, boot, t);
-		break;
-	case NV_MEM_TYPE_GDDR5:
-		ret = nouveau_mem_gddr5_mr(dev, freq, e, len, boot, t);
-		break;
-	case NV_MEM_TYPE_DDR2:
-		ret = nouveau_mem_ddr2_mr(dev, freq, e, len, boot, t);
-		break;
-	case NV_MEM_TYPE_DDR3:
-		ret = nouveau_mem_ddr3_mr(dev, freq, e, len, boot, t);
-		break;
-	default:
-		ret = -EINVAL;
-		break;
-	}
-
-	ramcfg = nouveau_perf_ramcfg(dev, freq, &ver, &len);
-	if (ramcfg) {
-		int dll_off;
-
-		if (ver == 0x00)
-			dll_off = !!(ramcfg[3] & 0x04);
-		else
-			dll_off = !!(ramcfg[2] & 0x40);
-
-		switch (pfb->ram->type) {
-		case NV_MEM_TYPE_GDDR3:
-			t->mr[1] &= ~0x00000040;
-			t->mr[1] |=  0x00000040 * dll_off;
-			break;
-		default:
-			t->mr[1] &= ~0x00000001;
-			t->mr[1] |=  0x00000001 * dll_off;
-			break;
-		}
-	}
-
-	return ret;
-}
-
-void
-nouveau_mem_timing_read(struct drm_device *dev, struct nouveau_pm_memtiming *t)
-{
-	struct nouveau_device *device = nouveau_dev(dev);
-	struct nouveau_fb *pfb = nouveau_fb(device);
-	u32 timing_base, timing_regs, mr_base;
-	int i;
-
-	if (device->card_type >= 0xC0) {
-		timing_base = 0x10f290;
-		mr_base = 0x10f300;
-	} else {
-		timing_base = 0x100220;
-		mr_base = 0x1002c0;
-	}
-
-	t->id = -1;
-
-	switch (device->card_type) {
-	case NV_50:
-		timing_regs = 9;
-		break;
-	case NV_C0:
-	case NV_D0:
-		timing_regs = 5;
-		break;
-	case NV_30:
-	case NV_40:
-		timing_regs = 3;
-		break;
-	default:
-		timing_regs = 0;
-		return;
-	}
-	for(i = 0; i < timing_regs; i++)
-		t->reg[i] = nv_rd32(device, timing_base + (0x04 * i));
-
-	t->tCWL = 0;
-	if (device->card_type < NV_C0) {
-		t->tCWL = ((nv_rd32(device, 0x100228) & 0x0f000000) >> 24) + 1;
-	} else if (device->card_type <= NV_D0) {
-		t->tCWL = ((nv_rd32(device, 0x10f294) & 0x00000f80) >> 7);
-	}
-
-	t->mr[0] = nv_rd32(device, mr_base);
-	t->mr[1] = nv_rd32(device, mr_base + 0x04);
-	t->mr[2] = nv_rd32(device, mr_base + 0x20);
-	t->mr[3] = nv_rd32(device, mr_base + 0x24);
-
-	t->odt = 0;
-	t->drive_strength = 0;
-
-	switch (pfb->ram->type) {
-	case NV_MEM_TYPE_DDR3:
-		t->odt |= (t->mr[1] & 0x200) >> 7;
-	case NV_MEM_TYPE_DDR2:
-		t->odt |= (t->mr[1] & 0x04) >> 2 |
-			  (t->mr[1] & 0x40) >> 5;
-		break;
-	case NV_MEM_TYPE_GDDR3:
-	case NV_MEM_TYPE_GDDR5:
-		t->drive_strength = t->mr[1] & 0x03;
-		t->odt = (t->mr[1] & 0x0c) >> 2;
-		break;
-	default:
-		break;
-	}
-}
-
-int
-nouveau_mem_exec(struct nouveau_mem_exec_func *exec,
-		 struct nouveau_pm_level *perflvl)
-{
-	struct nouveau_drm *drm = nouveau_drm(exec->dev);
-	struct nouveau_device *device = nouveau_dev(exec->dev);
-	struct nouveau_fb *pfb = nouveau_fb(device);
-	struct nouveau_pm_memtiming *info = &perflvl->timing;
-	u32 tMRD = 1000, tCKSRE = 0, tCKSRX = 0, tXS = 0, tDLLK = 0;
-	u32 mr[3] = { info->mr[0], info->mr[1], info->mr[2] };
-	u32 mr1_dlloff;
-
-	switch (pfb->ram->type) {
-	case NV_MEM_TYPE_DDR2:
-		tDLLK = 2000;
-		mr1_dlloff = 0x00000001;
-		break;
-	case NV_MEM_TYPE_DDR3:
-		tDLLK = 12000;
-		tCKSRE = 2000;
-		tXS = 1000;
-		mr1_dlloff = 0x00000001;
-		break;
-	case NV_MEM_TYPE_GDDR3:
-		tDLLK = 40000;
-		mr1_dlloff = 0x00000040;
-		break;
-	default:
-		NV_ERROR(drm, "cannot reclock unsupported memtype\n");
-		return -ENODEV;
-	}
-
-	/* fetch current MRs */
-	switch (pfb->ram->type) {
-	case NV_MEM_TYPE_GDDR3:
-	case NV_MEM_TYPE_DDR3:
-		mr[2] = exec->mrg(exec, 2);
-	default:
-		mr[1] = exec->mrg(exec, 1);
-		mr[0] = exec->mrg(exec, 0);
-		break;
-	}
-
-	/* DLL 'on' -> DLL 'off' mode, disable before entering self-refresh  */
-	if (!(mr[1] & mr1_dlloff) && (info->mr[1] & mr1_dlloff)) {
-		exec->precharge(exec);
-		exec->mrs (exec, 1, mr[1] | mr1_dlloff);
-		exec->wait(exec, tMRD);
-	}
-
-	/* enter self-refresh mode */
-	exec->precharge(exec);
-	exec->refresh(exec);
-	exec->refresh(exec);
-	exec->refresh_auto(exec, false);
-	exec->refresh_self(exec, true);
-	exec->wait(exec, tCKSRE);
-
-	/* modify input clock frequency */
-	exec->clock_set(exec);
-
-	/* exit self-refresh mode */
-	exec->wait(exec, tCKSRX);
-	exec->precharge(exec);
-	exec->refresh_self(exec, false);
-	exec->refresh_auto(exec, true);
-	exec->wait(exec, tXS);
-	exec->wait(exec, tXS);
-
-	/* update MRs */
-	if (mr[2] != info->mr[2]) {
-		exec->mrs (exec, 2, info->mr[2]);
-		exec->wait(exec, tMRD);
-	}
-
-	if (mr[1] != info->mr[1]) {
-		/* need to keep DLL off until later, at least on GDDR3 */
-		exec->mrs (exec, 1, info->mr[1] | (mr[1] & mr1_dlloff));
-		exec->wait(exec, tMRD);
-	}
-
-	if (mr[0] != info->mr[0]) {
-		exec->mrs (exec, 0, info->mr[0]);
-		exec->wait(exec, tMRD);
-	}
-
-	/* update PFB timing registers */
-	exec->timing_set(exec);
-
-	/* DLL (enable + ) reset */
-	if (!(info->mr[1] & mr1_dlloff)) {
-		if (mr[1] & mr1_dlloff) {
-			exec->mrs (exec, 1, info->mr[1]);
-			exec->wait(exec, tMRD);
-		}
-		exec->mrs (exec, 0, info->mr[0] | 0x00000100);
-		exec->wait(exec, tMRD);
-		exec->mrs (exec, 0, info->mr[0] | 0x00000000);
-		exec->wait(exec, tMRD);
-		exec->wait(exec, tDLLK);
-		if (pfb->ram->type == NV_MEM_TYPE_GDDR3)
-			exec->precharge(exec);
-	}
-
-	return 0;
-}
diff --git a/drivers/gpu/drm/nouveau/nouveau_perf.c b/drivers/gpu/drm/nouveau/nouveau_perf.c
deleted file mode 100644
index 4fe883c54918..000000000000
--- a/drivers/gpu/drm/nouveau/nouveau_perf.c
+++ /dev/null
@@ -1,416 +0,0 @@
-/*
- * Copyright 2010 Red Hat Inc.
- *
- * Permission is hereby granted, free of charge, to any person obtaining a
- * copy of this software and associated documentation files (the "Software"),
- * to deal in the Software without restriction, including without limitation
- * the rights to use, copy, modify, merge, publish, distribute, sublicense,
- * and/or sell copies of the Software, and to permit persons to whom the
- * Software is furnished to do so, subject to the following conditions:
- *
- * The above copyright notice and this permission notice shall be included in
- * all copies or substantial portions of the Software.
- *
- * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
- * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
- * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT.  IN NO EVENT SHALL
- * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR
- * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE,
- * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR
- * OTHER DEALINGS IN THE SOFTWARE.
- *
- * Authors: Ben Skeggs
- */
-
-#include <drm/drmP.h>
-
-#include "nouveau_drm.h"
-#include "nouveau_reg.h"
-#include "nouveau_pm.h"
-
-static u8 *
-nouveau_perf_table(struct drm_device *dev, u8 *ver)
-{
-	struct nouveau_drm *drm = nouveau_drm(dev);
-	struct nvbios *bios = &drm->vbios;
-	struct bit_entry P;
-
-	if (!bit_table(dev, 'P', &P) && P.version && P.version <= 2) {
-		u8 *perf = ROMPTR(dev, P.data[0]);
-		if (perf) {
-			*ver = perf[0];
-			return perf;
-		}
-	}
-
-	if (bios->type == NVBIOS_BMP) {
-		if (bios->data[bios->offset + 6] >= 0x25) {
-			u8 *perf = ROMPTR(dev, bios->data[bios->offset + 0x94]);
-			if (perf) {
-				*ver = perf[1];
-				return perf;
-			}
-		}
-	}
-
-	return NULL;
-}
-
-static u8 *
-nouveau_perf_entry(struct drm_device *dev, int idx,
-		   u8 *ver, u8 *hdr, u8 *cnt, u8 *len)
-{
-	u8 *perf = nouveau_perf_table(dev, ver);
-	if (perf) {
-		if (*ver >= 0x12 && *ver < 0x20 && idx < perf[2]) {
-			*hdr = perf[3];
-			*cnt = 0;
-			*len = 0;
-			return perf + perf[0] + idx * perf[3];
-		} else
-		if (*ver >= 0x20 && *ver < 0x40 && idx < perf[2]) {
-			*hdr = perf[3];
-			*cnt = perf[4];
-			*len = perf[5];
-			return perf + perf[1] + idx * (*hdr + (*cnt * *len));
-		} else
-		if (*ver >= 0x40 && *ver < 0x41 && idx < perf[5]) {
-			*hdr = perf[2];
-			*cnt = perf[4];
-			*len = perf[3];
-			return perf + perf[1] + idx * (*hdr + (*cnt * *len));
-		}
-	}
-	return NULL;
-}
-
-u8 *
-nouveau_perf_rammap(struct drm_device *dev, u32 freq,
-		    u8 *ver, u8 *hdr, u8 *cnt, u8 *len)
-{
-	struct nouveau_drm *drm = nouveau_drm(dev);
-	struct bit_entry P;
-	u8 *perf, i = 0;
-
-	if (!bit_table(dev, 'P', &P) && P.version == 2) {
-		u8 *rammap = ROMPTR(dev, P.data[4]);
-		if (rammap) {
-			u8 *ramcfg = rammap + rammap[1];
-
-			*ver = rammap[0];
-			*hdr = rammap[2];
-			*cnt = rammap[4];
-			*len = rammap[3];
-
-			freq /= 1000;
-			for (i = 0; i < rammap[5]; i++) {
-				if (freq >= ROM16(ramcfg[0]) &&
-				    freq <= ROM16(ramcfg[2]))
-					return ramcfg;
-
-				ramcfg += *hdr + (*cnt * *len);
-			}
-		}
-
-		return NULL;
-	}
-
-	if (nv_device(drm->device)->chipset == 0x49 ||
-	    nv_device(drm->device)->chipset == 0x4b)
-		freq /= 2;
-
-	while ((perf = nouveau_perf_entry(dev, i++, ver, hdr, cnt, len))) {
-		if (*ver >= 0x20 && *ver < 0x25) {
-			if (perf[0] != 0xff && freq <= ROM16(perf[11]) * 1000)
-				break;
-		} else
-		if (*ver >= 0x25 && *ver < 0x40) {
-			if (perf[0] != 0xff && freq <= ROM16(perf[12]) * 1000)
-				break;
-		}
-	}
-
-	if (perf) {
-		u8 *ramcfg = perf + *hdr;
-		*ver = 0x00;
-		*hdr = 0;
-		return ramcfg;
-	}
-
-	return NULL;
-}
-
-u8 *
-nouveau_perf_ramcfg(struct drm_device *dev, u32 freq, u8 *ver, u8 *len)
-{
-	struct nouveau_device *device = nouveau_dev(dev);
-	struct nouveau_drm *drm = nouveau_drm(dev);
-	struct nvbios *bios = &drm->vbios;
-	u8 strap, hdr, cnt;
-	u8 *rammap;
-
-	strap = (nv_rd32(device, 0x101000) & 0x0000003c) >> 2;
-	if (bios->ram_restrict_tbl_ptr)
-		strap = bios->data[bios->ram_restrict_tbl_ptr + strap];
-
-	rammap = nouveau_perf_rammap(dev, freq, ver, &hdr, &cnt, len);
-	if (rammap && strap < cnt)
-		return rammap + hdr + (strap * *len);
-
-	return NULL;
-}
-
-u8 *
-nouveau_perf_timing(struct drm_device *dev, u32 freq, u8 *ver, u8 *len)
-{
-	struct nouveau_drm *drm = nouveau_drm(dev);
-	struct nvbios *bios = &drm->vbios;
-	struct bit_entry P;
-	u8 *perf, *timing = NULL;
-	u8 i = 0, hdr, cnt;
-
-	if (bios->type == NVBIOS_BMP) {
-		while ((perf = nouveau_perf_entry(dev, i++, ver, &hdr, &cnt,
-						  len)) && *ver == 0x15) {
-			if (freq <= ROM32(perf[5]) * 20) {
-				*ver = 0x00;
-				*len = 14;
-				return perf + 41;
-			}
-		}
-		return NULL;
-	}
-
-	if (!bit_table(dev, 'P', &P)) {
-		if (P.version == 1)
-			timing = ROMPTR(dev, P.data[4]);
-		else
-		if (P.version == 2)
-			timing = ROMPTR(dev, P.data[8]);
-	}
-
-	if (timing && timing[0] == 0x10) {
-		u8 *ramcfg = nouveau_perf_ramcfg(dev, freq, ver, len);
-		if (ramcfg && ramcfg[1] < timing[2]) {
-			*ver = timing[0];
-			*len = timing[3];
-			return timing + timing[1] + (ramcfg[1] * timing[3]);
-		}
-	}
-
-	return NULL;
-}
-
-static void
-legacy_perf_init(struct drm_device *dev)
-{
-	struct nouveau_device *device = nouveau_dev(dev);
-	struct nouveau_drm *drm = nouveau_drm(dev);
-	struct nvbios *bios = &drm->vbios;
-	struct nouveau_pm *pm = nouveau_pm(dev);
-	char *perf, *entry, *bmp = &bios->data[bios->offset];
-	int headerlen, use_straps;
-
-	if (bmp[5] < 0x5 || bmp[6] < 0x14) {
-		NV_DEBUG(drm, "BMP version too old for perf\n");
-		return;
-	}
-
-	perf = ROMPTR(dev, bmp[0x73]);
-	if (!perf) {
-		NV_DEBUG(drm, "No memclock table pointer found.\n");
-		return;
-	}
-
-	switch (perf[0]) {
-	case 0x12:
-	case 0x14:
-	case 0x18:
-		use_straps = 0;
-		headerlen = 1;
-		break;
-	case 0x01:
-		use_straps = perf[1] & 1;
-		headerlen = (use_straps ? 8 : 2);
-		break;
-	default:
-		NV_WARN(drm, "Unknown memclock table version %x.\n", perf[0]);
-		return;
-	}
-
-	entry = perf + headerlen;
-	if (use_straps)
-		entry += (nv_rd32(device, NV_PEXTDEV_BOOT_0) & 0x3c) >> 1;
-
-	sprintf(pm->perflvl[0].name, "performance_level_0");
-	pm->perflvl[0].memory = ROM16(entry[0]) * 20;
-	pm->nr_perflvl = 1;
-}
-
-static void
-nouveau_perf_voltage(struct drm_device *dev, struct nouveau_pm_level *perflvl)
-{
-	struct nouveau_drm *drm = nouveau_drm(dev);
-	struct bit_entry P;
-	u8 *vmap;
-	int id;
-
-	id = perflvl->volt_min;
-	perflvl->volt_min = 0;
-
-	/* boards using voltage table version <0x40 store the voltage
-	 * level directly in the perflvl entry as a multiple of 10mV
-	 */
-	if (drm->pm->voltage.version < 0x40) {
-		perflvl->volt_min = id * 10000;
-		perflvl->volt_max = perflvl->volt_min;
-		return;
-	}
-
-	/* on newer ones, the perflvl stores an index into yet another
-	 * vbios table containing a min/max voltage value for the perflvl
-	 */
-	if (bit_table(dev, 'P', &P) || P.version != 2 || P.length < 34) {
-		NV_DEBUG(drm, "where's our volt map table ptr? %d %d\n",
-			 P.version, P.length);
-		return;
-	}
-
-	vmap = ROMPTR(dev, P.data[32]);
-	if (!vmap) {
-		NV_DEBUG(drm, "volt map table pointer invalid\n");
-		return;
-	}
-
-	if (id < vmap[3]) {
-		vmap += vmap[1] + (vmap[2] * id);
-		perflvl->volt_min = ROM32(vmap[0]);
-		perflvl->volt_max = ROM32(vmap[4]);
-	}
-}
-
-void
-nouveau_perf_init(struct drm_device *dev)
-{
-	struct nouveau_drm *drm = nouveau_drm(dev);
-	struct nouveau_pm *pm = nouveau_pm(dev);
-	struct nvbios *bios = &drm->vbios;
-	u8 *perf, ver, hdr, cnt, len;
-	int ret, vid, i = -1;
-
-	if (bios->type == NVBIOS_BMP && bios->data[bios->offset + 6] < 0x25) {
-		legacy_perf_init(dev);
-		return;
-	}
-
-	perf = nouveau_perf_table(dev, &ver);
-
-	while ((perf = nouveau_perf_entry(dev, ++i, &ver, &hdr, &cnt, &len))) {
-		struct nouveau_pm_level *perflvl = &pm->perflvl[pm->nr_perflvl];
-
-		if (perf[0] == 0xff)
-			continue;
-
-		switch (ver) {
-		case 0x12:
-		case 0x13:
-		case 0x15:
-			perflvl->fanspeed = perf[55];
-			if (hdr > 56)
-				perflvl->volt_min = perf[56];
-			perflvl->core = ROM32(perf[1]) * 10;
-			perflvl->memory = ROM32(perf[5]) * 20;
-			break;
-		case 0x21:
-		case 0x23:
-		case 0x24:
-			perflvl->fanspeed = perf[4];
-			perflvl->volt_min = perf[5];
-			perflvl->shader = ROM16(perf[6]) * 1000;
-			perflvl->core = perflvl->shader;
-			perflvl->core += (signed char)perf[8] * 1000;
-			if (nv_device(drm->device)->chipset == 0x49 ||
-			    nv_device(drm->device)->chipset == 0x4b)
-				perflvl->memory = ROM16(perf[11]) * 1000;
-			else
-				perflvl->memory = ROM16(perf[11]) * 2000;
-			break;
-		case 0x25:
-			perflvl->fanspeed = perf[4];
-			perflvl->volt_min = perf[5];
-			perflvl->core = ROM16(perf[6]) * 1000;
-			perflvl->shader = ROM16(perf[10]) * 1000;
-			perflvl->memory = ROM16(perf[12]) * 1000;
-			break;
-		case 0x30:
-			perflvl->memscript = ROM16(perf[2]);
-		case 0x35:
-			perflvl->fanspeed = perf[6];
-			perflvl->volt_min = perf[7];
-			perflvl->core = ROM16(perf[8]) * 1000;
-			perflvl->shader = ROM16(perf[10]) * 1000;
-			perflvl->memory = ROM16(perf[12]) * 1000;
-			perflvl->vdec = ROM16(perf[16]) * 1000;
-			perflvl->dom6 = ROM16(perf[20]) * 1000;
-			break;
-		case 0x40:
-#define subent(n) ((ROM16(perf[hdr + (n) * len]) & 0xfff) * 1000)
-			perflvl->fanspeed = 0; /*XXX*/
-			perflvl->volt_min = perf[2];
-			if (nv_device(drm->device)->card_type == NV_50) {
-				perflvl->core   = subent(0);
-				perflvl->shader = subent(1);
-				perflvl->memory = subent(2);
-				perflvl->vdec   = subent(3);
-				perflvl->unka0  = subent(4);
-			} else {
-				perflvl->hub06  = subent(0);
-				perflvl->hub01  = subent(1);
-				perflvl->copy   = subent(2);
-				perflvl->shader = subent(3);
-				perflvl->rop    = subent(4);
-				perflvl->memory = subent(5);
-				perflvl->vdec   = subent(6);
-				perflvl->daemon = subent(10);
-				perflvl->hub07  = subent(11);
-				perflvl->core   = perflvl->shader / 2;
-			}
-			break;
-		}
-
-		/* make sure vid is valid */
-		nouveau_perf_voltage(dev, perflvl);
-		if (pm->voltage.supported && perflvl->volt_min) {
-			vid = nouveau_volt_vid_lookup(dev, perflvl->volt_min);
-			if (vid < 0) {
-				NV_DEBUG(drm, "perflvl %d, bad vid\n", i);
-				continue;
-			}
-		}
-
-		/* get the corresponding memory timings */
-		ret = nouveau_mem_timing_calc(dev, perflvl->memory,
-					          &perflvl->timing);
-		if (ret) {
-			NV_DEBUG(drm, "perflvl %d, bad timing: %d\n", i, ret);
-			continue;
-		}
-
-		snprintf(perflvl->name, sizeof(perflvl->name),
-			 "performance_level_%d", i);
-		perflvl->id = i;
-
-		snprintf(perflvl->profile.name, sizeof(perflvl->profile.name),
-			 "%d", perflvl->id);
-		perflvl->profile.func = &nouveau_pm_static_profile_func;
-		list_add_tail(&perflvl->profile.head, &pm->profiles);
-
-
-		pm->nr_perflvl++;
-	}
-}
-
-void
-nouveau_perf_fini(struct drm_device *dev)
-{
-}
diff --git a/drivers/gpu/drm/nouveau/nouveau_pm.h b/drivers/gpu/drm/nouveau/nouveau_pm.h
deleted file mode 100644
index 73b789c230a9..000000000000
--- a/drivers/gpu/drm/nouveau/nouveau_pm.h
+++ /dev/null
@@ -1,283 +0,0 @@
-/*
- * Copyright 2010 Red Hat Inc.
- *
- * Permission is hereby granted, free of charge, to any person obtaining a
- * copy of this software and associated documentation files (the "Software"),
- * to deal in the Software without restriction, including without limitation
- * the rights to use, copy, modify, merge, publish, distribute, sublicense,
- * and/or sell copies of the Software, and to permit persons to whom the
- * Software is furnished to do so, subject to the following conditions:
- *
- * The above copyright notice and this permission notice shall be included in
- * all copies or substantial portions of the Software.
- *
- * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
- * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
- * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT.  IN NO EVENT SHALL
- * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR
- * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE,
- * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR
- * OTHER DEALINGS IN THE SOFTWARE.
- *
- * Authors: Ben Skeggs
- */
-
-#ifndef __NOUVEAU_PM_H__
-#define __NOUVEAU_PM_H__
-
-#include <subdev/bios/pll.h>
-#include <subdev/clock.h>
-
-struct nouveau_pm_voltage_level {
-	u32 voltage; /* microvolts */
-	u8  vid;
-};
-
-struct nouveau_pm_voltage {
-	bool supported;
-	u8 version;
-	u8 vid_mask;
-
-	struct nouveau_pm_voltage_level *level;
-	int nr_level;
-};
-
-/* Exclusive upper limits */
-#define NV_MEM_CL_DDR2_MAX 8
-#define NV_MEM_WR_DDR2_MAX 9
-#define NV_MEM_CL_DDR3_MAX 17
-#define NV_MEM_WR_DDR3_MAX 17
-#define NV_MEM_CL_GDDR3_MAX 16
-#define NV_MEM_WR_GDDR3_MAX 18
-#define NV_MEM_CL_GDDR5_MAX 21
-#define NV_MEM_WR_GDDR5_MAX 20
-
-struct nouveau_pm_memtiming {
-	int id;
-
-	u32 reg[9];
-	u32 mr[4];
-
-	u8 tCWL;
-
-	u8 odt;
-	u8 drive_strength;
-};
-
-struct nouveau_pm_tbl_header {
-	u8 version;
-	u8 header_len;
-	u8 entry_cnt;
-	u8 entry_len;
-};
-
-struct nouveau_pm_tbl_entry {
-	u8 tWR;
-	u8 tWTR;
-	u8 tCL;
-	u8 tRC;
-	u8 empty_4;
-	u8 tRFC;	/* Byte 5 */
-	u8 empty_6;
-	u8 tRAS;	/* Byte 7 */
-	u8 empty_8;
-	u8 tRP;		/* Byte 9 */
-	u8 tRCDRD;
-	u8 tRCDWR;
-	u8 tRRD;
-	u8 tUNK_13;
-	u8 RAM_FT1;		/* 14, a bitmask of random RAM features */
-	u8 empty_15;
-	u8 tUNK_16;
-	u8 empty_17;
-	u8 tUNK_18;
-	u8 tCWL;
-	u8 tUNK_20, tUNK_21;
-};
-
-struct nouveau_pm_profile;
-struct nouveau_pm_profile_func {
-	void (*destroy)(struct nouveau_pm_profile *);
-	void (*init)(struct nouveau_pm_profile *);
-	void (*fini)(struct nouveau_pm_profile *);
-	struct nouveau_pm_level *(*select)(struct nouveau_pm_profile *);
-};
-
-struct nouveau_pm_profile {
-	const struct nouveau_pm_profile_func *func;
-	struct list_head head;
-	char name[8];
-};
-
-#define NOUVEAU_PM_MAX_LEVEL 8
-struct nouveau_pm_level {
-	struct nouveau_pm_profile profile;
-	struct device_attribute dev_attr;
-	char name[32];
-	int id;
-
-	struct nouveau_pm_memtiming timing;
-	u32 memory;
-	u16 memscript;
-
-	u32 core;
-	u32 shader;
-	u32 rop;
-	u32 copy;
-	u32 daemon;
-	u32 vdec;
-	u32 dom6;
-	u32 unka0;	/* nva3:nvc0 */
-	u32 hub01;	/* nvc0- */
-	u32 hub06;	/* nvc0- */
-	u32 hub07;	/* nvc0- */
-
-	u32 volt_min; /* microvolts */
-	u32 volt_max;
-	u8  fanspeed;
-};
-
-struct nouveau_pm_temp_sensor_constants {
-	u16 offset_constant;
-	s16 offset_mult;
-	s16 offset_div;
-	s16 slope_mult;
-	s16 slope_div;
-};
-
-struct nouveau_pm_threshold_temp {
-	s16 critical;
-	s16 down_clock;
-};
-
-struct nouveau_pm {
-	struct drm_device *dev;
-
-	struct nouveau_pm_voltage voltage;
-	struct nouveau_pm_level perflvl[NOUVEAU_PM_MAX_LEVEL];
-	int nr_perflvl;
-	struct nouveau_pm_temp_sensor_constants sensor_constants;
-	struct nouveau_pm_threshold_temp threshold_temp;
-
-	struct nouveau_pm_profile *profile_ac;
-	struct nouveau_pm_profile *profile_dc;
-	struct nouveau_pm_profile *profile;
-	struct list_head profiles;
-
-	struct nouveau_pm_level boot;
-	struct nouveau_pm_level *cur;
-
-	struct device *hwmon;
-	struct notifier_block acpi_nb;
-
-	int  (*clocks_get)(struct drm_device *, struct nouveau_pm_level *);
-	void *(*clocks_pre)(struct drm_device *, struct nouveau_pm_level *);
-	int (*clocks_set)(struct drm_device *, void *);
-
-	int (*voltage_get)(struct drm_device *);
-	int (*voltage_set)(struct drm_device *, int voltage);
-};
-
-static inline struct nouveau_pm *
-nouveau_pm(struct drm_device *dev)
-{
-	return nouveau_drm(dev)->pm;
-}
-
-struct nouveau_mem_exec_func {
-	struct drm_device *dev;
-	void (*precharge)(struct nouveau_mem_exec_func *);
-	void (*refresh)(struct nouveau_mem_exec_func *);
-	void (*refresh_auto)(struct nouveau_mem_exec_func *, bool);
-	void (*refresh_self)(struct nouveau_mem_exec_func *, bool);
-	void (*wait)(struct nouveau_mem_exec_func *, u32 nsec);
-	u32  (*mrg)(struct nouveau_mem_exec_func *, int mr);
-	void (*mrs)(struct nouveau_mem_exec_func *, int mr, u32 data);
-	void (*clock_set)(struct nouveau_mem_exec_func *);
-	void (*timing_set)(struct nouveau_mem_exec_func *);
-	void *priv;
-};
-
-/* nouveau_mem.c */
-int  nouveau_mem_exec(struct nouveau_mem_exec_func *,
-		      struct nouveau_pm_level *);
-
-/* nouveau_pm.c */
-int  nouveau_pm_init(struct drm_device *dev);
-void nouveau_pm_fini(struct drm_device *dev);
-void nouveau_pm_resume(struct drm_device *dev);
-extern const struct nouveau_pm_profile_func nouveau_pm_static_profile_func;
-void nouveau_pm_trigger(struct drm_device *dev);
-
-/* nouveau_volt.c */
-void nouveau_volt_init(struct drm_device *);
-void nouveau_volt_fini(struct drm_device *);
-int  nouveau_volt_vid_lookup(struct drm_device *, int voltage);
-int  nouveau_volt_lvl_lookup(struct drm_device *, int vid);
-int  nouveau_voltage_gpio_get(struct drm_device *);
-int  nouveau_voltage_gpio_set(struct drm_device *, int voltage);
-
-/* nouveau_perf.c */
-void nouveau_perf_init(struct drm_device *);
-void nouveau_perf_fini(struct drm_device *);
-u8 *nouveau_perf_rammap(struct drm_device *, u32 freq, u8 *ver,
-			u8 *hdr, u8 *cnt, u8 *len);
-u8 *nouveau_perf_ramcfg(struct drm_device *, u32 freq, u8 *ver, u8 *len);
-u8 *nouveau_perf_timing(struct drm_device *, u32 freq, u8 *ver, u8 *len);
-
-/* nouveau_mem.c */
-void nouveau_mem_timing_init(struct drm_device *);
-void nouveau_mem_timing_fini(struct drm_device *);
-
-/* nv04_pm.c */
-int nv04_pm_clocks_get(struct drm_device *, struct nouveau_pm_level *);
-void *nv04_pm_clocks_pre(struct drm_device *, struct nouveau_pm_level *);
-int nv04_pm_clocks_set(struct drm_device *, void *);
-
-/* nv40_pm.c */
-int nv40_pm_clocks_get(struct drm_device *, struct nouveau_pm_level *);
-void *nv40_pm_clocks_pre(struct drm_device *, struct nouveau_pm_level *);
-int nv40_pm_clocks_set(struct drm_device *, void *);
-int nv40_pm_pwm_get(struct drm_device *, int, u32 *, u32 *);
-int nv40_pm_pwm_set(struct drm_device *, int, u32, u32);
-
-/* nv50_pm.c */
-int nv50_pm_clocks_get(struct drm_device *, struct nouveau_pm_level *);
-void *nv50_pm_clocks_pre(struct drm_device *, struct nouveau_pm_level *);
-int nv50_pm_clocks_set(struct drm_device *, void *);
-int nv50_pm_pwm_get(struct drm_device *, int, u32 *, u32 *);
-int nv50_pm_pwm_set(struct drm_device *, int, u32, u32);
-
-/* nva3_pm.c */
-int nva3_pm_clocks_get(struct drm_device *, struct nouveau_pm_level *);
-void *nva3_pm_clocks_pre(struct drm_device *, struct nouveau_pm_level *);
-int nva3_pm_clocks_set(struct drm_device *, void *);
-
-/* nvc0_pm.c */
-int nvc0_pm_clocks_get(struct drm_device *, struct nouveau_pm_level *);
-void *nvc0_pm_clocks_pre(struct drm_device *, struct nouveau_pm_level *);
-int nvc0_pm_clocks_set(struct drm_device *, void *);
-
-/* nouveau_mem.c */
-int  nouveau_mem_timing_calc(struct drm_device *, u32 freq,
-			     struct nouveau_pm_memtiming *);
-void nouveau_mem_timing_read(struct drm_device *,
-			     struct nouveau_pm_memtiming *);
-
-static inline int
-nva3_calc_pll(struct drm_device *dev, struct nvbios_pll *pll, u32 freq,
-	      int *N, int *fN, int *M, int *P)
-{
-	struct nouveau_device *device = nouveau_dev(dev);
-	struct nouveau_clock *clk = nouveau_clock(device);
-	struct nouveau_pll_vals pv;
-	int ret;
-
-	ret = clk->pll_calc(clk, pll, freq, &pv);
-	*N = pv.N1;
-	*M = pv.M1;
-	*P = pv.log2P;
-	return ret;
-}
-
-#endif
diff --git a/drivers/gpu/drm/nouveau/nouveau_prime.c b/drivers/gpu/drm/nouveau/nouveau_prime.c
index e90468d5e5c0..51a2cb102b44 100644
--- a/drivers/gpu/drm/nouveau/nouveau_prime.c
+++ b/drivers/gpu/drm/nouveau/nouveau_prime.c
@@ -71,14 +71,16 @@ struct drm_gem_object *nouveau_gem_prime_import_sg_table(struct drm_device *dev,
 		return ERR_PTR(ret);
 
 	nvbo->valid_domains = NOUVEAU_GEM_DOMAIN_GART;
-	nvbo->gem = drm_gem_object_alloc(dev, nvbo->bo.mem.size);
-	if (!nvbo->gem) {
+
+	/* Initialize the embedded gem-object. We return a single gem-reference
+	 * to the caller, instead of a normal nouveau_bo ttm reference. */
+	ret = drm_gem_object_init(dev, &nvbo->gem, nvbo->bo.mem.size);
+	if (ret) {
 		nouveau_bo_ref(NULL, &nvbo);
 		return ERR_PTR(-ENOMEM);
 	}
 
-	nvbo->gem->driver_private = nvbo;
-	return nvbo->gem;
+	return &nvbo->gem;
 }
 
 int nouveau_gem_prime_pin(struct drm_gem_object *obj)
diff --git a/drivers/gpu/drm/nouveau/nouveau_sysfs.c b/drivers/gpu/drm/nouveau/nouveau_sysfs.c
new file mode 100644
index 000000000000..89201a17ce75
--- /dev/null
+++ b/drivers/gpu/drm/nouveau/nouveau_sysfs.c
@@ -0,0 +1,162 @@
+/*
+ * Copyright 2013 Red Hat Inc.
+ *
+ * Permission is hereby granted, free of charge, to any person obtaining a
+ * copy of this software and associated documentation files (the "Software"),
+ * to deal in the Software without restriction, including without limitation
+ * the rights to use, copy, modify, merge, publish, distribute, sublicense,
+ * and/or sell copies of the Software, and to permit persons to whom the
+ * Software is furnished to do so, subject to the following conditions:
+ *
+ * The above copyright notice and this permission notice shall be included in
+ * all copies or substantial portions of the Software.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+ * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+ * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT.  IN NO EVENT SHALL
+ * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR
+ * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE,
+ * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR
+ * OTHER DEALINGS IN THE SOFTWARE.
+ *
+ * Authors: Ben Skeggs <bskeggs@redhat.com>
+ */
+
+#include "nouveau_sysfs.h"
+
+#include <core/object.h>
+#include <core/class.h>
+
+static inline struct drm_device *
+drm_device(struct device *d)
+{
+	return pci_get_drvdata(to_pci_dev(d));
+}
+
+#define snappendf(p,r,f,a...) do {                                             \
+	snprintf(p, r, f, ##a);                                                \
+	r -= strlen(p);                                                        \
+	p += strlen(p);                                                        \
+} while(0)
+
+static ssize_t
+nouveau_sysfs_pstate_get(struct device *d, struct device_attribute *a, char *b)
+{
+	struct nouveau_sysfs *sysfs = nouveau_sysfs(drm_device(d));
+	struct nv_control_pstate_info info;
+	size_t cnt = PAGE_SIZE;
+	char *buf = b;
+	int ret, i;
+
+	ret = nv_exec(sysfs->ctrl, NV_CONTROL_PSTATE_INFO, &info, sizeof(info));
+	if (ret)
+		return ret;
+
+	for (i = 0; i < info.count + 1; i++) {
+		const s32 state = i < info.count ? i :
+			NV_CONTROL_PSTATE_ATTR_STATE_CURRENT;
+		struct nv_control_pstate_attr attr = {
+			.state = state,
+			.index = 0,
+		};
+
+		ret = nv_exec(sysfs->ctrl, NV_CONTROL_PSTATE_ATTR,
+			     &attr, sizeof(attr));
+		if (ret)
+			return ret;
+
+		if (i < info.count)
+			snappendf(buf, cnt, "%02x:", attr.state);
+		else
+			snappendf(buf, cnt, "--:");
+
+		attr.index = 0;
+		do {
+			attr.state = state;
+			ret = nv_exec(sysfs->ctrl, NV_CONTROL_PSTATE_ATTR,
+				     &attr, sizeof(attr));
+			if (ret)
+				return ret;
+
+			snappendf(buf, cnt, " %s %d", attr.name, attr.min);
+			if (attr.min != attr.max)
+				snappendf(buf, cnt, "-%d", attr.max);
+			snappendf(buf, cnt, " %s", attr.unit);
+		} while (attr.index);
+
+		if ((state >= 0 && info.pstate == state) ||
+		    (state <  0 && info.ustate < 0))
+			snappendf(buf, cnt, " *");
+		snappendf(buf, cnt, "\n");
+	}
+
+	return strlen(b);
+}
+
+static ssize_t
+nouveau_sysfs_pstate_set(struct device *d, struct device_attribute *a,
+			 const char *buf, size_t count)
+{
+	struct nouveau_sysfs *sysfs = nouveau_sysfs(drm_device(d));
+	struct nv_control_pstate_user args;
+	long value, ret;
+	char *tmp;
+
+	if ((tmp = strchr(buf, '\n')))
+		*tmp = '\0';
+
+	if (!strcasecmp(buf, "none"))
+		args.state = NV_CONTROL_PSTATE_USER_STATE_UNKNOWN;
+	else
+	if (!strcasecmp(buf, "auto"))
+		args.state = NV_CONTROL_PSTATE_USER_STATE_PERFMON;
+	else {
+		ret = kstrtol(buf, 16, &value);
+		if (ret)
+			return ret;
+		args.state = value;
+	}
+
+	ret = nv_exec(sysfs->ctrl, NV_CONTROL_PSTATE_USER, &args, sizeof(args));
+	if (ret < 0)
+		return ret;
+
+	return count;
+}
+
+static DEVICE_ATTR(pstate, S_IRUGO | S_IWUSR,
+		   nouveau_sysfs_pstate_get, nouveau_sysfs_pstate_set);
+
+void
+nouveau_sysfs_fini(struct drm_device *dev)
+{
+	struct nouveau_sysfs *sysfs = nouveau_sysfs(dev);
+	struct nouveau_drm *drm = nouveau_drm(dev);
+
+	if (sysfs->ctrl) {
+		device_remove_file(&dev->pdev->dev, &dev_attr_pstate);
+		nouveau_object_del(nv_object(drm), NVDRM_DEVICE, NVDRM_CONTROL);
+	}
+
+	drm->sysfs = NULL;
+	kfree(sysfs);
+}
+
+int
+nouveau_sysfs_init(struct drm_device *dev)
+{
+	struct nouveau_drm *drm = nouveau_drm(dev);
+	struct nouveau_sysfs *sysfs;
+	int ret;
+
+	sysfs = drm->sysfs = kzalloc(sizeof(*sysfs), GFP_KERNEL);
+	if (!sysfs)
+		return -ENOMEM;
+
+	ret = nouveau_object_new(nv_object(drm), NVDRM_DEVICE, NVDRM_CONTROL,
+				 NV_CONTROL_CLASS, NULL, 0, &sysfs->ctrl);
+	if (ret == 0)
+		device_create_file(&dev->pdev->dev, &dev_attr_pstate);
+
+	return 0;
+}
diff --git a/drivers/gpu/drm/nouveau/nouveau_sysfs.h b/drivers/gpu/drm/nouveau/nouveau_sysfs.h
new file mode 100644
index 000000000000..74b47f1e01ed
--- /dev/null
+++ b/drivers/gpu/drm/nouveau/nouveau_sysfs.h
@@ -0,0 +1,19 @@
+#ifndef __NOUVEAU_SYSFS_H__
+#define __NOUVEAU_SYSFS_H__
+
+#include "nouveau_drm.h"
+
+struct nouveau_sysfs {
+	struct nouveau_object *ctrl;
+};
+
+static inline struct nouveau_sysfs *
+nouveau_sysfs(struct drm_device *dev)
+{
+	return nouveau_drm(dev)->sysfs;
+}
+
+int  nouveau_sysfs_init(struct drm_device *);
+void nouveau_sysfs_fini(struct drm_device *);
+
+#endif
diff --git a/drivers/gpu/drm/nouveau/nouveau_volt.c b/drivers/gpu/drm/nouveau/nouveau_volt.c
deleted file mode 100644
index 9976414cbe50..000000000000
--- a/drivers/gpu/drm/nouveau/nouveau_volt.c
+++ /dev/null
@@ -1,250 +0,0 @@
-/*
- * Copyright 2010 Red Hat Inc.
- *
- * Permission is hereby granted, free of charge, to any person obtaining a
- * copy of this software and associated documentation files (the "Software"),
- * to deal in the Software without restriction, including without limitation
- * the rights to use, copy, modify, merge, publish, distribute, sublicense,
- * and/or sell copies of the Software, and to permit persons to whom the
- * Software is furnished to do so, subject to the following conditions:
- *
- * The above copyright notice and this permission notice shall be included in
- * all copies or substantial portions of the Software.
- *
- * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
- * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
- * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT.  IN NO EVENT SHALL
- * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR
- * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE,
- * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR
- * OTHER DEALINGS IN THE SOFTWARE.
- *
- * Authors: Ben Skeggs
- */
-
-#include <drm/drmP.h>
-
-#include "nouveau_drm.h"
-#include "nouveau_pm.h"
-
-#include <subdev/bios/gpio.h>
-#include <subdev/gpio.h>
-
-static const enum dcb_gpio_func_name vidtag[] = { 0x04, 0x05, 0x06, 0x1a, 0x73 };
-static int nr_vidtag = sizeof(vidtag) / sizeof(vidtag[0]);
-
-int
-nouveau_voltage_gpio_get(struct drm_device *dev)
-{
-	struct nouveau_pm_voltage *volt = &nouveau_pm(dev)->voltage;
-	struct nouveau_device *device = nouveau_dev(dev);
-	struct nouveau_gpio *gpio = nouveau_gpio(device);
-	u8 vid = 0;
-	int i;
-
-	for (i = 0; i < nr_vidtag; i++) {
-		if (!(volt->vid_mask & (1 << i)))
-			continue;
-
-		vid |= gpio->get(gpio, 0, vidtag[i], 0xff) << i;
-	}
-
-	return nouveau_volt_lvl_lookup(dev, vid);
-}
-
-int
-nouveau_voltage_gpio_set(struct drm_device *dev, int voltage)
-{
-	struct nouveau_device *device = nouveau_dev(dev);
-	struct nouveau_gpio *gpio = nouveau_gpio(device);
-	struct nouveau_pm_voltage *volt = &nouveau_pm(dev)->voltage;
-	int vid, i;
-
-	vid = nouveau_volt_vid_lookup(dev, voltage);
-	if (vid < 0)
-		return vid;
-
-	for (i = 0; i < nr_vidtag; i++) {
-		if (!(volt->vid_mask & (1 << i)))
-			continue;
-
-		gpio->set(gpio, 0, vidtag[i], 0xff, !!(vid & (1 << i)));
-	}
-
-	return 0;
-}
-
-int
-nouveau_volt_vid_lookup(struct drm_device *dev, int voltage)
-{
-	struct nouveau_pm_voltage *volt = &nouveau_pm(dev)->voltage;
-	int i;
-
-	for (i = 0; i < volt->nr_level; i++) {
-		if (volt->level[i].voltage == voltage)
-			return volt->level[i].vid;
-	}
-
-	return -ENOENT;
-}
-
-int
-nouveau_volt_lvl_lookup(struct drm_device *dev, int vid)
-{
-	struct nouveau_pm_voltage *volt = &nouveau_pm(dev)->voltage;
-	int i;
-
-	for (i = 0; i < volt->nr_level; i++) {
-		if (volt->level[i].vid == vid)
-			return volt->level[i].voltage;
-	}
-
-	return -ENOENT;
-}
-
-void
-nouveau_volt_init(struct drm_device *dev)
-{
-	struct nouveau_drm *drm = nouveau_drm(dev);
-	struct nouveau_gpio *gpio = nouveau_gpio(drm->device);
-	struct nouveau_pm *pm = nouveau_pm(dev);
-	struct nouveau_pm_voltage *voltage = &pm->voltage;
-	struct nvbios *bios = &drm->vbios;
-	struct dcb_gpio_func func;
-	struct bit_entry P;
-	u8 *volt = NULL, *entry;
-	int i, headerlen, recordlen, entries, vidmask, vidshift;
-
-	if (bios->type == NVBIOS_BIT) {
-		if (bit_table(dev, 'P', &P))
-			return;
-
-		if (P.version == 1)
-			volt = ROMPTR(dev, P.data[16]);
-		else
-		if (P.version == 2)
-			volt = ROMPTR(dev, P.data[12]);
-		else {
-			NV_WARN(drm, "unknown volt for BIT P %d\n", P.version);
-		}
-	} else {
-		if (bios->data[bios->offset + 6] < 0x27) {
-			NV_DEBUG(drm, "BMP version too old for voltage\n");
-			return;
-		}
-
-		volt = ROMPTR(dev, bios->data[bios->offset + 0x98]);
-	}
-
-	if (!volt) {
-		NV_DEBUG(drm, "voltage table pointer invalid\n");
-		return;
-	}
-
-	switch (volt[0]) {
-	case 0x10:
-	case 0x11:
-	case 0x12:
-		headerlen = 5;
-		recordlen = volt[1];
-		entries   = volt[2];
-		vidshift  = 0;
-		vidmask   = volt[4];
-		break;
-	case 0x20:
-		headerlen = volt[1];
-		recordlen = volt[3];
-		entries   = volt[2];
-		vidshift  = 0; /* could be vidshift like 0x30? */
-		vidmask   = volt[5];
-		break;
-	case 0x30:
-		headerlen = volt[1];
-		recordlen = volt[2];
-		entries   = volt[3];
-		vidmask   = volt[4];
-		/* no longer certain what volt[5] is, if it's related to
-		 * the vid shift then it's definitely not a function of
-		 * how many bits are set.
-		 *
-		 * after looking at a number of nva3+ vbios images, they
-		 * all seem likely to have a static shift of 2.. lets
-		 * go with that for now until proven otherwise.
-		 */
-		vidshift  = 2;
-		break;
-	case 0x40:
-		headerlen = volt[1];
-		recordlen = volt[2];
-		entries   = volt[3]; /* not a clue what the entries are for.. */
-		vidmask   = volt[11]; /* guess.. */
-		vidshift  = 0;
-		break;
-	default:
-		NV_WARN(drm, "voltage table 0x%02x unknown\n", volt[0]);
-		return;
-	}
-
-	/* validate vid mask */
-	voltage->vid_mask = vidmask;
-	if (!voltage->vid_mask)
-		return;
-
-	i = 0;
-	while (vidmask) {
-		if (i > nr_vidtag) {
-			NV_DEBUG(drm, "vid bit %d unknown\n", i);
-			return;
-		}
-
-		if (gpio && gpio->find(gpio, 0, vidtag[i], 0xff, &func)) {
-			NV_DEBUG(drm, "vid bit %d has no gpio tag\n", i);
-			return;
-		}
-
-		vidmask >>= 1;
-		i++;
-	}
-
-	/* parse vbios entries into common format */
-	voltage->version = volt[0];
-	if (voltage->version < 0x40) {
-		voltage->nr_level = entries;
-		voltage->level =
-			kcalloc(entries, sizeof(*voltage->level), GFP_KERNEL);
-		if (!voltage->level)
-			return;
-
-		entry = volt + headerlen;
-		for (i = 0; i < entries; i++, entry += recordlen) {
-			voltage->level[i].voltage = entry[0] * 10000;
-			voltage->level[i].vid     = entry[1] >> vidshift;
-		}
-	} else {
-		u32 volt_uv = ROM32(volt[4]);
-		s16 step_uv = ROM16(volt[8]);
-		u8 vid;
-
-		voltage->nr_level = voltage->vid_mask + 1;
-		voltage->level = kcalloc(voltage->nr_level,
-					 sizeof(*voltage->level), GFP_KERNEL);
-		if (!voltage->level)
-			return;
-
-		for (vid = 0; vid <= voltage->vid_mask; vid++) {
-			voltage->level[vid].voltage = volt_uv;
-			voltage->level[vid].vid = vid;
-			volt_uv += step_uv;
-		}
-	}
-
-	voltage->supported = true;
-}
-
-void
-nouveau_volt_fini(struct drm_device *dev)
-{
-	struct nouveau_pm_voltage *volt = &nouveau_pm(dev)->voltage;
-
-	kfree(volt->level);
-}
diff --git a/drivers/gpu/drm/nouveau/nv04_fbcon.c b/drivers/gpu/drm/nouveau/nv04_fbcon.c
index 77dcc9c50777..8fe32bbed99a 100644
--- a/drivers/gpu/drm/nouveau/nv04_fbcon.c
+++ b/drivers/gpu/drm/nouveau/nv04_fbcon.c
@@ -255,6 +255,12 @@ nv04_fbcon_accel_init(struct fb_info *info)
 	OUT_RING(chan, NvCtxSurf2D);
 	BEGIN_NV04(chan, NvSubImageBlit, 0x02fc, 1);
 	OUT_RING(chan, 3);
+	if (device->chipset >= 0x11 /*XXX: oclass == 0x009f*/) {
+		BEGIN_NV04(chan, NvSubImageBlit, 0x0120, 3);
+		OUT_RING(chan, 0);
+		OUT_RING(chan, 1);
+		OUT_RING(chan, 2);
+	}
 
 	BEGIN_NV04(chan, NvSubGdiRect, 0x0000, 1);
 	OUT_RING(chan, NvGdiRect);
diff --git a/drivers/gpu/drm/nouveau/nv04_pm.c b/drivers/gpu/drm/nouveau/nv04_pm.c
deleted file mode 100644
index 27afc0ea28b0..000000000000
--- a/drivers/gpu/drm/nouveau/nv04_pm.c
+++ /dev/null
@@ -1,146 +0,0 @@
-/*
- * Copyright 2010 Red Hat Inc.
- *
- * Permission is hereby granted, free of charge, to any person obtaining a
- * copy of this software and associated documentation files (the "Software"),
- * to deal in the Software without restriction, including without limitation
- * the rights to use, copy, modify, merge, publish, distribute, sublicense,
- * and/or sell copies of the Software, and to permit persons to whom the
- * Software is furnished to do so, subject to the following conditions:
- *
- * The above copyright notice and this permission notice shall be included in
- * all copies or substantial portions of the Software.
- *
- * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
- * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
- * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT.  IN NO EVENT SHALL
- * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR
- * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE,
- * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR
- * OTHER DEALINGS IN THE SOFTWARE.
- *
- * Authors: Ben Skeggs
- */
-
-#include <drm/drmP.h>
-#include "nouveau_drm.h"
-#include "nouveau_reg.h"
-#include "dispnv04/hw.h"
-#include "nouveau_pm.h"
-
-#include <subdev/bios/pll.h>
-#include <subdev/clock.h>
-#include <subdev/timer.h>
-
-int
-nv04_pm_clocks_get(struct drm_device *dev, struct nouveau_pm_level *perflvl)
-{
-	int ret;
-
-	ret = nouveau_hw_get_clock(dev, PLL_CORE);
-	if (ret < 0)
-		return ret;
-	perflvl->core = ret;
-
-	ret = nouveau_hw_get_clock(dev, PLL_MEMORY);
-	if (ret < 0)
-		return ret;
-	perflvl->memory = ret;
-
-	return 0;
-}
-
-struct nv04_pm_clock {
-	struct nvbios_pll pll;
-	struct nouveau_pll_vals calc;
-};
-
-struct nv04_pm_state {
-	struct nv04_pm_clock core;
-	struct nv04_pm_clock memory;
-};
-
-static int
-calc_pll(struct drm_device *dev, u32 id, int khz, struct nv04_pm_clock *clk)
-{
-	struct nouveau_device *device = nouveau_dev(dev);
-	struct nouveau_bios *bios = nouveau_bios(device);
-	struct nouveau_clock *pclk = nouveau_clock(device);
-	int ret;
-
-	ret = nvbios_pll_parse(bios, id, &clk->pll);
-	if (ret)
-		return ret;
-
-	ret = pclk->pll_calc(pclk, &clk->pll, khz, &clk->calc);
-	if (!ret)
-		return -EINVAL;
-
-	return 0;
-}
-
-void *
-nv04_pm_clocks_pre(struct drm_device *dev, struct nouveau_pm_level *perflvl)
-{
-	struct nv04_pm_state *info;
-	int ret;
-
-	info = kzalloc(sizeof(*info), GFP_KERNEL);
-	if (!info)
-		return ERR_PTR(-ENOMEM);
-
-	ret = calc_pll(dev, PLL_CORE, perflvl->core, &info->core);
-	if (ret)
-		goto error;
-
-	if (perflvl->memory) {
-		ret = calc_pll(dev, PLL_MEMORY, perflvl->memory, &info->memory);
-		if (ret)
-			goto error;
-	}
-
-	return info;
-error:
-	kfree(info);
-	return ERR_PTR(ret);
-}
-
-static void
-prog_pll(struct drm_device *dev, struct nv04_pm_clock *clk)
-{
-	struct nouveau_device *device = nouveau_dev(dev);
-	struct nouveau_clock *pclk = nouveau_clock(device);
-	u32 reg = clk->pll.reg;
-
-	/* thank the insane nouveau_hw_setpll() interface for this */
-	if (device->card_type >= NV_40)
-		reg += 4;
-
-	pclk->pll_prog(pclk, reg, &clk->calc);
-}
-
-int
-nv04_pm_clocks_set(struct drm_device *dev, void *pre_state)
-{
-	struct nouveau_device *device = nouveau_dev(dev);
-	struct nouveau_timer *ptimer = nouveau_timer(device);
-	struct nv04_pm_state *state = pre_state;
-
-	prog_pll(dev, &state->core);
-
-	if (state->memory.pll.reg) {
-		prog_pll(dev, &state->memory);
-		if (device->card_type < NV_30) {
-			if (device->card_type == NV_20)
-				nv_mask(device, 0x1002c4, 0, 1 << 20);
-
-			/* Reset the DLLs */
-			nv_mask(device, 0x1002c0, 0, 1 << 8);
-		}
-	}
-
-	nv_ofuncs(ptimer)->init(nv_object(ptimer));
-
-	kfree(state);
-	return 0;
-}
diff --git a/drivers/gpu/drm/nouveau/nv40_pm.c b/drivers/gpu/drm/nouveau/nv40_pm.c
deleted file mode 100644
index 625f80d53dc2..000000000000
--- a/drivers/gpu/drm/nouveau/nv40_pm.c
+++ /dev/null
@@ -1,353 +0,0 @@
-/*
- * Copyright 2011 Red Hat Inc.
- *
- * Permission is hereby granted, free of charge, to any person obtaining a
- * copy of this software and associated documentation files (the "Software"),
- * to deal in the Software without restriction, including without limitation
- * the rights to use, copy, modify, merge, publish, distribute, sublicense,
- * and/or sell copies of the Software, and to permit persons to whom the
- * Software is furnished to do so, subject to the following conditions:
- *
- * The above copyright notice and this permission notice shall be included in
- * all copies or substantial portions of the Software.
- *
- * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
- * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
- * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT.  IN NO EVENT SHALL
- * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR
- * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE,
- * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR
- * OTHER DEALINGS IN THE SOFTWARE.
- *
- * Authors: Ben Skeggs
- */
-
-#include <drm/drmP.h>
-#include "nouveau_drm.h"
-#include "nouveau_bios.h"
-#include "nouveau_pm.h"
-#include "dispnv04/hw.h"
-
-#include <subdev/bios/pll.h>
-#include <subdev/clock.h>
-#include <subdev/timer.h>
-
-#include <engine/fifo.h>
-
-#define min2(a,b) ((a) < (b) ? (a) : (b))
-
-static u32
-read_pll_1(struct drm_device *dev, u32 reg)
-{
-	struct nouveau_device *device = nouveau_dev(dev);
-	u32 ctrl = nv_rd32(device, reg + 0x00);
-	int P = (ctrl & 0x00070000) >> 16;
-	int N = (ctrl & 0x0000ff00) >> 8;
-	int M = (ctrl & 0x000000ff) >> 0;
-	u32 ref = 27000, clk = 0;
-
-	if (ctrl & 0x80000000)
-		clk = ref * N / M;
-
-	return clk >> P;
-}
-
-static u32
-read_pll_2(struct drm_device *dev, u32 reg)
-{
-	struct nouveau_device *device = nouveau_dev(dev);
-	u32 ctrl = nv_rd32(device, reg + 0x00);
-	u32 coef = nv_rd32(device, reg + 0x04);
-	int N2 = (coef & 0xff000000) >> 24;
-	int M2 = (coef & 0x00ff0000) >> 16;
-	int N1 = (coef & 0x0000ff00) >> 8;
-	int M1 = (coef & 0x000000ff) >> 0;
-	int P = (ctrl & 0x00070000) >> 16;
-	u32 ref = 27000, clk = 0;
-
-	if ((ctrl & 0x80000000) && M1) {
-		clk = ref * N1 / M1;
-		if ((ctrl & 0x40000100) == 0x40000000) {
-			if (M2)
-				clk = clk * N2 / M2;
-			else
-				clk = 0;
-		}
-	}
-
-	return clk >> P;
-}
-
-static u32
-read_clk(struct drm_device *dev, u32 src)
-{
-	switch (src) {
-	case 3:
-		return read_pll_2(dev, 0x004000);
-	case 2:
-		return read_pll_1(dev, 0x004008);
-	default:
-		break;
-	}
-
-	return 0;
-}
-
-int
-nv40_pm_clocks_get(struct drm_device *dev, struct nouveau_pm_level *perflvl)
-{
-	struct nouveau_device *device = nouveau_dev(dev);
-	u32 ctrl = nv_rd32(device, 0x00c040);
-
-	perflvl->core   = read_clk(dev, (ctrl & 0x00000003) >> 0);
-	perflvl->shader = read_clk(dev, (ctrl & 0x00000030) >> 4);
-	perflvl->memory = read_pll_2(dev, 0x4020);
-	return 0;
-}
-
-struct nv40_pm_state {
-	u32 ctrl;
-	u32 npll_ctrl;
-	u32 npll_coef;
-	u32 spll;
-	u32 mpll_ctrl;
-	u32 mpll_coef;
-};
-
-static int
-nv40_calc_pll(struct drm_device *dev, u32 reg, struct nvbios_pll *pll,
-	      u32 clk, int *N1, int *M1, int *N2, int *M2, int *log2P)
-{
-	struct nouveau_device *device = nouveau_dev(dev);
-	struct nouveau_bios *bios = nouveau_bios(device);
-	struct nouveau_clock *pclk = nouveau_clock(device);
-	struct nouveau_pll_vals coef;
-	int ret;
-
-	ret = nvbios_pll_parse(bios, reg, pll);
-	if (ret)
-		return ret;
-
-	if (clk < pll->vco1.max_freq)
-		pll->vco2.max_freq = 0;
-
-	ret = pclk->pll_calc(pclk, pll, clk, &coef);
-	if (ret == 0)
-		return -ERANGE;
-
-	*N1 = coef.N1;
-	*M1 = coef.M1;
-	if (N2 && M2) {
-		if (pll->vco2.max_freq) {
-			*N2 = coef.N2;
-			*M2 = coef.M2;
-		} else {
-			*N2 = 1;
-			*M2 = 1;
-		}
-	}
-	*log2P = coef.log2P;
-	return 0;
-}
-
-void *
-nv40_pm_clocks_pre(struct drm_device *dev, struct nouveau_pm_level *perflvl)
-{
-	struct nv40_pm_state *info;
-	struct nvbios_pll pll;
-	int N1, N2, M1, M2, log2P;
-	int ret;
-
-	info = kmalloc(sizeof(*info), GFP_KERNEL);
-	if (!info)
-		return ERR_PTR(-ENOMEM);
-
-	/* core/geometric clock */
-	ret = nv40_calc_pll(dev, 0x004000, &pll, perflvl->core,
-			    &N1, &M1, &N2, &M2, &log2P);
-	if (ret < 0)
-		goto out;
-
-	if (N2 == M2) {
-		info->npll_ctrl = 0x80000100 | (log2P << 16);
-		info->npll_coef = (N1 << 8) | M1;
-	} else {
-		info->npll_ctrl = 0xc0000000 | (log2P << 16);
-		info->npll_coef = (N2 << 24) | (M2 << 16) | (N1 << 8) | M1;
-	}
-
-	/* use the second PLL for shader/rop clock, if it differs from core */
-	if (perflvl->shader && perflvl->shader != perflvl->core) {
-		ret = nv40_calc_pll(dev, 0x004008, &pll, perflvl->shader,
-				    &N1, &M1, NULL, NULL, &log2P);
-		if (ret < 0)
-			goto out;
-
-		info->spll = 0xc0000000 | (log2P << 16) | (N1 << 8) | M1;
-		info->ctrl = 0x00000223;
-	} else {
-		info->spll = 0x00000000;
-		info->ctrl = 0x00000333;
-	}
-
-	/* memory clock */
-	if (!perflvl->memory) {
-		info->mpll_ctrl = 0x00000000;
-		goto out;
-	}
-
-	ret = nv40_calc_pll(dev, 0x004020, &pll, perflvl->memory,
-			    &N1, &M1, &N2, &M2, &log2P);
-	if (ret < 0)
-		goto out;
-
-	info->mpll_ctrl  = 0x80000000 | (log2P << 16);
-	info->mpll_ctrl |= min2(pll.bias_p + log2P, pll.max_p) << 20;
-	if (N2 == M2) {
-		info->mpll_ctrl |= 0x00000100;
-		info->mpll_coef  = (N1 << 8) | M1;
-	} else {
-		info->mpll_ctrl |= 0x40000000;
-		info->mpll_coef  = (N2 << 24) | (M2 << 16) | (N1 << 8) | M1;
-	}
-
-out:
-	if (ret < 0) {
-		kfree(info);
-		info = ERR_PTR(ret);
-	}
-	return info;
-}
-
-static bool
-nv40_pm_gr_idle(void *data)
-{
-	struct drm_device *dev = data;
-	struct nouveau_device *device = nouveau_dev(dev);
-
-	if ((nv_rd32(device, 0x400760) & 0x000000f0) >> 4 !=
-	    (nv_rd32(device, 0x400760) & 0x0000000f))
-		return false;
-
-	if (nv_rd32(device, 0x400700))
-		return false;
-
-	return true;
-}
-
-int
-nv40_pm_clocks_set(struct drm_device *dev, void *pre_state)
-{
-	struct nouveau_device *device = nouveau_dev(dev);
-	struct nouveau_fifo *pfifo = nouveau_fifo(device);
-	struct nouveau_drm *drm = nouveau_drm(dev);
-	struct nv40_pm_state *info = pre_state;
-	unsigned long flags;
-	struct bit_entry M;
-	u32 crtc_mask = 0;
-	u8 sr1[2];
-	int i, ret = -EAGAIN;
-
-	/* determine which CRTCs are active, fetch VGA_SR1 for each */
-	for (i = 0; i < 2; i++) {
-		u32 vbl = nv_rd32(device, 0x600808 + (i * 0x2000));
-		u32 cnt = 0;
-		do {
-			if (vbl != nv_rd32(device, 0x600808 + (i * 0x2000))) {
-				nv_wr08(device, 0x0c03c4 + (i * 0x2000), 0x01);
-				sr1[i] = nv_rd08(device, 0x0c03c5 + (i * 0x2000));
-				if (!(sr1[i] & 0x20))
-					crtc_mask |= (1 << i);
-				break;
-			}
-			udelay(1);
-		} while (cnt++ < 32);
-	}
-
-	/* halt and idle engines */
-	pfifo->pause(pfifo, &flags);
-
-	if (!nv_wait_cb(device, nv40_pm_gr_idle, dev))
-		goto resume;
-
-	ret = 0;
-
-	/* set engine clocks */
-	nv_mask(device, 0x00c040, 0x00000333, 0x00000000);
-	nv_wr32(device, 0x004004, info->npll_coef);
-	nv_mask(device, 0x004000, 0xc0070100, info->npll_ctrl);
-	nv_mask(device, 0x004008, 0xc007ffff, info->spll);
-	mdelay(5);
-	nv_mask(device, 0x00c040, 0x00000333, info->ctrl);
-
-	if (!info->mpll_ctrl)
-		goto resume;
-
-	/* wait for vblank start on active crtcs, disable memory access */
-	for (i = 0; i < 2; i++) {
-		if (!(crtc_mask & (1 << i)))
-			continue;
-		nv_wait(device, 0x600808 + (i * 0x2000), 0x00010000, 0x00000000);
-		nv_wait(device, 0x600808 + (i * 0x2000), 0x00010000, 0x00010000);
-		nv_wr08(device, 0x0c03c4 + (i * 0x2000), 0x01);
-		nv_wr08(device, 0x0c03c5 + (i * 0x2000), sr1[i] | 0x20);
-	}
-
-	/* prepare ram for reclocking */
-	nv_wr32(device, 0x1002d4, 0x00000001); /* precharge */
-	nv_wr32(device, 0x1002d0, 0x00000001); /* refresh */
-	nv_wr32(device, 0x1002d0, 0x00000001); /* refresh */
-	nv_mask(device, 0x100210, 0x80000000, 0x00000000); /* no auto refresh */
-	nv_wr32(device, 0x1002dc, 0x00000001); /* enable self-refresh */
-
-	/* change the PLL of each memory partition */
-	nv_mask(device, 0x00c040, 0x0000c000, 0x00000000);
-	switch (nv_device(drm->device)->chipset) {
-	case 0x40:
-	case 0x45:
-	case 0x41:
-	case 0x42:
-	case 0x47:
-		nv_mask(device, 0x004044, 0xc0771100, info->mpll_ctrl);
-		nv_mask(device, 0x00402c, 0xc0771100, info->mpll_ctrl);
-		nv_wr32(device, 0x004048, info->mpll_coef);
-		nv_wr32(device, 0x004030, info->mpll_coef);
-	case 0x43:
-	case 0x49:
-	case 0x4b:
-		nv_mask(device, 0x004038, 0xc0771100, info->mpll_ctrl);
-		nv_wr32(device, 0x00403c, info->mpll_coef);
-	default:
-		nv_mask(device, 0x004020, 0xc0771100, info->mpll_ctrl);
-		nv_wr32(device, 0x004024, info->mpll_coef);
-		break;
-	}
-	udelay(100);
-	nv_mask(device, 0x00c040, 0x0000c000, 0x0000c000);
-
-	/* re-enable normal operation of memory controller */
-	nv_wr32(device, 0x1002dc, 0x00000000);
-	nv_mask(device, 0x100210, 0x80000000, 0x80000000);
-	udelay(100);
-
-	/* execute memory reset script from vbios */
-	if (!bit_table(dev, 'M', &M))
-		nouveau_bios_run_init_table(dev, ROM16(M.data[0]), NULL, 0);
-
-	/* make sure we're in vblank (hopefully the same one as before), and
-	 * then re-enable crtc memory access
-	 */
-	for (i = 0; i < 2; i++) {
-		if (!(crtc_mask & (1 << i)))
-			continue;
-		nv_wait(device, 0x600808 + (i * 0x2000), 0x00010000, 0x00010000);
-		nv_wr08(device, 0x0c03c4 + (i * 0x2000), 0x01);
-		nv_wr08(device, 0x0c03c5 + (i * 0x2000), sr1[i]);
-	}
-
-	/* resume engines */
-resume:
-	pfifo->start(pfifo, &flags);
-	kfree(info);
-	return ret;
-}
diff --git a/drivers/gpu/drm/nouveau/nv50_pm.c b/drivers/gpu/drm/nouveau/nv50_pm.c
deleted file mode 100644
index 4efc33fa73fc..000000000000
--- a/drivers/gpu/drm/nouveau/nv50_pm.c
+++ /dev/null
@@ -1,855 +0,0 @@
-/*
- * Copyright 2010 Red Hat Inc.
- *
- * Permission is hereby granted, free of charge, to any person obtaining a
- * copy of this software and associated documentation files (the "Software"),
- * to deal in the Software without restriction, including without limitation
- * the rights to use, copy, modify, merge, publish, distribute, sublicense,
- * and/or sell copies of the Software, and to permit persons to whom the
- * Software is furnished to do so, subject to the following conditions:
- *
- * The above copyright notice and this permission notice shall be included in
- * all copies or substantial portions of the Software.
- *
- * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
- * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
- * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT.  IN NO EVENT SHALL
- * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR
- * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE,
- * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR
- * OTHER DEALINGS IN THE SOFTWARE.
- *
- * Authors: Ben Skeggs
- */
-
-#include <drm/drmP.h>
-#include "nouveau_drm.h"
-#include "nouveau_bios.h"
-#include "dispnv04/hw.h"
-#include "nouveau_pm.h"
-#include "nouveau_hwsq.h"
-
-#include "nv50_display.h"
-
-#include <subdev/bios/pll.h>
-#include <subdev/clock.h>
-#include <subdev/timer.h>
-#include <subdev/fb.h>
-
-enum clk_src {
-	clk_src_crystal,
-	clk_src_href,
-	clk_src_hclk,
-	clk_src_hclkm3,
-	clk_src_hclkm3d2,
-	clk_src_host,
-	clk_src_nvclk,
-	clk_src_sclk,
-	clk_src_mclk,
-	clk_src_vdec,
-	clk_src_dom6
-};
-
-static u32 read_clk(struct drm_device *, enum clk_src);
-
-static u32
-read_div(struct drm_device *dev)
-{
-	struct nouveau_device *device = nouveau_dev(dev);
-	struct nouveau_drm *drm = nouveau_drm(dev);
-
-	switch (nv_device(drm->device)->chipset) {
-	case 0x50: /* it exists, but only has bit 31, not the dividers.. */
-	case 0x84:
-	case 0x86:
-	case 0x98:
-	case 0xa0:
-		return nv_rd32(device, 0x004700);
-	case 0x92:
-	case 0x94:
-	case 0x96:
-		return nv_rd32(device, 0x004800);
-	default:
-		return 0x00000000;
-	}
-}
-
-static u32
-read_pll_src(struct drm_device *dev, u32 base)
-{
-	struct nouveau_device *device = nouveau_dev(dev);
-	struct nouveau_drm *drm = nouveau_drm(dev);
-	u32 coef, ref = read_clk(dev, clk_src_crystal);
-	u32 rsel = nv_rd32(device, 0x00e18c);
-	int P, N, M, id;
-
-	switch (nv_device(drm->device)->chipset) {
-	case 0x50:
-	case 0xa0:
-		switch (base) {
-		case 0x4020:
-		case 0x4028: id = !!(rsel & 0x00000004); break;
-		case 0x4008: id = !!(rsel & 0x00000008); break;
-		case 0x4030: id = 0; break;
-		default:
-			NV_ERROR(drm, "ref: bad pll 0x%06x\n", base);
-			return 0;
-		}
-
-		coef = nv_rd32(device, 0x00e81c + (id * 0x0c));
-		ref *=  (coef & 0x01000000) ? 2 : 4;
-		P    =  (coef & 0x00070000) >> 16;
-		N    = ((coef & 0x0000ff00) >> 8) + 1;
-		M    = ((coef & 0x000000ff) >> 0) + 1;
-		break;
-	case 0x84:
-	case 0x86:
-	case 0x92:
-		coef = nv_rd32(device, 0x00e81c);
-		P    = (coef & 0x00070000) >> 16;
-		N    = (coef & 0x0000ff00) >> 8;
-		M    = (coef & 0x000000ff) >> 0;
-		break;
-	case 0x94:
-	case 0x96:
-	case 0x98:
-		rsel = nv_rd32(device, 0x00c050);
-		switch (base) {
-		case 0x4020: rsel = (rsel & 0x00000003) >> 0; break;
-		case 0x4008: rsel = (rsel & 0x0000000c) >> 2; break;
-		case 0x4028: rsel = (rsel & 0x00001800) >> 11; break;
-		case 0x4030: rsel = 3; break;
-		default:
-			NV_ERROR(drm, "ref: bad pll 0x%06x\n", base);
-			return 0;
-		}
-
-		switch (rsel) {
-		case 0: id = 1; break;
-		case 1: return read_clk(dev, clk_src_crystal);
-		case 2: return read_clk(dev, clk_src_href);
-		case 3: id = 0; break;
-		}
-
-		coef =  nv_rd32(device, 0x00e81c + (id * 0x28));
-		P    = (nv_rd32(device, 0x00e824 + (id * 0x28)) >> 16) & 7;
-		P   += (coef & 0x00070000) >> 16;
-		N    = (coef & 0x0000ff00) >> 8;
-		M    = (coef & 0x000000ff) >> 0;
-		break;
-	default:
-		BUG_ON(1);
-	}
-
-	if (M)
-		return (ref * N / M) >> P;
-	return 0;
-}
-
-static u32
-read_pll_ref(struct drm_device *dev, u32 base)
-{
-	struct nouveau_device *device = nouveau_dev(dev);
-	struct nouveau_drm *drm = nouveau_drm(dev);
-	u32 src, mast = nv_rd32(device, 0x00c040);
-
-	switch (base) {
-	case 0x004028:
-		src = !!(mast & 0x00200000);
-		break;
-	case 0x004020:
-		src = !!(mast & 0x00400000);
-		break;
-	case 0x004008:
-		src = !!(mast & 0x00010000);
-		break;
-	case 0x004030:
-		src = !!(mast & 0x02000000);
-		break;
-	case 0x00e810:
-		return read_clk(dev, clk_src_crystal);
-	default:
-		NV_ERROR(drm, "bad pll 0x%06x\n", base);
-		return 0;
-	}
-
-	if (src)
-		return read_clk(dev, clk_src_href);
-	return read_pll_src(dev, base);
-}
-
-static u32
-read_pll(struct drm_device *dev, u32 base)
-{
-	struct nouveau_device *device = nouveau_dev(dev);
-	struct nouveau_drm *drm = nouveau_drm(dev);
-	u32 mast = nv_rd32(device, 0x00c040);
-	u32 ctrl = nv_rd32(device, base + 0);
-	u32 coef = nv_rd32(device, base + 4);
-	u32 ref = read_pll_ref(dev, base);
-	u32 clk = 0;
-	int N1, N2, M1, M2;
-
-	if (base == 0x004028 && (mast & 0x00100000)) {
-		/* wtf, appears to only disable post-divider on nva0 */
-		if (nv_device(drm->device)->chipset != 0xa0)
-			return read_clk(dev, clk_src_dom6);
-	}
-
-	N2 = (coef & 0xff000000) >> 24;
-	M2 = (coef & 0x00ff0000) >> 16;
-	N1 = (coef & 0x0000ff00) >> 8;
-	M1 = (coef & 0x000000ff);
-	if ((ctrl & 0x80000000) && M1) {
-		clk = ref * N1 / M1;
-		if ((ctrl & 0x40000100) == 0x40000000) {
-			if (M2)
-				clk = clk * N2 / M2;
-			else
-				clk = 0;
-		}
-	}
-
-	return clk;
-}
-
-static u32
-read_clk(struct drm_device *dev, enum clk_src src)
-{
-	struct nouveau_device *device = nouveau_dev(dev);
-	struct nouveau_drm *drm = nouveau_drm(dev);
-	u32 mast = nv_rd32(device, 0x00c040);
-	u32 P = 0;
-
-	switch (src) {
-	case clk_src_crystal:
-		return device->crystal;
-	case clk_src_href:
-		return 100000; /* PCIE reference clock */
-	case clk_src_hclk:
-		return read_clk(dev, clk_src_href) * 27778 / 10000;
-	case clk_src_hclkm3:
-		return read_clk(dev, clk_src_hclk) * 3;
-	case clk_src_hclkm3d2:
-		return read_clk(dev, clk_src_hclk) * 3 / 2;
-	case clk_src_host:
-		switch (mast & 0x30000000) {
-		case 0x00000000: return read_clk(dev, clk_src_href);
-		case 0x10000000: break;
-		case 0x20000000: /* !0x50 */
-		case 0x30000000: return read_clk(dev, clk_src_hclk);
-		}
-		break;
-	case clk_src_nvclk:
-		if (!(mast & 0x00100000))
-			P = (nv_rd32(device, 0x004028) & 0x00070000) >> 16;
-		switch (mast & 0x00000003) {
-		case 0x00000000: return read_clk(dev, clk_src_crystal) >> P;
-		case 0x00000001: return read_clk(dev, clk_src_dom6);
-		case 0x00000002: return read_pll(dev, 0x004020) >> P;
-		case 0x00000003: return read_pll(dev, 0x004028) >> P;
-		}
-		break;
-	case clk_src_sclk:
-		P = (nv_rd32(device, 0x004020) & 0x00070000) >> 16;
-		switch (mast & 0x00000030) {
-		case 0x00000000:
-			if (mast & 0x00000080)
-				return read_clk(dev, clk_src_host) >> P;
-			return read_clk(dev, clk_src_crystal) >> P;
-		case 0x00000010: break;
-		case 0x00000020: return read_pll(dev, 0x004028) >> P;
-		case 0x00000030: return read_pll(dev, 0x004020) >> P;
-		}
-		break;
-	case clk_src_mclk:
-		P = (nv_rd32(device, 0x004008) & 0x00070000) >> 16;
-		if (nv_rd32(device, 0x004008) & 0x00000200) {
-			switch (mast & 0x0000c000) {
-			case 0x00000000:
-				return read_clk(dev, clk_src_crystal) >> P;
-			case 0x00008000:
-			case 0x0000c000:
-				return read_clk(dev, clk_src_href) >> P;
-			}
-		} else {
-			return read_pll(dev, 0x004008) >> P;
-		}
-		break;
-	case clk_src_vdec:
-		P = (read_div(dev) & 0x00000700) >> 8;
-		switch (nv_device(drm->device)->chipset) {
-		case 0x84:
-		case 0x86:
-		case 0x92:
-		case 0x94:
-		case 0x96:
-		case 0xa0:
-			switch (mast & 0x00000c00) {
-			case 0x00000000:
-				if (nv_device(drm->device)->chipset == 0xa0) /* wtf?? */
-					return read_clk(dev, clk_src_nvclk) >> P;
-				return read_clk(dev, clk_src_crystal) >> P;
-			case 0x00000400:
-				return 0;
-			case 0x00000800:
-				if (mast & 0x01000000)
-					return read_pll(dev, 0x004028) >> P;
-				return read_pll(dev, 0x004030) >> P;
-			case 0x00000c00:
-				return read_clk(dev, clk_src_nvclk) >> P;
-			}
-			break;
-		case 0x98:
-			switch (mast & 0x00000c00) {
-			case 0x00000000:
-				return read_clk(dev, clk_src_nvclk) >> P;
-			case 0x00000400:
-				return 0;
-			case 0x00000800:
-				return read_clk(dev, clk_src_hclkm3d2) >> P;
-			case 0x00000c00:
-				return read_clk(dev, clk_src_mclk) >> P;
-			}
-			break;
-		}
-		break;
-	case clk_src_dom6:
-		switch (nv_device(drm->device)->chipset) {
-		case 0x50:
-		case 0xa0:
-			return read_pll(dev, 0x00e810) >> 2;
-		case 0x84:
-		case 0x86:
-		case 0x92:
-		case 0x94:
-		case 0x96:
-		case 0x98:
-			P = (read_div(dev) & 0x00000007) >> 0;
-			switch (mast & 0x0c000000) {
-			case 0x00000000: return read_clk(dev, clk_src_href);
-			case 0x04000000: break;
-			case 0x08000000: return read_clk(dev, clk_src_hclk);
-			case 0x0c000000:
-				return read_clk(dev, clk_src_hclkm3) >> P;
-			}
-			break;
-		default:
-			break;
-		}
-	default:
-		break;
-	}
-
-	NV_DEBUG(drm, "unknown clock source %d 0x%08x\n", src, mast);
-	return 0;
-}
-
-int
-nv50_pm_clocks_get(struct drm_device *dev, struct nouveau_pm_level *perflvl)
-{
-	struct nouveau_drm *drm = nouveau_drm(dev);
-	if (nv_device(drm->device)->chipset == 0xaa ||
-	    nv_device(drm->device)->chipset == 0xac)
-		return 0;
-
-	perflvl->core   = read_clk(dev, clk_src_nvclk);
-	perflvl->shader = read_clk(dev, clk_src_sclk);
-	perflvl->memory = read_clk(dev, clk_src_mclk);
-	if (nv_device(drm->device)->chipset != 0x50) {
-		perflvl->vdec = read_clk(dev, clk_src_vdec);
-		perflvl->dom6 = read_clk(dev, clk_src_dom6);
-	}
-
-	return 0;
-}
-
-struct nv50_pm_state {
-	struct nouveau_pm_level *perflvl;
-	struct hwsq_ucode eclk_hwsq;
-	struct hwsq_ucode mclk_hwsq;
-	u32 mscript;
-	u32 mmast;
-	u32 mctrl;
-	u32 mcoef;
-};
-
-static u32
-calc_pll(struct drm_device *dev, u32 reg, struct nvbios_pll *pll,
-	 u32 clk, int *N1, int *M1, int *log2P)
-{
-	struct nouveau_device *device = nouveau_dev(dev);
-	struct nouveau_bios *bios = nouveau_bios(device);
-	struct nouveau_clock *pclk = nouveau_clock(device);
-	struct nouveau_pll_vals coef;
-	int ret;
-
-	ret = nvbios_pll_parse(bios, reg, pll);
-	if (ret)
-		return 0;
-
-	pll->vco2.max_freq = 0;
-	pll->refclk = read_pll_ref(dev, reg);
-	if (!pll->refclk)
-		return 0;
-
-	ret = pclk->pll_calc(pclk, pll, clk, &coef);
-	if (ret == 0)
-		return 0;
-
-	*N1 = coef.N1;
-	*M1 = coef.M1;
-	*log2P = coef.log2P;
-	return ret;
-}
-
-static inline u32
-calc_div(u32 src, u32 target, int *div)
-{
-	u32 clk0 = src, clk1 = src;
-	for (*div = 0; *div <= 7; (*div)++) {
-		if (clk0 <= target) {
-			clk1 = clk0 << (*div ? 1 : 0);
-			break;
-		}
-		clk0 >>= 1;
-	}
-
-	if (target - clk0 <= clk1 - target)
-		return clk0;
-	(*div)--;
-	return clk1;
-}
-
-static inline u32
-clk_same(u32 a, u32 b)
-{
-	return ((a / 1000) == (b / 1000));
-}
-
-static void
-mclk_precharge(struct nouveau_mem_exec_func *exec)
-{
-	struct nv50_pm_state *info = exec->priv;
-	struct hwsq_ucode *hwsq = &info->mclk_hwsq;
-
-	hwsq_wr32(hwsq, 0x1002d4, 0x00000001);
-}
-
-static void
-mclk_refresh(struct nouveau_mem_exec_func *exec)
-{
-	struct nv50_pm_state *info = exec->priv;
-	struct hwsq_ucode *hwsq = &info->mclk_hwsq;
-
-	hwsq_wr32(hwsq, 0x1002d0, 0x00000001);
-}
-
-static void
-mclk_refresh_auto(struct nouveau_mem_exec_func *exec, bool enable)
-{
-	struct nv50_pm_state *info = exec->priv;
-	struct hwsq_ucode *hwsq = &info->mclk_hwsq;
-
-	hwsq_wr32(hwsq, 0x100210, enable ? 0x80000000 : 0x00000000);
-}
-
-static void
-mclk_refresh_self(struct nouveau_mem_exec_func *exec, bool enable)
-{
-	struct nv50_pm_state *info = exec->priv;
-	struct hwsq_ucode *hwsq = &info->mclk_hwsq;
-
-	hwsq_wr32(hwsq, 0x1002dc, enable ? 0x00000001 : 0x00000000);
-}
-
-static void
-mclk_wait(struct nouveau_mem_exec_func *exec, u32 nsec)
-{
-	struct nv50_pm_state *info = exec->priv;
-	struct hwsq_ucode *hwsq = &info->mclk_hwsq;
-
-	if (nsec > 1000)
-		hwsq_usec(hwsq, (nsec + 500) / 1000);
-}
-
-static u32
-mclk_mrg(struct nouveau_mem_exec_func *exec, int mr)
-{
-	struct nouveau_device *device = nouveau_dev(exec->dev);
-	if (mr <= 1)
-		return nv_rd32(device, 0x1002c0 + ((mr - 0) * 4));
-	if (mr <= 3)
-		return nv_rd32(device, 0x1002e0 + ((mr - 2) * 4));
-	return 0;
-}
-
-static void
-mclk_mrs(struct nouveau_mem_exec_func *exec, int mr, u32 data)
-{
-	struct nouveau_device *device = nouveau_dev(exec->dev);
-	struct nouveau_fb *pfb = nouveau_fb(device);
-	struct nv50_pm_state *info = exec->priv;
-	struct hwsq_ucode *hwsq = &info->mclk_hwsq;
-
-	if (mr <= 1) {
-		if (pfb->ram->ranks > 1)
-			hwsq_wr32(hwsq, 0x1002c8 + ((mr - 0) * 4), data);
-		hwsq_wr32(hwsq, 0x1002c0 + ((mr - 0) * 4), data);
-	} else
-	if (mr <= 3) {
-		if (pfb->ram->ranks > 1)
-			hwsq_wr32(hwsq, 0x1002e8 + ((mr - 2) * 4), data);
-		hwsq_wr32(hwsq, 0x1002e0 + ((mr - 2) * 4), data);
-	}
-}
-
-static void
-mclk_clock_set(struct nouveau_mem_exec_func *exec)
-{
-	struct nouveau_device *device = nouveau_dev(exec->dev);
-	struct nv50_pm_state *info = exec->priv;
-	struct hwsq_ucode *hwsq = &info->mclk_hwsq;
-	u32 ctrl = nv_rd32(device, 0x004008);
-
-	info->mmast = nv_rd32(device, 0x00c040);
-	info->mmast &= ~0xc0000000; /* get MCLK_2 from HREF */
-	info->mmast |=  0x0000c000; /* use MCLK_2 as MPLL_BYPASS clock */
-
-	hwsq_wr32(hwsq, 0xc040, info->mmast);
-	hwsq_wr32(hwsq, 0x4008, ctrl | 0x00000200); /* bypass MPLL */
-	if (info->mctrl & 0x80000000)
-		hwsq_wr32(hwsq, 0x400c, info->mcoef);
-	hwsq_wr32(hwsq, 0x4008, info->mctrl);
-}
-
-static void
-mclk_timing_set(struct nouveau_mem_exec_func *exec)
-{
-	struct nouveau_device *device = nouveau_dev(exec->dev);
-	struct nv50_pm_state *info = exec->priv;
-	struct nouveau_pm_level *perflvl = info->perflvl;
-	struct hwsq_ucode *hwsq = &info->mclk_hwsq;
-	int i;
-
-	for (i = 0; i < 9; i++) {
-		u32 reg = 0x100220 + (i * 4);
-		u32 val = nv_rd32(device, reg);
-		if (val != perflvl->timing.reg[i])
-			hwsq_wr32(hwsq, reg, perflvl->timing.reg[i]);
-	}
-}
-
-static int
-calc_mclk(struct drm_device *dev, struct nouveau_pm_level *perflvl,
-	  struct nv50_pm_state *info)
-{
-	struct nouveau_drm *drm = nouveau_drm(dev);
-	struct nouveau_device *device = nouveau_dev(dev);
-	u32 crtc_mask = 0; /*XXX: nv50_display_active_crtcs(dev); */
-	struct nouveau_mem_exec_func exec = {
-		.dev = dev,
-		.precharge = mclk_precharge,
-		.refresh = mclk_refresh,
-		.refresh_auto = mclk_refresh_auto,
-		.refresh_self = mclk_refresh_self,
-		.wait = mclk_wait,
-		.mrg = mclk_mrg,
-		.mrs = mclk_mrs,
-		.clock_set = mclk_clock_set,
-		.timing_set = mclk_timing_set,
-		.priv = info
-	};
-	struct hwsq_ucode *hwsq = &info->mclk_hwsq;
-	struct nvbios_pll pll;
-	int N, M, P;
-	int ret;
-
-	/* use pcie refclock if possible, otherwise use mpll */
-	info->mctrl  = nv_rd32(device, 0x004008);
-	info->mctrl &= ~0x81ff0200;
-	if (clk_same(perflvl->memory, read_clk(dev, clk_src_href))) {
-		info->mctrl |= 0x00000200 | (pll.bias_p << 19);
-	} else {
-		ret = calc_pll(dev, 0x4008, &pll, perflvl->memory, &N, &M, &P);
-		if (ret == 0)
-			return -EINVAL;
-
-		info->mctrl |= 0x80000000 | (P << 22) | (P << 16);
-		info->mctrl |= pll.bias_p << 19;
-		info->mcoef  = (N << 8) | M;
-	}
-
-	/* build the ucode which will reclock the memory for us */
-	hwsq_init(hwsq);
-	if (crtc_mask) {
-		hwsq_op5f(hwsq, crtc_mask, 0x00); /* wait for scanout */
-		hwsq_op5f(hwsq, crtc_mask, 0x01); /* wait for vblank */
-	}
-	if (nv_device(drm->device)->chipset >= 0x92)
-		hwsq_wr32(hwsq, 0x611200, 0x00003300); /* disable scanout */
-	hwsq_setf(hwsq, 0x10, 0); /* disable bus access */
-	hwsq_op5f(hwsq, 0x00, 0x01); /* no idea :s */
-
-	ret = nouveau_mem_exec(&exec, perflvl);
-	if (ret)
-		return ret;
-
-	hwsq_setf(hwsq, 0x10, 1); /* enable bus access */
-	hwsq_op5f(hwsq, 0x00, 0x00); /* no idea, reverse of 0x00, 0x01? */
-	if (nv_device(drm->device)->chipset >= 0x92)
-		hwsq_wr32(hwsq, 0x611200, 0x00003330); /* enable scanout */
-	hwsq_fini(hwsq);
-	return 0;
-}
-
-void *
-nv50_pm_clocks_pre(struct drm_device *dev, struct nouveau_pm_level *perflvl)
-{
-	struct nouveau_device *device = nouveau_dev(dev);
-	struct nouveau_drm *drm = nouveau_drm(dev);
-	struct nv50_pm_state *info;
-	struct hwsq_ucode *hwsq;
-	struct nvbios_pll pll;
-	u32 out, mast, divs, ctrl;
-	int clk, ret = -EINVAL;
-	int N, M, P1, P2;
-
-	if (nv_device(drm->device)->chipset == 0xaa ||
-	    nv_device(drm->device)->chipset == 0xac)
-		return ERR_PTR(-ENODEV);
-
-	info = kmalloc(sizeof(*info), GFP_KERNEL);
-	if (!info)
-		return ERR_PTR(-ENOMEM);
-	info->perflvl = perflvl;
-
-	/* memory: build hwsq ucode which we'll use to reclock memory.
-	 *         use pcie refclock if possible, otherwise use mpll */
-	info->mclk_hwsq.len = 0;
-	if (perflvl->memory) {
-		ret = calc_mclk(dev, perflvl, info);
-		if (ret)
-			goto error;
-		info->mscript = perflvl->memscript;
-	}
-
-	divs = read_div(dev);
-	mast = info->mmast;
-
-	/* start building HWSQ script for engine reclocking */
-	hwsq = &info->eclk_hwsq;
-	hwsq_init(hwsq);
-	hwsq_setf(hwsq, 0x10, 0); /* disable bus access */
-	hwsq_op5f(hwsq, 0x00, 0x01); /* wait for access disabled? */
-
-	/* vdec/dom6: switch to "safe" clocks temporarily */
-	if (perflvl->vdec) {
-		mast &= ~0x00000c00;
-		divs &= ~0x00000700;
-	}
-
-	if (perflvl->dom6) {
-		mast &= ~0x0c000000;
-		divs &= ~0x00000007;
-	}
-
-	hwsq_wr32(hwsq, 0x00c040, mast);
-
-	/* vdec: avoid modifying xpll until we know exactly how the other
-	 * clock domains work, i suspect at least some of them can also be
-	 * tied to xpll...
-	 */
-	if (perflvl->vdec) {
-		/* see how close we can get using nvclk as a source */
-		clk = calc_div(perflvl->core, perflvl->vdec, &P1);
-
-		/* see how close we can get using xpll/hclk as a source */
-		if (nv_device(drm->device)->chipset != 0x98)
-			out = read_pll(dev, 0x004030);
-		else
-			out = read_clk(dev, clk_src_hclkm3d2);
-		out = calc_div(out, perflvl->vdec, &P2);
-
-		/* select whichever gets us closest */
-		if (abs((int)perflvl->vdec - clk) <=
-		    abs((int)perflvl->vdec - out)) {
-			if (nv_device(drm->device)->chipset != 0x98)
-				mast |= 0x00000c00;
-			divs |= P1 << 8;
-		} else {
-			mast |= 0x00000800;
-			divs |= P2 << 8;
-		}
-	}
-
-	/* dom6: nfi what this is, but we're limited to various combinations
-	 * of the host clock frequency
-	 */
-	if (perflvl->dom6) {
-		if (clk_same(perflvl->dom6, read_clk(dev, clk_src_href))) {
-			mast |= 0x00000000;
-		} else
-		if (clk_same(perflvl->dom6, read_clk(dev, clk_src_hclk))) {
-			mast |= 0x08000000;
-		} else {
-			clk = read_clk(dev, clk_src_hclk) * 3;
-			clk = calc_div(clk, perflvl->dom6, &P1);
-
-			mast |= 0x0c000000;
-			divs |= P1;
-		}
-	}
-
-	/* vdec/dom6: complete switch to new clocks */
-	switch (nv_device(drm->device)->chipset) {
-	case 0x92:
-	case 0x94:
-	case 0x96:
-		hwsq_wr32(hwsq, 0x004800, divs);
-		break;
-	default:
-		hwsq_wr32(hwsq, 0x004700, divs);
-		break;
-	}
-
-	hwsq_wr32(hwsq, 0x00c040, mast);
-
-	/* core/shader: make sure sclk/nvclk are disconnected from their
-	 * PLLs (nvclk to dom6, sclk to hclk)
-	 */
-	if (nv_device(drm->device)->chipset < 0x92)
-		mast = (mast & ~0x001000b0) | 0x00100080;
-	else
-		mast = (mast & ~0x000000b3) | 0x00000081;
-
-	hwsq_wr32(hwsq, 0x00c040, mast);
-
-	/* core: for the moment at least, always use nvpll */
-	clk = calc_pll(dev, 0x4028, &pll, perflvl->core, &N, &M, &P1);
-	if (clk == 0)
-		goto error;
-
-	ctrl  = nv_rd32(device, 0x004028) & ~0xc03f0100;
-	mast &= ~0x00100000;
-	mast |= 3;
-
-	hwsq_wr32(hwsq, 0x004028, 0x80000000 | (P1 << 19) | (P1 << 16) | ctrl);
-	hwsq_wr32(hwsq, 0x00402c, (N << 8) | M);
-
-	/* shader: tie to nvclk if possible, otherwise use spll.  have to be
-	 * very careful that the shader clock is at least twice the core, or
-	 * some chipsets will be very unhappy.  i expect most or all of these
-	 * cases will be handled by tying to nvclk, but it's possible there's
-	 * corners
-	 */
-	ctrl = nv_rd32(device, 0x004020) & ~0xc03f0100;
-
-	if (P1-- && perflvl->shader == (perflvl->core << 1)) {
-		hwsq_wr32(hwsq, 0x004020, (P1 << 19) | (P1 << 16) | ctrl);
-		hwsq_wr32(hwsq, 0x00c040, 0x00000020 | mast);
-	} else {
-		clk = calc_pll(dev, 0x4020, &pll, perflvl->shader, &N, &M, &P1);
-		if (clk == 0)
-			goto error;
-		ctrl |= 0x80000000;
-
-		hwsq_wr32(hwsq, 0x004020, (P1 << 19) | (P1 << 16) | ctrl);
-		hwsq_wr32(hwsq, 0x004024, (N << 8) | M);
-		hwsq_wr32(hwsq, 0x00c040, 0x00000030 | mast);
-	}
-
-	hwsq_setf(hwsq, 0x10, 1); /* enable bus access */
-	hwsq_op5f(hwsq, 0x00, 0x00); /* wait for access enabled? */
-	hwsq_fini(hwsq);
-
-	return info;
-error:
-	kfree(info);
-	return ERR_PTR(ret);
-}
-
-static int
-prog_hwsq(struct drm_device *dev, struct hwsq_ucode *hwsq)
-{
-	struct nouveau_device *device = nouveau_dev(dev);
-	struct nouveau_drm *drm = nouveau_drm(dev);
-	u32 hwsq_data, hwsq_kick;
-	int i;
-
-	if (nv_device(drm->device)->chipset < 0x94) {
-		hwsq_data = 0x001400;
-		hwsq_kick = 0x00000003;
-	} else {
-		hwsq_data = 0x080000;
-		hwsq_kick = 0x00000001;
-	}
-	/* upload hwsq ucode */
-	nv_mask(device, 0x001098, 0x00000008, 0x00000000);
-	nv_wr32(device, 0x001304, 0x00000000);
-	if (nv_device(drm->device)->chipset >= 0x92)
-		nv_wr32(device, 0x001318, 0x00000000);
-	for (i = 0; i < hwsq->len / 4; i++)
-		nv_wr32(device, hwsq_data + (i * 4), hwsq->ptr.u32[i]);
-	nv_mask(device, 0x001098, 0x00000018, 0x00000018);
-
-	/* launch, and wait for completion */
-	nv_wr32(device, 0x00130c, hwsq_kick);
-	if (!nv_wait(device, 0x001308, 0x00000100, 0x00000000)) {
-		NV_ERROR(drm, "hwsq ucode exec timed out\n");
-		NV_ERROR(drm, "0x001308: 0x%08x\n", nv_rd32(device, 0x001308));
-		for (i = 0; i < hwsq->len / 4; i++) {
-			NV_ERROR(drm, "0x%06x: 0x%08x\n", 0x1400 + (i * 4),
-				 nv_rd32(device, 0x001400 + (i * 4)));
-		}
-
-		return -EIO;
-	}
-
-	return 0;
-}
-
-int
-nv50_pm_clocks_set(struct drm_device *dev, void *data)
-{
-	struct nouveau_device *device = nouveau_dev(dev);
-	struct nv50_pm_state *info = data;
-	struct bit_entry M;
-	int ret = -EBUSY;
-
-	/* halt and idle execution engines */
-	nv_mask(device, 0x002504, 0x00000001, 0x00000001);
-	if (!nv_wait(device, 0x002504, 0x00000010, 0x00000010))
-		goto resume;
-	if (!nv_wait(device, 0x00251c, 0x0000003f, 0x0000003f))
-		goto resume;
-
-	/* program memory clock, if necessary - must come before engine clock
-	 * reprogramming due to how we construct the hwsq scripts in pre()
-	 */
-#define nouveau_bios_init_exec(a,b) nouveau_bios_run_init_table((a), (b), NULL, 0)
-	if (info->mclk_hwsq.len) {
-		/* execute some scripts that do ??? from the vbios.. */
-		if (!bit_table(dev, 'M', &M) && M.version == 1) {
-			if (M.length >= 6)
-				nouveau_bios_init_exec(dev, ROM16(M.data[5]));
-			if (M.length >= 8)
-				nouveau_bios_init_exec(dev, ROM16(M.data[7]));
-			if (M.length >= 10)
-				nouveau_bios_init_exec(dev, ROM16(M.data[9]));
-			nouveau_bios_init_exec(dev, info->mscript);
-		}
-
-		ret = prog_hwsq(dev, &info->mclk_hwsq);
-		if (ret)
-			goto resume;
-	}
-
-	/* program engine clocks */
-	ret = prog_hwsq(dev, &info->eclk_hwsq);
-
-resume:
-	nv_mask(device, 0x002504, 0x00000001, 0x00000000);
-	kfree(info);
-	return ret;
-}
diff --git a/drivers/gpu/drm/nouveau/nva3_pm.c b/drivers/gpu/drm/nouveau/nva3_pm.c
deleted file mode 100644
index 0d0ed597fea8..000000000000
--- a/drivers/gpu/drm/nouveau/nva3_pm.c
+++ /dev/null
@@ -1,624 +0,0 @@
-/*
- * Copyright 2010 Red Hat Inc.
- *
- * Permission is hereby granted, free of charge, to any person obtaining a
- * copy of this software and associated documentation files (the "Software"),
- * to deal in the Software without restriction, including without limitation
- * the rights to use, copy, modify, merge, publish, distribute, sublicense,
- * and/or sell copies of the Software, and to permit persons to whom the
- * Software is furnished to do so, subject to the following conditions:
- *
- * The above copyright notice and this permission notice shall be included in
- * all copies or substantial portions of the Software.
- *
- * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
- * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
- * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT.  IN NO EVENT SHALL
- * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR
- * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE,
- * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR
- * OTHER DEALINGS IN THE SOFTWARE.
- *
- * Authors: Ben Skeggs
- */
-
-#include <drm/drmP.h>
-#include "nouveau_drm.h"
-#include "nouveau_bios.h"
-#include "nouveau_pm.h"
-
-#include <subdev/bios/pll.h>
-#include <subdev/bios.h>
-#include <subdev/clock.h>
-#include <subdev/timer.h>
-#include <subdev/fb.h>
-
-static u32 read_clk(struct drm_device *, int, bool);
-static u32 read_pll(struct drm_device *, int, u32);
-
-static u32
-read_vco(struct drm_device *dev, int clk)
-{
-	struct nouveau_device *device = nouveau_dev(dev);
-	u32 sctl = nv_rd32(device, 0x4120 + (clk * 4));
-	if ((sctl & 0x00000030) != 0x00000030)
-		return read_pll(dev, 0x41, 0x00e820);
-	return read_pll(dev, 0x42, 0x00e8a0);
-}
-
-static u32
-read_clk(struct drm_device *dev, int clk, bool ignore_en)
-{
-	struct nouveau_device *device = nouveau_dev(dev);
-	struct nouveau_drm *drm = nouveau_drm(dev);
-	u32 sctl, sdiv, sclk;
-
-	/* refclk for the 0xe8xx plls is a fixed frequency */
-	if (clk >= 0x40) {
-		if (nv_device(drm->device)->chipset == 0xaf) {
-			/* no joke.. seriously.. sigh.. */
-			return nv_rd32(device, 0x00471c) * 1000;
-		}
-
-		return device->crystal;
-	}
-
-	sctl = nv_rd32(device, 0x4120 + (clk * 4));
-	if (!ignore_en && !(sctl & 0x00000100))
-		return 0;
-
-	switch (sctl & 0x00003000) {
-	case 0x00000000:
-		return device->crystal;
-	case 0x00002000:
-		if (sctl & 0x00000040)
-			return 108000;
-		return 100000;
-	case 0x00003000:
-		sclk = read_vco(dev, clk);
-		sdiv = ((sctl & 0x003f0000) >> 16) + 2;
-		return (sclk * 2) / sdiv;
-	default:
-		return 0;
-	}
-}
-
-static u32
-read_pll(struct drm_device *dev, int clk, u32 pll)
-{
-	struct nouveau_device *device = nouveau_dev(dev);
-	u32 ctrl = nv_rd32(device, pll + 0);
-	u32 sclk = 0, P = 1, N = 1, M = 1;
-
-	if (!(ctrl & 0x00000008)) {
-		if (ctrl & 0x00000001) {
-			u32 coef = nv_rd32(device, pll + 4);
-			M = (coef & 0x000000ff) >> 0;
-			N = (coef & 0x0000ff00) >> 8;
-			P = (coef & 0x003f0000) >> 16;
-
-			/* no post-divider on these.. */
-			if ((pll & 0x00ff00) == 0x00e800)
-				P = 1;
-
-			sclk = read_clk(dev, 0x00 + clk, false);
-		}
-	} else {
-		sclk = read_clk(dev, 0x10 + clk, false);
-	}
-
-	if (M * P)
-		return sclk * N / (M * P);
-	return 0;
-}
-
-struct creg {
-	u32 clk;
-	u32 pll;
-};
-
-static int
-calc_clk(struct drm_device *dev, int clk, u32 pll, u32 khz, struct creg *reg)
-{
-	struct nouveau_drm *drm = nouveau_drm(dev);
-	struct nouveau_device *device = nouveau_dev(dev);
-	struct nouveau_bios *bios = nouveau_bios(device);
-	struct nvbios_pll limits;
-	u32 oclk, sclk, sdiv;
-	int P, N, M, diff;
-	int ret;
-
-	reg->pll = 0;
-	reg->clk = 0;
-	if (!khz) {
-		NV_DEBUG(drm, "no clock for 0x%04x/0x%02x\n", pll, clk);
-		return 0;
-	}
-
-	switch (khz) {
-	case 27000:
-		reg->clk = 0x00000100;
-		return khz;
-	case 100000:
-		reg->clk = 0x00002100;
-		return khz;
-	case 108000:
-		reg->clk = 0x00002140;
-		return khz;
-	default:
-		sclk = read_vco(dev, clk);
-		sdiv = min((sclk * 2) / (khz - 2999), (u32)65);
-		/* if the clock has a PLL attached, and we can get a within
-		 * [-2, 3) MHz of a divider, we'll disable the PLL and use
-		 * the divider instead.
-		 *
-		 * divider can go as low as 2, limited here because NVIDIA
-		 * and the VBIOS on my NVA8 seem to prefer using the PLL
-		 * for 810MHz - is there a good reason?
-		 */
-		if (sdiv > 4) {
-			oclk = (sclk * 2) / sdiv;
-			diff = khz - oclk;
-			if (!pll || (diff >= -2000 && diff < 3000)) {
-				reg->clk = (((sdiv - 2) << 16) | 0x00003100);
-				return oclk;
-			}
-		}
-
-		if (!pll) {
-			NV_ERROR(drm, "bad freq %02x: %d %d\n", clk, khz, sclk);
-			return -ERANGE;
-		}
-
-		break;
-	}
-
-	ret = nvbios_pll_parse(bios, pll, &limits);
-	if (ret)
-		return ret;
-
-	limits.refclk = read_clk(dev, clk - 0x10, true);
-	if (!limits.refclk)
-		return -EINVAL;
-
-	ret = nva3_calc_pll(dev, &limits, khz, &N, NULL, &M, &P);
-	if (ret >= 0) {
-		reg->clk = nv_rd32(device, 0x4120 + (clk * 4));
-		reg->pll = (P << 16) | (N << 8) | M;
-	}
-
-	return ret;
-}
-
-static void
-prog_pll(struct drm_device *dev, int clk, u32 pll, struct creg *reg)
-{
-	struct nouveau_device *device = nouveau_dev(dev);
-	struct nouveau_drm *drm = nouveau_drm(dev);
-	const u32 src0 = 0x004120 + (clk * 4);
-	const u32 src1 = 0x004160 + (clk * 4);
-	const u32 ctrl = pll + 0;
-	const u32 coef = pll + 4;
-
-	if (!reg->clk && !reg->pll) {
-		NV_DEBUG(drm, "no clock for %02x\n", clk);
-		return;
-	}
-
-	if (reg->pll) {
-		nv_mask(device, src0, 0x00000101, 0x00000101);
-		nv_wr32(device, coef, reg->pll);
-		nv_mask(device, ctrl, 0x00000015, 0x00000015);
-		nv_mask(device, ctrl, 0x00000010, 0x00000000);
-		nv_wait(device, ctrl, 0x00020000, 0x00020000);
-		nv_mask(device, ctrl, 0x00000010, 0x00000010);
-		nv_mask(device, ctrl, 0x00000008, 0x00000000);
-		nv_mask(device, src1, 0x00000100, 0x00000000);
-		nv_mask(device, src1, 0x00000001, 0x00000000);
-	} else {
-		nv_mask(device, src1, 0x003f3141, 0x00000101 | reg->clk);
-		nv_mask(device, ctrl, 0x00000018, 0x00000018);
-		udelay(20);
-		nv_mask(device, ctrl, 0x00000001, 0x00000000);
-		nv_mask(device, src0, 0x00000100, 0x00000000);
-		nv_mask(device, src0, 0x00000001, 0x00000000);
-	}
-}
-
-static void
-prog_clk(struct drm_device *dev, int clk, struct creg *reg)
-{
-	struct nouveau_device *device = nouveau_dev(dev);
-	struct nouveau_drm *drm = nouveau_drm(dev);
-
-	if (!reg->clk) {
-		NV_DEBUG(drm, "no clock for %02x\n", clk);
-		return;
-	}
-
-	nv_mask(device, 0x004120 + (clk * 4), 0x003f3141, 0x00000101 | reg->clk);
-}
-
-int
-nva3_pm_clocks_get(struct drm_device *dev, struct nouveau_pm_level *perflvl)
-{
-	perflvl->core   = read_pll(dev, 0x00, 0x4200);
-	perflvl->shader = read_pll(dev, 0x01, 0x4220);
-	perflvl->memory = read_pll(dev, 0x02, 0x4000);
-	perflvl->unka0  = read_clk(dev, 0x20, false);
-	perflvl->vdec   = read_clk(dev, 0x21, false);
-	perflvl->daemon = read_clk(dev, 0x25, false);
-	perflvl->copy   = perflvl->core;
-	return 0;
-}
-
-struct nva3_pm_state {
-	struct nouveau_pm_level *perflvl;
-
-	struct creg nclk;
-	struct creg sclk;
-	struct creg vdec;
-	struct creg unka0;
-
-	struct creg mclk;
-	u8 *rammap;
-	u8  rammap_ver;
-	u8  rammap_len;
-	u8 *ramcfg;
-	u8  ramcfg_len;
-	u32 r004018;
-	u32 r100760;
-};
-
-void *
-nva3_pm_clocks_pre(struct drm_device *dev, struct nouveau_pm_level *perflvl)
-{
-	struct nva3_pm_state *info;
-	u8 ramcfg_cnt;
-	int ret;
-
-	info = kzalloc(sizeof(*info), GFP_KERNEL);
-	if (!info)
-		return ERR_PTR(-ENOMEM);
-
-	ret = calc_clk(dev, 0x10, 0x4200, perflvl->core, &info->nclk);
-	if (ret < 0)
-		goto out;
-
-	ret = calc_clk(dev, 0x11, 0x4220, perflvl->shader, &info->sclk);
-	if (ret < 0)
-		goto out;
-
-	ret = calc_clk(dev, 0x12, 0x4000, perflvl->memory, &info->mclk);
-	if (ret < 0)
-		goto out;
-
-	ret = calc_clk(dev, 0x20, 0x0000, perflvl->unka0, &info->unka0);
-	if (ret < 0)
-		goto out;
-
-	ret = calc_clk(dev, 0x21, 0x0000, perflvl->vdec, &info->vdec);
-	if (ret < 0)
-		goto out;
-
-	info->rammap = nouveau_perf_rammap(dev, perflvl->memory,
-					   &info->rammap_ver,
-					   &info->rammap_len,
-					   &ramcfg_cnt, &info->ramcfg_len);
-	if (info->rammap_ver != 0x10 || info->rammap_len < 5)
-		info->rammap = NULL;
-
-	info->ramcfg = nouveau_perf_ramcfg(dev, perflvl->memory,
-					   &info->rammap_ver,
-					   &info->ramcfg_len);
-	if (info->rammap_ver != 0x10)
-		info->ramcfg = NULL;
-
-	info->perflvl = perflvl;
-out:
-	if (ret < 0) {
-		kfree(info);
-		info = ERR_PTR(ret);
-	}
-	return info;
-}
-
-static bool
-nva3_pm_grcp_idle(void *data)
-{
-	struct drm_device *dev = data;
-	struct nouveau_device *device = nouveau_dev(dev);
-
-	if (!(nv_rd32(device, 0x400304) & 0x00000001))
-		return true;
-	if (nv_rd32(device, 0x400308) == 0x0050001c)
-		return true;
-	return false;
-}
-
-static void
-mclk_precharge(struct nouveau_mem_exec_func *exec)
-{
-	struct nouveau_device *device = nouveau_dev(exec->dev);
-	nv_wr32(device, 0x1002d4, 0x00000001);
-}
-
-static void
-mclk_refresh(struct nouveau_mem_exec_func *exec)
-{
-	struct nouveau_device *device = nouveau_dev(exec->dev);
-	nv_wr32(device, 0x1002d0, 0x00000001);
-}
-
-static void
-mclk_refresh_auto(struct nouveau_mem_exec_func *exec, bool enable)
-{
-	struct nouveau_device *device = nouveau_dev(exec->dev);
-	nv_wr32(device, 0x100210, enable ? 0x80000000 : 0x00000000);
-}
-
-static void
-mclk_refresh_self(struct nouveau_mem_exec_func *exec, bool enable)
-{
-	struct nouveau_device *device = nouveau_dev(exec->dev);
-	nv_wr32(device, 0x1002dc, enable ? 0x00000001 : 0x00000000);
-}
-
-static void
-mclk_wait(struct nouveau_mem_exec_func *exec, u32 nsec)
-{
-	struct nouveau_device *device = nouveau_dev(exec->dev);
-	volatile u32 post = nv_rd32(device, 0); (void)post;
-	udelay((nsec + 500) / 1000);
-}
-
-static u32
-mclk_mrg(struct nouveau_mem_exec_func *exec, int mr)
-{
-	struct nouveau_device *device = nouveau_dev(exec->dev);
-	if (mr <= 1)
-		return nv_rd32(device, 0x1002c0 + ((mr - 0) * 4));
-	if (mr <= 3)
-		return nv_rd32(device, 0x1002e0 + ((mr - 2) * 4));
-	return 0;
-}
-
-static void
-mclk_mrs(struct nouveau_mem_exec_func *exec, int mr, u32 data)
-{
-	struct nouveau_device *device = nouveau_dev(exec->dev);
-	struct nouveau_fb *pfb = nouveau_fb(device);
-	if (mr <= 1) {
-		if (pfb->ram->ranks > 1)
-			nv_wr32(device, 0x1002c8 + ((mr - 0) * 4), data);
-		nv_wr32(device, 0x1002c0 + ((mr - 0) * 4), data);
-	} else
-	if (mr <= 3) {
-		if (pfb->ram->ranks > 1)
-			nv_wr32(device, 0x1002e8 + ((mr - 2) * 4), data);
-		nv_wr32(device, 0x1002e0 + ((mr - 2) * 4), data);
-	}
-}
-
-static void
-mclk_clock_set(struct nouveau_mem_exec_func *exec)
-{
-	struct nouveau_device *device = nouveau_dev(exec->dev);
-	struct nva3_pm_state *info = exec->priv;
-	u32 ctrl;
-
-	ctrl = nv_rd32(device, 0x004000);
-	if (!(ctrl & 0x00000008) && info->mclk.pll) {
-		nv_wr32(device, 0x004000, (ctrl |=  0x00000008));
-		nv_mask(device, 0x1110e0, 0x00088000, 0x00088000);
-		nv_wr32(device, 0x004018, 0x00001000);
-		nv_wr32(device, 0x004000, (ctrl &= ~0x00000001));
-		nv_wr32(device, 0x004004, info->mclk.pll);
-		nv_wr32(device, 0x004000, (ctrl |=  0x00000001));
-		udelay(64);
-		nv_wr32(device, 0x004018, 0x00005000 | info->r004018);
-		udelay(20);
-	} else
-	if (!info->mclk.pll) {
-		nv_mask(device, 0x004168, 0x003f3040, info->mclk.clk);
-		nv_wr32(device, 0x004000, (ctrl |= 0x00000008));
-		nv_mask(device, 0x1110e0, 0x00088000, 0x00088000);
-		nv_wr32(device, 0x004018, 0x0000d000 | info->r004018);
-	}
-
-	if (info->rammap) {
-		if (info->ramcfg && (info->rammap[4] & 0x08)) {
-			u32 unk5a0 = (ROM16(info->ramcfg[5]) << 8) |
-				      info->ramcfg[5];
-			u32 unk5a4 = ROM16(info->ramcfg[7]);
-			u32 unk804 = (info->ramcfg[9] & 0xf0) << 16 |
-				     (info->ramcfg[3] & 0x0f) << 16 |
-				     (info->ramcfg[9] & 0x0f) |
-				     0x80000000;
-			nv_wr32(device, 0x1005a0, unk5a0);
-			nv_wr32(device, 0x1005a4, unk5a4);
-			nv_wr32(device, 0x10f804, unk804);
-			nv_mask(device, 0x10053c, 0x00001000, 0x00000000);
-		} else {
-			nv_mask(device, 0x10053c, 0x00001000, 0x00001000);
-			nv_mask(device, 0x10f804, 0x80000000, 0x00000000);
-			nv_mask(device, 0x100760, 0x22222222, info->r100760);
-			nv_mask(device, 0x1007a0, 0x22222222, info->r100760);
-			nv_mask(device, 0x1007e0, 0x22222222, info->r100760);
-		}
-	}
-
-	if (info->mclk.pll) {
-		nv_mask(device, 0x1110e0, 0x00088000, 0x00011000);
-		nv_wr32(device, 0x004000, (ctrl &= ~0x00000008));
-	}
-}
-
-static void
-mclk_timing_set(struct nouveau_mem_exec_func *exec)
-{
-	struct nouveau_device *device = nouveau_dev(exec->dev);
-	struct nva3_pm_state *info = exec->priv;
-	struct nouveau_pm_level *perflvl = info->perflvl;
-	int i;
-
-	for (i = 0; i < 9; i++)
-		nv_wr32(device, 0x100220 + (i * 4), perflvl->timing.reg[i]);
-
-	if (info->ramcfg) {
-		u32 data = (info->ramcfg[2] & 0x08) ? 0x00000000 : 0x00001000;
-		nv_mask(device, 0x100200, 0x00001000, data);
-	}
-
-	if (info->ramcfg) {
-		u32 unk714 = nv_rd32(device, 0x100714) & ~0xf0000010;
-		u32 unk718 = nv_rd32(device, 0x100718) & ~0x00000100;
-		u32 unk71c = nv_rd32(device, 0x10071c) & ~0x00000100;
-		if ( (info->ramcfg[2] & 0x20))
-			unk714 |= 0xf0000000;
-		if (!(info->ramcfg[2] & 0x04))
-			unk714 |= 0x00000010;
-		nv_wr32(device, 0x100714, unk714);
-
-		if (info->ramcfg[2] & 0x01)
-			unk71c |= 0x00000100;
-		nv_wr32(device, 0x10071c, unk71c);
-
-		if (info->ramcfg[2] & 0x02)
-			unk718 |= 0x00000100;
-		nv_wr32(device, 0x100718, unk718);
-
-		if (info->ramcfg[2] & 0x10)
-			nv_wr32(device, 0x111100, 0x48000000); /*XXX*/
-	}
-}
-
-static void
-prog_mem(struct drm_device *dev, struct nva3_pm_state *info)
-{
-	struct nouveau_device *device = nouveau_dev(dev);
-	struct nouveau_mem_exec_func exec = {
-		.dev = dev,
-		.precharge = mclk_precharge,
-		.refresh = mclk_refresh,
-		.refresh_auto = mclk_refresh_auto,
-		.refresh_self = mclk_refresh_self,
-		.wait = mclk_wait,
-		.mrg = mclk_mrg,
-		.mrs = mclk_mrs,
-		.clock_set = mclk_clock_set,
-		.timing_set = mclk_timing_set,
-		.priv = info
-	};
-	u32 ctrl;
-
-	/* XXX: where the fuck does 750MHz come from? */
-	if (info->perflvl->memory <= 750000) {
-		info->r004018 = 0x10000000;
-		info->r100760 = 0x22222222;
-	}
-
-	ctrl = nv_rd32(device, 0x004000);
-	if (ctrl & 0x00000008) {
-		if (info->mclk.pll) {
-			nv_mask(device, 0x004128, 0x00000101, 0x00000101);
-			nv_wr32(device, 0x004004, info->mclk.pll);
-			nv_wr32(device, 0x004000, (ctrl |= 0x00000001));
-			nv_wr32(device, 0x004000, (ctrl &= 0xffffffef));
-			nv_wait(device, 0x004000, 0x00020000, 0x00020000);
-			nv_wr32(device, 0x004000, (ctrl |= 0x00000010));
-			nv_wr32(device, 0x004018, 0x00005000 | info->r004018);
-			nv_wr32(device, 0x004000, (ctrl |= 0x00000004));
-		}
-	} else {
-		u32 ssel = 0x00000101;
-		if (info->mclk.clk)
-			ssel |= info->mclk.clk;
-		else
-			ssel |= 0x00080000; /* 324MHz, shouldn't matter... */
-		nv_mask(device, 0x004168, 0x003f3141, ctrl);
-	}
-
-	if (info->ramcfg) {
-		if (info->ramcfg[2] & 0x10) {
-			nv_mask(device, 0x111104, 0x00000600, 0x00000000);
-		} else {
-			nv_mask(device, 0x111100, 0x40000000, 0x40000000);
-			nv_mask(device, 0x111104, 0x00000180, 0x00000000);
-		}
-	}
-	if (info->rammap && !(info->rammap[4] & 0x02))
-		nv_mask(device, 0x100200, 0x00000800, 0x00000000);
-	nv_wr32(device, 0x611200, 0x00003300);
-	if (!(info->ramcfg[2] & 0x10))
-		nv_wr32(device, 0x111100, 0x4c020000); /*XXX*/
-
-	nouveau_mem_exec(&exec, info->perflvl);
-
-	nv_wr32(device, 0x611200, 0x00003330);
-	if (info->rammap && (info->rammap[4] & 0x02))
-		nv_mask(device, 0x100200, 0x00000800, 0x00000800);
-	if (info->ramcfg) {
-		if (info->ramcfg[2] & 0x10) {
-			nv_mask(device, 0x111104, 0x00000180, 0x00000180);
-			nv_mask(device, 0x111100, 0x40000000, 0x00000000);
-		} else {
-			nv_mask(device, 0x111104, 0x00000600, 0x00000600);
-		}
-	}
-
-	if (info->mclk.pll) {
-		nv_mask(device, 0x004168, 0x00000001, 0x00000000);
-		nv_mask(device, 0x004168, 0x00000100, 0x00000000);
-	} else {
-		nv_mask(device, 0x004000, 0x00000001, 0x00000000);
-		nv_mask(device, 0x004128, 0x00000001, 0x00000000);
-		nv_mask(device, 0x004128, 0x00000100, 0x00000000);
-	}
-}
-
-int
-nva3_pm_clocks_set(struct drm_device *dev, void *pre_state)
-{
-	struct nouveau_device *device = nouveau_dev(dev);
-	struct nouveau_drm *drm = nouveau_drm(dev);
-	struct nva3_pm_state *info = pre_state;
-	int ret = -EAGAIN;
-
-	/* prevent any new grctx switches from starting */
-	nv_wr32(device, 0x400324, 0x00000000);
-	nv_wr32(device, 0x400328, 0x0050001c); /* wait flag 0x1c */
-	/* wait for any pending grctx switches to complete */
-	if (!nv_wait_cb(device, nva3_pm_grcp_idle, dev)) {
-		NV_ERROR(drm, "pm: ctxprog didn't go idle\n");
-		goto cleanup;
-	}
-	/* freeze PFIFO */
-	nv_mask(device, 0x002504, 0x00000001, 0x00000001);
-	if (!nv_wait(device, 0x002504, 0x00000010, 0x00000010)) {
-		NV_ERROR(drm, "pm: fifo didn't go idle\n");
-		goto cleanup;
-	}
-
-	prog_pll(dev, 0x00, 0x004200, &info->nclk);
-	prog_pll(dev, 0x01, 0x004220, &info->sclk);
-	prog_clk(dev, 0x20, &info->unka0);
-	prog_clk(dev, 0x21, &info->vdec);
-
-	if (info->mclk.clk || info->mclk.pll)
-		prog_mem(dev, info);
-
-	ret = 0;
-
-cleanup:
-	/* unfreeze PFIFO */
-	nv_mask(device, 0x002504, 0x00000001, 0x00000000);
-	/* restore ctxprog to normal */
-	nv_wr32(device, 0x400324, 0x00000000);
-	nv_wr32(device, 0x400328, 0x0070009c); /* set flag 0x1c */
-	/* unblock it if necessary */
-	if (nv_rd32(device, 0x400308) == 0x0050001c)
-		nv_mask(device, 0x400824, 0x10000000, 0x10000000);
-	kfree(info);
-	return ret;
-}
diff --git a/drivers/gpu/drm/nouveau/nvc0_pm.c b/drivers/gpu/drm/nouveau/nvc0_pm.c
deleted file mode 100644
index 3b7041cb013f..000000000000
--- a/drivers/gpu/drm/nouveau/nvc0_pm.c
+++ /dev/null
@@ -1,599 +0,0 @@
-/*
- * Copyright 2011 Red Hat Inc.
- *
- * Permission is hereby granted, free of charge, to any person obtaining a
- * copy of this software and associated documentation files (the "Software"),
- * to deal in the Software without restriction, including without limitation
- * the rights to use, copy, modify, merge, publish, distribute, sublicense,
- * and/or sell copies of the Software, and to permit persons to whom the
- * Software is furnished to do so, subject to the following conditions:
- *
- * The above copyright notice and this permission notice shall be included in
- * all copies or substantial portions of the Software.
- *
- * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
- * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
- * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT.  IN NO EVENT SHALL
- * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR
- * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE,
- * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR
- * OTHER DEALINGS IN THE SOFTWARE.
- *
- * Authors: Ben Skeggs
- */
-
-#include "nouveau_drm.h"
-#include "nouveau_bios.h"
-#include "nouveau_pm.h"
-
-#include <subdev/bios/pll.h>
-#include <subdev/bios.h>
-#include <subdev/clock.h>
-#include <subdev/timer.h>
-#include <subdev/fb.h>
-
-static u32 read_div(struct drm_device *, int, u32, u32);
-static u32 read_pll(struct drm_device *, u32);
-
-static u32
-read_vco(struct drm_device *dev, u32 dsrc)
-{
-	struct nouveau_device *device = nouveau_dev(dev);
-	u32 ssrc = nv_rd32(device, dsrc);
-	if (!(ssrc & 0x00000100))
-		return read_pll(dev, 0x00e800);
-	return read_pll(dev, 0x00e820);
-}
-
-static u32
-read_pll(struct drm_device *dev, u32 pll)
-{
-	struct nouveau_device *device = nouveau_dev(dev);
-	u32 ctrl = nv_rd32(device, pll + 0);
-	u32 coef = nv_rd32(device, pll + 4);
-	u32 P = (coef & 0x003f0000) >> 16;
-	u32 N = (coef & 0x0000ff00) >> 8;
-	u32 M = (coef & 0x000000ff) >> 0;
-	u32 sclk, doff;
-
-	if (!(ctrl & 0x00000001))
-		return 0;
-
-	switch (pll & 0xfff000) {
-	case 0x00e000:
-		sclk = 27000;
-		P = 1;
-		break;
-	case 0x137000:
-		doff = (pll - 0x137000) / 0x20;
-		sclk = read_div(dev, doff, 0x137120, 0x137140);
-		break;
-	case 0x132000:
-		switch (pll) {
-		case 0x132000:
-			sclk = read_pll(dev, 0x132020);
-			break;
-		case 0x132020:
-			sclk = read_div(dev, 0, 0x137320, 0x137330);
-			break;
-		default:
-			return 0;
-		}
-		break;
-	default:
-		return 0;
-	}
-
-	return sclk * N / M / P;
-}
-
-static u32
-read_div(struct drm_device *dev, int doff, u32 dsrc, u32 dctl)
-{
-	struct nouveau_device *device = nouveau_dev(dev);
-	u32 ssrc = nv_rd32(device, dsrc + (doff * 4));
-	u32 sctl = nv_rd32(device, dctl + (doff * 4));
-
-	switch (ssrc & 0x00000003) {
-	case 0:
-		if ((ssrc & 0x00030000) != 0x00030000)
-			return 27000;
-		return 108000;
-	case 2:
-		return 100000;
-	case 3:
-		if (sctl & 0x80000000) {
-			u32 sclk = read_vco(dev, dsrc + (doff * 4));
-			u32 sdiv = (sctl & 0x0000003f) + 2;
-			return (sclk * 2) / sdiv;
-		}
-
-		return read_vco(dev, dsrc + (doff * 4));
-	default:
-		return 0;
-	}
-}
-
-static u32
-read_mem(struct drm_device *dev)
-{
-	struct nouveau_device *device = nouveau_dev(dev);
-	u32 ssel = nv_rd32(device, 0x1373f0);
-	if (ssel & 0x00000001)
-		return read_div(dev, 0, 0x137300, 0x137310);
-	return read_pll(dev, 0x132000);
-}
-
-static u32
-read_clk(struct drm_device *dev, int clk)
-{
-	struct nouveau_device *device = nouveau_dev(dev);
-	u32 sctl = nv_rd32(device, 0x137250 + (clk * 4));
-	u32 ssel = nv_rd32(device, 0x137100);
-	u32 sclk, sdiv;
-
-	if (ssel & (1 << clk)) {
-		if (clk < 7)
-			sclk = read_pll(dev, 0x137000 + (clk * 0x20));
-		else
-			sclk = read_pll(dev, 0x1370e0);
-		sdiv = ((sctl & 0x00003f00) >> 8) + 2;
-	} else {
-		sclk = read_div(dev, clk, 0x137160, 0x1371d0);
-		sdiv = ((sctl & 0x0000003f) >> 0) + 2;
-	}
-
-	if (sctl & 0x80000000)
-		return (sclk * 2) / sdiv;
-	return sclk;
-}
-
-int
-nvc0_pm_clocks_get(struct drm_device *dev, struct nouveau_pm_level *perflvl)
-{
-	perflvl->shader = read_clk(dev, 0x00);
-	perflvl->core   = perflvl->shader / 2;
-	perflvl->memory = read_mem(dev);
-	perflvl->rop    = read_clk(dev, 0x01);
-	perflvl->hub07  = read_clk(dev, 0x02);
-	perflvl->hub06  = read_clk(dev, 0x07);
-	perflvl->hub01  = read_clk(dev, 0x08);
-	perflvl->copy   = read_clk(dev, 0x09);
-	perflvl->daemon = read_clk(dev, 0x0c);
-	perflvl->vdec   = read_clk(dev, 0x0e);
-	return 0;
-}
-
-struct nvc0_pm_clock {
-	u32 freq;
-	u32 ssel;
-	u32 mdiv;
-	u32 dsrc;
-	u32 ddiv;
-	u32 coef;
-};
-
-struct nvc0_pm_state {
-	struct nouveau_pm_level *perflvl;
-	struct nvc0_pm_clock eng[16];
-	struct nvc0_pm_clock mem;
-};
-
-static u32
-calc_div(struct drm_device *dev, int clk, u32 ref, u32 freq, u32 *ddiv)
-{
-	u32 div = min((ref * 2) / freq, (u32)65);
-	if (div < 2)
-		div = 2;
-
-	*ddiv = div - 2;
-	return (ref * 2) / div;
-}
-
-static u32
-calc_src(struct drm_device *dev, int clk, u32 freq, u32 *dsrc, u32 *ddiv)
-{
-	u32 sclk;
-
-	/* use one of the fixed frequencies if possible */
-	*ddiv = 0x00000000;
-	switch (freq) {
-	case  27000:
-	case 108000:
-		*dsrc = 0x00000000;
-		if (freq == 108000)
-			*dsrc |= 0x00030000;
-		return freq;
-	case 100000:
-		*dsrc = 0x00000002;
-		return freq;
-	default:
-		*dsrc = 0x00000003;
-		break;
-	}
-
-	/* otherwise, calculate the closest divider */
-	sclk = read_vco(dev, clk);
-	if (clk < 7)
-		sclk = calc_div(dev, clk, sclk, freq, ddiv);
-	return sclk;
-}
-
-static u32
-calc_pll(struct drm_device *dev, int clk, u32 freq, u32 *coef)
-{
-	struct nouveau_device *device = nouveau_dev(dev);
-	struct nouveau_bios *bios = nouveau_bios(device);
-	struct nvbios_pll limits;
-	int N, M, P, ret;
-
-	ret = nvbios_pll_parse(bios, 0x137000 + (clk * 0x20), &limits);
-	if (ret)
-		return 0;
-
-	limits.refclk = read_div(dev, clk, 0x137120, 0x137140);
-	if (!limits.refclk)
-		return 0;
-
-	ret = nva3_calc_pll(dev, &limits, freq, &N, NULL, &M, &P);
-	if (ret <= 0)
-		return 0;
-
-	*coef = (P << 16) | (N << 8) | M;
-	return ret;
-}
-
-/* A (likely rather simplified and incomplete) view of the clock tree
- *
- * Key:
- *
- * S: source select
- * D: divider
- * P: pll
- * F: switch
- *
- * Engine clocks:
- *
- * 137250(D) ---- 137100(F0) ---- 137160(S)/1371d0(D) ------------------- ref
- *                      (F1) ---- 1370X0(P) ---- 137120(S)/137140(D) ---- ref
- *
- * Not all registers exist for all clocks.  For example: clocks >= 8 don't
- * have their own PLL (all tied to clock 7's PLL when in PLL mode), nor do
- * they have the divider at 1371d0, though the source selection at 137160
- * still exists.  You must use the divider at 137250 for these instead.
- *
- * Memory clock:
- *
- * TBD, read_mem() above is likely very wrong...
- *
- */
-
-static int
-calc_clk(struct drm_device *dev, int clk, struct nvc0_pm_clock *info, u32 freq)
-{
-	u32 src0, div0, div1D, div1P = 0;
-	u32 clk0, clk1 = 0;
-
-	/* invalid clock domain */
-	if (!freq)
-		return 0;
-
-	/* first possible path, using only dividers */
-	clk0 = calc_src(dev, clk, freq, &src0, &div0);
-	clk0 = calc_div(dev, clk, clk0, freq, &div1D);
-
-	/* see if we can get any closer using PLLs */
-	if (clk0 != freq && (0x00004387 & (1 << clk))) {
-		if (clk < 7)
-			clk1 = calc_pll(dev, clk, freq, &info->coef);
-		else
-			clk1 = read_pll(dev, 0x1370e0);
-		clk1 = calc_div(dev, clk, clk1, freq, &div1P);
-	}
-
-	/* select the method which gets closest to target freq */
-	if (abs((int)freq - clk0) <= abs((int)freq - clk1)) {
-		info->dsrc = src0;
-		if (div0) {
-			info->ddiv |= 0x80000000;
-			info->ddiv |= div0 << 8;
-			info->ddiv |= div0;
-		}
-		if (div1D) {
-			info->mdiv |= 0x80000000;
-			info->mdiv |= div1D;
-		}
-		info->ssel = 0;
-		info->freq = clk0;
-	} else {
-		if (div1P) {
-			info->mdiv |= 0x80000000;
-			info->mdiv |= div1P << 8;
-		}
-		info->ssel = (1 << clk);
-		info->freq = clk1;
-	}
-
-	return 0;
-}
-
-static int
-calc_mem(struct drm_device *dev, struct nvc0_pm_clock *info, u32 freq)
-{
-	struct nouveau_device *device = nouveau_dev(dev);
-	struct nouveau_bios *bios = nouveau_bios(device);
-	struct nvbios_pll pll;
-	int N, M, P, ret;
-	u32 ctrl;
-
-	/* mclk pll input freq comes from another pll, make sure it's on */
-	ctrl = nv_rd32(device, 0x132020);
-	if (!(ctrl & 0x00000001)) {
-		/* if not, program it to 567MHz.  nfi where this value comes
-		 * from - it looks like it's in the pll limits table for
-		 * 132000 but the binary driver ignores all my attempts to
-		 * change this value.
-		 */
-		nv_wr32(device, 0x137320, 0x00000103);
-		nv_wr32(device, 0x137330, 0x81200606);
-		nv_wait(device, 0x132020, 0x00010000, 0x00010000);
-		nv_wr32(device, 0x132024, 0x0001150f);
-		nv_mask(device, 0x132020, 0x00000001, 0x00000001);
-		nv_wait(device, 0x137390, 0x00020000, 0x00020000);
-		nv_mask(device, 0x132020, 0x00000004, 0x00000004);
-	}
-
-	/* for the moment, until the clock tree is better understood, use
-	 * pll mode for all clock frequencies
-	 */
-	ret = nvbios_pll_parse(bios, 0x132000, &pll);
-	if (ret == 0) {
-		pll.refclk = read_pll(dev, 0x132020);
-		if (pll.refclk) {
-			ret = nva3_calc_pll(dev, &pll, freq, &N, NULL, &M, &P);
-			if (ret > 0) {
-				info->coef = (P << 16) | (N << 8) | M;
-				return 0;
-			}
-		}
-	}
-
-	return -EINVAL;
-}
-
-void *
-nvc0_pm_clocks_pre(struct drm_device *dev, struct nouveau_pm_level *perflvl)
-{
-	struct nouveau_device *device = nouveau_dev(dev);
-	struct nvc0_pm_state *info;
-	int ret;
-
-	info = kzalloc(sizeof(*info), GFP_KERNEL);
-	if (!info)
-		return ERR_PTR(-ENOMEM);
-
-	/* NFI why this is still in the performance table, the ROPCs appear
-	 * to get their clock from clock 2 ("hub07", actually hub05 on this
-	 * chip, but, anyway...) as well.  nvatiming confirms hub05 and ROP
-	 * are always the same freq with the binary driver even when the
-	 * performance table says they should differ.
-	 */
-	if (device->chipset == 0xd9)
-		perflvl->rop = 0;
-
-	if ((ret = calc_clk(dev, 0x00, &info->eng[0x00], perflvl->shader)) ||
-	    (ret = calc_clk(dev, 0x01, &info->eng[0x01], perflvl->rop)) ||
-	    (ret = calc_clk(dev, 0x02, &info->eng[0x02], perflvl->hub07)) ||
-	    (ret = calc_clk(dev, 0x07, &info->eng[0x07], perflvl->hub06)) ||
-	    (ret = calc_clk(dev, 0x08, &info->eng[0x08], perflvl->hub01)) ||
-	    (ret = calc_clk(dev, 0x09, &info->eng[0x09], perflvl->copy)) ||
-	    (ret = calc_clk(dev, 0x0c, &info->eng[0x0c], perflvl->daemon)) ||
-	    (ret = calc_clk(dev, 0x0e, &info->eng[0x0e], perflvl->vdec))) {
-		kfree(info);
-		return ERR_PTR(ret);
-	}
-
-	if (perflvl->memory) {
-		ret = calc_mem(dev, &info->mem, perflvl->memory);
-		if (ret) {
-			kfree(info);
-			return ERR_PTR(ret);
-		}
-	}
-
-	info->perflvl = perflvl;
-	return info;
-}
-
-static void
-prog_clk(struct drm_device *dev, int clk, struct nvc0_pm_clock *info)
-{
-	struct nouveau_device *device = nouveau_dev(dev);
-
-	/* program dividers at 137160/1371d0 first */
-	if (clk < 7 && !info->ssel) {
-		nv_mask(device, 0x1371d0 + (clk * 0x04), 0x80003f3f, info->ddiv);
-		nv_wr32(device, 0x137160 + (clk * 0x04), info->dsrc);
-	}
-
-	/* switch clock to non-pll mode */
-	nv_mask(device, 0x137100, (1 << clk), 0x00000000);
-	nv_wait(device, 0x137100, (1 << clk), 0x00000000);
-
-	/* reprogram pll */
-	if (clk < 7) {
-		/* make sure it's disabled first... */
-		u32 base = 0x137000 + (clk * 0x20);
-		u32 ctrl = nv_rd32(device, base + 0x00);
-		if (ctrl & 0x00000001) {
-			nv_mask(device, base + 0x00, 0x00000004, 0x00000000);
-			nv_mask(device, base + 0x00, 0x00000001, 0x00000000);
-		}
-		/* program it to new values, if necessary */
-		if (info->ssel) {
-			nv_wr32(device, base + 0x04, info->coef);
-			nv_mask(device, base + 0x00, 0x00000001, 0x00000001);
-			nv_wait(device, base + 0x00, 0x00020000, 0x00020000);
-			nv_mask(device, base + 0x00, 0x00020004, 0x00000004);
-		}
-	}
-
-	/* select pll/non-pll mode, and program final clock divider */
-	nv_mask(device, 0x137100, (1 << clk), info->ssel);
-	nv_wait(device, 0x137100, (1 << clk), info->ssel);
-	nv_mask(device, 0x137250 + (clk * 0x04), 0x00003f3f, info->mdiv);
-}
-
-static void
-mclk_precharge(struct nouveau_mem_exec_func *exec)
-{
-}
-
-static void
-mclk_refresh(struct nouveau_mem_exec_func *exec)
-{
-}
-
-static void
-mclk_refresh_auto(struct nouveau_mem_exec_func *exec, bool enable)
-{
-	struct nouveau_device *device = nouveau_dev(exec->dev);
-	nv_wr32(device, 0x10f210, enable ? 0x80000000 : 0x00000000);
-}
-
-static void
-mclk_refresh_self(struct nouveau_mem_exec_func *exec, bool enable)
-{
-}
-
-static void
-mclk_wait(struct nouveau_mem_exec_func *exec, u32 nsec)
-{
-	udelay((nsec + 500) / 1000);
-}
-
-static u32
-mclk_mrg(struct nouveau_mem_exec_func *exec, int mr)
-{
-	struct nouveau_device *device = nouveau_dev(exec->dev);
-	struct nouveau_fb *pfb = nouveau_fb(device);
-	if (pfb->ram->type != NV_MEM_TYPE_GDDR5) {
-		if (mr <= 1)
-			return nv_rd32(device, 0x10f300 + ((mr - 0) * 4));
-		return nv_rd32(device, 0x10f320 + ((mr - 2) * 4));
-	} else {
-		if (mr == 0)
-			return nv_rd32(device, 0x10f300 + (mr * 4));
-		else
-		if (mr <= 7)
-			return nv_rd32(device, 0x10f32c + (mr * 4));
-		return nv_rd32(device, 0x10f34c);
-	}
-}
-
-static void
-mclk_mrs(struct nouveau_mem_exec_func *exec, int mr, u32 data)
-{
-	struct nouveau_device *device = nouveau_dev(exec->dev);
-	struct nouveau_fb *pfb = nouveau_fb(device);
-	if (pfb->ram->type != NV_MEM_TYPE_GDDR5) {
-		if (mr <= 1) {
-			nv_wr32(device, 0x10f300 + ((mr - 0) * 4), data);
-			if (pfb->ram->ranks > 1)
-				nv_wr32(device, 0x10f308 + ((mr - 0) * 4), data);
-		} else
-		if (mr <= 3) {
-			nv_wr32(device, 0x10f320 + ((mr - 2) * 4), data);
-			if (pfb->ram->ranks > 1)
-				nv_wr32(device, 0x10f328 + ((mr - 2) * 4), data);
-		}
-	} else {
-		if      (mr ==  0) nv_wr32(device, 0x10f300 + (mr * 4), data);
-		else if (mr <=  7) nv_wr32(device, 0x10f32c + (mr * 4), data);
-		else if (mr == 15) nv_wr32(device, 0x10f34c, data);
-	}
-}
-
-static void
-mclk_clock_set(struct nouveau_mem_exec_func *exec)
-{
-	struct nouveau_device *device = nouveau_dev(exec->dev);
-	struct nvc0_pm_state *info = exec->priv;
-	u32 ctrl = nv_rd32(device, 0x132000);
-
-	nv_wr32(device, 0x137360, 0x00000001);
-	nv_wr32(device, 0x137370, 0x00000000);
-	nv_wr32(device, 0x137380, 0x00000000);
-	if (ctrl & 0x00000001)
-		nv_wr32(device, 0x132000, (ctrl &= ~0x00000001));
-
-	nv_wr32(device, 0x132004, info->mem.coef);
-	nv_wr32(device, 0x132000, (ctrl |= 0x00000001));
-	nv_wait(device, 0x137390, 0x00000002, 0x00000002);
-	nv_wr32(device, 0x132018, 0x00005000);
-
-	nv_wr32(device, 0x137370, 0x00000001);
-	nv_wr32(device, 0x137380, 0x00000001);
-	nv_wr32(device, 0x137360, 0x00000000);
-}
-
-static void
-mclk_timing_set(struct nouveau_mem_exec_func *exec)
-{
-	struct nouveau_device *device = nouveau_dev(exec->dev);
-	struct nvc0_pm_state *info = exec->priv;
-	struct nouveau_pm_level *perflvl = info->perflvl;
-	int i;
-
-	for (i = 0; i < 5; i++)
-		nv_wr32(device, 0x10f290 + (i * 4), perflvl->timing.reg[i]);
-}
-
-static void
-prog_mem(struct drm_device *dev, struct nvc0_pm_state *info)
-{
-	struct nouveau_device *device = nouveau_dev(dev);
-	struct nouveau_mem_exec_func exec = {
-		.dev = dev,
-		.precharge = mclk_precharge,
-		.refresh = mclk_refresh,
-		.refresh_auto = mclk_refresh_auto,
-		.refresh_self = mclk_refresh_self,
-		.wait = mclk_wait,
-		.mrg = mclk_mrg,
-		.mrs = mclk_mrs,
-		.clock_set = mclk_clock_set,
-		.timing_set = mclk_timing_set,
-		.priv = info
-	};
-
-	if (device->chipset < 0xd0)
-		nv_wr32(device, 0x611200, 0x00003300);
-	else
-		nv_wr32(device, 0x62c000, 0x03030000);
-
-	nouveau_mem_exec(&exec, info->perflvl);
-
-	if (device->chipset < 0xd0)
-		nv_wr32(device, 0x611200, 0x00003330);
-	else
-		nv_wr32(device, 0x62c000, 0x03030300);
-}
-int
-nvc0_pm_clocks_set(struct drm_device *dev, void *data)
-{
-	struct nvc0_pm_state *info = data;
-	int i;
-
-	if (info->mem.coef)
-		prog_mem(dev, info);
-
-	for (i = 0; i < 16; i++) {
-		if (!info->eng[i].freq)
-			continue;
-		prog_clk(dev, i, &info->eng[i]);
-	}
-
-	kfree(info);
-	return 0;
-}
diff --git a/drivers/gpu/drm/omapdrm/Kconfig b/drivers/gpu/drm/omapdrm/Kconfig
index 20c41e73d448..6c220cd3497a 100644
--- a/drivers/gpu/drm/omapdrm/Kconfig
+++ b/drivers/gpu/drm/omapdrm/Kconfig
@@ -5,6 +5,7 @@ config DRM_OMAP
 	depends on ARCH_OMAP2PLUS || ARCH_MULTIPLATFORM
 	depends on OMAP2_DSS
 	select DRM_KMS_HELPER
+	select DRM_KMS_FB_HELPER
 	select FB_SYS_FILLRECT
 	select FB_SYS_COPYAREA
 	select FB_SYS_IMAGEBLIT
diff --git a/drivers/gpu/drm/omapdrm/omap_drv.c b/drivers/gpu/drm/omapdrm/omap_drv.c
index 2603d909f49c..e7fa3cd96743 100644
--- a/drivers/gpu/drm/omapdrm/omap_drv.c
+++ b/drivers/gpu/drm/omapdrm/omap_drv.c
@@ -620,7 +620,6 @@ static struct drm_driver omap_drm_driver = {
 		.prime_fd_to_handle = drm_gem_prime_fd_to_handle,
 		.gem_prime_export = omap_gem_prime_export,
 		.gem_prime_import = omap_gem_prime_import,
-		.gem_init_object = omap_gem_init_object,
 		.gem_free_object = omap_gem_free_object,
 		.gem_vm_ops = &omap_gem_vm_ops,
 		.dumb_create = omap_gem_dumb_create,
diff --git a/drivers/gpu/drm/omapdrm/omap_drv.h b/drivers/gpu/drm/omapdrm/omap_drv.h
index 30b95b736658..07847693cf49 100644
--- a/drivers/gpu/drm/omapdrm/omap_drv.h
+++ b/drivers/gpu/drm/omapdrm/omap_drv.h
@@ -220,7 +220,6 @@ struct drm_gem_object *omap_gem_new(struct drm_device *dev,
 int omap_gem_new_handle(struct drm_device *dev, struct drm_file *file,
 		union omap_gem_size gsize, uint32_t flags, uint32_t *handle);
 void omap_gem_free_object(struct drm_gem_object *obj);
-int omap_gem_init_object(struct drm_gem_object *obj);
 void *omap_gem_vaddr(struct drm_gem_object *obj);
 int omap_gem_dumb_map_offset(struct drm_file *file, struct drm_device *dev,
 		uint32_t handle, uint64_t *offset);
diff --git a/drivers/gpu/drm/omapdrm/omap_gem.c b/drivers/gpu/drm/omapdrm/omap_gem.c
index 533f6ebec531..5aec3e81fe24 100644
--- a/drivers/gpu/drm/omapdrm/omap_gem.c
+++ b/drivers/gpu/drm/omapdrm/omap_gem.c
@@ -1274,11 +1274,6 @@ unlock:
 	return ret;
 }
 
-int omap_gem_init_object(struct drm_gem_object *obj)
-{
-	return -EINVAL;          /* unused */
-}
-
 /* don't call directly.. called from GEM core when it is time to actually
  * free the object..
  */
diff --git a/drivers/gpu/drm/omapdrm/omap_irq.c b/drivers/gpu/drm/omapdrm/omap_irq.c
index 9263db117ff8..cb858600185f 100644
--- a/drivers/gpu/drm/omapdrm/omap_irq.c
+++ b/drivers/gpu/drm/omapdrm/omap_irq.c
@@ -261,7 +261,7 @@ int omap_drm_irq_install(struct drm_device *dev)
 		mutex_unlock(&dev->struct_mutex);
 		return -EBUSY;
 	}
-	dev->irq_enabled = 1;
+	dev->irq_enabled = true;
 	mutex_unlock(&dev->struct_mutex);
 
 	/* Before installing handler */
@@ -272,7 +272,7 @@ int omap_drm_irq_install(struct drm_device *dev)
 
 	if (ret < 0) {
 		mutex_lock(&dev->struct_mutex);
-		dev->irq_enabled = 0;
+		dev->irq_enabled = false;
 		mutex_unlock(&dev->struct_mutex);
 		return ret;
 	}
@@ -283,7 +283,7 @@ int omap_drm_irq_install(struct drm_device *dev)
 
 	if (ret < 0) {
 		mutex_lock(&dev->struct_mutex);
-		dev->irq_enabled = 0;
+		dev->irq_enabled = false;
 		mutex_unlock(&dev->struct_mutex);
 		dispc_free_irq(dev);
 	}
@@ -294,11 +294,12 @@ int omap_drm_irq_install(struct drm_device *dev)
 int omap_drm_irq_uninstall(struct drm_device *dev)
 {
 	unsigned long irqflags;
-	int irq_enabled, i;
+	bool irq_enabled;
+	int i;
 
 	mutex_lock(&dev->struct_mutex);
 	irq_enabled = dev->irq_enabled;
-	dev->irq_enabled = 0;
+	dev->irq_enabled = false;
 	mutex_unlock(&dev->struct_mutex);
 
 	/*
@@ -307,9 +308,9 @@ int omap_drm_irq_uninstall(struct drm_device *dev)
 	if (dev->num_crtcs) {
 		spin_lock_irqsave(&dev->vbl_lock, irqflags);
 		for (i = 0; i < dev->num_crtcs; i++) {
-			DRM_WAKEUP(&dev->vbl_queue[i]);
-			dev->vblank_enabled[i] = 0;
-			dev->last_vblank[i] =
+			DRM_WAKEUP(&dev->vblank[i].queue);
+			dev->vblank[i].enabled = false;
+			dev->vblank[i].last =
 				dev->driver->get_vblank_counter(dev, i);
 		}
 		spin_unlock_irqrestore(&dev->vbl_lock, irqflags);
diff --git a/drivers/gpu/drm/qxl/Kconfig b/drivers/gpu/drm/qxl/Kconfig
index d6c12796023c..037d324bf58f 100644
--- a/drivers/gpu/drm/qxl/Kconfig
+++ b/drivers/gpu/drm/qxl/Kconfig
@@ -6,6 +6,7 @@ config DRM_QXL
 	select FB_SYS_IMAGEBLIT
 	select FB_DEFERRED_IO
         select DRM_KMS_HELPER
+	select DRM_KMS_FB_HELPER
         select DRM_TTM
 	help
 		QXL virtual GPU for Spice virtualization desktop integration. Do not enable this driver unless your distro ships a corresponding X.org QXL driver that can handle kernel modesetting.
diff --git a/drivers/gpu/drm/qxl/qxl_display.c b/drivers/gpu/drm/qxl/qxl_display.c
index 835caba026d3..5e827c29d194 100644
--- a/drivers/gpu/drm/qxl/qxl_display.c
+++ b/drivers/gpu/drm/qxl/qxl_display.c
@@ -107,10 +107,17 @@ void qxl_display_read_client_monitors_config(struct qxl_device *qdev)
 		qxl_io_log(qdev, "failed crc check for client_monitors_config,"
 				 " retrying\n");
 	}
-	drm_helper_hpd_irq_event(qdev->ddev);
+
+	if (!drm_helper_hpd_irq_event(qdev->ddev)) {
+		/* notify that the monitor configuration changed, to
+		   adjust at the arbitrary resolution */
+		drm_kms_helper_hotplug_event(qdev->ddev);
+	}
 }
 
-static int qxl_add_monitors_config_modes(struct drm_connector *connector)
+static int qxl_add_monitors_config_modes(struct drm_connector *connector,
+                                         unsigned *pwidth,
+                                         unsigned *pheight)
 {
 	struct drm_device *dev = connector->dev;
 	struct qxl_device *qdev = dev->dev_private;
@@ -126,11 +133,15 @@ static int qxl_add_monitors_config_modes(struct drm_connector *connector)
 	mode = drm_cvt_mode(dev, head->width, head->height, 60, false, false,
 			    false);
 	mode->type |= DRM_MODE_TYPE_PREFERRED;
+	*pwidth = head->width;
+	*pheight = head->height;
 	drm_mode_probed_add(connector, mode);
 	return 1;
 }
 
-static int qxl_add_common_modes(struct drm_connector *connector)
+static int qxl_add_common_modes(struct drm_connector *connector,
+                                unsigned pwidth,
+                                unsigned pheight)
 {
 	struct drm_device *dev = connector->dev;
 	struct drm_display_mode *mode = NULL;
@@ -159,12 +170,9 @@ static int qxl_add_common_modes(struct drm_connector *connector)
 	};
 
 	for (i = 0; i < ARRAY_SIZE(common_modes); i++) {
-		if (common_modes[i].w < 320 || common_modes[i].h < 200)
-			continue;
-
 		mode = drm_cvt_mode(dev, common_modes[i].w, common_modes[i].h,
 				    60, false, false, false);
-		if (common_modes[i].w == 1024 && common_modes[i].h == 768)
+		if (common_modes[i].w == pwidth && common_modes[i].h == pheight)
 			mode->type |= DRM_MODE_TYPE_PREFERRED;
 		drm_mode_probed_add(connector, mode);
 	}
@@ -720,16 +728,18 @@ static int qxl_conn_get_modes(struct drm_connector *connector)
 {
 	int ret = 0;
 	struct qxl_device *qdev = connector->dev->dev_private;
+	unsigned pwidth = 1024;
+	unsigned pheight = 768;
 
 	DRM_DEBUG_KMS("monitors_config=%p\n", qdev->monitors_config);
 	/* TODO: what should we do here? only show the configured modes for the
 	 * device, or allow the full list, or both? */
 	if (qdev->monitors_config && qdev->monitors_config->count) {
-		ret = qxl_add_monitors_config_modes(connector);
+		ret = qxl_add_monitors_config_modes(connector, &pwidth, &pheight);
 		if (ret < 0)
 			return ret;
 	}
-	ret += qxl_add_common_modes(connector);
+	ret += qxl_add_common_modes(connector, pwidth, pheight);
 	return ret;
 }
 
@@ -793,7 +803,10 @@ static enum drm_connector_status qxl_conn_detect(
 		     qdev->client_monitors_config->count > output->index &&
 		     qxl_head_enabled(&qdev->client_monitors_config->heads[output->index]));
 
-	DRM_DEBUG("\n");
+	DRM_DEBUG("#%d connected: %d\n", output->index, connected);
+	if (!connected)
+		qxl_monitors_config_set(qdev, output->index, 0, 0, 0, 0, 0);
+
 	return connected ? connector_status_connected
 			 : connector_status_disconnected;
 }
@@ -835,8 +848,21 @@ static const struct drm_encoder_funcs qxl_enc_funcs = {
 	.destroy = qxl_enc_destroy,
 };
 
+static int qxl_mode_create_hotplug_mode_update_property(struct qxl_device *qdev)
+{
+	if (qdev->hotplug_mode_update_property)
+		return 0;
+
+	qdev->hotplug_mode_update_property =
+		drm_property_create_range(qdev->ddev, DRM_MODE_PROP_IMMUTABLE,
+					  "hotplug_mode_update", 0, 1);
+
+	return 0;
+}
+
 static int qdev_output_init(struct drm_device *dev, int num_output)
 {
+	struct qxl_device *qdev = dev->dev_private;
 	struct qxl_output *qxl_output;
 	struct drm_connector *connector;
 	struct drm_encoder *encoder;
@@ -863,6 +889,8 @@ static int qdev_output_init(struct drm_device *dev, int num_output)
 	drm_encoder_helper_add(encoder, &qxl_enc_helper_funcs);
 	drm_connector_helper_add(connector, &qxl_connector_helper_funcs);
 
+	drm_object_attach_property(&connector->base,
+				   qdev->hotplug_mode_update_property, 0);
 	drm_sysfs_connector_add(connector);
 	return 0;
 }
@@ -975,6 +1003,9 @@ int qxl_modeset_init(struct qxl_device *qdev)
 	qdev->ddev->mode_config.max_height = 8192;
 
 	qdev->ddev->mode_config.fb_base = qdev->vram_base;
+
+	qxl_mode_create_hotplug_mode_update_property(qdev);
+
 	for (i = 0 ; i < qxl_num_crtc; ++i) {
 		qdev_crtc_init(qdev->ddev, i);
 		qdev_output_init(qdev->ddev, i);
diff --git a/drivers/gpu/drm/qxl/qxl_drv.c b/drivers/gpu/drm/qxl/qxl_drv.c
index 514118ae72d4..fee8748bdca5 100644
--- a/drivers/gpu/drm/qxl/qxl_drv.c
+++ b/drivers/gpu/drm/qxl/qxl_drv.c
@@ -225,7 +225,6 @@ static struct drm_driver qxl_driver = {
 	.debugfs_init = qxl_debugfs_init,
 	.debugfs_cleanup = qxl_debugfs_takedown,
 #endif
-	.gem_init_object = qxl_gem_object_init,
 	.gem_free_object = qxl_gem_object_free,
 	.gem_open_object = qxl_gem_object_open,
 	.gem_close_object = qxl_gem_object_close,
diff --git a/drivers/gpu/drm/qxl/qxl_drv.h b/drivers/gpu/drm/qxl/qxl_drv.h
index f7c9adde46a0..7bda32f68d3b 100644
--- a/drivers/gpu/drm/qxl/qxl_drv.h
+++ b/drivers/gpu/drm/qxl/qxl_drv.h
@@ -323,6 +323,8 @@ struct qxl_device {
 	struct work_struct gc_work;
 
 	struct work_struct fb_work;
+
+	struct drm_property *hotplug_mode_update_property;
 };
 
 /* forward declaration for QXL_INFO_IO */
@@ -412,7 +414,6 @@ int qxl_gem_object_create_with_handle(struct qxl_device *qdev,
 				      struct qxl_surface *surf,
 				      struct qxl_bo **qobj,
 				      uint32_t *handle);
-int qxl_gem_object_init(struct drm_gem_object *obj);
 void qxl_gem_object_free(struct drm_gem_object *gobj);
 int qxl_gem_object_open(struct drm_gem_object *obj, struct drm_file *file_priv);
 void qxl_gem_object_close(struct drm_gem_object *obj,
diff --git a/drivers/gpu/drm/qxl/qxl_fb.c b/drivers/gpu/drm/qxl/qxl_fb.c
index 88722f233430..f437b30ce689 100644
--- a/drivers/gpu/drm/qxl/qxl_fb.c
+++ b/drivers/gpu/drm/qxl/qxl_fb.c
@@ -108,7 +108,7 @@ static void qxl_fb_dirty_flush(struct fb_info *info)
 	u32 x1, x2, y1, y2;
 
 	/* TODO: hard coding 32 bpp */
-	int stride = qfbdev->qfb.base.pitches[0] * 4;
+	int stride = qfbdev->qfb.base.pitches[0];
 
 	x1 = qfbdev->dirty.x1;
 	x2 = qfbdev->dirty.x2;
diff --git a/drivers/gpu/drm/qxl/qxl_gem.c b/drivers/gpu/drm/qxl/qxl_gem.c
index 1648e4125af7..b96f0c9d89b2 100644
--- a/drivers/gpu/drm/qxl/qxl_gem.c
+++ b/drivers/gpu/drm/qxl/qxl_gem.c
@@ -28,12 +28,6 @@
 #include "qxl_drv.h"
 #include "qxl_object.h"
 
-int qxl_gem_object_init(struct drm_gem_object *obj)
-{
-	/* we do nothings here */
-	return 0;
-}
-
 void qxl_gem_object_free(struct drm_gem_object *gobj)
 {
 	struct qxl_bo *qobj = gem_to_qxl_bo(gobj);
diff --git a/drivers/gpu/drm/qxl/qxl_kms.c b/drivers/gpu/drm/qxl/qxl_kms.c
index 9e8da9ee9731..e5ca498be920 100644
--- a/drivers/gpu/drm/qxl/qxl_kms.c
+++ b/drivers/gpu/drm/qxl/qxl_kms.c
@@ -120,7 +120,7 @@ int qxl_device_init(struct qxl_device *qdev,
 		    struct pci_dev *pdev,
 		    unsigned long flags)
 {
-	int r;
+	int r, sb;
 
 	qdev->dev = &pdev->dev;
 	qdev->ddev = ddev;
@@ -136,21 +136,39 @@ int qxl_device_init(struct qxl_device *qdev,
 	qdev->rom_base = pci_resource_start(pdev, 2);
 	qdev->rom_size = pci_resource_len(pdev, 2);
 	qdev->vram_base = pci_resource_start(pdev, 0);
-	qdev->surfaceram_base = pci_resource_start(pdev, 1);
-	qdev->surfaceram_size = pci_resource_len(pdev, 1);
 	qdev->io_base = pci_resource_start(pdev, 3);
 
 	qdev->vram_mapping = io_mapping_create_wc(qdev->vram_base, pci_resource_len(pdev, 0));
-	qdev->surface_mapping = io_mapping_create_wc(qdev->surfaceram_base, qdev->surfaceram_size);
-	DRM_DEBUG_KMS("qxl: vram %llx-%llx(%dM %dk), surface %llx-%llx(%dM %dk)\n",
+
+	if (pci_resource_len(pdev, 4) > 0) {
+		/* 64bit surface bar present */
+		sb = 4;
+		qdev->surfaceram_base = pci_resource_start(pdev, sb);
+		qdev->surfaceram_size = pci_resource_len(pdev, sb);
+		qdev->surface_mapping =
+			io_mapping_create_wc(qdev->surfaceram_base,
+					     qdev->surfaceram_size);
+	}
+	if (qdev->surface_mapping == NULL) {
+		/* 64bit surface bar not present (or mapping failed) */
+		sb = 1;
+		qdev->surfaceram_base = pci_resource_start(pdev, sb);
+		qdev->surfaceram_size = pci_resource_len(pdev, sb);
+		qdev->surface_mapping =
+			io_mapping_create_wc(qdev->surfaceram_base,
+					     qdev->surfaceram_size);
+	}
+
+	DRM_DEBUG_KMS("qxl: vram %llx-%llx(%dM %dk), surface %llx-%llx(%dM %dk, %s)\n",
 		 (unsigned long long)qdev->vram_base,
 		 (unsigned long long)pci_resource_end(pdev, 0),
 		 (int)pci_resource_len(pdev, 0) / 1024 / 1024,
 		 (int)pci_resource_len(pdev, 0) / 1024,
 		 (unsigned long long)qdev->surfaceram_base,
-		 (unsigned long long)pci_resource_end(pdev, 1),
+		 (unsigned long long)pci_resource_end(pdev, sb),
 		 (int)qdev->surfaceram_size / 1024 / 1024,
-		 (int)qdev->surfaceram_size / 1024);
+		 (int)qdev->surfaceram_size / 1024,
+		 (sb == 4) ? "64bit" : "32bit");
 
 	qdev->rom = ioremap(qdev->rom_base, qdev->rom_size);
 	if (!qdev->rom) {
@@ -230,9 +248,13 @@ int qxl_device_init(struct qxl_device *qdev,
 	qdev->surfaces_mem_slot = setup_slot(qdev, 1,
 		(unsigned long)qdev->surfaceram_base,
 		(unsigned long)qdev->surfaceram_base + qdev->surfaceram_size);
-	DRM_INFO("main mem slot %d [%lx,%x)\n",
-		qdev->main_mem_slot,
-		(unsigned long)qdev->vram_base, qdev->rom->ram_header_offset);
+	DRM_INFO("main mem slot %d [%lx,%x]\n",
+		 qdev->main_mem_slot,
+		 (unsigned long)qdev->vram_base, qdev->rom->ram_header_offset);
+	DRM_INFO("surface mem slot %d [%lx,%lx]\n",
+		 qdev->surfaces_mem_slot,
+		 (unsigned long)qdev->surfaceram_base,
+		 (unsigned long)qdev->surfaceram_size);
 
 
 	qdev->gc_queue = create_singlethread_workqueue("qxl_gc");
diff --git a/drivers/gpu/drm/qxl/qxl_ttm.c b/drivers/gpu/drm/qxl/qxl_ttm.c
index 037786d7c1dc..c7e7e6590c2b 100644
--- a/drivers/gpu/drm/qxl/qxl_ttm.c
+++ b/drivers/gpu/drm/qxl/qxl_ttm.c
@@ -516,6 +516,8 @@ int qxl_ttm_init(struct qxl_device *qdev)
 		 (unsigned)qdev->vram_size / (1024 * 1024));
 	DRM_INFO("qxl: %luM of IO pages memory ready (VRAM domain)\n",
 		 ((unsigned)num_io_pages * PAGE_SIZE) / (1024 * 1024));
+	DRM_INFO("qxl: %uM of Surface memory size\n",
+		 (unsigned)qdev->surfaceram_size / (1024 * 1024));
 	if (unlikely(qdev->mman.bdev.dev_mapping == NULL))
 		qdev->mman.bdev.dev_mapping = qdev->ddev->dev_mapping;
 	r = qxl_ttm_debugfs_init(qdev);
diff --git a/drivers/gpu/drm/radeon/atombios.h b/drivers/gpu/drm/radeon/atombios.h
index af10f8571d87..92be50c39ffd 100644
--- a/drivers/gpu/drm/radeon/atombios.h
+++ b/drivers/gpu/drm/radeon/atombios.h
@@ -1711,7 +1711,9 @@ typedef struct _PIXEL_CLOCK_PARAMETERS_V6
 #define PIXEL_CLOCK_V6_MISC_HDMI_BPP_MASK           0x0c
 #define PIXEL_CLOCK_V6_MISC_HDMI_24BPP              0x00
 #define PIXEL_CLOCK_V6_MISC_HDMI_36BPP              0x04
+#define PIXEL_CLOCK_V6_MISC_HDMI_36BPP_V6           0x08    //for V6, the correct defintion for 36bpp should be 2 for 36bpp(2:1)
 #define PIXEL_CLOCK_V6_MISC_HDMI_30BPP              0x08
+#define PIXEL_CLOCK_V6_MISC_HDMI_30BPP_V6           0x04    //for V6, the correct defintion for 30bpp should be 1 for 36bpp(5:4)
 #define PIXEL_CLOCK_V6_MISC_HDMI_48BPP              0x0c
 #define PIXEL_CLOCK_V6_MISC_REF_DIV_SRC             0x10
 #define PIXEL_CLOCK_V6_MISC_GEN_DPREFCLK            0x40
@@ -2223,7 +2225,7 @@ typedef struct	_SET_VOLTAGE_PARAMETERS_V2
   USHORT   usVoltageLevel;              // real voltage level
 }SET_VOLTAGE_PARAMETERS_V2;
 
-
+// used by both SetVoltageTable v1.3 and v1.4
 typedef struct	_SET_VOLTAGE_PARAMETERS_V1_3
 {
   UCHAR    ucVoltageType;               // To tell which voltage to set up, VDDC/MVDDC/MVDDQ/VDDCI
@@ -2290,15 +2292,36 @@ typedef struct  _GET_LEAKAGE_VOLTAGE_INFO_OUTPUT_PARAMETER_V1_1
 #define	ATOM_GET_VOLTAGE_VID                0x00
 #define ATOM_GET_VOTLAGE_INIT_SEQ           0x03
 #define ATOM_GET_VOLTTAGE_PHASE_PHASE_VID   0x04
-// for SI, this state map to 0xff02 voltage state in Power Play table, which is power boost state
-#define	ATOM_GET_VOLTAGE_STATE0_LEAKAGE_VID 0x10
+#define ATOM_GET_VOLTAGE_SVID2              0x07        //Get SVI2 Regulator Info
 
+// for SI, this state map to 0xff02 voltage state in Power Play table, which is power boost state
+#define ATOM_GET_VOLTAGE_STATE0_LEAKAGE_VID 0x10
 // for SI, this state map to 0xff01 voltage state in Power Play table, which is performance state
 #define	ATOM_GET_VOLTAGE_STATE1_LEAKAGE_VID 0x11
-// undefined power state
+
 #define	ATOM_GET_VOLTAGE_STATE2_LEAKAGE_VID 0x12
 #define	ATOM_GET_VOLTAGE_STATE3_LEAKAGE_VID 0x13
 
+// New Added from CI Hawaii for GetVoltageInfoTable, input parameter structure
+typedef struct  _GET_VOLTAGE_INFO_INPUT_PARAMETER_V1_2
+{
+  UCHAR    ucVoltageType;               // Input: To tell which voltage to set up, VDDC/MVDDC/MVDDQ/VDDCI
+  UCHAR    ucVoltageMode;               // Input: Indicate action: Get voltage info
+  USHORT   usVoltageLevel;              // Input: real voltage level in unit of mv or Voltage Phase (0, 1, 2, .. ) or Leakage Id 
+  ULONG    ulSCLKFreq;                  // Input: when ucVoltageMode= ATOM_GET_VOLTAGE_EVV_VOLTAGE, DPM state SCLK frequency, Define in PPTable SCLK/Voltage dependence table
+}GET_VOLTAGE_INFO_INPUT_PARAMETER_V1_2;
+
+// New in GetVoltageInfo v1.2 ucVoltageMode
+#define ATOM_GET_VOLTAGE_EVV_VOLTAGE        0x09        
+
+// New Added from CI Hawaii for EVV feature 
+typedef struct  _GET_EVV_VOLTAGE_INFO_OUTPUT_PARAMETER_V1_2
+{
+  USHORT   usVoltageLevel;                               // real voltage level in unit of mv
+  USHORT   usVoltageId;                                  // Voltage Id programmed in Voltage Regulator
+  ULONG    ulReseved;
+}GET_EVV_VOLTAGE_INFO_OUTPUT_PARAMETER_V1_2;
+
 /****************************************************************************/	
 // Structures used by TVEncoderControlTable
 /****************************************************************************/	
@@ -3864,6 +3887,8 @@ typedef struct _ATOM_GPIO_PIN_ASSIGNMENT
 #define PP_AC_DC_SWITCH_GPIO_PINID          60
 //from SMU7.x, if ucGPIO_ID=VDDC_REGULATOR_VRHOT_GPIO_PINID in GPIO_LUTable, VRHot feature is enable
 #define VDDC_VRHOT_GPIO_PINID               61
+//if ucGPIO_ID=VDDC_PCC_GPIO_PINID in GPIO_LUTable, Peak Current Control feature is enabled
+#define VDDC_PCC_GPIO_PINID                 62
 
 typedef struct _ATOM_GPIO_PIN_LUT
 {
@@ -4169,10 +4194,10 @@ typedef struct _ATOM_COMMON_RECORD_HEADER
 #define ATOM_OBJECT_LINK_RECORD_TYPE                   18 //Once this record is present under one object, it indicats the oobject is linked to another obj described by the record
 #define ATOM_CONNECTOR_REMOTE_CAP_RECORD_TYPE          19
 #define ATOM_ENCODER_CAP_RECORD_TYPE                   20
-
+#define ATOM_BRACKET_LAYOUT_RECORD_TYPE                21
 
 //Must be updated when new record type is added,equal to that record definition!
-#define ATOM_MAX_OBJECT_RECORD_NUMBER             ATOM_ENCODER_CAP_RECORD_TYPE
+#define ATOM_MAX_OBJECT_RECORD_NUMBER             ATOM_BRACKET_LAYOUT_RECORD_TYPE
 
 typedef struct  _ATOM_I2C_RECORD
 {
@@ -4397,6 +4422,31 @@ typedef struct _ATOM_CONNECTOR_REMOTE_CAP_RECORD
   USHORT                      usReserved;
 }ATOM_CONNECTOR_REMOTE_CAP_RECORD;
 
+typedef struct  _ATOM_CONNECTOR_LAYOUT_INFO
+{
+   USHORT usConnectorObjectId;
+   UCHAR  ucConnectorType;
+   UCHAR  ucPosition;
+}ATOM_CONNECTOR_LAYOUT_INFO;
+
+// define ATOM_CONNECTOR_LAYOUT_INFO.ucConnectorType to describe the display connector size
+#define CONNECTOR_TYPE_DVI_D                 1
+#define CONNECTOR_TYPE_DVI_I                 2
+#define CONNECTOR_TYPE_VGA                   3
+#define CONNECTOR_TYPE_HDMI                  4
+#define CONNECTOR_TYPE_DISPLAY_PORT          5
+#define CONNECTOR_TYPE_MINI_DISPLAY_PORT     6
+
+typedef struct  _ATOM_BRACKET_LAYOUT_RECORD
+{
+  ATOM_COMMON_RECORD_HEADER   sheader;
+  UCHAR                       ucLength;
+  UCHAR                       ucWidth;
+  UCHAR                       ucConnNum;
+  UCHAR                       ucReserved;
+  ATOM_CONNECTOR_LAYOUT_INFO  asConnInfo[1];
+}ATOM_BRACKET_LAYOUT_RECORD;
+
 /****************************************************************************/	
 // ASIC voltage data table
 /****************************************************************************/	
@@ -4524,8 +4574,9 @@ typedef struct _ATOM_VOLTAGE_OBJECT_HEADER_V3{
 #define VOLTAGE_OBJ_VR_I2C_INIT_SEQ          3        //VOLTAGE REGULATOR INIT sequece through I2C -> ATOM_I2C_VOLTAGE_OBJECT_V3
 #define VOLTAGE_OBJ_PHASE_LUT                4        //Set Vregulator Phase lookup table ->ATOM_GPIO_VOLTAGE_OBJECT_V3
 #define VOLTAGE_OBJ_SVID2                    7        //Indicate voltage control by SVID2 ->ATOM_SVID2_VOLTAGE_OBJECT_V3
-#define	VOLTAGE_OBJ_PWRBOOST_LEAKAGE_LUT     0x10     //Powerboost Voltage and LeakageId lookup table->ATOM_LEAKAGE_VOLTAGE_OBJECT_V3
-#define	VOLTAGE_OBJ_HIGH_STATE_LEAKAGE_LUT   0x11     //High voltage state Voltage and LeakageId lookup table->ATOM_LEAKAGE_VOLTAGE_OBJECT_V3
+#define VOLTAGE_OBJ_EVV                      8 
+#define VOLTAGE_OBJ_PWRBOOST_LEAKAGE_LUT     0x10     //Powerboost Voltage and LeakageId lookup table->ATOM_LEAKAGE_VOLTAGE_OBJECT_V3
+#define VOLTAGE_OBJ_HIGH_STATE_LEAKAGE_LUT   0x11     //High voltage state Voltage and LeakageId lookup table->ATOM_LEAKAGE_VOLTAGE_OBJECT_V3
 #define VOLTAGE_OBJ_HIGH1_STATE_LEAKAGE_LUT  0x12     //High1 voltage state Voltage and LeakageId lookup table->ATOM_LEAKAGE_VOLTAGE_OBJECT_V3
 
 typedef struct  _VOLTAGE_LUT_ENTRY_V2
@@ -4552,6 +4603,10 @@ typedef struct  _ATOM_I2C_VOLTAGE_OBJECT_V3
    VOLTAGE_LUT_ENTRY asVolI2cLut[1];        // end with 0xff
 }ATOM_I2C_VOLTAGE_OBJECT_V3;
 
+// ATOM_I2C_VOLTAGE_OBJECT_V3.ucVoltageControlFlag
+#define VOLTAGE_DATA_ONE_BYTE                0
+#define VOLTAGE_DATA_TWO_BYTE                1
+
 typedef struct  _ATOM_GPIO_VOLTAGE_OBJECT_V3
 {
    ATOM_VOLTAGE_OBJECT_HEADER_V3 sHeader;   // voltage mode = VOLTAGE_OBJ_GPIO_LUT or VOLTAGE_OBJ_PHASE_LUT
@@ -4584,7 +4639,8 @@ typedef struct  _ATOM_SVID2_VOLTAGE_OBJECT_V3
 // 1:0 – offset trim, 
    USHORT   usLoadLine_PSI;    
 // GPU GPIO pin Id to SVID2 regulator VRHot pin. possible value 0~31. 0 means GPIO0, 31 means GPIO31
-   UCHAR    ucReserved[2];
+   UCHAR    ucSVDGpioId;     //0~31 indicate GPIO0~31
+   UCHAR    ucSVCGpioId;     //0~31 indicate GPIO0~31
    ULONG    ulReserved;
 }ATOM_SVID2_VOLTAGE_OBJECT_V3;
 
@@ -4637,6 +4693,49 @@ typedef struct  _ATOM_ASIC_PROFILING_INFO_V2_1
   USHORT usElbVDDCI_LevelArrayOffset;    // offset of 2 dimension voltage level USHORT array
 }ATOM_ASIC_PROFILING_INFO_V2_1;
 
+typedef struct  _ATOM_ASIC_PROFILING_INFO_V3_1
+{
+  ATOM_COMMON_TABLE_HEADER         asHeader; 
+  ULONG  ulEvvDerateTdp;
+  ULONG  ulEvvDerateTdc;
+  ULONG  ulBoardCoreTemp;
+  ULONG  ulMaxVddc;
+  ULONG  ulMinVddc;
+  ULONG  ulLoadLineSlop;
+  ULONG  ulLeakageTemp;
+  ULONG  ulLeakageVoltage;
+  ULONG  ulCACmEncodeRange;
+  ULONG  ulCACmEncodeAverage;
+  ULONG  ulCACbEncodeRange;
+  ULONG  ulCACbEncodeAverage;
+  ULONG  ulKt_bEncodeRange;
+  ULONG  ulKt_bEncodeAverage;
+  ULONG  ulKv_mEncodeRange;
+  ULONG  ulKv_mEncodeAverage;
+  ULONG  ulKv_bEncodeRange;
+  ULONG  ulKv_bEncodeAverage;
+  ULONG  ulLkgEncodeLn_MaxDivMin;
+  ULONG  ulLkgEncodeMin;
+  ULONG  ulEfuseLogisticAlpha;
+  USHORT usPowerDpm0;
+  USHORT usCurrentDpm0;
+  USHORT usPowerDpm1;
+  USHORT usCurrentDpm1;
+  USHORT usPowerDpm2;
+  USHORT usCurrentDpm2;
+  USHORT usPowerDpm3;
+  USHORT usCurrentDpm3;
+  USHORT usPowerDpm4;
+  USHORT usCurrentDpm4;
+  USHORT usPowerDpm5;
+  USHORT usCurrentDpm5;
+  USHORT usPowerDpm6;
+  USHORT usCurrentDpm6;
+  USHORT usPowerDpm7;
+  USHORT usCurrentDpm7;
+}ATOM_ASIC_PROFILING_INFO_V3_1;
+
+
 typedef struct _ATOM_POWER_SOURCE_OBJECT
 {
 	UCHAR	ucPwrSrcId;													// Power source
@@ -5808,6 +5907,8 @@ typedef struct _ATOM_ASIC_INTERNAL_SS_INFO_V3
 #define ATOM_S7_DOS_MODE_PIXEL_DEPTHb0      0x0C
 #define ATOM_S7_DOS_MODE_PIXEL_FORMATb0     0xF0
 #define ATOM_S7_DOS_8BIT_DAC_ENb1           0x01
+#define ATOM_S7_ASIC_INIT_COMPLETEb1        0x02
+#define ATOM_S7_ASIC_INIT_COMPLETE_MASK     0x00000200
 #define ATOM_S7_DOS_MODE_NUMBERw1           0x0FFFF
 
 #define ATOM_S7_DOS_8BIT_DAC_EN_SHIFT       8
@@ -6242,6 +6343,7 @@ typedef struct _ATOM_MC_INIT_PARAM_TABLE
 #define _128Mx32            0x53
 #define _256Mx8             0x61
 #define _256Mx16            0x62
+#define _512Mx8             0x71
 
 #define SAMSUNG             0x1
 #define INFINEON            0x2
@@ -6987,9 +7089,10 @@ typedef struct _ATOM_DISP_OUT_INFO_V3
   UCHAR  ucMaxDispEngineNum;
   UCHAR  ucMaxActiveDispEngineNum;
   UCHAR  ucMaxPPLLNum;
-  UCHAR  ucCoreRefClkSource;                          // value of CORE_REF_CLK_SOURCE
-  UCHAR  ucReserved[3];
-	ASIC_TRANSMITTER_INFO_V2  asTransmitterInfo[1];     // for alligment only
+  UCHAR  ucCoreRefClkSource;                    // value of CORE_REF_CLK_SOURCE
+  UCHAR  ucDispCaps;
+  UCHAR  ucReserved[2];
+  ASIC_TRANSMITTER_INFO_V2  asTransmitterInfo[1];     // for alligment only
 }ATOM_DISP_OUT_INFO_V3;
 
 //ucDispCaps
diff --git a/drivers/gpu/drm/radeon/atombios_crtc.c b/drivers/gpu/drm/radeon/atombios_crtc.c
index bf87f6d435f8..80a20120e625 100644
--- a/drivers/gpu/drm/radeon/atombios_crtc.c
+++ b/drivers/gpu/drm/radeon/atombios_crtc.c
@@ -1753,7 +1753,7 @@ static int radeon_atom_pick_pll(struct drm_crtc *crtc)
 				if (pll != ATOM_PPLL_INVALID)
 					return pll;
 			}
-		} else {
+		} else if (!ASIC_IS_DCE41(rdev)) { /* Don't share PLLs on DCE4.1 chips */
 			/* use the same PPLL for all monitors with the same clock */
 			pll = radeon_get_shared_nondp_ppll(crtc);
 			if (pll != ATOM_PPLL_INVALID)
@@ -1910,6 +1910,21 @@ static void atombios_crtc_disable(struct drm_crtc *crtc)
 	int i;
 
 	atombios_crtc_dpms(crtc, DRM_MODE_DPMS_OFF);
+	if (crtc->fb) {
+		int r;
+		struct radeon_framebuffer *radeon_fb;
+		struct radeon_bo *rbo;
+
+		radeon_fb = to_radeon_framebuffer(crtc->fb);
+		rbo = gem_to_radeon_bo(radeon_fb->obj);
+		r = radeon_bo_reserve(rbo, false);
+		if (unlikely(r))
+			DRM_ERROR("failed to reserve rbo before unpin\n");
+		else {
+			radeon_bo_unpin(rbo);
+			radeon_bo_unreserve(rbo);
+		}
+	}
 	/* disable the GRPH */
 	if (ASIC_IS_DCE4(rdev))
 		WREG32(EVERGREEN_GRPH_ENABLE + radeon_crtc->crtc_offset, 0);
@@ -1940,7 +1955,9 @@ static void atombios_crtc_disable(struct drm_crtc *crtc)
 		break;
 	case ATOM_PPLL0:
 		/* disable the ppll */
-		if ((rdev->family == CHIP_ARUBA) || (rdev->family == CHIP_BONAIRE))
+		if ((rdev->family == CHIP_ARUBA) ||
+		    (rdev->family == CHIP_BONAIRE) ||
+		    (rdev->family == CHIP_HAWAII))
 			atombios_crtc_program_pll(crtc, radeon_crtc->crtc_id, radeon_crtc->pll_id,
 						  0, 0, ATOM_DISABLE, 0, 0, 0, 0, 0, false, &ss);
 		break;
diff --git a/drivers/gpu/drm/radeon/atombios_dp.c b/drivers/gpu/drm/radeon/atombios_dp.c
index 00885417ffff..fb3ae07a1469 100644
--- a/drivers/gpu/drm/radeon/atombios_dp.c
+++ b/drivers/gpu/drm/radeon/atombios_dp.c
@@ -690,8 +690,7 @@ static int radeon_dp_link_train_init(struct radeon_dp_link_train_info *dp_info)
 
 	/* set the lane count on the sink */
 	tmp = dp_info->dp_lane_count;
-	if (dp_info->dpcd[DP_DPCD_REV] >= 0x11 &&
-	    dp_info->dpcd[DP_MAX_LANE_COUNT] & DP_ENHANCED_FRAME_CAP)
+	if (drm_dp_enhanced_frame_cap(dp_info->dpcd))
 		tmp |= DP_LANE_COUNT_ENHANCED_FRAME_EN;
 	radeon_write_dpcd_reg(dp_info->radeon_connector, DP_LANE_COUNT_SET, tmp);
 
diff --git a/drivers/gpu/drm/radeon/atombios_encoders.c b/drivers/gpu/drm/radeon/atombios_encoders.c
index 5e891b226acf..a42d61571f49 100644
--- a/drivers/gpu/drm/radeon/atombios_encoders.c
+++ b/drivers/gpu/drm/radeon/atombios_encoders.c
@@ -213,7 +213,7 @@ void radeon_atom_backlight_init(struct radeon_encoder *radeon_encoder,
 	props.type = BACKLIGHT_RAW;
 	snprintf(bl_name, sizeof(bl_name),
 		 "radeon_bl%d", dev->primary->index);
-	bd = backlight_device_register(bl_name, &drm_connector->kdev,
+	bd = backlight_device_register(bl_name, drm_connector->kdev,
 				       pdata, &radeon_atom_backlight_ops, &props);
 	if (IS_ERR(bd)) {
 		DRM_ERROR("Backlight registration failed\n");
@@ -1662,19 +1662,11 @@ radeon_atom_encoder_dpms_dig(struct drm_encoder *encoder, int mode)
 			atombios_dig_encoder_setup(encoder, ATOM_ENCODER_CMD_SETUP, 0);
 			/* enable the transmitter */
 			atombios_dig_transmitter_setup(encoder, ATOM_TRANSMITTER_ACTION_ENABLE, 0, 0);
-			atombios_dig_transmitter_setup(encoder, ATOM_TRANSMITTER_ACTION_ENABLE_OUTPUT, 0, 0);
 		} else {
 			/* setup and enable the encoder and transmitter */
 			atombios_dig_encoder_setup(encoder, ATOM_ENABLE, 0);
 			atombios_dig_transmitter_setup(encoder, ATOM_TRANSMITTER_ACTION_SETUP, 0, 0);
 			atombios_dig_transmitter_setup(encoder, ATOM_TRANSMITTER_ACTION_ENABLE, 0, 0);
-			/* some dce3.x boards have a bug in their transmitter control table.
-			 * ACTION_ENABLE_OUTPUT can probably be dropped since ACTION_ENABLE
-			 * does the same thing and more.
-			 */
-			if ((rdev->family != CHIP_RV710) && (rdev->family != CHIP_RV730) &&
-			    (rdev->family != CHIP_RS780) && (rdev->family != CHIP_RS880))
-				atombios_dig_transmitter_setup(encoder, ATOM_TRANSMITTER_ACTION_ENABLE_OUTPUT, 0, 0);
 		}
 		if (ENCODER_MODE_IS_DP(atombios_get_encoder_mode(encoder)) && connector) {
 			if (connector->connector_type == DRM_MODE_CONNECTOR_eDP) {
@@ -1692,16 +1684,11 @@ radeon_atom_encoder_dpms_dig(struct drm_encoder *encoder, int mode)
 	case DRM_MODE_DPMS_STANDBY:
 	case DRM_MODE_DPMS_SUSPEND:
 	case DRM_MODE_DPMS_OFF:
-		if (ASIC_IS_DCE41(rdev) || ASIC_IS_DCE5(rdev)) {
-			/* disable the transmitter */
-			atombios_dig_transmitter_setup(encoder, ATOM_TRANSMITTER_ACTION_DISABLE, 0, 0);
-		} else if (ASIC_IS_DCE4(rdev)) {
+		if (ASIC_IS_DCE4(rdev)) {
 			/* disable the transmitter */
-			atombios_dig_transmitter_setup(encoder, ATOM_TRANSMITTER_ACTION_DISABLE_OUTPUT, 0, 0);
 			atombios_dig_transmitter_setup(encoder, ATOM_TRANSMITTER_ACTION_DISABLE, 0, 0);
 		} else {
 			/* disable the encoder and transmitter */
-			atombios_dig_transmitter_setup(encoder, ATOM_TRANSMITTER_ACTION_DISABLE_OUTPUT, 0, 0);
 			atombios_dig_transmitter_setup(encoder, ATOM_TRANSMITTER_ACTION_DISABLE, 0, 0);
 			atombios_dig_encoder_setup(encoder, ATOM_DISABLE, 0);
 		}
@@ -2410,6 +2397,15 @@ static void radeon_atom_encoder_prepare(struct drm_encoder *encoder)
 
 	/* this is needed for the pll/ss setup to work correctly in some cases */
 	atombios_set_encoder_crtc_source(encoder);
+	/* set up the FMT blocks */
+	if (ASIC_IS_DCE8(rdev))
+		dce8_program_fmt(encoder);
+	else if (ASIC_IS_DCE4(rdev))
+		dce4_program_fmt(encoder);
+	else if (ASIC_IS_DCE3(rdev))
+		dce3_program_fmt(encoder);
+	else if (ASIC_IS_AVIVO(rdev))
+		avivo_program_fmt(encoder);
 }
 
 static void radeon_atom_encoder_commit(struct drm_encoder *encoder)
diff --git a/drivers/gpu/drm/radeon/ci_dpm.c b/drivers/gpu/drm/radeon/ci_dpm.c
index 51e947a97edf..1ed479976358 100644
--- a/drivers/gpu/drm/radeon/ci_dpm.c
+++ b/drivers/gpu/drm/radeon/ci_dpm.c
@@ -40,6 +40,20 @@
 #define VOLTAGE_VID_OFFSET_SCALE1    625
 #define VOLTAGE_VID_OFFSET_SCALE2    100
 
+static const struct ci_pt_defaults defaults_hawaii_xt =
+{
+	1, 0xF, 0xFD, 0x19, 5, 0x14, 0, 0xB0000,
+	{ 0x84,  0x0,   0x0,   0x7F,  0x0,   0x0,   0x5A,  0x60,  0x51,  0x8E,  0x79,  0x6B,  0x5F,  0x90,  0x79  },
+	{ 0x1EA, 0x1EA, 0x1EA, 0x224, 0x224, 0x224, 0x24F, 0x24F, 0x24F, 0x28E, 0x28E, 0x28E, 0x2BC, 0x2BC, 0x2BC }
+};
+
+static const struct ci_pt_defaults defaults_hawaii_pro =
+{
+	1, 0xF, 0xFD, 0x19, 5, 0x14, 0, 0x65062,
+	{ 0x93,  0x0,   0x0,   0x97,  0x0,   0x0,   0x6B,  0x60,  0x51,  0x95,  0x79,  0x6B,  0x5F,  0x90,  0x79  },
+	{ 0x1EA, 0x1EA, 0x1EA, 0x224, 0x224, 0x224, 0x24F, 0x24F, 0x24F, 0x28E, 0x28E, 0x28E, 0x2BC, 0x2BC, 0x2BC }
+};
+
 static const struct ci_pt_defaults defaults_bonaire_xt =
 {
 	1, 0xF, 0xFD, 0x19, 5, 45, 0, 0xB0000,
@@ -187,22 +201,38 @@ static void ci_initialize_powertune_defaults(struct radeon_device *rdev)
 	struct ci_power_info *pi = ci_get_pi(rdev);
 
 	switch (rdev->pdev->device) {
-        case 0x6650:
-        case 0x6658:
-        case 0x665C:
-        default:
+	case 0x6650:
+	case 0x6658:
+	case 0x665C:
+	default:
 		pi->powertune_defaults = &defaults_bonaire_xt;
 		break;
-        case 0x6651:
-        case 0x665D:
+	case 0x6651:
+	case 0x665D:
 		pi->powertune_defaults = &defaults_bonaire_pro;
 		break;
-        case 0x6640:
+	case 0x6640:
 		pi->powertune_defaults = &defaults_saturn_xt;
 		break;
-        case 0x6641:
+	case 0x6641:
 		pi->powertune_defaults = &defaults_saturn_pro;
 		break;
+	case 0x67B8:
+	case 0x67B0:
+	case 0x67A0:
+	case 0x67A1:
+	case 0x67A2:
+	case 0x67A8:
+	case 0x67A9:
+	case 0x67AA:
+	case 0x67B9:
+	case 0x67BE:
+		pi->powertune_defaults = &defaults_hawaii_xt;
+		break;
+	case 0x67BA:
+	case 0x67B1:
+		pi->powertune_defaults = &defaults_hawaii_pro;
+		break;
 	}
 
 	pi->dte_tj_offset = 0;
@@ -5142,9 +5172,15 @@ int ci_dpm_init(struct radeon_device *rdev)
 	rdev->pm.dpm.dyn_state.valid_mclk_values.count = 0;
 	rdev->pm.dpm.dyn_state.valid_mclk_values.values = NULL;
 
-	pi->thermal_temp_setting.temperature_low = 99500;
-	pi->thermal_temp_setting.temperature_high = 100000;
-	pi->thermal_temp_setting.temperature_shutdown = 104000;
+	if (rdev->family == CHIP_HAWAII) {
+		pi->thermal_temp_setting.temperature_low = 94500;
+		pi->thermal_temp_setting.temperature_high = 95000;
+		pi->thermal_temp_setting.temperature_shutdown = 104000;
+	} else {
+		pi->thermal_temp_setting.temperature_low = 99500;
+		pi->thermal_temp_setting.temperature_high = 100000;
+		pi->thermal_temp_setting.temperature_shutdown = 104000;
+	}
 
 	pi->uvd_enabled = false;
 
diff --git a/drivers/gpu/drm/radeon/ci_smc.c b/drivers/gpu/drm/radeon/ci_smc.c
index 252e10a41cf5..9c745dd22438 100644
--- a/drivers/gpu/drm/radeon/ci_smc.c
+++ b/drivers/gpu/drm/radeon/ci_smc.c
@@ -217,6 +217,10 @@ int ci_load_smc_ucode(struct radeon_device *rdev, u32 limit)
 		ucode_start_address = BONAIRE_SMC_UCODE_START;
 		ucode_size = BONAIRE_SMC_UCODE_SIZE;
 		break;
+	case CHIP_HAWAII:
+		ucode_start_address = HAWAII_SMC_UCODE_START;
+		ucode_size = HAWAII_SMC_UCODE_SIZE;
+		break;
 	default:
 		DRM_ERROR("unknown asic in smc ucode loader\n");
 		BUG();
diff --git a/drivers/gpu/drm/radeon/cik.c b/drivers/gpu/drm/radeon/cik.c
index 9cd2bc989ac7..ae92aa041c6a 100644
--- a/drivers/gpu/drm/radeon/cik.c
+++ b/drivers/gpu/drm/radeon/cik.c
@@ -41,6 +41,14 @@ MODULE_FIRMWARE("radeon/BONAIRE_mc.bin");
 MODULE_FIRMWARE("radeon/BONAIRE_rlc.bin");
 MODULE_FIRMWARE("radeon/BONAIRE_sdma.bin");
 MODULE_FIRMWARE("radeon/BONAIRE_smc.bin");
+MODULE_FIRMWARE("radeon/HAWAII_pfp.bin");
+MODULE_FIRMWARE("radeon/HAWAII_me.bin");
+MODULE_FIRMWARE("radeon/HAWAII_ce.bin");
+MODULE_FIRMWARE("radeon/HAWAII_mec.bin");
+MODULE_FIRMWARE("radeon/HAWAII_mc.bin");
+MODULE_FIRMWARE("radeon/HAWAII_rlc.bin");
+MODULE_FIRMWARE("radeon/HAWAII_sdma.bin");
+MODULE_FIRMWARE("radeon/HAWAII_smc.bin");
 MODULE_FIRMWARE("radeon/KAVERI_pfp.bin");
 MODULE_FIRMWARE("radeon/KAVERI_me.bin");
 MODULE_FIRMWARE("radeon/KAVERI_ce.bin");
@@ -67,11 +75,6 @@ extern void si_init_uvd_internal_cg(struct radeon_device *rdev);
 extern int cik_sdma_resume(struct radeon_device *rdev);
 extern void cik_sdma_enable(struct radeon_device *rdev, bool enable);
 extern void cik_sdma_fini(struct radeon_device *rdev);
-extern void cik_sdma_vm_set_page(struct radeon_device *rdev,
-				 struct radeon_ib *ib,
-				 uint64_t pe,
-				 uint64_t addr, unsigned count,
-				 uint32_t incr, uint32_t flags);
 static void cik_rlc_stop(struct radeon_device *rdev);
 static void cik_pcie_gen3_enable(struct radeon_device *rdev);
 static void cik_program_aspm(struct radeon_device *rdev);
@@ -1302,6 +1305,171 @@ static const u32 kalindi_mgcg_cgcg_init[] =
 	0xd80c, 0xff000ff0, 0x00000100
 };
 
+static const u32 hawaii_golden_spm_registers[] =
+{
+	0x30800, 0xe0ffffff, 0xe0000000
+};
+
+static const u32 hawaii_golden_common_registers[] =
+{
+	0x30800, 0xffffffff, 0xe0000000,
+	0x28350, 0xffffffff, 0x3a00161a,
+	0x28354, 0xffffffff, 0x0000002e,
+	0x9a10, 0xffffffff, 0x00018208,
+	0x98f8, 0xffffffff, 0x12011003
+};
+
+static const u32 hawaii_golden_registers[] =
+{
+	0x3354, 0x00000333, 0x00000333,
+	0x9a10, 0x00010000, 0x00058208,
+	0x9830, 0xffffffff, 0x00000000,
+	0x9834, 0xf00fffff, 0x00000400,
+	0x9838, 0x0002021c, 0x00020200,
+	0xc78, 0x00000080, 0x00000000,
+	0x5bb0, 0x000000f0, 0x00000070,
+	0x5bc0, 0xf0311fff, 0x80300000,
+	0x350c, 0x00810000, 0x408af000,
+	0x7030, 0x31000111, 0x00000011,
+	0x2f48, 0x73773777, 0x12010001,
+	0x2120, 0x0000007f, 0x0000001b,
+	0x21dc, 0x00007fb6, 0x00002191,
+	0x3628, 0x0000003f, 0x0000000a,
+	0x362c, 0x0000003f, 0x0000000a,
+	0x2ae4, 0x00073ffe, 0x000022a2,
+	0x240c, 0x000007ff, 0x00000000,
+	0x8bf0, 0x00002001, 0x00000001,
+	0x8b24, 0xffffffff, 0x00ffffff,
+	0x30a04, 0x0000ff0f, 0x00000000,
+	0x28a4c, 0x07ffffff, 0x06000000,
+	0x3e78, 0x00000001, 0x00000002,
+	0xc768, 0x00000008, 0x00000008,
+	0xc770, 0x00000f00, 0x00000800,
+	0xc774, 0x00000f00, 0x00000800,
+	0xc798, 0x00ffffff, 0x00ff7fbf,
+	0xc79c, 0x00ffffff, 0x00ff7faf,
+	0x8c00, 0x000000ff, 0x00000800,
+	0xe40, 0x00001fff, 0x00001fff,
+	0x9060, 0x0000007f, 0x00000020,
+	0x9508, 0x00010000, 0x00010000,
+	0xae00, 0x00100000, 0x000ff07c,
+	0xac14, 0x000003ff, 0x0000000f,
+	0xac10, 0xffffffff, 0x7564fdec,
+	0xac0c, 0xffffffff, 0x3120b9a8,
+	0xac08, 0x20000000, 0x0f9c0000
+};
+
+static const u32 hawaii_mgcg_cgcg_init[] =
+{
+	0xc420, 0xffffffff, 0xfffffffd,
+	0x30800, 0xffffffff, 0xe0000000,
+	0x3c2a0, 0xffffffff, 0x00000100,
+	0x3c208, 0xffffffff, 0x00000100,
+	0x3c2c0, 0xffffffff, 0x00000100,
+	0x3c2c8, 0xffffffff, 0x00000100,
+	0x3c2c4, 0xffffffff, 0x00000100,
+	0x55e4, 0xffffffff, 0x00200100,
+	0x3c280, 0xffffffff, 0x00000100,
+	0x3c214, 0xffffffff, 0x06000100,
+	0x3c220, 0xffffffff, 0x00000100,
+	0x3c218, 0xffffffff, 0x06000100,
+	0x3c204, 0xffffffff, 0x00000100,
+	0x3c2e0, 0xffffffff, 0x00000100,
+	0x3c224, 0xffffffff, 0x00000100,
+	0x3c200, 0xffffffff, 0x00000100,
+	0x3c230, 0xffffffff, 0x00000100,
+	0x3c234, 0xffffffff, 0x00000100,
+	0x3c250, 0xffffffff, 0x00000100,
+	0x3c254, 0xffffffff, 0x00000100,
+	0x3c258, 0xffffffff, 0x00000100,
+	0x3c25c, 0xffffffff, 0x00000100,
+	0x3c260, 0xffffffff, 0x00000100,
+	0x3c27c, 0xffffffff, 0x00000100,
+	0x3c278, 0xffffffff, 0x00000100,
+	0x3c210, 0xffffffff, 0x06000100,
+	0x3c290, 0xffffffff, 0x00000100,
+	0x3c274, 0xffffffff, 0x00000100,
+	0x3c2b4, 0xffffffff, 0x00000100,
+	0x3c2b0, 0xffffffff, 0x00000100,
+	0x3c270, 0xffffffff, 0x00000100,
+	0x30800, 0xffffffff, 0xe0000000,
+	0x3c020, 0xffffffff, 0x00010000,
+	0x3c024, 0xffffffff, 0x00030002,
+	0x3c028, 0xffffffff, 0x00040007,
+	0x3c02c, 0xffffffff, 0x00060005,
+	0x3c030, 0xffffffff, 0x00090008,
+	0x3c034, 0xffffffff, 0x00010000,
+	0x3c038, 0xffffffff, 0x00030002,
+	0x3c03c, 0xffffffff, 0x00040007,
+	0x3c040, 0xffffffff, 0x00060005,
+	0x3c044, 0xffffffff, 0x00090008,
+	0x3c048, 0xffffffff, 0x00010000,
+	0x3c04c, 0xffffffff, 0x00030002,
+	0x3c050, 0xffffffff, 0x00040007,
+	0x3c054, 0xffffffff, 0x00060005,
+	0x3c058, 0xffffffff, 0x00090008,
+	0x3c05c, 0xffffffff, 0x00010000,
+	0x3c060, 0xffffffff, 0x00030002,
+	0x3c064, 0xffffffff, 0x00040007,
+	0x3c068, 0xffffffff, 0x00060005,
+	0x3c06c, 0xffffffff, 0x00090008,
+	0x3c070, 0xffffffff, 0x00010000,
+	0x3c074, 0xffffffff, 0x00030002,
+	0x3c078, 0xffffffff, 0x00040007,
+	0x3c07c, 0xffffffff, 0x00060005,
+	0x3c080, 0xffffffff, 0x00090008,
+	0x3c084, 0xffffffff, 0x00010000,
+	0x3c088, 0xffffffff, 0x00030002,
+	0x3c08c, 0xffffffff, 0x00040007,
+	0x3c090, 0xffffffff, 0x00060005,
+	0x3c094, 0xffffffff, 0x00090008,
+	0x3c098, 0xffffffff, 0x00010000,
+	0x3c09c, 0xffffffff, 0x00030002,
+	0x3c0a0, 0xffffffff, 0x00040007,
+	0x3c0a4, 0xffffffff, 0x00060005,
+	0x3c0a8, 0xffffffff, 0x00090008,
+	0x3c0ac, 0xffffffff, 0x00010000,
+	0x3c0b0, 0xffffffff, 0x00030002,
+	0x3c0b4, 0xffffffff, 0x00040007,
+	0x3c0b8, 0xffffffff, 0x00060005,
+	0x3c0bc, 0xffffffff, 0x00090008,
+	0x3c0c0, 0xffffffff, 0x00010000,
+	0x3c0c4, 0xffffffff, 0x00030002,
+	0x3c0c8, 0xffffffff, 0x00040007,
+	0x3c0cc, 0xffffffff, 0x00060005,
+	0x3c0d0, 0xffffffff, 0x00090008,
+	0x3c0d4, 0xffffffff, 0x00010000,
+	0x3c0d8, 0xffffffff, 0x00030002,
+	0x3c0dc, 0xffffffff, 0x00040007,
+	0x3c0e0, 0xffffffff, 0x00060005,
+	0x3c0e4, 0xffffffff, 0x00090008,
+	0x3c0e8, 0xffffffff, 0x00010000,
+	0x3c0ec, 0xffffffff, 0x00030002,
+	0x3c0f0, 0xffffffff, 0x00040007,
+	0x3c0f4, 0xffffffff, 0x00060005,
+	0x3c0f8, 0xffffffff, 0x00090008,
+	0xc318, 0xffffffff, 0x00020200,
+	0x3350, 0xffffffff, 0x00000200,
+	0x15c0, 0xffffffff, 0x00000400,
+	0x55e8, 0xffffffff, 0x00000000,
+	0x2f50, 0xffffffff, 0x00000902,
+	0x3c000, 0xffffffff, 0x96940200,
+	0x8708, 0xffffffff, 0x00900100,
+	0xc424, 0xffffffff, 0x0020003f,
+	0x38, 0xffffffff, 0x0140001c,
+	0x3c, 0x000f0000, 0x000f0000,
+	0x220, 0xffffffff, 0xc060000c,
+	0x224, 0xc0000fff, 0x00000100,
+	0xf90, 0xffffffff, 0x00000100,
+	0xf98, 0x00000101, 0x00000000,
+	0x20a8, 0xffffffff, 0x00000104,
+	0x55e4, 0xff000fff, 0x00000100,
+	0x30cc, 0xc0000fff, 0x00000104,
+	0xc1e4, 0x00000001, 0x00000001,
+	0xd00c, 0xff000ff0, 0x00000100,
+	0xd80c, 0xff000ff0, 0x00000100
+};
+
 static void cik_init_golden_registers(struct radeon_device *rdev)
 {
 	switch (rdev->family) {
@@ -1347,6 +1515,20 @@ static void cik_init_golden_registers(struct radeon_device *rdev)
 						 spectre_golden_spm_registers,
 						 (const u32)ARRAY_SIZE(spectre_golden_spm_registers));
 		break;
+	case CHIP_HAWAII:
+		radeon_program_register_sequence(rdev,
+						 hawaii_mgcg_cgcg_init,
+						 (const u32)ARRAY_SIZE(hawaii_mgcg_cgcg_init));
+		radeon_program_register_sequence(rdev,
+						 hawaii_golden_registers,
+						 (const u32)ARRAY_SIZE(hawaii_golden_registers));
+		radeon_program_register_sequence(rdev,
+						 hawaii_golden_common_registers,
+						 (const u32)ARRAY_SIZE(hawaii_golden_common_registers));
+		radeon_program_register_sequence(rdev,
+						 hawaii_golden_spm_registers,
+						 (const u32)ARRAY_SIZE(hawaii_golden_spm_registers));
+		break;
 	default:
 		break;
 	}
@@ -1454,6 +1636,35 @@ static const u32 bonaire_io_mc_regs[BONAIRE_IO_MC_REGS_SIZE][2] =
 	{0x0000009f, 0x00b48000}
 };
 
+#define HAWAII_IO_MC_REGS_SIZE 22
+
+static const u32 hawaii_io_mc_regs[HAWAII_IO_MC_REGS_SIZE][2] =
+{
+	{0x0000007d, 0x40000000},
+	{0x0000007e, 0x40180304},
+	{0x0000007f, 0x0000ff00},
+	{0x00000081, 0x00000000},
+	{0x00000083, 0x00000800},
+	{0x00000086, 0x00000000},
+	{0x00000087, 0x00000100},
+	{0x00000088, 0x00020100},
+	{0x00000089, 0x00000000},
+	{0x0000008b, 0x00040000},
+	{0x0000008c, 0x00000100},
+	{0x0000008e, 0xff010000},
+	{0x00000090, 0xffffefff},
+	{0x00000091, 0xfff3efff},
+	{0x00000092, 0xfff3efbf},
+	{0x00000093, 0xf7ffffff},
+	{0x00000094, 0xffffff7f},
+	{0x00000095, 0x00000fff},
+	{0x00000096, 0x00116fff},
+	{0x00000097, 0x60010000},
+	{0x00000098, 0x10010000},
+	{0x0000009f, 0x00c79000}
+};
+
+
 /**
  * cik_srbm_select - select specific register instances
  *
@@ -1498,11 +1709,17 @@ static int ci_mc_load_microcode(struct radeon_device *rdev)
 
 	switch (rdev->family) {
 	case CHIP_BONAIRE:
-	default:
 		io_mc_regs = (u32 *)&bonaire_io_mc_regs;
 		ucode_size = CIK_MC_UCODE_SIZE;
 		regs_size = BONAIRE_IO_MC_REGS_SIZE;
 		break;
+	case CHIP_HAWAII:
+		io_mc_regs = (u32 *)&hawaii_io_mc_regs;
+		ucode_size = HAWAII_MC_UCODE_SIZE;
+		regs_size = HAWAII_IO_MC_REGS_SIZE;
+		break;
+	default:
+		return -EINVAL;
 	}
 
 	running = RREG32(MC_SEQ_SUP_CNTL) & RUN_MASK;
@@ -1564,8 +1781,8 @@ static int cik_init_microcode(struct radeon_device *rdev)
 {
 	const char *chip_name;
 	size_t pfp_req_size, me_req_size, ce_req_size,
-		mec_req_size, rlc_req_size, mc_req_size,
-		sdma_req_size, smc_req_size;
+		mec_req_size, rlc_req_size, mc_req_size = 0,
+		sdma_req_size, smc_req_size = 0;
 	char fw_name[30];
 	int err;
 
@@ -1583,6 +1800,17 @@ static int cik_init_microcode(struct radeon_device *rdev)
 		sdma_req_size = CIK_SDMA_UCODE_SIZE * 4;
 		smc_req_size = ALIGN(BONAIRE_SMC_UCODE_SIZE, 4);
 		break;
+	case CHIP_HAWAII:
+		chip_name = "HAWAII";
+		pfp_req_size = CIK_PFP_UCODE_SIZE * 4;
+		me_req_size = CIK_ME_UCODE_SIZE * 4;
+		ce_req_size = CIK_CE_UCODE_SIZE * 4;
+		mec_req_size = CIK_MEC_UCODE_SIZE * 4;
+		rlc_req_size = BONAIRE_RLC_UCODE_SIZE * 4;
+		mc_req_size = HAWAII_MC_UCODE_SIZE * 4;
+		sdma_req_size = CIK_SDMA_UCODE_SIZE * 4;
+		smc_req_size = ALIGN(HAWAII_SMC_UCODE_SIZE, 4);
+		break;
 	case CHIP_KAVERI:
 		chip_name = "KAVERI";
 		pfp_req_size = CIK_PFP_UCODE_SIZE * 4;
@@ -1763,9 +1991,227 @@ static void cik_tiling_mode_table_init(struct radeon_device *rdev)
 
 	num_pipe_configs = rdev->config.cik.max_tile_pipes;
 	if (num_pipe_configs > 8)
-		num_pipe_configs = 8; /* ??? */
+		num_pipe_configs = 16;
 
-	if (num_pipe_configs == 8) {
+	if (num_pipe_configs == 16) {
+		for (reg_offset = 0; reg_offset < num_tile_mode_states; reg_offset++) {
+			switch (reg_offset) {
+			case 0:
+				gb_tile_moden = (ARRAY_MODE(ARRAY_2D_TILED_THIN1) |
+						 MICRO_TILE_MODE_NEW(ADDR_SURF_DEPTH_MICRO_TILING) |
+						 PIPE_CONFIG(ADDR_SURF_P16_32x32_16x16) |
+						 TILE_SPLIT(ADDR_SURF_TILE_SPLIT_64B));
+				break;
+			case 1:
+				gb_tile_moden = (ARRAY_MODE(ARRAY_2D_TILED_THIN1) |
+						 MICRO_TILE_MODE_NEW(ADDR_SURF_DEPTH_MICRO_TILING) |
+						 PIPE_CONFIG(ADDR_SURF_P16_32x32_16x16) |
+						 TILE_SPLIT(ADDR_SURF_TILE_SPLIT_128B));
+				break;
+			case 2:
+				gb_tile_moden = (ARRAY_MODE(ARRAY_2D_TILED_THIN1) |
+						 MICRO_TILE_MODE_NEW(ADDR_SURF_DEPTH_MICRO_TILING) |
+						 PIPE_CONFIG(ADDR_SURF_P16_32x32_16x16) |
+						 TILE_SPLIT(ADDR_SURF_TILE_SPLIT_256B));
+				break;
+			case 3:
+				gb_tile_moden = (ARRAY_MODE(ARRAY_2D_TILED_THIN1) |
+						 MICRO_TILE_MODE_NEW(ADDR_SURF_DEPTH_MICRO_TILING) |
+						 PIPE_CONFIG(ADDR_SURF_P16_32x32_16x16) |
+						 TILE_SPLIT(ADDR_SURF_TILE_SPLIT_512B));
+				break;
+			case 4:
+				gb_tile_moden = (ARRAY_MODE(ARRAY_2D_TILED_THIN1) |
+						 MICRO_TILE_MODE_NEW(ADDR_SURF_DEPTH_MICRO_TILING) |
+						 PIPE_CONFIG(ADDR_SURF_P16_32x32_16x16) |
+						 TILE_SPLIT(split_equal_to_row_size));
+				break;
+			case 5:
+				gb_tile_moden = (ARRAY_MODE(ARRAY_1D_TILED_THIN1) |
+						 MICRO_TILE_MODE_NEW(ADDR_SURF_DEPTH_MICRO_TILING));
+				break;
+			case 6:
+				gb_tile_moden = (ARRAY_MODE(ARRAY_PRT_2D_TILED_THIN1) |
+						 MICRO_TILE_MODE_NEW(ADDR_SURF_DEPTH_MICRO_TILING) |
+						 PIPE_CONFIG(ADDR_SURF_P16_32x32_16x16) |
+						 TILE_SPLIT(ADDR_SURF_TILE_SPLIT_256B));
+				break;
+			case 7:
+				gb_tile_moden = (ARRAY_MODE(ARRAY_PRT_2D_TILED_THIN1) |
+						 MICRO_TILE_MODE_NEW(ADDR_SURF_DEPTH_MICRO_TILING) |
+						 PIPE_CONFIG(ADDR_SURF_P16_32x32_16x16) |
+						 TILE_SPLIT(split_equal_to_row_size));
+				break;
+			case 8:
+				gb_tile_moden = (ARRAY_MODE(ARRAY_LINEAR_ALIGNED) |
+						 PIPE_CONFIG(ADDR_SURF_P16_32x32_16x16));
+				break;
+			case 9:
+				gb_tile_moden = (ARRAY_MODE(ARRAY_1D_TILED_THIN1) |
+						 MICRO_TILE_MODE_NEW(ADDR_SURF_DISPLAY_MICRO_TILING));
+				break;
+			case 10:
+				gb_tile_moden = (ARRAY_MODE(ARRAY_2D_TILED_THIN1) |
+						 MICRO_TILE_MODE_NEW(ADDR_SURF_DISPLAY_MICRO_TILING) |
+						 PIPE_CONFIG(ADDR_SURF_P16_32x32_16x16) |
+						 SAMPLE_SPLIT(ADDR_SURF_SAMPLE_SPLIT_2));
+				break;
+			case 11:
+				gb_tile_moden = (ARRAY_MODE(ARRAY_PRT_TILED_THIN1) |
+						 MICRO_TILE_MODE_NEW(ADDR_SURF_DISPLAY_MICRO_TILING) |
+						 PIPE_CONFIG(ADDR_SURF_P16_32x32_8x16) |
+						 SAMPLE_SPLIT(ADDR_SURF_SAMPLE_SPLIT_2));
+				break;
+			case 12:
+				gb_tile_moden = (ARRAY_MODE(ARRAY_PRT_2D_TILED_THIN1) |
+						 MICRO_TILE_MODE_NEW(ADDR_SURF_DISPLAY_MICRO_TILING) |
+						 PIPE_CONFIG(ADDR_SURF_P16_32x32_16x16) |
+						 SAMPLE_SPLIT(ADDR_SURF_SAMPLE_SPLIT_2));
+				break;
+			case 13:
+				gb_tile_moden = (ARRAY_MODE(ARRAY_1D_TILED_THIN1) |
+						 MICRO_TILE_MODE_NEW(ADDR_SURF_THIN_MICRO_TILING));
+				break;
+			case 14:
+				gb_tile_moden = (ARRAY_MODE(ARRAY_2D_TILED_THIN1) |
+						 MICRO_TILE_MODE_NEW(ADDR_SURF_THIN_MICRO_TILING) |
+						 PIPE_CONFIG(ADDR_SURF_P16_32x32_16x16) |
+						 SAMPLE_SPLIT(ADDR_SURF_SAMPLE_SPLIT_2));
+				break;
+			case 16:
+				gb_tile_moden = (ARRAY_MODE(ARRAY_PRT_TILED_THIN1) |
+						 MICRO_TILE_MODE_NEW(ADDR_SURF_THIN_MICRO_TILING) |
+						 PIPE_CONFIG(ADDR_SURF_P16_32x32_8x16) |
+						 SAMPLE_SPLIT(ADDR_SURF_SAMPLE_SPLIT_2));
+				break;
+			case 17:
+				gb_tile_moden = (ARRAY_MODE(ARRAY_PRT_2D_TILED_THIN1) |
+						 MICRO_TILE_MODE_NEW(ADDR_SURF_THIN_MICRO_TILING) |
+						 PIPE_CONFIG(ADDR_SURF_P16_32x32_16x16) |
+						 SAMPLE_SPLIT(ADDR_SURF_SAMPLE_SPLIT_2));
+				break;
+			case 27:
+				gb_tile_moden = (ARRAY_MODE(ARRAY_1D_TILED_THIN1) |
+						 MICRO_TILE_MODE_NEW(ADDR_SURF_ROTATED_MICRO_TILING));
+				break;
+			case 28:
+				gb_tile_moden = (ARRAY_MODE(ARRAY_2D_TILED_THIN1) |
+						 MICRO_TILE_MODE_NEW(ADDR_SURF_ROTATED_MICRO_TILING) |
+						 PIPE_CONFIG(ADDR_SURF_P16_32x32_16x16) |
+						 SAMPLE_SPLIT(ADDR_SURF_SAMPLE_SPLIT_2));
+				break;
+			case 29:
+				gb_tile_moden = (ARRAY_MODE(ARRAY_PRT_TILED_THIN1) |
+						 MICRO_TILE_MODE_NEW(ADDR_SURF_ROTATED_MICRO_TILING) |
+						 PIPE_CONFIG(ADDR_SURF_P16_32x32_8x16) |
+						 SAMPLE_SPLIT(ADDR_SURF_SAMPLE_SPLIT_2));
+				break;
+			case 30:
+				gb_tile_moden = (ARRAY_MODE(ARRAY_PRT_2D_TILED_THIN1) |
+						 MICRO_TILE_MODE_NEW(ADDR_SURF_ROTATED_MICRO_TILING) |
+						 PIPE_CONFIG(ADDR_SURF_P16_32x32_16x16) |
+						 SAMPLE_SPLIT(ADDR_SURF_SAMPLE_SPLIT_2));
+				break;
+			default:
+				gb_tile_moden = 0;
+				break;
+			}
+			rdev->config.cik.tile_mode_array[reg_offset] = gb_tile_moden;
+			WREG32(GB_TILE_MODE0 + (reg_offset * 4), gb_tile_moden);
+		}
+		for (reg_offset = 0; reg_offset < num_secondary_tile_mode_states; reg_offset++) {
+			switch (reg_offset) {
+			case 0:
+				gb_tile_moden = (BANK_WIDTH(ADDR_SURF_BANK_WIDTH_1) |
+						 BANK_HEIGHT(ADDR_SURF_BANK_HEIGHT_4) |
+						 MACRO_TILE_ASPECT(ADDR_SURF_MACRO_ASPECT_2) |
+						 NUM_BANKS(ADDR_SURF_16_BANK));
+				break;
+			case 1:
+				gb_tile_moden = (BANK_WIDTH(ADDR_SURF_BANK_WIDTH_1) |
+						 BANK_HEIGHT(ADDR_SURF_BANK_HEIGHT_2) |
+						 MACRO_TILE_ASPECT(ADDR_SURF_MACRO_ASPECT_2) |
+						 NUM_BANKS(ADDR_SURF_16_BANK));
+				break;
+			case 2:
+				gb_tile_moden = (BANK_WIDTH(ADDR_SURF_BANK_WIDTH_1) |
+						 BANK_HEIGHT(ADDR_SURF_BANK_HEIGHT_1) |
+						 MACRO_TILE_ASPECT(ADDR_SURF_MACRO_ASPECT_1) |
+						 NUM_BANKS(ADDR_SURF_16_BANK));
+				break;
+			case 3:
+				gb_tile_moden = (BANK_WIDTH(ADDR_SURF_BANK_WIDTH_1) |
+						 BANK_HEIGHT(ADDR_SURF_BANK_HEIGHT_1) |
+						 MACRO_TILE_ASPECT(ADDR_SURF_MACRO_ASPECT_1) |
+						 NUM_BANKS(ADDR_SURF_16_BANK));
+				break;
+			case 4:
+				gb_tile_moden = (BANK_WIDTH(ADDR_SURF_BANK_WIDTH_1) |
+						 BANK_HEIGHT(ADDR_SURF_BANK_HEIGHT_1) |
+						 MACRO_TILE_ASPECT(ADDR_SURF_MACRO_ASPECT_1) |
+						 NUM_BANKS(ADDR_SURF_8_BANK));
+				break;
+			case 5:
+				gb_tile_moden = (BANK_WIDTH(ADDR_SURF_BANK_WIDTH_1) |
+						 BANK_HEIGHT(ADDR_SURF_BANK_HEIGHT_1) |
+						 MACRO_TILE_ASPECT(ADDR_SURF_MACRO_ASPECT_1) |
+						 NUM_BANKS(ADDR_SURF_4_BANK));
+				break;
+			case 6:
+				gb_tile_moden = (BANK_WIDTH(ADDR_SURF_BANK_WIDTH_1) |
+						 BANK_HEIGHT(ADDR_SURF_BANK_HEIGHT_1) |
+						 MACRO_TILE_ASPECT(ADDR_SURF_MACRO_ASPECT_1) |
+						 NUM_BANKS(ADDR_SURF_2_BANK));
+				break;
+			case 8:
+				gb_tile_moden = (BANK_WIDTH(ADDR_SURF_BANK_WIDTH_1) |
+						 BANK_HEIGHT(ADDR_SURF_BANK_HEIGHT_4) |
+						 MACRO_TILE_ASPECT(ADDR_SURF_MACRO_ASPECT_2) |
+						 NUM_BANKS(ADDR_SURF_16_BANK));
+				break;
+			case 9:
+				gb_tile_moden = (BANK_WIDTH(ADDR_SURF_BANK_WIDTH_1) |
+						 BANK_HEIGHT(ADDR_SURF_BANK_HEIGHT_2) |
+						 MACRO_TILE_ASPECT(ADDR_SURF_MACRO_ASPECT_2) |
+						 NUM_BANKS(ADDR_SURF_16_BANK));
+				break;
+			case 10:
+				gb_tile_moden = (BANK_WIDTH(ADDR_SURF_BANK_WIDTH_1) |
+						 BANK_HEIGHT(ADDR_SURF_BANK_HEIGHT_1) |
+						 MACRO_TILE_ASPECT(ADDR_SURF_MACRO_ASPECT_1) |
+						 NUM_BANKS(ADDR_SURF_16_BANK));
+				break;
+			case 11:
+				gb_tile_moden = (BANK_WIDTH(ADDR_SURF_BANK_WIDTH_1) |
+						 BANK_HEIGHT(ADDR_SURF_BANK_HEIGHT_1) |
+						 MACRO_TILE_ASPECT(ADDR_SURF_MACRO_ASPECT_1) |
+						 NUM_BANKS(ADDR_SURF_8_BANK));
+				break;
+			case 12:
+				gb_tile_moden = (BANK_WIDTH(ADDR_SURF_BANK_WIDTH_1) |
+						 BANK_HEIGHT(ADDR_SURF_BANK_HEIGHT_1) |
+						 MACRO_TILE_ASPECT(ADDR_SURF_MACRO_ASPECT_1) |
+						 NUM_BANKS(ADDR_SURF_4_BANK));
+				break;
+			case 13:
+				gb_tile_moden = (BANK_WIDTH(ADDR_SURF_BANK_WIDTH_1) |
+						 BANK_HEIGHT(ADDR_SURF_BANK_HEIGHT_1) |
+						 MACRO_TILE_ASPECT(ADDR_SURF_MACRO_ASPECT_1) |
+						 NUM_BANKS(ADDR_SURF_2_BANK));
+				break;
+			case 14:
+				gb_tile_moden = (BANK_WIDTH(ADDR_SURF_BANK_WIDTH_1) |
+						 BANK_HEIGHT(ADDR_SURF_BANK_HEIGHT_1) |
+						 MACRO_TILE_ASPECT(ADDR_SURF_MACRO_ASPECT_1) |
+						 NUM_BANKS(ADDR_SURF_2_BANK));
+				break;
+			default:
+				gb_tile_moden = 0;
+				break;
+			}
+			WREG32(GB_MACROTILE_MODE0 + (reg_offset * 4), gb_tile_moden);
+		}
+	} else if (num_pipe_configs == 8) {
 		for (reg_offset = 0; reg_offset < num_tile_mode_states; reg_offset++) {
 			switch (reg_offset) {
 			case 0:
@@ -2650,7 +3096,10 @@ static void cik_setup_rb(struct radeon_device *rdev,
 		for (j = 0; j < sh_per_se; j++) {
 			cik_select_se_sh(rdev, i, j);
 			data = cik_get_rb_disabled(rdev, max_rb_num, se_num, sh_per_se);
-			disabled_rbs |= data << ((i * sh_per_se + j) * CIK_RB_BITMAP_WIDTH_PER_SH);
+			if (rdev->family == CHIP_HAWAII)
+				disabled_rbs |= data << ((i * sh_per_se + j) * HAWAII_RB_BITMAP_WIDTH_PER_SH);
+			else
+				disabled_rbs |= data << ((i * sh_per_se + j) * CIK_RB_BITMAP_WIDTH_PER_SH);
 		}
 	}
 	cik_select_se_sh(rdev, 0xffffffff, 0xffffffff);
@@ -2667,6 +3116,12 @@ static void cik_setup_rb(struct radeon_device *rdev,
 		data = 0;
 		for (j = 0; j < sh_per_se; j++) {
 			switch (enabled_rbs & 3) {
+			case 0:
+				if (j == 0)
+					data |= PKR_MAP(RASTER_CONFIG_RB_MAP_3);
+				else
+					data |= PKR_MAP(RASTER_CONFIG_RB_MAP_0);
+				break;
 			case 1:
 				data |= (RASTER_CONFIG_RB_MAP_0 << (i * sh_per_se + j) * 2);
 				break;
@@ -2719,6 +3174,23 @@ static void cik_gpu_init(struct radeon_device *rdev)
 		rdev->config.cik.sc_earlyz_tile_fifo_size = 0x130;
 		gb_addr_config = BONAIRE_GB_ADDR_CONFIG_GOLDEN;
 		break;
+	case CHIP_HAWAII:
+		rdev->config.cik.max_shader_engines = 4;
+		rdev->config.cik.max_tile_pipes = 16;
+		rdev->config.cik.max_cu_per_sh = 11;
+		rdev->config.cik.max_sh_per_se = 1;
+		rdev->config.cik.max_backends_per_se = 4;
+		rdev->config.cik.max_texture_channel_caches = 16;
+		rdev->config.cik.max_gprs = 256;
+		rdev->config.cik.max_gs_threads = 32;
+		rdev->config.cik.max_hw_contexts = 8;
+
+		rdev->config.cik.sc_prim_fifo_size_frontend = 0x20;
+		rdev->config.cik.sc_prim_fifo_size_backend = 0x100;
+		rdev->config.cik.sc_hiz_tile_fifo_size = 0x30;
+		rdev->config.cik.sc_earlyz_tile_fifo_size = 0x130;
+		gb_addr_config = HAWAII_GB_ADDR_CONFIG_GOLDEN;
+		break;
 	case CHIP_KAVERI:
 		rdev->config.cik.max_shader_engines = 1;
 		rdev->config.cik.max_tile_pipes = 4;
@@ -3097,6 +3569,85 @@ void cik_semaphore_ring_emit(struct radeon_device *rdev,
 	radeon_ring_write(ring, (upper_32_bits(addr) & 0xffff) | sel);
 }
 
+/**
+ * cik_copy_cpdma - copy pages using the CP DMA engine
+ *
+ * @rdev: radeon_device pointer
+ * @src_offset: src GPU address
+ * @dst_offset: dst GPU address
+ * @num_gpu_pages: number of GPU pages to xfer
+ * @fence: radeon fence object
+ *
+ * Copy GPU paging using the CP DMA engine (CIK+).
+ * Used by the radeon ttm implementation to move pages if
+ * registered as the asic copy callback.
+ */
+int cik_copy_cpdma(struct radeon_device *rdev,
+		   uint64_t src_offset, uint64_t dst_offset,
+		   unsigned num_gpu_pages,
+		   struct radeon_fence **fence)
+{
+	struct radeon_semaphore *sem = NULL;
+	int ring_index = rdev->asic->copy.blit_ring_index;
+	struct radeon_ring *ring = &rdev->ring[ring_index];
+	u32 size_in_bytes, cur_size_in_bytes, control;
+	int i, num_loops;
+	int r = 0;
+
+	r = radeon_semaphore_create(rdev, &sem);
+	if (r) {
+		DRM_ERROR("radeon: moving bo (%d).\n", r);
+		return r;
+	}
+
+	size_in_bytes = (num_gpu_pages << RADEON_GPU_PAGE_SHIFT);
+	num_loops = DIV_ROUND_UP(size_in_bytes, 0x1fffff);
+	r = radeon_ring_lock(rdev, ring, num_loops * 7 + 18);
+	if (r) {
+		DRM_ERROR("radeon: moving bo (%d).\n", r);
+		radeon_semaphore_free(rdev, &sem, NULL);
+		return r;
+	}
+
+	if (radeon_fence_need_sync(*fence, ring->idx)) {
+		radeon_semaphore_sync_rings(rdev, sem, (*fence)->ring,
+					    ring->idx);
+		radeon_fence_note_sync(*fence, ring->idx);
+	} else {
+		radeon_semaphore_free(rdev, &sem, NULL);
+	}
+
+	for (i = 0; i < num_loops; i++) {
+		cur_size_in_bytes = size_in_bytes;
+		if (cur_size_in_bytes > 0x1fffff)
+			cur_size_in_bytes = 0x1fffff;
+		size_in_bytes -= cur_size_in_bytes;
+		control = 0;
+		if (size_in_bytes == 0)
+			control |= PACKET3_DMA_DATA_CP_SYNC;
+		radeon_ring_write(ring, PACKET3(PACKET3_DMA_DATA, 5));
+		radeon_ring_write(ring, control);
+		radeon_ring_write(ring, lower_32_bits(src_offset));
+		radeon_ring_write(ring, upper_32_bits(src_offset));
+		radeon_ring_write(ring, lower_32_bits(dst_offset));
+		radeon_ring_write(ring, upper_32_bits(dst_offset));
+		radeon_ring_write(ring, cur_size_in_bytes);
+		src_offset += cur_size_in_bytes;
+		dst_offset += cur_size_in_bytes;
+	}
+
+	r = radeon_fence_emit(rdev, fence, ring->idx);
+	if (r) {
+		radeon_ring_unlock_undo(rdev, ring);
+		return r;
+	}
+
+	radeon_ring_unlock_commit(rdev, ring);
+	radeon_semaphore_free(rdev, &sem, *fence);
+
+	return r;
+}
+
 /*
  * IB stuff
  */
@@ -3403,7 +3954,8 @@ static int cik_cp_gfx_resume(struct radeon_device *rdev)
 	int r;
 
 	WREG32(CP_SEM_WAIT_TIMER, 0x0);
-	WREG32(CP_SEM_INCOMPLETE_TIMER_CNTL, 0x0);
+	if (rdev->family != CHIP_HAWAII)
+		WREG32(CP_SEM_INCOMPLETE_TIMER_CNTL, 0x0);
 
 	/* Set the write pointer delay */
 	WREG32(CP_RB_WPTR_DELAY, 0);
@@ -4740,12 +5292,17 @@ void cik_vm_fini(struct radeon_device *rdev)
 static void cik_vm_decode_fault(struct radeon_device *rdev,
 				u32 status, u32 addr, u32 mc_client)
 {
-	u32 mc_id = (status & MEMORY_CLIENT_ID_MASK) >> MEMORY_CLIENT_ID_SHIFT;
+	u32 mc_id;
 	u32 vmid = (status & FAULT_VMID_MASK) >> FAULT_VMID_SHIFT;
 	u32 protections = (status & PROTECTIONS_MASK) >> PROTECTIONS_SHIFT;
 	char block[5] = { mc_client >> 24, (mc_client >> 16) & 0xff,
 		(mc_client >> 8) & 0xff, mc_client & 0xff, 0 };
 
+	if (rdev->family == CHIP_HAWAII)
+		mc_id = (status & HAWAII_MEMORY_CLIENT_ID_MASK) >> MEMORY_CLIENT_ID_SHIFT;
+	else
+		mc_id = (status & MEMORY_CLIENT_ID_MASK) >> MEMORY_CLIENT_ID_SHIFT;
+
 	printk("VM fault (0x%02x, vmid %d) at page %u, %s from '%s' (0x%08x) (%d)\n",
 	       protections, vmid, addr,
 	       (status & MEMORY_CLIENT_RW_MASK) ? "write" : "read",
@@ -4834,62 +5391,6 @@ void cik_vm_flush(struct radeon_device *rdev, int ridx, struct radeon_vm *vm)
 	}
 }
 
-/**
- * cik_vm_set_page - update the page tables using sDMA
- *
- * @rdev: radeon_device pointer
- * @ib: indirect buffer to fill with commands
- * @pe: addr of the page entry
- * @addr: dst addr to write into pe
- * @count: number of page entries to update
- * @incr: increase next addr by incr bytes
- * @flags: access flags
- *
- * Update the page tables using CP or sDMA (CIK).
- */
-void cik_vm_set_page(struct radeon_device *rdev,
-		     struct radeon_ib *ib,
-		     uint64_t pe,
-		     uint64_t addr, unsigned count,
-		     uint32_t incr, uint32_t flags)
-{
-	uint32_t r600_flags = cayman_vm_page_flags(rdev, flags);
-	uint64_t value;
-	unsigned ndw;
-
-	if (rdev->asic->vm.pt_ring_index == RADEON_RING_TYPE_GFX_INDEX) {
-		/* CP */
-		while (count) {
-			ndw = 2 + count * 2;
-			if (ndw > 0x3FFE)
-				ndw = 0x3FFE;
-
-			ib->ptr[ib->length_dw++] = PACKET3(PACKET3_WRITE_DATA, ndw);
-			ib->ptr[ib->length_dw++] = (WRITE_DATA_ENGINE_SEL(0) |
-						    WRITE_DATA_DST_SEL(1));
-			ib->ptr[ib->length_dw++] = pe;
-			ib->ptr[ib->length_dw++] = upper_32_bits(pe);
-			for (; ndw > 2; ndw -= 2, --count, pe += 8) {
-				if (flags & RADEON_VM_PAGE_SYSTEM) {
-					value = radeon_vm_map_gart(rdev, addr);
-					value &= 0xFFFFFFFFFFFFF000ULL;
-				} else if (flags & RADEON_VM_PAGE_VALID) {
-					value = addr;
-				} else {
-					value = 0;
-				}
-				addr += incr;
-				value |= r600_flags;
-				ib->ptr[ib->length_dw++] = value;
-				ib->ptr[ib->length_dw++] = upper_32_bits(value);
-			}
-		}
-	} else {
-		/* DMA */
-		cik_sdma_vm_set_page(rdev, ib, pe, addr, count, incr, flags);
-	}
-}
-
 /*
  * RLC
  * The RLC is a multi-purpose microengine that handles a
@@ -5058,6 +5559,7 @@ static int cik_rlc_resume(struct radeon_device *rdev)
 
 	switch (rdev->family) {
 	case CHIP_BONAIRE:
+	case CHIP_HAWAII:
 	default:
 		size = BONAIRE_RLC_UCODE_SIZE;
 		break;
@@ -5556,7 +6058,7 @@ void cik_init_cp_pg_table(struct radeon_device *rdev)
 		}
 
 		for (i = 0; i < CP_ME_TABLE_SIZE; i ++) {
-			dst_ptr[bo_offset + i] = be32_to_cpu(fw_data[table_offset + i]);
+			dst_ptr[bo_offset + i] = cpu_to_le32(be32_to_cpu(fw_data[table_offset + i]));
 		}
 		bo_offset += CP_ME_TABLE_SIZE;
 	}
@@ -5778,52 +6280,57 @@ void cik_get_csb_buffer(struct radeon_device *rdev, volatile u32 *buffer)
 	if (buffer == NULL)
 		return;
 
-	buffer[count++] = PACKET3(PACKET3_PREAMBLE_CNTL, 0);
-	buffer[count++] = PACKET3_PREAMBLE_BEGIN_CLEAR_STATE;
+	buffer[count++] = cpu_to_le32(PACKET3(PACKET3_PREAMBLE_CNTL, 0));
+	buffer[count++] = cpu_to_le32(PACKET3_PREAMBLE_BEGIN_CLEAR_STATE);
 
-	buffer[count++] = PACKET3(PACKET3_CONTEXT_CONTROL, 1);
-	buffer[count++] = 0x80000000;
-	buffer[count++] = 0x80000000;
+	buffer[count++] = cpu_to_le32(PACKET3(PACKET3_CONTEXT_CONTROL, 1));
+	buffer[count++] = cpu_to_le32(0x80000000);
+	buffer[count++] = cpu_to_le32(0x80000000);
 
 	for (sect = rdev->rlc.cs_data; sect->section != NULL; ++sect) {
 		for (ext = sect->section; ext->extent != NULL; ++ext) {
 			if (sect->id == SECT_CONTEXT) {
-				buffer[count++] = PACKET3(PACKET3_SET_CONTEXT_REG, ext->reg_count);
-				buffer[count++] = ext->reg_index - 0xa000;
+				buffer[count++] =
+					cpu_to_le32(PACKET3(PACKET3_SET_CONTEXT_REG, ext->reg_count));
+				buffer[count++] = cpu_to_le32(ext->reg_index - 0xa000);
 				for (i = 0; i < ext->reg_count; i++)
-					buffer[count++] = ext->extent[i];
+					buffer[count++] = cpu_to_le32(ext->extent[i]);
 			} else {
 				return;
 			}
 		}
 	}
 
-	buffer[count++] = PACKET3(PACKET3_SET_CONTEXT_REG, 2);
-	buffer[count++] = PA_SC_RASTER_CONFIG - PACKET3_SET_CONTEXT_REG_START;
+	buffer[count++] = cpu_to_le32(PACKET3(PACKET3_SET_CONTEXT_REG, 2));
+	buffer[count++] = cpu_to_le32(PA_SC_RASTER_CONFIG - PACKET3_SET_CONTEXT_REG_START);
 	switch (rdev->family) {
 	case CHIP_BONAIRE:
-		buffer[count++] = 0x16000012;
-		buffer[count++] = 0x00000000;
+		buffer[count++] = cpu_to_le32(0x16000012);
+		buffer[count++] = cpu_to_le32(0x00000000);
 		break;
 	case CHIP_KAVERI:
-		buffer[count++] = 0x00000000; /* XXX */
-		buffer[count++] = 0x00000000;
+		buffer[count++] = cpu_to_le32(0x00000000); /* XXX */
+		buffer[count++] = cpu_to_le32(0x00000000);
 		break;
 	case CHIP_KABINI:
-		buffer[count++] = 0x00000000; /* XXX */
-		buffer[count++] = 0x00000000;
+		buffer[count++] = cpu_to_le32(0x00000000); /* XXX */
+		buffer[count++] = cpu_to_le32(0x00000000);
+		break;
+	case CHIP_HAWAII:
+		buffer[count++] = 0x3a00161a;
+		buffer[count++] = 0x0000002e;
 		break;
 	default:
-		buffer[count++] = 0x00000000;
-		buffer[count++] = 0x00000000;
+		buffer[count++] = cpu_to_le32(0x00000000);
+		buffer[count++] = cpu_to_le32(0x00000000);
 		break;
 	}
 
-	buffer[count++] = PACKET3(PACKET3_PREAMBLE_CNTL, 0);
-	buffer[count++] = PACKET3_PREAMBLE_END_CLEAR_STATE;
+	buffer[count++] = cpu_to_le32(PACKET3(PACKET3_PREAMBLE_CNTL, 0));
+	buffer[count++] = cpu_to_le32(PACKET3_PREAMBLE_END_CLEAR_STATE);
 
-	buffer[count++] = PACKET3(PACKET3_CLEAR_STATE, 0);
-	buffer[count++] = 0;
+	buffer[count++] = cpu_to_le32(PACKET3(PACKET3_CLEAR_STATE, 0));
+	buffer[count++] = cpu_to_le32(0);
 }
 
 static void cik_init_pg(struct radeon_device *rdev)
@@ -7118,7 +7625,7 @@ static int cik_startup(struct radeon_device *rdev)
 	ring = &rdev->ring[RADEON_RING_TYPE_GFX_INDEX];
 	r = radeon_ring_init(rdev, ring, ring->ring_size, RADEON_WB_CP_RPTR_OFFSET,
 			     CP_RB0_RPTR, CP_RB0_WPTR,
-			     RADEON_CP_PACKET2);
+			     PACKET3(PACKET3_NOP, 0x3FFF));
 	if (r)
 		return r;
 
@@ -7428,6 +7935,70 @@ void cik_fini(struct radeon_device *rdev)
 	rdev->bios = NULL;
 }
 
+void dce8_program_fmt(struct drm_encoder *encoder)
+{
+	struct drm_device *dev = encoder->dev;
+	struct radeon_device *rdev = dev->dev_private;
+	struct radeon_encoder *radeon_encoder = to_radeon_encoder(encoder);
+	struct radeon_crtc *radeon_crtc = to_radeon_crtc(encoder->crtc);
+	struct drm_connector *connector = radeon_get_connector_for_encoder(encoder);
+	int bpc = 0;
+	u32 tmp = 0;
+	enum radeon_connector_dither dither = RADEON_FMT_DITHER_DISABLE;
+
+	if (connector) {
+		struct radeon_connector *radeon_connector = to_radeon_connector(connector);
+		bpc = radeon_get_monitor_bpc(connector);
+		dither = radeon_connector->dither;
+	}
+
+	/* LVDS/eDP FMT is set up by atom */
+	if (radeon_encoder->devices & ATOM_DEVICE_LCD_SUPPORT)
+		return;
+
+	/* not needed for analog */
+	if ((radeon_encoder->encoder_id == ENCODER_OBJECT_ID_INTERNAL_KLDSCP_DAC1) ||
+	    (radeon_encoder->encoder_id == ENCODER_OBJECT_ID_INTERNAL_KLDSCP_DAC2))
+		return;
+
+	if (bpc == 0)
+		return;
+
+	switch (bpc) {
+	case 6:
+		if (dither == RADEON_FMT_DITHER_ENABLE)
+			/* XXX sort out optimal dither settings */
+			tmp |= (FMT_FRAME_RANDOM_ENABLE | FMT_HIGHPASS_RANDOM_ENABLE |
+				FMT_SPATIAL_DITHER_EN | FMT_SPATIAL_DITHER_DEPTH(0));
+		else
+			tmp |= (FMT_TRUNCATE_EN | FMT_TRUNCATE_DEPTH(0));
+		break;
+	case 8:
+		if (dither == RADEON_FMT_DITHER_ENABLE)
+			/* XXX sort out optimal dither settings */
+			tmp |= (FMT_FRAME_RANDOM_ENABLE | FMT_HIGHPASS_RANDOM_ENABLE |
+				FMT_RGB_RANDOM_ENABLE |
+				FMT_SPATIAL_DITHER_EN | FMT_SPATIAL_DITHER_DEPTH(1));
+		else
+			tmp |= (FMT_TRUNCATE_EN | FMT_TRUNCATE_DEPTH(1));
+		break;
+	case 10:
+		if (dither == RADEON_FMT_DITHER_ENABLE)
+			/* XXX sort out optimal dither settings */
+			tmp |= (FMT_FRAME_RANDOM_ENABLE | FMT_HIGHPASS_RANDOM_ENABLE |
+				FMT_RGB_RANDOM_ENABLE |
+				FMT_SPATIAL_DITHER_EN | FMT_SPATIAL_DITHER_DEPTH(2));
+		else
+			tmp |= (FMT_TRUNCATE_EN | FMT_TRUNCATE_DEPTH(2));
+		break;
+	default:
+		/* not needed */
+		break;
+	}
+
+	WREG32(FMT_BIT_DEPTH_CONTROL + radeon_crtc->crtc_offset, tmp);
+}
+
 /* display watermark setup */
 /**
  * dce8_line_buffer_adjust - Set up the line buffer
diff --git a/drivers/gpu/drm/radeon/cik_sdma.c b/drivers/gpu/drm/radeon/cik_sdma.c
index b6286068e111..9c9529de20ee 100644
--- a/drivers/gpu/drm/radeon/cik_sdma.c
+++ b/drivers/gpu/drm/radeon/cik_sdma.c
@@ -25,6 +25,7 @@
 #include <drm/drmP.h>
 #include "radeon.h"
 #include "radeon_asic.h"
+#include "radeon_trace.h"
 #include "cikd.h"
 
 /* sdma */
@@ -101,14 +102,6 @@ void cik_sdma_fence_ring_emit(struct radeon_device *rdev,
 {
 	struct radeon_ring *ring = &rdev->ring[fence->ring];
 	u64 addr = rdev->fence_drv[fence->ring].gpu_addr;
-	u32 extra_bits = (SDMA_POLL_REG_MEM_EXTRA_OP(1) |
-			  SDMA_POLL_REG_MEM_EXTRA_FUNC(3)); /* == */
-	u32 ref_and_mask;
-
-	if (fence->ring == R600_RING_TYPE_DMA_INDEX)
-		ref_and_mask = SDMA0;
-	else
-		ref_and_mask = SDMA1;
 
 	/* write the fence */
 	radeon_ring_write(ring, SDMA_PACKET(SDMA_OPCODE_FENCE, 0, 0));
@@ -118,12 +111,12 @@ void cik_sdma_fence_ring_emit(struct radeon_device *rdev,
 	/* generate an interrupt */
 	radeon_ring_write(ring, SDMA_PACKET(SDMA_OPCODE_TRAP, 0, 0));
 	/* flush HDP */
-	radeon_ring_write(ring, SDMA_PACKET(SDMA_OPCODE_POLL_REG_MEM, 0, extra_bits));
-	radeon_ring_write(ring, GPU_HDP_FLUSH_DONE);
-	radeon_ring_write(ring, GPU_HDP_FLUSH_REQ);
-	radeon_ring_write(ring, ref_and_mask); /* REFERENCE */
-	radeon_ring_write(ring, ref_and_mask); /* MASK */
-	radeon_ring_write(ring, (4 << 16) | 10); /* RETRY_COUNT, POLL_INTERVAL */
+	/* We should be using the new POLL_REG_MEM special op packet here
+	 * but it causes sDMA to hang sometimes
+	 */
+	radeon_ring_write(ring, SDMA_PACKET(SDMA_OPCODE_SRBM_WRITE, 0, 0xf000));
+	radeon_ring_write(ring, HDP_MEM_COHERENCY_FLUSH_CNTL >> 2);
+	radeon_ring_write(ring, 0);
 }
 
 /**
@@ -653,11 +646,12 @@ void cik_sdma_vm_set_page(struct radeon_device *rdev,
 			  uint64_t addr, unsigned count,
 			  uint32_t incr, uint32_t flags)
 {
-	uint32_t r600_flags = cayman_vm_page_flags(rdev, flags);
 	uint64_t value;
 	unsigned ndw;
 
-	if (flags & RADEON_VM_PAGE_SYSTEM) {
+	trace_radeon_vm_set_page(pe, addr, count, incr, flags);
+
+	if (flags & R600_PTE_SYSTEM) {
 		while (count) {
 			ndw = count * 2;
 			if (ndw > 0xFFFFE)
@@ -669,16 +663,10 @@ void cik_sdma_vm_set_page(struct radeon_device *rdev,
 			ib->ptr[ib->length_dw++] = upper_32_bits(pe);
 			ib->ptr[ib->length_dw++] = ndw;
 			for (; ndw > 0; ndw -= 2, --count, pe += 8) {
-				if (flags & RADEON_VM_PAGE_SYSTEM) {
-					value = radeon_vm_map_gart(rdev, addr);
-					value &= 0xFFFFFFFFFFFFF000ULL;
-				} else if (flags & RADEON_VM_PAGE_VALID) {
-					value = addr;
-				} else {
-					value = 0;
-				}
+				value = radeon_vm_map_gart(rdev, addr);
+				value &= 0xFFFFFFFFFFFFF000ULL;
 				addr += incr;
-				value |= r600_flags;
+				value |= flags;
 				ib->ptr[ib->length_dw++] = value;
 				ib->ptr[ib->length_dw++] = upper_32_bits(value);
 			}
@@ -689,7 +677,7 @@ void cik_sdma_vm_set_page(struct radeon_device *rdev,
 			if (ndw > 0x7FFFF)
 				ndw = 0x7FFFF;
 
-			if (flags & RADEON_VM_PAGE_VALID)
+			if (flags & R600_PTE_VALID)
 				value = addr;
 			else
 				value = 0;
@@ -697,7 +685,7 @@ void cik_sdma_vm_set_page(struct radeon_device *rdev,
 			ib->ptr[ib->length_dw++] = SDMA_PACKET(SDMA_OPCODE_GENERATE_PTE_PDE, 0, 0);
 			ib->ptr[ib->length_dw++] = pe; /* dst addr */
 			ib->ptr[ib->length_dw++] = upper_32_bits(pe);
-			ib->ptr[ib->length_dw++] = r600_flags; /* mask */
+			ib->ptr[ib->length_dw++] = flags; /* mask */
 			ib->ptr[ib->length_dw++] = 0;
 			ib->ptr[ib->length_dw++] = value; /* value */
 			ib->ptr[ib->length_dw++] = upper_32_bits(value);
@@ -724,18 +712,10 @@ void cik_sdma_vm_set_page(struct radeon_device *rdev,
 void cik_dma_vm_flush(struct radeon_device *rdev, int ridx, struct radeon_vm *vm)
 {
 	struct radeon_ring *ring = &rdev->ring[ridx];
-	u32 extra_bits = (SDMA_POLL_REG_MEM_EXTRA_OP(1) |
-			  SDMA_POLL_REG_MEM_EXTRA_FUNC(3)); /* == */
-	u32 ref_and_mask;
 
 	if (vm == NULL)
 		return;
 
-	if (ridx == R600_RING_TYPE_DMA_INDEX)
-		ref_and_mask = SDMA0;
-	else
-		ref_and_mask = SDMA1;
-
 	radeon_ring_write(ring, SDMA_PACKET(SDMA_OPCODE_SRBM_WRITE, 0, 0xf000));
 	if (vm->id < 8) {
 		radeon_ring_write(ring, (VM_CONTEXT0_PAGE_TABLE_BASE_ADDR + (vm->id << 2)) >> 2);
@@ -770,12 +750,12 @@ void cik_dma_vm_flush(struct radeon_device *rdev, int ridx, struct radeon_vm *vm
 	radeon_ring_write(ring, VMID(0));
 
 	/* flush HDP */
-	radeon_ring_write(ring, SDMA_PACKET(SDMA_OPCODE_POLL_REG_MEM, 0, extra_bits));
-	radeon_ring_write(ring, GPU_HDP_FLUSH_DONE);
-	radeon_ring_write(ring, GPU_HDP_FLUSH_REQ);
-	radeon_ring_write(ring, ref_and_mask); /* REFERENCE */
-	radeon_ring_write(ring, ref_and_mask); /* MASK */
-	radeon_ring_write(ring, (4 << 16) | 10); /* RETRY_COUNT, POLL_INTERVAL */
+	/* We should be using the new POLL_REG_MEM special op packet here
+	 * but it causes sDMA to hang sometimes
+	 */
+	radeon_ring_write(ring, SDMA_PACKET(SDMA_OPCODE_SRBM_WRITE, 0, 0xf000));
+	radeon_ring_write(ring, HDP_MEM_COHERENCY_FLUSH_CNTL >> 2);
+	radeon_ring_write(ring, 0);
 
 	/* flush TLB */
 	radeon_ring_write(ring, SDMA_PACKET(SDMA_OPCODE_SRBM_WRITE, 0, 0xf000));
diff --git a/drivers/gpu/drm/radeon/cikd.h b/drivers/gpu/drm/radeon/cikd.h
index 203d2a09a1f5..5964af5e5b2d 100644
--- a/drivers/gpu/drm/radeon/cikd.h
+++ b/drivers/gpu/drm/radeon/cikd.h
@@ -25,8 +25,10 @@
 #define CIK_H
 
 #define BONAIRE_GB_ADDR_CONFIG_GOLDEN        0x12010001
+#define HAWAII_GB_ADDR_CONFIG_GOLDEN         0x12011003
 
-#define CIK_RB_BITMAP_WIDTH_PER_SH  2
+#define CIK_RB_BITMAP_WIDTH_PER_SH     2
+#define HAWAII_RB_BITMAP_WIDTH_PER_SH  4
 
 /* DIDT IND registers */
 #define DIDT_SQ_CTRL0                                     0x0
@@ -499,6 +501,7 @@
 		 * bit 4: write
 		 */
 #define		MEMORY_CLIENT_ID_MASK			(0xff << 12)
+#define		HAWAII_MEMORY_CLIENT_ID_MASK		(0x1ff << 12)
 #define		MEMORY_CLIENT_ID_SHIFT			12
 #define		MEMORY_CLIENT_RW_MASK			(1 << 24)
 #define		MEMORY_CLIENT_RW_SHIFT			24
@@ -906,6 +909,39 @@
 #define DPG_PIPE_STUTTER_CONTROL                          0x6cd4
 #       define STUTTER_ENABLE                             (1 << 0)
 
+/* DCE8 FMT blocks */
+#define FMT_DYNAMIC_EXP_CNTL                 0x6fb4
+#       define FMT_DYNAMIC_EXP_EN            (1 << 0)
+#       define FMT_DYNAMIC_EXP_MODE          (1 << 4)
+        /* 0 = 10bit -> 12bit, 1 = 8bit -> 12bit */
+#define FMT_CONTROL                          0x6fb8
+#       define FMT_PIXEL_ENCODING            (1 << 16)
+        /* 0 = RGB 4:4:4 or YCbCr 4:4:4, 1 = YCbCr 4:2:2 */
+#define FMT_BIT_DEPTH_CONTROL                0x6fc8
+#       define FMT_TRUNCATE_EN               (1 << 0)
+#       define FMT_TRUNCATE_MODE             (1 << 1)
+#       define FMT_TRUNCATE_DEPTH(x)         ((x) << 4) /* 0 - 18bpp, 1 - 24bpp, 2 - 30bpp */
+#       define FMT_SPATIAL_DITHER_EN         (1 << 8)
+#       define FMT_SPATIAL_DITHER_MODE(x)    ((x) << 9)
+#       define FMT_SPATIAL_DITHER_DEPTH(x)   ((x) << 11) /* 0 - 18bpp, 1 - 24bpp, 2 - 30bpp */
+#       define FMT_FRAME_RANDOM_ENABLE       (1 << 13)
+#       define FMT_RGB_RANDOM_ENABLE         (1 << 14)
+#       define FMT_HIGHPASS_RANDOM_ENABLE    (1 << 15)
+#       define FMT_TEMPORAL_DITHER_EN        (1 << 16)
+#       define FMT_TEMPORAL_DITHER_DEPTH(x)  ((x) << 17) /* 0 - 18bpp, 1 - 24bpp, 2 - 30bpp */
+#       define FMT_TEMPORAL_DITHER_OFFSET(x) ((x) << 21)
+#       define FMT_TEMPORAL_LEVEL            (1 << 24)
+#       define FMT_TEMPORAL_DITHER_RESET     (1 << 25)
+#       define FMT_25FRC_SEL(x)              ((x) << 26)
+#       define FMT_50FRC_SEL(x)              ((x) << 28)
+#       define FMT_75FRC_SEL(x)              ((x) << 30)
+#define FMT_CLAMP_CONTROL                    0x6fe4
+#       define FMT_CLAMP_DATA_EN             (1 << 0)
+#       define FMT_CLAMP_COLOR_FORMAT(x)     ((x) << 16)
+#       define FMT_CLAMP_6BPC                0
+#       define FMT_CLAMP_8BPC                1
+#       define FMT_CLAMP_10BPC               2
+
 #define	GRBM_CNTL					0x8000
 #define		GRBM_READ_TIMEOUT(x)				((x) << 0)
 
@@ -1129,6 +1165,8 @@
 #              define	ADDR_SURF_P8_32x32_16x16		12
 #              define	ADDR_SURF_P8_32x32_16x32		13
 #              define	ADDR_SURF_P8_32x64_32x32		14
+#              define	ADDR_SURF_P16_32x32_8x16		16
+#              define	ADDR_SURF_P16_32x32_16x16		17
 #       define TILE_SPLIT(x)					((x) << 11)
 #              define	ADDR_SURF_TILE_SPLIT_64B		0
 #              define	ADDR_SURF_TILE_SPLIT_128B		1
@@ -1422,6 +1460,7 @@
 #       define RASTER_CONFIG_RB_MAP_1                   1
 #       define RASTER_CONFIG_RB_MAP_2                   2
 #       define RASTER_CONFIG_RB_MAP_3                   3
+#define		PKR_MAP(x)				((x) << 8)
 
 #define VGT_EVENT_INITIATOR                             0x28a90
 #       define SAMPLE_STREAMOUTSTATS1                   (1 << 0)
@@ -1714,6 +1753,68 @@
 #              define PACKET3_PREAMBLE_BEGIN_CLEAR_STATE     (2 << 28)
 #              define PACKET3_PREAMBLE_END_CLEAR_STATE       (3 << 28)
 #define	PACKET3_DMA_DATA				0x50
+/* 1. header
+ * 2. CONTROL
+ * 3. SRC_ADDR_LO or DATA [31:0]
+ * 4. SRC_ADDR_HI [31:0]
+ * 5. DST_ADDR_LO [31:0]
+ * 6. DST_ADDR_HI [7:0]
+ * 7. COMMAND [30:21] | BYTE_COUNT [20:0]
+ */
+/* CONTROL */
+#              define PACKET3_DMA_DATA_ENGINE(x)     ((x) << 0)
+                /* 0 - ME
+		 * 1 - PFP
+		 */
+#              define PACKET3_DMA_DATA_SRC_CACHE_POLICY(x) ((x) << 13)
+                /* 0 - LRU
+		 * 1 - Stream
+		 * 2 - Bypass
+		 */
+#              define PACKET3_DMA_DATA_SRC_VOLATILE (1 << 15)
+#              define PACKET3_DMA_DATA_DST_SEL(x)  ((x) << 20)
+                /* 0 - DST_ADDR using DAS
+		 * 1 - GDS
+		 * 3 - DST_ADDR using L2
+		 */
+#              define PACKET3_DMA_DATA_DST_CACHE_POLICY(x) ((x) << 25)
+                /* 0 - LRU
+		 * 1 - Stream
+		 * 2 - Bypass
+		 */
+#              define PACKET3_DMA_DATA_DST_VOLATILE (1 << 27)
+#              define PACKET3_DMA_DATA_SRC_SEL(x)  ((x) << 29)
+                /* 0 - SRC_ADDR using SAS
+		 * 1 - GDS
+		 * 2 - DATA
+		 * 3 - SRC_ADDR using L2
+		 */
+#              define PACKET3_DMA_DATA_CP_SYNC     (1 << 31)
+/* COMMAND */
+#              define PACKET3_DMA_DATA_DIS_WC      (1 << 21)
+#              define PACKET3_DMA_DATA_CMD_SRC_SWAP(x) ((x) << 22)
+                /* 0 - none
+		 * 1 - 8 in 16
+		 * 2 - 8 in 32
+		 * 3 - 8 in 64
+		 */
+#              define PACKET3_DMA_DATA_CMD_DST_SWAP(x) ((x) << 24)
+                /* 0 - none
+		 * 1 - 8 in 16
+		 * 2 - 8 in 32
+		 * 3 - 8 in 64
+		 */
+#              define PACKET3_DMA_DATA_CMD_SAS     (1 << 26)
+                /* 0 - memory
+		 * 1 - register
+		 */
+#              define PACKET3_DMA_DATA_CMD_DAS     (1 << 27)
+                /* 0 - memory
+		 * 1 - register
+		 */
+#              define PACKET3_DMA_DATA_CMD_SAIC    (1 << 28)
+#              define PACKET3_DMA_DATA_CMD_DAIC    (1 << 29)
+#              define PACKET3_DMA_DATA_CMD_RAW_WAIT  (1 << 30)
 #define	PACKET3_AQUIRE_MEM				0x58
 #define	PACKET3_REWIND					0x59
 #define	PACKET3_LOAD_UCONFIG_REG			0x5E
diff --git a/drivers/gpu/drm/radeon/dce6_afmt.c b/drivers/gpu/drm/radeon/dce6_afmt.c
index 9fcd338c0fcf..009f46e0ce72 100644
--- a/drivers/gpu/drm/radeon/dce6_afmt.c
+++ b/drivers/gpu/drm/radeon/dce6_afmt.c
@@ -102,6 +102,49 @@ void dce6_afmt_select_pin(struct drm_encoder *encoder)
 	       AFMT_AUDIO_SRC_SELECT(dig->afmt->pin->id));
 }
 
+void dce6_afmt_write_latency_fields(struct drm_encoder *encoder,
+				    struct drm_display_mode *mode)
+{
+	struct radeon_device *rdev = encoder->dev->dev_private;
+	struct radeon_encoder *radeon_encoder = to_radeon_encoder(encoder);
+	struct radeon_encoder_atom_dig *dig = radeon_encoder->enc_priv;
+	struct drm_connector *connector;
+	struct radeon_connector *radeon_connector = NULL;
+	u32 tmp = 0, offset;
+
+	if (!dig->afmt->pin)
+		return;
+
+	offset = dig->afmt->pin->offset;
+
+	list_for_each_entry(connector, &encoder->dev->mode_config.connector_list, head) {
+		if (connector->encoder == encoder) {
+			radeon_connector = to_radeon_connector(connector);
+			break;
+		}
+	}
+
+	if (!radeon_connector) {
+		DRM_ERROR("Couldn't find encoder's connector\n");
+		return;
+	}
+
+	if (mode->flags & DRM_MODE_FLAG_INTERLACE) {
+		if (connector->latency_present[1])
+			tmp = VIDEO_LIPSYNC(connector->video_latency[1]) |
+				AUDIO_LIPSYNC(connector->audio_latency[1]);
+		else
+			tmp = VIDEO_LIPSYNC(255) | AUDIO_LIPSYNC(255);
+	} else {
+		if (connector->latency_present[0])
+			tmp = VIDEO_LIPSYNC(connector->video_latency[0]) |
+				AUDIO_LIPSYNC(connector->audio_latency[0]);
+		else
+			tmp = VIDEO_LIPSYNC(255) | AUDIO_LIPSYNC(255);
+	}
+	WREG32_ENDPOINT(offset, AZ_F0_CODEC_PIN_CONTROL_RESPONSE_LIPSYNC, tmp);
+}
+
 void dce6_afmt_write_speaker_allocation(struct drm_encoder *encoder)
 {
 	struct radeon_device *rdev = encoder->dev->dev_private;
@@ -113,9 +156,6 @@ void dce6_afmt_write_speaker_allocation(struct drm_encoder *encoder)
 	u8 *sadb;
 	int sad_count;
 
-	/* XXX: setting this register causes hangs on some asics */
-	return;
-
 	if (!dig->afmt->pin)
 		return;
 
@@ -201,20 +241,30 @@ void dce6_afmt_write_sad_regs(struct drm_encoder *encoder)
 
 	for (i = 0; i < ARRAY_SIZE(eld_reg_to_type); i++) {
 		u32 value = 0;
+		u8 stereo_freqs = 0;
+		int max_channels = -1;
 		int j;
 
 		for (j = 0; j < sad_count; j++) {
 			struct cea_sad *sad = &sads[j];
 
 			if (sad->format == eld_reg_to_type[i][1]) {
-				value = MAX_CHANNELS(sad->channels) |
-					DESCRIPTOR_BYTE_2(sad->byte2) |
-					SUPPORTED_FREQUENCIES(sad->freq);
+				if (sad->channels > max_channels) {
+					value = MAX_CHANNELS(sad->channels) |
+						DESCRIPTOR_BYTE_2(sad->byte2) |
+						SUPPORTED_FREQUENCIES(sad->freq);
+					max_channels = sad->channels;
+				}
+
 				if (sad->format == HDMI_AUDIO_CODING_TYPE_PCM)
-					value |= SUPPORTED_FREQUENCIES_STEREO(sad->freq);
-				break;
+					stereo_freqs |= sad->freq;
+				else
+					break;
 			}
 		}
+
+		value |= SUPPORTED_FREQUENCIES_STEREO(stereo_freqs);
+
 		WREG32_ENDPOINT(offset, eld_reg_to_type[i][0], value);
 	}
 
diff --git a/drivers/gpu/drm/radeon/evergreen.c b/drivers/gpu/drm/radeon/evergreen.c
index 56f6bec34af5..9702e55e924e 100644
--- a/drivers/gpu/drm/radeon/evergreen.c
+++ b/drivers/gpu/drm/radeon/evergreen.c
@@ -1186,6 +1186,62 @@ void evergreen_fix_pci_max_read_req_size(struct radeon_device *rdev)
 		pcie_set_readrq(rdev->pdev, 512);
 }
 
+void dce4_program_fmt(struct drm_encoder *encoder)
+{
+	struct drm_device *dev = encoder->dev;
+	struct radeon_device *rdev = dev->dev_private;
+	struct radeon_encoder *radeon_encoder = to_radeon_encoder(encoder);
+	struct radeon_crtc *radeon_crtc = to_radeon_crtc(encoder->crtc);
+	struct drm_connector *connector = radeon_get_connector_for_encoder(encoder);
+	int bpc = 0;
+	u32 tmp = 0;
+	enum radeon_connector_dither dither = RADEON_FMT_DITHER_DISABLE;
+
+	if (connector) {
+		struct radeon_connector *radeon_connector = to_radeon_connector(connector);
+		bpc = radeon_get_monitor_bpc(connector);
+		dither = radeon_connector->dither;
+	}
+
+	/* LVDS/eDP FMT is set up by atom */
+	if (radeon_encoder->devices & ATOM_DEVICE_LCD_SUPPORT)
+		return;
+
+	/* not needed for analog */
+	if ((radeon_encoder->encoder_id == ENCODER_OBJECT_ID_INTERNAL_KLDSCP_DAC1) ||
+	    (radeon_encoder->encoder_id == ENCODER_OBJECT_ID_INTERNAL_KLDSCP_DAC2))
+		return;
+
+	if (bpc == 0)
+		return;
+
+	switch (bpc) {
+	case 6:
+		if (dither == RADEON_FMT_DITHER_ENABLE)
+			/* XXX sort out optimal dither settings */
+			tmp |= (FMT_FRAME_RANDOM_ENABLE | FMT_HIGHPASS_RANDOM_ENABLE |
+				FMT_SPATIAL_DITHER_EN);
+		else
+			tmp |= FMT_TRUNCATE_EN;
+		break;
+	case 8:
+		if (dither == RADEON_FMT_DITHER_ENABLE)
+			/* XXX sort out optimal dither settings */
+			tmp |= (FMT_FRAME_RANDOM_ENABLE | FMT_HIGHPASS_RANDOM_ENABLE |
+				FMT_RGB_RANDOM_ENABLE |
+				FMT_SPATIAL_DITHER_EN | FMT_SPATIAL_DITHER_DEPTH);
+		else
+			tmp |= (FMT_TRUNCATE_EN | FMT_TRUNCATE_DEPTH);
+		break;
+	case 10:
+	default:
+		/* not needed */
+		break;
+	}
+
+	WREG32(FMT_BIT_DEPTH_CONTROL + radeon_crtc->crtc_offset, tmp);
+}
+
 static bool dce4_is_in_vblank(struct radeon_device *rdev, int crtc)
 {
 	if (RREG32(EVERGREEN_CRTC_STATUS + crtc_offsets[crtc]) & EVERGREEN_CRTC_V_BLANK)
@@ -3956,7 +4012,7 @@ int sumo_rlc_init(struct radeon_device *rdev)
 		if (rdev->family >= CHIP_TAHITI) {
 			/* SI */
 			for (i = 0; i < rdev->rlc.reg_list_size; i++)
-				dst_ptr[i] = src_ptr[i];
+				dst_ptr[i] = cpu_to_le32(src_ptr[i]);
 		} else {
 			/* ON/LN/TN */
 			/* format:
@@ -3970,10 +4026,10 @@ int sumo_rlc_init(struct radeon_device *rdev)
 				if (i < dws)
 					data |= (src_ptr[i] >> 2) << 16;
 				j = (((i - 1) * 3) / 2);
-				dst_ptr[j] = data;
+				dst_ptr[j] = cpu_to_le32(data);
 			}
 			j = ((i * 3) / 2);
-			dst_ptr[j] = RLC_SAVE_RESTORE_LIST_END_MARKER;
+			dst_ptr[j] = cpu_to_le32(RLC_SAVE_RESTORE_LIST_END_MARKER);
 		}
 		radeon_bo_kunmap(rdev->rlc.save_restore_obj);
 		radeon_bo_unreserve(rdev->rlc.save_restore_obj);
@@ -4035,40 +4091,40 @@ int sumo_rlc_init(struct radeon_device *rdev)
 			cik_get_csb_buffer(rdev, dst_ptr);
 		} else if (rdev->family >= CHIP_TAHITI) {
 			reg_list_mc_addr = rdev->rlc.clear_state_gpu_addr + 256;
-			dst_ptr[0] = upper_32_bits(reg_list_mc_addr);
-			dst_ptr[1] = lower_32_bits(reg_list_mc_addr);
-			dst_ptr[2] = rdev->rlc.clear_state_size;
+			dst_ptr[0] = cpu_to_le32(upper_32_bits(reg_list_mc_addr));
+			dst_ptr[1] = cpu_to_le32(lower_32_bits(reg_list_mc_addr));
+			dst_ptr[2] = cpu_to_le32(rdev->rlc.clear_state_size);
 			si_get_csb_buffer(rdev, &dst_ptr[(256/4)]);
 		} else {
 			reg_list_hdr_blk_index = 0;
 			reg_list_mc_addr = rdev->rlc.clear_state_gpu_addr + (reg_list_blk_index * 4);
 			data = upper_32_bits(reg_list_mc_addr);
-			dst_ptr[reg_list_hdr_blk_index] = data;
+			dst_ptr[reg_list_hdr_blk_index] = cpu_to_le32(data);
 			reg_list_hdr_blk_index++;
 			for (i = 0; cs_data[i].section != NULL; i++) {
 				for (j = 0; cs_data[i].section[j].extent != NULL; j++) {
 					reg_num = cs_data[i].section[j].reg_count;
 					data = reg_list_mc_addr & 0xffffffff;
-					dst_ptr[reg_list_hdr_blk_index] = data;
+					dst_ptr[reg_list_hdr_blk_index] = cpu_to_le32(data);
 					reg_list_hdr_blk_index++;
 
 					data = (cs_data[i].section[j].reg_index * 4) & 0xffffffff;
-					dst_ptr[reg_list_hdr_blk_index] = data;
+					dst_ptr[reg_list_hdr_blk_index] = cpu_to_le32(data);
 					reg_list_hdr_blk_index++;
 
 					data = 0x08000000 | (reg_num * 4);
-					dst_ptr[reg_list_hdr_blk_index] = data;
+					dst_ptr[reg_list_hdr_blk_index] = cpu_to_le32(data);
 					reg_list_hdr_blk_index++;
 
 					for (k = 0; k < reg_num; k++) {
 						data = cs_data[i].section[j].extent[k];
-						dst_ptr[reg_list_blk_index + k] = data;
+						dst_ptr[reg_list_blk_index + k] = cpu_to_le32(data);
 					}
 					reg_list_mc_addr += reg_num * 4;
 					reg_list_blk_index += reg_num;
 				}
 			}
-			dst_ptr[reg_list_hdr_blk_index] = RLC_CLEAR_STATE_END_MARKER;
+			dst_ptr[reg_list_hdr_blk_index] = cpu_to_le32(RLC_CLEAR_STATE_END_MARKER);
 		}
 		radeon_bo_kunmap(rdev->rlc.clear_state_obj);
 		radeon_bo_unreserve(rdev->rlc.clear_state_obj);
diff --git a/drivers/gpu/drm/radeon/evergreen_hdmi.c b/drivers/gpu/drm/radeon/evergreen_hdmi.c
index 57fcc4b16a52..aa695c4feb3d 100644
--- a/drivers/gpu/drm/radeon/evergreen_hdmi.c
+++ b/drivers/gpu/drm/radeon/evergreen_hdmi.c
@@ -35,6 +35,8 @@
 extern void dce6_afmt_write_speaker_allocation(struct drm_encoder *encoder);
 extern void dce6_afmt_write_sad_regs(struct drm_encoder *encoder);
 extern void dce6_afmt_select_pin(struct drm_encoder *encoder);
+extern void dce6_afmt_write_latency_fields(struct drm_encoder *encoder,
+					   struct drm_display_mode *mode);
 
 /*
  * update the N and CTS parameters for a given pixel clock rate
@@ -58,6 +60,42 @@ static void evergreen_hdmi_update_ACR(struct drm_encoder *encoder, uint32_t cloc
 	WREG32(HDMI_ACR_48_1 + offset, acr.n_48khz);
 }
 
+static void dce4_afmt_write_latency_fields(struct drm_encoder *encoder,
+					   struct drm_display_mode *mode)
+{
+	struct radeon_device *rdev = encoder->dev->dev_private;
+	struct drm_connector *connector;
+	struct radeon_connector *radeon_connector = NULL;
+	u32 tmp = 0;
+
+	list_for_each_entry(connector, &encoder->dev->mode_config.connector_list, head) {
+		if (connector->encoder == encoder) {
+			radeon_connector = to_radeon_connector(connector);
+			break;
+		}
+	}
+
+	if (!radeon_connector) {
+		DRM_ERROR("Couldn't find encoder's connector\n");
+		return;
+	}
+
+	if (mode->flags & DRM_MODE_FLAG_INTERLACE) {
+		if (connector->latency_present[1])
+			tmp = VIDEO_LIPSYNC(connector->video_latency[1]) |
+				AUDIO_LIPSYNC(connector->audio_latency[1]);
+		else
+			tmp = VIDEO_LIPSYNC(255) | AUDIO_LIPSYNC(255);
+	} else {
+		if (connector->latency_present[0])
+			tmp = VIDEO_LIPSYNC(connector->video_latency[0]) |
+				AUDIO_LIPSYNC(connector->audio_latency[0]);
+		else
+			tmp = VIDEO_LIPSYNC(255) | AUDIO_LIPSYNC(255);
+	}
+	WREG32(AZ_F0_CODEC_PIN0_CONTROL_RESPONSE_LIPSYNC, tmp);
+}
+
 static void dce4_afmt_write_speaker_allocation(struct drm_encoder *encoder)
 {
 	struct radeon_device *rdev = encoder->dev->dev_private;
@@ -67,12 +105,11 @@ static void dce4_afmt_write_speaker_allocation(struct drm_encoder *encoder)
 	u8 *sadb;
 	int sad_count;
 
-	/* XXX: setting this register causes hangs on some asics */
-	return;
-
 	list_for_each_entry(connector, &encoder->dev->mode_config.connector_list, head) {
-		if (connector->encoder == encoder)
+		if (connector->encoder == encoder) {
 			radeon_connector = to_radeon_connector(connector);
+			break;
+		}
 	}
 
 	if (!radeon_connector) {
@@ -124,8 +161,10 @@ static void evergreen_hdmi_write_sad_regs(struct drm_encoder *encoder)
 	};
 
 	list_for_each_entry(connector, &encoder->dev->mode_config.connector_list, head) {
-		if (connector->encoder == encoder)
+		if (connector->encoder == encoder) {
 			radeon_connector = to_radeon_connector(connector);
+			break;
+		}
 	}
 
 	if (!radeon_connector) {
@@ -142,20 +181,30 @@ static void evergreen_hdmi_write_sad_regs(struct drm_encoder *encoder)
 
 	for (i = 0; i < ARRAY_SIZE(eld_reg_to_type); i++) {
 		u32 value = 0;
+		u8 stereo_freqs = 0;
+		int max_channels = -1;
 		int j;
 
 		for (j = 0; j < sad_count; j++) {
 			struct cea_sad *sad = &sads[j];
 
 			if (sad->format == eld_reg_to_type[i][1]) {
-				value = MAX_CHANNELS(sad->channels) |
-					DESCRIPTOR_BYTE_2(sad->byte2) |
-					SUPPORTED_FREQUENCIES(sad->freq);
+				if (sad->channels > max_channels) {
+					value = MAX_CHANNELS(sad->channels) |
+						DESCRIPTOR_BYTE_2(sad->byte2) |
+						SUPPORTED_FREQUENCIES(sad->freq);
+					max_channels = sad->channels;
+				}
+
 				if (sad->format == HDMI_AUDIO_CODING_TYPE_PCM)
-					value |= SUPPORTED_FREQUENCIES_STEREO(sad->freq);
-				break;
+					stereo_freqs |= sad->freq;
+				else
+					break;
 			}
 		}
+
+		value |= SUPPORTED_FREQUENCIES_STEREO(stereo_freqs);
+
 		WREG32(eld_reg_to_type[i][0], value);
 	}
 
@@ -324,8 +373,10 @@ void evergreen_hdmi_setmode(struct drm_encoder *encoder, struct drm_display_mode
 	if (ASIC_IS_DCE6(rdev)) {
 		dce6_afmt_select_pin(encoder);
 		dce6_afmt_write_sad_regs(encoder);
+		dce6_afmt_write_latency_fields(encoder, mode);
 	} else {
 		evergreen_hdmi_write_sad_regs(encoder);
+		dce4_afmt_write_latency_fields(encoder, mode);
 	}
 
 	err = drm_hdmi_avi_infoframe_from_display_mode(&frame, mode);
diff --git a/drivers/gpu/drm/radeon/evergreend.h b/drivers/gpu/drm/radeon/evergreend.h
index 4f6d2962767d..17f990798992 100644
--- a/drivers/gpu/drm/radeon/evergreend.h
+++ b/drivers/gpu/drm/radeon/evergreend.h
@@ -750,6 +750,44 @@
  * bit6 = 192 kHz
  */
 
+#define AZ_CHANNEL_COUNT_CONTROL                          0x5fe4
+#       define HBR_CHANNEL_COUNT(x)                       (((x) & 0x7) << 0)
+#       define COMPRESSED_CHANNEL_COUNT(x)                (((x) & 0x7) << 4)
+/* HBR_CHANNEL_COUNT, COMPRESSED_CHANNEL_COUNT
+ * 0   = use stream header
+ * 1-7 = channel count - 1
+ */
+#define AZ_F0_CODEC_PIN0_CONTROL_RESPONSE_LIPSYNC         0x5fe8
+#       define VIDEO_LIPSYNC(x)                           (((x) & 0xff) << 0)
+#       define AUDIO_LIPSYNC(x)                           (((x) & 0xff) << 8)
+/* VIDEO_LIPSYNC, AUDIO_LIPSYNC
+ * 0   = invalid
+ * x   = legal delay value
+ * 255 = sync not supported
+ */
+#define AZ_F0_CODEC_PIN0_CONTROL_RESPONSE_HBR             0x5fec
+#       define HBR_CAPABLE                                (1 << 0) /* enabled by default */
+
+#define AZ_F0_CODEC_PIN0_CONTROL_RESPONSE_AV_ASSOCIATION0 0x5ff4
+#       define DISPLAY0_TYPE(x)                           (((x) & 0x3) << 0)
+#       define DISPLAY_TYPE_NONE                   0
+#       define DISPLAY_TYPE_HDMI                   1
+#       define DISPLAY_TYPE_DP                     2
+#       define DISPLAY0_ID(x)                             (((x) & 0x3f) << 2)
+#       define DISPLAY1_TYPE(x)                           (((x) & 0x3) << 8)
+#       define DISPLAY1_ID(x)                             (((x) & 0x3f) << 10)
+#       define DISPLAY2_TYPE(x)                           (((x) & 0x3) << 16)
+#       define DISPLAY2_ID(x)                             (((x) & 0x3f) << 18)
+#       define DISPLAY3_TYPE(x)                           (((x) & 0x3) << 24)
+#       define DISPLAY3_ID(x)                             (((x) & 0x3f) << 26)
+#define AZ_F0_CODEC_PIN0_CONTROL_RESPONSE_AV_ASSOCIATION1 0x5ff8
+#       define DISPLAY4_TYPE(x)                           (((x) & 0x3) << 0)
+#       define DISPLAY4_ID(x)                             (((x) & 0x3f) << 2)
+#       define DISPLAY5_TYPE(x)                           (((x) & 0x3) << 8)
+#       define DISPLAY5_ID(x)                             (((x) & 0x3f) << 10)
+#define AZ_F0_CODEC_PIN0_CONTROL_RESPONSE_AV_NUMBER       0x5ffc
+#       define NUMBER_OF_DISPLAY_ID(x)                    (((x) & 0x7) << 0)
+
 #define AZ_HOT_PLUG_CONTROL                               0x5e78
 #       define AZ_FORCE_CODEC_WAKE                        (1 << 0)
 #       define PIN0_JACK_DETECTION_ENABLE                 (1 << 4)
@@ -1312,6 +1350,38 @@
 #       define DC_HPDx_RX_INT_TIMER(x)                    ((x) << 16)
 #       define DC_HPDx_EN                                 (1 << 28)
 
+/* DCE4/5/6 FMT blocks */
+#define FMT_DYNAMIC_EXP_CNTL                 0x6fb4
+#       define FMT_DYNAMIC_EXP_EN            (1 << 0)
+#       define FMT_DYNAMIC_EXP_MODE          (1 << 4)
+        /* 0 = 10bit -> 12bit, 1 = 8bit -> 12bit */
+#define FMT_CONTROL                          0x6fb8
+#       define FMT_PIXEL_ENCODING            (1 << 16)
+        /* 0 = RGB 4:4:4 or YCbCr 4:4:4, 1 = YCbCr 4:2:2 */
+#define FMT_BIT_DEPTH_CONTROL                0x6fc8
+#       define FMT_TRUNCATE_EN               (1 << 0)
+#       define FMT_TRUNCATE_DEPTH            (1 << 4)
+#       define FMT_SPATIAL_DITHER_EN         (1 << 8)
+#       define FMT_SPATIAL_DITHER_MODE(x)    ((x) << 9)
+#       define FMT_SPATIAL_DITHER_DEPTH      (1 << 12)
+#       define FMT_FRAME_RANDOM_ENABLE       (1 << 13)
+#       define FMT_RGB_RANDOM_ENABLE         (1 << 14)
+#       define FMT_HIGHPASS_RANDOM_ENABLE    (1 << 15)
+#       define FMT_TEMPORAL_DITHER_EN        (1 << 16)
+#       define FMT_TEMPORAL_DITHER_DEPTH     (1 << 20)
+#       define FMT_TEMPORAL_DITHER_OFFSET(x) ((x) << 21)
+#       define FMT_TEMPORAL_LEVEL            (1 << 24)
+#       define FMT_TEMPORAL_DITHER_RESET     (1 << 25)
+#       define FMT_25FRC_SEL(x)              ((x) << 26)
+#       define FMT_50FRC_SEL(x)              ((x) << 28)
+#       define FMT_75FRC_SEL(x)              ((x) << 30)
+#define FMT_CLAMP_CONTROL                    0x6fe4
+#       define FMT_CLAMP_DATA_EN             (1 << 0)
+#       define FMT_CLAMP_COLOR_FORMAT(x)     ((x) << 16)
+#       define FMT_CLAMP_6BPC                0
+#       define FMT_CLAMP_8BPC                1
+#       define FMT_CLAMP_10BPC               2
+
 /* ASYNC DMA */
 #define DMA_RB_RPTR                                       0xd008
 #define DMA_RB_WPTR                                       0xd00c
diff --git a/drivers/gpu/drm/radeon/ni.c b/drivers/gpu/drm/radeon/ni.c
index cac2866d79da..11aab2ab54ce 100644
--- a/drivers/gpu/drm/radeon/ni.c
+++ b/drivers/gpu/drm/radeon/ni.c
@@ -174,11 +174,6 @@ extern void evergreen_pcie_gen2_enable(struct radeon_device *rdev);
 extern void evergreen_program_aspm(struct radeon_device *rdev);
 extern void sumo_rlc_fini(struct radeon_device *rdev);
 extern int sumo_rlc_init(struct radeon_device *rdev);
-extern void cayman_dma_vm_set_page(struct radeon_device *rdev,
-				   struct radeon_ib *ib,
-				   uint64_t pe,
-				   uint64_t addr, unsigned count,
-				   uint32_t incr, uint32_t flags);
 
 /* Firmware Names */
 MODULE_FIRMWARE("radeon/BARTS_pfp.bin");
@@ -2400,77 +2395,6 @@ void cayman_vm_decode_fault(struct radeon_device *rdev,
 	       block, mc_id);
 }
 
-#define R600_ENTRY_VALID   (1 << 0)
-#define R600_PTE_SYSTEM    (1 << 1)
-#define R600_PTE_SNOOPED   (1 << 2)
-#define R600_PTE_READABLE  (1 << 5)
-#define R600_PTE_WRITEABLE (1 << 6)
-
-uint32_t cayman_vm_page_flags(struct radeon_device *rdev, uint32_t flags)
-{
-	uint32_t r600_flags = 0;
-	r600_flags |= (flags & RADEON_VM_PAGE_VALID) ? R600_ENTRY_VALID : 0;
-	r600_flags |= (flags & RADEON_VM_PAGE_READABLE) ? R600_PTE_READABLE : 0;
-	r600_flags |= (flags & RADEON_VM_PAGE_WRITEABLE) ? R600_PTE_WRITEABLE : 0;
-	if (flags & RADEON_VM_PAGE_SYSTEM) {
-		r600_flags |= R600_PTE_SYSTEM;
-		r600_flags |= (flags & RADEON_VM_PAGE_SNOOPED) ? R600_PTE_SNOOPED : 0;
-	}
-	return r600_flags;
-}
-
-/**
- * cayman_vm_set_page - update the page tables using the CP
- *
- * @rdev: radeon_device pointer
- * @ib: indirect buffer to fill with commands
- * @pe: addr of the page entry
- * @addr: dst addr to write into pe
- * @count: number of page entries to update
- * @incr: increase next addr by incr bytes
- * @flags: access flags
- *
- * Update the page tables using the CP (cayman/TN).
- */
-void cayman_vm_set_page(struct radeon_device *rdev,
-			struct radeon_ib *ib,
-			uint64_t pe,
-			uint64_t addr, unsigned count,
-			uint32_t incr, uint32_t flags)
-{
-	uint32_t r600_flags = cayman_vm_page_flags(rdev, flags);
-	uint64_t value;
-	unsigned ndw;
-
-	if (rdev->asic->vm.pt_ring_index == RADEON_RING_TYPE_GFX_INDEX) {
-		while (count) {
-			ndw = 1 + count * 2;
-			if (ndw > 0x3FFF)
-				ndw = 0x3FFF;
-
-			ib->ptr[ib->length_dw++] = PACKET3(PACKET3_ME_WRITE, ndw);
-			ib->ptr[ib->length_dw++] = pe;
-			ib->ptr[ib->length_dw++] = upper_32_bits(pe) & 0xff;
-			for (; ndw > 1; ndw -= 2, --count, pe += 8) {
-				if (flags & RADEON_VM_PAGE_SYSTEM) {
-					value = radeon_vm_map_gart(rdev, addr);
-					value &= 0xFFFFFFFFFFFFF000ULL;
-				} else if (flags & RADEON_VM_PAGE_VALID) {
-					value = addr;
-				} else {
-					value = 0;
-				}
-				addr += incr;
-				value |= r600_flags;
-				ib->ptr[ib->length_dw++] = value;
-				ib->ptr[ib->length_dw++] = upper_32_bits(value);
-			}
-		}
-	} else {
-		cayman_dma_vm_set_page(rdev, ib, pe, addr, count, incr, flags);
-	}
-}
-
 /**
  * cayman_vm_flush - vm flush using the CP
  *
diff --git a/drivers/gpu/drm/radeon/ni_dma.c b/drivers/gpu/drm/radeon/ni_dma.c
index dd6e9688fbef..bdeb65ed3658 100644
--- a/drivers/gpu/drm/radeon/ni_dma.c
+++ b/drivers/gpu/drm/radeon/ni_dma.c
@@ -24,6 +24,7 @@
 #include <drm/drmP.h>
 #include "radeon.h"
 #include "radeon_asic.h"
+#include "radeon_trace.h"
 #include "nid.h"
 
 u32 cayman_gpu_check_soft_reset(struct radeon_device *rdev);
@@ -245,8 +246,7 @@ bool cayman_dma_is_lockup(struct radeon_device *rdev, struct radeon_ring *ring)
  * @addr: dst addr to write into pe
  * @count: number of page entries to update
  * @incr: increase next addr by incr bytes
- * @flags: access flags
- * @r600_flags: hw access flags 
+ * @flags: hw access flags 
  *
  * Update the page tables using the DMA (cayman/TN).
  */
@@ -256,11 +256,12 @@ void cayman_dma_vm_set_page(struct radeon_device *rdev,
 			    uint64_t addr, unsigned count,
 			    uint32_t incr, uint32_t flags)
 {
-	uint32_t r600_flags = cayman_vm_page_flags(rdev, flags);
 	uint64_t value;
 	unsigned ndw;
 
-	if ((flags & RADEON_VM_PAGE_SYSTEM) || (count == 1)) {
+	trace_radeon_vm_set_page(pe, addr, count, incr, flags);
+
+	if ((flags & R600_PTE_SYSTEM) || (count == 1)) {
 		while (count) {
 			ndw = count * 2;
 			if (ndw > 0xFFFFE)
@@ -271,16 +272,16 @@ void cayman_dma_vm_set_page(struct radeon_device *rdev,
 			ib->ptr[ib->length_dw++] = pe;
 			ib->ptr[ib->length_dw++] = upper_32_bits(pe) & 0xff;
 			for (; ndw > 0; ndw -= 2, --count, pe += 8) {
-				if (flags & RADEON_VM_PAGE_SYSTEM) {
+				if (flags & R600_PTE_SYSTEM) {
 					value = radeon_vm_map_gart(rdev, addr);
 					value &= 0xFFFFFFFFFFFFF000ULL;
-				} else if (flags & RADEON_VM_PAGE_VALID) {
+				} else if (flags & R600_PTE_VALID) {
 					value = addr;
 				} else {
 					value = 0;
 				}
 				addr += incr;
-				value |= r600_flags;
+				value |= flags;
 				ib->ptr[ib->length_dw++] = value;
 				ib->ptr[ib->length_dw++] = upper_32_bits(value);
 			}
@@ -291,7 +292,7 @@ void cayman_dma_vm_set_page(struct radeon_device *rdev,
 			if (ndw > 0xFFFFE)
 				ndw = 0xFFFFE;
 
-			if (flags & RADEON_VM_PAGE_VALID)
+			if (flags & R600_PTE_VALID)
 				value = addr;
 			else
 				value = 0;
@@ -299,7 +300,7 @@ void cayman_dma_vm_set_page(struct radeon_device *rdev,
 			ib->ptr[ib->length_dw++] = DMA_PTE_PDE_PACKET(ndw);
 			ib->ptr[ib->length_dw++] = pe; /* dst addr */
 			ib->ptr[ib->length_dw++] = upper_32_bits(pe) & 0xff;
-			ib->ptr[ib->length_dw++] = r600_flags; /* mask */
+			ib->ptr[ib->length_dw++] = flags; /* mask */
 			ib->ptr[ib->length_dw++] = 0;
 			ib->ptr[ib->length_dw++] = value; /* value */
 			ib->ptr[ib->length_dw++] = upper_32_bits(value);
diff --git a/drivers/gpu/drm/radeon/r100.c b/drivers/gpu/drm/radeon/r100.c
index d71333033b2b..784983d78158 100644
--- a/drivers/gpu/drm/radeon/r100.c
+++ b/drivers/gpu/drm/radeon/r100.c
@@ -1434,7 +1434,7 @@ int r100_cs_packet_parse_vline(struct radeon_cs_parser *p)
 	obj = drm_mode_object_find(p->rdev->ddev, crtc_id, DRM_MODE_OBJECT_CRTC);
 	if (!obj) {
 		DRM_ERROR("cannot find crtc %d\n", crtc_id);
-		return -EINVAL;
+		return -ENOENT;
 	}
 	crtc = obj_to_crtc(obj);
 	radeon_crtc = to_radeon_crtc(crtc);
diff --git a/drivers/gpu/drm/radeon/r600.c b/drivers/gpu/drm/radeon/r600.c
index f9be22062df1..4e609e8a8d2b 100644
--- a/drivers/gpu/drm/radeon/r600.c
+++ b/drivers/gpu/drm/radeon/r600.c
@@ -124,6 +124,59 @@ int r600_set_uvd_clocks(struct radeon_device *rdev, u32 vclk, u32 dclk)
 	return 0;
 }
 
+void dce3_program_fmt(struct drm_encoder *encoder)
+{
+	struct drm_device *dev = encoder->dev;
+	struct radeon_device *rdev = dev->dev_private;
+	struct radeon_encoder *radeon_encoder = to_radeon_encoder(encoder);
+	struct radeon_crtc *radeon_crtc = to_radeon_crtc(encoder->crtc);
+	struct drm_connector *connector = radeon_get_connector_for_encoder(encoder);
+	int bpc = 0;
+	u32 tmp = 0;
+	enum radeon_connector_dither dither = RADEON_FMT_DITHER_DISABLE;
+
+	if (connector) {
+		struct radeon_connector *radeon_connector = to_radeon_connector(connector);
+		bpc = radeon_get_monitor_bpc(connector);
+		dither = radeon_connector->dither;
+	}
+
+	/* LVDS FMT is set up by atom */
+	if (radeon_encoder->devices & ATOM_DEVICE_LCD_SUPPORT)
+		return;
+
+	/* not needed for analog */
+	if ((radeon_encoder->encoder_id == ENCODER_OBJECT_ID_INTERNAL_KLDSCP_DAC1) ||
+	    (radeon_encoder->encoder_id == ENCODER_OBJECT_ID_INTERNAL_KLDSCP_DAC2))
+		return;
+
+	if (bpc == 0)
+		return;
+
+	switch (bpc) {
+	case 6:
+		if (dither == RADEON_FMT_DITHER_ENABLE)
+			/* XXX sort out optimal dither settings */
+			tmp |= FMT_SPATIAL_DITHER_EN;
+		else
+			tmp |= FMT_TRUNCATE_EN;
+		break;
+	case 8:
+		if (dither == RADEON_FMT_DITHER_ENABLE)
+			/* XXX sort out optimal dither settings */
+			tmp |= (FMT_SPATIAL_DITHER_EN | FMT_SPATIAL_DITHER_DEPTH);
+		else
+			tmp |= (FMT_TRUNCATE_EN | FMT_TRUNCATE_DEPTH);
+		break;
+	case 10:
+	default:
+		/* not needed */
+		break;
+	}
+
+	WREG32(FMT_BIT_DEPTH_CONTROL + radeon_crtc->crtc_offset, tmp);
+}
+
 /* get temperature in millidegrees */
 int rv6xx_get_temp(struct radeon_device *rdev)
 {
diff --git a/drivers/gpu/drm/radeon/r600_cs.c b/drivers/gpu/drm/radeon/r600_cs.c
index 01a3ec83f284..5dceea6f71ae 100644
--- a/drivers/gpu/drm/radeon/r600_cs.c
+++ b/drivers/gpu/drm/radeon/r600_cs.c
@@ -887,7 +887,7 @@ int r600_cs_common_vline_parse(struct radeon_cs_parser *p,
 	obj = drm_mode_object_find(p->rdev->ddev, crtc_id, DRM_MODE_OBJECT_CRTC);
 	if (!obj) {
 		DRM_ERROR("cannot find crtc %d\n", crtc_id);
-		return -EINVAL;
+		return -ENOENT;
 	}
 	crtc = obj_to_crtc(obj);
 	radeon_crtc = to_radeon_crtc(crtc);
@@ -2328,13 +2328,8 @@ static void r600_cs_parser_fini(struct radeon_cs_parser *parser, int error)
 	unsigned i;
 
 	kfree(parser->relocs);
-	for (i = 0; i < parser->nchunks; i++) {
-		kfree(parser->chunks[i].kdata);
-		if (parser->rdev && (parser->rdev->flags & RADEON_IS_AGP)) {
-			kfree(parser->chunks[i].kpage[0]);
-			kfree(parser->chunks[i].kpage[1]);
-		}
-	}
+	for (i = 0; i < parser->nchunks; i++)
+		drm_free_large(parser->chunks[i].kdata);
 	kfree(parser->chunks);
 	kfree(parser->chunks_array);
 }
@@ -2391,13 +2386,12 @@ int r600_cs_legacy(struct drm_device *dev, void *data, struct drm_file *filp,
 	ib_chunk = &parser.chunks[parser.chunk_ib_idx];
 	parser.ib.length_dw = ib_chunk->length_dw;
 	*l = parser.ib.length_dw;
-	r = r600_cs_parse(&parser);
-	if (r) {
-		DRM_ERROR("Invalid command stream !\n");
+	if (DRM_COPY_FROM_USER(ib, ib_chunk->user_ptr, ib_chunk->length_dw * 4)) {
+		r = -EFAULT;
 		r600_cs_parser_fini(&parser, r);
 		return r;
 	}
-	r = radeon_cs_finish_pages(&parser);
+	r = r600_cs_parse(&parser);
 	if (r) {
 		DRM_ERROR("Invalid command stream !\n");
 		r600_cs_parser_fini(&parser, r);
diff --git a/drivers/gpu/drm/radeon/r600_hdmi.c b/drivers/gpu/drm/radeon/r600_hdmi.c
index 06022e3b9c3b..4b89262f3f0e 100644
--- a/drivers/gpu/drm/radeon/r600_hdmi.c
+++ b/drivers/gpu/drm/radeon/r600_hdmi.c
@@ -24,6 +24,7 @@
  * Authors: Christian König
  */
 #include <linux/hdmi.h>
+#include <linux/gcd.h>
 #include <drm/drmP.h>
 #include <drm/radeon_drm.h>
 #include "radeon.h"
@@ -57,35 +58,57 @@ enum r600_hdmi_iec_status_bits {
 static const struct radeon_hdmi_acr r600_hdmi_predefined_acr[] = {
     /*	     32kHz	  44.1kHz	48kHz    */
     /* Clock      N     CTS      N     CTS      N     CTS */
-    {  25175,  4576,  28125,  7007,  31250,  6864,  28125 }, /*  25,20/1.001 MHz */
+    {  25175,  4096,  25175, 28224, 125875,  6144,  25175 }, /*  25,20/1.001 MHz */
     {  25200,  4096,  25200,  6272,  28000,  6144,  25200 }, /*  25.20       MHz */
     {  27000,  4096,  27000,  6272,  30000,  6144,  27000 }, /*  27.00       MHz */
     {  27027,  4096,  27027,  6272,  30030,  6144,  27027 }, /*  27.00*1.001 MHz */
     {  54000,  4096,  54000,  6272,  60000,  6144,  54000 }, /*  54.00       MHz */
     {  54054,  4096,  54054,  6272,  60060,  6144,  54054 }, /*  54.00*1.001 MHz */
-    {  74176, 11648, 210937, 17836, 234375, 11648, 140625 }, /*  74.25/1.001 MHz */
+    {  74176,  4096,  74176,  5733,  75335,  6144,  74176 }, /*  74.25/1.001 MHz */
     {  74250,  4096,  74250,  6272,  82500,  6144,  74250 }, /*  74.25       MHz */
-    { 148352, 11648, 421875,  8918, 234375,  5824, 140625 }, /* 148.50/1.001 MHz */
+    { 148352,  4096, 148352,  5733, 150670,  6144, 148352 }, /* 148.50/1.001 MHz */
     { 148500,  4096, 148500,  6272, 165000,  6144, 148500 }, /* 148.50       MHz */
-    {      0,  4096,      0,  6272,      0,  6144,      0 }  /* Other */
 };
 
+
 /*
- * calculate CTS value if it's not found in the table
+ * calculate CTS and N values if they are not found in the table
  */
-static void r600_hdmi_calc_cts(uint32_t clock, int *CTS, int N, int freq)
+static void r600_hdmi_calc_cts(uint32_t clock, int *CTS, int *N, int freq)
 {
-	u64 n;
-	u32 d;
-
-	if (*CTS == 0) {
-		n = (u64)clock * (u64)N * 1000ULL;
-		d = 128 * freq;
-		do_div(n, d);
-		*CTS = n;
-	}
-	DRM_DEBUG("Using ACR timing N=%d CTS=%d for frequency %d\n",
-		  N, *CTS, freq);
+	int n, cts;
+	unsigned long div, mul;
+
+	/* Safe, but overly large values */
+	n = 128 * freq;
+	cts = clock * 1000;
+
+	/* Smallest valid fraction */
+	div = gcd(n, cts);
+
+	n /= div;
+	cts /= div;
+
+	/*
+	 * The optimal N is 128*freq/1000. Calculate the closest larger
+	 * value that doesn't truncate any bits.
+	 */
+	mul = ((128*freq/1000) + (n-1))/n;
+
+	n *= mul;
+	cts *= mul;
+
+	/* Check that we are in spec (not always possible) */
+	if (n < (128*freq/1500))
+		printk(KERN_WARNING "Calculated ACR N value is too small. You may experience audio problems.\n");
+	if (n > (128*freq/300))
+		printk(KERN_WARNING "Calculated ACR N value is too large. You may experience audio problems.\n");
+
+	*N = n;
+	*CTS = cts;
+
+	DRM_DEBUG("Calculated ACR timing N=%d CTS=%d for frequency %d\n",
+		  *N, *CTS, freq);
 }
 
 struct radeon_hdmi_acr r600_hdmi_acr(uint32_t clock)
@@ -93,15 +116,16 @@ struct radeon_hdmi_acr r600_hdmi_acr(uint32_t clock)
 	struct radeon_hdmi_acr res;
 	u8 i;
 
-	for (i = 0; r600_hdmi_predefined_acr[i].clock != clock &&
-	     r600_hdmi_predefined_acr[i].clock != 0; i++)
-		;
-	res = r600_hdmi_predefined_acr[i];
+	/* Precalculated values for common clocks */
+	for (i = 0; i < ARRAY_SIZE(r600_hdmi_predefined_acr); i++) {
+		if (r600_hdmi_predefined_acr[i].clock == clock)
+			return r600_hdmi_predefined_acr[i];
+	}
 
-	/* In case some CTS are missing */
-	r600_hdmi_calc_cts(clock, &res.cts_32khz, res.n_32khz, 32000);
-	r600_hdmi_calc_cts(clock, &res.cts_44_1khz, res.n_44_1khz, 44100);
-	r600_hdmi_calc_cts(clock, &res.cts_48khz, res.n_48khz, 48000);
+	/* And odd clocks get manually calculated */
+	r600_hdmi_calc_cts(clock, &res.cts_32khz, &res.n_32khz, 32000);
+	r600_hdmi_calc_cts(clock, &res.cts_44_1khz, &res.n_44_1khz, 44100);
+	r600_hdmi_calc_cts(clock, &res.cts_48khz, &res.n_48khz, 48000);
 
 	return res;
 }
@@ -313,8 +337,10 @@ static void dce3_2_afmt_write_speaker_allocation(struct drm_encoder *encoder)
 	return;
 
 	list_for_each_entry(connector, &encoder->dev->mode_config.connector_list, head) {
-		if (connector->encoder == encoder)
+		if (connector->encoder == encoder) {
 			radeon_connector = to_radeon_connector(connector);
+			break;
+		}
 	}
 
 	if (!radeon_connector) {
@@ -366,8 +392,10 @@ static void dce3_2_afmt_write_sad_regs(struct drm_encoder *encoder)
 	};
 
 	list_for_each_entry(connector, &encoder->dev->mode_config.connector_list, head) {
-		if (connector->encoder == encoder)
+		if (connector->encoder == encoder) {
 			radeon_connector = to_radeon_connector(connector);
+			break;
+		}
 	}
 
 	if (!radeon_connector) {
@@ -384,20 +412,30 @@ static void dce3_2_afmt_write_sad_regs(struct drm_encoder *encoder)
 
 	for (i = 0; i < ARRAY_SIZE(eld_reg_to_type); i++) {
 		u32 value = 0;
+		u8 stereo_freqs = 0;
+		int max_channels = -1;
 		int j;
 
 		for (j = 0; j < sad_count; j++) {
 			struct cea_sad *sad = &sads[j];
 
 			if (sad->format == eld_reg_to_type[i][1]) {
-				value = MAX_CHANNELS(sad->channels) |
-					DESCRIPTOR_BYTE_2(sad->byte2) |
-					SUPPORTED_FREQUENCIES(sad->freq);
+				if (sad->channels > max_channels) {
+					value = MAX_CHANNELS(sad->channels) |
+						DESCRIPTOR_BYTE_2(sad->byte2) |
+						SUPPORTED_FREQUENCIES(sad->freq);
+					max_channels = sad->channels;
+				}
+
 				if (sad->format == HDMI_AUDIO_CODING_TYPE_PCM)
-					value |= SUPPORTED_FREQUENCIES_STEREO(sad->freq);
-				break;
+					stereo_freqs |= sad->freq;
+				else
+					break;
 			}
 		}
+
+		value |= SUPPORTED_FREQUENCIES_STEREO(stereo_freqs);
+
 		WREG32(eld_reg_to_type[i][0], value);
 	}
 
diff --git a/drivers/gpu/drm/radeon/r600d.h b/drivers/gpu/drm/radeon/r600d.h
index 7b3c7b5932c5..ebe38724a976 100644
--- a/drivers/gpu/drm/radeon/r600d.h
+++ b/drivers/gpu/drm/radeon/r600d.h
@@ -1199,6 +1199,34 @@
 #       define AFMT_AZ_FORMAT_WTRIG_ACK      (1 << 29)
 #       define AFMT_AZ_AUDIO_ENABLE_CHG_ACK  (1 << 30)
 
+/* DCE3 FMT blocks */
+#define FMT_CONTROL                          0x6700
+#       define FMT_PIXEL_ENCODING            (1 << 16)
+        /* 0 = RGB 4:4:4 or YCbCr 4:4:4, 1 = YCbCr 4:2:2 */
+#define FMT_BIT_DEPTH_CONTROL                0x6710
+#       define FMT_TRUNCATE_EN               (1 << 0)
+#       define FMT_TRUNCATE_DEPTH            (1 << 4)
+#       define FMT_SPATIAL_DITHER_EN         (1 << 8)
+#       define FMT_SPATIAL_DITHER_MODE(x)    ((x) << 9)
+#       define FMT_SPATIAL_DITHER_DEPTH      (1 << 12)
+#       define FMT_FRAME_RANDOM_ENABLE       (1 << 13)
+#       define FMT_RGB_RANDOM_ENABLE         (1 << 14)
+#       define FMT_HIGHPASS_RANDOM_ENABLE    (1 << 15)
+#       define FMT_TEMPORAL_DITHER_EN        (1 << 16)
+#       define FMT_TEMPORAL_DITHER_DEPTH     (1 << 20)
+#       define FMT_TEMPORAL_DITHER_OFFSET(x) ((x) << 21)
+#       define FMT_TEMPORAL_LEVEL            (1 << 24)
+#       define FMT_TEMPORAL_DITHER_RESET     (1 << 25)
+#       define FMT_25FRC_SEL(x)              ((x) << 26)
+#       define FMT_50FRC_SEL(x)              ((x) << 28)
+#       define FMT_75FRC_SEL(x)              ((x) << 30)
+#define FMT_CLAMP_CONTROL                    0x672c
+#       define FMT_CLAMP_DATA_EN             (1 << 0)
+#       define FMT_CLAMP_COLOR_FORMAT(x)     ((x) << 16)
+#       define FMT_CLAMP_6BPC                0
+#       define FMT_CLAMP_8BPC                1
+#       define FMT_CLAMP_10BPC               2
+
 /* Power management */
 #define CG_SPLL_FUNC_CNTL                                 0x600
 #       define SPLL_RESET                                (1 << 0)
diff --git a/drivers/gpu/drm/radeon/radeon.h b/drivers/gpu/drm/radeon/radeon.h
index 24f4960f59ee..b9ee99258602 100644
--- a/drivers/gpu/drm/radeon/radeon.h
+++ b/drivers/gpu/drm/radeon/radeon.h
@@ -98,6 +98,7 @@ extern int radeon_lockup_timeout;
 extern int radeon_fastfb;
 extern int radeon_dpm;
 extern int radeon_aspm;
+extern int radeon_runtime_pm;
 
 /*
  * Copy from radeon_drv.h so we don't have to include both and have conflicting
@@ -327,7 +328,6 @@ struct radeon_fence_driver {
 	/* sync_seq is protected by ring emission lock */
 	uint64_t			sync_seq[RADEON_NUM_RINGS];
 	atomic64_t			last_seq;
-	unsigned long			last_activity;
 	bool				initialized;
 };
 
@@ -832,6 +832,12 @@ struct radeon_mec {
 #define RADEON_VM_PTB_ALIGN_MASK (RADEON_VM_PTB_ALIGN_SIZE - 1)
 #define RADEON_VM_PTB_ALIGN(a) (((a) + RADEON_VM_PTB_ALIGN_MASK) & ~RADEON_VM_PTB_ALIGN_MASK)
 
+#define R600_PTE_VALID		(1 << 0)
+#define R600_PTE_SYSTEM		(1 << 1)
+#define R600_PTE_SNOOPED	(1 << 2)
+#define R600_PTE_READABLE	(1 << 5)
+#define R600_PTE_WRITEABLE	(1 << 6)
+
 struct radeon_vm {
 	struct list_head		list;
 	struct list_head		va;
@@ -967,12 +973,8 @@ struct radeon_cs_reloc {
 struct radeon_cs_chunk {
 	uint32_t		chunk_id;
 	uint32_t		length_dw;
-	int			kpage_idx[2];
-	uint32_t		*kpage[2];
 	uint32_t		*kdata;
 	void __user		*user_ptr;
-	int			last_copied_page;
-	int			last_page_index;
 };
 
 struct radeon_cs_parser {
@@ -1007,8 +1009,15 @@ struct radeon_cs_parser {
 	struct ww_acquire_ctx	ticket;
 };
 
-extern int radeon_cs_finish_pages(struct radeon_cs_parser *p);
-extern u32 radeon_get_ib_value(struct radeon_cs_parser *p, int idx);
+static inline u32 radeon_get_ib_value(struct radeon_cs_parser *p, int idx)
+{
+	struct radeon_cs_chunk *ibc = &p->chunks[p->chunk_ib_idx];
+
+	if (ibc->kdata)
+		return ibc->kdata[idx];
+	return p->ib.ptr[idx];
+}
+
 
 struct radeon_cs_packet {
 	unsigned	idx;
@@ -1675,8 +1684,6 @@ struct radeon_asic {
 	struct {
 		int (*init)(struct radeon_device *rdev);
 		void (*fini)(struct radeon_device *rdev);
-
-		u32 pt_ring_index;
 		void (*set_page)(struct radeon_device *rdev,
 				 struct radeon_ib *ib,
 				 uint64_t pe,
@@ -2170,6 +2177,7 @@ struct radeon_device {
 	bool				need_dma32;
 	bool				accel_working;
 	bool				fastfb_working; /* IGP feature*/
+	bool				needs_reset;
 	struct radeon_surface_reg surface_regs[RADEON_GEM_MAX_SURFACES];
 	const struct firmware *me_fw;	/* all family ME firmware */
 	const struct firmware *pfp_fw;	/* r6/700 PFP firmware */
@@ -2212,6 +2220,9 @@ struct radeon_device {
 	/* clock, powergating flags */
 	u32 cg_flags;
 	u32 pg_flags;
+
+	struct dev_pm_domain vga_pm_domain;
+	bool have_disp_power_ref;
 };
 
 int radeon_device_init(struct radeon_device *rdev,
@@ -2673,8 +2684,8 @@ extern void radeon_ttm_placement_from_domain(struct radeon_bo *rbo, u32 domain);
 extern bool radeon_ttm_bo_is_radeon_bo(struct ttm_buffer_object *bo);
 extern void radeon_vram_location(struct radeon_device *rdev, struct radeon_mc *mc, u64 base);
 extern void radeon_gtt_location(struct radeon_device *rdev, struct radeon_mc *mc);
-extern int radeon_resume_kms(struct drm_device *dev);
-extern int radeon_suspend_kms(struct drm_device *dev, pm_message_t state);
+extern int radeon_resume_kms(struct drm_device *dev, bool resume, bool fbcon);
+extern int radeon_suspend_kms(struct drm_device *dev, bool suspend, bool fbcon);
 extern void radeon_ttm_set_active_vram_size(struct radeon_device *rdev, u64 size);
 extern void radeon_program_register_sequence(struct radeon_device *rdev,
 					     const u32 *registers,
diff --git a/drivers/gpu/drm/radeon/radeon_asic.c b/drivers/gpu/drm/radeon/radeon_asic.c
index 8f7e04538fd6..50853c0cb49d 100644
--- a/drivers/gpu/drm/radeon/radeon_asic.c
+++ b/drivers/gpu/drm/radeon/radeon_asic.c
@@ -1622,8 +1622,7 @@ static struct radeon_asic cayman_asic = {
 	.vm = {
 		.init = &cayman_vm_init,
 		.fini = &cayman_vm_fini,
-		.pt_ring_index = R600_RING_TYPE_DMA_INDEX,
-		.set_page = &cayman_vm_set_page,
+		.set_page = &cayman_dma_vm_set_page,
 	},
 	.ring = {
 		[RADEON_RING_TYPE_GFX_INDEX] = &cayman_gfx_ring,
@@ -1723,8 +1722,7 @@ static struct radeon_asic trinity_asic = {
 	.vm = {
 		.init = &cayman_vm_init,
 		.fini = &cayman_vm_fini,
-		.pt_ring_index = R600_RING_TYPE_DMA_INDEX,
-		.set_page = &cayman_vm_set_page,
+		.set_page = &cayman_dma_vm_set_page,
 	},
 	.ring = {
 		[RADEON_RING_TYPE_GFX_INDEX] = &cayman_gfx_ring,
@@ -1854,8 +1852,7 @@ static struct radeon_asic si_asic = {
 	.vm = {
 		.init = &si_vm_init,
 		.fini = &si_vm_fini,
-		.pt_ring_index = R600_RING_TYPE_DMA_INDEX,
-		.set_page = &si_vm_set_page,
+		.set_page = &si_dma_vm_set_page,
 	},
 	.ring = {
 		[RADEON_RING_TYPE_GFX_INDEX] = &si_gfx_ring,
@@ -1879,7 +1876,7 @@ static struct radeon_asic si_asic = {
 		.hdmi_setmode = &evergreen_hdmi_setmode,
 	},
 	.copy = {
-		.blit = NULL,
+		.blit = &r600_copy_cpdma,
 		.blit_ring_index = RADEON_RING_TYPE_GFX_INDEX,
 		.dma = &si_copy_dma,
 		.dma_ring_index = R600_RING_TYPE_DMA_INDEX,
@@ -2000,8 +1997,7 @@ static struct radeon_asic ci_asic = {
 	.vm = {
 		.init = &cik_vm_init,
 		.fini = &cik_vm_fini,
-		.pt_ring_index = R600_RING_TYPE_DMA_INDEX,
-		.set_page = &cik_vm_set_page,
+		.set_page = &cik_sdma_vm_set_page,
 	},
 	.ring = {
 		[RADEON_RING_TYPE_GFX_INDEX] = &ci_gfx_ring,
@@ -2100,8 +2096,7 @@ static struct radeon_asic kv_asic = {
 	.vm = {
 		.init = &cik_vm_init,
 		.fini = &cik_vm_fini,
-		.pt_ring_index = R600_RING_TYPE_DMA_INDEX,
-		.set_page = &cik_vm_set_page,
+		.set_page = &cik_sdma_vm_set_page,
 	},
 	.ring = {
 		[RADEON_RING_TYPE_GFX_INDEX] = &ci_gfx_ring,
@@ -2442,27 +2437,48 @@ int radeon_asic_init(struct radeon_device *rdev)
 		}
 		break;
 	case CHIP_BONAIRE:
+	case CHIP_HAWAII:
 		rdev->asic = &ci_asic;
 		rdev->num_crtc = 6;
 		rdev->has_uvd = true;
-		rdev->cg_flags =
-			RADEON_CG_SUPPORT_GFX_MGCG |
-			RADEON_CG_SUPPORT_GFX_MGLS |
-			/*RADEON_CG_SUPPORT_GFX_CGCG |*/
-			RADEON_CG_SUPPORT_GFX_CGLS |
-			RADEON_CG_SUPPORT_GFX_CGTS |
-			RADEON_CG_SUPPORT_GFX_CGTS_LS |
-			RADEON_CG_SUPPORT_GFX_CP_LS |
-			RADEON_CG_SUPPORT_MC_LS |
-			RADEON_CG_SUPPORT_MC_MGCG |
-			RADEON_CG_SUPPORT_SDMA_MGCG |
-			RADEON_CG_SUPPORT_SDMA_LS |
-			RADEON_CG_SUPPORT_BIF_LS |
-			RADEON_CG_SUPPORT_VCE_MGCG |
-			RADEON_CG_SUPPORT_UVD_MGCG |
-			RADEON_CG_SUPPORT_HDP_LS |
-			RADEON_CG_SUPPORT_HDP_MGCG;
-		rdev->pg_flags = 0;
+		if (rdev->family == CHIP_BONAIRE) {
+			rdev->cg_flags =
+				RADEON_CG_SUPPORT_GFX_MGCG |
+				RADEON_CG_SUPPORT_GFX_MGLS |
+				/*RADEON_CG_SUPPORT_GFX_CGCG |*/
+				RADEON_CG_SUPPORT_GFX_CGLS |
+				RADEON_CG_SUPPORT_GFX_CGTS |
+				RADEON_CG_SUPPORT_GFX_CGTS_LS |
+				RADEON_CG_SUPPORT_GFX_CP_LS |
+				RADEON_CG_SUPPORT_MC_LS |
+				RADEON_CG_SUPPORT_MC_MGCG |
+				RADEON_CG_SUPPORT_SDMA_MGCG |
+				RADEON_CG_SUPPORT_SDMA_LS |
+				RADEON_CG_SUPPORT_BIF_LS |
+				RADEON_CG_SUPPORT_VCE_MGCG |
+				RADEON_CG_SUPPORT_UVD_MGCG |
+				RADEON_CG_SUPPORT_HDP_LS |
+				RADEON_CG_SUPPORT_HDP_MGCG;
+			rdev->pg_flags = 0;
+		} else {
+			rdev->cg_flags =
+				RADEON_CG_SUPPORT_GFX_MGCG |
+				RADEON_CG_SUPPORT_GFX_MGLS |
+				/*RADEON_CG_SUPPORT_GFX_CGCG |*/
+				RADEON_CG_SUPPORT_GFX_CGLS |
+				RADEON_CG_SUPPORT_GFX_CGTS |
+				RADEON_CG_SUPPORT_GFX_CP_LS |
+				RADEON_CG_SUPPORT_MC_LS |
+				RADEON_CG_SUPPORT_MC_MGCG |
+				RADEON_CG_SUPPORT_SDMA_MGCG |
+				RADEON_CG_SUPPORT_SDMA_LS |
+				RADEON_CG_SUPPORT_BIF_LS |
+				RADEON_CG_SUPPORT_VCE_MGCG |
+				RADEON_CG_SUPPORT_UVD_MGCG |
+				RADEON_CG_SUPPORT_HDP_LS |
+				RADEON_CG_SUPPORT_HDP_MGCG;
+			rdev->pg_flags = 0;
+		}
 		break;
 	case CHIP_KAVERI:
 	case CHIP_KABINI:
diff --git a/drivers/gpu/drm/radeon/radeon_asic.h b/drivers/gpu/drm/radeon/radeon_asic.h
index 70c29d5e080d..f2833ee3a613 100644
--- a/drivers/gpu/drm/radeon/radeon_asic.h
+++ b/drivers/gpu/drm/radeon/radeon_asic.h
@@ -581,17 +581,18 @@ int cayman_vm_init(struct radeon_device *rdev);
 void cayman_vm_fini(struct radeon_device *rdev);
 void cayman_vm_flush(struct radeon_device *rdev, int ridx, struct radeon_vm *vm);
 uint32_t cayman_vm_page_flags(struct radeon_device *rdev, uint32_t flags);
-void cayman_vm_set_page(struct radeon_device *rdev,
-			struct radeon_ib *ib,
-			uint64_t pe,
-			uint64_t addr, unsigned count,
-			uint32_t incr, uint32_t flags);
 int evergreen_ib_parse(struct radeon_device *rdev, struct radeon_ib *ib);
 int evergreen_dma_ib_parse(struct radeon_device *rdev, struct radeon_ib *ib);
 void cayman_dma_ring_ib_execute(struct radeon_device *rdev,
 				struct radeon_ib *ib);
 bool cayman_gfx_is_lockup(struct radeon_device *rdev, struct radeon_ring *ring);
 bool cayman_dma_is_lockup(struct radeon_device *rdev, struct radeon_ring *ring);
+void cayman_dma_vm_set_page(struct radeon_device *rdev,
+			    struct radeon_ib *ib,
+			    uint64_t pe,
+			    uint64_t addr, unsigned count,
+			    uint32_t incr, uint32_t flags);
+
 void cayman_dma_vm_flush(struct radeon_device *rdev, int ridx, struct radeon_vm *vm);
 
 int ni_dpm_init(struct radeon_device *rdev);
@@ -653,17 +654,17 @@ int si_irq_set(struct radeon_device *rdev);
 int si_irq_process(struct radeon_device *rdev);
 int si_vm_init(struct radeon_device *rdev);
 void si_vm_fini(struct radeon_device *rdev);
-void si_vm_set_page(struct radeon_device *rdev,
-		    struct radeon_ib *ib,
-		    uint64_t pe,
-		    uint64_t addr, unsigned count,
-		    uint32_t incr, uint32_t flags);
 void si_vm_flush(struct radeon_device *rdev, int ridx, struct radeon_vm *vm);
 int si_ib_parse(struct radeon_device *rdev, struct radeon_ib *ib);
 int si_copy_dma(struct radeon_device *rdev,
 		uint64_t src_offset, uint64_t dst_offset,
 		unsigned num_gpu_pages,
 		struct radeon_fence **fence);
+void si_dma_vm_set_page(struct radeon_device *rdev,
+			struct radeon_ib *ib,
+			uint64_t pe,
+			uint64_t addr, unsigned count,
+			uint32_t incr, uint32_t flags);
 void si_dma_vm_flush(struct radeon_device *rdev, int ridx, struct radeon_vm *vm);
 u32 si_get_xclk(struct radeon_device *rdev);
 uint64_t si_get_gpu_clock_counter(struct radeon_device *rdev);
@@ -705,6 +706,10 @@ int cik_copy_dma(struct radeon_device *rdev,
 		 uint64_t src_offset, uint64_t dst_offset,
 		 unsigned num_gpu_pages,
 		 struct radeon_fence **fence);
+int cik_copy_cpdma(struct radeon_device *rdev,
+		   uint64_t src_offset, uint64_t dst_offset,
+		   unsigned num_gpu_pages,
+		   struct radeon_fence **fence);
 int cik_sdma_ring_test(struct radeon_device *rdev, struct radeon_ring *ring);
 int cik_sdma_ib_test(struct radeon_device *rdev, struct radeon_ring *ring);
 bool cik_sdma_is_lockup(struct radeon_device *rdev, struct radeon_ring *ring);
@@ -731,11 +736,11 @@ int cik_irq_process(struct radeon_device *rdev);
 int cik_vm_init(struct radeon_device *rdev);
 void cik_vm_fini(struct radeon_device *rdev);
 void cik_vm_flush(struct radeon_device *rdev, int ridx, struct radeon_vm *vm);
-void cik_vm_set_page(struct radeon_device *rdev,
-		     struct radeon_ib *ib,
-		     uint64_t pe,
-		     uint64_t addr, unsigned count,
-		     uint32_t incr, uint32_t flags);
+void cik_sdma_vm_set_page(struct radeon_device *rdev,
+			  struct radeon_ib *ib,
+			  uint64_t pe,
+			  uint64_t addr, unsigned count,
+			  uint32_t incr, uint32_t flags);
 void cik_dma_vm_flush(struct radeon_device *rdev, int ridx, struct radeon_vm *vm);
 int cik_ib_parse(struct radeon_device *rdev, struct radeon_ib *ib);
 u32 cik_compute_ring_get_rptr(struct radeon_device *rdev,
diff --git a/drivers/gpu/drm/radeon/radeon_atpx_handler.c b/drivers/gpu/drm/radeon/radeon_atpx_handler.c
index 933bd946057b..9d302eaeea15 100644
--- a/drivers/gpu/drm/radeon/radeon_atpx_handler.c
+++ b/drivers/gpu/drm/radeon/radeon_atpx_handler.c
@@ -58,6 +58,10 @@ struct atpx_mux {
 	u16 mux;
 } __packed;
 
+bool radeon_is_px(void) {
+	return radeon_atpx_priv.atpx_detected;
+}
+
 /**
  * radeon_atpx_call - call an ATPX method
  *
diff --git a/drivers/gpu/drm/radeon/radeon_bios.c b/drivers/gpu/drm/radeon/radeon_bios.c
index 374eaba78a69..b3633d9a5317 100644
--- a/drivers/gpu/drm/radeon/radeon_bios.c
+++ b/drivers/gpu/drm/radeon/radeon_bios.c
@@ -499,7 +499,7 @@ static bool legacy_read_disabled_bios(struct radeon_device *rdev)
 	crtc_ext_cntl = RREG32(RADEON_CRTC_EXT_CNTL);
 	fp2_gen_cntl = 0;
 
-	if (rdev->ddev->pci_device == PCI_DEVICE_ID_ATI_RADEON_QY) {
+	if (rdev->ddev->pdev->device == PCI_DEVICE_ID_ATI_RADEON_QY) {
 		fp2_gen_cntl = RREG32(RADEON_FP2_GEN_CNTL);
 	}
 
@@ -536,7 +536,7 @@ static bool legacy_read_disabled_bios(struct radeon_device *rdev)
 		(RADEON_CRTC_SYNC_TRISTAT |
 		 RADEON_CRTC_DISPLAY_DIS)));
 
-	if (rdev->ddev->pci_device == PCI_DEVICE_ID_ATI_RADEON_QY) {
+	if (rdev->ddev->pdev->device == PCI_DEVICE_ID_ATI_RADEON_QY) {
 		WREG32(RADEON_FP2_GEN_CNTL, (fp2_gen_cntl & ~RADEON_FP2_ON));
 	}
 
@@ -554,7 +554,7 @@ static bool legacy_read_disabled_bios(struct radeon_device *rdev)
 		WREG32(RADEON_CRTC2_GEN_CNTL, crtc2_gen_cntl);
 	}
 	WREG32(RADEON_CRTC_EXT_CNTL, crtc_ext_cntl);
-	if (rdev->ddev->pci_device == PCI_DEVICE_ID_ATI_RADEON_QY) {
+	if (rdev->ddev->pdev->device == PCI_DEVICE_ID_ATI_RADEON_QY) {
 		WREG32(RADEON_FP2_GEN_CNTL, fp2_gen_cntl);
 	}
 	return r;
diff --git a/drivers/gpu/drm/radeon/radeon_connectors.c b/drivers/gpu/drm/radeon/radeon_connectors.c
index 64565732cb98..20a768ac89a8 100644
--- a/drivers/gpu/drm/radeon/radeon_connectors.c
+++ b/drivers/gpu/drm/radeon/radeon_connectors.c
@@ -31,6 +31,8 @@
 #include "radeon.h"
 #include "atom.h"
 
+#include <linux/pm_runtime.h>
+
 extern void
 radeon_combios_connected_scratch_regs(struct drm_connector *connector,
 				      struct drm_encoder *encoder,
@@ -411,6 +413,21 @@ static int radeon_connector_set_property(struct drm_connector *connector, struct
 		}
 	}
 
+	if (property == rdev->mode_info.dither_property) {
+		struct radeon_connector *radeon_connector = to_radeon_connector(connector);
+		/* need to find digital encoder on connector */
+		encoder = radeon_find_encoder(connector, DRM_MODE_ENCODER_TMDS);
+		if (!encoder)
+			return 0;
+
+		radeon_encoder = to_radeon_encoder(encoder);
+
+		if (radeon_connector->dither != val) {
+			radeon_connector->dither = val;
+			radeon_property_change_mode(&radeon_encoder->base);
+		}
+	}
+
 	if (property == rdev->mode_info.underscan_property) {
 		/* need to find digital encoder on connector */
 		encoder = radeon_find_encoder(connector, DRM_MODE_ENCODER_TMDS);
@@ -626,6 +643,11 @@ radeon_lvds_detect(struct drm_connector *connector, bool force)
 	struct radeon_connector *radeon_connector = to_radeon_connector(connector);
 	struct drm_encoder *encoder = radeon_best_single_encoder(connector);
 	enum drm_connector_status ret = connector_status_disconnected;
+	int r;
+
+	r = pm_runtime_get_sync(connector->dev->dev);
+	if (r < 0)
+		return connector_status_disconnected;
 
 	if (encoder) {
 		struct radeon_encoder *radeon_encoder = to_radeon_encoder(encoder);
@@ -651,6 +673,8 @@ radeon_lvds_detect(struct drm_connector *connector, bool force)
 	/* check acpi lid status ??? */
 
 	radeon_connector_update_scratch_regs(connector, ret);
+	pm_runtime_mark_last_busy(connector->dev->dev);
+	pm_runtime_put_autosuspend(connector->dev->dev);
 	return ret;
 }
 
@@ -750,6 +774,11 @@ radeon_vga_detect(struct drm_connector *connector, bool force)
 	struct drm_encoder_helper_funcs *encoder_funcs;
 	bool dret = false;
 	enum drm_connector_status ret = connector_status_disconnected;
+	int r;
+
+	r = pm_runtime_get_sync(connector->dev->dev);
+	if (r < 0)
+		return connector_status_disconnected;
 
 	encoder = radeon_best_single_encoder(connector);
 	if (!encoder)
@@ -790,9 +819,8 @@ radeon_vga_detect(struct drm_connector *connector, bool force)
 			 * detected a monitor via load.
 			 */
 			if (radeon_connector->detected_by_load)
-				return connector->status;
-			else
-				return ret;
+				ret = connector->status;
+			goto out;
 		}
 
 		if (radeon_connector->dac_load_detect && encoder) {
@@ -817,6 +845,11 @@ radeon_vga_detect(struct drm_connector *connector, bool force)
 	}
 
 	radeon_connector_update_scratch_regs(connector, ret);
+
+out:
+	pm_runtime_mark_last_busy(connector->dev->dev);
+	pm_runtime_put_autosuspend(connector->dev->dev);
+
 	return ret;
 }
 
@@ -873,10 +906,15 @@ radeon_tv_detect(struct drm_connector *connector, bool force)
 	struct drm_encoder_helper_funcs *encoder_funcs;
 	struct radeon_connector *radeon_connector = to_radeon_connector(connector);
 	enum drm_connector_status ret = connector_status_disconnected;
+	int r;
 
 	if (!radeon_connector->dac_load_detect)
 		return ret;
 
+	r = pm_runtime_get_sync(connector->dev->dev);
+	if (r < 0)
+		return connector_status_disconnected;
+
 	encoder = radeon_best_single_encoder(connector);
 	if (!encoder)
 		ret = connector_status_disconnected;
@@ -887,6 +925,8 @@ radeon_tv_detect(struct drm_connector *connector, bool force)
 	if (ret == connector_status_connected)
 		ret = radeon_connector_analog_encoder_conflict_solve(connector, encoder, ret, false);
 	radeon_connector_update_scratch_regs(connector, ret);
+	pm_runtime_mark_last_busy(connector->dev->dev);
+	pm_runtime_put_autosuspend(connector->dev->dev);
 	return ret;
 }
 
@@ -954,12 +994,18 @@ radeon_dvi_detect(struct drm_connector *connector, bool force)
 	struct drm_encoder *encoder = NULL;
 	struct drm_encoder_helper_funcs *encoder_funcs;
 	struct drm_mode_object *obj;
-	int i;
+	int i, r;
 	enum drm_connector_status ret = connector_status_disconnected;
 	bool dret = false, broken_edid = false;
 
-	if (!force && radeon_check_hpd_status_unchanged(connector))
-		return connector->status;
+	r = pm_runtime_get_sync(connector->dev->dev);
+	if (r < 0)
+		return connector_status_disconnected;
+
+	if (!force && radeon_check_hpd_status_unchanged(connector)) {
+		ret = connector->status;
+		goto exit;
+	}
 
 	if (radeon_connector->ddc_bus)
 		dret = radeon_ddc_probe(radeon_connector, false);
@@ -1110,6 +1156,11 @@ out:
 
 	/* updated in get modes as well since we need to know if it's analog or digital */
 	radeon_connector_update_scratch_regs(connector, ret);
+
+exit:
+	pm_runtime_mark_last_busy(connector->dev->dev);
+	pm_runtime_put_autosuspend(connector->dev->dev);
+
 	return ret;
 }
 
@@ -1377,9 +1428,16 @@ radeon_dp_detect(struct drm_connector *connector, bool force)
 	enum drm_connector_status ret = connector_status_disconnected;
 	struct radeon_connector_atom_dig *radeon_dig_connector = radeon_connector->con_priv;
 	struct drm_encoder *encoder = radeon_best_single_encoder(connector);
+	int r;
 
-	if (!force && radeon_check_hpd_status_unchanged(connector))
-		return connector->status;
+	r = pm_runtime_get_sync(connector->dev->dev);
+	if (r < 0)
+		return connector_status_disconnected;
+
+	if (!force && radeon_check_hpd_status_unchanged(connector)) {
+		ret = connector->status;
+		goto out;
+	}
 
 	if (radeon_connector->edid) {
 		kfree(radeon_connector->edid);
@@ -1443,6 +1501,10 @@ radeon_dp_detect(struct drm_connector *connector, bool force)
 	}
 
 	radeon_connector_update_scratch_regs(connector, ret);
+out:
+	pm_runtime_mark_last_busy(connector->dev->dev);
+	pm_runtime_put_autosuspend(connector->dev->dev);
+
 	return ret;
 }
 
@@ -1658,12 +1720,16 @@ radeon_add_atom_connector(struct drm_device *dev,
 			drm_object_attach_property(&radeon_connector->base.base,
 						      rdev->mode_info.underscan_vborder_property,
 						      0);
+
+			drm_object_attach_property(&radeon_connector->base.base,
+						   rdev->mode_info.dither_property,
+						   RADEON_FMT_DITHER_DISABLE);
+
 			if (radeon_audio != 0)
 				drm_object_attach_property(&radeon_connector->base.base,
 							   rdev->mode_info.audio_property,
-							   (radeon_audio == 1) ?
-							   RADEON_AUDIO_AUTO :
-							   RADEON_AUDIO_DISABLE);
+							   RADEON_AUDIO_AUTO);
+
 			subpixel_order = SubPixelHorizontalRGB;
 			connector->interlace_allowed = true;
 			if (connector_type == DRM_MODE_CONNECTOR_HDMIB)
@@ -1760,9 +1826,12 @@ radeon_add_atom_connector(struct drm_device *dev,
 			if (ASIC_IS_DCE2(rdev) && (radeon_audio != 0)) {
 				drm_object_attach_property(&radeon_connector->base.base,
 							   rdev->mode_info.audio_property,
-							   (radeon_audio == 1) ?
-							   RADEON_AUDIO_AUTO :
-							   RADEON_AUDIO_DISABLE);
+							   RADEON_AUDIO_AUTO);
+			}
+			if (ASIC_IS_AVIVO(rdev)) {
+				drm_object_attach_property(&radeon_connector->base.base,
+							   rdev->mode_info.dither_property,
+							   RADEON_FMT_DITHER_DISABLE);
 			}
 			if (connector_type == DRM_MODE_CONNECTOR_DVII) {
 				radeon_connector->dac_load_detect = true;
@@ -1807,9 +1876,12 @@ radeon_add_atom_connector(struct drm_device *dev,
 			if (ASIC_IS_DCE2(rdev) && (radeon_audio != 0)) {
 				drm_object_attach_property(&radeon_connector->base.base,
 							   rdev->mode_info.audio_property,
-							   (radeon_audio == 1) ?
-							   RADEON_AUDIO_AUTO :
-							   RADEON_AUDIO_DISABLE);
+							   RADEON_AUDIO_AUTO);
+			}
+			if (ASIC_IS_AVIVO(rdev)) {
+				drm_object_attach_property(&radeon_connector->base.base,
+							   rdev->mode_info.dither_property,
+							   RADEON_FMT_DITHER_DISABLE);
 			}
 			subpixel_order = SubPixelHorizontalRGB;
 			connector->interlace_allowed = true;
@@ -1853,9 +1925,13 @@ radeon_add_atom_connector(struct drm_device *dev,
 			if (ASIC_IS_DCE2(rdev) && (radeon_audio != 0)) {
 				drm_object_attach_property(&radeon_connector->base.base,
 							   rdev->mode_info.audio_property,
-							   (radeon_audio == 1) ?
-							   RADEON_AUDIO_AUTO :
-							   RADEON_AUDIO_DISABLE);
+							   RADEON_AUDIO_AUTO);
+			}
+			if (ASIC_IS_AVIVO(rdev)) {
+				drm_object_attach_property(&radeon_connector->base.base,
+							   rdev->mode_info.dither_property,
+							   RADEON_FMT_DITHER_DISABLE);
+
 			}
 			connector->interlace_allowed = true;
 			/* in theory with a DP to VGA converter... */
diff --git a/drivers/gpu/drm/radeon/radeon_cs.c b/drivers/gpu/drm/radeon/radeon_cs.c
index 80285e35bc65..26ca223d12d6 100644
--- a/drivers/gpu/drm/radeon/radeon_cs.c
+++ b/drivers/gpu/drm/radeon/radeon_cs.c
@@ -212,9 +212,7 @@ int radeon_cs_parser_init(struct radeon_cs_parser *p, void *data)
 			return -EFAULT;
 		}
 		p->chunks[i].length_dw = user_chunk.length_dw;
-		p->chunks[i].kdata = NULL;
 		p->chunks[i].chunk_id = user_chunk.chunk_id;
-		p->chunks[i].user_ptr = (void __user *)(unsigned long)user_chunk.chunk_data;
 		if (p->chunks[i].chunk_id == RADEON_CHUNK_ID_RELOCS) {
 			p->chunk_relocs_idx = i;
 		}
@@ -237,25 +235,31 @@ int radeon_cs_parser_init(struct radeon_cs_parser *p, void *data)
 				return -EINVAL;
 		}
 
-		cdata = (uint32_t *)(unsigned long)user_chunk.chunk_data;
-		if ((p->chunks[i].chunk_id == RADEON_CHUNK_ID_RELOCS) ||
-		    (p->chunks[i].chunk_id == RADEON_CHUNK_ID_FLAGS)) {
-			size = p->chunks[i].length_dw * sizeof(uint32_t);
-			p->chunks[i].kdata = kmalloc(size, GFP_KERNEL);
-			if (p->chunks[i].kdata == NULL) {
-				return -ENOMEM;
-			}
-			if (DRM_COPY_FROM_USER(p->chunks[i].kdata,
-					       p->chunks[i].user_ptr, size)) {
-				return -EFAULT;
-			}
-			if (p->chunks[i].chunk_id == RADEON_CHUNK_ID_FLAGS) {
-				p->cs_flags = p->chunks[i].kdata[0];
-				if (p->chunks[i].length_dw > 1)
-					ring = p->chunks[i].kdata[1];
-				if (p->chunks[i].length_dw > 2)
-					priority = (s32)p->chunks[i].kdata[2];
-			}
+		size = p->chunks[i].length_dw;
+		cdata = (void __user *)(unsigned long)user_chunk.chunk_data;
+		p->chunks[i].user_ptr = cdata;
+		if (p->chunks[i].chunk_id == RADEON_CHUNK_ID_CONST_IB)
+			continue;
+
+		if (p->chunks[i].chunk_id == RADEON_CHUNK_ID_IB) {
+			if (!p->rdev || !(p->rdev->flags & RADEON_IS_AGP))
+				continue;
+		}
+
+		p->chunks[i].kdata = drm_malloc_ab(size, sizeof(uint32_t));
+		size *= sizeof(uint32_t);
+		if (p->chunks[i].kdata == NULL) {
+			return -ENOMEM;
+		}
+		if (DRM_COPY_FROM_USER(p->chunks[i].kdata, cdata, size)) {
+			return -EFAULT;
+		}
+		if (p->chunks[i].chunk_id == RADEON_CHUNK_ID_FLAGS) {
+			p->cs_flags = p->chunks[i].kdata[0];
+			if (p->chunks[i].length_dw > 1)
+				ring = p->chunks[i].kdata[1];
+			if (p->chunks[i].length_dw > 2)
+				priority = (s32)p->chunks[i].kdata[2];
 		}
 	}
 
@@ -278,34 +282,6 @@ int radeon_cs_parser_init(struct radeon_cs_parser *p, void *data)
 		}
 	}
 
-	/* deal with non-vm */
-	if ((p->chunk_ib_idx != -1) &&
-	    ((p->cs_flags & RADEON_CS_USE_VM) == 0) &&
-	    (p->chunks[p->chunk_ib_idx].chunk_id == RADEON_CHUNK_ID_IB)) {
-		if (p->chunks[p->chunk_ib_idx].length_dw > (16 * 1024)) {
-			DRM_ERROR("cs IB too big: %d\n",
-				  p->chunks[p->chunk_ib_idx].length_dw);
-			return -EINVAL;
-		}
-		if (p->rdev && (p->rdev->flags & RADEON_IS_AGP)) {
-			p->chunks[p->chunk_ib_idx].kpage[0] = kmalloc(PAGE_SIZE, GFP_KERNEL);
-			p->chunks[p->chunk_ib_idx].kpage[1] = kmalloc(PAGE_SIZE, GFP_KERNEL);
-			if (p->chunks[p->chunk_ib_idx].kpage[0] == NULL ||
-			    p->chunks[p->chunk_ib_idx].kpage[1] == NULL) {
-				kfree(p->chunks[p->chunk_ib_idx].kpage[0]);
-				kfree(p->chunks[p->chunk_ib_idx].kpage[1]);
-				p->chunks[p->chunk_ib_idx].kpage[0] = NULL;
-				p->chunks[p->chunk_ib_idx].kpage[1] = NULL;
-				return -ENOMEM;
-			}
-		}
-		p->chunks[p->chunk_ib_idx].kpage_idx[0] = -1;
-		p->chunks[p->chunk_ib_idx].kpage_idx[1] = -1;
-		p->chunks[p->chunk_ib_idx].last_copied_page = -1;
-		p->chunks[p->chunk_ib_idx].last_page_index =
-			((p->chunks[p->chunk_ib_idx].length_dw * 4) - 1) / PAGE_SIZE;
-	}
-
 	return 0;
 }
 
@@ -339,13 +315,8 @@ static void radeon_cs_parser_fini(struct radeon_cs_parser *parser, int error, bo
 	kfree(parser->track);
 	kfree(parser->relocs);
 	kfree(parser->relocs_ptr);
-	for (i = 0; i < parser->nchunks; i++) {
-		kfree(parser->chunks[i].kdata);
-		if ((parser->rdev->flags & RADEON_IS_AGP)) {
-			kfree(parser->chunks[i].kpage[0]);
-			kfree(parser->chunks[i].kpage[1]);
-		}
-	}
+	for (i = 0; i < parser->nchunks; i++)
+		drm_free_large(parser->chunks[i].kdata);
 	kfree(parser->chunks);
 	kfree(parser->chunks_array);
 	radeon_ib_free(parser->rdev, &parser->ib);
@@ -355,7 +326,6 @@ static void radeon_cs_parser_fini(struct radeon_cs_parser *parser, int error, bo
 static int radeon_cs_ib_chunk(struct radeon_device *rdev,
 			      struct radeon_cs_parser *parser)
 {
-	struct radeon_cs_chunk *ib_chunk;
 	int r;
 
 	if (parser->chunk_ib_idx == -1)
@@ -364,28 +334,11 @@ static int radeon_cs_ib_chunk(struct radeon_device *rdev,
 	if (parser->cs_flags & RADEON_CS_USE_VM)
 		return 0;
 
-	ib_chunk = &parser->chunks[parser->chunk_ib_idx];
-	/* Copy the packet into the IB, the parser will read from the
-	 * input memory (cached) and write to the IB (which can be
-	 * uncached).
-	 */
-	r =  radeon_ib_get(rdev, parser->ring, &parser->ib,
-			   NULL, ib_chunk->length_dw * 4);
-	if (r) {
-		DRM_ERROR("Failed to get ib !\n");
-		return r;
-	}
-	parser->ib.length_dw = ib_chunk->length_dw;
 	r = radeon_cs_parse(rdev, parser->ring, parser);
 	if (r || parser->parser_error) {
 		DRM_ERROR("Invalid command stream !\n");
 		return r;
 	}
-	r = radeon_cs_finish_pages(parser);
-	if (r) {
-		DRM_ERROR("Invalid command stream !\n");
-		return r;
-	}
 
 	if (parser->ring == R600_RING_TYPE_UVD_INDEX)
 		radeon_uvd_note_usage(rdev);
@@ -423,7 +376,6 @@ static int radeon_bo_vm_update_pte(struct radeon_cs_parser *parser,
 static int radeon_cs_ib_vm_chunk(struct radeon_device *rdev,
 				 struct radeon_cs_parser *parser)
 {
-	struct radeon_cs_chunk *ib_chunk;
 	struct radeon_fpriv *fpriv = parser->filp->driver_priv;
 	struct radeon_vm *vm = &fpriv->vm;
 	int r;
@@ -433,49 +385,13 @@ static int radeon_cs_ib_vm_chunk(struct radeon_device *rdev,
 	if ((parser->cs_flags & RADEON_CS_USE_VM) == 0)
 		return 0;
 
-	if ((rdev->family >= CHIP_TAHITI) &&
-	    (parser->chunk_const_ib_idx != -1)) {
-		ib_chunk = &parser->chunks[parser->chunk_const_ib_idx];
-		if (ib_chunk->length_dw > RADEON_IB_VM_MAX_SIZE) {
-			DRM_ERROR("cs IB CONST too big: %d\n", ib_chunk->length_dw);
-			return -EINVAL;
-		}
-		r =  radeon_ib_get(rdev, parser->ring, &parser->const_ib,
-				   vm, ib_chunk->length_dw * 4);
-		if (r) {
-			DRM_ERROR("Failed to get const ib !\n");
-			return r;
-		}
-		parser->const_ib.is_const_ib = true;
-		parser->const_ib.length_dw = ib_chunk->length_dw;
-		/* Copy the packet into the IB */
-		if (DRM_COPY_FROM_USER(parser->const_ib.ptr, ib_chunk->user_ptr,
-				       ib_chunk->length_dw * 4)) {
-			return -EFAULT;
-		}
+	if (parser->const_ib.length_dw) {
 		r = radeon_ring_ib_parse(rdev, parser->ring, &parser->const_ib);
 		if (r) {
 			return r;
 		}
 	}
 
-	ib_chunk = &parser->chunks[parser->chunk_ib_idx];
-	if (ib_chunk->length_dw > RADEON_IB_VM_MAX_SIZE) {
-		DRM_ERROR("cs IB too big: %d\n", ib_chunk->length_dw);
-		return -EINVAL;
-	}
-	r =  radeon_ib_get(rdev, parser->ring, &parser->ib,
-			   vm, ib_chunk->length_dw * 4);
-	if (r) {
-		DRM_ERROR("Failed to get ib !\n");
-		return r;
-	}
-	parser->ib.length_dw = ib_chunk->length_dw;
-	/* Copy the packet into the IB */
-	if (DRM_COPY_FROM_USER(parser->ib.ptr, ib_chunk->user_ptr,
-			       ib_chunk->length_dw * 4)) {
-		return -EFAULT;
-	}
 	r = radeon_ring_ib_parse(rdev, parser->ring, &parser->ib);
 	if (r) {
 		return r;
@@ -527,6 +443,62 @@ static int radeon_cs_handle_lockup(struct radeon_device *rdev, int r)
 	return r;
 }
 
+static int radeon_cs_ib_fill(struct radeon_device *rdev, struct radeon_cs_parser *parser)
+{
+	struct radeon_cs_chunk *ib_chunk;
+	struct radeon_vm *vm = NULL;
+	int r;
+
+	if (parser->chunk_ib_idx == -1)
+		return 0;
+
+	if (parser->cs_flags & RADEON_CS_USE_VM) {
+		struct radeon_fpriv *fpriv = parser->filp->driver_priv;
+		vm = &fpriv->vm;
+
+		if ((rdev->family >= CHIP_TAHITI) &&
+		    (parser->chunk_const_ib_idx != -1)) {
+			ib_chunk = &parser->chunks[parser->chunk_const_ib_idx];
+			if (ib_chunk->length_dw > RADEON_IB_VM_MAX_SIZE) {
+				DRM_ERROR("cs IB CONST too big: %d\n", ib_chunk->length_dw);
+				return -EINVAL;
+			}
+			r =  radeon_ib_get(rdev, parser->ring, &parser->const_ib,
+					   vm, ib_chunk->length_dw * 4);
+			if (r) {
+				DRM_ERROR("Failed to get const ib !\n");
+				return r;
+			}
+			parser->const_ib.is_const_ib = true;
+			parser->const_ib.length_dw = ib_chunk->length_dw;
+			if (DRM_COPY_FROM_USER(parser->const_ib.ptr,
+					       ib_chunk->user_ptr,
+					       ib_chunk->length_dw * 4))
+				return -EFAULT;
+		}
+
+		ib_chunk = &parser->chunks[parser->chunk_ib_idx];
+		if (ib_chunk->length_dw > RADEON_IB_VM_MAX_SIZE) {
+			DRM_ERROR("cs IB too big: %d\n", ib_chunk->length_dw);
+			return -EINVAL;
+		}
+	}
+	ib_chunk = &parser->chunks[parser->chunk_ib_idx];
+
+	r =  radeon_ib_get(rdev, parser->ring, &parser->ib,
+			   vm, ib_chunk->length_dw * 4);
+	if (r) {
+		DRM_ERROR("Failed to get ib !\n");
+		return r;
+	}
+	parser->ib.length_dw = ib_chunk->length_dw;
+	if (ib_chunk->kdata)
+		memcpy(parser->ib.ptr, ib_chunk->kdata, ib_chunk->length_dw * 4);
+	else if (DRM_COPY_FROM_USER(parser->ib.ptr, ib_chunk->user_ptr, ib_chunk->length_dw * 4))
+		return -EFAULT;
+	return 0;
+}
+
 int radeon_cs_ioctl(struct drm_device *dev, void *data, struct drm_file *filp)
 {
 	struct radeon_device *rdev = dev->dev_private;
@@ -552,10 +524,15 @@ int radeon_cs_ioctl(struct drm_device *dev, void *data, struct drm_file *filp)
 		r = radeon_cs_handle_lockup(rdev, r);
 		return r;
 	}
-	r = radeon_cs_parser_relocs(&parser);
-	if (r) {
-		if (r != -ERESTARTSYS)
+
+	r = radeon_cs_ib_fill(rdev, &parser);
+	if (!r) {
+		r = radeon_cs_parser_relocs(&parser);
+		if (r && r != -ERESTARTSYS)
 			DRM_ERROR("Failed to parse relocation %d!\n", r);
+	}
+
+	if (r) {
 		radeon_cs_parser_fini(&parser, r, false);
 		up_read(&rdev->exclusive_lock);
 		r = radeon_cs_handle_lockup(rdev, r);
@@ -579,97 +556,6 @@ out:
 	return r;
 }
 
-int radeon_cs_finish_pages(struct radeon_cs_parser *p)
-{
-	struct radeon_cs_chunk *ibc = &p->chunks[p->chunk_ib_idx];
-	int i;
-	int size = PAGE_SIZE;
-
-	for (i = ibc->last_copied_page + 1; i <= ibc->last_page_index; i++) {
-		if (i == ibc->last_page_index) {
-			size = (ibc->length_dw * 4) % PAGE_SIZE;
-			if (size == 0)
-				size = PAGE_SIZE;
-		}
-		
-		if (DRM_COPY_FROM_USER(p->ib.ptr + (i * (PAGE_SIZE/4)),
-				       ibc->user_ptr + (i * PAGE_SIZE),
-				       size))
-			return -EFAULT;
-	}
-	return 0;
-}
-
-static int radeon_cs_update_pages(struct radeon_cs_parser *p, int pg_idx)
-{
-	int new_page;
-	struct radeon_cs_chunk *ibc = &p->chunks[p->chunk_ib_idx];
-	int i;
-	int size = PAGE_SIZE;
-	bool copy1 = (p->rdev && (p->rdev->flags & RADEON_IS_AGP)) ?
-		false : true;
-
-	for (i = ibc->last_copied_page + 1; i < pg_idx; i++) {
-		if (DRM_COPY_FROM_USER(p->ib.ptr + (i * (PAGE_SIZE/4)),
-				       ibc->user_ptr + (i * PAGE_SIZE),
-				       PAGE_SIZE)) {
-			p->parser_error = -EFAULT;
-			return 0;
-		}
-	}
-
-	if (pg_idx == ibc->last_page_index) {
-		size = (ibc->length_dw * 4) % PAGE_SIZE;
-		if (size == 0)
-			size = PAGE_SIZE;
-	}
-
-	new_page = ibc->kpage_idx[0] < ibc->kpage_idx[1] ? 0 : 1;
-	if (copy1)
-		ibc->kpage[new_page] = p->ib.ptr + (pg_idx * (PAGE_SIZE / 4));
-
-	if (DRM_COPY_FROM_USER(ibc->kpage[new_page],
-			       ibc->user_ptr + (pg_idx * PAGE_SIZE),
-			       size)) {
-		p->parser_error = -EFAULT;
-		return 0;
-	}
-
-	/* copy to IB for non single case */
-	if (!copy1)
-		memcpy((void *)(p->ib.ptr+(pg_idx*(PAGE_SIZE/4))), ibc->kpage[new_page], size);
-
-	ibc->last_copied_page = pg_idx;
-	ibc->kpage_idx[new_page] = pg_idx;
-
-	return new_page;
-}
-
-u32 radeon_get_ib_value(struct radeon_cs_parser *p, int idx)
-{
-	struct radeon_cs_chunk *ibc = &p->chunks[p->chunk_ib_idx];
-	u32 pg_idx, pg_offset;
-	u32 idx_value = 0;
-	int new_page;
-
-	pg_idx = (idx * 4) / PAGE_SIZE;
-	pg_offset = (idx * 4) % PAGE_SIZE;
-
-	if (ibc->kpage_idx[0] == pg_idx)
-		return ibc->kpage[0][pg_offset/4];
-	if (ibc->kpage_idx[1] == pg_idx)
-		return ibc->kpage[1][pg_offset/4];
-
-	new_page = radeon_cs_update_pages(p, pg_idx);
-	if (new_page < 0) {
-		p->parser_error = new_page;
-		return 0;
-	}
-
-	idx_value = ibc->kpage[new_page][pg_offset/4];
-	return idx_value;
-}
-
 /**
  * radeon_cs_packet_parse() - parse cp packet and point ib index to next packet
  * @parser:	parser structure holding parsing context.
diff --git a/drivers/gpu/drm/radeon/radeon_device.c b/drivers/gpu/drm/radeon/radeon_device.c
index 841d0e09be3e..b9234c43f43d 100644
--- a/drivers/gpu/drm/radeon/radeon_device.c
+++ b/drivers/gpu/drm/radeon/radeon_device.c
@@ -98,9 +98,16 @@ static const char radeon_family_name[][16] = {
 	"BONAIRE",
 	"KAVERI",
 	"KABINI",
+	"HAWAII",
 	"LAST",
 };
 
+#if defined(CONFIG_VGA_SWITCHEROO)
+bool radeon_is_px(void);
+#else
+static inline bool radeon_is_px(void) { return false; }
+#endif
+
 /**
  * radeon_program_register_sequence - program an array of registers.
  *
@@ -1076,7 +1083,10 @@ static bool radeon_switcheroo_quirk_long_wakeup(struct pci_dev *pdev)
 static void radeon_switcheroo_set_state(struct pci_dev *pdev, enum vga_switcheroo_state state)
 {
 	struct drm_device *dev = pci_get_drvdata(pdev);
-	pm_message_t pmm = { .event = PM_EVENT_SUSPEND };
+
+	if (radeon_is_px() && state == VGA_SWITCHEROO_OFF)
+		return;
+
 	if (state == VGA_SWITCHEROO_ON) {
 		unsigned d3_delay = dev->pdev->d3_delay;
 
@@ -1087,7 +1097,7 @@ static void radeon_switcheroo_set_state(struct pci_dev *pdev, enum vga_switchero
 		if (d3_delay < 20 && radeon_switcheroo_quirk_long_wakeup(pdev))
 			dev->pdev->d3_delay = 20;
 
-		radeon_resume_kms(dev);
+		radeon_resume_kms(dev, true, true);
 
 		dev->pdev->d3_delay = d3_delay;
 
@@ -1097,7 +1107,7 @@ static void radeon_switcheroo_set_state(struct pci_dev *pdev, enum vga_switchero
 		printk(KERN_INFO "radeon: switched off\n");
 		drm_kms_helper_poll_disable(dev);
 		dev->switch_power_state = DRM_SWITCH_POWER_CHANGING;
-		radeon_suspend_kms(dev, pmm);
+		radeon_suspend_kms(dev, true, true);
 		dev->switch_power_state = DRM_SWITCH_POWER_OFF;
 	}
 }
@@ -1147,6 +1157,7 @@ int radeon_device_init(struct radeon_device *rdev,
 {
 	int r, i;
 	int dma_bits;
+	bool runtime = false;
 
 	rdev->shutdown = false;
 	rdev->dev = &pdev->dev;
@@ -1293,7 +1304,14 @@ int radeon_device_init(struct radeon_device *rdev,
 	/* this will fail for cards that aren't VGA class devices, just
 	 * ignore it */
 	vga_client_register(rdev->pdev, rdev, NULL, radeon_vga_set_decode);
-	vga_switcheroo_register_client(rdev->pdev, &radeon_switcheroo_ops, false);
+
+	if (radeon_runtime_pm == 1)
+		runtime = true;
+	if ((radeon_runtime_pm == -1) && radeon_is_px())
+		runtime = true;
+	vga_switcheroo_register_client(rdev->pdev, &radeon_switcheroo_ops, runtime);
+	if (runtime)
+		vga_switcheroo_init_domain_pm_ops(rdev->dev, &rdev->vga_pm_domain);
 
 	r = radeon_init(rdev);
 	if (r)
@@ -1383,7 +1401,7 @@ void radeon_device_fini(struct radeon_device *rdev)
  * Returns 0 for success or an error on failure.
  * Called at driver suspend.
  */
-int radeon_suspend_kms(struct drm_device *dev, pm_message_t state)
+int radeon_suspend_kms(struct drm_device *dev, bool suspend, bool fbcon)
 {
 	struct radeon_device *rdev;
 	struct drm_crtc *crtc;
@@ -1394,9 +1412,7 @@ int radeon_suspend_kms(struct drm_device *dev, pm_message_t state)
 	if (dev == NULL || dev->dev_private == NULL) {
 		return -ENODEV;
 	}
-	if (state.event == PM_EVENT_PRETHAW) {
-		return 0;
-	}
+
 	rdev = dev->dev_private;
 
 	if (dev->switch_power_state == DRM_SWITCH_POWER_OFF)
@@ -1455,14 +1471,17 @@ int radeon_suspend_kms(struct drm_device *dev, pm_message_t state)
 	radeon_agp_suspend(rdev);
 
 	pci_save_state(dev->pdev);
-	if (state.event == PM_EVENT_SUSPEND) {
+	if (suspend) {
 		/* Shut down the device */
 		pci_disable_device(dev->pdev);
 		pci_set_power_state(dev->pdev, PCI_D3hot);
 	}
-	console_lock();
-	radeon_fbdev_set_suspend(rdev, 1);
-	console_unlock();
+
+	if (fbcon) {
+		console_lock();
+		radeon_fbdev_set_suspend(rdev, 1);
+		console_unlock();
+	}
 	return 0;
 }
 
@@ -1475,7 +1494,7 @@ int radeon_suspend_kms(struct drm_device *dev, pm_message_t state)
  * Returns 0 for success or an error on failure.
  * Called at driver resume.
  */
-int radeon_resume_kms(struct drm_device *dev)
+int radeon_resume_kms(struct drm_device *dev, bool resume, bool fbcon)
 {
 	struct drm_connector *connector;
 	struct radeon_device *rdev = dev->dev_private;
@@ -1484,12 +1503,17 @@ int radeon_resume_kms(struct drm_device *dev)
 	if (dev->switch_power_state == DRM_SWITCH_POWER_OFF)
 		return 0;
 
-	console_lock();
-	pci_set_power_state(dev->pdev, PCI_D0);
-	pci_restore_state(dev->pdev);
-	if (pci_enable_device(dev->pdev)) {
-		console_unlock();
-		return -1;
+	if (fbcon) {
+		console_lock();
+	}
+	if (resume) {
+		pci_set_power_state(dev->pdev, PCI_D0);
+		pci_restore_state(dev->pdev);
+		if (pci_enable_device(dev->pdev)) {
+			if (fbcon)
+				console_unlock();
+			return -1;
+		}
 	}
 	/* resume AGP if in use */
 	radeon_agp_resume(rdev);
@@ -1502,9 +1526,11 @@ int radeon_resume_kms(struct drm_device *dev)
 	radeon_pm_resume(rdev);
 	radeon_restore_bios_scratch_regs(rdev);
 
-	radeon_fbdev_set_suspend(rdev, 0);
-	console_unlock();
-
+	if (fbcon) {
+		radeon_fbdev_set_suspend(rdev, 0);
+		console_unlock();
+	}
+       
 	/* init dig PHYs, disp eng pll */
 	if (rdev->is_atom_bios) {
 		radeon_atom_encoder_init(rdev);
@@ -1549,6 +1575,14 @@ int radeon_gpu_reset(struct radeon_device *rdev)
 	int resched;
 
 	down_write(&rdev->exclusive_lock);
+
+	if (!rdev->needs_reset) {
+		up_write(&rdev->exclusive_lock);
+		return 0;
+	}
+
+	rdev->needs_reset = false;
+
 	radeon_save_bios_scratch_regs(rdev);
 	/* block TTM */
 	resched = ttm_bo_lock_delayed_workqueue(&rdev->mman.bdev);
diff --git a/drivers/gpu/drm/radeon/radeon_display.c b/drivers/gpu/drm/radeon/radeon_display.c
index 0d1aa050d41d..7b253815a323 100644
--- a/drivers/gpu/drm/radeon/radeon_display.c
+++ b/drivers/gpu/drm/radeon/radeon_display.c
@@ -30,6 +30,7 @@
 #include "atom.h"
 #include <asm/div64.h>
 
+#include <linux/pm_runtime.h>
 #include <drm/drm_crtc_helper.h>
 #include <drm/drm_edid.h>
 
@@ -306,7 +307,7 @@ void radeon_crtc_handle_flip(struct radeon_device *rdev, int crtc_id)
 	 */
 	if (update_pending &&
 	    (DRM_SCANOUTPOS_VALID & radeon_get_crtc_scanoutpos(rdev->ddev, crtc_id,
-							       &vpos, &hpos)) &&
+							       &vpos, &hpos, NULL, NULL)) &&
 	    ((vpos >= (99 * rdev->mode_info.crtcs[crtc_id]->base.hwmode.crtc_vdisplay)/100) ||
 	     (vpos < 0 && !ASIC_IS_AVIVO(rdev)))) {
 		/* crtc didn't flip in this target vblank interval,
@@ -494,11 +495,55 @@ unlock_free:
 	return r;
 }
 
+static int
+radeon_crtc_set_config(struct drm_mode_set *set)
+{
+	struct drm_device *dev;
+	struct radeon_device *rdev;
+	struct drm_crtc *crtc;
+	bool active = false;
+	int ret;
+
+	if (!set || !set->crtc)
+		return -EINVAL;
+
+	dev = set->crtc->dev;
+
+	ret = pm_runtime_get_sync(dev->dev);
+	if (ret < 0)
+		return ret;
+
+	ret = drm_crtc_helper_set_config(set);
+
+	list_for_each_entry(crtc, &dev->mode_config.crtc_list, head)
+		if (crtc->enabled)
+			active = true;
+
+	pm_runtime_mark_last_busy(dev->dev);
+
+	rdev = dev->dev_private;
+	/* if we have active crtcs and we don't have a power ref,
+	   take the current one */
+	if (active && !rdev->have_disp_power_ref) {
+		rdev->have_disp_power_ref = true;
+		return ret;
+	}
+	/* if we have no active crtcs, then drop the power ref
+	   we got before */
+	if (!active && rdev->have_disp_power_ref) {
+		pm_runtime_put_autosuspend(dev->dev);
+		rdev->have_disp_power_ref = false;
+	}
+
+	/* drop the power reference we got coming in here */
+	pm_runtime_put_autosuspend(dev->dev);
+	return ret;
+}
 static const struct drm_crtc_funcs radeon_crtc_funcs = {
 	.cursor_set = radeon_crtc_cursor_set,
 	.cursor_move = radeon_crtc_cursor_move,
 	.gamma_set = radeon_crtc_gamma_set,
-	.set_config = drm_crtc_helper_set_config,
+	.set_config = radeon_crtc_set_config,
 	.destroy = radeon_crtc_destroy,
 	.page_flip = radeon_crtc_page_flip,
 };
@@ -1178,6 +1223,12 @@ static struct drm_prop_enum_list radeon_audio_enum_list[] =
 	{ RADEON_AUDIO_AUTO, "auto" },
 };
 
+/* XXX support different dither options? spatial, temporal, both, etc. */
+static struct drm_prop_enum_list radeon_dither_enum_list[] =
+{	{ RADEON_FMT_DITHER_DISABLE, "off" },
+	{ RADEON_FMT_DITHER_ENABLE, "on" },
+};
+
 static int radeon_modeset_create_props(struct radeon_device *rdev)
 {
 	int sz;
@@ -1234,6 +1285,12 @@ static int radeon_modeset_create_props(struct radeon_device *rdev)
 					 "audio",
 					 radeon_audio_enum_list, sz);
 
+	sz = ARRAY_SIZE(radeon_dither_enum_list);
+	rdev->mode_info.dither_property =
+		drm_property_create_enum(rdev->ddev, 0,
+					 "dither",
+					 radeon_dither_enum_list, sz);
+
 	return 0;
 }
 
@@ -1539,12 +1596,17 @@ bool radeon_crtc_scaling_mode_fixup(struct drm_crtc *crtc,
 }
 
 /*
- * Retrieve current video scanout position of crtc on a given gpu.
+ * Retrieve current video scanout position of crtc on a given gpu, and
+ * an optional accurate timestamp of when query happened.
  *
  * \param dev Device to query.
  * \param crtc Crtc to query.
  * \param *vpos Location where vertical scanout position should be stored.
  * \param *hpos Location where horizontal scanout position should go.
+ * \param *stime Target location for timestamp taken immediately before
+ *               scanout position query. Can be NULL to skip timestamp.
+ * \param *etime Target location for timestamp taken immediately after
+ *               scanout position query. Can be NULL to skip timestamp.
  *
  * Returns vpos as a positive number while in active scanout area.
  * Returns vpos as a negative number inside vblank, counting the number
@@ -1560,7 +1622,8 @@ bool radeon_crtc_scaling_mode_fixup(struct drm_crtc *crtc,
  * unknown small number of scanlines wrt. real scanout position.
  *
  */
-int radeon_get_crtc_scanoutpos(struct drm_device *dev, int crtc, int *vpos, int *hpos)
+int radeon_get_crtc_scanoutpos(struct drm_device *dev, int crtc, int *vpos, int *hpos,
+			       ktime_t *stime, ktime_t *etime)
 {
 	u32 stat_crtc = 0, vbl = 0, position = 0;
 	int vbl_start, vbl_end, vtotal, ret = 0;
@@ -1568,6 +1631,12 @@ int radeon_get_crtc_scanoutpos(struct drm_device *dev, int crtc, int *vpos, int
 
 	struct radeon_device *rdev = dev->dev_private;
 
+	/* preempt_disable_rt() should go right here in PREEMPT_RT patchset. */
+
+	/* Get optional system timestamp before query. */
+	if (stime)
+		*stime = ktime_get();
+
 	if (ASIC_IS_DCE4(rdev)) {
 		if (crtc == 0) {
 			vbl = RREG32(EVERGREEN_CRTC_V_BLANK_START_END +
@@ -1650,6 +1719,12 @@ int radeon_get_crtc_scanoutpos(struct drm_device *dev, int crtc, int *vpos, int
 		}
 	}
 
+	/* Get optional system timestamp after query. */
+	if (etime)
+		*etime = ktime_get();
+
+	/* preempt_enable_rt() should go right here in PREEMPT_RT patchset. */
+
 	/* Decode into vertical and horizontal scanout position. */
 	*vpos = position & 0x1fff;
 	*hpos = (position >> 16) & 0x1fff;
diff --git a/drivers/gpu/drm/radeon/radeon_drv.c b/drivers/gpu/drm/radeon/radeon_drv.c
index 9c14a1ba1de4..1aee32213f66 100644
--- a/drivers/gpu/drm/radeon/radeon_drv.c
+++ b/drivers/gpu/drm/radeon/radeon_drv.c
@@ -36,8 +36,9 @@
 #include <drm/drm_pciids.h>
 #include <linux/console.h>
 #include <linux/module.h>
-
-
+#include <linux/pm_runtime.h>
+#include <linux/vga_switcheroo.h>
+#include "drm_crtc_helper.h"
 /*
  * KMS wrapper.
  * - 2.0.0 - initial interface
@@ -87,8 +88,8 @@ void radeon_driver_postclose_kms(struct drm_device *dev,
 				 struct drm_file *file_priv);
 void radeon_driver_preclose_kms(struct drm_device *dev,
 				struct drm_file *file_priv);
-int radeon_suspend_kms(struct drm_device *dev, pm_message_t state);
-int radeon_resume_kms(struct drm_device *dev);
+int radeon_suspend_kms(struct drm_device *dev, bool suspend, bool fbcon);
+int radeon_resume_kms(struct drm_device *dev, bool resume, bool fbcon);
 u32 radeon_get_vblank_counter_kms(struct drm_device *dev, int crtc);
 int radeon_enable_vblank_kms(struct drm_device *dev, int crtc);
 void radeon_disable_vblank_kms(struct drm_device *dev, int crtc);
@@ -100,14 +101,14 @@ void radeon_driver_irq_preinstall_kms(struct drm_device *dev);
 int radeon_driver_irq_postinstall_kms(struct drm_device *dev);
 void radeon_driver_irq_uninstall_kms(struct drm_device *dev);
 irqreturn_t radeon_driver_irq_handler_kms(DRM_IRQ_ARGS);
-int radeon_gem_object_init(struct drm_gem_object *obj);
 void radeon_gem_object_free(struct drm_gem_object *obj);
 int radeon_gem_object_open(struct drm_gem_object *obj,
 				struct drm_file *file_priv);
 void radeon_gem_object_close(struct drm_gem_object *obj,
 				struct drm_file *file_priv);
 extern int radeon_get_crtc_scanoutpos(struct drm_device *dev, int crtc,
-				      int *vpos, int *hpos);
+				      int *vpos, int *hpos, ktime_t *stime,
+				      ktime_t *etime);
 extern const struct drm_ioctl_desc radeon_ioctls_kms[];
 extern int radeon_max_kms_ioctl;
 int radeon_mmap(struct file *filp, struct vm_area_struct *vma);
@@ -137,9 +138,11 @@ void radeon_debugfs_cleanup(struct drm_minor *minor);
 #if defined(CONFIG_VGA_SWITCHEROO)
 void radeon_register_atpx_handler(void);
 void radeon_unregister_atpx_handler(void);
+bool radeon_is_px(void);
 #else
 static inline void radeon_register_atpx_handler(void) {}
 static inline void radeon_unregister_atpx_handler(void) {}
+static inline bool radeon_is_px(void) { return false; }
 #endif
 
 int radeon_no_wb;
@@ -162,6 +165,7 @@ int radeon_lockup_timeout = 10000;
 int radeon_fastfb = 0;
 int radeon_dpm = -1;
 int radeon_aspm = -1;
+int radeon_runtime_pm = -1;
 
 MODULE_PARM_DESC(no_wb, "Disable AGP writeback for scratch registers");
 module_param_named(no_wb, radeon_no_wb, int, 0444);
@@ -223,6 +227,9 @@ module_param_named(dpm, radeon_dpm, int, 0444);
 MODULE_PARM_DESC(aspm, "ASPM support (1 = enable, 0 = disable, -1 = auto)");
 module_param_named(aspm, radeon_aspm, int, 0444);
 
+MODULE_PARM_DESC(runpm, "PX runtime pm (1 = force enable, 0 = disable, -1 = PX only default)");
+module_param_named(runpm, radeon_runtime_pm, int, 0444);
+
 static struct pci_device_id pciidlist[] = {
 	radeon_PCI_IDS
 };
@@ -259,6 +266,7 @@ static int radeon_resume(struct drm_device *dev)
 	return 0;
 }
 
+
 static const struct file_operations radeon_driver_old_fops = {
 	.owner = THIS_MODULE,
 	.open = drm_open,
@@ -353,25 +361,144 @@ radeon_pci_remove(struct pci_dev *pdev)
 	drm_put_dev(dev);
 }
 
-static int
-radeon_pci_suspend(struct pci_dev *pdev, pm_message_t state)
+static int radeon_pmops_suspend(struct device *dev)
 {
-	struct drm_device *dev = pci_get_drvdata(pdev);
-	return radeon_suspend_kms(dev, state);
+	struct pci_dev *pdev = to_pci_dev(dev);
+	struct drm_device *drm_dev = pci_get_drvdata(pdev);
+	return radeon_suspend_kms(drm_dev, true, true);
 }
 
-static int
-radeon_pci_resume(struct pci_dev *pdev)
+static int radeon_pmops_resume(struct device *dev)
 {
-	struct drm_device *dev = pci_get_drvdata(pdev);
-	return radeon_resume_kms(dev);
+	struct pci_dev *pdev = to_pci_dev(dev);
+	struct drm_device *drm_dev = pci_get_drvdata(pdev);
+	return radeon_resume_kms(drm_dev, true, true);
+}
+
+static int radeon_pmops_freeze(struct device *dev)
+{
+	struct pci_dev *pdev = to_pci_dev(dev);
+	struct drm_device *drm_dev = pci_get_drvdata(pdev);
+	return radeon_suspend_kms(drm_dev, false, true);
+}
+
+static int radeon_pmops_thaw(struct device *dev)
+{
+	struct pci_dev *pdev = to_pci_dev(dev);
+	struct drm_device *drm_dev = pci_get_drvdata(pdev);
+	return radeon_resume_kms(drm_dev, false, true);
+}
+
+static int radeon_pmops_runtime_suspend(struct device *dev)
+{
+	struct pci_dev *pdev = to_pci_dev(dev);
+	struct drm_device *drm_dev = pci_get_drvdata(pdev);
+	int ret;
+
+	if (radeon_runtime_pm == 0)
+		return -EINVAL;
+
+	drm_dev->switch_power_state = DRM_SWITCH_POWER_CHANGING;
+	drm_kms_helper_poll_disable(drm_dev);
+	vga_switcheroo_set_dynamic_switch(pdev, VGA_SWITCHEROO_OFF);
+
+	ret = radeon_suspend_kms(drm_dev, false, false);
+	pci_save_state(pdev);
+	pci_disable_device(pdev);
+	pci_set_power_state(pdev, PCI_D3cold);
+	drm_dev->switch_power_state = DRM_SWITCH_POWER_DYNAMIC_OFF;
+
+	return 0;
+}
+
+static int radeon_pmops_runtime_resume(struct device *dev)
+{
+	struct pci_dev *pdev = to_pci_dev(dev);
+	struct drm_device *drm_dev = pci_get_drvdata(pdev);
+	int ret;
+
+	if (radeon_runtime_pm == 0)
+		return -EINVAL;
+
+	drm_dev->switch_power_state = DRM_SWITCH_POWER_CHANGING;
+
+	pci_set_power_state(pdev, PCI_D0);
+	pci_restore_state(pdev);
+	ret = pci_enable_device(pdev);
+	if (ret)
+		return ret;
+	pci_set_master(pdev);
+
+	ret = radeon_resume_kms(drm_dev, false, false);
+	drm_kms_helper_poll_enable(drm_dev);
+	vga_switcheroo_set_dynamic_switch(pdev, VGA_SWITCHEROO_ON);
+	drm_dev->switch_power_state = DRM_SWITCH_POWER_ON;
+	return 0;
 }
 
+static int radeon_pmops_runtime_idle(struct device *dev)
+{
+	struct pci_dev *pdev = to_pci_dev(dev);
+	struct drm_device *drm_dev = pci_get_drvdata(pdev);
+	struct drm_crtc *crtc;
+
+	if (radeon_runtime_pm == 0)
+		return -EBUSY;
+
+	/* are we PX enabled? */
+	if (radeon_runtime_pm == -1 && !radeon_is_px()) {
+		DRM_DEBUG_DRIVER("failing to power off - not px\n");
+		return -EBUSY;
+	}
+
+	list_for_each_entry(crtc, &drm_dev->mode_config.crtc_list, head) {
+		if (crtc->enabled) {
+			DRM_DEBUG_DRIVER("failing to power off - crtc active\n");
+			return -EBUSY;
+		}
+	}
+
+	pm_runtime_mark_last_busy(dev);
+	pm_runtime_autosuspend(dev);
+	/* we don't want the main rpm_idle to call suspend - we want to autosuspend */
+	return 1;
+}
+
+long radeon_drm_ioctl(struct file *filp,
+		      unsigned int cmd, unsigned long arg)
+{
+	struct drm_file *file_priv = filp->private_data;
+	struct drm_device *dev;
+	long ret;
+	dev = file_priv->minor->dev;
+	ret = pm_runtime_get_sync(dev->dev);
+	if (ret < 0)
+		return ret;
+
+	ret = drm_ioctl(filp, cmd, arg);
+	
+	pm_runtime_mark_last_busy(dev->dev);
+	pm_runtime_put_autosuspend(dev->dev);
+	return ret;
+}
+
+static const struct dev_pm_ops radeon_pm_ops = {
+	.suspend = radeon_pmops_suspend,
+	.resume = radeon_pmops_resume,
+	.freeze = radeon_pmops_freeze,
+	.thaw = radeon_pmops_thaw,
+	.poweroff = radeon_pmops_freeze,
+	.restore = radeon_pmops_resume,
+	.runtime_suspend = radeon_pmops_runtime_suspend,
+	.runtime_resume = radeon_pmops_runtime_resume,
+	.runtime_idle = radeon_pmops_runtime_idle,
+};
+
 static const struct file_operations radeon_driver_kms_fops = {
 	.owner = THIS_MODULE,
 	.open = drm_open,
 	.release = drm_release,
-	.unlocked_ioctl = drm_ioctl,
+	.unlocked_ioctl = radeon_drm_ioctl,
 	.mmap = radeon_mmap,
 	.poll = drm_poll,
 	.read = drm_read,
@@ -380,6 +507,15 @@ static const struct file_operations radeon_driver_kms_fops = {
 #endif
 };
 
+
+static void
+radeon_pci_shutdown(struct pci_dev *pdev)
+{
+	struct drm_device *dev = pci_get_drvdata(pdev);
+
+	radeon_driver_unload_kms(dev);
+}
+
 static struct drm_driver kms_driver = {
 	.driver_features =
 	    DRIVER_USE_AGP |
@@ -392,8 +528,6 @@ static struct drm_driver kms_driver = {
 	.postclose = radeon_driver_postclose_kms,
 	.lastclose = radeon_driver_lastclose_kms,
 	.unload = radeon_driver_unload_kms,
-	.suspend = radeon_suspend_kms,
-	.resume = radeon_resume_kms,
 	.get_vblank_counter = radeon_get_vblank_counter_kms,
 	.enable_vblank = radeon_enable_vblank_kms,
 	.disable_vblank = radeon_disable_vblank_kms,
@@ -408,7 +542,6 @@ static struct drm_driver kms_driver = {
 	.irq_uninstall = radeon_driver_irq_uninstall_kms,
 	.irq_handler = radeon_driver_irq_handler_kms,
 	.ioctls = radeon_ioctls_kms,
-	.gem_init_object = radeon_gem_object_init,
 	.gem_free_object = radeon_gem_object_free,
 	.gem_open_object = radeon_gem_object_open,
 	.gem_close_object = radeon_gem_object_close,
@@ -451,8 +584,8 @@ static struct pci_driver radeon_kms_pci_driver = {
 	.id_table = pciidlist,
 	.probe = radeon_pci_probe,
 	.remove = radeon_pci_remove,
-	.suspend = radeon_pci_suspend,
-	.resume = radeon_pci_resume,
+	.driver.pm = &radeon_pm_ops,
+	.shutdown = radeon_pci_shutdown,
 };
 
 static int __init radeon_init(void)
diff --git a/drivers/gpu/drm/radeon/radeon_drv.h b/drivers/gpu/drm/radeon/radeon_drv.h
index b369d42f7de5..543dcfae7e6f 100644
--- a/drivers/gpu/drm/radeon/radeon_drv.h
+++ b/drivers/gpu/drm/radeon/radeon_drv.h
@@ -113,6 +113,9 @@
 #define DRIVER_MINOR		33
 #define DRIVER_PATCHLEVEL	0
 
+long radeon_drm_ioctl(struct file *filp,
+		      unsigned int cmd, unsigned long arg);
+
 /* The rest of the file is DEPRECATED! */
 #ifdef CONFIG_DRM_RADEON_UMS
 
diff --git a/drivers/gpu/drm/radeon/radeon_family.h b/drivers/gpu/drm/radeon/radeon_family.h
index 3c8289083f9d..614ad549297f 100644
--- a/drivers/gpu/drm/radeon/radeon_family.h
+++ b/drivers/gpu/drm/radeon/radeon_family.h
@@ -96,6 +96,7 @@ enum radeon_family {
 	CHIP_BONAIRE,
 	CHIP_KAVERI,
 	CHIP_KABINI,
+	CHIP_HAWAII,
 	CHIP_LAST,
 };
 
diff --git a/drivers/gpu/drm/radeon/radeon_fence.c b/drivers/gpu/drm/radeon/radeon_fence.c
index ddb8f8e04eb5..281d14c22a47 100644
--- a/drivers/gpu/drm/radeon/radeon_fence.c
+++ b/drivers/gpu/drm/radeon/radeon_fence.c
@@ -190,10 +190,8 @@ void radeon_fence_process(struct radeon_device *rdev, int ring)
 		}
 	} while (atomic64_xchg(&rdev->fence_drv[ring].last_seq, seq) > seq);
 
-	if (wake) {
-		rdev->fence_drv[ring].last_activity = jiffies;
+	if (wake)
 		wake_up_all(&rdev->fence_queue);
-	}
 }
 
 /**
@@ -212,13 +210,13 @@ static void radeon_fence_destroy(struct kref *kref)
 }
 
 /**
- * radeon_fence_seq_signaled - check if a fence sequeuce number has signaled
+ * radeon_fence_seq_signaled - check if a fence sequence number has signaled
  *
  * @rdev: radeon device pointer
  * @seq: sequence number
  * @ring: ring index the fence is associated with
  *
- * Check if the last singled fence sequnce number is >= the requested
+ * Check if the last signaled fence sequnce number is >= the requested
  * sequence number (all asics).
  * Returns true if the fence has signaled (current fence value
  * is >= requested value) or false if it has not (current fence
@@ -263,113 +261,131 @@ bool radeon_fence_signaled(struct radeon_fence *fence)
 }
 
 /**
- * radeon_fence_wait_seq - wait for a specific sequence number
+ * radeon_fence_any_seq_signaled - check if any sequence number is signaled
  *
  * @rdev: radeon device pointer
- * @target_seq: sequence number we want to wait for
- * @ring: ring index the fence is associated with
+ * @seq: sequence numbers
+ *
+ * Check if the last signaled fence sequnce number is >= the requested
+ * sequence number (all asics).
+ * Returns true if any has signaled (current value is >= requested value)
+ * or false if it has not. Helper function for radeon_fence_wait_seq.
+ */
+static bool radeon_fence_any_seq_signaled(struct radeon_device *rdev, u64 *seq)
+{
+	unsigned i;
+
+	for (i = 0; i < RADEON_NUM_RINGS; ++i) {
+		if (seq[i] && radeon_fence_seq_signaled(rdev, seq[i], i))
+			return true;
+	}
+	return false;
+}
+
+/**
+ * radeon_fence_wait_seq - wait for a specific sequence numbers
+ *
+ * @rdev: radeon device pointer
+ * @target_seq: sequence number(s) we want to wait for
  * @intr: use interruptable sleep
  * @lock_ring: whether the ring should be locked or not
  *
- * Wait for the requested sequence number to be written (all asics).
+ * Wait for the requested sequence number(s) to be written by any ring
+ * (all asics).  Sequnce number array is indexed by ring id.
  * @intr selects whether to use interruptable (true) or non-interruptable
  * (false) sleep when waiting for the sequence number.  Helper function
- * for radeon_fence_wait(), et al.
+ * for radeon_fence_wait_*().
  * Returns 0 if the sequence number has passed, error for all other cases.
- * -EDEADLK is returned when a GPU lockup has been detected and the ring is
- * marked as not ready so no further jobs get scheduled until a successful
- * reset.
+ * -EDEADLK is returned when a GPU lockup has been detected.
  */
-static int radeon_fence_wait_seq(struct radeon_device *rdev, u64 target_seq,
-				 unsigned ring, bool intr, bool lock_ring)
+static int radeon_fence_wait_seq(struct radeon_device *rdev, u64 *target_seq,
+				 bool intr, bool lock_ring)
 {
-	unsigned long timeout, last_activity;
-	uint64_t seq;
-	unsigned i;
+	uint64_t last_seq[RADEON_NUM_RINGS];
 	bool signaled;
-	int r;
+	int i, r;
+
+	while (!radeon_fence_any_seq_signaled(rdev, target_seq)) {
+
+		/* Save current sequence values, used to check for GPU lockups */
+		for (i = 0; i < RADEON_NUM_RINGS; ++i) {
+			if (!target_seq[i])
+				continue;
 
-	while (target_seq > atomic64_read(&rdev->fence_drv[ring].last_seq)) {
-		if (!rdev->ring[ring].ready) {
-			return -EBUSY;
+			last_seq[i] = atomic64_read(&rdev->fence_drv[i].last_seq);
+			trace_radeon_fence_wait_begin(rdev->ddev, target_seq[i]);
+			radeon_irq_kms_sw_irq_get(rdev, i);
 		}
 
-		timeout = jiffies - RADEON_FENCE_JIFFIES_TIMEOUT;
-		if (time_after(rdev->fence_drv[ring].last_activity, timeout)) {
-			/* the normal case, timeout is somewhere before last_activity */
-			timeout = rdev->fence_drv[ring].last_activity - timeout;
+		if (intr) {
+			r = wait_event_interruptible_timeout(rdev->fence_queue, (
+				(signaled = radeon_fence_any_seq_signaled(rdev, target_seq))
+				 || rdev->needs_reset), RADEON_FENCE_JIFFIES_TIMEOUT);
 		} else {
-			/* either jiffies wrapped around, or no fence was signaled in the last 500ms
-			 * anyway we will just wait for the minimum amount and then check for a lockup
-			 */
-			timeout = 1;
+			r = wait_event_timeout(rdev->fence_queue, (
+				(signaled = radeon_fence_any_seq_signaled(rdev, target_seq))
+				 || rdev->needs_reset), RADEON_FENCE_JIFFIES_TIMEOUT);
 		}
-		seq = atomic64_read(&rdev->fence_drv[ring].last_seq);
-		/* Save current last activity valuee, used to check for GPU lockups */
-		last_activity = rdev->fence_drv[ring].last_activity;
 
-		trace_radeon_fence_wait_begin(rdev->ddev, seq);
-		radeon_irq_kms_sw_irq_get(rdev, ring);
-		if (intr) {
-			r = wait_event_interruptible_timeout(rdev->fence_queue,
-				(signaled = radeon_fence_seq_signaled(rdev, target_seq, ring)),
-				timeout);
-                } else {
-			r = wait_event_timeout(rdev->fence_queue,
-				(signaled = radeon_fence_seq_signaled(rdev, target_seq, ring)),
-				timeout);
+		for (i = 0; i < RADEON_NUM_RINGS; ++i) {
+			if (!target_seq[i])
+				continue;
+
+			radeon_irq_kms_sw_irq_put(rdev, i);
+			trace_radeon_fence_wait_end(rdev->ddev, target_seq[i]);
 		}
-		radeon_irq_kms_sw_irq_put(rdev, ring);
-		if (unlikely(r < 0)) {
+
+		if (unlikely(r < 0))
 			return r;
-		}
-		trace_radeon_fence_wait_end(rdev->ddev, seq);
 
 		if (unlikely(!signaled)) {
+			if (rdev->needs_reset)
+				return -EDEADLK;
+
 			/* we were interrupted for some reason and fence
 			 * isn't signaled yet, resume waiting */
-			if (r) {
+			if (r)
 				continue;
+
+			for (i = 0; i < RADEON_NUM_RINGS; ++i) {
+				if (!target_seq[i])
+					continue;
+
+				if (last_seq[i] != atomic64_read(&rdev->fence_drv[i].last_seq))
+					break;
 			}
 
-			/* check if sequence value has changed since last_activity */
-			if (seq != atomic64_read(&rdev->fence_drv[ring].last_seq)) {
+			if (i != RADEON_NUM_RINGS)
 				continue;
-			}
 
-			if (lock_ring) {
+			if (lock_ring)
 				mutex_lock(&rdev->ring_lock);
-			}
 
-			/* test if somebody else has already decided that this is a lockup */
-			if (last_activity != rdev->fence_drv[ring].last_activity) {
-				if (lock_ring) {
-					mutex_unlock(&rdev->ring_lock);
-				}
-				continue;
+			for (i = 0; i < RADEON_NUM_RINGS; ++i) {
+				if (!target_seq[i])
+					continue;
+
+				if (radeon_ring_is_lockup(rdev, i, &rdev->ring[i]))
+					break;
 			}
 
-			if (radeon_ring_is_lockup(rdev, ring, &rdev->ring[ring])) {
+			if (i < RADEON_NUM_RINGS) {
 				/* good news we believe it's a lockup */
-				dev_warn(rdev->dev, "GPU lockup (waiting for 0x%016llx last fence id 0x%016llx)\n",
-					 target_seq, seq);
-
-				/* change last activity so nobody else think there is a lockup */
-				for (i = 0; i < RADEON_NUM_RINGS; ++i) {
-					rdev->fence_drv[i].last_activity = jiffies;
-				}
-
-				/* mark the ring as not ready any more */
-				rdev->ring[ring].ready = false;
-				if (lock_ring) {
+				dev_warn(rdev->dev, "GPU lockup (waiting for "
+					 "0x%016llx last fence id 0x%016llx on"
+					 " ring %d)\n",
+					 target_seq[i], last_seq[i], i);
+
+				/* remember that we need an reset */
+				rdev->needs_reset = true;
+				if (lock_ring)
 					mutex_unlock(&rdev->ring_lock);
-				}
+				wake_up_all(&rdev->fence_queue);
 				return -EDEADLK;
 			}
 
-			if (lock_ring) {
+			if (lock_ring)
 				mutex_unlock(&rdev->ring_lock);
-			}
 		}
 	}
 	return 0;
@@ -388,6 +404,7 @@ static int radeon_fence_wait_seq(struct radeon_device *rdev, u64 target_seq,
  */
 int radeon_fence_wait(struct radeon_fence *fence, bool intr)
 {
+	uint64_t seq[RADEON_NUM_RINGS] = {};
 	int r;
 
 	if (fence == NULL) {
@@ -395,147 +412,15 @@ int radeon_fence_wait(struct radeon_fence *fence, bool intr)
 		return -EINVAL;
 	}
 
-	r = radeon_fence_wait_seq(fence->rdev, fence->seq,
-				  fence->ring, intr, true);
-	if (r) {
-		return r;
-	}
-	fence->seq = RADEON_FENCE_SIGNALED_SEQ;
-	return 0;
-}
-
-static bool radeon_fence_any_seq_signaled(struct radeon_device *rdev, u64 *seq)
-{
-	unsigned i;
-
-	for (i = 0; i < RADEON_NUM_RINGS; ++i) {
-		if (seq[i] && radeon_fence_seq_signaled(rdev, seq[i], i)) {
-			return true;
-		}
-	}
-	return false;
-}
+	seq[fence->ring] = fence->seq;
+	if (seq[fence->ring] == RADEON_FENCE_SIGNALED_SEQ)
+		return 0;
 
-/**
- * radeon_fence_wait_any_seq - wait for a sequence number on any ring
- *
- * @rdev: radeon device pointer
- * @target_seq: sequence number(s) we want to wait for
- * @intr: use interruptable sleep
- *
- * Wait for the requested sequence number(s) to be written by any ring
- * (all asics).  Sequnce number array is indexed by ring id.
- * @intr selects whether to use interruptable (true) or non-interruptable
- * (false) sleep when waiting for the sequence number.  Helper function
- * for radeon_fence_wait_any(), et al.
- * Returns 0 if the sequence number has passed, error for all other cases.
- */
-static int radeon_fence_wait_any_seq(struct radeon_device *rdev,
-				     u64 *target_seq, bool intr)
-{
-	unsigned long timeout, last_activity, tmp;
-	unsigned i, ring = RADEON_NUM_RINGS;
-	bool signaled;
-	int r;
-
-	for (i = 0, last_activity = 0; i < RADEON_NUM_RINGS; ++i) {
-		if (!target_seq[i]) {
-			continue;
-		}
-
-		/* use the most recent one as indicator */
-		if (time_after(rdev->fence_drv[i].last_activity, last_activity)) {
-			last_activity = rdev->fence_drv[i].last_activity;
-		}
-
-		/* For lockup detection just pick the lowest ring we are
-		 * actively waiting for
-		 */
-		if (i < ring) {
-			ring = i;
-		}
-	}
-
-	/* nothing to wait for ? */
-	if (ring == RADEON_NUM_RINGS) {
-		return -ENOENT;
-	}
-
-	while (!radeon_fence_any_seq_signaled(rdev, target_seq)) {
-		timeout = jiffies - RADEON_FENCE_JIFFIES_TIMEOUT;
-		if (time_after(last_activity, timeout)) {
-			/* the normal case, timeout is somewhere before last_activity */
-			timeout = last_activity - timeout;
-		} else {
-			/* either jiffies wrapped around, or no fence was signaled in the last 500ms
-			 * anyway we will just wait for the minimum amount and then check for a lockup
-			 */
-			timeout = 1;
-		}
-
-		trace_radeon_fence_wait_begin(rdev->ddev, target_seq[ring]);
-		for (i = 0; i < RADEON_NUM_RINGS; ++i) {
-			if (target_seq[i]) {
-				radeon_irq_kms_sw_irq_get(rdev, i);
-			}
-		}
-		if (intr) {
-			r = wait_event_interruptible_timeout(rdev->fence_queue,
-				(signaled = radeon_fence_any_seq_signaled(rdev, target_seq)),
-				timeout);
-		} else {
-			r = wait_event_timeout(rdev->fence_queue,
-				(signaled = radeon_fence_any_seq_signaled(rdev, target_seq)),
-				timeout);
-		}
-		for (i = 0; i < RADEON_NUM_RINGS; ++i) {
-			if (target_seq[i]) {
-				radeon_irq_kms_sw_irq_put(rdev, i);
-			}
-		}
-		if (unlikely(r < 0)) {
-			return r;
-		}
-		trace_radeon_fence_wait_end(rdev->ddev, target_seq[ring]);
-
-		if (unlikely(!signaled)) {
-			/* we were interrupted for some reason and fence
-			 * isn't signaled yet, resume waiting */
-			if (r) {
-				continue;
-			}
-
-			mutex_lock(&rdev->ring_lock);
-			for (i = 0, tmp = 0; i < RADEON_NUM_RINGS; ++i) {
-				if (time_after(rdev->fence_drv[i].last_activity, tmp)) {
-					tmp = rdev->fence_drv[i].last_activity;
-				}
-			}
-			/* test if somebody else has already decided that this is a lockup */
-			if (last_activity != tmp) {
-				last_activity = tmp;
-				mutex_unlock(&rdev->ring_lock);
-				continue;
-			}
-
-			if (radeon_ring_is_lockup(rdev, ring, &rdev->ring[ring])) {
-				/* good news we believe it's a lockup */
-				dev_warn(rdev->dev, "GPU lockup (waiting for 0x%016llx)\n",
-					 target_seq[ring]);
-
-				/* change last activity so nobody else think there is a lockup */
-				for (i = 0; i < RADEON_NUM_RINGS; ++i) {
-					rdev->fence_drv[i].last_activity = jiffies;
-				}
+	r = radeon_fence_wait_seq(fence->rdev, seq, intr, true);
+	if (r)
+		return r;
 
-				/* mark the ring as not ready any more */
-				rdev->ring[ring].ready = false;
-				mutex_unlock(&rdev->ring_lock);
-				return -EDEADLK;
-			}
-			mutex_unlock(&rdev->ring_lock);
-		}
-	}
+	fence->seq = RADEON_FENCE_SIGNALED_SEQ;
 	return 0;
 }
 
@@ -557,7 +442,7 @@ int radeon_fence_wait_any(struct radeon_device *rdev,
 			  bool intr)
 {
 	uint64_t seq[RADEON_NUM_RINGS];
-	unsigned i;
+	unsigned i, num_rings = 0;
 	int r;
 
 	for (i = 0; i < RADEON_NUM_RINGS; ++i) {
@@ -567,15 +452,19 @@ int radeon_fence_wait_any(struct radeon_device *rdev,
 			continue;
 		}
 
-		if (fences[i]->seq == RADEON_FENCE_SIGNALED_SEQ) {
-			/* something was allready signaled */
-			return 0;
-		}
-
 		seq[i] = fences[i]->seq;
+		++num_rings;
+
+		/* test if something was allready signaled */
+		if (seq[i] == RADEON_FENCE_SIGNALED_SEQ)
+			return 0;
 	}
 
-	r = radeon_fence_wait_any_seq(rdev, seq, intr);
+	/* nothing to wait for ? */
+	if (num_rings == 0)
+		return -ENOENT;
+
+	r = radeon_fence_wait_seq(rdev, seq, intr, true);
 	if (r) {
 		return r;
 	}
@@ -594,15 +483,15 @@ int radeon_fence_wait_any(struct radeon_device *rdev,
  */
 int radeon_fence_wait_next_locked(struct radeon_device *rdev, int ring)
 {
-	uint64_t seq;
+	uint64_t seq[RADEON_NUM_RINGS] = {};
 
-	seq = atomic64_read(&rdev->fence_drv[ring].last_seq) + 1ULL;
-	if (seq >= rdev->fence_drv[ring].sync_seq[ring]) {
+	seq[ring] = atomic64_read(&rdev->fence_drv[ring].last_seq) + 1ULL;
+	if (seq[ring] >= rdev->fence_drv[ring].sync_seq[ring]) {
 		/* nothing to wait for, last_seq is
 		   already the last emited fence */
 		return -ENOENT;
 	}
-	return radeon_fence_wait_seq(rdev, seq, ring, false, false);
+	return radeon_fence_wait_seq(rdev, seq, false, false);
 }
 
 /**
@@ -617,14 +506,18 @@ int radeon_fence_wait_next_locked(struct radeon_device *rdev, int ring)
  */
 int radeon_fence_wait_empty_locked(struct radeon_device *rdev, int ring)
 {
-	uint64_t seq = rdev->fence_drv[ring].sync_seq[ring];
+	uint64_t seq[RADEON_NUM_RINGS] = {};
 	int r;
 
-	r = radeon_fence_wait_seq(rdev, seq, ring, false, false);
+	seq[ring] = rdev->fence_drv[ring].sync_seq[ring];
+	if (!seq[ring])
+		return 0;
+
+	r = radeon_fence_wait_seq(rdev, seq, false, false);
 	if (r) {
-		if (r == -EDEADLK) {
+		if (r == -EDEADLK)
 			return -EDEADLK;
-		}
+
 		dev_err(rdev->dev, "error waiting for ring[%d] to become idle (%d)\n",
 			ring, r);
 	}
@@ -826,7 +719,6 @@ static void radeon_fence_driver_init_ring(struct radeon_device *rdev, int ring)
 	for (i = 0; i < RADEON_NUM_RINGS; ++i)
 		rdev->fence_drv[ring].sync_seq[i] = 0;
 	atomic64_set(&rdev->fence_drv[ring].last_seq, 0);
-	rdev->fence_drv[ring].last_activity = jiffies;
 	rdev->fence_drv[ring].initialized = false;
 }
 
diff --git a/drivers/gpu/drm/radeon/radeon_gart.c b/drivers/gpu/drm/radeon/radeon_gart.c
index b990b1a2bd50..8a83b89d4709 100644
--- a/drivers/gpu/drm/radeon/radeon_gart.c
+++ b/drivers/gpu/drm/radeon/radeon_gart.c
@@ -607,8 +607,8 @@ static int radeon_vm_evict(struct radeon_device *rdev, struct radeon_vm *vm)
  */
 int radeon_vm_alloc_pt(struct radeon_device *rdev, struct radeon_vm *vm)
 {
-	unsigned pd_size, pts_size;
-	u64 *pd_addr;
+	unsigned pd_size, pd_entries, pts_size;
+	struct radeon_ib ib;
 	int r;
 
 	if (vm == NULL) {
@@ -619,8 +619,10 @@ int radeon_vm_alloc_pt(struct radeon_device *rdev, struct radeon_vm *vm)
 		return 0;
 	}
 
-retry:
 	pd_size = radeon_vm_directory_size(rdev);
+	pd_entries = radeon_vm_num_pdes(rdev);
+
+retry:
 	r = radeon_sa_bo_new(rdev, &rdev->vm_manager.sa_manager,
 			     &vm->page_directory, pd_size,
 			     RADEON_VM_PTB_ALIGN_SIZE, false);
@@ -637,9 +639,31 @@ retry:
 	vm->pd_gpu_addr = radeon_sa_bo_gpu_addr(vm->page_directory);
 
 	/* Initially clear the page directory */
-	pd_addr = radeon_sa_bo_cpu_addr(vm->page_directory);
-	memset(pd_addr, 0, pd_size);
+	r = radeon_ib_get(rdev, R600_RING_TYPE_DMA_INDEX, &ib,
+			  NULL, pd_entries * 2 + 64);
+	if (r) {
+		radeon_sa_bo_free(rdev, &vm->page_directory, vm->fence);
+		return r;
+	}
+
+	ib.length_dw = 0;
+
+	radeon_asic_vm_set_page(rdev, &ib, vm->pd_gpu_addr,
+				0, pd_entries, 0, 0);
+
+	radeon_ib_sync_to(&ib, vm->fence);
+	r = radeon_ib_schedule(rdev, &ib, NULL);
+	if (r) {
+		radeon_ib_free(rdev, &ib);
+		radeon_sa_bo_free(rdev, &vm->page_directory, vm->fence);
+		return r;
+	}
+	radeon_fence_unref(&vm->fence);
+	vm->fence = radeon_fence_ref(ib.fence);
+	radeon_ib_free(rdev, &ib);
+	radeon_fence_unref(&vm->last_flush);
 
+	/* allocate page table array */
 	pts_size = radeon_vm_num_pdes(rdev) * sizeof(struct radeon_sa_bo *);
 	vm->page_tables = kzalloc(pts_size, GFP_KERNEL);
 
@@ -914,6 +938,26 @@ uint64_t radeon_vm_map_gart(struct radeon_device *rdev, uint64_t addr)
 }
 
 /**
+ * radeon_vm_page_flags - translate page flags to what the hw uses
+ *
+ * @flags: flags comming from userspace
+ *
+ * Translate the flags the userspace ABI uses to hw flags.
+ */
+static uint32_t radeon_vm_page_flags(uint32_t flags)
+{
+        uint32_t hw_flags = 0;
+        hw_flags |= (flags & RADEON_VM_PAGE_VALID) ? R600_PTE_VALID : 0;
+        hw_flags |= (flags & RADEON_VM_PAGE_READABLE) ? R600_PTE_READABLE : 0;
+        hw_flags |= (flags & RADEON_VM_PAGE_WRITEABLE) ? R600_PTE_WRITEABLE : 0;
+        if (flags & RADEON_VM_PAGE_SYSTEM) {
+                hw_flags |= R600_PTE_SYSTEM;
+                hw_flags |= (flags & RADEON_VM_PAGE_SNOOPED) ? R600_PTE_SNOOPED : 0;
+        }
+        return hw_flags;
+}
+
+/**
  * radeon_vm_update_pdes - make sure that page directory is valid
  *
  * @rdev: radeon_device pointer
@@ -974,7 +1018,11 @@ retry:
 			if (count) {
 				radeon_asic_vm_set_page(rdev, ib, last_pde,
 							last_pt, count, incr,
-							RADEON_VM_PAGE_VALID);
+							R600_PTE_VALID);
+
+				count *= RADEON_VM_PTE_COUNT;
+				radeon_asic_vm_set_page(rdev, ib, last_pt, 0,
+							count, 0, 0);
 			}
 
 			count = 1;
@@ -987,8 +1035,11 @@ retry:
 
 	if (count) {
 		radeon_asic_vm_set_page(rdev, ib, last_pde, last_pt, count,
-					incr, RADEON_VM_PAGE_VALID);
+					incr, R600_PTE_VALID);
 
+		count *= RADEON_VM_PTE_COUNT;
+		radeon_asic_vm_set_page(rdev, ib, last_pt, 0,
+					count, 0, 0);
 	}
 
 	return 0;
@@ -1082,7 +1133,6 @@ int radeon_vm_bo_update_pte(struct radeon_device *rdev,
 			    struct radeon_bo *bo,
 			    struct ttm_mem_reg *mem)
 {
-	unsigned ridx = rdev->asic->vm.pt_ring_index;
 	struct radeon_ib ib;
 	struct radeon_bo_va *bo_va;
 	unsigned nptes, npdes, ndw;
@@ -1151,11 +1201,14 @@ int radeon_vm_bo_update_pte(struct radeon_device *rdev,
 	/* reserve space for pde addresses */
 	ndw += npdes * 2;
 
+	/* reserve space for clearing new page tables */
+	ndw += npdes * 2 * RADEON_VM_PTE_COUNT;
+
 	/* update too big for an IB */
 	if (ndw > 0xfffff)
 		return -ENOMEM;
 
-	r = radeon_ib_get(rdev, ridx, &ib, NULL, ndw * 4);
+	r = radeon_ib_get(rdev, R600_RING_TYPE_DMA_INDEX, &ib, NULL, ndw * 4);
 	ib.length_dw = 0;
 
 	r = radeon_vm_update_pdes(rdev, vm, &ib, bo_va->soffset, bo_va->eoffset);
@@ -1165,7 +1218,7 @@ int radeon_vm_bo_update_pte(struct radeon_device *rdev,
 	}
 
 	radeon_vm_update_ptes(rdev, vm, &ib, bo_va->soffset, bo_va->eoffset,
-			      addr, bo_va->flags);
+			      addr, radeon_vm_page_flags(bo_va->flags));
 
 	radeon_ib_sync_to(&ib, vm->fence);
 	r = radeon_ib_schedule(rdev, &ib, NULL);
diff --git a/drivers/gpu/drm/radeon/radeon_gem.c b/drivers/gpu/drm/radeon/radeon_gem.c
index dce99c8a5835..805c5e566b9a 100644
--- a/drivers/gpu/drm/radeon/radeon_gem.c
+++ b/drivers/gpu/drm/radeon/radeon_gem.c
@@ -29,13 +29,6 @@
 #include <drm/radeon_drm.h>
 #include "radeon.h"
 
-int radeon_gem_object_init(struct drm_gem_object *obj)
-{
-	BUG();
-
-	return 0;
-}
-
 void radeon_gem_object_free(struct drm_gem_object *gobj)
 {
 	struct radeon_bo *robj = gem_to_radeon_bo(gobj);
diff --git a/drivers/gpu/drm/radeon/radeon_ioc32.c b/drivers/gpu/drm/radeon/radeon_ioc32.c
index c180df8e84db..bdb0f93e73bc 100644
--- a/drivers/gpu/drm/radeon/radeon_ioc32.c
+++ b/drivers/gpu/drm/radeon/radeon_ioc32.c
@@ -418,7 +418,7 @@ long radeon_kms_compat_ioctl(struct file *filp, unsigned int cmd, unsigned long
 	if (nr < DRM_COMMAND_BASE)
 		return drm_compat_ioctl(filp, cmd, arg);
 
-	ret = drm_ioctl(filp, cmd, arg);
+	ret = radeon_drm_ioctl(filp, cmd, arg);
 
 	return ret;
 }
diff --git a/drivers/gpu/drm/radeon/radeon_irq_kms.c b/drivers/gpu/drm/radeon/radeon_irq_kms.c
index cc9e8482cf30..ec6240b00469 100644
--- a/drivers/gpu/drm/radeon/radeon_irq_kms.c
+++ b/drivers/gpu/drm/radeon/radeon_irq_kms.c
@@ -32,6 +32,8 @@
 #include "radeon.h"
 #include "atom.h"
 
+#include <linux/pm_runtime.h>
+
 #define RADEON_WAIT_IDLE_TIMEOUT 200
 
 /**
@@ -47,8 +49,12 @@ irqreturn_t radeon_driver_irq_handler_kms(DRM_IRQ_ARGS)
 {
 	struct drm_device *dev = (struct drm_device *) arg;
 	struct radeon_device *rdev = dev->dev_private;
+	irqreturn_t ret;
 
-	return radeon_irq_process(rdev);
+	ret = radeon_irq_process(rdev);
+	if (ret == IRQ_HANDLED)
+		pm_runtime_mark_last_busy(dev->dev);
+	return ret;
 }
 
 /*
diff --git a/drivers/gpu/drm/radeon/radeon_kms.c b/drivers/gpu/drm/radeon/radeon_kms.c
index 61580ddc4eb2..bb8710531a1b 100644
--- a/drivers/gpu/drm/radeon/radeon_kms.c
+++ b/drivers/gpu/drm/radeon/radeon_kms.c
@@ -32,7 +32,7 @@
 
 #include <linux/vga_switcheroo.h>
 #include <linux/slab.h>
-
+#include <linux/pm_runtime.h>
 /**
  * radeon_driver_unload_kms - Main unload function for KMS.
  *
@@ -50,9 +50,14 @@ int radeon_driver_unload_kms(struct drm_device *dev)
 
 	if (rdev == NULL)
 		return 0;
+
 	if (rdev->rmmio == NULL)
 		goto done_free;
+
+	pm_runtime_get_sync(dev->dev);
+
 	radeon_acpi_fini(rdev);
+	
 	radeon_modeset_fini(rdev);
 	radeon_device_fini(rdev);
 
@@ -125,9 +130,20 @@ int radeon_driver_load_kms(struct drm_device *dev, unsigned long flags)
 				"Error during ACPI methods call\n");
 	}
 
+	if (radeon_runtime_pm != 0) {
+		pm_runtime_use_autosuspend(dev->dev);
+		pm_runtime_set_autosuspend_delay(dev->dev, 5000);
+		pm_runtime_set_active(dev->dev);
+		pm_runtime_allow(dev->dev);
+		pm_runtime_mark_last_busy(dev->dev);
+		pm_runtime_put_autosuspend(dev->dev);
+	}
+
 out:
 	if (r)
 		radeon_driver_unload_kms(dev);
+
+
 	return r;
 }
 
@@ -191,7 +207,7 @@ int radeon_info_ioctl(struct drm_device *dev, void *data, struct drm_file *filp)
 
 	switch (info->request) {
 	case RADEON_INFO_DEVICE_ID:
-		*value = dev->pci_device;
+		*value = dev->pdev->device;
 		break;
 	case RADEON_INFO_NUM_GB_PIPES:
 		*value = rdev->num_gb_pipes;
@@ -475,9 +491,14 @@ void radeon_driver_lastclose_kms(struct drm_device *dev)
 int radeon_driver_open_kms(struct drm_device *dev, struct drm_file *file_priv)
 {
 	struct radeon_device *rdev = dev->dev_private;
+	int r;
 
 	file_priv->driver_priv = NULL;
 
+	r = pm_runtime_get_sync(dev->dev);
+	if (r < 0)
+		return r;
+
 	/* new gpu have virtual address space support */
 	if (rdev->family >= CHIP_CAYMAN) {
 		struct radeon_fpriv *fpriv;
@@ -506,6 +527,9 @@ int radeon_driver_open_kms(struct drm_device *dev, struct drm_file *file_priv)
 
 		file_priv->driver_priv = fpriv;
 	}
+
+	pm_runtime_mark_last_busy(dev->dev);
+	pm_runtime_put_autosuspend(dev->dev);
 	return 0;
 }
 
diff --git a/drivers/gpu/drm/radeon/radeon_legacy_crtc.c b/drivers/gpu/drm/radeon/radeon_legacy_crtc.c
index 7cb178a34a0f..0c7b8c66301b 100644
--- a/drivers/gpu/drm/radeon/radeon_legacy_crtc.c
+++ b/drivers/gpu/drm/radeon/radeon_legacy_crtc.c
@@ -1056,6 +1056,26 @@ static void radeon_crtc_commit(struct drm_crtc *crtc)
 	}
 }
 
+static void radeon_crtc_disable(struct drm_crtc *crtc)
+{
+	radeon_crtc_dpms(crtc, DRM_MODE_DPMS_OFF);
+	if (crtc->fb) {
+		int r;
+		struct radeon_framebuffer *radeon_fb;
+		struct radeon_bo *rbo;
+
+		radeon_fb = to_radeon_framebuffer(crtc->fb);
+		rbo = gem_to_radeon_bo(radeon_fb->obj);
+		r = radeon_bo_reserve(rbo, false);
+		if (unlikely(r))
+			DRM_ERROR("failed to reserve rbo before unpin\n");
+		else {
+			radeon_bo_unpin(rbo);
+			radeon_bo_unreserve(rbo);
+		}
+	}
+}
+
 static const struct drm_crtc_helper_funcs legacy_helper_funcs = {
 	.dpms = radeon_crtc_dpms,
 	.mode_fixup = radeon_crtc_mode_fixup,
@@ -1065,6 +1085,7 @@ static const struct drm_crtc_helper_funcs legacy_helper_funcs = {
 	.prepare = radeon_crtc_prepare,
 	.commit = radeon_crtc_commit,
 	.load_lut = radeon_crtc_load_lut,
+	.disable = radeon_crtc_disable
 };
 
 
diff --git a/drivers/gpu/drm/radeon/radeon_legacy_encoders.c b/drivers/gpu/drm/radeon/radeon_legacy_encoders.c
index 62cd512f5c8d..c89971d904c3 100644
--- a/drivers/gpu/drm/radeon/radeon_legacy_encoders.c
+++ b/drivers/gpu/drm/radeon/radeon_legacy_encoders.c
@@ -392,7 +392,7 @@ void radeon_legacy_backlight_init(struct radeon_encoder *radeon_encoder,
 	props.type = BACKLIGHT_RAW;
 	snprintf(bl_name, sizeof(bl_name),
 		 "radeon_bl%d", dev->primary->index);
-	bd = backlight_device_register(bl_name, &drm_connector->kdev,
+	bd = backlight_device_register(bl_name, drm_connector->kdev,
 				       pdata, &radeon_backlight_ops, &props);
 	if (IS_ERR(bd)) {
 		DRM_ERROR("Backlight registration failed\n");
diff --git a/drivers/gpu/drm/radeon/radeon_mode.h b/drivers/gpu/drm/radeon/radeon_mode.h
index ef63d3f00b2f..3f0dd664af90 100644
--- a/drivers/gpu/drm/radeon/radeon_mode.h
+++ b/drivers/gpu/drm/radeon/radeon_mode.h
@@ -249,6 +249,8 @@ struct radeon_mode_info {
 	struct drm_property *underscan_vborder_property;
 	/* audio */
 	struct drm_property *audio_property;
+	/* FMT dithering */
+	struct drm_property *dither_property;
 	/* hardcoded DFP edid from BIOS */
 	struct edid *bios_hardcoded_edid;
 	int bios_hardcoded_edid_size;
@@ -479,6 +481,11 @@ enum radeon_connector_audio {
 	RADEON_AUDIO_AUTO = 2
 };
 
+enum radeon_connector_dither {
+	RADEON_FMT_DITHER_DISABLE = 0,
+	RADEON_FMT_DITHER_ENABLE = 1,
+};
+
 struct radeon_connector {
 	struct drm_connector base;
 	uint32_t connector_id;
@@ -498,6 +505,7 @@ struct radeon_connector {
 	struct radeon_router router;
 	struct radeon_i2c_chan *router_bus;
 	enum radeon_connector_audio audio;
+	enum radeon_connector_dither dither;
 };
 
 struct radeon_framebuffer {
@@ -758,7 +766,8 @@ extern int radeon_crtc_cursor_move(struct drm_crtc *crtc,
 				   int x, int y);
 
 extern int radeon_get_crtc_scanoutpos(struct drm_device *dev, int crtc,
-				      int *vpos, int *hpos);
+				      int *vpos, int *hpos, ktime_t *stime,
+				      ktime_t *etime);
 
 extern bool radeon_combios_check_hardcoded_edid(struct radeon_device *rdev);
 extern struct edid *
@@ -850,6 +859,12 @@ void radeon_legacy_tv_mode_set(struct drm_encoder *encoder,
 			       struct drm_display_mode *mode,
 			       struct drm_display_mode *adjusted_mode);
 
+/* fmt blocks */
+void avivo_program_fmt(struct drm_encoder *encoder);
+void dce3_program_fmt(struct drm_encoder *encoder);
+void dce4_program_fmt(struct drm_encoder *encoder);
+void dce8_program_fmt(struct drm_encoder *encoder);
+
 /* fbdev layer */
 int radeon_fbdev_init(struct radeon_device *rdev);
 void radeon_fbdev_fini(struct radeon_device *rdev);
diff --git a/drivers/gpu/drm/radeon/radeon_pm.c b/drivers/gpu/drm/radeon/radeon_pm.c
index 4f6b7fc7ad3c..866ace070b91 100644
--- a/drivers/gpu/drm/radeon/radeon_pm.c
+++ b/drivers/gpu/drm/radeon/radeon_pm.c
@@ -508,17 +508,21 @@ static ssize_t radeon_set_dpm_forced_performance_level(struct device *dev,
 	} else if (strncmp("auto", buf, strlen("auto")) == 0) {
 		level = RADEON_DPM_FORCED_LEVEL_AUTO;
 	} else {
-		mutex_unlock(&rdev->pm.mutex);
 		count = -EINVAL;
 		goto fail;
 	}
 	if (rdev->asic->dpm.force_performance_level) {
+		if (rdev->pm.dpm.thermal_active) {
+			count = -EINVAL;
+			goto fail;
+		}
 		ret = radeon_dpm_force_performance_level(rdev, level);
 		if (ret)
 			count = -EINVAL;
 	}
-	mutex_unlock(&rdev->pm.mutex);
 fail:
+	mutex_unlock(&rdev->pm.mutex);
+
 	return count;
 }
 
@@ -881,11 +885,12 @@ static void radeon_dpm_change_power_state_locked(struct radeon_device *rdev)
 		}
 	}
 
-	printk("switching from power state:\n");
-	radeon_dpm_print_power_state(rdev, rdev->pm.dpm.current_ps);
-	printk("switching to power state:\n");
-	radeon_dpm_print_power_state(rdev, rdev->pm.dpm.requested_ps);
-
+	if (radeon_dpm == 1) {
+		printk("switching from power state:\n");
+		radeon_dpm_print_power_state(rdev, rdev->pm.dpm.current_ps);
+		printk("switching to power state:\n");
+		radeon_dpm_print_power_state(rdev, rdev->pm.dpm.requested_ps);
+	}
 	mutex_lock(&rdev->ddev->struct_mutex);
 	down_write(&rdev->pm.mclk_lock);
 	mutex_lock(&rdev->ring_lock);
@@ -918,12 +923,16 @@ static void radeon_dpm_change_power_state_locked(struct radeon_device *rdev)
 	radeon_dpm_post_set_power_state(rdev);
 
 	if (rdev->asic->dpm.force_performance_level) {
-		if (rdev->pm.dpm.thermal_active)
+		if (rdev->pm.dpm.thermal_active) {
+			enum radeon_dpm_forced_level level = rdev->pm.dpm.forced_level;
 			/* force low perf level for thermal */
 			radeon_dpm_force_performance_level(rdev, RADEON_DPM_FORCED_LEVEL_LOW);
-		else
-			/* otherwise, enable auto */
-			radeon_dpm_force_performance_level(rdev, RADEON_DPM_FORCED_LEVEL_AUTO);
+			/* save the user's level */
+			rdev->pm.dpm.forced_level = level;
+		} else {
+			/* otherwise, user selected level */
+			radeon_dpm_force_performance_level(rdev, rdev->pm.dpm.forced_level);
+		}
 	}
 
 done:
@@ -1179,7 +1188,8 @@ static int radeon_pm_init_dpm(struct radeon_device *rdev)
 	mutex_lock(&rdev->pm.mutex);
 	radeon_dpm_init(rdev);
 	rdev->pm.dpm.current_ps = rdev->pm.dpm.requested_ps = rdev->pm.dpm.boot_ps;
-	radeon_dpm_print_power_states(rdev);
+	if (radeon_dpm == 1)
+		radeon_dpm_print_power_states(rdev);
 	radeon_dpm_setup_asic(rdev);
 	ret = radeon_dpm_enable(rdev);
 	mutex_unlock(&rdev->pm.mutex);
@@ -1241,6 +1251,24 @@ int radeon_pm_init(struct radeon_device *rdev)
 	case CHIP_RV670:
 	case CHIP_RS780:
 	case CHIP_RS880:
+	case CHIP_CAYMAN:
+	case CHIP_ARUBA:
+	case CHIP_BONAIRE:
+	case CHIP_KABINI:
+	case CHIP_KAVERI:
+	case CHIP_HAWAII:
+		/* DPM requires the RLC, RV770+ dGPU requires SMC */
+		if (!rdev->rlc_fw)
+			rdev->pm.pm_method = PM_METHOD_PROFILE;
+		else if ((rdev->family >= CHIP_RV770) &&
+			 (!(rdev->flags & RADEON_IS_IGP)) &&
+			 (!rdev->smc_fw))
+			rdev->pm.pm_method = PM_METHOD_PROFILE;
+		else if (radeon_dpm == 1)
+			rdev->pm.pm_method = PM_METHOD_DPM;
+		else
+			rdev->pm.pm_method = PM_METHOD_PROFILE;
+		break;
 	case CHIP_RV770:
 	case CHIP_RV730:
 	case CHIP_RV710:
@@ -1256,16 +1284,11 @@ int radeon_pm_init(struct radeon_device *rdev)
 	case CHIP_BARTS:
 	case CHIP_TURKS:
 	case CHIP_CAICOS:
-	case CHIP_CAYMAN:
-	case CHIP_ARUBA:
 	case CHIP_TAHITI:
 	case CHIP_PITCAIRN:
 	case CHIP_VERDE:
 	case CHIP_OLAND:
 	case CHIP_HAINAN:
-	case CHIP_BONAIRE:
-	case CHIP_KABINI:
-	case CHIP_KAVERI:
 		/* DPM requires the RLC, RV770+ dGPU requires SMC */
 		if (!rdev->rlc_fw)
 			rdev->pm.pm_method = PM_METHOD_PROFILE;
@@ -1273,10 +1296,10 @@ int radeon_pm_init(struct radeon_device *rdev)
 			 (!(rdev->flags & RADEON_IS_IGP)) &&
 			 (!rdev->smc_fw))
 			rdev->pm.pm_method = PM_METHOD_PROFILE;
-		else if (radeon_dpm == 1)
-			rdev->pm.pm_method = PM_METHOD_DPM;
-		else
+		else if (radeon_dpm == 0)
 			rdev->pm.pm_method = PM_METHOD_PROFILE;
+		else
+			rdev->pm.pm_method = PM_METHOD_DPM;
 		break;
 	default:
 		/* default to profile method */
@@ -1468,7 +1491,7 @@ static bool radeon_pm_in_vbl(struct radeon_device *rdev)
 	 */
 	for (crtc = 0; (crtc < rdev->num_crtc) && in_vbl; crtc++) {
 		if (rdev->pm.active_crtcs & (1 << crtc)) {
-			vbl_status = radeon_get_crtc_scanoutpos(rdev->ddev, crtc, &vpos, &hpos);
+			vbl_status = radeon_get_crtc_scanoutpos(rdev->ddev, crtc, &vpos, &hpos, NULL, NULL);
 			if ((vbl_status & DRM_SCANOUTPOS_VALID) &&
 			    !(vbl_status & DRM_SCANOUTPOS_INVBL))
 				in_vbl = false;
diff --git a/drivers/gpu/drm/radeon/radeon_trace.h b/drivers/gpu/drm/radeon/radeon_trace.h
index f7e367815964..811bca691b36 100644
--- a/drivers/gpu/drm/radeon/radeon_trace.h
+++ b/drivers/gpu/drm/radeon/radeon_trace.h
@@ -47,6 +47,30 @@ TRACE_EVENT(radeon_cs,
 		      __entry->fences)
 );
 
+TRACE_EVENT(radeon_vm_set_page,
+	    TP_PROTO(uint64_t pe, uint64_t addr, unsigned count,
+		     uint32_t incr, uint32_t flags),
+	    TP_ARGS(pe, addr, count, incr, flags),
+	    TP_STRUCT__entry(
+			     __field(u64, pe)
+			     __field(u64, addr)
+			     __field(u32, count)
+			     __field(u32, incr)
+			     __field(u32, flags)
+			     ),
+
+	    TP_fast_assign(
+			   __entry->pe = pe;
+			   __entry->addr = addr;
+			   __entry->count = count;
+			   __entry->incr = incr;
+			   __entry->flags = flags;
+			   ),
+	    TP_printk("pe=%010Lx, addr=%010Lx, incr=%u, flags=%08x, count=%u",
+		      __entry->pe, __entry->addr, __entry->incr,
+		      __entry->flags, __entry->count)
+);
+
 DECLARE_EVENT_CLASS(radeon_fence_request,
 
 	    TP_PROTO(struct drm_device *dev, u32 seqno),
diff --git a/drivers/gpu/drm/radeon/radeon_ucode.h b/drivers/gpu/drm/radeon/radeon_ucode.h
index 33858364fe89..a77cd274dfc3 100644
--- a/drivers/gpu/drm/radeon/radeon_ucode.h
+++ b/drivers/gpu/drm/radeon/radeon_ucode.h
@@ -59,6 +59,7 @@
 #define SI_MC_UCODE_SIZE             7769
 #define OLAND_MC_UCODE_SIZE          7863
 #define CIK_MC_UCODE_SIZE            7866
+#define HAWAII_MC_UCODE_SIZE         7933
 
 /* SDMA */
 #define CIK_SDMA_UCODE_SIZE          1050
@@ -143,4 +144,7 @@
 #define BONAIRE_SMC_UCODE_START      0x20000
 #define BONAIRE_SMC_UCODE_SIZE       0x1FDEC
 
+#define HAWAII_SMC_UCODE_START       0x20000
+#define HAWAII_SMC_UCODE_SIZE        0x1FDEC
+
 #endif
diff --git a/drivers/gpu/drm/radeon/radeon_uvd.c b/drivers/gpu/drm/radeon/radeon_uvd.c
index 308eff5be1b4..373d088bac66 100644
--- a/drivers/gpu/drm/radeon/radeon_uvd.c
+++ b/drivers/gpu/drm/radeon/radeon_uvd.c
@@ -97,6 +97,7 @@ int radeon_uvd_init(struct radeon_device *rdev)
 	case CHIP_BONAIRE:
 	case CHIP_KABINI:
 	case CHIP_KAVERI:
+	case CHIP_HAWAII:
 		fw_name = FIRMWARE_BONAIRE;
 		break;
 
@@ -240,6 +241,8 @@ void radeon_uvd_free_handles(struct radeon_device *rdev, struct drm_file *filp)
 		if (handle != 0 && rdev->uvd.filp[i] == filp) {
 			struct radeon_fence *fence;
 
+			radeon_uvd_note_usage(rdev);
+
 			r = radeon_uvd_get_destroy_msg(rdev,
 				R600_RING_TYPE_UVD_INDEX, handle, &fence);
 			if (r) {
@@ -620,7 +623,7 @@ static int radeon_uvd_send_msg(struct radeon_device *rdev,
 	if (r) 
 		goto err;
 
-	r = radeon_ib_get(rdev, ring, &ib, NULL, 16);
+	r = radeon_ib_get(rdev, ring, &ib, NULL, 64);
 	if (r)
 		goto err;
 
diff --git a/drivers/gpu/drm/radeon/rs600.c b/drivers/gpu/drm/radeon/rs600.c
index 6acba8017b9a..76cc8d3aafec 100644
--- a/drivers/gpu/drm/radeon/rs600.c
+++ b/drivers/gpu/drm/radeon/rs600.c
@@ -153,6 +153,70 @@ u32 rs600_page_flip(struct radeon_device *rdev, int crtc_id, u64 crtc_base)
 	return RREG32(AVIVO_D1GRPH_UPDATE + radeon_crtc->crtc_offset) & AVIVO_D1GRPH_SURFACE_UPDATE_PENDING;
 }
 
+void avivo_program_fmt(struct drm_encoder *encoder)
+{
+	struct drm_device *dev = encoder->dev;
+	struct radeon_device *rdev = dev->dev_private;
+	struct radeon_encoder *radeon_encoder = to_radeon_encoder(encoder);
+	struct drm_connector *connector = radeon_get_connector_for_encoder(encoder);
+	int bpc = 0;
+	u32 tmp = 0;
+	enum radeon_connector_dither dither = RADEON_FMT_DITHER_DISABLE;
+
+	if (connector) {
+		struct radeon_connector *radeon_connector = to_radeon_connector(connector);
+		bpc = radeon_get_monitor_bpc(connector);
+		dither = radeon_connector->dither;
+	}
+
+	/* LVDS FMT is set up by atom */
+	if (radeon_encoder->devices & ATOM_DEVICE_LCD_SUPPORT)
+		return;
+
+	if (bpc == 0)
+		return;
+
+	switch (bpc) {
+	case 6:
+		if (dither == RADEON_FMT_DITHER_ENABLE)
+			/* XXX sort out optimal dither settings */
+			tmp |= AVIVO_TMDS_BIT_DEPTH_CONTROL_SPATIAL_DITHER_EN;
+		else
+			tmp |= AVIVO_TMDS_BIT_DEPTH_CONTROL_TRUNCATE_EN;
+		break;
+	case 8:
+		if (dither == RADEON_FMT_DITHER_ENABLE)
+			/* XXX sort out optimal dither settings */
+			tmp |= (AVIVO_TMDS_BIT_DEPTH_CONTROL_SPATIAL_DITHER_EN |
+				AVIVO_TMDS_BIT_DEPTH_CONTROL_SPATIAL_DITHER_DEPTH);
+		else
+			tmp |= (AVIVO_TMDS_BIT_DEPTH_CONTROL_TRUNCATE_EN |
+				AVIVO_TMDS_BIT_DEPTH_CONTROL_TRUNCATE_DEPTH);
+		break;
+	case 10:
+	default:
+		/* not needed */
+		break;
+	}
+
+	switch (radeon_encoder->encoder_id) {
+	case ENCODER_OBJECT_ID_INTERNAL_KLDSCP_TMDS1:
+		WREG32(AVIVO_TMDSA_BIT_DEPTH_CONTROL, tmp);
+		break;
+	case ENCODER_OBJECT_ID_INTERNAL_LVTM1:
+		WREG32(AVIVO_LVTMA_BIT_DEPTH_CONTROL, tmp);
+		break;
+	case ENCODER_OBJECT_ID_INTERNAL_KLDSCP_DVO1:
+		WREG32(AVIVO_DVOA_BIT_DEPTH_CONTROL, tmp);
+		break;
+	case ENCODER_OBJECT_ID_INTERNAL_DDI:
+		WREG32(AVIVO_DDIA_BIT_DEPTH_CONTROL, tmp);
+		break;
+	default:
+		break;
+	}
+}
+
 void rs600_pm_misc(struct radeon_device *rdev)
 {
 	int requested_index = rdev->pm.requested_power_state_index;
diff --git a/drivers/gpu/drm/radeon/rs690.c b/drivers/gpu/drm/radeon/rs690.c
index 1447d794c22a..1c560629575a 100644
--- a/drivers/gpu/drm/radeon/rs690.c
+++ b/drivers/gpu/drm/radeon/rs690.c
@@ -345,9 +345,11 @@ static void rs690_crtc_bandwidth_compute(struct radeon_device *rdev,
 		if (max_bandwidth.full > rdev->pm.sideport_bandwidth.full &&
 			rdev->pm.sideport_bandwidth.full)
 			max_bandwidth = rdev->pm.sideport_bandwidth;
-		read_delay_latency.full = dfixed_const(370 * 800 * 1000);
-		read_delay_latency.full = dfixed_div(read_delay_latency,
-			rdev->pm.igp_sideport_mclk);
+		read_delay_latency.full = dfixed_const(370 * 800);
+		a.full = dfixed_const(1000);
+		b.full = dfixed_div(rdev->pm.igp_sideport_mclk, a);
+		read_delay_latency.full = dfixed_div(read_delay_latency, b);
+		read_delay_latency.full = dfixed_mul(read_delay_latency, a);
 	} else {
 		if (max_bandwidth.full > rdev->pm.k8_bandwidth.full &&
 			rdev->pm.k8_bandwidth.full)
@@ -488,14 +490,10 @@ static void rs690_compute_mode_priority(struct radeon_device *rdev,
 		}
 		if (wm0->priority_mark.full > priority_mark02.full)
 			priority_mark02.full = wm0->priority_mark.full;
-		if (dfixed_trunc(priority_mark02) < 0)
-			priority_mark02.full = 0;
 		if (wm0->priority_mark_max.full > priority_mark02.full)
 			priority_mark02.full = wm0->priority_mark_max.full;
 		if (wm1->priority_mark.full > priority_mark12.full)
 			priority_mark12.full = wm1->priority_mark.full;
-		if (dfixed_trunc(priority_mark12) < 0)
-			priority_mark12.full = 0;
 		if (wm1->priority_mark_max.full > priority_mark12.full)
 			priority_mark12.full = wm1->priority_mark_max.full;
 		*d1mode_priority_a_cnt = dfixed_trunc(priority_mark02);
@@ -526,8 +524,6 @@ static void rs690_compute_mode_priority(struct radeon_device *rdev,
 		}
 		if (wm0->priority_mark.full > priority_mark02.full)
 			priority_mark02.full = wm0->priority_mark.full;
-		if (dfixed_trunc(priority_mark02) < 0)
-			priority_mark02.full = 0;
 		if (wm0->priority_mark_max.full > priority_mark02.full)
 			priority_mark02.full = wm0->priority_mark_max.full;
 		*d1mode_priority_a_cnt = dfixed_trunc(priority_mark02);
@@ -555,8 +551,6 @@ static void rs690_compute_mode_priority(struct radeon_device *rdev,
 		}
 		if (wm1->priority_mark.full > priority_mark12.full)
 			priority_mark12.full = wm1->priority_mark.full;
-		if (dfixed_trunc(priority_mark12) < 0)
-			priority_mark12.full = 0;
 		if (wm1->priority_mark_max.full > priority_mark12.full)
 			priority_mark12.full = wm1->priority_mark_max.full;
 		*d2mode_priority_a_cnt = dfixed_trunc(priority_mark12);
diff --git a/drivers/gpu/drm/radeon/rv515.c b/drivers/gpu/drm/radeon/rv515.c
index 873eb4b193b4..5d1c316115ef 100644
--- a/drivers/gpu/drm/radeon/rv515.c
+++ b/drivers/gpu/drm/radeon/rv515.c
@@ -1155,14 +1155,10 @@ static void rv515_compute_mode_priority(struct radeon_device *rdev,
 		}
 		if (wm0->priority_mark.full > priority_mark02.full)
 			priority_mark02.full = wm0->priority_mark.full;
-		if (dfixed_trunc(priority_mark02) < 0)
-			priority_mark02.full = 0;
 		if (wm0->priority_mark_max.full > priority_mark02.full)
 			priority_mark02.full = wm0->priority_mark_max.full;
 		if (wm1->priority_mark.full > priority_mark12.full)
 			priority_mark12.full = wm1->priority_mark.full;
-		if (dfixed_trunc(priority_mark12) < 0)
-			priority_mark12.full = 0;
 		if (wm1->priority_mark_max.full > priority_mark12.full)
 			priority_mark12.full = wm1->priority_mark_max.full;
 		*d1mode_priority_a_cnt = dfixed_trunc(priority_mark02);
@@ -1193,8 +1189,6 @@ static void rv515_compute_mode_priority(struct radeon_device *rdev,
 		}
 		if (wm0->priority_mark.full > priority_mark02.full)
 			priority_mark02.full = wm0->priority_mark.full;
-		if (dfixed_trunc(priority_mark02) < 0)
-			priority_mark02.full = 0;
 		if (wm0->priority_mark_max.full > priority_mark02.full)
 			priority_mark02.full = wm0->priority_mark_max.full;
 		*d1mode_priority_a_cnt = dfixed_trunc(priority_mark02);
@@ -1222,8 +1216,6 @@ static void rv515_compute_mode_priority(struct radeon_device *rdev,
 		}
 		if (wm1->priority_mark.full > priority_mark12.full)
 			priority_mark12.full = wm1->priority_mark.full;
-		if (dfixed_trunc(priority_mark12) < 0)
-			priority_mark12.full = 0;
 		if (wm1->priority_mark_max.full > priority_mark12.full)
 			priority_mark12.full = wm1->priority_mark_max.full;
 		*d2mode_priority_a_cnt = dfixed_trunc(priority_mark12);
diff --git a/drivers/gpu/drm/radeon/rv6xx_dpm.c b/drivers/gpu/drm/radeon/rv6xx_dpm.c
index 5811d277a36a..26633a025252 100644
--- a/drivers/gpu/drm/radeon/rv6xx_dpm.c
+++ b/drivers/gpu/drm/radeon/rv6xx_dpm.c
@@ -407,9 +407,9 @@ static void rv6xx_enable_engine_feedback_and_reference_sync(struct radeon_device
 	WREG32_P(SPLL_CNTL_MODE, SPLL_DIV_SYNC, ~SPLL_DIV_SYNC);
 }
 
-static u64 rv6xx_clocks_per_unit(u32 unit)
+static u32 rv6xx_clocks_per_unit(u32 unit)
 {
-	u64 tmp = 1 << (2 * unit);
+	u32 tmp = 1 << (2 * unit);
 
 	return tmp;
 }
@@ -417,7 +417,7 @@ static u64 rv6xx_clocks_per_unit(u32 unit)
 static u32 rv6xx_scale_count_given_unit(struct radeon_device *rdev,
 					u32 unscaled_count, u32 unit)
 {
-	u32 count_per_unit = (u32)rv6xx_clocks_per_unit(unit);
+	u32 count_per_unit = rv6xx_clocks_per_unit(unit);
 
 	return (unscaled_count + count_per_unit - 1) / count_per_unit;
 }
diff --git a/drivers/gpu/drm/radeon/si.c b/drivers/gpu/drm/radeon/si.c
index d96f7cbca0a1..6a64ccaa0695 100644
--- a/drivers/gpu/drm/radeon/si.c
+++ b/drivers/gpu/drm/radeon/si.c
@@ -78,11 +78,6 @@ extern void evergreen_mc_resume(struct radeon_device *rdev, struct evergreen_mc_
 extern u32 evergreen_get_number_of_dram_channels(struct radeon_device *rdev);
 extern void evergreen_print_gpu_status_regs(struct radeon_device *rdev);
 extern bool evergreen_is_display_hung(struct radeon_device *rdev);
-extern void si_dma_vm_set_page(struct radeon_device *rdev,
-			       struct radeon_ib *ib,
-			       uint64_t pe,
-			       uint64_t addr, unsigned count,
-			       uint32_t incr, uint32_t flags);
 static void si_enable_gui_idle_interrupt(struct radeon_device *rdev,
 					 bool enable);
 static void si_fini_pg(struct radeon_device *rdev);
@@ -4673,61 +4668,6 @@ static void si_vm_decode_fault(struct radeon_device *rdev,
 	       block, mc_id);
 }
 
-/**
- * si_vm_set_page - update the page tables using the CP
- *
- * @rdev: radeon_device pointer
- * @ib: indirect buffer to fill with commands
- * @pe: addr of the page entry
- * @addr: dst addr to write into pe
- * @count: number of page entries to update
- * @incr: increase next addr by incr bytes
- * @flags: access flags
- *
- * Update the page tables using the CP (SI).
- */
-void si_vm_set_page(struct radeon_device *rdev,
-		    struct radeon_ib *ib,
-		    uint64_t pe,
-		    uint64_t addr, unsigned count,
-		    uint32_t incr, uint32_t flags)
-{
-	uint32_t r600_flags = cayman_vm_page_flags(rdev, flags);
-	uint64_t value;
-	unsigned ndw;
-
-	if (rdev->asic->vm.pt_ring_index == RADEON_RING_TYPE_GFX_INDEX) {
-		while (count) {
-			ndw = 2 + count * 2;
-			if (ndw > 0x3FFE)
-				ndw = 0x3FFE;
-
-			ib->ptr[ib->length_dw++] = PACKET3(PACKET3_WRITE_DATA, ndw);
-			ib->ptr[ib->length_dw++] = (WRITE_DATA_ENGINE_SEL(0) |
-					WRITE_DATA_DST_SEL(1));
-			ib->ptr[ib->length_dw++] = pe;
-			ib->ptr[ib->length_dw++] = upper_32_bits(pe);
-			for (; ndw > 2; ndw -= 2, --count, pe += 8) {
-				if (flags & RADEON_VM_PAGE_SYSTEM) {
-					value = radeon_vm_map_gart(rdev, addr);
-					value &= 0xFFFFFFFFFFFFF000ULL;
-				} else if (flags & RADEON_VM_PAGE_VALID) {
-					value = addr;
-				} else {
-					value = 0;
-				}
-				addr += incr;
-				value |= r600_flags;
-				ib->ptr[ib->length_dw++] = value;
-				ib->ptr[ib->length_dw++] = upper_32_bits(value);
-			}
-		}
-	} else {
-		/* DMA */
-		si_dma_vm_set_page(rdev, ib, pe, addr, count, incr, flags);
-	}
-}
-
 void si_vm_flush(struct radeon_device *rdev, int ridx, struct radeon_vm *vm)
 {
 	struct radeon_ring *ring = &rdev->ring[ridx];
@@ -5372,52 +5312,53 @@ void si_get_csb_buffer(struct radeon_device *rdev, volatile u32 *buffer)
 	if (buffer == NULL)
 		return;
 
-	buffer[count++] = PACKET3(PACKET3_PREAMBLE_CNTL, 0);
-	buffer[count++] = PACKET3_PREAMBLE_BEGIN_CLEAR_STATE;
+	buffer[count++] = cpu_to_le32(PACKET3(PACKET3_PREAMBLE_CNTL, 0));
+	buffer[count++] = cpu_to_le32(PACKET3_PREAMBLE_BEGIN_CLEAR_STATE);
 
-	buffer[count++] = PACKET3(PACKET3_CONTEXT_CONTROL, 1);
-	buffer[count++] = 0x80000000;
-	buffer[count++] = 0x80000000;
+	buffer[count++] = cpu_to_le32(PACKET3(PACKET3_CONTEXT_CONTROL, 1));
+	buffer[count++] = cpu_to_le32(0x80000000);
+	buffer[count++] = cpu_to_le32(0x80000000);
 
 	for (sect = rdev->rlc.cs_data; sect->section != NULL; ++sect) {
 		for (ext = sect->section; ext->extent != NULL; ++ext) {
 			if (sect->id == SECT_CONTEXT) {
-				buffer[count++] = PACKET3(PACKET3_SET_CONTEXT_REG, ext->reg_count);
-				buffer[count++] = ext->reg_index - 0xa000;
+				buffer[count++] =
+					cpu_to_le32(PACKET3(PACKET3_SET_CONTEXT_REG, ext->reg_count));
+				buffer[count++] = cpu_to_le32(ext->reg_index - 0xa000);
 				for (i = 0; i < ext->reg_count; i++)
-					buffer[count++] = ext->extent[i];
+					buffer[count++] = cpu_to_le32(ext->extent[i]);
 			} else {
 				return;
 			}
 		}
 	}
 
-	buffer[count++] = PACKET3(PACKET3_SET_CONTEXT_REG, 1);
-	buffer[count++] = PA_SC_RASTER_CONFIG - PACKET3_SET_CONTEXT_REG_START;
+	buffer[count++] = cpu_to_le32(PACKET3(PACKET3_SET_CONTEXT_REG, 1));
+	buffer[count++] = cpu_to_le32(PA_SC_RASTER_CONFIG - PACKET3_SET_CONTEXT_REG_START);
 	switch (rdev->family) {
 	case CHIP_TAHITI:
 	case CHIP_PITCAIRN:
-		buffer[count++] = 0x2a00126a;
+		buffer[count++] = cpu_to_le32(0x2a00126a);
 		break;
 	case CHIP_VERDE:
-		buffer[count++] = 0x0000124a;
+		buffer[count++] = cpu_to_le32(0x0000124a);
 		break;
 	case CHIP_OLAND:
-		buffer[count++] = 0x00000082;
+		buffer[count++] = cpu_to_le32(0x00000082);
 		break;
 	case CHIP_HAINAN:
-		buffer[count++] = 0x00000000;
+		buffer[count++] = cpu_to_le32(0x00000000);
 		break;
 	default:
-		buffer[count++] = 0x00000000;
+		buffer[count++] = cpu_to_le32(0x00000000);
 		break;
 	}
 
-	buffer[count++] = PACKET3(PACKET3_PREAMBLE_CNTL, 0);
-	buffer[count++] = PACKET3_PREAMBLE_END_CLEAR_STATE;
+	buffer[count++] = cpu_to_le32(PACKET3(PACKET3_PREAMBLE_CNTL, 0));
+	buffer[count++] = cpu_to_le32(PACKET3_PREAMBLE_END_CLEAR_STATE);
 
-	buffer[count++] = PACKET3(PACKET3_CLEAR_STATE, 0);
-	buffer[count++] = 0;
+	buffer[count++] = cpu_to_le32(PACKET3(PACKET3_CLEAR_STATE, 0));
+	buffer[count++] = cpu_to_le32(0);
 }
 
 static void si_init_pg(struct radeon_device *rdev)
diff --git a/drivers/gpu/drm/radeon/si_dma.c b/drivers/gpu/drm/radeon/si_dma.c
index 49909d23dfce..8e8f46133532 100644
--- a/drivers/gpu/drm/radeon/si_dma.c
+++ b/drivers/gpu/drm/radeon/si_dma.c
@@ -24,6 +24,7 @@
 #include <drm/drmP.h>
 #include "radeon.h"
 #include "radeon_asic.h"
+#include "radeon_trace.h"
 #include "sid.h"
 
 u32 si_gpu_check_soft_reset(struct radeon_device *rdev);
@@ -75,11 +76,12 @@ void si_dma_vm_set_page(struct radeon_device *rdev,
 			uint64_t addr, unsigned count,
 			uint32_t incr, uint32_t flags)
 {
-	uint32_t r600_flags = cayman_vm_page_flags(rdev, flags);
 	uint64_t value;
 	unsigned ndw;
 
-	if (flags & RADEON_VM_PAGE_SYSTEM) {
+	trace_radeon_vm_set_page(pe, addr, count, incr, flags);
+
+	if (flags & R600_PTE_SYSTEM) {
 		while (count) {
 			ndw = count * 2;
 			if (ndw > 0xFFFFE)
@@ -90,16 +92,10 @@ void si_dma_vm_set_page(struct radeon_device *rdev,
 			ib->ptr[ib->length_dw++] = pe;
 			ib->ptr[ib->length_dw++] = upper_32_bits(pe) & 0xff;
 			for (; ndw > 0; ndw -= 2, --count, pe += 8) {
-				if (flags & RADEON_VM_PAGE_SYSTEM) {
-					value = radeon_vm_map_gart(rdev, addr);
-					value &= 0xFFFFFFFFFFFFF000ULL;
-				} else if (flags & RADEON_VM_PAGE_VALID) {
-					value = addr;
-				} else {
-					value = 0;
-				}
+				value = radeon_vm_map_gart(rdev, addr);
+				value &= 0xFFFFFFFFFFFFF000ULL;
 				addr += incr;
-				value |= r600_flags;
+				value |= flags;
 				ib->ptr[ib->length_dw++] = value;
 				ib->ptr[ib->length_dw++] = upper_32_bits(value);
 			}
@@ -110,7 +106,7 @@ void si_dma_vm_set_page(struct radeon_device *rdev,
 			if (ndw > 0xFFFFE)
 				ndw = 0xFFFFE;
 
-			if (flags & RADEON_VM_PAGE_VALID)
+			if (flags & R600_PTE_VALID)
 				value = addr;
 			else
 				value = 0;
@@ -118,7 +114,7 @@ void si_dma_vm_set_page(struct radeon_device *rdev,
 			ib->ptr[ib->length_dw++] = DMA_PTE_PDE_PACKET(ndw);
 			ib->ptr[ib->length_dw++] = pe; /* dst addr */
 			ib->ptr[ib->length_dw++] = upper_32_bits(pe) & 0xff;
-			ib->ptr[ib->length_dw++] = r600_flags; /* mask */
+			ib->ptr[ib->length_dw++] = flags; /* mask */
 			ib->ptr[ib->length_dw++] = 0;
 			ib->ptr[ib->length_dw++] = value; /* value */
 			ib->ptr[ib->length_dw++] = upper_32_bits(value);
diff --git a/drivers/gpu/drm/radeon/si_dpm.c b/drivers/gpu/drm/radeon/si_dpm.c
index 2332aa1bf93c..0b00c790fb77 100644
--- a/drivers/gpu/drm/radeon/si_dpm.c
+++ b/drivers/gpu/drm/radeon/si_dpm.c
@@ -3589,7 +3589,12 @@ static void si_program_display_gap(struct radeon_device *rdev)
 		WREG32(DCCG_DISP_SLOW_SELECT_REG, tmp);
 	}
 
-	si_notify_smc_display_change(rdev, rdev->pm.dpm.new_active_crtc_count > 0);
+	/* Setting this to false forces the performance state to low if the crtcs are disabled.
+	 * This can be a problem on PowerXpress systems or if you want to use the card
+	 * for offscreen rendering or compute if there are no crtcs enabled.  Set it to
+	 * true for now so that performance scales even if the displays are off.
+	 */
+	si_notify_smc_display_change(rdev, true /*rdev->pm.dpm.new_active_crtc_count > 0*/);
 }
 
 static void si_enable_spread_spectrum(struct radeon_device *rdev, bool enable)
@@ -4553,7 +4558,7 @@ static int si_init_smc_table(struct radeon_device *rdev)
 		table->systemFlags |= PPSMC_SYSTEMFLAG_GDDR5;
 
 	if (rdev->pm.dpm.platform_caps & ATOM_PP_PLATFORM_CAP_REVERT_GPIO5_POLARITY)
-		table->systemFlags |= PPSMC_EXTRAFLAGS_AC2DC_GPIO5_POLARITY_HIGH;
+		table->extraFlags |= PPSMC_EXTRAFLAGS_AC2DC_GPIO5_POLARITY_HIGH;
 
 	if (rdev->pm.dpm.platform_caps & ATOM_PP_PLATFORM_CAP_VRHOT_GPIO_CONFIGURABLE) {
 		table->systemFlags |= PPSMC_SYSTEMFLAG_REGULATOR_HOT_PROG_GPIO;
diff --git a/drivers/gpu/drm/radeon/sid.h b/drivers/gpu/drm/radeon/sid.h
index 7e2e0ea66a00..b322acc48097 100644
--- a/drivers/gpu/drm/radeon/sid.h
+++ b/drivers/gpu/drm/radeon/sid.h
@@ -478,7 +478,7 @@
 #define		STATE3_MASK				(0x1f << 15)
 #define		STATE3_SHIFT				15
 
-#define	MC_SEQ_TRAIN_WAKEUP_CNTL			0x2808
+#define	MC_SEQ_TRAIN_WAKEUP_CNTL			0x28e8
 #define		TRAIN_DONE_D0      			(1 << 30)
 #define		TRAIN_DONE_D1      			(1 << 31)
 
@@ -683,6 +683,51 @@
  * bit5 = 176.4 kHz
  * bit6 = 192 kHz
  */
+
+#define AZ_F0_CODEC_PIN_CONTROL_RESPONSE_LIPSYNC         0x37
+#       define VIDEO_LIPSYNC(x)                           (((x) & 0xff) << 0)
+#       define AUDIO_LIPSYNC(x)                           (((x) & 0xff) << 8)
+/* VIDEO_LIPSYNC, AUDIO_LIPSYNC
+ * 0   = invalid
+ * x   = legal delay value
+ * 255 = sync not supported
+ */
+#define AZ_F0_CODEC_PIN_CONTROL_RESPONSE_HBR             0x38
+#       define HBR_CAPABLE                                (1 << 0) /* enabled by default */
+
+#define AZ_F0_CODEC_PIN_CONTROL_SINK_INFO0               0x3a
+#       define MANUFACTURER_ID(x)                        (((x) & 0xffff) << 0)
+#       define PRODUCT_ID(x)                             (((x) & 0xffff) << 16)
+#define AZ_F0_CODEC_PIN_CONTROL_SINK_INFO1               0x3b
+#       define SINK_DESCRIPTION_LEN(x)                   (((x) & 0xff) << 0)
+#define AZ_F0_CODEC_PIN_CONTROL_SINK_INFO2               0x3c
+#       define PORT_ID0(x)                               (((x) & 0xffffffff) << 0)
+#define AZ_F0_CODEC_PIN_CONTROL_SINK_INFO3               0x3d
+#       define PORT_ID1(x)                               (((x) & 0xffffffff) << 0)
+#define AZ_F0_CODEC_PIN_CONTROL_SINK_INFO4               0x3e
+#       define DESCRIPTION0(x)                           (((x) & 0xff) << 0)
+#       define DESCRIPTION1(x)                           (((x) & 0xff) << 8)
+#       define DESCRIPTION2(x)                           (((x) & 0xff) << 16)
+#       define DESCRIPTION3(x)                           (((x) & 0xff) << 24)
+#define AZ_F0_CODEC_PIN_CONTROL_SINK_INFO5               0x3f
+#       define DESCRIPTION4(x)                           (((x) & 0xff) << 0)
+#       define DESCRIPTION5(x)                           (((x) & 0xff) << 8)
+#       define DESCRIPTION6(x)                           (((x) & 0xff) << 16)
+#       define DESCRIPTION7(x)                           (((x) & 0xff) << 24)
+#define AZ_F0_CODEC_PIN_CONTROL_SINK_INFO6               0x40
+#       define DESCRIPTION8(x)                           (((x) & 0xff) << 0)
+#       define DESCRIPTION9(x)                           (((x) & 0xff) << 8)
+#       define DESCRIPTION10(x)                          (((x) & 0xff) << 16)
+#       define DESCRIPTION11(x)                          (((x) & 0xff) << 24)
+#define AZ_F0_CODEC_PIN_CONTROL_SINK_INFO7               0x41
+#       define DESCRIPTION12(x)                          (((x) & 0xff) << 0)
+#       define DESCRIPTION13(x)                          (((x) & 0xff) << 8)
+#       define DESCRIPTION14(x)                          (((x) & 0xff) << 16)
+#       define DESCRIPTION15(x)                          (((x) & 0xff) << 24)
+#define AZ_F0_CODEC_PIN_CONTROL_SINK_INFO8               0x42
+#       define DESCRIPTION16(x)                          (((x) & 0xff) << 0)
+#       define DESCRIPTION17(x)                          (((x) & 0xff) << 8)
+
 #define AZ_F0_CODEC_PIN_CONTROL_HOTPLUG_CONTROL          0x54
 #       define AUDIO_ENABLED                             (1 << 31)
 
diff --git a/drivers/gpu/drm/rcar-du/Kconfig b/drivers/gpu/drm/rcar-du/Kconfig
index c590cd9dca0b..d8e835ac2c5e 100644
--- a/drivers/gpu/drm/rcar-du/Kconfig
+++ b/drivers/gpu/drm/rcar-du/Kconfig
@@ -4,6 +4,7 @@ config DRM_RCAR_DU
 	select DRM_KMS_HELPER
 	select DRM_KMS_CMA_HELPER
 	select DRM_GEM_CMA_HELPER
+	select DRM_KMS_FB_HELPER
 	help
 	  Choose this option if you have an R-Car chipset.
 	  If M is selected the module will be called rcar-du-drm.
diff --git a/drivers/gpu/drm/shmobile/Kconfig b/drivers/gpu/drm/shmobile/Kconfig
index ca498d151a76..2ee44ca9d67f 100644
--- a/drivers/gpu/drm/shmobile/Kconfig
+++ b/drivers/gpu/drm/shmobile/Kconfig
@@ -1,7 +1,9 @@
 config DRM_SHMOBILE
 	tristate "DRM Support for SH Mobile"
 	depends on DRM && (ARM || SUPERH)
+	select BACKLIGHT_CLASS_DEVICE
 	select DRM_KMS_HELPER
+	select DRM_KMS_FB_HELPER
 	select DRM_KMS_CMA_HELPER
 	select DRM_GEM_CMA_HELPER
 	help
diff --git a/drivers/gpu/drm/shmobile/shmob_drm_crtc.c b/drivers/gpu/drm/shmobile/shmob_drm_crtc.c
index 54bad98e9477..562f9a401cf6 100644
--- a/drivers/gpu/drm/shmobile/shmob_drm_crtc.c
+++ b/drivers/gpu/drm/shmobile/shmob_drm_crtc.c
@@ -40,7 +40,7 @@
 static void shmob_drm_clk_on(struct shmob_drm_device *sdev)
 {
 	if (sdev->clock)
-		clk_enable(sdev->clock);
+		clk_prepare_enable(sdev->clock);
 #if 0
 	if (sdev->meram_dev && sdev->meram_dev->pdev)
 		pm_runtime_get_sync(&sdev->meram_dev->pdev->dev);
@@ -54,7 +54,7 @@ static void shmob_drm_clk_off(struct shmob_drm_device *sdev)
 		pm_runtime_put_sync(&sdev->meram_dev->pdev->dev);
 #endif
 	if (sdev->clock)
-		clk_disable(sdev->clock);
+		clk_disable_unprepare(sdev->clock);
 }
 
 /* -----------------------------------------------------------------------------
diff --git a/drivers/gpu/host1x/drm/Kconfig b/drivers/gpu/drm/tegra/Kconfig
index 69853a4de40a..8961ba6a34b8 100644
--- a/drivers/gpu/host1x/drm/Kconfig
+++ b/drivers/gpu/drm/tegra/Kconfig
@@ -1,7 +1,10 @@
 config DRM_TEGRA
 	bool "NVIDIA Tegra DRM"
+	depends on ARCH_TEGRA || ARCH_MULTIPLATFORM
 	depends on DRM
+	select TEGRA_HOST1X
 	select DRM_KMS_HELPER
+	select DRM_KMS_FB_HELPER
 	select FB_SYS_FILLRECT
 	select FB_SYS_COPYAREA
 	select FB_SYS_IMAGEBLIT
@@ -13,6 +16,11 @@ config DRM_TEGRA
 
 if DRM_TEGRA
 
+config DRM_TEGRA_DEBUG
+	bool "NVIDIA Tegra DRM debug support"
+	help
+	  Say yes here to enable debugging support.
+
 config DRM_TEGRA_STAGING
 	bool "Enable HOST1X interface"
 	depends on STAGING
@@ -21,9 +29,4 @@ config DRM_TEGRA_STAGING
 
 	  If unsure, choose N.
 
-config DRM_TEGRA_DEBUG
-	bool "NVIDIA Tegra DRM debug support"
-	help
-	  Say yes here to enable debugging support.
-
 endif
diff --git a/drivers/gpu/drm/tegra/Makefile b/drivers/gpu/drm/tegra/Makefile
new file mode 100644
index 000000000000..edc76abd58bb
--- /dev/null
+++ b/drivers/gpu/drm/tegra/Makefile
@@ -0,0 +1,15 @@
+ccflags-$(CONFIG_DRM_TEGRA_DEBUG) += -DDEBUG
+
+tegra-drm-y := \
+	bus.o \
+	drm.o \
+	gem.o \
+	fb.o \
+	dc.o \
+	output.o \
+	rgb.o \
+	hdmi.o \
+	gr2d.o \
+	gr3d.o
+
+obj-$(CONFIG_DRM_TEGRA) += tegra-drm.o
diff --git a/drivers/gpu/drm/tegra/bus.c b/drivers/gpu/drm/tegra/bus.c
new file mode 100644
index 000000000000..565f8f7b9a47
--- /dev/null
+++ b/drivers/gpu/drm/tegra/bus.c
@@ -0,0 +1,76 @@
+/*
+ * Copyright (C) 2013 NVIDIA Corporation
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ */
+
+#include "drm.h"
+
+static int drm_host1x_set_busid(struct drm_device *dev,
+				struct drm_master *master)
+{
+	const char *device = dev_name(dev->dev);
+	const char *driver = dev->driver->name;
+	const char *bus = dev->dev->bus->name;
+	int length;
+
+	master->unique_len = strlen(bus) + 1 + strlen(device);
+	master->unique_size = master->unique_len;
+
+	master->unique = kmalloc(master->unique_len + 1, GFP_KERNEL);
+	if (!master->unique)
+		return -ENOMEM;
+
+	snprintf(master->unique, master->unique_len + 1, "%s:%s", bus, device);
+
+	length = strlen(driver) + 1 + master->unique_len;
+
+	dev->devname = kmalloc(length + 1, GFP_KERNEL);
+	if (!dev->devname)
+		return -ENOMEM;
+
+	snprintf(dev->devname, length + 1, "%s@%s", driver, master->unique);
+
+	return 0;
+}
+
+static struct drm_bus drm_host1x_bus = {
+	.bus_type = DRIVER_BUS_HOST1X,
+	.set_busid = drm_host1x_set_busid,
+};
+
+int drm_host1x_init(struct drm_driver *driver, struct host1x_device *device)
+{
+	struct drm_device *drm;
+	int ret;
+
+	INIT_LIST_HEAD(&driver->device_list);
+	driver->bus = &drm_host1x_bus;
+
+	drm = drm_dev_alloc(driver, &device->dev);
+	if (!drm)
+		return -ENOMEM;
+
+	ret = drm_dev_register(drm, 0);
+	if (ret)
+		goto err_free;
+
+	DRM_INFO("Initialized %s %d.%d.%d %s on minor %d\n", driver->name,
+		 driver->major, driver->minor, driver->patchlevel,
+		 driver->date, drm->primary->index);
+
+	return 0;
+
+err_free:
+	drm_dev_free(drm);
+	return ret;
+}
+
+void drm_host1x_exit(struct drm_driver *driver, struct host1x_device *device)
+{
+	struct tegra_drm *tegra = dev_get_drvdata(&device->dev);
+
+	drm_put_dev(tegra->drm);
+}
diff --git a/drivers/gpu/host1x/drm/dc.c b/drivers/gpu/drm/tegra/dc.c
index b1a05ad901c3..ae1cb31ead7e 100644
--- a/drivers/gpu/host1x/drm/dc.c
+++ b/drivers/gpu/drm/tegra/dc.c
@@ -8,13 +8,9 @@
  */
 
 #include <linux/clk.h>
-#include <linux/debugfs.h>
-#include <linux/module.h>
-#include <linux/of.h>
-#include <linux/platform_device.h>
 #include <linux/clk/tegra.h>
+#include <linux/debugfs.h>
 
-#include "host1x_client.h"
 #include "dc.h"
 #include "drm.h"
 #include "gem.h"
@@ -51,6 +47,8 @@ static int tegra_plane_update(struct drm_plane *plane, struct drm_crtc *crtc,
 	window.dst.h = crtc_h;
 	window.format = tegra_dc_format(fb->pixel_format);
 	window.bits_per_pixel = fb->bits_per_pixel;
+	window.bottom_up = tegra_fb_is_bottom_up(fb);
+	window.tiled = tegra_fb_is_tiled(fb);
 
 	for (i = 0; i < drm_format_num_planes(fb->pixel_format); i++) {
 		struct tegra_bo *bo = tegra_fb_get_plane(fb, i);
@@ -97,8 +95,11 @@ static int tegra_plane_disable(struct drm_plane *plane)
 
 static void tegra_plane_destroy(struct drm_plane *plane)
 {
+	struct tegra_plane *p = to_tegra_plane(plane);
+
 	tegra_plane_disable(plane);
 	drm_plane_cleanup(plane);
+	kfree(p);
 }
 
 static const struct drm_plane_funcs tegra_plane_funcs = {
@@ -124,7 +125,7 @@ static int tegra_dc_add_planes(struct drm_device *drm, struct tegra_dc *dc)
 	for (i = 0; i < 2; i++) {
 		struct tegra_plane *plane;
 
-		plane = devm_kzalloc(drm->dev, sizeof(*plane), GFP_KERNEL);
+		plane = kzalloc(sizeof(*plane), GFP_KERNEL);
 		if (!plane)
 			return -ENOMEM;
 
@@ -133,8 +134,10 @@ static int tegra_dc_add_planes(struct drm_device *drm, struct tegra_dc *dc)
 		err = drm_plane_init(drm, &plane->base, 1 << dc->pipe,
 				     &tegra_plane_funcs, plane_formats,
 				     ARRAY_SIZE(plane_formats), false);
-		if (err < 0)
+		if (err < 0) {
+			kfree(plane);
 			return err;
+		}
 	}
 
 	return 0;
@@ -145,6 +148,7 @@ static int tegra_dc_set_base(struct tegra_dc *dc, int x, int y,
 {
 	unsigned int format = tegra_dc_format(fb->pixel_format);
 	struct tegra_bo *bo = tegra_fb_get_plane(fb, 0);
+	unsigned int h_offset = 0, v_offset = 0;
 	unsigned long value;
 
 	tegra_dc_writel(dc, WINDOW_A_SELECT, DC_CMD_DISPLAY_WINDOW_HEADER);
@@ -156,6 +160,32 @@ static int tegra_dc_set_base(struct tegra_dc *dc, int x, int y,
 	tegra_dc_writel(dc, fb->pitches[0], DC_WIN_LINE_STRIDE);
 	tegra_dc_writel(dc, format, DC_WIN_COLOR_DEPTH);
 
+	if (tegra_fb_is_tiled(fb)) {
+		value = DC_WIN_BUFFER_ADDR_MODE_TILE_UV |
+			DC_WIN_BUFFER_ADDR_MODE_TILE;
+	} else {
+		value = DC_WIN_BUFFER_ADDR_MODE_LINEAR_UV |
+			DC_WIN_BUFFER_ADDR_MODE_LINEAR;
+	}
+
+	tegra_dc_writel(dc, value, DC_WIN_BUFFER_ADDR_MODE);
+
+	/* make sure bottom-up buffers are properly displayed */
+	if (tegra_fb_is_bottom_up(fb)) {
+		value = tegra_dc_readl(dc, DC_WIN_WIN_OPTIONS);
+		value |= INVERT_V;
+		tegra_dc_writel(dc, value, DC_WIN_WIN_OPTIONS);
+
+		v_offset += fb->height - 1;
+	} else {
+		value = tegra_dc_readl(dc, DC_WIN_WIN_OPTIONS);
+		value &= ~INVERT_V;
+		tegra_dc_writel(dc, value, DC_WIN_WIN_OPTIONS);
+	}
+
+	tegra_dc_writel(dc, h_offset, DC_WINBUF_ADDR_H_OFFSET);
+	tegra_dc_writel(dc, v_offset, DC_WINBUF_ADDR_V_OFFSET);
+
 	value = GENERAL_UPDATE | WIN_A_UPDATE;
 	tegra_dc_writel(dc, value, DC_CMD_STATE_CONTROL);
 
@@ -255,14 +285,26 @@ static int tegra_dc_page_flip(struct drm_crtc *crtc, struct drm_framebuffer *fb,
 	return 0;
 }
 
+static void drm_crtc_clear(struct drm_crtc *crtc)
+{
+	memset(crtc, 0, sizeof(*crtc));
+}
+
+static void tegra_dc_destroy(struct drm_crtc *crtc)
+{
+	drm_crtc_cleanup(crtc);
+	drm_crtc_clear(crtc);
+}
+
 static const struct drm_crtc_funcs tegra_crtc_funcs = {
 	.page_flip = tegra_dc_page_flip,
 	.set_config = drm_crtc_helper_set_config,
-	.destroy = drm_crtc_cleanup,
+	.destroy = tegra_dc_destroy,
 };
 
 static void tegra_crtc_disable(struct drm_crtc *crtc)
 {
+	struct tegra_dc *dc = to_tegra_dc(crtc);
 	struct drm_device *drm = crtc->dev;
 	struct drm_plane *plane;
 
@@ -277,6 +319,8 @@ static void tegra_crtc_disable(struct drm_crtc *crtc)
 			}
 		}
 	}
+
+	drm_vblank_off(drm, dc->pipe);
 }
 
 static bool tegra_crtc_mode_fixup(struct drm_crtc *crtc,
@@ -491,9 +535,22 @@ int tegra_dc_setup_window(struct tegra_dc *dc, unsigned int index,
 		tegra_dc_writel(dc, window->stride[0], DC_WIN_LINE_STRIDE);
 	}
 
+	if (window->bottom_up)
+		v_offset += window->src.h - 1;
+
 	tegra_dc_writel(dc, h_offset, DC_WINBUF_ADDR_H_OFFSET);
 	tegra_dc_writel(dc, v_offset, DC_WINBUF_ADDR_V_OFFSET);
 
+	if (window->tiled) {
+		value = DC_WIN_BUFFER_ADDR_MODE_TILE_UV |
+			DC_WIN_BUFFER_ADDR_MODE_TILE;
+	} else {
+		value = DC_WIN_BUFFER_ADDR_MODE_LINEAR_UV |
+			DC_WIN_BUFFER_ADDR_MODE_LINEAR;
+	}
+
+	tegra_dc_writel(dc, value, DC_WIN_BUFFER_ADDR_MODE);
+
 	value = WIN_ENABLE;
 
 	if (yuv) {
@@ -512,6 +569,9 @@ int tegra_dc_setup_window(struct tegra_dc *dc, unsigned int index,
 		value |= COLOR_EXPAND;
 	}
 
+	if (window->bottom_up)
+		value |= INVERT_V;
+
 	tegra_dc_writel(dc, value, DC_WIN_WIN_OPTIONS);
 
 	/*
@@ -1041,30 +1101,30 @@ static int tegra_dc_debugfs_exit(struct tegra_dc *dc)
 	return 0;
 }
 
-static int tegra_dc_drm_init(struct host1x_client *client,
-			     struct drm_device *drm)
+static int tegra_dc_init(struct host1x_client *client)
 {
+	struct tegra_drm *tegra = dev_get_drvdata(client->parent);
 	struct tegra_dc *dc = host1x_client_to_dc(client);
 	int err;
 
-	dc->pipe = drm->mode_config.num_crtc;
+	dc->pipe = tegra->drm->mode_config.num_crtc;
 
-	drm_crtc_init(drm, &dc->base, &tegra_crtc_funcs);
+	drm_crtc_init(tegra->drm, &dc->base, &tegra_crtc_funcs);
 	drm_mode_crtc_set_gamma_size(&dc->base, 256);
 	drm_crtc_helper_add(&dc->base, &tegra_crtc_helper_funcs);
 
-	err = tegra_dc_rgb_init(drm, dc);
+	err = tegra_dc_rgb_init(tegra->drm, dc);
 	if (err < 0 && err != -ENODEV) {
 		dev_err(dc->dev, "failed to initialize RGB output: %d\n", err);
 		return err;
 	}
 
-	err = tegra_dc_add_planes(drm, dc);
+	err = tegra_dc_add_planes(tegra->drm, dc);
 	if (err < 0)
 		return err;
 
 	if (IS_ENABLED(CONFIG_DEBUG_FS)) {
-		err = tegra_dc_debugfs_init(dc, drm->primary);
+		err = tegra_dc_debugfs_init(dc, tegra->drm->primary);
 		if (err < 0)
 			dev_err(dc->dev, "debugfs setup failed: %d\n", err);
 	}
@@ -1080,7 +1140,7 @@ static int tegra_dc_drm_init(struct host1x_client *client,
 	return 0;
 }
 
-static int tegra_dc_drm_exit(struct host1x_client *client)
+static int tegra_dc_exit(struct host1x_client *client)
 {
 	struct tegra_dc *dc = host1x_client_to_dc(client);
 	int err;
@@ -1103,13 +1163,12 @@ static int tegra_dc_drm_exit(struct host1x_client *client)
 }
 
 static const struct host1x_client_ops dc_client_ops = {
-	.drm_init = tegra_dc_drm_init,
-	.drm_exit = tegra_dc_drm_exit,
+	.init = tegra_dc_init,
+	.exit = tegra_dc_exit,
 };
 
 static int tegra_dc_probe(struct platform_device *pdev)
 {
-	struct host1x_drm *host1x = host1x_get_drm_data(pdev->dev.parent);
 	struct resource *regs;
 	struct tegra_dc *dc;
 	int err;
@@ -1153,7 +1212,7 @@ static int tegra_dc_probe(struct platform_device *pdev)
 		return err;
 	}
 
-	err = host1x_register_client(host1x, &dc->client);
+	err = host1x_client_register(&dc->client);
 	if (err < 0) {
 		dev_err(&pdev->dev, "failed to register host1x client: %d\n",
 			err);
@@ -1167,17 +1226,22 @@ static int tegra_dc_probe(struct platform_device *pdev)
 
 static int tegra_dc_remove(struct platform_device *pdev)
 {
-	struct host1x_drm *host1x = host1x_get_drm_data(pdev->dev.parent);
 	struct tegra_dc *dc = platform_get_drvdata(pdev);
 	int err;
 
-	err = host1x_unregister_client(host1x, &dc->client);
+	err = host1x_client_unregister(&dc->client);
 	if (err < 0) {
 		dev_err(&pdev->dev, "failed to unregister host1x client: %d\n",
 			err);
 		return err;
 	}
 
+	err = tegra_dc_rgb_remove(dc);
+	if (err < 0) {
+		dev_err(&pdev->dev, "failed to remove RGB output: %d\n", err);
+		return err;
+	}
+
 	clk_disable_unprepare(dc->clk);
 
 	return 0;
diff --git a/drivers/gpu/host1x/drm/dc.h b/drivers/gpu/drm/tegra/dc.h
index 79eaec9aac77..91bbda291470 100644
--- a/drivers/gpu/host1x/drm/dc.h
+++ b/drivers/gpu/drm/tegra/dc.h
@@ -302,6 +302,7 @@
 #define DC_WIN_CSC_KVB				0x618
 
 #define DC_WIN_WIN_OPTIONS			0x700
+#define INVERT_V     (1 <<  2)
 #define COLOR_EXPAND (1 <<  6)
 #define CSC_ENABLE   (1 << 18)
 #define WIN_ENABLE   (1 << 30)
@@ -365,6 +366,10 @@
 #define DC_WIN_BUF_STRIDE			0x70b
 #define DC_WIN_UV_BUF_STRIDE			0x70c
 #define DC_WIN_BUFFER_ADDR_MODE			0x70d
+#define DC_WIN_BUFFER_ADDR_MODE_LINEAR		(0 <<  0)
+#define DC_WIN_BUFFER_ADDR_MODE_TILE		(1 <<  0)
+#define DC_WIN_BUFFER_ADDR_MODE_LINEAR_UV	(0 << 16)
+#define DC_WIN_BUFFER_ADDR_MODE_TILE_UV		(1 << 16)
 #define DC_WIN_DV_CONTROL			0x70e
 
 #define DC_WIN_BLEND_NOKEY			0x70f
diff --git a/drivers/gpu/drm/tegra/drm.c b/drivers/gpu/drm/tegra/drm.c
new file mode 100644
index 000000000000..28e178137718
--- /dev/null
+++ b/drivers/gpu/drm/tegra/drm.c
@@ -0,0 +1,714 @@
+/*
+ * Copyright (C) 2012 Avionic Design GmbH
+ * Copyright (C) 2012-2013 NVIDIA CORPORATION.  All rights reserved.
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ */
+
+#include <linux/host1x.h>
+
+#include "drm.h"
+#include "gem.h"
+
+#define DRIVER_NAME "tegra"
+#define DRIVER_DESC "NVIDIA Tegra graphics"
+#define DRIVER_DATE "20120330"
+#define DRIVER_MAJOR 0
+#define DRIVER_MINOR 0
+#define DRIVER_PATCHLEVEL 0
+
+struct tegra_drm_file {
+	struct list_head contexts;
+};
+
+static int tegra_drm_load(struct drm_device *drm, unsigned long flags)
+{
+	struct host1x_device *device = to_host1x_device(drm->dev);
+	struct tegra_drm *tegra;
+	int err;
+
+	tegra = kzalloc(sizeof(*tegra), GFP_KERNEL);
+	if (!tegra)
+		return -ENOMEM;
+
+	dev_set_drvdata(drm->dev, tegra);
+	mutex_init(&tegra->clients_lock);
+	INIT_LIST_HEAD(&tegra->clients);
+	drm->dev_private = tegra;
+	tegra->drm = drm;
+
+	drm_mode_config_init(drm);
+
+	err = host1x_device_init(device);
+	if (err < 0)
+		return err;
+
+	/*
+	 * We don't use the drm_irq_install() helpers provided by the DRM
+	 * core, so we need to set this manually in order to allow the
+	 * DRM_IOCTL_WAIT_VBLANK to operate correctly.
+	 */
+	drm->irq_enabled = true;
+
+	err = drm_vblank_init(drm, drm->mode_config.num_crtc);
+	if (err < 0)
+		return err;
+
+	err = tegra_drm_fb_init(drm);
+	if (err < 0)
+		return err;
+
+	drm_kms_helper_poll_init(drm);
+
+	return 0;
+}
+
+static int tegra_drm_unload(struct drm_device *drm)
+{
+	struct host1x_device *device = to_host1x_device(drm->dev);
+	int err;
+
+	drm_kms_helper_poll_fini(drm);
+	tegra_drm_fb_exit(drm);
+	drm_vblank_cleanup(drm);
+	drm_mode_config_cleanup(drm);
+
+	err = host1x_device_exit(device);
+	if (err < 0)
+		return err;
+
+	return 0;
+}
+
+static int tegra_drm_open(struct drm_device *drm, struct drm_file *filp)
+{
+	struct tegra_drm_file *fpriv;
+
+	fpriv = kzalloc(sizeof(*fpriv), GFP_KERNEL);
+	if (!fpriv)
+		return -ENOMEM;
+
+	INIT_LIST_HEAD(&fpriv->contexts);
+	filp->driver_priv = fpriv;
+
+	return 0;
+}
+
+static void tegra_drm_context_free(struct tegra_drm_context *context)
+{
+	context->client->ops->close_channel(context);
+	kfree(context);
+}
+
+static void tegra_drm_lastclose(struct drm_device *drm)
+{
+	struct tegra_drm *tegra = drm->dev_private;
+
+	tegra_fbdev_restore_mode(tegra->fbdev);
+}
+
+static struct host1x_bo *
+host1x_bo_lookup(struct drm_device *drm, struct drm_file *file, u32 handle)
+{
+	struct drm_gem_object *gem;
+	struct tegra_bo *bo;
+
+	gem = drm_gem_object_lookup(drm, file, handle);
+	if (!gem)
+		return NULL;
+
+	mutex_lock(&drm->struct_mutex);
+	drm_gem_object_unreference(gem);
+	mutex_unlock(&drm->struct_mutex);
+
+	bo = to_tegra_bo(gem);
+	return &bo->base;
+}
+
+int tegra_drm_submit(struct tegra_drm_context *context,
+		     struct drm_tegra_submit *args, struct drm_device *drm,
+		     struct drm_file *file)
+{
+	unsigned int num_cmdbufs = args->num_cmdbufs;
+	unsigned int num_relocs = args->num_relocs;
+	unsigned int num_waitchks = args->num_waitchks;
+	struct drm_tegra_cmdbuf __user *cmdbufs =
+		(void * __user)(uintptr_t)args->cmdbufs;
+	struct drm_tegra_reloc __user *relocs =
+		(void * __user)(uintptr_t)args->relocs;
+	struct drm_tegra_waitchk __user *waitchks =
+		(void * __user)(uintptr_t)args->waitchks;
+	struct drm_tegra_syncpt syncpt;
+	struct host1x_job *job;
+	int err;
+
+	/* We don't yet support other than one syncpt_incr struct per submit */
+	if (args->num_syncpts != 1)
+		return -EINVAL;
+
+	job = host1x_job_alloc(context->channel, args->num_cmdbufs,
+			       args->num_relocs, args->num_waitchks);
+	if (!job)
+		return -ENOMEM;
+
+	job->num_relocs = args->num_relocs;
+	job->num_waitchk = args->num_waitchks;
+	job->client = (u32)args->context;
+	job->class = context->client->base.class;
+	job->serialize = true;
+
+	while (num_cmdbufs) {
+		struct drm_tegra_cmdbuf cmdbuf;
+		struct host1x_bo *bo;
+
+		err = copy_from_user(&cmdbuf, cmdbufs, sizeof(cmdbuf));
+		if (err)
+			goto fail;
+
+		bo = host1x_bo_lookup(drm, file, cmdbuf.handle);
+		if (!bo) {
+			err = -ENOENT;
+			goto fail;
+		}
+
+		host1x_job_add_gather(job, bo, cmdbuf.words, cmdbuf.offset);
+		num_cmdbufs--;
+		cmdbufs++;
+	}
+
+	err = copy_from_user(job->relocarray, relocs,
+			     sizeof(*relocs) * num_relocs);
+	if (err)
+		goto fail;
+
+	while (num_relocs--) {
+		struct host1x_reloc *reloc = &job->relocarray[num_relocs];
+		struct host1x_bo *cmdbuf, *target;
+
+		cmdbuf = host1x_bo_lookup(drm, file, (u32)reloc->cmdbuf);
+		target = host1x_bo_lookup(drm, file, (u32)reloc->target);
+
+		reloc->cmdbuf = cmdbuf;
+		reloc->target = target;
+
+		if (!reloc->target || !reloc->cmdbuf) {
+			err = -ENOENT;
+			goto fail;
+		}
+	}
+
+	err = copy_from_user(job->waitchk, waitchks,
+			     sizeof(*waitchks) * num_waitchks);
+	if (err)
+		goto fail;
+
+	err = copy_from_user(&syncpt, (void * __user)(uintptr_t)args->syncpts,
+			     sizeof(syncpt));
+	if (err)
+		goto fail;
+
+	job->is_addr_reg = context->client->ops->is_addr_reg;
+	job->syncpt_incrs = syncpt.incrs;
+	job->syncpt_id = syncpt.id;
+	job->timeout = 10000;
+
+	if (args->timeout && args->timeout < 10000)
+		job->timeout = args->timeout;
+
+	err = host1x_job_pin(job, context->client->base.dev);
+	if (err)
+		goto fail;
+
+	err = host1x_job_submit(job);
+	if (err)
+		goto fail_submit;
+
+	args->fence = job->syncpt_end;
+
+	host1x_job_put(job);
+	return 0;
+
+fail_submit:
+	host1x_job_unpin(job);
+fail:
+	host1x_job_put(job);
+	return err;
+}
+
+
+#ifdef CONFIG_DRM_TEGRA_STAGING
+static struct tegra_drm_context *tegra_drm_get_context(__u64 context)
+{
+	return (struct tegra_drm_context *)(uintptr_t)context;
+}
+
+static bool tegra_drm_file_owns_context(struct tegra_drm_file *file,
+					struct tegra_drm_context *context)
+{
+	struct tegra_drm_context *ctx;
+
+	list_for_each_entry(ctx, &file->contexts, list)
+		if (ctx == context)
+			return true;
+
+	return false;
+}
+
+static int tegra_gem_create(struct drm_device *drm, void *data,
+			    struct drm_file *file)
+{
+	struct drm_tegra_gem_create *args = data;
+	struct tegra_bo *bo;
+
+	bo = tegra_bo_create_with_handle(file, drm, args->size, args->flags,
+					 &args->handle);
+	if (IS_ERR(bo))
+		return PTR_ERR(bo);
+
+	return 0;
+}
+
+static int tegra_gem_mmap(struct drm_device *drm, void *data,
+			  struct drm_file *file)
+{
+	struct drm_tegra_gem_mmap *args = data;
+	struct drm_gem_object *gem;
+	struct tegra_bo *bo;
+
+	gem = drm_gem_object_lookup(drm, file, args->handle);
+	if (!gem)
+		return -EINVAL;
+
+	bo = to_tegra_bo(gem);
+
+	args->offset = drm_vma_node_offset_addr(&bo->gem.vma_node);
+
+	drm_gem_object_unreference(gem);
+
+	return 0;
+}
+
+static int tegra_syncpt_read(struct drm_device *drm, void *data,
+			     struct drm_file *file)
+{
+	struct host1x *host = dev_get_drvdata(drm->dev->parent);
+	struct drm_tegra_syncpt_read *args = data;
+	struct host1x_syncpt *sp;
+
+	sp = host1x_syncpt_get(host, args->id);
+	if (!sp)
+		return -EINVAL;
+
+	args->value = host1x_syncpt_read_min(sp);
+	return 0;
+}
+
+static int tegra_syncpt_incr(struct drm_device *drm, void *data,
+			     struct drm_file *file)
+{
+	struct host1x *host1x = dev_get_drvdata(drm->dev->parent);
+	struct drm_tegra_syncpt_incr *args = data;
+	struct host1x_syncpt *sp;
+
+	sp = host1x_syncpt_get(host1x, args->id);
+	if (!sp)
+		return -EINVAL;
+
+	return host1x_syncpt_incr(sp);
+}
+
+static int tegra_syncpt_wait(struct drm_device *drm, void *data,
+			     struct drm_file *file)
+{
+	struct host1x *host1x = dev_get_drvdata(drm->dev->parent);
+	struct drm_tegra_syncpt_wait *args = data;
+	struct host1x_syncpt *sp;
+
+	sp = host1x_syncpt_get(host1x, args->id);
+	if (!sp)
+		return -EINVAL;
+
+	return host1x_syncpt_wait(sp, args->thresh, args->timeout,
+				  &args->value);
+}
+
+static int tegra_open_channel(struct drm_device *drm, void *data,
+			      struct drm_file *file)
+{
+	struct tegra_drm_file *fpriv = file->driver_priv;
+	struct tegra_drm *tegra = drm->dev_private;
+	struct drm_tegra_open_channel *args = data;
+	struct tegra_drm_context *context;
+	struct tegra_drm_client *client;
+	int err = -ENODEV;
+
+	context = kzalloc(sizeof(*context), GFP_KERNEL);
+	if (!context)
+		return -ENOMEM;
+
+	list_for_each_entry(client, &tegra->clients, list)
+		if (client->base.class == args->client) {
+			err = client->ops->open_channel(client, context);
+			if (err)
+				break;
+
+			list_add(&context->list, &fpriv->contexts);
+			args->context = (uintptr_t)context;
+			context->client = client;
+			return 0;
+		}
+
+	kfree(context);
+	return err;
+}
+
+static int tegra_close_channel(struct drm_device *drm, void *data,
+			       struct drm_file *file)
+{
+	struct tegra_drm_file *fpriv = file->driver_priv;
+	struct drm_tegra_close_channel *args = data;
+	struct tegra_drm_context *context;
+
+	context = tegra_drm_get_context(args->context);
+
+	if (!tegra_drm_file_owns_context(fpriv, context))
+		return -EINVAL;
+
+	list_del(&context->list);
+	tegra_drm_context_free(context);
+
+	return 0;
+}
+
+static int tegra_get_syncpt(struct drm_device *drm, void *data,
+			    struct drm_file *file)
+{
+	struct tegra_drm_file *fpriv = file->driver_priv;
+	struct drm_tegra_get_syncpt *args = data;
+	struct tegra_drm_context *context;
+	struct host1x_syncpt *syncpt;
+
+	context = tegra_drm_get_context(args->context);
+
+	if (!tegra_drm_file_owns_context(fpriv, context))
+		return -ENODEV;
+
+	if (args->index >= context->client->base.num_syncpts)
+		return -EINVAL;
+
+	syncpt = context->client->base.syncpts[args->index];
+	args->id = host1x_syncpt_id(syncpt);
+
+	return 0;
+}
+
+static int tegra_submit(struct drm_device *drm, void *data,
+			struct drm_file *file)
+{
+	struct tegra_drm_file *fpriv = file->driver_priv;
+	struct drm_tegra_submit *args = data;
+	struct tegra_drm_context *context;
+
+	context = tegra_drm_get_context(args->context);
+
+	if (!tegra_drm_file_owns_context(fpriv, context))
+		return -ENODEV;
+
+	return context->client->ops->submit(context, args, drm, file);
+}
+
+static int tegra_get_syncpt_base(struct drm_device *drm, void *data,
+				 struct drm_file *file)
+{
+	struct tegra_drm_file *fpriv = file->driver_priv;
+	struct drm_tegra_get_syncpt_base *args = data;
+	struct tegra_drm_context *context;
+	struct host1x_syncpt_base *base;
+	struct host1x_syncpt *syncpt;
+
+	context = tegra_drm_get_context(args->context);
+
+	if (!tegra_drm_file_owns_context(fpriv, context))
+		return -ENODEV;
+
+	if (args->syncpt >= context->client->base.num_syncpts)
+		return -EINVAL;
+
+	syncpt = context->client->base.syncpts[args->syncpt];
+
+	base = host1x_syncpt_get_base(syncpt);
+	if (!base)
+		return -ENXIO;
+
+	args->id = host1x_syncpt_base_id(base);
+
+	return 0;
+}
+#endif
+
+static const struct drm_ioctl_desc tegra_drm_ioctls[] = {
+#ifdef CONFIG_DRM_TEGRA_STAGING
+	DRM_IOCTL_DEF_DRV(TEGRA_GEM_CREATE, tegra_gem_create, DRM_UNLOCKED | DRM_AUTH),
+	DRM_IOCTL_DEF_DRV(TEGRA_GEM_MMAP, tegra_gem_mmap, DRM_UNLOCKED),
+	DRM_IOCTL_DEF_DRV(TEGRA_SYNCPT_READ, tegra_syncpt_read, DRM_UNLOCKED),
+	DRM_IOCTL_DEF_DRV(TEGRA_SYNCPT_INCR, tegra_syncpt_incr, DRM_UNLOCKED),
+	DRM_IOCTL_DEF_DRV(TEGRA_SYNCPT_WAIT, tegra_syncpt_wait, DRM_UNLOCKED),
+	DRM_IOCTL_DEF_DRV(TEGRA_OPEN_CHANNEL, tegra_open_channel, DRM_UNLOCKED),
+	DRM_IOCTL_DEF_DRV(TEGRA_CLOSE_CHANNEL, tegra_close_channel, DRM_UNLOCKED),
+	DRM_IOCTL_DEF_DRV(TEGRA_GET_SYNCPT, tegra_get_syncpt, DRM_UNLOCKED),
+	DRM_IOCTL_DEF_DRV(TEGRA_SUBMIT, tegra_submit, DRM_UNLOCKED),
+	DRM_IOCTL_DEF_DRV(TEGRA_GET_SYNCPT_BASE, tegra_get_syncpt_base, DRM_UNLOCKED),
+#endif
+};
+
+static const struct file_operations tegra_drm_fops = {
+	.owner = THIS_MODULE,
+	.open = drm_open,
+	.release = drm_release,
+	.unlocked_ioctl = drm_ioctl,
+	.mmap = tegra_drm_mmap,
+	.poll = drm_poll,
+	.read = drm_read,
+#ifdef CONFIG_COMPAT
+	.compat_ioctl = drm_compat_ioctl,
+#endif
+	.llseek = noop_llseek,
+};
+
+static struct drm_crtc *tegra_crtc_from_pipe(struct drm_device *drm, int pipe)
+{
+	struct drm_crtc *crtc;
+
+	list_for_each_entry(crtc, &drm->mode_config.crtc_list, head) {
+		struct tegra_dc *dc = to_tegra_dc(crtc);
+
+		if (dc->pipe == pipe)
+			return crtc;
+	}
+
+	return NULL;
+}
+
+static u32 tegra_drm_get_vblank_counter(struct drm_device *dev, int crtc)
+{
+	/* TODO: implement real hardware counter using syncpoints */
+	return drm_vblank_count(dev, crtc);
+}
+
+static int tegra_drm_enable_vblank(struct drm_device *drm, int pipe)
+{
+	struct drm_crtc *crtc = tegra_crtc_from_pipe(drm, pipe);
+	struct tegra_dc *dc = to_tegra_dc(crtc);
+
+	if (!crtc)
+		return -ENODEV;
+
+	tegra_dc_enable_vblank(dc);
+
+	return 0;
+}
+
+static void tegra_drm_disable_vblank(struct drm_device *drm, int pipe)
+{
+	struct drm_crtc *crtc = tegra_crtc_from_pipe(drm, pipe);
+	struct tegra_dc *dc = to_tegra_dc(crtc);
+
+	if (crtc)
+		tegra_dc_disable_vblank(dc);
+}
+
+static void tegra_drm_preclose(struct drm_device *drm, struct drm_file *file)
+{
+	struct tegra_drm_file *fpriv = file->driver_priv;
+	struct tegra_drm_context *context, *tmp;
+	struct drm_crtc *crtc;
+
+	list_for_each_entry(crtc, &drm->mode_config.crtc_list, head)
+		tegra_dc_cancel_page_flip(crtc, file);
+
+	list_for_each_entry_safe(context, tmp, &fpriv->contexts, list)
+		tegra_drm_context_free(context);
+
+	kfree(fpriv);
+}
+
+#ifdef CONFIG_DEBUG_FS
+static int tegra_debugfs_framebuffers(struct seq_file *s, void *data)
+{
+	struct drm_info_node *node = (struct drm_info_node *)s->private;
+	struct drm_device *drm = node->minor->dev;
+	struct drm_framebuffer *fb;
+
+	mutex_lock(&drm->mode_config.fb_lock);
+
+	list_for_each_entry(fb, &drm->mode_config.fb_list, head) {
+		seq_printf(s, "%3d: user size: %d x %d, depth %d, %d bpp, refcount %d\n",
+			   fb->base.id, fb->width, fb->height, fb->depth,
+			   fb->bits_per_pixel,
+			   atomic_read(&fb->refcount.refcount));
+	}
+
+	mutex_unlock(&drm->mode_config.fb_lock);
+
+	return 0;
+}
+
+static struct drm_info_list tegra_debugfs_list[] = {
+	{ "framebuffers", tegra_debugfs_framebuffers, 0 },
+};
+
+static int tegra_debugfs_init(struct drm_minor *minor)
+{
+	return drm_debugfs_create_files(tegra_debugfs_list,
+					ARRAY_SIZE(tegra_debugfs_list),
+					minor->debugfs_root, minor);
+}
+
+static void tegra_debugfs_cleanup(struct drm_minor *minor)
+{
+	drm_debugfs_remove_files(tegra_debugfs_list,
+				 ARRAY_SIZE(tegra_debugfs_list), minor);
+}
+#endif
+
+struct drm_driver tegra_drm_driver = {
+	.driver_features = DRIVER_MODESET | DRIVER_GEM,
+	.load = tegra_drm_load,
+	.unload = tegra_drm_unload,
+	.open = tegra_drm_open,
+	.preclose = tegra_drm_preclose,
+	.lastclose = tegra_drm_lastclose,
+
+	.get_vblank_counter = tegra_drm_get_vblank_counter,
+	.enable_vblank = tegra_drm_enable_vblank,
+	.disable_vblank = tegra_drm_disable_vblank,
+
+#if defined(CONFIG_DEBUG_FS)
+	.debugfs_init = tegra_debugfs_init,
+	.debugfs_cleanup = tegra_debugfs_cleanup,
+#endif
+
+	.gem_free_object = tegra_bo_free_object,
+	.gem_vm_ops = &tegra_bo_vm_ops,
+	.dumb_create = tegra_bo_dumb_create,
+	.dumb_map_offset = tegra_bo_dumb_map_offset,
+	.dumb_destroy = drm_gem_dumb_destroy,
+
+	.ioctls = tegra_drm_ioctls,
+	.num_ioctls = ARRAY_SIZE(tegra_drm_ioctls),
+	.fops = &tegra_drm_fops,
+
+	.name = DRIVER_NAME,
+	.desc = DRIVER_DESC,
+	.date = DRIVER_DATE,
+	.major = DRIVER_MAJOR,
+	.minor = DRIVER_MINOR,
+	.patchlevel = DRIVER_PATCHLEVEL,
+};
+
+int tegra_drm_register_client(struct tegra_drm *tegra,
+			      struct tegra_drm_client *client)
+{
+	mutex_lock(&tegra->clients_lock);
+	list_add_tail(&client->list, &tegra->clients);
+	mutex_unlock(&tegra->clients_lock);
+
+	return 0;
+}
+
+int tegra_drm_unregister_client(struct tegra_drm *tegra,
+				struct tegra_drm_client *client)
+{
+	mutex_lock(&tegra->clients_lock);
+	list_del_init(&client->list);
+	mutex_unlock(&tegra->clients_lock);
+
+	return 0;
+}
+
+static int host1x_drm_probe(struct host1x_device *device)
+{
+	return drm_host1x_init(&tegra_drm_driver, device);
+}
+
+static int host1x_drm_remove(struct host1x_device *device)
+{
+	drm_host1x_exit(&tegra_drm_driver, device);
+
+	return 0;
+}
+
+static const struct of_device_id host1x_drm_subdevs[] = {
+	{ .compatible = "nvidia,tegra20-dc", },
+	{ .compatible = "nvidia,tegra20-hdmi", },
+	{ .compatible = "nvidia,tegra20-gr2d", },
+	{ .compatible = "nvidia,tegra20-gr3d", },
+	{ .compatible = "nvidia,tegra30-dc", },
+	{ .compatible = "nvidia,tegra30-hdmi", },
+	{ .compatible = "nvidia,tegra30-gr2d", },
+	{ .compatible = "nvidia,tegra30-gr3d", },
+	{ .compatible = "nvidia,tegra114-hdmi", },
+	{ .compatible = "nvidia,tegra114-gr3d", },
+	{ /* sentinel */ }
+};
+
+static struct host1x_driver host1x_drm_driver = {
+	.name = "drm",
+	.probe = host1x_drm_probe,
+	.remove = host1x_drm_remove,
+	.subdevs = host1x_drm_subdevs,
+};
+
+static int __init host1x_drm_init(void)
+{
+	int err;
+
+	err = host1x_driver_register(&host1x_drm_driver);
+	if (err < 0)
+		return err;
+
+	err = platform_driver_register(&tegra_dc_driver);
+	if (err < 0)
+		goto unregister_host1x;
+
+	err = platform_driver_register(&tegra_hdmi_driver);
+	if (err < 0)
+		goto unregister_dc;
+
+	err = platform_driver_register(&tegra_gr2d_driver);
+	if (err < 0)
+		goto unregister_hdmi;
+
+	err = platform_driver_register(&tegra_gr3d_driver);
+	if (err < 0)
+		goto unregister_gr2d;
+
+	return 0;
+
+unregister_gr2d:
+	platform_driver_unregister(&tegra_gr2d_driver);
+unregister_hdmi:
+	platform_driver_unregister(&tegra_hdmi_driver);
+unregister_dc:
+	platform_driver_unregister(&tegra_dc_driver);
+unregister_host1x:
+	host1x_driver_unregister(&host1x_drm_driver);
+	return err;
+}
+module_init(host1x_drm_init);
+
+static void __exit host1x_drm_exit(void)
+{
+	platform_driver_unregister(&tegra_gr3d_driver);
+	platform_driver_unregister(&tegra_gr2d_driver);
+	platform_driver_unregister(&tegra_hdmi_driver);
+	platform_driver_unregister(&tegra_dc_driver);
+	host1x_driver_unregister(&host1x_drm_driver);
+}
+module_exit(host1x_drm_exit);
+
+MODULE_AUTHOR("Thierry Reding <thierry.reding@avionic-design.de>");
+MODULE_DESCRIPTION("NVIDIA Tegra DRM driver");
+MODULE_LICENSE("GPL v2");
diff --git a/drivers/gpu/host1x/drm/drm.h b/drivers/gpu/drm/tegra/drm.h
index 02ce020f2575..fdfe259ed7f8 100644
--- a/drivers/gpu/host1x/drm/drm.h
+++ b/drivers/gpu/drm/tegra/drm.h
@@ -10,14 +10,14 @@
 #ifndef HOST1X_DRM_H
 #define HOST1X_DRM_H 1
 
+#include <uapi/drm/tegra_drm.h>
+#include <linux/host1x.h>
+
 #include <drm/drmP.h>
 #include <drm/drm_crtc_helper.h>
 #include <drm/drm_edid.h>
 #include <drm/drm_fb_helper.h>
 #include <drm/drm_fixed.h>
-#include <uapi/drm/tegra_drm.h>
-
-#include "host1x.h"
 
 struct tegra_fb {
 	struct drm_framebuffer base;
@@ -30,17 +30,8 @@ struct tegra_fbdev {
 	struct tegra_fb *fb;
 };
 
-struct host1x_drm {
+struct tegra_drm {
 	struct drm_device *drm;
-	struct device *dev;
-	void __iomem *regs;
-	struct clk *clk;
-	int syncpt;
-	int irq;
-
-	struct mutex drm_clients_lock;
-	struct list_head drm_clients;
-	struct list_head drm_active;
 
 	struct mutex clients_lock;
 	struct list_head clients;
@@ -48,66 +39,60 @@ struct host1x_drm {
 	struct tegra_fbdev *fbdev;
 };
 
-struct host1x_client;
+struct tegra_drm_client;
 
-struct host1x_drm_context {
-	struct host1x_client *client;
+struct tegra_drm_context {
+	struct tegra_drm_client *client;
 	struct host1x_channel *channel;
 	struct list_head list;
 };
 
-struct host1x_client_ops {
-	int (*drm_init)(struct host1x_client *client, struct drm_device *drm);
-	int (*drm_exit)(struct host1x_client *client);
-	int (*open_channel)(struct host1x_client *client,
-			    struct host1x_drm_context *context);
-	void (*close_channel)(struct host1x_drm_context *context);
-	int (*submit)(struct host1x_drm_context *context,
+struct tegra_drm_client_ops {
+	int (*open_channel)(struct tegra_drm_client *client,
+			    struct tegra_drm_context *context);
+	void (*close_channel)(struct tegra_drm_context *context);
+	int (*is_addr_reg)(struct device *dev, u32 class, u32 offset);
+	int (*submit)(struct tegra_drm_context *context,
 		      struct drm_tegra_submit *args, struct drm_device *drm,
 		      struct drm_file *file);
 };
 
-struct host1x_drm_file {
-	struct list_head contexts;
-};
-
-struct host1x_client {
-	struct host1x_drm *host1x;
-	struct device *dev;
-
-	const struct host1x_client_ops *ops;
-
-	enum host1x_class class;
-	struct host1x_channel *channel;
-
-	struct host1x_syncpt **syncpts;
-	unsigned int num_syncpts;
+int tegra_drm_submit(struct tegra_drm_context *context,
+		     struct drm_tegra_submit *args, struct drm_device *drm,
+		     struct drm_file *file);
 
+struct tegra_drm_client {
+	struct host1x_client base;
 	struct list_head list;
+
+	const struct tegra_drm_client_ops *ops;
 };
 
-extern int host1x_drm_init(struct host1x_drm *host1x, struct drm_device *drm);
-extern int host1x_drm_exit(struct host1x_drm *host1x);
+static inline struct tegra_drm_client *
+host1x_to_drm_client(struct host1x_client *client)
+{
+	return container_of(client, struct tegra_drm_client, base);
+}
+
+extern int tegra_drm_register_client(struct tegra_drm *tegra,
+				     struct tegra_drm_client *client);
+extern int tegra_drm_unregister_client(struct tegra_drm *tegra,
+				       struct tegra_drm_client *client);
 
-extern int host1x_register_client(struct host1x_drm *host1x,
-				  struct host1x_client *client);
-extern int host1x_unregister_client(struct host1x_drm *host1x,
-				    struct host1x_client *client);
+extern int tegra_drm_init(struct tegra_drm *tegra, struct drm_device *drm);
+extern int tegra_drm_exit(struct tegra_drm *tegra);
 
 struct tegra_output;
 
 struct tegra_dc {
 	struct host1x_client client;
-	spinlock_t lock;
-
-	struct host1x_drm *host1x;
 	struct device *dev;
+	spinlock_t lock;
 
 	struct drm_crtc base;
 	int pipe;
 
 	struct clk *clk;
-
 	void __iomem *regs;
 	int irq;
 
@@ -123,7 +108,8 @@ struct tegra_dc {
 	struct drm_pending_vblank_event *event;
 };
 
-static inline struct tegra_dc *host1x_client_to_dc(struct host1x_client *client)
+static inline struct tegra_dc *
+host1x_client_to_dc(struct host1x_client *client)
 {
 	return container_of(client, struct tegra_dc, client);
 }
@@ -162,6 +148,8 @@ struct tegra_dc_window {
 	unsigned int format;
 	unsigned int stride[2];
 	unsigned long base[3];
+	bool bottom_up;
+	bool tiled;
 };
 
 /* from dc.c */
@@ -249,23 +237,34 @@ static inline int tegra_output_check_mode(struct tegra_output *output,
 	return output ? -ENOSYS : -EINVAL;
 }
 
+/* from bus.c */
+int drm_host1x_init(struct drm_driver *driver, struct host1x_device *device);
+void drm_host1x_exit(struct drm_driver *driver, struct host1x_device *device);
+
 /* from rgb.c */
 extern int tegra_dc_rgb_probe(struct tegra_dc *dc);
+extern int tegra_dc_rgb_remove(struct tegra_dc *dc);
 extern int tegra_dc_rgb_init(struct drm_device *drm, struct tegra_dc *dc);
 extern int tegra_dc_rgb_exit(struct tegra_dc *dc);
 
 /* from output.c */
-extern int tegra_output_parse_dt(struct tegra_output *output);
+extern int tegra_output_probe(struct tegra_output *output);
+extern int tegra_output_remove(struct tegra_output *output);
 extern int tegra_output_init(struct drm_device *drm, struct tegra_output *output);
 extern int tegra_output_exit(struct tegra_output *output);
 
 /* from fb.c */
 struct tegra_bo *tegra_fb_get_plane(struct drm_framebuffer *framebuffer,
 				    unsigned int index);
+bool tegra_fb_is_bottom_up(struct drm_framebuffer *framebuffer);
+bool tegra_fb_is_tiled(struct drm_framebuffer *framebuffer);
 extern int tegra_drm_fb_init(struct drm_device *drm);
 extern void tegra_drm_fb_exit(struct drm_device *drm);
 extern void tegra_fbdev_restore_mode(struct tegra_fbdev *fbdev);
 
-extern struct drm_driver tegra_drm_driver;
+extern struct platform_driver tegra_dc_driver;
+extern struct platform_driver tegra_hdmi_driver;
+extern struct platform_driver tegra_gr2d_driver;
+extern struct platform_driver tegra_gr3d_driver;
 
 #endif /* HOST1X_DRM_H */
diff --git a/drivers/gpu/host1x/drm/fb.c b/drivers/gpu/drm/tegra/fb.c
index 979a3e32b78b..490f7719e317 100644
--- a/drivers/gpu/host1x/drm/fb.c
+++ b/drivers/gpu/drm/tegra/fb.c
@@ -10,8 +10,6 @@
  * published by the Free Software Foundation.
  */
 
-#include <linux/module.h>
-
 #include "drm.h"
 #include "gem.h"
 
@@ -36,6 +34,26 @@ struct tegra_bo *tegra_fb_get_plane(struct drm_framebuffer *framebuffer,
 	return fb->planes[index];
 }
 
+bool tegra_fb_is_bottom_up(struct drm_framebuffer *framebuffer)
+{
+	struct tegra_fb *fb = to_tegra_fb(framebuffer);
+
+	if (fb->planes[0]->flags & TEGRA_BO_BOTTOM_UP)
+		return true;
+
+	return false;
+}
+
+bool tegra_fb_is_tiled(struct drm_framebuffer *framebuffer)
+{
+	struct tegra_fb *fb = to_tegra_fb(framebuffer);
+
+	if (fb->planes[0]->flags & TEGRA_BO_TILED)
+		return true;
+
+	return false;
+}
+
 static void tegra_fb_destroy(struct drm_framebuffer *framebuffer)
 {
 	struct tegra_fb *fb = to_tegra_fb(framebuffer);
@@ -190,7 +208,7 @@ static int tegra_fbdev_probe(struct drm_fb_helper *helper,
 
 	size = cmd.pitches[0] * cmd.height;
 
-	bo = tegra_bo_create(drm, size);
+	bo = tegra_bo_create(drm, size, 0);
 	if (IS_ERR(bo))
 		return PTR_ERR(bo);
 
@@ -323,10 +341,10 @@ static void tegra_fbdev_free(struct tegra_fbdev *fbdev)
 
 static void tegra_fb_output_poll_changed(struct drm_device *drm)
 {
-	struct host1x_drm *host1x = drm->dev_private;
+	struct tegra_drm *tegra = drm->dev_private;
 
-	if (host1x->fbdev)
-		drm_fb_helper_hotplug_event(&host1x->fbdev->base);
+	if (tegra->fbdev)
+		drm_fb_helper_hotplug_event(&tegra->fbdev->base);
 }
 
 static const struct drm_mode_config_funcs tegra_drm_mode_funcs = {
@@ -336,7 +354,7 @@ static const struct drm_mode_config_funcs tegra_drm_mode_funcs = {
 
 int tegra_drm_fb_init(struct drm_device *drm)
 {
-	struct host1x_drm *host1x = drm->dev_private;
+	struct tegra_drm *tegra = drm->dev_private;
 	struct tegra_fbdev *fbdev;
 
 	drm->mode_config.min_width = 0;
@@ -352,16 +370,16 @@ int tegra_drm_fb_init(struct drm_device *drm)
 	if (IS_ERR(fbdev))
 		return PTR_ERR(fbdev);
 
-	host1x->fbdev = fbdev;
+	tegra->fbdev = fbdev;
 
 	return 0;
 }
 
 void tegra_drm_fb_exit(struct drm_device *drm)
 {
-	struct host1x_drm *host1x = drm->dev_private;
+	struct tegra_drm *tegra = drm->dev_private;
 
-	tegra_fbdev_free(host1x->fbdev);
+	tegra_fbdev_free(tegra->fbdev);
 }
 
 void tegra_fbdev_restore_mode(struct tegra_fbdev *fbdev)
diff --git a/drivers/gpu/host1x/drm/gem.c b/drivers/gpu/drm/tegra/gem.c
index 59623de4ee15..28a9cbc07ab9 100644
--- a/drivers/gpu/host1x/drm/gem.c
+++ b/drivers/gpu/drm/tegra/gem.c
@@ -18,25 +18,18 @@
  * GNU General Public License for more details.
  */
 
-#include <linux/mm.h>
-#include <linux/slab.h>
-#include <linux/mutex.h>
-#include <linux/export.h>
-#include <linux/dma-mapping.h>
-
-#include <drm/drmP.h>
-#include <drm/drm.h>
+#include <drm/tegra_drm.h>
 
 #include "gem.h"
 
-static inline struct tegra_bo *host1x_to_drm_bo(struct host1x_bo *bo)
+static inline struct tegra_bo *host1x_to_tegra_bo(struct host1x_bo *bo)
 {
 	return container_of(bo, struct tegra_bo, base);
 }
 
 static void tegra_bo_put(struct host1x_bo *bo)
 {
-	struct tegra_bo *obj = host1x_to_drm_bo(bo);
+	struct tegra_bo *obj = host1x_to_tegra_bo(bo);
 	struct drm_device *drm = obj->gem.dev;
 
 	mutex_lock(&drm->struct_mutex);
@@ -46,7 +39,7 @@ static void tegra_bo_put(struct host1x_bo *bo)
 
 static dma_addr_t tegra_bo_pin(struct host1x_bo *bo, struct sg_table **sgt)
 {
-	struct tegra_bo *obj = host1x_to_drm_bo(bo);
+	struct tegra_bo *obj = host1x_to_tegra_bo(bo);
 
 	return obj->paddr;
 }
@@ -57,7 +50,7 @@ static void tegra_bo_unpin(struct host1x_bo *bo, struct sg_table *sgt)
 
 static void *tegra_bo_mmap(struct host1x_bo *bo)
 {
-	struct tegra_bo *obj = host1x_to_drm_bo(bo);
+	struct tegra_bo *obj = host1x_to_tegra_bo(bo);
 
 	return obj->vaddr;
 }
@@ -68,7 +61,7 @@ static void tegra_bo_munmap(struct host1x_bo *bo, void *addr)
 
 static void *tegra_bo_kmap(struct host1x_bo *bo, unsigned int page)
 {
-	struct tegra_bo *obj = host1x_to_drm_bo(bo);
+	struct tegra_bo *obj = host1x_to_tegra_bo(bo);
 
 	return obj->vaddr + page * PAGE_SIZE;
 }
@@ -80,7 +73,7 @@ static void tegra_bo_kunmap(struct host1x_bo *bo, unsigned int page,
 
 static struct host1x_bo *tegra_bo_get(struct host1x_bo *bo)
 {
-	struct tegra_bo *obj = host1x_to_drm_bo(bo);
+	struct tegra_bo *obj = host1x_to_tegra_bo(bo);
 	struct drm_device *drm = obj->gem.dev;
 
 	mutex_lock(&drm->struct_mutex);
@@ -106,7 +99,8 @@ static void tegra_bo_destroy(struct drm_device *drm, struct tegra_bo *bo)
 	dma_free_writecombine(drm->dev, bo->gem.size, bo->vaddr, bo->paddr);
 }
 
-struct tegra_bo *tegra_bo_create(struct drm_device *drm, unsigned int size)
+struct tegra_bo *tegra_bo_create(struct drm_device *drm, unsigned int size,
+				 unsigned long flags)
 {
 	struct tegra_bo *bo;
 	int err;
@@ -135,6 +129,12 @@ struct tegra_bo *tegra_bo_create(struct drm_device *drm, unsigned int size)
 	if (err)
 		goto err_mmap;
 
+	if (flags & DRM_TEGRA_GEM_CREATE_TILED)
+		bo->flags |= TEGRA_BO_TILED;
+
+	if (flags & DRM_TEGRA_GEM_CREATE_BOTTOM_UP)
+		bo->flags |= TEGRA_BO_BOTTOM_UP;
+
 	return bo;
 
 err_mmap:
@@ -149,14 +149,15 @@ err_dma:
 }
 
 struct tegra_bo *tegra_bo_create_with_handle(struct drm_file *file,
-					    struct drm_device *drm,
-					    unsigned int size,
-					    unsigned int *handle)
+					     struct drm_device *drm,
+					     unsigned int size,
+					     unsigned long flags,
+					     unsigned int *handle)
 {
 	struct tegra_bo *bo;
 	int ret;
 
-	bo = tegra_bo_create(drm, size);
+	bo = tegra_bo_create(drm, size, flags);
 	if (IS_ERR(bo))
 		return bo;
 
@@ -178,7 +179,6 @@ void tegra_bo_free_object(struct drm_gem_object *gem)
 	struct tegra_bo *bo = to_tegra_bo(gem);
 
 	drm_gem_free_mmap_offset(gem);
-
 	drm_gem_object_release(gem);
 	tegra_bo_destroy(gem->dev, bo);
 
@@ -197,8 +197,8 @@ int tegra_bo_dumb_create(struct drm_file *file, struct drm_device *drm,
 	if (args->size < args->pitch * args->height)
 		args->size = args->pitch * args->height;
 
-	bo = tegra_bo_create_with_handle(file, drm, args->size,
-					    &args->handle);
+	bo = tegra_bo_create_with_handle(file, drm, args->size, 0,
+					 &args->handle);
 	if (IS_ERR(bo))
 		return PTR_ERR(bo);
 
diff --git a/drivers/gpu/host1x/drm/gem.h b/drivers/gpu/drm/tegra/gem.h
index 492533a2dacb..7674000bf47d 100644
--- a/drivers/gpu/host1x/drm/gem.h
+++ b/drivers/gpu/drm/tegra/gem.h
@@ -19,14 +19,18 @@
 #ifndef __HOST1X_GEM_H
 #define __HOST1X_GEM_H
 
+#include <linux/host1x.h>
+
 #include <drm/drm.h>
 #include <drm/drmP.h>
 
-#include "host1x_bo.h"
+#define TEGRA_BO_TILED     (1 << 0)
+#define TEGRA_BO_BOTTOM_UP (1 << 1)
 
 struct tegra_bo {
 	struct drm_gem_object gem;
 	struct host1x_bo base;
+	unsigned long flags;
 	dma_addr_t paddr;
 	void *vaddr;
 };
@@ -38,11 +42,13 @@ static inline struct tegra_bo *to_tegra_bo(struct drm_gem_object *gem)
 
 extern const struct host1x_bo_ops tegra_bo_ops;
 
-struct tegra_bo *tegra_bo_create(struct drm_device *drm, unsigned int size);
+struct tegra_bo *tegra_bo_create(struct drm_device *drm, unsigned int size,
+				 unsigned long flags);
 struct tegra_bo *tegra_bo_create_with_handle(struct drm_file *file,
-					    struct drm_device *drm,
-					    unsigned int size,
-					    unsigned int *handle);
+					     struct drm_device *drm,
+					     unsigned int size,
+					     unsigned long flags,
+					     unsigned int *handle);
 void tegra_bo_free_object(struct drm_gem_object *gem);
 int tegra_bo_dumb_create(struct drm_file *file, struct drm_device *drm,
 			 struct drm_mode_create_dumb *args);
diff --git a/drivers/gpu/drm/tegra/gr2d.c b/drivers/gpu/drm/tegra/gr2d.c
new file mode 100644
index 000000000000..7ec4259ffded
--- /dev/null
+++ b/drivers/gpu/drm/tegra/gr2d.c
@@ -0,0 +1,227 @@
+/*
+ * Copyright (c) 2012-2013, NVIDIA Corporation.
+ *
+ * This program is free software; you can redistribute it and/or modify it
+ * under the terms and conditions of the GNU General Public License,
+ * version 2, as published by the Free Software Foundation.
+ *
+ * This program is distributed in the hope it will be useful, but WITHOUT
+ * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or
+ * FITNESS FOR A PARTICULAR PURPOSE.  See the GNU General Public License for
+ * more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program.  If not, see <http://www.gnu.org/licenses/>.
+ */
+
+#include <linux/clk.h>
+
+#include "drm.h"
+#include "gem.h"
+#include "gr2d.h"
+
+struct gr2d {
+	struct tegra_drm_client client;
+	struct host1x_channel *channel;
+	struct clk *clk;
+
+	DECLARE_BITMAP(addr_regs, GR2D_NUM_REGS);
+};
+
+static inline struct gr2d *to_gr2d(struct tegra_drm_client *client)
+{
+	return container_of(client, struct gr2d, client);
+}
+
+static int gr2d_init(struct host1x_client *client)
+{
+	struct tegra_drm_client *drm = host1x_to_drm_client(client);
+	struct tegra_drm *tegra = dev_get_drvdata(client->parent);
+	unsigned long flags = HOST1X_SYNCPT_HAS_BASE;
+	struct gr2d *gr2d = to_gr2d(drm);
+
+	gr2d->channel = host1x_channel_request(client->dev);
+	if (!gr2d->channel)
+		return -ENOMEM;
+
+	client->syncpts[0] = host1x_syncpt_request(client->dev, flags);
+	if (!client->syncpts[0]) {
+		host1x_channel_free(gr2d->channel);
+		return -ENOMEM;
+	}
+
+	return tegra_drm_register_client(tegra, drm);
+}
+
+static int gr2d_exit(struct host1x_client *client)
+{
+	struct tegra_drm_client *drm = host1x_to_drm_client(client);
+	struct tegra_drm *tegra = dev_get_drvdata(client->parent);
+	struct gr2d *gr2d = to_gr2d(drm);
+	int err;
+
+	err = tegra_drm_unregister_client(tegra, drm);
+	if (err < 0)
+		return err;
+
+	host1x_syncpt_free(client->syncpts[0]);
+	host1x_channel_free(gr2d->channel);
+
+	return 0;
+}
+
+static const struct host1x_client_ops gr2d_client_ops = {
+	.init = gr2d_init,
+	.exit = gr2d_exit,
+};
+
+static int gr2d_open_channel(struct tegra_drm_client *client,
+			     struct tegra_drm_context *context)
+{
+	struct gr2d *gr2d = to_gr2d(client);
+
+	context->channel = host1x_channel_get(gr2d->channel);
+	if (!context->channel)
+		return -ENOMEM;
+
+	return 0;
+}
+
+static void gr2d_close_channel(struct tegra_drm_context *context)
+{
+	host1x_channel_put(context->channel);
+}
+
+static int gr2d_is_addr_reg(struct device *dev, u32 class, u32 offset)
+{
+	struct gr2d *gr2d = dev_get_drvdata(dev);
+
+	switch (class) {
+	case HOST1X_CLASS_HOST1X:
+		if (offset == 0x2b)
+			return 1;
+
+		break;
+
+	case HOST1X_CLASS_GR2D:
+	case HOST1X_CLASS_GR2D_SB:
+		if (offset >= GR2D_NUM_REGS)
+			break;
+
+		if (test_bit(offset, gr2d->addr_regs))
+			return 1;
+
+		break;
+	}
+
+	return 0;
+}
+
+static const struct tegra_drm_client_ops gr2d_ops = {
+	.open_channel = gr2d_open_channel,
+	.close_channel = gr2d_close_channel,
+	.is_addr_reg = gr2d_is_addr_reg,
+	.submit = tegra_drm_submit,
+};
+
+static const struct of_device_id gr2d_match[] = {
+	{ .compatible = "nvidia,tegra30-gr2d" },
+	{ .compatible = "nvidia,tegra20-gr2d" },
+	{ },
+};
+
+static const u32 gr2d_addr_regs[] = {
+	GR2D_UA_BASE_ADDR,
+	GR2D_VA_BASE_ADDR,
+	GR2D_PAT_BASE_ADDR,
+	GR2D_DSTA_BASE_ADDR,
+	GR2D_DSTB_BASE_ADDR,
+	GR2D_DSTC_BASE_ADDR,
+	GR2D_SRCA_BASE_ADDR,
+	GR2D_SRCB_BASE_ADDR,
+	GR2D_SRC_BASE_ADDR_SB,
+	GR2D_DSTA_BASE_ADDR_SB,
+	GR2D_DSTB_BASE_ADDR_SB,
+	GR2D_UA_BASE_ADDR_SB,
+	GR2D_VA_BASE_ADDR_SB,
+};
+
+static int gr2d_probe(struct platform_device *pdev)
+{
+	struct device *dev = &pdev->dev;
+	struct host1x_syncpt **syncpts;
+	struct gr2d *gr2d;
+	unsigned int i;
+	int err;
+
+	gr2d = devm_kzalloc(dev, sizeof(*gr2d), GFP_KERNEL);
+	if (!gr2d)
+		return -ENOMEM;
+
+	syncpts = devm_kzalloc(dev, sizeof(*syncpts), GFP_KERNEL);
+	if (!syncpts)
+		return -ENOMEM;
+
+	gr2d->clk = devm_clk_get(dev, NULL);
+	if (IS_ERR(gr2d->clk)) {
+		dev_err(dev, "cannot get clock\n");
+		return PTR_ERR(gr2d->clk);
+	}
+
+	err = clk_prepare_enable(gr2d->clk);
+	if (err) {
+		dev_err(dev, "cannot turn on clock\n");
+		return err;
+	}
+
+	INIT_LIST_HEAD(&gr2d->client.base.list);
+	gr2d->client.base.ops = &gr2d_client_ops;
+	gr2d->client.base.dev = dev;
+	gr2d->client.base.class = HOST1X_CLASS_GR2D;
+	gr2d->client.base.syncpts = syncpts;
+	gr2d->client.base.num_syncpts = 1;
+
+	INIT_LIST_HEAD(&gr2d->client.list);
+	gr2d->client.ops = &gr2d_ops;
+
+	err = host1x_client_register(&gr2d->client.base);
+	if (err < 0) {
+		dev_err(dev, "failed to register host1x client: %d\n", err);
+		clk_disable_unprepare(gr2d->clk);
+		return err;
+	}
+
+	/* initialize address register map */
+	for (i = 0; i < ARRAY_SIZE(gr2d_addr_regs); i++)
+		set_bit(gr2d_addr_regs[i], gr2d->addr_regs);
+
+	platform_set_drvdata(pdev, gr2d);
+
+	return 0;
+}
+
+static int gr2d_remove(struct platform_device *pdev)
+{
+	struct gr2d *gr2d = platform_get_drvdata(pdev);
+	int err;
+
+	err = host1x_client_unregister(&gr2d->client.base);
+	if (err < 0) {
+		dev_err(&pdev->dev, "failed to unregister host1x client: %d\n",
+			err);
+		return err;
+	}
+
+	clk_disable_unprepare(gr2d->clk);
+
+	return 0;
+}
+
+struct platform_driver tegra_gr2d_driver = {
+	.driver = {
+		.name = "tegra-gr2d",
+		.of_match_table = gr2d_match,
+	},
+	.probe = gr2d_probe,
+	.remove = gr2d_remove,
+};
diff --git a/drivers/gpu/drm/tegra/gr2d.h b/drivers/gpu/drm/tegra/gr2d.h
new file mode 100644
index 000000000000..4d7304fb015e
--- /dev/null
+++ b/drivers/gpu/drm/tegra/gr2d.h
@@ -0,0 +1,28 @@
+/*
+ * Copyright (C) 2013 NVIDIA Corporation
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ */
+
+#ifndef TEGRA_GR2D_H
+#define TEGRA_GR2D_H
+
+#define GR2D_UA_BASE_ADDR		0x1a
+#define GR2D_VA_BASE_ADDR		0x1b
+#define GR2D_PAT_BASE_ADDR		0x26
+#define GR2D_DSTA_BASE_ADDR		0x2b
+#define GR2D_DSTB_BASE_ADDR		0x2c
+#define GR2D_DSTC_BASE_ADDR		0x2d
+#define GR2D_SRCA_BASE_ADDR		0x31
+#define GR2D_SRCB_BASE_ADDR		0x32
+#define GR2D_SRC_BASE_ADDR_SB		0x48
+#define GR2D_DSTA_BASE_ADDR_SB		0x49
+#define GR2D_DSTB_BASE_ADDR_SB		0x4a
+#define GR2D_UA_BASE_ADDR_SB		0x4b
+#define GR2D_VA_BASE_ADDR_SB		0x4c
+
+#define GR2D_NUM_REGS			0x4d
+
+#endif
diff --git a/drivers/gpu/drm/tegra/gr3d.c b/drivers/gpu/drm/tegra/gr3d.c
new file mode 100644
index 000000000000..4cec8f526af7
--- /dev/null
+++ b/drivers/gpu/drm/tegra/gr3d.c
@@ -0,0 +1,338 @@
+/*
+ * Copyright (C) 2013 Avionic Design GmbH
+ * Copyright (C) 2013 NVIDIA Corporation
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ */
+
+#include <linux/clk.h>
+#include <linux/host1x.h>
+#include <linux/module.h>
+#include <linux/platform_device.h>
+#include <linux/tegra-powergate.h>
+
+#include "drm.h"
+#include "gem.h"
+#include "gr3d.h"
+
+struct gr3d {
+	struct tegra_drm_client client;
+	struct host1x_channel *channel;
+	struct clk *clk_secondary;
+	struct clk *clk;
+
+	DECLARE_BITMAP(addr_regs, GR3D_NUM_REGS);
+};
+
+static inline struct gr3d *to_gr3d(struct tegra_drm_client *client)
+{
+	return container_of(client, struct gr3d, client);
+}
+
+static int gr3d_init(struct host1x_client *client)
+{
+	struct tegra_drm_client *drm = host1x_to_drm_client(client);
+	struct tegra_drm *tegra = dev_get_drvdata(client->parent);
+	unsigned long flags = HOST1X_SYNCPT_HAS_BASE;
+	struct gr3d *gr3d = to_gr3d(drm);
+
+	gr3d->channel = host1x_channel_request(client->dev);
+	if (!gr3d->channel)
+		return -ENOMEM;
+
+	client->syncpts[0] = host1x_syncpt_request(client->dev, flags);
+	if (!client->syncpts[0]) {
+		host1x_channel_free(gr3d->channel);
+		return -ENOMEM;
+	}
+
+	return tegra_drm_register_client(tegra, drm);
+}
+
+static int gr3d_exit(struct host1x_client *client)
+{
+	struct tegra_drm_client *drm = host1x_to_drm_client(client);
+	struct tegra_drm *tegra = dev_get_drvdata(client->parent);
+	struct gr3d *gr3d = to_gr3d(drm);
+	int err;
+
+	err = tegra_drm_unregister_client(tegra, drm);
+	if (err < 0)
+		return err;
+
+	host1x_syncpt_free(client->syncpts[0]);
+	host1x_channel_free(gr3d->channel);
+
+	return 0;
+}
+
+static const struct host1x_client_ops gr3d_client_ops = {
+	.init = gr3d_init,
+	.exit = gr3d_exit,
+};
+
+static int gr3d_open_channel(struct tegra_drm_client *client,
+			     struct tegra_drm_context *context)
+{
+	struct gr3d *gr3d = to_gr3d(client);
+
+	context->channel = host1x_channel_get(gr3d->channel);
+	if (!context->channel)
+		return -ENOMEM;
+
+	return 0;
+}
+
+static void gr3d_close_channel(struct tegra_drm_context *context)
+{
+	host1x_channel_put(context->channel);
+}
+
+static int gr3d_is_addr_reg(struct device *dev, u32 class, u32 offset)
+{
+	struct gr3d *gr3d = dev_get_drvdata(dev);
+
+	switch (class) {
+	case HOST1X_CLASS_HOST1X:
+		if (offset == 0x2b)
+			return 1;
+
+		break;
+
+	case HOST1X_CLASS_GR3D:
+		if (offset >= GR3D_NUM_REGS)
+			break;
+
+		if (test_bit(offset, gr3d->addr_regs))
+			return 1;
+
+		break;
+	}
+
+	return 0;
+}
+
+static const struct tegra_drm_client_ops gr3d_ops = {
+	.open_channel = gr3d_open_channel,
+	.close_channel = gr3d_close_channel,
+	.is_addr_reg = gr3d_is_addr_reg,
+	.submit = tegra_drm_submit,
+};
+
+static const struct of_device_id tegra_gr3d_match[] = {
+	{ .compatible = "nvidia,tegra114-gr3d" },
+	{ .compatible = "nvidia,tegra30-gr3d" },
+	{ .compatible = "nvidia,tegra20-gr3d" },
+	{ }
+};
+
+static const u32 gr3d_addr_regs[] = {
+	GR3D_IDX_ATTRIBUTE( 0),
+	GR3D_IDX_ATTRIBUTE( 1),
+	GR3D_IDX_ATTRIBUTE( 2),
+	GR3D_IDX_ATTRIBUTE( 3),
+	GR3D_IDX_ATTRIBUTE( 4),
+	GR3D_IDX_ATTRIBUTE( 5),
+	GR3D_IDX_ATTRIBUTE( 6),
+	GR3D_IDX_ATTRIBUTE( 7),
+	GR3D_IDX_ATTRIBUTE( 8),
+	GR3D_IDX_ATTRIBUTE( 9),
+	GR3D_IDX_ATTRIBUTE(10),
+	GR3D_IDX_ATTRIBUTE(11),
+	GR3D_IDX_ATTRIBUTE(12),
+	GR3D_IDX_ATTRIBUTE(13),
+	GR3D_IDX_ATTRIBUTE(14),
+	GR3D_IDX_ATTRIBUTE(15),
+	GR3D_IDX_INDEX_BASE,
+	GR3D_QR_ZTAG_ADDR,
+	GR3D_QR_CTAG_ADDR,
+	GR3D_QR_CZ_ADDR,
+	GR3D_TEX_TEX_ADDR( 0),
+	GR3D_TEX_TEX_ADDR( 1),
+	GR3D_TEX_TEX_ADDR( 2),
+	GR3D_TEX_TEX_ADDR( 3),
+	GR3D_TEX_TEX_ADDR( 4),
+	GR3D_TEX_TEX_ADDR( 5),
+	GR3D_TEX_TEX_ADDR( 6),
+	GR3D_TEX_TEX_ADDR( 7),
+	GR3D_TEX_TEX_ADDR( 8),
+	GR3D_TEX_TEX_ADDR( 9),
+	GR3D_TEX_TEX_ADDR(10),
+	GR3D_TEX_TEX_ADDR(11),
+	GR3D_TEX_TEX_ADDR(12),
+	GR3D_TEX_TEX_ADDR(13),
+	GR3D_TEX_TEX_ADDR(14),
+	GR3D_TEX_TEX_ADDR(15),
+	GR3D_DW_MEMORY_OUTPUT_ADDRESS,
+	GR3D_GLOBAL_SURFADDR( 0),
+	GR3D_GLOBAL_SURFADDR( 1),
+	GR3D_GLOBAL_SURFADDR( 2),
+	GR3D_GLOBAL_SURFADDR( 3),
+	GR3D_GLOBAL_SURFADDR( 4),
+	GR3D_GLOBAL_SURFADDR( 5),
+	GR3D_GLOBAL_SURFADDR( 6),
+	GR3D_GLOBAL_SURFADDR( 7),
+	GR3D_GLOBAL_SURFADDR( 8),
+	GR3D_GLOBAL_SURFADDR( 9),
+	GR3D_GLOBAL_SURFADDR(10),
+	GR3D_GLOBAL_SURFADDR(11),
+	GR3D_GLOBAL_SURFADDR(12),
+	GR3D_GLOBAL_SURFADDR(13),
+	GR3D_GLOBAL_SURFADDR(14),
+	GR3D_GLOBAL_SURFADDR(15),
+	GR3D_GLOBAL_SPILLSURFADDR,
+	GR3D_GLOBAL_SURFOVERADDR( 0),
+	GR3D_GLOBAL_SURFOVERADDR( 1),
+	GR3D_GLOBAL_SURFOVERADDR( 2),
+	GR3D_GLOBAL_SURFOVERADDR( 3),
+	GR3D_GLOBAL_SURFOVERADDR( 4),
+	GR3D_GLOBAL_SURFOVERADDR( 5),
+	GR3D_GLOBAL_SURFOVERADDR( 6),
+	GR3D_GLOBAL_SURFOVERADDR( 7),
+	GR3D_GLOBAL_SURFOVERADDR( 8),
+	GR3D_GLOBAL_SURFOVERADDR( 9),
+	GR3D_GLOBAL_SURFOVERADDR(10),
+	GR3D_GLOBAL_SURFOVERADDR(11),
+	GR3D_GLOBAL_SURFOVERADDR(12),
+	GR3D_GLOBAL_SURFOVERADDR(13),
+	GR3D_GLOBAL_SURFOVERADDR(14),
+	GR3D_GLOBAL_SURFOVERADDR(15),
+	GR3D_GLOBAL_SAMP01SURFADDR( 0),
+	GR3D_GLOBAL_SAMP01SURFADDR( 1),
+	GR3D_GLOBAL_SAMP01SURFADDR( 2),
+	GR3D_GLOBAL_SAMP01SURFADDR( 3),
+	GR3D_GLOBAL_SAMP01SURFADDR( 4),
+	GR3D_GLOBAL_SAMP01SURFADDR( 5),
+	GR3D_GLOBAL_SAMP01SURFADDR( 6),
+	GR3D_GLOBAL_SAMP01SURFADDR( 7),
+	GR3D_GLOBAL_SAMP01SURFADDR( 8),
+	GR3D_GLOBAL_SAMP01SURFADDR( 9),
+	GR3D_GLOBAL_SAMP01SURFADDR(10),
+	GR3D_GLOBAL_SAMP01SURFADDR(11),
+	GR3D_GLOBAL_SAMP01SURFADDR(12),
+	GR3D_GLOBAL_SAMP01SURFADDR(13),
+	GR3D_GLOBAL_SAMP01SURFADDR(14),
+	GR3D_GLOBAL_SAMP01SURFADDR(15),
+	GR3D_GLOBAL_SAMP23SURFADDR( 0),
+	GR3D_GLOBAL_SAMP23SURFADDR( 1),
+	GR3D_GLOBAL_SAMP23SURFADDR( 2),
+	GR3D_GLOBAL_SAMP23SURFADDR( 3),
+	GR3D_GLOBAL_SAMP23SURFADDR( 4),
+	GR3D_GLOBAL_SAMP23SURFADDR( 5),
+	GR3D_GLOBAL_SAMP23SURFADDR( 6),
+	GR3D_GLOBAL_SAMP23SURFADDR( 7),
+	GR3D_GLOBAL_SAMP23SURFADDR( 8),
+	GR3D_GLOBAL_SAMP23SURFADDR( 9),
+	GR3D_GLOBAL_SAMP23SURFADDR(10),
+	GR3D_GLOBAL_SAMP23SURFADDR(11),
+	GR3D_GLOBAL_SAMP23SURFADDR(12),
+	GR3D_GLOBAL_SAMP23SURFADDR(13),
+	GR3D_GLOBAL_SAMP23SURFADDR(14),
+	GR3D_GLOBAL_SAMP23SURFADDR(15),
+};
+
+static int gr3d_probe(struct platform_device *pdev)
+{
+	struct device_node *np = pdev->dev.of_node;
+	struct host1x_syncpt **syncpts;
+	struct gr3d *gr3d;
+	unsigned int i;
+	int err;
+
+	gr3d = devm_kzalloc(&pdev->dev, sizeof(*gr3d), GFP_KERNEL);
+	if (!gr3d)
+		return -ENOMEM;
+
+	syncpts = devm_kzalloc(&pdev->dev, sizeof(*syncpts), GFP_KERNEL);
+	if (!syncpts)
+		return -ENOMEM;
+
+	gr3d->clk = devm_clk_get(&pdev->dev, NULL);
+	if (IS_ERR(gr3d->clk)) {
+		dev_err(&pdev->dev, "cannot get clock\n");
+		return PTR_ERR(gr3d->clk);
+	}
+
+	if (of_device_is_compatible(np, "nvidia,tegra30-gr3d")) {
+		gr3d->clk_secondary = devm_clk_get(&pdev->dev, "3d2");
+		if (IS_ERR(gr3d->clk)) {
+			dev_err(&pdev->dev, "cannot get secondary clock\n");
+			return PTR_ERR(gr3d->clk);
+		}
+	}
+
+	err = tegra_powergate_sequence_power_up(TEGRA_POWERGATE_3D, gr3d->clk);
+	if (err < 0) {
+		dev_err(&pdev->dev, "failed to power up 3D unit\n");
+		return err;
+	}
+
+	if (gr3d->clk_secondary) {
+		err = tegra_powergate_sequence_power_up(TEGRA_POWERGATE_3D1,
+							gr3d->clk_secondary);
+		if (err < 0) {
+			dev_err(&pdev->dev,
+				"failed to power up secondary 3D unit\n");
+			return err;
+		}
+	}
+
+	INIT_LIST_HEAD(&gr3d->client.base.list);
+	gr3d->client.base.ops = &gr3d_client_ops;
+	gr3d->client.base.dev = &pdev->dev;
+	gr3d->client.base.class = HOST1X_CLASS_GR3D;
+	gr3d->client.base.syncpts = syncpts;
+	gr3d->client.base.num_syncpts = 1;
+
+	INIT_LIST_HEAD(&gr3d->client.list);
+	gr3d->client.ops = &gr3d_ops;
+
+	err = host1x_client_register(&gr3d->client.base);
+	if (err < 0) {
+		dev_err(&pdev->dev, "failed to register host1x client: %d\n",
+			err);
+		return err;
+	}
+
+	/* initialize address register map */
+	for (i = 0; i < ARRAY_SIZE(gr3d_addr_regs); i++)
+		set_bit(gr3d_addr_regs[i], gr3d->addr_regs);
+
+	platform_set_drvdata(pdev, gr3d);
+
+	return 0;
+}
+
+static int gr3d_remove(struct platform_device *pdev)
+{
+	struct gr3d *gr3d = platform_get_drvdata(pdev);
+	int err;
+
+	err = host1x_client_unregister(&gr3d->client.base);
+	if (err < 0) {
+		dev_err(&pdev->dev, "failed to unregister host1x client: %d\n",
+			err);
+		return err;
+	}
+
+	if (gr3d->clk_secondary) {
+		tegra_powergate_power_off(TEGRA_POWERGATE_3D1);
+		clk_disable_unprepare(gr3d->clk_secondary);
+	}
+
+	tegra_powergate_power_off(TEGRA_POWERGATE_3D);
+	clk_disable_unprepare(gr3d->clk);
+
+	return 0;
+}
+
+struct platform_driver tegra_gr3d_driver = {
+	.driver = {
+		.name = "tegra-gr3d",
+		.of_match_table = tegra_gr3d_match,
+	},
+	.probe = gr3d_probe,
+	.remove = gr3d_remove,
+};
diff --git a/drivers/gpu/drm/tegra/gr3d.h b/drivers/gpu/drm/tegra/gr3d.h
new file mode 100644
index 000000000000..0c30a1351c83
--- /dev/null
+++ b/drivers/gpu/drm/tegra/gr3d.h
@@ -0,0 +1,27 @@
+/*
+ * Copyright (C) 2013 NVIDIA Corporation
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ */
+
+#ifndef TEGRA_GR3D_H
+#define TEGRA_GR3D_H
+
+#define GR3D_IDX_ATTRIBUTE(x)		(0x100 + (x) * 2)
+#define GR3D_IDX_INDEX_BASE		0x121
+#define GR3D_QR_ZTAG_ADDR		0x415
+#define GR3D_QR_CTAG_ADDR		0x417
+#define GR3D_QR_CZ_ADDR			0x419
+#define GR3D_TEX_TEX_ADDR(x)		(0x710 + (x))
+#define GR3D_DW_MEMORY_OUTPUT_ADDRESS	0x904
+#define GR3D_GLOBAL_SURFADDR(x)		(0xe00 + (x))
+#define GR3D_GLOBAL_SPILLSURFADDR	0xe2a
+#define GR3D_GLOBAL_SURFOVERADDR(x)	(0xe30 + (x))
+#define GR3D_GLOBAL_SAMP01SURFADDR(x)	(0xe50 + (x))
+#define GR3D_GLOBAL_SAMP23SURFADDR(x)	(0xe60 + (x))
+
+#define GR3D_NUM_REGS			0xe88
+
+#endif
diff --git a/drivers/gpu/host1x/drm/hdmi.c b/drivers/gpu/drm/tegra/hdmi.c
index 644d95c7d489..0cd9bc2056e8 100644
--- a/drivers/gpu/host1x/drm/hdmi.c
+++ b/drivers/gpu/drm/tegra/hdmi.c
@@ -8,21 +8,33 @@
  */
 
 #include <linux/clk.h>
+#include <linux/clk/tegra.h>
 #include <linux/debugfs.h>
-#include <linux/gpio.h>
 #include <linux/hdmi.h>
-#include <linux/module.h>
-#include <linux/of.h>
-#include <linux/platform_device.h>
 #include <linux/regulator/consumer.h>
-#include <linux/clk/tegra.h>
-
-#include <drm/drm_edid.h>
 
 #include "hdmi.h"
 #include "drm.h"
 #include "dc.h"
-#include "host1x_client.h"
+
+struct tmds_config {
+	unsigned int pclk;
+	u32 pll0;
+	u32 pll1;
+	u32 pe_current;
+	u32 drive_current;
+	u32 peak_current;
+};
+
+struct tegra_hdmi_config {
+	const struct tmds_config *tmds;
+	unsigned int num_tmds;
+
+	unsigned long fuse_override_offset;
+	unsigned long fuse_override_value;
+
+	bool has_sor_io_peak_current;
+};
 
 struct tegra_hdmi {
 	struct host1x_client client;
@@ -38,6 +50,8 @@ struct tegra_hdmi {
 	struct clk *clk_parent;
 	struct clk *clk;
 
+	const struct tegra_hdmi_config *config;
+
 	unsigned int audio_source;
 	unsigned int audio_freq;
 	bool stereo;
@@ -143,15 +157,7 @@ static const struct tegra_hdmi_audio_config tegra_hdmi_audio_192k[] = {
 	{         0,     0,      0,     0 },
 };
 
-struct tmds_config {
-	unsigned int pclk;
-	u32 pll0;
-	u32 pll1;
-	u32 pe_current;
-	u32 drive_current;
-};
-
-static const struct tmds_config tegra2_tmds_config[] = {
+static const struct tmds_config tegra20_tmds_config[] = {
 	{ /* slow pixel clock modes */
 		.pclk = 27000000,
 		.pll0 = SOR_PLL_BG_V17_S(3) | SOR_PLL_ICHPMP(1) |
@@ -184,7 +190,7 @@ static const struct tmds_config tegra2_tmds_config[] = {
 	},
 };
 
-static const struct tmds_config tegra3_tmds_config[] = {
+static const struct tmds_config tegra30_tmds_config[] = {
 	{ /* 480p modes */
 		.pclk = 27000000,
 		.pll0 = SOR_PLL_BG_V17_S(3) | SOR_PLL_ICHPMP(1) |
@@ -230,6 +236,85 @@ static const struct tmds_config tegra3_tmds_config[] = {
 	},
 };
 
+static const struct tmds_config tegra114_tmds_config[] = {
+	{ /* 480p/576p / 25.2MHz/27MHz modes */
+		.pclk = 27000000,
+		.pll0 = SOR_PLL_ICHPMP(1) | SOR_PLL_BG_V17_S(3) |
+			SOR_PLL_VCOCAP(0) | SOR_PLL_RESISTORSEL,
+		.pll1 = SOR_PLL_LOADADJ(3) | SOR_PLL_TMDS_TERMADJ(0),
+		.pe_current = PE_CURRENT0(PE_CURRENT_0_mA_T114) |
+			PE_CURRENT1(PE_CURRENT_0_mA_T114) |
+			PE_CURRENT2(PE_CURRENT_0_mA_T114) |
+			PE_CURRENT3(PE_CURRENT_0_mA_T114),
+		.drive_current =
+			DRIVE_CURRENT_LANE0_T114(DRIVE_CURRENT_10_400_mA_T114) |
+			DRIVE_CURRENT_LANE1_T114(DRIVE_CURRENT_10_400_mA_T114) |
+			DRIVE_CURRENT_LANE2_T114(DRIVE_CURRENT_10_400_mA_T114) |
+			DRIVE_CURRENT_LANE3_T114(DRIVE_CURRENT_10_400_mA_T114),
+		.peak_current = PEAK_CURRENT_LANE0(PEAK_CURRENT_0_000_mA) |
+			PEAK_CURRENT_LANE1(PEAK_CURRENT_0_000_mA) |
+			PEAK_CURRENT_LANE2(PEAK_CURRENT_0_000_mA) |
+			PEAK_CURRENT_LANE3(PEAK_CURRENT_0_000_mA),
+	}, { /* 720p / 74.25MHz modes */
+		.pclk = 74250000,
+		.pll0 = SOR_PLL_ICHPMP(1) | SOR_PLL_BG_V17_S(3) |
+			SOR_PLL_VCOCAP(1) | SOR_PLL_RESISTORSEL,
+		.pll1 = SOR_PLL_PE_EN | SOR_PLL_LOADADJ(3) |
+			SOR_PLL_TMDS_TERMADJ(0),
+		.pe_current = PE_CURRENT0(PE_CURRENT_15_mA_T114) |
+			PE_CURRENT1(PE_CURRENT_15_mA_T114) |
+			PE_CURRENT2(PE_CURRENT_15_mA_T114) |
+			PE_CURRENT3(PE_CURRENT_15_mA_T114),
+		.drive_current =
+			DRIVE_CURRENT_LANE0_T114(DRIVE_CURRENT_10_400_mA_T114) |
+			DRIVE_CURRENT_LANE1_T114(DRIVE_CURRENT_10_400_mA_T114) |
+			DRIVE_CURRENT_LANE2_T114(DRIVE_CURRENT_10_400_mA_T114) |
+			DRIVE_CURRENT_LANE3_T114(DRIVE_CURRENT_10_400_mA_T114),
+		.peak_current = PEAK_CURRENT_LANE0(PEAK_CURRENT_0_000_mA) |
+			PEAK_CURRENT_LANE1(PEAK_CURRENT_0_000_mA) |
+			PEAK_CURRENT_LANE2(PEAK_CURRENT_0_000_mA) |
+			PEAK_CURRENT_LANE3(PEAK_CURRENT_0_000_mA),
+	}, { /* 1080p / 148.5MHz modes */
+		.pclk = 148500000,
+		.pll0 = SOR_PLL_ICHPMP(1) | SOR_PLL_BG_V17_S(3) |
+			SOR_PLL_VCOCAP(3) | SOR_PLL_RESISTORSEL,
+		.pll1 = SOR_PLL_PE_EN | SOR_PLL_LOADADJ(3) |
+			SOR_PLL_TMDS_TERMADJ(0),
+		.pe_current = PE_CURRENT0(PE_CURRENT_10_mA_T114) |
+			PE_CURRENT1(PE_CURRENT_10_mA_T114) |
+			PE_CURRENT2(PE_CURRENT_10_mA_T114) |
+			PE_CURRENT3(PE_CURRENT_10_mA_T114),
+		.drive_current =
+			DRIVE_CURRENT_LANE0_T114(DRIVE_CURRENT_12_400_mA_T114) |
+			DRIVE_CURRENT_LANE1_T114(DRIVE_CURRENT_12_400_mA_T114) |
+			DRIVE_CURRENT_LANE2_T114(DRIVE_CURRENT_12_400_mA_T114) |
+			DRIVE_CURRENT_LANE3_T114(DRIVE_CURRENT_12_400_mA_T114),
+		.peak_current = PEAK_CURRENT_LANE0(PEAK_CURRENT_0_000_mA) |
+			PEAK_CURRENT_LANE1(PEAK_CURRENT_0_000_mA) |
+			PEAK_CURRENT_LANE2(PEAK_CURRENT_0_000_mA) |
+			PEAK_CURRENT_LANE3(PEAK_CURRENT_0_000_mA),
+	}, { /* 225/297MHz modes */
+		.pclk = UINT_MAX,
+		.pll0 = SOR_PLL_ICHPMP(1) | SOR_PLL_BG_V17_S(3) |
+			SOR_PLL_VCOCAP(0xf) | SOR_PLL_RESISTORSEL,
+		.pll1 = SOR_PLL_LOADADJ(3) | SOR_PLL_TMDS_TERMADJ(7)
+			| SOR_PLL_TMDS_TERM_ENABLE,
+		.pe_current = PE_CURRENT0(PE_CURRENT_0_mA_T114) |
+			PE_CURRENT1(PE_CURRENT_0_mA_T114) |
+			PE_CURRENT2(PE_CURRENT_0_mA_T114) |
+			PE_CURRENT3(PE_CURRENT_0_mA_T114),
+		.drive_current =
+			DRIVE_CURRENT_LANE0_T114(DRIVE_CURRENT_25_200_mA_T114) |
+			DRIVE_CURRENT_LANE1_T114(DRIVE_CURRENT_25_200_mA_T114) |
+			DRIVE_CURRENT_LANE2_T114(DRIVE_CURRENT_25_200_mA_T114) |
+			DRIVE_CURRENT_LANE3_T114(DRIVE_CURRENT_19_200_mA_T114),
+		.peak_current = PEAK_CURRENT_LANE0(PEAK_CURRENT_3_000_mA) |
+			PEAK_CURRENT_LANE1(PEAK_CURRENT_3_000_mA) |
+			PEAK_CURRENT_LANE2(PEAK_CURRENT_3_000_mA) |
+			PEAK_CURRENT_LANE3(PEAK_CURRENT_0_800_mA),
+	},
+};
+
 static const struct tegra_hdmi_audio_config *
 tegra_hdmi_get_audio_config(unsigned int audio_freq, unsigned int pclk)
 {
@@ -511,7 +596,7 @@ static void tegra_hdmi_setup_audio_infoframe(struct tegra_hdmi *hdmi)
 
 	err = hdmi_audio_infoframe_init(&frame);
 	if (err < 0) {
-		dev_err(hdmi->dev, "failed to initialize audio infoframe: %d\n",
+		dev_err(hdmi->dev, "failed to setup audio infoframe: %zd\n",
 			err);
 		return;
 	}
@@ -531,7 +616,7 @@ static void tegra_hdmi_setup_audio_infoframe(struct tegra_hdmi *hdmi)
 	 * contain 7 bytes. Including the 3 byte header only the first 10
 	 * bytes can be programmed.
 	 */
-	tegra_hdmi_write_infopack(hdmi, buffer, min(10, err));
+	tegra_hdmi_write_infopack(hdmi, buffer, min_t(size_t, 10, err));
 
 	tegra_hdmi_writel(hdmi, INFOFRAME_CTRL_ENABLE,
 			  HDMI_NV_PDISP_HDMI_AUDIO_INFOFRAME_CTRL);
@@ -577,8 +662,28 @@ static void tegra_hdmi_setup_tmds(struct tegra_hdmi *hdmi,
 	tegra_hdmi_writel(hdmi, tmds->pll1, HDMI_NV_PDISP_SOR_PLL1);
 	tegra_hdmi_writel(hdmi, tmds->pe_current, HDMI_NV_PDISP_PE_CURRENT);
 
-	value = tmds->drive_current | DRIVE_CURRENT_FUSE_OVERRIDE;
-	tegra_hdmi_writel(hdmi, value, HDMI_NV_PDISP_SOR_LANE_DRIVE_CURRENT);
+	tegra_hdmi_writel(hdmi, tmds->drive_current,
+			  HDMI_NV_PDISP_SOR_LANE_DRIVE_CURRENT);
+
+	value = tegra_hdmi_readl(hdmi, hdmi->config->fuse_override_offset);
+	value |= hdmi->config->fuse_override_value;
+	tegra_hdmi_writel(hdmi, value, hdmi->config->fuse_override_offset);
+
+	if (hdmi->config->has_sor_io_peak_current)
+		tegra_hdmi_writel(hdmi, tmds->peak_current,
+				  HDMI_NV_PDISP_SOR_IO_PEAK_CURRENT);
+}
+
+static bool tegra_output_is_hdmi(struct tegra_output *output)
+{
+	struct edid *edid;
+
+	if (!output->connector.edid_blob_ptr)
+		return false;
+
+	edid = (struct edid *)output->connector.edid_blob_ptr->data;
+
+	return drm_detect_hdmi_monitor(edid);
 }
 
 static int tegra_output_hdmi_enable(struct tegra_output *output)
@@ -589,23 +694,17 @@ static int tegra_output_hdmi_enable(struct tegra_output *output)
 	struct tegra_hdmi *hdmi = to_hdmi(output);
 	struct device_node *node = hdmi->dev->of_node;
 	unsigned int pulse_start, div82, pclk;
-	const struct tmds_config *tmds;
-	unsigned int num_tmds;
 	unsigned long value;
 	int retries = 1000;
 	int err;
 
+	hdmi->dvi = !tegra_output_is_hdmi(output);
+
 	pclk = mode->clock * 1000;
 	h_sync_width = mode->hsync_end - mode->hsync_start;
 	h_back_porch = mode->htotal - mode->hsync_end;
 	h_front_porch = mode->hsync_start - mode->hdisplay;
 
-	err = regulator_enable(hdmi->vdd);
-	if (err < 0) {
-		dev_err(hdmi->dev, "failed to enable VDD regulator: %d\n", err);
-		return err;
-	}
-
 	err = regulator_enable(hdmi->pll);
 	if (err < 0) {
 		dev_err(hdmi->dev, "failed to enable PLL regulator: %d\n", err);
@@ -710,17 +809,9 @@ static int tegra_output_hdmi_enable(struct tegra_output *output)
 	tegra_hdmi_setup_stereo_infoframe(hdmi);
 
 	/* TMDS CONFIG */
-	if (of_device_is_compatible(node, "nvidia,tegra30-hdmi")) {
-		num_tmds = ARRAY_SIZE(tegra3_tmds_config);
-		tmds = tegra3_tmds_config;
-	} else {
-		num_tmds = ARRAY_SIZE(tegra2_tmds_config);
-		tmds = tegra2_tmds_config;
-	}
-
-	for (i = 0; i < num_tmds; i++) {
-		if (pclk <= tmds[i].pclk) {
-			tegra_hdmi_setup_tmds(hdmi, &tmds[i]);
+	for (i = 0; i < hdmi->config->num_tmds; i++) {
+		if (pclk <= hdmi->config->tmds[i].pclk) {
+			tegra_hdmi_setup_tmds(hdmi, &hdmi->config->tmds[i]);
 			break;
 		}
 	}
@@ -824,7 +915,6 @@ static int tegra_output_hdmi_disable(struct tegra_output *output)
 	tegra_periph_reset_assert(hdmi->clk);
 	clk_disable(hdmi->clk);
 	regulator_disable(hdmi->pll);
-	regulator_disable(hdmi->vdd);
 
 	return 0;
 }
@@ -1055,6 +1145,7 @@ static int tegra_hdmi_show_regs(struct seq_file *s, void *data)
 	DUMP_REG(HDMI_NV_PDISP_SOR_AUDIO_CNTRL0);
 	DUMP_REG(HDMI_NV_PDISP_SOR_AUDIO_HDA_ELD_BUFWR);
 	DUMP_REG(HDMI_NV_PDISP_SOR_AUDIO_HDA_PRESENSE);
+	DUMP_REG(HDMI_NV_PDISP_SOR_IO_PEAK_CURRENT);
 
 #undef DUMP_REG
 
@@ -1122,24 +1213,31 @@ static int tegra_hdmi_debugfs_exit(struct tegra_hdmi *hdmi)
 	return 0;
 }
 
-static int tegra_hdmi_drm_init(struct host1x_client *client,
-			       struct drm_device *drm)
+static int tegra_hdmi_init(struct host1x_client *client)
 {
+	struct tegra_drm *tegra = dev_get_drvdata(client->parent);
 	struct tegra_hdmi *hdmi = host1x_client_to_hdmi(client);
 	int err;
 
+	err = regulator_enable(hdmi->vdd);
+	if (err < 0) {
+		dev_err(client->dev, "failed to enable VDD regulator: %d\n",
+			err);
+		return err;
+	}
+
 	hdmi->output.type = TEGRA_OUTPUT_HDMI;
 	hdmi->output.dev = client->dev;
 	hdmi->output.ops = &hdmi_ops;
 
-	err = tegra_output_init(drm, &hdmi->output);
+	err = tegra_output_init(tegra->drm, &hdmi->output);
 	if (err < 0) {
 		dev_err(client->dev, "output setup failed: %d\n", err);
 		return err;
 	}
 
 	if (IS_ENABLED(CONFIG_DEBUG_FS)) {
-		err = tegra_hdmi_debugfs_init(hdmi, drm->primary);
+		err = tegra_hdmi_debugfs_init(hdmi, tegra->drm->primary);
 		if (err < 0)
 			dev_err(client->dev, "debugfs setup failed: %d\n", err);
 	}
@@ -1147,7 +1245,7 @@ static int tegra_hdmi_drm_init(struct host1x_client *client,
 	return 0;
 }
 
-static int tegra_hdmi_drm_exit(struct host1x_client *client)
+static int tegra_hdmi_exit(struct host1x_client *client)
 {
 	struct tegra_hdmi *hdmi = host1x_client_to_hdmi(client);
 	int err;
@@ -1171,25 +1269,63 @@ static int tegra_hdmi_drm_exit(struct host1x_client *client)
 		return err;
 	}
 
+	regulator_disable(hdmi->vdd);
+
 	return 0;
 }
 
 static const struct host1x_client_ops hdmi_client_ops = {
-	.drm_init = tegra_hdmi_drm_init,
-	.drm_exit = tegra_hdmi_drm_exit,
+	.init = tegra_hdmi_init,
+	.exit = tegra_hdmi_exit,
+};
+
+static const struct tegra_hdmi_config tegra20_hdmi_config = {
+	.tmds = tegra20_tmds_config,
+	.num_tmds = ARRAY_SIZE(tegra20_tmds_config),
+	.fuse_override_offset = HDMI_NV_PDISP_SOR_LANE_DRIVE_CURRENT,
+	.fuse_override_value = 1 << 31,
+	.has_sor_io_peak_current = false,
+};
+
+static const struct tegra_hdmi_config tegra30_hdmi_config = {
+	.tmds = tegra30_tmds_config,
+	.num_tmds = ARRAY_SIZE(tegra30_tmds_config),
+	.fuse_override_offset = HDMI_NV_PDISP_SOR_LANE_DRIVE_CURRENT,
+	.fuse_override_value = 1 << 31,
+	.has_sor_io_peak_current = false,
+};
+
+static const struct tegra_hdmi_config tegra114_hdmi_config = {
+	.tmds = tegra114_tmds_config,
+	.num_tmds = ARRAY_SIZE(tegra114_tmds_config),
+	.fuse_override_offset = HDMI_NV_PDISP_SOR_PAD_CTLS0,
+	.fuse_override_value = 1 << 31,
+	.has_sor_io_peak_current = true,
+};
+
+static const struct of_device_id tegra_hdmi_of_match[] = {
+	{ .compatible = "nvidia,tegra114-hdmi", .data = &tegra114_hdmi_config },
+	{ .compatible = "nvidia,tegra30-hdmi", .data = &tegra30_hdmi_config },
+	{ .compatible = "nvidia,tegra20-hdmi", .data = &tegra20_hdmi_config },
+	{ },
 };
 
 static int tegra_hdmi_probe(struct platform_device *pdev)
 {
-	struct host1x_drm *host1x = host1x_get_drm_data(pdev->dev.parent);
+	const struct of_device_id *match;
 	struct tegra_hdmi *hdmi;
 	struct resource *regs;
 	int err;
 
+	match = of_match_node(tegra_hdmi_of_match, pdev->dev.of_node);
+	if (!match)
+		return -ENODEV;
+
 	hdmi = devm_kzalloc(&pdev->dev, sizeof(*hdmi), GFP_KERNEL);
 	if (!hdmi)
 		return -ENOMEM;
 
+	hdmi->config = match->data;
 	hdmi->dev = &pdev->dev;
 	hdmi->audio_source = AUTO;
 	hdmi->audio_freq = 44100;
@@ -1234,7 +1370,7 @@ static int tegra_hdmi_probe(struct platform_device *pdev)
 
 	hdmi->output.dev = &pdev->dev;
 
-	err = tegra_output_parse_dt(&hdmi->output);
+	err = tegra_output_probe(&hdmi->output);
 	if (err < 0)
 		return err;
 
@@ -1252,11 +1388,11 @@ static int tegra_hdmi_probe(struct platform_device *pdev)
 
 	hdmi->irq = err;
 
-	hdmi->client.ops = &hdmi_client_ops;
 	INIT_LIST_HEAD(&hdmi->client.list);
+	hdmi->client.ops = &hdmi_client_ops;
 	hdmi->client.dev = &pdev->dev;
 
-	err = host1x_register_client(host1x, &hdmi->client);
+	err = host1x_client_register(&hdmi->client);
 	if (err < 0) {
 		dev_err(&pdev->dev, "failed to register host1x client: %d\n",
 			err);
@@ -1270,29 +1406,28 @@ static int tegra_hdmi_probe(struct platform_device *pdev)
 
 static int tegra_hdmi_remove(struct platform_device *pdev)
 {
-	struct host1x_drm *host1x = host1x_get_drm_data(pdev->dev.parent);
 	struct tegra_hdmi *hdmi = platform_get_drvdata(pdev);
 	int err;
 
-	err = host1x_unregister_client(host1x, &hdmi->client);
+	err = host1x_client_unregister(&hdmi->client);
 	if (err < 0) {
 		dev_err(&pdev->dev, "failed to unregister host1x client: %d\n",
 			err);
 		return err;
 	}
 
+	err = tegra_output_remove(&hdmi->output);
+	if (err < 0) {
+		dev_err(&pdev->dev, "failed to remove output: %d\n", err);
+		return err;
+	}
+
 	clk_unprepare(hdmi->clk_parent);
 	clk_unprepare(hdmi->clk);
 
 	return 0;
 }
 
-static struct of_device_id tegra_hdmi_of_match[] = {
-	{ .compatible = "nvidia,tegra30-hdmi", },
-	{ .compatible = "nvidia,tegra20-hdmi", },
-	{ },
-};
-
 struct platform_driver tegra_hdmi_driver = {
 	.driver = {
 		.name = "tegra-hdmi",
diff --git a/drivers/gpu/host1x/drm/hdmi.h b/drivers/gpu/drm/tegra/hdmi.h
index 52ac36e08ccb..0aebc485f7fa 100644
--- a/drivers/gpu/host1x/drm/hdmi.h
+++ b/drivers/gpu/drm/tegra/hdmi.h
@@ -233,7 +233,10 @@
 #define DRIVE_CURRENT_LANE1(x)      (((x) & 0x3f) <<  8)
 #define DRIVE_CURRENT_LANE2(x)      (((x) & 0x3f) << 16)
 #define DRIVE_CURRENT_LANE3(x)      (((x) & 0x3f) << 24)
-#define DRIVE_CURRENT_FUSE_OVERRIDE (1 << 31)
+#define DRIVE_CURRENT_LANE0_T114(x) (((x) & 0x7f) <<  0)
+#define DRIVE_CURRENT_LANE1_T114(x) (((x) & 0x7f) <<  8)
+#define DRIVE_CURRENT_LANE2_T114(x) (((x) & 0x7f) << 16)
+#define DRIVE_CURRENT_LANE3_T114(x) (((x) & 0x7f) << 24)
 
 #define DRIVE_CURRENT_1_500_mA  0x00
 #define DRIVE_CURRENT_1_875_mA  0x01
@@ -299,6 +302,79 @@
 #define DRIVE_CURRENT_24_375_mA 0x3d
 #define DRIVE_CURRENT_24_750_mA 0x3e
 
+#define DRIVE_CURRENT_0_000_mA_T114 0x00
+#define DRIVE_CURRENT_0_400_mA_T114 0x01
+#define DRIVE_CURRENT_0_800_mA_T114 0x02
+#define DRIVE_CURRENT_1_200_mA_T114 0x03
+#define DRIVE_CURRENT_1_600_mA_T114 0x04
+#define DRIVE_CURRENT_2_000_mA_T114 0x05
+#define DRIVE_CURRENT_2_400_mA_T114 0x06
+#define DRIVE_CURRENT_2_800_mA_T114 0x07
+#define DRIVE_CURRENT_3_200_mA_T114 0x08
+#define DRIVE_CURRENT_3_600_mA_T114 0x09
+#define DRIVE_CURRENT_4_000_mA_T114 0x0a
+#define DRIVE_CURRENT_4_400_mA_T114 0x0b
+#define DRIVE_CURRENT_4_800_mA_T114 0x0c
+#define DRIVE_CURRENT_5_200_mA_T114 0x0d
+#define DRIVE_CURRENT_5_600_mA_T114 0x0e
+#define DRIVE_CURRENT_6_000_mA_T114 0x0f
+#define DRIVE_CURRENT_6_400_mA_T114 0x10
+#define DRIVE_CURRENT_6_800_mA_T114 0x11
+#define DRIVE_CURRENT_7_200_mA_T114 0x12
+#define DRIVE_CURRENT_7_600_mA_T114 0x13
+#define DRIVE_CURRENT_8_000_mA_T114 0x14
+#define DRIVE_CURRENT_8_400_mA_T114 0x15
+#define DRIVE_CURRENT_8_800_mA_T114 0x16
+#define DRIVE_CURRENT_9_200_mA_T114 0x17
+#define DRIVE_CURRENT_9_600_mA_T114 0x18
+#define DRIVE_CURRENT_10_000_mA_T114 0x19
+#define DRIVE_CURRENT_10_400_mA_T114 0x1a
+#define DRIVE_CURRENT_10_800_mA_T114 0x1b
+#define DRIVE_CURRENT_11_200_mA_T114 0x1c
+#define DRIVE_CURRENT_11_600_mA_T114 0x1d
+#define DRIVE_CURRENT_12_000_mA_T114 0x1e
+#define DRIVE_CURRENT_12_400_mA_T114 0x1f
+#define DRIVE_CURRENT_12_800_mA_T114 0x20
+#define DRIVE_CURRENT_13_200_mA_T114 0x21
+#define DRIVE_CURRENT_13_600_mA_T114 0x22
+#define DRIVE_CURRENT_14_000_mA_T114 0x23
+#define DRIVE_CURRENT_14_400_mA_T114 0x24
+#define DRIVE_CURRENT_14_800_mA_T114 0x25
+#define DRIVE_CURRENT_15_200_mA_T114 0x26
+#define DRIVE_CURRENT_15_600_mA_T114 0x27
+#define DRIVE_CURRENT_16_000_mA_T114 0x28
+#define DRIVE_CURRENT_16_400_mA_T114 0x29
+#define DRIVE_CURRENT_16_800_mA_T114 0x2a
+#define DRIVE_CURRENT_17_200_mA_T114 0x2b
+#define DRIVE_CURRENT_17_600_mA_T114 0x2c
+#define DRIVE_CURRENT_18_000_mA_T114 0x2d
+#define DRIVE_CURRENT_18_400_mA_T114 0x2e
+#define DRIVE_CURRENT_18_800_mA_T114 0x2f
+#define DRIVE_CURRENT_19_200_mA_T114 0x30
+#define DRIVE_CURRENT_19_600_mA_T114 0x31
+#define DRIVE_CURRENT_20_000_mA_T114 0x32
+#define DRIVE_CURRENT_20_400_mA_T114 0x33
+#define DRIVE_CURRENT_20_800_mA_T114 0x34
+#define DRIVE_CURRENT_21_200_mA_T114 0x35
+#define DRIVE_CURRENT_21_600_mA_T114 0x36
+#define DRIVE_CURRENT_22_000_mA_T114 0x37
+#define DRIVE_CURRENT_22_400_mA_T114 0x38
+#define DRIVE_CURRENT_22_800_mA_T114 0x39
+#define DRIVE_CURRENT_23_200_mA_T114 0x3a
+#define DRIVE_CURRENT_23_600_mA_T114 0x3b
+#define DRIVE_CURRENT_24_000_mA_T114 0x3c
+#define DRIVE_CURRENT_24_400_mA_T114 0x3d
+#define DRIVE_CURRENT_24_800_mA_T114 0x3e
+#define DRIVE_CURRENT_25_200_mA_T114 0x3f
+#define DRIVE_CURRENT_25_400_mA_T114 0x40
+#define DRIVE_CURRENT_25_800_mA_T114 0x41
+#define DRIVE_CURRENT_26_200_mA_T114 0x42
+#define DRIVE_CURRENT_26_600_mA_T114 0x43
+#define DRIVE_CURRENT_27_000_mA_T114 0x44
+#define DRIVE_CURRENT_27_400_mA_T114 0x45
+#define DRIVE_CURRENT_27_800_mA_T114 0x46
+#define DRIVE_CURRENT_28_200_mA_T114 0x47
+
 #define HDMI_NV_PDISP_AUDIO_DEBUG0				0x7f
 #define HDMI_NV_PDISP_AUDIO_DEBUG1				0x80
 #define HDMI_NV_PDISP_AUDIO_DEBUG2				0x81
@@ -358,6 +434,23 @@
 #define PE_CURRENT_7_0_mA 0xe
 #define PE_CURRENT_7_5_mA 0xf
 
+#define PE_CURRENT_0_mA_T114 0x0
+#define PE_CURRENT_1_mA_T114 0x1
+#define PE_CURRENT_2_mA_T114 0x2
+#define PE_CURRENT_3_mA_T114 0x3
+#define PE_CURRENT_4_mA_T114 0x4
+#define PE_CURRENT_5_mA_T114 0x5
+#define PE_CURRENT_6_mA_T114 0x6
+#define PE_CURRENT_7_mA_T114 0x7
+#define PE_CURRENT_8_mA_T114 0x8
+#define PE_CURRENT_9_mA_T114 0x9
+#define PE_CURRENT_10_mA_T114 0xa
+#define PE_CURRENT_11_mA_T114 0xb
+#define PE_CURRENT_12_mA_T114 0xc
+#define PE_CURRENT_13_mA_T114 0xd
+#define PE_CURRENT_14_mA_T114 0xe
+#define PE_CURRENT_15_mA_T114 0xf
+
 #define HDMI_NV_PDISP_KEY_CTRL					0x9a
 #define HDMI_NV_PDISP_KEY_DEBUG0				0x9b
 #define HDMI_NV_PDISP_KEY_DEBUG1				0x9c
@@ -383,4 +476,61 @@
 #define HDMI_NV_PDISP_SOR_AUDIO_AVAL_1920    0xc5
 #define HDMI_NV_PDISP_SOR_AUDIO_AVAL_DEFAULT 0xc5
 
+#define HDMI_NV_PDISP_SOR_IO_PEAK_CURRENT		0xd1
+#define PEAK_CURRENT_LANE0(x) (((x) & 0x7f) <<  0)
+#define PEAK_CURRENT_LANE1(x) (((x) & 0x7f) <<  8)
+#define PEAK_CURRENT_LANE2(x) (((x) & 0x7f) << 16)
+#define PEAK_CURRENT_LANE3(x) (((x) & 0x7f) << 24)
+
+#define PEAK_CURRENT_0_000_mA 0x00
+#define PEAK_CURRENT_0_200_mA 0x01
+#define PEAK_CURRENT_0_400_mA 0x02
+#define PEAK_CURRENT_0_600_mA 0x03
+#define PEAK_CURRENT_0_800_mA 0x04
+#define PEAK_CURRENT_1_000_mA 0x05
+#define PEAK_CURRENT_1_200_mA 0x06
+#define PEAK_CURRENT_1_400_mA 0x07
+#define PEAK_CURRENT_1_600_mA 0x08
+#define PEAK_CURRENT_1_800_mA 0x09
+#define PEAK_CURRENT_2_000_mA 0x0a
+#define PEAK_CURRENT_2_200_mA 0x0b
+#define PEAK_CURRENT_2_400_mA 0x0c
+#define PEAK_CURRENT_2_600_mA 0x0d
+#define PEAK_CURRENT_2_800_mA 0x0e
+#define PEAK_CURRENT_3_000_mA 0x0f
+#define PEAK_CURRENT_3_200_mA 0x10
+#define PEAK_CURRENT_3_400_mA 0x11
+#define PEAK_CURRENT_3_600_mA 0x12
+#define PEAK_CURRENT_3_800_mA 0x13
+#define PEAK_CURRENT_4_000_mA 0x14
+#define PEAK_CURRENT_4_200_mA 0x15
+#define PEAK_CURRENT_4_400_mA 0x16
+#define PEAK_CURRENT_4_600_mA 0x17
+#define PEAK_CURRENT_4_800_mA 0x18
+#define PEAK_CURRENT_5_000_mA 0x19
+#define PEAK_CURRENT_5_200_mA 0x1a
+#define PEAK_CURRENT_5_400_mA 0x1b
+#define PEAK_CURRENT_5_600_mA 0x1c
+#define PEAK_CURRENT_5_800_mA 0x1d
+#define PEAK_CURRENT_6_000_mA 0x1e
+#define PEAK_CURRENT_6_200_mA 0x1f
+#define PEAK_CURRENT_6_400_mA 0x20
+#define PEAK_CURRENT_6_600_mA 0x21
+#define PEAK_CURRENT_6_800_mA 0x22
+#define PEAK_CURRENT_7_000_mA 0x23
+#define PEAK_CURRENT_7_200_mA 0x24
+#define PEAK_CURRENT_7_400_mA 0x25
+#define PEAK_CURRENT_7_600_mA 0x26
+#define PEAK_CURRENT_7_800_mA 0x27
+#define PEAK_CURRENT_8_000_mA 0x28
+#define PEAK_CURRENT_8_200_mA 0x29
+#define PEAK_CURRENT_8_400_mA 0x2a
+#define PEAK_CURRENT_8_600_mA 0x2b
+#define PEAK_CURRENT_8_800_mA 0x2c
+#define PEAK_CURRENT_9_000_mA 0x2d
+#define PEAK_CURRENT_9_200_mA 0x2e
+#define PEAK_CURRENT_9_400_mA 0x2f
+
+#define HDMI_NV_PDISP_SOR_PAD_CTLS0		0xd2
+
 #endif /* TEGRA_HDMI_H */
diff --git a/drivers/gpu/host1x/drm/output.c b/drivers/gpu/drm/tegra/output.c
index 137ae81ab80e..2cb0065e0578 100644
--- a/drivers/gpu/host1x/drm/output.c
+++ b/drivers/gpu/drm/tegra/output.c
@@ -7,9 +7,7 @@
  * published by the Free Software Foundation.
  */
 
-#include <linux/module.h>
 #include <linux/of_gpio.h>
-#include <linux/i2c.h>
 
 #include "drm.h"
 
@@ -81,10 +79,16 @@ tegra_connector_detect(struct drm_connector *connector, bool force)
 	return status;
 }
 
+static void drm_connector_clear(struct drm_connector *connector)
+{
+	memset(connector, 0, sizeof(*connector));
+}
+
 static void tegra_connector_destroy(struct drm_connector *connector)
 {
 	drm_sysfs_connector_remove(connector);
 	drm_connector_cleanup(connector);
+	drm_connector_clear(connector);
 }
 
 static const struct drm_connector_funcs connector_funcs = {
@@ -94,9 +98,15 @@ static const struct drm_connector_funcs connector_funcs = {
 	.destroy = tegra_connector_destroy,
 };
 
+static void drm_encoder_clear(struct drm_encoder *encoder)
+{
+	memset(encoder, 0, sizeof(*encoder));
+}
+
 static void tegra_encoder_destroy(struct drm_encoder *encoder)
 {
 	drm_encoder_cleanup(encoder);
+	drm_encoder_clear(encoder);
 }
 
 static const struct drm_encoder_funcs encoder_funcs = {
@@ -151,7 +161,7 @@ static irqreturn_t hpd_irq(int irq, void *data)
 	return IRQ_HANDLED;
 }
 
-int tegra_output_parse_dt(struct tegra_output *output)
+int tegra_output_probe(struct tegra_output *output)
 {
 	enum of_gpio_flags flags;
 	struct device_node *ddc;
@@ -181,14 +191,6 @@ int tegra_output_parse_dt(struct tegra_output *output)
 	output->hpd_gpio = of_get_named_gpio_flags(output->of_node,
 						   "nvidia,hpd-gpio", 0,
 						   &flags);
-
-	return 0;
-}
-
-int tegra_output_init(struct drm_device *drm, struct tegra_output *output)
-{
-	int connector, encoder, err;
-
 	if (gpio_is_valid(output->hpd_gpio)) {
 		unsigned long flags;
 
@@ -202,7 +204,8 @@ int tegra_output_init(struct drm_device *drm, struct tegra_output *output)
 		err = gpio_to_irq(output->hpd_gpio);
 		if (err < 0) {
 			dev_err(output->dev, "gpio_to_irq(): %d\n", err);
-			goto free_hpd;
+			gpio_free(output->hpd_gpio);
+			return err;
 		}
 
 		output->hpd_irq = err;
@@ -215,12 +218,33 @@ int tegra_output_init(struct drm_device *drm, struct tegra_output *output)
 		if (err < 0) {
 			dev_err(output->dev, "failed to request IRQ#%u: %d\n",
 				output->hpd_irq, err);
-			goto free_hpd;
+			gpio_free(output->hpd_gpio);
+			return err;
 		}
 
 		output->connector.polled = DRM_CONNECTOR_POLL_HPD;
 	}
 
+	return 0;
+}
+
+int tegra_output_remove(struct tegra_output *output)
+{
+	if (gpio_is_valid(output->hpd_gpio)) {
+		free_irq(output->hpd_irq, output);
+		gpio_free(output->hpd_gpio);
+	}
+
+	if (output->ddc)
+		put_device(&output->ddc->dev);
+
+	return 0;
+}
+
+int tegra_output_init(struct drm_device *drm, struct tegra_output *output)
+{
+	int connector, encoder;
+
 	switch (output->type) {
 	case TEGRA_OUTPUT_RGB:
 		connector = DRM_MODE_CONNECTOR_LVDS;
@@ -241,6 +265,7 @@ int tegra_output_init(struct drm_device *drm, struct tegra_output *output)
 	drm_connector_init(drm, &output->connector, &connector_funcs,
 			   connector);
 	drm_connector_helper_add(&output->connector, &connector_helper_funcs);
+	output->connector.dpms = DRM_MODE_DPMS_OFF;
 
 	drm_encoder_init(drm, &output->encoder, &encoder_funcs, encoder);
 	drm_encoder_helper_add(&output->encoder, &encoder_helper_funcs);
@@ -251,22 +276,9 @@ int tegra_output_init(struct drm_device *drm, struct tegra_output *output)
 	output->encoder.possible_crtcs = 0x3;
 
 	return 0;
-
-free_hpd:
-	gpio_free(output->hpd_gpio);
-
-	return err;
 }
 
 int tegra_output_exit(struct tegra_output *output)
 {
-	if (gpio_is_valid(output->hpd_gpio)) {
-		free_irq(output->hpd_irq, output);
-		gpio_free(output->hpd_gpio);
-	}
-
-	if (output->ddc)
-		put_device(&output->ddc->dev);
-
 	return 0;
 }
diff --git a/drivers/gpu/host1x/drm/rgb.c b/drivers/gpu/drm/tegra/rgb.c
index 5aa66ef7a946..ba47ca4fb880 100644
--- a/drivers/gpu/host1x/drm/rgb.c
+++ b/drivers/gpu/drm/tegra/rgb.c
@@ -8,9 +8,6 @@
  */
 
 #include <linux/clk.h>
-#include <linux/module.h>
-#include <linux/of.h>
-#include <linux/platform_device.h>
 
 #include "drm.h"
 #include "dc.h"
@@ -150,7 +147,7 @@ int tegra_dc_rgb_probe(struct tegra_dc *dc)
 	rgb->output.dev = dc->dev;
 	rgb->output.of_node = np;
 
-	err = tegra_output_parse_dt(&rgb->output);
+	err = tegra_output_probe(&rgb->output);
 	if (err < 0)
 		return err;
 
@@ -177,6 +174,20 @@ int tegra_dc_rgb_probe(struct tegra_dc *dc)
 	return 0;
 }
 
+int tegra_dc_rgb_remove(struct tegra_dc *dc)
+{
+	int err;
+
+	if (!dc->rgb)
+		return 0;
+
+	err = tegra_output_remove(dc->rgb);
+	if (err < 0)
+		return err;
+
+	return 0;
+}
+
 int tegra_dc_rgb_init(struct drm_device *drm, struct tegra_dc *dc)
 {
 	struct tegra_rgb *rgb = to_rgb(dc->rgb);
diff --git a/drivers/gpu/drm/tilcdc/Kconfig b/drivers/gpu/drm/tilcdc/Kconfig
index 7a4d10106906..7c3ef79fcb37 100644
--- a/drivers/gpu/drm/tilcdc/Kconfig
+++ b/drivers/gpu/drm/tilcdc/Kconfig
@@ -2,6 +2,7 @@ config DRM_TILCDC
 	tristate "DRM Support for TI LCDC Display Controller"
 	depends on DRM && OF && ARM
 	select DRM_KMS_HELPER
+	select DRM_KMS_FB_HELPER
 	select DRM_KMS_CMA_HELPER
 	select DRM_GEM_CMA_HELPER
 	select VIDEOMODE_HELPERS
diff --git a/drivers/gpu/drm/ttm/Makefile b/drivers/gpu/drm/ttm/Makefile
index b2b33dde2afb..b433b9f040c9 100644
--- a/drivers/gpu/drm/ttm/Makefile
+++ b/drivers/gpu/drm/ttm/Makefile
@@ -5,10 +5,6 @@ ccflags-y := -Iinclude/drm
 ttm-y := ttm_agp_backend.o ttm_memory.o ttm_tt.o ttm_bo.o \
 	ttm_bo_util.o ttm_bo_vm.o ttm_module.o \
 	ttm_object.o ttm_lock.o ttm_execbuf_util.o ttm_page_alloc.o \
-	ttm_bo_manager.o
-
-ifeq ($(CONFIG_SWIOTLB),y)
-ttm-y += ttm_page_alloc_dma.o
-endif
+	ttm_bo_manager.o ttm_page_alloc_dma.o
 
 obj-$(CONFIG_DRM_TTM) += ttm.o
diff --git a/drivers/gpu/drm/ttm/ttm_bo.c b/drivers/gpu/drm/ttm/ttm_bo.c
index f1a857ec1021..8d5a646ebe6a 100644
--- a/drivers/gpu/drm/ttm/ttm_bo.c
+++ b/drivers/gpu/drm/ttm/ttm_bo.c
@@ -429,8 +429,20 @@ static void ttm_bo_cleanup_refs_or_queue(struct ttm_buffer_object *bo)
 		sync_obj = driver->sync_obj_ref(bo->sync_obj);
 	spin_unlock(&bdev->fence_lock);
 
-	if (!ret)
+	if (!ret) {
+
+		/*
+		 * Make NO_EVICT bos immediately available to
+		 * shrinkers, now that they are queued for
+		 * destruction.
+		 */
+		if (bo->mem.placement & TTM_PL_FLAG_NO_EVICT) {
+			bo->mem.placement &= ~TTM_PL_FLAG_NO_EVICT;
+			ttm_bo_add_to_lru(bo);
+		}
+
 		ww_mutex_unlock(&bo->resv->lock);
+	}
 
 	kref_get(&bo->list_kref);
 	list_add_tail(&bo->ddestroy, &bdev->ddestroy);
@@ -986,24 +998,32 @@ out_unlock:
 	return ret;
 }
 
-static int ttm_bo_mem_compat(struct ttm_placement *placement,
-			     struct ttm_mem_reg *mem)
+static bool ttm_bo_mem_compat(struct ttm_placement *placement,
+			      struct ttm_mem_reg *mem,
+			      uint32_t *new_flags)
 {
 	int i;
 
 	if (mem->mm_node && placement->lpfn != 0 &&
 	    (mem->start < placement->fpfn ||
 	     mem->start + mem->num_pages > placement->lpfn))
-		return -1;
+		return false;
 
 	for (i = 0; i < placement->num_placement; i++) {
-		if ((placement->placement[i] & mem->placement &
-			TTM_PL_MASK_CACHING) &&
-			(placement->placement[i] & mem->placement &
-			TTM_PL_MASK_MEM))
-			return i;
+		*new_flags = placement->placement[i];
+		if ((*new_flags & mem->placement & TTM_PL_MASK_CACHING) &&
+		    (*new_flags & mem->placement & TTM_PL_MASK_MEM))
+			return true;
 	}
-	return -1;
+
+	for (i = 0; i < placement->num_busy_placement; i++) {
+		*new_flags = placement->busy_placement[i];
+		if ((*new_flags & mem->placement & TTM_PL_MASK_CACHING) &&
+		    (*new_flags & mem->placement & TTM_PL_MASK_MEM))
+			return true;
+	}
+
+	return false;
 }
 
 int ttm_bo_validate(struct ttm_buffer_object *bo,
@@ -1012,6 +1032,7 @@ int ttm_bo_validate(struct ttm_buffer_object *bo,
 			bool no_wait_gpu)
 {
 	int ret;
+	uint32_t new_flags;
 
 	lockdep_assert_held(&bo->resv->lock.base);
 	/* Check that range is valid */
@@ -1022,8 +1043,7 @@ int ttm_bo_validate(struct ttm_buffer_object *bo,
 	/*
 	 * Check whether we need to move buffer.
 	 */
-	ret = ttm_bo_mem_compat(placement, &bo->mem);
-	if (ret < 0) {
+	if (!ttm_bo_mem_compat(placement, &bo->mem, &new_flags)) {
 		ret = ttm_bo_move_buffer(bo, placement, interruptible,
 					 no_wait_gpu);
 		if (ret)
@@ -1033,7 +1053,7 @@ int ttm_bo_validate(struct ttm_buffer_object *bo,
 		 * Use the access and other non-mapping-related flag bits from
 		 * the compatible memory placement flags to the active flags
 		 */
-		ttm_flag_masked(&bo->mem.placement, placement->placement[ret],
+		ttm_flag_masked(&bo->mem.placement, new_flags,
 				~TTM_PL_MASK_MEMTYPE);
 	}
 	/*
diff --git a/drivers/gpu/drm/ttm/ttm_bo_util.c b/drivers/gpu/drm/ttm/ttm_bo_util.c
index 7cc904d3a4d1..4834c463c38b 100644
--- a/drivers/gpu/drm/ttm/ttm_bo_util.c
+++ b/drivers/gpu/drm/ttm/ttm_bo_util.c
@@ -343,19 +343,25 @@ int ttm_bo_move_memcpy(struct ttm_buffer_object *bo,
 	if (ret)
 		goto out;
 
+	/*
+	 * Single TTM move. NOP.
+	 */
 	if (old_iomap == NULL && new_iomap == NULL)
 		goto out2;
+
+	/*
+	 * Move nonexistent data. NOP.
+	 */
 	if (old_iomap == NULL && ttm == NULL)
 		goto out2;
 
-	if (ttm->state == tt_unpopulated) {
+	/*
+	 * TTM might be null for moves within the same region.
+	 */
+	if (ttm && ttm->state == tt_unpopulated) {
 		ret = ttm->bdev->driver->ttm_tt_populate(ttm);
-		if (ret) {
-			/* if we fail here don't nuke the mm node
-			 * as the bo still owns it */
-			old_copy.mm_node = NULL;
+		if (ret)
 			goto out1;
-		}
 	}
 
 	add = 0;
@@ -381,11 +387,8 @@ int ttm_bo_move_memcpy(struct ttm_buffer_object *bo,
 						   prot);
 		} else
 			ret = ttm_copy_io_page(new_iomap, old_iomap, page);
-		if (ret) {
-			/* failing here, means keep old copy as-is */
-			old_copy.mm_node = NULL;
+		if (ret)
 			goto out1;
-		}
 	}
 	mb();
 out2:
@@ -403,7 +406,12 @@ out1:
 	ttm_mem_reg_iounmap(bdev, old_mem, new_iomap);
 out:
 	ttm_mem_reg_iounmap(bdev, &old_copy, old_iomap);
-	ttm_bo_mem_put(bo, &old_copy);
+
+	/*
+	 * On error, keep the mm node!
+	 */
+	if (!ret)
+		ttm_bo_mem_put(bo, &old_copy);
 	return ret;
 }
 EXPORT_SYMBOL(ttm_bo_move_memcpy);
diff --git a/drivers/gpu/drm/ttm/ttm_bo_vm.c b/drivers/gpu/drm/ttm/ttm_bo_vm.c
index 1006c15445e9..ac617f3ecd0c 100644
--- a/drivers/gpu/drm/ttm/ttm_bo_vm.c
+++ b/drivers/gpu/drm/ttm/ttm_bo_vm.c
@@ -41,6 +41,51 @@
 
 #define TTM_BO_VM_NUM_PREFAULT 16
 
+static int ttm_bo_vm_fault_idle(struct ttm_buffer_object *bo,
+				struct vm_area_struct *vma,
+				struct vm_fault *vmf)
+{
+	struct ttm_bo_device *bdev = bo->bdev;
+	int ret = 0;
+
+	spin_lock(&bdev->fence_lock);
+	if (likely(!test_bit(TTM_BO_PRIV_FLAG_MOVING, &bo->priv_flags)))
+		goto out_unlock;
+
+	/*
+	 * Quick non-stalling check for idle.
+	 */
+	ret = ttm_bo_wait(bo, false, false, true);
+	if (likely(ret == 0))
+		goto out_unlock;
+
+	/*
+	 * If possible, avoid waiting for GPU with mmap_sem
+	 * held.
+	 */
+	if (vmf->flags & FAULT_FLAG_ALLOW_RETRY) {
+		ret = VM_FAULT_RETRY;
+		if (vmf->flags & FAULT_FLAG_RETRY_NOWAIT)
+			goto out_unlock;
+
+		up_read(&vma->vm_mm->mmap_sem);
+		(void) ttm_bo_wait(bo, false, true, false);
+		goto out_unlock;
+	}
+
+	/*
+	 * Ordinary wait.
+	 */
+	ret = ttm_bo_wait(bo, false, true, false);
+	if (unlikely(ret != 0))
+		ret = (ret != -ERESTARTSYS) ? VM_FAULT_SIGBUS :
+			VM_FAULT_NOPAGE;
+
+out_unlock:
+	spin_unlock(&bdev->fence_lock);
+	return ret;
+}
+
 static int ttm_bo_vm_fault(struct vm_area_struct *vma, struct vm_fault *vmf)
 {
 	struct ttm_buffer_object *bo = (struct ttm_buffer_object *)
@@ -57,6 +102,7 @@ static int ttm_bo_vm_fault(struct vm_area_struct *vma, struct vm_fault *vmf)
 	int retval = VM_FAULT_NOPAGE;
 	struct ttm_mem_type_manager *man =
 		&bdev->man[bo->mem.mem_type];
+	struct vm_area_struct cvma;
 
 	/*
 	 * Work around locking order reversal in fault / nopfn
@@ -91,18 +137,11 @@ static int ttm_bo_vm_fault(struct vm_area_struct *vma, struct vm_fault *vmf)
 	 * Wait for buffer data in transit, due to a pipelined
 	 * move.
 	 */
-
-	spin_lock(&bdev->fence_lock);
-	if (test_bit(TTM_BO_PRIV_FLAG_MOVING, &bo->priv_flags)) {
-		ret = ttm_bo_wait(bo, false, true, false);
-		spin_unlock(&bdev->fence_lock);
-		if (unlikely(ret != 0)) {
-			retval = (ret != -ERESTARTSYS) ?
-			    VM_FAULT_SIGBUS : VM_FAULT_NOPAGE;
-			goto out_unlock;
-		}
-	} else
-		spin_unlock(&bdev->fence_lock);
+	ret = ttm_bo_vm_fault_idle(bo, vma, vmf);
+	if (unlikely(ret != 0)) {
+		retval = ret;
+		goto out_unlock;
+	}
 
 	ret = ttm_mem_io_lock(man, true);
 	if (unlikely(ret != 0)) {
@@ -126,26 +165,21 @@ static int ttm_bo_vm_fault(struct vm_area_struct *vma, struct vm_fault *vmf)
 	}
 
 	/*
-	 * Strictly, we're not allowed to modify vma->vm_page_prot here,
-	 * since the mmap_sem is only held in read mode. However, we
-	 * modify only the caching bits of vma->vm_page_prot and
-	 * consider those bits protected by
-	 * the bo->mutex, as we should be the only writers.
-	 * There shouldn't really be any readers of these bits except
-	 * within vm_insert_mixed()? fork?
-	 *
-	 * TODO: Add a list of vmas to the bo, and change the
-	 * vma->vm_page_prot when the object changes caching policy, with
-	 * the correct locks held.
+	 * Make a local vma copy to modify the page_prot member
+	 * and vm_flags if necessary. The vma parameter is protected
+	 * by mmap_sem in write mode.
 	 */
+	cvma = *vma;
+	cvma.vm_page_prot = vm_get_page_prot(cvma.vm_flags);
+
 	if (bo->mem.bus.is_iomem) {
-		vma->vm_page_prot = ttm_io_prot(bo->mem.placement,
-						vma->vm_page_prot);
+		cvma.vm_page_prot = ttm_io_prot(bo->mem.placement,
+						cvma.vm_page_prot);
 	} else {
 		ttm = bo->ttm;
-		vma->vm_page_prot = (bo->mem.placement & TTM_PL_FLAG_CACHED) ?
-		    vm_get_page_prot(vma->vm_flags) :
-		    ttm_io_prot(bo->mem.placement, vma->vm_page_prot);
+		if (!(bo->mem.placement & TTM_PL_FLAG_CACHED))
+			cvma.vm_page_prot = ttm_io_prot(bo->mem.placement,
+							cvma.vm_page_prot);
 
 		/* Allocate all page at once, most common usage */
 		if (ttm->bdev->driver->ttm_tt_populate(ttm)) {
@@ -172,7 +206,7 @@ static int ttm_bo_vm_fault(struct vm_area_struct *vma, struct vm_fault *vmf)
 			pfn = page_to_pfn(page);
 		}
 
-		ret = vm_insert_mixed(vma, address, pfn);
+		ret = vm_insert_mixed(&cvma, address, pfn);
 		/*
 		 * Somebody beat us to this PTE or prefaulting to
 		 * an already populated PTE, or prefaulting error.
diff --git a/drivers/gpu/drm/ttm/ttm_page_alloc_dma.c b/drivers/gpu/drm/ttm/ttm_page_alloc_dma.c
index 7957beeeaf73..fb8259f69839 100644
--- a/drivers/gpu/drm/ttm/ttm_page_alloc_dma.c
+++ b/drivers/gpu/drm/ttm/ttm_page_alloc_dma.c
@@ -33,6 +33,7 @@
  *   when freed).
  */
 
+#if defined(CONFIG_SWIOTLB) || defined(CONFIG_INTEL_IOMMU)
 #define pr_fmt(fmt) "[TTM] " fmt
 
 #include <linux/dma-mapping.h>
@@ -1142,3 +1143,5 @@ int ttm_dma_page_alloc_debugfs(struct seq_file *m, void *data)
 	return 0;
 }
 EXPORT_SYMBOL_GPL(ttm_dma_page_alloc_debugfs);
+
+#endif
diff --git a/drivers/gpu/drm/udl/Kconfig b/drivers/gpu/drm/udl/Kconfig
index 6222af19f456..f02528686cd5 100644
--- a/drivers/gpu/drm/udl/Kconfig
+++ b/drivers/gpu/drm/udl/Kconfig
@@ -8,6 +8,7 @@ config DRM_UDL
 	select FB_SYS_IMAGEBLIT
 	select FB_DEFERRED_IO
 	select DRM_KMS_HELPER
+        select DRM_KMS_FB_HELPER
 	help
 	  This is a KMS driver for the USB displaylink video adapters.
           Say M/Y to add support for these devices via drm/kms interfaces.
diff --git a/drivers/gpu/drm/udl/udl_drv.c b/drivers/gpu/drm/udl/udl_drv.c
index 7650dc0d78ce..3ddd6cd98ac1 100644
--- a/drivers/gpu/drm/udl/udl_drv.c
+++ b/drivers/gpu/drm/udl/udl_drv.c
@@ -77,7 +77,6 @@ static struct drm_driver driver = {
 	.unload = udl_driver_unload,
 
 	/* gem hooks */
-	.gem_init_object = udl_gem_init_object,
 	.gem_free_object = udl_gem_free_object,
 	.gem_vm_ops = &udl_gem_vm_ops,
 
diff --git a/drivers/gpu/drm/udl/udl_drv.h b/drivers/gpu/drm/udl/udl_drv.h
index 56aec9409fa3..1fbf7b357f16 100644
--- a/drivers/gpu/drm/udl/udl_drv.h
+++ b/drivers/gpu/drm/udl/udl_drv.h
@@ -115,7 +115,6 @@ int udl_dumb_create(struct drm_file *file_priv,
 int udl_gem_mmap(struct drm_file *file_priv, struct drm_device *dev,
 		 uint32_t handle, uint64_t *offset);
 
-int udl_gem_init_object(struct drm_gem_object *obj);
 void udl_gem_free_object(struct drm_gem_object *gem_obj);
 struct udl_gem_object *udl_gem_alloc_object(struct drm_device *dev,
 					    size_t size);
diff --git a/drivers/gpu/drm/udl/udl_gem.c b/drivers/gpu/drm/udl/udl_gem.c
index 8bf646183bac..24ffbe990736 100644
--- a/drivers/gpu/drm/udl/udl_gem.c
+++ b/drivers/gpu/drm/udl/udl_gem.c
@@ -107,13 +107,6 @@ int udl_gem_fault(struct vm_area_struct *vma, struct vm_fault *vmf)
 	}
 }
 
-int udl_gem_init_object(struct drm_gem_object *obj)
-{
-	BUG();
-
-	return 0;
-}
-
 static int udl_gem_get_pages(struct udl_gem_object *obj, gfp_t gfpmask)
 {
 	struct page **pages;
diff --git a/drivers/gpu/drm/via/via_mm.c b/drivers/gpu/drm/via/via_mm.c
index 7e3ad87c366c..927889105483 100644
--- a/drivers/gpu/drm/via/via_mm.c
+++ b/drivers/gpu/drm/via/via_mm.c
@@ -79,7 +79,7 @@ int via_final_context(struct drm_device *dev, int context)
 
 	/* Linux specific until context tracking code gets ported to BSD */
 	/* Last context, perform cleanup */
-	if (dev->ctx_count == 1 && dev->dev_private) {
+	if (list_is_singular(&dev->ctxlist) && dev->dev_private) {
 		DRM_DEBUG("Last Context\n");
 		drm_irq_uninstall(dev);
 		via_cleanup_futex(dev_priv);
diff --git a/drivers/gpu/drm/vmwgfx/vmwgfx_buffer.c b/drivers/gpu/drm/vmwgfx/vmwgfx_buffer.c
index 96dc84dc34d0..7776e6f0aef6 100644
--- a/drivers/gpu/drm/vmwgfx/vmwgfx_buffer.c
+++ b/drivers/gpu/drm/vmwgfx/vmwgfx_buffer.c
@@ -141,37 +141,374 @@ struct ttm_placement vmw_srf_placement = {
 };
 
 struct vmw_ttm_tt {
-	struct ttm_tt ttm;
+	struct ttm_dma_tt dma_ttm;
 	struct vmw_private *dev_priv;
 	int gmr_id;
+	struct sg_table sgt;
+	struct vmw_sg_table vsgt;
+	uint64_t sg_alloc_size;
+	bool mapped;
 };
 
+/**
+ * Helper functions to advance a struct vmw_piter iterator.
+ *
+ * @viter: Pointer to the iterator.
+ *
+ * These functions return false if past the end of the list,
+ * true otherwise. Functions are selected depending on the current
+ * DMA mapping mode.
+ */
+static bool __vmw_piter_non_sg_next(struct vmw_piter *viter)
+{
+	return ++(viter->i) < viter->num_pages;
+}
+
+static bool __vmw_piter_sg_next(struct vmw_piter *viter)
+{
+	return __sg_page_iter_next(&viter->iter);
+}
+
+
+/**
+ * Helper functions to return a pointer to the current page.
+ *
+ * @viter: Pointer to the iterator
+ *
+ * These functions return a pointer to the page currently
+ * pointed to by @viter. Functions are selected depending on the
+ * current mapping mode.
+ */
+static struct page *__vmw_piter_non_sg_page(struct vmw_piter *viter)
+{
+	return viter->pages[viter->i];
+}
+
+static struct page *__vmw_piter_sg_page(struct vmw_piter *viter)
+{
+	return sg_page_iter_page(&viter->iter);
+}
+
+
+/**
+ * Helper functions to return the DMA address of the current page.
+ *
+ * @viter: Pointer to the iterator
+ *
+ * These functions return the DMA address of the page currently
+ * pointed to by @viter. Functions are selected depending on the
+ * current mapping mode.
+ */
+static dma_addr_t __vmw_piter_phys_addr(struct vmw_piter *viter)
+{
+	return page_to_phys(viter->pages[viter->i]);
+}
+
+static dma_addr_t __vmw_piter_dma_addr(struct vmw_piter *viter)
+{
+	return viter->addrs[viter->i];
+}
+
+static dma_addr_t __vmw_piter_sg_addr(struct vmw_piter *viter)
+{
+	return sg_page_iter_dma_address(&viter->iter);
+}
+
+
+/**
+ * vmw_piter_start - Initialize a struct vmw_piter.
+ *
+ * @viter: Pointer to the iterator to initialize
+ * @vsgt: Pointer to a struct vmw_sg_table to initialize from
+ *
+ * Note that we're following the convention of __sg_page_iter_start, so that
+ * the iterator doesn't point to a valid page after initialization; it has
+ * to be advanced one step first.
+ */
+void vmw_piter_start(struct vmw_piter *viter, const struct vmw_sg_table *vsgt,
+		     unsigned long p_offset)
+{
+	viter->i = p_offset - 1;
+	viter->num_pages = vsgt->num_pages;
+	switch (vsgt->mode) {
+	case vmw_dma_phys:
+		viter->next = &__vmw_piter_non_sg_next;
+		viter->dma_address = &__vmw_piter_phys_addr;
+		viter->page = &__vmw_piter_non_sg_page;
+		viter->pages = vsgt->pages;
+		break;
+	case vmw_dma_alloc_coherent:
+		viter->next = &__vmw_piter_non_sg_next;
+		viter->dma_address = &__vmw_piter_dma_addr;
+		viter->page = &__vmw_piter_non_sg_page;
+		viter->addrs = vsgt->addrs;
+		break;
+	case vmw_dma_map_populate:
+	case vmw_dma_map_bind:
+		viter->next = &__vmw_piter_sg_next;
+		viter->dma_address = &__vmw_piter_sg_addr;
+		viter->page = &__vmw_piter_sg_page;
+		__sg_page_iter_start(&viter->iter, vsgt->sgt->sgl,
+				     vsgt->sgt->orig_nents, p_offset);
+		break;
+	default:
+		BUG();
+	}
+}
+
+/**
+ * vmw_ttm_unmap_from_dma - unmap  device addresses previsouly mapped for
+ * TTM pages
+ *
+ * @vmw_tt: Pointer to a struct vmw_ttm_backend
+ *
+ * Used to free dma mappings previously mapped by vmw_ttm_map_for_dma.
+ */
+static void vmw_ttm_unmap_from_dma(struct vmw_ttm_tt *vmw_tt)
+{
+	struct device *dev = vmw_tt->dev_priv->dev->dev;
+
+	dma_unmap_sg(dev, vmw_tt->sgt.sgl, vmw_tt->sgt.nents,
+		DMA_BIDIRECTIONAL);
+	vmw_tt->sgt.nents = vmw_tt->sgt.orig_nents;
+}
+
+/**
+ * vmw_ttm_map_for_dma - map TTM pages to get device addresses
+ *
+ * @vmw_tt: Pointer to a struct vmw_ttm_backend
+ *
+ * This function is used to get device addresses from the kernel DMA layer.
+ * However, it's violating the DMA API in that when this operation has been
+ * performed, it's illegal for the CPU to write to the pages without first
+ * unmapping the DMA mappings, or calling dma_sync_sg_for_cpu(). It is
+ * therefore only legal to call this function if we know that the function
+ * dma_sync_sg_for_cpu() is a NOP, and dma_sync_sg_for_device() is at most
+ * a CPU write buffer flush.
+ */
+static int vmw_ttm_map_for_dma(struct vmw_ttm_tt *vmw_tt)
+{
+	struct device *dev = vmw_tt->dev_priv->dev->dev;
+	int ret;
+
+	ret = dma_map_sg(dev, vmw_tt->sgt.sgl, vmw_tt->sgt.orig_nents,
+			 DMA_BIDIRECTIONAL);
+	if (unlikely(ret == 0))
+		return -ENOMEM;
+
+	vmw_tt->sgt.nents = ret;
+
+	return 0;
+}
+
+/**
+ * vmw_ttm_map_dma - Make sure TTM pages are visible to the device
+ *
+ * @vmw_tt: Pointer to a struct vmw_ttm_tt
+ *
+ * Select the correct function for and make sure the TTM pages are
+ * visible to the device. Allocate storage for the device mappings.
+ * If a mapping has already been performed, indicated by the storage
+ * pointer being non NULL, the function returns success.
+ */
+static int vmw_ttm_map_dma(struct vmw_ttm_tt *vmw_tt)
+{
+	struct vmw_private *dev_priv = vmw_tt->dev_priv;
+	struct ttm_mem_global *glob = vmw_mem_glob(dev_priv);
+	struct vmw_sg_table *vsgt = &vmw_tt->vsgt;
+	struct vmw_piter iter;
+	dma_addr_t old;
+	int ret = 0;
+	static size_t sgl_size;
+	static size_t sgt_size;
+
+	if (vmw_tt->mapped)
+		return 0;
+
+	vsgt->mode = dev_priv->map_mode;
+	vsgt->pages = vmw_tt->dma_ttm.ttm.pages;
+	vsgt->num_pages = vmw_tt->dma_ttm.ttm.num_pages;
+	vsgt->addrs = vmw_tt->dma_ttm.dma_address;
+	vsgt->sgt = &vmw_tt->sgt;
+
+	switch (dev_priv->map_mode) {
+	case vmw_dma_map_bind:
+	case vmw_dma_map_populate:
+		if (unlikely(!sgl_size)) {
+			sgl_size = ttm_round_pot(sizeof(struct scatterlist));
+			sgt_size = ttm_round_pot(sizeof(struct sg_table));
+		}
+		vmw_tt->sg_alloc_size = sgt_size + sgl_size * vsgt->num_pages;
+		ret = ttm_mem_global_alloc(glob, vmw_tt->sg_alloc_size, false,
+					   true);
+		if (unlikely(ret != 0))
+			return ret;
+
+		ret = sg_alloc_table_from_pages(&vmw_tt->sgt, vsgt->pages,
+						vsgt->num_pages, 0,
+						(unsigned long)
+						vsgt->num_pages << PAGE_SHIFT,
+						GFP_KERNEL);
+		if (unlikely(ret != 0))
+			goto out_sg_alloc_fail;
+
+		if (vsgt->num_pages > vmw_tt->sgt.nents) {
+			uint64_t over_alloc =
+				sgl_size * (vsgt->num_pages -
+					    vmw_tt->sgt.nents);
+
+			ttm_mem_global_free(glob, over_alloc);
+			vmw_tt->sg_alloc_size -= over_alloc;
+		}
+
+		ret = vmw_ttm_map_for_dma(vmw_tt);
+		if (unlikely(ret != 0))
+			goto out_map_fail;
+
+		break;
+	default:
+		break;
+	}
+
+	old = ~((dma_addr_t) 0);
+	vmw_tt->vsgt.num_regions = 0;
+	for (vmw_piter_start(&iter, vsgt, 0); vmw_piter_next(&iter);) {
+		dma_addr_t cur = vmw_piter_dma_addr(&iter);
+
+		if (cur != old + PAGE_SIZE)
+			vmw_tt->vsgt.num_regions++;
+		old = cur;
+	}
+
+	vmw_tt->mapped = true;
+	return 0;
+
+out_map_fail:
+	sg_free_table(vmw_tt->vsgt.sgt);
+	vmw_tt->vsgt.sgt = NULL;
+out_sg_alloc_fail:
+	ttm_mem_global_free(glob, vmw_tt->sg_alloc_size);
+	return ret;
+}
+
+/**
+ * vmw_ttm_unmap_dma - Tear down any TTM page device mappings
+ *
+ * @vmw_tt: Pointer to a struct vmw_ttm_tt
+ *
+ * Tear down any previously set up device DMA mappings and free
+ * any storage space allocated for them. If there are no mappings set up,
+ * this function is a NOP.
+ */
+static void vmw_ttm_unmap_dma(struct vmw_ttm_tt *vmw_tt)
+{
+	struct vmw_private *dev_priv = vmw_tt->dev_priv;
+
+	if (!vmw_tt->vsgt.sgt)
+		return;
+
+	switch (dev_priv->map_mode) {
+	case vmw_dma_map_bind:
+	case vmw_dma_map_populate:
+		vmw_ttm_unmap_from_dma(vmw_tt);
+		sg_free_table(vmw_tt->vsgt.sgt);
+		vmw_tt->vsgt.sgt = NULL;
+		ttm_mem_global_free(vmw_mem_glob(dev_priv),
+				    vmw_tt->sg_alloc_size);
+		break;
+	default:
+		break;
+	}
+	vmw_tt->mapped = false;
+}
+
 static int vmw_ttm_bind(struct ttm_tt *ttm, struct ttm_mem_reg *bo_mem)
 {
-	struct vmw_ttm_tt *vmw_be = container_of(ttm, struct vmw_ttm_tt, ttm);
+	struct vmw_ttm_tt *vmw_be =
+		container_of(ttm, struct vmw_ttm_tt, dma_ttm.ttm);
+	int ret;
+
+	ret = vmw_ttm_map_dma(vmw_be);
+	if (unlikely(ret != 0))
+		return ret;
 
 	vmw_be->gmr_id = bo_mem->start;
 
-	return vmw_gmr_bind(vmw_be->dev_priv, ttm->pages,
+	return vmw_gmr_bind(vmw_be->dev_priv, &vmw_be->vsgt,
 			    ttm->num_pages, vmw_be->gmr_id);
 }
 
 static int vmw_ttm_unbind(struct ttm_tt *ttm)
 {
-	struct vmw_ttm_tt *vmw_be = container_of(ttm, struct vmw_ttm_tt, ttm);
+	struct vmw_ttm_tt *vmw_be =
+		container_of(ttm, struct vmw_ttm_tt, dma_ttm.ttm);
 
 	vmw_gmr_unbind(vmw_be->dev_priv, vmw_be->gmr_id);
+
+	if (vmw_be->dev_priv->map_mode == vmw_dma_map_bind)
+		vmw_ttm_unmap_dma(vmw_be);
+
 	return 0;
 }
 
 static void vmw_ttm_destroy(struct ttm_tt *ttm)
 {
-	struct vmw_ttm_tt *vmw_be = container_of(ttm, struct vmw_ttm_tt, ttm);
-
-	ttm_tt_fini(ttm);
+	struct vmw_ttm_tt *vmw_be =
+		container_of(ttm, struct vmw_ttm_tt, dma_ttm.ttm);
+
+	vmw_ttm_unmap_dma(vmw_be);
+	if (vmw_be->dev_priv->map_mode == vmw_dma_alloc_coherent)
+		ttm_dma_tt_fini(&vmw_be->dma_ttm);
+	else
+		ttm_tt_fini(ttm);
 	kfree(vmw_be);
 }
 
+static int vmw_ttm_populate(struct ttm_tt *ttm)
+{
+	struct vmw_ttm_tt *vmw_tt =
+		container_of(ttm, struct vmw_ttm_tt, dma_ttm.ttm);
+	struct vmw_private *dev_priv = vmw_tt->dev_priv;
+	struct ttm_mem_global *glob = vmw_mem_glob(dev_priv);
+	int ret;
+
+	if (ttm->state != tt_unpopulated)
+		return 0;
+
+	if (dev_priv->map_mode == vmw_dma_alloc_coherent) {
+		size_t size =
+			ttm_round_pot(ttm->num_pages * sizeof(dma_addr_t));
+		ret = ttm_mem_global_alloc(glob, size, false, true);
+		if (unlikely(ret != 0))
+			return ret;
+
+		ret = ttm_dma_populate(&vmw_tt->dma_ttm, dev_priv->dev->dev);
+		if (unlikely(ret != 0))
+			ttm_mem_global_free(glob, size);
+	} else
+		ret = ttm_pool_populate(ttm);
+
+	return ret;
+}
+
+static void vmw_ttm_unpopulate(struct ttm_tt *ttm)
+{
+	struct vmw_ttm_tt *vmw_tt = container_of(ttm, struct vmw_ttm_tt,
+						 dma_ttm.ttm);
+	struct vmw_private *dev_priv = vmw_tt->dev_priv;
+	struct ttm_mem_global *glob = vmw_mem_glob(dev_priv);
+
+	vmw_ttm_unmap_dma(vmw_tt);
+	if (dev_priv->map_mode == vmw_dma_alloc_coherent) {
+		size_t size =
+			ttm_round_pot(ttm->num_pages * sizeof(dma_addr_t));
+
+		ttm_dma_unpopulate(&vmw_tt->dma_ttm, dev_priv->dev->dev);
+		ttm_mem_global_free(glob, size);
+	} else
+		ttm_pool_unpopulate(ttm);
+}
+
 static struct ttm_backend_func vmw_ttm_func = {
 	.bind = vmw_ttm_bind,
 	.unbind = vmw_ttm_unbind,
@@ -183,20 +520,28 @@ struct ttm_tt *vmw_ttm_tt_create(struct ttm_bo_device *bdev,
 				 struct page *dummy_read_page)
 {
 	struct vmw_ttm_tt *vmw_be;
+	int ret;
 
-	vmw_be = kmalloc(sizeof(*vmw_be), GFP_KERNEL);
+	vmw_be = kzalloc(sizeof(*vmw_be), GFP_KERNEL);
 	if (!vmw_be)
 		return NULL;
 
-	vmw_be->ttm.func = &vmw_ttm_func;
+	vmw_be->dma_ttm.ttm.func = &vmw_ttm_func;
 	vmw_be->dev_priv = container_of(bdev, struct vmw_private, bdev);
 
-	if (ttm_tt_init(&vmw_be->ttm, bdev, size, page_flags, dummy_read_page)) {
-		kfree(vmw_be);
-		return NULL;
-	}
-
-	return &vmw_be->ttm;
+	if (vmw_be->dev_priv->map_mode == vmw_dma_alloc_coherent)
+		ret = ttm_dma_tt_init(&vmw_be->dma_ttm, bdev, size, page_flags,
+				      dummy_read_page);
+	else
+		ret = ttm_tt_init(&vmw_be->dma_ttm.ttm, bdev, size, page_flags,
+				  dummy_read_page);
+	if (unlikely(ret != 0))
+		goto out_no_init;
+
+	return &vmw_be->dma_ttm.ttm;
+out_no_init:
+	kfree(vmw_be);
+	return NULL;
 }
 
 int vmw_invalidate_caches(struct ttm_bo_device *bdev, uint32_t flags)
@@ -332,8 +677,8 @@ static int vmw_sync_obj_wait(void *sync_obj, bool lazy, bool interruptible)
 
 struct ttm_bo_driver vmw_bo_driver = {
 	.ttm_tt_create = &vmw_ttm_tt_create,
-	.ttm_tt_populate = &ttm_pool_populate,
-	.ttm_tt_unpopulate = &ttm_pool_unpopulate,
+	.ttm_tt_populate = &vmw_ttm_populate,
+	.ttm_tt_unpopulate = &vmw_ttm_unpopulate,
 	.invalidate_caches = vmw_invalidate_caches,
 	.init_mem_type = vmw_init_mem_type,
 	.evict_flags = vmw_evict_flags,
diff --git a/drivers/gpu/drm/vmwgfx/vmwgfx_drv.c b/drivers/gpu/drm/vmwgfx/vmwgfx_drv.c
index 0508f93b9795..20d5485eaf98 100644
--- a/drivers/gpu/drm/vmwgfx/vmwgfx_drv.c
+++ b/drivers/gpu/drm/vmwgfx/vmwgfx_drv.c
@@ -32,6 +32,7 @@
 #include <drm/ttm/ttm_bo_driver.h>
 #include <drm/ttm/ttm_object.h>
 #include <drm/ttm/ttm_module.h>
+#include <linux/dma_remapping.h>
 
 #define VMWGFX_DRIVER_NAME "vmwgfx"
 #define VMWGFX_DRIVER_DESC "Linux drm driver for VMware graphics devices"
@@ -185,6 +186,9 @@ static struct pci_device_id vmw_pci_id_list[] = {
 MODULE_DEVICE_TABLE(pci, vmw_pci_id_list);
 
 static int enable_fbdev = IS_ENABLED(CONFIG_DRM_VMWGFX_FBCON);
+static int vmw_force_iommu;
+static int vmw_restrict_iommu;
+static int vmw_force_coherent;
 
 static int vmw_probe(struct pci_dev *, const struct pci_device_id *);
 static void vmw_master_init(struct vmw_master *);
@@ -193,6 +197,13 @@ static int vmwgfx_pm_notifier(struct notifier_block *nb, unsigned long val,
 
 MODULE_PARM_DESC(enable_fbdev, "Enable vmwgfx fbdev");
 module_param_named(enable_fbdev, enable_fbdev, int, 0600);
+MODULE_PARM_DESC(force_dma_api, "Force using the DMA API for TTM pages");
+module_param_named(force_dma_api, vmw_force_iommu, int, 0600);
+MODULE_PARM_DESC(restrict_iommu, "Try to limit IOMMU usage for TTM pages");
+module_param_named(restrict_iommu, vmw_restrict_iommu, int, 0600);
+MODULE_PARM_DESC(force_coherent, "Force coherent TTM pages");
+module_param_named(force_coherent, vmw_force_coherent, int, 0600);
+
 
 static void vmw_print_capabilities(uint32_t capabilities)
 {
@@ -427,12 +438,85 @@ static void vmw_get_initial_size(struct vmw_private *dev_priv)
 	dev_priv->initial_height = height;
 }
 
+/**
+ * vmw_dma_select_mode - Determine how DMA mappings should be set up for this
+ * system.
+ *
+ * @dev_priv: Pointer to a struct vmw_private
+ *
+ * This functions tries to determine the IOMMU setup and what actions
+ * need to be taken by the driver to make system pages visible to the
+ * device.
+ * If this function decides that DMA is not possible, it returns -EINVAL.
+ * The driver may then try to disable features of the device that require
+ * DMA.
+ */
+static int vmw_dma_select_mode(struct vmw_private *dev_priv)
+{
+	static const char *names[vmw_dma_map_max] = {
+		[vmw_dma_phys] = "Using physical TTM page addresses.",
+		[vmw_dma_alloc_coherent] = "Using coherent TTM pages.",
+		[vmw_dma_map_populate] = "Keeping DMA mappings.",
+		[vmw_dma_map_bind] = "Giving up DMA mappings early."};
+#ifdef CONFIG_X86
+	const struct dma_map_ops *dma_ops = get_dma_ops(dev_priv->dev->dev);
+
+#ifdef CONFIG_INTEL_IOMMU
+	if (intel_iommu_enabled) {
+		dev_priv->map_mode = vmw_dma_map_populate;
+		goto out_fixup;
+	}
+#endif
+
+	if (!(vmw_force_iommu || vmw_force_coherent)) {
+		dev_priv->map_mode = vmw_dma_phys;
+		DRM_INFO("DMA map mode: %s\n", names[dev_priv->map_mode]);
+		return 0;
+	}
+
+	dev_priv->map_mode = vmw_dma_map_populate;
+
+	if (dma_ops->sync_single_for_cpu)
+		dev_priv->map_mode = vmw_dma_alloc_coherent;
+#ifdef CONFIG_SWIOTLB
+	if (swiotlb_nr_tbl() == 0)
+		dev_priv->map_mode = vmw_dma_map_populate;
+#endif
+
+#ifdef CONFIG_INTEL_IOMMU
+out_fixup:
+#endif
+	if (dev_priv->map_mode == vmw_dma_map_populate &&
+	    vmw_restrict_iommu)
+		dev_priv->map_mode = vmw_dma_map_bind;
+
+	if (vmw_force_coherent)
+		dev_priv->map_mode = vmw_dma_alloc_coherent;
+
+#if !defined(CONFIG_SWIOTLB) && !defined(CONFIG_INTEL_IOMMU)
+	/*
+	 * No coherent page pool
+	 */
+	if (dev_priv->map_mode == vmw_dma_alloc_coherent)
+		return -EINVAL;
+#endif
+
+#else /* CONFIG_X86 */
+	dev_priv->map_mode = vmw_dma_map_populate;
+#endif /* CONFIG_X86 */
+
+	DRM_INFO("DMA map mode: %s\n", names[dev_priv->map_mode]);
+
+	return 0;
+}
+
 static int vmw_driver_load(struct drm_device *dev, unsigned long chipset)
 {
 	struct vmw_private *dev_priv;
 	int ret;
 	uint32_t svga_id;
 	enum vmw_res_type i;
+	bool refuse_dma = false;
 
 	dev_priv = kzalloc(sizeof(*dev_priv), GFP_KERNEL);
 	if (unlikely(dev_priv == NULL)) {
@@ -481,6 +565,11 @@ static int vmw_driver_load(struct drm_device *dev, unsigned long chipset)
 	}
 
 	dev_priv->capabilities = vmw_read(dev_priv, SVGA_REG_CAPABILITIES);
+	ret = vmw_dma_select_mode(dev_priv);
+	if (unlikely(ret != 0)) {
+		DRM_INFO("Restricting capabilities due to IOMMU setup.\n");
+		refuse_dma = true;
+	}
 
 	dev_priv->vram_size = vmw_read(dev_priv, SVGA_REG_VRAM_SIZE);
 	dev_priv->mmio_size = vmw_read(dev_priv, SVGA_REG_MEM_SIZE);
@@ -558,8 +647,9 @@ static int vmw_driver_load(struct drm_device *dev, unsigned long chipset)
 	}
 
 	dev_priv->has_gmr = true;
-	if (ttm_bo_init_mm(&dev_priv->bdev, VMW_PL_GMR,
-			   dev_priv->max_gmr_ids) != 0) {
+	if (((dev_priv->capabilities & (SVGA_CAP_GMR | SVGA_CAP_GMR2)) == 0) ||
+	    refuse_dma || ttm_bo_init_mm(&dev_priv->bdev, VMW_PL_GMR,
+					 dev_priv->max_gmr_ids) != 0) {
 		DRM_INFO("No GMR memory available. "
 			 "Graphics memory resources are very limited.\n");
 		dev_priv->has_gmr = false;
diff --git a/drivers/gpu/drm/vmwgfx/vmwgfx_drv.h b/drivers/gpu/drm/vmwgfx/vmwgfx_drv.h
index 150ec64af617..e401d5dbcb96 100644
--- a/drivers/gpu/drm/vmwgfx/vmwgfx_drv.h
+++ b/drivers/gpu/drm/vmwgfx/vmwgfx_drv.h
@@ -177,6 +177,58 @@ struct vmw_res_cache_entry {
 	struct vmw_resource_val_node *node;
 };
 
+/**
+ * enum vmw_dma_map_mode - indicate how to perform TTM page dma mappings.
+ */
+enum vmw_dma_map_mode {
+	vmw_dma_phys,           /* Use physical page addresses */
+	vmw_dma_alloc_coherent, /* Use TTM coherent pages */
+	vmw_dma_map_populate,   /* Unmap from DMA just after unpopulate */
+	vmw_dma_map_bind,       /* Unmap from DMA just before unbind */
+	vmw_dma_map_max
+};
+
+/**
+ * struct vmw_sg_table - Scatter/gather table for binding, with additional
+ * device-specific information.
+ *
+ * @sgt: Pointer to a struct sg_table with binding information
+ * @num_regions: Number of regions with device-address contigous pages
+ */
+struct vmw_sg_table {
+	enum vmw_dma_map_mode mode;
+	struct page **pages;
+	const dma_addr_t *addrs;
+	struct sg_table *sgt;
+	unsigned long num_regions;
+	unsigned long num_pages;
+};
+
+/**
+ * struct vmw_piter - Page iterator that iterates over a list of pages
+ * and DMA addresses that could be either a scatter-gather list or
+ * arrays
+ *
+ * @pages: Array of page pointers to the pages.
+ * @addrs: DMA addresses to the pages if coherent pages are used.
+ * @iter: Scatter-gather page iterator. Current position in SG list.
+ * @i: Current position in arrays.
+ * @num_pages: Number of pages total.
+ * @next: Function to advance the iterator. Returns false if past the list
+ * of pages, true otherwise.
+ * @dma_address: Function to return the DMA address of the current page.
+ */
+struct vmw_piter {
+	struct page **pages;
+	const dma_addr_t *addrs;
+	struct sg_page_iter iter;
+	unsigned long i;
+	unsigned long num_pages;
+	bool (*next)(struct vmw_piter *);
+	dma_addr_t (*dma_address)(struct vmw_piter *);
+	struct page *(*page)(struct vmw_piter *);
+};
+
 struct vmw_sw_context{
 	struct drm_open_hash res_ht;
 	bool res_ht_initialized;
@@ -358,6 +410,11 @@ struct vmw_private {
 
 	struct list_head res_lru[vmw_res_max];
 	uint32_t used_memory_size;
+
+	/*
+	 * DMA mapping stuff.
+	 */
+	enum vmw_dma_map_mode map_mode;
 };
 
 static inline struct vmw_surface *vmw_res_to_srf(struct vmw_resource *res)
@@ -405,7 +462,7 @@ void vmw_3d_resource_dec(struct vmw_private *dev_priv, bool hide_svga);
  */
 
 extern int vmw_gmr_bind(struct vmw_private *dev_priv,
-			struct page *pages[],
+			const struct vmw_sg_table *vsgt,
 			unsigned long num_pages,
 			int gmr_id);
 extern void vmw_gmr_unbind(struct vmw_private *dev_priv, int gmr_id);
@@ -568,6 +625,45 @@ extern struct ttm_placement vmw_evictable_placement;
 extern struct ttm_placement vmw_srf_placement;
 extern struct ttm_bo_driver vmw_bo_driver;
 extern int vmw_dma_quiescent(struct drm_device *dev);
+extern void vmw_piter_start(struct vmw_piter *viter,
+			    const struct vmw_sg_table *vsgt,
+			    unsigned long p_offs);
+
+/**
+ * vmw_piter_next - Advance the iterator one page.
+ *
+ * @viter: Pointer to the iterator to advance.
+ *
+ * Returns false if past the list of pages, true otherwise.
+ */
+static inline bool vmw_piter_next(struct vmw_piter *viter)
+{
+	return viter->next(viter);
+}
+
+/**
+ * vmw_piter_dma_addr - Return the DMA address of the current page.
+ *
+ * @viter: Pointer to the iterator
+ *
+ * Returns the DMA address of the page pointed to by @viter.
+ */
+static inline dma_addr_t vmw_piter_dma_addr(struct vmw_piter *viter)
+{
+	return viter->dma_address(viter);
+}
+
+/**
+ * vmw_piter_page - Return a pointer to the current page.
+ *
+ * @viter: Pointer to the iterator
+ *
+ * Returns the DMA address of the page pointed to by @viter.
+ */
+static inline struct page *vmw_piter_page(struct vmw_piter *viter)
+{
+	return viter->page(viter);
+}
 
 /**
  * Command submission - vmwgfx_execbuf.c
diff --git a/drivers/gpu/drm/vmwgfx/vmwgfx_gmr.c b/drivers/gpu/drm/vmwgfx/vmwgfx_gmr.c
index 1a0bf07fe54b..6ef0b035becb 100644
--- a/drivers/gpu/drm/vmwgfx/vmwgfx_gmr.c
+++ b/drivers/gpu/drm/vmwgfx/vmwgfx_gmr.c
@@ -32,9 +32,11 @@
 #define VMW_PPN_SIZE (sizeof(unsigned long))
 /* A future safe maximum remap size. */
 #define VMW_PPN_PER_REMAP ((31 * 1024) / VMW_PPN_SIZE)
+#define DMA_ADDR_INVALID ((dma_addr_t) 0)
+#define DMA_PAGE_INVALID 0UL
 
 static int vmw_gmr2_bind(struct vmw_private *dev_priv,
-			 struct page *pages[],
+			 struct vmw_piter *iter,
 			 unsigned long num_pages,
 			 int gmr_id)
 {
@@ -81,11 +83,13 @@ static int vmw_gmr2_bind(struct vmw_private *dev_priv,
 
 		for (i = 0; i < nr; ++i) {
 			if (VMW_PPN_SIZE <= 4)
-				*cmd = page_to_pfn(*pages++);
+				*cmd = vmw_piter_dma_addr(iter) >> PAGE_SHIFT;
 			else
-				*((uint64_t *)cmd) = page_to_pfn(*pages++);
+				*((uint64_t *)cmd) = vmw_piter_dma_addr(iter) >>
+					PAGE_SHIFT;
 
 			cmd += VMW_PPN_SIZE / sizeof(*cmd);
+			vmw_piter_next(iter);
 		}
 
 		num_pages -= nr;
@@ -120,22 +124,56 @@ static void vmw_gmr2_unbind(struct vmw_private *dev_priv,
 	vmw_fifo_commit(dev_priv, define_size);
 }
 
+
+static void vmw_gmr_free_descriptors(struct device *dev, dma_addr_t desc_dma,
+				     struct list_head *desc_pages)
+{
+	struct page *page, *next;
+	struct svga_guest_mem_descriptor *page_virtual;
+	unsigned int desc_per_page = PAGE_SIZE /
+		sizeof(struct svga_guest_mem_descriptor) - 1;
+
+	if (list_empty(desc_pages))
+		return;
+
+	list_for_each_entry_safe(page, next, desc_pages, lru) {
+		list_del_init(&page->lru);
+
+		if (likely(desc_dma != DMA_ADDR_INVALID)) {
+			dma_unmap_page(dev, desc_dma, PAGE_SIZE,
+				       DMA_TO_DEVICE);
+		}
+
+		page_virtual = kmap_atomic(page);
+		desc_dma = (dma_addr_t)
+			le32_to_cpu(page_virtual[desc_per_page].ppn) <<
+			PAGE_SHIFT;
+		kunmap_atomic(page_virtual);
+
+		__free_page(page);
+	}
+}
+
 /**
  * FIXME: Adjust to the ttm lowmem / highmem storage to minimize
  * the number of used descriptors.
+ *
  */
 
-static int vmw_gmr_build_descriptors(struct list_head *desc_pages,
-				     struct page *pages[],
-				     unsigned long num_pages)
+static int vmw_gmr_build_descriptors(struct device *dev,
+				     struct list_head *desc_pages,
+				     struct vmw_piter *iter,
+				     unsigned long num_pages,
+				     dma_addr_t *first_dma)
 {
-	struct page *page, *next;
+	struct page *page;
 	struct svga_guest_mem_descriptor *page_virtual = NULL;
 	struct svga_guest_mem_descriptor *desc_virtual = NULL;
 	unsigned int desc_per_page;
 	unsigned long prev_pfn;
 	unsigned long pfn;
 	int ret;
+	dma_addr_t desc_dma;
 
 	desc_per_page = PAGE_SIZE /
 	    sizeof(struct svga_guest_mem_descriptor) - 1;
@@ -148,23 +186,12 @@ static int vmw_gmr_build_descriptors(struct list_head *desc_pages,
 		}
 
 		list_add_tail(&page->lru, desc_pages);
-
-		/*
-		 * Point previous page terminating descriptor to this
-		 * page before unmapping it.
-		 */
-
-		if (likely(page_virtual != NULL)) {
-			desc_virtual->ppn = page_to_pfn(page);
-			kunmap_atomic(page_virtual);
-		}
-
 		page_virtual = kmap_atomic(page);
 		desc_virtual = page_virtual - 1;
 		prev_pfn = ~(0UL);
 
 		while (likely(num_pages != 0)) {
-			pfn = page_to_pfn(*pages);
+			pfn = vmw_piter_dma_addr(iter) >> PAGE_SHIFT;
 
 			if (pfn != prev_pfn + 1) {
 
@@ -181,104 +208,82 @@ static int vmw_gmr_build_descriptors(struct list_head *desc_pages,
 			}
 			prev_pfn = pfn;
 			--num_pages;
-			++pages;
+			vmw_piter_next(iter);
 		}
 
-		(++desc_virtual)->ppn = cpu_to_le32(0);
+		(++desc_virtual)->ppn = DMA_PAGE_INVALID;
 		desc_virtual->num_pages = cpu_to_le32(0);
+		kunmap_atomic(page_virtual);
 	}
 
-	if (likely(page_virtual != NULL))
+	desc_dma = 0;
+	list_for_each_entry_reverse(page, desc_pages, lru) {
+		page_virtual = kmap_atomic(page);
+		page_virtual[desc_per_page].ppn = cpu_to_le32
+			(desc_dma >> PAGE_SHIFT);
 		kunmap_atomic(page_virtual);
+		desc_dma = dma_map_page(dev, page, 0, PAGE_SIZE,
+					DMA_TO_DEVICE);
+
+		if (unlikely(dma_mapping_error(dev, desc_dma)))
+			goto out_err;
+	}
+	*first_dma = desc_dma;
 
 	return 0;
 out_err:
-	list_for_each_entry_safe(page, next, desc_pages, lru) {
-		list_del_init(&page->lru);
-		__free_page(page);
-	}
+	vmw_gmr_free_descriptors(dev, DMA_ADDR_INVALID, desc_pages);
 	return ret;
 }
 
-static inline void vmw_gmr_free_descriptors(struct list_head *desc_pages)
-{
-	struct page *page, *next;
-
-	list_for_each_entry_safe(page, next, desc_pages, lru) {
-		list_del_init(&page->lru);
-		__free_page(page);
-	}
-}
-
 static void vmw_gmr_fire_descriptors(struct vmw_private *dev_priv,
-				     int gmr_id, struct list_head *desc_pages)
+				     int gmr_id, dma_addr_t desc_dma)
 {
-	struct page *page;
-
-	if (unlikely(list_empty(desc_pages)))
-		return;
-
-	page = list_entry(desc_pages->next, struct page, lru);
-
 	mutex_lock(&dev_priv->hw_mutex);
 
 	vmw_write(dev_priv, SVGA_REG_GMR_ID, gmr_id);
 	wmb();
-	vmw_write(dev_priv, SVGA_REG_GMR_DESCRIPTOR, page_to_pfn(page));
+	vmw_write(dev_priv, SVGA_REG_GMR_DESCRIPTOR, desc_dma >> PAGE_SHIFT);
 	mb();
 
 	mutex_unlock(&dev_priv->hw_mutex);
 
 }
 
-/**
- * FIXME: Adjust to the ttm lowmem / highmem storage to minimize
- * the number of used descriptors.
- */
-
-static unsigned long vmw_gmr_count_descriptors(struct page *pages[],
-					unsigned long num_pages)
-{
-	unsigned long prev_pfn = ~(0UL);
-	unsigned long pfn;
-	unsigned long descriptors = 0;
-
-	while (num_pages--) {
-		pfn = page_to_pfn(*pages++);
-		if (prev_pfn + 1 != pfn)
-			++descriptors;
-		prev_pfn = pfn;
-	}
-
-	return descriptors;
-}
-
 int vmw_gmr_bind(struct vmw_private *dev_priv,
-		 struct page *pages[],
+		 const struct vmw_sg_table *vsgt,
 		 unsigned long num_pages,
 		 int gmr_id)
 {
 	struct list_head desc_pages;
+	dma_addr_t desc_dma = 0;
+	struct device *dev = dev_priv->dev->dev;
+	struct vmw_piter data_iter;
 	int ret;
 
+	vmw_piter_start(&data_iter, vsgt, 0);
+
+	if (unlikely(!vmw_piter_next(&data_iter)))
+		return 0;
+
 	if (likely(dev_priv->capabilities & SVGA_CAP_GMR2))
-		return vmw_gmr2_bind(dev_priv, pages, num_pages, gmr_id);
+		return vmw_gmr2_bind(dev_priv, &data_iter, num_pages, gmr_id);
 
 	if (unlikely(!(dev_priv->capabilities & SVGA_CAP_GMR)))
 		return -EINVAL;
 
-	if (vmw_gmr_count_descriptors(pages, num_pages) >
-	    dev_priv->max_gmr_descriptors)
+	if (vsgt->num_regions > dev_priv->max_gmr_descriptors)
 		return -EINVAL;
 
 	INIT_LIST_HEAD(&desc_pages);
 
-	ret = vmw_gmr_build_descriptors(&desc_pages, pages, num_pages);
+	ret = vmw_gmr_build_descriptors(dev, &desc_pages, &data_iter,
+					num_pages, &desc_dma);
 	if (unlikely(ret != 0))
 		return ret;
 
-	vmw_gmr_fire_descriptors(dev_priv, gmr_id, &desc_pages);
-	vmw_gmr_free_descriptors(&desc_pages);
+	vmw_gmr_fire_descriptors(dev_priv, gmr_id, desc_dma);
+	vmw_gmr_free_descriptors(dev, desc_dma, &desc_pages);
 
 	return 0;
 }
diff --git a/drivers/gpu/drm/vmwgfx/vmwgfx_ioctl.c b/drivers/gpu/drm/vmwgfx/vmwgfx_ioctl.c
index c509d40c4897..a51f48e3e917 100644
--- a/drivers/gpu/drm/vmwgfx/vmwgfx_ioctl.c
+++ b/drivers/gpu/drm/vmwgfx/vmwgfx_ioctl.c
@@ -168,7 +168,7 @@ int vmw_present_ioctl(struct drm_device *dev, void *data,
 	fb = drm_framebuffer_lookup(dev, arg->fb_id);
 	if (!fb) {
 		DRM_ERROR("Invalid framebuffer id.\n");
-		ret = -EINVAL;
+		ret = -ENOENT;
 		goto out_no_fb;
 	}
 	vfb = vmw_framebuffer_to_vfb(fb);
@@ -252,7 +252,7 @@ int vmw_present_readback_ioctl(struct drm_device *dev, void *data,
 	fb = drm_framebuffer_lookup(dev, arg->fb_id);
 	if (!fb) {
 		DRM_ERROR("Invalid framebuffer id.\n");
-		ret = -EINVAL;
+		ret = -ENOENT;
 		goto out_no_fb;
 	}
 
diff --git a/drivers/gpu/drm/vmwgfx/vmwgfx_kms.c b/drivers/gpu/drm/vmwgfx/vmwgfx_kms.c
index fc43c0601236..ecb3d867b426 100644
--- a/drivers/gpu/drm/vmwgfx/vmwgfx_kms.c
+++ b/drivers/gpu/drm/vmwgfx/vmwgfx_kms.c
@@ -1508,7 +1508,7 @@ int vmw_kms_cursor_bypass_ioctl(struct drm_device *dev, void *data,
 
 	obj = drm_mode_object_find(dev, arg->crtc_id, DRM_MODE_OBJECT_CRTC);
 	if (!obj) {
-		ret = -EINVAL;
+		ret = -ENOENT;
 		goto out;
 	}
 
diff --git a/drivers/gpu/drm/vmwgfx/vmwgfx_resource.c b/drivers/gpu/drm/vmwgfx/vmwgfx_resource.c
index 37fb4befec82..252501a54def 100644
--- a/drivers/gpu/drm/vmwgfx/vmwgfx_resource.c
+++ b/drivers/gpu/drm/vmwgfx/vmwgfx_resource.c
@@ -32,6 +32,8 @@
 #include <drm/drmP.h>
 #include "vmwgfx_resource_priv.h"
 
+#define VMW_RES_EVICT_ERR_COUNT 10
+
 struct vmw_user_dma_buffer {
 	struct ttm_base_object base;
 	struct vmw_dma_buffer dma;
@@ -1091,8 +1093,9 @@ vmw_resource_backoff_reservation(struct ww_acquire_ctx *ticket,
  *                         to a backup buffer.
  *
  * @res:            The resource to evict.
+ * @interruptible:  Whether to wait interruptible.
  */
-int vmw_resource_do_evict(struct vmw_resource *res)
+int vmw_resource_do_evict(struct vmw_resource *res, bool interruptible)
 {
 	struct ttm_validate_buffer val_buf;
 	const struct vmw_res_func *func = res->func;
@@ -1102,7 +1105,8 @@ int vmw_resource_do_evict(struct vmw_resource *res)
 	BUG_ON(!func->may_evict);
 
 	val_buf.bo = NULL;
-	ret = vmw_resource_check_buffer(res, &ticket, true, &val_buf);
+	ret = vmw_resource_check_buffer(res, &ticket, interruptible,
+					&val_buf);
 	if (unlikely(ret != 0))
 		return ret;
 
@@ -1141,6 +1145,7 @@ int vmw_resource_validate(struct vmw_resource *res)
 	struct vmw_private *dev_priv = res->dev_priv;
 	struct list_head *lru_list = &dev_priv->res_lru[res->func->res_type];
 	struct ttm_validate_buffer val_buf;
+	unsigned err_count = 0;
 
 	if (likely(!res->func->may_evict))
 		return 0;
@@ -1155,7 +1160,7 @@ int vmw_resource_validate(struct vmw_resource *res)
 
 		write_lock(&dev_priv->resource_lock);
 		if (list_empty(lru_list) || !res->func->may_evict) {
-			DRM_ERROR("Out of device device id entries "
+			DRM_ERROR("Out of device device resources "
 				  "for %s.\n", res->func->type_name);
 			ret = -EBUSY;
 			write_unlock(&dev_priv->resource_lock);
@@ -1168,7 +1173,19 @@ int vmw_resource_validate(struct vmw_resource *res)
 		list_del_init(&evict_res->lru_head);
 
 		write_unlock(&dev_priv->resource_lock);
-		vmw_resource_do_evict(evict_res);
+
+		ret = vmw_resource_do_evict(evict_res, true);
+		if (unlikely(ret != 0)) {
+			write_lock(&dev_priv->resource_lock);
+			list_add_tail(&evict_res->lru_head, lru_list);
+			write_unlock(&dev_priv->resource_lock);
+			if (ret == -ERESTARTSYS ||
+			    ++err_count > VMW_RES_EVICT_ERR_COUNT) {
+				vmw_resource_unreference(&evict_res);
+				goto out_no_validate;
+			}
+		}
+
 		vmw_resource_unreference(&evict_res);
 	} while (1);
 
@@ -1253,13 +1270,15 @@ bool vmw_resource_needs_backup(const struct vmw_resource *res)
  * @type:           The resource type to evict
  *
  * To avoid thrashing starvation or as part of the hibernation sequence,
- * evict all evictable resources of a specific type.
+ * try to evict all evictable resources of a specific type.
  */
 static void vmw_resource_evict_type(struct vmw_private *dev_priv,
 				    enum vmw_res_type type)
 {
 	struct list_head *lru_list = &dev_priv->res_lru[type];
 	struct vmw_resource *evict_res;
+	unsigned err_count = 0;
+	int ret;
 
 	do {
 		write_lock(&dev_priv->resource_lock);
@@ -1272,7 +1291,18 @@ static void vmw_resource_evict_type(struct vmw_private *dev_priv,
 					 lru_head));
 		list_del_init(&evict_res->lru_head);
 		write_unlock(&dev_priv->resource_lock);
-		vmw_resource_do_evict(evict_res);
+
+		ret = vmw_resource_do_evict(evict_res, false);
+		if (unlikely(ret != 0)) {
+			write_lock(&dev_priv->resource_lock);
+			list_add_tail(&evict_res->lru_head, lru_list);
+			write_unlock(&dev_priv->resource_lock);
+			if (++err_count > VMW_RES_EVICT_ERR_COUNT) {
+				vmw_resource_unreference(&evict_res);
+				return;
+			}
+		}
+
 		vmw_resource_unreference(&evict_res);
 	} while (1);
 
diff --git a/drivers/gpu/host1x/Kconfig b/drivers/gpu/host1x/Kconfig
index ccfd42b23606..7d6bed222542 100644
--- a/drivers/gpu/host1x/Kconfig
+++ b/drivers/gpu/host1x/Kconfig
@@ -19,6 +19,4 @@ config TEGRA_HOST1X_FIREWALL
 
 	  If unsure, choose Y.
 
-source "drivers/gpu/host1x/drm/Kconfig"
-
 endif
diff --git a/drivers/gpu/host1x/Makefile b/drivers/gpu/host1x/Makefile
index 3b037b6e0298..afa1e9e4e512 100644
--- a/drivers/gpu/host1x/Makefile
+++ b/drivers/gpu/host1x/Makefile
@@ -1,6 +1,5 @@
-ccflags-y = -Idrivers/gpu/host1x
-
 host1x-y = \
+	bus.o \
 	syncpt.o \
 	dev.o \
 	intr.o \
@@ -8,13 +7,7 @@ host1x-y = \
 	channel.o \
 	job.o \
 	debug.o \
-	hw/host1x01.o
-
-ccflags-y += -Iinclude/drm
-ccflags-$(CONFIG_DRM_TEGRA_DEBUG) += -DDEBUG
+	hw/host1x01.o \
+	hw/host1x02.o
 
-host1x-$(CONFIG_DRM_TEGRA) += drm/drm.o drm/fb.o drm/dc.o
-host1x-$(CONFIG_DRM_TEGRA) += drm/output.o drm/rgb.o drm/hdmi.o
-host1x-$(CONFIG_DRM_TEGRA) += drm/gem.o
-host1x-$(CONFIG_DRM_TEGRA) += drm/gr2d.o
 obj-$(CONFIG_TEGRA_HOST1X) += host1x.o
diff --git a/drivers/gpu/host1x/bus.c b/drivers/gpu/host1x/bus.c
new file mode 100644
index 000000000000..509383f8be03
--- /dev/null
+++ b/drivers/gpu/host1x/bus.c
@@ -0,0 +1,550 @@
+/*
+ * Copyright (C) 2012 Avionic Design GmbH
+ * Copyright (C) 2012-2013, NVIDIA Corporation
+ *
+ * This program is free software; you can redistribute it and/or modify it
+ * under the terms and conditions of the GNU General Public License,
+ * version 2, as published by the Free Software Foundation.
+ *
+ * This program is distributed in the hope it will be useful, but WITHOUT
+ * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or
+ * FITNESS FOR A PARTICULAR PURPOSE.  See the GNU General Public License for
+ * more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program.  If not, see <http://www.gnu.org/licenses/>.
+ */
+
+#include <linux/host1x.h>
+#include <linux/of.h>
+#include <linux/slab.h>
+
+#include "dev.h"
+
+static DEFINE_MUTEX(clients_lock);
+static LIST_HEAD(clients);
+
+static DEFINE_MUTEX(drivers_lock);
+static LIST_HEAD(drivers);
+
+static DEFINE_MUTEX(devices_lock);
+static LIST_HEAD(devices);
+
+struct host1x_subdev {
+	struct host1x_client *client;
+	struct device_node *np;
+	struct list_head list;
+};
+
+/**
+ * host1x_subdev_add() - add a new subdevice with an associated device node
+ */
+static int host1x_subdev_add(struct host1x_device *device,
+			     struct device_node *np)
+{
+	struct host1x_subdev *subdev;
+
+	subdev = kzalloc(sizeof(*subdev), GFP_KERNEL);
+	if (!subdev)
+		return -ENOMEM;
+
+	INIT_LIST_HEAD(&subdev->list);
+	subdev->np = of_node_get(np);
+
+	mutex_lock(&device->subdevs_lock);
+	list_add_tail(&subdev->list, &device->subdevs);
+	mutex_unlock(&device->subdevs_lock);
+
+	return 0;
+}
+
+/**
+ * host1x_subdev_del() - remove subdevice
+ */
+static void host1x_subdev_del(struct host1x_subdev *subdev)
+{
+	list_del(&subdev->list);
+	of_node_put(subdev->np);
+	kfree(subdev);
+}
+
+/**
+ * host1x_device_parse_dt() - scan device tree and add matching subdevices
+ */
+static int host1x_device_parse_dt(struct host1x_device *device)
+{
+	struct device_node *np;
+	int err;
+
+	for_each_child_of_node(device->dev.parent->of_node, np) {
+		if (of_match_node(device->driver->subdevs, np) &&
+		    of_device_is_available(np)) {
+			err = host1x_subdev_add(device, np);
+			if (err < 0)
+				return err;
+		}
+	}
+
+	return 0;
+}
+
+static void host1x_subdev_register(struct host1x_device *device,
+				   struct host1x_subdev *subdev,
+				   struct host1x_client *client)
+{
+	int err;
+
+	/*
+	 * Move the subdevice to the list of active (registered) subdevices
+	 * and associate it with a client. At the same time, associate the
+	 * client with its parent device.
+	 */
+	mutex_lock(&device->subdevs_lock);
+	mutex_lock(&device->clients_lock);
+	list_move_tail(&client->list, &device->clients);
+	list_move_tail(&subdev->list, &device->active);
+	client->parent = &device->dev;
+	subdev->client = client;
+	mutex_unlock(&device->clients_lock);
+	mutex_unlock(&device->subdevs_lock);
+
+	/*
+	 * When all subdevices have been registered, the composite device is
+	 * ready to be probed.
+	 */
+	if (list_empty(&device->subdevs)) {
+		err = device->driver->probe(device);
+		if (err < 0)
+			dev_err(&device->dev, "probe failed: %d\n", err);
+	}
+}
+
+static void __host1x_subdev_unregister(struct host1x_device *device,
+				       struct host1x_subdev *subdev)
+{
+	struct host1x_client *client = subdev->client;
+	int err;
+
+	/*
+	 * If all subdevices have been activated, we're about to remove the
+	 * first active subdevice, so unload the driver first.
+	 */
+	if (list_empty(&device->subdevs)) {
+		err = device->driver->remove(device);
+		if (err < 0)
+			dev_err(&device->dev, "remove failed: %d\n", err);
+	}
+
+	/*
+	 * Move the subdevice back to the list of idle subdevices and remove
+	 * it from list of clients.
+	 */
+	mutex_lock(&device->clients_lock);
+	subdev->client = NULL;
+	client->parent = NULL;
+	list_move_tail(&subdev->list, &device->subdevs);
+	/*
+	 * XXX: Perhaps don't do this here, but rather explicitly remove it
+	 * when the device is about to be deleted.
+	 *
+	 * This is somewhat complicated by the fact that this function is
+	 * used to remove the subdevice when a client is unregistered but
+	 * also when the composite device is about to be removed.
+	 */
+	list_del_init(&client->list);
+	mutex_unlock(&device->clients_lock);
+}
+
+static void host1x_subdev_unregister(struct host1x_device *device,
+				     struct host1x_subdev *subdev)
+{
+	mutex_lock(&device->subdevs_lock);
+	__host1x_subdev_unregister(device, subdev);
+	mutex_unlock(&device->subdevs_lock);
+}
+
+int host1x_device_init(struct host1x_device *device)
+{
+	struct host1x_client *client;
+	int err;
+
+	mutex_lock(&device->clients_lock);
+
+	list_for_each_entry(client, &device->clients, list) {
+		if (client->ops && client->ops->init) {
+			err = client->ops->init(client);
+			if (err < 0) {
+				dev_err(&device->dev,
+					"failed to initialize %s: %d\n",
+					dev_name(client->dev), err);
+				mutex_unlock(&device->clients_lock);
+				return err;
+			}
+		}
+	}
+
+	mutex_unlock(&device->clients_lock);
+
+	return 0;
+}
+
+int host1x_device_exit(struct host1x_device *device)
+{
+	struct host1x_client *client;
+	int err;
+
+	mutex_lock(&device->clients_lock);
+
+	list_for_each_entry_reverse(client, &device->clients, list) {
+		if (client->ops && client->ops->exit) {
+			err = client->ops->exit(client);
+			if (err < 0) {
+				dev_err(&device->dev,
+					"failed to cleanup %s: %d\n",
+					dev_name(client->dev), err);
+				mutex_unlock(&device->clients_lock);
+				return err;
+			}
+		}
+	}
+
+	mutex_unlock(&device->clients_lock);
+
+	return 0;
+}
+
+static int host1x_register_client(struct host1x *host1x,
+				  struct host1x_client *client)
+{
+	struct host1x_device *device;
+	struct host1x_subdev *subdev;
+
+	mutex_lock(&host1x->devices_lock);
+
+	list_for_each_entry(device, &host1x->devices, list) {
+		list_for_each_entry(subdev, &device->subdevs, list) {
+			if (subdev->np == client->dev->of_node) {
+				host1x_subdev_register(device, subdev, client);
+				mutex_unlock(&host1x->devices_lock);
+				return 0;
+			}
+		}
+	}
+
+	mutex_unlock(&host1x->devices_lock);
+	return -ENODEV;
+}
+
+static int host1x_unregister_client(struct host1x *host1x,
+				    struct host1x_client *client)
+{
+	struct host1x_device *device, *dt;
+	struct host1x_subdev *subdev;
+
+	mutex_lock(&host1x->devices_lock);
+
+	list_for_each_entry_safe(device, dt, &host1x->devices, list) {
+		list_for_each_entry(subdev, &device->active, list) {
+			if (subdev->client == client) {
+				host1x_subdev_unregister(device, subdev);
+				mutex_unlock(&host1x->devices_lock);
+				return 0;
+			}
+		}
+	}
+
+	mutex_unlock(&host1x->devices_lock);
+	return -ENODEV;
+}
+
+struct bus_type host1x_bus_type = {
+	.name = "host1x",
+};
+
+int host1x_bus_init(void)
+{
+	return bus_register(&host1x_bus_type);
+}
+
+void host1x_bus_exit(void)
+{
+	bus_unregister(&host1x_bus_type);
+}
+
+static void host1x_device_release(struct device *dev)
+{
+	struct host1x_device *device = to_host1x_device(dev);
+
+	kfree(device);
+}
+
+static int host1x_device_add(struct host1x *host1x,
+			     struct host1x_driver *driver)
+{
+	struct host1x_client *client, *tmp;
+	struct host1x_subdev *subdev;
+	struct host1x_device *device;
+	int err;
+
+	device = kzalloc(sizeof(*device), GFP_KERNEL);
+	if (!device)
+		return -ENOMEM;
+
+	mutex_init(&device->subdevs_lock);
+	INIT_LIST_HEAD(&device->subdevs);
+	INIT_LIST_HEAD(&device->active);
+	mutex_init(&device->clients_lock);
+	INIT_LIST_HEAD(&device->clients);
+	INIT_LIST_HEAD(&device->list);
+	device->driver = driver;
+
+	device->dev.coherent_dma_mask = host1x->dev->coherent_dma_mask;
+	device->dev.dma_mask = &device->dev.coherent_dma_mask;
+	device->dev.release = host1x_device_release;
+	dev_set_name(&device->dev, driver->name);
+	device->dev.bus = &host1x_bus_type;
+	device->dev.parent = host1x->dev;
+
+	err = device_register(&device->dev);
+	if (err < 0)
+		return err;
+
+	err = host1x_device_parse_dt(device);
+	if (err < 0) {
+		device_unregister(&device->dev);
+		return err;
+	}
+
+	mutex_lock(&host1x->devices_lock);
+	list_add_tail(&device->list, &host1x->devices);
+	mutex_unlock(&host1x->devices_lock);
+
+	mutex_lock(&clients_lock);
+
+	list_for_each_entry_safe(client, tmp, &clients, list) {
+		list_for_each_entry(subdev, &device->subdevs, list) {
+			if (subdev->np == client->dev->of_node) {
+				host1x_subdev_register(device, subdev, client);
+				break;
+			}
+		}
+	}
+
+	mutex_unlock(&clients_lock);
+
+	return 0;
+}
+
+/*
+ * Removes a device by first unregistering any subdevices and then removing
+ * itself from the list of devices.
+ *
+ * This function must be called with the host1x->devices_lock held.
+ */
+static void host1x_device_del(struct host1x *host1x,
+			      struct host1x_device *device)
+{
+	struct host1x_subdev *subdev, *sd;
+	struct host1x_client *client, *cl;
+
+	mutex_lock(&device->subdevs_lock);
+
+	/* unregister subdevices */
+	list_for_each_entry_safe(subdev, sd, &device->active, list) {
+		/*
+		 * host1x_subdev_unregister() will remove the client from
+		 * any lists, so we'll need to manually add it back to the
+		 * list of idle clients.
+		 *
+		 * XXX: Alternatively, perhaps don't remove the client from
+		 * any lists in host1x_subdev_unregister() and instead do
+		 * that explicitly from host1x_unregister_client()?
+		 */
+		client = subdev->client;
+
+		__host1x_subdev_unregister(device, subdev);
+
+		/* add the client to the list of idle clients */
+		mutex_lock(&clients_lock);
+		list_add_tail(&client->list, &clients);
+		mutex_unlock(&clients_lock);
+	}
+
+	/* remove subdevices */
+	list_for_each_entry_safe(subdev, sd, &device->subdevs, list)
+		host1x_subdev_del(subdev);
+
+	mutex_unlock(&device->subdevs_lock);
+
+	/* move clients to idle list */
+	mutex_lock(&clients_lock);
+	mutex_lock(&device->clients_lock);
+
+	list_for_each_entry_safe(client, cl, &device->clients, list)
+		list_move_tail(&client->list, &clients);
+
+	mutex_unlock(&device->clients_lock);
+	mutex_unlock(&clients_lock);
+
+	/* finally remove the device */
+	list_del_init(&device->list);
+	device_unregister(&device->dev);
+}
+
+static void host1x_attach_driver(struct host1x *host1x,
+				 struct host1x_driver *driver)
+{
+	struct host1x_device *device;
+	int err;
+
+	mutex_lock(&host1x->devices_lock);
+
+	list_for_each_entry(device, &host1x->devices, list) {
+		if (device->driver == driver) {
+			mutex_unlock(&host1x->devices_lock);
+			return;
+		}
+	}
+
+	mutex_unlock(&host1x->devices_lock);
+
+	err = host1x_device_add(host1x, driver);
+	if (err < 0)
+		dev_err(host1x->dev, "failed to allocate device: %d\n", err);
+}
+
+static void host1x_detach_driver(struct host1x *host1x,
+				 struct host1x_driver *driver)
+{
+	struct host1x_device *device, *tmp;
+
+	mutex_lock(&host1x->devices_lock);
+
+	list_for_each_entry_safe(device, tmp, &host1x->devices, list)
+		if (device->driver == driver)
+			host1x_device_del(host1x, device);
+
+	mutex_unlock(&host1x->devices_lock);
+}
+
+int host1x_register(struct host1x *host1x)
+{
+	struct host1x_driver *driver;
+
+	mutex_lock(&devices_lock);
+	list_add_tail(&host1x->list, &devices);
+	mutex_unlock(&devices_lock);
+
+	mutex_lock(&drivers_lock);
+
+	list_for_each_entry(driver, &drivers, list)
+		host1x_attach_driver(host1x, driver);
+
+	mutex_unlock(&drivers_lock);
+
+	return 0;
+}
+
+int host1x_unregister(struct host1x *host1x)
+{
+	struct host1x_driver *driver;
+
+	mutex_lock(&drivers_lock);
+
+	list_for_each_entry(driver, &drivers, list)
+		host1x_detach_driver(host1x, driver);
+
+	mutex_unlock(&drivers_lock);
+
+	mutex_lock(&devices_lock);
+	list_del_init(&host1x->list);
+	mutex_unlock(&devices_lock);
+
+	return 0;
+}
+
+int host1x_driver_register(struct host1x_driver *driver)
+{
+	struct host1x *host1x;
+
+	INIT_LIST_HEAD(&driver->list);
+
+	mutex_lock(&drivers_lock);
+	list_add_tail(&driver->list, &drivers);
+	mutex_unlock(&drivers_lock);
+
+	mutex_lock(&devices_lock);
+
+	list_for_each_entry(host1x, &devices, list)
+		host1x_attach_driver(host1x, driver);
+
+	mutex_unlock(&devices_lock);
+
+	return 0;
+}
+EXPORT_SYMBOL(host1x_driver_register);
+
+void host1x_driver_unregister(struct host1x_driver *driver)
+{
+	mutex_lock(&drivers_lock);
+	list_del_init(&driver->list);
+	mutex_unlock(&drivers_lock);
+}
+EXPORT_SYMBOL(host1x_driver_unregister);
+
+int host1x_client_register(struct host1x_client *client)
+{
+	struct host1x *host1x;
+	int err;
+
+	mutex_lock(&devices_lock);
+
+	list_for_each_entry(host1x, &devices, list) {
+		err = host1x_register_client(host1x, client);
+		if (!err) {
+			mutex_unlock(&devices_lock);
+			return 0;
+		}
+	}
+
+	mutex_unlock(&devices_lock);
+
+	mutex_lock(&clients_lock);
+	list_add_tail(&client->list, &clients);
+	mutex_unlock(&clients_lock);
+
+	return 0;
+}
+EXPORT_SYMBOL(host1x_client_register);
+
+int host1x_client_unregister(struct host1x_client *client)
+{
+	struct host1x_client *c;
+	struct host1x *host1x;
+	int err;
+
+	mutex_lock(&devices_lock);
+
+	list_for_each_entry(host1x, &devices, list) {
+		err = host1x_unregister_client(host1x, client);
+		if (!err) {
+			mutex_unlock(&devices_lock);
+			return 0;
+		}
+	}
+
+	mutex_unlock(&devices_lock);
+	mutex_lock(&clients_lock);
+
+	list_for_each_entry(c, &clients, list) {
+		if (c == client) {
+			list_del_init(&c->list);
+			break;
+		}
+	}
+
+	mutex_unlock(&clients_lock);
+
+	return 0;
+}
+EXPORT_SYMBOL(host1x_client_unregister);
diff --git a/drivers/gpu/host1x/host1x_client.h b/drivers/gpu/host1x/bus.h
index 9b85f10f4a44..4099e99212c8 100644
--- a/drivers/gpu/host1x/host1x_client.h
+++ b/drivers/gpu/host1x/bus.h
@@ -1,5 +1,6 @@
 /*
- * Copyright (c) 2013, NVIDIA Corporation.
+ * Copyright (C) 2012 Avionic Design GmbH
+ * Copyright (C) 2012-2013, NVIDIA Corporation
  *
  * This program is free software; you can redistribute it and/or modify it
  * under the terms and conditions of the GNU General Public License,
@@ -14,22 +15,15 @@
  * along with this program.  If not, see <http://www.gnu.org/licenses/>.
  */
 
-#ifndef HOST1X_CLIENT_H
-#define HOST1X_CLIENT_H
+#ifndef HOST1X_BUS_H
+#define HOST1X_BUS_H
 
-struct device;
-struct platform_device;
+struct host1x;
 
-#ifdef CONFIG_DRM_TEGRA
-int host1x_drm_alloc(struct platform_device *pdev);
-#else
-static inline int host1x_drm_alloc(struct platform_device *pdev)
-{
-	return 0;
-}
-#endif
+int host1x_bus_init(void);
+void host1x_bus_exit(void);
 
-void host1x_set_drm_data(struct device *dev, void *data);
-void *host1x_get_drm_data(struct device *dev);
+int host1x_register(struct host1x *host1x);
+int host1x_unregister(struct host1x *host1x);
 
 #endif
diff --git a/drivers/gpu/host1x/cdma.c b/drivers/gpu/host1x/cdma.c
index de72172d3b5f..3995255b16c7 100644
--- a/drivers/gpu/host1x/cdma.c
+++ b/drivers/gpu/host1x/cdma.c
@@ -20,6 +20,7 @@
 #include <asm/cacheflush.h>
 #include <linux/device.h>
 #include <linux/dma-mapping.h>
+#include <linux/host1x.h>
 #include <linux/interrupt.h>
 #include <linux/kernel.h>
 #include <linux/kfifo.h>
@@ -30,7 +31,6 @@
 #include "channel.h"
 #include "dev.h"
 #include "debug.h"
-#include "host1x_bo.h"
 #include "job.h"
 
 /*
diff --git a/drivers/gpu/host1x/channel.h b/drivers/gpu/host1x/channel.h
index 48723b8eea42..df767cf90d51 100644
--- a/drivers/gpu/host1x/channel.h
+++ b/drivers/gpu/host1x/channel.h
@@ -40,12 +40,6 @@ struct host1x_channel {
 /* channel list operations */
 int host1x_channel_list_init(struct host1x *host);
 
-struct host1x_channel *host1x_channel_request(struct device *dev);
-void host1x_channel_free(struct host1x_channel *channel);
-struct host1x_channel *host1x_channel_get(struct host1x_channel *channel);
-void host1x_channel_put(struct host1x_channel *channel);
-int host1x_job_submit(struct host1x_job *job);
-
 #define host1x_for_each_channel(host, channel)				\
 	list_for_each_entry(channel, &host->chlist.list, list)
 
diff --git a/drivers/gpu/host1x/dev.c b/drivers/gpu/host1x/dev.c
index 471630299878..80da003d63de 100644
--- a/drivers/gpu/host1x/dev.c
+++ b/drivers/gpu/host1x/dev.c
@@ -27,24 +27,13 @@
 #define CREATE_TRACE_POINTS
 #include <trace/events/host1x.h>
 
+#include "bus.h"
 #include "dev.h"
 #include "intr.h"
 #include "channel.h"
 #include "debug.h"
 #include "hw/host1x01.h"
-#include "host1x_client.h"
-
-void host1x_set_drm_data(struct device *dev, void *data)
-{
-	struct host1x *host1x = dev_get_drvdata(dev);
-	host1x->drm_data = data;
-}
-
-void *host1x_get_drm_data(struct device *dev)
-{
-	struct host1x *host1x = dev_get_drvdata(dev);
-	return host1x ? host1x->drm_data : NULL;
-}
+#include "hw/host1x02.h"
 
 void host1x_sync_writel(struct host1x *host1x, u32 v, u32 r)
 {
@@ -79,7 +68,17 @@ static const struct host1x_info host1x01_info = {
 	.sync_offset	= 0x3000,
 };
 
+static const struct host1x_info host1x02_info = {
+	.nb_channels = 9,
+	.nb_pts = 32,
+	.nb_mlocks = 16,
+	.nb_bases = 12,
+	.init = host1x02_init,
+	.sync_offset = 0x3000,
+};
+
 static struct of_device_id host1x_of_match[] = {
+	{ .compatible = "nvidia,tegra114-host1x", .data = &host1x02_info, },
 	{ .compatible = "nvidia,tegra30-host1x", .data = &host1x01_info, },
 	{ .compatible = "nvidia,tegra20-host1x", .data = &host1x01_info, },
 	{ },
@@ -114,6 +113,9 @@ static int host1x_probe(struct platform_device *pdev)
 	if (!host)
 		return -ENOMEM;
 
+	mutex_init(&host->devices_lock);
+	INIT_LIST_HEAD(&host->devices);
+	INIT_LIST_HEAD(&host->list);
 	host->dev = &pdev->dev;
 	host->info = id->data;
 
@@ -152,7 +154,7 @@ static int host1x_probe(struct platform_device *pdev)
 	err = host1x_syncpt_init(host);
 	if (err) {
 		dev_err(&pdev->dev, "failed to initialize syncpts\n");
-		return err;
+		goto fail_unprepare_disable;
 	}
 
 	err = host1x_intr_init(host, syncpt_irq);
@@ -163,19 +165,26 @@ static int host1x_probe(struct platform_device *pdev)
 
 	host1x_debug_init(host);
 
-	host1x_drm_alloc(pdev);
+	err = host1x_register(host);
+	if (err < 0)
+		goto fail_deinit_intr;
 
 	return 0;
 
+fail_deinit_intr:
+	host1x_intr_deinit(host);
 fail_deinit_syncpt:
 	host1x_syncpt_deinit(host);
+fail_unprepare_disable:
+	clk_disable_unprepare(host->clk);
 	return err;
 }
 
-static int __exit host1x_remove(struct platform_device *pdev)
+static int host1x_remove(struct platform_device *pdev)
 {
 	struct host1x *host = platform_get_drvdata(pdev);
 
+	host1x_unregister(host);
 	host1x_intr_deinit(host);
 	host1x_syncpt_deinit(host);
 	clk_disable_unprepare(host->clk);
@@ -184,59 +193,36 @@ static int __exit host1x_remove(struct platform_device *pdev)
 }
 
 static struct platform_driver tegra_host1x_driver = {
-	.probe = host1x_probe,
-	.remove = __exit_p(host1x_remove),
 	.driver = {
-		.owner = THIS_MODULE,
 		.name = "tegra-host1x",
 		.of_match_table = host1x_of_match,
 	},
+	.probe = host1x_probe,
+	.remove = host1x_remove,
 };
 
 static int __init tegra_host1x_init(void)
 {
 	int err;
 
-	err = platform_driver_register(&tegra_host1x_driver);
+	err = host1x_bus_init();
 	if (err < 0)
 		return err;
 
-#ifdef CONFIG_DRM_TEGRA
-	err = platform_driver_register(&tegra_dc_driver);
-	if (err < 0)
-		goto unregister_host1x;
-
-	err = platform_driver_register(&tegra_hdmi_driver);
-	if (err < 0)
-		goto unregister_dc;
-
-	err = platform_driver_register(&tegra_gr2d_driver);
-	if (err < 0)
-		goto unregister_hdmi;
-#endif
+	err = platform_driver_register(&tegra_host1x_driver);
+	if (err < 0) {
+		host1x_bus_exit();
+		return err;
+	}
 
 	return 0;
-
-#ifdef CONFIG_DRM_TEGRA
-unregister_hdmi:
-	platform_driver_unregister(&tegra_hdmi_driver);
-unregister_dc:
-	platform_driver_unregister(&tegra_dc_driver);
-unregister_host1x:
-	platform_driver_unregister(&tegra_host1x_driver);
-	return err;
-#endif
 }
 module_init(tegra_host1x_init);
 
 static void __exit tegra_host1x_exit(void)
 {
-#ifdef CONFIG_DRM_TEGRA
-	platform_driver_unregister(&tegra_gr2d_driver);
-	platform_driver_unregister(&tegra_hdmi_driver);
-	platform_driver_unregister(&tegra_dc_driver);
-#endif
 	platform_driver_unregister(&tegra_host1x_driver);
+	host1x_bus_exit();
 }
 module_exit(tegra_host1x_exit);
 
diff --git a/drivers/gpu/host1x/dev.h b/drivers/gpu/host1x/dev.h
index bed90a8131be..a61a976e7a42 100644
--- a/drivers/gpu/host1x/dev.h
+++ b/drivers/gpu/host1x/dev.h
@@ -27,6 +27,7 @@
 #include "job.h"
 
 struct host1x_syncpt;
+struct host1x_syncpt_base;
 struct host1x_channel;
 struct host1x_cdma;
 struct host1x_job;
@@ -102,6 +103,7 @@ struct host1x {
 
 	void __iomem *regs;
 	struct host1x_syncpt *syncpt;
+	struct host1x_syncpt_base *bases;
 	struct device *dev;
 	struct clk *clk;
 
@@ -125,7 +127,10 @@ struct host1x {
 
 	struct dentry *debugfs;
 
-	void *drm_data;
+	struct mutex devices_lock;
+	struct list_head devices;
+
+	struct list_head list;
 };
 
 void host1x_sync_writel(struct host1x *host1x, u32 r, u32 v);
@@ -301,8 +306,4 @@ static inline void host1x_hw_show_mlocks(struct host1x *host, struct output *o)
 	host->debug_op->show_mlocks(host, o);
 }
 
-extern struct platform_driver tegra_dc_driver;
-extern struct platform_driver tegra_hdmi_driver;
-extern struct platform_driver tegra_gr2d_driver;
-
 #endif
diff --git a/drivers/gpu/host1x/drm/drm.c b/drivers/gpu/host1x/drm/drm.c
deleted file mode 100644
index 8c61ceeaa12d..000000000000
--- a/drivers/gpu/host1x/drm/drm.c
+++ /dev/null
@@ -1,647 +0,0 @@
-/*
- * Copyright (C) 2012 Avionic Design GmbH
- * Copyright (C) 2012-2013 NVIDIA CORPORATION.  All rights reserved.
- *
- * This program is free software; you can redistribute it and/or modify
- * it under the terms of the GNU General Public License version 2 as
- * published by the Free Software Foundation.
- */
-
-#include <linux/module.h>
-#include <linux/of_address.h>
-#include <linux/of_platform.h>
-
-#include <linux/dma-mapping.h>
-#include <asm/dma-iommu.h>
-
-#include <drm/drm.h>
-#include <drm/drmP.h>
-
-#include "host1x_client.h"
-#include "dev.h"
-#include "drm.h"
-#include "gem.h"
-#include "syncpt.h"
-
-#define DRIVER_NAME "tegra"
-#define DRIVER_DESC "NVIDIA Tegra graphics"
-#define DRIVER_DATE "20120330"
-#define DRIVER_MAJOR 0
-#define DRIVER_MINOR 0
-#define DRIVER_PATCHLEVEL 0
-
-struct host1x_drm_client {
-	struct host1x_client *client;
-	struct device_node *np;
-	struct list_head list;
-};
-
-static int host1x_add_drm_client(struct host1x_drm *host1x,
-				 struct device_node *np)
-{
-	struct host1x_drm_client *client;
-
-	client = kzalloc(sizeof(*client), GFP_KERNEL);
-	if (!client)
-		return -ENOMEM;
-
-	INIT_LIST_HEAD(&client->list);
-	client->np = of_node_get(np);
-
-	list_add_tail(&client->list, &host1x->drm_clients);
-
-	return 0;
-}
-
-static int host1x_activate_drm_client(struct host1x_drm *host1x,
-				      struct host1x_drm_client *drm,
-				      struct host1x_client *client)
-{
-	mutex_lock(&host1x->drm_clients_lock);
-	list_del_init(&drm->list);
-	list_add_tail(&drm->list, &host1x->drm_active);
-	drm->client = client;
-	mutex_unlock(&host1x->drm_clients_lock);
-
-	return 0;
-}
-
-static int host1x_remove_drm_client(struct host1x_drm *host1x,
-				    struct host1x_drm_client *client)
-{
-	mutex_lock(&host1x->drm_clients_lock);
-	list_del_init(&client->list);
-	mutex_unlock(&host1x->drm_clients_lock);
-
-	of_node_put(client->np);
-	kfree(client);
-
-	return 0;
-}
-
-static int host1x_parse_dt(struct host1x_drm *host1x)
-{
-	static const char * const compat[] = {
-		"nvidia,tegra20-dc",
-		"nvidia,tegra20-hdmi",
-		"nvidia,tegra20-gr2d",
-		"nvidia,tegra30-dc",
-		"nvidia,tegra30-hdmi",
-		"nvidia,tegra30-gr2d",
-	};
-	unsigned int i;
-	int err;
-
-	for (i = 0; i < ARRAY_SIZE(compat); i++) {
-		struct device_node *np;
-
-		for_each_child_of_node(host1x->dev->of_node, np) {
-			if (of_device_is_compatible(np, compat[i]) &&
-			    of_device_is_available(np)) {
-				err = host1x_add_drm_client(host1x, np);
-				if (err < 0)
-					return err;
-			}
-		}
-	}
-
-	return 0;
-}
-
-int host1x_drm_alloc(struct platform_device *pdev)
-{
-	struct host1x_drm *host1x;
-	int err;
-
-	host1x = devm_kzalloc(&pdev->dev, sizeof(*host1x), GFP_KERNEL);
-	if (!host1x)
-		return -ENOMEM;
-
-	mutex_init(&host1x->drm_clients_lock);
-	INIT_LIST_HEAD(&host1x->drm_clients);
-	INIT_LIST_HEAD(&host1x->drm_active);
-	mutex_init(&host1x->clients_lock);
-	INIT_LIST_HEAD(&host1x->clients);
-	host1x->dev = &pdev->dev;
-
-	err = host1x_parse_dt(host1x);
-	if (err < 0) {
-		dev_err(&pdev->dev, "failed to parse DT: %d\n", err);
-		return err;
-	}
-
-	host1x_set_drm_data(&pdev->dev, host1x);
-
-	return 0;
-}
-
-int host1x_drm_init(struct host1x_drm *host1x, struct drm_device *drm)
-{
-	struct host1x_client *client;
-
-	mutex_lock(&host1x->clients_lock);
-
-	list_for_each_entry(client, &host1x->clients, list) {
-		if (client->ops && client->ops->drm_init) {
-			int err = client->ops->drm_init(client, drm);
-			if (err < 0) {
-				dev_err(host1x->dev,
-					"DRM setup failed for %s: %d\n",
-					dev_name(client->dev), err);
-				mutex_unlock(&host1x->clients_lock);
-				return err;
-			}
-		}
-	}
-
-	mutex_unlock(&host1x->clients_lock);
-
-	return 0;
-}
-
-int host1x_drm_exit(struct host1x_drm *host1x)
-{
-	struct platform_device *pdev = to_platform_device(host1x->dev);
-	struct host1x_client *client;
-
-	if (!host1x->drm)
-		return 0;
-
-	mutex_lock(&host1x->clients_lock);
-
-	list_for_each_entry_reverse(client, &host1x->clients, list) {
-		if (client->ops && client->ops->drm_exit) {
-			int err = client->ops->drm_exit(client);
-			if (err < 0) {
-				dev_err(host1x->dev,
-					"DRM cleanup failed for %s: %d\n",
-					dev_name(client->dev), err);
-				mutex_unlock(&host1x->clients_lock);
-				return err;
-			}
-		}
-	}
-
-	mutex_unlock(&host1x->clients_lock);
-
-	drm_platform_exit(&tegra_drm_driver, pdev);
-	host1x->drm = NULL;
-
-	return 0;
-}
-
-int host1x_register_client(struct host1x_drm *host1x,
-			   struct host1x_client *client)
-{
-	struct host1x_drm_client *drm, *tmp;
-	int err;
-
-	mutex_lock(&host1x->clients_lock);
-	list_add_tail(&client->list, &host1x->clients);
-	mutex_unlock(&host1x->clients_lock);
-
-	list_for_each_entry_safe(drm, tmp, &host1x->drm_clients, list)
-		if (drm->np == client->dev->of_node)
-			host1x_activate_drm_client(host1x, drm, client);
-
-	if (list_empty(&host1x->drm_clients)) {
-		struct platform_device *pdev = to_platform_device(host1x->dev);
-
-		err = drm_platform_init(&tegra_drm_driver, pdev);
-		if (err < 0) {
-			dev_err(host1x->dev, "drm_platform_init(): %d\n", err);
-			return err;
-		}
-	}
-
-	return 0;
-}
-
-int host1x_unregister_client(struct host1x_drm *host1x,
-			     struct host1x_client *client)
-{
-	struct host1x_drm_client *drm, *tmp;
-	int err;
-
-	list_for_each_entry_safe(drm, tmp, &host1x->drm_active, list) {
-		if (drm->client == client) {
-			err = host1x_drm_exit(host1x);
-			if (err < 0) {
-				dev_err(host1x->dev, "host1x_drm_exit(): %d\n",
-					err);
-				return err;
-			}
-
-			host1x_remove_drm_client(host1x, drm);
-			break;
-		}
-	}
-
-	mutex_lock(&host1x->clients_lock);
-	list_del_init(&client->list);
-	mutex_unlock(&host1x->clients_lock);
-
-	return 0;
-}
-
-static int tegra_drm_load(struct drm_device *drm, unsigned long flags)
-{
-	struct host1x_drm *host1x;
-	int err;
-
-	host1x = host1x_get_drm_data(drm->dev);
-	drm->dev_private = host1x;
-	host1x->drm = drm;
-
-	drm_mode_config_init(drm);
-
-	err = host1x_drm_init(host1x, drm);
-	if (err < 0)
-		return err;
-
-	/*
-	 * We don't use the drm_irq_install() helpers provided by the DRM
-	 * core, so we need to set this manually in order to allow the
-	 * DRM_IOCTL_WAIT_VBLANK to operate correctly.
-	 */
-	drm->irq_enabled = 1;
-
-	err = drm_vblank_init(drm, drm->mode_config.num_crtc);
-	if (err < 0)
-		return err;
-
-	err = tegra_drm_fb_init(drm);
-	if (err < 0)
-		return err;
-
-	drm_kms_helper_poll_init(drm);
-
-	return 0;
-}
-
-static int tegra_drm_unload(struct drm_device *drm)
-{
-	drm_kms_helper_poll_fini(drm);
-	tegra_drm_fb_exit(drm);
-
-	drm_mode_config_cleanup(drm);
-
-	return 0;
-}
-
-static int tegra_drm_open(struct drm_device *drm, struct drm_file *filp)
-{
-	struct host1x_drm_file *fpriv;
-
-	fpriv = kzalloc(sizeof(*fpriv), GFP_KERNEL);
-	if (!fpriv)
-		return -ENOMEM;
-
-	INIT_LIST_HEAD(&fpriv->contexts);
-	filp->driver_priv = fpriv;
-
-	return 0;
-}
-
-static void host1x_drm_context_free(struct host1x_drm_context *context)
-{
-	context->client->ops->close_channel(context);
-	kfree(context);
-}
-
-static void tegra_drm_lastclose(struct drm_device *drm)
-{
-	struct host1x_drm *host1x = drm->dev_private;
-
-	tegra_fbdev_restore_mode(host1x->fbdev);
-}
-
-#ifdef CONFIG_DRM_TEGRA_STAGING
-static bool host1x_drm_file_owns_context(struct host1x_drm_file *file,
-					 struct host1x_drm_context *context)
-{
-	struct host1x_drm_context *ctx;
-
-	list_for_each_entry(ctx, &file->contexts, list)
-		if (ctx == context)
-			return true;
-
-	return false;
-}
-
-static int tegra_gem_create(struct drm_device *drm, void *data,
-			    struct drm_file *file)
-{
-	struct drm_tegra_gem_create *args = data;
-	struct tegra_bo *bo;
-
-	bo = tegra_bo_create_with_handle(file, drm, args->size,
-					 &args->handle);
-	if (IS_ERR(bo))
-		return PTR_ERR(bo);
-
-	return 0;
-}
-
-static int tegra_gem_mmap(struct drm_device *drm, void *data,
-			  struct drm_file *file)
-{
-	struct drm_tegra_gem_mmap *args = data;
-	struct drm_gem_object *gem;
-	struct tegra_bo *bo;
-
-	gem = drm_gem_object_lookup(drm, file, args->handle);
-	if (!gem)
-		return -EINVAL;
-
-	bo = to_tegra_bo(gem);
-
-	args->offset = drm_vma_node_offset_addr(&bo->gem.vma_node);
-
-	drm_gem_object_unreference(gem);
-
-	return 0;
-}
-
-static int tegra_syncpt_read(struct drm_device *drm, void *data,
-			     struct drm_file *file)
-{
-	struct drm_tegra_syncpt_read *args = data;
-	struct host1x *host = dev_get_drvdata(drm->dev);
-	struct host1x_syncpt *sp = host1x_syncpt_get(host, args->id);
-
-	if (!sp)
-		return -EINVAL;
-
-	args->value = host1x_syncpt_read_min(sp);
-	return 0;
-}
-
-static int tegra_syncpt_incr(struct drm_device *drm, void *data,
-			     struct drm_file *file)
-{
-	struct drm_tegra_syncpt_incr *args = data;
-	struct host1x *host = dev_get_drvdata(drm->dev);
-	struct host1x_syncpt *sp = host1x_syncpt_get(host, args->id);
-
-	if (!sp)
-		return -EINVAL;
-
-	return host1x_syncpt_incr(sp);
-}
-
-static int tegra_syncpt_wait(struct drm_device *drm, void *data,
-			     struct drm_file *file)
-{
-	struct drm_tegra_syncpt_wait *args = data;
-	struct host1x *host = dev_get_drvdata(drm->dev);
-	struct host1x_syncpt *sp = host1x_syncpt_get(host, args->id);
-
-	if (!sp)
-		return -EINVAL;
-
-	return host1x_syncpt_wait(sp, args->thresh, args->timeout,
-				  &args->value);
-}
-
-static int tegra_open_channel(struct drm_device *drm, void *data,
-			      struct drm_file *file)
-{
-	struct drm_tegra_open_channel *args = data;
-	struct host1x_client *client;
-	struct host1x_drm_context *context;
-	struct host1x_drm_file *fpriv = file->driver_priv;
-	struct host1x_drm *host1x = drm->dev_private;
-	int err = -ENODEV;
-
-	context = kzalloc(sizeof(*context), GFP_KERNEL);
-	if (!context)
-		return -ENOMEM;
-
-	list_for_each_entry(client, &host1x->clients, list)
-		if (client->class == args->client) {
-			err = client->ops->open_channel(client, context);
-			if (err)
-				break;
-
-			context->client = client;
-			list_add(&context->list, &fpriv->contexts);
-			args->context = (uintptr_t)context;
-			return 0;
-		}
-
-	kfree(context);
-	return err;
-}
-
-static int tegra_close_channel(struct drm_device *drm, void *data,
-			       struct drm_file *file)
-{
-	struct drm_tegra_close_channel *args = data;
-	struct host1x_drm_file *fpriv = file->driver_priv;
-	struct host1x_drm_context *context =
-		(struct host1x_drm_context *)(uintptr_t)args->context;
-
-	if (!host1x_drm_file_owns_context(fpriv, context))
-		return -EINVAL;
-
-	list_del(&context->list);
-	host1x_drm_context_free(context);
-
-	return 0;
-}
-
-static int tegra_get_syncpt(struct drm_device *drm, void *data,
-			    struct drm_file *file)
-{
-	struct drm_tegra_get_syncpt *args = data;
-	struct host1x_drm_file *fpriv = file->driver_priv;
-	struct host1x_drm_context *context =
-		(struct host1x_drm_context *)(uintptr_t)args->context;
-	struct host1x_syncpt *syncpt;
-
-	if (!host1x_drm_file_owns_context(fpriv, context))
-		return -ENODEV;
-
-	if (args->index >= context->client->num_syncpts)
-		return -EINVAL;
-
-	syncpt = context->client->syncpts[args->index];
-	args->id = host1x_syncpt_id(syncpt);
-
-	return 0;
-}
-
-static int tegra_submit(struct drm_device *drm, void *data,
-			struct drm_file *file)
-{
-	struct drm_tegra_submit *args = data;
-	struct host1x_drm_file *fpriv = file->driver_priv;
-	struct host1x_drm_context *context =
-		(struct host1x_drm_context *)(uintptr_t)args->context;
-
-	if (!host1x_drm_file_owns_context(fpriv, context))
-		return -ENODEV;
-
-	return context->client->ops->submit(context, args, drm, file);
-}
-#endif
-
-static const struct drm_ioctl_desc tegra_drm_ioctls[] = {
-#ifdef CONFIG_DRM_TEGRA_STAGING
-	DRM_IOCTL_DEF_DRV(TEGRA_GEM_CREATE, tegra_gem_create, DRM_UNLOCKED | DRM_AUTH),
-	DRM_IOCTL_DEF_DRV(TEGRA_GEM_MMAP, tegra_gem_mmap, DRM_UNLOCKED),
-	DRM_IOCTL_DEF_DRV(TEGRA_SYNCPT_READ, tegra_syncpt_read, DRM_UNLOCKED),
-	DRM_IOCTL_DEF_DRV(TEGRA_SYNCPT_INCR, tegra_syncpt_incr, DRM_UNLOCKED),
-	DRM_IOCTL_DEF_DRV(TEGRA_SYNCPT_WAIT, tegra_syncpt_wait, DRM_UNLOCKED),
-	DRM_IOCTL_DEF_DRV(TEGRA_OPEN_CHANNEL, tegra_open_channel, DRM_UNLOCKED),
-	DRM_IOCTL_DEF_DRV(TEGRA_CLOSE_CHANNEL, tegra_close_channel, DRM_UNLOCKED),
-	DRM_IOCTL_DEF_DRV(TEGRA_GET_SYNCPT, tegra_get_syncpt, DRM_UNLOCKED),
-	DRM_IOCTL_DEF_DRV(TEGRA_SUBMIT, tegra_submit, DRM_UNLOCKED),
-#endif
-};
-
-static const struct file_operations tegra_drm_fops = {
-	.owner = THIS_MODULE,
-	.open = drm_open,
-	.release = drm_release,
-	.unlocked_ioctl = drm_ioctl,
-	.mmap = tegra_drm_mmap,
-	.poll = drm_poll,
-	.read = drm_read,
-#ifdef CONFIG_COMPAT
-	.compat_ioctl = drm_compat_ioctl,
-#endif
-	.llseek = noop_llseek,
-};
-
-static struct drm_crtc *tegra_crtc_from_pipe(struct drm_device *drm, int pipe)
-{
-	struct drm_crtc *crtc;
-
-	list_for_each_entry(crtc, &drm->mode_config.crtc_list, head) {
-		struct tegra_dc *dc = to_tegra_dc(crtc);
-
-		if (dc->pipe == pipe)
-			return crtc;
-	}
-
-	return NULL;
-}
-
-static u32 tegra_drm_get_vblank_counter(struct drm_device *dev, int crtc)
-{
-	/* TODO: implement real hardware counter using syncpoints */
-	return drm_vblank_count(dev, crtc);
-}
-
-static int tegra_drm_enable_vblank(struct drm_device *drm, int pipe)
-{
-	struct drm_crtc *crtc = tegra_crtc_from_pipe(drm, pipe);
-	struct tegra_dc *dc = to_tegra_dc(crtc);
-
-	if (!crtc)
-		return -ENODEV;
-
-	tegra_dc_enable_vblank(dc);
-
-	return 0;
-}
-
-static void tegra_drm_disable_vblank(struct drm_device *drm, int pipe)
-{
-	struct drm_crtc *crtc = tegra_crtc_from_pipe(drm, pipe);
-	struct tegra_dc *dc = to_tegra_dc(crtc);
-
-	if (crtc)
-		tegra_dc_disable_vblank(dc);
-}
-
-static void tegra_drm_preclose(struct drm_device *drm, struct drm_file *file)
-{
-	struct host1x_drm_file *fpriv = file->driver_priv;
-	struct host1x_drm_context *context, *tmp;
-	struct drm_crtc *crtc;
-
-	list_for_each_entry(crtc, &drm->mode_config.crtc_list, head)
-		tegra_dc_cancel_page_flip(crtc, file);
-
-	list_for_each_entry_safe(context, tmp, &fpriv->contexts, list)
-		host1x_drm_context_free(context);
-
-	kfree(fpriv);
-}
-
-#ifdef CONFIG_DEBUG_FS
-static int tegra_debugfs_framebuffers(struct seq_file *s, void *data)
-{
-	struct drm_info_node *node = (struct drm_info_node *)s->private;
-	struct drm_device *drm = node->minor->dev;
-	struct drm_framebuffer *fb;
-
-	mutex_lock(&drm->mode_config.fb_lock);
-
-	list_for_each_entry(fb, &drm->mode_config.fb_list, head) {
-		seq_printf(s, "%3d: user size: %d x %d, depth %d, %d bpp, refcount %d\n",
-			   fb->base.id, fb->width, fb->height, fb->depth,
-			   fb->bits_per_pixel,
-			   atomic_read(&fb->refcount.refcount));
-	}
-
-	mutex_unlock(&drm->mode_config.fb_lock);
-
-	return 0;
-}
-
-static struct drm_info_list tegra_debugfs_list[] = {
-	{ "framebuffers", tegra_debugfs_framebuffers, 0 },
-};
-
-static int tegra_debugfs_init(struct drm_minor *minor)
-{
-	return drm_debugfs_create_files(tegra_debugfs_list,
-					ARRAY_SIZE(tegra_debugfs_list),
-					minor->debugfs_root, minor);
-}
-
-static void tegra_debugfs_cleanup(struct drm_minor *minor)
-{
-	drm_debugfs_remove_files(tegra_debugfs_list,
-				 ARRAY_SIZE(tegra_debugfs_list), minor);
-}
-#endif
-
-struct drm_driver tegra_drm_driver = {
-	.driver_features = DRIVER_MODESET | DRIVER_GEM,
-	.load = tegra_drm_load,
-	.unload = tegra_drm_unload,
-	.open = tegra_drm_open,
-	.preclose = tegra_drm_preclose,
-	.lastclose = tegra_drm_lastclose,
-
-	.get_vblank_counter = tegra_drm_get_vblank_counter,
-	.enable_vblank = tegra_drm_enable_vblank,
-	.disable_vblank = tegra_drm_disable_vblank,
-
-#if defined(CONFIG_DEBUG_FS)
-	.debugfs_init = tegra_debugfs_init,
-	.debugfs_cleanup = tegra_debugfs_cleanup,
-#endif
-
-	.gem_free_object = tegra_bo_free_object,
-	.gem_vm_ops = &tegra_bo_vm_ops,
-	.dumb_create = tegra_bo_dumb_create,
-	.dumb_map_offset = tegra_bo_dumb_map_offset,
-	.dumb_destroy = drm_gem_dumb_destroy,
-
-	.ioctls = tegra_drm_ioctls,
-	.num_ioctls = ARRAY_SIZE(tegra_drm_ioctls),
-	.fops = &tegra_drm_fops,
-
-	.name = DRIVER_NAME,
-	.desc = DRIVER_DESC,
-	.date = DRIVER_DATE,
-	.major = DRIVER_MAJOR,
-	.minor = DRIVER_MINOR,
-	.patchlevel = DRIVER_PATCHLEVEL,
-};
diff --git a/drivers/gpu/host1x/drm/gr2d.c b/drivers/gpu/host1x/drm/gr2d.c
deleted file mode 100644
index 27ffcf15a4b4..000000000000
--- a/drivers/gpu/host1x/drm/gr2d.c
+++ /dev/null
@@ -1,343 +0,0 @@
-/*
- * drivers/video/tegra/host/gr2d/gr2d.c
- *
- * Tegra Graphics 2D
- *
- * Copyright (c) 2012-2013, NVIDIA Corporation.
- *
- * This program is free software; you can redistribute it and/or modify it
- * under the terms and conditions of the GNU General Public License,
- * version 2, as published by the Free Software Foundation.
- *
- * This program is distributed in the hope it will be useful, but WITHOUT
- * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or
- * FITNESS FOR A PARTICULAR PURPOSE.  See the GNU General Public License for
- * more details.
- *
- * You should have received a copy of the GNU General Public License
- * along with this program.  If not, see <http://www.gnu.org/licenses/>.
- */
-
-#include <linux/export.h>
-#include <linux/of.h>
-#include <linux/of_device.h>
-#include <linux/clk.h>
-
-#include "channel.h"
-#include "drm.h"
-#include "gem.h"
-#include "job.h"
-#include "host1x.h"
-#include "host1x_bo.h"
-#include "host1x_client.h"
-#include "syncpt.h"
-
-struct gr2d {
-	struct host1x_client client;
-	struct clk *clk;
-	struct host1x_channel *channel;
-	unsigned long *addr_regs;
-};
-
-static inline struct gr2d *to_gr2d(struct host1x_client *client)
-{
-	return container_of(client, struct gr2d, client);
-}
-
-static int gr2d_is_addr_reg(struct device *dev, u32 class, u32 reg);
-
-static int gr2d_client_init(struct host1x_client *client,
-			    struct drm_device *drm)
-{
-	return 0;
-}
-
-static int gr2d_client_exit(struct host1x_client *client)
-{
-	return 0;
-}
-
-static int gr2d_open_channel(struct host1x_client *client,
-			     struct host1x_drm_context *context)
-{
-	struct gr2d *gr2d = to_gr2d(client);
-
-	context->channel = host1x_channel_get(gr2d->channel);
-
-	if (!context->channel)
-		return -ENOMEM;
-
-	return 0;
-}
-
-static void gr2d_close_channel(struct host1x_drm_context *context)
-{
-	host1x_channel_put(context->channel);
-}
-
-static struct host1x_bo *host1x_bo_lookup(struct drm_device *drm,
-					  struct drm_file *file,
-					  u32 handle)
-{
-	struct drm_gem_object *gem;
-	struct tegra_bo *bo;
-
-	gem = drm_gem_object_lookup(drm, file, handle);
-	if (!gem)
-		return NULL;
-
-	mutex_lock(&drm->struct_mutex);
-	drm_gem_object_unreference(gem);
-	mutex_unlock(&drm->struct_mutex);
-
-	bo = to_tegra_bo(gem);
-	return &bo->base;
-}
-
-static int gr2d_submit(struct host1x_drm_context *context,
-		       struct drm_tegra_submit *args, struct drm_device *drm,
-		       struct drm_file *file)
-{
-	struct host1x_job *job;
-	unsigned int num_cmdbufs = args->num_cmdbufs;
-	unsigned int num_relocs = args->num_relocs;
-	unsigned int num_waitchks = args->num_waitchks;
-	struct drm_tegra_cmdbuf __user *cmdbufs =
-		(void * __user)(uintptr_t)args->cmdbufs;
-	struct drm_tegra_reloc __user *relocs =
-		(void * __user)(uintptr_t)args->relocs;
-	struct drm_tegra_waitchk __user *waitchks =
-		(void * __user)(uintptr_t)args->waitchks;
-	struct drm_tegra_syncpt syncpt;
-	int err;
-
-	/* We don't yet support other than one syncpt_incr struct per submit */
-	if (args->num_syncpts != 1)
-		return -EINVAL;
-
-	job = host1x_job_alloc(context->channel, args->num_cmdbufs,
-			       args->num_relocs, args->num_waitchks);
-	if (!job)
-		return -ENOMEM;
-
-	job->num_relocs = args->num_relocs;
-	job->num_waitchk = args->num_waitchks;
-	job->client = (u32)args->context;
-	job->class = context->client->class;
-	job->serialize = true;
-
-	while (num_cmdbufs) {
-		struct drm_tegra_cmdbuf cmdbuf;
-		struct host1x_bo *bo;
-
-		err = copy_from_user(&cmdbuf, cmdbufs, sizeof(cmdbuf));
-		if (err)
-			goto fail;
-
-		bo = host1x_bo_lookup(drm, file, cmdbuf.handle);
-		if (!bo) {
-			err = -ENOENT;
-			goto fail;
-		}
-
-		host1x_job_add_gather(job, bo, cmdbuf.words, cmdbuf.offset);
-		num_cmdbufs--;
-		cmdbufs++;
-	}
-
-	err = copy_from_user(job->relocarray, relocs,
-			     sizeof(*relocs) * num_relocs);
-	if (err)
-		goto fail;
-
-	while (num_relocs--) {
-		struct host1x_reloc *reloc = &job->relocarray[num_relocs];
-		struct host1x_bo *cmdbuf, *target;
-
-		cmdbuf = host1x_bo_lookup(drm, file, (u32)reloc->cmdbuf);
-		target = host1x_bo_lookup(drm, file, (u32)reloc->target);
-
-		reloc->cmdbuf = cmdbuf;
-		reloc->target = target;
-
-		if (!reloc->target || !reloc->cmdbuf) {
-			err = -ENOENT;
-			goto fail;
-		}
-	}
-
-	err = copy_from_user(job->waitchk, waitchks,
-			     sizeof(*waitchks) * num_waitchks);
-	if (err)
-		goto fail;
-
-	err = copy_from_user(&syncpt, (void * __user)(uintptr_t)args->syncpts,
-			     sizeof(syncpt));
-	if (err)
-		goto fail;
-
-	job->syncpt_id = syncpt.id;
-	job->syncpt_incrs = syncpt.incrs;
-	job->timeout = 10000;
-	job->is_addr_reg = gr2d_is_addr_reg;
-
-	if (args->timeout && args->timeout < 10000)
-		job->timeout = args->timeout;
-
-	err = host1x_job_pin(job, context->client->dev);
-	if (err)
-		goto fail;
-
-	err = host1x_job_submit(job);
-	if (err)
-		goto fail_submit;
-
-	args->fence = job->syncpt_end;
-
-	host1x_job_put(job);
-	return 0;
-
-fail_submit:
-	host1x_job_unpin(job);
-fail:
-	host1x_job_put(job);
-	return err;
-}
-
-static struct host1x_client_ops gr2d_client_ops = {
-	.drm_init = gr2d_client_init,
-	.drm_exit = gr2d_client_exit,
-	.open_channel = gr2d_open_channel,
-	.close_channel = gr2d_close_channel,
-	.submit = gr2d_submit,
-};
-
-static void gr2d_init_addr_reg_map(struct device *dev, struct gr2d *gr2d)
-{
-	const u32 gr2d_addr_regs[] = {0x1a, 0x1b, 0x26, 0x2b, 0x2c, 0x2d, 0x31,
-				      0x32, 0x48, 0x49, 0x4a, 0x4b, 0x4c};
-	unsigned long *bitmap;
-	int i;
-
-	bitmap = devm_kzalloc(dev, DIV_ROUND_UP(256, BITS_PER_BYTE),
-			      GFP_KERNEL);
-
-	for (i = 0; i < ARRAY_SIZE(gr2d_addr_regs); ++i) {
-		u32 reg = gr2d_addr_regs[i];
-		bitmap[BIT_WORD(reg)] |= BIT_MASK(reg);
-	}
-
-	gr2d->addr_regs = bitmap;
-}
-
-static int gr2d_is_addr_reg(struct device *dev, u32 class, u32 reg)
-{
-	struct gr2d *gr2d = dev_get_drvdata(dev);
-
-	switch (class) {
-	case HOST1X_CLASS_HOST1X:
-		return reg == 0x2b;
-	case HOST1X_CLASS_GR2D:
-	case HOST1X_CLASS_GR2D_SB:
-		reg &= 0xff;
-		if (gr2d->addr_regs[BIT_WORD(reg)] & BIT_MASK(reg))
-			return 1;
-	default:
-		return 0;
-	}
-}
-
-static const struct of_device_id gr2d_match[] = {
-	{ .compatible = "nvidia,tegra30-gr2d" },
-	{ .compatible = "nvidia,tegra20-gr2d" },
-	{ },
-};
-
-static int gr2d_probe(struct platform_device *pdev)
-{
-	struct device *dev = &pdev->dev;
-	struct host1x_drm *host1x = host1x_get_drm_data(dev->parent);
-	int err;
-	struct gr2d *gr2d = NULL;
-	struct host1x_syncpt **syncpts;
-
-	gr2d = devm_kzalloc(dev, sizeof(*gr2d), GFP_KERNEL);
-	if (!gr2d)
-		return -ENOMEM;
-
-	syncpts = devm_kzalloc(dev, sizeof(*syncpts), GFP_KERNEL);
-	if (!syncpts)
-		return -ENOMEM;
-
-	gr2d->clk = devm_clk_get(dev, NULL);
-	if (IS_ERR(gr2d->clk)) {
-		dev_err(dev, "cannot get clock\n");
-		return PTR_ERR(gr2d->clk);
-	}
-
-	err = clk_prepare_enable(gr2d->clk);
-	if (err) {
-		dev_err(dev, "cannot turn on clock\n");
-		return err;
-	}
-
-	gr2d->channel = host1x_channel_request(dev);
-	if (!gr2d->channel)
-		return -ENOMEM;
-
-	*syncpts = host1x_syncpt_request(dev, false);
-	if (!(*syncpts)) {
-		host1x_channel_free(gr2d->channel);
-		return -ENOMEM;
-	}
-
-	gr2d->client.ops = &gr2d_client_ops;
-	gr2d->client.dev = dev;
-	gr2d->client.class = HOST1X_CLASS_GR2D;
-	gr2d->client.syncpts = syncpts;
-	gr2d->client.num_syncpts = 1;
-
-	err = host1x_register_client(host1x, &gr2d->client);
-	if (err < 0) {
-		dev_err(dev, "failed to register host1x client: %d\n", err);
-		return err;
-	}
-
-	gr2d_init_addr_reg_map(dev, gr2d);
-
-	platform_set_drvdata(pdev, gr2d);
-
-	return 0;
-}
-
-static int __exit gr2d_remove(struct platform_device *pdev)
-{
-	struct host1x_drm *host1x = host1x_get_drm_data(pdev->dev.parent);
-	struct gr2d *gr2d = platform_get_drvdata(pdev);
-	unsigned int i;
-	int err;
-
-	err = host1x_unregister_client(host1x, &gr2d->client);
-	if (err < 0) {
-		dev_err(&pdev->dev, "failed to unregister client: %d\n", err);
-		return err;
-	}
-
-	for (i = 0; i < gr2d->client.num_syncpts; i++)
-		host1x_syncpt_free(gr2d->client.syncpts[i]);
-
-	host1x_channel_free(gr2d->channel);
-	clk_disable_unprepare(gr2d->clk);
-
-	return 0;
-}
-
-struct platform_driver tegra_gr2d_driver = {
-	.probe = gr2d_probe,
-	.remove = __exit_p(gr2d_remove),
-	.driver = {
-		.owner = THIS_MODULE,
-		.name = "gr2d",
-		.of_match_table = gr2d_match,
-	}
-};
diff --git a/drivers/gpu/host1x/host1x.h b/drivers/gpu/host1x/host1x.h
deleted file mode 100644
index a2bc1e65e972..000000000000
--- a/drivers/gpu/host1x/host1x.h
+++ /dev/null
@@ -1,30 +0,0 @@
-/*
- * Tegra host1x driver
- *
- * Copyright (c) 2009-2013, NVIDIA Corporation. All rights reserved.
- *
- * This program is free software; you can redistribute it and/or modify
- * it under the terms of the GNU General Public License as published by
- * the Free Software Foundation; either version 2 of the License, or
- * (at your option) any later version.
- *
- * This program is distributed in the hope that it will be useful, but WITHOUT
- * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or
- * FITNESS FOR A PARTICULAR PURPOSE.  See the GNU General Public License for
- * more details.
- *
- * You should have received a copy of the GNU General Public License along
- * with this program; if not, write to the Free Software Foundation, Inc.,
- * 51 Franklin Street, Fifth Floor, Boston, MA  02110-1301, USA.
- */
-
-#ifndef __LINUX_HOST1X_H
-#define __LINUX_HOST1X_H
-
-enum host1x_class {
-	HOST1X_CLASS_HOST1X	= 0x1,
-	HOST1X_CLASS_GR2D	= 0x51,
-	HOST1X_CLASS_GR2D_SB    = 0x52
-};
-
-#endif
diff --git a/drivers/gpu/host1x/host1x_bo.h b/drivers/gpu/host1x/host1x_bo.h
deleted file mode 100644
index 4c1f10bd773d..000000000000
--- a/drivers/gpu/host1x/host1x_bo.h
+++ /dev/null
@@ -1,87 +0,0 @@
-/*
- * Tegra host1x Memory Management Abstraction header
- *
- * Copyright (c) 2012-2013, NVIDIA Corporation.
- *
- * This program is free software; you can redistribute it and/or modify it
- * under the terms and conditions of the GNU General Public License,
- * version 2, as published by the Free Software Foundation.
- *
- * This program is distributed in the hope it will be useful, but WITHOUT
- * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or
- * FITNESS FOR A PARTICULAR PURPOSE.  See the GNU General Public License for
- * more details.
- *
- * You should have received a copy of the GNU General Public License
- * along with this program.  If not, see <http://www.gnu.org/licenses/>.
- */
-
-#ifndef _HOST1X_BO_H
-#define _HOST1X_BO_H
-
-struct host1x_bo;
-
-struct host1x_bo_ops {
-	struct host1x_bo *(*get)(struct host1x_bo *bo);
-	void (*put)(struct host1x_bo *bo);
-	dma_addr_t (*pin)(struct host1x_bo *bo, struct sg_table **sgt);
-	void (*unpin)(struct host1x_bo *bo, struct sg_table *sgt);
-	void *(*mmap)(struct host1x_bo *bo);
-	void (*munmap)(struct host1x_bo *bo, void *addr);
-	void *(*kmap)(struct host1x_bo *bo, unsigned int pagenum);
-	void (*kunmap)(struct host1x_bo *bo, unsigned int pagenum, void *addr);
-};
-
-struct host1x_bo {
-	const struct host1x_bo_ops *ops;
-};
-
-static inline void host1x_bo_init(struct host1x_bo *bo,
-				  const struct host1x_bo_ops *ops)
-{
-	bo->ops = ops;
-}
-
-static inline struct host1x_bo *host1x_bo_get(struct host1x_bo *bo)
-{
-	return bo->ops->get(bo);
-}
-
-static inline void host1x_bo_put(struct host1x_bo *bo)
-{
-	bo->ops->put(bo);
-}
-
-static inline dma_addr_t host1x_bo_pin(struct host1x_bo *bo,
-				       struct sg_table **sgt)
-{
-	return bo->ops->pin(bo, sgt);
-}
-
-static inline void host1x_bo_unpin(struct host1x_bo *bo, struct sg_table *sgt)
-{
-	bo->ops->unpin(bo, sgt);
-}
-
-static inline void *host1x_bo_mmap(struct host1x_bo *bo)
-{
-	return bo->ops->mmap(bo);
-}
-
-static inline void host1x_bo_munmap(struct host1x_bo *bo, void *addr)
-{
-	bo->ops->munmap(bo, addr);
-}
-
-static inline void *host1x_bo_kmap(struct host1x_bo *bo, unsigned int pagenum)
-{
-	return bo->ops->kmap(bo, pagenum);
-}
-
-static inline void host1x_bo_kunmap(struct host1x_bo *bo,
-				    unsigned int pagenum, void *addr)
-{
-	bo->ops->kunmap(bo, pagenum, addr);
-}
-
-#endif
diff --git a/drivers/gpu/host1x/hw/Makefile b/drivers/gpu/host1x/hw/Makefile
deleted file mode 100644
index 9b50863a2236..000000000000
--- a/drivers/gpu/host1x/hw/Makefile
+++ /dev/null
@@ -1,6 +0,0 @@
-ccflags-y = -Idrivers/gpu/host1x
-
-host1x-hw-objs  = \
-	host1x01.o
-
-obj-$(CONFIG_TEGRA_HOST1X) += host1x-hw.o
diff --git a/drivers/gpu/host1x/hw/cdma_hw.c b/drivers/gpu/host1x/hw/cdma_hw.c
index 2ee4ad55c4db..37e2a63241a9 100644
--- a/drivers/gpu/host1x/hw/cdma_hw.c
+++ b/drivers/gpu/host1x/hw/cdma_hw.c
@@ -20,10 +20,10 @@
 #include <linux/scatterlist.h>
 #include <linux/dma-mapping.h>
 
-#include "cdma.h"
-#include "channel.h"
-#include "dev.h"
-#include "debug.h"
+#include "../cdma.h"
+#include "../channel.h"
+#include "../dev.h"
+#include "../debug.h"
 
 /*
  * Put the restart at the end of pushbuffer memor
diff --git a/drivers/gpu/host1x/hw/channel_hw.c b/drivers/gpu/host1x/hw/channel_hw.c
index ee199623e365..4608257ab656 100644
--- a/drivers/gpu/host1x/hw/channel_hw.c
+++ b/drivers/gpu/host1x/hw/channel_hw.c
@@ -16,15 +16,15 @@
  * along with this program.  If not, see <http://www.gnu.org/licenses/>.
  */
 
+#include <linux/host1x.h>
 #include <linux/slab.h>
+
 #include <trace/events/host1x.h>
 
-#include "host1x.h"
-#include "host1x_bo.h"
-#include "channel.h"
-#include "dev.h"
-#include "intr.h"
-#include "job.h"
+#include "../channel.h"
+#include "../dev.h"
+#include "../intr.h"
+#include "../job.h"
 
 #define HOST1X_CHANNEL_SIZE 16384
 #define TRACE_MAX_LENGTH 128U
@@ -67,6 +67,22 @@ static void submit_gathers(struct host1x_job *job)
 	}
 }
 
+static inline void synchronize_syncpt_base(struct host1x_job *job)
+{
+	struct host1x *host = dev_get_drvdata(job->channel->dev->parent);
+	struct host1x_syncpt *sp = host->syncpt + job->syncpt_id;
+	u32 id, value;
+
+	value = host1x_syncpt_read_max(sp);
+	id = sp->base->id;
+
+	host1x_cdma_push(&job->channel->cdma,
+			 host1x_opcode_setclass(HOST1X_CLASS_HOST1X,
+				HOST1X_UCLASS_LOAD_SYNCPT_BASE, 1),
+			 HOST1X_UCLASS_LOAD_SYNCPT_BASE_BASE_INDX_F(id) |
+			 HOST1X_UCLASS_LOAD_SYNCPT_BASE_VALUE_F(value));
+}
+
 static int channel_submit(struct host1x_job *job)
 {
 	struct host1x_channel *ch = job->channel;
@@ -118,6 +134,10 @@ static int channel_submit(struct host1x_job *job)
 					host1x_syncpt_read_max(sp)));
 	}
 
+	/* Synchronize base register to allow using it for relative waiting */
+	if (sp->base)
+		synchronize_syncpt_base(job);
+
 	syncval = host1x_syncpt_incr_max(sp, user_syncpt_incrs);
 
 	job->syncpt_end = syncval;
diff --git a/drivers/gpu/host1x/hw/debug_hw.c b/drivers/gpu/host1x/hw/debug_hw.c
index 334c038052f5..640c75ca5a8b 100644
--- a/drivers/gpu/host1x/hw/debug_hw.c
+++ b/drivers/gpu/host1x/hw/debug_hw.c
@@ -15,18 +15,10 @@
  *
  */
 
-#include <linux/debugfs.h>
-#include <linux/seq_file.h>
-#include <linux/mm.h>
-#include <linux/scatterlist.h>
-
-#include <linux/io.h>
-
-#include "dev.h"
-#include "debug.h"
-#include "cdma.h"
-#include "channel.h"
-#include "host1x_bo.h"
+#include "../dev.h"
+#include "../debug.h"
+#include "../cdma.h"
+#include "../channel.h"
 
 #define HOST1X_DEBUG_MAX_PAGE_OFFSET 102400
 
diff --git a/drivers/gpu/host1x/hw/host1x01.c b/drivers/gpu/host1x/hw/host1x01.c
index a14e91cd1e58..859b73beb4d0 100644
--- a/drivers/gpu/host1x/hw/host1x01.c
+++ b/drivers/gpu/host1x/hw/host1x01.c
@@ -17,17 +17,17 @@
  */
 
 /* include hw specification */
-#include "hw/host1x01.h"
-#include "hw/host1x01_hardware.h"
+#include "host1x01.h"
+#include "host1x01_hardware.h"
 
 /* include code */
-#include "hw/cdma_hw.c"
-#include "hw/channel_hw.c"
-#include "hw/debug_hw.c"
-#include "hw/intr_hw.c"
-#include "hw/syncpt_hw.c"
+#include "cdma_hw.c"
+#include "channel_hw.c"
+#include "debug_hw.c"
+#include "intr_hw.c"
+#include "syncpt_hw.c"
 
-#include "dev.h"
+#include "../dev.h"
 
 int host1x01_init(struct host1x *host)
 {
diff --git a/drivers/gpu/host1x/hw/host1x02.c b/drivers/gpu/host1x/hw/host1x02.c
new file mode 100644
index 000000000000..e98caca0ca42
--- /dev/null
+++ b/drivers/gpu/host1x/hw/host1x02.c
@@ -0,0 +1,42 @@
+/*
+ * Host1x init for Tegra114 SoCs
+ *
+ * Copyright (c) 2013 NVIDIA Corporation.
+ *
+ * This program is free software; you can redistribute it and/or modify it
+ * under the terms and conditions of the GNU General Public License,
+ * version 2, as published by the Free Software Foundation.
+ *
+ * This program is distributed in the hope it will be useful, but WITHOUT
+ * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or
+ * FITNESS FOR A PARTICULAR PURPOSE.  See the GNU General Public License for
+ * more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program.  If not, see <http://www.gnu.org/licenses/>.
+ */
+
+/* include hw specification */
+#include "host1x01.h"
+#include "host1x01_hardware.h"
+
+/* include code */
+#include "cdma_hw.c"
+#include "channel_hw.c"
+#include "debug_hw.c"
+#include "intr_hw.c"
+#include "syncpt_hw.c"
+
+#include "../dev.h"
+
+int host1x02_init(struct host1x *host)
+{
+	host->channel_op = &host1x_channel_ops;
+	host->cdma_op = &host1x_cdma_ops;
+	host->cdma_pb_op = &host1x_pushbuffer_ops;
+	host->syncpt_op = &host1x_syncpt_ops;
+	host->intr_op = &host1x_intr_ops;
+	host->debug_op = &host1x_debug_ops;
+
+	return 0;
+}
diff --git a/drivers/gpu/host1x/hw/host1x02.h b/drivers/gpu/host1x/hw/host1x02.h
new file mode 100644
index 000000000000..f7486609a90e
--- /dev/null
+++ b/drivers/gpu/host1x/hw/host1x02.h
@@ -0,0 +1,26 @@
+/*
+ * Host1x init for Tegra114 SoCs
+ *
+ * Copyright (c) 2013 NVIDIA Corporation.
+ *
+ * This program is free software; you can redistribute it and/or modify it
+ * under the terms and conditions of the GNU General Public License,
+ * version 2, as published by the Free Software Foundation.
+ *
+ * This program is distributed in the hope it will be useful, but WITHOUT
+ * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or
+ * FITNESS FOR A PARTICULAR PURPOSE.  See the GNU General Public License for
+ * more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program.  If not, see <http://www.gnu.org/licenses/>.
+ */
+
+#ifndef HOST1X_HOST1X02_H
+#define HOST1X_HOST1X02_H
+
+struct host1x;
+
+int host1x02_init(struct host1x *host);
+
+#endif
diff --git a/drivers/gpu/host1x/hw/hw_host1x01_uclass.h b/drivers/gpu/host1x/hw/hw_host1x01_uclass.h
index 42f3ce19ca32..f7553599ee27 100644
--- a/drivers/gpu/host1x/hw/hw_host1x01_uclass.h
+++ b/drivers/gpu/host1x/hw/hw_host1x01_uclass.h
@@ -111,6 +111,12 @@ static inline u32 host1x_uclass_wait_syncpt_base_offset_f(u32 v)
 }
 #define HOST1X_UCLASS_WAIT_SYNCPT_BASE_OFFSET_F(v) \
 	host1x_uclass_wait_syncpt_base_offset_f(v)
+static inline u32 host1x_uclass_load_syncpt_base_r(void)
+{
+	return 0xb;
+}
+#define HOST1X_UCLASS_LOAD_SYNCPT_BASE \
+	host1x_uclass_load_syncpt_base_r()
 static inline u32 host1x_uclass_load_syncpt_base_base_indx_f(u32 v)
 {
 	return (v & 0xff) << 24;
diff --git a/drivers/gpu/host1x/hw/hw_host1x02_channel.h b/drivers/gpu/host1x/hw/hw_host1x02_channel.h
new file mode 100644
index 000000000000..e490bcde33fe
--- /dev/null
+++ b/drivers/gpu/host1x/hw/hw_host1x02_channel.h
@@ -0,0 +1,121 @@
+/*
+ * Copyright (c) 2013 NVIDIA Corporation.
+ *
+ * This program is free software; you can redistribute it and/or modify it
+ * under the terms and conditions of the GNU General Public License,
+ * version 2, as published by the Free Software Foundation.
+ *
+ * This program is distributed in the hope it will be useful, but WITHOUT
+ * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or
+ * FITNESS FOR A PARTICULAR PURPOSE.  See the GNU General Public License for
+ * more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program.  If not, see <http://www.gnu.org/licenses/>.
+ *
+ */
+
+ /*
+  * Function naming determines intended use:
+  *
+  *     <x>_r(void) : Returns the offset for register <x>.
+  *
+  *     <x>_w(void) : Returns the word offset for word (4 byte) element <x>.
+  *
+  *     <x>_<y>_s(void) : Returns size of field <y> of register <x> in bits.
+  *
+  *     <x>_<y>_f(u32 v) : Returns a value based on 'v' which has been shifted
+  *         and masked to place it at field <y> of register <x>.  This value
+  *         can be |'d with others to produce a full register value for
+  *         register <x>.
+  *
+  *     <x>_<y>_m(void) : Returns a mask for field <y> of register <x>.  This
+  *         value can be ~'d and then &'d to clear the value of field <y> for
+  *         register <x>.
+  *
+  *     <x>_<y>_<z>_f(void) : Returns the constant value <z> after being shifted
+  *         to place it at field <y> of register <x>.  This value can be |'d
+  *         with others to produce a full register value for <x>.
+  *
+  *     <x>_<y>_v(u32 r) : Returns the value of field <y> from a full register
+  *         <x> value 'r' after being shifted to place its LSB at bit 0.
+  *         This value is suitable for direct comparison with other unshifted
+  *         values appropriate for use in field <y> of register <x>.
+  *
+  *     <x>_<y>_<z>_v(void) : Returns the constant value for <z> defined for
+  *         field <y> of register <x>.  This value is suitable for direct
+  *         comparison with unshifted values appropriate for use in field <y>
+  *         of register <x>.
+  */
+
+#ifndef HOST1X_HW_HOST1X02_CHANNEL_H
+#define HOST1X_HW_HOST1X02_CHANNEL_H
+
+static inline u32 host1x_channel_fifostat_r(void)
+{
+	return 0x0;
+}
+#define HOST1X_CHANNEL_FIFOSTAT \
+	host1x_channel_fifostat_r()
+static inline u32 host1x_channel_fifostat_cfempty_v(u32 r)
+{
+	return (r >> 11) & 0x1;
+}
+#define HOST1X_CHANNEL_FIFOSTAT_CFEMPTY_V(r) \
+	host1x_channel_fifostat_cfempty_v(r)
+static inline u32 host1x_channel_dmastart_r(void)
+{
+	return 0x14;
+}
+#define HOST1X_CHANNEL_DMASTART \
+	host1x_channel_dmastart_r()
+static inline u32 host1x_channel_dmaput_r(void)
+{
+	return 0x18;
+}
+#define HOST1X_CHANNEL_DMAPUT \
+	host1x_channel_dmaput_r()
+static inline u32 host1x_channel_dmaget_r(void)
+{
+	return 0x1c;
+}
+#define HOST1X_CHANNEL_DMAGET \
+	host1x_channel_dmaget_r()
+static inline u32 host1x_channel_dmaend_r(void)
+{
+	return 0x20;
+}
+#define HOST1X_CHANNEL_DMAEND \
+	host1x_channel_dmaend_r()
+static inline u32 host1x_channel_dmactrl_r(void)
+{
+	return 0x24;
+}
+#define HOST1X_CHANNEL_DMACTRL \
+	host1x_channel_dmactrl_r()
+static inline u32 host1x_channel_dmactrl_dmastop(void)
+{
+	return 1 << 0;
+}
+#define HOST1X_CHANNEL_DMACTRL_DMASTOP \
+	host1x_channel_dmactrl_dmastop()
+static inline u32 host1x_channel_dmactrl_dmastop_v(u32 r)
+{
+	return (r >> 0) & 0x1;
+}
+#define HOST1X_CHANNEL_DMACTRL_DMASTOP_V(r) \
+	host1x_channel_dmactrl_dmastop_v(r)
+static inline u32 host1x_channel_dmactrl_dmagetrst(void)
+{
+	return 1 << 1;
+}
+#define HOST1X_CHANNEL_DMACTRL_DMAGETRST \
+	host1x_channel_dmactrl_dmagetrst()
+static inline u32 host1x_channel_dmactrl_dmainitget(void)
+{
+	return 1 << 2;
+}
+#define HOST1X_CHANNEL_DMACTRL_DMAINITGET \
+	host1x_channel_dmactrl_dmainitget()
+
+#endif
diff --git a/drivers/gpu/host1x/hw/hw_host1x02_sync.h b/drivers/gpu/host1x/hw/hw_host1x02_sync.h
new file mode 100644
index 000000000000..4495401525e8
--- /dev/null
+++ b/drivers/gpu/host1x/hw/hw_host1x02_sync.h
@@ -0,0 +1,243 @@
+/*
+ * Copyright (c) 2013 NVIDIA Corporation.
+ *
+ * This program is free software; you can redistribute it and/or modify it
+ * under the terms and conditions of the GNU General Public License,
+ * version 2, as published by the Free Software Foundation.
+ *
+ * This program is distributed in the hope it will be useful, but WITHOUT
+ * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or
+ * FITNESS FOR A PARTICULAR PURPOSE.  See the GNU General Public License for
+ * more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program.  If not, see <http://www.gnu.org/licenses/>.
+ *
+ */
+
+ /*
+  * Function naming determines intended use:
+  *
+  *     <x>_r(void) : Returns the offset for register <x>.
+  *
+  *     <x>_w(void) : Returns the word offset for word (4 byte) element <x>.
+  *
+  *     <x>_<y>_s(void) : Returns size of field <y> of register <x> in bits.
+  *
+  *     <x>_<y>_f(u32 v) : Returns a value based on 'v' which has been shifted
+  *         and masked to place it at field <y> of register <x>.  This value
+  *         can be |'d with others to produce a full register value for
+  *         register <x>.
+  *
+  *     <x>_<y>_m(void) : Returns a mask for field <y> of register <x>.  This
+  *         value can be ~'d and then &'d to clear the value of field <y> for
+  *         register <x>.
+  *
+  *     <x>_<y>_<z>_f(void) : Returns the constant value <z> after being shifted
+  *         to place it at field <y> of register <x>.  This value can be |'d
+  *         with others to produce a full register value for <x>.
+  *
+  *     <x>_<y>_v(u32 r) : Returns the value of field <y> from a full register
+  *         <x> value 'r' after being shifted to place its LSB at bit 0.
+  *         This value is suitable for direct comparison with other unshifted
+  *         values appropriate for use in field <y> of register <x>.
+  *
+  *     <x>_<y>_<z>_v(void) : Returns the constant value for <z> defined for
+  *         field <y> of register <x>.  This value is suitable for direct
+  *         comparison with unshifted values appropriate for use in field <y>
+  *         of register <x>.
+  */
+
+#ifndef HOST1X_HW_HOST1X02_SYNC_H
+#define HOST1X_HW_HOST1X02_SYNC_H
+
+#define REGISTER_STRIDE	4
+
+static inline u32 host1x_sync_syncpt_r(unsigned int id)
+{
+	return 0x400 + id * REGISTER_STRIDE;
+}
+#define HOST1X_SYNC_SYNCPT(id) \
+	host1x_sync_syncpt_r(id)
+static inline u32 host1x_sync_syncpt_thresh_cpu0_int_status_r(unsigned int id)
+{
+	return 0x40 + id * REGISTER_STRIDE;
+}
+#define HOST1X_SYNC_SYNCPT_THRESH_CPU0_INT_STATUS(id) \
+	host1x_sync_syncpt_thresh_cpu0_int_status_r(id)
+static inline u32 host1x_sync_syncpt_thresh_int_disable_r(unsigned int id)
+{
+	return 0x60 + id * REGISTER_STRIDE;
+}
+#define HOST1X_SYNC_SYNCPT_THRESH_INT_DISABLE(id) \
+	host1x_sync_syncpt_thresh_int_disable_r(id)
+static inline u32 host1x_sync_syncpt_thresh_int_enable_cpu0_r(unsigned int id)
+{
+	return 0x68 + id * REGISTER_STRIDE;
+}
+#define HOST1X_SYNC_SYNCPT_THRESH_INT_ENABLE_CPU0(id) \
+	host1x_sync_syncpt_thresh_int_enable_cpu0_r(id)
+static inline u32 host1x_sync_cf_setup_r(unsigned int channel)
+{
+	return 0x80 + channel * REGISTER_STRIDE;
+}
+#define HOST1X_SYNC_CF_SETUP(channel) \
+	host1x_sync_cf_setup_r(channel)
+static inline u32 host1x_sync_cf_setup_base_v(u32 r)
+{
+	return (r >> 0) & 0x3ff;
+}
+#define HOST1X_SYNC_CF_SETUP_BASE_V(r) \
+	host1x_sync_cf_setup_base_v(r)
+static inline u32 host1x_sync_cf_setup_limit_v(u32 r)
+{
+	return (r >> 16) & 0x3ff;
+}
+#define HOST1X_SYNC_CF_SETUP_LIMIT_V(r) \
+	host1x_sync_cf_setup_limit_v(r)
+static inline u32 host1x_sync_cmdproc_stop_r(void)
+{
+	return 0xac;
+}
+#define HOST1X_SYNC_CMDPROC_STOP \
+	host1x_sync_cmdproc_stop_r()
+static inline u32 host1x_sync_ch_teardown_r(void)
+{
+	return 0xb0;
+}
+#define HOST1X_SYNC_CH_TEARDOWN \
+	host1x_sync_ch_teardown_r()
+static inline u32 host1x_sync_usec_clk_r(void)
+{
+	return 0x1a4;
+}
+#define HOST1X_SYNC_USEC_CLK \
+	host1x_sync_usec_clk_r()
+static inline u32 host1x_sync_ctxsw_timeout_cfg_r(void)
+{
+	return 0x1a8;
+}
+#define HOST1X_SYNC_CTXSW_TIMEOUT_CFG \
+	host1x_sync_ctxsw_timeout_cfg_r()
+static inline u32 host1x_sync_ip_busy_timeout_r(void)
+{
+	return 0x1bc;
+}
+#define HOST1X_SYNC_IP_BUSY_TIMEOUT \
+	host1x_sync_ip_busy_timeout_r()
+static inline u32 host1x_sync_mlock_owner_r(unsigned int id)
+{
+	return 0x340 + id * REGISTER_STRIDE;
+}
+#define HOST1X_SYNC_MLOCK_OWNER(id) \
+	host1x_sync_mlock_owner_r(id)
+static inline u32 host1x_sync_mlock_owner_chid_f(u32 v)
+{
+	return (v & 0xf) << 8;
+}
+#define HOST1X_SYNC_MLOCK_OWNER_CHID_F(v) \
+	host1x_sync_mlock_owner_chid_f(v)
+static inline u32 host1x_sync_mlock_owner_cpu_owns_v(u32 r)
+{
+	return (r >> 1) & 0x1;
+}
+#define HOST1X_SYNC_MLOCK_OWNER_CPU_OWNS_V(r) \
+	host1x_sync_mlock_owner_cpu_owns_v(r)
+static inline u32 host1x_sync_mlock_owner_ch_owns_v(u32 r)
+{
+	return (r >> 0) & 0x1;
+}
+#define HOST1X_SYNC_MLOCK_OWNER_CH_OWNS_V(r) \
+	host1x_sync_mlock_owner_ch_owns_v(r)
+static inline u32 host1x_sync_syncpt_int_thresh_r(unsigned int id)
+{
+	return 0x500 + id * REGISTER_STRIDE;
+}
+#define HOST1X_SYNC_SYNCPT_INT_THRESH(id) \
+	host1x_sync_syncpt_int_thresh_r(id)
+static inline u32 host1x_sync_syncpt_base_r(unsigned int id)
+{
+	return 0x600 + id * REGISTER_STRIDE;
+}
+#define HOST1X_SYNC_SYNCPT_BASE(id) \
+	host1x_sync_syncpt_base_r(id)
+static inline u32 host1x_sync_syncpt_cpu_incr_r(unsigned int id)
+{
+	return 0x700 + id * REGISTER_STRIDE;
+}
+#define HOST1X_SYNC_SYNCPT_CPU_INCR(id) \
+	host1x_sync_syncpt_cpu_incr_r(id)
+static inline u32 host1x_sync_cbread_r(unsigned int channel)
+{
+	return 0x720 + channel * REGISTER_STRIDE;
+}
+#define HOST1X_SYNC_CBREAD(channel) \
+	host1x_sync_cbread_r(channel)
+static inline u32 host1x_sync_cfpeek_ctrl_r(void)
+{
+	return 0x74c;
+}
+#define HOST1X_SYNC_CFPEEK_CTRL \
+	host1x_sync_cfpeek_ctrl_r()
+static inline u32 host1x_sync_cfpeek_ctrl_addr_f(u32 v)
+{
+	return (v & 0x3ff) << 0;
+}
+#define HOST1X_SYNC_CFPEEK_CTRL_ADDR_F(v) \
+	host1x_sync_cfpeek_ctrl_addr_f(v)
+static inline u32 host1x_sync_cfpeek_ctrl_channr_f(u32 v)
+{
+	return (v & 0xf) << 16;
+}
+#define HOST1X_SYNC_CFPEEK_CTRL_CHANNR_F(v) \
+	host1x_sync_cfpeek_ctrl_channr_f(v)
+static inline u32 host1x_sync_cfpeek_ctrl_ena_f(u32 v)
+{
+	return (v & 0x1) << 31;
+}
+#define HOST1X_SYNC_CFPEEK_CTRL_ENA_F(v) \
+	host1x_sync_cfpeek_ctrl_ena_f(v)
+static inline u32 host1x_sync_cfpeek_read_r(void)
+{
+	return 0x750;
+}
+#define HOST1X_SYNC_CFPEEK_READ \
+	host1x_sync_cfpeek_read_r()
+static inline u32 host1x_sync_cfpeek_ptrs_r(void)
+{
+	return 0x754;
+}
+#define HOST1X_SYNC_CFPEEK_PTRS \
+	host1x_sync_cfpeek_ptrs_r()
+static inline u32 host1x_sync_cfpeek_ptrs_cf_rd_ptr_v(u32 r)
+{
+	return (r >> 0) & 0x3ff;
+}
+#define HOST1X_SYNC_CFPEEK_PTRS_CF_RD_PTR_V(r) \
+	host1x_sync_cfpeek_ptrs_cf_rd_ptr_v(r)
+static inline u32 host1x_sync_cfpeek_ptrs_cf_wr_ptr_v(u32 r)
+{
+	return (r >> 16) & 0x3ff;
+}
+#define HOST1X_SYNC_CFPEEK_PTRS_CF_WR_PTR_V(r) \
+	host1x_sync_cfpeek_ptrs_cf_wr_ptr_v(r)
+static inline u32 host1x_sync_cbstat_r(unsigned int channel)
+{
+	return 0x758 + channel * REGISTER_STRIDE;
+}
+#define HOST1X_SYNC_CBSTAT(channel) \
+	host1x_sync_cbstat_r(channel)
+static inline u32 host1x_sync_cbstat_cboffset_v(u32 r)
+{
+	return (r >> 0) & 0xffff;
+}
+#define HOST1X_SYNC_CBSTAT_CBOFFSET_V(r) \
+	host1x_sync_cbstat_cboffset_v(r)
+static inline u32 host1x_sync_cbstat_cbclass_v(u32 r)
+{
+	return (r >> 16) & 0x3ff;
+}
+#define HOST1X_SYNC_CBSTAT_CBCLASS_V(r) \
+	host1x_sync_cbstat_cbclass_v(r)
+
+#endif
diff --git a/drivers/gpu/host1x/hw/hw_host1x02_uclass.h b/drivers/gpu/host1x/hw/hw_host1x02_uclass.h
new file mode 100644
index 000000000000..a3b3c9874413
--- /dev/null
+++ b/drivers/gpu/host1x/hw/hw_host1x02_uclass.h
@@ -0,0 +1,175 @@
+/*
+ * Copyright (c) 2013 NVIDIA Corporation.
+ *
+ * This program is free software; you can redistribute it and/or modify it
+ * under the terms and conditions of the GNU General Public License,
+ * version 2, as published by the Free Software Foundation.
+ *
+ * This program is distributed in the hope it will be useful, but WITHOUT
+ * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or
+ * FITNESS FOR A PARTICULAR PURPOSE.  See the GNU General Public License for
+ * more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program.  If not, see <http://www.gnu.org/licenses/>.
+ *
+ */
+
+ /*
+  * Function naming determines intended use:
+  *
+  *     <x>_r(void) : Returns the offset for register <x>.
+  *
+  *     <x>_w(void) : Returns the word offset for word (4 byte) element <x>.
+  *
+  *     <x>_<y>_s(void) : Returns size of field <y> of register <x> in bits.
+  *
+  *     <x>_<y>_f(u32 v) : Returns a value based on 'v' which has been shifted
+  *         and masked to place it at field <y> of register <x>.  This value
+  *         can be |'d with others to produce a full register value for
+  *         register <x>.
+  *
+  *     <x>_<y>_m(void) : Returns a mask for field <y> of register <x>.  This
+  *         value can be ~'d and then &'d to clear the value of field <y> for
+  *         register <x>.
+  *
+  *     <x>_<y>_<z>_f(void) : Returns the constant value <z> after being shifted
+  *         to place it at field <y> of register <x>.  This value can be |'d
+  *         with others to produce a full register value for <x>.
+  *
+  *     <x>_<y>_v(u32 r) : Returns the value of field <y> from a full register
+  *         <x> value 'r' after being shifted to place its LSB at bit 0.
+  *         This value is suitable for direct comparison with other unshifted
+  *         values appropriate for use in field <y> of register <x>.
+  *
+  *     <x>_<y>_<z>_v(void) : Returns the constant value for <z> defined for
+  *         field <y> of register <x>.  This value is suitable for direct
+  *         comparison with unshifted values appropriate for use in field <y>
+  *         of register <x>.
+  */
+
+#ifndef HOST1X_HW_HOST1X02_UCLASS_H
+#define HOST1X_HW_HOST1X02_UCLASS_H
+
+static inline u32 host1x_uclass_incr_syncpt_r(void)
+{
+	return 0x0;
+}
+#define HOST1X_UCLASS_INCR_SYNCPT \
+	host1x_uclass_incr_syncpt_r()
+static inline u32 host1x_uclass_incr_syncpt_cond_f(u32 v)
+{
+	return (v & 0xff) << 8;
+}
+#define HOST1X_UCLASS_INCR_SYNCPT_COND_F(v) \
+	host1x_uclass_incr_syncpt_cond_f(v)
+static inline u32 host1x_uclass_incr_syncpt_indx_f(u32 v)
+{
+	return (v & 0xff) << 0;
+}
+#define HOST1X_UCLASS_INCR_SYNCPT_INDX_F(v) \
+	host1x_uclass_incr_syncpt_indx_f(v)
+static inline u32 host1x_uclass_wait_syncpt_r(void)
+{
+	return 0x8;
+}
+#define HOST1X_UCLASS_WAIT_SYNCPT \
+	host1x_uclass_wait_syncpt_r()
+static inline u32 host1x_uclass_wait_syncpt_indx_f(u32 v)
+{
+	return (v & 0xff) << 24;
+}
+#define HOST1X_UCLASS_WAIT_SYNCPT_INDX_F(v) \
+	host1x_uclass_wait_syncpt_indx_f(v)
+static inline u32 host1x_uclass_wait_syncpt_thresh_f(u32 v)
+{
+	return (v & 0xffffff) << 0;
+}
+#define HOST1X_UCLASS_WAIT_SYNCPT_THRESH_F(v) \
+	host1x_uclass_wait_syncpt_thresh_f(v)
+static inline u32 host1x_uclass_wait_syncpt_base_r(void)
+{
+	return 0x9;
+}
+#define HOST1X_UCLASS_WAIT_SYNCPT_BASE \
+	host1x_uclass_wait_syncpt_base_r()
+static inline u32 host1x_uclass_wait_syncpt_base_indx_f(u32 v)
+{
+	return (v & 0xff) << 24;
+}
+#define HOST1X_UCLASS_WAIT_SYNCPT_BASE_INDX_F(v) \
+	host1x_uclass_wait_syncpt_base_indx_f(v)
+static inline u32 host1x_uclass_wait_syncpt_base_base_indx_f(u32 v)
+{
+	return (v & 0xff) << 16;
+}
+#define HOST1X_UCLASS_WAIT_SYNCPT_BASE_BASE_INDX_F(v) \
+	host1x_uclass_wait_syncpt_base_base_indx_f(v)
+static inline u32 host1x_uclass_wait_syncpt_base_offset_f(u32 v)
+{
+	return (v & 0xffff) << 0;
+}
+#define HOST1X_UCLASS_WAIT_SYNCPT_BASE_OFFSET_F(v) \
+	host1x_uclass_wait_syncpt_base_offset_f(v)
+static inline u32 host1x_uclass_load_syncpt_base_base_indx_f(u32 v)
+{
+	return (v & 0xff) << 24;
+}
+#define HOST1X_UCLASS_LOAD_SYNCPT_BASE_BASE_INDX_F(v) \
+	host1x_uclass_load_syncpt_base_base_indx_f(v)
+static inline u32 host1x_uclass_load_syncpt_base_value_f(u32 v)
+{
+	return (v & 0xffffff) << 0;
+}
+#define HOST1X_UCLASS_LOAD_SYNCPT_BASE_VALUE_F(v) \
+	host1x_uclass_load_syncpt_base_value_f(v)
+static inline u32 host1x_uclass_incr_syncpt_base_base_indx_f(u32 v)
+{
+	return (v & 0xff) << 24;
+}
+#define HOST1X_UCLASS_INCR_SYNCPT_BASE_BASE_INDX_F(v) \
+	host1x_uclass_incr_syncpt_base_base_indx_f(v)
+static inline u32 host1x_uclass_incr_syncpt_base_offset_f(u32 v)
+{
+	return (v & 0xffffff) << 0;
+}
+#define HOST1X_UCLASS_INCR_SYNCPT_BASE_OFFSET_F(v) \
+	host1x_uclass_incr_syncpt_base_offset_f(v)
+static inline u32 host1x_uclass_indoff_r(void)
+{
+	return 0x2d;
+}
+#define HOST1X_UCLASS_INDOFF \
+	host1x_uclass_indoff_r()
+static inline u32 host1x_uclass_indoff_indbe_f(u32 v)
+{
+	return (v & 0xf) << 28;
+}
+#define HOST1X_UCLASS_INDOFF_INDBE_F(v) \
+	host1x_uclass_indoff_indbe_f(v)
+static inline u32 host1x_uclass_indoff_autoinc_f(u32 v)
+{
+	return (v & 0x1) << 27;
+}
+#define HOST1X_UCLASS_INDOFF_AUTOINC_F(v) \
+	host1x_uclass_indoff_autoinc_f(v)
+static inline u32 host1x_uclass_indoff_indmodid_f(u32 v)
+{
+	return (v & 0xff) << 18;
+}
+#define HOST1X_UCLASS_INDOFF_INDMODID_F(v) \
+	host1x_uclass_indoff_indmodid_f(v)
+static inline u32 host1x_uclass_indoff_indroffset_f(u32 v)
+{
+	return (v & 0xffff) << 2;
+}
+#define HOST1X_UCLASS_INDOFF_INDROFFSET_F(v) \
+	host1x_uclass_indoff_indroffset_f(v)
+static inline u32 host1x_uclass_indoff_rwn_read_v(void)
+{
+	return 1;
+}
+#define HOST1X_UCLASS_INDOFF_INDROFFSET_F(v) \
+	host1x_uclass_indoff_indroffset_f(v)
+
+#endif
diff --git a/drivers/gpu/host1x/hw/intr_hw.c b/drivers/gpu/host1x/hw/intr_hw.c
index b592eef1efcb..b26dcc83bc1b 100644
--- a/drivers/gpu/host1x/hw/intr_hw.c
+++ b/drivers/gpu/host1x/hw/intr_hw.c
@@ -22,8 +22,8 @@
 #include <linux/io.h>
 #include <asm/mach/irq.h>
 
-#include "intr.h"
-#include "dev.h"
+#include "../intr.h"
+#include "../dev.h"
 
 /*
  * Sync point threshold interrupt service function
diff --git a/drivers/gpu/host1x/hw/syncpt_hw.c b/drivers/gpu/host1x/hw/syncpt_hw.c
index 0cf6095d3367..56e85395ac24 100644
--- a/drivers/gpu/host1x/hw/syncpt_hw.c
+++ b/drivers/gpu/host1x/hw/syncpt_hw.c
@@ -18,8 +18,8 @@
 
 #include <linux/io.h>
 
-#include "dev.h"
-#include "syncpt.h"
+#include "../dev.h"
+#include "../syncpt.h"
 
 /*
  * Write the current syncpoint value back to hw.
diff --git a/drivers/gpu/host1x/job.c b/drivers/gpu/host1x/job.c
index c4e1050f2252..de5ec333ce1a 100644
--- a/drivers/gpu/host1x/job.c
+++ b/drivers/gpu/host1x/job.c
@@ -18,6 +18,7 @@
 
 #include <linux/dma-mapping.h>
 #include <linux/err.h>
+#include <linux/host1x.h>
 #include <linux/kref.h>
 #include <linux/module.h>
 #include <linux/scatterlist.h>
@@ -27,7 +28,6 @@
 
 #include "channel.h"
 #include "dev.h"
-#include "host1x_bo.h"
 #include "job.h"
 #include "syncpt.h"
 
@@ -264,7 +264,7 @@ static unsigned int do_relocs(struct host1x_job *job, struct host1x_bo *cmdbuf)
 }
 
 static bool check_reloc(struct host1x_reloc *reloc, struct host1x_bo *cmdbuf,
-		       unsigned int offset)
+			unsigned int offset)
 {
 	offset *= sizeof(u32);
 
@@ -281,7 +281,7 @@ struct host1x_firewall {
 	unsigned int num_relocs;
 	struct host1x_reloc *reloc;
 
-	struct host1x_bo *cmdbuf_id;
+	struct host1x_bo *cmdbuf;
 	unsigned int offset;
 
 	u32 words;
@@ -291,25 +291,37 @@ struct host1x_firewall {
 	u32 count;
 };
 
+static int check_register(struct host1x_firewall *fw, unsigned long offset)
+{
+	if (fw->job->is_addr_reg(fw->dev, fw->class, offset)) {
+		if (!fw->num_relocs)
+			return -EINVAL;
+
+		if (!check_reloc(fw->reloc, fw->cmdbuf, fw->offset))
+			return -EINVAL;
+
+		fw->num_relocs--;
+		fw->reloc++;
+	}
+
+	return 0;
+}
+
 static int check_mask(struct host1x_firewall *fw)
 {
 	u32 mask = fw->mask;
 	u32 reg = fw->reg;
+	int ret;
 
 	while (mask) {
 		if (fw->words == 0)
 			return -EINVAL;
 
 		if (mask & 1) {
-			if (fw->job->is_addr_reg(fw->dev, fw->class, reg)) {
-				if (!fw->num_relocs)
-					return -EINVAL;
-				if (!check_reloc(fw->reloc, fw->cmdbuf_id,
-						 fw->offset))
-					return -EINVAL;
-				fw->reloc++;
-				fw->num_relocs--;
-			}
+			ret = check_register(fw, reg);
+			if (ret < 0)
+				return ret;
+
 			fw->words--;
 			fw->offset++;
 		}
@@ -324,19 +336,16 @@ static int check_incr(struct host1x_firewall *fw)
 {
 	u32 count = fw->count;
 	u32 reg = fw->reg;
+	int ret;
 
 	while (count) {
 		if (fw->words == 0)
 			return -EINVAL;
 
-		if (fw->job->is_addr_reg(fw->dev, fw->class, reg)) {
-			if (!fw->num_relocs)
-				return -EINVAL;
-			if (!check_reloc(fw->reloc, fw->cmdbuf_id, fw->offset))
-				return -EINVAL;
-			fw->reloc++;
-			fw->num_relocs--;
-		}
+		ret = check_register(fw, reg);
+		if (ret < 0)
+			return ret;
+
 		reg++;
 		fw->words--;
 		fw->offset++;
@@ -348,21 +357,17 @@ static int check_incr(struct host1x_firewall *fw)
 
 static int check_nonincr(struct host1x_firewall *fw)
 {
-	int is_addr_reg = fw->job->is_addr_reg(fw->dev, fw->class, fw->reg);
 	u32 count = fw->count;
+	int ret;
 
 	while (count) {
 		if (fw->words == 0)
 			return -EINVAL;
 
-		if (is_addr_reg) {
-			if (!fw->num_relocs)
-				return -EINVAL;
-			if (!check_reloc(fw->reloc, fw->cmdbuf_id, fw->offset))
-				return -EINVAL;
-			fw->reloc++;
-			fw->num_relocs--;
-		}
+		ret = check_register(fw, fw->reg);
+		if (ret < 0)
+			return ret;
+
 		fw->words--;
 		fw->offset++;
 		count--;
@@ -381,7 +386,7 @@ static int validate(struct host1x_firewall *fw, struct host1x_job_gather *g)
 		return 0;
 
 	fw->words = g->words;
-	fw->cmdbuf_id = g->bo;
+	fw->cmdbuf = g->bo;
 	fw->offset = 0;
 
 	while (fw->words && !err) {
@@ -436,10 +441,6 @@ static int validate(struct host1x_firewall *fw, struct host1x_job_gather *g)
 		}
 	}
 
-	/* No relocs should remain at this point */
-	if (fw->num_relocs)
-		err = -EINVAL;
-
 out:
 	return err;
 }
@@ -493,6 +494,10 @@ static inline int copy_gathers(struct host1x_job *job, struct device *dev)
 		offset += g->words * sizeof(u32);
 	}
 
+	/* No relocs should remain at this point */
+	if (fw.num_relocs)
+		return -EINVAL;
+
 	return 0;
 }
 
diff --git a/drivers/gpu/host1x/job.h b/drivers/gpu/host1x/job.h
index fba45f20458e..33a697d6dcef 100644
--- a/drivers/gpu/host1x/job.h
+++ b/drivers/gpu/host1x/job.h
@@ -34,15 +34,6 @@ struct host1x_cmdbuf {
 	u32 pad;
 };
 
-struct host1x_reloc {
-	struct host1x_bo *cmdbuf;
-	u32 cmdbuf_offset;
-	struct host1x_bo *target;
-	u32 target_offset;
-	u32 shift;
-	u32 pad;
-};
-
 struct host1x_waitchk {
 	struct host1x_bo *bo;
 	u32 offset;
@@ -56,105 +47,6 @@ struct host1x_job_unpin_data {
 };
 
 /*
- * Each submit is tracked as a host1x_job.
- */
-struct host1x_job {
-	/* When refcount goes to zero, job can be freed */
-	struct kref ref;
-
-	/* List entry */
-	struct list_head list;
-
-	/* Channel where job is submitted to */
-	struct host1x_channel *channel;
-
-	u32 client;
-
-	/* Gathers and their memory */
-	struct host1x_job_gather *gathers;
-	unsigned int num_gathers;
-
-	/* Wait checks to be processed at submit time */
-	struct host1x_waitchk *waitchk;
-	unsigned int num_waitchk;
-	u32 waitchk_mask;
-
-	/* Array of handles to be pinned & unpinned */
-	struct host1x_reloc *relocarray;
-	unsigned int num_relocs;
-	struct host1x_job_unpin_data *unpins;
-	unsigned int num_unpins;
-
-	dma_addr_t *addr_phys;
-	dma_addr_t *gather_addr_phys;
-	dma_addr_t *reloc_addr_phys;
-
-	/* Sync point id, number of increments and end related to the submit */
-	u32 syncpt_id;
-	u32 syncpt_incrs;
-	u32 syncpt_end;
-
-	/* Maximum time to wait for this job */
-	unsigned int timeout;
-
-	/* Index and number of slots used in the push buffer */
-	unsigned int first_get;
-	unsigned int num_slots;
-
-	/* Copy of gathers */
-	size_t gather_copy_size;
-	dma_addr_t gather_copy;
-	u8 *gather_copy_mapped;
-
-	/* Check if register is marked as an address reg */
-	int (*is_addr_reg)(struct device *dev, u32 reg, u32 class);
-
-	/* Request a SETCLASS to this class */
-	u32 class;
-
-	/* Add a channel wait for previous ops to complete */
-	bool serialize;
-};
-/*
- * Allocate memory for a job. Just enough memory will be allocated to
- * accomodate the submit.
- */
-struct host1x_job *host1x_job_alloc(struct host1x_channel *ch,
-				    u32 num_cmdbufs, u32 num_relocs,
-				    u32 num_waitchks);
-
-/*
- * Add a gather to a job.
- */
-void host1x_job_add_gather(struct host1x_job *job, struct host1x_bo *mem_id,
-			   u32 words, u32 offset);
-
-/*
- * Increment reference going to host1x_job.
- */
-struct host1x_job *host1x_job_get(struct host1x_job *job);
-
-/*
- * Decrement reference job, free if goes to zero.
- */
-void host1x_job_put(struct host1x_job *job);
-
-/*
- * Pin memory related to job. This handles relocation of addresses to the
- * host1x address space. Handles both the gather memory and any other memory
- * referred to from the gather buffers.
- *
- * Handles also patching out host waits that would wait for an expired sync
- * point value.
- */
-int host1x_job_pin(struct host1x_job *job, struct device *dev);
-
-/*
- * Unpin memory related to job.
- */
-void host1x_job_unpin(struct host1x_job *job);
-
-/*
  * Dump contents of job to debug output.
  */
 void host1x_job_dump(struct device *dev, struct host1x_job *job);
diff --git a/drivers/gpu/host1x/syncpt.c b/drivers/gpu/host1x/syncpt.c
index 409745b949db..159c479829c9 100644
--- a/drivers/gpu/host1x/syncpt.c
+++ b/drivers/gpu/host1x/syncpt.c
@@ -30,9 +30,32 @@
 #define SYNCPT_CHECK_PERIOD (2 * HZ)
 #define MAX_STUCK_CHECK_COUNT 15
 
-static struct host1x_syncpt *_host1x_syncpt_alloc(struct host1x *host,
-						  struct device *dev,
-						  bool client_managed)
+static struct host1x_syncpt_base *
+host1x_syncpt_base_request(struct host1x *host)
+{
+	struct host1x_syncpt_base *bases = host->bases;
+	unsigned int i;
+
+	for (i = 0; i < host->info->nb_bases; i++)
+		if (!bases[i].requested)
+			break;
+
+	if (i >= host->info->nb_bases)
+		return NULL;
+
+	bases[i].requested = true;
+	return &bases[i];
+}
+
+static void host1x_syncpt_base_free(struct host1x_syncpt_base *base)
+{
+	if (base)
+		base->requested = false;
+}
+
+static struct host1x_syncpt *host1x_syncpt_alloc(struct host1x *host,
+						 struct device *dev,
+						 unsigned long flags)
 {
 	int i;
 	struct host1x_syncpt *sp = host->syncpt;
@@ -44,6 +67,12 @@ static struct host1x_syncpt *_host1x_syncpt_alloc(struct host1x *host,
 	if (i >= host->info->nb_pts)
 		return NULL;
 
+	if (flags & HOST1X_SYNCPT_HAS_BASE) {
+		sp->base = host1x_syncpt_base_request(host);
+		if (!sp->base)
+			return NULL;
+	}
+
 	name = kasprintf(GFP_KERNEL, "%02d-%s", sp->id,
 			dev ? dev_name(dev) : NULL);
 	if (!name)
@@ -51,7 +80,11 @@ static struct host1x_syncpt *_host1x_syncpt_alloc(struct host1x *host,
 
 	sp->dev = dev;
 	sp->name = name;
-	sp->client_managed = client_managed;
+
+	if (flags & HOST1X_SYNCPT_CLIENT_MANAGED)
+		sp->client_managed = true;
+	else
+		sp->client_managed = false;
 
 	return sp;
 }
@@ -303,25 +336,35 @@ int host1x_syncpt_patch_wait(struct host1x_syncpt *sp, void *patch_addr)
 
 int host1x_syncpt_init(struct host1x *host)
 {
+	struct host1x_syncpt_base *bases;
 	struct host1x_syncpt *syncpt;
 	int i;
 
 	syncpt = devm_kzalloc(host->dev, sizeof(*syncpt) * host->info->nb_pts,
-		GFP_KERNEL);
+			      GFP_KERNEL);
 	if (!syncpt)
 		return -ENOMEM;
 
-	for (i = 0; i < host->info->nb_pts; ++i) {
+	bases = devm_kzalloc(host->dev, sizeof(*bases) * host->info->nb_bases,
+			     GFP_KERNEL);
+	if (!bases)
+		return -ENOMEM;
+
+	for (i = 0; i < host->info->nb_pts; i++) {
 		syncpt[i].id = i;
 		syncpt[i].host = host;
 	}
 
+	for (i = 0; i < host->info->nb_bases; i++)
+		bases[i].id = i;
+
 	host->syncpt = syncpt;
+	host->bases = bases;
 
 	host1x_syncpt_restore(host);
 
 	/* Allocate sync point to use for clearing waits for expired fences */
-	host->nop_sp = _host1x_syncpt_alloc(host, NULL, false);
+	host->nop_sp = host1x_syncpt_alloc(host, NULL, 0);
 	if (!host->nop_sp)
 		return -ENOMEM;
 
@@ -329,10 +372,10 @@ int host1x_syncpt_init(struct host1x *host)
 }
 
 struct host1x_syncpt *host1x_syncpt_request(struct device *dev,
-					    bool client_managed)
+					    unsigned long flags)
 {
 	struct host1x *host = dev_get_drvdata(dev->parent);
-	return _host1x_syncpt_alloc(host, dev, client_managed);
+	return host1x_syncpt_alloc(host, dev, flags);
 }
 
 void host1x_syncpt_free(struct host1x_syncpt *sp)
@@ -340,7 +383,9 @@ void host1x_syncpt_free(struct host1x_syncpt *sp)
 	if (!sp)
 		return;
 
+	host1x_syncpt_base_free(sp->base);
 	kfree(sp->name);
+	sp->base = NULL;
 	sp->dev = NULL;
 	sp->name = NULL;
 	sp->client_managed = false;
@@ -354,6 +399,25 @@ void host1x_syncpt_deinit(struct host1x *host)
 		kfree(sp->name);
 }
 
+/*
+ * Read max. It indicates how many operations there are in queue, either in
+ * channel or in a software thread.
+ * */
+u32 host1x_syncpt_read_max(struct host1x_syncpt *sp)
+{
+	smp_rmb();
+	return (u32)atomic_read(&sp->max_val);
+}
+
+/*
+ * Read min, which is a shadow of the current sync point value in hardware.
+ */
+u32 host1x_syncpt_read_min(struct host1x_syncpt *sp)
+{
+	smp_rmb();
+	return (u32)atomic_read(&sp->min_val);
+}
+
 int host1x_syncpt_nb_pts(struct host1x *host)
 {
 	return host->info->nb_pts;
@@ -375,3 +439,13 @@ struct host1x_syncpt *host1x_syncpt_get(struct host1x *host, u32 id)
 		return NULL;
 	return host->syncpt + id;
 }
+
+struct host1x_syncpt_base *host1x_syncpt_get_base(struct host1x_syncpt *sp)
+{
+	return sp ? sp->base : NULL;
+}
+
+u32 host1x_syncpt_base_id(struct host1x_syncpt_base *base)
+{
+	return base->id;
+}
diff --git a/drivers/gpu/host1x/syncpt.h b/drivers/gpu/host1x/syncpt.h
index 267c0b9d3647..9056465ecd3f 100644
--- a/drivers/gpu/host1x/syncpt.h
+++ b/drivers/gpu/host1x/syncpt.h
@@ -20,6 +20,7 @@
 #define __HOST1X_SYNCPT_H
 
 #include <linux/atomic.h>
+#include <linux/host1x.h>
 #include <linux/kernel.h>
 #include <linux/sched.h>
 
@@ -30,6 +31,11 @@ struct host1x;
 /* Reserved for replacing an expired wait with a NOP */
 #define HOST1X_SYNCPT_RESERVED			0
 
+struct host1x_syncpt_base {
+	unsigned int id;
+	bool requested;
+};
+
 struct host1x_syncpt {
 	int id;
 	atomic_t min_val;
@@ -39,6 +45,7 @@ struct host1x_syncpt {
 	bool client_managed;
 	struct host1x *host;
 	struct device *dev;
+	struct host1x_syncpt_base *base;
 
 	/* interrupt data */
 	struct host1x_syncpt_intr intr;
@@ -50,25 +57,6 @@ int host1x_syncpt_init(struct host1x *host);
 /*  Free sync point array */
 void host1x_syncpt_deinit(struct host1x *host);
 
-/*
- * Read max. It indicates how many operations there are in queue, either in
- * channel or in a software thread.
- * */
-static inline u32 host1x_syncpt_read_max(struct host1x_syncpt *sp)
-{
-	smp_rmb();
-	return (u32)atomic_read(&sp->max_val);
-}
-
-/*
- * Read min, which is a shadow of the current sync point value in hardware.
- */
-static inline u32 host1x_syncpt_read_min(struct host1x_syncpt *sp)
-{
-	smp_rmb();
-	return (u32)atomic_read(&sp->min_val);
-}
-
 /* Return number of sync point supported. */
 int host1x_syncpt_nb_pts(struct host1x *host);
 
@@ -112,9 +100,6 @@ static inline bool host1x_syncpt_idle(struct host1x_syncpt *sp)
 	return (min == max);
 }
 
-/* Return pointer to struct denoting sync point id. */
-struct host1x_syncpt *host1x_syncpt_get(struct host1x *host, u32 id);
-
 /* Load current value from hardware to the shadow register. */
 u32 host1x_syncpt_load(struct host1x_syncpt *sp);
 
@@ -130,16 +115,9 @@ void host1x_syncpt_restore(struct host1x *host);
 /* Read current wait base value into shadow register and return it. */
 u32 host1x_syncpt_load_wait_base(struct host1x_syncpt *sp);
 
-/* Request incrementing a sync point. */
-int host1x_syncpt_incr(struct host1x_syncpt *sp);
-
 /* Indicate future operations by incrementing the sync point max. */
 u32 host1x_syncpt_incr_max(struct host1x_syncpt *sp, u32 incrs);
 
-/* Wait until sync point reaches a threshold value, or a timeout. */
-int host1x_syncpt_wait(struct host1x_syncpt *sp, u32 thresh,
-			long timeout, u32 *value);
-
 /* Check if sync point id is valid. */
 static inline int host1x_syncpt_is_valid(struct host1x_syncpt *sp)
 {
@@ -149,14 +127,4 @@ static inline int host1x_syncpt_is_valid(struct host1x_syncpt *sp)
 /* Patch a wait by replacing it with a wait for syncpt 0 value 0 */
 int host1x_syncpt_patch_wait(struct host1x_syncpt *sp, void *patch_addr);
 
-/* Return id of the sync point */
-u32 host1x_syncpt_id(struct host1x_syncpt *sp);
-
-/* Allocate a sync point for a device. */
-struct host1x_syncpt *host1x_syncpt_request(struct device *dev,
-					    bool client_managed);
-
-/* Free a sync point. */
-void host1x_syncpt_free(struct host1x_syncpt *sp);
-
 #endif
diff --git a/drivers/hid/Kconfig b/drivers/hid/Kconfig
index c91d547191dd..329fbb9b5976 100644
--- a/drivers/hid/Kconfig
+++ b/drivers/hid/Kconfig
@@ -242,6 +242,7 @@ config HID_HOLTEK
 	  - Tracer Sniper TRM-503 / NOVA Gaming Slider X200 /
 	    Zalman ZM-GM1
 	  - SHARKOON DarkGlider Gaming mouse
+	  - LEETGION Hellion Gaming Mouse
 
 config HOLTEK_FF
 	bool "Holtek On Line Grip force feedback support"
@@ -323,7 +324,7 @@ config HID_LCPOWER
 
 config HID_LENOVO_TPKBD
 	tristate "Lenovo ThinkPad USB Keyboard with TrackPoint"
-	depends on USB_HID
+	depends on HID
 	select NEW_LEDS
 	select LEDS_CLASS
 	---help---
@@ -362,19 +363,20 @@ config LOGITECH_FF
 	  - Logitech WingMan Force 3D
 	  - Logitech Formula Force EX
 	  - Logitech WingMan Formula Force GP
-	  - Logitech MOMO Force wheel
 
 	  and if you want to enable force feedback for them.
 	  Note: if you say N here, this device will still be supported, but without
 	  force feedback.
 
 config LOGIRUMBLEPAD2_FF
-	bool "Logitech RumblePad/Rumblepad 2 force feedback support"
+	bool "Logitech force feedback support (variant 2)"
 	depends on HID_LOGITECH
 	select INPUT_FF_MEMLESS
 	help
-	  Say Y here if you want to enable force feedback support for Logitech
-	  RumblePad and Rumblepad 2 devices.
+	  Say Y here if you want to enable force feedback support for:
+	  - Logitech RumblePad
+	  - Logitech Rumblepad 2
+	  - Logitech Formula Vibration Feedback Wheel
 
 config LOGIG940_FF
 	bool "Logitech Flight System G940 force feedback support"
@@ -437,6 +439,7 @@ config HID_MULTITOUCH
 	  - Chunghwa panels
 	  - CVTouch panels
 	  - Cypress TrueTouch panels
+	  - Elan Microelectronics touch panels
 	  - Elo TouchSystems IntelliTouch Plus panels
 	  - GeneralTouch 'Sensing Win7-TwoFinger' panels
 	  - GoodTouch panels
@@ -453,6 +456,7 @@ config HID_MULTITOUCH
 	  - Pixcir dual touch panels
 	  - Quanta panels
 	  - eGalax dual-touch panels, including the Joojoo and Wetab tablets
+	  - SiS multitouch panels
 	  - Stantum multitouch panels
 	  - Touch International Panels
 	  - Unitec Panels
@@ -614,6 +618,14 @@ config HID_SONY
 	  * Sony PS3 Blue-ray Disk Remote Control (Bluetooth)
 	  * Logitech Harmony adapter for Sony Playstation 3 (Bluetooth)
 
+config SONY_FF
+	bool "Sony PS2/3 accessories force feedback support"
+	depends on HID_SONY
+	select INPUT_FF_MEMLESS
+	---help---
+	Say Y here if you have a Sony PS2/3 accessory and want to enable force
+	feedback support for it.
+
 config HID_SPEEDLINK
 	tristate "Speedlink VAD Cezanne mouse support"
 	depends on HID
diff --git a/drivers/hid/Makefile b/drivers/hid/Makefile
index a959f4aecaf5..30e44318f87f 100644
--- a/drivers/hid/Makefile
+++ b/drivers/hid/Makefile
@@ -95,7 +95,7 @@ obj-$(CONFIG_HID_PRIMAX)	+= hid-primax.o
 obj-$(CONFIG_HID_ROCCAT)	+= hid-roccat.o hid-roccat-common.o \
 	hid-roccat-arvo.o hid-roccat-isku.o hid-roccat-kone.o \
 	hid-roccat-koneplus.o hid-roccat-konepure.o hid-roccat-kovaplus.o \
-	hid-roccat-lua.o hid-roccat-pyra.o hid-roccat-savu.o
+	hid-roccat-lua.o hid-roccat-pyra.o hid-roccat-ryos.o hid-roccat-savu.o
 obj-$(CONFIG_HID_SAITEK)	+= hid-saitek.o
 obj-$(CONFIG_HID_SAMSUNG)	+= hid-samsung.o
 obj-$(CONFIG_HID_SMARTJOYPLUS)	+= hid-sjoy.o
diff --git a/drivers/hid/hid-apple.c b/drivers/hid/hid-apple.c
index 881cf7b4f9a4..497558127bb3 100644
--- a/drivers/hid/hid-apple.c
+++ b/drivers/hid/hid-apple.c
@@ -46,6 +46,12 @@ module_param(iso_layout, uint, 0644);
 MODULE_PARM_DESC(iso_layout, "Enable/Disable hardcoded ISO-layout of the keyboard. "
 		"(0 = disabled, [1] = enabled)");
 
+static unsigned int swap_opt_cmd;
+module_param(swap_opt_cmd, uint, 0644);
+MODULE_PARM_DESC(swap_opt_cmd, "Swap the Option (\"Alt\") and Command (\"Flag\") keys. "
+		"(For people who want to keep Windows PC keyboard muscle memory. "
+		"[0] = as-is, Mac layout. 1 = swapped, Windows layout.)");
+
 struct apple_sc {
 	unsigned long quirks;
 	unsigned int fn_on;
@@ -150,6 +156,14 @@ static const struct apple_key_translation apple_iso_keyboard[] = {
 	{ }
 };
 
+static const struct apple_key_translation swapped_option_cmd_keys[] = {
+	{ KEY_LEFTALT,	KEY_LEFTMETA },
+	{ KEY_LEFTMETA,	KEY_LEFTALT },
+	{ KEY_RIGHTALT,	KEY_RIGHTMETA },
+	{ KEY_RIGHTMETA,KEY_RIGHTALT },
+	{ }
+};
+
 static const struct apple_key_translation *apple_find_translation(
 		const struct apple_key_translation *table, u16 from)
 {
@@ -242,6 +256,14 @@ static int hidinput_apple_event(struct hid_device *hid, struct input_dev *input,
 		}
 	}
 
+	if (swap_opt_cmd) {
+		trans = apple_find_translation(swapped_option_cmd_keys, usage->code);
+		if (trans) {
+			input_event(input, usage->type, trans->to, value);
+			return 1;
+		}
+	}
+
 	return 0;
 }
 
diff --git a/drivers/hid/hid-axff.c b/drivers/hid/hid-axff.c
index 64ab94a55aa7..a594e478a1e2 100644
--- a/drivers/hid/hid-axff.c
+++ b/drivers/hid/hid-axff.c
@@ -95,7 +95,7 @@ static int axff_init(struct hid_device *hid)
 		}
 	}
 
-	if (field_count < 4) {
+	if (field_count < 4 && hid->product != 0xf705) {
 		hid_err(hid, "not enough fields in the report: %d\n",
 			field_count);
 		return -ENODEV;
@@ -180,6 +180,7 @@ static void ax_remove(struct hid_device *hdev)
 
 static const struct hid_device_id ax_devices[] = {
 	{ HID_USB_DEVICE(USB_VENDOR_ID_ACRUX, 0x0802), },
+	{ HID_USB_DEVICE(USB_VENDOR_ID_ACRUX, 0xf705), },
 	{ }
 };
 MODULE_DEVICE_TABLE(hid, ax_devices);
diff --git a/drivers/hid/hid-core.c b/drivers/hid/hid-core.c
index e80da62363bc..8c10f2742233 100644
--- a/drivers/hid/hid-core.c
+++ b/drivers/hid/hid-core.c
@@ -1418,10 +1418,8 @@ int hid_input_report(struct hid_device *hid, int type, u8 *data, int size, int i
 
 	if (hdrv && hdrv->raw_event && hid_match_report(hid, report)) {
 		ret = hdrv->raw_event(hid, report, data, size);
-		if (ret < 0) {
-			ret = ret < 0 ? ret : 0;
+		if (ret < 0)
 			goto unlock;
-		}
 	}
 
 	ret = hid_report_raw_event(hid, type, data, size, interrupt);
@@ -1605,6 +1603,7 @@ static const struct hid_device_id hid_have_special_driver[] = {
 	{ HID_USB_DEVICE(USB_VENDOR_ID_A4TECH, USB_DEVICE_ID_A4TECH_X5_005D) },
 	{ HID_USB_DEVICE(USB_VENDOR_ID_A4TECH, USB_DEVICE_ID_A4TECH_RP_649) },
 	{ HID_USB_DEVICE(USB_VENDOR_ID_ACRUX, 0x0802) },
+	{ HID_USB_DEVICE(USB_VENDOR_ID_ACRUX, 0xf705) },
 	{ HID_USB_DEVICE(USB_VENDOR_ID_APPLE, USB_DEVICE_ID_APPLE_MIGHTYMOUSE) },
 	{ HID_BLUETOOTH_DEVICE(USB_VENDOR_ID_APPLE, USB_DEVICE_ID_APPLE_MAGICMOUSE) },
 	{ HID_BLUETOOTH_DEVICE(USB_VENDOR_ID_APPLE, USB_DEVICE_ID_APPLE_MAGICTRACKPAD) },
@@ -1716,6 +1715,7 @@ static const struct hid_device_id hid_have_special_driver[] = {
 	{ HID_USB_DEVICE(USB_VENDOR_ID_HOLTEK_ALT, USB_DEVICE_ID_HOLTEK_ALT_KEYBOARD) },
 	{ HID_USB_DEVICE(USB_VENDOR_ID_HOLTEK_ALT, USB_DEVICE_ID_HOLTEK_ALT_MOUSE_A04A) },
 	{ HID_USB_DEVICE(USB_VENDOR_ID_HOLTEK_ALT, USB_DEVICE_ID_HOLTEK_ALT_MOUSE_A067) },
+	{ HID_USB_DEVICE(USB_VENDOR_ID_HOLTEK_ALT, USB_DEVICE_ID_HOLTEK_ALT_MOUSE_A072) },
 	{ HID_USB_DEVICE(USB_VENDOR_ID_HOLTEK_ALT, USB_DEVICE_ID_HOLTEK_ALT_MOUSE_A081) },
 	{ HID_USB_DEVICE(USB_VENDOR_ID_HUION, USB_DEVICE_ID_HUION_580) },
 	{ HID_USB_DEVICE(USB_VENDOR_ID_JESS2, USB_DEVICE_ID_JESS2_COLOR_RUMBLE_PAD) },
@@ -1754,6 +1754,7 @@ static const struct hid_device_id hid_have_special_driver[] = {
 	{ HID_USB_DEVICE(USB_VENDOR_ID_LOGITECH, USB_DEVICE_ID_LOGITECH_FLIGHT_SYSTEM_G940) },
 	{ HID_USB_DEVICE(USB_VENDOR_ID_LOGITECH, USB_DEVICE_ID_LOGITECH_MOMO_WHEEL) },
 	{ HID_USB_DEVICE(USB_VENDOR_ID_LOGITECH, USB_DEVICE_ID_LOGITECH_MOMO_WHEEL2) },
+	{ HID_USB_DEVICE(USB_VENDOR_ID_LOGITECH, USB_DEVICE_ID_LOGITECH_VIBRATION_WHEEL) },
 	{ HID_USB_DEVICE(USB_VENDOR_ID_LOGITECH, USB_DEVICE_ID_LOGITECH_DFP_WHEEL) },
 	{ HID_USB_DEVICE(USB_VENDOR_ID_LOGITECH, USB_DEVICE_ID_LOGITECH_DFGT_WHEEL) },
 	{ HID_USB_DEVICE(USB_VENDOR_ID_LOGITECH, USB_DEVICE_ID_LOGITECH_G25_WHEEL) },
@@ -1801,21 +1802,28 @@ static const struct hid_device_id hid_have_special_driver[] = {
 	{ HID_USB_DEVICE(USB_VENDOR_ID_PETALYNX, USB_DEVICE_ID_PETALYNX_MAXTER_REMOTE) },
 	{ HID_USB_DEVICE(USB_VENDOR_ID_PRIMAX, USB_DEVICE_ID_PRIMAX_KEYBOARD) },
 #if IS_ENABLED(CONFIG_HID_ROCCAT)
-	{ HID_USB_DEVICE(USB_VENDOR_ID_ROCCAT, USB_DEVICE_ID_ROCCAT_KONE) },
 	{ HID_USB_DEVICE(USB_VENDOR_ID_ROCCAT, USB_DEVICE_ID_ROCCAT_ARVO) },
 	{ HID_USB_DEVICE(USB_VENDOR_ID_ROCCAT, USB_DEVICE_ID_ROCCAT_ISKU) },
+	{ HID_USB_DEVICE(USB_VENDOR_ID_ROCCAT, USB_DEVICE_ID_ROCCAT_ISKUFX) },
+	{ HID_USB_DEVICE(USB_VENDOR_ID_ROCCAT, USB_DEVICE_ID_ROCCAT_KONE) },
 	{ HID_USB_DEVICE(USB_VENDOR_ID_ROCCAT, USB_DEVICE_ID_ROCCAT_KONEPLUS) },
 	{ HID_USB_DEVICE(USB_VENDOR_ID_ROCCAT, USB_DEVICE_ID_ROCCAT_KONEPURE) },
 	{ HID_USB_DEVICE(USB_VENDOR_ID_ROCCAT, USB_DEVICE_ID_ROCCAT_KONEPURE_OPTICAL) },
+	{ HID_USB_DEVICE(USB_VENDOR_ID_ROCCAT, USB_DEVICE_ID_ROCCAT_KONEXTD) },
 	{ HID_USB_DEVICE(USB_VENDOR_ID_ROCCAT, USB_DEVICE_ID_ROCCAT_KOVAPLUS) },
 	{ HID_USB_DEVICE(USB_VENDOR_ID_ROCCAT, USB_DEVICE_ID_ROCCAT_LUA) },
 	{ HID_USB_DEVICE(USB_VENDOR_ID_ROCCAT, USB_DEVICE_ID_ROCCAT_PYRA_WIRED) },
 	{ HID_USB_DEVICE(USB_VENDOR_ID_ROCCAT, USB_DEVICE_ID_ROCCAT_PYRA_WIRELESS) },
+	{ HID_USB_DEVICE(USB_VENDOR_ID_ROCCAT, USB_DEVICE_ID_ROCCAT_RYOS_MK) },
+	{ HID_USB_DEVICE(USB_VENDOR_ID_ROCCAT, USB_DEVICE_ID_ROCCAT_RYOS_MK_GLOW) },
+	{ HID_USB_DEVICE(USB_VENDOR_ID_ROCCAT, USB_DEVICE_ID_ROCCAT_RYOS_MK_PRO) },
 	{ HID_USB_DEVICE(USB_VENDOR_ID_ROCCAT, USB_DEVICE_ID_ROCCAT_SAVU) },
 #endif
 	{ HID_USB_DEVICE(USB_VENDOR_ID_SAITEK, USB_DEVICE_ID_SAITEK_PS1000) },
 	{ HID_USB_DEVICE(USB_VENDOR_ID_SAMSUNG, USB_DEVICE_ID_SAMSUNG_IR_REMOTE) },
 	{ HID_USB_DEVICE(USB_VENDOR_ID_SAMSUNG, USB_DEVICE_ID_SAMSUNG_WIRELESS_KBD_MOUSE) },
+	{ HID_USB_DEVICE(USB_VENDOR_ID_SIS2_TOUCH, USB_DEVICE_ID_SIS9200_TOUCH) },
+	{ HID_USB_DEVICE(USB_VENDOR_ID_SIS2_TOUCH, USB_DEVICE_ID_SIS817_TOUCH) },
 	{ HID_USB_DEVICE(USB_VENDOR_ID_SKYCABLE, USB_DEVICE_ID_SKYCABLE_WIRELESS_PRESENTER) },
 	{ HID_USB_DEVICE(USB_VENDOR_ID_SONY, USB_DEVICE_ID_SONY_BUZZ_CONTROLLER) },
 	{ HID_USB_DEVICE(USB_VENDOR_ID_SONY, USB_DEVICE_ID_SONY_WIRELESS_BUZZ_CONTROLLER) },
@@ -2376,15 +2384,6 @@ bool hid_ignore(struct hid_device *hdev)
 				hdev->type == HID_TYPE_USBNONE)
 			return true;
 		break;
-	case USB_VENDOR_ID_DWAV:
-		/* These are handled by usbtouchscreen. hdev->type is probably
-		 * HID_TYPE_USBNONE, but we say !HID_TYPE_USBMOUSE to match
-		 * usbtouchscreen. */
-		if ((hdev->product == USB_DEVICE_ID_EGALAX_TOUCHCONTROLLER ||
-		     hdev->product == USB_DEVICE_ID_DWAV_TOUCHCONTROLLER) &&
-		    hdev->type != HID_TYPE_USBMOUSE)
-			return true;
-		break;
 	case USB_VENDOR_ID_VELLEMAN:
 		/* These are not HID devices.  They are handled by comedi. */
 		if ((hdev->product >= USB_DEVICE_ID_VELLEMAN_K8055_FIRST &&
diff --git a/drivers/hid/hid-elo.c b/drivers/hid/hid-elo.c
index f042a6cf8b18..4e49462870ab 100644
--- a/drivers/hid/hid-elo.c
+++ b/drivers/hid/hid-elo.c
@@ -181,7 +181,40 @@ fail:
  */
 static bool elo_broken_firmware(struct usb_device *dev)
 {
-	return use_fw_quirk && le16_to_cpu(dev->descriptor.bcdDevice) == 0x10d;
+	struct usb_device *hub = dev->parent;
+	struct usb_device *child = NULL;
+	u16 fw_lvl = le16_to_cpu(dev->descriptor.bcdDevice);
+	u16 child_vid, child_pid;
+	int i;
+    
+	if (!use_fw_quirk)
+		return false;
+	if (fw_lvl != 0x10d)
+		return false;
+
+	/* iterate sibling devices of the touch controller */
+	usb_hub_for_each_child(hub, i, child) {
+		child_vid = le16_to_cpu(child->descriptor.idVendor);
+		child_pid = le16_to_cpu(child->descriptor.idProduct);
+
+		/*
+		 * If one of the devices below is present attached as a sibling of 
+		 * the touch controller then  this is a newer IBM 4820 monitor that 
+		 * does not need the IBM-requested workaround if fw level is
+		 * 0x010d - aka 'M'.
+		 * No other HW can have this combination.
+		 */
+		if (child_vid==0x04b3) {
+			switch (child_pid) {
+			case 0x4676: /* 4820 21x Video */
+			case 0x4677: /* 4820 51x Video */
+			case 0x4678: /* 4820 2Lx Video */
+			case 0x4679: /* 4820 5Lx Video */
+				return false;
+			}
+		}
+	}
+	return true;
 }
 
 static int elo_probe(struct hid_device *hdev, const struct hid_device_id *id)
diff --git a/drivers/hid/hid-holtek-mouse.c b/drivers/hid/hid-holtek-mouse.c
index e696566cde46..0caa676de622 100644
--- a/drivers/hid/hid-holtek-mouse.c
+++ b/drivers/hid/hid-holtek-mouse.c
@@ -28,6 +28,7 @@
  * - USB ID 04d9:a04a, sold as Tracer Sniper TRM-503, NOVA Gaming Slider X200
  *   and Zalman ZM-GM1
  * - USB ID 04d9:a081, sold as SHARKOON DarkGlider Gaming mouse
+ * - USB ID 04d9:a072, sold as LEETGION Hellion Gaming Mouse
  */
 
 static __u8 *holtek_mouse_report_fixup(struct hid_device *hdev, __u8 *rdesc,
@@ -40,6 +41,7 @@ static __u8 *holtek_mouse_report_fixup(struct hid_device *hdev, __u8 *rdesc,
 		 * 0x2fff, so they don't exceed HID_MAX_USAGES */
 		switch (hdev->product) {
 		case USB_DEVICE_ID_HOLTEK_ALT_MOUSE_A067:
+		case USB_DEVICE_ID_HOLTEK_ALT_MOUSE_A072:
 			if (*rsize >= 122 && rdesc[115] == 0xff && rdesc[116] == 0x7f
 					&& rdesc[120] == 0xff && rdesc[121] == 0x7f) {
 				hid_info(hdev, "Fixing up report descriptor\n");
@@ -66,6 +68,8 @@ static const struct hid_device_id holtek_mouse_devices[] = {
 	{ HID_USB_DEVICE(USB_VENDOR_ID_HOLTEK_ALT,
 			USB_DEVICE_ID_HOLTEK_ALT_MOUSE_A04A) },
 	{ HID_USB_DEVICE(USB_VENDOR_ID_HOLTEK_ALT,
+			USB_DEVICE_ID_HOLTEK_ALT_MOUSE_A072) },
+	{ HID_USB_DEVICE(USB_VENDOR_ID_HOLTEK_ALT,
 			USB_DEVICE_ID_HOLTEK_ALT_MOUSE_A081) },
 	{ }
 };
diff --git a/drivers/hid/hid-ids.h b/drivers/hid/hid-ids.h
index f0296a50be5f..76559629568c 100644
--- a/drivers/hid/hid-ids.h
+++ b/drivers/hid/hid-ids.h
@@ -332,6 +332,11 @@
 #define USB_VENDOR_ID_GENERAL_TOUCH	0x0dfc
 #define USB_DEVICE_ID_GENERAL_TOUCH_WIN7_TWOFINGERS 0x0003
 #define USB_DEVICE_ID_GENERAL_TOUCH_WIN8_PWT_TENFINGERS 0x0100
+#define USB_DEVICE_ID_GENERAL_TOUCH_WIN8_PIT_0101 0x0101
+#define USB_DEVICE_ID_GENERAL_TOUCH_WIN8_PIT_0102 0x0102
+#define USB_DEVICE_ID_GENERAL_TOUCH_WIN8_PIT_0106 0x0106
+#define USB_DEVICE_ID_GENERAL_TOUCH_WIN8_PIT_010A 0x010a
+#define USB_DEVICE_ID_GENERAL_TOUCH_WIN8_PIT_E100 0xe100
 
 #define USB_VENDOR_ID_GLAB		0x06c2
 #define USB_DEVICE_ID_4_PHIDGETSERVO_30	0x0038
@@ -448,8 +453,9 @@
 
 #define USB_VENDOR_ID_HOLTEK_ALT		0x04d9
 #define USB_DEVICE_ID_HOLTEK_ALT_KEYBOARD	0xa055
-#define USB_DEVICE_ID_HOLTEK_ALT_MOUSE_A067	0xa067
 #define USB_DEVICE_ID_HOLTEK_ALT_MOUSE_A04A	0xa04a
+#define USB_DEVICE_ID_HOLTEK_ALT_MOUSE_A067	0xa067
+#define USB_DEVICE_ID_HOLTEK_ALT_MOUSE_A072	0xa072
 #define USB_DEVICE_ID_HOLTEK_ALT_MOUSE_A081	0xa081
 
 #define USB_VENDOR_ID_IMATION		0x0718
@@ -571,6 +577,7 @@
 #define USB_DEVICE_ID_DINOVO_EDGE	0xc714
 #define USB_DEVICE_ID_DINOVO_MINI	0xc71f
 #define USB_DEVICE_ID_LOGITECH_MOMO_WHEEL2	0xca03
+#define USB_DEVICE_ID_LOGITECH_VIBRATION_WHEEL	0xca04
 
 #define USB_VENDOR_ID_LUMIO		0x202e
 #define USB_DEVICE_ID_CRYSTALTOUCH	0x0006
@@ -726,6 +733,9 @@
 #define USB_DEVICE_ID_ROCCAT_LUA	0x2c2e
 #define USB_DEVICE_ID_ROCCAT_PYRA_WIRED	0x2c24
 #define USB_DEVICE_ID_ROCCAT_PYRA_WIRELESS	0x2cf6
+#define USB_DEVICE_ID_ROCCAT_RYOS_MK	0x3138
+#define USB_DEVICE_ID_ROCCAT_RYOS_MK_GLOW	0x31ce
+#define USB_DEVICE_ID_ROCCAT_RYOS_MK_PRO	0x3232
 #define USB_DEVICE_ID_ROCCAT_SAVU	0x2d5a
 
 #define USB_VENDOR_ID_SAITEK		0x06a3
@@ -745,6 +755,10 @@
 #define USB_VENDOR_ID_SIGMATEL		0x066F
 #define USB_DEVICE_ID_SIGMATEL_STMP3780	0x3780
 
+#define USB_VENDOR_ID_SIS2_TOUCH	0x0457
+#define USB_DEVICE_ID_SIS9200_TOUCH	0x9200
+#define USB_DEVICE_ID_SIS817_TOUCH	0x0817
+
 #define USB_VENDOR_ID_SKYCABLE			0x1223
 #define	USB_DEVICE_ID_SKYCABLE_WIRELESS_PRESENTER	0x3F07
 
diff --git a/drivers/hid/hid-lenovo-tpkbd.c b/drivers/hid/hid-lenovo-tpkbd.c
index 31cf29a6ba17..2d25b6cbbc05 100644
--- a/drivers/hid/hid-lenovo-tpkbd.c
+++ b/drivers/hid/hid-lenovo-tpkbd.c
@@ -14,11 +14,9 @@
 #include <linux/module.h>
 #include <linux/sysfs.h>
 #include <linux/device.h>
-#include <linux/usb.h>
 #include <linux/hid.h>
 #include <linux/input.h>
 #include <linux/leds.h>
-#include "usbhid/usbhid.h"
 
 #include "hid-ids.h"
 
@@ -41,10 +39,9 @@ static int tpkbd_input_mapping(struct hid_device *hdev,
 		struct hid_input *hi, struct hid_field *field,
 		struct hid_usage *usage, unsigned long **bit, int *max)
 {
-	struct usbhid_device *uhdev;
-
-	uhdev = (struct usbhid_device *) hdev->driver_data;
-	if (uhdev->ifnum == 1 && usage->hid == (HID_UP_BUTTON | 0x0010)) {
+	if (usage->hid == (HID_UP_BUTTON | 0x0010)) {
+		/* mark the device as pointer */
+		hid_set_drvdata(hdev, (void *)1);
 		map_key_clear(KEY_MICMUTE);
 		return 1;
 	}
@@ -339,7 +336,7 @@ static int tpkbd_probe_tp(struct hid_device *hdev)
 	struct tpkbd_data_pointer *data_pointer;
 	size_t name_sz = strlen(dev_name(dev)) + 16;
 	char *name_mute, *name_micmute;
-	int i, ret;
+	int i;
 
 	/* Validate required reports. */
 	for (i = 0; i < 4; i++) {
@@ -354,7 +351,9 @@ static int tpkbd_probe_tp(struct hid_device *hdev)
 		hid_warn(hdev, "Could not create sysfs group\n");
 	}
 
-	data_pointer = kzalloc(sizeof(struct tpkbd_data_pointer), GFP_KERNEL);
+	data_pointer = devm_kzalloc(&hdev->dev,
+				    sizeof(struct tpkbd_data_pointer),
+				    GFP_KERNEL);
 	if (data_pointer == NULL) {
 		hid_err(hdev, "Could not allocate memory for driver data\n");
 		return -ENOMEM;
@@ -364,20 +363,13 @@ static int tpkbd_probe_tp(struct hid_device *hdev)
 	data_pointer->sensitivity = 0xa0;
 	data_pointer->press_speed = 0x38;
 
-	name_mute = kzalloc(name_sz, GFP_KERNEL);
-	if (name_mute == NULL) {
+	name_mute = devm_kzalloc(&hdev->dev, name_sz, GFP_KERNEL);
+	name_micmute = devm_kzalloc(&hdev->dev, name_sz, GFP_KERNEL);
+	if (name_mute == NULL || name_micmute == NULL) {
 		hid_err(hdev, "Could not allocate memory for led data\n");
-		ret = -ENOMEM;
-		goto err;
+		return -ENOMEM;
 	}
 	snprintf(name_mute, name_sz, "%s:amber:mute", dev_name(dev));
-
-	name_micmute = kzalloc(name_sz, GFP_KERNEL);
-	if (name_micmute == NULL) {
-		hid_err(hdev, "Could not allocate memory for led data\n");
-		ret = -ENOMEM;
-		goto err2;
-	}
 	snprintf(name_micmute, name_sz, "%s:amber:micmute", dev_name(dev));
 
 	hid_set_drvdata(hdev, data_pointer);
@@ -397,19 +389,12 @@ static int tpkbd_probe_tp(struct hid_device *hdev)
 	tpkbd_features_set(hdev);
 
 	return 0;
-
-err2:
-	kfree(name_mute);
-err:
-	kfree(data_pointer);
-	return ret;
 }
 
 static int tpkbd_probe(struct hid_device *hdev,
 		const struct hid_device_id *id)
 {
 	int ret;
-	struct usbhid_device *uhdev;
 
 	ret = hid_parse(hdev);
 	if (ret) {
@@ -423,9 +408,8 @@ static int tpkbd_probe(struct hid_device *hdev,
 		goto err;
 	}
 
-	uhdev = (struct usbhid_device *) hdev->driver_data;
-
-	if (uhdev->ifnum == 1) {
+	if (hid_get_drvdata(hdev)) {
+		hid_set_drvdata(hdev, NULL);
 		ret = tpkbd_probe_tp(hdev);
 		if (ret)
 			goto err_hid;
@@ -449,17 +433,11 @@ static void tpkbd_remove_tp(struct hid_device *hdev)
 	led_classdev_unregister(&data_pointer->led_mute);
 
 	hid_set_drvdata(hdev, NULL);
-	kfree(data_pointer->led_micmute.name);
-	kfree(data_pointer->led_mute.name);
-	kfree(data_pointer);
 }
 
 static void tpkbd_remove(struct hid_device *hdev)
 {
-	struct usbhid_device *uhdev;
-
-	uhdev = (struct usbhid_device *) hdev->driver_data;
-	if (uhdev->ifnum == 1)
+	if (hid_get_drvdata(hdev))
 		tpkbd_remove_tp(hdev);
 
 	hid_hw_stop(hdev);
diff --git a/drivers/hid/hid-lg.c b/drivers/hid/hid-lg.c
index 6f12ecd36c88..06eb45fa6331 100644
--- a/drivers/hid/hid-lg.c
+++ b/drivers/hid/hid-lg.c
@@ -45,7 +45,9 @@
 /* Size of the original descriptors of the Driving Force (and Pro) wheels */
 #define DF_RDESC_ORIG_SIZE	130
 #define DFP_RDESC_ORIG_SIZE	97
+#define FV_RDESC_ORIG_SIZE	130
 #define MOMO_RDESC_ORIG_SIZE	87
+#define MOMO2_RDESC_ORIG_SIZE	87
 
 /* Fixed report descriptors for Logitech Driving Force (and Pro)
  * wheel controllers
@@ -170,6 +172,73 @@ static __u8 dfp_rdesc_fixed[] = {
 0xC0                /*  End Collection                          */
 };
 
+static __u8 fv_rdesc_fixed[] = {
+0x05, 0x01,         /*  Usage Page (Desktop),                   */
+0x09, 0x04,         /*  Usage (Joystik),                        */
+0xA1, 0x01,         /*  Collection (Application),               */
+0xA1, 0x02,         /*      Collection (Logical),               */
+0x95, 0x01,         /*          Report Count (1),               */
+0x75, 0x0A,         /*          Report Size (10),               */
+0x15, 0x00,         /*          Logical Minimum (0),            */
+0x26, 0xFF, 0x03,   /*          Logical Maximum (1023),         */
+0x35, 0x00,         /*          Physical Minimum (0),           */
+0x46, 0xFF, 0x03,   /*          Physical Maximum (1023),        */
+0x09, 0x30,         /*          Usage (X),                      */
+0x81, 0x02,         /*          Input (Variable),               */
+0x95, 0x0C,         /*          Report Count (12),              */
+0x75, 0x01,         /*          Report Size (1),                */
+0x25, 0x01,         /*          Logical Maximum (1),            */
+0x45, 0x01,         /*          Physical Maximum (1),           */
+0x05, 0x09,         /*          Usage Page (Button),            */
+0x19, 0x01,         /*          Usage Minimum (01h),            */
+0x29, 0x0C,         /*          Usage Maximum (0Ch),            */
+0x81, 0x02,         /*          Input (Variable),               */
+0x95, 0x02,         /*          Report Count (2),               */
+0x06, 0x00, 0xFF,   /*          Usage Page (FF00h),             */
+0x09, 0x01,         /*          Usage (01h),                    */
+0x81, 0x02,         /*          Input (Variable),               */
+0x09, 0x02,         /*          Usage (02h),                    */
+0x26, 0xFF, 0x00,   /*          Logical Maximum (255),          */
+0x46, 0xFF, 0x00,   /*          Physical Maximum (255),         */
+0x95, 0x01,         /*          Report Count (1),               */
+0x75, 0x08,         /*          Report Size (8),                */
+0x81, 0x02,         /*          Input (Variable),               */
+0x05, 0x01,         /*          Usage Page (Desktop),           */
+0x25, 0x07,         /*          Logical Maximum (7),            */
+0x46, 0x3B, 0x01,   /*          Physical Maximum (315),         */
+0x75, 0x04,         /*          Report Size (4),                */
+0x65, 0x14,         /*          Unit (Degrees),                 */
+0x09, 0x39,         /*          Usage (Hat Switch),             */
+0x81, 0x42,         /*          Input (Variable, Null State),   */
+0x75, 0x01,         /*          Report Size (1),                */
+0x95, 0x04,         /*          Report Count (4),               */
+0x65, 0x00,         /*          Unit,                           */
+0x06, 0x00, 0xFF,   /*          Usage Page (FF00h),             */
+0x09, 0x01,         /*          Usage (01h),                    */
+0x25, 0x01,         /*          Logical Maximum (1),            */
+0x45, 0x01,         /*          Physical Maximum (1),           */
+0x81, 0x02,         /*          Input (Variable),               */
+0x05, 0x01,         /*          Usage Page (Desktop),           */
+0x95, 0x01,         /*          Report Count (1),               */
+0x75, 0x08,         /*          Report Size (8),                */
+0x26, 0xFF, 0x00,   /*          Logical Maximum (255),          */
+0x46, 0xFF, 0x00,   /*          Physical Maximum (255),         */
+0x09, 0x31,         /*          Usage (Y),                      */
+0x81, 0x02,         /*          Input (Variable),               */
+0x09, 0x32,         /*          Usage (Z),                      */
+0x81, 0x02,         /*          Input (Variable),               */
+0xC0,               /*      End Collection,                     */
+0xA1, 0x02,         /*      Collection (Logical),               */
+0x26, 0xFF, 0x00,   /*          Logical Maximum (255),          */
+0x46, 0xFF, 0x00,   /*          Physical Maximum (255),         */
+0x95, 0x07,         /*          Report Count (7),               */
+0x75, 0x08,         /*          Report Size (8),                */
+0x09, 0x03,         /*          Usage (03h),                    */
+0x91, 0x02,         /*          Output (Variable),              */
+0xC0,               /*      End Collection,                     */
+0xC0                /*  End Collection                          */
+};
+
 static __u8 momo_rdesc_fixed[] = {
 0x05, 0x01,         /*  Usage Page (Desktop),               */
 0x09, 0x04,         /*  Usage (Joystik),                    */
@@ -216,6 +285,54 @@ static __u8 momo_rdesc_fixed[] = {
 0xC0                /*  End Collection                      */
 };
 
+static __u8 momo2_rdesc_fixed[] = {
+0x05, 0x01,         /*  Usage Page (Desktop),               */
+0x09, 0x04,         /*  Usage (Joystik),                    */
+0xA1, 0x01,         /*  Collection (Application),           */
+0xA1, 0x02,         /*      Collection (Logical),           */
+0x95, 0x01,         /*          Report Count (1),           */
+0x75, 0x0A,         /*          Report Size (10),           */
+0x15, 0x00,         /*          Logical Minimum (0),        */
+0x26, 0xFF, 0x03,   /*          Logical Maximum (1023),     */
+0x35, 0x00,         /*          Physical Minimum (0),       */
+0x46, 0xFF, 0x03,   /*          Physical Maximum (1023),    */
+0x09, 0x30,         /*          Usage (X),                  */
+0x81, 0x02,         /*          Input (Variable),           */
+0x95, 0x0A,         /*          Report Count (10),          */
+0x75, 0x01,         /*          Report Size (1),            */
+0x25, 0x01,         /*          Logical Maximum (1),        */
+0x45, 0x01,         /*          Physical Maximum (1),       */
+0x05, 0x09,         /*          Usage Page (Button),        */
+0x19, 0x01,         /*          Usage Minimum (01h),        */
+0x29, 0x0A,         /*          Usage Maximum (0Ah),        */
+0x81, 0x02,         /*          Input (Variable),           */
+0x06, 0x00, 0xFF,   /*          Usage Page (FF00h),         */
+0x09, 0x00,         /*          Usage (00h),                */
+0x95, 0x04,         /*          Report Count (4),           */
+0x81, 0x02,         /*          Input (Variable),           */
+0x95, 0x01,         /*          Report Count (1),           */
+0x75, 0x08,         /*          Report Size (8),            */
+0x26, 0xFF, 0x00,   /*          Logical Maximum (255),      */
+0x46, 0xFF, 0x00,   /*          Physical Maximum (255),     */
+0x09, 0x01,         /*          Usage (01h),                */
+0x81, 0x02,         /*          Input (Variable),           */
+0x05, 0x01,         /*          Usage Page (Desktop),       */
+0x09, 0x31,         /*          Usage (Y),                  */
+0x81, 0x02,         /*          Input (Variable),           */
+0x09, 0x32,         /*          Usage (Z),                  */
+0x81, 0x02,         /*          Input (Variable),           */
+0x06, 0x00, 0xFF,   /*          Usage Page (FF00h),         */
+0x09, 0x00,         /*          Usage (00h),                */
+0x81, 0x02,         /*          Input (Variable),           */
+0xC0,               /*      End Collection,                 */
+0xA1, 0x02,         /*      Collection (Logical),           */
+0x09, 0x02,         /*          Usage (02h),                */
+0x95, 0x07,         /*          Report Count (7),           */
+0x91, 0x02,         /*          Output (Variable),          */
+0xC0,               /*      End Collection,                 */
+0xC0                /*  End Collection                      */
+};
+
 /*
  * Certain Logitech keyboards send in report #3 keys which are far
  * above the logical maximum described in descriptor. This extends
@@ -275,6 +392,24 @@ static __u8 *lg_report_fixup(struct hid_device *hdev, __u8 *rdesc,
 		}
 		break;
 
+	case USB_DEVICE_ID_LOGITECH_MOMO_WHEEL2:
+		if (*rsize == MOMO2_RDESC_ORIG_SIZE) {
+			hid_info(hdev,
+				"fixing up Logitech Momo Racing Force (Black) report descriptor\n");
+			rdesc = momo2_rdesc_fixed;
+			*rsize = sizeof(momo2_rdesc_fixed);
+		}
+		break;
+
+	case USB_DEVICE_ID_LOGITECH_VIBRATION_WHEEL:
+		if (*rsize == FV_RDESC_ORIG_SIZE) {
+			hid_info(hdev,
+				"fixing up Logitech Formula Vibration report descriptor\n");
+			rdesc = fv_rdesc_fixed;
+			*rsize = sizeof(fv_rdesc_fixed);
+		}
+		break;
+
 	case USB_DEVICE_ID_LOGITECH_DFP_WHEEL:
 		if (*rsize == DFP_RDESC_ORIG_SIZE) {
 			hid_info(hdev,
@@ -492,6 +627,7 @@ static int lg_input_mapped(struct hid_device *hdev, struct hid_input *hi,
 		case USB_DEVICE_ID_LOGITECH_G27_WHEEL:
 		case USB_DEVICE_ID_LOGITECH_WII_WHEEL:
 		case USB_DEVICE_ID_LOGITECH_MOMO_WHEEL2:
+		case USB_DEVICE_ID_LOGITECH_VIBRATION_WHEEL:
 			field->application = HID_GD_MULTIAXIS;
 			break;
 		default:
@@ -639,6 +775,8 @@ static const struct hid_device_id lg_devices[] = {
 		.driver_data = LG_NOGET | LG_FF4 },
 	{ HID_USB_DEVICE(USB_VENDOR_ID_LOGITECH, USB_DEVICE_ID_LOGITECH_MOMO_WHEEL2),
 		.driver_data = LG_FF4 },
+	{ HID_USB_DEVICE(USB_VENDOR_ID_LOGITECH, USB_DEVICE_ID_LOGITECH_VIBRATION_WHEEL),
+		.driver_data = LG_FF2 },
 	{ HID_USB_DEVICE(USB_VENDOR_ID_LOGITECH, USB_DEVICE_ID_LOGITECH_G25_WHEEL),
 		.driver_data = LG_FF4 },
 	{ HID_USB_DEVICE(USB_VENDOR_ID_LOGITECH, USB_DEVICE_ID_LOGITECH_DFGT_WHEEL),
diff --git a/drivers/hid/hid-lg2ff.c b/drivers/hid/hid-lg2ff.c
index 1a42eaa6ca02..0e3fb1a7e421 100644
--- a/drivers/hid/hid-lg2ff.c
+++ b/drivers/hid/hid-lg2ff.c
@@ -95,7 +95,7 @@ int lg2ff_init(struct hid_device *hid)
 
 	hid_hw_request(hid, report, HID_REQ_SET_REPORT);
 
-	hid_info(hid, "Force feedback for Logitech RumblePad/Rumblepad 2 by Anssi Hannula <anssi.hannula@gmail.com>\n");
+	hid_info(hid, "Force feedback for Logitech variant 2 rumble devices by Anssi Hannula <anssi.hannula@gmail.com>\n");
 
 	return 0;
 }
diff --git a/drivers/hid/hid-lg4ff.c b/drivers/hid/hid-lg4ff.c
index 8782fe1aaa07..befe0e336471 100644
--- a/drivers/hid/hid-lg4ff.c
+++ b/drivers/hid/hid-lg4ff.c
@@ -196,6 +196,21 @@ static int hid_lg4ff_play(struct input_dev *dev, void *data, struct ff_effect *e
 	case FF_CONSTANT:
 		x = effect->u.ramp.start_level + 0x80;	/* 0x80 is no force */
 		CLAMP(x);
+
+		if (x == 0x80) {
+			/* De-activate force in slot-1*/
+			value[0] = 0x13;
+			value[1] = 0x00;
+			value[2] = 0x00;
+			value[3] = 0x00;
+			value[4] = 0x00;
+			value[5] = 0x00;
+			value[6] = 0x00;
+
+			hid_hw_request(hid, report, HID_REQ_SET_REPORT);
+			return 0;
+		}
+
 		value[0] = 0x11;	/* Slot 1 */
 		value[1] = 0x08;
 		value[2] = x;
@@ -218,12 +233,70 @@ static void hid_lg4ff_set_autocenter_default(struct input_dev *dev, u16 magnitud
 	struct list_head *report_list = &hid->report_enum[HID_OUTPUT_REPORT].report_list;
 	struct hid_report *report = list_entry(report_list->next, struct hid_report, list);
 	__s32 *value = report->field[0]->value;
+	__u32 expand_a, expand_b;
+	struct lg4ff_device_entry *entry;
+	struct lg_drv_data *drv_data;
+
+	drv_data = hid_get_drvdata(hid);
+	if (!drv_data) {
+		hid_err(hid, "Private driver data not found!\n");
+		return;
+	}
+
+	entry = drv_data->device_props;
+	if (!entry) {
+		hid_err(hid, "Device properties not found!\n");
+		return;
+	}
+
+	/* De-activate Auto-Center */
+	if (magnitude == 0) {
+		value[0] = 0xf5;
+		value[1] = 0x00;
+		value[2] = 0x00;
+		value[3] = 0x00;
+		value[4] = 0x00;
+		value[5] = 0x00;
+		value[6] = 0x00;
+
+		hid_hw_request(hid, report, HID_REQ_SET_REPORT);
+		return;
+	}
+
+	if (magnitude <= 0xaaaa) {
+		expand_a = 0x0c * magnitude;
+		expand_b = 0x80 * magnitude;
+	} else {
+		expand_a = (0x0c * 0xaaaa) + 0x06 * (magnitude - 0xaaaa);
+		expand_b = (0x80 * 0xaaaa) + 0xff * (magnitude - 0xaaaa);
+	}
+
+	/* Adjust for non-MOMO wheels */
+	switch (entry->product_id) {
+	case USB_DEVICE_ID_LOGITECH_MOMO_WHEEL:
+	case USB_DEVICE_ID_LOGITECH_MOMO_WHEEL2:
+		break;
+	default:
+		expand_a = expand_a >> 1;
+		break;
+	}
 
 	value[0] = 0xfe;
 	value[1] = 0x0d;
-	value[2] = magnitude >> 13;
-	value[3] = magnitude >> 13;
-	value[4] = magnitude >> 8;
+	value[2] = expand_a / 0xaaaa;
+	value[3] = expand_a / 0xaaaa;
+	value[4] = expand_b / 0xaaaa;
+	value[5] = 0x00;
+	value[6] = 0x00;
+
+	hid_hw_request(hid, report, HID_REQ_SET_REPORT);
+
+	/* Activate Auto-Center */
+	value[0] = 0x14;
+	value[1] = 0x00;
+	value[2] = 0x00;
+	value[3] = 0x00;
+	value[4] = 0x00;
 	value[5] = 0x00;
 	value[6] = 0x00;
 
@@ -540,17 +613,6 @@ int lg4ff_init(struct hid_device *hid)
 	if (error)
 		return error;
 
-	/* Check if autocentering is available and
-	 * set the centering force to zero by default */
-	if (test_bit(FF_AUTOCENTER, dev->ffbit)) {
-		if (rev_maj == FFEX_REV_MAJ && rev_min == FFEX_REV_MIN)	/* Formula Force EX expects different autocentering command */
-			dev->ff->set_autocenter = hid_lg4ff_set_autocenter_ffex;
-		else
-			dev->ff->set_autocenter = hid_lg4ff_set_autocenter_default;
-
-		dev->ff->set_autocenter(dev, 0);
-	}
-
 	/* Get private driver data */
 	drv_data = hid_get_drvdata(hid);
 	if (!drv_data) {
@@ -571,6 +633,17 @@ int lg4ff_init(struct hid_device *hid)
 	entry->max_range = lg4ff_devices[i].max_range;
 	entry->set_range = lg4ff_devices[i].set_range;
 
+	/* Check if autocentering is available and
+	 * set the centering force to zero by default */
+	if (test_bit(FF_AUTOCENTER, dev->ffbit)) {
+		if (rev_maj == FFEX_REV_MAJ && rev_min == FFEX_REV_MIN)	/* Formula Force EX expects different autocentering command */
+			dev->ff->set_autocenter = hid_lg4ff_set_autocenter_ffex;
+		else
+			dev->ff->set_autocenter = hid_lg4ff_set_autocenter_default;
+
+		dev->ff->set_autocenter(dev, 0);
+	}
+
 	/* Create sysfs interface */
 	error = device_create_file(&hid->dev, &dev_attr_range);
 	if (error)
diff --git a/drivers/hid/hid-logitech-dj.c b/drivers/hid/hid-logitech-dj.c
index 2e5302462efb..a7947d8251a8 100644
--- a/drivers/hid/hid-logitech-dj.c
+++ b/drivers/hid/hid-logitech-dj.c
@@ -542,9 +542,9 @@ static int logi_dj_output_hidraw_report(struct hid_device *hid, u8 * buf,
 	return 0;
 }
 
-static void rdcat(char **rdesc, unsigned int *rsize, const char *data, unsigned int size)
+static void rdcat(char *rdesc, unsigned int *rsize, const char *data, unsigned int size)
 {
-	memcpy(*rdesc + *rsize, data, size);
+	memcpy(rdesc + *rsize, data, size);
 	*rsize += size;
 }
 
@@ -567,31 +567,31 @@ static int logi_dj_ll_parse(struct hid_device *hid)
 	if (djdev->reports_supported & STD_KEYBOARD) {
 		dbg_hid("%s: sending a kbd descriptor, reports_supported: %x\n",
 			__func__, djdev->reports_supported);
-		rdcat(&rdesc, &rsize, kbd_descriptor, sizeof(kbd_descriptor));
+		rdcat(rdesc, &rsize, kbd_descriptor, sizeof(kbd_descriptor));
 	}
 
 	if (djdev->reports_supported & STD_MOUSE) {
 		dbg_hid("%s: sending a mouse descriptor, reports_supported: "
 			"%x\n", __func__, djdev->reports_supported);
-		rdcat(&rdesc, &rsize, mse_descriptor, sizeof(mse_descriptor));
+		rdcat(rdesc, &rsize, mse_descriptor, sizeof(mse_descriptor));
 	}
 
 	if (djdev->reports_supported & MULTIMEDIA) {
 		dbg_hid("%s: sending a multimedia report descriptor: %x\n",
 			__func__, djdev->reports_supported);
-		rdcat(&rdesc, &rsize, consumer_descriptor, sizeof(consumer_descriptor));
+		rdcat(rdesc, &rsize, consumer_descriptor, sizeof(consumer_descriptor));
 	}
 
 	if (djdev->reports_supported & POWER_KEYS) {
 		dbg_hid("%s: sending a power keys report descriptor: %x\n",
 			__func__, djdev->reports_supported);
-		rdcat(&rdesc, &rsize, syscontrol_descriptor, sizeof(syscontrol_descriptor));
+		rdcat(rdesc, &rsize, syscontrol_descriptor, sizeof(syscontrol_descriptor));
 	}
 
 	if (djdev->reports_supported & MEDIA_CENTER) {
 		dbg_hid("%s: sending a media center report descriptor: %x\n",
 			__func__, djdev->reports_supported);
-		rdcat(&rdesc, &rsize, media_descriptor, sizeof(media_descriptor));
+		rdcat(rdesc, &rsize, media_descriptor, sizeof(media_descriptor));
 	}
 
 	if (djdev->reports_supported & KBD_LEDS) {
diff --git a/drivers/hid/hid-multitouch.c b/drivers/hid/hid-multitouch.c
index 5e5fe1b8eebb..a2cedb8ae1c0 100644
--- a/drivers/hid/hid-multitouch.c
+++ b/drivers/hid/hid-multitouch.c
@@ -250,12 +250,12 @@ static struct mt_class mt_classes[] = {
 	{ .name	= MT_CLS_GENERALTOUCH_TWOFINGERS,
 		.quirks	= MT_QUIRK_NOT_SEEN_MEANS_UP |
 			MT_QUIRK_VALID_IS_INRANGE |
-			MT_QUIRK_SLOT_IS_CONTACTNUMBER,
+			MT_QUIRK_SLOT_IS_CONTACTID,
 		.maxcontacts = 2
 	},
 	{ .name	= MT_CLS_GENERALTOUCH_PWT_TENFINGERS,
 		.quirks	= MT_QUIRK_NOT_SEEN_MEANS_UP |
-			MT_QUIRK_SLOT_IS_CONTACTNUMBER
+			MT_QUIRK_SLOT_IS_CONTACTID
 	},
 
 	{ .name = MT_CLS_FLATFROG,
@@ -1173,6 +1173,21 @@ static const struct hid_device_id mt_devices[] = {
 	{ .driver_data = MT_CLS_GENERALTOUCH_PWT_TENFINGERS,
 		MT_USB_DEVICE(USB_VENDOR_ID_GENERAL_TOUCH,
 			USB_DEVICE_ID_GENERAL_TOUCH_WIN8_PWT_TENFINGERS) },
+	{ .driver_data = MT_CLS_GENERALTOUCH_TWOFINGERS,
+		MT_USB_DEVICE(USB_VENDOR_ID_GENERAL_TOUCH,
+			USB_DEVICE_ID_GENERAL_TOUCH_WIN8_PIT_0101) },
+	{ .driver_data = MT_CLS_GENERALTOUCH_PWT_TENFINGERS,
+		MT_USB_DEVICE(USB_VENDOR_ID_GENERAL_TOUCH,
+			USB_DEVICE_ID_GENERAL_TOUCH_WIN8_PIT_0102) },
+	{ .driver_data = MT_CLS_GENERALTOUCH_PWT_TENFINGERS,
+		MT_USB_DEVICE(USB_VENDOR_ID_GENERAL_TOUCH,
+			USB_DEVICE_ID_GENERAL_TOUCH_WIN8_PIT_0106) },
+	{ .driver_data = MT_CLS_GENERALTOUCH_PWT_TENFINGERS,
+		MT_USB_DEVICE(USB_VENDOR_ID_GENERAL_TOUCH,
+			USB_DEVICE_ID_GENERAL_TOUCH_WIN8_PIT_010A) },
+	{ .driver_data = MT_CLS_GENERALTOUCH_PWT_TENFINGERS,
+		MT_USB_DEVICE(USB_VENDOR_ID_GENERAL_TOUCH,
+			USB_DEVICE_ID_GENERAL_TOUCH_WIN8_PIT_E100) },
 
 	/* Gametel game controller */
 	{ .driver_data = MT_CLS_NSMU,
@@ -1284,6 +1299,14 @@ static const struct hid_device_id mt_devices[] = {
 		MT_USB_DEVICE(USB_VENDOR_ID_QUANTA,
 			USB_DEVICE_ID_QUANTA_OPTICAL_TOUCH_3008) },
 
+	/* SiS panels */
+	{ .driver_data = MT_CLS_DEFAULT,
+		HID_USB_DEVICE(USB_VENDOR_ID_SIS2_TOUCH,
+		USB_DEVICE_ID_SIS9200_TOUCH) },
+	{ .driver_data = MT_CLS_DEFAULT,
+		HID_USB_DEVICE(USB_VENDOR_ID_SIS2_TOUCH,
+		USB_DEVICE_ID_SIS817_TOUCH) },
+
 	/* Stantum panels */
 	{ .driver_data = MT_CLS_CONFIDENCE,
 		MT_USB_DEVICE(USB_VENDOR_ID_STANTUM,
diff --git a/drivers/hid/hid-roccat-common.c b/drivers/hid/hid-roccat-common.c
index 74f704032627..02e28e9f4ea7 100644
--- a/drivers/hid/hid-roccat-common.c
+++ b/drivers/hid/hid-roccat-common.c
@@ -65,10 +65,11 @@ int roccat_common2_send(struct usb_device *usb_dev, uint report_id,
 EXPORT_SYMBOL_GPL(roccat_common2_send);
 
 enum roccat_common2_control_states {
-	ROCCAT_COMMON_CONTROL_STATUS_OVERLOAD = 0,
+	ROCCAT_COMMON_CONTROL_STATUS_CRITICAL = 0,
 	ROCCAT_COMMON_CONTROL_STATUS_OK = 1,
 	ROCCAT_COMMON_CONTROL_STATUS_INVALID = 2,
-	ROCCAT_COMMON_CONTROL_STATUS_WAIT = 3,
+	ROCCAT_COMMON_CONTROL_STATUS_BUSY = 3,
+	ROCCAT_COMMON_CONTROL_STATUS_CRITICAL_NEW = 4,
 };
 
 static int roccat_common2_receive_control_status(struct usb_device *usb_dev)
@@ -88,13 +89,12 @@ static int roccat_common2_receive_control_status(struct usb_device *usb_dev)
 		switch (control.value) {
 		case ROCCAT_COMMON_CONTROL_STATUS_OK:
 			return 0;
-		case ROCCAT_COMMON_CONTROL_STATUS_WAIT:
+		case ROCCAT_COMMON_CONTROL_STATUS_BUSY:
 			msleep(500);
 			continue;
 		case ROCCAT_COMMON_CONTROL_STATUS_INVALID:
-
-		case ROCCAT_COMMON_CONTROL_STATUS_OVERLOAD:
-			/* seems to be critical - replug necessary */
+		case ROCCAT_COMMON_CONTROL_STATUS_CRITICAL:
+		case ROCCAT_COMMON_CONTROL_STATUS_CRITICAL_NEW:
 			return -EINVAL;
 		default:
 			dev_err(&usb_dev->dev,
@@ -122,6 +122,59 @@ int roccat_common2_send_with_status(struct usb_device *usb_dev,
 }
 EXPORT_SYMBOL_GPL(roccat_common2_send_with_status);
 
+int roccat_common2_device_init_struct(struct usb_device *usb_dev,
+		struct roccat_common2_device *dev)
+{
+	mutex_init(&dev->lock);
+	return 0;
+}
+EXPORT_SYMBOL_GPL(roccat_common2_device_init_struct);
+
+ssize_t roccat_common2_sysfs_read(struct file *fp, struct kobject *kobj,
+		char *buf, loff_t off, size_t count,
+		size_t real_size, uint command)
+{
+	struct device *dev =
+			container_of(kobj, struct device, kobj)->parent->parent;
+	struct roccat_common2_device *roccat_dev = hid_get_drvdata(dev_get_drvdata(dev));
+	struct usb_device *usb_dev = interface_to_usbdev(to_usb_interface(dev));
+	int retval;
+
+	if (off >= real_size)
+		return 0;
+
+	if (off != 0 || count != real_size)
+		return -EINVAL;
+
+	mutex_lock(&roccat_dev->lock);
+	retval = roccat_common2_receive(usb_dev, command, buf, real_size);
+	mutex_unlock(&roccat_dev->lock);
+
+	return retval ? retval : real_size;
+}
+EXPORT_SYMBOL_GPL(roccat_common2_sysfs_read);
+
+ssize_t roccat_common2_sysfs_write(struct file *fp, struct kobject *kobj,
+		void const *buf, loff_t off, size_t count,
+		size_t real_size, uint command)
+{
+	struct device *dev =
+			container_of(kobj, struct device, kobj)->parent->parent;
+	struct roccat_common2_device *roccat_dev = hid_get_drvdata(dev_get_drvdata(dev));
+	struct usb_device *usb_dev = interface_to_usbdev(to_usb_interface(dev));
+	int retval;
+
+	if (off != 0 || count != real_size)
+		return -EINVAL;
+
+	mutex_lock(&roccat_dev->lock);
+	retval = roccat_common2_send_with_status(usb_dev, command, buf, real_size);
+	mutex_unlock(&roccat_dev->lock);
+
+	return retval ? retval : real_size;
+}
+EXPORT_SYMBOL_GPL(roccat_common2_sysfs_write);
+
 MODULE_AUTHOR("Stefan Achatz");
 MODULE_DESCRIPTION("USB Roccat common driver");
 MODULE_LICENSE("GPL v2");
diff --git a/drivers/hid/hid-roccat-common.h b/drivers/hid/hid-roccat-common.h
index a97746a63b70..eaa56eb7d5d1 100644
--- a/drivers/hid/hid-roccat-common.h
+++ b/drivers/hid/hid-roccat-common.h
@@ -32,4 +32,66 @@ int roccat_common2_send(struct usb_device *usb_dev, uint report_id,
 int roccat_common2_send_with_status(struct usb_device *usb_dev,
 		uint command, void const *buf, uint size);
 
+struct roccat_common2_device {
+	int roccat_claimed;
+	int chrdev_minor;
+	struct mutex lock;
+};
+
+int roccat_common2_device_init_struct(struct usb_device *usb_dev,
+		struct roccat_common2_device *dev);
+ssize_t roccat_common2_sysfs_read(struct file *fp, struct kobject *kobj,
+		char *buf, loff_t off, size_t count,
+		size_t real_size, uint command);
+ssize_t roccat_common2_sysfs_write(struct file *fp, struct kobject *kobj,
+		void const *buf, loff_t off, size_t count,
+		size_t real_size, uint command);
+
+#define ROCCAT_COMMON2_SYSFS_W(thingy, COMMAND, SIZE) \
+static ssize_t roccat_common2_sysfs_write_ ## thingy(struct file *fp, \
+		struct kobject *kobj, struct bin_attribute *attr, char *buf, \
+		loff_t off, size_t count) \
+{ \
+	return roccat_common2_sysfs_write(fp, kobj, buf, off, count, \
+			SIZE, COMMAND); \
+}
+
+#define ROCCAT_COMMON2_SYSFS_R(thingy, COMMAND, SIZE) \
+static ssize_t roccat_common2_sysfs_read_ ## thingy(struct file *fp, \
+		struct kobject *kobj, struct bin_attribute *attr, char *buf, \
+		loff_t off, size_t count) \
+{ \
+	return roccat_common2_sysfs_read(fp, kobj, buf, off, count, \
+			SIZE, COMMAND); \
+}
+
+#define ROCCAT_COMMON2_SYSFS_RW(thingy, COMMAND, SIZE) \
+ROCCAT_COMMON2_SYSFS_W(thingy, COMMAND, SIZE) \
+ROCCAT_COMMON2_SYSFS_R(thingy, COMMAND, SIZE)
+
+#define ROCCAT_COMMON2_BIN_ATTRIBUTE_RW(thingy, COMMAND, SIZE) \
+ROCCAT_COMMON2_SYSFS_RW(thingy, COMMAND, SIZE); \
+static struct bin_attribute bin_attr_ ## thingy = { \
+	.attr = { .name = #thingy, .mode = 0660 }, \
+	.size = SIZE, \
+	.read = roccat_common2_sysfs_read_ ## thingy, \
+	.write = roccat_common2_sysfs_write_ ## thingy \
+}
+
+#define ROCCAT_COMMON2_BIN_ATTRIBUTE_R(thingy, COMMAND, SIZE) \
+ROCCAT_COMMON2_SYSFS_R(thingy, COMMAND, SIZE); \
+static struct bin_attribute bin_attr_ ## thingy = { \
+	.attr = { .name = #thingy, .mode = 0440 }, \
+	.size = SIZE, \
+	.read = roccat_common2_sysfs_read_ ## thingy, \
+}
+
+#define ROCCAT_COMMON2_BIN_ATTRIBUTE_W(thingy, COMMAND, SIZE) \
+ROCCAT_COMMON2_SYSFS_W(thingy, COMMAND, SIZE); \
+static struct bin_attribute bin_attr_ ## thingy = { \
+	.attr = { .name = #thingy, .mode = 0220 }, \
+	.size = SIZE, \
+	.write = roccat_common2_sysfs_write_ ## thingy \
+}
+
 #endif
diff --git a/drivers/hid/hid-roccat-konepure.c b/drivers/hid/hid-roccat-konepure.c
index 99a605ebb665..07de2f9014c6 100644
--- a/drivers/hid/hid-roccat-konepure.c
+++ b/drivers/hid/hid-roccat-konepure.c
@@ -15,6 +15,7 @@
  * Roccat KonePure is a smaller version of KoneXTD with less buttons and lights.
  */
 
+#include <linux/types.h>
 #include <linux/device.h>
 #include <linux/input.h>
 #include <linux/hid.h>
@@ -23,128 +24,50 @@
 #include <linux/hid-roccat.h>
 #include "hid-ids.h"
 #include "hid-roccat-common.h"
-#include "hid-roccat-konepure.h"
 
-static struct class *konepure_class;
-
-static ssize_t konepure_sysfs_read(struct file *fp, struct kobject *kobj,
-		char *buf, loff_t off, size_t count,
-		size_t real_size, uint command)
-{
-	struct device *dev =
-			container_of(kobj, struct device, kobj)->parent->parent;
-	struct konepure_device *konepure = hid_get_drvdata(dev_get_drvdata(dev));
-	struct usb_device *usb_dev = interface_to_usbdev(to_usb_interface(dev));
-	int retval;
-
-	if (off >= real_size)
-		return 0;
-
-	if (off != 0 || count != real_size)
-		return -EINVAL;
-
-	mutex_lock(&konepure->konepure_lock);
-	retval = roccat_common2_receive(usb_dev, command, buf, real_size);
-	mutex_unlock(&konepure->konepure_lock);
-
-	return retval ? retval : real_size;
-}
-
-static ssize_t konepure_sysfs_write(struct file *fp, struct kobject *kobj,
-		void const *buf, loff_t off, size_t count,
-		size_t real_size, uint command)
-{
-	struct device *dev =
-			container_of(kobj, struct device, kobj)->parent->parent;
-	struct konepure_device *konepure = hid_get_drvdata(dev_get_drvdata(dev));
-	struct usb_device *usb_dev = interface_to_usbdev(to_usb_interface(dev));
-	int retval;
-
-	if (off != 0 || count != real_size)
-		return -EINVAL;
-
-	mutex_lock(&konepure->konepure_lock);
-	retval = roccat_common2_send_with_status(usb_dev, command,
-			(void *)buf, real_size);
-	mutex_unlock(&konepure->konepure_lock);
-
-	return retval ? retval : real_size;
-}
-
-#define KONEPURE_SYSFS_W(thingy, THINGY) \
-static ssize_t konepure_sysfs_write_ ## thingy(struct file *fp, \
-		struct kobject *kobj, struct bin_attribute *attr, char *buf, \
-		loff_t off, size_t count) \
-{ \
-	return konepure_sysfs_write(fp, kobj, buf, off, count, \
-			KONEPURE_SIZE_ ## THINGY, KONEPURE_COMMAND_ ## THINGY); \
-}
-
-#define KONEPURE_SYSFS_R(thingy, THINGY) \
-static ssize_t konepure_sysfs_read_ ## thingy(struct file *fp, \
-		struct kobject *kobj, struct bin_attribute *attr, char *buf, \
-		loff_t off, size_t count) \
-{ \
-	return konepure_sysfs_read(fp, kobj, buf, off, count, \
-			KONEPURE_SIZE_ ## THINGY, KONEPURE_COMMAND_ ## THINGY); \
-}
+enum {
+	KONEPURE_MOUSE_REPORT_NUMBER_BUTTON = 3,
+};
 
-#define KONEPURE_SYSFS_RW(thingy, THINGY) \
-KONEPURE_SYSFS_W(thingy, THINGY) \
-KONEPURE_SYSFS_R(thingy, THINGY)
-
-#define KONEPURE_BIN_ATTRIBUTE_RW(thingy, THINGY) \
-KONEPURE_SYSFS_RW(thingy, THINGY); \
-static struct bin_attribute bin_attr_##thingy = { \
-	.attr = { .name = #thingy, .mode = 0660 }, \
-	.size = KONEPURE_SIZE_ ## THINGY, \
-	.read = konepure_sysfs_read_ ## thingy, \
-	.write = konepure_sysfs_write_ ## thingy \
-}
+struct konepure_mouse_report_button {
+	uint8_t report_number; /* always KONEPURE_MOUSE_REPORT_NUMBER_BUTTON */
+	uint8_t zero;
+	uint8_t type;
+	uint8_t data1;
+	uint8_t data2;
+	uint8_t zero2;
+	uint8_t unknown[2];
+} __packed;
 
-#define KONEPURE_BIN_ATTRIBUTE_R(thingy, THINGY) \
-KONEPURE_SYSFS_R(thingy, THINGY); \
-static struct bin_attribute bin_attr_##thingy = { \
-	.attr = { .name = #thingy, .mode = 0440 }, \
-	.size = KONEPURE_SIZE_ ## THINGY, \
-	.read = konepure_sysfs_read_ ## thingy, \
-}
-
-#define KONEPURE_BIN_ATTRIBUTE_W(thingy, THINGY) \
-KONEPURE_SYSFS_W(thingy, THINGY); \
-static struct bin_attribute bin_attr_##thingy = { \
-	.attr = { .name = #thingy, .mode = 0220 }, \
-	.size = KONEPURE_SIZE_ ## THINGY, \
-	.write = konepure_sysfs_write_ ## thingy \
-}
+static struct class *konepure_class;
 
-KONEPURE_BIN_ATTRIBUTE_RW(actual_profile, ACTUAL_PROFILE);
-KONEPURE_BIN_ATTRIBUTE_RW(info, INFO);
-KONEPURE_BIN_ATTRIBUTE_RW(sensor, SENSOR);
-KONEPURE_BIN_ATTRIBUTE_RW(tcu, TCU);
-KONEPURE_BIN_ATTRIBUTE_RW(profile_settings, PROFILE_SETTINGS);
-KONEPURE_BIN_ATTRIBUTE_RW(profile_buttons, PROFILE_BUTTONS);
-KONEPURE_BIN_ATTRIBUTE_W(control, CONTROL);
-KONEPURE_BIN_ATTRIBUTE_W(talk, TALK);
-KONEPURE_BIN_ATTRIBUTE_W(macro, MACRO);
-KONEPURE_BIN_ATTRIBUTE_R(tcu_image, TCU_IMAGE);
-
-static struct bin_attribute *konepure_bin_attributes[] = {
+ROCCAT_COMMON2_BIN_ATTRIBUTE_W(control, 0x04, 0x03);
+ROCCAT_COMMON2_BIN_ATTRIBUTE_RW(actual_profile, 0x05, 0x03);
+ROCCAT_COMMON2_BIN_ATTRIBUTE_RW(profile_settings, 0x06, 0x1f);
+ROCCAT_COMMON2_BIN_ATTRIBUTE_RW(profile_buttons, 0x07, 0x3b);
+ROCCAT_COMMON2_BIN_ATTRIBUTE_W(macro, 0x08, 0x0822);
+ROCCAT_COMMON2_BIN_ATTRIBUTE_RW(info, 0x09, 0x06);
+ROCCAT_COMMON2_BIN_ATTRIBUTE_RW(tcu, 0x0c, 0x04);
+ROCCAT_COMMON2_BIN_ATTRIBUTE_R(tcu_image, 0x0c, 0x0404);
+ROCCAT_COMMON2_BIN_ATTRIBUTE_RW(sensor, 0x0f, 0x06);
+ROCCAT_COMMON2_BIN_ATTRIBUTE_W(talk, 0x10, 0x10);
+
+static struct bin_attribute *konepure_bin_attrs[] = {
 	&bin_attr_actual_profile,
+	&bin_attr_control,
 	&bin_attr_info,
+	&bin_attr_talk,
+	&bin_attr_macro,
 	&bin_attr_sensor,
 	&bin_attr_tcu,
+	&bin_attr_tcu_image,
 	&bin_attr_profile_settings,
 	&bin_attr_profile_buttons,
-	&bin_attr_control,
-	&bin_attr_talk,
-	&bin_attr_macro,
-	&bin_attr_tcu_image,
 	NULL,
 };
 
 static const struct attribute_group konepure_group = {
-	.bin_attrs = konepure_bin_attributes,
+	.bin_attrs = konepure_bin_attrs,
 };
 
 static const struct attribute_group *konepure_groups[] = {
@@ -152,20 +75,11 @@ static const struct attribute_group *konepure_groups[] = {
 	NULL,
 };
 
-
-static int konepure_init_konepure_device_struct(struct usb_device *usb_dev,
-		struct konepure_device *konepure)
-{
-	mutex_init(&konepure->konepure_lock);
-
-	return 0;
-}
-
 static int konepure_init_specials(struct hid_device *hdev)
 {
 	struct usb_interface *intf = to_usb_interface(hdev->dev.parent);
 	struct usb_device *usb_dev = interface_to_usbdev(intf);
-	struct konepure_device *konepure;
+	struct roccat_common2_device *konepure;
 	int retval;
 
 	if (intf->cur_altsetting->desc.bInterfaceProtocol
@@ -181,9 +95,9 @@ static int konepure_init_specials(struct hid_device *hdev)
 	}
 	hid_set_drvdata(hdev, konepure);
 
-	retval = konepure_init_konepure_device_struct(usb_dev, konepure);
+	retval = roccat_common2_device_init_struct(usb_dev, konepure);
 	if (retval) {
-		hid_err(hdev, "couldn't init struct konepure_device\n");
+		hid_err(hdev, "couldn't init KonePure device\n");
 		goto exit_free;
 	}
 
@@ -205,7 +119,7 @@ exit_free:
 static void konepure_remove_specials(struct hid_device *hdev)
 {
 	struct usb_interface *intf = to_usb_interface(hdev->dev.parent);
-	struct konepure_device *konepure;
+	struct roccat_common2_device *konepure;
 
 	if (intf->cur_altsetting->desc.bInterfaceProtocol
 			!= USB_INTERFACE_PROTOCOL_MOUSE)
@@ -258,7 +172,7 @@ static int konepure_raw_event(struct hid_device *hdev,
 		struct hid_report *report, u8 *data, int size)
 {
 	struct usb_interface *intf = to_usb_interface(hdev->dev.parent);
-	struct konepure_device *konepure = hid_get_drvdata(hdev);
+	struct roccat_common2_device *konepure = hid_get_drvdata(hdev);
 
 	if (intf->cur_altsetting->desc.bInterfaceProtocol
 			!= USB_INTERFACE_PROTOCOL_MOUSE)
diff --git a/drivers/hid/hid-roccat-konepure.h b/drivers/hid/hid-roccat-konepure.h
deleted file mode 100644
index 2cd24e93dfd6..000000000000
--- a/drivers/hid/hid-roccat-konepure.h
+++ /dev/null
@@ -1,72 +0,0 @@
-#ifndef __HID_ROCCAT_KONEPURE_H
-#define __HID_ROCCAT_KONEPURE_H
-
-/*
- * Copyright (c) 2012 Stefan Achatz <erazor_de@users.sourceforge.net>
- */
-
-/*
- * This program is free software; you can redistribute it and/or modify it
- * under the terms of the GNU General Public License as published by the Free
- * Software Foundation; either version 2 of the License, or (at your option)
- * any later version.
- */
-
-#include <linux/types.h>
-
-enum {
-	KONEPURE_SIZE_ACTUAL_PROFILE = 0x03,
-	KONEPURE_SIZE_CONTROL = 0x03,
-	KONEPURE_SIZE_FIRMWARE_WRITE = 0x0402,
-	KONEPURE_SIZE_INFO = 0x06,
-	KONEPURE_SIZE_MACRO = 0x0822,
-	KONEPURE_SIZE_PROFILE_SETTINGS = 0x1f,
-	KONEPURE_SIZE_PROFILE_BUTTONS = 0x3b,
-	KONEPURE_SIZE_SENSOR = 0x06,
-	KONEPURE_SIZE_TALK = 0x10,
-	KONEPURE_SIZE_TCU = 0x04,
-	KONEPURE_SIZE_TCU_IMAGE = 0x0404,
-};
-
-enum konepure_control_requests {
-	KONEPURE_CONTROL_REQUEST_GENERAL = 0x80,
-	KONEPURE_CONTROL_REQUEST_BUTTONS = 0x90,
-};
-
-enum konepure_commands {
-	KONEPURE_COMMAND_CONTROL = 0x04,
-	KONEPURE_COMMAND_ACTUAL_PROFILE = 0x05,
-	KONEPURE_COMMAND_PROFILE_SETTINGS = 0x06,
-	KONEPURE_COMMAND_PROFILE_BUTTONS = 0x07,
-	KONEPURE_COMMAND_MACRO = 0x08,
-	KONEPURE_COMMAND_INFO = 0x09,
-	KONEPURE_COMMAND_TCU = 0x0c,
-	KONEPURE_COMMAND_TCU_IMAGE = 0x0c,
-	KONEPURE_COMMAND_E = 0x0e,
-	KONEPURE_COMMAND_SENSOR = 0x0f,
-	KONEPURE_COMMAND_TALK = 0x10,
-	KONEPURE_COMMAND_FIRMWARE_WRITE = 0x1b,
-	KONEPURE_COMMAND_FIRMWARE_WRITE_CONTROL = 0x1c,
-};
-
-enum {
-	KONEPURE_MOUSE_REPORT_NUMBER_BUTTON = 3,
-};
-
-struct konepure_mouse_report_button {
-	uint8_t report_number; /* always KONEPURE_MOUSE_REPORT_NUMBER_BUTTON */
-	uint8_t zero;
-	uint8_t type;
-	uint8_t data1;
-	uint8_t data2;
-	uint8_t zero2;
-	uint8_t unknown[2];
-} __packed;
-
-struct konepure_device {
-	int roccat_claimed;
-	int chrdev_minor;
-	struct mutex konepure_lock;
-};
-
-#endif
diff --git a/drivers/hid/hid-roccat-kovaplus.c b/drivers/hid/hid-roccat-kovaplus.c
index 0c8e1ef0b67d..966047711fbf 100644
--- a/drivers/hid/hid-roccat-kovaplus.c
+++ b/drivers/hid/hid-roccat-kovaplus.c
@@ -554,9 +554,13 @@ static void kovaplus_keep_values_up_to_date(struct kovaplus_device *kovaplus,
 		break;
 	case KOVAPLUS_MOUSE_REPORT_BUTTON_TYPE_CPI:
 		kovaplus->actual_cpi = kovaplus_convert_event_cpi(button_report->data1);
+		break;
 	case KOVAPLUS_MOUSE_REPORT_BUTTON_TYPE_SENSITIVITY:
 		kovaplus->actual_x_sensitivity = button_report->data1;
 		kovaplus->actual_y_sensitivity = button_report->data2;
+		break;
+	default:
+		break;
 	}
 }
 
diff --git a/drivers/hid/hid-roccat-ryos.c b/drivers/hid/hid-roccat-ryos.c
new file mode 100644
index 000000000000..47cc8f30ff6d
--- /dev/null
+++ b/drivers/hid/hid-roccat-ryos.c
@@ -0,0 +1,241 @@
+/*
+ * Roccat Ryos driver for Linux
+ *
+ * Copyright (c) 2013 Stefan Achatz <erazor_de@users.sourceforge.net>
+ */
+
+/*
+ * This program is free software; you can redistribute it and/or modify it
+ * under the terms of the GNU General Public License as published by the Free
+ * Software Foundation; either version 2 of the License, or (at your option)
+ * any later version.
+ */
+
+#include <linux/types.h>
+#include <linux/device.h>
+#include <linux/input.h>
+#include <linux/hid.h>
+#include <linux/module.h>
+#include <linux/slab.h>
+#include <linux/hid-roccat.h>
+#include "hid-ids.h"
+#include "hid-roccat-common.h"
+
+enum {
+	RYOS_REPORT_NUMBER_SPECIAL = 3,
+	RYOS_USB_INTERFACE_PROTOCOL = 0,
+};
+
+struct ryos_report_special {
+	uint8_t number; /* RYOS_REPORT_NUMBER_SPECIAL */
+	uint8_t data[4];
+} __packed;
+
+static struct class *ryos_class;
+
+ROCCAT_COMMON2_BIN_ATTRIBUTE_W(control, 0x04, 0x03);
+ROCCAT_COMMON2_BIN_ATTRIBUTE_RW(profile, 0x05, 0x03);
+ROCCAT_COMMON2_BIN_ATTRIBUTE_RW(keys_primary, 0x06, 0x7d);
+ROCCAT_COMMON2_BIN_ATTRIBUTE_RW(keys_function, 0x07, 0x5f);
+ROCCAT_COMMON2_BIN_ATTRIBUTE_RW(keys_macro, 0x08, 0x23);
+ROCCAT_COMMON2_BIN_ATTRIBUTE_RW(keys_thumbster, 0x09, 0x17);
+ROCCAT_COMMON2_BIN_ATTRIBUTE_RW(keys_extra, 0x0a, 0x08);
+ROCCAT_COMMON2_BIN_ATTRIBUTE_RW(keys_easyzone, 0x0b, 0x126);
+ROCCAT_COMMON2_BIN_ATTRIBUTE_RW(key_mask, 0x0c, 0x06);
+ROCCAT_COMMON2_BIN_ATTRIBUTE_RW(light, 0x0d, 0x10);
+ROCCAT_COMMON2_BIN_ATTRIBUTE_RW(macro, 0x0e, 0x7d2);
+ROCCAT_COMMON2_BIN_ATTRIBUTE_R(info, 0x0f, 0x08);
+ROCCAT_COMMON2_BIN_ATTRIBUTE_W(reset, 0x11, 0x03);
+ROCCAT_COMMON2_BIN_ATTRIBUTE_W(light_control, 0x13, 0x08);
+ROCCAT_COMMON2_BIN_ATTRIBUTE_W(talk, 0x16, 0x10);
+ROCCAT_COMMON2_BIN_ATTRIBUTE_RW(stored_lights, 0x17, 0x0566);
+ROCCAT_COMMON2_BIN_ATTRIBUTE_W(custom_lights, 0x18, 0x14);
+ROCCAT_COMMON2_BIN_ATTRIBUTE_RW(light_macro, 0x19, 0x07d2);
+
+static struct bin_attribute *ryos_bin_attrs[] = {
+	&bin_attr_control,
+	&bin_attr_profile,
+	&bin_attr_keys_primary,
+	&bin_attr_keys_function,
+	&bin_attr_keys_macro,
+	&bin_attr_keys_thumbster,
+	&bin_attr_keys_extra,
+	&bin_attr_keys_easyzone,
+	&bin_attr_key_mask,
+	&bin_attr_light,
+	&bin_attr_macro,
+	&bin_attr_info,
+	&bin_attr_reset,
+	&bin_attr_light_control,
+	&bin_attr_talk,
+	&bin_attr_stored_lights,
+	&bin_attr_custom_lights,
+	&bin_attr_light_macro,
+	NULL,
+};
+
+static const struct attribute_group ryos_group = {
+	.bin_attrs = ryos_bin_attrs,
+};
+
+static const struct attribute_group *ryos_groups[] = {
+	&ryos_group,
+	NULL,
+};
+
+static int ryos_init_specials(struct hid_device *hdev)
+{
+	struct usb_interface *intf = to_usb_interface(hdev->dev.parent);
+	struct usb_device *usb_dev = interface_to_usbdev(intf);
+	struct roccat_common2_device *ryos;
+	int retval;
+
+	if (intf->cur_altsetting->desc.bInterfaceProtocol
+			!= RYOS_USB_INTERFACE_PROTOCOL) {
+		hid_set_drvdata(hdev, NULL);
+		return 0;
+	}
+
+	ryos = kzalloc(sizeof(*ryos), GFP_KERNEL);
+	if (!ryos) {
+		hid_err(hdev, "can't alloc device descriptor\n");
+		return -ENOMEM;
+	}
+	hid_set_drvdata(hdev, ryos);
+
+	retval = roccat_common2_device_init_struct(usb_dev, ryos);
+	if (retval) {
+		hid_err(hdev, "couldn't init Ryos device\n");
+		goto exit_free;
+	}
+
+	retval = roccat_connect(ryos_class, hdev,
+			sizeof(struct ryos_report_special));
+	if (retval < 0) {
+		hid_err(hdev, "couldn't init char dev\n");
+	} else {
+		ryos->chrdev_minor = retval;
+		ryos->roccat_claimed = 1;
+	}
+
+	return 0;
+exit_free:
+	kfree(ryos);
+	return retval;
+}
+
+static void ryos_remove_specials(struct hid_device *hdev)
+{
+	struct usb_interface *intf = to_usb_interface(hdev->dev.parent);
+	struct roccat_common2_device *ryos;
+
+	if (intf->cur_altsetting->desc.bInterfaceProtocol
+			!= RYOS_USB_INTERFACE_PROTOCOL)
+		return;
+
+	ryos = hid_get_drvdata(hdev);
+	if (ryos->roccat_claimed)
+		roccat_disconnect(ryos->chrdev_minor);
+	kfree(ryos);
+}
+
+static int ryos_probe(struct hid_device *hdev,
+		const struct hid_device_id *id)
+{
+	int retval;
+
+	retval = hid_parse(hdev);
+	if (retval) {
+		hid_err(hdev, "parse failed\n");
+		goto exit;
+	}
+
+	retval = hid_hw_start(hdev, HID_CONNECT_DEFAULT);
+	if (retval) {
+		hid_err(hdev, "hw start failed\n");
+		goto exit;
+	}
+
+	retval = ryos_init_specials(hdev);
+	if (retval) {
+		hid_err(hdev, "couldn't install mouse\n");
+		goto exit_stop;
+	}
+
+	return 0;
+
+exit_stop:
+	hid_hw_stop(hdev);
+exit:
+	return retval;
+}
+
+static void ryos_remove(struct hid_device *hdev)
+{
+	ryos_remove_specials(hdev);
+	hid_hw_stop(hdev);
+}
+
+static int ryos_raw_event(struct hid_device *hdev,
+		struct hid_report *report, u8 *data, int size)
+{
+	struct usb_interface *intf = to_usb_interface(hdev->dev.parent);
+	struct roccat_common2_device *ryos = hid_get_drvdata(hdev);
+
+	if (intf->cur_altsetting->desc.bInterfaceProtocol
+			!= RYOS_USB_INTERFACE_PROTOCOL)
+		return 0;
+
+	if (data[0] != RYOS_REPORT_NUMBER_SPECIAL)
+		return 0;
+
+	if (ryos != NULL && ryos->roccat_claimed)
+		roccat_report_event(ryos->chrdev_minor, data);
+
+	return 0;
+}
+
+static const struct hid_device_id ryos_devices[] = {
+	{ HID_USB_DEVICE(USB_VENDOR_ID_ROCCAT, USB_DEVICE_ID_ROCCAT_RYOS_MK) },
+	{ HID_USB_DEVICE(USB_VENDOR_ID_ROCCAT, USB_DEVICE_ID_ROCCAT_RYOS_MK_GLOW) },
+	{ HID_USB_DEVICE(USB_VENDOR_ID_ROCCAT, USB_DEVICE_ID_ROCCAT_RYOS_MK_PRO) },
+	{ }
+};
+
+MODULE_DEVICE_TABLE(hid, ryos_devices);
+
+static struct hid_driver ryos_driver = {
+		.name = "ryos",
+		.id_table = ryos_devices,
+		.probe = ryos_probe,
+		.remove = ryos_remove,
+		.raw_event = ryos_raw_event
+};
+
+static int __init ryos_init(void)
+{
+	int retval;
+
+	ryos_class = class_create(THIS_MODULE, "ryos");
+	if (IS_ERR(ryos_class))
+		return PTR_ERR(ryos_class);
+	ryos_class->dev_groups = ryos_groups;
+
+	retval = hid_register_driver(&ryos_driver);
+	if (retval)
+		class_destroy(ryos_class);
+	return retval;
+}
+
+static void __exit ryos_exit(void)
+{
+	hid_unregister_driver(&ryos_driver);
+	class_destroy(ryos_class);
+}
+
+module_init(ryos_init);
+module_exit(ryos_exit);
+
+MODULE_AUTHOR("Stefan Achatz");
+MODULE_DESCRIPTION("USB Roccat Ryos MK/Glow/Pro driver");
+MODULE_LICENSE("GPL v2");
diff --git a/drivers/hid/hid-roccat-savu.c b/drivers/hid/hid-roccat-savu.c
index 0332267199d5..6dbf6e04dce7 100644
--- a/drivers/hid/hid-roccat-savu.c
+++ b/drivers/hid/hid-roccat-savu.c
@@ -27,98 +27,15 @@
 
 static struct class *savu_class;
 
-static ssize_t savu_sysfs_read(struct file *fp, struct kobject *kobj,
-		char *buf, loff_t off, size_t count,
-		size_t real_size, uint command)
-{
-	struct device *dev =
-			container_of(kobj, struct device, kobj)->parent->parent;
-	struct savu_device *savu = hid_get_drvdata(dev_get_drvdata(dev));
-	struct usb_device *usb_dev = interface_to_usbdev(to_usb_interface(dev));
-	int retval;
-
-	if (off >= real_size)
-		return 0;
-
-	if (off != 0 || count != real_size)
-		return -EINVAL;
-
-	mutex_lock(&savu->savu_lock);
-	retval = roccat_common2_receive(usb_dev, command, buf, real_size);
-	mutex_unlock(&savu->savu_lock);
-
-	return retval ? retval : real_size;
-}
-
-static ssize_t savu_sysfs_write(struct file *fp, struct kobject *kobj,
-		void const *buf, loff_t off, size_t count,
-		size_t real_size, uint command)
-{
-	struct device *dev =
-			container_of(kobj, struct device, kobj)->parent->parent;
-	struct savu_device *savu = hid_get_drvdata(dev_get_drvdata(dev));
-	struct usb_device *usb_dev = interface_to_usbdev(to_usb_interface(dev));
-	int retval;
-
-	if (off != 0 || count != real_size)
-		return -EINVAL;
-
-	mutex_lock(&savu->savu_lock);
-	retval = roccat_common2_send_with_status(usb_dev, command,
-			(void *)buf, real_size);
-	mutex_unlock(&savu->savu_lock);
-
-	return retval ? retval : real_size;
-}
-
-#define SAVU_SYSFS_W(thingy, THINGY) \
-static ssize_t savu_sysfs_write_ ## thingy(struct file *fp, \
-		struct kobject *kobj, struct bin_attribute *attr, char *buf, \
-		loff_t off, size_t count) \
-{ \
-	return savu_sysfs_write(fp, kobj, buf, off, count, \
-			SAVU_SIZE_ ## THINGY, SAVU_COMMAND_ ## THINGY); \
-}
-
-#define SAVU_SYSFS_R(thingy, THINGY) \
-static ssize_t savu_sysfs_read_ ## thingy(struct file *fp, \
-		struct kobject *kobj, struct bin_attribute *attr, char *buf, \
-		loff_t off, size_t count) \
-{ \
-	return savu_sysfs_read(fp, kobj, buf, off, count, \
-			SAVU_SIZE_ ## THINGY, SAVU_COMMAND_ ## THINGY); \
-}
-
-#define SAVU_SYSFS_RW(thingy, THINGY) \
-SAVU_SYSFS_W(thingy, THINGY) \
-SAVU_SYSFS_R(thingy, THINGY)
-
-#define SAVU_BIN_ATTRIBUTE_RW(thingy, THINGY) \
-SAVU_SYSFS_RW(thingy, THINGY); \
-static struct bin_attribute bin_attr_##thingy = { \
-	.attr = { .name = #thingy, .mode = 0660 }, \
-	.size = SAVU_SIZE_ ## THINGY, \
-	.read = savu_sysfs_read_ ## thingy, \
-	.write = savu_sysfs_write_ ## thingy \
-}
-
-#define SAVU_BIN_ATTRIBUTE_W(thingy, THINGY) \
-SAVU_SYSFS_W(thingy, THINGY); \
-static struct bin_attribute bin_attr_##thingy = { \
-	.attr = { .name = #thingy, .mode = 0220 }, \
-	.size = SAVU_SIZE_ ## THINGY, \
-	.write = savu_sysfs_write_ ## thingy \
-}
-
-SAVU_BIN_ATTRIBUTE_W(control, CONTROL);
-SAVU_BIN_ATTRIBUTE_RW(profile, PROFILE);
-SAVU_BIN_ATTRIBUTE_RW(general, GENERAL);
-SAVU_BIN_ATTRIBUTE_RW(buttons, BUTTONS);
-SAVU_BIN_ATTRIBUTE_RW(macro, MACRO);
-SAVU_BIN_ATTRIBUTE_RW(info, INFO);
-SAVU_BIN_ATTRIBUTE_RW(sensor, SENSOR);
-
-static struct bin_attribute *savu_bin_attributes[] = {
+ROCCAT_COMMON2_BIN_ATTRIBUTE_W(control, 0x4, 0x03);
+ROCCAT_COMMON2_BIN_ATTRIBUTE_RW(profile, 0x5, 0x03);
+ROCCAT_COMMON2_BIN_ATTRIBUTE_RW(general, 0x6, 0x10);
+ROCCAT_COMMON2_BIN_ATTRIBUTE_RW(buttons, 0x7, 0x2f);
+ROCCAT_COMMON2_BIN_ATTRIBUTE_RW(macro, 0x8, 0x0823);
+ROCCAT_COMMON2_BIN_ATTRIBUTE_RW(info, 0x9, 0x08);
+ROCCAT_COMMON2_BIN_ATTRIBUTE_RW(sensor, 0xc, 0x04);
+
+static struct bin_attribute *savu_bin_attrs[] = {
 	&bin_attr_control,
 	&bin_attr_profile,
 	&bin_attr_general,
@@ -130,7 +47,7 @@ static struct bin_attribute *savu_bin_attributes[] = {
 };
 
 static const struct attribute_group savu_group = {
-	.bin_attrs = savu_bin_attributes,
+	.bin_attrs = savu_bin_attrs,
 };
 
 static const struct attribute_group *savu_groups[] = {
@@ -138,19 +55,11 @@ static const struct attribute_group *savu_groups[] = {
 	NULL,
 };
 
-static int savu_init_savu_device_struct(struct usb_device *usb_dev,
-		struct savu_device *savu)
-{
-	mutex_init(&savu->savu_lock);
-
-	return 0;
-}
-
 static int savu_init_specials(struct hid_device *hdev)
 {
 	struct usb_interface *intf = to_usb_interface(hdev->dev.parent);
 	struct usb_device *usb_dev = interface_to_usbdev(intf);
-	struct savu_device *savu;
+	struct roccat_common2_device *savu;
 	int retval;
 
 	if (intf->cur_altsetting->desc.bInterfaceProtocol
@@ -166,9 +75,9 @@ static int savu_init_specials(struct hid_device *hdev)
 	}
 	hid_set_drvdata(hdev, savu);
 
-	retval = savu_init_savu_device_struct(usb_dev, savu);
+	retval = roccat_common2_device_init_struct(usb_dev, savu);
 	if (retval) {
-		hid_err(hdev, "couldn't init struct savu_device\n");
+		hid_err(hdev, "couldn't init Savu device\n");
 		goto exit_free;
 	}
 
@@ -190,7 +99,7 @@ exit_free:
 static void savu_remove_specials(struct hid_device *hdev)
 {
 	struct usb_interface *intf = to_usb_interface(hdev->dev.parent);
-	struct savu_device *savu;
+	struct roccat_common2_device *savu;
 
 	if (intf->cur_altsetting->desc.bInterfaceProtocol
 			!= USB_INTERFACE_PROTOCOL_MOUSE)
@@ -239,7 +148,7 @@ static void savu_remove(struct hid_device *hdev)
 	hid_hw_stop(hdev);
 }
 
-static void savu_report_to_chrdev(struct savu_device const *savu,
+static void savu_report_to_chrdev(struct roccat_common2_device const *savu,
 		u8 const *data)
 {
 	struct savu_roccat_report roccat_report;
@@ -261,7 +170,7 @@ static int savu_raw_event(struct hid_device *hdev,
 		struct hid_report *report, u8 *data, int size)
 {
 	struct usb_interface *intf = to_usb_interface(hdev->dev.parent);
-	struct savu_device *savu = hid_get_drvdata(hdev);
+	struct roccat_common2_device *savu = hid_get_drvdata(hdev);
 
 	if (intf->cur_altsetting->desc.bInterfaceProtocol
 			!= USB_INTERFACE_PROTOCOL_MOUSE)
diff --git a/drivers/hid/hid-roccat-savu.h b/drivers/hid/hid-roccat-savu.h
index 9120ba72087f..d23217bd2b86 100644
--- a/drivers/hid/hid-roccat-savu.h
+++ b/drivers/hid/hid-roccat-savu.h
@@ -14,31 +14,6 @@
 
 #include <linux/types.h>
 
-enum {
-	SAVU_SIZE_CONTROL = 0x03,
-	SAVU_SIZE_PROFILE = 0x03,
-	SAVU_SIZE_GENERAL = 0x10,
-	SAVU_SIZE_BUTTONS = 0x2f,
-	SAVU_SIZE_MACRO = 0x0823,
-	SAVU_SIZE_INFO = 0x08,
-	SAVU_SIZE_SENSOR = 0x04,
-};
-
-enum savu_control_requests {
-	SAVU_CONTROL_REQUEST_GENERAL = 0x80,
-	SAVU_CONTROL_REQUEST_BUTTONS = 0x90,
-};
-
-enum savu_commands {
-	SAVU_COMMAND_CONTROL = 0x4,
-	SAVU_COMMAND_PROFILE = 0x5,
-	SAVU_COMMAND_GENERAL = 0x6,
-	SAVU_COMMAND_BUTTONS = 0x7,
-	SAVU_COMMAND_MACRO = 0x8,
-	SAVU_COMMAND_INFO = 0x9,
-	SAVU_COMMAND_SENSOR = 0xc,
-};
-
 struct savu_mouse_report_special {
 	uint8_t report_number; /* always 3 */
 	uint8_t zero;
@@ -77,11 +52,4 @@ struct savu_roccat_report {
 	uint8_t data[2];
 } __packed;
 
-struct savu_device {
-	int roccat_claimed;
-	int chrdev_minor;
-
-	struct mutex savu_lock;
-};
-
 #endif
diff --git a/drivers/hid/hid-sensor-hub.c b/drivers/hid/hid-sensor-hub.c
index 88fc5aefcd96..a184e1921c11 100644
--- a/drivers/hid/hid-sensor-hub.c
+++ b/drivers/hid/hid-sensor-hub.c
@@ -326,7 +326,8 @@ int sensor_hub_input_get_attribute_info(struct hid_sensor_hub_device *hsdev,
 				field->logical == attr_usage_id) {
 				sensor_hub_fill_attr_info(info, i, report->id,
 					field->unit, field->unit_exponent,
-					field->report_size);
+					field->report_size *
+							field->report_count);
 				ret = 0;
 			} else {
 				for (j = 0; j < field->maxusage; ++j) {
@@ -338,7 +339,8 @@ int sensor_hub_input_get_attribute_info(struct hid_sensor_hub_device *hsdev,
 							i, report->id,
 							field->unit,
 							field->unit_exponent,
-							field->report_size);
+							field->report_size *
+							field->report_count);
 						ret = 0;
 						break;
 					}
@@ -425,9 +427,10 @@ static int sensor_hub_raw_event(struct hid_device *hdev,
 		hid_dbg(hdev, "%d collection_index:%x hid:%x sz:%x\n",
 				i, report->field[i]->usage->collection_index,
 				report->field[i]->usage->hid,
-				report->field[i]->report_size/8);
-
-		sz = report->field[i]->report_size/8;
+				(report->field[i]->report_size *
+					report->field[i]->report_count)/8);
+		sz = (report->field[i]->report_size *
+					report->field[i]->report_count)/8;
 		if (pdata->pending.status && pdata->pending.attr_usage_id ==
 				report->field[i]->usage->hid) {
 			hid_dbg(hdev, "data was pending ...\n");
diff --git a/drivers/hid/hid-sony.c b/drivers/hid/hid-sony.c
index b18320db5f7d..da551d113762 100644
--- a/drivers/hid/hid-sony.c
+++ b/drivers/hid/hid-sony.c
@@ -419,21 +419,14 @@ static int sixaxis_usb_output_raw_report(struct hid_device *hid, __u8 *buf,
  */
 static int sixaxis_set_operational_usb(struct hid_device *hdev)
 {
-	struct usb_interface *intf = to_usb_interface(hdev->dev.parent);
-	struct usb_device *dev = interface_to_usbdev(intf);
-	__u16 ifnum = intf->cur_altsetting->desc.bInterfaceNumber;
 	int ret;
 	char *buf = kmalloc(18, GFP_KERNEL);
 
 	if (!buf)
 		return -ENOMEM;
 
-	ret = usb_control_msg(dev, usb_rcvctrlpipe(dev, 0),
-				 HID_REQ_GET_REPORT,
-				 USB_DIR_IN | USB_TYPE_CLASS |
-				 USB_RECIP_INTERFACE,
-				 (3 << 8) | 0xf2, ifnum, buf, 17,
-				 USB_CTRL_GET_TIMEOUT);
+	ret = hdev->hid_get_raw_report(hdev, 0xf2, buf, 17, HID_FEATURE_REPORT);
+
 	if (ret < 0)
 		hid_err(hdev, "can't set operational mode\n");
 
@@ -621,6 +614,54 @@ static void buzz_remove(struct hid_device *hdev)
 	drv_data->extra = NULL;
 }
 
+#ifdef CONFIG_SONY_FF
+static int sony_play_effect(struct input_dev *dev, void *data,
+			    struct ff_effect *effect)
+{
+	unsigned char buf[] = {
+		0x01,
+		0x00, 0xff, 0x00, 0xff, 0x00,
+		0x00, 0x00, 0x00, 0x00, 0x03,
+		0xff, 0x27, 0x10, 0x00, 0x32,
+		0xff, 0x27, 0x10, 0x00, 0x32,
+		0xff, 0x27, 0x10, 0x00, 0x32,
+		0xff, 0x27, 0x10, 0x00, 0x32,
+		0x00, 0x00, 0x00, 0x00, 0x00
+	};
+	__u8 left;
+	__u8 right;
+	struct hid_device *hid = input_get_drvdata(dev);
+
+	if (effect->type != FF_RUMBLE)
+		return 0;
+
+	left = effect->u.rumble.strong_magnitude / 256;
+	right = effect->u.rumble.weak_magnitude ? 1 : 0;
+
+	buf[3] = right;
+	buf[5] = left;
+
+	return hid->hid_output_raw_report(hid, buf, sizeof(buf),
+					  HID_OUTPUT_REPORT);
+}
+
+static int sony_init_ff(struct hid_device *hdev)
+{
+	struct hid_input *hidinput = list_entry(hdev->inputs.next,
+						struct hid_input, list);
+	struct input_dev *input_dev = hidinput->input;
+
+	input_set_capability(input_dev, EV_FF, FF_RUMBLE);
+	return input_ff_create_memless(input_dev, NULL, sony_play_effect);
+}
+
+#else
+static int sony_init_ff(struct hid_device *hdev)
+{
+	return 0;
+}
+#endif
+
 static int sony_probe(struct hid_device *hdev, const struct hid_device_id *id)
 {
 	int ret;
@@ -670,6 +711,10 @@ static int sony_probe(struct hid_device *hdev, const struct hid_device_id *id)
 	if (ret < 0)
 		goto err_stop;
 
+	ret = sony_init_ff(hdev);
+	if (ret < 0)
+		goto err_stop;
+
 	return 0;
 err_stop:
 	hid_hw_stop(hdev);
diff --git a/drivers/hid/hid-wiimote-modules.c b/drivers/hid/hid-wiimote-modules.c
index 71adf9e60b13..6b61f01e01e7 100644
--- a/drivers/hid/hid-wiimote-modules.c
+++ b/drivers/hid/hid-wiimote-modules.c
@@ -1655,10 +1655,39 @@ static void wiimod_pro_in_ext(struct wiimote_data *wdata, const __u8 *ext)
 	ly = (ext[4] & 0xff) | ((ext[5] & 0x0f) << 8);
 	ry = (ext[6] & 0xff) | ((ext[7] & 0x0f) << 8);
 
-	input_report_abs(wdata->extension.input, ABS_X, lx - 0x800);
-	input_report_abs(wdata->extension.input, ABS_Y, ly - 0x800);
-	input_report_abs(wdata->extension.input, ABS_RX, rx - 0x800);
-	input_report_abs(wdata->extension.input, ABS_RY, ry - 0x800);
+	/* zero-point offsets */
+	lx -= 0x800;
+	ly = 0x800 - ly;
+	rx -= 0x800;
+	ry = 0x800 - ry;
+
+	/* Trivial automatic calibration. We don't know any calibration data
+	 * in the EEPROM so we must use the first report to calibrate the
+	 * null-position of the analog sticks. Users can retrigger calibration
+	 * via sysfs, or set it explicitly. If data is off more than abs(500),
+	 * we skip calibration as the sticks are likely to be moved already. */
+	if (!(wdata->state.flags & WIIPROTO_FLAG_PRO_CALIB_DONE)) {
+		wdata->state.flags |= WIIPROTO_FLAG_PRO_CALIB_DONE;
+		if (abs(lx) < 500)
+			wdata->state.calib_pro_sticks[0] = -lx;
+		if (abs(ly) < 500)
+			wdata->state.calib_pro_sticks[1] = -ly;
+		if (abs(rx) < 500)
+			wdata->state.calib_pro_sticks[2] = -rx;
+		if (abs(ry) < 500)
+			wdata->state.calib_pro_sticks[3] = -ry;
+	}
+
+	/* apply calibration data */
+	lx += wdata->state.calib_pro_sticks[0];
+	ly += wdata->state.calib_pro_sticks[1];
+	rx += wdata->state.calib_pro_sticks[2];
+	ry += wdata->state.calib_pro_sticks[3];
+
+	input_report_abs(wdata->extension.input, ABS_X, lx);
+	input_report_abs(wdata->extension.input, ABS_Y, ly);
+	input_report_abs(wdata->extension.input, ABS_RX, rx);
+	input_report_abs(wdata->extension.input, ABS_RY, ry);
 
 	input_report_key(wdata->extension.input,
 			 wiimod_pro_map[WIIMOD_PRO_KEY_RIGHT],
@@ -1766,12 +1795,70 @@ static int wiimod_pro_play(struct input_dev *dev, void *data,
 	return 0;
 }
 
+static ssize_t wiimod_pro_calib_show(struct device *dev,
+				     struct device_attribute *attr,
+				     char *out)
+{
+	struct wiimote_data *wdata = dev_to_wii(dev);
+	int r;
+
+	r = 0;
+	r += sprintf(&out[r], "%+06hd:", wdata->state.calib_pro_sticks[0]);
+	r += sprintf(&out[r], "%+06hd ", wdata->state.calib_pro_sticks[1]);
+	r += sprintf(&out[r], "%+06hd:", wdata->state.calib_pro_sticks[2]);
+	r += sprintf(&out[r], "%+06hd\n", wdata->state.calib_pro_sticks[3]);
+
+	return r;
+}
+
+static ssize_t wiimod_pro_calib_store(struct device *dev,
+				      struct device_attribute *attr,
+				      const char *buf, size_t count)
+{
+	struct wiimote_data *wdata = dev_to_wii(dev);
+	int r;
+	s16 x1, y1, x2, y2;
+
+	if (!strncmp(buf, "scan\n", 5)) {
+		spin_lock_irq(&wdata->state.lock);
+		wdata->state.flags &= ~WIIPROTO_FLAG_PRO_CALIB_DONE;
+		spin_unlock_irq(&wdata->state.lock);
+	} else {
+		r = sscanf(buf, "%hd:%hd %hd:%hd", &x1, &y1, &x2, &y2);
+		if (r != 4)
+			return -EINVAL;
+
+		spin_lock_irq(&wdata->state.lock);
+		wdata->state.flags |= WIIPROTO_FLAG_PRO_CALIB_DONE;
+		spin_unlock_irq(&wdata->state.lock);
+
+		wdata->state.calib_pro_sticks[0] = x1;
+		wdata->state.calib_pro_sticks[1] = y1;
+		wdata->state.calib_pro_sticks[2] = x2;
+		wdata->state.calib_pro_sticks[3] = y2;
+	}
+
+	return strnlen(buf, PAGE_SIZE);
+}
+
+static DEVICE_ATTR(pro_calib, S_IRUGO|S_IWUSR|S_IWGRP, wiimod_pro_calib_show,
+		   wiimod_pro_calib_store);
+
 static int wiimod_pro_probe(const struct wiimod_ops *ops,
 			    struct wiimote_data *wdata)
 {
 	int ret, i;
+	unsigned long flags;
 
 	INIT_WORK(&wdata->rumble_worker, wiimod_rumble_worker);
+	wdata->state.calib_pro_sticks[0] = 0;
+	wdata->state.calib_pro_sticks[1] = 0;
+	wdata->state.calib_pro_sticks[2] = 0;
+	wdata->state.calib_pro_sticks[3] = 0;
+
+	spin_lock_irqsave(&wdata->state.lock, flags);
+	wdata->state.flags &= ~WIIPROTO_FLAG_PRO_CALIB_DONE;
+	spin_unlock_irqrestore(&wdata->state.lock, flags);
 
 	wdata->extension.input = input_allocate_device();
 	if (!wdata->extension.input)
@@ -1786,6 +1873,13 @@ static int wiimod_pro_probe(const struct wiimod_ops *ops,
 		goto err_free;
 	}
 
+	ret = device_create_file(&wdata->hdev->dev,
+				 &dev_attr_pro_calib);
+	if (ret) {
+		hid_err(wdata->hdev, "cannot create sysfs attribute\n");
+		goto err_free;
+	}
+
 	wdata->extension.input->open = wiimod_pro_open;
 	wdata->extension.input->close = wiimod_pro_close;
 	wdata->extension.input->dev.parent = &wdata->hdev->dev;
@@ -1806,20 +1900,23 @@ static int wiimod_pro_probe(const struct wiimod_ops *ops,
 	set_bit(ABS_RX, wdata->extension.input->absbit);
 	set_bit(ABS_RY, wdata->extension.input->absbit);
 	input_set_abs_params(wdata->extension.input,
-			     ABS_X, -0x800, 0x800, 2, 4);
+			     ABS_X, -0x400, 0x400, 4, 100);
 	input_set_abs_params(wdata->extension.input,
-			     ABS_Y, -0x800, 0x800, 2, 4);
+			     ABS_Y, -0x400, 0x400, 4, 100);
 	input_set_abs_params(wdata->extension.input,
-			     ABS_RX, -0x800, 0x800, 2, 4);
+			     ABS_RX, -0x400, 0x400, 4, 100);
 	input_set_abs_params(wdata->extension.input,
-			     ABS_RY, -0x800, 0x800, 2, 4);
+			     ABS_RY, -0x400, 0x400, 4, 100);
 
 	ret = input_register_device(wdata->extension.input);
 	if (ret)
-		goto err_free;
+		goto err_file;
 
 	return 0;
 
+err_file:
+	device_remove_file(&wdata->hdev->dev,
+			   &dev_attr_pro_calib);
 err_free:
 	input_free_device(wdata->extension.input);
 	wdata->extension.input = NULL;
@@ -1837,6 +1934,8 @@ static void wiimod_pro_remove(const struct wiimod_ops *ops,
 	input_unregister_device(wdata->extension.input);
 	wdata->extension.input = NULL;
 	cancel_work_sync(&wdata->rumble_worker);
+	device_remove_file(&wdata->hdev->dev,
+			   &dev_attr_pro_calib);
 
 	spin_lock_irqsave(&wdata->state.lock, flags);
 	wiiproto_req_rumble(wdata, 0);
diff --git a/drivers/hid/hid-wiimote.h b/drivers/hid/hid-wiimote.h
index 75db0c400037..10934aa129fb 100644
--- a/drivers/hid/hid-wiimote.h
+++ b/drivers/hid/hid-wiimote.h
@@ -46,6 +46,7 @@
 #define WIIPROTO_FLAG_DRM_LOCKED	0x8000
 #define WIIPROTO_FLAG_BUILTIN_MP	0x010000
 #define WIIPROTO_FLAG_NO_MP		0x020000
+#define WIIPROTO_FLAG_PRO_CALIB_DONE	0x040000
 
 #define WIIPROTO_FLAGS_LEDS (WIIPROTO_FLAG_LED1 | WIIPROTO_FLAG_LED2 | \
 					WIIPROTO_FLAG_LED3 | WIIPROTO_FLAG_LED4)
@@ -135,6 +136,7 @@ struct wiimote_state {
 
 	/* calibration/cache data */
 	__u16 calib_bboard[4][3];
+	__s16 calib_pro_sticks[4];
 	__u8 cache_rumble;
 };
 
@@ -327,7 +329,7 @@ static inline void wiimote_cmd_acquire_noint(struct wiimote_data *wdata)
 static inline void wiimote_cmd_set(struct wiimote_data *wdata, int cmd,
 								__u32 opt)
 {
-	INIT_COMPLETION(wdata->state.ready);
+	reinit_completion(&wdata->state.ready);
 	wdata->state.cmd = cmd;
 	wdata->state.opt = opt;
 }
diff --git a/drivers/hid/i2c-hid/i2c-hid.c b/drivers/hid/i2c-hid/i2c-hid.c
index b7ecc3623a43..5f7e55f4b7f0 100644
--- a/drivers/hid/i2c-hid/i2c-hid.c
+++ b/drivers/hid/i2c-hid/i2c-hid.c
@@ -455,10 +455,6 @@ static void i2c_hid_init_reports(struct hid_device *hid)
 	}
 
 	list_for_each_entry(report,
-		&hid->report_enum[HID_INPUT_REPORT].report_list, list)
-		i2c_hid_init_report(report, inbuf, ihid->bufsize);
-
-	list_for_each_entry(report,
 		&hid->report_enum[HID_FEATURE_REPORT].report_list, list)
 		i2c_hid_init_report(report, inbuf, ihid->bufsize);
 
diff --git a/drivers/hid/usbhid/hid-quirks.c b/drivers/hid/usbhid/hid-quirks.c
index 3fca3be08337..0db9a67278ba 100644
--- a/drivers/hid/usbhid/hid-quirks.c
+++ b/drivers/hid/usbhid/hid-quirks.c
@@ -84,6 +84,8 @@ static const struct hid_blacklist {
 	{ USB_VENDOR_ID_REALTEK, USB_DEVICE_ID_REALTEK_READER, HID_QUIRK_NO_INIT_REPORTS },
 	{ USB_VENDOR_ID_SENNHEISER, USB_DEVICE_ID_SENNHEISER_BTD500USB, HID_QUIRK_NOGET },
 	{ USB_VENDOR_ID_SIGMATEL, USB_DEVICE_ID_SIGMATEL_STMP3780, HID_QUIRK_NOGET },
+	{ USB_VENDOR_ID_SIS2_TOUCH, USB_DEVICE_ID_SIS9200_TOUCH, HID_QUIRK_NOGET },
+	{ USB_VENDOR_ID_SIS2_TOUCH, USB_DEVICE_ID_SIS817_TOUCH, HID_QUIRK_NOGET },
 	{ USB_VENDOR_ID_SUN, USB_DEVICE_ID_RARITAN_KVM_DONGLE, HID_QUIRK_NOGET },
 	{ USB_VENDOR_ID_SYMBOL, USB_DEVICE_ID_SYMBOL_SCANNER_1, HID_QUIRK_NOGET },
 	{ USB_VENDOR_ID_SYMBOL, USB_DEVICE_ID_SYMBOL_SCANNER_2, HID_QUIRK_NOGET },
diff --git a/drivers/hwmon/jz4740-hwmon.c b/drivers/hwmon/jz4740-hwmon.c
index e0d66b9590ab..a183e488db78 100644
--- a/drivers/hwmon/jz4740-hwmon.c
+++ b/drivers/hwmon/jz4740-hwmon.c
@@ -66,7 +66,7 @@ static ssize_t jz4740_hwmon_read_adcin(struct device *dev,
 
 	mutex_lock(&hwmon->lock);
 
-	INIT_COMPLETION(*completion);
+	reinit_completion(completion);
 
 	enable_irq(hwmon->irq);
 	hwmon->cell->enable(to_platform_device(dev));
diff --git a/drivers/hwmon/lm90.c b/drivers/hwmon/lm90.c
index cdff74282955..4c4c1421bf28 100644
--- a/drivers/hwmon/lm90.c
+++ b/drivers/hwmon/lm90.c
@@ -60,6 +60,11 @@
  * This driver also supports the G781 from GMT. This device is compatible
  * with the ADM1032.
  *
+ * This driver also supports TMP451 from Texas Instruments. This device is
+ * supported in both compatibility and extended mode. It's mostly compatible
+ * with ADT7461 except for local temperature low byte register and max
+ * conversion rate.
+ *
  * Since the LM90 was the first chipset supported by this driver, most
  * comments will refer to this chipset, but are actually general and
  * concern all supported chipsets, unless mentioned otherwise.
@@ -89,6 +94,8 @@
 #include <linux/err.h>
 #include <linux/mutex.h>
 #include <linux/sysfs.h>
+#include <linux/interrupt.h>
+#include <linux/regulator/consumer.h>
 
 /*
  * Addresses to scan
@@ -110,7 +117,7 @@ static const unsigned short normal_i2c[] = {
 	0x4d, 0x4e, 0x4f, I2C_CLIENT_END };
 
 enum chips { lm90, adm1032, lm99, lm86, max6657, max6659, adt7461, max6680,
-	max6646, w83l771, max6696, sa56004, g781 };
+	max6646, w83l771, max6696, sa56004, g781, tmp451 };
 
 /*
  * The LM90 registers
@@ -167,6 +174,9 @@ enum chips { lm90, adm1032, lm99, lm86, max6657, max6659, adt7461, max6680,
 #define LM90_DEF_CONVRATE_RVAL	6	/* Def conversion rate register value */
 #define LM90_MAX_CONVRATE_MS	16000	/* Maximum conversion rate in ms */
 
+/* TMP451 registers */
+#define TMP451_REG_R_LOCAL_TEMPL	0x15
+
 /*
  * Device flags
  */
@@ -179,6 +189,23 @@ enum chips { lm90, adm1032, lm99, lm86, max6657, max6659, adt7461, max6680,
 #define LM90_HAVE_TEMP3		(1 << 6) /* 3rd temperature sensor	*/
 #define LM90_HAVE_BROKEN_ALERT	(1 << 7) /* Broken alert		*/
 
+/* LM90 status */
+#define LM90_STATUS_LTHRM	(1 << 0) /* local THERM limit tripped */
+#define LM90_STATUS_RTHRM	(1 << 1) /* remote THERM limit tripped */
+#define LM90_STATUS_ROPEN	(1 << 2) /* remote is an open circuit */
+#define LM90_STATUS_RLOW	(1 << 3) /* remote low temp limit tripped */
+#define LM90_STATUS_RHIGH	(1 << 4) /* remote high temp limit tripped */
+#define LM90_STATUS_LLOW	(1 << 5) /* local low temp limit tripped */
+#define LM90_STATUS_LHIGH	(1 << 6) /* local high temp limit tripped */
+
+#define MAX6696_STATUS2_R2THRM	(1 << 1) /* remote2 THERM limit tripped */
+#define MAX6696_STATUS2_R2OPEN	(1 << 2) /* remote2 is an open circuit */
+#define MAX6696_STATUS2_R2LOW	(1 << 3) /* remote2 low temp limit tripped */
+#define MAX6696_STATUS2_R2HIGH	(1 << 4) /* remote2 high temp limit tripped */
+#define MAX6696_STATUS2_ROT2	(1 << 5) /* remote emergency limit tripped */
+#define MAX6696_STATUS2_R2OT2	(1 << 6) /* remote2 emergency limit tripped */
+#define MAX6696_STATUS2_LOT2	(1 << 7) /* local emergency limit tripped */
+
 /*
  * Driver data (common to all clients)
  */
@@ -205,6 +232,7 @@ static const struct i2c_device_id lm90_id[] = {
 	{ "nct1008", adt7461 },
 	{ "w83l771", w83l771 },
 	{ "sa56004", sa56004 },
+	{ "tmp451", tmp451 },
 	{ }
 };
 MODULE_DEVICE_TABLE(i2c, lm90_id);
@@ -278,7 +306,7 @@ static const struct lm90_params lm90_params[] = {
 	[max6696] = {
 		.flags = LM90_HAVE_EMERGENCY
 		  | LM90_HAVE_EMERGENCY_ALARM | LM90_HAVE_TEMP3,
-		.alert_alarms = 0x187c,
+		.alert_alarms = 0x1c7c,
 		.max_convrate = 6,
 		.reg_local_ext = MAX6657_REG_R_LOCAL_TEMPL,
 	},
@@ -293,6 +321,43 @@ static const struct lm90_params lm90_params[] = {
 		.max_convrate = 9,
 		.reg_local_ext = SA56004_REG_R_LOCAL_TEMPL,
 	},
+	[tmp451] = {
+		.flags = LM90_HAVE_OFFSET | LM90_HAVE_REM_LIMIT_EXT
+		  | LM90_HAVE_BROKEN_ALERT,
+		.alert_alarms = 0x7c,
+		.max_convrate = 9,
+		.reg_local_ext = TMP451_REG_R_LOCAL_TEMPL,
+	}
+};
+
+/*
+ * TEMP8 register index
+ */
+enum lm90_temp8_reg_index {
+	LOCAL_LOW = 0,
+	LOCAL_HIGH,
+	LOCAL_CRIT,
+	REMOTE_CRIT,
+	LOCAL_EMERG,	/* max6659 and max6695/96 */
+	REMOTE_EMERG,	/* max6659 and max6695/96 */
+	REMOTE2_CRIT,	/* max6695/96 only */
+	REMOTE2_EMERG,	/* max6695/96 only */
+	TEMP8_REG_NUM
+};
+
+/*
+ * TEMP11 register index
+ */
+enum lm90_temp11_reg_index {
+	REMOTE_TEMP = 0,
+	REMOTE_LOW,
+	REMOTE_HIGH,
+	REMOTE_OFFSET,	/* except max6646, max6657/58/59, and max6695/96 */
+	LOCAL_TEMP,
+	REMOTE2_TEMP,	/* max6695/96 only */
+	REMOTE2_LOW,	/* max6695/96 only */
+	REMOTE2_HIGH,	/* max6695/96 only */
+	TEMP11_REG_NUM
 };
 
 /*
@@ -302,6 +367,7 @@ static const struct lm90_params lm90_params[] = {
 struct lm90_data {
 	struct device *hwmon_dev;
 	struct mutex update_lock;
+	struct regulator *regulator;
 	char valid; /* zero until following fields are valid */
 	unsigned long last_updated; /* in jiffies */
 	int kind;
@@ -317,25 +383,8 @@ struct lm90_data {
 	u8 reg_local_ext;	/* local extension register offset */
 
 	/* registers values */
-	s8 temp8[8];	/* 0: local low limit
-			 * 1: local high limit
-			 * 2: local critical limit
-			 * 3: remote critical limit
-			 * 4: local emergency limit (max6659 and max6695/96)
-			 * 5: remote emergency limit (max6659 and max6695/96)
-			 * 6: remote 2 critical limit (max6695/96 only)
-			 * 7: remote 2 emergency limit (max6695/96 only)
-			 */
-	s16 temp11[8];	/* 0: remote input
-			 * 1: remote low limit
-			 * 2: remote high limit
-			 * 3: remote offset (except max6646, max6657/58/59,
-			 *		     and max6695/96)
-			 * 4: local input
-			 * 5: remote 2 input (max6695/96 only)
-			 * 6: remote 2 low limit (max6695/96 only)
-			 * 7: remote 2 high limit (max6695/96 only)
-			 */
+	s8 temp8[TEMP8_REG_NUM];
+	s16 temp11[TEMP11_REG_NUM];
 	u8 temp_hyst;
 	u16 alarms; /* bitvector (upper 8 bits for max6695/96) */
 };
@@ -477,37 +526,42 @@ static struct lm90_data *lm90_update_device(struct device *dev)
 		u8 alarms;
 
 		dev_dbg(&client->dev, "Updating lm90 data.\n");
-		lm90_read_reg(client, LM90_REG_R_LOCAL_LOW, &data->temp8[0]);
-		lm90_read_reg(client, LM90_REG_R_LOCAL_HIGH, &data->temp8[1]);
-		lm90_read_reg(client, LM90_REG_R_LOCAL_CRIT, &data->temp8[2]);
-		lm90_read_reg(client, LM90_REG_R_REMOTE_CRIT, &data->temp8[3]);
+		lm90_read_reg(client, LM90_REG_R_LOCAL_LOW,
+			      &data->temp8[LOCAL_LOW]);
+		lm90_read_reg(client, LM90_REG_R_LOCAL_HIGH,
+			      &data->temp8[LOCAL_HIGH]);
+		lm90_read_reg(client, LM90_REG_R_LOCAL_CRIT,
+			      &data->temp8[LOCAL_CRIT]);
+		lm90_read_reg(client, LM90_REG_R_REMOTE_CRIT,
+			      &data->temp8[REMOTE_CRIT]);
 		lm90_read_reg(client, LM90_REG_R_TCRIT_HYST, &data->temp_hyst);
 
 		if (data->reg_local_ext) {
 			lm90_read16(client, LM90_REG_R_LOCAL_TEMP,
 				    data->reg_local_ext,
-				    &data->temp11[4]);
+				    &data->temp11[LOCAL_TEMP]);
 		} else {
 			if (lm90_read_reg(client, LM90_REG_R_LOCAL_TEMP,
 					  &h) == 0)
-				data->temp11[4] = h << 8;
+				data->temp11[LOCAL_TEMP] = h << 8;
 		}
 		lm90_read16(client, LM90_REG_R_REMOTE_TEMPH,
-			    LM90_REG_R_REMOTE_TEMPL, &data->temp11[0]);
+			    LM90_REG_R_REMOTE_TEMPL,
+			    &data->temp11[REMOTE_TEMP]);
 
 		if (lm90_read_reg(client, LM90_REG_R_REMOTE_LOWH, &h) == 0) {
-			data->temp11[1] = h << 8;
+			data->temp11[REMOTE_LOW] = h << 8;
 			if ((data->flags & LM90_HAVE_REM_LIMIT_EXT)
 			 && lm90_read_reg(client, LM90_REG_R_REMOTE_LOWL,
 					  &l) == 0)
-				data->temp11[1] |= l;
+				data->temp11[REMOTE_LOW] |= l;
 		}
 		if (lm90_read_reg(client, LM90_REG_R_REMOTE_HIGHH, &h) == 0) {
-			data->temp11[2] = h << 8;
+			data->temp11[REMOTE_HIGH] = h << 8;
 			if ((data->flags & LM90_HAVE_REM_LIMIT_EXT)
 			 && lm90_read_reg(client, LM90_REG_R_REMOTE_HIGHL,
 					  &l) == 0)
-				data->temp11[2] |= l;
+				data->temp11[REMOTE_HIGH] |= l;
 		}
 
 		if (data->flags & LM90_HAVE_OFFSET) {
@@ -515,13 +569,13 @@ static struct lm90_data *lm90_update_device(struct device *dev)
 					  &h) == 0
 			 && lm90_read_reg(client, LM90_REG_R_REMOTE_OFFSL,
 					  &l) == 0)
-				data->temp11[3] = (h << 8) | l;
+				data->temp11[REMOTE_OFFSET] = (h << 8) | l;
 		}
 		if (data->flags & LM90_HAVE_EMERGENCY) {
 			lm90_read_reg(client, MAX6659_REG_R_LOCAL_EMERG,
-				      &data->temp8[4]);
+				      &data->temp8[LOCAL_EMERG]);
 			lm90_read_reg(client, MAX6659_REG_R_REMOTE_EMERG,
-				      &data->temp8[5]);
+				      &data->temp8[REMOTE_EMERG]);
 		}
 		lm90_read_reg(client, LM90_REG_R_STATUS, &alarms);
 		data->alarms = alarms;	/* save as 16 bit value */
@@ -529,15 +583,16 @@ static struct lm90_data *lm90_update_device(struct device *dev)
 		if (data->kind == max6696) {
 			lm90_select_remote_channel(client, data, 1);
 			lm90_read_reg(client, LM90_REG_R_REMOTE_CRIT,
-				      &data->temp8[6]);
+				      &data->temp8[REMOTE2_CRIT]);
 			lm90_read_reg(client, MAX6659_REG_R_REMOTE_EMERG,
-				      &data->temp8[7]);
+				      &data->temp8[REMOTE2_EMERG]);
 			lm90_read16(client, LM90_REG_R_REMOTE_TEMPH,
-				    LM90_REG_R_REMOTE_TEMPL, &data->temp11[5]);
+				    LM90_REG_R_REMOTE_TEMPL,
+				    &data->temp11[REMOTE2_TEMP]);
 			if (!lm90_read_reg(client, LM90_REG_R_REMOTE_LOWH, &h))
-				data->temp11[6] = h << 8;
+				data->temp11[REMOTE2_LOW] = h << 8;
 			if (!lm90_read_reg(client, LM90_REG_R_REMOTE_HIGHH, &h))
-				data->temp11[7] = h << 8;
+				data->temp11[REMOTE2_HIGH] = h << 8;
 			lm90_select_remote_channel(client, data, 0);
 
 			if (!lm90_read_reg(client, MAX6696_REG_R_STATUS2,
@@ -709,7 +764,7 @@ static ssize_t show_temp8(struct device *dev, struct device_attribute *devattr,
 	struct lm90_data *data = lm90_update_device(dev);
 	int temp;
 
-	if (data->kind == adt7461)
+	if (data->kind == adt7461 || data->kind == tmp451)
 		temp = temp_from_u8_adt7461(data, data->temp8[attr->index]);
 	else if (data->kind == max6646)
 		temp = temp_from_u8(data->temp8[attr->index]);
@@ -726,7 +781,7 @@ static ssize_t show_temp8(struct device *dev, struct device_attribute *devattr,
 static ssize_t set_temp8(struct device *dev, struct device_attribute *devattr,
 			 const char *buf, size_t count)
 {
-	static const u8 reg[8] = {
+	static const u8 reg[TEMP8_REG_NUM] = {
 		LM90_REG_W_LOCAL_LOW,
 		LM90_REG_W_LOCAL_HIGH,
 		LM90_REG_W_LOCAL_CRIT,
@@ -753,7 +808,7 @@ static ssize_t set_temp8(struct device *dev, struct device_attribute *devattr,
 		val -= 16000;
 
 	mutex_lock(&data->update_lock);
-	if (data->kind == adt7461)
+	if (data->kind == adt7461 || data->kind == tmp451)
 		data->temp8[nr] = temp_to_u8_adt7461(data, val);
 	else if (data->kind == max6646)
 		data->temp8[nr] = temp_to_u8(val);
@@ -775,7 +830,7 @@ static ssize_t show_temp11(struct device *dev, struct device_attribute *devattr,
 	struct lm90_data *data = lm90_update_device(dev);
 	int temp;
 
-	if (data->kind == adt7461)
+	if (data->kind == adt7461 || data->kind == tmp451)
 		temp = temp_from_u16_adt7461(data, data->temp11[attr->index]);
 	else if (data->kind == max6646)
 		temp = temp_from_u16(data->temp11[attr->index]);
@@ -821,7 +876,7 @@ static ssize_t set_temp11(struct device *dev, struct device_attribute *devattr,
 		val -= 16000;
 
 	mutex_lock(&data->update_lock);
-	if (data->kind == adt7461)
+	if (data->kind == adt7461 || data->kind == tmp451)
 		data->temp11[index] = temp_to_u16_adt7461(data, val);
 	else if (data->kind == max6646)
 		data->temp11[index] = temp_to_u8(val) << 8;
@@ -850,7 +905,7 @@ static ssize_t show_temphyst(struct device *dev,
 	struct lm90_data *data = lm90_update_device(dev);
 	int temp;
 
-	if (data->kind == adt7461)
+	if (data->kind == adt7461 || data->kind == tmp451)
 		temp = temp_from_u8_adt7461(data, data->temp8[attr->index]);
 	else if (data->kind == max6646)
 		temp = temp_from_u8(data->temp8[attr->index]);
@@ -878,12 +933,12 @@ static ssize_t set_temphyst(struct device *dev, struct device_attribute *dummy,
 		return err;
 
 	mutex_lock(&data->update_lock);
-	if (data->kind == adt7461)
-		temp = temp_from_u8_adt7461(data, data->temp8[2]);
+	if (data->kind == adt7461 || data->kind == tmp451)
+		temp = temp_from_u8_adt7461(data, data->temp8[LOCAL_CRIT]);
 	else if (data->kind == max6646)
-		temp = temp_from_u8(data->temp8[2]);
+		temp = temp_from_u8(data->temp8[LOCAL_CRIT]);
 	else
-		temp = temp_from_s8(data->temp8[2]);
+		temp = temp_from_s8(data->temp8[LOCAL_CRIT]);
 
 	data->temp_hyst = hyst_to_reg(temp - val);
 	i2c_smbus_write_byte_data(client, LM90_REG_W_TCRIT_HYST,
@@ -937,25 +992,28 @@ static ssize_t set_update_interval(struct device *dev,
 	return count;
 }
 
-static SENSOR_DEVICE_ATTR_2(temp1_input, S_IRUGO, show_temp11, NULL, 0, 4);
-static SENSOR_DEVICE_ATTR_2(temp2_input, S_IRUGO, show_temp11, NULL, 0, 0);
+static SENSOR_DEVICE_ATTR_2(temp1_input, S_IRUGO, show_temp11, NULL,
+	0, LOCAL_TEMP);
+static SENSOR_DEVICE_ATTR_2(temp2_input, S_IRUGO, show_temp11, NULL,
+	0, REMOTE_TEMP);
 static SENSOR_DEVICE_ATTR(temp1_min, S_IWUSR | S_IRUGO, show_temp8,
-	set_temp8, 0);
+	set_temp8, LOCAL_LOW);
 static SENSOR_DEVICE_ATTR_2(temp2_min, S_IWUSR | S_IRUGO, show_temp11,
-	set_temp11, 0, 1);
+	set_temp11, 0, REMOTE_LOW);
 static SENSOR_DEVICE_ATTR(temp1_max, S_IWUSR | S_IRUGO, show_temp8,
-	set_temp8, 1);
+	set_temp8, LOCAL_HIGH);
 static SENSOR_DEVICE_ATTR_2(temp2_max, S_IWUSR | S_IRUGO, show_temp11,
-	set_temp11, 1, 2);
+	set_temp11, 1, REMOTE_HIGH);
 static SENSOR_DEVICE_ATTR(temp1_crit, S_IWUSR | S_IRUGO, show_temp8,
-	set_temp8, 2);
+	set_temp8, LOCAL_CRIT);
 static SENSOR_DEVICE_ATTR(temp2_crit, S_IWUSR | S_IRUGO, show_temp8,
-	set_temp8, 3);
+	set_temp8, REMOTE_CRIT);
 static SENSOR_DEVICE_ATTR(temp1_crit_hyst, S_IWUSR | S_IRUGO, show_temphyst,
-	set_temphyst, 2);
-static SENSOR_DEVICE_ATTR(temp2_crit_hyst, S_IRUGO, show_temphyst, NULL, 3);
+	set_temphyst, LOCAL_CRIT);
+static SENSOR_DEVICE_ATTR(temp2_crit_hyst, S_IRUGO, show_temphyst, NULL,
+	REMOTE_CRIT);
 static SENSOR_DEVICE_ATTR_2(temp2_offset, S_IWUSR | S_IRUGO, show_temp11,
-	set_temp11, 2, 3);
+	set_temp11, 2, REMOTE_OFFSET);
 
 /* Individual alarm files */
 static SENSOR_DEVICE_ATTR(temp1_crit_alarm, S_IRUGO, show_alarm, NULL, 0);
@@ -1003,13 +1061,13 @@ static const struct attribute_group lm90_group = {
  * Additional attributes for devices with emergency sensors
  */
 static SENSOR_DEVICE_ATTR(temp1_emergency, S_IWUSR | S_IRUGO, show_temp8,
-	set_temp8, 4);
+	set_temp8, LOCAL_EMERG);
 static SENSOR_DEVICE_ATTR(temp2_emergency, S_IWUSR | S_IRUGO, show_temp8,
-	set_temp8, 5);
+	set_temp8, REMOTE_EMERG);
 static SENSOR_DEVICE_ATTR(temp1_emergency_hyst, S_IRUGO, show_temphyst,
-			  NULL, 4);
+			  NULL, LOCAL_EMERG);
 static SENSOR_DEVICE_ATTR(temp2_emergency_hyst, S_IRUGO, show_temphyst,
-			  NULL, 5);
+			  NULL, REMOTE_EMERG);
 
 static struct attribute *lm90_emergency_attributes[] = {
 	&sensor_dev_attr_temp1_emergency.dev_attr.attr,
@@ -1039,18 +1097,20 @@ static const struct attribute_group lm90_emergency_alarm_group = {
 /*
  * Additional attributes for devices with 3 temperature sensors
  */
-static SENSOR_DEVICE_ATTR_2(temp3_input, S_IRUGO, show_temp11, NULL, 0, 5);
+static SENSOR_DEVICE_ATTR_2(temp3_input, S_IRUGO, show_temp11, NULL,
+	0, REMOTE2_TEMP);
 static SENSOR_DEVICE_ATTR_2(temp3_min, S_IWUSR | S_IRUGO, show_temp11,
-	set_temp11, 3, 6);
+	set_temp11, 3, REMOTE2_LOW);
 static SENSOR_DEVICE_ATTR_2(temp3_max, S_IWUSR | S_IRUGO, show_temp11,
-	set_temp11, 4, 7);
+	set_temp11, 4, REMOTE2_HIGH);
 static SENSOR_DEVICE_ATTR(temp3_crit, S_IWUSR | S_IRUGO, show_temp8,
-	set_temp8, 6);
-static SENSOR_DEVICE_ATTR(temp3_crit_hyst, S_IRUGO, show_temphyst, NULL, 6);
+	set_temp8, REMOTE2_CRIT);
+static SENSOR_DEVICE_ATTR(temp3_crit_hyst, S_IRUGO, show_temphyst, NULL,
+	REMOTE2_CRIT);
 static SENSOR_DEVICE_ATTR(temp3_emergency, S_IWUSR | S_IRUGO, show_temp8,
-	set_temp8, 7);
+	set_temp8, REMOTE2_EMERG);
 static SENSOR_DEVICE_ATTR(temp3_emergency_hyst, S_IRUGO, show_temphyst,
-			  NULL, 7);
+			  NULL, REMOTE2_EMERG);
 
 static SENSOR_DEVICE_ATTR(temp3_crit_alarm, S_IRUGO, show_alarm, NULL, 9);
 static SENSOR_DEVICE_ATTR(temp3_fault, S_IRUGO, show_alarm, NULL, 10);
@@ -1306,6 +1366,19 @@ static int lm90_detect(struct i2c_client *client,
 		 && (config1 & 0x3F) == 0x00
 		 && convrate <= 0x08)
 			name = "g781";
+	} else
+	if (address == 0x4C
+	 && man_id == 0x55) { /* Texas Instruments */
+		int local_ext;
+
+		local_ext = i2c_smbus_read_byte_data(client,
+						     TMP451_REG_R_LOCAL_TEMPL);
+
+		if (chip_id == 0x00 /* TMP451 */
+		 && (config1 & 0x1B) == 0x00
+		 && convrate <= 0x09
+		 && (local_ext & 0x0F) == 0x00)
+			name = "tmp451";
 	}
 
 	if (!name) { /* identification failed */
@@ -1367,7 +1440,7 @@ static void lm90_init_client(struct i2c_client *client)
 	data->config_orig = config;
 
 	/* Check Temperature Range Select */
-	if (data->kind == adt7461) {
+	if (data->kind == adt7461 || data->kind == tmp451) {
 		if (config & 0x04)
 			data->flags |= LM90_FLAG_ADT7461_EXT;
 	}
@@ -1391,14 +1464,74 @@ static void lm90_init_client(struct i2c_client *client)
 		i2c_smbus_write_byte_data(client, LM90_REG_W_CONFIG1, config);
 }
 
+static bool lm90_is_tripped(struct i2c_client *client, u16 *status)
+{
+	struct lm90_data *data = i2c_get_clientdata(client);
+	u8 st, st2 = 0;
+
+	lm90_read_reg(client, LM90_REG_R_STATUS, &st);
+
+	if (data->kind == max6696)
+		lm90_read_reg(client, MAX6696_REG_R_STATUS2, &st2);
+
+	*status = st | (st2 << 8);
+
+	if ((st & 0x7f) == 0 && (st2 & 0xfe) == 0)
+		return false;
+
+	if ((st & (LM90_STATUS_LLOW | LM90_STATUS_LHIGH | LM90_STATUS_LTHRM)) ||
+	    (st2 & MAX6696_STATUS2_LOT2))
+		dev_warn(&client->dev,
+			 "temp%d out of range, please check!\n", 1);
+	if ((st & (LM90_STATUS_RLOW | LM90_STATUS_RHIGH | LM90_STATUS_RTHRM)) ||
+	    (st2 & MAX6696_STATUS2_ROT2))
+		dev_warn(&client->dev,
+			 "temp%d out of range, please check!\n", 2);
+	if (st & LM90_STATUS_ROPEN)
+		dev_warn(&client->dev,
+			 "temp%d diode open, please check!\n", 2);
+	if (st2 & (MAX6696_STATUS2_R2LOW | MAX6696_STATUS2_R2HIGH |
+		   MAX6696_STATUS2_R2THRM | MAX6696_STATUS2_R2OT2))
+		dev_warn(&client->dev,
+			 "temp%d out of range, please check!\n", 3);
+	if (st2 & MAX6696_STATUS2_R2OPEN)
+		dev_warn(&client->dev,
+			 "temp%d diode open, please check!\n", 3);
+
+	return true;
+}
+
+static irqreturn_t lm90_irq_thread(int irq, void *dev_id)
+{
+	struct i2c_client *client = dev_id;
+	u16 status;
+
+	if (lm90_is_tripped(client, &status))
+		return IRQ_HANDLED;
+	else
+		return IRQ_NONE;
+}
+
 static int lm90_probe(struct i2c_client *client,
 		      const struct i2c_device_id *id)
 {
 	struct device *dev = &client->dev;
 	struct i2c_adapter *adapter = to_i2c_adapter(dev->parent);
 	struct lm90_data *data;
+	struct regulator *regulator;
 	int err;
 
+	regulator = devm_regulator_get(dev, "vcc");
+	if (IS_ERR(regulator))
+		return PTR_ERR(regulator);
+
+	err = regulator_enable(regulator);
+	if (err < 0) {
+		dev_err(&client->dev,
+			"Failed to enable regulator: %d\n", err);
+		return err;
+	}
+
 	data = devm_kzalloc(&client->dev, sizeof(struct lm90_data), GFP_KERNEL);
 	if (!data)
 		return -ENOMEM;
@@ -1406,6 +1539,8 @@ static int lm90_probe(struct i2c_client *client,
 	i2c_set_clientdata(client, data);
 	mutex_init(&data->update_lock);
 
+	data->regulator = regulator;
+
 	/* Set the device type */
 	data->kind = id->driver_data;
 	if (data->kind == adm1032) {
@@ -1467,12 +1602,26 @@ static int lm90_probe(struct i2c_client *client,
 		goto exit_remove_files;
 	}
 
+	if (client->irq) {
+		dev_dbg(dev, "IRQ: %d\n", client->irq);
+		err = devm_request_threaded_irq(dev, client->irq,
+						NULL, lm90_irq_thread,
+						IRQF_TRIGGER_LOW | IRQF_ONESHOT,
+						"lm90", client);
+		if (err < 0) {
+			dev_err(dev, "cannot request IRQ %d\n", client->irq);
+			goto exit_remove_files;
+		}
+	}
+
 	return 0;
 
 exit_remove_files:
 	lm90_remove_files(client, data);
 exit_restore:
 	lm90_restore_conf(client, data);
+	regulator_disable(data->regulator);
+
 	return err;
 }
 
@@ -1483,49 +1632,33 @@ static int lm90_remove(struct i2c_client *client)
 	hwmon_device_unregister(data->hwmon_dev);
 	lm90_remove_files(client, data);
 	lm90_restore_conf(client, data);
+	regulator_disable(data->regulator);
 
 	return 0;
 }
 
 static void lm90_alert(struct i2c_client *client, unsigned int flag)
 {
-	struct lm90_data *data = i2c_get_clientdata(client);
-	u8 config, alarms, alarms2 = 0;
-
-	lm90_read_reg(client, LM90_REG_R_STATUS, &alarms);
-
-	if (data->kind == max6696)
-		lm90_read_reg(client, MAX6696_REG_R_STATUS2, &alarms2);
-
-	if ((alarms & 0x7f) == 0 && (alarms2 & 0xfe) == 0) {
-		dev_info(&client->dev, "Everything OK\n");
-	} else {
-		if (alarms & 0x61)
-			dev_warn(&client->dev,
-				 "temp%d out of range, please check!\n", 1);
-		if (alarms & 0x1a)
-			dev_warn(&client->dev,
-				 "temp%d out of range, please check!\n", 2);
-		if (alarms & 0x04)
-			dev_warn(&client->dev,
-				 "temp%d diode open, please check!\n", 2);
-
-		if (alarms2 & 0x18)
-			dev_warn(&client->dev,
-				 "temp%d out of range, please check!\n", 3);
+	u16 alarms;
 
+	if (lm90_is_tripped(client, &alarms)) {
 		/*
 		 * Disable ALERT# output, because these chips don't implement
 		 * SMBus alert correctly; they should only hold the alert line
 		 * low briefly.
 		 */
+		struct lm90_data *data = i2c_get_clientdata(client);
+
 		if ((data->flags & LM90_HAVE_BROKEN_ALERT)
 		 && (alarms & data->alert_alarms)) {
+			u8 config;
 			dev_dbg(&client->dev, "Disabling ALERT#\n");
 			lm90_read_reg(client, LM90_REG_R_CONFIG1, &config);
 			i2c_smbus_write_byte_data(client, LM90_REG_W_CONFIG1,
 						  config | 0x80);
 		}
+	} else {
+		dev_info(&client->dev, "Everything OK\n");
 	}
 }
 
diff --git a/drivers/i2c/busses/Kconfig b/drivers/i2c/busses/Kconfig
index cdcbd8368ed3..3b26129f6055 100644
--- a/drivers/i2c/busses/Kconfig
+++ b/drivers/i2c/busses/Kconfig
@@ -109,6 +109,7 @@ config I2C_I801
 	    Avoton (SOC)
 	    Wellsburg (PCH)
 	    Coleto Creek (PCH)
+	    Wildcat Point-LP (PCH)
 
 	  This driver can also be built as a module.  If so, the module
 	  will be called i2c-i801.
@@ -345,6 +346,16 @@ config I2C_BCM2835
 	  This support is also available as a module.  If so, the module
 	  will be called i2c-bcm2835.
 
+config I2C_BCM_KONA
+	tristate "BCM Kona I2C adapter"
+	depends on ARCH_BCM_MOBILE
+	default y
+	help
+	  If you say yes to this option, support will be included for the
+	  I2C interface on the Broadcom Kona family of processors.
+
+	  If you do not need KONA I2C inteface, say N.
+
 config I2C_BLACKFIN_TWI
 	tristate "Blackfin TWI I2C support"
 	depends on BLACKFIN
@@ -436,6 +447,13 @@ config I2C_EG20T
 	  ML7213/ML7223/ML7831 is companion chip for Intel Atom E6xx series.
 	  ML7213/ML7223/ML7831 is completely compatible for Intel EG20T PCH.
 
+config I2C_EXYNOS5
+	tristate "Exynos5 high-speed I2C driver"
+	depends on ARCH_EXYNOS5 && OF
+	help
+	  Say Y here to include support for high-speed I2C controller in the
+	  Exynos5 based Samsung SoCs.
+
 config I2C_GPIO
 	tristate "GPIO-based bitbanging I2C"
 	depends on GPIOLIB
@@ -665,7 +683,7 @@ config I2C_SH7760
 
 config I2C_SH_MOBILE
 	tristate "SuperH Mobile I2C Controller"
-	depends on SUPERH || ARCH_SHMOBILE
+	depends on SUPERH || ARM || COMPILE_TEST
 	help
 	  If you say yes to this option, support will be included for the
 	  built-in I2C interface on the Renesas SH-Mobile processor.
@@ -695,6 +713,16 @@ config I2C_SIRF
 	  This driver can also be built as a module.  If so, the module
 	  will be called i2c-sirf.
 
+config I2C_ST
+	tristate "STMicroelectronics SSC I2C support"
+	depends on ARCH_STI
+	help
+	  Enable this option to add support for STMicroelectronics SoCs
+	  hardware SSC (Synchronous Serial Controller) as an I2C controller.
+
+	  This driver can also be built as module. If so, the module
+	  will be called i2c-st.
+
 config I2C_STU300
 	tristate "ST Microelectronics DDC I2C interface"
 	depends on MACH_U300
@@ -768,7 +796,7 @@ config I2C_XLR
 
 config I2C_RCAR
 	tristate "Renesas R-Car I2C Controller"
-	depends on ARCH_SHMOBILE && I2C
+	depends on ARM || COMPILE_TEST
 	help
 	  If you say yes to this option, support will be included for the
 	  R-Car I2C controller.
diff --git a/drivers/i2c/busses/Makefile b/drivers/i2c/busses/Makefile
index d00997f3eb3b..c73eb0ea788e 100644
--- a/drivers/i2c/busses/Makefile
+++ b/drivers/i2c/busses/Makefile
@@ -42,6 +42,7 @@ i2c-designware-platform-objs := i2c-designware-platdrv.o
 obj-$(CONFIG_I2C_DESIGNWARE_PCI)	+= i2c-designware-pci.o
 i2c-designware-pci-objs := i2c-designware-pcidrv.o
 obj-$(CONFIG_I2C_EG20T)		+= i2c-eg20t.o
+obj-$(CONFIG_I2C_EXYNOS5)	+= i2c-exynos5.o
 obj-$(CONFIG_I2C_GPIO)		+= i2c-gpio.o
 obj-$(CONFIG_I2C_HIGHLANDER)	+= i2c-highlander.o
 obj-$(CONFIG_I2C_IBM_IIC)	+= i2c-ibm_iic.o
@@ -68,6 +69,7 @@ obj-$(CONFIG_I2C_SH7760)	+= i2c-sh7760.o
 obj-$(CONFIG_I2C_SH_MOBILE)	+= i2c-sh_mobile.o
 obj-$(CONFIG_I2C_SIMTEC)	+= i2c-simtec.o
 obj-$(CONFIG_I2C_SIRF)		+= i2c-sirf.o
+obj-$(CONFIG_I2C_ST)		+= i2c-st.o
 obj-$(CONFIG_I2C_STU300)	+= i2c-stu300.o
 obj-$(CONFIG_I2C_TEGRA)		+= i2c-tegra.o
 obj-$(CONFIG_I2C_VERSATILE)	+= i2c-versatile.o
@@ -87,6 +89,7 @@ obj-$(CONFIG_I2C_VIPERBOARD)	+= i2c-viperboard.o
 
 # Other I2C/SMBus bus drivers
 obj-$(CONFIG_I2C_ACORN)		+= i2c-acorn.o
+obj-$(CONFIG_I2C_BCM_KONA)	+= i2c-bcm-kona.o
 obj-$(CONFIG_I2C_ELEKTOR)	+= i2c-elektor.o
 obj-$(CONFIG_I2C_PCA_ISA)	+= i2c-pca-isa.o
 obj-$(CONFIG_I2C_SIBYTE)	+= i2c-sibyte.o
diff --git a/drivers/i2c/busses/i2c-at91.c b/drivers/i2c/busses/i2c-at91.c
index fd059308affa..8edba9de76df 100644
--- a/drivers/i2c/busses/i2c-at91.c
+++ b/drivers/i2c/busses/i2c-at91.c
@@ -371,7 +371,7 @@ static int at91_do_twi_transfer(struct at91_twi_dev *dev)
 	dev_dbg(dev->dev, "transfer: %s %d bytes.\n",
 		(dev->msg->flags & I2C_M_RD) ? "read" : "write", dev->buf_len);
 
-	INIT_COMPLETION(dev->cmd_complete);
+	reinit_completion(&dev->cmd_complete);
 	dev->transfer_status = 0;
 
 	if (!dev->buf_len) {
diff --git a/drivers/i2c/busses/i2c-bcm-kona.c b/drivers/i2c/busses/i2c-bcm-kona.c
new file mode 100644
index 000000000000..036cf03aeb61
--- /dev/null
+++ b/drivers/i2c/busses/i2c-bcm-kona.c
@@ -0,0 +1,909 @@
+/*
+ * Copyright (C) 2013 Broadcom Corporation
+ *
+ * This program is free software; you can redistribute it and/or
+ * modify it under the terms of the GNU General Public License as
+ * published by the Free Software Foundation version 2.
+ *
+ * This program is distributed "as is" WITHOUT ANY WARRANTY of any
+ * kind, whether express or implied; without even the implied warranty
+ * of MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+ * GNU General Public License for more details.
+ */
+
+#include <linux/device.h>
+#include <linux/kernel.h>
+#include <linux/module.h>
+#include <linux/sched.h>
+#include <linux/i2c.h>
+#include <linux/interrupt.h>
+#include <linux/platform_device.h>
+#include <linux/clk.h>
+#include <linux/io.h>
+#include <linux/clk.h>
+#include <linux/slab.h>
+
+/* Hardware register offsets and field defintions */
+#define CS_OFFSET				0x00000020
+#define CS_ACK_SHIFT				3
+#define CS_ACK_MASK				0x00000008
+#define CS_ACK_CMD_GEN_START			0x00000000
+#define CS_ACK_CMD_GEN_RESTART			0x00000001
+#define CS_CMD_SHIFT				1
+#define CS_CMD_CMD_NO_ACTION			0x00000000
+#define CS_CMD_CMD_START_RESTART		0x00000001
+#define CS_CMD_CMD_STOP				0x00000002
+#define CS_EN_SHIFT				0
+#define CS_EN_CMD_ENABLE_BSC			0x00000001
+
+#define TIM_OFFSET				0x00000024
+#define TIM_PRESCALE_SHIFT			6
+#define TIM_P_SHIFT				3
+#define TIM_NO_DIV_SHIFT			2
+#define TIM_DIV_SHIFT				0
+
+#define DAT_OFFSET				0x00000028
+
+#define TOUT_OFFSET				0x0000002c
+
+#define TXFCR_OFFSET				0x0000003c
+#define TXFCR_FIFO_FLUSH_MASK			0x00000080
+#define TXFCR_FIFO_EN_MASK			0x00000040
+
+#define IER_OFFSET				0x00000044
+#define IER_READ_COMPLETE_INT_MASK		0x00000010
+#define IER_I2C_INT_EN_MASK			0x00000008
+#define IER_FIFO_INT_EN_MASK			0x00000002
+#define IER_NOACK_EN_MASK			0x00000001
+
+#define ISR_OFFSET				0x00000048
+#define ISR_RESERVED_MASK			0xffffff60
+#define ISR_CMDBUSY_MASK			0x00000080
+#define ISR_READ_COMPLETE_MASK			0x00000010
+#define ISR_SES_DONE_MASK			0x00000008
+#define ISR_ERR_MASK				0x00000004
+#define ISR_TXFIFOEMPTY_MASK			0x00000002
+#define ISR_NOACK_MASK				0x00000001
+
+#define CLKEN_OFFSET				0x0000004C
+#define CLKEN_AUTOSENSE_OFF_MASK		0x00000080
+#define CLKEN_M_SHIFT				4
+#define CLKEN_N_SHIFT				1
+#define CLKEN_CLKEN_MASK			0x00000001
+
+#define FIFO_STATUS_OFFSET			0x00000054
+#define FIFO_STATUS_RXFIFO_EMPTY_MASK		0x00000004
+#define FIFO_STATUS_TXFIFO_EMPTY_MASK		0x00000010
+
+#define HSTIM_OFFSET				0x00000058
+#define HSTIM_HS_MODE_MASK			0x00008000
+#define HSTIM_HS_HOLD_SHIFT			10
+#define HSTIM_HS_HIGH_PHASE_SHIFT		5
+#define HSTIM_HS_SETUP_SHIFT			0
+
+#define PADCTL_OFFSET				0x0000005c
+#define PADCTL_PAD_OUT_EN_MASK			0x00000004
+
+#define RXFCR_OFFSET				0x00000068
+#define RXFCR_NACK_EN_SHIFT			7
+#define RXFCR_READ_COUNT_SHIFT			0
+#define RXFIFORDOUT_OFFSET			0x0000006c
+
+/* Locally used constants */
+#define MAX_RX_FIFO_SIZE		64U /* bytes */
+#define MAX_TX_FIFO_SIZE		64U /* bytes */
+
+#define STD_EXT_CLK_FREQ		13000000UL
+#define HS_EXT_CLK_FREQ			104000000UL
+
+#define MASTERCODE			0x08 /* Mastercodes are 0000_1xxxb */
+
+#define I2C_TIMEOUT			100 /* msecs */
+
+/* Operations that can be commanded to the controller */
+enum bcm_kona_cmd_t {
+	BCM_CMD_NOACTION = 0,
+	BCM_CMD_START,
+	BCM_CMD_RESTART,
+	BCM_CMD_STOP,
+};
+
+enum bus_speed_index {
+	BCM_SPD_100K = 0,
+	BCM_SPD_400K,
+	BCM_SPD_1MHZ,
+};
+
+enum hs_bus_speed_index {
+	BCM_SPD_3P4MHZ = 0,
+};
+
+/* Internal divider settings for standard mode, fast mode and fast mode plus */
+struct bus_speed_cfg {
+	uint8_t time_m;		/* Number of cycles for setup time */
+	uint8_t time_n;		/* Number of cycles for hold time */
+	uint8_t prescale;	/* Prescale divider */
+	uint8_t time_p;		/* Timing coefficient */
+	uint8_t no_div;		/* Disable clock divider */
+	uint8_t time_div;	/* Post-prescale divider */
+};
+
+/* Internal divider settings for high-speed mode */
+struct hs_bus_speed_cfg {
+	uint8_t hs_hold;	/* Number of clock cycles SCL stays low until
+				   the end of bit period */
+	uint8_t hs_high_phase;	/* Number of clock cycles SCL stays high
+				   before it falls */
+	uint8_t hs_setup;	/* Number of clock cycles SCL stays low
+				   before it rises  */
+	uint8_t prescale;	/* Prescale divider */
+	uint8_t time_p;		/* Timing coefficient */
+	uint8_t no_div;		/* Disable clock divider */
+	uint8_t time_div;	/* Post-prescale divider */
+};
+
+static const struct bus_speed_cfg std_cfg_table[] = {
+	[BCM_SPD_100K] = {0x01, 0x01, 0x03, 0x06, 0x00, 0x02},
+	[BCM_SPD_400K] = {0x05, 0x01, 0x03, 0x05, 0x01, 0x02},
+	[BCM_SPD_1MHZ] = {0x01, 0x01, 0x03, 0x01, 0x01, 0x03},
+};
+
+static const struct hs_bus_speed_cfg hs_cfg_table[] = {
+	[BCM_SPD_3P4MHZ] = {0x01, 0x08, 0x14, 0x00, 0x06, 0x01, 0x00},
+};
+
+struct bcm_kona_i2c_dev {
+	struct device *device;
+
+	void __iomem *base;
+	int irq;
+	struct clk *external_clk;
+
+	struct i2c_adapter adapter;
+
+	struct completion done;
+
+	const struct bus_speed_cfg *std_cfg;
+	const struct hs_bus_speed_cfg *hs_cfg;
+};
+
+static void bcm_kona_i2c_send_cmd_to_ctrl(struct bcm_kona_i2c_dev *dev,
+					  enum bcm_kona_cmd_t cmd)
+{
+	dev_dbg(dev->device, "%s, %d\n", __func__, cmd);
+
+	switch (cmd) {
+	case BCM_CMD_NOACTION:
+		writel((CS_CMD_CMD_NO_ACTION << CS_CMD_SHIFT) |
+		       (CS_EN_CMD_ENABLE_BSC << CS_EN_SHIFT),
+		       dev->base + CS_OFFSET);
+		break;
+
+	case BCM_CMD_START:
+		writel((CS_ACK_CMD_GEN_START << CS_ACK_SHIFT) |
+		       (CS_CMD_CMD_START_RESTART << CS_CMD_SHIFT) |
+		       (CS_EN_CMD_ENABLE_BSC << CS_EN_SHIFT),
+		       dev->base + CS_OFFSET);
+		break;
+
+	case BCM_CMD_RESTART:
+		writel((CS_ACK_CMD_GEN_RESTART << CS_ACK_SHIFT) |
+		       (CS_CMD_CMD_START_RESTART << CS_CMD_SHIFT) |
+		       (CS_EN_CMD_ENABLE_BSC << CS_EN_SHIFT),
+		       dev->base + CS_OFFSET);
+		break;
+
+	case BCM_CMD_STOP:
+		writel((CS_CMD_CMD_STOP << CS_CMD_SHIFT) |
+		       (CS_EN_CMD_ENABLE_BSC << CS_EN_SHIFT),
+		       dev->base + CS_OFFSET);
+		break;
+
+	default:
+		dev_err(dev->device, "Unknown command %d\n", cmd);
+	}
+}
+
+static void bcm_kona_i2c_enable_clock(struct bcm_kona_i2c_dev *dev)
+{
+	writel(readl(dev->base + CLKEN_OFFSET) | CLKEN_CLKEN_MASK,
+	       dev->base + CLKEN_OFFSET);
+}
+
+static void bcm_kona_i2c_disable_clock(struct bcm_kona_i2c_dev *dev)
+{
+	writel(readl(dev->base + CLKEN_OFFSET) & ~CLKEN_CLKEN_MASK,
+	       dev->base + CLKEN_OFFSET);
+}
+
+static irqreturn_t bcm_kona_i2c_isr(int irq, void *devid)
+{
+	struct bcm_kona_i2c_dev *dev = devid;
+	uint32_t status = readl(dev->base + ISR_OFFSET);
+
+	if ((status & ~ISR_RESERVED_MASK) == 0)
+		return IRQ_NONE;
+
+	/* Must flush the TX FIFO when NAK detected */
+	if (status & ISR_NOACK_MASK)
+		writel(TXFCR_FIFO_FLUSH_MASK | TXFCR_FIFO_EN_MASK,
+		       dev->base + TXFCR_OFFSET);
+
+	writel(status & ~ISR_RESERVED_MASK, dev->base + ISR_OFFSET);
+	complete_all(&dev->done);
+
+	return IRQ_HANDLED;
+}
+
+/* Wait for ISR_CMDBUSY_MASK to go low before writing to CS, DAT, or RCD */
+static int bcm_kona_i2c_wait_if_busy(struct bcm_kona_i2c_dev *dev)
+{
+	unsigned long timeout = jiffies + msecs_to_jiffies(I2C_TIMEOUT);
+
+	while (readl(dev->base + ISR_OFFSET) & ISR_CMDBUSY_MASK)
+		if (time_after(jiffies, timeout)) {
+			dev_err(dev->device, "CMDBUSY timeout\n");
+			return -ETIMEDOUT;
+		}
+
+	return 0;
+}
+
+/* Send command to I2C bus */
+static int bcm_kona_send_i2c_cmd(struct bcm_kona_i2c_dev *dev,
+				 enum bcm_kona_cmd_t cmd)
+{
+	int rc;
+	unsigned long time_left = msecs_to_jiffies(I2C_TIMEOUT);
+
+	/* Make sure the hardware is ready */
+	rc = bcm_kona_i2c_wait_if_busy(dev);
+	if (rc < 0)
+		return rc;
+
+	/* Unmask the session done interrupt */
+	writel(IER_I2C_INT_EN_MASK, dev->base + IER_OFFSET);
+
+	/* Mark as incomplete before sending the command */
+	reinit_completion(&dev->done);
+
+	/* Send the command */
+	bcm_kona_i2c_send_cmd_to_ctrl(dev, cmd);
+
+	/* Wait for transaction to finish or timeout */
+	time_left = wait_for_completion_timeout(&dev->done, time_left);
+
+	/* Mask all interrupts */
+	writel(0, dev->base + IER_OFFSET);
+
+	if (!time_left) {
+		dev_err(dev->device, "controller timed out\n");
+		rc = -ETIMEDOUT;
+	}
+
+	/* Clear command */
+	bcm_kona_i2c_send_cmd_to_ctrl(dev, BCM_CMD_NOACTION);
+
+	return rc;
+}
+
+/* Read a single RX FIFO worth of data from the i2c bus */
+static int bcm_kona_i2c_read_fifo_single(struct bcm_kona_i2c_dev *dev,
+					 uint8_t *buf, unsigned int len,
+					 unsigned int last_byte_nak)
+{
+	unsigned long time_left = msecs_to_jiffies(I2C_TIMEOUT);
+
+	/* Mark as incomplete before starting the RX FIFO */
+	reinit_completion(&dev->done);
+
+	/* Unmask the read complete interrupt */
+	writel(IER_READ_COMPLETE_INT_MASK, dev->base + IER_OFFSET);
+
+	/* Start the RX FIFO */
+	writel((last_byte_nak << RXFCR_NACK_EN_SHIFT) |
+	       (len << RXFCR_READ_COUNT_SHIFT),
+		dev->base + RXFCR_OFFSET);
+
+	/* Wait for FIFO read to complete */
+	time_left = wait_for_completion_timeout(&dev->done, time_left);
+
+	/* Mask all interrupts */
+	writel(0, dev->base + IER_OFFSET);
+
+	if (!time_left) {
+		dev_err(dev->device, "RX FIFO time out\n");
+		return -EREMOTEIO;
+	}
+
+	/* Read data from FIFO */
+	for (; len > 0; len--, buf++)
+		*buf = readl(dev->base + RXFIFORDOUT_OFFSET);
+
+	return 0;
+}
+
+/* Read any amount of data using the RX FIFO from the i2c bus */
+static int bcm_kona_i2c_read_fifo(struct bcm_kona_i2c_dev *dev,
+				  struct i2c_msg *msg)
+{
+	unsigned int bytes_to_read = MAX_RX_FIFO_SIZE;
+	unsigned int last_byte_nak = 0;
+	unsigned int bytes_read = 0;
+	int rc;
+
+	uint8_t *tmp_buf = msg->buf;
+
+	while (bytes_read < msg->len) {
+		if (msg->len - bytes_read <= MAX_RX_FIFO_SIZE) {
+			last_byte_nak = 1; /* NAK last byte of transfer */
+			bytes_to_read = msg->len - bytes_read;
+		}
+
+		rc = bcm_kona_i2c_read_fifo_single(dev, tmp_buf, bytes_to_read,
+						   last_byte_nak);
+		if (rc < 0)
+			return -EREMOTEIO;
+
+		bytes_read += bytes_to_read;
+		tmp_buf += bytes_to_read;
+	}
+
+	return 0;
+}
+
+/* Write a single byte of data to the i2c bus */
+static int bcm_kona_i2c_write_byte(struct bcm_kona_i2c_dev *dev, uint8_t data,
+				   unsigned int nak_expected)
+{
+	int rc;
+	unsigned long time_left = msecs_to_jiffies(I2C_TIMEOUT);
+	unsigned int nak_received;
+
+	/* Make sure the hardware is ready */
+	rc = bcm_kona_i2c_wait_if_busy(dev);
+	if (rc < 0)
+		return rc;
+
+	/* Clear pending session done interrupt */
+	writel(ISR_SES_DONE_MASK, dev->base + ISR_OFFSET);
+
+	/* Unmask the session done interrupt */
+	writel(IER_I2C_INT_EN_MASK, dev->base + IER_OFFSET);
+
+	/* Mark as incomplete before sending the data */
+	reinit_completion(&dev->done);
+
+	/* Send one byte of data */
+	writel(data, dev->base + DAT_OFFSET);
+
+	/* Wait for byte to be written */
+	time_left = wait_for_completion_timeout(&dev->done, time_left);
+
+	/* Mask all interrupts */
+	writel(0, dev->base + IER_OFFSET);
+
+	if (!time_left) {
+		dev_dbg(dev->device, "controller timed out\n");
+		return -ETIMEDOUT;
+	}
+
+	nak_received = readl(dev->base + CS_OFFSET) & CS_ACK_MASK ? 1 : 0;
+
+	if (nak_received ^ nak_expected) {
+		dev_dbg(dev->device, "unexpected NAK/ACK\n");
+		return -EREMOTEIO;
+	}
+
+	return 0;
+}
+
+/* Write a single TX FIFO worth of data to the i2c bus */
+static int bcm_kona_i2c_write_fifo_single(struct bcm_kona_i2c_dev *dev,
+					  uint8_t *buf, unsigned int len)
+{
+	int k;
+	unsigned long time_left = msecs_to_jiffies(I2C_TIMEOUT);
+	unsigned int fifo_status;
+
+	/* Mark as incomplete before sending data to the TX FIFO */
+	reinit_completion(&dev->done);
+
+	/* Unmask the fifo empty and nak interrupt */
+	writel(IER_FIFO_INT_EN_MASK | IER_NOACK_EN_MASK,
+	       dev->base + IER_OFFSET);
+
+	/* Disable IRQ to load a FIFO worth of data without interruption */
+	disable_irq(dev->irq);
+
+	/* Write data into FIFO */
+	for (k = 0; k < len; k++)
+		writel(buf[k], (dev->base + DAT_OFFSET));
+
+	/* Enable IRQ now that data has been loaded */
+	enable_irq(dev->irq);
+
+	/* Wait for FIFO to empty */
+	do {
+		time_left = wait_for_completion_timeout(&dev->done, time_left);
+		fifo_status = readl(dev->base + FIFO_STATUS_OFFSET);
+	} while (time_left && !(fifo_status & FIFO_STATUS_TXFIFO_EMPTY_MASK));
+
+	/* Mask all interrupts */
+	writel(0, dev->base + IER_OFFSET);
+
+	/* Check if there was a NAK */
+	if (readl(dev->base + CS_OFFSET) & CS_ACK_MASK) {
+		dev_err(dev->device, "unexpected NAK\n");
+		return -EREMOTEIO;
+	}
+
+	/* Check if a timeout occured */
+	if (!time_left) {
+		dev_err(dev->device, "completion timed out\n");
+		return -EREMOTEIO;
+	}
+
+	return 0;
+}
+
+
+/* Write any amount of data using TX FIFO to the i2c bus */
+static int bcm_kona_i2c_write_fifo(struct bcm_kona_i2c_dev *dev,
+				   struct i2c_msg *msg)
+{
+	unsigned int bytes_to_write = MAX_TX_FIFO_SIZE;
+	unsigned int bytes_written = 0;
+	int rc;
+
+	uint8_t *tmp_buf = msg->buf;
+
+	while (bytes_written < msg->len) {
+		if (msg->len - bytes_written <= MAX_TX_FIFO_SIZE)
+			bytes_to_write = msg->len - bytes_written;
+
+		rc = bcm_kona_i2c_write_fifo_single(dev, tmp_buf,
+						    bytes_to_write);
+		if (rc < 0)
+			return -EREMOTEIO;
+
+		bytes_written += bytes_to_write;
+		tmp_buf += bytes_to_write;
+	}
+
+	return 0;
+}
+
+/* Send i2c address */
+static int bcm_kona_i2c_do_addr(struct bcm_kona_i2c_dev *dev,
+				     struct i2c_msg *msg)
+{
+	unsigned char addr;
+
+	if (msg->flags & I2C_M_TEN) {
+		/* First byte is 11110XX0 where XX is upper 2 bits */
+		addr = 0xF0 | ((msg->addr & 0x300) >> 7);
+		if (bcm_kona_i2c_write_byte(dev, addr, 0) < 0)
+			return -EREMOTEIO;
+
+		/* Second byte is the remaining 8 bits */
+		addr = msg->addr & 0xFF;
+		if (bcm_kona_i2c_write_byte(dev, addr, 0) < 0)
+			return -EREMOTEIO;
+
+		if (msg->flags & I2C_M_RD) {
+			/* For read, send restart command */
+			if (bcm_kona_send_i2c_cmd(dev, BCM_CMD_RESTART) < 0)
+				return -EREMOTEIO;
+
+			/* Then re-send the first byte with the read bit set */
+			addr = 0xF0 | ((msg->addr & 0x300) >> 7) | 0x01;
+			if (bcm_kona_i2c_write_byte(dev, addr, 0) < 0)
+				return -EREMOTEIO;
+		}
+	} else {
+		addr = msg->addr << 1;
+
+		if (msg->flags & I2C_M_RD)
+			addr |= 1;
+
+		if (bcm_kona_i2c_write_byte(dev, addr, 0) < 0)
+			return -EREMOTEIO;
+	}
+
+	return 0;
+}
+
+static void bcm_kona_i2c_enable_autosense(struct bcm_kona_i2c_dev *dev)
+{
+	writel(readl(dev->base + CLKEN_OFFSET) & ~CLKEN_AUTOSENSE_OFF_MASK,
+	       dev->base + CLKEN_OFFSET);
+}
+
+static void bcm_kona_i2c_config_timing(struct bcm_kona_i2c_dev *dev)
+{
+	writel(readl(dev->base + HSTIM_OFFSET) & ~HSTIM_HS_MODE_MASK,
+	       dev->base + HSTIM_OFFSET);
+
+	writel((dev->std_cfg->prescale << TIM_PRESCALE_SHIFT) |
+	       (dev->std_cfg->time_p << TIM_P_SHIFT) |
+	       (dev->std_cfg->no_div << TIM_NO_DIV_SHIFT) |
+	       (dev->std_cfg->time_div	<< TIM_DIV_SHIFT),
+	       dev->base + TIM_OFFSET);
+
+	writel((dev->std_cfg->time_m << CLKEN_M_SHIFT) |
+	       (dev->std_cfg->time_n << CLKEN_N_SHIFT) |
+	       CLKEN_CLKEN_MASK,
+	       dev->base + CLKEN_OFFSET);
+}
+
+static void bcm_kona_i2c_config_timing_hs(struct bcm_kona_i2c_dev *dev)
+{
+	writel((dev->hs_cfg->prescale << TIM_PRESCALE_SHIFT) |
+	       (dev->hs_cfg->time_p << TIM_P_SHIFT) |
+	       (dev->hs_cfg->no_div << TIM_NO_DIV_SHIFT) |
+	       (dev->hs_cfg->time_div << TIM_DIV_SHIFT),
+	       dev->base + TIM_OFFSET);
+
+	writel((dev->hs_cfg->hs_hold << HSTIM_HS_HOLD_SHIFT) |
+	       (dev->hs_cfg->hs_high_phase << HSTIM_HS_HIGH_PHASE_SHIFT) |
+	       (dev->hs_cfg->hs_setup << HSTIM_HS_SETUP_SHIFT),
+	       dev->base + HSTIM_OFFSET);
+
+	writel(readl(dev->base + HSTIM_OFFSET) | HSTIM_HS_MODE_MASK,
+	       dev->base + HSTIM_OFFSET);
+}
+
+static int bcm_kona_i2c_switch_to_hs(struct bcm_kona_i2c_dev *dev)
+{
+	int rc;
+
+	/* Send mastercode at standard speed */
+	rc = bcm_kona_i2c_write_byte(dev, MASTERCODE, 1);
+	if (rc < 0) {
+		pr_err("High speed handshake failed\n");
+		return rc;
+	}
+
+	/* Configure external clock to higher frequency */
+	rc = clk_set_rate(dev->external_clk, HS_EXT_CLK_FREQ);
+	if (rc) {
+		dev_err(dev->device, "%s: clk_set_rate returned %d\n",
+			__func__, rc);
+		return rc;
+	}
+
+	/* Reconfigure internal dividers */
+	bcm_kona_i2c_config_timing_hs(dev);
+
+	/* Send a restart command */
+	rc = bcm_kona_send_i2c_cmd(dev, BCM_CMD_RESTART);
+	if (rc < 0)
+		dev_err(dev->device, "High speed restart command failed\n");
+
+	return rc;
+}
+
+static int bcm_kona_i2c_switch_to_std(struct bcm_kona_i2c_dev *dev)
+{
+	int rc;
+
+	/* Reconfigure internal dividers */
+	bcm_kona_i2c_config_timing(dev);
+
+	/* Configure external clock to lower frequency */
+	rc = clk_set_rate(dev->external_clk, STD_EXT_CLK_FREQ);
+	if (rc) {
+		dev_err(dev->device, "%s: clk_set_rate returned %d\n",
+			__func__, rc);
+	}
+
+	return rc;
+}
+
+/* Master transfer function */
+static int bcm_kona_i2c_xfer(struct i2c_adapter *adapter,
+			     struct i2c_msg msgs[], int num)
+{
+	struct bcm_kona_i2c_dev *dev = i2c_get_adapdata(adapter);
+	struct i2c_msg *pmsg;
+	int rc = 0;
+	int i;
+
+	rc = clk_prepare_enable(dev->external_clk);
+	if (rc) {
+		dev_err(dev->device, "%s: peri clock enable failed. err %d\n",
+			__func__, rc);
+		return rc;
+	}
+
+	/* Enable pad output */
+	writel(0, dev->base + PADCTL_OFFSET);
+
+	/* Enable internal clocks */
+	bcm_kona_i2c_enable_clock(dev);
+
+	/* Send start command */
+	rc = bcm_kona_send_i2c_cmd(dev, BCM_CMD_START);
+	if (rc < 0) {
+		dev_err(dev->device, "Start command failed rc = %d\n", rc);
+		goto xfer_disable_pad;
+	}
+
+	/* Switch to high speed if applicable */
+	if (dev->hs_cfg) {
+		rc = bcm_kona_i2c_switch_to_hs(dev);
+		if (rc < 0)
+			goto xfer_send_stop;
+	}
+
+	/* Loop through all messages */
+	for (i = 0; i < num; i++) {
+		pmsg = &msgs[i];
+
+		/* Send restart for subsequent messages */
+		if ((i != 0) && ((pmsg->flags & I2C_M_NOSTART) == 0)) {
+			rc = bcm_kona_send_i2c_cmd(dev, BCM_CMD_RESTART);
+			if (rc < 0) {
+				dev_err(dev->device,
+					"restart cmd failed rc = %d\n", rc);
+					goto xfer_send_stop;
+			}
+		}
+
+		/* Send slave address */
+		if (!(pmsg->flags & I2C_M_NOSTART)) {
+			rc = bcm_kona_i2c_do_addr(dev, pmsg);
+			if (rc < 0) {
+				dev_err(dev->device,
+					"NAK from addr %2.2x msg#%d rc = %d\n",
+					pmsg->addr, i, rc);
+				goto xfer_send_stop;
+			}
+		}
+
+		/* Perform data transfer */
+		if (pmsg->flags & I2C_M_RD) {
+			rc = bcm_kona_i2c_read_fifo(dev, pmsg);
+			if (rc < 0) {
+				dev_err(dev->device, "read failure\n");
+				goto xfer_send_stop;
+			}
+		} else {
+			rc = bcm_kona_i2c_write_fifo(dev, pmsg);
+			if (rc < 0) {
+				dev_err(dev->device, "write failure");
+				goto xfer_send_stop;
+			}
+		}
+	}
+
+	rc = num;
+
+xfer_send_stop:
+	/* Send a STOP command */
+	bcm_kona_send_i2c_cmd(dev, BCM_CMD_STOP);
+
+	/* Return from high speed if applicable */
+	if (dev->hs_cfg) {
+		int hs_rc = bcm_kona_i2c_switch_to_std(dev);
+
+		if (hs_rc)
+			rc = hs_rc;
+	}
+
+xfer_disable_pad:
+	/* Disable pad output */
+	writel(PADCTL_PAD_OUT_EN_MASK, dev->base + PADCTL_OFFSET);
+
+	/* Stop internal clock */
+	bcm_kona_i2c_disable_clock(dev);
+
+	clk_disable_unprepare(dev->external_clk);
+
+	return rc;
+}
+
+static uint32_t bcm_kona_i2c_functionality(struct i2c_adapter *adap)
+{
+	return I2C_FUNC_I2C | I2C_FUNC_SMBUS_EMUL | I2C_FUNC_10BIT_ADDR |
+	    I2C_FUNC_NOSTART;
+}
+
+static const struct i2c_algorithm bcm_algo = {
+	.master_xfer = bcm_kona_i2c_xfer,
+	.functionality = bcm_kona_i2c_functionality,
+};
+
+static int bcm_kona_i2c_assign_bus_speed(struct bcm_kona_i2c_dev *dev)
+{
+	unsigned int bus_speed;
+	int ret = of_property_read_u32(dev->device->of_node, "clock-frequency",
+				       &bus_speed);
+	if (ret < 0) {
+		dev_err(dev->device, "missing clock-frequency property\n");
+		return -ENODEV;
+	}
+
+	switch (bus_speed) {
+	case 100000:
+		dev->std_cfg = &std_cfg_table[BCM_SPD_100K];
+		break;
+	case 400000:
+		dev->std_cfg = &std_cfg_table[BCM_SPD_400K];
+		break;
+	case 1000000:
+		dev->std_cfg = &std_cfg_table[BCM_SPD_1MHZ];
+		break;
+	case 3400000:
+		/* Send mastercode at 100k */
+		dev->std_cfg = &std_cfg_table[BCM_SPD_100K];
+		dev->hs_cfg = &hs_cfg_table[BCM_SPD_3P4MHZ];
+		break;
+	default:
+		pr_err("%d hz bus speed not supported\n", bus_speed);
+		pr_err("Valid speeds are 100khz, 400khz, 1mhz, and 3.4mhz\n");
+		return -EINVAL;
+	}
+
+	return 0;
+}
+
+static int bcm_kona_i2c_probe(struct platform_device *pdev)
+{
+	int rc = 0;
+	struct bcm_kona_i2c_dev *dev;
+	struct i2c_adapter *adap;
+	struct resource *iomem;
+
+	/* Allocate memory for private data structure */
+	dev = devm_kzalloc(&pdev->dev, sizeof(*dev), GFP_KERNEL);
+	if (!dev)
+		return -ENOMEM;
+
+	platform_set_drvdata(pdev, dev);
+	dev->device = &pdev->dev;
+	init_completion(&dev->done);
+
+	/* Map hardware registers */
+	iomem = platform_get_resource(pdev, IORESOURCE_MEM, 0);
+	dev->base = devm_ioremap_resource(dev->device, iomem);
+	if (IS_ERR(dev->base))
+		return -ENOMEM;
+
+	/* Get and enable external clock */
+	dev->external_clk = devm_clk_get(dev->device, NULL);
+	if (IS_ERR(dev->external_clk)) {
+		dev_err(dev->device, "couldn't get clock\n");
+		return -ENODEV;
+	}
+
+	rc = clk_set_rate(dev->external_clk, STD_EXT_CLK_FREQ);
+	if (rc) {
+		dev_err(dev->device, "%s: clk_set_rate returned %d\n",
+			__func__, rc);
+		return rc;
+	}
+
+	rc = clk_prepare_enable(dev->external_clk);
+	if (rc) {
+		dev_err(dev->device, "couldn't enable clock\n");
+		return rc;
+	}
+
+	/* Parse bus speed */
+	rc = bcm_kona_i2c_assign_bus_speed(dev);
+	if (rc)
+		goto probe_disable_clk;
+
+	/* Enable internal clocks */
+	bcm_kona_i2c_enable_clock(dev);
+
+	/* Configure internal dividers */
+	bcm_kona_i2c_config_timing(dev);
+
+	/* Disable timeout */
+	writel(0, dev->base + TOUT_OFFSET);
+
+	/* Enable autosense */
+	bcm_kona_i2c_enable_autosense(dev);
+
+	/* Enable TX FIFO */
+	writel(TXFCR_FIFO_FLUSH_MASK | TXFCR_FIFO_EN_MASK,
+	       dev->base + TXFCR_OFFSET);
+
+	/* Mask all interrupts */
+	writel(0, dev->base + IER_OFFSET);
+
+	/* Clear all pending interrupts */
+	writel(ISR_CMDBUSY_MASK |
+	       ISR_READ_COMPLETE_MASK |
+	       ISR_SES_DONE_MASK |
+	       ISR_ERR_MASK |
+	       ISR_TXFIFOEMPTY_MASK |
+	       ISR_NOACK_MASK,
+	       dev->base + ISR_OFFSET);
+
+	/* Get the interrupt number */
+	dev->irq = platform_get_irq(pdev, 0);
+	if (dev->irq < 0) {
+		dev_err(dev->device, "no irq resource\n");
+		rc = -ENODEV;
+		goto probe_disable_clk;
+	}
+
+	/* register the ISR handler */
+	rc = devm_request_irq(&pdev->dev, dev->irq, bcm_kona_i2c_isr,
+			      IRQF_SHARED, pdev->name, dev);
+	if (rc) {
+		dev_err(dev->device, "failed to request irq %i\n", dev->irq);
+		goto probe_disable_clk;
+	}
+
+	/* Enable the controller but leave it idle */
+	bcm_kona_i2c_send_cmd_to_ctrl(dev, BCM_CMD_NOACTION);
+
+	/* Disable pad output */
+	writel(PADCTL_PAD_OUT_EN_MASK, dev->base + PADCTL_OFFSET);
+
+	/* Disable internal clock */
+	bcm_kona_i2c_disable_clock(dev);
+
+	/* Disable external clock */
+	clk_disable_unprepare(dev->external_clk);
+
+	/* Add the i2c adapter */
+	adap = &dev->adapter;
+	i2c_set_adapdata(adap, dev);
+	adap->owner = THIS_MODULE;
+	strlcpy(adap->name, "Broadcom I2C adapter", sizeof(adap->name));
+	adap->algo = &bcm_algo;
+	adap->dev.parent = &pdev->dev;
+	adap->dev.of_node = pdev->dev.of_node;
+
+	rc = i2c_add_adapter(adap);
+	if (rc) {
+		dev_err(dev->device, "failed to add adapter\n");
+		return rc;
+	}
+
+	dev_info(dev->device, "device registered successfully\n");
+
+	return 0;
+
+probe_disable_clk:
+	bcm_kona_i2c_disable_clock(dev);
+	clk_disable_unprepare(dev->external_clk);
+
+	return rc;
+}
+
+static int bcm_kona_i2c_remove(struct platform_device *pdev)
+{
+	struct bcm_kona_i2c_dev *dev = platform_get_drvdata(pdev);
+
+	i2c_del_adapter(&dev->adapter);
+
+	return 0;
+}
+
+static const struct of_device_id bcm_kona_i2c_of_match[] = {
+	{.compatible = "brcm,kona-i2c",},
+	{},
+};
+MODULE_DEVICE_TABLE(of, kona_i2c_of_match);
+
+static struct platform_driver bcm_kona_i2c_driver = {
+	.driver = {
+		   .name = "bcm-kona-i2c",
+		   .owner = THIS_MODULE,
+		   .of_match_table = bcm_kona_i2c_of_match,
+		   },
+	.probe = bcm_kona_i2c_probe,
+	.remove = bcm_kona_i2c_remove,
+};
+module_platform_driver(bcm_kona_i2c_driver);
+
+MODULE_AUTHOR("Tim Kryger <tkryger@broadcom.com>");
+MODULE_DESCRIPTION("Broadcom Kona I2C Driver");
+MODULE_LICENSE("GPL v2");
diff --git a/drivers/i2c/busses/i2c-bcm2835.c b/drivers/i2c/busses/i2c-bcm2835.c
index ea4b08fc3353..d7e8600f31fb 100644
--- a/drivers/i2c/busses/i2c-bcm2835.c
+++ b/drivers/i2c/busses/i2c-bcm2835.c
@@ -151,7 +151,7 @@ static int bcm2835_i2c_xfer_msg(struct bcm2835_i2c_dev *i2c_dev,
 
 	i2c_dev->msg_buf = msg->buf;
 	i2c_dev->msg_buf_remaining = msg->len;
-	INIT_COMPLETION(i2c_dev->completion);
+	reinit_completion(&i2c_dev->completion);
 
 	bcm2835_i2c_writel(i2c_dev, BCM2835_I2C_C, BCM2835_I2C_C_CLEAR);
 
diff --git a/drivers/i2c/busses/i2c-bfin-twi.c b/drivers/i2c/busses/i2c-bfin-twi.c
index 35a473ba3d81..3b9bd9a3f2b0 100644
--- a/drivers/i2c/busses/i2c-bfin-twi.c
+++ b/drivers/i2c/busses/i2c-bfin-twi.c
@@ -675,7 +675,7 @@ static int i2c_bfin_twi_probe(struct platform_device *pdev)
 	p_adap->retries = 3;
 
 	rc = peripheral_request_list(
-			(unsigned short *)dev_get_platdata(&pdev->dev),
+			dev_get_platdata(&pdev->dev),
 			"i2c-bfin-twi");
 	if (rc) {
 		dev_err(&pdev->dev, "Can't setup pin mux!\n");
@@ -723,7 +723,7 @@ out_error_add_adapter:
 	free_irq(iface->irq, iface);
 out_error_req_irq:
 out_error_no_irq:
-	peripheral_free_list((unsigned short *)dev_get_platdata(&pdev->dev));
+	peripheral_free_list(dev_get_platdata(&pdev->dev));
 out_error_pin_mux:
 	iounmap(iface->regs_base);
 out_error_ioremap:
@@ -739,7 +739,7 @@ static int i2c_bfin_twi_remove(struct platform_device *pdev)
 
 	i2c_del_adapter(&(iface->adap));
 	free_irq(iface->irq, iface);
-	peripheral_free_list((unsigned short *)dev_get_platdata(&pdev->dev));
+	peripheral_free_list(dev_get_platdata(&pdev->dev));
 	iounmap(iface->regs_base);
 	kfree(iface);
 
diff --git a/drivers/i2c/busses/i2c-cbus-gpio.c b/drivers/i2c/busses/i2c-cbus-gpio.c
index 2d46f13adfdf..ce7ffba2b020 100644
--- a/drivers/i2c/busses/i2c-cbus-gpio.c
+++ b/drivers/i2c/busses/i2c-cbus-gpio.c
@@ -246,6 +246,7 @@ static int cbus_i2c_probe(struct platform_device *pdev)
 	adapter->owner		= THIS_MODULE;
 	adapter->class		= I2C_CLASS_HWMON;
 	adapter->dev.parent	= &pdev->dev;
+	adapter->dev.of_node	= pdev->dev.of_node;
 	adapter->nr		= pdev->id;
 	adapter->timeout	= HZ;
 	adapter->algo		= &cbus_i2c_algo;
@@ -289,6 +290,7 @@ static struct platform_driver cbus_i2c_driver = {
 	.driver	= {
 		.owner	= THIS_MODULE,
 		.name	= "i2c-cbus-gpio",
+		.of_match_table = of_match_ptr(i2c_cbus_dt_ids),
 	},
 };
 module_platform_driver(cbus_i2c_driver);
diff --git a/drivers/i2c/busses/i2c-davinci.c b/drivers/i2c/busses/i2c-davinci.c
index 132369fad4e0..ff05d9fef4a8 100644
--- a/drivers/i2c/busses/i2c-davinci.c
+++ b/drivers/i2c/busses/i2c-davinci.c
@@ -323,7 +323,7 @@ i2c_davinci_xfer_msg(struct i2c_adapter *adap, struct i2c_msg *msg, int stop)
 
 	davinci_i2c_write_reg(dev, DAVINCI_I2C_CNT_REG, dev->buf_len);
 
-	INIT_COMPLETION(dev->cmd_complete);
+	reinit_completion(&dev->cmd_complete);
 	dev->cmd_err = 0;
 
 	/* Take I2C out of reset and configure it as master */
@@ -795,7 +795,7 @@ static struct platform_driver davinci_i2c_driver = {
 		.name	= "i2c_davinci",
 		.owner	= THIS_MODULE,
 		.pm	= davinci_i2c_pm_ops,
-		.of_match_table = of_match_ptr(davinci_i2c_of_match),
+		.of_match_table = davinci_i2c_of_match,
 	},
 };
 
diff --git a/drivers/i2c/busses/i2c-designware-core.c b/drivers/i2c/busses/i2c-designware-core.c
index 5888feef1ac5..e89e3e2145e5 100644
--- a/drivers/i2c/busses/i2c-designware-core.c
+++ b/drivers/i2c/busses/i2c-designware-core.c
@@ -613,7 +613,7 @@ i2c_dw_xfer(struct i2c_adapter *adap, struct i2c_msg msgs[], int num)
 	mutex_lock(&dev->lock);
 	pm_runtime_get_sync(dev->dev);
 
-	INIT_COMPLETION(dev->cmd_complete);
+	reinit_completion(&dev->cmd_complete);
 	dev->msgs = msgs;
 	dev->msgs_num = num;
 	dev->cmd_err = 0;
diff --git a/drivers/i2c/busses/i2c-designware-platdrv.c b/drivers/i2c/busses/i2c-designware-platdrv.c
index 0aa01136f8d9..d0bdac0498ce 100644
--- a/drivers/i2c/busses/i2c-designware-platdrv.c
+++ b/drivers/i2c/busses/i2c-designware-platdrv.c
@@ -103,6 +103,8 @@ static int dw_i2c_acpi_configure(struct platform_device *pdev)
 static const struct acpi_device_id dw_i2c_acpi_match[] = {
 	{ "INT33C2", 0 },
 	{ "INT33C3", 0 },
+	{ "INT3432", 0 },
+	{ "INT3433", 0 },
 	{ "80860F41", 0 },
 	{ }
 };
diff --git a/drivers/i2c/busses/i2c-eg20t.c b/drivers/i2c/busses/i2c-eg20t.c
index 0f3752967c4b..ff15ae90aaf5 100644
--- a/drivers/i2c/busses/i2c-eg20t.c
+++ b/drivers/i2c/busses/i2c-eg20t.c
@@ -312,24 +312,6 @@ static void pch_i2c_start(struct i2c_algo_pch_data *adap)
 }
 
 /**
- * pch_i2c_getack() - to confirm ACK/NACK
- * @adap:	Pointer to struct i2c_algo_pch_data.
- */
-static s32 pch_i2c_getack(struct i2c_algo_pch_data *adap)
-{
-	u32 reg_val;
-	void __iomem *p = adap->pch_base_address;
-	reg_val = ioread32(p + PCH_I2CSR) & PCH_GETACK;
-
-	if (reg_val != 0) {
-		pch_err(adap, "return%d\n", -EPROTO);
-		return -EPROTO;
-	}
-
-	return 0;
-}
-
-/**
  * pch_i2c_stop() - generate stop condition in normal mode.
  * @adap:	Pointer to struct i2c_algo_pch_data.
  */
@@ -344,6 +326,7 @@ static void pch_i2c_stop(struct i2c_algo_pch_data *adap)
 static int pch_i2c_wait_for_check_xfer(struct i2c_algo_pch_data *adap)
 {
 	long ret;
+	void __iomem *p = adap->pch_base_address;
 
 	ret = wait_event_timeout(pch_event,
 			(adap->pch_event_flag != 0), msecs_to_jiffies(1000));
@@ -366,10 +349,9 @@ static int pch_i2c_wait_for_check_xfer(struct i2c_algo_pch_data *adap)
 
 	adap->pch_event_flag = 0;
 
-	if (pch_i2c_getack(adap)) {
-		pch_dbg(adap, "Receive NACK for slave address"
-			"setting\n");
-		return -EIO;
+	if (ioread32(p + PCH_I2CSR) & PCH_GETACK) {
+		pch_dbg(adap, "Receive NACK for slave address setting\n");
+		return -ENXIO;
 	}
 
 	return 0;
diff --git a/drivers/i2c/busses/i2c-exynos5.c b/drivers/i2c/busses/i2c-exynos5.c
new file mode 100644
index 000000000000..c1ef228095b5
--- /dev/null
+++ b/drivers/i2c/busses/i2c-exynos5.c
@@ -0,0 +1,769 @@
+/**
+ * i2c-exynos5.c - Samsung Exynos5 I2C Controller Driver
+ *
+ * Copyright (C) 2013 Samsung Electronics Co., Ltd.
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+*/
+
+#include <linux/kernel.h>
+#include <linux/module.h>
+
+#include <linux/i2c.h>
+#include <linux/init.h>
+#include <linux/time.h>
+#include <linux/interrupt.h>
+#include <linux/delay.h>
+#include <linux/errno.h>
+#include <linux/err.h>
+#include <linux/platform_device.h>
+#include <linux/clk.h>
+#include <linux/slab.h>
+#include <linux/io.h>
+#include <linux/of_address.h>
+#include <linux/of_irq.h>
+#include <linux/spinlock.h>
+
+/*
+ * HSI2C controller from Samsung supports 2 modes of operation
+ * 1. Auto mode: Where in master automatically controls the whole transaction
+ * 2. Manual mode: Software controls the transaction by issuing commands
+ *    START, READ, WRITE, STOP, RESTART in I2C_MANUAL_CMD register.
+ *
+ * Operation mode can be selected by setting AUTO_MODE bit in I2C_CONF register
+ *
+ * Special bits are available for both modes of operation to set commands
+ * and for checking transfer status
+ */
+
+/* Register Map */
+#define HSI2C_CTL		0x00
+#define HSI2C_FIFO_CTL		0x04
+#define HSI2C_TRAILIG_CTL	0x08
+#define HSI2C_CLK_CTL		0x0C
+#define HSI2C_CLK_SLOT		0x10
+#define HSI2C_INT_ENABLE	0x20
+#define HSI2C_INT_STATUS	0x24
+#define HSI2C_ERR_STATUS	0x2C
+#define HSI2C_FIFO_STATUS	0x30
+#define HSI2C_TX_DATA		0x34
+#define HSI2C_RX_DATA		0x38
+#define HSI2C_CONF		0x40
+#define HSI2C_AUTO_CONF		0x44
+#define HSI2C_TIMEOUT		0x48
+#define HSI2C_MANUAL_CMD	0x4C
+#define HSI2C_TRANS_STATUS	0x50
+#define HSI2C_TIMING_HS1	0x54
+#define HSI2C_TIMING_HS2	0x58
+#define HSI2C_TIMING_HS3	0x5C
+#define HSI2C_TIMING_FS1	0x60
+#define HSI2C_TIMING_FS2	0x64
+#define HSI2C_TIMING_FS3	0x68
+#define HSI2C_TIMING_SLA	0x6C
+#define HSI2C_ADDR		0x70
+
+/* I2C_CTL Register bits */
+#define HSI2C_FUNC_MODE_I2C			(1u << 0)
+#define HSI2C_MASTER				(1u << 3)
+#define HSI2C_RXCHON				(1u << 6)
+#define HSI2C_TXCHON				(1u << 7)
+#define HSI2C_SW_RST				(1u << 31)
+
+/* I2C_FIFO_CTL Register bits */
+#define HSI2C_RXFIFO_EN				(1u << 0)
+#define HSI2C_TXFIFO_EN				(1u << 1)
+#define HSI2C_RXFIFO_TRIGGER_LEVEL(x)		((x) << 4)
+#define HSI2C_TXFIFO_TRIGGER_LEVEL(x)		((x) << 16)
+
+/* As per user manual FIFO max depth is 64bytes */
+#define HSI2C_FIFO_MAX				0x40
+/* default trigger levels for Tx and Rx FIFOs */
+#define HSI2C_DEF_TXFIFO_LVL			(HSI2C_FIFO_MAX - 0x30)
+#define HSI2C_DEF_RXFIFO_LVL			(HSI2C_FIFO_MAX - 0x10)
+
+/* I2C_TRAILING_CTL Register bits */
+#define HSI2C_TRAILING_COUNT			(0xf)
+
+/* I2C_INT_EN Register bits */
+#define HSI2C_INT_TX_ALMOSTEMPTY_EN		(1u << 0)
+#define HSI2C_INT_RX_ALMOSTFULL_EN		(1u << 1)
+#define HSI2C_INT_TRAILING_EN			(1u << 6)
+#define HSI2C_INT_I2C_EN			(1u << 9)
+
+/* I2C_INT_STAT Register bits */
+#define HSI2C_INT_TX_ALMOSTEMPTY		(1u << 0)
+#define HSI2C_INT_RX_ALMOSTFULL			(1u << 1)
+#define HSI2C_INT_TX_UNDERRUN			(1u << 2)
+#define HSI2C_INT_TX_OVERRUN			(1u << 3)
+#define HSI2C_INT_RX_UNDERRUN			(1u << 4)
+#define HSI2C_INT_RX_OVERRUN			(1u << 5)
+#define HSI2C_INT_TRAILING			(1u << 6)
+#define HSI2C_INT_I2C				(1u << 9)
+
+/* I2C_FIFO_STAT Register bits */
+#define HSI2C_RX_FIFO_EMPTY			(1u << 24)
+#define HSI2C_RX_FIFO_FULL			(1u << 23)
+#define HSI2C_RX_FIFO_LVL(x)			((x >> 16) & 0x7f)
+#define HSI2C_TX_FIFO_EMPTY			(1u << 8)
+#define HSI2C_TX_FIFO_FULL			(1u << 7)
+#define HSI2C_TX_FIFO_LVL(x)			((x >> 0) & 0x7f)
+
+/* I2C_CONF Register bits */
+#define HSI2C_AUTO_MODE				(1u << 31)
+#define HSI2C_10BIT_ADDR_MODE			(1u << 30)
+#define HSI2C_HS_MODE				(1u << 29)
+
+/* I2C_AUTO_CONF Register bits */
+#define HSI2C_READ_WRITE			(1u << 16)
+#define HSI2C_STOP_AFTER_TRANS			(1u << 17)
+#define HSI2C_MASTER_RUN			(1u << 31)
+
+/* I2C_TIMEOUT Register bits */
+#define HSI2C_TIMEOUT_EN			(1u << 31)
+#define HSI2C_TIMEOUT_MASK			0xff
+
+/* I2C_TRANS_STATUS register bits */
+#define HSI2C_MASTER_BUSY			(1u << 17)
+#define HSI2C_SLAVE_BUSY			(1u << 16)
+#define HSI2C_TIMEOUT_AUTO			(1u << 4)
+#define HSI2C_NO_DEV				(1u << 3)
+#define HSI2C_NO_DEV_ACK			(1u << 2)
+#define HSI2C_TRANS_ABORT			(1u << 1)
+#define HSI2C_TRANS_DONE			(1u << 0)
+
+/* I2C_ADDR register bits */
+#define HSI2C_SLV_ADDR_SLV(x)			((x & 0x3ff) << 0)
+#define HSI2C_SLV_ADDR_MAS(x)			((x & 0x3ff) << 10)
+#define HSI2C_MASTER_ID(x)			((x & 0xff) << 24)
+#define MASTER_ID(x)				((x & 0x7) + 0x08)
+
+/*
+ * Controller operating frequency, timing values for operation
+ * are calculated against this frequency
+ */
+#define HSI2C_HS_TX_CLOCK	1000000
+#define HSI2C_FS_TX_CLOCK	100000
+#define HSI2C_HIGH_SPD		1
+#define HSI2C_FAST_SPD		0
+
+#define EXYNOS5_I2C_TIMEOUT (msecs_to_jiffies(1000))
+
+struct exynos5_i2c {
+	struct i2c_adapter	adap;
+	unsigned int		suspended:1;
+
+	struct i2c_msg		*msg;
+	struct completion	msg_complete;
+	unsigned int		msg_ptr;
+
+	unsigned int		irq;
+
+	void __iomem		*regs;
+	struct clk		*clk;
+	struct device		*dev;
+	int			state;
+
+	spinlock_t		lock;		/* IRQ synchronization */
+
+	/*
+	 * Since the TRANS_DONE bit is cleared on read, and we may read it
+	 * either during an IRQ or after a transaction, keep track of its
+	 * state here.
+	 */
+	int			trans_done;
+
+	/* Controller operating frequency */
+	unsigned int		fs_clock;
+	unsigned int		hs_clock;
+
+	/*
+	 * HSI2C Controller can operate in
+	 * 1. High speed upto 3.4Mbps
+	 * 2. Fast speed upto 1Mbps
+	 */
+	int			speed_mode;
+};
+
+static const struct of_device_id exynos5_i2c_match[] = {
+	{ .compatible = "samsung,exynos5-hsi2c" },
+	{},
+};
+MODULE_DEVICE_TABLE(of, exynos5_i2c_match);
+
+static void exynos5_i2c_clr_pend_irq(struct exynos5_i2c *i2c)
+{
+	writel(readl(i2c->regs + HSI2C_INT_STATUS),
+				i2c->regs + HSI2C_INT_STATUS);
+}
+
+/*
+ * exynos5_i2c_set_timing: updates the registers with appropriate
+ * timing values calculated
+ *
+ * Returns 0 on success, -EINVAL if the cycle length cannot
+ * be calculated.
+ */
+static int exynos5_i2c_set_timing(struct exynos5_i2c *i2c, int mode)
+{
+	u32 i2c_timing_s1;
+	u32 i2c_timing_s2;
+	u32 i2c_timing_s3;
+	u32 i2c_timing_sla;
+	unsigned int t_start_su, t_start_hd;
+	unsigned int t_stop_su;
+	unsigned int t_data_su, t_data_hd;
+	unsigned int t_scl_l, t_scl_h;
+	unsigned int t_sr_release;
+	unsigned int t_ftl_cycle;
+	unsigned int clkin = clk_get_rate(i2c->clk);
+	unsigned int div, utemp0 = 0, utemp1 = 0, clk_cycle;
+	unsigned int op_clk = (mode == HSI2C_HIGH_SPD) ?
+				i2c->hs_clock : i2c->fs_clock;
+
+	/*
+	 * FPCLK / FI2C =
+	 * (CLK_DIV + 1) * (TSCLK_L + TSCLK_H + 2) + 8 + 2 * FLT_CYCLE
+	 * utemp0 = (CLK_DIV + 1) * (TSCLK_L + TSCLK_H + 2)
+	 * utemp1 = (TSCLK_L + TSCLK_H + 2)
+	 */
+	t_ftl_cycle = (readl(i2c->regs + HSI2C_CONF) >> 16) & 0x7;
+	utemp0 = (clkin / op_clk) - 8 - 2 * t_ftl_cycle;
+
+	/* CLK_DIV max is 256 */
+	for (div = 0; div < 256; div++) {
+		utemp1 = utemp0 / (div + 1);
+
+		/*
+		 * SCL_L and SCL_H each has max value of 255
+		 * Hence, For the clk_cycle to the have right value
+		 * utemp1 has to be less then 512 and more than 4.
+		 */
+		if ((utemp1 < 512) && (utemp1 > 4)) {
+			clk_cycle = utemp1 - 2;
+			break;
+		} else if (div == 255) {
+			dev_warn(i2c->dev, "Failed to calculate divisor");
+			return -EINVAL;
+		}
+	}
+
+	t_scl_l = clk_cycle / 2;
+	t_scl_h = clk_cycle / 2;
+	t_start_su = t_scl_l;
+	t_start_hd = t_scl_l;
+	t_stop_su = t_scl_l;
+	t_data_su = t_scl_l / 2;
+	t_data_hd = t_scl_l / 2;
+	t_sr_release = clk_cycle;
+
+	i2c_timing_s1 = t_start_su << 24 | t_start_hd << 16 | t_stop_su << 8;
+	i2c_timing_s2 = t_data_su << 24 | t_scl_l << 8 | t_scl_h << 0;
+	i2c_timing_s3 = div << 16 | t_sr_release << 0;
+	i2c_timing_sla = t_data_hd << 0;
+
+	dev_dbg(i2c->dev, "tSTART_SU: %X, tSTART_HD: %X, tSTOP_SU: %X\n",
+		t_start_su, t_start_hd, t_stop_su);
+	dev_dbg(i2c->dev, "tDATA_SU: %X, tSCL_L: %X, tSCL_H: %X\n",
+		t_data_su, t_scl_l, t_scl_h);
+	dev_dbg(i2c->dev, "nClkDiv: %X, tSR_RELEASE: %X\n",
+		div, t_sr_release);
+	dev_dbg(i2c->dev, "tDATA_HD: %X\n", t_data_hd);
+
+	if (mode == HSI2C_HIGH_SPD) {
+		writel(i2c_timing_s1, i2c->regs + HSI2C_TIMING_HS1);
+		writel(i2c_timing_s2, i2c->regs + HSI2C_TIMING_HS2);
+		writel(i2c_timing_s3, i2c->regs + HSI2C_TIMING_HS3);
+	} else {
+		writel(i2c_timing_s1, i2c->regs + HSI2C_TIMING_FS1);
+		writel(i2c_timing_s2, i2c->regs + HSI2C_TIMING_FS2);
+		writel(i2c_timing_s3, i2c->regs + HSI2C_TIMING_FS3);
+	}
+	writel(i2c_timing_sla, i2c->regs + HSI2C_TIMING_SLA);
+
+	return 0;
+}
+
+static int exynos5_hsi2c_clock_setup(struct exynos5_i2c *i2c)
+{
+	/*
+	 * Configure the Fast speed timing values
+	 * Even the High Speed mode initially starts with Fast mode
+	 */
+	if (exynos5_i2c_set_timing(i2c, HSI2C_FAST_SPD)) {
+		dev_err(i2c->dev, "HSI2C FS Clock set up failed\n");
+		return -EINVAL;
+	}
+
+	/* configure the High speed timing values */
+	if (i2c->speed_mode == HSI2C_HIGH_SPD) {
+		if (exynos5_i2c_set_timing(i2c, HSI2C_HIGH_SPD)) {
+			dev_err(i2c->dev, "HSI2C HS Clock set up failed\n");
+			return -EINVAL;
+		}
+	}
+
+	return 0;
+}
+
+/*
+ * exynos5_i2c_init: configures the controller for I2C functionality
+ * Programs I2C controller for Master mode operation
+ */
+static void exynos5_i2c_init(struct exynos5_i2c *i2c)
+{
+	u32 i2c_conf = readl(i2c->regs + HSI2C_CONF);
+	u32 i2c_timeout = readl(i2c->regs + HSI2C_TIMEOUT);
+
+	/* Clear to disable Timeout */
+	i2c_timeout &= ~HSI2C_TIMEOUT_EN;
+	writel(i2c_timeout, i2c->regs + HSI2C_TIMEOUT);
+
+	writel((HSI2C_FUNC_MODE_I2C | HSI2C_MASTER),
+					i2c->regs + HSI2C_CTL);
+	writel(HSI2C_TRAILING_COUNT, i2c->regs + HSI2C_TRAILIG_CTL);
+
+	if (i2c->speed_mode == HSI2C_HIGH_SPD) {
+		writel(HSI2C_MASTER_ID(MASTER_ID(i2c->adap.nr)),
+					i2c->regs + HSI2C_ADDR);
+		i2c_conf |= HSI2C_HS_MODE;
+	}
+
+	writel(i2c_conf | HSI2C_AUTO_MODE, i2c->regs + HSI2C_CONF);
+}
+
+static void exynos5_i2c_reset(struct exynos5_i2c *i2c)
+{
+	u32 i2c_ctl;
+
+	/* Set and clear the bit for reset */
+	i2c_ctl = readl(i2c->regs + HSI2C_CTL);
+	i2c_ctl |= HSI2C_SW_RST;
+	writel(i2c_ctl, i2c->regs + HSI2C_CTL);
+
+	i2c_ctl = readl(i2c->regs + HSI2C_CTL);
+	i2c_ctl &= ~HSI2C_SW_RST;
+	writel(i2c_ctl, i2c->regs + HSI2C_CTL);
+
+	/* We don't expect calculations to fail during the run */
+	exynos5_hsi2c_clock_setup(i2c);
+	/* Initialize the configure registers */
+	exynos5_i2c_init(i2c);
+}
+
+/*
+ * exynos5_i2c_irq: top level IRQ servicing routine
+ *
+ * INT_STATUS registers gives the interrupt details. Further,
+ * FIFO_STATUS or TRANS_STATUS registers are to be check for detailed
+ * state of the bus.
+ */
+static irqreturn_t exynos5_i2c_irq(int irqno, void *dev_id)
+{
+	struct exynos5_i2c *i2c = dev_id;
+	u32 fifo_level, int_status, fifo_status, trans_status;
+	unsigned char byte;
+	int len = 0;
+
+	i2c->state = -EINVAL;
+
+	spin_lock(&i2c->lock);
+
+	int_status = readl(i2c->regs + HSI2C_INT_STATUS);
+	writel(int_status, i2c->regs + HSI2C_INT_STATUS);
+	fifo_status = readl(i2c->regs + HSI2C_FIFO_STATUS);
+
+	/* handle interrupt related to the transfer status */
+	if (int_status & HSI2C_INT_I2C) {
+		trans_status = readl(i2c->regs + HSI2C_TRANS_STATUS);
+		if (trans_status & HSI2C_NO_DEV_ACK) {
+			dev_dbg(i2c->dev, "No ACK from device\n");
+			i2c->state = -ENXIO;
+			goto stop;
+		} else if (trans_status & HSI2C_NO_DEV) {
+			dev_dbg(i2c->dev, "No device\n");
+			i2c->state = -ENXIO;
+			goto stop;
+		} else if (trans_status & HSI2C_TRANS_ABORT) {
+			dev_dbg(i2c->dev, "Deal with arbitration lose\n");
+			i2c->state = -EAGAIN;
+			goto stop;
+		} else if (trans_status & HSI2C_TIMEOUT_AUTO) {
+			dev_dbg(i2c->dev, "Accessing device timed out\n");
+			i2c->state = -EAGAIN;
+			goto stop;
+		} else if (trans_status & HSI2C_TRANS_DONE) {
+			i2c->trans_done = 1;
+			i2c->state = 0;
+		}
+	}
+
+	if ((i2c->msg->flags & I2C_M_RD) && (int_status &
+			(HSI2C_INT_TRAILING | HSI2C_INT_RX_ALMOSTFULL))) {
+		fifo_status = readl(i2c->regs + HSI2C_FIFO_STATUS);
+		fifo_level = HSI2C_RX_FIFO_LVL(fifo_status);
+		len = min(fifo_level, i2c->msg->len - i2c->msg_ptr);
+
+		while (len > 0) {
+			byte = (unsigned char)
+				readl(i2c->regs + HSI2C_RX_DATA);
+			i2c->msg->buf[i2c->msg_ptr++] = byte;
+			len--;
+		}
+		i2c->state = 0;
+	} else if (int_status & HSI2C_INT_TX_ALMOSTEMPTY) {
+		fifo_status = readl(i2c->regs + HSI2C_FIFO_STATUS);
+		fifo_level = HSI2C_TX_FIFO_LVL(fifo_status);
+
+		len = HSI2C_FIFO_MAX - fifo_level;
+		if (len > (i2c->msg->len - i2c->msg_ptr))
+			len = i2c->msg->len - i2c->msg_ptr;
+
+		while (len > 0) {
+			byte = i2c->msg->buf[i2c->msg_ptr++];
+			writel(byte, i2c->regs + HSI2C_TX_DATA);
+			len--;
+		}
+		i2c->state = 0;
+	}
+
+ stop:
+	if ((i2c->trans_done && (i2c->msg->len == i2c->msg_ptr)) ||
+	    (i2c->state < 0)) {
+		writel(0, i2c->regs + HSI2C_INT_ENABLE);
+		exynos5_i2c_clr_pend_irq(i2c);
+		complete(&i2c->msg_complete);
+	}
+
+	spin_unlock(&i2c->lock);
+
+	return IRQ_HANDLED;
+}
+
+/*
+ * exynos5_i2c_wait_bus_idle
+ *
+ * Wait for the bus to go idle, indicated by the MASTER_BUSY bit being
+ * cleared.
+ *
+ * Returns -EBUSY if the bus cannot be bought to idle
+ */
+static int exynos5_i2c_wait_bus_idle(struct exynos5_i2c *i2c)
+{
+	unsigned long stop_time;
+	u32 trans_status;
+
+	/* wait for 100 milli seconds for the bus to be idle */
+	stop_time = jiffies + msecs_to_jiffies(100) + 1;
+	do {
+		trans_status = readl(i2c->regs + HSI2C_TRANS_STATUS);
+		if (!(trans_status & HSI2C_MASTER_BUSY))
+			return 0;
+
+		usleep_range(50, 200);
+	} while (time_before(jiffies, stop_time));
+
+	return -EBUSY;
+}
+
+/*
+ * exynos5_i2c_message_start: Configures the bus and starts the xfer
+ * i2c: struct exynos5_i2c pointer for the current bus
+ * stop: Enables stop after transfer if set. Set for last transfer of
+ *       in the list of messages.
+ *
+ * Configures the bus for read/write function
+ * Sets chip address to talk to, message length to be sent.
+ * Enables appropriate interrupts and sends start xfer command.
+ */
+static void exynos5_i2c_message_start(struct exynos5_i2c *i2c, int stop)
+{
+	u32 i2c_ctl;
+	u32 int_en = HSI2C_INT_I2C_EN;
+	u32 i2c_auto_conf = 0;
+	u32 fifo_ctl;
+	unsigned long flags;
+
+	i2c_ctl = readl(i2c->regs + HSI2C_CTL);
+	i2c_ctl &= ~(HSI2C_TXCHON | HSI2C_RXCHON);
+	fifo_ctl = HSI2C_RXFIFO_EN | HSI2C_TXFIFO_EN;
+
+	if (i2c->msg->flags & I2C_M_RD) {
+		i2c_ctl |= HSI2C_RXCHON;
+
+		i2c_auto_conf = HSI2C_READ_WRITE;
+
+		fifo_ctl |= HSI2C_RXFIFO_TRIGGER_LEVEL(HSI2C_DEF_TXFIFO_LVL);
+		int_en |= (HSI2C_INT_RX_ALMOSTFULL_EN |
+			HSI2C_INT_TRAILING_EN);
+	} else {
+		i2c_ctl |= HSI2C_TXCHON;
+
+		fifo_ctl |= HSI2C_TXFIFO_TRIGGER_LEVEL(HSI2C_DEF_RXFIFO_LVL);
+		int_en |= HSI2C_INT_TX_ALMOSTEMPTY_EN;
+	}
+
+	writel(HSI2C_SLV_ADDR_MAS(i2c->msg->addr), i2c->regs + HSI2C_ADDR);
+
+	writel(fifo_ctl, i2c->regs + HSI2C_FIFO_CTL);
+	writel(i2c_ctl, i2c->regs + HSI2C_CTL);
+
+
+	/*
+	 * Enable interrupts before starting the transfer so that we don't
+	 * miss any INT_I2C interrupts.
+	 */
+	spin_lock_irqsave(&i2c->lock, flags);
+	writel(int_en, i2c->regs + HSI2C_INT_ENABLE);
+
+	if (stop == 1)
+		i2c_auto_conf |= HSI2C_STOP_AFTER_TRANS;
+	i2c_auto_conf |= i2c->msg->len;
+	i2c_auto_conf |= HSI2C_MASTER_RUN;
+	writel(i2c_auto_conf, i2c->regs + HSI2C_AUTO_CONF);
+	spin_unlock_irqrestore(&i2c->lock, flags);
+}
+
+static int exynos5_i2c_xfer_msg(struct exynos5_i2c *i2c,
+			      struct i2c_msg *msgs, int stop)
+{
+	unsigned long timeout;
+	int ret;
+
+	i2c->msg = msgs;
+	i2c->msg_ptr = 0;
+	i2c->trans_done = 0;
+
+	reinit_completion(&i2c->msg_complete);
+
+	exynos5_i2c_message_start(i2c, stop);
+
+	timeout = wait_for_completion_timeout(&i2c->msg_complete,
+					      EXYNOS5_I2C_TIMEOUT);
+	if (timeout == 0)
+		ret = -ETIMEDOUT;
+	else
+		ret = i2c->state;
+
+	/*
+	 * If this is the last message to be transfered (stop == 1)
+	 * Then check if the bus can be brought back to idle.
+	 */
+	if (ret == 0 && stop)
+		ret = exynos5_i2c_wait_bus_idle(i2c);
+
+	if (ret < 0) {
+		exynos5_i2c_reset(i2c);
+		if (ret == -ETIMEDOUT)
+			dev_warn(i2c->dev, "%s timeout\n",
+				 (msgs->flags & I2C_M_RD) ? "rx" : "tx");
+	}
+
+	/* Return the state as in interrupt routine */
+	return ret;
+}
+
+static int exynos5_i2c_xfer(struct i2c_adapter *adap,
+			struct i2c_msg *msgs, int num)
+{
+	struct exynos5_i2c *i2c = (struct exynos5_i2c *)adap->algo_data;
+	int i = 0, ret = 0, stop = 0;
+
+	if (i2c->suspended) {
+		dev_err(i2c->dev, "HS-I2C is not initialzed.\n");
+		return -EIO;
+	}
+
+	clk_prepare_enable(i2c->clk);
+
+	for (i = 0; i < num; i++, msgs++) {
+		stop = (i == num - 1);
+
+		ret = exynos5_i2c_xfer_msg(i2c, msgs, stop);
+
+		if (ret < 0)
+			goto out;
+	}
+
+	if (i == num) {
+		ret = num;
+	} else {
+		/* Only one message, cannot access the device */
+		if (i == 1)
+			ret = -EREMOTEIO;
+		else
+			ret = i;
+
+		dev_warn(i2c->dev, "xfer message failed\n");
+	}
+
+ out:
+	clk_disable_unprepare(i2c->clk);
+	return ret;
+}
+
+static u32 exynos5_i2c_func(struct i2c_adapter *adap)
+{
+	return I2C_FUNC_I2C | (I2C_FUNC_SMBUS_EMUL & ~I2C_FUNC_SMBUS_QUICK);
+}
+
+static const struct i2c_algorithm exynos5_i2c_algorithm = {
+	.master_xfer		= exynos5_i2c_xfer,
+	.functionality		= exynos5_i2c_func,
+};
+
+static int exynos5_i2c_probe(struct platform_device *pdev)
+{
+	struct device_node *np = pdev->dev.of_node;
+	struct exynos5_i2c *i2c;
+	struct resource *mem;
+	unsigned int op_clock;
+	int ret;
+
+	i2c = devm_kzalloc(&pdev->dev, sizeof(struct exynos5_i2c), GFP_KERNEL);
+	if (!i2c) {
+		dev_err(&pdev->dev, "no memory for state\n");
+		return -ENOMEM;
+	}
+
+	if (of_property_read_u32(np, "clock-frequency", &op_clock)) {
+		i2c->speed_mode = HSI2C_FAST_SPD;
+		i2c->fs_clock = HSI2C_FS_TX_CLOCK;
+	} else {
+		if (op_clock >= HSI2C_HS_TX_CLOCK) {
+			i2c->speed_mode = HSI2C_HIGH_SPD;
+			i2c->fs_clock = HSI2C_FS_TX_CLOCK;
+			i2c->hs_clock = op_clock;
+		} else {
+			i2c->speed_mode = HSI2C_FAST_SPD;
+			i2c->fs_clock = op_clock;
+		}
+	}
+
+	strlcpy(i2c->adap.name, "exynos5-i2c", sizeof(i2c->adap.name));
+	i2c->adap.owner   = THIS_MODULE;
+	i2c->adap.algo    = &exynos5_i2c_algorithm;
+	i2c->adap.retries = 3;
+
+	i2c->dev = &pdev->dev;
+	i2c->clk = devm_clk_get(&pdev->dev, "hsi2c");
+	if (IS_ERR(i2c->clk)) {
+		dev_err(&pdev->dev, "cannot get clock\n");
+		return -ENOENT;
+	}
+
+	clk_prepare_enable(i2c->clk);
+
+	mem = platform_get_resource(pdev, IORESOURCE_MEM, 0);
+	i2c->regs = devm_ioremap_resource(&pdev->dev, mem);
+	if (IS_ERR(i2c->regs)) {
+		ret = PTR_ERR(i2c->regs);
+		goto err_clk;
+	}
+
+	i2c->adap.dev.of_node = np;
+	i2c->adap.algo_data = i2c;
+	i2c->adap.dev.parent = &pdev->dev;
+
+	/* Clear pending interrupts from u-boot or misc causes */
+	exynos5_i2c_clr_pend_irq(i2c);
+
+	spin_lock_init(&i2c->lock);
+	init_completion(&i2c->msg_complete);
+
+	i2c->irq = ret = platform_get_irq(pdev, 0);
+	if (ret <= 0) {
+		dev_err(&pdev->dev, "cannot find HS-I2C IRQ\n");
+		ret = -EINVAL;
+		goto err_clk;
+	}
+
+	ret = devm_request_irq(&pdev->dev, i2c->irq, exynos5_i2c_irq,
+				IRQF_NO_SUSPEND | IRQF_ONESHOT,
+				dev_name(&pdev->dev), i2c);
+
+	if (ret != 0) {
+		dev_err(&pdev->dev, "cannot request HS-I2C IRQ %d\n", i2c->irq);
+		goto err_clk;
+	}
+
+	ret = exynos5_hsi2c_clock_setup(i2c);
+	if (ret)
+		goto err_clk;
+
+	exynos5_i2c_init(i2c);
+
+	ret = i2c_add_adapter(&i2c->adap);
+	if (ret < 0) {
+		dev_err(&pdev->dev, "failed to add bus to i2c core\n");
+		goto err_clk;
+	}
+
+	platform_set_drvdata(pdev, i2c);
+
+ err_clk:
+	clk_disable_unprepare(i2c->clk);
+	return ret;
+}
+
+static int exynos5_i2c_remove(struct platform_device *pdev)
+{
+	struct exynos5_i2c *i2c = platform_get_drvdata(pdev);
+
+	i2c_del_adapter(&i2c->adap);
+
+	return 0;
+}
+
+static int exynos5_i2c_suspend_noirq(struct device *dev)
+{
+	struct platform_device *pdev = to_platform_device(dev);
+	struct exynos5_i2c *i2c = platform_get_drvdata(pdev);
+
+	i2c->suspended = 1;
+
+	return 0;
+}
+
+static int exynos5_i2c_resume_noirq(struct device *dev)
+{
+	struct platform_device *pdev = to_platform_device(dev);
+	struct exynos5_i2c *i2c = platform_get_drvdata(pdev);
+	int ret = 0;
+
+	clk_prepare_enable(i2c->clk);
+
+	ret = exynos5_hsi2c_clock_setup(i2c);
+	if (ret) {
+		clk_disable_unprepare(i2c->clk);
+		return ret;
+	}
+
+	exynos5_i2c_init(i2c);
+	clk_disable_unprepare(i2c->clk);
+	i2c->suspended = 0;
+
+	return 0;
+}
+
+static SIMPLE_DEV_PM_OPS(exynos5_i2c_dev_pm_ops, exynos5_i2c_suspend_noirq,
+			 exynos5_i2c_resume_noirq);
+
+static struct platform_driver exynos5_i2c_driver = {
+	.probe		= exynos5_i2c_probe,
+	.remove		= exynos5_i2c_remove,
+	.driver		= {
+		.owner	= THIS_MODULE,
+		.name	= "exynos5-hsi2c",
+		.pm	= &exynos5_i2c_dev_pm_ops,
+		.of_match_table = exynos5_i2c_match,
+	},
+};
+
+module_platform_driver(exynos5_i2c_driver);
+
+MODULE_DESCRIPTION("Exynos5 HS-I2C Bus driver");
+MODULE_AUTHOR("Naveen Krishna Chatradhi, <ch.naveen@samsung.com>");
+MODULE_AUTHOR("Taekgyun Ko, <taeggyun.ko@samsung.com>");
+MODULE_LICENSE("GPL v2");
diff --git a/drivers/i2c/busses/i2c-gpio.c b/drivers/i2c/busses/i2c-gpio.c
index bfa02c6c2dda..d9f7e186a4c7 100644
--- a/drivers/i2c/busses/i2c-gpio.c
+++ b/drivers/i2c/busses/i2c-gpio.c
@@ -15,6 +15,7 @@
 #include <linux/slab.h>
 #include <linux/platform_device.h>
 #include <linux/gpio.h>
+#include <linux/of.h>
 #include <linux/of_gpio.h>
 
 struct i2c_gpio_private_data {
diff --git a/drivers/i2c/busses/i2c-i801.c b/drivers/i2c/busses/i2c-i801.c
index 4296d1721272..737e29866887 100644
--- a/drivers/i2c/busses/i2c-i801.c
+++ b/drivers/i2c/busses/i2c-i801.c
@@ -59,6 +59,7 @@
   Wellsburg (PCH) MS    0x8d7e     32     hard     yes     yes     yes
   Wellsburg (PCH) MS    0x8d7f     32     hard     yes     yes     yes
   Coleto Creek (PCH)    0x23b0     32     hard     yes     yes     yes
+  Wildcat Point-LP (PCH)   0x9ca2     32     hard     yes     yes     yes
 
   Features supported by this driver:
   Software PEC                     no
@@ -177,6 +178,7 @@
 #define PCI_DEVICE_ID_INTEL_WELLSBURG_SMBUS_MS1	0x8d7e
 #define PCI_DEVICE_ID_INTEL_WELLSBURG_SMBUS_MS2	0x8d7f
 #define PCI_DEVICE_ID_INTEL_LYNXPOINT_LP_SMBUS	0x9c22
+#define PCI_DEVICE_ID_INTEL_WILDCATPOINT_LP_SMBUS	0x9ca2
 
 struct i801_mux_config {
 	char *gpio_chip;
@@ -819,6 +821,7 @@ static DEFINE_PCI_DEVICE_TABLE(i801_ids) = {
 	{ PCI_DEVICE(PCI_VENDOR_ID_INTEL, PCI_DEVICE_ID_INTEL_WELLSBURG_SMBUS_MS1) },
 	{ PCI_DEVICE(PCI_VENDOR_ID_INTEL, PCI_DEVICE_ID_INTEL_WELLSBURG_SMBUS_MS2) },
 	{ PCI_DEVICE(PCI_VENDOR_ID_INTEL, PCI_DEVICE_ID_INTEL_COLETOCREEK_SMBUS) },
+	{ PCI_DEVICE(PCI_VENDOR_ID_INTEL, PCI_DEVICE_ID_INTEL_WILDCATPOINT_LP_SMBUS) },
 	{ 0, }
 };
 
diff --git a/drivers/i2c/busses/i2c-ismt.c b/drivers/i2c/busses/i2c-ismt.c
index 1672effbcebb..0043ede234c2 100644
--- a/drivers/i2c/busses/i2c-ismt.c
+++ b/drivers/i2c/busses/i2c-ismt.c
@@ -541,7 +541,7 @@ static int ismt_access(struct i2c_adapter *adap, u16 addr,
 		desc->dptr_high = upper_32_bits(dma_addr);
 	}
 
-	INIT_COMPLETION(priv->cmp);
+	reinit_completion(&priv->cmp);
 
 	/* Add the descriptor */
 	ismt_submit_desc(priv);
diff --git a/drivers/i2c/busses/i2c-mv64xxx.c b/drivers/i2c/busses/i2c-mv64xxx.c
index d3e9cc3153a9..8be7e42aa4de 100644
--- a/drivers/i2c/busses/i2c-mv64xxx.c
+++ b/drivers/i2c/busses/i2c-mv64xxx.c
@@ -911,7 +911,7 @@ static struct platform_driver mv64xxx_i2c_driver = {
 	.driver	= {
 		.owner	= THIS_MODULE,
 		.name	= MV64XXX_I2C_CTLR_NAME,
-		.of_match_table = of_match_ptr(mv64xxx_i2c_of_match_table),
+		.of_match_table = mv64xxx_i2c_of_match_table,
 	},
 };
 
diff --git a/drivers/i2c/busses/i2c-mxs.c b/drivers/i2c/busses/i2c-mxs.c
index b7c857774708..0cde4e6ab2b2 100644
--- a/drivers/i2c/busses/i2c-mxs.c
+++ b/drivers/i2c/busses/i2c-mxs.c
@@ -1,6 +1,7 @@
 /*
  * Freescale MXS I2C bus driver
  *
+ * Copyright (C) 2012-2013 Marek Vasut <marex@denx.de>
  * Copyright (C) 2011-2012 Wolfram Sang, Pengutronix e.K.
  *
  * based on a (non-working) driver which was:
@@ -34,10 +35,12 @@
 
 #define MXS_I2C_CTRL0		(0x00)
 #define MXS_I2C_CTRL0_SET	(0x04)
+#define MXS_I2C_CTRL0_CLR	(0x08)
 
 #define MXS_I2C_CTRL0_SFTRST			0x80000000
 #define MXS_I2C_CTRL0_RUN			0x20000000
 #define MXS_I2C_CTRL0_SEND_NAK_ON_LAST		0x02000000
+#define MXS_I2C_CTRL0_PIO_MODE			0x01000000
 #define MXS_I2C_CTRL0_RETAIN_CLOCK		0x00200000
 #define MXS_I2C_CTRL0_POST_SEND_STOP		0x00100000
 #define MXS_I2C_CTRL0_PRE_SEND_START		0x00080000
@@ -64,13 +67,13 @@
 #define MXS_I2C_CTRL1_SLAVE_IRQ			0x01
 
 #define MXS_I2C_STAT		(0x50)
+#define MXS_I2C_STAT_GOT_A_NAK			0x10000000
 #define MXS_I2C_STAT_BUS_BUSY			0x00000800
 #define MXS_I2C_STAT_CLK_GEN_BUSY		0x00000400
 
-#define MXS_I2C_DATA		(0xa0)
+#define MXS_I2C_DATA(i2c)	((i2c->dev_type == MXS_I2C_V1) ? 0x60 : 0xa0)
 
-#define MXS_I2C_DEBUG0		(0xb0)
-#define MXS_I2C_DEBUG0_CLR	(0xb8)
+#define MXS_I2C_DEBUG0_CLR(i2c)	((i2c->dev_type == MXS_I2C_V1) ? 0x78 : 0xb8)
 
 #define MXS_I2C_DEBUG0_DMAREQ	0x80000000
 
@@ -95,10 +98,17 @@
 #define MXS_CMD_I2C_READ	(MXS_I2C_CTRL0_SEND_NAK_ON_LAST | \
 				 MXS_I2C_CTRL0_MASTER_MODE)
 
+enum mxs_i2c_devtype {
+	MXS_I2C_UNKNOWN = 0,
+	MXS_I2C_V1,
+	MXS_I2C_V2,
+};
+
 /**
  * struct mxs_i2c_dev - per device, private MXS-I2C data
  *
  * @dev: driver model device node
+ * @dev_type: distinguish i.MX23/i.MX28 features
  * @regs: IO registers pointer
  * @cmd_complete: completion object for transaction wait
  * @cmd_err: error code for last transaction
@@ -106,6 +116,7 @@
  */
 struct mxs_i2c_dev {
 	struct device *dev;
+	enum mxs_i2c_devtype dev_type;
 	void __iomem *regs;
 	struct completion cmd_complete;
 	int cmd_err;
@@ -291,48 +302,11 @@ write_init_pio_fail:
 	return -EINVAL;
 }
 
-static int mxs_i2c_pio_wait_dmareq(struct mxs_i2c_dev *i2c)
+static int mxs_i2c_pio_wait_xfer_end(struct mxs_i2c_dev *i2c)
 {
 	unsigned long timeout = jiffies + msecs_to_jiffies(1000);
 
-	while (!(readl(i2c->regs + MXS_I2C_DEBUG0) &
-		MXS_I2C_DEBUG0_DMAREQ)) {
-		if (time_after(jiffies, timeout))
-			return -ETIMEDOUT;
-		cond_resched();
-	}
-
-	return 0;
-}
-
-static int mxs_i2c_pio_wait_cplt(struct mxs_i2c_dev *i2c, int last)
-{
-	unsigned long timeout = jiffies + msecs_to_jiffies(1000);
-
-	/*
-	 * We do not use interrupts in the PIO mode. Due to the
-	 * maximum transfer length being 8 bytes in PIO mode, the
-	 * overhead of interrupt would be too large and this would
-	 * neglect the gain from using the PIO mode.
-	 */
-
-	while (!(readl(i2c->regs + MXS_I2C_CTRL1) &
-		MXS_I2C_CTRL1_DATA_ENGINE_CMPLT_IRQ)) {
-		if (time_after(jiffies, timeout))
-			return -ETIMEDOUT;
-		cond_resched();
-	}
-
-	writel(MXS_I2C_CTRL1_DATA_ENGINE_CMPLT_IRQ,
-		i2c->regs + MXS_I2C_CTRL1_CLR);
-
-	/*
-	 * When ending a transfer with a stop, we have to wait for the bus to
-	 * go idle before we report the transfer as completed. Otherwise the
-	 * start of the next transfer may race with the end of the current one.
-	 */
-	while (last && (readl(i2c->regs + MXS_I2C_STAT) &
-			(MXS_I2C_STAT_BUS_BUSY | MXS_I2C_STAT_CLK_GEN_BUSY))) {
+	while (readl(i2c->regs + MXS_I2C_CTRL0) & MXS_I2C_CTRL0_RUN) {
 		if (time_after(jiffies, timeout))
 			return -ETIMEDOUT;
 		cond_resched();
@@ -370,106 +344,215 @@ static void mxs_i2c_pio_trigger_cmd(struct mxs_i2c_dev *i2c, u32 cmd)
 	writel(reg, i2c->regs + MXS_I2C_CTRL0);
 }
 
+/*
+ * Start WRITE transaction on the I2C bus. By studying i.MX23 datasheet,
+ * CTRL0::PIO_MODE bit description clarifies the order in which the registers
+ * must be written during PIO mode operation. First, the CTRL0 register has
+ * to be programmed with all the necessary bits but the RUN bit. Then the
+ * payload has to be written into the DATA register. Finally, the transmission
+ * is executed by setting the RUN bit in CTRL0.
+ */
+static void mxs_i2c_pio_trigger_write_cmd(struct mxs_i2c_dev *i2c, u32 cmd,
+					  u32 data)
+{
+	writel(cmd, i2c->regs + MXS_I2C_CTRL0);
+
+	if (i2c->dev_type == MXS_I2C_V1)
+		writel(MXS_I2C_CTRL0_PIO_MODE, i2c->regs + MXS_I2C_CTRL0_SET);
+
+	writel(data, i2c->regs + MXS_I2C_DATA(i2c));
+	writel(MXS_I2C_CTRL0_RUN, i2c->regs + MXS_I2C_CTRL0_SET);
+}
+
 static int mxs_i2c_pio_setup_xfer(struct i2c_adapter *adap,
 			struct i2c_msg *msg, uint32_t flags)
 {
 	struct mxs_i2c_dev *i2c = i2c_get_adapdata(adap);
 	uint32_t addr_data = msg->addr << 1;
 	uint32_t data = 0;
-	int i, shifts_left, ret;
+	int i, ret, xlen = 0, xmit = 0;
+	uint32_t start;
 
 	/* Mute IRQs coming from this block. */
 	writel(MXS_I2C_IRQ_MASK << 8, i2c->regs + MXS_I2C_CTRL1_CLR);
 
+	/*
+	 * MX23 idea:
+	 * - Enable CTRL0::PIO_MODE (1 << 24)
+	 * - Enable CTRL1::ACK_MODE (1 << 27)
+	 *
+	 * WARNING! The MX23 is broken in some way, even if it claims
+	 * to support PIO, when we try to transfer any amount of data
+	 * that is not aligned to 4 bytes, the DMA engine will have
+	 * bits in DEBUG1::DMA_BYTES_ENABLES still set even after the
+	 * transfer. This in turn will mess up the next transfer as
+	 * the block it emit one byte write onto the bus terminated
+	 * with a NAK+STOP. A possible workaround is to reset the IP
+	 * block after every PIO transmission, which might just work.
+	 *
+	 * NOTE: The CTRL0::PIO_MODE description is important, since
+	 * it outlines how the PIO mode is really supposed to work.
+	 */
 	if (msg->flags & I2C_M_RD) {
+		/*
+		 * PIO READ transfer:
+		 *
+		 * This transfer MUST be limited to 4 bytes maximum. It is not
+		 * possible to transfer more than four bytes via PIO, since we
+		 * can not in any way make sure we can read the data from the
+		 * DATA register fast enough. Besides, the RX FIFO is only four
+		 * bytes deep, thus we can only really read up to four bytes at
+		 * time. Finally, there is no bit indicating us that new data
+		 * arrived at the FIFO and can thus be fetched from the DATA
+		 * register.
+		 */
+		BUG_ON(msg->len > 4);
+
 		addr_data |= I2C_SMBUS_READ;
 
 		/* SELECT command. */
-		mxs_i2c_pio_trigger_cmd(i2c, MXS_CMD_I2C_SELECT);
-
-		ret = mxs_i2c_pio_wait_dmareq(i2c);
-		if (ret)
-			return ret;
-
-		writel(addr_data, i2c->regs + MXS_I2C_DATA);
-		writel(MXS_I2C_DEBUG0_DMAREQ, i2c->regs + MXS_I2C_DEBUG0_CLR);
+		mxs_i2c_pio_trigger_write_cmd(i2c, MXS_CMD_I2C_SELECT,
+					      addr_data);
 
-		ret = mxs_i2c_pio_wait_cplt(i2c, 0);
-		if (ret)
-			return ret;
-
-		if (mxs_i2c_pio_check_error_state(i2c))
+		ret = mxs_i2c_pio_wait_xfer_end(i2c);
+		if (ret) {
+			dev_err(i2c->dev,
+				"PIO: Failed to send SELECT command!\n");
 			goto cleanup;
+		}
 
 		/* READ command. */
 		mxs_i2c_pio_trigger_cmd(i2c,
 					MXS_CMD_I2C_READ | flags |
 					MXS_I2C_CTRL0_XFER_COUNT(msg->len));
 
+		ret = mxs_i2c_pio_wait_xfer_end(i2c);
+		if (ret) {
+			dev_err(i2c->dev,
+				"PIO: Failed to send SELECT command!\n");
+			goto cleanup;
+		}
+
+		data = readl(i2c->regs + MXS_I2C_DATA(i2c));
 		for (i = 0; i < msg->len; i++) {
-			if ((i & 3) == 0) {
-				ret = mxs_i2c_pio_wait_dmareq(i2c);
-				if (ret)
-					return ret;
-				data = readl(i2c->regs + MXS_I2C_DATA);
-				writel(MXS_I2C_DEBUG0_DMAREQ,
-				       i2c->regs + MXS_I2C_DEBUG0_CLR);
-			}
 			msg->buf[i] = data & 0xff;
 			data >>= 8;
 		}
 	} else {
+		/*
+		 * PIO WRITE transfer:
+		 *
+		 * The code below implements clock stretching to circumvent
+		 * the possibility of kernel not being able to supply data
+		 * fast enough. It is possible to transfer arbitrary amount
+		 * of data using PIO write.
+		 */
 		addr_data |= I2C_SMBUS_WRITE;
 
-		/* WRITE command. */
-		mxs_i2c_pio_trigger_cmd(i2c,
-					MXS_CMD_I2C_WRITE | flags |
-					MXS_I2C_CTRL0_XFER_COUNT(msg->len + 1));
-
 		/*
 		 * The LSB of data buffer is the first byte blasted across
 		 * the bus. Higher order bytes follow. Thus the following
 		 * filling schematic.
 		 */
+
 		data = addr_data << 24;
+
+		/* Start the transfer with START condition. */
+		start = MXS_I2C_CTRL0_PRE_SEND_START;
+
+		/* If the transfer is long, use clock stretching. */
+		if (msg->len > 3)
+			start |= MXS_I2C_CTRL0_RETAIN_CLOCK;
+
 		for (i = 0; i < msg->len; i++) {
 			data >>= 8;
 			data |= (msg->buf[i] << 24);
-			if ((i & 3) == 2) {
-				ret = mxs_i2c_pio_wait_dmareq(i2c);
-				if (ret)
-					return ret;
-				writel(data, i2c->regs + MXS_I2C_DATA);
-				writel(MXS_I2C_DEBUG0_DMAREQ,
-				       i2c->regs + MXS_I2C_DEBUG0_CLR);
+
+			xmit = 0;
+
+			/* This is the last transfer of the message. */
+			if (i + 1 == msg->len) {
+				/* Add optional STOP flag. */
+				start |= flags;
+				/* Remove RETAIN_CLOCK bit. */
+				start &= ~MXS_I2C_CTRL0_RETAIN_CLOCK;
+				xmit = 1;
 			}
-		}
 
-		shifts_left = 24 - (i & 3) * 8;
-		if (shifts_left) {
-			data >>= shifts_left;
-			ret = mxs_i2c_pio_wait_dmareq(i2c);
-			if (ret)
-				return ret;
-			writel(data, i2c->regs + MXS_I2C_DATA);
+			/* Four bytes are ready in the "data" variable. */
+			if ((i & 3) == 2)
+				xmit = 1;
+
+			/* Nothing interesting happened, continue stuffing. */
+			if (!xmit)
+				continue;
+
+			/*
+			 * Compute the size of the transfer and shift the
+			 * data accordingly.
+			 *
+			 * i = (4k + 0) .... xlen = 2
+			 * i = (4k + 1) .... xlen = 3
+			 * i = (4k + 2) .... xlen = 4
+			 * i = (4k + 3) .... xlen = 1
+			 */
+
+			if ((i % 4) == 3)
+				xlen = 1;
+			else
+				xlen = (i % 4) + 2;
+
+			data >>= (4 - xlen) * 8;
+
+			dev_dbg(i2c->dev,
+				"PIO: len=%i pos=%i total=%i [W%s%s%s]\n",
+				xlen, i, msg->len,
+				start & MXS_I2C_CTRL0_PRE_SEND_START ? "S" : "",
+				start & MXS_I2C_CTRL0_POST_SEND_STOP ? "E" : "",
+				start & MXS_I2C_CTRL0_RETAIN_CLOCK ? "C" : "");
+
 			writel(MXS_I2C_DEBUG0_DMAREQ,
-			       i2c->regs + MXS_I2C_DEBUG0_CLR);
+			       i2c->regs + MXS_I2C_DEBUG0_CLR(i2c));
+
+			mxs_i2c_pio_trigger_write_cmd(i2c,
+				start | MXS_I2C_CTRL0_MASTER_MODE |
+				MXS_I2C_CTRL0_DIRECTION |
+				MXS_I2C_CTRL0_XFER_COUNT(xlen), data);
+
+			/* The START condition is sent only once. */
+			start &= ~MXS_I2C_CTRL0_PRE_SEND_START;
+
+			/* Wait for the end of the transfer. */
+			ret = mxs_i2c_pio_wait_xfer_end(i2c);
+			if (ret) {
+				dev_err(i2c->dev,
+					"PIO: Failed to finish WRITE cmd!\n");
+				break;
+			}
+
+			/* Check NAK here. */
+			ret = readl(i2c->regs + MXS_I2C_STAT) &
+				    MXS_I2C_STAT_GOT_A_NAK;
+			if (ret) {
+				ret = -ENXIO;
+				goto cleanup;
+			}
 		}
 	}
 
-	ret = mxs_i2c_pio_wait_cplt(i2c, flags & MXS_I2C_CTRL0_POST_SEND_STOP);
-	if (ret)
-		return ret;
-
 	/* make sure we capture any occurred error into cmd_err */
-	mxs_i2c_pio_check_error_state(i2c);
+	ret = mxs_i2c_pio_check_error_state(i2c);
 
 cleanup:
 	/* Clear any dangling IRQs and re-enable interrupts. */
 	writel(MXS_I2C_IRQ_MASK, i2c->regs + MXS_I2C_CTRL1_CLR);
 	writel(MXS_I2C_IRQ_MASK << 8, i2c->regs + MXS_I2C_CTRL1_SET);
 
-	return 0;
+	/* Clear the PIO_MODE on i.MX23 */
+	if (i2c->dev_type == MXS_I2C_V1)
+		writel(MXS_I2C_CTRL0_PIO_MODE, i2c->regs + MXS_I2C_CTRL0_CLR);
+
+	return ret;
 }
 
 /*
@@ -479,8 +562,9 @@ static int mxs_i2c_xfer_msg(struct i2c_adapter *adap, struct i2c_msg *msg,
 				int stop)
 {
 	struct mxs_i2c_dev *i2c = i2c_get_adapdata(adap);
-	int ret, err;
+	int ret;
 	int flags;
+	int use_pio = 0;
 
 	flags = stop ? MXS_I2C_CTRL0_POST_SEND_STOP : 0;
 
@@ -491,21 +575,23 @@ static int mxs_i2c_xfer_msg(struct i2c_adapter *adap, struct i2c_msg *msg,
 		return -EINVAL;
 
 	/*
-	 * The current boundary to select between PIO/DMA transfer method
-	 * is set to 8 bytes, transfers shorter than 8 bytes are transfered
-	 * using PIO mode while longer transfers use DMA. The 8 byte border is
-	 * based on this empirical measurement and a lot of previous frobbing.
+	 * The MX28 I2C IP block can only do PIO READ for transfer of to up
+	 * 4 bytes of length. The write transfer is not limited as it can use
+	 * clock stretching to avoid FIFO underruns.
 	 */
+	if ((msg->flags & I2C_M_RD) && (msg->len <= 4))
+		use_pio = 1;
+	if (!(msg->flags & I2C_M_RD) && (msg->len < 7))
+		use_pio = 1;
+
 	i2c->cmd_err = 0;
-	if (0) {	/* disable PIO mode until a proper fix is made */
+	if (use_pio) {
 		ret = mxs_i2c_pio_setup_xfer(adap, msg, flags);
-		if (ret) {
-			err = mxs_i2c_reset(i2c);
-			if (err)
-				return err;
-		}
+		/* No need to reset the block if NAK was received. */
+		if (ret && (ret != -ENXIO))
+			mxs_i2c_reset(i2c);
 	} else {
-		INIT_COMPLETION(i2c->cmd_complete);
+		reinit_completion(&i2c->cmd_complete);
 		ret = mxs_i2c_dma_setup_xfer(adap, msg, flags);
 		if (ret)
 			return ret;
@@ -514,9 +600,11 @@ static int mxs_i2c_xfer_msg(struct i2c_adapter *adap, struct i2c_msg *msg,
 						msecs_to_jiffies(1000));
 		if (ret == 0)
 			goto timeout;
+
+		ret = i2c->cmd_err;
 	}
 
-	if (i2c->cmd_err == -ENXIO) {
+	if (ret == -ENXIO) {
 		/*
 		 * If the transfer fails with a NAK from the slave the
 		 * controller halts until it gets told to return to idle state.
@@ -525,7 +613,19 @@ static int mxs_i2c_xfer_msg(struct i2c_adapter *adap, struct i2c_msg *msg,
 		       i2c->regs + MXS_I2C_CTRL1_SET);
 	}
 
-	ret = i2c->cmd_err;
+	/*
+	 * WARNING!
+	 * The i.MX23 is strange. After each and every operation, it's I2C IP
+	 * block must be reset, otherwise the IP block will misbehave. This can
+	 * be observed on the bus by the block sending out one single byte onto
+	 * the bus. In case such an error happens, bit 27 will be set in the
+	 * DEBUG0 register. This bit is not documented in the i.MX23 datasheet
+	 * and is marked as "TBD" instead. To reset this bit to a correct state,
+	 * reset the whole block. Since the block reset does not take long, do
+	 * reset the block after every transfer to play safe.
+	 */
+	if (i2c->dev_type == MXS_I2C_V1)
+		mxs_i2c_reset(i2c);
 
 	dev_dbg(i2c->dev, "Done with err=%d\n", ret);
 
@@ -680,8 +780,28 @@ static int mxs_i2c_get_ofdata(struct mxs_i2c_dev *i2c)
 	return 0;
 }
 
+static struct platform_device_id mxs_i2c_devtype[] = {
+	{
+		.name = "imx23-i2c",
+		.driver_data = MXS_I2C_V1,
+	}, {
+		.name = "imx28-i2c",
+		.driver_data = MXS_I2C_V2,
+	}, { /* sentinel */ }
+};
+MODULE_DEVICE_TABLE(platform, mxs_i2c_devtype);
+
+static const struct of_device_id mxs_i2c_dt_ids[] = {
+	{ .compatible = "fsl,imx23-i2c", .data = &mxs_i2c_devtype[0], },
+	{ .compatible = "fsl,imx28-i2c", .data = &mxs_i2c_devtype[1], },
+	{ /* sentinel */ }
+};
+MODULE_DEVICE_TABLE(of, mxs_i2c_dt_ids);
+
 static int mxs_i2c_probe(struct platform_device *pdev)
 {
+	const struct of_device_id *of_id =
+				of_match_device(mxs_i2c_dt_ids, &pdev->dev);
 	struct device *dev = &pdev->dev;
 	struct mxs_i2c_dev *i2c;
 	struct i2c_adapter *adap;
@@ -693,6 +813,11 @@ static int mxs_i2c_probe(struct platform_device *pdev)
 	if (!i2c)
 		return -ENOMEM;
 
+	if (of_id) {
+		const struct platform_device_id *device_id = of_id->data;
+		i2c->dev_type = device_id->driver_data;
+	}
+
 	res = platform_get_resource(pdev, IORESOURCE_MEM, 0);
 	irq = platform_get_irq(pdev, 0);
 
@@ -768,12 +893,6 @@ static int mxs_i2c_remove(struct platform_device *pdev)
 	return 0;
 }
 
-static const struct of_device_id mxs_i2c_dt_ids[] = {
-	{ .compatible = "fsl,imx28-i2c", },
-	{ /* sentinel */ }
-};
-MODULE_DEVICE_TABLE(of, mxs_i2c_dt_ids);
-
 static struct platform_driver mxs_i2c_driver = {
 	.driver = {
 		   .name = DRIVER_NAME,
@@ -796,6 +915,7 @@ static void __exit mxs_i2c_exit(void)
 }
 module_exit(mxs_i2c_exit);
 
+MODULE_AUTHOR("Marek Vasut <marex@denx.de>");
 MODULE_AUTHOR("Wolfram Sang <w.sang@pengutronix.de>");
 MODULE_DESCRIPTION("MXS I2C Bus Driver");
 MODULE_LICENSE("GPL");
diff --git a/drivers/i2c/busses/i2c-omap.c b/drivers/i2c/busses/i2c-omap.c
index 9967a6f9c2ff..a6a891d7970d 100644
--- a/drivers/i2c/busses/i2c-omap.c
+++ b/drivers/i2c/busses/i2c-omap.c
@@ -543,7 +543,7 @@ static int omap_i2c_xfer_msg(struct i2c_adapter *adap,
 	w |= OMAP_I2C_BUF_RXFIF_CLR | OMAP_I2C_BUF_TXFIF_CLR;
 	omap_i2c_write_reg(dev, OMAP_I2C_BUF_REG, w);
 
-	INIT_COMPLETION(dev->cmd_complete);
+	reinit_completion(&dev->cmd_complete);
 	dev->cmd_err = 0;
 
 	w = OMAP_I2C_CON_EN | OMAP_I2C_CON_MST | OMAP_I2C_CON_STT;
diff --git a/drivers/i2c/busses/i2c-pnx.c b/drivers/i2c/busses/i2c-pnx.c
index 1a9ea25f2314..c9a352f0a9a5 100644
--- a/drivers/i2c/busses/i2c-pnx.c
+++ b/drivers/i2c/busses/i2c-pnx.c
@@ -23,6 +23,7 @@
 #include <linux/err.h>
 #include <linux/clk.h>
 #include <linux/slab.h>
+#include <linux/of.h>
 
 #define I2C_PNX_TIMEOUT_DEFAULT		10 /* msec */
 #define I2C_PNX_SPEED_KHZ_DEFAULT	100
diff --git a/drivers/i2c/busses/i2c-rcar.c b/drivers/i2c/busses/i2c-rcar.c
index d2fe11da5e82..2c2fd7c2b116 100644
--- a/drivers/i2c/busses/i2c-rcar.c
+++ b/drivers/i2c/busses/i2c-rcar.c
@@ -33,6 +33,7 @@
 #include <linux/i2c/i2c-rcar.h>
 #include <linux/kernel.h>
 #include <linux/module.h>
+#include <linux/of_device.h>
 #include <linux/platform_device.h>
 #include <linux/pm_runtime.h>
 #include <linux/slab.h>
@@ -102,8 +103,8 @@ enum {
 #define ID_NACK		(1 << 4)
 
 enum rcar_i2c_type {
-	I2C_RCAR_H1,
-	I2C_RCAR_H2,
+	I2C_RCAR_GEN1,
+	I2C_RCAR_GEN2,
 };
 
 struct rcar_i2c_priv {
@@ -226,22 +227,23 @@ static int rcar_i2c_clock_calculate(struct rcar_i2c_priv *priv,
 				    u32 bus_speed,
 				    struct device *dev)
 {
-	struct clk *clkp = clk_get(NULL, "peripheral_clk");
+	struct clk *clkp = clk_get(dev, NULL);
 	u32 scgd, cdf;
 	u32 round, ick;
 	u32 scl;
 	u32 cdf_width;
+	unsigned long rate;
 
-	if (!clkp) {
-		dev_err(dev, "there is no peripheral_clk\n");
-		return -EIO;
+	if (IS_ERR(clkp)) {
+		dev_err(dev, "couldn't get clock\n");
+		return PTR_ERR(clkp);
 	}
 
 	switch (priv->devtype) {
-	case I2C_RCAR_H1:
+	case I2C_RCAR_GEN1:
 		cdf_width = 2;
 		break;
-	case I2C_RCAR_H2:
+	case I2C_RCAR_GEN2:
 		cdf_width = 3;
 		break;
 	default:
@@ -264,15 +266,14 @@ static int rcar_i2c_clock_calculate(struct rcar_i2c_priv *priv,
 	 * clkp : peripheral_clk
 	 * F[]  : integer up-valuation
 	 */
-	for (cdf = 0; cdf < (1 << cdf_width); cdf++) {
-		ick = clk_get_rate(clkp) / (1 + cdf);
-		if (ick < 20000000)
-			goto ick_find;
+	rate = clk_get_rate(clkp);
+	cdf = rate / 20000000;
+	if (cdf >= 1 << cdf_width) {
+		dev_err(dev, "Input clock %lu too high\n", rate);
+		return -EIO;
 	}
-	dev_err(dev, "there is no best CDF\n");
-	return -EIO;
+	ick = rate / (cdf + 1);
 
-ick_find:
 	/*
 	 * it is impossible to calculate large scale
 	 * number on u32. separate it
@@ -290,6 +291,12 @@ ick_find:
 	 *
 	 * Calculation result (= SCL) should be less than
 	 * bus_speed for hardware safety
+	 *
+	 * We could use something along the lines of
+	 *	div = ick / (bus_speed + 1) + 1;
+	 *	scgd = (div - 20 - round + 7) / 8;
+	 *	scl = ick / (20 + (scgd * 8) + round);
+	 * (not fully verified) but that would get pretty involved
 	 */
 	for (scgd = 0; scgd < 0x40; scgd++) {
 		scl = ick / (20 + (scgd * 8) + round);
@@ -306,7 +313,7 @@ scgd_find:
 	/*
 	 * keep icccr value
 	 */
-	priv->icccr = (scgd << (cdf_width) | cdf);
+	priv->icccr = scgd << cdf_width | cdf;
 
 	return 0;
 }
@@ -632,6 +639,15 @@ static const struct i2c_algorithm rcar_i2c_algo = {
 	.functionality	= rcar_i2c_func,
 };
 
+static const struct of_device_id rcar_i2c_dt_ids[] = {
+	{ .compatible = "renesas,i2c-rcar", .data = (void *)I2C_RCAR_GEN1 },
+	{ .compatible = "renesas,i2c-r8a7778", .data = (void *)I2C_RCAR_GEN1 },
+	{ .compatible = "renesas,i2c-r8a7779", .data = (void *)I2C_RCAR_GEN1 },
+	{ .compatible = "renesas,i2c-r8a7790", .data = (void *)I2C_RCAR_GEN2 },
+	{},
+};
+MODULE_DEVICE_TABLE(of, rcar_i2c_dt_ids);
+
 static int rcar_i2c_probe(struct platform_device *pdev)
 {
 	struct i2c_rcar_platform_data *pdata = dev_get_platdata(&pdev->dev);
@@ -649,10 +665,15 @@ static int rcar_i2c_probe(struct platform_device *pdev)
 	}
 
 	bus_speed = 100000; /* default 100 kHz */
-	if (pdata && pdata->bus_speed)
+	ret = of_property_read_u32(dev->of_node, "clock-frequency", &bus_speed);
+	if (ret < 0 && pdata && pdata->bus_speed)
 		bus_speed = pdata->bus_speed;
 
-	priv->devtype = platform_get_device_id(pdev)->driver_data;
+	if (pdev->dev.of_node)
+		priv->devtype = (long)of_match_device(rcar_i2c_dt_ids,
+						      dev)->data;
+	else
+		priv->devtype = platform_get_device_id(pdev)->driver_data;
 
 	ret = rcar_i2c_clock_calculate(priv, bus_speed, dev);
 	if (ret < 0)
@@ -673,6 +694,7 @@ static int rcar_i2c_probe(struct platform_device *pdev)
 	adap->class		= I2C_CLASS_HWMON | I2C_CLASS_SPD;
 	adap->retries		= 3;
 	adap->dev.parent	= dev;
+	adap->dev.of_node	= dev->of_node;
 	i2c_set_adapdata(adap, priv);
 	strlcpy(adap->name, pdev->name, sizeof(adap->name));
 
@@ -709,9 +731,9 @@ static int rcar_i2c_remove(struct platform_device *pdev)
 }
 
 static struct platform_device_id rcar_i2c_id_table[] = {
-	{ "i2c-rcar",		I2C_RCAR_H1 },
-	{ "i2c-rcar_h1",	I2C_RCAR_H1 },
-	{ "i2c-rcar_h2",	I2C_RCAR_H2 },
+	{ "i2c-rcar",		I2C_RCAR_GEN1 },
+	{ "i2c-rcar_gen1",	I2C_RCAR_GEN1 },
+	{ "i2c-rcar_gen2",	I2C_RCAR_GEN2 },
 	{},
 };
 MODULE_DEVICE_TABLE(platform, rcar_i2c_id_table);
@@ -720,6 +742,7 @@ static struct platform_driver rcar_i2c_driver = {
 	.driver	= {
 		.name	= "i2c-rcar",
 		.owner	= THIS_MODULE,
+		.of_match_table = rcar_i2c_dt_ids,
 	},
 	.probe		= rcar_i2c_probe,
 	.remove		= rcar_i2c_remove,
diff --git a/drivers/i2c/busses/i2c-s3c2410.c b/drivers/i2c/busses/i2c-s3c2410.c
index 3747b9bf67d6..bf8fb94ebc5d 100644
--- a/drivers/i2c/busses/i2c-s3c2410.c
+++ b/drivers/i2c/busses/i2c-s3c2410.c
@@ -36,6 +36,7 @@
 #include <linux/cpufreq.h>
 #include <linux/slab.h>
 #include <linux/io.h>
+#include <linux/of.h>
 #include <linux/of_gpio.h>
 #include <linux/pinctrl/consumer.h>
 
diff --git a/drivers/i2c/busses/i2c-scmi.c b/drivers/i2c/busses/i2c-scmi.c
index c447e8d40b78..599235514138 100644
--- a/drivers/i2c/busses/i2c-scmi.c
+++ b/drivers/i2c/busses/i2c-scmi.c
@@ -223,7 +223,7 @@ acpi_smbus_cmi_access(struct i2c_adapter *adap, u16 addr, unsigned short flags,
 		goto out;
 
 	obj = pkg->package.elements + 1;
-	if (obj == NULL || obj->type != ACPI_TYPE_INTEGER) {
+	if (obj->type != ACPI_TYPE_INTEGER) {
 		ACPI_ERROR((AE_INFO, "Invalid argument type"));
 		result = -EIO;
 		goto out;
@@ -235,7 +235,7 @@ acpi_smbus_cmi_access(struct i2c_adapter *adap, u16 addr, unsigned short flags,
 	case I2C_SMBUS_BYTE:
 	case I2C_SMBUS_BYTE_DATA:
 	case I2C_SMBUS_WORD_DATA:
-		if (obj == NULL || obj->type != ACPI_TYPE_INTEGER) {
+		if (obj->type != ACPI_TYPE_INTEGER) {
 			ACPI_ERROR((AE_INFO, "Invalid argument type"));
 			result = -EIO;
 			goto out;
@@ -246,7 +246,7 @@ acpi_smbus_cmi_access(struct i2c_adapter *adap, u16 addr, unsigned short flags,
 			data->byte = obj->integer.value;
 		break;
 	case I2C_SMBUS_BLOCK_DATA:
-		if (obj == NULL || obj->type != ACPI_TYPE_BUFFER) {
+		if (obj->type != ACPI_TYPE_BUFFER) {
 			ACPI_ERROR((AE_INFO, "Invalid argument type"));
 			result = -EIO;
 			goto out;
diff --git a/drivers/i2c/busses/i2c-sh_mobile.c b/drivers/i2c/busses/i2c-sh_mobile.c
index 55110ddbed1f..1d79585ba4b3 100644
--- a/drivers/i2c/busses/i2c-sh_mobile.c
+++ b/drivers/i2c/busses/i2c-sh_mobile.c
@@ -235,7 +235,7 @@ static void sh_mobile_i2c_init(struct sh_mobile_i2c_data *pd)
 	int offset;
 
 	/* Get clock rate after clock is enabled */
-	clk_enable(pd->clk);
+	clk_prepare_enable(pd->clk);
 	i2c_clk_khz = clk_get_rate(pd->clk) / 1000;
 	i2c_clk_khz /= pd->clks_per_count;
 
@@ -270,14 +270,14 @@ static void sh_mobile_i2c_init(struct sh_mobile_i2c_data *pd)
 		pd->icic &= ~ICIC_ICCHB8;
 
 out:
-	clk_disable(pd->clk);
+	clk_disable_unprepare(pd->clk);
 }
 
 static void activate_ch(struct sh_mobile_i2c_data *pd)
 {
 	/* Wake up device and enable clock */
 	pm_runtime_get_sync(pd->dev);
-	clk_enable(pd->clk);
+	clk_prepare_enable(pd->clk);
 
 	/* Enable channel and configure rx ack */
 	iic_set_clr(pd, ICCR, ICCR_ICE, 0);
@@ -300,7 +300,7 @@ static void deactivate_ch(struct sh_mobile_i2c_data *pd)
 	iic_set_clr(pd, ICCR, 0, ICCR_ICE);
 
 	/* Disable clock and mark device as idle */
-	clk_disable(pd->clk);
+	clk_disable_unprepare(pd->clk);
 	pm_runtime_put_sync(pd->dev);
 }
 
diff --git a/drivers/i2c/busses/i2c-st.c b/drivers/i2c/busses/i2c-st.c
new file mode 100644
index 000000000000..9cf715d69551
--- /dev/null
+++ b/drivers/i2c/busses/i2c-st.c
@@ -0,0 +1,872 @@
+/*
+ * Copyright (C) 2013 STMicroelectronics
+ *
+ * I2C master mode controller driver, used in STMicroelectronics devices.
+ *
+ * Author: Maxime Coquelin <maxime.coquelin@st.com>
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2, as
+ * published by the Free Software Foundation.
+ */
+
+#include <linux/module.h>
+#include <linux/platform_device.h>
+#include <linux/i2c.h>
+#include <linux/clk.h>
+#include <linux/io.h>
+#include <linux/delay.h>
+#include <linux/interrupt.h>
+#include <linux/err.h>
+#include <linux/of.h>
+#include <linux/of_address.h>
+#include <linux/of_irq.h>
+
+/* SSC registers */
+#define SSC_BRG				0x000
+#define SSC_TBUF			0x004
+#define SSC_RBUF			0x008
+#define SSC_CTL				0x00C
+#define SSC_IEN				0x010
+#define SSC_STA				0x014
+#define SSC_I2C				0x018
+#define SSC_SLAD			0x01C
+#define SSC_REP_START_HOLD		0x020
+#define SSC_START_HOLD			0x024
+#define SSC_REP_START_SETUP		0x028
+#define SSC_DATA_SETUP			0x02C
+#define SSC_STOP_SETUP			0x030
+#define SSC_BUS_FREE			0x034
+#define SSC_TX_FSTAT			0x038
+#define SSC_RX_FSTAT			0x03C
+#define SSC_PRE_SCALER_BRG		0x040
+#define SSC_CLR				0x080
+#define SSC_NOISE_SUPP_WIDTH		0x100
+#define SSC_PRSCALER			0x104
+#define SSC_NOISE_SUPP_WIDTH_DATAOUT	0x108
+#define SSC_PRSCALER_DATAOUT		0x10c
+
+/* SSC Control */
+#define SSC_CTL_DATA_WIDTH_9		0x8
+#define SSC_CTL_DATA_WIDTH_MSK		0xf
+#define SSC_CTL_BM			0xf
+#define SSC_CTL_HB			BIT(4)
+#define SSC_CTL_PH			BIT(5)
+#define SSC_CTL_PO			BIT(6)
+#define SSC_CTL_SR			BIT(7)
+#define SSC_CTL_MS			BIT(8)
+#define SSC_CTL_EN			BIT(9)
+#define SSC_CTL_LPB			BIT(10)
+#define SSC_CTL_EN_TX_FIFO		BIT(11)
+#define SSC_CTL_EN_RX_FIFO		BIT(12)
+#define SSC_CTL_EN_CLST_RX		BIT(13)
+
+/* SSC Interrupt Enable */
+#define SSC_IEN_RIEN			BIT(0)
+#define SSC_IEN_TIEN			BIT(1)
+#define SSC_IEN_TEEN			BIT(2)
+#define SSC_IEN_REEN			BIT(3)
+#define SSC_IEN_PEEN			BIT(4)
+#define SSC_IEN_AASEN			BIT(6)
+#define SSC_IEN_STOPEN			BIT(7)
+#define SSC_IEN_ARBLEN			BIT(8)
+#define SSC_IEN_NACKEN			BIT(10)
+#define SSC_IEN_REPSTRTEN		BIT(11)
+#define SSC_IEN_TX_FIFO_HALF		BIT(12)
+#define SSC_IEN_RX_FIFO_HALF_FULL	BIT(14)
+
+/* SSC Status */
+#define SSC_STA_RIR			BIT(0)
+#define SSC_STA_TIR			BIT(1)
+#define SSC_STA_TE			BIT(2)
+#define SSC_STA_RE			BIT(3)
+#define SSC_STA_PE			BIT(4)
+#define SSC_STA_CLST			BIT(5)
+#define SSC_STA_AAS			BIT(6)
+#define SSC_STA_STOP			BIT(7)
+#define SSC_STA_ARBL			BIT(8)
+#define SSC_STA_BUSY			BIT(9)
+#define SSC_STA_NACK			BIT(10)
+#define SSC_STA_REPSTRT			BIT(11)
+#define SSC_STA_TX_FIFO_HALF		BIT(12)
+#define SSC_STA_TX_FIFO_FULL		BIT(13)
+#define SSC_STA_RX_FIFO_HALF		BIT(14)
+
+/* SSC I2C Control */
+#define SSC_I2C_I2CM			BIT(0)
+#define SSC_I2C_STRTG			BIT(1)
+#define SSC_I2C_STOPG			BIT(2)
+#define SSC_I2C_ACKG			BIT(3)
+#define SSC_I2C_AD10			BIT(4)
+#define SSC_I2C_TXENB			BIT(5)
+#define SSC_I2C_REPSTRTG		BIT(11)
+#define SSC_I2C_SLAVE_DISABLE		BIT(12)
+
+/* SSC Tx FIFO Status */
+#define SSC_TX_FSTAT_STATUS		0x07
+
+/* SSC Rx FIFO Status */
+#define SSC_RX_FSTAT_STATUS		0x07
+
+/* SSC Clear bit operation */
+#define SSC_CLR_SSCAAS			BIT(6)
+#define SSC_CLR_SSCSTOP			BIT(7)
+#define SSC_CLR_SSCARBL			BIT(8)
+#define SSC_CLR_NACK			BIT(10)
+#define SSC_CLR_REPSTRT			BIT(11)
+
+/* SSC Clock Prescaler */
+#define SSC_PRSC_VALUE			0x0f
+
+
+#define SSC_TXFIFO_SIZE			0x8
+#define SSC_RXFIFO_SIZE			0x8
+
+enum st_i2c_mode {
+	I2C_MODE_STANDARD,
+	I2C_MODE_FAST,
+	I2C_MODE_END,
+};
+
+/**
+ * struct st_i2c_timings - per-Mode tuning parameters
+ * @rate: I2C bus rate
+ * @rep_start_hold: I2C repeated start hold time requirement
+ * @rep_start_setup: I2C repeated start set up time requirement
+ * @start_hold: I2C start hold time requirement
+ * @data_setup_time: I2C data set up time requirement
+ * @stop_setup_time: I2C stop set up time requirement
+ * @bus_free_time: I2C bus free time requirement
+ * @sda_pulse_min_limit: I2C SDA pulse mini width limit
+ */
+struct st_i2c_timings {
+	u32 rate;
+	u32 rep_start_hold;
+	u32 rep_start_setup;
+	u32 start_hold;
+	u32 data_setup_time;
+	u32 stop_setup_time;
+	u32 bus_free_time;
+	u32 sda_pulse_min_limit;
+};
+
+/**
+ * struct st_i2c_client - client specific data
+ * @addr: 8-bit slave addr, including r/w bit
+ * @count: number of bytes to be transfered
+ * @xfered: number of bytes already transferred
+ * @buf: data buffer
+ * @result: result of the transfer
+ * @stop: last I2C msg to be sent, i.e. STOP to be generated
+ */
+struct st_i2c_client {
+	u8	addr;
+	u32	count;
+	u32	xfered;
+	u8	*buf;
+	int	result;
+	bool	stop;
+};
+
+/**
+ * struct st_i2c_dev - private data of the controller
+ * @adap: I2C adapter for this controller
+ * @dev: device for this controller
+ * @base: virtual memory area
+ * @complete: completion of I2C message
+ * @irq: interrupt line for th controller
+ * @clk: hw ssc block clock
+ * @mode: I2C mode of the controller. Standard or Fast only supported
+ * @scl_min_width_us: SCL line minimum pulse width in us
+ * @sda_min_width_us: SDA line minimum pulse width in us
+ * @client: I2C transfert information
+ * @busy: I2C transfer on-going
+ */
+struct st_i2c_dev {
+	struct i2c_adapter	adap;
+	struct device		*dev;
+	void __iomem		*base;
+	struct completion	complete;
+	int			irq;
+	struct clk		*clk;
+	int			mode;
+	u32			scl_min_width_us;
+	u32			sda_min_width_us;
+	struct st_i2c_client	client;
+	bool			busy;
+};
+
+static inline void st_i2c_set_bits(void __iomem *reg, u32 mask)
+{
+	writel_relaxed(readl_relaxed(reg) | mask, reg);
+}
+
+static inline void st_i2c_clr_bits(void __iomem *reg, u32 mask)
+{
+	writel_relaxed(readl_relaxed(reg) & ~mask, reg);
+}
+
+/* From I2C Specifications v0.5 */
+static struct st_i2c_timings i2c_timings[] = {
+	[I2C_MODE_STANDARD] = {
+		.rate			= 100000,
+		.rep_start_hold		= 4000,
+		.rep_start_setup	= 4700,
+		.start_hold		= 4000,
+		.data_setup_time	= 250,
+		.stop_setup_time	= 4000,
+		.bus_free_time		= 4700,
+	},
+	[I2C_MODE_FAST] = {
+		.rate			= 400000,
+		.rep_start_hold		= 600,
+		.rep_start_setup	= 600,
+		.start_hold		= 600,
+		.data_setup_time	= 100,
+		.stop_setup_time	= 600,
+		.bus_free_time		= 1300,
+	},
+};
+
+static void st_i2c_flush_rx_fifo(struct st_i2c_dev *i2c_dev)
+{
+	int count, i;
+
+	/*
+	 * Counter only counts up to 7 but fifo size is 8...
+	 * When fifo is full, counter is 0 and RIR bit of status register is
+	 * set
+	 */
+	if (readl_relaxed(i2c_dev->base + SSC_STA) & SSC_STA_RIR)
+		count = SSC_RXFIFO_SIZE;
+	else
+		count = readl_relaxed(i2c_dev->base + SSC_RX_FSTAT) &
+			SSC_RX_FSTAT_STATUS;
+
+	for (i = 0; i < count; i++)
+		readl_relaxed(i2c_dev->base + SSC_RBUF);
+}
+
+static void st_i2c_soft_reset(struct st_i2c_dev *i2c_dev)
+{
+	/*
+	 * FIFO needs to be emptied before reseting the IP,
+	 * else the controller raises a BUSY error.
+	 */
+	st_i2c_flush_rx_fifo(i2c_dev);
+
+	st_i2c_set_bits(i2c_dev->base + SSC_CTL, SSC_CTL_SR);
+	st_i2c_clr_bits(i2c_dev->base + SSC_CTL, SSC_CTL_SR);
+}
+
+/**
+ * st_i2c_hw_config() - Prepare SSC block, calculate and apply tuning timings
+ * @i2c_dev: Controller's private data
+ */
+static void st_i2c_hw_config(struct st_i2c_dev *i2c_dev)
+{
+	unsigned long rate;
+	u32 val, ns_per_clk;
+	struct st_i2c_timings *t = &i2c_timings[i2c_dev->mode];
+
+	st_i2c_soft_reset(i2c_dev);
+
+	val = SSC_CLR_REPSTRT | SSC_CLR_NACK | SSC_CLR_SSCARBL |
+		SSC_CLR_SSCAAS | SSC_CLR_SSCSTOP;
+	writel_relaxed(val, i2c_dev->base + SSC_CLR);
+
+	/* SSC Control register setup */
+	val = SSC_CTL_PO | SSC_CTL_PH | SSC_CTL_HB | SSC_CTL_DATA_WIDTH_9;
+	writel_relaxed(val, i2c_dev->base + SSC_CTL);
+
+	rate = clk_get_rate(i2c_dev->clk);
+	ns_per_clk = 1000000000 / rate;
+
+	/* Baudrate */
+	val = rate / (2 * t->rate);
+	writel_relaxed(val, i2c_dev->base + SSC_BRG);
+
+	/* Pre-scaler baudrate */
+	writel_relaxed(1, i2c_dev->base + SSC_PRE_SCALER_BRG);
+
+	/* Enable I2C mode */
+	writel_relaxed(SSC_I2C_I2CM, i2c_dev->base + SSC_I2C);
+
+	/* Repeated start hold time */
+	val = t->rep_start_hold / ns_per_clk;
+	writel_relaxed(val, i2c_dev->base + SSC_REP_START_HOLD);
+
+	/* Repeated start set up time */
+	val = t->rep_start_setup / ns_per_clk;
+	writel_relaxed(val, i2c_dev->base + SSC_REP_START_SETUP);
+
+	/* Start hold time */
+	val = t->start_hold / ns_per_clk;
+	writel_relaxed(val, i2c_dev->base + SSC_START_HOLD);
+
+	/* Data set up time */
+	val = t->data_setup_time / ns_per_clk;
+	writel_relaxed(val, i2c_dev->base + SSC_DATA_SETUP);
+
+	/* Stop set up time */
+	val = t->stop_setup_time / ns_per_clk;
+	writel_relaxed(val, i2c_dev->base + SSC_STOP_SETUP);
+
+	/* Bus free time */
+	val = t->bus_free_time / ns_per_clk;
+	writel_relaxed(val, i2c_dev->base + SSC_BUS_FREE);
+
+	/* Prescalers set up */
+	val = rate / 10000000;
+	writel_relaxed(val, i2c_dev->base + SSC_PRSCALER);
+	writel_relaxed(val, i2c_dev->base + SSC_PRSCALER_DATAOUT);
+
+	/* Noise suppression witdh */
+	val = i2c_dev->scl_min_width_us * rate / 100000000;
+	writel_relaxed(val, i2c_dev->base + SSC_NOISE_SUPP_WIDTH);
+
+	/* Noise suppression max output data delay width */
+	val = i2c_dev->sda_min_width_us * rate / 100000000;
+	writel_relaxed(val, i2c_dev->base + SSC_NOISE_SUPP_WIDTH_DATAOUT);
+}
+
+static int st_i2c_wait_free_bus(struct st_i2c_dev *i2c_dev)
+{
+	u32 sta;
+	int i;
+
+	for (i = 0; i < 10; i++) {
+		sta = readl_relaxed(i2c_dev->base + SSC_STA);
+		if (!(sta & SSC_STA_BUSY))
+			return 0;
+
+		usleep_range(2000, 4000);
+	}
+
+	dev_err(i2c_dev->dev, "bus not free (status = 0x%08x)\n", sta);
+
+	return -EBUSY;
+}
+
+/**
+ * st_i2c_write_tx_fifo() - Write a byte in the Tx FIFO
+ * @i2c_dev: Controller's private data
+ * @byte: Data to write in the Tx FIFO
+ */
+static inline void st_i2c_write_tx_fifo(struct st_i2c_dev *i2c_dev, u8 byte)
+{
+	u16 tbuf = byte << 1;
+
+	writel_relaxed(tbuf | 1, i2c_dev->base + SSC_TBUF);
+}
+
+/**
+ * st_i2c_wr_fill_tx_fifo() - Fill the Tx FIFO in write mode
+ * @i2c_dev: Controller's private data
+ *
+ * This functions fills the Tx FIFO with I2C transfert buffer when
+ * in write mode.
+ */
+static void st_i2c_wr_fill_tx_fifo(struct st_i2c_dev *i2c_dev)
+{
+	struct st_i2c_client *c = &i2c_dev->client;
+	u32 tx_fstat, sta;
+	int i;
+
+	sta = readl_relaxed(i2c_dev->base + SSC_STA);
+	if (sta & SSC_STA_TX_FIFO_FULL)
+		return;
+
+	tx_fstat = readl_relaxed(i2c_dev->base + SSC_TX_FSTAT);
+	tx_fstat &= SSC_TX_FSTAT_STATUS;
+
+	if (c->count < (SSC_TXFIFO_SIZE - tx_fstat))
+		i = c->count;
+	else
+		i = SSC_TXFIFO_SIZE - tx_fstat;
+
+	for (; i > 0; i--, c->count--, c->buf++)
+		st_i2c_write_tx_fifo(i2c_dev, *c->buf);
+}
+
+/**
+ * st_i2c_rd_fill_tx_fifo() - Fill the Tx FIFO in read mode
+ * @i2c_dev: Controller's private data
+ *
+ * This functions fills the Tx FIFO with fixed pattern when
+ * in read mode to trigger clock.
+ */
+static void st_i2c_rd_fill_tx_fifo(struct st_i2c_dev *i2c_dev, int max)
+{
+	struct st_i2c_client *c = &i2c_dev->client;
+	u32 tx_fstat, sta;
+	int i;
+
+	sta = readl_relaxed(i2c_dev->base + SSC_STA);
+	if (sta & SSC_STA_TX_FIFO_FULL)
+		return;
+
+	tx_fstat = readl_relaxed(i2c_dev->base + SSC_TX_FSTAT);
+	tx_fstat &= SSC_TX_FSTAT_STATUS;
+
+	if (max < (SSC_TXFIFO_SIZE - tx_fstat))
+		i = max;
+	else
+		i = SSC_TXFIFO_SIZE - tx_fstat;
+
+	for (; i > 0; i--, c->xfered++)
+		st_i2c_write_tx_fifo(i2c_dev, 0xff);
+}
+
+static void st_i2c_read_rx_fifo(struct st_i2c_dev *i2c_dev)
+{
+	struct st_i2c_client *c = &i2c_dev->client;
+	u32 i, sta;
+	u16 rbuf;
+
+	sta = readl_relaxed(i2c_dev->base + SSC_STA);
+	if (sta & SSC_STA_RIR) {
+		i = SSC_RXFIFO_SIZE;
+	} else {
+		i = readl_relaxed(i2c_dev->base + SSC_RX_FSTAT);
+		i &= SSC_RX_FSTAT_STATUS;
+	}
+
+	for (; (i > 0) && (c->count > 0); i--, c->count--) {
+		rbuf = readl_relaxed(i2c_dev->base + SSC_RBUF) >> 1;
+		*c->buf++ = (u8)rbuf & 0xff;
+	}
+
+	if (i) {
+		dev_err(i2c_dev->dev, "Unexpected %d bytes in rx fifo\n", i);
+		st_i2c_flush_rx_fifo(i2c_dev);
+	}
+}
+
+/**
+ * st_i2c_terminate_xfer() - Send either STOP or REPSTART condition
+ * @i2c_dev: Controller's private data
+ */
+static void st_i2c_terminate_xfer(struct st_i2c_dev *i2c_dev)
+{
+	struct st_i2c_client *c = &i2c_dev->client;
+
+	st_i2c_clr_bits(i2c_dev->base + SSC_IEN, SSC_IEN_TEEN);
+	st_i2c_clr_bits(i2c_dev->base + SSC_I2C, SSC_I2C_STRTG);
+
+	if (c->stop) {
+		st_i2c_set_bits(i2c_dev->base + SSC_IEN, SSC_IEN_STOPEN);
+		st_i2c_set_bits(i2c_dev->base + SSC_I2C, SSC_I2C_STOPG);
+	} else {
+		st_i2c_set_bits(i2c_dev->base + SSC_IEN, SSC_IEN_REPSTRTEN);
+		st_i2c_set_bits(i2c_dev->base + SSC_I2C, SSC_I2C_REPSTRTG);
+	}
+}
+
+/**
+ * st_i2c_handle_write() - Handle FIFO empty interrupt in case of write
+ * @i2c_dev: Controller's private data
+ */
+static void st_i2c_handle_write(struct st_i2c_dev *i2c_dev)
+{
+	struct st_i2c_client *c = &i2c_dev->client;
+
+	st_i2c_flush_rx_fifo(i2c_dev);
+
+	if (!c->count)
+		/* End of xfer, send stop or repstart */
+		st_i2c_terminate_xfer(i2c_dev);
+	else
+		st_i2c_wr_fill_tx_fifo(i2c_dev);
+}
+
+/**
+ * st_i2c_handle_write() - Handle FIFO enmpty interrupt in case of read
+ * @i2c_dev: Controller's private data
+ */
+static void st_i2c_handle_read(struct st_i2c_dev *i2c_dev)
+{
+	struct st_i2c_client *c = &i2c_dev->client;
+	u32 ien;
+
+	/* Trash the address read back */
+	if (!c->xfered) {
+		readl_relaxed(i2c_dev->base + SSC_RBUF);
+		st_i2c_clr_bits(i2c_dev->base + SSC_I2C, SSC_I2C_TXENB);
+	} else {
+		st_i2c_read_rx_fifo(i2c_dev);
+	}
+
+	if (!c->count) {
+		/* End of xfer, send stop or repstart */
+		st_i2c_terminate_xfer(i2c_dev);
+	} else if (c->count == 1) {
+		/* Penultimate byte to xfer, disable ACK gen. */
+		st_i2c_clr_bits(i2c_dev->base + SSC_I2C, SSC_I2C_ACKG);
+
+		/* Last received byte is to be handled by NACK interrupt */
+		ien = SSC_IEN_NACKEN | SSC_IEN_ARBLEN;
+		writel_relaxed(ien, i2c_dev->base + SSC_IEN);
+
+		st_i2c_rd_fill_tx_fifo(i2c_dev, c->count);
+	} else {
+		st_i2c_rd_fill_tx_fifo(i2c_dev, c->count - 1);
+	}
+}
+
+/**
+ * st_i2c_isr() - Interrupt routine
+ * @irq: interrupt number
+ * @data: Controller's private data
+ */
+static irqreturn_t st_i2c_isr_thread(int irq, void *data)
+{
+	struct st_i2c_dev *i2c_dev = data;
+	struct st_i2c_client *c = &i2c_dev->client;
+	u32 sta, ien;
+	int it;
+
+	ien = readl_relaxed(i2c_dev->base + SSC_IEN);
+	sta = readl_relaxed(i2c_dev->base + SSC_STA);
+
+	/* Use __fls() to check error bits first */
+	it = __fls(sta & ien);
+	if (it < 0) {
+		dev_dbg(i2c_dev->dev, "spurious it (sta=0x%04x, ien=0x%04x)\n",
+				sta, ien);
+		return IRQ_NONE;
+	}
+
+	switch (1 << it) {
+	case SSC_STA_TE:
+		if (c->addr & I2C_M_RD)
+			st_i2c_handle_read(i2c_dev);
+		else
+			st_i2c_handle_write(i2c_dev);
+		break;
+
+	case SSC_STA_STOP:
+	case SSC_STA_REPSTRT:
+		writel_relaxed(0, i2c_dev->base + SSC_IEN);
+		complete(&i2c_dev->complete);
+		break;
+
+	case SSC_STA_NACK:
+		writel_relaxed(SSC_CLR_NACK, i2c_dev->base + SSC_CLR);
+
+		/* Last received byte handled by NACK interrupt */
+		if ((c->addr & I2C_M_RD) && (c->count == 1) && (c->xfered)) {
+			st_i2c_handle_read(i2c_dev);
+			break;
+		}
+
+		it = SSC_IEN_STOPEN | SSC_IEN_ARBLEN;
+		writel_relaxed(it, i2c_dev->base + SSC_IEN);
+
+		st_i2c_set_bits(i2c_dev->base + SSC_I2C, SSC_I2C_STOPG);
+		c->result = -EIO;
+		break;
+
+	case SSC_STA_ARBL:
+		writel_relaxed(SSC_CLR_SSCARBL, i2c_dev->base + SSC_CLR);
+
+		it = SSC_IEN_STOPEN | SSC_IEN_ARBLEN;
+		writel_relaxed(it, i2c_dev->base + SSC_IEN);
+
+		st_i2c_set_bits(i2c_dev->base + SSC_I2C, SSC_I2C_STOPG);
+		c->result = -EIO;
+		break;
+
+	default:
+		dev_err(i2c_dev->dev,
+				"it %d unhandled (sta=0x%04x)\n", it, sta);
+	}
+
+	/*
+	 * Read IEN register to ensure interrupt mask write is effective
+	 * before re-enabling interrupt at GIC level, and thus avoid spurious
+	 * interrupts.
+	 */
+	readl(i2c_dev->base + SSC_IEN);
+
+	return IRQ_HANDLED;
+}
+
+/**
+ * st_i2c_xfer_msg() - Transfer a single I2C message
+ * @i2c_dev: Controller's private data
+ * @msg: I2C message to transfer
+ * @is_first: first message of the sequence
+ * @is_last: last message of the sequence
+ */
+static int st_i2c_xfer_msg(struct st_i2c_dev *i2c_dev, struct i2c_msg *msg,
+			    bool is_first, bool is_last)
+{
+	struct st_i2c_client *c = &i2c_dev->client;
+	u32 ctl, i2c, it;
+	unsigned long timeout;
+	int ret;
+
+	c->addr		= (u8)(msg->addr << 1);
+	c->addr		|= (msg->flags & I2C_M_RD);
+	c->buf		= msg->buf;
+	c->count	= msg->len;
+	c->xfered	= 0;
+	c->result	= 0;
+	c->stop		= is_last;
+
+	reinit_completion(&i2c_dev->complete);
+
+	ctl = SSC_CTL_EN | SSC_CTL_MS |	SSC_CTL_EN_RX_FIFO | SSC_CTL_EN_TX_FIFO;
+	st_i2c_set_bits(i2c_dev->base + SSC_CTL, ctl);
+
+	i2c = SSC_I2C_TXENB;
+	if (c->addr & I2C_M_RD)
+		i2c |= SSC_I2C_ACKG;
+	st_i2c_set_bits(i2c_dev->base + SSC_I2C, i2c);
+
+	/* Write slave address */
+	st_i2c_write_tx_fifo(i2c_dev, c->addr);
+
+	/* Pre-fill Tx fifo with data in case of write */
+	if (!(c->addr & I2C_M_RD))
+		st_i2c_wr_fill_tx_fifo(i2c_dev);
+
+	it = SSC_IEN_NACKEN | SSC_IEN_TEEN | SSC_IEN_ARBLEN;
+	writel_relaxed(it, i2c_dev->base + SSC_IEN);
+
+	if (is_first) {
+		ret = st_i2c_wait_free_bus(i2c_dev);
+		if (ret)
+			return ret;
+
+		st_i2c_set_bits(i2c_dev->base + SSC_I2C, SSC_I2C_STRTG);
+	}
+
+	timeout = wait_for_completion_timeout(&i2c_dev->complete,
+			i2c_dev->adap.timeout);
+	ret = c->result;
+
+	if (!timeout) {
+		dev_err(i2c_dev->dev, "Write to slave 0x%x timed out\n",
+				c->addr);
+		ret = -ETIMEDOUT;
+	}
+
+	i2c = SSC_I2C_STOPG | SSC_I2C_REPSTRTG;
+	st_i2c_clr_bits(i2c_dev->base + SSC_I2C, i2c);
+
+	writel_relaxed(SSC_CLR_SSCSTOP | SSC_CLR_REPSTRT,
+			i2c_dev->base + SSC_CLR);
+
+	return ret;
+}
+
+/**
+ * st_i2c_xfer() - Transfer a single I2C message
+ * @i2c_adap: Adapter pointer to the controller
+ * @msgs: Pointer to data to be written.
+ * @num: Number of messages to be executed
+ */
+static int st_i2c_xfer(struct i2c_adapter *i2c_adap,
+			struct i2c_msg msgs[], int num)
+{
+	struct st_i2c_dev *i2c_dev = i2c_get_adapdata(i2c_adap);
+	int ret, i;
+
+	i2c_dev->busy = true;
+
+	ret = clk_prepare_enable(i2c_dev->clk);
+	if (ret) {
+		dev_err(i2c_dev->dev, "Failed to prepare_enable clock\n");
+		return ret;
+	}
+
+	pinctrl_pm_select_default_state(i2c_dev->dev);
+
+	st_i2c_hw_config(i2c_dev);
+
+	for (i = 0; (i < num) && !ret; i++)
+		ret = st_i2c_xfer_msg(i2c_dev, &msgs[i], i == 0, i == num - 1);
+
+	pinctrl_pm_select_idle_state(i2c_dev->dev);
+
+	clk_disable_unprepare(i2c_dev->clk);
+
+	i2c_dev->busy = false;
+
+	return (ret < 0) ? ret : i;
+}
+
+#ifdef CONFIG_PM_SLEEP
+static int st_i2c_suspend(struct device *dev)
+{
+	struct platform_device *pdev =
+		container_of(dev, struct platform_device, dev);
+	struct st_i2c_dev *i2c_dev = platform_get_drvdata(pdev);
+
+	if (i2c_dev->busy)
+		return -EBUSY;
+
+	pinctrl_pm_select_sleep_state(dev);
+
+	return 0;
+}
+
+static int st_i2c_resume(struct device *dev)
+{
+	pinctrl_pm_select_default_state(dev);
+	/* Go in idle state if available */
+	pinctrl_pm_select_idle_state(dev);
+
+	return 0;
+}
+
+static SIMPLE_DEV_PM_OPS(st_i2c_pm, st_i2c_suspend, st_i2c_resume);
+#define ST_I2C_PM	(&st_i2c_pm)
+#else
+#define ST_I2C_PM	NULL
+#endif
+
+static u32 st_i2c_func(struct i2c_adapter *adap)
+{
+	return I2C_FUNC_I2C | I2C_FUNC_SMBUS_EMUL;
+}
+
+static struct i2c_algorithm st_i2c_algo = {
+	.master_xfer = st_i2c_xfer,
+	.functionality = st_i2c_func,
+};
+
+static int st_i2c_of_get_deglitch(struct device_node *np,
+		struct st_i2c_dev *i2c_dev)
+{
+	int ret;
+
+	ret = of_property_read_u32(np, "st,i2c-min-scl-pulse-width-us",
+			&i2c_dev->scl_min_width_us);
+	if ((ret == -ENODATA) || (ret == -EOVERFLOW)) {
+		dev_err(i2c_dev->dev, "st,i2c-min-scl-pulse-width-us invalid\n");
+		return ret;
+	}
+
+	ret = of_property_read_u32(np, "st,i2c-min-sda-pulse-width-us",
+			&i2c_dev->sda_min_width_us);
+	if ((ret == -ENODATA) || (ret == -EOVERFLOW)) {
+		dev_err(i2c_dev->dev, "st,i2c-min-sda-pulse-width-us invalid\n");
+		return ret;
+	}
+
+	return 0;
+}
+
+static int st_i2c_probe(struct platform_device *pdev)
+{
+	struct device_node *np = pdev->dev.of_node;
+	struct st_i2c_dev *i2c_dev;
+	struct resource *res;
+	u32 clk_rate;
+	struct i2c_adapter *adap;
+	int ret;
+
+	i2c_dev = devm_kzalloc(&pdev->dev, sizeof(*i2c_dev), GFP_KERNEL);
+	if (!i2c_dev)
+		return -ENOMEM;
+
+	res = platform_get_resource(pdev, IORESOURCE_MEM, 0);
+	i2c_dev->base = devm_ioremap_resource(&pdev->dev, res);
+	if (IS_ERR(i2c_dev->base))
+		return PTR_ERR(i2c_dev->base);
+
+	i2c_dev->irq = irq_of_parse_and_map(np, 0);
+	if (!i2c_dev->irq) {
+		dev_err(&pdev->dev, "IRQ missing or invalid\n");
+		return -EINVAL;
+	}
+
+	i2c_dev->clk = of_clk_get_by_name(np, "ssc");
+	if (IS_ERR(i2c_dev->clk)) {
+		dev_err(&pdev->dev, "Unable to request clock\n");
+		return PTR_ERR(i2c_dev->clk);
+	}
+
+	i2c_dev->mode = I2C_MODE_STANDARD;
+	ret = of_property_read_u32(np, "clock-frequency", &clk_rate);
+	if ((!ret) && (clk_rate == 400000))
+		i2c_dev->mode = I2C_MODE_FAST;
+
+	i2c_dev->dev = &pdev->dev;
+
+	ret = devm_request_threaded_irq(&pdev->dev, i2c_dev->irq,
+			NULL, st_i2c_isr_thread,
+			IRQF_ONESHOT, pdev->name, i2c_dev);
+	if (ret) {
+		dev_err(&pdev->dev, "Failed to request irq %i\n", i2c_dev->irq);
+		return ret;
+	}
+
+	pinctrl_pm_select_default_state(i2c_dev->dev);
+	/* In case idle state available, select it */
+	pinctrl_pm_select_idle_state(i2c_dev->dev);
+
+	ret = st_i2c_of_get_deglitch(np, i2c_dev);
+	if (ret)
+		return ret;
+
+	adap = &i2c_dev->adap;
+	i2c_set_adapdata(adap, i2c_dev);
+	snprintf(adap->name, sizeof(adap->name), "ST I2C(0x%x)", res->start);
+	adap->owner = THIS_MODULE;
+	adap->timeout = 2 * HZ;
+	adap->retries = 0;
+	adap->algo = &st_i2c_algo;
+	adap->dev.parent = &pdev->dev;
+	adap->dev.of_node = pdev->dev.of_node;
+
+	init_completion(&i2c_dev->complete);
+
+	ret = i2c_add_adapter(adap);
+	if (ret) {
+		dev_err(&pdev->dev, "Failed to add adapter\n");
+		return ret;
+	}
+
+	platform_set_drvdata(pdev, i2c_dev);
+
+	dev_info(i2c_dev->dev, "%s initialized\n", adap->name);
+
+	return 0;
+}
+
+static int st_i2c_remove(struct platform_device *pdev)
+{
+	struct st_i2c_dev *i2c_dev = platform_get_drvdata(pdev);
+
+	i2c_del_adapter(&i2c_dev->adap);
+
+	return 0;
+}
+
+static struct of_device_id st_i2c_match[] = {
+	{ .compatible = "st,comms-ssc-i2c", },
+	{ .compatible = "st,comms-ssc4-i2c", },
+	{},
+};
+MODULE_DEVICE_TABLE(of, st_i2c_match);
+
+static struct platform_driver st_i2c_driver = {
+	.driver = {
+		.name = "st-i2c",
+		.owner = THIS_MODULE,
+		.of_match_table = st_i2c_match,
+		.pm = ST_I2C_PM,
+	},
+	.probe = st_i2c_probe,
+	.remove = st_i2c_remove,
+};
+
+module_platform_driver(st_i2c_driver);
+
+MODULE_AUTHOR("Maxime Coquelin <maxime.coquelin@st.com>");
+MODULE_DESCRIPTION("STMicroelectronics I2C driver");
+MODULE_LICENSE("GPL v2");
diff --git a/drivers/i2c/busses/i2c-tegra.c b/drivers/i2c/busses/i2c-tegra.c
index c457cb447c66..e661edee4d0c 100644
--- a/drivers/i2c/busses/i2c-tegra.c
+++ b/drivers/i2c/busses/i2c-tegra.c
@@ -544,7 +544,7 @@ static int tegra_i2c_xfer_msg(struct tegra_i2c_dev *i2c_dev,
 	i2c_dev->msg_buf_remaining = msg->len;
 	i2c_dev->msg_err = I2C_ERR_NONE;
 	i2c_dev->msg_read = (msg->flags & I2C_M_RD);
-	INIT_COMPLETION(i2c_dev->msg_complete);
+	reinit_completion(&i2c_dev->msg_complete);
 
 	packet_header = (0 << PACKET_HEADER0_HEADER_SIZE_SHIFT) |
 			PACKET_HEADER0_PROTOCOL_I2C |
diff --git a/drivers/i2c/busses/i2c-wmt.c b/drivers/i2c/busses/i2c-wmt.c
index c65da3d913a0..2c8a3e4f9008 100644
--- a/drivers/i2c/busses/i2c-wmt.c
+++ b/drivers/i2c/busses/i2c-wmt.c
@@ -158,7 +158,7 @@ static int wmt_i2c_write(struct i2c_adapter *adap, struct i2c_msg *pmsg,
 		writew(val, i2c_dev->base + REG_CR);
 	}
 
-	INIT_COMPLETION(i2c_dev->complete);
+	reinit_completion(&i2c_dev->complete);
 
 	if (i2c_dev->mode == I2C_MODE_STANDARD)
 		tcr_val = TCR_STANDARD_MODE;
@@ -247,7 +247,7 @@ static int wmt_i2c_read(struct i2c_adapter *adap, struct i2c_msg *pmsg,
 		writew(val, i2c_dev->base + REG_CR);
 	}
 
-	INIT_COMPLETION(i2c_dev->complete);
+	reinit_completion(&i2c_dev->complete);
 
 	if (i2c_dev->mode == I2C_MODE_STANDARD)
 		tcr_val = TCR_STANDARD_MODE;
@@ -349,6 +349,7 @@ static int wmt_i2c_reset_hardware(struct wmt_i2c_dev *i2c_dev)
 	err = clk_set_rate(i2c_dev->clk, 20000000);
 	if (err) {
 		dev_err(i2c_dev->dev, "failed to set clock = 20Mhz\n");
+		clk_disable_unprepare(i2c_dev->clk);
 		return err;
 	}
 
diff --git a/drivers/i2c/busses/i2c-xiic.c b/drivers/i2c/busses/i2c-xiic.c
index 4c8b368d463b..fc2716afdfd9 100644
--- a/drivers/i2c/busses/i2c-xiic.c
+++ b/drivers/i2c/busses/i2c-xiic.c
@@ -40,6 +40,7 @@
 #include <linux/i2c-xiic.h>
 #include <linux/io.h>
 #include <linux/slab.h>
+#include <linux/of.h>
 
 #define DRIVER_NAME "xiic-i2c"
 
@@ -702,7 +703,7 @@ static int xiic_i2c_probe(struct platform_device *pdev)
 	if (irq < 0)
 		goto resource_missing;
 
-	pdata = (struct xiic_i2c_platform_data *)dev_get_platdata(&pdev->dev);
+	pdata = dev_get_platdata(&pdev->dev);
 
 	i2c = kzalloc(sizeof(*i2c), GFP_KERNEL);
 	if (!i2c)
diff --git a/drivers/i2c/i2c-core.c b/drivers/i2c/i2c-core.c
index bc9a294c07f7..d74c0b34248e 100644
--- a/drivers/i2c/i2c-core.c
+++ b/drivers/i2c/i2c-core.c
@@ -248,7 +248,7 @@ static int i2c_device_probe(struct device *dev)
 	driver = to_i2c_driver(dev->driver);
 	if (!driver->probe || !driver->id_table)
 		return -ENODEV;
-	client->driver = driver;
+
 	if (!device_can_wakeup(&client->dev))
 		device_init_wakeup(&client->dev,
 					client->flags & I2C_CLIENT_WAKE);
@@ -257,7 +257,6 @@ static int i2c_device_probe(struct device *dev)
 	acpi_dev_pm_attach(&client->dev, true);
 	status = driver->probe(client, i2c_match_id(driver->id_table, client));
 	if (status) {
-		client->driver = NULL;
 		i2c_set_clientdata(client, NULL);
 		acpi_dev_pm_detach(&client->dev, true);
 	}
@@ -281,10 +280,8 @@ static int i2c_device_remove(struct device *dev)
 		dev->driver = NULL;
 		status = 0;
 	}
-	if (status == 0) {
-		client->driver = NULL;
+	if (status == 0)
 		i2c_set_clientdata(client, NULL);
-	}
 	acpi_dev_pm_detach(&client->dev, true);
 	return status;
 }
@@ -1627,9 +1624,14 @@ static int i2c_cmd(struct device *dev, void *_arg)
 {
 	struct i2c_client	*client = i2c_verify_client(dev);
 	struct i2c_cmd_arg	*arg = _arg;
+	struct i2c_driver	*driver;
+
+	if (!client || !client->dev.driver)
+		return 0;
 
-	if (client && client->driver && client->driver->command)
-		client->driver->command(client, arg->cmd, arg->arg);
+	driver = to_i2c_driver(client->dev.driver);
+	if (driver->command)
+		driver->command(client, arg->cmd, arg->arg);
 	return 0;
 }
 
diff --git a/drivers/i2c/i2c-dev.c b/drivers/i2c/i2c-dev.c
index c3ccdea3d180..80b47e8ce030 100644
--- a/drivers/i2c/i2c-dev.c
+++ b/drivers/i2c/i2c-dev.c
@@ -102,8 +102,8 @@ static void return_i2c_dev(struct i2c_dev *i2c_dev)
 	kfree(i2c_dev);
 }
 
-static ssize_t show_adapter_name(struct device *dev,
-				 struct device_attribute *attr, char *buf)
+static ssize_t name_show(struct device *dev,
+			 struct device_attribute *attr, char *buf)
 {
 	struct i2c_dev *i2c_dev = i2c_dev_get_by_minor(MINOR(dev->devt));
 
@@ -111,7 +111,13 @@ static ssize_t show_adapter_name(struct device *dev,
 		return -ENODEV;
 	return sprintf(buf, "%s\n", i2c_dev->adap->name);
 }
-static DEVICE_ATTR(name, S_IRUGO, show_adapter_name, NULL);
+static DEVICE_ATTR_RO(name);
+
+static struct attribute *i2c_attrs[] = {
+	&dev_attr_name.attr,
+	NULL,
+};
+ATTRIBUTE_GROUPS(i2c);
 
 /* ------------------------------------------------------------------------- */
 
@@ -562,15 +568,10 @@ static int i2cdev_attach_adapter(struct device *dev, void *dummy)
 		res = PTR_ERR(i2c_dev->dev);
 		goto error;
 	}
-	res = device_create_file(i2c_dev->dev, &dev_attr_name);
-	if (res)
-		goto error_destroy;
 
 	pr_debug("i2c-dev: adapter [%s] registered as minor %d\n",
 		 adap->name, adap->nr);
 	return 0;
-error_destroy:
-	device_destroy(i2c_dev_class, MKDEV(I2C_MAJOR, adap->nr));
 error:
 	return_i2c_dev(i2c_dev);
 	return res;
@@ -589,7 +590,6 @@ static int i2cdev_detach_adapter(struct device *dev, void *dummy)
 	if (!i2c_dev) /* attach_adapter must have failed */
 		return 0;
 
-	device_remove_file(i2c_dev->dev, &dev_attr_name);
 	return_i2c_dev(i2c_dev);
 	device_destroy(i2c_dev_class, MKDEV(I2C_MAJOR, adap->nr));
 
@@ -637,6 +637,7 @@ static int __init i2c_dev_init(void)
 		res = PTR_ERR(i2c_dev_class);
 		goto out_unreg_chrdev;
 	}
+	i2c_dev_class->dev_groups = i2c_groups;
 
 	/* Keep track of adapters which will be added or removed later */
 	res = bus_register_notifier(&i2c_bus_type, &i2cdev_notifier);
diff --git a/drivers/i2c/i2c-smbus.c b/drivers/i2c/i2c-smbus.c
index 44d4c6071c15..c99b22987366 100644
--- a/drivers/i2c/i2c-smbus.c
+++ b/drivers/i2c/i2c-smbus.c
@@ -46,6 +46,7 @@ static int smbus_do_alert(struct device *dev, void *addrp)
 {
 	struct i2c_client *client = i2c_verify_client(dev);
 	struct alert_data *data = addrp;
+	struct i2c_driver *driver;
 
 	if (!client || client->addr != data->addr)
 		return 0;
@@ -54,12 +55,13 @@ static int smbus_do_alert(struct device *dev, void *addrp)
 
 	/*
 	 * Drivers should either disable alerts, or provide at least
-	 * a minimal handler.  Lock so client->driver won't change.
+	 * a minimal handler.  Lock so the driver won't change.
 	 */
 	device_lock(dev);
-	if (client->driver) {
-		if (client->driver->alert)
-			client->driver->alert(client, data->flag);
+	if (client->dev.driver) {
+		driver = to_i2c_driver(client->dev.driver);
+		if (driver->alert)
+			driver->alert(client, data->flag);
 		else
 			dev_warn(&client->dev, "no driver alert()!\n");
 	} else
diff --git a/drivers/i2c/muxes/i2c-arb-gpio-challenge.c b/drivers/i2c/muxes/i2c-arb-gpio-challenge.c
index 928656e241dd..c58e093b6032 100644
--- a/drivers/i2c/muxes/i2c-arb-gpio-challenge.c
+++ b/drivers/i2c/muxes/i2c-arb-gpio-challenge.c
@@ -238,7 +238,7 @@ static struct platform_driver i2c_arbitrator_driver = {
 	.driver	= {
 		.owner	= THIS_MODULE,
 		.name	= "i2c-arb-gpio-challenge",
-		.of_match_table = of_match_ptr(i2c_arbitrator_of_match),
+		.of_match_table = i2c_arbitrator_of_match,
 	},
 };
 
diff --git a/drivers/i2c/muxes/i2c-mux-gpio.c b/drivers/i2c/muxes/i2c-mux-gpio.c
index a764da777f08..8a8c56f4b026 100644
--- a/drivers/i2c/muxes/i2c-mux-gpio.c
+++ b/drivers/i2c/muxes/i2c-mux-gpio.c
@@ -30,15 +30,15 @@ static void i2c_mux_gpio_set(const struct gpiomux *mux, unsigned val)
 	int i;
 
 	for (i = 0; i < mux->data.n_gpios; i++)
-		gpio_set_value(mux->gpio_base + mux->data.gpios[i],
-			       val & (1 << i));
+		gpio_set_value_cansleep(mux->gpio_base + mux->data.gpios[i],
+					val & (1 << i));
 }
 
 static int i2c_mux_gpio_select(struct i2c_adapter *adap, void *data, u32 chan)
 {
 	struct gpiomux *mux = data;
 
-	i2c_mux_gpio_set(mux, mux->data.values[chan]);
+	i2c_mux_gpio_set(mux, chan);
 
 	return 0;
 }
@@ -228,7 +228,7 @@ static int i2c_mux_gpio_probe(struct platform_device *pdev)
 		unsigned int class = mux->data.classes ? mux->data.classes[i] : 0;
 
 		mux->adap[i] = i2c_add_mux_adapter(parent, &pdev->dev, mux, nr,
-						   i, class,
+						   mux->data.values[i], class,
 						   i2c_mux_gpio_select, deselect);
 		if (!mux->adap[i]) {
 			ret = -ENODEV;
@@ -283,7 +283,7 @@ static struct platform_driver i2c_mux_gpio_driver = {
 	.driver	= {
 		.owner	= THIS_MODULE,
 		.name	= "i2c-mux-gpio",
-		.of_match_table = of_match_ptr(i2c_mux_gpio_of_match),
+		.of_match_table = i2c_mux_gpio_of_match,
 	},
 };
 
diff --git a/drivers/i2c/muxes/i2c-mux-pinctrl.c b/drivers/i2c/muxes/i2c-mux-pinctrl.c
index 68a37157377d..d7978dc4ad0b 100644
--- a/drivers/i2c/muxes/i2c-mux-pinctrl.c
+++ b/drivers/i2c/muxes/i2c-mux-pinctrl.c
@@ -24,6 +24,7 @@
 #include <linux/i2c-mux-pinctrl.h>
 #include <linux/platform_device.h>
 #include <linux/slab.h>
+#include <linux/of.h>
 
 struct i2c_mux_pinctrl {
 	struct device *dev;
diff --git a/drivers/ide/cs5536.c b/drivers/ide/cs5536.c
index 24214ab60ac0..de9185db41d4 100644
--- a/drivers/ide/cs5536.c
+++ b/drivers/ide/cs5536.c
@@ -295,15 +295,7 @@ static struct pci_driver cs5536_pci_driver = {
 	.resume		= ide_pci_resume,
 };
 
-static int __init cs5536_init(void)
-{
-	return pci_register_driver(&cs5536_pci_driver);
-}
-
-static void __exit cs5536_exit(void)
-{
-	pci_unregister_driver(&cs5536_pci_driver);
-}
+module_pci_driver(cs5536_pci_driver);
 
 MODULE_AUTHOR("Martin K. Petersen, Bartlomiej Zolnierkiewicz");
 MODULE_DESCRIPTION("low-level driver for the CS5536 IDE controller");
@@ -312,6 +304,3 @@ MODULE_DEVICE_TABLE(pci, cs5536_pci_tbl);
 
 module_param_named(msr, use_msr, int, 0644);
 MODULE_PARM_DESC(msr, "Force using MSR to configure IDE function (Default: 0)");
-
-module_init(cs5536_init);
-module_exit(cs5536_exit);
diff --git a/drivers/ide/pmac.c b/drivers/ide/pmac.c
index bf83d7bb6bc6..2db803cd095c 100644
--- a/drivers/ide/pmac.c
+++ b/drivers/ide/pmac.c
@@ -416,8 +416,7 @@ static int pmac_ide_init_dma(ide_hwif_t *, const struct ide_port_info *);
 static void pmac_ide_apply_timings(ide_drive_t *drive)
 {
 	ide_hwif_t *hwif = drive->hwif;
-	pmac_ide_hwif_t *pmif =
-		(pmac_ide_hwif_t *)dev_get_drvdata(hwif->gendev.parent);
+	pmac_ide_hwif_t *pmif = dev_get_drvdata(hwif->gendev.parent);
 
 	if (drive->dn & 1)
 		writel(pmif->timings[1], PMAC_IDE_REG(IDE_TIMING_CONFIG));
@@ -434,8 +433,7 @@ static void pmac_ide_apply_timings(ide_drive_t *drive)
 static void pmac_ide_kauai_apply_timings(ide_drive_t *drive)
 {
 	ide_hwif_t *hwif = drive->hwif;
-	pmac_ide_hwif_t *pmif =
-		(pmac_ide_hwif_t *)dev_get_drvdata(hwif->gendev.parent);
+	pmac_ide_hwif_t *pmif = dev_get_drvdata(hwif->gendev.parent);
 
 	if (drive->dn & 1) {
 		writel(pmif->timings[1], PMAC_IDE_REG(IDE_KAUAI_PIO_CONFIG));
@@ -454,8 +452,7 @@ static void
 pmac_ide_do_update_timings(ide_drive_t *drive)
 {
 	ide_hwif_t *hwif = drive->hwif;
-	pmac_ide_hwif_t *pmif =
-		(pmac_ide_hwif_t *)dev_get_drvdata(hwif->gendev.parent);
+	pmac_ide_hwif_t *pmif = dev_get_drvdata(hwif->gendev.parent);
 
 	if (pmif->kind == controller_sh_ata6 ||
 	    pmif->kind == controller_un_ata6 ||
@@ -500,8 +497,7 @@ static void pmac_write_devctl(ide_hwif_t *hwif, u8 ctl)
  */
 static void pmac_ide_set_pio_mode(ide_hwif_t *hwif, ide_drive_t *drive)
 {
-	pmac_ide_hwif_t *pmif =
-		(pmac_ide_hwif_t *)dev_get_drvdata(hwif->gendev.parent);
+	pmac_ide_hwif_t *pmif = dev_get_drvdata(hwif->gendev.parent);
 	const u8 pio = drive->pio_mode - XFER_PIO_0;
 	struct ide_timing *tim = ide_timing_find_mode(XFER_PIO_0 + pio);
 	u32 *timings, t;
@@ -781,8 +777,7 @@ set_timings_mdma(ide_drive_t *drive, int intf_type, u32 *timings, u32 *timings2,
 
 static void pmac_ide_set_dma_mode(ide_hwif_t *hwif, ide_drive_t *drive)
 {
-	pmac_ide_hwif_t *pmif =
-		(pmac_ide_hwif_t *)dev_get_drvdata(hwif->gendev.parent);
+	pmac_ide_hwif_t *pmif = dev_get_drvdata(hwif->gendev.parent);
 	int ret = 0;
 	u32 *timings, *timings2, tl[2];
 	u8 unit = drive->dn & 1;
@@ -919,8 +914,7 @@ static int pmac_ide_do_resume(pmac_ide_hwif_t *pmif)
 
 static u8 pmac_ide_cable_detect(ide_hwif_t *hwif)
 {
-	pmac_ide_hwif_t *pmif =
-		(pmac_ide_hwif_t *)dev_get_drvdata(hwif->gendev.parent);
+	pmac_ide_hwif_t *pmif = dev_get_drvdata(hwif->gendev.parent);
 	struct device_node *np = pmif->node;
 	const char *cable = of_get_property(np, "cable-type", NULL);
 	struct device_node *root = of_find_node_by_path("/");
@@ -951,8 +945,7 @@ static u8 pmac_ide_cable_detect(ide_hwif_t *hwif)
 static void pmac_ide_init_dev(ide_drive_t *drive)
 {
 	ide_hwif_t *hwif = drive->hwif;
-	pmac_ide_hwif_t *pmif =
-		(pmac_ide_hwif_t *)dev_get_drvdata(hwif->gendev.parent);
+	pmac_ide_hwif_t *pmif = dev_get_drvdata(hwif->gendev.parent);
 
 	if (on_media_bay(pmif)) {
 		if (check_media_bay(pmif->mdev->media_bay) == MB_CD) {
@@ -1228,8 +1221,7 @@ out_free_pmif:
 static int
 pmac_ide_macio_suspend(struct macio_dev *mdev, pm_message_t mesg)
 {
-	pmac_ide_hwif_t *pmif =
-		(pmac_ide_hwif_t *)dev_get_drvdata(&mdev->ofdev.dev);
+	pmac_ide_hwif_t *pmif = dev_get_drvdata(&mdev->ofdev.dev);
 	int rc = 0;
 
 	if (mesg.event != mdev->ofdev.dev.power.power_state.event
@@ -1245,8 +1237,7 @@ pmac_ide_macio_suspend(struct macio_dev *mdev, pm_message_t mesg)
 static int
 pmac_ide_macio_resume(struct macio_dev *mdev)
 {
-	pmac_ide_hwif_t *pmif =
-		(pmac_ide_hwif_t *)dev_get_drvdata(&mdev->ofdev.dev);
+	pmac_ide_hwif_t *pmif = dev_get_drvdata(&mdev->ofdev.dev);
 	int rc = 0;
 
 	if (mdev->ofdev.dev.power.power_state.event != PM_EVENT_ON) {
@@ -1318,7 +1309,6 @@ static int pmac_ide_pci_attach(struct pci_dev *pdev,
 	rc = pmac_ide_setup_device(pmif, &hw);
 	if (rc != 0) {
 		/* The inteface is released to the common IDE layer */
-		pci_set_drvdata(pdev, NULL);
 		iounmap(base);
 		pci_release_regions(pdev);
 		kfree(pmif);
@@ -1365,8 +1355,7 @@ pmac_ide_pci_resume(struct pci_dev *pdev)
 #ifdef CONFIG_PMAC_MEDIABAY
 static void pmac_ide_macio_mb_event(struct macio_dev* mdev, int mb_state)
 {
-	pmac_ide_hwif_t *pmif =
-		(pmac_ide_hwif_t *)dev_get_drvdata(&mdev->ofdev.dev);
+	pmac_ide_hwif_t *pmif = dev_get_drvdata(&mdev->ofdev.dev);
 
 	switch(mb_state) {
 	case MB_CD:
@@ -1468,8 +1457,7 @@ out:
 static int pmac_ide_build_dmatable(ide_drive_t *drive, struct ide_cmd *cmd)
 {
 	ide_hwif_t *hwif = drive->hwif;
-	pmac_ide_hwif_t *pmif =
-		(pmac_ide_hwif_t *)dev_get_drvdata(hwif->gendev.parent);
+	pmac_ide_hwif_t *pmif = dev_get_drvdata(hwif->gendev.parent);
 	struct dbdma_cmd *table;
 	volatile struct dbdma_regs __iomem *dma = pmif->dma_regs;
 	struct scatterlist *sg;
@@ -1546,8 +1534,7 @@ static int pmac_ide_build_dmatable(ide_drive_t *drive, struct ide_cmd *cmd)
 static int pmac_ide_dma_setup(ide_drive_t *drive, struct ide_cmd *cmd)
 {
 	ide_hwif_t *hwif = drive->hwif;
-	pmac_ide_hwif_t *pmif =
-		(pmac_ide_hwif_t *)dev_get_drvdata(hwif->gendev.parent);
+	pmac_ide_hwif_t *pmif = dev_get_drvdata(hwif->gendev.parent);
 	u8 unit = drive->dn & 1, ata4 = (pmif->kind == controller_kl_ata4);
 	u8 write = !!(cmd->tf_flags & IDE_TFLAG_WRITE);
 
@@ -1572,8 +1559,7 @@ static void
 pmac_ide_dma_start(ide_drive_t *drive)
 {
 	ide_hwif_t *hwif = drive->hwif;
-	pmac_ide_hwif_t *pmif =
-		(pmac_ide_hwif_t *)dev_get_drvdata(hwif->gendev.parent);
+	pmac_ide_hwif_t *pmif = dev_get_drvdata(hwif->gendev.parent);
 	volatile struct dbdma_regs __iomem *dma;
 
 	dma = pmif->dma_regs;
@@ -1590,8 +1576,7 @@ static int
 pmac_ide_dma_end (ide_drive_t *drive)
 {
 	ide_hwif_t *hwif = drive->hwif;
-	pmac_ide_hwif_t *pmif =
-		(pmac_ide_hwif_t *)dev_get_drvdata(hwif->gendev.parent);
+	pmac_ide_hwif_t *pmif = dev_get_drvdata(hwif->gendev.parent);
 	volatile struct dbdma_regs __iomem *dma = pmif->dma_regs;
 	u32 dstat;
 
@@ -1615,8 +1600,7 @@ static int
 pmac_ide_dma_test_irq (ide_drive_t *drive)
 {
 	ide_hwif_t *hwif = drive->hwif;
-	pmac_ide_hwif_t *pmif =
-		(pmac_ide_hwif_t *)dev_get_drvdata(hwif->gendev.parent);
+	pmac_ide_hwif_t *pmif = dev_get_drvdata(hwif->gendev.parent);
 	volatile struct dbdma_regs __iomem *dma = pmif->dma_regs;
 	unsigned long status, timeout;
 
@@ -1670,8 +1654,7 @@ static void
 pmac_ide_dma_lost_irq (ide_drive_t *drive)
 {
 	ide_hwif_t *hwif = drive->hwif;
-	pmac_ide_hwif_t *pmif =
-		(pmac_ide_hwif_t *)dev_get_drvdata(hwif->gendev.parent);
+	pmac_ide_hwif_t *pmif = dev_get_drvdata(hwif->gendev.parent);
 	volatile struct dbdma_regs __iomem *dma = pmif->dma_regs;
 	unsigned long status = readl(&dma->status);
 
@@ -1693,8 +1676,7 @@ static const struct ide_dma_ops pmac_dma_ops = {
  */
 static int pmac_ide_init_dma(ide_hwif_t *hwif, const struct ide_port_info *d)
 {
-	pmac_ide_hwif_t *pmif =
-		(pmac_ide_hwif_t *)dev_get_drvdata(hwif->gendev.parent);
+	pmac_ide_hwif_t *pmif = dev_get_drvdata(hwif->gendev.parent);
 	struct pci_dev *dev = to_pci_dev(hwif->dev);
 
 	/* We won't need pci_dev if we switch to generic consistent
diff --git a/drivers/iio/adc/ad_sigma_delta.c b/drivers/iio/adc/ad_sigma_delta.c
index e6fbd3e70981..9a4e0e32a771 100644
--- a/drivers/iio/adc/ad_sigma_delta.c
+++ b/drivers/iio/adc/ad_sigma_delta.c
@@ -188,7 +188,7 @@ static int ad_sd_calibrate(struct ad_sigma_delta *sigma_delta,
 
 	spi_bus_lock(sigma_delta->spi->master);
 	sigma_delta->bus_locked = true;
-	INIT_COMPLETION(sigma_delta->completion);
+	reinit_completion(&sigma_delta->completion);
 
 	ret = ad_sigma_delta_set_mode(sigma_delta, mode);
 	if (ret < 0)
@@ -259,7 +259,7 @@ int ad_sigma_delta_single_conversion(struct iio_dev *indio_dev,
 
 	spi_bus_lock(sigma_delta->spi->master);
 	sigma_delta->bus_locked = true;
-	INIT_COMPLETION(sigma_delta->completion);
+	reinit_completion(&sigma_delta->completion);
 
 	ad_sigma_delta_set_mode(sigma_delta, AD_SD_MODE_SINGLE);
 
@@ -343,7 +343,7 @@ static int ad_sd_buffer_postdisable(struct iio_dev *indio_dev)
 {
 	struct ad_sigma_delta *sigma_delta = iio_device_get_drvdata(indio_dev);
 
-	INIT_COMPLETION(sigma_delta->completion);
+	reinit_completion(&sigma_delta->completion);
 	wait_for_completion_timeout(&sigma_delta->completion, HZ);
 
 	if (!sigma_delta->irq_dis) {
diff --git a/drivers/iio/adc/nau7802.c b/drivers/iio/adc/nau7802.c
index 54c5babe6746..e525aa6475c4 100644
--- a/drivers/iio/adc/nau7802.c
+++ b/drivers/iio/adc/nau7802.c
@@ -190,7 +190,7 @@ static int nau7802_read_irq(struct iio_dev *indio_dev,
 	struct nau7802_state *st = iio_priv(indio_dev);
 	int ret;
 
-	INIT_COMPLETION(st->value_ok);
+	reinit_completion(&st->value_ok);
 	enable_irq(st->client->irq);
 
 	nau7802_sync(st);
diff --git a/drivers/iio/industrialio-event.c b/drivers/iio/industrialio-event.c
index dac15b9f9df8..c10eab64bc05 100644
--- a/drivers/iio/industrialio-event.c
+++ b/drivers/iio/industrialio-event.c
@@ -56,7 +56,7 @@ int iio_push_event(struct iio_dev *indio_dev, u64 ev_code, s64 timestamp)
 		ev.id = ev_code;
 		ev.timestamp = timestamp;
 
-		copied = kfifo_put(&ev_int->det_events, &ev);
+		copied = kfifo_put(&ev_int->det_events, ev);
 		if (copied != 0)
 			wake_up_locked_poll(&ev_int->wait, POLLIN);
 	}
diff --git a/drivers/infiniband/Kconfig b/drivers/infiniband/Kconfig
index b84791f03a27..5ceda710f516 100644
--- a/drivers/infiniband/Kconfig
+++ b/drivers/infiniband/Kconfig
@@ -31,17 +31,6 @@ config INFINIBAND_USER_ACCESS
 	  libibverbs, libibcm and a hardware driver library from
 	  <http://www.openfabrics.org/git/>.
 
-config INFINIBAND_EXPERIMENTAL_UVERBS_FLOW_STEERING
-       bool "Experimental and unstable ABI for userspace access to flow steering verbs"
-       depends on INFINIBAND_USER_ACCESS
-       depends on STAGING
-	---help---
-	  The final ABI for userspace access to flow steering verbs
-	  has not been defined.  To use the current ABI, *WHICH WILL
-	  CHANGE IN THE FUTURE*, say Y here.
-
-	  If unsure, say N.
-
 config INFINIBAND_USER_MEM
 	bool
 	depends on INFINIBAND_USER_ACCESS != n
diff --git a/drivers/infiniband/core/cm.c b/drivers/infiniband/core/cm.c
index 784b97cb05b0..f2ef7ef0f36f 100644
--- a/drivers/infiniband/core/cm.c
+++ b/drivers/infiniband/core/cm.c
@@ -383,14 +383,11 @@ static int cm_alloc_id(struct cm_id_private *cm_id_priv)
 {
 	unsigned long flags;
 	int id;
-	static int next_id;
 
 	idr_preload(GFP_KERNEL);
 	spin_lock_irqsave(&cm.lock, flags);
 
-	id = idr_alloc(&cm.local_id_table, cm_id_priv, next_id, 0, GFP_NOWAIT);
-	if (id >= 0)
-		next_id = max(id + 1, 0);
+	id = idr_alloc_cyclic(&cm.local_id_table, cm_id_priv, 0, 0, GFP_NOWAIT);
 
 	spin_unlock_irqrestore(&cm.lock, flags);
 	idr_preload_end();
diff --git a/drivers/infiniband/core/cma.c b/drivers/infiniband/core/cma.c
index d2172e71f985..8e49db690f33 100644
--- a/drivers/infiniband/core/cma.c
+++ b/drivers/infiniband/core/cma.c
@@ -328,28 +328,6 @@ static int cma_set_qkey(struct rdma_id_private *id_priv, u32 qkey)
 	return ret;
 }
 
-static int find_gid_port(struct ib_device *device, union ib_gid *gid, u8 port_num)
-{
-	int i;
-	int err;
-	struct ib_port_attr props;
-	union ib_gid tmp;
-
-	err = ib_query_port(device, port_num, &props);
-	if (err)
-		return err;
-
-	for (i = 0; i < props.gid_tbl_len; ++i) {
-		err = ib_query_gid(device, port_num, i, &tmp);
-		if (err)
-			return err;
-		if (!memcmp(&tmp, gid, sizeof tmp))
-			return 0;
-	}
-
-	return -EADDRNOTAVAIL;
-}
-
 static void cma_translate_ib(struct sockaddr_ib *sib, struct rdma_dev_addr *dev_addr)
 {
 	dev_addr->dev_type = ARPHRD_INFINIBAND;
@@ -371,13 +349,14 @@ static int cma_translate_addr(struct sockaddr *addr, struct rdma_dev_addr *dev_a
 	return ret;
 }
 
-static int cma_acquire_dev(struct rdma_id_private *id_priv)
+static int cma_acquire_dev(struct rdma_id_private *id_priv,
+			   struct rdma_id_private *listen_id_priv)
 {
 	struct rdma_dev_addr *dev_addr = &id_priv->id.route.addr.dev_addr;
 	struct cma_device *cma_dev;
 	union ib_gid gid, iboe_gid;
 	int ret = -ENODEV;
-	u8 port;
+	u8 port, found_port;
 	enum rdma_link_layer dev_ll = dev_addr->dev_type == ARPHRD_INFINIBAND ?
 		IB_LINK_LAYER_INFINIBAND : IB_LINK_LAYER_ETHERNET;
 
@@ -389,17 +368,39 @@ static int cma_acquire_dev(struct rdma_id_private *id_priv)
 	iboe_addr_get_sgid(dev_addr, &iboe_gid);
 	memcpy(&gid, dev_addr->src_dev_addr +
 	       rdma_addr_gid_offset(dev_addr), sizeof gid);
+	if (listen_id_priv &&
+	    rdma_port_get_link_layer(listen_id_priv->id.device,
+				     listen_id_priv->id.port_num) == dev_ll) {
+		cma_dev = listen_id_priv->cma_dev;
+		port = listen_id_priv->id.port_num;
+		if (rdma_node_get_transport(cma_dev->device->node_type) == RDMA_TRANSPORT_IB &&
+		    rdma_port_get_link_layer(cma_dev->device, port) == IB_LINK_LAYER_ETHERNET)
+			ret = ib_find_cached_gid(cma_dev->device, &iboe_gid,
+						 &found_port, NULL);
+		else
+			ret = ib_find_cached_gid(cma_dev->device, &gid,
+						 &found_port, NULL);
+
+		if (!ret && (port  == found_port)) {
+			id_priv->id.port_num = found_port;
+			goto out;
+		}
+	}
 	list_for_each_entry(cma_dev, &dev_list, list) {
 		for (port = 1; port <= cma_dev->device->phys_port_cnt; ++port) {
+			if (listen_id_priv &&
+			    listen_id_priv->cma_dev == cma_dev &&
+			    listen_id_priv->id.port_num == port)
+				continue;
 			if (rdma_port_get_link_layer(cma_dev->device, port) == dev_ll) {
 				if (rdma_node_get_transport(cma_dev->device->node_type) == RDMA_TRANSPORT_IB &&
 				    rdma_port_get_link_layer(cma_dev->device, port) == IB_LINK_LAYER_ETHERNET)
-					ret = find_gid_port(cma_dev->device, &iboe_gid, port);
+					ret = ib_find_cached_gid(cma_dev->device, &iboe_gid, &found_port, NULL);
 				else
-					ret = find_gid_port(cma_dev->device, &gid, port);
+					ret = ib_find_cached_gid(cma_dev->device, &gid, &found_port, NULL);
 
-				if (!ret) {
-					id_priv->id.port_num = port;
+				if (!ret && (port == found_port)) {
+					id_priv->id.port_num = found_port;
 					goto out;
 				}
 			}
@@ -1292,7 +1293,7 @@ static int cma_req_handler(struct ib_cm_id *cm_id, struct ib_cm_event *ib_event)
 	}
 
 	mutex_lock_nested(&conn_id->handler_mutex, SINGLE_DEPTH_NESTING);
-	ret = cma_acquire_dev(conn_id);
+	ret = cma_acquire_dev(conn_id, listen_id);
 	if (ret)
 		goto err2;
 
@@ -1451,7 +1452,6 @@ static int iw_conn_req_handler(struct iw_cm_id *cm_id,
 {
 	struct rdma_cm_id *new_cm_id;
 	struct rdma_id_private *listen_id, *conn_id;
-	struct net_device *dev = NULL;
 	struct rdma_cm_event event;
 	int ret;
 	struct ib_device_attr attr;
@@ -1481,7 +1481,7 @@ static int iw_conn_req_handler(struct iw_cm_id *cm_id,
 		goto out;
 	}
 
-	ret = cma_acquire_dev(conn_id);
+	ret = cma_acquire_dev(conn_id, listen_id);
 	if (ret) {
 		mutex_unlock(&conn_id->handler_mutex);
 		rdma_destroy_id(new_cm_id);
@@ -1529,8 +1529,6 @@ static int iw_conn_req_handler(struct iw_cm_id *cm_id,
 	cma_deref_id(conn_id);
 
 out:
-	if (dev)
-		dev_put(dev);
 	mutex_unlock(&listen_id->handler_mutex);
 	return ret;
 }
@@ -2066,7 +2064,7 @@ static void addr_handler(int status, struct sockaddr *src_addr,
 		goto out;
 
 	if (!status && !id_priv->cma_dev)
-		status = cma_acquire_dev(id_priv);
+		status = cma_acquire_dev(id_priv, NULL);
 
 	if (status) {
 		if (!cma_comp_exch(id_priv, RDMA_CM_ADDR_RESOLVED,
@@ -2563,7 +2561,7 @@ int rdma_bind_addr(struct rdma_cm_id *id, struct sockaddr *addr)
 		if (ret)
 			goto err1;
 
-		ret = cma_acquire_dev(id_priv);
+		ret = cma_acquire_dev(id_priv, NULL);
 		if (ret)
 			goto err1;
 	}
diff --git a/drivers/infiniband/core/netlink.c b/drivers/infiniband/core/netlink.c
index da06abde9e0d..a1e9cba84944 100644
--- a/drivers/infiniband/core/netlink.c
+++ b/drivers/infiniband/core/netlink.c
@@ -148,7 +148,7 @@ static int ibnl_rcv_msg(struct sk_buff *skb, struct nlmsghdr *nlh)
 	list_for_each_entry(client, &client_list, list) {
 		if (client->index == index) {
 			if (op < 0 || op >= client->nops ||
-			    !client->cb_table[RDMA_NL_GET_OP(op)].dump)
+			    !client->cb_table[op].dump)
 				return -EINVAL;
 
 			{
diff --git a/drivers/infiniband/core/sysfs.c b/drivers/infiniband/core/sysfs.c
index cde1e7b5b85d..faad2caf22b1 100644
--- a/drivers/infiniband/core/sysfs.c
+++ b/drivers/infiniband/core/sysfs.c
@@ -612,6 +612,7 @@ static ssize_t show_node_type(struct device *device,
 	switch (dev->node_type) {
 	case RDMA_NODE_IB_CA:	  return sprintf(buf, "%d: CA\n", dev->node_type);
 	case RDMA_NODE_RNIC:	  return sprintf(buf, "%d: RNIC\n", dev->node_type);
+	case RDMA_NODE_USNIC:	  return sprintf(buf, "%d: usNIC\n", dev->node_type);
 	case RDMA_NODE_IB_SWITCH: return sprintf(buf, "%d: switch\n", dev->node_type);
 	case RDMA_NODE_IB_ROUTER: return sprintf(buf, "%d: router\n", dev->node_type);
 	default:		  return sprintf(buf, "%d: <unknown>\n", dev->node_type);
diff --git a/drivers/infiniband/core/ucma.c b/drivers/infiniband/core/ucma.c
index b0f189be543b..ab8b1c30b36b 100644
--- a/drivers/infiniband/core/ucma.c
+++ b/drivers/infiniband/core/ucma.c
@@ -57,7 +57,7 @@ MODULE_LICENSE("Dual BSD/GPL");
 static unsigned int max_backlog = 1024;
 
 static struct ctl_table_header *ucma_ctl_table_hdr;
-static ctl_table ucma_ctl_table[] = {
+static struct ctl_table ucma_ctl_table[] = {
 	{
 		.procname	= "max_backlog",
 		.data		= &max_backlog,
@@ -271,7 +271,7 @@ static int ucma_event_handler(struct rdma_cm_id *cm_id,
 			goto out;
 		}
 		ctx->backlog--;
-	} else if (!ctx->uid) {
+	} else if (!ctx->uid || ctx->cm_id != cm_id) {
 		/*
 		 * We ignore events for new connections until userspace has set
 		 * their context.  This can only happen if an error occurs on a
diff --git a/drivers/infiniband/core/uverbs.h b/drivers/infiniband/core/uverbs.h
index d8f9c6c272d7..bdc842e9faef 100644
--- a/drivers/infiniband/core/uverbs.h
+++ b/drivers/infiniband/core/uverbs.h
@@ -47,6 +47,14 @@
 #include <rdma/ib_umem.h>
 #include <rdma/ib_user_verbs.h>
 
+#define INIT_UDATA(udata, ibuf, obuf, ilen, olen)			\
+	do {								\
+		(udata)->inbuf  = (void __user *) (ibuf);		\
+		(udata)->outbuf = (void __user *) (obuf);		\
+		(udata)->inlen  = (ilen);				\
+		(udata)->outlen = (olen);				\
+	} while (0)
+
 /*
  * Our lifetime rules for these structs are the following:
  *
@@ -178,6 +186,22 @@ void ib_uverbs_event_handler(struct ib_event_handler *handler,
 			     struct ib_event *event);
 void ib_uverbs_dealloc_xrcd(struct ib_uverbs_device *dev, struct ib_xrcd *xrcd);
 
+struct ib_uverbs_flow_spec {
+	union {
+		union {
+			struct ib_uverbs_flow_spec_hdr hdr;
+			struct {
+				__u32 type;
+				__u16 size;
+				__u16 reserved;
+			};
+		};
+		struct ib_uverbs_flow_spec_eth     eth;
+		struct ib_uverbs_flow_spec_ipv4    ipv4;
+		struct ib_uverbs_flow_spec_tcp_udp tcp_udp;
+	};
+};
+
 #define IB_UVERBS_DECLARE_CMD(name)					\
 	ssize_t ib_uverbs_##name(struct ib_uverbs_file *file,		\
 				 const char __user *buf, int in_len,	\
@@ -217,9 +241,13 @@ IB_UVERBS_DECLARE_CMD(destroy_srq);
 IB_UVERBS_DECLARE_CMD(create_xsrq);
 IB_UVERBS_DECLARE_CMD(open_xrcd);
 IB_UVERBS_DECLARE_CMD(close_xrcd);
-#ifdef CONFIG_INFINIBAND_EXPERIMENTAL_UVERBS_FLOW_STEERING
-IB_UVERBS_DECLARE_CMD(create_flow);
-IB_UVERBS_DECLARE_CMD(destroy_flow);
-#endif /* CONFIG_INFINIBAND_EXPERIMENTAL_UVERBS_FLOW_STEERING */
+
+#define IB_UVERBS_DECLARE_EX_CMD(name)				\
+	int ib_uverbs_ex_##name(struct ib_uverbs_file *file,	\
+				struct ib_udata *ucore,		\
+				struct ib_udata *uhw)
+
+IB_UVERBS_DECLARE_EX_CMD(create_flow);
+IB_UVERBS_DECLARE_EX_CMD(destroy_flow);
 
 #endif /* UVERBS_H */
diff --git a/drivers/infiniband/core/uverbs_cmd.c b/drivers/infiniband/core/uverbs_cmd.c
index 2f0f01b70e3b..65f6e7dc380c 100644
--- a/drivers/infiniband/core/uverbs_cmd.c
+++ b/drivers/infiniband/core/uverbs_cmd.c
@@ -54,17 +54,7 @@ static struct uverbs_lock_class qp_lock_class	= { .name = "QP-uobj" };
 static struct uverbs_lock_class ah_lock_class	= { .name = "AH-uobj" };
 static struct uverbs_lock_class srq_lock_class	= { .name = "SRQ-uobj" };
 static struct uverbs_lock_class xrcd_lock_class = { .name = "XRCD-uobj" };
-#ifdef CONFIG_INFINIBAND_EXPERIMENTAL_UVERBS_FLOW_STEERING
 static struct uverbs_lock_class rule_lock_class = { .name = "RULE-uobj" };
-#endif /* CONFIG_INFINIBAND_EXPERIMENTAL_UVERBS_FLOW_STEERING */
-
-#define INIT_UDATA(udata, ibuf, obuf, ilen, olen)			\
-	do {								\
-		(udata)->inbuf  = (void __user *) (ibuf);		\
-		(udata)->outbuf = (void __user *) (obuf);		\
-		(udata)->inlen  = (ilen);				\
-		(udata)->outlen = (olen);				\
-	} while (0)
 
 /*
  * The ib_uobject locking scheme is as follows:
@@ -939,13 +929,9 @@ ssize_t ib_uverbs_reg_mr(struct ib_uverbs_file *file,
 	if ((cmd.start & ~PAGE_MASK) != (cmd.hca_va & ~PAGE_MASK))
 		return -EINVAL;
 
-	/*
-	 * Local write permission is required if remote write or
-	 * remote atomic permission is also requested.
-	 */
-	if (cmd.access_flags & (IB_ACCESS_REMOTE_ATOMIC | IB_ACCESS_REMOTE_WRITE) &&
-	    !(cmd.access_flags & IB_ACCESS_LOCAL_WRITE))
-		return -EINVAL;
+	ret = ib_check_mr_access(cmd.access_flags);
+	if (ret)
+		return ret;
 
 	uobj = kmalloc(sizeof *uobj, GFP_KERNEL);
 	if (!uobj)
@@ -2128,6 +2114,9 @@ ssize_t ib_uverbs_post_send(struct ib_uverbs_file *file,
 			}
 			next->wr.ud.remote_qpn  = user_wr->wr.ud.remote_qpn;
 			next->wr.ud.remote_qkey = user_wr->wr.ud.remote_qkey;
+			if (next->opcode == IB_WR_SEND_WITH_IMM)
+				next->ex.imm_data =
+					(__be32 __force) user_wr->ex.imm_data;
 		} else {
 			switch (next->opcode) {
 			case IB_WR_RDMA_WRITE_WITH_IMM:
@@ -2601,8 +2590,7 @@ out_put:
 	return ret ? ret : in_len;
 }
 
-#ifdef CONFIG_INFINIBAND_EXPERIMENTAL_UVERBS_FLOW_STEERING
-static int kern_spec_to_ib_spec(struct ib_kern_spec *kern_spec,
+static int kern_spec_to_ib_spec(struct ib_uverbs_flow_spec *kern_spec,
 				union ib_flow_spec *ib_spec)
 {
 	ib_spec->type = kern_spec->type;
@@ -2642,28 +2630,31 @@ static int kern_spec_to_ib_spec(struct ib_kern_spec *kern_spec,
 	return 0;
 }
 
-ssize_t ib_uverbs_create_flow(struct ib_uverbs_file *file,
-			      const char __user *buf, int in_len,
-			      int out_len)
+int ib_uverbs_ex_create_flow(struct ib_uverbs_file *file,
+			     struct ib_udata *ucore,
+			     struct ib_udata *uhw)
 {
 	struct ib_uverbs_create_flow	  cmd;
 	struct ib_uverbs_create_flow_resp resp;
 	struct ib_uobject		  *uobj;
 	struct ib_flow			  *flow_id;
-	struct ib_kern_flow_attr	  *kern_flow_attr;
+	struct ib_uverbs_flow_attr	  *kern_flow_attr;
 	struct ib_flow_attr		  *flow_attr;
 	struct ib_qp			  *qp;
 	int err = 0;
 	void *kern_spec;
 	void *ib_spec;
 	int i;
-	int kern_attr_size;
 
-	if (out_len < sizeof(resp))
+	if (ucore->outlen < sizeof(resp))
 		return -ENOSPC;
 
-	if (copy_from_user(&cmd, buf, sizeof(cmd)))
-		return -EFAULT;
+	err = ib_copy_from_udata(&cmd, ucore, sizeof(cmd));
+	if (err)
+		return err;
+
+	ucore->inbuf += sizeof(cmd);
+	ucore->inlen -= sizeof(cmd);
 
 	if (cmd.comp_mask)
 		return -EINVAL;
@@ -2672,32 +2663,27 @@ ssize_t ib_uverbs_create_flow(struct ib_uverbs_file *file,
 	     !capable(CAP_NET_ADMIN)) || !capable(CAP_NET_RAW))
 		return -EPERM;
 
-	if (cmd.flow_attr.num_of_specs < 0 ||
-	    cmd.flow_attr.num_of_specs > IB_FLOW_SPEC_SUPPORT_LAYERS)
+	if (cmd.flow_attr.num_of_specs > IB_FLOW_SPEC_SUPPORT_LAYERS)
 		return -EINVAL;
 
-	kern_attr_size = cmd.flow_attr.size - sizeof(cmd) -
-			 sizeof(struct ib_uverbs_cmd_hdr_ex);
-
-	if (cmd.flow_attr.size < 0 || cmd.flow_attr.size > in_len ||
-	    kern_attr_size < 0 || kern_attr_size >
-	    (cmd.flow_attr.num_of_specs * sizeof(struct ib_kern_spec)))
+	if (cmd.flow_attr.size > ucore->inlen ||
+	    cmd.flow_attr.size >
+	    (cmd.flow_attr.num_of_specs * sizeof(struct ib_uverbs_flow_spec)))
 		return -EINVAL;
 
 	if (cmd.flow_attr.num_of_specs) {
-		kern_flow_attr = kmalloc(cmd.flow_attr.size, GFP_KERNEL);
+		kern_flow_attr = kmalloc(sizeof(*kern_flow_attr) + cmd.flow_attr.size,
+					 GFP_KERNEL);
 		if (!kern_flow_attr)
 			return -ENOMEM;
 
 		memcpy(kern_flow_attr, &cmd.flow_attr, sizeof(*kern_flow_attr));
-		if (copy_from_user(kern_flow_attr + 1, buf + sizeof(cmd),
-				   kern_attr_size)) {
-			err = -EFAULT;
+		err = ib_copy_from_udata(kern_flow_attr + 1, ucore,
+					 cmd.flow_attr.size);
+		if (err)
 			goto err_free_attr;
-		}
 	} else {
 		kern_flow_attr = &cmd.flow_attr;
-		kern_attr_size = sizeof(cmd.flow_attr);
 	}
 
 	uobj = kmalloc(sizeof(*uobj), GFP_KERNEL);
@@ -2714,7 +2700,7 @@ ssize_t ib_uverbs_create_flow(struct ib_uverbs_file *file,
 		goto err_uobj;
 	}
 
-	flow_attr = kmalloc(cmd.flow_attr.size, GFP_KERNEL);
+	flow_attr = kmalloc(sizeof(*flow_attr) + cmd.flow_attr.size, GFP_KERNEL);
 	if (!flow_attr) {
 		err = -ENOMEM;
 		goto err_put;
@@ -2729,19 +2715,22 @@ ssize_t ib_uverbs_create_flow(struct ib_uverbs_file *file,
 
 	kern_spec = kern_flow_attr + 1;
 	ib_spec = flow_attr + 1;
-	for (i = 0; i < flow_attr->num_of_specs && kern_attr_size > 0; i++) {
+	for (i = 0; i < flow_attr->num_of_specs &&
+	     cmd.flow_attr.size > offsetof(struct ib_uverbs_flow_spec, reserved) &&
+	     cmd.flow_attr.size >=
+	     ((struct ib_uverbs_flow_spec *)kern_spec)->size; i++) {
 		err = kern_spec_to_ib_spec(kern_spec, ib_spec);
 		if (err)
 			goto err_free;
 		flow_attr->size +=
 			((union ib_flow_spec *) ib_spec)->size;
-		kern_attr_size -= ((struct ib_kern_spec *) kern_spec)->size;
-		kern_spec += ((struct ib_kern_spec *) kern_spec)->size;
+		cmd.flow_attr.size -= ((struct ib_uverbs_flow_spec *)kern_spec)->size;
+		kern_spec += ((struct ib_uverbs_flow_spec *) kern_spec)->size;
 		ib_spec += ((union ib_flow_spec *) ib_spec)->size;
 	}
-	if (kern_attr_size) {
-		pr_warn("create flow failed, %d bytes left from uverb cmd\n",
-			kern_attr_size);
+	if (cmd.flow_attr.size || (i != flow_attr->num_of_specs)) {
+		pr_warn("create flow failed, flow %d: %d bytes left from uverb cmd\n",
+			i, cmd.flow_attr.size);
 		goto err_free;
 	}
 	flow_id = ib_create_flow(qp, flow_attr, IB_FLOW_DOMAIN_USER);
@@ -2760,11 +2749,10 @@ ssize_t ib_uverbs_create_flow(struct ib_uverbs_file *file,
 	memset(&resp, 0, sizeof(resp));
 	resp.flow_handle = uobj->id;
 
-	if (copy_to_user((void __user *)(unsigned long) cmd.response,
-			 &resp, sizeof(resp))) {
-		err = -EFAULT;
+	err = ib_copy_to_udata(ucore,
+			       &resp, sizeof(resp));
+	if (err)
 		goto err_copy;
-	}
 
 	put_qp_read(qp);
 	mutex_lock(&file->mutex);
@@ -2777,7 +2765,7 @@ ssize_t ib_uverbs_create_flow(struct ib_uverbs_file *file,
 	kfree(flow_attr);
 	if (cmd.flow_attr.num_of_specs)
 		kfree(kern_flow_attr);
-	return in_len;
+	return 0;
 err_copy:
 	idr_remove_uobj(&ib_uverbs_rule_idr, uobj);
 destroy_flow:
@@ -2794,16 +2782,18 @@ err_free_attr:
 	return err;
 }
 
-ssize_t ib_uverbs_destroy_flow(struct ib_uverbs_file *file,
-			       const char __user *buf, int in_len,
-			       int out_len) {
+int ib_uverbs_ex_destroy_flow(struct ib_uverbs_file *file,
+			      struct ib_udata *ucore,
+			      struct ib_udata *uhw)
+{
 	struct ib_uverbs_destroy_flow	cmd;
 	struct ib_flow			*flow_id;
 	struct ib_uobject		*uobj;
 	int				ret;
 
-	if (copy_from_user(&cmd, buf, sizeof(cmd)))
-		return -EFAULT;
+	ret = ib_copy_from_udata(&cmd, ucore, sizeof(cmd));
+	if (ret)
+		return ret;
 
 	uobj = idr_write_uobj(&ib_uverbs_rule_idr, cmd.flow_handle,
 			      file->ucontext);
@@ -2825,9 +2815,8 @@ ssize_t ib_uverbs_destroy_flow(struct ib_uverbs_file *file,
 
 	put_uobj(uobj);
 
-	return ret ? ret : in_len;
+	return ret;
 }
-#endif /* CONFIG_INFINIBAND_EXPERIMENTAL_UVERBS_FLOW_STEERING */
 
 static int __uverbs_create_xsrq(struct ib_uverbs_file *file,
 				struct ib_uverbs_create_xsrq *cmd,
diff --git a/drivers/infiniband/core/uverbs_main.c b/drivers/infiniband/core/uverbs_main.c
index 2df31f68ea09..34386943ebcf 100644
--- a/drivers/infiniband/core/uverbs_main.c
+++ b/drivers/infiniband/core/uverbs_main.c
@@ -115,10 +115,13 @@ static ssize_t (*uverbs_cmd_table[])(struct ib_uverbs_file *file,
 	[IB_USER_VERBS_CMD_CLOSE_XRCD]		= ib_uverbs_close_xrcd,
 	[IB_USER_VERBS_CMD_CREATE_XSRQ]		= ib_uverbs_create_xsrq,
 	[IB_USER_VERBS_CMD_OPEN_QP]		= ib_uverbs_open_qp,
-#ifdef CONFIG_INFINIBAND_EXPERIMENTAL_UVERBS_FLOW_STEERING
-	[IB_USER_VERBS_CMD_CREATE_FLOW]		= ib_uverbs_create_flow,
-	[IB_USER_VERBS_CMD_DESTROY_FLOW]	= ib_uverbs_destroy_flow
-#endif /* CONFIG_INFINIBAND_EXPERIMENTAL_UVERBS_FLOW_STEERING */
+};
+
+static int (*uverbs_ex_cmd_table[])(struct ib_uverbs_file *file,
+				    struct ib_udata *ucore,
+				    struct ib_udata *uhw) = {
+	[IB_USER_VERBS_EX_CMD_CREATE_FLOW]	= ib_uverbs_ex_create_flow,
+	[IB_USER_VERBS_EX_CMD_DESTROY_FLOW]	= ib_uverbs_ex_destroy_flow
 };
 
 static void ib_uverbs_add_one(struct ib_device *device);
@@ -589,6 +592,7 @@ static ssize_t ib_uverbs_write(struct file *filp, const char __user *buf,
 {
 	struct ib_uverbs_file *file = filp->private_data;
 	struct ib_uverbs_cmd_hdr hdr;
+	__u32 flags;
 
 	if (count < sizeof hdr)
 		return -EINVAL;
@@ -596,45 +600,105 @@ static ssize_t ib_uverbs_write(struct file *filp, const char __user *buf,
 	if (copy_from_user(&hdr, buf, sizeof hdr))
 		return -EFAULT;
 
-	if (hdr.command >= ARRAY_SIZE(uverbs_cmd_table) ||
-	    !uverbs_cmd_table[hdr.command])
-		return -EINVAL;
+	flags = (hdr.command &
+		 IB_USER_VERBS_CMD_FLAGS_MASK) >> IB_USER_VERBS_CMD_FLAGS_SHIFT;
 
-	if (!file->ucontext &&
-	    hdr.command != IB_USER_VERBS_CMD_GET_CONTEXT)
-		return -EINVAL;
+	if (!flags) {
+		__u32 command;
 
-	if (!(file->device->ib_dev->uverbs_cmd_mask & (1ull << hdr.command)))
-		return -ENOSYS;
+		if (hdr.command & ~(__u32)(IB_USER_VERBS_CMD_FLAGS_MASK |
+					   IB_USER_VERBS_CMD_COMMAND_MASK))
+			return -EINVAL;
 
-#ifdef CONFIG_INFINIBAND_EXPERIMENTAL_UVERBS_FLOW_STEERING
-	if (hdr.command >= IB_USER_VERBS_CMD_THRESHOLD) {
-		struct ib_uverbs_cmd_hdr_ex hdr_ex;
+		command = hdr.command & IB_USER_VERBS_CMD_COMMAND_MASK;
 
-		if (copy_from_user(&hdr_ex, buf, sizeof(hdr_ex)))
-			return -EFAULT;
+		if (command >= ARRAY_SIZE(uverbs_cmd_table) ||
+		    !uverbs_cmd_table[command])
+			return -EINVAL;
 
-		if (((hdr_ex.in_words + hdr_ex.provider_in_words) * 4) != count)
+		if (!file->ucontext &&
+		    command != IB_USER_VERBS_CMD_GET_CONTEXT)
 			return -EINVAL;
 
-		return uverbs_cmd_table[hdr.command](file,
-						     buf + sizeof(hdr_ex),
-						     (hdr_ex.in_words +
-						      hdr_ex.provider_in_words) * 4,
-						     (hdr_ex.out_words +
-						      hdr_ex.provider_out_words) * 4);
-	} else {
-#endif /* CONFIG_INFINIBAND_EXPERIMENTAL_UVERBS_FLOW_STEERING */
+		if (!(file->device->ib_dev->uverbs_cmd_mask & (1ull << command)))
+			return -ENOSYS;
+
 		if (hdr.in_words * 4 != count)
 			return -EINVAL;
 
-		return uverbs_cmd_table[hdr.command](file,
-						     buf + sizeof(hdr),
-						     hdr.in_words * 4,
-						     hdr.out_words * 4);
-#ifdef CONFIG_INFINIBAND_EXPERIMENTAL_UVERBS_FLOW_STEERING
+		return uverbs_cmd_table[command](file,
+						 buf + sizeof(hdr),
+						 hdr.in_words * 4,
+						 hdr.out_words * 4);
+
+	} else if (flags == IB_USER_VERBS_CMD_FLAG_EXTENDED) {
+		__u32 command;
+
+		struct ib_uverbs_ex_cmd_hdr ex_hdr;
+		struct ib_udata ucore;
+		struct ib_udata uhw;
+		int err;
+		size_t written_count = count;
+
+		if (hdr.command & ~(__u32)(IB_USER_VERBS_CMD_FLAGS_MASK |
+					   IB_USER_VERBS_CMD_COMMAND_MASK))
+			return -EINVAL;
+
+		command = hdr.command & IB_USER_VERBS_CMD_COMMAND_MASK;
+
+		if (command >= ARRAY_SIZE(uverbs_ex_cmd_table) ||
+		    !uverbs_ex_cmd_table[command])
+			return -ENOSYS;
+
+		if (!file->ucontext)
+			return -EINVAL;
+
+		if (!(file->device->ib_dev->uverbs_ex_cmd_mask & (1ull << command)))
+			return -ENOSYS;
+
+		if (count < (sizeof(hdr) + sizeof(ex_hdr)))
+			return -EINVAL;
+
+		if (copy_from_user(&ex_hdr, buf + sizeof(hdr), sizeof(ex_hdr)))
+			return -EFAULT;
+
+		count -= sizeof(hdr) + sizeof(ex_hdr);
+		buf += sizeof(hdr) + sizeof(ex_hdr);
+
+		if ((hdr.in_words + ex_hdr.provider_in_words) * 8 != count)
+			return -EINVAL;
+
+		if (ex_hdr.response) {
+			if (!hdr.out_words && !ex_hdr.provider_out_words)
+				return -EINVAL;
+		} else {
+			if (hdr.out_words || ex_hdr.provider_out_words)
+				return -EINVAL;
+		}
+
+		INIT_UDATA(&ucore,
+			   (hdr.in_words) ? buf : 0,
+			   (unsigned long)ex_hdr.response,
+			   hdr.in_words * 8,
+			   hdr.out_words * 8);
+
+		INIT_UDATA(&uhw,
+			   (ex_hdr.provider_in_words) ? buf + ucore.inlen : 0,
+			   (ex_hdr.provider_out_words) ? (unsigned long)ex_hdr.response + ucore.outlen : 0,
+			   ex_hdr.provider_in_words * 8,
+			   ex_hdr.provider_out_words * 8);
+
+		err = uverbs_ex_cmd_table[command](file,
+						   &ucore,
+						   &uhw);
+
+		if (err)
+			return err;
+
+		return written_count;
 	}
-#endif /* CONFIG_INFINIBAND_EXPERIMENTAL_UVERBS_FLOW_STEERING */
+
+	return -ENOSYS;
 }
 
 static int ib_uverbs_mmap(struct file *filp, struct vm_area_struct *vma)
diff --git a/drivers/infiniband/core/verbs.c b/drivers/infiniband/core/verbs.c
index a321df28bab2..d4f6ddf72ffa 100644
--- a/drivers/infiniband/core/verbs.c
+++ b/drivers/infiniband/core/verbs.c
@@ -114,6 +114,8 @@ rdma_node_get_transport(enum rdma_node_type node_type)
 		return RDMA_TRANSPORT_IB;
 	case RDMA_NODE_RNIC:
 		return RDMA_TRANSPORT_IWARP;
+	case RDMA_NODE_USNIC:
+		return RDMA_TRANSPORT_USNIC;
 	default:
 		BUG();
 		return 0;
@@ -130,6 +132,7 @@ enum rdma_link_layer rdma_port_get_link_layer(struct ib_device *device, u8 port_
 	case RDMA_TRANSPORT_IB:
 		return IB_LINK_LAYER_INFINIBAND;
 	case RDMA_TRANSPORT_IWARP:
+	case RDMA_TRANSPORT_USNIC:
 		return IB_LINK_LAYER_ETHERNET;
 	default:
 		return IB_LINK_LAYER_UNSPECIFIED;
@@ -958,6 +961,11 @@ EXPORT_SYMBOL(ib_resize_cq);
 struct ib_mr *ib_get_dma_mr(struct ib_pd *pd, int mr_access_flags)
 {
 	struct ib_mr *mr;
+	int err;
+
+	err = ib_check_mr_access(mr_access_flags);
+	if (err)
+		return ERR_PTR(err);
 
 	mr = pd->device->get_dma_mr(pd, mr_access_flags);
 
@@ -980,6 +988,11 @@ struct ib_mr *ib_reg_phys_mr(struct ib_pd *pd,
 			     u64 *iova_start)
 {
 	struct ib_mr *mr;
+	int err;
+
+	err = ib_check_mr_access(mr_access_flags);
+	if (err)
+		return ERR_PTR(err);
 
 	if (!pd->device->reg_phys_mr)
 		return ERR_PTR(-ENOSYS);
@@ -1010,6 +1023,10 @@ int ib_rereg_phys_mr(struct ib_mr *mr,
 	struct ib_pd *old_pd;
 	int ret;
 
+	ret = ib_check_mr_access(mr_access_flags);
+	if (ret)
+		return ret;
+
 	if (!mr->device->rereg_phys_mr)
 		return -ENOSYS;
 
diff --git a/drivers/infiniband/hw/cxgb4/device.c b/drivers/infiniband/hw/cxgb4/device.c
index 33d2cc6ab562..4a033853312e 100644
--- a/drivers/infiniband/hw/cxgb4/device.c
+++ b/drivers/infiniband/hw/cxgb4/device.c
@@ -602,10 +602,10 @@ static int c4iw_rdev_open(struct c4iw_rdev *rdev)
 	     rdev->lldi.vr->qp.size,
 	     rdev->lldi.vr->cq.start,
 	     rdev->lldi.vr->cq.size);
-	PDBG("udb len 0x%x udb base %p db_reg %p gts_reg %p qpshift %lu "
+	PDBG("udb len 0x%x udb base %llx db_reg %p gts_reg %p qpshift %lu "
 	     "qpmask 0x%x cqshift %lu cqmask 0x%x\n",
 	     (unsigned)pci_resource_len(rdev->lldi.pdev, 2),
-	     (void *)(unsigned long)pci_resource_start(rdev->lldi.pdev, 2),
+	     (u64)pci_resource_start(rdev->lldi.pdev, 2),
 	     rdev->lldi.db_reg,
 	     rdev->lldi.gts_reg,
 	     rdev->qpshift, rdev->qpmask,
diff --git a/drivers/infiniband/hw/ipath/ipath_user_sdma.c b/drivers/infiniband/hw/ipath/ipath_user_sdma.c
index f5cb13b21445..cc04b7ba3488 100644
--- a/drivers/infiniband/hw/ipath/ipath_user_sdma.c
+++ b/drivers/infiniband/hw/ipath/ipath_user_sdma.c
@@ -280,9 +280,7 @@ static int ipath_user_sdma_pin_pages(const struct ipath_devdata *dd,
 	int j;
 	int ret;
 
-	ret = get_user_pages(current, current->mm, addr,
-			     npages, 0, 1, pages, NULL);
-
+	ret = get_user_pages_fast(addr, npages, 0, pages);
 	if (ret != npages) {
 		int i;
 
@@ -811,10 +809,7 @@ int ipath_user_sdma_writev(struct ipath_devdata *dd,
 	while (dim) {
 		const int mxp = 8;
 
-		down_write(&current->mm->mmap_sem);
 		ret = ipath_user_sdma_queue_pkts(dd, pq, &list, iov, dim, mxp);
-		up_write(&current->mm->mmap_sem);
-
 		if (ret <= 0)
 			goto done_unlock;
 		else {
diff --git a/drivers/infiniband/hw/mlx4/cq.c b/drivers/infiniband/hw/mlx4/cq.c
index d5e60f44ba5a..66dbf8062374 100644
--- a/drivers/infiniband/hw/mlx4/cq.c
+++ b/drivers/infiniband/hw/mlx4/cq.c
@@ -324,7 +324,7 @@ static int mlx4_ib_get_outstanding_cqes(struct mlx4_ib_cq *cq)
 	u32 i;
 
 	i = cq->mcq.cons_index;
-	while (get_sw_cqe(cq, i & cq->ibcq.cqe))
+	while (get_sw_cqe(cq, i))
 		++i;
 
 	return i - cq->mcq.cons_index;
@@ -365,7 +365,7 @@ int mlx4_ib_resize_cq(struct ib_cq *ibcq, int entries, struct ib_udata *udata)
 
 	mutex_lock(&cq->resize_mutex);
 
-	if (entries < 1 || entries > dev->dev->caps.max_cqes) {
+	if (entries < 1) {
 		err = -EINVAL;
 		goto out;
 	}
@@ -376,6 +376,11 @@ int mlx4_ib_resize_cq(struct ib_cq *ibcq, int entries, struct ib_udata *udata)
 		goto out;
 	}
 
+	if (entries > dev->dev->caps.max_cqes) {
+		err = -EINVAL;
+		goto out;
+	}
+
 	if (ibcq->uobject) {
 		err = mlx4_alloc_resize_umem(dev, cq, entries, udata);
 		if (err)
diff --git a/drivers/infiniband/hw/mlx4/main.c b/drivers/infiniband/hw/mlx4/main.c
index 6a0a0d29660d..1958c5ca792a 100644
--- a/drivers/infiniband/hw/mlx4/main.c
+++ b/drivers/infiniband/hw/mlx4/main.c
@@ -1685,11 +1685,9 @@ static void *mlx4_ib_add(struct mlx4_dev *dev)
 		ibdev->ib_dev.create_flow	= mlx4_ib_create_flow;
 		ibdev->ib_dev.destroy_flow	= mlx4_ib_destroy_flow;
 
-#ifdef CONFIG_INFINIBAND_EXPERIMENTAL_UVERBS_FLOW_STEERING
-		ibdev->ib_dev.uverbs_cmd_mask	|=
-			(1ull << IB_USER_VERBS_CMD_CREATE_FLOW) |
-			(1ull << IB_USER_VERBS_CMD_DESTROY_FLOW);
-#endif /* CONFIG_INFINIBAND_EXPERIMENTAL_UVERBS_FLOW_STEERING */
+		ibdev->ib_dev.uverbs_ex_cmd_mask	|=
+			(1ull << IB_USER_VERBS_EX_CMD_CREATE_FLOW) |
+			(1ull << IB_USER_VERBS_EX_CMD_DESTROY_FLOW);
 	}
 
 	mlx4_ib_alloc_eqs(dev, ibdev);
diff --git a/drivers/infiniband/hw/mlx5/cq.c b/drivers/infiniband/hw/mlx5/cq.c
index 344ab03948a3..b72627429745 100644
--- a/drivers/infiniband/hw/mlx5/cq.c
+++ b/drivers/infiniband/hw/mlx5/cq.c
@@ -556,7 +556,7 @@ static int create_cq_user(struct mlx5_ib_dev *dev, struct ib_udata *udata,
 		goto err_db;
 	}
 	mlx5_ib_populate_pas(dev, cq->buf.umem, page_shift, (*cqb)->pas, 0);
-	(*cqb)->ctx.log_pg_sz = page_shift - PAGE_SHIFT;
+	(*cqb)->ctx.log_pg_sz = page_shift - MLX5_ADAPTER_PAGE_SHIFT;
 
 	*index = to_mucontext(context)->uuari.uars[0].index;
 
@@ -620,7 +620,7 @@ static int create_cq_kernel(struct mlx5_ib_dev *dev, struct mlx5_ib_cq *cq,
 	}
 	mlx5_fill_page_array(&cq->buf.buf, (*cqb)->pas);
 
-	(*cqb)->ctx.log_pg_sz = cq->buf.buf.page_shift - PAGE_SHIFT;
+	(*cqb)->ctx.log_pg_sz = cq->buf.buf.page_shift - MLX5_ADAPTER_PAGE_SHIFT;
 	*index = dev->mdev.priv.uuari.uars[0].index;
 
 	return 0;
@@ -653,8 +653,11 @@ struct ib_cq *mlx5_ib_create_cq(struct ib_device *ibdev, int entries,
 	int eqn;
 	int err;
 
+	if (entries < 0)
+		return ERR_PTR(-EINVAL);
+
 	entries = roundup_pow_of_two(entries + 1);
-	if (entries < 1 || entries > dev->mdev.caps.max_cqes)
+	if (entries > dev->mdev.caps.max_cqes)
 		return ERR_PTR(-EINVAL);
 
 	cq = kzalloc(sizeof(*cq), GFP_KERNEL);
@@ -747,17 +750,9 @@ int mlx5_ib_destroy_cq(struct ib_cq *cq)
 	return 0;
 }
 
-static int is_equal_rsn(struct mlx5_cqe64 *cqe64, struct mlx5_ib_srq *srq,
-			u32 rsn)
+static int is_equal_rsn(struct mlx5_cqe64 *cqe64, u32 rsn)
 {
-	u32 lrsn;
-
-	if (srq)
-		lrsn = be32_to_cpu(cqe64->srqn) & 0xffffff;
-	else
-		lrsn = be32_to_cpu(cqe64->sop_drop_qpn) & 0xffffff;
-
-	return rsn == lrsn;
+	return rsn == (ntohl(cqe64->sop_drop_qpn) & 0xffffff);
 }
 
 void __mlx5_ib_cq_clean(struct mlx5_ib_cq *cq, u32 rsn, struct mlx5_ib_srq *srq)
@@ -787,8 +782,8 @@ void __mlx5_ib_cq_clean(struct mlx5_ib_cq *cq, u32 rsn, struct mlx5_ib_srq *srq)
 	while ((int) --prod_index - (int) cq->mcq.cons_index >= 0) {
 		cqe = get_cqe(cq, prod_index & cq->ibcq.cqe);
 		cqe64 = (cq->mcq.cqe_sz == 64) ? cqe : cqe + 64;
-		if (is_equal_rsn(cqe64, srq, rsn)) {
-			if (srq)
+		if (is_equal_rsn(cqe64, rsn)) {
+			if (srq && (ntohl(cqe64->srqn) & 0xffffff))
 				mlx5_ib_free_srq_wqe(srq, be16_to_cpu(cqe64->wqe_counter));
 			++nfreed;
 		} else if (nfreed) {
diff --git a/drivers/infiniband/hw/mlx5/main.c b/drivers/infiniband/hw/mlx5/main.c
index b1a6cb3a2809..306534109627 100644
--- a/drivers/infiniband/hw/mlx5/main.c
+++ b/drivers/infiniband/hw/mlx5/main.c
@@ -745,7 +745,8 @@ static int alloc_pa_mkey(struct mlx5_ib_dev *dev, u32 *key, u32 pdn)
 	seg->qpn_mkey7_0 = cpu_to_be32(0xffffff << 8);
 	seg->start_addr = 0;
 
-	err = mlx5_core_create_mkey(&dev->mdev, &mr, in, sizeof(*in));
+	err = mlx5_core_create_mkey(&dev->mdev, &mr, in, sizeof(*in),
+				    NULL, NULL, NULL);
 	if (err) {
 		mlx5_ib_warn(dev, "failed to create mkey, %d\n", err);
 		goto err_in;
diff --git a/drivers/infiniband/hw/mlx5/mlx5_ib.h b/drivers/infiniband/hw/mlx5/mlx5_ib.h
index 836be9157242..4c134d93d4fc 100644
--- a/drivers/infiniband/hw/mlx5/mlx5_ib.h
+++ b/drivers/infiniband/hw/mlx5/mlx5_ib.h
@@ -262,6 +262,9 @@ struct mlx5_ib_mr {
 	int			npages;
 	struct completion	done;
 	enum ib_wc_status	status;
+	struct mlx5_ib_dev     *dev;
+	struct mlx5_create_mkey_mbox_out out;
+	unsigned long		start;
 };
 
 struct mlx5_ib_fast_reg_page_list {
@@ -323,6 +326,7 @@ struct mlx5_cache_ent {
 	struct mlx5_ib_dev     *dev;
 	struct work_struct	work;
 	struct delayed_work	dwork;
+	int			pending;
 };
 
 struct mlx5_mr_cache {
@@ -358,6 +362,8 @@ struct mlx5_ib_dev {
 	spinlock_t			mr_lock;
 	struct mlx5_ib_resources	devr;
 	struct mlx5_mr_cache		cache;
+	struct timer_list		delay_timer;
+	int				fill_delay;
 };
 
 static inline struct mlx5_ib_cq *to_mibcq(struct mlx5_core_cq *mcq)
diff --git a/drivers/infiniband/hw/mlx5/mr.c b/drivers/infiniband/hw/mlx5/mr.c
index 3453580b1eb2..039c3e40fcb4 100644
--- a/drivers/infiniband/hw/mlx5/mr.c
+++ b/drivers/infiniband/hw/mlx5/mr.c
@@ -35,11 +35,12 @@
 #include <linux/random.h>
 #include <linux/debugfs.h>
 #include <linux/export.h>
+#include <linux/delay.h>
 #include <rdma/ib_umem.h>
 #include "mlx5_ib.h"
 
 enum {
-	DEF_CACHE_SIZE	= 10,
+	MAX_PENDING_REG_MR = 8,
 };
 
 enum {
@@ -63,6 +64,51 @@ static int order2idx(struct mlx5_ib_dev *dev, int order)
 		return order - cache->ent[0].order;
 }
 
+static void reg_mr_callback(int status, void *context)
+{
+	struct mlx5_ib_mr *mr = context;
+	struct mlx5_ib_dev *dev = mr->dev;
+	struct mlx5_mr_cache *cache = &dev->cache;
+	int c = order2idx(dev, mr->order);
+	struct mlx5_cache_ent *ent = &cache->ent[c];
+	u8 key;
+	unsigned long flags;
+
+	spin_lock_irqsave(&ent->lock, flags);
+	ent->pending--;
+	spin_unlock_irqrestore(&ent->lock, flags);
+	if (status) {
+		mlx5_ib_warn(dev, "async reg mr failed. status %d\n", status);
+		kfree(mr);
+		dev->fill_delay = 1;
+		mod_timer(&dev->delay_timer, jiffies + HZ);
+		return;
+	}
+
+	if (mr->out.hdr.status) {
+		mlx5_ib_warn(dev, "failed - status %d, syndorme 0x%x\n",
+			     mr->out.hdr.status,
+			     be32_to_cpu(mr->out.hdr.syndrome));
+		kfree(mr);
+		dev->fill_delay = 1;
+		mod_timer(&dev->delay_timer, jiffies + HZ);
+		return;
+	}
+
+	spin_lock_irqsave(&dev->mdev.priv.mkey_lock, flags);
+	key = dev->mdev.priv.mkey_key++;
+	spin_unlock_irqrestore(&dev->mdev.priv.mkey_lock, flags);
+	mr->mmr.key = mlx5_idx_to_mkey(be32_to_cpu(mr->out.mkey) & 0xffffff) | key;
+
+	cache->last_add = jiffies;
+
+	spin_lock_irqsave(&ent->lock, flags);
+	list_add_tail(&mr->list, &ent->head);
+	ent->cur++;
+	ent->size++;
+	spin_unlock_irqrestore(&ent->lock, flags);
+}
+
 static int add_keys(struct mlx5_ib_dev *dev, int c, int num)
 {
 	struct mlx5_mr_cache *cache = &dev->cache;
@@ -78,36 +124,39 @@ static int add_keys(struct mlx5_ib_dev *dev, int c, int num)
 		return -ENOMEM;
 
 	for (i = 0; i < num; i++) {
+		if (ent->pending >= MAX_PENDING_REG_MR) {
+			err = -EAGAIN;
+			break;
+		}
+
 		mr = kzalloc(sizeof(*mr), GFP_KERNEL);
 		if (!mr) {
 			err = -ENOMEM;
-			goto out;
+			break;
 		}
 		mr->order = ent->order;
 		mr->umred = 1;
+		mr->dev = dev;
 		in->seg.status = 1 << 6;
 		in->seg.xlt_oct_size = cpu_to_be32((npages + 1) / 2);
 		in->seg.qpn_mkey7_0 = cpu_to_be32(0xffffff << 8);
 		in->seg.flags = MLX5_ACCESS_MODE_MTT | MLX5_PERM_UMR_EN;
 		in->seg.log2_page_size = 12;
 
+		spin_lock_irq(&ent->lock);
+		ent->pending++;
+		spin_unlock_irq(&ent->lock);
+		mr->start = jiffies;
 		err = mlx5_core_create_mkey(&dev->mdev, &mr->mmr, in,
-					    sizeof(*in));
+					    sizeof(*in), reg_mr_callback,
+					    mr, &mr->out);
 		if (err) {
 			mlx5_ib_warn(dev, "create mkey failed %d\n", err);
 			kfree(mr);
-			goto out;
+			break;
 		}
-		cache->last_add = jiffies;
-
-		spin_lock(&ent->lock);
-		list_add_tail(&mr->list, &ent->head);
-		ent->cur++;
-		ent->size++;
-		spin_unlock(&ent->lock);
 	}
 
-out:
 	kfree(in);
 	return err;
 }
@@ -121,16 +170,16 @@ static void remove_keys(struct mlx5_ib_dev *dev, int c, int num)
 	int i;
 
 	for (i = 0; i < num; i++) {
-		spin_lock(&ent->lock);
+		spin_lock_irq(&ent->lock);
 		if (list_empty(&ent->head)) {
-			spin_unlock(&ent->lock);
+			spin_unlock_irq(&ent->lock);
 			return;
 		}
 		mr = list_first_entry(&ent->head, struct mlx5_ib_mr, list);
 		list_del(&mr->list);
 		ent->cur--;
 		ent->size--;
-		spin_unlock(&ent->lock);
+		spin_unlock_irq(&ent->lock);
 		err = mlx5_core_destroy_mkey(&dev->mdev, &mr->mmr);
 		if (err)
 			mlx5_ib_warn(dev, "failed destroy mkey\n");
@@ -162,9 +211,13 @@ static ssize_t size_write(struct file *filp, const char __user *buf,
 		return -EINVAL;
 
 	if (var > ent->size) {
-		err = add_keys(dev, c, var - ent->size);
-		if (err)
-			return err;
+		do {
+			err = add_keys(dev, c, var - ent->size);
+			if (err && err != -EAGAIN)
+				return err;
+
+			usleep_range(3000, 5000);
+		} while (err);
 	} else if (var < ent->size) {
 		remove_keys(dev, c, ent->size - var);
 	}
@@ -280,23 +333,37 @@ static void __cache_work_func(struct mlx5_cache_ent *ent)
 	struct mlx5_ib_dev *dev = ent->dev;
 	struct mlx5_mr_cache *cache = &dev->cache;
 	int i = order2idx(dev, ent->order);
+	int err;
 
 	if (cache->stopped)
 		return;
 
 	ent = &dev->cache.ent[i];
-	if (ent->cur < 2 * ent->limit) {
-		add_keys(dev, i, 1);
-		if (ent->cur < 2 * ent->limit)
-			queue_work(cache->wq, &ent->work);
+	if (ent->cur < 2 * ent->limit && !dev->fill_delay) {
+		err = add_keys(dev, i, 1);
+		if (ent->cur < 2 * ent->limit) {
+			if (err == -EAGAIN) {
+				mlx5_ib_dbg(dev, "returned eagain, order %d\n",
+					    i + 2);
+				queue_delayed_work(cache->wq, &ent->dwork,
+						   msecs_to_jiffies(3));
+			} else if (err) {
+				mlx5_ib_warn(dev, "command failed order %d, err %d\n",
+					     i + 2, err);
+				queue_delayed_work(cache->wq, &ent->dwork,
+						   msecs_to_jiffies(1000));
+			} else {
+				queue_work(cache->wq, &ent->work);
+			}
+		}
 	} else if (ent->cur > 2 * ent->limit) {
 		if (!someone_adding(cache) &&
-		    time_after(jiffies, cache->last_add + 60 * HZ)) {
+		    time_after(jiffies, cache->last_add + 300 * HZ)) {
 			remove_keys(dev, i, 1);
 			if (ent->cur > ent->limit)
 				queue_work(cache->wq, &ent->work);
 		} else {
-			queue_delayed_work(cache->wq, &ent->dwork, 60 * HZ);
+			queue_delayed_work(cache->wq, &ent->dwork, 300 * HZ);
 		}
 	}
 }
@@ -336,18 +403,18 @@ static struct mlx5_ib_mr *alloc_cached_mr(struct mlx5_ib_dev *dev, int order)
 
 		mlx5_ib_dbg(dev, "order %d, cache index %d\n", ent->order, i);
 
-		spin_lock(&ent->lock);
+		spin_lock_irq(&ent->lock);
 		if (!list_empty(&ent->head)) {
 			mr = list_first_entry(&ent->head, struct mlx5_ib_mr,
 					      list);
 			list_del(&mr->list);
 			ent->cur--;
-			spin_unlock(&ent->lock);
+			spin_unlock_irq(&ent->lock);
 			if (ent->cur < ent->limit)
 				queue_work(cache->wq, &ent->work);
 			break;
 		}
-		spin_unlock(&ent->lock);
+		spin_unlock_irq(&ent->lock);
 
 		queue_work(cache->wq, &ent->work);
 
@@ -374,12 +441,12 @@ static void free_cached_mr(struct mlx5_ib_dev *dev, struct mlx5_ib_mr *mr)
 		return;
 	}
 	ent = &cache->ent[c];
-	spin_lock(&ent->lock);
+	spin_lock_irq(&ent->lock);
 	list_add_tail(&mr->list, &ent->head);
 	ent->cur++;
 	if (ent->cur > 2 * ent->limit)
 		shrink = 1;
-	spin_unlock(&ent->lock);
+	spin_unlock_irq(&ent->lock);
 
 	if (shrink)
 		queue_work(cache->wq, &ent->work);
@@ -394,16 +461,16 @@ static void clean_keys(struct mlx5_ib_dev *dev, int c)
 
 	cancel_delayed_work(&ent->dwork);
 	while (1) {
-		spin_lock(&ent->lock);
+		spin_lock_irq(&ent->lock);
 		if (list_empty(&ent->head)) {
-			spin_unlock(&ent->lock);
+			spin_unlock_irq(&ent->lock);
 			return;
 		}
 		mr = list_first_entry(&ent->head, struct mlx5_ib_mr, list);
 		list_del(&mr->list);
 		ent->cur--;
 		ent->size--;
-		spin_unlock(&ent->lock);
+		spin_unlock_irq(&ent->lock);
 		err = mlx5_core_destroy_mkey(&dev->mdev, &mr->mmr);
 		if (err)
 			mlx5_ib_warn(dev, "failed destroy mkey\n");
@@ -464,12 +531,18 @@ static void mlx5_mr_cache_debugfs_cleanup(struct mlx5_ib_dev *dev)
 	debugfs_remove_recursive(dev->cache.root);
 }
 
+static void delay_time_func(unsigned long ctx)
+{
+	struct mlx5_ib_dev *dev = (struct mlx5_ib_dev *)ctx;
+
+	dev->fill_delay = 0;
+}
+
 int mlx5_mr_cache_init(struct mlx5_ib_dev *dev)
 {
 	struct mlx5_mr_cache *cache = &dev->cache;
 	struct mlx5_cache_ent *ent;
 	int limit;
-	int size;
 	int err;
 	int i;
 
@@ -479,6 +552,7 @@ int mlx5_mr_cache_init(struct mlx5_ib_dev *dev)
 		return -ENOMEM;
 	}
 
+	setup_timer(&dev->delay_timer, delay_time_func, (unsigned long)dev);
 	for (i = 0; i < MAX_MR_CACHE_ENTRIES; i++) {
 		INIT_LIST_HEAD(&cache->ent[i].head);
 		spin_lock_init(&cache->ent[i].lock);
@@ -489,13 +563,11 @@ int mlx5_mr_cache_init(struct mlx5_ib_dev *dev)
 		ent->order = i + 2;
 		ent->dev = dev;
 
-		if (dev->mdev.profile->mask & MLX5_PROF_MASK_MR_CACHE) {
-			size = dev->mdev.profile->mr_cache[i].size;
+		if (dev->mdev.profile->mask & MLX5_PROF_MASK_MR_CACHE)
 			limit = dev->mdev.profile->mr_cache[i].limit;
-		} else {
-			size = DEF_CACHE_SIZE;
+		else
 			limit = 0;
-		}
+
 		INIT_WORK(&ent->work, cache_work_func);
 		INIT_DELAYED_WORK(&ent->dwork, delayed_cache_work_func);
 		ent->limit = limit;
@@ -522,6 +594,7 @@ int mlx5_mr_cache_cleanup(struct mlx5_ib_dev *dev)
 		clean_keys(dev, i);
 
 	destroy_workqueue(dev->cache.wq);
+	del_timer_sync(&dev->delay_timer);
 
 	return 0;
 }
@@ -551,7 +624,8 @@ struct ib_mr *mlx5_ib_get_dma_mr(struct ib_pd *pd, int acc)
 	seg->qpn_mkey7_0 = cpu_to_be32(0xffffff << 8);
 	seg->start_addr = 0;
 
-	err = mlx5_core_create_mkey(mdev, &mr->mmr, in, sizeof(*in));
+	err = mlx5_core_create_mkey(mdev, &mr->mmr, in, sizeof(*in), NULL, NULL,
+				    NULL);
 	if (err)
 		goto err_in;
 
@@ -660,14 +734,14 @@ static struct mlx5_ib_mr *reg_umr(struct ib_pd *pd, struct ib_umem *umem,
 	int err;
 	int i;
 
-	for (i = 0; i < 10; i++) {
+	for (i = 0; i < 1; i++) {
 		mr = alloc_cached_mr(dev, order);
 		if (mr)
 			break;
 
 		err = add_keys(dev, order2idx(dev, order), 1);
-		if (err) {
-			mlx5_ib_warn(dev, "add_keys failed\n");
+		if (err && err != -EAGAIN) {
+			mlx5_ib_warn(dev, "add_keys failed, err %d\n", err);
 			break;
 		}
 	}
@@ -759,8 +833,10 @@ static struct mlx5_ib_mr *reg_create(struct ib_pd *pd, u64 virt_addr,
 	in->seg.xlt_oct_size = cpu_to_be32(get_octo_len(virt_addr, length, 1 << page_shift));
 	in->seg.log2_page_size = page_shift;
 	in->seg.qpn_mkey7_0 = cpu_to_be32(0xffffff << 8);
-	in->xlat_oct_act_size = cpu_to_be32(get_octo_len(virt_addr, length, 1 << page_shift));
-	err = mlx5_core_create_mkey(&dev->mdev, &mr->mmr, in, inlen);
+	in->xlat_oct_act_size = cpu_to_be32(get_octo_len(virt_addr, length,
+							 1 << page_shift));
+	err = mlx5_core_create_mkey(&dev->mdev, &mr->mmr, in, inlen, NULL,
+				    NULL, NULL);
 	if (err) {
 		mlx5_ib_warn(dev, "create mkey failed\n");
 		goto err_2;
@@ -944,7 +1020,8 @@ struct ib_mr *mlx5_ib_alloc_fast_reg_mr(struct ib_pd *pd,
 	 * TBD not needed - issue 197292 */
 	in->seg.log2_page_size = PAGE_SHIFT;
 
-	err = mlx5_core_create_mkey(&dev->mdev, &mr->mmr, in, sizeof(*in));
+	err = mlx5_core_create_mkey(&dev->mdev, &mr->mmr, in, sizeof(*in), NULL,
+				    NULL, NULL);
 	kfree(in);
 	if (err)
 		goto err_free;
diff --git a/drivers/infiniband/hw/mlx5/qp.c b/drivers/infiniband/hw/mlx5/qp.c
index 5659ea880741..7c6b4ba49bec 100644
--- a/drivers/infiniband/hw/mlx5/qp.c
+++ b/drivers/infiniband/hw/mlx5/qp.c
@@ -551,7 +551,7 @@ static int create_user_qp(struct mlx5_ib_dev *dev, struct ib_pd *pd,
 	}
 	mlx5_ib_populate_pas(dev, qp->umem, page_shift, (*in)->pas, 0);
 	(*in)->ctx.log_pg_sz_remote_qpn =
-		cpu_to_be32((page_shift - PAGE_SHIFT) << 24);
+		cpu_to_be32((page_shift - MLX5_ADAPTER_PAGE_SHIFT) << 24);
 	(*in)->ctx.params2 = cpu_to_be32(offset << 6);
 
 	(*in)->ctx.qp_counter_set_usr_page = cpu_to_be32(uar_index);
@@ -648,7 +648,8 @@ static int create_kernel_qp(struct mlx5_ib_dev *dev,
 		goto err_buf;
 	}
 	(*in)->ctx.qp_counter_set_usr_page = cpu_to_be32(uar_index);
-	(*in)->ctx.log_pg_sz_remote_qpn = cpu_to_be32((qp->buf.page_shift - PAGE_SHIFT) << 24);
+	(*in)->ctx.log_pg_sz_remote_qpn =
+		cpu_to_be32((qp->buf.page_shift - MLX5_ADAPTER_PAGE_SHIFT) << 24);
 	/* Set "fast registration enabled" for all kernel QPs */
 	(*in)->ctx.params1 |= cpu_to_be32(1 << 11);
 	(*in)->ctx.sq_crq_size |= cpu_to_be16(1 << 4);
@@ -1317,9 +1318,11 @@ static enum mlx5_qp_optpar opt_mask[MLX5_QP_NUM_STATE][MLX5_QP_NUM_STATE][MLX5_Q
 					  MLX5_QP_OPTPAR_RAE		|
 					  MLX5_QP_OPTPAR_RWE		|
 					  MLX5_QP_OPTPAR_RNR_TIMEOUT	|
-					  MLX5_QP_OPTPAR_PM_STATE,
+					  MLX5_QP_OPTPAR_PM_STATE	|
+					  MLX5_QP_OPTPAR_ALT_ADDR_PATH,
 			[MLX5_QP_ST_UC] = MLX5_QP_OPTPAR_RWE		|
-					  MLX5_QP_OPTPAR_PM_STATE,
+					  MLX5_QP_OPTPAR_PM_STATE	|
+					  MLX5_QP_OPTPAR_ALT_ADDR_PATH,
 			[MLX5_QP_ST_UD] = MLX5_QP_OPTPAR_Q_KEY		|
 					  MLX5_QP_OPTPAR_SRQN		|
 					  MLX5_QP_OPTPAR_CQN_RCV,
@@ -1550,7 +1553,7 @@ static int __mlx5_ib_modify_qp(struct ib_qp *ibqp,
 	mlx5_cur = to_mlx5_state(cur_state);
 	mlx5_new = to_mlx5_state(new_state);
 	mlx5_st = to_mlx5_st(ibqp->qp_type);
-	if (mlx5_cur < 0 || mlx5_new < 0 || mlx5_st < 0)
+	if (mlx5_st < 0)
 		goto out;
 
 	optpar = ib_mask_to_mlx5_opt(attr_mask);
@@ -1744,6 +1747,7 @@ static void set_reg_umr_segment(struct mlx5_wqe_umr_ctrl_seg *umr,
 			MLX5_MKEY_MASK_PD		|
 			MLX5_MKEY_MASK_LR		|
 			MLX5_MKEY_MASK_LW		|
+			MLX5_MKEY_MASK_KEY		|
 			MLX5_MKEY_MASK_RR		|
 			MLX5_MKEY_MASK_RW		|
 			MLX5_MKEY_MASK_A		|
@@ -1800,7 +1804,8 @@ static void set_reg_mkey_segment(struct mlx5_mkey_seg *seg, struct ib_send_wr *w
 	seg->start_addr = cpu_to_be64(wr->wr.fast_reg.iova_start);
 	seg->len = cpu_to_be64(wr->wr.fast_reg.length);
 	seg->log2_page_size = wr->wr.fast_reg.page_shift;
-	seg->qpn_mkey7_0 = cpu_to_be32(0xffffff << 8);
+	seg->qpn_mkey7_0 = cpu_to_be32(0xffffff00 |
+				       mlx5_mkey_variant(wr->wr.fast_reg.rkey));
 }
 
 static void set_frwr_pages(struct mlx5_wqe_data_seg *dseg,
@@ -1913,6 +1918,10 @@ static int set_frwr_li_wr(void **seg, struct ib_send_wr *wr, int *size,
 	if (unlikely((*seg == qp->sq.qend)))
 		*seg = mlx5_get_send_wqe(qp, 0);
 	if (!li) {
+		if (unlikely(wr->wr.fast_reg.page_list_len >
+			     wr->wr.fast_reg.page_list->max_page_list_len))
+			return	-ENOMEM;
+
 		set_frwr_pages(*seg, wr, mdev, pd, writ);
 		*seg += sizeof(struct mlx5_wqe_data_seg);
 		*size += (sizeof(struct mlx5_wqe_data_seg) / 16);
diff --git a/drivers/infiniband/hw/mlx5/srq.c b/drivers/infiniband/hw/mlx5/srq.c
index 0aa478bc291a..210b3eaf188a 100644
--- a/drivers/infiniband/hw/mlx5/srq.c
+++ b/drivers/infiniband/hw/mlx5/srq.c
@@ -123,7 +123,7 @@ static int create_srq_user(struct ib_pd *pd, struct mlx5_ib_srq *srq,
 		goto err_in;
 	}
 
-	(*in)->ctx.log_pg_sz = page_shift - PAGE_SHIFT;
+	(*in)->ctx.log_pg_sz = page_shift - MLX5_ADAPTER_PAGE_SHIFT;
 	(*in)->ctx.pgoff_cqn = cpu_to_be32(offset << 26);
 
 	return 0;
@@ -192,7 +192,7 @@ static int create_srq_kernel(struct mlx5_ib_dev *dev, struct mlx5_ib_srq *srq,
 	}
 	srq->wq_sig = !!srq_signature;
 
-	(*in)->ctx.log_pg_sz = page_shift - PAGE_SHIFT;
+	(*in)->ctx.log_pg_sz = page_shift - MLX5_ADAPTER_PAGE_SHIFT;
 
 	return 0;
 
@@ -390,9 +390,7 @@ int mlx5_ib_destroy_srq(struct ib_srq *srq)
 		mlx5_ib_db_unmap_user(to_mucontext(srq->uobject->context), &msrq->db);
 		ib_umem_release(msrq->umem);
 	} else {
-		kfree(msrq->wrid);
-		mlx5_buf_free(&dev->mdev, &msrq->buf);
-		mlx5_db_free(&dev->mdev, &msrq->db);
+		destroy_srq_kernel(dev, msrq);
 	}
 
 	kfree(srq);
diff --git a/drivers/infiniband/hw/nes/nes_verbs.c b/drivers/infiniband/hw/nes/nes_verbs.c
index 5b53ca5a2284..8308e3634767 100644
--- a/drivers/infiniband/hw/nes/nes_verbs.c
+++ b/drivers/infiniband/hw/nes/nes_verbs.c
@@ -2834,7 +2834,7 @@ static int nes_query_qp(struct ib_qp *ibqp, struct ib_qp_attr *attr,
 	init_attr->qp_context = nesqp->ibqp.qp_context;
 	init_attr->send_cq = nesqp->ibqp.send_cq;
 	init_attr->recv_cq = nesqp->ibqp.recv_cq;
-	init_attr->srq = nesqp->ibqp.srq = nesqp->ibqp.srq;
+	init_attr->srq = nesqp->ibqp.srq;
 	init_attr->cap = attr->cap;
 
 	return 0;
diff --git a/drivers/infiniband/hw/ocrdma/ocrdma.h b/drivers/infiniband/hw/ocrdma/ocrdma.h
index adc11d14f878..294dd27b601e 100644
--- a/drivers/infiniband/hw/ocrdma/ocrdma.h
+++ b/drivers/infiniband/hw/ocrdma/ocrdma.h
@@ -122,6 +122,32 @@ struct mqe_ctx {
 	bool cmd_done;
 };
 
+struct ocrdma_hw_mr {
+	u32 lkey;
+	u8 fr_mr;
+	u8 remote_atomic;
+	u8 remote_rd;
+	u8 remote_wr;
+	u8 local_rd;
+	u8 local_wr;
+	u8 mw_bind;
+	u8 rsvd;
+	u64 len;
+	struct ocrdma_pbl *pbl_table;
+	u32 num_pbls;
+	u32 num_pbes;
+	u32 pbl_size;
+	u32 pbe_size;
+	u64 fbo;
+	u64 va;
+};
+
+struct ocrdma_mr {
+	struct ib_mr ibmr;
+	struct ib_umem *umem;
+	struct ocrdma_hw_mr hwmr;
+};
+
 struct ocrdma_dev {
 	struct ib_device ibdev;
 	struct ocrdma_dev_attr attr;
@@ -169,7 +195,7 @@ struct ocrdma_dev {
 	struct list_head entry;
 	struct rcu_head rcu;
 	int id;
-	u64 stag_arr[OCRDMA_MAX_STAG];
+	struct ocrdma_mr *stag_arr[OCRDMA_MAX_STAG];
 	u16 pvid;
 };
 
@@ -294,31 +320,6 @@ struct ocrdma_qp {
 	u16 db_cache;
 };
 
-struct ocrdma_hw_mr {
-	u32 lkey;
-	u8 fr_mr;
-	u8 remote_atomic;
-	u8 remote_rd;
-	u8 remote_wr;
-	u8 local_rd;
-	u8 local_wr;
-	u8 mw_bind;
-	u8 rsvd;
-	u64 len;
-	struct ocrdma_pbl *pbl_table;
-	u32 num_pbls;
-	u32 num_pbes;
-	u32 pbl_size;
-	u32 pbe_size;
-	u64 fbo;
-	u64 va;
-};
-
-struct ocrdma_mr {
-	struct ib_mr ibmr;
-	struct ib_umem *umem;
-	struct ocrdma_hw_mr hwmr;
-};
 
 struct ocrdma_ucontext {
 	struct ib_ucontext ibucontext;
diff --git a/drivers/infiniband/hw/ocrdma/ocrdma_hw.c b/drivers/infiniband/hw/ocrdma/ocrdma_hw.c
index 50219ab2279d..56bf32fcb62c 100644
--- a/drivers/infiniband/hw/ocrdma/ocrdma_hw.c
+++ b/drivers/infiniband/hw/ocrdma/ocrdma_hw.c
@@ -1783,7 +1783,7 @@ static int ocrdma_set_create_qp_sq_cmd(struct ocrdma_create_qp_req *cmd,
 	u32 max_sges = attrs->cap.max_send_sge;
 
 	/* QP1 may exceed 127 */
-	max_wqe_allocated = min_t(int, attrs->cap.max_send_wr + 1,
+	max_wqe_allocated = min_t(u32, attrs->cap.max_send_wr + 1,
 				dev->attr.max_wqe);
 
 	status = ocrdma_build_q_conf(&max_wqe_allocated,
diff --git a/drivers/infiniband/hw/ocrdma/ocrdma_main.c b/drivers/infiniband/hw/ocrdma/ocrdma_main.c
index 0ce7674621ea..91443bcb9e0e 100644
--- a/drivers/infiniband/hw/ocrdma/ocrdma_main.c
+++ b/drivers/infiniband/hw/ocrdma/ocrdma_main.c
@@ -452,9 +452,6 @@ static void ocrdma_remove_free(struct rcu_head *rcu)
 {
 	struct ocrdma_dev *dev = container_of(rcu, struct ocrdma_dev, rcu);
 
-	ocrdma_free_resources(dev);
-	ocrdma_cleanup_hw(dev);
-
 	idr_remove(&ocrdma_dev_id, dev->id);
 	kfree(dev->mbx_cmd);
 	ib_dealloc_device(&dev->ibdev);
@@ -470,6 +467,10 @@ static void ocrdma_remove(struct ocrdma_dev *dev)
 	spin_lock(&ocrdma_devlist_lock);
 	list_del_rcu(&dev->entry);
 	spin_unlock(&ocrdma_devlist_lock);
+
+	ocrdma_free_resources(dev);
+	ocrdma_cleanup_hw(dev);
+
 	call_rcu(&dev->rcu, ocrdma_remove_free);
 }
 
diff --git a/drivers/infiniband/hw/ocrdma/ocrdma_verbs.c b/drivers/infiniband/hw/ocrdma/ocrdma_verbs.c
index 69f1d1221a6b..7686dceadd29 100644
--- a/drivers/infiniband/hw/ocrdma/ocrdma_verbs.c
+++ b/drivers/infiniband/hw/ocrdma/ocrdma_verbs.c
@@ -1981,9 +1981,7 @@ static int ocrdma_build_fr(struct ocrdma_qp *qp, struct ocrdma_hdr_wqe *hdr,
 
 	wqe_size = roundup(wqe_size, OCRDMA_WQE_ALIGN_BYTES);
 
-	if ((wr->wr.fast_reg.page_list_len >
-		qp->dev->attr.max_pages_per_frmr) ||
-		(wr->wr.fast_reg.length > 0xffffffffULL))
+	if (wr->wr.fast_reg.page_list_len > qp->dev->attr.max_pages_per_frmr)
 		return -EINVAL;
 
 	hdr->cw |= (OCRDMA_FR_MR << OCRDMA_WQE_OPCODE_SHIFT);
@@ -2839,7 +2837,7 @@ struct ib_mr *ocrdma_alloc_frmr(struct ib_pd *ibpd, int max_page_list_len)
 		goto mbx_err;
 	mr->ibmr.rkey = mr->hwmr.lkey;
 	mr->ibmr.lkey = mr->hwmr.lkey;
-	dev->stag_arr[(mr->hwmr.lkey >> 8) & (OCRDMA_MAX_STAG - 1)] = (unsigned long) mr;
+	dev->stag_arr[(mr->hwmr.lkey >> 8) & (OCRDMA_MAX_STAG - 1)] = mr;
 	return &mr->ibmr;
 mbx_err:
 	ocrdma_free_mr_pbl_tbl(dev, &mr->hwmr);
diff --git a/drivers/infiniband/hw/qib/qib_iba7322.c b/drivers/infiniband/hw/qib/qib_iba7322.c
index 016e7429adf6..5bfc02f450e6 100644
--- a/drivers/infiniband/hw/qib/qib_iba7322.c
+++ b/drivers/infiniband/hw/qib/qib_iba7322.c
@@ -6190,21 +6190,20 @@ static int setup_txselect(const char *str, struct kernel_param *kp)
 {
 	struct qib_devdata *dd;
 	unsigned long val;
-	int ret;
-
+	char *n;
 	if (strlen(str) >= MAX_ATTEN_LEN) {
 		pr_info("txselect_values string too long\n");
 		return -ENOSPC;
 	}
-	ret = kstrtoul(str, 0, &val);
-	if (ret || val >= (TXDDS_TABLE_SZ + TXDDS_EXTRA_SZ +
+	val = simple_strtoul(str, &n, 0);
+	if (n == str || val >= (TXDDS_TABLE_SZ + TXDDS_EXTRA_SZ +
 				TXDDS_MFG_SZ)) {
 		pr_info("txselect_values must start with a number < %d\n",
 			TXDDS_TABLE_SZ + TXDDS_EXTRA_SZ + TXDDS_MFG_SZ);
-		return ret ? ret : -EINVAL;
+		return -EINVAL;
 	}
-
 	strcpy(txselect_list, str);
+
 	list_for_each_entry(dd, &qib_dev_list, list)
 		if (dd->deviceid == PCI_DEVICE_ID_QLOGIC_IB_7322)
 			set_no_qsfp_atten(dd, 1);
diff --git a/drivers/infiniband/hw/qib/qib_mad.h b/drivers/infiniband/hw/qib/qib_mad.h
index 28874f8606f8..941d4d50d8e7 100644
--- a/drivers/infiniband/hw/qib/qib_mad.h
+++ b/drivers/infiniband/hw/qib/qib_mad.h
@@ -54,7 +54,7 @@ struct ib_node_info {
 	__be32 revision;
 	u8 local_port_num;
 	u8 vendor_id[3];
-} __attribute__ ((packed));
+} __packed;
 
 struct ib_mad_notice_attr {
 	u8 generic_type;
@@ -73,7 +73,7 @@ struct ib_mad_notice_attr {
 			__be16	reserved;
 			__be16	lid;		/* where violation happened */
 			u8	port_num;	/* where violation happened */
-		} __attribute__ ((packed)) ntc_129_131;
+		} __packed ntc_129_131;
 
 		struct {
 			__be16	reserved;
@@ -83,14 +83,14 @@ struct ib_mad_notice_attr {
 			__be32	new_cap_mask;	/* new capability mask */
 			u8	reserved3;
 			u8	change_flags;	/* low 3 bits only */
-		} __attribute__ ((packed)) ntc_144;
+		} __packed ntc_144;
 
 		struct {
 			__be16	reserved;
 			__be16	lid;		/* lid where sys guid changed */
 			__be16	reserved2;
 			__be64	new_sys_guid;
-		} __attribute__ ((packed)) ntc_145;
+		} __packed ntc_145;
 
 		struct {
 			__be16	reserved;
@@ -104,7 +104,7 @@ struct ib_mad_notice_attr {
 			u8	reserved3;
 			u8	dr_trunc_hop;
 			u8	dr_rtn_path[30];
-		} __attribute__ ((packed)) ntc_256;
+		} __packed ntc_256;
 
 		struct {
 			__be16		reserved;
@@ -115,7 +115,7 @@ struct ib_mad_notice_attr {
 			__be32		qp2;	/* high 8 bits reserved */
 			union ib_gid	gid1;
 			union ib_gid	gid2;
-		} __attribute__ ((packed)) ntc_257_258;
+		} __packed ntc_257_258;
 
 	} details;
 };
@@ -209,7 +209,7 @@ struct ib_pma_portcounters_cong {
 	__be64 port_rcv_packets;
 	__be64 port_xmit_wait;
 	__be64 port_adr_events;
-} __attribute__ ((packed));
+} __packed;
 
 #define IB_PMA_CONG_HW_CONTROL_TIMER            0x00
 #define IB_PMA_CONG_HW_CONTROL_SAMPLE           0x01
diff --git a/drivers/infiniband/hw/qib/qib_user_sdma.c b/drivers/infiniband/hw/qib/qib_user_sdma.c
index d0a0ea0c14d6..165aee2ca8a0 100644
--- a/drivers/infiniband/hw/qib/qib_user_sdma.c
+++ b/drivers/infiniband/hw/qib/qib_user_sdma.c
@@ -594,8 +594,7 @@ static int qib_user_sdma_pin_pages(const struct qib_devdata *dd,
 		else
 			j = npages;
 
-		ret = get_user_pages(current, current->mm, addr,
-			     j, 0, 1, pages, NULL);
+		ret = get_user_pages_fast(addr, j, 0, pages);
 		if (ret != j) {
 			i = 0;
 			j = ret;
@@ -1294,11 +1293,8 @@ int qib_user_sdma_writev(struct qib_ctxtdata *rcd,
 		int mxp = 8;
 		int ndesc = 0;
 
-		down_write(&current->mm->mmap_sem);
 		ret = qib_user_sdma_queue_pkts(dd, ppd, pq,
 				iov, dim, &list, &mxp, &ndesc);
-		up_write(&current->mm->mmap_sem);
-
 		if (ret < 0)
 			goto done_unlock;
 		else {
diff --git a/drivers/infiniband/hw/qib/qib_verbs.h b/drivers/infiniband/hw/qib/qib_verbs.h
index 012e2c7575ad..a01c7d2cf541 100644
--- a/drivers/infiniband/hw/qib/qib_verbs.h
+++ b/drivers/infiniband/hw/qib/qib_verbs.h
@@ -150,14 +150,14 @@ struct ib_reth {
 	__be64 vaddr;
 	__be32 rkey;
 	__be32 length;
-} __attribute__ ((packed));
+} __packed;
 
 struct ib_atomic_eth {
 	__be32 vaddr[2];        /* unaligned so access as 2 32-bit words */
 	__be32 rkey;
 	__be64 swap_data;
 	__be64 compare_data;
-} __attribute__ ((packed));
+} __packed;
 
 struct qib_other_headers {
 	__be32 bth[3];
@@ -178,7 +178,7 @@ struct qib_other_headers {
 		__be32 aeth;
 		struct ib_atomic_eth atomic_eth;
 	} u;
-} __attribute__ ((packed));
+} __packed;
 
 /*
  * Note that UD packets with a GRH header are 8+40+12+8 = 68 bytes
@@ -195,12 +195,12 @@ struct qib_ib_header {
 		} l;
 		struct qib_other_headers oth;
 	} u;
-} __attribute__ ((packed));
+} __packed;
 
 struct qib_pio_header {
 	__le32 pbc[2];
 	struct qib_ib_header hdr;
-} __attribute__ ((packed));
+} __packed;
 
 /*
  * There is one struct qib_mcast for each multicast GID.
diff --git a/drivers/infiniband/ulp/ipoib/ipoib.h b/drivers/infiniband/ulp/ipoib/ipoib.h
index eb71aaa26a9a..c639f90cfda4 100644
--- a/drivers/infiniband/ulp/ipoib/ipoib.h
+++ b/drivers/infiniband/ulp/ipoib/ipoib.h
@@ -101,6 +101,7 @@ enum {
 	IPOIB_MCAST_FLAG_SENDONLY = 1,
 	IPOIB_MCAST_FLAG_BUSY	  = 2,	/* joining or already joined */
 	IPOIB_MCAST_FLAG_ATTACHED = 3,
+	IPOIB_MCAST_JOIN_STARTED  = 4,
 
 	MAX_SEND_CQE		  = 16,
 	IPOIB_CM_COPYBREAK	  = 256,
@@ -151,6 +152,7 @@ struct ipoib_mcast {
 	struct sk_buff_head pkt_queue;
 
 	struct net_device *dev;
+	struct completion done;
 };
 
 struct ipoib_rx_buf {
@@ -299,7 +301,7 @@ struct ipoib_dev_priv {
 
 	unsigned long flags;
 
-	struct mutex vlan_mutex;
+	struct rw_semaphore vlan_rwsem;
 
 	struct rb_root  path_tree;
 	struct list_head path_list;
diff --git a/drivers/infiniband/ulp/ipoib/ipoib_cm.c b/drivers/infiniband/ulp/ipoib/ipoib_cm.c
index 7a3175400b2a..1377f85911c2 100644
--- a/drivers/infiniband/ulp/ipoib/ipoib_cm.c
+++ b/drivers/infiniband/ulp/ipoib/ipoib_cm.c
@@ -140,7 +140,8 @@ static int ipoib_cm_post_receive_nonsrq(struct net_device *dev,
 static struct sk_buff *ipoib_cm_alloc_rx_skb(struct net_device *dev,
 					     struct ipoib_cm_rx_buf *rx_ring,
 					     int id, int frags,
-					     u64 mapping[IPOIB_CM_RX_SG])
+					     u64 mapping[IPOIB_CM_RX_SG],
+					     gfp_t gfp)
 {
 	struct ipoib_dev_priv *priv = netdev_priv(dev);
 	struct sk_buff *skb;
@@ -164,7 +165,7 @@ static struct sk_buff *ipoib_cm_alloc_rx_skb(struct net_device *dev,
 	}
 
 	for (i = 0; i < frags; i++) {
-		struct page *page = alloc_page(GFP_ATOMIC);
+		struct page *page = alloc_page(gfp);
 
 		if (!page)
 			goto partial_error;
@@ -382,7 +383,8 @@ static int ipoib_cm_nonsrq_init_rx(struct net_device *dev, struct ib_cm_id *cm_i
 
 	for (i = 0; i < ipoib_recvq_size; ++i) {
 		if (!ipoib_cm_alloc_rx_skb(dev, rx->rx_ring, i, IPOIB_CM_RX_SG - 1,
-					   rx->rx_ring[i].mapping)) {
+					   rx->rx_ring[i].mapping,
+					   GFP_KERNEL)) {
 			ipoib_warn(priv, "failed to allocate receive buffer %d\n", i);
 				ret = -ENOMEM;
 				goto err_count;
@@ -639,7 +641,8 @@ void ipoib_cm_handle_rx_wc(struct net_device *dev, struct ib_wc *wc)
 	frags = PAGE_ALIGN(wc->byte_len - min(wc->byte_len,
 					      (unsigned)IPOIB_CM_HEAD_SIZE)) / PAGE_SIZE;
 
-	newskb = ipoib_cm_alloc_rx_skb(dev, rx_ring, wr_id, frags, mapping);
+	newskb = ipoib_cm_alloc_rx_skb(dev, rx_ring, wr_id, frags,
+				       mapping, GFP_ATOMIC);
 	if (unlikely(!newskb)) {
 		/*
 		 * If we can't allocate a new RX buffer, dump
@@ -1556,7 +1559,8 @@ int ipoib_cm_dev_init(struct net_device *dev)
 		for (i = 0; i < ipoib_recvq_size; ++i) {
 			if (!ipoib_cm_alloc_rx_skb(dev, priv->cm.srq_ring, i,
 						   priv->cm.num_frags - 1,
-						   priv->cm.srq_ring[i].mapping)) {
+						   priv->cm.srq_ring[i].mapping,
+						   GFP_KERNEL)) {
 				ipoib_warn(priv, "failed to allocate "
 					   "receive buffer %d\n", i);
 				ipoib_cm_dev_cleanup(dev);
diff --git a/drivers/infiniband/ulp/ipoib/ipoib_ib.c b/drivers/infiniband/ulp/ipoib/ipoib_ib.c
index 196b1d13cbcb..6a7003ddb0be 100644
--- a/drivers/infiniband/ulp/ipoib/ipoib_ib.c
+++ b/drivers/infiniband/ulp/ipoib/ipoib_ib.c
@@ -685,15 +685,13 @@ int ipoib_ib_dev_open(struct net_device *dev)
 	ret = ipoib_ib_post_receives(dev);
 	if (ret) {
 		ipoib_warn(priv, "ipoib_ib_post_receives returned %d\n", ret);
-		ipoib_ib_dev_stop(dev, 1);
-		return -1;
+		goto dev_stop;
 	}
 
 	ret = ipoib_cm_dev_open(dev);
 	if (ret) {
 		ipoib_warn(priv, "ipoib_cm_dev_open returned %d\n", ret);
-		ipoib_ib_dev_stop(dev, 1);
-		return -1;
+		goto dev_stop;
 	}
 
 	clear_bit(IPOIB_STOP_REAPER, &priv->flags);
@@ -704,6 +702,11 @@ int ipoib_ib_dev_open(struct net_device *dev)
 		napi_enable(&priv->napi);
 
 	return 0;
+dev_stop:
+	if (!test_and_set_bit(IPOIB_FLAG_INITIALIZED, &priv->flags))
+		napi_enable(&priv->napi);
+	ipoib_ib_dev_stop(dev, 1);
+	return -1;
 }
 
 static void ipoib_pkey_dev_check_presence(struct net_device *dev)
@@ -746,10 +749,8 @@ int ipoib_ib_dev_down(struct net_device *dev, int flush)
 	if (!test_bit(IPOIB_PKEY_ASSIGNED, &priv->flags)) {
 		mutex_lock(&pkey_mutex);
 		set_bit(IPOIB_PKEY_STOP, &priv->flags);
-		cancel_delayed_work(&priv->pkey_poll_task);
+		cancel_delayed_work_sync(&priv->pkey_poll_task);
 		mutex_unlock(&pkey_mutex);
-		if (flush)
-			flush_workqueue(ipoib_workqueue);
 	}
 
 	ipoib_mcast_stop_thread(dev, flush);
@@ -974,7 +975,7 @@ static void __ipoib_ib_dev_flush(struct ipoib_dev_priv *priv,
 	u16 new_index;
 	int result;
 
-	mutex_lock(&priv->vlan_mutex);
+	down_read(&priv->vlan_rwsem);
 
 	/*
 	 * Flush any child interfaces too -- they might be up even if
@@ -983,7 +984,7 @@ static void __ipoib_ib_dev_flush(struct ipoib_dev_priv *priv,
 	list_for_each_entry(cpriv, &priv->child_intfs, list)
 		__ipoib_ib_dev_flush(cpriv, level);
 
-	mutex_unlock(&priv->vlan_mutex);
+	up_read(&priv->vlan_rwsem);
 
 	if (!test_bit(IPOIB_FLAG_INITIALIZED, &priv->flags)) {
 		/* for non-child devices must check/update the pkey value here */
@@ -1081,6 +1082,11 @@ void ipoib_ib_dev_cleanup(struct net_device *dev)
 	struct ipoib_dev_priv *priv = netdev_priv(dev);
 
 	ipoib_dbg(priv, "cleaning up ib_dev\n");
+	/*
+	 * We must make sure there are no more (path) completions
+	 * that may wish to touch priv fields that are no longer valid
+	 */
+	ipoib_flush_paths(dev);
 
 	ipoib_mcast_stop_thread(dev, 1);
 	ipoib_mcast_dev_flush(dev);
diff --git a/drivers/infiniband/ulp/ipoib/ipoib_main.c b/drivers/infiniband/ulp/ipoib/ipoib_main.c
index 82cec1af902c..d64ed05fb082 100644
--- a/drivers/infiniband/ulp/ipoib/ipoib_main.c
+++ b/drivers/infiniband/ulp/ipoib/ipoib_main.c
@@ -119,7 +119,7 @@ int ipoib_open(struct net_device *dev)
 		struct ipoib_dev_priv *cpriv;
 
 		/* Bring up any child interfaces too */
-		mutex_lock(&priv->vlan_mutex);
+		down_read(&priv->vlan_rwsem);
 		list_for_each_entry(cpriv, &priv->child_intfs, list) {
 			int flags;
 
@@ -129,7 +129,7 @@ int ipoib_open(struct net_device *dev)
 
 			dev_change_flags(cpriv->dev, flags | IFF_UP);
 		}
-		mutex_unlock(&priv->vlan_mutex);
+		up_read(&priv->vlan_rwsem);
 	}
 
 	netif_start_queue(dev);
@@ -162,7 +162,7 @@ static int ipoib_stop(struct net_device *dev)
 		struct ipoib_dev_priv *cpriv;
 
 		/* Bring down any child interfaces too */
-		mutex_lock(&priv->vlan_mutex);
+		down_read(&priv->vlan_rwsem);
 		list_for_each_entry(cpriv, &priv->child_intfs, list) {
 			int flags;
 
@@ -172,7 +172,7 @@ static int ipoib_stop(struct net_device *dev)
 
 			dev_change_flags(cpriv->dev, flags & ~IFF_UP);
 		}
-		mutex_unlock(&priv->vlan_mutex);
+		up_read(&priv->vlan_rwsem);
 	}
 
 	return 0;
@@ -1350,7 +1350,7 @@ void ipoib_setup(struct net_device *dev)
 
 	ipoib_set_ethtool_ops(dev);
 
-	netif_napi_add(dev, &priv->napi, ipoib_poll, 100);
+	netif_napi_add(dev, &priv->napi, ipoib_poll, NAPI_POLL_WEIGHT);
 
 	dev->watchdog_timeo	 = HZ;
 
@@ -1372,7 +1372,7 @@ void ipoib_setup(struct net_device *dev)
 
 	spin_lock_init(&priv->lock);
 
-	mutex_init(&priv->vlan_mutex);
+	init_rwsem(&priv->vlan_rwsem);
 
 	INIT_LIST_HEAD(&priv->path_list);
 	INIT_LIST_HEAD(&priv->child_intfs);
diff --git a/drivers/infiniband/ulp/ipoib/ipoib_multicast.c b/drivers/infiniband/ulp/ipoib/ipoib_multicast.c
index cecb98a4c662..d4e005720d01 100644
--- a/drivers/infiniband/ulp/ipoib/ipoib_multicast.c
+++ b/drivers/infiniband/ulp/ipoib/ipoib_multicast.c
@@ -386,8 +386,10 @@ static int ipoib_mcast_join_complete(int status,
 			mcast->mcmember.mgid.raw, status);
 
 	/* We trap for port events ourselves. */
-	if (status == -ENETRESET)
-		return 0;
+	if (status == -ENETRESET) {
+		status = 0;
+		goto out;
+	}
 
 	if (!status)
 		status = ipoib_mcast_join_finish(mcast, &multicast->rec);
@@ -407,7 +409,8 @@ static int ipoib_mcast_join_complete(int status,
 		if (mcast == priv->broadcast)
 			queue_work(ipoib_workqueue, &priv->carrier_on_task);
 
-		return 0;
+		status = 0;
+		goto out;
 	}
 
 	if (mcast->logcount++ < 20) {
@@ -434,7 +437,8 @@ static int ipoib_mcast_join_complete(int status,
 				   mcast->backoff * HZ);
 	spin_unlock_irq(&priv->lock);
 	mutex_unlock(&mcast_mutex);
-
+out:
+	complete(&mcast->done);
 	return status;
 }
 
@@ -484,11 +488,15 @@ static void ipoib_mcast_join(struct net_device *dev, struct ipoib_mcast *mcast,
 	}
 
 	set_bit(IPOIB_MCAST_FLAG_BUSY, &mcast->flags);
+	init_completion(&mcast->done);
+	set_bit(IPOIB_MCAST_JOIN_STARTED, &mcast->flags);
+
 	mcast->mc = ib_sa_join_multicast(&ipoib_sa_client, priv->ca, priv->port,
 					 &rec, comp_mask, GFP_KERNEL,
 					 ipoib_mcast_join_complete, mcast);
 	if (IS_ERR(mcast->mc)) {
 		clear_bit(IPOIB_MCAST_FLAG_BUSY, &mcast->flags);
+		complete(&mcast->done);
 		ret = PTR_ERR(mcast->mc);
 		ipoib_warn(priv, "ib_sa_join_multicast failed, status %d\n", ret);
 
@@ -510,10 +518,18 @@ void ipoib_mcast_join_task(struct work_struct *work)
 	struct ipoib_dev_priv *priv =
 		container_of(work, struct ipoib_dev_priv, mcast_task.work);
 	struct net_device *dev = priv->dev;
+	struct ib_port_attr port_attr;
 
 	if (!test_bit(IPOIB_MCAST_RUN, &priv->flags))
 		return;
 
+	if (ib_query_port(priv->ca, priv->port, &port_attr) ||
+	    port_attr.state != IB_PORT_ACTIVE) {
+		ipoib_dbg(priv, "port state is not ACTIVE (state = %d) suspending join task\n",
+			  port_attr.state);
+		return;
+	}
+
 	if (ib_query_gid(priv->ca, priv->port, 0, &priv->local_gid))
 		ipoib_warn(priv, "ib_query_gid() failed\n");
 	else
@@ -751,6 +767,11 @@ void ipoib_mcast_dev_flush(struct net_device *dev)
 
 	spin_unlock_irqrestore(&priv->lock, flags);
 
+	/* seperate between the wait to the leave*/
+	list_for_each_entry_safe(mcast, tmcast, &remove_list, list)
+		if (test_bit(IPOIB_MCAST_JOIN_STARTED, &mcast->flags))
+			wait_for_completion(&mcast->done);
+
 	list_for_each_entry_safe(mcast, tmcast, &remove_list, list) {
 		ipoib_mcast_leave(dev, mcast);
 		ipoib_mcast_free(mcast);
diff --git a/drivers/infiniband/ulp/ipoib/ipoib_netlink.c b/drivers/infiniband/ulp/ipoib/ipoib_netlink.c
index f81abe16cf09..c29b5c838833 100644
--- a/drivers/infiniband/ulp/ipoib/ipoib_netlink.c
+++ b/drivers/infiniband/ulp/ipoib/ipoib_netlink.c
@@ -142,10 +142,10 @@ static void ipoib_unregister_child_dev(struct net_device *dev, struct list_head
 	priv = netdev_priv(dev);
 	ppriv = netdev_priv(priv->parent);
 
-	mutex_lock(&ppriv->vlan_mutex);
+	down_write(&ppriv->vlan_rwsem);
 	unregister_netdevice_queue(dev, head);
 	list_del(&priv->list);
-	mutex_unlock(&ppriv->vlan_mutex);
+	up_write(&ppriv->vlan_rwsem);
 }
 
 static size_t ipoib_get_size(const struct net_device *dev)
diff --git a/drivers/infiniband/ulp/ipoib/ipoib_vlan.c b/drivers/infiniband/ulp/ipoib/ipoib_vlan.c
index 8292554bccb5..9fad7b5ac8b9 100644
--- a/drivers/infiniband/ulp/ipoib/ipoib_vlan.c
+++ b/drivers/infiniband/ulp/ipoib/ipoib_vlan.c
@@ -140,7 +140,7 @@ int ipoib_vlan_add(struct net_device *pdev, unsigned short pkey)
 	if (!rtnl_trylock())
 		return restart_syscall();
 
-	mutex_lock(&ppriv->vlan_mutex);
+	down_write(&ppriv->vlan_rwsem);
 
 	/*
 	 * First ensure this isn't a duplicate. We check the parent device and
@@ -163,7 +163,7 @@ int ipoib_vlan_add(struct net_device *pdev, unsigned short pkey)
 	result = __ipoib_vlan_add(ppriv, priv, pkey, IPOIB_LEGACY_CHILD);
 
 out:
-	mutex_unlock(&ppriv->vlan_mutex);
+	up_write(&ppriv->vlan_rwsem);
 
 	if (result)
 		free_netdev(priv->dev);
@@ -185,7 +185,8 @@ int ipoib_vlan_delete(struct net_device *pdev, unsigned short pkey)
 
 	if (!rtnl_trylock())
 		return restart_syscall();
-	mutex_lock(&ppriv->vlan_mutex);
+
+	down_write(&ppriv->vlan_rwsem);
 	list_for_each_entry_safe(priv, tpriv, &ppriv->child_intfs, list) {
 		if (priv->pkey == pkey &&
 		    priv->child_type == IPOIB_LEGACY_CHILD) {
@@ -195,7 +196,8 @@ int ipoib_vlan_delete(struct net_device *pdev, unsigned short pkey)
 			break;
 		}
 	}
-	mutex_unlock(&ppriv->vlan_mutex);
+	up_write(&ppriv->vlan_rwsem);
+
 	rtnl_unlock();
 
 	if (dev) {
diff --git a/drivers/infiniband/ulp/isert/Kconfig b/drivers/infiniband/ulp/isert/Kconfig
index ce3fd32167dc..02f9759ebb1a 100644
--- a/drivers/infiniband/ulp/isert/Kconfig
+++ b/drivers/infiniband/ulp/isert/Kconfig
@@ -1,5 +1,5 @@
 config INFINIBAND_ISERT
-	tristate "iSCSI Extentions for RDMA (iSER) target support"
+	tristate "iSCSI Extensions for RDMA (iSER) target support"
 	depends on INET && INFINIBAND_ADDR_TRANS && TARGET_CORE && ISCSI_TARGET
 	---help---
-	Support for iSCSI Extentions for RDMA (iSER) Target on Infiniband fabrics.
+	Support for iSCSI Extensions for RDMA (iSER) Target on Infiniband fabrics.
diff --git a/drivers/infiniband/ulp/srp/ib_srp.c b/drivers/infiniband/ulp/srp/ib_srp.c
index f93baf8254c4..a88631918e85 100644
--- a/drivers/infiniband/ulp/srp/ib_srp.c
+++ b/drivers/infiniband/ulp/srp/ib_srp.c
@@ -46,6 +46,7 @@
 #include <scsi/scsi.h>
 #include <scsi/scsi_device.h>
 #include <scsi/scsi_dbg.h>
+#include <scsi/scsi_tcq.h>
 #include <scsi/srp.h>
 #include <scsi/scsi_transport_srp.h>
 
@@ -86,6 +87,32 @@ module_param(topspin_workarounds, int, 0444);
 MODULE_PARM_DESC(topspin_workarounds,
 		 "Enable workarounds for Topspin/Cisco SRP target bugs if != 0");
 
+static struct kernel_param_ops srp_tmo_ops;
+
+static int srp_reconnect_delay = 10;
+module_param_cb(reconnect_delay, &srp_tmo_ops, &srp_reconnect_delay,
+		S_IRUGO | S_IWUSR);
+MODULE_PARM_DESC(reconnect_delay, "Time between successive reconnect attempts");
+
+static int srp_fast_io_fail_tmo = 15;
+module_param_cb(fast_io_fail_tmo, &srp_tmo_ops, &srp_fast_io_fail_tmo,
+		S_IRUGO | S_IWUSR);
+MODULE_PARM_DESC(fast_io_fail_tmo,
+		 "Number of seconds between the observation of a transport"
+		 " layer error and failing all I/O. \"off\" means that this"
+		 " functionality is disabled.");
+
+static int srp_dev_loss_tmo = 600;
+module_param_cb(dev_loss_tmo, &srp_tmo_ops, &srp_dev_loss_tmo,
+		S_IRUGO | S_IWUSR);
+MODULE_PARM_DESC(dev_loss_tmo,
+		 "Maximum number of seconds that the SRP transport should"
+		 " insulate transport layer errors. After this time has been"
+		 " exceeded the SCSI host is removed. Should be"
+		 " between 1 and " __stringify(SCSI_DEVICE_BLOCK_MAX_TIMEOUT)
+		 " if fast_io_fail_tmo has not been set. \"off\" means that"
+		 " this functionality is disabled.");
+
 static void srp_add_one(struct ib_device *device);
 static void srp_remove_one(struct ib_device *device);
 static void srp_recv_completion(struct ib_cq *cq, void *target_ptr);
@@ -102,6 +129,48 @@ static struct ib_client srp_client = {
 
 static struct ib_sa_client srp_sa_client;
 
+static int srp_tmo_get(char *buffer, const struct kernel_param *kp)
+{
+	int tmo = *(int *)kp->arg;
+
+	if (tmo >= 0)
+		return sprintf(buffer, "%d", tmo);
+	else
+		return sprintf(buffer, "off");
+}
+
+static int srp_tmo_set(const char *val, const struct kernel_param *kp)
+{
+	int tmo, res;
+
+	if (strncmp(val, "off", 3) != 0) {
+		res = kstrtoint(val, 0, &tmo);
+		if (res)
+			goto out;
+	} else {
+		tmo = -1;
+	}
+	if (kp->arg == &srp_reconnect_delay)
+		res = srp_tmo_valid(tmo, srp_fast_io_fail_tmo,
+				    srp_dev_loss_tmo);
+	else if (kp->arg == &srp_fast_io_fail_tmo)
+		res = srp_tmo_valid(srp_reconnect_delay, tmo, srp_dev_loss_tmo);
+	else
+		res = srp_tmo_valid(srp_reconnect_delay, srp_fast_io_fail_tmo,
+				    tmo);
+	if (res)
+		goto out;
+	*(int *)kp->arg = tmo;
+
+out:
+	return res;
+}
+
+static struct kernel_param_ops srp_tmo_ops = {
+	.get = srp_tmo_get,
+	.set = srp_tmo_set,
+};
+
 static inline struct srp_target_port *host_to_target(struct Scsi_Host *host)
 {
 	return (struct srp_target_port *) host->hostdata;
@@ -231,16 +300,16 @@ static int srp_create_target_ib(struct srp_target_port *target)
 		return -ENOMEM;
 
 	recv_cq = ib_create_cq(target->srp_host->srp_dev->dev,
-			       srp_recv_completion, NULL, target, SRP_RQ_SIZE,
-			       target->comp_vector);
+			       srp_recv_completion, NULL, target,
+			       target->queue_size, target->comp_vector);
 	if (IS_ERR(recv_cq)) {
 		ret = PTR_ERR(recv_cq);
 		goto err;
 	}
 
 	send_cq = ib_create_cq(target->srp_host->srp_dev->dev,
-			       srp_send_completion, NULL, target, SRP_SQ_SIZE,
-			       target->comp_vector);
+			       srp_send_completion, NULL, target,
+			       target->queue_size, target->comp_vector);
 	if (IS_ERR(send_cq)) {
 		ret = PTR_ERR(send_cq);
 		goto err_recv_cq;
@@ -249,8 +318,8 @@ static int srp_create_target_ib(struct srp_target_port *target)
 	ib_req_notify_cq(recv_cq, IB_CQ_NEXT_COMP);
 
 	init_attr->event_handler       = srp_qp_event;
-	init_attr->cap.max_send_wr     = SRP_SQ_SIZE;
-	init_attr->cap.max_recv_wr     = SRP_RQ_SIZE;
+	init_attr->cap.max_send_wr     = target->queue_size;
+	init_attr->cap.max_recv_wr     = target->queue_size;
 	init_attr->cap.max_recv_sge    = 1;
 	init_attr->cap.max_send_sge    = 1;
 	init_attr->sq_sig_type         = IB_SIGNAL_ALL_WR;
@@ -296,6 +365,10 @@ err:
 	return ret;
 }
 
+/*
+ * Note: this function may be called without srp_alloc_iu_bufs() having been
+ * invoked. Hence the target->[rt]x_ring checks.
+ */
 static void srp_free_target_ib(struct srp_target_port *target)
 {
 	int i;
@@ -307,10 +380,18 @@ static void srp_free_target_ib(struct srp_target_port *target)
 	target->qp = NULL;
 	target->send_cq = target->recv_cq = NULL;
 
-	for (i = 0; i < SRP_RQ_SIZE; ++i)
-		srp_free_iu(target->srp_host, target->rx_ring[i]);
-	for (i = 0; i < SRP_SQ_SIZE; ++i)
-		srp_free_iu(target->srp_host, target->tx_ring[i]);
+	if (target->rx_ring) {
+		for (i = 0; i < target->queue_size; ++i)
+			srp_free_iu(target->srp_host, target->rx_ring[i]);
+		kfree(target->rx_ring);
+		target->rx_ring = NULL;
+	}
+	if (target->tx_ring) {
+		for (i = 0; i < target->queue_size; ++i)
+			srp_free_iu(target->srp_host, target->tx_ring[i]);
+		kfree(target->tx_ring);
+		target->tx_ring = NULL;
+	}
 }
 
 static void srp_path_rec_completion(int status,
@@ -390,7 +471,7 @@ static int srp_send_req(struct srp_target_port *target)
 	req->param.responder_resources	      = 4;
 	req->param.remote_cm_response_timeout = 20;
 	req->param.local_cm_response_timeout  = 20;
-	req->param.retry_count 		      = 7;
+	req->param.retry_count                = target->tl_retry_count;
 	req->param.rnr_retry_count 	      = 7;
 	req->param.max_cm_retries 	      = 15;
 
@@ -496,7 +577,11 @@ static void srp_free_req_data(struct srp_target_port *target)
 	struct srp_request *req;
 	int i;
 
-	for (i = 0, req = target->req_ring; i < SRP_CMD_SQ_SIZE; ++i, ++req) {
+	if (!target->req_ring)
+		return;
+
+	for (i = 0; i < target->req_ring_size; ++i) {
+		req = &target->req_ring[i];
 		kfree(req->fmr_list);
 		kfree(req->map_page);
 		if (req->indirect_dma_addr) {
@@ -506,6 +591,50 @@ static void srp_free_req_data(struct srp_target_port *target)
 		}
 		kfree(req->indirect_desc);
 	}
+
+	kfree(target->req_ring);
+	target->req_ring = NULL;
+}
+
+static int srp_alloc_req_data(struct srp_target_port *target)
+{
+	struct srp_device *srp_dev = target->srp_host->srp_dev;
+	struct ib_device *ibdev = srp_dev->dev;
+	struct srp_request *req;
+	dma_addr_t dma_addr;
+	int i, ret = -ENOMEM;
+
+	INIT_LIST_HEAD(&target->free_reqs);
+
+	target->req_ring = kzalloc(target->req_ring_size *
+				   sizeof(*target->req_ring), GFP_KERNEL);
+	if (!target->req_ring)
+		goto out;
+
+	for (i = 0; i < target->req_ring_size; ++i) {
+		req = &target->req_ring[i];
+		req->fmr_list = kmalloc(target->cmd_sg_cnt * sizeof(void *),
+					GFP_KERNEL);
+		req->map_page = kmalloc(SRP_FMR_SIZE * sizeof(void *),
+					GFP_KERNEL);
+		req->indirect_desc = kmalloc(target->indirect_size, GFP_KERNEL);
+		if (!req->fmr_list || !req->map_page || !req->indirect_desc)
+			goto out;
+
+		dma_addr = ib_dma_map_single(ibdev, req->indirect_desc,
+					     target->indirect_size,
+					     DMA_TO_DEVICE);
+		if (ib_dma_mapping_error(ibdev, dma_addr))
+			goto out;
+
+		req->indirect_dma_addr = dma_addr;
+		req->index = i;
+		list_add_tail(&req->list, &target->free_reqs);
+	}
+	ret = 0;
+
+out:
+	return ret;
 }
 
 /**
@@ -528,12 +657,20 @@ static void srp_remove_target(struct srp_target_port *target)
 	WARN_ON_ONCE(target->state != SRP_TARGET_REMOVED);
 
 	srp_del_scsi_host_attr(target->scsi_host);
+	srp_rport_get(target->rport);
 	srp_remove_host(target->scsi_host);
 	scsi_remove_host(target->scsi_host);
 	srp_disconnect_target(target);
 	ib_destroy_cm_id(target->cm_id);
 	srp_free_target_ib(target);
+	cancel_work_sync(&target->tl_err_work);
+	srp_rport_put(target->rport);
 	srp_free_req_data(target);
+
+	spin_lock(&target->srp_host->target_lock);
+	list_del(&target->list);
+	spin_unlock(&target->srp_host->target_lock);
+
 	scsi_host_put(target->scsi_host);
 }
 
@@ -545,10 +682,6 @@ static void srp_remove_work(struct work_struct *work)
 	WARN_ON_ONCE(target->state != SRP_TARGET_REMOVED);
 
 	srp_remove_target(target);
-
-	spin_lock(&target->srp_host->target_lock);
-	list_del(&target->list);
-	spin_unlock(&target->srp_host->target_lock);
 }
 
 static void srp_rport_delete(struct srp_rport *rport)
@@ -686,23 +819,42 @@ static void srp_free_req(struct srp_target_port *target,
 	spin_unlock_irqrestore(&target->lock, flags);
 }
 
-static void srp_reset_req(struct srp_target_port *target, struct srp_request *req)
+static void srp_finish_req(struct srp_target_port *target,
+			   struct srp_request *req, int result)
 {
 	struct scsi_cmnd *scmnd = srp_claim_req(target, req, NULL);
 
 	if (scmnd) {
 		srp_free_req(target, req, scmnd, 0);
-		scmnd->result = DID_RESET << 16;
+		scmnd->result = result;
 		scmnd->scsi_done(scmnd);
 	}
 }
 
-static int srp_reconnect_target(struct srp_target_port *target)
+static void srp_terminate_io(struct srp_rport *rport)
 {
-	struct Scsi_Host *shost = target->scsi_host;
-	int i, ret;
+	struct srp_target_port *target = rport->lld_data;
+	int i;
 
-	scsi_target_block(&shost->shost_gendev);
+	for (i = 0; i < target->req_ring_size; ++i) {
+		struct srp_request *req = &target->req_ring[i];
+		srp_finish_req(target, req, DID_TRANSPORT_FAILFAST << 16);
+	}
+}
+
+/*
+ * It is up to the caller to ensure that srp_rport_reconnect() calls are
+ * serialized and that no concurrent srp_queuecommand(), srp_abort(),
+ * srp_reset_device() or srp_reset_host() calls will occur while this function
+ * is in progress. One way to realize that is not to call this function
+ * directly but to call srp_reconnect_rport() instead since that last function
+ * serializes calls of this function via rport->mutex and also blocks
+ * srp_queuecommand() calls before invoking this function.
+ */
+static int srp_rport_reconnect(struct srp_rport *rport)
+{
+	struct srp_target_port *target = rport->lld_data;
+	int i, ret;
 
 	srp_disconnect_target(target);
 	/*
@@ -721,41 +873,21 @@ static int srp_reconnect_target(struct srp_target_port *target)
 	else
 		srp_create_target_ib(target);
 
-	for (i = 0; i < SRP_CMD_SQ_SIZE; ++i) {
+	for (i = 0; i < target->req_ring_size; ++i) {
 		struct srp_request *req = &target->req_ring[i];
-		if (req->scmnd)
-			srp_reset_req(target, req);
+		srp_finish_req(target, req, DID_RESET << 16);
 	}
 
 	INIT_LIST_HEAD(&target->free_tx);
-	for (i = 0; i < SRP_SQ_SIZE; ++i)
+	for (i = 0; i < target->queue_size; ++i)
 		list_add(&target->tx_ring[i]->list, &target->free_tx);
 
 	if (ret == 0)
 		ret = srp_connect_target(target);
 
-	scsi_target_unblock(&shost->shost_gendev, ret == 0 ? SDEV_RUNNING :
-			    SDEV_TRANSPORT_OFFLINE);
-	target->transport_offline = !!ret;
-
-	if (ret)
-		goto err;
-
-	shost_printk(KERN_INFO, target->scsi_host, PFX "reconnect succeeded\n");
-
-	return ret;
-
-err:
-	shost_printk(KERN_ERR, target->scsi_host,
-		     PFX "reconnect failed (%d), removing target port.\n", ret);
-
-	/*
-	 * We couldn't reconnect, so kill our target port off.
-	 * However, we have to defer the real removal because we
-	 * are in the context of the SCSI error handler now, which
-	 * will deadlock if we call scsi_remove_host().
-	 */
-	srp_queue_remove_work(target);
+	if (ret == 0)
+		shost_printk(KERN_INFO, target->scsi_host,
+			     PFX "reconnect succeeded\n");
 
 	return ret;
 }
@@ -1302,15 +1434,30 @@ static void srp_handle_recv(struct srp_target_port *target, struct ib_wc *wc)
 			     PFX "Recv failed with error code %d\n", res);
 }
 
-static void srp_handle_qp_err(enum ib_wc_status wc_status,
-			      enum ib_wc_opcode wc_opcode,
+/**
+ * srp_tl_err_work() - handle a transport layer error
+ *
+ * Note: This function may get invoked before the rport has been created,
+ * hence the target->rport test.
+ */
+static void srp_tl_err_work(struct work_struct *work)
+{
+	struct srp_target_port *target;
+
+	target = container_of(work, struct srp_target_port, tl_err_work);
+	if (target->rport)
+		srp_start_tl_fail_timers(target->rport);
+}
+
+static void srp_handle_qp_err(enum ib_wc_status wc_status, bool send_err,
 			      struct srp_target_port *target)
 {
 	if (target->connected && !target->qp_in_error) {
 		shost_printk(KERN_ERR, target->scsi_host,
 			     PFX "failed %s status %d\n",
-			     wc_opcode & IB_WC_RECV ? "receive" : "send",
+			     send_err ? "send" : "receive",
 			     wc_status);
+		queue_work(system_long_wq, &target->tl_err_work);
 	}
 	target->qp_in_error = true;
 }
@@ -1325,7 +1472,7 @@ static void srp_recv_completion(struct ib_cq *cq, void *target_ptr)
 		if (likely(wc.status == IB_WC_SUCCESS)) {
 			srp_handle_recv(target, &wc);
 		} else {
-			srp_handle_qp_err(wc.status, wc.opcode, target);
+			srp_handle_qp_err(wc.status, false, target);
 		}
 	}
 }
@@ -1341,7 +1488,7 @@ static void srp_send_completion(struct ib_cq *cq, void *target_ptr)
 			iu = (struct srp_iu *) (uintptr_t) wc.wr_id;
 			list_add(&iu->list, &target->free_tx);
 		} else {
-			srp_handle_qp_err(wc.status, wc.opcode, target);
+			srp_handle_qp_err(wc.status, true, target);
 		}
 	}
 }
@@ -1349,17 +1496,29 @@ static void srp_send_completion(struct ib_cq *cq, void *target_ptr)
 static int srp_queuecommand(struct Scsi_Host *shost, struct scsi_cmnd *scmnd)
 {
 	struct srp_target_port *target = host_to_target(shost);
+	struct srp_rport *rport = target->rport;
 	struct srp_request *req;
 	struct srp_iu *iu;
 	struct srp_cmd *cmd;
 	struct ib_device *dev;
 	unsigned long flags;
-	int len;
+	int len, result;
+	const bool in_scsi_eh = !in_interrupt() && current == shost->ehandler;
+
+	/*
+	 * The SCSI EH thread is the only context from which srp_queuecommand()
+	 * can get invoked for blocked devices (SDEV_BLOCK /
+	 * SDEV_CREATED_BLOCK). Avoid racing with srp_reconnect_rport() by
+	 * locking the rport mutex if invoked from inside the SCSI EH.
+	 */
+	if (in_scsi_eh)
+		mutex_lock(&rport->mutex);
 
-	if (unlikely(target->transport_offline)) {
-		scmnd->result = DID_NO_CONNECT << 16;
+	result = srp_chkready(target->rport);
+	if (unlikely(result)) {
+		scmnd->result = result;
 		scmnd->scsi_done(scmnd);
-		return 0;
+		goto unlock_rport;
 	}
 
 	spin_lock_irqsave(&target->lock, flags);
@@ -1404,6 +1563,10 @@ static int srp_queuecommand(struct Scsi_Host *shost, struct scsi_cmnd *scmnd)
 		goto err_unmap;
 	}
 
+unlock_rport:
+	if (in_scsi_eh)
+		mutex_unlock(&rport->mutex);
+
 	return 0;
 
 err_unmap:
@@ -1418,14 +1581,30 @@ err_iu:
 err_unlock:
 	spin_unlock_irqrestore(&target->lock, flags);
 
+	if (in_scsi_eh)
+		mutex_unlock(&rport->mutex);
+
 	return SCSI_MLQUEUE_HOST_BUSY;
 }
 
+/*
+ * Note: the resources allocated in this function are freed in
+ * srp_free_target_ib().
+ */
 static int srp_alloc_iu_bufs(struct srp_target_port *target)
 {
 	int i;
 
-	for (i = 0; i < SRP_RQ_SIZE; ++i) {
+	target->rx_ring = kzalloc(target->queue_size * sizeof(*target->rx_ring),
+				  GFP_KERNEL);
+	if (!target->rx_ring)
+		goto err_no_ring;
+	target->tx_ring = kzalloc(target->queue_size * sizeof(*target->tx_ring),
+				  GFP_KERNEL);
+	if (!target->tx_ring)
+		goto err_no_ring;
+
+	for (i = 0; i < target->queue_size; ++i) {
 		target->rx_ring[i] = srp_alloc_iu(target->srp_host,
 						  target->max_ti_iu_len,
 						  GFP_KERNEL, DMA_FROM_DEVICE);
@@ -1433,7 +1612,7 @@ static int srp_alloc_iu_bufs(struct srp_target_port *target)
 			goto err;
 	}
 
-	for (i = 0; i < SRP_SQ_SIZE; ++i) {
+	for (i = 0; i < target->queue_size; ++i) {
 		target->tx_ring[i] = srp_alloc_iu(target->srp_host,
 						  target->max_iu_len,
 						  GFP_KERNEL, DMA_TO_DEVICE);
@@ -1446,16 +1625,18 @@ static int srp_alloc_iu_bufs(struct srp_target_port *target)
 	return 0;
 
 err:
-	for (i = 0; i < SRP_RQ_SIZE; ++i) {
+	for (i = 0; i < target->queue_size; ++i) {
 		srp_free_iu(target->srp_host, target->rx_ring[i]);
-		target->rx_ring[i] = NULL;
-	}
-
-	for (i = 0; i < SRP_SQ_SIZE; ++i) {
 		srp_free_iu(target->srp_host, target->tx_ring[i]);
-		target->tx_ring[i] = NULL;
 	}
 
+
+err_no_ring:
+	kfree(target->tx_ring);
+	target->tx_ring = NULL;
+	kfree(target->rx_ring);
+	target->rx_ring = NULL;
+
 	return -ENOMEM;
 }
 
@@ -1506,6 +1687,9 @@ static void srp_cm_rep_handler(struct ib_cm_id *cm_id,
 		target->scsi_host->can_queue
 			= min(target->req_lim - SRP_TSK_MGMT_SQ_SIZE,
 			      target->scsi_host->can_queue);
+		target->scsi_host->cmd_per_lun
+			= min_t(int, target->scsi_host->can_queue,
+				target->scsi_host->cmd_per_lun);
 	} else {
 		shost_printk(KERN_WARNING, target->scsi_host,
 			     PFX "Unhandled RSP opcode %#x\n", lrsp->opcode);
@@ -1513,7 +1697,7 @@ static void srp_cm_rep_handler(struct ib_cm_id *cm_id,
 		goto error;
 	}
 
-	if (!target->rx_ring[0]) {
+	if (!target->rx_ring) {
 		ret = srp_alloc_iu_bufs(target);
 		if (ret)
 			goto error;
@@ -1533,7 +1717,7 @@ static void srp_cm_rep_handler(struct ib_cm_id *cm_id,
 	if (ret)
 		goto error_free;
 
-	for (i = 0; i < SRP_RQ_SIZE; i++) {
+	for (i = 0; i < target->queue_size; i++) {
 		struct srp_iu *iu = target->rx_ring[i];
 		ret = srp_post_recv(target, iu);
 		if (ret)
@@ -1672,6 +1856,7 @@ static int srp_cm_handler(struct ib_cm_id *cm_id, struct ib_cm_event *event)
 		if (ib_send_cm_drep(cm_id, NULL, 0))
 			shost_printk(KERN_ERR, target->scsi_host,
 				     PFX "Sending CM DREP failed\n");
+		queue_work(system_long_wq, &target->tl_err_work);
 		break;
 
 	case IB_CM_TIMEWAIT_EXIT:
@@ -1698,9 +1883,61 @@ static int srp_cm_handler(struct ib_cm_id *cm_id, struct ib_cm_event *event)
 	return 0;
 }
 
+/**
+ * srp_change_queue_type - changing device queue tag type
+ * @sdev: scsi device struct
+ * @tag_type: requested tag type
+ *
+ * Returns queue tag type.
+ */
+static int
+srp_change_queue_type(struct scsi_device *sdev, int tag_type)
+{
+	if (sdev->tagged_supported) {
+		scsi_set_tag_type(sdev, tag_type);
+		if (tag_type)
+			scsi_activate_tcq(sdev, sdev->queue_depth);
+		else
+			scsi_deactivate_tcq(sdev, sdev->queue_depth);
+	} else
+		tag_type = 0;
+
+	return tag_type;
+}
+
+/**
+ * srp_change_queue_depth - setting device queue depth
+ * @sdev: scsi device struct
+ * @qdepth: requested queue depth
+ * @reason: SCSI_QDEPTH_DEFAULT/SCSI_QDEPTH_QFULL/SCSI_QDEPTH_RAMP_UP
+ * (see include/scsi/scsi_host.h for definition)
+ *
+ * Returns queue depth.
+ */
+static int
+srp_change_queue_depth(struct scsi_device *sdev, int qdepth, int reason)
+{
+	struct Scsi_Host *shost = sdev->host;
+	int max_depth;
+	if (reason == SCSI_QDEPTH_DEFAULT || reason == SCSI_QDEPTH_RAMP_UP) {
+		max_depth = shost->can_queue;
+		if (!sdev->tagged_supported)
+			max_depth = 1;
+		if (qdepth > max_depth)
+			qdepth = max_depth;
+		scsi_adjust_queue_depth(sdev, scsi_get_tag_type(sdev), qdepth);
+	} else if (reason == SCSI_QDEPTH_QFULL)
+		scsi_track_queue_full(sdev, qdepth);
+	else
+		return -EOPNOTSUPP;
+
+	return sdev->queue_depth;
+}
+
 static int srp_send_tsk_mgmt(struct srp_target_port *target,
 			     u64 req_tag, unsigned int lun, u8 func)
 {
+	struct srp_rport *rport = target->rport;
 	struct ib_device *dev = target->srp_host->srp_dev->dev;
 	struct srp_iu *iu;
 	struct srp_tsk_mgmt *tsk_mgmt;
@@ -1710,12 +1947,20 @@ static int srp_send_tsk_mgmt(struct srp_target_port *target,
 
 	init_completion(&target->tsk_mgmt_done);
 
+	/*
+	 * Lock the rport mutex to avoid that srp_create_target_ib() is
+	 * invoked while a task management function is being sent.
+	 */
+	mutex_lock(&rport->mutex);
 	spin_lock_irq(&target->lock);
 	iu = __srp_get_tx_iu(target, SRP_IU_TSK_MGMT);
 	spin_unlock_irq(&target->lock);
 
-	if (!iu)
+	if (!iu) {
+		mutex_unlock(&rport->mutex);
+
 		return -1;
+	}
 
 	ib_dma_sync_single_for_cpu(dev, iu->dma, sizeof *tsk_mgmt,
 				   DMA_TO_DEVICE);
@@ -1732,8 +1977,11 @@ static int srp_send_tsk_mgmt(struct srp_target_port *target,
 				      DMA_TO_DEVICE);
 	if (srp_post_send(target, iu, sizeof *tsk_mgmt)) {
 		srp_put_tx_iu(target, iu, SRP_IU_TSK_MGMT);
+		mutex_unlock(&rport->mutex);
+
 		return -1;
 	}
+	mutex_unlock(&rport->mutex);
 
 	if (!wait_for_completion_timeout(&target->tsk_mgmt_done,
 					 msecs_to_jiffies(SRP_ABORT_TIMEOUT_MS)))
@@ -1751,11 +1999,11 @@ static int srp_abort(struct scsi_cmnd *scmnd)
 	shost_printk(KERN_ERR, target->scsi_host, "SRP abort called\n");
 
 	if (!req || !srp_claim_req(target, req, scmnd))
-		return FAILED;
+		return SUCCESS;
 	if (srp_send_tsk_mgmt(target, req->index, scmnd->device->lun,
 			      SRP_TSK_ABORT_TASK) == 0)
 		ret = SUCCESS;
-	else if (target->transport_offline)
+	else if (target->rport->state == SRP_RPORT_LOST)
 		ret = FAST_IO_FAIL;
 	else
 		ret = FAILED;
@@ -1779,10 +2027,10 @@ static int srp_reset_device(struct scsi_cmnd *scmnd)
 	if (target->tsk_mgmt_status)
 		return FAILED;
 
-	for (i = 0; i < SRP_CMD_SQ_SIZE; ++i) {
+	for (i = 0; i < target->req_ring_size; ++i) {
 		struct srp_request *req = &target->req_ring[i];
 		if (req->scmnd && req->scmnd->device == scmnd->device)
-			srp_reset_req(target, req);
+			srp_finish_req(target, req, DID_RESET << 16);
 	}
 
 	return SUCCESS;
@@ -1791,14 +2039,10 @@ static int srp_reset_device(struct scsi_cmnd *scmnd)
 static int srp_reset_host(struct scsi_cmnd *scmnd)
 {
 	struct srp_target_port *target = host_to_target(scmnd->device->host);
-	int ret = FAILED;
 
 	shost_printk(KERN_ERR, target->scsi_host, PFX "SRP reset_host called\n");
 
-	if (!srp_reconnect_target(target))
-		ret = SUCCESS;
-
-	return ret;
+	return srp_reconnect_rport(target->rport) == 0 ? SUCCESS : FAILED;
 }
 
 static int srp_slave_configure(struct scsi_device *sdev)
@@ -1851,6 +2095,14 @@ static ssize_t show_pkey(struct device *dev, struct device_attribute *attr,
 	return sprintf(buf, "0x%04x\n", be16_to_cpu(target->path.pkey));
 }
 
+static ssize_t show_sgid(struct device *dev, struct device_attribute *attr,
+			 char *buf)
+{
+	struct srp_target_port *target = host_to_target(class_to_shost(dev));
+
+	return sprintf(buf, "%pI6\n", target->path.sgid.raw);
+}
+
 static ssize_t show_dgid(struct device *dev, struct device_attribute *attr,
 			 char *buf)
 {
@@ -1907,6 +2159,14 @@ static ssize_t show_comp_vector(struct device *dev,
 	return sprintf(buf, "%d\n", target->comp_vector);
 }
 
+static ssize_t show_tl_retry_count(struct device *dev,
+				   struct device_attribute *attr, char *buf)
+{
+	struct srp_target_port *target = host_to_target(class_to_shost(dev));
+
+	return sprintf(buf, "%d\n", target->tl_retry_count);
+}
+
 static ssize_t show_cmd_sg_entries(struct device *dev,
 				   struct device_attribute *attr, char *buf)
 {
@@ -1927,6 +2187,7 @@ static DEVICE_ATTR(id_ext,	    S_IRUGO, show_id_ext,	   NULL);
 static DEVICE_ATTR(ioc_guid,	    S_IRUGO, show_ioc_guid,	   NULL);
 static DEVICE_ATTR(service_id,	    S_IRUGO, show_service_id,	   NULL);
 static DEVICE_ATTR(pkey,	    S_IRUGO, show_pkey,		   NULL);
+static DEVICE_ATTR(sgid,	    S_IRUGO, show_sgid,		   NULL);
 static DEVICE_ATTR(dgid,	    S_IRUGO, show_dgid,		   NULL);
 static DEVICE_ATTR(orig_dgid,	    S_IRUGO, show_orig_dgid,	   NULL);
 static DEVICE_ATTR(req_lim,         S_IRUGO, show_req_lim,         NULL);
@@ -1934,6 +2195,7 @@ static DEVICE_ATTR(zero_req_lim,    S_IRUGO, show_zero_req_lim,	   NULL);
 static DEVICE_ATTR(local_ib_port,   S_IRUGO, show_local_ib_port,   NULL);
 static DEVICE_ATTR(local_ib_device, S_IRUGO, show_local_ib_device, NULL);
 static DEVICE_ATTR(comp_vector,     S_IRUGO, show_comp_vector,     NULL);
+static DEVICE_ATTR(tl_retry_count,  S_IRUGO, show_tl_retry_count,  NULL);
 static DEVICE_ATTR(cmd_sg_entries,  S_IRUGO, show_cmd_sg_entries,  NULL);
 static DEVICE_ATTR(allow_ext_sg,    S_IRUGO, show_allow_ext_sg,    NULL);
 
@@ -1942,6 +2204,7 @@ static struct device_attribute *srp_host_attrs[] = {
 	&dev_attr_ioc_guid,
 	&dev_attr_service_id,
 	&dev_attr_pkey,
+	&dev_attr_sgid,
 	&dev_attr_dgid,
 	&dev_attr_orig_dgid,
 	&dev_attr_req_lim,
@@ -1949,6 +2212,7 @@ static struct device_attribute *srp_host_attrs[] = {
 	&dev_attr_local_ib_port,
 	&dev_attr_local_ib_device,
 	&dev_attr_comp_vector,
+	&dev_attr_tl_retry_count,
 	&dev_attr_cmd_sg_entries,
 	&dev_attr_allow_ext_sg,
 	NULL
@@ -1961,14 +2225,16 @@ static struct scsi_host_template srp_template = {
 	.slave_configure		= srp_slave_configure,
 	.info				= srp_target_info,
 	.queuecommand			= srp_queuecommand,
+	.change_queue_depth             = srp_change_queue_depth,
+	.change_queue_type              = srp_change_queue_type,
 	.eh_abort_handler		= srp_abort,
 	.eh_device_reset_handler	= srp_reset_device,
 	.eh_host_reset_handler		= srp_reset_host,
 	.skip_settle_delay		= true,
 	.sg_tablesize			= SRP_DEF_SG_TABLESIZE,
-	.can_queue			= SRP_CMD_SQ_SIZE,
+	.can_queue			= SRP_DEFAULT_CMD_SQ_SIZE,
 	.this_id			= -1,
-	.cmd_per_lun			= SRP_CMD_SQ_SIZE,
+	.cmd_per_lun			= SRP_DEFAULT_CMD_SQ_SIZE,
 	.use_clustering			= ENABLE_CLUSTERING,
 	.shost_attrs			= srp_host_attrs
 };
@@ -1994,6 +2260,7 @@ static int srp_add_target(struct srp_host *host, struct srp_target_port *target)
 	}
 
 	rport->lld_data = target;
+	target->rport = rport;
 
 	spin_lock(&host->target_lock);
 	list_add_tail(&target->list, &host->target_list);
@@ -2073,6 +2340,8 @@ enum {
 	SRP_OPT_ALLOW_EXT_SG	= 1 << 10,
 	SRP_OPT_SG_TABLESIZE	= 1 << 11,
 	SRP_OPT_COMP_VECTOR	= 1 << 12,
+	SRP_OPT_TL_RETRY_COUNT	= 1 << 13,
+	SRP_OPT_QUEUE_SIZE	= 1 << 14,
 	SRP_OPT_ALL		= (SRP_OPT_ID_EXT	|
 				   SRP_OPT_IOC_GUID	|
 				   SRP_OPT_DGID		|
@@ -2094,6 +2363,8 @@ static const match_table_t srp_opt_tokens = {
 	{ SRP_OPT_ALLOW_EXT_SG,		"allow_ext_sg=%u"	},
 	{ SRP_OPT_SG_TABLESIZE,		"sg_tablesize=%u"	},
 	{ SRP_OPT_COMP_VECTOR,		"comp_vector=%u"	},
+	{ SRP_OPT_TL_RETRY_COUNT,	"tl_retry_count=%u"	},
+	{ SRP_OPT_QUEUE_SIZE,		"queue_size=%d"		},
 	{ SRP_OPT_ERR,			NULL 			}
 };
 
@@ -2188,13 +2459,25 @@ static int srp_parse_options(const char *buf, struct srp_target_port *target)
 			target->scsi_host->max_sectors = token;
 			break;
 
+		case SRP_OPT_QUEUE_SIZE:
+			if (match_int(args, &token) || token < 1) {
+				pr_warn("bad queue_size parameter '%s'\n", p);
+				goto out;
+			}
+			target->scsi_host->can_queue = token;
+			target->queue_size = token + SRP_RSP_SQ_SIZE +
+					     SRP_TSK_MGMT_SQ_SIZE;
+			if (!(opt_mask & SRP_OPT_MAX_CMD_PER_LUN))
+				target->scsi_host->cmd_per_lun = token;
+			break;
+
 		case SRP_OPT_MAX_CMD_PER_LUN:
-			if (match_int(args, &token)) {
+			if (match_int(args, &token) || token < 1) {
 				pr_warn("bad max cmd_per_lun parameter '%s'\n",
 					p);
 				goto out;
 			}
-			target->scsi_host->cmd_per_lun = min(token, SRP_CMD_SQ_SIZE);
+			target->scsi_host->cmd_per_lun = token;
 			break;
 
 		case SRP_OPT_IO_CLASS:
@@ -2257,6 +2540,15 @@ static int srp_parse_options(const char *buf, struct srp_target_port *target)
 			target->comp_vector = token;
 			break;
 
+		case SRP_OPT_TL_RETRY_COUNT:
+			if (match_int(args, &token) || token < 2 || token > 7) {
+				pr_warn("bad tl_retry_count parameter '%s' (must be a number between 2 and 7)\n",
+					p);
+				goto out;
+			}
+			target->tl_retry_count = token;
+			break;
+
 		default:
 			pr_warn("unknown parameter or missing value '%s' in target creation request\n",
 				p);
@@ -2273,6 +2565,12 @@ static int srp_parse_options(const char *buf, struct srp_target_port *target)
 				pr_warn("target creation request is missing parameter '%s'\n",
 					srp_opt_tokens[i].pattern);
 
+	if (target->scsi_host->cmd_per_lun > target->scsi_host->can_queue
+	    && (opt_mask & SRP_OPT_MAX_CMD_PER_LUN))
+		pr_warn("cmd_per_lun = %d > queue_size = %d\n",
+			target->scsi_host->cmd_per_lun,
+			target->scsi_host->can_queue);
+
 out:
 	kfree(options);
 	return ret;
@@ -2287,8 +2585,7 @@ static ssize_t srp_create_target(struct device *dev,
 	struct Scsi_Host *target_host;
 	struct srp_target_port *target;
 	struct ib_device *ibdev = host->srp_dev->dev;
-	dma_addr_t dma_addr;
-	int i, ret;
+	int ret;
 
 	target_host = scsi_host_alloc(&srp_template,
 				      sizeof (struct srp_target_port));
@@ -2311,11 +2608,15 @@ static ssize_t srp_create_target(struct device *dev,
 	target->cmd_sg_cnt	= cmd_sg_entries;
 	target->sg_tablesize	= indirect_sg_entries ? : cmd_sg_entries;
 	target->allow_ext_sg	= allow_ext_sg;
+	target->tl_retry_count	= 7;
+	target->queue_size	= SRP_DEFAULT_QUEUE_SIZE;
 
 	ret = srp_parse_options(buf, target);
 	if (ret)
 		goto err;
 
+	target->req_ring_size = target->queue_size - SRP_TSK_MGMT_SQ_SIZE;
+
 	if (!srp_conn_unique(target->srp_host, target)) {
 		shost_printk(KERN_INFO, target->scsi_host,
 			     PFX "Already connected to target port with id_ext=%016llx;ioc_guid=%016llx;initiator_ext=%016llx\n",
@@ -2339,31 +2640,13 @@ static ssize_t srp_create_target(struct device *dev,
 			     sizeof (struct srp_indirect_buf) +
 			     target->cmd_sg_cnt * sizeof (struct srp_direct_buf);
 
+	INIT_WORK(&target->tl_err_work, srp_tl_err_work);
 	INIT_WORK(&target->remove_work, srp_remove_work);
 	spin_lock_init(&target->lock);
 	INIT_LIST_HEAD(&target->free_tx);
-	INIT_LIST_HEAD(&target->free_reqs);
-	for (i = 0; i < SRP_CMD_SQ_SIZE; ++i) {
-		struct srp_request *req = &target->req_ring[i];
-
-		req->fmr_list = kmalloc(target->cmd_sg_cnt * sizeof (void *),
-					GFP_KERNEL);
-		req->map_page = kmalloc(SRP_FMR_SIZE * sizeof (void *),
-					GFP_KERNEL);
-		req->indirect_desc = kmalloc(target->indirect_size, GFP_KERNEL);
-		if (!req->fmr_list || !req->map_page || !req->indirect_desc)
-			goto err_free_mem;
-
-		dma_addr = ib_dma_map_single(ibdev, req->indirect_desc,
-					     target->indirect_size,
-					     DMA_TO_DEVICE);
-		if (ib_dma_mapping_error(ibdev, dma_addr))
-			goto err_free_mem;
-
-		req->indirect_dma_addr = dma_addr;
-		req->index = i;
-		list_add_tail(&req->list, &target->free_reqs);
-	}
+	ret = srp_alloc_req_data(target);
+	if (ret)
+		goto err_free_mem;
 
 	ib_query_gid(ibdev, host->port, 0, &target->path.sgid);
 
@@ -2612,7 +2895,14 @@ static void srp_remove_one(struct ib_device *device)
 }
 
 static struct srp_function_template ib_srp_transport_functions = {
+	.has_rport_state	 = true,
+	.reset_timer_if_blocked	 = true,
+	.reconnect_delay	 = &srp_reconnect_delay,
+	.fast_io_fail_tmo	 = &srp_fast_io_fail_tmo,
+	.dev_loss_tmo		 = &srp_dev_loss_tmo,
+	.reconnect		 = srp_rport_reconnect,
 	.rport_delete		 = srp_rport_delete,
+	.terminate_rport_io	 = srp_terminate_io,
 };
 
 static int __init srp_init_module(void)
diff --git a/drivers/infiniband/ulp/srp/ib_srp.h b/drivers/infiniband/ulp/srp/ib_srp.h
index e641088c14dc..575681063f38 100644
--- a/drivers/infiniband/ulp/srp/ib_srp.h
+++ b/drivers/infiniband/ulp/srp/ib_srp.h
@@ -57,14 +57,11 @@ enum {
 	SRP_MAX_LUN		= 512,
 	SRP_DEF_SG_TABLESIZE	= 12,
 
-	SRP_RQ_SHIFT    	= 6,
-	SRP_RQ_SIZE		= 1 << SRP_RQ_SHIFT,
-
-	SRP_SQ_SIZE		= SRP_RQ_SIZE,
+	SRP_DEFAULT_QUEUE_SIZE	= 1 << 6,
 	SRP_RSP_SQ_SIZE		= 1,
-	SRP_REQ_SQ_SIZE		= SRP_SQ_SIZE - SRP_RSP_SQ_SIZE,
 	SRP_TSK_MGMT_SQ_SIZE	= 1,
-	SRP_CMD_SQ_SIZE		= SRP_REQ_SQ_SIZE - SRP_TSK_MGMT_SQ_SIZE,
+	SRP_DEFAULT_CMD_SQ_SIZE = SRP_DEFAULT_QUEUE_SIZE - SRP_RSP_SQ_SIZE -
+				  SRP_TSK_MGMT_SQ_SIZE,
 
 	SRP_TAG_NO_REQ		= ~0U,
 	SRP_TAG_TSK_MGMT	= 1U << 31,
@@ -140,7 +137,6 @@ struct srp_target_port {
 	unsigned int		cmd_sg_cnt;
 	unsigned int		indirect_size;
 	bool			allow_ext_sg;
-	bool			transport_offline;
 
 	/* Everything above this point is used in the hot path of
 	 * command processing. Try to keep them packed into cachelines.
@@ -153,10 +149,14 @@ struct srp_target_port {
 	u16			io_class;
 	struct srp_host	       *srp_host;
 	struct Scsi_Host       *scsi_host;
+	struct srp_rport       *rport;
 	char			target_name[32];
 	unsigned int		scsi_id;
 	unsigned int		sg_tablesize;
+	int			queue_size;
+	int			req_ring_size;
 	int			comp_vector;
+	int			tl_retry_count;
 
 	struct ib_sa_path_rec	path;
 	__be16			orig_dgid[8];
@@ -172,10 +172,11 @@ struct srp_target_port {
 
 	int			zero_req_lim;
 
-	struct srp_iu	       *tx_ring[SRP_SQ_SIZE];
-	struct srp_iu	       *rx_ring[SRP_RQ_SIZE];
-	struct srp_request	req_ring[SRP_CMD_SQ_SIZE];
+	struct srp_iu	       **tx_ring;
+	struct srp_iu	       **rx_ring;
+	struct srp_request	*req_ring;
 
+	struct work_struct	tl_err_work;
 	struct work_struct	remove_work;
 
 	struct list_head	list;
diff --git a/drivers/input/Kconfig b/drivers/input/Kconfig
index 38b523a1ece0..a11ff74a5127 100644
--- a/drivers/input/Kconfig
+++ b/drivers/input/Kconfig
@@ -80,7 +80,7 @@ config INPUT_MATRIXKMAP
 comment "Userland interfaces"
 
 config INPUT_MOUSEDEV
-	tristate "Mouse interface" if EXPERT
+	tristate "Mouse interface"
 	default y
 	help
 	  Say Y here if you want your mouse to be accessible as char devices
diff --git a/drivers/input/evdev.c b/drivers/input/evdev.c
index b6ded17b3be3..a06e12552886 100644
--- a/drivers/input/evdev.c
+++ b/drivers/input/evdev.c
@@ -18,6 +18,8 @@
 #include <linux/poll.h>
 #include <linux/sched.h>
 #include <linux/slab.h>
+#include <linux/vmalloc.h>
+#include <linux/mm.h>
 #include <linux/module.h>
 #include <linux/init.h>
 #include <linux/input/mt.h>
@@ -369,7 +371,11 @@ static int evdev_release(struct inode *inode, struct file *file)
 	mutex_unlock(&evdev->mutex);
 
 	evdev_detach_client(evdev, client);
-	kfree(client);
+
+	if (is_vmalloc_addr(client))
+		vfree(client);
+	else
+		kfree(client);
 
 	evdev_close_device(evdev);
 
@@ -389,12 +395,14 @@ static int evdev_open(struct inode *inode, struct file *file)
 {
 	struct evdev *evdev = container_of(inode->i_cdev, struct evdev, cdev);
 	unsigned int bufsize = evdev_compute_buffer_size(evdev->handle.dev);
+	unsigned int size = sizeof(struct evdev_client) +
+					bufsize * sizeof(struct input_event);
 	struct evdev_client *client;
 	int error;
 
-	client = kzalloc(sizeof(struct evdev_client) +
-				bufsize * sizeof(struct input_event),
-			 GFP_KERNEL);
+	client = kzalloc(size, GFP_KERNEL | __GFP_NOWARN);
+	if (!client)
+		client = vzalloc(size);
 	if (!client)
 		return -ENOMEM;
 
diff --git a/drivers/input/input.c b/drivers/input/input.c
index e75d015024a1..846ccdd905b1 100644
--- a/drivers/input/input.c
+++ b/drivers/input/input.c
@@ -2052,7 +2052,7 @@ int input_register_device(struct input_dev *dev)
 	if (dev->hint_events_per_packet < packet_size)
 		dev->hint_events_per_packet = packet_size;
 
-	dev->max_vals = max(dev->hint_events_per_packet, packet_size) + 2;
+	dev->max_vals = dev->hint_events_per_packet + 2;
 	dev->vals = kcalloc(dev->max_vals, sizeof(*dev->vals), GFP_KERNEL);
 	if (!dev->vals) {
 		error = -ENOMEM;
diff --git a/drivers/input/keyboard/Kconfig b/drivers/input/keyboard/Kconfig
index c1edd39bc5ba..bb174c1a9886 100644
--- a/drivers/input/keyboard/Kconfig
+++ b/drivers/input/keyboard/Kconfig
@@ -2,7 +2,7 @@
 # Input core configuration
 #
 menuconfig INPUT_KEYBOARD
-	bool "Keyboards" if EXPERT || !X86
+	bool "Keyboards"
 	default y
 	help
 	  Say Y here, and a list of supported keyboards will be displayed.
@@ -67,7 +67,7 @@ config KEYBOARD_ATARI
 	  module will be called atakbd.
 
 config KEYBOARD_ATKBD
-	tristate "AT keyboard" if EXPERT || !X86
+	tristate "AT keyboard"
 	default y
 	select SERIO
 	select SERIO_LIBPS2
@@ -525,7 +525,7 @@ config KEYBOARD_SUNKBD
 
 config KEYBOARD_SH_KEYSC
 	tristate "SuperH KEYSC keypad support"
-	depends on SUPERH || ARCH_SHMOBILE
+	depends on SUPERH || ARM || COMPILE_TEST
 	help
 	  Say Y here if you want to use a keypad attached to the KEYSC block
 	  on SuperH processors such as sh7722 and sh7343.
diff --git a/drivers/input/keyboard/gpio_keys.c b/drivers/input/keyboard/gpio_keys.c
index 440ce32462ba..2db13246eb8e 100644
--- a/drivers/input/keyboard/gpio_keys.c
+++ b/drivers/input/keyboard/gpio_keys.c
@@ -26,6 +26,7 @@
 #include <linux/gpio_keys.h>
 #include <linux/workqueue.h>
 #include <linux/gpio.h>
+#include <linux/of.h>
 #include <linux/of_platform.h>
 #include <linux/of_gpio.h>
 #include <linux/spinlock.h>
diff --git a/drivers/input/keyboard/gpio_keys_polled.c b/drivers/input/keyboard/gpio_keys_polled.c
index cd5ed9e22168..4e428199e580 100644
--- a/drivers/input/keyboard/gpio_keys_polled.c
+++ b/drivers/input/keyboard/gpio_keys_polled.c
@@ -25,6 +25,7 @@
 #include <linux/platform_device.h>
 #include <linux/gpio.h>
 #include <linux/gpio_keys.h>
+#include <linux/of.h>
 #include <linux/of_platform.h>
 #include <linux/of_gpio.h>
 
diff --git a/drivers/input/keyboard/lpc32xx-keys.c b/drivers/input/keyboard/lpc32xx-keys.c
index 42181435fe67..8b1b01361ec6 100644
--- a/drivers/input/keyboard/lpc32xx-keys.c
+++ b/drivers/input/keyboard/lpc32xx-keys.c
@@ -383,7 +383,7 @@ static struct platform_driver lpc32xx_kscan_driver = {
 		.name	= DRV_NAME,
 		.owner	= THIS_MODULE,
 		.pm	= &lpc32xx_kscan_pm_ops,
-		.of_match_table = of_match_ptr(lpc32xx_kscan_match),
+		.of_match_table = lpc32xx_kscan_match,
 	}
 };
 
diff --git a/drivers/input/keyboard/nspire-keypad.c b/drivers/input/keyboard/nspire-keypad.c
index b3e3edab6d9f..b31064981e96 100644
--- a/drivers/input/keyboard/nspire-keypad.c
+++ b/drivers/input/keyboard/nspire-keypad.c
@@ -143,8 +143,10 @@ static int nspire_keypad_open(struct input_dev *input)
 		return error;
 
 	error = nspire_keypad_chip_init(keypad);
-	if (error)
+	if (error) {
+		clk_disable_unprepare(keypad->clk);
 		return error;
+	}
 
 	return 0;
 }
@@ -267,7 +269,7 @@ static struct platform_driver nspire_keypad_driver = {
 	.driver = {
 		.name = "nspire-keypad",
 		.owner = THIS_MODULE,
-		.of_match_table = of_match_ptr(nspire_keypad_dt_match),
+		.of_match_table = nspire_keypad_dt_match,
 	},
 	.probe = nspire_keypad_probe,
 };
diff --git a/drivers/input/keyboard/pxa27x_keypad.c b/drivers/input/keyboard/pxa27x_keypad.c
index a2e758d27584..186138c720c7 100644
--- a/drivers/input/keyboard/pxa27x_keypad.c
+++ b/drivers/input/keyboard/pxa27x_keypad.c
@@ -27,6 +27,7 @@
 #include <linux/err.h>
 #include <linux/input/matrix_keypad.h>
 #include <linux/slab.h>
+#include <linux/of.h>
 
 #include <asm/mach/arch.h>
 #include <asm/mach/map.h>
diff --git a/drivers/input/keyboard/tegra-kbc.c b/drivers/input/keyboard/tegra-kbc.c
index 9cd20e6905a0..8508879f6faf 100644
--- a/drivers/input/keyboard/tegra-kbc.c
+++ b/drivers/input/keyboard/tegra-kbc.c
@@ -614,7 +614,7 @@ static int tegra_kbc_probe(struct platform_device *pdev)
 	unsigned int keymap_rows;
 	const struct of_device_id *match;
 
-	match = of_match_device(of_match_ptr(tegra_kbc_of_match), &pdev->dev);
+	match = of_match_device(tegra_kbc_of_match, &pdev->dev);
 
 	kbc = devm_kzalloc(&pdev->dev, sizeof(*kbc), GFP_KERNEL);
 	if (!kbc) {
diff --git a/drivers/input/keyboard/tnetv107x-keypad.c b/drivers/input/keyboard/tnetv107x-keypad.c
index 5f7b427dd7ed..8bd24d52bf1b 100644
--- a/drivers/input/keyboard/tnetv107x-keypad.c
+++ b/drivers/input/keyboard/tnetv107x-keypad.c
@@ -60,8 +60,8 @@ struct keypad_data {
 	struct clk			*clk;
 	struct device			*dev;
 	spinlock_t			lock;
-	u32				irq_press;
-	u32				irq_release;
+	int				irq_press;
+	int				irq_release;
 	int				rows, cols, row_shift;
 	int				debounce_ms, active_low;
 	u32				prev_keys[3];
diff --git a/drivers/input/misc/Kconfig b/drivers/input/misc/Kconfig
index aa51baaa9b1e..5f4967d01bc3 100644
--- a/drivers/input/misc/Kconfig
+++ b/drivers/input/misc/Kconfig
@@ -156,7 +156,7 @@ config INPUT_MAX8925_ONKEY
 
 config INPUT_MAX8997_HAPTIC
 	tristate "MAXIM MAX8997 haptic controller support"
-	depends on HAVE_PWM && MFD_MAX8997
+	depends on PWM && HAVE_PWM && MFD_MAX8997
 	select INPUT_FF_MEMLESS
 	help
 	  This option enables device driver support for the haptic controller
@@ -461,7 +461,7 @@ config INPUT_PCF8574
 
 config INPUT_PWM_BEEPER
 	tristate "PWM beeper support"
-	depends on HAVE_PWM || PWM
+	depends on PWM && HAVE_PWM
 	help
 	  Say Y here to get support for PWM based beeper devices.
 
diff --git a/drivers/input/misc/ad714x-spi.c b/drivers/input/misc/ad714x-spi.c
index 61891486067c..3a90b710e309 100644
--- a/drivers/input/misc/ad714x-spi.c
+++ b/drivers/input/misc/ad714x-spi.c
@@ -108,7 +108,6 @@ static int ad714x_spi_remove(struct spi_device *spi)
 	struct ad714x_chip *chip = spi_get_drvdata(spi);
 
 	ad714x_remove(chip);
-	spi_set_drvdata(spi, NULL);
 
 	return 0;
 }
diff --git a/drivers/input/misc/cobalt_btns.c b/drivers/input/misc/cobalt_btns.c
index 4f77f87847e8..b5d71d245854 100644
--- a/drivers/input/misc/cobalt_btns.c
+++ b/drivers/input/misc/cobalt_btns.c
@@ -131,7 +131,6 @@ static int cobalt_buttons_probe(struct platform_device *pdev)
  err_free_mem:
 	input_free_polled_device(poll_dev);
 	kfree(bdev);
-	dev_set_drvdata(&pdev->dev, NULL);
 	return error;
 }
 
@@ -144,7 +143,6 @@ static int cobalt_buttons_remove(struct platform_device *pdev)
 	input_free_polled_device(bdev->poll_dev);
 	iounmap(bdev->reg);
 	kfree(bdev);
-	dev_set_drvdata(dev, NULL);
 
 	return 0;
 }
diff --git a/drivers/input/misc/mma8450.c b/drivers/input/misc/mma8450.c
index f3309696d053..59d4dcddf6de 100644
--- a/drivers/input/misc/mma8450.c
+++ b/drivers/input/misc/mma8450.c
@@ -168,7 +168,7 @@ static void mma8450_close(struct input_polled_dev *dev)
  * I2C init/probing/exit functions
  */
 static int mma8450_probe(struct i2c_client *c,
-				   const struct i2c_device_id *id)
+			 const struct i2c_device_id *id)
 {
 	struct input_polled_dev *idev;
 	struct mma8450 *m;
@@ -204,6 +204,8 @@ static int mma8450_probe(struct i2c_client *c,
 		goto err_free_mem;
 	}
 
+	i2c_set_clientdata(c, m);
+
 	return 0;
 
 err_free_mem:
diff --git a/drivers/input/misc/mpu3050.c b/drivers/input/misc/mpu3050.c
index dce0d95943c5..6983ffbbfb94 100644
--- a/drivers/input/misc/mpu3050.c
+++ b/drivers/input/misc/mpu3050.c
@@ -383,6 +383,7 @@ static int mpu3050_probe(struct i2c_client *client,
 
 	pm_runtime_enable(&client->dev);
 	pm_runtime_set_autosuspend_delay(&client->dev, MPU3050_AUTO_DELAY);
+	i2c_set_clientdata(client, sensor);
 
 	return 0;
 
diff --git a/drivers/input/misc/pwm-beeper.c b/drivers/input/misc/pwm-beeper.c
index 2ff4d1c78ab8..940566e7be13 100644
--- a/drivers/input/misc/pwm-beeper.c
+++ b/drivers/input/misc/pwm-beeper.c
@@ -16,6 +16,7 @@
 #include <linux/input.h>
 #include <linux/module.h>
 #include <linux/kernel.h>
+#include <linux/of.h>
 #include <linux/platform_device.h>
 #include <linux/pwm.h>
 #include <linux/slab.h>
diff --git a/drivers/input/misc/rb532_button.c b/drivers/input/misc/rb532_button.c
index fb4f8ac3343b..83fff38b86b3 100644
--- a/drivers/input/misc/rb532_button.c
+++ b/drivers/input/misc/rb532_button.c
@@ -87,7 +87,6 @@ static int rb532_button_remove(struct platform_device *pdev)
 
 	input_unregister_polled_device(poll_dev);
 	input_free_polled_device(poll_dev);
-	dev_set_drvdata(&pdev->dev, NULL);
 
 	return 0;
 }
diff --git a/drivers/input/misc/rotary_encoder.c b/drivers/input/misc/rotary_encoder.c
index 5b1aff825138..f920ba7ab51f 100644
--- a/drivers/input/misc/rotary_encoder.c
+++ b/drivers/input/misc/rotary_encoder.c
@@ -24,6 +24,7 @@
 #include <linux/gpio.h>
 #include <linux/rotary_encoder.h>
 #include <linux/slab.h>
+#include <linux/of.h>
 #include <linux/of_platform.h>
 #include <linux/of_gpio.h>
 
diff --git a/drivers/input/misc/sirfsoc-onkey.c b/drivers/input/misc/sirfsoc-onkey.c
index 0621c367049a..7b8b03e0d0be 100644
--- a/drivers/input/misc/sirfsoc-onkey.c
+++ b/drivers/input/misc/sirfsoc-onkey.c
@@ -153,7 +153,7 @@ static struct platform_driver sirfsoc_pwrc_driver = {
 		.name	= "sirfsoc-pwrc",
 		.owner	= THIS_MODULE,
 		.pm	= &sirfsoc_pwrc_pm_ops,
-		.of_match_table = of_match_ptr(sirfsoc_pwrc_of_match),
+		.of_match_table = sirfsoc_pwrc_of_match,
 	}
 };
 
diff --git a/drivers/input/misc/uinput.c b/drivers/input/misc/uinput.c
index a0a4bbaef02c..772835938a52 100644
--- a/drivers/input/misc/uinput.c
+++ b/drivers/input/misc/uinput.c
@@ -430,20 +430,30 @@ static int uinput_setup_device(struct uinput_device *udev,
 	return retval;
 }
 
-static ssize_t uinput_inject_event(struct uinput_device *udev,
-				   const char __user *buffer, size_t count)
+static ssize_t uinput_inject_events(struct uinput_device *udev,
+				    const char __user *buffer, size_t count)
 {
 	struct input_event ev;
+	size_t bytes = 0;
 
-	if (count < input_event_size())
+	if (count != 0 && count < input_event_size())
 		return -EINVAL;
 
-	if (input_event_from_user(buffer, &ev))
-		return -EFAULT;
+	while (bytes + input_event_size() <= count) {
+		/*
+		 * Note that even if some events were fetched successfully
+		 * we are still going to return EFAULT instead of partial
+		 * count to let userspace know that it got it's buffers
+		 * all wrong.
+		 */
+		if (input_event_from_user(buffer + bytes, &ev))
+			return -EFAULT;
 
-	input_event(udev->dev, ev.type, ev.code, ev.value);
+		input_event(udev->dev, ev.type, ev.code, ev.value);
+		bytes += input_event_size();
+	}
 
-	return input_event_size();
+	return bytes;
 }
 
 static ssize_t uinput_write(struct file *file, const char __user *buffer,
@@ -460,7 +470,7 @@ static ssize_t uinput_write(struct file *file, const char __user *buffer,
 		return retval;
 
 	retval = udev->state == UIST_CREATED ?
-			uinput_inject_event(udev, buffer, count) :
+			uinput_inject_events(udev, buffer, count) :
 			uinput_setup_device(udev, buffer, count);
 
 	mutex_unlock(&udev->mutex);
diff --git a/drivers/input/mouse/alps.c b/drivers/input/mouse/alps.c
index 83658472ad25..ca7a26f1dce8 100644
--- a/drivers/input/mouse/alps.c
+++ b/drivers/input/mouse/alps.c
@@ -103,7 +103,6 @@ static const struct alps_model_info alps_model_data[] = {
 	/* Dell Latitude E5500, E6400, E6500, Precision M4400 */
 	{ { 0x62, 0x02, 0x14 }, 0x00, ALPS_PROTO_V2, 0xcf, 0xcf,
 		ALPS_PASS | ALPS_DUALPOINT | ALPS_PS2_INTERLEAVED },
-	{ { 0x73, 0x00, 0x14 }, 0x00, ALPS_PROTO_V2, 0xcf, 0xcf, ALPS_DUALPOINT },		/* Dell XT2 */
 	{ { 0x73, 0x02, 0x50 }, 0x00, ALPS_PROTO_V2, 0xcf, 0xcf, ALPS_FOUR_BUTTONS },		/* Dell Vostro 1400 */
 	{ { 0x52, 0x01, 0x14 }, 0x00, ALPS_PROTO_V2, 0xff, 0xff,
 		ALPS_PASS | ALPS_DUALPOINT | ALPS_PS2_INTERLEAVED },				/* Toshiba Tecra A11-11L */
@@ -1793,7 +1792,7 @@ int alps_init(struct psmouse *psmouse)
 	snprintf(priv->phys, sizeof(priv->phys), "%s/input1", psmouse->ps2dev.serio->phys);
 	dev2->phys = priv->phys;
 	dev2->name = (priv->flags & ALPS_DUALPOINT) ?
-		     "DualPoint Stick" : "PS/2 Mouse";
+		     "DualPoint Stick" : "ALPS PS/2 Device";
 	dev2->id.bustype = BUS_I8042;
 	dev2->id.vendor  = 0x0002;
 	dev2->id.product = PSMOUSE_ALPS;
diff --git a/drivers/input/mouse/cypress_ps2.c b/drivers/input/mouse/cypress_ps2.c
index f51765fff054..a5869a856ea5 100644
--- a/drivers/input/mouse/cypress_ps2.c
+++ b/drivers/input/mouse/cypress_ps2.c
@@ -439,7 +439,7 @@ static int cypress_get_finger_count(unsigned char header_byte)
 			case 2: return 5;
 			default:
 				/* Invalid contact (e.g. palm). Ignore it. */
-				return -1;
+				return 0;
 		}
 	}
 
@@ -452,17 +452,10 @@ static int cypress_parse_packet(struct psmouse *psmouse,
 {
 	unsigned char *packet = psmouse->packet;
 	unsigned char header_byte = packet[0];
-	int contact_cnt;
 
 	memset(report_data, 0, sizeof(struct cytp_report_data));
 
-	contact_cnt = cypress_get_finger_count(header_byte);
-
-	if (contact_cnt < 0) /* e.g. palm detect */
-		return -EINVAL;
-
-	report_data->contact_cnt = contact_cnt;
-
+	report_data->contact_cnt = cypress_get_finger_count(header_byte);
 	report_data->tap = (header_byte & ABS_MULTIFINGER_TAP) ? 1 : 0;
 
 	if (report_data->contact_cnt == 1) {
@@ -535,11 +528,9 @@ static void cypress_process_packet(struct psmouse *psmouse, bool zero_pkt)
 	int slots[CYTP_MAX_MT_SLOTS];
 	int n;
 
-	if (cypress_parse_packet(psmouse, cytp, &report_data))
-		return;
+	cypress_parse_packet(psmouse, cytp, &report_data);
 
 	n = report_data.contact_cnt;
-
 	if (n > CYTP_MAX_MT_SLOTS)
 		n = CYTP_MAX_MT_SLOTS;
 
@@ -605,10 +596,6 @@ static psmouse_ret_t cypress_validate_byte(struct psmouse *psmouse)
 		return PSMOUSE_BAD_DATA;
 
 	contact_cnt = cypress_get_finger_count(packet[0]);
-
-	if (contact_cnt < 0)
-		return PSMOUSE_BAD_DATA;
-
 	if (cytp->mode & CYTP_BIT_ABS_NO_PRESSURE)
 		cypress_set_packet_size(psmouse, contact_cnt == 2 ? 7 : 4);
 	else
@@ -679,15 +666,15 @@ int cypress_init(struct psmouse *psmouse)
 {
 	struct cytp_data *cytp;
 
-	cytp = (struct cytp_data *)kzalloc(sizeof(struct cytp_data), GFP_KERNEL);
-	psmouse->private = (void *)cytp;
-	if (cytp == NULL)
+	cytp = kzalloc(sizeof(struct cytp_data), GFP_KERNEL);
+	if (!cytp)
 		return -ENOMEM;
 
-	cypress_reset(psmouse);
-
+	psmouse->private = cytp;
 	psmouse->pktsize = 8;
 
+	cypress_reset(psmouse);
+
 	if (cypress_query_hardware(psmouse)) {
 		psmouse_err(psmouse, "Unable to query Trackpad hardware.\n");
 		goto err_exit;
diff --git a/drivers/input/serio/Kconfig b/drivers/input/serio/Kconfig
index 1de1e5f8f795..8541f949778d 100644
--- a/drivers/input/serio/Kconfig
+++ b/drivers/input/serio/Kconfig
@@ -2,7 +2,7 @@
 # Input core configuration
 #
 config SERIO
-	tristate "Serial I/O support" if EXPERT || !X86
+	tristate "Serial I/O support"
 	default y
 	help
 	  Say Yes here if you have any input device that uses serial I/O to
@@ -19,7 +19,7 @@ config SERIO
 if SERIO
 
 config SERIO_I8042
-	tristate "i8042 PC Keyboard controller" if EXPERT || !X86
+	tristate "i8042 PC Keyboard controller"
 	default y
 	depends on !PARISC && (!ARM || FOOTBRIDGE_HOST) && \
 		   (!SUPERH || SH_CAYMAN) && !M68K && !BLACKFIN && !S390 && \
@@ -170,7 +170,7 @@ config SERIO_MACEPS2
 	  module will be called maceps2.
 
 config SERIO_LIBPS2
-	tristate "PS/2 driver library" if EXPERT
+	tristate "PS/2 driver library"
 	depends on SERIO_I8042 || SERIO_I8042=n
 	help
 	  Say Y here if you are using a driver for device connected
@@ -266,4 +266,14 @@ config SERIO_OLPC_APSP
 	  To compile this driver as a module, choose M here: the module will
 	  be called olpc_apsp.
 
+config HYPERV_KEYBOARD
+	tristate "Microsoft Synthetic Keyboard driver"
+	depends on HYPERV
+	default HYPERV
+	help
+	  Select this option to enable the Hyper-V Keyboard driver.
+
+	  To compile this driver as a module, choose M here: the module will
+	  be called hyperv_keyboard.
+
 endif
diff --git a/drivers/input/serio/Makefile b/drivers/input/serio/Makefile
index 12298b1c0e71..815d874fe724 100644
--- a/drivers/input/serio/Makefile
+++ b/drivers/input/serio/Makefile
@@ -28,3 +28,4 @@ obj-$(CONFIG_SERIO_ALTERA_PS2)	+= altera_ps2.o
 obj-$(CONFIG_SERIO_ARC_PS2)	+= arc_ps2.o
 obj-$(CONFIG_SERIO_APBPS2)	+= apbps2.o
 obj-$(CONFIG_SERIO_OLPC_APSP)	+= olpc_apsp.o
+obj-$(CONFIG_HYPERV_KEYBOARD)	+= hyperv-keyboard.o
diff --git a/drivers/input/serio/hyperv-keyboard.c b/drivers/input/serio/hyperv-keyboard.c
new file mode 100644
index 000000000000..3a83c3c14b23
--- /dev/null
+++ b/drivers/input/serio/hyperv-keyboard.c
@@ -0,0 +1,437 @@
+/*
+ *  Copyright (c) 2013, Microsoft Corporation.
+ *
+ *  This program is free software; you can redistribute it and/or modify it
+ *  under the terms and conditions of the GNU General Public License,
+ *  version 2, as published by the Free Software Foundation.
+ *
+ *  This program is distributed in the hope it will be useful, but WITHOUT
+ *  ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or
+ *  FITNESS FOR A PARTICULAR PURPOSE.  See the GNU General Public License for
+ *  more details.
+ */
+
+#include <linux/init.h>
+#include <linux/module.h>
+#include <linux/device.h>
+#include <linux/completion.h>
+#include <linux/hyperv.h>
+#include <linux/serio.h>
+#include <linux/slab.h>
+
+/*
+ * Current version 1.0
+ *
+ */
+#define SYNTH_KBD_VERSION_MAJOR 1
+#define SYNTH_KBD_VERSION_MINOR	0
+#define SYNTH_KBD_VERSION		(SYNTH_KBD_VERSION_MINOR | \
+					 (SYNTH_KBD_VERSION_MAJOR << 16))
+
+
+/*
+ * Message types in the synthetic input protocol
+ */
+enum synth_kbd_msg_type {
+	SYNTH_KBD_PROTOCOL_REQUEST = 1,
+	SYNTH_KBD_PROTOCOL_RESPONSE = 2,
+	SYNTH_KBD_EVENT = 3,
+	SYNTH_KBD_LED_INDICATORS = 4,
+};
+
+/*
+ * Basic message structures.
+ */
+struct synth_kbd_msg_hdr {
+	__le32 type;
+};
+
+struct synth_kbd_msg {
+	struct synth_kbd_msg_hdr header;
+	char data[]; /* Enclosed message */
+};
+
+union synth_kbd_version {
+	__le32 version;
+};
+
+/*
+ * Protocol messages
+ */
+struct synth_kbd_protocol_request {
+	struct synth_kbd_msg_hdr header;
+	union synth_kbd_version version_requested;
+};
+
+#define PROTOCOL_ACCEPTED	BIT(0)
+struct synth_kbd_protocol_response {
+	struct synth_kbd_msg_hdr header;
+	__le32 proto_status;
+};
+
+#define IS_UNICODE	BIT(0)
+#define IS_BREAK	BIT(1)
+#define IS_E0		BIT(2)
+#define IS_E1		BIT(3)
+struct synth_kbd_keystroke {
+	struct synth_kbd_msg_hdr header;
+	__le16 make_code;
+	__le16 reserved0;
+	__le32 info; /* Additional information */
+};
+
+
+#define HK_MAXIMUM_MESSAGE_SIZE 256
+
+#define KBD_VSC_SEND_RING_BUFFER_SIZE		(10 * PAGE_SIZE)
+#define KBD_VSC_RECV_RING_BUFFER_SIZE		(10 * PAGE_SIZE)
+
+#define XTKBD_EMUL0     0xe0
+#define XTKBD_EMUL1     0xe1
+#define XTKBD_RELEASE   0x80
+
+
+/*
+ * Represents a keyboard device
+ */
+struct hv_kbd_dev {
+	struct hv_device *hv_dev;
+	struct serio *hv_serio;
+	struct synth_kbd_protocol_request protocol_req;
+	struct synth_kbd_protocol_response protocol_resp;
+	/* Synchronize the request/response if needed */
+	struct completion wait_event;
+	spinlock_t lock; /* protects 'started' field */
+	bool started;
+};
+
+static void hv_kbd_on_receive(struct hv_device *hv_dev,
+			      struct synth_kbd_msg *msg, u32 msg_length)
+{
+	struct hv_kbd_dev *kbd_dev = hv_get_drvdata(hv_dev);
+	struct synth_kbd_keystroke *ks_msg;
+	unsigned long flags;
+	u32 msg_type = __le32_to_cpu(msg->header.type);
+	u32 info;
+	u16 scan_code;
+
+	switch (msg_type) {
+	case SYNTH_KBD_PROTOCOL_RESPONSE:
+		/*
+		 * Validate the information provided by the host.
+		 * If the host is giving us a bogus packet,
+		 * drop the packet (hoping the problem
+		 * goes away).
+		 */
+		if (msg_length < sizeof(struct synth_kbd_protocol_response)) {
+			dev_err(&hv_dev->device,
+				"Illegal protocol response packet (len: %d)\n",
+				msg_length);
+			break;
+		}
+
+		memcpy(&kbd_dev->protocol_resp, msg,
+			sizeof(struct synth_kbd_protocol_response));
+		complete(&kbd_dev->wait_event);
+		break;
+
+	case SYNTH_KBD_EVENT:
+		/*
+		 * Validate the information provided by the host.
+		 * If the host is giving us a bogus packet,
+		 * drop the packet (hoping the problem
+		 * goes away).
+		 */
+		if (msg_length < sizeof(struct  synth_kbd_keystroke)) {
+			dev_err(&hv_dev->device,
+				"Illegal keyboard event packet (len: %d)\n",
+				msg_length);
+			break;
+		}
+
+		ks_msg = (struct synth_kbd_keystroke *)msg;
+		info = __le32_to_cpu(ks_msg->info);
+
+		/*
+		 * Inject the information through the serio interrupt.
+		 */
+		spin_lock_irqsave(&kbd_dev->lock, flags);
+		if (kbd_dev->started) {
+			if (info & IS_E0)
+				serio_interrupt(kbd_dev->hv_serio,
+						XTKBD_EMUL0, 0);
+
+			scan_code = __le16_to_cpu(ks_msg->make_code);
+			if (info & IS_BREAK)
+				scan_code |= XTKBD_RELEASE;
+
+			serio_interrupt(kbd_dev->hv_serio, scan_code, 0);
+		}
+		spin_unlock_irqrestore(&kbd_dev->lock, flags);
+		break;
+
+	default:
+		dev_err(&hv_dev->device,
+			"unhandled message type %d\n", msg_type);
+	}
+}
+
+static void hv_kbd_handle_received_packet(struct hv_device *hv_dev,
+					  struct vmpacket_descriptor *desc,
+					  u32 bytes_recvd,
+					  u64 req_id)
+{
+	struct synth_kbd_msg *msg;
+	u32 msg_sz;
+
+	switch (desc->type) {
+	case VM_PKT_COMP:
+		break;
+
+	case VM_PKT_DATA_INBAND:
+		/*
+		 * We have a packet that has "inband" data. The API used
+		 * for retrieving the packet guarantees that the complete
+		 * packet is read. So, minimally, we should be able to
+		 * parse the payload header safely (assuming that the host
+		 * can be trusted.  Trusting the host seems to be a
+		 * reasonable assumption because in a virtualized
+		 * environment there is not whole lot you can do if you
+		 * don't trust the host.
+		 *
+		 * Nonetheless, let us validate if the host can be trusted
+		 * (in a trivial way).  The interesting aspect of this
+		 * validation is how do you recover if we discover that the
+		 * host is not to be trusted? Simply dropping the packet, I
+		 * don't think is an appropriate recovery.  In the interest
+		 * of failing fast, it may be better to crash the guest.
+		 * For now, I will just drop the packet!
+		 */
+
+		msg_sz = bytes_recvd - (desc->offset8 << 3);
+		if (msg_sz <= sizeof(struct synth_kbd_msg_hdr)) {
+			/*
+			 * Drop the packet and hope
+			 * the problem magically goes away.
+			 */
+			dev_err(&hv_dev->device,
+				"Illegal packet (type: %d, tid: %llx, size: %d)\n",
+				desc->type, req_id, msg_sz);
+			break;
+		}
+
+		msg = (void *)desc + (desc->offset8 << 3);
+		hv_kbd_on_receive(hv_dev, msg, msg_sz);
+		break;
+
+	default:
+		dev_err(&hv_dev->device,
+			"unhandled packet type %d, tid %llx len %d\n",
+			desc->type, req_id, bytes_recvd);
+		break;
+	}
+}
+
+static void hv_kbd_on_channel_callback(void *context)
+{
+	struct hv_device *hv_dev = context;
+	void *buffer;
+	int bufferlen = 0x100; /* Start with sensible size */
+	u32 bytes_recvd;
+	u64 req_id;
+	int error;
+
+	buffer = kmalloc(bufferlen, GFP_ATOMIC);
+	if (!buffer)
+		return;
+
+	while (1) {
+		error = vmbus_recvpacket_raw(hv_dev->channel, buffer, bufferlen,
+					     &bytes_recvd, &req_id);
+		switch (error) {
+		case 0:
+			if (bytes_recvd == 0) {
+				kfree(buffer);
+				return;
+			}
+
+			hv_kbd_handle_received_packet(hv_dev, buffer,
+						      bytes_recvd, req_id);
+			break;
+
+		case -ENOBUFS:
+			kfree(buffer);
+			/* Handle large packet */
+			bufferlen = bytes_recvd;
+			buffer = kmalloc(bytes_recvd, GFP_ATOMIC);
+			if (!buffer)
+				return;
+			break;
+		}
+	}
+}
+
+static int hv_kbd_connect_to_vsp(struct hv_device *hv_dev)
+{
+	struct hv_kbd_dev *kbd_dev = hv_get_drvdata(hv_dev);
+	struct synth_kbd_protocol_request *request;
+	struct synth_kbd_protocol_response *response;
+	u32 proto_status;
+	int error;
+
+	request = &kbd_dev->protocol_req;
+	memset(request, 0, sizeof(struct synth_kbd_protocol_request));
+	request->header.type = __cpu_to_le32(SYNTH_KBD_PROTOCOL_REQUEST);
+	request->version_requested.version = __cpu_to_le32(SYNTH_KBD_VERSION);
+
+	error = vmbus_sendpacket(hv_dev->channel, request,
+				 sizeof(struct synth_kbd_protocol_request),
+				 (unsigned long)request,
+				 VM_PKT_DATA_INBAND,
+				 VMBUS_DATA_PACKET_FLAG_COMPLETION_REQUESTED);
+	if (error)
+		return error;
+
+	if (!wait_for_completion_timeout(&kbd_dev->wait_event, 10 * HZ))
+		return -ETIMEDOUT;
+
+	response = &kbd_dev->protocol_resp;
+	proto_status = __le32_to_cpu(response->proto_status);
+	if (!(proto_status & PROTOCOL_ACCEPTED)) {
+		dev_err(&hv_dev->device,
+			"synth_kbd protocol request failed (version %d)\n",
+		        SYNTH_KBD_VERSION);
+		return -ENODEV;
+	}
+
+	return 0;
+}
+
+static int hv_kbd_start(struct serio *serio)
+{
+	struct hv_kbd_dev *kbd_dev = serio->port_data;
+	unsigned long flags;
+
+	spin_lock_irqsave(&kbd_dev->lock, flags);
+	kbd_dev->started = true;
+	spin_unlock_irqrestore(&kbd_dev->lock, flags);
+
+	return 0;
+}
+
+static void hv_kbd_stop(struct serio *serio)
+{
+	struct hv_kbd_dev *kbd_dev = serio->port_data;
+	unsigned long flags;
+
+	spin_lock_irqsave(&kbd_dev->lock, flags);
+	kbd_dev->started = false;
+	spin_unlock_irqrestore(&kbd_dev->lock, flags);
+}
+
+static int hv_kbd_probe(struct hv_device *hv_dev,
+			const struct hv_vmbus_device_id *dev_id)
+{
+	struct hv_kbd_dev *kbd_dev;
+	struct serio *hv_serio;
+	int error;
+
+	kbd_dev = kzalloc(sizeof(struct hv_kbd_dev), GFP_KERNEL);
+	hv_serio = kzalloc(sizeof(struct serio), GFP_KERNEL);
+	if (!kbd_dev || !hv_serio) {
+		error = -ENOMEM;
+		goto err_free_mem;
+	}
+
+	kbd_dev->hv_dev = hv_dev;
+	kbd_dev->hv_serio = hv_serio;
+	spin_lock_init(&kbd_dev->lock);
+	init_completion(&kbd_dev->wait_event);
+	hv_set_drvdata(hv_dev, kbd_dev);
+
+	hv_serio->dev.parent  = &hv_dev->device;
+	hv_serio->id.type = SERIO_8042_XL;
+	hv_serio->port_data = kbd_dev;
+	strlcpy(hv_serio->name, dev_name(&hv_dev->device),
+		sizeof(hv_serio->name));
+	strlcpy(hv_serio->phys, dev_name(&hv_dev->device),
+		sizeof(hv_serio->phys));
+
+	hv_serio->start = hv_kbd_start;
+	hv_serio->stop = hv_kbd_stop;
+
+	error = vmbus_open(hv_dev->channel,
+			   KBD_VSC_SEND_RING_BUFFER_SIZE,
+			   KBD_VSC_RECV_RING_BUFFER_SIZE,
+			   NULL, 0,
+			   hv_kbd_on_channel_callback,
+			   hv_dev);
+	if (error)
+		goto err_free_mem;
+
+	error = hv_kbd_connect_to_vsp(hv_dev);
+	if (error)
+		goto err_close_vmbus;
+
+	serio_register_port(kbd_dev->hv_serio);
+	return 0;
+
+err_close_vmbus:
+	vmbus_close(hv_dev->channel);
+err_free_mem:
+	kfree(hv_serio);
+	kfree(kbd_dev);
+	return error;
+}
+
+static int hv_kbd_remove(struct hv_device *hv_dev)
+{
+	struct hv_kbd_dev *kbd_dev = hv_get_drvdata(hv_dev);
+
+	serio_unregister_port(kbd_dev->hv_serio);
+	vmbus_close(hv_dev->channel);
+	kfree(kbd_dev);
+
+	hv_set_drvdata(hv_dev, NULL);
+
+	return 0;
+}
+
+/*
+ * Keyboard GUID
+ * {f912ad6d-2b17-48ea-bd65-f927a61c7684}
+ */
+#define HV_KBD_GUID \
+	.guid = { \
+			0x6d, 0xad, 0x12, 0xf9, 0x17, 0x2b, 0xea, 0x48, \
+			0xbd, 0x65, 0xf9, 0x27, 0xa6, 0x1c, 0x76, 0x84 \
+	}
+
+static const struct hv_vmbus_device_id id_table[] = {
+	/* Keyboard guid */
+	{ HV_KBD_GUID, },
+	{ },
+};
+
+MODULE_DEVICE_TABLE(vmbus, id_table);
+
+static struct  hv_driver hv_kbd_drv = {
+	.name = KBUILD_MODNAME,
+	.id_table = id_table,
+	.probe = hv_kbd_probe,
+	.remove = hv_kbd_remove,
+};
+
+static int __init hv_kbd_init(void)
+{
+	return vmbus_driver_register(&hv_kbd_drv);
+}
+
+static void __exit hv_kbd_exit(void)
+{
+	vmbus_driver_unregister(&hv_kbd_drv);
+}
+
+MODULE_LICENSE("GPL");
+module_init(hv_kbd_init);
+module_exit(hv_kbd_exit);
diff --git a/drivers/input/serio/i8042-x86ia64io.h b/drivers/input/serio/i8042-x86ia64io.h
index 5f306f79da0c..0ec9abbe31fe 100644
--- a/drivers/input/serio/i8042-x86ia64io.h
+++ b/drivers/input/serio/i8042-x86ia64io.h
@@ -765,6 +765,7 @@ static struct pnp_device_id pnp_kbd_devids[] = {
 	{ .id = "CPQA0D7", .driver_data = 0 },
 	{ .id = "", },
 };
+MODULE_DEVICE_TABLE(pnp, pnp_kbd_devids);
 
 static struct pnp_driver i8042_pnp_kbd_driver = {
 	.name           = "i8042 kbd",
@@ -786,6 +787,7 @@ static struct pnp_device_id pnp_aux_devids[] = {
 	{ .id = "SYN0801", .driver_data = 0 },
 	{ .id = "", },
 };
+MODULE_DEVICE_TABLE(pnp, pnp_aux_devids);
 
 static struct pnp_driver i8042_pnp_aux_driver = {
 	.name           = "i8042 aux",
diff --git a/drivers/input/serio/i8042.c b/drivers/input/serio/i8042.c
index 52c9ebf94729..020053fa5aaa 100644
--- a/drivers/input/serio/i8042.c
+++ b/drivers/input/serio/i8042.c
@@ -1036,7 +1036,7 @@ static void i8042_controller_reset(bool force_reset)
 /*
  * i8042_panic_blink() will turn the keyboard LEDs on or off and is called
  * when kernel panics. Flashing LEDs is useful for users running X who may
- * not see the console and will help distingushing panics from "real"
+ * not see the console and will help distinguishing panics from "real"
  * lockups.
  *
  * Note that DELAY has a limit of 10ms so we will not get stuck here
diff --git a/drivers/input/tablet/wacom_sys.c b/drivers/input/tablet/wacom_sys.c
index e53416a4d7f3..867e7c33ac55 100644
--- a/drivers/input/tablet/wacom_sys.c
+++ b/drivers/input/tablet/wacom_sys.c
@@ -524,9 +524,6 @@ static int wacom_set_device_mode(struct usb_interface *intf, int report_id, int
 
 		error = wacom_set_report(intf, WAC_HID_FEATURE_REPORT,
 		                         report_id, rep_data, length, 1);
-		if (error >= 0)
-			error = wacom_get_report(intf, WAC_HID_FEATURE_REPORT,
-			                         report_id, rep_data, length, 1);
 	} while ((error < 0 || rep_data[1] != mode) && limit++ < WAC_MSG_RETRIES);
 
 	kfree(rep_data);
@@ -548,7 +545,7 @@ static int wacom_query_tablet_data(struct usb_interface *intf, struct wacom_feat
 			/* MT Tablet PC touch */
 			return wacom_set_device_mode(intf, 3, 4, 4);
 		}
-		else if (features->type == WACOM_24HDT) {
+		else if (features->type == WACOM_24HDT || features->type == CINTIQ_HYBRID) {
 			return wacom_set_device_mode(intf, 18, 3, 2);
 		}
 	} else if (features->device_type == BTN_TOOL_PEN) {
@@ -719,7 +716,7 @@ static int wacom_led_control(struct wacom *wacom)
 		return -ENOMEM;
 
 	if (wacom->wacom_wac.features.type >= INTUOS5S &&
-	    wacom->wacom_wac.features.type <= INTUOS5L)	{
+	    wacom->wacom_wac.features.type <= INTUOSPL) {
 		/*
 		 * Touch Ring and crop mark LED luminance may take on
 		 * one of four values:
@@ -981,14 +978,20 @@ static int wacom_initialize_leds(struct wacom *wacom)
 	case INTUOS5S:
 	case INTUOS5:
 	case INTUOS5L:
-		wacom->led.select[0] = 0;
-		wacom->led.select[1] = 0;
-		wacom->led.llv = 32;
-		wacom->led.hlv = 0;
-		wacom->led.img_lum = 0;
-
-		error = sysfs_create_group(&wacom->intf->dev.kobj,
-					   &intuos5_led_attr_group);
+	case INTUOSPS:
+	case INTUOSPM:
+	case INTUOSPL:
+		if (wacom->wacom_wac.features.device_type == BTN_TOOL_PEN) {
+			wacom->led.select[0] = 0;
+			wacom->led.select[1] = 0;
+			wacom->led.llv = 32;
+			wacom->led.hlv = 0;
+			wacom->led.img_lum = 0;
+
+			error = sysfs_create_group(&wacom->intf->dev.kobj,
+						  &intuos5_led_attr_group);
+		} else
+			return 0;
 		break;
 
 	default:
@@ -1024,8 +1027,12 @@ static void wacom_destroy_leds(struct wacom *wacom)
 	case INTUOS5S:
 	case INTUOS5:
 	case INTUOS5L:
-		sysfs_remove_group(&wacom->intf->dev.kobj,
-				   &intuos5_led_attr_group);
+	case INTUOSPS:
+	case INTUOSPM:
+	case INTUOSPL:
+		if (wacom->wacom_wac.features.device_type == BTN_TOOL_PEN)
+			sysfs_remove_group(&wacom->intf->dev.kobj,
+					   &intuos5_led_attr_group);
 		break;
 	}
 }
@@ -1185,34 +1192,47 @@ static void wacom_wireless_work(struct work_struct *work)
 		wacom_wac1->features =
 			*((struct wacom_features *)id->driver_info);
 		wacom_wac1->features.device_type = BTN_TOOL_PEN;
+		snprintf(wacom_wac1->name, WACOM_NAME_MAX, "%s (WL) Pen",
+			 wacom_wac1->features.name);
 		error = wacom_register_input(wacom1);
 		if (error)
-			goto fail1;
+			goto fail;
 
 		/* Touch interface */
-		wacom_wac2->features =
-			*((struct wacom_features *)id->driver_info);
-		wacom_wac2->features.pktlen = WACOM_PKGLEN_BBTOUCH3;
-		wacom_wac2->features.device_type = BTN_TOOL_FINGER;
-		wacom_wac2->features.x_max = wacom_wac2->features.y_max = 4096;
-		error = wacom_register_input(wacom2);
-		if (error)
-			goto fail2;
+		if (wacom_wac1->features.touch_max) {
+			wacom_wac2->features =
+				*((struct wacom_features *)id->driver_info);
+			wacom_wac2->features.pktlen = WACOM_PKGLEN_BBTOUCH3;
+			wacom_wac2->features.device_type = BTN_TOOL_FINGER;
+			wacom_wac2->features.x_max = wacom_wac2->features.y_max = 4096;
+			if (wacom_wac2->features.touch_max)
+				snprintf(wacom_wac2->name, WACOM_NAME_MAX,
+					 "%s (WL) Finger",wacom_wac2->features.name);
+			else
+				snprintf(wacom_wac2->name, WACOM_NAME_MAX,
+					 "%s (WL) Pad",wacom_wac2->features.name);
+			error = wacom_register_input(wacom2);
+			if (error)
+				goto fail;
+		}
 
 		error = wacom_initialize_battery(wacom);
 		if (error)
-			goto fail3;
+			goto fail;
 	}
 
 	return;
 
-fail3:
-	input_unregister_device(wacom_wac2->input);
-	wacom_wac2->input = NULL;
-fail2:
-	input_unregister_device(wacom_wac1->input);
-	wacom_wac1->input = NULL;
-fail1:
+fail:
+	if (wacom_wac2->input) {
+		input_unregister_device(wacom_wac2->input);
+		wacom_wac2->input = NULL;
+	}
+
+	if (wacom_wac1->input) {
+		input_unregister_device(wacom_wac1->input);
+		wacom_wac1->input = NULL;
+	}
 	return;
 }
 
@@ -1302,7 +1322,7 @@ static int wacom_probe(struct usb_interface *intf, const struct usb_device_id *i
 	 * HID descriptor. If this is the touch interface (wMaxPacketSize
 	 * of WACOM_PKGLEN_BBTOUCH3), override the table values.
 	 */
-	if (features->type >= INTUOS5S && features->type <= INTUOS5L) {
+	if (features->type >= INTUOS5S && features->type <= INTUOSPL) {
 		if (endpoint->wMaxPacketSize == WACOM_PKGLEN_BBTOUCH3) {
 			features->device_type = BTN_TOOL_FINGER;
 			features->pktlen = WACOM_PKGLEN_BBTOUCH3;
@@ -1329,10 +1349,12 @@ static int wacom_probe(struct usb_interface *intf, const struct usb_device_id *i
 		struct usb_device *other_dev;
 
 		/* Append the device type to the name */
-		strlcat(wacom_wac->name,
-			features->device_type == BTN_TOOL_PEN ?
-				" Pen" : " Finger",
-			sizeof(wacom_wac->name));
+		if (features->device_type != BTN_TOOL_FINGER)
+			strlcat(wacom_wac->name, " Pen", WACOM_NAME_MAX);
+		else if (features->touch_max)
+			strlcat(wacom_wac->name, " Finger", WACOM_NAME_MAX);
+		else
+			strlcat(wacom_wac->name, " Pad", WACOM_NAME_MAX);
 
 		other_dev = wacom_get_sibling(dev, features->oVid, features->oPid);
 		if (other_dev == NULL || wacom_get_usbdev_data(other_dev) == NULL)
diff --git a/drivers/input/tablet/wacom_wac.c b/drivers/input/tablet/wacom_wac.c
index c59b797eeafa..782c2535f1d8 100644
--- a/drivers/input/tablet/wacom_wac.c
+++ b/drivers/input/tablet/wacom_wac.c
@@ -427,6 +427,13 @@ static int wacom_intuos_inout(struct wacom_wac *wacom)
 			(features->type == WACOM_21UX2))
 		return 1;
 
+	/* Range Report */
+	if ((data[1] & 0xfe) == 0x20) {
+		input_report_key(input, BTN_TOUCH, 0);
+		input_report_abs(input, ABS_PRESSURE, 0);
+		input_report_abs(input, ABS_DISTANCE, wacom->features.distance_max);
+	}
+
 	/* Exit report */
 	if ((data[1] & 0xfe) == 0x80) {
 		if (features->quirks == WACOM_QUIRK_MULTI_INPUT)
@@ -477,7 +484,7 @@ static void wacom_intuos_general(struct wacom_wac *wacom)
 	/* general pen packet */
 	if ((data[1] & 0xb8) == 0xa0) {
 		t = (data[6] << 2) | ((data[7] >> 6) & 3);
-		if (features->type >= INTUOS4S && features->type <= WACOM_24HD) {
+		if (features->type >= INTUOS4S && features->type <= CINTIQ_HYBRID) {
 			t = (t << 1) | (data[1] & 1);
 		}
 		input_report_abs(input, ABS_PRESSURE, t);
@@ -621,14 +628,30 @@ static int wacom_intuos_irq(struct wacom_wac *wacom)
 			} else {
 				input_report_abs(input, ABS_MISC, 0);
 			}
-		} else if (features->type >= INTUOS5S && features->type <= INTUOS5L) {
+		} else if (features->type == CINTIQ_HYBRID) {
+			/*
+			 * Do not send hardware buttons under Android. They
+			 * are already sent to the system through GPIO (and
+			 * have different meaning).
+			 */
+			input_report_key(input, BTN_1, (data[4] & 0x01));
+			input_report_key(input, BTN_2, (data[4] & 0x02));
+			input_report_key(input, BTN_3, (data[4] & 0x04));
+			input_report_key(input, BTN_4, (data[4] & 0x08));
+
+			input_report_key(input, BTN_5, (data[4] & 0x10));  /* Right  */
+			input_report_key(input, BTN_6, (data[4] & 0x20));  /* Up     */
+			input_report_key(input, BTN_7, (data[4] & 0x40));  /* Left   */
+			input_report_key(input, BTN_8, (data[4] & 0x80));  /* Down   */
+			input_report_key(input, BTN_0, (data[3] & 0x01));  /* Center */
+		} else if (features->type >= INTUOS5S && features->type <= INTUOSPL) {
 			int i;
 
 			/* Touch ring mode switch has no capacitive sensor */
 			input_report_key(input, BTN_0, (data[3] & 0x01));
 
 			/*
-			 * ExpressKeys on Intuos5 have a capacitive sensor in
+			 * ExpressKeys on Intuos5/Intuos Pro have a capacitive sensor in
 			 * addition to the mechanical switch. Switch data is
 			 * stored in data[4], capacitive data in data[5].
 			 */
@@ -716,7 +739,9 @@ static int wacom_intuos_irq(struct wacom_wac *wacom)
 	     features->type == INTUOS4 ||
 	     features->type == INTUOS4S ||
 	     features->type == INTUOS5 ||
-	     features->type == INTUOS5S)) {
+	     features->type == INTUOS5S ||
+	     features->type == INTUOSPM ||
+	     features->type == INTUOSPS)) {
 
 		return 0;
 	}
@@ -769,8 +794,7 @@ static int wacom_intuos_irq(struct wacom_wac *wacom)
 
 		} else if (wacom->tool[idx] == BTN_TOOL_MOUSE) {
 			/* I4 mouse */
-			if ((features->type >= INTUOS4S && features->type <= INTUOS4L) ||
-			    (features->type >= INTUOS5S && features->type <= INTUOS5L)) {
+			if (features->type >= INTUOS4S && features->type <= INTUOSPL) {
 				input_report_key(input, BTN_LEFT,   data[6] & 0x01);
 				input_report_key(input, BTN_MIDDLE, data[6] & 0x02);
 				input_report_key(input, BTN_RIGHT,  data[6] & 0x04);
@@ -797,7 +821,8 @@ static int wacom_intuos_irq(struct wacom_wac *wacom)
 				}
 			}
 		} else if ((features->type < INTUOS3S || features->type == INTUOS3L ||
-				features->type == INTUOS4L || features->type == INTUOS5L) &&
+				features->type == INTUOS4L || features->type == INTUOS5L ||
+				features->type == INTUOSPL) &&
 			   wacom->tool[idx] == BTN_TOOL_LENS) {
 			/* Lens cursor packets */
 			input_report_key(input, BTN_LEFT,   data[8] & 0x01);
@@ -1107,6 +1132,7 @@ static int wacom_bpt_touch(struct wacom_wac *wacom)
 
 static void wacom_bpt3_touch_msg(struct wacom_wac *wacom, unsigned char *data)
 {
+	struct wacom_features *features = &wacom->features;
 	struct input_dev *input = wacom->input;
 	bool touch = data[1] & 0x80;
 	int slot = input_mt_get_slot_by_key(input, data[0]);
@@ -1122,14 +1148,23 @@ static void wacom_bpt3_touch_msg(struct wacom_wac *wacom, unsigned char *data)
 	if (touch) {
 		int x = (data[2] << 4) | (data[4] >> 4);
 		int y = (data[3] << 4) | (data[4] & 0x0f);
-		int a = data[5];
+		int width, height;
 
-		// "a" is a scaled-down area which we assume is roughly
-		// circular and which can be described as: a=(pi*r^2)/C.
-		int x_res  = input_abs_get_res(input, ABS_X);
-		int y_res  = input_abs_get_res(input, ABS_Y);
-		int width  = 2 * int_sqrt(a * WACOM_CONTACT_AREA_SCALE);
-		int height = width * y_res / x_res;
+		if (features->type >= INTUOSPS && features->type <= INTUOSPL) {
+			width  = data[5];
+			height = data[6];
+		} else {
+			/*
+			 * "a" is a scaled-down area which we assume is
+			 * roughly circular and which can be described as:
+			 * a=(pi*r^2)/C.
+			 */
+			int a = data[5];
+			int x_res  = input_abs_get_res(input, ABS_X);
+			int y_res  = input_abs_get_res(input, ABS_Y);
+			width  = 2 * int_sqrt(a * WACOM_CONTACT_AREA_SCALE);
+			height = width * y_res / x_res;
+		}
 
 		input_report_abs(input, ABS_MT_POSITION_X, x);
 		input_report_abs(input, ABS_MT_POSITION_Y, y);
@@ -1327,6 +1362,7 @@ void wacom_wac_irq(struct wacom_wac *wacom_wac, size_t len)
 	case WACOM_22HD:
 	case WACOM_24HD:
 	case DTK:
+	case CINTIQ_HYBRID:
 		sync = wacom_intuos_irq(wacom_wac);
 		break;
 
@@ -1337,6 +1373,9 @@ void wacom_wac_irq(struct wacom_wac *wacom_wac, size_t len)
 	case INTUOS5S:
 	case INTUOS5:
 	case INTUOS5L:
+	case INTUOSPS:
+	case INTUOSPM:
+	case INTUOSPL:
 		if (len == WACOM_PKGLEN_BBTOUCH3)
 			sync = wacom_bpt3_touch(wacom_wac);
 		else
@@ -1420,7 +1459,7 @@ void wacom_setup_device_quirks(struct wacom_features *features)
 
 	/* these device have multiple inputs */
 	if (features->type >= WIRELESS ||
-	    (features->type >= INTUOS5S && features->type <= INTUOS5L) ||
+	    (features->type >= INTUOS5S && features->type <= INTUOSPL) ||
 	    (features->oVid && features->oPid))
 		features->quirks |= WACOM_QUIRK_MULTI_INPUT;
 
@@ -1627,6 +1666,8 @@ int wacom_setup_input_capabilities(struct input_dev *input_dev,
 
 	case INTUOS5:
 	case INTUOS5L:
+	case INTUOSPM:
+	case INTUOSPL:
 		if (features->device_type == BTN_TOOL_PEN) {
 			__set_bit(BTN_7, input_dev->keybit);
 			__set_bit(BTN_8, input_dev->keybit);
@@ -1634,6 +1675,7 @@ int wacom_setup_input_capabilities(struct input_dev *input_dev,
 		/* fall through */
 
 	case INTUOS5S:
+	case INTUOSPS:
 		__set_bit(INPUT_PROP_POINTER, input_dev->propbit);
 
 		if (features->device_type == BTN_TOOL_PEN) {
@@ -1765,6 +1807,24 @@ int wacom_setup_input_capabilities(struct input_dev *input_dev,
 					      0, 0);
 		}
 		break;
+
+	case CINTIQ_HYBRID:
+		__set_bit(BTN_1, input_dev->keybit);
+		__set_bit(BTN_2, input_dev->keybit);
+		__set_bit(BTN_3, input_dev->keybit);
+		__set_bit(BTN_4, input_dev->keybit);
+
+		__set_bit(BTN_5, input_dev->keybit);
+		__set_bit(BTN_6, input_dev->keybit);
+		__set_bit(BTN_7, input_dev->keybit);
+		__set_bit(BTN_8, input_dev->keybit);
+		__set_bit(BTN_0, input_dev->keybit);
+
+		input_set_abs_params(input_dev, ABS_Z, -900, 899, 0, 0);
+		__set_bit(INPUT_PROP_DIRECT, input_dev->propbit);
+
+		wacom_setup_cintiq(wacom_wac);
+		break;
 	}
 	return 0;
 }
@@ -1952,6 +2012,18 @@ static const struct wacom_features wacom_features_0x29 =
 static const struct wacom_features wacom_features_0x2A =
 	{ "Wacom Intuos5 M", WACOM_PKGLEN_INTUOS,  44704, 27940, 2047,
 	  63, INTUOS5, WACOM_INTUOS3_RES, WACOM_INTUOS3_RES };
+static const struct wacom_features wacom_features_0x314 =
+	{ "Wacom Intuos Pro S", WACOM_PKGLEN_INTUOS,  31496, 19685, 2047,
+	  63, INTUOSPS, WACOM_INTUOS3_RES, WACOM_INTUOS3_RES,
+	  .touch_max = 16 };
+static const struct wacom_features wacom_features_0x315 =
+	{ "Wacom Intuos Pro M", WACOM_PKGLEN_INTUOS,  44704, 27940, 2047,
+	  63, INTUOSPM, WACOM_INTUOS3_RES, WACOM_INTUOS3_RES,
+	  .touch_max = 16 };
+static const struct wacom_features wacom_features_0x317 =
+	{ "Wacom Intuos Pro L", WACOM_PKGLEN_INTUOS,  65024, 40640, 2047,
+	  63, INTUOSPL, WACOM_INTUOS3_RES, WACOM_INTUOS3_RES,
+	  .touch_max = 16 };
 static const struct wacom_features wacom_features_0xF4 =
 	{ "Wacom Cintiq 24HD",       WACOM_PKGLEN_INTUOS,   104480, 65600, 2047,
 	  63, WACOM_24HD, WACOM_INTUOS3_RES, WACOM_INTUOS3_RES };
@@ -2131,6 +2203,13 @@ static const struct wacom_features wacom_features_0x301 =
 static const struct wacom_features wacom_features_0x6004 =
 	{ "ISD-V4",               WACOM_PKGLEN_GRAPHIRE,  12800,  8000,  255,
 	  0, TABLETPC, WACOM_INTUOS_RES, WACOM_INTUOS_RES };
+static const struct wacom_features wacom_features_0x0307 =
+	{ "Wacom ISDv5 307", WACOM_PKGLEN_INTUOS,  59552,  33848, 2047,
+	  63, CINTIQ_HYBRID, WACOM_INTUOS3_RES, WACOM_INTUOS3_RES,
+	  .oVid = USB_VENDOR_ID_WACOM, .oPid = 0x309 };
+static const struct wacom_features wacom_features_0x0309 =
+	{ "Wacom ISDv5 309", .type = WACOM_24HDT, /* Touch */
+	  .oVid = USB_VENDOR_ID_WACOM, .oPid = 0x0307, .touch_max = 10 };
 
 #define USB_DEVICE_WACOM(prod)					\
 	USB_DEVICE(USB_VENDOR_ID_WACOM, prod),			\
@@ -2259,12 +2338,17 @@ const struct usb_device_id wacom_ids[] = {
 	{ USB_DEVICE_WACOM(0x300) },
 	{ USB_DEVICE_WACOM(0x301) },
 	{ USB_DEVICE_WACOM(0x304) },
+	{ USB_DEVICE_DETAILED(0x314, USB_CLASS_HID, 0, 0) },
+	{ USB_DEVICE_DETAILED(0x315, USB_CLASS_HID, 0, 0) },
+	{ USB_DEVICE_DETAILED(0x317, USB_CLASS_HID, 0, 0) },
 	{ USB_DEVICE_WACOM(0x4001) },
 	{ USB_DEVICE_WACOM(0x47) },
 	{ USB_DEVICE_WACOM(0xF4) },
 	{ USB_DEVICE_WACOM(0xF8) },
 	{ USB_DEVICE_DETAILED(0xF6, USB_CLASS_HID, 0, 0) },
 	{ USB_DEVICE_WACOM(0xFA) },
+	{ USB_DEVICE_WACOM(0x0307) },
+	{ USB_DEVICE_DETAILED(0x0309, USB_CLASS_HID, 0, 0) },
 	{ USB_DEVICE_LENOVO(0x6004) },
 	{ }
 };
diff --git a/drivers/input/tablet/wacom_wac.h b/drivers/input/tablet/wacom_wac.h
index dfc9e08e7f70..fd23a3790605 100644
--- a/drivers/input/tablet/wacom_wac.h
+++ b/drivers/input/tablet/wacom_wac.h
@@ -14,6 +14,8 @@
 /* maximum packet length for USB devices */
 #define WACOM_PKGLEN_MAX	64
 
+#define WACOM_NAME_MAX		64
+
 /* packet length for individual models */
 #define WACOM_PKGLEN_PENPRTN	 7
 #define WACOM_PKGLEN_GRAPHIRE	 8
@@ -76,10 +78,14 @@ enum {
 	INTUOS5S,
 	INTUOS5,
 	INTUOS5L,
+	INTUOSPS,
+	INTUOSPM,
+	INTUOSPL,
 	WACOM_21UX2,
 	WACOM_22HD,
 	DTK,
 	WACOM_24HD,
+	CINTIQ_HYBRID,
 	CINTIQ,
 	WACOM_BEE,
 	WACOM_13HD,
@@ -126,7 +132,7 @@ struct wacom_shared {
 };
 
 struct wacom_wac {
-	char name[64];
+	char name[WACOM_NAME_MAX];
 	unsigned char *data;
 	int tool[2];
 	int id[2];
diff --git a/drivers/input/touchscreen/Kconfig b/drivers/input/touchscreen/Kconfig
index e09ec67957a3..00d1e547b211 100644
--- a/drivers/input/touchscreen/Kconfig
+++ b/drivers/input/touchscreen/Kconfig
@@ -919,4 +919,17 @@ config TOUCHSCREEN_TPS6507X
 	  To compile this driver as a module, choose M here: the
 	  module will be called tps6507x_ts.
 
+config TOUCHSCREEN_ZFORCE
+	tristate "Neonode zForce infrared touchscreens"
+	depends on I2C
+	depends on GPIOLIB
+	help
+	  Say Y here if you have a touchscreen using the zforce
+	  infraread technology from Neonode.
+
+	  If unsure, say N.
+
+	  To compile this driver as a module, choose M here: the
+	  module will be called zforce_ts.
+
 endif
diff --git a/drivers/input/touchscreen/Makefile b/drivers/input/touchscreen/Makefile
index f5216c1bf53e..7587883b8d38 100644
--- a/drivers/input/touchscreen/Makefile
+++ b/drivers/input/touchscreen/Makefile
@@ -75,3 +75,4 @@ obj-$(CONFIG_TOUCHSCREEN_WM97XX_MAINSTONE)	+= mainstone-wm97xx.o
 obj-$(CONFIG_TOUCHSCREEN_WM97XX_ZYLONITE)	+= zylonite-wm97xx.o
 obj-$(CONFIG_TOUCHSCREEN_W90X900)	+= w90p910_ts.o
 obj-$(CONFIG_TOUCHSCREEN_TPS6507X)	+= tps6507x-ts.o
+obj-$(CONFIG_TOUCHSCREEN_ZFORCE)	+= zforce_ts.o
diff --git a/drivers/input/touchscreen/ad7877.c b/drivers/input/touchscreen/ad7877.c
index f3a174a83c82..69834dd3c313 100644
--- a/drivers/input/touchscreen/ad7877.c
+++ b/drivers/input/touchscreen/ad7877.c
@@ -806,7 +806,6 @@ err_free_irq:
 err_free_mem:
 	input_free_device(input_dev);
 	kfree(ts);
-	spi_set_drvdata(spi, NULL);
 	return err;
 }
 
@@ -823,7 +822,6 @@ static int ad7877_remove(struct spi_device *spi)
 	kfree(ts);
 
 	dev_dbg(&spi->dev, "unregistered touchscreen\n");
-	spi_set_drvdata(spi, NULL);
 
 	return 0;
 }
diff --git a/drivers/input/touchscreen/ad7879-spi.c b/drivers/input/touchscreen/ad7879-spi.c
index 606da5bd6115..1a7b1143536e 100644
--- a/drivers/input/touchscreen/ad7879-spi.c
+++ b/drivers/input/touchscreen/ad7879-spi.c
@@ -142,7 +142,6 @@ static int ad7879_spi_remove(struct spi_device *spi)
 	struct ad7879 *ts = spi_get_drvdata(spi);
 
 	ad7879_remove(ts);
-	spi_set_drvdata(spi, NULL);
 
 	return 0;
 }
diff --git a/drivers/input/touchscreen/cyttsp4_core.c b/drivers/input/touchscreen/cyttsp4_core.c
index d038575f49db..42d830efa316 100644
--- a/drivers/input/touchscreen/cyttsp4_core.c
+++ b/drivers/input/touchscreen/cyttsp4_core.c
@@ -2113,7 +2113,6 @@ error_startup:
 error_request_irq:
 	if (cd->cpdata->init)
 		cd->cpdata->init(cd->cpdata, 0, dev);
-	dev_set_drvdata(dev, NULL);
 error_free_xfer:
 	kfree(cd->xfer_buf);
 error_free_cd:
@@ -2151,7 +2150,6 @@ int cyttsp4_remove(struct cyttsp4 *cd)
 	free_irq(cd->irq, cd);
 	if (cd->cpdata->init)
 		cd->cpdata->init(cd->cpdata, 0, dev);
-	dev_set_drvdata(dev, NULL);
 	cyttsp4_free_si_ptrs(cd);
 	kfree(cd);
 	return 0;
diff --git a/drivers/input/touchscreen/cyttsp4_spi.c b/drivers/input/touchscreen/cyttsp4_spi.c
index a71e1141d638..b19434cebbf6 100644
--- a/drivers/input/touchscreen/cyttsp4_spi.c
+++ b/drivers/input/touchscreen/cyttsp4_spi.c
@@ -171,10 +171,7 @@ static int cyttsp4_spi_probe(struct spi_device *spi)
 	ts = cyttsp4_probe(&cyttsp_spi_bus_ops, &spi->dev, spi->irq,
 			  CY_SPI_DATA_BUF_SIZE);
 
-	if (IS_ERR(ts))
-		return PTR_ERR(ts);
-
-	return 0;
+	return PTR_ERR_OR_ZERO(ts);
 }
 
 static int cyttsp4_spi_remove(struct spi_device *spi)
diff --git a/drivers/input/touchscreen/cyttsp_core.c b/drivers/input/touchscreen/cyttsp_core.c
index d53e0b72a407..4204841cdc49 100644
--- a/drivers/input/touchscreen/cyttsp_core.c
+++ b/drivers/input/touchscreen/cyttsp_core.c
@@ -242,7 +242,7 @@ static int cyttsp_soft_reset(struct cyttsp *ts)
 	int retval;
 
 	/* wait for interrupt to set ready completion */
-	INIT_COMPLETION(ts->bl_ready);
+	reinit_completion(&ts->bl_ready);
 	ts->state = CY_BL_STATE;
 
 	enable_irq(ts->irq);
diff --git a/drivers/input/touchscreen/egalax_ts.c b/drivers/input/touchscreen/egalax_ts.c
index ef5fcb0945e9..054d22583248 100644
--- a/drivers/input/touchscreen/egalax_ts.c
+++ b/drivers/input/touchscreen/egalax_ts.c
@@ -273,7 +273,7 @@ static struct i2c_driver egalax_ts_driver = {
 		.name	= "egalax_ts",
 		.owner	= THIS_MODULE,
 		.pm	= &egalax_ts_pm_ops,
-		.of_match_table	= of_match_ptr(egalax_ts_dt_ids),
+		.of_match_table	= egalax_ts_dt_ids,
 	},
 	.id_table	= egalax_ts_id,
 	.probe		= egalax_ts_probe,
diff --git a/drivers/input/touchscreen/htcpen.c b/drivers/input/touchscreen/htcpen.c
index 66500852341b..92e2243fb77d 100644
--- a/drivers/input/touchscreen/htcpen.c
+++ b/drivers/input/touchscreen/htcpen.c
@@ -186,8 +186,6 @@ static int htcpen_isa_remove(struct device *dev, unsigned int id)
 	release_region(HTCPEN_PORT_INIT, 1);
 	release_region(HTCPEN_PORT_IRQ_CLEAR, 1);
 
-	dev_set_drvdata(dev, NULL);
-
 	return 0;
 }
 
diff --git a/drivers/input/touchscreen/st1232.c b/drivers/input/touchscreen/st1232.c
index 1740a2496371..2f03b2f289dd 100644
--- a/drivers/input/touchscreen/st1232.c
+++ b/drivers/input/touchscreen/st1232.c
@@ -24,6 +24,7 @@
 #include <linux/input.h>
 #include <linux/interrupt.h>
 #include <linux/module.h>
+#include <linux/of.h>
 #include <linux/of_gpio.h>
 #include <linux/pm_qos.h>
 #include <linux/slab.h>
diff --git a/drivers/input/touchscreen/ti_am335x_tsc.c b/drivers/input/touchscreen/ti_am335x_tsc.c
index 24e625c0b531..68beadaabceb 100644
--- a/drivers/input/touchscreen/ti_am335x_tsc.c
+++ b/drivers/input/touchscreen/ti_am335x_tsc.c
@@ -354,9 +354,16 @@ static int titsc_parse_dt(struct platform_device *pdev,
 	if (err < 0)
 		return err;
 
-	err = of_property_read_u32(node, "ti,coordiante-readouts",
+	/*
+	 * Try with the new binding first. If it fails, try again with
+	 * bogus, miss-spelled version.
+	 */
+	err = of_property_read_u32(node, "ti,coordinate-readouts",
 			&ts_dev->coordinate_readouts);
 	if (err < 0)
+		err = of_property_read_u32(node, "ti,coordiante-readouts",
+				&ts_dev->coordinate_readouts);
+	if (err < 0)
 		return err;
 
 	return of_property_read_u32_array(node, "ti,wire-config",
@@ -511,7 +518,7 @@ static struct platform_driver ti_tsc_driver = {
 		.name   = "TI-am335x-tsc",
 		.owner	= THIS_MODULE,
 		.pm	= TITSC_PM_OPS,
-		.of_match_table = of_match_ptr(ti_tsc_dt_ids),
+		.of_match_table = ti_tsc_dt_ids,
 	},
 };
 module_platform_driver(ti_tsc_driver);
diff --git a/drivers/input/touchscreen/tsc2005.c b/drivers/input/touchscreen/tsc2005.c
index 7213e8b07e79..811353353917 100644
--- a/drivers/input/touchscreen/tsc2005.c
+++ b/drivers/input/touchscreen/tsc2005.c
@@ -678,7 +678,6 @@ static int tsc2005_probe(struct spi_device *spi)
 err_remove_sysfs:
 	sysfs_remove_group(&spi->dev.kobj, &tsc2005_attr_group);
 err_clear_drvdata:
-	spi_set_drvdata(spi, NULL);
 	free_irq(spi->irq, ts);
 err_free_mem:
 	input_free_device(input_dev);
@@ -696,7 +695,6 @@ static int tsc2005_remove(struct spi_device *spi)
 	input_unregister_device(ts->idev);
 	kfree(ts);
 
-	spi_set_drvdata(spi, NULL);
 	return 0;
 }
 
diff --git a/drivers/input/touchscreen/usbtouchscreen.c b/drivers/input/touchscreen/usbtouchscreen.c
index 721fdb3597ca..ae4b6b903629 100644
--- a/drivers/input/touchscreen/usbtouchscreen.c
+++ b/drivers/input/touchscreen/usbtouchscreen.c
@@ -146,12 +146,10 @@ enum {
 
 #define USB_DEVICE_HID_CLASS(vend, prod) \
 	.match_flags = USB_DEVICE_ID_MATCH_INT_CLASS \
-		| USB_DEVICE_ID_MATCH_INT_PROTOCOL \
 		| USB_DEVICE_ID_MATCH_DEVICE, \
 	.idVendor = (vend), \
 	.idProduct = (prod), \
-	.bInterfaceClass = USB_INTERFACE_CLASS_HID, \
-	.bInterfaceProtocol = USB_INTERFACE_PROTOCOL_MOUSE
+	.bInterfaceClass = USB_INTERFACE_CLASS_HID
 
 static const struct usb_device_id usbtouch_devices[] = {
 #ifdef CONFIG_TOUCHSCREEN_USB_EGALAX
diff --git a/drivers/input/touchscreen/zforce_ts.c b/drivers/input/touchscreen/zforce_ts.c
new file mode 100644
index 000000000000..75762d6ff3ba
--- /dev/null
+++ b/drivers/input/touchscreen/zforce_ts.c
@@ -0,0 +1,836 @@
+/*
+ * Copyright (C) 2012-2013 MundoReader S.L.
+ * Author: Heiko Stuebner <heiko@sntech.de>
+ *
+ * based in parts on Nook zforce driver
+ *
+ * Copyright (C) 2010 Barnes & Noble, Inc.
+ * Author: Pieter Truter<ptruter@intrinsyc.com>
+ *
+ * This software is licensed under the terms of the GNU General Public
+ * License version 2, as published by the Free Software Foundation, and
+ * may be copied, distributed, and modified under those terms.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+ * GNU General Public License for more details.
+ */
+
+#include <linux/module.h>
+#include <linux/hrtimer.h>
+#include <linux/slab.h>
+#include <linux/input.h>
+#include <linux/interrupt.h>
+#include <linux/i2c.h>
+#include <linux/delay.h>
+#include <linux/gpio.h>
+#include <linux/device.h>
+#include <linux/sysfs.h>
+#include <linux/input/mt.h>
+#include <linux/platform_data/zforce_ts.h>
+
+#define WAIT_TIMEOUT		msecs_to_jiffies(1000)
+
+#define FRAME_START		0xee
+
+/* Offsets of the different parts of the payload the controller sends */
+#define PAYLOAD_HEADER		0
+#define PAYLOAD_LENGTH		1
+#define PAYLOAD_BODY		2
+
+/* Response offsets */
+#define RESPONSE_ID		0
+#define RESPONSE_DATA		1
+
+/* Commands */
+#define COMMAND_DEACTIVATE	0x00
+#define COMMAND_INITIALIZE	0x01
+#define COMMAND_RESOLUTION	0x02
+#define COMMAND_SETCONFIG	0x03
+#define COMMAND_DATAREQUEST	0x04
+#define COMMAND_SCANFREQ	0x08
+#define COMMAND_STATUS		0X1e
+
+/*
+ * Responses the controller sends as a result of
+ * command requests
+ */
+#define RESPONSE_DEACTIVATE	0x00
+#define RESPONSE_INITIALIZE	0x01
+#define RESPONSE_RESOLUTION	0x02
+#define RESPONSE_SETCONFIG	0x03
+#define RESPONSE_SCANFREQ	0x08
+#define RESPONSE_STATUS		0X1e
+
+/*
+ * Notifications are send by the touch controller without
+ * being requested by the driver and include for example
+ * touch indications
+ */
+#define NOTIFICATION_TOUCH		0x04
+#define NOTIFICATION_BOOTCOMPLETE	0x07
+#define NOTIFICATION_OVERRUN		0x25
+#define NOTIFICATION_PROXIMITY		0x26
+#define NOTIFICATION_INVALID_COMMAND	0xfe
+
+#define ZFORCE_REPORT_POINTS		2
+#define ZFORCE_MAX_AREA			0xff
+
+#define STATE_DOWN			0
+#define STATE_MOVE			1
+#define STATE_UP			2
+
+#define SETCONFIG_DUALTOUCH		(1 << 0)
+
+struct zforce_point {
+	int coord_x;
+	int coord_y;
+	int state;
+	int id;
+	int area_major;
+	int area_minor;
+	int orientation;
+	int pressure;
+	int prblty;
+};
+
+/*
+ * @client		the i2c_client
+ * @input		the input device
+ * @suspending		in the process of going to suspend (don't emit wakeup
+ *			events for commands executed to suspend the device)
+ * @suspended		device suspended
+ * @access_mutex	serialize i2c-access, to keep multipart reads together
+ * @command_done	completion to wait for the command result
+ * @command_mutex	serialize commands send to the ic
+ * @command_waiting	the id of the command that that is currently waiting
+ *			for a result
+ * @command_result	returned result of the command
+ */
+struct zforce_ts {
+	struct i2c_client	*client;
+	struct input_dev	*input;
+	const struct zforce_ts_platdata *pdata;
+	char			phys[32];
+
+	bool			suspending;
+	bool			suspended;
+	bool			boot_complete;
+
+	/* Firmware version information */
+	u16			version_major;
+	u16			version_minor;
+	u16			version_build;
+	u16			version_rev;
+
+	struct mutex		access_mutex;
+
+	struct completion	command_done;
+	struct mutex		command_mutex;
+	int			command_waiting;
+	int			command_result;
+};
+
+static int zforce_command(struct zforce_ts *ts, u8 cmd)
+{
+	struct i2c_client *client = ts->client;
+	char buf[3];
+	int ret;
+
+	dev_dbg(&client->dev, "%s: 0x%x\n", __func__, cmd);
+
+	buf[0] = FRAME_START;
+	buf[1] = 1; /* data size, command only */
+	buf[2] = cmd;
+
+	mutex_lock(&ts->access_mutex);
+	ret = i2c_master_send(client, &buf[0], ARRAY_SIZE(buf));
+	mutex_unlock(&ts->access_mutex);
+	if (ret < 0) {
+		dev_err(&client->dev, "i2c send data request error: %d\n", ret);
+		return ret;
+	}
+
+	return 0;
+}
+
+static int zforce_send_wait(struct zforce_ts *ts, const char *buf, int len)
+{
+	struct i2c_client *client = ts->client;
+	int ret;
+
+	ret = mutex_trylock(&ts->command_mutex);
+	if (!ret) {
+		dev_err(&client->dev, "already waiting for a command\n");
+		return -EBUSY;
+	}
+
+	dev_dbg(&client->dev, "sending %d bytes for command 0x%x\n",
+		buf[1], buf[2]);
+
+	ts->command_waiting = buf[2];
+
+	mutex_lock(&ts->access_mutex);
+	ret = i2c_master_send(client, buf, len);
+	mutex_unlock(&ts->access_mutex);
+	if (ret < 0) {
+		dev_err(&client->dev, "i2c send data request error: %d\n", ret);
+		goto unlock;
+	}
+
+	dev_dbg(&client->dev, "waiting for result for command 0x%x\n", buf[2]);
+
+	if (wait_for_completion_timeout(&ts->command_done, WAIT_TIMEOUT) == 0) {
+		ret = -ETIME;
+		goto unlock;
+	}
+
+	ret = ts->command_result;
+
+unlock:
+	mutex_unlock(&ts->command_mutex);
+	return ret;
+}
+
+static int zforce_command_wait(struct zforce_ts *ts, u8 cmd)
+{
+	struct i2c_client *client = ts->client;
+	char buf[3];
+	int ret;
+
+	dev_dbg(&client->dev, "%s: 0x%x\n", __func__, cmd);
+
+	buf[0] = FRAME_START;
+	buf[1] = 1; /* data size, command only */
+	buf[2] = cmd;
+
+	ret = zforce_send_wait(ts, &buf[0], ARRAY_SIZE(buf));
+	if (ret < 0) {
+		dev_err(&client->dev, "i2c send data request error: %d\n", ret);
+		return ret;
+	}
+
+	return 0;
+}
+
+static int zforce_resolution(struct zforce_ts *ts, u16 x, u16 y)
+{
+	struct i2c_client *client = ts->client;
+	char buf[7] = { FRAME_START, 5, COMMAND_RESOLUTION,
+			(x & 0xff), ((x >> 8) & 0xff),
+			(y & 0xff), ((y >> 8) & 0xff) };
+
+	dev_dbg(&client->dev, "set resolution to (%d,%d)\n", x, y);
+
+	return zforce_send_wait(ts, &buf[0], ARRAY_SIZE(buf));
+}
+
+static int zforce_scan_frequency(struct zforce_ts *ts, u16 idle, u16 finger,
+				 u16 stylus)
+{
+	struct i2c_client *client = ts->client;
+	char buf[9] = { FRAME_START, 7, COMMAND_SCANFREQ,
+			(idle & 0xff), ((idle >> 8) & 0xff),
+			(finger & 0xff), ((finger >> 8) & 0xff),
+			(stylus & 0xff), ((stylus >> 8) & 0xff) };
+
+	dev_dbg(&client->dev, "set scan frequency to (idle: %d, finger: %d, stylus: %d)\n",
+		idle, finger, stylus);
+
+	return zforce_send_wait(ts, &buf[0], ARRAY_SIZE(buf));
+}
+
+static int zforce_setconfig(struct zforce_ts *ts, char b1)
+{
+	struct i2c_client *client = ts->client;
+	char buf[7] = { FRAME_START, 5, COMMAND_SETCONFIG,
+			b1, 0, 0, 0 };
+
+	dev_dbg(&client->dev, "set config to (%d)\n", b1);
+
+	return zforce_send_wait(ts, &buf[0], ARRAY_SIZE(buf));
+}
+
+static int zforce_start(struct zforce_ts *ts)
+{
+	struct i2c_client *client = ts->client;
+	const struct zforce_ts_platdata *pdata = dev_get_platdata(&client->dev);
+	int ret;
+
+	dev_dbg(&client->dev, "starting device\n");
+
+	ret = zforce_command_wait(ts, COMMAND_INITIALIZE);
+	if (ret) {
+		dev_err(&client->dev, "Unable to initialize, %d\n", ret);
+		return ret;
+	}
+
+	ret = zforce_resolution(ts, pdata->x_max, pdata->y_max);
+	if (ret) {
+		dev_err(&client->dev, "Unable to set resolution, %d\n", ret);
+		goto error;
+	}
+
+	ret = zforce_scan_frequency(ts, 10, 50, 50);
+	if (ret) {
+		dev_err(&client->dev, "Unable to set scan frequency, %d\n",
+			ret);
+		goto error;
+	}
+
+	if (zforce_setconfig(ts, SETCONFIG_DUALTOUCH)) {
+		dev_err(&client->dev, "Unable to set config\n");
+		goto error;
+	}
+
+	/* start sending touch events */
+	ret = zforce_command(ts, COMMAND_DATAREQUEST);
+	if (ret) {
+		dev_err(&client->dev, "Unable to request data\n");
+		goto error;
+	}
+
+	/*
+	 * Per NN, initial cal. take max. of 200msec.
+	 * Allow time to complete this calibration
+	 */
+	msleep(200);
+
+	return 0;
+
+error:
+	zforce_command_wait(ts, COMMAND_DEACTIVATE);
+	return ret;
+}
+
+static int zforce_stop(struct zforce_ts *ts)
+{
+	struct i2c_client *client = ts->client;
+	int ret;
+
+	dev_dbg(&client->dev, "stopping device\n");
+
+	/* Deactivates touch sensing and puts the device into sleep. */
+	ret = zforce_command_wait(ts, COMMAND_DEACTIVATE);
+	if (ret != 0) {
+		dev_err(&client->dev, "could not deactivate device, %d\n",
+			ret);
+		return ret;
+	}
+
+	return 0;
+}
+
+static int zforce_touch_event(struct zforce_ts *ts, u8 *payload)
+{
+	struct i2c_client *client = ts->client;
+	const struct zforce_ts_platdata *pdata = dev_get_platdata(&client->dev);
+	struct zforce_point point;
+	int count, i, num = 0;
+
+	count = payload[0];
+	if (count > ZFORCE_REPORT_POINTS) {
+		dev_warn(&client->dev, "to many coordinates %d, expected max %d\n",
+			 count, ZFORCE_REPORT_POINTS);
+		count = ZFORCE_REPORT_POINTS;
+	}
+
+	for (i = 0; i < count; i++) {
+		point.coord_x =
+			payload[9 * i + 2] << 8 | payload[9 * i + 1];
+		point.coord_y =
+			payload[9 * i + 4] << 8 | payload[9 * i + 3];
+
+		if (point.coord_x > pdata->x_max ||
+		    point.coord_y > pdata->y_max) {
+			dev_warn(&client->dev, "coordinates (%d,%d) invalid\n",
+				point.coord_x, point.coord_y);
+			point.coord_x = point.coord_y = 0;
+		}
+
+		point.state = payload[9 * i + 5] & 0x03;
+		point.id = (payload[9 * i + 5] & 0xfc) >> 2;
+
+		/* determine touch major, minor and orientation */
+		point.area_major = max(payload[9 * i + 6],
+					  payload[9 * i + 7]);
+		point.area_minor = min(payload[9 * i + 6],
+					  payload[9 * i + 7]);
+		point.orientation = payload[9 * i + 6] > payload[9 * i + 7];
+
+		point.pressure = payload[9 * i + 8];
+		point.prblty = payload[9 * i + 9];
+
+		dev_dbg(&client->dev,
+			"point %d/%d: state %d, id %d, pressure %d, prblty %d, x %d, y %d, amajor %d, aminor %d, ori %d\n",
+			i, count, point.state, point.id,
+			point.pressure, point.prblty,
+			point.coord_x, point.coord_y,
+			point.area_major, point.area_minor,
+			point.orientation);
+
+		/* the zforce id starts with "1", so needs to be decreased */
+		input_mt_slot(ts->input, point.id - 1);
+
+		input_mt_report_slot_state(ts->input, MT_TOOL_FINGER,
+						point.state != STATE_UP);
+
+		if (point.state != STATE_UP) {
+			input_report_abs(ts->input, ABS_MT_POSITION_X,
+					 point.coord_x);
+			input_report_abs(ts->input, ABS_MT_POSITION_Y,
+					 point.coord_y);
+			input_report_abs(ts->input, ABS_MT_TOUCH_MAJOR,
+					 point.area_major);
+			input_report_abs(ts->input, ABS_MT_TOUCH_MINOR,
+					 point.area_minor);
+			input_report_abs(ts->input, ABS_MT_ORIENTATION,
+					 point.orientation);
+			num++;
+		}
+	}
+
+	input_mt_sync_frame(ts->input);
+
+	input_mt_report_finger_count(ts->input, num);
+
+	input_sync(ts->input);
+
+	return 0;
+}
+
+static int zforce_read_packet(struct zforce_ts *ts, u8 *buf)
+{
+	struct i2c_client *client = ts->client;
+	int ret;
+
+	mutex_lock(&ts->access_mutex);
+
+	/* read 2 byte message header */
+	ret = i2c_master_recv(client, buf, 2);
+	if (ret < 0) {
+		dev_err(&client->dev, "error reading header: %d\n", ret);
+		goto unlock;
+	}
+
+	if (buf[PAYLOAD_HEADER] != FRAME_START) {
+		dev_err(&client->dev, "invalid frame start: %d\n", buf[0]);
+		ret = -EIO;
+		goto unlock;
+	}
+
+	if (buf[PAYLOAD_LENGTH] <= 0 || buf[PAYLOAD_LENGTH] > 255) {
+		dev_err(&client->dev, "invalid payload length: %d\n",
+			buf[PAYLOAD_LENGTH]);
+		ret = -EIO;
+		goto unlock;
+	}
+
+	/* read the message */
+	ret = i2c_master_recv(client, &buf[PAYLOAD_BODY], buf[PAYLOAD_LENGTH]);
+	if (ret < 0) {
+		dev_err(&client->dev, "error reading payload: %d\n", ret);
+		goto unlock;
+	}
+
+	dev_dbg(&client->dev, "read %d bytes for response command 0x%x\n",
+		buf[PAYLOAD_LENGTH], buf[PAYLOAD_BODY]);
+
+unlock:
+	mutex_unlock(&ts->access_mutex);
+	return ret;
+}
+
+static void zforce_complete(struct zforce_ts *ts, int cmd, int result)
+{
+	struct i2c_client *client = ts->client;
+
+	if (ts->command_waiting == cmd) {
+		dev_dbg(&client->dev, "completing command 0x%x\n", cmd);
+		ts->command_result = result;
+		complete(&ts->command_done);
+	} else {
+		dev_dbg(&client->dev, "command %d not for us\n", cmd);
+	}
+}
+
+static irqreturn_t zforce_interrupt(int irq, void *dev_id)
+{
+	struct zforce_ts *ts = dev_id;
+	struct i2c_client *client = ts->client;
+	const struct zforce_ts_platdata *pdata = dev_get_platdata(&client->dev);
+	int ret;
+	u8 payload_buffer[512];
+	u8 *payload;
+
+	/*
+	 * When suspended, emit a wakeup signal if necessary and return.
+	 * Due to the level-interrupt we will get re-triggered later.
+	 */
+	if (ts->suspended) {
+		if (device_may_wakeup(&client->dev))
+			pm_wakeup_event(&client->dev, 500);
+		msleep(20);
+		return IRQ_HANDLED;
+	}
+
+	dev_dbg(&client->dev, "handling interrupt\n");
+
+	/* Don't emit wakeup events from commands run by zforce_suspend */
+	if (!ts->suspending && device_may_wakeup(&client->dev))
+		pm_stay_awake(&client->dev);
+
+	while (!gpio_get_value(pdata->gpio_int)) {
+		ret = zforce_read_packet(ts, payload_buffer);
+		if (ret < 0) {
+			dev_err(&client->dev, "could not read packet, ret: %d\n",
+				ret);
+			break;
+		}
+
+		payload =  &payload_buffer[PAYLOAD_BODY];
+
+		switch (payload[RESPONSE_ID]) {
+		case NOTIFICATION_TOUCH:
+			/*
+			 * Always report touch-events received while
+			 * suspending, when being a wakeup source
+			 */
+			if (ts->suspending && device_may_wakeup(&client->dev))
+				pm_wakeup_event(&client->dev, 500);
+			zforce_touch_event(ts, &payload[RESPONSE_DATA]);
+			break;
+
+		case NOTIFICATION_BOOTCOMPLETE:
+			ts->boot_complete = payload[RESPONSE_DATA];
+			zforce_complete(ts, payload[RESPONSE_ID], 0);
+			break;
+
+		case RESPONSE_INITIALIZE:
+		case RESPONSE_DEACTIVATE:
+		case RESPONSE_SETCONFIG:
+		case RESPONSE_RESOLUTION:
+		case RESPONSE_SCANFREQ:
+			zforce_complete(ts, payload[RESPONSE_ID],
+					payload[RESPONSE_DATA]);
+			break;
+
+		case RESPONSE_STATUS:
+			/*
+			 * Version Payload Results
+			 * [2:major] [2:minor] [2:build] [2:rev]
+			 */
+			ts->version_major = (payload[RESPONSE_DATA + 1] << 8) |
+						payload[RESPONSE_DATA];
+			ts->version_minor = (payload[RESPONSE_DATA + 3] << 8) |
+						payload[RESPONSE_DATA + 2];
+			ts->version_build = (payload[RESPONSE_DATA + 5] << 8) |
+						payload[RESPONSE_DATA + 4];
+			ts->version_rev   = (payload[RESPONSE_DATA + 7] << 8) |
+						payload[RESPONSE_DATA + 6];
+			dev_dbg(&ts->client->dev, "Firmware Version %04x:%04x %04x:%04x\n",
+				ts->version_major, ts->version_minor,
+				ts->version_build, ts->version_rev);
+
+			zforce_complete(ts, payload[RESPONSE_ID], 0);
+			break;
+
+		case NOTIFICATION_INVALID_COMMAND:
+			dev_err(&ts->client->dev, "invalid command: 0x%x\n",
+				payload[RESPONSE_DATA]);
+			break;
+
+		default:
+			dev_err(&ts->client->dev, "unrecognized response id: 0x%x\n",
+				payload[RESPONSE_ID]);
+			break;
+		}
+	}
+
+	if (!ts->suspending && device_may_wakeup(&client->dev))
+		pm_relax(&client->dev);
+
+	dev_dbg(&client->dev, "finished interrupt\n");
+
+	return IRQ_HANDLED;
+}
+
+static int zforce_input_open(struct input_dev *dev)
+{
+	struct zforce_ts *ts = input_get_drvdata(dev);
+	int ret;
+
+	ret = zforce_start(ts);
+	if (ret)
+		return ret;
+
+	return 0;
+}
+
+static void zforce_input_close(struct input_dev *dev)
+{
+	struct zforce_ts *ts = input_get_drvdata(dev);
+	struct i2c_client *client = ts->client;
+	int ret;
+
+	ret = zforce_stop(ts);
+	if (ret)
+		dev_warn(&client->dev, "stopping zforce failed\n");
+
+	return;
+}
+
+#ifdef CONFIG_PM_SLEEP
+static int zforce_suspend(struct device *dev)
+{
+	struct i2c_client *client = to_i2c_client(dev);
+	struct zforce_ts *ts = i2c_get_clientdata(client);
+	struct input_dev *input = ts->input;
+	int ret = 0;
+
+	mutex_lock(&input->mutex);
+	ts->suspending = true;
+
+	/*
+	 * When configured as a wakeup source device should always wake
+	 * the system, therefore start device if necessary.
+	 */
+	if (device_may_wakeup(&client->dev)) {
+		dev_dbg(&client->dev, "suspend while being a wakeup source\n");
+
+		/* Need to start device, if not open, to be a wakeup source. */
+		if (!input->users) {
+			ret = zforce_start(ts);
+			if (ret)
+				goto unlock;
+		}
+
+		enable_irq_wake(client->irq);
+	} else if (input->users) {
+		dev_dbg(&client->dev, "suspend without being a wakeup source\n");
+
+		ret = zforce_stop(ts);
+		if (ret)
+			goto unlock;
+
+		disable_irq(client->irq);
+	}
+
+	ts->suspended = true;
+
+unlock:
+	ts->suspending = false;
+	mutex_unlock(&input->mutex);
+
+	return ret;
+}
+
+static int zforce_resume(struct device *dev)
+{
+	struct i2c_client *client = to_i2c_client(dev);
+	struct zforce_ts *ts = i2c_get_clientdata(client);
+	struct input_dev *input = ts->input;
+	int ret = 0;
+
+	mutex_lock(&input->mutex);
+
+	ts->suspended = false;
+
+	if (device_may_wakeup(&client->dev)) {
+		dev_dbg(&client->dev, "resume from being a wakeup source\n");
+
+		disable_irq_wake(client->irq);
+
+		/* need to stop device if it was not open on suspend */
+		if (!input->users) {
+			ret = zforce_stop(ts);
+			if (ret)
+				goto unlock;
+		}
+	} else if (input->users) {
+		dev_dbg(&client->dev, "resume without being a wakeup source\n");
+
+		enable_irq(client->irq);
+
+		ret = zforce_start(ts);
+		if (ret < 0)
+			goto unlock;
+	}
+
+unlock:
+	mutex_unlock(&input->mutex);
+
+	return ret;
+}
+#endif
+
+static SIMPLE_DEV_PM_OPS(zforce_pm_ops, zforce_suspend, zforce_resume);
+
+static void zforce_reset(void *data)
+{
+	struct zforce_ts *ts = data;
+
+	gpio_set_value(ts->pdata->gpio_rst, 0);
+}
+
+static int zforce_probe(struct i2c_client *client,
+			const struct i2c_device_id *id)
+{
+	const struct zforce_ts_platdata *pdata = dev_get_platdata(&client->dev);
+	struct zforce_ts *ts;
+	struct input_dev *input_dev;
+	int ret;
+
+	if (!pdata)
+		return -EINVAL;
+
+	ts = devm_kzalloc(&client->dev, sizeof(struct zforce_ts), GFP_KERNEL);
+	if (!ts)
+		return -ENOMEM;
+
+	ret = devm_gpio_request_one(&client->dev, pdata->gpio_int, GPIOF_IN,
+				    "zforce_ts_int");
+	if (ret) {
+		dev_err(&client->dev, "request of gpio %d failed, %d\n",
+			pdata->gpio_int, ret);
+		return ret;
+	}
+
+	ret = devm_gpio_request_one(&client->dev, pdata->gpio_rst,
+				    GPIOF_OUT_INIT_LOW, "zforce_ts_rst");
+	if (ret) {
+		dev_err(&client->dev, "request of gpio %d failed, %d\n",
+			pdata->gpio_rst, ret);
+		return ret;
+	}
+
+	ret = devm_add_action(&client->dev, zforce_reset, ts);
+	if (ret) {
+		dev_err(&client->dev, "failed to register reset action, %d\n",
+			ret);
+		return ret;
+	}
+
+	snprintf(ts->phys, sizeof(ts->phys),
+		 "%s/input0", dev_name(&client->dev));
+
+	input_dev = devm_input_allocate_device(&client->dev);
+	if (!input_dev) {
+		dev_err(&client->dev, "could not allocate input device\n");
+		return -ENOMEM;
+	}
+
+	mutex_init(&ts->access_mutex);
+	mutex_init(&ts->command_mutex);
+
+	ts->pdata = pdata;
+	ts->client = client;
+	ts->input = input_dev;
+
+	input_dev->name = "Neonode zForce touchscreen";
+	input_dev->phys = ts->phys;
+	input_dev->id.bustype = BUS_I2C;
+
+	input_dev->open = zforce_input_open;
+	input_dev->close = zforce_input_close;
+
+	__set_bit(EV_KEY, input_dev->evbit);
+	__set_bit(EV_SYN, input_dev->evbit);
+	__set_bit(EV_ABS, input_dev->evbit);
+
+	/* For multi touch */
+	input_set_abs_params(input_dev, ABS_MT_POSITION_X, 0,
+			     pdata->x_max, 0, 0);
+	input_set_abs_params(input_dev, ABS_MT_POSITION_Y, 0,
+			     pdata->y_max, 0, 0);
+
+	input_set_abs_params(input_dev, ABS_MT_TOUCH_MAJOR, 0,
+			     ZFORCE_MAX_AREA, 0, 0);
+	input_set_abs_params(input_dev, ABS_MT_TOUCH_MINOR, 0,
+			     ZFORCE_MAX_AREA, 0, 0);
+	input_set_abs_params(input_dev, ABS_MT_ORIENTATION, 0, 1, 0, 0);
+	input_mt_init_slots(input_dev, ZFORCE_REPORT_POINTS, INPUT_MT_DIRECT);
+
+	input_set_drvdata(ts->input, ts);
+
+	init_completion(&ts->command_done);
+
+	/*
+	 * The zforce pulls the interrupt low when it has data ready.
+	 * After it is triggered the isr thread runs until all the available
+	 * packets have been read and the interrupt is high again.
+	 * Therefore we can trigger the interrupt anytime it is low and do
+	 * not need to limit it to the interrupt edge.
+	 */
+	ret = devm_request_threaded_irq(&client->dev, client->irq, NULL,
+					zforce_interrupt,
+					IRQF_TRIGGER_LOW | IRQF_ONESHOT,
+					input_dev->name, ts);
+	if (ret) {
+		dev_err(&client->dev, "irq %d request failed\n", client->irq);
+		return ret;
+	}
+
+	i2c_set_clientdata(client, ts);
+
+	/* let the controller boot */
+	gpio_set_value(pdata->gpio_rst, 1);
+
+	ts->command_waiting = NOTIFICATION_BOOTCOMPLETE;
+	if (wait_for_completion_timeout(&ts->command_done, WAIT_TIMEOUT) == 0)
+		dev_warn(&client->dev, "bootcomplete timed out\n");
+
+	/* need to start device to get version information */
+	ret = zforce_command_wait(ts, COMMAND_INITIALIZE);
+	if (ret) {
+		dev_err(&client->dev, "unable to initialize, %d\n", ret);
+		return ret;
+	}
+
+	/* this gets the firmware version among other informations */
+	ret = zforce_command_wait(ts, COMMAND_STATUS);
+	if (ret < 0) {
+		dev_err(&client->dev, "couldn't get status, %d\n", ret);
+		zforce_stop(ts);
+		return ret;
+	}
+
+	/* stop device and put it into sleep until it is opened */
+	ret = zforce_stop(ts);
+	if (ret < 0)
+		return ret;
+
+	device_set_wakeup_capable(&client->dev, true);
+
+	ret = input_register_device(input_dev);
+	if (ret) {
+		dev_err(&client->dev, "could not register input device, %d\n",
+			ret);
+		return ret;
+	}
+
+	return 0;
+}
+
+static struct i2c_device_id zforce_idtable[] = {
+	{ "zforce-ts", 0 },
+	{ }
+};
+MODULE_DEVICE_TABLE(i2c, zforce_idtable);
+
+static struct i2c_driver zforce_driver = {
+	.driver = {
+		.owner	= THIS_MODULE,
+		.name	= "zforce-ts",
+		.pm	= &zforce_pm_ops,
+	},
+	.probe		= zforce_probe,
+	.id_table	= zforce_idtable,
+};
+
+module_i2c_driver(zforce_driver);
+
+MODULE_AUTHOR("Heiko Stuebner <heiko@sntech.de>");
+MODULE_DESCRIPTION("zForce TouchScreen Driver");
+MODULE_LICENSE("GPL");
diff --git a/drivers/iommu/Kconfig b/drivers/iommu/Kconfig
index c880ebaf1553..3e7fdbb4916b 100644
--- a/drivers/iommu/Kconfig
+++ b/drivers/iommu/Kconfig
@@ -206,7 +206,7 @@ config SHMOBILE_IPMMU_TLB
 config SHMOBILE_IOMMU
 	bool "IOMMU for Renesas IPMMU/IPMMUI"
 	default n
-	depends on (ARM && ARCH_SHMOBILE)
+	depends on ARM
 	select IOMMU_API
 	select ARM_DMA_USE_IOMMU
 	select SHMOBILE_IPMMU
diff --git a/drivers/iommu/Makefile b/drivers/iommu/Makefile
index 14c1f474cf11..5d58bf16e9e3 100644
--- a/drivers/iommu/Makefile
+++ b/drivers/iommu/Makefile
@@ -1,4 +1,5 @@
 obj-$(CONFIG_IOMMU_API) += iommu.o
+obj-$(CONFIG_IOMMU_API) += iommu-traces.o
 obj-$(CONFIG_OF_IOMMU)	+= of_iommu.o
 obj-$(CONFIG_MSM_IOMMU) += msm_iommu.o msm_iommu_dev.o
 obj-$(CONFIG_AMD_IOMMU) += amd_iommu.o amd_iommu_init.o
diff --git a/drivers/iommu/arm-smmu.c b/drivers/iommu/arm-smmu.c
index 181c9ba929cd..1abfb5684ab7 100644
--- a/drivers/iommu/arm-smmu.c
+++ b/drivers/iommu/arm-smmu.c
@@ -590,6 +590,9 @@ static irqreturn_t arm_smmu_context_fault(int irq, void *dev)
 		ret = IRQ_HANDLED;
 		resume = RESUME_RETRY;
 	} else {
+		dev_err_ratelimited(smmu->dev,
+		    "Unhandled context fault: iova=0x%08lx, fsynr=0x%x, cb=%d\n",
+		    iova, fsynr, root_cfg->cbndx);
 		ret = IRQ_NONE;
 		resume = RESUME_TERMINATE;
 	}
@@ -778,7 +781,7 @@ static void arm_smmu_init_context_bank(struct arm_smmu_domain *smmu_domain)
 #ifdef __BIG_ENDIAN
 	reg |= SCTLR_E;
 #endif
-	writel(reg, cb_base + ARM_SMMU_CB_SCTLR);
+	writel_relaxed(reg, cb_base + ARM_SMMU_CB_SCTLR);
 }
 
 static int arm_smmu_init_domain_context(struct iommu_domain *domain,
@@ -1212,7 +1215,10 @@ static int arm_smmu_alloc_init_pte(struct arm_smmu_device *smmu, pmd_t *pmd,
 
 		arm_smmu_flush_pgtable(smmu, page_address(table),
 				       ARM_SMMU_PTE_HWTABLE_SIZE);
-		pgtable_page_ctor(table);
+		if (!pgtable_page_ctor(table)) {
+			__free_page(table);
+			return -ENOMEM;
+		}
 		pmd_populate(NULL, pmd, table);
 		arm_smmu_flush_pgtable(smmu, pmd, sizeof(*pmd));
 	}
@@ -1559,9 +1565,13 @@ static struct iommu_ops arm_smmu_ops = {
 static void arm_smmu_device_reset(struct arm_smmu_device *smmu)
 {
 	void __iomem *gr0_base = ARM_SMMU_GR0(smmu);
-	void __iomem *sctlr_base = ARM_SMMU_CB_BASE(smmu) + ARM_SMMU_CB_SCTLR;
+	void __iomem *cb_base;
 	int i = 0;
-	u32 scr0 = readl_relaxed(gr0_base + ARM_SMMU_GR0_sCR0);
+	u32 reg;
+
+	/* Clear Global FSR */
+	reg = readl_relaxed(gr0_base + ARM_SMMU_GR0_sGFSR);
+	writel(reg, gr0_base + ARM_SMMU_GR0_sGFSR);
 
 	/* Mark all SMRn as invalid and all S2CRn as bypass */
 	for (i = 0; i < smmu->num_mapping_groups; ++i) {
@@ -1569,33 +1579,38 @@ static void arm_smmu_device_reset(struct arm_smmu_device *smmu)
 		writel_relaxed(S2CR_TYPE_BYPASS, gr0_base + ARM_SMMU_GR0_S2CR(i));
 	}
 
-	/* Make sure all context banks are disabled */
-	for (i = 0; i < smmu->num_context_banks; ++i)
-		writel_relaxed(0, sctlr_base + ARM_SMMU_CB(smmu, i));
+	/* Make sure all context banks are disabled and clear CB_FSR  */
+	for (i = 0; i < smmu->num_context_banks; ++i) {
+		cb_base = ARM_SMMU_CB_BASE(smmu) + ARM_SMMU_CB(smmu, i);
+		writel_relaxed(0, cb_base + ARM_SMMU_CB_SCTLR);
+		writel_relaxed(FSR_FAULT, cb_base + ARM_SMMU_CB_FSR);
+	}
 
 	/* Invalidate the TLB, just in case */
 	writel_relaxed(0, gr0_base + ARM_SMMU_GR0_STLBIALL);
 	writel_relaxed(0, gr0_base + ARM_SMMU_GR0_TLBIALLH);
 	writel_relaxed(0, gr0_base + ARM_SMMU_GR0_TLBIALLNSNH);
 
+	reg = readl_relaxed(gr0_base + ARM_SMMU_GR0_sCR0);
+
 	/* Enable fault reporting */
-	scr0 |= (sCR0_GFRE | sCR0_GFIE | sCR0_GCFGFRE | sCR0_GCFGFIE);
+	reg |= (sCR0_GFRE | sCR0_GFIE | sCR0_GCFGFRE | sCR0_GCFGFIE);
 
 	/* Disable TLB broadcasting. */
-	scr0 |= (sCR0_VMIDPNE | sCR0_PTM);
+	reg |= (sCR0_VMIDPNE | sCR0_PTM);
 
 	/* Enable client access, but bypass when no mapping is found */
-	scr0 &= ~(sCR0_CLIENTPD | sCR0_USFCFG);
+	reg &= ~(sCR0_CLIENTPD | sCR0_USFCFG);
 
 	/* Disable forced broadcasting */
-	scr0 &= ~sCR0_FB;
+	reg &= ~sCR0_FB;
 
 	/* Don't upgrade barriers */
-	scr0 &= ~(sCR0_BSU_MASK << sCR0_BSU_SHIFT);
+	reg &= ~(sCR0_BSU_MASK << sCR0_BSU_SHIFT);
 
 	/* Push the button */
 	arm_smmu_tlb_sync(smmu);
-	writel(scr0, gr0_base + ARM_SMMU_GR0_sCR0);
+	writel_relaxed(reg, gr0_base + ARM_SMMU_GR0_sCR0);
 }
 
 static int arm_smmu_id_size_to_bits(int size)
@@ -1700,13 +1715,12 @@ static int arm_smmu_device_cfg_probe(struct arm_smmu_device *smmu)
 	id = readl_relaxed(gr0_base + ARM_SMMU_GR0_ID1);
 	smmu->pagesize = (id & ID1_PAGESIZE) ? SZ_64K : SZ_4K;
 
-	/* Check that we ioremapped enough */
+	/* Check for size mismatch of SMMU address space from mapped region */
 	size = 1 << (((id >> ID1_NUMPAGENDXB_SHIFT) & ID1_NUMPAGENDXB_MASK) + 1);
 	size *= (smmu->pagesize << 1);
-	if (smmu->size < size)
-		dev_warn(smmu->dev,
-			 "device is 0x%lx bytes but only mapped 0x%lx!\n",
-			 size, smmu->size);
+	if (smmu->size != size)
+		dev_warn(smmu->dev, "SMMU address space size (0x%lx) differs "
+			"from mapped region size (0x%lx)!\n", size, smmu->size);
 
 	smmu->num_s2_context_banks = (id >> ID1_NUMS2CB_SHIFT) &
 				      ID1_NUMS2CB_MASK;
@@ -1781,15 +1795,10 @@ static int arm_smmu_device_dt_probe(struct platform_device *pdev)
 	smmu->dev = dev;
 
 	res = platform_get_resource(pdev, IORESOURCE_MEM, 0);
-	if (!res) {
-		dev_err(dev, "missing base address/size\n");
-		return -ENODEV;
-	}
-
+	smmu->base = devm_ioremap_resource(dev, res);
+	if (IS_ERR(smmu->base))
+		return PTR_ERR(smmu->base);
 	smmu->size = resource_size(res);
-	smmu->base = devm_request_and_ioremap(dev, res);
-	if (!smmu->base)
-		return -EADDRNOTAVAIL;
 
 	if (of_property_read_u32(dev->of_node, "#global-interrupts",
 				 &smmu->num_global_irqs)) {
@@ -1804,12 +1813,11 @@ static int arm_smmu_device_dt_probe(struct platform_device *pdev)
 			smmu->num_context_irqs++;
 	}
 
-	if (num_irqs < smmu->num_global_irqs) {
-		dev_warn(dev, "found %d interrupts but expected at least %d\n",
-			 num_irqs, smmu->num_global_irqs);
-		smmu->num_global_irqs = num_irqs;
+	if (!smmu->num_context_irqs) {
+		dev_err(dev, "found %d interrupts but expected at least %d\n",
+			num_irqs, smmu->num_global_irqs + 1);
+		return -ENODEV;
 	}
-	smmu->num_context_irqs = num_irqs - smmu->num_global_irqs;
 
 	smmu->irqs = devm_kzalloc(dev, sizeof(*smmu->irqs) * num_irqs,
 				  GFP_KERNEL);
@@ -1933,7 +1941,7 @@ static int arm_smmu_device_remove(struct platform_device *pdev)
 		free_irq(smmu->irqs[i], smmu);
 
 	/* Turn the thing off */
-	writel(sCR0_CLIENTPD, ARM_SMMU_GR0(smmu) + ARM_SMMU_GR0_sCR0);
+	writel_relaxed(sCR0_CLIENTPD, ARM_SMMU_GR0(smmu) + ARM_SMMU_GR0_sCR0);
 	return 0;
 }
 
@@ -1981,7 +1989,7 @@ static void __exit arm_smmu_exit(void)
 	return platform_driver_unregister(&arm_smmu_driver);
 }
 
-module_init(arm_smmu_init);
+subsys_initcall(arm_smmu_init);
 module_exit(arm_smmu_exit);
 
 MODULE_DESCRIPTION("IOMMU API for ARM architected SMMU implementations");
diff --git a/drivers/iommu/dmar.c b/drivers/iommu/dmar.c
index 900946950230..8b452c9676d9 100644
--- a/drivers/iommu/dmar.c
+++ b/drivers/iommu/dmar.c
@@ -403,7 +403,7 @@ dmar_find_matched_drhd_unit(struct pci_dev *dev)
 
 	dev = pci_physfn(dev);
 
-	list_for_each_entry(dmaru, &dmar_drhd_units, list) {
+	for_each_drhd_unit(dmaru) {
 		drhd = container_of(dmaru->hdr,
 				    struct acpi_dmar_hardware_unit,
 				    header);
diff --git a/drivers/iommu/intel-iommu.c b/drivers/iommu/intel-iommu.c
index 15e9b57e9cf0..43b9bfea48fa 100644
--- a/drivers/iommu/intel-iommu.c
+++ b/drivers/iommu/intel-iommu.c
@@ -782,7 +782,11 @@ static struct dma_pte *pfn_to_dma_pte(struct dmar_domain *domain,
 	int offset;
 
 	BUG_ON(!domain->pgd);
-	BUG_ON(addr_width < BITS_PER_LONG && pfn >> addr_width);
+
+	if (addr_width < BITS_PER_LONG && pfn >> addr_width)
+		/* Address beyond IOMMU's addressing capabilities. */
+		return NULL;
+
 	parent = domain->pgd;
 
 	while (level > 0) {
@@ -3777,11 +3781,10 @@ static void iommu_detach_dependent_devices(struct intel_iommu *iommu,
 static void domain_remove_one_dev_info(struct dmar_domain *domain,
 					  struct pci_dev *pdev)
 {
-	struct device_domain_info *info;
+	struct device_domain_info *info, *tmp;
 	struct intel_iommu *iommu;
 	unsigned long flags;
 	int found = 0;
-	struct list_head *entry, *tmp;
 
 	iommu = device_to_iommu(pci_domain_nr(pdev->bus), pdev->bus->number,
 				pdev->devfn);
@@ -3789,8 +3792,7 @@ static void domain_remove_one_dev_info(struct dmar_domain *domain,
 		return;
 
 	spin_lock_irqsave(&device_domain_lock, flags);
-	list_for_each_safe(entry, tmp, &domain->devices) {
-		info = list_entry(entry, struct device_domain_info, link);
+	list_for_each_entry_safe(info, tmp, &domain->devices, link) {
 		if (info->segment == pci_domain_nr(pdev->bus) &&
 		    info->bus == pdev->bus->number &&
 		    info->devfn == pdev->devfn) {
diff --git a/drivers/iommu/intel_irq_remapping.c b/drivers/iommu/intel_irq_remapping.c
index ab86902fd9ff..bab10b1002fb 100644
--- a/drivers/iommu/intel_irq_remapping.c
+++ b/drivers/iommu/intel_irq_remapping.c
@@ -525,12 +525,13 @@ static int __init intel_irq_remapping_supported(void)
 	if (disable_irq_remap)
 		return 0;
 	if (irq_remap_broken) {
-		WARN_TAINT(1, TAINT_FIRMWARE_WORKAROUND,
-			   "This system BIOS has enabled interrupt remapping\n"
-			   "on a chipset that contains an erratum making that\n"
-			   "feature unstable.  To maintain system stability\n"
-			   "interrupt remapping is being disabled.  Please\n"
-			   "contact your BIOS vendor for an update\n");
+		printk(KERN_WARNING
+			"This system BIOS has enabled interrupt remapping\n"
+			"on a chipset that contains an erratum making that\n"
+			"feature unstable.  To maintain system stability\n"
+			"interrupt remapping is being disabled.  Please\n"
+			"contact your BIOS vendor for an update\n");
+		add_taint(TAINT_FIRMWARE_WORKAROUND, LOCKDEP_STILL_OK);
 		disable_irq_remap = 1;
 		return 0;
 	}
diff --git a/drivers/iommu/iommu-traces.c b/drivers/iommu/iommu-traces.c
new file mode 100644
index 000000000000..bf3b317ff0c1
--- /dev/null
+++ b/drivers/iommu/iommu-traces.c
@@ -0,0 +1,27 @@
+/*
+ * iommu trace points
+ *
+ * Copyright (C) 2013 Shuah Khan <shuah.kh@samsung.com>
+ *
+ */
+
+#include <linux/string.h>
+#include <linux/types.h>
+
+#define CREATE_TRACE_POINTS
+#include <trace/events/iommu.h>
+
+/* iommu_group_event */
+EXPORT_TRACEPOINT_SYMBOL_GPL(add_device_to_group);
+EXPORT_TRACEPOINT_SYMBOL_GPL(remove_device_from_group);
+
+/* iommu_device_event */
+EXPORT_TRACEPOINT_SYMBOL_GPL(attach_device_to_domain);
+EXPORT_TRACEPOINT_SYMBOL_GPL(detach_device_from_domain);
+
+/* iommu_map_unmap */
+EXPORT_TRACEPOINT_SYMBOL_GPL(map);
+EXPORT_TRACEPOINT_SYMBOL_GPL(unmap);
+
+/* iommu_error */
+EXPORT_TRACEPOINT_SYMBOL_GPL(io_page_fault);
diff --git a/drivers/iommu/iommu.c b/drivers/iommu/iommu.c
index fbe9ca734f8f..e5555fcfe703 100644
--- a/drivers/iommu/iommu.c
+++ b/drivers/iommu/iommu.c
@@ -29,6 +29,7 @@
 #include <linux/idr.h>
 #include <linux/notifier.h>
 #include <linux/err.h>
+#include <trace/events/iommu.h>
 
 static struct kset *iommu_group_kset;
 static struct ida iommu_group_ida;
@@ -363,6 +364,8 @@ rename:
 	/* Notify any listeners about change to group. */
 	blocking_notifier_call_chain(&group->notifier,
 				     IOMMU_GROUP_NOTIFY_ADD_DEVICE, dev);
+
+	trace_add_device_to_group(group->id, dev);
 	return 0;
 }
 EXPORT_SYMBOL_GPL(iommu_group_add_device);
@@ -399,6 +402,8 @@ void iommu_group_remove_device(struct device *dev)
 	sysfs_remove_link(group->devices_kobj, device->name);
 	sysfs_remove_link(&dev->kobj, "iommu_group");
 
+	trace_remove_device_from_group(group->id, dev);
+
 	kfree(device->name);
 	kfree(device);
 	dev->iommu_group = NULL;
@@ -680,10 +685,14 @@ EXPORT_SYMBOL_GPL(iommu_domain_free);
 
 int iommu_attach_device(struct iommu_domain *domain, struct device *dev)
 {
+	int ret;
 	if (unlikely(domain->ops->attach_dev == NULL))
 		return -ENODEV;
 
-	return domain->ops->attach_dev(domain, dev);
+	ret = domain->ops->attach_dev(domain, dev);
+	if (!ret)
+		trace_attach_device_to_domain(dev);
+	return ret;
 }
 EXPORT_SYMBOL_GPL(iommu_attach_device);
 
@@ -693,6 +702,7 @@ void iommu_detach_device(struct iommu_domain *domain, struct device *dev)
 		return;
 
 	domain->ops->detach_dev(domain, dev);
+	trace_detach_device_from_domain(dev);
 }
 EXPORT_SYMBOL_GPL(iommu_detach_device);
 
@@ -807,17 +817,17 @@ int iommu_map(struct iommu_domain *domain, unsigned long iova,
 	 * size of the smallest page supported by the hardware
 	 */
 	if (!IS_ALIGNED(iova | paddr | size, min_pagesz)) {
-		pr_err("unaligned: iova 0x%lx pa 0x%pa size 0x%zx min_pagesz 0x%x\n",
+		pr_err("unaligned: iova 0x%lx pa %pa size 0x%zx min_pagesz 0x%x\n",
 		       iova, &paddr, size, min_pagesz);
 		return -EINVAL;
 	}
 
-	pr_debug("map: iova 0x%lx pa 0x%pa size 0x%zx\n", iova, &paddr, size);
+	pr_debug("map: iova 0x%lx pa %pa size 0x%zx\n", iova, &paddr, size);
 
 	while (size) {
 		size_t pgsize = iommu_pgsize(domain, iova | paddr, size);
 
-		pr_debug("mapping: iova 0x%lx pa 0x%pa pgsize 0x%zx\n",
+		pr_debug("mapping: iova 0x%lx pa %pa pgsize 0x%zx\n",
 			 iova, &paddr, pgsize);
 
 		ret = domain->ops->map(domain, iova, paddr, pgsize, prot);
@@ -832,6 +842,8 @@ int iommu_map(struct iommu_domain *domain, unsigned long iova,
 	/* unroll mapping in case something went wrong */
 	if (ret)
 		iommu_unmap(domain, orig_iova, orig_size - size);
+	else
+		trace_map(iova, paddr, size);
 
 	return ret;
 }
@@ -880,6 +892,7 @@ size_t iommu_unmap(struct iommu_domain *domain, unsigned long iova, size_t size)
 		unmapped += unmapped_page;
 	}
 
+	trace_unmap(iova, 0, size);
 	return unmapped;
 }
 EXPORT_SYMBOL_GPL(iommu_unmap);
diff --git a/drivers/iommu/tegra-gart.c b/drivers/iommu/tegra-gart.c
index 108c0e9c24d9..dba1a9fd5070 100644
--- a/drivers/iommu/tegra-gart.c
+++ b/drivers/iommu/tegra-gart.c
@@ -252,7 +252,7 @@ static int gart_iommu_map(struct iommu_domain *domain, unsigned long iova,
 	spin_lock_irqsave(&gart->pte_lock, flags);
 	pfn = __phys_to_pfn(pa);
 	if (!pfn_valid(pfn)) {
-		dev_err(gart->dev, "Invalid page: %08x\n", pa);
+		dev_err(gart->dev, "Invalid page: %pa\n", &pa);
 		spin_unlock_irqrestore(&gart->pte_lock, flags);
 		return -EINVAL;
 	}
@@ -295,8 +295,8 @@ static phys_addr_t gart_iommu_iova_to_phys(struct iommu_domain *domain,
 
 	pa = (pte & GART_PAGE_MASK);
 	if (!pfn_valid(__phys_to_pfn(pa))) {
-		dev_err(gart->dev, "No entry for %08llx:%08x\n",
-			 (unsigned long long)iova, pa);
+		dev_err(gart->dev, "No entry for %08llx:%pa\n",
+			 (unsigned long long)iova, &pa);
 		gart_dump_table(gart);
 		return -EINVAL;
 	}
@@ -351,7 +351,6 @@ static int tegra_gart_probe(struct platform_device *pdev)
 	struct gart_device *gart;
 	struct resource *res, *res_remap;
 	void __iomem *gart_regs;
-	int err;
 	struct device *dev = &pdev->dev;
 
 	if (gart_handle)
@@ -376,8 +375,7 @@ static int tegra_gart_probe(struct platform_device *pdev)
 	gart_regs = devm_ioremap(dev, res->start, resource_size(res));
 	if (!gart_regs) {
 		dev_err(dev, "failed to remap GART registers\n");
-		err = -ENXIO;
-		goto fail;
+		return -ENXIO;
 	}
 
 	gart->dev = &pdev->dev;
@@ -391,8 +389,7 @@ static int tegra_gart_probe(struct platform_device *pdev)
 	gart->savedata = vmalloc(sizeof(u32) * gart->page_count);
 	if (!gart->savedata) {
 		dev_err(dev, "failed to allocate context save area\n");
-		err = -ENOMEM;
-		goto fail;
+		return -ENOMEM;
 	}
 
 	platform_set_drvdata(pdev, gart);
@@ -401,32 +398,20 @@ static int tegra_gart_probe(struct platform_device *pdev)
 	gart_handle = gart;
 	bus_set_iommu(&platform_bus_type, &gart_iommu_ops);
 	return 0;
-
-fail:
-	if (gart_regs)
-		devm_iounmap(dev, gart_regs);
-	if (gart && gart->savedata)
-		vfree(gart->savedata);
-	devm_kfree(dev, gart);
-	return err;
 }
 
 static int tegra_gart_remove(struct platform_device *pdev)
 {
 	struct gart_device *gart = platform_get_drvdata(pdev);
-	struct device *dev = gart->dev;
 
 	writel(0, gart->regs + GART_CONFIG);
 	if (gart->savedata)
 		vfree(gart->savedata);
-	if (gart->regs)
-		devm_iounmap(dev, gart->regs);
-	devm_kfree(dev, gart);
 	gart_handle = NULL;
 	return 0;
 }
 
-const struct dev_pm_ops tegra_gart_pm_ops = {
+static const struct dev_pm_ops tegra_gart_pm_ops = {
 	.suspend	= tegra_gart_suspend,
 	.resume		= tegra_gart_resume,
 };
diff --git a/drivers/iommu/tegra-smmu.c b/drivers/iommu/tegra-smmu.c
index e0665603afd9..605b5b46a903 100644
--- a/drivers/iommu/tegra-smmu.c
+++ b/drivers/iommu/tegra-smmu.c
@@ -731,7 +731,7 @@ static int smmu_iommu_map(struct iommu_domain *domain, unsigned long iova,
 	unsigned long pfn = __phys_to_pfn(pa);
 	unsigned long flags;
 
-	dev_dbg(as->smmu->dev, "[%d] %08lx:%08x\n", as->asid, iova, pa);
+	dev_dbg(as->smmu->dev, "[%d] %08lx:%pa\n", as->asid, iova, &pa);
 
 	if (!pfn_valid(pfn))
 		return -ENOMEM;
@@ -1254,7 +1254,7 @@ static int tegra_smmu_remove(struct platform_device *pdev)
 	return 0;
 }
 
-const struct dev_pm_ops tegra_smmu_pm_ops = {
+static const struct dev_pm_ops tegra_smmu_pm_ops = {
 	.suspend	= tegra_smmu_suspend,
 	.resume		= tegra_smmu_resume,
 };
diff --git a/drivers/isdn/isdnloop/isdnloop.c b/drivers/isdn/isdnloop/isdnloop.c
index baf2686aa8eb..02125e6a9109 100644
--- a/drivers/isdn/isdnloop/isdnloop.c
+++ b/drivers/isdn/isdnloop/isdnloop.c
@@ -1083,8 +1083,10 @@ isdnloop_start(isdnloop_card *card, isdnloop_sdef *sdefp)
 			spin_unlock_irqrestore(&card->isdnloop_lock, flags);
 			return -ENOMEM;
 		}
-		for (i = 0; i < 3; i++)
-			strcpy(card->s0num[i], sdef.num[i]);
+		for (i = 0; i < 3; i++) {
+			strlcpy(card->s0num[i], sdef.num[i],
+				sizeof(card->s0num[0]));
+		}
 		break;
 	case ISDN_PTYPE_1TR6:
 		if (isdnloop_fake(card, "DRV1.04TC-1TR6-CAPI-CNS-BASIS-29.11.95",
@@ -1097,7 +1099,7 @@ isdnloop_start(isdnloop_card *card, isdnloop_sdef *sdefp)
 			spin_unlock_irqrestore(&card->isdnloop_lock, flags);
 			return -ENOMEM;
 		}
-		strcpy(card->s0num[0], sdef.num[0]);
+		strlcpy(card->s0num[0], sdef.num[0], sizeof(card->s0num[0]));
 		card->s0num[1][0] = '\0';
 		card->s0num[2][0] = '\0';
 		break;
diff --git a/drivers/lguest/lguest_device.c b/drivers/lguest/lguest_device.c
index b3256ff0d426..d0a1d8a45c81 100644
--- a/drivers/lguest/lguest_device.c
+++ b/drivers/lguest/lguest_device.c
@@ -229,7 +229,7 @@ struct lguest_vq_info {
  * make a hypercall.  We hand the physical address of the virtqueue so the Host
  * knows which virtqueue we're talking about.
  */
-static void lg_notify(struct virtqueue *vq)
+static bool lg_notify(struct virtqueue *vq)
 {
 	/*
 	 * We store our virtqueue information in the "priv" pointer of the
@@ -238,6 +238,7 @@ static void lg_notify(struct virtqueue *vq)
 	struct lguest_vq_info *lvq = vq->priv;
 
 	hcall(LHCALL_NOTIFY, lvq->config.pfn << PAGE_SHIFT, 0, 0, 0);
+	return true;
 }
 
 /* An extern declaration inside a C file is bad form.  Don't do it. */
diff --git a/drivers/lguest/x86/core.c b/drivers/lguest/x86/core.c
index 516923926335..922a1acbf652 100644
--- a/drivers/lguest/x86/core.c
+++ b/drivers/lguest/x86/core.c
@@ -157,7 +157,7 @@ static void run_guest_once(struct lg_cpu *cpu, struct lguest_pages *pages)
 	 * stack, then the address of this call.  This stack layout happens to
 	 * exactly match the stack layout created by an interrupt...
 	 */
-	asm volatile("pushf; lcall *lguest_entry"
+	asm volatile("pushf; lcall *%4"
 		     /*
 		      * This is how we tell GCC that %eax ("a") and %ebx ("b")
 		      * are changed by this routine.  The "=" means output.
@@ -169,7 +169,9 @@ static void run_guest_once(struct lg_cpu *cpu, struct lguest_pages *pages)
 		      * physical address of the Guest's top-level page
 		      * directory.
 		      */
-		     : "0"(pages), "1"(__pa(cpu->lg->pgdirs[cpu->cpu_pgd].pgdir))
+		     : "0"(pages), 
+		       "1"(__pa(cpu->lg->pgdirs[cpu->cpu_pgd].pgdir)),
+		       "m"(lguest_entry)
 		     /*
 		      * We tell gcc that all these registers could change,
 		      * which means we don't have to save and restore them in
diff --git a/drivers/md/bcache/Kconfig b/drivers/md/bcache/Kconfig
index f950c9d29f3e..2638417b19aa 100644
--- a/drivers/md/bcache/Kconfig
+++ b/drivers/md/bcache/Kconfig
@@ -13,15 +13,8 @@ config BCACHE_DEBUG
 	---help---
 	Don't select this option unless you're a developer
 
-	Enables extra debugging tools (primarily a fuzz tester)
-
-config BCACHE_EDEBUG
-	bool "Extended runtime checks"
-	depends on BCACHE
-	---help---
-	Don't select this option unless you're a developer
-
-	Enables extra runtime checks which significantly affect performance
+	Enables extra debugging tools, allows expensive runtime checks to be
+	turned on.
 
 config BCACHE_CLOSURES_DEBUG
 	bool "Debug closures"
diff --git a/drivers/md/bcache/alloc.c b/drivers/md/bcache/alloc.c
index e45f5575fd4d..2b46bf1d7e40 100644
--- a/drivers/md/bcache/alloc.c
+++ b/drivers/md/bcache/alloc.c
@@ -63,13 +63,12 @@
 #include "bcache.h"
 #include "btree.h"
 
+#include <linux/blkdev.h>
 #include <linux/freezer.h>
 #include <linux/kthread.h>
 #include <linux/random.h>
 #include <trace/events/bcache.h>
 
-#define MAX_IN_FLIGHT_DISCARDS		8U
-
 /* Bucket heap / gen */
 
 uint8_t bch_inc_gen(struct cache *ca, struct bucket *b)
@@ -121,75 +120,6 @@ void bch_rescale_priorities(struct cache_set *c, int sectors)
 	mutex_unlock(&c->bucket_lock);
 }
 
-/* Discard/TRIM */
-
-struct discard {
-	struct list_head	list;
-	struct work_struct	work;
-	struct cache		*ca;
-	long			bucket;
-
-	struct bio		bio;
-	struct bio_vec		bv;
-};
-
-static void discard_finish(struct work_struct *w)
-{
-	struct discard *d = container_of(w, struct discard, work);
-	struct cache *ca = d->ca;
-	char buf[BDEVNAME_SIZE];
-
-	if (!test_bit(BIO_UPTODATE, &d->bio.bi_flags)) {
-		pr_notice("discard error on %s, disabling",
-			 bdevname(ca->bdev, buf));
-		d->ca->discard = 0;
-	}
-
-	mutex_lock(&ca->set->bucket_lock);
-
-	fifo_push(&ca->free, d->bucket);
-	list_add(&d->list, &ca->discards);
-	atomic_dec(&ca->discards_in_flight);
-
-	mutex_unlock(&ca->set->bucket_lock);
-
-	closure_wake_up(&ca->set->bucket_wait);
-	wake_up_process(ca->alloc_thread);
-
-	closure_put(&ca->set->cl);
-}
-
-static void discard_endio(struct bio *bio, int error)
-{
-	struct discard *d = container_of(bio, struct discard, bio);
-	schedule_work(&d->work);
-}
-
-static void do_discard(struct cache *ca, long bucket)
-{
-	struct discard *d = list_first_entry(&ca->discards,
-					     struct discard, list);
-
-	list_del(&d->list);
-	d->bucket = bucket;
-
-	atomic_inc(&ca->discards_in_flight);
-	closure_get(&ca->set->cl);
-
-	bio_init(&d->bio);
-
-	d->bio.bi_sector	= bucket_to_sector(ca->set, d->bucket);
-	d->bio.bi_bdev		= ca->bdev;
-	d->bio.bi_rw		= REQ_WRITE|REQ_DISCARD;
-	d->bio.bi_max_vecs	= 1;
-	d->bio.bi_io_vec	= d->bio.bi_inline_vecs;
-	d->bio.bi_size		= bucket_bytes(ca);
-	d->bio.bi_end_io	= discard_endio;
-	bio_set_prio(&d->bio, IOPRIO_PRIO_VALUE(IOPRIO_CLASS_IDLE, 0));
-
-	submit_bio(0, &d->bio);
-}
-
 /* Allocation */
 
 static inline bool can_inc_bucket_gen(struct bucket *b)
@@ -280,7 +210,7 @@ static void invalidate_buckets_lru(struct cache *ca)
 			 * multiple times when it can't do anything
 			 */
 			ca->invalidate_needs_gc = 1;
-			bch_queue_gc(ca->set);
+			wake_up_gc(ca->set);
 			return;
 		}
 
@@ -305,7 +235,7 @@ static void invalidate_buckets_fifo(struct cache *ca)
 
 		if (++checked >= ca->sb.nbuckets) {
 			ca->invalidate_needs_gc = 1;
-			bch_queue_gc(ca->set);
+			wake_up_gc(ca->set);
 			return;
 		}
 	}
@@ -330,7 +260,7 @@ static void invalidate_buckets_random(struct cache *ca)
 
 		if (++checked >= ca->sb.nbuckets / 2) {
 			ca->invalidate_needs_gc = 1;
-			bch_queue_gc(ca->set);
+			wake_up_gc(ca->set);
 			return;
 		}
 	}
@@ -398,16 +328,18 @@ static int bch_allocator_thread(void *arg)
 			else
 				break;
 
-			allocator_wait(ca, (int) fifo_free(&ca->free) >
-				       atomic_read(&ca->discards_in_flight));
-
 			if (ca->discard) {
-				allocator_wait(ca, !list_empty(&ca->discards));
-				do_discard(ca, bucket);
-			} else {
-				fifo_push(&ca->free, bucket);
-				closure_wake_up(&ca->set->bucket_wait);
+				mutex_unlock(&ca->set->bucket_lock);
+				blkdev_issue_discard(ca->bdev,
+					bucket_to_sector(ca->set, bucket),
+					ca->sb.block_size, GFP_KERNEL, 0);
+				mutex_lock(&ca->set->bucket_lock);
 			}
+
+			allocator_wait(ca, !fifo_full(&ca->free));
+
+			fifo_push(&ca->free, bucket);
+			wake_up(&ca->set->bucket_wait);
 		}
 
 		/*
@@ -433,16 +365,40 @@ static int bch_allocator_thread(void *arg)
 	}
 }
 
-long bch_bucket_alloc(struct cache *ca, unsigned watermark, struct closure *cl)
+long bch_bucket_alloc(struct cache *ca, unsigned watermark, bool wait)
 {
-	long r = -1;
-again:
+	DEFINE_WAIT(w);
+	struct bucket *b;
+	long r;
+
+	/* fastpath */
+	if (fifo_used(&ca->free) > ca->watermark[watermark]) {
+		fifo_pop(&ca->free, r);
+		goto out;
+	}
+
+	if (!wait)
+		return -1;
+
+	while (1) {
+		if (fifo_used(&ca->free) > ca->watermark[watermark]) {
+			fifo_pop(&ca->free, r);
+			break;
+		}
+
+		prepare_to_wait(&ca->set->bucket_wait, &w,
+				TASK_UNINTERRUPTIBLE);
+
+		mutex_unlock(&ca->set->bucket_lock);
+		schedule();
+		mutex_lock(&ca->set->bucket_lock);
+	}
+
+	finish_wait(&ca->set->bucket_wait, &w);
+out:
 	wake_up_process(ca->alloc_thread);
 
-	if (fifo_used(&ca->free) > ca->watermark[watermark] &&
-	    fifo_pop(&ca->free, r)) {
-		struct bucket *b = ca->buckets + r;
-#ifdef CONFIG_BCACHE_EDEBUG
+	if (expensive_debug_checks(ca->set)) {
 		size_t iter;
 		long i;
 
@@ -455,36 +411,23 @@ again:
 			BUG_ON(i == r);
 		fifo_for_each(i, &ca->unused, iter)
 			BUG_ON(i == r);
-#endif
-		BUG_ON(atomic_read(&b->pin) != 1);
-
-		SET_GC_SECTORS_USED(b, ca->sb.bucket_size);
-
-		if (watermark <= WATERMARK_METADATA) {
-			SET_GC_MARK(b, GC_MARK_METADATA);
-			b->prio = BTREE_PRIO;
-		} else {
-			SET_GC_MARK(b, GC_MARK_RECLAIMABLE);
-			b->prio = INITIAL_PRIO;
-		}
-
-		return r;
 	}
 
-	trace_bcache_alloc_fail(ca);
+	b = ca->buckets + r;
 
-	if (cl) {
-		closure_wait(&ca->set->bucket_wait, cl);
+	BUG_ON(atomic_read(&b->pin) != 1);
 
-		if (closure_blocking(cl)) {
-			mutex_unlock(&ca->set->bucket_lock);
-			closure_sync(cl);
-			mutex_lock(&ca->set->bucket_lock);
-			goto again;
-		}
+	SET_GC_SECTORS_USED(b, ca->sb.bucket_size);
+
+	if (watermark <= WATERMARK_METADATA) {
+		SET_GC_MARK(b, GC_MARK_METADATA);
+		b->prio = BTREE_PRIO;
+	} else {
+		SET_GC_MARK(b, GC_MARK_RECLAIMABLE);
+		b->prio = INITIAL_PRIO;
 	}
 
-	return -1;
+	return r;
 }
 
 void bch_bucket_free(struct cache_set *c, struct bkey *k)
@@ -501,7 +444,7 @@ void bch_bucket_free(struct cache_set *c, struct bkey *k)
 }
 
 int __bch_bucket_alloc_set(struct cache_set *c, unsigned watermark,
-			   struct bkey *k, int n, struct closure *cl)
+			   struct bkey *k, int n, bool wait)
 {
 	int i;
 
@@ -514,7 +457,7 @@ int __bch_bucket_alloc_set(struct cache_set *c, unsigned watermark,
 
 	for (i = 0; i < n; i++) {
 		struct cache *ca = c->cache_by_alloc[i];
-		long b = bch_bucket_alloc(ca, watermark, cl);
+		long b = bch_bucket_alloc(ca, watermark, wait);
 
 		if (b == -1)
 			goto err;
@@ -529,22 +472,202 @@ int __bch_bucket_alloc_set(struct cache_set *c, unsigned watermark,
 	return 0;
 err:
 	bch_bucket_free(c, k);
-	__bkey_put(c, k);
+	bkey_put(c, k);
 	return -1;
 }
 
 int bch_bucket_alloc_set(struct cache_set *c, unsigned watermark,
-			 struct bkey *k, int n, struct closure *cl)
+			 struct bkey *k, int n, bool wait)
 {
 	int ret;
 	mutex_lock(&c->bucket_lock);
-	ret = __bch_bucket_alloc_set(c, watermark, k, n, cl);
+	ret = __bch_bucket_alloc_set(c, watermark, k, n, wait);
 	mutex_unlock(&c->bucket_lock);
 	return ret;
 }
 
+/* Sector allocator */
+
+struct open_bucket {
+	struct list_head	list;
+	unsigned		last_write_point;
+	unsigned		sectors_free;
+	BKEY_PADDED(key);
+};
+
+/*
+ * We keep multiple buckets open for writes, and try to segregate different
+ * write streams for better cache utilization: first we look for a bucket where
+ * the last write to it was sequential with the current write, and failing that
+ * we look for a bucket that was last used by the same task.
+ *
+ * The ideas is if you've got multiple tasks pulling data into the cache at the
+ * same time, you'll get better cache utilization if you try to segregate their
+ * data and preserve locality.
+ *
+ * For example, say you've starting Firefox at the same time you're copying a
+ * bunch of files. Firefox will likely end up being fairly hot and stay in the
+ * cache awhile, but the data you copied might not be; if you wrote all that
+ * data to the same buckets it'd get invalidated at the same time.
+ *
+ * Both of those tasks will be doing fairly random IO so we can't rely on
+ * detecting sequential IO to segregate their data, but going off of the task
+ * should be a sane heuristic.
+ */
+static struct open_bucket *pick_data_bucket(struct cache_set *c,
+					    const struct bkey *search,
+					    unsigned write_point,
+					    struct bkey *alloc)
+{
+	struct open_bucket *ret, *ret_task = NULL;
+
+	list_for_each_entry_reverse(ret, &c->data_buckets, list)
+		if (!bkey_cmp(&ret->key, search))
+			goto found;
+		else if (ret->last_write_point == write_point)
+			ret_task = ret;
+
+	ret = ret_task ?: list_first_entry(&c->data_buckets,
+					   struct open_bucket, list);
+found:
+	if (!ret->sectors_free && KEY_PTRS(alloc)) {
+		ret->sectors_free = c->sb.bucket_size;
+		bkey_copy(&ret->key, alloc);
+		bkey_init(alloc);
+	}
+
+	if (!ret->sectors_free)
+		ret = NULL;
+
+	return ret;
+}
+
+/*
+ * Allocates some space in the cache to write to, and k to point to the newly
+ * allocated space, and updates KEY_SIZE(k) and KEY_OFFSET(k) (to point to the
+ * end of the newly allocated space).
+ *
+ * May allocate fewer sectors than @sectors, KEY_SIZE(k) indicates how many
+ * sectors were actually allocated.
+ *
+ * If s->writeback is true, will not fail.
+ */
+bool bch_alloc_sectors(struct cache_set *c, struct bkey *k, unsigned sectors,
+		       unsigned write_point, unsigned write_prio, bool wait)
+{
+	struct open_bucket *b;
+	BKEY_PADDED(key) alloc;
+	unsigned i;
+
+	/*
+	 * We might have to allocate a new bucket, which we can't do with a
+	 * spinlock held. So if we have to allocate, we drop the lock, allocate
+	 * and then retry. KEY_PTRS() indicates whether alloc points to
+	 * allocated bucket(s).
+	 */
+
+	bkey_init(&alloc.key);
+	spin_lock(&c->data_bucket_lock);
+
+	while (!(b = pick_data_bucket(c, k, write_point, &alloc.key))) {
+		unsigned watermark = write_prio
+			? WATERMARK_MOVINGGC
+			: WATERMARK_NONE;
+
+		spin_unlock(&c->data_bucket_lock);
+
+		if (bch_bucket_alloc_set(c, watermark, &alloc.key, 1, wait))
+			return false;
+
+		spin_lock(&c->data_bucket_lock);
+	}
+
+	/*
+	 * If we had to allocate, we might race and not need to allocate the
+	 * second time we call find_data_bucket(). If we allocated a bucket but
+	 * didn't use it, drop the refcount bch_bucket_alloc_set() took:
+	 */
+	if (KEY_PTRS(&alloc.key))
+		bkey_put(c, &alloc.key);
+
+	for (i = 0; i < KEY_PTRS(&b->key); i++)
+		EBUG_ON(ptr_stale(c, &b->key, i));
+
+	/* Set up the pointer to the space we're allocating: */
+
+	for (i = 0; i < KEY_PTRS(&b->key); i++)
+		k->ptr[i] = b->key.ptr[i];
+
+	sectors = min(sectors, b->sectors_free);
+
+	SET_KEY_OFFSET(k, KEY_OFFSET(k) + sectors);
+	SET_KEY_SIZE(k, sectors);
+	SET_KEY_PTRS(k, KEY_PTRS(&b->key));
+
+	/*
+	 * Move b to the end of the lru, and keep track of what this bucket was
+	 * last used for:
+	 */
+	list_move_tail(&b->list, &c->data_buckets);
+	bkey_copy_key(&b->key, k);
+	b->last_write_point = write_point;
+
+	b->sectors_free	-= sectors;
+
+	for (i = 0; i < KEY_PTRS(&b->key); i++) {
+		SET_PTR_OFFSET(&b->key, i, PTR_OFFSET(&b->key, i) + sectors);
+
+		atomic_long_add(sectors,
+				&PTR_CACHE(c, &b->key, i)->sectors_written);
+	}
+
+	if (b->sectors_free < c->sb.block_size)
+		b->sectors_free = 0;
+
+	/*
+	 * k takes refcounts on the buckets it points to until it's inserted
+	 * into the btree, but if we're done with this bucket we just transfer
+	 * get_data_bucket()'s refcount.
+	 */
+	if (b->sectors_free)
+		for (i = 0; i < KEY_PTRS(&b->key); i++)
+			atomic_inc(&PTR_BUCKET(c, &b->key, i)->pin);
+
+	spin_unlock(&c->data_bucket_lock);
+	return true;
+}
+
 /* Init */
 
+void bch_open_buckets_free(struct cache_set *c)
+{
+	struct open_bucket *b;
+
+	while (!list_empty(&c->data_buckets)) {
+		b = list_first_entry(&c->data_buckets,
+				     struct open_bucket, list);
+		list_del(&b->list);
+		kfree(b);
+	}
+}
+
+int bch_open_buckets_alloc(struct cache_set *c)
+{
+	int i;
+
+	spin_lock_init(&c->data_bucket_lock);
+
+	for (i = 0; i < 6; i++) {
+		struct open_bucket *b = kzalloc(sizeof(*b), GFP_KERNEL);
+		if (!b)
+			return -ENOMEM;
+
+		list_add(&b->list, &c->data_buckets);
+	}
+
+	return 0;
+}
+
 int bch_cache_allocator_start(struct cache *ca)
 {
 	struct task_struct *k = kthread_run(bch_allocator_thread,
@@ -556,22 +679,8 @@ int bch_cache_allocator_start(struct cache *ca)
 	return 0;
 }
 
-void bch_cache_allocator_exit(struct cache *ca)
-{
-	struct discard *d;
-
-	while (!list_empty(&ca->discards)) {
-		d = list_first_entry(&ca->discards, struct discard, list);
-		cancel_work_sync(&d->work);
-		list_del(&d->list);
-		kfree(d);
-	}
-}
-
 int bch_cache_allocator_init(struct cache *ca)
 {
-	unsigned i;
-
 	/*
 	 * Reserve:
 	 * Prio/gen writes first
@@ -589,15 +698,5 @@ int bch_cache_allocator_init(struct cache *ca)
 	ca->watermark[WATERMARK_NONE] = ca->free.size / 2 +
 		ca->watermark[WATERMARK_MOVINGGC];
 
-	for (i = 0; i < MAX_IN_FLIGHT_DISCARDS; i++) {
-		struct discard *d = kzalloc(sizeof(*d), GFP_KERNEL);
-		if (!d)
-			return -ENOMEM;
-
-		d->ca = ca;
-		INIT_WORK(&d->work, discard_finish);
-		list_add(&d->list, &ca->discards);
-	}
-
 	return 0;
 }
diff --git a/drivers/md/bcache/bcache.h b/drivers/md/bcache/bcache.h
index 0f12382aa35d..4beb55a0ff30 100644
--- a/drivers/md/bcache/bcache.h
+++ b/drivers/md/bcache/bcache.h
@@ -177,6 +177,7 @@
 
 #define pr_fmt(fmt) "bcache: %s() " fmt "\n", __func__
 
+#include <linux/bcache.h>
 #include <linux/bio.h>
 #include <linux/kobject.h>
 #include <linux/list.h>
@@ -210,168 +211,6 @@ BITMASK(GC_MARK,	 struct bucket, gc_mark, 0, 2);
 #define GC_MARK_METADATA	2
 BITMASK(GC_SECTORS_USED, struct bucket, gc_mark, 2, 14);
 
-struct bkey {
-	uint64_t	high;
-	uint64_t	low;
-	uint64_t	ptr[];
-};
-
-/* Enough for a key with 6 pointers */
-#define BKEY_PAD		8
-
-#define BKEY_PADDED(key)					\
-	union { struct bkey key; uint64_t key ## _pad[BKEY_PAD]; }
-
-/* Version 0: Cache device
- * Version 1: Backing device
- * Version 2: Seed pointer into btree node checksum
- * Version 3: Cache device with new UUID format
- * Version 4: Backing device with data offset
- */
-#define BCACHE_SB_VERSION_CDEV			0
-#define BCACHE_SB_VERSION_BDEV			1
-#define BCACHE_SB_VERSION_CDEV_WITH_UUID	3
-#define BCACHE_SB_VERSION_BDEV_WITH_OFFSET	4
-#define BCACHE_SB_MAX_VERSION			4
-
-#define SB_SECTOR		8
-#define SB_SIZE			4096
-#define SB_LABEL_SIZE		32
-#define SB_JOURNAL_BUCKETS	256U
-/* SB_JOURNAL_BUCKETS must be divisible by BITS_PER_LONG */
-#define MAX_CACHES_PER_SET	8
-
-#define BDEV_DATA_START_DEFAULT	16	/* sectors */
-
-struct cache_sb {
-	uint64_t		csum;
-	uint64_t		offset;	/* sector where this sb was written */
-	uint64_t		version;
-
-	uint8_t			magic[16];
-
-	uint8_t			uuid[16];
-	union {
-		uint8_t		set_uuid[16];
-		uint64_t	set_magic;
-	};
-	uint8_t			label[SB_LABEL_SIZE];
-
-	uint64_t		flags;
-	uint64_t		seq;
-	uint64_t		pad[8];
-
-	union {
-	struct {
-		/* Cache devices */
-		uint64_t	nbuckets;	/* device size */
-
-		uint16_t	block_size;	/* sectors */
-		uint16_t	bucket_size;	/* sectors */
-
-		uint16_t	nr_in_set;
-		uint16_t	nr_this_dev;
-	};
-	struct {
-		/* Backing devices */
-		uint64_t	data_offset;
-
-		/*
-		 * block_size from the cache device section is still used by
-		 * backing devices, so don't add anything here until we fix
-		 * things to not need it for backing devices anymore
-		 */
-	};
-	};
-
-	uint32_t		last_mount;	/* time_t */
-
-	uint16_t		first_bucket;
-	union {
-		uint16_t	njournal_buckets;
-		uint16_t	keys;
-	};
-	uint64_t		d[SB_JOURNAL_BUCKETS];	/* journal buckets */
-};
-
-BITMASK(CACHE_SYNC,		struct cache_sb, flags, 0, 1);
-BITMASK(CACHE_DISCARD,		struct cache_sb, flags, 1, 1);
-BITMASK(CACHE_REPLACEMENT,	struct cache_sb, flags, 2, 3);
-#define CACHE_REPLACEMENT_LRU	0U
-#define CACHE_REPLACEMENT_FIFO	1U
-#define CACHE_REPLACEMENT_RANDOM 2U
-
-BITMASK(BDEV_CACHE_MODE,	struct cache_sb, flags, 0, 4);
-#define CACHE_MODE_WRITETHROUGH	0U
-#define CACHE_MODE_WRITEBACK	1U
-#define CACHE_MODE_WRITEAROUND	2U
-#define CACHE_MODE_NONE		3U
-BITMASK(BDEV_STATE,		struct cache_sb, flags, 61, 2);
-#define BDEV_STATE_NONE		0U
-#define BDEV_STATE_CLEAN	1U
-#define BDEV_STATE_DIRTY	2U
-#define BDEV_STATE_STALE	3U
-
-/* Version 1: Seed pointer into btree node checksum
- */
-#define BCACHE_BSET_VERSION	1
-
-/*
- * This is the on disk format for btree nodes - a btree node on disk is a list
- * of these; within each set the keys are sorted
- */
-struct bset {
-	uint64_t		csum;
-	uint64_t		magic;
-	uint64_t		seq;
-	uint32_t		version;
-	uint32_t		keys;
-
-	union {
-		struct bkey	start[0];
-		uint64_t	d[0];
-	};
-};
-
-/*
- * On disk format for priorities and gens - see super.c near prio_write() for
- * more.
- */
-struct prio_set {
-	uint64_t		csum;
-	uint64_t		magic;
-	uint64_t		seq;
-	uint32_t		version;
-	uint32_t		pad;
-
-	uint64_t		next_bucket;
-
-	struct bucket_disk {
-		uint16_t	prio;
-		uint8_t		gen;
-	} __attribute((packed)) data[];
-};
-
-struct uuid_entry {
-	union {
-		struct {
-			uint8_t		uuid[16];
-			uint8_t		label[32];
-			uint32_t	first_reg;
-			uint32_t	last_reg;
-			uint32_t	invalidated;
-
-			uint32_t	flags;
-			/* Size of flash only volumes */
-			uint64_t	sectors;
-		};
-
-		uint8_t	pad[128];
-	};
-};
-
-BITMASK(UUID_FLASH_ONLY,	struct uuid_entry, flags, 0, 1);
-
 #include "journal.h"
 #include "stats.h"
 struct search;
@@ -384,8 +223,6 @@ struct keybuf_key {
 	void			*private;
 };
 
-typedef bool (keybuf_pred_fn)(struct keybuf *, struct bkey *);
-
 struct keybuf {
 	struct bkey		last_scanned;
 	spinlock_t		lock;
@@ -400,7 +237,7 @@ struct keybuf {
 
 	struct rb_root		keys;
 
-#define KEYBUF_NR		100
+#define KEYBUF_NR		500
 	DECLARE_ARRAY_ALLOCATOR(struct keybuf_key, freelist, KEYBUF_NR);
 };
 
@@ -429,16 +266,15 @@ struct bcache_device {
 
 	struct gendisk		*disk;
 
-	/* If nonzero, we're closing */
-	atomic_t		closing;
-
-	/* If nonzero, we're detaching/unregistering from cache set */
-	atomic_t		detaching;
-	int			flush_done;
+	unsigned long		flags;
+#define BCACHE_DEV_CLOSING	0
+#define BCACHE_DEV_DETACHING	1
+#define BCACHE_DEV_UNLINK_DONE	2
 
-	uint64_t		nr_stripes;
-	unsigned		stripe_size_bits;
+	unsigned		nr_stripes;
+	unsigned		stripe_size;
 	atomic_t		*stripe_sectors_dirty;
+	unsigned long		*full_dirty_stripes;
 
 	unsigned long		sectors_dirty_last;
 	long			sectors_dirty_derivative;
@@ -509,7 +345,7 @@ struct cached_dev {
 
 	/* Limit number of writeback bios in flight */
 	struct semaphore	in_flight;
-	struct closure_with_timer writeback;
+	struct task_struct	*writeback_thread;
 
 	struct keybuf		writeback_keys;
 
@@ -527,8 +363,8 @@ struct cached_dev {
 	unsigned		sequential_cutoff;
 	unsigned		readahead;
 
-	unsigned		sequential_merge:1;
 	unsigned		verify:1;
+	unsigned		bypass_torture_test:1;
 
 	unsigned		partial_stripes_expensive:1;
 	unsigned		writeback_metadata:1;
@@ -620,15 +456,6 @@ struct cache {
 
 	bool			discard; /* Get rid of? */
 
-	/*
-	 * We preallocate structs for issuing discards to buckets, and keep them
-	 * on this list when they're not in use; do_discard() issues discards
-	 * whenever there's work to do and is called by free_some_buckets() and
-	 * when a discard finishes.
-	 */
-	atomic_t		discards_in_flight;
-	struct list_head	discards;
-
 	struct journal_device	journal;
 
 	/* The rest of this all shows up in sysfs */
@@ -649,7 +476,6 @@ struct gc_stat {
 
 	size_t			nkeys;
 	uint64_t		data;	/* sectors */
-	uint64_t		dirty;	/* sectors */
 	unsigned		in_use; /* percent */
 };
 
@@ -744,8 +570,8 @@ struct cache_set {
 	 * basically a lock for this that we can wait on asynchronously. The
 	 * btree_root() macro releases the lock when it returns.
 	 */
-	struct closure		*try_harder;
-	struct closure_waitlist	try_wait;
+	struct task_struct	*try_harder;
+	wait_queue_head_t	try_wait;
 	uint64_t		try_harder_start;
 
 	/*
@@ -759,7 +585,7 @@ struct cache_set {
 	 * written.
 	 */
 	atomic_t		prio_blocked;
-	struct closure_waitlist	bucket_wait;
+	wait_queue_head_t	bucket_wait;
 
 	/*
 	 * For any bio we don't skip we subtract the number of sectors from
@@ -782,7 +608,7 @@ struct cache_set {
 	struct gc_stat		gc_stats;
 	size_t			nbuckets;
 
-	struct closure_with_waitlist gc;
+	struct task_struct	*gc_thread;
 	/* Where in the btree gc currently is */
 	struct bkey		gc_done;
 
@@ -795,11 +621,10 @@ struct cache_set {
 	/* Counts how many sectors bio_insert has added to the cache */
 	atomic_t		sectors_to_gc;
 
-	struct closure		moving_gc;
-	struct closure_waitlist	moving_gc_wait;
+	wait_queue_head_t	moving_gc_wait;
 	struct keybuf		moving_gc_keys;
 	/* Number of moving GC bios in flight */
-	atomic_t		in_flight;
+	struct semaphore	moving_in_flight;
 
 	struct btree		*root;
 
@@ -841,22 +666,27 @@ struct cache_set {
 	unsigned		congested_read_threshold_us;
 	unsigned		congested_write_threshold_us;
 
-	spinlock_t		sort_time_lock;
 	struct time_stats	sort_time;
 	struct time_stats	btree_gc_time;
 	struct time_stats	btree_split_time;
-	spinlock_t		btree_read_time_lock;
 	struct time_stats	btree_read_time;
 	struct time_stats	try_harder_time;
 
 	atomic_long_t		cache_read_races;
 	atomic_long_t		writeback_keys_done;
 	atomic_long_t		writeback_keys_failed;
+
+	enum			{
+		ON_ERROR_UNREGISTER,
+		ON_ERROR_PANIC,
+	}			on_error;
 	unsigned		error_limit;
 	unsigned		error_decay;
+
 	unsigned short		journal_delay_ms;
 	unsigned		verify:1;
 	unsigned		key_merging_disabled:1;
+	unsigned		expensive_debug_checks:1;
 	unsigned		gc_always_rewrite:1;
 	unsigned		shrinker_disabled:1;
 	unsigned		copy_gc_enabled:1;
@@ -865,21 +695,6 @@ struct cache_set {
 	struct hlist_head	bucket_hash[1 << BUCKET_HASH_BITS];
 };
 
-static inline bool key_merging_disabled(struct cache_set *c)
-{
-#ifdef CONFIG_BCACHE_DEBUG
-	return c->key_merging_disabled;
-#else
-	return 0;
-#endif
-}
-
-static inline bool SB_IS_BDEV(const struct cache_sb *sb)
-{
-	return sb->version == BCACHE_SB_VERSION_BDEV
-		|| sb->version == BCACHE_SB_VERSION_BDEV_WITH_OFFSET;
-}
-
 struct bbio {
 	unsigned		submit_time_us;
 	union {
@@ -933,59 +748,6 @@ static inline unsigned local_clock_us(void)
 #define prio_buckets(c)					\
 	DIV_ROUND_UP((size_t) (c)->sb.nbuckets, prios_per_bucket(c))
 
-#define JSET_MAGIC		0x245235c1a3625032ULL
-#define PSET_MAGIC		0x6750e15f87337f91ULL
-#define BSET_MAGIC		0x90135c78b99e07f5ULL
-
-#define jset_magic(c)		((c)->sb.set_magic ^ JSET_MAGIC)
-#define pset_magic(c)		((c)->sb.set_magic ^ PSET_MAGIC)
-#define bset_magic(c)		((c)->sb.set_magic ^ BSET_MAGIC)
-
-/* Bkey fields: all units are in sectors */
-
-#define KEY_FIELD(name, field, offset, size)				\
-	BITMASK(name, struct bkey, field, offset, size)
-
-#define PTR_FIELD(name, offset, size)					\
-	static inline uint64_t name(const struct bkey *k, unsigned i)	\
-	{ return (k->ptr[i] >> offset) & ~(((uint64_t) ~0) << size); }	\
-									\
-	static inline void SET_##name(struct bkey *k, unsigned i, uint64_t v)\
-	{								\
-		k->ptr[i] &= ~(~((uint64_t) ~0 << size) << offset);	\
-		k->ptr[i] |= v << offset;				\
-	}
-
-KEY_FIELD(KEY_PTRS,	high, 60, 3)
-KEY_FIELD(HEADER_SIZE,	high, 58, 2)
-KEY_FIELD(KEY_CSUM,	high, 56, 2)
-KEY_FIELD(KEY_PINNED,	high, 55, 1)
-KEY_FIELD(KEY_DIRTY,	high, 36, 1)
-
-KEY_FIELD(KEY_SIZE,	high, 20, 16)
-KEY_FIELD(KEY_INODE,	high, 0,  20)
-
-/* Next time I change the on disk format, KEY_OFFSET() won't be 64 bits */
-
-static inline uint64_t KEY_OFFSET(const struct bkey *k)
-{
-	return k->low;
-}
-
-static inline void SET_KEY_OFFSET(struct bkey *k, uint64_t v)
-{
-	k->low = v;
-}
-
-PTR_FIELD(PTR_DEV,		51, 12)
-PTR_FIELD(PTR_OFFSET,		8,  43)
-PTR_FIELD(PTR_GEN,		0,  8)
-
-#define PTR_CHECK_DEV		((1 << 12) - 1)
-
-#define PTR(gen, offset, dev)						\
-	((((uint64_t) dev) << 51) | ((uint64_t) offset) << 8 | gen)
-
 static inline size_t sector_to_bucket(struct cache_set *c, sector_t s)
 {
 	return s >> c->bucket_bits;
@@ -1024,27 +786,11 @@ static inline struct bucket *PTR_BUCKET(struct cache_set *c,
 
 /* Btree key macros */
 
-/*
- * The high bit being set is a relic from when we used it to do binary
- * searches - it told you where a key started. It's not used anymore,
- * and can probably be safely dropped.
- */
-#define KEY(dev, sector, len)						\
-((struct bkey) {							\
-	.high = (1ULL << 63) | ((uint64_t) (len) << 20) | (dev),	\
-	.low = (sector)							\
-})
-
 static inline void bkey_init(struct bkey *k)
 {
-	*k = KEY(0, 0, 0);
+	*k = ZERO_KEY;
 }
 
-#define KEY_START(k)		(KEY_OFFSET(k) - KEY_SIZE(k))
-#define START_KEY(k)		KEY(KEY_INODE(k), KEY_START(k), 0)
-#define MAX_KEY			KEY(~(~0 << 20), ((uint64_t) ~0) >> 1, 0)
-#define ZERO_KEY		KEY(0, 0, 0)
-
 /*
  * This is used for various on disk data structures - cache_sb, prio_set, bset,
  * jset: The checksum is _always_ the first 8 bytes of these structs
@@ -1094,14 +840,6 @@ do {									\
 	for (b = (ca)->buckets + (ca)->sb.first_bucket;			\
 	     b < (ca)->buckets + (ca)->sb.nbuckets; b++)
 
-static inline void __bkey_put(struct cache_set *c, struct bkey *k)
-{
-	unsigned i;
-
-	for (i = 0; i < KEY_PTRS(k); i++)
-		atomic_dec_bug(&PTR_BUCKET(c, k, i)->pin);
-}
-
 static inline void cached_dev_put(struct cached_dev *dc)
 {
 	if (atomic_dec_and_test(&dc->count))
@@ -1173,13 +911,15 @@ uint8_t bch_inc_gen(struct cache *, struct bucket *);
 void bch_rescale_priorities(struct cache_set *, int);
 bool bch_bucket_add_unused(struct cache *, struct bucket *);
 
-long bch_bucket_alloc(struct cache *, unsigned, struct closure *);
+long bch_bucket_alloc(struct cache *, unsigned, bool);
 void bch_bucket_free(struct cache_set *, struct bkey *);
 
 int __bch_bucket_alloc_set(struct cache_set *, unsigned,
-			   struct bkey *, int, struct closure *);
+			   struct bkey *, int, bool);
 int bch_bucket_alloc_set(struct cache_set *, unsigned,
-			 struct bkey *, int, struct closure *);
+			 struct bkey *, int, bool);
+bool bch_alloc_sectors(struct cache_set *, struct bkey *, unsigned,
+		       unsigned, unsigned, bool);
 
 __printf(2, 3)
 bool bch_cache_set_error(struct cache_set *, const char *, ...);
@@ -1187,7 +927,7 @@ bool bch_cache_set_error(struct cache_set *, const char *, ...);
 void bch_prio_write(struct cache *);
 void bch_write_bdev_super(struct cached_dev *, struct closure *);
 
-extern struct workqueue_struct *bcache_wq, *bch_gc_wq;
+extern struct workqueue_struct *bcache_wq;
 extern const char * const bch_cache_modes[];
 extern struct mutex bch_register_lock;
 extern struct list_head bch_cache_sets;
@@ -1220,15 +960,14 @@ struct cache_set *bch_cache_set_alloc(struct cache_sb *);
 void bch_btree_cache_free(struct cache_set *);
 int bch_btree_cache_alloc(struct cache_set *);
 void bch_moving_init_cache_set(struct cache_set *);
+int bch_open_buckets_alloc(struct cache_set *);
+void bch_open_buckets_free(struct cache_set *);
 
 int bch_cache_allocator_start(struct cache *ca);
-void bch_cache_allocator_exit(struct cache *ca);
 int bch_cache_allocator_init(struct cache *ca);
 
 void bch_debug_exit(void);
 int bch_debug_init(struct kobject *);
-void bch_writeback_exit(void);
-int bch_writeback_init(void);
 void bch_request_exit(void);
 int bch_request_init(void);
 void bch_btree_exit(void);
diff --git a/drivers/md/bcache/bset.c b/drivers/md/bcache/bset.c
index 22d1ae72c282..7d388b8bb50e 100644
--- a/drivers/md/bcache/bset.c
+++ b/drivers/md/bcache/bset.c
@@ -14,22 +14,12 @@
 
 /* Keylists */
 
-void bch_keylist_copy(struct keylist *dest, struct keylist *src)
-{
-	*dest = *src;
-
-	if (src->list == src->d) {
-		size_t n = (uint64_t *) src->top - src->d;
-		dest->top = (struct bkey *) &dest->d[n];
-		dest->list = dest->d;
-	}
-}
-
 int bch_keylist_realloc(struct keylist *l, int nptrs, struct cache_set *c)
 {
-	unsigned oldsize = (uint64_t *) l->top - l->list;
-	unsigned newsize = oldsize + 2 + nptrs;
-	uint64_t *new;
+	size_t oldsize = bch_keylist_nkeys(l);
+	size_t newsize = oldsize + 2 + nptrs;
+	uint64_t *old_keys = l->keys_p == l->inline_keys ? NULL : l->keys_p;
+	uint64_t *new_keys;
 
 	/* The journalling code doesn't handle the case where the keys to insert
 	 * is bigger than an empty write: If we just return -ENOMEM here,
@@ -45,24 +35,23 @@ int bch_keylist_realloc(struct keylist *l, int nptrs, struct cache_set *c)
 	    roundup_pow_of_two(oldsize) == newsize)
 		return 0;
 
-	new = krealloc(l->list == l->d ? NULL : l->list,
-		       sizeof(uint64_t) * newsize, GFP_NOIO);
+	new_keys = krealloc(old_keys, sizeof(uint64_t) * newsize, GFP_NOIO);
 
-	if (!new)
+	if (!new_keys)
 		return -ENOMEM;
 
-	if (l->list == l->d)
-		memcpy(new, l->list, sizeof(uint64_t) * KEYLIST_INLINE);
+	if (!old_keys)
+		memcpy(new_keys, l->inline_keys, sizeof(uint64_t) * oldsize);
 
-	l->list = new;
-	l->top = (struct bkey *) (&l->list[oldsize]);
+	l->keys_p = new_keys;
+	l->top_p = new_keys + oldsize;
 
 	return 0;
 }
 
 struct bkey *bch_keylist_pop(struct keylist *l)
 {
-	struct bkey *k = l->bottom;
+	struct bkey *k = l->keys;
 
 	if (k == l->top)
 		return NULL;
@@ -73,21 +62,20 @@ struct bkey *bch_keylist_pop(struct keylist *l)
 	return l->top = k;
 }
 
-/* Pointer validation */
-
-bool __bch_ptr_invalid(struct cache_set *c, int level, const struct bkey *k)
+void bch_keylist_pop_front(struct keylist *l)
 {
-	unsigned i;
-	char buf[80];
+	l->top_p -= bkey_u64s(l->keys);
 
-	if (level && (!KEY_PTRS(k) || !KEY_SIZE(k) || KEY_DIRTY(k)))
-		goto bad;
+	memmove(l->keys,
+		bkey_next(l->keys),
+		bch_keylist_bytes(l));
+}
 
-	if (!level && KEY_SIZE(k) > KEY_OFFSET(k))
-		goto bad;
+/* Pointer validation */
 
-	if (!KEY_SIZE(k))
-		return true;
+static bool __ptr_invalid(struct cache_set *c, const struct bkey *k)
+{
+	unsigned i;
 
 	for (i = 0; i < KEY_PTRS(k); i++)
 		if (ptr_available(c, k, i)) {
@@ -98,13 +86,83 @@ bool __bch_ptr_invalid(struct cache_set *c, int level, const struct bkey *k)
 			if (KEY_SIZE(k) + r > c->sb.bucket_size ||
 			    bucket <  ca->sb.first_bucket ||
 			    bucket >= ca->sb.nbuckets)
-				goto bad;
+				return true;
 		}
 
 	return false;
+}
+
+bool bch_btree_ptr_invalid(struct cache_set *c, const struct bkey *k)
+{
+	char buf[80];
+
+	if (!KEY_PTRS(k) || !KEY_SIZE(k) || KEY_DIRTY(k))
+		goto bad;
+
+	if (__ptr_invalid(c, k))
+		goto bad;
+
+	return false;
+bad:
+	bch_bkey_to_text(buf, sizeof(buf), k);
+	cache_bug(c, "spotted btree ptr %s: %s", buf, bch_ptr_status(c, k));
+	return true;
+}
+
+bool bch_extent_ptr_invalid(struct cache_set *c, const struct bkey *k)
+{
+	char buf[80];
+
+	if (!KEY_SIZE(k))
+		return true;
+
+	if (KEY_SIZE(k) > KEY_OFFSET(k))
+		goto bad;
+
+	if (__ptr_invalid(c, k))
+		goto bad;
+
+	return false;
 bad:
 	bch_bkey_to_text(buf, sizeof(buf), k);
-	cache_bug(c, "spotted bad key %s: %s", buf, bch_ptr_status(c, k));
+	cache_bug(c, "spotted extent %s: %s", buf, bch_ptr_status(c, k));
+	return true;
+}
+
+static bool ptr_bad_expensive_checks(struct btree *b, const struct bkey *k,
+				     unsigned ptr)
+{
+	struct bucket *g = PTR_BUCKET(b->c, k, ptr);
+	char buf[80];
+
+	if (mutex_trylock(&b->c->bucket_lock)) {
+		if (b->level) {
+			if (KEY_DIRTY(k) ||
+			    g->prio != BTREE_PRIO ||
+			    (b->c->gc_mark_valid &&
+			     GC_MARK(g) != GC_MARK_METADATA))
+				goto err;
+
+		} else {
+			if (g->prio == BTREE_PRIO)
+				goto err;
+
+			if (KEY_DIRTY(k) &&
+			    b->c->gc_mark_valid &&
+			    GC_MARK(g) != GC_MARK_DIRTY)
+				goto err;
+		}
+		mutex_unlock(&b->c->bucket_lock);
+	}
+
+	return false;
+err:
+	mutex_unlock(&b->c->bucket_lock);
+	bch_bkey_to_text(buf, sizeof(buf), k);
+	btree_bug(b,
+"inconsistent pointer %s: bucket %zu pin %i prio %i gen %i last_gc %i mark %llu gc_gen %i",
+		  buf, PTR_BUCKET_NR(b->c, k, ptr), atomic_read(&g->pin),
+		  g->prio, g->gen, g->last_gc, GC_MARK(g), g->gc_gen);
 	return true;
 }
 
@@ -118,64 +176,29 @@ bool bch_ptr_bad(struct btree *b, const struct bkey *k)
 	    bch_ptr_invalid(b, k))
 		return true;
 
-	if (KEY_PTRS(k) && PTR_DEV(k, 0) == PTR_CHECK_DEV)
-		return true;
+	for (i = 0; i < KEY_PTRS(k); i++) {
+		if (!ptr_available(b->c, k, i))
+			return true;
 
-	for (i = 0; i < KEY_PTRS(k); i++)
-		if (ptr_available(b->c, k, i)) {
-			g = PTR_BUCKET(b->c, k, i);
-			stale = ptr_stale(b->c, k, i);
+		g = PTR_BUCKET(b->c, k, i);
+		stale = ptr_stale(b->c, k, i);
 
-			btree_bug_on(stale > 96, b,
-				     "key too stale: %i, need_gc %u",
-				     stale, b->c->need_gc);
+		btree_bug_on(stale > 96, b,
+			     "key too stale: %i, need_gc %u",
+			     stale, b->c->need_gc);
 
-			btree_bug_on(stale && KEY_DIRTY(k) && KEY_SIZE(k),
-				     b, "stale dirty pointer");
+		btree_bug_on(stale && KEY_DIRTY(k) && KEY_SIZE(k),
+			     b, "stale dirty pointer");
 
-			if (stale)
-				return true;
+		if (stale)
+			return true;
 
-#ifdef CONFIG_BCACHE_EDEBUG
-			if (!mutex_trylock(&b->c->bucket_lock))
-				continue;
-
-			if (b->level) {
-				if (KEY_DIRTY(k) ||
-				    g->prio != BTREE_PRIO ||
-				    (b->c->gc_mark_valid &&
-				     GC_MARK(g) != GC_MARK_METADATA))
-					goto bug;
-
-			} else {
-				if (g->prio == BTREE_PRIO)
-					goto bug;
-
-				if (KEY_DIRTY(k) &&
-				    b->c->gc_mark_valid &&
-				    GC_MARK(g) != GC_MARK_DIRTY)
-					goto bug;
-			}
-			mutex_unlock(&b->c->bucket_lock);
-#endif
-		}
+		if (expensive_debug_checks(b->c) &&
+		    ptr_bad_expensive_checks(b, k, i))
+			return true;
+	}
 
 	return false;
-#ifdef CONFIG_BCACHE_EDEBUG
-bug:
-	mutex_unlock(&b->c->bucket_lock);
-
-	{
-		char buf[80];
-
-		bch_bkey_to_text(buf, sizeof(buf), k);
-		btree_bug(b,
-"inconsistent pointer %s: bucket %zu pin %i prio %i gen %i last_gc %i mark %llu gc_gen %i",
-			  buf, PTR_BUCKET_NR(b->c, k, i), atomic_read(&g->pin),
-			  g->prio, g->gen, g->last_gc, GC_MARK(g), g->gc_gen);
-	}
-	return true;
-#endif
 }
 
 /* Key/pointer manipulation */
@@ -458,16 +481,8 @@ static struct bkey *table_to_bkey(struct bset_tree *t, unsigned cacheline)
 
 static inline uint64_t shrd128(uint64_t high, uint64_t low, uint8_t shift)
 {
-#ifdef CONFIG_X86_64
-	asm("shrd %[shift],%[high],%[low]"
-	    : [low] "+Rm" (low)
-	    : [high] "R" (high),
-	    [shift] "ci" (shift)
-	    : "cc");
-#else
 	low >>= shift;
 	low  |= (high << 1) << (63U - shift);
-#endif
 	return low;
 }
 
@@ -686,7 +701,7 @@ void bch_bset_init_next(struct btree *b)
 	} else
 		get_random_bytes(&i->seq, sizeof(uint64_t));
 
-	i->magic	= bset_magic(b->c);
+	i->magic	= bset_magic(&b->c->sb);
 	i->version	= 0;
 	i->keys		= 0;
 
@@ -824,16 +839,16 @@ struct bkey *__bch_bset_search(struct btree *b, struct bset_tree *t,
 	} else
 		i = bset_search_write_set(b, t, search);
 
-#ifdef CONFIG_BCACHE_EDEBUG
-	BUG_ON(bset_written(b, t) &&
-	       i.l != t->data->start &&
-	       bkey_cmp(tree_to_prev_bkey(t,
-		  inorder_to_tree(bkey_to_cacheline(t, i.l), t)),
-			search) > 0);
+	if (expensive_debug_checks(b->c)) {
+		BUG_ON(bset_written(b, t) &&
+		       i.l != t->data->start &&
+		       bkey_cmp(tree_to_prev_bkey(t,
+			  inorder_to_tree(bkey_to_cacheline(t, i.l), t)),
+				search) > 0);
 
-	BUG_ON(i.r != end(t->data) &&
-	       bkey_cmp(i.r, search) <= 0);
-#endif
+		BUG_ON(i.r != end(t->data) &&
+		       bkey_cmp(i.r, search) <= 0);
+	}
 
 	while (likely(i.l != i.r) &&
 	       bkey_cmp(i.l, search) <= 0)
@@ -844,6 +859,13 @@ struct bkey *__bch_bset_search(struct btree *b, struct bset_tree *t,
 
 /* Btree iterator */
 
+/*
+ * Returns true if l > r - unless l == r, in which case returns true if l is
+ * older than r.
+ *
+ * Necessary for btree_sort_fixup() - if there are multiple keys that compare
+ * equal in different sets, we have to process them newest to oldest.
+ */
 static inline bool btree_iter_cmp(struct btree_iter_set l,
 				  struct btree_iter_set r)
 {
@@ -867,12 +889,16 @@ void bch_btree_iter_push(struct btree_iter *iter, struct bkey *k,
 }
 
 struct bkey *__bch_btree_iter_init(struct btree *b, struct btree_iter *iter,
-			       struct bkey *search, struct bset_tree *start)
+				   struct bkey *search, struct bset_tree *start)
 {
 	struct bkey *ret = NULL;
 	iter->size = ARRAY_SIZE(iter->data);
 	iter->used = 0;
 
+#ifdef CONFIG_BCACHE_DEBUG
+	iter->b = b;
+#endif
+
 	for (; start <= &b->sets[b->nsets]; start++) {
 		ret = bch_bset_search(b, start, search);
 		bch_btree_iter_push(iter, ret, end(start->data));
@@ -887,6 +913,8 @@ struct bkey *bch_btree_iter_next(struct btree_iter *iter)
 	struct bkey *ret = NULL;
 
 	if (!btree_iter_end(iter)) {
+		bch_btree_iter_next_check(iter);
+
 		ret = iter->data->k;
 		iter->data->k = bkey_next(iter->data->k);
 
@@ -916,14 +944,6 @@ struct bkey *bch_btree_iter_next_filter(struct btree_iter *iter,
 	return ret;
 }
 
-struct bkey *bch_next_recurse_key(struct btree *b, struct bkey *search)
-{
-	struct btree_iter iter;
-
-	bch_btree_iter_init(b, &iter, search);
-	return bch_btree_iter_next_filter(&iter, b, bch_ptr_bad);
-}
-
 /* Mergesort */
 
 static void sort_key_next(struct btree_iter *iter,
@@ -998,7 +1018,6 @@ static void btree_mergesort(struct btree *b, struct bset *out,
 	out->keys = last ? (uint64_t *) bkey_next(last) - out->d : 0;
 
 	pr_debug("sorted %i keys", out->keys);
-	bch_check_key_order(b, out);
 }
 
 static void __btree_sort(struct btree *b, struct btree_iter *iter,
@@ -1029,7 +1048,7 @@ static void __btree_sort(struct btree *b, struct btree_iter *iter,
 		 * memcpy()
 		 */
 
-		out->magic	= bset_magic(b->c);
+		out->magic	= bset_magic(&b->c->sb);
 		out->seq	= b->sets[0].data->seq;
 		out->version	= b->sets[0].data->version;
 		swap(out, b->sets[0].data);
@@ -1050,24 +1069,21 @@ static void __btree_sort(struct btree *b, struct btree_iter *iter,
 	if (b->written)
 		bset_build_written_tree(b);
 
-	if (!start) {
-		spin_lock(&b->c->sort_time_lock);
+	if (!start)
 		bch_time_stats_update(&b->c->sort_time, start_time);
-		spin_unlock(&b->c->sort_time_lock);
-	}
 }
 
 void bch_btree_sort_partial(struct btree *b, unsigned start)
 {
-	size_t oldsize = 0, order = b->page_order, keys = 0;
+	size_t order = b->page_order, keys = 0;
 	struct btree_iter iter;
+	int oldsize = bch_count_data(b);
+
 	__bch_btree_iter_init(b, &iter, NULL, &b->sets[start]);
 
 	BUG_ON(b->sets[b->nsets].data == write_block(b) &&
 	       (b->sets[b->nsets].size || b->nsets));
 
-	if (b->written)
-		oldsize = bch_count_data(b);
 
 	if (start) {
 		unsigned i;
@@ -1083,7 +1099,7 @@ void bch_btree_sort_partial(struct btree *b, unsigned start)
 
 	__btree_sort(b, &iter, start, order, false);
 
-	EBUG_ON(b->written && bch_count_data(b) != oldsize);
+	EBUG_ON(b->written && oldsize >= 0 && bch_count_data(b) != oldsize);
 }
 
 void bch_btree_sort_and_fix_extents(struct btree *b, struct btree_iter *iter)
@@ -1101,9 +1117,7 @@ void bch_btree_sort_into(struct btree *b, struct btree *new)
 
 	btree_mergesort(b, new->sets->data, &iter, false, true);
 
-	spin_lock(&b->c->sort_time_lock);
 	bch_time_stats_update(&b->c->sort_time, start_time);
-	spin_unlock(&b->c->sort_time_lock);
 
 	bkey_copy_key(&new->key, &b->key);
 	new->sets->size = 0;
@@ -1148,16 +1162,16 @@ out:
 /* Sysfs stuff */
 
 struct bset_stats {
+	struct btree_op op;
 	size_t nodes;
 	size_t sets_written, sets_unwritten;
 	size_t bytes_written, bytes_unwritten;
 	size_t floats, failed;
 };
 
-static int bch_btree_bset_stats(struct btree *b, struct btree_op *op,
-			    struct bset_stats *stats)
+static int btree_bset_stats(struct btree_op *op, struct btree *b)
 {
-	struct bkey *k;
+	struct bset_stats *stats = container_of(op, struct bset_stats, op);
 	unsigned i;
 
 	stats->nodes++;
@@ -1182,30 +1196,19 @@ static int bch_btree_bset_stats(struct btree *b, struct btree_op *op,
 		}
 	}
 
-	if (b->level) {
-		struct btree_iter iter;
-
-		for_each_key_filter(b, k, &iter, bch_ptr_bad) {
-			int ret = btree(bset_stats, k, b, op, stats);
-			if (ret)
-				return ret;
-		}
-	}
-
-	return 0;
+	return MAP_CONTINUE;
 }
 
 int bch_bset_print_stats(struct cache_set *c, char *buf)
 {
-	struct btree_op op;
 	struct bset_stats t;
 	int ret;
 
-	bch_btree_op_init_stack(&op);
 	memset(&t, 0, sizeof(struct bset_stats));
+	bch_btree_op_init(&t.op, -1);
 
-	ret = btree_root(bset_stats, c, &op, &t);
-	if (ret)
+	ret = bch_btree_map_nodes(&t.op, c, &ZERO_KEY, btree_bset_stats);
+	if (ret < 0)
 		return ret;
 
 	return snprintf(buf, PAGE_SIZE,
diff --git a/drivers/md/bcache/bset.h b/drivers/md/bcache/bset.h
index ae115a253d73..1d3c24f9fa0e 100644
--- a/drivers/md/bcache/bset.h
+++ b/drivers/md/bcache/bset.h
@@ -148,6 +148,9 @@
 
 struct btree_iter {
 	size_t size, used;
+#ifdef CONFIG_BCACHE_DEBUG
+	struct btree *b;
+#endif
 	struct btree_iter_set {
 		struct bkey *k, *end;
 	} data[MAX_BSETS];
@@ -193,54 +196,26 @@ static __always_inline int64_t bkey_cmp(const struct bkey *l,
 		: (int64_t) KEY_OFFSET(l) - (int64_t) KEY_OFFSET(r);
 }
 
-static inline size_t bkey_u64s(const struct bkey *k)
-{
-	BUG_ON(KEY_CSUM(k) > 1);
-	return 2 + KEY_PTRS(k) + (KEY_CSUM(k) ? 1 : 0);
-}
-
-static inline size_t bkey_bytes(const struct bkey *k)
-{
-	return bkey_u64s(k) * sizeof(uint64_t);
-}
-
-static inline void bkey_copy(struct bkey *dest, const struct bkey *src)
-{
-	memcpy(dest, src, bkey_bytes(src));
-}
-
-static inline void bkey_copy_key(struct bkey *dest, const struct bkey *src)
-{
-	if (!src)
-		src = &KEY(0, 0, 0);
-
-	SET_KEY_INODE(dest, KEY_INODE(src));
-	SET_KEY_OFFSET(dest, KEY_OFFSET(src));
-}
-
-static inline struct bkey *bkey_next(const struct bkey *k)
-{
-	uint64_t *d = (void *) k;
-	return (struct bkey *) (d + bkey_u64s(k));
-}
-
 /* Keylists */
 
 struct keylist {
-	struct bkey		*top;
 	union {
-		uint64_t		*list;
-		struct bkey		*bottom;
+		struct bkey		*keys;
+		uint64_t		*keys_p;
+	};
+	union {
+		struct bkey		*top;
+		uint64_t		*top_p;
 	};
 
 	/* Enough room for btree_split's keys without realloc */
 #define KEYLIST_INLINE		16
-	uint64_t		d[KEYLIST_INLINE];
+	uint64_t		inline_keys[KEYLIST_INLINE];
 };
 
 static inline void bch_keylist_init(struct keylist *l)
 {
-	l->top = (void *) (l->list = l->d);
+	l->top_p = l->keys_p = l->inline_keys;
 }
 
 static inline void bch_keylist_push(struct keylist *l)
@@ -256,17 +231,32 @@ static inline void bch_keylist_add(struct keylist *l, struct bkey *k)
 
 static inline bool bch_keylist_empty(struct keylist *l)
 {
-	return l->top == (void *) l->list;
+	return l->top == l->keys;
+}
+
+static inline void bch_keylist_reset(struct keylist *l)
+{
+	l->top = l->keys;
 }
 
 static inline void bch_keylist_free(struct keylist *l)
 {
-	if (l->list != l->d)
-		kfree(l->list);
+	if (l->keys_p != l->inline_keys)
+		kfree(l->keys_p);
+}
+
+static inline size_t bch_keylist_nkeys(struct keylist *l)
+{
+	return l->top_p - l->keys_p;
+}
+
+static inline size_t bch_keylist_bytes(struct keylist *l)
+{
+	return bch_keylist_nkeys(l) * sizeof(uint64_t);
 }
 
-void bch_keylist_copy(struct keylist *, struct keylist *);
 struct bkey *bch_keylist_pop(struct keylist *);
+void bch_keylist_pop_front(struct keylist *);
 int bch_keylist_realloc(struct keylist *, int, struct cache_set *);
 
 void bch_bkey_copy_single_ptr(struct bkey *, const struct bkey *,
@@ -287,7 +277,9 @@ static inline bool bch_cut_back(const struct bkey *where, struct bkey *k)
 }
 
 const char *bch_ptr_status(struct cache_set *, const struct bkey *);
-bool __bch_ptr_invalid(struct cache_set *, int level, const struct bkey *);
+bool bch_btree_ptr_invalid(struct cache_set *, const struct bkey *);
+bool bch_extent_ptr_invalid(struct cache_set *, const struct bkey *);
+
 bool bch_ptr_bad(struct btree *, const struct bkey *);
 
 static inline uint8_t gen_after(uint8_t a, uint8_t b)
@@ -311,7 +303,6 @@ static inline bool ptr_available(struct cache_set *c, const struct bkey *k,
 
 typedef bool (*ptr_filter_fn)(struct btree *, const struct bkey *);
 
-struct bkey *bch_next_recurse_key(struct btree *, struct bkey *);
 struct bkey *bch_btree_iter_next(struct btree_iter *);
 struct bkey *bch_btree_iter_next_filter(struct btree_iter *,
 					struct btree *, ptr_filter_fn);
@@ -361,12 +352,30 @@ void bch_bset_fix_lookup_table(struct btree *, struct bkey *);
 struct bkey *__bch_bset_search(struct btree *, struct bset_tree *,
 			   const struct bkey *);
 
+/*
+ * Returns the first key that is strictly greater than search
+ */
 static inline struct bkey *bch_bset_search(struct btree *b, struct bset_tree *t,
 					   const struct bkey *search)
 {
 	return search ? __bch_bset_search(b, t, search) : t->data->start;
 }
 
+#define PRECEDING_KEY(_k)					\
+({								\
+	struct bkey *_ret = NULL;				\
+								\
+	if (KEY_INODE(_k) || KEY_OFFSET(_k)) {			\
+		_ret = &KEY(KEY_INODE(_k), KEY_OFFSET(_k), 0);	\
+								\
+		if (!_ret->low)					\
+			_ret->high--;				\
+		_ret->low--;					\
+	}							\
+								\
+	_ret;							\
+})
+
 bool bch_bkey_try_merge(struct btree *, struct bkey *, struct bkey *);
 void bch_btree_sort_lazy(struct btree *);
 void bch_btree_sort_into(struct btree *, struct btree *);
diff --git a/drivers/md/bcache/btree.c b/drivers/md/bcache/btree.c
index f42fc7ed9cd6..5e2765aadce1 100644
--- a/drivers/md/bcache/btree.c
+++ b/drivers/md/bcache/btree.c
@@ -23,12 +23,13 @@
 #include "bcache.h"
 #include "btree.h"
 #include "debug.h"
-#include "request.h"
 #include "writeback.h"
 
 #include <linux/slab.h>
 #include <linux/bitops.h>
+#include <linux/freezer.h>
 #include <linux/hash.h>
+#include <linux/kthread.h>
 #include <linux/prefetch.h>
 #include <linux/random.h>
 #include <linux/rcupdate.h>
@@ -88,15 +89,13 @@
  * Test module load/unload
  */
 
-static const char * const op_types[] = {
-	"insert", "replace"
+enum {
+	BTREE_INSERT_STATUS_INSERT,
+	BTREE_INSERT_STATUS_BACK_MERGE,
+	BTREE_INSERT_STATUS_OVERWROTE,
+	BTREE_INSERT_STATUS_FRONT_MERGE,
 };
 
-static const char *op_type(struct btree_op *op)
-{
-	return op_types[op->type];
-}
-
 #define MAX_NEED_GC		64
 #define MAX_SAVE_PRIO		72
 
@@ -105,23 +104,89 @@ static const char *op_type(struct btree_op *op)
 #define PTR_HASH(c, k)							\
 	(((k)->ptr[0] >> c->bucket_bits) | PTR_GEN(k, 0))
 
-struct workqueue_struct *bch_gc_wq;
 static struct workqueue_struct *btree_io_wq;
 
-void bch_btree_op_init_stack(struct btree_op *op)
+static inline bool should_split(struct btree *b)
 {
-	memset(op, 0, sizeof(struct btree_op));
-	closure_init_stack(&op->cl);
-	op->lock = -1;
-	bch_keylist_init(&op->keys);
+	struct bset *i = write_block(b);
+	return b->written >= btree_blocks(b) ||
+		(b->written + __set_blocks(i, i->keys + 15, b->c)
+		 > btree_blocks(b));
 }
 
+#define insert_lock(s, b)	((b)->level <= (s)->lock)
+
+/*
+ * These macros are for recursing down the btree - they handle the details of
+ * locking and looking up nodes in the cache for you. They're best treated as
+ * mere syntax when reading code that uses them.
+ *
+ * op->lock determines whether we take a read or a write lock at a given depth.
+ * If you've got a read lock and find that you need a write lock (i.e. you're
+ * going to have to split), set op->lock and return -EINTR; btree_root() will
+ * call you again and you'll have the correct lock.
+ */
+
+/**
+ * btree - recurse down the btree on a specified key
+ * @fn:		function to call, which will be passed the child node
+ * @key:	key to recurse on
+ * @b:		parent btree node
+ * @op:		pointer to struct btree_op
+ */
+#define btree(fn, key, b, op, ...)					\
+({									\
+	int _r, l = (b)->level - 1;					\
+	bool _w = l <= (op)->lock;					\
+	struct btree *_child = bch_btree_node_get((b)->c, key, l, _w);	\
+	if (!IS_ERR(_child)) {						\
+		_child->parent = (b);					\
+		_r = bch_btree_ ## fn(_child, op, ##__VA_ARGS__);	\
+		rw_unlock(_w, _child);					\
+	} else								\
+		_r = PTR_ERR(_child);					\
+	_r;								\
+})
+
+/**
+ * btree_root - call a function on the root of the btree
+ * @fn:		function to call, which will be passed the child node
+ * @c:		cache set
+ * @op:		pointer to struct btree_op
+ */
+#define btree_root(fn, c, op, ...)					\
+({									\
+	int _r = -EINTR;						\
+	do {								\
+		struct btree *_b = (c)->root;				\
+		bool _w = insert_lock(op, _b);				\
+		rw_lock(_w, _b, _b->level);				\
+		if (_b == (c)->root &&					\
+		    _w == insert_lock(op, _b)) {			\
+			_b->parent = NULL;				\
+			_r = bch_btree_ ## fn(_b, op, ##__VA_ARGS__);	\
+		}							\
+		rw_unlock(_w, _b);					\
+		bch_cannibalize_unlock(c);				\
+		if (_r == -ENOSPC) {					\
+			wait_event((c)->try_wait,			\
+				   !(c)->try_harder);			\
+			_r = -EINTR;					\
+		}							\
+	} while (_r == -EINTR);						\
+									\
+	_r;								\
+})
+
 /* Btree key manipulation */
 
-static void bkey_put(struct cache_set *c, struct bkey *k, int level)
+void bkey_put(struct cache_set *c, struct bkey *k)
 {
-	if ((level && KEY_OFFSET(k)) || !level)
-		__bkey_put(c, k);
+	unsigned i;
+
+	for (i = 0; i < KEY_PTRS(k); i++)
+		if (ptr_available(c, k, i))
+			atomic_dec_bug(&PTR_BUCKET(c, k, i)->pin);
 }
 
 /* Btree IO */
@@ -145,6 +210,10 @@ static void bch_btree_node_read_done(struct btree *b)
 	iter->size = b->c->sb.bucket_size / b->c->sb.block_size;
 	iter->used = 0;
 
+#ifdef CONFIG_BCACHE_DEBUG
+	iter->b = b;
+#endif
+
 	if (!i->seq)
 		goto err;
 
@@ -160,7 +229,7 @@ static void bch_btree_node_read_done(struct btree *b)
 			goto err;
 
 		err = "bad magic";
-		if (i->magic != bset_magic(b->c))
+		if (i->magic != bset_magic(&b->c->sb))
 			goto err;
 
 		err = "bad checksum";
@@ -248,10 +317,7 @@ void bch_btree_node_read(struct btree *b)
 		goto err;
 
 	bch_btree_node_read_done(b);
-
-	spin_lock(&b->c->btree_read_time_lock);
 	bch_time_stats_update(&b->c->btree_read_time, start_time);
-	spin_unlock(&b->c->btree_read_time_lock);
 
 	return;
 err:
@@ -327,7 +393,7 @@ static void do_btree_node_write(struct btree *b)
 	b->bio = bch_bbio_alloc(b->c);
 
 	b->bio->bi_end_io	= btree_node_write_endio;
-	b->bio->bi_private	= &b->io.cl;
+	b->bio->bi_private	= cl;
 	b->bio->bi_rw		= REQ_META|WRITE_SYNC|REQ_FUA;
 	b->bio->bi_size		= set_blocks(i, b->c) * block_bytes(b->c);
 	bch_bio_map(b->bio, i);
@@ -383,7 +449,7 @@ void bch_btree_node_write(struct btree *b, struct closure *parent)
 	BUG_ON(b->written >= btree_blocks(b));
 	BUG_ON(b->written && !i->keys);
 	BUG_ON(b->sets->data->seq != i->seq);
-	bch_check_key_order(b, i);
+	bch_check_keys(b, "writing");
 
 	cancel_delayed_work(&b->work);
 
@@ -405,6 +471,15 @@ void bch_btree_node_write(struct btree *b, struct closure *parent)
 		bch_bset_init_next(b);
 }
 
+static void bch_btree_node_write_sync(struct btree *b)
+{
+	struct closure cl;
+
+	closure_init_stack(&cl);
+	bch_btree_node_write(b, &cl);
+	closure_sync(&cl);
+}
+
 static void btree_node_write_work(struct work_struct *w)
 {
 	struct btree *b = container_of(to_delayed_work(w), struct btree, work);
@@ -416,7 +491,7 @@ static void btree_node_write_work(struct work_struct *w)
 	rw_unlock(true, b);
 }
 
-static void bch_btree_leaf_dirty(struct btree *b, struct btree_op *op)
+static void bch_btree_leaf_dirty(struct btree *b, atomic_t *journal_ref)
 {
 	struct bset *i = b->sets[b->nsets].data;
 	struct btree_write *w = btree_current_write(b);
@@ -429,15 +504,15 @@ static void bch_btree_leaf_dirty(struct btree *b, struct btree_op *op)
 
 	set_btree_node_dirty(b);
 
-	if (op && op->journal) {
+	if (journal_ref) {
 		if (w->journal &&
-		    journal_pin_cmp(b->c, w, op)) {
+		    journal_pin_cmp(b->c, w->journal, journal_ref)) {
 			atomic_dec_bug(w->journal);
 			w->journal = NULL;
 		}
 
 		if (!w->journal) {
-			w->journal = op->journal;
+			w->journal = journal_ref;
 			atomic_inc(w->journal);
 		}
 	}
@@ -566,33 +641,32 @@ static struct btree *mca_bucket_alloc(struct cache_set *c,
 	return b;
 }
 
-static int mca_reap(struct btree *b, struct closure *cl, unsigned min_order)
+static int mca_reap(struct btree *b, unsigned min_order, bool flush)
 {
+	struct closure cl;
+
+	closure_init_stack(&cl);
 	lockdep_assert_held(&b->c->bucket_lock);
 
 	if (!down_write_trylock(&b->lock))
 		return -ENOMEM;
 
-	if (b->page_order < min_order) {
+	BUG_ON(btree_node_dirty(b) && !b->sets[0].data);
+
+	if (b->page_order < min_order ||
+	    (!flush &&
+	     (btree_node_dirty(b) ||
+	      atomic_read(&b->io.cl.remaining) != -1))) {
 		rw_unlock(true, b);
 		return -ENOMEM;
 	}
 
-	BUG_ON(btree_node_dirty(b) && !b->sets[0].data);
-
-	if (cl && btree_node_dirty(b))
-		bch_btree_node_write(b, NULL);
-
-	if (cl)
-		closure_wait_event_async(&b->io.wait, cl,
-			 atomic_read(&b->io.cl.remaining) == -1);
+	if (btree_node_dirty(b))
+		bch_btree_node_write_sync(b);
 
-	if (btree_node_dirty(b) ||
-	    !closure_is_unlocked(&b->io.cl) ||
-	    work_pending(&b->work.work)) {
-		rw_unlock(true, b);
-		return -EAGAIN;
-	}
+	/* wait for any in flight btree write */
+	closure_wait_event(&b->io.wait, &cl,
+			   atomic_read(&b->io.cl.remaining) == -1);
 
 	return 0;
 }
@@ -633,7 +707,7 @@ static unsigned long bch_mca_scan(struct shrinker *shrink,
 			break;
 
 		if (++i > 3 &&
-		    !mca_reap(b, NULL, 0)) {
+		    !mca_reap(b, 0, false)) {
 			mca_data_free(b);
 			rw_unlock(true, b);
 			freed++;
@@ -652,7 +726,7 @@ static unsigned long bch_mca_scan(struct shrinker *shrink,
 		list_rotate_left(&c->btree_cache);
 
 		if (!b->accessed &&
-		    !mca_reap(b, NULL, 0)) {
+		    !mca_reap(b, 0, false)) {
 			mca_bucket_free(b);
 			mca_data_free(b);
 			rw_unlock(true, b);
@@ -723,12 +797,9 @@ int bch_btree_cache_alloc(struct cache_set *c)
 {
 	unsigned i;
 
-	/* XXX: doesn't check for errors */
-
-	closure_init_unlocked(&c->gc);
-
 	for (i = 0; i < mca_reserve(c); i++)
-		mca_bucket_alloc(c, &ZERO_KEY, GFP_KERNEL);
+		if (!mca_bucket_alloc(c, &ZERO_KEY, GFP_KERNEL))
+			return -ENOMEM;
 
 	list_splice_init(&c->btree_cache,
 			 &c->btree_cache_freeable);
@@ -775,52 +846,27 @@ out:
 	return b;
 }
 
-static struct btree *mca_cannibalize(struct cache_set *c, struct bkey *k,
-				     int level, struct closure *cl)
+static struct btree *mca_cannibalize(struct cache_set *c, struct bkey *k)
 {
-	int ret = -ENOMEM;
-	struct btree *i;
+	struct btree *b;
 
 	trace_bcache_btree_cache_cannibalize(c);
 
-	if (!cl)
-		return ERR_PTR(-ENOMEM);
-
-	/*
-	 * Trying to free up some memory - i.e. reuse some btree nodes - may
-	 * require initiating IO to flush the dirty part of the node. If we're
-	 * running under generic_make_request(), that IO will never finish and
-	 * we would deadlock. Returning -EAGAIN causes the cache lookup code to
-	 * punt to workqueue and retry.
-	 */
-	if (current->bio_list)
-		return ERR_PTR(-EAGAIN);
-
-	if (c->try_harder && c->try_harder != cl) {
-		closure_wait_event_async(&c->try_wait, cl, !c->try_harder);
-		return ERR_PTR(-EAGAIN);
-	}
+	if (!c->try_harder) {
+		c->try_harder = current;
+		c->try_harder_start = local_clock();
+	} else if (c->try_harder != current)
+		return ERR_PTR(-ENOSPC);
 
-	c->try_harder = cl;
-	c->try_harder_start = local_clock();
-retry:
-	list_for_each_entry_reverse(i, &c->btree_cache, list) {
-		int r = mca_reap(i, cl, btree_order(k));
-		if (!r)
-			return i;
-		if (r != -ENOMEM)
-			ret = r;
-	}
+	list_for_each_entry_reverse(b, &c->btree_cache, list)
+		if (!mca_reap(b, btree_order(k), false))
+			return b;
 
-	if (ret == -EAGAIN &&
-	    closure_blocking(cl)) {
-		mutex_unlock(&c->bucket_lock);
-		closure_sync(cl);
-		mutex_lock(&c->bucket_lock);
-		goto retry;
-	}
+	list_for_each_entry_reverse(b, &c->btree_cache, list)
+		if (!mca_reap(b, btree_order(k), true))
+			return b;
 
-	return ERR_PTR(ret);
+	return ERR_PTR(-ENOMEM);
 }
 
 /*
@@ -829,20 +875,21 @@ retry:
  * cannibalize_bucket() will take. This means every time we unlock the root of
  * the btree, we need to release this lock if we have it held.
  */
-void bch_cannibalize_unlock(struct cache_set *c, struct closure *cl)
+static void bch_cannibalize_unlock(struct cache_set *c)
 {
-	if (c->try_harder == cl) {
+	if (c->try_harder == current) {
 		bch_time_stats_update(&c->try_harder_time, c->try_harder_start);
 		c->try_harder = NULL;
-		__closure_wake_up(&c->try_wait);
+		wake_up(&c->try_wait);
 	}
 }
 
-static struct btree *mca_alloc(struct cache_set *c, struct bkey *k,
-			       int level, struct closure *cl)
+static struct btree *mca_alloc(struct cache_set *c, struct bkey *k, int level)
 {
 	struct btree *b;
 
+	BUG_ON(current->bio_list);
+
 	lockdep_assert_held(&c->bucket_lock);
 
 	if (mca_find(c, k))
@@ -852,14 +899,14 @@ static struct btree *mca_alloc(struct cache_set *c, struct bkey *k,
 	 * the list. Check if there's any freed nodes there:
 	 */
 	list_for_each_entry(b, &c->btree_cache_freeable, list)
-		if (!mca_reap(b, NULL, btree_order(k)))
+		if (!mca_reap(b, btree_order(k), false))
 			goto out;
 
 	/* We never free struct btree itself, just the memory that holds the on
 	 * disk node. Check the freed list before allocating a new one:
 	 */
 	list_for_each_entry(b, &c->btree_cache_freed, list)
-		if (!mca_reap(b, NULL, 0)) {
+		if (!mca_reap(b, 0, false)) {
 			mca_data_alloc(b, k, __GFP_NOWARN|GFP_NOIO);
 			if (!b->sets[0].data)
 				goto err;
@@ -884,6 +931,7 @@ out:
 
 	lock_set_subclass(&b->lock.dep_map, level + 1, _THIS_IP_);
 	b->level	= level;
+	b->parent	= (void *) ~0UL;
 
 	mca_reinit(b);
 
@@ -892,7 +940,7 @@ err:
 	if (b)
 		rw_unlock(true, b);
 
-	b = mca_cannibalize(c, k, level, cl);
+	b = mca_cannibalize(c, k);
 	if (!IS_ERR(b))
 		goto out;
 
@@ -903,17 +951,15 @@ err:
  * bch_btree_node_get - find a btree node in the cache and lock it, reading it
  * in from disk if necessary.
  *
- * If IO is necessary, it uses the closure embedded in struct btree_op to wait;
- * if that closure is in non blocking mode, will return -EAGAIN.
+ * If IO is necessary and running under generic_make_request, returns -EAGAIN.
  *
  * The btree node will have either a read or a write lock held, depending on
  * level and op->lock.
  */
 struct btree *bch_btree_node_get(struct cache_set *c, struct bkey *k,
-				 int level, struct btree_op *op)
+				 int level, bool write)
 {
 	int i = 0;
-	bool write = level <= op->lock;
 	struct btree *b;
 
 	BUG_ON(level < 0);
@@ -925,7 +971,7 @@ retry:
 			return ERR_PTR(-EAGAIN);
 
 		mutex_lock(&c->bucket_lock);
-		b = mca_alloc(c, k, level, &op->cl);
+		b = mca_alloc(c, k, level);
 		mutex_unlock(&c->bucket_lock);
 
 		if (!b)
@@ -971,7 +1017,7 @@ static void btree_node_prefetch(struct cache_set *c, struct bkey *k, int level)
 	struct btree *b;
 
 	mutex_lock(&c->bucket_lock);
-	b = mca_alloc(c, k, level, NULL);
+	b = mca_alloc(c, k, level);
 	mutex_unlock(&c->bucket_lock);
 
 	if (!IS_ERR_OR_NULL(b)) {
@@ -982,17 +1028,12 @@ static void btree_node_prefetch(struct cache_set *c, struct bkey *k, int level)
 
 /* Btree alloc */
 
-static void btree_node_free(struct btree *b, struct btree_op *op)
+static void btree_node_free(struct btree *b)
 {
 	unsigned i;
 
 	trace_bcache_btree_node_free(b);
 
-	/*
-	 * The BUG_ON() in btree_node_get() implies that we must have a write
-	 * lock on parent to free or even invalidate a node
-	 */
-	BUG_ON(op->lock <= b->level);
 	BUG_ON(b == b->c->root);
 
 	if (btree_node_dirty(b))
@@ -1015,27 +1056,26 @@ static void btree_node_free(struct btree *b, struct btree_op *op)
 	mutex_unlock(&b->c->bucket_lock);
 }
 
-struct btree *bch_btree_node_alloc(struct cache_set *c, int level,
-				   struct closure *cl)
+struct btree *bch_btree_node_alloc(struct cache_set *c, int level, bool wait)
 {
 	BKEY_PADDED(key) k;
 	struct btree *b = ERR_PTR(-EAGAIN);
 
 	mutex_lock(&c->bucket_lock);
 retry:
-	if (__bch_bucket_alloc_set(c, WATERMARK_METADATA, &k.key, 1, cl))
+	if (__bch_bucket_alloc_set(c, WATERMARK_METADATA, &k.key, 1, wait))
 		goto err;
 
+	bkey_put(c, &k.key);
 	SET_KEY_SIZE(&k.key, c->btree_pages * PAGE_SECTORS);
 
-	b = mca_alloc(c, &k.key, level, cl);
+	b = mca_alloc(c, &k.key, level);
 	if (IS_ERR(b))
 		goto err_free;
 
 	if (!b) {
 		cache_bug(c,
 			"Tried to allocate bucket that was in btree cache");
-		__bkey_put(c, &k.key);
 		goto retry;
 	}
 
@@ -1048,7 +1088,6 @@ retry:
 	return b;
 err_free:
 	bch_bucket_free(c, &k.key);
-	__bkey_put(c, &k.key);
 err:
 	mutex_unlock(&c->bucket_lock);
 
@@ -1056,16 +1095,31 @@ err:
 	return b;
 }
 
-static struct btree *btree_node_alloc_replacement(struct btree *b,
-						  struct closure *cl)
+static struct btree *btree_node_alloc_replacement(struct btree *b, bool wait)
 {
-	struct btree *n = bch_btree_node_alloc(b->c, b->level, cl);
+	struct btree *n = bch_btree_node_alloc(b->c, b->level, wait);
 	if (!IS_ERR_OR_NULL(n))
 		bch_btree_sort_into(b, n);
 
 	return n;
 }
 
+static void make_btree_freeing_key(struct btree *b, struct bkey *k)
+{
+	unsigned i;
+
+	bkey_copy(k, &b->key);
+	bkey_copy_key(k, &ZERO_KEY);
+
+	for (i = 0; i < KEY_PTRS(k); i++) {
+		uint8_t g = PTR_BUCKET(b->c, k, i)->gen + 1;
+
+		SET_PTR_GEN(k, i, g);
+	}
+
+	atomic_inc(&b->c->prio_blocked);
+}
+
 /* Garbage collection */
 
 uint8_t __bch_btree_mark_key(struct cache_set *c, int level, struct bkey *k)
@@ -1119,12 +1173,10 @@ uint8_t __bch_btree_mark_key(struct cache_set *c, int level, struct bkey *k)
 
 #define btree_mark_key(b, k)	__bch_btree_mark_key(b->c, b->level, k)
 
-static int btree_gc_mark_node(struct btree *b, unsigned *keys,
-			      struct gc_stat *gc)
+static bool btree_gc_mark_node(struct btree *b, struct gc_stat *gc)
 {
 	uint8_t stale = 0;
-	unsigned last_dev = -1;
-	struct bcache_device *d = NULL;
+	unsigned keys = 0, good_keys = 0;
 	struct bkey *k;
 	struct btree_iter iter;
 	struct bset_tree *t;
@@ -1132,27 +1184,17 @@ static int btree_gc_mark_node(struct btree *b, unsigned *keys,
 	gc->nodes++;
 
 	for_each_key_filter(b, k, &iter, bch_ptr_invalid) {
-		if (last_dev != KEY_INODE(k)) {
-			last_dev = KEY_INODE(k);
-
-			d = KEY_INODE(k) < b->c->nr_uuids
-				? b->c->devices[last_dev]
-				: NULL;
-		}
-
 		stale = max(stale, btree_mark_key(b, k));
+		keys++;
 
 		if (bch_ptr_bad(b, k))
 			continue;
 
-		*keys += bkey_u64s(k);
-
 		gc->key_bytes += bkey_u64s(k);
 		gc->nkeys++;
+		good_keys++;
 
 		gc->data += KEY_SIZE(k);
-		if (KEY_DIRTY(k))
-			gc->dirty += KEY_SIZE(k);
 	}
 
 	for (t = b->sets; t <= &b->sets[b->nsets]; t++)
@@ -1161,78 +1203,74 @@ static int btree_gc_mark_node(struct btree *b, unsigned *keys,
 			     bkey_cmp(&b->key, &t->end) < 0,
 			     b, "found short btree key in gc");
 
-	return stale;
-}
-
-static struct btree *btree_gc_alloc(struct btree *b, struct bkey *k,
-				    struct btree_op *op)
-{
-	/*
-	 * We block priorities from being written for the duration of garbage
-	 * collection, so we can't sleep in btree_alloc() ->
-	 * bch_bucket_alloc_set(), or we'd risk deadlock - so we don't pass it
-	 * our closure.
-	 */
-	struct btree *n = btree_node_alloc_replacement(b, NULL);
-
-	if (!IS_ERR_OR_NULL(n)) {
-		swap(b, n);
-		__bkey_put(b->c, &b->key);
+	if (b->c->gc_always_rewrite)
+		return true;
 
-		memcpy(k->ptr, b->key.ptr,
-		       sizeof(uint64_t) * KEY_PTRS(&b->key));
+	if (stale > 10)
+		return true;
 
-		btree_node_free(n, op);
-		up_write(&n->lock);
-	}
+	if ((keys - good_keys) * 2 > keys)
+		return true;
 
-	return b;
+	return false;
 }
 
-/*
- * Leaving this at 2 until we've got incremental garbage collection done; it
- * could be higher (and has been tested with 4) except that garbage collection
- * could take much longer, adversely affecting latency.
- */
-#define GC_MERGE_NODES	2U
+#define GC_MERGE_NODES	4U
 
 struct gc_merge_info {
 	struct btree	*b;
-	struct bkey	*k;
 	unsigned	keys;
 };
 
-static void btree_gc_coalesce(struct btree *b, struct btree_op *op,
-			      struct gc_stat *gc, struct gc_merge_info *r)
+static int bch_btree_insert_node(struct btree *, struct btree_op *,
+				 struct keylist *, atomic_t *, struct bkey *);
+
+static int btree_gc_coalesce(struct btree *b, struct btree_op *op,
+			     struct keylist *keylist, struct gc_stat *gc,
+			     struct gc_merge_info *r)
 {
-	unsigned nodes = 0, keys = 0, blocks;
-	int i;
+	unsigned i, nodes = 0, keys = 0, blocks;
+	struct btree *new_nodes[GC_MERGE_NODES];
+	struct closure cl;
+	struct bkey *k;
+
+	memset(new_nodes, 0, sizeof(new_nodes));
+	closure_init_stack(&cl);
 
-	while (nodes < GC_MERGE_NODES && r[nodes].b)
+	while (nodes < GC_MERGE_NODES && !IS_ERR_OR_NULL(r[nodes].b))
 		keys += r[nodes++].keys;
 
 	blocks = btree_default_blocks(b->c) * 2 / 3;
 
 	if (nodes < 2 ||
 	    __set_blocks(b->sets[0].data, keys, b->c) > blocks * (nodes - 1))
-		return;
-
-	for (i = nodes - 1; i >= 0; --i) {
-		if (r[i].b->written)
-			r[i].b = btree_gc_alloc(r[i].b, r[i].k, op);
+		return 0;
 
-		if (r[i].b->written)
-			return;
+	for (i = 0; i < nodes; i++) {
+		new_nodes[i] = btree_node_alloc_replacement(r[i].b, false);
+		if (IS_ERR_OR_NULL(new_nodes[i]))
+			goto out_nocoalesce;
 	}
 
 	for (i = nodes - 1; i > 0; --i) {
-		struct bset *n1 = r[i].b->sets->data;
-		struct bset *n2 = r[i - 1].b->sets->data;
+		struct bset *n1 = new_nodes[i]->sets->data;
+		struct bset *n2 = new_nodes[i - 1]->sets->data;
 		struct bkey *k, *last = NULL;
 
 		keys = 0;
 
-		if (i == 1) {
+		if (i > 1) {
+			for (k = n2->start;
+			     k < end(n2);
+			     k = bkey_next(k)) {
+				if (__set_blocks(n1, n1->keys + keys +
+						 bkey_u64s(k), b->c) > blocks)
+					break;
+
+				last = k;
+				keys += bkey_u64s(k);
+			}
+		} else {
 			/*
 			 * Last node we're not getting rid of - we're getting
 			 * rid of the node at r[0]. Have to try and fit all of
@@ -1241,37 +1279,27 @@ static void btree_gc_coalesce(struct btree *b, struct btree_op *op,
 			 * length keys (shouldn't be possible in practice,
 			 * though)
 			 */
-			if (__set_blocks(n1, n1->keys + r->keys,
-					 b->c) > btree_blocks(r[i].b))
-				return;
+			if (__set_blocks(n1, n1->keys + n2->keys,
+					 b->c) > btree_blocks(new_nodes[i]))
+				goto out_nocoalesce;
 
 			keys = n2->keys;
+			/* Take the key of the node we're getting rid of */
 			last = &r->b->key;
-		} else
-			for (k = n2->start;
-			     k < end(n2);
-			     k = bkey_next(k)) {
-				if (__set_blocks(n1, n1->keys + keys +
-						 bkey_u64s(k), b->c) > blocks)
-					break;
-
-				last = k;
-				keys += bkey_u64s(k);
-			}
+		}
 
 		BUG_ON(__set_blocks(n1, n1->keys + keys,
-				    b->c) > btree_blocks(r[i].b));
+				    b->c) > btree_blocks(new_nodes[i]));
 
-		if (last) {
-			bkey_copy_key(&r[i].b->key, last);
-			bkey_copy_key(r[i].k, last);
-		}
+		if (last)
+			bkey_copy_key(&new_nodes[i]->key, last);
 
 		memcpy(end(n1),
 		       n2->start,
 		       (void *) node(n2, keys) - (void *) n2->start);
 
 		n1->keys += keys;
+		r[i].keys = n1->keys;
 
 		memmove(n2->start,
 			node(n2, keys),
@@ -1279,95 +1307,176 @@ static void btree_gc_coalesce(struct btree *b, struct btree_op *op,
 
 		n2->keys -= keys;
 
-		r[i].keys	= n1->keys;
-		r[i - 1].keys	= n2->keys;
+		if (bch_keylist_realloc(keylist,
+					KEY_PTRS(&new_nodes[i]->key), b->c))
+			goto out_nocoalesce;
+
+		bch_btree_node_write(new_nodes[i], &cl);
+		bch_keylist_add(keylist, &new_nodes[i]->key);
 	}
 
-	btree_node_free(r->b, op);
-	up_write(&r->b->lock);
+	for (i = 0; i < nodes; i++) {
+		if (bch_keylist_realloc(keylist, KEY_PTRS(&r[i].b->key), b->c))
+			goto out_nocoalesce;
 
-	trace_bcache_btree_gc_coalesce(nodes);
+		make_btree_freeing_key(r[i].b, keylist->top);
+		bch_keylist_push(keylist);
+	}
+
+	/* We emptied out this node */
+	BUG_ON(new_nodes[0]->sets->data->keys);
+	btree_node_free(new_nodes[0]);
+	rw_unlock(true, new_nodes[0]);
+
+	closure_sync(&cl);
+
+	for (i = 0; i < nodes; i++) {
+		btree_node_free(r[i].b);
+		rw_unlock(true, r[i].b);
+
+		r[i].b = new_nodes[i];
+	}
+
+	bch_btree_insert_node(b, op, keylist, NULL, NULL);
+	BUG_ON(!bch_keylist_empty(keylist));
+
+	memmove(r, r + 1, sizeof(r[0]) * (nodes - 1));
+	r[nodes - 1].b = ERR_PTR(-EINTR);
 
+	trace_bcache_btree_gc_coalesce(nodes);
 	gc->nodes--;
-	nodes--;
 
-	memmove(&r[0], &r[1], sizeof(struct gc_merge_info) * nodes);
-	memset(&r[nodes], 0, sizeof(struct gc_merge_info));
+	/* Invalidated our iterator */
+	return -EINTR;
+
+out_nocoalesce:
+	closure_sync(&cl);
+
+	while ((k = bch_keylist_pop(keylist)))
+		if (!bkey_cmp(k, &ZERO_KEY))
+			atomic_dec(&b->c->prio_blocked);
+
+	for (i = 0; i < nodes; i++)
+		if (!IS_ERR_OR_NULL(new_nodes[i])) {
+			btree_node_free(new_nodes[i]);
+			rw_unlock(true, new_nodes[i]);
+		}
+	return 0;
 }
 
-static int btree_gc_recurse(struct btree *b, struct btree_op *op,
-			    struct closure *writes, struct gc_stat *gc)
+static unsigned btree_gc_count_keys(struct btree *b)
 {
-	void write(struct btree *r)
-	{
-		if (!r->written)
-			bch_btree_node_write(r, &op->cl);
-		else if (btree_node_dirty(r))
-			bch_btree_node_write(r, writes);
+	struct bkey *k;
+	struct btree_iter iter;
+	unsigned ret = 0;
 
-		up_write(&r->lock);
-	}
+	for_each_key_filter(b, k, &iter, bch_ptr_bad)
+		ret += bkey_u64s(k);
+
+	return ret;
+}
 
-	int ret = 0, stale;
+static int btree_gc_recurse(struct btree *b, struct btree_op *op,
+			    struct closure *writes, struct gc_stat *gc)
+{
 	unsigned i;
+	int ret = 0;
+	bool should_rewrite;
+	struct btree *n;
+	struct bkey *k;
+	struct keylist keys;
+	struct btree_iter iter;
 	struct gc_merge_info r[GC_MERGE_NODES];
+	struct gc_merge_info *last = r + GC_MERGE_NODES - 1;
 
-	memset(r, 0, sizeof(r));
+	bch_keylist_init(&keys);
+	bch_btree_iter_init(b, &iter, &b->c->gc_done);
 
-	while ((r->k = bch_next_recurse_key(b, &b->c->gc_done))) {
-		r->b = bch_btree_node_get(b->c, r->k, b->level - 1, op);
+	for (i = 0; i < GC_MERGE_NODES; i++)
+		r[i].b = ERR_PTR(-EINTR);
 
-		if (IS_ERR(r->b)) {
-			ret = PTR_ERR(r->b);
-			break;
+	while (1) {
+		k = bch_btree_iter_next_filter(&iter, b, bch_ptr_bad);
+		if (k) {
+			r->b = bch_btree_node_get(b->c, k, b->level - 1, true);
+			if (IS_ERR(r->b)) {
+				ret = PTR_ERR(r->b);
+				break;
+			}
+
+			r->keys = btree_gc_count_keys(r->b);
+
+			ret = btree_gc_coalesce(b, op, &keys, gc, r);
+			if (ret)
+				break;
 		}
 
-		r->keys	= 0;
-		stale = btree_gc_mark_node(r->b, &r->keys, gc);
+		if (!last->b)
+			break;
 
-		if (!b->written &&
-		    (r->b->level || stale > 10 ||
-		     b->c->gc_always_rewrite))
-			r->b = btree_gc_alloc(r->b, r->k, op);
+		if (!IS_ERR(last->b)) {
+			should_rewrite = btree_gc_mark_node(last->b, gc);
+			if (should_rewrite) {
+				n = btree_node_alloc_replacement(last->b,
+								 false);
 
-		if (r->b->level)
-			ret = btree_gc_recurse(r->b, op, writes, gc);
+				if (!IS_ERR_OR_NULL(n)) {
+					bch_btree_node_write_sync(n);
+					bch_keylist_add(&keys, &n->key);
 
-		if (ret) {
-			write(r->b);
-			break;
-		}
+					make_btree_freeing_key(last->b,
+							       keys.top);
+					bch_keylist_push(&keys);
+
+					btree_node_free(last->b);
+
+					bch_btree_insert_node(b, op, &keys,
+							      NULL, NULL);
+					BUG_ON(!bch_keylist_empty(&keys));
 
-		bkey_copy_key(&b->c->gc_done, r->k);
+					rw_unlock(true, last->b);
+					last->b = n;
 
-		if (!b->written)
-			btree_gc_coalesce(b, op, gc, r);
+					/* Invalidated our iterator */
+					ret = -EINTR;
+					break;
+				}
+			}
 
-		if (r[GC_MERGE_NODES - 1].b)
-			write(r[GC_MERGE_NODES - 1].b);
+			if (last->b->level) {
+				ret = btree_gc_recurse(last->b, op, writes, gc);
+				if (ret)
+					break;
+			}
 
-		memmove(&r[1], &r[0],
-			sizeof(struct gc_merge_info) * (GC_MERGE_NODES - 1));
+			bkey_copy_key(&b->c->gc_done, &last->b->key);
+
+			/*
+			 * Must flush leaf nodes before gc ends, since replace
+			 * operations aren't journalled
+			 */
+			if (btree_node_dirty(last->b))
+				bch_btree_node_write(last->b, writes);
+			rw_unlock(true, last->b);
+		}
+
+		memmove(r + 1, r, sizeof(r[0]) * (GC_MERGE_NODES - 1));
+		r->b = NULL;
 
-		/* When we've got incremental GC working, we'll want to do
-		 * if (should_resched())
-		 *	return -EAGAIN;
-		 */
-		cond_resched();
-#if 0
 		if (need_resched()) {
 			ret = -EAGAIN;
 			break;
 		}
-#endif
 	}
 
-	for (i = 1; i < GC_MERGE_NODES && r[i].b; i++)
-		write(r[i].b);
+	for (i = 0; i < GC_MERGE_NODES; i++)
+		if (!IS_ERR_OR_NULL(r[i].b)) {
+			if (btree_node_dirty(r[i].b))
+				bch_btree_node_write(r[i].b, writes);
+			rw_unlock(true, r[i].b);
+		}
 
-	/* Might have freed some children, must remove their keys */
-	if (!b->written)
-		bch_btree_sort(b);
+	bch_keylist_free(&keys);
 
 	return ret;
 }
@@ -1376,29 +1485,31 @@ static int bch_btree_gc_root(struct btree *b, struct btree_op *op,
 			     struct closure *writes, struct gc_stat *gc)
 {
 	struct btree *n = NULL;
-	unsigned keys = 0;
-	int ret = 0, stale = btree_gc_mark_node(b, &keys, gc);
-
-	if (b->level || stale > 10)
-		n = btree_node_alloc_replacement(b, NULL);
+	int ret = 0;
+	bool should_rewrite;
 
-	if (!IS_ERR_OR_NULL(n))
-		swap(b, n);
+	should_rewrite = btree_gc_mark_node(b, gc);
+	if (should_rewrite) {
+		n = btree_node_alloc_replacement(b, false);
 
-	if (b->level)
-		ret = btree_gc_recurse(b, op, writes, gc);
+		if (!IS_ERR_OR_NULL(n)) {
+			bch_btree_node_write_sync(n);
+			bch_btree_set_root(n);
+			btree_node_free(b);
+			rw_unlock(true, n);
 
-	if (!b->written || btree_node_dirty(b)) {
-		bch_btree_node_write(b, n ? &op->cl : NULL);
+			return -EINTR;
+		}
 	}
 
-	if (!IS_ERR_OR_NULL(n)) {
-		closure_sync(&op->cl);
-		bch_btree_set_root(b);
-		btree_node_free(n, op);
-		rw_unlock(true, b);
+	if (b->level) {
+		ret = btree_gc_recurse(b, op, writes, gc);
+		if (ret)
+			return ret;
 	}
 
+	bkey_copy_key(&b->c->gc_done, &b->key);
+
 	return ret;
 }
 
@@ -1479,9 +1590,8 @@ size_t bch_btree_gc_finish(struct cache_set *c)
 	return available;
 }
 
-static void bch_btree_gc(struct closure *cl)
+static void bch_btree_gc(struct cache_set *c)
 {
-	struct cache_set *c = container_of(cl, struct cache_set, gc.cl);
 	int ret;
 	unsigned long available;
 	struct gc_stat stats;
@@ -1493,47 +1603,73 @@ static void bch_btree_gc(struct closure *cl)
 
 	memset(&stats, 0, sizeof(struct gc_stat));
 	closure_init_stack(&writes);
-	bch_btree_op_init_stack(&op);
-	op.lock = SHRT_MAX;
+	bch_btree_op_init(&op, SHRT_MAX);
 
 	btree_gc_start(c);
 
-	atomic_inc(&c->prio_blocked);
-
-	ret = btree_root(gc_root, c, &op, &writes, &stats);
-	closure_sync(&op.cl);
-	closure_sync(&writes);
-
-	if (ret) {
-		pr_warn("gc failed!");
-		continue_at(cl, bch_btree_gc, bch_gc_wq);
-	}
+	do {
+		ret = btree_root(gc_root, c, &op, &writes, &stats);
+		closure_sync(&writes);
 
-	/* Possibly wait for new UUIDs or whatever to hit disk */
-	bch_journal_meta(c, &op.cl);
-	closure_sync(&op.cl);
+		if (ret && ret != -EAGAIN)
+			pr_warn("gc failed!");
+	} while (ret);
 
 	available = bch_btree_gc_finish(c);
-
-	atomic_dec(&c->prio_blocked);
 	wake_up_allocators(c);
 
 	bch_time_stats_update(&c->btree_gc_time, start_time);
 
 	stats.key_bytes *= sizeof(uint64_t);
-	stats.dirty	<<= 9;
 	stats.data	<<= 9;
 	stats.in_use	= (c->nbuckets - available) * 100 / c->nbuckets;
 	memcpy(&c->gc_stats, &stats, sizeof(struct gc_stat));
 
 	trace_bcache_gc_end(c);
 
-	continue_at(cl, bch_moving_gc, bch_gc_wq);
+	bch_moving_gc(c);
+}
+
+static int bch_gc_thread(void *arg)
+{
+	struct cache_set *c = arg;
+	struct cache *ca;
+	unsigned i;
+
+	while (1) {
+again:
+		bch_btree_gc(c);
+
+		set_current_state(TASK_INTERRUPTIBLE);
+		if (kthread_should_stop())
+			break;
+
+		mutex_lock(&c->bucket_lock);
+
+		for_each_cache(ca, c, i)
+			if (ca->invalidate_needs_gc) {
+				mutex_unlock(&c->bucket_lock);
+				set_current_state(TASK_RUNNING);
+				goto again;
+			}
+
+		mutex_unlock(&c->bucket_lock);
+
+		try_to_freeze();
+		schedule();
+	}
+
+	return 0;
 }
 
-void bch_queue_gc(struct cache_set *c)
+int bch_gc_thread_start(struct cache_set *c)
 {
-	closure_trylock_call(&c->gc.cl, bch_btree_gc, bch_gc_wq, &c->cl);
+	c->gc_thread = kthread_create(bch_gc_thread, c, "bcache_gc");
+	if (IS_ERR(c->gc_thread))
+		return PTR_ERR(c->gc_thread);
+
+	set_task_state(c->gc_thread, TASK_INTERRUPTIBLE);
+	return 0;
 }
 
 /* Initial partial gc */
@@ -1541,9 +1677,9 @@ void bch_queue_gc(struct cache_set *c)
 static int bch_btree_check_recurse(struct btree *b, struct btree_op *op,
 				   unsigned long **seen)
 {
-	int ret;
+	int ret = 0;
 	unsigned i;
-	struct bkey *k;
+	struct bkey *k, *p = NULL;
 	struct bucket *g;
 	struct btree_iter iter;
 
@@ -1570,31 +1706,32 @@ static int bch_btree_check_recurse(struct btree *b, struct btree_op *op,
 	}
 
 	if (b->level) {
-		k = bch_next_recurse_key(b, &ZERO_KEY);
+		bch_btree_iter_init(b, &iter, NULL);
 
-		while (k) {
-			struct bkey *p = bch_next_recurse_key(b, k);
-			if (p)
-				btree_node_prefetch(b->c, p, b->level - 1);
+		do {
+			k = bch_btree_iter_next_filter(&iter, b, bch_ptr_bad);
+			if (k)
+				btree_node_prefetch(b->c, k, b->level - 1);
 
-			ret = btree(check_recurse, k, b, op, seen);
-			if (ret)
-				return ret;
+			if (p)
+				ret = btree(check_recurse, p, b, op, seen);
 
-			k = p;
-		}
+			p = k;
+		} while (p && !ret);
 	}
 
 	return 0;
 }
 
-int bch_btree_check(struct cache_set *c, struct btree_op *op)
+int bch_btree_check(struct cache_set *c)
 {
 	int ret = -ENOMEM;
 	unsigned i;
 	unsigned long *seen[MAX_CACHES_PER_SET];
+	struct btree_op op;
 
 	memset(seen, 0, sizeof(seen));
+	bch_btree_op_init(&op, SHRT_MAX);
 
 	for (i = 0; c->cache[i]; i++) {
 		size_t n = DIV_ROUND_UP(c->cache[i]->sb.nbuckets, 8);
@@ -1606,7 +1743,7 @@ int bch_btree_check(struct cache_set *c, struct btree_op *op)
 		memset(seen[i], 0xFF, n);
 	}
 
-	ret = btree_root(check_recurse, c, op, seen);
+	ret = btree_root(check_recurse, c, &op, seen);
 err:
 	for (i = 0; i < MAX_CACHES_PER_SET; i++)
 		kfree(seen[i]);
@@ -1628,10 +1765,9 @@ static void shift_keys(struct btree *b, struct bkey *where, struct bkey *insert)
 	bch_bset_fix_lookup_table(b, where);
 }
 
-static bool fix_overlapping_extents(struct btree *b,
-				    struct bkey *insert,
+static bool fix_overlapping_extents(struct btree *b, struct bkey *insert,
 				    struct btree_iter *iter,
-				    struct btree_op *op)
+				    struct bkey *replace_key)
 {
 	void subtract_dirty(struct bkey *k, uint64_t offset, int sectors)
 	{
@@ -1659,39 +1795,38 @@ static bool fix_overlapping_extents(struct btree *b,
 		 * We might overlap with 0 size extents; we can't skip these
 		 * because if they're in the set we're inserting to we have to
 		 * adjust them so they don't overlap with the key we're
-		 * inserting. But we don't want to check them for BTREE_REPLACE
+		 * inserting. But we don't want to check them for replace
 		 * operations.
 		 */
 
-		if (op->type == BTREE_REPLACE &&
-		    KEY_SIZE(k)) {
+		if (replace_key && KEY_SIZE(k)) {
 			/*
 			 * k might have been split since we inserted/found the
 			 * key we're replacing
 			 */
 			unsigned i;
 			uint64_t offset = KEY_START(k) -
-				KEY_START(&op->replace);
+				KEY_START(replace_key);
 
 			/* But it must be a subset of the replace key */
-			if (KEY_START(k) < KEY_START(&op->replace) ||
-			    KEY_OFFSET(k) > KEY_OFFSET(&op->replace))
+			if (KEY_START(k) < KEY_START(replace_key) ||
+			    KEY_OFFSET(k) > KEY_OFFSET(replace_key))
 				goto check_failed;
 
 			/* We didn't find a key that we were supposed to */
 			if (KEY_START(k) > KEY_START(insert) + sectors_found)
 				goto check_failed;
 
-			if (KEY_PTRS(&op->replace) != KEY_PTRS(k))
+			if (KEY_PTRS(replace_key) != KEY_PTRS(k))
 				goto check_failed;
 
 			/* skip past gen */
 			offset <<= 8;
 
-			BUG_ON(!KEY_PTRS(&op->replace));
+			BUG_ON(!KEY_PTRS(replace_key));
 
-			for (i = 0; i < KEY_PTRS(&op->replace); i++)
-				if (k->ptr[i] != op->replace.ptr[i] + offset)
+			for (i = 0; i < KEY_PTRS(replace_key); i++)
+				if (k->ptr[i] != replace_key->ptr[i] + offset)
 					goto check_failed;
 
 			sectors_found = KEY_OFFSET(k) - KEY_START(insert);
@@ -1742,6 +1877,9 @@ static bool fix_overlapping_extents(struct btree *b,
 		if (bkey_cmp(insert, k) < 0) {
 			bch_cut_front(insert, k);
 		} else {
+			if (bkey_cmp(&START_KEY(insert), &START_KEY(k)) > 0)
+				old_offset = KEY_START(insert);
+
 			if (bkey_written(b, k) &&
 			    bkey_cmp(&START_KEY(insert), &START_KEY(k)) <= 0) {
 				/*
@@ -1759,9 +1897,8 @@ static bool fix_overlapping_extents(struct btree *b,
 	}
 
 check_failed:
-	if (op->type == BTREE_REPLACE) {
+	if (replace_key) {
 		if (!sectors_found) {
-			op->insert_collision = true;
 			return true;
 		} else if (sectors_found < KEY_SIZE(insert)) {
 			SET_KEY_OFFSET(insert, KEY_OFFSET(insert) -
@@ -1774,7 +1911,7 @@ check_failed:
 }
 
 static bool btree_insert_key(struct btree *b, struct btree_op *op,
-			     struct bkey *k)
+			     struct bkey *k, struct bkey *replace_key)
 {
 	struct bset *i = b->sets[b->nsets].data;
 	struct bkey *m, *prev;
@@ -1786,22 +1923,23 @@ static bool btree_insert_key(struct btree *b, struct btree_op *op,
 
 	if (!b->level) {
 		struct btree_iter iter;
-		struct bkey search = KEY(KEY_INODE(k), KEY_START(k), 0);
 
 		/*
 		 * bset_search() returns the first key that is strictly greater
 		 * than the search key - but for back merging, we want to find
-		 * the first key that is greater than or equal to KEY_START(k) -
-		 * unless KEY_START(k) is 0.
+		 * the previous key.
 		 */
-		if (KEY_OFFSET(&search))
-			SET_KEY_OFFSET(&search, KEY_OFFSET(&search) - 1);
-
 		prev = NULL;
-		m = bch_btree_iter_init(b, &iter, &search);
+		m = bch_btree_iter_init(b, &iter, PRECEDING_KEY(&START_KEY(k)));
 
-		if (fix_overlapping_extents(b, k, &iter, op))
+		if (fix_overlapping_extents(b, k, &iter, replace_key)) {
+			op->insert_collision = true;
 			return false;
+		}
+
+		if (KEY_DIRTY(k))
+			bcache_dev_sectors_dirty_add(b->c, KEY_INODE(k),
+						     KEY_START(k), KEY_SIZE(k));
 
 		while (m != end(i) &&
 		       bkey_cmp(k, &START_KEY(m)) > 0)
@@ -1825,84 +1963,80 @@ static bool btree_insert_key(struct btree *b, struct btree_op *op,
 		if (m != end(i) &&
 		    bch_bkey_try_merge(b, k, m))
 			goto copy;
-	} else
+	} else {
+		BUG_ON(replace_key);
 		m = bch_bset_search(b, &b->sets[b->nsets], k);
+	}
 
 insert:	shift_keys(b, m, k);
 copy:	bkey_copy(m, k);
 merged:
-	if (KEY_DIRTY(k))
-		bcache_dev_sectors_dirty_add(b->c, KEY_INODE(k),
-					     KEY_START(k), KEY_SIZE(k));
-
-	bch_check_keys(b, "%u for %s", status, op_type(op));
+	bch_check_keys(b, "%u for %s", status,
+		       replace_key ? "replace" : "insert");
 
 	if (b->level && !KEY_OFFSET(k))
 		btree_current_write(b)->prio_blocked++;
 
-	trace_bcache_btree_insert_key(b, k, op->type, status);
+	trace_bcache_btree_insert_key(b, k, replace_key != NULL, status);
 
 	return true;
 }
 
-static bool bch_btree_insert_keys(struct btree *b, struct btree_op *op)
+static bool bch_btree_insert_keys(struct btree *b, struct btree_op *op,
+				  struct keylist *insert_keys,
+				  struct bkey *replace_key)
 {
 	bool ret = false;
-	struct bkey *k;
-	unsigned oldsize = bch_count_data(b);
-
-	while ((k = bch_keylist_pop(&op->keys))) {
-		bkey_put(b->c, k, b->level);
-		ret |= btree_insert_key(b, op, k);
-	}
-
-	BUG_ON(bch_count_data(b) < oldsize);
-	return ret;
-}
+	int oldsize = bch_count_data(b);
 
-bool bch_btree_insert_check_key(struct btree *b, struct btree_op *op,
-				   struct bio *bio)
-{
-	bool ret = false;
-	uint64_t btree_ptr = b->key.ptr[0];
-	unsigned long seq = b->seq;
-	BKEY_PADDED(k) tmp;
+	while (!bch_keylist_empty(insert_keys)) {
+		struct bset *i = write_block(b);
+		struct bkey *k = insert_keys->keys;
 
-	rw_unlock(false, b);
-	rw_lock(true, b, b->level);
+		if (b->written + __set_blocks(i, i->keys + bkey_u64s(k), b->c)
+		    > btree_blocks(b))
+			break;
 
-	if (b->key.ptr[0] != btree_ptr ||
-	    b->seq != seq + 1 ||
-	    should_split(b))
-		goto out;
+		if (bkey_cmp(k, &b->key) <= 0) {
+			if (!b->level)
+				bkey_put(b->c, k);
 
-	op->replace = KEY(op->inode, bio_end_sector(bio), bio_sectors(bio));
+			ret |= btree_insert_key(b, op, k, replace_key);
+			bch_keylist_pop_front(insert_keys);
+		} else if (bkey_cmp(&START_KEY(k), &b->key) < 0) {
+			BKEY_PADDED(key) temp;
+			bkey_copy(&temp.key, insert_keys->keys);
 
-	SET_KEY_PTRS(&op->replace, 1);
-	get_random_bytes(&op->replace.ptr[0], sizeof(uint64_t));
+			bch_cut_back(&b->key, &temp.key);
+			bch_cut_front(&b->key, insert_keys->keys);
 
-	SET_PTR_DEV(&op->replace, 0, PTR_CHECK_DEV);
+			ret |= btree_insert_key(b, op, &temp.key, replace_key);
+			break;
+		} else {
+			break;
+		}
+	}
 
-	bkey_copy(&tmp.k, &op->replace);
+	BUG_ON(!bch_keylist_empty(insert_keys) && b->level);
 
-	BUG_ON(op->type != BTREE_INSERT);
-	BUG_ON(!btree_insert_key(b, op, &tmp.k));
-	ret = true;
-out:
-	downgrade_write(&b->lock);
+	BUG_ON(bch_count_data(b) < oldsize);
 	return ret;
 }
 
-static int btree_split(struct btree *b, struct btree_op *op)
+static int btree_split(struct btree *b, struct btree_op *op,
+		       struct keylist *insert_keys,
+		       struct bkey *replace_key)
 {
-	bool split, root = b == b->c->root;
+	bool split;
 	struct btree *n1, *n2 = NULL, *n3 = NULL;
 	uint64_t start_time = local_clock();
+	struct closure cl;
+	struct keylist parent_keys;
 
-	if (b->level)
-		set_closure_blocking(&op->cl);
+	closure_init_stack(&cl);
+	bch_keylist_init(&parent_keys);
 
-	n1 = btree_node_alloc_replacement(b, &op->cl);
+	n1 = btree_node_alloc_replacement(b, true);
 	if (IS_ERR(n1))
 		goto err;
 
@@ -1913,19 +2047,20 @@ static int btree_split(struct btree *b, struct btree_op *op)
 
 		trace_bcache_btree_node_split(b, n1->sets[0].data->keys);
 
-		n2 = bch_btree_node_alloc(b->c, b->level, &op->cl);
+		n2 = bch_btree_node_alloc(b->c, b->level, true);
 		if (IS_ERR(n2))
 			goto err_free1;
 
-		if (root) {
-			n3 = bch_btree_node_alloc(b->c, b->level + 1, &op->cl);
+		if (!b->parent) {
+			n3 = bch_btree_node_alloc(b->c, b->level + 1, true);
 			if (IS_ERR(n3))
 				goto err_free2;
 		}
 
-		bch_btree_insert_keys(n1, op);
+		bch_btree_insert_keys(n1, op, insert_keys, replace_key);
 
-		/* Has to be a linear search because we don't have an auxiliary
+		/*
+		 * Has to be a linear search because we don't have an auxiliary
 		 * search tree yet
 		 */
 
@@ -1944,60 +2079,57 @@ static int btree_split(struct btree *b, struct btree_op *op)
 
 		bkey_copy_key(&n2->key, &b->key);
 
-		bch_keylist_add(&op->keys, &n2->key);
-		bch_btree_node_write(n2, &op->cl);
+		bch_keylist_add(&parent_keys, &n2->key);
+		bch_btree_node_write(n2, &cl);
 		rw_unlock(true, n2);
 	} else {
 		trace_bcache_btree_node_compact(b, n1->sets[0].data->keys);
 
-		bch_btree_insert_keys(n1, op);
+		bch_btree_insert_keys(n1, op, insert_keys, replace_key);
 	}
 
-	bch_keylist_add(&op->keys, &n1->key);
-	bch_btree_node_write(n1, &op->cl);
+	bch_keylist_add(&parent_keys, &n1->key);
+	bch_btree_node_write(n1, &cl);
 
 	if (n3) {
+		/* Depth increases, make a new root */
 		bkey_copy_key(&n3->key, &MAX_KEY);
-		bch_btree_insert_keys(n3, op);
-		bch_btree_node_write(n3, &op->cl);
+		bch_btree_insert_keys(n3, op, &parent_keys, NULL);
+		bch_btree_node_write(n3, &cl);
 
-		closure_sync(&op->cl);
+		closure_sync(&cl);
 		bch_btree_set_root(n3);
 		rw_unlock(true, n3);
-	} else if (root) {
-		op->keys.top = op->keys.bottom;
-		closure_sync(&op->cl);
-		bch_btree_set_root(n1);
-	} else {
-		unsigned i;
 
-		bkey_copy(op->keys.top, &b->key);
-		bkey_copy_key(op->keys.top, &ZERO_KEY);
+		btree_node_free(b);
+	} else if (!b->parent) {
+		/* Root filled up but didn't need to be split */
+		closure_sync(&cl);
+		bch_btree_set_root(n1);
 
-		for (i = 0; i < KEY_PTRS(&b->key); i++) {
-			uint8_t g = PTR_BUCKET(b->c, &b->key, i)->gen + 1;
+		btree_node_free(b);
+	} else {
+		/* Split a non root node */
+		closure_sync(&cl);
+		make_btree_freeing_key(b, parent_keys.top);
+		bch_keylist_push(&parent_keys);
 
-			SET_PTR_GEN(op->keys.top, i, g);
-		}
+		btree_node_free(b);
 
-		bch_keylist_push(&op->keys);
-		closure_sync(&op->cl);
-		atomic_inc(&b->c->prio_blocked);
+		bch_btree_insert_node(b->parent, op, &parent_keys, NULL, NULL);
+		BUG_ON(!bch_keylist_empty(&parent_keys));
 	}
 
 	rw_unlock(true, n1);
-	btree_node_free(b, op);
 
 	bch_time_stats_update(&b->c->btree_split_time, start_time);
 
 	return 0;
 err_free2:
-	__bkey_put(n2->c, &n2->key);
-	btree_node_free(n2, op);
+	btree_node_free(n2);
 	rw_unlock(true, n2);
 err_free1:
-	__bkey_put(n1->c, &n1->key);
-	btree_node_free(n1, op);
+	btree_node_free(n1);
 	rw_unlock(true, n1);
 err:
 	if (n3 == ERR_PTR(-EAGAIN) ||
@@ -2009,116 +2141,126 @@ err:
 	return -ENOMEM;
 }
 
-static int bch_btree_insert_recurse(struct btree *b, struct btree_op *op,
-				    struct keylist *stack_keys)
+static int bch_btree_insert_node(struct btree *b, struct btree_op *op,
+				 struct keylist *insert_keys,
+				 atomic_t *journal_ref,
+				 struct bkey *replace_key)
 {
-	if (b->level) {
-		int ret;
-		struct bkey *insert = op->keys.bottom;
-		struct bkey *k = bch_next_recurse_key(b, &START_KEY(insert));
-
-		if (!k) {
-			btree_bug(b, "no key to recurse on at level %i/%i",
-				  b->level, b->c->root->level);
+	BUG_ON(b->level && replace_key);
 
-			op->keys.top = op->keys.bottom;
-			return -EIO;
+	if (should_split(b)) {
+		if (current->bio_list) {
+			op->lock = b->c->root->level + 1;
+			return -EAGAIN;
+		} else if (op->lock <= b->c->root->level) {
+			op->lock = b->c->root->level + 1;
+			return -EINTR;
+		} else {
+			/* Invalidated all iterators */
+			return btree_split(b, op, insert_keys, replace_key) ?:
+				-EINTR;
 		}
+	} else {
+		BUG_ON(write_block(b) != b->sets[b->nsets].data);
 
-		if (bkey_cmp(insert, k) > 0) {
-			unsigned i;
-
-			if (op->type == BTREE_REPLACE) {
-				__bkey_put(b->c, insert);
-				op->keys.top = op->keys.bottom;
-				op->insert_collision = true;
-				return 0;
-			}
+		if (bch_btree_insert_keys(b, op, insert_keys, replace_key)) {
+			if (!b->level)
+				bch_btree_leaf_dirty(b, journal_ref);
+			else
+				bch_btree_node_write_sync(b);
+		}
 
-			for (i = 0; i < KEY_PTRS(insert); i++)
-				atomic_inc(&PTR_BUCKET(b->c, insert, i)->pin);
+		return 0;
+	}
+}
 
-			bkey_copy(stack_keys->top, insert);
+int bch_btree_insert_check_key(struct btree *b, struct btree_op *op,
+			       struct bkey *check_key)
+{
+	int ret = -EINTR;
+	uint64_t btree_ptr = b->key.ptr[0];
+	unsigned long seq = b->seq;
+	struct keylist insert;
+	bool upgrade = op->lock == -1;
 
-			bch_cut_back(k, insert);
-			bch_cut_front(k, stack_keys->top);
+	bch_keylist_init(&insert);
 
-			bch_keylist_push(stack_keys);
-		}
+	if (upgrade) {
+		rw_unlock(false, b);
+		rw_lock(true, b, b->level);
 
-		ret = btree(insert_recurse, k, b, op, stack_keys);
-		if (ret)
-			return ret;
+		if (b->key.ptr[0] != btree_ptr ||
+		    b->seq != seq + 1)
+			goto out;
 	}
 
-	if (!bch_keylist_empty(&op->keys)) {
-		if (should_split(b)) {
-			if (op->lock <= b->c->root->level) {
-				BUG_ON(b->level);
-				op->lock = b->c->root->level + 1;
-				return -EINTR;
-			}
-			return btree_split(b, op);
-		}
+	SET_KEY_PTRS(check_key, 1);
+	get_random_bytes(&check_key->ptr[0], sizeof(uint64_t));
 
-		BUG_ON(write_block(b) != b->sets[b->nsets].data);
+	SET_PTR_DEV(check_key, 0, PTR_CHECK_DEV);
 
-		if (bch_btree_insert_keys(b, op)) {
-			if (!b->level)
-				bch_btree_leaf_dirty(b, op);
-			else
-				bch_btree_node_write(b, &op->cl);
-		}
-	}
+	bch_keylist_add(&insert, check_key);
 
-	return 0;
+	ret = bch_btree_insert_node(b, op, &insert, NULL, NULL);
+
+	BUG_ON(!ret && !bch_keylist_empty(&insert));
+out:
+	if (upgrade)
+		downgrade_write(&b->lock);
+	return ret;
 }
 
-int bch_btree_insert(struct btree_op *op, struct cache_set *c)
+struct btree_insert_op {
+	struct btree_op	op;
+	struct keylist	*keys;
+	atomic_t	*journal_ref;
+	struct bkey	*replace_key;
+};
+
+int btree_insert_fn(struct btree_op *b_op, struct btree *b)
 {
-	int ret = 0;
-	struct keylist stack_keys;
+	struct btree_insert_op *op = container_of(b_op,
+					struct btree_insert_op, op);
 
-	/*
-	 * Don't want to block with the btree locked unless we have to,
-	 * otherwise we get deadlocks with try_harder and between split/gc
-	 */
-	clear_closure_blocking(&op->cl);
-
-	BUG_ON(bch_keylist_empty(&op->keys));
-	bch_keylist_copy(&stack_keys, &op->keys);
-	bch_keylist_init(&op->keys);
-
-	while (!bch_keylist_empty(&stack_keys) ||
-	       !bch_keylist_empty(&op->keys)) {
-		if (bch_keylist_empty(&op->keys)) {
-			bch_keylist_add(&op->keys,
-					bch_keylist_pop(&stack_keys));
-			op->lock = 0;
-		}
+	int ret = bch_btree_insert_node(b, &op->op, op->keys,
+					op->journal_ref, op->replace_key);
+	if (ret && !bch_keylist_empty(op->keys))
+		return ret;
+	else
+		return MAP_DONE;
+}
 
-		ret = btree_root(insert_recurse, c, op, &stack_keys);
+int bch_btree_insert(struct cache_set *c, struct keylist *keys,
+		     atomic_t *journal_ref, struct bkey *replace_key)
+{
+	struct btree_insert_op op;
+	int ret = 0;
 
-		if (ret == -EAGAIN) {
-			ret = 0;
-			closure_sync(&op->cl);
-		} else if (ret) {
-			struct bkey *k;
+	BUG_ON(current->bio_list);
+	BUG_ON(bch_keylist_empty(keys));
+
+	bch_btree_op_init(&op.op, 0);
+	op.keys		= keys;
+	op.journal_ref	= journal_ref;
+	op.replace_key	= replace_key;
+
+	while (!ret && !bch_keylist_empty(keys)) {
+		op.op.lock = 0;
+		ret = bch_btree_map_leaf_nodes(&op.op, c,
+					       &START_KEY(keys->keys),
+					       btree_insert_fn);
+	}
 
-			pr_err("error %i trying to insert key for %s",
-			       ret, op_type(op));
+	if (ret) {
+		struct bkey *k;
 
-			while ((k = bch_keylist_pop(&stack_keys) ?:
-				    bch_keylist_pop(&op->keys)))
-				bkey_put(c, k, 0);
-		}
-	}
+		pr_err("error %i", ret);
 
-	bch_keylist_free(&stack_keys);
+		while ((k = bch_keylist_pop(keys)))
+			bkey_put(c, k);
+	} else if (op.op.insert_collision)
+		ret = -ESRCH;
 
-	if (op->journal)
-		atomic_dec_bug(op->journal);
-	op->journal = NULL;
 	return ret;
 }
 
@@ -2141,132 +2283,81 @@ void bch_btree_set_root(struct btree *b)
 	mutex_unlock(&b->c->bucket_lock);
 
 	b->c->root = b;
-	__bkey_put(b->c, &b->key);
 
 	bch_journal_meta(b->c, &cl);
 	closure_sync(&cl);
 }
 
-/* Cache lookup */
+/* Map across nodes or keys */
 
-static int submit_partial_cache_miss(struct btree *b, struct btree_op *op,
-				     struct bkey *k)
+static int bch_btree_map_nodes_recurse(struct btree *b, struct btree_op *op,
+				       struct bkey *from,
+				       btree_map_nodes_fn *fn, int flags)
 {
-	struct search *s = container_of(op, struct search, op);
-	struct bio *bio = &s->bio.bio;
-	int ret = 0;
+	int ret = MAP_CONTINUE;
+
+	if (b->level) {
+		struct bkey *k;
+		struct btree_iter iter;
 
-	while (!ret &&
-	       !op->lookup_done) {
-		unsigned sectors = INT_MAX;
+		bch_btree_iter_init(b, &iter, from);
 
-		if (KEY_INODE(k) == op->inode) {
-			if (KEY_START(k) <= bio->bi_sector)
-				break;
+		while ((k = bch_btree_iter_next_filter(&iter, b,
+						       bch_ptr_bad))) {
+			ret = btree(map_nodes_recurse, k, b,
+				    op, from, fn, flags);
+			from = NULL;
 
-			sectors = min_t(uint64_t, sectors,
-					KEY_START(k) - bio->bi_sector);
+			if (ret != MAP_CONTINUE)
+				return ret;
 		}
-
-		ret = s->d->cache_miss(b, s, bio, sectors);
 	}
 
+	if (!b->level || flags == MAP_ALL_NODES)
+		ret = fn(op, b);
+
 	return ret;
 }
 
-/*
- * Read from a single key, handling the initial cache miss if the key starts in
- * the middle of the bio
- */
-static int submit_partial_cache_hit(struct btree *b, struct btree_op *op,
-				    struct bkey *k)
+int __bch_btree_map_nodes(struct btree_op *op, struct cache_set *c,
+			  struct bkey *from, btree_map_nodes_fn *fn, int flags)
 {
-	struct search *s = container_of(op, struct search, op);
-	struct bio *bio = &s->bio.bio;
-	unsigned ptr;
-	struct bio *n;
-
-	int ret = submit_partial_cache_miss(b, op, k);
-	if (ret || op->lookup_done)
-		return ret;
-
-	/* XXX: figure out best pointer - for multiple cache devices */
-	ptr = 0;
-
-	PTR_BUCKET(b->c, k, ptr)->prio = INITIAL_PRIO;
-
-	while (!op->lookup_done &&
-	       KEY_INODE(k) == op->inode &&
-	       bio->bi_sector < KEY_OFFSET(k)) {
-		struct bkey *bio_key;
-		sector_t sector = PTR_OFFSET(k, ptr) +
-			(bio->bi_sector - KEY_START(k));
-		unsigned sectors = min_t(uint64_t, INT_MAX,
-					 KEY_OFFSET(k) - bio->bi_sector);
-
-		n = bch_bio_split(bio, sectors, GFP_NOIO, s->d->bio_split);
-		if (n == bio)
-			op->lookup_done = true;
-
-		bio_key = &container_of(n, struct bbio, bio)->key;
-
-		/*
-		 * The bucket we're reading from might be reused while our bio
-		 * is in flight, and we could then end up reading the wrong
-		 * data.
-		 *
-		 * We guard against this by checking (in cache_read_endio()) if
-		 * the pointer is stale again; if so, we treat it as an error
-		 * and reread from the backing device (but we don't pass that
-		 * error up anywhere).
-		 */
-
-		bch_bkey_copy_single_ptr(bio_key, k, ptr);
-		SET_PTR_OFFSET(bio_key, 0, sector);
-
-		n->bi_end_io	= bch_cache_read_endio;
-		n->bi_private	= &s->cl;
-
-		__bch_submit_bbio(n, b->c);
-	}
-
-	return 0;
+	return btree_root(map_nodes_recurse, c, op, from, fn, flags);
 }
 
-int bch_btree_search_recurse(struct btree *b, struct btree_op *op)
+static int bch_btree_map_keys_recurse(struct btree *b, struct btree_op *op,
+				      struct bkey *from, btree_map_keys_fn *fn,
+				      int flags)
 {
-	struct search *s = container_of(op, struct search, op);
-	struct bio *bio = &s->bio.bio;
-
-	int ret = 0;
+	int ret = MAP_CONTINUE;
 	struct bkey *k;
 	struct btree_iter iter;
-	bch_btree_iter_init(b, &iter, &KEY(op->inode, bio->bi_sector, 0));
 
-	do {
-		k = bch_btree_iter_next_filter(&iter, b, bch_ptr_bad);
-		if (!k) {
-			/*
-			 * b->key would be exactly what we want, except that
-			 * pointers to btree nodes have nonzero size - we
-			 * wouldn't go far enough
-			 */
+	bch_btree_iter_init(b, &iter, from);
 
-			ret = submit_partial_cache_miss(b, op,
-					&KEY(KEY_INODE(&b->key),
-					     KEY_OFFSET(&b->key), 0));
-			break;
-		}
+	while ((k = bch_btree_iter_next_filter(&iter, b, bch_ptr_bad))) {
+		ret = !b->level
+			? fn(op, b, k)
+			: btree(map_keys_recurse, k, b, op, from, fn, flags);
+		from = NULL;
+
+		if (ret != MAP_CONTINUE)
+			return ret;
+	}
 
-		ret = b->level
-			? btree(search_recurse, k, b, op)
-			: submit_partial_cache_hit(b, op, k);
-	} while (!ret &&
-		 !op->lookup_done);
+	if (!b->level && (flags & MAP_END_KEY))
+		ret = fn(op, b, &KEY(KEY_INODE(&b->key),
+				     KEY_OFFSET(&b->key), 0));
 
 	return ret;
 }
 
+int bch_btree_map_keys(struct btree_op *op, struct cache_set *c,
+		       struct bkey *from, btree_map_keys_fn *fn, int flags)
+{
+	return btree_root(map_keys_recurse, c, op, from, fn, flags);
+}
+
 /* Keybuf code */
 
 static inline int keybuf_cmp(struct keybuf_key *l, struct keybuf_key *r)
@@ -2285,80 +2376,79 @@ static inline int keybuf_nonoverlapping_cmp(struct keybuf_key *l,
 	return clamp_t(int64_t, bkey_cmp(&l->key, &r->key), -1, 1);
 }
 
-static int bch_btree_refill_keybuf(struct btree *b, struct btree_op *op,
-				   struct keybuf *buf, struct bkey *end,
-				   keybuf_pred_fn *pred)
-{
-	struct btree_iter iter;
-	bch_btree_iter_init(b, &iter, &buf->last_scanned);
-
-	while (!array_freelist_empty(&buf->freelist)) {
-		struct bkey *k = bch_btree_iter_next_filter(&iter, b,
-							    bch_ptr_bad);
-
-		if (!b->level) {
-			if (!k) {
-				buf->last_scanned = b->key;
-				break;
-			}
+struct refill {
+	struct btree_op	op;
+	unsigned	nr_found;
+	struct keybuf	*buf;
+	struct bkey	*end;
+	keybuf_pred_fn	*pred;
+};
 
-			buf->last_scanned = *k;
-			if (bkey_cmp(&buf->last_scanned, end) >= 0)
-				break;
+static int refill_keybuf_fn(struct btree_op *op, struct btree *b,
+			    struct bkey *k)
+{
+	struct refill *refill = container_of(op, struct refill, op);
+	struct keybuf *buf = refill->buf;
+	int ret = MAP_CONTINUE;
 
-			if (pred(buf, k)) {
-				struct keybuf_key *w;
+	if (bkey_cmp(k, refill->end) >= 0) {
+		ret = MAP_DONE;
+		goto out;
+	}
 
-				spin_lock(&buf->lock);
+	if (!KEY_SIZE(k)) /* end key */
+		goto out;
 
-				w = array_alloc(&buf->freelist);
+	if (refill->pred(buf, k)) {
+		struct keybuf_key *w;
 
-				w->private = NULL;
-				bkey_copy(&w->key, k);
+		spin_lock(&buf->lock);
 
-				if (RB_INSERT(&buf->keys, w, node, keybuf_cmp))
-					array_free(&buf->freelist, w);
+		w = array_alloc(&buf->freelist);
+		if (!w) {
+			spin_unlock(&buf->lock);
+			return MAP_DONE;
+		}
 
-				spin_unlock(&buf->lock);
-			}
-		} else {
-			if (!k)
-				break;
+		w->private = NULL;
+		bkey_copy(&w->key, k);
 
-			btree(refill_keybuf, k, b, op, buf, end, pred);
-			/*
-			 * Might get an error here, but can't really do anything
-			 * and it'll get logged elsewhere. Just read what we
-			 * can.
-			 */
+		if (RB_INSERT(&buf->keys, w, node, keybuf_cmp))
+			array_free(&buf->freelist, w);
+		else
+			refill->nr_found++;
 
-			if (bkey_cmp(&buf->last_scanned, end) >= 0)
-				break;
+		if (array_freelist_empty(&buf->freelist))
+			ret = MAP_DONE;
 
-			cond_resched();
-		}
+		spin_unlock(&buf->lock);
 	}
-
-	return 0;
+out:
+	buf->last_scanned = *k;
+	return ret;
 }
 
 void bch_refill_keybuf(struct cache_set *c, struct keybuf *buf,
 		       struct bkey *end, keybuf_pred_fn *pred)
 {
 	struct bkey start = buf->last_scanned;
-	struct btree_op op;
-	bch_btree_op_init_stack(&op);
+	struct refill refill;
 
 	cond_resched();
 
-	btree_root(refill_keybuf, c, &op, buf, end, pred);
-	closure_sync(&op.cl);
+	bch_btree_op_init(&refill.op, -1);
+	refill.nr_found	= 0;
+	refill.buf	= buf;
+	refill.end	= end;
+	refill.pred	= pred;
+
+	bch_btree_map_keys(&refill.op, c, &buf->last_scanned,
+			   refill_keybuf_fn, MAP_END_KEY);
 
-	pr_debug("found %s keys from %llu:%llu to %llu:%llu",
-		 RB_EMPTY_ROOT(&buf->keys) ? "no" :
-		 array_freelist_empty(&buf->freelist) ? "some" : "a few",
-		 KEY_INODE(&start), KEY_OFFSET(&start),
-		 KEY_INODE(&buf->last_scanned), KEY_OFFSET(&buf->last_scanned));
+	trace_bcache_keyscan(refill.nr_found,
+			     KEY_INODE(&start), KEY_OFFSET(&start),
+			     KEY_INODE(&buf->last_scanned),
+			     KEY_OFFSET(&buf->last_scanned));
 
 	spin_lock(&buf->lock);
 
@@ -2436,9 +2526,9 @@ struct keybuf_key *bch_keybuf_next(struct keybuf *buf)
 }
 
 struct keybuf_key *bch_keybuf_next_rescan(struct cache_set *c,
-					     struct keybuf *buf,
-					     struct bkey *end,
-					     keybuf_pred_fn *pred)
+					  struct keybuf *buf,
+					  struct bkey *end,
+					  keybuf_pred_fn *pred)
 {
 	struct keybuf_key *ret;
 
@@ -2471,14 +2561,12 @@ void bch_btree_exit(void)
 {
 	if (btree_io_wq)
 		destroy_workqueue(btree_io_wq);
-	if (bch_gc_wq)
-		destroy_workqueue(bch_gc_wq);
 }
 
 int __init bch_btree_init(void)
 {
-	if (!(bch_gc_wq = create_singlethread_workqueue("bch_btree_gc")) ||
-	    !(btree_io_wq = create_singlethread_workqueue("bch_btree_io")))
+	btree_io_wq = create_singlethread_workqueue("bch_btree_io");
+	if (!btree_io_wq)
 		return -ENOMEM;
 
 	return 0;
diff --git a/drivers/md/bcache/btree.h b/drivers/md/bcache/btree.h
index 3333d3723633..767e75570896 100644
--- a/drivers/md/bcache/btree.h
+++ b/drivers/md/bcache/btree.h
@@ -125,6 +125,7 @@ struct btree {
 	unsigned long		seq;
 	struct rw_semaphore	lock;
 	struct cache_set	*c;
+	struct btree		*parent;
 
 	unsigned long		flags;
 	uint16_t		written;	/* would be nice to kill */
@@ -200,12 +201,7 @@ static inline bool bkey_written(struct btree *b, struct bkey *k)
 
 static inline void set_gc_sectors(struct cache_set *c)
 {
-	atomic_set(&c->sectors_to_gc, c->sb.bucket_size * c->nbuckets / 8);
-}
-
-static inline bool bch_ptr_invalid(struct btree *b, const struct bkey *k)
-{
-	return __bch_ptr_invalid(b->c, b->level, k);
+	atomic_set(&c->sectors_to_gc, c->sb.bucket_size * c->nbuckets / 16);
 }
 
 static inline struct bkey *bch_btree_iter_init(struct btree *b,
@@ -215,6 +211,16 @@ static inline struct bkey *bch_btree_iter_init(struct btree *b,
 	return __bch_btree_iter_init(b, iter, search, b->sets);
 }
 
+static inline bool bch_ptr_invalid(struct btree *b, const struct bkey *k)
+{
+	if (b->level)
+		return bch_btree_ptr_invalid(b->c, k);
+	else
+		return bch_extent_ptr_invalid(b->c, k);
+}
+
+void bkey_put(struct cache_set *c, struct bkey *k);
+
 /* Looping macros */
 
 #define for_each_cached_btree(b, c, iter)				\
@@ -234,51 +240,17 @@ static inline struct bkey *bch_btree_iter_init(struct btree *b,
 /* Recursing down the btree */
 
 struct btree_op {
-	struct closure		cl;
-	struct cache_set	*c;
-
-	/* Journal entry we have a refcount on */
-	atomic_t		*journal;
-
-	/* Bio to be inserted into the cache */
-	struct bio		*cache_bio;
-
-	unsigned		inode;
-
-	uint16_t		write_prio;
-
 	/* Btree level at which we start taking write locks */
 	short			lock;
 
-	/* Btree insertion type */
-	enum {
-		BTREE_INSERT,
-		BTREE_REPLACE
-	} type:8;
-
-	unsigned		csum:1;
-	unsigned		skip:1;
-	unsigned		flush_journal:1;
-
-	unsigned		insert_data_done:1;
-	unsigned		lookup_done:1;
 	unsigned		insert_collision:1;
-
-	/* Anything after this point won't get zeroed in do_bio_hook() */
-
-	/* Keys to be inserted */
-	struct keylist		keys;
-	BKEY_PADDED(replace);
 };
 
-enum {
-	BTREE_INSERT_STATUS_INSERT,
-	BTREE_INSERT_STATUS_BACK_MERGE,
-	BTREE_INSERT_STATUS_OVERWROTE,
-	BTREE_INSERT_STATUS_FRONT_MERGE,
-};
-
-void bch_btree_op_init_stack(struct btree_op *);
+static inline void bch_btree_op_init(struct btree_op *op, int write_lock_level)
+{
+	memset(op, 0, sizeof(struct btree_op));
+	op->lock = write_lock_level;
+}
 
 static inline void rw_lock(bool w, struct btree *b, int level)
 {
@@ -290,108 +262,71 @@ static inline void rw_lock(bool w, struct btree *b, int level)
 
 static inline void rw_unlock(bool w, struct btree *b)
 {
-#ifdef CONFIG_BCACHE_EDEBUG
-	unsigned i;
-
-	if (w && b->key.ptr[0])
-		for (i = 0; i <= b->nsets; i++)
-			bch_check_key_order(b, b->sets[i].data);
-#endif
-
 	if (w)
 		b->seq++;
 	(w ? up_write : up_read)(&b->lock);
 }
 
-#define insert_lock(s, b)	((b)->level <= (s)->lock)
+void bch_btree_node_read(struct btree *);
+void bch_btree_node_write(struct btree *, struct closure *);
 
-/*
- * These macros are for recursing down the btree - they handle the details of
- * locking and looking up nodes in the cache for you. They're best treated as
- * mere syntax when reading code that uses them.
- *
- * op->lock determines whether we take a read or a write lock at a given depth.
- * If you've got a read lock and find that you need a write lock (i.e. you're
- * going to have to split), set op->lock and return -EINTR; btree_root() will
- * call you again and you'll have the correct lock.
- */
+void bch_btree_set_root(struct btree *);
+struct btree *bch_btree_node_alloc(struct cache_set *, int, bool);
+struct btree *bch_btree_node_get(struct cache_set *, struct bkey *, int, bool);
 
-/**
- * btree - recurse down the btree on a specified key
- * @fn:		function to call, which will be passed the child node
- * @key:	key to recurse on
- * @b:		parent btree node
- * @op:		pointer to struct btree_op
- */
-#define btree(fn, key, b, op, ...)					\
-({									\
-	int _r, l = (b)->level - 1;					\
-	bool _w = l <= (op)->lock;					\
-	struct btree *_b = bch_btree_node_get((b)->c, key, l, op);	\
-	if (!IS_ERR(_b)) {						\
-		_r = bch_btree_ ## fn(_b, op, ##__VA_ARGS__);		\
-		rw_unlock(_w, _b);					\
-	} else								\
-		_r = PTR_ERR(_b);					\
-	_r;								\
-})
-
-/**
- * btree_root - call a function on the root of the btree
- * @fn:		function to call, which will be passed the child node
- * @c:		cache set
- * @op:		pointer to struct btree_op
- */
-#define btree_root(fn, c, op, ...)					\
-({									\
-	int _r = -EINTR;						\
-	do {								\
-		struct btree *_b = (c)->root;				\
-		bool _w = insert_lock(op, _b);				\
-		rw_lock(_w, _b, _b->level);				\
-		if (_b == (c)->root &&					\
-		    _w == insert_lock(op, _b))				\
-			_r = bch_btree_ ## fn(_b, op, ##__VA_ARGS__);	\
-		rw_unlock(_w, _b);					\
-		bch_cannibalize_unlock(c, &(op)->cl);		\
-	} while (_r == -EINTR);						\
-									\
-	_r;								\
-})
+int bch_btree_insert_check_key(struct btree *, struct btree_op *,
+			       struct bkey *);
+int bch_btree_insert(struct cache_set *, struct keylist *,
+		     atomic_t *, struct bkey *);
+
+int bch_gc_thread_start(struct cache_set *);
+size_t bch_btree_gc_finish(struct cache_set *);
+void bch_moving_gc(struct cache_set *);
+int bch_btree_check(struct cache_set *);
+uint8_t __bch_btree_mark_key(struct cache_set *, int, struct bkey *);
 
-static inline bool should_split(struct btree *b)
+static inline void wake_up_gc(struct cache_set *c)
 {
-	struct bset *i = write_block(b);
-	return b->written >= btree_blocks(b) ||
-		(i->seq == b->sets[0].data->seq &&
-		 b->written + __set_blocks(i, i->keys + 15, b->c)
-		 > btree_blocks(b));
+	if (c->gc_thread)
+		wake_up_process(c->gc_thread);
 }
 
-void bch_btree_node_read(struct btree *);
-void bch_btree_node_write(struct btree *, struct closure *);
+#define MAP_DONE	0
+#define MAP_CONTINUE	1
 
-void bch_cannibalize_unlock(struct cache_set *, struct closure *);
-void bch_btree_set_root(struct btree *);
-struct btree *bch_btree_node_alloc(struct cache_set *, int, struct closure *);
-struct btree *bch_btree_node_get(struct cache_set *, struct bkey *,
-				int, struct btree_op *);
+#define MAP_ALL_NODES	0
+#define MAP_LEAF_NODES	1
 
-bool bch_btree_insert_check_key(struct btree *, struct btree_op *,
-				   struct bio *);
-int bch_btree_insert(struct btree_op *, struct cache_set *);
+#define MAP_END_KEY	1
 
-int bch_btree_search_recurse(struct btree *, struct btree_op *);
+typedef int (btree_map_nodes_fn)(struct btree_op *, struct btree *);
+int __bch_btree_map_nodes(struct btree_op *, struct cache_set *,
+			  struct bkey *, btree_map_nodes_fn *, int);
 
-void bch_queue_gc(struct cache_set *);
-size_t bch_btree_gc_finish(struct cache_set *);
-void bch_moving_gc(struct closure *);
-int bch_btree_check(struct cache_set *, struct btree_op *);
-uint8_t __bch_btree_mark_key(struct cache_set *, int, struct bkey *);
+static inline int bch_btree_map_nodes(struct btree_op *op, struct cache_set *c,
+				      struct bkey *from, btree_map_nodes_fn *fn)
+{
+	return __bch_btree_map_nodes(op, c, from, fn, MAP_ALL_NODES);
+}
+
+static inline int bch_btree_map_leaf_nodes(struct btree_op *op,
+					   struct cache_set *c,
+					   struct bkey *from,
+					   btree_map_nodes_fn *fn)
+{
+	return __bch_btree_map_nodes(op, c, from, fn, MAP_LEAF_NODES);
+}
+
+typedef int (btree_map_keys_fn)(struct btree_op *, struct btree *,
+				struct bkey *);
+int bch_btree_map_keys(struct btree_op *, struct cache_set *,
+		       struct bkey *, btree_map_keys_fn *, int);
+
+typedef bool (keybuf_pred_fn)(struct keybuf *, struct bkey *);
 
 void bch_keybuf_init(struct keybuf *);
-void bch_refill_keybuf(struct cache_set *, struct keybuf *, struct bkey *,
-		       keybuf_pred_fn *);
+void bch_refill_keybuf(struct cache_set *, struct keybuf *,
+		       struct bkey *, keybuf_pred_fn *);
 bool bch_keybuf_check_overlapping(struct keybuf *, struct bkey *,
 				  struct bkey *);
 void bch_keybuf_del(struct keybuf *, struct keybuf_key *);
diff --git a/drivers/md/bcache/closure.c b/drivers/md/bcache/closure.c
index 9aba2017f0d1..dfff2410322e 100644
--- a/drivers/md/bcache/closure.c
+++ b/drivers/md/bcache/closure.c
@@ -11,17 +11,6 @@
 
 #include "closure.h"
 
-void closure_queue(struct closure *cl)
-{
-	struct workqueue_struct *wq = cl->wq;
-	if (wq) {
-		INIT_WORK(&cl->work, cl->work.func);
-		BUG_ON(!queue_work(wq, &cl->work));
-	} else
-		cl->fn(cl);
-}
-EXPORT_SYMBOL_GPL(closure_queue);
-
 #define CL_FIELD(type, field)					\
 	case TYPE_ ## type:					\
 	return &container_of(cl, struct type, cl)->field
@@ -30,17 +19,6 @@ static struct closure_waitlist *closure_waitlist(struct closure *cl)
 {
 	switch (cl->type) {
 		CL_FIELD(closure_with_waitlist, wait);
-		CL_FIELD(closure_with_waitlist_and_timer, wait);
-	default:
-		return NULL;
-	}
-}
-
-static struct timer_list *closure_timer(struct closure *cl)
-{
-	switch (cl->type) {
-		CL_FIELD(closure_with_timer, timer);
-		CL_FIELD(closure_with_waitlist_and_timer, timer);
 	default:
 		return NULL;
 	}
@@ -51,7 +29,7 @@ static inline void closure_put_after_sub(struct closure *cl, int flags)
 	int r = flags & CLOSURE_REMAINING_MASK;
 
 	BUG_ON(flags & CLOSURE_GUARD_MASK);
-	BUG_ON(!r && (flags & ~(CLOSURE_DESTRUCTOR|CLOSURE_BLOCKING)));
+	BUG_ON(!r && (flags & ~CLOSURE_DESTRUCTOR));
 
 	/* Must deliver precisely one wakeup */
 	if (r == 1 && (flags & CLOSURE_SLEEPING))
@@ -59,7 +37,6 @@ static inline void closure_put_after_sub(struct closure *cl, int flags)
 
 	if (!r) {
 		if (cl->fn && !(flags & CLOSURE_DESTRUCTOR)) {
-			/* CLOSURE_BLOCKING might be set - clear it */
 			atomic_set(&cl->remaining,
 				   CLOSURE_REMAINING_INITIALIZER);
 			closure_queue(cl);
@@ -90,13 +67,13 @@ void closure_sub(struct closure *cl, int v)
 {
 	closure_put_after_sub(cl, atomic_sub_return(v, &cl->remaining));
 }
-EXPORT_SYMBOL_GPL(closure_sub);
+EXPORT_SYMBOL(closure_sub);
 
 void closure_put(struct closure *cl)
 {
 	closure_put_after_sub(cl, atomic_dec_return(&cl->remaining));
 }
-EXPORT_SYMBOL_GPL(closure_put);
+EXPORT_SYMBOL(closure_put);
 
 static void set_waiting(struct closure *cl, unsigned long f)
 {
@@ -133,7 +110,7 @@ void __closure_wake_up(struct closure_waitlist *wait_list)
 		closure_sub(cl, CLOSURE_WAITING + 1);
 	}
 }
-EXPORT_SYMBOL_GPL(__closure_wake_up);
+EXPORT_SYMBOL(__closure_wake_up);
 
 bool closure_wait(struct closure_waitlist *list, struct closure *cl)
 {
@@ -146,7 +123,7 @@ bool closure_wait(struct closure_waitlist *list, struct closure *cl)
 
 	return true;
 }
-EXPORT_SYMBOL_GPL(closure_wait);
+EXPORT_SYMBOL(closure_wait);
 
 /**
  * closure_sync() - sleep until a closure a closure has nothing left to wait on
@@ -169,7 +146,7 @@ void closure_sync(struct closure *cl)
 
 	__closure_end_sleep(cl);
 }
-EXPORT_SYMBOL_GPL(closure_sync);
+EXPORT_SYMBOL(closure_sync);
 
 /**
  * closure_trylock() - try to acquire the closure, without waiting
@@ -183,17 +160,17 @@ bool closure_trylock(struct closure *cl, struct closure *parent)
 			   CLOSURE_REMAINING_INITIALIZER) != -1)
 		return false;
 
-	closure_set_ret_ip(cl);
-
 	smp_mb();
+
 	cl->parent = parent;
 	if (parent)
 		closure_get(parent);
 
+	closure_set_ret_ip(cl);
 	closure_debug_create(cl);
 	return true;
 }
-EXPORT_SYMBOL_GPL(closure_trylock);
+EXPORT_SYMBOL(closure_trylock);
 
 void __closure_lock(struct closure *cl, struct closure *parent,
 		    struct closure_waitlist *wait_list)
@@ -205,57 +182,11 @@ void __closure_lock(struct closure *cl, struct closure *parent,
 		if (closure_trylock(cl, parent))
 			return;
 
-		closure_wait_event_sync(wait_list, &wait,
-					atomic_read(&cl->remaining) == -1);
+		closure_wait_event(wait_list, &wait,
+				   atomic_read(&cl->remaining) == -1);
 	}
 }
-EXPORT_SYMBOL_GPL(__closure_lock);
-
-static void closure_delay_timer_fn(unsigned long data)
-{
-	struct closure *cl = (struct closure *) data;
-	closure_sub(cl, CLOSURE_TIMER + 1);
-}
-
-void do_closure_timer_init(struct closure *cl)
-{
-	struct timer_list *timer = closure_timer(cl);
-
-	init_timer(timer);
-	timer->data	= (unsigned long) cl;
-	timer->function = closure_delay_timer_fn;
-}
-EXPORT_SYMBOL_GPL(do_closure_timer_init);
-
-bool __closure_delay(struct closure *cl, unsigned long delay,
-		     struct timer_list *timer)
-{
-	if (atomic_read(&cl->remaining) & CLOSURE_TIMER)
-		return false;
-
-	BUG_ON(timer_pending(timer));
-
-	timer->expires	= jiffies + delay;
-
-	atomic_add(CLOSURE_TIMER + 1, &cl->remaining);
-	add_timer(timer);
-	return true;
-}
-EXPORT_SYMBOL_GPL(__closure_delay);
-
-void __closure_flush(struct closure *cl, struct timer_list *timer)
-{
-	if (del_timer(timer))
-		closure_sub(cl, CLOSURE_TIMER + 1);
-}
-EXPORT_SYMBOL_GPL(__closure_flush);
-
-void __closure_flush_sync(struct closure *cl, struct timer_list *timer)
-{
-	if (del_timer_sync(timer))
-		closure_sub(cl, CLOSURE_TIMER + 1);
-}
-EXPORT_SYMBOL_GPL(__closure_flush_sync);
+EXPORT_SYMBOL(__closure_lock);
 
 #ifdef CONFIG_BCACHE_CLOSURES_DEBUG
 
@@ -273,7 +204,7 @@ void closure_debug_create(struct closure *cl)
 	list_add(&cl->all, &closure_list);
 	spin_unlock_irqrestore(&closure_list_lock, flags);
 }
-EXPORT_SYMBOL_GPL(closure_debug_create);
+EXPORT_SYMBOL(closure_debug_create);
 
 void closure_debug_destroy(struct closure *cl)
 {
@@ -286,7 +217,7 @@ void closure_debug_destroy(struct closure *cl)
 	list_del(&cl->all);
 	spin_unlock_irqrestore(&closure_list_lock, flags);
 }
-EXPORT_SYMBOL_GPL(closure_debug_destroy);
+EXPORT_SYMBOL(closure_debug_destroy);
 
 static struct dentry *debug;
 
@@ -304,14 +235,12 @@ static int debug_seq_show(struct seq_file *f, void *data)
 			   cl, (void *) cl->ip, cl->fn, cl->parent,
 			   r & CLOSURE_REMAINING_MASK);
 
-		seq_printf(f, "%s%s%s%s%s%s\n",
+		seq_printf(f, "%s%s%s%s\n",
 			   test_bit(WORK_STRUCT_PENDING,
 				    work_data_bits(&cl->work)) ? "Q" : "",
 			   r & CLOSURE_RUNNING	? "R" : "",
-			   r & CLOSURE_BLOCKING	? "B" : "",
 			   r & CLOSURE_STACK	? "S" : "",
-			   r & CLOSURE_SLEEPING	? "Sl" : "",
-			   r & CLOSURE_TIMER	? "T" : "");
+			   r & CLOSURE_SLEEPING	? "Sl" : "");
 
 		if (r & CLOSURE_WAITING)
 			seq_printf(f, " W %pF\n",
diff --git a/drivers/md/bcache/closure.h b/drivers/md/bcache/closure.h
index 00039924ea9d..9762f1be3304 100644
--- a/drivers/md/bcache/closure.h
+++ b/drivers/md/bcache/closure.h
@@ -155,21 +155,6 @@
  * delayed_work embeds a work item and a timer_list. The important thing is, use
  * it exactly like you would a regular closure and closure_put() will magically
  * handle everything for you.
- *
- * We've got closures that embed timers, too. They're called, appropriately
- * enough:
- * struct closure_with_timer;
- *
- * This gives you access to closure_delay(). It takes a refcount for a specified
- * number of jiffies - you could then call closure_sync() (for a slightly
- * convoluted version of msleep()) or continue_at() - which gives you the same
- * effect as using a delayed work item, except you can reuse the work_struct
- * already embedded in struct closure.
- *
- * Lastly, there's struct closure_with_waitlist_and_timer. It does what you
- * probably expect, if you happen to need the features of both. (You don't
- * really want to know how all this is implemented, but if I've done my job
- * right you shouldn't have to care).
  */
 
 struct closure;
@@ -182,16 +167,11 @@ struct closure_waitlist {
 enum closure_type {
 	TYPE_closure				= 0,
 	TYPE_closure_with_waitlist		= 1,
-	TYPE_closure_with_timer			= 2,
-	TYPE_closure_with_waitlist_and_timer	= 3,
-	MAX_CLOSURE_TYPE			= 3,
+	MAX_CLOSURE_TYPE			= 1,
 };
 
 enum closure_state {
 	/*
-	 * CLOSURE_BLOCKING: Causes closure_wait_event() to block, instead of
-	 * waiting asynchronously
-	 *
 	 * CLOSURE_WAITING: Set iff the closure is on a waitlist. Must be set by
 	 * the thread that owns the closure, and cleared by the thread that's
 	 * waking up the closure.
@@ -200,10 +180,6 @@ enum closure_state {
 	 * - indicates that cl->task is valid and closure_put() may wake it up.
 	 * Only set or cleared by the thread that owns the closure.
 	 *
-	 * CLOSURE_TIMER: Analagous to CLOSURE_WAITING, indicates that a closure
-	 * has an outstanding timer. Must be set by the thread that owns the
-	 * closure, and cleared by the timer function when the timer goes off.
-	 *
 	 * The rest are for debugging and don't affect behaviour:
 	 *
 	 * CLOSURE_RUNNING: Set when a closure is running (i.e. by
@@ -218,19 +194,17 @@ enum closure_state {
 	 * closure with this flag set
 	 */
 
-	CLOSURE_BITS_START	= (1 << 19),
-	CLOSURE_DESTRUCTOR	= (1 << 19),
-	CLOSURE_BLOCKING	= (1 << 21),
-	CLOSURE_WAITING		= (1 << 23),
-	CLOSURE_SLEEPING	= (1 << 25),
-	CLOSURE_TIMER		= (1 << 27),
+	CLOSURE_BITS_START	= (1 << 23),
+	CLOSURE_DESTRUCTOR	= (1 << 23),
+	CLOSURE_WAITING		= (1 << 25),
+	CLOSURE_SLEEPING	= (1 << 27),
 	CLOSURE_RUNNING		= (1 << 29),
 	CLOSURE_STACK		= (1 << 31),
 };
 
 #define CLOSURE_GUARD_MASK					\
-	((CLOSURE_DESTRUCTOR|CLOSURE_BLOCKING|CLOSURE_WAITING|	\
-	  CLOSURE_SLEEPING|CLOSURE_TIMER|CLOSURE_RUNNING|CLOSURE_STACK) << 1)
+	((CLOSURE_DESTRUCTOR|CLOSURE_WAITING|CLOSURE_SLEEPING|	\
+	  CLOSURE_RUNNING|CLOSURE_STACK) << 1)
 
 #define CLOSURE_REMAINING_MASK		(CLOSURE_BITS_START - 1)
 #define CLOSURE_REMAINING_INITIALIZER	(1|CLOSURE_RUNNING)
@@ -268,17 +242,6 @@ struct closure_with_waitlist {
 	struct closure_waitlist	wait;
 };
 
-struct closure_with_timer {
-	struct closure		cl;
-	struct timer_list	timer;
-};
-
-struct closure_with_waitlist_and_timer {
-	struct closure		cl;
-	struct closure_waitlist	wait;
-	struct timer_list	timer;
-};
-
 extern unsigned invalid_closure_type(void);
 
 #define __CLOSURE_TYPE(cl, _t)						\
@@ -289,14 +252,11 @@ extern unsigned invalid_closure_type(void);
 (									\
 	__CLOSURE_TYPE(cl, closure)					\
 	__CLOSURE_TYPE(cl, closure_with_waitlist)			\
-	__CLOSURE_TYPE(cl, closure_with_timer)				\
-	__CLOSURE_TYPE(cl, closure_with_waitlist_and_timer)		\
 	invalid_closure_type()						\
 )
 
 void closure_sub(struct closure *cl, int v);
 void closure_put(struct closure *cl);
-void closure_queue(struct closure *cl);
 void __closure_wake_up(struct closure_waitlist *list);
 bool closure_wait(struct closure_waitlist *list, struct closure *cl);
 void closure_sync(struct closure *cl);
@@ -305,12 +265,6 @@ bool closure_trylock(struct closure *cl, struct closure *parent);
 void __closure_lock(struct closure *cl, struct closure *parent,
 		    struct closure_waitlist *wait_list);
 
-void do_closure_timer_init(struct closure *cl);
-bool __closure_delay(struct closure *cl, unsigned long delay,
-		     struct timer_list *timer);
-void __closure_flush(struct closure *cl, struct timer_list *timer);
-void __closure_flush_sync(struct closure *cl, struct timer_list *timer);
-
 #ifdef CONFIG_BCACHE_CLOSURES_DEBUG
 
 void closure_debug_init(void);
@@ -354,11 +308,6 @@ static inline void closure_set_stopped(struct closure *cl)
 	atomic_sub(CLOSURE_RUNNING, &cl->remaining);
 }
 
-static inline bool closure_is_stopped(struct closure *cl)
-{
-	return !(atomic_read(&cl->remaining) & CLOSURE_RUNNING);
-}
-
 static inline bool closure_is_unlocked(struct closure *cl)
 {
 	return atomic_read(&cl->remaining) == -1;
@@ -367,14 +316,6 @@ static inline bool closure_is_unlocked(struct closure *cl)
 static inline void do_closure_init(struct closure *cl, struct closure *parent,
 				   bool running)
 {
-	switch (cl->type) {
-	case TYPE_closure_with_timer:
-	case TYPE_closure_with_waitlist_and_timer:
-		do_closure_timer_init(cl);
-	default:
-		break;
-	}
-
 	cl->parent = parent;
 	if (parent)
 		closure_get(parent);
@@ -429,8 +370,7 @@ do {								\
 static inline void closure_init_stack(struct closure *cl)
 {
 	memset(cl, 0, sizeof(struct closure));
-	atomic_set(&cl->remaining, CLOSURE_REMAINING_INITIALIZER|
-		   CLOSURE_BLOCKING|CLOSURE_STACK);
+	atomic_set(&cl->remaining, CLOSURE_REMAINING_INITIALIZER|CLOSURE_STACK);
 }
 
 /**
@@ -461,24 +401,6 @@ do {								\
 #define closure_lock(cl, parent)				\
 	__closure_lock(__to_internal_closure(cl), parent, &(cl)->wait)
 
-/**
- * closure_delay() - delay some number of jiffies
- * @cl:		the closure that will sleep
- * @delay:	the delay in jiffies
- *
- * Takes a refcount on @cl which will be released after @delay jiffies; this may
- * be used to have a function run after a delay with continue_at(), or
- * closure_sync() may be used for a convoluted version of msleep().
- */
-#define closure_delay(cl, delay)			\
-	__closure_delay(__to_internal_closure(cl), delay, &(cl)->timer)
-
-#define closure_flush(cl)				\
-	__closure_flush(__to_internal_closure(cl), &(cl)->timer)
-
-#define closure_flush_sync(cl)				\
-	__closure_flush_sync(__to_internal_closure(cl), &(cl)->timer)
-
 static inline void __closure_end_sleep(struct closure *cl)
 {
 	__set_current_state(TASK_RUNNING);
@@ -498,40 +420,6 @@ static inline void __closure_start_sleep(struct closure *cl)
 }
 
 /**
- * closure_blocking() - returns true if the closure is in blocking mode.
- *
- * If a closure is in blocking mode, closure_wait_event() will sleep until the
- * condition is true instead of waiting asynchronously.
- */
-static inline bool closure_blocking(struct closure *cl)
-{
-	return atomic_read(&cl->remaining) & CLOSURE_BLOCKING;
-}
-
-/**
- * set_closure_blocking() - put a closure in blocking mode.
- *
- * If a closure is in blocking mode, closure_wait_event() will sleep until the
- * condition is true instead of waiting asynchronously.
- *
- * Not thread safe - can only be called by the thread running the closure.
- */
-static inline void set_closure_blocking(struct closure *cl)
-{
-	if (!closure_blocking(cl))
-		atomic_add(CLOSURE_BLOCKING, &cl->remaining);
-}
-
-/*
- * Not thread safe - can only be called by the thread running the closure.
- */
-static inline void clear_closure_blocking(struct closure *cl)
-{
-	if (closure_blocking(cl))
-		atomic_sub(CLOSURE_BLOCKING, &cl->remaining);
-}
-
-/**
  * closure_wake_up() - wake up all closures on a wait list.
  */
 static inline void closure_wake_up(struct closure_waitlist *list)
@@ -561,63 +449,36 @@ static inline void closure_wake_up(struct closure_waitlist *list)
  * refcount on our closure. If this was a stack allocated closure, that would be
  * bad.
  */
-#define __closure_wait_event(list, cl, condition, _block)		\
+#define closure_wait_event(list, cl, condition)				\
 ({									\
-	bool block = _block;						\
 	typeof(condition) ret;						\
 									\
 	while (1) {							\
 		ret = (condition);					\
 		if (ret) {						\
 			__closure_wake_up(list);			\
-			if (block)					\
-				closure_sync(cl);			\
-									\
+			closure_sync(cl);				\
 			break;						\
 		}							\
 									\
-		if (block)						\
-			__closure_start_sleep(cl);			\
-									\
-		if (!closure_wait(list, cl)) {				\
-			if (!block)					\
-				break;					\
+		__closure_start_sleep(cl);				\
 									\
+		if (!closure_wait(list, cl))				\
 			schedule();					\
-		}							\
 	}								\
 									\
 	ret;								\
 })
 
-/**
- * closure_wait_event() - wait on a condition, synchronously or asynchronously.
- * @list:	the wait list to wait on
- * @cl:		the closure that is doing the waiting
- * @condition:	a C expression for the event to wait for
- *
- * If the closure is in blocking mode, sleeps until the @condition evaluates to
- * true - exactly like wait_event().
- *
- * If the closure is not in blocking mode, waits asynchronously; if the
- * condition is currently false the @cl is put onto @list and returns. @list
- * owns a refcount on @cl; closure_sync() or continue_at() may be used later to
- * wait for another thread to wake up @list, which drops the refcount on @cl.
- *
- * Returns the value of @condition; @cl will be on @list iff @condition was
- * false.
- *
- * closure_wake_up(@list) must be called after changing any variable that could
- * cause @condition to become true.
- */
-#define closure_wait_event(list, cl, condition)				\
-	__closure_wait_event(list, cl, condition, closure_blocking(cl))
-
-#define closure_wait_event_async(list, cl, condition)			\
-	__closure_wait_event(list, cl, condition, false)
-
-#define closure_wait_event_sync(list, cl, condition)			\
-	__closure_wait_event(list, cl, condition, true)
+static inline void closure_queue(struct closure *cl)
+{
+	struct workqueue_struct *wq = cl->wq;
+	if (wq) {
+		INIT_WORK(&cl->work, cl->work.func);
+		BUG_ON(!queue_work(wq, &cl->work));
+	} else
+		cl->fn(cl);
+}
 
 static inline void set_closure_fn(struct closure *cl, closure_fn *fn,
 				  struct workqueue_struct *wq)
@@ -642,7 +503,7 @@ do {									\
 #define continue_at_nobarrier(_cl, _fn, _wq)				\
 do {									\
 	set_closure_fn(_cl, _fn, _wq);					\
-	closure_queue(cl);						\
+	closure_queue(_cl);						\
 	return;								\
 } while (0)
 
diff --git a/drivers/md/bcache/debug.c b/drivers/md/bcache/debug.c
index 88e6411eab4f..264fcfbd6290 100644
--- a/drivers/md/bcache/debug.c
+++ b/drivers/md/bcache/debug.c
@@ -8,7 +8,6 @@
 #include "bcache.h"
 #include "btree.h"
 #include "debug.h"
-#include "request.h"
 
 #include <linux/console.h>
 #include <linux/debugfs.h>
@@ -77,29 +76,17 @@ int bch_bkey_to_text(char *buf, size_t size, const struct bkey *k)
 	return out - buf;
 }
 
-int bch_btree_to_text(char *buf, size_t size, const struct btree *b)
-{
-	return scnprintf(buf, size, "%zu level %i/%i",
-			 PTR_BUCKET_NR(b->c, &b->key, 0),
-			 b->level, b->c->root ? b->c->root->level : -1);
-}
-
-#if defined(CONFIG_BCACHE_DEBUG) || defined(CONFIG_BCACHE_EDEBUG)
-
-static bool skipped_backwards(struct btree *b, struct bkey *k)
-{
-	return bkey_cmp(k, (!b->level)
-			? &START_KEY(bkey_next(k))
-			: bkey_next(k)) > 0;
-}
+#ifdef CONFIG_BCACHE_DEBUG
 
 static void dump_bset(struct btree *b, struct bset *i)
 {
-	struct bkey *k;
+	struct bkey *k, *next;
 	unsigned j;
 	char buf[80];
 
-	for (k = i->start; k < end(i); k = bkey_next(k)) {
+	for (k = i->start; k < end(i); k = next) {
+		next = bkey_next(k);
+
 		bch_bkey_to_text(buf, sizeof(buf), k);
 		printk(KERN_ERR "block %zu key %zi/%u: %s", index(i, b),
 		       (uint64_t *) k - i->d, i->keys, buf);
@@ -115,15 +102,21 @@ static void dump_bset(struct btree *b, struct bset *i)
 
 		printk(" %s\n", bch_ptr_status(b->c, k));
 
-		if (bkey_next(k) < end(i) &&
-		    skipped_backwards(b, k))
+		if (next < end(i) &&
+		    bkey_cmp(k, !b->level ? &START_KEY(next) : next) > 0)
 			printk(KERN_ERR "Key skipped backwards\n");
 	}
 }
 
-#endif
+static void bch_dump_bucket(struct btree *b)
+{
+	unsigned i;
 
-#ifdef CONFIG_BCACHE_DEBUG
+	console_lock();
+	for (i = 0; i <= b->nsets; i++)
+		dump_bset(b, b->sets[i].data);
+	console_unlock();
+}
 
 void bch_btree_verify(struct btree *b, struct bset *new)
 {
@@ -176,66 +169,44 @@ void bch_btree_verify(struct btree *b, struct bset *new)
 	mutex_unlock(&b->c->verify_lock);
 }
 
-static void data_verify_endio(struct bio *bio, int error)
-{
-	struct closure *cl = bio->bi_private;
-	closure_put(cl);
-}
-
-void bch_data_verify(struct search *s)
+void bch_data_verify(struct cached_dev *dc, struct bio *bio)
 {
 	char name[BDEVNAME_SIZE];
-	struct cached_dev *dc = container_of(s->d, struct cached_dev, disk);
-	struct closure *cl = &s->cl;
 	struct bio *check;
 	struct bio_vec *bv;
 	int i;
 
-	if (!s->unaligned_bvec)
-		bio_for_each_segment(bv, s->orig_bio, i)
-			bv->bv_offset = 0, bv->bv_len = PAGE_SIZE;
-
-	check = bio_clone(s->orig_bio, GFP_NOIO);
+	check = bio_clone(bio, GFP_NOIO);
 	if (!check)
 		return;
 
 	if (bio_alloc_pages(check, GFP_NOIO))
 		goto out_put;
 
-	check->bi_rw		= READ_SYNC;
-	check->bi_private	= cl;
-	check->bi_end_io	= data_verify_endio;
-
-	closure_bio_submit(check, cl, &dc->disk);
-	closure_sync(cl);
+	submit_bio_wait(READ_SYNC, check);
 
-	bio_for_each_segment(bv, s->orig_bio, i) {
-		void *p1 = kmap(bv->bv_page);
-		void *p2 = kmap(check->bi_io_vec[i].bv_page);
+	bio_for_each_segment(bv, bio, i) {
+		void *p1 = kmap_atomic(bv->bv_page);
+		void *p2 = page_address(check->bi_io_vec[i].bv_page);
 
-		if (memcmp(p1 + bv->bv_offset,
-			   p2 + bv->bv_offset,
-			   bv->bv_len))
-			printk(KERN_ERR
-			       "bcache (%s): verify failed at sector %llu\n",
-			       bdevname(dc->bdev, name),
-			       (uint64_t) s->orig_bio->bi_sector);
+		cache_set_err_on(memcmp(p1 + bv->bv_offset,
+					p2 + bv->bv_offset,
+					bv->bv_len),
+				 dc->disk.c,
+				 "verify failed at dev %s sector %llu",
+				 bdevname(dc->bdev, name),
+				 (uint64_t) bio->bi_sector);
 
-		kunmap(bv->bv_page);
-		kunmap(check->bi_io_vec[i].bv_page);
+		kunmap_atomic(p1);
 	}
 
-	__bio_for_each_segment(bv, check, i, 0)
+	bio_for_each_segment_all(bv, check, i)
 		__free_page(bv->bv_page);
 out_put:
 	bio_put(check);
 }
 
-#endif
-
-#ifdef CONFIG_BCACHE_EDEBUG
-
-unsigned bch_count_data(struct btree *b)
+int __bch_count_data(struct btree *b)
 {
 	unsigned ret = 0;
 	struct btree_iter iter;
@@ -247,72 +218,60 @@ unsigned bch_count_data(struct btree *b)
 	return ret;
 }
 
-static void vdump_bucket_and_panic(struct btree *b, const char *fmt,
-				   va_list args)
-{
-	unsigned i;
-	char buf[80];
-
-	console_lock();
-
-	for (i = 0; i <= b->nsets; i++)
-		dump_bset(b, b->sets[i].data);
-
-	vprintk(fmt, args);
-
-	console_unlock();
-
-	bch_btree_to_text(buf, sizeof(buf), b);
-	panic("at %s\n", buf);
-}
-
-void bch_check_key_order_msg(struct btree *b, struct bset *i,
-			     const char *fmt, ...)
-{
-	struct bkey *k;
-
-	if (!i->keys)
-		return;
-
-	for (k = i->start; bkey_next(k) < end(i); k = bkey_next(k))
-		if (skipped_backwards(b, k)) {
-			va_list args;
-			va_start(args, fmt);
-
-			vdump_bucket_and_panic(b, fmt, args);
-			va_end(args);
-		}
-}
-
-void bch_check_keys(struct btree *b, const char *fmt, ...)
+void __bch_check_keys(struct btree *b, const char *fmt, ...)
 {
 	va_list args;
 	struct bkey *k, *p = NULL;
 	struct btree_iter iter;
-
-	if (b->level)
-		return;
+	const char *err;
 
 	for_each_key(b, k, &iter) {
-		if (p && bkey_cmp(&START_KEY(p), &START_KEY(k)) > 0) {
-			printk(KERN_ERR "Keys out of order:\n");
-			goto bug;
-		}
-
-		if (bch_ptr_invalid(b, k))
-			continue;
-
-		if (p && bkey_cmp(p, &START_KEY(k)) > 0) {
-			printk(KERN_ERR "Overlapping keys:\n");
-			goto bug;
+		if (!b->level) {
+			err = "Keys out of order";
+			if (p && bkey_cmp(&START_KEY(p), &START_KEY(k)) > 0)
+				goto bug;
+
+			if (bch_ptr_invalid(b, k))
+				continue;
+
+			err =  "Overlapping keys";
+			if (p && bkey_cmp(p, &START_KEY(k)) > 0)
+				goto bug;
+		} else {
+			if (bch_ptr_bad(b, k))
+				continue;
+
+			err = "Duplicate keys";
+			if (p && !bkey_cmp(p, k))
+				goto bug;
 		}
 		p = k;
 	}
+
+	err = "Key larger than btree node key";
+	if (p && bkey_cmp(p, &b->key) > 0)
+		goto bug;
+
 	return;
 bug:
+	bch_dump_bucket(b);
+
 	va_start(args, fmt);
-	vdump_bucket_and_panic(b, fmt, args);
+	vprintk(fmt, args);
 	va_end(args);
+
+	panic("bcache error: %s:\n", err);
+}
+
+void bch_btree_iter_next_check(struct btree_iter *iter)
+{
+	struct bkey *k = iter->data->k, *next = bkey_next(k);
+
+	if (next < iter->data->end &&
+	    bkey_cmp(k, iter->b->level ? next : &START_KEY(next)) > 0) {
+		bch_dump_bucket(iter->b);
+		panic("Key skipped backwards\n");
+	}
 }
 
 #endif
diff --git a/drivers/md/bcache/debug.h b/drivers/md/bcache/debug.h
index 1c39b5a2489b..2ede60e31874 100644
--- a/drivers/md/bcache/debug.h
+++ b/drivers/md/bcache/debug.h
@@ -4,40 +4,44 @@
 /* Btree/bkey debug printing */
 
 int bch_bkey_to_text(char *buf, size_t size, const struct bkey *k);
-int bch_btree_to_text(char *buf, size_t size, const struct btree *b);
-
-#ifdef CONFIG_BCACHE_EDEBUG
-
-unsigned bch_count_data(struct btree *);
-void bch_check_key_order_msg(struct btree *, struct bset *, const char *, ...);
-void bch_check_keys(struct btree *, const char *, ...);
-
-#define bch_check_key_order(b, i)			\
-	bch_check_key_order_msg(b, i, "keys out of order")
-#define EBUG_ON(cond)		BUG_ON(cond)
-
-#else /* EDEBUG */
-
-#define bch_count_data(b)				0
-#define bch_check_key_order(b, i)			do {} while (0)
-#define bch_check_key_order_msg(b, i, ...)		do {} while (0)
-#define bch_check_keys(b, ...)				do {} while (0)
-#define EBUG_ON(cond)					do {} while (0)
-
-#endif
 
 #ifdef CONFIG_BCACHE_DEBUG
 
 void bch_btree_verify(struct btree *, struct bset *);
-void bch_data_verify(struct search *);
+void bch_data_verify(struct cached_dev *, struct bio *);
+int __bch_count_data(struct btree *);
+void __bch_check_keys(struct btree *, const char *, ...);
+void bch_btree_iter_next_check(struct btree_iter *);
+
+#define EBUG_ON(cond)			BUG_ON(cond)
+#define expensive_debug_checks(c)	((c)->expensive_debug_checks)
+#define key_merging_disabled(c)		((c)->key_merging_disabled)
+#define bypass_torture_test(d)		((d)->bypass_torture_test)
 
 #else /* DEBUG */
 
 static inline void bch_btree_verify(struct btree *b, struct bset *i) {}
-static inline void bch_data_verify(struct search *s) {};
+static inline void bch_data_verify(struct cached_dev *dc, struct bio *bio) {}
+static inline int __bch_count_data(struct btree *b) { return -1; }
+static inline void __bch_check_keys(struct btree *b, const char *fmt, ...) {}
+static inline void bch_btree_iter_next_check(struct btree_iter *iter) {}
+
+#define EBUG_ON(cond)			do { if (cond); } while (0)
+#define expensive_debug_checks(c)	0
+#define key_merging_disabled(c)		0
+#define bypass_torture_test(d)		0
 
 #endif
 
+#define bch_count_data(b)						\
+	(expensive_debug_checks((b)->c) ? __bch_count_data(b) : -1)
+
+#define bch_check_keys(b, ...)						\
+do {									\
+	if (expensive_debug_checks((b)->c))				\
+		__bch_check_keys(b, __VA_ARGS__);			\
+} while (0)
+
 #ifdef CONFIG_DEBUG_FS
 void bch_debug_init_cache_set(struct cache_set *);
 #else
diff --git a/drivers/md/bcache/journal.c b/drivers/md/bcache/journal.c
index 8435f81e5d85..ecdaa671bd50 100644
--- a/drivers/md/bcache/journal.c
+++ b/drivers/md/bcache/journal.c
@@ -7,7 +7,6 @@
 #include "bcache.h"
 #include "btree.h"
 #include "debug.h"
-#include "request.h"
 
 #include <trace/events/bcache.h>
 
@@ -31,17 +30,20 @@ static void journal_read_endio(struct bio *bio, int error)
 }
 
 static int journal_read_bucket(struct cache *ca, struct list_head *list,
-			       struct btree_op *op, unsigned bucket_index)
+			       unsigned bucket_index)
 {
 	struct journal_device *ja = &ca->journal;
 	struct bio *bio = &ja->bio;
 
 	struct journal_replay *i;
 	struct jset *j, *data = ca->set->journal.w[0].data;
+	struct closure cl;
 	unsigned len, left, offset = 0;
 	int ret = 0;
 	sector_t bucket = bucket_to_sector(ca->set, ca->sb.d[bucket_index]);
 
+	closure_init_stack(&cl);
+
 	pr_debug("reading %llu", (uint64_t) bucket);
 
 	while (offset < ca->sb.bucket_size) {
@@ -55,11 +57,11 @@ reread:		left = ca->sb.bucket_size - offset;
 		bio->bi_size	= len << 9;
 
 		bio->bi_end_io	= journal_read_endio;
-		bio->bi_private = &op->cl;
+		bio->bi_private = &cl;
 		bch_bio_map(bio, data);
 
-		closure_bio_submit(bio, &op->cl, ca);
-		closure_sync(&op->cl);
+		closure_bio_submit(bio, &cl, ca);
+		closure_sync(&cl);
 
 		/* This function could be simpler now since we no longer write
 		 * journal entries that overlap bucket boundaries; this means
@@ -72,7 +74,7 @@ reread:		left = ca->sb.bucket_size - offset;
 			struct list_head *where;
 			size_t blocks, bytes = set_bytes(j);
 
-			if (j->magic != jset_magic(ca->set))
+			if (j->magic != jset_magic(&ca->sb))
 				return ret;
 
 			if (bytes > left << 9)
@@ -129,12 +131,11 @@ next_set:
 	return ret;
 }
 
-int bch_journal_read(struct cache_set *c, struct list_head *list,
-			struct btree_op *op)
+int bch_journal_read(struct cache_set *c, struct list_head *list)
 {
 #define read_bucket(b)							\
 	({								\
-		int ret = journal_read_bucket(ca, list, op, b);		\
+		int ret = journal_read_bucket(ca, list, b);		\
 		__set_bit(b, bitmap);					\
 		if (ret < 0)						\
 			return ret;					\
@@ -292,8 +293,7 @@ void bch_journal_mark(struct cache_set *c, struct list_head *list)
 	}
 }
 
-int bch_journal_replay(struct cache_set *s, struct list_head *list,
-			  struct btree_op *op)
+int bch_journal_replay(struct cache_set *s, struct list_head *list)
 {
 	int ret = 0, keys = 0, entries = 0;
 	struct bkey *k;
@@ -301,31 +301,30 @@ int bch_journal_replay(struct cache_set *s, struct list_head *list,
 		list_entry(list->prev, struct journal_replay, list);
 
 	uint64_t start = i->j.last_seq, end = i->j.seq, n = start;
+	struct keylist keylist;
+
+	bch_keylist_init(&keylist);
 
 	list_for_each_entry(i, list, list) {
 		BUG_ON(i->pin && atomic_read(i->pin) != 1);
 
-		if (n != i->j.seq)
-			pr_err(
-		"journal entries %llu-%llu missing! (replaying %llu-%llu)\n",
-		n, i->j.seq - 1, start, end);
+		cache_set_err_on(n != i->j.seq, s,
+"bcache: journal entries %llu-%llu missing! (replaying %llu-%llu)",
+				 n, i->j.seq - 1, start, end);
 
 		for (k = i->j.start;
 		     k < end(&i->j);
 		     k = bkey_next(k)) {
 			trace_bcache_journal_replay_key(k);
 
-			bkey_copy(op->keys.top, k);
-			bch_keylist_push(&op->keys);
-
-			op->journal = i->pin;
-			atomic_inc(op->journal);
+			bkey_copy(keylist.top, k);
+			bch_keylist_push(&keylist);
 
-			ret = bch_btree_insert(op, s);
+			ret = bch_btree_insert(s, &keylist, i->pin, NULL);
 			if (ret)
 				goto err;
 
-			BUG_ON(!bch_keylist_empty(&op->keys));
+			BUG_ON(!bch_keylist_empty(&keylist));
 			keys++;
 
 			cond_resched();
@@ -339,14 +338,13 @@ int bch_journal_replay(struct cache_set *s, struct list_head *list,
 
 	pr_info("journal replay done, %i keys in %i entries, seq %llu",
 		keys, entries, end);
-
+err:
 	while (!list_empty(list)) {
 		i = list_first_entry(list, struct journal_replay, list);
 		list_del(&i->list);
 		kfree(i);
 	}
-err:
-	closure_sync(&op->cl);
+
 	return ret;
 }
 
@@ -358,48 +356,35 @@ static void btree_flush_write(struct cache_set *c)
 	 * Try to find the btree node with that references the oldest journal
 	 * entry, best is our current candidate and is locked if non NULL:
 	 */
-	struct btree *b, *best = NULL;
-	unsigned iter;
+	struct btree *b, *best;
+	unsigned i;
+retry:
+	best = NULL;
+
+	for_each_cached_btree(b, c, i)
+		if (btree_current_write(b)->journal) {
+			if (!best)
+				best = b;
+			else if (journal_pin_cmp(c,
+					btree_current_write(best)->journal,
+					btree_current_write(b)->journal)) {
+				best = b;
+			}
+		}
 
-	for_each_cached_btree(b, c, iter) {
-		if (!down_write_trylock(&b->lock))
-			continue;
+	b = best;
+	if (b) {
+		rw_lock(true, b, b->level);
 
-		if (!btree_node_dirty(b) ||
-		    !btree_current_write(b)->journal) {
+		if (!btree_current_write(b)->journal) {
 			rw_unlock(true, b);
-			continue;
+			/* We raced */
+			goto retry;
 		}
 
-		if (!best)
-			best = b;
-		else if (journal_pin_cmp(c,
-					 btree_current_write(best),
-					 btree_current_write(b))) {
-			rw_unlock(true, best);
-			best = b;
-		} else
-			rw_unlock(true, b);
+		bch_btree_node_write(b, NULL);
+		rw_unlock(true, b);
 	}
-
-	if (best)
-		goto out;
-
-	/* We can't find the best btree node, just pick the first */
-	list_for_each_entry(b, &c->btree_cache, list)
-		if (!b->level && btree_node_dirty(b)) {
-			best = b;
-			rw_lock(true, best, best->level);
-			goto found;
-		}
-
-out:
-	if (!best)
-		return;
-found:
-	if (btree_node_dirty(best))
-		bch_btree_node_write(best, NULL);
-	rw_unlock(true, best);
 }
 
 #define last_seq(j)	((j)->seq - fifo_used(&(j)->pin) + 1)
@@ -495,7 +480,7 @@ static void journal_reclaim(struct cache_set *c)
 		do_journal_discard(ca);
 
 	if (c->journal.blocks_free)
-		return;
+		goto out;
 
 	/*
 	 * Allocate:
@@ -521,7 +506,7 @@ static void journal_reclaim(struct cache_set *c)
 
 	if (n)
 		c->journal.blocks_free = c->sb.bucket_size >> c->block_bits;
-
+out:
 	if (!journal_full(&c->journal))
 		__closure_wake_up(&c->journal.wait);
 }
@@ -554,32 +539,26 @@ static void journal_write_endio(struct bio *bio, int error)
 	struct journal_write *w = bio->bi_private;
 
 	cache_set_err_on(error, w->c, "journal io error");
-	closure_put(&w->c->journal.io.cl);
+	closure_put(&w->c->journal.io);
 }
 
 static void journal_write(struct closure *);
 
 static void journal_write_done(struct closure *cl)
 {
-	struct journal *j = container_of(cl, struct journal, io.cl);
-	struct cache_set *c = container_of(j, struct cache_set, journal);
-
+	struct journal *j = container_of(cl, struct journal, io);
 	struct journal_write *w = (j->cur == j->w)
 		? &j->w[1]
 		: &j->w[0];
 
 	__closure_wake_up(&w->wait);
-
-	if (c->journal_delay_ms)
-		closure_delay(&j->io, msecs_to_jiffies(c->journal_delay_ms));
-
-	continue_at(cl, journal_write, system_wq);
+	continue_at_nobarrier(cl, journal_write, system_wq);
 }
 
 static void journal_write_unlocked(struct closure *cl)
 	__releases(c->journal.lock)
 {
-	struct cache_set *c = container_of(cl, struct cache_set, journal.io.cl);
+	struct cache_set *c = container_of(cl, struct cache_set, journal.io);
 	struct cache *ca;
 	struct journal_write *w = c->journal.cur;
 	struct bkey *k = &c->journal.key;
@@ -617,7 +596,7 @@ static void journal_write_unlocked(struct closure *cl)
 	for_each_cache(ca, c, i)
 		w->data->prio_bucket[ca->sb.nr_this_dev] = ca->prio_buckets[0];
 
-	w->data->magic		= jset_magic(c);
+	w->data->magic		= jset_magic(&c->sb);
 	w->data->version	= BCACHE_JSET_VERSION;
 	w->data->last_seq	= last_seq(&c->journal);
 	w->data->csum		= csum_set(w->data);
@@ -660,121 +639,134 @@ static void journal_write_unlocked(struct closure *cl)
 
 static void journal_write(struct closure *cl)
 {
-	struct cache_set *c = container_of(cl, struct cache_set, journal.io.cl);
+	struct cache_set *c = container_of(cl, struct cache_set, journal.io);
 
 	spin_lock(&c->journal.lock);
 	journal_write_unlocked(cl);
 }
 
-static void __journal_try_write(struct cache_set *c, bool noflush)
+static void journal_try_write(struct cache_set *c)
 	__releases(c->journal.lock)
 {
-	struct closure *cl = &c->journal.io.cl;
+	struct closure *cl = &c->journal.io;
+	struct journal_write *w = c->journal.cur;
 
-	if (!closure_trylock(cl, &c->cl))
-		spin_unlock(&c->journal.lock);
-	else if (noflush && journal_full(&c->journal)) {
-		spin_unlock(&c->journal.lock);
-		continue_at(cl, journal_write, system_wq);
-	} else
+	w->need_write = true;
+
+	if (closure_trylock(cl, &c->cl))
 		journal_write_unlocked(cl);
+	else
+		spin_unlock(&c->journal.lock);
 }
 
-#define journal_try_write(c)	__journal_try_write(c, false)
-
-void bch_journal_meta(struct cache_set *c, struct closure *cl)
+static struct journal_write *journal_wait_for_write(struct cache_set *c,
+						    unsigned nkeys)
 {
-	struct journal_write *w;
+	size_t sectors;
+	struct closure cl;
 
-	if (CACHE_SYNC(&c->sb)) {
-		spin_lock(&c->journal.lock);
+	closure_init_stack(&cl);
+
+	spin_lock(&c->journal.lock);
 
-		w = c->journal.cur;
-		w->need_write = true;
+	while (1) {
+		struct journal_write *w = c->journal.cur;
 
-		if (cl)
-			BUG_ON(!closure_wait(&w->wait, cl));
+		sectors = __set_blocks(w->data, w->data->keys + nkeys,
+				       c) * c->sb.block_size;
 
-		closure_flush(&c->journal.io);
-		__journal_try_write(c, true);
+		if (sectors <= min_t(size_t,
+				     c->journal.blocks_free * c->sb.block_size,
+				     PAGE_SECTORS << JSET_BITS))
+			return w;
+
+		/* XXX: tracepoint */
+		if (!journal_full(&c->journal)) {
+			trace_bcache_journal_entry_full(c);
+
+			/*
+			 * XXX: If we were inserting so many keys that they
+			 * won't fit in an _empty_ journal write, we'll
+			 * deadlock. For now, handle this in
+			 * bch_keylist_realloc() - but something to think about.
+			 */
+			BUG_ON(!w->data->keys);
+
+			closure_wait(&w->wait, &cl);
+			journal_try_write(c); /* unlocks */
+		} else {
+			trace_bcache_journal_full(c);
+
+			closure_wait(&c->journal.wait, &cl);
+			journal_reclaim(c);
+			spin_unlock(&c->journal.lock);
+
+			btree_flush_write(c);
+		}
+
+		closure_sync(&cl);
+		spin_lock(&c->journal.lock);
 	}
 }
 
+static void journal_write_work(struct work_struct *work)
+{
+	struct cache_set *c = container_of(to_delayed_work(work),
+					   struct cache_set,
+					   journal.work);
+	spin_lock(&c->journal.lock);
+	journal_try_write(c);
+}
+
 /*
  * Entry point to the journalling code - bio_insert() and btree_invalidate()
  * pass bch_journal() a list of keys to be journalled, and then
  * bch_journal() hands those same keys off to btree_insert_async()
  */
 
-void bch_journal(struct closure *cl)
+atomic_t *bch_journal(struct cache_set *c,
+		      struct keylist *keys,
+		      struct closure *parent)
 {
-	struct btree_op *op = container_of(cl, struct btree_op, cl);
-	struct cache_set *c = op->c;
 	struct journal_write *w;
-	size_t b, n = ((uint64_t *) op->keys.top) - op->keys.list;
-
-	if (op->type != BTREE_INSERT ||
-	    !CACHE_SYNC(&c->sb))
-		goto out;
+	atomic_t *ret;
 
-	/*
-	 * If we're looping because we errored, might already be waiting on
-	 * another journal write:
-	 */
-	while (atomic_read(&cl->parent->remaining) & CLOSURE_WAITING)
-		closure_sync(cl->parent);
+	if (!CACHE_SYNC(&c->sb))
+		return NULL;
 
-	spin_lock(&c->journal.lock);
+	w = journal_wait_for_write(c, bch_keylist_nkeys(keys));
 
-	if (journal_full(&c->journal)) {
-		trace_bcache_journal_full(c);
+	memcpy(end(w->data), keys->keys, bch_keylist_bytes(keys));
+	w->data->keys += bch_keylist_nkeys(keys);
 
-		closure_wait(&c->journal.wait, cl);
+	ret = &fifo_back(&c->journal.pin);
+	atomic_inc(ret);
 
-		journal_reclaim(c);
+	if (parent) {
+		closure_wait(&w->wait, parent);
+		journal_try_write(c);
+	} else if (!w->need_write) {
+		schedule_delayed_work(&c->journal.work,
+				      msecs_to_jiffies(c->journal_delay_ms));
+		spin_unlock(&c->journal.lock);
+	} else {
 		spin_unlock(&c->journal.lock);
-
-		btree_flush_write(c);
-		continue_at(cl, bch_journal, bcache_wq);
 	}
 
-	w = c->journal.cur;
-	w->need_write = true;
-	b = __set_blocks(w->data, w->data->keys + n, c);
-
-	if (b * c->sb.block_size > PAGE_SECTORS << JSET_BITS ||
-	    b > c->journal.blocks_free) {
-		trace_bcache_journal_entry_full(c);
-
-		/*
-		 * XXX: If we were inserting so many keys that they won't fit in
-		 * an _empty_ journal write, we'll deadlock. For now, handle
-		 * this in bch_keylist_realloc() - but something to think about.
-		 */
-		BUG_ON(!w->data->keys);
-
-		BUG_ON(!closure_wait(&w->wait, cl));
-
-		closure_flush(&c->journal.io);
 
-		journal_try_write(c);
-		continue_at(cl, bch_journal, bcache_wq);
-	}
-
-	memcpy(end(w->data), op->keys.list, n * sizeof(uint64_t));
-	w->data->keys += n;
+	return ret;
+}
 
-	op->journal = &fifo_back(&c->journal.pin);
-	atomic_inc(op->journal);
+void bch_journal_meta(struct cache_set *c, struct closure *cl)
+{
+	struct keylist keys;
+	atomic_t *ref;
 
-	if (op->flush_journal) {
-		closure_flush(&c->journal.io);
-		closure_wait(&w->wait, cl->parent);
-	}
+	bch_keylist_init(&keys);
 
-	journal_try_write(c);
-out:
-	bch_btree_insert_async(cl);
+	ref = bch_journal(c, &keys, cl);
+	if (ref)
+		atomic_dec_bug(ref);
 }
 
 void bch_journal_free(struct cache_set *c)
@@ -790,6 +782,7 @@ int bch_journal_alloc(struct cache_set *c)
 
 	closure_init_unlocked(&j->io);
 	spin_lock_init(&j->lock);
+	INIT_DELAYED_WORK(&j->work, journal_write_work);
 
 	c->journal_delay_ms = 100;
 
diff --git a/drivers/md/bcache/journal.h b/drivers/md/bcache/journal.h
index 3d7851274b04..a6472fda94b2 100644
--- a/drivers/md/bcache/journal.h
+++ b/drivers/md/bcache/journal.h
@@ -75,43 +75,6 @@
  * nodes that are pinning the oldest journal entries first.
  */
 
-#define BCACHE_JSET_VERSION_UUIDv1	1
-/* Always latest UUID format */
-#define BCACHE_JSET_VERSION_UUID	1
-#define BCACHE_JSET_VERSION		1
-
-/*
- * On disk format for a journal entry:
- * seq is monotonically increasing; every journal entry has its own unique
- * sequence number.
- *
- * last_seq is the oldest journal entry that still has keys the btree hasn't
- * flushed to disk yet.
- *
- * version is for on disk format changes.
- */
-struct jset {
-	uint64_t		csum;
-	uint64_t		magic;
-	uint64_t		seq;
-	uint32_t		version;
-	uint32_t		keys;
-
-	uint64_t		last_seq;
-
-	BKEY_PADDED(uuid_bucket);
-	BKEY_PADDED(btree_root);
-	uint16_t		btree_level;
-	uint16_t		pad[3];
-
-	uint64_t		prio_bucket[MAX_CACHES_PER_SET];
-
-	union {
-		struct bkey	start[0];
-		uint64_t	d[0];
-	};
-};
-
 /*
  * Only used for holding the journal entries we read in btree_journal_read()
  * during cache_registration
@@ -140,7 +103,8 @@ struct journal {
 	spinlock_t		lock;
 	/* used when waiting because the journal was full */
 	struct closure_waitlist	wait;
-	struct closure_with_timer io;
+	struct closure		io;
+	struct delayed_work	work;
 
 	/* Number of blocks free in the bucket(s) we're currently writing to */
 	unsigned		blocks_free;
@@ -188,8 +152,7 @@ struct journal_device {
 };
 
 #define journal_pin_cmp(c, l, r)				\
-	(fifo_idx(&(c)->journal.pin, (l)->journal) >		\
-	 fifo_idx(&(c)->journal.pin, (r)->journal))
+	(fifo_idx(&(c)->journal.pin, (l)) > fifo_idx(&(c)->journal.pin, (r)))
 
 #define JOURNAL_PIN	20000
 
@@ -199,15 +162,14 @@ struct journal_device {
 struct closure;
 struct cache_set;
 struct btree_op;
+struct keylist;
 
-void bch_journal(struct closure *);
+atomic_t *bch_journal(struct cache_set *, struct keylist *, struct closure *);
 void bch_journal_next(struct journal *);
 void bch_journal_mark(struct cache_set *, struct list_head *);
 void bch_journal_meta(struct cache_set *, struct closure *);
-int bch_journal_read(struct cache_set *, struct list_head *,
-			struct btree_op *);
-int bch_journal_replay(struct cache_set *, struct list_head *,
-			  struct btree_op *);
+int bch_journal_read(struct cache_set *, struct list_head *);
+int bch_journal_replay(struct cache_set *, struct list_head *);
 
 void bch_journal_free(struct cache_set *);
 int bch_journal_alloc(struct cache_set *);
diff --git a/drivers/md/bcache/movinggc.c b/drivers/md/bcache/movinggc.c
index 1a3b4f4786c3..7c1275e66025 100644
--- a/drivers/md/bcache/movinggc.c
+++ b/drivers/md/bcache/movinggc.c
@@ -12,8 +12,9 @@
 #include <trace/events/bcache.h>
 
 struct moving_io {
+	struct closure		cl;
 	struct keybuf_key	*w;
-	struct search		s;
+	struct data_insert_op	op;
 	struct bbio		bio;
 };
 
@@ -38,13 +39,13 @@ static bool moving_pred(struct keybuf *buf, struct bkey *k)
 
 static void moving_io_destructor(struct closure *cl)
 {
-	struct moving_io *io = container_of(cl, struct moving_io, s.cl);
+	struct moving_io *io = container_of(cl, struct moving_io, cl);
 	kfree(io);
 }
 
 static void write_moving_finish(struct closure *cl)
 {
-	struct moving_io *io = container_of(cl, struct moving_io, s.cl);
+	struct moving_io *io = container_of(cl, struct moving_io, cl);
 	struct bio *bio = &io->bio.bio;
 	struct bio_vec *bv;
 	int i;
@@ -52,13 +53,12 @@ static void write_moving_finish(struct closure *cl)
 	bio_for_each_segment_all(bv, bio, i)
 		__free_page(bv->bv_page);
 
-	if (io->s.op.insert_collision)
+	if (io->op.replace_collision)
 		trace_bcache_gc_copy_collision(&io->w->key);
 
-	bch_keybuf_del(&io->s.op.c->moving_gc_keys, io->w);
+	bch_keybuf_del(&io->op.c->moving_gc_keys, io->w);
 
-	atomic_dec_bug(&io->s.op.c->in_flight);
-	closure_wake_up(&io->s.op.c->moving_gc_wait);
+	up(&io->op.c->moving_in_flight);
 
 	closure_return_with_destructor(cl, moving_io_destructor);
 }
@@ -66,12 +66,12 @@ static void write_moving_finish(struct closure *cl)
 static void read_moving_endio(struct bio *bio, int error)
 {
 	struct moving_io *io = container_of(bio->bi_private,
-					    struct moving_io, s.cl);
+					    struct moving_io, cl);
 
 	if (error)
-		io->s.error = error;
+		io->op.error = error;
 
-	bch_bbio_endio(io->s.op.c, bio, error, "reading data to move");
+	bch_bbio_endio(io->op.c, bio, error, "reading data to move");
 }
 
 static void moving_init(struct moving_io *io)
@@ -85,54 +85,53 @@ static void moving_init(struct moving_io *io)
 	bio->bi_size		= KEY_SIZE(&io->w->key) << 9;
 	bio->bi_max_vecs	= DIV_ROUND_UP(KEY_SIZE(&io->w->key),
 					       PAGE_SECTORS);
-	bio->bi_private		= &io->s.cl;
+	bio->bi_private		= &io->cl;
 	bio->bi_io_vec		= bio->bi_inline_vecs;
 	bch_bio_map(bio, NULL);
 }
 
 static void write_moving(struct closure *cl)
 {
-	struct search *s = container_of(cl, struct search, cl);
-	struct moving_io *io = container_of(s, struct moving_io, s);
+	struct moving_io *io = container_of(cl, struct moving_io, cl);
+	struct data_insert_op *op = &io->op;
 
-	if (!s->error) {
+	if (!op->error) {
 		moving_init(io);
 
-		io->bio.bio.bi_sector	= KEY_START(&io->w->key);
-		s->op.lock		= -1;
-		s->op.write_prio	= 1;
-		s->op.cache_bio		= &io->bio.bio;
+		io->bio.bio.bi_sector = KEY_START(&io->w->key);
+		op->write_prio		= 1;
+		op->bio			= &io->bio.bio;
 
-		s->writeback		= KEY_DIRTY(&io->w->key);
-		s->op.csum		= KEY_CSUM(&io->w->key);
+		op->writeback		= KEY_DIRTY(&io->w->key);
+		op->csum		= KEY_CSUM(&io->w->key);
 
-		s->op.type = BTREE_REPLACE;
-		bkey_copy(&s->op.replace, &io->w->key);
+		bkey_copy(&op->replace_key, &io->w->key);
+		op->replace		= true;
 
-		closure_init(&s->op.cl, cl);
-		bch_insert_data(&s->op.cl);
+		closure_call(&op->cl, bch_data_insert, NULL, cl);
 	}
 
-	continue_at(cl, write_moving_finish, NULL);
+	continue_at(cl, write_moving_finish, system_wq);
 }
 
 static void read_moving_submit(struct closure *cl)
 {
-	struct search *s = container_of(cl, struct search, cl);
-	struct moving_io *io = container_of(s, struct moving_io, s);
+	struct moving_io *io = container_of(cl, struct moving_io, cl);
 	struct bio *bio = &io->bio.bio;
 
-	bch_submit_bbio(bio, s->op.c, &io->w->key, 0);
+	bch_submit_bbio(bio, io->op.c, &io->w->key, 0);
 
-	continue_at(cl, write_moving, bch_gc_wq);
+	continue_at(cl, write_moving, system_wq);
 }
 
-static void read_moving(struct closure *cl)
+static void read_moving(struct cache_set *c)
 {
-	struct cache_set *c = container_of(cl, struct cache_set, moving_gc);
 	struct keybuf_key *w;
 	struct moving_io *io;
 	struct bio *bio;
+	struct closure cl;
+
+	closure_init_stack(&cl);
 
 	/* XXX: if we error, background writeback could stall indefinitely */
 
@@ -150,8 +149,8 @@ static void read_moving(struct closure *cl)
 
 		w->private	= io;
 		io->w		= w;
-		io->s.op.inode	= KEY_INODE(&w->key);
-		io->s.op.c	= c;
+		io->op.inode	= KEY_INODE(&w->key);
+		io->op.c	= c;
 
 		moving_init(io);
 		bio = &io->bio.bio;
@@ -164,13 +163,8 @@ static void read_moving(struct closure *cl)
 
 		trace_bcache_gc_copy(&w->key);
 
-		closure_call(&io->s.cl, read_moving_submit, NULL, &c->gc.cl);
-
-		if (atomic_inc_return(&c->in_flight) >= 64) {
-			closure_wait_event(&c->moving_gc_wait, cl,
-					   atomic_read(&c->in_flight) < 64);
-			continue_at(cl, read_moving, bch_gc_wq);
-		}
+		down(&c->moving_in_flight);
+		closure_call(&io->cl, read_moving_submit, NULL, &cl);
 	}
 
 	if (0) {
@@ -180,7 +174,7 @@ err:		if (!IS_ERR_OR_NULL(w->private))
 		bch_keybuf_del(&c->moving_gc_keys, w);
 	}
 
-	closure_return(cl);
+	closure_sync(&cl);
 }
 
 static bool bucket_cmp(struct bucket *l, struct bucket *r)
@@ -193,15 +187,14 @@ static unsigned bucket_heap_top(struct cache *ca)
 	return GC_SECTORS_USED(heap_peek(&ca->heap));
 }
 
-void bch_moving_gc(struct closure *cl)
+void bch_moving_gc(struct cache_set *c)
 {
-	struct cache_set *c = container_of(cl, struct cache_set, gc.cl);
 	struct cache *ca;
 	struct bucket *b;
 	unsigned i;
 
 	if (!c->copy_gc_enabled)
-		closure_return(cl);
+		return;
 
 	mutex_lock(&c->bucket_lock);
 
@@ -242,13 +235,11 @@ void bch_moving_gc(struct closure *cl)
 
 	c->moving_gc_keys.last_scanned = ZERO_KEY;
 
-	closure_init(&c->moving_gc, cl);
-	read_moving(&c->moving_gc);
-
-	closure_return(cl);
+	read_moving(c);
 }
 
 void bch_moving_init_cache_set(struct cache_set *c)
 {
 	bch_keybuf_init(&c->moving_gc_keys);
+	sema_init(&c->moving_in_flight, 64);
 }
diff --git a/drivers/md/bcache/request.c b/drivers/md/bcache/request.c
index 2a7f0dd6abab..fbcc851ed5a5 100644
--- a/drivers/md/bcache/request.c
+++ b/drivers/md/bcache/request.c
@@ -25,7 +25,7 @@
 
 struct kmem_cache *bch_search_cache;
 
-static void check_should_skip(struct cached_dev *, struct search *);
+static void bch_data_insert_start(struct closure *);
 
 /* Cgroup interface */
 
@@ -213,221 +213,79 @@ static void bio_csum(struct bio *bio, struct bkey *k)
 
 /* Insert data into cache */
 
-static void bio_invalidate(struct closure *cl)
+static void bch_data_insert_keys(struct closure *cl)
 {
-	struct btree_op *op = container_of(cl, struct btree_op, cl);
-	struct bio *bio = op->cache_bio;
-
-	pr_debug("invalidating %i sectors from %llu",
-		 bio_sectors(bio), (uint64_t) bio->bi_sector);
-
-	while (bio_sectors(bio)) {
-		unsigned len = min(bio_sectors(bio), 1U << 14);
-
-		if (bch_keylist_realloc(&op->keys, 0, op->c))
-			goto out;
-
-		bio->bi_sector	+= len;
-		bio->bi_size	-= len << 9;
-
-		bch_keylist_add(&op->keys,
-				&KEY(op->inode, bio->bi_sector, len));
-	}
-
-	op->insert_data_done = true;
-	bio_put(bio);
-out:
-	continue_at(cl, bch_journal, bcache_wq);
-}
-
-struct open_bucket {
-	struct list_head	list;
-	struct task_struct	*last;
-	unsigned		sectors_free;
-	BKEY_PADDED(key);
-};
-
-void bch_open_buckets_free(struct cache_set *c)
-{
-	struct open_bucket *b;
-
-	while (!list_empty(&c->data_buckets)) {
-		b = list_first_entry(&c->data_buckets,
-				     struct open_bucket, list);
-		list_del(&b->list);
-		kfree(b);
-	}
-}
-
-int bch_open_buckets_alloc(struct cache_set *c)
-{
-	int i;
-
-	spin_lock_init(&c->data_bucket_lock);
-
-	for (i = 0; i < 6; i++) {
-		struct open_bucket *b = kzalloc(sizeof(*b), GFP_KERNEL);
-		if (!b)
-			return -ENOMEM;
-
-		list_add(&b->list, &c->data_buckets);
-	}
-
-	return 0;
-}
-
-/*
- * We keep multiple buckets open for writes, and try to segregate different
- * write streams for better cache utilization: first we look for a bucket where
- * the last write to it was sequential with the current write, and failing that
- * we look for a bucket that was last used by the same task.
- *
- * The ideas is if you've got multiple tasks pulling data into the cache at the
- * same time, you'll get better cache utilization if you try to segregate their
- * data and preserve locality.
- *
- * For example, say you've starting Firefox at the same time you're copying a
- * bunch of files. Firefox will likely end up being fairly hot and stay in the
- * cache awhile, but the data you copied might not be; if you wrote all that
- * data to the same buckets it'd get invalidated at the same time.
- *
- * Both of those tasks will be doing fairly random IO so we can't rely on
- * detecting sequential IO to segregate their data, but going off of the task
- * should be a sane heuristic.
- */
-static struct open_bucket *pick_data_bucket(struct cache_set *c,
-					    const struct bkey *search,
-					    struct task_struct *task,
-					    struct bkey *alloc)
-{
-	struct open_bucket *ret, *ret_task = NULL;
-
-	list_for_each_entry_reverse(ret, &c->data_buckets, list)
-		if (!bkey_cmp(&ret->key, search))
-			goto found;
-		else if (ret->last == task)
-			ret_task = ret;
-
-	ret = ret_task ?: list_first_entry(&c->data_buckets,
-					   struct open_bucket, list);
-found:
-	if (!ret->sectors_free && KEY_PTRS(alloc)) {
-		ret->sectors_free = c->sb.bucket_size;
-		bkey_copy(&ret->key, alloc);
-		bkey_init(alloc);
-	}
-
-	if (!ret->sectors_free)
-		ret = NULL;
-
-	return ret;
-}
-
-/*
- * Allocates some space in the cache to write to, and k to point to the newly
- * allocated space, and updates KEY_SIZE(k) and KEY_OFFSET(k) (to point to the
- * end of the newly allocated space).
- *
- * May allocate fewer sectors than @sectors, KEY_SIZE(k) indicates how many
- * sectors were actually allocated.
- *
- * If s->writeback is true, will not fail.
- */
-static bool bch_alloc_sectors(struct bkey *k, unsigned sectors,
-			      struct search *s)
-{
-	struct cache_set *c = s->op.c;
-	struct open_bucket *b;
-	BKEY_PADDED(key) alloc;
-	struct closure cl, *w = NULL;
-	unsigned i;
-
-	if (s->writeback) {
-		closure_init_stack(&cl);
-		w = &cl;
-	}
+	struct data_insert_op *op = container_of(cl, struct data_insert_op, cl);
+	atomic_t *journal_ref = NULL;
+	struct bkey *replace_key = op->replace ? &op->replace_key : NULL;
+	int ret;
 
 	/*
-	 * We might have to allocate a new bucket, which we can't do with a
-	 * spinlock held. So if we have to allocate, we drop the lock, allocate
-	 * and then retry. KEY_PTRS() indicates whether alloc points to
-	 * allocated bucket(s).
+	 * If we're looping, might already be waiting on
+	 * another journal write - can't wait on more than one journal write at
+	 * a time
+	 *
+	 * XXX: this looks wrong
 	 */
+#if 0
+	while (atomic_read(&s->cl.remaining) & CLOSURE_WAITING)
+		closure_sync(&s->cl);
+#endif
 
-	bkey_init(&alloc.key);
-	spin_lock(&c->data_bucket_lock);
-
-	while (!(b = pick_data_bucket(c, k, s->task, &alloc.key))) {
-		unsigned watermark = s->op.write_prio
-			? WATERMARK_MOVINGGC
-			: WATERMARK_NONE;
-
-		spin_unlock(&c->data_bucket_lock);
-
-		if (bch_bucket_alloc_set(c, watermark, &alloc.key, 1, w))
-			return false;
+	if (!op->replace)
+		journal_ref = bch_journal(op->c, &op->insert_keys,
+					  op->flush_journal ? cl : NULL);
 
-		spin_lock(&c->data_bucket_lock);
+	ret = bch_btree_insert(op->c, &op->insert_keys,
+			       journal_ref, replace_key);
+	if (ret == -ESRCH) {
+		op->replace_collision = true;
+	} else if (ret) {
+		op->error		= -ENOMEM;
+		op->insert_data_done	= true;
 	}
 
-	/*
-	 * If we had to allocate, we might race and not need to allocate the
-	 * second time we call find_data_bucket(). If we allocated a bucket but
-	 * didn't use it, drop the refcount bch_bucket_alloc_set() took:
-	 */
-	if (KEY_PTRS(&alloc.key))
-		__bkey_put(c, &alloc.key);
-
-	for (i = 0; i < KEY_PTRS(&b->key); i++)
-		EBUG_ON(ptr_stale(c, &b->key, i));
+	if (journal_ref)
+		atomic_dec_bug(journal_ref);
 
-	/* Set up the pointer to the space we're allocating: */
+	if (!op->insert_data_done)
+		continue_at(cl, bch_data_insert_start, bcache_wq);
 
-	for (i = 0; i < KEY_PTRS(&b->key); i++)
-		k->ptr[i] = b->key.ptr[i];
+	bch_keylist_free(&op->insert_keys);
+	closure_return(cl);
+}
 
-	sectors = min(sectors, b->sectors_free);
+static void bch_data_invalidate(struct closure *cl)
+{
+	struct data_insert_op *op = container_of(cl, struct data_insert_op, cl);
+	struct bio *bio = op->bio;
 
-	SET_KEY_OFFSET(k, KEY_OFFSET(k) + sectors);
-	SET_KEY_SIZE(k, sectors);
-	SET_KEY_PTRS(k, KEY_PTRS(&b->key));
+	pr_debug("invalidating %i sectors from %llu",
+		 bio_sectors(bio), (uint64_t) bio->bi_sector);
 
-	/*
-	 * Move b to the end of the lru, and keep track of what this bucket was
-	 * last used for:
-	 */
-	list_move_tail(&b->list, &c->data_buckets);
-	bkey_copy_key(&b->key, k);
-	b->last = s->task;
+	while (bio_sectors(bio)) {
+		unsigned sectors = min(bio_sectors(bio),
+				       1U << (KEY_SIZE_BITS - 1));
 
-	b->sectors_free	-= sectors;
+		if (bch_keylist_realloc(&op->insert_keys, 0, op->c))
+			goto out;
 
-	for (i = 0; i < KEY_PTRS(&b->key); i++) {
-		SET_PTR_OFFSET(&b->key, i, PTR_OFFSET(&b->key, i) + sectors);
+		bio->bi_sector	+= sectors;
+		bio->bi_size	-= sectors << 9;
 
-		atomic_long_add(sectors,
-				&PTR_CACHE(c, &b->key, i)->sectors_written);
+		bch_keylist_add(&op->insert_keys,
+				&KEY(op->inode, bio->bi_sector, sectors));
 	}
 
-	if (b->sectors_free < c->sb.block_size)
-		b->sectors_free = 0;
-
-	/*
-	 * k takes refcounts on the buckets it points to until it's inserted
-	 * into the btree, but if we're done with this bucket we just transfer
-	 * get_data_bucket()'s refcount.
-	 */
-	if (b->sectors_free)
-		for (i = 0; i < KEY_PTRS(&b->key); i++)
-			atomic_inc(&PTR_BUCKET(c, &b->key, i)->pin);
-
-	spin_unlock(&c->data_bucket_lock);
-	return true;
+	op->insert_data_done = true;
+	bio_put(bio);
+out:
+	continue_at(cl, bch_data_insert_keys, bcache_wq);
 }
 
-static void bch_insert_data_error(struct closure *cl)
+static void bch_data_insert_error(struct closure *cl)
 {
-	struct btree_op *op = container_of(cl, struct btree_op, cl);
+	struct data_insert_op *op = container_of(cl, struct data_insert_op, cl);
 
 	/*
 	 * Our data write just errored, which means we've got a bunch of keys to
@@ -438,35 +296,34 @@ static void bch_insert_data_error(struct closure *cl)
 	 * from the keys we'll accomplish just that.
 	 */
 
-	struct bkey *src = op->keys.bottom, *dst = op->keys.bottom;
+	struct bkey *src = op->insert_keys.keys, *dst = op->insert_keys.keys;
 
-	while (src != op->keys.top) {
+	while (src != op->insert_keys.top) {
 		struct bkey *n = bkey_next(src);
 
 		SET_KEY_PTRS(src, 0);
-		bkey_copy(dst, src);
+		memmove(dst, src, bkey_bytes(src));
 
 		dst = bkey_next(dst);
 		src = n;
 	}
 
-	op->keys.top = dst;
+	op->insert_keys.top = dst;
 
-	bch_journal(cl);
+	bch_data_insert_keys(cl);
 }
 
-static void bch_insert_data_endio(struct bio *bio, int error)
+static void bch_data_insert_endio(struct bio *bio, int error)
 {
 	struct closure *cl = bio->bi_private;
-	struct btree_op *op = container_of(cl, struct btree_op, cl);
-	struct search *s = container_of(op, struct search, op);
+	struct data_insert_op *op = container_of(cl, struct data_insert_op, cl);
 
 	if (error) {
 		/* TODO: We could try to recover from this. */
-		if (s->writeback)
-			s->error = error;
-		else if (s->write)
-			set_closure_fn(cl, bch_insert_data_error, bcache_wq);
+		if (op->writeback)
+			op->error = error;
+		else if (!op->replace)
+			set_closure_fn(cl, bch_data_insert_error, bcache_wq);
 		else
 			set_closure_fn(cl, NULL, NULL);
 	}
@@ -474,18 +331,17 @@ static void bch_insert_data_endio(struct bio *bio, int error)
 	bch_bbio_endio(op->c, bio, error, "writing data to cache");
 }
 
-static void bch_insert_data_loop(struct closure *cl)
+static void bch_data_insert_start(struct closure *cl)
 {
-	struct btree_op *op = container_of(cl, struct btree_op, cl);
-	struct search *s = container_of(op, struct search, op);
-	struct bio *bio = op->cache_bio, *n;
+	struct data_insert_op *op = container_of(cl, struct data_insert_op, cl);
+	struct bio *bio = op->bio, *n;
 
-	if (op->skip)
-		return bio_invalidate(cl);
+	if (op->bypass)
+		return bch_data_invalidate(cl);
 
 	if (atomic_sub_return(bio_sectors(bio), &op->c->sectors_to_gc) < 0) {
 		set_gc_sectors(op->c);
-		bch_queue_gc(op->c);
+		wake_up_gc(op->c);
 	}
 
 	/*
@@ -497,29 +353,30 @@ static void bch_insert_data_loop(struct closure *cl)
 	do {
 		unsigned i;
 		struct bkey *k;
-		struct bio_set *split = s->d
-			? s->d->bio_split : op->c->bio_split;
+		struct bio_set *split = op->c->bio_split;
 
 		/* 1 for the device pointer and 1 for the chksum */
-		if (bch_keylist_realloc(&op->keys,
+		if (bch_keylist_realloc(&op->insert_keys,
 					1 + (op->csum ? 1 : 0),
 					op->c))
-			continue_at(cl, bch_journal, bcache_wq);
+			continue_at(cl, bch_data_insert_keys, bcache_wq);
 
-		k = op->keys.top;
+		k = op->insert_keys.top;
 		bkey_init(k);
 		SET_KEY_INODE(k, op->inode);
 		SET_KEY_OFFSET(k, bio->bi_sector);
 
-		if (!bch_alloc_sectors(k, bio_sectors(bio), s))
+		if (!bch_alloc_sectors(op->c, k, bio_sectors(bio),
+				       op->write_point, op->write_prio,
+				       op->writeback))
 			goto err;
 
 		n = bch_bio_split(bio, KEY_SIZE(k), GFP_NOIO, split);
 
-		n->bi_end_io	= bch_insert_data_endio;
+		n->bi_end_io	= bch_data_insert_endio;
 		n->bi_private	= cl;
 
-		if (s->writeback) {
+		if (op->writeback) {
 			SET_KEY_DIRTY(k, true);
 
 			for (i = 0; i < KEY_PTRS(k); i++)
@@ -532,17 +389,17 @@ static void bch_insert_data_loop(struct closure *cl)
 			bio_csum(n, k);
 
 		trace_bcache_cache_insert(k);
-		bch_keylist_push(&op->keys);
+		bch_keylist_push(&op->insert_keys);
 
 		n->bi_rw |= REQ_WRITE;
 		bch_submit_bbio(n, op->c, k, 0);
 	} while (n != bio);
 
 	op->insert_data_done = true;
-	continue_at(cl, bch_journal, bcache_wq);
+	continue_at(cl, bch_data_insert_keys, bcache_wq);
 err:
 	/* bch_alloc_sectors() blocks if s->writeback = true */
-	BUG_ON(s->writeback);
+	BUG_ON(op->writeback);
 
 	/*
 	 * But if it's not a writeback write we'd rather just bail out if
@@ -550,15 +407,15 @@ err:
 	 * we might be starving btree writes for gc or something.
 	 */
 
-	if (s->write) {
+	if (!op->replace) {
 		/*
 		 * Writethrough write: We can't complete the write until we've
 		 * updated the index. But we don't want to delay the write while
 		 * we wait for buckets to be freed up, so just invalidate the
 		 * rest of the write.
 		 */
-		op->skip = true;
-		return bio_invalidate(cl);
+		op->bypass = true;
+		return bch_data_invalidate(cl);
 	} else {
 		/*
 		 * From a cache miss, we can just insert the keys for the data
@@ -567,15 +424,15 @@ err:
 		op->insert_data_done = true;
 		bio_put(bio);
 
-		if (!bch_keylist_empty(&op->keys))
-			continue_at(cl, bch_journal, bcache_wq);
+		if (!bch_keylist_empty(&op->insert_keys))
+			continue_at(cl, bch_data_insert_keys, bcache_wq);
 		else
 			closure_return(cl);
 	}
 }
 
 /**
- * bch_insert_data - stick some data in the cache
+ * bch_data_insert - stick some data in the cache
  *
  * This is the starting point for any data to end up in a cache device; it could
  * be from a normal write, or a writeback write, or a write to a flash only
@@ -587,56 +444,179 @@ err:
  * data is written it calls bch_journal, and after the keys have been added to
  * the next journal write they're inserted into the btree.
  *
- * It inserts the data in op->cache_bio; bi_sector is used for the key offset,
+ * It inserts the data in s->cache_bio; bi_sector is used for the key offset,
  * and op->inode is used for the key inode.
  *
- * If op->skip is true, instead of inserting the data it invalidates the region
- * of the cache represented by op->cache_bio and op->inode.
+ * If s->bypass is true, instead of inserting the data it invalidates the
+ * region of the cache represented by s->cache_bio and op->inode.
  */
-void bch_insert_data(struct closure *cl)
+void bch_data_insert(struct closure *cl)
 {
-	struct btree_op *op = container_of(cl, struct btree_op, cl);
+	struct data_insert_op *op = container_of(cl, struct data_insert_op, cl);
+
+	trace_bcache_write(op->bio, op->writeback, op->bypass);
 
-	bch_keylist_init(&op->keys);
-	bio_get(op->cache_bio);
-	bch_insert_data_loop(cl);
+	bch_keylist_init(&op->insert_keys);
+	bio_get(op->bio);
+	bch_data_insert_start(cl);
 }
 
-void bch_btree_insert_async(struct closure *cl)
+/* Congested? */
+
+unsigned bch_get_congested(struct cache_set *c)
 {
-	struct btree_op *op = container_of(cl, struct btree_op, cl);
-	struct search *s = container_of(op, struct search, op);
+	int i;
+	long rand;
 
-	if (bch_btree_insert(op, op->c)) {
-		s->error		= -ENOMEM;
-		op->insert_data_done	= true;
-	}
+	if (!c->congested_read_threshold_us &&
+	    !c->congested_write_threshold_us)
+		return 0;
+
+	i = (local_clock_us() - c->congested_last_us) / 1024;
+	if (i < 0)
+		return 0;
+
+	i += atomic_read(&c->congested);
+	if (i >= 0)
+		return 0;
 
-	if (op->insert_data_done) {
-		bch_keylist_free(&op->keys);
-		closure_return(cl);
-	} else
-		continue_at(cl, bch_insert_data_loop, bcache_wq);
+	i += CONGESTED_MAX;
+
+	if (i > 0)
+		i = fract_exp_two(i, 6);
+
+	rand = get_random_int();
+	i -= bitmap_weight(&rand, BITS_PER_LONG);
+
+	return i > 0 ? i : 1;
 }
 
-/* Common code for the make_request functions */
+static void add_sequential(struct task_struct *t)
+{
+	ewma_add(t->sequential_io_avg,
+		 t->sequential_io, 8, 0);
 
-static void request_endio(struct bio *bio, int error)
+	t->sequential_io = 0;
+}
+
+static struct hlist_head *iohash(struct cached_dev *dc, uint64_t k)
 {
-	struct closure *cl = bio->bi_private;
+	return &dc->io_hash[hash_64(k, RECENT_IO_BITS)];
+}
 
-	if (error) {
-		struct search *s = container_of(cl, struct search, cl);
-		s->error = error;
-		/* Only cache read errors are recoverable */
-		s->recoverable = false;
+static bool check_should_bypass(struct cached_dev *dc, struct bio *bio)
+{
+	struct cache_set *c = dc->disk.c;
+	unsigned mode = cache_mode(dc, bio);
+	unsigned sectors, congested = bch_get_congested(c);
+	struct task_struct *task = current;
+	struct io *i;
+
+	if (test_bit(BCACHE_DEV_DETACHING, &dc->disk.flags) ||
+	    c->gc_stats.in_use > CUTOFF_CACHE_ADD ||
+	    (bio->bi_rw & REQ_DISCARD))
+		goto skip;
+
+	if (mode == CACHE_MODE_NONE ||
+	    (mode == CACHE_MODE_WRITEAROUND &&
+	     (bio->bi_rw & REQ_WRITE)))
+		goto skip;
+
+	if (bio->bi_sector & (c->sb.block_size - 1) ||
+	    bio_sectors(bio) & (c->sb.block_size - 1)) {
+		pr_debug("skipping unaligned io");
+		goto skip;
 	}
 
-	bio_put(bio);
-	closure_put(cl);
+	if (bypass_torture_test(dc)) {
+		if ((get_random_int() & 3) == 3)
+			goto skip;
+		else
+			goto rescale;
+	}
+
+	if (!congested && !dc->sequential_cutoff)
+		goto rescale;
+
+	if (!congested &&
+	    mode == CACHE_MODE_WRITEBACK &&
+	    (bio->bi_rw & REQ_WRITE) &&
+	    (bio->bi_rw & REQ_SYNC))
+		goto rescale;
+
+	spin_lock(&dc->io_lock);
+
+	hlist_for_each_entry(i, iohash(dc, bio->bi_sector), hash)
+		if (i->last == bio->bi_sector &&
+		    time_before(jiffies, i->jiffies))
+			goto found;
+
+	i = list_first_entry(&dc->io_lru, struct io, lru);
+
+	add_sequential(task);
+	i->sequential = 0;
+found:
+	if (i->sequential + bio->bi_size > i->sequential)
+		i->sequential	+= bio->bi_size;
+
+	i->last			 = bio_end_sector(bio);
+	i->jiffies		 = jiffies + msecs_to_jiffies(5000);
+	task->sequential_io	 = i->sequential;
+
+	hlist_del(&i->hash);
+	hlist_add_head(&i->hash, iohash(dc, i->last));
+	list_move_tail(&i->lru, &dc->io_lru);
+
+	spin_unlock(&dc->io_lock);
+
+	sectors = max(task->sequential_io,
+		      task->sequential_io_avg) >> 9;
+
+	if (dc->sequential_cutoff &&
+	    sectors >= dc->sequential_cutoff >> 9) {
+		trace_bcache_bypass_sequential(bio);
+		goto skip;
+	}
+
+	if (congested && sectors >= congested) {
+		trace_bcache_bypass_congested(bio);
+		goto skip;
+	}
+
+rescale:
+	bch_rescale_priorities(c, bio_sectors(bio));
+	return false;
+skip:
+	bch_mark_sectors_bypassed(c, dc, bio_sectors(bio));
+	return true;
 }
 
-void bch_cache_read_endio(struct bio *bio, int error)
+/* Cache lookup */
+
+struct search {
+	/* Stack frame for bio_complete */
+	struct closure		cl;
+
+	struct bcache_device	*d;
+
+	struct bbio		bio;
+	struct bio		*orig_bio;
+	struct bio		*cache_miss;
+
+	unsigned		insert_bio_sectors;
+
+	unsigned		recoverable:1;
+	unsigned		unaligned_bvec:1;
+	unsigned		write:1;
+	unsigned		read_dirty_data:1;
+
+	unsigned long		start_time;
+
+	struct btree_op		op;
+	struct data_insert_op	iop;
+};
+
+static void bch_cache_read_endio(struct bio *bio, int error)
 {
 	struct bbio *b = container_of(bio, struct bbio, bio);
 	struct closure *cl = bio->bi_private;
@@ -650,13 +630,113 @@ void bch_cache_read_endio(struct bio *bio, int error)
 	 */
 
 	if (error)
-		s->error = error;
-	else if (ptr_stale(s->op.c, &b->key, 0)) {
-		atomic_long_inc(&s->op.c->cache_read_races);
-		s->error = -EINTR;
+		s->iop.error = error;
+	else if (ptr_stale(s->iop.c, &b->key, 0)) {
+		atomic_long_inc(&s->iop.c->cache_read_races);
+		s->iop.error = -EINTR;
 	}
 
-	bch_bbio_endio(s->op.c, bio, error, "reading from cache");
+	bch_bbio_endio(s->iop.c, bio, error, "reading from cache");
+}
+
+/*
+ * Read from a single key, handling the initial cache miss if the key starts in
+ * the middle of the bio
+ */
+static int cache_lookup_fn(struct btree_op *op, struct btree *b, struct bkey *k)
+{
+	struct search *s = container_of(op, struct search, op);
+	struct bio *n, *bio = &s->bio.bio;
+	struct bkey *bio_key;
+	unsigned ptr;
+
+	if (bkey_cmp(k, &KEY(s->iop.inode, bio->bi_sector, 0)) <= 0)
+		return MAP_CONTINUE;
+
+	if (KEY_INODE(k) != s->iop.inode ||
+	    KEY_START(k) > bio->bi_sector) {
+		unsigned bio_sectors = bio_sectors(bio);
+		unsigned sectors = KEY_INODE(k) == s->iop.inode
+			? min_t(uint64_t, INT_MAX,
+				KEY_START(k) - bio->bi_sector)
+			: INT_MAX;
+
+		int ret = s->d->cache_miss(b, s, bio, sectors);
+		if (ret != MAP_CONTINUE)
+			return ret;
+
+		/* if this was a complete miss we shouldn't get here */
+		BUG_ON(bio_sectors <= sectors);
+	}
+
+	if (!KEY_SIZE(k))
+		return MAP_CONTINUE;
+
+	/* XXX: figure out best pointer - for multiple cache devices */
+	ptr = 0;
+
+	PTR_BUCKET(b->c, k, ptr)->prio = INITIAL_PRIO;
+
+	if (KEY_DIRTY(k))
+		s->read_dirty_data = true;
+
+	n = bch_bio_split(bio, min_t(uint64_t, INT_MAX,
+				     KEY_OFFSET(k) - bio->bi_sector),
+			  GFP_NOIO, s->d->bio_split);
+
+	bio_key = &container_of(n, struct bbio, bio)->key;
+	bch_bkey_copy_single_ptr(bio_key, k, ptr);
+
+	bch_cut_front(&KEY(s->iop.inode, n->bi_sector, 0), bio_key);
+	bch_cut_back(&KEY(s->iop.inode, bio_end_sector(n), 0), bio_key);
+
+	n->bi_end_io	= bch_cache_read_endio;
+	n->bi_private	= &s->cl;
+
+	/*
+	 * The bucket we're reading from might be reused while our bio
+	 * is in flight, and we could then end up reading the wrong
+	 * data.
+	 *
+	 * We guard against this by checking (in cache_read_endio()) if
+	 * the pointer is stale again; if so, we treat it as an error
+	 * and reread from the backing device (but we don't pass that
+	 * error up anywhere).
+	 */
+
+	__bch_submit_bbio(n, b->c);
+	return n == bio ? MAP_DONE : MAP_CONTINUE;
+}
+
+static void cache_lookup(struct closure *cl)
+{
+	struct search *s = container_of(cl, struct search, iop.cl);
+	struct bio *bio = &s->bio.bio;
+
+	int ret = bch_btree_map_keys(&s->op, s->iop.c,
+				     &KEY(s->iop.inode, bio->bi_sector, 0),
+				     cache_lookup_fn, MAP_END_KEY);
+	if (ret == -EAGAIN)
+		continue_at(cl, cache_lookup, bcache_wq);
+
+	closure_return(cl);
+}
+
+/* Common code for the make_request functions */
+
+static void request_endio(struct bio *bio, int error)
+{
+	struct closure *cl = bio->bi_private;
+
+	if (error) {
+		struct search *s = container_of(cl, struct search, cl);
+		s->iop.error = error;
+		/* Only cache read errors are recoverable */
+		s->recoverable = false;
+	}
+
+	bio_put(bio);
+	closure_put(cl);
 }
 
 static void bio_complete(struct search *s)
@@ -670,8 +750,8 @@ static void bio_complete(struct search *s)
 		part_stat_add(cpu, &s->d->disk->part0, ticks[rw], duration);
 		part_stat_unlock();
 
-		trace_bcache_request_end(s, s->orig_bio);
-		bio_endio(s->orig_bio, s->error);
+		trace_bcache_request_end(s->d, s->orig_bio);
+		bio_endio(s->orig_bio, s->iop.error);
 		s->orig_bio = NULL;
 	}
 }
@@ -691,8 +771,8 @@ static void search_free(struct closure *cl)
 	struct search *s = container_of(cl, struct search, cl);
 	bio_complete(s);
 
-	if (s->op.cache_bio)
-		bio_put(s->op.cache_bio);
+	if (s->iop.bio)
+		bio_put(s->iop.bio);
 
 	if (s->unaligned_bvec)
 		mempool_free(s->bio.bio.bi_io_vec, s->d->unaligned_bvec);
@@ -703,21 +783,22 @@ static void search_free(struct closure *cl)
 
 static struct search *search_alloc(struct bio *bio, struct bcache_device *d)
 {
+	struct search *s;
 	struct bio_vec *bv;
-	struct search *s = mempool_alloc(d->c->search, GFP_NOIO);
-	memset(s, 0, offsetof(struct search, op.keys));
+
+	s = mempool_alloc(d->c->search, GFP_NOIO);
+	memset(s, 0, offsetof(struct search, iop.insert_keys));
 
 	__closure_init(&s->cl, NULL);
 
-	s->op.inode		= d->id;
-	s->op.c			= d->c;
+	s->iop.inode		= d->id;
+	s->iop.c		= d->c;
 	s->d			= d;
 	s->op.lock		= -1;
-	s->task			= current;
+	s->iop.write_point	= hash_long((unsigned long) current, 16);
 	s->orig_bio		= bio;
 	s->write		= (bio->bi_rw & REQ_WRITE) != 0;
-	s->op.flush_journal	= (bio->bi_rw & (REQ_FLUSH|REQ_FUA)) != 0;
-	s->op.skip		= (bio->bi_rw & REQ_DISCARD) != 0;
+	s->iop.flush_journal	= (bio->bi_rw & (REQ_FLUSH|REQ_FUA)) != 0;
 	s->recoverable		= 1;
 	s->start_time		= jiffies;
 	do_bio_hook(s);
@@ -734,18 +815,6 @@ static struct search *search_alloc(struct bio *bio, struct bcache_device *d)
 	return s;
 }
 
-static void btree_read_async(struct closure *cl)
-{
-	struct btree_op *op = container_of(cl, struct btree_op, cl);
-
-	int ret = btree_root(search_recurse, op->c, op);
-
-	if (ret == -EAGAIN)
-		continue_at(cl, btree_read_async, bcache_wq);
-
-	closure_return(cl);
-}
-
 /* Cached devices */
 
 static void cached_dev_bio_complete(struct closure *cl)
@@ -759,27 +828,28 @@ static void cached_dev_bio_complete(struct closure *cl)
 
 /* Process reads */
 
-static void cached_dev_read_complete(struct closure *cl)
+static void cached_dev_cache_miss_done(struct closure *cl)
 {
 	struct search *s = container_of(cl, struct search, cl);
 
-	if (s->op.insert_collision)
-		bch_mark_cache_miss_collision(s);
+	if (s->iop.replace_collision)
+		bch_mark_cache_miss_collision(s->iop.c, s->d);
 
-	if (s->op.cache_bio) {
+	if (s->iop.bio) {
 		int i;
 		struct bio_vec *bv;
 
-		__bio_for_each_segment(bv, s->op.cache_bio, i, 0)
+		bio_for_each_segment_all(bv, s->iop.bio, i)
 			__free_page(bv->bv_page);
 	}
 
 	cached_dev_bio_complete(cl);
 }
 
-static void request_read_error(struct closure *cl)
+static void cached_dev_read_error(struct closure *cl)
 {
 	struct search *s = container_of(cl, struct search, cl);
+	struct bio *bio = &s->bio.bio;
 	struct bio_vec *bv;
 	int i;
 
@@ -787,7 +857,7 @@ static void request_read_error(struct closure *cl)
 		/* Retry from the backing device: */
 		trace_bcache_read_retry(s->orig_bio);
 
-		s->error = 0;
+		s->iop.error = 0;
 		bv = s->bio.bio.bi_io_vec;
 		do_bio_hook(s);
 		s->bio.bio.bi_io_vec = bv;
@@ -803,146 +873,148 @@ static void request_read_error(struct closure *cl)
 
 		/* XXX: invalidate cache */
 
-		closure_bio_submit(&s->bio.bio, &s->cl, s->d);
+		closure_bio_submit(bio, cl, s->d);
 	}
 
-	continue_at(cl, cached_dev_read_complete, NULL);
+	continue_at(cl, cached_dev_cache_miss_done, NULL);
 }
 
-static void request_read_done(struct closure *cl)
+static void cached_dev_read_done(struct closure *cl)
 {
 	struct search *s = container_of(cl, struct search, cl);
 	struct cached_dev *dc = container_of(s->d, struct cached_dev, disk);
 
 	/*
-	 * s->cache_bio != NULL implies that we had a cache miss; cache_bio now
-	 * contains data ready to be inserted into the cache.
+	 * We had a cache miss; cache_bio now contains data ready to be inserted
+	 * into the cache.
 	 *
 	 * First, we copy the data we just read from cache_bio's bounce buffers
 	 * to the buffers the original bio pointed to:
 	 */
 
-	if (s->op.cache_bio) {
-		bio_reset(s->op.cache_bio);
-		s->op.cache_bio->bi_sector	= s->cache_miss->bi_sector;
-		s->op.cache_bio->bi_bdev	= s->cache_miss->bi_bdev;
-		s->op.cache_bio->bi_size	= s->cache_bio_sectors << 9;
-		bch_bio_map(s->op.cache_bio, NULL);
+	if (s->iop.bio) {
+		bio_reset(s->iop.bio);
+		s->iop.bio->bi_sector = s->cache_miss->bi_sector;
+		s->iop.bio->bi_bdev = s->cache_miss->bi_bdev;
+		s->iop.bio->bi_size = s->insert_bio_sectors << 9;
+		bch_bio_map(s->iop.bio, NULL);
 
-		bio_copy_data(s->cache_miss, s->op.cache_bio);
+		bio_copy_data(s->cache_miss, s->iop.bio);
 
 		bio_put(s->cache_miss);
 		s->cache_miss = NULL;
 	}
 
-	if (verify(dc, &s->bio.bio) && s->recoverable)
-		bch_data_verify(s);
+	if (verify(dc, &s->bio.bio) && s->recoverable &&
+	    !s->unaligned_bvec && !s->read_dirty_data)
+		bch_data_verify(dc, s->orig_bio);
 
 	bio_complete(s);
 
-	if (s->op.cache_bio &&
-	    !test_bit(CACHE_SET_STOPPING, &s->op.c->flags)) {
-		s->op.type = BTREE_REPLACE;
-		closure_call(&s->op.cl, bch_insert_data, NULL, cl);
+	if (s->iop.bio &&
+	    !test_bit(CACHE_SET_STOPPING, &s->iop.c->flags)) {
+		BUG_ON(!s->iop.replace);
+		closure_call(&s->iop.cl, bch_data_insert, NULL, cl);
 	}
 
-	continue_at(cl, cached_dev_read_complete, NULL);
+	continue_at(cl, cached_dev_cache_miss_done, NULL);
 }
 
-static void request_read_done_bh(struct closure *cl)
+static void cached_dev_read_done_bh(struct closure *cl)
 {
 	struct search *s = container_of(cl, struct search, cl);
 	struct cached_dev *dc = container_of(s->d, struct cached_dev, disk);
 
-	bch_mark_cache_accounting(s, !s->cache_miss, s->op.skip);
-	trace_bcache_read(s->orig_bio, !s->cache_miss, s->op.skip);
+	bch_mark_cache_accounting(s->iop.c, s->d,
+				  !s->cache_miss, s->iop.bypass);
+	trace_bcache_read(s->orig_bio, !s->cache_miss, s->iop.bypass);
 
-	if (s->error)
-		continue_at_nobarrier(cl, request_read_error, bcache_wq);
-	else if (s->op.cache_bio || verify(dc, &s->bio.bio))
-		continue_at_nobarrier(cl, request_read_done, bcache_wq);
+	if (s->iop.error)
+		continue_at_nobarrier(cl, cached_dev_read_error, bcache_wq);
+	else if (s->iop.bio || verify(dc, &s->bio.bio))
+		continue_at_nobarrier(cl, cached_dev_read_done, bcache_wq);
 	else
-		continue_at_nobarrier(cl, cached_dev_read_complete, NULL);
+		continue_at_nobarrier(cl, cached_dev_bio_complete, NULL);
 }
 
 static int cached_dev_cache_miss(struct btree *b, struct search *s,
 				 struct bio *bio, unsigned sectors)
 {
-	int ret = 0;
-	unsigned reada;
+	int ret = MAP_CONTINUE;
+	unsigned reada = 0;
 	struct cached_dev *dc = container_of(s->d, struct cached_dev, disk);
-	struct bio *miss;
-
-	miss = bch_bio_split(bio, sectors, GFP_NOIO, s->d->bio_split);
-	if (miss == bio)
-		s->op.lookup_done = true;
+	struct bio *miss, *cache_bio;
 
-	miss->bi_end_io		= request_endio;
-	miss->bi_private	= &s->cl;
-
-	if (s->cache_miss || s->op.skip)
+	if (s->cache_miss || s->iop.bypass) {
+		miss = bch_bio_split(bio, sectors, GFP_NOIO, s->d->bio_split);
+		ret = miss == bio ? MAP_DONE : MAP_CONTINUE;
 		goto out_submit;
-
-	if (miss != bio ||
-	    (bio->bi_rw & REQ_RAHEAD) ||
-	    (bio->bi_rw & REQ_META) ||
-	    s->op.c->gc_stats.in_use >= CUTOFF_CACHE_READA)
-		reada = 0;
-	else {
-		reada = min(dc->readahead >> 9,
-			    sectors - bio_sectors(miss));
-
-		if (bio_end_sector(miss) + reada > bdev_sectors(miss->bi_bdev))
-			reada = bdev_sectors(miss->bi_bdev) -
-				bio_end_sector(miss);
 	}
 
-	s->cache_bio_sectors = bio_sectors(miss) + reada;
-	s->op.cache_bio = bio_alloc_bioset(GFP_NOWAIT,
-			DIV_ROUND_UP(s->cache_bio_sectors, PAGE_SECTORS),
-			dc->disk.bio_split);
+	if (!(bio->bi_rw & REQ_RAHEAD) &&
+	    !(bio->bi_rw & REQ_META) &&
+	    s->iop.c->gc_stats.in_use < CUTOFF_CACHE_READA)
+		reada = min_t(sector_t, dc->readahead >> 9,
+			      bdev_sectors(bio->bi_bdev) - bio_end_sector(bio));
 
-	if (!s->op.cache_bio)
-		goto out_submit;
+	s->insert_bio_sectors = min(sectors, bio_sectors(bio) + reada);
 
-	s->op.cache_bio->bi_sector	= miss->bi_sector;
-	s->op.cache_bio->bi_bdev	= miss->bi_bdev;
-	s->op.cache_bio->bi_size	= s->cache_bio_sectors << 9;
+	s->iop.replace_key = KEY(s->iop.inode,
+				 bio->bi_sector + s->insert_bio_sectors,
+				 s->insert_bio_sectors);
 
-	s->op.cache_bio->bi_end_io	= request_endio;
-	s->op.cache_bio->bi_private	= &s->cl;
+	ret = bch_btree_insert_check_key(b, &s->op, &s->iop.replace_key);
+	if (ret)
+		return ret;
+
+	s->iop.replace = true;
+
+	miss = bch_bio_split(bio, sectors, GFP_NOIO, s->d->bio_split);
 
 	/* btree_search_recurse()'s btree iterator is no good anymore */
-	ret = -EINTR;
-	if (!bch_btree_insert_check_key(b, &s->op, s->op.cache_bio))
-		goto out_put;
+	ret = miss == bio ? MAP_DONE : -EINTR;
+
+	cache_bio = bio_alloc_bioset(GFP_NOWAIT,
+			DIV_ROUND_UP(s->insert_bio_sectors, PAGE_SECTORS),
+			dc->disk.bio_split);
+	if (!cache_bio)
+		goto out_submit;
+
+	cache_bio->bi_sector	= miss->bi_sector;
+	cache_bio->bi_bdev	= miss->bi_bdev;
+	cache_bio->bi_size	= s->insert_bio_sectors << 9;
+
+	cache_bio->bi_end_io	= request_endio;
+	cache_bio->bi_private	= &s->cl;
 
-	bch_bio_map(s->op.cache_bio, NULL);
-	if (bio_alloc_pages(s->op.cache_bio, __GFP_NOWARN|GFP_NOIO))
+	bch_bio_map(cache_bio, NULL);
+	if (bio_alloc_pages(cache_bio, __GFP_NOWARN|GFP_NOIO))
 		goto out_put;
 
-	s->cache_miss = miss;
-	bio_get(s->op.cache_bio);
+	if (reada)
+		bch_mark_cache_readahead(s->iop.c, s->d);
 
-	closure_bio_submit(s->op.cache_bio, &s->cl, s->d);
+	s->cache_miss	= miss;
+	s->iop.bio	= cache_bio;
+	bio_get(cache_bio);
+	closure_bio_submit(cache_bio, &s->cl, s->d);
 
 	return ret;
 out_put:
-	bio_put(s->op.cache_bio);
-	s->op.cache_bio = NULL;
+	bio_put(cache_bio);
 out_submit:
+	miss->bi_end_io		= request_endio;
+	miss->bi_private	= &s->cl;
 	closure_bio_submit(miss, &s->cl, s->d);
 	return ret;
 }
 
-static void request_read(struct cached_dev *dc, struct search *s)
+static void cached_dev_read(struct cached_dev *dc, struct search *s)
 {
 	struct closure *cl = &s->cl;
 
-	check_should_skip(dc, s);
-	closure_call(&s->op.cl, btree_read_async, NULL, cl);
-
-	continue_at(cl, request_read_done_bh, NULL);
+	closure_call(&s->iop.cl, cache_lookup, NULL, cl);
+	continue_at(cl, cached_dev_read_done_bh, NULL);
 }
 
 /* Process writes */
@@ -956,47 +1028,52 @@ static void cached_dev_write_complete(struct closure *cl)
 	cached_dev_bio_complete(cl);
 }
 
-static void request_write(struct cached_dev *dc, struct search *s)
+static void cached_dev_write(struct cached_dev *dc, struct search *s)
 {
 	struct closure *cl = &s->cl;
 	struct bio *bio = &s->bio.bio;
-	struct bkey start, end;
-	start = KEY(dc->disk.id, bio->bi_sector, 0);
-	end = KEY(dc->disk.id, bio_end_sector(bio), 0);
+	struct bkey start = KEY(dc->disk.id, bio->bi_sector, 0);
+	struct bkey end = KEY(dc->disk.id, bio_end_sector(bio), 0);
 
-	bch_keybuf_check_overlapping(&s->op.c->moving_gc_keys, &start, &end);
+	bch_keybuf_check_overlapping(&s->iop.c->moving_gc_keys, &start, &end);
 
-	check_should_skip(dc, s);
 	down_read_non_owner(&dc->writeback_lock);
-
 	if (bch_keybuf_check_overlapping(&dc->writeback_keys, &start, &end)) {
-		s->op.skip	= false;
-		s->writeback	= true;
+		/*
+		 * We overlap with some dirty data undergoing background
+		 * writeback, force this write to writeback
+		 */
+		s->iop.bypass = false;
+		s->iop.writeback = true;
 	}
 
+	/*
+	 * Discards aren't _required_ to do anything, so skipping if
+	 * check_overlapping returned true is ok
+	 *
+	 * But check_overlapping drops dirty keys for which io hasn't started,
+	 * so we still want to call it.
+	 */
 	if (bio->bi_rw & REQ_DISCARD)
-		goto skip;
+		s->iop.bypass = true;
 
 	if (should_writeback(dc, s->orig_bio,
 			     cache_mode(dc, bio),
-			     s->op.skip)) {
-		s->op.skip = false;
-		s->writeback = true;
+			     s->iop.bypass)) {
+		s->iop.bypass = false;
+		s->iop.writeback = true;
 	}
 
-	if (s->op.skip)
-		goto skip;
-
-	trace_bcache_write(s->orig_bio, s->writeback, s->op.skip);
+	if (s->iop.bypass) {
+		s->iop.bio = s->orig_bio;
+		bio_get(s->iop.bio);
 
-	if (!s->writeback) {
-		s->op.cache_bio = bio_clone_bioset(bio, GFP_NOIO,
-						   dc->disk.bio_split);
-
-		closure_bio_submit(bio, cl, s->d);
-	} else {
+		if (!(bio->bi_rw & REQ_DISCARD) ||
+		    blk_queue_discard(bdev_get_queue(dc->bdev)))
+			closure_bio_submit(bio, cl, s->d);
+	} else if (s->iop.writeback) {
 		bch_writeback_add(dc);
-		s->op.cache_bio = bio;
+		s->iop.bio = bio;
 
 		if (bio->bi_rw & REQ_FLUSH) {
 			/* Also need to send a flush to the backing device */
@@ -1010,36 +1087,26 @@ static void request_write(struct cached_dev *dc, struct search *s)
 
 			closure_bio_submit(flush, cl, s->d);
 		}
-	}
-out:
-	closure_call(&s->op.cl, bch_insert_data, NULL, cl);
-	continue_at(cl, cached_dev_write_complete, NULL);
-skip:
-	s->op.skip = true;
-	s->op.cache_bio = s->orig_bio;
-	bio_get(s->op.cache_bio);
+	} else {
+		s->iop.bio = bio_clone_bioset(bio, GFP_NOIO,
+					      dc->disk.bio_split);
 
-	if ((bio->bi_rw & REQ_DISCARD) &&
-	    !blk_queue_discard(bdev_get_queue(dc->bdev)))
-		goto out;
+		closure_bio_submit(bio, cl, s->d);
+	}
 
-	closure_bio_submit(bio, cl, s->d);
-	goto out;
+	closure_call(&s->iop.cl, bch_data_insert, NULL, cl);
+	continue_at(cl, cached_dev_write_complete, NULL);
 }
 
-static void request_nodata(struct cached_dev *dc, struct search *s)
+static void cached_dev_nodata(struct closure *cl)
 {
-	struct closure *cl = &s->cl;
+	struct search *s = container_of(cl, struct search, cl);
 	struct bio *bio = &s->bio.bio;
 
-	if (bio->bi_rw & REQ_DISCARD) {
-		request_write(dc, s);
-		return;
-	}
-
-	if (s->op.flush_journal)
-		bch_journal_meta(s->op.c, cl);
+	if (s->iop.flush_journal)
+		bch_journal_meta(s->iop.c, cl);
 
+	/* If it's a flush, we send the flush to the backing device too */
 	closure_bio_submit(bio, cl, s->d);
 
 	continue_at(cl, cached_dev_bio_complete, NULL);
@@ -1047,134 +1114,6 @@ static void request_nodata(struct cached_dev *dc, struct search *s)
 
 /* Cached devices - read & write stuff */
 
-unsigned bch_get_congested(struct cache_set *c)
-{
-	int i;
-	long rand;
-
-	if (!c->congested_read_threshold_us &&
-	    !c->congested_write_threshold_us)
-		return 0;
-
-	i = (local_clock_us() - c->congested_last_us) / 1024;
-	if (i < 0)
-		return 0;
-
-	i += atomic_read(&c->congested);
-	if (i >= 0)
-		return 0;
-
-	i += CONGESTED_MAX;
-
-	if (i > 0)
-		i = fract_exp_two(i, 6);
-
-	rand = get_random_int();
-	i -= bitmap_weight(&rand, BITS_PER_LONG);
-
-	return i > 0 ? i : 1;
-}
-
-static void add_sequential(struct task_struct *t)
-{
-	ewma_add(t->sequential_io_avg,
-		 t->sequential_io, 8, 0);
-
-	t->sequential_io = 0;
-}
-
-static struct hlist_head *iohash(struct cached_dev *dc, uint64_t k)
-{
-	return &dc->io_hash[hash_64(k, RECENT_IO_BITS)];
-}
-
-static void check_should_skip(struct cached_dev *dc, struct search *s)
-{
-	struct cache_set *c = s->op.c;
-	struct bio *bio = &s->bio.bio;
-	unsigned mode = cache_mode(dc, bio);
-	unsigned sectors, congested = bch_get_congested(c);
-
-	if (atomic_read(&dc->disk.detaching) ||
-	    c->gc_stats.in_use > CUTOFF_CACHE_ADD ||
-	    (bio->bi_rw & REQ_DISCARD))
-		goto skip;
-
-	if (mode == CACHE_MODE_NONE ||
-	    (mode == CACHE_MODE_WRITEAROUND &&
-	     (bio->bi_rw & REQ_WRITE)))
-		goto skip;
-
-	if (bio->bi_sector   & (c->sb.block_size - 1) ||
-	    bio_sectors(bio) & (c->sb.block_size - 1)) {
-		pr_debug("skipping unaligned io");
-		goto skip;
-	}
-
-	if (!congested && !dc->sequential_cutoff)
-		goto rescale;
-
-	if (!congested &&
-	    mode == CACHE_MODE_WRITEBACK &&
-	    (bio->bi_rw & REQ_WRITE) &&
-	    (bio->bi_rw & REQ_SYNC))
-		goto rescale;
-
-	if (dc->sequential_merge) {
-		struct io *i;
-
-		spin_lock(&dc->io_lock);
-
-		hlist_for_each_entry(i, iohash(dc, bio->bi_sector), hash)
-			if (i->last == bio->bi_sector &&
-			    time_before(jiffies, i->jiffies))
-				goto found;
-
-		i = list_first_entry(&dc->io_lru, struct io, lru);
-
-		add_sequential(s->task);
-		i->sequential = 0;
-found:
-		if (i->sequential + bio->bi_size > i->sequential)
-			i->sequential	+= bio->bi_size;
-
-		i->last			 = bio_end_sector(bio);
-		i->jiffies		 = jiffies + msecs_to_jiffies(5000);
-		s->task->sequential_io	 = i->sequential;
-
-		hlist_del(&i->hash);
-		hlist_add_head(&i->hash, iohash(dc, i->last));
-		list_move_tail(&i->lru, &dc->io_lru);
-
-		spin_unlock(&dc->io_lock);
-	} else {
-		s->task->sequential_io = bio->bi_size;
-
-		add_sequential(s->task);
-	}
-
-	sectors = max(s->task->sequential_io,
-		      s->task->sequential_io_avg) >> 9;
-
-	if (dc->sequential_cutoff &&
-	    sectors >= dc->sequential_cutoff >> 9) {
-		trace_bcache_bypass_sequential(s->orig_bio);
-		goto skip;
-	}
-
-	if (congested && sectors >= congested) {
-		trace_bcache_bypass_congested(s->orig_bio);
-		goto skip;
-	}
-
-rescale:
-	bch_rescale_priorities(c, bio_sectors(bio));
-	return;
-skip:
-	bch_mark_sectors_bypassed(s, bio_sectors(bio));
-	s->op.skip = true;
-}
-
 static void cached_dev_make_request(struct request_queue *q, struct bio *bio)
 {
 	struct search *s;
@@ -1192,14 +1131,24 @@ static void cached_dev_make_request(struct request_queue *q, struct bio *bio)
 
 	if (cached_dev_get(dc)) {
 		s = search_alloc(bio, d);
-		trace_bcache_request_start(s, bio);
-
-		if (!bio_has_data(bio))
-			request_nodata(dc, s);
-		else if (rw)
-			request_write(dc, s);
-		else
-			request_read(dc, s);
+		trace_bcache_request_start(s->d, bio);
+
+		if (!bio->bi_size) {
+			/*
+			 * can't call bch_journal_meta from under
+			 * generic_make_request
+			 */
+			continue_at_nobarrier(&s->cl,
+					      cached_dev_nodata,
+					      bcache_wq);
+		} else {
+			s->iop.bypass = check_should_bypass(dc, bio);
+
+			if (rw)
+				cached_dev_write(dc, s);
+			else
+				cached_dev_read(dc, s);
+		}
 	} else {
 		if ((bio->bi_rw & REQ_DISCARD) &&
 		    !blk_queue_discard(bdev_get_queue(dc->bdev)))
@@ -1274,9 +1223,19 @@ static int flash_dev_cache_miss(struct btree *b, struct search *s,
 	bio_advance(bio, min(sectors << 9, bio->bi_size));
 
 	if (!bio->bi_size)
-		s->op.lookup_done = true;
+		return MAP_DONE;
 
-	return 0;
+	return MAP_CONTINUE;
+}
+
+static void flash_dev_nodata(struct closure *cl)
+{
+	struct search *s = container_of(cl, struct search, cl);
+
+	if (s->iop.flush_journal)
+		bch_journal_meta(s->iop.c, cl);
+
+	continue_at(cl, search_free, NULL);
 }
 
 static void flash_dev_make_request(struct request_queue *q, struct bio *bio)
@@ -1295,23 +1254,28 @@ static void flash_dev_make_request(struct request_queue *q, struct bio *bio)
 	cl = &s->cl;
 	bio = &s->bio.bio;
 
-	trace_bcache_request_start(s, bio);
+	trace_bcache_request_start(s->d, bio);
 
-	if (bio_has_data(bio) && !rw) {
-		closure_call(&s->op.cl, btree_read_async, NULL, cl);
-	} else if (bio_has_data(bio) || s->op.skip) {
-		bch_keybuf_check_overlapping(&s->op.c->moving_gc_keys,
+	if (!bio->bi_size) {
+		/*
+		 * can't call bch_journal_meta from under
+		 * generic_make_request
+		 */
+		continue_at_nobarrier(&s->cl,
+				      flash_dev_nodata,
+				      bcache_wq);
+	} else if (rw) {
+		bch_keybuf_check_overlapping(&s->iop.c->moving_gc_keys,
 					&KEY(d->id, bio->bi_sector, 0),
 					&KEY(d->id, bio_end_sector(bio), 0));
 
-		s->writeback	= true;
-		s->op.cache_bio	= bio;
+		s->iop.bypass		= (bio->bi_rw & REQ_DISCARD) != 0;
+		s->iop.writeback	= true;
+		s->iop.bio		= bio;
 
-		closure_call(&s->op.cl, bch_insert_data, NULL, cl);
+		closure_call(&s->iop.cl, bch_data_insert, NULL, cl);
 	} else {
-		/* No data - probably a cache flush */
-		if (s->op.flush_journal)
-			bch_journal_meta(s->op.c, cl);
+		closure_call(&s->iop.cl, cache_lookup, NULL, cl);
 	}
 
 	continue_at(cl, search_free, NULL);
diff --git a/drivers/md/bcache/request.h b/drivers/md/bcache/request.h
index 57dc4784f4f4..2cd65bf073c2 100644
--- a/drivers/md/bcache/request.h
+++ b/drivers/md/bcache/request.h
@@ -3,40 +3,33 @@
 
 #include <linux/cgroup.h>
 
-struct search {
-	/* Stack frame for bio_complete */
+struct data_insert_op {
 	struct closure		cl;
+	struct cache_set	*c;
+	struct bio		*bio;
 
-	struct bcache_device	*d;
-	struct task_struct	*task;
-
-	struct bbio		bio;
-	struct bio		*orig_bio;
-	struct bio		*cache_miss;
-	unsigned		cache_bio_sectors;
-
-	unsigned		recoverable:1;
-	unsigned		unaligned_bvec:1;
+	unsigned		inode;
+	uint16_t		write_point;
+	uint16_t		write_prio;
+	short			error;
 
-	unsigned		write:1;
+	unsigned		bypass:1;
 	unsigned		writeback:1;
+	unsigned		flush_journal:1;
+	unsigned		csum:1;
 
-	/* IO error returned to s->bio */
-	short			error;
-	unsigned long		start_time;
+	unsigned		replace:1;
+	unsigned		replace_collision:1;
+
+	unsigned		insert_data_done:1;
 
-	/* Anything past op->keys won't get zeroed in do_bio_hook */
-	struct btree_op		op;
+	/* Anything past this point won't get zeroed in search_alloc() */
+	struct keylist		insert_keys;
+	BKEY_PADDED(replace_key);
 };
 
-void bch_cache_read_endio(struct bio *, int);
 unsigned bch_get_congested(struct cache_set *);
-void bch_insert_data(struct closure *cl);
-void bch_btree_insert_async(struct closure *);
-void bch_cache_read_endio(struct bio *, int);
-
-void bch_open_buckets_free(struct cache_set *);
-int bch_open_buckets_alloc(struct cache_set *);
+void bch_data_insert(struct closure *cl);
 
 void bch_cached_dev_request_init(struct cached_dev *dc);
 void bch_flash_dev_request_init(struct bcache_device *d);
diff --git a/drivers/md/bcache/stats.c b/drivers/md/bcache/stats.c
index b8730e714d69..84d0782f702e 100644
--- a/drivers/md/bcache/stats.c
+++ b/drivers/md/bcache/stats.c
@@ -7,7 +7,6 @@
 #include "bcache.h"
 #include "stats.h"
 #include "btree.h"
-#include "request.h"
 #include "sysfs.h"
 
 /*
@@ -196,35 +195,36 @@ static void mark_cache_stats(struct cache_stat_collector *stats,
 			atomic_inc(&stats->cache_bypass_misses);
 }
 
-void bch_mark_cache_accounting(struct search *s, bool hit, bool bypass)
+void bch_mark_cache_accounting(struct cache_set *c, struct bcache_device *d,
+			       bool hit, bool bypass)
 {
-	struct cached_dev *dc = container_of(s->d, struct cached_dev, disk);
+	struct cached_dev *dc = container_of(d, struct cached_dev, disk);
 	mark_cache_stats(&dc->accounting.collector, hit, bypass);
-	mark_cache_stats(&s->op.c->accounting.collector, hit, bypass);
+	mark_cache_stats(&c->accounting.collector, hit, bypass);
 #ifdef CONFIG_CGROUP_BCACHE
 	mark_cache_stats(&(bch_bio_to_cgroup(s->orig_bio)->stats), hit, bypass);
 #endif
 }
 
-void bch_mark_cache_readahead(struct search *s)
+void bch_mark_cache_readahead(struct cache_set *c, struct bcache_device *d)
 {
-	struct cached_dev *dc = container_of(s->d, struct cached_dev, disk);
+	struct cached_dev *dc = container_of(d, struct cached_dev, disk);
 	atomic_inc(&dc->accounting.collector.cache_readaheads);
-	atomic_inc(&s->op.c->accounting.collector.cache_readaheads);
+	atomic_inc(&c->accounting.collector.cache_readaheads);
 }
 
-void bch_mark_cache_miss_collision(struct search *s)
+void bch_mark_cache_miss_collision(struct cache_set *c, struct bcache_device *d)
 {
-	struct cached_dev *dc = container_of(s->d, struct cached_dev, disk);
+	struct cached_dev *dc = container_of(d, struct cached_dev, disk);
 	atomic_inc(&dc->accounting.collector.cache_miss_collisions);
-	atomic_inc(&s->op.c->accounting.collector.cache_miss_collisions);
+	atomic_inc(&c->accounting.collector.cache_miss_collisions);
 }
 
-void bch_mark_sectors_bypassed(struct search *s, int sectors)
+void bch_mark_sectors_bypassed(struct cache_set *c, struct cached_dev *dc,
+			       int sectors)
 {
-	struct cached_dev *dc = container_of(s->d, struct cached_dev, disk);
 	atomic_add(sectors, &dc->accounting.collector.sectors_bypassed);
-	atomic_add(sectors, &s->op.c->accounting.collector.sectors_bypassed);
+	atomic_add(sectors, &c->accounting.collector.sectors_bypassed);
 }
 
 void bch_cache_accounting_init(struct cache_accounting *acc,
diff --git a/drivers/md/bcache/stats.h b/drivers/md/bcache/stats.h
index c7c7a8fd29fe..adbff141c887 100644
--- a/drivers/md/bcache/stats.h
+++ b/drivers/md/bcache/stats.h
@@ -38,7 +38,9 @@ struct cache_accounting {
 	struct cache_stats day;
 };
 
-struct search;
+struct cache_set;
+struct cached_dev;
+struct bcache_device;
 
 void bch_cache_accounting_init(struct cache_accounting *acc,
 			       struct closure *parent);
@@ -50,9 +52,10 @@ void bch_cache_accounting_clear(struct cache_accounting *acc);
 
 void bch_cache_accounting_destroy(struct cache_accounting *acc);
 
-void bch_mark_cache_accounting(struct search *s, bool hit, bool bypass);
-void bch_mark_cache_readahead(struct search *s);
-void bch_mark_cache_miss_collision(struct search *s);
-void bch_mark_sectors_bypassed(struct search *s, int sectors);
+void bch_mark_cache_accounting(struct cache_set *, struct bcache_device *,
+			       bool, bool);
+void bch_mark_cache_readahead(struct cache_set *, struct bcache_device *);
+void bch_mark_cache_miss_collision(struct cache_set *, struct bcache_device *);
+void bch_mark_sectors_bypassed(struct cache_set *, struct cached_dev *, int);
 
 #endif /* _BCACHE_STATS_H_ */
diff --git a/drivers/md/bcache/super.c b/drivers/md/bcache/super.c
index 547c4c57b052..dec15cd2d797 100644
--- a/drivers/md/bcache/super.c
+++ b/drivers/md/bcache/super.c
@@ -16,6 +16,7 @@
 #include <linux/buffer_head.h>
 #include <linux/debugfs.h>
 #include <linux/genhd.h>
+#include <linux/idr.h>
 #include <linux/kthread.h>
 #include <linux/module.h>
 #include <linux/random.h>
@@ -45,21 +46,13 @@ const char * const bch_cache_modes[] = {
 	NULL
 };
 
-struct uuid_entry_v0 {
-	uint8_t		uuid[16];
-	uint8_t		label[32];
-	uint32_t	first_reg;
-	uint32_t	last_reg;
-	uint32_t	invalidated;
-	uint32_t	pad;
-};
-
 static struct kobject *bcache_kobj;
 struct mutex bch_register_lock;
 LIST_HEAD(bch_cache_sets);
 static LIST_HEAD(uncached_devices);
 
-static int bcache_major, bcache_minor;
+static int bcache_major;
+static DEFINE_IDA(bcache_minor);
 static wait_queue_head_t unregister_wait;
 struct workqueue_struct *bcache_wq;
 
@@ -382,7 +375,7 @@ static char *uuid_read(struct cache_set *c, struct jset *j, struct closure *cl)
 {
 	struct bkey *k = &j->uuid_bucket;
 
-	if (__bch_ptr_invalid(c, 1, k))
+	if (bch_btree_ptr_invalid(c, k))
 		return "bad uuid pointer";
 
 	bkey_copy(&c->uuid_bucket, k);
@@ -427,7 +420,7 @@ static int __uuid_write(struct cache_set *c)
 
 	lockdep_assert_held(&bch_register_lock);
 
-	if (bch_bucket_alloc_set(c, WATERMARK_METADATA, &k.key, 1, &cl))
+	if (bch_bucket_alloc_set(c, WATERMARK_METADATA, &k.key, 1, true))
 		return 1;
 
 	SET_KEY_SIZE(&k.key, c->sb.bucket_size);
@@ -435,7 +428,7 @@ static int __uuid_write(struct cache_set *c)
 	closure_sync(&cl);
 
 	bkey_copy(&c->uuid_bucket, &k.key);
-	__bkey_put(c, &k.key);
+	bkey_put(c, &k.key);
 	return 0;
 }
 
@@ -562,10 +555,10 @@ void bch_prio_write(struct cache *ca)
 		}
 
 		p->next_bucket	= ca->prio_buckets[i + 1];
-		p->magic	= pset_magic(ca);
+		p->magic	= pset_magic(&ca->sb);
 		p->csum		= bch_crc64(&p->magic, bucket_bytes(ca) - 8);
 
-		bucket = bch_bucket_alloc(ca, WATERMARK_PRIO, &cl);
+		bucket = bch_bucket_alloc(ca, WATERMARK_PRIO, true);
 		BUG_ON(bucket == -1);
 
 		mutex_unlock(&ca->set->bucket_lock);
@@ -613,7 +606,7 @@ static void prio_read(struct cache *ca, uint64_t bucket)
 			if (p->csum != bch_crc64(&p->magic, bucket_bytes(ca) - 8))
 				pr_warn("bad csum reading priorities");
 
-			if (p->magic != pset_magic(ca))
+			if (p->magic != pset_magic(&ca->sb))
 				pr_warn("bad magic reading priorities");
 
 			bucket = p->next_bucket;
@@ -630,7 +623,7 @@ static void prio_read(struct cache *ca, uint64_t bucket)
 static int open_dev(struct block_device *b, fmode_t mode)
 {
 	struct bcache_device *d = b->bd_disk->private_data;
-	if (atomic_read(&d->closing))
+	if (test_bit(BCACHE_DEV_CLOSING, &d->flags))
 		return -ENXIO;
 
 	closure_get(&d->cl);
@@ -659,20 +652,24 @@ static const struct block_device_operations bcache_ops = {
 
 void bcache_device_stop(struct bcache_device *d)
 {
-	if (!atomic_xchg(&d->closing, 1))
+	if (!test_and_set_bit(BCACHE_DEV_CLOSING, &d->flags))
 		closure_queue(&d->cl);
 }
 
 static void bcache_device_unlink(struct bcache_device *d)
 {
-	unsigned i;
-	struct cache *ca;
+	lockdep_assert_held(&bch_register_lock);
 
-	sysfs_remove_link(&d->c->kobj, d->name);
-	sysfs_remove_link(&d->kobj, "cache");
+	if (d->c && !test_and_set_bit(BCACHE_DEV_UNLINK_DONE, &d->flags)) {
+		unsigned i;
+		struct cache *ca;
 
-	for_each_cache(ca, d->c, i)
-		bd_unlink_disk_holder(ca->bdev, d->disk);
+		sysfs_remove_link(&d->c->kobj, d->name);
+		sysfs_remove_link(&d->kobj, "cache");
+
+		for_each_cache(ca, d->c, i)
+			bd_unlink_disk_holder(ca->bdev, d->disk);
+	}
 }
 
 static void bcache_device_link(struct bcache_device *d, struct cache_set *c,
@@ -696,19 +693,16 @@ static void bcache_device_detach(struct bcache_device *d)
 {
 	lockdep_assert_held(&bch_register_lock);
 
-	if (atomic_read(&d->detaching)) {
+	if (test_bit(BCACHE_DEV_DETACHING, &d->flags)) {
 		struct uuid_entry *u = d->c->uuids + d->id;
 
 		SET_UUID_FLASH_ONLY(u, 0);
 		memcpy(u->uuid, invalid_uuid, 16);
 		u->invalidated = cpu_to_le32(get_seconds());
 		bch_uuid_write(d->c);
-
-		atomic_set(&d->detaching, 0);
 	}
 
-	if (!d->flush_done)
-		bcache_device_unlink(d);
+	bcache_device_unlink(d);
 
 	d->c->devices[d->id] = NULL;
 	closure_put(&d->c->caching);
@@ -739,14 +733,20 @@ static void bcache_device_free(struct bcache_device *d)
 		del_gendisk(d->disk);
 	if (d->disk && d->disk->queue)
 		blk_cleanup_queue(d->disk->queue);
-	if (d->disk)
+	if (d->disk) {
+		ida_simple_remove(&bcache_minor, d->disk->first_minor);
 		put_disk(d->disk);
+	}
 
 	bio_split_pool_free(&d->bio_split_hook);
 	if (d->unaligned_bvec)
 		mempool_destroy(d->unaligned_bvec);
 	if (d->bio_split)
 		bioset_free(d->bio_split);
+	if (is_vmalloc_addr(d->full_dirty_stripes))
+		vfree(d->full_dirty_stripes);
+	else
+		kfree(d->full_dirty_stripes);
 	if (is_vmalloc_addr(d->stripe_sectors_dirty))
 		vfree(d->stripe_sectors_dirty);
 	else
@@ -760,15 +760,19 @@ static int bcache_device_init(struct bcache_device *d, unsigned block_size,
 {
 	struct request_queue *q;
 	size_t n;
+	int minor;
 
-	if (!d->stripe_size_bits)
-		d->stripe_size_bits = 31;
+	if (!d->stripe_size)
+		d->stripe_size = 1 << 31;
 
-	d->nr_stripes = round_up(sectors, 1 << d->stripe_size_bits) >>
-		d->stripe_size_bits;
+	d->nr_stripes = DIV_ROUND_UP_ULL(sectors, d->stripe_size);
 
-	if (!d->nr_stripes || d->nr_stripes > SIZE_MAX / sizeof(atomic_t))
+	if (!d->nr_stripes ||
+	    d->nr_stripes > INT_MAX ||
+	    d->nr_stripes > SIZE_MAX / sizeof(atomic_t)) {
+		pr_err("nr_stripes too large");
 		return -ENOMEM;
+	}
 
 	n = d->nr_stripes * sizeof(atomic_t);
 	d->stripe_sectors_dirty = n < PAGE_SIZE << 6
@@ -777,22 +781,38 @@ static int bcache_device_init(struct bcache_device *d, unsigned block_size,
 	if (!d->stripe_sectors_dirty)
 		return -ENOMEM;
 
+	n = BITS_TO_LONGS(d->nr_stripes) * sizeof(unsigned long);
+	d->full_dirty_stripes = n < PAGE_SIZE << 6
+		? kzalloc(n, GFP_KERNEL)
+		: vzalloc(n);
+	if (!d->full_dirty_stripes)
+		return -ENOMEM;
+
+	minor = ida_simple_get(&bcache_minor, 0, MINORMASK + 1, GFP_KERNEL);
+	if (minor < 0)
+		return minor;
+
 	if (!(d->bio_split = bioset_create(4, offsetof(struct bbio, bio))) ||
 	    !(d->unaligned_bvec = mempool_create_kmalloc_pool(1,
 				sizeof(struct bio_vec) * BIO_MAX_PAGES)) ||
 	    bio_split_pool_init(&d->bio_split_hook) ||
-	    !(d->disk = alloc_disk(1)) ||
-	    !(q = blk_alloc_queue(GFP_KERNEL)))
+	    !(d->disk = alloc_disk(1))) {
+		ida_simple_remove(&bcache_minor, minor);
 		return -ENOMEM;
+	}
 
 	set_capacity(d->disk, sectors);
-	snprintf(d->disk->disk_name, DISK_NAME_LEN, "bcache%i", bcache_minor);
+	snprintf(d->disk->disk_name, DISK_NAME_LEN, "bcache%i", minor);
 
 	d->disk->major		= bcache_major;
-	d->disk->first_minor	= bcache_minor++;
+	d->disk->first_minor	= minor;
 	d->disk->fops		= &bcache_ops;
 	d->disk->private_data	= d;
 
+	q = blk_alloc_queue(GFP_KERNEL);
+	if (!q)
+		return -ENOMEM;
+
 	blk_queue_make_request(q, NULL);
 	d->disk->queue			= q;
 	q->queuedata			= d;
@@ -874,7 +894,7 @@ static void cached_dev_detach_finish(struct work_struct *w)
 	struct closure cl;
 	closure_init_stack(&cl);
 
-	BUG_ON(!atomic_read(&dc->disk.detaching));
+	BUG_ON(!test_bit(BCACHE_DEV_DETACHING, &dc->disk.flags));
 	BUG_ON(atomic_read(&dc->count));
 
 	mutex_lock(&bch_register_lock);
@@ -888,6 +908,8 @@ static void cached_dev_detach_finish(struct work_struct *w)
 	bcache_device_detach(&dc->disk);
 	list_move(&dc->list, &uncached_devices);
 
+	clear_bit(BCACHE_DEV_DETACHING, &dc->disk.flags);
+
 	mutex_unlock(&bch_register_lock);
 
 	pr_info("Caching disabled for %s", bdevname(dc->bdev, buf));
@@ -900,10 +922,10 @@ void bch_cached_dev_detach(struct cached_dev *dc)
 {
 	lockdep_assert_held(&bch_register_lock);
 
-	if (atomic_read(&dc->disk.closing))
+	if (test_bit(BCACHE_DEV_CLOSING, &dc->disk.flags))
 		return;
 
-	if (atomic_xchg(&dc->disk.detaching, 1))
+	if (test_and_set_bit(BCACHE_DEV_DETACHING, &dc->disk.flags))
 		return;
 
 	/*
@@ -1030,6 +1052,7 @@ static void cached_dev_free(struct closure *cl)
 	struct cached_dev *dc = container_of(cl, struct cached_dev, disk.cl);
 
 	cancel_delayed_work_sync(&dc->writeback_rate_update);
+	kthread_stop(dc->writeback_thread);
 
 	mutex_lock(&bch_register_lock);
 
@@ -1058,11 +1081,7 @@ static void cached_dev_flush(struct closure *cl)
 	struct bcache_device *d = &dc->disk;
 
 	mutex_lock(&bch_register_lock);
-	d->flush_done = 1;
-
-	if (d->c)
-		bcache_device_unlink(d);
-
+	bcache_device_unlink(d);
 	mutex_unlock(&bch_register_lock);
 
 	bch_cache_accounting_destroy(&dc->accounting);
@@ -1088,7 +1107,6 @@ static int cached_dev_init(struct cached_dev *dc, unsigned block_size)
 	spin_lock_init(&dc->io_lock);
 	bch_cache_accounting_init(&dc->accounting, &dc->disk.cl);
 
-	dc->sequential_merge		= true;
 	dc->sequential_cutoff		= 4 << 20;
 
 	for (io = dc->io; io < dc->io + RECENT_IO; io++) {
@@ -1260,7 +1278,8 @@ bool bch_cache_set_error(struct cache_set *c, const char *fmt, ...)
 {
 	va_list args;
 
-	if (test_bit(CACHE_SET_STOPPING, &c->flags))
+	if (c->on_error != ON_ERROR_PANIC &&
+	    test_bit(CACHE_SET_STOPPING, &c->flags))
 		return false;
 
 	/* XXX: we can be called from atomic context
@@ -1275,6 +1294,9 @@ bool bch_cache_set_error(struct cache_set *c, const char *fmt, ...)
 
 	printk(", disabling caching\n");
 
+	if (c->on_error == ON_ERROR_PANIC)
+		panic("panic forced after error\n");
+
 	bch_cache_set_unregister(c);
 	return true;
 }
@@ -1339,6 +1361,9 @@ static void cache_set_flush(struct closure *cl)
 	kobject_put(&c->internal);
 	kobject_del(&c->kobj);
 
+	if (c->gc_thread)
+		kthread_stop(c->gc_thread);
+
 	if (!IS_ERR_OR_NULL(c->root))
 		list_add(&c->root->list, &c->btree_cache);
 
@@ -1433,12 +1458,19 @@ struct cache_set *bch_cache_set_alloc(struct cache_sb *sb)
 
 	c->sort_crit_factor = int_sqrt(c->btree_pages);
 
-	mutex_init(&c->bucket_lock);
-	mutex_init(&c->sort_lock);
-	spin_lock_init(&c->sort_time_lock);
 	closure_init_unlocked(&c->sb_write);
+	mutex_init(&c->bucket_lock);
+	init_waitqueue_head(&c->try_wait);
+	init_waitqueue_head(&c->bucket_wait);
 	closure_init_unlocked(&c->uuid_write);
-	spin_lock_init(&c->btree_read_time_lock);
+	mutex_init(&c->sort_lock);
+
+	spin_lock_init(&c->sort_time.lock);
+	spin_lock_init(&c->btree_gc_time.lock);
+	spin_lock_init(&c->btree_split_time.lock);
+	spin_lock_init(&c->btree_read_time.lock);
+	spin_lock_init(&c->try_harder_time.lock);
+
 	bch_moving_init_cache_set(c);
 
 	INIT_LIST_HEAD(&c->list);
@@ -1483,11 +1515,10 @@ static void run_cache_set(struct cache_set *c)
 	const char *err = "cannot allocate memory";
 	struct cached_dev *dc, *t;
 	struct cache *ca;
+	struct closure cl;
 	unsigned i;
 
-	struct btree_op op;
-	bch_btree_op_init_stack(&op);
-	op.lock = SHRT_MAX;
+	closure_init_stack(&cl);
 
 	for_each_cache(ca, c, i)
 		c->nbuckets += ca->sb.nbuckets;
@@ -1498,7 +1529,7 @@ static void run_cache_set(struct cache_set *c)
 		struct jset *j;
 
 		err = "cannot allocate memory for journal";
-		if (bch_journal_read(c, &journal, &op))
+		if (bch_journal_read(c, &journal))
 			goto err;
 
 		pr_debug("btree_journal_read() done");
@@ -1522,23 +1553,23 @@ static void run_cache_set(struct cache_set *c)
 		k = &j->btree_root;
 
 		err = "bad btree root";
-		if (__bch_ptr_invalid(c, j->btree_level + 1, k))
+		if (bch_btree_ptr_invalid(c, k))
 			goto err;
 
 		err = "error reading btree root";
-		c->root = bch_btree_node_get(c, k, j->btree_level, &op);
+		c->root = bch_btree_node_get(c, k, j->btree_level, true);
 		if (IS_ERR_OR_NULL(c->root))
 			goto err;
 
 		list_del_init(&c->root->list);
 		rw_unlock(true, c->root);
 
-		err = uuid_read(c, j, &op.cl);
+		err = uuid_read(c, j, &cl);
 		if (err)
 			goto err;
 
 		err = "error in recovery";
-		if (bch_btree_check(c, &op))
+		if (bch_btree_check(c))
 			goto err;
 
 		bch_journal_mark(c, &journal);
@@ -1570,11 +1601,9 @@ static void run_cache_set(struct cache_set *c)
 		if (j->version < BCACHE_JSET_VERSION_UUID)
 			__uuid_write(c);
 
-		bch_journal_replay(c, &journal, &op);
+		bch_journal_replay(c, &journal);
 	} else {
 		pr_notice("invalidating existing data");
-		/* Don't want invalidate_buckets() to queue a gc yet */
-		closure_lock(&c->gc, NULL);
 
 		for_each_cache(ca, c, i) {
 			unsigned j;
@@ -1600,15 +1629,15 @@ static void run_cache_set(struct cache_set *c)
 
 		err = "cannot allocate new UUID bucket";
 		if (__uuid_write(c))
-			goto err_unlock_gc;
+			goto err;
 
 		err = "cannot allocate new btree root";
-		c->root = bch_btree_node_alloc(c, 0, &op.cl);
+		c->root = bch_btree_node_alloc(c, 0, true);
 		if (IS_ERR_OR_NULL(c->root))
-			goto err_unlock_gc;
+			goto err;
 
 		bkey_copy_key(&c->root->key, &MAX_KEY);
-		bch_btree_node_write(c->root, &op.cl);
+		bch_btree_node_write(c->root, &cl);
 
 		bch_btree_set_root(c->root);
 		rw_unlock(true, c->root);
@@ -1621,14 +1650,14 @@ static void run_cache_set(struct cache_set *c)
 		SET_CACHE_SYNC(&c->sb, true);
 
 		bch_journal_next(&c->journal);
-		bch_journal_meta(c, &op.cl);
-
-		/* Unlock */
-		closure_set_stopped(&c->gc.cl);
-		closure_put(&c->gc.cl);
+		bch_journal_meta(c, &cl);
 	}
 
-	closure_sync(&op.cl);
+	err = "error starting gc thread";
+	if (bch_gc_thread_start(c))
+		goto err;
+
+	closure_sync(&cl);
 	c->sb.last_mount = get_seconds();
 	bcache_write_super(c);
 
@@ -1638,13 +1667,10 @@ static void run_cache_set(struct cache_set *c)
 	flash_devs_run(c);
 
 	return;
-err_unlock_gc:
-	closure_set_stopped(&c->gc.cl);
-	closure_put(&c->gc.cl);
 err:
-	closure_sync(&op.cl);
+	closure_sync(&cl);
 	/* XXX: test this, it's broken */
-	bch_cache_set_error(c, err);
+	bch_cache_set_error(c, "%s", err);
 }
 
 static bool can_attach_cache(struct cache *ca, struct cache_set *c)
@@ -1725,8 +1751,6 @@ void bch_cache_release(struct kobject *kobj)
 	if (ca->set)
 		ca->set->cache[ca->sb.nr_this_dev] = NULL;
 
-	bch_cache_allocator_exit(ca);
-
 	bio_split_pool_free(&ca->bio_split_hook);
 
 	free_pages((unsigned long) ca->disk_buckets, ilog2(bucket_pages(ca)));
@@ -1758,8 +1782,6 @@ static int cache_alloc(struct cache_sb *sb, struct cache *ca)
 	__module_get(THIS_MODULE);
 	kobject_init(&ca->kobj, &bch_cache_ktype);
 
-	INIT_LIST_HEAD(&ca->discards);
-
 	bio_init(&ca->journal.bio);
 	ca->journal.bio.bi_max_vecs = 8;
 	ca->journal.bio.bi_io_vec = ca->journal.bio.bi_inline_vecs;
@@ -2006,7 +2028,6 @@ static struct notifier_block reboot = {
 static void bcache_exit(void)
 {
 	bch_debug_exit();
-	bch_writeback_exit();
 	bch_request_exit();
 	bch_btree_exit();
 	if (bcache_kobj)
@@ -2039,7 +2060,6 @@ static int __init bcache_init(void)
 	    sysfs_create_files(bcache_kobj, files) ||
 	    bch_btree_init() ||
 	    bch_request_init() ||
-	    bch_writeback_init() ||
 	    bch_debug_init(bcache_kobj))
 		goto err;
 
diff --git a/drivers/md/bcache/sysfs.c b/drivers/md/bcache/sysfs.c
index 924dcfdae111..80d4c2bee18a 100644
--- a/drivers/md/bcache/sysfs.c
+++ b/drivers/md/bcache/sysfs.c
@@ -21,6 +21,12 @@ static const char * const cache_replacement_policies[] = {
 	NULL
 };
 
+static const char * const error_actions[] = {
+	"unregister",
+	"panic",
+	NULL
+};
+
 write_attribute(attach);
 write_attribute(detach);
 write_attribute(unregister);
@@ -66,7 +72,6 @@ rw_attribute(congested_read_threshold_us);
 rw_attribute(congested_write_threshold_us);
 
 rw_attribute(sequential_cutoff);
-rw_attribute(sequential_merge);
 rw_attribute(data_csum);
 rw_attribute(cache_mode);
 rw_attribute(writeback_metadata);
@@ -90,11 +95,14 @@ rw_attribute(discard);
 rw_attribute(running);
 rw_attribute(label);
 rw_attribute(readahead);
+rw_attribute(errors);
 rw_attribute(io_error_limit);
 rw_attribute(io_error_halflife);
 rw_attribute(verify);
+rw_attribute(bypass_torture_test);
 rw_attribute(key_merging_disabled);
 rw_attribute(gc_always_rewrite);
+rw_attribute(expensive_debug_checks);
 rw_attribute(freelist_percent);
 rw_attribute(cache_replacement_policy);
 rw_attribute(btree_shrinker_disabled);
@@ -116,6 +124,7 @@ SHOW(__bch_cached_dev)
 
 	sysfs_printf(data_csum,		"%i", dc->disk.data_csum);
 	var_printf(verify,		"%i");
+	var_printf(bypass_torture_test,	"%i");
 	var_printf(writeback_metadata,	"%i");
 	var_printf(writeback_running,	"%i");
 	var_print(writeback_delay);
@@ -150,10 +159,9 @@ SHOW(__bch_cached_dev)
 	sysfs_hprint(dirty_data,
 		     bcache_dev_sectors_dirty(&dc->disk) << 9);
 
-	sysfs_hprint(stripe_size,	(1 << dc->disk.stripe_size_bits) << 9);
+	sysfs_hprint(stripe_size,	dc->disk.stripe_size << 9);
 	var_printf(partial_stripes_expensive,	"%u");
 
-	var_printf(sequential_merge,	"%i");
 	var_hprint(sequential_cutoff);
 	var_hprint(readahead);
 
@@ -185,6 +193,7 @@ STORE(__cached_dev)
 
 	sysfs_strtoul(data_csum,	dc->disk.data_csum);
 	d_strtoul(verify);
+	d_strtoul(bypass_torture_test);
 	d_strtoul(writeback_metadata);
 	d_strtoul(writeback_running);
 	d_strtoul(writeback_delay);
@@ -199,7 +208,6 @@ STORE(__cached_dev)
 			    dc->writeback_rate_p_term_inverse, 1, INT_MAX);
 	d_strtoul(writeback_rate_d_smooth);
 
-	d_strtoul(sequential_merge);
 	d_strtoi_h(sequential_cutoff);
 	d_strtoi_h(readahead);
 
@@ -311,7 +319,6 @@ static struct attribute *bch_cached_dev_files[] = {
 	&sysfs_stripe_size,
 	&sysfs_partial_stripes_expensive,
 	&sysfs_sequential_cutoff,
-	&sysfs_sequential_merge,
 	&sysfs_clear_stats,
 	&sysfs_running,
 	&sysfs_state,
@@ -319,6 +326,7 @@ static struct attribute *bch_cached_dev_files[] = {
 	&sysfs_readahead,
 #ifdef CONFIG_BCACHE_DEBUG
 	&sysfs_verify,
+	&sysfs_bypass_torture_test,
 #endif
 	NULL
 };
@@ -366,7 +374,7 @@ STORE(__bch_flash_dev)
 	}
 
 	if (attr == &sysfs_unregister) {
-		atomic_set(&d->detaching, 1);
+		set_bit(BCACHE_DEV_DETACHING, &d->flags);
 		bcache_device_stop(d);
 	}
 
@@ -481,7 +489,6 @@ lock_root:
 
 	sysfs_print(btree_used_percent,	btree_used(c));
 	sysfs_print(btree_nodes,	c->gc_stats.nodes);
-	sysfs_hprint(dirty_data,	c->gc_stats.dirty);
 	sysfs_hprint(average_key_size,	average_key_size(c));
 
 	sysfs_print(cache_read_races,
@@ -492,6 +499,10 @@ lock_root:
 	sysfs_print(writeback_keys_failed,
 		    atomic_long_read(&c->writeback_keys_failed));
 
+	if (attr == &sysfs_errors)
+		return bch_snprint_string_list(buf, PAGE_SIZE, error_actions,
+					       c->on_error);
+
 	/* See count_io_errors for why 88 */
 	sysfs_print(io_error_halflife,	c->error_decay * 88);
 	sysfs_print(io_error_limit,	c->error_limit >> IO_ERROR_SHIFT);
@@ -506,6 +517,8 @@ lock_root:
 	sysfs_print(active_journal_entries,	fifo_used(&c->journal.pin));
 	sysfs_printf(verify,			"%i", c->verify);
 	sysfs_printf(key_merging_disabled,	"%i", c->key_merging_disabled);
+	sysfs_printf(expensive_debug_checks,
+		     "%i", c->expensive_debug_checks);
 	sysfs_printf(gc_always_rewrite,		"%i", c->gc_always_rewrite);
 	sysfs_printf(btree_shrinker_disabled,	"%i", c->shrinker_disabled);
 	sysfs_printf(copy_gc_enabled,		"%i", c->copy_gc_enabled);
@@ -555,7 +568,7 @@ STORE(__bch_cache_set)
 	}
 
 	if (attr == &sysfs_trigger_gc)
-		bch_queue_gc(c);
+		wake_up_gc(c);
 
 	if (attr == &sysfs_prune_cache) {
 		struct shrink_control sc;
@@ -569,6 +582,15 @@ STORE(__bch_cache_set)
 	sysfs_strtoul(congested_write_threshold_us,
 		      c->congested_write_threshold_us);
 
+	if (attr == &sysfs_errors) {
+		ssize_t v = bch_read_string_list(buf, error_actions);
+
+		if (v < 0)
+			return v;
+
+		c->on_error = v;
+	}
+
 	if (attr == &sysfs_io_error_limit)
 		c->error_limit = strtoul_or_return(buf) << IO_ERROR_SHIFT;
 
@@ -579,6 +601,7 @@ STORE(__bch_cache_set)
 	sysfs_strtoul(journal_delay_ms,		c->journal_delay_ms);
 	sysfs_strtoul(verify,			c->verify);
 	sysfs_strtoul(key_merging_disabled,	c->key_merging_disabled);
+	sysfs_strtoul(expensive_debug_checks,	c->expensive_debug_checks);
 	sysfs_strtoul(gc_always_rewrite,	c->gc_always_rewrite);
 	sysfs_strtoul(btree_shrinker_disabled,	c->shrinker_disabled);
 	sysfs_strtoul(copy_gc_enabled,		c->copy_gc_enabled);
@@ -618,8 +641,8 @@ static struct attribute *bch_cache_set_files[] = {
 	&sysfs_cache_available_percent,
 
 	&sysfs_average_key_size,
-	&sysfs_dirty_data,
 
+	&sysfs_errors,
 	&sysfs_io_error_limit,
 	&sysfs_io_error_halflife,
 	&sysfs_congested,
@@ -653,6 +676,7 @@ static struct attribute *bch_cache_set_internal_files[] = {
 #ifdef CONFIG_BCACHE_DEBUG
 	&sysfs_verify,
 	&sysfs_key_merging_disabled,
+	&sysfs_expensive_debug_checks,
 #endif
 	&sysfs_gc_always_rewrite,
 	&sysfs_btree_shrinker_disabled,
diff --git a/drivers/md/bcache/trace.c b/drivers/md/bcache/trace.c
index f7b6c197f90f..adbc3df17a80 100644
--- a/drivers/md/bcache/trace.c
+++ b/drivers/md/bcache/trace.c
@@ -1,6 +1,5 @@
 #include "bcache.h"
 #include "btree.h"
-#include "request.h"
 
 #include <linux/blktrace_api.h>
 #include <linux/module.h>
diff --git a/drivers/md/bcache/util.c b/drivers/md/bcache/util.c
index 420dad545c7d..462214eeacbe 100644
--- a/drivers/md/bcache/util.c
+++ b/drivers/md/bcache/util.c
@@ -168,10 +168,14 @@ int bch_parse_uuid(const char *s, char *uuid)
 
 void bch_time_stats_update(struct time_stats *stats, uint64_t start_time)
 {
-	uint64_t now		= local_clock();
-	uint64_t duration	= time_after64(now, start_time)
+	uint64_t now, duration, last;
+
+	spin_lock(&stats->lock);
+
+	now		= local_clock();
+	duration	= time_after64(now, start_time)
 		? now - start_time : 0;
-	uint64_t last		= time_after64(now, stats->last)
+	last		= time_after64(now, stats->last)
 		? now - stats->last : 0;
 
 	stats->max_duration = max(stats->max_duration, duration);
@@ -188,6 +192,8 @@ void bch_time_stats_update(struct time_stats *stats, uint64_t start_time)
 	}
 
 	stats->last = now ?: 1;
+
+	spin_unlock(&stats->lock);
 }
 
 /**
diff --git a/drivers/md/bcache/util.h b/drivers/md/bcache/util.h
index ea345c6896f4..362c4b3f8b4a 100644
--- a/drivers/md/bcache/util.h
+++ b/drivers/md/bcache/util.h
@@ -15,28 +15,18 @@
 
 struct closure;
 
-#ifdef CONFIG_BCACHE_EDEBUG
+#ifdef CONFIG_BCACHE_DEBUG
 
 #define atomic_dec_bug(v)	BUG_ON(atomic_dec_return(v) < 0)
 #define atomic_inc_bug(v, i)	BUG_ON(atomic_inc_return(v) <= i)
 
-#else /* EDEBUG */
+#else /* DEBUG */
 
 #define atomic_dec_bug(v)	atomic_dec(v)
 #define atomic_inc_bug(v, i)	atomic_inc(v)
 
 #endif
 
-#define BITMASK(name, type, field, offset, size)		\
-static inline uint64_t name(const type *k)			\
-{ return (k->field >> offset) & ~(((uint64_t) ~0) << size); }	\
-								\
-static inline void SET_##name(type *k, uint64_t v)		\
-{								\
-	k->field &= ~(~((uint64_t) ~0 << size) << offset);	\
-	k->field |= v << offset;				\
-}
-
 #define DECLARE_HEAP(type, name)					\
 	struct {							\
 		size_t size, used;					\
@@ -388,6 +378,7 @@ ssize_t bch_snprint_string_list(char *buf, size_t size, const char * const list[
 ssize_t bch_read_string_list(const char *buf, const char * const list[]);
 
 struct time_stats {
+	spinlock_t	lock;
 	/*
 	 * all fields are in nanoseconds, averages are ewmas stored left shifted
 	 * by 8
diff --git a/drivers/md/bcache/writeback.c b/drivers/md/bcache/writeback.c
index ba3ee48320f2..99053b1251be 100644
--- a/drivers/md/bcache/writeback.c
+++ b/drivers/md/bcache/writeback.c
@@ -11,18 +11,11 @@
 #include "debug.h"
 #include "writeback.h"
 
+#include <linux/delay.h>
+#include <linux/freezer.h>
+#include <linux/kthread.h>
 #include <trace/events/bcache.h>
 
-static struct workqueue_struct *dirty_wq;
-
-static void read_dirty(struct closure *);
-
-struct dirty_io {
-	struct closure		cl;
-	struct cached_dev	*dc;
-	struct bio		bio;
-};
-
 /* Rate limiting */
 
 static void __update_writeback_rate(struct cached_dev *dc)
@@ -72,9 +65,6 @@ out:
 	dc->writeback_rate_derivative = derivative;
 	dc->writeback_rate_change = change;
 	dc->writeback_rate_target = target;
-
-	schedule_delayed_work(&dc->writeback_rate_update,
-			      dc->writeback_rate_update_seconds * HZ);
 }
 
 static void update_writeback_rate(struct work_struct *work)
@@ -90,13 +80,16 @@ static void update_writeback_rate(struct work_struct *work)
 		__update_writeback_rate(dc);
 
 	up_read(&dc->writeback_lock);
+
+	schedule_delayed_work(&dc->writeback_rate_update,
+			      dc->writeback_rate_update_seconds * HZ);
 }
 
 static unsigned writeback_delay(struct cached_dev *dc, unsigned sectors)
 {
 	uint64_t ret;
 
-	if (atomic_read(&dc->disk.detaching) ||
+	if (test_bit(BCACHE_DEV_DETACHING, &dc->disk.flags) ||
 	    !dc->writeback_percent)
 		return 0;
 
@@ -105,37 +98,11 @@ static unsigned writeback_delay(struct cached_dev *dc, unsigned sectors)
 	return min_t(uint64_t, ret, HZ);
 }
 
-/* Background writeback */
-
-static bool dirty_pred(struct keybuf *buf, struct bkey *k)
-{
-	return KEY_DIRTY(k);
-}
-
-static bool dirty_full_stripe_pred(struct keybuf *buf, struct bkey *k)
-{
-	uint64_t stripe;
-	unsigned nr_sectors = KEY_SIZE(k);
-	struct cached_dev *dc = container_of(buf, struct cached_dev,
-					     writeback_keys);
-	unsigned stripe_size = 1 << dc->disk.stripe_size_bits;
-
-	if (!KEY_DIRTY(k))
-		return false;
-
-	stripe = KEY_START(k) >> dc->disk.stripe_size_bits;
-	while (1) {
-		if (atomic_read(dc->disk.stripe_sectors_dirty + stripe) !=
-		    stripe_size)
-			return false;
-
-		if (nr_sectors <= stripe_size)
-			return true;
-
-		nr_sectors -= stripe_size;
-		stripe++;
-	}
-}
+struct dirty_io {
+	struct closure		cl;
+	struct cached_dev	*dc;
+	struct bio		bio;
+};
 
 static void dirty_init(struct keybuf_key *w)
 {
@@ -153,131 +120,6 @@ static void dirty_init(struct keybuf_key *w)
 	bch_bio_map(bio, NULL);
 }
 
-static void refill_dirty(struct closure *cl)
-{
-	struct cached_dev *dc = container_of(cl, struct cached_dev,
-					     writeback.cl);
-	struct keybuf *buf = &dc->writeback_keys;
-	bool searched_from_start = false;
-	struct bkey end = MAX_KEY;
-	SET_KEY_INODE(&end, dc->disk.id);
-
-	if (!atomic_read(&dc->disk.detaching) &&
-	    !dc->writeback_running)
-		closure_return(cl);
-
-	down_write(&dc->writeback_lock);
-
-	if (!atomic_read(&dc->has_dirty)) {
-		SET_BDEV_STATE(&dc->sb, BDEV_STATE_CLEAN);
-		bch_write_bdev_super(dc, NULL);
-
-		up_write(&dc->writeback_lock);
-		closure_return(cl);
-	}
-
-	if (bkey_cmp(&buf->last_scanned, &end) >= 0) {
-		buf->last_scanned = KEY(dc->disk.id, 0, 0);
-		searched_from_start = true;
-	}
-
-	if (dc->partial_stripes_expensive) {
-		uint64_t i;
-
-		for (i = 0; i < dc->disk.nr_stripes; i++)
-			if (atomic_read(dc->disk.stripe_sectors_dirty + i) ==
-			    1 << dc->disk.stripe_size_bits)
-				goto full_stripes;
-
-		goto normal_refill;
-full_stripes:
-		bch_refill_keybuf(dc->disk.c, buf, &end,
-				  dirty_full_stripe_pred);
-	} else {
-normal_refill:
-		bch_refill_keybuf(dc->disk.c, buf, &end, dirty_pred);
-	}
-
-	if (bkey_cmp(&buf->last_scanned, &end) >= 0 && searched_from_start) {
-		/* Searched the entire btree  - delay awhile */
-
-		if (RB_EMPTY_ROOT(&buf->keys)) {
-			atomic_set(&dc->has_dirty, 0);
-			cached_dev_put(dc);
-		}
-
-		if (!atomic_read(&dc->disk.detaching))
-			closure_delay(&dc->writeback, dc->writeback_delay * HZ);
-	}
-
-	up_write(&dc->writeback_lock);
-
-	bch_ratelimit_reset(&dc->writeback_rate);
-
-	/* Punt to workqueue only so we don't recurse and blow the stack */
-	continue_at(cl, read_dirty, dirty_wq);
-}
-
-void bch_writeback_queue(struct cached_dev *dc)
-{
-	if (closure_trylock(&dc->writeback.cl, &dc->disk.cl)) {
-		if (!atomic_read(&dc->disk.detaching))
-			closure_delay(&dc->writeback, dc->writeback_delay * HZ);
-
-		continue_at(&dc->writeback.cl, refill_dirty, dirty_wq);
-	}
-}
-
-void bch_writeback_add(struct cached_dev *dc)
-{
-	if (!atomic_read(&dc->has_dirty) &&
-	    !atomic_xchg(&dc->has_dirty, 1)) {
-		atomic_inc(&dc->count);
-
-		if (BDEV_STATE(&dc->sb) != BDEV_STATE_DIRTY) {
-			SET_BDEV_STATE(&dc->sb, BDEV_STATE_DIRTY);
-			/* XXX: should do this synchronously */
-			bch_write_bdev_super(dc, NULL);
-		}
-
-		bch_writeback_queue(dc);
-
-		if (dc->writeback_percent)
-			schedule_delayed_work(&dc->writeback_rate_update,
-				      dc->writeback_rate_update_seconds * HZ);
-	}
-}
-
-void bcache_dev_sectors_dirty_add(struct cache_set *c, unsigned inode,
-				  uint64_t offset, int nr_sectors)
-{
-	struct bcache_device *d = c->devices[inode];
-	unsigned stripe_size, stripe_offset;
-	uint64_t stripe;
-
-	if (!d)
-		return;
-
-	stripe_size = 1 << d->stripe_size_bits;
-	stripe = offset >> d->stripe_size_bits;
-	stripe_offset = offset & (stripe_size - 1);
-
-	while (nr_sectors) {
-		int s = min_t(unsigned, abs(nr_sectors),
-			      stripe_size - stripe_offset);
-
-		if (nr_sectors < 0)
-			s = -s;
-
-		atomic_add(s, d->stripe_sectors_dirty + stripe);
-		nr_sectors -= s;
-		stripe_offset = 0;
-		stripe++;
-	}
-}
-
-/* Background writeback - IO loop */
-
 static void dirty_io_destructor(struct closure *cl)
 {
 	struct dirty_io *io = container_of(cl, struct dirty_io, cl);
@@ -297,26 +139,25 @@ static void write_dirty_finish(struct closure *cl)
 
 	/* This is kind of a dumb way of signalling errors. */
 	if (KEY_DIRTY(&w->key)) {
+		int ret;
 		unsigned i;
-		struct btree_op op;
-		bch_btree_op_init_stack(&op);
+		struct keylist keys;
 
-		op.type = BTREE_REPLACE;
-		bkey_copy(&op.replace, &w->key);
+		bch_keylist_init(&keys);
 
-		SET_KEY_DIRTY(&w->key, false);
-		bch_keylist_add(&op.keys, &w->key);
+		bkey_copy(keys.top, &w->key);
+		SET_KEY_DIRTY(keys.top, false);
+		bch_keylist_push(&keys);
 
 		for (i = 0; i < KEY_PTRS(&w->key); i++)
 			atomic_inc(&PTR_BUCKET(dc->disk.c, &w->key, i)->pin);
 
-		bch_btree_insert(&op, dc->disk.c);
-		closure_sync(&op.cl);
+		ret = bch_btree_insert(dc->disk.c, &keys, NULL, &w->key);
 
-		if (op.insert_collision)
+		if (ret)
 			trace_bcache_writeback_collision(&w->key);
 
-		atomic_long_inc(op.insert_collision
+		atomic_long_inc(ret
 				? &dc->disk.c->writeback_keys_failed
 				: &dc->disk.c->writeback_keys_done);
 	}
@@ -374,30 +215,33 @@ static void read_dirty_submit(struct closure *cl)
 	continue_at(cl, write_dirty, system_wq);
 }
 
-static void read_dirty(struct closure *cl)
+static void read_dirty(struct cached_dev *dc)
 {
-	struct cached_dev *dc = container_of(cl, struct cached_dev,
-					     writeback.cl);
-	unsigned delay = writeback_delay(dc, 0);
+	unsigned delay = 0;
 	struct keybuf_key *w;
 	struct dirty_io *io;
+	struct closure cl;
+
+	closure_init_stack(&cl);
 
 	/*
 	 * XXX: if we error, background writeback just spins. Should use some
 	 * mempools.
 	 */
 
-	while (1) {
+	while (!kthread_should_stop()) {
+		try_to_freeze();
+
 		w = bch_keybuf_next(&dc->writeback_keys);
 		if (!w)
 			break;
 
 		BUG_ON(ptr_stale(dc->disk.c, &w->key, 0));
 
-		if (delay > 0 &&
-		    (KEY_START(&w->key) != dc->last_read ||
-		     jiffies_to_msecs(delay) > 50))
-			delay = schedule_timeout_uninterruptible(delay);
+		if (KEY_START(&w->key) != dc->last_read ||
+		    jiffies_to_msecs(delay) > 50)
+			while (!kthread_should_stop() && delay)
+				delay = schedule_timeout_interruptible(delay);
 
 		dc->last_read	= KEY_OFFSET(&w->key);
 
@@ -423,7 +267,7 @@ static void read_dirty(struct closure *cl)
 		trace_bcache_writeback(&w->key);
 
 		down(&dc->in_flight);
-		closure_call(&io->cl, read_dirty_submit, NULL, cl);
+		closure_call(&io->cl, read_dirty_submit, NULL, &cl);
 
 		delay = writeback_delay(dc, KEY_SIZE(&w->key));
 	}
@@ -439,52 +283,205 @@ err:
 	 * Wait for outstanding writeback IOs to finish (and keybuf slots to be
 	 * freed) before refilling again
 	 */
-	continue_at(cl, refill_dirty, dirty_wq);
+	closure_sync(&cl);
 }
 
-/* Init */
+/* Scan for dirty data */
+
+void bcache_dev_sectors_dirty_add(struct cache_set *c, unsigned inode,
+				  uint64_t offset, int nr_sectors)
+{
+	struct bcache_device *d = c->devices[inode];
+	unsigned stripe_offset, stripe, sectors_dirty;
+
+	if (!d)
+		return;
+
+	stripe = offset_to_stripe(d, offset);
+	stripe_offset = offset & (d->stripe_size - 1);
+
+	while (nr_sectors) {
+		int s = min_t(unsigned, abs(nr_sectors),
+			      d->stripe_size - stripe_offset);
+
+		if (nr_sectors < 0)
+			s = -s;
+
+		if (stripe >= d->nr_stripes)
+			return;
+
+		sectors_dirty = atomic_add_return(s,
+					d->stripe_sectors_dirty + stripe);
+		if (sectors_dirty == d->stripe_size)
+			set_bit(stripe, d->full_dirty_stripes);
+		else
+			clear_bit(stripe, d->full_dirty_stripes);
+
+		nr_sectors -= s;
+		stripe_offset = 0;
+		stripe++;
+	}
+}
 
-static int bch_btree_sectors_dirty_init(struct btree *b, struct btree_op *op,
-					struct cached_dev *dc)
+static bool dirty_pred(struct keybuf *buf, struct bkey *k)
 {
-	struct bkey *k;
-	struct btree_iter iter;
-
-	bch_btree_iter_init(b, &iter, &KEY(dc->disk.id, 0, 0));
-	while ((k = bch_btree_iter_next_filter(&iter, b, bch_ptr_bad)))
-		if (!b->level) {
-			if (KEY_INODE(k) > dc->disk.id)
-				break;
-
-			if (KEY_DIRTY(k))
-				bcache_dev_sectors_dirty_add(b->c, dc->disk.id,
-							     KEY_START(k),
-							     KEY_SIZE(k));
-		} else {
-			btree(sectors_dirty_init, k, b, op, dc);
-			if (KEY_INODE(k) > dc->disk.id)
-				break;
-
-			cond_resched();
+	return KEY_DIRTY(k);
+}
+
+static void refill_full_stripes(struct cached_dev *dc)
+{
+	struct keybuf *buf = &dc->writeback_keys;
+	unsigned start_stripe, stripe, next_stripe;
+	bool wrapped = false;
+
+	stripe = offset_to_stripe(&dc->disk, KEY_OFFSET(&buf->last_scanned));
+
+	if (stripe >= dc->disk.nr_stripes)
+		stripe = 0;
+
+	start_stripe = stripe;
+
+	while (1) {
+		stripe = find_next_bit(dc->disk.full_dirty_stripes,
+				       dc->disk.nr_stripes, stripe);
+
+		if (stripe == dc->disk.nr_stripes)
+			goto next;
+
+		next_stripe = find_next_zero_bit(dc->disk.full_dirty_stripes,
+						 dc->disk.nr_stripes, stripe);
+
+		buf->last_scanned = KEY(dc->disk.id,
+					stripe * dc->disk.stripe_size, 0);
+
+		bch_refill_keybuf(dc->disk.c, buf,
+				  &KEY(dc->disk.id,
+				       next_stripe * dc->disk.stripe_size, 0),
+				  dirty_pred);
+
+		if (array_freelist_empty(&buf->freelist))
+			return;
+
+		stripe = next_stripe;
+next:
+		if (wrapped && stripe > start_stripe)
+			return;
+
+		if (stripe == dc->disk.nr_stripes) {
+			stripe = 0;
+			wrapped = true;
 		}
+	}
+}
+
+static bool refill_dirty(struct cached_dev *dc)
+{
+	struct keybuf *buf = &dc->writeback_keys;
+	struct bkey end = KEY(dc->disk.id, MAX_KEY_OFFSET, 0);
+	bool searched_from_start = false;
+
+	if (dc->partial_stripes_expensive) {
+		refill_full_stripes(dc);
+		if (array_freelist_empty(&buf->freelist))
+			return false;
+	}
+
+	if (bkey_cmp(&buf->last_scanned, &end) >= 0) {
+		buf->last_scanned = KEY(dc->disk.id, 0, 0);
+		searched_from_start = true;
+	}
+
+	bch_refill_keybuf(dc->disk.c, buf, &end, dirty_pred);
+
+	return bkey_cmp(&buf->last_scanned, &end) >= 0 && searched_from_start;
+}
+
+static int bch_writeback_thread(void *arg)
+{
+	struct cached_dev *dc = arg;
+	bool searched_full_index;
+
+	while (!kthread_should_stop()) {
+		down_write(&dc->writeback_lock);
+		if (!atomic_read(&dc->has_dirty) ||
+		    (!test_bit(BCACHE_DEV_DETACHING, &dc->disk.flags) &&
+		     !dc->writeback_running)) {
+			up_write(&dc->writeback_lock);
+			set_current_state(TASK_INTERRUPTIBLE);
+
+			if (kthread_should_stop())
+				return 0;
+
+			try_to_freeze();
+			schedule();
+			continue;
+		}
+
+		searched_full_index = refill_dirty(dc);
+
+		if (searched_full_index &&
+		    RB_EMPTY_ROOT(&dc->writeback_keys.keys)) {
+			atomic_set(&dc->has_dirty, 0);
+			cached_dev_put(dc);
+			SET_BDEV_STATE(&dc->sb, BDEV_STATE_CLEAN);
+			bch_write_bdev_super(dc, NULL);
+		}
+
+		up_write(&dc->writeback_lock);
+
+		bch_ratelimit_reset(&dc->writeback_rate);
+		read_dirty(dc);
+
+		if (searched_full_index) {
+			unsigned delay = dc->writeback_delay * HZ;
+
+			while (delay &&
+			       !kthread_should_stop() &&
+			       !test_bit(BCACHE_DEV_DETACHING, &dc->disk.flags))
+				delay = schedule_timeout_interruptible(delay);
+		}
+	}
 
 	return 0;
 }
 
+/* Init */
+
+struct sectors_dirty_init {
+	struct btree_op	op;
+	unsigned	inode;
+};
+
+static int sectors_dirty_init_fn(struct btree_op *_op, struct btree *b,
+				 struct bkey *k)
+{
+	struct sectors_dirty_init *op = container_of(_op,
+						struct sectors_dirty_init, op);
+	if (KEY_INODE(k) > op->inode)
+		return MAP_DONE;
+
+	if (KEY_DIRTY(k))
+		bcache_dev_sectors_dirty_add(b->c, KEY_INODE(k),
+					     KEY_START(k), KEY_SIZE(k));
+
+	return MAP_CONTINUE;
+}
+
 void bch_sectors_dirty_init(struct cached_dev *dc)
 {
-	struct btree_op op;
+	struct sectors_dirty_init op;
+
+	bch_btree_op_init(&op.op, -1);
+	op.inode = dc->disk.id;
 
-	bch_btree_op_init_stack(&op);
-	btree_root(sectors_dirty_init, dc->disk.c, &op, dc);
+	bch_btree_map_keys(&op.op, dc->disk.c, &KEY(op.inode, 0, 0),
+			   sectors_dirty_init_fn, 0);
 }
 
-void bch_cached_dev_writeback_init(struct cached_dev *dc)
+int bch_cached_dev_writeback_init(struct cached_dev *dc)
 {
 	sema_init(&dc->in_flight, 64);
-	closure_init_unlocked(&dc->writeback);
 	init_rwsem(&dc->writeback_lock);
-
 	bch_keybuf_init(&dc->writeback_keys);
 
 	dc->writeback_metadata		= true;
@@ -498,22 +495,16 @@ void bch_cached_dev_writeback_init(struct cached_dev *dc)
 	dc->writeback_rate_p_term_inverse = 64;
 	dc->writeback_rate_d_smooth	= 8;
 
+	dc->writeback_thread = kthread_create(bch_writeback_thread, dc,
+					      "bcache_writeback");
+	if (IS_ERR(dc->writeback_thread))
+		return PTR_ERR(dc->writeback_thread);
+
+	set_task_state(dc->writeback_thread, TASK_INTERRUPTIBLE);
+
 	INIT_DELAYED_WORK(&dc->writeback_rate_update, update_writeback_rate);
 	schedule_delayed_work(&dc->writeback_rate_update,
 			      dc->writeback_rate_update_seconds * HZ);
-}
-
-void bch_writeback_exit(void)
-{
-	if (dirty_wq)
-		destroy_workqueue(dirty_wq);
-}
-
-int __init bch_writeback_init(void)
-{
-	dirty_wq = create_workqueue("bcache_writeback");
-	if (!dirty_wq)
-		return -ENOMEM;
 
 	return 0;
 }
diff --git a/drivers/md/bcache/writeback.h b/drivers/md/bcache/writeback.h
index c91f61bb95b6..c9ddcf4614b9 100644
--- a/drivers/md/bcache/writeback.h
+++ b/drivers/md/bcache/writeback.h
@@ -14,20 +14,27 @@ static inline uint64_t bcache_dev_sectors_dirty(struct bcache_device *d)
 	return ret;
 }
 
-static inline bool bcache_dev_stripe_dirty(struct bcache_device *d,
+static inline unsigned offset_to_stripe(struct bcache_device *d,
+					uint64_t offset)
+{
+	do_div(offset, d->stripe_size);
+	return offset;
+}
+
+static inline bool bcache_dev_stripe_dirty(struct cached_dev *dc,
 					   uint64_t offset,
 					   unsigned nr_sectors)
 {
-	uint64_t stripe = offset >> d->stripe_size_bits;
+	unsigned stripe = offset_to_stripe(&dc->disk, offset);
 
 	while (1) {
-		if (atomic_read(d->stripe_sectors_dirty + stripe))
+		if (atomic_read(dc->disk.stripe_sectors_dirty + stripe))
 			return true;
 
-		if (nr_sectors <= 1 << d->stripe_size_bits)
+		if (nr_sectors <= dc->disk.stripe_size)
 			return false;
 
-		nr_sectors -= 1 << d->stripe_size_bits;
+		nr_sectors -= dc->disk.stripe_size;
 		stripe++;
 	}
 }
@@ -38,12 +45,12 @@ static inline bool should_writeback(struct cached_dev *dc, struct bio *bio,
 	unsigned in_use = dc->disk.c->gc_stats.in_use;
 
 	if (cache_mode != CACHE_MODE_WRITEBACK ||
-	    atomic_read(&dc->disk.detaching) ||
+	    test_bit(BCACHE_DEV_DETACHING, &dc->disk.flags) ||
 	    in_use > CUTOFF_WRITEBACK_SYNC)
 		return false;
 
 	if (dc->partial_stripes_expensive &&
-	    bcache_dev_stripe_dirty(&dc->disk, bio->bi_sector,
+	    bcache_dev_stripe_dirty(dc, bio->bi_sector,
 				    bio_sectors(bio)))
 		return true;
 
@@ -54,11 +61,30 @@ static inline bool should_writeback(struct cached_dev *dc, struct bio *bio,
 		in_use <= CUTOFF_WRITEBACK;
 }
 
+static inline void bch_writeback_queue(struct cached_dev *dc)
+{
+	wake_up_process(dc->writeback_thread);
+}
+
+static inline void bch_writeback_add(struct cached_dev *dc)
+{
+	if (!atomic_read(&dc->has_dirty) &&
+	    !atomic_xchg(&dc->has_dirty, 1)) {
+		atomic_inc(&dc->count);
+
+		if (BDEV_STATE(&dc->sb) != BDEV_STATE_DIRTY) {
+			SET_BDEV_STATE(&dc->sb, BDEV_STATE_DIRTY);
+			/* XXX: should do this synchronously */
+			bch_write_bdev_super(dc, NULL);
+		}
+
+		bch_writeback_queue(dc);
+	}
+}
+
 void bcache_dev_sectors_dirty_add(struct cache_set *, unsigned, uint64_t, int);
-void bch_writeback_queue(struct cached_dev *);
-void bch_writeback_add(struct cached_dev *);
 
 void bch_sectors_dirty_init(struct cached_dev *dc);
-void bch_cached_dev_writeback_init(struct cached_dev *);
+int bch_cached_dev_writeback_init(struct cached_dev *);
 
 #endif
diff --git a/drivers/md/dm-crypt.c b/drivers/md/dm-crypt.c
index 50ea7ed24dce..81b0fa660452 100644
--- a/drivers/md/dm-crypt.c
+++ b/drivers/md/dm-crypt.c
@@ -950,7 +950,7 @@ static int crypt_convert(struct crypt_config *cc,
 		/* async */
 		case -EBUSY:
 			wait_for_completion(&ctx->restart);
-			INIT_COMPLETION(ctx->restart);
+			reinit_completion(&ctx->restart);
 			/* fall through*/
 		case -EINPROGRESS:
 			this_cc->req = NULL;
diff --git a/drivers/md/md.c b/drivers/md/md.c
index 8766eabb0014..b6b7a2866c9e 100644
--- a/drivers/md/md.c
+++ b/drivers/md/md.c
@@ -112,7 +112,7 @@ static inline int speed_max(struct mddev *mddev)
 
 static struct ctl_table_header *raid_table_header;
 
-static ctl_table raid_table[] = {
+static struct ctl_table raid_table[] = {
 	{
 		.procname	= "speed_limit_min",
 		.data		= &sysctl_speed_limit_min,
@@ -130,7 +130,7 @@ static ctl_table raid_table[] = {
 	{ }
 };
 
-static ctl_table raid_dir_table[] = {
+static struct ctl_table raid_dir_table[] = {
 	{
 		.procname	= "raid",
 		.maxlen		= 0,
@@ -140,7 +140,7 @@ static ctl_table raid_dir_table[] = {
 	{ }
 };
 
-static ctl_table raid_root_table[] = {
+static struct ctl_table raid_root_table[] = {
 	{
 		.procname	= "dev",
 		.maxlen		= 0,
@@ -562,11 +562,19 @@ static struct mddev * mddev_find(dev_t unit)
 	goto retry;
 }
 
-static inline int mddev_lock(struct mddev * mddev)
+static inline int __must_check mddev_lock(struct mddev * mddev)
 {
 	return mutex_lock_interruptible(&mddev->reconfig_mutex);
 }
 
+/* Sometimes we need to take the lock in a situation where
+ * failure due to interrupts is not acceptable.
+ */
+static inline void mddev_lock_nointr(struct mddev * mddev)
+{
+	mutex_lock(&mddev->reconfig_mutex);
+}
+
 static inline int mddev_is_locked(struct mddev *mddev)
 {
 	return mutex_is_locked(&mddev->reconfig_mutex);
@@ -2978,7 +2986,7 @@ rdev_size_store(struct md_rdev *rdev, const char *buf, size_t len)
 		for_each_mddev(mddev, tmp) {
 			struct md_rdev *rdev2;
 
-			mddev_lock(mddev);
+			mddev_lock_nointr(mddev);
 			rdev_for_each(rdev2, mddev)
 				if (rdev->bdev == rdev2->bdev &&
 				    rdev != rdev2 &&
@@ -2994,7 +3002,7 @@ rdev_size_store(struct md_rdev *rdev, const char *buf, size_t len)
 				break;
 			}
 		}
-		mddev_lock(my_mddev);
+		mddev_lock_nointr(my_mddev);
 		if (overlap) {
 			/* Someone else could have slipped in a size
 			 * change here, but doing so is just silly.
@@ -3580,6 +3588,7 @@ level_store(struct mddev *mddev, const char *buf, size_t len)
 		mddev->in_sync = 1;
 		del_timer_sync(&mddev->safemode_timer);
 	}
+	blk_set_stacking_limits(&mddev->queue->limits);
 	pers->run(mddev);
 	set_bit(MD_CHANGE_DEVS, &mddev->flags);
 	mddev_resume(mddev);
@@ -5258,7 +5267,7 @@ static void __md_stop_writes(struct mddev *mddev)
 
 void md_stop_writes(struct mddev *mddev)
 {
-	mddev_lock(mddev);
+	mddev_lock_nointr(mddev);
 	__md_stop_writes(mddev);
 	mddev_unlock(mddev);
 }
@@ -5291,20 +5300,35 @@ EXPORT_SYMBOL_GPL(md_stop);
 static int md_set_readonly(struct mddev *mddev, struct block_device *bdev)
 {
 	int err = 0;
+	int did_freeze = 0;
+
+	if (!test_bit(MD_RECOVERY_FROZEN, &mddev->recovery)) {
+		did_freeze = 1;
+		set_bit(MD_RECOVERY_FROZEN, &mddev->recovery);
+		md_wakeup_thread(mddev->thread);
+	}
+	if (mddev->sync_thread) {
+		set_bit(MD_RECOVERY_INTR, &mddev->recovery);
+		/* Thread might be blocked waiting for metadata update
+		 * which will now never happen */
+		wake_up_process(mddev->sync_thread->tsk);
+	}
+	mddev_unlock(mddev);
+	wait_event(resync_wait, mddev->sync_thread == NULL);
+	mddev_lock_nointr(mddev);
+
 	mutex_lock(&mddev->open_mutex);
-	if (atomic_read(&mddev->openers) > !!bdev) {
+	if (atomic_read(&mddev->openers) > !!bdev ||
+	    mddev->sync_thread ||
+	    (bdev && !test_bit(MD_STILL_CLOSED, &mddev->flags))) {
 		printk("md: %s still in use.\n",mdname(mddev));
+		if (did_freeze) {
+			clear_bit(MD_RECOVERY_FROZEN, &mddev->recovery);
+			md_wakeup_thread(mddev->thread);
+		}
 		err = -EBUSY;
 		goto out;
 	}
-	if (bdev && !test_bit(MD_STILL_CLOSED, &mddev->flags)) {
-		/* Someone opened the device since we flushed it
-		 * so page cache could be dirty and it is too late
-		 * to flush.  So abort
-		 */
-		mutex_unlock(&mddev->open_mutex);
-		return -EBUSY;
-	}
 	if (mddev->pers) {
 		__md_stop_writes(mddev);
 
@@ -5315,7 +5339,7 @@ static int md_set_readonly(struct mddev *mddev, struct block_device *bdev)
 		set_disk_ro(mddev->gendisk, 1);
 		clear_bit(MD_RECOVERY_FROZEN, &mddev->recovery);
 		sysfs_notify_dirent_safe(mddev->sysfs_state);
-		err = 0;	
+		err = 0;
 	}
 out:
 	mutex_unlock(&mddev->open_mutex);
@@ -5331,20 +5355,34 @@ static int do_md_stop(struct mddev * mddev, int mode,
 {
 	struct gendisk *disk = mddev->gendisk;
 	struct md_rdev *rdev;
+	int did_freeze = 0;
+
+	if (!test_bit(MD_RECOVERY_FROZEN, &mddev->recovery)) {
+		did_freeze = 1;
+		set_bit(MD_RECOVERY_FROZEN, &mddev->recovery);
+		md_wakeup_thread(mddev->thread);
+	}
+	if (mddev->sync_thread) {
+		set_bit(MD_RECOVERY_INTR, &mddev->recovery);
+		/* Thread might be blocked waiting for metadata update
+		 * which will now never happen */
+		wake_up_process(mddev->sync_thread->tsk);
+	}
+	mddev_unlock(mddev);
+	wait_event(resync_wait, mddev->sync_thread == NULL);
+	mddev_lock_nointr(mddev);
 
 	mutex_lock(&mddev->open_mutex);
 	if (atomic_read(&mddev->openers) > !!bdev ||
-	    mddev->sysfs_active) {
+	    mddev->sysfs_active ||
+	    mddev->sync_thread ||
+	    (bdev && !test_bit(MD_STILL_CLOSED, &mddev->flags))) {
 		printk("md: %s still in use.\n",mdname(mddev));
 		mutex_unlock(&mddev->open_mutex);
-		return -EBUSY;
-	}
-	if (bdev && !test_bit(MD_STILL_CLOSED, &mddev->flags)) {
-		/* Someone opened the device since we flushed it
-		 * so page cache could be dirty and it is too late
-		 * to flush.  So abort
-		 */
-		mutex_unlock(&mddev->open_mutex);
+		if (did_freeze) {
+			clear_bit(MD_RECOVERY_FROZEN, &mddev->recovery);
+			md_wakeup_thread(mddev->thread);
+		}
 		return -EBUSY;
 	}
 	if (mddev->pers) {
@@ -6551,7 +6589,7 @@ static int md_ioctl(struct block_device *bdev, fmode_t mode,
 				wait_event(mddev->sb_wait,
 					   !test_bit(MD_CHANGE_DEVS, &mddev->flags) &&
 					   !test_bit(MD_CHANGE_PENDING, &mddev->flags));
-				mddev_lock(mddev);
+				mddev_lock_nointr(mddev);
 			}
 		} else {
 			err = -EROFS;
@@ -7361,9 +7399,6 @@ void md_do_sync(struct md_thread *thread)
 		mddev->curr_resync = 2;
 
 	try_again:
-		if (kthread_should_stop())
-			set_bit(MD_RECOVERY_INTR, &mddev->recovery);
-
 		if (test_bit(MD_RECOVERY_INTR, &mddev->recovery))
 			goto skip;
 		for_each_mddev(mddev2, tmp) {
@@ -7388,7 +7423,7 @@ void md_do_sync(struct md_thread *thread)
 				 * be caught by 'softlockup'
 				 */
 				prepare_to_wait(&resync_wait, &wq, TASK_INTERRUPTIBLE);
-				if (!kthread_should_stop() &&
+				if (!test_bit(MD_RECOVERY_INTR, &mddev->recovery) &&
 				    mddev2->curr_resync >= mddev->curr_resync) {
 					printk(KERN_INFO "md: delaying %s of %s"
 					       " until %s has finished (they"
@@ -7464,7 +7499,7 @@ void md_do_sync(struct md_thread *thread)
 	last_check = 0;
 
 	if (j>2) {
-		printk(KERN_INFO 
+		printk(KERN_INFO
 		       "md: resuming %s of %s from checkpoint.\n",
 		       desc, mdname(mddev));
 		mddev->curr_resync = j;
@@ -7501,7 +7536,8 @@ void md_do_sync(struct md_thread *thread)
 			sysfs_notify(&mddev->kobj, NULL, "sync_completed");
 		}
 
-		while (j >= mddev->resync_max && !kthread_should_stop()) {
+		while (j >= mddev->resync_max &&
+		       !test_bit(MD_RECOVERY_INTR, &mddev->recovery)) {
 			/* As this condition is controlled by user-space,
 			 * we can block indefinitely, so use '_interruptible'
 			 * to avoid triggering warnings.
@@ -7509,17 +7545,18 @@ void md_do_sync(struct md_thread *thread)
 			flush_signals(current); /* just in case */
 			wait_event_interruptible(mddev->recovery_wait,
 						 mddev->resync_max > j
-						 || kthread_should_stop());
+						 || test_bit(MD_RECOVERY_INTR,
+							     &mddev->recovery));
 		}
 
-		if (kthread_should_stop())
-			goto interrupted;
+		if (test_bit(MD_RECOVERY_INTR, &mddev->recovery))
+			break;
 
 		sectors = mddev->pers->sync_request(mddev, j, &skipped,
 						  currspeed < speed_min(mddev));
 		if (sectors == 0) {
 			set_bit(MD_RECOVERY_INTR, &mddev->recovery);
-			goto out;
+			break;
 		}
 
 		if (!skipped) { /* actual IO requested */
@@ -7556,10 +7593,8 @@ void md_do_sync(struct md_thread *thread)
 			last_mark = next;
 		}
 
-
-		if (kthread_should_stop())
-			goto interrupted;
-
+		if (test_bit(MD_RECOVERY_INTR, &mddev->recovery))
+			break;
 
 		/*
 		 * this loop exits only if either when we are slower than
@@ -7582,11 +7617,12 @@ void md_do_sync(struct md_thread *thread)
 			}
 		}
 	}
-	printk(KERN_INFO "md: %s: %s done.\n",mdname(mddev), desc);
+	printk(KERN_INFO "md: %s: %s %s.\n",mdname(mddev), desc,
+	       test_bit(MD_RECOVERY_INTR, &mddev->recovery)
+	       ? "interrupted" : "done");
 	/*
 	 * this also signals 'finished resyncing' to md_stop
 	 */
- out:
 	blk_finish_plug(&plug);
 	wait_event(mddev->recovery_wait, !atomic_read(&mddev->recovery_active));
 
@@ -7640,16 +7676,6 @@ void md_do_sync(struct md_thread *thread)
 	set_bit(MD_RECOVERY_DONE, &mddev->recovery);
 	md_wakeup_thread(mddev->thread);
 	return;
-
- interrupted:
-	/*
-	 * got a signal, exit.
-	 */
-	printk(KERN_INFO
-	       "md: md_do_sync() got signal ... exiting\n");
-	set_bit(MD_RECOVERY_INTR, &mddev->recovery);
-	goto out;
-
 }
 EXPORT_SYMBOL_GPL(md_do_sync);
 
@@ -7894,6 +7920,7 @@ void md_reap_sync_thread(struct mddev *mddev)
 
 	/* resync has finished, collect result */
 	md_unregister_thread(&mddev->sync_thread);
+	wake_up(&resync_wait);
 	if (!test_bit(MD_RECOVERY_INTR, &mddev->recovery) &&
 	    !test_bit(MD_RECOVERY_REQUESTED, &mddev->recovery)) {
 		/* success...*/
diff --git a/drivers/md/raid1.c b/drivers/md/raid1.c
index af6681b19776..1e5a540995e9 100644
--- a/drivers/md/raid1.c
+++ b/drivers/md/raid1.c
@@ -66,7 +66,8 @@
  */
 static int max_queued_requests = 1024;
 
-static void allow_barrier(struct r1conf *conf);
+static void allow_barrier(struct r1conf *conf, sector_t start_next_window,
+			  sector_t bi_sector);
 static void lower_barrier(struct r1conf *conf);
 
 static void * r1bio_pool_alloc(gfp_t gfp_flags, void *data)
@@ -84,10 +85,12 @@ static void r1bio_pool_free(void *r1_bio, void *data)
 }
 
 #define RESYNC_BLOCK_SIZE (64*1024)
-//#define RESYNC_BLOCK_SIZE PAGE_SIZE
+#define RESYNC_DEPTH 32
 #define RESYNC_SECTORS (RESYNC_BLOCK_SIZE >> 9)
 #define RESYNC_PAGES ((RESYNC_BLOCK_SIZE + PAGE_SIZE-1) / PAGE_SIZE)
-#define RESYNC_WINDOW (2048*1024)
+#define RESYNC_WINDOW (RESYNC_BLOCK_SIZE * RESYNC_DEPTH)
+#define RESYNC_WINDOW_SECTORS (RESYNC_WINDOW >> 9)
+#define NEXT_NORMALIO_DISTANCE (3 * RESYNC_WINDOW_SECTORS)
 
 static void * r1buf_pool_alloc(gfp_t gfp_flags, void *data)
 {
@@ -225,6 +228,8 @@ static void call_bio_endio(struct r1bio *r1_bio)
 	struct bio *bio = r1_bio->master_bio;
 	int done;
 	struct r1conf *conf = r1_bio->mddev->private;
+	sector_t start_next_window = r1_bio->start_next_window;
+	sector_t bi_sector = bio->bi_sector;
 
 	if (bio->bi_phys_segments) {
 		unsigned long flags;
@@ -232,6 +237,11 @@ static void call_bio_endio(struct r1bio *r1_bio)
 		bio->bi_phys_segments--;
 		done = (bio->bi_phys_segments == 0);
 		spin_unlock_irqrestore(&conf->device_lock, flags);
+		/*
+		 * make_request() might be waiting for
+		 * bi_phys_segments to decrease
+		 */
+		wake_up(&conf->wait_barrier);
 	} else
 		done = 1;
 
@@ -243,7 +253,7 @@ static void call_bio_endio(struct r1bio *r1_bio)
 		 * Wake up any possible resync thread that waits for the device
 		 * to go idle.
 		 */
-		allow_barrier(conf);
+		allow_barrier(conf, start_next_window, bi_sector);
 	}
 }
 
@@ -814,8 +824,6 @@ static void flush_pending_writes(struct r1conf *conf)
  *    there is no normal IO happeing.  It must arrange to call
  *    lower_barrier when the particular background IO completes.
  */
-#define RESYNC_DEPTH 32
-
 static void raise_barrier(struct r1conf *conf)
 {
 	spin_lock_irq(&conf->resync_lock);
@@ -827,9 +835,19 @@ static void raise_barrier(struct r1conf *conf)
 	/* block any new IO from starting */
 	conf->barrier++;
 
-	/* Now wait for all pending IO to complete */
+	/* For these conditions we must wait:
+	 * A: while the array is in frozen state
+	 * B: while barrier >= RESYNC_DEPTH, meaning resync reach
+	 *    the max count which allowed.
+	 * C: next_resync + RESYNC_SECTORS > start_next_window, meaning
+	 *    next resync will reach to the window which normal bios are
+	 *    handling.
+	 */
 	wait_event_lock_irq(conf->wait_barrier,
-			    !conf->nr_pending && conf->barrier < RESYNC_DEPTH,
+			    !conf->array_frozen &&
+			    conf->barrier < RESYNC_DEPTH &&
+			    (conf->start_next_window >=
+			     conf->next_resync + RESYNC_SECTORS),
 			    conf->resync_lock);
 
 	spin_unlock_irq(&conf->resync_lock);
@@ -845,10 +863,33 @@ static void lower_barrier(struct r1conf *conf)
 	wake_up(&conf->wait_barrier);
 }
 
-static void wait_barrier(struct r1conf *conf)
+static bool need_to_wait_for_sync(struct r1conf *conf, struct bio *bio)
 {
+	bool wait = false;
+
+	if (conf->array_frozen || !bio)
+		wait = true;
+	else if (conf->barrier && bio_data_dir(bio) == WRITE) {
+		if (conf->next_resync < RESYNC_WINDOW_SECTORS)
+			wait = true;
+		else if ((conf->next_resync - RESYNC_WINDOW_SECTORS
+				>= bio_end_sector(bio)) ||
+			 (conf->next_resync + NEXT_NORMALIO_DISTANCE
+				<= bio->bi_sector))
+			wait = false;
+		else
+			wait = true;
+	}
+
+	return wait;
+}
+
+static sector_t wait_barrier(struct r1conf *conf, struct bio *bio)
+{
+	sector_t sector = 0;
+
 	spin_lock_irq(&conf->resync_lock);
-	if (conf->barrier) {
+	if (need_to_wait_for_sync(conf, bio)) {
 		conf->nr_waiting++;
 		/* Wait for the barrier to drop.
 		 * However if there are already pending
@@ -860,22 +901,67 @@ static void wait_barrier(struct r1conf *conf)
 		 * count down.
 		 */
 		wait_event_lock_irq(conf->wait_barrier,
-				    !conf->barrier ||
-				    (conf->nr_pending &&
+				    !conf->array_frozen &&
+				    (!conf->barrier ||
+				    ((conf->start_next_window <
+				      conf->next_resync + RESYNC_SECTORS) &&
 				     current->bio_list &&
-				     !bio_list_empty(current->bio_list)),
+				     !bio_list_empty(current->bio_list))),
 				    conf->resync_lock);
 		conf->nr_waiting--;
 	}
+
+	if (bio && bio_data_dir(bio) == WRITE) {
+		if (conf->next_resync + NEXT_NORMALIO_DISTANCE
+		    <= bio->bi_sector) {
+			if (conf->start_next_window == MaxSector)
+				conf->start_next_window =
+					conf->next_resync +
+					NEXT_NORMALIO_DISTANCE;
+
+			if ((conf->start_next_window + NEXT_NORMALIO_DISTANCE)
+			    <= bio->bi_sector)
+				conf->next_window_requests++;
+			else
+				conf->current_window_requests++;
+		}
+		if (bio->bi_sector >= conf->start_next_window)
+			sector = conf->start_next_window;
+	}
+
 	conf->nr_pending++;
 	spin_unlock_irq(&conf->resync_lock);
+	return sector;
 }
 
-static void allow_barrier(struct r1conf *conf)
+static void allow_barrier(struct r1conf *conf, sector_t start_next_window,
+			  sector_t bi_sector)
 {
 	unsigned long flags;
+
 	spin_lock_irqsave(&conf->resync_lock, flags);
 	conf->nr_pending--;
+	if (start_next_window) {
+		if (start_next_window == conf->start_next_window) {
+			if (conf->start_next_window + NEXT_NORMALIO_DISTANCE
+			    <= bi_sector)
+				conf->next_window_requests--;
+			else
+				conf->current_window_requests--;
+		} else
+			conf->current_window_requests--;
+
+		if (!conf->current_window_requests) {
+			if (conf->next_window_requests) {
+				conf->current_window_requests =
+					conf->next_window_requests;
+				conf->next_window_requests = 0;
+				conf->start_next_window +=
+					NEXT_NORMALIO_DISTANCE;
+			} else
+				conf->start_next_window = MaxSector;
+		}
+	}
 	spin_unlock_irqrestore(&conf->resync_lock, flags);
 	wake_up(&conf->wait_barrier);
 }
@@ -884,8 +970,7 @@ static void freeze_array(struct r1conf *conf, int extra)
 {
 	/* stop syncio and normal IO and wait for everything to
 	 * go quite.
-	 * We increment barrier and nr_waiting, and then
-	 * wait until nr_pending match nr_queued+extra
+	 * We wait until nr_pending match nr_queued+extra
 	 * This is called in the context of one normal IO request
 	 * that has failed. Thus any sync request that might be pending
 	 * will be blocked by nr_pending, and we need to wait for
@@ -895,8 +980,7 @@ static void freeze_array(struct r1conf *conf, int extra)
 	 * we continue.
 	 */
 	spin_lock_irq(&conf->resync_lock);
-	conf->barrier++;
-	conf->nr_waiting++;
+	conf->array_frozen = 1;
 	wait_event_lock_irq_cmd(conf->wait_barrier,
 				conf->nr_pending == conf->nr_queued+extra,
 				conf->resync_lock,
@@ -907,8 +991,7 @@ static void unfreeze_array(struct r1conf *conf)
 {
 	/* reverse the effect of the freeze */
 	spin_lock_irq(&conf->resync_lock);
-	conf->barrier--;
-	conf->nr_waiting--;
+	conf->array_frozen = 0;
 	wake_up(&conf->wait_barrier);
 	spin_unlock_irq(&conf->resync_lock);
 }
@@ -1013,6 +1096,7 @@ static void make_request(struct mddev *mddev, struct bio * bio)
 	int first_clone;
 	int sectors_handled;
 	int max_sectors;
+	sector_t start_next_window;
 
 	/*
 	 * Register the new request and wait if the reconstruction
@@ -1042,7 +1126,7 @@ static void make_request(struct mddev *mddev, struct bio * bio)
 		finish_wait(&conf->wait_barrier, &w);
 	}
 
-	wait_barrier(conf);
+	start_next_window = wait_barrier(conf, bio);
 
 	bitmap = mddev->bitmap;
 
@@ -1163,6 +1247,7 @@ read_again:
 
 	disks = conf->raid_disks * 2;
  retry_write:
+	r1_bio->start_next_window = start_next_window;
 	blocked_rdev = NULL;
 	rcu_read_lock();
 	max_sectors = r1_bio->sectors;
@@ -1231,14 +1316,24 @@ read_again:
 	if (unlikely(blocked_rdev)) {
 		/* Wait for this device to become unblocked */
 		int j;
+		sector_t old = start_next_window;
 
 		for (j = 0; j < i; j++)
 			if (r1_bio->bios[j])
 				rdev_dec_pending(conf->mirrors[j].rdev, mddev);
 		r1_bio->state = 0;
-		allow_barrier(conf);
+		allow_barrier(conf, start_next_window, bio->bi_sector);
 		md_wait_for_blocked_rdev(blocked_rdev, mddev);
-		wait_barrier(conf);
+		start_next_window = wait_barrier(conf, bio);
+		/*
+		 * We must make sure the multi r1bios of bio have
+		 * the same value of bi_phys_segments
+		 */
+		if (bio->bi_phys_segments && old &&
+		    old != start_next_window)
+			/* Wait for the former r1bio(s) to complete */
+			wait_event(conf->wait_barrier,
+				   bio->bi_phys_segments == 1);
 		goto retry_write;
 	}
 
@@ -1438,11 +1533,14 @@ static void print_conf(struct r1conf *conf)
 
 static void close_sync(struct r1conf *conf)
 {
-	wait_barrier(conf);
-	allow_barrier(conf);
+	wait_barrier(conf, NULL);
+	allow_barrier(conf, 0, 0);
 
 	mempool_destroy(conf->r1buf_pool);
 	conf->r1buf_pool = NULL;
+
+	conf->next_resync = 0;
+	conf->start_next_window = MaxSector;
 }
 
 static int raid1_spare_active(struct mddev *mddev)
@@ -2714,6 +2812,9 @@ static struct r1conf *setup_conf(struct mddev *mddev)
 	conf->pending_count = 0;
 	conf->recovery_disabled = mddev->recovery_disabled - 1;
 
+	conf->start_next_window = MaxSector;
+	conf->current_window_requests = conf->next_window_requests = 0;
+
 	err = -EIO;
 	for (i = 0; i < conf->raid_disks * 2; i++) {
 
@@ -2871,8 +2972,8 @@ static int stop(struct mddev *mddev)
 			   atomic_read(&bitmap->behind_writes) == 0);
 	}
 
-	raise_barrier(conf);
-	lower_barrier(conf);
+	freeze_array(conf, 0);
+	unfreeze_array(conf);
 
 	md_unregister_thread(&mddev->thread);
 	if (conf->r1bio_pool)
@@ -3031,10 +3132,10 @@ static void raid1_quiesce(struct mddev *mddev, int state)
 		wake_up(&conf->wait_barrier);
 		break;
 	case 1:
-		raise_barrier(conf);
+		freeze_array(conf, 0);
 		break;
 	case 0:
-		lower_barrier(conf);
+		unfreeze_array(conf);
 		break;
 	}
 }
@@ -3051,7 +3152,8 @@ static void *raid1_takeover(struct mddev *mddev)
 		mddev->new_chunk_sectors = 0;
 		conf = setup_conf(mddev);
 		if (!IS_ERR(conf))
-			conf->barrier = 1;
+			/* Array must appear to be quiesced */
+			conf->array_frozen = 1;
 		return conf;
 	}
 	return ERR_PTR(-EINVAL);
diff --git a/drivers/md/raid1.h b/drivers/md/raid1.h
index 0ff3715fb7eb..9bebca7bff2f 100644
--- a/drivers/md/raid1.h
+++ b/drivers/md/raid1.h
@@ -41,6 +41,19 @@ struct r1conf {
 	 */
 	sector_t		next_resync;
 
+	/* When raid1 starts resync, we divide array into four partitions
+	 * |---------|--------------|---------------------|-------------|
+	 *        next_resync   start_next_window       end_window
+	 * start_next_window = next_resync + NEXT_NORMALIO_DISTANCE
+	 * end_window = start_next_window + NEXT_NORMALIO_DISTANCE
+	 * current_window_requests means the count of normalIO between
+	 *   start_next_window and end_window.
+	 * next_window_requests means the count of normalIO after end_window.
+	 * */
+	sector_t		start_next_window;
+	int			current_window_requests;
+	int			next_window_requests;
+
 	spinlock_t		device_lock;
 
 	/* list of 'struct r1bio' that need to be processed by raid1d,
@@ -65,6 +78,7 @@ struct r1conf {
 	int			nr_waiting;
 	int			nr_queued;
 	int			barrier;
+	int			array_frozen;
 
 	/* Set to 1 if a full sync is needed, (fresh device added).
 	 * Cleared when a sync completes.
@@ -111,6 +125,7 @@ struct r1bio {
 						 * in this BehindIO request
 						 */
 	sector_t		sector;
+	sector_t		start_next_window;
 	int			sectors;
 	unsigned long		state;
 	struct mddev		*mddev;
diff --git a/drivers/md/raid10.c b/drivers/md/raid10.c
index 7c3508abb5e1..c504e8389e69 100644
--- a/drivers/md/raid10.c
+++ b/drivers/md/raid10.c
@@ -4384,7 +4384,11 @@ static sector_t reshape_request(struct mddev *mddev, sector_t sector_nr,
 		set_bit(MD_CHANGE_DEVS, &mddev->flags);
 		md_wakeup_thread(mddev->thread);
 		wait_event(mddev->sb_wait, mddev->flags == 0 ||
-			   kthread_should_stop());
+			   test_bit(MD_RECOVERY_INTR, &mddev->recovery));
+		if (test_bit(MD_RECOVERY_INTR, &mddev->recovery)) {
+			allow_barrier(conf);
+			return sectors_done;
+		}
 		conf->reshape_safe = mddev->reshape_position;
 		allow_barrier(conf);
 	}
diff --git a/drivers/md/raid5.c b/drivers/md/raid5.c
index f8b906843926..47da0af6322b 100644
--- a/drivers/md/raid5.c
+++ b/drivers/md/raid5.c
@@ -85,6 +85,42 @@ static inline struct hlist_head *stripe_hash(struct r5conf *conf, sector_t sect)
 	return &conf->stripe_hashtbl[hash];
 }
 
+static inline int stripe_hash_locks_hash(sector_t sect)
+{
+	return (sect >> STRIPE_SHIFT) & STRIPE_HASH_LOCKS_MASK;
+}
+
+static inline void lock_device_hash_lock(struct r5conf *conf, int hash)
+{
+	spin_lock_irq(conf->hash_locks + hash);
+	spin_lock(&conf->device_lock);
+}
+
+static inline void unlock_device_hash_lock(struct r5conf *conf, int hash)
+{
+	spin_unlock(&conf->device_lock);
+	spin_unlock_irq(conf->hash_locks + hash);
+}
+
+static inline void lock_all_device_hash_locks_irq(struct r5conf *conf)
+{
+	int i;
+	local_irq_disable();
+	spin_lock(conf->hash_locks);
+	for (i = 1; i < NR_STRIPE_HASH_LOCKS; i++)
+		spin_lock_nest_lock(conf->hash_locks + i, conf->hash_locks);
+	spin_lock(&conf->device_lock);
+}
+
+static inline void unlock_all_device_hash_locks_irq(struct r5conf *conf)
+{
+	int i;
+	spin_unlock(&conf->device_lock);
+	for (i = NR_STRIPE_HASH_LOCKS; i; i--)
+		spin_unlock(conf->hash_locks + i - 1);
+	local_irq_enable();
+}
+
 /* bio's attached to a stripe+device for I/O are linked together in bi_sector
  * order without overlap.  There may be several bio's per stripe+device, and
  * a bio could span several devices.
@@ -249,7 +285,8 @@ static void raid5_wakeup_stripe_thread(struct stripe_head *sh)
 	}
 }
 
-static void do_release_stripe(struct r5conf *conf, struct stripe_head *sh)
+static void do_release_stripe(struct r5conf *conf, struct stripe_head *sh,
+			      struct list_head *temp_inactive_list)
 {
 	BUG_ON(!list_empty(&sh->lru));
 	BUG_ON(atomic_read(&conf->active_stripes)==0);
@@ -278,37 +315,68 @@ static void do_release_stripe(struct r5conf *conf, struct stripe_head *sh)
 			    < IO_THRESHOLD)
 				md_wakeup_thread(conf->mddev->thread);
 		atomic_dec(&conf->active_stripes);
-		if (!test_bit(STRIPE_EXPANDING, &sh->state)) {
-			list_add_tail(&sh->lru, &conf->inactive_list);
-			wake_up(&conf->wait_for_stripe);
-			if (conf->retry_read_aligned)
-				md_wakeup_thread(conf->mddev->thread);
-		}
+		if (!test_bit(STRIPE_EXPANDING, &sh->state))
+			list_add_tail(&sh->lru, temp_inactive_list);
 	}
 }
 
-static void __release_stripe(struct r5conf *conf, struct stripe_head *sh)
+static void __release_stripe(struct r5conf *conf, struct stripe_head *sh,
+			     struct list_head *temp_inactive_list)
 {
 	if (atomic_dec_and_test(&sh->count))
-		do_release_stripe(conf, sh);
+		do_release_stripe(conf, sh, temp_inactive_list);
 }
 
-static struct llist_node *llist_reverse_order(struct llist_node *head)
+/*
+ * @hash could be NR_STRIPE_HASH_LOCKS, then we have a list of inactive_list
+ *
+ * Be careful: Only one task can add/delete stripes from temp_inactive_list at
+ * given time. Adding stripes only takes device lock, while deleting stripes
+ * only takes hash lock.
+ */
+static void release_inactive_stripe_list(struct r5conf *conf,
+					 struct list_head *temp_inactive_list,
+					 int hash)
 {
-	struct llist_node *new_head = NULL;
+	int size;
+	bool do_wakeup = false;
+	unsigned long flags;
 
-	while (head) {
-		struct llist_node *tmp = head;
-		head = head->next;
-		tmp->next = new_head;
-		new_head = tmp;
+	if (hash == NR_STRIPE_HASH_LOCKS) {
+		size = NR_STRIPE_HASH_LOCKS;
+		hash = NR_STRIPE_HASH_LOCKS - 1;
+	} else
+		size = 1;
+	while (size) {
+		struct list_head *list = &temp_inactive_list[size - 1];
+
+		/*
+		 * We don't hold any lock here yet, get_active_stripe() might
+		 * remove stripes from the list
+		 */
+		if (!list_empty_careful(list)) {
+			spin_lock_irqsave(conf->hash_locks + hash, flags);
+			if (list_empty(conf->inactive_list + hash) &&
+			    !list_empty(list))
+				atomic_dec(&conf->empty_inactive_list_nr);
+			list_splice_tail_init(list, conf->inactive_list + hash);
+			do_wakeup = true;
+			spin_unlock_irqrestore(conf->hash_locks + hash, flags);
+		}
+		size--;
+		hash--;
 	}
 
-	return new_head;
+	if (do_wakeup) {
+		wake_up(&conf->wait_for_stripe);
+		if (conf->retry_read_aligned)
+			md_wakeup_thread(conf->mddev->thread);
+	}
 }
 
 /* should hold conf->device_lock already */
-static int release_stripe_list(struct r5conf *conf)
+static int release_stripe_list(struct r5conf *conf,
+			       struct list_head *temp_inactive_list)
 {
 	struct stripe_head *sh;
 	int count = 0;
@@ -317,6 +385,8 @@ static int release_stripe_list(struct r5conf *conf)
 	head = llist_del_all(&conf->released_stripes);
 	head = llist_reverse_order(head);
 	while (head) {
+		int hash;
+
 		sh = llist_entry(head, struct stripe_head, release_list);
 		head = llist_next(head);
 		/* sh could be readded after STRIPE_ON_RELEASE_LIST is cleard */
@@ -327,7 +397,8 @@ static int release_stripe_list(struct r5conf *conf)
 		 * again, the count is always > 1. This is true for
 		 * STRIPE_ON_UNPLUG_LIST bit too.
 		 */
-		__release_stripe(conf, sh);
+		hash = sh->hash_lock_index;
+		__release_stripe(conf, sh, &temp_inactive_list[hash]);
 		count++;
 	}
 
@@ -338,9 +409,12 @@ static void release_stripe(struct stripe_head *sh)
 {
 	struct r5conf *conf = sh->raid_conf;
 	unsigned long flags;
+	struct list_head list;
+	int hash;
 	bool wakeup;
 
-	if (test_and_set_bit(STRIPE_ON_RELEASE_LIST, &sh->state))
+	if (unlikely(!conf->mddev->thread) ||
+		test_and_set_bit(STRIPE_ON_RELEASE_LIST, &sh->state))
 		goto slow_path;
 	wakeup = llist_add(&sh->release_list, &conf->released_stripes);
 	if (wakeup)
@@ -350,8 +424,11 @@ slow_path:
 	local_irq_save(flags);
 	/* we are ok here if STRIPE_ON_RELEASE_LIST is set or not */
 	if (atomic_dec_and_lock(&sh->count, &conf->device_lock)) {
-		do_release_stripe(conf, sh);
+		INIT_LIST_HEAD(&list);
+		hash = sh->hash_lock_index;
+		do_release_stripe(conf, sh, &list);
 		spin_unlock(&conf->device_lock);
+		release_inactive_stripe_list(conf, &list, hash);
 	}
 	local_irq_restore(flags);
 }
@@ -376,18 +453,21 @@ static inline void insert_hash(struct r5conf *conf, struct stripe_head *sh)
 
 
 /* find an idle stripe, make sure it is unhashed, and return it. */
-static struct stripe_head *get_free_stripe(struct r5conf *conf)
+static struct stripe_head *get_free_stripe(struct r5conf *conf, int hash)
 {
 	struct stripe_head *sh = NULL;
 	struct list_head *first;
 
-	if (list_empty(&conf->inactive_list))
+	if (list_empty(conf->inactive_list + hash))
 		goto out;
-	first = conf->inactive_list.next;
+	first = (conf->inactive_list + hash)->next;
 	sh = list_entry(first, struct stripe_head, lru);
 	list_del_init(first);
 	remove_hash(sh);
 	atomic_inc(&conf->active_stripes);
+	BUG_ON(hash != sh->hash_lock_index);
+	if (list_empty(conf->inactive_list + hash))
+		atomic_inc(&conf->empty_inactive_list_nr);
 out:
 	return sh;
 }
@@ -430,7 +510,7 @@ static void stripe_set_idx(sector_t stripe, struct r5conf *conf, int previous,
 static void init_stripe(struct stripe_head *sh, sector_t sector, int previous)
 {
 	struct r5conf *conf = sh->raid_conf;
-	int i;
+	int i, seq;
 
 	BUG_ON(atomic_read(&sh->count) != 0);
 	BUG_ON(test_bit(STRIPE_HANDLE, &sh->state));
@@ -440,7 +520,8 @@ static void init_stripe(struct stripe_head *sh, sector_t sector, int previous)
 		(unsigned long long)sh->sector);
 
 	remove_hash(sh);
-
+retry:
+	seq = read_seqcount_begin(&conf->gen_lock);
 	sh->generation = conf->generation - previous;
 	sh->disks = previous ? conf->previous_raid_disks : conf->raid_disks;
 	sh->sector = sector;
@@ -462,6 +543,8 @@ static void init_stripe(struct stripe_head *sh, sector_t sector, int previous)
 		dev->flags = 0;
 		raid5_build_block(sh, i, previous);
 	}
+	if (read_seqcount_retry(&conf->gen_lock, seq))
+		goto retry;
 	insert_hash(conf, sh);
 	sh->cpu = smp_processor_id();
 }
@@ -566,29 +649,31 @@ get_active_stripe(struct r5conf *conf, sector_t sector,
 		  int previous, int noblock, int noquiesce)
 {
 	struct stripe_head *sh;
+	int hash = stripe_hash_locks_hash(sector);
 
 	pr_debug("get_stripe, sector %llu\n", (unsigned long long)sector);
 
-	spin_lock_irq(&conf->device_lock);
+	spin_lock_irq(conf->hash_locks + hash);
 
 	do {
 		wait_event_lock_irq(conf->wait_for_stripe,
 				    conf->quiesce == 0 || noquiesce,
-				    conf->device_lock);
+				    *(conf->hash_locks + hash));
 		sh = __find_stripe(conf, sector, conf->generation - previous);
 		if (!sh) {
 			if (!conf->inactive_blocked)
-				sh = get_free_stripe(conf);
+				sh = get_free_stripe(conf, hash);
 			if (noblock && sh == NULL)
 				break;
 			if (!sh) {
 				conf->inactive_blocked = 1;
-				wait_event_lock_irq(conf->wait_for_stripe,
-						    !list_empty(&conf->inactive_list) &&
-						    (atomic_read(&conf->active_stripes)
-						     < (conf->max_nr_stripes *3/4)
-						     || !conf->inactive_blocked),
-						    conf->device_lock);
+				wait_event_lock_irq(
+					conf->wait_for_stripe,
+					!list_empty(conf->inactive_list + hash) &&
+					(atomic_read(&conf->active_stripes)
+					 < (conf->max_nr_stripes * 3 / 4)
+					 || !conf->inactive_blocked),
+					*(conf->hash_locks + hash));
 				conf->inactive_blocked = 0;
 			} else
 				init_stripe(sh, sector, previous);
@@ -599,9 +684,11 @@ get_active_stripe(struct r5conf *conf, sector_t sector,
 				    && !test_bit(STRIPE_ON_UNPLUG_LIST, &sh->state)
 				    && !test_bit(STRIPE_ON_RELEASE_LIST, &sh->state));
 			} else {
+				spin_lock(&conf->device_lock);
 				if (!test_bit(STRIPE_HANDLE, &sh->state))
 					atomic_inc(&conf->active_stripes);
 				if (list_empty(&sh->lru) &&
+				    !test_bit(STRIPE_ON_RELEASE_LIST, &sh->state) &&
 				    !test_bit(STRIPE_EXPANDING, &sh->state))
 					BUG();
 				list_del_init(&sh->lru);
@@ -609,6 +696,7 @@ get_active_stripe(struct r5conf *conf, sector_t sector,
 					sh->group->stripes_cnt--;
 					sh->group = NULL;
 				}
+				spin_unlock(&conf->device_lock);
 			}
 		}
 	} while (sh == NULL);
@@ -616,7 +704,7 @@ get_active_stripe(struct r5conf *conf, sector_t sector,
 	if (sh)
 		atomic_inc(&sh->count);
 
-	spin_unlock_irq(&conf->device_lock);
+	spin_unlock_irq(conf->hash_locks + hash);
 	return sh;
 }
 
@@ -772,7 +860,7 @@ static void ops_run_io(struct stripe_head *sh, struct stripe_head_state *s)
 				bi->bi_sector = (sh->sector
 						 + rdev->data_offset);
 			if (test_bit(R5_ReadNoMerge, &sh->dev[i].flags))
-				bi->bi_rw |= REQ_FLUSH;
+				bi->bi_rw |= REQ_NOMERGE;
 
 			bi->bi_vcnt = 1;
 			bi->bi_io_vec[0].bv_len = STRIPE_SIZE;
@@ -1596,7 +1684,7 @@ static void raid_run_ops(struct stripe_head *sh, unsigned long ops_request)
 	put_cpu();
 }
 
-static int grow_one_stripe(struct r5conf *conf)
+static int grow_one_stripe(struct r5conf *conf, int hash)
 {
 	struct stripe_head *sh;
 	sh = kmem_cache_zalloc(conf->slab_cache, GFP_KERNEL);
@@ -1612,6 +1700,7 @@ static int grow_one_stripe(struct r5conf *conf)
 		kmem_cache_free(conf->slab_cache, sh);
 		return 0;
 	}
+	sh->hash_lock_index = hash;
 	/* we just created an active stripe so... */
 	atomic_set(&sh->count, 1);
 	atomic_inc(&conf->active_stripes);
@@ -1624,6 +1713,7 @@ static int grow_stripes(struct r5conf *conf, int num)
 {
 	struct kmem_cache *sc;
 	int devs = max(conf->raid_disks, conf->previous_raid_disks);
+	int hash;
 
 	if (conf->mddev->gendisk)
 		sprintf(conf->cache_name[0],
@@ -1641,9 +1731,13 @@ static int grow_stripes(struct r5conf *conf, int num)
 		return 1;
 	conf->slab_cache = sc;
 	conf->pool_size = devs;
-	while (num--)
-		if (!grow_one_stripe(conf))
+	hash = conf->max_nr_stripes % NR_STRIPE_HASH_LOCKS;
+	while (num--) {
+		if (!grow_one_stripe(conf, hash))
 			return 1;
+		conf->max_nr_stripes++;
+		hash = (hash + 1) % NR_STRIPE_HASH_LOCKS;
+	}
 	return 0;
 }
 
@@ -1701,6 +1795,7 @@ static int resize_stripes(struct r5conf *conf, int newsize)
 	int err;
 	struct kmem_cache *sc;
 	int i;
+	int hash, cnt;
 
 	if (newsize <= conf->pool_size)
 		return 0; /* never bother to shrink */
@@ -1740,19 +1835,29 @@ static int resize_stripes(struct r5conf *conf, int newsize)
 	 * OK, we have enough stripes, start collecting inactive
 	 * stripes and copying them over
 	 */
+	hash = 0;
+	cnt = 0;
 	list_for_each_entry(nsh, &newstripes, lru) {
-		spin_lock_irq(&conf->device_lock);
-		wait_event_lock_irq(conf->wait_for_stripe,
-				    !list_empty(&conf->inactive_list),
-				    conf->device_lock);
-		osh = get_free_stripe(conf);
-		spin_unlock_irq(&conf->device_lock);
+		lock_device_hash_lock(conf, hash);
+		wait_event_cmd(conf->wait_for_stripe,
+				    !list_empty(conf->inactive_list + hash),
+				    unlock_device_hash_lock(conf, hash),
+				    lock_device_hash_lock(conf, hash));
+		osh = get_free_stripe(conf, hash);
+		unlock_device_hash_lock(conf, hash);
 		atomic_set(&nsh->count, 1);
 		for(i=0; i<conf->pool_size; i++)
 			nsh->dev[i].page = osh->dev[i].page;
 		for( ; i<newsize; i++)
 			nsh->dev[i].page = NULL;
+		nsh->hash_lock_index = hash;
 		kmem_cache_free(conf->slab_cache, osh);
+		cnt++;
+		if (cnt >= conf->max_nr_stripes / NR_STRIPE_HASH_LOCKS +
+		    !!((conf->max_nr_stripes % NR_STRIPE_HASH_LOCKS) > hash)) {
+			hash++;
+			cnt = 0;
+		}
 	}
 	kmem_cache_destroy(conf->slab_cache);
 
@@ -1811,13 +1916,13 @@ static int resize_stripes(struct r5conf *conf, int newsize)
 	return err;
 }
 
-static int drop_one_stripe(struct r5conf *conf)
+static int drop_one_stripe(struct r5conf *conf, int hash)
 {
 	struct stripe_head *sh;
 
-	spin_lock_irq(&conf->device_lock);
-	sh = get_free_stripe(conf);
-	spin_unlock_irq(&conf->device_lock);
+	spin_lock_irq(conf->hash_locks + hash);
+	sh = get_free_stripe(conf, hash);
+	spin_unlock_irq(conf->hash_locks + hash);
 	if (!sh)
 		return 0;
 	BUG_ON(atomic_read(&sh->count));
@@ -1829,8 +1934,10 @@ static int drop_one_stripe(struct r5conf *conf)
 
 static void shrink_stripes(struct r5conf *conf)
 {
-	while (drop_one_stripe(conf))
-		;
+	int hash;
+	for (hash = 0; hash < NR_STRIPE_HASH_LOCKS; hash++)
+		while (drop_one_stripe(conf, hash))
+			;
 
 	if (conf->slab_cache)
 		kmem_cache_destroy(conf->slab_cache);
@@ -1935,6 +2042,9 @@ static void raid5_end_read_request(struct bio * bi, int error)
 			       mdname(conf->mddev), bdn);
 		else
 			retry = 1;
+		if (set_bad && test_bit(In_sync, &rdev->flags)
+		    && !test_bit(R5_ReadNoMerge, &sh->dev[i].flags))
+			retry = 1;
 		if (retry)
 			if (test_bit(R5_ReadNoMerge, &sh->dev[i].flags)) {
 				set_bit(R5_ReadError, &sh->dev[i].flags);
@@ -3914,7 +4024,8 @@ static void raid5_activate_delayed(struct r5conf *conf)
 	}
 }
 
-static void activate_bit_delay(struct r5conf *conf)
+static void activate_bit_delay(struct r5conf *conf,
+	struct list_head *temp_inactive_list)
 {
 	/* device_lock is held */
 	struct list_head head;
@@ -3922,9 +4033,11 @@ static void activate_bit_delay(struct r5conf *conf)
 	list_del_init(&conf->bitmap_list);
 	while (!list_empty(&head)) {
 		struct stripe_head *sh = list_entry(head.next, struct stripe_head, lru);
+		int hash;
 		list_del_init(&sh->lru);
 		atomic_inc(&sh->count);
-		__release_stripe(conf, sh);
+		hash = sh->hash_lock_index;
+		__release_stripe(conf, sh, &temp_inactive_list[hash]);
 	}
 }
 
@@ -3940,7 +4053,7 @@ int md_raid5_congested(struct mddev *mddev, int bits)
 		return 1;
 	if (conf->quiesce)
 		return 1;
-	if (list_empty_careful(&conf->inactive_list))
+	if (atomic_read(&conf->empty_inactive_list_nr))
 		return 1;
 
 	return 0;
@@ -4270,6 +4383,7 @@ static struct stripe_head *__get_priority_stripe(struct r5conf *conf, int group)
 struct raid5_plug_cb {
 	struct blk_plug_cb	cb;
 	struct list_head	list;
+	struct list_head	temp_inactive_list[NR_STRIPE_HASH_LOCKS];
 };
 
 static void raid5_unplug(struct blk_plug_cb *blk_cb, bool from_schedule)
@@ -4280,6 +4394,7 @@ static void raid5_unplug(struct blk_plug_cb *blk_cb, bool from_schedule)
 	struct mddev *mddev = cb->cb.data;
 	struct r5conf *conf = mddev->private;
 	int cnt = 0;
+	int hash;
 
 	if (cb->list.next && !list_empty(&cb->list)) {
 		spin_lock_irq(&conf->device_lock);
@@ -4297,11 +4412,14 @@ static void raid5_unplug(struct blk_plug_cb *blk_cb, bool from_schedule)
 			 * STRIPE_ON_RELEASE_LIST could be set here. In that
 			 * case, the count is always > 1 here
 			 */
-			__release_stripe(conf, sh);
+			hash = sh->hash_lock_index;
+			__release_stripe(conf, sh, &cb->temp_inactive_list[hash]);
 			cnt++;
 		}
 		spin_unlock_irq(&conf->device_lock);
 	}
+	release_inactive_stripe_list(conf, cb->temp_inactive_list,
+				     NR_STRIPE_HASH_LOCKS);
 	if (mddev->queue)
 		trace_block_unplug(mddev->queue, cnt, !from_schedule);
 	kfree(cb);
@@ -4322,8 +4440,12 @@ static void release_stripe_plug(struct mddev *mddev,
 
 	cb = container_of(blk_cb, struct raid5_plug_cb, cb);
 
-	if (cb->list.next == NULL)
+	if (cb->list.next == NULL) {
+		int i;
 		INIT_LIST_HEAD(&cb->list);
+		for (i = 0; i < NR_STRIPE_HASH_LOCKS; i++)
+			INIT_LIST_HEAD(cb->temp_inactive_list + i);
+	}
 
 	if (!test_and_set_bit(STRIPE_ON_UNPLUG_LIST, &sh->state))
 		list_add_tail(&sh->lru, &cb->list);
@@ -4706,14 +4828,19 @@ static sector_t reshape_request(struct mddev *mddev, sector_t sector_nr, int *sk
 	    time_after(jiffies, conf->reshape_checkpoint + 10*HZ)) {
 		/* Cannot proceed until we've updated the superblock... */
 		wait_event(conf->wait_for_overlap,
-			   atomic_read(&conf->reshape_stripes)==0);
+			   atomic_read(&conf->reshape_stripes)==0
+			   || test_bit(MD_RECOVERY_INTR, &mddev->recovery));
+		if (atomic_read(&conf->reshape_stripes) != 0)
+			return 0;
 		mddev->reshape_position = conf->reshape_progress;
 		mddev->curr_resync_completed = sector_nr;
 		conf->reshape_checkpoint = jiffies;
 		set_bit(MD_CHANGE_DEVS, &mddev->flags);
 		md_wakeup_thread(mddev->thread);
 		wait_event(mddev->sb_wait, mddev->flags == 0 ||
-			   kthread_should_stop());
+			   test_bit(MD_RECOVERY_INTR, &mddev->recovery));
+		if (test_bit(MD_RECOVERY_INTR, &mddev->recovery))
+			return 0;
 		spin_lock_irq(&conf->device_lock);
 		conf->reshape_safe = mddev->reshape_position;
 		spin_unlock_irq(&conf->device_lock);
@@ -4796,7 +4923,10 @@ static sector_t reshape_request(struct mddev *mddev, sector_t sector_nr, int *sk
 	    >= mddev->resync_max - mddev->curr_resync_completed) {
 		/* Cannot proceed until we've updated the superblock... */
 		wait_event(conf->wait_for_overlap,
-			   atomic_read(&conf->reshape_stripes) == 0);
+			   atomic_read(&conf->reshape_stripes) == 0
+			   || test_bit(MD_RECOVERY_INTR, &mddev->recovery));
+		if (atomic_read(&conf->reshape_stripes) != 0)
+			goto ret;
 		mddev->reshape_position = conf->reshape_progress;
 		mddev->curr_resync_completed = sector_nr;
 		conf->reshape_checkpoint = jiffies;
@@ -4804,13 +4934,16 @@ static sector_t reshape_request(struct mddev *mddev, sector_t sector_nr, int *sk
 		md_wakeup_thread(mddev->thread);
 		wait_event(mddev->sb_wait,
 			   !test_bit(MD_CHANGE_DEVS, &mddev->flags)
-			   || kthread_should_stop());
+			   || test_bit(MD_RECOVERY_INTR, &mddev->recovery));
+		if (test_bit(MD_RECOVERY_INTR, &mddev->recovery))
+			goto ret;
 		spin_lock_irq(&conf->device_lock);
 		conf->reshape_safe = mddev->reshape_position;
 		spin_unlock_irq(&conf->device_lock);
 		wake_up(&conf->wait_for_overlap);
 		sysfs_notify(&mddev->kobj, NULL, "sync_completed");
 	}
+ret:
 	return reshape_sectors;
 }
 
@@ -4968,27 +5101,45 @@ static int  retry_aligned_read(struct r5conf *conf, struct bio *raid_bio)
 }
 
 static int handle_active_stripes(struct r5conf *conf, int group,
-				 struct r5worker *worker)
+				 struct r5worker *worker,
+				 struct list_head *temp_inactive_list)
 {
 	struct stripe_head *batch[MAX_STRIPE_BATCH], *sh;
-	int i, batch_size = 0;
+	int i, batch_size = 0, hash;
+	bool release_inactive = false;
 
 	while (batch_size < MAX_STRIPE_BATCH &&
 			(sh = __get_priority_stripe(conf, group)) != NULL)
 		batch[batch_size++] = sh;
 
-	if (batch_size == 0)
-		return batch_size;
+	if (batch_size == 0) {
+		for (i = 0; i < NR_STRIPE_HASH_LOCKS; i++)
+			if (!list_empty(temp_inactive_list + i))
+				break;
+		if (i == NR_STRIPE_HASH_LOCKS)
+			return batch_size;
+		release_inactive = true;
+	}
 	spin_unlock_irq(&conf->device_lock);
 
+	release_inactive_stripe_list(conf, temp_inactive_list,
+				     NR_STRIPE_HASH_LOCKS);
+
+	if (release_inactive) {
+		spin_lock_irq(&conf->device_lock);
+		return 0;
+	}
+
 	for (i = 0; i < batch_size; i++)
 		handle_stripe(batch[i]);
 
 	cond_resched();
 
 	spin_lock_irq(&conf->device_lock);
-	for (i = 0; i < batch_size; i++)
-		__release_stripe(conf, batch[i]);
+	for (i = 0; i < batch_size; i++) {
+		hash = batch[i]->hash_lock_index;
+		__release_stripe(conf, batch[i], &temp_inactive_list[hash]);
+	}
 	return batch_size;
 }
 
@@ -5009,9 +5160,10 @@ static void raid5_do_work(struct work_struct *work)
 	while (1) {
 		int batch_size, released;
 
-		released = release_stripe_list(conf);
+		released = release_stripe_list(conf, worker->temp_inactive_list);
 
-		batch_size = handle_active_stripes(conf, group_id, worker);
+		batch_size = handle_active_stripes(conf, group_id, worker,
+						   worker->temp_inactive_list);
 		worker->working = false;
 		if (!batch_size && !released)
 			break;
@@ -5050,7 +5202,7 @@ static void raid5d(struct md_thread *thread)
 		struct bio *bio;
 		int batch_size, released;
 
-		released = release_stripe_list(conf);
+		released = release_stripe_list(conf, conf->temp_inactive_list);
 
 		if (
 		    !list_empty(&conf->bitmap_list)) {
@@ -5060,7 +5212,7 @@ static void raid5d(struct md_thread *thread)
 			bitmap_unplug(mddev->bitmap);
 			spin_lock_irq(&conf->device_lock);
 			conf->seq_write = conf->seq_flush;
-			activate_bit_delay(conf);
+			activate_bit_delay(conf, conf->temp_inactive_list);
 		}
 		raid5_activate_delayed(conf);
 
@@ -5074,7 +5226,8 @@ static void raid5d(struct md_thread *thread)
 			handled++;
 		}
 
-		batch_size = handle_active_stripes(conf, ANY_GROUP, NULL);
+		batch_size = handle_active_stripes(conf, ANY_GROUP, NULL,
+						   conf->temp_inactive_list);
 		if (!batch_size && !released)
 			break;
 		handled += batch_size;
@@ -5110,22 +5263,29 @@ raid5_set_cache_size(struct mddev *mddev, int size)
 {
 	struct r5conf *conf = mddev->private;
 	int err;
+	int hash;
 
 	if (size <= 16 || size > 32768)
 		return -EINVAL;
+	hash = (conf->max_nr_stripes - 1) % NR_STRIPE_HASH_LOCKS;
 	while (size < conf->max_nr_stripes) {
-		if (drop_one_stripe(conf))
+		if (drop_one_stripe(conf, hash))
 			conf->max_nr_stripes--;
 		else
 			break;
+		hash--;
+		if (hash < 0)
+			hash = NR_STRIPE_HASH_LOCKS - 1;
 	}
 	err = md_allow_write(mddev);
 	if (err)
 		return err;
+	hash = conf->max_nr_stripes % NR_STRIPE_HASH_LOCKS;
 	while (size > conf->max_nr_stripes) {
-		if (grow_one_stripe(conf))
+		if (grow_one_stripe(conf, hash))
 			conf->max_nr_stripes++;
 		else break;
+		hash = (hash + 1) % NR_STRIPE_HASH_LOCKS;
 	}
 	return 0;
 }
@@ -5213,15 +5373,18 @@ raid5_show_group_thread_cnt(struct mddev *mddev, char *page)
 		return 0;
 }
 
-static int alloc_thread_groups(struct r5conf *conf, int cnt);
+static int alloc_thread_groups(struct r5conf *conf, int cnt,
+			       int *group_cnt,
+			       int *worker_cnt_per_group,
+			       struct r5worker_group **worker_groups);
 static ssize_t
 raid5_store_group_thread_cnt(struct mddev *mddev, const char *page, size_t len)
 {
 	struct r5conf *conf = mddev->private;
 	unsigned long new;
 	int err;
-	struct r5worker_group *old_groups;
-	int old_group_cnt;
+	struct r5worker_group *new_groups, *old_groups;
+	int group_cnt, worker_cnt_per_group;
 
 	if (len >= PAGE_SIZE)
 		return -EINVAL;
@@ -5237,14 +5400,19 @@ raid5_store_group_thread_cnt(struct mddev *mddev, const char *page, size_t len)
 	mddev_suspend(mddev);
 
 	old_groups = conf->worker_groups;
-	old_group_cnt = conf->worker_cnt_per_group;
+	if (old_groups)
+		flush_workqueue(raid5_wq);
+
+	err = alloc_thread_groups(conf, new,
+				  &group_cnt, &worker_cnt_per_group,
+				  &new_groups);
+	if (!err) {
+		spin_lock_irq(&conf->device_lock);
+		conf->group_cnt = group_cnt;
+		conf->worker_cnt_per_group = worker_cnt_per_group;
+		conf->worker_groups = new_groups;
+		spin_unlock_irq(&conf->device_lock);
 
-	conf->worker_groups = NULL;
-	err = alloc_thread_groups(conf, new);
-	if (err) {
-		conf->worker_groups = old_groups;
-		conf->worker_cnt_per_group = old_group_cnt;
-	} else {
 		if (old_groups)
 			kfree(old_groups[0].workers);
 		kfree(old_groups);
@@ -5274,40 +5442,47 @@ static struct attribute_group raid5_attrs_group = {
 	.attrs = raid5_attrs,
 };
 
-static int alloc_thread_groups(struct r5conf *conf, int cnt)
+static int alloc_thread_groups(struct r5conf *conf, int cnt,
+			       int *group_cnt,
+			       int *worker_cnt_per_group,
+			       struct r5worker_group **worker_groups)
 {
-	int i, j;
+	int i, j, k;
 	ssize_t size;
 	struct r5worker *workers;
 
-	conf->worker_cnt_per_group = cnt;
+	*worker_cnt_per_group = cnt;
 	if (cnt == 0) {
-		conf->worker_groups = NULL;
+		*group_cnt = 0;
+		*worker_groups = NULL;
 		return 0;
 	}
-	conf->group_cnt = num_possible_nodes();
+	*group_cnt = num_possible_nodes();
 	size = sizeof(struct r5worker) * cnt;
-	workers = kzalloc(size * conf->group_cnt, GFP_NOIO);
-	conf->worker_groups = kzalloc(sizeof(struct r5worker_group) *
-				conf->group_cnt, GFP_NOIO);
-	if (!conf->worker_groups || !workers) {
+	workers = kzalloc(size * *group_cnt, GFP_NOIO);
+	*worker_groups = kzalloc(sizeof(struct r5worker_group) *
+				*group_cnt, GFP_NOIO);
+	if (!*worker_groups || !workers) {
 		kfree(workers);
-		kfree(conf->worker_groups);
-		conf->worker_groups = NULL;
+		kfree(*worker_groups);
 		return -ENOMEM;
 	}
 
-	for (i = 0; i < conf->group_cnt; i++) {
+	for (i = 0; i < *group_cnt; i++) {
 		struct r5worker_group *group;
 
-		group = &conf->worker_groups[i];
+		group = worker_groups[i];
 		INIT_LIST_HEAD(&group->handle_list);
 		group->conf = conf;
 		group->workers = workers + i * cnt;
 
 		for (j = 0; j < cnt; j++) {
-			group->workers[j].group = group;
-			INIT_WORK(&group->workers[j].work, raid5_do_work);
+			struct r5worker *worker = group->workers + j;
+			worker->group = group;
+			INIT_WORK(&worker->work, raid5_do_work);
+
+			for (k = 0; k < NR_STRIPE_HASH_LOCKS; k++)
+				INIT_LIST_HEAD(worker->temp_inactive_list + k);
 		}
 	}
 
@@ -5458,6 +5633,9 @@ static struct r5conf *setup_conf(struct mddev *mddev)
 	struct md_rdev *rdev;
 	struct disk_info *disk;
 	char pers_name[6];
+	int i;
+	int group_cnt, worker_cnt_per_group;
+	struct r5worker_group *new_group;
 
 	if (mddev->new_level != 5
 	    && mddev->new_level != 4
@@ -5492,7 +5670,12 @@ static struct r5conf *setup_conf(struct mddev *mddev)
 	if (conf == NULL)
 		goto abort;
 	/* Don't enable multi-threading by default*/
-	if (alloc_thread_groups(conf, 0))
+	if (!alloc_thread_groups(conf, 0, &group_cnt, &worker_cnt_per_group,
+				 &new_group)) {
+		conf->group_cnt = group_cnt;
+		conf->worker_cnt_per_group = worker_cnt_per_group;
+		conf->worker_groups = new_group;
+	} else
 		goto abort;
 	spin_lock_init(&conf->device_lock);
 	seqcount_init(&conf->gen_lock);
@@ -5502,7 +5685,6 @@ static struct r5conf *setup_conf(struct mddev *mddev)
 	INIT_LIST_HEAD(&conf->hold_list);
 	INIT_LIST_HEAD(&conf->delayed_list);
 	INIT_LIST_HEAD(&conf->bitmap_list);
-	INIT_LIST_HEAD(&conf->inactive_list);
 	init_llist_head(&conf->released_stripes);
 	atomic_set(&conf->active_stripes, 0);
 	atomic_set(&conf->preread_active_stripes, 0);
@@ -5528,6 +5710,21 @@ static struct r5conf *setup_conf(struct mddev *mddev)
 	if ((conf->stripe_hashtbl = kzalloc(PAGE_SIZE, GFP_KERNEL)) == NULL)
 		goto abort;
 
+	/* We init hash_locks[0] separately to that it can be used
+	 * as the reference lock in the spin_lock_nest_lock() call
+	 * in lock_all_device_hash_locks_irq in order to convince
+	 * lockdep that we know what we are doing.
+	 */
+	spin_lock_init(conf->hash_locks);
+	for (i = 1; i < NR_STRIPE_HASH_LOCKS; i++)
+		spin_lock_init(conf->hash_locks + i);
+
+	for (i = 0; i < NR_STRIPE_HASH_LOCKS; i++)
+		INIT_LIST_HEAD(conf->inactive_list + i);
+
+	for (i = 0; i < NR_STRIPE_HASH_LOCKS; i++)
+		INIT_LIST_HEAD(conf->temp_inactive_list + i);
+
 	conf->level = mddev->new_level;
 	if (raid5_alloc_percpu(conf) != 0)
 		goto abort;
@@ -5568,7 +5765,6 @@ static struct r5conf *setup_conf(struct mddev *mddev)
 	else
 		conf->max_degraded = 1;
 	conf->algorithm = mddev->new_layout;
-	conf->max_nr_stripes = NR_STRIPES;
 	conf->reshape_progress = mddev->reshape_position;
 	if (conf->reshape_progress != MaxSector) {
 		conf->prev_chunk_sectors = mddev->chunk_sectors;
@@ -5577,7 +5773,8 @@ static struct r5conf *setup_conf(struct mddev *mddev)
 
 	memory = conf->max_nr_stripes * (sizeof(struct stripe_head) +
 		 max_disks * ((sizeof(struct bio) + PAGE_SIZE))) / 1024;
-	if (grow_stripes(conf, conf->max_nr_stripes)) {
+	atomic_set(&conf->empty_inactive_list_nr, NR_STRIPE_HASH_LOCKS);
+	if (grow_stripes(conf, NR_STRIPES)) {
 		printk(KERN_ERR
 		       "md/raid:%s: couldn't allocate %dkB for buffers\n",
 		       mdname(mddev), memory);
@@ -6383,12 +6580,18 @@ static int raid5_start_reshape(struct mddev *mddev)
 	if (!mddev->sync_thread) {
 		mddev->recovery = 0;
 		spin_lock_irq(&conf->device_lock);
+		write_seqcount_begin(&conf->gen_lock);
 		mddev->raid_disks = conf->raid_disks = conf->previous_raid_disks;
+		mddev->new_chunk_sectors =
+			conf->chunk_sectors = conf->prev_chunk_sectors;
+		mddev->new_layout = conf->algorithm = conf->prev_algo;
 		rdev_for_each(rdev, mddev)
 			rdev->new_data_offset = rdev->data_offset;
 		smp_wmb();
+		conf->generation --;
 		conf->reshape_progress = MaxSector;
 		mddev->reshape_position = MaxSector;
+		write_seqcount_end(&conf->gen_lock);
 		spin_unlock_irq(&conf->device_lock);
 		return -EAGAIN;
 	}
@@ -6476,27 +6679,28 @@ static void raid5_quiesce(struct mddev *mddev, int state)
 		break;
 
 	case 1: /* stop all writes */
-		spin_lock_irq(&conf->device_lock);
+		lock_all_device_hash_locks_irq(conf);
 		/* '2' tells resync/reshape to pause so that all
 		 * active stripes can drain
 		 */
 		conf->quiesce = 2;
-		wait_event_lock_irq(conf->wait_for_stripe,
+		wait_event_cmd(conf->wait_for_stripe,
 				    atomic_read(&conf->active_stripes) == 0 &&
 				    atomic_read(&conf->active_aligned_reads) == 0,
-				    conf->device_lock);
+				    unlock_all_device_hash_locks_irq(conf),
+				    lock_all_device_hash_locks_irq(conf));
 		conf->quiesce = 1;
-		spin_unlock_irq(&conf->device_lock);
+		unlock_all_device_hash_locks_irq(conf);
 		/* allow reshape to continue */
 		wake_up(&conf->wait_for_overlap);
 		break;
 
 	case 0: /* re-enable writes */
-		spin_lock_irq(&conf->device_lock);
+		lock_all_device_hash_locks_irq(conf);
 		conf->quiesce = 0;
 		wake_up(&conf->wait_for_stripe);
 		wake_up(&conf->wait_for_overlap);
-		spin_unlock_irq(&conf->device_lock);
+		unlock_all_device_hash_locks_irq(conf);
 		break;
 	}
 }
diff --git a/drivers/md/raid5.h b/drivers/md/raid5.h
index 2113ffa82c7a..01ad8ae8f578 100644
--- a/drivers/md/raid5.h
+++ b/drivers/md/raid5.h
@@ -49,7 +49,7 @@
  * can't distinguish between a clean block that has been generated
  * from parity calculations, and a clean block that has been
  * successfully written to the spare ( or to parity when resyncing).
- * To distingush these states we have a stripe bit STRIPE_INSYNC that
+ * To distinguish these states we have a stripe bit STRIPE_INSYNC that
  * is set whenever a write is scheduled to the spare, or to the parity
  * disc if there is no spare.  A sync request clears this bit, and
  * when we find it set with no buffers locked, we know the sync is
@@ -205,6 +205,7 @@ struct stripe_head {
 	short			pd_idx;		/* parity disk index */
 	short			qd_idx;		/* 'Q' disk index for raid6 */
 	short			ddf_layout;/* use DDF ordering to calculate Q */
+	short			hash_lock_index;
 	unsigned long		state;		/* state flags */
 	atomic_t		count;	      /* nr of active thread/requests */
 	int			bm_seq;	/* sequence number for bitmap flushes */
@@ -367,9 +368,18 @@ struct disk_info {
 	struct md_rdev	*rdev, *replacement;
 };
 
+/* NOTE NR_STRIPE_HASH_LOCKS must remain below 64.
+ * This is because we sometimes take all the spinlocks
+ * and creating that much locking depth can cause
+ * problems.
+ */
+#define NR_STRIPE_HASH_LOCKS 8
+#define STRIPE_HASH_LOCKS_MASK (NR_STRIPE_HASH_LOCKS - 1)
+
 struct r5worker {
 	struct work_struct work;
 	struct r5worker_group *group;
+	struct list_head temp_inactive_list[NR_STRIPE_HASH_LOCKS];
 	bool working;
 };
 
@@ -382,6 +392,8 @@ struct r5worker_group {
 
 struct r5conf {
 	struct hlist_head	*stripe_hashtbl;
+	/* only protect corresponding hash list and inactive_list */
+	spinlock_t		hash_locks[NR_STRIPE_HASH_LOCKS];
 	struct mddev		*mddev;
 	int			chunk_sectors;
 	int			level, algorithm;
@@ -462,7 +474,8 @@ struct r5conf {
 	 * Free stripes pool
 	 */
 	atomic_t		active_stripes;
-	struct list_head	inactive_list;
+	struct list_head	inactive_list[NR_STRIPE_HASH_LOCKS];
+	atomic_t		empty_inactive_list_nr;
 	struct llist_head	released_stripes;
 	wait_queue_head_t	wait_for_stripe;
 	wait_queue_head_t	wait_for_overlap;
@@ -477,6 +490,7 @@ struct r5conf {
 	 * the new thread here until we fully activate the array.
 	 */
 	struct md_thread	*thread;
+	struct list_head	temp_inactive_list[NR_STRIPE_HASH_LOCKS];
 	struct r5worker_group	*worker_groups;
 	int			group_cnt;
 	int			worker_cnt_per_group;
diff --git a/drivers/media/common/b2c2/flexcop-sram.c b/drivers/media/common/b2c2/flexcop-sram.c
index f2199e43e803..185c285f70fc 100644
--- a/drivers/media/common/b2c2/flexcop-sram.c
+++ b/drivers/media/common/b2c2/flexcop-sram.c
@@ -85,7 +85,7 @@ static void flexcop_sram_write(struct adapter *adapter, u32 bank, u32 addr, u8 *
 		while (((read_reg_dw(adapter, 0x700) & 0x80000000) != 0) && (retries > 0)) {
 			mdelay(1);
 			retries--;
-		};
+		}
 
 		if (retries == 0)
 			printk("%s: SRAM timeout\n", __func__);
@@ -110,7 +110,7 @@ static void flex_sram_read(struct adapter *adapter, u32 bank, u32 addr, u8 *buf,
 		while (((read_reg_dw(adapter, 0x700) & 0x80000000) != 0) && (retries > 0)) {
 			mdelay(1);
 			retries--;
-		};
+		}
 
 		if (retries == 0)
 			printk("%s: SRAM timeout\n", __func__);
@@ -122,7 +122,7 @@ static void flex_sram_read(struct adapter *adapter, u32 bank, u32 addr, u8 *buf,
 		while (((read_reg_dw(adapter, 0x700) & 0x80000000) != 0) && (retries > 0)) {
 			mdelay(1);
 			retries--;
-		};
+		}
 
 		if (retries == 0)
 			printk("%s: SRAM timeout\n", __func__);
diff --git a/drivers/media/common/saa7146/saa7146_core.c b/drivers/media/common/saa7146/saa7146_core.c
index bb6ee5191eb1..34b0d0ddeef3 100644
--- a/drivers/media/common/saa7146/saa7146_core.c
+++ b/drivers/media/common/saa7146/saa7146_core.c
@@ -411,7 +411,7 @@ static int saa7146_init_one(struct pci_dev *pci, const struct pci_device_id *ent
 	saa7146_write(dev, MC2, 0xf8000000);
 
 	/* request an interrupt for the saa7146 */
-	err = request_irq(pci->irq, interrupt_hw, IRQF_SHARED | IRQF_DISABLED,
+	err = request_irq(pci->irq, interrupt_hw, IRQF_SHARED,
 			  dev->name, dev);
 	if (err < 0) {
 		ERR("request_irq() failed\n");
@@ -524,8 +524,6 @@ static void saa7146_remove_one(struct pci_dev *pdev)
 	DEB_EE("dev:%p\n", dev);
 
 	dev->ext->detach(dev);
-	/* Zero the PCI drvdata after use. */
-	pci_set_drvdata(pdev, NULL);
 
 	/* shut down all video dma transfers */
 	saa7146_write(dev, MC1, 0x00ff0000);
diff --git a/drivers/media/common/siano/smscoreapi.c b/drivers/media/common/siano/smscoreapi.c
index a142f7942a01..050984c5b1e3 100644
--- a/drivers/media/common/siano/smscoreapi.c
+++ b/drivers/media/common/siano/smscoreapi.c
@@ -922,8 +922,8 @@ static int smscore_load_firmware_family2(struct smscore_device_t *coredev,
 	u32 i, *ptr;
 	u8 *payload = firmware->payload;
 	int rc = 0;
-	firmware->start_address = le32_to_cpu(firmware->start_address);
-	firmware->length = le32_to_cpu(firmware->length);
+	firmware->start_address = le32_to_cpup((__le32 *)&firmware->start_address);
+	firmware->length = le32_to_cpup((__le32 *)&firmware->length);
 
 	mem_address = firmware->start_address;
 
@@ -982,7 +982,7 @@ static int smscore_load_firmware_family2(struct smscore_device_t *coredev,
 	if (rc < 0)
 		goto exit_fw_download;
 
-	sms_err("sending MSG_SMS_DATA_VALIDITY_REQ expecting 0x%x",
+	sms_debug("sending MSG_SMS_DATA_VALIDITY_REQ expecting 0x%x",
 		calc_checksum);
 	SMS_INIT_MSG(&msg->x_msg_header, MSG_SMS_DATA_VALIDITY_REQ,
 			sizeof(msg->x_msg_header) +
@@ -1562,7 +1562,7 @@ void smscore_onresponse(struct smscore_device_t *coredev,
 		{
 			struct sms_msg_data *validity = (struct sms_msg_data *) phdr;
 
-			sms_err("MSG_SMS_DATA_VALIDITY_RES, checksum = 0x%x",
+			sms_debug("MSG_SMS_DATA_VALIDITY_RES, checksum = 0x%x",
 				validity->msg_data[0]);
 			complete(&coredev->data_validity_done);
 			break;
diff --git a/drivers/media/common/siano/smsdvb-main.c b/drivers/media/common/siano/smsdvb-main.c
index 63676a8b024c..85151efdd94c 100644
--- a/drivers/media/common/siano/smsdvb-main.c
+++ b/drivers/media/common/siano/smsdvb-main.c
@@ -44,14 +44,14 @@ module_param_named(debug, sms_dbg, int, 0644);
 MODULE_PARM_DESC(debug, "set debug level (info=1, adv=2 (or-able))");
 
 
-u32 sms_to_guard_interval_table[] = {
+static u32 sms_to_guard_interval_table[] = {
 	[0] = GUARD_INTERVAL_1_32,
 	[1] = GUARD_INTERVAL_1_16,
 	[2] = GUARD_INTERVAL_1_8,
 	[3] = GUARD_INTERVAL_1_4,
 };
 
-u32 sms_to_code_rate_table[] = {
+static u32 sms_to_code_rate_table[] = {
 	[0] = FEC_1_2,
 	[1] = FEC_2_3,
 	[2] = FEC_3_4,
@@ -60,14 +60,14 @@ u32 sms_to_code_rate_table[] = {
 };
 
 
-u32 sms_to_hierarchy_table[] = {
+static u32 sms_to_hierarchy_table[] = {
 	[0] = HIERARCHY_NONE,
 	[1] = HIERARCHY_1,
 	[2] = HIERARCHY_2,
 	[3] = HIERARCHY_4,
 };
 
-u32 sms_to_modulation_table[] = {
+static u32 sms_to_modulation_table[] = {
 	[0] = QPSK,
 	[1] = QAM_16,
 	[2] = QAM_64,
diff --git a/drivers/media/dvb-core/dvb_demux.c b/drivers/media/dvb-core/dvb_demux.c
index 3485655fa082..58de4410c525 100644
--- a/drivers/media/dvb-core/dvb_demux.c
+++ b/drivers/media/dvb-core/dvb_demux.c
@@ -476,7 +476,9 @@ static void dvb_dmx_swfilter_packet(struct dvb_demux *demux, const u8 *buf)
 void dvb_dmx_swfilter_packets(struct dvb_demux *demux, const u8 *buf,
 			      size_t count)
 {
-	spin_lock(&demux->lock);
+	unsigned long flags;
+
+	spin_lock_irqsave(&demux->lock, flags);
 
 	while (count--) {
 		if (buf[0] == 0x47)
@@ -484,7 +486,7 @@ void dvb_dmx_swfilter_packets(struct dvb_demux *demux, const u8 *buf,
 		buf += 188;
 	}
 
-	spin_unlock(&demux->lock);
+	spin_unlock_irqrestore(&demux->lock, flags);
 }
 
 EXPORT_SYMBOL(dvb_dmx_swfilter_packets);
@@ -519,8 +521,9 @@ static inline void _dvb_dmx_swfilter(struct dvb_demux *demux, const u8 *buf,
 {
 	int p = 0, i, j;
 	const u8 *q;
+	unsigned long flags;
 
-	spin_lock(&demux->lock);
+	spin_lock_irqsave(&demux->lock, flags);
 
 	if (demux->tsbufp) { /* tsbuf[0] is now 0x47. */
 		i = demux->tsbufp;
@@ -564,7 +567,7 @@ static inline void _dvb_dmx_swfilter(struct dvb_demux *demux, const u8 *buf,
 	}
 
 bailout:
-	spin_unlock(&demux->lock);
+	spin_unlock_irqrestore(&demux->lock, flags);
 }
 
 void dvb_dmx_swfilter(struct dvb_demux *demux, const u8 *buf, size_t count)
@@ -581,11 +584,13 @@ EXPORT_SYMBOL(dvb_dmx_swfilter_204);
 
 void dvb_dmx_swfilter_raw(struct dvb_demux *demux, const u8 *buf, size_t count)
 {
-	spin_lock(&demux->lock);
+	unsigned long flags;
+
+	spin_lock_irqsave(&demux->lock, flags);
 
 	demux->feed->cb.ts(buf, count, NULL, 0, &demux->feed->feed.ts, DMX_OK);
 
-	spin_unlock(&demux->lock);
+	spin_unlock_irqrestore(&demux->lock, flags);
 }
 EXPORT_SYMBOL(dvb_dmx_swfilter_raw);
 
diff --git a/drivers/media/dvb-frontends/Kconfig b/drivers/media/dvb-frontends/Kconfig
index 0e2ec6f73b05..bddbab43a2df 100644
--- a/drivers/media/dvb-frontends/Kconfig
+++ b/drivers/media/dvb-frontends/Kconfig
@@ -200,6 +200,13 @@ config DVB_CX24116
 	help
 	  A DVB-S/S2 tuner module. Say Y when you want to support this frontend.
 
+config DVB_CX24117
+	tristate "Conexant CX24117 based"
+	depends on DVB_CORE && I2C
+	default m if !MEDIA_SUBDRV_AUTOSELECT
+	help
+	  A Dual DVB-S/S2 tuner module. Say Y when you want to support this frontend.
+
 config DVB_SI21XX
 	tristate "Silicon Labs SI21XX based"
 	depends on DVB_CORE && I2C
diff --git a/drivers/media/dvb-frontends/Makefile b/drivers/media/dvb-frontends/Makefile
index cebc0faffab5..f9cb43d9aed9 100644
--- a/drivers/media/dvb-frontends/Makefile
+++ b/drivers/media/dvb-frontends/Makefile
@@ -76,6 +76,7 @@ obj-$(CONFIG_DVB_ATBM8830) += atbm8830.o
 obj-$(CONFIG_DVB_DUMMY_FE) += dvb_dummy_fe.o
 obj-$(CONFIG_DVB_AF9013) += af9013.o
 obj-$(CONFIG_DVB_CX24116) += cx24116.o
+obj-$(CONFIG_DVB_CX24117) += cx24117.o
 obj-$(CONFIG_DVB_SI21XX) += si21xx.o
 obj-$(CONFIG_DVB_STV0288) += stv0288.o
 obj-$(CONFIG_DVB_STB6000) += stb6000.o
diff --git a/drivers/media/dvb-frontends/af9013.c b/drivers/media/dvb-frontends/af9013.c
index a204f2828820..fb504f1e9125 100644
--- a/drivers/media/dvb-frontends/af9013.c
+++ b/drivers/media/dvb-frontends/af9013.c
@@ -24,6 +24,9 @@
 
 #include "af9013_priv.h"
 
+/* Max transfer size done by I2C transfer functions */
+#define MAX_XFER_SIZE  64
+
 struct af9013_state {
 	struct i2c_adapter *i2c;
 	struct dvb_frontend fe;
@@ -50,16 +53,23 @@ static int af9013_wr_regs_i2c(struct af9013_state *priv, u8 mbox, u16 reg,
 	const u8 *val, int len)
 {
 	int ret;
-	u8 buf[3+len];
+	u8 buf[MAX_XFER_SIZE];
 	struct i2c_msg msg[1] = {
 		{
 			.addr = priv->config.i2c_addr,
 			.flags = 0,
-			.len = sizeof(buf),
+			.len = 3 + len,
 			.buf = buf,
 		}
 	};
 
+	if (3 + len > sizeof(buf)) {
+		dev_warn(&priv->i2c->dev,
+			 "%s: i2c wr reg=%04x: len=%d is too big!\n",
+			 KBUILD_MODNAME, reg, len);
+		return -EINVAL;
+	}
+
 	buf[0] = (reg >> 8) & 0xff;
 	buf[1] = (reg >> 0) & 0xff;
 	buf[2] = mbox;
diff --git a/drivers/media/dvb-frontends/af9033.c b/drivers/media/dvb-frontends/af9033.c
index a777b4b944eb..30ee59052157 100644
--- a/drivers/media/dvb-frontends/af9033.c
+++ b/drivers/media/dvb-frontends/af9033.c
@@ -21,6 +21,9 @@
 
 #include "af9033_priv.h"
 
+/* Max transfer size done by I2C transfer functions */
+#define MAX_XFER_SIZE  64
+
 struct af9033_state {
 	struct i2c_adapter *i2c;
 	struct dvb_frontend fe;
@@ -40,16 +43,23 @@ static int af9033_wr_regs(struct af9033_state *state, u32 reg, const u8 *val,
 		int len)
 {
 	int ret;
-	u8 buf[3 + len];
+	u8 buf[MAX_XFER_SIZE];
 	struct i2c_msg msg[1] = {
 		{
 			.addr = state->cfg.i2c_addr,
 			.flags = 0,
-			.len = sizeof(buf),
+			.len = 3 + len,
 			.buf = buf,
 		}
 	};
 
+	if (3 + len > sizeof(buf)) {
+		dev_warn(&state->i2c->dev,
+			 "%s: i2c wr reg=%04x: len=%d is too big!\n",
+			 KBUILD_MODNAME, reg, len);
+		return -EINVAL;
+	}
+
 	buf[0] = (reg >> 16) & 0xff;
 	buf[1] = (reg >>  8) & 0xff;
 	buf[2] = (reg >>  0) & 0xff;
@@ -161,7 +171,14 @@ static int af9033_wr_reg_val_tab(struct af9033_state *state,
 		const struct reg_val *tab, int tab_len)
 {
 	int ret, i, j;
-	u8 buf[tab_len];
+	u8 buf[MAX_XFER_SIZE];
+
+	if (tab_len > sizeof(buf)) {
+		dev_warn(&state->i2c->dev,
+			 "%s: i2c wr len=%d is too big!\n",
+			 KBUILD_MODNAME, tab_len);
+		return -EINVAL;
+	}
 
 	dev_dbg(&state->i2c->dev, "%s: tab_len=%d\n", __func__, tab_len);
 
diff --git a/drivers/media/dvb-frontends/bcm3510.c b/drivers/media/dvb-frontends/bcm3510.c
index 1b77909c0c71..39a29dd29519 100644
--- a/drivers/media/dvb-frontends/bcm3510.c
+++ b/drivers/media/dvb-frontends/bcm3510.c
@@ -44,6 +44,9 @@
 #include "bcm3510.h"
 #include "bcm3510_priv.h"
 
+/* Max transfer size done by bcm3510_do_hab_cmd() function */
+#define MAX_XFER_SIZE	128
+
 struct bcm3510_state {
 
 	struct i2c_adapter* i2c;
@@ -201,9 +204,19 @@ static int bcm3510_hab_send_request(struct bcm3510_state *st, u8 *buf, int len)
 
 static int bcm3510_do_hab_cmd(struct bcm3510_state *st, u8 cmd, u8 msgid, u8 *obuf, u8 olen, u8 *ibuf, u8 ilen)
 {
-	u8 ob[olen+2],ib[ilen+2];
+	u8 ob[MAX_XFER_SIZE], ib[MAX_XFER_SIZE];
 	int ret = 0;
 
+	if (ilen + 2 > sizeof(ib)) {
+		deb_hab("do_hab_cmd: ilen=%d is too big!\n", ilen);
+		return -EINVAL;
+	}
+
+	if (olen + 2 > sizeof(ob)) {
+		deb_hab("do_hab_cmd: olen=%d is too big!\n", olen);
+		return -EINVAL;
+	}
+
 	ob[0] = cmd;
 	ob[1] = msgid;
 	memcpy(&ob[2],obuf,olen);
diff --git a/drivers/media/dvb-frontends/cx24110.c b/drivers/media/dvb-frontends/cx24110.c
index 0cd6927e654c..95b981cd7115 100644
--- a/drivers/media/dvb-frontends/cx24110.c
+++ b/drivers/media/dvb-frontends/cx24110.c
@@ -378,7 +378,7 @@ static int cx24110_set_voltage (struct dvb_frontend* fe, fe_sec_voltage_t voltag
 		return cx24110_writereg(state,0x76,(cx24110_readreg(state,0x76)&0x3b)|0x40);
 	default:
 		return -EINVAL;
-	};
+	}
 }
 
 static int cx24110_diseqc_send_burst(struct dvb_frontend* fe, fe_sec_mini_cmd_t burst)
diff --git a/drivers/media/dvb-frontends/cx24117.c b/drivers/media/dvb-frontends/cx24117.c
new file mode 100644
index 000000000000..476b422ccf19
--- /dev/null
+++ b/drivers/media/dvb-frontends/cx24117.c
@@ -0,0 +1,1650 @@
+/*
+    Conexant cx24117/cx24132 - Dual DVBS/S2 Satellite demod/tuner driver
+
+    Copyright (C) 2013 Luis Alves <ljalvs@gmail.com>
+	July, 6th 2013
+	    First release based on cx24116 driver by:
+	    Steven Toth and Georg Acher, Darron Broad, Igor Liplianin
+	    Cards currently supported:
+		TBS6980 - Dual DVBS/S2 PCIe card
+		TBS6981 - Dual DVBS/S2 PCIe card
+
+    This program is free software; you can redistribute it and/or modify
+    it under the terms of the GNU General Public License as published by
+    the Free Software Foundation; either version 2 of the License, or
+    (at your option) any later version.
+
+    This program is distributed in the hope that it will be useful,
+    but WITHOUT ANY WARRANTY; without even the implied warranty of
+    MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+    GNU General Public License for more details.
+
+    You should have received a copy of the GNU General Public License
+    along with this program; if not, write to the Free Software
+    Foundation, Inc., 675 Mass Ave, Cambridge, MA 02139, USA.
+*/
+
+#include <linux/slab.h>
+#include <linux/kernel.h>
+#include <linux/module.h>
+#include <linux/moduleparam.h>
+#include <linux/init.h>
+#include <linux/firmware.h>
+
+#include "tuner-i2c.h"
+#include "dvb_frontend.h"
+#include "cx24117.h"
+
+
+#define CX24117_DEFAULT_FIRMWARE "dvb-fe-cx24117.fw"
+#define CX24117_SEARCH_RANGE_KHZ 5000
+
+/* known registers */
+#define CX24117_REG_COMMAND      (0x00)      /* command buffer */
+#define CX24117_REG_EXECUTE      (0x1f)      /* execute command */
+
+#define CX24117_REG_FREQ3_0      (0x34)      /* frequency */
+#define CX24117_REG_FREQ2_0      (0x35)
+#define CX24117_REG_FREQ1_0      (0x36)
+#define CX24117_REG_STATE0       (0x39)
+#define CX24117_REG_SSTATUS0     (0x3a)      /* demod0 signal high / status */
+#define CX24117_REG_SIGNAL0      (0x3b)
+#define CX24117_REG_FREQ5_0      (0x3c)      /* +-freq */
+#define CX24117_REG_FREQ6_0      (0x3d)
+#define CX24117_REG_SRATE2_0     (0x3e)      /* +- 1000 * srate */
+#define CX24117_REG_SRATE1_0     (0x3f)
+#define CX24117_REG_QUALITY2_0   (0x40)
+#define CX24117_REG_QUALITY1_0   (0x41)
+
+#define CX24117_REG_BER4_0       (0x47)
+#define CX24117_REG_BER3_0       (0x48)
+#define CX24117_REG_BER2_0       (0x49)
+#define CX24117_REG_BER1_0       (0x4a)
+#define CX24117_REG_DVBS_UCB2_0  (0x4b)
+#define CX24117_REG_DVBS_UCB1_0  (0x4c)
+#define CX24117_REG_DVBS2_UCB2_0 (0x50)
+#define CX24117_REG_DVBS2_UCB1_0 (0x51)
+#define CX24117_REG_QSTATUS0     (0x93)
+#define CX24117_REG_CLKDIV0      (0xe6)
+#define CX24117_REG_RATEDIV0     (0xf0)
+
+
+#define CX24117_REG_FREQ3_1      (0x55)      /* frequency */
+#define CX24117_REG_FREQ2_1      (0x56)
+#define CX24117_REG_FREQ1_1      (0x57)
+#define CX24117_REG_STATE1       (0x5a)
+#define CX24117_REG_SSTATUS1     (0x5b)      /* demod1 signal high / status */
+#define CX24117_REG_SIGNAL1      (0x5c)
+#define CX24117_REG_FREQ5_1      (0x5d)      /* +- freq */
+#define CX24117_REG_FREQ4_1      (0x5e)
+#define CX24117_REG_SRATE2_1     (0x5f)
+#define CX24117_REG_SRATE1_1     (0x60)
+#define CX24117_REG_QUALITY2_1   (0x61)
+#define CX24117_REG_QUALITY1_1   (0x62)
+#define CX24117_REG_BER4_1       (0x68)
+#define CX24117_REG_BER3_1       (0x69)
+#define CX24117_REG_BER2_1       (0x6a)
+#define CX24117_REG_BER1_1       (0x6b)
+#define CX24117_REG_DVBS_UCB2_1  (0x6c)
+#define CX24117_REG_DVBS_UCB1_1  (0x6d)
+#define CX24117_REG_DVBS2_UCB2_1 (0x71)
+#define CX24117_REG_DVBS2_UCB1_1 (0x72)
+#define CX24117_REG_QSTATUS1     (0x9f)
+#define CX24117_REG_CLKDIV1      (0xe7)
+#define CX24117_REG_RATEDIV1     (0xf1)
+
+
+/* arg buffer size */
+#define CX24117_ARGLEN       (0x1e)
+
+/* rolloff */
+#define CX24117_ROLLOFF_020  (0x00)
+#define CX24117_ROLLOFF_025  (0x01)
+#define CX24117_ROLLOFF_035  (0x02)
+
+/* pilot bit */
+#define CX24117_PILOT_OFF    (0x00)
+#define CX24117_PILOT_ON     (0x40)
+#define CX24117_PILOT_AUTO   (0x80)
+
+/* signal status */
+#define CX24117_HAS_SIGNAL   (0x01)
+#define CX24117_HAS_CARRIER  (0x02)
+#define CX24117_HAS_VITERBI  (0x04)
+#define CX24117_HAS_SYNCLOCK (0x08)
+#define CX24117_STATUS_MASK  (0x0f)
+#define CX24117_SIGNAL_MASK  (0xc0)
+
+
+/* arg offset for DiSEqC */
+#define CX24117_DISEQC_DEMOD  (1)
+#define CX24117_DISEQC_BURST  (2)
+#define CX24117_DISEQC_ARG3_2 (3)   /* unknown value=2 */
+#define CX24117_DISEQC_ARG4_0 (4)   /* unknown value=0 */
+#define CX24117_DISEQC_ARG5_0 (5)   /* unknown value=0 */
+#define CX24117_DISEQC_MSGLEN (6)
+#define CX24117_DISEQC_MSGOFS (7)
+
+/* DiSEqC burst */
+#define CX24117_DISEQC_MINI_A (0)
+#define CX24117_DISEQC_MINI_B (1)
+
+
+#define CX24117_PNE	(0) /* 0 disabled / 2 enabled */
+#define CX24117_OCC	(1) /* 0 disabled / 1 enabled */
+
+
+enum cmds {
+	CMD_SET_VCO     = 0x10,
+	CMD_TUNEREQUEST = 0x11,
+	CMD_MPEGCONFIG  = 0x13,
+	CMD_TUNERINIT   = 0x14,
+	CMD_LNBSEND     = 0x21, /* Formerly CMD_SEND_DISEQC */
+	CMD_LNBDCLEVEL  = 0x22,
+	CMD_SET_TONE    = 0x23,
+	CMD_UPDFWVERS   = 0x35,
+	CMD_TUNERSLEEP  = 0x36,
+};
+
+static LIST_HEAD(hybrid_tuner_instance_list);
+static DEFINE_MUTEX(cx24117_list_mutex);
+
+/* The Demod/Tuner can't easily provide these, we cache them */
+struct cx24117_tuning {
+	u32 frequency;
+	u32 symbol_rate;
+	fe_spectral_inversion_t inversion;
+	fe_code_rate_t fec;
+
+	fe_delivery_system_t delsys;
+	fe_modulation_t modulation;
+	fe_pilot_t pilot;
+	fe_rolloff_t rolloff;
+
+	/* Demod values */
+	u8 fec_val;
+	u8 fec_mask;
+	u8 inversion_val;
+	u8 pilot_val;
+	u8 rolloff_val;
+};
+
+/* Basic commands that are sent to the firmware */
+struct cx24117_cmd {
+	u8 len;
+	u8 args[CX24117_ARGLEN];
+};
+
+/* common to both fe's */
+struct cx24117_priv {
+	u8 demod_address;
+	struct i2c_adapter *i2c;
+	u8 skip_fw_load;
+	struct mutex fe_lock;
+
+	/* Used for sharing this struct between demods */
+	struct tuner_i2c_props i2c_props;
+	struct list_head hybrid_tuner_instance_list;
+};
+
+/* one per each fe */
+struct cx24117_state {
+	struct cx24117_priv *priv;
+	struct dvb_frontend frontend;
+
+	struct cx24117_tuning dcur;
+	struct cx24117_tuning dnxt;
+	struct cx24117_cmd dsec_cmd;
+
+	int demod;
+};
+
+/* modfec (modulation and FEC) lookup table */
+/* Check cx24116.c for a detailed description of each field */
+static struct cx24117_modfec {
+	fe_delivery_system_t delivery_system;
+	fe_modulation_t modulation;
+	fe_code_rate_t fec;
+	u8 mask;	/* In DVBS mode this is used to autodetect */
+	u8 val;		/* Passed to the firmware to indicate mode selection */
+} cx24117_modfec_modes[] = {
+	/* QPSK. For unknown rates we set hardware to auto detect 0xfe 0x30 */
+
+	/*mod   fec       mask  val */
+	{ SYS_DVBS, QPSK, FEC_NONE, 0xfe, 0x30 },
+	{ SYS_DVBS, QPSK, FEC_1_2,  0x02, 0x2e }, /* 00000010 00101110 */
+	{ SYS_DVBS, QPSK, FEC_2_3,  0x04, 0x2f }, /* 00000100 00101111 */
+	{ SYS_DVBS, QPSK, FEC_3_4,  0x08, 0x30 }, /* 00001000 00110000 */
+	{ SYS_DVBS, QPSK, FEC_4_5,  0xfe, 0x30 }, /* 000?0000 ?        */
+	{ SYS_DVBS, QPSK, FEC_5_6,  0x20, 0x31 }, /* 00100000 00110001 */
+	{ SYS_DVBS, QPSK, FEC_6_7,  0xfe, 0x30 }, /* 0?000000 ?        */
+	{ SYS_DVBS, QPSK, FEC_7_8,  0x80, 0x32 }, /* 10000000 00110010 */
+	{ SYS_DVBS, QPSK, FEC_8_9,  0xfe, 0x30 }, /* 0000000? ?        */
+	{ SYS_DVBS, QPSK, FEC_AUTO, 0xfe, 0x30 },
+	/* NBC-QPSK */
+	{ SYS_DVBS2, QPSK, FEC_NONE, 0x00, 0x00 },
+	{ SYS_DVBS2, QPSK, FEC_1_2,  0x00, 0x04 },
+	{ SYS_DVBS2, QPSK, FEC_3_5,  0x00, 0x05 },
+	{ SYS_DVBS2, QPSK, FEC_2_3,  0x00, 0x06 },
+	{ SYS_DVBS2, QPSK, FEC_3_4,  0x00, 0x07 },
+	{ SYS_DVBS2, QPSK, FEC_4_5,  0x00, 0x08 },
+	{ SYS_DVBS2, QPSK, FEC_5_6,  0x00, 0x09 },
+	{ SYS_DVBS2, QPSK, FEC_8_9,  0x00, 0x0a },
+	{ SYS_DVBS2, QPSK, FEC_9_10, 0x00, 0x0b },
+	{ SYS_DVBS2, QPSK, FEC_AUTO, 0x00, 0x00 },
+	/* 8PSK */
+	{ SYS_DVBS2, PSK_8, FEC_NONE, 0x00, 0x00 },
+	{ SYS_DVBS2, PSK_8, FEC_3_5,  0x00, 0x0c },
+	{ SYS_DVBS2, PSK_8, FEC_2_3,  0x00, 0x0d },
+	{ SYS_DVBS2, PSK_8, FEC_3_4,  0x00, 0x0e },
+	{ SYS_DVBS2, PSK_8, FEC_5_6,  0x00, 0x0f },
+	{ SYS_DVBS2, PSK_8, FEC_8_9,  0x00, 0x10 },
+	{ SYS_DVBS2, PSK_8, FEC_9_10, 0x00, 0x11 },
+	{ SYS_DVBS2, PSK_8, FEC_AUTO, 0x00, 0x00 },
+	/*
+	 * 'val' can be found in the FECSTATUS register when tuning.
+	 * FECSTATUS will give the actual FEC in use if tuning was successful.
+	 */
+};
+
+
+static int cx24117_writereg(struct cx24117_state *state, u8 reg, u8 data)
+{
+	u8 buf[] = { reg, data };
+	struct i2c_msg msg = { .addr = state->priv->demod_address,
+		.flags = 0, .buf = buf, .len = 2 };
+	int ret;
+
+	dev_dbg(&state->priv->i2c->dev,
+			"%s() demod%d i2c wr @0x%02x=0x%02x\n",
+			__func__, state->demod, reg, data);
+
+	ret = i2c_transfer(state->priv->i2c, &msg, 1);
+	if (ret < 0) {
+		dev_warn(&state->priv->i2c->dev,
+			"%s: demod%d i2c wr err(%i) @0x%02x=0x%02x\n",
+			KBUILD_MODNAME, state->demod, ret, reg, data);
+		return ret;
+	}
+	return 0;
+}
+
+static int cx24117_writecmd(struct cx24117_state *state,
+	struct cx24117_cmd *cmd)
+{
+	struct i2c_msg msg;
+	u8 buf[CX24117_ARGLEN+1];
+	int ret;
+
+	dev_dbg(&state->priv->i2c->dev,
+			"%s() demod%d i2c wr cmd len=%d\n",
+			__func__, state->demod, cmd->len);
+
+	buf[0] = CX24117_REG_COMMAND;
+	memcpy(&buf[1], cmd->args, cmd->len);
+
+	msg.addr = state->priv->demod_address;
+	msg.flags = 0;
+	msg.len = cmd->len+1;
+	msg.buf = buf;
+	ret = i2c_transfer(state->priv->i2c, &msg, 1);
+	if (ret < 0) {
+		dev_warn(&state->priv->i2c->dev,
+			"%s: demod%d i2c wr cmd err(%i) len=%d\n",
+			KBUILD_MODNAME, state->demod, ret, cmd->len);
+		return ret;
+	}
+	return 0;
+}
+
+static int cx24117_readreg(struct cx24117_state *state, u8 reg)
+{
+	int ret;
+	u8 recv = 0;
+	struct i2c_msg msg[] = {
+		{ .addr = state->priv->demod_address, .flags = 0,
+			.buf = &reg, .len = 1 },
+		{ .addr = state->priv->demod_address, .flags = I2C_M_RD,
+			.buf = &recv, .len = 1 }
+	};
+
+	ret = i2c_transfer(state->priv->i2c, msg, 2);
+	if (ret < 0) {
+		dev_warn(&state->priv->i2c->dev,
+			"%s: demod%d i2c rd err(%d) @0x%x\n",
+			KBUILD_MODNAME, state->demod, ret, reg);
+		return ret;
+	}
+
+	dev_dbg(&state->priv->i2c->dev, "%s() demod%d i2c rd @0x%02x=0x%02x\n",
+		__func__, state->demod, reg, recv);
+
+	return recv;
+}
+
+static int cx24117_readregN(struct cx24117_state *state,
+	u8 reg, u8 *buf, int len)
+{
+	int ret;
+	struct i2c_msg msg[] = {
+		{ .addr = state->priv->demod_address, .flags = 0,
+			.buf = &reg, .len = 1 },
+		{ .addr = state->priv->demod_address, .flags = I2C_M_RD,
+			.buf = buf, .len = len }
+	};
+
+	ret = i2c_transfer(state->priv->i2c, msg, 2);
+	if (ret < 0) {
+		dev_warn(&state->priv->i2c->dev,
+			"%s: demod%d i2c rd err(%d) @0x%x\n",
+			KBUILD_MODNAME, state->demod, ret, reg);
+		return ret;
+	}
+	return 0;
+}
+
+static int cx24117_set_inversion(struct cx24117_state *state,
+	fe_spectral_inversion_t inversion)
+{
+	dev_dbg(&state->priv->i2c->dev, "%s(%d) demod%d\n",
+		__func__, inversion, state->demod);
+
+	switch (inversion) {
+	case INVERSION_OFF:
+		state->dnxt.inversion_val = 0x00;
+		break;
+	case INVERSION_ON:
+		state->dnxt.inversion_val = 0x04;
+		break;
+	case INVERSION_AUTO:
+		state->dnxt.inversion_val = 0x0C;
+		break;
+	default:
+		return -EINVAL;
+	}
+
+	state->dnxt.inversion = inversion;
+
+	return 0;
+}
+
+static int cx24117_lookup_fecmod(struct cx24117_state *state,
+	fe_delivery_system_t d, fe_modulation_t m, fe_code_rate_t f)
+{
+	int i, ret = -EINVAL;
+
+	dev_dbg(&state->priv->i2c->dev,
+		"%s(demod(0x%02x,0x%02x) demod%d\n",
+		__func__, m, f, state->demod);
+
+	for (i = 0; i < ARRAY_SIZE(cx24117_modfec_modes); i++) {
+		if ((d == cx24117_modfec_modes[i].delivery_system) &&
+			(m == cx24117_modfec_modes[i].modulation) &&
+			(f == cx24117_modfec_modes[i].fec)) {
+				ret = i;
+				break;
+			}
+	}
+
+	return ret;
+}
+
+static int cx24117_set_fec(struct cx24117_state *state,
+	fe_delivery_system_t delsys, fe_modulation_t mod, fe_code_rate_t fec)
+{
+	int ret;
+
+	dev_dbg(&state->priv->i2c->dev,
+		"%s(0x%02x,0x%02x) demod%d\n",
+		__func__, mod, fec, state->demod);
+
+	ret = cx24117_lookup_fecmod(state, delsys, mod, fec);
+	if (ret < 0)
+		return ret;
+
+	state->dnxt.fec = fec;
+	state->dnxt.fec_val = cx24117_modfec_modes[ret].val;
+	state->dnxt.fec_mask = cx24117_modfec_modes[ret].mask;
+	dev_dbg(&state->priv->i2c->dev,
+		"%s() demod%d mask/val = 0x%02x/0x%02x\n", __func__,
+		state->demod, state->dnxt.fec_mask, state->dnxt.fec_val);
+
+	return 0;
+}
+
+static int cx24117_set_symbolrate(struct cx24117_state *state, u32 rate)
+{
+	dev_dbg(&state->priv->i2c->dev, "%s(%d) demod%d\n",
+		__func__, rate, state->demod);
+
+	state->dnxt.symbol_rate = rate;
+
+	dev_dbg(&state->priv->i2c->dev,
+		"%s() demod%d symbol_rate = %d\n",
+		__func__, state->demod, rate);
+
+	return 0;
+}
+
+static int cx24117_load_firmware(struct dvb_frontend *fe,
+	const struct firmware *fw);
+
+static int cx24117_firmware_ondemand(struct dvb_frontend *fe)
+{
+	struct cx24117_state *state = fe->demodulator_priv;
+	const struct firmware *fw;
+	int ret = 0;
+
+	dev_dbg(&state->priv->i2c->dev, "%s() demod%d skip_fw_load=%d\n",
+		__func__, state->demod, state->priv->skip_fw_load);
+
+	if (state->priv->skip_fw_load)
+		return 0;
+
+	/* check if firmware if already running */
+	if (cx24117_readreg(state, 0xeb) != 0xa) {
+		/* Load firmware */
+		/* request the firmware, this will block until loaded */
+		dev_dbg(&state->priv->i2c->dev,
+			"%s: Waiting for firmware upload (%s)...\n",
+			__func__, CX24117_DEFAULT_FIRMWARE);
+		ret = request_firmware(&fw, CX24117_DEFAULT_FIRMWARE,
+			state->priv->i2c->dev.parent);
+		dev_dbg(&state->priv->i2c->dev,
+			"%s: Waiting for firmware upload(2)...\n", __func__);
+		if (ret) {
+			dev_err(&state->priv->i2c->dev,
+				"%s: No firmware uploaded "
+				"(timeout or file not found?)\n", __func__);
+			return ret;
+		}
+
+		/* Make sure we don't recurse back through here
+		 * during loading */
+		state->priv->skip_fw_load = 1;
+
+		ret = cx24117_load_firmware(fe, fw);
+		if (ret)
+			dev_err(&state->priv->i2c->dev,
+				"%s: Writing firmware failed\n", __func__);
+		release_firmware(fw);
+
+		dev_info(&state->priv->i2c->dev,
+			"%s: Firmware upload %s\n", __func__,
+			ret == 0 ? "complete" : "failed");
+
+		/* Ensure firmware is always loaded if required */
+		state->priv->skip_fw_load = 0;
+	}
+
+	return ret;
+}
+
+/* Take a basic firmware command structure, format it
+ * and forward it for processing
+ */
+static int cx24117_cmd_execute_nolock(struct dvb_frontend *fe,
+	struct cx24117_cmd *cmd)
+{
+	struct cx24117_state *state = fe->demodulator_priv;
+	int i, ret;
+
+	dev_dbg(&state->priv->i2c->dev, "%s() demod%d\n",
+		__func__, state->demod);
+
+	/* Load the firmware if required */
+	ret = cx24117_firmware_ondemand(fe);
+	if (ret != 0)
+		return ret;
+
+	/* Write the command */
+	cx24117_writecmd(state, cmd);
+
+	/* Start execution and wait for cmd to terminate */
+	cx24117_writereg(state, CX24117_REG_EXECUTE, 0x01);
+	i = 0;
+	while (cx24117_readreg(state, CX24117_REG_EXECUTE)) {
+		msleep(20);
+		if (i++ > 40) {
+			/* Avoid looping forever if the firmware does
+				not respond */
+			dev_warn(&state->priv->i2c->dev,
+				"%s() Firmware not responding\n", __func__);
+			return -EIO;
+		}
+	}
+	return 0;
+}
+
+static int cx24117_cmd_execute(struct dvb_frontend *fe, struct cx24117_cmd *cmd)
+{
+	struct cx24117_state *state = fe->demodulator_priv;
+	int ret;
+
+	mutex_lock(&state->priv->fe_lock);
+	ret = cx24117_cmd_execute_nolock(fe, cmd);
+	mutex_unlock(&state->priv->fe_lock);
+
+	return ret;
+}
+
+static int cx24117_load_firmware(struct dvb_frontend *fe,
+	const struct firmware *fw)
+{
+	struct cx24117_state *state = fe->demodulator_priv;
+	struct cx24117_cmd cmd;
+	int i, ret;
+	unsigned char vers[4];
+
+	struct i2c_msg msg;
+	u8 *buf;
+
+	dev_dbg(&state->priv->i2c->dev,
+		"%s() demod%d FW is %zu bytes (%02x %02x .. %02x %02x)\n",
+		__func__, state->demod, fw->size, fw->data[0], fw->data[1],
+		fw->data[fw->size - 2], fw->data[fw->size - 1]);
+
+	cx24117_writereg(state, 0xea, 0x00);
+	cx24117_writereg(state, 0xea, 0x01);
+	cx24117_writereg(state, 0xea, 0x00);
+
+	cx24117_writereg(state, 0xce, 0x92);
+
+	cx24117_writereg(state, 0xfb, 0x00);
+	cx24117_writereg(state, 0xfc, 0x00);
+
+	cx24117_writereg(state, 0xc3, 0x04);
+	cx24117_writereg(state, 0xc4, 0x04);
+
+	cx24117_writereg(state, 0xce, 0x00);
+	cx24117_writereg(state, 0xcf, 0x00);
+
+	cx24117_writereg(state, 0xea, 0x00);
+	cx24117_writereg(state, 0xeb, 0x0c);
+	cx24117_writereg(state, 0xec, 0x06);
+	cx24117_writereg(state, 0xed, 0x05);
+	cx24117_writereg(state, 0xee, 0x03);
+	cx24117_writereg(state, 0xef, 0x05);
+
+	cx24117_writereg(state, 0xf3, 0x03);
+	cx24117_writereg(state, 0xf4, 0x44);
+
+	cx24117_writereg(state, CX24117_REG_RATEDIV0, 0x04);
+	cx24117_writereg(state, CX24117_REG_CLKDIV0, 0x02);
+
+	cx24117_writereg(state, CX24117_REG_RATEDIV1, 0x04);
+	cx24117_writereg(state, CX24117_REG_CLKDIV1, 0x02);
+
+	cx24117_writereg(state, 0xf2, 0x04);
+	cx24117_writereg(state, 0xe8, 0x02);
+	cx24117_writereg(state, 0xea, 0x01);
+	cx24117_writereg(state, 0xc8, 0x00);
+	cx24117_writereg(state, 0xc9, 0x00);
+	cx24117_writereg(state, 0xca, 0x00);
+	cx24117_writereg(state, 0xcb, 0x00);
+	cx24117_writereg(state, 0xcc, 0x00);
+	cx24117_writereg(state, 0xcd, 0x00);
+	cx24117_writereg(state, 0xe4, 0x03);
+	cx24117_writereg(state, 0xeb, 0x0a);
+
+	cx24117_writereg(state, 0xfb, 0x00);
+	cx24117_writereg(state, 0xe0, 0x76);
+	cx24117_writereg(state, 0xf7, 0x81);
+	cx24117_writereg(state, 0xf8, 0x00);
+	cx24117_writereg(state, 0xf9, 0x00);
+
+	buf = kmalloc(fw->size + 1, GFP_KERNEL);
+	if (buf == NULL) {
+		state->priv->skip_fw_load = 0;
+		return -ENOMEM;
+	}
+
+	/* fw upload reg */
+	buf[0] = 0xfa;
+	memcpy(&buf[1], fw->data, fw->size);
+
+	/* prepare i2c message to send */
+	msg.addr = state->priv->demod_address;
+	msg.flags = 0;
+	msg.len = fw->size + 1;
+	msg.buf = buf;
+
+	/* send fw */
+	ret = i2c_transfer(state->priv->i2c, &msg, 1);
+	if (ret < 0)
+		return ret;
+
+	kfree(buf);
+
+	cx24117_writereg(state, 0xf7, 0x0c);
+	cx24117_writereg(state, 0xe0, 0x00);
+
+	/* CMD 1B */
+	cmd.args[0] = 0x1b;
+	cmd.args[1] = 0x00;
+	cmd.args[2] = 0x01;
+	cmd.args[3] = 0x00;
+	cmd.len = 4;
+	ret = cx24117_cmd_execute_nolock(fe, &cmd);
+	if (ret != 0)
+		goto error;
+
+	/* CMD 10 */
+	cmd.args[0] = CMD_SET_VCO;
+	cmd.args[1] = 0x06;
+	cmd.args[2] = 0x2b;
+	cmd.args[3] = 0xd8;
+	cmd.args[4] = 0xa5;
+	cmd.args[5] = 0xee;
+	cmd.args[6] = 0x03;
+	cmd.args[7] = 0x9d;
+	cmd.args[8] = 0xfc;
+	cmd.args[9] = 0x06;
+	cmd.args[10] = 0x02;
+	cmd.args[11] = 0x9d;
+	cmd.args[12] = 0xfc;
+	cmd.len = 13;
+	ret = cx24117_cmd_execute_nolock(fe, &cmd);
+	if (ret != 0)
+		goto error;
+
+	/* CMD 15 */
+	cmd.args[0] = 0x15;
+	cmd.args[1] = 0x00;
+	cmd.args[2] = 0x01;
+	cmd.args[3] = 0x00;
+	cmd.args[4] = 0x00;
+	cmd.args[5] = 0x01;
+	cmd.args[6] = 0x01;
+	cmd.args[7] = 0x01;
+	cmd.args[8] = 0x00;
+	cmd.args[9] = 0x05;
+	cmd.args[10] = 0x02;
+	cmd.args[11] = 0x02;
+	cmd.args[12] = 0x00;
+	cmd.len = 13;
+	ret = cx24117_cmd_execute_nolock(fe, &cmd);
+	if (ret != 0)
+		goto error;
+
+	/* CMD 13 */
+	cmd.args[0] = CMD_MPEGCONFIG;
+	cmd.args[1] = 0x00;
+	cmd.args[2] = 0x00;
+	cmd.args[3] = 0x00;
+	cmd.args[4] = 0x01;
+	cmd.args[5] = 0x00;
+	cmd.len = 6;
+	ret = cx24117_cmd_execute_nolock(fe, &cmd);
+	if (ret != 0)
+		goto error;
+
+	/* CMD 14 */
+	for (i = 0; i < 2; i++) {
+		cmd.args[0] = CMD_TUNERINIT;
+		cmd.args[1] = (u8) i;
+		cmd.args[2] = 0x00;
+		cmd.args[3] = 0x05;
+		cmd.args[4] = 0x00;
+		cmd.args[5] = 0x00;
+		cmd.args[6] = 0x55;
+		cmd.args[7] = 0x00;
+		cmd.len = 8;
+		ret = cx24117_cmd_execute_nolock(fe, &cmd);
+		if (ret != 0)
+			goto error;
+	}
+
+	cx24117_writereg(state, 0xce, 0xc0);
+	cx24117_writereg(state, 0xcf, 0x00);
+	cx24117_writereg(state, 0xe5, 0x04);
+
+	/* Firmware CMD 35: Get firmware version */
+	cmd.args[0] = CMD_UPDFWVERS;
+	cmd.len = 2;
+	for (i = 0; i < 4; i++) {
+		cmd.args[1] = i;
+		ret = cx24117_cmd_execute_nolock(fe, &cmd);
+		if (ret != 0)
+			goto error;
+		vers[i] = cx24117_readreg(state, 0x33);
+	}
+	dev_info(&state->priv->i2c->dev,
+		"%s: FW version %i.%i.%i.%i\n", __func__,
+		vers[0], vers[1], vers[2], vers[3]);
+	return 0;
+error:
+	state->priv->skip_fw_load = 0;
+	dev_err(&state->priv->i2c->dev, "%s() Error running FW.\n", __func__);
+	return ret;
+}
+
+static int cx24117_read_status(struct dvb_frontend *fe, fe_status_t *status)
+{
+	struct cx24117_state *state = fe->demodulator_priv;
+	int lock;
+
+	lock = cx24117_readreg(state,
+		(state->demod == 0) ? CX24117_REG_SSTATUS0 :
+				      CX24117_REG_SSTATUS1) &
+		CX24117_STATUS_MASK;
+
+	dev_dbg(&state->priv->i2c->dev, "%s() demod%d status = 0x%02x\n",
+		__func__, state->demod, lock);
+
+	*status = 0;
+
+	if (lock & CX24117_HAS_SIGNAL)
+		*status |= FE_HAS_SIGNAL;
+	if (lock & CX24117_HAS_CARRIER)
+		*status |= FE_HAS_CARRIER;
+	if (lock & CX24117_HAS_VITERBI)
+		*status |= FE_HAS_VITERBI;
+	if (lock & CX24117_HAS_SYNCLOCK)
+		*status |= FE_HAS_SYNC | FE_HAS_LOCK;
+
+	return 0;
+}
+
+static int cx24117_read_ber(struct dvb_frontend *fe, u32 *ber)
+{
+	struct cx24117_state *state = fe->demodulator_priv;
+	int ret;
+	u8 buf[4];
+	u8 base_reg = (state->demod == 0) ?
+			CX24117_REG_BER4_0 :
+			CX24117_REG_BER4_1;
+
+	ret = cx24117_readregN(state, base_reg, buf, 4);
+	if (ret != 0)
+		return ret;
+
+	*ber = (buf[0] << 24) | (buf[1] << 16) |
+		(buf[1] << 8) | buf[0];
+
+	dev_dbg(&state->priv->i2c->dev, "%s() demod%d ber=0x%04x\n",
+		__func__, state->demod, *ber);
+
+	return 0;
+}
+
+static int cx24117_read_signal_strength(struct dvb_frontend *fe,
+	u16 *signal_strength)
+{
+	struct cx24117_state *state = fe->demodulator_priv;
+	struct cx24117_cmd cmd;
+	int ret;
+	u16 sig_reading;
+	u8 buf[2];
+	u8 reg = (state->demod == 0) ?
+		CX24117_REG_SSTATUS0 : CX24117_REG_SSTATUS1;
+
+	/* Firmware CMD 1A */
+	cmd.args[0] = 0x1a;
+	cmd.args[1] = (u8) state->demod;
+	cmd.len = 2;
+	ret = cx24117_cmd_execute(fe, &cmd);
+	if (ret != 0)
+		return ret;
+
+	ret = cx24117_readregN(state, reg, buf, 2);
+	if (ret != 0)
+		return ret;
+	sig_reading = ((buf[0] & CX24117_SIGNAL_MASK) << 2) | buf[1];
+
+	*signal_strength = -100 * sig_reading + 94324;
+
+	dev_dbg(&state->priv->i2c->dev,
+		"%s() demod%d raw / cooked = 0x%04x / 0x%04x\n",
+		__func__, state->demod, sig_reading, *signal_strength);
+
+	return 0;
+}
+
+static int cx24117_read_snr(struct dvb_frontend *fe, u16 *snr)
+{
+	struct cx24117_state *state = fe->demodulator_priv;
+	int ret;
+	u8 buf[2];
+	u8 reg = (state->demod == 0) ?
+		CX24117_REG_QUALITY2_0 : CX24117_REG_QUALITY2_1;
+
+	ret = cx24117_readregN(state, reg, buf, 2);
+	if (ret != 0)
+		return ret;
+
+	*snr = (buf[0] << 8) | buf[1];
+
+	dev_dbg(&state->priv->i2c->dev,
+		"%s() demod%d snr = 0x%04x\n",
+		__func__, state->demod, *snr);
+
+	return ret;
+}
+
+static int cx24117_read_ucblocks(struct dvb_frontend *fe, u32 *ucblocks)
+{
+	struct cx24117_state *state = fe->demodulator_priv;
+	fe_delivery_system_t delsys = fe->dtv_property_cache.delivery_system;
+	int ret;
+	u8 buf[2];
+	u8 reg = (state->demod == 0) ?
+		CX24117_REG_DVBS_UCB2_0 :
+		CX24117_REG_DVBS_UCB2_1;
+
+	switch (delsys) {
+	case SYS_DVBS:
+		break;
+	case SYS_DVBS2:
+		reg += (CX24117_REG_DVBS2_UCB2_0 - CX24117_REG_DVBS_UCB2_0);
+		break;
+	default:
+		return -EINVAL;
+	}
+
+	ret = cx24117_readregN(state, reg, buf, 2);
+	if (ret != 0)
+		return ret;
+	*ucblocks = (buf[0] << 8) | buf[1];
+
+	dev_dbg(&state->priv->i2c->dev, "%s() demod%d ucb=0x%04x\n",
+		__func__, state->demod, *ucblocks);
+
+	return 0;
+}
+
+/* Overwrite the current tuning params, we are about to tune */
+static void cx24117_clone_params(struct dvb_frontend *fe)
+{
+	struct cx24117_state *state = fe->demodulator_priv;
+	state->dcur = state->dnxt;
+}
+
+/* Wait for LNB */
+static int cx24117_wait_for_lnb(struct dvb_frontend *fe)
+{
+	struct cx24117_state *state = fe->demodulator_priv;
+	int i;
+	u8 val, reg = (state->demod == 0) ? CX24117_REG_QSTATUS0 :
+					    CX24117_REG_QSTATUS1;
+
+	dev_dbg(&state->priv->i2c->dev, "%s() demod%d qstatus = 0x%02x\n",
+		__func__, state->demod, cx24117_readreg(state, reg));
+
+	/* Wait for up to 300 ms */
+	for (i = 0; i < 10; i++) {
+		val = cx24117_readreg(state, reg) & 0x01;
+		if (val != 0)
+			return 0;
+		msleep(30);
+	}
+
+	dev_warn(&state->priv->i2c->dev, "%s: demod%d LNB not ready\n",
+		KBUILD_MODNAME, state->demod);
+
+	return -ETIMEDOUT; /* -EBUSY ? */
+}
+
+static int cx24117_set_voltage(struct dvb_frontend *fe,
+	fe_sec_voltage_t voltage)
+{
+	struct cx24117_state *state = fe->demodulator_priv;
+	struct cx24117_cmd cmd;
+	int ret;
+	u8 reg = (state->demod == 0) ? 0x10 : 0x20;
+
+	dev_dbg(&state->priv->i2c->dev, "%s() demod%d %s\n",
+		__func__, state->demod,
+		voltage == SEC_VOLTAGE_13 ? "SEC_VOLTAGE_13" :
+		voltage == SEC_VOLTAGE_18 ? "SEC_VOLTAGE_18" :
+		"SEC_VOLTAGE_OFF");
+
+	/* CMD 32 */
+	cmd.args[0] = 0x32;
+	cmd.args[1] = reg;
+	cmd.args[2] = reg;
+	cmd.len = 3;
+	ret = cx24117_cmd_execute(fe, &cmd);
+	if (ret)
+		return ret;
+
+	if ((voltage == SEC_VOLTAGE_13) ||
+	    (voltage == SEC_VOLTAGE_18)) {
+		/* CMD 33 */
+		cmd.args[0] = 0x33;
+		cmd.args[1] = reg;
+		cmd.args[2] = reg;
+		cmd.len = 3;
+		ret = cx24117_cmd_execute(fe, &cmd);
+		if (ret != 0)
+			return ret;
+
+		ret = cx24117_wait_for_lnb(fe);
+		if (ret != 0)
+			return ret;
+
+		/* Wait for voltage/min repeat delay */
+		msleep(100);
+
+		/* CMD 22 - CMD_LNBDCLEVEL */
+		cmd.args[0] = CMD_LNBDCLEVEL;
+		cmd.args[1] = state->demod ? 0 : 1;
+		cmd.args[2] = (voltage == SEC_VOLTAGE_18 ? 0x01 : 0x00);
+		cmd.len = 3;
+
+		/* Min delay time before DiSEqC send */
+		msleep(20);
+	} else {
+		cmd.args[0] = 0x33;
+		cmd.args[1] = 0x00;
+		cmd.args[2] = reg;
+		cmd.len = 3;
+	}
+
+	return cx24117_cmd_execute(fe, &cmd);
+}
+
+static int cx24117_set_tone(struct dvb_frontend *fe,
+	fe_sec_tone_mode_t tone)
+{
+	struct cx24117_state *state = fe->demodulator_priv;
+	struct cx24117_cmd cmd;
+	int ret;
+
+	dev_dbg(&state->priv->i2c->dev, "%s(%d) demod%d\n",
+		__func__, state->demod, tone);
+	if ((tone != SEC_TONE_ON) && (tone != SEC_TONE_OFF)) {
+		dev_warn(&state->priv->i2c->dev, "%s: demod%d invalid tone=%d\n",
+			KBUILD_MODNAME, state->demod, tone);
+		return -EINVAL;
+	}
+
+	/* Wait for LNB ready */
+	ret = cx24117_wait_for_lnb(fe);
+	if (ret != 0)
+		return ret;
+
+	/* Min delay time after DiSEqC send */
+	msleep(20);
+
+	/* Set the tone */
+	/* CMD 23 - CMD_SET_TONE */
+	cmd.args[0] = CMD_SET_TONE;
+	cmd.args[1] = (state->demod ? 0 : 1);
+	cmd.args[2] = 0x00;
+	cmd.args[3] = 0x00;
+	cmd.len = 5;
+	switch (tone) {
+	case SEC_TONE_ON:
+		cmd.args[4] = 0x01;
+		break;
+	case SEC_TONE_OFF:
+		cmd.args[4] = 0x00;
+		break;
+	}
+
+	msleep(20);
+
+	return cx24117_cmd_execute(fe, &cmd);
+}
+
+/* Initialise DiSEqC */
+static int cx24117_diseqc_init(struct dvb_frontend *fe)
+{
+	struct cx24117_state *state = fe->demodulator_priv;
+
+	/* Prepare a DiSEqC command */
+	state->dsec_cmd.args[0] = CMD_LNBSEND;
+
+	/* demod */
+	state->dsec_cmd.args[CX24117_DISEQC_DEMOD] = state->demod ? 0 : 1;
+
+	/* DiSEqC burst */
+	state->dsec_cmd.args[CX24117_DISEQC_BURST] = CX24117_DISEQC_MINI_A;
+
+	/* Unknown */
+	state->dsec_cmd.args[CX24117_DISEQC_ARG3_2] = 0x02;
+	state->dsec_cmd.args[CX24117_DISEQC_ARG4_0] = 0x00;
+
+	/* Continuation flag? */
+	state->dsec_cmd.args[CX24117_DISEQC_ARG5_0] = 0x00;
+
+	/* DiSEqC message length */
+	state->dsec_cmd.args[CX24117_DISEQC_MSGLEN] = 0x00;
+
+	/* Command length */
+	state->dsec_cmd.len = 7;
+
+	return 0;
+}
+
+/* Send DiSEqC message */
+static int cx24117_send_diseqc_msg(struct dvb_frontend *fe,
+	struct dvb_diseqc_master_cmd *d)
+{
+	struct cx24117_state *state = fe->demodulator_priv;
+	int i, ret;
+
+	/* Dump DiSEqC message */
+	dev_dbg(&state->priv->i2c->dev, "%s: demod %d (",
+		__func__, state->demod);
+	for (i = 0; i < d->msg_len; i++)
+		dev_dbg(&state->priv->i2c->dev, "0x%02x ", d->msg[i]);
+	dev_dbg(&state->priv->i2c->dev, ")\n");
+
+	/* Validate length */
+	if (d->msg_len > 15)
+		return -EINVAL;
+
+	/* DiSEqC message */
+	for (i = 0; i < d->msg_len; i++)
+		state->dsec_cmd.args[CX24117_DISEQC_MSGOFS + i] = d->msg[i];
+
+	/* DiSEqC message length */
+	state->dsec_cmd.args[CX24117_DISEQC_MSGLEN] = d->msg_len;
+
+	/* Command length */
+	state->dsec_cmd.len = CX24117_DISEQC_MSGOFS +
+		state->dsec_cmd.args[CX24117_DISEQC_MSGLEN];
+
+	/*
+	 * Message is sent with derived else cached burst
+	 *
+	 * WRITE PORT GROUP COMMAND 38
+	 *
+	 * 0/A/A: E0 10 38 F0..F3
+	 * 1/B/B: E0 10 38 F4..F7
+	 * 2/C/A: E0 10 38 F8..FB
+	 * 3/D/B: E0 10 38 FC..FF
+	 *
+	 * databyte[3]= 8421:8421
+	 *              ABCD:WXYZ
+	 *              CLR :SET
+	 *
+	 *              WX= PORT SELECT 0..3    (X=TONEBURST)
+	 *              Y = VOLTAGE             (0=13V, 1=18V)
+	 *              Z = BAND                (0=LOW, 1=HIGH(22K))
+	 */
+	if (d->msg_len >= 4 && d->msg[2] == 0x38)
+		state->dsec_cmd.args[CX24117_DISEQC_BURST] =
+			((d->msg[3] & 4) >> 2);
+
+	dev_dbg(&state->priv->i2c->dev, "%s() demod%d burst=%d\n",
+		__func__, state->demod,
+		state->dsec_cmd.args[CX24117_DISEQC_BURST]);
+
+	/* Wait for LNB ready */
+	ret = cx24117_wait_for_lnb(fe);
+	if (ret != 0)
+		return ret;
+
+	/* Wait for voltage/min repeat delay */
+	msleep(100);
+
+	/* Command */
+	ret = cx24117_cmd_execute(fe, &state->dsec_cmd);
+	if (ret != 0)
+		return ret;
+	/*
+	 * Wait for send
+	 *
+	 * Eutelsat spec:
+	 * >15ms delay          + (XXX determine if FW does this, see set_tone)
+	 *  13.5ms per byte     +
+	 * >15ms delay          +
+	 *  12.5ms burst        +
+	 * >15ms delay            (XXX determine if FW does this, see set_tone)
+	 */
+	msleep((state->dsec_cmd.args[CX24117_DISEQC_MSGLEN] << 4) + 60);
+
+	return 0;
+}
+
+/* Send DiSEqC burst */
+static int cx24117_diseqc_send_burst(struct dvb_frontend *fe,
+	fe_sec_mini_cmd_t burst)
+{
+	struct cx24117_state *state = fe->demodulator_priv;
+
+	dev_dbg(&state->priv->i2c->dev, "%s(%d) demod=%d\n",
+		__func__, burst, state->demod);
+
+	/* DiSEqC burst */
+	if (burst == SEC_MINI_A)
+		state->dsec_cmd.args[CX24117_DISEQC_BURST] =
+			CX24117_DISEQC_MINI_A;
+	else if (burst == SEC_MINI_B)
+		state->dsec_cmd.args[CX24117_DISEQC_BURST] =
+			CX24117_DISEQC_MINI_B;
+	else
+		return -EINVAL;
+
+	return 0;
+}
+
+static int cx24117_get_priv(struct cx24117_priv **priv,
+	struct i2c_adapter *i2c, u8 client_address)
+{
+	int ret;
+
+	mutex_lock(&cx24117_list_mutex);
+	ret = hybrid_tuner_request_state(struct cx24117_priv, (*priv),
+		hybrid_tuner_instance_list, i2c, client_address, "cx24117");
+	mutex_unlock(&cx24117_list_mutex);
+
+	return ret;
+}
+
+static void cx24117_release_priv(struct cx24117_priv *priv)
+{
+	mutex_lock(&cx24117_list_mutex);
+	if (priv != NULL)
+		hybrid_tuner_release_state(priv);
+	mutex_unlock(&cx24117_list_mutex);
+}
+
+static void cx24117_release(struct dvb_frontend *fe)
+{
+	struct cx24117_state *state = fe->demodulator_priv;
+	dev_dbg(&state->priv->i2c->dev, "%s demod%d\n",
+		__func__, state->demod);
+	cx24117_release_priv(state->priv);
+	kfree(state);
+}
+
+static struct dvb_frontend_ops cx24117_ops;
+
+struct dvb_frontend *cx24117_attach(const struct cx24117_config *config,
+	struct i2c_adapter *i2c)
+{
+	struct cx24117_state *state = NULL;
+	struct cx24117_priv *priv = NULL;
+	int demod = 0;
+
+	/* get the common data struct for both demods */
+	demod = cx24117_get_priv(&priv, i2c, config->demod_address);
+
+	switch (demod) {
+	case 0:
+		dev_err(&state->priv->i2c->dev,
+			"%s: Error attaching frontend %d\n",
+			KBUILD_MODNAME, demod);
+		goto error1;
+		break;
+	case 1:
+		/* new priv instance */
+		priv->i2c = i2c;
+		priv->demod_address = config->demod_address;
+		mutex_init(&priv->fe_lock);
+		break;
+	default:
+		/* existing priv instance */
+		break;
+	}
+
+	/* allocate memory for the internal state */
+	state = kzalloc(sizeof(struct cx24117_state), GFP_KERNEL);
+	if (state == NULL)
+		goto error2;
+
+	state->demod = demod - 1;
+	state->priv = priv;
+
+	/* test i2c bus for ack */
+	if (demod == 0) {
+		if (cx24117_readreg(state, 0x00) < 0)
+			goto error3;
+	}
+
+	dev_info(&state->priv->i2c->dev,
+		"%s: Attaching frontend %d\n",
+		KBUILD_MODNAME, state->demod);
+
+	/* create dvb_frontend */
+	memcpy(&state->frontend.ops, &cx24117_ops,
+		sizeof(struct dvb_frontend_ops));
+	state->frontend.demodulator_priv = state;
+	return &state->frontend;
+
+error3:
+	kfree(state);
+error2:
+	cx24117_release_priv(priv);
+error1:
+	return NULL;
+}
+EXPORT_SYMBOL_GPL(cx24117_attach);
+
+/*
+ * Initialise or wake up device
+ *
+ * Power config will reset and load initial firmware if required
+ */
+static int cx24117_initfe(struct dvb_frontend *fe)
+{
+	struct cx24117_state *state = fe->demodulator_priv;
+	struct cx24117_cmd cmd;
+	int ret;
+
+	dev_dbg(&state->priv->i2c->dev, "%s() demod%d\n",
+		__func__, state->demod);
+
+	mutex_lock(&state->priv->fe_lock);
+
+	/* Firmware CMD 36: Power config */
+	cmd.args[0] = CMD_TUNERSLEEP;
+	cmd.args[1] = (state->demod ? 1 : 0);
+	cmd.args[2] = 0;
+	cmd.len = 3;
+	ret = cx24117_cmd_execute_nolock(fe, &cmd);
+	if (ret != 0)
+		goto exit;
+
+	ret = cx24117_diseqc_init(fe);
+	if (ret != 0)
+		goto exit;
+
+	/* CMD 3C */
+	cmd.args[0] = 0x3c;
+	cmd.args[1] = (state->demod ? 1 : 0);
+	cmd.args[2] = 0x10;
+	cmd.args[3] = 0x10;
+	cmd.len = 4;
+	ret = cx24117_cmd_execute_nolock(fe, &cmd);
+	if (ret != 0)
+		goto exit;
+
+	/* CMD 34 */
+	cmd.args[0] = 0x34;
+	cmd.args[1] = (state->demod ? 1 : 0);
+	cmd.args[2] = CX24117_OCC;
+	cmd.len = 3;
+	ret = cx24117_cmd_execute_nolock(fe, &cmd);
+
+exit:
+	mutex_unlock(&state->priv->fe_lock);
+
+	return ret;
+}
+
+/*
+ * Put device to sleep
+ */
+static int cx24117_sleep(struct dvb_frontend *fe)
+{
+	struct cx24117_state *state = fe->demodulator_priv;
+	struct cx24117_cmd cmd;
+
+	dev_dbg(&state->priv->i2c->dev, "%s() demod%d\n",
+		__func__, state->demod);
+
+	/* Firmware CMD 36: Power config */
+	cmd.args[0] = CMD_TUNERSLEEP;
+	cmd.args[1] = (state->demod ? 1 : 0);
+	cmd.args[2] = 1;
+	cmd.len = 3;
+	return cx24117_cmd_execute(fe, &cmd);
+}
+
+/* dvb-core told us to tune, the tv property cache will be complete,
+ * it's safe for is to pull values and use them for tuning purposes.
+ */
+static int cx24117_set_frontend(struct dvb_frontend *fe)
+{
+	struct cx24117_state *state = fe->demodulator_priv;
+	struct dtv_frontend_properties *c = &fe->dtv_property_cache;
+	struct cx24117_cmd cmd;
+	fe_status_t tunerstat;
+	int i, status, ret, retune = 1;
+	u8 reg_clkdiv, reg_ratediv;
+
+	dev_dbg(&state->priv->i2c->dev, "%s() demod%d\n",
+		__func__, state->demod);
+
+	switch (c->delivery_system) {
+	case SYS_DVBS:
+		dev_dbg(&state->priv->i2c->dev, "%s() demod%d DVB-S\n",
+			__func__, state->demod);
+
+		/* Only QPSK is supported for DVB-S */
+		if (c->modulation != QPSK) {
+			dev_dbg(&state->priv->i2c->dev,
+				"%s() demod%d unsupported modulation (%d)\n",
+				__func__, state->demod, c->modulation);
+			return -EINVAL;
+		}
+
+		/* Pilot doesn't exist in DVB-S, turn bit off */
+		state->dnxt.pilot_val = CX24117_PILOT_OFF;
+
+		/* DVB-S only supports 0.35 */
+		state->dnxt.rolloff_val = CX24117_ROLLOFF_035;
+		break;
+
+	case SYS_DVBS2:
+		dev_dbg(&state->priv->i2c->dev, "%s() demod%d DVB-S2\n",
+			__func__, state->demod);
+
+		/*
+		 * NBC 8PSK/QPSK with DVB-S is supported for DVB-S2,
+		 * but not hardware auto detection
+		 */
+		if (c->modulation != PSK_8 && c->modulation != QPSK) {
+			dev_dbg(&state->priv->i2c->dev,
+				"%s() demod%d unsupported modulation (%d)\n",
+				__func__, state->demod, c->modulation);
+			return -EOPNOTSUPP;
+		}
+
+		switch (c->pilot) {
+		case PILOT_AUTO:
+			state->dnxt.pilot_val = CX24117_PILOT_AUTO;
+			break;
+		case PILOT_OFF:
+			state->dnxt.pilot_val = CX24117_PILOT_OFF;
+			break;
+		case PILOT_ON:
+			state->dnxt.pilot_val = CX24117_PILOT_ON;
+			break;
+		default:
+			dev_dbg(&state->priv->i2c->dev,
+				"%s() demod%d unsupported pilot mode (%d)\n",
+				__func__, state->demod, c->pilot);
+			return -EOPNOTSUPP;
+		}
+
+		switch (c->rolloff) {
+		case ROLLOFF_20:
+			state->dnxt.rolloff_val = CX24117_ROLLOFF_020;
+			break;
+		case ROLLOFF_25:
+			state->dnxt.rolloff_val = CX24117_ROLLOFF_025;
+			break;
+		case ROLLOFF_35:
+			state->dnxt.rolloff_val = CX24117_ROLLOFF_035;
+			break;
+		case ROLLOFF_AUTO:
+			state->dnxt.rolloff_val = CX24117_ROLLOFF_035;
+			/* soft-auto rolloff */
+			retune = 3;
+			break;
+		default:
+			dev_warn(&state->priv->i2c->dev,
+				"%s: demod%d unsupported rolloff (%d)\n",
+				KBUILD_MODNAME, state->demod, c->rolloff);
+			return -EOPNOTSUPP;
+		}
+		break;
+
+	default:
+		dev_warn(&state->priv->i2c->dev,
+			"%s: demod %d unsupported delivery system (%d)\n",
+			KBUILD_MODNAME, state->demod, c->delivery_system);
+		return -EINVAL;
+	}
+
+	state->dnxt.delsys = c->delivery_system;
+	state->dnxt.modulation = c->modulation;
+	state->dnxt.frequency = c->frequency;
+	state->dnxt.pilot = c->pilot;
+	state->dnxt.rolloff = c->rolloff;
+
+	ret = cx24117_set_inversion(state, c->inversion);
+	if (ret !=  0)
+		return ret;
+
+	ret = cx24117_set_fec(state,
+		c->delivery_system, c->modulation, c->fec_inner);
+	if (ret !=  0)
+		return ret;
+
+	ret = cx24117_set_symbolrate(state, c->symbol_rate);
+	if (ret !=  0)
+		return ret;
+
+	/* discard the 'current' tuning parameters and prepare to tune */
+	cx24117_clone_params(fe);
+
+	dev_dbg(&state->priv->i2c->dev,
+		"%s: delsys      = %d\n", __func__, state->dcur.delsys);
+	dev_dbg(&state->priv->i2c->dev,
+		"%s: modulation  = %d\n", __func__, state->dcur.modulation);
+	dev_dbg(&state->priv->i2c->dev,
+		"%s: frequency   = %d\n", __func__, state->dcur.frequency);
+	dev_dbg(&state->priv->i2c->dev,
+		"%s: pilot       = %d (val = 0x%02x)\n", __func__,
+		state->dcur.pilot, state->dcur.pilot_val);
+	dev_dbg(&state->priv->i2c->dev,
+		"%s: retune      = %d\n", __func__, retune);
+	dev_dbg(&state->priv->i2c->dev,
+		"%s: rolloff     = %d (val = 0x%02x)\n", __func__,
+		state->dcur.rolloff, state->dcur.rolloff_val);
+	dev_dbg(&state->priv->i2c->dev,
+		"%s: symbol_rate = %d\n", __func__, state->dcur.symbol_rate);
+	dev_dbg(&state->priv->i2c->dev,
+		"%s: FEC         = %d (mask/val = 0x%02x/0x%02x)\n", __func__,
+		state->dcur.fec, state->dcur.fec_mask, state->dcur.fec_val);
+	dev_dbg(&state->priv->i2c->dev,
+		"%s: Inversion   = %d (val = 0x%02x)\n", __func__,
+		state->dcur.inversion, state->dcur.inversion_val);
+
+	/* Prepare a tune request */
+	cmd.args[0] = CMD_TUNEREQUEST;
+
+	/* demod */
+	cmd.args[1] = state->demod;
+
+	/* Frequency */
+	cmd.args[2] = (state->dcur.frequency & 0xff0000) >> 16;
+	cmd.args[3] = (state->dcur.frequency & 0x00ff00) >> 8;
+	cmd.args[4] = (state->dcur.frequency & 0x0000ff);
+
+	/* Symbol Rate */
+	cmd.args[5] = ((state->dcur.symbol_rate / 1000) & 0xff00) >> 8;
+	cmd.args[6] = ((state->dcur.symbol_rate / 1000) & 0x00ff);
+
+	/* Automatic Inversion */
+	cmd.args[7] = state->dcur.inversion_val;
+
+	/* Modulation / FEC / Pilot */
+	cmd.args[8] = state->dcur.fec_val | state->dcur.pilot_val;
+
+	cmd.args[9] = CX24117_SEARCH_RANGE_KHZ >> 8;
+	cmd.args[10] = CX24117_SEARCH_RANGE_KHZ & 0xff;
+
+	cmd.args[11] = state->dcur.rolloff_val;
+	cmd.args[12] = state->dcur.fec_mask;
+
+	if (state->dcur.symbol_rate > 30000000) {
+		reg_ratediv = 0x04;
+		reg_clkdiv = 0x02;
+	} else if (state->dcur.symbol_rate > 10000000) {
+		reg_ratediv = 0x06;
+		reg_clkdiv = 0x03;
+	} else {
+		reg_ratediv = 0x0a;
+		reg_clkdiv = 0x05;
+	}
+
+	cmd.args[13] = reg_ratediv;
+	cmd.args[14] = reg_clkdiv;
+
+	cx24117_writereg(state, (state->demod == 0) ?
+		CX24117_REG_CLKDIV0 : CX24117_REG_CLKDIV1, reg_clkdiv);
+	cx24117_writereg(state, (state->demod == 0) ?
+		CX24117_REG_RATEDIV0 : CX24117_REG_RATEDIV1, reg_ratediv);
+
+	cmd.args[15] = CX24117_PNE;
+	cmd.len = 16;
+
+	do {
+		/* Reset status register */
+		status = cx24117_readreg(state, (state->demod == 0) ?
+			CX24117_REG_SSTATUS0 : CX24117_REG_SSTATUS1) &
+			CX24117_SIGNAL_MASK;
+
+		dev_dbg(&state->priv->i2c->dev,
+			"%s() demod%d status_setfe = %02x\n",
+			__func__, state->demod, status);
+
+		cx24117_writereg(state, (state->demod == 0) ?
+			CX24117_REG_SSTATUS0 : CX24117_REG_SSTATUS1, status);
+
+		/* Tune */
+		ret = cx24117_cmd_execute(fe, &cmd);
+		if (ret != 0)
+			break;
+
+		/*
+		 * Wait for up to 500 ms before retrying
+		 *
+		 * If we are able to tune then generally it occurs within 100ms.
+		 * If it takes longer, try a different rolloff setting.
+		 */
+		for (i = 0; i < 50; i++) {
+			cx24117_read_status(fe, &tunerstat);
+			status = tunerstat & (FE_HAS_SIGNAL | FE_HAS_SYNC);
+			if (status == (FE_HAS_SIGNAL | FE_HAS_SYNC)) {
+				dev_dbg(&state->priv->i2c->dev,
+					"%s() demod%d tuned\n",
+					__func__, state->demod);
+				return 0;
+			}
+			msleep(20);
+		}
+
+		dev_dbg(&state->priv->i2c->dev, "%s() demod%d not tuned\n",
+			__func__, state->demod);
+
+		/* try next rolloff value */
+		if (state->dcur.rolloff == 3)
+			cmd.args[11]--;
+
+	} while (--retune);
+	return -EINVAL;
+}
+
+static int cx24117_tune(struct dvb_frontend *fe, bool re_tune,
+	unsigned int mode_flags, unsigned int *delay, fe_status_t *status)
+{
+	struct cx24117_state *state = fe->demodulator_priv;
+
+	dev_dbg(&state->priv->i2c->dev, "%s() demod%d\n",
+		__func__, state->demod);
+
+	*delay = HZ / 5;
+	if (re_tune) {
+		int ret = cx24117_set_frontend(fe);
+		if (ret)
+			return ret;
+	}
+	return cx24117_read_status(fe, status);
+}
+
+static int cx24117_get_algo(struct dvb_frontend *fe)
+{
+	return DVBFE_ALGO_HW;
+}
+
+static int cx24117_get_frontend(struct dvb_frontend *fe)
+{
+	struct cx24117_state *state = fe->demodulator_priv;
+	struct dtv_frontend_properties *c = &fe->dtv_property_cache;
+	struct cx24117_cmd cmd;
+	u8 reg, st, inv;
+	int ret, idx;
+	unsigned int freq;
+	short srate_os, freq_os;
+
+	u8 buf[0x1f-4];
+
+	cmd.args[0] = 0x1c;
+	cmd.args[1] = (u8) state->demod;
+	cmd.len = 2;
+	ret = cx24117_cmd_execute(fe, &cmd);
+	if (ret != 0)
+		return ret;
+
+	/* read all required regs at once */
+	reg = (state->demod == 0) ? CX24117_REG_FREQ3_0 : CX24117_REG_FREQ3_1;
+	ret = cx24117_readregN(state, reg, buf, 0x1f-4);
+	if (ret != 0)
+		return ret;
+
+	st = buf[5];
+
+	/* get spectral inversion */
+	inv = (((state->demod == 0) ? ~st : st) >> 6) & 1;
+	if (inv == 0)
+		c->inversion = INVERSION_OFF;
+	else
+		c->inversion = INVERSION_ON;
+
+	/* modulation and fec */
+	idx = st & 0x3f;
+	if (c->delivery_system == SYS_DVBS2) {
+		if (idx > 11)
+			idx += 9;
+		else
+			idx += 7;
+	}
+
+	c->modulation = cx24117_modfec_modes[idx].modulation;
+	c->fec_inner = cx24117_modfec_modes[idx].fec;
+
+	/* frequency */
+	freq = (buf[0] << 16) | (buf[1] << 8) | buf[2];
+	freq_os = (buf[8] << 8) | buf[9];
+	c->frequency = freq + freq_os;
+
+	/* symbol rate */
+	srate_os = (buf[10] << 8) | buf[11];
+	c->symbol_rate = -1000 * srate_os + state->dcur.symbol_rate;
+	return 0;
+}
+
+static struct dvb_frontend_ops cx24117_ops = {
+	.delsys = { SYS_DVBS, SYS_DVBS2 },
+	.info = {
+		.name = "Conexant CX24117/CX24132",
+		.frequency_min = 950000,
+		.frequency_max = 2150000,
+		.frequency_stepsize = 1011, /* kHz for QPSK frontends */
+		.frequency_tolerance = 5000,
+		.symbol_rate_min = 1000000,
+		.symbol_rate_max = 45000000,
+		.caps = FE_CAN_INVERSION_AUTO |
+			FE_CAN_FEC_1_2 | FE_CAN_FEC_2_3 | FE_CAN_FEC_3_4 |
+			FE_CAN_FEC_4_5 | FE_CAN_FEC_5_6 | FE_CAN_FEC_6_7 |
+			FE_CAN_FEC_7_8 | FE_CAN_FEC_AUTO |
+			FE_CAN_2G_MODULATION |
+			FE_CAN_QPSK | FE_CAN_RECOVER
+	},
+
+	.release = cx24117_release,
+
+	.init = cx24117_initfe,
+	.sleep = cx24117_sleep,
+	.read_status = cx24117_read_status,
+	.read_ber = cx24117_read_ber,
+	.read_signal_strength = cx24117_read_signal_strength,
+	.read_snr = cx24117_read_snr,
+	.read_ucblocks = cx24117_read_ucblocks,
+	.set_tone = cx24117_set_tone,
+	.set_voltage = cx24117_set_voltage,
+	.diseqc_send_master_cmd = cx24117_send_diseqc_msg,
+	.diseqc_send_burst = cx24117_diseqc_send_burst,
+	.get_frontend_algo = cx24117_get_algo,
+	.tune = cx24117_tune,
+
+	.set_frontend = cx24117_set_frontend,
+	.get_frontend = cx24117_get_frontend,
+};
+
+
+MODULE_DESCRIPTION("DVB Frontend module for Conexant cx24117/cx24132 hardware");
+MODULE_AUTHOR("Luis Alves (ljalvs@gmail.com)");
+MODULE_LICENSE("GPL");
+MODULE_VERSION("1.1");
+MODULE_FIRMWARE(CX24117_DEFAULT_FIRMWARE);
+
diff --git a/drivers/media/dvb-frontends/cx24117.h b/drivers/media/dvb-frontends/cx24117.h
new file mode 100644
index 000000000000..4e59e9574fa7
--- /dev/null
+++ b/drivers/media/dvb-frontends/cx24117.h
@@ -0,0 +1,47 @@
+/*
+    Conexant cx24117/cx24132 - Dual DVBS/S2 Satellite demod/tuner driver
+
+    Copyright (C) 2013 Luis Alves <ljalvs@gmail.com>
+	(based on cx24116.h by Steven Toth)
+
+    This program is free software; you can redistribute it and/or modify
+    it under the terms of the GNU General Public License as published by
+    the Free Software Foundation; either version 2 of the License, or
+    (at your option) any later version.
+
+    This program is distributed in the hope that it will be useful,
+    but WITHOUT ANY WARRANTY; without even the implied warranty of
+    MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+    GNU General Public License for more details.
+
+    You should have received a copy of the GNU General Public License
+    along with this program; if not, write to the Free Software
+    Foundation, Inc., 675 Mass Ave, Cambridge, MA 02139, USA.
+*/
+
+#ifndef CX24117_H
+#define CX24117_H
+
+#include <linux/kconfig.h>
+#include <linux/dvb/frontend.h>
+
+struct cx24117_config {
+	/* the demodulator's i2c address */
+	u8 demod_address;
+};
+
+#if IS_ENABLED(CONFIG_DVB_CX24117)
+extern struct dvb_frontend *cx24117_attach(
+	const struct cx24117_config *config,
+	struct i2c_adapter *i2c);
+#else
+static inline struct dvb_frontend *cx24117_attach(
+	const struct cx24117_config *config,
+	struct i2c_adapter *i2c)
+{
+	dev_warn(&i2c->dev, "%s: driver disabled by Kconfig\n", __func__);
+	return NULL;
+}
+#endif
+
+#endif /* CX24117_H */
diff --git a/drivers/media/dvb-frontends/cx24123.c b/drivers/media/dvb-frontends/cx24123.c
index a771da3e9f99..72fb5838cae0 100644
--- a/drivers/media/dvb-frontends/cx24123.c
+++ b/drivers/media/dvb-frontends/cx24123.c
@@ -739,7 +739,7 @@ static int cx24123_set_voltage(struct dvb_frontend *fe,
 		return 0;
 	default:
 		return -EINVAL;
-	};
+	}
 
 	return 0;
 }
diff --git a/drivers/media/dvb-frontends/cxd2820r_core.c b/drivers/media/dvb-frontends/cxd2820r_core.c
index 7ca5c69dd200..03930d5e9fea 100644
--- a/drivers/media/dvb-frontends/cxd2820r_core.c
+++ b/drivers/media/dvb-frontends/cxd2820r_core.c
@@ -21,21 +21,31 @@
 
 #include "cxd2820r_priv.h"
 
+/* Max transfer size done by I2C transfer functions */
+#define MAX_XFER_SIZE  64
+
 /* write multiple registers */
 static int cxd2820r_wr_regs_i2c(struct cxd2820r_priv *priv, u8 i2c, u8 reg,
 	u8 *val, int len)
 {
 	int ret;
-	u8 buf[len+1];
+	u8 buf[MAX_XFER_SIZE];
 	struct i2c_msg msg[1] = {
 		{
 			.addr = i2c,
 			.flags = 0,
-			.len = sizeof(buf),
+			.len = len + 1,
 			.buf = buf,
 		}
 	};
 
+	if (1 + len > sizeof(buf)) {
+		dev_warn(&priv->i2c->dev,
+			 "%s: i2c wr reg=%04x: len=%d is too big!\n",
+			 KBUILD_MODNAME, reg, len);
+		return -EINVAL;
+	}
+
 	buf[0] = reg;
 	memcpy(&buf[1], val, len);
 
@@ -55,7 +65,7 @@ static int cxd2820r_rd_regs_i2c(struct cxd2820r_priv *priv, u8 i2c, u8 reg,
 	u8 *val, int len)
 {
 	int ret;
-	u8 buf[len];
+	u8 buf[MAX_XFER_SIZE];
 	struct i2c_msg msg[2] = {
 		{
 			.addr = i2c,
@@ -65,11 +75,18 @@ static int cxd2820r_rd_regs_i2c(struct cxd2820r_priv *priv, u8 i2c, u8 reg,
 		}, {
 			.addr = i2c,
 			.flags = I2C_M_RD,
-			.len = sizeof(buf),
+			.len = len,
 			.buf = buf,
 		}
 	};
 
+	if (len > sizeof(buf)) {
+		dev_warn(&priv->i2c->dev,
+			 "%s: i2c wr reg=%04x: len=%d is too big!\n",
+			 KBUILD_MODNAME, reg, len);
+		return -EINVAL;
+	}
+
 	ret = i2c_transfer(priv->i2c, msg, 2);
 	if (ret == 2) {
 		memcpy(val, buf, len);
diff --git a/drivers/media/dvb-frontends/dib9000.c b/drivers/media/dvb-frontends/dib9000.c
index 6201c59a78dd..e540cfb13bac 100644
--- a/drivers/media/dvb-frontends/dib9000.c
+++ b/drivers/media/dvb-frontends/dib9000.c
@@ -649,9 +649,9 @@ static int dib9000_risc_debug_buf(struct dib9000_state *state, u16 * data, u8 si
 	b[2 * (size - 2) - 1] = '\0';	/* Bullet proof the buffer */
 	if (*b == '~') {
 		b++;
-		dprintk(b);
+		dprintk("%s", b);
 	} else
-		dprintk("RISC%d: %d.%04d %s", state->fe_id, ts / 10000, ts % 10000, *b ? b : "<emtpy>");
+		dprintk("RISC%d: %d.%04d %s", state->fe_id, ts / 10000, ts % 10000, *b ? b : "<empty>");
 	return 1;
 }
 
diff --git a/drivers/media/dvb-frontends/drxd_hard.c b/drivers/media/dvb-frontends/drxd_hard.c
index 9a2134792cfa..959ae36403b8 100644
--- a/drivers/media/dvb-frontends/drxd_hard.c
+++ b/drivers/media/dvb-frontends/drxd_hard.c
@@ -46,10 +46,6 @@
 #define DRX_I2C_MODEFLAGS     0xC0
 #define DRX_I2C_FLAGS         0xF0
 
-#ifndef SIZEOF_ARRAY
-#define SIZEOF_ARRAY(array) (sizeof((array))/sizeof((array)[0]))
-#endif
-
 #define DEFAULT_LOCK_TIMEOUT    1100
 
 #define DRX_CHANNEL_AUTO 0
@@ -1018,7 +1014,7 @@ static int HI_CfgCommand(struct drxd_state *state)
 		status = Write16(state, HI_RA_RAM_SRV_CMD__A,
 				 HI_RA_RAM_SRV_CMD_CONFIG, 0);
 	else
-		status = HI_Command(state, HI_RA_RAM_SRV_CMD_CONFIG, 0);
+		status = HI_Command(state, HI_RA_RAM_SRV_CMD_CONFIG, NULL);
 	mutex_unlock(&state->mutex);
 	return status;
 }
@@ -1039,7 +1035,7 @@ static int HI_ResetCommand(struct drxd_state *state)
 	status = Write16(state, HI_RA_RAM_SRV_RST_KEY__A,
 			 HI_RA_RAM_SRV_RST_KEY_ACT, 0);
 	if (status == 0)
-		status = HI_Command(state, HI_RA_RAM_SRV_CMD_RESET, 0);
+		status = HI_Command(state, HI_RA_RAM_SRV_CMD_RESET, NULL);
 	mutex_unlock(&state->mutex);
 	msleep(1);
 	return status;
@@ -2837,7 +2833,7 @@ static int drxd_init(struct dvb_frontend *fe)
 	int err = 0;
 
 /*	if (request_firmware(&state->fw, "drxd.fw", state->dev)<0) */
-	return DRXD_init(state, 0, 0);
+	return DRXD_init(state, NULL, 0);
 
 	err = DRXD_init(state, state->fw->data, state->fw->size);
 	release_firmware(state->fw);
@@ -2973,7 +2969,7 @@ struct dvb_frontend *drxd_attach(const struct drxd_config *config,
 
 	mutex_init(&state->mutex);
 
-	if (Read16(state, 0, 0, 0) < 0)
+	if (Read16(state, 0, NULL, 0) < 0)
 		goto error;
 
 	state->frontend.ops = drxd_ops;
diff --git a/drivers/media/dvb-frontends/drxk_hard.c b/drivers/media/dvb-frontends/drxk_hard.c
index 082014de6875..d416c15691da 100644
--- a/drivers/media/dvb-frontends/drxk_hard.c
+++ b/drivers/media/dvb-frontends/drxk_hard.c
@@ -1083,7 +1083,7 @@ static int hi_cfg_command(struct drxk_state *state)
 			 SIO_HI_RA_RAM_PAR_1_PAR1_SEC_KEY);
 	if (status < 0)
 		goto error;
-	status = hi_command(state, SIO_HI_RA_RAM_CMD_CONFIG, 0);
+	status = hi_command(state, SIO_HI_RA_RAM_CMD_CONFIG, NULL);
 	if (status < 0)
 		goto error;
 
@@ -2781,7 +2781,7 @@ static int ConfigureI2CBridge(struct drxk_state *state, bool b_enable_bridge)
 			goto error;
 	}
 
-	status = hi_command(state, SIO_HI_RA_RAM_CMD_BRDCTRL, 0);
+	status = hi_command(state, SIO_HI_RA_RAM_CMD_BRDCTRL, NULL);
 
 error:
 	if (status < 0)
diff --git a/drivers/media/dvb-frontends/itd1000.c b/drivers/media/dvb-frontends/itd1000.c
index c1c3400b2173..cadcae4cff89 100644
--- a/drivers/media/dvb-frontends/itd1000.c
+++ b/drivers/media/dvb-frontends/itd1000.c
@@ -31,6 +31,9 @@
 #include "itd1000.h"
 #include "itd1000_priv.h"
 
+/* Max transfer size done by I2C transfer functions */
+#define MAX_XFER_SIZE  64
+
 static int debug;
 module_param(debug, int, 0644);
 MODULE_PARM_DESC(debug, "Turn on/off debugging (default:off).");
@@ -52,10 +55,18 @@ MODULE_PARM_DESC(debug, "Turn on/off debugging (default:off).");
 /* don't write more than one byte with flexcop behind */
 static int itd1000_write_regs(struct itd1000_state *state, u8 reg, u8 v[], u8 len)
 {
-	u8 buf[1+len];
+	u8 buf[MAX_XFER_SIZE];
 	struct i2c_msg msg = {
 		.addr = state->cfg->i2c_address, .flags = 0, .buf = buf, .len = len+1
 	};
+
+	if (1 + len > sizeof(buf)) {
+		printk(KERN_WARNING
+		       "itd1000: i2c wr reg=%04x: len=%d is too big!\n",
+		       reg, len);
+		return -EINVAL;
+	}
+
 	buf[0] = reg;
 	memcpy(&buf[1], v, len);
 
diff --git a/drivers/media/dvb-frontends/mt312.c b/drivers/media/dvb-frontends/mt312.c
index ec388c1d6913..a74ac0ddb833 100644
--- a/drivers/media/dvb-frontends/mt312.c
+++ b/drivers/media/dvb-frontends/mt312.c
@@ -36,6 +36,8 @@
 #include "mt312_priv.h"
 #include "mt312.h"
 
+/* Max transfer size done by I2C transfer functions */
+#define MAX_XFER_SIZE  64
 
 struct mt312_state {
 	struct i2c_adapter *i2c;
@@ -96,9 +98,15 @@ static int mt312_write(struct mt312_state *state, const enum mt312_reg_addr reg,
 		       const u8 *src, const size_t count)
 {
 	int ret;
-	u8 buf[count + 1];
+	u8 buf[MAX_XFER_SIZE];
 	struct i2c_msg msg;
 
+	if (1 + count > sizeof(buf)) {
+		printk(KERN_WARNING
+		       "mt312: write: len=%zd is too big!\n", count);
+		return -EINVAL;
+	}
+
 	if (debug) {
 		int i;
 		dprintk("W(%d):", reg & 0x7f);
diff --git a/drivers/media/dvb-frontends/nxt200x.c b/drivers/media/dvb-frontends/nxt200x.c
index 8e288940a61f..fbca9856313a 100644
--- a/drivers/media/dvb-frontends/nxt200x.c
+++ b/drivers/media/dvb-frontends/nxt200x.c
@@ -39,6 +39,9 @@
  */
 #define pr_fmt(fmt) KBUILD_MODNAME ": " fmt
 
+/* Max transfer size done by I2C transfer functions */
+#define MAX_XFER_SIZE  64
+
 #define NXT2002_DEFAULT_FIRMWARE "dvb-fe-nxt2002.fw"
 #define NXT2004_DEFAULT_FIRMWARE "dvb-fe-nxt2004.fw"
 #define CRC_CCIT_MASK 0x1021
@@ -95,10 +98,16 @@ static int i2c_readbytes(struct nxt200x_state *state, u8 addr, u8 *buf, u8 len)
 static int nxt200x_writebytes (struct nxt200x_state* state, u8 reg,
 			       const u8 *buf, u8 len)
 {
-	u8 buf2 [len+1];
+	u8 buf2[MAX_XFER_SIZE];
 	int err;
 	struct i2c_msg msg = { .addr = state->config->demod_address, .flags = 0, .buf = buf2, .len = len + 1 };
 
+	if (1 + len > sizeof(buf2)) {
+		pr_warn("%s: i2c wr reg=%04x: len=%d is too big!\n",
+			 __func__, reg, len);
+		return -EINVAL;
+	}
+
 	buf2[0] = reg;
 	memcpy(&buf2[1], buf, len);
 
diff --git a/drivers/media/dvb-frontends/rtl2830.c b/drivers/media/dvb-frontends/rtl2830.c
index 362d26d11e82..7efb796c472c 100644
--- a/drivers/media/dvb-frontends/rtl2830.c
+++ b/drivers/media/dvb-frontends/rtl2830.c
@@ -27,20 +27,30 @@
 
 #include "rtl2830_priv.h"
 
+/* Max transfer size done by I2C transfer functions */
+#define MAX_XFER_SIZE  64
+
 /* write multiple hardware registers */
 static int rtl2830_wr(struct rtl2830_priv *priv, u8 reg, const u8 *val, int len)
 {
 	int ret;
-	u8 buf[1+len];
+	u8 buf[MAX_XFER_SIZE];
 	struct i2c_msg msg[1] = {
 		{
 			.addr = priv->cfg.i2c_addr,
 			.flags = 0,
-			.len = 1+len,
+			.len = 1 + len,
 			.buf = buf,
 		}
 	};
 
+	if (1 + len > sizeof(buf)) {
+		dev_warn(&priv->i2c->dev,
+			 "%s: i2c wr reg=%04x: len=%d is too big!\n",
+			 KBUILD_MODNAME, reg, len);
+		return -EINVAL;
+	}
+
 	buf[0] = reg;
 	memcpy(&buf[1], val, len);
 
diff --git a/drivers/media/dvb-frontends/rtl2832.c b/drivers/media/dvb-frontends/rtl2832.c
index facb84841518..ff73da9365e3 100644
--- a/drivers/media/dvb-frontends/rtl2832.c
+++ b/drivers/media/dvb-frontends/rtl2832.c
@@ -22,6 +22,9 @@
 #include "dvb_math.h"
 #include <linux/bitops.h>
 
+/* Max transfer size done by I2C transfer functions */
+#define MAX_XFER_SIZE  64
+
 int rtl2832_debug;
 module_param_named(debug, rtl2832_debug, int, 0644);
 MODULE_PARM_DESC(debug, "Turn on/off frontend debugging (default:off).");
@@ -162,16 +165,23 @@ static const struct rtl2832_reg_entry registers[] = {
 static int rtl2832_wr(struct rtl2832_priv *priv, u8 reg, u8 *val, int len)
 {
 	int ret;
-	u8 buf[1+len];
+	u8 buf[MAX_XFER_SIZE];
 	struct i2c_msg msg[1] = {
 		{
 			.addr = priv->cfg.i2c_addr,
 			.flags = 0,
-			.len = 1+len,
+			.len = 1 + len,
 			.buf = buf,
 		}
 	};
 
+	if (1 + len > sizeof(buf)) {
+		dev_warn(&priv->i2c->dev,
+			 "%s: i2c wr reg=%04x: len=%d is too big!\n",
+			 KBUILD_MODNAME, reg, len);
+		return -EINVAL;
+	}
+
 	buf[0] = reg;
 	memcpy(&buf[1], val, len);
 
@@ -489,6 +499,7 @@ static int rtl2832_init(struct dvb_frontend *fe)
 		init = rtl2832_tuner_init_e4000;
 		break;
 	case RTL2832_TUNER_R820T:
+	case RTL2832_TUNER_R828D:
 		len = ARRAY_SIZE(rtl2832_tuner_init_r820t);
 		init = rtl2832_tuner_init_r820t;
 		break;
diff --git a/drivers/media/dvb-frontends/rtl2832.h b/drivers/media/dvb-frontends/rtl2832.h
index 91b2dcf5a6ea..2cfbb6a97061 100644
--- a/drivers/media/dvb-frontends/rtl2832.h
+++ b/drivers/media/dvb-frontends/rtl2832.h
@@ -53,6 +53,7 @@ struct rtl2832_config {
 #define RTL2832_TUNER_E4000     0x27
 #define RTL2832_TUNER_FC0013    0x29
 #define RTL2832_TUNER_R820T	0x2a
+#define RTL2832_TUNER_R828D	0x2b
 	u8 tuner;
 };
 
diff --git a/drivers/media/dvb-frontends/s5h1420.c b/drivers/media/dvb-frontends/s5h1420.c
index e2fec9ebf947..93eeaf7118fd 100644
--- a/drivers/media/dvb-frontends/s5h1420.c
+++ b/drivers/media/dvb-frontends/s5h1420.c
@@ -836,9 +836,16 @@ static u32 s5h1420_tuner_i2c_func(struct i2c_adapter *adapter)
 static int s5h1420_tuner_i2c_tuner_xfer(struct i2c_adapter *i2c_adap, struct i2c_msg msg[], int num)
 {
 	struct s5h1420_state *state = i2c_get_adapdata(i2c_adap);
-	struct i2c_msg m[1 + num];
+	struct i2c_msg m[3];
 	u8 tx_open[2] = { CON_1, state->CON_1_val | 1 }; /* repeater stops once there was a stop condition */
 
+	if (1 + num > ARRAY_SIZE(m)) {
+		printk(KERN_WARNING
+		       "%s: i2c xfer: num=%d is too big!\n",
+		       KBUILD_MODNAME, num);
+		return  -EOPNOTSUPP;
+	}
+
 	memset(m, 0, sizeof(struct i2c_msg) * (1 + num));
 
 	m[0].addr = state->config->demod_address;
@@ -847,7 +854,7 @@ static int s5h1420_tuner_i2c_tuner_xfer(struct i2c_adapter *i2c_adap, struct i2c
 
 	memcpy(&m[1], msg, sizeof(struct i2c_msg) * num);
 
-	return i2c_transfer(state->i2c, m, 1+num) == 1 + num ? num : -EIO;
+	return i2c_transfer(state->i2c, m, 1 + num) == 1 + num ? num : -EIO;
 }
 
 static struct i2c_algorithm s5h1420_tuner_i2c_algo = {
diff --git a/drivers/media/dvb-frontends/stb0899_drv.c b/drivers/media/dvb-frontends/stb0899_drv.c
index 3dd5714eadba..07cd5ea7a038 100644
--- a/drivers/media/dvb-frontends/stb0899_drv.c
+++ b/drivers/media/dvb-frontends/stb0899_drv.c
@@ -32,6 +32,9 @@
 #include "stb0899_priv.h"
 #include "stb0899_reg.h"
 
+/* Max transfer size done by I2C transfer functions */
+#define MAX_XFER_SIZE  64
+
 static unsigned int verbose = 0;//1;
 module_param(verbose, int, 0644);
 
@@ -499,7 +502,7 @@ err:
 int stb0899_write_regs(struct stb0899_state *state, unsigned int reg, u8 *data, u32 count)
 {
 	int ret;
-	u8 buf[2 + count];
+	u8 buf[MAX_XFER_SIZE];
 	struct i2c_msg i2c_msg = {
 		.addr	= state->config->demod_address,
 		.flags	= 0,
@@ -507,6 +510,13 @@ int stb0899_write_regs(struct stb0899_state *state, unsigned int reg, u8 *data,
 		.len	= 2 + count
 	};
 
+	if (2 + count > sizeof(buf)) {
+		printk(KERN_WARNING
+		       "%s: i2c wr reg=%04x: len=%d is too big!\n",
+		       KBUILD_MODNAME, reg, count);
+		return -EINVAL;
+	}
+
 	buf[0] = reg >> 8;
 	buf[1] = reg & 0xff;
 	memcpy(&buf[2], data, count);
diff --git a/drivers/media/dvb-frontends/stb6100.c b/drivers/media/dvb-frontends/stb6100.c
index 45f9523f968f..cea175d19890 100644
--- a/drivers/media/dvb-frontends/stb6100.c
+++ b/drivers/media/dvb-frontends/stb6100.c
@@ -31,6 +31,8 @@
 static unsigned int verbose;
 module_param(verbose, int, 0644);
 
+/* Max transfer size done by I2C transfer functions */
+#define MAX_XFER_SIZE  64
 
 #define FE_ERROR		0
 #define FE_NOTICE		1
@@ -183,7 +185,7 @@ static int stb6100_read_reg(struct stb6100_state *state, u8 reg)
 static int stb6100_write_reg_range(struct stb6100_state *state, u8 buf[], int start, int len)
 {
 	int rc;
-	u8 cmdbuf[len + 1];
+	u8 cmdbuf[MAX_XFER_SIZE];
 	struct i2c_msg msg = {
 		.addr	= state->config->tuner_address,
 		.flags	= 0,
@@ -191,6 +193,13 @@ static int stb6100_write_reg_range(struct stb6100_state *state, u8 buf[], int st
 		.len	= len + 1
 	};
 
+	if (1 + len > sizeof(buf)) {
+		printk(KERN_WARNING
+		       "%s: i2c wr: len=%d is too big!\n",
+		       KBUILD_MODNAME, len);
+		return -EINVAL;
+	}
+
 	if (unlikely(start < 1 || start + len > STB6100_NUMREGS)) {
 		dprintk(verbose, FE_ERROR, 1, "Invalid register range %d:%d",
 			start, len);
diff --git a/drivers/media/dvb-frontends/stv0367.c b/drivers/media/dvb-frontends/stv0367.c
index 7b6dba3ce55e..458772739423 100644
--- a/drivers/media/dvb-frontends/stv0367.c
+++ b/drivers/media/dvb-frontends/stv0367.c
@@ -33,6 +33,9 @@
 #include "stv0367_regs.h"
 #include "stv0367_priv.h"
 
+/* Max transfer size done by I2C transfer functions */
+#define MAX_XFER_SIZE  64
+
 static int stvdebug;
 module_param_named(debug, stvdebug, int, 0644);
 
@@ -767,7 +770,7 @@ static struct st_register def0367cab[STV0367CAB_NBREGS] = {
 static
 int stv0367_writeregs(struct stv0367_state *state, u16 reg, u8 *data, int len)
 {
-	u8 buf[len + 2];
+	u8 buf[MAX_XFER_SIZE];
 	struct i2c_msg msg = {
 		.addr = state->config->demod_address,
 		.flags = 0,
@@ -776,6 +779,14 @@ int stv0367_writeregs(struct stv0367_state *state, u16 reg, u8 *data, int len)
 	};
 	int ret;
 
+	if (2 + len > sizeof(buf)) {
+		printk(KERN_WARNING
+		       "%s: i2c wr reg=%04x: len=%d is too big!\n",
+		       KBUILD_MODNAME, reg, len);
+		return -EINVAL;
+	}
+
+
 	buf[0] = MSB(reg);
 	buf[1] = LSB(reg);
 	memcpy(buf + 2, data, len);
diff --git a/drivers/media/dvb-frontends/stv090x.c b/drivers/media/dvb-frontends/stv090x.c
index 56d470ad5a82..23e872f84742 100644
--- a/drivers/media/dvb-frontends/stv090x.c
+++ b/drivers/media/dvb-frontends/stv090x.c
@@ -35,6 +35,9 @@
 #include "stv090x.h"
 #include "stv090x_priv.h"
 
+/* Max transfer size done by I2C transfer functions */
+#define MAX_XFER_SIZE  64
+
 static unsigned int verbose;
 module_param(verbose, int, 0644);
 
@@ -722,9 +725,16 @@ static int stv090x_write_regs(struct stv090x_state *state, unsigned int reg, u8
 {
 	const struct stv090x_config *config = state->config;
 	int ret;
-	u8 buf[2 + count];
+	u8 buf[MAX_XFER_SIZE];
 	struct i2c_msg i2c_msg = { .addr = config->address, .flags = 0, .buf = buf, .len = 2 + count };
 
+	if (2 + count > sizeof(buf)) {
+		printk(KERN_WARNING
+		       "%s: i2c wr reg=%04x: len=%d is too big!\n",
+		       KBUILD_MODNAME, reg, count);
+		return -EINVAL;
+	}
+
 	buf[0] = reg >> 8;
 	buf[1] = reg & 0xff;
 	memcpy(&buf[2], data, count);
diff --git a/drivers/media/dvb-frontends/stv6110.c b/drivers/media/dvb-frontends/stv6110.c
index 20b5fa92c53e..b1425830a24e 100644
--- a/drivers/media/dvb-frontends/stv6110.c
+++ b/drivers/media/dvb-frontends/stv6110.c
@@ -30,6 +30,9 @@
 
 #include "stv6110.h"
 
+/* Max transfer size done by I2C transfer functions */
+#define MAX_XFER_SIZE  64
+
 static int debug;
 
 struct stv6110_priv {
@@ -68,7 +71,7 @@ static int stv6110_write_regs(struct dvb_frontend *fe, u8 buf[],
 {
 	struct stv6110_priv *priv = fe->tuner_priv;
 	int rc;
-	u8 cmdbuf[len + 1];
+	u8 cmdbuf[MAX_XFER_SIZE];
 	struct i2c_msg msg = {
 		.addr	= priv->i2c_address,
 		.flags	= 0,
@@ -78,6 +81,13 @@ static int stv6110_write_regs(struct dvb_frontend *fe, u8 buf[],
 
 	dprintk("%s\n", __func__);
 
+	if (1 + len > sizeof(cmdbuf)) {
+		printk(KERN_WARNING
+		       "%s: i2c wr: len=%d is too big!\n",
+		       KBUILD_MODNAME, len);
+		return -EINVAL;
+	}
+
 	if (start + len > 8)
 		return -EINVAL;
 
diff --git a/drivers/media/dvb-frontends/stv6110x.c b/drivers/media/dvb-frontends/stv6110x.c
index f36cab12bdc7..e66154e5c1d7 100644
--- a/drivers/media/dvb-frontends/stv6110x.c
+++ b/drivers/media/dvb-frontends/stv6110x.c
@@ -32,6 +32,9 @@
 #include "stv6110x.h"
 #include "stv6110x_priv.h"
 
+/* Max transfer size done by I2C transfer functions */
+#define MAX_XFER_SIZE  64
+
 static unsigned int verbose;
 module_param(verbose, int, 0644);
 MODULE_PARM_DESC(verbose, "Set Verbosity level");
@@ -61,7 +64,8 @@ static int stv6110x_write_regs(struct stv6110x_state *stv6110x, int start, u8 da
 {
 	int ret;
 	const struct stv6110x_config *config = stv6110x->config;
-	u8 buf[len + 1];
+	u8 buf[MAX_XFER_SIZE];
+
 	struct i2c_msg msg = {
 		.addr = config->addr,
 		.flags = 0,
@@ -69,6 +73,13 @@ static int stv6110x_write_regs(struct stv6110x_state *stv6110x, int start, u8 da
 		.len = len + 1
 	};
 
+	if (1 + len > sizeof(buf)) {
+		printk(KERN_WARNING
+		       "%s: i2c wr: len=%d is too big!\n",
+		       KBUILD_MODNAME, len);
+		return -EINVAL;
+	}
+
 	if (start + len > 8)
 		return -EINVAL;
 
diff --git a/drivers/media/dvb-frontends/tda10071.c b/drivers/media/dvb-frontends/tda10071.c
index e79749cfec81..8ad3a57cf640 100644
--- a/drivers/media/dvb-frontends/tda10071.c
+++ b/drivers/media/dvb-frontends/tda10071.c
@@ -20,6 +20,9 @@
 
 #include "tda10071_priv.h"
 
+/* Max transfer size done by I2C transfer functions */
+#define MAX_XFER_SIZE  64
+
 static struct dvb_frontend_ops tda10071_ops;
 
 /* write multiple registers */
@@ -27,16 +30,23 @@ static int tda10071_wr_regs(struct tda10071_priv *priv, u8 reg, u8 *val,
 	int len)
 {
 	int ret;
-	u8 buf[len+1];
+	u8 buf[MAX_XFER_SIZE];
 	struct i2c_msg msg[1] = {
 		{
 			.addr = priv->cfg.demod_i2c_addr,
 			.flags = 0,
-			.len = sizeof(buf),
+			.len = 1 + len,
 			.buf = buf,
 		}
 	};
 
+	if (1 + len > sizeof(buf)) {
+		dev_warn(&priv->i2c->dev,
+			 "%s: i2c wr reg=%04x: len=%d is too big!\n",
+			 KBUILD_MODNAME, reg, len);
+		return -EINVAL;
+	}
+
 	buf[0] = reg;
 	memcpy(&buf[1], val, len);
 
@@ -56,7 +66,7 @@ static int tda10071_rd_regs(struct tda10071_priv *priv, u8 reg, u8 *val,
 	int len)
 {
 	int ret;
-	u8 buf[len];
+	u8 buf[MAX_XFER_SIZE];
 	struct i2c_msg msg[2] = {
 		{
 			.addr = priv->cfg.demod_i2c_addr,
@@ -66,11 +76,18 @@ static int tda10071_rd_regs(struct tda10071_priv *priv, u8 reg, u8 *val,
 		}, {
 			.addr = priv->cfg.demod_i2c_addr,
 			.flags = I2C_M_RD,
-			.len = sizeof(buf),
+			.len = len,
 			.buf = buf,
 		}
 	};
 
+	if (len > sizeof(buf)) {
+		dev_warn(&priv->i2c->dev,
+			 "%s: i2c wr reg=%04x: len=%d is too big!\n",
+			 KBUILD_MODNAME, reg, len);
+		return -EINVAL;
+	}
+
 	ret = i2c_transfer(priv->i2c, msg, 2);
 	if (ret == 2) {
 		memcpy(val, buf, len);
diff --git a/drivers/media/dvb-frontends/tda18271c2dd.c b/drivers/media/dvb-frontends/tda18271c2dd.c
index d281f77d5c28..2c54586ac07f 100644
--- a/drivers/media/dvb-frontends/tda18271c2dd.c
+++ b/drivers/media/dvb-frontends/tda18271c2dd.c
@@ -34,6 +34,9 @@
 #include "dvb_frontend.h"
 #include "tda18271c2dd.h"
 
+/* Max transfer size done by I2C transfer functions */
+#define MAX_XFER_SIZE  64
+
 struct SStandardParam {
 	s32   m_IFFrequency;
 	u32   m_BandWidth;
@@ -139,11 +142,18 @@ static int i2c_write(struct i2c_adapter *adap, u8 adr, u8 *data, int len)
 static int WriteRegs(struct tda_state *state,
 		     u8 SubAddr, u8 *Regs, u16 nRegs)
 {
-	u8 data[nRegs+1];
+	u8 data[MAX_XFER_SIZE];
+
+	if (1 + nRegs > sizeof(data)) {
+		printk(KERN_WARNING
+		       "%s: i2c wr: len=%d is too big!\n",
+		       KBUILD_MODNAME, nRegs);
+		return -EINVAL;
+	}
 
 	data[0] = SubAddr;
 	memcpy(data + 1, Regs, nRegs);
-	return i2c_write(state->i2c, state->adr, data, nRegs+1);
+	return i2c_write(state->i2c, state->adr, data, nRegs + 1);
 }
 
 static int WriteReg(struct tda_state *state, u8 SubAddr, u8 Reg)
diff --git a/drivers/media/dvb-frontends/tda8083.c b/drivers/media/dvb-frontends/tda8083.c
index 9d08350fe4b0..69e62f42e2e1 100644
--- a/drivers/media/dvb-frontends/tda8083.c
+++ b/drivers/media/dvb-frontends/tda8083.c
@@ -189,7 +189,7 @@ static int tda8083_set_tone (struct tda8083_state* state, fe_sec_tone_mode_t ton
 		return tda8083_writereg (state, 0x29, 0x80);
 	default:
 		return -EINVAL;
-	};
+	}
 }
 
 static int tda8083_set_voltage (struct tda8083_state* state, fe_sec_voltage_t voltage)
@@ -201,7 +201,7 @@ static int tda8083_set_voltage (struct tda8083_state* state, fe_sec_voltage_t vo
 		return tda8083_writereg (state, 0x20, 0x11);
 	default:
 		return -EINVAL;
-	};
+	}
 }
 
 static int tda8083_send_diseqc_burst (struct tda8083_state* state, fe_sec_mini_cmd_t burst)
diff --git a/drivers/media/dvb-frontends/ts2020.c b/drivers/media/dvb-frontends/ts2020.c
index ad7ad857ab2a..9aba044dabed 100644
--- a/drivers/media/dvb-frontends/ts2020.c
+++ b/drivers/media/dvb-frontends/ts2020.c
@@ -31,6 +31,7 @@ struct ts2020_priv {
 	struct i2c_adapter *i2c;
 	u8 clk_out_div;
 	u32 frequency;
+	u32 frequency_div;
 };
 
 static int ts2020_release(struct dvb_frontend *fe)
@@ -193,7 +194,7 @@ static int ts2020_set_params(struct dvb_frontend *fe)
 	u8 lo = 0x01, div4 = 0x0;
 
 	/* Calculate frequency divider */
-	if (frequency < 1060000) {
+	if (frequency < priv->frequency_div) {
 		lo |= 0x10;
 		div4 = 0x1;
 		ndiv = (frequency * 14 * 4) / TS2020_XTAL_FREQ;
@@ -340,8 +341,12 @@ struct dvb_frontend *ts2020_attach(struct dvb_frontend *fe,
 	priv->i2c_address = config->tuner_address;
 	priv->i2c = i2c;
 	priv->clk_out_div = config->clk_out_div;
+	priv->frequency_div = config->frequency_div;
 	fe->tuner_priv = priv;
 
+	if (!priv->frequency_div)
+		priv->frequency_div = 1060000;
+
 	/* Wake Up the tuner */
 	if ((0x03 & ts2020_readreg(fe, 0x00)) == 0x00) {
 		ts2020_writereg(fe, 0x00, 0x01);
diff --git a/drivers/media/dvb-frontends/ts2020.h b/drivers/media/dvb-frontends/ts2020.h
index 5bcb9a71ca80..b2fe6bb3a38b 100644
--- a/drivers/media/dvb-frontends/ts2020.h
+++ b/drivers/media/dvb-frontends/ts2020.h
@@ -28,6 +28,7 @@
 struct ts2020_config {
 	u8 tuner_address;
 	u8 clk_out_div;
+	u32 frequency_div;
 };
 
 #if IS_ENABLED(CONFIG_DVB_TS2020)
diff --git a/drivers/media/dvb-frontends/zl10039.c b/drivers/media/dvb-frontends/zl10039.c
index eff9c5fde50a..91b6b2e9b792 100644
--- a/drivers/media/dvb-frontends/zl10039.c
+++ b/drivers/media/dvb-frontends/zl10039.c
@@ -30,6 +30,9 @@
 
 static int debug;
 
+/* Max transfer size done by I2C transfer functions */
+#define MAX_XFER_SIZE  64
+
 #define dprintk(args...) \
 	do { \
 		if (debug) \
@@ -98,7 +101,7 @@ static int zl10039_write(struct zl10039_state *state,
 			const enum zl10039_reg_addr reg, const u8 *src,
 			const size_t count)
 {
-	u8 buf[count + 1];
+	u8 buf[MAX_XFER_SIZE];
 	struct i2c_msg msg = {
 		.addr = state->i2c_addr,
 		.flags = 0,
@@ -106,6 +109,13 @@ static int zl10039_write(struct zl10039_state *state,
 		.len = count + 1,
 	};
 
+	if (1 + count > sizeof(buf)) {
+		printk(KERN_WARNING
+		       "%s: i2c wr reg=%04x: len=%zd is too big!\n",
+		       KBUILD_MODNAME, reg, count);
+		return -EINVAL;
+	}
+
 	dprintk("%s\n", __func__);
 	/* Write register address and data in one go */
 	buf[0] = reg;
diff --git a/drivers/media/i2c/Kconfig b/drivers/media/i2c/Kconfig
index d18be19c96cd..842654d33317 100644
--- a/drivers/media/i2c/Kconfig
+++ b/drivers/media/i2c/Kconfig
@@ -621,6 +621,15 @@ config VIDEO_AS3645A
 	  This is a driver for the AS3645A and LM3555 flash controllers. It has
 	  build in control for flash, torch and indicator LEDs.
 
+config VIDEO_LM3560
+	tristate "LM3560 dual flash driver support"
+	depends on I2C && VIDEO_V4L2 && MEDIA_CONTROLLER
+	depends on MEDIA_CAMERA_SUPPORT
+	select REGMAP_I2C
+	---help---
+	  This is a driver for the lm3560 dual flash controllers. It controls
+	  flash, torch LEDs.
+
 comment "Video improvement chips"
 
 config VIDEO_UPD64031A
@@ -646,7 +655,7 @@ config VIDEO_UPD64083
 	  To compile this driver as a module, choose M here: the
 	  module will be called upd64083.
 
-comment "Miscelaneous helper chips"
+comment "Miscellaneous helper chips"
 
 config VIDEO_THS7303
 	tristate "THS7303/53 Video Amplifier"
diff --git a/drivers/media/i2c/Makefile b/drivers/media/i2c/Makefile
index 9f462df77b4a..e03f1776f4f4 100644
--- a/drivers/media/i2c/Makefile
+++ b/drivers/media/i2c/Makefile
@@ -70,6 +70,7 @@ obj-$(CONFIG_VIDEO_S5K4ECGX)	+= s5k4ecgx.o
 obj-$(CONFIG_VIDEO_S5C73M3)	+= s5c73m3/
 obj-$(CONFIG_VIDEO_ADP1653)	+= adp1653.o
 obj-$(CONFIG_VIDEO_AS3645A)	+= as3645a.o
+obj-$(CONFIG_VIDEO_LM3560)	+= lm3560.o
 obj-$(CONFIG_VIDEO_SMIAPP_PLL)	+= smiapp-pll.o
 obj-$(CONFIG_VIDEO_AK881X)		+= ak881x.o
 obj-$(CONFIG_VIDEO_IR_I2C)  += ir-kbd-i2c.o
diff --git a/drivers/media/i2c/adv7183.c b/drivers/media/i2c/adv7183.c
index 6f738d8e3a8f..d45e0e3a781d 100644
--- a/drivers/media/i2c/adv7183.c
+++ b/drivers/media/i2c/adv7183.c
@@ -178,7 +178,7 @@ static int adv7183_log_status(struct v4l2_subdev *sd)
 			adv7183_read(sd, ADV7183_VS_FIELD_CTRL_1),
 			adv7183_read(sd, ADV7183_VS_FIELD_CTRL_2),
 			adv7183_read(sd, ADV7183_VS_FIELD_CTRL_3));
-	v4l2_info(sd, "adv7183: Hsync positon control 1 2 and 3 = 0x%02x 0x%02x 0x%02x\n",
+	v4l2_info(sd, "adv7183: Hsync position control 1 2 and 3 = 0x%02x 0x%02x 0x%02x\n",
 			adv7183_read(sd, ADV7183_HS_POS_CTRL_1),
 			adv7183_read(sd, ADV7183_HS_POS_CTRL_2),
 			adv7183_read(sd, ADV7183_HS_POS_CTRL_3));
diff --git a/drivers/media/i2c/adv7343.c b/drivers/media/i2c/adv7343.c
index aeb56c53e39f..d4e15a617c3b 100644
--- a/drivers/media/i2c/adv7343.c
+++ b/drivers/media/i2c/adv7343.c
@@ -25,6 +25,7 @@
 #include <linux/module.h>
 #include <linux/videodev2.h>
 #include <linux/uaccess.h>
+#include <linux/of.h>
 
 #include <media/adv7343.h>
 #include <media/v4l2-async.h>
diff --git a/drivers/media/i2c/lm3560.c b/drivers/media/i2c/lm3560.c
new file mode 100644
index 000000000000..3317a9ae3961
--- /dev/null
+++ b/drivers/media/i2c/lm3560.c
@@ -0,0 +1,488 @@
+/*
+ * drivers/media/i2c/lm3560.c
+ * General device driver for TI lm3560, FLASH LED Driver
+ *
+ * Copyright (C) 2013 Texas Instruments
+ *
+ * Contact: Daniel Jeong <gshark.jeong@gmail.com>
+ *			Ldd-Mlp <ldd-mlp@list.ti.com>
+ *
+ * This program is free software; you can redistribute it and/or
+ * modify it under the terms of the GNU General Public License
+ * version 2 as published by the Free Software Foundation.
+ *
+ * This program is distributed in the hope that it will be useful, but
+ * WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the GNU
+ * General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program; if not, write to the Free Software
+ * Foundation, Inc., 51 Franklin St, Fifth Floor, Boston, MA
+ * 02110-1301 USA
+ *
+ */
+
+#include <linux/delay.h>
+#include <linux/module.h>
+#include <linux/i2c.h>
+#include <linux/slab.h>
+#include <linux/mutex.h>
+#include <linux/regmap.h>
+#include <linux/videodev2.h>
+#include <media/lm3560.h>
+#include <media/v4l2-ctrls.h>
+#include <media/v4l2-device.h>
+
+/* registers definitions */
+#define REG_ENABLE		0x10
+#define REG_TORCH_BR	0xa0
+#define REG_FLASH_BR	0xb0
+#define REG_FLASH_TOUT	0xc0
+#define REG_FLAG		0xd0
+#define REG_CONFIG1		0xe0
+
+/* Fault Mask */
+#define FAULT_TIMEOUT	(1<<0)
+#define FAULT_OVERTEMP	(1<<1)
+#define FAULT_SHORT_CIRCUIT	(1<<2)
+
+enum led_enable {
+	MODE_SHDN = 0x0,
+	MODE_TORCH = 0x2,
+	MODE_FLASH = 0x3,
+};
+
+/* struct lm3560_flash
+ *
+ * @pdata: platform data
+ * @regmap: reg. map for i2c
+ * @lock: muxtex for serial access.
+ * @led_mode: V4L2 LED mode
+ * @ctrls_led: V4L2 contols
+ * @subdev_led: V4L2 subdev
+ */
+struct lm3560_flash {
+	struct device *dev;
+	struct lm3560_platform_data *pdata;
+	struct regmap *regmap;
+	struct mutex lock;
+
+	enum v4l2_flash_led_mode led_mode;
+	struct v4l2_ctrl_handler ctrls_led[LM3560_LED_MAX];
+	struct v4l2_subdev subdev_led[LM3560_LED_MAX];
+};
+
+#define to_lm3560_flash(_ctrl, _no)	\
+	container_of(_ctrl->handler, struct lm3560_flash, ctrls_led[_no])
+
+/* enable mode control */
+static int lm3560_mode_ctrl(struct lm3560_flash *flash)
+{
+	int rval = -EINVAL;
+
+	switch (flash->led_mode) {
+	case V4L2_FLASH_LED_MODE_NONE:
+		rval = regmap_update_bits(flash->regmap,
+					  REG_ENABLE, 0x03, MODE_SHDN);
+		break;
+	case V4L2_FLASH_LED_MODE_TORCH:
+		rval = regmap_update_bits(flash->regmap,
+					  REG_ENABLE, 0x03, MODE_TORCH);
+		break;
+	case V4L2_FLASH_LED_MODE_FLASH:
+		rval = regmap_update_bits(flash->regmap,
+					  REG_ENABLE, 0x03, MODE_FLASH);
+		break;
+	}
+	return rval;
+}
+
+/* led1/2  enable/disable */
+static int lm3560_enable_ctrl(struct lm3560_flash *flash,
+			      enum lm3560_led_id led_no, bool on)
+{
+	int rval;
+
+	if (led_no == LM3560_LED0) {
+		if (on == true)
+			rval = regmap_update_bits(flash->regmap,
+						  REG_ENABLE, 0x08, 0x08);
+		else
+			rval = regmap_update_bits(flash->regmap,
+						  REG_ENABLE, 0x08, 0x00);
+	} else {
+		if (on == true)
+			rval = regmap_update_bits(flash->regmap,
+						  REG_ENABLE, 0x10, 0x10);
+		else
+			rval = regmap_update_bits(flash->regmap,
+						  REG_ENABLE, 0x10, 0x00);
+	}
+	return rval;
+}
+
+/* torch1/2 brightness control */
+static int lm3560_torch_brt_ctrl(struct lm3560_flash *flash,
+				 enum lm3560_led_id led_no, unsigned int brt)
+{
+	int rval;
+	u8 br_bits;
+
+	if (brt < LM3560_TORCH_BRT_MIN)
+		return lm3560_enable_ctrl(flash, led_no, false);
+	else
+		rval = lm3560_enable_ctrl(flash, led_no, true);
+
+	br_bits = LM3560_TORCH_BRT_uA_TO_REG(brt);
+	if (led_no == LM3560_LED0)
+		rval = regmap_update_bits(flash->regmap,
+					  REG_TORCH_BR, 0x07, br_bits);
+	else
+		rval = regmap_update_bits(flash->regmap,
+					  REG_TORCH_BR, 0x38, br_bits << 3);
+
+	return rval;
+}
+
+/* flash1/2 brightness control */
+static int lm3560_flash_brt_ctrl(struct lm3560_flash *flash,
+				 enum lm3560_led_id led_no, unsigned int brt)
+{
+	int rval;
+	u8 br_bits;
+
+	if (brt < LM3560_FLASH_BRT_MIN)
+		return lm3560_enable_ctrl(flash, led_no, false);
+	else
+		rval = lm3560_enable_ctrl(flash, led_no, true);
+
+	br_bits = LM3560_FLASH_BRT_uA_TO_REG(brt);
+	if (led_no == LM3560_LED0)
+		rval = regmap_update_bits(flash->regmap,
+					  REG_FLASH_BR, 0x0f, br_bits);
+	else
+		rval = regmap_update_bits(flash->regmap,
+					  REG_FLASH_BR, 0xf0, br_bits << 4);
+
+	return rval;
+}
+
+/* V4L2 controls  */
+static int lm3560_get_ctrl(struct v4l2_ctrl *ctrl, enum lm3560_led_id led_no)
+{
+	struct lm3560_flash *flash = to_lm3560_flash(ctrl, led_no);
+
+	mutex_lock(&flash->lock);
+
+	if (ctrl->id == V4L2_CID_FLASH_FAULT) {
+		int rval;
+		s32 fault = 0;
+		unsigned int reg_val;
+		rval = regmap_read(flash->regmap, REG_FLAG, &reg_val);
+		if (rval < 0)
+			return rval;
+		if (rval & FAULT_SHORT_CIRCUIT)
+			fault |= V4L2_FLASH_FAULT_SHORT_CIRCUIT;
+		if (rval & FAULT_OVERTEMP)
+			fault |= V4L2_FLASH_FAULT_OVER_TEMPERATURE;
+		if (rval & FAULT_TIMEOUT)
+			fault |= V4L2_FLASH_FAULT_TIMEOUT;
+		ctrl->cur.val = fault;
+		return 0;
+	}
+
+	mutex_unlock(&flash->lock);
+	return -EINVAL;
+}
+
+static int lm3560_set_ctrl(struct v4l2_ctrl *ctrl, enum lm3560_led_id led_no)
+{
+	struct lm3560_flash *flash = to_lm3560_flash(ctrl, led_no);
+	u8 tout_bits;
+	int rval = -EINVAL;
+
+	mutex_lock(&flash->lock);
+
+	switch (ctrl->id) {
+	case V4L2_CID_FLASH_LED_MODE:
+		flash->led_mode = ctrl->val;
+		if (flash->led_mode != V4L2_FLASH_LED_MODE_FLASH)
+			rval = lm3560_mode_ctrl(flash);
+		break;
+
+	case V4L2_CID_FLASH_STROBE_SOURCE:
+		rval = regmap_update_bits(flash->regmap,
+					  REG_CONFIG1, 0x04, (ctrl->val) << 2);
+		if (rval < 0)
+			goto err_out;
+		break;
+
+	case V4L2_CID_FLASH_STROBE:
+		if (flash->led_mode != V4L2_FLASH_LED_MODE_FLASH)
+			return -EBUSY;
+		flash->led_mode = V4L2_FLASH_LED_MODE_FLASH;
+		rval = lm3560_mode_ctrl(flash);
+		break;
+
+	case V4L2_CID_FLASH_STROBE_STOP:
+		if (flash->led_mode != V4L2_FLASH_LED_MODE_FLASH)
+			return -EBUSY;
+		flash->led_mode = V4L2_FLASH_LED_MODE_NONE;
+		rval = lm3560_mode_ctrl(flash);
+		break;
+
+	case V4L2_CID_FLASH_TIMEOUT:
+		tout_bits = LM3560_FLASH_TOUT_ms_TO_REG(ctrl->val);
+		rval = regmap_update_bits(flash->regmap,
+					  REG_FLASH_TOUT, 0x1f, tout_bits);
+		break;
+
+	case V4L2_CID_FLASH_INTENSITY:
+		rval = lm3560_flash_brt_ctrl(flash, led_no, ctrl->val);
+		break;
+
+	case V4L2_CID_FLASH_TORCH_INTENSITY:
+		rval = lm3560_torch_brt_ctrl(flash, led_no, ctrl->val);
+		break;
+	}
+
+	mutex_unlock(&flash->lock);
+err_out:
+	return rval;
+}
+
+static int lm3560_led1_get_ctrl(struct v4l2_ctrl *ctrl)
+{
+	return lm3560_get_ctrl(ctrl, LM3560_LED1);
+}
+
+static int lm3560_led1_set_ctrl(struct v4l2_ctrl *ctrl)
+{
+	return lm3560_set_ctrl(ctrl, LM3560_LED1);
+}
+
+static int lm3560_led0_get_ctrl(struct v4l2_ctrl *ctrl)
+{
+	return lm3560_get_ctrl(ctrl, LM3560_LED0);
+}
+
+static int lm3560_led0_set_ctrl(struct v4l2_ctrl *ctrl)
+{
+	return lm3560_set_ctrl(ctrl, LM3560_LED0);
+}
+
+static const struct v4l2_ctrl_ops lm3560_led_ctrl_ops[LM3560_LED_MAX] = {
+	[LM3560_LED0] = {
+			 .g_volatile_ctrl = lm3560_led0_get_ctrl,
+			 .s_ctrl = lm3560_led0_set_ctrl,
+			 },
+	[LM3560_LED1] = {
+			 .g_volatile_ctrl = lm3560_led1_get_ctrl,
+			 .s_ctrl = lm3560_led1_set_ctrl,
+			 }
+};
+
+static int lm3560_init_controls(struct lm3560_flash *flash,
+				enum lm3560_led_id led_no)
+{
+	struct v4l2_ctrl *fault;
+	u32 max_flash_brt = flash->pdata->max_flash_brt[led_no];
+	u32 max_torch_brt = flash->pdata->max_torch_brt[led_no];
+	struct v4l2_ctrl_handler *hdl = &flash->ctrls_led[led_no];
+	const struct v4l2_ctrl_ops *ops = &lm3560_led_ctrl_ops[led_no];
+
+	v4l2_ctrl_handler_init(hdl, 8);
+	/* flash mode */
+	v4l2_ctrl_new_std_menu(hdl, ops, V4L2_CID_FLASH_LED_MODE,
+			       V4L2_FLASH_LED_MODE_TORCH, ~0x7,
+			       V4L2_FLASH_LED_MODE_NONE);
+	flash->led_mode = V4L2_FLASH_LED_MODE_NONE;
+
+	/* flash source */
+	v4l2_ctrl_new_std_menu(hdl, ops, V4L2_CID_FLASH_STROBE_SOURCE,
+			       0x1, ~0x3, V4L2_FLASH_STROBE_SOURCE_SOFTWARE);
+
+	/* flash strobe */
+	v4l2_ctrl_new_std(hdl, ops, V4L2_CID_FLASH_STROBE, 0, 0, 0, 0);
+	/* flash strobe stop */
+	v4l2_ctrl_new_std(hdl, ops, V4L2_CID_FLASH_STROBE_STOP, 0, 0, 0, 0);
+
+	/* flash strobe timeout */
+	v4l2_ctrl_new_std(hdl, ops, V4L2_CID_FLASH_TIMEOUT,
+			  LM3560_FLASH_TOUT_MIN,
+			  flash->pdata->max_flash_timeout,
+			  LM3560_FLASH_TOUT_STEP,
+			  flash->pdata->max_flash_timeout);
+
+	/* flash brt */
+	v4l2_ctrl_new_std(hdl, ops, V4L2_CID_FLASH_INTENSITY,
+			  LM3560_FLASH_BRT_MIN, max_flash_brt,
+			  LM3560_FLASH_BRT_STEP, max_flash_brt);
+
+	/* torch brt */
+	v4l2_ctrl_new_std(hdl, ops, V4L2_CID_FLASH_TORCH_INTENSITY,
+			  LM3560_TORCH_BRT_MIN, max_torch_brt,
+			  LM3560_TORCH_BRT_STEP, max_torch_brt);
+
+	/* fault */
+	fault = v4l2_ctrl_new_std(hdl, ops, V4L2_CID_FLASH_FAULT, 0,
+				  V4L2_FLASH_FAULT_OVER_VOLTAGE
+				  | V4L2_FLASH_FAULT_OVER_TEMPERATURE
+				  | V4L2_FLASH_FAULT_SHORT_CIRCUIT
+				  | V4L2_FLASH_FAULT_TIMEOUT, 0, 0);
+	if (fault != NULL)
+		fault->flags |= V4L2_CTRL_FLAG_VOLATILE;
+
+	if (hdl->error)
+		return hdl->error;
+
+	flash->subdev_led[led_no].ctrl_handler = hdl;
+	return 0;
+}
+
+/* initialize device */
+static const struct v4l2_subdev_ops lm3560_ops = {
+	.core = NULL,
+};
+
+static const struct regmap_config lm3560_regmap = {
+	.reg_bits = 8,
+	.val_bits = 8,
+	.max_register = 0xFF,
+};
+
+static int lm3560_subdev_init(struct lm3560_flash *flash,
+			      enum lm3560_led_id led_no, char *led_name)
+{
+	struct i2c_client *client = to_i2c_client(flash->dev);
+	int rval;
+
+	v4l2_i2c_subdev_init(&flash->subdev_led[led_no], client, &lm3560_ops);
+	flash->subdev_led[led_no].flags |= V4L2_SUBDEV_FL_HAS_DEVNODE;
+	strcpy(flash->subdev_led[led_no].name, led_name);
+	rval = lm3560_init_controls(flash, led_no);
+	if (rval)
+		goto err_out;
+	rval = media_entity_init(&flash->subdev_led[led_no].entity, 0, NULL, 0);
+	if (rval < 0)
+		goto err_out;
+	flash->subdev_led[led_no].entity.type = MEDIA_ENT_T_V4L2_SUBDEV_FLASH;
+
+	return rval;
+
+err_out:
+	v4l2_ctrl_handler_free(&flash->ctrls_led[led_no]);
+	return rval;
+}
+
+static int lm3560_init_device(struct lm3560_flash *flash)
+{
+	int rval;
+	unsigned int reg_val;
+
+	/* set peak current */
+	rval = regmap_update_bits(flash->regmap,
+				  REG_FLASH_TOUT, 0x60, flash->pdata->peak);
+	if (rval < 0)
+		return rval;
+	/* output disable */
+	flash->led_mode = V4L2_FLASH_LED_MODE_NONE;
+	rval = lm3560_mode_ctrl(flash);
+	if (rval < 0)
+		return rval;
+	/* Reset faults */
+	rval = regmap_read(flash->regmap, REG_FLAG, &reg_val);
+	return rval;
+}
+
+static int lm3560_probe(struct i2c_client *client,
+			const struct i2c_device_id *devid)
+{
+	struct lm3560_flash *flash;
+	struct lm3560_platform_data *pdata = dev_get_platdata(&client->dev);
+	int rval;
+
+	flash = devm_kzalloc(&client->dev, sizeof(*flash), GFP_KERNEL);
+	if (flash == NULL)
+		return -ENOMEM;
+
+	flash->regmap = devm_regmap_init_i2c(client, &lm3560_regmap);
+	if (IS_ERR(flash->regmap)) {
+		rval = PTR_ERR(flash->regmap);
+		return rval;
+	}
+
+	/* if there is no platform data, use chip default value */
+	if (pdata == NULL) {
+		pdata =
+		    kzalloc(sizeof(struct lm3560_platform_data), GFP_KERNEL);
+		if (pdata == NULL)
+			return -ENODEV;
+		pdata->peak = LM3560_PEAK_3600mA;
+		pdata->max_flash_timeout = LM3560_FLASH_TOUT_MAX;
+		/* led 1 */
+		pdata->max_flash_brt[LM3560_LED0] = LM3560_FLASH_BRT_MAX;
+		pdata->max_torch_brt[LM3560_LED0] = LM3560_TORCH_BRT_MAX;
+		/* led 2 */
+		pdata->max_flash_brt[LM3560_LED1] = LM3560_FLASH_BRT_MAX;
+		pdata->max_torch_brt[LM3560_LED1] = LM3560_TORCH_BRT_MAX;
+	}
+	flash->pdata = pdata;
+	flash->dev = &client->dev;
+	mutex_init(&flash->lock);
+
+	rval = lm3560_subdev_init(flash, LM3560_LED0, "lm3560-led0");
+	if (rval < 0)
+		return rval;
+
+	rval = lm3560_subdev_init(flash, LM3560_LED1, "lm3560-led1");
+	if (rval < 0)
+		return rval;
+
+	rval = lm3560_init_device(flash);
+	if (rval < 0)
+		return rval;
+
+	return 0;
+}
+
+static int lm3560_remove(struct i2c_client *client)
+{
+	struct v4l2_subdev *subdev = i2c_get_clientdata(client);
+	struct lm3560_flash *flash = container_of(subdev, struct lm3560_flash,
+						  subdev_led[LM3560_LED_MAX]);
+	unsigned int i;
+
+	for (i = LM3560_LED0; i < LM3560_LED_MAX; i++) {
+		v4l2_device_unregister_subdev(&flash->subdev_led[i]);
+		v4l2_ctrl_handler_free(&flash->ctrls_led[i]);
+		media_entity_cleanup(&flash->subdev_led[i].entity);
+	}
+
+	return 0;
+}
+
+static const struct i2c_device_id lm3560_id_table[] = {
+	{LM3560_NAME, 0},
+	{}
+};
+
+MODULE_DEVICE_TABLE(i2c, lm3560_id_table);
+
+static struct i2c_driver lm3560_i2c_driver = {
+	.driver = {
+		   .name = LM3560_NAME,
+		   .pm = NULL,
+		   },
+	.probe = lm3560_probe,
+	.remove = lm3560_remove,
+	.id_table = lm3560_id_table,
+};
+
+module_i2c_driver(lm3560_i2c_driver);
+
+MODULE_AUTHOR("Daniel Jeong <gshark.jeong@gmail.com>");
+MODULE_AUTHOR("Ldd Mlp <ldd-mlp@list.ti.com>");
+MODULE_DESCRIPTION("Texas Instruments LM3560 LED flash driver");
+MODULE_LICENSE("GPL");
diff --git a/drivers/media/i2c/s5c73m3/s5c73m3-core.c b/drivers/media/i2c/s5c73m3/s5c73m3-core.c
index b76ec0e7e685..6fec9384d86e 100644
--- a/drivers/media/i2c/s5c73m3/s5c73m3-core.c
+++ b/drivers/media/i2c/s5c73m3/s5c73m3-core.c
@@ -1581,7 +1581,7 @@ static int s5c73m3_probe(struct i2c_client *client,
 	oif_sd = &state->oif_sd;
 
 	v4l2_subdev_init(sd, &s5c73m3_subdev_ops);
-	sd->owner = client->driver->driver.owner;
+	sd->owner = client->dev.driver->owner;
 	v4l2_set_subdevdata(sd, state);
 	strlcpy(sd->name, "S5C73M3", sizeof(sd->name));
 
@@ -1651,7 +1651,7 @@ static int s5c73m3_probe(struct i2c_client *client,
 	if (ret < 0)
 		goto out_err;
 
-	v4l2_info(sd, "%s: completed succesfully\n", __func__);
+	v4l2_info(sd, "%s: completed successfully\n", __func__);
 	return 0;
 
 out_err:
diff --git a/drivers/media/i2c/soc_camera/imx074.c b/drivers/media/i2c/soc_camera/imx074.c
index 1d384a371b41..5b915936c3f3 100644
--- a/drivers/media/i2c/soc_camera/imx074.c
+++ b/drivers/media/i2c/soc_camera/imx074.c
@@ -451,7 +451,9 @@ static int imx074_probe(struct i2c_client *client,
 	if (ret < 0)
 		goto eprobe;
 
-	return v4l2_async_register_subdev(&priv->subdev);
+	ret = v4l2_async_register_subdev(&priv->subdev);
+	if (!ret)
+		return 0;
 
 epwrinit:
 eprobe:
diff --git a/drivers/media/i2c/soc_camera/ov9640.c b/drivers/media/i2c/soc_camera/ov9640.c
index e968c3fdbd9e..bc74224503e7 100644
--- a/drivers/media/i2c/soc_camera/ov9640.c
+++ b/drivers/media/i2c/soc_camera/ov9640.c
@@ -371,7 +371,7 @@ static void ov9640_alter_regs(enum v4l2_mbus_pixelcode code,
 		alt->com13	= OV9640_COM13_RGB_AVG;
 		alt->com15	= OV9640_COM15_RGB_565;
 		break;
-	};
+	}
 }
 
 /* Setup registers according to resolution and color encoding */
diff --git a/drivers/media/i2c/ths8200.c b/drivers/media/i2c/ths8200.c
index d9f65d7e3e58..04139eec8c4e 100644
--- a/drivers/media/i2c/ths8200.c
+++ b/drivers/media/i2c/ths8200.c
@@ -19,6 +19,7 @@
 
 #include <linux/i2c.h>
 #include <linux/module.h>
+#include <linux/of.h>
 #include <linux/v4l2-dv-timings.h>
 
 #include <media/v4l2-dv-timings.h>
diff --git a/drivers/media/i2c/tvp514x.c b/drivers/media/i2c/tvp514x.c
index 91f3dd4cda1b..83d85df4853a 100644
--- a/drivers/media/i2c/tvp514x.c
+++ b/drivers/media/i2c/tvp514x.c
@@ -35,6 +35,7 @@
 #include <linux/videodev2.h>
 #include <linux/module.h>
 #include <linux/v4l2-mediabus.h>
+#include <linux/of.h>
 
 #include <media/v4l2-async.h>
 #include <media/v4l2-device.h>
diff --git a/drivers/media/i2c/tvp7002.c b/drivers/media/i2c/tvp7002.c
index 24a08fa7e328..912e1cccdd1c 100644
--- a/drivers/media/i2c/tvp7002.c
+++ b/drivers/media/i2c/tvp7002.c
@@ -29,6 +29,7 @@
 #include <linux/slab.h>
 #include <linux/videodev2.h>
 #include <linux/module.h>
+#include <linux/of.h>
 #include <linux/v4l2-dv-timings.h>
 #include <media/tvp7002.h>
 #include <media/v4l2-async.h>
diff --git a/drivers/media/pci/b2c2/flexcop-pci.c b/drivers/media/pci/b2c2/flexcop-pci.c
index 447afbd904a4..8b5e0b3a92a0 100644
--- a/drivers/media/pci/b2c2/flexcop-pci.c
+++ b/drivers/media/pci/b2c2/flexcop-pci.c
@@ -319,7 +319,6 @@ static int flexcop_pci_init(struct flexcop_pci *fc_pci)
 
 err_pci_iounmap:
 	pci_iounmap(fc_pci->pdev, fc_pci->io_mem);
-	pci_set_drvdata(fc_pci->pdev, NULL);
 err_pci_release_regions:
 	pci_release_regions(fc_pci->pdev);
 err_pci_disable_device:
@@ -332,7 +331,6 @@ static void flexcop_pci_exit(struct flexcop_pci *fc_pci)
 	if (fc_pci->init_state & FC_PCI_INIT) {
 		free_irq(fc_pci->pdev->irq, fc_pci);
 		pci_iounmap(fc_pci->pdev, fc_pci->io_mem);
-		pci_set_drvdata(fc_pci->pdev, NULL);
 		pci_release_regions(fc_pci->pdev);
 		pci_disable_device(fc_pci->pdev);
 	}
diff --git a/drivers/media/pci/bt8xx/bt878.c b/drivers/media/pci/bt8xx/bt878.c
index 66eb0baab0e9..d0c281f41a0a 100644
--- a/drivers/media/pci/bt8xx/bt878.c
+++ b/drivers/media/pci/bt8xx/bt878.c
@@ -488,8 +488,7 @@ static int bt878_probe(struct pci_dev *dev, const struct pci_device_id *pci_id)
 	btwrite(0, BT848_INT_MASK);
 
 	result = request_irq(bt->irq, bt878_irq,
-			     IRQF_SHARED | IRQF_DISABLED, "bt878",
-			     (void *) bt);
+			     IRQF_SHARED, "bt878", (void *) bt);
 	if (result == -EINVAL) {
 		printk(KERN_ERR "bt878(%d): Bad irq number or handler\n",
 		       bt878_num);
@@ -563,7 +562,6 @@ static void bt878_remove(struct pci_dev *pci_dev)
 	bt->shutdown = 1;
 	bt878_mem_free(bt);
 
-	pci_set_drvdata(pci_dev, NULL);
 	pci_disable_device(pci_dev);
 	return;
 }
diff --git a/drivers/media/pci/bt8xx/bttv-driver.c b/drivers/media/pci/bt8xx/bttv-driver.c
index c6532de0eac7..a3b1ee9c00d7 100644
--- a/drivers/media/pci/bt8xx/bttv-driver.c
+++ b/drivers/media/pci/bt8xx/bttv-driver.c
@@ -4086,7 +4086,7 @@ static int bttv_probe(struct pci_dev *dev, const struct pci_device_id *pci_id)
 	/* disable irqs, register irq handler */
 	btwrite(0, BT848_INT_MASK);
 	result = request_irq(btv->c.pci->irq, bttv_irq,
-	    IRQF_SHARED | IRQF_DISABLED, btv->c.v4l2_dev.name, (void *)btv);
+	    IRQF_SHARED, btv->c.v4l2_dev.name, (void *)btv);
 	if (result < 0) {
 		pr_err("%d: can't get IRQ %d\n",
 		       bttv_num, btv->c.pci->irq);
diff --git a/drivers/media/pci/cx18/cx18-driver.c b/drivers/media/pci/cx18/cx18-driver.c
index 004d8ace5019..c1f8cc6f14b2 100644
--- a/drivers/media/pci/cx18/cx18-driver.c
+++ b/drivers/media/pci/cx18/cx18-driver.c
@@ -324,23 +324,24 @@ static void cx18_eeprom_dump(struct cx18 *cx, unsigned char *eedata, int len)
 /* Hauppauge card? get values from tveeprom */
 void cx18_read_eeprom(struct cx18 *cx, struct tveeprom *tv)
 {
-	struct i2c_client c;
+	struct i2c_client *c;
 	u8 eedata[256];
 
-	memset(&c, 0, sizeof(c));
-	strlcpy(c.name, "cx18 tveeprom tmp", sizeof(c.name));
-	c.adapter = &cx->i2c_adap[0];
-	c.addr = 0xA0 >> 1;
+	c = kzalloc(sizeof(*c), GFP_KERNEL);
+
+	strlcpy(c->name, "cx18 tveeprom tmp", sizeof(c->name));
+	c->adapter = &cx->i2c_adap[0];
+	c->addr = 0xa0 >> 1;
 
 	memset(tv, 0, sizeof(*tv));
-	if (tveeprom_read(&c, eedata, sizeof(eedata)))
-		return;
+	if (tveeprom_read(c, eedata, sizeof(eedata)))
+		goto ret;
 
 	switch (cx->card->type) {
 	case CX18_CARD_HVR_1600_ESMT:
 	case CX18_CARD_HVR_1600_SAMSUNG:
 	case CX18_CARD_HVR_1600_S5H1411:
-		tveeprom_hauppauge_analog(&c, tv, eedata);
+		tveeprom_hauppauge_analog(c, tv, eedata);
 		break;
 	case CX18_CARD_YUAN_MPC718:
 	case CX18_CARD_GOTVIEW_PCI_DVD3:
@@ -354,6 +355,9 @@ void cx18_read_eeprom(struct cx18 *cx, struct tveeprom *tv)
 		cx18_eeprom_dump(cx, eedata, sizeof(eedata));
 		break;
 	}
+
+ret:
+	kfree(c);
 }
 
 static void cx18_process_eeprom(struct cx18 *cx)
@@ -1031,8 +1035,7 @@ static int cx18_probe(struct pci_dev *pci_dev,
 
 	/* Register IRQ */
 	retval = request_irq(cx->pci_dev->irq, cx18_irq_handler,
-			     IRQF_SHARED | IRQF_DISABLED,
-			     cx->v4l2_dev.name, (void *)cx);
+			     IRQF_SHARED, cx->v4l2_dev.name, (void *)cx);
 	if (retval) {
 		CX18_ERR("Failed to register irq %d\n", retval);
 		goto free_i2c;
diff --git a/drivers/media/pci/cx23885/Kconfig b/drivers/media/pci/cx23885/Kconfig
index 5104c802f72f..d1dcb1d2e087 100644
--- a/drivers/media/pci/cx23885/Kconfig
+++ b/drivers/media/pci/cx23885/Kconfig
@@ -23,6 +23,7 @@ config VIDEO_CX23885
 	select DVB_STB6100 if MEDIA_SUBDRV_AUTOSELECT
 	select DVB_STV6110 if MEDIA_SUBDRV_AUTOSELECT
 	select DVB_CX24116 if MEDIA_SUBDRV_AUTOSELECT
+	select DVB_CX24117 if MEDIA_SUBDRV_AUTOSELECT
 	select DVB_STV0900 if MEDIA_SUBDRV_AUTOSELECT
 	select DVB_DS3000 if MEDIA_SUBDRV_AUTOSELECT
 	select DVB_TS2020 if MEDIA_SUBDRV_AUTOSELECT
diff --git a/drivers/media/pci/cx23885/cimax2.c b/drivers/media/pci/cx23885/cimax2.c
index 7344849183a7..16fa7ea4d4aa 100644
--- a/drivers/media/pci/cx23885/cimax2.c
+++ b/drivers/media/pci/cx23885/cimax2.c
@@ -26,6 +26,10 @@
 #include "cx23885.h"
 #include "cimax2.h"
 #include "dvb_ca_en50221.h"
+
+/* Max transfer size done by I2C transfer functions */
+#define MAX_XFER_SIZE  64
+
 /**** Bit definitions for MC417_RWD and MC417_OEN registers  ***
   bits 31-16
 +-----------+
@@ -125,7 +129,7 @@ static int netup_write_i2c(struct i2c_adapter *i2c_adap, u8 addr, u8 reg,
 						u8 *buf, int len)
 {
 	int ret;
-	u8 buffer[len + 1];
+	u8 buffer[MAX_XFER_SIZE];
 
 	struct i2c_msg msg = {
 		.addr	= addr,
@@ -134,6 +138,13 @@ static int netup_write_i2c(struct i2c_adapter *i2c_adap, u8 addr, u8 reg,
 		.len	= len + 1
 	};
 
+	if (1 + len > sizeof(buffer)) {
+		printk(KERN_WARNING
+		       "%s: i2c wr reg=%04x: len=%d is too big!\n",
+		       KBUILD_MODNAME, reg, len);
+		return -EINVAL;
+	}
+
 	buffer[0] = reg;
 	memcpy(&buffer[1], buf, len);
 
diff --git a/drivers/media/pci/cx23885/cx23885-cards.c b/drivers/media/pci/cx23885/cx23885-cards.c
index 6a71a965e757..79f20c8c842e 100644
--- a/drivers/media/pci/cx23885/cx23885-cards.c
+++ b/drivers/media/pci/cx23885/cx23885-cards.c
@@ -223,6 +223,39 @@ struct cx23885_board cx23885_boards[] = {
 		.name		= "Leadtek Winfast PxDVR3200 H",
 		.portc		= CX23885_MPEG_DVB,
 	},
+	[CX23885_BOARD_LEADTEK_WINFAST_PXPVR2200] = {
+		.name		= "Leadtek Winfast PxPVR2200",
+		.porta		= CX23885_ANALOG_VIDEO,
+		.tuner_type	= TUNER_XC2028,
+		.tuner_addr	= 0x61,
+		.tuner_bus	= 1,
+		.input		= {{
+			.type	= CX23885_VMUX_TELEVISION,
+			.vmux	= CX25840_VIN2_CH1 |
+				  CX25840_VIN5_CH2,
+			.amux	= CX25840_AUDIO8,
+			.gpio0	= 0x704040,
+		}, {
+			.type	= CX23885_VMUX_COMPOSITE1,
+			.vmux	= CX25840_COMPOSITE1,
+			.amux	= CX25840_AUDIO7,
+			.gpio0	= 0x704040,
+		}, {
+			.type	= CX23885_VMUX_SVIDEO,
+			.vmux	= CX25840_SVIDEO_LUMA3 |
+				  CX25840_SVIDEO_CHROMA4,
+			.amux	= CX25840_AUDIO7,
+			.gpio0	= 0x704040,
+		}, {
+			.type	= CX23885_VMUX_COMPONENT,
+			.vmux	= CX25840_VIN7_CH1 |
+				  CX25840_VIN6_CH2 |
+				  CX25840_VIN8_CH3 |
+				  CX25840_COMPONENT_ON,
+			.amux	= CX25840_AUDIO7,
+			.gpio0	= 0x704040,
+		} },
+	},
 	[CX23885_BOARD_LEADTEK_WINFAST_PXDVR3200_H_XC4000] = {
 		.name		= "Leadtek Winfast PxDVR3200 H XC4000",
 		.porta		= CX23885_ANALOG_VIDEO,
@@ -259,6 +292,16 @@ struct cx23885_board cx23885_boards[] = {
 		.name		= "TurboSight TBS 6920",
 		.portb		= CX23885_MPEG_DVB,
 	},
+	[CX23885_BOARD_TBS_6980] = {
+		.name		= "TurboSight TBS 6980",
+		.portb		= CX23885_MPEG_DVB,
+		.portc		= CX23885_MPEG_DVB,
+	},
+	[CX23885_BOARD_TBS_6981] = {
+		.name		= "TurboSight TBS 6981",
+		.portb		= CX23885_MPEG_DVB,
+		.portc		= CX23885_MPEG_DVB,
+	},
 	[CX23885_BOARD_TEVII_S470] = {
 		.name		= "TeVii S470",
 		.portb		= CX23885_MPEG_DVB,
@@ -688,6 +731,10 @@ struct cx23885_subid cx23885_subids[] = {
 		.card      = CX23885_BOARD_LEADTEK_WINFAST_PXDVR3200_H,
 	}, {
 		.subvendor = 0x107d,
+		.subdevice = 0x6f21,
+		.card      = CX23885_BOARD_LEADTEK_WINFAST_PXPVR2200,
+	}, {
+		.subvendor = 0x107d,
 		.subdevice = 0x6f39,
 		.card	   = CX23885_BOARD_LEADTEK_WINFAST_PXDVR3200_H_XC4000,
 	}, {
@@ -699,6 +746,14 @@ struct cx23885_subid cx23885_subids[] = {
 		.subdevice = 0x8888,
 		.card      = CX23885_BOARD_TBS_6920,
 	}, {
+		.subvendor = 0x6980,
+		.subdevice = 0x8888,
+		.card      = CX23885_BOARD_TBS_6980,
+	}, {
+		.subvendor = 0x6981,
+		.subdevice = 0x8888,
+		.card      = CX23885_BOARD_TBS_6981,
+	}, {
 		.subvendor = 0xd470,
 		.subdevice = 0x9022,
 		.card      = CX23885_BOARD_TEVII_S470,
@@ -1023,6 +1078,35 @@ static void hauppauge_eeprom(struct cx23885_dev *dev, u8 *eeprom_data)
 			dev->name, tv.model);
 }
 
+/* Some TBS cards require initing a chip using a bitbanged SPI attached
+   to the cx23885 gpio's. If this chip doesn't get init'ed the demod
+   doesn't respond to any command. */
+static void tbs_card_init(struct cx23885_dev *dev)
+{
+	int i;
+	const u8 buf[] = {
+		0xe0, 0x06, 0x66, 0x33, 0x65,
+		0x01, 0x17, 0x06, 0xde};
+
+	switch (dev->board) {
+	case CX23885_BOARD_TBS_6980:
+	case CX23885_BOARD_TBS_6981:
+		cx_set(GP0_IO, 0x00070007);
+		usleep_range(1000, 10000);
+		cx_clear(GP0_IO, 2);
+		usleep_range(1000, 10000);
+		for (i = 0; i < 9 * 8; i++) {
+			cx_clear(GP0_IO, 7);
+			usleep_range(1000, 10000);
+			cx_set(GP0_IO,
+				((buf[i >> 3] >> (7 - (i & 7))) & 1) | 4);
+			usleep_range(1000, 10000);
+		}
+		cx_set(GP0_IO, 7);
+		break;
+	}
+}
+
 int cx23885_tuner_callback(void *priv, int component, int command, int arg)
 {
 	struct cx23885_tsport *port = priv;
@@ -1043,6 +1127,7 @@ int cx23885_tuner_callback(void *priv, int component, int command, int arg)
 	case CX23885_BOARD_HAUPPAUGE_HVR1500:
 	case CX23885_BOARD_HAUPPAUGE_HVR1500Q:
 	case CX23885_BOARD_LEADTEK_WINFAST_PXDVR3200_H:
+	case CX23885_BOARD_LEADTEK_WINFAST_PXPVR2200:
 	case CX23885_BOARD_LEADTEK_WINFAST_PXDVR3200_H_XC4000:
 	case CX23885_BOARD_COMPRO_VIDEOMATE_E650F:
 	case CX23885_BOARD_COMPRO_VIDEOMATE_E800:
@@ -1208,6 +1293,7 @@ void cx23885_gpio_setup(struct cx23885_dev *dev)
 		cx_set(GP0_IO, 0x000f000f);
 		break;
 	case CX23885_BOARD_LEADTEK_WINFAST_PXDVR3200_H:
+	case CX23885_BOARD_LEADTEK_WINFAST_PXPVR2200:
 	case CX23885_BOARD_LEADTEK_WINFAST_PXDVR3200_H_XC4000:
 	case CX23885_BOARD_COMPRO_VIDEOMATE_E650F:
 	case CX23885_BOARD_COMPRO_VIDEOMATE_E800:
@@ -1225,6 +1311,8 @@ void cx23885_gpio_setup(struct cx23885_dev *dev)
 		cx_set(GP0_IO, 0x00040004);
 		break;
 	case CX23885_BOARD_TBS_6920:
+	case CX23885_BOARD_TBS_6980:
+	case CX23885_BOARD_TBS_6981:
 	case CX23885_BOARD_PROF_8000:
 		cx_write(MC417_CTL, 0x00000036);
 		cx_write(MC417_OEN, 0x00001000);
@@ -1473,6 +1561,8 @@ int cx23885_ir_init(struct cx23885_dev *dev)
 	case CX23885_BOARD_TERRATEC_CINERGY_T_PCIE_DUAL:
 	case CX23885_BOARD_TEVII_S470:
 	case CX23885_BOARD_MYGICA_X8507:
+	case CX23885_BOARD_TBS_6980:
+	case CX23885_BOARD_TBS_6981:
 		if (!enable_885_ir)
 			break;
 		dev->sd_ir = cx23885_find_hw(dev, CX23885_HW_AV_CORE);
@@ -1516,6 +1606,8 @@ void cx23885_ir_fini(struct cx23885_dev *dev)
 	case CX23885_BOARD_TEVII_S470:
 	case CX23885_BOARD_HAUPPAUGE_HVR1250:
 	case CX23885_BOARD_MYGICA_X8507:
+	case CX23885_BOARD_TBS_6980:
+	case CX23885_BOARD_TBS_6981:
 		cx23885_irq_remove(dev, PCI_MSK_AV_CORE);
 		/* sd_ir is a duplicate pointer to the AV Core, just clear it */
 		dev->sd_ir = NULL;
@@ -1561,6 +1653,8 @@ void cx23885_ir_pci_int_enable(struct cx23885_dev *dev)
 	case CX23885_BOARD_TEVII_S470:
 	case CX23885_BOARD_HAUPPAUGE_HVR1250:
 	case CX23885_BOARD_MYGICA_X8507:
+	case CX23885_BOARD_TBS_6980:
+	case CX23885_BOARD_TBS_6981:
 		if (dev->sd_ir)
 			cx23885_irq_add_enable(dev, PCI_MSK_AV_CORE);
 		break;
@@ -1676,6 +1770,16 @@ void cx23885_card_setup(struct cx23885_dev *dev)
 		ts2->ts_clk_en_val = 0x1; /* Enable TS_CLK */
 		ts2->src_sel_val   = CX23885_SRC_SEL_PARALLEL_MPEG_VIDEO;
 		break;
+	case CX23885_BOARD_TBS_6980:
+	case CX23885_BOARD_TBS_6981:
+		ts1->gen_ctrl_val  = 0xc; /* Serial bus + punctured clock */
+		ts1->ts_clk_en_val = 0x1; /* Enable TS_CLK */
+		ts1->src_sel_val   = CX23885_SRC_SEL_PARALLEL_MPEG_VIDEO;
+		ts2->gen_ctrl_val  = 0xc; /* Serial bus + punctured clock */
+		ts2->ts_clk_en_val = 0x1; /* Enable TS_CLK */
+		ts2->src_sel_val   = CX23885_SRC_SEL_PARALLEL_MPEG_VIDEO;
+		tbs_card_init(dev);
+		break;
 	case CX23885_BOARD_MYGICA_X8506:
 	case CX23885_BOARD_MAGICPRO_PROHDTVE2:
 	case CX23885_BOARD_MYGICA_X8507:
@@ -1704,6 +1808,7 @@ void cx23885_card_setup(struct cx23885_dev *dev)
 	case CX23885_BOARD_HAUPPAUGE_HVR1700:
 	case CX23885_BOARD_HAUPPAUGE_HVR1400:
 	case CX23885_BOARD_LEADTEK_WINFAST_PXDVR3200_H:
+	case CX23885_BOARD_LEADTEK_WINFAST_PXPVR2200:
 	case CX23885_BOARD_LEADTEK_WINFAST_PXDVR3200_H_XC4000:
 	case CX23885_BOARD_COMPRO_VIDEOMATE_E650F:
 	case CX23885_BOARD_HAUPPAUGE_HVR1270:
@@ -1733,6 +1838,7 @@ void cx23885_card_setup(struct cx23885_dev *dev)
 	case CX23885_BOARD_HAUPPAUGE_HVR1800lp:
 	case CX23885_BOARD_HAUPPAUGE_HVR1700:
 	case CX23885_BOARD_LEADTEK_WINFAST_PXDVR3200_H:
+	case CX23885_BOARD_LEADTEK_WINFAST_PXPVR2200:
 	case CX23885_BOARD_LEADTEK_WINFAST_PXDVR3200_H_XC4000:
 	case CX23885_BOARD_COMPRO_VIDEOMATE_E650F:
 	case CX23885_BOARD_NETUP_DUAL_DVBS2_CI:
@@ -1752,6 +1858,8 @@ void cx23885_card_setup(struct cx23885_dev *dev)
 	case CX23885_BOARD_MYGICA_X8507:
 	case CX23885_BOARD_TERRATEC_CINERGY_T_PCIE_DUAL:
 	case CX23885_BOARD_AVERMEDIA_HC81R:
+	case CX23885_BOARD_TBS_6980:
+	case CX23885_BOARD_TBS_6981:
 		dev->sd_cx25840 = v4l2_i2c_new_subdev(&dev->v4l2_dev,
 				&dev->i2c_bus[2].i2c_adap,
 				"cx25840", 0x88 >> 1, NULL);
diff --git a/drivers/media/pci/cx23885/cx23885-core.c b/drivers/media/pci/cx23885/cx23885-core.c
index 9f63d93239ec..edcd79db1e4e 100644
--- a/drivers/media/pci/cx23885/cx23885-core.c
+++ b/drivers/media/pci/cx23885/cx23885-core.c
@@ -2129,7 +2129,7 @@ static int cx23885_initdev(struct pci_dev *pci_dev,
 	}
 
 	err = request_irq(pci_dev->irq, cx23885_irq,
-			  IRQF_SHARED | IRQF_DISABLED, dev->name, dev);
+			  IRQF_SHARED, dev->name, dev);
 	if (err < 0) {
 		printk(KERN_ERR "%s: can't get IRQ %d\n",
 		       dev->name, pci_dev->irq);
diff --git a/drivers/media/pci/cx23885/cx23885-dvb.c b/drivers/media/pci/cx23885/cx23885-dvb.c
index 971e4ff1b87f..05492053b473 100644
--- a/drivers/media/pci/cx23885/cx23885-dvb.c
+++ b/drivers/media/pci/cx23885/cx23885-dvb.c
@@ -51,6 +51,7 @@
 #include "stv6110.h"
 #include "lnbh24.h"
 #include "cx24116.h"
+#include "cx24117.h"
 #include "cimax2.h"
 #include "lgs8gxx.h"
 #include "netup-eeprom.h"
@@ -461,6 +462,10 @@ static struct cx24116_config tbs_cx24116_config = {
 	.demod_address = 0x55,
 };
 
+static struct cx24117_config tbs_cx24117_config = {
+	.demod_address = 0x55,
+};
+
 static struct ds3000_config tevii_ds3000_config = {
 	.demod_address = 0x68,
 };
@@ -1044,6 +1049,25 @@ static int dvb_register(struct cx23885_tsport *port)
 			fe0->dvb.frontend->ops.set_voltage = f300_set_voltage;
 
 		break;
+	case CX23885_BOARD_TBS_6980:
+	case CX23885_BOARD_TBS_6981:
+		i2c_bus = &dev->i2c_bus[1];
+
+		switch (port->nr) {
+		/* PORT B */
+		case 1:
+			fe0->dvb.frontend = dvb_attach(cx24117_attach,
+					&tbs_cx24117_config,
+					&i2c_bus->i2c_adap);
+			break;
+		/* PORT C */
+		case 2:
+			fe0->dvb.frontend = dvb_attach(cx24117_attach,
+					&tbs_cx24117_config,
+					&i2c_bus->i2c_adap);
+			break;
+		}
+		break;
 	case CX23885_BOARD_TEVII_S470:
 		i2c_bus = &dev->i2c_bus[1];
 
diff --git a/drivers/media/pci/cx23885/cx23885-input.c b/drivers/media/pci/cx23885/cx23885-input.c
index 7875dfbe09ff..8a49e7c9eddd 100644
--- a/drivers/media/pci/cx23885/cx23885-input.c
+++ b/drivers/media/pci/cx23885/cx23885-input.c
@@ -90,6 +90,8 @@ void cx23885_input_rx_work_handler(struct cx23885_dev *dev, u32 events)
 	case CX23885_BOARD_TEVII_S470:
 	case CX23885_BOARD_HAUPPAUGE_HVR1250:
 	case CX23885_BOARD_MYGICA_X8507:
+	case CX23885_BOARD_TBS_6980:
+	case CX23885_BOARD_TBS_6981:
 		/*
 		 * The only boards we handle right now.  However other boards
 		 * using the CX2388x integrated IR controller should be similar
@@ -168,6 +170,8 @@ static int cx23885_input_ir_start(struct cx23885_dev *dev)
 		break;
 	case CX23885_BOARD_TERRATEC_CINERGY_T_PCIE_DUAL:
 	case CX23885_BOARD_TEVII_S470:
+	case CX23885_BOARD_TBS_6980:
+	case CX23885_BOARD_TBS_6981:
 		/*
 		 * The IR controller on this board only returns pulse widths.
 		 * Any other mode setting will fail to set up the device.
@@ -298,6 +302,14 @@ int cx23885_input_init(struct cx23885_dev *dev)
 		/* A guess at the remote */
 		rc_map = RC_MAP_TOTAL_MEDIA_IN_HAND_02;
 		break;
+	case CX23885_BOARD_TBS_6980:
+	case CX23885_BOARD_TBS_6981:
+		/* Integrated CX23885 IR controller */
+		driver_type = RC_DRIVER_IR_RAW;
+		allowed_protos = RC_BIT_ALL;
+		/* A guess at the remote */
+		rc_map = RC_MAP_TBS_NEC;
+		break;
 	default:
 		return -ENODEV;
 	}
diff --git a/drivers/media/pci/cx23885/cx23885-video.c b/drivers/media/pci/cx23885/cx23885-video.c
index 161686832b20..7891f34157d1 100644
--- a/drivers/media/pci/cx23885/cx23885-video.c
+++ b/drivers/media/pci/cx23885/cx23885-video.c
@@ -1865,7 +1865,8 @@ int cx23885_video_register(struct cx23885_dev *dev)
 
 			v4l2_subdev_call(sd, tuner, s_type_addr, &tun_setup);
 
-			if (dev->board == CX23885_BOARD_LEADTEK_WINFAST_PXTV1200) {
+			if ((dev->board == CX23885_BOARD_LEADTEK_WINFAST_PXTV1200) ||
+			    (dev->board == CX23885_BOARD_LEADTEK_WINFAST_PXPVR2200)) {
 				struct xc2028_ctrl ctrl = {
 					.fname = XC2028_DEFAULT_FIRMWARE,
 					.max_len = 64
diff --git a/drivers/media/pci/cx23885/cx23885.h b/drivers/media/pci/cx23885/cx23885.h
index 038caf53908b..0fa4048ab872 100644
--- a/drivers/media/pci/cx23885/cx23885.h
+++ b/drivers/media/pci/cx23885/cx23885.h
@@ -93,6 +93,9 @@
 #define CX23885_BOARD_PROF_8000                37
 #define CX23885_BOARD_HAUPPAUGE_HVR4400        38
 #define CX23885_BOARD_AVERMEDIA_HC81R          39
+#define CX23885_BOARD_TBS_6981                 40
+#define CX23885_BOARD_TBS_6980                 41
+#define CX23885_BOARD_LEADTEK_WINFAST_PXPVR2200 42
 
 #define GPIO_0 0x00000001
 #define GPIO_1 0x00000002
diff --git a/drivers/media/pci/cx25821/cx25821-cards.c b/drivers/media/pci/cx25821/cx25821-cards.c
index 3b409feb03d8..f2ebc989b303 100644
--- a/drivers/media/pci/cx25821/cx25821-cards.c
+++ b/drivers/media/pci/cx25821/cx25821-cards.c
@@ -45,5 +45,3 @@ struct cx25821_board cx25821_boards[] = {
 	},
 
 };
-
-const unsigned int cx25821_bcount = ARRAY_SIZE(cx25821_boards);
diff --git a/drivers/media/pci/cx25821/cx25821-medusa-video.c b/drivers/media/pci/cx25821/cx25821-medusa-video.c
index 22fa04415ccc..43bdfa4dfba1 100644
--- a/drivers/media/pci/cx25821/cx25821-medusa-video.c
+++ b/drivers/media/pci/cx25821/cx25821-medusa-video.c
@@ -438,7 +438,7 @@ void medusa_set_resolution(struct cx25821_dev *dev, int width,
 		decoder_count = decoder_select + 1;
 	} else {
 		decoder = 0;
-		decoder_count = _num_decoders;
+		decoder_count = dev->_max_num_decoders;
 	}
 
 	switch (width) {
@@ -506,8 +506,6 @@ static void medusa_set_decoderduration(struct cx25821_dev *dev, int decoder,
 		break;
 	}
 
-	_display_field_cnt[decoder] = duration;
-
 	/* update hardware */
 	fld_cnt = cx25821_i2c_read(&dev->i2c_bus[0], disp_cnt_reg, &tmp);
 
@@ -667,8 +665,6 @@ int medusa_video_init(struct cx25821_dev *dev)
 	int ret_val = 0;
 	int i = 0;
 
-	_num_decoders = dev->_max_num_decoders;
-
 	/* disable Auto source selection on all video decoders */
 	value = cx25821_i2c_read(&dev->i2c_bus[0], MON_A_CTRL, &tmp);
 	value &= 0xFFFFF0FF;
@@ -685,8 +681,14 @@ int medusa_video_init(struct cx25821_dev *dev)
 	if (ret_val < 0)
 		goto error;
 
-	for (i = 0; i < _num_decoders; i++)
-		medusa_set_decoderduration(dev, i, _display_field_cnt[i]);
+	/*
+	 * FIXME: due to a coding bug the duration was always 0. It's
+	 * likely that it really should be something else, but due to the
+	 * lack of documentation I have no idea what it should be. For
+	 * now just fill in 0 as the duration.
+	 */
+	for (i = 0; i < dev->_max_num_decoders; i++)
+		medusa_set_decoderduration(dev, i, 0);
 
 	/* Select monitor as DENC A input, power up the DAC */
 	value = cx25821_i2c_read(&dev->i2c_bus[0], DENC_AB_CTRL, &tmp);
@@ -717,7 +719,7 @@ int medusa_video_init(struct cx25821_dev *dev)
 	/* Turn on all of the data out and control output pins. */
 	value = cx25821_i2c_read(&dev->i2c_bus[0], PIN_OE_CTRL, &tmp);
 	value &= 0xFEF0FE00;
-	if (_num_decoders == MAX_DECODERS) {
+	if (dev->_max_num_decoders == MAX_DECODERS) {
 		/*
 		 * Note: The octal board does not support control pins(bit16-19)
 		 * These bits are ignored in the octal board.
diff --git a/drivers/media/pci/cx25821/cx25821-medusa-video.h b/drivers/media/pci/cx25821/cx25821-medusa-video.h
index 6175e0961855..8bf602ff27b1 100644
--- a/drivers/media/pci/cx25821/cx25821-medusa-video.h
+++ b/drivers/media/pci/cx25821/cx25821-medusa-video.h
@@ -40,10 +40,4 @@
 #define CONTRAST_DEFAULT                5000
 #define HUE_DEFAULT                     5000
 
-unsigned short _num_decoders;
-unsigned short _num_cameras;
-
-unsigned int _video_standard;
-int _display_field_cnt[MAX_DECODERS];
-
 #endif
diff --git a/drivers/media/pci/cx25821/cx25821-video-upstream.c b/drivers/media/pci/cx25821/cx25821-video-upstream.c
index 88ffef410c50..1f43be0b04c8 100644
--- a/drivers/media/pci/cx25821/cx25821-video-upstream.c
+++ b/drivers/media/pci/cx25821/cx25821-video-upstream.c
@@ -159,10 +159,10 @@ static __le32 *cx25821_risc_field_upstream(struct cx25821_channel *chan, __le32
 		 * For the upstream video channel, the risc engine will enable
 		 * the FIFO. */
 		if (fifo_enable && line == 3) {
-			*(rp++) = RISC_WRITECR;
-			*(rp++) = sram_ch->dma_ctl;
-			*(rp++) = FLD_VID_FIFO_EN;
-			*(rp++) = 0x00000001;
+			*(rp++) = cpu_to_le32(RISC_WRITECR);
+			*(rp++) = cpu_to_le32(sram_ch->dma_ctl);
+			*(rp++) = cpu_to_le32(FLD_VID_FIFO_EN);
+			*(rp++) = cpu_to_le32(0x00000001);
 		}
 	}
 
diff --git a/drivers/media/pci/cx88/cx88-alsa.c b/drivers/media/pci/cx88/cx88-alsa.c
index aba5b1c649e6..400eb1c42d3f 100644
--- a/drivers/media/pci/cx88/cx88-alsa.c
+++ b/drivers/media/pci/cx88/cx88-alsa.c
@@ -834,7 +834,7 @@ static int snd_cx88_create(struct snd_card *card, struct pci_dev *pci,
 
 	/* get irq */
 	err = request_irq(chip->pci->irq, cx8801_irq,
-			  IRQF_SHARED | IRQF_DISABLED, chip->core->name, chip);
+			  IRQF_SHARED, chip->core->name, chip);
 	if (err < 0) {
 		dprintk(0, "%s: can't get IRQ %d\n",
 		       chip->core->name, chip->pci->irq);
@@ -935,8 +935,6 @@ static void cx88_audio_finidev(struct pci_dev *pci)
 
 	snd_card_free((void *)card);
 
-	pci_set_drvdata(pci, NULL);
-
 	devno--;
 }
 
@@ -951,27 +949,4 @@ static struct pci_driver cx88_audio_pci_driver = {
 	.remove   = cx88_audio_finidev,
 };
 
-/****************************************************************************
-				LINUX MODULE INIT
- ****************************************************************************/
-
-/*
- * module init
- */
-static int __init cx88_audio_init(void)
-{
-	printk(KERN_INFO "cx2388x alsa driver version %s loaded\n",
-	       CX88_VERSION);
-	return pci_register_driver(&cx88_audio_pci_driver);
-}
-
-/*
- * module remove
- */
-static void __exit cx88_audio_fini(void)
-{
-	pci_unregister_driver(&cx88_audio_pci_driver);
-}
-
-module_init(cx88_audio_init);
-module_exit(cx88_audio_fini);
+module_pci_driver(cx88_audio_pci_driver);
diff --git a/drivers/media/pci/cx88/cx88-mpeg.c b/drivers/media/pci/cx88/cx88-mpeg.c
index 2d3507eb4897..74b7b8614c23 100644
--- a/drivers/media/pci/cx88/cx88-mpeg.c
+++ b/drivers/media/pci/cx88/cx88-mpeg.c
@@ -499,7 +499,7 @@ static int cx8802_init_common(struct cx8802_dev *dev)
 
 	/* get irq */
 	err = request_irq(dev->pci->irq, cx8802_irq,
-			  IRQF_SHARED | IRQF_DISABLED, dev->core->name, dev);
+			  IRQF_SHARED, dev->core->name, dev);
 	if (err < 0) {
 		printk(KERN_ERR "%s: can't get IRQ %d\n",
 		       dev->core->name, dev->pci->irq);
@@ -520,7 +520,6 @@ static void cx8802_fini_common(struct cx8802_dev *dev)
 
 	/* unregister stuff */
 	free_irq(dev->pci->irq, dev);
-	pci_set_drvdata(dev->pci, NULL);
 
 	/* free memory */
 	btcx_riscmem_free(dev->pci,&dev->mpegq.stopper);
@@ -903,20 +902,8 @@ static struct pci_driver cx8802_pci_driver = {
 	.remove   = cx8802_remove,
 };
 
-static int __init cx8802_init(void)
-{
-	printk(KERN_INFO "cx88/2: cx2388x MPEG-TS Driver Manager version %s loaded\n",
-	       CX88_VERSION);
-	return pci_register_driver(&cx8802_pci_driver);
-}
-
-static void __exit cx8802_fini(void)
-{
-	pci_unregister_driver(&cx8802_pci_driver);
-}
+module_pci_driver(cx8802_pci_driver);
 
-module_init(cx8802_init);
-module_exit(cx8802_fini);
 EXPORT_SYMBOL(cx8802_buf_prepare);
 EXPORT_SYMBOL(cx8802_buf_queue);
 EXPORT_SYMBOL(cx8802_cancel_buffers);
diff --git a/drivers/media/pci/cx88/cx88-video.c b/drivers/media/pci/cx88/cx88-video.c
index ecf21d9f1f34..ed8cb9037b6f 100644
--- a/drivers/media/pci/cx88/cx88-video.c
+++ b/drivers/media/pci/cx88/cx88-video.c
@@ -1738,7 +1738,7 @@ static int cx8800_initdev(struct pci_dev *pci_dev,
 
 	/* get irq */
 	err = request_irq(pci_dev->irq, cx8800_irq,
-			  IRQF_SHARED | IRQF_DISABLED, core->name, dev);
+			  IRQF_SHARED, core->name, dev);
 	if (err < 0) {
 		printk(KERN_ERR "%s/0: can't get IRQ %d\n",
 		       core->name,pci_dev->irq);
@@ -1922,7 +1922,6 @@ static void cx8800_finidev(struct pci_dev *pci_dev)
 
 	free_irq(pci_dev->irq, dev);
 	cx8800_unregister_video(dev);
-	pci_set_drvdata(pci_dev, NULL);
 
 	/* free memory */
 	btcx_riscmem_free(dev->pci,&dev->vidq.stopper);
@@ -2039,17 +2038,4 @@ static struct pci_driver cx8800_pci_driver = {
 #endif
 };
 
-static int __init cx8800_init(void)
-{
-	printk(KERN_INFO "cx88/0: cx2388x v4l2 driver version %s loaded\n",
-	       CX88_VERSION);
-	return pci_register_driver(&cx8800_pci_driver);
-}
-
-static void __exit cx8800_fini(void)
-{
-	pci_unregister_driver(&cx8800_pci_driver);
-}
-
-module_init(cx8800_init);
-module_exit(cx8800_fini);
+module_pci_driver(cx8800_pci_driver);
diff --git a/drivers/media/pci/ddbridge/ddbridge-core.c b/drivers/media/pci/ddbridge/ddbridge-core.c
index 36e34522b9a8..9375f30d9a81 100644
--- a/drivers/media/pci/ddbridge/ddbridge-core.c
+++ b/drivers/media/pci/ddbridge/ddbridge-core.c
@@ -1544,7 +1544,7 @@ static void ddb_unmap(struct ddb *dev)
 
 static void ddb_remove(struct pci_dev *pdev)
 {
-	struct ddb *dev = (struct ddb *) pci_get_drvdata(pdev);
+	struct ddb *dev = pci_get_drvdata(pdev);
 
 	ddb_ports_detach(dev);
 	ddb_i2c_release(dev);
diff --git a/drivers/media/pci/dm1105/dm1105.c b/drivers/media/pci/dm1105/dm1105.c
index ab797fe466d2..e60ac35fc10c 100644
--- a/drivers/media/pci/dm1105/dm1105.c
+++ b/drivers/media/pci/dm1105/dm1105.c
@@ -1178,7 +1178,6 @@ err_pci_release_regions:
 err_pci_disable_device:
 	pci_disable_device(pdev);
 err_kfree:
-	pci_set_drvdata(pdev, NULL);
 	kfree(dev);
 	return ret;
 }
@@ -1202,8 +1201,7 @@ static void dm1105_remove(struct pci_dev *pdev)
 	dvb_dmxdev_release(&dev->dmxdev);
 	dvb_dmx_release(dvbdemux);
 	dvb_unregister_adapter(dvb_adapter);
-	if (&dev->i2c_adap)
-		i2c_del_adapter(&dev->i2c_adap);
+	i2c_del_adapter(&dev->i2c_adap);
 
 	dm1105_hw_exit(dev);
 	synchronize_irq(pdev->irq);
@@ -1211,7 +1209,6 @@ static void dm1105_remove(struct pci_dev *pdev)
 	pci_iounmap(pdev, dev->io_mem);
 	pci_release_regions(pdev);
 	pci_disable_device(pdev);
-	pci_set_drvdata(pdev, NULL);
 	dm1105_devcount--;
 	kfree(dev);
 }
diff --git a/drivers/media/pci/ivtv/ivtv-driver.c b/drivers/media/pci/ivtv/ivtv-driver.c
index c08ae3eb9554..802642d26643 100644
--- a/drivers/media/pci/ivtv/ivtv-driver.c
+++ b/drivers/media/pci/ivtv/ivtv-driver.c
@@ -1261,7 +1261,7 @@ static int ivtv_probe(struct pci_dev *pdev, const struct pci_device_id *pci_id)
 
 	/* Register IRQ */
 	retval = request_irq(itv->pdev->irq, ivtv_irq_handler,
-	     IRQF_SHARED | IRQF_DISABLED, itv->v4l2_dev.name, (void *)itv);
+	     IRQF_SHARED, itv->v4l2_dev.name, (void *)itv);
 	if (retval) {
 		IVTV_ERR("Failed to register irq %d\n", retval);
 		goto free_i2c;
diff --git a/drivers/media/pci/mantis/mantis_pci.c b/drivers/media/pci/mantis/mantis_pci.c
index a846036ea022..9e89e045213a 100644
--- a/drivers/media/pci/mantis/mantis_pci.c
+++ b/drivers/media/pci/mantis/mantis_pci.c
@@ -143,7 +143,6 @@ fail1:
 
 fail0:
 	dprintk(MANTIS_ERROR, 1, "ERROR: <%d> exiting", ret);
-	pci_set_drvdata(pdev, NULL);
 	return ret;
 }
 EXPORT_SYMBOL_GPL(mantis_pci_init);
@@ -161,7 +160,6 @@ void mantis_pci_exit(struct mantis_pci *mantis)
 	}
 
 	pci_disable_device(pdev);
-	pci_set_drvdata(pdev, NULL);
 }
 EXPORT_SYMBOL_GPL(mantis_pci_exit);
 
diff --git a/drivers/media/pci/meye/meye.c b/drivers/media/pci/meye/meye.c
index 2381b05432e6..54d5c821007c 100644
--- a/drivers/media/pci/meye/meye.c
+++ b/drivers/media/pci/meye/meye.c
@@ -1698,7 +1698,7 @@ static int meye_probe(struct pci_dev *pcidev, const struct pci_device_id *ent)
 
 	meye.mchip_irq = pcidev->irq;
 	if (request_irq(meye.mchip_irq, meye_irq,
-			IRQF_DISABLED | IRQF_SHARED, "meye", meye_irq)) {
+			IRQF_SHARED, "meye", meye_irq)) {
 		v4l2_err(v4l2_dev, "request_irq failed\n");
 		goto outreqirq;
 	}
diff --git a/drivers/media/pci/ngene/ngene-core.c b/drivers/media/pci/ngene/ngene-core.c
index 37ebc42392ad..970e83308525 100644
--- a/drivers/media/pci/ngene/ngene-core.c
+++ b/drivers/media/pci/ngene/ngene-core.c
@@ -1622,7 +1622,7 @@ static void ngene_unlink(struct ngene *dev)
 
 void ngene_shutdown(struct pci_dev *pdev)
 {
-	struct ngene *dev = (struct ngene *)pci_get_drvdata(pdev);
+	struct ngene *dev = pci_get_drvdata(pdev);
 
 	if (!dev || !shutdown_workaround)
 		return;
@@ -1648,7 +1648,6 @@ void ngene_remove(struct pci_dev *pdev)
 		cxd_detach(dev);
 	ngene_stop(dev);
 	ngene_release_buffers(dev);
-	pci_set_drvdata(pdev, NULL);
 	pci_disable_device(pdev);
 }
 
@@ -1702,6 +1701,5 @@ fail1:
 	ngene_release_buffers(dev);
 fail0:
 	pci_disable_device(pci_dev);
-	pci_set_drvdata(pci_dev, NULL);
 	return stat;
 }
diff --git a/drivers/media/pci/pluto2/pluto2.c b/drivers/media/pci/pluto2/pluto2.c
index 493828500055..8164d74b46a4 100644
--- a/drivers/media/pci/pluto2/pluto2.c
+++ b/drivers/media/pci/pluto2/pluto2.c
@@ -736,7 +736,6 @@ err_pci_release_regions:
 err_pci_disable_device:
 	pci_disable_device(pdev);
 err_kfree:
-	pci_set_drvdata(pdev, NULL);
 	kfree(pluto);
 	goto out;
 }
@@ -765,7 +764,6 @@ static void pluto2_remove(struct pci_dev *pdev)
 	pci_iounmap(pdev, pluto->io_mem);
 	pci_release_regions(pdev);
 	pci_disable_device(pdev);
-	pci_set_drvdata(pdev, NULL);
 	kfree(pluto);
 }
 
diff --git a/drivers/media/pci/pt1/pt1.c b/drivers/media/pci/pt1/pt1.c
index 75ce14229e03..db887b0c37b1 100644
--- a/drivers/media/pci/pt1/pt1.c
+++ b/drivers/media/pci/pt1/pt1.c
@@ -1076,7 +1076,6 @@ static void pt1_remove(struct pci_dev *pdev)
 	pt1_update_power(pt1);
 	pt1_cleanup_adapters(pt1);
 	i2c_del_adapter(&pt1->i2c_adap);
-	pci_set_drvdata(pdev, NULL);
 	kfree(pt1);
 	pci_iounmap(pdev, regs);
 	pci_release_regions(pdev);
@@ -1198,7 +1197,6 @@ err_i2c_del_adapter:
 err_pt1_cleanup_adapters:
 	pt1_cleanup_adapters(pt1);
 err_kfree:
-	pci_set_drvdata(pdev, NULL);
 	kfree(pt1);
 err_pci_iounmap:
 	pci_iounmap(pdev, regs);
diff --git a/drivers/media/pci/saa7134/saa7134-alsa.c b/drivers/media/pci/saa7134/saa7134-alsa.c
index dbcdfbf8aed0..dd67c8a400cc 100644
--- a/drivers/media/pci/saa7134/saa7134-alsa.c
+++ b/drivers/media/pci/saa7134/saa7134-alsa.c
@@ -1096,7 +1096,7 @@ static int alsa_card_saa7134_create(struct saa7134_dev *dev, int devnum)
 
 
 	err = request_irq(dev->pci->irq, saa7134_alsa_irq,
-				IRQF_SHARED | IRQF_DISABLED, dev->name,
+				IRQF_SHARED, dev->name,
 				(void*) &dev->dmasound);
 
 	if (err < 0) {
diff --git a/drivers/media/pci/saa7134/saa7134-core.c b/drivers/media/pci/saa7134/saa7134-core.c
index 45f0aca597ae..27d7ee709c58 100644
--- a/drivers/media/pci/saa7134/saa7134-core.c
+++ b/drivers/media/pci/saa7134/saa7134-core.c
@@ -992,7 +992,7 @@ static int saa7134_initdev(struct pci_dev *pci_dev,
 
 	/* get irq */
 	err = request_irq(pci_dev->irq, saa7134_irq,
-			  IRQF_SHARED | IRQF_DISABLED, dev->name, dev);
+			  IRQF_SHARED, dev->name, dev);
 	if (err < 0) {
 		printk(KERN_ERR "%s: can't get IRQ %d\n",
 		       dev->name,pci_dev->irq);
diff --git a/drivers/media/pci/saa7164/saa7164-core.c b/drivers/media/pci/saa7164/saa7164-core.c
index d37ee37aaefe..57ef5456f1e8 100644
--- a/drivers/media/pci/saa7164/saa7164-core.c
+++ b/drivers/media/pci/saa7164/saa7164-core.c
@@ -1232,7 +1232,7 @@ static int saa7164_initdev(struct pci_dev *pci_dev,
 	}
 
 	err = request_irq(pci_dev->irq, saa7164_irq,
-		IRQF_SHARED | IRQF_DISABLED, dev->name, dev);
+		IRQF_SHARED, dev->name, dev);
 	if (err < 0) {
 		printk(KERN_ERR "%s: can't get IRQ %d\n", dev->name,
 			pci_dev->irq);
@@ -1439,7 +1439,6 @@ static void saa7164_finidev(struct pci_dev *pci_dev)
 
 	/* unregister stuff */
 	free_irq(pci_dev->irq, dev);
-	pci_set_drvdata(pci_dev, NULL);
 
 	mutex_lock(&devlist);
 	list_del(&dev->devlist);
diff --git a/drivers/media/pci/ttpci/av7110_hw.c b/drivers/media/pci/ttpci/av7110_hw.c
index f1cbfe526989..6299d5dadb82 100644
--- a/drivers/media/pci/ttpci/av7110_hw.c
+++ b/drivers/media/pci/ttpci/av7110_hw.c
@@ -22,7 +22,7 @@
  * Foundation, Inc., 59 Temple Place - Suite 330, Boston, MA 02111-1307, USA.
  * Or, point your browser to http://www.gnu.org/copyleft/gpl.html
  *
- * the project's page is at http://www.linuxtv.org/ 
+ * the project's page is at http://www.linuxtv.org/
  */
 
 /* for debugging ARM communication: */
@@ -40,6 +40,14 @@
 
 #define _NOHANDSHAKE
 
+/*
+ * Max transfer size done by av7110_fw_cmd()
+ *
+ * The maximum size passed to this function is 6 bytes. The buffer also
+ * uses two additional ones for type and size. So, 8 bytes is enough.
+ */
+#define MAX_XFER_SIZE  8
+
 /****************************************************************************
  * DEBI functions
  ****************************************************************************/
@@ -488,11 +496,18 @@ static int av7110_send_fw_cmd(struct av7110 *av7110, u16* buf, int length)
 int av7110_fw_cmd(struct av7110 *av7110, int type, int com, int num, ...)
 {
 	va_list args;
-	u16 buf[num + 2];
+	u16 buf[MAX_XFER_SIZE];
 	int i, ret;
 
 //	dprintk(4, "%p\n", av7110);
 
+	if (2 + num > sizeof(buf)) {
+		printk(KERN_WARNING
+		       "%s: %s len=%d is too big!\n",
+		       KBUILD_MODNAME, __func__, num);
+		return -EINVAL;
+	}
+
 	buf[0] = ((type << 8) | com);
 	buf[1] = num;
 
diff --git a/drivers/media/pci/zoran/Kconfig b/drivers/media/pci/zoran/Kconfig
index 26ca8702e33f..39ec35bd21a5 100644
--- a/drivers/media/pci/zoran/Kconfig
+++ b/drivers/media/pci/zoran/Kconfig
@@ -1,6 +1,7 @@
 config VIDEO_ZORAN
 	tristate "Zoran ZR36057/36067 Video For Linux"
 	depends on PCI && I2C_ALGOBIT && VIDEO_V4L2 && VIRT_TO_BUS
+	depends on !ALPHA
 	help
 	  Say Y for support for MJPEG capture cards based on the Zoran
 	  36057/36067 PCI controller chipset. This includes the Iomega
diff --git a/drivers/media/pci/zoran/zoran_card.c b/drivers/media/pci/zoran/zoran_card.c
index 923d59a321f8..cec5b7553f28 100644
--- a/drivers/media/pci/zoran/zoran_card.c
+++ b/drivers/media/pci/zoran/zoran_card.c
@@ -1293,7 +1293,7 @@ static int zoran_probe(struct pci_dev *pdev, const struct pci_device_id *ent)
 	}
 
 	result = request_irq(zr->pci_dev->irq, zoran_irq,
-			     IRQF_SHARED | IRQF_DISABLED, ZR_DEVNAME(zr), zr);
+			     IRQF_SHARED, ZR_DEVNAME(zr), zr);
 	if (result < 0) {
 		if (result == -EINVAL) {
 			dprintk(1,
diff --git a/drivers/media/platform/Kconfig b/drivers/media/platform/Kconfig
index eb70dda8cbf3..d7f0249e4050 100644
--- a/drivers/media/platform/Kconfig
+++ b/drivers/media/platform/Kconfig
@@ -143,6 +143,7 @@ if V4L_MEM2MEM_DRIVERS
 config VIDEO_CODA
 	tristate "Chips&Media Coda multi-standard codec IP"
 	depends on VIDEO_DEV && VIDEO_V4L2 && ARCH_MXC
+	select SRAM
 	select VIDEOBUF2_DMA_CONTIG
 	select V4L2_MEM2MEM_DEV
 	---help---
@@ -212,7 +213,7 @@ config VIDEO_SH_VEU
 
 config VIDEO_RENESAS_VSP1
 	tristate "Renesas VSP1 Video Processing Engine"
-	depends on VIDEO_V4L2 && VIDEO_V4L2_SUBDEV_API
+	depends on VIDEO_V4L2 && VIDEO_V4L2_SUBDEV_API && HAS_DMA
 	select VIDEOBUF2_DMA_CONTIG
 	---help---
 	  This is a V4L2 driver for the Renesas VSP1 video processing engine.
@@ -220,6 +221,22 @@ config VIDEO_RENESAS_VSP1
 	  To compile this driver as a module, choose M here: the module
 	  will be called vsp1.
 
+config VIDEO_TI_VPE
+	tristate "TI VPE (Video Processing Engine) driver"
+	depends on VIDEO_DEV && VIDEO_V4L2 && SOC_DRA7XX
+	select VIDEOBUF2_DMA_CONTIG
+	select V4L2_MEM2MEM_DEV
+	default n
+	---help---
+	  Support for the TI VPE(Video Processing Engine) block
+	  found on DRA7XX SoC.
+
+config VIDEO_TI_VPE_DEBUG
+	bool "VPE debug messages"
+	depends on VIDEO_TI_VPE
+	---help---
+	  Enable debug messages on VPE driver.
+
 endif # V4L_MEM2MEM_DRIVERS
 
 menuconfig V4L_TEST_DRIVERS
diff --git a/drivers/media/platform/Makefile b/drivers/media/platform/Makefile
index 4e4da482c522..1348ba1faf92 100644
--- a/drivers/media/platform/Makefile
+++ b/drivers/media/platform/Makefile
@@ -22,6 +22,8 @@ obj-$(CONFIG_VIDEO_VIVI) += vivi.o
 
 obj-$(CONFIG_VIDEO_MEM2MEM_TESTDEV) += mem2mem_testdev.o
 
+obj-$(CONFIG_VIDEO_TI_VPE)		+= ti-vpe/
+
 obj-$(CONFIG_VIDEO_MX2_EMMAPRP)		+= mx2_emmaprp.o
 obj-$(CONFIG_VIDEO_CODA) 		+= coda.o
 
diff --git a/drivers/media/platform/blackfin/bfin_capture.c b/drivers/media/platform/blackfin/bfin_capture.c
index 4c1105977090..281916591437 100644
--- a/drivers/media/platform/blackfin/bfin_capture.c
+++ b/drivers/media/platform/blackfin/bfin_capture.c
@@ -422,7 +422,7 @@ static int bcap_start_streaming(struct vb2_queue *vq, unsigned int count)
 		return ret;
 	}
 
-	INIT_COMPLETION(bcap_dev->comp);
+	reinit_completion(&bcap_dev->comp);
 	bcap_dev->stop = false;
 	return 0;
 }
diff --git a/drivers/media/platform/coda.c b/drivers/media/platform/coda.c
index 4993610051ee..bd72fb97fea5 100644
--- a/drivers/media/platform/coda.c
+++ b/drivers/media/platform/coda.c
@@ -39,7 +39,7 @@
 
 #define CODA_NAME		"coda"
 
-#define CODA_MAX_INSTANCES	4
+#define CODADX6_MAX_INSTANCES	4
 
 #define CODA_FMO_BUF_SIZE	32
 #define CODADX6_WORK_BUF_SIZE	(288 * 1024 + CODA_FMO_BUF_SIZE * 8 * 1024)
@@ -54,8 +54,6 @@
 
 #define CODA_MAX_FRAMEBUFFERS	8
 
-#define MAX_W		8192
-#define MAX_H		8192
 #define CODA_MAX_FRAME_SIZE	0x100000
 #define FMO_SLICE_SAVE_BUF_SIZE         (32)
 #define CODA_DEFAULT_GAMMA		4096
@@ -394,14 +392,57 @@ static struct coda_codec *coda_find_codec(struct coda_dev *dev, int src_fourcc,
 	return &codecs[k];
 }
 
+static void coda_get_max_dimensions(struct coda_dev *dev,
+				    struct coda_codec *codec,
+				    int *max_w, int *max_h)
+{
+	struct coda_codec *codecs = dev->devtype->codecs;
+	int num_codecs = dev->devtype->num_codecs;
+	unsigned int w, h;
+	int k;
+
+	if (codec) {
+		w = codec->max_w;
+		h = codec->max_h;
+	} else {
+		for (k = 0, w = 0, h = 0; k < num_codecs; k++) {
+			w = max(w, codecs[k].max_w);
+			h = max(h, codecs[k].max_h);
+		}
+	}
+
+	if (max_w)
+		*max_w = w;
+	if (max_h)
+		*max_h = h;
+}
+
+static char *coda_product_name(int product)
+{
+	static char buf[9];
+
+	switch (product) {
+	case CODA_DX6:
+		return "CodaDx6";
+	case CODA_7541:
+		return "CODA7541";
+	default:
+		snprintf(buf, sizeof(buf), "(0x%04x)", product);
+		return buf;
+	}
+}
+
 /*
  * V4L2 ioctl() operations.
  */
-static int vidioc_querycap(struct file *file, void *priv,
-			   struct v4l2_capability *cap)
+static int coda_querycap(struct file *file, void *priv,
+			 struct v4l2_capability *cap)
 {
+	struct coda_ctx *ctx = fh_to_ctx(priv);
+
 	strlcpy(cap->driver, CODA_NAME, sizeof(cap->driver));
-	strlcpy(cap->card, CODA_NAME, sizeof(cap->card));
+	strlcpy(cap->card, coda_product_name(ctx->dev->devtype->product),
+		sizeof(cap->card));
 	strlcpy(cap->bus_info, "platform:" CODA_NAME, sizeof(cap->bus_info));
 	/*
 	 * This is only a mem-to-mem video device. The capture and output
@@ -457,6 +498,8 @@ static int enum_fmt(void *priv, struct v4l2_fmtdesc *f,
 		fmt = &formats[i];
 		strlcpy(f->description, fmt->name, sizeof(f->description));
 		f->pixelformat = fmt->fourcc;
+		if (!coda_format_is_yuv(fmt->fourcc))
+			f->flags |= V4L2_FMT_FLAG_COMPRESSED;
 		return 0;
 	}
 
@@ -464,8 +507,8 @@ static int enum_fmt(void *priv, struct v4l2_fmtdesc *f,
 	return -EINVAL;
 }
 
-static int vidioc_enum_fmt_vid_cap(struct file *file, void *priv,
-				   struct v4l2_fmtdesc *f)
+static int coda_enum_fmt_vid_cap(struct file *file, void *priv,
+				 struct v4l2_fmtdesc *f)
 {
 	struct coda_ctx *ctx = fh_to_ctx(priv);
 	struct vb2_queue *src_vq;
@@ -483,13 +526,14 @@ static int vidioc_enum_fmt_vid_cap(struct file *file, void *priv,
 	return enum_fmt(priv, f, V4L2_BUF_TYPE_VIDEO_CAPTURE, 0);
 }
 
-static int vidioc_enum_fmt_vid_out(struct file *file, void *priv,
-				   struct v4l2_fmtdesc *f)
+static int coda_enum_fmt_vid_out(struct file *file, void *priv,
+				 struct v4l2_fmtdesc *f)
 {
 	return enum_fmt(priv, f, V4L2_BUF_TYPE_VIDEO_OUTPUT, 0);
 }
 
-static int vidioc_g_fmt(struct file *file, void *priv, struct v4l2_format *f)
+static int coda_g_fmt(struct file *file, void *priv,
+		      struct v4l2_format *f)
 {
 	struct vb2_queue *vq;
 	struct coda_q_data *q_data;
@@ -516,8 +560,11 @@ static int vidioc_g_fmt(struct file *file, void *priv, struct v4l2_format *f)
 	return 0;
 }
 
-static int vidioc_try_fmt(struct coda_codec *codec, struct v4l2_format *f)
+static int coda_try_fmt(struct coda_ctx *ctx, struct coda_codec *codec,
+			struct v4l2_format *f)
 {
+	struct coda_dev *dev = ctx->dev;
+	struct coda_q_data *q_data;
 	unsigned int max_w, max_h;
 	enum v4l2_field field;
 
@@ -531,32 +578,48 @@ static int vidioc_try_fmt(struct coda_codec *codec, struct v4l2_format *f)
 	 * if any of the dimensions is unsupported */
 	f->fmt.pix.field = field;
 
-	if (codec) {
-		max_w = codec->max_w;
-		max_h = codec->max_h;
-	} else {
-		max_w = MAX_W;
-		max_h = MAX_H;
+	coda_get_max_dimensions(dev, codec, &max_w, &max_h);
+	v4l_bound_align_image(&f->fmt.pix.width, MIN_W, max_w, W_ALIGN,
+			      &f->fmt.pix.height, MIN_H, max_h, H_ALIGN,
+			      S_ALIGN);
+
+	switch (f->fmt.pix.pixelformat) {
+	case V4L2_PIX_FMT_YUV420:
+	case V4L2_PIX_FMT_YVU420:
+	case V4L2_PIX_FMT_H264:
+	case V4L2_PIX_FMT_MPEG4:
+	case V4L2_PIX_FMT_JPEG:
+		break;
+	default:
+		q_data = get_q_data(ctx, f->type);
+		f->fmt.pix.pixelformat = q_data->fourcc;
 	}
-	v4l_bound_align_image(&f->fmt.pix.width, MIN_W, max_w,
-			      W_ALIGN, &f->fmt.pix.height,
-			      MIN_H, max_h, H_ALIGN, S_ALIGN);
 
-	if (coda_format_is_yuv(f->fmt.pix.pixelformat)) {
+	switch (f->fmt.pix.pixelformat) {
+	case V4L2_PIX_FMT_YUV420:
+	case V4L2_PIX_FMT_YVU420:
 		/* Frame stride must be multiple of 8 */
 		f->fmt.pix.bytesperline = round_up(f->fmt.pix.width, 8);
 		f->fmt.pix.sizeimage = f->fmt.pix.bytesperline *
 					f->fmt.pix.height * 3 / 2;
-	} else { /*encoded formats h.264/mpeg4 */
+		break;
+	case V4L2_PIX_FMT_H264:
+	case V4L2_PIX_FMT_MPEG4:
+	case V4L2_PIX_FMT_JPEG:
 		f->fmt.pix.bytesperline = 0;
 		f->fmt.pix.sizeimage = CODA_MAX_FRAME_SIZE;
+		break;
+	default:
+		BUG();
 	}
 
+	f->fmt.pix.priv = 0;
+
 	return 0;
 }
 
-static int vidioc_try_fmt_vid_cap(struct file *file, void *priv,
-				  struct v4l2_format *f)
+static int coda_try_fmt_vid_cap(struct file *file, void *priv,
+				struct v4l2_format *f)
 {
 	struct coda_ctx *ctx = fh_to_ctx(priv);
 	struct coda_codec *codec;
@@ -584,7 +647,7 @@ static int vidioc_try_fmt_vid_cap(struct file *file, void *priv,
 
 	f->fmt.pix.colorspace = ctx->colorspace;
 
-	ret = vidioc_try_fmt(codec, f);
+	ret = coda_try_fmt(ctx, codec, f);
 	if (ret < 0)
 		return ret;
 
@@ -600,8 +663,8 @@ static int vidioc_try_fmt_vid_cap(struct file *file, void *priv,
 	return 0;
 }
 
-static int vidioc_try_fmt_vid_out(struct file *file, void *priv,
-				  struct v4l2_format *f)
+static int coda_try_fmt_vid_out(struct file *file, void *priv,
+				struct v4l2_format *f)
 {
 	struct coda_ctx *ctx = fh_to_ctx(priv);
 	struct coda_codec *codec;
@@ -613,10 +676,10 @@ static int vidioc_try_fmt_vid_out(struct file *file, void *priv,
 	if (!f->fmt.pix.colorspace)
 		f->fmt.pix.colorspace = V4L2_COLORSPACE_REC709;
 
-	return vidioc_try_fmt(codec, f);
+	return coda_try_fmt(ctx, codec, f);
 }
 
-static int vidioc_s_fmt(struct coda_ctx *ctx, struct v4l2_format *f)
+static int coda_s_fmt(struct coda_ctx *ctx, struct v4l2_format *f)
 {
 	struct coda_q_data *q_data;
 	struct vb2_queue *vq;
@@ -646,61 +709,62 @@ static int vidioc_s_fmt(struct coda_ctx *ctx, struct v4l2_format *f)
 	return 0;
 }
 
-static int vidioc_s_fmt_vid_cap(struct file *file, void *priv,
-				struct v4l2_format *f)
+static int coda_s_fmt_vid_cap(struct file *file, void *priv,
+			      struct v4l2_format *f)
 {
 	struct coda_ctx *ctx = fh_to_ctx(priv);
 	int ret;
 
-	ret = vidioc_try_fmt_vid_cap(file, priv, f);
+	ret = coda_try_fmt_vid_cap(file, priv, f);
 	if (ret)
 		return ret;
 
-	return vidioc_s_fmt(ctx, f);
+	return coda_s_fmt(ctx, f);
 }
 
-static int vidioc_s_fmt_vid_out(struct file *file, void *priv,
-				struct v4l2_format *f)
+static int coda_s_fmt_vid_out(struct file *file, void *priv,
+			      struct v4l2_format *f)
 {
 	struct coda_ctx *ctx = fh_to_ctx(priv);
 	int ret;
 
-	ret = vidioc_try_fmt_vid_out(file, priv, f);
+	ret = coda_try_fmt_vid_out(file, priv, f);
 	if (ret)
 		return ret;
 
-	ret = vidioc_s_fmt(ctx, f);
+	ret = coda_s_fmt(ctx, f);
 	if (ret)
 		ctx->colorspace = f->fmt.pix.colorspace;
 
 	return ret;
 }
 
-static int vidioc_reqbufs(struct file *file, void *priv,
-			  struct v4l2_requestbuffers *reqbufs)
+static int coda_reqbufs(struct file *file, void *priv,
+			struct v4l2_requestbuffers *reqbufs)
 {
 	struct coda_ctx *ctx = fh_to_ctx(priv);
 
 	return v4l2_m2m_reqbufs(file, ctx->m2m_ctx, reqbufs);
 }
 
-static int vidioc_querybuf(struct file *file, void *priv,
-			   struct v4l2_buffer *buf)
+static int coda_querybuf(struct file *file, void *priv,
+			 struct v4l2_buffer *buf)
 {
 	struct coda_ctx *ctx = fh_to_ctx(priv);
 
 	return v4l2_m2m_querybuf(file, ctx->m2m_ctx, buf);
 }
 
-static int vidioc_qbuf(struct file *file, void *priv, struct v4l2_buffer *buf)
+static int coda_qbuf(struct file *file, void *priv,
+		     struct v4l2_buffer *buf)
 {
 	struct coda_ctx *ctx = fh_to_ctx(priv);
 
 	return v4l2_m2m_qbuf(file, ctx->m2m_ctx, buf);
 }
 
-static int vidioc_expbuf(struct file *file, void *priv,
-			 struct v4l2_exportbuffer *eb)
+static int coda_expbuf(struct file *file, void *priv,
+		       struct v4l2_exportbuffer *eb)
 {
 	struct coda_ctx *ctx = fh_to_ctx(priv);
 
@@ -718,7 +782,8 @@ static bool coda_buf_is_end_of_stream(struct coda_ctx *ctx,
 		(buf->sequence == (ctx->qsequence - 1)));
 }
 
-static int vidioc_dqbuf(struct file *file, void *priv, struct v4l2_buffer *buf)
+static int coda_dqbuf(struct file *file, void *priv,
+		      struct v4l2_buffer *buf)
 {
 	struct coda_ctx *ctx = fh_to_ctx(priv);
 	int ret;
@@ -738,24 +803,24 @@ static int vidioc_dqbuf(struct file *file, void *priv, struct v4l2_buffer *buf)
 	return ret;
 }
 
-static int vidioc_create_bufs(struct file *file, void *priv,
-			      struct v4l2_create_buffers *create)
+static int coda_create_bufs(struct file *file, void *priv,
+			    struct v4l2_create_buffers *create)
 {
 	struct coda_ctx *ctx = fh_to_ctx(priv);
 
 	return v4l2_m2m_create_bufs(file, ctx->m2m_ctx, create);
 }
 
-static int vidioc_streamon(struct file *file, void *priv,
-			   enum v4l2_buf_type type)
+static int coda_streamon(struct file *file, void *priv,
+			 enum v4l2_buf_type type)
 {
 	struct coda_ctx *ctx = fh_to_ctx(priv);
 
 	return v4l2_m2m_streamon(file, ctx->m2m_ctx, type);
 }
 
-static int vidioc_streamoff(struct file *file, void *priv,
-			    enum v4l2_buf_type type)
+static int coda_streamoff(struct file *file, void *priv,
+			  enum v4l2_buf_type type)
 {
 	struct coda_ctx *ctx = fh_to_ctx(priv);
 	int ret;
@@ -772,23 +837,34 @@ static int vidioc_streamoff(struct file *file, void *priv,
 	return ret;
 }
 
-static int vidioc_decoder_cmd(struct file *file, void *fh,
-			      struct v4l2_decoder_cmd *dc)
+static int coda_try_decoder_cmd(struct file *file, void *fh,
+				struct v4l2_decoder_cmd *dc)
 {
-	struct coda_ctx *ctx = fh_to_ctx(fh);
-
 	if (dc->cmd != V4L2_DEC_CMD_STOP)
 		return -EINVAL;
 
-	if ((dc->flags & V4L2_DEC_CMD_STOP_TO_BLACK) ||
-	    (dc->flags & V4L2_DEC_CMD_STOP_IMMEDIATELY))
+	if (dc->flags & V4L2_DEC_CMD_STOP_TO_BLACK)
 		return -EINVAL;
 
-	if (dc->stop.pts != 0)
+	if (!(dc->flags & V4L2_DEC_CMD_STOP_IMMEDIATELY) && (dc->stop.pts != 0))
 		return -EINVAL;
 
+	return 0;
+}
+
+static int coda_decoder_cmd(struct file *file, void *fh,
+			    struct v4l2_decoder_cmd *dc)
+{
+	struct coda_ctx *ctx = fh_to_ctx(fh);
+	int ret;
+
+	ret = coda_try_decoder_cmd(file, fh, dc);
+	if (ret < 0)
+		return ret;
+
+	/* Ignore decoder stop command silently in encoder context */
 	if (ctx->inst_type != CODA_INST_DECODER)
-		return -EINVAL;
+		return 0;
 
 	/* Set the strem-end flag on this context */
 	ctx->bit_stream_param |= CODA_BIT_STREAM_END_FLAG;
@@ -796,8 +872,8 @@ static int vidioc_decoder_cmd(struct file *file, void *fh,
 	return 0;
 }
 
-static int vidioc_subscribe_event(struct v4l2_fh *fh,
-				  const struct v4l2_event_subscription *sub)
+static int coda_subscribe_event(struct v4l2_fh *fh,
+				const struct v4l2_event_subscription *sub)
 {
 	switch (sub->type) {
 	case V4L2_EVENT_EOS:
@@ -808,32 +884,33 @@ static int vidioc_subscribe_event(struct v4l2_fh *fh,
 }
 
 static const struct v4l2_ioctl_ops coda_ioctl_ops = {
-	.vidioc_querycap	= vidioc_querycap,
+	.vidioc_querycap	= coda_querycap,
 
-	.vidioc_enum_fmt_vid_cap = vidioc_enum_fmt_vid_cap,
-	.vidioc_g_fmt_vid_cap	= vidioc_g_fmt,
-	.vidioc_try_fmt_vid_cap	= vidioc_try_fmt_vid_cap,
-	.vidioc_s_fmt_vid_cap	= vidioc_s_fmt_vid_cap,
+	.vidioc_enum_fmt_vid_cap = coda_enum_fmt_vid_cap,
+	.vidioc_g_fmt_vid_cap	= coda_g_fmt,
+	.vidioc_try_fmt_vid_cap	= coda_try_fmt_vid_cap,
+	.vidioc_s_fmt_vid_cap	= coda_s_fmt_vid_cap,
 
-	.vidioc_enum_fmt_vid_out = vidioc_enum_fmt_vid_out,
-	.vidioc_g_fmt_vid_out	= vidioc_g_fmt,
-	.vidioc_try_fmt_vid_out	= vidioc_try_fmt_vid_out,
-	.vidioc_s_fmt_vid_out	= vidioc_s_fmt_vid_out,
+	.vidioc_enum_fmt_vid_out = coda_enum_fmt_vid_out,
+	.vidioc_g_fmt_vid_out	= coda_g_fmt,
+	.vidioc_try_fmt_vid_out	= coda_try_fmt_vid_out,
+	.vidioc_s_fmt_vid_out	= coda_s_fmt_vid_out,
 
-	.vidioc_reqbufs		= vidioc_reqbufs,
-	.vidioc_querybuf	= vidioc_querybuf,
+	.vidioc_reqbufs		= coda_reqbufs,
+	.vidioc_querybuf	= coda_querybuf,
 
-	.vidioc_qbuf		= vidioc_qbuf,
-	.vidioc_expbuf		= vidioc_expbuf,
-	.vidioc_dqbuf		= vidioc_dqbuf,
-	.vidioc_create_bufs	= vidioc_create_bufs,
+	.vidioc_qbuf		= coda_qbuf,
+	.vidioc_expbuf		= coda_expbuf,
+	.vidioc_dqbuf		= coda_dqbuf,
+	.vidioc_create_bufs	= coda_create_bufs,
 
-	.vidioc_streamon	= vidioc_streamon,
-	.vidioc_streamoff	= vidioc_streamoff,
+	.vidioc_streamon	= coda_streamon,
+	.vidioc_streamoff	= coda_streamoff,
 
-	.vidioc_decoder_cmd	= vidioc_decoder_cmd,
+	.vidioc_try_decoder_cmd	= coda_try_decoder_cmd,
+	.vidioc_decoder_cmd	= coda_decoder_cmd,
 
-	.vidioc_subscribe_event = vidioc_subscribe_event,
+	.vidioc_subscribe_event = coda_subscribe_event,
 	.vidioc_unsubscribe_event = v4l2_event_unsubscribe,
 };
 
@@ -1928,8 +2005,9 @@ static int coda_start_streaming(struct vb2_queue *q, unsigned int count)
 	if (!(ctx->streamon_out & ctx->streamon_cap))
 		return 0;
 
-	/* Allow device_run with no buffers queued and after streamoff */
-	v4l2_m2m_set_src_buffered(ctx->m2m_ctx, true);
+	/* Allow decoder device_run with no new buffers queued */
+	if (ctx->inst_type == CODA_INST_DECODER)
+		v4l2_m2m_set_src_buffered(ctx->m2m_ctx, true);
 
 	ctx->gopcounter = ctx->params.gop_size - 1;
 	buf = v4l2_m2m_next_dst_buf(ctx->m2m_ctx);
@@ -2071,10 +2149,8 @@ static int coda_start_streaming(struct vb2_queue *q, unsigned int count)
 	coda_setup_iram(ctx);
 
 	if (dst_fourcc == V4L2_PIX_FMT_H264) {
-		value  = (FMO_SLICE_SAVE_BUF_SIZE << 7);
-		value |= (0 & CODA_FMOPARAM_TYPE_MASK) << CODA_FMOPARAM_TYPE_OFFSET;
-		value |=  0 & CODA_FMOPARAM_SLICENUM_MASK;
 		if (dev->devtype->product == CODA_DX6) {
+			value = FMO_SLICE_SAVE_BUF_SIZE << 7;
 			coda_write(dev, value, CODADX6_CMD_ENC_SEQ_FMO);
 		} else {
 			coda_write(dev, ctx->iram_info.search_ram_paddr,
@@ -2371,7 +2447,13 @@ static int coda_queue_init(void *priv, struct vb2_queue *src_vq,
 
 static int coda_next_free_instance(struct coda_dev *dev)
 {
-	return ffz(dev->instance_mask);
+	int idx = ffz(dev->instance_mask);
+
+	if ((idx < 0) ||
+	    (dev->devtype->product == CODA_DX6 && idx > CODADX6_MAX_INSTANCES))
+		return -EBUSY;
+
+	return idx;
 }
 
 static int coda_open(struct file *file)
@@ -2386,8 +2468,8 @@ static int coda_open(struct file *file)
 		return -ENOMEM;
 
 	idx = coda_next_free_instance(dev);
-	if (idx >= CODA_MAX_INSTANCES) {
-		ret = -EBUSY;
+	if (idx < 0) {
+		ret = idx;
 		goto err_coda_max;
 	}
 	set_bit(idx, &dev->instance_mask);
@@ -2719,7 +2801,6 @@ static void coda_finish_encode(struct coda_ctx *ctx)
 	dst_buf = v4l2_m2m_dst_buf_remove(ctx->m2m_ctx);
 
 	/* Get results from the coda */
-	coda_read(dev, CODA_RET_ENC_PIC_TYPE);
 	start_ptr = coda_read(dev, CODA_CMD_ENC_PIC_BB_START);
 	wr_ptr = coda_read(dev, CODA_REG_BIT_WR_PTR(ctx->reg_idx));
 
@@ -2739,7 +2820,7 @@ static void coda_finish_encode(struct coda_ctx *ctx)
 	coda_read(dev, CODA_RET_ENC_PIC_SLICE_NUM);
 	coda_read(dev, CODA_RET_ENC_PIC_FLAG);
 
-	if (src_buf->v4l2_buf.flags & V4L2_BUF_FLAG_KEYFRAME) {
+	if (coda_read(dev, CODA_RET_ENC_PIC_TYPE) == 0) {
 		dst_buf->v4l2_buf.flags |= V4L2_BUF_FLAG_KEYFRAME;
 		dst_buf->v4l2_buf.flags &= ~V4L2_BUF_FLAG_PFRAME;
 	} else {
@@ -2861,21 +2942,6 @@ static bool coda_firmware_supported(u32 vernum)
 	return false;
 }
 
-static char *coda_product_name(int product)
-{
-	static char buf[9];
-
-	switch (product) {
-	case CODA_DX6:
-		return "CodaDx6";
-	case CODA_7541:
-		return "CODA7541";
-	default:
-		snprintf(buf, sizeof(buf), "(0x%04x)", product);
-		return buf;
-	}
-}
-
 static int coda_hw_init(struct coda_dev *dev)
 {
 	u16 product, major, minor, release;
diff --git a/drivers/media/platform/davinci/vpbe_display.c b/drivers/media/platform/davinci/vpbe_display.c
index 04609cc6eba7..eac472b5ae83 100644
--- a/drivers/media/platform/davinci/vpbe_display.c
+++ b/drivers/media/platform/davinci/vpbe_display.c
@@ -1785,7 +1785,7 @@ static int vpbe_display_probe(struct platform_device *pdev)
 	}
 
 	irq = res->start;
-	err = devm_request_irq(&pdev->dev, irq, venc_isr, IRQF_DISABLED,
+	err = devm_request_irq(&pdev->dev, irq, venc_isr, 0,
 			       VPBE_DISPLAY_DRIVER, disp_dev);
 	if (err) {
 		v4l2_err(&disp_dev->vpbe_dev->v4l2_dev,
diff --git a/drivers/media/platform/davinci/vpfe_capture.c b/drivers/media/platform/davinci/vpfe_capture.c
index 93609091cb23..d762246eabf5 100644
--- a/drivers/media/platform/davinci/vpfe_capture.c
+++ b/drivers/media/platform/davinci/vpfe_capture.c
@@ -688,7 +688,7 @@ static int vpfe_attach_irq(struct vpfe_device *vpfe_dev)
 	frame_format = ccdc_dev->hw_ops.get_frame_format();
 	if (frame_format == CCDC_FRMFMT_PROGRESSIVE) {
 		return request_irq(vpfe_dev->ccdc_irq1, vdint1_isr,
-				    IRQF_DISABLED, "vpfe_capture1",
+				    0, "vpfe_capture1",
 				    vpfe_dev);
 	}
 	return 0;
@@ -1863,7 +1863,7 @@ static int vpfe_probe(struct platform_device *pdev)
 	}
 	vpfe_dev->ccdc_irq1 = res1->start;
 
-	ret = request_irq(vpfe_dev->ccdc_irq0, vpfe_isr, IRQF_DISABLED,
+	ret = request_irq(vpfe_dev->ccdc_irq0, vpfe_isr, 0,
 			  "vpfe_capture0", vpfe_dev);
 
 	if (0 != ret) {
diff --git a/drivers/media/platform/davinci/vpif_capture.c b/drivers/media/platform/davinci/vpif_capture.c
index 1089834a4efe..52ac5e6c8625 100644
--- a/drivers/media/platform/davinci/vpif_capture.c
+++ b/drivers/media/platform/davinci/vpif_capture.c
@@ -2154,7 +2154,7 @@ static __init int vpif_probe(struct platform_device *pdev)
 
 			if (!vpif_obj.sd[i]) {
 				vpif_err("Error registering v4l2 subdevice\n");
-				err = -ENOMEM;
+				err = -ENODEV;
 				goto probe_subdev_out;
 			}
 			v4l2_info(&vpif_obj.v4l2_dev,
diff --git a/drivers/media/platform/exynos-gsc/gsc-core.h b/drivers/media/platform/exynos-gsc/gsc-core.h
index 76435d3bf62d..ef0a6564cef9 100644
--- a/drivers/media/platform/exynos-gsc/gsc-core.h
+++ b/drivers/media/platform/exynos-gsc/gsc-core.h
@@ -45,6 +45,7 @@
 #define GSC_DST_FMT			(1 << 2)
 #define GSC_CTX_M2M			(1 << 3)
 #define GSC_CTX_STOP_REQ		(1 << 6)
+#define	GSC_CTX_ABORT			(1 << 7)
 
 enum gsc_dev_flags {
 	/* for global */
diff --git a/drivers/media/platform/exynos-gsc/gsc-m2m.c b/drivers/media/platform/exynos-gsc/gsc-m2m.c
index e576ff2de3de..810c3e13970c 100644
--- a/drivers/media/platform/exynos-gsc/gsc-m2m.c
+++ b/drivers/media/platform/exynos-gsc/gsc-m2m.c
@@ -46,6 +46,17 @@ static int gsc_m2m_ctx_stop_req(struct gsc_ctx *ctx)
 	return ret == 0 ? -ETIMEDOUT : ret;
 }
 
+static void __gsc_m2m_job_abort(struct gsc_ctx *ctx)
+{
+	int ret;
+
+	ret = gsc_m2m_ctx_stop_req(ctx);
+	if ((ret == -ETIMEDOUT) || (ctx->state & GSC_CTX_ABORT)) {
+		gsc_ctx_state_lock_clear(GSC_CTX_STOP_REQ | GSC_CTX_ABORT, ctx);
+		gsc_m2m_job_finish(ctx, VB2_BUF_STATE_ERROR);
+	}
+}
+
 static int gsc_m2m_start_streaming(struct vb2_queue *q, unsigned int count)
 {
 	struct gsc_ctx *ctx = q->drv_priv;
@@ -58,11 +69,8 @@ static int gsc_m2m_start_streaming(struct vb2_queue *q, unsigned int count)
 static int gsc_m2m_stop_streaming(struct vb2_queue *q)
 {
 	struct gsc_ctx *ctx = q->drv_priv;
-	int ret;
 
-	ret = gsc_m2m_ctx_stop_req(ctx);
-	if (ret == -ETIMEDOUT)
-		gsc_m2m_job_finish(ctx, VB2_BUF_STATE_ERROR);
+	__gsc_m2m_job_abort(ctx);
 
 	pm_runtime_put(&ctx->gsc_dev->pdev->dev);
 
@@ -91,15 +99,9 @@ void gsc_m2m_job_finish(struct gsc_ctx *ctx, int vb_state)
 	}
 }
 
-
 static void gsc_m2m_job_abort(void *priv)
 {
-	struct gsc_ctx *ctx = priv;
-	int ret;
-
-	ret = gsc_m2m_ctx_stop_req(ctx);
-	if (ret == -ETIMEDOUT)
-		gsc_m2m_job_finish(ctx, VB2_BUF_STATE_ERROR);
+	__gsc_m2m_job_abort((struct gsc_ctx *)priv);
 }
 
 static int gsc_get_bufs(struct gsc_ctx *ctx)
@@ -150,9 +152,10 @@ static void gsc_m2m_device_run(void *priv)
 		gsc->m2m.ctx = ctx;
 	}
 
-	is_set = (ctx->state & GSC_CTX_STOP_REQ) ? 1 : 0;
-	ctx->state &= ~GSC_CTX_STOP_REQ;
+	is_set = ctx->state & GSC_CTX_STOP_REQ;
 	if (is_set) {
+		ctx->state &= ~GSC_CTX_STOP_REQ;
+		ctx->state |= GSC_CTX_ABORT;
 		wake_up(&gsc->irq_queue);
 		goto put_device;
 	}
diff --git a/drivers/media/platform/exynos4-is/fimc-isp.c b/drivers/media/platform/exynos4-is/fimc-isp.c
index d2e6cba3566d..f3c6136aa5b4 100644
--- a/drivers/media/platform/exynos4-is/fimc-isp.c
+++ b/drivers/media/platform/exynos4-is/fimc-isp.c
@@ -511,7 +511,7 @@ static int __ctrl_set_metering(struct fimc_is *is, unsigned int value)
 		break;
 	default:
 		return -EINVAL;
-	};
+	}
 
 	__is_set_isp_metering(is, IS_METERING_CONFIG_CMD, val);
 	return 0;
diff --git a/drivers/media/platform/exynos4-is/media-dev.c b/drivers/media/platform/exynos4-is/media-dev.c
index a83511278317..7a4ee4c0449d 100644
--- a/drivers/media/platform/exynos4-is/media-dev.c
+++ b/drivers/media/platform/exynos4-is/media-dev.c
@@ -411,8 +411,8 @@ static int fimc_md_of_add_sensor(struct fimc_md *fmd,
 
 	device_lock(&client->dev);
 
-	if (!client->driver ||
-	    !try_module_get(client->driver->driver.owner)) {
+	if (!client->dev.driver ||
+	    !try_module_get(client->dev.driver->owner)) {
 		ret = -EPROBE_DEFER;
 		v4l2_info(&fmd->v4l2_dev, "No driver found for %s\n",
 						node->full_name);
@@ -442,7 +442,7 @@ static int fimc_md_of_add_sensor(struct fimc_md *fmd,
 	fmd->num_sensors++;
 
 mod_put:
-	module_put(client->driver->driver.owner);
+	module_put(client->dev.driver->owner);
 dev_put:
 	device_unlock(&client->dev);
 	put_device(&client->dev);
diff --git a/drivers/media/platform/m2m-deinterlace.c b/drivers/media/platform/m2m-deinterlace.c
index 540516ca872c..65cab70fefcb 100644
--- a/drivers/media/platform/m2m-deinterlace.c
+++ b/drivers/media/platform/m2m-deinterlace.c
@@ -341,8 +341,7 @@ static void deinterlace_issue_dma(struct deinterlace_ctx *ctx, int op,
 	ctx->xt->dir = DMA_MEM_TO_MEM;
 	ctx->xt->src_sgl = false;
 	ctx->xt->dst_sgl = true;
-	flags = DMA_CTRL_ACK | DMA_PREP_INTERRUPT |
-		DMA_COMPL_SKIP_DEST_UNMAP | DMA_COMPL_SKIP_SRC_UNMAP;
+	flags = DMA_CTRL_ACK | DMA_PREP_INTERRUPT;
 
 	tx = dmadev->device_prep_interleaved_dma(chan, ctx->xt, flags);
 	if (tx == NULL) {
@@ -1084,8 +1083,7 @@ free_dev:
 
 static int deinterlace_remove(struct platform_device *pdev)
 {
-	struct deinterlace_dev *pcdev =
-		(struct deinterlace_dev *)platform_get_drvdata(pdev);
+	struct deinterlace_dev *pcdev = platform_get_drvdata(pdev);
 
 	v4l2_info(&pcdev->v4l2_dev, "Removing " MEM2MEM_TEST_MODULE_NAME);
 	v4l2_m2m_release(pcdev->m2m_dev);
diff --git a/drivers/media/platform/marvell-ccic/mcam-core.c b/drivers/media/platform/marvell-ccic/mcam-core.c
index 5184887b155c..32fab30a9105 100644
--- a/drivers/media/platform/marvell-ccic/mcam-core.c
+++ b/drivers/media/platform/marvell-ccic/mcam-core.c
@@ -1221,16 +1221,16 @@ static int mcam_vb_sg_buf_prepare(struct vb2_buffer *vb)
 {
 	struct mcam_vb_buffer *mvb = vb_to_mvb(vb);
 	struct mcam_camera *cam = vb2_get_drv_priv(vb->vb2_queue);
-	struct vb2_dma_sg_desc *sgd = vb2_dma_sg_plane_desc(vb, 0);
+	struct sg_table *sg_table = vb2_dma_sg_plane_desc(vb, 0);
 	struct mcam_dma_desc *desc = mvb->dma_desc;
 	struct scatterlist *sg;
 	int i;
 
-	mvb->dma_desc_nent = dma_map_sg(cam->dev, sgd->sglist, sgd->num_pages,
-			DMA_FROM_DEVICE);
+	mvb->dma_desc_nent = dma_map_sg(cam->dev, sg_table->sgl,
+			sg_table->nents, DMA_FROM_DEVICE);
 	if (mvb->dma_desc_nent <= 0)
 		return -EIO;  /* Not sure what's right here */
-	for_each_sg(sgd->sglist, sg, mvb->dma_desc_nent, i) {
+	for_each_sg(sg_table->sgl, sg, mvb->dma_desc_nent, i) {
 		desc->dma_addr = sg_dma_address(sg);
 		desc->segment_len = sg_dma_len(sg);
 		desc++;
@@ -1241,9 +1241,11 @@ static int mcam_vb_sg_buf_prepare(struct vb2_buffer *vb)
 static int mcam_vb_sg_buf_finish(struct vb2_buffer *vb)
 {
 	struct mcam_camera *cam = vb2_get_drv_priv(vb->vb2_queue);
-	struct vb2_dma_sg_desc *sgd = vb2_dma_sg_plane_desc(vb, 0);
+	struct sg_table *sg_table = vb2_dma_sg_plane_desc(vb, 0);
 
-	dma_unmap_sg(cam->dev, sgd->sglist, sgd->num_pages, DMA_FROM_DEVICE);
+	if (sg_table)
+		dma_unmap_sg(cam->dev, sg_table->sgl,
+				sg_table->nents, DMA_FROM_DEVICE);
 	return 0;
 }
 
diff --git a/drivers/media/platform/marvell-ccic/mmp-driver.c b/drivers/media/platform/marvell-ccic/mmp-driver.c
index b5a19af5c587..3458fa0e2fd5 100644
--- a/drivers/media/platform/marvell-ccic/mmp-driver.c
+++ b/drivers/media/platform/marvell-ccic/mmp-driver.c
@@ -481,7 +481,6 @@ static int mmpcam_remove(struct mmp_camera *cam)
 	struct mmp_camera_platform_data *pdata;
 
 	mmpcam_remove_device(cam);
-	free_irq(cam->irq, mcam);
 	mccic_shutdown(mcam);
 	mmpcam_power_down(mcam);
 	pdata = cam->pdev->dev.platform_data;
diff --git a/drivers/media/platform/mem2mem_testdev.c b/drivers/media/platform/mem2mem_testdev.c
index 6a17676f9d72..8df5975b700a 100644
--- a/drivers/media/platform/mem2mem_testdev.c
+++ b/drivers/media/platform/mem2mem_testdev.c
@@ -1090,8 +1090,7 @@ unreg_dev:
 
 static int m2mtest_remove(struct platform_device *pdev)
 {
-	struct m2mtest_dev *dev =
-		(struct m2mtest_dev *)platform_get_drvdata(pdev);
+	struct m2mtest_dev *dev = platform_get_drvdata(pdev);
 
 	v4l2_info(&dev->v4l2_dev, "Removing " MEM2MEM_TEST_MODULE_NAME);
 	v4l2_m2m_release(dev->m2m_dev);
diff --git a/drivers/media/platform/s5p-g2d/g2d.c b/drivers/media/platform/s5p-g2d/g2d.c
index fd6289d60cde..0b2948376aee 100644
--- a/drivers/media/platform/s5p-g2d/g2d.c
+++ b/drivers/media/platform/s5p-g2d/g2d.c
@@ -840,7 +840,7 @@ put_clk:
 
 static int g2d_remove(struct platform_device *pdev)
 {
-	struct g2d_dev *dev = (struct g2d_dev *)platform_get_drvdata(pdev);
+	struct g2d_dev *dev = platform_get_drvdata(pdev);
 
 	v4l2_info(&dev->v4l2_dev, "Removing " G2D_NAME);
 	v4l2_m2m_release(dev->m2m_dev);
diff --git a/drivers/media/platform/s5p-mfc/s5p_mfc.c b/drivers/media/platform/s5p-mfc/s5p_mfc.c
index 084263dd126f..5f2c4ad6c2cb 100644
--- a/drivers/media/platform/s5p-mfc/s5p_mfc.c
+++ b/drivers/media/platform/s5p-mfc/s5p_mfc.c
@@ -404,7 +404,11 @@ leave_handle_frame:
 	if (test_and_clear_bit(0, &dev->hw_lock) == 0)
 		BUG();
 	s5p_mfc_clock_off();
-	s5p_mfc_hw_call(dev->mfc_ops, try_run, dev);
+	/* if suspending, wake up device and do not try_run again*/
+	if (test_bit(0, &dev->enter_suspend))
+		wake_up_dev(dev, reason, err);
+	else
+		s5p_mfc_hw_call(dev->mfc_ops, try_run, dev);
 }
 
 /* Error handling for interrupt */
@@ -1101,7 +1105,7 @@ static int s5p_mfc_probe(struct platform_device *pdev)
 	}
 	dev->irq = res->start;
 	ret = devm_request_irq(&pdev->dev, dev->irq, s5p_mfc_irq,
-					IRQF_DISABLED, pdev->name, dev);
+					0, pdev->name, dev);
 	if (ret) {
 		dev_err(&pdev->dev, "Failed to install irq (%d)\n", ret);
 		goto err_res;
@@ -1286,9 +1290,7 @@ static int s5p_mfc_suspend(struct device *dev)
 		/* Try and lock the HW */
 		/* Wait on the interrupt waitqueue */
 		ret = wait_event_interruptible_timeout(m_dev->queue,
-			m_dev->int_cond || m_dev->ctx[m_dev->curr_ctx]->int_cond,
-			msecs_to_jiffies(MFC_INT_TIMEOUT));
-
+			m_dev->int_cond, msecs_to_jiffies(MFC_INT_TIMEOUT));
 		if (ret == 0) {
 			mfc_err("Waiting for hardware to finish timed out\n");
 			return -EIO;
diff --git a/drivers/media/platform/s5p-mfc/s5p_mfc_cmd_v5.c b/drivers/media/platform/s5p-mfc/s5p_mfc_cmd_v5.c
index ad4f1df0a18e..9a6efd6c1329 100644
--- a/drivers/media/platform/s5p-mfc/s5p_mfc_cmd_v5.c
+++ b/drivers/media/platform/s5p-mfc/s5p_mfc_cmd_v5.c
@@ -111,7 +111,7 @@ static int s5p_mfc_open_inst_cmd_v5(struct s5p_mfc_ctx *ctx)
 		break;
 	default:
 		h2r_args.arg[0] = S5P_FIMV_CODEC_NONE;
-	};
+	}
 	h2r_args.arg[1] = 0; /* no crc & no pixelcache */
 	h2r_args.arg[2] = ctx->ctx.ofs;
 	h2r_args.arg[3] = ctx->ctx.size;
diff --git a/drivers/media/platform/s5p-mfc/s5p_mfc_cmd_v6.c b/drivers/media/platform/s5p-mfc/s5p_mfc_cmd_v6.c
index db796c8e7874..ec1a5947ed7d 100644
--- a/drivers/media/platform/s5p-mfc/s5p_mfc_cmd_v6.c
+++ b/drivers/media/platform/s5p-mfc/s5p_mfc_cmd_v6.c
@@ -113,7 +113,7 @@ static int s5p_mfc_open_inst_cmd_v6(struct s5p_mfc_ctx *ctx)
 		break;
 	default:
 		codec_type = S5P_FIMV_CODEC_NONE_V6;
-	};
+	}
 	mfc_write(dev, codec_type, S5P_FIMV_CODEC_TYPE_V6);
 	mfc_write(dev, ctx->ctx.dma, S5P_FIMV_CONTEXT_MEM_ADDR_V6);
 	mfc_write(dev, ctx->ctx.size, S5P_FIMV_CONTEXT_MEM_SIZE_V6);
diff --git a/drivers/media/platform/s5p-mfc/s5p_mfc_enc.c b/drivers/media/platform/s5p-mfc/s5p_mfc_enc.c
index 41f5a3c10dbd..4ff3b6cd6842 100644
--- a/drivers/media/platform/s5p-mfc/s5p_mfc_enc.c
+++ b/drivers/media/platform/s5p-mfc/s5p_mfc_enc.c
@@ -113,7 +113,7 @@ static struct mfc_control controls[] = {
 		.minimum = 0,
 		.maximum = (1 << 16) - 1,
 		.step = 1,
-		.default_value = 0,
+		.default_value = 12,
 	},
 	{
 		.id = V4L2_CID_MPEG_VIDEO_MULTI_SLICE_MODE,
@@ -356,7 +356,7 @@ static struct mfc_control controls[] = {
 		.minimum = 0,
 		.maximum = 51,
 		.step = 1,
-		.default_value = 1,
+		.default_value = 51,
 	},
 	{
 		.id = V4L2_CID_MPEG_VIDEO_H264_P_FRAME_QP,
@@ -399,7 +399,7 @@ static struct mfc_control controls[] = {
 		.minimum = 1,
 		.maximum = 31,
 		.step = 1,
-		.default_value = 1,
+		.default_value = 31,
 	},
 	{
 		.id = V4L2_CID_MPEG_VIDEO_H263_P_FRAME_QP,
@@ -444,7 +444,7 @@ static struct mfc_control controls[] = {
 		.minimum = 0,
 		.maximum = 51,
 		.step = 1,
-		.default_value = 1,
+		.default_value = 51,
 	},
 	{
 		.id = V4L2_CID_MPEG_VIDEO_MPEG4_P_FRAME_QP,
diff --git a/drivers/media/platform/s5p-mfc/s5p_mfc_opr_v5.c b/drivers/media/platform/s5p-mfc/s5p_mfc_opr_v5.c
index 368582b091bf..58ec7bb26ebc 100644
--- a/drivers/media/platform/s5p-mfc/s5p_mfc_opr_v5.c
+++ b/drivers/media/platform/s5p-mfc/s5p_mfc_opr_v5.c
@@ -1582,7 +1582,7 @@ static int s5p_mfc_get_int_reason_v5(struct s5p_mfc_dev *dev)
 		break;
 	default:
 		reason = S5P_MFC_R2H_CMD_EMPTY;
-	};
+	}
 	return reason;
 }
 
diff --git a/drivers/media/platform/s5p-tv/mixer_grp_layer.c b/drivers/media/platform/s5p-tv/mixer_grp_layer.c
index b93a21f5aa13..74344c764daa 100644
--- a/drivers/media/platform/s5p-tv/mixer_grp_layer.c
+++ b/drivers/media/platform/s5p-tv/mixer_grp_layer.c
@@ -226,7 +226,7 @@ static void mxr_graph_fix_geometry(struct mxr_layer *layer,
 			src->width + src->x_offset, 32767);
 		src->full_height = clamp_val(src->full_height,
 			src->height + src->y_offset, 2047);
-	};
+	}
 }
 
 /* PUBLIC API */
diff --git a/drivers/media/platform/s5p-tv/mixer_vp_layer.c b/drivers/media/platform/s5p-tv/mixer_vp_layer.c
index 3d13a636877b..c9388c45ad75 100644
--- a/drivers/media/platform/s5p-tv/mixer_vp_layer.c
+++ b/drivers/media/platform/s5p-tv/mixer_vp_layer.c
@@ -197,7 +197,7 @@ static void mxr_vp_fix_geometry(struct mxr_layer *layer,
 			ALIGN(src->width + src->x_offset, 8), 8192U);
 		src->full_height = clamp(src->full_height,
 			src->height + src->y_offset, 8192U);
-	};
+	}
 }
 
 /* PUBLIC API */
diff --git a/drivers/media/platform/soc_camera/rcar_vin.c b/drivers/media/platform/soc_camera/rcar_vin.c
index d02a7e0b773f..6866bb4fbebc 100644
--- a/drivers/media/platform/soc_camera/rcar_vin.c
+++ b/drivers/media/platform/soc_camera/rcar_vin.c
@@ -16,6 +16,7 @@
 
 #include <linux/delay.h>
 #include <linux/interrupt.h>
+#include <linux/io.h>
 #include <linux/kernel.h>
 #include <linux/module.h>
 #include <linux/platform_data/camera-rcar.h>
@@ -105,6 +106,7 @@
 #define VIN_MAX_HEIGHT		2048
 
 enum chip_id {
+	RCAR_H2,
 	RCAR_H1,
 	RCAR_M1,
 	RCAR_E1,
@@ -300,7 +302,8 @@ static int rcar_vin_setup(struct rcar_vin_priv *priv)
 		dmr = 0;
 		break;
 	case V4L2_PIX_FMT_RGB32:
-		if (priv->chip == RCAR_H1 || priv->chip == RCAR_E1) {
+		if (priv->chip == RCAR_H2 || priv->chip == RCAR_H1 ||
+		    priv->chip == RCAR_E1) {
 			dmr = VNDMR_EXRGB;
 			break;
 		}
@@ -1381,6 +1384,7 @@ static struct soc_camera_host_ops rcar_vin_host_ops = {
 };
 
 static struct platform_device_id rcar_vin_id_table[] = {
+	{ "r8a7790-vin",  RCAR_H2 },
 	{ "r8a7779-vin",  RCAR_H1 },
 	{ "r8a7778-vin",  RCAR_M1 },
 	{ "uPD35004-vin", RCAR_E1 },
diff --git a/drivers/media/platform/soc_camera/sh_mobile_ceu_camera.c b/drivers/media/platform/soc_camera/sh_mobile_ceu_camera.c
index 8df22f779175..150bd4df413c 100644
--- a/drivers/media/platform/soc_camera/sh_mobile_ceu_camera.c
+++ b/drivers/media/platform/soc_camera/sh_mobile_ceu_camera.c
@@ -1800,7 +1800,7 @@ static int sh_mobile_ceu_probe(struct platform_device *pdev)
 
 	/* request irq */
 	err = devm_request_irq(&pdev->dev, pcdev->irq, sh_mobile_ceu_irq,
-			       IRQF_DISABLED, dev_name(&pdev->dev), pcdev);
+			       0, dev_name(&pdev->dev), pcdev);
 	if (err) {
 		dev_err(&pdev->dev, "Unable to register CEU interrupt.\n");
 		goto exit_release_mem;
diff --git a/drivers/media/platform/soc_camera/soc_camera.c b/drivers/media/platform/soc_camera/soc_camera.c
index 387a232d95a4..4b8c024fc487 100644
--- a/drivers/media/platform/soc_camera/soc_camera.c
+++ b/drivers/media/platform/soc_camera/soc_camera.c
@@ -71,13 +71,23 @@ static int video_dev_create(struct soc_camera_device *icd);
 int soc_camera_power_on(struct device *dev, struct soc_camera_subdev_desc *ssdd,
 			struct v4l2_clk *clk)
 {
-	int ret = clk ? v4l2_clk_enable(clk) : 0;
-	if (ret < 0) {
-		dev_err(dev, "Cannot enable clock: %d\n", ret);
-		return ret;
+	int ret;
+	bool clock_toggle;
+
+	if (clk && (!ssdd->unbalanced_power ||
+		    !test_and_set_bit(0, &ssdd->clock_state))) {
+		ret = v4l2_clk_enable(clk);
+		if (ret < 0) {
+			dev_err(dev, "Cannot enable clock: %d\n", ret);
+			return ret;
+		}
+		clock_toggle = true;
+	} else {
+		clock_toggle = false;
 	}
-	ret = regulator_bulk_enable(ssdd->num_regulators,
-					ssdd->regulators);
+
+	ret = regulator_bulk_enable(ssdd->sd_pdata.num_regulators,
+				    ssdd->sd_pdata.regulators);
 	if (ret < 0) {
 		dev_err(dev, "Cannot enable regulators\n");
 		goto eregenable;
@@ -95,10 +105,10 @@ int soc_camera_power_on(struct device *dev, struct soc_camera_subdev_desc *ssdd,
 	return 0;
 
 epwron:
-	regulator_bulk_disable(ssdd->num_regulators,
-			       ssdd->regulators);
+	regulator_bulk_disable(ssdd->sd_pdata.num_regulators,
+			       ssdd->sd_pdata.regulators);
 eregenable:
-	if (clk)
+	if (clock_toggle)
 		v4l2_clk_disable(clk);
 
 	return ret;
@@ -120,14 +130,14 @@ int soc_camera_power_off(struct device *dev, struct soc_camera_subdev_desc *ssdd
 		}
 	}
 
-	err = regulator_bulk_disable(ssdd->num_regulators,
-				     ssdd->regulators);
+	err = regulator_bulk_disable(ssdd->sd_pdata.num_regulators,
+				     ssdd->sd_pdata.regulators);
 	if (err < 0) {
 		dev_err(dev, "Cannot disable regulators\n");
 		ret = ret ? : err;
 	}
 
-	if (clk)
+	if (clk && (!ssdd->unbalanced_power || test_and_clear_bit(0, &ssdd->clock_state)))
 		v4l2_clk_disable(clk);
 
 	return ret;
@@ -137,8 +147,8 @@ EXPORT_SYMBOL(soc_camera_power_off);
 int soc_camera_power_init(struct device *dev, struct soc_camera_subdev_desc *ssdd)
 {
 	/* Should not have any effect in synchronous case */
-	return devm_regulator_bulk_get(dev, ssdd->num_regulators,
-				       ssdd->regulators);
+	return devm_regulator_bulk_get(dev, ssdd->sd_pdata.num_regulators,
+				       ssdd->sd_pdata.regulators);
 }
 EXPORT_SYMBOL(soc_camera_power_init);
 
@@ -1346,8 +1356,8 @@ static int soc_camera_i2c_init(struct soc_camera_device *icd,
 	 * soc_camera_pdrv_probe(), make sure the subdevice driver doesn't try
 	 * to allocate them again.
 	 */
-	ssdd->num_regulators = 0;
-	ssdd->regulators = NULL;
+	ssdd->sd_pdata.num_regulators = 0;
+	ssdd->sd_pdata.regulators = NULL;
 	shd->board_info->platform_data = ssdd;
 
 	snprintf(clk_name, sizeof(clk_name), "%d-%04x",
@@ -2020,8 +2030,8 @@ static int soc_camera_pdrv_probe(struct platform_device *pdev)
 	 * that case regulators are attached to the I2C device and not to the
 	 * camera platform device.
 	 */
-	ret = devm_regulator_bulk_get(&pdev->dev, ssdd->num_regulators,
-				      ssdd->regulators);
+	ret = devm_regulator_bulk_get(&pdev->dev, ssdd->sd_pdata.num_regulators,
+				      ssdd->sd_pdata.regulators);
 	if (ret < 0)
 		return ret;
 
diff --git a/drivers/media/platform/ti-vpe/Makefile b/drivers/media/platform/ti-vpe/Makefile
new file mode 100644
index 000000000000..cbf0a806ba1d
--- /dev/null
+++ b/drivers/media/platform/ti-vpe/Makefile
@@ -0,0 +1,5 @@
+obj-$(CONFIG_VIDEO_TI_VPE) += ti-vpe.o
+
+ti-vpe-y := vpe.o vpdma.o
+
+ccflags-$(CONFIG_VIDEO_TI_VPE_DEBUG) += -DDEBUG
diff --git a/drivers/media/platform/ti-vpe/vpdma.c b/drivers/media/platform/ti-vpe/vpdma.c
new file mode 100644
index 000000000000..af0a5ffcaa98
--- /dev/null
+++ b/drivers/media/platform/ti-vpe/vpdma.c
@@ -0,0 +1,846 @@
+/*
+ * VPDMA helper library
+ *
+ * Copyright (c) 2013 Texas Instruments Inc.
+ *
+ * David Griego, <dagriego@biglakesoftware.com>
+ * Dale Farnsworth, <dale@farnsworth.org>
+ * Archit Taneja, <archit@ti.com>
+ *
+ * This program is free software; you can redistribute it and/or modify it
+ * under the terms of the GNU General Public License version 2 as published by
+ * the Free Software Foundation.
+ */
+
+#include <linux/delay.h>
+#include <linux/dma-mapping.h>
+#include <linux/err.h>
+#include <linux/firmware.h>
+#include <linux/io.h>
+#include <linux/module.h>
+#include <linux/platform_device.h>
+#include <linux/sched.h>
+#include <linux/slab.h>
+#include <linux/videodev2.h>
+
+#include "vpdma.h"
+#include "vpdma_priv.h"
+
+#define VPDMA_FIRMWARE	"vpdma-1b8.bin"
+
+const struct vpdma_data_format vpdma_yuv_fmts[] = {
+	[VPDMA_DATA_FMT_Y444] = {
+		.data_type	= DATA_TYPE_Y444,
+		.depth		= 8,
+	},
+	[VPDMA_DATA_FMT_Y422] = {
+		.data_type	= DATA_TYPE_Y422,
+		.depth		= 8,
+	},
+	[VPDMA_DATA_FMT_Y420] = {
+		.data_type	= DATA_TYPE_Y420,
+		.depth		= 8,
+	},
+	[VPDMA_DATA_FMT_C444] = {
+		.data_type	= DATA_TYPE_C444,
+		.depth		= 8,
+	},
+	[VPDMA_DATA_FMT_C422] = {
+		.data_type	= DATA_TYPE_C422,
+		.depth		= 8,
+	},
+	[VPDMA_DATA_FMT_C420] = {
+		.data_type	= DATA_TYPE_C420,
+		.depth		= 4,
+	},
+	[VPDMA_DATA_FMT_YC422] = {
+		.data_type	= DATA_TYPE_YC422,
+		.depth		= 16,
+	},
+	[VPDMA_DATA_FMT_YC444] = {
+		.data_type	= DATA_TYPE_YC444,
+		.depth		= 24,
+	},
+	[VPDMA_DATA_FMT_CY422] = {
+		.data_type	= DATA_TYPE_CY422,
+		.depth		= 16,
+	},
+};
+
+const struct vpdma_data_format vpdma_rgb_fmts[] = {
+	[VPDMA_DATA_FMT_RGB565] = {
+		.data_type	= DATA_TYPE_RGB16_565,
+		.depth		= 16,
+	},
+	[VPDMA_DATA_FMT_ARGB16_1555] = {
+		.data_type	= DATA_TYPE_ARGB_1555,
+		.depth		= 16,
+	},
+	[VPDMA_DATA_FMT_ARGB16] = {
+		.data_type	= DATA_TYPE_ARGB_4444,
+		.depth		= 16,
+	},
+	[VPDMA_DATA_FMT_RGBA16_5551] = {
+		.data_type	= DATA_TYPE_RGBA_5551,
+		.depth		= 16,
+	},
+	[VPDMA_DATA_FMT_RGBA16] = {
+		.data_type	= DATA_TYPE_RGBA_4444,
+		.depth		= 16,
+	},
+	[VPDMA_DATA_FMT_ARGB24] = {
+		.data_type	= DATA_TYPE_ARGB24_6666,
+		.depth		= 24,
+	},
+	[VPDMA_DATA_FMT_RGB24] = {
+		.data_type	= DATA_TYPE_RGB24_888,
+		.depth		= 24,
+	},
+	[VPDMA_DATA_FMT_ARGB32] = {
+		.data_type	= DATA_TYPE_ARGB32_8888,
+		.depth		= 32,
+	},
+	[VPDMA_DATA_FMT_RGBA24] = {
+		.data_type	= DATA_TYPE_RGBA24_6666,
+		.depth		= 24,
+	},
+	[VPDMA_DATA_FMT_RGBA32] = {
+		.data_type	= DATA_TYPE_RGBA32_8888,
+		.depth		= 32,
+	},
+	[VPDMA_DATA_FMT_BGR565] = {
+		.data_type	= DATA_TYPE_BGR16_565,
+		.depth		= 16,
+	},
+	[VPDMA_DATA_FMT_ABGR16_1555] = {
+		.data_type	= DATA_TYPE_ABGR_1555,
+		.depth		= 16,
+	},
+	[VPDMA_DATA_FMT_ABGR16] = {
+		.data_type	= DATA_TYPE_ABGR_4444,
+		.depth		= 16,
+	},
+	[VPDMA_DATA_FMT_BGRA16_5551] = {
+		.data_type	= DATA_TYPE_BGRA_5551,
+		.depth		= 16,
+	},
+	[VPDMA_DATA_FMT_BGRA16] = {
+		.data_type	= DATA_TYPE_BGRA_4444,
+		.depth		= 16,
+	},
+	[VPDMA_DATA_FMT_ABGR24] = {
+		.data_type	= DATA_TYPE_ABGR24_6666,
+		.depth		= 24,
+	},
+	[VPDMA_DATA_FMT_BGR24] = {
+		.data_type	= DATA_TYPE_BGR24_888,
+		.depth		= 24,
+	},
+	[VPDMA_DATA_FMT_ABGR32] = {
+		.data_type	= DATA_TYPE_ABGR32_8888,
+		.depth		= 32,
+	},
+	[VPDMA_DATA_FMT_BGRA24] = {
+		.data_type	= DATA_TYPE_BGRA24_6666,
+		.depth		= 24,
+	},
+	[VPDMA_DATA_FMT_BGRA32] = {
+		.data_type	= DATA_TYPE_BGRA32_8888,
+		.depth		= 32,
+	},
+};
+
+const struct vpdma_data_format vpdma_misc_fmts[] = {
+	[VPDMA_DATA_FMT_MV] = {
+		.data_type	= DATA_TYPE_MV,
+		.depth		= 4,
+	},
+};
+
+struct vpdma_channel_info {
+	int num;		/* VPDMA channel number */
+	int cstat_offset;	/* client CSTAT register offset */
+};
+
+static const struct vpdma_channel_info chan_info[] = {
+	[VPE_CHAN_LUMA1_IN] = {
+		.num		= VPE_CHAN_NUM_LUMA1_IN,
+		.cstat_offset	= VPDMA_DEI_LUMA1_CSTAT,
+	},
+	[VPE_CHAN_CHROMA1_IN] = {
+		.num		= VPE_CHAN_NUM_CHROMA1_IN,
+		.cstat_offset	= VPDMA_DEI_CHROMA1_CSTAT,
+	},
+	[VPE_CHAN_LUMA2_IN] = {
+		.num		= VPE_CHAN_NUM_LUMA2_IN,
+		.cstat_offset	= VPDMA_DEI_LUMA2_CSTAT,
+	},
+	[VPE_CHAN_CHROMA2_IN] = {
+		.num		= VPE_CHAN_NUM_CHROMA2_IN,
+		.cstat_offset	= VPDMA_DEI_CHROMA2_CSTAT,
+	},
+	[VPE_CHAN_LUMA3_IN] = {
+		.num		= VPE_CHAN_NUM_LUMA3_IN,
+		.cstat_offset	= VPDMA_DEI_LUMA3_CSTAT,
+	},
+	[VPE_CHAN_CHROMA3_IN] = {
+		.num		= VPE_CHAN_NUM_CHROMA3_IN,
+		.cstat_offset	= VPDMA_DEI_CHROMA3_CSTAT,
+	},
+	[VPE_CHAN_MV_IN] = {
+		.num		= VPE_CHAN_NUM_MV_IN,
+		.cstat_offset	= VPDMA_DEI_MV_IN_CSTAT,
+	},
+	[VPE_CHAN_MV_OUT] = {
+		.num		= VPE_CHAN_NUM_MV_OUT,
+		.cstat_offset	= VPDMA_DEI_MV_OUT_CSTAT,
+	},
+	[VPE_CHAN_LUMA_OUT] = {
+		.num		= VPE_CHAN_NUM_LUMA_OUT,
+		.cstat_offset	= VPDMA_VIP_UP_Y_CSTAT,
+	},
+	[VPE_CHAN_CHROMA_OUT] = {
+		.num		= VPE_CHAN_NUM_CHROMA_OUT,
+		.cstat_offset	= VPDMA_VIP_UP_UV_CSTAT,
+	},
+	[VPE_CHAN_RGB_OUT] = {
+		.num		= VPE_CHAN_NUM_RGB_OUT,
+		.cstat_offset	= VPDMA_VIP_UP_Y_CSTAT,
+	},
+};
+
+static u32 read_reg(struct vpdma_data *vpdma, int offset)
+{
+	return ioread32(vpdma->base + offset);
+}
+
+static void write_reg(struct vpdma_data *vpdma, int offset, u32 value)
+{
+	iowrite32(value, vpdma->base + offset);
+}
+
+static int read_field_reg(struct vpdma_data *vpdma, int offset,
+		u32 mask, int shift)
+{
+	return (read_reg(vpdma, offset) & (mask << shift)) >> shift;
+}
+
+static void write_field_reg(struct vpdma_data *vpdma, int offset, u32 field,
+		u32 mask, int shift)
+{
+	u32 val = read_reg(vpdma, offset);
+
+	val &= ~(mask << shift);
+	val |= (field & mask) << shift;
+
+	write_reg(vpdma, offset, val);
+}
+
+void vpdma_dump_regs(struct vpdma_data *vpdma)
+{
+	struct device *dev = &vpdma->pdev->dev;
+
+#define DUMPREG(r) dev_dbg(dev, "%-35s %08x\n", #r, read_reg(vpdma, VPDMA_##r))
+
+	dev_dbg(dev, "VPDMA Registers:\n");
+
+	DUMPREG(PID);
+	DUMPREG(LIST_ADDR);
+	DUMPREG(LIST_ATTR);
+	DUMPREG(LIST_STAT_SYNC);
+	DUMPREG(BG_RGB);
+	DUMPREG(BG_YUV);
+	DUMPREG(SETUP);
+	DUMPREG(MAX_SIZE1);
+	DUMPREG(MAX_SIZE2);
+	DUMPREG(MAX_SIZE3);
+
+	/*
+	 * dumping registers of only group0 and group3, because VPE channels
+	 * lie within group0 and group3 registers
+	 */
+	DUMPREG(INT_CHAN_STAT(0));
+	DUMPREG(INT_CHAN_MASK(0));
+	DUMPREG(INT_CHAN_STAT(3));
+	DUMPREG(INT_CHAN_MASK(3));
+	DUMPREG(INT_CLIENT0_STAT);
+	DUMPREG(INT_CLIENT0_MASK);
+	DUMPREG(INT_CLIENT1_STAT);
+	DUMPREG(INT_CLIENT1_MASK);
+	DUMPREG(INT_LIST0_STAT);
+	DUMPREG(INT_LIST0_MASK);
+
+	/*
+	 * these are registers specific to VPE clients, we can make this
+	 * function dump client registers specific to VPE or VIP based on
+	 * who is using it
+	 */
+	DUMPREG(DEI_CHROMA1_CSTAT);
+	DUMPREG(DEI_LUMA1_CSTAT);
+	DUMPREG(DEI_CHROMA2_CSTAT);
+	DUMPREG(DEI_LUMA2_CSTAT);
+	DUMPREG(DEI_CHROMA3_CSTAT);
+	DUMPREG(DEI_LUMA3_CSTAT);
+	DUMPREG(DEI_MV_IN_CSTAT);
+	DUMPREG(DEI_MV_OUT_CSTAT);
+	DUMPREG(VIP_UP_Y_CSTAT);
+	DUMPREG(VIP_UP_UV_CSTAT);
+	DUMPREG(VPI_CTL_CSTAT);
+}
+
+/*
+ * Allocate a DMA buffer
+ */
+int vpdma_alloc_desc_buf(struct vpdma_buf *buf, size_t size)
+{
+	buf->size = size;
+	buf->mapped = false;
+	buf->addr = kzalloc(size, GFP_KERNEL);
+	if (!buf->addr)
+		return -ENOMEM;
+
+	WARN_ON((u32) buf->addr & VPDMA_DESC_ALIGN);
+
+	return 0;
+}
+
+void vpdma_free_desc_buf(struct vpdma_buf *buf)
+{
+	WARN_ON(buf->mapped);
+	kfree(buf->addr);
+	buf->addr = NULL;
+	buf->size = 0;
+}
+
+/*
+ * map descriptor/payload DMA buffer, enabling DMA access
+ */
+int vpdma_map_desc_buf(struct vpdma_data *vpdma, struct vpdma_buf *buf)
+{
+	struct device *dev = &vpdma->pdev->dev;
+
+	WARN_ON(buf->mapped);
+	buf->dma_addr = dma_map_single(dev, buf->addr, buf->size,
+				DMA_TO_DEVICE);
+	if (dma_mapping_error(dev, buf->dma_addr)) {
+		dev_err(dev, "failed to map buffer\n");
+		return -EINVAL;
+	}
+
+	buf->mapped = true;
+
+	return 0;
+}
+
+/*
+ * unmap descriptor/payload DMA buffer, disabling DMA access and
+ * allowing the main processor to acces the data
+ */
+void vpdma_unmap_desc_buf(struct vpdma_data *vpdma, struct vpdma_buf *buf)
+{
+	struct device *dev = &vpdma->pdev->dev;
+
+	if (buf->mapped)
+		dma_unmap_single(dev, buf->dma_addr, buf->size, DMA_TO_DEVICE);
+
+	buf->mapped = false;
+}
+
+/*
+ * create a descriptor list, the user of this list will append configuration,
+ * control and data descriptors to this list, this list will be submitted to
+ * VPDMA. VPDMA's list parser will go through each descriptor and perform the
+ * required DMA operations
+ */
+int vpdma_create_desc_list(struct vpdma_desc_list *list, size_t size, int type)
+{
+	int r;
+
+	r = vpdma_alloc_desc_buf(&list->buf, size);
+	if (r)
+		return r;
+
+	list->next = list->buf.addr;
+
+	list->type = type;
+
+	return 0;
+}
+
+/*
+ * once a descriptor list is parsed by VPDMA, we reset the list by emptying it,
+ * to allow new descriptors to be added to the list.
+ */
+void vpdma_reset_desc_list(struct vpdma_desc_list *list)
+{
+	list->next = list->buf.addr;
+}
+
+/*
+ * free the buffer allocated fot the VPDMA descriptor list, this should be
+ * called when the user doesn't want to use VPDMA any more.
+ */
+void vpdma_free_desc_list(struct vpdma_desc_list *list)
+{
+	vpdma_free_desc_buf(&list->buf);
+
+	list->next = NULL;
+}
+
+static bool vpdma_list_busy(struct vpdma_data *vpdma, int list_num)
+{
+	return read_reg(vpdma, VPDMA_LIST_STAT_SYNC) & BIT(list_num + 16);
+}
+
+/*
+ * submit a list of DMA descriptors to the VPE VPDMA, do not wait for completion
+ */
+int vpdma_submit_descs(struct vpdma_data *vpdma, struct vpdma_desc_list *list)
+{
+	/* we always use the first list */
+	int list_num = 0;
+	int list_size;
+
+	if (vpdma_list_busy(vpdma, list_num))
+		return -EBUSY;
+
+	/* 16-byte granularity */
+	list_size = (list->next - list->buf.addr) >> 4;
+
+	write_reg(vpdma, VPDMA_LIST_ADDR, (u32) list->buf.dma_addr);
+
+	write_reg(vpdma, VPDMA_LIST_ATTR,
+			(list_num << VPDMA_LIST_NUM_SHFT) |
+			(list->type << VPDMA_LIST_TYPE_SHFT) |
+			list_size);
+
+	return 0;
+}
+
+static void dump_cfd(struct vpdma_cfd *cfd)
+{
+	int class;
+
+	class = cfd_get_class(cfd);
+
+	pr_debug("config descriptor of payload class: %s\n",
+		class == CFD_CLS_BLOCK ? "simple block" :
+		"address data block");
+
+	if (class == CFD_CLS_BLOCK)
+		pr_debug("word0: dst_addr_offset = 0x%08x\n",
+			cfd->dest_addr_offset);
+
+	if (class == CFD_CLS_BLOCK)
+		pr_debug("word1: num_data_wrds = %d\n", cfd->block_len);
+
+	pr_debug("word2: payload_addr = 0x%08x\n", cfd->payload_addr);
+
+	pr_debug("word3: pkt_type = %d, direct = %d, class = %d, dest = %d, "
+		"payload_len = %d\n", cfd_get_pkt_type(cfd),
+		cfd_get_direct(cfd), class, cfd_get_dest(cfd),
+		cfd_get_payload_len(cfd));
+}
+
+/*
+ * append a configuration descriptor to the given descriptor list, where the
+ * payload is in the form of a simple data block specified in the descriptor
+ * header, this is used to upload scaler coefficients to the scaler module
+ */
+void vpdma_add_cfd_block(struct vpdma_desc_list *list, int client,
+		struct vpdma_buf *blk, u32 dest_offset)
+{
+	struct vpdma_cfd *cfd;
+	int len = blk->size;
+
+	WARN_ON(blk->dma_addr & VPDMA_DESC_ALIGN);
+
+	cfd = list->next;
+	WARN_ON((void *)(cfd + 1) > (list->buf.addr + list->buf.size));
+
+	cfd->dest_addr_offset = dest_offset;
+	cfd->block_len = len;
+	cfd->payload_addr = (u32) blk->dma_addr;
+	cfd->ctl_payload_len = cfd_pkt_payload_len(CFD_INDIRECT, CFD_CLS_BLOCK,
+				client, len >> 4);
+
+	list->next = cfd + 1;
+
+	dump_cfd(cfd);
+}
+
+/*
+ * append a configuration descriptor to the given descriptor list, where the
+ * payload is in the address data block format, this is used to a configure a
+ * discontiguous set of MMRs
+ */
+void vpdma_add_cfd_adb(struct vpdma_desc_list *list, int client,
+		struct vpdma_buf *adb)
+{
+	struct vpdma_cfd *cfd;
+	unsigned int len = adb->size;
+
+	WARN_ON(len & VPDMA_ADB_SIZE_ALIGN);
+	WARN_ON(adb->dma_addr & VPDMA_DESC_ALIGN);
+
+	cfd = list->next;
+	BUG_ON((void *)(cfd + 1) > (list->buf.addr + list->buf.size));
+
+	cfd->w0 = 0;
+	cfd->w1 = 0;
+	cfd->payload_addr = (u32) adb->dma_addr;
+	cfd->ctl_payload_len = cfd_pkt_payload_len(CFD_INDIRECT, CFD_CLS_ADB,
+				client, len >> 4);
+
+	list->next = cfd + 1;
+
+	dump_cfd(cfd);
+};
+
+/*
+ * control descriptor format change based on what type of control descriptor it
+ * is, we only use 'sync on channel' control descriptors for now, so assume it's
+ * that
+ */
+static void dump_ctd(struct vpdma_ctd *ctd)
+{
+	pr_debug("control descriptor\n");
+
+	pr_debug("word3: pkt_type = %d, source = %d, ctl_type = %d\n",
+		ctd_get_pkt_type(ctd), ctd_get_source(ctd), ctd_get_ctl(ctd));
+}
+
+/*
+ * append a 'sync on channel' type control descriptor to the given descriptor
+ * list, this descriptor stalls the VPDMA list till the time DMA is completed
+ * on the specified channel
+ */
+void vpdma_add_sync_on_channel_ctd(struct vpdma_desc_list *list,
+		enum vpdma_channel chan)
+{
+	struct vpdma_ctd *ctd;
+
+	ctd = list->next;
+	WARN_ON((void *)(ctd + 1) > (list->buf.addr + list->buf.size));
+
+	ctd->w0 = 0;
+	ctd->w1 = 0;
+	ctd->w2 = 0;
+	ctd->type_source_ctl = ctd_type_source_ctl(chan_info[chan].num,
+				CTD_TYPE_SYNC_ON_CHANNEL);
+
+	list->next = ctd + 1;
+
+	dump_ctd(ctd);
+}
+
+static void dump_dtd(struct vpdma_dtd *dtd)
+{
+	int dir, chan;
+
+	dir = dtd_get_dir(dtd);
+	chan = dtd_get_chan(dtd);
+
+	pr_debug("%s data transfer descriptor for channel %d\n",
+		dir == DTD_DIR_OUT ? "outbound" : "inbound", chan);
+
+	pr_debug("word0: data_type = %d, notify = %d, field = %d, 1D = %d, "
+		"even_ln_skp = %d, odd_ln_skp = %d, line_stride = %d\n",
+		dtd_get_data_type(dtd), dtd_get_notify(dtd), dtd_get_field(dtd),
+		dtd_get_1d(dtd), dtd_get_even_line_skip(dtd),
+		dtd_get_odd_line_skip(dtd), dtd_get_line_stride(dtd));
+
+	if (dir == DTD_DIR_IN)
+		pr_debug("word1: line_length = %d, xfer_height = %d\n",
+			dtd_get_line_length(dtd), dtd_get_xfer_height(dtd));
+
+	pr_debug("word2: start_addr = 0x%08x\n", dtd->start_addr);
+
+	pr_debug("word3: pkt_type = %d, mode = %d, dir = %d, chan = %d, "
+		"pri = %d, next_chan = %d\n", dtd_get_pkt_type(dtd),
+		dtd_get_mode(dtd), dir, chan, dtd_get_priority(dtd),
+		dtd_get_next_chan(dtd));
+
+	if (dir == DTD_DIR_IN)
+		pr_debug("word4: frame_width = %d, frame_height = %d\n",
+			dtd_get_frame_width(dtd), dtd_get_frame_height(dtd));
+	else
+		pr_debug("word4: desc_write_addr = 0x%08x, write_desc = %d, "
+			"drp_data = %d, use_desc_reg = %d\n",
+			dtd_get_desc_write_addr(dtd), dtd_get_write_desc(dtd),
+			dtd_get_drop_data(dtd), dtd_get_use_desc(dtd));
+
+	if (dir == DTD_DIR_IN)
+		pr_debug("word5: hor_start = %d, ver_start = %d\n",
+			dtd_get_h_start(dtd), dtd_get_v_start(dtd));
+	else
+		pr_debug("word5: max_width %d, max_height %d\n",
+			dtd_get_max_width(dtd), dtd_get_max_height(dtd));
+
+	pr_debug("word6: client specfic attr0 = 0x%08x\n", dtd->client_attr0);
+	pr_debug("word7: client specfic attr1 = 0x%08x\n", dtd->client_attr1);
+}
+
+/*
+ * append an outbound data transfer descriptor to the given descriptor list,
+ * this sets up a 'client to memory' VPDMA transfer for the given VPDMA channel
+ */
+void vpdma_add_out_dtd(struct vpdma_desc_list *list, struct v4l2_rect *c_rect,
+		const struct vpdma_data_format *fmt, dma_addr_t dma_addr,
+		enum vpdma_channel chan, u32 flags)
+{
+	int priority = 0;
+	int field = 0;
+	int notify = 1;
+	int channel, next_chan;
+	int depth = fmt->depth;
+	int stride;
+	struct vpdma_dtd *dtd;
+
+	channel = next_chan = chan_info[chan].num;
+
+	if (fmt->data_type == DATA_TYPE_C420)
+		depth = 8;
+
+	stride = (depth * c_rect->width) >> 3;
+	dma_addr += (c_rect->left * depth) >> 3;
+
+	dtd = list->next;
+	WARN_ON((void *)(dtd + 1) > (list->buf.addr + list->buf.size));
+
+	dtd->type_ctl_stride = dtd_type_ctl_stride(fmt->data_type,
+					notify,
+					field,
+					!!(flags & VPDMA_DATA_FRAME_1D),
+					!!(flags & VPDMA_DATA_EVEN_LINE_SKIP),
+					!!(flags & VPDMA_DATA_ODD_LINE_SKIP),
+					stride);
+	dtd->w1 = 0;
+	dtd->start_addr = (u32) dma_addr;
+	dtd->pkt_ctl = dtd_pkt_ctl(!!(flags & VPDMA_DATA_MODE_TILED),
+				DTD_DIR_OUT, channel, priority, next_chan);
+	dtd->desc_write_addr = dtd_desc_write_addr(0, 0, 0, 0);
+	dtd->max_width_height = dtd_max_width_height(MAX_OUT_WIDTH_1920,
+					MAX_OUT_HEIGHT_1080);
+	dtd->client_attr0 = 0;
+	dtd->client_attr1 = 0;
+
+	list->next = dtd + 1;
+
+	dump_dtd(dtd);
+}
+
+/*
+ * append an inbound data transfer descriptor to the given descriptor list,
+ * this sets up a 'memory to client' VPDMA transfer for the given VPDMA channel
+ */
+void vpdma_add_in_dtd(struct vpdma_desc_list *list, int frame_width,
+		int frame_height, struct v4l2_rect *c_rect,
+		const struct vpdma_data_format *fmt, dma_addr_t dma_addr,
+		enum vpdma_channel chan, int field, u32 flags)
+{
+	int priority = 0;
+	int notify = 1;
+	int depth = fmt->depth;
+	int channel, next_chan;
+	int stride;
+	int height = c_rect->height;
+	struct vpdma_dtd *dtd;
+
+	channel = next_chan = chan_info[chan].num;
+
+	if (fmt->data_type == DATA_TYPE_C420) {
+		height >>= 1;
+		frame_height >>= 1;
+		depth = 8;
+	}
+
+	stride = (depth * c_rect->width) >> 3;
+	dma_addr += (c_rect->left * depth) >> 3;
+
+	dtd = list->next;
+	WARN_ON((void *)(dtd + 1) > (list->buf.addr + list->buf.size));
+
+	dtd->type_ctl_stride = dtd_type_ctl_stride(fmt->data_type,
+					notify,
+					field,
+					!!(flags & VPDMA_DATA_FRAME_1D),
+					!!(flags & VPDMA_DATA_EVEN_LINE_SKIP),
+					!!(flags & VPDMA_DATA_ODD_LINE_SKIP),
+					stride);
+
+	dtd->xfer_length_height = dtd_xfer_length_height(c_rect->width, height);
+	dtd->start_addr = (u32) dma_addr;
+	dtd->pkt_ctl = dtd_pkt_ctl(!!(flags & VPDMA_DATA_MODE_TILED),
+				DTD_DIR_IN, channel, priority, next_chan);
+	dtd->frame_width_height = dtd_frame_width_height(frame_width,
+					frame_height);
+	dtd->start_h_v = dtd_start_h_v(c_rect->left, c_rect->top);
+	dtd->client_attr0 = 0;
+	dtd->client_attr1 = 0;
+
+	list->next = dtd + 1;
+
+	dump_dtd(dtd);
+}
+
+/* set or clear the mask for list complete interrupt */
+void vpdma_enable_list_complete_irq(struct vpdma_data *vpdma, int list_num,
+		bool enable)
+{
+	u32 val;
+
+	val = read_reg(vpdma, VPDMA_INT_LIST0_MASK);
+	if (enable)
+		val |= (1 << (list_num * 2));
+	else
+		val &= ~(1 << (list_num * 2));
+	write_reg(vpdma, VPDMA_INT_LIST0_MASK, val);
+}
+
+/* clear previosuly occured list intterupts in the LIST_STAT register */
+void vpdma_clear_list_stat(struct vpdma_data *vpdma)
+{
+	write_reg(vpdma, VPDMA_INT_LIST0_STAT,
+		read_reg(vpdma, VPDMA_INT_LIST0_STAT));
+}
+
+/*
+ * configures the output mode of the line buffer for the given client, the
+ * line buffer content can either be mirrored(each line repeated twice) or
+ * passed to the client as is
+ */
+void vpdma_set_line_mode(struct vpdma_data *vpdma, int line_mode,
+		enum vpdma_channel chan)
+{
+	int client_cstat = chan_info[chan].cstat_offset;
+
+	write_field_reg(vpdma, client_cstat, line_mode,
+		VPDMA_CSTAT_LINE_MODE_MASK, VPDMA_CSTAT_LINE_MODE_SHIFT);
+}
+
+/*
+ * configures the event which should trigger VPDMA transfer for the given
+ * client
+ */
+void vpdma_set_frame_start_event(struct vpdma_data *vpdma,
+		enum vpdma_frame_start_event fs_event,
+		enum vpdma_channel chan)
+{
+	int client_cstat = chan_info[chan].cstat_offset;
+
+	write_field_reg(vpdma, client_cstat, fs_event,
+		VPDMA_CSTAT_FRAME_START_MASK, VPDMA_CSTAT_FRAME_START_SHIFT);
+}
+
+static void vpdma_firmware_cb(const struct firmware *f, void *context)
+{
+	struct vpdma_data *vpdma = context;
+	struct vpdma_buf fw_dma_buf;
+	int i, r;
+
+	dev_dbg(&vpdma->pdev->dev, "firmware callback\n");
+
+	if (!f || !f->data) {
+		dev_err(&vpdma->pdev->dev, "couldn't get firmware\n");
+		return;
+	}
+
+	/* already initialized */
+	if (read_field_reg(vpdma, VPDMA_LIST_ATTR, VPDMA_LIST_RDY_MASK,
+			VPDMA_LIST_RDY_SHFT)) {
+		vpdma->ready = true;
+		return;
+	}
+
+	r = vpdma_alloc_desc_buf(&fw_dma_buf, f->size);
+	if (r) {
+		dev_err(&vpdma->pdev->dev,
+			"failed to allocate dma buffer for firmware\n");
+		goto rel_fw;
+	}
+
+	memcpy(fw_dma_buf.addr, f->data, f->size);
+
+	vpdma_map_desc_buf(vpdma, &fw_dma_buf);
+
+	write_reg(vpdma, VPDMA_LIST_ADDR, (u32) fw_dma_buf.dma_addr);
+
+	for (i = 0; i < 100; i++) {		/* max 1 second */
+		msleep_interruptible(10);
+
+		if (read_field_reg(vpdma, VPDMA_LIST_ATTR, VPDMA_LIST_RDY_MASK,
+				VPDMA_LIST_RDY_SHFT))
+			break;
+	}
+
+	if (i == 100) {
+		dev_err(&vpdma->pdev->dev, "firmware upload failed\n");
+		goto free_buf;
+	}
+
+	vpdma->ready = true;
+
+free_buf:
+	vpdma_unmap_desc_buf(vpdma, &fw_dma_buf);
+
+	vpdma_free_desc_buf(&fw_dma_buf);
+rel_fw:
+	release_firmware(f);
+}
+
+static int vpdma_load_firmware(struct vpdma_data *vpdma)
+{
+	int r;
+	struct device *dev = &vpdma->pdev->dev;
+
+	r = request_firmware_nowait(THIS_MODULE, 1,
+		(const char *) VPDMA_FIRMWARE, dev, GFP_KERNEL, vpdma,
+		vpdma_firmware_cb);
+	if (r) {
+		dev_err(dev, "firmware not available %s\n", VPDMA_FIRMWARE);
+		return r;
+	} else {
+		dev_info(dev, "loading firmware %s\n", VPDMA_FIRMWARE);
+	}
+
+	return 0;
+}
+
+struct vpdma_data *vpdma_create(struct platform_device *pdev)
+{
+	struct resource *res;
+	struct vpdma_data *vpdma;
+	int r;
+
+	dev_dbg(&pdev->dev, "vpdma_create\n");
+
+	vpdma = devm_kzalloc(&pdev->dev, sizeof(*vpdma), GFP_KERNEL);
+	if (!vpdma) {
+		dev_err(&pdev->dev, "couldn't alloc vpdma_dev\n");
+		return ERR_PTR(-ENOMEM);
+	}
+
+	vpdma->pdev = pdev;
+
+	res = platform_get_resource_byname(pdev, IORESOURCE_MEM, "vpdma");
+	if (res == NULL) {
+		dev_err(&pdev->dev, "missing platform resources data\n");
+		return ERR_PTR(-ENODEV);
+	}
+
+	vpdma->base = devm_ioremap(&pdev->dev, res->start, resource_size(res));
+	if (!vpdma->base) {
+		dev_err(&pdev->dev, "failed to ioremap\n");
+		return ERR_PTR(-ENOMEM);
+	}
+
+	r = vpdma_load_firmware(vpdma);
+	if (r) {
+		pr_err("failed to load firmware %s\n", VPDMA_FIRMWARE);
+		return ERR_PTR(r);
+	}
+
+	return vpdma;
+}
+MODULE_FIRMWARE(VPDMA_FIRMWARE);
diff --git a/drivers/media/platform/ti-vpe/vpdma.h b/drivers/media/platform/ti-vpe/vpdma.h
new file mode 100644
index 000000000000..eaa2a71a5db9
--- /dev/null
+++ b/drivers/media/platform/ti-vpe/vpdma.h
@@ -0,0 +1,203 @@
+/*
+ * Copyright (c) 2013 Texas Instruments Inc.
+ *
+ * David Griego, <dagriego@biglakesoftware.com>
+ * Dale Farnsworth, <dale@farnsworth.org>
+ * Archit Taneja, <archit@ti.com>
+ *
+ * This program is free software; you can redistribute it and/or modify it
+ * under the terms of the GNU General Public License version 2 as published by
+ * the Free Software Foundation.
+ */
+
+#ifndef __TI_VPDMA_H_
+#define __TI_VPDMA_H_
+
+/*
+ * A vpdma_buf tracks the size, DMA address and mapping status of each
+ * driver DMA area.
+ */
+struct vpdma_buf {
+	void			*addr;
+	dma_addr_t		dma_addr;
+	size_t			size;
+	bool			mapped;
+};
+
+struct vpdma_desc_list {
+	struct vpdma_buf buf;
+	void *next;
+	int type;
+};
+
+struct vpdma_data {
+	void __iomem		*base;
+
+	struct platform_device	*pdev;
+
+	/* tells whether vpdma firmware is loaded or not */
+	bool ready;
+};
+
+struct vpdma_data_format {
+	int data_type;
+	u8 depth;
+};
+
+#define VPDMA_DESC_ALIGN		16	/* 16-byte descriptor alignment */
+
+#define VPDMA_DTD_DESC_SIZE		32	/* 8 words */
+#define VPDMA_CFD_CTD_DESC_SIZE		16	/* 4 words */
+
+#define VPDMA_LIST_TYPE_NORMAL		0
+#define VPDMA_LIST_TYPE_SELF_MODIFYING	1
+#define VPDMA_LIST_TYPE_DOORBELL	2
+
+enum vpdma_yuv_formats {
+	VPDMA_DATA_FMT_Y444 = 0,
+	VPDMA_DATA_FMT_Y422,
+	VPDMA_DATA_FMT_Y420,
+	VPDMA_DATA_FMT_C444,
+	VPDMA_DATA_FMT_C422,
+	VPDMA_DATA_FMT_C420,
+	VPDMA_DATA_FMT_YC422,
+	VPDMA_DATA_FMT_YC444,
+	VPDMA_DATA_FMT_CY422,
+};
+
+enum vpdma_rgb_formats {
+	VPDMA_DATA_FMT_RGB565 = 0,
+	VPDMA_DATA_FMT_ARGB16_1555,
+	VPDMA_DATA_FMT_ARGB16,
+	VPDMA_DATA_FMT_RGBA16_5551,
+	VPDMA_DATA_FMT_RGBA16,
+	VPDMA_DATA_FMT_ARGB24,
+	VPDMA_DATA_FMT_RGB24,
+	VPDMA_DATA_FMT_ARGB32,
+	VPDMA_DATA_FMT_RGBA24,
+	VPDMA_DATA_FMT_RGBA32,
+	VPDMA_DATA_FMT_BGR565,
+	VPDMA_DATA_FMT_ABGR16_1555,
+	VPDMA_DATA_FMT_ABGR16,
+	VPDMA_DATA_FMT_BGRA16_5551,
+	VPDMA_DATA_FMT_BGRA16,
+	VPDMA_DATA_FMT_ABGR24,
+	VPDMA_DATA_FMT_BGR24,
+	VPDMA_DATA_FMT_ABGR32,
+	VPDMA_DATA_FMT_BGRA24,
+	VPDMA_DATA_FMT_BGRA32,
+};
+
+enum vpdma_misc_formats {
+	VPDMA_DATA_FMT_MV = 0,
+};
+
+extern const struct vpdma_data_format vpdma_yuv_fmts[];
+extern const struct vpdma_data_format vpdma_rgb_fmts[];
+extern const struct vpdma_data_format vpdma_misc_fmts[];
+
+enum vpdma_frame_start_event {
+	VPDMA_FSEVENT_HDMI_FID = 0,
+	VPDMA_FSEVENT_DVO2_FID,
+	VPDMA_FSEVENT_HDCOMP_FID,
+	VPDMA_FSEVENT_SD_FID,
+	VPDMA_FSEVENT_LM_FID0,
+	VPDMA_FSEVENT_LM_FID1,
+	VPDMA_FSEVENT_LM_FID2,
+	VPDMA_FSEVENT_CHANNEL_ACTIVE,
+};
+
+/*
+ * VPDMA channel numbers
+ */
+enum vpdma_channel {
+	VPE_CHAN_LUMA1_IN,
+	VPE_CHAN_CHROMA1_IN,
+	VPE_CHAN_LUMA2_IN,
+	VPE_CHAN_CHROMA2_IN,
+	VPE_CHAN_LUMA3_IN,
+	VPE_CHAN_CHROMA3_IN,
+	VPE_CHAN_MV_IN,
+	VPE_CHAN_MV_OUT,
+	VPE_CHAN_LUMA_OUT,
+	VPE_CHAN_CHROMA_OUT,
+	VPE_CHAN_RGB_OUT,
+};
+
+/* flags for VPDMA data descriptors */
+#define VPDMA_DATA_ODD_LINE_SKIP	(1 << 0)
+#define VPDMA_DATA_EVEN_LINE_SKIP	(1 << 1)
+#define VPDMA_DATA_FRAME_1D		(1 << 2)
+#define VPDMA_DATA_MODE_TILED		(1 << 3)
+
+/*
+ * client identifiers used for configuration descriptors
+ */
+#define CFD_MMR_CLIENT		0
+#define CFD_SC_CLIENT		4
+
+/* Address data block header format */
+struct vpdma_adb_hdr {
+	u32			offset;
+	u32			nwords;
+	u32			reserved0;
+	u32			reserved1;
+};
+
+/* helpers for creating ADB headers for config descriptors MMRs as client */
+#define ADB_ADDR(dma_buf, str, fld)	((dma_buf)->addr + offsetof(str, fld))
+#define MMR_ADB_ADDR(buf, str, fld)	ADB_ADDR(&(buf), struct str, fld)
+
+#define VPDMA_SET_MMR_ADB_HDR(buf, str, hdr, regs, offset_a)	\
+	do {							\
+		struct vpdma_adb_hdr *h;			\
+		struct str *adb = NULL;				\
+		h = MMR_ADB_ADDR(buf, str, hdr);		\
+		h->offset = (offset_a);				\
+		h->nwords = sizeof(adb->regs) >> 2;		\
+	} while (0)
+
+/* vpdma descriptor buffer allocation and management */
+int vpdma_alloc_desc_buf(struct vpdma_buf *buf, size_t size);
+void vpdma_free_desc_buf(struct vpdma_buf *buf);
+int vpdma_map_desc_buf(struct vpdma_data *vpdma, struct vpdma_buf *buf);
+void vpdma_unmap_desc_buf(struct vpdma_data *vpdma, struct vpdma_buf *buf);
+
+/* vpdma descriptor list funcs */
+int vpdma_create_desc_list(struct vpdma_desc_list *list, size_t size, int type);
+void vpdma_reset_desc_list(struct vpdma_desc_list *list);
+void vpdma_free_desc_list(struct vpdma_desc_list *list);
+int vpdma_submit_descs(struct vpdma_data *vpdma, struct vpdma_desc_list *list);
+
+/* helpers for creating vpdma descriptors */
+void vpdma_add_cfd_block(struct vpdma_desc_list *list, int client,
+		struct vpdma_buf *blk, u32 dest_offset);
+void vpdma_add_cfd_adb(struct vpdma_desc_list *list, int client,
+		struct vpdma_buf *adb);
+void vpdma_add_sync_on_channel_ctd(struct vpdma_desc_list *list,
+		enum vpdma_channel chan);
+void vpdma_add_out_dtd(struct vpdma_desc_list *list, struct v4l2_rect *c_rect,
+		const struct vpdma_data_format *fmt, dma_addr_t dma_addr,
+		enum vpdma_channel chan, u32 flags);
+void vpdma_add_in_dtd(struct vpdma_desc_list *list, int frame_width,
+		int frame_height, struct v4l2_rect *c_rect,
+		const struct vpdma_data_format *fmt, dma_addr_t dma_addr,
+		enum vpdma_channel chan, int field, u32 flags);
+
+/* vpdma list interrupt management */
+void vpdma_enable_list_complete_irq(struct vpdma_data *vpdma, int list_num,
+		bool enable);
+void vpdma_clear_list_stat(struct vpdma_data *vpdma);
+
+/* vpdma client configuration */
+void vpdma_set_line_mode(struct vpdma_data *vpdma, int line_mode,
+		enum vpdma_channel chan);
+void vpdma_set_frame_start_event(struct vpdma_data *vpdma,
+		enum vpdma_frame_start_event fs_event, enum vpdma_channel chan);
+
+void vpdma_dump_regs(struct vpdma_data *vpdma);
+
+/* initialize vpdma, passed with VPE's platform device pointer */
+struct vpdma_data *vpdma_create(struct platform_device *pdev);
+
+#endif
diff --git a/drivers/media/platform/ti-vpe/vpdma_priv.h b/drivers/media/platform/ti-vpe/vpdma_priv.h
new file mode 100644
index 000000000000..f0e9a8038c1b
--- /dev/null
+++ b/drivers/media/platform/ti-vpe/vpdma_priv.h
@@ -0,0 +1,641 @@
+/*
+ * Copyright (c) 2013 Texas Instruments Inc.
+ *
+ * David Griego, <dagriego@biglakesoftware.com>
+ * Dale Farnsworth, <dale@farnsworth.org>
+ * Archit Taneja, <archit@ti.com>
+ *
+ * This program is free software; you can redistribute it and/or modify it
+ * under the terms of the GNU General Public License version 2 as published by
+ * the Free Software Foundation.
+ */
+
+#ifndef _TI_VPDMA_PRIV_H_
+#define _TI_VPDMA_PRIV_H_
+
+/*
+ * VPDMA Register offsets
+ */
+
+/* Top level */
+#define VPDMA_PID		0x00
+#define VPDMA_LIST_ADDR		0x04
+#define VPDMA_LIST_ATTR		0x08
+#define VPDMA_LIST_STAT_SYNC	0x0c
+#define VPDMA_BG_RGB		0x18
+#define VPDMA_BG_YUV		0x1c
+#define VPDMA_SETUP		0x30
+#define VPDMA_MAX_SIZE1		0x34
+#define VPDMA_MAX_SIZE2		0x38
+#define VPDMA_MAX_SIZE3		0x3c
+
+/* Interrupts */
+#define VPDMA_INT_CHAN_STAT(grp)	(0x40 + grp * 8)
+#define VPDMA_INT_CHAN_MASK(grp)	(VPDMA_INT_CHAN_STAT(grp) + 4)
+#define VPDMA_INT_CLIENT0_STAT		0x78
+#define VPDMA_INT_CLIENT0_MASK		0x7c
+#define VPDMA_INT_CLIENT1_STAT		0x80
+#define VPDMA_INT_CLIENT1_MASK		0x84
+#define VPDMA_INT_LIST0_STAT		0x88
+#define VPDMA_INT_LIST0_MASK		0x8c
+
+#define VPDMA_PERFMON(i)		(0x200 + i * 4)
+
+/* VPE specific client registers */
+#define VPDMA_DEI_CHROMA1_CSTAT		0x0300
+#define VPDMA_DEI_LUMA1_CSTAT		0x0304
+#define VPDMA_DEI_LUMA2_CSTAT		0x0308
+#define VPDMA_DEI_CHROMA2_CSTAT		0x030c
+#define VPDMA_DEI_LUMA3_CSTAT		0x0310
+#define VPDMA_DEI_CHROMA3_CSTAT		0x0314
+#define VPDMA_DEI_MV_IN_CSTAT		0x0330
+#define VPDMA_DEI_MV_OUT_CSTAT		0x033c
+#define VPDMA_VIP_UP_Y_CSTAT		0x0390
+#define VPDMA_VIP_UP_UV_CSTAT		0x0394
+#define VPDMA_VPI_CTL_CSTAT		0x03d0
+
+/* Reg field info for VPDMA_CLIENT_CSTAT registers */
+#define VPDMA_CSTAT_LINE_MODE_MASK	0x03
+#define VPDMA_CSTAT_LINE_MODE_SHIFT	8
+#define VPDMA_CSTAT_FRAME_START_MASK	0xf
+#define VPDMA_CSTAT_FRAME_START_SHIFT	10
+
+#define VPDMA_LIST_NUM_MASK		0x07
+#define VPDMA_LIST_NUM_SHFT		24
+#define VPDMA_LIST_STOP_SHFT		20
+#define VPDMA_LIST_RDY_MASK		0x01
+#define VPDMA_LIST_RDY_SHFT		19
+#define VPDMA_LIST_TYPE_MASK		0x03
+#define VPDMA_LIST_TYPE_SHFT		16
+#define VPDMA_LIST_SIZE_MASK		0xffff
+
+/* VPDMA data type values for data formats */
+#define DATA_TYPE_Y444				0x0
+#define DATA_TYPE_Y422				0x1
+#define DATA_TYPE_Y420				0x2
+#define DATA_TYPE_C444				0x4
+#define DATA_TYPE_C422				0x5
+#define DATA_TYPE_C420				0x6
+#define DATA_TYPE_YC422				0x7
+#define DATA_TYPE_YC444				0x8
+#define DATA_TYPE_CY422				0x23
+
+#define DATA_TYPE_RGB16_565			0x0
+#define DATA_TYPE_ARGB_1555			0x1
+#define DATA_TYPE_ARGB_4444			0x2
+#define DATA_TYPE_RGBA_5551			0x3
+#define DATA_TYPE_RGBA_4444			0x4
+#define DATA_TYPE_ARGB24_6666			0x5
+#define DATA_TYPE_RGB24_888			0x6
+#define DATA_TYPE_ARGB32_8888			0x7
+#define DATA_TYPE_RGBA24_6666			0x8
+#define DATA_TYPE_RGBA32_8888			0x9
+#define DATA_TYPE_BGR16_565			0x10
+#define DATA_TYPE_ABGR_1555			0x11
+#define DATA_TYPE_ABGR_4444			0x12
+#define DATA_TYPE_BGRA_5551			0x13
+#define DATA_TYPE_BGRA_4444			0x14
+#define DATA_TYPE_ABGR24_6666			0x15
+#define DATA_TYPE_BGR24_888			0x16
+#define DATA_TYPE_ABGR32_8888			0x17
+#define DATA_TYPE_BGRA24_6666			0x18
+#define DATA_TYPE_BGRA32_8888			0x19
+
+#define DATA_TYPE_MV				0x3
+
+/* VPDMA channel numbers(only VPE channels for now) */
+#define	VPE_CHAN_NUM_LUMA1_IN		0
+#define	VPE_CHAN_NUM_CHROMA1_IN		1
+#define	VPE_CHAN_NUM_LUMA2_IN		2
+#define	VPE_CHAN_NUM_CHROMA2_IN		3
+#define	VPE_CHAN_NUM_LUMA3_IN		4
+#define	VPE_CHAN_NUM_CHROMA3_IN		5
+#define	VPE_CHAN_NUM_MV_IN		12
+#define	VPE_CHAN_NUM_MV_OUT		15
+#define	VPE_CHAN_NUM_LUMA_OUT		102
+#define	VPE_CHAN_NUM_CHROMA_OUT		103
+#define	VPE_CHAN_NUM_RGB_OUT		106
+
+/*
+ * a VPDMA address data block payload for a configuration descriptor needs to
+ * have each sub block length as a multiple of 16 bytes. Therefore, the overall
+ * size of the payload also needs to be a multiple of 16 bytes. The sub block
+ * lengths should be ensured to be aligned by the VPDMA user.
+ */
+#define VPDMA_ADB_SIZE_ALIGN		0x0f
+
+/*
+ * data transfer descriptor
+ */
+struct vpdma_dtd {
+	u32			type_ctl_stride;
+	union {
+		u32		xfer_length_height;
+		u32		w1;
+	};
+	dma_addr_t		start_addr;
+	u32			pkt_ctl;
+	union {
+		u32		frame_width_height;	/* inbound */
+		dma_addr_t	desc_write_addr;	/* outbound */
+	};
+	union {
+		u32		start_h_v;		/* inbound */
+		u32		max_width_height;	/* outbound */
+	};
+	u32			client_attr0;
+	u32			client_attr1;
+};
+
+/* Data Transfer Descriptor specifics */
+#define DTD_NO_NOTIFY		0
+#define DTD_NOTIFY		1
+
+#define DTD_PKT_TYPE		0xa
+#define DTD_DIR_IN		0
+#define DTD_DIR_OUT		1
+
+/* type_ctl_stride */
+#define DTD_DATA_TYPE_MASK	0x3f
+#define DTD_DATA_TYPE_SHFT	26
+#define DTD_NOTIFY_MASK		0x01
+#define DTD_NOTIFY_SHFT		25
+#define DTD_FIELD_MASK		0x01
+#define DTD_FIELD_SHFT		24
+#define DTD_1D_MASK		0x01
+#define DTD_1D_SHFT		23
+#define DTD_EVEN_LINE_SKIP_MASK	0x01
+#define DTD_EVEN_LINE_SKIP_SHFT	20
+#define DTD_ODD_LINE_SKIP_MASK	0x01
+#define DTD_ODD_LINE_SKIP_SHFT	16
+#define DTD_LINE_STRIDE_MASK	0xffff
+#define DTD_LINE_STRIDE_SHFT	0
+
+/* xfer_length_height */
+#define DTD_LINE_LENGTH_MASK	0xffff
+#define DTD_LINE_LENGTH_SHFT	16
+#define DTD_XFER_HEIGHT_MASK	0xffff
+#define DTD_XFER_HEIGHT_SHFT	0
+
+/* pkt_ctl */
+#define DTD_PKT_TYPE_MASK	0x1f
+#define DTD_PKT_TYPE_SHFT	27
+#define DTD_MODE_MASK		0x01
+#define DTD_MODE_SHFT		26
+#define DTD_DIR_MASK		0x01
+#define DTD_DIR_SHFT		25
+#define DTD_CHAN_MASK		0x01ff
+#define DTD_CHAN_SHFT		16
+#define DTD_PRI_MASK		0x0f
+#define DTD_PRI_SHFT		9
+#define DTD_NEXT_CHAN_MASK	0x01ff
+#define DTD_NEXT_CHAN_SHFT	0
+
+/* frame_width_height */
+#define DTD_FRAME_WIDTH_MASK	0xffff
+#define DTD_FRAME_WIDTH_SHFT	16
+#define DTD_FRAME_HEIGHT_MASK	0xffff
+#define DTD_FRAME_HEIGHT_SHFT	0
+
+/* start_h_v */
+#define DTD_H_START_MASK	0xffff
+#define DTD_H_START_SHFT	16
+#define DTD_V_START_MASK	0xffff
+#define DTD_V_START_SHFT	0
+
+#define DTD_DESC_START_SHIFT	5
+#define DTD_WRITE_DESC_MASK	0x01
+#define DTD_WRITE_DESC_SHIFT	2
+#define DTD_DROP_DATA_MASK	0x01
+#define DTD_DROP_DATA_SHIFT	1
+#define DTD_USE_DESC_MASK	0x01
+#define DTD_USE_DESC_SHIFT	0
+
+/* max_width_height */
+#define DTD_MAX_WIDTH_MASK	0x07
+#define DTD_MAX_WIDTH_SHFT	4
+#define DTD_MAX_HEIGHT_MASK	0x07
+#define DTD_MAX_HEIGHT_SHFT	0
+
+/* max width configurations */
+ /* unlimited width */
+#define	MAX_OUT_WIDTH_UNLIMITED		0
+/* as specified in max_size1 reg */
+#define MAX_OUT_WIDTH_REG1		1
+/* as specified in max_size2 reg */
+#define MAX_OUT_WIDTH_REG2		2
+/* as specified in max_size3 reg */
+#define	MAX_OUT_WIDTH_REG3		3
+/* maximum of 352 pixels as width */
+#define MAX_OUT_WIDTH_352		4
+/* maximum of 768 pixels as width */
+#define	MAX_OUT_WIDTH_768		5
+/* maximum of 1280 pixels width */
+#define	MAX_OUT_WIDTH_1280		6
+/* maximum of 1920 pixels as width */
+#define	MAX_OUT_WIDTH_1920		7
+
+/* max height configurations */
+ /* unlimited height */
+#define	MAX_OUT_HEIGHT_UNLIMITED	0
+/* as specified in max_size1 reg */
+#define MAX_OUT_HEIGHT_REG1		1
+/* as specified in max_size2 reg */
+#define MAX_OUT_HEIGHT_REG2		2
+/* as specified in max_size3 reg */
+#define	MAX_OUT_HEIGHT_REG3		3
+/* maximum of 288 lines as height */
+#define MAX_OUT_HEIGHT_288		4
+/* maximum of 576 lines as height */
+#define	MAX_OUT_HEIGHT_576		5
+/* maximum of 720 lines as height */
+#define	MAX_OUT_HEIGHT_720		6
+/* maximum of 1080 lines as height */
+#define	MAX_OUT_HEIGHT_1080		7
+
+static inline u32 dtd_type_ctl_stride(int type, bool notify, int field,
+			bool one_d, bool even_line_skip, bool odd_line_skip,
+			int line_stride)
+{
+	return (type << DTD_DATA_TYPE_SHFT) | (notify << DTD_NOTIFY_SHFT) |
+		(field << DTD_FIELD_SHFT) | (one_d << DTD_1D_SHFT) |
+		(even_line_skip << DTD_EVEN_LINE_SKIP_SHFT) |
+		(odd_line_skip << DTD_ODD_LINE_SKIP_SHFT) |
+		line_stride;
+}
+
+static inline u32 dtd_xfer_length_height(int line_length, int xfer_height)
+{
+	return (line_length << DTD_LINE_LENGTH_SHFT) | xfer_height;
+}
+
+static inline u32 dtd_pkt_ctl(bool mode, bool dir, int chan, int pri,
+			int next_chan)
+{
+	return (DTD_PKT_TYPE << DTD_PKT_TYPE_SHFT) | (mode << DTD_MODE_SHFT) |
+		(dir << DTD_DIR_SHFT) | (chan << DTD_CHAN_SHFT) |
+		(pri << DTD_PRI_SHFT) | next_chan;
+}
+
+static inline u32 dtd_frame_width_height(int width, int height)
+{
+	return (width << DTD_FRAME_WIDTH_SHFT) | height;
+}
+
+static inline u32 dtd_desc_write_addr(unsigned int addr, bool write_desc,
+			bool drop_data, bool use_desc)
+{
+	return (addr << DTD_DESC_START_SHIFT) |
+		(write_desc << DTD_WRITE_DESC_SHIFT) |
+		(drop_data << DTD_DROP_DATA_SHIFT) |
+		use_desc;
+}
+
+static inline u32 dtd_start_h_v(int h_start, int v_start)
+{
+	return (h_start << DTD_H_START_SHFT) | v_start;
+}
+
+static inline u32 dtd_max_width_height(int max_width, int max_height)
+{
+	return (max_width << DTD_MAX_WIDTH_SHFT) | max_height;
+}
+
+static inline int dtd_get_data_type(struct vpdma_dtd *dtd)
+{
+	return dtd->type_ctl_stride >> DTD_DATA_TYPE_SHFT;
+}
+
+static inline bool dtd_get_notify(struct vpdma_dtd *dtd)
+{
+	return (dtd->type_ctl_stride >> DTD_NOTIFY_SHFT) & DTD_NOTIFY_MASK;
+}
+
+static inline int dtd_get_field(struct vpdma_dtd *dtd)
+{
+	return (dtd->type_ctl_stride >> DTD_FIELD_SHFT) & DTD_FIELD_MASK;
+}
+
+static inline bool dtd_get_1d(struct vpdma_dtd *dtd)
+{
+	return (dtd->type_ctl_stride >> DTD_1D_SHFT) & DTD_1D_MASK;
+}
+
+static inline bool dtd_get_even_line_skip(struct vpdma_dtd *dtd)
+{
+	return (dtd->type_ctl_stride >> DTD_EVEN_LINE_SKIP_SHFT)
+		& DTD_EVEN_LINE_SKIP_MASK;
+}
+
+static inline bool dtd_get_odd_line_skip(struct vpdma_dtd *dtd)
+{
+	return (dtd->type_ctl_stride >> DTD_ODD_LINE_SKIP_SHFT)
+		& DTD_ODD_LINE_SKIP_MASK;
+}
+
+static inline int dtd_get_line_stride(struct vpdma_dtd *dtd)
+{
+	return dtd->type_ctl_stride & DTD_LINE_STRIDE_MASK;
+}
+
+static inline int dtd_get_line_length(struct vpdma_dtd *dtd)
+{
+	return dtd->xfer_length_height >> DTD_LINE_LENGTH_SHFT;
+}
+
+static inline int dtd_get_xfer_height(struct vpdma_dtd *dtd)
+{
+	return dtd->xfer_length_height & DTD_XFER_HEIGHT_MASK;
+}
+
+static inline int dtd_get_pkt_type(struct vpdma_dtd *dtd)
+{
+	return dtd->pkt_ctl >> DTD_PKT_TYPE_SHFT;
+}
+
+static inline bool dtd_get_mode(struct vpdma_dtd *dtd)
+{
+	return (dtd->pkt_ctl >> DTD_MODE_SHFT) & DTD_MODE_MASK;
+}
+
+static inline bool dtd_get_dir(struct vpdma_dtd *dtd)
+{
+	return (dtd->pkt_ctl >> DTD_DIR_SHFT) & DTD_DIR_MASK;
+}
+
+static inline int dtd_get_chan(struct vpdma_dtd *dtd)
+{
+	return (dtd->pkt_ctl >> DTD_CHAN_SHFT) & DTD_CHAN_MASK;
+}
+
+static inline int dtd_get_priority(struct vpdma_dtd *dtd)
+{
+	return (dtd->pkt_ctl >> DTD_PRI_SHFT) & DTD_PRI_MASK;
+}
+
+static inline int dtd_get_next_chan(struct vpdma_dtd *dtd)
+{
+	return (dtd->pkt_ctl >> DTD_NEXT_CHAN_SHFT) & DTD_NEXT_CHAN_MASK;
+}
+
+static inline int dtd_get_frame_width(struct vpdma_dtd *dtd)
+{
+	return dtd->frame_width_height >> DTD_FRAME_WIDTH_SHFT;
+}
+
+static inline int dtd_get_frame_height(struct vpdma_dtd *dtd)
+{
+	return dtd->frame_width_height & DTD_FRAME_HEIGHT_MASK;
+}
+
+static inline int dtd_get_desc_write_addr(struct vpdma_dtd *dtd)
+{
+	return dtd->desc_write_addr >> DTD_DESC_START_SHIFT;
+}
+
+static inline bool dtd_get_write_desc(struct vpdma_dtd *dtd)
+{
+	return (dtd->desc_write_addr >> DTD_WRITE_DESC_SHIFT) &
+							DTD_WRITE_DESC_MASK;
+}
+
+static inline bool dtd_get_drop_data(struct vpdma_dtd *dtd)
+{
+	return (dtd->desc_write_addr >> DTD_DROP_DATA_SHIFT) &
+							DTD_DROP_DATA_MASK;
+}
+
+static inline bool dtd_get_use_desc(struct vpdma_dtd *dtd)
+{
+	return dtd->desc_write_addr & DTD_USE_DESC_MASK;
+}
+
+static inline int dtd_get_h_start(struct vpdma_dtd *dtd)
+{
+	return dtd->start_h_v >> DTD_H_START_SHFT;
+}
+
+static inline int dtd_get_v_start(struct vpdma_dtd *dtd)
+{
+	return dtd->start_h_v & DTD_V_START_MASK;
+}
+
+static inline int dtd_get_max_width(struct vpdma_dtd *dtd)
+{
+	return (dtd->max_width_height >> DTD_MAX_WIDTH_SHFT) &
+							DTD_MAX_WIDTH_MASK;
+}
+
+static inline int dtd_get_max_height(struct vpdma_dtd *dtd)
+{
+	return (dtd->max_width_height >> DTD_MAX_HEIGHT_SHFT) &
+							DTD_MAX_HEIGHT_MASK;
+}
+
+/*
+ * configuration descriptor
+ */
+struct vpdma_cfd {
+	union {
+		u32	dest_addr_offset;
+		u32	w0;
+	};
+	union {
+		u32	block_len;		/* in words */
+		u32	w1;
+	};
+	u32		payload_addr;
+	u32		ctl_payload_len;	/* in words */
+};
+
+/* Configuration descriptor specifics */
+
+#define CFD_PKT_TYPE		0xb
+
+#define CFD_DIRECT		1
+#define CFD_INDIRECT		0
+#define CFD_CLS_ADB		0
+#define CFD_CLS_BLOCK		1
+
+/* block_len */
+#define CFD__BLOCK_LEN_MASK	0xffff
+#define CFD__BLOCK_LEN_SHFT	0
+
+/* ctl_payload_len */
+#define CFD_PKT_TYPE_MASK	0x1f
+#define CFD_PKT_TYPE_SHFT	27
+#define CFD_DIRECT_MASK		0x01
+#define CFD_DIRECT_SHFT		26
+#define CFD_CLASS_MASK		0x03
+#define CFD_CLASS_SHFT		24
+#define CFD_DEST_MASK		0xff
+#define CFD_DEST_SHFT		16
+#define CFD_PAYLOAD_LEN_MASK	0xffff
+#define CFD_PAYLOAD_LEN_SHFT	0
+
+static inline u32 cfd_pkt_payload_len(bool direct, int cls, int dest,
+		int payload_len)
+{
+	return (CFD_PKT_TYPE << CFD_PKT_TYPE_SHFT) |
+		(direct << CFD_DIRECT_SHFT) |
+		(cls << CFD_CLASS_SHFT) |
+		(dest << CFD_DEST_SHFT) |
+		payload_len;
+}
+
+static inline int cfd_get_pkt_type(struct vpdma_cfd *cfd)
+{
+	return cfd->ctl_payload_len >> CFD_PKT_TYPE_SHFT;
+}
+
+static inline bool cfd_get_direct(struct vpdma_cfd *cfd)
+{
+	return (cfd->ctl_payload_len >> CFD_DIRECT_SHFT) & CFD_DIRECT_MASK;
+}
+
+static inline bool cfd_get_class(struct vpdma_cfd *cfd)
+{
+	return (cfd->ctl_payload_len >> CFD_CLASS_SHFT) & CFD_CLASS_MASK;
+}
+
+static inline int cfd_get_dest(struct vpdma_cfd *cfd)
+{
+	return (cfd->ctl_payload_len >> CFD_DEST_SHFT) & CFD_DEST_MASK;
+}
+
+static inline int cfd_get_payload_len(struct vpdma_cfd *cfd)
+{
+	return cfd->ctl_payload_len & CFD_PAYLOAD_LEN_MASK;
+}
+
+/*
+ * control descriptor
+ */
+struct vpdma_ctd {
+	union {
+		u32	timer_value;
+		u32	list_addr;
+		u32	w0;
+	};
+	union {
+		u32	pixel_line_count;
+		u32	list_size;
+		u32	w1;
+	};
+	union {
+		u32	event;
+		u32	fid_ctl;
+		u32	w2;
+	};
+	u32		type_source_ctl;
+};
+
+/* control descriptor types */
+#define CTD_TYPE_SYNC_ON_CLIENT		0
+#define CTD_TYPE_SYNC_ON_LIST		1
+#define CTD_TYPE_SYNC_ON_EXT		2
+#define CTD_TYPE_SYNC_ON_LM_TIMER	3
+#define CTD_TYPE_SYNC_ON_CHANNEL	4
+#define CTD_TYPE_CHNG_CLIENT_IRQ	5
+#define CTD_TYPE_SEND_IRQ		6
+#define CTD_TYPE_RELOAD_LIST		7
+#define CTD_TYPE_ABORT_CHANNEL		8
+
+#define CTD_PKT_TYPE		0xc
+
+/* timer_value */
+#define CTD_TIMER_VALUE_MASK	0xffff
+#define CTD_TIMER_VALUE_SHFT	0
+
+/* pixel_line_count */
+#define CTD_PIXEL_COUNT_MASK	0xffff
+#define CTD_PIXEL_COUNT_SHFT	16
+#define CTD_LINE_COUNT_MASK	0xffff
+#define CTD_LINE_COUNT_SHFT	0
+
+/* list_size */
+#define CTD_LIST_SIZE_MASK	0xffff
+#define CTD_LIST_SIZE_SHFT	0
+
+/* event */
+#define CTD_EVENT_MASK		0x0f
+#define CTD_EVENT_SHFT		0
+
+/* fid_ctl */
+#define CTD_FID2_MASK		0x03
+#define CTD_FID2_SHFT		4
+#define CTD_FID1_MASK		0x03
+#define CTD_FID1_SHFT		2
+#define CTD_FID0_MASK		0x03
+#define CTD_FID0_SHFT		0
+
+/* type_source_ctl */
+#define CTD_PKT_TYPE_MASK	0x1f
+#define CTD_PKT_TYPE_SHFT	27
+#define CTD_SOURCE_MASK		0xff
+#define CTD_SOURCE_SHFT		16
+#define CTD_CONTROL_MASK	0x0f
+#define CTD_CONTROL_SHFT	0
+
+static inline u32 ctd_pixel_line_count(int pixel_count, int line_count)
+{
+	return (pixel_count << CTD_PIXEL_COUNT_SHFT) | line_count;
+}
+
+static inline u32 ctd_set_fid_ctl(int fid0, int fid1, int fid2)
+{
+	return (fid2 << CTD_FID2_SHFT) | (fid1 << CTD_FID1_SHFT) | fid0;
+}
+
+static inline u32 ctd_type_source_ctl(int source, int control)
+{
+	return (CTD_PKT_TYPE << CTD_PKT_TYPE_SHFT) |
+		(source << CTD_SOURCE_SHFT) | control;
+}
+
+static inline u32 ctd_get_pixel_count(struct vpdma_ctd *ctd)
+{
+	return ctd->pixel_line_count >> CTD_PIXEL_COUNT_SHFT;
+}
+
+static inline int ctd_get_line_count(struct vpdma_ctd *ctd)
+{
+	return ctd->pixel_line_count & CTD_LINE_COUNT_MASK;
+}
+
+static inline int ctd_get_event(struct vpdma_ctd *ctd)
+{
+	return ctd->event & CTD_EVENT_MASK;
+}
+
+static inline int ctd_get_fid2_ctl(struct vpdma_ctd *ctd)
+{
+	return (ctd->fid_ctl >> CTD_FID2_SHFT) & CTD_FID2_MASK;
+}
+
+static inline int ctd_get_fid1_ctl(struct vpdma_ctd *ctd)
+{
+	return (ctd->fid_ctl >> CTD_FID1_SHFT) & CTD_FID1_MASK;
+}
+
+static inline int ctd_get_fid0_ctl(struct vpdma_ctd *ctd)
+{
+	return ctd->fid_ctl & CTD_FID2_MASK;
+}
+
+static inline int ctd_get_pkt_type(struct vpdma_ctd *ctd)
+{
+	return ctd->type_source_ctl >> CTD_PKT_TYPE_SHFT;
+}
+
+static inline int ctd_get_source(struct vpdma_ctd *ctd)
+{
+	return (ctd->type_source_ctl >> CTD_SOURCE_SHFT) & CTD_SOURCE_MASK;
+}
+
+static inline int ctd_get_ctl(struct vpdma_ctd *ctd)
+{
+	return ctd->type_source_ctl & CTD_CONTROL_MASK;
+}
+
+#endif
diff --git a/drivers/media/platform/ti-vpe/vpe.c b/drivers/media/platform/ti-vpe/vpe.c
new file mode 100644
index 000000000000..4e58069e24ff
--- /dev/null
+++ b/drivers/media/platform/ti-vpe/vpe.c
@@ -0,0 +1,2099 @@
+/*
+ * TI VPE mem2mem driver, based on the virtual v4l2-mem2mem example driver
+ *
+ * Copyright (c) 2013 Texas Instruments Inc.
+ * David Griego, <dagriego@biglakesoftware.com>
+ * Dale Farnsworth, <dale@farnsworth.org>
+ * Archit Taneja, <archit@ti.com>
+ *
+ * Copyright (c) 2009-2010 Samsung Electronics Co., Ltd.
+ * Pawel Osciak, <pawel@osciak.com>
+ * Marek Szyprowski, <m.szyprowski@samsung.com>
+ *
+ * Based on the virtual v4l2-mem2mem example device
+ *
+ * This program is free software; you can redistribute it and/or modify it
+ * under the terms of the GNU General Public License version 2 as published by
+ * the Free Software Foundation
+ */
+
+#include <linux/delay.h>
+#include <linux/dma-mapping.h>
+#include <linux/err.h>
+#include <linux/fs.h>
+#include <linux/interrupt.h>
+#include <linux/io.h>
+#include <linux/ioctl.h>
+#include <linux/module.h>
+#include <linux/platform_device.h>
+#include <linux/pm_runtime.h>
+#include <linux/sched.h>
+#include <linux/slab.h>
+#include <linux/videodev2.h>
+
+#include <media/v4l2-common.h>
+#include <media/v4l2-ctrls.h>
+#include <media/v4l2-device.h>
+#include <media/v4l2-event.h>
+#include <media/v4l2-ioctl.h>
+#include <media/v4l2-mem2mem.h>
+#include <media/videobuf2-core.h>
+#include <media/videobuf2-dma-contig.h>
+
+#include "vpdma.h"
+#include "vpe_regs.h"
+
+#define VPE_MODULE_NAME "vpe"
+
+/* minimum and maximum frame sizes */
+#define MIN_W		128
+#define MIN_H		128
+#define MAX_W		1920
+#define MAX_H		1080
+
+/* required alignments */
+#define S_ALIGN		0	/* multiple of 1 */
+#define H_ALIGN		1	/* multiple of 2 */
+#define W_ALIGN		1	/* multiple of 2 */
+
+/* multiple of 128 bits, line stride, 16 bytes */
+#define L_ALIGN		4
+
+/* flags that indicate a format can be used for capture/output */
+#define VPE_FMT_TYPE_CAPTURE	(1 << 0)
+#define VPE_FMT_TYPE_OUTPUT	(1 << 1)
+
+/* used as plane indices */
+#define VPE_MAX_PLANES	2
+#define VPE_LUMA	0
+#define VPE_CHROMA	1
+
+/* per m2m context info */
+#define VPE_MAX_SRC_BUFS	3	/* need 3 src fields to de-interlace */
+
+#define VPE_DEF_BUFS_PER_JOB	1	/* default one buffer per batch job */
+
+/*
+ * each VPE context can need up to 3 config desciptors, 7 input descriptors,
+ * 3 output descriptors, and 10 control descriptors
+ */
+#define VPE_DESC_LIST_SIZE	(10 * VPDMA_DTD_DESC_SIZE +	\
+					13 * VPDMA_CFD_CTD_DESC_SIZE)
+
+#define vpe_dbg(vpedev, fmt, arg...)	\
+		dev_dbg((vpedev)->v4l2_dev.dev, fmt, ##arg)
+#define vpe_err(vpedev, fmt, arg...)	\
+		dev_err((vpedev)->v4l2_dev.dev, fmt, ##arg)
+
+struct vpe_us_coeffs {
+	unsigned short	anchor_fid0_c0;
+	unsigned short	anchor_fid0_c1;
+	unsigned short	anchor_fid0_c2;
+	unsigned short	anchor_fid0_c3;
+	unsigned short	interp_fid0_c0;
+	unsigned short	interp_fid0_c1;
+	unsigned short	interp_fid0_c2;
+	unsigned short	interp_fid0_c3;
+	unsigned short	anchor_fid1_c0;
+	unsigned short	anchor_fid1_c1;
+	unsigned short	anchor_fid1_c2;
+	unsigned short	anchor_fid1_c3;
+	unsigned short	interp_fid1_c0;
+	unsigned short	interp_fid1_c1;
+	unsigned short	interp_fid1_c2;
+	unsigned short	interp_fid1_c3;
+};
+
+/*
+ * Default upsampler coefficients
+ */
+static const struct vpe_us_coeffs us_coeffs[] = {
+	{
+		/* Coefficients for progressive input */
+		0x00C8, 0x0348, 0x0018, 0x3FD8, 0x3FB8, 0x0378, 0x00E8, 0x3FE8,
+		0x00C8, 0x0348, 0x0018, 0x3FD8, 0x3FB8, 0x0378, 0x00E8, 0x3FE8,
+	},
+	{
+		/* Coefficients for Top Field Interlaced input */
+		0x0051, 0x03D5, 0x3FE3, 0x3FF7, 0x3FB5, 0x02E9, 0x018F, 0x3FD3,
+		/* Coefficients for Bottom Field Interlaced input */
+		0x016B, 0x0247, 0x00B1, 0x3F9D, 0x3FCF, 0x03DB, 0x005D, 0x3FF9,
+	},
+};
+
+/*
+ * the following registers are for configuring some of the parameters of the
+ * motion and edge detection blocks inside DEI, these generally remain the same,
+ * these could be passed later via userspace if some one needs to tweak these.
+ */
+struct vpe_dei_regs {
+	unsigned long mdt_spacial_freq_thr_reg;		/* VPE_DEI_REG2 */
+	unsigned long edi_config_reg;			/* VPE_DEI_REG3 */
+	unsigned long edi_lut_reg0;			/* VPE_DEI_REG4 */
+	unsigned long edi_lut_reg1;			/* VPE_DEI_REG5 */
+	unsigned long edi_lut_reg2;			/* VPE_DEI_REG6 */
+	unsigned long edi_lut_reg3;			/* VPE_DEI_REG7 */
+};
+
+/*
+ * default expert DEI register values, unlikely to be modified.
+ */
+static const struct vpe_dei_regs dei_regs = {
+	0x020C0804u,
+	0x0118100Fu,
+	0x08040200u,
+	0x1010100Cu,
+	0x10101010u,
+	0x10101010u,
+};
+
+/*
+ * The port_data structure contains per-port data.
+ */
+struct vpe_port_data {
+	enum vpdma_channel channel;	/* VPDMA channel */
+	u8	vb_index;		/* input frame f, f-1, f-2 index */
+	u8	vb_part;		/* plane index for co-panar formats */
+};
+
+/*
+ * Define indices into the port_data tables
+ */
+#define VPE_PORT_LUMA1_IN	0
+#define VPE_PORT_CHROMA1_IN	1
+#define VPE_PORT_LUMA2_IN	2
+#define VPE_PORT_CHROMA2_IN	3
+#define VPE_PORT_LUMA3_IN	4
+#define VPE_PORT_CHROMA3_IN	5
+#define VPE_PORT_MV_IN		6
+#define VPE_PORT_MV_OUT		7
+#define VPE_PORT_LUMA_OUT	8
+#define VPE_PORT_CHROMA_OUT	9
+#define VPE_PORT_RGB_OUT	10
+
+static const struct vpe_port_data port_data[11] = {
+	[VPE_PORT_LUMA1_IN] = {
+		.channel	= VPE_CHAN_LUMA1_IN,
+		.vb_index	= 0,
+		.vb_part	= VPE_LUMA,
+	},
+	[VPE_PORT_CHROMA1_IN] = {
+		.channel	= VPE_CHAN_CHROMA1_IN,
+		.vb_index	= 0,
+		.vb_part	= VPE_CHROMA,
+	},
+	[VPE_PORT_LUMA2_IN] = {
+		.channel	= VPE_CHAN_LUMA2_IN,
+		.vb_index	= 1,
+		.vb_part	= VPE_LUMA,
+	},
+	[VPE_PORT_CHROMA2_IN] = {
+		.channel	= VPE_CHAN_CHROMA2_IN,
+		.vb_index	= 1,
+		.vb_part	= VPE_CHROMA,
+	},
+	[VPE_PORT_LUMA3_IN] = {
+		.channel	= VPE_CHAN_LUMA3_IN,
+		.vb_index	= 2,
+		.vb_part	= VPE_LUMA,
+	},
+	[VPE_PORT_CHROMA3_IN] = {
+		.channel	= VPE_CHAN_CHROMA3_IN,
+		.vb_index	= 2,
+		.vb_part	= VPE_CHROMA,
+	},
+	[VPE_PORT_MV_IN] = {
+		.channel	= VPE_CHAN_MV_IN,
+	},
+	[VPE_PORT_MV_OUT] = {
+		.channel	= VPE_CHAN_MV_OUT,
+	},
+	[VPE_PORT_LUMA_OUT] = {
+		.channel	= VPE_CHAN_LUMA_OUT,
+		.vb_part	= VPE_LUMA,
+	},
+	[VPE_PORT_CHROMA_OUT] = {
+		.channel	= VPE_CHAN_CHROMA_OUT,
+		.vb_part	= VPE_CHROMA,
+	},
+	[VPE_PORT_RGB_OUT] = {
+		.channel	= VPE_CHAN_RGB_OUT,
+		.vb_part	= VPE_LUMA,
+	},
+};
+
+
+/* driver info for each of the supported video formats */
+struct vpe_fmt {
+	char	*name;			/* human-readable name */
+	u32	fourcc;			/* standard format identifier */
+	u8	types;			/* CAPTURE and/or OUTPUT */
+	u8	coplanar;		/* set for unpacked Luma and Chroma */
+	/* vpdma format info for each plane */
+	struct vpdma_data_format const *vpdma_fmt[VPE_MAX_PLANES];
+};
+
+static struct vpe_fmt vpe_formats[] = {
+	{
+		.name		= "YUV 422 co-planar",
+		.fourcc		= V4L2_PIX_FMT_NV16,
+		.types		= VPE_FMT_TYPE_CAPTURE | VPE_FMT_TYPE_OUTPUT,
+		.coplanar	= 1,
+		.vpdma_fmt	= { &vpdma_yuv_fmts[VPDMA_DATA_FMT_Y444],
+				    &vpdma_yuv_fmts[VPDMA_DATA_FMT_C444],
+				  },
+	},
+	{
+		.name		= "YUV 420 co-planar",
+		.fourcc		= V4L2_PIX_FMT_NV12,
+		.types		= VPE_FMT_TYPE_CAPTURE | VPE_FMT_TYPE_OUTPUT,
+		.coplanar	= 1,
+		.vpdma_fmt	= { &vpdma_yuv_fmts[VPDMA_DATA_FMT_Y420],
+				    &vpdma_yuv_fmts[VPDMA_DATA_FMT_C420],
+				  },
+	},
+	{
+		.name		= "YUYV 422 packed",
+		.fourcc		= V4L2_PIX_FMT_YUYV,
+		.types		= VPE_FMT_TYPE_CAPTURE | VPE_FMT_TYPE_OUTPUT,
+		.coplanar	= 0,
+		.vpdma_fmt	= { &vpdma_yuv_fmts[VPDMA_DATA_FMT_YC422],
+				  },
+	},
+	{
+		.name		= "UYVY 422 packed",
+		.fourcc		= V4L2_PIX_FMT_UYVY,
+		.types		= VPE_FMT_TYPE_CAPTURE | VPE_FMT_TYPE_OUTPUT,
+		.coplanar	= 0,
+		.vpdma_fmt	= { &vpdma_yuv_fmts[VPDMA_DATA_FMT_CY422],
+				  },
+	},
+};
+
+/*
+ * per-queue, driver-specific private data.
+ * there is one source queue and one destination queue for each m2m context.
+ */
+struct vpe_q_data {
+	unsigned int		width;				/* frame width */
+	unsigned int		height;				/* frame height */
+	unsigned int		bytesperline[VPE_MAX_PLANES];	/* bytes per line in memory */
+	enum v4l2_colorspace	colorspace;
+	enum v4l2_field		field;				/* supported field value */
+	unsigned int		flags;
+	unsigned int		sizeimage[VPE_MAX_PLANES];	/* image size in memory */
+	struct v4l2_rect	c_rect;				/* crop/compose rectangle */
+	struct vpe_fmt		*fmt;				/* format info */
+};
+
+/* vpe_q_data flag bits */
+#define	Q_DATA_FRAME_1D		(1 << 0)
+#define	Q_DATA_MODE_TILED	(1 << 1)
+#define	Q_DATA_INTERLACED	(1 << 2)
+
+enum {
+	Q_DATA_SRC = 0,
+	Q_DATA_DST = 1,
+};
+
+/* find our format description corresponding to the passed v4l2_format */
+static struct vpe_fmt *find_format(struct v4l2_format *f)
+{
+	struct vpe_fmt *fmt;
+	unsigned int k;
+
+	for (k = 0; k < ARRAY_SIZE(vpe_formats); k++) {
+		fmt = &vpe_formats[k];
+		if (fmt->fourcc == f->fmt.pix.pixelformat)
+			return fmt;
+	}
+
+	return NULL;
+}
+
+/*
+ * there is one vpe_dev structure in the driver, it is shared by
+ * all instances.
+ */
+struct vpe_dev {
+	struct v4l2_device	v4l2_dev;
+	struct video_device	vfd;
+	struct v4l2_m2m_dev	*m2m_dev;
+
+	atomic_t		num_instances;	/* count of driver instances */
+	dma_addr_t		loaded_mmrs;	/* shadow mmrs in device */
+	struct mutex		dev_mutex;
+	spinlock_t		lock;
+
+	int			irq;
+	void __iomem		*base;
+
+	struct vb2_alloc_ctx	*alloc_ctx;
+	struct vpdma_data	*vpdma;		/* vpdma data handle */
+};
+
+/*
+ * There is one vpe_ctx structure for each m2m context.
+ */
+struct vpe_ctx {
+	struct v4l2_fh		fh;
+	struct vpe_dev		*dev;
+	struct v4l2_m2m_ctx	*m2m_ctx;
+	struct v4l2_ctrl_handler hdl;
+
+	unsigned int		field;			/* current field */
+	unsigned int		sequence;		/* current frame/field seq */
+	unsigned int		aborting;		/* abort after next irq */
+
+	unsigned int		bufs_per_job;		/* input buffers per batch */
+	unsigned int		bufs_completed;		/* bufs done in this batch */
+
+	struct vpe_q_data	q_data[2];		/* src & dst queue data */
+	struct vb2_buffer	*src_vbs[VPE_MAX_SRC_BUFS];
+	struct vb2_buffer	*dst_vb;
+
+	dma_addr_t		mv_buf_dma[2];		/* dma addrs of motion vector in/out bufs */
+	void			*mv_buf[2];		/* virtual addrs of motion vector bufs */
+	size_t			mv_buf_size;		/* current motion vector buffer size */
+	struct vpdma_buf	mmr_adb;		/* shadow reg addr/data block */
+	struct vpdma_desc_list	desc_list;		/* DMA descriptor list */
+
+	bool			deinterlacing;		/* using de-interlacer */
+	bool			load_mmrs;		/* have new shadow reg values */
+
+	unsigned int		src_mv_buf_selector;
+};
+
+
+/*
+ * M2M devices get 2 queues.
+ * Return the queue given the type.
+ */
+static struct vpe_q_data *get_q_data(struct vpe_ctx *ctx,
+				     enum v4l2_buf_type type)
+{
+	switch (type) {
+	case V4L2_BUF_TYPE_VIDEO_OUTPUT_MPLANE:
+		return &ctx->q_data[Q_DATA_SRC];
+	case V4L2_BUF_TYPE_VIDEO_CAPTURE_MPLANE:
+		return &ctx->q_data[Q_DATA_DST];
+	default:
+		BUG();
+	}
+	return NULL;
+}
+
+static u32 read_reg(struct vpe_dev *dev, int offset)
+{
+	return ioread32(dev->base + offset);
+}
+
+static void write_reg(struct vpe_dev *dev, int offset, u32 value)
+{
+	iowrite32(value, dev->base + offset);
+}
+
+/* register field read/write helpers */
+static int get_field(u32 value, u32 mask, int shift)
+{
+	return (value & (mask << shift)) >> shift;
+}
+
+static int read_field_reg(struct vpe_dev *dev, int offset, u32 mask, int shift)
+{
+	return get_field(read_reg(dev, offset), mask, shift);
+}
+
+static void write_field(u32 *valp, u32 field, u32 mask, int shift)
+{
+	u32 val = *valp;
+
+	val &= ~(mask << shift);
+	val |= (field & mask) << shift;
+	*valp = val;
+}
+
+static void write_field_reg(struct vpe_dev *dev, int offset, u32 field,
+		u32 mask, int shift)
+{
+	u32 val = read_reg(dev, offset);
+
+	write_field(&val, field, mask, shift);
+
+	write_reg(dev, offset, val);
+}
+
+/*
+ * DMA address/data block for the shadow registers
+ */
+struct vpe_mmr_adb {
+	struct vpdma_adb_hdr	out_fmt_hdr;
+	u32			out_fmt_reg[1];
+	u32			out_fmt_pad[3];
+	struct vpdma_adb_hdr	us1_hdr;
+	u32			us1_regs[8];
+	struct vpdma_adb_hdr	us2_hdr;
+	u32			us2_regs[8];
+	struct vpdma_adb_hdr	us3_hdr;
+	u32			us3_regs[8];
+	struct vpdma_adb_hdr	dei_hdr;
+	u32			dei_regs[8];
+	struct vpdma_adb_hdr	sc_hdr;
+	u32			sc_regs[1];
+	u32			sc_pad[3];
+	struct vpdma_adb_hdr	csc_hdr;
+	u32			csc_regs[6];
+	u32			csc_pad[2];
+};
+
+#define VPE_SET_MMR_ADB_HDR(ctx, hdr, regs, offset_a)	\
+	VPDMA_SET_MMR_ADB_HDR(ctx->mmr_adb, vpe_mmr_adb, hdr, regs, offset_a)
+/*
+ * Set the headers for all of the address/data block structures.
+ */
+static void init_adb_hdrs(struct vpe_ctx *ctx)
+{
+	VPE_SET_MMR_ADB_HDR(ctx, out_fmt_hdr, out_fmt_reg, VPE_CLK_FORMAT_SELECT);
+	VPE_SET_MMR_ADB_HDR(ctx, us1_hdr, us1_regs, VPE_US1_R0);
+	VPE_SET_MMR_ADB_HDR(ctx, us2_hdr, us2_regs, VPE_US2_R0);
+	VPE_SET_MMR_ADB_HDR(ctx, us3_hdr, us3_regs, VPE_US3_R0);
+	VPE_SET_MMR_ADB_HDR(ctx, dei_hdr, dei_regs, VPE_DEI_FRAME_SIZE);
+	VPE_SET_MMR_ADB_HDR(ctx, sc_hdr, sc_regs, VPE_SC_MP_SC0);
+	VPE_SET_MMR_ADB_HDR(ctx, csc_hdr, csc_regs, VPE_CSC_CSC00);
+};
+
+/*
+ * Allocate or re-allocate the motion vector DMA buffers
+ * There are two buffers, one for input and one for output.
+ * However, the roles are reversed after each field is processed.
+ * In other words, after each field is processed, the previous
+ * output (dst) MV buffer becomes the new input (src) MV buffer.
+ */
+static int realloc_mv_buffers(struct vpe_ctx *ctx, size_t size)
+{
+	struct device *dev = ctx->dev->v4l2_dev.dev;
+
+	if (ctx->mv_buf_size == size)
+		return 0;
+
+	if (ctx->mv_buf[0])
+		dma_free_coherent(dev, ctx->mv_buf_size, ctx->mv_buf[0],
+			ctx->mv_buf_dma[0]);
+
+	if (ctx->mv_buf[1])
+		dma_free_coherent(dev, ctx->mv_buf_size, ctx->mv_buf[1],
+			ctx->mv_buf_dma[1]);
+
+	if (size == 0)
+		return 0;
+
+	ctx->mv_buf[0] = dma_alloc_coherent(dev, size, &ctx->mv_buf_dma[0],
+				GFP_KERNEL);
+	if (!ctx->mv_buf[0]) {
+		vpe_err(ctx->dev, "failed to allocate motion vector buffer\n");
+		return -ENOMEM;
+	}
+
+	ctx->mv_buf[1] = dma_alloc_coherent(dev, size, &ctx->mv_buf_dma[1],
+				GFP_KERNEL);
+	if (!ctx->mv_buf[1]) {
+		vpe_err(ctx->dev, "failed to allocate motion vector buffer\n");
+		dma_free_coherent(dev, size, ctx->mv_buf[0],
+			ctx->mv_buf_dma[0]);
+
+		return -ENOMEM;
+	}
+
+	ctx->mv_buf_size = size;
+	ctx->src_mv_buf_selector = 0;
+
+	return 0;
+}
+
+static void free_mv_buffers(struct vpe_ctx *ctx)
+{
+	realloc_mv_buffers(ctx, 0);
+}
+
+/*
+ * While de-interlacing, we keep the two most recent input buffers
+ * around.  This function frees those two buffers when we have
+ * finished processing the current stream.
+ */
+static void free_vbs(struct vpe_ctx *ctx)
+{
+	struct vpe_dev *dev = ctx->dev;
+	unsigned long flags;
+
+	if (ctx->src_vbs[2] == NULL)
+		return;
+
+	spin_lock_irqsave(&dev->lock, flags);
+	if (ctx->src_vbs[2]) {
+		v4l2_m2m_buf_done(ctx->src_vbs[2], VB2_BUF_STATE_DONE);
+		v4l2_m2m_buf_done(ctx->src_vbs[1], VB2_BUF_STATE_DONE);
+	}
+	spin_unlock_irqrestore(&dev->lock, flags);
+}
+
+/*
+ * Enable or disable the VPE clocks
+ */
+static void vpe_set_clock_enable(struct vpe_dev *dev, bool on)
+{
+	u32 val = 0;
+
+	if (on)
+		val = VPE_DATA_PATH_CLK_ENABLE | VPE_VPEDMA_CLK_ENABLE;
+	write_reg(dev, VPE_CLK_ENABLE, val);
+}
+
+static void vpe_top_reset(struct vpe_dev *dev)
+{
+
+	write_field_reg(dev, VPE_CLK_RESET, 1, VPE_DATA_PATH_CLK_RESET_MASK,
+		VPE_DATA_PATH_CLK_RESET_SHIFT);
+
+	usleep_range(100, 150);
+
+	write_field_reg(dev, VPE_CLK_RESET, 0, VPE_DATA_PATH_CLK_RESET_MASK,
+		VPE_DATA_PATH_CLK_RESET_SHIFT);
+}
+
+static void vpe_top_vpdma_reset(struct vpe_dev *dev)
+{
+	write_field_reg(dev, VPE_CLK_RESET, 1, VPE_VPDMA_CLK_RESET_MASK,
+		VPE_VPDMA_CLK_RESET_SHIFT);
+
+	usleep_range(100, 150);
+
+	write_field_reg(dev, VPE_CLK_RESET, 0, VPE_VPDMA_CLK_RESET_MASK,
+		VPE_VPDMA_CLK_RESET_SHIFT);
+}
+
+/*
+ * Load the correct of upsampler coefficients into the shadow MMRs
+ */
+static void set_us_coefficients(struct vpe_ctx *ctx)
+{
+	struct vpe_mmr_adb *mmr_adb = ctx->mmr_adb.addr;
+	struct vpe_q_data *s_q_data = &ctx->q_data[Q_DATA_SRC];
+	u32 *us1_reg = &mmr_adb->us1_regs[0];
+	u32 *us2_reg = &mmr_adb->us2_regs[0];
+	u32 *us3_reg = &mmr_adb->us3_regs[0];
+	const unsigned short *cp, *end_cp;
+
+	cp = &us_coeffs[0].anchor_fid0_c0;
+
+	if (s_q_data->flags & Q_DATA_INTERLACED)	/* interlaced */
+		cp += sizeof(us_coeffs[0]) / sizeof(*cp);
+
+	end_cp = cp + sizeof(us_coeffs[0]) / sizeof(*cp);
+
+	while (cp < end_cp) {
+		write_field(us1_reg, *cp++, VPE_US_C0_MASK, VPE_US_C0_SHIFT);
+		write_field(us1_reg, *cp++, VPE_US_C1_MASK, VPE_US_C1_SHIFT);
+		*us2_reg++ = *us1_reg;
+		*us3_reg++ = *us1_reg++;
+	}
+	ctx->load_mmrs = true;
+}
+
+/*
+ * Set the upsampler config mode and the VPDMA line mode in the shadow MMRs.
+ */
+static void set_cfg_and_line_modes(struct vpe_ctx *ctx)
+{
+	struct vpe_fmt *fmt = ctx->q_data[Q_DATA_SRC].fmt;
+	struct vpe_mmr_adb *mmr_adb = ctx->mmr_adb.addr;
+	u32 *us1_reg0 = &mmr_adb->us1_regs[0];
+	u32 *us2_reg0 = &mmr_adb->us2_regs[0];
+	u32 *us3_reg0 = &mmr_adb->us3_regs[0];
+	int line_mode = 1;
+	int cfg_mode = 1;
+
+	/*
+	 * Cfg Mode 0: YUV420 source, enable upsampler, DEI is de-interlacing.
+	 * Cfg Mode 1: YUV422 source, disable upsampler, DEI is de-interlacing.
+	 */
+
+	if (fmt->fourcc == V4L2_PIX_FMT_NV12) {
+		cfg_mode = 0;
+		line_mode = 0;		/* double lines to line buffer */
+	}
+
+	write_field(us1_reg0, cfg_mode, VPE_US_MODE_MASK, VPE_US_MODE_SHIFT);
+	write_field(us2_reg0, cfg_mode, VPE_US_MODE_MASK, VPE_US_MODE_SHIFT);
+	write_field(us3_reg0, cfg_mode, VPE_US_MODE_MASK, VPE_US_MODE_SHIFT);
+
+	/* regs for now */
+	vpdma_set_line_mode(ctx->dev->vpdma, line_mode, VPE_CHAN_CHROMA1_IN);
+	vpdma_set_line_mode(ctx->dev->vpdma, line_mode, VPE_CHAN_CHROMA2_IN);
+	vpdma_set_line_mode(ctx->dev->vpdma, line_mode, VPE_CHAN_CHROMA3_IN);
+
+	/* frame start for input luma */
+	vpdma_set_frame_start_event(ctx->dev->vpdma, VPDMA_FSEVENT_CHANNEL_ACTIVE,
+		VPE_CHAN_LUMA1_IN);
+	vpdma_set_frame_start_event(ctx->dev->vpdma, VPDMA_FSEVENT_CHANNEL_ACTIVE,
+		VPE_CHAN_LUMA2_IN);
+	vpdma_set_frame_start_event(ctx->dev->vpdma, VPDMA_FSEVENT_CHANNEL_ACTIVE,
+		VPE_CHAN_LUMA3_IN);
+
+	/* frame start for input chroma */
+	vpdma_set_frame_start_event(ctx->dev->vpdma, VPDMA_FSEVENT_CHANNEL_ACTIVE,
+		VPE_CHAN_CHROMA1_IN);
+	vpdma_set_frame_start_event(ctx->dev->vpdma, VPDMA_FSEVENT_CHANNEL_ACTIVE,
+		VPE_CHAN_CHROMA2_IN);
+	vpdma_set_frame_start_event(ctx->dev->vpdma, VPDMA_FSEVENT_CHANNEL_ACTIVE,
+		VPE_CHAN_CHROMA3_IN);
+
+	/* frame start for MV in client */
+	vpdma_set_frame_start_event(ctx->dev->vpdma, VPDMA_FSEVENT_CHANNEL_ACTIVE,
+		VPE_CHAN_MV_IN);
+
+	ctx->load_mmrs = true;
+}
+
+/*
+ * Set the shadow registers that are modified when the source
+ * format changes.
+ */
+static void set_src_registers(struct vpe_ctx *ctx)
+{
+	set_us_coefficients(ctx);
+}
+
+/*
+ * Set the shadow registers that are modified when the destination
+ * format changes.
+ */
+static void set_dst_registers(struct vpe_ctx *ctx)
+{
+	struct vpe_mmr_adb *mmr_adb = ctx->mmr_adb.addr;
+	struct vpe_fmt *fmt = ctx->q_data[Q_DATA_DST].fmt;
+	u32 val = 0;
+
+	/* select RGB path when color space conversion is supported in future */
+	if (fmt->fourcc == V4L2_PIX_FMT_RGB24)
+		val |= VPE_RGB_OUT_SELECT | VPE_CSC_SRC_DEI_SCALER;
+	else if (fmt->fourcc == V4L2_PIX_FMT_NV16)
+		val |= VPE_COLOR_SEPARATE_422;
+
+	/* The source of CHR_DS is always the scaler, whether it's used or not */
+	val |= VPE_DS_SRC_DEI_SCALER;
+
+	if (fmt->fourcc != V4L2_PIX_FMT_NV12)
+		val |= VPE_DS_BYPASS;
+
+	mmr_adb->out_fmt_reg[0] = val;
+
+	ctx->load_mmrs = true;
+}
+
+/*
+ * Set the de-interlacer shadow register values
+ */
+static void set_dei_regs(struct vpe_ctx *ctx)
+{
+	struct vpe_mmr_adb *mmr_adb = ctx->mmr_adb.addr;
+	struct vpe_q_data *s_q_data = &ctx->q_data[Q_DATA_SRC];
+	unsigned int src_h = s_q_data->c_rect.height;
+	unsigned int src_w = s_q_data->c_rect.width;
+	u32 *dei_mmr0 = &mmr_adb->dei_regs[0];
+	bool deinterlace = true;
+	u32 val = 0;
+
+	/*
+	 * according to TRM, we should set DEI in progressive bypass mode when
+	 * the input content is progressive, however, DEI is bypassed correctly
+	 * for both progressive and interlace content in interlace bypass mode.
+	 * It has been recommended not to use progressive bypass mode.
+	 */
+	if ((!ctx->deinterlacing && (s_q_data->flags & Q_DATA_INTERLACED)) ||
+			!(s_q_data->flags & Q_DATA_INTERLACED)) {
+		deinterlace = false;
+		val = VPE_DEI_INTERLACE_BYPASS;
+	}
+
+	src_h = deinterlace ? src_h * 2 : src_h;
+
+	val |= (src_h << VPE_DEI_HEIGHT_SHIFT) |
+		(src_w << VPE_DEI_WIDTH_SHIFT) |
+		VPE_DEI_FIELD_FLUSH;
+
+	*dei_mmr0 = val;
+
+	ctx->load_mmrs = true;
+}
+
+static void set_dei_shadow_registers(struct vpe_ctx *ctx)
+{
+	struct vpe_mmr_adb *mmr_adb = ctx->mmr_adb.addr;
+	u32 *dei_mmr = &mmr_adb->dei_regs[0];
+	const struct vpe_dei_regs *cur = &dei_regs;
+
+	dei_mmr[2]  = cur->mdt_spacial_freq_thr_reg;
+	dei_mmr[3]  = cur->edi_config_reg;
+	dei_mmr[4]  = cur->edi_lut_reg0;
+	dei_mmr[5]  = cur->edi_lut_reg1;
+	dei_mmr[6]  = cur->edi_lut_reg2;
+	dei_mmr[7]  = cur->edi_lut_reg3;
+
+	ctx->load_mmrs = true;
+}
+
+static void set_csc_coeff_bypass(struct vpe_ctx *ctx)
+{
+	struct vpe_mmr_adb *mmr_adb = ctx->mmr_adb.addr;
+	u32 *shadow_csc_reg5 = &mmr_adb->csc_regs[5];
+
+	*shadow_csc_reg5 |= VPE_CSC_BYPASS;
+
+	ctx->load_mmrs = true;
+}
+
+static void set_sc_regs_bypass(struct vpe_ctx *ctx)
+{
+	struct vpe_mmr_adb *mmr_adb = ctx->mmr_adb.addr;
+	u32 *sc_reg0 = &mmr_adb->sc_regs[0];
+	u32 val = 0;
+
+	val |= VPE_SC_BYPASS;
+	*sc_reg0 = val;
+
+	ctx->load_mmrs = true;
+}
+
+/*
+ * Set the shadow registers whose values are modified when either the
+ * source or destination format is changed.
+ */
+static int set_srcdst_params(struct vpe_ctx *ctx)
+{
+	struct vpe_q_data *s_q_data =  &ctx->q_data[Q_DATA_SRC];
+	struct vpe_q_data *d_q_data =  &ctx->q_data[Q_DATA_DST];
+	size_t mv_buf_size;
+	int ret;
+
+	ctx->sequence = 0;
+	ctx->field = V4L2_FIELD_TOP;
+
+	if ((s_q_data->flags & Q_DATA_INTERLACED) &&
+			!(d_q_data->flags & Q_DATA_INTERLACED)) {
+		const struct vpdma_data_format *mv =
+			&vpdma_misc_fmts[VPDMA_DATA_FMT_MV];
+
+		ctx->deinterlacing = 1;
+		mv_buf_size =
+			(s_q_data->width * s_q_data->height * mv->depth) >> 3;
+	} else {
+		ctx->deinterlacing = 0;
+		mv_buf_size = 0;
+	}
+
+	free_vbs(ctx);
+
+	ret = realloc_mv_buffers(ctx, mv_buf_size);
+	if (ret)
+		return ret;
+
+	set_cfg_and_line_modes(ctx);
+	set_dei_regs(ctx);
+	set_csc_coeff_bypass(ctx);
+	set_sc_regs_bypass(ctx);
+
+	return 0;
+}
+
+/*
+ * Return the vpe_ctx structure for a given struct file
+ */
+static struct vpe_ctx *file2ctx(struct file *file)
+{
+	return container_of(file->private_data, struct vpe_ctx, fh);
+}
+
+/*
+ * mem2mem callbacks
+ */
+
+/**
+ * job_ready() - check whether an instance is ready to be scheduled to run
+ */
+static int job_ready(void *priv)
+{
+	struct vpe_ctx *ctx = priv;
+	int needed = ctx->bufs_per_job;
+
+	if (ctx->deinterlacing && ctx->src_vbs[2] == NULL)
+		needed += 2;	/* need additional two most recent fields */
+
+	if (v4l2_m2m_num_src_bufs_ready(ctx->m2m_ctx) < needed)
+		return 0;
+
+	return 1;
+}
+
+static void job_abort(void *priv)
+{
+	struct vpe_ctx *ctx = priv;
+
+	/* Will cancel the transaction in the next interrupt handler */
+	ctx->aborting = 1;
+}
+
+/*
+ * Lock access to the device
+ */
+static void vpe_lock(void *priv)
+{
+	struct vpe_ctx *ctx = priv;
+	struct vpe_dev *dev = ctx->dev;
+	mutex_lock(&dev->dev_mutex);
+}
+
+static void vpe_unlock(void *priv)
+{
+	struct vpe_ctx *ctx = priv;
+	struct vpe_dev *dev = ctx->dev;
+	mutex_unlock(&dev->dev_mutex);
+}
+
+static void vpe_dump_regs(struct vpe_dev *dev)
+{
+#define DUMPREG(r) vpe_dbg(dev, "%-35s %08x\n", #r, read_reg(dev, VPE_##r))
+
+	vpe_dbg(dev, "VPE Registers:\n");
+
+	DUMPREG(PID);
+	DUMPREG(SYSCONFIG);
+	DUMPREG(INT0_STATUS0_RAW);
+	DUMPREG(INT0_STATUS0);
+	DUMPREG(INT0_ENABLE0);
+	DUMPREG(INT0_STATUS1_RAW);
+	DUMPREG(INT0_STATUS1);
+	DUMPREG(INT0_ENABLE1);
+	DUMPREG(CLK_ENABLE);
+	DUMPREG(CLK_RESET);
+	DUMPREG(CLK_FORMAT_SELECT);
+	DUMPREG(CLK_RANGE_MAP);
+	DUMPREG(US1_R0);
+	DUMPREG(US1_R1);
+	DUMPREG(US1_R2);
+	DUMPREG(US1_R3);
+	DUMPREG(US1_R4);
+	DUMPREG(US1_R5);
+	DUMPREG(US1_R6);
+	DUMPREG(US1_R7);
+	DUMPREG(US2_R0);
+	DUMPREG(US2_R1);
+	DUMPREG(US2_R2);
+	DUMPREG(US2_R3);
+	DUMPREG(US2_R4);
+	DUMPREG(US2_R5);
+	DUMPREG(US2_R6);
+	DUMPREG(US2_R7);
+	DUMPREG(US3_R0);
+	DUMPREG(US3_R1);
+	DUMPREG(US3_R2);
+	DUMPREG(US3_R3);
+	DUMPREG(US3_R4);
+	DUMPREG(US3_R5);
+	DUMPREG(US3_R6);
+	DUMPREG(US3_R7);
+	DUMPREG(DEI_FRAME_SIZE);
+	DUMPREG(MDT_BYPASS);
+	DUMPREG(MDT_SF_THRESHOLD);
+	DUMPREG(EDI_CONFIG);
+	DUMPREG(DEI_EDI_LUT_R0);
+	DUMPREG(DEI_EDI_LUT_R1);
+	DUMPREG(DEI_EDI_LUT_R2);
+	DUMPREG(DEI_EDI_LUT_R3);
+	DUMPREG(DEI_FMD_WINDOW_R0);
+	DUMPREG(DEI_FMD_WINDOW_R1);
+	DUMPREG(DEI_FMD_CONTROL_R0);
+	DUMPREG(DEI_FMD_CONTROL_R1);
+	DUMPREG(DEI_FMD_STATUS_R0);
+	DUMPREG(DEI_FMD_STATUS_R1);
+	DUMPREG(DEI_FMD_STATUS_R2);
+	DUMPREG(SC_MP_SC0);
+	DUMPREG(SC_MP_SC1);
+	DUMPREG(SC_MP_SC2);
+	DUMPREG(SC_MP_SC3);
+	DUMPREG(SC_MP_SC4);
+	DUMPREG(SC_MP_SC5);
+	DUMPREG(SC_MP_SC6);
+	DUMPREG(SC_MP_SC8);
+	DUMPREG(SC_MP_SC9);
+	DUMPREG(SC_MP_SC10);
+	DUMPREG(SC_MP_SC11);
+	DUMPREG(SC_MP_SC12);
+	DUMPREG(SC_MP_SC13);
+	DUMPREG(SC_MP_SC17);
+	DUMPREG(SC_MP_SC18);
+	DUMPREG(SC_MP_SC19);
+	DUMPREG(SC_MP_SC20);
+	DUMPREG(SC_MP_SC21);
+	DUMPREG(SC_MP_SC22);
+	DUMPREG(SC_MP_SC23);
+	DUMPREG(SC_MP_SC24);
+	DUMPREG(SC_MP_SC25);
+	DUMPREG(CSC_CSC00);
+	DUMPREG(CSC_CSC01);
+	DUMPREG(CSC_CSC02);
+	DUMPREG(CSC_CSC03);
+	DUMPREG(CSC_CSC04);
+	DUMPREG(CSC_CSC05);
+#undef DUMPREG
+}
+
+static void add_out_dtd(struct vpe_ctx *ctx, int port)
+{
+	struct vpe_q_data *q_data = &ctx->q_data[Q_DATA_DST];
+	const struct vpe_port_data *p_data = &port_data[port];
+	struct vb2_buffer *vb = ctx->dst_vb;
+	struct v4l2_rect *c_rect = &q_data->c_rect;
+	struct vpe_fmt *fmt = q_data->fmt;
+	const struct vpdma_data_format *vpdma_fmt;
+	int mv_buf_selector = !ctx->src_mv_buf_selector;
+	dma_addr_t dma_addr;
+	u32 flags = 0;
+
+	if (port == VPE_PORT_MV_OUT) {
+		vpdma_fmt = &vpdma_misc_fmts[VPDMA_DATA_FMT_MV];
+		dma_addr = ctx->mv_buf_dma[mv_buf_selector];
+	} else {
+		/* to incorporate interleaved formats */
+		int plane = fmt->coplanar ? p_data->vb_part : 0;
+
+		vpdma_fmt = fmt->vpdma_fmt[plane];
+		dma_addr = vb2_dma_contig_plane_dma_addr(vb, plane);
+		if (!dma_addr) {
+			vpe_err(ctx->dev,
+				"acquiring output buffer(%d) dma_addr failed\n",
+				port);
+			return;
+		}
+	}
+
+	if (q_data->flags & Q_DATA_FRAME_1D)
+		flags |= VPDMA_DATA_FRAME_1D;
+	if (q_data->flags & Q_DATA_MODE_TILED)
+		flags |= VPDMA_DATA_MODE_TILED;
+
+	vpdma_add_out_dtd(&ctx->desc_list, c_rect, vpdma_fmt, dma_addr,
+		p_data->channel, flags);
+}
+
+static void add_in_dtd(struct vpe_ctx *ctx, int port)
+{
+	struct vpe_q_data *q_data = &ctx->q_data[Q_DATA_SRC];
+	const struct vpe_port_data *p_data = &port_data[port];
+	struct vb2_buffer *vb = ctx->src_vbs[p_data->vb_index];
+	struct v4l2_rect *c_rect = &q_data->c_rect;
+	struct vpe_fmt *fmt = q_data->fmt;
+	const struct vpdma_data_format *vpdma_fmt;
+	int mv_buf_selector = ctx->src_mv_buf_selector;
+	int field = vb->v4l2_buf.field == V4L2_FIELD_BOTTOM;
+	dma_addr_t dma_addr;
+	u32 flags = 0;
+
+	if (port == VPE_PORT_MV_IN) {
+		vpdma_fmt = &vpdma_misc_fmts[VPDMA_DATA_FMT_MV];
+		dma_addr = ctx->mv_buf_dma[mv_buf_selector];
+	} else {
+		/* to incorporate interleaved formats */
+		int plane = fmt->coplanar ? p_data->vb_part : 0;
+
+		vpdma_fmt = fmt->vpdma_fmt[plane];
+
+		dma_addr = vb2_dma_contig_plane_dma_addr(vb, plane);
+		if (!dma_addr) {
+			vpe_err(ctx->dev,
+				"acquiring input buffer(%d) dma_addr failed\n",
+				port);
+			return;
+		}
+	}
+
+	if (q_data->flags & Q_DATA_FRAME_1D)
+		flags |= VPDMA_DATA_FRAME_1D;
+	if (q_data->flags & Q_DATA_MODE_TILED)
+		flags |= VPDMA_DATA_MODE_TILED;
+
+	vpdma_add_in_dtd(&ctx->desc_list, q_data->width, q_data->height,
+		c_rect, vpdma_fmt, dma_addr, p_data->channel, field, flags);
+}
+
+/*
+ * Enable the expected IRQ sources
+ */
+static void enable_irqs(struct vpe_ctx *ctx)
+{
+	write_reg(ctx->dev, VPE_INT0_ENABLE0_SET, VPE_INT0_LIST0_COMPLETE);
+	write_reg(ctx->dev, VPE_INT0_ENABLE1_SET, VPE_DEI_ERROR_INT |
+				VPE_DS1_UV_ERROR_INT);
+
+	vpdma_enable_list_complete_irq(ctx->dev->vpdma, 0, true);
+}
+
+static void disable_irqs(struct vpe_ctx *ctx)
+{
+	write_reg(ctx->dev, VPE_INT0_ENABLE0_CLR, 0xffffffff);
+	write_reg(ctx->dev, VPE_INT0_ENABLE1_CLR, 0xffffffff);
+
+	vpdma_enable_list_complete_irq(ctx->dev->vpdma, 0, false);
+}
+
+/* device_run() - prepares and starts the device
+ *
+ * This function is only called when both the source and destination
+ * buffers are in place.
+ */
+static void device_run(void *priv)
+{
+	struct vpe_ctx *ctx = priv;
+	struct vpe_q_data *d_q_data = &ctx->q_data[Q_DATA_DST];
+
+	if (ctx->deinterlacing && ctx->src_vbs[2] == NULL) {
+		ctx->src_vbs[2] = v4l2_m2m_src_buf_remove(ctx->m2m_ctx);
+		WARN_ON(ctx->src_vbs[2] == NULL);
+		ctx->src_vbs[1] = v4l2_m2m_src_buf_remove(ctx->m2m_ctx);
+		WARN_ON(ctx->src_vbs[1] == NULL);
+	}
+
+	ctx->src_vbs[0] = v4l2_m2m_src_buf_remove(ctx->m2m_ctx);
+	WARN_ON(ctx->src_vbs[0] == NULL);
+	ctx->dst_vb = v4l2_m2m_dst_buf_remove(ctx->m2m_ctx);
+	WARN_ON(ctx->dst_vb == NULL);
+
+	/* config descriptors */
+	if (ctx->dev->loaded_mmrs != ctx->mmr_adb.dma_addr || ctx->load_mmrs) {
+		vpdma_map_desc_buf(ctx->dev->vpdma, &ctx->mmr_adb);
+		vpdma_add_cfd_adb(&ctx->desc_list, CFD_MMR_CLIENT, &ctx->mmr_adb);
+		ctx->dev->loaded_mmrs = ctx->mmr_adb.dma_addr;
+		ctx->load_mmrs = false;
+	}
+
+	/* output data descriptors */
+	if (ctx->deinterlacing)
+		add_out_dtd(ctx, VPE_PORT_MV_OUT);
+
+	add_out_dtd(ctx, VPE_PORT_LUMA_OUT);
+	if (d_q_data->fmt->coplanar)
+		add_out_dtd(ctx, VPE_PORT_CHROMA_OUT);
+
+	/* input data descriptors */
+	if (ctx->deinterlacing) {
+		add_in_dtd(ctx, VPE_PORT_LUMA3_IN);
+		add_in_dtd(ctx, VPE_PORT_CHROMA3_IN);
+
+		add_in_dtd(ctx, VPE_PORT_LUMA2_IN);
+		add_in_dtd(ctx, VPE_PORT_CHROMA2_IN);
+	}
+
+	add_in_dtd(ctx, VPE_PORT_LUMA1_IN);
+	add_in_dtd(ctx, VPE_PORT_CHROMA1_IN);
+
+	if (ctx->deinterlacing)
+		add_in_dtd(ctx, VPE_PORT_MV_IN);
+
+	/* sync on channel control descriptors for input ports */
+	vpdma_add_sync_on_channel_ctd(&ctx->desc_list, VPE_CHAN_LUMA1_IN);
+	vpdma_add_sync_on_channel_ctd(&ctx->desc_list, VPE_CHAN_CHROMA1_IN);
+
+	if (ctx->deinterlacing) {
+		vpdma_add_sync_on_channel_ctd(&ctx->desc_list,
+			VPE_CHAN_LUMA2_IN);
+		vpdma_add_sync_on_channel_ctd(&ctx->desc_list,
+			VPE_CHAN_CHROMA2_IN);
+
+		vpdma_add_sync_on_channel_ctd(&ctx->desc_list,
+			VPE_CHAN_LUMA3_IN);
+		vpdma_add_sync_on_channel_ctd(&ctx->desc_list,
+			VPE_CHAN_CHROMA3_IN);
+
+		vpdma_add_sync_on_channel_ctd(&ctx->desc_list, VPE_CHAN_MV_IN);
+	}
+
+	/* sync on channel control descriptors for output ports */
+	vpdma_add_sync_on_channel_ctd(&ctx->desc_list, VPE_CHAN_LUMA_OUT);
+	if (d_q_data->fmt->coplanar)
+		vpdma_add_sync_on_channel_ctd(&ctx->desc_list, VPE_CHAN_CHROMA_OUT);
+
+	if (ctx->deinterlacing)
+		vpdma_add_sync_on_channel_ctd(&ctx->desc_list, VPE_CHAN_MV_OUT);
+
+	enable_irqs(ctx);
+
+	vpdma_map_desc_buf(ctx->dev->vpdma, &ctx->desc_list.buf);
+	vpdma_submit_descs(ctx->dev->vpdma, &ctx->desc_list);
+}
+
+static void dei_error(struct vpe_ctx *ctx)
+{
+	dev_warn(ctx->dev->v4l2_dev.dev,
+		"received DEI error interrupt\n");
+}
+
+static void ds1_uv_error(struct vpe_ctx *ctx)
+{
+	dev_warn(ctx->dev->v4l2_dev.dev,
+		"received downsampler error interrupt\n");
+}
+
+static irqreturn_t vpe_irq(int irq_vpe, void *data)
+{
+	struct vpe_dev *dev = (struct vpe_dev *)data;
+	struct vpe_ctx *ctx;
+	struct vpe_q_data *d_q_data;
+	struct vb2_buffer *s_vb, *d_vb;
+	struct v4l2_buffer *s_buf, *d_buf;
+	unsigned long flags;
+	u32 irqst0, irqst1;
+
+	irqst0 = read_reg(dev, VPE_INT0_STATUS0);
+	if (irqst0) {
+		write_reg(dev, VPE_INT0_STATUS0_CLR, irqst0);
+		vpe_dbg(dev, "INT0_STATUS0 = 0x%08x\n", irqst0);
+	}
+
+	irqst1 = read_reg(dev, VPE_INT0_STATUS1);
+	if (irqst1) {
+		write_reg(dev, VPE_INT0_STATUS1_CLR, irqst1);
+		vpe_dbg(dev, "INT0_STATUS1 = 0x%08x\n", irqst1);
+	}
+
+	ctx = v4l2_m2m_get_curr_priv(dev->m2m_dev);
+	if (!ctx) {
+		vpe_err(dev, "instance released before end of transaction\n");
+		goto handled;
+	}
+
+	if (irqst1) {
+		if (irqst1 & VPE_DEI_ERROR_INT) {
+			irqst1 &= ~VPE_DEI_ERROR_INT;
+			dei_error(ctx);
+		}
+		if (irqst1 & VPE_DS1_UV_ERROR_INT) {
+			irqst1 &= ~VPE_DS1_UV_ERROR_INT;
+			ds1_uv_error(ctx);
+		}
+	}
+
+	if (irqst0) {
+		if (irqst0 & VPE_INT0_LIST0_COMPLETE)
+			vpdma_clear_list_stat(ctx->dev->vpdma);
+
+		irqst0 &= ~(VPE_INT0_LIST0_COMPLETE);
+	}
+
+	if (irqst0 | irqst1) {
+		dev_warn(dev->v4l2_dev.dev, "Unexpected interrupt: "
+			"INT0_STATUS0 = 0x%08x, INT0_STATUS1 = 0x%08x\n",
+			irqst0, irqst1);
+	}
+
+	disable_irqs(ctx);
+
+	vpdma_unmap_desc_buf(dev->vpdma, &ctx->desc_list.buf);
+	vpdma_unmap_desc_buf(dev->vpdma, &ctx->mmr_adb);
+
+	vpdma_reset_desc_list(&ctx->desc_list);
+
+	 /* the previous dst mv buffer becomes the next src mv buffer */
+	ctx->src_mv_buf_selector = !ctx->src_mv_buf_selector;
+
+	if (ctx->aborting)
+		goto finished;
+
+	s_vb = ctx->src_vbs[0];
+	d_vb = ctx->dst_vb;
+	s_buf = &s_vb->v4l2_buf;
+	d_buf = &d_vb->v4l2_buf;
+
+	d_buf->timestamp = s_buf->timestamp;
+	if (s_buf->flags & V4L2_BUF_FLAG_TIMECODE) {
+		d_buf->flags |= V4L2_BUF_FLAG_TIMECODE;
+		d_buf->timecode = s_buf->timecode;
+	}
+	d_buf->sequence = ctx->sequence;
+	d_buf->field = ctx->field;
+
+	d_q_data = &ctx->q_data[Q_DATA_DST];
+	if (d_q_data->flags & Q_DATA_INTERLACED) {
+		if (ctx->field == V4L2_FIELD_BOTTOM) {
+			ctx->sequence++;
+			ctx->field = V4L2_FIELD_TOP;
+		} else {
+			WARN_ON(ctx->field != V4L2_FIELD_TOP);
+			ctx->field = V4L2_FIELD_BOTTOM;
+		}
+	} else {
+		ctx->sequence++;
+	}
+
+	if (ctx->deinterlacing)
+		s_vb = ctx->src_vbs[2];
+
+	spin_lock_irqsave(&dev->lock, flags);
+	v4l2_m2m_buf_done(s_vb, VB2_BUF_STATE_DONE);
+	v4l2_m2m_buf_done(d_vb, VB2_BUF_STATE_DONE);
+	spin_unlock_irqrestore(&dev->lock, flags);
+
+	if (ctx->deinterlacing) {
+		ctx->src_vbs[2] = ctx->src_vbs[1];
+		ctx->src_vbs[1] = ctx->src_vbs[0];
+	}
+
+	ctx->bufs_completed++;
+	if (ctx->bufs_completed < ctx->bufs_per_job) {
+		device_run(ctx);
+		goto handled;
+	}
+
+finished:
+	vpe_dbg(ctx->dev, "finishing transaction\n");
+	ctx->bufs_completed = 0;
+	v4l2_m2m_job_finish(dev->m2m_dev, ctx->m2m_ctx);
+handled:
+	return IRQ_HANDLED;
+}
+
+/*
+ * video ioctls
+ */
+static int vpe_querycap(struct file *file, void *priv,
+			struct v4l2_capability *cap)
+{
+	strncpy(cap->driver, VPE_MODULE_NAME, sizeof(cap->driver) - 1);
+	strncpy(cap->card, VPE_MODULE_NAME, sizeof(cap->card) - 1);
+	strlcpy(cap->bus_info, VPE_MODULE_NAME, sizeof(cap->bus_info));
+	cap->device_caps  = V4L2_CAP_VIDEO_M2M | V4L2_CAP_STREAMING;
+	cap->capabilities = cap->device_caps | V4L2_CAP_DEVICE_CAPS;
+	return 0;
+}
+
+static int __enum_fmt(struct v4l2_fmtdesc *f, u32 type)
+{
+	int i, index;
+	struct vpe_fmt *fmt = NULL;
+
+	index = 0;
+	for (i = 0; i < ARRAY_SIZE(vpe_formats); ++i) {
+		if (vpe_formats[i].types & type) {
+			if (index == f->index) {
+				fmt = &vpe_formats[i];
+				break;
+			}
+			index++;
+		}
+	}
+
+	if (!fmt)
+		return -EINVAL;
+
+	strncpy(f->description, fmt->name, sizeof(f->description) - 1);
+	f->pixelformat = fmt->fourcc;
+	return 0;
+}
+
+static int vpe_enum_fmt(struct file *file, void *priv,
+				struct v4l2_fmtdesc *f)
+{
+	if (V4L2_TYPE_IS_OUTPUT(f->type))
+		return __enum_fmt(f, VPE_FMT_TYPE_OUTPUT);
+
+	return __enum_fmt(f, VPE_FMT_TYPE_CAPTURE);
+}
+
+static int vpe_g_fmt(struct file *file, void *priv, struct v4l2_format *f)
+{
+	struct v4l2_pix_format_mplane *pix = &f->fmt.pix_mp;
+	struct vpe_ctx *ctx = file2ctx(file);
+	struct vb2_queue *vq;
+	struct vpe_q_data *q_data;
+	int i;
+
+	vq = v4l2_m2m_get_vq(ctx->m2m_ctx, f->type);
+	if (!vq)
+		return -EINVAL;
+
+	q_data = get_q_data(ctx, f->type);
+
+	pix->width = q_data->width;
+	pix->height = q_data->height;
+	pix->pixelformat = q_data->fmt->fourcc;
+	pix->field = q_data->field;
+
+	if (V4L2_TYPE_IS_OUTPUT(f->type)) {
+		pix->colorspace = q_data->colorspace;
+	} else {
+		struct vpe_q_data *s_q_data;
+
+		/* get colorspace from the source queue */
+		s_q_data = get_q_data(ctx, V4L2_BUF_TYPE_VIDEO_OUTPUT_MPLANE);
+
+		pix->colorspace = s_q_data->colorspace;
+	}
+
+	pix->num_planes = q_data->fmt->coplanar ? 2 : 1;
+
+	for (i = 0; i < pix->num_planes; i++) {
+		pix->plane_fmt[i].bytesperline = q_data->bytesperline[i];
+		pix->plane_fmt[i].sizeimage = q_data->sizeimage[i];
+	}
+
+	return 0;
+}
+
+static int __vpe_try_fmt(struct vpe_ctx *ctx, struct v4l2_format *f,
+		       struct vpe_fmt *fmt, int type)
+{
+	struct v4l2_pix_format_mplane *pix = &f->fmt.pix_mp;
+	struct v4l2_plane_pix_format *plane_fmt;
+	int i;
+
+	if (!fmt || !(fmt->types & type)) {
+		vpe_err(ctx->dev, "Fourcc format (0x%08x) invalid.\n",
+			pix->pixelformat);
+		return -EINVAL;
+	}
+
+	if (pix->field != V4L2_FIELD_NONE && pix->field != V4L2_FIELD_ALTERNATE)
+		pix->field = V4L2_FIELD_NONE;
+
+	v4l_bound_align_image(&pix->width, MIN_W, MAX_W, W_ALIGN,
+			      &pix->height, MIN_H, MAX_H, H_ALIGN,
+			      S_ALIGN);
+
+	pix->num_planes = fmt->coplanar ? 2 : 1;
+	pix->pixelformat = fmt->fourcc;
+
+	if (type == VPE_FMT_TYPE_CAPTURE) {
+		struct vpe_q_data *s_q_data;
+
+		/* get colorspace from the source queue */
+		s_q_data = get_q_data(ctx, V4L2_BUF_TYPE_VIDEO_OUTPUT_MPLANE);
+
+		pix->colorspace = s_q_data->colorspace;
+	} else {
+		if (!pix->colorspace)
+			pix->colorspace = V4L2_COLORSPACE_SMPTE240M;
+	}
+
+	for (i = 0; i < pix->num_planes; i++) {
+		int depth;
+
+		plane_fmt = &pix->plane_fmt[i];
+		depth = fmt->vpdma_fmt[i]->depth;
+
+		if (i == VPE_LUMA)
+			plane_fmt->bytesperline =
+					round_up((pix->width * depth) >> 3,
+						1 << L_ALIGN);
+		else
+			plane_fmt->bytesperline = pix->width;
+
+		plane_fmt->sizeimage =
+				(pix->height * pix->width * depth) >> 3;
+	}
+
+	return 0;
+}
+
+static int vpe_try_fmt(struct file *file, void *priv, struct v4l2_format *f)
+{
+	struct vpe_ctx *ctx = file2ctx(file);
+	struct vpe_fmt *fmt = find_format(f);
+
+	if (V4L2_TYPE_IS_OUTPUT(f->type))
+		return __vpe_try_fmt(ctx, f, fmt, VPE_FMT_TYPE_OUTPUT);
+	else
+		return __vpe_try_fmt(ctx, f, fmt, VPE_FMT_TYPE_CAPTURE);
+}
+
+static int __vpe_s_fmt(struct vpe_ctx *ctx, struct v4l2_format *f)
+{
+	struct v4l2_pix_format_mplane *pix = &f->fmt.pix_mp;
+	struct v4l2_plane_pix_format *plane_fmt;
+	struct vpe_q_data *q_data;
+	struct vb2_queue *vq;
+	int i;
+
+	vq = v4l2_m2m_get_vq(ctx->m2m_ctx, f->type);
+	if (!vq)
+		return -EINVAL;
+
+	if (vb2_is_busy(vq)) {
+		vpe_err(ctx->dev, "queue busy\n");
+		return -EBUSY;
+	}
+
+	q_data = get_q_data(ctx, f->type);
+	if (!q_data)
+		return -EINVAL;
+
+	q_data->fmt		= find_format(f);
+	q_data->width		= pix->width;
+	q_data->height		= pix->height;
+	q_data->colorspace	= pix->colorspace;
+	q_data->field		= pix->field;
+
+	for (i = 0; i < pix->num_planes; i++) {
+		plane_fmt = &pix->plane_fmt[i];
+
+		q_data->bytesperline[i]	= plane_fmt->bytesperline;
+		q_data->sizeimage[i]	= plane_fmt->sizeimage;
+	}
+
+	q_data->c_rect.left	= 0;
+	q_data->c_rect.top	= 0;
+	q_data->c_rect.width	= q_data->width;
+	q_data->c_rect.height	= q_data->height;
+
+	if (q_data->field == V4L2_FIELD_ALTERNATE)
+		q_data->flags |= Q_DATA_INTERLACED;
+	else
+		q_data->flags &= ~Q_DATA_INTERLACED;
+
+	vpe_dbg(ctx->dev, "Setting format for type %d, wxh: %dx%d, fmt: %d bpl_y %d",
+		f->type, q_data->width, q_data->height, q_data->fmt->fourcc,
+		q_data->bytesperline[VPE_LUMA]);
+	if (q_data->fmt->coplanar)
+		vpe_dbg(ctx->dev, " bpl_uv %d\n",
+			q_data->bytesperline[VPE_CHROMA]);
+
+	return 0;
+}
+
+static int vpe_s_fmt(struct file *file, void *priv, struct v4l2_format *f)
+{
+	int ret;
+	struct vpe_ctx *ctx = file2ctx(file);
+
+	ret = vpe_try_fmt(file, priv, f);
+	if (ret)
+		return ret;
+
+	ret = __vpe_s_fmt(ctx, f);
+	if (ret)
+		return ret;
+
+	if (V4L2_TYPE_IS_OUTPUT(f->type))
+		set_src_registers(ctx);
+	else
+		set_dst_registers(ctx);
+
+	return set_srcdst_params(ctx);
+}
+
+static int vpe_reqbufs(struct file *file, void *priv,
+		       struct v4l2_requestbuffers *reqbufs)
+{
+	struct vpe_ctx *ctx = file2ctx(file);
+
+	return v4l2_m2m_reqbufs(file, ctx->m2m_ctx, reqbufs);
+}
+
+static int vpe_querybuf(struct file *file, void *priv, struct v4l2_buffer *buf)
+{
+	struct vpe_ctx *ctx = file2ctx(file);
+
+	return v4l2_m2m_querybuf(file, ctx->m2m_ctx, buf);
+}
+
+static int vpe_qbuf(struct file *file, void *priv, struct v4l2_buffer *buf)
+{
+	struct vpe_ctx *ctx = file2ctx(file);
+
+	return v4l2_m2m_qbuf(file, ctx->m2m_ctx, buf);
+}
+
+static int vpe_dqbuf(struct file *file, void *priv, struct v4l2_buffer *buf)
+{
+	struct vpe_ctx *ctx = file2ctx(file);
+
+	return v4l2_m2m_dqbuf(file, ctx->m2m_ctx, buf);
+}
+
+static int vpe_streamon(struct file *file, void *priv, enum v4l2_buf_type type)
+{
+	struct vpe_ctx *ctx = file2ctx(file);
+
+	return v4l2_m2m_streamon(file, ctx->m2m_ctx, type);
+}
+
+static int vpe_streamoff(struct file *file, void *priv, enum v4l2_buf_type type)
+{
+	struct vpe_ctx *ctx = file2ctx(file);
+
+	vpe_dump_regs(ctx->dev);
+	vpdma_dump_regs(ctx->dev->vpdma);
+
+	return v4l2_m2m_streamoff(file, ctx->m2m_ctx, type);
+}
+
+/*
+ * defines number of buffers/frames a context can process with VPE before
+ * switching to a different context. default value is 1 buffer per context
+ */
+#define V4L2_CID_VPE_BUFS_PER_JOB		(V4L2_CID_USER_TI_VPE_BASE + 0)
+
+static int vpe_s_ctrl(struct v4l2_ctrl *ctrl)
+{
+	struct vpe_ctx *ctx =
+		container_of(ctrl->handler, struct vpe_ctx, hdl);
+
+	switch (ctrl->id) {
+	case V4L2_CID_VPE_BUFS_PER_JOB:
+		ctx->bufs_per_job = ctrl->val;
+		break;
+
+	default:
+		vpe_err(ctx->dev, "Invalid control\n");
+		return -EINVAL;
+	}
+
+	return 0;
+}
+
+static const struct v4l2_ctrl_ops vpe_ctrl_ops = {
+	.s_ctrl = vpe_s_ctrl,
+};
+
+static const struct v4l2_ioctl_ops vpe_ioctl_ops = {
+	.vidioc_querycap	= vpe_querycap,
+
+	.vidioc_enum_fmt_vid_cap_mplane = vpe_enum_fmt,
+	.vidioc_g_fmt_vid_cap_mplane	= vpe_g_fmt,
+	.vidioc_try_fmt_vid_cap_mplane	= vpe_try_fmt,
+	.vidioc_s_fmt_vid_cap_mplane	= vpe_s_fmt,
+
+	.vidioc_enum_fmt_vid_out_mplane = vpe_enum_fmt,
+	.vidioc_g_fmt_vid_out_mplane	= vpe_g_fmt,
+	.vidioc_try_fmt_vid_out_mplane	= vpe_try_fmt,
+	.vidioc_s_fmt_vid_out_mplane	= vpe_s_fmt,
+
+	.vidioc_reqbufs		= vpe_reqbufs,
+	.vidioc_querybuf	= vpe_querybuf,
+
+	.vidioc_qbuf		= vpe_qbuf,
+	.vidioc_dqbuf		= vpe_dqbuf,
+
+	.vidioc_streamon	= vpe_streamon,
+	.vidioc_streamoff	= vpe_streamoff,
+	.vidioc_subscribe_event = v4l2_ctrl_subscribe_event,
+	.vidioc_unsubscribe_event = v4l2_event_unsubscribe,
+};
+
+/*
+ * Queue operations
+ */
+static int vpe_queue_setup(struct vb2_queue *vq,
+			   const struct v4l2_format *fmt,
+			   unsigned int *nbuffers, unsigned int *nplanes,
+			   unsigned int sizes[], void *alloc_ctxs[])
+{
+	int i;
+	struct vpe_ctx *ctx = vb2_get_drv_priv(vq);
+	struct vpe_q_data *q_data;
+
+	q_data = get_q_data(ctx, vq->type);
+
+	*nplanes = q_data->fmt->coplanar ? 2 : 1;
+
+	for (i = 0; i < *nplanes; i++) {
+		sizes[i] = q_data->sizeimage[i];
+		alloc_ctxs[i] = ctx->dev->alloc_ctx;
+	}
+
+	vpe_dbg(ctx->dev, "get %d buffer(s) of size %d", *nbuffers,
+		sizes[VPE_LUMA]);
+	if (q_data->fmt->coplanar)
+		vpe_dbg(ctx->dev, " and %d\n", sizes[VPE_CHROMA]);
+
+	return 0;
+}
+
+static int vpe_buf_prepare(struct vb2_buffer *vb)
+{
+	struct vpe_ctx *ctx = vb2_get_drv_priv(vb->vb2_queue);
+	struct vpe_q_data *q_data;
+	int i, num_planes;
+
+	vpe_dbg(ctx->dev, "type: %d\n", vb->vb2_queue->type);
+
+	q_data = get_q_data(ctx, vb->vb2_queue->type);
+	num_planes = q_data->fmt->coplanar ? 2 : 1;
+
+	for (i = 0; i < num_planes; i++) {
+		if (vb2_plane_size(vb, i) < q_data->sizeimage[i]) {
+			vpe_err(ctx->dev,
+				"data will not fit into plane (%lu < %lu)\n",
+				vb2_plane_size(vb, i),
+				(long) q_data->sizeimage[i]);
+			return -EINVAL;
+		}
+	}
+
+	for (i = 0; i < num_planes; i++)
+		vb2_set_plane_payload(vb, i, q_data->sizeimage[i]);
+
+	return 0;
+}
+
+static void vpe_buf_queue(struct vb2_buffer *vb)
+{
+	struct vpe_ctx *ctx = vb2_get_drv_priv(vb->vb2_queue);
+	v4l2_m2m_buf_queue(ctx->m2m_ctx, vb);
+}
+
+static void vpe_wait_prepare(struct vb2_queue *q)
+{
+	struct vpe_ctx *ctx = vb2_get_drv_priv(q);
+	vpe_unlock(ctx);
+}
+
+static void vpe_wait_finish(struct vb2_queue *q)
+{
+	struct vpe_ctx *ctx = vb2_get_drv_priv(q);
+	vpe_lock(ctx);
+}
+
+static struct vb2_ops vpe_qops = {
+	.queue_setup	 = vpe_queue_setup,
+	.buf_prepare	 = vpe_buf_prepare,
+	.buf_queue	 = vpe_buf_queue,
+	.wait_prepare	 = vpe_wait_prepare,
+	.wait_finish	 = vpe_wait_finish,
+};
+
+static int queue_init(void *priv, struct vb2_queue *src_vq,
+		      struct vb2_queue *dst_vq)
+{
+	struct vpe_ctx *ctx = priv;
+	int ret;
+
+	memset(src_vq, 0, sizeof(*src_vq));
+	src_vq->type = V4L2_BUF_TYPE_VIDEO_OUTPUT_MPLANE;
+	src_vq->io_modes = VB2_MMAP;
+	src_vq->drv_priv = ctx;
+	src_vq->buf_struct_size = sizeof(struct v4l2_m2m_buffer);
+	src_vq->ops = &vpe_qops;
+	src_vq->mem_ops = &vb2_dma_contig_memops;
+	src_vq->timestamp_type = V4L2_BUF_FLAG_TIMESTAMP_COPY;
+
+	ret = vb2_queue_init(src_vq);
+	if (ret)
+		return ret;
+
+	memset(dst_vq, 0, sizeof(*dst_vq));
+	dst_vq->type = V4L2_BUF_TYPE_VIDEO_CAPTURE_MPLANE;
+	dst_vq->io_modes = VB2_MMAP;
+	dst_vq->drv_priv = ctx;
+	dst_vq->buf_struct_size = sizeof(struct v4l2_m2m_buffer);
+	dst_vq->ops = &vpe_qops;
+	dst_vq->mem_ops = &vb2_dma_contig_memops;
+	dst_vq->timestamp_type = V4L2_BUF_FLAG_TIMESTAMP_COPY;
+
+	return vb2_queue_init(dst_vq);
+}
+
+static const struct v4l2_ctrl_config vpe_bufs_per_job = {
+	.ops = &vpe_ctrl_ops,
+	.id = V4L2_CID_VPE_BUFS_PER_JOB,
+	.name = "Buffers Per Transaction",
+	.type = V4L2_CTRL_TYPE_INTEGER,
+	.def = VPE_DEF_BUFS_PER_JOB,
+	.min = 1,
+	.max = VIDEO_MAX_FRAME,
+	.step = 1,
+};
+
+/*
+ * File operations
+ */
+static int vpe_open(struct file *file)
+{
+	struct vpe_dev *dev = video_drvdata(file);
+	struct vpe_ctx *ctx = NULL;
+	struct vpe_q_data *s_q_data;
+	struct v4l2_ctrl_handler *hdl;
+	int ret;
+
+	vpe_dbg(dev, "vpe_open\n");
+
+	if (!dev->vpdma->ready) {
+		vpe_err(dev, "vpdma firmware not loaded\n");
+		return -ENODEV;
+	}
+
+	ctx = kzalloc(sizeof(*ctx), GFP_KERNEL);
+	if (!ctx)
+		return -ENOMEM;
+
+	ctx->dev = dev;
+
+	if (mutex_lock_interruptible(&dev->dev_mutex)) {
+		ret = -ERESTARTSYS;
+		goto free_ctx;
+	}
+
+	ret = vpdma_create_desc_list(&ctx->desc_list, VPE_DESC_LIST_SIZE,
+			VPDMA_LIST_TYPE_NORMAL);
+	if (ret != 0)
+		goto unlock;
+
+	ret = vpdma_alloc_desc_buf(&ctx->mmr_adb, sizeof(struct vpe_mmr_adb));
+	if (ret != 0)
+		goto free_desc_list;
+
+	init_adb_hdrs(ctx);
+
+	v4l2_fh_init(&ctx->fh, video_devdata(file));
+	file->private_data = &ctx->fh;
+
+	hdl = &ctx->hdl;
+	v4l2_ctrl_handler_init(hdl, 1);
+	v4l2_ctrl_new_custom(hdl, &vpe_bufs_per_job, NULL);
+	if (hdl->error) {
+		ret = hdl->error;
+		goto exit_fh;
+	}
+	ctx->fh.ctrl_handler = hdl;
+	v4l2_ctrl_handler_setup(hdl);
+
+	s_q_data = &ctx->q_data[Q_DATA_SRC];
+	s_q_data->fmt = &vpe_formats[2];
+	s_q_data->width = 1920;
+	s_q_data->height = 1080;
+	s_q_data->sizeimage[VPE_LUMA] = (s_q_data->width * s_q_data->height *
+			s_q_data->fmt->vpdma_fmt[VPE_LUMA]->depth) >> 3;
+	s_q_data->colorspace = V4L2_COLORSPACE_SMPTE240M;
+	s_q_data->field = V4L2_FIELD_NONE;
+	s_q_data->c_rect.left = 0;
+	s_q_data->c_rect.top = 0;
+	s_q_data->c_rect.width = s_q_data->width;
+	s_q_data->c_rect.height = s_q_data->height;
+	s_q_data->flags = 0;
+
+	ctx->q_data[Q_DATA_DST] = *s_q_data;
+
+	set_dei_shadow_registers(ctx);
+	set_src_registers(ctx);
+	set_dst_registers(ctx);
+	ret = set_srcdst_params(ctx);
+	if (ret)
+		goto exit_fh;
+
+	ctx->m2m_ctx = v4l2_m2m_ctx_init(dev->m2m_dev, ctx, &queue_init);
+
+	if (IS_ERR(ctx->m2m_ctx)) {
+		ret = PTR_ERR(ctx->m2m_ctx);
+		goto exit_fh;
+	}
+
+	v4l2_fh_add(&ctx->fh);
+
+	/*
+	 * for now, just report the creation of the first instance, we can later
+	 * optimize the driver to enable or disable clocks when the first
+	 * instance is created or the last instance released
+	 */
+	if (atomic_inc_return(&dev->num_instances) == 1)
+		vpe_dbg(dev, "first instance created\n");
+
+	ctx->bufs_per_job = VPE_DEF_BUFS_PER_JOB;
+
+	ctx->load_mmrs = true;
+
+	vpe_dbg(dev, "created instance %p, m2m_ctx: %p\n",
+		ctx, ctx->m2m_ctx);
+
+	mutex_unlock(&dev->dev_mutex);
+
+	return 0;
+exit_fh:
+	v4l2_ctrl_handler_free(hdl);
+	v4l2_fh_exit(&ctx->fh);
+	vpdma_free_desc_buf(&ctx->mmr_adb);
+free_desc_list:
+	vpdma_free_desc_list(&ctx->desc_list);
+unlock:
+	mutex_unlock(&dev->dev_mutex);
+free_ctx:
+	kfree(ctx);
+	return ret;
+}
+
+static int vpe_release(struct file *file)
+{
+	struct vpe_dev *dev = video_drvdata(file);
+	struct vpe_ctx *ctx = file2ctx(file);
+
+	vpe_dbg(dev, "releasing instance %p\n", ctx);
+
+	mutex_lock(&dev->dev_mutex);
+	free_vbs(ctx);
+	free_mv_buffers(ctx);
+	vpdma_free_desc_list(&ctx->desc_list);
+	vpdma_free_desc_buf(&ctx->mmr_adb);
+
+	v4l2_fh_del(&ctx->fh);
+	v4l2_fh_exit(&ctx->fh);
+	v4l2_ctrl_handler_free(&ctx->hdl);
+	v4l2_m2m_ctx_release(ctx->m2m_ctx);
+
+	kfree(ctx);
+
+	/*
+	 * for now, just report the release of the last instance, we can later
+	 * optimize the driver to enable or disable clocks when the first
+	 * instance is created or the last instance released
+	 */
+	if (atomic_dec_return(&dev->num_instances) == 0)
+		vpe_dbg(dev, "last instance released\n");
+
+	mutex_unlock(&dev->dev_mutex);
+
+	return 0;
+}
+
+static unsigned int vpe_poll(struct file *file,
+			     struct poll_table_struct *wait)
+{
+	struct vpe_ctx *ctx = file2ctx(file);
+	struct vpe_dev *dev = ctx->dev;
+	int ret;
+
+	mutex_lock(&dev->dev_mutex);
+	ret = v4l2_m2m_poll(file, ctx->m2m_ctx, wait);
+	mutex_unlock(&dev->dev_mutex);
+	return ret;
+}
+
+static int vpe_mmap(struct file *file, struct vm_area_struct *vma)
+{
+	struct vpe_ctx *ctx = file2ctx(file);
+	struct vpe_dev *dev = ctx->dev;
+	int ret;
+
+	if (mutex_lock_interruptible(&dev->dev_mutex))
+		return -ERESTARTSYS;
+	ret = v4l2_m2m_mmap(file, ctx->m2m_ctx, vma);
+	mutex_unlock(&dev->dev_mutex);
+	return ret;
+}
+
+static const struct v4l2_file_operations vpe_fops = {
+	.owner		= THIS_MODULE,
+	.open		= vpe_open,
+	.release	= vpe_release,
+	.poll		= vpe_poll,
+	.unlocked_ioctl	= video_ioctl2,
+	.mmap		= vpe_mmap,
+};
+
+static struct video_device vpe_videodev = {
+	.name		= VPE_MODULE_NAME,
+	.fops		= &vpe_fops,
+	.ioctl_ops	= &vpe_ioctl_ops,
+	.minor		= -1,
+	.release	= video_device_release,
+	.vfl_dir	= VFL_DIR_M2M,
+};
+
+static struct v4l2_m2m_ops m2m_ops = {
+	.device_run	= device_run,
+	.job_ready	= job_ready,
+	.job_abort	= job_abort,
+	.lock		= vpe_lock,
+	.unlock		= vpe_unlock,
+};
+
+static int vpe_runtime_get(struct platform_device *pdev)
+{
+	int r;
+
+	dev_dbg(&pdev->dev, "vpe_runtime_get\n");
+
+	r = pm_runtime_get_sync(&pdev->dev);
+	WARN_ON(r < 0);
+	return r < 0 ? r : 0;
+}
+
+static void vpe_runtime_put(struct platform_device *pdev)
+{
+
+	int r;
+
+	dev_dbg(&pdev->dev, "vpe_runtime_put\n");
+
+	r = pm_runtime_put_sync(&pdev->dev);
+	WARN_ON(r < 0 && r != -ENOSYS);
+}
+
+static int vpe_probe(struct platform_device *pdev)
+{
+	struct vpe_dev *dev;
+	struct video_device *vfd;
+	struct resource *res;
+	int ret, irq, func;
+
+	dev = devm_kzalloc(&pdev->dev, sizeof(*dev), GFP_KERNEL);
+	if (IS_ERR(dev))
+		return PTR_ERR(dev);
+
+	spin_lock_init(&dev->lock);
+
+	ret = v4l2_device_register(&pdev->dev, &dev->v4l2_dev);
+	if (ret)
+		return ret;
+
+	atomic_set(&dev->num_instances, 0);
+	mutex_init(&dev->dev_mutex);
+
+	res = platform_get_resource_byname(pdev, IORESOURCE_MEM, "vpe_top");
+	/*
+	 * HACK: we get resource info from device tree in the form of a list of
+	 * VPE sub blocks, the driver currently uses only the base of vpe_top
+	 * for register access, the driver should be changed later to access
+	 * registers based on the sub block base addresses
+	 */
+	dev->base = devm_ioremap(&pdev->dev, res->start, SZ_32K);
+	if (IS_ERR(dev->base)) {
+		ret = PTR_ERR(dev->base);
+		goto v4l2_dev_unreg;
+	}
+
+	irq = platform_get_irq(pdev, 0);
+	ret = devm_request_irq(&pdev->dev, irq, vpe_irq, 0, VPE_MODULE_NAME,
+			dev);
+	if (ret)
+		goto v4l2_dev_unreg;
+
+	platform_set_drvdata(pdev, dev);
+
+	dev->alloc_ctx = vb2_dma_contig_init_ctx(&pdev->dev);
+	if (IS_ERR(dev->alloc_ctx)) {
+		vpe_err(dev, "Failed to alloc vb2 context\n");
+		ret = PTR_ERR(dev->alloc_ctx);
+		goto v4l2_dev_unreg;
+	}
+
+	dev->m2m_dev = v4l2_m2m_init(&m2m_ops);
+	if (IS_ERR(dev->m2m_dev)) {
+		vpe_err(dev, "Failed to init mem2mem device\n");
+		ret = PTR_ERR(dev->m2m_dev);
+		goto rel_ctx;
+	}
+
+	pm_runtime_enable(&pdev->dev);
+
+	ret = vpe_runtime_get(pdev);
+	if (ret)
+		goto rel_m2m;
+
+	/* Perform clk enable followed by reset */
+	vpe_set_clock_enable(dev, 1);
+
+	vpe_top_reset(dev);
+
+	func = read_field_reg(dev, VPE_PID, VPE_PID_FUNC_MASK,
+		VPE_PID_FUNC_SHIFT);
+	vpe_dbg(dev, "VPE PID function %x\n", func);
+
+	vpe_top_vpdma_reset(dev);
+
+	dev->vpdma = vpdma_create(pdev);
+	if (IS_ERR(dev->vpdma))
+		goto runtime_put;
+
+	vfd = &dev->vfd;
+	*vfd = vpe_videodev;
+	vfd->lock = &dev->dev_mutex;
+	vfd->v4l2_dev = &dev->v4l2_dev;
+
+	ret = video_register_device(vfd, VFL_TYPE_GRABBER, 0);
+	if (ret) {
+		vpe_err(dev, "Failed to register video device\n");
+		goto runtime_put;
+	}
+
+	video_set_drvdata(vfd, dev);
+	snprintf(vfd->name, sizeof(vfd->name), "%s", vpe_videodev.name);
+	dev_info(dev->v4l2_dev.dev, "Device registered as /dev/video%d\n",
+		vfd->num);
+
+	return 0;
+
+runtime_put:
+	vpe_runtime_put(pdev);
+rel_m2m:
+	pm_runtime_disable(&pdev->dev);
+	v4l2_m2m_release(dev->m2m_dev);
+rel_ctx:
+	vb2_dma_contig_cleanup_ctx(dev->alloc_ctx);
+v4l2_dev_unreg:
+	v4l2_device_unregister(&dev->v4l2_dev);
+
+	return ret;
+}
+
+static int vpe_remove(struct platform_device *pdev)
+{
+	struct vpe_dev *dev =
+		(struct vpe_dev *) platform_get_drvdata(pdev);
+
+	v4l2_info(&dev->v4l2_dev, "Removing " VPE_MODULE_NAME);
+
+	v4l2_m2m_release(dev->m2m_dev);
+	video_unregister_device(&dev->vfd);
+	v4l2_device_unregister(&dev->v4l2_dev);
+	vb2_dma_contig_cleanup_ctx(dev->alloc_ctx);
+
+	vpe_set_clock_enable(dev, 0);
+	vpe_runtime_put(pdev);
+	pm_runtime_disable(&pdev->dev);
+
+	return 0;
+}
+
+#if defined(CONFIG_OF)
+static const struct of_device_id vpe_of_match[] = {
+	{
+		.compatible = "ti,vpe",
+	},
+	{},
+};
+#else
+#define vpe_of_match NULL
+#endif
+
+static struct platform_driver vpe_pdrv = {
+	.probe		= vpe_probe,
+	.remove		= vpe_remove,
+	.driver		= {
+		.name	= VPE_MODULE_NAME,
+		.owner	= THIS_MODULE,
+		.of_match_table = vpe_of_match,
+	},
+};
+
+static void __exit vpe_exit(void)
+{
+	platform_driver_unregister(&vpe_pdrv);
+}
+
+static int __init vpe_init(void)
+{
+	return platform_driver_register(&vpe_pdrv);
+}
+
+module_init(vpe_init);
+module_exit(vpe_exit);
+
+MODULE_DESCRIPTION("TI VPE driver");
+MODULE_AUTHOR("Dale Farnsworth, <dale@farnsworth.org>");
+MODULE_LICENSE("GPL");
diff --git a/drivers/media/platform/ti-vpe/vpe_regs.h b/drivers/media/platform/ti-vpe/vpe_regs.h
new file mode 100644
index 000000000000..ed214e828398
--- /dev/null
+++ b/drivers/media/platform/ti-vpe/vpe_regs.h
@@ -0,0 +1,496 @@
+/*
+ * Copyright (c) 2013 Texas Instruments Inc.
+ *
+ * David Griego, <dagriego@biglakesoftware.com>
+ * Dale Farnsworth, <dale@farnsworth.org>
+ * Archit Taneja, <archit@ti.com>
+ *
+ * This program is free software; you can redistribute it and/or modify it
+ * under the terms of the GNU General Public License version 2 as published by
+ * the Free Software Foundation.
+ */
+
+#ifndef __TI_VPE_REGS_H
+#define __TI_VPE_REGS_H
+
+/* VPE register offsets and field selectors */
+
+/* VPE top level regs */
+#define VPE_PID				0x0000
+#define VPE_PID_MINOR_MASK		0x3f
+#define VPE_PID_MINOR_SHIFT		0
+#define VPE_PID_CUSTOM_MASK		0x03
+#define VPE_PID_CUSTOM_SHIFT		6
+#define VPE_PID_MAJOR_MASK		0x07
+#define VPE_PID_MAJOR_SHIFT		8
+#define VPE_PID_RTL_MASK		0x1f
+#define VPE_PID_RTL_SHIFT		11
+#define VPE_PID_FUNC_MASK		0xfff
+#define VPE_PID_FUNC_SHIFT		16
+#define VPE_PID_SCHEME_MASK		0x03
+#define VPE_PID_SCHEME_SHIFT		30
+
+#define VPE_SYSCONFIG			0x0010
+#define VPE_SYSCONFIG_IDLE_MASK		0x03
+#define VPE_SYSCONFIG_IDLE_SHIFT	2
+#define VPE_SYSCONFIG_STANDBY_MASK	0x03
+#define VPE_SYSCONFIG_STANDBY_SHIFT	4
+#define VPE_FORCE_IDLE_MODE		0
+#define VPE_NO_IDLE_MODE		1
+#define VPE_SMART_IDLE_MODE		2
+#define VPE_SMART_IDLE_WAKEUP_MODE	3
+#define VPE_FORCE_STANDBY_MODE		0
+#define VPE_NO_STANDBY_MODE		1
+#define VPE_SMART_STANDBY_MODE		2
+#define VPE_SMART_STANDBY_WAKEUP_MODE	3
+
+#define VPE_INT0_STATUS0_RAW_SET	0x0020
+#define VPE_INT0_STATUS0_RAW		VPE_INT0_STATUS0_RAW_SET
+#define VPE_INT0_STATUS0_CLR		0x0028
+#define VPE_INT0_STATUS0		VPE_INT0_STATUS0_CLR
+#define VPE_INT0_ENABLE0_SET		0x0030
+#define VPE_INT0_ENABLE0		VPE_INT0_ENABLE0_SET
+#define VPE_INT0_ENABLE0_CLR		0x0038
+#define VPE_INT0_LIST0_COMPLETE		(1 << 0)
+#define VPE_INT0_LIST0_NOTIFY		(1 << 1)
+#define VPE_INT0_LIST1_COMPLETE		(1 << 2)
+#define VPE_INT0_LIST1_NOTIFY		(1 << 3)
+#define VPE_INT0_LIST2_COMPLETE		(1 << 4)
+#define VPE_INT0_LIST2_NOTIFY		(1 << 5)
+#define VPE_INT0_LIST3_COMPLETE		(1 << 6)
+#define VPE_INT0_LIST3_NOTIFY		(1 << 7)
+#define VPE_INT0_LIST4_COMPLETE		(1 << 8)
+#define VPE_INT0_LIST4_NOTIFY		(1 << 9)
+#define VPE_INT0_LIST5_COMPLETE		(1 << 10)
+#define VPE_INT0_LIST5_NOTIFY		(1 << 11)
+#define VPE_INT0_LIST6_COMPLETE		(1 << 12)
+#define VPE_INT0_LIST6_NOTIFY		(1 << 13)
+#define VPE_INT0_LIST7_COMPLETE		(1 << 14)
+#define VPE_INT0_LIST7_NOTIFY		(1 << 15)
+#define VPE_INT0_DESCRIPTOR		(1 << 16)
+#define VPE_DEI_FMD_INT			(1 << 18)
+
+#define VPE_INT0_STATUS1_RAW_SET	0x0024
+#define VPE_INT0_STATUS1_RAW		VPE_INT0_STATUS1_RAW_SET
+#define VPE_INT0_STATUS1_CLR		0x002c
+#define VPE_INT0_STATUS1		VPE_INT0_STATUS1_CLR
+#define VPE_INT0_ENABLE1_SET		0x0034
+#define VPE_INT0_ENABLE1		VPE_INT0_ENABLE1_SET
+#define VPE_INT0_ENABLE1_CLR		0x003c
+#define VPE_INT0_CHANNEL_GROUP0		(1 << 0)
+#define VPE_INT0_CHANNEL_GROUP1		(1 << 1)
+#define VPE_INT0_CHANNEL_GROUP2		(1 << 2)
+#define VPE_INT0_CHANNEL_GROUP3		(1 << 3)
+#define VPE_INT0_CHANNEL_GROUP4		(1 << 4)
+#define VPE_INT0_CHANNEL_GROUP5		(1 << 5)
+#define VPE_INT0_CLIENT			(1 << 7)
+#define VPE_DEI_ERROR_INT		(1 << 16)
+#define VPE_DS1_UV_ERROR_INT		(1 << 22)
+
+#define VPE_INTC_EOI			0x00a0
+
+#define VPE_CLK_ENABLE			0x0100
+#define VPE_VPEDMA_CLK_ENABLE		(1 << 0)
+#define VPE_DATA_PATH_CLK_ENABLE	(1 << 1)
+
+#define VPE_CLK_RESET			0x0104
+#define VPE_VPDMA_CLK_RESET_MASK	0x1
+#define VPE_VPDMA_CLK_RESET_SHIFT	0
+#define VPE_DATA_PATH_CLK_RESET_MASK	0x1
+#define VPE_DATA_PATH_CLK_RESET_SHIFT	1
+#define VPE_MAIN_RESET_MASK		0x1
+#define VPE_MAIN_RESET_SHIFT		31
+
+#define VPE_CLK_FORMAT_SELECT		0x010c
+#define VPE_CSC_SRC_SELECT_MASK		0x03
+#define VPE_CSC_SRC_SELECT_SHIFT	0
+#define VPE_RGB_OUT_SELECT		(1 << 8)
+#define VPE_DS_SRC_SELECT_MASK		0x07
+#define VPE_DS_SRC_SELECT_SHIFT		9
+#define VPE_DS_BYPASS			(1 << 16)
+#define VPE_COLOR_SEPARATE_422		(1 << 18)
+
+#define VPE_DS_SRC_DEI_SCALER		(5 << VPE_DS_SRC_SELECT_SHIFT)
+#define VPE_CSC_SRC_DEI_SCALER		(3 << VPE_CSC_SRC_SELECT_SHIFT)
+
+#define VPE_CLK_RANGE_MAP		0x011c
+#define VPE_RANGE_RANGE_MAP_Y_MASK	0x07
+#define VPE_RANGE_RANGE_MAP_Y_SHIFT	0
+#define VPE_RANGE_RANGE_MAP_UV_MASK	0x07
+#define VPE_RANGE_RANGE_MAP_UV_SHIFT	3
+#define VPE_RANGE_MAP_ON		(1 << 6)
+#define VPE_RANGE_REDUCTION_ON		(1 << 28)
+
+/* VPE chrominance upsampler regs */
+#define VPE_US1_R0			0x0304
+#define VPE_US2_R0			0x0404
+#define VPE_US3_R0			0x0504
+#define VPE_US_C1_MASK			0x3fff
+#define VPE_US_C1_SHIFT			2
+#define VPE_US_C0_MASK			0x3fff
+#define VPE_US_C0_SHIFT			18
+#define VPE_US_MODE_MASK		0x03
+#define VPE_US_MODE_SHIFT		16
+#define VPE_ANCHOR_FID0_C1_MASK		0x3fff
+#define VPE_ANCHOR_FID0_C1_SHIFT	2
+#define VPE_ANCHOR_FID0_C0_MASK		0x3fff
+#define VPE_ANCHOR_FID0_C0_SHIFT	18
+
+#define VPE_US1_R1			0x0308
+#define VPE_US2_R1			0x0408
+#define VPE_US3_R1			0x0508
+#define VPE_ANCHOR_FID0_C3_MASK		0x3fff
+#define VPE_ANCHOR_FID0_C3_SHIFT	2
+#define VPE_ANCHOR_FID0_C2_MASK		0x3fff
+#define VPE_ANCHOR_FID0_C2_SHIFT	18
+
+#define VPE_US1_R2			0x030c
+#define VPE_US2_R2			0x040c
+#define VPE_US3_R2			0x050c
+#define VPE_INTERP_FID0_C1_MASK		0x3fff
+#define VPE_INTERP_FID0_C1_SHIFT	2
+#define VPE_INTERP_FID0_C0_MASK		0x3fff
+#define VPE_INTERP_FID0_C0_SHIFT	18
+
+#define VPE_US1_R3			0x0310
+#define VPE_US2_R3			0x0410
+#define VPE_US3_R3			0x0510
+#define VPE_INTERP_FID0_C3_MASK		0x3fff
+#define VPE_INTERP_FID0_C3_SHIFT	2
+#define VPE_INTERP_FID0_C2_MASK		0x3fff
+#define VPE_INTERP_FID0_C2_SHIFT	18
+
+#define VPE_US1_R4			0x0314
+#define VPE_US2_R4			0x0414
+#define VPE_US3_R4			0x0514
+#define VPE_ANCHOR_FID1_C1_MASK		0x3fff
+#define VPE_ANCHOR_FID1_C1_SHIFT	2
+#define VPE_ANCHOR_FID1_C0_MASK		0x3fff
+#define VPE_ANCHOR_FID1_C0_SHIFT	18
+
+#define VPE_US1_R5			0x0318
+#define VPE_US2_R5			0x0418
+#define VPE_US3_R5			0x0518
+#define VPE_ANCHOR_FID1_C3_MASK		0x3fff
+#define VPE_ANCHOR_FID1_C3_SHIFT	2
+#define VPE_ANCHOR_FID1_C2_MASK		0x3fff
+#define VPE_ANCHOR_FID1_C2_SHIFT	18
+
+#define VPE_US1_R6			0x031c
+#define VPE_US2_R6			0x041c
+#define VPE_US3_R6			0x051c
+#define VPE_INTERP_FID1_C1_MASK		0x3fff
+#define VPE_INTERP_FID1_C1_SHIFT	2
+#define VPE_INTERP_FID1_C0_MASK		0x3fff
+#define VPE_INTERP_FID1_C0_SHIFT	18
+
+#define VPE_US1_R7			0x0320
+#define VPE_US2_R7			0x0420
+#define VPE_US3_R7			0x0520
+#define VPE_INTERP_FID0_C3_MASK		0x3fff
+#define VPE_INTERP_FID0_C3_SHIFT	2
+#define VPE_INTERP_FID0_C2_MASK		0x3fff
+#define VPE_INTERP_FID0_C2_SHIFT	18
+
+/* VPE de-interlacer regs */
+#define VPE_DEI_FRAME_SIZE		0x0600
+#define VPE_DEI_WIDTH_MASK		0x07ff
+#define VPE_DEI_WIDTH_SHIFT		0
+#define VPE_DEI_HEIGHT_MASK		0x07ff
+#define VPE_DEI_HEIGHT_SHIFT		16
+#define VPE_DEI_INTERLACE_BYPASS	(1 << 29)
+#define VPE_DEI_FIELD_FLUSH		(1 << 30)
+#define VPE_DEI_PROGRESSIVE		(1 << 31)
+
+#define VPE_MDT_BYPASS			0x0604
+#define VPE_MDT_TEMPMAX_BYPASS		(1 << 0)
+#define VPE_MDT_SPATMAX_BYPASS		(1 << 1)
+
+#define VPE_MDT_SF_THRESHOLD		0x0608
+#define VPE_MDT_SF_SC_THR1_MASK		0xff
+#define VPE_MDT_SF_SC_THR1_SHIFT	0
+#define VPE_MDT_SF_SC_THR2_MASK		0xff
+#define VPE_MDT_SF_SC_THR2_SHIFT	0
+#define VPE_MDT_SF_SC_THR3_MASK		0xff
+#define VPE_MDT_SF_SC_THR3_SHIFT	0
+
+#define VPE_EDI_CONFIG			0x060c
+#define VPE_EDI_INP_MODE_MASK		0x03
+#define VPE_EDI_INP_MODE_SHIFT		0
+#define VPE_EDI_ENABLE_3D		(1 << 2)
+#define VPE_EDI_ENABLE_CHROMA_3D	(1 << 3)
+#define VPE_EDI_CHROMA3D_COR_THR_MASK	0xff
+#define VPE_EDI_CHROMA3D_COR_THR_SHIFT	8
+#define VPE_EDI_DIR_COR_LOWER_THR_MASK	0xff
+#define VPE_EDI_DIR_COR_LOWER_THR_SHIFT	16
+#define VPE_EDI_COR_SCALE_FACTOR_MASK	0xff
+#define VPE_EDI_COR_SCALE_FACTOR_SHIFT	23
+
+#define VPE_DEI_EDI_LUT_R0		0x0610
+#define VPE_EDI_LUT0_MASK		0x1f
+#define VPE_EDI_LUT0_SHIFT		0
+#define VPE_EDI_LUT1_MASK		0x1f
+#define VPE_EDI_LUT1_SHIFT		8
+#define VPE_EDI_LUT2_MASK		0x1f
+#define VPE_EDI_LUT2_SHIFT		16
+#define VPE_EDI_LUT3_MASK		0x1f
+#define VPE_EDI_LUT3_SHIFT		24
+
+#define VPE_DEI_EDI_LUT_R1		0x0614
+#define VPE_EDI_LUT0_MASK		0x1f
+#define VPE_EDI_LUT0_SHIFT		0
+#define VPE_EDI_LUT1_MASK		0x1f
+#define VPE_EDI_LUT1_SHIFT		8
+#define VPE_EDI_LUT2_MASK		0x1f
+#define VPE_EDI_LUT2_SHIFT		16
+#define VPE_EDI_LUT3_MASK		0x1f
+#define VPE_EDI_LUT3_SHIFT		24
+
+#define VPE_DEI_EDI_LUT_R2		0x0618
+#define VPE_EDI_LUT4_MASK		0x1f
+#define VPE_EDI_LUT4_SHIFT		0
+#define VPE_EDI_LUT5_MASK		0x1f
+#define VPE_EDI_LUT5_SHIFT		8
+#define VPE_EDI_LUT6_MASK		0x1f
+#define VPE_EDI_LUT6_SHIFT		16
+#define VPE_EDI_LUT7_MASK		0x1f
+#define VPE_EDI_LUT7_SHIFT		24
+
+#define VPE_DEI_EDI_LUT_R3		0x061c
+#define VPE_EDI_LUT8_MASK		0x1f
+#define VPE_EDI_LUT8_SHIFT		0
+#define VPE_EDI_LUT9_MASK		0x1f
+#define VPE_EDI_LUT9_SHIFT		8
+#define VPE_EDI_LUT10_MASK		0x1f
+#define VPE_EDI_LUT10_SHIFT		16
+#define VPE_EDI_LUT11_MASK		0x1f
+#define VPE_EDI_LUT11_SHIFT		24
+
+#define VPE_DEI_FMD_WINDOW_R0		0x0620
+#define VPE_FMD_WINDOW_MINX_MASK	0x07ff
+#define VPE_FMD_WINDOW_MINX_SHIFT	0
+#define VPE_FMD_WINDOW_MAXX_MASK	0x07ff
+#define VPE_FMD_WINDOW_MAXX_SHIFT	16
+#define VPE_FMD_WINDOW_ENABLE		(1 << 31)
+
+#define VPE_DEI_FMD_WINDOW_R1		0x0624
+#define VPE_FMD_WINDOW_MINY_MASK	0x07ff
+#define VPE_FMD_WINDOW_MINY_SHIFT	0
+#define VPE_FMD_WINDOW_MAXY_MASK	0x07ff
+#define VPE_FMD_WINDOW_MAXY_SHIFT	16
+
+#define VPE_DEI_FMD_CONTROL_R0		0x0628
+#define VPE_FMD_ENABLE			(1 << 0)
+#define VPE_FMD_LOCK			(1 << 1)
+#define VPE_FMD_JAM_DIR			(1 << 2)
+#define VPE_FMD_BED_ENABLE		(1 << 3)
+#define VPE_FMD_CAF_FIELD_THR_MASK	0xff
+#define VPE_FMD_CAF_FIELD_THR_SHIFT	16
+#define VPE_FMD_CAF_LINE_THR_MASK	0xff
+#define VPE_FMD_CAF_LINE_THR_SHIFT	24
+
+#define VPE_DEI_FMD_CONTROL_R1		0x062c
+#define VPE_FMD_CAF_THR_MASK		0x000fffff
+#define VPE_FMD_CAF_THR_SHIFT		0
+
+#define VPE_DEI_FMD_STATUS_R0		0x0630
+#define VPE_FMD_CAF_MASK		0x000fffff
+#define VPE_FMD_CAF_SHIFT		0
+#define VPE_FMD_RESET			(1 << 24)
+
+#define VPE_DEI_FMD_STATUS_R1		0x0634
+#define VPE_FMD_FIELD_DIFF_MASK		0x0fffffff
+#define VPE_FMD_FIELD_DIFF_SHIFT	0
+
+#define VPE_DEI_FMD_STATUS_R2		0x0638
+#define VPE_FMD_FRAME_DIFF_MASK		0x000fffff
+#define VPE_FMD_FRAME_DIFF_SHIFT	0
+
+/* VPE scaler regs */
+#define VPE_SC_MP_SC0			0x0700
+#define VPE_INTERLACE_O			(1 << 0)
+#define VPE_LINEAR			(1 << 1)
+#define VPE_SC_BYPASS			(1 << 2)
+#define VPE_INVT_FID			(1 << 3)
+#define VPE_USE_RAV			(1 << 4)
+#define VPE_ENABLE_EV			(1 << 5)
+#define VPE_AUTO_HS			(1 << 6)
+#define VPE_DCM_2X			(1 << 7)
+#define VPE_DCM_4X			(1 << 8)
+#define VPE_HP_BYPASS			(1 << 9)
+#define VPE_INTERLACE_I			(1 << 10)
+#define VPE_ENABLE_SIN2_VER_INTP	(1 << 11)
+#define VPE_Y_PK_EN			(1 << 14)
+#define VPE_TRIM			(1 << 15)
+#define VPE_SELFGEN_FID			(1 << 16)
+
+#define VPE_SC_MP_SC1			0x0704
+#define VPE_ROW_ACC_INC_MASK		0x07ffffff
+#define VPE_ROW_ACC_INC_SHIFT		0
+
+#define VPE_SC_MP_SC2			0x0708
+#define VPE_ROW_ACC_OFFSET_MASK		0x0fffffff
+#define VPE_ROW_ACC_OFFSET_SHIFT	0
+
+#define VPE_SC_MP_SC3			0x070c
+#define VPE_ROW_ACC_OFFSET_B_MASK	0x0fffffff
+#define VPE_ROW_ACC_OFFSET_B_SHIFT	0
+
+#define VPE_SC_MP_SC4			0x0710
+#define VPE_TAR_H_MASK			0x07ff
+#define VPE_TAR_H_SHIFT			0
+#define VPE_TAR_W_MASK			0x07ff
+#define VPE_TAR_W_SHIFT			12
+#define VPE_LIN_ACC_INC_U_MASK		0x07
+#define VPE_LIN_ACC_INC_U_SHIFT		24
+#define VPE_NLIN_ACC_INIT_U_MASK	0x07
+#define VPE_NLIN_ACC_INIT_U_SHIFT	28
+
+#define VPE_SC_MP_SC5			0x0714
+#define VPE_SRC_H_MASK			0x07ff
+#define VPE_SRC_H_SHIFT			0
+#define VPE_SRC_W_MASK			0x07ff
+#define VPE_SRC_W_SHIFT			12
+#define VPE_NLIN_ACC_INC_U_MASK		0x07
+#define VPE_NLIN_ACC_INC_U_SHIFT	24
+
+#define VPE_SC_MP_SC6			0x0718
+#define VPE_ROW_ACC_INIT_RAV_MASK	0x03ff
+#define VPE_ROW_ACC_INIT_RAV_SHIFT	0
+#define VPE_ROW_ACC_INIT_RAV_B_MASK	0x03ff
+#define VPE_ROW_ACC_INIT_RAV_B_SHIFT	10
+
+#define VPE_SC_MP_SC8			0x0720
+#define VPE_NLIN_LEFT_MASK		0x07ff
+#define VPE_NLIN_LEFT_SHIFT		0
+#define VPE_NLIN_RIGHT_MASK		0x07ff
+#define VPE_NLIN_RIGHT_SHIFT		12
+
+#define VPE_SC_MP_SC9			0x0724
+#define VPE_LIN_ACC_INC			VPE_SC_MP_SC9
+
+#define VPE_SC_MP_SC10			0x0728
+#define VPE_NLIN_ACC_INIT		VPE_SC_MP_SC10
+
+#define VPE_SC_MP_SC11			0x072c
+#define VPE_NLIN_ACC_INC		VPE_SC_MP_SC11
+
+#define VPE_SC_MP_SC12			0x0730
+#define VPE_COL_ACC_OFFSET_MASK		0x01ffffff
+#define VPE_COL_ACC_OFFSET_SHIFT	0
+
+#define VPE_SC_MP_SC13			0x0734
+#define VPE_SC_FACTOR_RAV_MASK		0x03ff
+#define VPE_SC_FACTOR_RAV_SHIFT		0
+#define VPE_CHROMA_INTP_THR_MASK	0x03ff
+#define VPE_CHROMA_INTP_THR_SHIFT	12
+#define VPE_DELTA_CHROMA_THR_MASK	0x0f
+#define VPE_DELTA_CHROMA_THR_SHIFT	24
+
+#define VPE_SC_MP_SC17			0x0744
+#define VPE_EV_THR_MASK			0x03ff
+#define VPE_EV_THR_SHIFT		12
+#define VPE_DELTA_LUMA_THR_MASK		0x0f
+#define VPE_DELTA_LUMA_THR_SHIFT	24
+#define VPE_DELTA_EV_THR_MASK		0x0f
+#define VPE_DELTA_EV_THR_SHIFT		28
+
+#define VPE_SC_MP_SC18			0x0748
+#define VPE_HS_FACTOR_MASK		0x03ff
+#define VPE_HS_FACTOR_SHIFT		0
+#define VPE_CONF_DEFAULT_MASK		0x01ff
+#define VPE_CONF_DEFAULT_SHIFT		16
+
+#define VPE_SC_MP_SC19			0x074c
+#define VPE_HPF_COEFF0_MASK		0xff
+#define VPE_HPF_COEFF0_SHIFT		0
+#define VPE_HPF_COEFF1_MASK		0xff
+#define VPE_HPF_COEFF1_SHIFT		8
+#define VPE_HPF_COEFF2_MASK		0xff
+#define VPE_HPF_COEFF2_SHIFT		16
+#define VPE_HPF_COEFF3_MASK		0xff
+#define VPE_HPF_COEFF3_SHIFT		23
+
+#define VPE_SC_MP_SC20			0x0750
+#define VPE_HPF_COEFF4_MASK		0xff
+#define VPE_HPF_COEFF4_SHIFT		0
+#define VPE_HPF_COEFF5_MASK		0xff
+#define VPE_HPF_COEFF5_SHIFT		8
+#define VPE_HPF_NORM_SHIFT_MASK		0x07
+#define VPE_HPF_NORM_SHIFT_SHIFT	16
+#define VPE_NL_LIMIT_MASK		0x1ff
+#define VPE_NL_LIMIT_SHIFT		20
+
+#define VPE_SC_MP_SC21			0x0754
+#define VPE_NL_LO_THR_MASK		0x01ff
+#define VPE_NL_LO_THR_SHIFT		0
+#define VPE_NL_LO_SLOPE_MASK		0xff
+#define VPE_NL_LO_SLOPE_SHIFT		16
+
+#define VPE_SC_MP_SC22			0x0758
+#define VPE_NL_HI_THR_MASK		0x01ff
+#define VPE_NL_HI_THR_SHIFT		0
+#define VPE_NL_HI_SLOPE_SH_MASK		0x07
+#define VPE_NL_HI_SLOPE_SH_SHIFT	16
+
+#define VPE_SC_MP_SC23			0x075c
+#define VPE_GRADIENT_THR_MASK		0x07ff
+#define VPE_GRADIENT_THR_SHIFT		0
+#define VPE_GRADIENT_THR_RANGE_MASK	0x0f
+#define VPE_GRADIENT_THR_RANGE_SHIFT	12
+#define VPE_MIN_GY_THR_MASK		0xff
+#define VPE_MIN_GY_THR_SHIFT		16
+#define VPE_MIN_GY_THR_RANGE_MASK	0x0f
+#define VPE_MIN_GY_THR_RANGE_SHIFT	28
+
+#define VPE_SC_MP_SC24			0x0760
+#define VPE_ORG_H_MASK			0x07ff
+#define VPE_ORG_H_SHIFT			0
+#define VPE_ORG_W_MASK			0x07ff
+#define VPE_ORG_W_SHIFT			16
+
+#define VPE_SC_MP_SC25			0x0764
+#define VPE_OFF_H_MASK			0x07ff
+#define VPE_OFF_H_SHIFT			0
+#define VPE_OFF_W_MASK			0x07ff
+#define VPE_OFF_W_SHIFT			16
+
+/* VPE color space converter regs */
+#define VPE_CSC_CSC00			0x5700
+#define VPE_CSC_A0_MASK			0x1fff
+#define VPE_CSC_A0_SHIFT		0
+#define VPE_CSC_B0_MASK			0x1fff
+#define VPE_CSC_B0_SHIFT		16
+
+#define VPE_CSC_CSC01			0x5704
+#define VPE_CSC_C0_MASK			0x1fff
+#define VPE_CSC_C0_SHIFT		0
+#define VPE_CSC_A1_MASK			0x1fff
+#define VPE_CSC_A1_SHIFT		16
+
+#define VPE_CSC_CSC02			0x5708
+#define VPE_CSC_B1_MASK			0x1fff
+#define VPE_CSC_B1_SHIFT		0
+#define VPE_CSC_C1_MASK			0x1fff
+#define VPE_CSC_C1_SHIFT		16
+
+#define VPE_CSC_CSC03			0x570c
+#define VPE_CSC_A2_MASK			0x1fff
+#define VPE_CSC_A2_SHIFT		0
+#define VPE_CSC_B2_MASK			0x1fff
+#define VPE_CSC_B2_SHIFT		16
+
+#define VPE_CSC_CSC04			0x5710
+#define VPE_CSC_C2_MASK			0x1fff
+#define VPE_CSC_C2_SHIFT		0
+#define VPE_CSC_D0_MASK			0x0fff
+#define VPE_CSC_D0_SHIFT		16
+
+#define VPE_CSC_CSC05			0x5714
+#define VPE_CSC_D1_MASK			0x0fff
+#define VPE_CSC_D1_SHIFT		0
+#define VPE_CSC_D2_MASK			0x0fff
+#define VPE_CSC_D2_SHIFT		16
+#define VPE_CSC_BYPASS			(1 << 28)
+
+#endif
diff --git a/drivers/media/platform/timblogiw.c b/drivers/media/platform/timblogiw.c
index b557caf5b1a4..ccdadd623a3a 100644
--- a/drivers/media/platform/timblogiw.c
+++ b/drivers/media/platform/timblogiw.c
@@ -403,7 +403,7 @@ static int timblogiw_s_input(struct file *file, void  *priv, unsigned int input)
 	return 0;
 }
 
-static int timblogiw_streamon(struct file *file, void  *priv, unsigned int type)
+static int timblogiw_streamon(struct file *file, void  *priv, enum v4l2_buf_type type)
 {
 	struct video_device *vdev = video_devdata(file);
 	struct timblogiw_fh *fh = priv;
@@ -420,7 +420,7 @@ static int timblogiw_streamon(struct file *file, void  *priv, unsigned int type)
 }
 
 static int timblogiw_streamoff(struct file *file, void  *priv,
-	unsigned int type)
+	enum v4l2_buf_type type)
 {
 	struct video_device *vdev = video_devdata(file);
 	struct timblogiw_fh *fh = priv;
@@ -565,7 +565,7 @@ static void buffer_queue(struct videobuf_queue *vq, struct videobuf_buffer *vb)
 
 	desc = dmaengine_prep_slave_sg(fh->chan,
 		buf->sg, sg_elems, DMA_DEV_TO_MEM,
-		DMA_PREP_INTERRUPT | DMA_COMPL_SKIP_SRC_UNMAP);
+		DMA_PREP_INTERRUPT);
 	if (!desc) {
 		spin_lock_irq(&fh->queue_lock);
 		list_del_init(&vb->queue);
diff --git a/drivers/media/radio/radio-keene.c b/drivers/media/radio/radio-keene.c
index 21db23b196be..fa3964022b96 100644
--- a/drivers/media/radio/radio-keene.c
+++ b/drivers/media/radio/radio-keene.c
@@ -123,7 +123,7 @@ static int keene_cmd_set(struct keene_device *radio)
 	/* If bit 0 is set, then transmit mono, otherwise stereo.
 	   If bit 2 is set, then enable 75 us preemphasis, otherwise
 	   it is 50 us. */
-	radio->buffer[3] = (!radio->stereo) | (radio->preemph_75_us ? 4 : 0);
+	radio->buffer[3] = (radio->stereo ? 0 : 1) | (radio->preemph_75_us ? 4 : 0);
 	radio->buffer[4] = 0x00;
 	radio->buffer[5] = 0x00;
 	radio->buffer[6] = 0x00;
diff --git a/drivers/media/radio/radio-sf16fmr2.c b/drivers/media/radio/radio-sf16fmr2.c
index f1e3714b5f16..93d864eb8306 100644
--- a/drivers/media/radio/radio-sf16fmr2.c
+++ b/drivers/media/radio/radio-sf16fmr2.c
@@ -74,8 +74,8 @@ static u8 fmr2_tea575x_get_pins(struct snd_tea575x *tea)
 	struct fmr2 *fmr2 = tea->private_data;
 	u8 bits = inb(fmr2->io);
 
-	return  (bits & STR_DATA) ? TEA575X_DATA : 0 |
-		(bits & STR_MOST) ? TEA575X_MOST : 0;
+	return  ((bits & STR_DATA) ? TEA575X_DATA : 0) |
+		((bits & STR_MOST) ? TEA575X_MOST : 0);
 }
 
 static void fmr2_tea575x_set_direction(struct snd_tea575x *tea, bool output)
@@ -295,7 +295,6 @@ static void fmr2_remove(struct fmr2 *fmr2)
 static int fmr2_isa_remove(struct device *pdev, unsigned int ndev)
 {
 	fmr2_remove(dev_get_drvdata(pdev));
-	dev_set_drvdata(pdev, NULL);
 
 	return 0;
 }
diff --git a/drivers/media/radio/radio-shark.c b/drivers/media/radio/radio-shark.c
index b91477212413..3db8a8cfe1a8 100644
--- a/drivers/media/radio/radio-shark.c
+++ b/drivers/media/radio/radio-shark.c
@@ -271,6 +271,7 @@ static void shark_unregister_leds(struct shark_device *shark)
 	cancel_work_sync(&shark->led_work);
 }
 
+#ifdef CONFIG_PM
 static void shark_resume_leds(struct shark_device *shark)
 {
 	if (test_bit(BLUE_IS_PULSE, &shark->brightness_new))
@@ -280,6 +281,7 @@ static void shark_resume_leds(struct shark_device *shark)
 	set_bit(RED_LED, &shark->brightness_new);
 	schedule_work(&shark->led_work);
 }
+#endif
 #else
 static int shark_register_leds(struct shark_device *shark, struct device *dev)
 {
diff --git a/drivers/media/radio/radio-shark2.c b/drivers/media/radio/radio-shark2.c
index 9fb669721e66..d86d90dab8bf 100644
--- a/drivers/media/radio/radio-shark2.c
+++ b/drivers/media/radio/radio-shark2.c
@@ -237,6 +237,7 @@ static void shark_unregister_leds(struct shark_device *shark)
 	cancel_work_sync(&shark->led_work);
 }
 
+#ifdef CONFIG_PM
 static void shark_resume_leds(struct shark_device *shark)
 {
 	int i;
@@ -246,6 +247,7 @@ static void shark_resume_leds(struct shark_device *shark)
 
 	schedule_work(&shark->led_work);
 }
+#endif
 #else
 static int shark_register_leds(struct shark_device *shark, struct device *dev)
 {
diff --git a/drivers/media/radio/radio-wl1273.c b/drivers/media/radio/radio-wl1273.c
index 97c2c18803ef..9cf6731fb816 100644
--- a/drivers/media/radio/radio-wl1273.c
+++ b/drivers/media/radio/radio-wl1273.c
@@ -375,7 +375,7 @@ static int wl1273_fm_set_tx_freq(struct wl1273_device *radio, unsigned int freq)
 	if (r)
 		return r;
 
-	INIT_COMPLETION(radio->busy);
+	reinit_completion(&radio->busy);
 
 	/* wait for the FR IRQ */
 	r = wait_for_completion_timeout(&radio->busy, msecs_to_jiffies(2000));
@@ -389,7 +389,7 @@ static int wl1273_fm_set_tx_freq(struct wl1273_device *radio, unsigned int freq)
 	if (r)
 		return r;
 
-	INIT_COMPLETION(radio->busy);
+	reinit_completion(&radio->busy);
 
 	/* wait for the POWER_ENB IRQ */
 	r = wait_for_completion_timeout(&radio->busy, msecs_to_jiffies(1000));
@@ -444,7 +444,7 @@ static int wl1273_fm_set_rx_freq(struct wl1273_device *radio, unsigned int freq)
 		goto err;
 	}
 
-	INIT_COMPLETION(radio->busy);
+	reinit_completion(&radio->busy);
 
 	r = wait_for_completion_timeout(&radio->busy, msecs_to_jiffies(2000));
 	if (!r) {
@@ -805,7 +805,7 @@ static int wl1273_fm_set_seek(struct wl1273_device *radio,
 	if (level < SCHAR_MIN || level > SCHAR_MAX)
 		return -EINVAL;
 
-	INIT_COMPLETION(radio->busy);
+	reinit_completion(&radio->busy);
 	dev_dbg(radio->dev, "%s: BUSY\n", __func__);
 
 	r = core->write(core, WL1273_INT_MASK_SET, radio->irq_flags);
@@ -847,7 +847,7 @@ static int wl1273_fm_set_seek(struct wl1273_device *radio,
 	if (r)
 		goto out;
 
-	INIT_COMPLETION(radio->busy);
+	reinit_completion(&radio->busy);
 	dev_dbg(radio->dev, "%s: BUSY\n", __func__);
 
 	r = core->write(core, WL1273_TUNER_MODE_SET, TUNER_MODE_AUTO_SEEK);
diff --git a/drivers/media/radio/si470x/radio-si470x-common.c b/drivers/media/radio/si470x/radio-si470x-common.c
index 5c57e5b0f949..0e750aef656a 100644
--- a/drivers/media/radio/si470x/radio-si470x-common.c
+++ b/drivers/media/radio/si470x/radio-si470x-common.c
@@ -218,7 +218,7 @@ static int si470x_set_chan(struct si470x_device *radio, unsigned short chan)
 		goto done;
 
 	/* wait till tune operation has completed */
-	INIT_COMPLETION(radio->completion);
+	reinit_completion(&radio->completion);
 	retval = wait_for_completion_timeout(&radio->completion,
 			msecs_to_jiffies(tune_timeout));
 	if (!retval)
@@ -254,7 +254,7 @@ static unsigned int si470x_get_step(struct si470x_device *radio)
 	/* 2:  50 kHz */
 	default:
 		return 50 * 16;
-	};
+	}
 }
 
 
@@ -341,7 +341,7 @@ static int si470x_set_seek(struct si470x_device *radio,
 		return retval;
 
 	/* wait till tune operation has completed */
-	INIT_COMPLETION(radio->completion);
+	reinit_completion(&radio->completion);
 	retval = wait_for_completion_timeout(&radio->completion,
 			msecs_to_jiffies(seek_timeout));
 	if (!retval)
diff --git a/drivers/media/radio/si470x/radio-si470x-i2c.c b/drivers/media/radio/si470x/radio-si470x-i2c.c
index e5fc9acd0c4f..2a497c80c77f 100644
--- a/drivers/media/radio/si470x/radio-si470x-i2c.c
+++ b/drivers/media/radio/si470x/radio-si470x-i2c.c
@@ -463,7 +463,7 @@ static int si470x_i2c_remove(struct i2c_client *client)
 }
 
 
-#ifdef CONFIG_PM
+#ifdef CONFIG_PM_SLEEP
 /*
  * si470x_i2c_suspend - suspend the device
  */
@@ -509,7 +509,7 @@ static struct i2c_driver si470x_i2c_driver = {
 	.driver = {
 		.name		= "si470x",
 		.owner		= THIS_MODULE,
-#ifdef CONFIG_PM
+#ifdef CONFIG_PM_SLEEP
 		.pm		= &si470x_i2c_pm,
 #endif
 	},
diff --git a/drivers/media/radio/si4713-i2c.c b/drivers/media/radio/si4713-i2c.c
index fe160882ee10..9ec48ccbcf0b 100644
--- a/drivers/media/radio/si4713-i2c.c
+++ b/drivers/media/radio/si4713-i2c.c
@@ -1456,7 +1456,7 @@ static int si4713_probe(struct i2c_client *client,
 
 	if (client->irq) {
 		rval = request_irq(client->irq,
-			si4713_handler, IRQF_TRIGGER_FALLING | IRQF_DISABLED,
+			si4713_handler, IRQF_TRIGGER_FALLING,
 			client->name, sdev);
 		if (rval < 0) {
 			v4l2_err(&sdev->sd, "Could not request IRQ\n");
diff --git a/drivers/media/radio/tef6862.c b/drivers/media/radio/tef6862.c
index 06ac69245ca1..69e3245a58a0 100644
--- a/drivers/media/radio/tef6862.c
+++ b/drivers/media/radio/tef6862.c
@@ -48,15 +48,15 @@
 #define WM_SUB_TEST		0xF
 
 /* Different modes of the MSA register */
-#define MODE_BUFFER		0x0
-#define MODE_PRESET		0x1
-#define MODE_SEARCH		0x2
-#define MODE_AF_UPDATE		0x3
-#define MODE_JUMP		0x4
-#define MODE_CHECK		0x5
-#define MODE_LOAD		0x6
-#define MODE_END		0x7
-#define MODE_SHIFT		5
+#define MSA_MODE_BUFFER		0x0
+#define MSA_MODE_PRESET		0x1
+#define MSA_MODE_SEARCH		0x2
+#define MSA_MODE_AF_UPDATE	0x3
+#define MSA_MODE_JUMP		0x4
+#define MSA_MODE_CHECK		0x5
+#define MSA_MODE_LOAD		0x6
+#define MSA_MODE_END		0x7
+#define MSA_MODE_SHIFT		5
 
 struct tef6862_state {
 	struct v4l2_subdev sd;
@@ -114,7 +114,7 @@ static int tef6862_s_frequency(struct v4l2_subdev *sd, const struct v4l2_frequen
 
 	clamp(freq, TEF6862_LO_FREQ, TEF6862_HI_FREQ);
 	pll = 1964 + ((freq - TEF6862_LO_FREQ) * 20) / FREQ_MUL;
-	i2cmsg[0] = (MODE_PRESET << MODE_SHIFT) | WM_SUB_PLLM;
+	i2cmsg[0] = (MSA_MODE_PRESET << MSA_MODE_SHIFT) | WM_SUB_PLLM;
 	i2cmsg[1] = (pll >> 8) & 0xff;
 	i2cmsg[2] = pll & 0xff;
 
diff --git a/drivers/media/radio/wl128x/fmdrv_common.c b/drivers/media/radio/wl128x/fmdrv_common.c
index 253f307f0b37..4b2e9e8298e1 100644
--- a/drivers/media/radio/wl128x/fmdrv_common.c
+++ b/drivers/media/radio/wl128x/fmdrv_common.c
@@ -175,7 +175,7 @@ static int_handler_prototype int_handler_table[] = {
 	fm_irq_handle_intmsk_cmd_resp
 };
 
-long (*g_st_write) (struct sk_buff *skb);
+static long (*g_st_write) (struct sk_buff *skb);
 static struct completion wait_for_fmdrv_reg_comp;
 
 static inline void fm_irq_call(struct fmdev *fmdev)
diff --git a/drivers/media/rc/Kconfig b/drivers/media/rc/Kconfig
index 11e84bcc23a1..904f11367c29 100644
--- a/drivers/media/rc/Kconfig
+++ b/drivers/media/rc/Kconfig
@@ -322,4 +322,14 @@ config IR_GPIO_CIR
 	   To compile this driver as a module, choose M here: the module will
 	   be called gpio-ir-recv.
 
+config RC_ST
+	tristate "ST remote control receiver"
+	depends on ARCH_STI && RC_CORE
+	help
+	 Say Y here if you want support for ST remote control driver
+	 which allows both IR and UHF RX.
+	 The driver passes raw pulse and space information to the LIRC decoder.
+
+	 If you're not sure, select N here.
+
 endif #RC_DEVICES
diff --git a/drivers/media/rc/Makefile b/drivers/media/rc/Makefile
index 56bacf07b361..f4eb32c0a455 100644
--- a/drivers/media/rc/Makefile
+++ b/drivers/media/rc/Makefile
@@ -30,3 +30,4 @@ obj-$(CONFIG_RC_LOOPBACK) += rc-loopback.o
 obj-$(CONFIG_IR_GPIO_CIR) += gpio-ir-recv.o
 obj-$(CONFIG_IR_IGUANA) += iguanair.o
 obj-$(CONFIG_IR_TTUSBIR) += ttusbir.o
+obj-$(CONFIG_RC_ST) += st_rc.o
diff --git a/drivers/media/rc/fintek-cir.h b/drivers/media/rc/fintek-cir.h
index 82516a1d39b0..b698f3d2ced9 100644
--- a/drivers/media/rc/fintek-cir.h
+++ b/drivers/media/rc/fintek-cir.h
@@ -76,8 +76,8 @@ struct fintek_dev {
 	} tx;
 
 	/* Config register index/data port pair */
-	u8 cr_ip;
-	u8 cr_dp;
+	u32 cr_ip;
+	u32 cr_dp;
 
 	/* hardware I/O settings */
 	unsigned long cir_addr;
diff --git a/drivers/media/rc/gpio-ir-recv.c b/drivers/media/rc/gpio-ir-recv.c
index 07aacfa5903d..80c611c2e8c2 100644
--- a/drivers/media/rc/gpio-ir-recv.c
+++ b/drivers/media/rc/gpio-ir-recv.c
@@ -16,6 +16,7 @@
 #include <linux/interrupt.h>
 #include <linux/gpio.h>
 #include <linux/slab.h>
+#include <linux/of.h>
 #include <linux/of_gpio.h>
 #include <linux/platform_device.h>
 #include <linux/irq.h>
diff --git a/drivers/media/rc/iguanair.c b/drivers/media/rc/iguanair.c
index 19632b1c2190..fdae05c4f377 100644
--- a/drivers/media/rc/iguanair.c
+++ b/drivers/media/rc/iguanair.c
@@ -207,7 +207,7 @@ static int iguanair_send(struct iguanair *ir, unsigned size)
 {
 	int rc;
 
-	INIT_COMPLETION(ir->completion);
+	reinit_completion(&ir->completion);
 
 	ir->urb_out->transfer_buffer_length = size;
 	rc = usb_submit_urb(ir->urb_out, GFP_KERNEL);
@@ -308,22 +308,12 @@ static int iguanair_set_tx_carrier(struct rc_dev *dev, uint32_t carrier)
 		cycles = DIV_ROUND_CLOSEST(24000000, carrier * 2) -
 							ir->cycle_overhead;
 
-		/*  make up the the remainer of 4-cycle blocks */
-		switch (cycles & 3) {
-		case 0:
-			sevens = 0;
-			break;
-		case 1:
-			sevens = 3;
-			break;
-		case 2:
-			sevens = 2;
-			break;
-		case 3:
-			sevens = 1;
-			break;
-		}
-
+		/*
+		 * Calculate minimum number of 7 cycles needed so
+		 * we are left with a multiple of 4; so we want to have
+		 * (sevens * 7) & 3 == cycles & 3
+		 */
+		sevens = (4 - cycles) & 3;
 		fours = (cycles - sevens * 7) / 4;
 
 		/* magic happens here */
diff --git a/drivers/media/rc/ir-rx51.c b/drivers/media/rc/ir-rx51.c
index 31b955bf7664..b1e19a26208d 100644
--- a/drivers/media/rc/ir-rx51.c
+++ b/drivers/media/rc/ir-rx51.c
@@ -201,8 +201,7 @@ static int lirc_rx51_init_port(struct lirc_rx51 *lirc_rx51)
 
 	lirc_rx51->irq_num = omap_dm_timer_get_irq(lirc_rx51->pulse_timer);
 	retval = request_irq(lirc_rx51->irq_num, lirc_rx51_interrupt_handler,
-			     IRQF_DISABLED | IRQF_SHARED,
-			     "lirc_pulse_timer", lirc_rx51);
+			     IRQF_SHARED, "lirc_pulse_timer", lirc_rx51);
 	if (retval) {
 		dev_err(lirc_rx51->dev, ": Failed to request interrupt line\n");
 		goto err2;
diff --git a/drivers/media/rc/keymaps/rc-dib0700-nec.c b/drivers/media/rc/keymaps/rc-dib0700-nec.c
index 4d13a7f2e5c3..492a05ade7e1 100644
--- a/drivers/media/rc/keymaps/rc-dib0700-nec.c
+++ b/drivers/media/rc/keymaps/rc-dib0700-nec.c
@@ -5,7 +5,7 @@
  * TODO: This table is a real mess, as it merges RC codes from several
  * devices into a big table. It also has both RC-5 and NEC codes inside.
  * It should be broken into small tables, and the protocols should properly
- * be indentificated.
+ * be identificated.
  *
  * The table were imported from dib0700_devices.c.
  *
diff --git a/drivers/media/rc/keymaps/rc-dib0700-rc5.c b/drivers/media/rc/keymaps/rc-dib0700-rc5.c
index ba81d9697cfc..454ea596a7ee 100644
--- a/drivers/media/rc/keymaps/rc-dib0700-rc5.c
+++ b/drivers/media/rc/keymaps/rc-dib0700-rc5.c
@@ -5,7 +5,7 @@
  * TODO: This table is a real mess, as it merges RC codes from several
  * devices into a big table. It also has both RC-5 and NEC codes inside.
  * It should be broken into small tables, and the protocols should properly
- * be indentificated.
+ * be identificated.
  *
  * The table were imported from dib0700_devices.c.
  *
diff --git a/drivers/media/rc/nuvoton-cir.h b/drivers/media/rc/nuvoton-cir.h
index 7c3674ff5ea2..07e83108df0f 100644
--- a/drivers/media/rc/nuvoton-cir.h
+++ b/drivers/media/rc/nuvoton-cir.h
@@ -84,8 +84,8 @@ struct nvt_dev {
 	} tx;
 
 	/* EFER Config register index/data pair */
-	u8 cr_efir;
-	u8 cr_efdr;
+	u32 cr_efir;
+	u32 cr_efdr;
 
 	/* hardware I/O settings */
 	unsigned long cir_addr;
diff --git a/drivers/media/rc/st_rc.c b/drivers/media/rc/st_rc.c
new file mode 100644
index 000000000000..65120c2d47ad
--- /dev/null
+++ b/drivers/media/rc/st_rc.c
@@ -0,0 +1,395 @@
+/*
+ * Copyright (C) 2013 STMicroelectronics Limited
+ * Author: Srinivas Kandagatla <srinivas.kandagatla@st.com>
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation; either version 2 of the License, or
+ * (at your option) any later version.
+ */
+#include <linux/kernel.h>
+#include <linux/clk.h>
+#include <linux/interrupt.h>
+#include <linux/module.h>
+#include <linux/of.h>
+#include <linux/platform_device.h>
+#include <media/rc-core.h>
+#include <linux/pinctrl/consumer.h>
+
+struct st_rc_device {
+	struct device			*dev;
+	int				irq;
+	int				irq_wake;
+	struct clk			*sys_clock;
+	void				*base;	/* Register base address */
+	void				*rx_base;/* RX Register base address */
+	struct rc_dev			*rdev;
+	bool				overclocking;
+	int				sample_mult;
+	int				sample_div;
+	bool				rxuhfmode;
+};
+
+/* Registers */
+#define IRB_SAMPLE_RATE_COMM	0x64	/* sample freq divisor*/
+#define IRB_CLOCK_SEL		0x70	/* clock select       */
+#define IRB_CLOCK_SEL_STATUS	0x74	/* clock status       */
+/* IRB IR/UHF receiver registers */
+#define IRB_RX_ON               0x40	/* pulse time capture */
+#define IRB_RX_SYS              0X44	/* sym period capture */
+#define IRB_RX_INT_EN           0x48	/* IRQ enable (R/W)   */
+#define IRB_RX_INT_STATUS       0x4c	/* IRQ status (R/W)   */
+#define IRB_RX_EN               0x50	/* Receive enable     */
+#define IRB_MAX_SYM_PERIOD      0x54	/* max sym value      */
+#define IRB_RX_INT_CLEAR        0x58	/* overrun status     */
+#define IRB_RX_STATUS           0x6c	/* receive status     */
+#define IRB_RX_NOISE_SUPPR      0x5c	/* noise suppression  */
+#define IRB_RX_POLARITY_INV     0x68	/* polarity inverter  */
+
+/**
+ * IRQ set: Enable full FIFO                 1  -> bit  3;
+ *          Enable overrun IRQ               1  -> bit  2;
+ *          Enable last symbol IRQ           1  -> bit  1:
+ *          Enable RX interrupt              1  -> bit  0;
+ */
+#define IRB_RX_INTS		0x0f
+#define IRB_RX_OVERRUN_INT	0x04
+ /* maximum symbol period (microsecs),timeout to detect end of symbol train */
+#define MAX_SYMB_TIME		0x5000
+#define IRB_SAMPLE_FREQ		10000000
+#define	IRB_FIFO_NOT_EMPTY	0xff00
+#define IRB_OVERFLOW		0x4
+#define IRB_TIMEOUT		0xffff
+#define IR_ST_NAME "st-rc"
+
+static void st_rc_send_lirc_timeout(struct rc_dev *rdev)
+{
+	DEFINE_IR_RAW_EVENT(ev);
+	ev.timeout = true;
+	ir_raw_event_store(rdev, &ev);
+}
+
+/**
+ * RX graphical example to better understand the difference between ST IR block
+ * output and standard definition used by LIRC (and most of the world!)
+ *
+ *           mark                                     mark
+ *      |-IRB_RX_ON-|                            |-IRB_RX_ON-|
+ *      ___  ___  ___                            ___  ___  ___             _
+ *      | |  | |  | |                            | |  | |  | |             |
+ *      | |  | |  | |         space 0            | |  | |  | |   space 1   |
+ * _____| |__| |__| |____________________________| |__| |__| |_____________|
+ *
+ *      |--------------- IRB_RX_SYS -------------|------ IRB_RX_SYS -------|
+ *
+ *      |------------- encoding bit 0 -----------|---- encoding bit 1 -----|
+ *
+ * ST hardware returns mark (IRB_RX_ON) and total symbol time (IRB_RX_SYS), so
+ * convert to standard mark/space we have to calculate space=(IRB_RX_SYS-mark)
+ * The mark time represents the amount of time the carrier (usually 36-40kHz)
+ * is detected.The above examples shows Pulse Width Modulation encoding where
+ * bit 0 is represented by space>mark.
+ */
+
+static irqreturn_t st_rc_rx_interrupt(int irq, void *data)
+{
+	unsigned int symbol, mark = 0;
+	struct st_rc_device *dev = data;
+	int last_symbol = 0;
+	u32 status;
+	DEFINE_IR_RAW_EVENT(ev);
+
+	if (dev->irq_wake)
+		pm_wakeup_event(dev->dev, 0);
+
+	status  = readl(dev->rx_base + IRB_RX_STATUS);
+
+	while (status & (IRB_FIFO_NOT_EMPTY | IRB_OVERFLOW)) {
+		u32 int_status = readl(dev->rx_base + IRB_RX_INT_STATUS);
+		if (unlikely(int_status & IRB_RX_OVERRUN_INT)) {
+			/* discard the entire collection in case of errors!  */
+			ir_raw_event_reset(dev->rdev);
+			dev_info(dev->dev, "IR RX overrun\n");
+			writel(IRB_RX_OVERRUN_INT,
+					dev->rx_base + IRB_RX_INT_CLEAR);
+			continue;
+		}
+
+		symbol = readl(dev->rx_base + IRB_RX_SYS);
+		mark = readl(dev->rx_base + IRB_RX_ON);
+
+		if (symbol == IRB_TIMEOUT)
+			last_symbol = 1;
+
+		 /* Ignore any noise */
+		if ((mark > 2) && (symbol > 1)) {
+			symbol -= mark;
+			if (dev->overclocking) { /* adjustments to timings */
+				symbol *= dev->sample_mult;
+				symbol /= dev->sample_div;
+				mark *= dev->sample_mult;
+				mark /= dev->sample_div;
+			}
+
+			ev.duration = US_TO_NS(mark);
+			ev.pulse = true;
+			ir_raw_event_store(dev->rdev, &ev);
+
+			if (!last_symbol) {
+				ev.duration = US_TO_NS(symbol);
+				ev.pulse = false;
+				ir_raw_event_store(dev->rdev, &ev);
+			} else  {
+				st_rc_send_lirc_timeout(dev->rdev);
+			}
+
+		}
+		last_symbol = 0;
+		status  = readl(dev->rx_base + IRB_RX_STATUS);
+	}
+
+	writel(IRB_RX_INTS, dev->rx_base + IRB_RX_INT_CLEAR);
+
+	/* Empty software fifo */
+	ir_raw_event_handle(dev->rdev);
+	return IRQ_HANDLED;
+}
+
+static void st_rc_hardware_init(struct st_rc_device *dev)
+{
+	int baseclock, freqdiff;
+	unsigned int rx_max_symbol_per = MAX_SYMB_TIME;
+	unsigned int rx_sampling_freq_div;
+
+	clk_prepare_enable(dev->sys_clock);
+	baseclock = clk_get_rate(dev->sys_clock);
+
+	/* IRB input pins are inverted internally from high to low. */
+	writel(1, dev->rx_base + IRB_RX_POLARITY_INV);
+
+	rx_sampling_freq_div = baseclock / IRB_SAMPLE_FREQ;
+	writel(rx_sampling_freq_div, dev->base + IRB_SAMPLE_RATE_COMM);
+
+	freqdiff = baseclock - (rx_sampling_freq_div * IRB_SAMPLE_FREQ);
+	if (freqdiff) { /* over clocking, workout the adjustment factors */
+		dev->overclocking = true;
+		dev->sample_mult = 1000;
+		dev->sample_div = baseclock / (10000 * rx_sampling_freq_div);
+		rx_max_symbol_per = (rx_max_symbol_per * 1000)/dev->sample_div;
+	}
+
+	writel(rx_max_symbol_per, dev->rx_base + IRB_MAX_SYM_PERIOD);
+}
+
+static int st_rc_remove(struct platform_device *pdev)
+{
+	struct st_rc_device *rc_dev = platform_get_drvdata(pdev);
+	clk_disable_unprepare(rc_dev->sys_clock);
+	rc_unregister_device(rc_dev->rdev);
+	return 0;
+}
+
+static int st_rc_open(struct rc_dev *rdev)
+{
+	struct st_rc_device *dev = rdev->priv;
+	unsigned long flags;
+	local_irq_save(flags);
+	/* enable interrupts and receiver */
+	writel(IRB_RX_INTS, dev->rx_base + IRB_RX_INT_EN);
+	writel(0x01, dev->rx_base + IRB_RX_EN);
+	local_irq_restore(flags);
+
+	return 0;
+}
+
+static void st_rc_close(struct rc_dev *rdev)
+{
+	struct st_rc_device *dev = rdev->priv;
+	/* disable interrupts and receiver */
+	writel(0x00, dev->rx_base + IRB_RX_EN);
+	writel(0x00, dev->rx_base + IRB_RX_INT_EN);
+}
+
+static int st_rc_probe(struct platform_device *pdev)
+{
+	int ret = -EINVAL;
+	struct rc_dev *rdev;
+	struct device *dev = &pdev->dev;
+	struct resource *res;
+	struct st_rc_device *rc_dev;
+	struct device_node *np = pdev->dev.of_node;
+	const char *rx_mode;
+
+	rc_dev = devm_kzalloc(dev, sizeof(struct st_rc_device), GFP_KERNEL);
+
+	if (!rc_dev)
+		return -ENOMEM;
+
+	rdev = rc_allocate_device();
+
+	if (!rdev)
+		return -ENOMEM;
+
+	if (np && !of_property_read_string(np, "rx-mode", &rx_mode)) {
+
+		if (!strcmp(rx_mode, "uhf")) {
+			rc_dev->rxuhfmode = true;
+		} else if (!strcmp(rx_mode, "infrared")) {
+			rc_dev->rxuhfmode = false;
+		} else {
+			dev_err(dev, "Unsupported rx mode [%s]\n", rx_mode);
+			goto err;
+		}
+
+	} else {
+		goto err;
+	}
+
+	rc_dev->sys_clock = devm_clk_get(dev, NULL);
+	if (IS_ERR(rc_dev->sys_clock)) {
+		dev_err(dev, "System clock not found\n");
+		ret = PTR_ERR(rc_dev->sys_clock);
+		goto err;
+	}
+
+	rc_dev->irq = platform_get_irq(pdev, 0);
+	if (rc_dev->irq < 0) {
+		ret = rc_dev->irq;
+		goto err;
+	}
+
+	res = platform_get_resource(pdev, IORESOURCE_MEM, 0);
+
+	rc_dev->base = devm_ioremap_resource(dev, res);
+	if (IS_ERR(rc_dev->base)) {
+		ret = PTR_ERR(rc_dev->base);
+		goto err;
+	}
+
+	if (rc_dev->rxuhfmode)
+		rc_dev->rx_base = rc_dev->base + 0x40;
+	else
+		rc_dev->rx_base = rc_dev->base;
+
+	rc_dev->dev = dev;
+	platform_set_drvdata(pdev, rc_dev);
+	st_rc_hardware_init(rc_dev);
+
+	rdev->driver_type = RC_DRIVER_IR_RAW;
+	rdev->allowed_protos = RC_BIT_ALL;
+	/* rx sampling rate is 10Mhz */
+	rdev->rx_resolution = 100;
+	rdev->timeout = US_TO_NS(MAX_SYMB_TIME);
+	rdev->priv = rc_dev;
+	rdev->open = st_rc_open;
+	rdev->close = st_rc_close;
+	rdev->driver_name = IR_ST_NAME;
+	rdev->map_name = RC_MAP_LIRC;
+	rdev->input_name = "ST Remote Control Receiver";
+
+	/* enable wake via this device */
+	device_set_wakeup_capable(dev, true);
+	device_set_wakeup_enable(dev, true);
+
+	ret = rc_register_device(rdev);
+	if (ret < 0)
+		goto clkerr;
+
+	rc_dev->rdev = rdev;
+	if (devm_request_irq(dev, rc_dev->irq, st_rc_rx_interrupt,
+			IRQF_NO_SUSPEND, IR_ST_NAME, rc_dev) < 0) {
+		dev_err(dev, "IRQ %d register failed\n", rc_dev->irq);
+		ret = -EINVAL;
+		goto rcerr;
+	}
+
+	/**
+	 * for LIRC_MODE_MODE2 or LIRC_MODE_PULSE or LIRC_MODE_RAW
+	 * lircd expects a long space first before a signal train to sync.
+	 */
+	st_rc_send_lirc_timeout(rdev);
+
+	dev_info(dev, "setup in %s mode\n", rc_dev->rxuhfmode ? "UHF" : "IR");
+
+	return ret;
+rcerr:
+	rc_unregister_device(rdev);
+	rdev = NULL;
+clkerr:
+	clk_disable_unprepare(rc_dev->sys_clock);
+err:
+	rc_free_device(rdev);
+	dev_err(dev, "Unable to register device (%d)\n", ret);
+	return ret;
+}
+
+#ifdef CONFIG_PM
+static int st_rc_suspend(struct device *dev)
+{
+	struct st_rc_device *rc_dev = dev_get_drvdata(dev);
+
+	if (device_may_wakeup(dev)) {
+		if (!enable_irq_wake(rc_dev->irq))
+			rc_dev->irq_wake = 1;
+		else
+			return -EINVAL;
+	} else {
+		pinctrl_pm_select_sleep_state(dev);
+		writel(0x00, rc_dev->rx_base + IRB_RX_EN);
+		writel(0x00, rc_dev->rx_base + IRB_RX_INT_EN);
+		clk_disable_unprepare(rc_dev->sys_clock);
+	}
+
+	return 0;
+}
+
+static int st_rc_resume(struct device *dev)
+{
+	struct st_rc_device *rc_dev = dev_get_drvdata(dev);
+	struct rc_dev	*rdev = rc_dev->rdev;
+
+	if (rc_dev->irq_wake) {
+		disable_irq_wake(rc_dev->irq);
+		rc_dev->irq_wake = 0;
+	} else {
+		pinctrl_pm_select_default_state(dev);
+		st_rc_hardware_init(rc_dev);
+		if (rdev->users) {
+			writel(IRB_RX_INTS, rc_dev->rx_base + IRB_RX_INT_EN);
+			writel(0x01, rc_dev->rx_base + IRB_RX_EN);
+		}
+	}
+
+	return 0;
+}
+
+static SIMPLE_DEV_PM_OPS(st_rc_pm_ops, st_rc_suspend, st_rc_resume);
+#endif
+
+#ifdef CONFIG_OF
+static struct of_device_id st_rc_match[] = {
+	{ .compatible = "st,comms-irb", },
+	{},
+};
+
+MODULE_DEVICE_TABLE(of, st_rc_match);
+#endif
+
+static struct platform_driver st_rc_driver = {
+	.driver = {
+		.name = IR_ST_NAME,
+		.owner	= THIS_MODULE,
+		.of_match_table = of_match_ptr(st_rc_match),
+#ifdef CONFIG_PM
+		.pm     = &st_rc_pm_ops,
+#endif
+	},
+	.probe = st_rc_probe,
+	.remove = st_rc_remove,
+};
+
+module_platform_driver(st_rc_driver);
+
+MODULE_DESCRIPTION("RC Transceiver driver for STMicroelectronics platforms");
+MODULE_AUTHOR("STMicroelectronics (R&D) Ltd");
+MODULE_LICENSE("GPL");
diff --git a/drivers/media/rc/winbond-cir.c b/drivers/media/rc/winbond-cir.c
index 98bd4960c75e..904baf4eec28 100644
--- a/drivers/media/rc/winbond-cir.c
+++ b/drivers/media/rc/winbond-cir.c
@@ -1110,7 +1110,7 @@ wbcir_probe(struct pnp_dev *device, const struct pnp_device_id *dev_id)
 	}
 
 	err = request_irq(data->irq, wbcir_irq_handler,
-			  IRQF_DISABLED, DRVNAME, device);
+			  0, DRVNAME, device);
 	if (err) {
 		dev_err(dev, "Failed to claim IRQ %u\n", data->irq);
 		err = -EBUSY;
diff --git a/drivers/media/tuners/e4000.c b/drivers/media/tuners/e4000.c
index 6c96e4898777..72971a8d3c37 100644
--- a/drivers/media/tuners/e4000.c
+++ b/drivers/media/tuners/e4000.c
@@ -21,20 +21,30 @@
 #include "e4000_priv.h"
 #include <linux/math64.h>
 
+/* Max transfer size done by I2C transfer functions */
+#define MAX_XFER_SIZE  64
+
 /* write multiple registers */
 static int e4000_wr_regs(struct e4000_priv *priv, u8 reg, u8 *val, int len)
 {
 	int ret;
-	u8 buf[1 + len];
+	u8 buf[MAX_XFER_SIZE];
 	struct i2c_msg msg[1] = {
 		{
 			.addr = priv->cfg->i2c_addr,
 			.flags = 0,
-			.len = sizeof(buf),
+			.len = 1 + len,
 			.buf = buf,
 		}
 	};
 
+	if (1 + len > sizeof(buf)) {
+		dev_warn(&priv->i2c->dev,
+			 "%s: i2c wr reg=%04x: len=%d is too big!\n",
+			 KBUILD_MODNAME, reg, len);
+		return -EINVAL;
+	}
+
 	buf[0] = reg;
 	memcpy(&buf[1], val, len);
 
@@ -54,7 +64,7 @@ static int e4000_wr_regs(struct e4000_priv *priv, u8 reg, u8 *val, int len)
 static int e4000_rd_regs(struct e4000_priv *priv, u8 reg, u8 *val, int len)
 {
 	int ret;
-	u8 buf[len];
+	u8 buf[MAX_XFER_SIZE];
 	struct i2c_msg msg[2] = {
 		{
 			.addr = priv->cfg->i2c_addr,
@@ -64,11 +74,18 @@ static int e4000_rd_regs(struct e4000_priv *priv, u8 reg, u8 *val, int len)
 		}, {
 			.addr = priv->cfg->i2c_addr,
 			.flags = I2C_M_RD,
-			.len = sizeof(buf),
+			.len = len,
 			.buf = buf,
 		}
 	};
 
+	if (len > sizeof(buf)) {
+		dev_warn(&priv->i2c->dev,
+			 "%s: i2c rd reg=%04x: len=%d is too big!\n",
+			 KBUILD_MODNAME, reg, len);
+		return -EINVAL;
+	}
+
 	ret = i2c_transfer(priv->i2c, msg, 2);
 	if (ret == 2) {
 		memcpy(val, buf, len);
diff --git a/drivers/media/tuners/fc0012.c b/drivers/media/tuners/fc0012.c
index f4d0e797a6cc..d74e92056810 100644
--- a/drivers/media/tuners/fc0012.c
+++ b/drivers/media/tuners/fc0012.c
@@ -139,7 +139,7 @@ static int fc0012_set_params(struct dvb_frontend *fe)
 	unsigned char reg[7], am, pm, multi, tmp;
 	unsigned long f_vco;
 	unsigned short xtal_freq_khz_2, xin, xdiv;
-	int vco_select = false;
+	bool vco_select = false;
 
 	if (fe->callback) {
 		ret = fe->callback(priv->i2c, DVB_FRONTEND_COMPONENT_TUNER,
diff --git a/drivers/media/tuners/fc0013.c b/drivers/media/tuners/fc0013.c
index bd8f0f1e8f3b..b4162315773d 100644
--- a/drivers/media/tuners/fc0013.c
+++ b/drivers/media/tuners/fc0013.c
@@ -233,7 +233,7 @@ static int fc0013_set_params(struct dvb_frontend *fe)
 	unsigned char reg[7], am, pm, multi, tmp;
 	unsigned long f_vco;
 	unsigned short xtal_freq_khz_2, xin, xdiv;
-	int vco_select = false;
+	bool vco_select = false;
 
 	if (fe->callback) {
 		ret = fe->callback(priv->i2c, DVB_FRONTEND_COMPONENT_TUNER,
diff --git a/drivers/media/tuners/fc2580.c b/drivers/media/tuners/fc2580.c
index 81f38aae9c66..3aecaf465094 100644
--- a/drivers/media/tuners/fc2580.c
+++ b/drivers/media/tuners/fc2580.c
@@ -20,6 +20,9 @@
 
 #include "fc2580_priv.h"
 
+/* Max transfer size done by I2C transfer functions */
+#define MAX_XFER_SIZE  64
+
 /*
  * TODO:
  * I2C write and read works only for one single register. Multiple registers
@@ -41,16 +44,23 @@
 static int fc2580_wr_regs(struct fc2580_priv *priv, u8 reg, u8 *val, int len)
 {
 	int ret;
-	u8 buf[1 + len];
+	u8 buf[MAX_XFER_SIZE];
 	struct i2c_msg msg[1] = {
 		{
 			.addr = priv->cfg->i2c_addr,
 			.flags = 0,
-			.len = sizeof(buf),
+			.len = 1 + len,
 			.buf = buf,
 		}
 	};
 
+	if (1 + len > sizeof(buf)) {
+		dev_warn(&priv->i2c->dev,
+			 "%s: i2c wr reg=%04x: len=%d is too big!\n",
+			 KBUILD_MODNAME, reg, len);
+		return -EINVAL;
+	}
+
 	buf[0] = reg;
 	memcpy(&buf[1], val, len);
 
@@ -69,7 +79,7 @@ static int fc2580_wr_regs(struct fc2580_priv *priv, u8 reg, u8 *val, int len)
 static int fc2580_rd_regs(struct fc2580_priv *priv, u8 reg, u8 *val, int len)
 {
 	int ret;
-	u8 buf[len];
+	u8 buf[MAX_XFER_SIZE];
 	struct i2c_msg msg[2] = {
 		{
 			.addr = priv->cfg->i2c_addr,
@@ -79,11 +89,18 @@ static int fc2580_rd_regs(struct fc2580_priv *priv, u8 reg, u8 *val, int len)
 		}, {
 			.addr = priv->cfg->i2c_addr,
 			.flags = I2C_M_RD,
-			.len = sizeof(buf),
+			.len = len,
 			.buf = buf,
 		}
 	};
 
+	if (len > sizeof(buf)) {
+		dev_warn(&priv->i2c->dev,
+			 "%s: i2c rd reg=%04x: len=%d is too big!\n",
+			 KBUILD_MODNAME, reg, len);
+		return -EINVAL;
+	}
+
 	ret = i2c_transfer(priv->i2c, msg, 2);
 	if (ret == 2) {
 		memcpy(val, buf, len);
diff --git a/drivers/media/tuners/r820t.c b/drivers/media/tuners/r820t.c
index 1c23666468cf..d9ee43fae62d 100644
--- a/drivers/media/tuners/r820t.c
+++ b/drivers/media/tuners/r820t.c
@@ -612,10 +612,19 @@ static int r820t_set_pll(struct r820t_priv *priv, enum v4l2_tuner_type type,
 
 	vco_fine_tune = (data[4] & 0x30) >> 4;
 
-	if (vco_fine_tune > VCO_POWER_REF)
-		div_num = div_num - 1;
-	else if (vco_fine_tune < VCO_POWER_REF)
-		div_num = div_num + 1;
+	tuner_dbg("mix_div=%d div_num=%d vco_fine_tune=%d\n",
+			mix_div, div_num, vco_fine_tune);
+
+	/*
+	 * XXX: R828D/16MHz seems to have always vco_fine_tune=1.
+	 * Due to that, this calculation goes wrong.
+	 */
+	if (priv->cfg->rafael_chip != CHIP_R828D) {
+		if (vco_fine_tune > VCO_POWER_REF)
+			div_num = div_num - 1;
+		else if (vco_fine_tune < VCO_POWER_REF)
+			div_num = div_num + 1;
+	}
 
 	rc = r820t_write_reg_mask(priv, 0x10, div_num << 5, 0xe0);
 	if (rc < 0)
@@ -637,11 +646,6 @@ static int r820t_set_pll(struct r820t_priv *priv, enum v4l2_tuner_type type,
 		vco_fra = pll_ref * 129 / 128;
 	}
 
-	if (nint > 63) {
-		tuner_info("No valid PLL values for %u kHz!\n", freq);
-		return -EINVAL;
-	}
-
 	ni = (nint - 13) / 4;
 	si = nint - 4 * ni - 13;
 
diff --git a/drivers/media/tuners/tda18212.c b/drivers/media/tuners/tda18212.c
index e4a84ee231cf..abe256e1f843 100644
--- a/drivers/media/tuners/tda18212.c
+++ b/drivers/media/tuners/tda18212.c
@@ -20,6 +20,9 @@
 
 #include "tda18212.h"
 
+/* Max transfer size done by I2C transfer functions */
+#define MAX_XFER_SIZE  64
+
 struct tda18212_priv {
 	struct tda18212_config *cfg;
 	struct i2c_adapter *i2c;
@@ -32,16 +35,23 @@ static int tda18212_wr_regs(struct tda18212_priv *priv, u8 reg, u8 *val,
 	int len)
 {
 	int ret;
-	u8 buf[len+1];
+	u8 buf[MAX_XFER_SIZE];
 	struct i2c_msg msg[1] = {
 		{
 			.addr = priv->cfg->i2c_address,
 			.flags = 0,
-			.len = sizeof(buf),
+			.len = 1 + len,
 			.buf = buf,
 		}
 	};
 
+	if (1 + len > sizeof(buf)) {
+		dev_warn(&priv->i2c->dev,
+			 "%s: i2c wr reg=%04x: len=%d is too big!\n",
+			 KBUILD_MODNAME, reg, len);
+		return -EINVAL;
+	}
+
 	buf[0] = reg;
 	memcpy(&buf[1], val, len);
 
@@ -61,7 +71,7 @@ static int tda18212_rd_regs(struct tda18212_priv *priv, u8 reg, u8 *val,
 	int len)
 {
 	int ret;
-	u8 buf[len];
+	u8 buf[MAX_XFER_SIZE];
 	struct i2c_msg msg[2] = {
 		{
 			.addr = priv->cfg->i2c_address,
@@ -71,11 +81,18 @@ static int tda18212_rd_regs(struct tda18212_priv *priv, u8 reg, u8 *val,
 		}, {
 			.addr = priv->cfg->i2c_address,
 			.flags = I2C_M_RD,
-			.len = sizeof(buf),
+			.len = len,
 			.buf = buf,
 		}
 	};
 
+	if (len > sizeof(buf)) {
+		dev_warn(&priv->i2c->dev,
+			 "%s: i2c rd reg=%04x: len=%d is too big!\n",
+			 KBUILD_MODNAME, reg, len);
+		return -EINVAL;
+	}
+
 	ret = i2c_transfer(priv->i2c, msg, 2);
 	if (ret == 2) {
 		memcpy(val, buf, len);
diff --git a/drivers/media/tuners/tda18218.c b/drivers/media/tuners/tda18218.c
index 2d31aeb6b088..9300e9361e3b 100644
--- a/drivers/media/tuners/tda18218.c
+++ b/drivers/media/tuners/tda18218.c
@@ -20,11 +20,14 @@
 
 #include "tda18218_priv.h"
 
+/* Max transfer size done by I2C transfer functions */
+#define MAX_XFER_SIZE  64
+
 /* write multiple registers */
 static int tda18218_wr_regs(struct tda18218_priv *priv, u8 reg, u8 *val, u8 len)
 {
 	int ret = 0, len2, remaining;
-	u8 buf[1 + len];
+	u8 buf[MAX_XFER_SIZE];
 	struct i2c_msg msg[1] = {
 		{
 			.addr = priv->cfg->i2c_address,
@@ -33,6 +36,13 @@ static int tda18218_wr_regs(struct tda18218_priv *priv, u8 reg, u8 *val, u8 len)
 		}
 	};
 
+	if (1 + len > sizeof(buf)) {
+		dev_warn(&priv->i2c->dev,
+			 "%s: i2c wr reg=%04x: len=%d is too big!\n",
+			 KBUILD_MODNAME, reg, len);
+		return -EINVAL;
+	}
+
 	for (remaining = len; remaining > 0;
 			remaining -= (priv->cfg->i2c_wr_max - 1)) {
 		len2 = remaining;
@@ -63,7 +73,7 @@ static int tda18218_wr_regs(struct tda18218_priv *priv, u8 reg, u8 *val, u8 len)
 static int tda18218_rd_regs(struct tda18218_priv *priv, u8 reg, u8 *val, u8 len)
 {
 	int ret;
-	u8 buf[reg+len]; /* we must start read always from reg 0x00 */
+	u8 buf[MAX_XFER_SIZE]; /* we must start read always from reg 0x00 */
 	struct i2c_msg msg[2] = {
 		{
 			.addr = priv->cfg->i2c_address,
@@ -73,11 +83,18 @@ static int tda18218_rd_regs(struct tda18218_priv *priv, u8 reg, u8 *val, u8 len)
 		}, {
 			.addr = priv->cfg->i2c_address,
 			.flags = I2C_M_RD,
-			.len = sizeof(buf),
+			.len = reg + len,
 			.buf = buf,
 		}
 	};
 
+	if (reg + len > sizeof(buf)) {
+		dev_warn(&priv->i2c->dev,
+			 "%s: i2c wr reg=%04x: len=%d is too big!\n",
+			 KBUILD_MODNAME, reg, len);
+		return -EINVAL;
+	}
+
 	ret = i2c_transfer(priv->i2c, msg, 2);
 	if (ret == 2) {
 		memcpy(val, &buf[reg], len);
diff --git a/drivers/media/tuners/tda9887.c b/drivers/media/tuners/tda9887.c
index 300005c535ba..9823248d743f 100644
--- a/drivers/media/tuners/tda9887.c
+++ b/drivers/media/tuners/tda9887.c
@@ -536,8 +536,8 @@ static int tda9887_status(struct dvb_frontend *fe)
 	unsigned char buf[1];
 	int rc;
 
-	memset(buf,0,sizeof(buf));
-	if (1 != (rc = tuner_i2c_xfer_recv(&priv->i2c_props,buf,1)))
+	rc = tuner_i2c_xfer_recv(&priv->i2c_props, buf, 1);
+	if (rc != 1)
 		tuner_info("i2c i/o error: rc == %d (should be 1)\n", rc);
 	dump_read_message(fe, buf);
 	return 0;
diff --git a/drivers/media/tuners/tuner-xc2028.c b/drivers/media/tuners/tuner-xc2028.c
index 878d2c4d9e8e..4be5cf808a40 100644
--- a/drivers/media/tuners/tuner-xc2028.c
+++ b/drivers/media/tuners/tuner-xc2028.c
@@ -24,6 +24,9 @@
 #include <linux/dvb/frontend.h>
 #include "dvb_frontend.h"
 
+/* Max transfer size done by I2C transfer functions */
+#define MAX_XFER_SIZE  80
+
 /* Registers (Write-only) */
 #define XREG_INIT         0x00
 #define XREG_RF_FREQ      0x02
@@ -547,7 +550,10 @@ static int load_firmware(struct dvb_frontend *fe, unsigned int type,
 {
 	struct xc2028_data *priv = fe->tuner_priv;
 	int                pos, rc;
-	unsigned char      *p, *endp, buf[priv->ctrl.max_len];
+	unsigned char      *p, *endp, buf[MAX_XFER_SIZE];
+
+	if (priv->ctrl.max_len > sizeof(buf))
+		priv->ctrl.max_len = sizeof(buf);
 
 	tuner_dbg("%s called\n", __func__);
 
@@ -572,7 +578,7 @@ static int load_firmware(struct dvb_frontend *fe, unsigned int type,
 			return -EINVAL;
 		}
 
-		size = le16_to_cpu(*(__u16 *) p);
+		size = le16_to_cpu(*(__le16 *) p);
 		p += sizeof(size);
 
 		if (size == 0xffff)
@@ -683,7 +689,7 @@ static int load_scode(struct dvb_frontend *fe, unsigned int type,
 		/* 16 SCODE entries per file; each SCODE entry is 12 bytes and
 		 * has a 2-byte size header in the firmware format. */
 		if (priv->firm[pos].size != 14 * 16 || scode >= 16 ||
-		    le16_to_cpu(*(__u16 *)(p + 14 * scode)) != 12)
+		    le16_to_cpu(*(__le16 *)(p + 14 * scode)) != 12)
 			return -EINVAL;
 		p += 14 * scode + 2;
 	}
diff --git a/drivers/media/usb/b2c2/flexcop-usb.c b/drivers/media/usb/b2c2/flexcop-usb.c
index 8b6275f85908..0bd969063392 100644
--- a/drivers/media/usb/b2c2/flexcop-usb.c
+++ b/drivers/media/usb/b2c2/flexcop-usb.c
@@ -390,7 +390,7 @@ static void flexcop_usb_transfer_exit(struct flexcop_usb *fc_usb)
 		}
 
 	if (fc_usb->iso_buffer != NULL)
-		pci_free_consistent(NULL,
+		usb_free_coherent(fc_usb->udev,
 			fc_usb->buffer_size, fc_usb->iso_buffer,
 			fc_usb->dma_addr);
 }
@@ -407,8 +407,8 @@ static int flexcop_usb_transfer_init(struct flexcop_usb *fc_usb)
 			"each of %d bytes size = %d.\n", B2C2_USB_NUM_ISO_URB,
 			B2C2_USB_FRAMES_PER_ISO, frame_size, bufsize);
 
-	fc_usb->iso_buffer = pci_alloc_consistent(NULL,
-			bufsize, &fc_usb->dma_addr);
+	fc_usb->iso_buffer = usb_alloc_coherent(fc_usb->udev,
+			bufsize, GFP_KERNEL, &fc_usb->dma_addr);
 	if (fc_usb->iso_buffer == NULL)
 		return -ENOMEM;
 
diff --git a/drivers/media/usb/cpia2/cpia2_usb.c b/drivers/media/usb/cpia2/cpia2_usb.c
index be1719283609..351a78a84c3d 100644
--- a/drivers/media/usb/cpia2/cpia2_usb.c
+++ b/drivers/media/usb/cpia2/cpia2_usb.c
@@ -209,7 +209,7 @@ static void cpia2_usb_complete(struct urb *urb)
 {
 	int i;
 	unsigned char *cdata;
-	static int frame_ready = false;
+	static bool frame_ready = false;
 	struct camera_data *cam = (struct camera_data *) urb->context;
 
 	if (urb->status!=0) {
diff --git a/drivers/media/usb/cx231xx/cx231xx-cards.c b/drivers/media/usb/cx231xx/cx231xx-cards.c
index a384f80f595e..e9d017bea377 100644
--- a/drivers/media/usb/cx231xx/cx231xx-cards.c
+++ b/drivers/media/usb/cx231xx/cx231xx-cards.c
@@ -978,7 +978,6 @@ static int cx231xx_init_dev(struct cx231xx *dev, struct usb_device *udev,
 			    int minor)
 {
 	int retval = -ENOMEM;
-	int errCode;
 	unsigned int maxh, maxw;
 
 	dev->udev = udev;
@@ -1014,8 +1013,8 @@ static int cx231xx_init_dev(struct cx231xx *dev, struct usb_device *udev,
 	/* Cx231xx pre card setup */
 	cx231xx_pre_card_setup(dev);
 
-	errCode = cx231xx_config(dev);
-	if (errCode) {
+	retval = cx231xx_config(dev);
+	if (retval) {
 		cx231xx_errdev("error configuring device\n");
 		return -ENOMEM;
 	}
@@ -1024,12 +1023,11 @@ static int cx231xx_init_dev(struct cx231xx *dev, struct usb_device *udev,
 	dev->norm = dev->board.norm;
 
 	/* register i2c bus */
-	errCode = cx231xx_dev_init(dev);
-	if (errCode < 0) {
-		cx231xx_dev_uninit(dev);
+	retval = cx231xx_dev_init(dev);
+	if (retval) {
 		cx231xx_errdev("%s: cx231xx_i2c_register - errCode [%d]!\n",
-			       __func__, errCode);
-		return errCode;
+			       __func__, retval);
+		goto err_dev_init;
 	}
 
 	/* Do board specific init */
@@ -1047,11 +1045,11 @@ static int cx231xx_init_dev(struct cx231xx *dev, struct usb_device *udev,
 	dev->interlaced = 0;
 	dev->video_input = 0;
 
-	errCode = cx231xx_config(dev);
-	if (errCode < 0) {
+	retval = cx231xx_config(dev);
+	if (retval) {
 		cx231xx_errdev("%s: cx231xx_config - errCode [%d]!\n",
-			       __func__, errCode);
-		return errCode;
+			       __func__, retval);
+		goto err_dev_init;
 	}
 
 	/* init video dma queues */
@@ -1075,9 +1073,9 @@ static int cx231xx_init_dev(struct cx231xx *dev, struct usb_device *udev,
 	}
 
 	retval = cx231xx_register_analog_devices(dev);
-	if (retval < 0) {
-		cx231xx_release_resources(dev);
-		return retval;
+	if (retval) {
+		cx231xx_release_analog_resources(dev);
+		goto err_analog;
 	}
 
 	cx231xx_ir_init(dev);
@@ -1085,6 +1083,11 @@ static int cx231xx_init_dev(struct cx231xx *dev, struct usb_device *udev,
 	cx231xx_init_extension(dev);
 
 	return 0;
+err_analog:
+	cx231xx_remove_from_devlist(dev);
+err_dev_init:
+	cx231xx_dev_uninit(dev);
+	return retval;
 }
 
 #if defined(CONFIG_MODULES) && defined(MODULE)
@@ -1132,7 +1135,6 @@ static int cx231xx_usb_probe(struct usb_interface *interface,
 	char *speed;
 	struct usb_interface_assoc_descriptor *assoc_desc;
 
-	udev = usb_get_dev(interface_to_usbdev(interface));
 	ifnum = interface->altsetting[0].desc.bInterfaceNumber;
 
 	/*
@@ -1161,6 +1163,8 @@ static int cx231xx_usb_probe(struct usb_interface *interface,
 		return -ENOMEM;
 	}
 
+	udev = usb_get_dev(interface_to_usbdev(interface));
+
 	snprintf(dev->name, 29, "cx231xx #%d", nr);
 	dev->devno = nr;
 	dev->model = id->driver_info;
@@ -1223,10 +1227,8 @@ static int cx231xx_usb_probe(struct usb_interface *interface,
 	if (assoc_desc->bFirstInterface != ifnum) {
 		cx231xx_err(DRIVER_NAME ": Not found "
 			    "matching IAD interface\n");
-		clear_bit(dev->devno, &cx231xx_devused);
-		kfree(dev);
-		dev = NULL;
-		return -ENODEV;
+		retval = -ENODEV;
+		goto err_if;
 	}
 
 	cx231xx_info("registering interface %d\n", ifnum);
@@ -1242,22 +1244,13 @@ static int cx231xx_usb_probe(struct usb_interface *interface,
 	retval = v4l2_device_register(&interface->dev, &dev->v4l2_dev);
 	if (retval) {
 		cx231xx_errdev("v4l2_device_register failed\n");
-		clear_bit(dev->devno, &cx231xx_devused);
-		kfree(dev);
-		dev = NULL;
-		return -EIO;
+		retval = -EIO;
+		goto err_v4l2;
 	}
 	/* allocate device struct */
 	retval = cx231xx_init_dev(dev, udev, nr);
-	if (retval) {
-		clear_bit(dev->devno, &cx231xx_devused);
-		v4l2_device_unregister(&dev->v4l2_dev);
-		kfree(dev);
-		dev = NULL;
-		usb_set_intfdata(interface, NULL);
-
-		return retval;
-	}
+	if (retval)
+		goto err_init;
 
 	/* compute alternate max packet sizes for video */
 	uif = udev->actconfig->interface[dev->current_pcb_config.
@@ -1275,11 +1268,8 @@ static int cx231xx_usb_probe(struct usb_interface *interface,
 
 	if (dev->video_mode.alt_max_pkt_size == NULL) {
 		cx231xx_errdev("out of memory!\n");
-		clear_bit(dev->devno, &cx231xx_devused);
-		v4l2_device_unregister(&dev->v4l2_dev);
-		kfree(dev);
-		dev = NULL;
-		return -ENOMEM;
+		retval = -ENOMEM;
+		goto err_video_alt;
 	}
 
 	for (i = 0; i < dev->video_mode.num_alt; i++) {
@@ -1309,11 +1299,8 @@ static int cx231xx_usb_probe(struct usb_interface *interface,
 
 	if (dev->vbi_mode.alt_max_pkt_size == NULL) {
 		cx231xx_errdev("out of memory!\n");
-		clear_bit(dev->devno, &cx231xx_devused);
-		v4l2_device_unregister(&dev->v4l2_dev);
-		kfree(dev);
-		dev = NULL;
-		return -ENOMEM;
+		retval = -ENOMEM;
+		goto err_vbi_alt;
 	}
 
 	for (i = 0; i < dev->vbi_mode.num_alt; i++) {
@@ -1344,11 +1331,8 @@ static int cx231xx_usb_probe(struct usb_interface *interface,
 
 	if (dev->sliced_cc_mode.alt_max_pkt_size == NULL) {
 		cx231xx_errdev("out of memory!\n");
-		clear_bit(dev->devno, &cx231xx_devused);
-		v4l2_device_unregister(&dev->v4l2_dev);
-		kfree(dev);
-		dev = NULL;
-		return -ENOMEM;
+		retval = -ENOMEM;
+		goto err_sliced_cc_alt;
 	}
 
 	for (i = 0; i < dev->sliced_cc_mode.num_alt; i++) {
@@ -1380,11 +1364,8 @@ static int cx231xx_usb_probe(struct usb_interface *interface,
 
 		if (dev->ts1_mode.alt_max_pkt_size == NULL) {
 			cx231xx_errdev("out of memory!\n");
-			clear_bit(dev->devno, &cx231xx_devused);
-			v4l2_device_unregister(&dev->v4l2_dev);
-			kfree(dev);
-			dev = NULL;
-			return -ENOMEM;
+			retval = -ENOMEM;
+			goto err_ts1_alt;
 		}
 
 		for (i = 0; i < dev->ts1_mode.num_alt; i++) {
@@ -1411,6 +1392,29 @@ static int cx231xx_usb_probe(struct usb_interface *interface,
 	request_modules(dev);
 
 	return 0;
+err_ts1_alt:
+	kfree(dev->sliced_cc_mode.alt_max_pkt_size);
+err_sliced_cc_alt:
+	kfree(dev->vbi_mode.alt_max_pkt_size);
+err_vbi_alt:
+	kfree(dev->video_mode.alt_max_pkt_size);
+err_video_alt:
+	/* cx231xx_uninit_dev: */
+	cx231xx_close_extension(dev);
+	cx231xx_ir_exit(dev);
+	cx231xx_release_analog_resources(dev);
+	cx231xx_417_unregister(dev);
+	cx231xx_remove_from_devlist(dev);
+	cx231xx_dev_uninit(dev);
+err_init:
+	v4l2_device_unregister(&dev->v4l2_dev);
+err_v4l2:
+	usb_set_intfdata(interface, NULL);
+err_if:
+	usb_put_dev(udev);
+	kfree(dev);
+	clear_bit(dev->devno, &cx231xx_devused);
+	return retval;
 }
 
 /*
diff --git a/drivers/media/usb/cx231xx/cx231xx-pcb-cfg.c b/drivers/media/usb/cx231xx/cx231xx-pcb-cfg.c
index d7308ab7a90f..2a34ceee4802 100644
--- a/drivers/media/usb/cx231xx/cx231xx-pcb-cfg.c
+++ b/drivers/media/usb/cx231xx/cx231xx-pcb-cfg.c
@@ -28,7 +28,7 @@ MODULE_PARM_DESC(pcb_debug, "enable pcb config debug messages [video]");
 
 /******************************************************************************/
 
-struct pcb_config cx231xx_Scenario[] = {
+static struct pcb_config cx231xx_Scenario[] = {
 	{
 	 INDEX_SELFPOWER_DIGITAL_ONLY,	/* index */
 	 USB_SELF_POWER,	/* power_type */
@@ -672,7 +672,7 @@ u32 initialize_cx231xx(struct cx231xx *dev)
 	pcb config it is related to */
 	cx231xx_read_ctrl_reg(dev, VRT_GET_REGISTER, BOARD_CFG_STAT, data, 4);
 
-	config_info = le32_to_cpu(*((u32 *) data));
+	config_info = le32_to_cpu(*((__le32 *)data));
 	usb_speed = (u8) (config_info & 0x1);
 
 	/* Verify this device belongs to Bus power or Self power device */
diff --git a/drivers/media/usb/dvb-usb-v2/af9015.c b/drivers/media/usb/dvb-usb-v2/af9015.c
index d556042cf312..da47d2392f2a 100644
--- a/drivers/media/usb/dvb-usb-v2/af9015.c
+++ b/drivers/media/usb/dvb-usb-v2/af9015.c
@@ -397,12 +397,13 @@ error:
 	return ret;
 }
 
+#define AF9015_EEPROM_SIZE 256
+
 /* hash (and dump) eeprom */
 static int af9015_eeprom_hash(struct dvb_usb_device *d)
 {
 	struct af9015_state *state = d_to_priv(d);
 	int ret, i;
-	static const unsigned int AF9015_EEPROM_SIZE = 256;
 	u8 buf[AF9015_EEPROM_SIZE];
 	struct req_t req = {READ_I2C, AF9015_I2C_EEPROM, 0, 0, 1, 1, NULL};
 
diff --git a/drivers/media/usb/dvb-usb-v2/af9035.c b/drivers/media/usb/dvb-usb-v2/af9035.c
index 1ea17dc2a76e..c8fcd78425bd 100644
--- a/drivers/media/usb/dvb-usb-v2/af9035.c
+++ b/drivers/media/usb/dvb-usb-v2/af9035.c
@@ -21,6 +21,9 @@
 
 #include "af9035.h"
 
+/* Max transfer size done by I2C transfer functions */
+#define MAX_XFER_SIZE  64
+
 DVB_DEFINE_MOD_OPT_ADAPTER_NR(adapter_nr);
 
 static u16 af9035_checksum(const u8 *buf, size_t len)
@@ -126,10 +129,16 @@ exit:
 /* write multiple registers */
 static int af9035_wr_regs(struct dvb_usb_device *d, u32 reg, u8 *val, int len)
 {
-	u8 wbuf[6 + len];
+	u8 wbuf[MAX_XFER_SIZE];
 	u8 mbox = (reg >> 16) & 0xff;
 	struct usb_req req = { CMD_MEM_WR, mbox, sizeof(wbuf), wbuf, 0, NULL };
 
+	if (6 + len > sizeof(wbuf)) {
+		dev_warn(&d->udev->dev, "%s: i2c wr: len=%d is too big!\n",
+			 KBUILD_MODNAME, len);
+		return -EOPNOTSUPP;
+	}
+
 	wbuf[0] = len;
 	wbuf[1] = 2;
 	wbuf[2] = 0;
@@ -228,9 +237,16 @@ static int af9035_i2c_master_xfer(struct i2c_adapter *adap,
 					msg[1].len);
 		} else {
 			/* I2C */
-			u8 buf[5 + msg[0].len];
+			u8 buf[MAX_XFER_SIZE];
 			struct usb_req req = { CMD_I2C_RD, 0, sizeof(buf),
 					buf, msg[1].len, msg[1].buf };
+
+			if (5 + msg[0].len > sizeof(buf)) {
+				dev_warn(&d->udev->dev,
+					 "%s: i2c xfer: len=%d is too big!\n",
+					 KBUILD_MODNAME, msg[0].len);
+				return -EOPNOTSUPP;
+			}
 			req.mbox |= ((msg[0].addr & 0x80)  >>  3);
 			buf[0] = msg[1].len;
 			buf[1] = msg[0].addr << 1;
@@ -257,9 +273,16 @@ static int af9035_i2c_master_xfer(struct i2c_adapter *adap,
 					msg[0].len - 3);
 		} else {
 			/* I2C */
-			u8 buf[5 + msg[0].len];
+			u8 buf[MAX_XFER_SIZE];
 			struct usb_req req = { CMD_I2C_WR, 0, sizeof(buf), buf,
 					0, NULL };
+
+			if (5 + msg[0].len > sizeof(buf)) {
+				dev_warn(&d->udev->dev,
+					 "%s: i2c xfer: len=%d is too big!\n",
+					 KBUILD_MODNAME, msg[0].len);
+				return -EOPNOTSUPP;
+			}
 			req.mbox |= ((msg[0].addr & 0x80)  >>  3);
 			buf[0] = msg[0].len;
 			buf[1] = msg[0].addr << 1;
diff --git a/drivers/media/usb/dvb-usb-v2/mxl111sf.c b/drivers/media/usb/dvb-usb-v2/mxl111sf.c
index e97964ef7f56..2627553f7de1 100644
--- a/drivers/media/usb/dvb-usb-v2/mxl111sf.c
+++ b/drivers/media/usb/dvb-usb-v2/mxl111sf.c
@@ -23,6 +23,9 @@
 #include "lgdt3305.h"
 #include "lg2160.h"
 
+/* Max transfer size done by I2C transfer functions */
+#define MAX_XFER_SIZE  64
+
 int dvb_usb_mxl111sf_debug;
 module_param_named(debug, dvb_usb_mxl111sf_debug, int, 0644);
 MODULE_PARM_DESC(debug, "set debugging level "
@@ -57,7 +60,12 @@ int mxl111sf_ctrl_msg(struct dvb_usb_device *d,
 {
 	int wo = (rbuf == NULL || rlen == 0); /* write-only */
 	int ret;
-	u8 sndbuf[1+wlen];
+	u8 sndbuf[MAX_XFER_SIZE];
+
+	if (1 + wlen > sizeof(sndbuf)) {
+		pr_warn("%s: len=%d is too big!\n", __func__, wlen);
+		return -EOPNOTSUPP;
+	}
 
 	pr_debug("%s(wlen = %d, rlen = %d)\n", __func__, wlen, rlen);
 
diff --git a/drivers/media/usb/dvb-usb-v2/rtl28xxu.c b/drivers/media/usb/dvb-usb-v2/rtl28xxu.c
index c0cd0848631b..ecca03667f98 100644
--- a/drivers/media/usb/dvb-usb-v2/rtl28xxu.c
+++ b/drivers/media/usb/dvb-usb-v2/rtl28xxu.c
@@ -377,6 +377,7 @@ static int rtl2832u_read_config(struct dvb_usb_device *d)
 	struct rtl28xxu_req req_e4000 = {0x02c8, CMD_I2C_RD, 1, buf};
 	struct rtl28xxu_req req_tda18272 = {0x00c0, CMD_I2C_RD, 2, buf};
 	struct rtl28xxu_req req_r820t = {0x0034, CMD_I2C_RD, 1, buf};
+	struct rtl28xxu_req req_r828d = {0x0074, CMD_I2C_RD, 1, buf};
 
 	dev_dbg(&d->udev->dev, "%s:\n", __func__);
 
@@ -489,6 +490,15 @@ static int rtl2832u_read_config(struct dvb_usb_device *d)
 		goto found;
 	}
 
+	/* check R828D ID register; reg=00 val=69 */
+	ret = rtl28xxu_ctrl_msg(d, &req_r828d);
+	if (ret == 0 && buf[0] == 0x69) {
+		priv->tuner = TUNER_RTL2832_R828D;
+		priv->tuner_name = "R828D";
+		goto found;
+	}
+
+
 found:
 	dev_dbg(&d->udev->dev, "%s: tuner=%s\n", __func__, priv->tuner_name);
 
@@ -745,6 +755,7 @@ static int rtl2832u_frontend_attach(struct dvb_usb_adapter *adap)
 		rtl2832_config = &rtl28xxu_rtl2832_e4000_config;
 		break;
 	case TUNER_RTL2832_R820T:
+	case TUNER_RTL2832_R828D:
 		rtl2832_config = &rtl28xxu_rtl2832_r820t_config;
 		break;
 	default:
@@ -866,6 +877,13 @@ static const struct r820t_config rtl2832u_r820t_config = {
 	.rafael_chip = CHIP_R820T,
 };
 
+static const struct r820t_config rtl2832u_r828d_config = {
+	.i2c_addr = 0x3a,
+	.xtal = 16000000,
+	.max_i2c_msg_len = 2,
+	.rafael_chip = CHIP_R828D,
+};
+
 static int rtl2832u_tuner_attach(struct dvb_usb_adapter *adap)
 {
 	int ret;
@@ -923,6 +941,27 @@ static int rtl2832u_tuner_attach(struct dvb_usb_adapter *adap)
 		adap->fe[0]->ops.read_signal_strength =
 				adap->fe[0]->ops.tuner_ops.get_rf_strength;
 		break;
+	case TUNER_RTL2832_R828D:
+		/* power off mn88472 demod on GPIO0 */
+		ret = rtl28xx_wr_reg_mask(d, SYS_GPIO_OUT_VAL, 0x00, 0x01);
+		if (ret)
+			goto err;
+
+		ret = rtl28xx_wr_reg_mask(d, SYS_GPIO_DIR, 0x00, 0x01);
+		if (ret)
+			goto err;
+
+		ret = rtl28xx_wr_reg_mask(d, SYS_GPIO_OUT_EN, 0x01, 0x01);
+		if (ret)
+			goto err;
+
+		fe = dvb_attach(r820t_attach, adap->fe[0], &d->i2c_adap,
+				&rtl2832u_r828d_config);
+
+		/* Use tuner to get the signal strength */
+		adap->fe[0]->ops.read_signal_strength =
+				adap->fe[0]->ops.tuner_ops.get_rf_strength;
+		break;
 	default:
 		fe = NULL;
 		dev_err(&d->udev->dev, "%s: unknown tuner=%d\n", KBUILD_MODNAME,
@@ -1388,6 +1427,9 @@ static const struct usb_device_id rtl28xxu_id_table[] = {
 		&rtl2832u_props, "Leadtek WinFast DTV Dongle mini", NULL) },
 	{ DVB_USB_DEVICE(USB_VID_GTEK, USB_PID_CPYTO_REDI_PC50A,
 		&rtl2832u_props, "Crypto ReDi PC 50 A", NULL) },
+
+	{ DVB_USB_DEVICE(USB_VID_HANFTEK, 0x0131,
+		&rtl2832u_props, "Astrometa DVB-T2", NULL) },
 	{ }
 };
 MODULE_DEVICE_TABLE(usb, rtl28xxu_id_table);
diff --git a/drivers/media/usb/dvb-usb-v2/rtl28xxu.h b/drivers/media/usb/dvb-usb-v2/rtl28xxu.h
index 729b3540c2f9..2142bcb41b41 100644
--- a/drivers/media/usb/dvb-usb-v2/rtl28xxu.h
+++ b/drivers/media/usb/dvb-usb-v2/rtl28xxu.h
@@ -83,6 +83,7 @@ enum rtl28xxu_tuner {
 	TUNER_RTL2832_TDA18272,
 	TUNER_RTL2832_FC0013,
 	TUNER_RTL2832_R820T,
+	TUNER_RTL2832_R828D,
 };
 
 struct rtl28xxu_req {
diff --git a/drivers/media/usb/dvb-usb/az6027.c b/drivers/media/usb/dvb-usb/az6027.c
index ea2d5ee86576..c11138ebf6fb 100644
--- a/drivers/media/usb/dvb-usb/az6027.c
+++ b/drivers/media/usb/dvb-usb/az6027.c
@@ -254,7 +254,7 @@ static const struct stb0899_s1_reg az6027_stb0899_s1_init_3[] = {
 
 
 
-struct stb0899_config az6027_stb0899_config = {
+static struct stb0899_config az6027_stb0899_config = {
 	.init_dev		= az6027_stb0899_s1_init_1,
 	.init_s2_demod		= stb0899_s2_init_2,
 	.init_s1_demod		= az6027_stb0899_s1_init_3,
@@ -291,7 +291,7 @@ struct stb0899_config az6027_stb0899_config = {
 	.tuner_set_rfsiggain	= NULL,
 };
 
-struct stb6100_config az6027_stb6100_config = {
+static struct stb6100_config az6027_stb6100_config = {
 	.tuner_address	= 0xc0,
 	.refclock	= 27000000,
 };
diff --git a/drivers/media/usb/dvb-usb/cxusb.c b/drivers/media/usb/dvb-usb/cxusb.c
index 3940bb0f9ef6..20e345d9fe8f 100644
--- a/drivers/media/usb/dvb-usb/cxusb.c
+++ b/drivers/media/usb/dvb-usb/cxusb.c
@@ -43,6 +43,9 @@
 #include "lgs8gxx.h"
 #include "atbm8830.h"
 
+/* Max transfer size done by I2C transfer functions */
+#define MAX_XFER_SIZE  64
+
 /* debug */
 static int dvb_usb_cxusb_debug;
 module_param_named(debug, dvb_usb_cxusb_debug, int, 0644);
@@ -57,7 +60,14 @@ static int cxusb_ctrl_msg(struct dvb_usb_device *d,
 			  u8 cmd, u8 *wbuf, int wlen, u8 *rbuf, int rlen)
 {
 	int wo = (rbuf == NULL || rlen == 0); /* write-only */
-	u8 sndbuf[1+wlen];
+	u8 sndbuf[MAX_XFER_SIZE];
+
+	if (1 + wlen > sizeof(sndbuf)) {
+		warn("i2c wr: len=%d is too big!\n",
+		     wlen);
+		return -EOPNOTSUPP;
+	}
+
 	memset(sndbuf, 0, 1+wlen);
 
 	sndbuf[0] = cmd;
@@ -158,7 +168,13 @@ static int cxusb_i2c_xfer(struct i2c_adapter *adap, struct i2c_msg msg[],
 
 		if (msg[i].flags & I2C_M_RD) {
 			/* read only */
-			u8 obuf[3], ibuf[1+msg[i].len];
+			u8 obuf[3], ibuf[MAX_XFER_SIZE];
+
+			if (1 + msg[i].len > sizeof(ibuf)) {
+				warn("i2c rd: len=%d is too big!\n",
+				     msg[i].len);
+				return -EOPNOTSUPP;
+			}
 			obuf[0] = 0;
 			obuf[1] = msg[i].len;
 			obuf[2] = msg[i].addr;
@@ -172,7 +188,18 @@ static int cxusb_i2c_xfer(struct i2c_adapter *adap, struct i2c_msg msg[],
 		} else if (i+1 < num && (msg[i+1].flags & I2C_M_RD) &&
 			   msg[i].addr == msg[i+1].addr) {
 			/* write to then read from same address */
-			u8 obuf[3+msg[i].len], ibuf[1+msg[i+1].len];
+			u8 obuf[MAX_XFER_SIZE], ibuf[MAX_XFER_SIZE];
+
+			if (3 + msg[i].len > sizeof(obuf)) {
+				warn("i2c wr: len=%d is too big!\n",
+				     msg[i].len);
+				return -EOPNOTSUPP;
+			}
+			if (1 + msg[i + 1].len > sizeof(ibuf)) {
+				warn("i2c rd: len=%d is too big!\n",
+				     msg[i + 1].len);
+				return -EOPNOTSUPP;
+			}
 			obuf[0] = msg[i].len;
 			obuf[1] = msg[i+1].len;
 			obuf[2] = msg[i].addr;
@@ -191,7 +218,13 @@ static int cxusb_i2c_xfer(struct i2c_adapter *adap, struct i2c_msg msg[],
 			i++;
 		} else {
 			/* write only */
-			u8 obuf[2+msg[i].len], ibuf;
+			u8 obuf[MAX_XFER_SIZE], ibuf;
+
+			if (2 + msg[i].len > sizeof(obuf)) {
+				warn("i2c wr: len=%d is too big!\n",
+				     msg[i].len);
+				return -EOPNOTSUPP;
+			}
 			obuf[0] = msg[i].addr;
 			obuf[1] = msg[i].len;
 			memcpy(&obuf[2], msg[i].buf, msg[i].len);
diff --git a/drivers/media/usb/dvb-usb/dibusb-common.c b/drivers/media/usb/dvb-usb/dibusb-common.c
index c2dded92f1d3..6d68af0c49c8 100644
--- a/drivers/media/usb/dvb-usb/dibusb-common.c
+++ b/drivers/media/usb/dvb-usb/dibusb-common.c
@@ -12,6 +12,9 @@
 #include <linux/kconfig.h>
 #include "dibusb.h"
 
+/* Max transfer size done by I2C transfer functions */
+#define MAX_XFER_SIZE  64
+
 static int debug;
 module_param(debug, int, 0644);
 MODULE_PARM_DESC(debug, "set debugging level (1=info (|-able))." DVB_USB_DEBUG_STATUS);
@@ -105,11 +108,16 @@ EXPORT_SYMBOL(dibusb2_0_power_ctrl);
 static int dibusb_i2c_msg(struct dvb_usb_device *d, u8 addr,
 			  u8 *wbuf, u16 wlen, u8 *rbuf, u16 rlen)
 {
-	u8 sndbuf[wlen+4]; /* lead(1) devaddr,direction(1) addr(2) data(wlen) (len(2) (when reading)) */
+	u8 sndbuf[MAX_XFER_SIZE]; /* lead(1) devaddr,direction(1) addr(2) data(wlen) (len(2) (when reading)) */
 	/* write only ? */
 	int wo = (rbuf == NULL || rlen == 0),
 		len = 2 + wlen + (wo ? 0 : 2);
 
+	if (4 + wlen > sizeof(sndbuf)) {
+		warn("i2c wr: len=%d is too big!\n", wlen);
+		return -EOPNOTSUPP;
+	}
+
 	sndbuf[0] = wo ? DIBUSB_REQ_I2C_WRITE : DIBUSB_REQ_I2C_READ;
 	sndbuf[1] = (addr << 1) | (wo ? 0 : 1);
 
diff --git a/drivers/media/usb/dvb-usb/dw2102.c b/drivers/media/usb/dvb-usb/dw2102.c
index 6e237b6dd0a8..c1a63b2a6baa 100644
--- a/drivers/media/usb/dvb-usb/dw2102.c
+++ b/drivers/media/usb/dvb-usb/dw2102.c
@@ -30,6 +30,9 @@
 #include "stb6100_proc.h"
 #include "m88rs2000.h"
 
+/* Max transfer size done by I2C transfer functions */
+#define MAX_XFER_SIZE  64
+
 #ifndef USB_PID_DW2102
 #define USB_PID_DW2102 0x2102
 #endif
@@ -308,7 +311,14 @@ static int dw2102_earda_i2c_transfer(struct i2c_adapter *adap, struct i2c_msg ms
 	case 2: {
 		/* read */
 		/* first write first register number */
-		u8 ibuf[msg[1].len + 2], obuf[3];
+		u8 ibuf[MAX_XFER_SIZE], obuf[3];
+
+		if (2 + msg[1].len > sizeof(ibuf)) {
+			warn("i2c rd: len=%d is too big!\n",
+			     msg[1].len);
+			return -EOPNOTSUPP;
+		}
+
 		obuf[0] = msg[0].addr << 1;
 		obuf[1] = msg[0].len;
 		obuf[2] = msg[0].buf[0];
@@ -325,7 +335,14 @@ static int dw2102_earda_i2c_transfer(struct i2c_adapter *adap, struct i2c_msg ms
 		switch (msg[0].addr) {
 		case 0x68: {
 			/* write to register */
-			u8 obuf[msg[0].len + 2];
+			u8 obuf[MAX_XFER_SIZE];
+
+			if (2 + msg[0].len > sizeof(obuf)) {
+				warn("i2c wr: len=%d is too big!\n",
+				     msg[1].len);
+				return -EOPNOTSUPP;
+			}
+
 			obuf[0] = msg[0].addr << 1;
 			obuf[1] = msg[0].len;
 			memcpy(obuf + 2, msg[0].buf, msg[0].len);
@@ -335,7 +352,14 @@ static int dw2102_earda_i2c_transfer(struct i2c_adapter *adap, struct i2c_msg ms
 		}
 		case 0x61: {
 			/* write to tuner */
-			u8 obuf[msg[0].len + 2];
+			u8 obuf[MAX_XFER_SIZE];
+
+			if (2 + msg[0].len > sizeof(obuf)) {
+				warn("i2c wr: len=%d is too big!\n",
+				     msg[1].len);
+				return -EOPNOTSUPP;
+			}
+
 			obuf[0] = msg[0].addr << 1;
 			obuf[1] = msg[0].len;
 			memcpy(obuf + 2, msg[0].buf, msg[0].len);
@@ -401,7 +425,14 @@ static int dw2104_i2c_transfer(struct i2c_adapter *adap, struct i2c_msg msg[], i
 		default: {
 			if (msg[j].flags == I2C_M_RD) {
 				/* read registers */
-				u8  ibuf[msg[j].len + 2];
+				u8  ibuf[MAX_XFER_SIZE];
+
+				if (2 + msg[j].len > sizeof(ibuf)) {
+					warn("i2c rd: len=%d is too big!\n",
+					     msg[j].len);
+					return -EOPNOTSUPP;
+				}
+
 				dw210x_op_rw(d->udev, 0xc3,
 						(msg[j].addr << 1) + 1, 0,
 						ibuf, msg[j].len + 2,
@@ -430,7 +461,14 @@ static int dw2104_i2c_transfer(struct i2c_adapter *adap, struct i2c_msg msg[], i
 				} while (len > 0);
 			} else {
 				/* write registers */
-				u8 obuf[msg[j].len + 2];
+				u8 obuf[MAX_XFER_SIZE];
+
+				if (2 + msg[j].len > sizeof(obuf)) {
+					warn("i2c wr: len=%d is too big!\n",
+					     msg[j].len);
+					return -EOPNOTSUPP;
+				}
+
 				obuf[0] = msg[j].addr << 1;
 				obuf[1] = msg[j].len;
 				memcpy(obuf + 2, msg[j].buf, msg[j].len);
@@ -463,7 +501,13 @@ static int dw3101_i2c_transfer(struct i2c_adapter *adap, struct i2c_msg msg[],
 	case 2: {
 		/* read */
 		/* first write first register number */
-		u8 ibuf[msg[1].len + 2], obuf[3];
+		u8 ibuf[MAX_XFER_SIZE], obuf[3];
+
+		if (2 + msg[1].len > sizeof(ibuf)) {
+			warn("i2c rd: len=%d is too big!\n",
+			     msg[1].len);
+			return -EOPNOTSUPP;
+		}
 		obuf[0] = msg[0].addr << 1;
 		obuf[1] = msg[0].len;
 		obuf[2] = msg[0].buf[0];
@@ -481,7 +525,13 @@ static int dw3101_i2c_transfer(struct i2c_adapter *adap, struct i2c_msg msg[],
 		case 0x60:
 		case 0x0c: {
 			/* write to register */
-			u8 obuf[msg[0].len + 2];
+			u8 obuf[MAX_XFER_SIZE];
+
+			if (2 + msg[0].len > sizeof(obuf)) {
+				warn("i2c wr: len=%d is too big!\n",
+				     msg[0].len);
+				return -EOPNOTSUPP;
+			}
 			obuf[0] = msg[0].addr << 1;
 			obuf[1] = msg[0].len;
 			memcpy(obuf + 2, msg[0].buf, msg[0].len);
@@ -563,7 +613,14 @@ static int s6x0_i2c_transfer(struct i2c_adapter *adap, struct i2c_msg msg[],
 		default: {
 			if (msg[j].flags == I2C_M_RD) {
 				/* read registers */
-				u8 ibuf[msg[j].len];
+				u8 ibuf[MAX_XFER_SIZE];
+
+				if (msg[j].len > sizeof(ibuf)) {
+					warn("i2c rd: len=%d is too big!\n",
+					     msg[j].len);
+					return -EOPNOTSUPP;
+				}
+
 				dw210x_op_rw(d->udev, 0x91, 0, 0,
 						ibuf, msg[j].len,
 						DW210X_READ_MSG);
@@ -590,7 +647,14 @@ static int s6x0_i2c_transfer(struct i2c_adapter *adap, struct i2c_msg msg[],
 				} while (len > 0);
 			} else if (j < (num - 1)) {
 				/* write register addr before read */
-				u8 obuf[msg[j].len + 2];
+				u8 obuf[MAX_XFER_SIZE];
+
+				if (2 + msg[j].len > sizeof(obuf)) {
+					warn("i2c wr: len=%d is too big!\n",
+					     msg[j].len);
+					return -EOPNOTSUPP;
+				}
+
 				obuf[0] = msg[j + 1].len;
 				obuf[1] = (msg[j].addr << 1);
 				memcpy(obuf + 2, msg[j].buf, msg[j].len);
@@ -602,7 +666,13 @@ static int s6x0_i2c_transfer(struct i2c_adapter *adap, struct i2c_msg msg[],
 				break;
 			} else {
 				/* write registers */
-				u8 obuf[msg[j].len + 2];
+				u8 obuf[MAX_XFER_SIZE];
+
+				if (2 + msg[j].len > sizeof(obuf)) {
+					warn("i2c wr: len=%d is too big!\n",
+					     msg[j].len);
+					return -EOPNOTSUPP;
+				}
 				obuf[0] = msg[j].len + 1;
 				obuf[1] = (msg[j].addr << 1);
 				memcpy(obuf + 2, msg[j].buf, msg[j].len);
@@ -955,9 +1025,10 @@ static struct ds3000_config dw2104_ds3000_config = {
 	.demod_address = 0x68,
 };
 
-static struct ts2020_config dw2104_ts2020_config  = {
+static struct ts2020_config dw2104_ts2020_config = {
 	.tuner_address = 0x60,
 	.clk_out_div = 1,
+	.frequency_div = 1060000,
 };
 
 static struct ds3000_config s660_ds3000_config = {
@@ -966,6 +1037,12 @@ static struct ds3000_config s660_ds3000_config = {
 	.set_lock_led = dw210x_led_ctrl,
 };
 
+static struct ts2020_config s660_ts2020_config = {
+	.tuner_address = 0x60,
+	.clk_out_div = 1,
+	.frequency_div = 1146000,
+};
+
 static struct stv0900_config dw2104a_stv0900_config = {
 	.demod_address = 0x6a,
 	.demod_mode = 0,
@@ -1205,7 +1282,7 @@ static int ds3000_frontend_attach(struct dvb_usb_adapter *d)
 	if (d->fe_adap[0].fe == NULL)
 		return -EIO;
 
-	dvb_attach(ts2020_attach, d->fe_adap[0].fe, &dw2104_ts2020_config,
+	dvb_attach(ts2020_attach, d->fe_adap[0].fe, &s660_ts2020_config,
 		&d->dev->i2c_adap);
 
 	st->old_set_voltage = d->fe_adap[0].fe->ops.set_voltage;
@@ -1213,7 +1290,7 @@ static int ds3000_frontend_attach(struct dvb_usb_adapter *d)
 
 	dw210x_op_rw(d->dev->udev, 0x8a, 0, 0, obuf, 2, DW210X_WRITE_MSG);
 
-	info("Attached ds3000+ds2020!\n");
+	info("Attached ds3000+ts2020!\n");
 
 	return 0;
 }
diff --git a/drivers/media/usb/em28xx/em28xx-camera.c b/drivers/media/usb/em28xx/em28xx-camera.c
index 73cc50afa5e1..d666741797d4 100644
--- a/drivers/media/usb/em28xx/em28xx-camera.c
+++ b/drivers/media/usb/em28xx/em28xx-camera.c
@@ -22,6 +22,7 @@
 #include <linux/i2c.h>
 #include <media/soc_camera.h>
 #include <media/mt9v011.h>
+#include <media/v4l2-clk.h>
 #include <media/v4l2-common.h>
 
 #include "em28xx.h"
@@ -47,6 +48,7 @@ static struct soc_camera_link camlink = {
 	.bus_id = 0,
 	.flags = 0,
 	.module_name = "em28xx",
+	.unbalanced_power = true,
 };
 
 
@@ -325,13 +327,24 @@ int em28xx_detect_sensor(struct em28xx *dev)
 
 int em28xx_init_camera(struct em28xx *dev)
 {
+	char clk_name[V4L2_SUBDEV_NAME_SIZE];
+	struct i2c_client *client = &dev->i2c_client[dev->def_i2c_bus];
+	struct i2c_adapter *adap = &dev->i2c_adap[dev->def_i2c_bus];
+	int ret = 0;
+
+	v4l2_clk_name_i2c(clk_name, sizeof(clk_name),
+			  i2c_adapter_id(adap), client->addr);
+	dev->clk = v4l2_clk_register_fixed(clk_name, "mclk", -EINVAL);
+	if (IS_ERR(dev->clk))
+		return PTR_ERR(dev->clk);
+
 	switch (dev->em28xx_sensor) {
 	case EM28XX_MT9V011:
 	{
 		struct mt9v011_platform_data pdata;
 		struct i2c_board_info mt9v011_info = {
 			.type = "mt9v011",
-			.addr = dev->i2c_client[dev->def_i2c_bus].addr,
+			.addr = client->addr,
 			.platform_data = &pdata,
 		};
 
@@ -352,10 +365,11 @@ int em28xx_init_camera(struct em28xx *dev)
 		dev->sensor_xtal = 4300000;
 		pdata.xtal = dev->sensor_xtal;
 		if (NULL ==
-		    v4l2_i2c_new_subdev_board(&dev->v4l2_dev,
-					      &dev->i2c_adap[dev->def_i2c_bus],
-					      &mt9v011_info, NULL))
-			return -ENODEV;
+		    v4l2_i2c_new_subdev_board(&dev->v4l2_dev, adap,
+					      &mt9v011_info, NULL)) {
+			ret = -ENODEV;
+			break;
+		}
 		/* probably means GRGB 16 bit bayer */
 		dev->vinmode = 0x0d;
 		dev->vinctl = 0x00;
@@ -391,7 +405,7 @@ int em28xx_init_camera(struct em28xx *dev)
 		struct i2c_board_info ov2640_info = {
 			.type = "ov2640",
 			.flags = I2C_CLIENT_SCCB,
-			.addr = dev->i2c_client[dev->def_i2c_bus].addr,
+			.addr = client->addr,
 			.platform_data = &camlink,
 		};
 		struct v4l2_mbus_framefmt fmt;
@@ -408,9 +422,12 @@ int em28xx_init_camera(struct em28xx *dev)
 		dev->sensor_yres = 480;
 
 		subdev =
-		     v4l2_i2c_new_subdev_board(&dev->v4l2_dev,
-					       &dev->i2c_adap[dev->def_i2c_bus],
+		     v4l2_i2c_new_subdev_board(&dev->v4l2_dev, adap,
 					       &ov2640_info, NULL);
+		if (NULL == subdev) {
+			ret = -ENODEV;
+			break;
+		}
 
 		fmt.code = V4L2_MBUS_FMT_YUYV8_2X8;
 		fmt.width = 640;
@@ -427,8 +444,13 @@ int em28xx_init_camera(struct em28xx *dev)
 	}
 	case EM28XX_NOSENSOR:
 	default:
-		return -EINVAL;
+		ret = -EINVAL;
 	}
 
-	return 0;
+	if (ret < 0) {
+		v4l2_clk_unregister_fixed(dev->clk);
+		dev->clk = NULL;
+	}
+
+	return ret;
 }
diff --git a/drivers/media/usb/em28xx/em28xx-cards.c b/drivers/media/usb/em28xx/em28xx-cards.c
index dc65742c4bbc..a5196697627f 100644
--- a/drivers/media/usb/em28xx/em28xx-cards.c
+++ b/drivers/media/usb/em28xx/em28xx-cards.c
@@ -36,6 +36,7 @@
 #include <media/tvaudio.h>
 #include <media/i2c-addr.h>
 #include <media/tveeprom.h>
+#include <media/v4l2-clk.h>
 #include <media/v4l2-common.h>
 
 #include "em28xx.h"
@@ -95,8 +96,8 @@ static struct em28xx_reg_seq default_digital[] = {
 /* Board Hauppauge WinTV HVR 900 analog */
 static struct em28xx_reg_seq hauppauge_wintv_hvr_900_analog[] = {
 	{EM2820_R08_GPIO_CTRL,	0x2d,	~EM_GPIO_4,	10},
-	{0x05,			0xff,	0x10,		10},
-	{  -1,			-1,	-1,		-1},
+	{	0x05,		0xff,	0x10,		10},
+	{	-1,		-1,	-1,		-1},
 };
 
 /* Board Hauppauge WinTV HVR 900 digital */
@@ -104,20 +105,20 @@ static struct em28xx_reg_seq hauppauge_wintv_hvr_900_digital[] = {
 	{EM2820_R08_GPIO_CTRL,	0x2e,	~EM_GPIO_4,	10},
 	{EM2880_R04_GPO,	0x04,	0x0f,		10},
 	{EM2880_R04_GPO,	0x0c,	0x0f,		10},
-	{ -1,			-1,	-1,		-1},
+	{	-1,		-1,	-1,		-1},
 };
 
 /* Board Hauppauge WinTV HVR 900 (R2) digital */
 static struct em28xx_reg_seq hauppauge_wintv_hvr_900R2_digital[] = {
 	{EM2820_R08_GPIO_CTRL,	0x2e,	~EM_GPIO_4,	10},
 	{EM2880_R04_GPO,	0x0c,	0x0f,		10},
-	{ -1,			-1,	-1,		-1},
+	{	-1,		-1,	-1,		-1},
 };
 
 /* Boards - EM2880 MSI DIGIVOX AD and EM2880_BOARD_MSI_DIGIVOX_AD_II */
 static struct em28xx_reg_seq em2880_msi_digivox_ad_analog[] = {
-	{EM2820_R08_GPIO_CTRL,       0x69,   ~EM_GPIO_4,	 10},
-	{	-1,		-1,	-1,		 -1},
+	{EM2820_R08_GPIO_CTRL,	0x69,   ~EM_GPIO_4,	10},
+	{	-1,		-1,	-1,		-1},
 };
 
 /* Boards - EM2880 MSI DIGIVOX AD and EM2880_BOARD_MSI_DIGIVOX_AD_II */
@@ -132,7 +133,7 @@ static struct em28xx_reg_seq em2882_kworld_315u_digital[] = {
 	{EM2880_R04_GPO,	0x04,	0xff,		10},
 	{EM2880_R04_GPO,	0x0c,	0xff,		10},
 	{EM2820_R08_GPIO_CTRL,	0x7e,	0xff,		10},
-	{  -1,			-1,	-1,		-1},
+	{	-1,		-1,	-1,		-1},
 };
 
 static struct em28xx_reg_seq em2882_kworld_315u_tuner_gpio[] = {
@@ -140,19 +141,19 @@ static struct em28xx_reg_seq em2882_kworld_315u_tuner_gpio[] = {
 	{EM2880_R04_GPO,	0x0c,	0xff,		10},
 	{EM2880_R04_GPO,	0x08,	0xff,		10},
 	{EM2880_R04_GPO,	0x0c,	0xff,		10},
-	{  -1,			-1,	-1,		-1},
+	{	-1,		-1,	-1,		-1},
 };
 
 static struct em28xx_reg_seq kworld_330u_analog[] = {
 	{EM2820_R08_GPIO_CTRL,	0x6d,	~EM_GPIO_4,	10},
 	{EM2880_R04_GPO,	0x00,	0xff,		10},
-	{ -1,			-1,	-1,		-1},
+	{	-1,		-1,	-1,		-1},
 };
 
 static struct em28xx_reg_seq kworld_330u_digital[] = {
 	{EM2820_R08_GPIO_CTRL,	0x6e,	~EM_GPIO_4,	10},
 	{EM2880_R04_GPO,	0x08,	0xff,		10},
-	{ -1,			-1,	-1,		-1},
+	{	-1,		-1,	-1,		-1},
 };
 
 /* Evga inDtube
@@ -170,11 +171,11 @@ static struct em28xx_reg_seq evga_indtube_digital[] = {
 	{EM2820_R08_GPIO_CTRL,	0x7a,	0xff,		 1},
 	{EM2880_R04_GPO,	0x04,	0xff,		10},
 	{EM2880_R04_GPO,	0x0c,	0xff,		 1},
-	{ -1,			-1,	-1,		-1},
+	{	-1,		-1,	-1,		-1},
 };
 
 /*
- * KWorld PlusTV 340U and UB435-Q (ATSC) GPIOs map:
+ * KWorld PlusTV 340U, UB435-Q and UB435-Q V2 (ATSC) GPIOs map:
  * EM_GPIO_0 - currently unknown
  * EM_GPIO_1 - LED disable/enable (1 = off, 0 = on)
  * EM_GPIO_2 - currently unknown
@@ -185,8 +186,8 @@ static struct em28xx_reg_seq evga_indtube_digital[] = {
  * EM_GPIO_7 - currently unknown
  */
 static struct em28xx_reg_seq kworld_a340_digital[] = {
-	{EM2820_R08_GPIO_CTRL,	0x6d,		~EM_GPIO_4,	10},
-	{ -1,			-1,		-1,		-1},
+	{EM2820_R08_GPIO_CTRL,	0x6d,	~EM_GPIO_4,	10},
+	{	-1,		-1,	-1,		-1},
 };
 
 /* Pinnacle Hybrid Pro eb1a:2881 */
@@ -205,13 +206,13 @@ static struct em28xx_reg_seq pinnacle_hybrid_pro_digital[] = {
 static struct em28xx_reg_seq terratec_cinergy_USB_XS_FR_analog[] = {
 	{EM2820_R08_GPIO_CTRL,	0x6d,	~EM_GPIO_4,	10},
 	{EM2880_R04_GPO,	0x00,	0xff,		10},
-	{ -1,			-1,	-1,		-1},
+	{	-1,		-1,	-1,		-1},
 };
 
 static struct em28xx_reg_seq terratec_cinergy_USB_XS_FR_digital[] = {
 	{EM2820_R08_GPIO_CTRL,	0x6e,	~EM_GPIO_4,	10},
 	{EM2880_R04_GPO,	0x08,	0xff,		10},
-	{ -1,			-1,	-1,		-1},
+	{	-1,		-1,	-1,		-1},
 };
 
 /* eb1a:2868 Reddo DVB-C USB TV Box
@@ -225,7 +226,7 @@ static struct em28xx_reg_seq reddo_dvb_c_usb_box[] = {
 	{EM2820_R08_GPIO_CTRL,	0x7f,	0xff,		10},
 	{EM2820_R08_GPIO_CTRL,	0x6f,	0xff,		10},
 	{EM2820_R08_GPIO_CTRL,	0xff,	0xff,		10},
-	{-1,			-1,	-1,		-1},
+	{	-1,		-1,	-1,		-1},
 };
 
 /* Callback for the most boards */
@@ -233,23 +234,23 @@ static struct em28xx_reg_seq default_tuner_gpio[] = {
 	{EM2820_R08_GPIO_CTRL,	EM_GPIO_4,	EM_GPIO_4,	10},
 	{EM2820_R08_GPIO_CTRL,	0,		EM_GPIO_4,	10},
 	{EM2820_R08_GPIO_CTRL,	EM_GPIO_4,	EM_GPIO_4,	10},
-	{  -1,			-1,		-1,		-1},
+	{	-1,		-1,		-1,		-1},
 };
 
 /* Mute/unmute */
 static struct em28xx_reg_seq compro_unmute_tv_gpio[] = {
-	{EM2820_R08_GPIO_CTRL,	5,		7,		10},
-	{  -1,			-1,		-1,		-1},
+	{EM2820_R08_GPIO_CTRL,	5,	7,	10},
+	{	-1,		-1,	-1,	-1},
 };
 
 static struct em28xx_reg_seq compro_unmute_svid_gpio[] = {
-	{EM2820_R08_GPIO_CTRL,	4,		7,		10},
-	{  -1,			-1,		-1,		-1},
+	{EM2820_R08_GPIO_CTRL,	4,	7,	10},
+	{	-1,		-1,	-1,	-1},
 };
 
 static struct em28xx_reg_seq compro_mute_gpio[] = {
-	{EM2820_R08_GPIO_CTRL,	6,		7,		10},
-	{  -1,			-1,		-1,		-1},
+	{EM2820_R08_GPIO_CTRL,	6,	7,	10},
+	{	-1,		-1,	-1,	-1},
 };
 
 /* Terratec AV350 */
@@ -279,21 +280,21 @@ static struct em28xx_reg_seq vc211a_enable[] = {
 static struct em28xx_reg_seq dikom_dk300_digital[] = {
 	{EM2820_R08_GPIO_CTRL,	0x6e,	~EM_GPIO_4,	10},
 	{EM2880_R04_GPO,	0x08,	0xff,		10},
-	{ -1,			-1,	-1,		-1},
+	{	-1,		-1,	-1,		-1},
 };
 
 
 /* Reset for the most [digital] boards */
 static struct em28xx_reg_seq leadership_digital[] = {
 	{EM2874_R80_GPIO_P0_CTRL,	0x70,	0xff,	10},
-	{	-1,		-1,	-1,	-1},
+	{	-1,			-1,	-1,	-1},
 };
 
 static struct em28xx_reg_seq leadership_reset[] = {
 	{EM2874_R80_GPIO_P0_CTRL,	0xf0,	0xff,	10},
 	{EM2874_R80_GPIO_P0_CTRL,	0xb0,	0xff,	10},
 	{EM2874_R80_GPIO_P0_CTRL,	0xf0,	0xff,	10},
-	{	-1,		-1,	-1,	-1},
+	{	-1,			-1,	-1,	-1},
 };
 
 /* 2013:024f PCTV nanoStick T2 290e
@@ -304,7 +305,7 @@ static struct em28xx_reg_seq pctv_290e[] = {
 	{EM2874_R80_GPIO_P0_CTRL,	0x00,	0xff,	80},
 	{EM2874_R80_GPIO_P0_CTRL,	0x40,	0xff,	80}, /* GPIO_6 = 1 */
 	{EM2874_R80_GPIO_P0_CTRL,	0xc0,	0xff,	80}, /* GPIO_7 = 1 */
-	{-1,			-1,	-1,		-1},
+	{	-1,			-1,	-1,	-1},
 };
 
 #if 0
@@ -313,14 +314,14 @@ static struct em28xx_reg_seq terratec_h5_gpio[] = {
 	{EM2874_R80_GPIO_P0_CTRL,	0xf6,	0xff,	100},
 	{EM2874_R80_GPIO_P0_CTRL,	0xf2,	0xff,	50},
 	{EM2874_R80_GPIO_P0_CTRL,	0xf6,	0xff,	50},
-	{ -1,			-1,	-1,	-1},
+	{	-1,			-1,	-1,	-1},
 };
 
 static struct em28xx_reg_seq terratec_h5_digital[] = {
 	{EM2874_R80_GPIO_P0_CTRL,	0xf6,	0xff,	10},
 	{EM2874_R80_GPIO_P0_CTRL,	0xe6,	0xff,	100},
 	{EM2874_R80_GPIO_P0_CTRL,	0xa6,	0xff,	10},
-	{ -1,			-1,	-1,	-1},
+	{	-1,			-1,	-1,	-1},
 };
 #endif
 
@@ -335,12 +336,12 @@ static struct em28xx_reg_seq terratec_h5_digital[] = {
  * GPIO_7 - LED (green LED)
  */
 static struct em28xx_reg_seq pctv_460e[] = {
-	{EM2874_R80_GPIO_P0_CTRL, 0x01, 0xff,  50},
-	{0x0d,            0xff, 0xff,  50},
-	{EM2874_R80_GPIO_P0_CTRL, 0x41, 0xff,  50}, /* GPIO_6=1 */
-	{0x0d,            0x42, 0xff,  50},
-	{EM2874_R80_GPIO_P0_CTRL, 0x61, 0xff,  50}, /* GPIO_5=1 */
-	{             -1,   -1,   -1,  -1},
+	{EM2874_R80_GPIO_P0_CTRL,	0x01,	0xff,	50},
+	{	0x0d,			0xff,	0xff,	50},
+	{EM2874_R80_GPIO_P0_CTRL,	0x41,	0xff,	50}, /* GPIO_6=1 */
+	{	0x0d,			0x42,	0xff,	50},
+	{EM2874_R80_GPIO_P0_CTRL,	0x61,	0xff,	50}, /* GPIO_5=1 */
+	{	-1,			-1,	-1,	-1},
 };
 
 static struct em28xx_reg_seq c3tech_digital_duo_digital[] = {
@@ -352,7 +353,7 @@ static struct em28xx_reg_seq c3tech_digital_duo_digital[] = {
 	{EM2874_R80_GPIO_P0_CTRL,	0xfe,	0xff,	10},
 	{EM2874_R80_GPIO_P0_CTRL,	0xbe,	0xff,	10},
 	{EM2874_R80_GPIO_P0_CTRL,	0xfe,	0xff,	20},
-	{ -1,			-1,	-1,	-1},
+	{	-1,			-1,	-1,	-1},
 };
 
 #if 0
@@ -361,14 +362,14 @@ static struct em28xx_reg_seq hauppauge_930c_gpio[] = {
 	{EM2874_R80_GPIO_P0_CTRL,	0x4f,	0xff,	10}, /* xc5000 reset */
 	{EM2874_R80_GPIO_P0_CTRL,	0x6f,	0xff,	10},
 	{EM2874_R80_GPIO_P0_CTRL,	0x4f,	0xff,	10},
-	{ -1,			-1,	-1,	-1},
+	{	-1,			-1,	-1,	-1},
 };
 
 static struct em28xx_reg_seq hauppauge_930c_digital[] = {
 	{EM2874_R80_GPIO_P0_CTRL,	0xf6,	0xff,	10},
 	{EM2874_R80_GPIO_P0_CTRL,	0xe6,	0xff,	100},
 	{EM2874_R80_GPIO_P0_CTRL,	0xa6,	0xff,	10},
-	{ -1,			-1,	-1,	-1},
+	{	-1,			-1,	-1,	-1},
 };
 #endif
 
@@ -378,10 +379,10 @@ static struct em28xx_reg_seq hauppauge_930c_digital[] = {
  * GPIO_7 - LED, 0=active
  */
 static struct em28xx_reg_seq maxmedia_ub425_tc[] = {
-	{EM2874_R80_GPIO_P0_CTRL,  0x83,  0xff,  100},
-	{EM2874_R80_GPIO_P0_CTRL,  0xc3,  0xff,  100}, /* GPIO_6 = 1 */
-	{EM2874_R80_GPIO_P0_CTRL,  0x43,  0xff,  000}, /* GPIO_7 = 0 */
-	{-1,                 -1,    -1,   -1},
+	{EM2874_R80_GPIO_P0_CTRL,	0x83,	0xff,	100},
+	{EM2874_R80_GPIO_P0_CTRL,	0xc3,	0xff,	100}, /* GPIO_6 = 1 */
+	{EM2874_R80_GPIO_P0_CTRL,	0x43,	0xff,	000}, /* GPIO_7 = 0 */
+	{	-1,			-1,	-1,	-1},
 };
 
 /* 2304:0242 PCTV QuatroStick (510e)
@@ -391,10 +392,10 @@ static struct em28xx_reg_seq maxmedia_ub425_tc[] = {
  * GPIO_7: LED, 1=active
  */
 static struct em28xx_reg_seq pctv_510e[] = {
-	{EM2874_R80_GPIO_P0_CTRL, 0x10, 0xff, 100},
-	{EM2874_R80_GPIO_P0_CTRL, 0x14, 0xff, 100}, /* GPIO_2 = 1 */
-	{EM2874_R80_GPIO_P0_CTRL, 0x54, 0xff, 050}, /* GPIO_6 = 1 */
-	{             -1,   -1,   -1,  -1},
+	{EM2874_R80_GPIO_P0_CTRL,	0x10,	0xff,	100},
+	{EM2874_R80_GPIO_P0_CTRL,	0x14,	0xff,	100}, /* GPIO_2 = 1 */
+	{EM2874_R80_GPIO_P0_CTRL,	0x54,	0xff,	050}, /* GPIO_6 = 1 */
+	{	-1,			-1,	-1,	-1},
 };
 
 /* 2013:0251 PCTV QuatroStick nano (520e)
@@ -404,11 +405,11 @@ static struct em28xx_reg_seq pctv_510e[] = {
  * GPIO_7: LED, 1=active
  */
 static struct em28xx_reg_seq pctv_520e[] = {
-	{EM2874_R80_GPIO_P0_CTRL, 0x10, 0xff, 100},
-	{EM2874_R80_GPIO_P0_CTRL, 0x14, 0xff, 100}, /* GPIO_2 = 1 */
-	{EM2874_R80_GPIO_P0_CTRL, 0x54, 0xff, 050}, /* GPIO_6 = 1 */
-	{EM2874_R80_GPIO_P0_CTRL, 0xd4, 0xff, 000}, /* GPIO_7 = 1 */
-	{             -1,   -1,   -1,  -1},
+	{EM2874_R80_GPIO_P0_CTRL,	0x10,	0xff,	100},
+	{EM2874_R80_GPIO_P0_CTRL,	0x14,	0xff,	100}, /* GPIO_2 = 1 */
+	{EM2874_R80_GPIO_P0_CTRL,	0x54,	0xff,	050}, /* GPIO_6 = 1 */
+	{EM2874_R80_GPIO_P0_CTRL,	0xd4,	0xff,	000}, /* GPIO_7 = 1 */
+	{	-1,			-1,	-1,	-1},
 };
 
 /*
@@ -2030,6 +2031,18 @@ struct em28xx_board em28xx_boards[] = {
 		.i2c_speed     = EM28XX_I2C_CLK_WAIT_ENABLE |
 				EM28XX_I2C_FREQ_400_KHZ,
 	},
+	/*
+	 * 1b80:e346 KWorld USB ATSC TV Stick UB435-Q V2
+	 * Empia EM2874B + LG DT3305 + NXP TDA18271HDC2
+	 */
+	[EM2874_BOARD_KWORLD_UB435Q_V2] = {
+		.name		= "KWorld USB ATSC TV Stick UB435-Q V2",
+		.tuner_type	= TUNER_ABSENT,
+		.has_dvb	= 1,
+		.dvb_gpio	= kworld_a340_digital,
+		.tuner_gpio	= default_tuner_gpio,
+		.def_i2c_bus	= 1,
+	},
 };
 const unsigned int em28xx_bcount = ARRAY_SIZE(em28xx_boards);
 
@@ -2173,6 +2186,8 @@ struct usb_device_id em28xx_id_table[] = {
 			.driver_info = EM2860_BOARD_GADMEI_UTV330 },
 	{ USB_DEVICE(0x1b80, 0xa340),
 			.driver_info = EM2870_BOARD_KWORLD_A340 },
+	{ USB_DEVICE(0x1b80, 0xe346),
+			.driver_info = EM2874_BOARD_KWORLD_UB435Q_V2 },
 	{ USB_DEVICE(0x2013, 0x024f),
 			.driver_info = EM28174_BOARD_PCTV_290E },
 	{ USB_DEVICE(0x2013, 0x024c),
@@ -2857,6 +2872,8 @@ void em28xx_release_resources(struct em28xx *dev)
 	if (dev->def_i2c_bus)
 		em28xx_i2c_unregister(dev, 1);
 	em28xx_i2c_unregister(dev, 0);
+	if (dev->clk)
+		v4l2_clk_unregister_fixed(dev->clk);
 
 	v4l2_ctrl_handler_free(&dev->ctrl_handler);
 
diff --git a/drivers/media/usb/em28xx/em28xx-dvb.c b/drivers/media/usb/em28xx/em28xx-dvb.c
index bb1e8dca80cd..344042bb845c 100644
--- a/drivers/media/usb/em28xx/em28xx-dvb.c
+++ b/drivers/media/usb/em28xx/em28xx-dvb.c
@@ -298,6 +298,18 @@ static struct lgdt3305_config em2870_lgdt3304_dev = {
 	.qam_if_khz         = 4000,
 };
 
+static struct lgdt3305_config em2874_lgdt3305_dev = {
+	.i2c_addr           = 0x0e,
+	.demod_chip         = LGDT3305,
+	.spectral_inversion = 1,
+	.deny_i2c_rptr      = 0,
+	.mpeg_mode          = LGDT3305_MPEG_SERIAL,
+	.tpclk_edge         = LGDT3305_TPCLK_FALLING_EDGE,
+	.tpvalid_polarity   = LGDT3305_TP_VALID_HIGH,
+	.vsb_if_khz         = 3250,
+	.qam_if_khz         = 4000,
+};
+
 static struct s921_config sharp_isdbt = {
 	.demod_address = 0x30 >> 1
 };
@@ -329,6 +341,11 @@ static struct tda18271_config kworld_a340_config = {
 	.std_map           = &kworld_a340_std_map,
 };
 
+static struct tda18271_config kworld_ub435q_v2_config = {
+	.std_map	= &kworld_a340_std_map,
+	.gate		= TDA18271_GATE_DIGITAL,
+};
+
 static struct zl10353_config em28xx_zl10353_xc3028_no_i2c_gate = {
 	.demod_address = (0x1e >> 1),
 	.no_tuner = 1,
@@ -384,7 +401,10 @@ static struct drxk_config maxmedia_ub425_tc_drxk = {
 	.adr = 0x29,
 	.single_master = 1,
 	.no_i2c_bridge = 1,
+	.microcode_name = "dvb-demod-drxk-01.fw",
+	.chunk_size = 62,
 	.load_firmware_sync = true,
+	.qam_demod_parameter_count = 2,
 };
 
 static struct drxk_config pctv_520e_drxk = {
@@ -424,7 +444,7 @@ static void hauppauge_hvr930c_init(struct em28xx *dev)
 		{EM2874_R80_GPIO_P0_CTRL,	0xff,	0xff,	0x65},
 		{EM2874_R80_GPIO_P0_CTRL,	0xfb,	0xff,	0x32},
 		{EM2874_R80_GPIO_P0_CTRL,	0xff,	0xff,	0xb8},
-		{ -1,                   -1,     -1,     -1},
+		{	-1,			-1,	-1,	-1},
 	};
 	struct em28xx_reg_seq hauppauge_hvr930c_end[] = {
 		{EM2874_R80_GPIO_P0_CTRL,	0xef,	0xff,	0x01},
@@ -439,7 +459,7 @@ static void hauppauge_hvr930c_init(struct em28xx *dev)
 		{EM2874_R80_GPIO_P0_CTRL,	0xcf,	0xff,	0x0b},
 		{EM2874_R80_GPIO_P0_CTRL,	0xef,	0xff,	0x65},
 
-		{ -1,                   -1,     -1,     -1},
+		{	-1,			-1,	-1,	-1},
 	};
 
 	struct {
@@ -491,13 +511,13 @@ static void terratec_h5_init(struct em28xx *dev)
 		{EM2874_R80_GPIO_P0_CTRL,	0xf6,	0xff,	100},
 		{EM2874_R80_GPIO_P0_CTRL,	0xf2,	0xff,	50},
 		{EM2874_R80_GPIO_P0_CTRL,	0xf6,	0xff,	100},
-		{ -1,                   -1,     -1,     -1},
+		{	-1,			-1,	-1,	-1},
 	};
 	struct em28xx_reg_seq terratec_h5_end[] = {
 		{EM2874_R80_GPIO_P0_CTRL,	0xe6,	0xff,	100},
 		{EM2874_R80_GPIO_P0_CTRL,	0xa6,	0xff,	50},
 		{EM2874_R80_GPIO_P0_CTRL,	0xe6,	0xff,	100},
-		{ -1,                   -1,     -1,     -1},
+		{	-1,			-1,	-1,	-1},
 	};
 	struct {
 		unsigned char r[4];
@@ -547,12 +567,12 @@ static void terratec_htc_stick_init(struct em28xx *dev)
 		{EM2874_R80_GPIO_P0_CTRL,	0xf6,	0xff,	100},
 		{EM2874_R80_GPIO_P0_CTRL,	0xe6,	0xff,	50},
 		{EM2874_R80_GPIO_P0_CTRL,	0xf6,	0xff,	100},
-		{ -1,                   -1,     -1,     -1},
+		{	-1,			-1,	-1,	-1},
 	};
 	struct em28xx_reg_seq terratec_htc_stick_end[] = {
 		{EM2874_R80_GPIO_P0_CTRL,	0xb6,	0xff,	100},
 		{EM2874_R80_GPIO_P0_CTRL,	0xf6,	0xff,	50},
-		{ -1,                   -1,     -1,     -1},
+		{	-1,			-1,	-1,	-1},
 	};
 
 	/*
@@ -594,13 +614,13 @@ static void terratec_htc_usb_xs_init(struct em28xx *dev)
 		{EM2874_R80_GPIO_P0_CTRL,	0xb2,	0xff,	100},
 		{EM2874_R80_GPIO_P0_CTRL,	0xb2,	0xff,	50},
 		{EM2874_R80_GPIO_P0_CTRL,	0xb6,	0xff,	100},
-		{ -1,                   -1,     -1,     -1},
+		{	-1,			-1,	-1,	-1},
 	};
 	struct em28xx_reg_seq terratec_htc_usb_xs_end[] = {
 		{EM2874_R80_GPIO_P0_CTRL,	0xa6,	0xff,	100},
 		{EM2874_R80_GPIO_P0_CTRL,	0xa6,	0xff,	50},
 		{EM2874_R80_GPIO_P0_CTRL,	0xe6,	0xff,	100},
-		{ -1,                   -1,     -1,     -1},
+		{	-1,			-1,	-1,	-1},
 	};
 
 	/*
@@ -1227,18 +1247,14 @@ static int em28xx_dvb_init(struct em28xx *dev)
 			dvb->fe[0]->ops.i2c_gate_ctrl = NULL;
 
 			/* attach tuner */
-			if (!dvb_attach(tda18271c2dd_attach, dvb->fe[0],
-					&dev->i2c_adap[dev->def_i2c_bus], 0x60)) {
+			if (!dvb_attach(tda18271_attach, dvb->fe[0], 0x60,
+					&dev->i2c_adap[dev->def_i2c_bus],
+					&em28xx_cxd2820r_tda18271_config)) {
 				dvb_frontend_detach(dvb->fe[0]);
 				result = -EINVAL;
 				goto out_free;
 			}
 		}
-
-		/* TODO: we need drx-3913k firmware in order to support DVB-T */
-		em28xx_info("MaxMedia UB425-TC/Delock 61959: only DVB-C " \
-				"supported by that driver version\n");
-
 		break;
 	case EM2884_BOARD_PCTV_510E:
 	case EM2884_BOARD_PCTV_520E:
@@ -1297,6 +1313,23 @@ static int em28xx_dvb_init(struct em28xx *dev)
 			goto out_free;
 		}
 		break;
+	case EM2874_BOARD_KWORLD_UB435Q_V2:
+		dvb->fe[0] = dvb_attach(lgdt3305_attach,
+					&em2874_lgdt3305_dev,
+					&dev->i2c_adap[dev->def_i2c_bus]);
+		if (!dvb->fe[0]) {
+			result = -EINVAL;
+			goto out_free;
+		}
+
+		/* Attach the demodulator. */
+		if (!dvb_attach(tda18271_attach, dvb->fe[0], 0x60,
+				&dev->i2c_adap[dev->def_i2c_bus],
+				&kworld_ub435q_v2_config)) {
+			result = -EINVAL;
+			goto out_free;
+		}
+		break;
 	default:
 		em28xx_errdev("/2: The frontend of your DVB/ATSC card"
 				" isn't supported yet\n");
diff --git a/drivers/media/usb/em28xx/em28xx-video.c b/drivers/media/usb/em28xx/em28xx-video.c
index 9d103344f34a..fc5d60efd4ab 100644
--- a/drivers/media/usb/em28xx/em28xx-video.c
+++ b/drivers/media/usb/em28xx/em28xx-video.c
@@ -638,7 +638,7 @@ int em28xx_start_analog_streaming(struct vb2_queue *vq, unsigned int count)
 	if (rc)
 		return rc;
 
-	if (dev->streaming_users++ == 0) {
+	if (dev->streaming_users == 0) {
 		/* First active streaming user, so allocate all the URBs */
 
 		/* Allocate the USB bandwidth */
@@ -657,7 +657,7 @@ int em28xx_start_analog_streaming(struct vb2_queue *vq, unsigned int count)
 					  dev->packet_multiplier,
 					  em28xx_urb_data_copy);
 		if (rc < 0)
-			goto fail;
+			return rc;
 
 		/*
 		 * djh: it's not clear whether this code is still needed.  I'm
@@ -675,7 +675,8 @@ int em28xx_start_analog_streaming(struct vb2_queue *vq, unsigned int count)
 		v4l2_device_call_all(&dev->v4l2_dev, 0, tuner, s_frequency, &f);
 	}
 
-fail:
+	dev->streaming_users++;
+
 	return rc;
 }
 
diff --git a/drivers/media/usb/em28xx/em28xx.h b/drivers/media/usb/em28xx/em28xx.h
index 205e9038b1c0..f8726ad5d0a8 100644
--- a/drivers/media/usb/em28xx/em28xx.h
+++ b/drivers/media/usb/em28xx/em28xx.h
@@ -131,6 +131,7 @@
 #define EM2884_BOARD_TERRATEC_HTC_USB_XS	  87
 #define EM2884_BOARD_C3TECH_DIGITAL_DUO		  88
 #define EM2874_BOARD_DELOCK_61959		  89
+#define EM2874_BOARD_KWORLD_UB435Q_V2		  90
 
 /* Limits minimum and default number of buffers */
 #define EM28XX_MIN_BUF 4
@@ -492,6 +493,7 @@ struct em28xx {
 
 	struct v4l2_device v4l2_dev;
 	struct v4l2_ctrl_handler ctrl_handler;
+	struct v4l2_clk *clk;
 	struct em28xx_board board;
 
 	/* Webcam specific fields */
diff --git a/drivers/media/usb/gspca/conex.c b/drivers/media/usb/gspca/conex.c
index 38714df31ac4..2e15c80d6e3d 100644
--- a/drivers/media/usb/gspca/conex.c
+++ b/drivers/media/usb/gspca/conex.c
@@ -783,7 +783,8 @@ static int sd_start(struct gspca_dev *gspca_dev)
 	struct sd *sd = (struct sd *) gspca_dev;
 
 	/* create the JPEG header */
-	jpeg_define(sd->jpeg_hdr, gspca_dev->height, gspca_dev->width,
+	jpeg_define(sd->jpeg_hdr, gspca_dev->pixfmt.height,
+			gspca_dev->pixfmt.width,
 			0x22);		/* JPEG 411 */
 	jpeg_set_qual(sd->jpeg_hdr, QUALITY);
 
diff --git a/drivers/media/usb/gspca/cpia1.c b/drivers/media/usb/gspca/cpia1.c
index 064b53043b15..f23df4a9d8c5 100644
--- a/drivers/media/usb/gspca/cpia1.c
+++ b/drivers/media/usb/gspca/cpia1.c
@@ -1553,9 +1553,9 @@ static int sd_start(struct gspca_dev *gspca_dev)
 		sd->params.format.videoSize = VIDEOSIZE_CIF;
 
 	sd->params.roi.colEnd = sd->params.roi.colStart +
-				(gspca_dev->width >> 3);
+				(gspca_dev->pixfmt.width >> 3);
 	sd->params.roi.rowEnd = sd->params.roi.rowStart +
-				(gspca_dev->height >> 2);
+				(gspca_dev->pixfmt.height >> 2);
 
 	/* And now set the camera to a known state */
 	ret = do_command(gspca_dev, CPIA_COMMAND_SetGrabMode,
diff --git a/drivers/media/usb/gspca/gspca.c b/drivers/media/usb/gspca/gspca.c
index 048507b27bb2..f3a7ace0fac9 100644
--- a/drivers/media/usb/gspca/gspca.c
+++ b/drivers/media/usb/gspca/gspca.c
@@ -504,8 +504,7 @@ static int frame_alloc(struct gspca_dev *gspca_dev, struct file *file,
 	unsigned int frsz;
 	int i;
 
-	i = gspca_dev->curr_mode;
-	frsz = gspca_dev->cam.cam_mode[i].sizeimage;
+	frsz = gspca_dev->pixfmt.sizeimage;
 	PDEBUG(D_STREAM, "frame alloc frsz: %d", frsz);
 	frsz = PAGE_ALIGN(frsz);
 	if (count >= GSPCA_MAX_FRAMES)
@@ -627,16 +626,14 @@ static struct usb_host_endpoint *alt_xfer(struct usb_host_interface *alt,
 static u32 which_bandwidth(struct gspca_dev *gspca_dev)
 {
 	u32 bandwidth;
-	int i;
 
 	/* get the (max) image size */
-	i = gspca_dev->curr_mode;
-	bandwidth = gspca_dev->cam.cam_mode[i].sizeimage;
+	bandwidth = gspca_dev->pixfmt.sizeimage;
 
 	/* if the image is compressed, estimate its mean size */
 	if (!gspca_dev->cam.needs_full_bandwidth &&
-	    bandwidth < gspca_dev->cam.cam_mode[i].width *
-				gspca_dev->cam.cam_mode[i].height)
+	    bandwidth < gspca_dev->pixfmt.width *
+				gspca_dev->pixfmt.height)
 		bandwidth = bandwidth * 3 / 8;	/* 0.375 */
 
 	/* estimate the frame rate */
@@ -650,7 +647,7 @@ static u32 which_bandwidth(struct gspca_dev *gspca_dev)
 
 		/* don't hope more than 15 fps with USB 1.1 and
 		 * image resolution >= 640x480 */
-		if (gspca_dev->width >= 640
+		if (gspca_dev->pixfmt.width >= 640
 		 && gspca_dev->dev->speed == USB_SPEED_FULL)
 			bandwidth *= 15;		/* 15 fps */
 		else
@@ -982,9 +979,7 @@ static void gspca_set_default_mode(struct gspca_dev *gspca_dev)
 
 	i = gspca_dev->cam.nmodes - 1;	/* take the highest mode */
 	gspca_dev->curr_mode = i;
-	gspca_dev->width = gspca_dev->cam.cam_mode[i].width;
-	gspca_dev->height = gspca_dev->cam.cam_mode[i].height;
-	gspca_dev->pixfmt = gspca_dev->cam.cam_mode[i].pixelformat;
+	gspca_dev->pixfmt = gspca_dev->cam.cam_mode[i];
 
 	/* does nothing if ctrl_handler == NULL */
 	v4l2_ctrl_handler_setup(gspca_dev->vdev.ctrl_handler);
@@ -1105,10 +1100,8 @@ static int vidioc_g_fmt_vid_cap(struct file *file, void *priv,
 			    struct v4l2_format *fmt)
 {
 	struct gspca_dev *gspca_dev = video_drvdata(file);
-	int mode;
 
-	mode = gspca_dev->curr_mode;
-	fmt->fmt.pix = gspca_dev->cam.cam_mode[mode];
+	fmt->fmt.pix = gspca_dev->pixfmt;
 	/* some drivers use priv internally, zero it before giving it to
 	   userspace */
 	fmt->fmt.pix.priv = 0;
@@ -1140,6 +1133,12 @@ static int try_fmt_vid_cap(struct gspca_dev *gspca_dev,
 			mode = mode2;
 	}
 	fmt->fmt.pix = gspca_dev->cam.cam_mode[mode];
+	if (gspca_dev->sd_desc->try_fmt) {
+		/* pass original resolution to subdriver try_fmt */
+		fmt->fmt.pix.width = w;
+		fmt->fmt.pix.height = h;
+		gspca_dev->sd_desc->try_fmt(gspca_dev, fmt);
+	}
 	/* some drivers use priv internally, zero it before giving it to
 	   userspace */
 	fmt->fmt.pix.priv = 0;
@@ -1178,19 +1177,16 @@ static int vidioc_s_fmt_vid_cap(struct file *file, void *priv,
 		goto out;
 	}
 
-	if (ret == gspca_dev->curr_mode) {
-		ret = 0;
-		goto out;			/* same mode */
-	}
-
 	if (gspca_dev->streaming) {
 		ret = -EBUSY;
 		goto out;
 	}
-	gspca_dev->width = fmt->fmt.pix.width;
-	gspca_dev->height = fmt->fmt.pix.height;
-	gspca_dev->pixfmt = fmt->fmt.pix.pixelformat;
 	gspca_dev->curr_mode = ret;
+	if (gspca_dev->sd_desc->try_fmt)
+		/* subdriver try_fmt can modify format parameters */
+		gspca_dev->pixfmt = fmt->fmt.pix;
+	else
+		gspca_dev->pixfmt = gspca_dev->cam.cam_mode[ret];
 
 	ret = 0;
 out:
@@ -1205,6 +1201,9 @@ static int vidioc_enum_framesizes(struct file *file, void *priv,
 	int i;
 	__u32 index = 0;
 
+	if (gspca_dev->sd_desc->enum_framesizes)
+		return gspca_dev->sd_desc->enum_framesizes(gspca_dev, fsize);
+
 	for (i = 0; i < gspca_dev->cam.nmodes; i++) {
 		if (fsize->pixel_format !=
 				gspca_dev->cam.cam_mode[i].pixelformat)
@@ -1471,8 +1470,9 @@ static int vidioc_streamon(struct file *file, void *priv,
 		if (ret < 0)
 			goto out;
 	}
-	PDEBUG_MODE(gspca_dev, D_STREAM, "stream on OK", gspca_dev->pixfmt,
-		    gspca_dev->width, gspca_dev->height);
+	PDEBUG_MODE(gspca_dev, D_STREAM, "stream on OK",
+		    gspca_dev->pixfmt.pixelformat,
+		    gspca_dev->pixfmt.width, gspca_dev->pixfmt.height);
 	ret = 0;
 out:
 	mutex_unlock(&gspca_dev->queue_lock);
diff --git a/drivers/media/usb/gspca/gspca.h b/drivers/media/usb/gspca/gspca.h
index ac0b11f46f50..300642dc1a17 100644
--- a/drivers/media/usb/gspca/gspca.h
+++ b/drivers/media/usb/gspca/gspca.h
@@ -88,6 +88,10 @@ typedef void (*cam_pkt_op) (struct gspca_dev *gspca_dev,
 typedef int (*cam_int_pkt_op) (struct gspca_dev *gspca_dev,
 				u8 *data,
 				int len);
+typedef void (*cam_format_op) (struct gspca_dev *gspca_dev,
+				struct v4l2_format *fmt);
+typedef int (*cam_frmsize_op) (struct gspca_dev *gspca_dev,
+				struct v4l2_frmsizeenum *fsize);
 
 /* subdriver description */
 struct sd_desc {
@@ -109,6 +113,8 @@ struct sd_desc {
 	cam_set_jpg_op set_jcomp;
 	cam_streamparm_op get_streamparm;
 	cam_streamparm_op set_streamparm;
+	cam_format_op try_fmt;
+	cam_frmsize_op enum_framesizes;
 #ifdef CONFIG_VIDEO_ADV_DEBUG
 	cam_set_reg_op set_register;
 	cam_get_reg_op get_register;
@@ -183,9 +189,7 @@ struct gspca_dev {
 	__u8 streaming;			/* protected by both mutexes (*) */
 
 	__u8 curr_mode;			/* current camera mode */
-	__u32 pixfmt;			/* current mode parameters */
-	__u16 width;
-	__u16 height;
+	struct v4l2_pix_format pixfmt;	/* current mode parameters */
 	__u32 sequence;			/* frame sequence number */
 
 	wait_queue_head_t wq;		/* wait queue */
diff --git a/drivers/media/usb/gspca/jeilinj.c b/drivers/media/usb/gspca/jeilinj.c
index 8da3dde38385..19736e237b37 100644
--- a/drivers/media/usb/gspca/jeilinj.c
+++ b/drivers/media/usb/gspca/jeilinj.c
@@ -378,11 +378,12 @@ static int sd_start(struct gspca_dev *gspca_dev)
 	struct sd *dev = (struct sd *) gspca_dev;
 
 	/* create the JPEG header */
-	jpeg_define(dev->jpeg_hdr, gspca_dev->height, gspca_dev->width,
+	jpeg_define(dev->jpeg_hdr, gspca_dev->pixfmt.height,
+			gspca_dev->pixfmt.width,
 			0x21);          /* JPEG 422 */
 	jpeg_set_qual(dev->jpeg_hdr, dev->quality);
 	PDEBUG(D_STREAM, "Start streaming at %dx%d",
-		gspca_dev->height, gspca_dev->width);
+		gspca_dev->pixfmt.height, gspca_dev->pixfmt.width);
 	jlj_start(gspca_dev);
 	return gspca_dev->usb_err;
 }
diff --git a/drivers/media/usb/gspca/jl2005bcd.c b/drivers/media/usb/gspca/jl2005bcd.c
index fdaeeb14453f..5b481fa43099 100644
--- a/drivers/media/usb/gspca/jl2005bcd.c
+++ b/drivers/media/usb/gspca/jl2005bcd.c
@@ -455,7 +455,7 @@ static int sd_start(struct gspca_dev *gspca_dev)
 	struct sd *sd = (struct sd *) gspca_dev;
 	sd->cap_mode = gspca_dev->cam.cam_mode;
 
-	switch (gspca_dev->width) {
+	switch (gspca_dev->pixfmt.width) {
 	case 640:
 		PDEBUG(D_STREAM, "Start streaming at vga resolution");
 		jl2005c_stream_start_vga_lg(gspca_dev);
diff --git a/drivers/media/usb/gspca/m5602/m5602_mt9m111.c b/drivers/media/usb/gspca/m5602/m5602_mt9m111.c
index cfa4663f8934..27fcef11aef4 100644
--- a/drivers/media/usb/gspca/m5602/m5602_mt9m111.c
+++ b/drivers/media/usb/gspca/m5602/m5602_mt9m111.c
@@ -266,7 +266,7 @@ static int mt9m111_set_hvflip(struct gspca_dev *gspca_dev)
 		return err;
 
 	data[0] = MT9M111_RMB_OVER_SIZED;
-	if (gspca_dev->width == 640) {
+	if (gspca_dev->pixfmt.width == 640) {
 		data[1] = MT9M111_RMB_ROW_SKIP_2X |
 			  MT9M111_RMB_COLUMN_SKIP_2X |
 			  (hflip << 1) | vflip;
diff --git a/drivers/media/usb/gspca/mars.c b/drivers/media/usb/gspca/mars.c
index ff2c5abf115b..779a8785f421 100644
--- a/drivers/media/usb/gspca/mars.c
+++ b/drivers/media/usb/gspca/mars.c
@@ -254,7 +254,8 @@ static int sd_start(struct gspca_dev *gspca_dev)
 	int i;
 
 	/* create the JPEG header */
-	jpeg_define(sd->jpeg_hdr, gspca_dev->height, gspca_dev->width,
+	jpeg_define(sd->jpeg_hdr, gspca_dev->pixfmt.height,
+			gspca_dev->pixfmt.width,
 			0x21);		/* JPEG 422 */
 	jpeg_set_qual(sd->jpeg_hdr, QUALITY);
 
@@ -270,8 +271,8 @@ static int sd_start(struct gspca_dev *gspca_dev)
 	data[0] = 0x00;		/* address */
 	data[1] = 0x0c | 0x01;	/* reg 0 */
 	data[2] = 0x01;		/* reg 1 */
-	data[3] = gspca_dev->width / 8;		/* h_size , reg 2 */
-	data[4] = gspca_dev->height / 8;	/* v_size , reg 3 */
+	data[3] = gspca_dev->pixfmt.width / 8;	/* h_size , reg 2 */
+	data[4] = gspca_dev->pixfmt.height / 8;	/* v_size , reg 3 */
 	data[5] = 0x30;		/* reg 4, MI, PAS5101 :
 				 *	0x30 for 24mhz , 0x28 for 12mhz */
 	data[6] = 0x02;		/* reg 5, H start - was 0x04 */
diff --git a/drivers/media/usb/gspca/mr97310a.c b/drivers/media/usb/gspca/mr97310a.c
index 68bb2f359666..f006e29ca019 100644
--- a/drivers/media/usb/gspca/mr97310a.c
+++ b/drivers/media/usb/gspca/mr97310a.c
@@ -521,7 +521,7 @@ static int start_cif_cam(struct gspca_dev *gspca_dev)
 	if (sd->sensor_type)
 		data[5] = 0xbb;
 
-	switch (gspca_dev->width) {
+	switch (gspca_dev->pixfmt.width) {
 	case 160:
 		data[9] |= 0x04;  /* reg 8, 2:1 scale down from 320 */
 		/* fall thru */
@@ -618,7 +618,7 @@ static int start_vga_cam(struct gspca_dev *gspca_dev)
 		data[10] = 0x18;
 	}
 
-	switch (gspca_dev->width) {
+	switch (gspca_dev->pixfmt.width) {
 	case 160:
 		data[9] |= 0x0c;  /* reg 8, 4:1 scale down */
 		/* fall thru */
@@ -847,7 +847,7 @@ static void setexposure(struct gspca_dev *gspca_dev, s32 expo, s32 min_clockdiv)
 		u8 clockdiv = (60 * expo + 7999) / 8000;
 
 		/* Limit framerate to not exceed usb bandwidth */
-		if (clockdiv < min_clockdiv && gspca_dev->width >= 320)
+		if (clockdiv < min_clockdiv && gspca_dev->pixfmt.width >= 320)
 			clockdiv = min_clockdiv;
 		else if (clockdiv < 2)
 			clockdiv = 2;
diff --git a/drivers/media/usb/gspca/nw80x.c b/drivers/media/usb/gspca/nw80x.c
index 44c9964b1b3e..599f755e75b8 100644
--- a/drivers/media/usb/gspca/nw80x.c
+++ b/drivers/media/usb/gspca/nw80x.c
@@ -1708,7 +1708,7 @@ static void setautogain(struct gspca_dev *gspca_dev, s32 val)
 
 	reg_r(gspca_dev, 0x1004, 1);
 	if (gspca_dev->usb_buf[0] & 0x04) {	/* if AE_FULL_FRM */
-		sd->ae_res = gspca_dev->width * gspca_dev->height;
+		sd->ae_res = gspca_dev->pixfmt.width * gspca_dev->pixfmt.height;
 	} else {				/* get the AE window size */
 		reg_r(gspca_dev, 0x1011, 8);
 		w = (gspca_dev->usb_buf[1] << 8) + gspca_dev->usb_buf[0]
@@ -1717,7 +1717,8 @@ static void setautogain(struct gspca_dev *gspca_dev, s32 val)
 		  - (gspca_dev->usb_buf[7] << 8) - gspca_dev->usb_buf[6];
 		sd->ae_res = h * w;
 		if (sd->ae_res == 0)
-			sd->ae_res = gspca_dev->width * gspca_dev->height;
+			sd->ae_res = gspca_dev->pixfmt.width *
+					gspca_dev->pixfmt.height;
 	}
 }
 
@@ -1856,21 +1857,21 @@ static int sd_start(struct gspca_dev *gspca_dev)
 	reg_w_buf(gspca_dev, cmd);
 	switch (sd->webcam) {
 	case P35u:
-		if (gspca_dev->width == 320)
+		if (gspca_dev->pixfmt.width == 320)
 			reg_w_buf(gspca_dev, nw801_start_qvga);
 		else
 			reg_w_buf(gspca_dev, nw801_start_vga);
 		reg_w_buf(gspca_dev, nw801_start_2);
 		break;
 	case Kr651us:
-		if (gspca_dev->width == 320)
+		if (gspca_dev->pixfmt.width == 320)
 			reg_w_buf(gspca_dev, kr651_start_qvga);
 		else
 			reg_w_buf(gspca_dev, kr651_start_vga);
 		reg_w_buf(gspca_dev, kr651_start_2);
 		break;
 	case Proscope:
-		if (gspca_dev->width == 320)
+		if (gspca_dev->pixfmt.width == 320)
 			reg_w_buf(gspca_dev, proscope_start_qvga);
 		else
 			reg_w_buf(gspca_dev, proscope_start_vga);
diff --git a/drivers/media/usb/gspca/ov519.c b/drivers/media/usb/gspca/ov519.c
index 8937d79fd176..c95f32a0c02b 100644
--- a/drivers/media/usb/gspca/ov519.c
+++ b/drivers/media/usb/gspca/ov519.c
@@ -3468,7 +3468,7 @@ static int sd_isoc_init(struct gspca_dev *gspca_dev)
 
 	switch (sd->bridge) {
 	case BRIDGE_OVFX2:
-		if (gspca_dev->width != 800)
+		if (gspca_dev->pixfmt.width != 800)
 			gspca_dev->cam.bulk_size = OVFX2_BULK_SIZE;
 		else
 			gspca_dev->cam.bulk_size = 7 * 4096;
@@ -3507,8 +3507,8 @@ static void ov511_mode_init_regs(struct sd *sd)
 	/* Here I'm assuming that snapshot size == image size.
 	 * I hope that's always true. --claudio
 	 */
-	hsegs = (sd->gspca_dev.width >> 3) - 1;
-	vsegs = (sd->gspca_dev.height >> 3) - 1;
+	hsegs = (sd->gspca_dev.pixfmt.width >> 3) - 1;
+	vsegs = (sd->gspca_dev.pixfmt.height >> 3) - 1;
 
 	reg_w(sd, R511_CAM_PXCNT, hsegs);
 	reg_w(sd, R511_CAM_LNCNT, vsegs);
@@ -3541,7 +3541,7 @@ static void ov511_mode_init_regs(struct sd *sd)
 	case SEN_OV7640:
 	case SEN_OV7648:
 	case SEN_OV76BE:
-		if (sd->gspca_dev.width == 320)
+		if (sd->gspca_dev.pixfmt.width == 320)
 			interlaced = 1;
 		/* Fall through */
 	case SEN_OV6630:
@@ -3551,7 +3551,7 @@ static void ov511_mode_init_regs(struct sd *sd)
 		case 30:
 		case 25:
 			/* Not enough bandwidth to do 640x480 @ 30 fps */
-			if (sd->gspca_dev.width != 640) {
+			if (sd->gspca_dev.pixfmt.width != 640) {
 				sd->clockdiv = 0;
 				break;
 			}
@@ -3584,7 +3584,8 @@ static void ov511_mode_init_regs(struct sd *sd)
 
 	/* Check if we have enough bandwidth to disable compression */
 	fps = (interlaced ? 60 : 30) / (sd->clockdiv + 1) + 1;
-	needed = fps * sd->gspca_dev.width * sd->gspca_dev.height * 3 / 2;
+	needed = fps * sd->gspca_dev.pixfmt.width *
+			sd->gspca_dev.pixfmt.height * 3 / 2;
 	/* 1000 isoc packets/sec */
 	if (needed > 1000 * packet_size) {
 		/* Enable Y and UV quantization and compression */
@@ -3646,8 +3647,8 @@ static void ov518_mode_init_regs(struct sd *sd)
 		reg_w(sd, 0x38, 0x80);
 	}
 
-	hsegs = sd->gspca_dev.width / 16;
-	vsegs = sd->gspca_dev.height / 4;
+	hsegs = sd->gspca_dev.pixfmt.width / 16;
+	vsegs = sd->gspca_dev.pixfmt.height / 4;
 
 	reg_w(sd, 0x29, hsegs);
 	reg_w(sd, 0x2a, vsegs);
@@ -3686,7 +3687,8 @@ static void ov518_mode_init_regs(struct sd *sd)
 			 * happened to be with revision < 2 cams using an
 			 * OV7620 and revision 2 cams using an OV7620AE.
 			 */
-			if (sd->revision > 0 && sd->gspca_dev.width == 640) {
+			if (sd->revision > 0 &&
+					sd->gspca_dev.pixfmt.width == 640) {
 				reg_w(sd, 0x20, 0x60);
 				reg_w(sd, 0x21, 0x1f);
 			} else {
@@ -3812,8 +3814,8 @@ static void ov519_mode_init_regs(struct sd *sd)
 		break;
 	}
 
-	reg_w(sd, OV519_R10_H_SIZE,	sd->gspca_dev.width >> 4);
-	reg_w(sd, OV519_R11_V_SIZE,	sd->gspca_dev.height >> 3);
+	reg_w(sd, OV519_R10_H_SIZE,	sd->gspca_dev.pixfmt.width >> 4);
+	reg_w(sd, OV519_R11_V_SIZE,	sd->gspca_dev.pixfmt.height >> 3);
 	if (sd->sensor == SEN_OV7670 &&
 	    sd->gspca_dev.cam.cam_mode[sd->gspca_dev.curr_mode].priv)
 		reg_w(sd, OV519_R12_X_OFFSETL, 0x04);
@@ -3947,14 +3949,16 @@ static void mode_init_ov_sensor_regs(struct sd *sd)
 	    }
 	case SEN_OV3610:
 		if (qvga) {
-			xstart = (1040 - gspca_dev->width) / 2 + (0x1f << 4);
-			ystart = (776 - gspca_dev->height) / 2;
+			xstart = (1040 - gspca_dev->pixfmt.width) / 2 +
+				(0x1f << 4);
+			ystart = (776 - gspca_dev->pixfmt.height) / 2;
 		} else {
-			xstart = (2076 - gspca_dev->width) / 2 + (0x10 << 4);
-			ystart = (1544 - gspca_dev->height) / 2;
+			xstart = (2076 - gspca_dev->pixfmt.width) / 2 +
+				(0x10 << 4);
+			ystart = (1544 - gspca_dev->pixfmt.height) / 2;
 		}
-		xend = xstart + gspca_dev->width;
-		yend = ystart + gspca_dev->height;
+		xend = xstart + gspca_dev->pixfmt.width;
+		yend = ystart + gspca_dev->pixfmt.height;
 		/* Writing to the COMH register resets the other windowing regs
 		   to their default values, so we must do this first. */
 		i2c_w_mask(sd, 0x12, qvga ? 0x40 : 0x00, 0xf0);
@@ -4229,8 +4233,8 @@ static int sd_start(struct gspca_dev *gspca_dev)
 	struct sd *sd = (struct sd *) gspca_dev;
 
 	/* Default for most bridges, allow bridge_mode_init_regs to override */
-	sd->sensor_width = sd->gspca_dev.width;
-	sd->sensor_height = sd->gspca_dev.height;
+	sd->sensor_width = sd->gspca_dev.pixfmt.width;
+	sd->sensor_height = sd->gspca_dev.pixfmt.height;
 
 	switch (sd->bridge) {
 	case BRIDGE_OV511:
@@ -4345,12 +4349,13 @@ static void ov511_pkt_scan(struct gspca_dev *gspca_dev,
 		ov51x_handle_button(gspca_dev, (in[8] >> 2) & 1);
 		if (in[8] & 0x80) {
 			/* Frame end */
-			if ((in[9] + 1) * 8 != gspca_dev->width ||
-			    (in[10] + 1) * 8 != gspca_dev->height) {
+			if ((in[9] + 1) * 8 != gspca_dev->pixfmt.width ||
+			    (in[10] + 1) * 8 != gspca_dev->pixfmt.height) {
 				PERR("Invalid frame size, got: %dx%d,"
 					" requested: %dx%d\n",
 					(in[9] + 1) * 8, (in[10] + 1) * 8,
-					gspca_dev->width, gspca_dev->height);
+					gspca_dev->pixfmt.width,
+					gspca_dev->pixfmt.height);
 				gspca_dev->last_packet_type = DISCARD_PACKET;
 				return;
 			}
@@ -4470,7 +4475,8 @@ static void ovfx2_pkt_scan(struct gspca_dev *gspca_dev,
 		if (sd->first_frame) {
 			sd->first_frame--;
 			if (gspca_dev->image_len <
-				  sd->gspca_dev.width * sd->gspca_dev.height)
+				  sd->gspca_dev.pixfmt.width *
+					sd->gspca_dev.pixfmt.height)
 				gspca_dev->last_packet_type = DISCARD_PACKET;
 		}
 		gspca_frame_add(gspca_dev, LAST_PACKET, NULL, 0);
diff --git a/drivers/media/usb/gspca/ov534.c b/drivers/media/usb/gspca/ov534.c
index 03a33c46ca2c..90f0d637cd9d 100644
--- a/drivers/media/usb/gspca/ov534.c
+++ b/drivers/media/usb/gspca/ov534.c
@@ -1440,9 +1440,10 @@ static void sd_pkt_scan(struct gspca_dev *gspca_dev,
 		/* If this packet is marked as EOF, end the frame */
 		} else if (data[1] & UVC_STREAM_EOF) {
 			sd->last_pts = 0;
-			if (gspca_dev->pixfmt == V4L2_PIX_FMT_YUYV
+			if (gspca_dev->pixfmt.pixelformat == V4L2_PIX_FMT_YUYV
 			 && gspca_dev->image_len + len - 12 !=
-				   gspca_dev->width * gspca_dev->height * 2) {
+				   gspca_dev->pixfmt.width *
+					gspca_dev->pixfmt.height * 2) {
 				PDEBUG(D_PACK, "wrong sized frame");
 				goto discard;
 			}
diff --git a/drivers/media/usb/gspca/ov534_9.c b/drivers/media/usb/gspca/ov534_9.c
index c4cd028fe0b4..47085cf2d723 100644
--- a/drivers/media/usb/gspca/ov534_9.c
+++ b/drivers/media/usb/gspca/ov534_9.c
@@ -59,6 +59,7 @@ enum sensors {
 	SENSOR_OV965x,		/* ov9657 */
 	SENSOR_OV971x,		/* ov9712 */
 	SENSOR_OV562x,		/* ov5621 */
+	SENSOR_OV361x,		/* ov3610 */
 	NSENSORS
 };
 
@@ -106,6 +107,274 @@ static const struct v4l2_pix_format ov562x_mode[] = {
 	}
 };
 
+enum ov361x {
+	ov361x_2048 = 0,
+	ov361x_1600,
+	ov361x_1024,
+	ov361x_640,
+	ov361x_320,
+	ov361x_160,
+	ov361x_last
+};
+
+static const struct v4l2_pix_format ov361x_mode[] = {
+	{0x800, 0x600, V4L2_PIX_FMT_SBGGR8, V4L2_FIELD_NONE,
+		.bytesperline = 0x800,
+		.sizeimage = 0x800 * 0x600,
+		.colorspace = V4L2_COLORSPACE_SRGB},
+	{1600, 1200, V4L2_PIX_FMT_SBGGR8, V4L2_FIELD_NONE,
+		.bytesperline = 1600,
+		.sizeimage = 1600 * 1200,
+		.colorspace = V4L2_COLORSPACE_SRGB},
+	{1024, 768, V4L2_PIX_FMT_SBGGR8, V4L2_FIELD_NONE,
+		.bytesperline = 768,
+		.sizeimage = 1024 * 768,
+		.colorspace = V4L2_COLORSPACE_SRGB},
+	{640, 480, V4L2_PIX_FMT_SBGGR8, V4L2_FIELD_NONE,
+		.bytesperline = 640,
+		.sizeimage = 640 * 480,
+		.colorspace = V4L2_COLORSPACE_SRGB},
+	{320, 240, V4L2_PIX_FMT_SBGGR8, V4L2_FIELD_NONE,
+		.bytesperline = 320,
+		.sizeimage = 320 * 240,
+		.colorspace = V4L2_COLORSPACE_SRGB},
+	{160, 120, V4L2_PIX_FMT_SBGGR8, V4L2_FIELD_NONE,
+		.bytesperline = 160,
+		.sizeimage = 160 * 120,
+		.colorspace = V4L2_COLORSPACE_SRGB}
+};
+
+static const u8 ov361x_start_2048[][2] = {
+	{0x12, 0x80},
+	{0x13, 0xcf},
+	{0x14, 0x40},
+	{0x15, 0x00},
+	{0x01, 0x80},
+	{0x02, 0x80},
+	{0x04, 0x70},
+	{0x0d, 0x40},
+	{0x0f, 0x47},
+	{0x11, 0x81},
+	{0x32, 0x36},
+	{0x33, 0x0c},
+	{0x34, 0x00},
+	{0x35, 0x90},
+	{0x12, 0x00},
+	{0x17, 0x10},
+	{0x18, 0x90},
+	{0x19, 0x00},
+	{0x1a, 0xc0},
+};
+static const u8 ov361x_bridge_start_2048[][2] = {
+	{0xf1, 0x60},
+	{0x88, 0x00},
+	{0x89, 0x08},
+	{0x8a, 0x00},
+	{0x8b, 0x06},
+	{0x8c, 0x01},
+	{0x8d, 0x10},
+	{0x1c, 0x00},
+	{0x1d, 0x48},
+	{0x1d, 0x00},
+	{0x1d, 0xff},
+	{0x1c, 0x0a},
+	{0x1d, 0x2e},
+	{0x1d, 0x1e},
+};
+
+static const u8 ov361x_start_1600[][2] = {
+	{0x12, 0x80},
+	{0x13, 0xcf},
+	{0x14, 0x40},
+	{0x15, 0x00},
+	{0x01, 0x80},
+	{0x02, 0x80},
+	{0x04, 0x70},
+	{0x0d, 0x40},
+	{0x0f, 0x47},
+	{0x11, 0x81},
+	{0x32, 0x36},
+	{0x33, 0x0C},
+	{0x34, 0x00},
+	{0x35, 0x90},
+	{0x12, 0x00},
+	{0x17, 0x10},
+	{0x18, 0x90},
+	{0x19, 0x00},
+	{0x1a, 0xc0},
+};
+static const u8 ov361x_bridge_start_1600[][2] = {
+	{0xf1, 0x60},  /* Hsize[7:0] */
+	{0x88, 0x00},  /* Hsize[15:8] Write Only, can't read */
+	{0x89, 0x08},  /* Vsize[7:0] */
+	{0x8a, 0x00},  /* Vsize[15:8] Write Only, can't read */
+	{0x8b, 0x06},  /* for Iso */
+	{0x8c, 0x01},  /* RAW input */
+	{0x8d, 0x10},
+	{0x1c, 0x00},  /* RAW output, Iso transfer */
+	{0x1d, 0x48},
+	{0x1d, 0x00},
+	{0x1d, 0xff},
+	{0x1c, 0x0a},  /* turn off JPEG, Iso mode */
+	{0x1d, 0x2e},  /* for Iso */
+	{0x1d, 0x1e},
+};
+
+static const u8 ov361x_start_1024[][2] = {
+	{0x12, 0x80},
+	{0x13, 0xcf},
+	{0x14, 0x40},
+	{0x15, 0x00},
+	{0x01, 0x80},
+	{0x02, 0x80},
+	{0x04, 0x70},
+	{0x0d, 0x40},
+	{0x0f, 0x47},
+	{0x11, 0x81},
+	{0x32, 0x36},
+	{0x33, 0x0C},
+	{0x34, 0x00},
+	{0x35, 0x90},
+	{0x12, 0x40},
+	{0x17, 0x1f},
+	{0x18, 0x5f},
+	{0x19, 0x00},
+	{0x1a, 0x68},
+};
+static const u8 ov361x_bridge_start_1024[][2] = {
+	{0xf1, 0x60},  /* Hsize[7:0] */
+	{0x88, 0x00},  /* Hsize[15:8] Write Only, can't read */
+	{0x89, 0x04},  /* Vsize[7:0] */
+	{0x8a, 0x00},  /* Vsize[15:8] Write Only, can't read */
+	{0x8b, 0x03},  /* for Iso */
+	{0x8c, 0x01},  /* RAW input  */
+	{0x8d, 0x10},
+	{0x1c, 0x00},  /* RAW output, Iso transfer */
+	{0x1d, 0x48},
+	{0x1d, 0x00},
+	{0x1d, 0xff},
+	{0x1c, 0x0a},  /* turn off JPEG, Iso mode */
+	{0x1d, 0x2e},  /* for Iso */
+	{0x1d, 0x1e},
+};
+
+static const u8 ov361x_start_640[][2] = {
+	{0x12, 0x80},
+	{0x13, 0xcf},
+	{0x14, 0x40},
+	{0x15, 0x00},
+	{0x01, 0x80},
+	{0x02, 0x80},
+	{0x04, 0x70},
+	{0x0d, 0x40},
+	{0x0f, 0x47},
+	{0x11, 0x81},
+	{0x32, 0x36},
+	{0x33, 0x0C},
+	{0x34, 0x00},
+	{0x35, 0x90},
+	{0x12, 0x40},
+	{0x17, 0x1f},
+	{0x18, 0x5f},
+	{0x19, 0x00},
+	{0x1a, 0x68},
+};
+
+static const u8 ov361x_bridge_start_640[][2] = {
+	{0xf1, 0x60},  /* Hsize[7:0]*/
+	{0x88, 0x00},  /* Hsize[15:8] Write Only, can't read */
+	{0x89, 0x04},  /* Vsize[7:0] */
+	{0x8a, 0x00},  /* Vsize[15:8] Write Only, can't read */
+	{0x8b, 0x03},  /* for Iso */
+	{0x8c, 0x01},  /* RAW input */
+	{0x8d, 0x10},
+	{0x1c, 0x00},  /* RAW output, Iso transfer */
+	{0x1d, 0x48},
+	{0x1d, 0x00},
+	{0x1d, 0xff},
+	{0x1c, 0x0a},  /* turn off JPEG, Iso mode */
+	{0x1d, 0x2e},  /* for Iso */
+	{0x1d, 0x1e},
+};
+
+static const u8 ov361x_start_320[][2] = {
+	{0x12, 0x80},
+	{0x13, 0xcf},
+	{0x14, 0x40},
+	{0x15, 0x00},
+	{0x01, 0x80},
+	{0x02, 0x80},
+	{0x04, 0x70},
+	{0x0d, 0x40},
+	{0x0f, 0x47},
+	{0x11, 0x81},
+	{0x32, 0x36},
+	{0x33, 0x0C},
+	{0x34, 0x00},
+	{0x35, 0x90},
+	{0x12, 0x40},
+	{0x17, 0x1f},
+	{0x18, 0x5f},
+	{0x19, 0x00},
+	{0x1a, 0x68},
+};
+
+static const u8 ov361x_bridge_start_320[][2] = {
+	{0xf1, 0x60},  /* Hsize[7:0] */
+	{0x88, 0x00},  /* Hsize[15:8] Write Only, can't read */
+	{0x89, 0x04},  /* Vsize[7:0] */
+	{0x8a, 0x00},  /* Vsize[15:8] Write Only, can't read */
+	{0x8b, 0x03},  /* for Iso */
+	{0x8c, 0x01},  /* RAW input */
+	{0x8d, 0x10},
+	{0x1c, 0x00},  /* RAW output, Iso transfer; */
+	{0x1d, 0x48},
+	{0x1d, 0x00},
+	{0x1d, 0xff},
+	{0x1c, 0x0a},  /* turn off JPEG, Iso mode */
+	{0x1d, 0x2e},  /* for Iso */
+	{0x1d, 0x1e},
+};
+
+static const u8 ov361x_start_160[][2] = {
+	{0x12, 0x80},
+	{0x13, 0xcf},
+	{0x14, 0x40},
+	{0x15, 0x00},
+	{0x01, 0x80},
+	{0x02, 0x80},
+	{0x04, 0x70},
+	{0x0d, 0x40},
+	{0x0f, 0x47},
+	{0x11, 0x81},
+	{0x32, 0x36},
+	{0x33, 0x0C},
+	{0x34, 0x00},
+	{0x35, 0x90},
+	{0x12, 0x40},
+	{0x17, 0x1f},
+	{0x18, 0x5f},
+	{0x19, 0x00},
+	{0x1a, 0x68},
+};
+
+static const u8 ov361x_bridge_start_160[][2] = {
+	{0xf1, 0x60},  /* Hsize[7:0] */
+	{0x88, 0x00},  /* Hsize[15:8] Write Only, can't read */
+	{0x89, 0x04},  /* Vsize[7:0] */
+	{0x8a, 0x00},  /* Vsize[15:8] Write Only, can't read */
+	{0x8b, 0x03},  /* for Iso */
+	{0x8c, 0x01},  /* RAW input */
+	{0x8d, 0x10},
+	{0x1c, 0x00},  /* RAW output, Iso transfer */
+	{0x1d, 0x48},
+	{0x1d, 0x00},
+	{0x1d, 0xff},
+	{0x1c, 0x0a},  /* turn off JPEG, Iso mode */
+	{0x1d, 0x2e},  /* for Iso */
+	{0x1d, 0x1e},
+};
+
 static const u8 bridge_init[][2] = {
 	{0x88, 0xf8},
 	{0x89, 0xff},
@@ -898,7 +1167,7 @@ static int sccb_check_status(struct gspca_dev *gspca_dev)
 	int i;
 
 	for (i = 0; i < 5; i++) {
-		msleep(10);
+		msleep(20);
 		data = reg_r(gspca_dev, OV534_REG_STATUS);
 
 		switch (data) {
@@ -1221,6 +1490,13 @@ static int sd_init(struct gspca_dev *gspca_dev)
 		sccb_w_array(gspca_dev, ov562x_init_2,
 				ARRAY_SIZE(ov562x_init_2));
 		reg_w(gspca_dev, 0xe0, 0x00);
+	} else if ((sensor_id & 0xfff0) == 0x3610) {
+		sd->sensor = SENSOR_OV361x;
+		gspca_dev->cam.cam_mode = ov361x_mode;
+		gspca_dev->cam.nmodes = ARRAY_SIZE(ov361x_mode);
+		reg_w(gspca_dev, 0xe7, 0x3a);
+		reg_w(gspca_dev, 0xf1, 0x60);
+		sccb_write(gspca_dev, 0x12, 0x80);
 	} else {
 		pr_err("Unknown sensor %04x", sensor_id);
 		return -EINVAL;
@@ -1229,6 +1505,53 @@ static int sd_init(struct gspca_dev *gspca_dev)
 	return gspca_dev->usb_err;
 }
 
+static int sd_start_ov361x(struct gspca_dev *gspca_dev)
+{
+	sccb_write(gspca_dev, 0x12, 0x80);
+	msleep(20);
+	switch (gspca_dev->curr_mode % (ov361x_last)) {
+	case ov361x_2048:
+		reg_w_array(gspca_dev, ov361x_bridge_start_2048,
+			    ARRAY_SIZE(ov361x_bridge_start_2048));
+		sccb_w_array(gspca_dev, ov361x_start_2048,
+			     ARRAY_SIZE(ov361x_start_2048));
+		break;
+	case ov361x_1600:
+		reg_w_array(gspca_dev, ov361x_bridge_start_1600,
+			    ARRAY_SIZE(ov361x_bridge_start_1600));
+		sccb_w_array(gspca_dev, ov361x_start_1600,
+			     ARRAY_SIZE(ov361x_start_1600));
+		break;
+	case ov361x_1024:
+		reg_w_array(gspca_dev, ov361x_bridge_start_1024,
+			    ARRAY_SIZE(ov361x_bridge_start_1024));
+		sccb_w_array(gspca_dev, ov361x_start_1024,
+			     ARRAY_SIZE(ov361x_start_1024));
+		break;
+	case ov361x_640:
+		reg_w_array(gspca_dev, ov361x_bridge_start_640,
+			    ARRAY_SIZE(ov361x_bridge_start_640));
+		sccb_w_array(gspca_dev, ov361x_start_640,
+			     ARRAY_SIZE(ov361x_start_640));
+		break;
+	case ov361x_320:
+		reg_w_array(gspca_dev, ov361x_bridge_start_320,
+			    ARRAY_SIZE(ov361x_bridge_start_320));
+		sccb_w_array(gspca_dev, ov361x_start_320,
+			     ARRAY_SIZE(ov361x_start_320));
+		break;
+	case ov361x_160:
+		reg_w_array(gspca_dev, ov361x_bridge_start_160,
+			    ARRAY_SIZE(ov361x_bridge_start_160));
+		sccb_w_array(gspca_dev, ov361x_start_160,
+			     ARRAY_SIZE(ov361x_start_160));
+		break;
+	}
+	reg_w(gspca_dev, 0xe0, 0x00); /* start transfer */
+
+	return gspca_dev->usb_err;
+}
+
 static int sd_start(struct gspca_dev *gspca_dev)
 {
 	struct sd *sd = (struct sd *) gspca_dev;
@@ -1237,6 +1560,8 @@ static int sd_start(struct gspca_dev *gspca_dev)
 		return gspca_dev->usb_err;
 	if (sd->sensor == SENSOR_OV562x)
 		return gspca_dev->usb_err;
+	if (sd->sensor == SENSOR_OV361x)
+		return sd_start_ov361x(gspca_dev);
 
 	switch (gspca_dev->curr_mode) {
 	case QVGA_MODE:			/* 320x240 */
@@ -1290,6 +1615,11 @@ static int sd_start(struct gspca_dev *gspca_dev)
 
 static void sd_stopN(struct gspca_dev *gspca_dev)
 {
+	if (((struct sd *)gspca_dev)->sensor == SENSOR_OV361x) {
+		reg_w(gspca_dev, 0xe0, 0x01); /* stop transfer */
+		/* reg_w(gspca_dev, 0x31, 0x09); */
+		return;
+	}
 	reg_w(gspca_dev, 0xe0, 0x01);
 	set_led(gspca_dev, 0);
 	reg_w(gspca_dev, 0xe0, 0x00);
@@ -1425,6 +1755,8 @@ static int sd_init_controls(struct gspca_dev *gspca_dev)
 
 	if (sd->sensor == SENSOR_OV971x)
 		return 0;
+	if (sd->sensor == SENSOR_OV361x)
+		return 0;
 	gspca_dev->vdev.ctrl_handler = hdl;
 	v4l2_ctrl_handler_init(hdl, 7);
 	if (sd->sensor == SENSOR_OV562x) {
diff --git a/drivers/media/usb/gspca/pac207.c b/drivers/media/usb/gspca/pac207.c
index 83519be94e58..cd79c180f67b 100644
--- a/drivers/media/usb/gspca/pac207.c
+++ b/drivers/media/usb/gspca/pac207.c
@@ -299,7 +299,7 @@ static int sd_start(struct gspca_dev *gspca_dev)
 	pac207_write_regs(gspca_dev, 0x0042, pac207_sensor_init[3], 8);
 
 	/* Compression Balance */
-	if (gspca_dev->width == 176)
+	if (gspca_dev->pixfmt.width == 176)
 		pac207_write_reg(gspca_dev, 0x4a, 0xff);
 	else
 		pac207_write_reg(gspca_dev, 0x4a, 0x30);
@@ -317,7 +317,7 @@ static int sd_start(struct gspca_dev *gspca_dev)
 		mode = 0x00;
 	else
 		mode = 0x02;
-	if (gspca_dev->width == 176) {	/* 176x144 */
+	if (gspca_dev->pixfmt.width == 176) {	/* 176x144 */
 		mode |= 0x01;
 		PDEBUG(D_STREAM, "pac207_start mode 176x144");
 	} else {				/* 352x288 */
diff --git a/drivers/media/usb/gspca/pac7311.c b/drivers/media/usb/gspca/pac7311.c
index 1a5bdc853a80..25f86b1e74a8 100644
--- a/drivers/media/usb/gspca/pac7311.c
+++ b/drivers/media/usb/gspca/pac7311.c
@@ -326,7 +326,7 @@ static void setexposure(struct gspca_dev *gspca_dev, s32 val)
 	 *  640x480 mode and page 4 reg 2 <= 3 then it must be 9
 	 */
 	reg_w(gspca_dev, 0xff, 0x01);
-	if (gspca_dev->width != 640 && val <= 3)
+	if (gspca_dev->pixfmt.width != 640 && val <= 3)
 		reg_w(gspca_dev, 0x08, 0x09);
 	else
 		reg_w(gspca_dev, 0x08, 0x08);
@@ -337,7 +337,7 @@ static void setexposure(struct gspca_dev *gspca_dev, s32 val)
 	 * camera to use higher compression or we may run out of
 	 * bandwidth.
 	 */
-	if (gspca_dev->width == 640 && val == 2)
+	if (gspca_dev->pixfmt.width == 640 && val == 2)
 		reg_w(gspca_dev, 0x80, 0x01);
 	else
 		reg_w(gspca_dev, 0x80, 0x1c);
@@ -615,7 +615,7 @@ static void sd_pkt_scan(struct gspca_dev *gspca_dev,
 
 		/* Start the new frame with the jpeg header */
 		pac_start_frame(gspca_dev,
-			gspca_dev->height, gspca_dev->width);
+			gspca_dev->pixfmt.height, gspca_dev->pixfmt.width);
 	}
 	gspca_frame_add(gspca_dev, INTER_PACKET, data, len);
 }
diff --git a/drivers/media/usb/gspca/se401.c b/drivers/media/usb/gspca/se401.c
index 5f729b8aa2bd..5102cea50471 100644
--- a/drivers/media/usb/gspca/se401.c
+++ b/drivers/media/usb/gspca/se401.c
@@ -354,9 +354,9 @@ static int sd_start(struct gspca_dev *gspca_dev)
 
 	/* set size + mode */
 	se401_write_req(gspca_dev, SE401_REQ_SET_WIDTH,
-			gspca_dev->width * mult, 0);
+			gspca_dev->pixfmt.width * mult, 0);
 	se401_write_req(gspca_dev, SE401_REQ_SET_HEIGHT,
-			gspca_dev->height * mult, 0);
+			gspca_dev->pixfmt.height * mult, 0);
 	/*
 	 * HDG: disabled this as it does not seem to do anything
 	 * se401_write_req(gspca_dev, SE401_REQ_SET_OUTPUT_MODE,
@@ -480,7 +480,7 @@ static void sd_complete_frame(struct gspca_dev *gspca_dev, u8 *data, int len)
 static void sd_pkt_scan_janggu(struct gspca_dev *gspca_dev, u8 *data, int len)
 {
 	struct sd *sd = (struct sd *)gspca_dev;
-	int imagesize = gspca_dev->width * gspca_dev->height;
+	int imagesize = gspca_dev->pixfmt.width * gspca_dev->pixfmt.height;
 	int i, plen, bits, pixels, info, count;
 
 	if (sd->restart_stream)
diff --git a/drivers/media/usb/gspca/sn9c20x.c b/drivers/media/usb/gspca/sn9c20x.c
index f4453d52801b..2a38621cf718 100644
--- a/drivers/media/usb/gspca/sn9c20x.c
+++ b/drivers/media/usb/gspca/sn9c20x.c
@@ -1955,7 +1955,7 @@ static int sd_isoc_init(struct gspca_dev *gspca_dev)
 			return 0;
 		}
 
-		switch (gspca_dev->width) {
+		switch (gspca_dev->pixfmt.width) {
 		case 160: /* 160x120 */
 			gspca_dev->alt = 2;
 			break;
@@ -1985,8 +1985,8 @@ static int sd_start(struct gspca_dev *gspca_dev)
 {
 	struct sd *sd = (struct sd *) gspca_dev;
 	int mode = gspca_dev->cam.cam_mode[(int) gspca_dev->curr_mode].priv;
-	int width = gspca_dev->width;
-	int height = gspca_dev->height;
+	int width = gspca_dev->pixfmt.width;
+	int height = gspca_dev->pixfmt.height;
 	u8 fmt, scale = 0;
 
 	jpeg_define(sd->jpeg_hdr, height, width,
diff --git a/drivers/media/usb/gspca/sonixb.c b/drivers/media/usb/gspca/sonixb.c
index d7ff3b9687c5..7277dbd2afcd 100644
--- a/drivers/media/usb/gspca/sonixb.c
+++ b/drivers/media/usb/gspca/sonixb.c
@@ -513,10 +513,7 @@ static void i2c_w(struct gspca_dev *gspca_dev, const u8 *buf)
 		if (gspca_dev->usb_buf[0] & 0x04) {
 			if (gspca_dev->usb_buf[0] & 0x08) {
 				dev_err(gspca_dev->v4l2_dev.dev,
-					"i2c error writing %02x %02x %02x %02x"
-					" %02x %02x %02x %02x\n",
-					buf[0], buf[1], buf[2], buf[3],
-					buf[4], buf[5], buf[6], buf[7]);
+					"i2c error writing %8ph\n", buf);
 				gspca_dev->usb_err = -EIO;
 			}
 			return;
@@ -753,7 +750,7 @@ static void setexposure(struct gspca_dev *gspca_dev)
 		/* In 640x480, if the reg11 has less than 4, the image is
 		   unstable (the bridge goes into a higher compression mode
 		   which we have not reverse engineered yet). */
-		if (gspca_dev->width == 640 && reg11 < 4)
+		if (gspca_dev->pixfmt.width == 640 && reg11 < 4)
 			reg11 = 4;
 
 		/* frame exposure time in ms = 1000 * reg11 / 30    ->
diff --git a/drivers/media/usb/gspca/sonixj.c b/drivers/media/usb/gspca/sonixj.c
index 3b5ccb1c4cdf..c69b45d7cfbf 100644
--- a/drivers/media/usb/gspca/sonixj.c
+++ b/drivers/media/usb/gspca/sonixj.c
@@ -2204,7 +2204,8 @@ static int sd_start(struct gspca_dev *gspca_dev)
 				{ 0x14, 0xe7, 0x1e, 0xdd };
 
 	/* create the JPEG header */
-	jpeg_define(sd->jpeg_hdr, gspca_dev->height, gspca_dev->width,
+	jpeg_define(sd->jpeg_hdr, gspca_dev->pixfmt.height,
+			gspca_dev->pixfmt.width,
 			0x21);		/* JPEG 422 */
 
 	/* initialize the bridge */
diff --git a/drivers/media/usb/gspca/spca1528.c b/drivers/media/usb/gspca/spca1528.c
index 688592b289ea..f38fd8949609 100644
--- a/drivers/media/usb/gspca/spca1528.c
+++ b/drivers/media/usb/gspca/spca1528.c
@@ -255,7 +255,8 @@ static int sd_start(struct gspca_dev *gspca_dev)
 	struct sd *sd = (struct sd *) gspca_dev;
 
 	/* initialize the JPEG header */
-	jpeg_define(sd->jpeg_hdr, gspca_dev->height, gspca_dev->width,
+	jpeg_define(sd->jpeg_hdr, gspca_dev->pixfmt.height,
+			gspca_dev->pixfmt.width,
 			0x22);		/* JPEG 411 */
 
 	/* the JPEG quality shall be 85% */
diff --git a/drivers/media/usb/gspca/spca500.c b/drivers/media/usb/gspca/spca500.c
index 9f8bf51fd64b..f011a309dd65 100644
--- a/drivers/media/usb/gspca/spca500.c
+++ b/drivers/media/usb/gspca/spca500.c
@@ -608,7 +608,8 @@ static int sd_start(struct gspca_dev *gspca_dev)
 	__u8 xmult, ymult;
 
 	/* create the JPEG header */
-	jpeg_define(sd->jpeg_hdr, gspca_dev->height, gspca_dev->width,
+	jpeg_define(sd->jpeg_hdr, gspca_dev->pixfmt.height,
+			gspca_dev->pixfmt.width,
 			0x22);		/* JPEG 411 */
 	jpeg_set_qual(sd->jpeg_hdr, QUALITY);
 
diff --git a/drivers/media/usb/gspca/sq905c.c b/drivers/media/usb/gspca/sq905c.c
index acb19fb9a3df..aa21edc9502d 100644
--- a/drivers/media/usb/gspca/sq905c.c
+++ b/drivers/media/usb/gspca/sq905c.c
@@ -272,7 +272,7 @@ static int sd_start(struct gspca_dev *gspca_dev)
 
 	dev->cap_mode = gspca_dev->cam.cam_mode;
 	/* "Open the shutter" and set size, to start capture */
-	switch (gspca_dev->width) {
+	switch (gspca_dev->pixfmt.width) {
 	case 640:
 		PDEBUG(D_STREAM, "Start streaming at high resolution");
 		dev->cap_mode++;
diff --git a/drivers/media/usb/gspca/sq930x.c b/drivers/media/usb/gspca/sq930x.c
index b10d0821111c..e274cf19a3ea 100644
--- a/drivers/media/usb/gspca/sq930x.c
+++ b/drivers/media/usb/gspca/sq930x.c
@@ -906,7 +906,8 @@ static int sd_isoc_init(struct gspca_dev *gspca_dev)
 
 	gspca_dev->cam.bulk_nurbs = 1;	/* there must be one URB only */
 	sd->do_ctrl = 0;
-	gspca_dev->cam.bulk_size = gspca_dev->width * gspca_dev->height + 8;
+	gspca_dev->cam.bulk_size = gspca_dev->pixfmt.width *
+			gspca_dev->pixfmt.height + 8;
 	return 0;
 }
 
diff --git a/drivers/media/usb/gspca/stk014.c b/drivers/media/usb/gspca/stk014.c
index 8c0982607f25..b0c70fea760b 100644
--- a/drivers/media/usb/gspca/stk014.c
+++ b/drivers/media/usb/gspca/stk014.c
@@ -250,7 +250,8 @@ static int sd_start(struct gspca_dev *gspca_dev)
 	int ret, value;
 
 	/* create the JPEG header */
-	jpeg_define(sd->jpeg_hdr, gspca_dev->height, gspca_dev->width,
+	jpeg_define(sd->jpeg_hdr, gspca_dev->pixfmt.height,
+			gspca_dev->pixfmt.width,
 			0x22);		/* JPEG 411 */
 	jpeg_set_qual(sd->jpeg_hdr, QUALITY);
 
@@ -261,7 +262,7 @@ static int sd_start(struct gspca_dev *gspca_dev)
 	set_par(gspca_dev, 0x00000000);
 	set_par(gspca_dev, 0x8002e001);
 	set_par(gspca_dev, 0x14000000);
-	if (gspca_dev->width > 320)
+	if (gspca_dev->pixfmt.width > 320)
 		value = 0x8002e001;		/* 640x480 */
 	else
 		value = 0x4001f000;		/* 320x240 */
diff --git a/drivers/media/usb/gspca/stk1135.c b/drivers/media/usb/gspca/stk1135.c
index 585868835ace..1fc80af2a189 100644
--- a/drivers/media/usb/gspca/stk1135.c
+++ b/drivers/media/usb/gspca/stk1135.c
@@ -48,42 +48,11 @@ struct sd {
 };
 
 static const struct v4l2_pix_format stk1135_modes[] = {
-	{160, 120, V4L2_PIX_FMT_SBGGR8, V4L2_FIELD_NONE,
-		.bytesperline = 160,
-		.sizeimage = 160 * 120,
-		.colorspace = V4L2_COLORSPACE_SRGB},
-	{176, 144, V4L2_PIX_FMT_SBGGR8, V4L2_FIELD_NONE,
-		.bytesperline = 176,
-		.sizeimage = 176 * 144,
-		.colorspace = V4L2_COLORSPACE_SRGB},
-	{320, 240, V4L2_PIX_FMT_SBGGR8, V4L2_FIELD_NONE,
-		.bytesperline = 320,
-		.sizeimage = 320 * 240,
-		.colorspace = V4L2_COLORSPACE_SRGB},
-	{352, 288, V4L2_PIX_FMT_SBGGR8, V4L2_FIELD_NONE,
-		.bytesperline = 352,
-		.sizeimage = 352 * 288,
-		.colorspace = V4L2_COLORSPACE_SRGB},
+	/* default mode (this driver supports variable resolution) */
 	{640, 480, V4L2_PIX_FMT_SBGGR8, V4L2_FIELD_NONE,
 		.bytesperline = 640,
 		.sizeimage = 640 * 480,
 		.colorspace = V4L2_COLORSPACE_SRGB},
-	{720, 576, V4L2_PIX_FMT_SBGGR8, V4L2_FIELD_NONE,
-		.bytesperline = 720,
-		.sizeimage = 720 * 576,
-		.colorspace = V4L2_COLORSPACE_SRGB},
-	{800, 600, V4L2_PIX_FMT_SBGGR8, V4L2_FIELD_NONE,
-		.bytesperline = 800,
-		.sizeimage = 800 * 600,
-		.colorspace = V4L2_COLORSPACE_SRGB},
-	{1024, 768, V4L2_PIX_FMT_SBGGR8, V4L2_FIELD_NONE,
-		.bytesperline = 1024,
-		.sizeimage = 1024 * 768,
-		.colorspace = V4L2_COLORSPACE_SRGB},
-	{1280, 1024, V4L2_PIX_FMT_SBGGR8, V4L2_FIELD_NONE,
-		.bytesperline = 1280,
-		.sizeimage = 1280 * 1024,
-		.colorspace = V4L2_COLORSPACE_SRGB},
 };
 
 /* -- read a register -- */
@@ -347,16 +316,16 @@ static void stk1135_configure_mt9m112(struct gspca_dev *gspca_dev)
 		sensor_write(gspca_dev, cfg[i].reg, cfg[i].val);
 
 	/* set output size */
-	width = gspca_dev->cam.cam_mode[gspca_dev->curr_mode].width;
-	height = gspca_dev->cam.cam_mode[gspca_dev->curr_mode].height;
-	if (width <= 640) { /* use context A (half readout speed by default) */
+	width = gspca_dev->pixfmt.width;
+	height = gspca_dev->pixfmt.height;
+	if (width <= 640 && height <= 512) { /* context A (half readout speed)*/
 		sensor_write(gspca_dev, 0x1a7, width);
 		sensor_write(gspca_dev, 0x1aa, height);
 		/* set read mode context A */
 		sensor_write(gspca_dev, 0x0c8, 0x0000);
 		/* set resize, read mode, vblank, hblank context A */
 		sensor_write(gspca_dev, 0x2c8, 0x0000);
-	} else { /* use context B (full readout speed by default) */
+	} else { /* context B (full readout speed) */
 		sensor_write(gspca_dev, 0x1a1, width);
 		sensor_write(gspca_dev, 0x1a4, height);
 		/* set read mode context B */
@@ -484,8 +453,8 @@ static int sd_start(struct gspca_dev *gspca_dev)
 	reg_w(gspca_dev, STK1135_REG_CISPO + 3, 0x00);
 
 	/* set capture end position */
-	width = gspca_dev->cam.cam_mode[gspca_dev->curr_mode].width;
-	height = gspca_dev->cam.cam_mode[gspca_dev->curr_mode].height;
+	width = gspca_dev->pixfmt.width;
+	height = gspca_dev->pixfmt.height;
 	reg_w(gspca_dev, STK1135_REG_CIEPO + 0, width & 0xff);
 	reg_w(gspca_dev, STK1135_REG_CIEPO + 1, width >> 8);
 	reg_w(gspca_dev, STK1135_REG_CIEPO + 2, height & 0xff);
@@ -643,6 +612,35 @@ static int sd_init_controls(struct gspca_dev *gspca_dev)
 	return 0;
 }
 
+static void stk1135_try_fmt(struct gspca_dev *gspca_dev, struct v4l2_format *fmt)
+{
+	fmt->fmt.pix.width = clamp(fmt->fmt.pix.width, 32U, 1280U);
+	fmt->fmt.pix.height = clamp(fmt->fmt.pix.height, 32U, 1024U);
+	/* round up to even numbers */
+	fmt->fmt.pix.width += (fmt->fmt.pix.width & 1);
+	fmt->fmt.pix.height += (fmt->fmt.pix.height & 1);
+
+	fmt->fmt.pix.bytesperline = fmt->fmt.pix.width;
+	fmt->fmt.pix.sizeimage = fmt->fmt.pix.width * fmt->fmt.pix.height;
+}
+
+static int stk1135_enum_framesizes(struct gspca_dev *gspca_dev,
+			struct v4l2_frmsizeenum *fsize)
+{
+	if (fsize->index != 0 || fsize->pixel_format != V4L2_PIX_FMT_SBGGR8)
+		return -EINVAL;
+
+	fsize->type = V4L2_FRMSIZE_TYPE_STEPWISE;
+	fsize->stepwise.min_width = 32;
+	fsize->stepwise.min_height = 32;
+	fsize->stepwise.max_width = 1280;
+	fsize->stepwise.max_height = 1024;
+	fsize->stepwise.step_width = 2;
+	fsize->stepwise.step_height = 2;
+
+	return 0;
+}
+
 /* sub-driver description */
 static const struct sd_desc sd_desc = {
 	.name = MODULE_NAME,
@@ -653,6 +651,8 @@ static const struct sd_desc sd_desc = {
 	.stopN = sd_stopN,
 	.pkt_scan = sd_pkt_scan,
 	.dq_callback = stk1135_dq_callback,
+	.try_fmt = stk1135_try_fmt,
+	.enum_framesizes = stk1135_enum_framesizes,
 };
 
 /* -- module initialisation -- */
diff --git a/drivers/media/usb/gspca/stv06xx/stv06xx.c b/drivers/media/usb/gspca/stv06xx/stv06xx.c
index 55ee7a61c67f..49d209bbf9ee 100644
--- a/drivers/media/usb/gspca/stv06xx/stv06xx.c
+++ b/drivers/media/usb/gspca/stv06xx/stv06xx.c
@@ -452,7 +452,7 @@ frame_data:
 					NULL, 0);
 
 			if (sd->bridge == BRIDGE_ST6422)
-				sd->to_skip = gspca_dev->width * 4;
+				sd->to_skip = gspca_dev->pixfmt.width * 4;
 
 			if (chunk_len)
 				PERR("Chunk length is "
diff --git a/drivers/media/usb/gspca/stv06xx/stv06xx_pb0100.c b/drivers/media/usb/gspca/stv06xx/stv06xx_pb0100.c
index 8206b7743300..8d785edcccf2 100644
--- a/drivers/media/usb/gspca/stv06xx/stv06xx_pb0100.c
+++ b/drivers/media/usb/gspca/stv06xx/stv06xx_pb0100.c
@@ -421,7 +421,7 @@ static int pb0100_set_autogain_target(struct gspca_dev *gspca_dev, __s32 val)
 
 	/* Number of pixels counted by the sensor when subsampling the pixels.
 	 * Slightly larger than the real value to avoid oscillation */
-	totalpixels = gspca_dev->width * gspca_dev->height;
+	totalpixels = gspca_dev->pixfmt.width * gspca_dev->pixfmt.height;
 	totalpixels = totalpixels/(8*8) + totalpixels/(64*64);
 
 	brightpixels = (totalpixels * val) >> 8;
diff --git a/drivers/media/usb/gspca/sunplus.c b/drivers/media/usb/gspca/sunplus.c
index af8767a9bd4c..a517d185febe 100644
--- a/drivers/media/usb/gspca/sunplus.c
+++ b/drivers/media/usb/gspca/sunplus.c
@@ -715,7 +715,8 @@ static int sd_start(struct gspca_dev *gspca_dev)
 	int enable;
 
 	/* create the JPEG header */
-	jpeg_define(sd->jpeg_hdr, gspca_dev->height, gspca_dev->width,
+	jpeg_define(sd->jpeg_hdr, gspca_dev->pixfmt.height,
+			gspca_dev->pixfmt.width,
 			0x22);		/* JPEG 411 */
 	jpeg_set_qual(sd->jpeg_hdr, QUALITY);
 
diff --git a/drivers/media/usb/gspca/topro.c b/drivers/media/usb/gspca/topro.c
index 4cb511ccc5f6..640c2fe760b3 100644
--- a/drivers/media/usb/gspca/topro.c
+++ b/drivers/media/usb/gspca/topro.c
@@ -3856,7 +3856,7 @@ static void setsharpness(struct gspca_dev *gspca_dev, s32 val)
 
 	if (sd->bridge == BRIDGE_TP6800) {
 		val |= 0x08;		/* grid compensation enable */
-		if (gspca_dev->width == 640)
+		if (gspca_dev->pixfmt.width == 640)
 			reg_w(gspca_dev, TP6800_R78_FORMAT, 0x00); /* vga */
 		else
 			val |= 0x04;		/* scaling down enable */
@@ -3880,7 +3880,7 @@ static void set_resolution(struct gspca_dev *gspca_dev)
 	struct sd *sd = (struct sd *) gspca_dev;
 
 	reg_w(gspca_dev, TP6800_R21_ENDP_1_CTL, 0x00);
-	if (gspca_dev->width == 320) {
+	if (gspca_dev->pixfmt.width == 320) {
 		reg_w(gspca_dev, TP6800_R3F_FRAME_RATE, 0x06);
 		msleep(100);
 		i2c_w(gspca_dev, CX0342_AUTO_ADC_CALIB, 0x01);
@@ -3924,7 +3924,7 @@ static int get_fr_idx(struct gspca_dev *gspca_dev)
 
 		/* 640x480 * 30 fps does not work */
 		if (i == 6			/* if 30 fps */
-		 && gspca_dev->width == 640)
+		 && gspca_dev->pixfmt.width == 640)
 			i = 0x05;		/* 15 fps */
 	} else {
 		for (i = 0; i < ARRAY_SIZE(rates_6810) - 1; i++) {
@@ -3935,7 +3935,7 @@ static int get_fr_idx(struct gspca_dev *gspca_dev)
 
 		/* 640x480 * 30 fps does not work */
 		if (i == 7			/* if 30 fps */
-		 && gspca_dev->width == 640)
+		 && gspca_dev->pixfmt.width == 640)
 			i = 6;			/* 15 fps */
 		i |= 0x80;			/* clock * 1 */
 	}
@@ -4554,7 +4554,8 @@ static int sd_start(struct gspca_dev *gspca_dev)
 {
 	struct sd *sd = (struct sd *) gspca_dev;
 
-	jpeg_define(sd->jpeg_hdr, gspca_dev->height, gspca_dev->width);
+	jpeg_define(sd->jpeg_hdr, gspca_dev->pixfmt.height,
+			gspca_dev->pixfmt.width);
 	set_dqt(gspca_dev, sd->quality);
 	if (sd->bridge == BRIDGE_TP6800) {
 		if (sd->sensor == SENSOR_CX0342)
@@ -4737,7 +4738,7 @@ static void sd_dq_callback(struct gspca_dev *gspca_dev)
 			(gspca_dev->usb_buf[26] << 8) + gspca_dev->usb_buf[25] +
 			(gspca_dev->usb_buf[29] << 8) + gspca_dev->usb_buf[28])
 				/ 8;
-		if (gspca_dev->width == 640)
+		if (gspca_dev->pixfmt.width == 640)
 			luma /= 4;
 		reg_w(gspca_dev, 0x7d, 0x00);
 
diff --git a/drivers/media/usb/gspca/tv8532.c b/drivers/media/usb/gspca/tv8532.c
index 8591324a53e1..d497ba38af0d 100644
--- a/drivers/media/usb/gspca/tv8532.c
+++ b/drivers/media/usb/gspca/tv8532.c
@@ -268,7 +268,7 @@ static void sd_pkt_scan(struct gspca_dev *gspca_dev,
 	packet_type0 = packet_type1 = INTER_PACKET;
 	if (gspca_dev->empty_packet) {
 		gspca_dev->empty_packet = 0;
-		sd->packet = gspca_dev->height / 2;
+		sd->packet = gspca_dev->pixfmt.height / 2;
 		packet_type0 = FIRST_PACKET;
 	} else if (sd->packet == 0)
 		return;			/* 2 more lines in 352x288 ! */
@@ -284,9 +284,10 @@ static void sd_pkt_scan(struct gspca_dev *gspca_dev,
 	 * - 4 bytes
 	 */
 	gspca_frame_add(gspca_dev, packet_type0,
-			data + 2, gspca_dev->width);
+			data + 2, gspca_dev->pixfmt.width);
 	gspca_frame_add(gspca_dev, packet_type1,
-			data + gspca_dev->width + 5, gspca_dev->width);
+			data + gspca_dev->pixfmt.width + 5,
+			gspca_dev->pixfmt.width);
 }
 
 static int sd_s_ctrl(struct v4l2_ctrl *ctrl)
diff --git a/drivers/media/usb/gspca/vicam.c b/drivers/media/usb/gspca/vicam.c
index a2275cfe0b81..103f6c4236b0 100644
--- a/drivers/media/usb/gspca/vicam.c
+++ b/drivers/media/usb/gspca/vicam.c
@@ -121,13 +121,13 @@ static int vicam_read_frame(struct gspca_dev *gspca_dev, u8 *data, int size)
 
 	memset(req_data, 0, 16);
 	req_data[0] = gain;
-	if (gspca_dev->width == 256)
+	if (gspca_dev->pixfmt.width == 256)
 		req_data[1] |= 0x01; /* low nibble x-scale */
-	if (gspca_dev->height <= 122) {
+	if (gspca_dev->pixfmt.height <= 122) {
 		req_data[1] |= 0x10; /* high nibble y-scale */
-		unscaled_height = gspca_dev->height * 2;
+		unscaled_height = gspca_dev->pixfmt.height * 2;
 	} else
-		unscaled_height = gspca_dev->height;
+		unscaled_height = gspca_dev->pixfmt.height;
 	req_data[2] = 0x90; /* unknown, does not seem to do anything */
 	if (unscaled_height <= 200)
 		req_data[3] = 0x06; /* vend? */
diff --git a/drivers/media/usb/gspca/w996Xcf.c b/drivers/media/usb/gspca/w996Xcf.c
index 2165da0c7ce1..fb9fe2ef3a6f 100644
--- a/drivers/media/usb/gspca/w996Xcf.c
+++ b/drivers/media/usb/gspca/w996Xcf.c
@@ -430,11 +430,11 @@ static void w9968cf_set_crop_window(struct sd *sd)
 	#define SC(x) ((x) << 10)
 
 	/* Scaling factors */
-	fw = SC(sd->gspca_dev.width) / max_width;
-	fh = SC(sd->gspca_dev.height) / max_height;
+	fw = SC(sd->gspca_dev.pixfmt.width) / max_width;
+	fh = SC(sd->gspca_dev.pixfmt.height) / max_height;
 
-	cw = (fw >= fh) ? max_width : SC(sd->gspca_dev.width) / fh;
-	ch = (fw >= fh) ? SC(sd->gspca_dev.height) / fw : max_height;
+	cw = (fw >= fh) ? max_width : SC(sd->gspca_dev.pixfmt.width) / fh;
+	ch = (fw >= fh) ? SC(sd->gspca_dev.pixfmt.height) / fw : max_height;
 
 	sd->sensor_width = max_width;
 	sd->sensor_height = max_height;
@@ -454,34 +454,34 @@ static void w9968cf_mode_init_regs(struct sd *sd)
 
 	w9968cf_set_crop_window(sd);
 
-	reg_w(sd, 0x14, sd->gspca_dev.width);
-	reg_w(sd, 0x15, sd->gspca_dev.height);
+	reg_w(sd, 0x14, sd->gspca_dev.pixfmt.width);
+	reg_w(sd, 0x15, sd->gspca_dev.pixfmt.height);
 
 	/* JPEG width & height */
-	reg_w(sd, 0x30, sd->gspca_dev.width);
-	reg_w(sd, 0x31, sd->gspca_dev.height);
+	reg_w(sd, 0x30, sd->gspca_dev.pixfmt.width);
+	reg_w(sd, 0x31, sd->gspca_dev.pixfmt.height);
 
 	/* Y & UV frame buffer strides (in WORD) */
 	if (w9968cf_vga_mode[sd->gspca_dev.curr_mode].pixelformat ==
 	    V4L2_PIX_FMT_JPEG) {
-		reg_w(sd, 0x2c, sd->gspca_dev.width / 2);
-		reg_w(sd, 0x2d, sd->gspca_dev.width / 4);
+		reg_w(sd, 0x2c, sd->gspca_dev.pixfmt.width / 2);
+		reg_w(sd, 0x2d, sd->gspca_dev.pixfmt.width / 4);
 	} else
-		reg_w(sd, 0x2c, sd->gspca_dev.width);
+		reg_w(sd, 0x2c, sd->gspca_dev.pixfmt.width);
 
 	reg_w(sd, 0x00, 0xbf17); /* reset everything */
 	reg_w(sd, 0x00, 0xbf10); /* normal operation */
 
 	/* Transfer size in WORDS (for UYVY format only) */
-	val = sd->gspca_dev.width * sd->gspca_dev.height;
+	val = sd->gspca_dev.pixfmt.width * sd->gspca_dev.pixfmt.height;
 	reg_w(sd, 0x3d, val & 0xffff); /* low bits */
 	reg_w(sd, 0x3e, val >> 16);    /* high bits */
 
 	if (w9968cf_vga_mode[sd->gspca_dev.curr_mode].pixelformat ==
 	    V4L2_PIX_FMT_JPEG) {
 		/* We may get called multiple times (usb isoc bw negotiat.) */
-		jpeg_define(sd->jpeg_hdr, sd->gspca_dev.height,
-			    sd->gspca_dev.width, 0x22); /* JPEG 420 */
+		jpeg_define(sd->jpeg_hdr, sd->gspca_dev.pixfmt.height,
+			    sd->gspca_dev.pixfmt.width, 0x22); /* JPEG 420 */
 		jpeg_set_qual(sd->jpeg_hdr, v4l2_ctrl_g_ctrl(sd->jpegqual));
 		w9968cf_upload_quantizationtables(sd);
 		v4l2_ctrl_grab(sd->jpegqual, true);
diff --git a/drivers/media/usb/gspca/xirlink_cit.c b/drivers/media/usb/gspca/xirlink_cit.c
index 7eaf64eb867c..a41aa7817c54 100644
--- a/drivers/media/usb/gspca/xirlink_cit.c
+++ b/drivers/media/usb/gspca/xirlink_cit.c
@@ -1471,14 +1471,14 @@ static int cit_get_clock_div(struct gspca_dev *gspca_dev)
 
 	while (clock_div > 3 &&
 			1000 * packet_size >
-			gspca_dev->width * gspca_dev->height *
+			gspca_dev->pixfmt.width * gspca_dev->pixfmt.height *
 			fps[clock_div - 1] * 3 / 2)
 		clock_div--;
 
 	PDEBUG(D_PROBE,
 	       "PacketSize: %d, res: %dx%d -> using clockdiv: %d (%d fps)",
-	       packet_size, gspca_dev->width, gspca_dev->height, clock_div,
-	       fps[clock_div]);
+	       packet_size, gspca_dev->pixfmt.width, gspca_dev->pixfmt.height,
+	       clock_div, fps[clock_div]);
 
 	return clock_div;
 }
@@ -1502,7 +1502,7 @@ static int cit_start_model0(struct gspca_dev *gspca_dev)
 	cit_write_reg(gspca_dev, 0x0002, 0x0426);
 	cit_write_reg(gspca_dev, 0x0014, 0x0427);
 
-	switch (gspca_dev->width) {
+	switch (gspca_dev->pixfmt.width) {
 	case 160: /* 160x120 */
 		cit_write_reg(gspca_dev, 0x0004, 0x010b);
 		cit_write_reg(gspca_dev, 0x0001, 0x010a);
@@ -1643,7 +1643,7 @@ static int cit_start_model1(struct gspca_dev *gspca_dev)
 	cit_write_reg(gspca_dev, 0x00, 0x0101);
 	cit_write_reg(gspca_dev, 0x00, 0x010a);
 
-	switch (gspca_dev->width) {
+	switch (gspca_dev->pixfmt.width) {
 	case 128: /* 128x96 */
 		cit_write_reg(gspca_dev, 0x80, 0x0103);
 		cit_write_reg(gspca_dev, 0x60, 0x0105);
@@ -1700,7 +1700,7 @@ static int cit_start_model1(struct gspca_dev *gspca_dev)
 	}
 
 	/* Assorted init */
-	switch (gspca_dev->width) {
+	switch (gspca_dev->pixfmt.width) {
 	case 128: /* 128x96 */
 		cit_Packet_Format1(gspca_dev, 0x2b, 0x1e);
 		cit_write_reg(gspca_dev, 0xc9, 0x0119);	/* Same everywhere */
@@ -1753,7 +1753,7 @@ static int cit_start_model2(struct gspca_dev *gspca_dev)
 	cit_write_reg(gspca_dev, 0x0000, 0x0108);
 	cit_write_reg(gspca_dev, 0x0001, 0x0133);
 	cit_write_reg(gspca_dev, 0x0001, 0x0102);
-	switch (gspca_dev->width) {
+	switch (gspca_dev->pixfmt.width) {
 	case 176: /* 176x144 */
 		cit_write_reg(gspca_dev, 0x002c, 0x0103);	/* All except 320x240 */
 		cit_write_reg(gspca_dev, 0x0000, 0x0104);	/* Same */
@@ -1792,7 +1792,7 @@ static int cit_start_model2(struct gspca_dev *gspca_dev)
 
 	cit_write_reg(gspca_dev, 0x0000, 0x0100);	/* LED on */
 
-	switch (gspca_dev->width) {
+	switch (gspca_dev->pixfmt.width) {
 	case 176: /* 176x144 */
 		cit_write_reg(gspca_dev, 0x0050, 0x0111);
 		cit_write_reg(gspca_dev, 0x00d0, 0x0111);
@@ -1840,7 +1840,7 @@ static int cit_start_model2(struct gspca_dev *gspca_dev)
 	 * Magic control of CMOS sensor. Only lower values like
 	 * 0-3 work, and picture shifts left or right. Don't change.
 	 */
-	switch (gspca_dev->width) {
+	switch (gspca_dev->pixfmt.width) {
 	case 176: /* 176x144 */
 		cit_model2_Packet1(gspca_dev, 0x0014, 0x0002);
 		cit_model2_Packet1(gspca_dev, 0x0016, 0x0002); /* Horizontal shift */
@@ -1899,7 +1899,7 @@ static int cit_start_model2(struct gspca_dev *gspca_dev)
 	 * does not allow arbitrary values and apparently is a bit mask, to
 	 * be activated only at appropriate time. Don't change it randomly!
 	 */
-	switch (gspca_dev->width) {
+	switch (gspca_dev->pixfmt.width) {
 	case 176: /* 176x144 */
 		cit_model2_Packet1(gspca_dev, 0x0026, 0x00c2);
 		break;
@@ -2023,7 +2023,7 @@ static int cit_start_model3(struct gspca_dev *gspca_dev)
 	cit_model3_Packet1(gspca_dev, 0x009e, 0x0096);
 	cit_model3_Packet1(gspca_dev, 0x009f, 0x000a);
 
-	switch (gspca_dev->width) {
+	switch (gspca_dev->pixfmt.width) {
 	case 160:
 		cit_write_reg(gspca_dev, 0x0000, 0x0101); /* Same on 160x120, 320x240 */
 		cit_write_reg(gspca_dev, 0x00a0, 0x0103); /* Same on 160x120, 320x240 */
@@ -2134,7 +2134,7 @@ static int cit_start_model3(struct gspca_dev *gspca_dev)
 	   like with the IBM netcam pro). */
 	cit_write_reg(gspca_dev, clock_div, 0x0111); /* Clock Divider */
 
-	switch (gspca_dev->width) {
+	switch (gspca_dev->pixfmt.width) {
 	case 160:
 		cit_model3_Packet1(gspca_dev, 0x001f, 0x0000); /* Same */
 		cit_model3_Packet1(gspca_dev, 0x0039, 0x001f); /* Same */
@@ -2211,7 +2211,7 @@ static int cit_start_model4(struct gspca_dev *gspca_dev)
 	cit_write_reg(gspca_dev, 0xfffa, 0x0124);
 	cit_model4_Packet1(gspca_dev, 0x0034, 0x0000);
 
-	switch (gspca_dev->width) {
+	switch (gspca_dev->pixfmt.width) {
 	case 128: /* 128x96 */
 		cit_write_reg(gspca_dev, 0x0070, 0x0119);
 		cit_write_reg(gspca_dev, 0x00d0, 0x0111);
@@ -2531,7 +2531,7 @@ static int cit_start_ibm_netcam_pro(struct gspca_dev *gspca_dev)
 	cit_write_reg(gspca_dev, 0x00fc, 0x012b); /* Same */
 	cit_write_reg(gspca_dev, 0x0022, 0x012a); /* Same */
 
-	switch (gspca_dev->width) {
+	switch (gspca_dev->pixfmt.width) {
 	case 160: /* 160x120 */
 		cit_write_reg(gspca_dev, 0x0024, 0x010b);
 		cit_write_reg(gspca_dev, 0x0089, 0x0119);
@@ -2635,7 +2635,7 @@ static int sd_isoc_init(struct gspca_dev *gspca_dev)
 	struct usb_host_interface *alt;
 	int max_packet_size;
 
-	switch (gspca_dev->width) {
+	switch (gspca_dev->pixfmt.width) {
 	case 160:
 		max_packet_size = 450;
 		break;
@@ -2659,7 +2659,7 @@ static int sd_isoc_nego(struct gspca_dev *gspca_dev)
 	int ret, packet_size, min_packet_size;
 	struct usb_host_interface *alt;
 
-	switch (gspca_dev->width) {
+	switch (gspca_dev->pixfmt.width) {
 	case 160:
 		min_packet_size = 200;
 		break;
@@ -2780,7 +2780,7 @@ static u8 *cit_find_sof(struct gspca_dev *gspca_dev, u8 *data, int len)
 	case CIT_MODEL1:
 	case CIT_MODEL3:
 	case CIT_IBM_NETCAM_PRO:
-		switch (gspca_dev->width) {
+		switch (gspca_dev->pixfmt.width) {
 		case 160: /* 160x120 */
 			byte3 = 0x02;
 			byte4 = 0x0a;
@@ -2864,20 +2864,16 @@ static u8 *cit_find_sof(struct gspca_dev *gspca_dev, u8 *data, int len)
 				if (data[i] == 0xff) {
 					if (i >= 4)
 						PDEBUG(D_FRAM,
-						       "header found at offset: %d: %02x %02x 00 %02x %02x %02x\n",
+						       "header found at offset: %d: %02x %02x 00 %3ph\n",
 						       i - 1,
 						       data[i - 4],
 						       data[i - 3],
-						       data[i],
-						       data[i + 1],
-						       data[i + 2]);
+						       &data[i]);
 					else
 						PDEBUG(D_FRAM,
-						       "header found at offset: %d: 00 %02x %02x %02x\n",
+						       "header found at offset: %d: 00 %3ph\n",
 						       i - 1,
-						       data[i],
-						       data[i + 1],
-						       data[i + 2]);
+						       &data[i]);
 					return data + i + (sd->sof_len - 1);
 				}
 				break;
diff --git a/drivers/media/usb/gspca/zc3xx.c b/drivers/media/usb/gspca/zc3xx.c
index cbfc2f921427..7b95d8e88a20 100644
--- a/drivers/media/usb/gspca/zc3xx.c
+++ b/drivers/media/usb/gspca/zc3xx.c
@@ -6700,7 +6700,8 @@ static int sd_start(struct gspca_dev *gspca_dev)
 	};
 
 	/* create the JPEG header */
-	jpeg_define(sd->jpeg_hdr, gspca_dev->height, gspca_dev->width,
+	jpeg_define(sd->jpeg_hdr, gspca_dev->pixfmt.height,
+			gspca_dev->pixfmt.width,
 			0x21);		/* JPEG 422 */
 
 	mode = gspca_dev->cam.cam_mode[gspca_dev->curr_mode].priv;
diff --git a/drivers/media/usb/hdpvr/hdpvr-core.c b/drivers/media/usb/hdpvr/hdpvr-core.c
index 6e5070774dc2..2f0c89cbac76 100644
--- a/drivers/media/usb/hdpvr/hdpvr-core.c
+++ b/drivers/media/usb/hdpvr/hdpvr-core.c
@@ -78,7 +78,8 @@ void hdpvr_delete(struct hdpvr_device *dev)
 
 static void challenge(u8 *bytes)
 {
-	u64 *i64P, tmp64;
+	__le64 *i64P;
+	u64 tmp64;
 	uint i, idx;
 
 	for (idx = 0; idx < 32; ++idx) {
@@ -106,10 +107,10 @@ static void challenge(u8 *bytes)
 			for (i = 0; i < 3; i++)
 				bytes[1] *= bytes[6] + 1;
 			for (i = 0; i < 3; i++) {
-				i64P = (u64 *)bytes;
+				i64P = (__le64 *)bytes;
 				tmp64 = le64_to_cpup(i64P);
-				tmp64 <<= bytes[7] & 0x0f;
-				*i64P += cpu_to_le64(tmp64);
+				tmp64 = tmp64 + (tmp64 << (bytes[7] & 0x0f));
+				*i64P = cpu_to_le64(tmp64);
 			}
 			break;
 		}
@@ -301,8 +302,6 @@ static int hdpvr_probe(struct usb_interface *interface,
 		goto error;
 	}
 
-	dev->workqueue = 0;
-
 	/* init video transfer queues first of all */
 	/* to prevent oops in hdpvr_delete() on error paths */
 	INIT_LIST_HEAD(&dev->free_buff_list);
diff --git a/drivers/media/usb/pvrusb2/pvrusb2-hdw.c b/drivers/media/usb/pvrusb2/pvrusb2-hdw.c
index c4d51d78f837..ea05f678b559 100644
--- a/drivers/media/usb/pvrusb2/pvrusb2-hdw.c
+++ b/drivers/media/usb/pvrusb2/pvrusb2-hdw.c
@@ -2868,7 +2868,7 @@ static void pvr2_subdev_set_control(struct pvr2_hdw *hdw, int id,
 		pvr2_subdev_set_control(hdw, id, #lab, (hdw)->lab##_val); \
 	}
 
-v4l2_std_id pvr2_hdw_get_detected_std(struct pvr2_hdw *hdw)
+static v4l2_std_id pvr2_hdw_get_detected_std(struct pvr2_hdw *hdw)
 {
 	v4l2_std_id std;
 	std = (v4l2_std_id)hdw->std_mask_avail;
diff --git a/drivers/media/usb/siano/smsusb.c b/drivers/media/usb/siano/smsusb.c
index 03761c6f472f..05bd91a60c09 100644
--- a/drivers/media/usb/siano/smsusb.c
+++ b/drivers/media/usb/siano/smsusb.c
@@ -209,8 +209,10 @@ static int smsusb_sendrequest(void *context, void *buffer, size_t size)
 	struct sms_msg_hdr *phdr = (struct sms_msg_hdr *) buffer;
 	int dummy;
 
-	if (dev->state != SMSUSB_ACTIVE)
+	if (dev->state != SMSUSB_ACTIVE) {
+		sms_debug("Device not active yet");
 		return -ENOENT;
+	}
 
 	sms_debug("sending %s(%d) size: %d",
 		  smscore_translate_msg(phdr->msg_type), phdr->msg_type,
@@ -243,6 +245,9 @@ static int smsusb1_load_firmware(struct usb_device *udev, int id, int board_id)
 	int rc, dummy;
 	char *fw_filename;
 
+	if (id < 0)
+		id = sms_get_board(board_id)->default_mode;
+
 	if (id < DEVICE_MODE_DVBT || id > DEVICE_MODE_DVBT_BDA) {
 		sms_err("invalid firmware id specified %d", id);
 		return -EINVAL;
@@ -445,14 +450,15 @@ static int smsusb_probe(struct usb_interface *intf,
 	char devpath[32];
 	int i, rc;
 
-	sms_info("interface number %d",
+	sms_info("board id=%lu, interface number %d",
+		 id->driver_info,
 		 intf->cur_altsetting->desc.bInterfaceNumber);
 
 	if (sms_get_board(id->driver_info)->intf_num !=
 	    intf->cur_altsetting->desc.bInterfaceNumber) {
-		sms_err("interface number is %d expecting %d",
-			sms_get_board(id->driver_info)->intf_num,
-			intf->cur_altsetting->desc.bInterfaceNumber);
+		sms_debug("interface %d won't be used. Expecting interface %d to popup",
+			intf->cur_altsetting->desc.bInterfaceNumber,
+			sms_get_board(id->driver_info)->intf_num);
 		return -ENODEV;
 	}
 
@@ -483,22 +489,32 @@ static int smsusb_probe(struct usb_interface *intf,
 	}
 	if ((udev->actconfig->desc.bNumInterfaces == 2) &&
 	    (intf->cur_altsetting->desc.bInterfaceNumber == 0)) {
-		sms_err("rom interface 0 is not used");
+		sms_debug("rom interface 0 is not used");
 		return -ENODEV;
 	}
 
 	if (id->driver_info == SMS1XXX_BOARD_SIANO_STELLAR_ROM) {
-		sms_info("stellar device was found.");
+		/* Detected a Siano Stellar uninitialized */
+
 		snprintf(devpath, sizeof(devpath), "usb\\%d-%s",
 			 udev->bus->busnum, udev->devpath);
-		sms_info("stellar device was found.");
-		return smsusb1_load_firmware(
+		sms_info("stellar device in cold state was found at %s.", devpath);
+		rc = smsusb1_load_firmware(
 				udev, smscore_registry_getmode(devpath),
 				id->driver_info);
+
+		/* This device will reset and gain another USB ID */
+		if (!rc)
+			sms_info("stellar device now in warm state");
+		else
+			sms_err("Failed to put stellar in warm state. Error: %d", rc);
+
+		return rc;
+	} else {
+		rc = smsusb_init_device(intf, id->driver_info);
 	}
 
-	rc = smsusb_init_device(intf, id->driver_info);
-	sms_info("rc %d", rc);
+	sms_info("Device initialized with return code %d", rc);
 	sms_board_load_modules(id->driver_info);
 	return rc;
 }
@@ -550,10 +566,13 @@ static int smsusb_resume(struct usb_interface *intf)
 }
 
 static const struct usb_device_id smsusb_id_table[] = {
+	/* This device is only present before firmware load */
 	{ USB_DEVICE(0x187f, 0x0010),
-		.driver_info = SMS1XXX_BOARD_SIANO_STELLAR },
+		.driver_info = SMS1XXX_BOARD_SIANO_STELLAR_ROM },
+	/* This device pops up after firmware load */
 	{ USB_DEVICE(0x187f, 0x0100),
 		.driver_info = SMS1XXX_BOARD_SIANO_STELLAR },
+
 	{ USB_DEVICE(0x187f, 0x0200),
 		.driver_info = SMS1XXX_BOARD_SIANO_NOVA_A },
 	{ USB_DEVICE(0x187f, 0x0201),
diff --git a/drivers/media/usb/tlg2300/pd-main.c b/drivers/media/usb/tlg2300/pd-main.c
index 95f94e5aa66d..3316caa4733b 100644
--- a/drivers/media/usb/tlg2300/pd-main.c
+++ b/drivers/media/usb/tlg2300/pd-main.c
@@ -232,7 +232,7 @@ static int firmware_download(struct usb_device *udev)
 		goto out;
 	}
 
-	max_packet_size = udev->ep_out[0x1]->desc.wMaxPacketSize;
+	max_packet_size = le16_to_cpu(udev->ep_out[0x1]->desc.wMaxPacketSize);
 	log("\t\t download size : %d", (int)max_packet_size);
 
 	for (offset = 0; offset < fwlength; offset += max_packet_size) {
diff --git a/drivers/media/usb/ttusb-dec/ttusb_dec.c b/drivers/media/usb/ttusb-dec/ttusb_dec.c
index e52c3b97f304..29724af9b9ab 100644
--- a/drivers/media/usb/ttusb-dec/ttusb_dec.c
+++ b/drivers/media/usb/ttusb-dec/ttusb_dec.c
@@ -366,7 +366,7 @@ static int ttusb_dec_get_stb_state (struct ttusb_dec *dec, unsigned int *mode,
 		}
 		return 0;
 	} else {
-		return -1;
+		return -ENOENT;
 	}
 }
 
@@ -1241,6 +1241,8 @@ static void ttusb_dec_init_v_pes(struct ttusb_dec *dec)
 
 static int ttusb_dec_init_usb(struct ttusb_dec *dec)
 {
+	int result;
+
 	dprintk("%s\n", __func__);
 
 	mutex_init(&dec->usb_mutex);
@@ -1258,7 +1260,7 @@ static int ttusb_dec_init_usb(struct ttusb_dec *dec)
 			return -ENOMEM;
 		}
 		dec->irq_buffer = usb_alloc_coherent(dec->udev,IRQ_PACKET_SIZE,
-					GFP_ATOMIC, &dec->irq_dma_handle);
+					GFP_KERNEL, &dec->irq_dma_handle);
 		if(!dec->irq_buffer) {
 			usb_free_urb(dec->irq_urb);
 			return -ENOMEM;
@@ -1270,7 +1272,13 @@ static int ttusb_dec_init_usb(struct ttusb_dec *dec)
 		dec->irq_urb->transfer_flags |= URB_NO_TRANSFER_DMA_MAP;
 	}
 
-	return ttusb_dec_alloc_iso_urbs(dec);
+	result = ttusb_dec_alloc_iso_urbs(dec);
+	if (result) {
+		usb_free_urb(dec->irq_urb);
+		usb_free_coherent(dec->udev, IRQ_PACKET_SIZE,
+				  dec->irq_buffer, dec->irq_dma_handle);
+	}
+	return result;
 }
 
 static int ttusb_dec_boot_dsp(struct ttusb_dec *dec)
@@ -1293,10 +1301,11 @@ static int ttusb_dec_boot_dsp(struct ttusb_dec *dec)
 
 	dprintk("%s\n", __func__);
 
-	if (request_firmware(&fw_entry, dec->firmware_name, &dec->udev->dev)) {
+	result = request_firmware(&fw_entry, dec->firmware_name, &dec->udev->dev);
+	if (result) {
 		printk(KERN_ERR "%s: Firmware (%s) unavailable.\n",
 		       __func__, dec->firmware_name);
-		return 1;
+		return result;
 	}
 
 	firmware = fw_entry->data;
@@ -1306,7 +1315,7 @@ static int ttusb_dec_boot_dsp(struct ttusb_dec *dec)
 		printk("%s: firmware size too small for DSP code (%zu < 60).\n",
 			__func__, firmware_size);
 		release_firmware(fw_entry);
-		return -1;
+		return -ENOENT;
 	}
 
 	/* a 32 bit checksum over the first 56 bytes of the DSP Code is stored
@@ -1320,7 +1329,7 @@ static int ttusb_dec_boot_dsp(struct ttusb_dec *dec)
 		       "0x%08x != 0x%08x in file), file invalid.\n",
 			__func__, crc32_csum, crc32_check);
 		release_firmware(fw_entry);
-		return -1;
+		return -ENOENT;
 	}
 	memcpy(idstring, &firmware[36], 20);
 	idstring[20] = '\0';
@@ -1389,55 +1398,48 @@ static int ttusb_dec_init_stb(struct ttusb_dec *dec)
 	dprintk("%s\n", __func__);
 
 	result = ttusb_dec_get_stb_state(dec, &mode, &model, &version);
+	if (result)
+		return result;
 
-	if (!result) {
-		if (!mode) {
-			if (version == 0xABCDEFAB)
-				printk(KERN_INFO "ttusb_dec: no version "
-				       "info in Firmware\n");
-			else
-				printk(KERN_INFO "ttusb_dec: Firmware "
-				       "%x.%02x%c%c\n",
-				       version >> 24, (version >> 16) & 0xff,
-				       (version >> 8) & 0xff, version & 0xff);
-
-			result = ttusb_dec_boot_dsp(dec);
-			if (result)
-				return result;
-			else
-				return 1;
-		} else {
-			/* We can't trust the USB IDs that some firmwares
-			   give the box */
-			switch (model) {
-			case 0x00070001:
-			case 0x00070008:
-			case 0x0007000c:
-				ttusb_dec_set_model(dec, TTUSB_DEC3000S);
-				break;
-			case 0x00070009:
-			case 0x00070013:
-				ttusb_dec_set_model(dec, TTUSB_DEC2000T);
-				break;
-			case 0x00070011:
-				ttusb_dec_set_model(dec, TTUSB_DEC2540T);
-				break;
-			default:
-				printk(KERN_ERR "%s: unknown model returned "
-				       "by firmware (%08x) - please report\n",
-				       __func__, model);
-				return -1;
-				break;
-			}
+	if (!mode) {
+		if (version == 0xABCDEFAB)
+			printk(KERN_INFO "ttusb_dec: no version "
+			       "info in Firmware\n");
+		else
+			printk(KERN_INFO "ttusb_dec: Firmware "
+			       "%x.%02x%c%c\n",
+			       version >> 24, (version >> 16) & 0xff,
+			       (version >> 8) & 0xff, version & 0xff);
 
+		result = ttusb_dec_boot_dsp(dec);
+		if (result)
+			return result;
+	} else {
+		/* We can't trust the USB IDs that some firmwares
+		   give the box */
+		switch (model) {
+		case 0x00070001:
+		case 0x00070008:
+		case 0x0007000c:
+			ttusb_dec_set_model(dec, TTUSB_DEC3000S);
+			break;
+		case 0x00070009:
+		case 0x00070013:
+			ttusb_dec_set_model(dec, TTUSB_DEC2000T);
+			break;
+		case 0x00070011:
+			ttusb_dec_set_model(dec, TTUSB_DEC2540T);
+			break;
+		default:
+			printk(KERN_ERR "%s: unknown model returned "
+			       "by firmware (%08x) - please report\n",
+			       __func__, model);
+			return -ENOENT;
+		}
 			if (version >= 0x01770000)
 				dec->can_playback = 1;
-
-			return 0;
-		}
 	}
-	else
-		return result;
+	return 0;
 }
 
 static int ttusb_dec_init_dvb(struct ttusb_dec *dec)
@@ -1539,19 +1541,7 @@ static void ttusb_dec_exit_dvb(struct ttusb_dec *dec)
 
 static void ttusb_dec_exit_rc(struct ttusb_dec *dec)
 {
-
 	dprintk("%s\n", __func__);
-	/* we have to check whether the irq URB is already submitted.
-	  * As the irq is submitted after the interface is changed,
-	  * this is the best method i figured out.
-	  * Any others?*/
-	if (dec->interface == TTUSB_DEC_INTERFACE_IN)
-		usb_kill_urb(dec->irq_urb);
-
-	usb_free_urb(dec->irq_urb);
-
-	usb_free_coherent(dec->udev,IRQ_PACKET_SIZE,
-			  dec->irq_buffer, dec->irq_dma_handle);
 
 	if (dec->rc_input_dev) {
 		input_unregister_device(dec->rc_input_dev);
@@ -1566,6 +1556,20 @@ static void ttusb_dec_exit_usb(struct ttusb_dec *dec)
 
 	dprintk("%s\n", __func__);
 
+	if (enable_rc) {
+		/* we have to check whether the irq URB is already submitted.
+		 * As the irq is submitted after the interface is changed,
+		 * this is the best method i figured out.
+		 * Any others?*/
+		if (dec->interface == TTUSB_DEC_INTERFACE_IN)
+			usb_kill_urb(dec->irq_urb);
+
+		usb_free_urb(dec->irq_urb);
+
+		usb_free_coherent(dec->udev, IRQ_PACKET_SIZE,
+				  dec->irq_buffer, dec->irq_dma_handle);
+	}
+
 	dec->iso_stream_count = 0;
 
 	for (i = 0; i < ISO_BUF_COUNT; i++)
@@ -1623,6 +1627,7 @@ static int ttusb_dec_probe(struct usb_interface *intf,
 {
 	struct usb_device *udev;
 	struct ttusb_dec *dec;
+	int result;
 
 	dprintk("%s\n", __func__);
 
@@ -1651,13 +1656,15 @@ static int ttusb_dec_probe(struct usb_interface *intf,
 
 	dec->udev = udev;
 
-	if (ttusb_dec_init_usb(dec))
-		return 0;
-	if (ttusb_dec_init_stb(dec)) {
-		ttusb_dec_exit_usb(dec);
-		return 0;
-	}
-	ttusb_dec_init_dvb(dec);
+	result = ttusb_dec_init_usb(dec);
+	if (result)
+		goto err_usb;
+	result = ttusb_dec_init_stb(dec);
+	if (result)
+		goto err_stb;
+	result = ttusb_dec_init_dvb(dec);
+	if (result)
+		goto err_stb;
 
 	dec->adapter.priv = dec;
 	switch (id->idProduct) {
@@ -1696,6 +1703,11 @@ static int ttusb_dec_probe(struct usb_interface *intf,
 		ttusb_init_rc(dec);
 
 	return 0;
+err_stb:
+	ttusb_dec_exit_usb(dec);
+err_usb:
+	kfree(dec);
+	return result;
 }
 
 static void ttusb_dec_disconnect(struct usb_interface *intf)
diff --git a/drivers/media/usb/uvc/uvc_ctrl.c b/drivers/media/usb/uvc/uvc_ctrl.c
index a2f4501c23ca..0eb82106d2ff 100644
--- a/drivers/media/usb/uvc/uvc_ctrl.c
+++ b/drivers/media/usb/uvc/uvc_ctrl.c
@@ -664,7 +664,7 @@ static struct uvc_control_mapping uvc_ctrl_mappings[] = {
 		.size		= 32,
 		.offset		= 0,
 		.v4l2_type	= V4L2_CTRL_TYPE_INTEGER,
-		.data_type	= UVC_CTRL_DATA_TYPE_UNSIGNED,
+		.data_type	= UVC_CTRL_DATA_TYPE_SIGNED,
 	},
 	{
 		.id		= V4L2_CID_TILT_ABSOLUTE,
@@ -674,7 +674,7 @@ static struct uvc_control_mapping uvc_ctrl_mappings[] = {
 		.size		= 32,
 		.offset		= 32,
 		.v4l2_type	= V4L2_CTRL_TYPE_INTEGER,
-		.data_type	= UVC_CTRL_DATA_TYPE_UNSIGNED,
+		.data_type	= UVC_CTRL_DATA_TYPE_SIGNED,
 	},
 	{
 		.id		= V4L2_CID_PRIVACY,
diff --git a/drivers/media/usb/uvc/uvc_video.c b/drivers/media/usb/uvc/uvc_video.c
index 3394c3432011..899cb6d1c4a4 100644
--- a/drivers/media/usb/uvc/uvc_video.c
+++ b/drivers/media/usb/uvc/uvc_video.c
@@ -680,7 +680,8 @@ void uvc_video_clock_update(struct uvc_streaming *stream,
 		  stream->dev->name,
 		  sof >> 16, div_u64(((u64)sof & 0xffff) * 1000000LLU, 65536),
 		  y, ts.tv_sec, ts.tv_nsec / NSEC_PER_USEC,
-		  v4l2_buf->timestamp.tv_sec, v4l2_buf->timestamp.tv_usec,
+		  v4l2_buf->timestamp.tv_sec,
+		  (unsigned long)v4l2_buf->timestamp.tv_usec,
 		  x1, first->host_sof, first->dev_sof,
 		  x2, last->host_sof, last->dev_sof, y1, y2);
 
diff --git a/drivers/media/v4l2-core/tuner-core.c b/drivers/media/v4l2-core/tuner-core.c
index ddc9379eb276..20c09229a08e 100644
--- a/drivers/media/v4l2-core/tuner-core.c
+++ b/drivers/media/v4l2-core/tuner-core.c
@@ -43,7 +43,7 @@
 
 #define UNSET (-1U)
 
-#define PREFIX (t->i2c->driver->driver.name)
+#define PREFIX (t->i2c->dev.driver->name)
 
 /*
  * Driver modprobe parameters
@@ -247,7 +247,7 @@ static const struct analog_demod_ops tuner_analog_ops = {
 /**
  * set_type - Sets the tuner type for a given device
  *
- * @c:			i2c_client descriptoy
+ * @c:			i2c_client descriptor
  * @type:		type of the tuner (e. g. tuner number)
  * @new_mode_mask:	Indicates if tuner supports TV and/or Radio
  * @new_config:		an optional parameter used by a few tuners to adjust
@@ -452,7 +452,7 @@ static void set_type(struct i2c_client *c, unsigned int type,
 	}
 
 	tuner_dbg("%s %s I2C addr 0x%02x with type %d used for 0x%02x\n",
-		  c->adapter->name, c->driver->driver.name, c->addr << 1, type,
+		  c->adapter->name, c->dev.driver->name, c->addr << 1, type,
 		  t->mode_mask);
 	return;
 
@@ -556,7 +556,7 @@ static void tuner_lookup(struct i2c_adapter *adap,
 		int mode_mask;
 
 		if (pos->i2c->adapter != adap ||
-		    strcmp(pos->i2c->driver->driver.name, "tuner"))
+		    strcmp(pos->i2c->dev.driver->name, "tuner"))
 			continue;
 
 		mode_mask = pos->mode_mask;
diff --git a/drivers/media/v4l2-core/v4l2-async.c b/drivers/media/v4l2-core/v4l2-async.c
index c85d69da35bd..85a6a34128a8 100644
--- a/drivers/media/v4l2-core/v4l2-async.c
+++ b/drivers/media/v4l2-core/v4l2-async.c
@@ -189,30 +189,53 @@ void v4l2_async_notifier_unregister(struct v4l2_async_notifier *notifier)
 	struct v4l2_subdev *sd, *tmp;
 	unsigned int notif_n_subdev = notifier->num_subdevs;
 	unsigned int n_subdev = min(notif_n_subdev, V4L2_MAX_SUBDEVS);
-	struct device *dev[n_subdev];
+	struct device **dev;
 	int i = 0;
 
 	if (!notifier->v4l2_dev)
 		return;
 
+	dev = kmalloc(n_subdev * sizeof(*dev), GFP_KERNEL);
+	if (!dev) {
+		dev_err(notifier->v4l2_dev->dev,
+			"Failed to allocate device cache!\n");
+	}
+
 	mutex_lock(&list_lock);
 
 	list_del(&notifier->list);
 
 	list_for_each_entry_safe(sd, tmp, &notifier->done, async_list) {
-		dev[i] = get_device(sd->dev);
+		struct device *d;
+
+		d = get_device(sd->dev);
 
 		v4l2_async_cleanup(sd);
 
 		/* If we handled USB devices, we'd have to lock the parent too */
-		device_release_driver(dev[i++]);
+		device_release_driver(d);
 
 		if (notifier->unbind)
 			notifier->unbind(notifier, sd, sd->asd);
+
+		/*
+		 * Store device at the device cache, in order to call
+		 * put_device() on the final step
+		 */
+		if (dev)
+			dev[i++] = d;
+		else
+			put_device(d);
 	}
 
 	mutex_unlock(&list_lock);
 
+	/*
+	 * Call device_attach() to reprobe devices
+	 *
+	 * NOTE: If dev allocation fails, i is 0, and the whole loop won't be
+	 * executed.
+	 */
 	while (i--) {
 		struct device *d = dev[i];
 
@@ -228,6 +251,7 @@ void v4l2_async_notifier_unregister(struct v4l2_async_notifier *notifier)
 		}
 		put_device(d);
 	}
+	kfree(dev);
 
 	notifier->v4l2_dev = NULL;
 
diff --git a/drivers/media/v4l2-core/v4l2-clk.c b/drivers/media/v4l2-core/v4l2-clk.c
index b67de8642b5a..e18cc0469cf8 100644
--- a/drivers/media/v4l2-core/v4l2-clk.c
+++ b/drivers/media/v4l2-core/v4l2-clk.c
@@ -240,3 +240,42 @@ void v4l2_clk_unregister(struct v4l2_clk *clk)
 	kfree(clk);
 }
 EXPORT_SYMBOL(v4l2_clk_unregister);
+
+struct v4l2_clk_fixed {
+	unsigned long rate;
+	struct v4l2_clk_ops ops;
+};
+
+static unsigned long fixed_get_rate(struct v4l2_clk *clk)
+{
+	struct v4l2_clk_fixed *priv = clk->priv;
+	return priv->rate;
+}
+
+struct v4l2_clk *__v4l2_clk_register_fixed(const char *dev_id,
+		const char *id, unsigned long rate, struct module *owner)
+{
+	struct v4l2_clk *clk;
+	struct v4l2_clk_fixed *priv = kzalloc(sizeof(*priv), GFP_KERNEL);
+
+	if (!priv)
+		return ERR_PTR(-ENOMEM);
+
+	priv->rate = rate;
+	priv->ops.get_rate = fixed_get_rate;
+	priv->ops.owner = owner;
+
+	clk = v4l2_clk_register(&priv->ops, dev_id, id, priv);
+	if (IS_ERR(clk))
+		kfree(priv);
+
+	return clk;
+}
+EXPORT_SYMBOL(__v4l2_clk_register_fixed);
+
+void v4l2_clk_unregister_fixed(struct v4l2_clk *clk)
+{
+	kfree(clk->priv);
+	v4l2_clk_unregister(clk);
+}
+EXPORT_SYMBOL(v4l2_clk_unregister_fixed);
diff --git a/drivers/media/v4l2-core/v4l2-common.c b/drivers/media/v4l2-core/v4l2-common.c
index 037d7a55aa8c..433d6d77942e 100644
--- a/drivers/media/v4l2-core/v4l2-common.c
+++ b/drivers/media/v4l2-core/v4l2-common.c
@@ -236,14 +236,14 @@ void v4l2_i2c_subdev_init(struct v4l2_subdev *sd, struct i2c_client *client,
 	v4l2_subdev_init(sd, ops);
 	sd->flags |= V4L2_SUBDEV_FL_IS_I2C;
 	/* the owner is the same as the i2c_client's driver owner */
-	sd->owner = client->driver->driver.owner;
+	sd->owner = client->dev.driver->owner;
 	sd->dev = &client->dev;
 	/* i2c_client and v4l2_subdev point to one another */
 	v4l2_set_subdevdata(sd, client);
 	i2c_set_clientdata(client, sd);
 	/* initialize name */
 	snprintf(sd->name, sizeof(sd->name), "%s %d-%04x",
-		client->driver->driver.name, i2c_adapter_id(client->adapter),
+		client->dev.driver->name, i2c_adapter_id(client->adapter),
 		client->addr);
 }
 EXPORT_SYMBOL_GPL(v4l2_i2c_subdev_init);
@@ -274,11 +274,11 @@ struct v4l2_subdev *v4l2_i2c_new_subdev_board(struct v4l2_device *v4l2_dev,
 	   loaded. This delay-load mechanism doesn't work if other drivers
 	   want to use the i2c device, so explicitly loading the module
 	   is the best alternative. */
-	if (client == NULL || client->driver == NULL)
+	if (client == NULL || client->dev.driver == NULL)
 		goto error;
 
 	/* Lock the module so we can safely get the v4l2_subdev pointer */
-	if (!try_module_get(client->driver->driver.owner))
+	if (!try_module_get(client->dev.driver->owner))
 		goto error;
 	sd = i2c_get_clientdata(client);
 
@@ -287,7 +287,7 @@ struct v4l2_subdev *v4l2_i2c_new_subdev_board(struct v4l2_device *v4l2_dev,
 	if (v4l2_device_register_subdev(v4l2_dev, sd))
 		sd = NULL;
 	/* Decrease the module use count to match the first try_module_get. */
-	module_put(client->driver->driver.owner);
+	module_put(client->dev.driver->owner);
 
 error:
 	/* If we have a client but no subdev, then something went wrong and
diff --git a/drivers/media/v4l2-core/v4l2-ctrls.c b/drivers/media/v4l2-core/v4l2-ctrls.c
index c3f080388684..60dcc0f3b32e 100644
--- a/drivers/media/v4l2-core/v4l2-ctrls.c
+++ b/drivers/media/v4l2-core/v4l2-ctrls.c
@@ -565,13 +565,13 @@ EXPORT_SYMBOL(v4l2_ctrl_get_menu);
  * Returns NULL or an s64 type array containing the menu for given
  * control ID. The total number of the menu items is returned in @len.
  */
-const s64 const *v4l2_ctrl_get_int_menu(u32 id, u32 *len)
+const s64 *v4l2_ctrl_get_int_menu(u32 id, u32 *len)
 {
-	static const s64 const qmenu_int_vpx_num_partitions[] = {
+	static const s64 qmenu_int_vpx_num_partitions[] = {
 		1, 2, 4, 8,
 	};
 
-	static const s64 const qmenu_int_vpx_num_ref_frames[] = {
+	static const s64 qmenu_int_vpx_num_ref_frames[] = {
 		1, 2, 3,
 	};
 
@@ -583,7 +583,7 @@ const s64 const *v4l2_ctrl_get_int_menu(u32 id, u32 *len)
 	default:
 		*len = 0;
 		return NULL;
-	};
+	}
 }
 EXPORT_SYMBOL(v4l2_ctrl_get_int_menu);
 
diff --git a/drivers/media/v4l2-core/v4l2-mem2mem.c b/drivers/media/v4l2-core/v4l2-mem2mem.c
index 7c4371288215..73035ee0f4de 100644
--- a/drivers/media/v4l2-core/v4l2-mem2mem.c
+++ b/drivers/media/v4l2-core/v4l2-mem2mem.c
@@ -41,6 +41,8 @@ module_param(debug, bool, 0644);
 #define TRANS_QUEUED		(1 << 0)
 /* Instance is currently running in hardware */
 #define TRANS_RUNNING		(1 << 1)
+/* Instance is currently aborting */
+#define TRANS_ABORT		(1 << 2)
 
 
 /* Offset base for buffers on the destination queue - used to distinguish
@@ -221,6 +223,14 @@ static void v4l2_m2m_try_schedule(struct v4l2_m2m_ctx *m2m_ctx)
 	}
 
 	spin_lock_irqsave(&m2m_dev->job_spinlock, flags_job);
+
+	/* If the context is aborted then don't schedule it */
+	if (m2m_ctx->job_flags & TRANS_ABORT) {
+		spin_unlock_irqrestore(&m2m_dev->job_spinlock, flags_job);
+		dprintk("Aborted context\n");
+		return;
+	}
+
 	if (m2m_ctx->job_flags & TRANS_QUEUED) {
 		spin_unlock_irqrestore(&m2m_dev->job_spinlock, flags_job);
 		dprintk("On job queue already\n");
@@ -280,6 +290,8 @@ static void v4l2_m2m_cancel_job(struct v4l2_m2m_ctx *m2m_ctx)
 
 	m2m_dev = m2m_ctx->m2m_dev;
 	spin_lock_irqsave(&m2m_dev->job_spinlock, flags);
+
+	m2m_ctx->job_flags |= TRANS_ABORT;
 	if (m2m_ctx->job_flags & TRANS_RUNNING) {
 		spin_unlock_irqrestore(&m2m_dev->job_spinlock, flags);
 		m2m_dev->m2m_ops->job_abort(m2m_ctx->priv);
@@ -480,13 +492,15 @@ int v4l2_m2m_streamoff(struct file *file, struct v4l2_m2m_ctx *m2m_ctx,
 	m2m_dev = m2m_ctx->m2m_dev;
 	spin_lock_irqsave(&m2m_dev->job_spinlock, flags_job);
 	/* We should not be scheduled anymore, since we're dropping a queue. */
-	INIT_LIST_HEAD(&m2m_ctx->queue);
+	if (m2m_ctx->job_flags & TRANS_QUEUED)
+		list_del(&m2m_ctx->queue);
 	m2m_ctx->job_flags = 0;
 
 	spin_lock_irqsave(&q_ctx->rdy_spinlock, flags);
 	/* Drop queue, since streamoff returns device to the same state as after
 	 * calling reqbufs. */
 	INIT_LIST_HEAD(&q_ctx->rdy_queue);
+	q_ctx->num_rdy = 0;
 	spin_unlock_irqrestore(&q_ctx->rdy_spinlock, flags);
 
 	if (m2m_dev->curr_ctx == m2m_ctx) {
diff --git a/drivers/media/v4l2-core/videobuf2-core.c b/drivers/media/v4l2-core/videobuf2-core.c
index de0e87f0b2c3..b19b306c8f7f 100644
--- a/drivers/media/v4l2-core/videobuf2-core.c
+++ b/drivers/media/v4l2-core/videobuf2-core.c
@@ -241,7 +241,8 @@ static int __vb2_queue_alloc(struct vb2_queue *q, enum v4l2_memory memory,
 		q->bufs[q->num_buffers + buffer] = vb;
 	}
 
-	__setup_offsets(q, buffer);
+	if (memory == V4L2_MEMORY_MMAP)
+		__setup_offsets(q, buffer);
 
 	dprintk(1, "Allocated %d buffers, %d plane(s) each\n",
 			buffer, num_planes);
@@ -1015,6 +1016,10 @@ static int __qbuf_userptr(struct vb2_buffer *vb, const struct v4l2_buffer *b)
 
 		/* Check if the provided plane buffer is large enough */
 		if (planes[plane].length < q->plane_sizes[plane]) {
+			dprintk(1, "qbuf: provided buffer size %u is less than "
+						"setup size %u for plane %d\n",
+						planes[plane].length,
+						q->plane_sizes[plane], plane);
 			ret = -EINVAL;
 			goto err;
 		}
@@ -1205,8 +1210,11 @@ static int __buf_prepare(struct vb2_buffer *vb, const struct v4l2_buffer *b)
 	int ret;
 
 	ret = __verify_length(vb, b);
-	if (ret < 0)
+	if (ret < 0) {
+		dprintk(1, "%s(): plane parameters verification failed: %d\n",
+			__func__, ret);
 		return ret;
+	}
 
 	switch (q->memory) {
 	case V4L2_MEMORY_MMAP:
@@ -2469,10 +2477,11 @@ size_t vb2_read(struct vb2_queue *q, char __user *data, size_t count,
 }
 EXPORT_SYMBOL_GPL(vb2_read);
 
-size_t vb2_write(struct vb2_queue *q, char __user *data, size_t count,
+size_t vb2_write(struct vb2_queue *q, const char __user *data, size_t count,
 		loff_t *ppos, int nonblocking)
 {
-	return __vb2_perform_fileio(q, data, count, ppos, nonblocking, 0);
+	return __vb2_perform_fileio(q, (char __user *) data, count,
+							ppos, nonblocking, 0);
 }
 EXPORT_SYMBOL_GPL(vb2_write);
 
@@ -2633,7 +2642,7 @@ int vb2_fop_release(struct file *file)
 }
 EXPORT_SYMBOL_GPL(vb2_fop_release);
 
-ssize_t vb2_fop_write(struct file *file, char __user *buf,
+ssize_t vb2_fop_write(struct file *file, const char __user *buf,
 		size_t count, loff_t *ppos)
 {
 	struct video_device *vdev = video_devdata(file);
diff --git a/drivers/media/v4l2-core/videobuf2-dma-sg.c b/drivers/media/v4l2-core/videobuf2-dma-sg.c
index 16ae3dcc7e29..2f860543912c 100644
--- a/drivers/media/v4l2-core/videobuf2-dma-sg.c
+++ b/drivers/media/v4l2-core/videobuf2-dma-sg.c
@@ -35,17 +35,61 @@ struct vb2_dma_sg_buf {
 	struct page			**pages;
 	int				write;
 	int				offset;
-	struct vb2_dma_sg_desc		sg_desc;
+	struct sg_table			sg_table;
+	size_t				size;
+	unsigned int			num_pages;
 	atomic_t			refcount;
 	struct vb2_vmarea_handler	handler;
 };
 
 static void vb2_dma_sg_put(void *buf_priv);
 
+static int vb2_dma_sg_alloc_compacted(struct vb2_dma_sg_buf *buf,
+		gfp_t gfp_flags)
+{
+	unsigned int last_page = 0;
+	int size = buf->size;
+
+	while (size > 0) {
+		struct page *pages;
+		int order;
+		int i;
+
+		order = get_order(size);
+		/* Dont over allocate*/
+		if ((PAGE_SIZE << order) > size)
+			order--;
+
+		pages = NULL;
+		while (!pages) {
+			pages = alloc_pages(GFP_KERNEL | __GFP_ZERO |
+					__GFP_NOWARN | gfp_flags, order);
+			if (pages)
+				break;
+
+			if (order == 0) {
+				while (last_page--)
+					__free_page(buf->pages[last_page]);
+				return -ENOMEM;
+			}
+			order--;
+		}
+
+		split_page(pages, order);
+		for (i = 0; i < (1 << order); i++)
+			buf->pages[last_page++] = &pages[i];
+
+		size -= PAGE_SIZE << order;
+	}
+
+	return 0;
+}
+
 static void *vb2_dma_sg_alloc(void *alloc_ctx, unsigned long size, gfp_t gfp_flags)
 {
 	struct vb2_dma_sg_buf *buf;
-	int i;
+	int ret;
+	int num_pages;
 
 	buf = kzalloc(sizeof *buf, GFP_KERNEL);
 	if (!buf)
@@ -54,29 +98,23 @@ static void *vb2_dma_sg_alloc(void *alloc_ctx, unsigned long size, gfp_t gfp_fla
 	buf->vaddr = NULL;
 	buf->write = 0;
 	buf->offset = 0;
-	buf->sg_desc.size = size;
+	buf->size = size;
 	/* size is already page aligned */
-	buf->sg_desc.num_pages = size >> PAGE_SHIFT;
+	buf->num_pages = size >> PAGE_SHIFT;
 
-	buf->sg_desc.sglist = vzalloc(buf->sg_desc.num_pages *
-				      sizeof(*buf->sg_desc.sglist));
-	if (!buf->sg_desc.sglist)
-		goto fail_sglist_alloc;
-	sg_init_table(buf->sg_desc.sglist, buf->sg_desc.num_pages);
-
-	buf->pages = kzalloc(buf->sg_desc.num_pages * sizeof(struct page *),
+	buf->pages = kzalloc(buf->num_pages * sizeof(struct page *),
 			     GFP_KERNEL);
 	if (!buf->pages)
 		goto fail_pages_array_alloc;
 
-	for (i = 0; i < buf->sg_desc.num_pages; ++i) {
-		buf->pages[i] = alloc_page(GFP_KERNEL | __GFP_ZERO |
-					   __GFP_NOWARN | gfp_flags);
-		if (NULL == buf->pages[i])
-			goto fail_pages_alloc;
-		sg_set_page(&buf->sg_desc.sglist[i],
-			    buf->pages[i], PAGE_SIZE, 0);
-	}
+	ret = vb2_dma_sg_alloc_compacted(buf, gfp_flags);
+	if (ret)
+		goto fail_pages_alloc;
+
+	ret = sg_alloc_table_from_pages(&buf->sg_table, buf->pages,
+			buf->num_pages, 0, size, gfp_flags);
+	if (ret)
+		goto fail_table_alloc;
 
 	buf->handler.refcount = &buf->refcount;
 	buf->handler.put = vb2_dma_sg_put;
@@ -85,18 +123,16 @@ static void *vb2_dma_sg_alloc(void *alloc_ctx, unsigned long size, gfp_t gfp_fla
 	atomic_inc(&buf->refcount);
 
 	dprintk(1, "%s: Allocated buffer of %d pages\n",
-		__func__, buf->sg_desc.num_pages);
+		__func__, buf->num_pages);
 	return buf;
 
+fail_table_alloc:
+	num_pages = buf->num_pages;
+	while (num_pages--)
+		__free_page(buf->pages[num_pages]);
 fail_pages_alloc:
-	while (--i >= 0)
-		__free_page(buf->pages[i]);
 	kfree(buf->pages);
-
 fail_pages_array_alloc:
-	vfree(buf->sg_desc.sglist);
-
-fail_sglist_alloc:
 	kfree(buf);
 	return NULL;
 }
@@ -104,14 +140,14 @@ fail_sglist_alloc:
 static void vb2_dma_sg_put(void *buf_priv)
 {
 	struct vb2_dma_sg_buf *buf = buf_priv;
-	int i = buf->sg_desc.num_pages;
+	int i = buf->num_pages;
 
 	if (atomic_dec_and_test(&buf->refcount)) {
 		dprintk(1, "%s: Freeing buffer of %d pages\n", __func__,
-			buf->sg_desc.num_pages);
+			buf->num_pages);
 		if (buf->vaddr)
-			vm_unmap_ram(buf->vaddr, buf->sg_desc.num_pages);
-		vfree(buf->sg_desc.sglist);
+			vm_unmap_ram(buf->vaddr, buf->num_pages);
+		sg_free_table(&buf->sg_table);
 		while (--i >= 0)
 			__free_page(buf->pages[i]);
 		kfree(buf->pages);
@@ -124,7 +160,7 @@ static void *vb2_dma_sg_get_userptr(void *alloc_ctx, unsigned long vaddr,
 {
 	struct vb2_dma_sg_buf *buf;
 	unsigned long first, last;
-	int num_pages_from_user, i;
+	int num_pages_from_user;
 
 	buf = kzalloc(sizeof *buf, GFP_KERNEL);
 	if (!buf)
@@ -133,56 +169,41 @@ static void *vb2_dma_sg_get_userptr(void *alloc_ctx, unsigned long vaddr,
 	buf->vaddr = NULL;
 	buf->write = write;
 	buf->offset = vaddr & ~PAGE_MASK;
-	buf->sg_desc.size = size;
+	buf->size = size;
 
 	first = (vaddr           & PAGE_MASK) >> PAGE_SHIFT;
 	last  = ((vaddr + size - 1) & PAGE_MASK) >> PAGE_SHIFT;
-	buf->sg_desc.num_pages = last - first + 1;
-
-	buf->sg_desc.sglist = vzalloc(
-		buf->sg_desc.num_pages * sizeof(*buf->sg_desc.sglist));
-	if (!buf->sg_desc.sglist)
-		goto userptr_fail_sglist_alloc;
-
-	sg_init_table(buf->sg_desc.sglist, buf->sg_desc.num_pages);
+	buf->num_pages = last - first + 1;
 
-	buf->pages = kzalloc(buf->sg_desc.num_pages * sizeof(struct page *),
+	buf->pages = kzalloc(buf->num_pages * sizeof(struct page *),
 			     GFP_KERNEL);
 	if (!buf->pages)
-		goto userptr_fail_pages_array_alloc;
+		return NULL;
 
 	num_pages_from_user = get_user_pages(current, current->mm,
 					     vaddr & PAGE_MASK,
-					     buf->sg_desc.num_pages,
+					     buf->num_pages,
 					     write,
 					     1, /* force */
 					     buf->pages,
 					     NULL);
 
-	if (num_pages_from_user != buf->sg_desc.num_pages)
+	if (num_pages_from_user != buf->num_pages)
 		goto userptr_fail_get_user_pages;
 
-	sg_set_page(&buf->sg_desc.sglist[0], buf->pages[0],
-		    PAGE_SIZE - buf->offset, buf->offset);
-	size -= PAGE_SIZE - buf->offset;
-	for (i = 1; i < buf->sg_desc.num_pages; ++i) {
-		sg_set_page(&buf->sg_desc.sglist[i], buf->pages[i],
-			    min_t(size_t, PAGE_SIZE, size), 0);
-		size -= min_t(size_t, PAGE_SIZE, size);
-	}
+	if (sg_alloc_table_from_pages(&buf->sg_table, buf->pages,
+			buf->num_pages, buf->offset, size, 0))
+		goto userptr_fail_alloc_table_from_pages;
+
 	return buf;
 
+userptr_fail_alloc_table_from_pages:
 userptr_fail_get_user_pages:
 	dprintk(1, "get_user_pages requested/got: %d/%d]\n",
-	       num_pages_from_user, buf->sg_desc.num_pages);
+	       num_pages_from_user, buf->num_pages);
 	while (--num_pages_from_user >= 0)
 		put_page(buf->pages[num_pages_from_user]);
 	kfree(buf->pages);
-
-userptr_fail_pages_array_alloc:
-	vfree(buf->sg_desc.sglist);
-
-userptr_fail_sglist_alloc:
 	kfree(buf);
 	return NULL;
 }
@@ -194,18 +215,18 @@ userptr_fail_sglist_alloc:
 static void vb2_dma_sg_put_userptr(void *buf_priv)
 {
 	struct vb2_dma_sg_buf *buf = buf_priv;
-	int i = buf->sg_desc.num_pages;
+	int i = buf->num_pages;
 
 	dprintk(1, "%s: Releasing userspace buffer of %d pages\n",
-	       __func__, buf->sg_desc.num_pages);
+	       __func__, buf->num_pages);
 	if (buf->vaddr)
-		vm_unmap_ram(buf->vaddr, buf->sg_desc.num_pages);
+		vm_unmap_ram(buf->vaddr, buf->num_pages);
+	sg_free_table(&buf->sg_table);
 	while (--i >= 0) {
 		if (buf->write)
 			set_page_dirty_lock(buf->pages[i]);
 		put_page(buf->pages[i]);
 	}
-	vfree(buf->sg_desc.sglist);
 	kfree(buf->pages);
 	kfree(buf);
 }
@@ -218,7 +239,7 @@ static void *vb2_dma_sg_vaddr(void *buf_priv)
 
 	if (!buf->vaddr)
 		buf->vaddr = vm_map_ram(buf->pages,
-					buf->sg_desc.num_pages,
+					buf->num_pages,
 					-1,
 					PAGE_KERNEL);
 
@@ -274,7 +295,7 @@ static void *vb2_dma_sg_cookie(void *buf_priv)
 {
 	struct vb2_dma_sg_buf *buf = buf_priv;
 
-	return &buf->sg_desc;
+	return &buf->sg_table;
 }
 
 const struct vb2_mem_ops vb2_dma_sg_memops = {
diff --git a/drivers/memstick/core/memstick.c b/drivers/memstick/core/memstick.c
index bbf4aea1627d..a0547dbf9806 100644
--- a/drivers/memstick/core/memstick.c
+++ b/drivers/memstick/core/memstick.c
@@ -253,7 +253,7 @@ void memstick_new_req(struct memstick_host *host)
 {
 	if (host->card) {
 		host->retries = cmd_retries;
-		INIT_COMPLETION(host->card->mrq_complete);
+		reinit_completion(&host->card->mrq_complete);
 		host->request(host);
 	}
 }
diff --git a/drivers/memstick/core/ms_block.c b/drivers/memstick/core/ms_block.c
index 9188ef5d677e..24f2f8473dee 100644
--- a/drivers/memstick/core/ms_block.c
+++ b/drivers/memstick/core/ms_block.c
@@ -401,10 +401,10 @@ again:
 			sizeof(struct ms_status_register)))
 			return 0;
 
-		msb->state = MSB_RP_RECIVE_STATUS_REG;
+		msb->state = MSB_RP_RECEIVE_STATUS_REG;
 		return 0;
 
-	case MSB_RP_RECIVE_STATUS_REG:
+	case MSB_RP_RECEIVE_STATUS_REG:
 		msb->regs.status = *(struct ms_status_register *)mrq->data;
 		msb->state = MSB_RP_SEND_OOB_READ;
 		/* fallthrough */
diff --git a/drivers/memstick/core/ms_block.h b/drivers/memstick/core/ms_block.h
index 96e637550988..c75198dbf139 100644
--- a/drivers/memstick/core/ms_block.h
+++ b/drivers/memstick/core/ms_block.h
@@ -223,7 +223,7 @@ enum msb_readpage_states {
 	MSB_RP_RECEIVE_INT_REQ_RESULT,
 
 	MSB_RP_SEND_READ_STATUS_REG,
-	MSB_RP_RECIVE_STATUS_REG,
+	MSB_RP_RECEIVE_STATUS_REG,
 
 	MSB_RP_SEND_OOB_READ,
 	MSB_RP_RECEIVE_OOB_READ,
diff --git a/drivers/memstick/host/r592.c b/drivers/memstick/host/r592.c
index 1b6e91345222..31727bf285d0 100644
--- a/drivers/memstick/host/r592.c
+++ b/drivers/memstick/host/r592.c
@@ -290,7 +290,7 @@ static int r592_transfer_fifo_dma(struct r592_device *dev)
 	dbg_verbose("doing dma transfer");
 
 	dev->dma_error = 0;
-	INIT_COMPLETION(dev->dma_done);
+	reinit_completion(&dev->dma_done);
 
 	/* TODO: hidden assumption about nenth beeing always 1 */
 	sg_count = dma_map_sg(&dev->pci_dev->dev, &dev->req->sg, 1, is_write ?
diff --git a/drivers/mfd/88pm860x-core.c b/drivers/mfd/88pm860x-core.c
index 7ebe9ef1eba6..c9b1f6422941 100644
--- a/drivers/mfd/88pm860x-core.c
+++ b/drivers/mfd/88pm860x-core.c
@@ -1247,7 +1247,7 @@ static struct i2c_driver pm860x_driver = {
 		.name	= "88PM860x",
 		.owner	= THIS_MODULE,
 		.pm     = &pm860x_pm_ops,
-		.of_match_table	= of_match_ptr(pm860x_dt_ids),
+		.of_match_table	= pm860x_dt_ids,
 	},
 	.probe		= pm860x_probe,
 	.remove		= pm860x_remove,
diff --git a/drivers/mfd/Kconfig b/drivers/mfd/Kconfig
index 914c3d142f78..62a60caa5d1f 100644
--- a/drivers/mfd/Kconfig
+++ b/drivers/mfd/Kconfig
@@ -27,6 +27,18 @@ config MFD_AS3711
 	help
 	  Support for the AS3711 PMIC from AMS
 
+config MFD_AS3722
+	bool "ams AS3722 Power Management IC"
+	select MFD_CORE
+	select REGMAP_I2C
+	select REGMAP_IRQ
+	depends on I2C && OF
+	help
+	  The ams AS3722 is a compact system PMU suitable for mobile phones,
+	  tablets etc. It has 4 DC/DC step-down regulators, 3 DC/DC step-down
+	  controllers, 11 LDOs, RTC, automatic battery, temperature and
+	  over current monitoring, GPIOs, ADC and a watchdog.
+
 config PMIC_ADP5520
 	bool "Analog Devices ADP5520/01 MFD PMIC Core Support"
 	depends on I2C=y
@@ -664,14 +676,14 @@ menu "STMicroelectronics STMPE Interface Drivers"
 depends on MFD_STMPE
 
 config STMPE_I2C
-	bool "STMicroelectronics STMPE I2C Inteface"
+	bool "STMicroelectronics STMPE I2C Interface"
 	depends on I2C=y
 	default y
 	help
 	  This is used to enable I2C interface of STMPE
 
 config STMPE_SPI
-	bool "STMicroelectronics STMPE SPI Inteface"
+	bool "STMicroelectronics STMPE SPI Interface"
 	depends on SPI_MASTER
 	help
 	  This is used to enable SPI interface of STMPE
@@ -1151,6 +1163,16 @@ config MFD_WM8994
 	  core support for the WM8994, in order to use the actual
 	  functionaltiy of the device other drivers must be enabled.
 
+config MFD_STW481X
+	bool "Support for ST Microelectronics STw481x"
+	depends on I2C && ARCH_NOMADIK
+	select REGMAP_I2C
+	select MFD_CORE
+	help
+	  Select this option to enable the STw481x chip driver used
+	  in various ST Microelectronics and ST-Ericsson embedded
+	  Nomadik series.
+
 endmenu
 endif
 
diff --git a/drivers/mfd/Makefile b/drivers/mfd/Makefile
index 15b905c6553c..8a28dc90fe78 100644
--- a/drivers/mfd/Makefile
+++ b/drivers/mfd/Makefile
@@ -162,3 +162,5 @@ obj-$(CONFIG_MFD_LM3533)	+= lm3533-core.o lm3533-ctrlbank.o
 obj-$(CONFIG_VEXPRESS_CONFIG)	+= vexpress-config.o vexpress-sysreg.o
 obj-$(CONFIG_MFD_RETU)		+= retu-mfd.o
 obj-$(CONFIG_MFD_AS3711)	+= as3711.o
+obj-$(CONFIG_MFD_AS3722)	+= as3722.o
+obj-$(CONFIG_MFD_STW481X)	+= stw481x.o
diff --git a/drivers/mfd/aat2870-core.c b/drivers/mfd/aat2870-core.c
index 6f68472e0ca6..14d9542a4eed 100644
--- a/drivers/mfd/aat2870-core.c
+++ b/drivers/mfd/aat2870-core.c
@@ -293,7 +293,7 @@ static ssize_t aat2870_reg_write_file(struct file *file,
 	unsigned long addr, val;
 	int ret;
 
-	buf_size = min(count, (sizeof(buf)-1));
+	buf_size = min(count, (size_t)(sizeof(buf)-1));
 	if (copy_from_user(buf, user_buf, buf_size)) {
 		dev_err(aat2870->dev, "Failed to copy from user\n");
 		return -EFAULT;
diff --git a/drivers/mfd/arizona-core.c b/drivers/mfd/arizona-core.c
index 022b1863d36c..75e180ceecf3 100644
--- a/drivers/mfd/arizona-core.c
+++ b/drivers/mfd/arizona-core.c
@@ -540,7 +540,7 @@ static int arizona_of_get_core_pdata(struct arizona *arizona)
 		for (i = 0; i < ARRAY_SIZE(arizona->pdata.gpio_defaults); i++) {
 			if (arizona->pdata.gpio_defaults[i] > 0xffff)
 				arizona->pdata.gpio_defaults[i] = 0;
-			if (arizona->pdata.gpio_defaults[i] == 0)
+			else if (arizona->pdata.gpio_defaults[i] == 0)
 				arizona->pdata.gpio_defaults[i] = 0x10000;
 		}
 	} else {
@@ -633,11 +633,11 @@ int arizona_dev_init(struct arizona *arizona)
 	dev_set_drvdata(arizona->dev, arizona);
 	mutex_init(&arizona->clk_lock);
 
-	arizona_of_get_core_pdata(arizona);
-
 	if (dev_get_platdata(arizona->dev))
 		memcpy(&arizona->pdata, dev_get_platdata(arizona->dev),
 		       sizeof(arizona->pdata));
+	else
+		arizona_of_get_core_pdata(arizona);
 
 	regcache_cache_only(arizona->regmap, true);
 
diff --git a/drivers/mfd/arizona-i2c.c b/drivers/mfd/arizona-i2c.c
index 51dbabf7c021..beccb790c9ba 100644
--- a/drivers/mfd/arizona-i2c.c
+++ b/drivers/mfd/arizona-i2c.c
@@ -17,6 +17,7 @@
 #include <linux/regmap.h>
 #include <linux/regulator/consumer.h>
 #include <linux/slab.h>
+#include <linux/of.h>
 
 #include <linux/mfd/arizona/core.h>
 
diff --git a/drivers/mfd/arizona-spi.c b/drivers/mfd/arizona-spi.c
index 47be7b35b5c5..1ca554b18bef 100644
--- a/drivers/mfd/arizona-spi.c
+++ b/drivers/mfd/arizona-spi.c
@@ -17,6 +17,7 @@
 #include <linux/regulator/consumer.h>
 #include <linux/slab.h>
 #include <linux/spi/spi.h>
+#include <linux/of.h>
 
 #include <linux/mfd/arizona/core.h>
 
diff --git a/drivers/mfd/as3711.c b/drivers/mfd/as3711.c
index abd3ab7c0908..ec684fcedb42 100644
--- a/drivers/mfd/as3711.c
+++ b/drivers/mfd/as3711.c
@@ -17,6 +17,7 @@
 #include <linux/mfd/as3711.h>
 #include <linux/mfd/core.h>
 #include <linux/module.h>
+#include <linux/of.h>
 #include <linux/regmap.h>
 #include <linux/slab.h>
 
diff --git a/drivers/mfd/as3722.c b/drivers/mfd/as3722.c
new file mode 100644
index 000000000000..f161f2e00df7
--- /dev/null
+++ b/drivers/mfd/as3722.c
@@ -0,0 +1,449 @@
+/*
+ * Core driver for ams AS3722 PMICs
+ *
+ * Copyright (C) 2013 AMS AG
+ * Copyright (c) 2013, NVIDIA Corporation. All rights reserved.
+ *
+ * Author: Florian Lobmaier <florian.lobmaier@ams.com>
+ * Author: Laxman Dewangan <ldewangan@nvidia.com>
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation; either version 2 of the License, or
+ * (at your option) any later version.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+ * GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program; if not, write to the Free Software
+ * Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA  02111-1307 USA
+ */
+
+#include <linux/err.h>
+#include <linux/i2c.h>
+#include <linux/interrupt.h>
+#include <linux/irq.h>
+#include <linux/kernel.h>
+#include <linux/module.h>
+#include <linux/mfd/core.h>
+#include <linux/mfd/as3722.h>
+#include <linux/of.h>
+#include <linux/regmap.h>
+#include <linux/slab.h>
+
+#define AS3722_DEVICE_ID	0x0C
+
+static const struct resource as3722_rtc_resource[] = {
+	{
+		.name = "as3722-rtc-alarm",
+		.start = AS3722_IRQ_RTC_ALARM,
+		.end = AS3722_IRQ_RTC_ALARM,
+		.flags = IORESOURCE_IRQ,
+	},
+};
+
+static const struct resource as3722_adc_resource[] = {
+	{
+		.name = "as3722-adc",
+		.start = AS3722_IRQ_ADC,
+		.end = AS3722_IRQ_ADC,
+		.flags = IORESOURCE_IRQ,
+	},
+};
+
+static struct mfd_cell as3722_devs[] = {
+	{
+		.name = "as3722-pinctrl",
+	},
+	{
+		.name = "as3722-regulator",
+	},
+	{
+		.name = "as3722-rtc",
+		.num_resources = ARRAY_SIZE(as3722_rtc_resource),
+		.resources = as3722_rtc_resource,
+	},
+	{
+		.name = "as3722-adc",
+		.num_resources = ARRAY_SIZE(as3722_adc_resource),
+		.resources = as3722_adc_resource,
+	},
+	{
+		.name = "as3722-power-off",
+	},
+};
+
+static const struct regmap_irq as3722_irqs[] = {
+	/* INT1 IRQs */
+	[AS3722_IRQ_LID] = {
+		.mask = AS3722_INTERRUPT_MASK1_LID,
+	},
+	[AS3722_IRQ_ACOK] = {
+		.mask = AS3722_INTERRUPT_MASK1_ACOK,
+	},
+	[AS3722_IRQ_ENABLE1] = {
+		.mask = AS3722_INTERRUPT_MASK1_ENABLE1,
+	},
+	[AS3722_IRQ_OCCUR_ALARM_SD0] = {
+		.mask = AS3722_INTERRUPT_MASK1_OCURR_ALARM_SD0,
+	},
+	[AS3722_IRQ_ONKEY_LONG_PRESS] = {
+		.mask = AS3722_INTERRUPT_MASK1_ONKEY_LONG,
+	},
+	[AS3722_IRQ_ONKEY] = {
+		.mask = AS3722_INTERRUPT_MASK1_ONKEY,
+	},
+	[AS3722_IRQ_OVTMP] = {
+		.mask = AS3722_INTERRUPT_MASK1_OVTMP,
+	},
+	[AS3722_IRQ_LOWBAT] = {
+		.mask = AS3722_INTERRUPT_MASK1_LOWBAT,
+	},
+
+	/* INT2 IRQs */
+	[AS3722_IRQ_SD0_LV] = {
+		.mask = AS3722_INTERRUPT_MASK2_SD0_LV,
+		.reg_offset = 1,
+	},
+	[AS3722_IRQ_SD1_LV] = {
+		.mask = AS3722_INTERRUPT_MASK2_SD1_LV,
+		.reg_offset = 1,
+	},
+	[AS3722_IRQ_SD2_LV] = {
+		.mask = AS3722_INTERRUPT_MASK2_SD2345_LV,
+		.reg_offset = 1,
+	},
+	[AS3722_IRQ_PWM1_OV_PROT] = {
+		.mask = AS3722_INTERRUPT_MASK2_PWM1_OV_PROT,
+		.reg_offset = 1,
+	},
+	[AS3722_IRQ_PWM2_OV_PROT] = {
+		.mask = AS3722_INTERRUPT_MASK2_PWM2_OV_PROT,
+		.reg_offset = 1,
+	},
+	[AS3722_IRQ_ENABLE2] = {
+		.mask = AS3722_INTERRUPT_MASK2_ENABLE2,
+		.reg_offset = 1,
+	},
+	[AS3722_IRQ_SD6_LV] = {
+		.mask = AS3722_INTERRUPT_MASK2_SD6_LV,
+		.reg_offset = 1,
+	},
+	[AS3722_IRQ_RTC_REP] = {
+		.mask = AS3722_INTERRUPT_MASK2_RTC_REP,
+		.reg_offset = 1,
+	},
+
+	/* INT3 IRQs */
+	[AS3722_IRQ_RTC_ALARM] = {
+		.mask = AS3722_INTERRUPT_MASK3_RTC_ALARM,
+		.reg_offset = 2,
+	},
+	[AS3722_IRQ_GPIO1] = {
+		.mask = AS3722_INTERRUPT_MASK3_GPIO1,
+		.reg_offset = 2,
+	},
+	[AS3722_IRQ_GPIO2] = {
+		.mask = AS3722_INTERRUPT_MASK3_GPIO2,
+		.reg_offset = 2,
+	},
+	[AS3722_IRQ_GPIO3] = {
+		.mask = AS3722_INTERRUPT_MASK3_GPIO3,
+		.reg_offset = 2,
+	},
+	[AS3722_IRQ_GPIO4] = {
+		.mask = AS3722_INTERRUPT_MASK3_GPIO4,
+		.reg_offset = 2,
+	},
+	[AS3722_IRQ_GPIO5] = {
+		.mask = AS3722_INTERRUPT_MASK3_GPIO5,
+		.reg_offset = 2,
+	},
+	[AS3722_IRQ_WATCHDOG] = {
+		.mask = AS3722_INTERRUPT_MASK3_WATCHDOG,
+		.reg_offset = 2,
+	},
+	[AS3722_IRQ_ENABLE3] = {
+		.mask = AS3722_INTERRUPT_MASK3_ENABLE3,
+		.reg_offset = 2,
+	},
+
+	/* INT4 IRQs */
+	[AS3722_IRQ_TEMP_SD0_SHUTDOWN] = {
+		.mask = AS3722_INTERRUPT_MASK4_TEMP_SD0_SHUTDOWN,
+		.reg_offset = 3,
+	},
+	[AS3722_IRQ_TEMP_SD1_SHUTDOWN] = {
+		.mask = AS3722_INTERRUPT_MASK4_TEMP_SD1_SHUTDOWN,
+		.reg_offset = 3,
+	},
+	[AS3722_IRQ_TEMP_SD2_SHUTDOWN] = {
+		.mask = AS3722_INTERRUPT_MASK4_TEMP_SD6_SHUTDOWN,
+		.reg_offset = 3,
+	},
+	[AS3722_IRQ_TEMP_SD0_ALARM] = {
+		.mask = AS3722_INTERRUPT_MASK4_TEMP_SD0_ALARM,
+		.reg_offset = 3,
+	},
+	[AS3722_IRQ_TEMP_SD1_ALARM] = {
+		.mask = AS3722_INTERRUPT_MASK4_TEMP_SD1_ALARM,
+		.reg_offset = 3,
+	},
+	[AS3722_IRQ_TEMP_SD6_ALARM] = {
+		.mask = AS3722_INTERRUPT_MASK4_TEMP_SD6_ALARM,
+		.reg_offset = 3,
+	},
+	[AS3722_IRQ_OCCUR_ALARM_SD6] = {
+		.mask = AS3722_INTERRUPT_MASK4_OCCUR_ALARM_SD6,
+		.reg_offset = 3,
+	},
+	[AS3722_IRQ_ADC] = {
+		.mask = AS3722_INTERRUPT_MASK4_ADC,
+		.reg_offset = 3,
+	},
+};
+
+static const struct regmap_irq_chip as3722_irq_chip = {
+	.name = "as3722",
+	.irqs = as3722_irqs,
+	.num_irqs = ARRAY_SIZE(as3722_irqs),
+	.num_regs = 4,
+	.status_base = AS3722_INTERRUPT_STATUS1_REG,
+	.mask_base = AS3722_INTERRUPT_MASK1_REG,
+};
+
+static int as3722_check_device_id(struct as3722 *as3722)
+{
+	u32 val;
+	int ret;
+
+	/* Check that this is actually a AS3722 */
+	ret = as3722_read(as3722, AS3722_ASIC_ID1_REG, &val);
+	if (ret < 0) {
+		dev_err(as3722->dev, "ASIC_ID1 read failed: %d\n", ret);
+		return ret;
+	}
+
+	if (val != AS3722_DEVICE_ID) {
+		dev_err(as3722->dev, "Device is not AS3722, ID is 0x%x\n", val);
+		return -ENODEV;
+	}
+
+	ret = as3722_read(as3722, AS3722_ASIC_ID2_REG, &val);
+	if (ret < 0) {
+		dev_err(as3722->dev, "ASIC_ID2 read failed: %d\n", ret);
+		return ret;
+	}
+
+	dev_info(as3722->dev, "AS3722 with revision 0x%x found\n", val);
+	return 0;
+}
+
+static int as3722_configure_pullups(struct as3722 *as3722)
+{
+	int ret;
+	u32 val = 0;
+
+	if (as3722->en_intern_int_pullup)
+		val |= AS3722_INT_PULL_UP;
+	if (as3722->en_intern_i2c_pullup)
+		val |= AS3722_I2C_PULL_UP;
+
+	ret = as3722_update_bits(as3722, AS3722_IOVOLTAGE_REG,
+			AS3722_INT_PULL_UP | AS3722_I2C_PULL_UP, val);
+	if (ret < 0)
+		dev_err(as3722->dev, "IOVOLTAGE_REG update failed: %d\n", ret);
+	return ret;
+}
+
+static const struct regmap_range as3722_readable_ranges[] = {
+	regmap_reg_range(AS3722_SD0_VOLTAGE_REG, AS3722_SD6_VOLTAGE_REG),
+	regmap_reg_range(AS3722_GPIO0_CONTROL_REG, AS3722_LDO7_VOLTAGE_REG),
+	regmap_reg_range(AS3722_LDO9_VOLTAGE_REG, AS3722_REG_SEQU_MOD3_REG),
+	regmap_reg_range(AS3722_SD_PHSW_CTRL_REG, AS3722_PWM_CONTROL_H_REG),
+	regmap_reg_range(AS3722_WATCHDOG_TIMER_REG, AS3722_WATCHDOG_TIMER_REG),
+	regmap_reg_range(AS3722_WATCHDOG_SOFTWARE_SIGNAL_REG,
+					AS3722_BATTERY_VOLTAGE_MONITOR2_REG),
+	regmap_reg_range(AS3722_SD_CONTROL_REG, AS3722_PWM_VCONTROL4_REG),
+	regmap_reg_range(AS3722_BB_CHARGER_REG, AS3722_SRAM_REG),
+	regmap_reg_range(AS3722_RTC_ACCESS_REG, AS3722_RTC_ACCESS_REG),
+	regmap_reg_range(AS3722_RTC_STATUS_REG, AS3722_TEMP_STATUS_REG),
+	regmap_reg_range(AS3722_ADC0_CONTROL_REG, AS3722_ADC_CONFIGURATION_REG),
+	regmap_reg_range(AS3722_ASIC_ID1_REG, AS3722_ASIC_ID2_REG),
+	regmap_reg_range(AS3722_LOCK_REG, AS3722_LOCK_REG),
+};
+
+static const struct regmap_access_table as3722_readable_table = {
+	.yes_ranges = as3722_readable_ranges,
+	.n_yes_ranges = ARRAY_SIZE(as3722_readable_ranges),
+};
+
+static const struct regmap_range as3722_writable_ranges[] = {
+	regmap_reg_range(AS3722_SD0_VOLTAGE_REG, AS3722_SD6_VOLTAGE_REG),
+	regmap_reg_range(AS3722_GPIO0_CONTROL_REG, AS3722_LDO7_VOLTAGE_REG),
+	regmap_reg_range(AS3722_LDO9_VOLTAGE_REG, AS3722_GPIO_SIGNAL_OUT_REG),
+	regmap_reg_range(AS3722_REG_SEQU_MOD1_REG, AS3722_REG_SEQU_MOD3_REG),
+	regmap_reg_range(AS3722_SD_PHSW_CTRL_REG, AS3722_PWM_CONTROL_H_REG),
+	regmap_reg_range(AS3722_WATCHDOG_TIMER_REG, AS3722_WATCHDOG_TIMER_REG),
+	regmap_reg_range(AS3722_WATCHDOG_SOFTWARE_SIGNAL_REG,
+					AS3722_BATTERY_VOLTAGE_MONITOR2_REG),
+	regmap_reg_range(AS3722_SD_CONTROL_REG, AS3722_PWM_VCONTROL4_REG),
+	regmap_reg_range(AS3722_BB_CHARGER_REG, AS3722_SRAM_REG),
+	regmap_reg_range(AS3722_INTERRUPT_MASK1_REG, AS3722_TEMP_STATUS_REG),
+	regmap_reg_range(AS3722_ADC0_CONTROL_REG, AS3722_ADC1_CONTROL_REG),
+	regmap_reg_range(AS3722_ADC1_THRESHOLD_HI_MSB_REG,
+					AS3722_ADC_CONFIGURATION_REG),
+	regmap_reg_range(AS3722_LOCK_REG, AS3722_LOCK_REG),
+};
+
+static const struct regmap_access_table as3722_writable_table = {
+	.yes_ranges = as3722_writable_ranges,
+	.n_yes_ranges = ARRAY_SIZE(as3722_writable_ranges),
+};
+
+static const struct regmap_range as3722_cacheable_ranges[] = {
+	regmap_reg_range(AS3722_SD0_VOLTAGE_REG, AS3722_LDO11_VOLTAGE_REG),
+	regmap_reg_range(AS3722_SD_CONTROL_REG, AS3722_LDOCONTROL1_REG),
+};
+
+static const struct regmap_access_table as3722_volatile_table = {
+	.no_ranges = as3722_cacheable_ranges,
+	.n_no_ranges = ARRAY_SIZE(as3722_cacheable_ranges),
+};
+
+static const struct regmap_config as3722_regmap_config = {
+	.reg_bits = 8,
+	.val_bits = 8,
+	.max_register = AS3722_MAX_REGISTER,
+	.cache_type = REGCACHE_RBTREE,
+	.rd_table = &as3722_readable_table,
+	.wr_table = &as3722_writable_table,
+	.volatile_table = &as3722_volatile_table,
+};
+
+static int as3722_i2c_of_probe(struct i2c_client *i2c,
+			struct as3722 *as3722)
+{
+	struct device_node *np = i2c->dev.of_node;
+	struct irq_data *irq_data;
+
+	if (!np) {
+		dev_err(&i2c->dev, "Device Tree not found\n");
+		return -EINVAL;
+	}
+
+	irq_data = irq_get_irq_data(i2c->irq);
+	if (!irq_data) {
+		dev_err(&i2c->dev, "Invalid IRQ: %d\n", i2c->irq);
+		return -EINVAL;
+	}
+
+	as3722->en_intern_int_pullup = of_property_read_bool(np,
+					"ams,enable-internal-int-pullup");
+	as3722->en_intern_i2c_pullup = of_property_read_bool(np,
+					"ams,enable-internal-i2c-pullup");
+	as3722->irq_flags = irqd_get_trigger_type(irq_data);
+	dev_dbg(&i2c->dev, "IRQ flags are 0x%08lx\n", as3722->irq_flags);
+	return 0;
+}
+
+static int as3722_i2c_probe(struct i2c_client *i2c,
+			const struct i2c_device_id *id)
+{
+	struct as3722 *as3722;
+	unsigned long irq_flags;
+	int ret;
+
+	as3722 = devm_kzalloc(&i2c->dev, sizeof(struct as3722), GFP_KERNEL);
+	if (!as3722)
+		return -ENOMEM;
+
+	as3722->dev = &i2c->dev;
+	as3722->chip_irq = i2c->irq;
+	i2c_set_clientdata(i2c, as3722);
+
+	ret = as3722_i2c_of_probe(i2c, as3722);
+	if (ret < 0)
+		return ret;
+
+	as3722->regmap = devm_regmap_init_i2c(i2c, &as3722_regmap_config);
+	if (IS_ERR(as3722->regmap)) {
+		ret = PTR_ERR(as3722->regmap);
+		dev_err(&i2c->dev, "regmap init failed: %d\n", ret);
+		return ret;
+	}
+
+	ret = as3722_check_device_id(as3722);
+	if (ret < 0)
+		return ret;
+
+	irq_flags = as3722->irq_flags | IRQF_ONESHOT;
+	ret = regmap_add_irq_chip(as3722->regmap, as3722->chip_irq,
+			irq_flags, -1, &as3722_irq_chip,
+			&as3722->irq_data);
+	if (ret < 0) {
+		dev_err(as3722->dev, "Failed to add regmap irq: %d\n", ret);
+		return ret;
+	}
+
+	ret = as3722_configure_pullups(as3722);
+	if (ret < 0)
+		goto scrub;
+
+	ret = mfd_add_devices(&i2c->dev, -1, as3722_devs,
+			ARRAY_SIZE(as3722_devs), NULL, 0,
+			regmap_irq_get_domain(as3722->irq_data));
+	if (ret) {
+		dev_err(as3722->dev, "Failed to add MFD devices: %d\n", ret);
+		goto scrub;
+	}
+
+	dev_dbg(as3722->dev, "AS3722 core driver initialized successfully\n");
+	return 0;
+
+scrub:
+	regmap_del_irq_chip(as3722->chip_irq, as3722->irq_data);
+	return ret;
+}
+
+static int as3722_i2c_remove(struct i2c_client *i2c)
+{
+	struct as3722 *as3722 = i2c_get_clientdata(i2c);
+
+	mfd_remove_devices(as3722->dev);
+	regmap_del_irq_chip(as3722->chip_irq, as3722->irq_data);
+	return 0;
+}
+
+static const struct of_device_id as3722_of_match[] = {
+	{ .compatible = "ams,as3722", },
+	{},
+};
+MODULE_DEVICE_TABLE(of, as3722_of_match);
+
+static const struct i2c_device_id as3722_i2c_id[] = {
+	{ "as3722", 0 },
+	{},
+};
+MODULE_DEVICE_TABLE(i2c, as3722_i2c_id);
+
+static struct i2c_driver as3722_i2c_driver = {
+	.driver = {
+		.name = "as3722",
+		.owner = THIS_MODULE,
+		.of_match_table = as3722_of_match,
+	},
+	.probe = as3722_i2c_probe,
+	.remove = as3722_i2c_remove,
+	.id_table = as3722_i2c_id,
+};
+
+module_i2c_driver(as3722_i2c_driver);
+
+MODULE_DESCRIPTION("I2C support for AS3722 PMICs");
+MODULE_AUTHOR("Florian Lobmaier <florian.lobmaier@ams.com>");
+MODULE_AUTHOR("Laxman Dewangan <ldewangan@nvidia.com>");
+MODULE_LICENSE("GPL");
diff --git a/drivers/mfd/da9052-i2c.c b/drivers/mfd/da9052-i2c.c
index 6a9fec40d018..c319c4ef5d49 100644
--- a/drivers/mfd/da9052-i2c.c
+++ b/drivers/mfd/da9052-i2c.c
@@ -86,7 +86,11 @@ static int da9052_i2c_fix(struct da9052 *da9052, unsigned char reg)
 	return 0;
 }
 
-static int da9052_i2c_enable_multiwrite(struct da9052 *da9052)
+/*
+ * According to errata item 24, multiwrite mode should be avoided
+ * in order to prevent register data corruption after power-down.
+ */
+static int da9052_i2c_disable_multiwrite(struct da9052 *da9052)
 {
 	int reg_val, ret;
 
@@ -94,8 +98,8 @@ static int da9052_i2c_enable_multiwrite(struct da9052 *da9052)
 	if (ret < 0)
 		return ret;
 
-	if (reg_val & DA9052_CONTROL_B_WRITEMODE) {
-		reg_val &= ~DA9052_CONTROL_B_WRITEMODE;
+	if (!(reg_val & DA9052_CONTROL_B_WRITEMODE)) {
+		reg_val |= DA9052_CONTROL_B_WRITEMODE;
 		ret = regmap_write(da9052->regmap, DA9052_CONTROL_B_REG,
 				   reg_val);
 		if (ret < 0)
@@ -154,7 +158,7 @@ static int da9052_i2c_probe(struct i2c_client *client,
 		return ret;
 	}
 
-	ret = da9052_i2c_enable_multiwrite(da9052);
+	ret = da9052_i2c_disable_multiwrite(da9052);
 	if (ret < 0)
 		return ret;
 
diff --git a/drivers/mfd/ezx-pcap.c b/drivers/mfd/ezx-pcap.c
index 7245b0c5b794..2ed774e7d342 100644
--- a/drivers/mfd/ezx-pcap.c
+++ b/drivers/mfd/ezx-pcap.c
@@ -394,16 +394,12 @@ static int pcap_add_subdev(struct pcap_chip *pcap,
 static int ezx_pcap_remove(struct spi_device *spi)
 {
 	struct pcap_chip *pcap = spi_get_drvdata(spi);
-	struct pcap_platform_data *pdata = dev_get_platdata(&spi->dev);
-	int i, adc_irq;
+	int i;
 
 	/* remove all registered subdevs */
 	device_for_each_child(&spi->dev, NULL, pcap_remove_subdev);
 
 	/* cleanup ADC */
-	adc_irq = pcap_to_irq(pcap, (pdata->config & PCAP_SECOND_PORT) ?
-				PCAP_IRQ_ADCDONE2 : PCAP_IRQ_ADCDONE);
-	devm_free_irq(&spi->dev, adc_irq, pcap);
 	mutex_lock(&pcap->adc_mutex);
 	for (i = 0; i < PCAP_ADC_MAXQ; i++)
 		kfree(pcap->adc_queue[i]);
@@ -509,8 +505,6 @@ static int ezx_pcap_probe(struct spi_device *spi)
 
 remove_subdevs:
 	device_for_each_child(&spi->dev, NULL, pcap_remove_subdev);
-/* free_adc: */
-	devm_free_irq(&spi->dev, adc_irq, pcap);
 free_irqchip:
 	for (i = pcap->irq_base; i < (pcap->irq_base + PCAP_NIRQS); i++)
 		irq_set_chip_and_handler(i, NULL, NULL);
diff --git a/drivers/mfd/lpc_ich.c b/drivers/mfd/lpc_ich.c
index 9483bc8472a5..da1c6566d93d 100644
--- a/drivers/mfd/lpc_ich.c
+++ b/drivers/mfd/lpc_ich.c
@@ -53,6 +53,7 @@
  *	document number TBD : Wellsburg
  *	document number TBD : Avoton SoC
  *	document number TBD : Coleto Creek
+ *	document number TBD : Wildcat Point-LP
  */
 
 #define pr_fmt(fmt) KBUILD_MODNAME ": " fmt
@@ -211,6 +212,7 @@ enum lpc_chipsets {
 	LPC_WBG,	/* Wellsburg */
 	LPC_AVN,	/* Avoton SoC */
 	LPC_COLETO,	/* Coleto Creek */
+	LPC_WPT_LP,	/* Wildcat Point-LP */
 };
 
 static struct lpc_ich_info lpc_chipset_info[] = {
@@ -503,6 +505,10 @@ static struct lpc_ich_info lpc_chipset_info[] = {
 		.name = "Coleto Creek",
 		.iTCO_version = 2,
 	},
+	[LPC_WPT_LP] = {
+		.name = "Lynx Point_LP",
+		.iTCO_version = 2,
+	},
 };
 
 /*
@@ -721,6 +727,13 @@ static DEFINE_PCI_DEVICE_TABLE(lpc_ich_ids) = {
 	{ PCI_VDEVICE(INTEL, 0x1f3a), LPC_AVN},
 	{ PCI_VDEVICE(INTEL, 0x1f3b), LPC_AVN},
 	{ PCI_VDEVICE(INTEL, 0x2390), LPC_COLETO},
+	{ PCI_VDEVICE(INTEL, 0x9cc1), LPC_WPT_LP},
+	{ PCI_VDEVICE(INTEL, 0x9cc2), LPC_WPT_LP},
+	{ PCI_VDEVICE(INTEL, 0x9cc3), LPC_WPT_LP},
+	{ PCI_VDEVICE(INTEL, 0x9cc5), LPC_WPT_LP},
+	{ PCI_VDEVICE(INTEL, 0x9cc6), LPC_WPT_LP},
+	{ PCI_VDEVICE(INTEL, 0x9cc7), LPC_WPT_LP},
+	{ PCI_VDEVICE(INTEL, 0x9cc9), LPC_WPT_LP},
 	{ 0, },			/* End of list */
 };
 MODULE_DEVICE_TABLE(pci, lpc_ich_ids);
@@ -969,7 +982,6 @@ static int lpc_ich_probe(struct pci_dev *dev,
 	if (!cell_added) {
 		dev_warn(&dev->dev, "No MFD cells added\n");
 		lpc_ich_restore_config_space(dev);
-		pci_set_drvdata(dev, NULL);
 		return -ENODEV;
 	}
 
@@ -980,7 +992,6 @@ static void lpc_ich_remove(struct pci_dev *dev)
 {
 	mfd_remove_devices(&dev->dev);
 	lpc_ich_restore_config_space(dev);
-	pci_set_drvdata(dev, NULL);
 }
 
 static struct pci_driver lpc_ich_driver = {
diff --git a/drivers/mfd/lpc_sch.c b/drivers/mfd/lpc_sch.c
index 8cc6aac27cb2..fbfbf0b7f97a 100644
--- a/drivers/mfd/lpc_sch.c
+++ b/drivers/mfd/lpc_sch.c
@@ -59,18 +59,21 @@ static struct mfd_cell isch_smbus_cell = {
 	.name = "isch_smbus",
 	.num_resources = 1,
 	.resources = &smbus_sch_resource,
+	.ignore_resource_conflicts = true,
 };
 
 static struct mfd_cell sch_gpio_cell = {
 	.name = "sch_gpio",
 	.num_resources = 1,
 	.resources = &gpio_sch_resource,
+	.ignore_resource_conflicts = true,
 };
 
 static struct mfd_cell wdt_sch_cell = {
 	.name = "ie6xx_wdt",
 	.num_resources = 1,
 	.resources = &wdt_sch_resource,
+	.ignore_resource_conflicts = true,
 };
 
 static DEFINE_PCI_DEVICE_TABLE(lpc_sch_ids) = {
diff --git a/drivers/mfd/max77686.c b/drivers/mfd/max77686.c
index 522be67b2e68..34520cbe8afb 100644
--- a/drivers/mfd/max77686.c
+++ b/drivers/mfd/max77686.c
@@ -31,6 +31,7 @@
 #include <linux/mfd/max77686.h>
 #include <linux/mfd/max77686-private.h>
 #include <linux/err.h>
+#include <linux/of.h>
 
 #define I2C_ADDR_RTC	(0x0C >> 1)
 
diff --git a/drivers/mfd/max77693-irq.c b/drivers/mfd/max77693-irq.c
index 1029d018c739..66b58fe77094 100644
--- a/drivers/mfd/max77693-irq.c
+++ b/drivers/mfd/max77693-irq.c
@@ -128,7 +128,8 @@ static void max77693_irq_sync_unlock(struct irq_data *data)
 static const inline struct max77693_irq_data *
 irq_to_max77693_irq(struct max77693_dev *max77693, int irq)
 {
-	return &max77693_irqs[irq];
+	struct irq_data *data = irq_get_irq_data(irq);
+	return &max77693_irqs[data->hwirq];
 }
 
 static void max77693_irq_mask(struct irq_data *data)
diff --git a/drivers/mfd/max77693.c b/drivers/mfd/max77693.c
index c04723efc707..9f92463f4f7e 100644
--- a/drivers/mfd/max77693.c
+++ b/drivers/mfd/max77693.c
@@ -28,6 +28,7 @@
 #include <linux/i2c.h>
 #include <linux/err.h>
 #include <linux/interrupt.h>
+#include <linux/of.h>
 #include <linux/pm_runtime.h>
 #include <linux/mutex.h>
 #include <linux/mfd/core.h>
@@ -110,15 +111,9 @@ static int max77693_i2c_probe(struct i2c_client *i2c,
 			      const struct i2c_device_id *id)
 {
 	struct max77693_dev *max77693;
-	struct max77693_platform_data *pdata = dev_get_platdata(&i2c->dev);
 	u8 reg_data;
 	int ret = 0;
 
-	if (!pdata) {
-		dev_err(&i2c->dev, "No platform data found.\n");
-		return -EINVAL;
-	}
-
 	max77693 = devm_kzalloc(&i2c->dev,
 			sizeof(struct max77693_dev), GFP_KERNEL);
 	if (max77693 == NULL)
@@ -138,8 +133,6 @@ static int max77693_i2c_probe(struct i2c_client *i2c,
 		return ret;
 	}
 
-	max77693->wakeup = pdata->wakeup;
-
 	ret = max77693_read_reg(max77693->regmap, MAX77693_PMIC_REG_PMIC_ID2,
 				&reg_data);
 	if (ret < 0) {
@@ -179,8 +172,6 @@ static int max77693_i2c_probe(struct i2c_client *i2c,
 	if (ret < 0)
 		goto err_mfd;
 
-	device_init_wakeup(max77693->dev, pdata->wakeup);
-
 	return ret;
 
 err_mfd:
@@ -235,11 +226,19 @@ static const struct dev_pm_ops max77693_pm = {
 	.resume = max77693_resume,
 };
 
+#ifdef CONFIG_OF
+static struct of_device_id max77693_dt_match[] = {
+	{ .compatible = "maxim,max77693" },
+	{},
+};
+#endif
+
 static struct i2c_driver max77693_i2c_driver = {
 	.driver = {
 		   .name = "max77693",
 		   .owner = THIS_MODULE,
 		   .pm = &max77693_pm,
+		   .of_match_table = of_match_ptr(max77693_dt_match),
 	},
 	.probe = max77693_i2c_probe,
 	.remove = max77693_i2c_remove,
diff --git a/drivers/mfd/max8907.c b/drivers/mfd/max8907.c
index e9b1c93a3ade..3bbfedc07f41 100644
--- a/drivers/mfd/max8907.c
+++ b/drivers/mfd/max8907.c
@@ -17,6 +17,7 @@
 #include <linux/mfd/core.h>
 #include <linux/mfd/max8907.h>
 #include <linux/module.h>
+#include <linux/of.h>
 #include <linux/of_device.h>
 #include <linux/regmap.h>
 #include <linux/slab.h>
diff --git a/drivers/mfd/max8925-i2c.c b/drivers/mfd/max8925-i2c.c
index de7fb80a6052..176aa26fc787 100644
--- a/drivers/mfd/max8925-i2c.c
+++ b/drivers/mfd/max8925-i2c.c
@@ -238,7 +238,7 @@ static struct i2c_driver max8925_driver = {
 		.name	= "max8925",
 		.owner	= THIS_MODULE,
 		.pm     = &max8925_pm_ops,
-		.of_match_table = of_match_ptr(max8925_dt_ids),
+		.of_match_table = max8925_dt_ids,
 	},
 	.probe		= max8925_probe,
 	.remove		= max8925_remove,
diff --git a/drivers/mfd/max8997.c b/drivers/mfd/max8997.c
index cee098c0dae3..791aea3e96ce 100644
--- a/drivers/mfd/max8997.c
+++ b/drivers/mfd/max8997.c
@@ -24,6 +24,7 @@
 #include <linux/err.h>
 #include <linux/slab.h>
 #include <linux/i2c.h>
+#include <linux/of.h>
 #include <linux/of_irq.h>
 #include <linux/interrupt.h>
 #include <linux/pm_runtime.h>
diff --git a/drivers/mfd/mc13xxx-i2c.c b/drivers/mfd/mc13xxx-i2c.c
index f745e27ee874..898bd335cd8e 100644
--- a/drivers/mfd/mc13xxx-i2c.c
+++ b/drivers/mfd/mc13xxx-i2c.c
@@ -78,7 +78,6 @@ static int mc13xxx_i2c_probe(struct i2c_client *client,
 		ret = PTR_ERR(mc13xxx->regmap);
 		dev_err(mc13xxx->dev, "Failed to initialize register map: %d\n",
 				ret);
-		dev_set_drvdata(&client->dev, NULL);
 		return ret;
 	}
 
diff --git a/drivers/mfd/mfd-core.c b/drivers/mfd/mfd-core.c
index adc8ea36e7c4..267649244737 100644
--- a/drivers/mfd/mfd-core.c
+++ b/drivers/mfd/mfd-core.c
@@ -64,7 +64,8 @@ int mfd_cell_disable(struct platform_device *pdev)
 EXPORT_SYMBOL(mfd_cell_disable);
 
 static int mfd_platform_add_cell(struct platform_device *pdev,
-				 const struct mfd_cell *cell)
+				 const struct mfd_cell *cell,
+				 atomic_t *usage_count)
 {
 	if (!cell)
 		return 0;
@@ -73,11 +74,12 @@ static int mfd_platform_add_cell(struct platform_device *pdev,
 	if (!pdev->mfd_cell)
 		return -ENOMEM;
 
+	pdev->mfd_cell->usage_count = usage_count;
 	return 0;
 }
 
 static int mfd_add_device(struct device *parent, int id,
-			  const struct mfd_cell *cell,
+			  const struct mfd_cell *cell, atomic_t *usage_count,
 			  struct resource *mem_base,
 			  int irq_base, struct irq_domain *domain)
 {
@@ -123,7 +125,7 @@ static int mfd_add_device(struct device *parent, int id,
 			goto fail_alias;
 	}
 
-	ret = mfd_platform_add_cell(pdev, cell);
+	ret = mfd_platform_add_cell(pdev, cell, usage_count);
 	if (ret)
 		goto fail_alias;
 
@@ -192,12 +194,12 @@ fail_alloc:
 }
 
 int mfd_add_devices(struct device *parent, int id,
-		    struct mfd_cell *cells, int n_devs,
+		    const struct mfd_cell *cells, int n_devs,
 		    struct resource *mem_base,
 		    int irq_base, struct irq_domain *domain)
 {
 	int i;
-	int ret = 0;
+	int ret;
 	atomic_t *cnts;
 
 	/* initialize reference counting for all cells */
@@ -207,16 +209,19 @@ int mfd_add_devices(struct device *parent, int id,
 
 	for (i = 0; i < n_devs; i++) {
 		atomic_set(&cnts[i], 0);
-		cells[i].usage_count = &cnts[i];
-		ret = mfd_add_device(parent, id, cells + i, mem_base,
+		ret = mfd_add_device(parent, id, cells + i, cnts + i, mem_base,
 				     irq_base, domain);
 		if (ret)
-			break;
+			goto fail;
 	}
 
-	if (ret)
-		mfd_remove_devices(parent);
+	return 0;
 
+fail:
+	if (i)
+		mfd_remove_devices(parent);
+	else
+		kfree(cnts);
 	return ret;
 }
 EXPORT_SYMBOL(mfd_add_devices);
@@ -271,8 +276,8 @@ int mfd_clone_cell(const char *cell, const char **clones, size_t n_clones)
 	for (i = 0; i < n_clones; i++) {
 		cell_entry.name = clones[i];
 		/* don't give up if a single call fails; just report error */
-		if (mfd_add_device(pdev->dev.parent, -1, &cell_entry, NULL, 0,
-				   NULL))
+		if (mfd_add_device(pdev->dev.parent, -1, &cell_entry,
+				   cell_entry.usage_count, NULL, 0, NULL))
 			dev_err(dev, "failed to create platform device '%s'\n",
 					clones[i]);
 	}
diff --git a/drivers/mfd/omap-usb-host.c b/drivers/mfd/omap-usb-host.c
index 29ee54d68512..142650fdc058 100644
--- a/drivers/mfd/omap-usb-host.c
+++ b/drivers/mfd/omap-usb-host.c
@@ -328,13 +328,13 @@ static int usbhs_runtime_resume(struct device *dev)
 	omap_tll_enable(pdata);
 
 	if (!IS_ERR(omap->ehci_logic_fck))
-		clk_enable(omap->ehci_logic_fck);
+		clk_prepare_enable(omap->ehci_logic_fck);
 
 	for (i = 0; i < omap->nports; i++) {
 		switch (pdata->port_mode[i]) {
 		case OMAP_EHCI_PORT_MODE_HSIC:
 			if (!IS_ERR(omap->hsic60m_clk[i])) {
-				r = clk_enable(omap->hsic60m_clk[i]);
+				r = clk_prepare_enable(omap->hsic60m_clk[i]);
 				if (r) {
 					dev_err(dev,
 					 "Can't enable port %d hsic60m clk:%d\n",
@@ -343,7 +343,7 @@ static int usbhs_runtime_resume(struct device *dev)
 			}
 
 			if (!IS_ERR(omap->hsic480m_clk[i])) {
-				r = clk_enable(omap->hsic480m_clk[i]);
+				r = clk_prepare_enable(omap->hsic480m_clk[i]);
 				if (r) {
 					dev_err(dev,
 					 "Can't enable port %d hsic480m clk:%d\n",
@@ -354,7 +354,7 @@ static int usbhs_runtime_resume(struct device *dev)
 
 		case OMAP_EHCI_PORT_MODE_TLL:
 			if (!IS_ERR(omap->utmi_clk[i])) {
-				r = clk_enable(omap->utmi_clk[i]);
+				r = clk_prepare_enable(omap->utmi_clk[i]);
 				if (r) {
 					dev_err(dev,
 					 "Can't enable port %d clk : %d\n",
@@ -382,15 +382,15 @@ static int usbhs_runtime_suspend(struct device *dev)
 		switch (pdata->port_mode[i]) {
 		case OMAP_EHCI_PORT_MODE_HSIC:
 			if (!IS_ERR(omap->hsic60m_clk[i]))
-				clk_disable(omap->hsic60m_clk[i]);
+				clk_disable_unprepare(omap->hsic60m_clk[i]);
 
 			if (!IS_ERR(omap->hsic480m_clk[i]))
-				clk_disable(omap->hsic480m_clk[i]);
+				clk_disable_unprepare(omap->hsic480m_clk[i]);
 		/* Fall through as utmi_clks were used in HSIC mode */
 
 		case OMAP_EHCI_PORT_MODE_TLL:
 			if (!IS_ERR(omap->utmi_clk[i]))
-				clk_disable(omap->utmi_clk[i]);
+				clk_disable_unprepare(omap->utmi_clk[i]);
 			break;
 		default:
 			break;
@@ -398,7 +398,7 @@ static int usbhs_runtime_suspend(struct device *dev)
 	}
 
 	if (!IS_ERR(omap->ehci_logic_fck))
-		clk_disable(omap->ehci_logic_fck);
+		clk_disable_unprepare(omap->ehci_logic_fck);
 
 	omap_tll_disable(pdata);
 
@@ -893,7 +893,7 @@ static struct platform_driver usbhs_omap_driver = {
 		.name		= (char *)usbhs_driver_name,
 		.owner		= THIS_MODULE,
 		.pm		= &usbhsomap_dev_pm_ops,
-		.of_match_table = of_match_ptr(usbhs_omap_dt_ids),
+		.of_match_table = usbhs_omap_dt_ids,
 	},
 	.remove		= usbhs_omap_remove,
 };
diff --git a/drivers/mfd/omap-usb-tll.c b/drivers/mfd/omap-usb-tll.c
index e59ac4cbac96..0d946ae14453 100644
--- a/drivers/mfd/omap-usb-tll.c
+++ b/drivers/mfd/omap-usb-tll.c
@@ -320,7 +320,7 @@ static struct platform_driver usbtll_omap_driver = {
 	.driver = {
 		.name		= (char *)usbtll_driver_name,
 		.owner		= THIS_MODULE,
-		.of_match_table = of_match_ptr(usbtll_omap_dt_ids),
+		.of_match_table = usbtll_omap_dt_ids,
 	},
 	.probe		= usbtll_omap_probe,
 	.remove		= usbtll_omap_remove,
@@ -429,7 +429,7 @@ int omap_tll_enable(struct usbhs_omap_platform_data *pdata)
 			if (IS_ERR(tll->ch_clk[i]))
 				continue;
 
-			r = clk_enable(tll->ch_clk[i]);
+			r = clk_prepare_enable(tll->ch_clk[i]);
 			if (r) {
 				dev_err(tll_dev,
 				 "Error enabling ch %d clock: %d\n", i, r);
@@ -460,7 +460,7 @@ int omap_tll_disable(struct usbhs_omap_platform_data *pdata)
 	for (i = 0; i < tll->nch; i++) {
 		if (omap_usb_mode_needs_tll(pdata->port_mode[i])) {
 			if (!IS_ERR(tll->ch_clk[i]))
-				clk_disable(tll->ch_clk[i]);
+				clk_disable_unprepare(tll->ch_clk[i]);
 		}
 	}
 
diff --git a/drivers/mfd/palmas.c b/drivers/mfd/palmas.c
index 135afabe4ae2..d280d789e55a 100644
--- a/drivers/mfd/palmas.c
+++ b/drivers/mfd/palmas.c
@@ -368,6 +368,7 @@ static const struct of_device_id of_palmas_match_tbl[] = {
 	},
 	{ },
 };
+MODULE_DEVICE_TABLE(of, of_palmas_match_tbl);
 
 static int palmas_i2c_probe(struct i2c_client *i2c,
 			    const struct i2c_device_id *id)
@@ -402,7 +403,7 @@ static int palmas_i2c_probe(struct i2c_client *i2c,
 	palmas->dev = &i2c->dev;
 	palmas->irq = i2c->irq;
 
-	match = of_match_device(of_match_ptr(of_palmas_match_tbl), &i2c->dev);
+	match = of_match_device(of_palmas_match_tbl, &i2c->dev);
 
 	if (!match)
 		return -ENODATA;
@@ -421,7 +422,7 @@ static int palmas_i2c_probe(struct i2c_client *i2c,
 				dev_err(palmas->dev,
 					"can't attach client %d\n", i);
 				ret = -ENOMEM;
-				goto err;
+				goto err_i2c;
 			}
 			palmas->i2c_clients[i]->dev.of_node = of_node_get(node);
 		}
@@ -432,7 +433,7 @@ static int palmas_i2c_probe(struct i2c_client *i2c,
 			dev_err(palmas->dev,
 				"Failed to allocate regmap %d, err: %d\n",
 				i, ret);
-			goto err;
+			goto err_i2c;
 		}
 	}
 
@@ -451,7 +452,7 @@ static int palmas_i2c_probe(struct i2c_client *i2c,
 			reg);
 	if (ret < 0) {
 		dev_err(palmas->dev, "POLARITY_CTRL updat failed: %d\n", ret);
-		goto err;
+		goto err_i2c;
 	}
 
 	/* Change IRQ into clear on read mode for efficiency */
@@ -465,7 +466,7 @@ static int palmas_i2c_probe(struct i2c_client *i2c,
 			IRQF_ONESHOT | pdata->irq_flags, 0, &palmas_irq_chip,
 			&palmas->irq_data);
 	if (ret < 0)
-		goto err;
+		goto err_i2c;
 
 no_irq:
 	slave = PALMAS_BASE_TO_SLAVE(PALMAS_PU_PD_OD_BASE);
@@ -551,7 +552,6 @@ no_irq:
 		} else if (pdata->pm_off && !pm_power_off) {
 			palmas_dev = palmas;
 			pm_power_off = palmas_power_off;
-			return ret;
 		}
 	}
 
@@ -559,17 +559,31 @@ no_irq:
 
 err_irq:
 	regmap_del_irq_chip(palmas->irq, palmas->irq_data);
-err:
+err_i2c:
+	for (i = 1; i < PALMAS_NUM_CLIENTS; i++) {
+		if (palmas->i2c_clients[i])
+			i2c_unregister_device(palmas->i2c_clients[i]);
+	}
 	return ret;
 }
 
 static int palmas_i2c_remove(struct i2c_client *i2c)
 {
 	struct palmas *palmas = i2c_get_clientdata(i2c);
+	int i;
 
-	mfd_remove_devices(palmas->dev);
 	regmap_del_irq_chip(palmas->irq, palmas->irq_data);
 
+	for (i = 1; i < PALMAS_NUM_CLIENTS; i++) {
+		if (palmas->i2c_clients[i])
+			i2c_unregister_device(palmas->i2c_clients[i]);
+	}
+
+	if (palmas == palmas_dev) {
+		pm_power_off = NULL;
+		palmas_dev = NULL;
+	}
+
 	return 0;
 }
 
diff --git a/drivers/mfd/pm8921-core.c b/drivers/mfd/pm8921-core.c
index a6841f77aa5e..484fe66e6c88 100644
--- a/drivers/mfd/pm8921-core.c
+++ b/drivers/mfd/pm8921-core.c
@@ -171,11 +171,12 @@ static int pm8921_remove(struct platform_device *pdev)
 	drvdata = platform_get_drvdata(pdev);
 	if (drvdata)
 		pmic = drvdata->pm_chip_data;
-	if (pmic)
+	if (pmic) {
 		mfd_remove_devices(pmic->dev);
-	if (pmic->irq_chip) {
-		pm8xxx_irq_exit(pmic->irq_chip);
-		pmic->irq_chip = NULL;
+		if (pmic->irq_chip) {
+			pm8xxx_irq_exit(pmic->irq_chip);
+			pmic->irq_chip = NULL;
+		}
 	}
 
 	return 0;
diff --git a/drivers/mfd/rts5249.c b/drivers/mfd/rts5249.c
index 3b835f593e35..573de7bfcced 100644
--- a/drivers/mfd/rts5249.c
+++ b/drivers/mfd/rts5249.c
@@ -130,13 +130,57 @@ static int rts5249_optimize_phy(struct rtsx_pcr *pcr)
 {
 	int err;
 
-	err = rtsx_pci_write_phy_register(pcr, PHY_REG_REV, 0xFE46);
+	err = rtsx_pci_write_phy_register(pcr, PHY_REG_REV,
+			PHY_REG_REV_RESV | PHY_REG_REV_RXIDLE_LATCHED |
+			PHY_REG_REV_P1_EN | PHY_REG_REV_RXIDLE_EN |
+			PHY_REG_REV_RX_PWST | PHY_REG_REV_CLKREQ_DLY_TIMER_1_0 |
+			PHY_REG_REV_STOP_CLKRD | PHY_REG_REV_STOP_CLKWR);
 	if (err < 0)
 		return err;
 
 	msleep(1);
 
-	return rtsx_pci_write_phy_register(pcr, PHY_BPCR, 0x05C0);
+	err = rtsx_pci_write_phy_register(pcr, PHY_BPCR,
+			PHY_BPCR_IBRXSEL | PHY_BPCR_IBTXSEL |
+			PHY_BPCR_IB_FILTER | PHY_BPCR_CMIRROR_EN);
+	if (err < 0)
+		return err;
+	err = rtsx_pci_write_phy_register(pcr, PHY_PCR,
+			PHY_PCR_FORCE_CODE | PHY_PCR_OOBS_CALI_50 |
+			PHY_PCR_OOBS_VCM_08 | PHY_PCR_OOBS_SEN_90 |
+			PHY_PCR_RSSI_EN);
+	if (err < 0)
+		return err;
+	err = rtsx_pci_write_phy_register(pcr, PHY_RCR2,
+			PHY_RCR2_EMPHASE_EN | PHY_RCR2_NADJR |
+			PHY_RCR2_CDR_CP_10 | PHY_RCR2_CDR_SR_2 |
+			PHY_RCR2_FREQSEL_12 | PHY_RCR2_CPADJEN |
+			PHY_RCR2_CDR_SC_8 | PHY_RCR2_CALIB_LATE);
+	if (err < 0)
+		return err;
+	err = rtsx_pci_write_phy_register(pcr, PHY_FLD4,
+			PHY_FLD4_FLDEN_SEL | PHY_FLD4_REQ_REF |
+			PHY_FLD4_RXAMP_OFF | PHY_FLD4_REQ_ADDA |
+			PHY_FLD4_BER_COUNT | PHY_FLD4_BER_TIMER |
+			PHY_FLD4_BER_CHK_EN);
+	if (err < 0)
+		return err;
+	err = rtsx_pci_write_phy_register(pcr, PHY_RDR, PHY_RDR_RXDSEL_1_9);
+	if (err < 0)
+		return err;
+	err = rtsx_pci_write_phy_register(pcr, PHY_RCR1,
+			PHY_RCR1_ADP_TIME | PHY_RCR1_VCO_COARSE);
+	if (err < 0)
+		return err;
+	err = rtsx_pci_write_phy_register(pcr, PHY_FLD3,
+			PHY_FLD3_TIMER_4 | PHY_FLD3_TIMER_6 |
+			PHY_FLD3_RXDELINK);
+	if (err < 0)
+		return err;
+	return rtsx_pci_write_phy_register(pcr, PHY_TUNE,
+			PHY_TUNE_TUNEREF_1_0 | PHY_TUNE_VBGSEL_1252 |
+			PHY_TUNE_SDBUS_33 | PHY_TUNE_TUNED18 |
+			PHY_TUNE_TUNED12);
 }
 
 static int rts5249_turn_on_led(struct rtsx_pcr *pcr)
diff --git a/drivers/mfd/rtsx_pcr.c b/drivers/mfd/rtsx_pcr.c
index e6ae7720f9e1..11e20afbdcac 100644
--- a/drivers/mfd/rtsx_pcr.c
+++ b/drivers/mfd/rtsx_pcr.c
@@ -1149,7 +1149,7 @@ static int rtsx_pci_probe(struct pci_dev *pcidev,
 	pcr->remap_addr = ioremap_nocache(base, len);
 	if (!pcr->remap_addr) {
 		ret = -ENOMEM;
-		goto free_host;
+		goto free_handle;
 	}
 
 	pcr->rtsx_resv_buf = dma_alloc_coherent(&(pcidev->dev),
@@ -1209,8 +1209,6 @@ disable_msi:
 			pcr->rtsx_resv_buf, pcr->rtsx_resv_buf_addr);
 unmap:
 	iounmap(pcr->remap_addr);
-free_host:
-	dev_set_drvdata(&pcidev->dev, NULL);
 free_handle:
 	kfree(handle);
 free_pcr:
@@ -1242,7 +1240,6 @@ static void rtsx_pci_remove(struct pci_dev *pcidev)
 		pci_disable_msi(pcr->pci);
 	iounmap(pcr->remap_addr);
 
-	dev_set_drvdata(&pcidev->dev, NULL);
 	pci_release_regions(pcidev);
 	pci_disable_device(pcidev);
 
diff --git a/drivers/mfd/sec-core.c b/drivers/mfd/sec-core.c
index f530e4b73f19..34c18fb8c089 100644
--- a/drivers/mfd/sec-core.c
+++ b/drivers/mfd/sec-core.c
@@ -17,6 +17,7 @@
 #include <linux/err.h>
 #include <linux/slab.h>
 #include <linux/i2c.h>
+#include <linux/of.h>
 #include <linux/of_irq.h>
 #include <linux/interrupt.h>
 #include <linux/pm_runtime.h>
diff --git a/drivers/mfd/sm501.c b/drivers/mfd/sm501.c
index 33f040c558d0..c2c8c91c6c7b 100644
--- a/drivers/mfd/sm501.c
+++ b/drivers/mfd/sm501.c
@@ -1232,7 +1232,7 @@ static ssize_t sm501_dbg_regs(struct device *dev,
 }
 
 
-static DEVICE_ATTR(dbg_regs, 0666, sm501_dbg_regs, NULL);
+static DEVICE_ATTR(dbg_regs, 0444, sm501_dbg_regs, NULL);
 
 /* sm501_init_reg
  *
@@ -1660,7 +1660,6 @@ static int sm501_pci_probe(struct pci_dev *dev,
  err3:
 	pci_disable_device(dev);
  err2:
-	pci_set_drvdata(dev, NULL);
 	kfree(sm);
  err1:
 	return err;
@@ -1695,7 +1694,6 @@ static void sm501_pci_remove(struct pci_dev *dev)
 	release_resource(sm->regs_claim);
 	kfree(sm->regs_claim);
 
-	pci_set_drvdata(dev, NULL);
 	pci_disable_device(dev);
 }
 
diff --git a/drivers/mfd/stw481x.c b/drivers/mfd/stw481x.c
new file mode 100644
index 000000000000..1243d5c6a448
--- /dev/null
+++ b/drivers/mfd/stw481x.c
@@ -0,0 +1,250 @@
+/*
+ * Core driver for STw4810/STw4811
+ *
+ * Copyright (C) 2013 ST-Ericsson SA
+ * Written on behalf of Linaro for ST-Ericsson
+ *
+ * Author: Linus Walleij <linus.walleij@linaro.org>
+ *
+ * License terms: GNU General Public License (GPL) version 2
+ */
+
+#include <linux/err.h>
+#include <linux/i2c.h>
+#include <linux/init.h>
+#include <linux/mfd/core.h>
+#include <linux/mfd/stw481x.h>
+#include <linux/module.h>
+#include <linux/regmap.h>
+#include <linux/spinlock.h>
+#include <linux/slab.h>
+
+/*
+ * This driver can only access the non-USB portions of STw4811, the register
+ * range 0x00-0x10 dealing with USB is bound to the two special I2C pins used
+ * for USB control.
+ */
+
+/* Registers inside the power control address space */
+#define STW_PC_VCORE_SEL	0x05U
+#define STW_PC_VAUX_SEL		0x06U
+#define STW_PC_VPLL_SEL		0x07U
+
+/**
+ * stw481x_get_pctl_reg() - get a power control register
+ * @stw481x: handle to the stw481x chip
+ * @reg: power control register to fetch
+ *
+ * The power control registers is a set of one-time-programmable registers
+ * in its own register space, accessed by writing addess bits to these
+ * two registers: bits 7,6,5 of PCTL_REG_LO corresponds to the 3 LSBs of
+ * the address and bits 8,9 of PCTL_REG_HI corresponds to the 2 MSBs of
+ * the address, forming an address space of 5 bits, i.e. 32 registers
+ * 0x00 ... 0x1f can be obtained.
+ */
+static int stw481x_get_pctl_reg(struct stw481x *stw481x, u8 reg)
+{
+	u8 msb = (reg >> 3) & 0x03;
+	u8 lsb = (reg << 5) & 0xe0;
+	unsigned int val;
+	u8 vrfy;
+	int ret;
+
+	ret = regmap_write(stw481x->map, STW_PCTL_REG_HI, msb);
+	if (ret)
+		return ret;
+	ret = regmap_write(stw481x->map, STW_PCTL_REG_LO, lsb);
+	if (ret)
+		return ret;
+	ret = regmap_read(stw481x->map, STW_PCTL_REG_HI, &val);
+	if (ret)
+		return ret;
+	vrfy = (val & 0x03) << 3;
+	ret = regmap_read(stw481x->map, STW_PCTL_REG_LO, &val);
+	if (ret)
+		return ret;
+	vrfy |= ((val >> 5) & 0x07);
+	if (vrfy != reg)
+		return -EIO;
+	return (val >> 1) & 0x0f;
+}
+
+static int stw481x_startup(struct stw481x *stw481x)
+{
+	/* Voltages multiplied by 100 */
+	u8 vcore_val[] = { 100, 105, 110, 115, 120, 122, 124, 126, 128,
+			   130, 132, 134, 136, 138, 140, 145 };
+	u8 vpll_val[] = { 105, 120, 130, 180 };
+	u8 vaux_val[] = { 15, 18, 25, 28 };
+	u8 vcore;
+	u8 vcore_slp;
+	u8 vpll;
+	u8 vaux;
+	bool vaux_en;
+	bool it_warn;
+	int ret;
+	unsigned int val;
+
+	ret = regmap_read(stw481x->map, STW_CONF1, &val);
+	if (ret)
+		return ret;
+	vaux_en = !!(val & STW_CONF1_PDN_VAUX);
+	it_warn = !!(val & STW_CONF1_IT_WARN);
+
+	dev_info(&stw481x->client->dev, "voltages %s\n",
+		(val & STW_CONF1_V_MONITORING) ? "OK" : "LOW");
+	dev_info(&stw481x->client->dev, "MMC level shifter %s\n",
+		(val & STW_CONF1_MMC_LS_STATUS) ? "high impedance" : "ON");
+	dev_info(&stw481x->client->dev, "VMMC: %s\n",
+		(val & STW_CONF1_PDN_VMMC) ? "ON" : "disabled");
+
+	dev_info(&stw481x->client->dev, "STw481x power control registers:\n");
+
+	ret = stw481x_get_pctl_reg(stw481x, STW_PC_VCORE_SEL);
+	if (ret < 0)
+		return ret;
+	vcore = ret & 0x0f;
+
+	ret = stw481x_get_pctl_reg(stw481x, STW_PC_VAUX_SEL);
+	if (ret < 0)
+		return ret;
+	vaux = (ret >> 2) & 3;
+	vpll = (ret >> 4) & 1; /* Save bit 4 */
+
+	ret = stw481x_get_pctl_reg(stw481x, STW_PC_VPLL_SEL);
+	if (ret < 0)
+		return ret;
+	vpll |= (ret >> 1) & 2;
+
+	dev_info(&stw481x->client->dev, "VCORE: %u.%uV %s\n",
+		vcore_val[vcore] / 100, vcore_val[vcore] % 100,
+		(ret & 4) ? "ON" : "OFF");
+
+	dev_info(&stw481x->client->dev, "VPLL:  %u.%uV %s\n",
+		vpll_val[vpll] / 100, vpll_val[vpll] % 100,
+		(ret & 0x10) ? "ON" : "OFF");
+
+	dev_info(&stw481x->client->dev, "VAUX:  %u.%uV %s\n",
+		vaux_val[vaux] / 10, vaux_val[vaux] % 10,
+		vaux_en ? "ON" : "OFF");
+
+	ret = regmap_read(stw481x->map, STW_CONF2, &val);
+	if (ret)
+		return ret;
+
+	dev_info(&stw481x->client->dev, "TWARN: %s threshold, %s\n",
+		it_warn ? "below" : "above",
+		(val & STW_CONF2_MASK_TWARN) ?
+		 "enabled" : "mask through VDDOK");
+	dev_info(&stw481x->client->dev, "VMMC: %s\n",
+		(val & STW_CONF2_VMMC_EXT) ? "internal" : "external");
+	dev_info(&stw481x->client->dev, "IT WAKE UP: %s\n",
+		(val & STW_CONF2_MASK_IT_WAKE_UP) ? "enabled" : "masked");
+	dev_info(&stw481x->client->dev, "GPO1: %s\n",
+		(val & STW_CONF2_GPO1) ? "low" : "high impedance");
+	dev_info(&stw481x->client->dev, "GPO2: %s\n",
+		(val & STW_CONF2_GPO2) ? "low" : "high impedance");
+
+	ret = regmap_read(stw481x->map, STW_VCORE_SLEEP, &val);
+	if (ret)
+		return ret;
+	vcore_slp = val & 0x0f;
+	dev_info(&stw481x->client->dev, "VCORE SLEEP: %u.%uV\n",
+		vcore_val[vcore_slp] / 100, vcore_val[vcore_slp] % 100);
+
+	return 0;
+}
+
+/*
+ * MFD cells - we have one cell which is selected operation
+ * mode, and we always have a GPIO cell.
+ */
+static struct mfd_cell stw481x_cells[] = {
+	{
+		.of_compatible = "st,stw481x-vmmc",
+		.name = "stw481x-vmmc-regulator",
+		.id = -1,
+	},
+};
+
+const struct regmap_config stw481x_regmap_config = {
+	.reg_bits = 8,
+	.val_bits = 8,
+};
+
+static int stw481x_probe(struct i2c_client *client,
+			 const struct i2c_device_id *id)
+{
+	struct stw481x			*stw481x;
+	int ret;
+	int i;
+
+	stw481x = devm_kzalloc(&client->dev, sizeof(*stw481x), GFP_KERNEL);
+	if (!stw481x)
+		return -ENOMEM;
+
+	i2c_set_clientdata(client, stw481x);
+	stw481x->client = client;
+	stw481x->map = devm_regmap_init_i2c(client, &stw481x_regmap_config);
+
+	ret = stw481x_startup(stw481x);
+	if (ret) {
+		dev_err(&client->dev, "chip initialization failed\n");
+		return ret;
+	}
+
+	/* Set up and register the platform devices. */
+	for (i = 0; i < ARRAY_SIZE(stw481x_cells); i++) {
+		/* One state holder for all drivers, this is simple */
+		stw481x_cells[i].platform_data = stw481x;
+		stw481x_cells[i].pdata_size = sizeof(*stw481x);
+	}
+
+	ret = mfd_add_devices(&client->dev, 0, stw481x_cells,
+			ARRAY_SIZE(stw481x_cells), NULL, 0, NULL);
+	if (ret)
+		return ret;
+
+	dev_info(&client->dev, "initialized STw481x device\n");
+
+	return ret;
+}
+
+static int stw481x_remove(struct i2c_client *client)
+{
+	mfd_remove_devices(&client->dev);
+	return 0;
+}
+
+/*
+ * This ID table is completely unused, as this is a pure
+ * device-tree probed driver, but it has to be here due to
+ * the structure of the I2C core.
+ */
+static const struct i2c_device_id stw481x_id[] = {
+	{ "stw481x", 0 },
+	{ },
+};
+
+static const struct of_device_id stw481x_match[] = {
+	{ .compatible = "st,stw4810", },
+	{ .compatible = "st,stw4811", },
+	{ },
+};
+MODULE_DEVICE_TABLE(of, stw481x_match);
+
+static struct i2c_driver stw481x_driver = {
+	.driver = {
+		.name	= "stw481x",
+		.of_match_table = stw481x_match,
+	},
+	.probe		= stw481x_probe,
+	.remove		= stw481x_remove,
+	.id_table	= stw481x_id,
+};
+
+module_i2c_driver(stw481x_driver);
+
+MODULE_AUTHOR("Linus Walleij");
+MODULE_DESCRIPTION("STw481x PMIC driver");
+MODULE_LICENSE("GPL v2");
diff --git a/drivers/mfd/tc3589x.c b/drivers/mfd/tc3589x.c
index 70f4909fee13..87ea51dc6234 100644
--- a/drivers/mfd/tc3589x.c
+++ b/drivers/mfd/tc3589x.c
@@ -16,6 +16,19 @@
 #include <linux/mfd/core.h>
 #include <linux/mfd/tc3589x.h>
 
+/**
+ * enum tc3589x_version - indicates the TC3589x version
+ */
+enum tc3589x_version {
+	TC3589X_TC35890,
+	TC3589X_TC35892,
+	TC3589X_TC35893,
+	TC3589X_TC35894,
+	TC3589X_TC35895,
+	TC3589X_TC35896,
+	TC3589X_UNKNOWN,
+};
+
 #define TC3589x_CLKMODE_MODCTL_SLEEP		0x0
 #define TC3589x_CLKMODE_MODCTL_OPERATION	(1 << 0)
 
@@ -361,7 +374,21 @@ static int tc3589x_probe(struct i2c_client *i2c,
 	tc3589x->i2c = i2c;
 	tc3589x->pdata = pdata;
 	tc3589x->irq_base = pdata->irq_base;
-	tc3589x->num_gpio = id->driver_data;
+
+	switch (id->driver_data) {
+	case TC3589X_TC35893:
+	case TC3589X_TC35895:
+	case TC3589X_TC35896:
+		tc3589x->num_gpio = 20;
+		break;
+	case TC3589X_TC35890:
+	case TC3589X_TC35892:
+	case TC3589X_TC35894:
+	case TC3589X_UNKNOWN:
+	default:
+		tc3589x->num_gpio = 24;
+		break;
+	}
 
 	i2c_set_clientdata(i2c, tc3589x);
 
@@ -432,7 +459,13 @@ static int tc3589x_resume(struct device *dev)
 static SIMPLE_DEV_PM_OPS(tc3589x_dev_pm_ops, tc3589x_suspend, tc3589x_resume);
 
 static const struct i2c_device_id tc3589x_id[] = {
-	{ "tc3589x", 24 },
+	{ "tc35890", TC3589X_TC35890 },
+	{ "tc35892", TC3589X_TC35892 },
+	{ "tc35893", TC3589X_TC35893 },
+	{ "tc35894", TC3589X_TC35894 },
+	{ "tc35895", TC3589X_TC35895 },
+	{ "tc35896", TC3589X_TC35896 },
+	{ "tc3589x", TC3589X_UNKNOWN },
 	{ }
 };
 MODULE_DEVICE_TABLE(i2c, tc3589x_id);
diff --git a/drivers/mfd/ti-ssp.c b/drivers/mfd/ti-ssp.c
index 1c2b994e1f6c..71e3e0c5bf73 100644
--- a/drivers/mfd/ti-ssp.c
+++ b/drivers/mfd/ti-ssp.c
@@ -445,7 +445,6 @@ static int ti_ssp_remove(struct platform_device *pdev)
 	iounmap(ssp->regs);
 	release_mem_region(ssp->res->start, resource_size(ssp->res));
 	kfree(ssp);
-	dev_set_drvdata(dev, NULL);
 	return 0;
 }
 
diff --git a/drivers/mfd/ti_am335x_tscadc.c b/drivers/mfd/ti_am335x_tscadc.c
index baaf5a8123bb..88718abfb9ba 100644
--- a/drivers/mfd/ti_am335x_tscadc.c
+++ b/drivers/mfd/ti_am335x_tscadc.c
@@ -56,21 +56,25 @@ EXPORT_SYMBOL_GPL(am335x_tsc_se_update);
 
 void am335x_tsc_se_set(struct ti_tscadc_dev *tsadc, u32 val)
 {
-	spin_lock(&tsadc->reg_lock);
+	unsigned long flags;
+
+	spin_lock_irqsave(&tsadc->reg_lock, flags);
 	tsadc->reg_se_cache = tscadc_readl(tsadc, REG_SE);
 	tsadc->reg_se_cache |= val;
 	am335x_tsc_se_update(tsadc);
-	spin_unlock(&tsadc->reg_lock);
+	spin_unlock_irqrestore(&tsadc->reg_lock, flags);
 }
 EXPORT_SYMBOL_GPL(am335x_tsc_se_set);
 
 void am335x_tsc_se_clr(struct ti_tscadc_dev *tsadc, u32 val)
 {
-	spin_lock(&tsadc->reg_lock);
+	unsigned long flags;
+
+	spin_lock_irqsave(&tsadc->reg_lock, flags);
 	tsadc->reg_se_cache = tscadc_readl(tsadc, REG_SE);
 	tsadc->reg_se_cache &= ~val;
 	am335x_tsc_se_update(tsadc);
-	spin_unlock(&tsadc->reg_lock);
+	spin_unlock_irqrestore(&tsadc->reg_lock, flags);
 }
 EXPORT_SYMBOL_GPL(am335x_tsc_se_clr);
 
@@ -95,7 +99,7 @@ static	int ti_tscadc_probe(struct platform_device *pdev)
 	const __be32            *cur;
 	u32			val;
 	int			err, ctrl;
-	int			clk_value, clock_rate;
+	int			clock_rate;
 	int			tsc_wires = 0, adc_channels = 0, total_channels;
 	int			readouts = 0;
 
@@ -196,11 +200,11 @@ static	int ti_tscadc_probe(struct platform_device *pdev)
 	}
 	clock_rate = clk_get_rate(clk);
 	clk_put(clk);
-	clk_value = clock_rate / ADC_CLK;
+	tscadc->clk_div = clock_rate / ADC_CLK;
 
 	/* TSCADC_CLKDIV needs to be configured to the value minus 1 */
-	clk_value = clk_value - 1;
-	tscadc_writel(tscadc, REG_CLKDIV, clk_value);
+	tscadc->clk_div--;
+	tscadc_writel(tscadc, REG_CLKDIV, tscadc->clk_div);
 
 	/* Set the control register bits */
 	ctrl = CNTRLREG_STEPCONFIGWRT |
@@ -303,6 +307,8 @@ static int tscadc_resume(struct device *dev)
 	tscadc_writel(tscadc_dev, REG_CTRL,
 			(restore | CNTRLREG_TSCSSENB));
 
+	tscadc_writel(tscadc_dev, REG_CLKDIV, tscadc_dev->clk_div);
+
 	return 0;
 }
 
@@ -326,7 +332,7 @@ static struct platform_driver ti_tscadc_driver = {
 		.name   = "ti_am3359-tscadc",
 		.owner	= THIS_MODULE,
 		.pm	= TSCADC_PM_OPS,
-		.of_match_table = of_match_ptr(ti_tscadc_dt_ids),
+		.of_match_table = ti_tscadc_dt_ids,
 	},
 	.probe	= ti_tscadc_probe,
 	.remove	= ti_tscadc_remove,
diff --git a/drivers/mfd/timberdale.c b/drivers/mfd/timberdale.c
index a6755ec7bd6a..dbb34f94e5e3 100644
--- a/drivers/mfd/timberdale.c
+++ b/drivers/mfd/timberdale.c
@@ -678,7 +678,7 @@ static int timb_probe(struct pci_dev *dev,
 	priv->ctl_mapbase = mapbase + CHIPCTLOFFSET;
 	if (!request_mem_region(priv->ctl_mapbase, CHIPCTLSIZE, "timb-ctl")) {
 		dev_err(&dev->dev, "Failed to request ctl mem\n");
-		goto err_request;
+		goto err_start;
 	}
 
 	priv->ctl_membase = ioremap(priv->ctl_mapbase, CHIPCTLSIZE);
@@ -828,13 +828,10 @@ err_config:
 	iounmap(priv->ctl_membase);
 err_ioremap:
 	release_mem_region(priv->ctl_mapbase, CHIPCTLSIZE);
-err_request:
-	pci_set_drvdata(dev, NULL);
 err_start:
 	pci_disable_device(dev);
 err_enable:
 	kfree(priv);
-	pci_set_drvdata(dev, NULL);
 	return -ENODEV;
 }
 
@@ -851,7 +848,6 @@ static void timb_remove(struct pci_dev *dev)
 
 	pci_disable_msix(dev);
 	pci_disable_device(dev);
-	pci_set_drvdata(dev, NULL);
 	kfree(priv);
 }
 
diff --git a/drivers/mfd/tps6507x.c b/drivers/mfd/tps6507x.c
index 5ad4b772b097..a081b925d10b 100644
--- a/drivers/mfd/tps6507x.c
+++ b/drivers/mfd/tps6507x.c
@@ -19,6 +19,7 @@
 #include <linux/init.h>
 #include <linux/slab.h>
 #include <linux/i2c.h>
+#include <linux/of.h>
 #include <linux/of_device.h>
 #include <linux/mfd/core.h>
 #include <linux/mfd/tps6507x.h>
diff --git a/drivers/mfd/tps65217.c b/drivers/mfd/tps65217.c
index b8f48647661e..b7be0b295575 100644
--- a/drivers/mfd/tps65217.c
+++ b/drivers/mfd/tps65217.c
@@ -245,7 +245,7 @@ static struct i2c_driver tps65217_driver = {
 	.driver		= {
 		.name	= "tps65217",
 		.owner	= THIS_MODULE,
-		.of_match_table = of_match_ptr(tps65217_of_match),
+		.of_match_table = tps65217_of_match,
 	},
 	.id_table	= tps65217_id_table,
 	.probe		= tps65217_probe,
diff --git a/drivers/mfd/tps6586x.c b/drivers/mfd/tps6586x.c
index f54fe4d4f77b..ee61fd7c198d 100644
--- a/drivers/mfd/tps6586x.c
+++ b/drivers/mfd/tps6586x.c
@@ -26,6 +26,7 @@
 #include <linux/i2c.h>
 #include <linux/platform_device.h>
 #include <linux/regmap.h>
+#include <linux/of.h>
 
 #include <linux/mfd/core.h>
 #include <linux/mfd/tps6586x.h>
@@ -124,6 +125,7 @@ struct tps6586x {
 	struct i2c_client	*client;
 	struct regmap		*regmap;
 
+	int			irq;
 	struct irq_chip		irq_chip;
 	struct mutex		irq_lock;
 	int			irq_base;
@@ -261,12 +263,23 @@ static void tps6586x_irq_sync_unlock(struct irq_data *data)
 	mutex_unlock(&tps6586x->irq_lock);
 }
 
+#ifdef CONFIG_PM_SLEEP
+static int tps6586x_irq_set_wake(struct irq_data *irq_data, unsigned int on)
+{
+	struct tps6586x *tps6586x = irq_data_get_irq_chip_data(irq_data);
+	return irq_set_irq_wake(tps6586x->irq, on);
+}
+#else
+#define tps6586x_irq_set_wake NULL
+#endif
+
 static struct irq_chip tps6586x_irq_chip = {
 	.name = "tps6586x",
 	.irq_bus_lock = tps6586x_irq_lock,
 	.irq_bus_sync_unlock = tps6586x_irq_sync_unlock,
 	.irq_disable = tps6586x_irq_disable,
 	.irq_enable = tps6586x_irq_enable,
+	.irq_set_wake = tps6586x_irq_set_wake,
 };
 
 static int tps6586x_irq_map(struct irq_domain *h, unsigned int virq,
@@ -331,6 +344,8 @@ static int tps6586x_irq_init(struct tps6586x *tps6586x, int irq,
 	int new_irq_base;
 	int irq_num = ARRAY_SIZE(tps6586x_irqs);
 
+	tps6586x->irq = irq;
+
 	mutex_init(&tps6586x->irq_lock);
 	for (i = 0; i < 5; i++) {
 		tps6586x->mask_reg[i] = 0xff;
@@ -360,10 +375,8 @@ static int tps6586x_irq_init(struct tps6586x *tps6586x, int irq,
 	ret = request_threaded_irq(irq, NULL, tps6586x_irq, IRQF_ONESHOT,
 				   "tps6586x", tps6586x);
 
-	if (!ret) {
+	if (!ret)
 		device_init_wakeup(tps6586x->dev, 1);
-		enable_irq_wake(irq);
-	}
 
 	return ret;
 }
diff --git a/drivers/mfd/tps65910.c b/drivers/mfd/tps65910.c
index d79277204835..c0f608e3ca9e 100644
--- a/drivers/mfd/tps65910.c
+++ b/drivers/mfd/tps65910.c
@@ -25,6 +25,7 @@
 #include <linux/mfd/core.h>
 #include <linux/regmap.h>
 #include <linux/mfd/tps65910.h>
+#include <linux/of.h>
 #include <linux/of_device.h>
 
 static struct resource rtc_resources[] = {
@@ -410,14 +411,10 @@ static struct tps65910_board *tps65910_parse_dt(struct i2c_client *client,
 	ret = of_property_read_u32(np, "ti,vmbch-threshold", &prop);
 	if (!ret)
 		board_info->vmbch_threshold = prop;
-	else if (*chip_id == TPS65911)
-		dev_warn(&client->dev, "VMBCH-Threshold not specified");
 
 	ret = of_property_read_u32(np, "ti,vmbch2-threshold", &prop);
 	if (!ret)
 		board_info->vmbch2_threshold = prop;
-	else if (*chip_id == TPS65911)
-		dev_warn(&client->dev, "VMBCH2-Threshold not specified");
 
 	prop = of_property_read_bool(np, "ti,en-ck32k-xtal");
 	board_info->en_ck32k_xtal = prop;
diff --git a/drivers/mfd/twl6040.c b/drivers/mfd/twl6040.c
index daf66942071c..0779d5ab9ab1 100644
--- a/drivers/mfd/twl6040.c
+++ b/drivers/mfd/twl6040.c
@@ -565,13 +565,13 @@ static int twl6040_probe(struct i2c_client *client,
 				      twl6040->supplies);
 	if (ret != 0) {
 		dev_err(&client->dev, "Failed to get supplies: %d\n", ret);
-		goto regulator_get_err;
+		return ret;
 	}
 
 	ret = regulator_bulk_enable(TWL6040_NUM_SUPPLIES, twl6040->supplies);
 	if (ret != 0) {
 		dev_err(&client->dev, "Failed to enable supplies: %d\n", ret);
-		goto regulator_get_err;
+		return ret;
 	}
 
 	twl6040->dev = &client->dev;
@@ -619,7 +619,7 @@ static int twl6040_probe(struct i2c_client *client,
 					"twl6040_irq_th", twl6040);
 	if (ret) {
 		dev_err(twl6040->dev, "Thermal IRQ request failed: %d\n", ret);
-		goto thirq_err;
+		goto readyirq_err;
 	}
 
 	/* dual-access registers controlled by I2C only */
@@ -659,21 +659,14 @@ static int twl6040_probe(struct i2c_client *client,
 	ret = mfd_add_devices(&client->dev, -1, twl6040->cells, children,
 			      NULL, 0, NULL);
 	if (ret)
-		goto mfd_err;
+		goto readyirq_err;
 
 	return 0;
 
-mfd_err:
-	devm_free_irq(&client->dev, twl6040->irq_th, twl6040);
-thirq_err:
-	devm_free_irq(&client->dev, twl6040->irq_ready, twl6040);
 readyirq_err:
 	regmap_del_irq_chip(twl6040->irq, twl6040->irq_data);
 gpio_err:
 	regulator_bulk_disable(TWL6040_NUM_SUPPLIES, twl6040->supplies);
-regulator_get_err:
-	i2c_set_clientdata(client, NULL);
-
 	return ret;
 }
 
@@ -684,12 +677,9 @@ static int twl6040_remove(struct i2c_client *client)
 	if (twl6040->power_count)
 		twl6040_power(twl6040, 0);
 
-	devm_free_irq(&client->dev, twl6040->irq_ready, twl6040);
-	devm_free_irq(&client->dev, twl6040->irq_th, twl6040);
 	regmap_del_irq_chip(twl6040->irq, twl6040->irq_data);
 
 	mfd_remove_devices(&client->dev);
-	i2c_set_clientdata(client, NULL);
 
 	regulator_bulk_disable(TWL6040_NUM_SUPPLIES, twl6040->supplies);
 
diff --git a/drivers/mfd/ucb1x00-core.c b/drivers/mfd/ucb1x00-core.c
index d5966e6b5a7d..0313f839e8fa 100644
--- a/drivers/mfd/ucb1x00-core.c
+++ b/drivers/mfd/ucb1x00-core.c
@@ -553,6 +553,7 @@ static int ucb1x00_probe(struct mcp *mcp)
 	if (ucb->irq_base < 0) {
 		dev_err(&ucb->dev, "unable to allocate 16 irqs: %d\n",
 			ucb->irq_base);
+		ret = ucb->irq_base;
 		goto err_irq_alloc;
 	}
 
diff --git a/drivers/mfd/wm5102-tables.c b/drivers/mfd/wm5102-tables.c
index 802dd3cb18cf..1e9a4b2102f9 100644
--- a/drivers/mfd/wm5102-tables.c
+++ b/drivers/mfd/wm5102-tables.c
@@ -903,7 +903,6 @@ static const struct reg_default wm5102_reg_default[] = {
 	{ 0x00000D1B, 0xFFFF },   /* R3355  - IRQ2 Status 4 Mask */ 
 	{ 0x00000D1C, 0xFFFF },   /* R3356  - IRQ2 Status 5 Mask */ 
 	{ 0x00000D1F, 0x0000 },   /* R3359  - IRQ2 Control */ 
-	{ 0x00000D50, 0x0000 },   /* R3408  - AOD wkup and trig */
 	{ 0x00000D53, 0xFFFF },   /* R3411  - AOD IRQ Mask IRQ1 */ 
 	{ 0x00000D54, 0xFFFF },   /* R3412  - AOD IRQ Mask IRQ2 */ 
 	{ 0x00000D56, 0x0000 },   /* R3414  - Jack detect debounce */ 
diff --git a/drivers/mfd/wm5110-tables.c b/drivers/mfd/wm5110-tables.c
index 3113e39b318e..bf8b3b5ad1fe 100644
--- a/drivers/mfd/wm5110-tables.c
+++ b/drivers/mfd/wm5110-tables.c
@@ -243,6 +243,12 @@ int wm5110_patch(struct arizona *arizona)
 EXPORT_SYMBOL_GPL(wm5110_patch);
 
 static const struct regmap_irq wm5110_aod_irqs[ARIZONA_NUM_IRQ] = {
+	[ARIZONA_IRQ_MICD_CLAMP_FALL] = {
+		.mask = ARIZONA_MICD_CLAMP_FALL_EINT1
+	},
+	[ARIZONA_IRQ_MICD_CLAMP_RISE] = {
+		.mask = ARIZONA_MICD_CLAMP_RISE_EINT1
+	},
 	[ARIZONA_IRQ_GP5_FALL] = { .mask = ARIZONA_GP5_FALL_EINT1 },
 	[ARIZONA_IRQ_GP5_RISE] = { .mask = ARIZONA_GP5_RISE_EINT1 },
 	[ARIZONA_IRQ_JD_FALL] = { .mask = ARIZONA_JD1_FALL_EINT1 },
@@ -505,6 +511,7 @@ static const struct reg_default wm5110_reg_default[] = {
 	{ 0x00000293, 0x0000 },    /* R659   - Accessory Detect Mode 1 */
 	{ 0x0000029B, 0x0020 },    /* R667   - Headphone Detect 1 */
 	{ 0x0000029C, 0x0000 },    /* R668   - Headphone Detect 2 */
+	{ 0x000002A2, 0x0000 },    /* R674   - Micd clamp control */
 	{ 0x000002A3, 0x1102 },    /* R675   - Mic Detect 1 */
 	{ 0x000002A4, 0x009F },    /* R676   - Mic Detect 2 */
 	{ 0x000002A5, 0x0000 },    /* R677   - Mic Detect 3 */
@@ -592,7 +599,7 @@ static const struct reg_default wm5110_reg_default[] = {
 	{ 0x0000043E, 0x0080 },    /* R1086  - DAC Volume Limit 6R */
 	{ 0x0000043F, 0x0800 },    /* R1087  - Noise Gate Select 6R */
 	{ 0x00000450, 0x0000 },    /* R1104  - DAC AEC Control 1 */
-	{ 0x00000458, 0x0001 },    /* R1112  - Noise Gate Control */
+	{ 0x00000458, 0x0000 },    /* R1112  - Noise Gate Control */
 	{ 0x00000480, 0x0040 },    /* R1152  - Class W ANC Threshold 1 */
 	{ 0x00000481, 0x0040 },    /* R1153  - Class W ANC Threshold 2 */
 	{ 0x00000490, 0x0069 },    /* R1168  - PDM SPK1 CTRL 1 */
@@ -1204,7 +1211,6 @@ static const struct reg_default wm5110_reg_default[] = {
 	{ 0x00000D1B, 0xFFFF },    /* R3355  - IRQ2 Status 4 Mask */
 	{ 0x00000D1C, 0xFFFF },    /* R3356  - IRQ2 Status 5 Mask */
 	{ 0x00000D1F, 0x0000 },    /* R3359  - IRQ2 Control */
-	{ 0x00000D50, 0x0000 },    /* R3408  - AOD wkup and trig */
 	{ 0x00000D53, 0xFFFF },    /* R3411  - AOD IRQ Mask IRQ1 */
 	{ 0x00000D54, 0xFFFF },    /* R3412  - AOD IRQ Mask IRQ2 */
 	{ 0x00000D56, 0x0000 },    /* R3414  - Jack detect debounce */
@@ -1440,6 +1446,7 @@ static bool wm5110_readable_register(struct device *dev, unsigned int reg)
 	case ARIZONA_ACCESSORY_DETECT_MODE_1:
 	case ARIZONA_HEADPHONE_DETECT_1:
 	case ARIZONA_HEADPHONE_DETECT_2:
+	case ARIZONA_MICD_CLAMP_CONTROL:
 	case ARIZONA_MIC_DETECT_1:
 	case ARIZONA_MIC_DETECT_2:
 	case ARIZONA_MIC_DETECT_3:
@@ -2291,21 +2298,37 @@ static bool wm5110_readable_register(struct device *dev, unsigned int reg)
 	case ARIZONA_DSP1_STATUS_1:
 	case ARIZONA_DSP1_STATUS_2:
 	case ARIZONA_DSP1_STATUS_3:
+	case ARIZONA_DSP1_SCRATCH_0:
+	case ARIZONA_DSP1_SCRATCH_1:
+	case ARIZONA_DSP1_SCRATCH_2:
+	case ARIZONA_DSP1_SCRATCH_3:
 	case ARIZONA_DSP2_CONTROL_1:
 	case ARIZONA_DSP2_CLOCKING_1:
 	case ARIZONA_DSP2_STATUS_1:
 	case ARIZONA_DSP2_STATUS_2:
 	case ARIZONA_DSP2_STATUS_3:
+	case ARIZONA_DSP2_SCRATCH_0:
+	case ARIZONA_DSP2_SCRATCH_1:
+	case ARIZONA_DSP2_SCRATCH_2:
+	case ARIZONA_DSP2_SCRATCH_3:
 	case ARIZONA_DSP3_CONTROL_1:
 	case ARIZONA_DSP3_CLOCKING_1:
 	case ARIZONA_DSP3_STATUS_1:
 	case ARIZONA_DSP3_STATUS_2:
 	case ARIZONA_DSP3_STATUS_3:
+	case ARIZONA_DSP3_SCRATCH_0:
+	case ARIZONA_DSP3_SCRATCH_1:
+	case ARIZONA_DSP3_SCRATCH_2:
+	case ARIZONA_DSP3_SCRATCH_3:
 	case ARIZONA_DSP4_CONTROL_1:
 	case ARIZONA_DSP4_CLOCKING_1:
 	case ARIZONA_DSP4_STATUS_1:
 	case ARIZONA_DSP4_STATUS_2:
 	case ARIZONA_DSP4_STATUS_3:
+	case ARIZONA_DSP4_SCRATCH_0:
+	case ARIZONA_DSP4_SCRATCH_1:
+	case ARIZONA_DSP4_SCRATCH_2:
+	case ARIZONA_DSP4_SCRATCH_3:
 		return true;
 	default:
 		return false;
@@ -2347,25 +2370,41 @@ static bool wm5110_volatile_register(struct device *dev, unsigned int reg)
 	case ARIZONA_INTERRUPT_RAW_STATUS_7:
 	case ARIZONA_INTERRUPT_RAW_STATUS_8:
 	case ARIZONA_IRQ_PIN_STATUS:
+	case ARIZONA_AOD_WKUP_AND_TRIG:
 	case ARIZONA_AOD_IRQ1:
 	case ARIZONA_AOD_IRQ2:
+	case ARIZONA_AOD_IRQ_RAW_STATUS:
 	case ARIZONA_FX_CTRL2:
 	case ARIZONA_ASRC_STATUS:
 	case ARIZONA_DSP_STATUS:
-	case ARIZONA_DSP1_CONTROL_1:
-	case ARIZONA_DSP1_CLOCKING_1:
 	case ARIZONA_DSP1_STATUS_1:
 	case ARIZONA_DSP1_STATUS_2:
 	case ARIZONA_DSP1_STATUS_3:
+	case ARIZONA_DSP1_SCRATCH_0:
+	case ARIZONA_DSP1_SCRATCH_1:
+	case ARIZONA_DSP1_SCRATCH_2:
+	case ARIZONA_DSP1_SCRATCH_3:
 	case ARIZONA_DSP2_STATUS_1:
 	case ARIZONA_DSP2_STATUS_2:
 	case ARIZONA_DSP2_STATUS_3:
+	case ARIZONA_DSP2_SCRATCH_0:
+	case ARIZONA_DSP2_SCRATCH_1:
+	case ARIZONA_DSP2_SCRATCH_2:
+	case ARIZONA_DSP2_SCRATCH_3:
 	case ARIZONA_DSP3_STATUS_1:
 	case ARIZONA_DSP3_STATUS_2:
 	case ARIZONA_DSP3_STATUS_3:
+	case ARIZONA_DSP3_SCRATCH_0:
+	case ARIZONA_DSP3_SCRATCH_1:
+	case ARIZONA_DSP3_SCRATCH_2:
+	case ARIZONA_DSP3_SCRATCH_3:
 	case ARIZONA_DSP4_STATUS_1:
 	case ARIZONA_DSP4_STATUS_2:
 	case ARIZONA_DSP4_STATUS_3:
+	case ARIZONA_DSP4_SCRATCH_0:
+	case ARIZONA_DSP4_SCRATCH_1:
+	case ARIZONA_DSP4_SCRATCH_2:
+	case ARIZONA_DSP4_SCRATCH_3:
 		return true;
 	default:
 		return false;
diff --git a/drivers/mfd/wm8994-core.c b/drivers/mfd/wm8994-core.c
index e1c283e6d4e5..030827511667 100644
--- a/drivers/mfd/wm8994-core.c
+++ b/drivers/mfd/wm8994-core.c
@@ -33,84 +33,6 @@
 
 #include "wm8994.h"
 
-/**
- * wm8994_reg_read: Read a single WM8994 register.
- *
- * @wm8994: Device to read from.
- * @reg: Register to read.
- */
-int wm8994_reg_read(struct wm8994 *wm8994, unsigned short reg)
-{
-	unsigned int val;
-	int ret;
-
-	ret = regmap_read(wm8994->regmap, reg, &val);
-
-	if (ret < 0)
-		return ret;
-	else
-		return val;
-}
-EXPORT_SYMBOL_GPL(wm8994_reg_read);
-
-/**
- * wm8994_bulk_read: Read multiple WM8994 registers
- *
- * @wm8994: Device to read from
- * @reg: First register
- * @count: Number of registers
- * @buf: Buffer to fill.  The data will be returned big endian.
- */
-int wm8994_bulk_read(struct wm8994 *wm8994, unsigned short reg,
-		     int count, u16 *buf)
-{
-	return regmap_bulk_read(wm8994->regmap, reg, buf, count);
-}
-
-/**
- * wm8994_reg_write: Write a single WM8994 register.
- *
- * @wm8994: Device to write to.
- * @reg: Register to write to.
- * @val: Value to write.
- */
-int wm8994_reg_write(struct wm8994 *wm8994, unsigned short reg,
-		     unsigned short val)
-{
-	return regmap_write(wm8994->regmap, reg, val);
-}
-EXPORT_SYMBOL_GPL(wm8994_reg_write);
-
-/**
- * wm8994_bulk_write: Write multiple WM8994 registers
- *
- * @wm8994: Device to write to
- * @reg: First register
- * @count: Number of registers
- * @buf: Buffer to write from.  Data must be big-endian formatted.
- */
-int wm8994_bulk_write(struct wm8994 *wm8994, unsigned short reg,
-		      int count, const u16 *buf)
-{
-	return regmap_raw_write(wm8994->regmap, reg, buf, count * sizeof(u16));
-}
-EXPORT_SYMBOL_GPL(wm8994_bulk_write);
-
-/**
- * wm8994_set_bits: Set the value of a bitfield in a WM8994 register
- *
- * @wm8994: Device to write to.
- * @reg: Register to write to.
- * @mask: Mask of bits to set.
- * @val: Value to set (unshifted)
- */
-int wm8994_set_bits(struct wm8994 *wm8994, unsigned short reg,
-		    unsigned short mask, unsigned short val)
-{
-	return regmap_update_bits(wm8994->regmap, reg, mask, val);
-}
-EXPORT_SYMBOL_GPL(wm8994_set_bits);
-
 static struct mfd_cell wm8994_regulator_devs[] = {
 	{
 		.name = "wm8994-ldo",
diff --git a/drivers/misc/Kconfig b/drivers/misc/Kconfig
index e760715bd9cb..a3e291d0df9a 100644
--- a/drivers/misc/Kconfig
+++ b/drivers/misc/Kconfig
@@ -381,19 +381,6 @@ config HMC6352
 	  This driver provides support for the Honeywell HMC6352 compass,
 	  providing configuration and heading data via sysfs.
 
-config EP93XX_PWM
-	tristate "EP93xx PWM support"
-	depends on ARCH_EP93XX
-	help
-	  This option enables device driver support for the PWM channels
-	  on the Cirrus EP93xx processors.  The EP9307 chip only has one
-	  PWM channel all the others have two, the second channel is an
-	  alternate function of the EGPIO14 pin.  A sysfs interface is
-	  provided to control the PWM channels.
-
-	  To compile this driver as a module, choose M here: the module will
-	  be called ep93xx_pwm.
-
 config DS1682
 	tristate "Dallas DS1682 Total Elapsed Time Recorder with Alarm"
 	depends on I2C
diff --git a/drivers/misc/Makefile b/drivers/misc/Makefile
index 0b7ea3ea8bb8..f45473e68bf7 100644
--- a/drivers/misc/Makefile
+++ b/drivers/misc/Makefile
@@ -33,7 +33,6 @@ obj-$(CONFIG_APDS9802ALS)	+= apds9802als.o
 obj-$(CONFIG_ISL29003)		+= isl29003.o
 obj-$(CONFIG_ISL29020)		+= isl29020.o
 obj-$(CONFIG_SENSORS_TSL2550)	+= tsl2550.o
-obj-$(CONFIG_EP93XX_PWM)	+= ep93xx_pwm.o
 obj-$(CONFIG_DS1682)		+= ds1682.o
 obj-$(CONFIG_TI_DAC7512)	+= ti_dac7512.o
 obj-$(CONFIG_C2PORT)		+= c2port/
diff --git a/drivers/misc/carma/carma-fpga.c b/drivers/misc/carma/carma-fpga.c
index 08b18f3f5264..9e2b985293fc 100644
--- a/drivers/misc/carma/carma-fpga.c
+++ b/drivers/misc/carma/carma-fpga.c
@@ -633,8 +633,7 @@ static int data_submit_dma(struct fpga_device *priv, struct data_buf *buf)
 	struct dma_async_tx_descriptor *tx;
 	dma_cookie_t cookie;
 	dma_addr_t dst, src;
-	unsigned long dma_flags = DMA_COMPL_SKIP_DEST_UNMAP |
-				  DMA_COMPL_SKIP_SRC_UNMAP;
+	unsigned long dma_flags = 0;
 
 	dst_sg = buf->vb.sglist;
 	dst_nents = buf->vb.sglen;
diff --git a/drivers/misc/eeprom/at24.c b/drivers/misc/eeprom/at24.c
index 94b8a3324319..d87f77f790d6 100644
--- a/drivers/misc/eeprom/at24.c
+++ b/drivers/misc/eeprom/at24.c
@@ -22,7 +22,7 @@
 #include <linux/jiffies.h>
 #include <linux/of.h>
 #include <linux/i2c.h>
-#include <linux/i2c/at24.h>
+#include <linux/platform_data/at24.h>
 
 /*
  * I2C EEPROMs from most vendors are inexpensive and mostly interchangeable.
diff --git a/drivers/misc/ep93xx_pwm.c b/drivers/misc/ep93xx_pwm.c
deleted file mode 100644
index cdb67a9c1959..000000000000
--- a/drivers/misc/ep93xx_pwm.c
+++ /dev/null
@@ -1,286 +0,0 @@
-/*
- *  Simple PWM driver for EP93XX
- *
- *	(c) Copyright 2009  Matthieu Crapet <mcrapet@gmail.com>
- *	(c) Copyright 2009  H Hartley Sweeten <hsweeten@visionengravers.com>
- *
- *	This program is free software; you can redistribute it and/or
- *	modify it under the terms of the GNU General Public License
- *	as published by the Free Software Foundation; either version
- *	2 of the License, or (at your option) any later version.
- *
- *  EP9307 has only one channel:
- *    - PWMOUT
- *
- *  EP9301/02/12/15 have two channels:
- *    - PWMOUT
- *    - PWMOUT1 (alternate function for EGPIO14)
- */
-
-#include <linux/module.h>
-#include <linux/platform_device.h>
-#include <linux/slab.h>
-#include <linux/clk.h>
-#include <linux/err.h>
-#include <linux/io.h>
-
-#include <mach/platform.h>
-
-#define EP93XX_PWMx_TERM_COUNT	0x00
-#define EP93XX_PWMx_DUTY_CYCLE	0x04
-#define EP93XX_PWMx_ENABLE	0x08
-#define EP93XX_PWMx_INVERT	0x0C
-
-#define EP93XX_PWM_MAX_COUNT	0xFFFF
-
-struct ep93xx_pwm {
-	void __iomem	*mmio_base;
-	struct clk	*clk;
-	u32		duty_percent;
-};
-
-/*
- * /sys/devices/platform/ep93xx-pwm.N
- *   /min_freq      read-only   minimum pwm output frequency
- *   /max_req       read-only   maximum pwm output frequency
- *   /freq          read-write  pwm output frequency (0 = disable output)
- *   /duty_percent  read-write  pwm duty cycle percent (1..99)
- *   /invert        read-write  invert pwm output
- */
-
-static ssize_t ep93xx_pwm_get_min_freq(struct device *dev,
-		struct device_attribute *attr, char *buf)
-{
-	struct platform_device *pdev = to_platform_device(dev);
-	struct ep93xx_pwm *pwm = platform_get_drvdata(pdev);
-	unsigned long rate = clk_get_rate(pwm->clk);
-
-	return sprintf(buf, "%ld\n", rate / (EP93XX_PWM_MAX_COUNT + 1));
-}
-
-static ssize_t ep93xx_pwm_get_max_freq(struct device *dev,
-		struct device_attribute *attr, char *buf)
-{
-	struct platform_device *pdev = to_platform_device(dev);
-	struct ep93xx_pwm *pwm = platform_get_drvdata(pdev);
-	unsigned long rate = clk_get_rate(pwm->clk);
-
-	return sprintf(buf, "%ld\n", rate / 2);
-}
-
-static ssize_t ep93xx_pwm_get_freq(struct device *dev,
-		struct device_attribute *attr, char *buf)
-{
-	struct platform_device *pdev = to_platform_device(dev);
-	struct ep93xx_pwm *pwm = platform_get_drvdata(pdev);
-
-	if (readl(pwm->mmio_base + EP93XX_PWMx_ENABLE) & 0x1) {
-		unsigned long rate = clk_get_rate(pwm->clk);
-		u16 term = readl(pwm->mmio_base + EP93XX_PWMx_TERM_COUNT);
-
-		return sprintf(buf, "%ld\n", rate / (term + 1));
-	} else {
-		return sprintf(buf, "disabled\n");
-	}
-}
-
-static ssize_t ep93xx_pwm_set_freq(struct device *dev,
-		struct device_attribute *attr, const char *buf, size_t count)
-{
-	struct platform_device *pdev = to_platform_device(dev);
-	struct ep93xx_pwm *pwm = platform_get_drvdata(pdev);
-	long val;
-	int err;
-
-	err = kstrtol(buf, 10, &val);
-	if (err)
-		return -EINVAL;
-
-	if (val == 0) {
-		writel(0x0, pwm->mmio_base + EP93XX_PWMx_ENABLE);
-	} else if (val <= (clk_get_rate(pwm->clk) / 2)) {
-		u32 term, duty;
-
-		val = (clk_get_rate(pwm->clk) / val) - 1;
-		if (val > EP93XX_PWM_MAX_COUNT)
-			val = EP93XX_PWM_MAX_COUNT;
-		if (val < 1)
-			val = 1;
-
-		term = readl(pwm->mmio_base + EP93XX_PWMx_TERM_COUNT);
-		duty = ((val + 1) * pwm->duty_percent / 100) - 1;
-
-		/* If pwm is running, order is important */
-		if (val > term) {
-			writel(val, pwm->mmio_base + EP93XX_PWMx_TERM_COUNT);
-			writel(duty, pwm->mmio_base + EP93XX_PWMx_DUTY_CYCLE);
-		} else {
-			writel(duty, pwm->mmio_base + EP93XX_PWMx_DUTY_CYCLE);
-			writel(val, pwm->mmio_base + EP93XX_PWMx_TERM_COUNT);
-		}
-
-		if (!readl(pwm->mmio_base + EP93XX_PWMx_ENABLE) & 0x1)
-			writel(0x1, pwm->mmio_base + EP93XX_PWMx_ENABLE);
-	} else {
-		return -EINVAL;
-	}
-
-	return count;
-}
-
-static ssize_t ep93xx_pwm_get_duty_percent(struct device *dev,
-		struct device_attribute *attr, char *buf)
-{
-	struct platform_device *pdev = to_platform_device(dev);
-	struct ep93xx_pwm *pwm = platform_get_drvdata(pdev);
-
-	return sprintf(buf, "%d\n", pwm->duty_percent);
-}
-
-static ssize_t ep93xx_pwm_set_duty_percent(struct device *dev,
-		struct device_attribute *attr, const char *buf, size_t count)
-{
-	struct platform_device *pdev = to_platform_device(dev);
-	struct ep93xx_pwm *pwm = platform_get_drvdata(pdev);
-	long val;
-	int err;
-
-	err = kstrtol(buf, 10, &val);
-	if (err)
-		return -EINVAL;
-
-	if (val > 0 && val < 100) {
-		u32 term = readl(pwm->mmio_base + EP93XX_PWMx_TERM_COUNT);
-		u32 duty = ((term + 1) * val / 100) - 1;
-
-		writel(duty, pwm->mmio_base + EP93XX_PWMx_DUTY_CYCLE);
-		pwm->duty_percent = val;
-		return count;
-	}
-
-	return -EINVAL;
-}
-
-static ssize_t ep93xx_pwm_get_invert(struct device *dev,
-		struct device_attribute *attr, char *buf)
-{
-	struct platform_device *pdev = to_platform_device(dev);
-	struct ep93xx_pwm *pwm = platform_get_drvdata(pdev);
-	int inverted = readl(pwm->mmio_base + EP93XX_PWMx_INVERT) & 0x1;
-
-	return sprintf(buf, "%d\n", inverted);
-}
-
-static ssize_t ep93xx_pwm_set_invert(struct device *dev,
-		struct device_attribute *attr, const char *buf, size_t count)
-{
-	struct platform_device *pdev = to_platform_device(dev);
-	struct ep93xx_pwm *pwm = platform_get_drvdata(pdev);
-	long val;
-	int err;
-
-	err = kstrtol(buf, 10, &val);
-	if (err)
-		return -EINVAL;
-
-	if (val == 0)
-		writel(0x0, pwm->mmio_base + EP93XX_PWMx_INVERT);
-	else if (val == 1)
-		writel(0x1, pwm->mmio_base + EP93XX_PWMx_INVERT);
-	else
-		return -EINVAL;
-
-	return count;
-}
-
-static DEVICE_ATTR(min_freq, S_IRUGO, ep93xx_pwm_get_min_freq, NULL);
-static DEVICE_ATTR(max_freq, S_IRUGO, ep93xx_pwm_get_max_freq, NULL);
-static DEVICE_ATTR(freq, S_IWUSR | S_IRUGO,
-		   ep93xx_pwm_get_freq, ep93xx_pwm_set_freq);
-static DEVICE_ATTR(duty_percent, S_IWUSR | S_IRUGO,
-		   ep93xx_pwm_get_duty_percent, ep93xx_pwm_set_duty_percent);
-static DEVICE_ATTR(invert, S_IWUSR | S_IRUGO,
-		   ep93xx_pwm_get_invert, ep93xx_pwm_set_invert);
-
-static struct attribute *ep93xx_pwm_attrs[] = {
-	&dev_attr_min_freq.attr,
-	&dev_attr_max_freq.attr,
-	&dev_attr_freq.attr,
-	&dev_attr_duty_percent.attr,
-	&dev_attr_invert.attr,
-	NULL
-};
-
-static const struct attribute_group ep93xx_pwm_sysfs_files = {
-	.attrs	= ep93xx_pwm_attrs,
-};
-
-static int ep93xx_pwm_probe(struct platform_device *pdev)
-{
-	struct ep93xx_pwm *pwm;
-	struct resource *res;
-	int ret;
-
-	pwm = devm_kzalloc(&pdev->dev, sizeof(*pwm), GFP_KERNEL);
-	if (!pwm)
-		return -ENOMEM;
-
-	pwm->clk = devm_clk_get(&pdev->dev, "pwm_clk");
-	if (IS_ERR(pwm->clk))
-		return PTR_ERR(pwm->clk);
-
-	res = platform_get_resource(pdev, IORESOURCE_MEM, 0);
-	pwm->mmio_base = devm_ioremap_resource(&pdev->dev, res);
-	if (IS_ERR(pwm->mmio_base))
-		return PTR_ERR(pwm->mmio_base);
-
-	ret = ep93xx_pwm_acquire_gpio(pdev);
-	if (ret)
-		return ret;
-
-	ret = sysfs_create_group(&pdev->dev.kobj, &ep93xx_pwm_sysfs_files);
-	if (ret) {
-		ep93xx_pwm_release_gpio(pdev);
-		return ret;
-	}
-
-	pwm->duty_percent = 50;
-
-	/* disable pwm at startup. Avoids zero value. */
-	writel(0x0, pwm->mmio_base + EP93XX_PWMx_ENABLE);
-	writel(EP93XX_PWM_MAX_COUNT, pwm->mmio_base + EP93XX_PWMx_TERM_COUNT);
-	writel(EP93XX_PWM_MAX_COUNT/2, pwm->mmio_base + EP93XX_PWMx_DUTY_CYCLE);
-
-	clk_enable(pwm->clk);
-
-	platform_set_drvdata(pdev, pwm);
-	return 0;
-}
-
-static int ep93xx_pwm_remove(struct platform_device *pdev)
-{
-	struct ep93xx_pwm *pwm = platform_get_drvdata(pdev);
-
-	writel(0x0, pwm->mmio_base + EP93XX_PWMx_ENABLE);
-	clk_disable(pwm->clk);
-	sysfs_remove_group(&pdev->dev.kobj, &ep93xx_pwm_sysfs_files);
-	ep93xx_pwm_release_gpio(pdev);
-
-	return 0;
-}
-
-static struct platform_driver ep93xx_pwm_driver = {
-	.driver		= {
-		.name	= "ep93xx-pwm",
-		.owner	= THIS_MODULE,
-	},
-	.probe		= ep93xx_pwm_probe,
-	.remove		= ep93xx_pwm_remove,
-};
-module_platform_driver(ep93xx_pwm_driver);
-
-MODULE_AUTHOR("Matthieu Crapet <mcrapet@gmail.com>, "
-	      "H Hartley Sweeten <hsweeten@visionengravers.com>");
-MODULE_DESCRIPTION("EP93xx PWM driver");
-MODULE_LICENSE("GPL");
-MODULE_ALIAS("platform:ep93xx-pwm");
diff --git a/drivers/misc/mic/card/mic_virtio.c b/drivers/misc/mic/card/mic_virtio.c
index 914cc9b2caad..8aa42e738acc 100644
--- a/drivers/misc/mic/card/mic_virtio.c
+++ b/drivers/misc/mic/card/mic_virtio.c
@@ -493,7 +493,7 @@ static int mic_remove_device(struct mic_device_desc __iomem *d,
 			ioread8(&dc->config_change), ioread8(&d->type), mvdev);
 
 		status = ioread8(&d->status);
-		INIT_COMPLETION(mvdev->reset_done);
+		reinit_completion(&mvdev->reset_done);
 		unregister_virtio_device(&mvdev->vdev);
 		mic_free_card_irq(mvdev->virtio_cookie, mvdev);
 		if (status & VIRTIO_CONFIG_S_DRIVER_OK)
diff --git a/drivers/misc/mic/host/mic_boot.c b/drivers/misc/mic/host/mic_boot.c
index b079c65eed6d..7558d9186438 100644
--- a/drivers/misc/mic/host/mic_boot.c
+++ b/drivers/misc/mic/host/mic_boot.c
@@ -38,7 +38,7 @@ static void mic_reset(struct mic_device *mdev)
 
 #define MIC_RESET_TO (45)
 
-	INIT_COMPLETION(mdev->reset_wait);
+	reinit_completion(&mdev->reset_wait);
 	mdev->ops->reset_fw_ready(mdev);
 	mdev->ops->reset(mdev);
 
diff --git a/drivers/misc/ti-st/st_kim.c b/drivers/misc/ti-st/st_kim.c
index 83907c720594..96853a09788a 100644
--- a/drivers/misc/ti-st/st_kim.c
+++ b/drivers/misc/ti-st/st_kim.c
@@ -218,7 +218,7 @@ static long read_local_version(struct kim_data_s *kim_gdata, char *bts_scr_name)
 
 	pr_debug("%s", __func__);
 
-	INIT_COMPLETION(kim_gdata->kim_rcvd);
+	reinit_completion(&kim_gdata->kim_rcvd);
 	if (4 != st_int_write(kim_gdata->core_data, read_ver_cmd, 4)) {
 		pr_err("kim: couldn't write 4 bytes");
 		return -EIO;
@@ -229,7 +229,7 @@ static long read_local_version(struct kim_data_s *kim_gdata, char *bts_scr_name)
 		pr_err(" waiting for ver info- timed out ");
 		return -ETIMEDOUT;
 	}
-	INIT_COMPLETION(kim_gdata->kim_rcvd);
+	reinit_completion(&kim_gdata->kim_rcvd);
 	/* the positions 12 & 13 in the response buffer provide with the
 	 * chip, major & minor numbers
 	 */
@@ -362,7 +362,7 @@ static long download_firmware(struct kim_data_s *kim_gdata)
 			/* reinit completion before sending for the
 			 * relevant wait
 			 */
-			INIT_COMPLETION(kim_gdata->kim_rcvd);
+			reinit_completion(&kim_gdata->kim_rcvd);
 
 			/*
 			 * Free space found in uart buffer, call st_int_write
@@ -398,7 +398,7 @@ static long download_firmware(struct kim_data_s *kim_gdata)
 				release_firmware(kim_gdata->fw_entry);
 				return -ETIMEDOUT;
 			}
-			INIT_COMPLETION(kim_gdata->kim_rcvd);
+			reinit_completion(&kim_gdata->kim_rcvd);
 			break;
 		case ACTION_DELAY:	/* sleep */
 			pr_info("sleep command in scr");
@@ -474,7 +474,7 @@ long st_kim_start(void *kim_data)
 		gpio_set_value(kim_gdata->nshutdown, GPIO_HIGH);
 		mdelay(100);
 		/* re-initialize the completion */
-		INIT_COMPLETION(kim_gdata->ldisc_installed);
+		reinit_completion(&kim_gdata->ldisc_installed);
 		/* send notification to UIM */
 		kim_gdata->ldisc_install = 1;
 		pr_info("ldisc_install = 1");
@@ -525,7 +525,7 @@ long st_kim_stop(void *kim_data)
 		kim_gdata->kim_pdev->dev.platform_data;
 	struct tty_struct	*tty = kim_gdata->core_data->tty;
 
-	INIT_COMPLETION(kim_gdata->ldisc_installed);
+	reinit_completion(&kim_gdata->ldisc_installed);
 
 	if (tty) {	/* can be called before ldisc is installed */
 		/* Flush any pending characters in the driver and discipline. */
diff --git a/drivers/mmc/card/block.c b/drivers/mmc/card/block.c
index 1a3163f1407e..29d5d988a51c 100644
--- a/drivers/mmc/card/block.c
+++ b/drivers/mmc/card/block.c
@@ -2448,7 +2448,6 @@ static int _mmc_blk_suspend(struct mmc_card *card)
 	struct mmc_blk_data *md = mmc_get_drvdata(card);
 
 	if (md) {
-		pm_runtime_get_sync(&card->dev);
 		mmc_queue_suspend(&md->queue);
 		list_for_each_entry(part_md, &md->part, part) {
 			mmc_queue_suspend(&part_md->queue);
@@ -2483,7 +2482,6 @@ static int mmc_blk_resume(struct mmc_card *card)
 		list_for_each_entry(part_md, &md->part, part) {
 			mmc_queue_resume(&part_md->queue);
 		}
-		pm_runtime_put(&card->dev);
 	}
 	return 0;
 }
diff --git a/drivers/mmc/core/bus.c b/drivers/mmc/core/bus.c
index 3e227bd91e81..64145a32b917 100644
--- a/drivers/mmc/core/bus.c
+++ b/drivers/mmc/core/bus.c
@@ -342,7 +342,7 @@ int mmc_add_card(struct mmc_card *card)
 		break;
 	}
 
-	if (mmc_sd_card_uhs(card) &&
+	if (mmc_card_uhs(card) &&
 		(card->sd_bus_speed < ARRAY_SIZE(uhs_speeds)))
 		uhs_bus_speed_mode = uhs_speeds[card->sd_bus_speed];
 
diff --git a/drivers/mmc/core/core.c b/drivers/mmc/core/core.c
index bf18b6bfce48..57a2b403bf8e 100644
--- a/drivers/mmc/core/core.c
+++ b/drivers/mmc/core/core.c
@@ -23,6 +23,7 @@
 #include <linux/log2.h>
 #include <linux/regulator/consumer.h>
 #include <linux/pm_runtime.h>
+#include <linux/pm_wakeup.h>
 #include <linux/suspend.h>
 #include <linux/fault-inject.h>
 #include <linux/random.h>
@@ -301,7 +302,7 @@ void mmc_start_bkops(struct mmc_card *card, bool from_exception)
 	}
 
 	err = __mmc_switch(card, EXT_CSD_CMD_SET_NORMAL,
-			EXT_CSD_BKOPS_START, 1, timeout, use_busy_signal);
+			EXT_CSD_BKOPS_START, 1, timeout, use_busy_signal, true);
 	if (err) {
 		pr_warn("%s: Error %d starting bkops\n",
 			mmc_hostname(card->host), err);
@@ -918,31 +919,6 @@ int __mmc_claim_host(struct mmc_host *host, atomic_t *abort)
 EXPORT_SYMBOL(__mmc_claim_host);
 
 /**
- *	mmc_try_claim_host - try exclusively to claim a host
- *	@host: mmc host to claim
- *
- *	Returns %1 if the host is claimed, %0 otherwise.
- */
-int mmc_try_claim_host(struct mmc_host *host)
-{
-	int claimed_host = 0;
-	unsigned long flags;
-
-	spin_lock_irqsave(&host->lock, flags);
-	if (!host->claimed || host->claimer == current) {
-		host->claimed = 1;
-		host->claimer = current;
-		host->claim_cnt += 1;
-		claimed_host = 1;
-	}
-	spin_unlock_irqrestore(&host->lock, flags);
-	if (host->ops->enable && claimed_host && host->claim_cnt == 1)
-		host->ops->enable(host);
-	return claimed_host;
-}
-EXPORT_SYMBOL(mmc_try_claim_host);
-
-/**
  *	mmc_release_host - release a host
  *	@host: mmc host to release
  *
@@ -1382,22 +1358,31 @@ u32 mmc_select_voltage(struct mmc_host *host, u32 ocr)
 {
 	int bit;
 
-	ocr &= host->ocr_avail;
+	/*
+	 * Sanity check the voltages that the card claims to
+	 * support.
+	 */
+	if (ocr & 0x7F) {
+		dev_warn(mmc_dev(host),
+		"card claims to support voltages below defined range\n");
+		ocr &= ~0x7F;
+	}
 
-	bit = ffs(ocr);
-	if (bit) {
-		bit -= 1;
+	ocr &= host->ocr_avail;
+	if (!ocr) {
+		dev_warn(mmc_dev(host), "no support for card's volts\n");
+		return 0;
+	}
 
+	if (host->caps2 & MMC_CAP2_FULL_PWR_CYCLE) {
+		bit = ffs(ocr) - 1;
 		ocr &= 3 << bit;
-
-		mmc_host_clk_hold(host);
-		host->ios.vdd = bit;
-		mmc_set_ios(host);
-		mmc_host_clk_release(host);
+		mmc_power_cycle(host, ocr);
 	} else {
-		pr_warning("%s: host doesn't support card's voltages\n",
-				mmc_hostname(host));
-		ocr = 0;
+		bit = fls(ocr) - 1;
+		ocr &= 3 << bit;
+		if (bit != host->ios.vdd)
+			dev_warn(mmc_dev(host), "exceeding card's volts\n");
 	}
 
 	return ocr;
@@ -1422,7 +1407,7 @@ int __mmc_set_signal_voltage(struct mmc_host *host, int signal_voltage)
 
 }
 
-int mmc_set_signal_voltage(struct mmc_host *host, int signal_voltage)
+int mmc_set_signal_voltage(struct mmc_host *host, int signal_voltage, u32 ocr)
 {
 	struct mmc_command cmd = {0};
 	int err = 0;
@@ -1504,7 +1489,7 @@ power_cycle:
 	if (err) {
 		pr_debug("%s: Signal voltage switch failed, "
 			"power cycling card\n", mmc_hostname(host));
-		mmc_power_cycle(host);
+		mmc_power_cycle(host, ocr);
 	}
 
 	mmc_host_clk_release(host);
@@ -1545,22 +1530,14 @@ void mmc_set_driver_type(struct mmc_host *host, unsigned int drv_type)
  * If a host does all the power sequencing itself, ignore the
  * initial MMC_POWER_UP stage.
  */
-void mmc_power_up(struct mmc_host *host)
+void mmc_power_up(struct mmc_host *host, u32 ocr)
 {
-	int bit;
-
 	if (host->ios.power_mode == MMC_POWER_ON)
 		return;
 
 	mmc_host_clk_hold(host);
 
-	/* If ocr is set, we use it */
-	if (host->ocr)
-		bit = ffs(host->ocr) - 1;
-	else
-		bit = fls(host->ocr_avail) - 1;
-
-	host->ios.vdd = bit;
+	host->ios.vdd = fls(ocr) - 1;
 	if (mmc_host_is_spi(host))
 		host->ios.chip_select = MMC_CS_HIGH;
 	else
@@ -1604,13 +1581,6 @@ void mmc_power_off(struct mmc_host *host)
 	host->ios.clock = 0;
 	host->ios.vdd = 0;
 
-
-	/*
-	 * Reset ocr mask to be the highest possible voltage supported for
-	 * this mmc host. This value will be used at next power up.
-	 */
-	host->ocr = 1 << (fls(host->ocr_avail) - 1);
-
 	if (!mmc_host_is_spi(host)) {
 		host->ios.bus_mode = MMC_BUSMODE_OPENDRAIN;
 		host->ios.chip_select = MMC_CS_DONTCARE;
@@ -1630,12 +1600,12 @@ void mmc_power_off(struct mmc_host *host)
 	mmc_host_clk_release(host);
 }
 
-void mmc_power_cycle(struct mmc_host *host)
+void mmc_power_cycle(struct mmc_host *host, u32 ocr)
 {
 	mmc_power_off(host);
 	/* Wait at least 1 ms according to SD spec */
 	mmc_delay(1);
-	mmc_power_up(host);
+	mmc_power_up(host, ocr);
 }
 
 /*
@@ -1723,6 +1693,28 @@ void mmc_detach_bus(struct mmc_host *host)
 	mmc_bus_put(host);
 }
 
+static void _mmc_detect_change(struct mmc_host *host, unsigned long delay,
+				bool cd_irq)
+{
+#ifdef CONFIG_MMC_DEBUG
+	unsigned long flags;
+	spin_lock_irqsave(&host->lock, flags);
+	WARN_ON(host->removed);
+	spin_unlock_irqrestore(&host->lock, flags);
+#endif
+
+	/*
+	 * If the device is configured as wakeup, we prevent a new sleep for
+	 * 5 s to give provision for user space to consume the event.
+	 */
+	if (cd_irq && !(host->caps & MMC_CAP_NEEDS_POLL) &&
+		device_can_wakeup(mmc_dev(host)))
+		pm_wakeup_event(mmc_dev(host), 5000);
+
+	host->detect_change = 1;
+	mmc_schedule_delayed_work(&host->detect, delay);
+}
+
 /**
  *	mmc_detect_change - process change of state on a MMC socket
  *	@host: host which changed state.
@@ -1735,16 +1727,8 @@ void mmc_detach_bus(struct mmc_host *host)
  */
 void mmc_detect_change(struct mmc_host *host, unsigned long delay)
 {
-#ifdef CONFIG_MMC_DEBUG
-	unsigned long flags;
-	spin_lock_irqsave(&host->lock, flags);
-	WARN_ON(host->removed);
-	spin_unlock_irqrestore(&host->lock, flags);
-#endif
-	host->detect_change = 1;
-	mmc_schedule_delayed_work(&host->detect, delay);
+	_mmc_detect_change(host, delay, true);
 }
-
 EXPORT_SYMBOL(mmc_detect_change);
 
 void mmc_init_erase(struct mmc_card *card)
@@ -2334,7 +2318,7 @@ static int mmc_rescan_try_freq(struct mmc_host *host, unsigned freq)
 	pr_info("%s: %s: trying to init card at %u Hz\n",
 		mmc_hostname(host), __func__, host->f_init);
 #endif
-	mmc_power_up(host);
+	mmc_power_up(host, host->ocr_avail);
 
 	/*
 	 * Some eMMCs (with VCCQ always on) may not be reset after power up, so
@@ -2423,7 +2407,7 @@ int mmc_detect_card_removed(struct mmc_host *host)
 			 * rescan handle the card removal.
 			 */
 			cancel_delayed_work(&host->detect);
-			mmc_detect_change(host, 0);
+			_mmc_detect_change(host, 0, false);
 		}
 	}
 
@@ -2504,8 +2488,8 @@ void mmc_start_host(struct mmc_host *host)
 	if (host->caps2 & MMC_CAP2_NO_PRESCAN_POWERUP)
 		mmc_power_off(host);
 	else
-		mmc_power_up(host);
-	mmc_detect_change(host, 0);
+		mmc_power_up(host, host->ocr_avail);
+	_mmc_detect_change(host, 0, false);
 }
 
 void mmc_stop_host(struct mmc_host *host)
@@ -2583,7 +2567,7 @@ int mmc_power_restore_host(struct mmc_host *host)
 		return -EINVAL;
 	}
 
-	mmc_power_up(host);
+	mmc_power_up(host, host->card->ocr);
 	ret = host->bus_ops->power_restore(host);
 
 	mmc_bus_put(host);
@@ -2657,28 +2641,6 @@ EXPORT_SYMBOL(mmc_cache_ctrl);
 
 #ifdef CONFIG_PM
 
-/**
- *	mmc_suspend_host - suspend a host
- *	@host: mmc host
- */
-int mmc_suspend_host(struct mmc_host *host)
-{
-	/* This function is deprecated */
-	return 0;
-}
-EXPORT_SYMBOL(mmc_suspend_host);
-
-/**
- *	mmc_resume_host - resume a previously suspended host
- *	@host: mmc host
- */
-int mmc_resume_host(struct mmc_host *host)
-{
-	/* This function is deprecated */
-	return 0;
-}
-EXPORT_SYMBOL(mmc_resume_host);
-
 /* Do the card removal on suspend if card is assumed removeable
  * Do that in pm notifier while userspace isn't yet frozen, so we will be able
    to sync the card.
@@ -2724,7 +2686,7 @@ int mmc_pm_notify(struct notifier_block *notify_block,
 		spin_lock_irqsave(&host->lock, flags);
 		host->rescan_disable = 0;
 		spin_unlock_irqrestore(&host->lock, flags);
-		mmc_detect_change(host, 0);
+		_mmc_detect_change(host, 0, false);
 
 	}
 
diff --git a/drivers/mmc/core/core.h b/drivers/mmc/core/core.h
index 5345d156493e..443a584660f0 100644
--- a/drivers/mmc/core/core.h
+++ b/drivers/mmc/core/core.h
@@ -42,13 +42,13 @@ void mmc_set_ungated(struct mmc_host *host);
 void mmc_set_bus_mode(struct mmc_host *host, unsigned int mode);
 void mmc_set_bus_width(struct mmc_host *host, unsigned int width);
 u32 mmc_select_voltage(struct mmc_host *host, u32 ocr);
-int mmc_set_signal_voltage(struct mmc_host *host, int signal_voltage);
+int mmc_set_signal_voltage(struct mmc_host *host, int signal_voltage, u32 ocr);
 int __mmc_set_signal_voltage(struct mmc_host *host, int signal_voltage);
 void mmc_set_timing(struct mmc_host *host, unsigned int timing);
 void mmc_set_driver_type(struct mmc_host *host, unsigned int drv_type);
-void mmc_power_up(struct mmc_host *host);
+void mmc_power_up(struct mmc_host *host, u32 ocr);
 void mmc_power_off(struct mmc_host *host);
-void mmc_power_cycle(struct mmc_host *host);
+void mmc_power_cycle(struct mmc_host *host, u32 ocr);
 
 static inline void mmc_delay(unsigned int ms)
 {
diff --git a/drivers/mmc/core/mmc.c b/drivers/mmc/core/mmc.c
index 6d02012a1d0b..f631f5a9bf79 100644
--- a/drivers/mmc/core/mmc.c
+++ b/drivers/mmc/core/mmc.c
@@ -13,6 +13,7 @@
 #include <linux/err.h>
 #include <linux/slab.h>
 #include <linux/stat.h>
+#include <linux/pm_runtime.h>
 
 #include <linux/mmc/host.h>
 #include <linux/mmc/card.h>
@@ -934,6 +935,7 @@ static int mmc_init_card(struct mmc_host *host, u32 ocr,
 			goto err;
 		}
 
+		card->ocr = ocr;
 		card->type = MMC_TYPE_MMC;
 		card->rca = 1;
 		memcpy(card->raw_cid, cid, sizeof(card->raw_cid));
@@ -1404,9 +1406,9 @@ static int mmc_poweroff_notify(struct mmc_card *card, unsigned int notify_type)
 	if (notify_type == EXT_CSD_POWER_OFF_LONG)
 		timeout = card->ext_csd.power_off_longtime;
 
-	err = mmc_switch(card, EXT_CSD_CMD_SET_NORMAL,
-			 EXT_CSD_POWER_OFF_NOTIFICATION,
-			 notify_type, timeout);
+	err = __mmc_switch(card, EXT_CSD_CMD_SET_NORMAL,
+			EXT_CSD_POWER_OFF_NOTIFICATION,
+			notify_type, timeout, true, false);
 	if (err)
 		pr_err("%s: Power Off Notification timed out, %u\n",
 		       mmc_hostname(card->host), timeout);
@@ -1477,6 +1479,9 @@ static int _mmc_suspend(struct mmc_host *host, bool is_suspend)
 
 	mmc_claim_host(host);
 
+	if (mmc_card_suspended(host->card))
+		goto out;
+
 	if (mmc_card_doing_bkops(host->card)) {
 		err = mmc_stop_bkops(host->card);
 		if (err)
@@ -1496,51 +1501,93 @@ static int _mmc_suspend(struct mmc_host *host, bool is_suspend)
 		err = mmc_deselect_cards(host);
 	host->card->state &= ~(MMC_STATE_HIGHSPEED | MMC_STATE_HIGHSPEED_200);
 
-	if (!err)
+	if (!err) {
 		mmc_power_off(host);
+		mmc_card_set_suspended(host->card);
+	}
 out:
 	mmc_release_host(host);
 	return err;
 }
 
 /*
- * Suspend callback from host.
+ * Suspend callback
  */
 static int mmc_suspend(struct mmc_host *host)
 {
-	return _mmc_suspend(host, true);
-}
+	int err;
 
-/*
- * Shutdown callback
- */
-static int mmc_shutdown(struct mmc_host *host)
-{
-	return _mmc_suspend(host, false);
+	err = _mmc_suspend(host, true);
+	if (!err) {
+		pm_runtime_disable(&host->card->dev);
+		pm_runtime_set_suspended(&host->card->dev);
+	}
+
+	return err;
 }
 
 /*
- * Resume callback from host.
- *
  * This function tries to determine if the same card is still present
  * and, if so, restore all state to it.
  */
-static int mmc_resume(struct mmc_host *host)
+static int _mmc_resume(struct mmc_host *host)
 {
-	int err;
+	int err = 0;
 
 	BUG_ON(!host);
 	BUG_ON(!host->card);
 
 	mmc_claim_host(host);
-	mmc_power_up(host);
-	mmc_select_voltage(host, host->ocr);
-	err = mmc_init_card(host, host->ocr, host->card);
+
+	if (!mmc_card_suspended(host->card))
+		goto out;
+
+	mmc_power_up(host, host->card->ocr);
+	err = mmc_init_card(host, host->card->ocr, host->card);
+	mmc_card_clr_suspended(host->card);
+
+out:
 	mmc_release_host(host);
+	return err;
+}
+
+/*
+ * Shutdown callback
+ */
+static int mmc_shutdown(struct mmc_host *host)
+{
+	int err = 0;
+
+	/*
+	 * In a specific case for poweroff notify, we need to resume the card
+	 * before we can shutdown it properly.
+	 */
+	if (mmc_can_poweroff_notify(host->card) &&
+		!(host->caps2 & MMC_CAP2_FULL_PWR_CYCLE))
+		err = _mmc_resume(host);
+
+	if (!err)
+		err = _mmc_suspend(host, false);
 
 	return err;
 }
 
+/*
+ * Callback for resume.
+ */
+static int mmc_resume(struct mmc_host *host)
+{
+	int err = 0;
+
+	if (!(host->caps & MMC_CAP_RUNTIME_RESUME)) {
+		err = _mmc_resume(host);
+		pm_runtime_set_active(&host->card->dev);
+		pm_runtime_mark_last_busy(&host->card->dev);
+	}
+	pm_runtime_enable(&host->card->dev);
+
+	return err;
+}
 
 /*
  * Callback for runtime_suspend.
@@ -1552,18 +1599,11 @@ static int mmc_runtime_suspend(struct mmc_host *host)
 	if (!(host->caps & MMC_CAP_AGGRESSIVE_PM))
 		return 0;
 
-	mmc_claim_host(host);
-
-	err = mmc_suspend(host);
-	if (err) {
+	err = _mmc_suspend(host, true);
+	if (err)
 		pr_err("%s: error %d doing aggessive suspend\n",
 			mmc_hostname(host), err);
-		goto out;
-	}
-	mmc_power_off(host);
 
-out:
-	mmc_release_host(host);
 	return err;
 }
 
@@ -1574,18 +1614,14 @@ static int mmc_runtime_resume(struct mmc_host *host)
 {
 	int err;
 
-	if (!(host->caps & MMC_CAP_AGGRESSIVE_PM))
+	if (!(host->caps & (MMC_CAP_AGGRESSIVE_PM | MMC_CAP_RUNTIME_RESUME)))
 		return 0;
 
-	mmc_claim_host(host);
-
-	mmc_power_up(host);
-	err = mmc_resume(host);
+	err = _mmc_resume(host);
 	if (err)
 		pr_err("%s: error %d doing aggessive resume\n",
 			mmc_hostname(host), err);
 
-	mmc_release_host(host);
 	return 0;
 }
 
@@ -1595,7 +1631,7 @@ static int mmc_power_restore(struct mmc_host *host)
 
 	host->card->state &= ~(MMC_STATE_HIGHSPEED | MMC_STATE_HIGHSPEED_200);
 	mmc_claim_host(host);
-	ret = mmc_init_card(host, host->ocr, host->card);
+	ret = mmc_init_card(host, host->card->ocr, host->card);
 	mmc_release_host(host);
 
 	return ret;
@@ -1640,7 +1676,7 @@ static void mmc_attach_bus_ops(struct mmc_host *host)
 int mmc_attach_mmc(struct mmc_host *host)
 {
 	int err;
-	u32 ocr;
+	u32 ocr, rocr;
 
 	BUG_ON(!host);
 	WARN_ON(!host->claimed);
@@ -1666,23 +1702,12 @@ int mmc_attach_mmc(struct mmc_host *host)
 			goto err;
 	}
 
-	/*
-	 * Sanity check the voltages that the card claims to
-	 * support.
-	 */
-	if (ocr & 0x7F) {
-		pr_warning("%s: card claims to support voltages "
-		       "below the defined range. These will be ignored.\n",
-		       mmc_hostname(host));
-		ocr &= ~0x7F;
-	}
-
-	host->ocr = mmc_select_voltage(host, ocr);
+	rocr = mmc_select_voltage(host, ocr);
 
 	/*
 	 * Can we support the voltage of the card?
 	 */
-	if (!host->ocr) {
+	if (!rocr) {
 		err = -EINVAL;
 		goto err;
 	}
@@ -1690,7 +1715,7 @@ int mmc_attach_mmc(struct mmc_host *host)
 	/*
 	 * Detect and init the card.
 	 */
-	err = mmc_init_card(host, host->ocr, NULL);
+	err = mmc_init_card(host, rocr, NULL);
 	if (err)
 		goto err;
 
diff --git a/drivers/mmc/core/mmc_ops.c b/drivers/mmc/core/mmc_ops.c
index ef183483d5b6..e5b5eeb548d1 100644
--- a/drivers/mmc/core/mmc_ops.c
+++ b/drivers/mmc/core/mmc_ops.c
@@ -23,6 +23,40 @@
 
 #define MMC_OPS_TIMEOUT_MS	(10 * 60 * 1000) /* 10 minute timeout */
 
+static inline int __mmc_send_status(struct mmc_card *card, u32 *status,
+				    bool ignore_crc)
+{
+	int err;
+	struct mmc_command cmd = {0};
+
+	BUG_ON(!card);
+	BUG_ON(!card->host);
+
+	cmd.opcode = MMC_SEND_STATUS;
+	if (!mmc_host_is_spi(card->host))
+		cmd.arg = card->rca << 16;
+	cmd.flags = MMC_RSP_SPI_R2 | MMC_RSP_R1 | MMC_CMD_AC;
+	if (ignore_crc)
+		cmd.flags &= ~MMC_RSP_CRC;
+
+	err = mmc_wait_for_cmd(card->host, &cmd, MMC_CMD_RETRIES);
+	if (err)
+		return err;
+
+	/* NOTE: callers are required to understand the difference
+	 * between "native" and SPI format status words!
+	 */
+	if (status)
+		*status = cmd.resp[0];
+
+	return 0;
+}
+
+int mmc_send_status(struct mmc_card *card, u32 *status)
+{
+	return __mmc_send_status(card, status, false);
+}
+
 static int _mmc_select_card(struct mmc_host *host, struct mmc_card *card)
 {
 	int err;
@@ -370,16 +404,18 @@ int mmc_spi_set_crc(struct mmc_host *host, int use_crc)
  *	@timeout_ms: timeout (ms) for operation performed by register write,
  *                   timeout of zero implies maximum possible timeout
  *	@use_busy_signal: use the busy signal as response type
+ *	@send_status: send status cmd to poll for busy
  *
  *	Modifies the EXT_CSD register for selected card.
  */
 int __mmc_switch(struct mmc_card *card, u8 set, u8 index, u8 value,
-	       unsigned int timeout_ms, bool use_busy_signal)
+		unsigned int timeout_ms, bool use_busy_signal, bool send_status)
 {
 	int err;
 	struct mmc_command cmd = {0};
 	unsigned long timeout;
-	u32 status;
+	u32 status = 0;
+	bool ignore_crc = false;
 
 	BUG_ON(!card);
 	BUG_ON(!card->host);
@@ -408,17 +444,37 @@ int __mmc_switch(struct mmc_card *card, u8 set, u8 index, u8 value,
 	if (!use_busy_signal)
 		return 0;
 
-	/* Must check status to be sure of no errors */
+	/*
+	 * Must check status to be sure of no errors
+	 * If CMD13 is to check the busy completion of the timing change,
+	 * disable the check of CRC error.
+	 */
+	if (index == EXT_CSD_HS_TIMING &&
+	    !(card->host->caps & MMC_CAP_WAIT_WHILE_BUSY))
+		ignore_crc = true;
+
 	timeout = jiffies + msecs_to_jiffies(MMC_OPS_TIMEOUT_MS);
 	do {
-		err = mmc_send_status(card, &status);
-		if (err)
-			return err;
+		if (send_status) {
+			err = __mmc_send_status(card, &status, ignore_crc);
+			if (err)
+				return err;
+		}
 		if (card->host->caps & MMC_CAP_WAIT_WHILE_BUSY)
 			break;
 		if (mmc_host_is_spi(card->host))
 			break;
 
+		/*
+		 * We are not allowed to issue a status command and the host
+		 * does'nt support MMC_CAP_WAIT_WHILE_BUSY, then we can only
+		 * rely on waiting for the stated timeout to be sufficient.
+		 */
+		if (!send_status) {
+			mmc_delay(timeout_ms);
+			return 0;
+		}
+
 		/* Timeout if the device never leaves the program state. */
 		if (time_after(jiffies, timeout)) {
 			pr_err("%s: Card stuck in programming state! %s\n",
@@ -445,36 +501,10 @@ EXPORT_SYMBOL_GPL(__mmc_switch);
 int mmc_switch(struct mmc_card *card, u8 set, u8 index, u8 value,
 		unsigned int timeout_ms)
 {
-	return __mmc_switch(card, set, index, value, timeout_ms, true);
+	return __mmc_switch(card, set, index, value, timeout_ms, true, true);
 }
 EXPORT_SYMBOL_GPL(mmc_switch);
 
-int mmc_send_status(struct mmc_card *card, u32 *status)
-{
-	int err;
-	struct mmc_command cmd = {0};
-
-	BUG_ON(!card);
-	BUG_ON(!card->host);
-
-	cmd.opcode = MMC_SEND_STATUS;
-	if (!mmc_host_is_spi(card->host))
-		cmd.arg = card->rca << 16;
-	cmd.flags = MMC_RSP_SPI_R2 | MMC_RSP_R1 | MMC_CMD_AC;
-
-	err = mmc_wait_for_cmd(card->host, &cmd, MMC_CMD_RETRIES);
-	if (err)
-		return err;
-
-	/* NOTE: callers are required to understand the difference
-	 * between "native" and SPI format status words!
-	 */
-	if (status)
-		*status = cmd.resp[0];
-
-	return 0;
-}
-
 static int
 mmc_send_bus_test(struct mmc_card *card, struct mmc_host *host, u8 opcode,
 		  u8 len)
diff --git a/drivers/mmc/core/sd.c b/drivers/mmc/core/sd.c
index 5e8823dc3ef6..6f42050b7ccc 100644
--- a/drivers/mmc/core/sd.c
+++ b/drivers/mmc/core/sd.c
@@ -13,6 +13,7 @@
 #include <linux/err.h>
 #include <linux/slab.h>
 #include <linux/stat.h>
+#include <linux/pm_runtime.h>
 
 #include <linux/mmc/host.h>
 #include <linux/mmc/card.h>
@@ -721,6 +722,7 @@ int mmc_sd_get_cid(struct mmc_host *host, u32 ocr, u32 *cid, u32 *rocr)
 	int err;
 	u32 max_current;
 	int retries = 10;
+	u32 pocr = ocr;
 
 try_again:
 	if (!retries) {
@@ -773,7 +775,8 @@ try_again:
 	 */
 	if (!mmc_host_is_spi(host) && rocr &&
 	   ((*rocr & 0x41000000) == 0x41000000)) {
-		err = mmc_set_signal_voltage(host, MMC_SIGNAL_VOLTAGE_180);
+		err = mmc_set_signal_voltage(host, MMC_SIGNAL_VOLTAGE_180,
+					pocr);
 		if (err == -EAGAIN) {
 			retries--;
 			goto try_again;
@@ -935,6 +938,7 @@ static int mmc_sd_init_card(struct mmc_host *host, u32 ocr,
 		if (IS_ERR(card))
 			return PTR_ERR(card);
 
+		card->ocr = ocr;
 		card->type = MMC_TYPE_SD;
 		memcpy(card->raw_cid, cid, sizeof(card->raw_cid));
 	}
@@ -1064,10 +1068,7 @@ static void mmc_sd_detect(struct mmc_host *host)
 	}
 }
 
-/*
- * Suspend callback from host.
- */
-static int mmc_sd_suspend(struct mmc_host *host)
+static int _mmc_sd_suspend(struct mmc_host *host)
 {
 	int err = 0;
 
@@ -1075,34 +1076,77 @@ static int mmc_sd_suspend(struct mmc_host *host)
 	BUG_ON(!host->card);
 
 	mmc_claim_host(host);
+
+	if (mmc_card_suspended(host->card))
+		goto out;
+
 	if (!mmc_host_is_spi(host))
 		err = mmc_deselect_cards(host);
 	host->card->state &= ~MMC_STATE_HIGHSPEED;
-	if (!err)
+	if (!err) {
 		mmc_power_off(host);
+		mmc_card_set_suspended(host->card);
+	}
+
+out:
 	mmc_release_host(host);
+	return err;
+}
+
+/*
+ * Callback for suspend
+ */
+static int mmc_sd_suspend(struct mmc_host *host)
+{
+	int err;
+
+	err = _mmc_sd_suspend(host);
+	if (!err) {
+		pm_runtime_disable(&host->card->dev);
+		pm_runtime_set_suspended(&host->card->dev);
+	}
 
 	return err;
 }
 
 /*
- * Resume callback from host.
- *
  * This function tries to determine if the same card is still present
  * and, if so, restore all state to it.
  */
-static int mmc_sd_resume(struct mmc_host *host)
+static int _mmc_sd_resume(struct mmc_host *host)
 {
-	int err;
+	int err = 0;
 
 	BUG_ON(!host);
 	BUG_ON(!host->card);
 
 	mmc_claim_host(host);
-	mmc_power_up(host);
-	mmc_select_voltage(host, host->ocr);
-	err = mmc_sd_init_card(host, host->ocr, host->card);
+
+	if (!mmc_card_suspended(host->card))
+		goto out;
+
+	mmc_power_up(host, host->card->ocr);
+	err = mmc_sd_init_card(host, host->card->ocr, host->card);
+	mmc_card_clr_suspended(host->card);
+
+out:
 	mmc_release_host(host);
+	return err;
+}
+
+/*
+ * Callback for resume
+ */
+static int mmc_sd_resume(struct mmc_host *host)
+{
+	int err = 0;
+
+	if (!(host->caps & MMC_CAP_RUNTIME_RESUME)) {
+		err = _mmc_sd_resume(host);
+		pm_runtime_set_active(&host->card->dev);
+		pm_runtime_mark_last_busy(&host->card->dev);
+	}
+	pm_runtime_enable(&host->card->dev);
 
 	return err;
 }
@@ -1117,18 +1161,11 @@ static int mmc_sd_runtime_suspend(struct mmc_host *host)
 	if (!(host->caps & MMC_CAP_AGGRESSIVE_PM))
 		return 0;
 
-	mmc_claim_host(host);
-
-	err = mmc_sd_suspend(host);
-	if (err) {
+	err = _mmc_sd_suspend(host);
+	if (err)
 		pr_err("%s: error %d doing aggessive suspend\n",
 			mmc_hostname(host), err);
-		goto out;
-	}
-	mmc_power_off(host);
 
-out:
-	mmc_release_host(host);
 	return err;
 }
 
@@ -1139,18 +1176,14 @@ static int mmc_sd_runtime_resume(struct mmc_host *host)
 {
 	int err;
 
-	if (!(host->caps & MMC_CAP_AGGRESSIVE_PM))
+	if (!(host->caps & (MMC_CAP_AGGRESSIVE_PM | MMC_CAP_RUNTIME_RESUME)))
 		return 0;
 
-	mmc_claim_host(host);
-
-	mmc_power_up(host);
-	err = mmc_sd_resume(host);
+	err = _mmc_sd_resume(host);
 	if (err)
 		pr_err("%s: error %d doing aggessive resume\n",
 			mmc_hostname(host), err);
 
-	mmc_release_host(host);
 	return 0;
 }
 
@@ -1160,7 +1193,7 @@ static int mmc_sd_power_restore(struct mmc_host *host)
 
 	host->card->state &= ~MMC_STATE_HIGHSPEED;
 	mmc_claim_host(host);
-	ret = mmc_sd_init_card(host, host->ocr, host->card);
+	ret = mmc_sd_init_card(host, host->card->ocr, host->card);
 	mmc_release_host(host);
 
 	return ret;
@@ -1205,7 +1238,7 @@ static void mmc_sd_attach_bus_ops(struct mmc_host *host)
 int mmc_attach_sd(struct mmc_host *host)
 {
 	int err;
-	u32 ocr;
+	u32 ocr, rocr;
 
 	BUG_ON(!host);
 	WARN_ON(!host->claimed);
@@ -1229,31 +1262,12 @@ int mmc_attach_sd(struct mmc_host *host)
 			goto err;
 	}
 
-	/*
-	 * Sanity check the voltages that the card claims to
-	 * support.
-	 */
-	if (ocr & 0x7F) {
-		pr_warning("%s: card claims to support voltages "
-		       "below the defined range. These will be ignored.\n",
-		       mmc_hostname(host));
-		ocr &= ~0x7F;
-	}
-
-	if ((ocr & MMC_VDD_165_195) &&
-	    !(host->ocr_avail_sd & MMC_VDD_165_195)) {
-		pr_warning("%s: SD card claims to support the "
-		       "incompletely defined 'low voltage range'. This "
-		       "will be ignored.\n", mmc_hostname(host));
-		ocr &= ~MMC_VDD_165_195;
-	}
-
-	host->ocr = mmc_select_voltage(host, ocr);
+	rocr = mmc_select_voltage(host, ocr);
 
 	/*
 	 * Can we support the voltage(s) of the card(s)?
 	 */
-	if (!host->ocr) {
+	if (!rocr) {
 		err = -EINVAL;
 		goto err;
 	}
@@ -1261,7 +1275,7 @@ int mmc_attach_sd(struct mmc_host *host)
 	/*
 	 * Detect and init the card.
 	 */
-	err = mmc_sd_init_card(host, host->ocr, NULL);
+	err = mmc_sd_init_card(host, rocr, NULL);
 	if (err)
 		goto err;
 
diff --git a/drivers/mmc/core/sdio.c b/drivers/mmc/core/sdio.c
index 80d89cff7306..4d721c6e2af0 100644
--- a/drivers/mmc/core/sdio.c
+++ b/drivers/mmc/core/sdio.c
@@ -593,23 +593,28 @@ static int mmc_sdio_init_card(struct mmc_host *host, u32 ocr,
 	struct mmc_card *card;
 	int err;
 	int retries = 10;
+	u32 rocr = 0;
+	u32 ocr_card = ocr;
 
 	BUG_ON(!host);
 	WARN_ON(!host->claimed);
 
+	/* to query card if 1.8V signalling is supported */
+	if (mmc_host_uhs(host))
+		ocr |= R4_18V_PRESENT;
+
 try_again:
 	if (!retries) {
 		pr_warning("%s: Skipping voltage switch\n",
 				mmc_hostname(host));
 		ocr &= ~R4_18V_PRESENT;
-		host->ocr &= ~R4_18V_PRESENT;
 	}
 
 	/*
 	 * Inform the card of the voltage
 	 */
 	if (!powered_resume) {
-		err = mmc_send_io_op_cond(host, host->ocr, &ocr);
+		err = mmc_send_io_op_cond(host, ocr, &rocr);
 		if (err)
 			goto err;
 	}
@@ -632,8 +637,8 @@ try_again:
 		goto err;
 	}
 
-	if ((ocr & R4_MEMORY_PRESENT) &&
-	    mmc_sd_get_cid(host, host->ocr & ocr, card->raw_cid, NULL) == 0) {
+	if ((rocr & R4_MEMORY_PRESENT) &&
+	    mmc_sd_get_cid(host, ocr & rocr, card->raw_cid, NULL) == 0) {
 		card->type = MMC_TYPE_SD_COMBO;
 
 		if (oldcard && (oldcard->type != MMC_TYPE_SD_COMBO ||
@@ -663,8 +668,9 @@ try_again:
 	 * systems that claim 1.8v signalling in fact do not support
 	 * it.
 	 */
-	if (!powered_resume && (ocr & R4_18V_PRESENT) && mmc_host_uhs(host)) {
-		err = mmc_set_signal_voltage(host, MMC_SIGNAL_VOLTAGE_180);
+	if (!powered_resume && (rocr & ocr & R4_18V_PRESENT)) {
+		err = mmc_set_signal_voltage(host, MMC_SIGNAL_VOLTAGE_180,
+					ocr);
 		if (err == -EAGAIN) {
 			sdio_reset(host);
 			mmc_go_idle(host);
@@ -674,12 +680,10 @@ try_again:
 			goto try_again;
 		} else if (err) {
 			ocr &= ~R4_18V_PRESENT;
-			host->ocr &= ~R4_18V_PRESENT;
 		}
 		err = 0;
 	} else {
 		ocr &= ~R4_18V_PRESENT;
-		host->ocr &= ~R4_18V_PRESENT;
 	}
 
 	/*
@@ -759,6 +763,7 @@ try_again:
 
 		card = oldcard;
 	}
+	card->ocr = ocr_card;
 	mmc_fixup_device(card, NULL);
 
 	if (card->type == MMC_TYPE_SD_COMBO) {
@@ -981,8 +986,7 @@ static int mmc_sdio_resume(struct mmc_host *host)
 
 	/* Restore power if needed */
 	if (!mmc_card_keep_power(host)) {
-		mmc_power_up(host);
-		mmc_select_voltage(host, host->ocr);
+		mmc_power_up(host, host->card->ocr);
 		/*
 		 * Tell runtime PM core we just powered up the card,
 		 * since it still believes the card is powered off.
@@ -1000,7 +1004,7 @@ static int mmc_sdio_resume(struct mmc_host *host)
 	if (mmc_card_is_removable(host) || !mmc_card_keep_power(host)) {
 		sdio_reset(host);
 		mmc_go_idle(host);
-		err = mmc_sdio_init_card(host, host->ocr, host->card,
+		err = mmc_sdio_init_card(host, host->card->ocr, host->card,
 					mmc_card_keep_power(host));
 	} else if (mmc_card_keep_power(host) && mmc_card_wake_sdio_irq(host)) {
 		/* We may have switched to 1-bit mode during suspend */
@@ -1040,7 +1044,6 @@ static int mmc_sdio_resume(struct mmc_host *host)
 static int mmc_sdio_power_restore(struct mmc_host *host)
 {
 	int ret;
-	u32 ocr;
 
 	BUG_ON(!host);
 	BUG_ON(!host->card);
@@ -1062,32 +1065,17 @@ static int mmc_sdio_power_restore(struct mmc_host *host)
 	 * for OLPC SD8686 (which expects a [CMD5,5,3,7] init sequence), and
 	 * harmless in other situations.
 	 *
-	 * With these steps taken, mmc_select_voltage() is also required to
-	 * restore the correct voltage setting of the card.
 	 */
 
 	sdio_reset(host);
 	mmc_go_idle(host);
 	mmc_send_if_cond(host, host->ocr_avail);
 
-	ret = mmc_send_io_op_cond(host, 0, &ocr);
+	ret = mmc_send_io_op_cond(host, 0, NULL);
 	if (ret)
 		goto out;
 
-	if (host->ocr_avail_sdio)
-		host->ocr_avail = host->ocr_avail_sdio;
-
-	host->ocr = mmc_select_voltage(host, ocr & ~0x7F);
-	if (!host->ocr) {
-		ret = -EINVAL;
-		goto out;
-	}
-
-	if (mmc_host_uhs(host))
-		/* to query card if 1.8V signalling is supported */
-		host->ocr |= R4_18V_PRESENT;
-
-	ret = mmc_sdio_init_card(host, host->ocr, host->card,
+	ret = mmc_sdio_init_card(host, host->card->ocr, host->card,
 				mmc_card_keep_power(host));
 	if (!ret && host->sdio_irqs)
 		mmc_signal_sdio_irq(host);
@@ -1108,7 +1096,7 @@ static int mmc_sdio_runtime_suspend(struct mmc_host *host)
 static int mmc_sdio_runtime_resume(struct mmc_host *host)
 {
 	/* Restore power and re-initialize. */
-	mmc_power_up(host);
+	mmc_power_up(host, host->card->ocr);
 	return mmc_sdio_power_restore(host);
 }
 
@@ -1131,7 +1119,7 @@ static const struct mmc_bus_ops mmc_sdio_ops = {
 int mmc_attach_sdio(struct mmc_host *host)
 {
 	int err, i, funcs;
-	u32 ocr;
+	u32 ocr, rocr;
 	struct mmc_card *card;
 
 	BUG_ON(!host);
@@ -1145,23 +1133,13 @@ int mmc_attach_sdio(struct mmc_host *host)
 	if (host->ocr_avail_sdio)
 		host->ocr_avail = host->ocr_avail_sdio;
 
-	/*
-	 * Sanity check the voltages that the card claims to
-	 * support.
-	 */
-	if (ocr & 0x7F) {
-		pr_warning("%s: card claims to support voltages "
-		       "below the defined range. These will be ignored.\n",
-		       mmc_hostname(host));
-		ocr &= ~0x7F;
-	}
 
-	host->ocr = mmc_select_voltage(host, ocr);
+	rocr = mmc_select_voltage(host, ocr);
 
 	/*
 	 * Can we support the voltage(s) of the card(s)?
 	 */
-	if (!host->ocr) {
+	if (!rocr) {
 		err = -EINVAL;
 		goto err;
 	}
@@ -1169,22 +1147,10 @@ int mmc_attach_sdio(struct mmc_host *host)
 	/*
 	 * Detect and init the card.
 	 */
-	if (mmc_host_uhs(host))
-		/* to query card if 1.8V signalling is supported */
-		host->ocr |= R4_18V_PRESENT;
+	err = mmc_sdio_init_card(host, rocr, NULL, 0);
+	if (err)
+		goto err;
 
-	err = mmc_sdio_init_card(host, host->ocr, NULL, 0);
-	if (err) {
-		if (err == -EAGAIN) {
-			/*
-			 * Retry initialization with S18R set to 0.
-			 */
-			host->ocr &= ~R4_18V_PRESENT;
-			err = mmc_sdio_init_card(host, host->ocr, NULL, 0);
-		}
-		if (err)
-			goto err;
-	}
 	card = host->card;
 
 	/*
diff --git a/drivers/mmc/host/atmel-mci.c b/drivers/mmc/host/atmel-mci.c
index 69e438ee043e..2cbb4516d353 100644
--- a/drivers/mmc/host/atmel-mci.c
+++ b/drivers/mmc/host/atmel-mci.c
@@ -255,7 +255,6 @@ struct atmel_mci_slot {
 #define ATMCI_CARD_PRESENT	0
 #define ATMCI_CARD_NEED_INIT	1
 #define ATMCI_SHUTDOWN		2
-#define ATMCI_SUSPENDED		3
 
 	int			detect_pin;
 	int			wp_pin;
@@ -589,6 +588,13 @@ static void atmci_timeout_timer(unsigned long data)
 	if (host->mrq->cmd->data) {
 		host->mrq->cmd->data->error = -ETIMEDOUT;
 		host->data = NULL;
+		/*
+		 * With some SDIO modules, sometimes DMA transfer hangs. If
+		 * stop_transfer() is not called then the DMA request is not
+		 * removed, following ones are queued and never computed.
+		 */
+		if (host->state == STATE_DATA_XFER)
+			host->stop_transfer(host);
 	} else {
 		host->mrq->cmd->error = -ETIMEDOUT;
 		host->cmd = NULL;
@@ -1803,12 +1809,14 @@ static void atmci_tasklet_func(unsigned long priv)
 			if (unlikely(status)) {
 				host->stop_transfer(host);
 				host->data = NULL;
-				if (status & ATMCI_DTOE) {
-					data->error = -ETIMEDOUT;
-				} else if (status & ATMCI_DCRCE) {
-					data->error = -EILSEQ;
-				} else {
-					data->error = -EIO;
+				if (data) {
+					if (status & ATMCI_DTOE) {
+						data->error = -ETIMEDOUT;
+					} else if (status & ATMCI_DCRCE) {
+						data->error = -EILSEQ;
+					} else {
+						data->error = -EIO;
+					}
 				}
 			}
 
@@ -2520,70 +2528,10 @@ static int __exit atmci_remove(struct platform_device *pdev)
 	return 0;
 }
 
-#ifdef CONFIG_PM_SLEEP
-static int atmci_suspend(struct device *dev)
-{
-	struct atmel_mci *host = dev_get_drvdata(dev);
-	int i;
-
-	 for (i = 0; i < ATMCI_MAX_NR_SLOTS; i++) {
-		struct atmel_mci_slot *slot = host->slot[i];
-		int ret;
-
-		if (!slot)
-			continue;
-		ret = mmc_suspend_host(slot->mmc);
-		if (ret < 0) {
-			while (--i >= 0) {
-				slot = host->slot[i];
-				if (slot
-				&& test_bit(ATMCI_SUSPENDED, &slot->flags)) {
-					mmc_resume_host(host->slot[i]->mmc);
-					clear_bit(ATMCI_SUSPENDED, &slot->flags);
-				}
-			}
-			return ret;
-		} else {
-			set_bit(ATMCI_SUSPENDED, &slot->flags);
-		}
-	}
-
-	return 0;
-}
-
-static int atmci_resume(struct device *dev)
-{
-	struct atmel_mci *host = dev_get_drvdata(dev);
-	int i;
-	int ret = 0;
-
-	for (i = 0; i < ATMCI_MAX_NR_SLOTS; i++) {
-		struct atmel_mci_slot *slot = host->slot[i];
-		int err;
-
-		slot = host->slot[i];
-		if (!slot)
-			continue;
-		if (!test_bit(ATMCI_SUSPENDED, &slot->flags))
-			continue;
-		err = mmc_resume_host(slot->mmc);
-		if (err < 0)
-			ret = err;
-		else
-			clear_bit(ATMCI_SUSPENDED, &slot->flags);
-	}
-
-	return ret;
-}
-#endif
-
-static SIMPLE_DEV_PM_OPS(atmci_pm, atmci_suspend, atmci_resume);
-
 static struct platform_driver atmci_driver = {
 	.remove		= __exit_p(atmci_remove),
 	.driver		= {
 		.name		= "atmel_mci",
-		.pm		= &atmci_pm,
 		.of_match_table	= of_match_ptr(atmci_dt_ids),
 	},
 };
diff --git a/drivers/mmc/host/au1xmmc.c b/drivers/mmc/host/au1xmmc.c
index df9becdd2e99..f5443a6c4915 100644
--- a/drivers/mmc/host/au1xmmc.c
+++ b/drivers/mmc/host/au1xmmc.c
@@ -1157,11 +1157,6 @@ static int au1xmmc_remove(struct platform_device *pdev)
 static int au1xmmc_suspend(struct platform_device *pdev, pm_message_t state)
 {
 	struct au1xmmc_host *host = platform_get_drvdata(pdev);
-	int ret;
-
-	ret = mmc_suspend_host(host->mmc);
-	if (ret)
-		return ret;
 
 	au_writel(0, HOST_CONFIG2(host));
 	au_writel(0, HOST_CONFIG(host));
@@ -1178,7 +1173,7 @@ static int au1xmmc_resume(struct platform_device *pdev)
 
 	au1xmmc_reset_controller(host);
 
-	return mmc_resume_host(host->mmc);
+	return 0;
 }
 #else
 #define au1xmmc_suspend NULL
diff --git a/drivers/mmc/host/bfin_sdh.c b/drivers/mmc/host/bfin_sdh.c
index 94fae2f1baaf..2b7f37e82ca9 100644
--- a/drivers/mmc/host/bfin_sdh.c
+++ b/drivers/mmc/host/bfin_sdh.c
@@ -391,6 +391,7 @@ static void sdh_set_ios(struct mmc_host *mmc, struct mmc_ios *ios)
 		/* Disable 4 bit SDIO */
 		cfg &= ~SD4E;
 	}
+	bfin_write_SDH_CFG(cfg);
 
 	host->power_mode = ios->power_mode;
 #ifndef RSI_BLKSZ
@@ -415,7 +416,6 @@ static void sdh_set_ios(struct mmc_host *mmc, struct mmc_ios *ios)
 		cfg &= ~SD_CMD_OD;
 # endif
 
-
 	if (ios->power_mode != MMC_POWER_OFF)
 		cfg |= PWR_ON;
 	else
@@ -433,7 +433,6 @@ static void sdh_set_ios(struct mmc_host *mmc, struct mmc_ios *ios)
 		clk_ctl |= CLK_E;
 		host->clk_div = clk_div;
 		bfin_write_SDH_CLK_CTL(clk_ctl);
-
 	} else
 		sdh_stop_clock(host);
 
@@ -640,21 +639,15 @@ static int sdh_remove(struct platform_device *pdev)
 #ifdef CONFIG_PM
 static int sdh_suspend(struct platform_device *dev, pm_message_t state)
 {
-	struct mmc_host *mmc = platform_get_drvdata(dev);
 	struct bfin_sd_host *drv_data = get_sdh_data(dev);
-	int ret = 0;
-
-	if (mmc)
-		ret = mmc_suspend_host(mmc);
 
 	peripheral_free_list(drv_data->pin_req);
 
-	return ret;
+	return 0;
 }
 
 static int sdh_resume(struct platform_device *dev)
 {
-	struct mmc_host *mmc = platform_get_drvdata(dev);
 	struct bfin_sd_host *drv_data = get_sdh_data(dev);
 	int ret = 0;
 
@@ -665,10 +658,6 @@ static int sdh_resume(struct platform_device *dev)
 	}
 
 	sdh_reset();
-
-	if (mmc)
-		ret = mmc_resume_host(mmc);
-
 	return ret;
 }
 #else
diff --git a/drivers/mmc/host/cb710-mmc.c b/drivers/mmc/host/cb710-mmc.c
index 9d6e2b844404..1087b4c79cd6 100644
--- a/drivers/mmc/host/cb710-mmc.c
+++ b/drivers/mmc/host/cb710-mmc.c
@@ -667,12 +667,6 @@ static const struct mmc_host_ops cb710_mmc_host = {
 static int cb710_mmc_suspend(struct platform_device *pdev, pm_message_t state)
 {
 	struct cb710_slot *slot = cb710_pdev_to_slot(pdev);
-	struct mmc_host *mmc = cb710_slot_to_mmc(slot);
-	int err;
-
-	err = mmc_suspend_host(mmc);
-	if (err)
-		return err;
 
 	cb710_mmc_enable_irq(slot, 0, ~0);
 	return 0;
@@ -681,11 +675,9 @@ static int cb710_mmc_suspend(struct platform_device *pdev, pm_message_t state)
 static int cb710_mmc_resume(struct platform_device *pdev)
 {
 	struct cb710_slot *slot = cb710_pdev_to_slot(pdev);
-	struct mmc_host *mmc = cb710_slot_to_mmc(slot);
 
 	cb710_mmc_enable_irq(slot, 0, ~0);
-
-	return mmc_resume_host(mmc);
+	return 0;
 }
 
 #endif /* CONFIG_PM */
diff --git a/drivers/mmc/host/davinci_mmc.c b/drivers/mmc/host/davinci_mmc.c
index e9fa87df909c..d6153740b77f 100644
--- a/drivers/mmc/host/davinci_mmc.c
+++ b/drivers/mmc/host/davinci_mmc.c
@@ -193,7 +193,6 @@ struct mmc_davinci_host {
 #define DAVINCI_MMC_DATADIR_READ	1
 #define DAVINCI_MMC_DATADIR_WRITE	2
 	unsigned char data_dir;
-	unsigned char suspended;
 
 	/* buffer is used during PIO of one scatterlist segment, and
 	 * is updated along with buffer_bytes_left.  bytes_left applies
@@ -1435,38 +1434,23 @@ static int davinci_mmcsd_suspend(struct device *dev)
 {
 	struct platform_device *pdev = to_platform_device(dev);
 	struct mmc_davinci_host *host = platform_get_drvdata(pdev);
-	int ret;
 
-	ret = mmc_suspend_host(host->mmc);
-	if (!ret) {
-		writel(0, host->base + DAVINCI_MMCIM);
-		mmc_davinci_reset_ctrl(host, 1);
-		clk_disable(host->clk);
-		host->suspended = 1;
-	} else {
-		host->suspended = 0;
-	}
+	writel(0, host->base + DAVINCI_MMCIM);
+	mmc_davinci_reset_ctrl(host, 1);
+	clk_disable(host->clk);
 
-	return ret;
+	return 0;
 }
 
 static int davinci_mmcsd_resume(struct device *dev)
 {
 	struct platform_device *pdev = to_platform_device(dev);
 	struct mmc_davinci_host *host = platform_get_drvdata(pdev);
-	int ret;
-
-	if (!host->suspended)
-		return 0;
 
 	clk_enable(host->clk);
-
 	mmc_davinci_reset_ctrl(host, 0);
-	ret = mmc_resume_host(host->mmc);
-	if (!ret)
-		host->suspended = 0;
 
-	return ret;
+	return 0;
 }
 
 static const struct dev_pm_ops davinci_mmcsd_pm = {
diff --git a/drivers/mmc/host/dw_mmc-exynos.c b/drivers/mmc/host/dw_mmc-exynos.c
index 6a1fa2110a05..3423c5ed50c7 100644
--- a/drivers/mmc/host/dw_mmc-exynos.c
+++ b/drivers/mmc/host/dw_mmc-exynos.c
@@ -14,8 +14,10 @@
 #include <linux/clk.h>
 #include <linux/mmc/host.h>
 #include <linux/mmc/dw_mmc.h>
+#include <linux/mmc/mmc.h>
 #include <linux/of.h>
 #include <linux/of_gpio.h>
+#include <linux/slab.h>
 
 #include "dw_mmc.h"
 #include "dw_mmc-pltfm.h"
@@ -30,16 +32,39 @@
 #define SDMMC_CLKSEL_TIMING(x, y, z)	(SDMMC_CLKSEL_CCLK_SAMPLE(x) |	\
 					SDMMC_CLKSEL_CCLK_DRIVE(y) |	\
 					SDMMC_CLKSEL_CCLK_DIVIDER(z))
+#define SDMMC_CLKSEL_WAKEUP_INT		BIT(11)
 
 #define EXYNOS4210_FIXED_CIU_CLK_DIV	2
 #define EXYNOS4412_FIXED_CIU_CLK_DIV	4
 
+/* Block number in eMMC */
+#define DWMCI_BLOCK_NUM		0xFFFFFFFF
+
+#define SDMMC_EMMCP_BASE	0x1000
+#define SDMMC_MPSECURITY	(SDMMC_EMMCP_BASE + 0x0010)
+#define SDMMC_MPSBEGIN0		(SDMMC_EMMCP_BASE + 0x0200)
+#define SDMMC_MPSEND0		(SDMMC_EMMCP_BASE + 0x0204)
+#define SDMMC_MPSCTRL0		(SDMMC_EMMCP_BASE + 0x020C)
+
+/* SMU control bits */
+#define DWMCI_MPSCTRL_SECURE_READ_BIT		BIT(7)
+#define DWMCI_MPSCTRL_SECURE_WRITE_BIT		BIT(6)
+#define DWMCI_MPSCTRL_NON_SECURE_READ_BIT	BIT(5)
+#define DWMCI_MPSCTRL_NON_SECURE_WRITE_BIT	BIT(4)
+#define DWMCI_MPSCTRL_USE_FUSE_KEY		BIT(3)
+#define DWMCI_MPSCTRL_ECB_MODE			BIT(2)
+#define DWMCI_MPSCTRL_ENCRYPTION		BIT(1)
+#define DWMCI_MPSCTRL_VALID			BIT(0)
+
+#define EXYNOS_CCLKIN_MIN	50000000	/* unit: HZ */
+
 /* Variations in Exynos specific dw-mshc controller */
 enum dw_mci_exynos_type {
 	DW_MCI_TYPE_EXYNOS4210,
 	DW_MCI_TYPE_EXYNOS4412,
 	DW_MCI_TYPE_EXYNOS5250,
 	DW_MCI_TYPE_EXYNOS5420,
+	DW_MCI_TYPE_EXYNOS5420_SMU,
 };
 
 /* Exynos implementation specific driver private data */
@@ -48,6 +73,7 @@ struct dw_mci_exynos_priv_data {
 	u8				ciu_div;
 	u32				sdr_timing;
 	u32				ddr_timing;
+	u32				cur_speed;
 };
 
 static struct dw_mci_exynos_compatible {
@@ -66,44 +92,80 @@ static struct dw_mci_exynos_compatible {
 	}, {
 		.compatible	= "samsung,exynos5420-dw-mshc",
 		.ctrl_type	= DW_MCI_TYPE_EXYNOS5420,
+	}, {
+		.compatible	= "samsung,exynos5420-dw-mshc-smu",
+		.ctrl_type	= DW_MCI_TYPE_EXYNOS5420_SMU,
 	},
 };
 
 static int dw_mci_exynos_priv_init(struct dw_mci *host)
 {
-	struct dw_mci_exynos_priv_data *priv;
-	int idx;
-
-	priv = devm_kzalloc(host->dev, sizeof(*priv), GFP_KERNEL);
-	if (!priv) {
-		dev_err(host->dev, "mem alloc failed for private data\n");
-		return -ENOMEM;
-	}
+	struct dw_mci_exynos_priv_data *priv = host->priv;
 
-	for (idx = 0; idx < ARRAY_SIZE(exynos_compat); idx++) {
-		if (of_device_is_compatible(host->dev->of_node,
-					exynos_compat[idx].compatible))
-			priv->ctrl_type = exynos_compat[idx].ctrl_type;
+	if (priv->ctrl_type == DW_MCI_TYPE_EXYNOS5420_SMU) {
+		mci_writel(host, MPSBEGIN0, 0);
+		mci_writel(host, MPSEND0, DWMCI_BLOCK_NUM);
+		mci_writel(host, MPSCTRL0, DWMCI_MPSCTRL_SECURE_WRITE_BIT |
+			   DWMCI_MPSCTRL_NON_SECURE_READ_BIT |
+			   DWMCI_MPSCTRL_VALID |
+			   DWMCI_MPSCTRL_NON_SECURE_WRITE_BIT);
 	}
 
-	host->priv = priv;
 	return 0;
 }
 
 static int dw_mci_exynos_setup_clock(struct dw_mci *host)
 {
 	struct dw_mci_exynos_priv_data *priv = host->priv;
+	unsigned long rate = clk_get_rate(host->ciu_clk);
 
-	if (priv->ctrl_type == DW_MCI_TYPE_EXYNOS5250 ||
-		priv->ctrl_type == DW_MCI_TYPE_EXYNOS5420)
-		host->bus_hz /= (priv->ciu_div + 1);
-	else if (priv->ctrl_type == DW_MCI_TYPE_EXYNOS4412)
-		host->bus_hz /= EXYNOS4412_FIXED_CIU_CLK_DIV;
-	else if (priv->ctrl_type == DW_MCI_TYPE_EXYNOS4210)
-		host->bus_hz /= EXYNOS4210_FIXED_CIU_CLK_DIV;
+	host->bus_hz = rate / (priv->ciu_div + 1);
+	return 0;
+}
+
+#ifdef CONFIG_PM_SLEEP
+static int dw_mci_exynos_suspend(struct device *dev)
+{
+	struct dw_mci *host = dev_get_drvdata(dev);
+
+	return dw_mci_suspend(host);
+}
+
+static int dw_mci_exynos_resume(struct device *dev)
+{
+	struct dw_mci *host = dev_get_drvdata(dev);
+
+	dw_mci_exynos_priv_init(host);
+	return dw_mci_resume(host);
+}
+
+/**
+ * dw_mci_exynos_resume_noirq - Exynos-specific resume code
+ *
+ * On exynos5420 there is a silicon errata that will sometimes leave the
+ * WAKEUP_INT bit in the CLKSEL register asserted.  This bit is 1 to indicate
+ * that it fired and we can clear it by writing a 1 back.  Clear it to prevent
+ * interrupts from going off constantly.
+ *
+ * We run this code on all exynos variants because it doesn't hurt.
+ */
+
+static int dw_mci_exynos_resume_noirq(struct device *dev)
+{
+	struct dw_mci *host = dev_get_drvdata(dev);
+	u32 clksel;
+
+	clksel = mci_readl(host, CLKSEL);
+	if (clksel & SDMMC_CLKSEL_WAKEUP_INT)
+		mci_writel(host, CLKSEL, clksel);
 
 	return 0;
 }
+#else
+#define dw_mci_exynos_suspend		NULL
+#define dw_mci_exynos_resume		NULL
+#define dw_mci_exynos_resume_noirq	NULL
+#endif /* CONFIG_PM_SLEEP */
 
 static void dw_mci_exynos_prepare_command(struct dw_mci *host, u32 *cmdr)
 {
@@ -121,23 +183,68 @@ static void dw_mci_exynos_prepare_command(struct dw_mci *host, u32 *cmdr)
 static void dw_mci_exynos_set_ios(struct dw_mci *host, struct mmc_ios *ios)
 {
 	struct dw_mci_exynos_priv_data *priv = host->priv;
+	unsigned int wanted = ios->clock;
+	unsigned long actual;
+	u8 div = priv->ciu_div + 1;
 
-	if (ios->timing == MMC_TIMING_UHS_DDR50)
+	if (ios->timing == MMC_TIMING_UHS_DDR50) {
 		mci_writel(host, CLKSEL, priv->ddr_timing);
-	else
+		/* Should be double rate for DDR mode */
+		if (ios->bus_width == MMC_BUS_WIDTH_8)
+			wanted <<= 1;
+	} else {
 		mci_writel(host, CLKSEL, priv->sdr_timing);
+	}
+
+	/* Don't care if wanted clock is zero */
+	if (!wanted)
+		return;
+
+	/* Guaranteed minimum frequency for cclkin */
+	if (wanted < EXYNOS_CCLKIN_MIN)
+		wanted = EXYNOS_CCLKIN_MIN;
+
+	if (wanted != priv->cur_speed) {
+		int ret = clk_set_rate(host->ciu_clk, wanted * div);
+		if (ret)
+			dev_warn(host->dev,
+				"failed to set clk-rate %u error: %d\n",
+				 wanted * div, ret);
+		actual = clk_get_rate(host->ciu_clk);
+		host->bus_hz = actual / div;
+		priv->cur_speed = wanted;
+		host->current_speed = 0;
+	}
 }
 
 static int dw_mci_exynos_parse_dt(struct dw_mci *host)
 {
-	struct dw_mci_exynos_priv_data *priv = host->priv;
+	struct dw_mci_exynos_priv_data *priv;
 	struct device_node *np = host->dev->of_node;
 	u32 timing[2];
 	u32 div = 0;
+	int idx;
 	int ret;
 
-	of_property_read_u32(np, "samsung,dw-mshc-ciu-div", &div);
-	priv->ciu_div = div;
+	priv = devm_kzalloc(host->dev, sizeof(*priv), GFP_KERNEL);
+	if (!priv) {
+		dev_err(host->dev, "mem alloc failed for private data\n");
+		return -ENOMEM;
+	}
+
+	for (idx = 0; idx < ARRAY_SIZE(exynos_compat); idx++) {
+		if (of_device_is_compatible(np, exynos_compat[idx].compatible))
+			priv->ctrl_type = exynos_compat[idx].ctrl_type;
+	}
+
+	if (priv->ctrl_type == DW_MCI_TYPE_EXYNOS4412)
+		priv->ciu_div = EXYNOS4412_FIXED_CIU_CLK_DIV - 1;
+	else if (priv->ctrl_type == DW_MCI_TYPE_EXYNOS4210)
+		priv->ciu_div = EXYNOS4210_FIXED_CIU_CLK_DIV - 1;
+	else {
+		of_property_read_u32(np, "samsung,dw-mshc-ciu-div", &div);
+		priv->ciu_div = div;
+	}
 
 	ret = of_property_read_u32_array(np,
 			"samsung,dw-mshc-sdr-timing", timing, 2);
@@ -152,9 +259,131 @@ static int dw_mci_exynos_parse_dt(struct dw_mci *host)
 		return ret;
 
 	priv->ddr_timing = SDMMC_CLKSEL_TIMING(timing[0], timing[1], div);
+	host->priv = priv;
 	return 0;
 }
 
+static inline u8 dw_mci_exynos_get_clksmpl(struct dw_mci *host)
+{
+	return SDMMC_CLKSEL_CCLK_SAMPLE(mci_readl(host, CLKSEL));
+}
+
+static inline void dw_mci_exynos_set_clksmpl(struct dw_mci *host, u8 sample)
+{
+	u32 clksel;
+	clksel = mci_readl(host, CLKSEL);
+	clksel = (clksel & ~0x7) | SDMMC_CLKSEL_CCLK_SAMPLE(sample);
+	mci_writel(host, CLKSEL, clksel);
+}
+
+static inline u8 dw_mci_exynos_move_next_clksmpl(struct dw_mci *host)
+{
+	u32 clksel;
+	u8 sample;
+
+	clksel = mci_readl(host, CLKSEL);
+	sample = (clksel + 1) & 0x7;
+	clksel = (clksel & ~0x7) | sample;
+	mci_writel(host, CLKSEL, clksel);
+	return sample;
+}
+
+static s8 dw_mci_exynos_get_best_clksmpl(u8 candiates)
+{
+	const u8 iter = 8;
+	u8 __c;
+	s8 i, loc = -1;
+
+	for (i = 0; i < iter; i++) {
+		__c = ror8(candiates, i);
+		if ((__c & 0xc7) == 0xc7) {
+			loc = i;
+			goto out;
+		}
+	}
+
+	for (i = 0; i < iter; i++) {
+		__c = ror8(candiates, i);
+		if ((__c & 0x83) == 0x83) {
+			loc = i;
+			goto out;
+		}
+	}
+
+out:
+	return loc;
+}
+
+static int dw_mci_exynos_execute_tuning(struct dw_mci_slot *slot, u32 opcode,
+					struct dw_mci_tuning_data *tuning_data)
+{
+	struct dw_mci *host = slot->host;
+	struct mmc_host *mmc = slot->mmc;
+	const u8 *blk_pattern = tuning_data->blk_pattern;
+	u8 *blk_test;
+	unsigned int blksz = tuning_data->blksz;
+	u8 start_smpl, smpl, candiates = 0;
+	s8 found = -1;
+	int ret = 0;
+
+	blk_test = kmalloc(blksz, GFP_KERNEL);
+	if (!blk_test)
+		return -ENOMEM;
+
+	start_smpl = dw_mci_exynos_get_clksmpl(host);
+
+	do {
+		struct mmc_request mrq = {NULL};
+		struct mmc_command cmd = {0};
+		struct mmc_command stop = {0};
+		struct mmc_data data = {0};
+		struct scatterlist sg;
+
+		cmd.opcode = opcode;
+		cmd.arg = 0;
+		cmd.flags = MMC_RSP_R1 | MMC_CMD_ADTC;
+
+		stop.opcode = MMC_STOP_TRANSMISSION;
+		stop.arg = 0;
+		stop.flags = MMC_RSP_R1B | MMC_CMD_AC;
+
+		data.blksz = blksz;
+		data.blocks = 1;
+		data.flags = MMC_DATA_READ;
+		data.sg = &sg;
+		data.sg_len = 1;
+
+		sg_init_one(&sg, blk_test, blksz);
+		mrq.cmd = &cmd;
+		mrq.stop = &stop;
+		mrq.data = &data;
+		host->mrq = &mrq;
+
+		mci_writel(host, TMOUT, ~0);
+		smpl = dw_mci_exynos_move_next_clksmpl(host);
+
+		mmc_wait_for_req(mmc, &mrq);
+
+		if (!cmd.error && !data.error) {
+			if (!memcmp(blk_pattern, blk_test, blksz))
+				candiates |= (1 << smpl);
+		} else {
+			dev_dbg(host->dev,
+				"Tuning error: cmd.error:%d, data.error:%d\n",
+				cmd.error, data.error);
+		}
+	} while (start_smpl != smpl);
+
+	found = dw_mci_exynos_get_best_clksmpl(candiates);
+	if (found >= 0)
+		dw_mci_exynos_set_clksmpl(host, found);
+	else
+		ret = -EIO;
+
+	kfree(blk_test);
+	return ret;
+}
+
 /* Common capabilities of Exynos4/Exynos5 SoC */
 static unsigned long exynos_dwmmc_caps[4] = {
 	MMC_CAP_UHS_DDR50 | MMC_CAP_1_8V_DDR |
@@ -171,6 +400,7 @@ static const struct dw_mci_drv_data exynos_drv_data = {
 	.prepare_command	= dw_mci_exynos_prepare_command,
 	.set_ios		= dw_mci_exynos_set_ios,
 	.parse_dt		= dw_mci_exynos_parse_dt,
+	.execute_tuning		= dw_mci_exynos_execute_tuning,
 };
 
 static const struct of_device_id dw_mci_exynos_match[] = {
@@ -180,6 +410,8 @@ static const struct of_device_id dw_mci_exynos_match[] = {
 			.data = &exynos_drv_data, },
 	{ .compatible = "samsung,exynos5420-dw-mshc",
 			.data = &exynos_drv_data, },
+	{ .compatible = "samsung,exynos5420-dw-mshc-smu",
+			.data = &exynos_drv_data, },
 	{},
 };
 MODULE_DEVICE_TABLE(of, dw_mci_exynos_match);
@@ -194,13 +426,20 @@ static int dw_mci_exynos_probe(struct platform_device *pdev)
 	return dw_mci_pltfm_register(pdev, drv_data);
 }
 
+const struct dev_pm_ops dw_mci_exynos_pmops = {
+	SET_SYSTEM_SLEEP_PM_OPS(dw_mci_exynos_suspend, dw_mci_exynos_resume)
+	.resume_noirq = dw_mci_exynos_resume_noirq,
+	.thaw_noirq = dw_mci_exynos_resume_noirq,
+	.restore_noirq = dw_mci_exynos_resume_noirq,
+};
+
 static struct platform_driver dw_mci_exynos_pltfm_driver = {
 	.probe		= dw_mci_exynos_probe,
 	.remove		= __exit_p(dw_mci_pltfm_remove),
 	.driver		= {
 		.name		= "dwmmc_exynos",
 		.of_match_table	= dw_mci_exynos_match,
-		.pm		= &dw_mci_pltfm_pmops,
+		.pm		= &dw_mci_exynos_pmops,
 	},
 };
 
diff --git a/drivers/mmc/host/dw_mmc-pltfm.c b/drivers/mmc/host/dw_mmc-pltfm.c
index 20897529ea5e..5c4965655297 100644
--- a/drivers/mmc/host/dw_mmc-pltfm.c
+++ b/drivers/mmc/host/dw_mmc-pltfm.c
@@ -39,7 +39,6 @@ int dw_mci_pltfm_register(struct platform_device *pdev,
 {
 	struct dw_mci *host;
 	struct resource	*regs;
-	int ret;
 
 	host = devm_kzalloc(&pdev->dev, sizeof(struct dw_mci), GFP_KERNEL);
 	if (!host)
@@ -59,12 +58,6 @@ int dw_mci_pltfm_register(struct platform_device *pdev,
 	if (IS_ERR(host->regs))
 		return PTR_ERR(host->regs);
 
-	if (drv_data && drv_data->init) {
-		ret = drv_data->init(host);
-		if (ret)
-			return ret;
-	}
-
 	platform_set_drvdata(pdev, host);
 	return dw_mci_probe(host);
 }
diff --git a/drivers/mmc/host/dw_mmc-socfpga.c b/drivers/mmc/host/dw_mmc-socfpga.c
index 14b5961a851c..3e8e53ae3302 100644
--- a/drivers/mmc/host/dw_mmc-socfpga.c
+++ b/drivers/mmc/host/dw_mmc-socfpga.c
@@ -38,21 +38,6 @@ struct dw_mci_socfpga_priv_data {
 
 static int dw_mci_socfpga_priv_init(struct dw_mci *host)
 {
-	struct dw_mci_socfpga_priv_data *priv;
-
-	priv = devm_kzalloc(host->dev, sizeof(*priv), GFP_KERNEL);
-	if (!priv) {
-		dev_err(host->dev, "mem alloc failed for private data\n");
-		return -ENOMEM;
-	}
-
-	priv->sysreg = syscon_regmap_lookup_by_compatible("altr,sys-mgr");
-	if (IS_ERR(priv->sysreg)) {
-		dev_err(host->dev, "regmap for altr,sys-mgr lookup failed.\n");
-		return PTR_ERR(priv->sysreg);
-	}
-	host->priv = priv;
-
 	return 0;
 }
 
@@ -79,12 +64,24 @@ static void dw_mci_socfpga_prepare_command(struct dw_mci *host, u32 *cmdr)
 
 static int dw_mci_socfpga_parse_dt(struct dw_mci *host)
 {
-	struct dw_mci_socfpga_priv_data *priv = host->priv;
+	struct dw_mci_socfpga_priv_data *priv;
 	struct device_node *np = host->dev->of_node;
 	u32 timing[2];
 	u32 div = 0;
 	int ret;
 
+	priv = devm_kzalloc(host->dev, sizeof(*priv), GFP_KERNEL);
+	if (!priv) {
+		dev_err(host->dev, "mem alloc failed for private data\n");
+		return -ENOMEM;
+	}
+
+	priv->sysreg = syscon_regmap_lookup_by_compatible("altr,sys-mgr");
+	if (IS_ERR(priv->sysreg)) {
+		dev_err(host->dev, "regmap for altr,sys-mgr lookup failed.\n");
+		return PTR_ERR(priv->sysreg);
+	}
+
 	ret = of_property_read_u32(np, "altr,dw-mshc-ciu-div", &div);
 	if (ret)
 		dev_info(host->dev, "No dw-mshc-ciu-div specified, assuming 1");
@@ -96,6 +93,7 @@ static int dw_mci_socfpga_parse_dt(struct dw_mci *host)
 		return ret;
 
 	priv->hs_timing = SYSMGR_SDMMC_CTRL_SET(timing[0], timing[1]);
+	host->priv = priv;
 	return 0;
 }
 
@@ -113,7 +111,7 @@ static const struct of_device_id dw_mci_socfpga_match[] = {
 };
 MODULE_DEVICE_TABLE(of, dw_mci_socfpga_match);
 
-int dw_mci_socfpga_probe(struct platform_device *pdev)
+static int dw_mci_socfpga_probe(struct platform_device *pdev)
 {
 	const struct dw_mci_drv_data *drv_data;
 	const struct of_device_id *match;
@@ -128,7 +126,7 @@ static struct platform_driver dw_mci_socfpga_pltfm_driver = {
 	.remove		= __exit_p(dw_mci_pltfm_remove),
 	.driver		= {
 		.name		= "dwmmc_socfpga",
-		.of_match_table	= of_match_ptr(dw_mci_socfpga_match),
+		.of_match_table	= dw_mci_socfpga_match,
 		.pm		= &dw_mci_pltfm_pmops,
 	},
 };
diff --git a/drivers/mmc/host/dw_mmc.c b/drivers/mmc/host/dw_mmc.c
index 018f365e5ae4..4bce0deec362 100644
--- a/drivers/mmc/host/dw_mmc.c
+++ b/drivers/mmc/host/dw_mmc.c
@@ -29,6 +29,7 @@
 #include <linux/irq.h>
 #include <linux/mmc/host.h>
 #include <linux/mmc/mmc.h>
+#include <linux/mmc/sdio.h>
 #include <linux/mmc/dw_mmc.h>
 #include <linux/bitops.h>
 #include <linux/regulator/consumer.h>
@@ -50,6 +51,9 @@
 #define DW_MCI_RECV_STATUS	2
 #define DW_MCI_DMA_THRESHOLD	16
 
+#define DW_MCI_FREQ_MAX	200000000	/* unit: HZ */
+#define DW_MCI_FREQ_MIN	400000		/* unit: HZ */
+
 #ifdef CONFIG_MMC_DW_IDMAC
 #define IDMAC_INT_CLR		(SDMMC_IDMAC_INT_AI | SDMMC_IDMAC_INT_NI | \
 				 SDMMC_IDMAC_INT_CES | SDMMC_IDMAC_INT_DU | \
@@ -76,42 +80,39 @@ struct idmac_desc {
 };
 #endif /* CONFIG_MMC_DW_IDMAC */
 
-/**
- * struct dw_mci_slot - MMC slot state
- * @mmc: The mmc_host representing this slot.
- * @host: The MMC controller this slot is using.
- * @quirks: Slot-level quirks (DW_MCI_SLOT_QUIRK_XXX)
- * @wp_gpio: If gpio_is_valid() we'll use this to read write protect.
- * @ctype: Card type for this slot.
- * @mrq: mmc_request currently being processed or waiting to be
- *	processed, or NULL when the slot is idle.
- * @queue_node: List node for placing this node in the @queue list of
- *	&struct dw_mci.
- * @clock: Clock rate configured by set_ios(). Protected by host->lock.
- * @flags: Random state bits associated with the slot.
- * @id: Number of this slot.
- * @last_detect_state: Most recently observed card detect state.
- */
-struct dw_mci_slot {
-	struct mmc_host		*mmc;
-	struct dw_mci		*host;
-
-	int			quirks;
-	int			wp_gpio;
-
-	u32			ctype;
-
-	struct mmc_request	*mrq;
-	struct list_head	queue_node;
+static const u8 tuning_blk_pattern_4bit[] = {
+	0xff, 0x0f, 0xff, 0x00, 0xff, 0xcc, 0xc3, 0xcc,
+	0xc3, 0x3c, 0xcc, 0xff, 0xfe, 0xff, 0xfe, 0xef,
+	0xff, 0xdf, 0xff, 0xdd, 0xff, 0xfb, 0xff, 0xfb,
+	0xbf, 0xff, 0x7f, 0xff, 0x77, 0xf7, 0xbd, 0xef,
+	0xff, 0xf0, 0xff, 0xf0, 0x0f, 0xfc, 0xcc, 0x3c,
+	0xcc, 0x33, 0xcc, 0xcf, 0xff, 0xef, 0xff, 0xee,
+	0xff, 0xfd, 0xff, 0xfd, 0xdf, 0xff, 0xbf, 0xff,
+	0xbb, 0xff, 0xf7, 0xff, 0xf7, 0x7f, 0x7b, 0xde,
+};
 
-	unsigned int		clock;
-	unsigned long		flags;
-#define DW_MMC_CARD_PRESENT	0
-#define DW_MMC_CARD_NEED_INIT	1
-	int			id;
-	int			last_detect_state;
+static const u8 tuning_blk_pattern_8bit[] = {
+	0xff, 0xff, 0x00, 0xff, 0xff, 0xff, 0x00, 0x00,
+	0xff, 0xff, 0xcc, 0xcc, 0xcc, 0x33, 0xcc, 0xcc,
+	0xcc, 0x33, 0x33, 0xcc, 0xcc, 0xcc, 0xff, 0xff,
+	0xff, 0xee, 0xff, 0xff, 0xff, 0xee, 0xee, 0xff,
+	0xff, 0xff, 0xdd, 0xff, 0xff, 0xff, 0xdd, 0xdd,
+	0xff, 0xff, 0xff, 0xbb, 0xff, 0xff, 0xff, 0xbb,
+	0xbb, 0xff, 0xff, 0xff, 0x77, 0xff, 0xff, 0xff,
+	0x77, 0x77, 0xff, 0x77, 0xbb, 0xdd, 0xee, 0xff,
+	0xff, 0xff, 0xff, 0x00, 0xff, 0xff, 0xff, 0x00,
+	0x00, 0xff, 0xff, 0xcc, 0xcc, 0xcc, 0x33, 0xcc,
+	0xcc, 0xcc, 0x33, 0x33, 0xcc, 0xcc, 0xcc, 0xff,
+	0xff, 0xff, 0xee, 0xff, 0xff, 0xff, 0xee, 0xee,
+	0xff, 0xff, 0xff, 0xdd, 0xff, 0xff, 0xff, 0xdd,
+	0xdd, 0xff, 0xff, 0xff, 0xbb, 0xff, 0xff, 0xff,
+	0xbb, 0xbb, 0xff, 0xff, 0xff, 0x77, 0xff, 0xff,
+	0xff, 0x77, 0x77, 0xff, 0x77, 0xbb, 0xdd, 0xee,
 };
 
+static inline bool dw_mci_fifo_reset(struct dw_mci *host);
+static inline bool dw_mci_ctrl_all_reset(struct dw_mci *host);
+
 #if defined(CONFIG_DEBUG_FS)
 static int dw_mci_req_show(struct seq_file *s, void *v)
 {
@@ -249,10 +250,15 @@ static u32 dw_mci_prepare_command(struct mmc_host *mmc, struct mmc_command *cmd)
 
 	cmdr = cmd->opcode;
 
-	if (cmdr == MMC_STOP_TRANSMISSION)
+	if (cmd->opcode == MMC_STOP_TRANSMISSION ||
+	    cmd->opcode == MMC_GO_IDLE_STATE ||
+	    cmd->opcode == MMC_GO_INACTIVE_STATE ||
+	    (cmd->opcode == SD_IO_RW_DIRECT &&
+	     ((cmd->arg >> 9) & 0x1FFFF) == SDIO_CCCR_ABORT))
 		cmdr |= SDMMC_CMD_STOP;
 	else
-		cmdr |= SDMMC_CMD_PRV_DAT_WAIT;
+		if (cmd->opcode != MMC_SEND_STATUS && cmd->data)
+			cmdr |= SDMMC_CMD_PRV_DAT_WAIT;
 
 	if (cmd->flags & MMC_RSP_PRESENT) {
 		/* We expect a response, so set this bit */
@@ -279,6 +285,40 @@ static u32 dw_mci_prepare_command(struct mmc_host *mmc, struct mmc_command *cmd)
 	return cmdr;
 }
 
+static u32 dw_mci_prep_stop_abort(struct dw_mci *host, struct mmc_command *cmd)
+{
+	struct mmc_command *stop;
+	u32 cmdr;
+
+	if (!cmd->data)
+		return 0;
+
+	stop = &host->stop_abort;
+	cmdr = cmd->opcode;
+	memset(stop, 0, sizeof(struct mmc_command));
+
+	if (cmdr == MMC_READ_SINGLE_BLOCK ||
+	    cmdr == MMC_READ_MULTIPLE_BLOCK ||
+	    cmdr == MMC_WRITE_BLOCK ||
+	    cmdr == MMC_WRITE_MULTIPLE_BLOCK) {
+		stop->opcode = MMC_STOP_TRANSMISSION;
+		stop->arg = 0;
+		stop->flags = MMC_RSP_R1B | MMC_CMD_AC;
+	} else if (cmdr == SD_IO_RW_EXTENDED) {
+		stop->opcode = SD_IO_RW_DIRECT;
+		stop->arg |= (1 << 31) | (0 << 28) | (SDIO_CCCR_ABORT << 9) |
+			     ((cmd->arg >> 28) & 0x7);
+		stop->flags = MMC_RSP_SPI_R5 | MMC_RSP_R5 | MMC_CMD_AC;
+	} else {
+		return 0;
+	}
+
+	cmdr = stop->opcode | SDMMC_CMD_STOP |
+		SDMMC_CMD_RESP_CRC | SDMMC_CMD_RESP_EXP;
+
+	return cmdr;
+}
+
 static void dw_mci_start_command(struct dw_mci *host,
 				 struct mmc_command *cmd, u32 cmd_flags)
 {
@@ -293,9 +333,10 @@ static void dw_mci_start_command(struct dw_mci *host,
 	mci_writel(host, CMD, cmd_flags | SDMMC_CMD_START);
 }
 
-static void send_stop_cmd(struct dw_mci *host, struct mmc_data *data)
+static inline void send_stop_abort(struct dw_mci *host, struct mmc_data *data)
 {
-	dw_mci_start_command(host, data->stop, host->stop_cmdr);
+	struct mmc_command *stop = data->stop ? data->stop : &host->stop_abort;
+	dw_mci_start_command(host, stop, host->stop_cmdr);
 }
 
 /* DMA interface functions */
@@ -304,10 +345,10 @@ static void dw_mci_stop_dma(struct dw_mci *host)
 	if (host->using_dma) {
 		host->dma_ops->stop(host);
 		host->dma_ops->cleanup(host);
-	} else {
-		/* Data transfer was stopped by the interrupt handler */
-		set_bit(EVENT_XFER_COMPLETE, &host->pending_events);
 	}
+
+	/* Data transfer was stopped by the interrupt handler */
+	set_bit(EVENT_XFER_COMPLETE, &host->pending_events);
 }
 
 static int dw_mci_get_dma_dir(struct mmc_data *data)
@@ -331,6 +372,14 @@ static void dw_mci_dma_cleanup(struct dw_mci *host)
 				     dw_mci_get_dma_dir(data));
 }
 
+static void dw_mci_idmac_reset(struct dw_mci *host)
+{
+	u32 bmod = mci_readl(host, BMOD);
+	/* Software reset of DMA */
+	bmod |= SDMMC_IDMAC_SWRESET;
+	mci_writel(host, BMOD, bmod);
+}
+
 static void dw_mci_idmac_stop_dma(struct dw_mci *host)
 {
 	u32 temp;
@@ -344,6 +393,7 @@ static void dw_mci_idmac_stop_dma(struct dw_mci *host)
 	/* Stop the IDMAC running */
 	temp = mci_readl(host, BMOD);
 	temp &= ~(SDMMC_IDMAC_ENABLE | SDMMC_IDMAC_FB);
+	temp |= SDMMC_IDMAC_SWRESET;
 	mci_writel(host, BMOD, temp);
 }
 
@@ -435,7 +485,7 @@ static int dw_mci_idmac_init(struct dw_mci *host)
 	p->des3 = host->sg_dma;
 	p->des0 = IDMAC_DES0_ER;
 
-	mci_writel(host, BMOD, SDMMC_IDMAC_SWRESET);
+	dw_mci_idmac_reset(host);
 
 	/* Mask out interrupts - get Tx & Rx complete only */
 	mci_writel(host, IDSTS, IDMAC_INT_CLR);
@@ -532,6 +582,78 @@ static void dw_mci_post_req(struct mmc_host *mmc,
 	data->host_cookie = 0;
 }
 
+static void dw_mci_adjust_fifoth(struct dw_mci *host, struct mmc_data *data)
+{
+#ifdef CONFIG_MMC_DW_IDMAC
+	unsigned int blksz = data->blksz;
+	const u32 mszs[] = {1, 4, 8, 16, 32, 64, 128, 256};
+	u32 fifo_width = 1 << host->data_shift;
+	u32 blksz_depth = blksz / fifo_width, fifoth_val;
+	u32 msize = 0, rx_wmark = 1, tx_wmark, tx_wmark_invers;
+	int idx = (sizeof(mszs) / sizeof(mszs[0])) - 1;
+
+	tx_wmark = (host->fifo_depth) / 2;
+	tx_wmark_invers = host->fifo_depth - tx_wmark;
+
+	/*
+	 * MSIZE is '1',
+	 * if blksz is not a multiple of the FIFO width
+	 */
+	if (blksz % fifo_width) {
+		msize = 0;
+		rx_wmark = 1;
+		goto done;
+	}
+
+	do {
+		if (!((blksz_depth % mszs[idx]) ||
+		     (tx_wmark_invers % mszs[idx]))) {
+			msize = idx;
+			rx_wmark = mszs[idx] - 1;
+			break;
+		}
+	} while (--idx > 0);
+	/*
+	 * If idx is '0', it won't be tried
+	 * Thus, initial values are uesed
+	 */
+done:
+	fifoth_val = SDMMC_SET_FIFOTH(msize, rx_wmark, tx_wmark);
+	mci_writel(host, FIFOTH, fifoth_val);
+#endif
+}
+
+static void dw_mci_ctrl_rd_thld(struct dw_mci *host, struct mmc_data *data)
+{
+	unsigned int blksz = data->blksz;
+	u32 blksz_depth, fifo_depth;
+	u16 thld_size;
+
+	WARN_ON(!(data->flags & MMC_DATA_READ));
+
+	if (host->timing != MMC_TIMING_MMC_HS200 &&
+	    host->timing != MMC_TIMING_UHS_SDR104)
+		goto disable;
+
+	blksz_depth = blksz / (1 << host->data_shift);
+	fifo_depth = host->fifo_depth;
+
+	if (blksz_depth > fifo_depth)
+		goto disable;
+
+	/*
+	 * If (blksz_depth) >= (fifo_depth >> 1), should be 'thld_size <= blksz'
+	 * If (blksz_depth) <  (fifo_depth >> 1), should be thld_size = blksz
+	 * Currently just choose blksz.
+	 */
+	thld_size = blksz;
+	mci_writel(host, CDTHRCTL, SDMMC_SET_RD_THLD(thld_size, 1));
+	return;
+
+disable:
+	mci_writel(host, CDTHRCTL, SDMMC_SET_RD_THLD(0, 0));
+}
+
 static int dw_mci_submit_data_dma(struct dw_mci *host, struct mmc_data *data)
 {
 	int sg_len;
@@ -556,6 +678,14 @@ static int dw_mci_submit_data_dma(struct dw_mci *host, struct mmc_data *data)
 		 (unsigned long)host->sg_cpu, (unsigned long)host->sg_dma,
 		 sg_len);
 
+	/*
+	 * Decide the MSIZE and RX/TX Watermark.
+	 * If current block size is same with previous size,
+	 * no need to update fifoth.
+	 */
+	if (host->prev_blksz != data->blksz)
+		dw_mci_adjust_fifoth(host, data);
+
 	/* Enable the DMA interface */
 	temp = mci_readl(host, CTRL);
 	temp |= SDMMC_CTRL_DMA_ENABLE;
@@ -581,10 +711,12 @@ static void dw_mci_submit_data(struct dw_mci *host, struct mmc_data *data)
 	host->sg = NULL;
 	host->data = data;
 
-	if (data->flags & MMC_DATA_READ)
+	if (data->flags & MMC_DATA_READ) {
 		host->dir_status = DW_MCI_RECV_STATUS;
-	else
+		dw_mci_ctrl_rd_thld(host, data);
+	} else {
 		host->dir_status = DW_MCI_SEND_STATUS;
+	}
 
 	if (dw_mci_submit_data_dma(host, data)) {
 		int flags = SG_MITER_ATOMIC;
@@ -606,6 +738,21 @@ static void dw_mci_submit_data(struct dw_mci *host, struct mmc_data *data)
 		temp = mci_readl(host, CTRL);
 		temp &= ~SDMMC_CTRL_DMA_ENABLE;
 		mci_writel(host, CTRL, temp);
+
+		/*
+		 * Use the initial fifoth_val for PIO mode.
+		 * If next issued data may be transfered by DMA mode,
+		 * prev_blksz should be invalidated.
+		 */
+		mci_writel(host, FIFOTH, host->fifoth_val);
+		host->prev_blksz = 0;
+	} else {
+		/*
+		 * Keep the current block size.
+		 * It will be used to decide whether to update
+		 * fifoth register next time.
+		 */
+		host->prev_blksz = data->blksz;
 	}
 }
 
@@ -632,24 +779,31 @@ static void mci_send_cmd(struct dw_mci_slot *slot, u32 cmd, u32 arg)
 static void dw_mci_setup_bus(struct dw_mci_slot *slot, bool force_clkinit)
 {
 	struct dw_mci *host = slot->host;
+	unsigned int clock = slot->clock;
 	u32 div;
 	u32 clk_en_a;
 
-	if (slot->clock != host->current_speed || force_clkinit) {
-		div = host->bus_hz / slot->clock;
-		if (host->bus_hz % slot->clock && host->bus_hz > slot->clock)
+	if (!clock) {
+		mci_writel(host, CLKENA, 0);
+		mci_send_cmd(slot,
+			     SDMMC_CMD_UPD_CLK | SDMMC_CMD_PRV_DAT_WAIT, 0);
+	} else if (clock != host->current_speed || force_clkinit) {
+		div = host->bus_hz / clock;
+		if (host->bus_hz % clock && host->bus_hz > clock)
 			/*
 			 * move the + 1 after the divide to prevent
 			 * over-clocking the card.
 			 */
 			div += 1;
 
-		div = (host->bus_hz != slot->clock) ? DIV_ROUND_UP(div, 2) : 0;
+		div = (host->bus_hz != clock) ? DIV_ROUND_UP(div, 2) : 0;
 
-		dev_info(&slot->mmc->class_dev,
-			 "Bus speed (slot %d) = %dHz (slot req %dHz, actual %dHZ"
-			 " div = %d)\n", slot->id, host->bus_hz, slot->clock,
-			 div ? ((host->bus_hz / div) >> 1) : host->bus_hz, div);
+		if ((clock << div) != slot->__clk_old || force_clkinit)
+			dev_info(&slot->mmc->class_dev,
+				 "Bus speed (slot %d) = %dHz (slot req %dHz, actual %dHZ div = %d)\n",
+				 slot->id, host->bus_hz, clock,
+				 div ? ((host->bus_hz / div) >> 1) :
+				 host->bus_hz, div);
 
 		/* disable clock */
 		mci_writel(host, CLKENA, 0);
@@ -676,9 +830,12 @@ static void dw_mci_setup_bus(struct dw_mci_slot *slot, bool force_clkinit)
 		mci_send_cmd(slot,
 			     SDMMC_CMD_UPD_CLK | SDMMC_CMD_PRV_DAT_WAIT, 0);
 
-		host->current_speed = slot->clock;
+		/* keep the clock with reflecting clock dividor */
+		slot->__clk_old = clock << div;
 	}
 
+	host->current_speed = clock;
+
 	/* Set the current slot bus width */
 	mci_writel(host, CTYPE, (slot->ctype << slot->id));
 }
@@ -700,7 +857,9 @@ static void __dw_mci_start_request(struct dw_mci *host,
 
 	host->pending_events = 0;
 	host->completed_events = 0;
+	host->cmd_status = 0;
 	host->data_status = 0;
+	host->dir_status = 0;
 
 	data = cmd->data;
 	if (data) {
@@ -724,6 +883,8 @@ static void __dw_mci_start_request(struct dw_mci *host,
 
 	if (mrq->stop)
 		host->stop_cmdr = dw_mci_prepare_command(slot->mmc, mrq->stop);
+	else
+		host->stop_cmdr = dw_mci_prep_stop_abort(host, cmd);
 }
 
 static void dw_mci_start_request(struct dw_mci *host,
@@ -806,14 +967,13 @@ static void dw_mci_set_ios(struct mmc_host *mmc, struct mmc_ios *ios)
 		regs &= ~((0x1 << slot->id) << 16);
 
 	mci_writel(slot->host, UHS_REG, regs);
+	slot->host->timing = ios->timing;
 
-	if (ios->clock) {
-		/*
-		 * Use mirror of ios->clock to prevent race with mmc
-		 * core ios update when finding the minimum.
-		 */
-		slot->clock = ios->clock;
-	}
+	/*
+	 * Use mirror of ios->clock to prevent race with mmc
+	 * core ios update when finding the minimum.
+	 */
+	slot->clock = ios->clock;
 
 	if (drv_data && drv_data->set_ios)
 		drv_data->set_ios(slot->host, ios);
@@ -939,6 +1099,38 @@ static void dw_mci_enable_sdio_irq(struct mmc_host *mmc, int enb)
 	}
 }
 
+static int dw_mci_execute_tuning(struct mmc_host *mmc, u32 opcode)
+{
+	struct dw_mci_slot *slot = mmc_priv(mmc);
+	struct dw_mci *host = slot->host;
+	const struct dw_mci_drv_data *drv_data = host->drv_data;
+	struct dw_mci_tuning_data tuning_data;
+	int err = -ENOSYS;
+
+	if (opcode == MMC_SEND_TUNING_BLOCK_HS200) {
+		if (mmc->ios.bus_width == MMC_BUS_WIDTH_8) {
+			tuning_data.blk_pattern = tuning_blk_pattern_8bit;
+			tuning_data.blksz = sizeof(tuning_blk_pattern_8bit);
+		} else if (mmc->ios.bus_width == MMC_BUS_WIDTH_4) {
+			tuning_data.blk_pattern = tuning_blk_pattern_4bit;
+			tuning_data.blksz = sizeof(tuning_blk_pattern_4bit);
+		} else {
+			return -EINVAL;
+		}
+	} else if (opcode == MMC_SEND_TUNING_BLOCK) {
+		tuning_data.blk_pattern = tuning_blk_pattern_4bit;
+		tuning_data.blksz = sizeof(tuning_blk_pattern_4bit);
+	} else {
+		dev_err(host->dev,
+			"Undefined command(%d) for tuning\n", opcode);
+		return -EINVAL;
+	}
+
+	if (drv_data && drv_data->execute_tuning)
+		err = drv_data->execute_tuning(slot, opcode, &tuning_data);
+	return err;
+}
+
 static const struct mmc_host_ops dw_mci_ops = {
 	.request		= dw_mci_request,
 	.pre_req		= dw_mci_pre_req,
@@ -947,6 +1139,7 @@ static const struct mmc_host_ops dw_mci_ops = {
 	.get_ro			= dw_mci_get_ro,
 	.get_cd			= dw_mci_get_cd,
 	.enable_sdio_irq	= dw_mci_enable_sdio_irq,
+	.execute_tuning		= dw_mci_execute_tuning,
 };
 
 static void dw_mci_request_end(struct dw_mci *host, struct mmc_request *mrq)
@@ -978,7 +1171,7 @@ static void dw_mci_request_end(struct dw_mci *host, struct mmc_request *mrq)
 	spin_lock(&host->lock);
 }
 
-static void dw_mci_command_complete(struct dw_mci *host, struct mmc_command *cmd)
+static int dw_mci_command_complete(struct dw_mci *host, struct mmc_command *cmd)
 {
 	u32 status = host->cmd_status;
 
@@ -1012,12 +1205,52 @@ static void dw_mci_command_complete(struct dw_mci *host, struct mmc_command *cmd
 		/* newer ip versions need a delay between retries */
 		if (host->quirks & DW_MCI_QUIRK_RETRY_DELAY)
 			mdelay(20);
+	}
 
-		if (cmd->data) {
-			dw_mci_stop_dma(host);
-			host->data = NULL;
+	return cmd->error;
+}
+
+static int dw_mci_data_complete(struct dw_mci *host, struct mmc_data *data)
+{
+	u32 status = host->data_status;
+
+	if (status & DW_MCI_DATA_ERROR_FLAGS) {
+		if (status & SDMMC_INT_DRTO) {
+			data->error = -ETIMEDOUT;
+		} else if (status & SDMMC_INT_DCRC) {
+			data->error = -EILSEQ;
+		} else if (status & SDMMC_INT_EBE) {
+			if (host->dir_status ==
+				DW_MCI_SEND_STATUS) {
+				/*
+				 * No data CRC status was returned.
+				 * The number of bytes transferred
+				 * will be exaggerated in PIO mode.
+				 */
+				data->bytes_xfered = 0;
+				data->error = -ETIMEDOUT;
+			} else if (host->dir_status ==
+					DW_MCI_RECV_STATUS) {
+				data->error = -EIO;
+			}
+		} else {
+			/* SDMMC_INT_SBE is included */
+			data->error = -EIO;
 		}
+
+		dev_err(host->dev, "data error, status 0x%08x\n", status);
+
+		/*
+		 * After an error, there may be data lingering
+		 * in the FIFO
+		 */
+		dw_mci_fifo_reset(host);
+	} else {
+		data->bytes_xfered = data->blocks * data->blksz;
+		data->error = 0;
 	}
+
+	return data->error;
 }
 
 static void dw_mci_tasklet_func(unsigned long priv)
@@ -1025,14 +1258,16 @@ static void dw_mci_tasklet_func(unsigned long priv)
 	struct dw_mci *host = (struct dw_mci *)priv;
 	struct mmc_data	*data;
 	struct mmc_command *cmd;
+	struct mmc_request *mrq;
 	enum dw_mci_state state;
 	enum dw_mci_state prev_state;
-	u32 status, ctrl;
+	unsigned int err;
 
 	spin_lock(&host->lock);
 
 	state = host->state;
 	data = host->data;
+	mrq = host->mrq;
 
 	do {
 		prev_state = state;
@@ -1049,16 +1284,23 @@ static void dw_mci_tasklet_func(unsigned long priv)
 			cmd = host->cmd;
 			host->cmd = NULL;
 			set_bit(EVENT_CMD_COMPLETE, &host->completed_events);
-			dw_mci_command_complete(host, cmd);
-			if (cmd == host->mrq->sbc && !cmd->error) {
+			err = dw_mci_command_complete(host, cmd);
+			if (cmd == mrq->sbc && !err) {
 				prev_state = state = STATE_SENDING_CMD;
 				__dw_mci_start_request(host, host->cur_slot,
-						       host->mrq->cmd);
+						       mrq->cmd);
 				goto unlock;
 			}
 
-			if (!host->mrq->data || cmd->error) {
-				dw_mci_request_end(host, host->mrq);
+			if (cmd->data && err) {
+				dw_mci_stop_dma(host);
+				send_stop_abort(host, data);
+				state = STATE_SENDING_STOP;
+				break;
+			}
+
+			if (!cmd->data || err) {
+				dw_mci_request_end(host, mrq);
 				goto unlock;
 			}
 
@@ -1069,8 +1311,7 @@ static void dw_mci_tasklet_func(unsigned long priv)
 			if (test_and_clear_bit(EVENT_DATA_ERROR,
 					       &host->pending_events)) {
 				dw_mci_stop_dma(host);
-				if (data->stop)
-					send_stop_cmd(host, data);
+				send_stop_abort(host, data);
 				state = STATE_DATA_ERROR;
 				break;
 			}
@@ -1090,60 +1331,27 @@ static void dw_mci_tasklet_func(unsigned long priv)
 
 			host->data = NULL;
 			set_bit(EVENT_DATA_COMPLETE, &host->completed_events);
-			status = host->data_status;
-
-			if (status & DW_MCI_DATA_ERROR_FLAGS) {
-				if (status & SDMMC_INT_DRTO) {
-					data->error = -ETIMEDOUT;
-				} else if (status & SDMMC_INT_DCRC) {
-					data->error = -EILSEQ;
-				} else if (status & SDMMC_INT_EBE &&
-					   host->dir_status ==
-							DW_MCI_SEND_STATUS) {
-					/*
-					 * No data CRC status was returned.
-					 * The number of bytes transferred will
-					 * be exaggerated in PIO mode.
-					 */
-					data->bytes_xfered = 0;
-					data->error = -ETIMEDOUT;
-				} else {
-					dev_err(host->dev,
-						"data FIFO error "
-						"(status=%08x)\n",
-						status);
-					data->error = -EIO;
-				}
-				/*
-				 * After an error, there may be data lingering
-				 * in the FIFO, so reset it - doing so
-				 * generates a block interrupt, hence setting
-				 * the scatter-gather pointer to NULL.
-				 */
-				sg_miter_stop(&host->sg_miter);
-				host->sg = NULL;
-				ctrl = mci_readl(host, CTRL);
-				ctrl |= SDMMC_CTRL_FIFO_RESET;
-				mci_writel(host, CTRL, ctrl);
-			} else {
-				data->bytes_xfered = data->blocks * data->blksz;
-				data->error = 0;
-			}
+			err = dw_mci_data_complete(host, data);
 
-			if (!data->stop) {
-				dw_mci_request_end(host, host->mrq);
-				goto unlock;
-			}
+			if (!err) {
+				if (!data->stop || mrq->sbc) {
+					if (mrq->sbc)
+						data->stop->error = 0;
+					dw_mci_request_end(host, mrq);
+					goto unlock;
+				}
 
-			if (host->mrq->sbc && !data->error) {
-				data->stop->error = 0;
-				dw_mci_request_end(host, host->mrq);
-				goto unlock;
+				/* stop command for open-ended transfer*/
+				if (data->stop)
+					send_stop_abort(host, data);
 			}
 
+			/*
+			 * If err has non-zero,
+			 * stop-abort command has been already issued.
+			 */
 			prev_state = state = STATE_SENDING_STOP;
-			if (!data->error)
-				send_stop_cmd(host, data);
+
 			/* fall through */
 
 		case STATE_SENDING_STOP:
@@ -1151,9 +1359,19 @@ static void dw_mci_tasklet_func(unsigned long priv)
 						&host->pending_events))
 				break;
 
+			/* CMD error in data command */
+			if (mrq->cmd->error && mrq->data)
+				dw_mci_fifo_reset(host);
+
 			host->cmd = NULL;
-			dw_mci_command_complete(host, host->mrq->stop);
-			dw_mci_request_end(host, host->mrq);
+			host->data = NULL;
+
+			if (mrq->stop)
+				dw_mci_command_complete(host, mrq->stop);
+			else
+				host->cmd_status = 0;
+
+			dw_mci_request_end(host, mrq);
 			goto unlock;
 
 		case STATE_DATA_ERROR:
@@ -1697,7 +1915,6 @@ static void dw_mci_work_routine_card(struct work_struct *work)
 		struct mmc_host *mmc = slot->mmc;
 		struct mmc_request *mrq;
 		int present;
-		u32 ctrl;
 
 		present = dw_mci_get_cd(mmc);
 		while (present != slot->last_detect_state) {
@@ -1736,11 +1953,10 @@ static void dw_mci_work_routine_card(struct work_struct *work)
 					case STATE_DATA_ERROR:
 						if (mrq->data->error == -EINPROGRESS)
 							mrq->data->error = -ENOMEDIUM;
-						if (!mrq->stop)
-							break;
 						/* fall through */
 					case STATE_SENDING_STOP:
-						mrq->stop->error = -ENOMEDIUM;
+						if (mrq->stop)
+							mrq->stop->error = -ENOMEDIUM;
 						break;
 					}
 
@@ -1763,23 +1979,10 @@ static void dw_mci_work_routine_card(struct work_struct *work)
 			if (present == 0) {
 				clear_bit(DW_MMC_CARD_PRESENT, &slot->flags);
 
-				/*
-				 * Clear down the FIFO - doing so generates a
-				 * block interrupt, hence setting the
-				 * scatter-gather pointer to NULL.
-				 */
-				sg_miter_stop(&host->sg_miter);
-				host->sg = NULL;
-
-				ctrl = mci_readl(host, CTRL);
-				ctrl |= SDMMC_CTRL_FIFO_RESET;
-				mci_writel(host, CTRL, ctrl);
-
+				/* Clear down the FIFO */
+				dw_mci_fifo_reset(host);
 #ifdef CONFIG_MMC_DW_IDMAC
-				ctrl = mci_readl(host, BMOD);
-				/* Software reset of DMA */
-				ctrl |= SDMMC_IDMAC_SWRESET;
-				mci_writel(host, BMOD, ctrl);
+				dw_mci_idmac_reset(host);
 #endif
 
 			}
@@ -1901,6 +2104,7 @@ static int dw_mci_init_slot(struct dw_mci *host, unsigned int id)
 	struct dw_mci_slot *slot;
 	const struct dw_mci_drv_data *drv_data = host->drv_data;
 	int ctrl_id, ret;
+	u32 freq[2];
 	u8 bus_width;
 
 	mmc = mmc_alloc_host(sizeof(struct dw_mci_slot), host->dev);
@@ -1916,8 +2120,14 @@ static int dw_mci_init_slot(struct dw_mci *host, unsigned int id)
 	slot->quirks = dw_mci_of_get_slot_quirks(host->dev, slot->id);
 
 	mmc->ops = &dw_mci_ops;
-	mmc->f_min = DIV_ROUND_UP(host->bus_hz, 510);
-	mmc->f_max = host->bus_hz;
+	if (of_property_read_u32_array(host->dev->of_node,
+				       "clock-freq-min-max", freq, 2)) {
+		mmc->f_min = DW_MCI_FREQ_MIN;
+		mmc->f_max = DW_MCI_FREQ_MAX;
+	} else {
+		mmc->f_min = freq[0];
+		mmc->f_max = freq[1];
+	}
 
 	if (host->pdata->get_ocr)
 		mmc->ocr_avail = host->pdata->get_ocr(id);
@@ -1964,9 +2174,6 @@ static int dw_mci_init_slot(struct dw_mci *host, unsigned int id)
 		mmc->caps |= MMC_CAP_4_BIT_DATA;
 	}
 
-	if (host->pdata->quirks & DW_MCI_QUIRK_HIGHSPEED)
-		mmc->caps |= MMC_CAP_SD_HIGHSPEED | MMC_CAP_MMC_HIGHSPEED;
-
 	if (host->pdata->blk_settings) {
 		mmc->max_segs = host->pdata->blk_settings->max_segs;
 		mmc->max_blk_size = host->pdata->blk_settings->max_blk_size;
@@ -2008,12 +2215,6 @@ static int dw_mci_init_slot(struct dw_mci *host, unsigned int id)
 	/* Card initially undetected */
 	slot->last_detect_state = 0;
 
-	/*
-	 * Card may have been plugged in prior to boot so we
-	 * need to run the detect tasklet
-	 */
-	queue_work(host->card_workqueue, &host->card_work);
-
 	return 0;
 
 err_setup_bus:
@@ -2074,36 +2275,57 @@ no_dma:
 	return;
 }
 
-static bool mci_wait_reset(struct device *dev, struct dw_mci *host)
+static bool dw_mci_ctrl_reset(struct dw_mci *host, u32 reset)
 {
 	unsigned long timeout = jiffies + msecs_to_jiffies(500);
-	unsigned int ctrl;
+	u32 ctrl;
 
-	mci_writel(host, CTRL, (SDMMC_CTRL_RESET | SDMMC_CTRL_FIFO_RESET |
-				SDMMC_CTRL_DMA_RESET));
+	ctrl = mci_readl(host, CTRL);
+	ctrl |= reset;
+	mci_writel(host, CTRL, ctrl);
 
 	/* wait till resets clear */
 	do {
 		ctrl = mci_readl(host, CTRL);
-		if (!(ctrl & (SDMMC_CTRL_RESET | SDMMC_CTRL_FIFO_RESET |
-			      SDMMC_CTRL_DMA_RESET)))
+		if (!(ctrl & reset))
 			return true;
 	} while (time_before(jiffies, timeout));
 
-	dev_err(dev, "Timeout resetting block (ctrl %#x)\n", ctrl);
+	dev_err(host->dev,
+		"Timeout resetting block (ctrl reset %#x)\n",
+		ctrl & reset);
 
 	return false;
 }
 
+static inline bool dw_mci_fifo_reset(struct dw_mci *host)
+{
+	/*
+	 * Reseting generates a block interrupt, hence setting
+	 * the scatter-gather pointer to NULL.
+	 */
+	if (host->sg) {
+		sg_miter_stop(&host->sg_miter);
+		host->sg = NULL;
+	}
+
+	return dw_mci_ctrl_reset(host, SDMMC_CTRL_FIFO_RESET);
+}
+
+static inline bool dw_mci_ctrl_all_reset(struct dw_mci *host)
+{
+	return dw_mci_ctrl_reset(host,
+				 SDMMC_CTRL_FIFO_RESET |
+				 SDMMC_CTRL_RESET |
+				 SDMMC_CTRL_DMA_RESET);
+}
+
 #ifdef CONFIG_OF
 static struct dw_mci_of_quirks {
 	char *quirk;
 	int id;
 } of_quirks[] = {
 	{
-		.quirk	= "supports-highspeed",
-		.id	= DW_MCI_QUIRK_HIGHSPEED,
-	}, {
 		.quirk	= "broken-cd",
 		.id	= DW_MCI_QUIRK_BROKEN_CARD_DETECTION,
 	},
@@ -2158,6 +2380,15 @@ static struct dw_mci_board *dw_mci_parse_dt(struct dw_mci *host)
 	if (of_find_property(np, "enable-sdio-wakeup", NULL))
 		pdata->pm_caps |= MMC_PM_WAKE_SDIO_IRQ;
 
+	if (of_find_property(np, "supports-highspeed", NULL))
+		pdata->caps |= MMC_CAP_SD_HIGHSPEED | MMC_CAP_MMC_HIGHSPEED;
+
+	if (of_find_property(np, "caps2-mmc-hs200-1_8v", NULL))
+		pdata->caps2 |= MMC_CAP2_HS200_1_8V_SDR;
+
+	if (of_find_property(np, "caps2-mmc-hs200-1_2v", NULL))
+		pdata->caps2 |= MMC_CAP2_HS200_1_2V_SDR;
+
 	return pdata;
 }
 
@@ -2221,6 +2452,15 @@ int dw_mci_probe(struct dw_mci *host)
 		host->bus_hz = clk_get_rate(host->ciu_clk);
 	}
 
+	if (drv_data && drv_data->init) {
+		ret = drv_data->init(host);
+		if (ret) {
+			dev_err(host->dev,
+				"implementation specific init failed\n");
+			goto err_clk_ciu;
+		}
+	}
+
 	if (drv_data && drv_data->setup_clock) {
 		ret = drv_data->setup_clock(host);
 		if (ret) {
@@ -2287,7 +2527,7 @@ int dw_mci_probe(struct dw_mci *host)
 	}
 
 	/* Reset all blocks */
-	if (!mci_wait_reset(host->dev, host))
+	if (!dw_mci_ctrl_all_reset(host))
 		return -ENODEV;
 
 	host->dma_ops = host->pdata->dma_ops;
@@ -2317,8 +2557,8 @@ int dw_mci_probe(struct dw_mci *host)
 		fifo_size = host->pdata->fifo_depth;
 	}
 	host->fifo_depth = fifo_size;
-	host->fifoth_val = ((0x2 << 28) | ((fifo_size/2 - 1) << 16) |
-			((fifo_size/2) << 0));
+	host->fifoth_val =
+		SDMMC_SET_FIFOTH(0x2, fifo_size / 2 - 1, fifo_size / 2);
 	mci_writel(host, FIFOTH, host->fifoth_val);
 
 	/* disable clock to CIU */
@@ -2456,23 +2696,6 @@ EXPORT_SYMBOL(dw_mci_remove);
  */
 int dw_mci_suspend(struct dw_mci *host)
 {
-	int i, ret = 0;
-
-	for (i = 0; i < host->num_slots; i++) {
-		struct dw_mci_slot *slot = host->slot[i];
-		if (!slot)
-			continue;
-		ret = mmc_suspend_host(slot->mmc);
-		if (ret < 0) {
-			while (--i >= 0) {
-				slot = host->slot[i];
-				if (slot)
-					mmc_resume_host(host->slot[i]->mmc);
-			}
-			return ret;
-		}
-	}
-
 	if (host->vmmc)
 		regulator_disable(host->vmmc);
 
@@ -2493,7 +2716,7 @@ int dw_mci_resume(struct dw_mci *host)
 		}
 	}
 
-	if (!mci_wait_reset(host->dev, host)) {
+	if (!dw_mci_ctrl_all_reset(host)) {
 		ret = -ENODEV;
 		return ret;
 	}
@@ -2501,8 +2724,15 @@ int dw_mci_resume(struct dw_mci *host)
 	if (host->use_dma && host->dma_ops->init)
 		host->dma_ops->init(host);
 
-	/* Restore the old value at FIFOTH register */
+	/*
+	 * Restore the initial value at FIFOTH register
+	 * And Invalidate the prev_blksz with zero
+	 */
 	mci_writel(host, FIFOTH, host->fifoth_val);
+	host->prev_blksz = 0;
+
+	/* Put in max timeout */
+	mci_writel(host, TMOUT, 0xFFFFFFFF);
 
 	mci_writel(host, RINTSTS, 0xFFFFFFFF);
 	mci_writel(host, INTMASK, SDMMC_INT_CMD_DONE | SDMMC_INT_DATA_OVER |
@@ -2518,10 +2748,6 @@ int dw_mci_resume(struct dw_mci *host)
 			dw_mci_set_ios(slot->mmc, &slot->mmc->ios);
 			dw_mci_setup_bus(slot, true);
 		}
-
-		ret = mmc_resume_host(host->slot[i]->mmc);
-		if (ret < 0)
-			return ret;
 	}
 	return 0;
 }
diff --git a/drivers/mmc/host/dw_mmc.h b/drivers/mmc/host/dw_mmc.h
index 81b29941c5b9..6bf24ab917e6 100644
--- a/drivers/mmc/host/dw_mmc.h
+++ b/drivers/mmc/host/dw_mmc.h
@@ -53,6 +53,7 @@
 #define SDMMC_IDINTEN		0x090
 #define SDMMC_DSCADDR		0x094
 #define SDMMC_BUFADDR		0x098
+#define SDMMC_CDTHRCTL		0x100
 #define SDMMC_DATA(x)		(x)
 
 /*
@@ -128,6 +129,10 @@
 #define SDMMC_CMD_INDX(n)		((n) & 0x1F)
 /* Status register defines */
 #define SDMMC_GET_FCNT(x)		(((x)>>17) & 0x1FFF)
+/* FIFOTH register defines */
+#define SDMMC_SET_FIFOTH(m, r, t)	(((m) & 0x7) << 28 | \
+					 ((r) & 0xFFF) << 16 | \
+					 ((t) & 0xFFF))
 /* Internal DMAC interrupt defines */
 #define SDMMC_IDMAC_INT_AI		BIT(9)
 #define SDMMC_IDMAC_INT_NI		BIT(8)
@@ -142,6 +147,8 @@
 #define SDMMC_IDMAC_SWRESET		BIT(0)
 /* Version ID register define */
 #define SDMMC_GET_VERID(x)		((x) & 0xFFFF)
+/* Card read threshold */
+#define SDMMC_SET_RD_THLD(v, x)		(((v) & 0x1FFF) << 16 | (x))
 
 /* Register access macros */
 #define mci_readl(dev, reg)			\
@@ -184,6 +191,52 @@ extern int dw_mci_resume(struct dw_mci *host);
 #endif
 
 /**
+ * struct dw_mci_slot - MMC slot state
+ * @mmc: The mmc_host representing this slot.
+ * @host: The MMC controller this slot is using.
+ * @quirks: Slot-level quirks (DW_MCI_SLOT_QUIRK_XXX)
+ * @wp_gpio: If gpio_is_valid() we'll use this to read write protect.
+ * @ctype: Card type for this slot.
+ * @mrq: mmc_request currently being processed or waiting to be
+ *	processed, or NULL when the slot is idle.
+ * @queue_node: List node for placing this node in the @queue list of
+ *	&struct dw_mci.
+ * @clock: Clock rate configured by set_ios(). Protected by host->lock.
+ * @__clk_old: The last updated clock with reflecting clock divider.
+ *	Keeping track of this helps us to avoid spamming the console
+ *	with CONFIG_MMC_CLKGATE.
+ * @flags: Random state bits associated with the slot.
+ * @id: Number of this slot.
+ * @last_detect_state: Most recently observed card detect state.
+ */
+struct dw_mci_slot {
+	struct mmc_host		*mmc;
+	struct dw_mci		*host;
+
+	int			quirks;
+	int			wp_gpio;
+
+	u32			ctype;
+
+	struct mmc_request	*mrq;
+	struct list_head	queue_node;
+
+	unsigned int		clock;
+	unsigned int		__clk_old;
+
+	unsigned long		flags;
+#define DW_MMC_CARD_PRESENT	0
+#define DW_MMC_CARD_NEED_INIT	1
+	int			id;
+	int			last_detect_state;
+};
+
+struct dw_mci_tuning_data {
+	const u8 *blk_pattern;
+	unsigned int blksz;
+};
+
+/**
  * dw_mci driver data - dw-mshc implementation specific driver data.
  * @caps: mmc subsystem specified capabilities of the controller(s).
  * @init: early implementation specific initialization.
@@ -203,5 +256,7 @@ struct dw_mci_drv_data {
 	void		(*prepare_command)(struct dw_mci *host, u32 *cmdr);
 	void		(*set_ios)(struct dw_mci *host, struct mmc_ios *ios);
 	int		(*parse_dt)(struct dw_mci *host);
+	int		(*execute_tuning)(struct dw_mci_slot *slot, u32 opcode,
+					struct dw_mci_tuning_data *tuning_data);
 };
 #endif /* _DW_MMC_H_ */
diff --git a/drivers/mmc/host/jz4740_mmc.c b/drivers/mmc/host/jz4740_mmc.c
index 66516339e3a0..de2139cf3444 100644
--- a/drivers/mmc/host/jz4740_mmc.c
+++ b/drivers/mmc/host/jz4740_mmc.c
@@ -880,8 +880,6 @@ static int jz4740_mmc_suspend(struct device *dev)
 {
 	struct jz4740_mmc_host *host = dev_get_drvdata(dev);
 
-	mmc_suspend_host(host->mmc);
-
 	jz_gpio_bulk_suspend(jz4740_mmc_pins, jz4740_mmc_num_pins(host));
 
 	return 0;
@@ -893,8 +891,6 @@ static int jz4740_mmc_resume(struct device *dev)
 
 	jz_gpio_bulk_resume(jz4740_mmc_pins, jz4740_mmc_num_pins(host));
 
-	mmc_resume_host(host->mmc);
-
 	return 0;
 }
 
diff --git a/drivers/mmc/host/mmci.c b/drivers/mmc/host/mmci.c
index d135c76c4855..f32057972dd7 100644
--- a/drivers/mmc/host/mmci.c
+++ b/drivers/mmc/host/mmci.c
@@ -1730,37 +1730,28 @@ static int mmci_suspend(struct device *dev)
 {
 	struct amba_device *adev = to_amba_device(dev);
 	struct mmc_host *mmc = amba_get_drvdata(adev);
-	int ret = 0;
 
 	if (mmc) {
 		struct mmci_host *host = mmc_priv(mmc);
-
-		ret = mmc_suspend_host(mmc);
-		if (ret == 0) {
-			pm_runtime_get_sync(dev);
-			writel(0, host->base + MMCIMASK0);
-		}
+		pm_runtime_get_sync(dev);
+		writel(0, host->base + MMCIMASK0);
 	}
 
-	return ret;
+	return 0;
 }
 
 static int mmci_resume(struct device *dev)
 {
 	struct amba_device *adev = to_amba_device(dev);
 	struct mmc_host *mmc = amba_get_drvdata(adev);
-	int ret = 0;
 
 	if (mmc) {
 		struct mmci_host *host = mmc_priv(mmc);
-
 		writel(MCI_IRQENABLE, host->base + MMCIMASK0);
 		pm_runtime_put(dev);
-
-		ret = mmc_resume_host(mmc);
 	}
 
-	return ret;
+	return 0;
 }
 #endif
 
diff --git a/drivers/mmc/host/msm_sdcc.c b/drivers/mmc/host/msm_sdcc.c
index b900de4e7e94..9405ecdaf6cf 100644
--- a/drivers/mmc/host/msm_sdcc.c
+++ b/drivers/mmc/host/msm_sdcc.c
@@ -1416,28 +1416,10 @@ ioremap_free:
 }
 
 #ifdef CONFIG_PM
-#ifdef CONFIG_MMC_MSM7X00A_RESUME_IN_WQ
-static void
-do_resume_work(struct work_struct *work)
-{
-	struct msmsdcc_host *host =
-		container_of(work, struct msmsdcc_host, resume_task);
-	struct mmc_host	*mmc = host->mmc;
-
-	if (mmc) {
-		mmc_resume_host(mmc);
-		if (host->stat_irq)
-			enable_irq(host->stat_irq);
-	}
-}
-#endif
-
-
 static int
 msmsdcc_suspend(struct platform_device *dev, pm_message_t state)
 {
 	struct mmc_host *mmc = mmc_get_drvdata(dev);
-	int rc = 0;
 
 	if (mmc) {
 		struct msmsdcc_host *host = mmc_priv(mmc);
@@ -1445,14 +1427,11 @@ msmsdcc_suspend(struct platform_device *dev, pm_message_t state)
 		if (host->stat_irq)
 			disable_irq(host->stat_irq);
 
-		if (mmc->card && mmc->card->type != MMC_TYPE_SDIO)
-			rc = mmc_suspend_host(mmc);
-		if (!rc)
-			msmsdcc_writel(host, 0, MMCIMASK0);
+		msmsdcc_writel(host, 0, MMCIMASK0);
 		if (host->clks_on)
 			msmsdcc_disable_clocks(host, 0);
 	}
-	return rc;
+	return 0;
 }
 
 static int
@@ -1467,8 +1446,6 @@ msmsdcc_resume(struct platform_device *dev)
 
 		msmsdcc_writel(host, host->saved_irq0mask, MMCIMASK0);
 
-		if (mmc->card && mmc->card->type != MMC_TYPE_SDIO)
-			mmc_resume_host(mmc);
 		if (host->stat_irq)
 			enable_irq(host->stat_irq);
 #if BUSCLK_PWRSAVE
diff --git a/drivers/mmc/host/mvsdio.c b/drivers/mmc/host/mvsdio.c
index deecee08c288..45aa2206741d 100644
--- a/drivers/mmc/host/mvsdio.c
+++ b/drivers/mmc/host/mvsdio.c
@@ -775,9 +775,9 @@ static int mvsd_probe(struct platform_device *pdev)
 
 	spin_lock_init(&host->lock);
 
-	host->base = devm_request_and_ioremap(&pdev->dev, r);
-	if (!host->base) {
-		ret = -ENOMEM;
+	host->base = devm_ioremap_resource(&pdev->dev, r);
+	if (IS_ERR(host->base)) {
+		ret = PTR_ERR(host->base);
 		goto out;
 	}
 
@@ -838,33 +838,6 @@ static int mvsd_remove(struct platform_device *pdev)
 	return 0;
 }
 
-#ifdef CONFIG_PM
-static int mvsd_suspend(struct platform_device *dev, pm_message_t state)
-{
-	struct mmc_host *mmc = platform_get_drvdata(dev);
-	int ret = 0;
-
-	if (mmc)
-		ret = mmc_suspend_host(mmc);
-
-	return ret;
-}
-
-static int mvsd_resume(struct platform_device *dev)
-{
-	struct mmc_host *mmc = platform_get_drvdata(dev);
-	int ret = 0;
-
-	if (mmc)
-		ret = mmc_resume_host(mmc);
-
-	return ret;
-}
-#else
-#define mvsd_suspend	NULL
-#define mvsd_resume	NULL
-#endif
-
 static const struct of_device_id mvsdio_dt_ids[] = {
 	{ .compatible = "marvell,orion-sdio" },
 	{ /* sentinel */ }
@@ -874,8 +847,6 @@ MODULE_DEVICE_TABLE(of, mvsdio_dt_ids);
 static struct platform_driver mvsd_driver = {
 	.probe		= mvsd_probe,
 	.remove		= mvsd_remove,
-	.suspend	= mvsd_suspend,
-	.resume		= mvsd_resume,
 	.driver		= {
 		.name	= DRIVER_NAME,
 		.of_match_table = mvsdio_dt_ids,
diff --git a/drivers/mmc/host/mxcmmc.c b/drivers/mmc/host/mxcmmc.c
index c174c6a0d224..f7199c83f5cf 100644
--- a/drivers/mmc/host/mxcmmc.c
+++ b/drivers/mmc/host/mxcmmc.c
@@ -1250,28 +1250,20 @@ static int mxcmci_suspend(struct device *dev)
 {
 	struct mmc_host *mmc = dev_get_drvdata(dev);
 	struct mxcmci_host *host = mmc_priv(mmc);
-	int ret = 0;
 
-	if (mmc)
-		ret = mmc_suspend_host(mmc);
 	clk_disable_unprepare(host->clk_per);
 	clk_disable_unprepare(host->clk_ipg);
-
-	return ret;
+	return 0;
 }
 
 static int mxcmci_resume(struct device *dev)
 {
 	struct mmc_host *mmc = dev_get_drvdata(dev);
 	struct mxcmci_host *host = mmc_priv(mmc);
-	int ret = 0;
 
 	clk_prepare_enable(host->clk_per);
 	clk_prepare_enable(host->clk_ipg);
-	if (mmc)
-		ret = mmc_resume_host(mmc);
-
-	return ret;
+	return 0;
 }
 
 static const struct dev_pm_ops mxcmci_pm_ops = {
diff --git a/drivers/mmc/host/mxs-mmc.c b/drivers/mmc/host/mxs-mmc.c
index e1fa3ef735e0..50fc9df791b2 100644
--- a/drivers/mmc/host/mxs-mmc.c
+++ b/drivers/mmc/host/mxs-mmc.c
@@ -724,13 +724,9 @@ static int mxs_mmc_suspend(struct device *dev)
 	struct mmc_host *mmc = dev_get_drvdata(dev);
 	struct mxs_mmc_host *host = mmc_priv(mmc);
 	struct mxs_ssp *ssp = &host->ssp;
-	int ret = 0;
-
-	ret = mmc_suspend_host(mmc);
 
 	clk_disable_unprepare(ssp->clk);
-
-	return ret;
+	return 0;
 }
 
 static int mxs_mmc_resume(struct device *dev)
@@ -738,13 +734,9 @@ static int mxs_mmc_resume(struct device *dev)
 	struct mmc_host *mmc = dev_get_drvdata(dev);
 	struct mxs_mmc_host *host = mmc_priv(mmc);
 	struct mxs_ssp *ssp = &host->ssp;
-	int ret = 0;
 
 	clk_prepare_enable(ssp->clk);
-
-	ret = mmc_resume_host(mmc);
-
-	return ret;
+	return 0;
 }
 
 static const struct dev_pm_ops mxs_mmc_pm_ops = {
diff --git a/drivers/mmc/host/omap.c b/drivers/mmc/host/omap.c
index b94f38ec2a83..0b10a9030f4e 100644
--- a/drivers/mmc/host/omap.c
+++ b/drivers/mmc/host/omap.c
@@ -128,7 +128,6 @@ struct mmc_omap_slot {
 
 struct mmc_omap_host {
 	int			initialized;
-	int			suspended;
 	struct mmc_request *	mrq;
 	struct mmc_command *	cmd;
 	struct mmc_data *	data;
@@ -1513,61 +1512,9 @@ static int mmc_omap_remove(struct platform_device *pdev)
 	return 0;
 }
 
-#ifdef CONFIG_PM
-static int mmc_omap_suspend(struct platform_device *pdev, pm_message_t mesg)
-{
-	int i, ret = 0;
-	struct mmc_omap_host *host = platform_get_drvdata(pdev);
-
-	if (host == NULL || host->suspended)
-		return 0;
-
-	for (i = 0; i < host->nr_slots; i++) {
-		struct mmc_omap_slot *slot;
-
-		slot = host->slots[i];
-		ret = mmc_suspend_host(slot->mmc);
-		if (ret < 0) {
-			while (--i >= 0) {
-				slot = host->slots[i];
-				mmc_resume_host(slot->mmc);
-			}
-			return ret;
-		}
-	}
-	host->suspended = 1;
-	return 0;
-}
-
-static int mmc_omap_resume(struct platform_device *pdev)
-{
-	int i, ret = 0;
-	struct mmc_omap_host *host = platform_get_drvdata(pdev);
-
-	if (host == NULL || !host->suspended)
-		return 0;
-
-	for (i = 0; i < host->nr_slots; i++) {
-		struct mmc_omap_slot *slot;
-		slot = host->slots[i];
-		ret = mmc_resume_host(slot->mmc);
-		if (ret < 0)
-			return ret;
-
-		host->suspended = 0;
-	}
-	return 0;
-}
-#else
-#define mmc_omap_suspend	NULL
-#define mmc_omap_resume		NULL
-#endif
-
 static struct platform_driver mmc_omap_driver = {
 	.probe		= mmc_omap_probe,
 	.remove		= mmc_omap_remove,
-	.suspend	= mmc_omap_suspend,
-	.resume		= mmc_omap_resume,
 	.driver		= {
 		.name	= DRIVER_NAME,
 		.owner	= THIS_MODULE,
diff --git a/drivers/mmc/host/omap_hsmmc.c b/drivers/mmc/host/omap_hsmmc.c
index 6ac63df645c4..dbd32ad3b749 100644
--- a/drivers/mmc/host/omap_hsmmc.c
+++ b/drivers/mmc/host/omap_hsmmc.c
@@ -75,6 +75,7 @@
 #define ICE			0x1
 #define ICS			0x2
 #define CEN			(1 << 2)
+#define CLKD_MAX		0x3FF		/* max clock divisor: 1023 */
 #define CLKD_MASK		0x0000FFC0
 #define CLKD_SHIFT		6
 #define DTO_MASK		0x000F0000
@@ -119,7 +120,8 @@
 		BRR_EN | BWR_EN | TC_EN | CC_EN)
 
 #define MMC_AUTOSUSPEND_DELAY	100
-#define MMC_TIMEOUT_MS		20
+#define MMC_TIMEOUT_MS		20		/* 20 mSec */
+#define MMC_TIMEOUT_US		20000		/* 20000 micro Sec */
 #define OMAP_MMC_MIN_CLOCK	400000
 #define OMAP_MMC_MAX_CLOCK	52000000
 #define DRIVER_NAME		"omap_hsmmc"
@@ -171,6 +173,10 @@ struct omap_hsmmc_host {
 	unsigned char		bus_mode;
 	unsigned char		power_mode;
 	int			suspended;
+	u32			con;
+	u32			hctl;
+	u32			sysctl;
+	u32			capa;
 	int			irq;
 	int			use_dma, dma_ch;
 	struct dma_chan		*tx_chan;
@@ -183,7 +189,6 @@ struct omap_hsmmc_host {
 	int			use_reg;
 	int			req_in_progress;
 	struct omap_hsmmc_next	next_data;
-
 	struct	omap_mmc_platform_data	*pdata;
 };
 
@@ -493,8 +498,8 @@ static u16 calc_divisor(struct omap_hsmmc_host *host, struct mmc_ios *ios)
 
 	if (ios->clock) {
 		dsor = DIV_ROUND_UP(clk_get_rate(host->fclk), ios->clock);
-		if (dsor > 250)
-			dsor = 250;
+		if (dsor > CLKD_MAX)
+			dsor = CLKD_MAX;
 	}
 
 	return dsor;
@@ -597,25 +602,20 @@ static void omap_hsmmc_set_bus_mode(struct omap_hsmmc_host *host)
 static int omap_hsmmc_context_restore(struct omap_hsmmc_host *host)
 {
 	struct mmc_ios *ios = &host->mmc->ios;
-	struct omap_mmc_platform_data *pdata = host->pdata;
-	int context_loss = 0;
 	u32 hctl, capa;
 	unsigned long timeout;
 
-	if (pdata->get_context_loss_count) {
-		context_loss = pdata->get_context_loss_count(host->dev);
-		if (context_loss < 0)
-			return 1;
-	}
-
-	dev_dbg(mmc_dev(host->mmc), "context was %slost\n",
-		context_loss == host->context_loss ? "not " : "");
-	if (host->context_loss == context_loss)
-		return 1;
-
 	if (!OMAP_HSMMC_READ(host->base, SYSSTATUS) & RESETDONE)
 		return 1;
 
+	if (host->con == OMAP_HSMMC_READ(host->base, CON) &&
+	    host->hctl == OMAP_HSMMC_READ(host->base, HCTL) &&
+	    host->sysctl == OMAP_HSMMC_READ(host->base, SYSCTL) &&
+	    host->capa == OMAP_HSMMC_READ(host->base, CAPA))
+		return 0;
+
+	host->context_loss++;
+
 	if (host->pdata->controller_flags & OMAP_HSMMC_SUPPORTS_DUAL_VOLT) {
 		if (host->power_mode != MMC_POWER_OFF &&
 		    (1 << ios->vdd) <= MMC_VDD_23_24)
@@ -655,9 +655,8 @@ static int omap_hsmmc_context_restore(struct omap_hsmmc_host *host)
 	omap_hsmmc_set_bus_mode(host);
 
 out:
-	host->context_loss = context_loss;
-
-	dev_dbg(mmc_dev(host->mmc), "context is restored\n");
+	dev_dbg(mmc_dev(host->mmc), "context is restored: restore count %d\n",
+		host->context_loss);
 	return 0;
 }
 
@@ -666,15 +665,10 @@ out:
  */
 static void omap_hsmmc_context_save(struct omap_hsmmc_host *host)
 {
-	struct omap_mmc_platform_data *pdata = host->pdata;
-	int context_loss;
-
-	if (pdata->get_context_loss_count) {
-		context_loss = pdata->get_context_loss_count(host->dev);
-		if (context_loss < 0)
-			return;
-		host->context_loss = context_loss;
-	}
+	host->con =  OMAP_HSMMC_READ(host->base, CON);
+	host->hctl = OMAP_HSMMC_READ(host->base, HCTL);
+	host->sysctl =  OMAP_HSMMC_READ(host->base, SYSCTL);
+	host->capa = OMAP_HSMMC_READ(host->base, CAPA);
 }
 
 #else
@@ -975,8 +969,7 @@ static inline void omap_hsmmc_reset_controller_fsm(struct omap_hsmmc_host *host,
 						   unsigned long bit)
 {
 	unsigned long i = 0;
-	unsigned long limit = (loops_per_jiffy *
-				msecs_to_jiffies(MMC_TIMEOUT_MS));
+	unsigned long limit = MMC_TIMEOUT_US;
 
 	OMAP_HSMMC_WRITE(host->base, SYSCTL,
 			 OMAP_HSMMC_READ(host->base, SYSCTL) | bit);
@@ -988,13 +981,13 @@ static inline void omap_hsmmc_reset_controller_fsm(struct omap_hsmmc_host *host,
 	if (mmc_slot(host).features & HSMMC_HAS_UPDATED_RESET) {
 		while ((!(OMAP_HSMMC_READ(host->base, SYSCTL) & bit))
 					&& (i++ < limit))
-			cpu_relax();
+			udelay(1);
 	}
 	i = 0;
 
 	while ((OMAP_HSMMC_READ(host->base, SYSCTL) & bit) &&
 		(i++ < limit))
-		cpu_relax();
+		udelay(1);
 
 	if (OMAP_HSMMC_READ(host->base, SYSCTL) & bit)
 		dev_err(mmc_dev(host->mmc),
@@ -1178,9 +1171,6 @@ static irqreturn_t omap_hsmmc_detect(int irq, void *dev_id)
 	struct omap_mmc_slot_data *slot = &mmc_slot(host);
 	int carddetect;
 
-	if (host->suspended)
-		return IRQ_HANDLED;
-
 	sysfs_notify(&host->mmc->class_dev.kobj, NULL, "cover_switch");
 
 	if (slot->card_detect)
@@ -1635,18 +1625,9 @@ static int omap_hsmmc_regs_show(struct seq_file *s, void *data)
 {
 	struct mmc_host *mmc = s->private;
 	struct omap_hsmmc_host *host = mmc_priv(mmc);
-	int context_loss = 0;
-
-	if (host->pdata->get_context_loss_count)
-		context_loss = host->pdata->get_context_loss_count(host->dev);
 
-	seq_printf(s, "mmc%d:\n ctx_loss:\t%d:%d\n\nregs:\n",
-			mmc->index, host->context_loss, context_loss);
-
-	if (host->suspended) {
-		seq_printf(s, "host suspended, can't read registers\n");
-		return 0;
-	}
+	seq_printf(s, "mmc%d:\n ctx_loss:\t%d\n\nregs:\n",
+			mmc->index, host->context_loss);
 
 	pm_runtime_get_sync(host->dev);
 
@@ -1838,13 +1819,6 @@ static int omap_hsmmc_probe(struct platform_device *pdev)
 
 	mmc->ops	= &omap_hsmmc_ops;
 
-	/*
-	 * If regulator_disable can only put vcc_aux to sleep then there is
-	 * no off state.
-	 */
-	if (mmc_slot(host).vcc_aux_disable_is_sleep)
-		mmc_slot(host).no_off = 1;
-
 	mmc->f_min = OMAP_MMC_MIN_CLOCK;
 
 	if (pdata->max_freq > 0)
@@ -1874,7 +1848,7 @@ static int omap_hsmmc_probe(struct platform_device *pdev)
 	omap_hsmmc_context_save(host);
 
 	/* This can be removed once we support PBIAS with DT */
-	if (host->dev->of_node && host->mapbase == 0x4809c000)
+	if (host->dev->of_node && res->start == 0x4809c000)
 		host->pbias_disable = 1;
 
 	host->dbclk = clk_get(&pdev->dev, "mmchsdb_fck");
@@ -2119,23 +2093,12 @@ static void omap_hsmmc_complete(struct device *dev)
 
 static int omap_hsmmc_suspend(struct device *dev)
 {
-	int ret = 0;
 	struct omap_hsmmc_host *host = dev_get_drvdata(dev);
 
 	if (!host)
 		return 0;
 
-	if (host && host->suspended)
-		return 0;
-
 	pm_runtime_get_sync(host->dev);
-	host->suspended = 1;
-	ret = mmc_suspend_host(host->mmc);
-
-	if (ret) {
-		host->suspended = 0;
-		goto err;
-	}
 
 	if (!(host->mmc->pm_flags & MMC_PM_KEEP_POWER)) {
 		omap_hsmmc_disable_irq(host);
@@ -2145,23 +2108,19 @@ static int omap_hsmmc_suspend(struct device *dev)
 
 	if (host->dbclk)
 		clk_disable_unprepare(host->dbclk);
-err:
+
 	pm_runtime_put_sync(host->dev);
-	return ret;
+	return 0;
 }
 
 /* Routine to resume the MMC device */
 static int omap_hsmmc_resume(struct device *dev)
 {
-	int ret = 0;
 	struct omap_hsmmc_host *host = dev_get_drvdata(dev);
 
 	if (!host)
 		return 0;
 
-	if (host && !host->suspended)
-		return 0;
-
 	pm_runtime_get_sync(host->dev);
 
 	if (host->dbclk)
@@ -2172,16 +2131,9 @@ static int omap_hsmmc_resume(struct device *dev)
 
 	omap_hsmmc_protect_card(host);
 
-	/* Notify the core to resume the host */
-	ret = mmc_resume_host(host->mmc);
-	if (ret == 0)
-		host->suspended = 0;
-
 	pm_runtime_mark_last_busy(host->dev);
 	pm_runtime_put_autosuspend(host->dev);
-
-	return ret;
-
+	return 0;
 }
 
 #else
diff --git a/drivers/mmc/host/pxamci.c b/drivers/mmc/host/pxamci.c
index 1956a3df7cf3..32fe11323f39 100644
--- a/drivers/mmc/host/pxamci.c
+++ b/drivers/mmc/host/pxamci.c
@@ -880,35 +880,6 @@ static int pxamci_remove(struct platform_device *pdev)
 	return 0;
 }
 
-#ifdef CONFIG_PM
-static int pxamci_suspend(struct device *dev)
-{
-	struct mmc_host *mmc = dev_get_drvdata(dev);
-	int ret = 0;
-
-	if (mmc)
-		ret = mmc_suspend_host(mmc);
-
-	return ret;
-}
-
-static int pxamci_resume(struct device *dev)
-{
-	struct mmc_host *mmc = dev_get_drvdata(dev);
-	int ret = 0;
-
-	if (mmc)
-		ret = mmc_resume_host(mmc);
-
-	return ret;
-}
-
-static const struct dev_pm_ops pxamci_pm_ops = {
-	.suspend	= pxamci_suspend,
-	.resume		= pxamci_resume,
-};
-#endif
-
 static struct platform_driver pxamci_driver = {
 	.probe		= pxamci_probe,
 	.remove		= pxamci_remove,
@@ -916,9 +887,6 @@ static struct platform_driver pxamci_driver = {
 		.name	= DRIVER_NAME,
 		.owner	= THIS_MODULE,
 		.of_match_table = of_match_ptr(pxa_mmc_dt_ids),
-#ifdef CONFIG_PM
-		.pm	= &pxamci_pm_ops,
-#endif
 	},
 };
 
diff --git a/drivers/mmc/host/rtsx_pci_sdmmc.c b/drivers/mmc/host/rtsx_pci_sdmmc.c
index 375a880e0c5f..c46feda07d56 100644
--- a/drivers/mmc/host/rtsx_pci_sdmmc.c
+++ b/drivers/mmc/host/rtsx_pci_sdmmc.c
@@ -364,7 +364,7 @@ static int sd_rw_multi(struct realtek_pci_sdmmc *host, struct mmc_request *mrq)
 	struct mmc_host *mmc = host->mmc;
 	struct mmc_card *card = mmc->card;
 	struct mmc_data *data = mrq->data;
-	int uhs = mmc_sd_card_uhs(card);
+	int uhs = mmc_card_uhs(card);
 	int read = (data->flags & MMC_DATA_READ) ? 1 : 0;
 	u8 cfg2, trans_mode;
 	int err;
@@ -1197,37 +1197,6 @@ static const struct mmc_host_ops realtek_pci_sdmmc_ops = {
 	.execute_tuning = sdmmc_execute_tuning,
 };
 
-#ifdef CONFIG_PM
-static int rtsx_pci_sdmmc_suspend(struct platform_device *pdev,
-		pm_message_t state)
-{
-	struct realtek_pci_sdmmc *host = platform_get_drvdata(pdev);
-	struct mmc_host *mmc = host->mmc;
-	int err;
-
-	dev_dbg(sdmmc_dev(host), "--> %s\n", __func__);
-
-	err = mmc_suspend_host(mmc);
-	if (err)
-		return err;
-
-	return 0;
-}
-
-static int rtsx_pci_sdmmc_resume(struct platform_device *pdev)
-{
-	struct realtek_pci_sdmmc *host = platform_get_drvdata(pdev);
-	struct mmc_host *mmc = host->mmc;
-
-	dev_dbg(sdmmc_dev(host), "--> %s\n", __func__);
-
-	return mmc_resume_host(mmc);
-}
-#else /* CONFIG_PM */
-#define rtsx_pci_sdmmc_suspend NULL
-#define rtsx_pci_sdmmc_resume NULL
-#endif /* CONFIG_PM */
-
 static void init_extra_caps(struct realtek_pci_sdmmc *host)
 {
 	struct mmc_host *mmc = host->mmc;
@@ -1367,8 +1336,6 @@ static struct platform_driver rtsx_pci_sdmmc_driver = {
 	.probe		= rtsx_pci_sdmmc_drv_probe,
 	.remove		= rtsx_pci_sdmmc_drv_remove,
 	.id_table       = rtsx_pci_sdmmc_ids,
-	.suspend	= rtsx_pci_sdmmc_suspend,
-	.resume		= rtsx_pci_sdmmc_resume,
 	.driver		= {
 		.owner	= THIS_MODULE,
 		.name	= DRV_NAME_RTSX_PCI_SDMMC,
diff --git a/drivers/mmc/host/s3cmci.c b/drivers/mmc/host/s3cmci.c
index 8d6794cdf899..2fce5ea5eb39 100644
--- a/drivers/mmc/host/s3cmci.c
+++ b/drivers/mmc/host/s3cmci.c
@@ -1949,39 +1949,10 @@ static struct platform_device_id s3cmci_driver_ids[] = {
 
 MODULE_DEVICE_TABLE(platform, s3cmci_driver_ids);
 
-
-#ifdef CONFIG_PM
-
-static int s3cmci_suspend(struct device *dev)
-{
-	struct mmc_host *mmc = platform_get_drvdata(to_platform_device(dev));
-
-	return mmc_suspend_host(mmc);
-}
-
-static int s3cmci_resume(struct device *dev)
-{
-	struct mmc_host *mmc = platform_get_drvdata(to_platform_device(dev));
-
-	return mmc_resume_host(mmc);
-}
-
-static const struct dev_pm_ops s3cmci_pm = {
-	.suspend	= s3cmci_suspend,
-	.resume		= s3cmci_resume,
-};
-
-#define s3cmci_pm_ops &s3cmci_pm
-#else /* CONFIG_PM */
-#define s3cmci_pm_ops NULL
-#endif /* CONFIG_PM */
-
-
 static struct platform_driver s3cmci_driver = {
 	.driver	= {
 		.name	= "s3c-sdi",
 		.owner	= THIS_MODULE,
-		.pm	= s3cmci_pm_ops,
 	},
 	.id_table	= s3cmci_driver_ids,
 	.probe		= s3cmci_probe,
diff --git a/drivers/mmc/host/sdhci-bcm-kona.c b/drivers/mmc/host/sdhci-bcm-kona.c
index 85472d3fd37f..7a190fe4dff1 100644
--- a/drivers/mmc/host/sdhci-bcm-kona.c
+++ b/drivers/mmc/host/sdhci-bcm-kona.c
@@ -316,19 +316,7 @@ err_pltfm_free:
 
 static int __exit sdhci_bcm_kona_remove(struct platform_device *pdev)
 {
-	struct sdhci_host *host = platform_get_drvdata(pdev);
-	int dead;
-	u32 scratch;
-
-	dead = 0;
-	scratch = readl(host->ioaddr + SDHCI_INT_STATUS);
-	if (scratch == (u32)-1)
-		dead = 1;
-	sdhci_remove_host(host, dead);
-
-	sdhci_free_host(host);
-
-	return 0;
+	return sdhci_pltfm_unregister(pdev);
 }
 
 static struct platform_driver sdhci_bcm_kona_driver = {
diff --git a/drivers/mmc/host/sdhci-bcm2835.c b/drivers/mmc/host/sdhci-bcm2835.c
index 36fa2df04660..f6d8d67c545f 100644
--- a/drivers/mmc/host/sdhci-bcm2835.c
+++ b/drivers/mmc/host/sdhci-bcm2835.c
@@ -178,13 +178,7 @@ err:
 
 static int bcm2835_sdhci_remove(struct platform_device *pdev)
 {
-	struct sdhci_host *host = platform_get_drvdata(pdev);
-	int dead = (readl(host->ioaddr + SDHCI_INT_STATUS) == 0xffffffff);
-
-	sdhci_remove_host(host, dead);
-	sdhci_pltfm_free(pdev);
-
-	return 0;
+	return sdhci_pltfm_unregister(pdev);
 }
 
 static const struct of_device_id bcm2835_sdhci_of_match[] = {
diff --git a/drivers/mmc/host/sdhci-esdhc-imx.c b/drivers/mmc/host/sdhci-esdhc-imx.c
index abc8cf01e6e3..461a4c3f4ef7 100644
--- a/drivers/mmc/host/sdhci-esdhc-imx.c
+++ b/drivers/mmc/host/sdhci-esdhc-imx.c
@@ -34,12 +34,40 @@
 /* VENDOR SPEC register */
 #define ESDHC_VENDOR_SPEC		0xc0
 #define  ESDHC_VENDOR_SPEC_SDIO_QUIRK	(1 << 1)
+#define  ESDHC_VENDOR_SPEC_VSELECT	(1 << 1)
+#define  ESDHC_VENDOR_SPEC_FRC_SDCLK_ON	(1 << 8)
 #define ESDHC_WTMK_LVL			0x44
 #define ESDHC_MIX_CTRL			0x48
+#define  ESDHC_MIX_CTRL_DDREN		(1 << 3)
 #define  ESDHC_MIX_CTRL_AC23EN		(1 << 7)
+#define  ESDHC_MIX_CTRL_EXE_TUNE	(1 << 22)
+#define  ESDHC_MIX_CTRL_SMPCLK_SEL	(1 << 23)
+#define  ESDHC_MIX_CTRL_FBCLK_SEL	(1 << 25)
 /* Bits 3 and 6 are not SDHCI standard definitions */
 #define  ESDHC_MIX_CTRL_SDHCI_MASK	0xb7
 
+/* dll control register */
+#define ESDHC_DLL_CTRL			0x60
+#define ESDHC_DLL_OVERRIDE_VAL_SHIFT	9
+#define ESDHC_DLL_OVERRIDE_EN_SHIFT	8
+
+/* tune control register */
+#define ESDHC_TUNE_CTRL_STATUS		0x68
+#define  ESDHC_TUNE_CTRL_STEP		1
+#define  ESDHC_TUNE_CTRL_MIN		0
+#define  ESDHC_TUNE_CTRL_MAX		((1 << 7) - 1)
+
+#define ESDHC_TUNING_CTRL		0xcc
+#define ESDHC_STD_TUNING_EN		(1 << 24)
+/* NOTE: the minimum valid tuning start tap for mx6sl is 1 */
+#define ESDHC_TUNING_START_TAP		0x1
+
+#define ESDHC_TUNING_BLOCK_PATTERN_LEN	64
+
+/* pinctrl state */
+#define ESDHC_PINCTRL_STATE_100MHZ	"state_100mhz"
+#define ESDHC_PINCTRL_STATE_200MHZ	"state_200mhz"
+
 /*
  * Our interpretation of the SDHCI_HOST_CONTROL register
  */
@@ -66,21 +94,60 @@
  * As a result, the TC flag is not asserted and SW  received timeout
  * exeception. Bit1 of Vendor Spec registor is used to fix it.
  */
-#define ESDHC_FLAG_MULTIBLK_NO_INT	(1 << 1)
-
-enum imx_esdhc_type {
-	IMX25_ESDHC,
-	IMX35_ESDHC,
-	IMX51_ESDHC,
-	IMX53_ESDHC,
-	IMX6Q_USDHC,
+#define ESDHC_FLAG_MULTIBLK_NO_INT	BIT(1)
+/*
+ * The flag enables the workaround for ESDHC errata ENGcm07207 which
+ * affects i.MX25 and i.MX35.
+ */
+#define ESDHC_FLAG_ENGCM07207		BIT(2)
+/*
+ * The flag tells that the ESDHC controller is an USDHC block that is
+ * integrated on the i.MX6 series.
+ */
+#define ESDHC_FLAG_USDHC		BIT(3)
+/* The IP supports manual tuning process */
+#define ESDHC_FLAG_MAN_TUNING		BIT(4)
+/* The IP supports standard tuning process */
+#define ESDHC_FLAG_STD_TUNING		BIT(5)
+/* The IP has SDHCI_CAPABILITIES_1 register */
+#define ESDHC_FLAG_HAVE_CAP1		BIT(6)
+
+struct esdhc_soc_data {
+	u32 flags;
+};
+
+static struct esdhc_soc_data esdhc_imx25_data = {
+	.flags = ESDHC_FLAG_ENGCM07207,
+};
+
+static struct esdhc_soc_data esdhc_imx35_data = {
+	.flags = ESDHC_FLAG_ENGCM07207,
+};
+
+static struct esdhc_soc_data esdhc_imx51_data = {
+	.flags = 0,
+};
+
+static struct esdhc_soc_data esdhc_imx53_data = {
+	.flags = ESDHC_FLAG_MULTIBLK_NO_INT,
+};
+
+static struct esdhc_soc_data usdhc_imx6q_data = {
+	.flags = ESDHC_FLAG_USDHC | ESDHC_FLAG_MAN_TUNING,
+};
+
+static struct esdhc_soc_data usdhc_imx6sl_data = {
+	.flags = ESDHC_FLAG_USDHC | ESDHC_FLAG_STD_TUNING
+			| ESDHC_FLAG_HAVE_CAP1,
 };
 
 struct pltfm_imx_data {
-	int flags;
 	u32 scratchpad;
-	enum imx_esdhc_type devtype;
 	struct pinctrl *pinctrl;
+	struct pinctrl_state *pins_default;
+	struct pinctrl_state *pins_100mhz;
+	struct pinctrl_state *pins_200mhz;
+	const struct esdhc_soc_data *socdata;
 	struct esdhc_platform_data boarddata;
 	struct clk *clk_ipg;
 	struct clk *clk_ahb;
@@ -90,25 +157,20 @@ struct pltfm_imx_data {
 		MULTIBLK_IN_PROCESS, /* exact multiblock cmd in process */
 		WAIT_FOR_INT,        /* sent CMD12, waiting for response INT */
 	} multiblock_status;
-
+	u32 uhs_mode;
+	u32 is_ddr;
 };
 
 static struct platform_device_id imx_esdhc_devtype[] = {
 	{
 		.name = "sdhci-esdhc-imx25",
-		.driver_data = IMX25_ESDHC,
+		.driver_data = (kernel_ulong_t) &esdhc_imx25_data,
 	}, {
 		.name = "sdhci-esdhc-imx35",
-		.driver_data = IMX35_ESDHC,
+		.driver_data = (kernel_ulong_t) &esdhc_imx35_data,
 	}, {
 		.name = "sdhci-esdhc-imx51",
-		.driver_data = IMX51_ESDHC,
-	}, {
-		.name = "sdhci-esdhc-imx53",
-		.driver_data = IMX53_ESDHC,
-	}, {
-		.name = "sdhci-usdhc-imx6q",
-		.driver_data = IMX6Q_USDHC,
+		.driver_data = (kernel_ulong_t) &esdhc_imx51_data,
 	}, {
 		/* sentinel */
 	}
@@ -116,38 +178,34 @@ static struct platform_device_id imx_esdhc_devtype[] = {
 MODULE_DEVICE_TABLE(platform, imx_esdhc_devtype);
 
 static const struct of_device_id imx_esdhc_dt_ids[] = {
-	{ .compatible = "fsl,imx25-esdhc", .data = &imx_esdhc_devtype[IMX25_ESDHC], },
-	{ .compatible = "fsl,imx35-esdhc", .data = &imx_esdhc_devtype[IMX35_ESDHC], },
-	{ .compatible = "fsl,imx51-esdhc", .data = &imx_esdhc_devtype[IMX51_ESDHC], },
-	{ .compatible = "fsl,imx53-esdhc", .data = &imx_esdhc_devtype[IMX53_ESDHC], },
-	{ .compatible = "fsl,imx6q-usdhc", .data = &imx_esdhc_devtype[IMX6Q_USDHC], },
+	{ .compatible = "fsl,imx25-esdhc", .data = &esdhc_imx25_data, },
+	{ .compatible = "fsl,imx35-esdhc", .data = &esdhc_imx35_data, },
+	{ .compatible = "fsl,imx51-esdhc", .data = &esdhc_imx51_data, },
+	{ .compatible = "fsl,imx53-esdhc", .data = &esdhc_imx53_data, },
+	{ .compatible = "fsl,imx6sl-usdhc", .data = &usdhc_imx6sl_data, },
+	{ .compatible = "fsl,imx6q-usdhc", .data = &usdhc_imx6q_data, },
 	{ /* sentinel */ }
 };
 MODULE_DEVICE_TABLE(of, imx_esdhc_dt_ids);
 
 static inline int is_imx25_esdhc(struct pltfm_imx_data *data)
 {
-	return data->devtype == IMX25_ESDHC;
-}
-
-static inline int is_imx35_esdhc(struct pltfm_imx_data *data)
-{
-	return data->devtype == IMX35_ESDHC;
+	return data->socdata == &esdhc_imx25_data;
 }
 
-static inline int is_imx51_esdhc(struct pltfm_imx_data *data)
+static inline int is_imx53_esdhc(struct pltfm_imx_data *data)
 {
-	return data->devtype == IMX51_ESDHC;
+	return data->socdata == &esdhc_imx53_data;
 }
 
-static inline int is_imx53_esdhc(struct pltfm_imx_data *data)
+static inline int is_imx6q_usdhc(struct pltfm_imx_data *data)
 {
-	return data->devtype == IMX53_ESDHC;
+	return data->socdata == &usdhc_imx6q_data;
 }
 
-static inline int is_imx6q_usdhc(struct pltfm_imx_data *data)
+static inline int esdhc_is_usdhc(struct pltfm_imx_data *data)
 {
-	return data->devtype == IMX6Q_USDHC;
+	return !!(data->socdata->flags & ESDHC_FLAG_USDHC);
 }
 
 static inline void esdhc_clrset_le(struct sdhci_host *host, u32 mask, u32 val, int reg)
@@ -164,7 +222,21 @@ static u32 esdhc_readl_le(struct sdhci_host *host, int reg)
 	struct pltfm_imx_data *imx_data = pltfm_host->priv;
 	u32 val = readl(host->ioaddr + reg);
 
+	if (unlikely(reg == SDHCI_PRESENT_STATE)) {
+		u32 fsl_prss = val;
+		/* save the least 20 bits */
+		val = fsl_prss & 0x000FFFFF;
+		/* move dat[0-3] bits */
+		val |= (fsl_prss & 0x0F000000) >> 4;
+		/* move cmd line bit */
+		val |= (fsl_prss & 0x00800000) << 1;
+	}
+
 	if (unlikely(reg == SDHCI_CAPABILITIES)) {
+		/* ignore bit[0-15] as it stores cap_1 register val for mx6sl */
+		if (imx_data->socdata->flags & ESDHC_FLAG_HAVE_CAP1)
+			val &= 0xffff0000;
+
 		/* In FSL esdhc IC module, only bit20 is used to indicate the
 		 * ADMA2 capability of esdhc, but this bit is messed up on
 		 * some SOCs (e.g. on MX25, MX35 this bit is set, but they
@@ -178,6 +250,25 @@ static u32 esdhc_readl_le(struct sdhci_host *host, int reg)
 		}
 	}
 
+	if (unlikely(reg == SDHCI_CAPABILITIES_1)) {
+		if (esdhc_is_usdhc(imx_data)) {
+			if (imx_data->socdata->flags & ESDHC_FLAG_HAVE_CAP1)
+				val = readl(host->ioaddr + SDHCI_CAPABILITIES) & 0xFFFF;
+			else
+				/* imx6q/dl does not have cap_1 register, fake one */
+				val = SDHCI_SUPPORT_DDR50 | SDHCI_SUPPORT_SDR104
+					| SDHCI_SUPPORT_SDR50
+					| SDHCI_USE_SDR50_TUNING;
+		}
+	}
+
+	if (unlikely(reg == SDHCI_MAX_CURRENT) && esdhc_is_usdhc(imx_data)) {
+		val = 0;
+		val |= 0xFF << SDHCI_MAX_CURRENT_330_SHIFT;
+		val |= 0xFF << SDHCI_MAX_CURRENT_300_SHIFT;
+		val |= 0xFF << SDHCI_MAX_CURRENT_180_SHIFT;
+	}
+
 	if (unlikely(reg == SDHCI_INT_STATUS)) {
 		if (val & ESDHC_INT_VENDOR_SPEC_DMA_ERR) {
 			val &= ~ESDHC_INT_VENDOR_SPEC_DMA_ERR;
@@ -224,7 +315,7 @@ static void esdhc_writel_le(struct sdhci_host *host, u32 val, int reg)
 		}
 	}
 
-	if (unlikely((imx_data->flags & ESDHC_FLAG_MULTIBLK_NO_INT)
+	if (unlikely((imx_data->socdata->flags & ESDHC_FLAG_MULTIBLK_NO_INT)
 				&& (reg == SDHCI_INT_STATUS)
 				&& (val & SDHCI_INT_DATA_END))) {
 			u32 v;
@@ -256,10 +347,12 @@ static u16 esdhc_readw_le(struct sdhci_host *host, int reg)
 {
 	struct sdhci_pltfm_host *pltfm_host = sdhci_priv(host);
 	struct pltfm_imx_data *imx_data = pltfm_host->priv;
+	u16 ret = 0;
+	u32 val;
 
 	if (unlikely(reg == SDHCI_HOST_VERSION)) {
 		reg ^= 2;
-		if (is_imx6q_usdhc(imx_data)) {
+		if (esdhc_is_usdhc(imx_data)) {
 			/*
 			 * The usdhc register returns a wrong host version.
 			 * Correct it here.
@@ -268,6 +361,30 @@ static u16 esdhc_readw_le(struct sdhci_host *host, int reg)
 		}
 	}
 
+	if (unlikely(reg == SDHCI_HOST_CONTROL2)) {
+		val = readl(host->ioaddr + ESDHC_VENDOR_SPEC);
+		if (val & ESDHC_VENDOR_SPEC_VSELECT)
+			ret |= SDHCI_CTRL_VDD_180;
+
+		if (esdhc_is_usdhc(imx_data)) {
+			if (imx_data->socdata->flags & ESDHC_FLAG_MAN_TUNING)
+				val = readl(host->ioaddr + ESDHC_MIX_CTRL);
+			else if (imx_data->socdata->flags & ESDHC_FLAG_STD_TUNING)
+				/* the std tuning bits is in ACMD12_ERR for imx6sl */
+				val = readl(host->ioaddr + SDHCI_ACMD12_ERR);
+		}
+
+		if (val & ESDHC_MIX_CTRL_EXE_TUNE)
+			ret |= SDHCI_CTRL_EXEC_TUNING;
+		if (val & ESDHC_MIX_CTRL_SMPCLK_SEL)
+			ret |= SDHCI_CTRL_TUNED_CLK;
+
+		ret |= (imx_data->uhs_mode & SDHCI_CTRL_UHS_MASK);
+		ret &= ~SDHCI_CTRL_PRESET_VAL_ENABLE;
+
+		return ret;
+	}
+
 	return readw(host->ioaddr + reg);
 }
 
@@ -275,10 +392,59 @@ static void esdhc_writew_le(struct sdhci_host *host, u16 val, int reg)
 {
 	struct sdhci_pltfm_host *pltfm_host = sdhci_priv(host);
 	struct pltfm_imx_data *imx_data = pltfm_host->priv;
+	u32 new_val = 0;
 
 	switch (reg) {
+	case SDHCI_CLOCK_CONTROL:
+		new_val = readl(host->ioaddr + ESDHC_VENDOR_SPEC);
+		if (val & SDHCI_CLOCK_CARD_EN)
+			new_val |= ESDHC_VENDOR_SPEC_FRC_SDCLK_ON;
+		else
+			new_val &= ~ESDHC_VENDOR_SPEC_FRC_SDCLK_ON;
+			writel(new_val, host->ioaddr + ESDHC_VENDOR_SPEC);
+		return;
+	case SDHCI_HOST_CONTROL2:
+		new_val = readl(host->ioaddr + ESDHC_VENDOR_SPEC);
+		if (val & SDHCI_CTRL_VDD_180)
+			new_val |= ESDHC_VENDOR_SPEC_VSELECT;
+		else
+			new_val &= ~ESDHC_VENDOR_SPEC_VSELECT;
+		writel(new_val, host->ioaddr + ESDHC_VENDOR_SPEC);
+		imx_data->uhs_mode = val & SDHCI_CTRL_UHS_MASK;
+		if (imx_data->socdata->flags & ESDHC_FLAG_MAN_TUNING) {
+			new_val = readl(host->ioaddr + ESDHC_MIX_CTRL);
+			if (val & SDHCI_CTRL_TUNED_CLK)
+				new_val |= ESDHC_MIX_CTRL_SMPCLK_SEL;
+			else
+				new_val &= ~ESDHC_MIX_CTRL_SMPCLK_SEL;
+			writel(new_val , host->ioaddr + ESDHC_MIX_CTRL);
+		} else if (imx_data->socdata->flags & ESDHC_FLAG_STD_TUNING) {
+			u32 v = readl(host->ioaddr + SDHCI_ACMD12_ERR);
+			u32 m = readl(host->ioaddr + ESDHC_MIX_CTRL);
+			new_val = readl(host->ioaddr + ESDHC_TUNING_CTRL);
+			if (val & SDHCI_CTRL_EXEC_TUNING) {
+				new_val |= ESDHC_STD_TUNING_EN |
+						ESDHC_TUNING_START_TAP;
+				v |= ESDHC_MIX_CTRL_EXE_TUNE;
+				m |= ESDHC_MIX_CTRL_FBCLK_SEL;
+			} else {
+				new_val &= ~ESDHC_STD_TUNING_EN;
+				v &= ~ESDHC_MIX_CTRL_EXE_TUNE;
+				m &= ~ESDHC_MIX_CTRL_FBCLK_SEL;
+			}
+
+			if (val & SDHCI_CTRL_TUNED_CLK)
+				v |= ESDHC_MIX_CTRL_SMPCLK_SEL;
+			else
+				v &= ~ESDHC_MIX_CTRL_SMPCLK_SEL;
+
+			writel(new_val, host->ioaddr + ESDHC_TUNING_CTRL);
+			writel(v, host->ioaddr + SDHCI_ACMD12_ERR);
+			writel(m, host->ioaddr + ESDHC_MIX_CTRL);
+		}
+		return;
 	case SDHCI_TRANSFER_MODE:
-		if ((imx_data->flags & ESDHC_FLAG_MULTIBLK_NO_INT)
+		if ((imx_data->socdata->flags & ESDHC_FLAG_MULTIBLK_NO_INT)
 				&& (host->cmd->opcode == SD_IO_RW_EXTENDED)
 				&& (host->cmd->data->blocks > 1)
 				&& (host->cmd->data->flags & MMC_DATA_READ)) {
@@ -288,7 +454,7 @@ static void esdhc_writew_le(struct sdhci_host *host, u16 val, int reg)
 			writel(v, host->ioaddr + ESDHC_VENDOR_SPEC);
 		}
 
-		if (is_imx6q_usdhc(imx_data)) {
+		if (esdhc_is_usdhc(imx_data)) {
 			u32 m = readl(host->ioaddr + ESDHC_MIX_CTRL);
 			/* Swap AC23 bit */
 			if (val & SDHCI_TRNS_AUTO_CMD23) {
@@ -310,10 +476,10 @@ static void esdhc_writew_le(struct sdhci_host *host, u16 val, int reg)
 			val |= SDHCI_CMD_ABORTCMD;
 
 		if ((host->cmd->opcode == MMC_SET_BLOCK_COUNT) &&
-		    (imx_data->flags & ESDHC_FLAG_MULTIBLK_NO_INT))
+		    (imx_data->socdata->flags & ESDHC_FLAG_MULTIBLK_NO_INT))
 			imx_data->multiblock_status = MULTIBLK_IN_PROCESS;
 
-		if (is_imx6q_usdhc(imx_data))
+		if (esdhc_is_usdhc(imx_data))
 			writel(val << 16,
 			       host->ioaddr + SDHCI_TRANSFER_MODE);
 		else
@@ -379,8 +545,10 @@ static void esdhc_writeb_le(struct sdhci_host *host, u8 val, int reg)
 		 * The reset on usdhc fails to clear MIX_CTRL register.
 		 * Do it manually here.
 		 */
-		if (is_imx6q_usdhc(imx_data))
+		if (esdhc_is_usdhc(imx_data)) {
 			writel(0, host->ioaddr + ESDHC_MIX_CTRL);
+			imx_data->is_ddr = 0;
+		}
 	}
 }
 
@@ -409,8 +577,60 @@ static inline void esdhc_pltfm_set_clock(struct sdhci_host *host,
 					 unsigned int clock)
 {
 	struct sdhci_pltfm_host *pltfm_host = sdhci_priv(host);
+	struct pltfm_imx_data *imx_data = pltfm_host->priv;
+	unsigned int host_clock = clk_get_rate(pltfm_host->clk);
+	int pre_div = 2;
+	int div = 1;
+	u32 temp, val;
+
+	if (clock == 0) {
+		if (esdhc_is_usdhc(imx_data)) {
+			val = readl(host->ioaddr + ESDHC_VENDOR_SPEC);
+			writel(val & ~ESDHC_VENDOR_SPEC_FRC_SDCLK_ON,
+					host->ioaddr + ESDHC_VENDOR_SPEC);
+		}
+		goto out;
+	}
+
+	if (esdhc_is_usdhc(imx_data) && !imx_data->is_ddr)
+		pre_div = 1;
+
+	temp = sdhci_readl(host, ESDHC_SYSTEM_CONTROL);
+	temp &= ~(ESDHC_CLOCK_IPGEN | ESDHC_CLOCK_HCKEN | ESDHC_CLOCK_PEREN
+		| ESDHC_CLOCK_MASK);
+	sdhci_writel(host, temp, ESDHC_SYSTEM_CONTROL);
+
+	while (host_clock / pre_div / 16 > clock && pre_div < 256)
+		pre_div *= 2;
+
+	while (host_clock / pre_div / div > clock && div < 16)
+		div++;
 
-	esdhc_set_clock(host, clock, clk_get_rate(pltfm_host->clk));
+	host->mmc->actual_clock = host_clock / pre_div / div;
+	dev_dbg(mmc_dev(host->mmc), "desired SD clock: %d, actual: %d\n",
+		clock, host->mmc->actual_clock);
+
+	if (imx_data->is_ddr)
+		pre_div >>= 2;
+	else
+		pre_div >>= 1;
+	div--;
+
+	temp = sdhci_readl(host, ESDHC_SYSTEM_CONTROL);
+	temp |= (ESDHC_CLOCK_IPGEN | ESDHC_CLOCK_HCKEN | ESDHC_CLOCK_PEREN
+		| (div << ESDHC_DIVIDER_SHIFT)
+		| (pre_div << ESDHC_PREDIV_SHIFT));
+	sdhci_writel(host, temp, ESDHC_SYSTEM_CONTROL);
+
+	if (esdhc_is_usdhc(imx_data)) {
+		val = readl(host->ioaddr + ESDHC_VENDOR_SPEC);
+		writel(val | ESDHC_VENDOR_SPEC_FRC_SDCLK_ON,
+		host->ioaddr + ESDHC_VENDOR_SPEC);
+	}
+
+	mdelay(1);
+out:
+	host->clock = clock;
 }
 
 static unsigned int esdhc_pltfm_get_ro(struct sdhci_host *host)
@@ -454,7 +674,192 @@ static int esdhc_pltfm_bus_width(struct sdhci_host *host, int width)
 	return 0;
 }
 
-static const struct sdhci_ops sdhci_esdhc_ops = {
+static void esdhc_prepare_tuning(struct sdhci_host *host, u32 val)
+{
+	u32 reg;
+
+	/* FIXME: delay a bit for card to be ready for next tuning due to errors */
+	mdelay(1);
+
+	reg = readl(host->ioaddr + ESDHC_MIX_CTRL);
+	reg |= ESDHC_MIX_CTRL_EXE_TUNE | ESDHC_MIX_CTRL_SMPCLK_SEL |
+			ESDHC_MIX_CTRL_FBCLK_SEL;
+	writel(reg, host->ioaddr + ESDHC_MIX_CTRL);
+	writel(val << 8, host->ioaddr + ESDHC_TUNE_CTRL_STATUS);
+	dev_dbg(mmc_dev(host->mmc),
+		"tunning with delay 0x%x ESDHC_TUNE_CTRL_STATUS 0x%x\n",
+			val, readl(host->ioaddr + ESDHC_TUNE_CTRL_STATUS));
+}
+
+static void esdhc_request_done(struct mmc_request *mrq)
+{
+	complete(&mrq->completion);
+}
+
+static int esdhc_send_tuning_cmd(struct sdhci_host *host, u32 opcode)
+{
+	struct mmc_command cmd = {0};
+	struct mmc_request mrq = {0};
+	struct mmc_data data = {0};
+	struct scatterlist sg;
+	char tuning_pattern[ESDHC_TUNING_BLOCK_PATTERN_LEN];
+
+	cmd.opcode = opcode;
+	cmd.arg = 0;
+	cmd.flags = MMC_RSP_R1 | MMC_CMD_ADTC;
+
+	data.blksz = ESDHC_TUNING_BLOCK_PATTERN_LEN;
+	data.blocks = 1;
+	data.flags = MMC_DATA_READ;
+	data.sg = &sg;
+	data.sg_len = 1;
+
+	sg_init_one(&sg, tuning_pattern, sizeof(tuning_pattern));
+
+	mrq.cmd = &cmd;
+	mrq.cmd->mrq = &mrq;
+	mrq.data = &data;
+	mrq.data->mrq = &mrq;
+	mrq.cmd->data = mrq.data;
+
+	mrq.done = esdhc_request_done;
+	init_completion(&(mrq.completion));
+
+	disable_irq(host->irq);
+	spin_lock(&host->lock);
+	host->mrq = &mrq;
+
+	sdhci_send_command(host, mrq.cmd);
+
+	spin_unlock(&host->lock);
+	enable_irq(host->irq);
+
+	wait_for_completion(&mrq.completion);
+
+	if (cmd.error)
+		return cmd.error;
+	if (data.error)
+		return data.error;
+
+	return 0;
+}
+
+static void esdhc_post_tuning(struct sdhci_host *host)
+{
+	u32 reg;
+
+	reg = readl(host->ioaddr + ESDHC_MIX_CTRL);
+	reg &= ~ESDHC_MIX_CTRL_EXE_TUNE;
+	writel(reg, host->ioaddr + ESDHC_MIX_CTRL);
+}
+
+static int esdhc_executing_tuning(struct sdhci_host *host, u32 opcode)
+{
+	int min, max, avg, ret;
+
+	/* find the mininum delay first which can pass tuning */
+	min = ESDHC_TUNE_CTRL_MIN;
+	while (min < ESDHC_TUNE_CTRL_MAX) {
+		esdhc_prepare_tuning(host, min);
+		if (!esdhc_send_tuning_cmd(host, opcode))
+			break;
+		min += ESDHC_TUNE_CTRL_STEP;
+	}
+
+	/* find the maxinum delay which can not pass tuning */
+	max = min + ESDHC_TUNE_CTRL_STEP;
+	while (max < ESDHC_TUNE_CTRL_MAX) {
+		esdhc_prepare_tuning(host, max);
+		if (esdhc_send_tuning_cmd(host, opcode)) {
+			max -= ESDHC_TUNE_CTRL_STEP;
+			break;
+		}
+		max += ESDHC_TUNE_CTRL_STEP;
+	}
+
+	/* use average delay to get the best timing */
+	avg = (min + max) / 2;
+	esdhc_prepare_tuning(host, avg);
+	ret = esdhc_send_tuning_cmd(host, opcode);
+	esdhc_post_tuning(host);
+
+	dev_dbg(mmc_dev(host->mmc), "tunning %s at 0x%x ret %d\n",
+		ret ? "failed" : "passed", avg, ret);
+
+	return ret;
+}
+
+static int esdhc_change_pinstate(struct sdhci_host *host,
+						unsigned int uhs)
+{
+	struct sdhci_pltfm_host *pltfm_host = sdhci_priv(host);
+	struct pltfm_imx_data *imx_data = pltfm_host->priv;
+	struct pinctrl_state *pinctrl;
+
+	dev_dbg(mmc_dev(host->mmc), "change pinctrl state for uhs %d\n", uhs);
+
+	if (IS_ERR(imx_data->pinctrl) ||
+		IS_ERR(imx_data->pins_default) ||
+		IS_ERR(imx_data->pins_100mhz) ||
+		IS_ERR(imx_data->pins_200mhz))
+		return -EINVAL;
+
+	switch (uhs) {
+	case MMC_TIMING_UHS_SDR50:
+		pinctrl = imx_data->pins_100mhz;
+		break;
+	case MMC_TIMING_UHS_SDR104:
+		pinctrl = imx_data->pins_200mhz;
+		break;
+	default:
+		/* back to default state for other legacy timing */
+		pinctrl = imx_data->pins_default;
+	}
+
+	return pinctrl_select_state(imx_data->pinctrl, pinctrl);
+}
+
+static int esdhc_set_uhs_signaling(struct sdhci_host *host, unsigned int uhs)
+{
+	struct sdhci_pltfm_host *pltfm_host = sdhci_priv(host);
+	struct pltfm_imx_data *imx_data = pltfm_host->priv;
+	struct esdhc_platform_data *boarddata = &imx_data->boarddata;
+
+	switch (uhs) {
+	case MMC_TIMING_UHS_SDR12:
+		imx_data->uhs_mode = SDHCI_CTRL_UHS_SDR12;
+		break;
+	case MMC_TIMING_UHS_SDR25:
+		imx_data->uhs_mode = SDHCI_CTRL_UHS_SDR25;
+		break;
+	case MMC_TIMING_UHS_SDR50:
+		imx_data->uhs_mode = SDHCI_CTRL_UHS_SDR50;
+		break;
+	case MMC_TIMING_UHS_SDR104:
+		imx_data->uhs_mode = SDHCI_CTRL_UHS_SDR104;
+		break;
+	case MMC_TIMING_UHS_DDR50:
+		imx_data->uhs_mode = SDHCI_CTRL_UHS_DDR50;
+		writel(readl(host->ioaddr + ESDHC_MIX_CTRL) |
+				ESDHC_MIX_CTRL_DDREN,
+				host->ioaddr + ESDHC_MIX_CTRL);
+		imx_data->is_ddr = 1;
+		if (boarddata->delay_line) {
+			u32 v;
+			v = boarddata->delay_line <<
+				ESDHC_DLL_OVERRIDE_VAL_SHIFT |
+				(1 << ESDHC_DLL_OVERRIDE_EN_SHIFT);
+			if (is_imx53_esdhc(imx_data))
+				v <<= 1;
+			writel(v, host->ioaddr + ESDHC_DLL_CTRL);
+		}
+		break;
+	}
+
+	return esdhc_change_pinstate(host, uhs);
+}
+
+static struct sdhci_ops sdhci_esdhc_ops = {
 	.read_l = esdhc_readl_le,
 	.read_w = esdhc_readw_le,
 	.write_l = esdhc_writel_le,
@@ -465,6 +870,7 @@ static const struct sdhci_ops sdhci_esdhc_ops = {
 	.get_min_clock = esdhc_pltfm_get_min_clock,
 	.get_ro = esdhc_pltfm_get_ro,
 	.platform_bus_width = esdhc_pltfm_bus_width,
+	.set_uhs_signaling = esdhc_set_uhs_signaling,
 };
 
 static const struct sdhci_pltfm_data sdhci_esdhc_imx_pdata = {
@@ -506,6 +912,14 @@ sdhci_esdhc_imx_probe_dt(struct platform_device *pdev,
 
 	of_property_read_u32(np, "max-frequency", &boarddata->f_max);
 
+	if (of_find_property(np, "no-1-8-v", NULL))
+		boarddata->support_vsel = false;
+	else
+		boarddata->support_vsel = true;
+
+	if (of_property_read_u32(np, "fsl,delay-line", &boarddata->delay_line))
+		boarddata->delay_line = 0;
+
 	return 0;
 }
 #else
@@ -539,9 +953,8 @@ static int sdhci_esdhc_imx_probe(struct platform_device *pdev)
 		goto free_sdhci;
 	}
 
-	if (of_id)
-		pdev->id_entry = of_id->data;
-	imx_data->devtype = pdev->id_entry->driver_data;
+	imx_data->socdata = of_id ? of_id->data : (struct esdhc_soc_data *)
+						  pdev->id_entry->driver_data;
 	pltfm_host->priv = imx_data;
 
 	imx_data->clk_ipg = devm_clk_get(&pdev->dev, "ipg");
@@ -568,29 +981,39 @@ static int sdhci_esdhc_imx_probe(struct platform_device *pdev)
 	clk_prepare_enable(imx_data->clk_ipg);
 	clk_prepare_enable(imx_data->clk_ahb);
 
-	imx_data->pinctrl = devm_pinctrl_get_select_default(&pdev->dev);
+	imx_data->pinctrl = devm_pinctrl_get(&pdev->dev);
 	if (IS_ERR(imx_data->pinctrl)) {
 		err = PTR_ERR(imx_data->pinctrl);
 		goto disable_clk;
 	}
 
+	imx_data->pins_default = pinctrl_lookup_state(imx_data->pinctrl,
+						PINCTRL_STATE_DEFAULT);
+	if (IS_ERR(imx_data->pins_default)) {
+		err = PTR_ERR(imx_data->pins_default);
+		dev_err(mmc_dev(host->mmc), "could not get default state\n");
+		goto disable_clk;
+	}
+
 	host->quirks |= SDHCI_QUIRK_BROKEN_TIMEOUT_VAL;
 
-	if (is_imx25_esdhc(imx_data) || is_imx35_esdhc(imx_data))
+	if (imx_data->socdata->flags & ESDHC_FLAG_ENGCM07207)
 		/* Fix errata ENGcm07207 present on i.MX25 and i.MX35 */
 		host->quirks |= SDHCI_QUIRK_NO_MULTIBLOCK
 			| SDHCI_QUIRK_BROKEN_ADMA;
 
-	if (is_imx53_esdhc(imx_data))
-		imx_data->flags |= ESDHC_FLAG_MULTIBLK_NO_INT;
-
 	/*
 	 * The imx6q ROM code will change the default watermark level setting
 	 * to something insane.  Change it back here.
 	 */
-	if (is_imx6q_usdhc(imx_data))
+	if (esdhc_is_usdhc(imx_data)) {
 		writel(0x08100810, host->ioaddr + ESDHC_WTMK_LVL);
+		host->quirks2 |= SDHCI_QUIRK2_PRESET_VALUE_BROKEN;
+	}
 
+	if (imx_data->socdata->flags & ESDHC_FLAG_MAN_TUNING)
+		sdhci_esdhc_ops.platform_execute_tuning =
+					esdhc_executing_tuning;
 	boarddata = &imx_data->boarddata;
 	if (sdhci_esdhc_imx_probe_dt(pdev, boarddata) < 0) {
 		if (!host->mmc->parent->platform_data) {
@@ -650,6 +1073,23 @@ static int sdhci_esdhc_imx_probe(struct platform_device *pdev)
 		break;
 	}
 
+	/* sdr50 and sdr104 needs work on 1.8v signal voltage */
+	if ((boarddata->support_vsel) && esdhc_is_usdhc(imx_data)) {
+		imx_data->pins_100mhz = pinctrl_lookup_state(imx_data->pinctrl,
+						ESDHC_PINCTRL_STATE_100MHZ);
+		imx_data->pins_200mhz = pinctrl_lookup_state(imx_data->pinctrl,
+						ESDHC_PINCTRL_STATE_200MHZ);
+		if (IS_ERR(imx_data->pins_100mhz) ||
+				IS_ERR(imx_data->pins_200mhz)) {
+			dev_warn(mmc_dev(host->mmc),
+				"could not get ultra high speed state, work on normal mode\n");
+			/* fall back to not support uhs by specify no 1.8v quirk */
+			host->quirks2 |= SDHCI_QUIRK2_NO_1_8_V;
+		}
+	} else {
+		host->quirks2 |= SDHCI_QUIRK2_NO_1_8_V;
+	}
+
 	err = sdhci_add_host(host);
 	if (err)
 		goto disable_clk;
diff --git a/drivers/mmc/host/sdhci-esdhc.h b/drivers/mmc/host/sdhci-esdhc.h
index a2a06420e463..a7d9f95a7b03 100644
--- a/drivers/mmc/host/sdhci-esdhc.h
+++ b/drivers/mmc/host/sdhci-esdhc.h
@@ -49,41 +49,4 @@
 
 #define ESDHC_HOST_CONTROL_RES	0x05
 
-static inline void esdhc_set_clock(struct sdhci_host *host, unsigned int clock,
-				   unsigned int host_clock)
-{
-	int pre_div = 2;
-	int div = 1;
-	u32 temp;
-
-	if (clock == 0)
-		goto out;
-
-	temp = sdhci_readl(host, ESDHC_SYSTEM_CONTROL);
-	temp &= ~(ESDHC_CLOCK_IPGEN | ESDHC_CLOCK_HCKEN | ESDHC_CLOCK_PEREN
-		| ESDHC_CLOCK_MASK);
-	sdhci_writel(host, temp, ESDHC_SYSTEM_CONTROL);
-
-	while (host_clock / pre_div / 16 > clock && pre_div < 256)
-		pre_div *= 2;
-
-	while (host_clock / pre_div / div > clock && div < 16)
-		div++;
-
-	dev_dbg(mmc_dev(host->mmc), "desired SD clock: %d, actual: %d\n",
-		clock, host_clock / pre_div / div);
-
-	pre_div >>= 1;
-	div--;
-
-	temp = sdhci_readl(host, ESDHC_SYSTEM_CONTROL);
-	temp |= (ESDHC_CLOCK_IPGEN | ESDHC_CLOCK_HCKEN | ESDHC_CLOCK_PEREN
-		| (div << ESDHC_DIVIDER_SHIFT)
-		| (pre_div << ESDHC_PREDIV_SHIFT));
-	sdhci_writel(host, temp, ESDHC_SYSTEM_CONTROL);
-	mdelay(1);
-out:
-	host->clock = clock;
-}
-
 #endif /* _DRIVERS_MMC_SDHCI_ESDHC_H */
diff --git a/drivers/mmc/host/sdhci-of-esdhc.c b/drivers/mmc/host/sdhci-of-esdhc.c
index e328252ebf2a..0b249970b119 100644
--- a/drivers/mmc/host/sdhci-of-esdhc.c
+++ b/drivers/mmc/host/sdhci-of-esdhc.c
@@ -199,6 +199,14 @@ static unsigned int esdhc_of_get_min_clock(struct sdhci_host *host)
 
 static void esdhc_of_set_clock(struct sdhci_host *host, unsigned int clock)
 {
+
+	int pre_div = 2;
+	int div = 1;
+	u32 temp;
+
+	if (clock == 0)
+		goto out;
+
 	/* Workaround to reduce the clock frequency for p1010 esdhc */
 	if (of_find_compatible_node(NULL, NULL, "fsl,p1010-esdhc")) {
 		if (clock > 20000000)
@@ -207,8 +215,31 @@ static void esdhc_of_set_clock(struct sdhci_host *host, unsigned int clock)
 			clock -= 5000000;
 	}
 
-	/* Set the clock */
-	esdhc_set_clock(host, clock, host->max_clk);
+	temp = sdhci_readl(host, ESDHC_SYSTEM_CONTROL);
+	temp &= ~(ESDHC_CLOCK_IPGEN | ESDHC_CLOCK_HCKEN | ESDHC_CLOCK_PEREN
+		| ESDHC_CLOCK_MASK);
+	sdhci_writel(host, temp, ESDHC_SYSTEM_CONTROL);
+
+	while (host->max_clk / pre_div / 16 > clock && pre_div < 256)
+		pre_div *= 2;
+
+	while (host->max_clk / pre_div / div > clock && div < 16)
+		div++;
+
+	dev_dbg(mmc_dev(host->mmc), "desired SD clock: %d, actual: %d\n",
+		clock, host->max_clk / pre_div / div);
+
+	pre_div >>= 1;
+	div--;
+
+	temp = sdhci_readl(host, ESDHC_SYSTEM_CONTROL);
+	temp |= (ESDHC_CLOCK_IPGEN | ESDHC_CLOCK_HCKEN | ESDHC_CLOCK_PEREN
+		| (div << ESDHC_DIVIDER_SHIFT)
+		| (pre_div << ESDHC_PREDIV_SHIFT));
+	sdhci_writel(host, temp, ESDHC_SYSTEM_CONTROL);
+	mdelay(1);
+out:
+	host->clock = clock;
 }
 
 #ifdef CONFIG_PM
diff --git a/drivers/mmc/host/sdhci-pci.c b/drivers/mmc/host/sdhci-pci.c
index d7d6bc8968d2..8f753811fc7a 100644
--- a/drivers/mmc/host/sdhci-pci.c
+++ b/drivers/mmc/host/sdhci-pci.c
@@ -37,6 +37,12 @@
 #define PCI_DEVICE_ID_INTEL_BYT_SDIO	0x0f15
 #define PCI_DEVICE_ID_INTEL_BYT_SD	0x0f16
 #define PCI_DEVICE_ID_INTEL_BYT_EMMC2	0x0f50
+#define PCI_DEVICE_ID_INTEL_MRFL_MMC	0x1190
+#define PCI_DEVICE_ID_INTEL_CLV_SDIO0	0x08f9
+#define PCI_DEVICE_ID_INTEL_CLV_SDIO1	0x08fa
+#define PCI_DEVICE_ID_INTEL_CLV_SDIO2	0x08fb
+#define PCI_DEVICE_ID_INTEL_CLV_EMMC0	0x08e5
+#define PCI_DEVICE_ID_INTEL_CLV_EMMC1	0x08e6
 
 /*
  * PCI registers
@@ -356,6 +362,28 @@ static const struct sdhci_pci_fixes sdhci_intel_byt_sd = {
 	.allow_runtime_pm = true,
 };
 
+/* Define Host controllers for Intel Merrifield platform */
+#define INTEL_MRFL_EMMC_0	0
+#define INTEL_MRFL_EMMC_1	1
+
+static int intel_mrfl_mmc_probe_slot(struct sdhci_pci_slot *slot)
+{
+	if ((PCI_FUNC(slot->chip->pdev->devfn) != INTEL_MRFL_EMMC_0) &&
+	    (PCI_FUNC(slot->chip->pdev->devfn) != INTEL_MRFL_EMMC_1))
+		/* SD support is not ready yet */
+		return -ENODEV;
+
+	slot->host->mmc->caps |= MMC_CAP_8_BIT_DATA | MMC_CAP_NONREMOVABLE |
+				 MMC_CAP_1_8V_DDR;
+
+	return 0;
+}
+
+static const struct sdhci_pci_fixes sdhci_intel_mrfl_mmc = {
+	.quirks		= SDHCI_QUIRK_NO_ENDATTR_IN_NOPDESC,
+	.probe_slot	= intel_mrfl_mmc_probe_slot,
+};
+
 /* O2Micro extra registers */
 #define O2_SD_LOCK_WP		0xD3
 #define O2_SD_MULTI_VCC3V	0xEE
@@ -939,6 +967,54 @@ static const struct pci_device_id pci_ids[] = {
 		.driver_data	= (kernel_ulong_t)&sdhci_intel_byt_emmc,
 	},
 
+
+	{
+		.vendor		= PCI_VENDOR_ID_INTEL,
+		.device		= PCI_DEVICE_ID_INTEL_CLV_SDIO0,
+		.subvendor	= PCI_ANY_ID,
+		.subdevice	= PCI_ANY_ID,
+		.driver_data	= (kernel_ulong_t)&sdhci_intel_mfd_sd,
+	},
+
+	{
+		.vendor		= PCI_VENDOR_ID_INTEL,
+		.device		= PCI_DEVICE_ID_INTEL_CLV_SDIO1,
+		.subvendor	= PCI_ANY_ID,
+		.subdevice	= PCI_ANY_ID,
+		.driver_data	= (kernel_ulong_t)&sdhci_intel_mfd_sdio,
+	},
+
+	{
+		.vendor		= PCI_VENDOR_ID_INTEL,
+		.device		= PCI_DEVICE_ID_INTEL_CLV_SDIO2,
+		.subvendor	= PCI_ANY_ID,
+		.subdevice	= PCI_ANY_ID,
+		.driver_data	= (kernel_ulong_t)&sdhci_intel_mfd_sdio,
+	},
+
+	{
+		.vendor		= PCI_VENDOR_ID_INTEL,
+		.device		= PCI_DEVICE_ID_INTEL_CLV_EMMC0,
+		.subvendor	= PCI_ANY_ID,
+		.subdevice	= PCI_ANY_ID,
+		.driver_data	= (kernel_ulong_t)&sdhci_intel_mfd_emmc,
+	},
+
+	{
+		.vendor		= PCI_VENDOR_ID_INTEL,
+		.device		= PCI_DEVICE_ID_INTEL_CLV_EMMC1,
+		.subvendor	= PCI_ANY_ID,
+		.subdevice	= PCI_ANY_ID,
+		.driver_data	= (kernel_ulong_t)&sdhci_intel_mfd_emmc,
+	},
+
+	{
+		.vendor		= PCI_VENDOR_ID_INTEL,
+		.device		= PCI_DEVICE_ID_INTEL_MRFL_MMC,
+		.subvendor	= PCI_ANY_ID,
+		.subdevice	= PCI_ANY_ID,
+		.driver_data	= (kernel_ulong_t)&sdhci_intel_mrfl_mmc,
+	},
 	{
 		.vendor		= PCI_VENDOR_ID_O2,
 		.device		= PCI_DEVICE_ID_O2_8120,
diff --git a/drivers/mmc/host/sdhci.c b/drivers/mmc/host/sdhci.c
index 7a7fb4f0d5a4..bd8a0982aec3 100644
--- a/drivers/mmc/host/sdhci.c
+++ b/drivers/mmc/host/sdhci.c
@@ -49,7 +49,6 @@ static unsigned int debug_quirks2;
 
 static void sdhci_finish_data(struct sdhci_host *);
 
-static void sdhci_send_command(struct sdhci_host *, struct mmc_command *);
 static void sdhci_finish_command(struct sdhci_host *);
 static int sdhci_execute_tuning(struct mmc_host *mmc, u32 opcode);
 static void sdhci_tuning_timer(unsigned long data);
@@ -981,7 +980,7 @@ static void sdhci_finish_data(struct sdhci_host *host)
 		tasklet_schedule(&host->finish_tasklet);
 }
 
-static void sdhci_send_command(struct sdhci_host *host, struct mmc_command *cmd)
+void sdhci_send_command(struct sdhci_host *host, struct mmc_command *cmd)
 {
 	int flags;
 	u32 mask;
@@ -1053,6 +1052,7 @@ static void sdhci_send_command(struct sdhci_host *host, struct mmc_command *cmd)
 
 	sdhci_writew(host, SDHCI_MAKE_CMD(cmd->opcode, flags), SDHCI_COMMAND);
 }
+EXPORT_SYMBOL_GPL(sdhci_send_command);
 
 static void sdhci_finish_command(struct sdhci_host *host)
 {
@@ -1435,7 +1435,8 @@ static void sdhci_do_set_ios(struct sdhci_host *host, struct mmc_ios *ios)
 	}
 
 	if (host->version >= SDHCI_SPEC_300 &&
-		(ios->power_mode == MMC_POWER_UP))
+		(ios->power_mode == MMC_POWER_UP) &&
+		!(host->quirks2 & SDHCI_QUIRK2_PRESET_VALUE_BROKEN))
 		sdhci_enable_preset_value(host, false);
 
 	sdhci_set_clock(host, ios->clock);
@@ -1875,6 +1876,14 @@ static int sdhci_execute_tuning(struct mmc_host *mmc, u32 opcode)
 		return 0;
 	}
 
+	if (host->ops->platform_execute_tuning) {
+		spin_unlock(&host->lock);
+		enable_irq(host->irq);
+		err = host->ops->platform_execute_tuning(host, opcode);
+		sdhci_runtime_pm_put(host);
+		return err;
+	}
+
 	sdhci_writew(host, ctrl, SDHCI_HOST_CONTROL2);
 
 	/*
@@ -1981,6 +1990,7 @@ static int sdhci_execute_tuning(struct mmc_host *mmc, u32 opcode)
 	if (!tuning_loop_counter || !timeout) {
 		ctrl &= ~SDHCI_CTRL_TUNED_CLK;
 		sdhci_writew(host, ctrl, SDHCI_HOST_CONTROL2);
+		err = -EIO;
 	} else {
 		if (!(ctrl & SDHCI_CTRL_TUNED_CLK)) {
 			pr_info(DRIVER_NAME ": Tuning procedure"
@@ -2491,6 +2501,14 @@ again:
 	result = IRQ_HANDLED;
 
 	intmask = sdhci_readl(host, SDHCI_INT_STATUS);
+
+	/*
+	 * If we know we'll call the driver to signal SDIO IRQ, disregard
+	 * further indications of Card Interrupt in the status to avoid a
+	 * needless loop.
+	 */
+	if (cardint)
+		intmask &= ~SDHCI_INT_CARD_INT;
 	if (intmask && --max_loops)
 		goto again;
 out:
@@ -2546,8 +2564,6 @@ EXPORT_SYMBOL_GPL(sdhci_disable_irq_wakeups);
 
 int sdhci_suspend_host(struct sdhci_host *host)
 {
-	int ret;
-
 	if (host->ops->platform_suspend)
 		host->ops->platform_suspend(host);
 
@@ -2559,19 +2575,6 @@ int sdhci_suspend_host(struct sdhci_host *host)
 		host->flags &= ~SDHCI_NEEDS_RETUNING;
 	}
 
-	ret = mmc_suspend_host(host->mmc);
-	if (ret) {
-		if (host->flags & SDHCI_USING_RETUNING_TIMER) {
-			host->flags |= SDHCI_NEEDS_RETUNING;
-			mod_timer(&host->tuning_timer, jiffies +
-					host->tuning_count * HZ);
-		}
-
-		sdhci_enable_card_detection(host);
-
-		return ret;
-	}
-
 	if (!device_may_wakeup(mmc_dev(host->mmc))) {
 		sdhci_mask_irqs(host, SDHCI_INT_ALL_MASK);
 		free_irq(host->irq, host);
@@ -2579,14 +2582,14 @@ int sdhci_suspend_host(struct sdhci_host *host)
 		sdhci_enable_irq_wakeups(host);
 		enable_irq_wake(host->irq);
 	}
-	return ret;
+	return 0;
 }
 
 EXPORT_SYMBOL_GPL(sdhci_suspend_host);
 
 int sdhci_resume_host(struct sdhci_host *host)
 {
-	int ret;
+	int ret = 0;
 
 	if (host->flags & (SDHCI_USE_SDMA | SDHCI_USE_ADMA)) {
 		if (host->ops->enable_dma)
@@ -2615,7 +2618,6 @@ int sdhci_resume_host(struct sdhci_host *host)
 		mmiowb();
 	}
 
-	ret = mmc_resume_host(host->mmc);
 	sdhci_enable_card_detection(host);
 
 	if (host->ops->platform_resume)
diff --git a/drivers/mmc/host/sdhci.h b/drivers/mmc/host/sdhci.h
index b037f188fe44..0a3ed01887db 100644
--- a/drivers/mmc/host/sdhci.h
+++ b/drivers/mmc/host/sdhci.h
@@ -288,6 +288,7 @@ struct sdhci_ops {
 	unsigned int    (*get_ro)(struct sdhci_host *host);
 	void	(*platform_reset_enter)(struct sdhci_host *host, u8 mask);
 	void	(*platform_reset_exit)(struct sdhci_host *host, u8 mask);
+	int	(*platform_execute_tuning)(struct sdhci_host *host, u32 opcode);
 	int	(*set_uhs_signaling)(struct sdhci_host *host, unsigned int uhs);
 	void	(*hw_reset)(struct sdhci_host *host);
 	void	(*platform_suspend)(struct sdhci_host *host);
@@ -393,6 +394,8 @@ static inline void *sdhci_priv(struct sdhci_host *host)
 extern void sdhci_card_detect(struct sdhci_host *host);
 extern int sdhci_add_host(struct sdhci_host *host);
 extern void sdhci_remove_host(struct sdhci_host *host, int dead);
+extern void sdhci_send_command(struct sdhci_host *host,
+				struct mmc_command *cmd);
 
 #ifdef CONFIG_PM
 extern int sdhci_suspend_host(struct sdhci_host *host);
diff --git a/drivers/mmc/host/sdricoh_cs.c b/drivers/mmc/host/sdricoh_cs.c
index 50adbd155f35..b7e305775314 100644
--- a/drivers/mmc/host/sdricoh_cs.c
+++ b/drivers/mmc/host/sdricoh_cs.c
@@ -516,9 +516,7 @@ static void sdricoh_pcmcia_detach(struct pcmcia_device *link)
 #ifdef CONFIG_PM
 static int sdricoh_pcmcia_suspend(struct pcmcia_device *link)
 {
-	struct mmc_host *mmc = link->priv;
 	dev_dbg(&link->dev, "suspend\n");
-	mmc_suspend_host(mmc);
 	return 0;
 }
 
@@ -527,7 +525,6 @@ static int sdricoh_pcmcia_resume(struct pcmcia_device *link)
 	struct mmc_host *mmc = link->priv;
 	dev_dbg(&link->dev, "resume\n");
 	sdricoh_reset(mmc_priv(mmc));
-	mmc_resume_host(mmc);
 	return 0;
 }
 #else
diff --git a/drivers/mmc/host/sh_mmcif.c b/drivers/mmc/host/sh_mmcif.c
index 36629a024aa1..d032b080ac4d 100644
--- a/drivers/mmc/host/sh_mmcif.c
+++ b/drivers/mmc/host/sh_mmcif.c
@@ -964,7 +964,7 @@ static void sh_mmcif_request(struct mmc_host *mmc, struct mmc_request *mrq)
 
 static int sh_mmcif_clk_update(struct sh_mmcif_host *host)
 {
-	int ret = clk_enable(host->hclk);
+	int ret = clk_prepare_enable(host->hclk);
 
 	if (!ret) {
 		host->clk = clk_get_rate(host->hclk);
@@ -1018,7 +1018,7 @@ static void sh_mmcif_set_ios(struct mmc_host *mmc, struct mmc_ios *ios)
 		}
 		if (host->power) {
 			pm_runtime_put_sync(&host->pd->dev);
-			clk_disable(host->hclk);
+			clk_disable_unprepare(host->hclk);
 			host->power = false;
 			if (ios->power_mode == MMC_POWER_OFF)
 				sh_mmcif_set_power(host, ios);
@@ -1466,7 +1466,7 @@ static int sh_mmcif_probe(struct platform_device *pdev)
 
 	mutex_init(&host->thread_lock);
 
-	clk_disable(host->hclk);
+	clk_disable_unprepare(host->hclk);
 	ret = mmc_add_host(mmc);
 	if (ret < 0)
 		goto emmcaddh;
@@ -1487,7 +1487,7 @@ ereqirq1:
 ereqirq0:
 	pm_runtime_suspend(&pdev->dev);
 eresume:
-	clk_disable(host->hclk);
+	clk_disable_unprepare(host->hclk);
 eclkupdate:
 	clk_put(host->hclk);
 eclkget:
@@ -1505,7 +1505,7 @@ static int sh_mmcif_remove(struct platform_device *pdev)
 	int irq[2];
 
 	host->dying = true;
-	clk_enable(host->hclk);
+	clk_prepare_enable(host->hclk);
 	pm_runtime_get_sync(&pdev->dev);
 
 	dev_pm_qos_hide_latency_limit(&pdev->dev);
@@ -1530,7 +1530,7 @@ static int sh_mmcif_remove(struct platform_device *pdev)
 	if (irq[1] >= 0)
 		free_irq(irq[1], host);
 
-	clk_disable(host->hclk);
+	clk_disable_unprepare(host->hclk);
 	mmc_free_host(host->mmc);
 	pm_runtime_put_sync(&pdev->dev);
 	pm_runtime_disable(&pdev->dev);
@@ -1538,28 +1538,21 @@ static int sh_mmcif_remove(struct platform_device *pdev)
 	return 0;
 }
 
-#ifdef CONFIG_PM
+#ifdef CONFIG_PM_SLEEP
 static int sh_mmcif_suspend(struct device *dev)
 {
 	struct sh_mmcif_host *host = dev_get_drvdata(dev);
-	int ret = mmc_suspend_host(host->mmc);
 
-	if (!ret)
-		sh_mmcif_writel(host->addr, MMCIF_CE_INT_MASK, MASK_ALL);
+	sh_mmcif_writel(host->addr, MMCIF_CE_INT_MASK, MASK_ALL);
 
-	return ret;
+	return 0;
 }
 
 static int sh_mmcif_resume(struct device *dev)
 {
-	struct sh_mmcif_host *host = dev_get_drvdata(dev);
-
-	return mmc_resume_host(host->mmc);
+	return 0;
 }
-#else
-#define sh_mmcif_suspend	NULL
-#define sh_mmcif_resume		NULL
-#endif	/* CONFIG_PM */
+#endif
 
 static const struct of_device_id mmcif_of_match[] = {
 	{ .compatible = "renesas,sh-mmcif" },
@@ -1568,8 +1561,7 @@ static const struct of_device_id mmcif_of_match[] = {
 MODULE_DEVICE_TABLE(of, mmcif_of_match);
 
 static const struct dev_pm_ops sh_mmcif_dev_pm_ops = {
-	.suspend = sh_mmcif_suspend,
-	.resume = sh_mmcif_resume,
+	SET_SYSTEM_SLEEP_PM_OPS(sh_mmcif_suspend, sh_mmcif_resume)
 };
 
 static struct platform_driver sh_mmcif_driver = {
diff --git a/drivers/mmc/host/tifm_sd.c b/drivers/mmc/host/tifm_sd.c
index 43d962829f8e..d1760ebcac03 100644
--- a/drivers/mmc/host/tifm_sd.c
+++ b/drivers/mmc/host/tifm_sd.c
@@ -1030,7 +1030,7 @@ static void tifm_sd_remove(struct tifm_dev *sock)
 
 static int tifm_sd_suspend(struct tifm_dev *sock, pm_message_t state)
 {
-	return mmc_suspend_host(tifm_get_drvdata(sock));
+	return 0;
 }
 
 static int tifm_sd_resume(struct tifm_dev *sock)
@@ -1044,8 +1044,6 @@ static int tifm_sd_resume(struct tifm_dev *sock)
 
 	if (rc)
 		host->eject = 1;
-	else
-		rc = mmc_resume_host(mmc);
 
 	return rc;
 }
diff --git a/drivers/mmc/host/tmio_mmc_pio.c b/drivers/mmc/host/tmio_mmc_pio.c
index b3802256f954..f3b2d8ca1eca 100644
--- a/drivers/mmc/host/tmio_mmc_pio.c
+++ b/drivers/mmc/host/tmio_mmc_pio.c
@@ -1145,12 +1145,9 @@ int tmio_mmc_host_suspend(struct device *dev)
 {
 	struct mmc_host *mmc = dev_get_drvdata(dev);
 	struct tmio_mmc_host *host = mmc_priv(mmc);
-	int ret = mmc_suspend_host(mmc);
 
-	if (!ret)
-		tmio_mmc_disable_mmc_irqs(host, TMIO_MASK_ALL);
-
-	return ret;
+	tmio_mmc_disable_mmc_irqs(host, TMIO_MASK_ALL);
+	return 0;
 }
 EXPORT_SYMBOL(tmio_mmc_host_suspend);
 
@@ -1163,7 +1160,7 @@ int tmio_mmc_host_resume(struct device *dev)
 
 	/* The MMC core will perform the complete set up */
 	host->resuming = true;
-	return mmc_resume_host(mmc);
+	return 0;
 }
 EXPORT_SYMBOL(tmio_mmc_host_resume);
 
diff --git a/drivers/mmc/host/via-sdmmc.c b/drivers/mmc/host/via-sdmmc.c
index 4f84586c6e9e..63fac78b3d46 100644
--- a/drivers/mmc/host/via-sdmmc.c
+++ b/drivers/mmc/host/via-sdmmc.c
@@ -1269,21 +1269,18 @@ static void via_init_sdc_pm(struct via_crdr_mmc_host *host)
 static int via_sd_suspend(struct pci_dev *pcidev, pm_message_t state)
 {
 	struct via_crdr_mmc_host *host;
-	int ret = 0;
 
 	host = pci_get_drvdata(pcidev);
 
 	via_save_pcictrlreg(host);
 	via_save_sdcreg(host);
 
-	ret = mmc_suspend_host(host->mmc);
-
 	pci_save_state(pcidev);
 	pci_enable_wake(pcidev, pci_choose_state(pcidev, state), 0);
 	pci_disable_device(pcidev);
 	pci_set_power_state(pcidev, pci_choose_state(pcidev, state));
 
-	return ret;
+	return 0;
 }
 
 static int via_sd_resume(struct pci_dev *pcidev)
@@ -1316,8 +1313,6 @@ static int via_sd_resume(struct pci_dev *pcidev)
 	via_restore_pcictrlreg(sdhost);
 	via_init_sdc_pm(sdhost);
 
-	ret = mmc_resume_host(sdhost->mmc);
-
 	return ret;
 }
 
diff --git a/drivers/mmc/host/vub300.c b/drivers/mmc/host/vub300.c
index e9028ad05ffb..4262296c12fa 100644
--- a/drivers/mmc/host/vub300.c
+++ b/drivers/mmc/host/vub300.c
@@ -2392,26 +2392,12 @@ static void vub300_disconnect(struct usb_interface *interface)
 #ifdef CONFIG_PM
 static int vub300_suspend(struct usb_interface *intf, pm_message_t message)
 {
-	struct vub300_mmc_host *vub300 = usb_get_intfdata(intf);
-	if (!vub300 || !vub300->mmc) {
-		return 0;
-	} else {
-		struct mmc_host *mmc = vub300->mmc;
-		mmc_suspend_host(mmc);
-		return 0;
-	}
+	return 0;
 }
 
 static int vub300_resume(struct usb_interface *intf)
 {
-	struct vub300_mmc_host *vub300 = usb_get_intfdata(intf);
-	if (!vub300 || !vub300->mmc) {
-		return 0;
-	} else {
-		struct mmc_host *mmc = vub300->mmc;
-		mmc_resume_host(mmc);
-		return 0;
-	}
+	return 0;
 }
 #else
 #define vub300_suspend NULL
diff --git a/drivers/mmc/host/wbsd.c b/drivers/mmc/host/wbsd.c
index e954b7758876..1defd5ed3236 100644
--- a/drivers/mmc/host/wbsd.c
+++ b/drivers/mmc/host/wbsd.c
@@ -1814,28 +1814,11 @@ static void wbsd_pnp_remove(struct pnp_dev *dev)
 
 #ifdef CONFIG_PM
 
-static int wbsd_suspend(struct wbsd_host *host, pm_message_t state)
-{
-	BUG_ON(host == NULL);
-
-	return mmc_suspend_host(host->mmc);
-}
-
-static int wbsd_resume(struct wbsd_host *host)
-{
-	BUG_ON(host == NULL);
-
-	wbsd_init_device(host);
-
-	return mmc_resume_host(host->mmc);
-}
-
 static int wbsd_platform_suspend(struct platform_device *dev,
 				 pm_message_t state)
 {
 	struct mmc_host *mmc = platform_get_drvdata(dev);
 	struct wbsd_host *host;
-	int ret;
 
 	if (mmc == NULL)
 		return 0;
@@ -1844,12 +1827,7 @@ static int wbsd_platform_suspend(struct platform_device *dev,
 
 	host = mmc_priv(mmc);
 
-	ret = wbsd_suspend(host, state);
-	if (ret)
-		return ret;
-
 	wbsd_chip_poweroff(host);
-
 	return 0;
 }
 
@@ -1872,7 +1850,8 @@ static int wbsd_platform_resume(struct platform_device *dev)
 	 */
 	mdelay(5);
 
-	return wbsd_resume(host);
+	wbsd_init_device(host);
+	return 0;
 }
 
 #ifdef CONFIG_PNP
@@ -1880,16 +1859,12 @@ static int wbsd_platform_resume(struct platform_device *dev)
 static int wbsd_pnp_suspend(struct pnp_dev *pnp_dev, pm_message_t state)
 {
 	struct mmc_host *mmc = dev_get_drvdata(&pnp_dev->dev);
-	struct wbsd_host *host;
 
 	if (mmc == NULL)
 		return 0;
 
 	DBGF("Suspending...\n");
-
-	host = mmc_priv(mmc);
-
-	return wbsd_suspend(host, state);
+	return 0;
 }
 
 static int wbsd_pnp_resume(struct pnp_dev *pnp_dev)
@@ -1922,7 +1897,8 @@ static int wbsd_pnp_resume(struct pnp_dev *pnp_dev)
 	 */
 	mdelay(5);
 
-	return wbsd_resume(host);
+	wbsd_init_device(host);
+	return 0;
 }
 
 #endif /* CONFIG_PNP */
diff --git a/drivers/mmc/host/wmt-sdmmc.c b/drivers/mmc/host/wmt-sdmmc.c
index 34231d5168fc..e902ed7846b0 100644
--- a/drivers/mmc/host/wmt-sdmmc.c
+++ b/drivers/mmc/host/wmt-sdmmc.c
@@ -212,28 +212,14 @@ struct wmt_mci_priv {
 
 static void wmt_set_sd_power(struct wmt_mci_priv *priv, int enable)
 {
-	u32 reg_tmp;
-	if (enable) {
-		if (priv->power_inverted) {
-			reg_tmp = readb(priv->sdmmc_base + SDMMC_BUSMODE);
-			writeb(reg_tmp | BM_SD_OFF,
-			       priv->sdmmc_base + SDMMC_BUSMODE);
-		} else {
-			reg_tmp = readb(priv->sdmmc_base + SDMMC_BUSMODE);
-			writeb(reg_tmp & (~BM_SD_OFF),
-			       priv->sdmmc_base + SDMMC_BUSMODE);
-		}
-	} else {
-		if (priv->power_inverted) {
-			reg_tmp = readb(priv->sdmmc_base + SDMMC_BUSMODE);
-			writeb(reg_tmp & (~BM_SD_OFF),
-			       priv->sdmmc_base + SDMMC_BUSMODE);
-		} else {
-			reg_tmp = readb(priv->sdmmc_base + SDMMC_BUSMODE);
-			writeb(reg_tmp | BM_SD_OFF,
-			       priv->sdmmc_base + SDMMC_BUSMODE);
-		}
-	}
+	u32 reg_tmp = readb(priv->sdmmc_base + SDMMC_BUSMODE);
+
+	if (enable ^ priv->power_inverted)
+		reg_tmp &= ~BM_SD_OFF;
+	else
+		reg_tmp |= BM_SD_OFF;
+
+	writeb(reg_tmp, priv->sdmmc_base + SDMMC_BUSMODE);
 }
 
 static void wmt_mci_read_response(struct mmc_host *mmc)
@@ -939,28 +925,23 @@ static int wmt_mci_suspend(struct device *dev)
 	struct platform_device *pdev = to_platform_device(dev);
 	struct mmc_host *mmc = platform_get_drvdata(pdev);
 	struct wmt_mci_priv *priv;
-	int ret;
 
 	if (!mmc)
 		return 0;
 
 	priv = mmc_priv(mmc);
-	ret = mmc_suspend_host(mmc);
-
-	if (!ret) {
-		reg_tmp = readb(priv->sdmmc_base + SDMMC_BUSMODE);
-		writeb(reg_tmp | BM_SOFT_RESET, priv->sdmmc_base +
-		       SDMMC_BUSMODE);
+	reg_tmp = readb(priv->sdmmc_base + SDMMC_BUSMODE);
+	writeb(reg_tmp | BM_SOFT_RESET, priv->sdmmc_base +
+	       SDMMC_BUSMODE);
 
-		reg_tmp = readw(priv->sdmmc_base + SDMMC_BLKLEN);
-		writew(reg_tmp & 0x5FFF, priv->sdmmc_base + SDMMC_BLKLEN);
+	reg_tmp = readw(priv->sdmmc_base + SDMMC_BLKLEN);
+	writew(reg_tmp & 0x5FFF, priv->sdmmc_base + SDMMC_BLKLEN);
 
-		writeb(0xFF, priv->sdmmc_base + SDMMC_STS0);
-		writeb(0xFF, priv->sdmmc_base + SDMMC_STS1);
+	writeb(0xFF, priv->sdmmc_base + SDMMC_STS0);
+	writeb(0xFF, priv->sdmmc_base + SDMMC_STS1);
 
-		clk_disable(priv->clk_sdmmc);
-	}
-	return ret;
+	clk_disable(priv->clk_sdmmc);
+	return 0;
 }
 
 static int wmt_mci_resume(struct device *dev)
@@ -969,7 +950,6 @@ static int wmt_mci_resume(struct device *dev)
 	struct platform_device *pdev = to_platform_device(dev);
 	struct mmc_host *mmc = platform_get_drvdata(pdev);
 	struct wmt_mci_priv *priv;
-	int ret = 0;
 
 	if (mmc) {
 		priv = mmc_priv(mmc);
@@ -987,10 +967,9 @@ static int wmt_mci_resume(struct device *dev)
 		writeb(reg_tmp | INT0_DI_INT_EN, priv->sdmmc_base +
 		       SDMMC_INTMASK0);
 
-		ret = mmc_resume_host(mmc);
 	}
 
-	return ret;
+	return 0;
 }
 
 static const struct dev_pm_ops wmt_mci_pm = {
diff --git a/drivers/mtd/nand/atmel_nand.c b/drivers/mtd/nand/atmel_nand.c
index d78a97d4153a..59f08c44abdb 100644
--- a/drivers/mtd/nand/atmel_nand.c
+++ b/drivers/mtd/nand/atmel_nand.c
@@ -375,8 +375,7 @@ static int atmel_nand_dma_op(struct mtd_info *mtd, void *buf, int len,
 
 	dma_dev = host->dma_chan->device;
 
-	flags = DMA_CTRL_ACK | DMA_PREP_INTERRUPT | DMA_COMPL_SKIP_SRC_UNMAP |
-		DMA_COMPL_SKIP_DEST_UNMAP;
+	flags = DMA_CTRL_ACK | DMA_PREP_INTERRUPT;
 
 	phys_addr = dma_map_single(dma_dev->dev, p, len, dir);
 	if (dma_mapping_error(dma_dev->dev, phys_addr)) {
diff --git a/drivers/mtd/nand/docg4.c b/drivers/mtd/nand/docg4.c
index bd1cb672034f..1b0265e85a06 100644
--- a/drivers/mtd/nand/docg4.c
+++ b/drivers/mtd/nand/docg4.c
@@ -491,7 +491,7 @@ static uint8_t docg4_read_byte(struct mtd_info *mtd)
 		return status;
 	}
 
-	dev_warn(doc->dev, "unexpectd call to read_byte()\n");
+	dev_warn(doc->dev, "unexpected call to read_byte()\n");
 
 	return 0;
 }
diff --git a/drivers/mtd/nand/fsmc_nand.c b/drivers/mtd/nand/fsmc_nand.c
index 3dc1a7564d87..8b2752263db9 100644
--- a/drivers/mtd/nand/fsmc_nand.c
+++ b/drivers/mtd/nand/fsmc_nand.c
@@ -573,8 +573,6 @@ static int dma_xfer(struct fsmc_nand_data *host, void *buffer, int len,
 	dma_dev = chan->device;
 	dma_addr = dma_map_single(dma_dev->dev, buffer, len, direction);
 
-	flags |= DMA_COMPL_SKIP_SRC_UNMAP | DMA_COMPL_SKIP_DEST_UNMAP;
-
 	if (direction == DMA_TO_DEVICE) {
 		dma_src = dma_addr;
 		dma_dst = host->data_pa;
diff --git a/drivers/mtd/nand/mxc_nand.c b/drivers/mtd/nand/mxc_nand.c
index 4edea7f4462f..9dfdb06c508b 100644
--- a/drivers/mtd/nand/mxc_nand.c
+++ b/drivers/mtd/nand/mxc_nand.c
@@ -396,7 +396,7 @@ static void wait_op_done(struct mxc_nand_host *host, int useirq)
 
 	if (useirq) {
 		if (!host->devtype_data->check_int(host)) {
-			INIT_COMPLETION(host->op_completion);
+			reinit_completion(&host->op_completion);
 			irq_control(host, 1);
 			wait_for_completion(&host->op_completion);
 		}
diff --git a/drivers/mtd/nand/r852.c b/drivers/mtd/nand/r852.c
index 9dcf02d22aa8..325930db3f04 100644
--- a/drivers/mtd/nand/r852.c
+++ b/drivers/mtd/nand/r852.c
@@ -181,7 +181,7 @@ static void r852_do_dma(struct r852_device *dev, uint8_t *buf, int do_read)
 	/* Set dma direction */
 	dev->dma_dir = do_read;
 	dev->dma_stage = 1;
-	INIT_COMPLETION(dev->dma_done);
+	reinit_completion(&dev->dma_done);
 
 	dbg_verbose("doing dma %s ", do_read ? "read" : "write");
 
diff --git a/drivers/mtd/onenand/omap2.c b/drivers/mtd/onenand/omap2.c
index 2362909d20c0..6547c84afc3a 100644
--- a/drivers/mtd/onenand/omap2.c
+++ b/drivers/mtd/onenand/omap2.c
@@ -159,7 +159,7 @@ static int omap2_onenand_wait(struct mtd_info *mtd, int state)
 				syscfg = read_reg(c, ONENAND_REG_SYS_CFG1);
 		}
 
-		INIT_COMPLETION(c->irq_done);
+		reinit_completion(&c->irq_done);
 		if (c->gpio_irq) {
 			result = gpio_get_value(c->gpio_irq);
 			if (result == -1) {
@@ -349,7 +349,7 @@ static int omap3_onenand_read_bufferram(struct mtd_info *mtd, int area,
 	omap_set_dma_dest_params(c->dma_channel, 0, OMAP_DMA_AMODE_POST_INC,
 				 dma_dst, 0, 0);
 
-	INIT_COMPLETION(c->dma_done);
+	reinit_completion(&c->dma_done);
 	omap_start_dma(c->dma_channel);
 
 	timeout = jiffies + msecs_to_jiffies(20);
@@ -420,7 +420,7 @@ static int omap3_onenand_write_bufferram(struct mtd_info *mtd, int area,
 	omap_set_dma_dest_params(c->dma_channel, 0, OMAP_DMA_AMODE_POST_INC,
 				 dma_dst, 0, 0);
 
-	INIT_COMPLETION(c->dma_done);
+	reinit_completion(&c->dma_done);
 	omap_start_dma(c->dma_channel);
 
 	timeout = jiffies + msecs_to_jiffies(20);
@@ -499,7 +499,7 @@ static int omap2_onenand_read_bufferram(struct mtd_info *mtd, int area,
 	omap_set_dma_dest_params(c->dma_channel, 0, OMAP_DMA_AMODE_POST_INC,
 				 dma_dst, 0, 0);
 
-	INIT_COMPLETION(c->dma_done);
+	reinit_completion(&c->dma_done);
 	omap_start_dma(c->dma_channel);
 	wait_for_completion(&c->dma_done);
 
@@ -544,7 +544,7 @@ static int omap2_onenand_write_bufferram(struct mtd_info *mtd, int area,
 	omap_set_dma_dest_params(c->dma_channel, 0, OMAP_DMA_AMODE_POST_INC,
 				 dma_dst, 0, 0);
 
-	INIT_COMPLETION(c->dma_done);
+	reinit_completion(&c->dma_done);
 	omap_start_dma(c->dma_channel);
 	wait_for_completion(&c->dma_done);
 
diff --git a/drivers/net/bonding/bond_sysfs.c b/drivers/net/bonding/bond_sysfs.c
index bc8fd362a5aa..0ec2a7e8c8a9 100644
--- a/drivers/net/bonding/bond_sysfs.c
+++ b/drivers/net/bonding/bond_sysfs.c
@@ -524,8 +524,9 @@ static ssize_t bonding_store_arp_interval(struct device *d,
 		goto out;
 	}
 	if (bond->params.mode == BOND_MODE_ALB ||
-	    bond->params.mode == BOND_MODE_TLB) {
-		pr_info("%s: ARP monitoring cannot be used with ALB/TLB. Only MII monitoring is supported on %s.\n",
+	    bond->params.mode == BOND_MODE_TLB ||
+	    bond->params.mode == BOND_MODE_8023AD) {
+		pr_info("%s: ARP monitoring cannot be used with ALB/TLB/802.3ad. Only MII monitoring is supported on %s.\n",
 			bond->dev->name, bond->dev->name);
 		ret = -EINVAL;
 		goto out;
@@ -603,15 +604,14 @@ static ssize_t bonding_store_arp_targets(struct device *d,
 		return restart_syscall();
 
 	targets = bond->params.arp_targets;
-	newtarget = in_aton(buf + 1);
+	if (!in4_pton(buf + 1, -1, (u8 *)&newtarget, -1, NULL) ||
+	    IS_IP_TARGET_UNUSABLE_ADDRESS(newtarget)) {
+		pr_err("%s: invalid ARP target %pI4 specified for addition\n",
+		       bond->dev->name, &newtarget);
+		goto out;
+	}
 	/* look for adds */
 	if (buf[0] == '+') {
-		if ((newtarget == 0) || (newtarget == htonl(INADDR_BROADCAST))) {
-			pr_err("%s: invalid ARP target %pI4 specified for addition\n",
-			       bond->dev->name, &newtarget);
-			goto out;
-		}
-
 		if (bond_get_targets_ip(targets, newtarget) != -1) { /* dup */
 			pr_err("%s: ARP target %pI4 is already present\n",
 			       bond->dev->name, &newtarget);
@@ -634,12 +634,6 @@ static ssize_t bonding_store_arp_targets(struct device *d,
 		targets[ind] = newtarget;
 		write_unlock_bh(&bond->lock);
 	} else if (buf[0] == '-')	{
-		if ((newtarget == 0) || (newtarget == htonl(INADDR_BROADCAST))) {
-			pr_err("%s: invalid ARP target %pI4 specified for removal\n",
-			       bond->dev->name, &newtarget);
-			goto out;
-		}
-
 		ind = bond_get_targets_ip(targets, newtarget);
 		if (ind == -1) {
 			pr_err("%s: unable to remove nonexistent ARP target %pI4.\n",
@@ -701,6 +695,8 @@ static ssize_t bonding_store_downdelay(struct device *d,
 	int new_value, ret = count;
 	struct bonding *bond = to_bond(d);
 
+	if (!rtnl_trylock())
+		return restart_syscall();
 	if (!(bond->params.miimon)) {
 		pr_err("%s: Unable to set down delay as MII monitoring is disabled\n",
 		       bond->dev->name);
@@ -734,6 +730,7 @@ static ssize_t bonding_store_downdelay(struct device *d,
 	}
 
 out:
+	rtnl_unlock();
 	return ret;
 }
 static DEVICE_ATTR(downdelay, S_IRUGO | S_IWUSR,
@@ -756,6 +753,8 @@ static ssize_t bonding_store_updelay(struct device *d,
 	int new_value, ret = count;
 	struct bonding *bond = to_bond(d);
 
+	if (!rtnl_trylock())
+		return restart_syscall();
 	if (!(bond->params.miimon)) {
 		pr_err("%s: Unable to set up delay as MII monitoring is disabled\n",
 		       bond->dev->name);
@@ -789,6 +788,7 @@ static ssize_t bonding_store_updelay(struct device *d,
 	}
 
 out:
+	rtnl_unlock();
 	return ret;
 }
 static DEVICE_ATTR(updelay, S_IRUGO | S_IWUSR,
diff --git a/drivers/net/bonding/bonding.h b/drivers/net/bonding/bonding.h
index 77a07a12e77f..ca31286aa028 100644
--- a/drivers/net/bonding/bonding.h
+++ b/drivers/net/bonding/bonding.h
@@ -63,6 +63,9 @@
 		(((mode) == BOND_MODE_TLB) ||	\
 		 ((mode) == BOND_MODE_ALB))
 
+#define IS_IP_TARGET_UNUSABLE_ADDRESS(a)	\
+	((htonl(INADDR_BROADCAST) == a) ||	\
+	 ipv4_is_zeronet(a))
 /*
  * Less bad way to call ioctl from within the kernel; this needs to be
  * done some other way to get the call out of interrupt context.
diff --git a/drivers/net/caif/caif_virtio.c b/drivers/net/caif/caif_virtio.c
index b9ed1288ce2d..985608634f8c 100644
--- a/drivers/net/caif/caif_virtio.c
+++ b/drivers/net/caif/caif_virtio.c
@@ -686,18 +686,19 @@ static int cfv_probe(struct virtio_device *vdev)
 		goto err;
 
 	/* Get the CAIF configuration from virtio config space, if available */
-#define GET_VIRTIO_CONFIG_OPS(_v, _var, _f) \
-	((_v)->config->get(_v, offsetof(struct virtio_caif_transf_config, _f), \
-			   &_var, \
-			   FIELD_SIZEOF(struct virtio_caif_transf_config, _f)))
-
 	if (vdev->config->get) {
-		GET_VIRTIO_CONFIG_OPS(vdev, cfv->tx_hr, headroom);
-		GET_VIRTIO_CONFIG_OPS(vdev, cfv->rx_hr, headroom);
-		GET_VIRTIO_CONFIG_OPS(vdev, cfv->tx_tr, tailroom);
-		GET_VIRTIO_CONFIG_OPS(vdev, cfv->rx_tr, tailroom);
-		GET_VIRTIO_CONFIG_OPS(vdev, cfv->mtu, mtu);
-		GET_VIRTIO_CONFIG_OPS(vdev, cfv->mru, mtu);
+		virtio_cread(vdev, struct virtio_caif_transf_config, headroom,
+			     &cfv->tx_hr);
+		virtio_cread(vdev, struct virtio_caif_transf_config, headroom,
+			     &cfv->rx_hr);
+		virtio_cread(vdev, struct virtio_caif_transf_config, tailroom,
+			     &cfv->tx_tr);
+		virtio_cread(vdev, struct virtio_caif_transf_config, tailroom,
+			     &cfv->rx_tr);
+		virtio_cread(vdev, struct virtio_caif_transf_config, mtu,
+			     &cfv->mtu);
+		virtio_cread(vdev, struct virtio_caif_transf_config, mtu,
+			     &cfv->mru);
 	} else {
 		cfv->tx_hr = CFV_DEF_HEADROOM;
 		cfv->rx_hr = CFV_DEF_HEADROOM;
diff --git a/drivers/net/dummy.c b/drivers/net/dummy.c
index b710c6b2d659..bd8f84b0b894 100644
--- a/drivers/net/dummy.c
+++ b/drivers/net/dummy.c
@@ -88,10 +88,16 @@ static netdev_tx_t dummy_xmit(struct sk_buff *skb, struct net_device *dev)
 
 static int dummy_dev_init(struct net_device *dev)
 {
+	int i;
 	dev->dstats = alloc_percpu(struct pcpu_dstats);
 	if (!dev->dstats)
 		return -ENOMEM;
 
+	for_each_possible_cpu(i) {
+		struct pcpu_dstats *dstats;
+		dstats = per_cpu_ptr(dev->dstats, i);
+		u64_stats_init(&dstats->syncp);
+	}
 	return 0;
 }
 
diff --git a/drivers/net/ethernet/atheros/alx/main.c b/drivers/net/ethernet/atheros/alx/main.c
index 5aa5e8146496..c3c4c266b846 100644
--- a/drivers/net/ethernet/atheros/alx/main.c
+++ b/drivers/net/ethernet/atheros/alx/main.c
@@ -1388,6 +1388,9 @@ static int alx_resume(struct device *dev)
 {
 	struct pci_dev *pdev = to_pci_dev(dev);
 	struct alx_priv *alx = pci_get_drvdata(pdev);
+	struct alx_hw *hw = &alx->hw;
+
+	alx_reset_phy(hw);
 
 	if (!netif_running(alx->dev))
 		return 0;
diff --git a/drivers/net/ethernet/broadcom/bnx2x/bnx2x.h b/drivers/net/ethernet/broadcom/bnx2x/bnx2x.h
index 4e01c57d8c8d..a1f66e2c9a86 100644
--- a/drivers/net/ethernet/broadcom/bnx2x/bnx2x.h
+++ b/drivers/net/ethernet/broadcom/bnx2x/bnx2x.h
@@ -1376,7 +1376,6 @@ enum {
 	BNX2X_SP_RTNL_RX_MODE,
 	BNX2X_SP_RTNL_HYPERVISOR_VLAN,
 	BNX2X_SP_RTNL_TX_STOP,
-	BNX2X_SP_RTNL_TX_RESUME,
 };
 
 struct bnx2x_prev_path_list {
diff --git a/drivers/net/ethernet/broadcom/bnx2x/bnx2x_cmn.c b/drivers/net/ethernet/broadcom/bnx2x/bnx2x_cmn.c
index dcafbda3e5be..ec96130533cc 100644
--- a/drivers/net/ethernet/broadcom/bnx2x/bnx2x_cmn.c
+++ b/drivers/net/ethernet/broadcom/bnx2x/bnx2x_cmn.c
@@ -2959,6 +2959,10 @@ int bnx2x_nic_unload(struct bnx2x *bp, int unload_mode, bool keep_link)
 
 	bp->port.pmf = 0;
 
+	/* clear pending work in rtnl task */
+	bp->sp_rtnl_state = 0;
+	smp_mb();
+
 	/* Free SKBs, SGEs, TPA pool and driver internals */
 	bnx2x_free_skbs(bp);
 	if (CNIC_LOADED(bp))
diff --git a/drivers/net/ethernet/broadcom/bnx2x/bnx2x_dcb.c b/drivers/net/ethernet/broadcom/bnx2x/bnx2x_dcb.c
index fcf2761d8828..fdace204b054 100644
--- a/drivers/net/ethernet/broadcom/bnx2x/bnx2x_dcb.c
+++ b/drivers/net/ethernet/broadcom/bnx2x/bnx2x_dcb.c
@@ -778,11 +778,6 @@ void bnx2x_dcbx_set_params(struct bnx2x *bp, u32 state)
 
 		/* ets may affect cmng configuration: reinit it in hw */
 		bnx2x_set_local_cmng(bp);
-
-		set_bit(BNX2X_SP_RTNL_TX_RESUME, &bp->sp_rtnl_state);
-
-		schedule_delayed_work(&bp->sp_rtnl_task, 0);
-
 		return;
 	case BNX2X_DCBX_STATE_TX_RELEASED:
 		DP(BNX2X_MSG_DCB, "BNX2X_DCBX_STATE_TX_RELEASED\n");
diff --git a/drivers/net/ethernet/broadcom/bnx2x/bnx2x_main.c b/drivers/net/ethernet/broadcom/bnx2x/bnx2x_main.c
index e622cc1f96ff..814d0eca9b33 100644
--- a/drivers/net/ethernet/broadcom/bnx2x/bnx2x_main.c
+++ b/drivers/net/ethernet/broadcom/bnx2x/bnx2x_main.c
@@ -577,7 +577,9 @@ void bnx2x_write_dmae(struct bnx2x *bp, dma_addr_t dma_addr, u32 dst_addr,
 	rc = bnx2x_issue_dmae_with_comp(bp, &dmae, bnx2x_sp(bp, wb_comp));
 	if (rc) {
 		BNX2X_ERR("DMAE returned failure %d\n", rc);
+#ifdef BNX2X_STOP_ON_ERROR
 		bnx2x_panic();
+#endif
 	}
 }
 
@@ -614,7 +616,9 @@ void bnx2x_read_dmae(struct bnx2x *bp, u32 src_addr, u32 len32)
 	rc = bnx2x_issue_dmae_with_comp(bp, &dmae, bnx2x_sp(bp, wb_comp));
 	if (rc) {
 		BNX2X_ERR("DMAE returned failure %d\n", rc);
+#ifdef BNX2X_STOP_ON_ERROR
 		bnx2x_panic();
+#endif
 	}
 }
 
@@ -5231,18 +5235,18 @@ static void bnx2x_eq_int(struct bnx2x *bp)
 
 		case EVENT_RING_OPCODE_STOP_TRAFFIC:
 			DP(BNX2X_MSG_SP | BNX2X_MSG_DCB, "got STOP TRAFFIC\n");
+			bnx2x_dcbx_set_params(bp, BNX2X_DCBX_STATE_TX_PAUSED);
 			if (f_obj->complete_cmd(bp, f_obj,
 						BNX2X_F_CMD_TX_STOP))
 				break;
-			bnx2x_dcbx_set_params(bp, BNX2X_DCBX_STATE_TX_PAUSED);
 			goto next_spqe;
 
 		case EVENT_RING_OPCODE_START_TRAFFIC:
 			DP(BNX2X_MSG_SP | BNX2X_MSG_DCB, "got START TRAFFIC\n");
+			bnx2x_dcbx_set_params(bp, BNX2X_DCBX_STATE_TX_RELEASED);
 			if (f_obj->complete_cmd(bp, f_obj,
 						BNX2X_F_CMD_TX_START))
 				break;
-			bnx2x_dcbx_set_params(bp, BNX2X_DCBX_STATE_TX_RELEASED);
 			goto next_spqe;
 
 		case EVENT_RING_OPCODE_FUNCTION_UPDATE:
@@ -9352,6 +9356,10 @@ static int bnx2x_process_kill(struct bnx2x *bp, bool global)
 	bnx2x_process_kill_chip_reset(bp, global);
 	barrier();
 
+	/* clear errors in PGB */
+	if (!CHIP_IS_E1x(bp))
+		REG_WR(bp, PGLUE_B_REG_LATCHED_ERRORS_CLR, 0x7f);
+
 	/* Recover after reset: */
 	/* MCP */
 	if (global && bnx2x_reset_mcp_comp(bp, val))
@@ -9706,11 +9714,10 @@ sp_rtnl_not_reset:
 			       &bp->sp_rtnl_state))
 		bnx2x_pf_set_vfs_vlan(bp);
 
-	if (test_and_clear_bit(BNX2X_SP_RTNL_TX_STOP, &bp->sp_rtnl_state))
+	if (test_and_clear_bit(BNX2X_SP_RTNL_TX_STOP, &bp->sp_rtnl_state)) {
 		bnx2x_dcbx_stop_hw_tx(bp);
-
-	if (test_and_clear_bit(BNX2X_SP_RTNL_TX_RESUME, &bp->sp_rtnl_state))
 		bnx2x_dcbx_resume_hw_tx(bp);
+	}
 
 	/* work which needs rtnl lock not-taken (as it takes the lock itself and
 	 * can be called from other contexts as well)
diff --git a/drivers/net/ethernet/broadcom/bnx2x/bnx2x_reg.h b/drivers/net/ethernet/broadcom/bnx2x/bnx2x_reg.h
index 5ecf267dc4cc..3efbb35267c8 100644
--- a/drivers/net/ethernet/broadcom/bnx2x/bnx2x_reg.h
+++ b/drivers/net/ethernet/broadcom/bnx2x/bnx2x_reg.h
@@ -2864,6 +2864,17 @@
 #define PGLUE_B_REG_INTERNAL_PFID_ENABLE_TARGET_READ		 0x9430
 #define PGLUE_B_REG_INTERNAL_PFID_ENABLE_TARGET_WRITE		 0x9434
 #define PGLUE_B_REG_INTERNAL_VFID_ENABLE			 0x9438
+/* [W 7] Writing 1 to each bit in this register clears a corresponding error
+ * details register and enables logging new error details. Bit 0 - clears
+ * INCORRECT_RCV_DETAILS; Bit 1 - clears RX_ERR_DETAILS; Bit 2 - clears
+ * TX_ERR_WR_ADD_31_0 TX_ERR_WR_ADD_63_32 TX_ERR_WR_DETAILS
+ * TX_ERR_WR_DETAILS2 TX_ERR_RD_ADD_31_0 TX_ERR_RD_ADD_63_32
+ * TX_ERR_RD_DETAILS TX_ERR_RD_DETAILS2 TX_ERR_WR_DETAILS_ICPL; Bit 3 -
+ * clears VF_LENGTH_VIOLATION_DETAILS. Bit 4 - clears
+ * VF_GRC_SPACE_VIOLATION_DETAILS. Bit 5 - clears RX_TCPL_ERR_DETAILS. Bit 6
+ * - clears TCPL_IN_TWO_RCBS_DETAILS. */
+#define PGLUE_B_REG_LATCHED_ERRORS_CLR				 0x943c
+
 /* [R 9] Interrupt register #0 read */
 #define PGLUE_B_REG_PGLUE_B_INT_STS				 0x9298
 /* [RC 9] Interrupt register #0 read clear */
diff --git a/drivers/net/ethernet/broadcom/bnx2x/bnx2x_vfpf.c b/drivers/net/ethernet/broadcom/bnx2x/bnx2x_vfpf.c
index 9199adf32d33..efa8a151d789 100644
--- a/drivers/net/ethernet/broadcom/bnx2x/bnx2x_vfpf.c
+++ b/drivers/net/ethernet/broadcom/bnx2x/bnx2x_vfpf.c
@@ -152,7 +152,7 @@ static int bnx2x_send_msg2pf(struct bnx2x *bp, u8 *done, dma_addr_t msg_mapping)
 	if (bp->old_bulletin.valid_bitmap & 1 << CHANNEL_DOWN) {
 		DP(BNX2X_MSG_IOV, "detecting channel down. Aborting message\n");
 		*done = PFVF_STATUS_SUCCESS;
-		return 0;
+		return -EINVAL;
 	}
 
 	/* Write message address */
diff --git a/drivers/net/ethernet/broadcom/tg3.c b/drivers/net/ethernet/broadcom/tg3.c
index 00c5be8c55b8..a9e068423ba0 100644
--- a/drivers/net/ethernet/broadcom/tg3.c
+++ b/drivers/net/ethernet/broadcom/tg3.c
@@ -13618,16 +13618,9 @@ static int tg3_hwtstamp_ioctl(struct net_device *dev,
 	if (stmpconf.flags)
 		return -EINVAL;
 
-	switch (stmpconf.tx_type) {
-	case HWTSTAMP_TX_ON:
-		tg3_flag_set(tp, TX_TSTAMP_EN);
-		break;
-	case HWTSTAMP_TX_OFF:
-		tg3_flag_clear(tp, TX_TSTAMP_EN);
-		break;
-	default:
+	if (stmpconf.tx_type != HWTSTAMP_TX_ON &&
+	    stmpconf.tx_type != HWTSTAMP_TX_OFF)
 		return -ERANGE;
-	}
 
 	switch (stmpconf.rx_filter) {
 	case HWTSTAMP_FILTER_NONE:
@@ -13689,6 +13682,11 @@ static int tg3_hwtstamp_ioctl(struct net_device *dev,
 		tw32(TG3_RX_PTP_CTL,
 		     tp->rxptpctl | TG3_RX_PTP_CTL_HWTS_INTERLOCK);
 
+	if (stmpconf.tx_type == HWTSTAMP_TX_ON)
+		tg3_flag_set(tp, TX_TSTAMP_EN);
+	else
+		tg3_flag_clear(tp, TX_TSTAMP_EN);
+
 	return copy_to_user(ifr->ifr_data, &stmpconf, sizeof(stmpconf)) ?
 		-EFAULT : 0;
 }
diff --git a/drivers/net/ethernet/emulex/benet/be_cmds.c b/drivers/net/ethernet/emulex/benet/be_cmds.c
index 7fb0edfe3d24..dbcd5262c016 100644
--- a/drivers/net/ethernet/emulex/benet/be_cmds.c
+++ b/drivers/net/ethernet/emulex/benet/be_cmds.c
@@ -1758,7 +1758,7 @@ err:
 
 /* Uses sycnhronous mcc */
 int be_cmd_vlan_config(struct be_adapter *adapter, u32 if_id, u16 *vtag_array,
-			u32 num, bool untagged, bool promiscuous)
+		       u32 num, bool promiscuous)
 {
 	struct be_mcc_wrb *wrb;
 	struct be_cmd_req_vlan_config *req;
@@ -1778,7 +1778,7 @@ int be_cmd_vlan_config(struct be_adapter *adapter, u32 if_id, u16 *vtag_array,
 
 	req->interface_id = if_id;
 	req->promiscuous = promiscuous;
-	req->untagged = untagged;
+	req->untagged = BE_IF_FLAGS_UNTAGGED & be_if_cap_flags(adapter) ? 1 : 0;
 	req->num_vlan = num;
 	if (!promiscuous) {
 		memcpy(req->normal_vlan, vtag_array,
@@ -1847,7 +1847,19 @@ int be_cmd_rx_filter(struct be_adapter *adapter, u32 flags, u32 value)
 			memcpy(req->mcast_mac[i++].byte, ha->addr, ETH_ALEN);
 	}
 
+	if ((req->if_flags_mask & cpu_to_le32(be_if_cap_flags(adapter))) !=
+	     req->if_flags_mask) {
+		dev_warn(&adapter->pdev->dev,
+			 "Cannot set rx filter flags 0x%x\n",
+			 req->if_flags_mask);
+		dev_warn(&adapter->pdev->dev,
+			 "Interface is capable of 0x%x flags only\n",
+			 be_if_cap_flags(adapter));
+	}
+	req->if_flags_mask &= cpu_to_le32(be_if_cap_flags(adapter));
+
 	status = be_mcc_notify_wait(adapter);
+
 err:
 	spin_unlock_bh(&adapter->mcc_lock);
 	return status;
diff --git a/drivers/net/ethernet/emulex/benet/be_cmds.h b/drivers/net/ethernet/emulex/benet/be_cmds.h
index edf3e8a0ff83..0075686276aa 100644
--- a/drivers/net/ethernet/emulex/benet/be_cmds.h
+++ b/drivers/net/ethernet/emulex/benet/be_cmds.h
@@ -1984,7 +1984,7 @@ int be_cmd_get_fw_ver(struct be_adapter *adapter, char *fw_ver,
 		      char *fw_on_flash);
 int be_cmd_modify_eqd(struct be_adapter *adapter, struct be_set_eqd *, int num);
 int be_cmd_vlan_config(struct be_adapter *adapter, u32 if_id, u16 *vtag_array,
-		       u32 num, bool untagged, bool promiscuous);
+		       u32 num, bool promiscuous);
 int be_cmd_rx_filter(struct be_adapter *adapter, u32 flags, u32 status);
 int be_cmd_set_flow_control(struct be_adapter *adapter, u32 tx_fc, u32 rx_fc);
 int be_cmd_get_flow_control(struct be_adapter *adapter, u32 *tx_fc, u32 *rx_fc);
diff --git a/drivers/net/ethernet/emulex/benet/be_main.c b/drivers/net/ethernet/emulex/benet/be_main.c
index cb2bb6fccbc8..abde97471636 100644
--- a/drivers/net/ethernet/emulex/benet/be_main.c
+++ b/drivers/net/ethernet/emulex/benet/be_main.c
@@ -1079,7 +1079,7 @@ static int be_vid_config(struct be_adapter *adapter)
 			vids[num++] = cpu_to_le16(i);
 
 	status = be_cmd_vlan_config(adapter, adapter->if_handle,
-				    vids, num, 1, 0);
+				    vids, num, 0);
 
 	if (status) {
 		/* Set to VLAN promisc mode as setting VLAN filter failed */
@@ -2148,6 +2148,9 @@ static int be_tx_qs_create(struct be_adapter *adapter)
 		if (status)
 			return status;
 
+		u64_stats_init(&txo->stats.sync);
+		u64_stats_init(&txo->stats.sync_compl);
+
 		/* If num_evt_qs is less than num_tx_qs, then more than
 		 * one txq share an eq
 		 */
@@ -2209,6 +2212,7 @@ static int be_rx_cqs_create(struct be_adapter *adapter)
 		if (rc)
 			return rc;
 
+		u64_stats_init(&rxo->stats.sync);
 		eq = &adapter->eq_obj[i % adapter->num_evt_qs].q;
 		rc = be_cmd_cq_create(adapter, cq, eq, false, 3);
 		if (rc)
@@ -2672,6 +2676,11 @@ static int be_close(struct net_device *netdev)
 
 	be_rx_qs_destroy(adapter);
 
+	for (i = 1; i < (adapter->uc_macs + 1); i++)
+		be_cmd_pmac_del(adapter, adapter->if_handle,
+				adapter->pmac_id[i], 0);
+	adapter->uc_macs = 0;
+
 	for_all_evt_queues(adapter, eqo, i) {
 		if (msix_enabled(adapter))
 			synchronize_irq(be_msix_vec_get(adapter, eqo));
diff --git a/drivers/net/ethernet/freescale/fec_main.c b/drivers/net/ethernet/freescale/fec_main.c
index b2793b91cc55..4cbebf3d80eb 100644
--- a/drivers/net/ethernet/freescale/fec_main.c
+++ b/drivers/net/ethernet/freescale/fec_main.c
@@ -386,7 +386,14 @@ fec_enet_start_xmit(struct sk_buff *skb, struct net_device *ndev)
 	 */
 	bdp->cbd_bufaddr = dma_map_single(&fep->pdev->dev, bufaddr,
 			FEC_ENET_TX_FRSIZE, DMA_TO_DEVICE);
-
+	if (dma_mapping_error(&fep->pdev->dev, bdp->cbd_bufaddr)) {
+		bdp->cbd_bufaddr = 0;
+		fep->tx_skbuff[index] = NULL;
+		dev_kfree_skb_any(skb);
+		if (net_ratelimit())
+			netdev_err(ndev, "Tx DMA memory map failed\n");
+		return NETDEV_TX_OK;
+	}
 	/* Send it on its way.  Tell FEC it's ready, interrupt when done,
 	 * it's the last BD of the frame, and to put the CRC on the end.
 	 */
@@ -861,6 +868,7 @@ fec_enet_rx(struct net_device *ndev, int budget)
 	struct	bufdesc_ex *ebdp = NULL;
 	bool	vlan_packet_rcvd = false;
 	u16	vlan_tag;
+	int	index = 0;
 
 #ifdef CONFIG_M532x
 	flush_cache_all();
@@ -916,10 +924,15 @@ fec_enet_rx(struct net_device *ndev, int budget)
 		ndev->stats.rx_packets++;
 		pkt_len = bdp->cbd_datlen;
 		ndev->stats.rx_bytes += pkt_len;
-		data = (__u8*)__va(bdp->cbd_bufaddr);
 
-		dma_unmap_single(&fep->pdev->dev, bdp->cbd_bufaddr,
-				FEC_ENET_TX_FRSIZE, DMA_FROM_DEVICE);
+		if (fep->bufdesc_ex)
+			index = (struct bufdesc_ex *)bdp -
+				(struct bufdesc_ex *)fep->rx_bd_base;
+		else
+			index = bdp - fep->rx_bd_base;
+		data = fep->rx_skbuff[index]->data;
+		dma_sync_single_for_cpu(&fep->pdev->dev, bdp->cbd_bufaddr,
+					FEC_ENET_RX_FRSIZE, DMA_FROM_DEVICE);
 
 		if (id_entry->driver_data & FEC_QUIRK_SWAP_FRAME)
 			swap_buffer(data, pkt_len);
@@ -999,8 +1012,8 @@ fec_enet_rx(struct net_device *ndev, int budget)
 			napi_gro_receive(&fep->napi, skb);
 		}
 
-		bdp->cbd_bufaddr = dma_map_single(&fep->pdev->dev, data,
-				FEC_ENET_TX_FRSIZE, DMA_FROM_DEVICE);
+		dma_sync_single_for_device(&fep->pdev->dev, bdp->cbd_bufaddr,
+					FEC_ENET_RX_FRSIZE, DMA_FROM_DEVICE);
 rx_processing_done:
 		/* Clear the status flags for this buffer */
 		status &= ~BD_ENET_RX_STATS;
@@ -1719,6 +1732,12 @@ static int fec_enet_alloc_buffers(struct net_device *ndev)
 
 		bdp->cbd_bufaddr = dma_map_single(&fep->pdev->dev, skb->data,
 				FEC_ENET_RX_FRSIZE, DMA_FROM_DEVICE);
+		if (dma_mapping_error(&fep->pdev->dev, bdp->cbd_bufaddr)) {
+			fec_enet_free_buffers(ndev);
+			if (net_ratelimit())
+				netdev_err(ndev, "Rx DMA memory map failed\n");
+			return -ENOMEM;
+		}
 		bdp->cbd_sc = BD_ENET_RX_EMPTY;
 
 		if (fep->bufdesc_ex) {
diff --git a/drivers/net/ethernet/intel/e1000e/netdev.c b/drivers/net/ethernet/intel/e1000e/netdev.c
index aedd5736a87d..8d3945ab7334 100644
--- a/drivers/net/ethernet/intel/e1000e/netdev.c
+++ b/drivers/net/ethernet/intel/e1000e/netdev.c
@@ -3482,10 +3482,10 @@ s32 e1000e_get_base_timinca(struct e1000_adapter *adapter, u32 *timinca)
  * specified. Matching the kind of event packet is not supported, with the
  * exception of "all V2 events regardless of level 2 or 4".
  **/
-static int e1000e_config_hwtstamp(struct e1000_adapter *adapter)
+static int e1000e_config_hwtstamp(struct e1000_adapter *adapter,
+				  struct hwtstamp_config *config)
 {
 	struct e1000_hw *hw = &adapter->hw;
-	struct hwtstamp_config *config = &adapter->hwtstamp_config;
 	u32 tsync_tx_ctl = E1000_TSYNCTXCTL_ENABLED;
 	u32 tsync_rx_ctl = E1000_TSYNCRXCTL_ENABLED;
 	u32 rxmtrl = 0;
@@ -3586,6 +3586,8 @@ static int e1000e_config_hwtstamp(struct e1000_adapter *adapter)
 		return -ERANGE;
 	}
 
+	adapter->hwtstamp_config = *config;
+
 	/* enable/disable Tx h/w time stamping */
 	regval = er32(TSYNCTXCTL);
 	regval &= ~E1000_TSYNCTXCTL_ENABLED;
@@ -3874,7 +3876,7 @@ void e1000e_reset(struct e1000_adapter *adapter)
 	e1000e_reset_adaptive(hw);
 
 	/* initialize systim and reset the ns time counter */
-	e1000e_config_hwtstamp(adapter);
+	e1000e_config_hwtstamp(adapter, &adapter->hwtstamp_config);
 
 	/* Set EEE advertisement as appropriate */
 	if (adapter->flags2 & FLAG2_HAS_EEE) {
@@ -5797,14 +5799,10 @@ static int e1000e_hwtstamp_ioctl(struct net_device *netdev, struct ifreq *ifr)
 	if (copy_from_user(&config, ifr->ifr_data, sizeof(config)))
 		return -EFAULT;
 
-	adapter->hwtstamp_config = config;
-
-	ret_val = e1000e_config_hwtstamp(adapter);
+	ret_val = e1000e_config_hwtstamp(adapter, &config);
 	if (ret_val)
 		return ret_val;
 
-	config = adapter->hwtstamp_config;
-
 	switch (config.rx_filter) {
 	case HWTSTAMP_FILTER_PTP_V2_L4_SYNC:
 	case HWTSTAMP_FILTER_PTP_V2_L2_SYNC:
diff --git a/drivers/net/ethernet/intel/igb/igb_main.c b/drivers/net/ethernet/intel/igb/igb_main.c
index 2ac14bdd5fbb..025e5f4b7481 100644
--- a/drivers/net/ethernet/intel/igb/igb_main.c
+++ b/drivers/net/ethernet/intel/igb/igb_main.c
@@ -1224,6 +1224,9 @@ static int igb_alloc_q_vector(struct igb_adapter *adapter,
 		ring->count = adapter->tx_ring_count;
 		ring->queue_index = txr_idx;
 
+		u64_stats_init(&ring->tx_syncp);
+		u64_stats_init(&ring->tx_syncp2);
+
 		/* assign ring to adapter */
 		adapter->tx_ring[txr_idx] = ring;
 
@@ -1257,6 +1260,8 @@ static int igb_alloc_q_vector(struct igb_adapter *adapter,
 		ring->count = adapter->rx_ring_count;
 		ring->queue_index = rxr_idx;
 
+		u64_stats_init(&ring->rx_syncp);
+
 		/* assign ring to adapter */
 		adapter->rx_ring[rxr_idx] = ring;
 	}
diff --git a/drivers/net/ethernet/intel/ixgbe/ixgbe_main.c b/drivers/net/ethernet/intel/ixgbe/ixgbe_main.c
index 0066f0aefbfa..0c55079ebee3 100644
--- a/drivers/net/ethernet/intel/ixgbe/ixgbe_main.c
+++ b/drivers/net/ethernet/intel/ixgbe/ixgbe_main.c
@@ -5085,6 +5085,8 @@ int ixgbe_setup_tx_resources(struct ixgbe_ring *tx_ring)
 	if (!tx_ring->tx_buffer_info)
 		goto err;
 
+	u64_stats_init(&tx_ring->syncp);
+
 	/* round up to nearest 4K */
 	tx_ring->size = tx_ring->count * sizeof(union ixgbe_adv_tx_desc);
 	tx_ring->size = ALIGN(tx_ring->size, 4096);
@@ -5167,6 +5169,8 @@ int ixgbe_setup_rx_resources(struct ixgbe_ring *rx_ring)
 	if (!rx_ring->rx_buffer_info)
 		goto err;
 
+	u64_stats_init(&rx_ring->syncp);
+
 	/* Round up to nearest 4K */
 	rx_ring->size = rx_ring->count * sizeof(union ixgbe_adv_rx_desc);
 	rx_ring->size = ALIGN(rx_ring->size, 4096);
diff --git a/drivers/net/ethernet/marvell/mv643xx_eth.c b/drivers/net/ethernet/marvell/mv643xx_eth.c
index 00cd36e08601..61088a6a9424 100644
--- a/drivers/net/ethernet/marvell/mv643xx_eth.c
+++ b/drivers/net/ethernet/marvell/mv643xx_eth.c
@@ -2890,7 +2890,8 @@ static int mv643xx_eth_probe(struct platform_device *pdev)
 					 PHY_INTERFACE_MODE_GMII);
 		if (!mp->phy)
 			err = -ENODEV;
-		phy_addr_set(mp, mp->phy->addr);
+		else
+			phy_addr_set(mp, mp->phy->addr);
 	} else if (pd->phy_addr != MV643XX_ETH_PHY_NONE) {
 		mp->phy = phy_scan(mp, pd->phy_addr);
 
diff --git a/drivers/net/ethernet/marvell/mvneta.c b/drivers/net/ethernet/marvell/mvneta.c
index 7d99e695a110..b8e232b4ea2d 100644
--- a/drivers/net/ethernet/marvell/mvneta.c
+++ b/drivers/net/ethernet/marvell/mvneta.c
@@ -2792,6 +2792,9 @@ static int mvneta_probe(struct platform_device *pdev)
 
 	pp = netdev_priv(dev);
 
+	u64_stats_init(&pp->tx_stats.syncp);
+	u64_stats_init(&pp->rx_stats.syncp);
+
 	pp->weight = MVNETA_RX_POLL_WEIGHT;
 	pp->phy_node = phy_node;
 	pp->phy_interface = phy_mode;
diff --git a/drivers/net/ethernet/marvell/sky2.c b/drivers/net/ethernet/marvell/sky2.c
index a7df981d2123..43aa7acd84a6 100644
--- a/drivers/net/ethernet/marvell/sky2.c
+++ b/drivers/net/ethernet/marvell/sky2.c
@@ -4763,6 +4763,9 @@ static struct net_device *sky2_init_netdev(struct sky2_hw *hw, unsigned port,
 	sky2->hw = hw;
 	sky2->msg_enable = netif_msg_init(debug, default_msg);
 
+	u64_stats_init(&sky2->tx_stats.syncp);
+	u64_stats_init(&sky2->rx_stats.syncp);
+
 	/* Auto speed and flow control */
 	sky2->flags = SKY2_FLAG_AUTO_SPEED | SKY2_FLAG_AUTO_PAUSE;
 	if (hw->chip_id != CHIP_ID_YUKON_XL)
diff --git a/drivers/net/ethernet/mellanox/mlx4/fw.c b/drivers/net/ethernet/mellanox/mlx4/fw.c
index fda26679f7d5..194928214606 100644
--- a/drivers/net/ethernet/mellanox/mlx4/fw.c
+++ b/drivers/net/ethernet/mellanox/mlx4/fw.c
@@ -1774,7 +1774,7 @@ void mlx4_opreq_action(struct work_struct *work)
 				   MLX4_CMD_GET_OP_REQ, MLX4_CMD_TIME_CLASS_A,
 				   MLX4_CMD_NATIVE);
 		if (err) {
-			mlx4_err(dev, "Failed to retreive required operation: %d\n",
+			mlx4_err(dev, "Failed to retrieve required operation: %d\n",
 				 err);
 			return;
 		}
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/cmd.c b/drivers/net/ethernet/mellanox/mlx5/core/cmd.c
index 6ca30739625f..8675d26a678b 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/cmd.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/cmd.c
@@ -98,6 +98,7 @@ enum {
 static struct mlx5_cmd_work_ent *alloc_cmd(struct mlx5_cmd *cmd,
 					   struct mlx5_cmd_msg *in,
 					   struct mlx5_cmd_msg *out,
+					   void *uout, int uout_size,
 					   mlx5_cmd_cbk_t cbk,
 					   void *context, int page_queue)
 {
@@ -110,6 +111,8 @@ static struct mlx5_cmd_work_ent *alloc_cmd(struct mlx5_cmd *cmd,
 
 	ent->in		= in;
 	ent->out	= out;
+	ent->uout	= uout;
+	ent->uout_size	= uout_size;
 	ent->callback	= cbk;
 	ent->context	= context;
 	ent->cmd	= cmd;
@@ -534,6 +537,7 @@ static void cmd_work_handler(struct work_struct *work)
 	ent->lay = lay;
 	memset(lay, 0, sizeof(*lay));
 	memcpy(lay->in, ent->in->first.data, sizeof(lay->in));
+	ent->op = be32_to_cpu(lay->in[0]) >> 16;
 	if (ent->in->next)
 		lay->in_ptr = cpu_to_be64(ent->in->next->dma);
 	lay->inlen = cpu_to_be32(ent->in->len);
@@ -628,7 +632,8 @@ static int wait_func(struct mlx5_core_dev *dev, struct mlx5_cmd_work_ent *ent)
  *    2. page queue commands do not support asynchrous completion
  */
 static int mlx5_cmd_invoke(struct mlx5_core_dev *dev, struct mlx5_cmd_msg *in,
-			   struct mlx5_cmd_msg *out, mlx5_cmd_cbk_t callback,
+			   struct mlx5_cmd_msg *out, void *uout, int uout_size,
+			   mlx5_cmd_cbk_t callback,
 			   void *context, int page_queue, u8 *status)
 {
 	struct mlx5_cmd *cmd = &dev->cmd;
@@ -642,7 +647,8 @@ static int mlx5_cmd_invoke(struct mlx5_core_dev *dev, struct mlx5_cmd_msg *in,
 	if (callback && page_queue)
 		return -EINVAL;
 
-	ent = alloc_cmd(cmd, in, out, callback, context, page_queue);
+	ent = alloc_cmd(cmd, in, out, uout, uout_size, callback, context,
+			page_queue);
 	if (IS_ERR(ent))
 		return PTR_ERR(ent);
 
@@ -670,10 +676,10 @@ static int mlx5_cmd_invoke(struct mlx5_core_dev *dev, struct mlx5_cmd_msg *in,
 		op = be16_to_cpu(((struct mlx5_inbox_hdr *)in->first.data)->opcode);
 		if (op < ARRAY_SIZE(cmd->stats)) {
 			stats = &cmd->stats[op];
-			spin_lock(&stats->lock);
+			spin_lock_irq(&stats->lock);
 			stats->sum += ds;
 			++stats->n;
-			spin_unlock(&stats->lock);
+			spin_unlock_irq(&stats->lock);
 		}
 		mlx5_core_dbg_mask(dev, 1 << MLX5_CMD_TIME,
 				   "fw exec time for %s is %lld nsec\n",
@@ -826,7 +832,7 @@ static struct mlx5_cmd_msg *mlx5_alloc_cmd_msg(struct mlx5_core_dev *dev,
 	int n;
 	int i;
 
-	msg = kzalloc(sizeof(*msg), GFP_KERNEL);
+	msg = kzalloc(sizeof(*msg), flags);
 	if (!msg)
 		return ERR_PTR(-ENOMEM);
 
@@ -1109,6 +1115,19 @@ void mlx5_cmd_use_polling(struct mlx5_core_dev *dev)
 		up(&cmd->sem);
 }
 
+static void free_msg(struct mlx5_core_dev *dev, struct mlx5_cmd_msg *msg)
+{
+	unsigned long flags;
+
+	if (msg->cache) {
+		spin_lock_irqsave(&msg->cache->lock, flags);
+		list_add_tail(&msg->list, &msg->cache->head);
+		spin_unlock_irqrestore(&msg->cache->lock, flags);
+	} else {
+		mlx5_free_cmd_msg(dev, msg);
+	}
+}
+
 void mlx5_cmd_comp_handler(struct mlx5_core_dev *dev, unsigned long vector)
 {
 	struct mlx5_cmd *cmd = &dev->cmd;
@@ -1117,6 +1136,10 @@ void mlx5_cmd_comp_handler(struct mlx5_core_dev *dev, unsigned long vector)
 	void *context;
 	int err;
 	int i;
+	ktime_t t1, t2, delta;
+	s64 ds;
+	struct mlx5_cmd_stats *stats;
+	unsigned long flags;
 
 	for (i = 0; i < (1 << cmd->log_sz); i++) {
 		if (test_bit(i, &vector)) {
@@ -1141,9 +1164,29 @@ void mlx5_cmd_comp_handler(struct mlx5_core_dev *dev, unsigned long vector)
 			}
 			free_ent(cmd, ent->idx);
 			if (ent->callback) {
+				t1 = timespec_to_ktime(ent->ts1);
+				t2 = timespec_to_ktime(ent->ts2);
+				delta = ktime_sub(t2, t1);
+				ds = ktime_to_ns(delta);
+				if (ent->op < ARRAY_SIZE(cmd->stats)) {
+					stats = &cmd->stats[ent->op];
+					spin_lock_irqsave(&stats->lock, flags);
+					stats->sum += ds;
+					++stats->n;
+					spin_unlock_irqrestore(&stats->lock, flags);
+				}
+
 				callback = ent->callback;
 				context = ent->context;
 				err = ent->ret;
+				if (!err)
+					err = mlx5_copy_from_msg(ent->uout,
+								 ent->out,
+								 ent->uout_size);
+
+				mlx5_free_cmd_msg(dev, ent->out);
+				free_msg(dev, ent->in);
+
 				free_cmd(ent);
 				callback(err, context);
 			} else {
@@ -1160,7 +1203,8 @@ static int status_to_err(u8 status)
 	return status ? -1 : 0; /* TBD more meaningful codes */
 }
 
-static struct mlx5_cmd_msg *alloc_msg(struct mlx5_core_dev *dev, int in_size)
+static struct mlx5_cmd_msg *alloc_msg(struct mlx5_core_dev *dev, int in_size,
+				      gfp_t gfp)
 {
 	struct mlx5_cmd_msg *msg = ERR_PTR(-ENOMEM);
 	struct mlx5_cmd *cmd = &dev->cmd;
@@ -1172,7 +1216,7 @@ static struct mlx5_cmd_msg *alloc_msg(struct mlx5_core_dev *dev, int in_size)
 		ent = &cmd->cache.med;
 
 	if (ent) {
-		spin_lock(&ent->lock);
+		spin_lock_irq(&ent->lock);
 		if (!list_empty(&ent->head)) {
 			msg = list_entry(ent->head.next, typeof(*msg), list);
 			/* For cached lists, we must explicitly state what is
@@ -1181,43 +1225,34 @@ static struct mlx5_cmd_msg *alloc_msg(struct mlx5_core_dev *dev, int in_size)
 			msg->len = in_size;
 			list_del(&msg->list);
 		}
-		spin_unlock(&ent->lock);
+		spin_unlock_irq(&ent->lock);
 	}
 
 	if (IS_ERR(msg))
-		msg = mlx5_alloc_cmd_msg(dev, GFP_KERNEL, in_size);
+		msg = mlx5_alloc_cmd_msg(dev, gfp, in_size);
 
 	return msg;
 }
 
-static void free_msg(struct mlx5_core_dev *dev, struct mlx5_cmd_msg *msg)
-{
-	if (msg->cache) {
-		spin_lock(&msg->cache->lock);
-		list_add_tail(&msg->list, &msg->cache->head);
-		spin_unlock(&msg->cache->lock);
-	} else {
-		mlx5_free_cmd_msg(dev, msg);
-	}
-}
-
 static int is_manage_pages(struct mlx5_inbox_hdr *in)
 {
 	return be16_to_cpu(in->opcode) == MLX5_CMD_OP_MANAGE_PAGES;
 }
 
-int mlx5_cmd_exec(struct mlx5_core_dev *dev, void *in, int in_size, void *out,
-		  int out_size)
+static int cmd_exec(struct mlx5_core_dev *dev, void *in, int in_size, void *out,
+		    int out_size, mlx5_cmd_cbk_t callback, void *context)
 {
 	struct mlx5_cmd_msg *inb;
 	struct mlx5_cmd_msg *outb;
 	int pages_queue;
+	gfp_t gfp;
 	int err;
 	u8 status = 0;
 
 	pages_queue = is_manage_pages(in);
+	gfp = callback ? GFP_ATOMIC : GFP_KERNEL;
 
-	inb = alloc_msg(dev, in_size);
+	inb = alloc_msg(dev, in_size, gfp);
 	if (IS_ERR(inb)) {
 		err = PTR_ERR(inb);
 		return err;
@@ -1229,13 +1264,14 @@ int mlx5_cmd_exec(struct mlx5_core_dev *dev, void *in, int in_size, void *out,
 		goto out_in;
 	}
 
-	outb = mlx5_alloc_cmd_msg(dev, GFP_KERNEL, out_size);
+	outb = mlx5_alloc_cmd_msg(dev, gfp, out_size);
 	if (IS_ERR(outb)) {
 		err = PTR_ERR(outb);
 		goto out_in;
 	}
 
-	err = mlx5_cmd_invoke(dev, inb, outb, NULL, NULL, pages_queue, &status);
+	err = mlx5_cmd_invoke(dev, inb, outb, out, out_size, callback, context,
+			      pages_queue, &status);
 	if (err)
 		goto out_out;
 
@@ -1248,14 +1284,30 @@ int mlx5_cmd_exec(struct mlx5_core_dev *dev, void *in, int in_size, void *out,
 	err = mlx5_copy_from_msg(out, outb, out_size);
 
 out_out:
-	mlx5_free_cmd_msg(dev, outb);
+	if (!callback)
+		mlx5_free_cmd_msg(dev, outb);
 
 out_in:
-	free_msg(dev, inb);
+	if (!callback)
+		free_msg(dev, inb);
 	return err;
 }
+
+int mlx5_cmd_exec(struct mlx5_core_dev *dev, void *in, int in_size, void *out,
+		  int out_size)
+{
+	return cmd_exec(dev, in, in_size, out, out_size, NULL, NULL);
+}
 EXPORT_SYMBOL(mlx5_cmd_exec);
 
+int mlx5_cmd_exec_cb(struct mlx5_core_dev *dev, void *in, int in_size,
+		     void *out, int out_size, mlx5_cmd_cbk_t callback,
+		     void *context)
+{
+	return cmd_exec(dev, in, in_size, out, out_size, callback, context);
+}
+EXPORT_SYMBOL(mlx5_cmd_exec_cb);
+
 static void destroy_msg_cache(struct mlx5_core_dev *dev)
 {
 	struct mlx5_cmd *cmd = &dev->cmd;
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/debugfs.c b/drivers/net/ethernet/mellanox/mlx5/core/debugfs.c
index 9c7194b26ee2..80f6d127257a 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/debugfs.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/debugfs.c
@@ -154,10 +154,10 @@ static ssize_t average_read(struct file *filp, char __user *buf, size_t count,
 		return 0;
 
 	stats = filp->private_data;
-	spin_lock(&stats->lock);
+	spin_lock_irq(&stats->lock);
 	if (stats->n)
 		field = div64_u64(stats->sum, stats->n);
-	spin_unlock(&stats->lock);
+	spin_unlock_irq(&stats->lock);
 	ret = snprintf(tbuf, sizeof(tbuf), "%llu\n", field);
 	if (ret > 0) {
 		if (copy_to_user(buf, tbuf, ret))
@@ -175,10 +175,10 @@ static ssize_t average_write(struct file *filp, const char __user *buf,
 	struct mlx5_cmd_stats *stats;
 
 	stats = filp->private_data;
-	spin_lock(&stats->lock);
+	spin_lock_irq(&stats->lock);
 	stats->sum = 0;
 	stats->n = 0;
-	spin_unlock(&stats->lock);
+	spin_unlock_irq(&stats->lock);
 
 	*pos += count;
 
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/eq.c b/drivers/net/ethernet/mellanox/mlx5/core/eq.c
index 2231d93cc7ad..64a61b286b2c 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/eq.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/eq.c
@@ -354,7 +354,7 @@ int mlx5_create_map_eq(struct mlx5_core_dev *dev, struct mlx5_eq *eq, u8 vecidx,
 	in->hdr.opcode = cpu_to_be16(MLX5_CMD_OP_CREATE_EQ);
 	in->ctx.log_sz_usr_page = cpu_to_be32(ilog2(eq->nent) << 24 | uar->index);
 	in->ctx.intr = vecidx;
-	in->ctx.log_page_size = PAGE_SHIFT - 12;
+	in->ctx.log_page_size = eq->buf.page_shift - MLX5_ADAPTER_PAGE_SHIFT;
 	in->events_mask = cpu_to_be64(mask);
 
 	err = mlx5_cmd_exec(dev, in, inlen, &out, sizeof(out));
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/main.c b/drivers/net/ethernet/mellanox/mlx5/core/main.c
index bc0f5fb66e24..40a9f5ed814d 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/main.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/main.c
@@ -159,6 +159,36 @@ struct mlx5_reg_host_endianess {
 	u8      rsvd[15];
 };
 
+
+#define CAP_MASK(pos, size) ((u64)((1 << (size)) - 1) << (pos))
+
+enum {
+	MLX5_CAP_BITS_RW_MASK	= CAP_MASK(MLX5_CAP_OFF_CMDIF_CSUM, 2) |
+				  CAP_MASK(MLX5_CAP_OFF_DCT, 1),
+};
+
+/* selectively copy writable fields clearing any reserved area
+ */
+static void copy_rw_fields(struct mlx5_hca_cap *to, struct mlx5_hca_cap *from)
+{
+	u64 v64;
+
+	to->log_max_qp = from->log_max_qp & 0x1f;
+	to->log_max_ra_req_dc = from->log_max_ra_req_dc & 0x3f;
+	to->log_max_ra_res_dc = from->log_max_ra_res_dc & 0x3f;
+	to->log_max_ra_req_qp = from->log_max_ra_req_qp & 0x3f;
+	to->log_max_ra_res_qp = from->log_max_ra_res_qp & 0x3f;
+	to->log_max_atomic_size_qp = from->log_max_atomic_size_qp;
+	to->log_max_atomic_size_dc = from->log_max_atomic_size_dc;
+	v64 = be64_to_cpu(from->flags) & MLX5_CAP_BITS_RW_MASK;
+	to->flags = cpu_to_be64(v64);
+}
+
+enum {
+	HCA_CAP_OPMOD_GET_MAX	= 0,
+	HCA_CAP_OPMOD_GET_CUR	= 1,
+};
+
 static int handle_hca_cap(struct mlx5_core_dev *dev)
 {
 	struct mlx5_cmd_query_hca_cap_mbox_out *query_out = NULL;
@@ -180,7 +210,7 @@ static int handle_hca_cap(struct mlx5_core_dev *dev)
 	}
 
 	query_ctx.hdr.opcode = cpu_to_be16(MLX5_CMD_OP_QUERY_HCA_CAP);
-	query_ctx.hdr.opmod  = cpu_to_be16(0x1);
+	query_ctx.hdr.opmod  = cpu_to_be16(HCA_CAP_OPMOD_GET_CUR);
 	err = mlx5_cmd_exec(dev, &query_ctx, sizeof(query_ctx),
 				 query_out, sizeof(*query_out));
 	if (err)
@@ -192,8 +222,7 @@ static int handle_hca_cap(struct mlx5_core_dev *dev)
 		goto query_ex;
 	}
 
-	memcpy(&set_ctx->hca_cap, &query_out->hca_cap,
-	       sizeof(set_ctx->hca_cap));
+	copy_rw_fields(&set_ctx->hca_cap, &query_out->hca_cap);
 
 	if (dev->profile->mask & MLX5_PROF_MASK_QP_SIZE)
 		set_ctx->hca_cap.log_max_qp = dev->profile->log_max_qp;
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/mr.c b/drivers/net/ethernet/mellanox/mlx5/core/mr.c
index 5b44e2e46daf..35e514dc7b7d 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/mr.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/mr.c
@@ -37,31 +37,41 @@
 #include "mlx5_core.h"
 
 int mlx5_core_create_mkey(struct mlx5_core_dev *dev, struct mlx5_core_mr *mr,
-			  struct mlx5_create_mkey_mbox_in *in, int inlen)
+			  struct mlx5_create_mkey_mbox_in *in, int inlen,
+			  mlx5_cmd_cbk_t callback, void *context,
+			  struct mlx5_create_mkey_mbox_out *out)
 {
-	struct mlx5_create_mkey_mbox_out out;
+	struct mlx5_create_mkey_mbox_out lout;
 	int err;
 	u8 key;
 
-	memset(&out, 0, sizeof(out));
-	spin_lock(&dev->priv.mkey_lock);
+	memset(&lout, 0, sizeof(lout));
+	spin_lock_irq(&dev->priv.mkey_lock);
 	key = dev->priv.mkey_key++;
-	spin_unlock(&dev->priv.mkey_lock);
+	spin_unlock_irq(&dev->priv.mkey_lock);
 	in->seg.qpn_mkey7_0 |= cpu_to_be32(key);
 	in->hdr.opcode = cpu_to_be16(MLX5_CMD_OP_CREATE_MKEY);
-	err = mlx5_cmd_exec(dev, in, inlen, &out, sizeof(out));
+	if (callback) {
+		err = mlx5_cmd_exec_cb(dev, in, inlen, out, sizeof(*out),
+				       callback, context);
+		return err;
+	} else {
+		err = mlx5_cmd_exec(dev, in, inlen, &lout, sizeof(lout));
+	}
+
 	if (err) {
 		mlx5_core_dbg(dev, "cmd exec faile %d\n", err);
 		return err;
 	}
 
-	if (out.hdr.status) {
-		mlx5_core_dbg(dev, "status %d\n", out.hdr.status);
-		return mlx5_cmd_status_to_err(&out.hdr);
+	if (lout.hdr.status) {
+		mlx5_core_dbg(dev, "status %d\n", lout.hdr.status);
+		return mlx5_cmd_status_to_err(&lout.hdr);
 	}
 
-	mr->key = mlx5_idx_to_mkey(be32_to_cpu(out.mkey) & 0xffffff) | key;
-	mlx5_core_dbg(dev, "out 0x%x, key 0x%x, mkey 0x%x\n", be32_to_cpu(out.mkey), key, mr->key);
+	mr->key = mlx5_idx_to_mkey(be32_to_cpu(lout.mkey) & 0xffffff) | key;
+	mlx5_core_dbg(dev, "out 0x%x, key 0x%x, mkey 0x%x\n",
+		      be32_to_cpu(lout.mkey), key, mr->key);
 
 	return err;
 }
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/pagealloc.c b/drivers/net/ethernet/mellanox/mlx5/core/pagealloc.c
index 7b12acf210f8..37b6ad1f9a1b 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/pagealloc.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/pagealloc.c
@@ -57,10 +57,13 @@ struct mlx5_pages_req {
 };
 
 struct fw_page {
-	struct rb_node	rb_node;
-	u64		addr;
-	struct page	*page;
-	u16		func_id;
+	struct rb_node		rb_node;
+	u64			addr;
+	struct page	       *page;
+	u16			func_id;
+	unsigned long		bitmask;
+	struct list_head	list;
+	unsigned		free_count;
 };
 
 struct mlx5_query_pages_inbox {
@@ -94,6 +97,11 @@ enum {
 	MAX_RECLAIM_TIME_MSECS	= 5000,
 };
 
+enum {
+	MLX5_MAX_RECLAIM_TIME_MILI	= 5000,
+	MLX5_NUM_4K_IN_PAGE		= PAGE_SIZE / 4096,
+};
+
 static int insert_page(struct mlx5_core_dev *dev, u64 addr, struct page *page, u16 func_id)
 {
 	struct rb_root *root = &dev->priv.page_root;
@@ -101,6 +109,7 @@ static int insert_page(struct mlx5_core_dev *dev, u64 addr, struct page *page, u
 	struct rb_node *parent = NULL;
 	struct fw_page *nfp;
 	struct fw_page *tfp;
+	int i;
 
 	while (*new) {
 		parent = *new;
@@ -113,25 +122,29 @@ static int insert_page(struct mlx5_core_dev *dev, u64 addr, struct page *page, u
 			return -EEXIST;
 	}
 
-	nfp = kmalloc(sizeof(*nfp), GFP_KERNEL);
+	nfp = kzalloc(sizeof(*nfp), GFP_KERNEL);
 	if (!nfp)
 		return -ENOMEM;
 
 	nfp->addr = addr;
 	nfp->page = page;
 	nfp->func_id = func_id;
+	nfp->free_count = MLX5_NUM_4K_IN_PAGE;
+	for (i = 0; i < MLX5_NUM_4K_IN_PAGE; i++)
+		set_bit(i, &nfp->bitmask);
 
 	rb_link_node(&nfp->rb_node, parent, new);
 	rb_insert_color(&nfp->rb_node, root);
+	list_add(&nfp->list, &dev->priv.free_list);
 
 	return 0;
 }
 
-static struct page *remove_page(struct mlx5_core_dev *dev, u64 addr)
+static struct fw_page *find_fw_page(struct mlx5_core_dev *dev, u64 addr)
 {
 	struct rb_root *root = &dev->priv.page_root;
 	struct rb_node *tmp = root->rb_node;
-	struct page *result = NULL;
+	struct fw_page *result = NULL;
 	struct fw_page *tfp;
 
 	while (tmp) {
@@ -141,9 +154,7 @@ static struct page *remove_page(struct mlx5_core_dev *dev, u64 addr)
 		} else if (tfp->addr > addr) {
 			tmp = tmp->rb_right;
 		} else {
-			rb_erase(&tfp->rb_node, root);
-			result = tfp->page;
-			kfree(tfp);
+			result = tfp;
 			break;
 		}
 	}
@@ -176,12 +187,98 @@ static int mlx5_cmd_query_pages(struct mlx5_core_dev *dev, u16 *func_id,
 	return err;
 }
 
+static int alloc_4k(struct mlx5_core_dev *dev, u64 *addr)
+{
+	struct fw_page *fp;
+	unsigned n;
+
+	if (list_empty(&dev->priv.free_list)) {
+		return -ENOMEM;
+		mlx5_core_warn(dev, "\n");
+	}
+
+	fp = list_entry(dev->priv.free_list.next, struct fw_page, list);
+	n = find_first_bit(&fp->bitmask, 8 * sizeof(fp->bitmask));
+	if (n >= MLX5_NUM_4K_IN_PAGE) {
+		mlx5_core_warn(dev, "alloc 4k bug\n");
+		return -ENOENT;
+	}
+	clear_bit(n, &fp->bitmask);
+	fp->free_count--;
+	if (!fp->free_count)
+		list_del(&fp->list);
+
+	*addr = fp->addr + n * 4096;
+
+	return 0;
+}
+
+static void free_4k(struct mlx5_core_dev *dev, u64 addr)
+{
+	struct fw_page *fwp;
+	int n;
+
+	fwp = find_fw_page(dev, addr & PAGE_MASK);
+	if (!fwp) {
+		mlx5_core_warn(dev, "page not found\n");
+		return;
+	}
+
+	n = (addr & ~PAGE_MASK) % 4096;
+	fwp->free_count++;
+	set_bit(n, &fwp->bitmask);
+	if (fwp->free_count == MLX5_NUM_4K_IN_PAGE) {
+		rb_erase(&fwp->rb_node, &dev->priv.page_root);
+		if (fwp->free_count != 1)
+			list_del(&fwp->list);
+		dma_unmap_page(&dev->pdev->dev, addr, PAGE_SIZE, DMA_BIDIRECTIONAL);
+		__free_page(fwp->page);
+		kfree(fwp);
+	} else if (fwp->free_count == 1) {
+		list_add(&fwp->list, &dev->priv.free_list);
+	}
+}
+
+static int alloc_system_page(struct mlx5_core_dev *dev, u16 func_id)
+{
+	struct page *page;
+	u64 addr;
+	int err;
+
+	page = alloc_page(GFP_HIGHUSER);
+	if (!page) {
+		mlx5_core_warn(dev, "failed to allocate page\n");
+		return -ENOMEM;
+	}
+	addr = dma_map_page(&dev->pdev->dev, page, 0,
+			    PAGE_SIZE, DMA_BIDIRECTIONAL);
+	if (dma_mapping_error(&dev->pdev->dev, addr)) {
+		mlx5_core_warn(dev, "failed dma mapping page\n");
+		err = -ENOMEM;
+		goto out_alloc;
+	}
+	err = insert_page(dev, addr, page, func_id);
+	if (err) {
+		mlx5_core_err(dev, "failed to track allocated page\n");
+		goto out_mapping;
+	}
+
+	return 0;
+
+out_mapping:
+	dma_unmap_page(&dev->pdev->dev, addr, PAGE_SIZE, DMA_BIDIRECTIONAL);
+
+out_alloc:
+	__free_page(page);
+
+	return err;
+}
 static int give_pages(struct mlx5_core_dev *dev, u16 func_id, int npages,
 		      int notify_fail)
 {
 	struct mlx5_manage_pages_inbox *in;
 	struct mlx5_manage_pages_outbox out;
-	struct page *page;
+	struct mlx5_manage_pages_inbox *nin;
 	int inlen;
 	u64 addr;
 	int err;
@@ -196,27 +293,15 @@ static int give_pages(struct mlx5_core_dev *dev, u16 func_id, int npages,
 	memset(&out, 0, sizeof(out));
 
 	for (i = 0; i < npages; i++) {
-		page = alloc_page(GFP_HIGHUSER);
-		if (!page) {
-			err = -ENOMEM;
-			mlx5_core_warn(dev, "failed to allocate page\n");
-			goto out_alloc;
-		}
-		addr = dma_map_page(&dev->pdev->dev, page, 0,
-				    PAGE_SIZE, DMA_BIDIRECTIONAL);
-		if (dma_mapping_error(&dev->pdev->dev, addr)) {
-			mlx5_core_warn(dev, "failed dma mapping page\n");
-			__free_page(page);
-			err = -ENOMEM;
-			goto out_alloc;
-		}
-		err = insert_page(dev, addr, page, func_id);
+retry:
+		err = alloc_4k(dev, &addr);
 		if (err) {
-			mlx5_core_err(dev, "failed to track allocated page\n");
-			dma_unmap_page(&dev->pdev->dev, addr, PAGE_SIZE, DMA_BIDIRECTIONAL);
-			__free_page(page);
-			err = -ENOMEM;
-			goto out_alloc;
+			if (err == -ENOMEM)
+				err = alloc_system_page(dev, func_id);
+			if (err)
+				goto out_4k;
+
+			goto retry;
 		}
 		in->pas[i] = cpu_to_be64(addr);
 	}
@@ -226,7 +311,6 @@ static int give_pages(struct mlx5_core_dev *dev, u16 func_id, int npages,
 	in->func_id = cpu_to_be16(func_id);
 	in->num_entries = cpu_to_be32(npages);
 	err = mlx5_cmd_exec(dev, in, inlen, &out, sizeof(out));
-	mlx5_core_dbg(dev, "err %d\n", err);
 	if (err) {
 		mlx5_core_warn(dev, "func_id 0x%x, npages %d, err %d\n", func_id, npages, err);
 		goto out_alloc;
@@ -247,25 +331,22 @@ static int give_pages(struct mlx5_core_dev *dev, u16 func_id, int npages,
 
 out_alloc:
 	if (notify_fail) {
-		memset(in, 0, inlen);
-		memset(&out, 0, sizeof(out));
-		in->hdr.opcode = cpu_to_be16(MLX5_CMD_OP_MANAGE_PAGES);
-		in->hdr.opmod = cpu_to_be16(MLX5_PAGES_CANT_GIVE);
-		if (mlx5_cmd_exec(dev, in, sizeof(*in), &out, sizeof(out)))
-			mlx5_core_warn(dev, "\n");
-	}
-	for (i--; i >= 0; i--) {
-		addr = be64_to_cpu(in->pas[i]);
-		page = remove_page(dev, addr);
-		if (!page) {
-			mlx5_core_err(dev, "BUG: can't remove page at addr 0x%llx\n",
-				      addr);
-			continue;
+		nin = kzalloc(sizeof(*nin), GFP_KERNEL);
+		if (!nin) {
+			mlx5_core_warn(dev, "allocation failed\n");
+			goto out_4k;
 		}
-		dma_unmap_page(&dev->pdev->dev, addr, PAGE_SIZE, DMA_BIDIRECTIONAL);
-		__free_page(page);
+		memset(&out, 0, sizeof(out));
+		nin->hdr.opcode = cpu_to_be16(MLX5_CMD_OP_MANAGE_PAGES);
+		nin->hdr.opmod = cpu_to_be16(MLX5_PAGES_CANT_GIVE);
+		if (mlx5_cmd_exec(dev, nin, sizeof(*nin), &out, sizeof(out)))
+			mlx5_core_warn(dev, "page notify failed\n");
+		kfree(nin);
 	}
 
+out_4k:
+	for (i--; i >= 0; i--)
+		free_4k(dev, be64_to_cpu(in->pas[i]));
 out_free:
 	mlx5_vfree(in);
 	return err;
@@ -276,7 +357,6 @@ static int reclaim_pages(struct mlx5_core_dev *dev, u32 func_id, int npages,
 {
 	struct mlx5_manage_pages_inbox   in;
 	struct mlx5_manage_pages_outbox *out;
-	struct page *page;
 	int num_claimed;
 	int outlen;
 	u64 addr;
@@ -315,13 +395,7 @@ static int reclaim_pages(struct mlx5_core_dev *dev, u32 func_id, int npages,
 
 	for (i = 0; i < num_claimed; i++) {
 		addr = be64_to_cpu(out->pas[i]);
-		page = remove_page(dev, addr);
-		if (!page) {
-			mlx5_core_warn(dev, "FW reported unknown DMA address 0x%llx\n", addr);
-		} else {
-			dma_unmap_page(&dev->pdev->dev, addr, PAGE_SIZE, DMA_BIDIRECTIONAL);
-			__free_page(page);
-		}
+		free_4k(dev, addr);
 	}
 
 out_free:
@@ -381,14 +455,19 @@ int mlx5_satisfy_startup_pages(struct mlx5_core_dev *dev, int boot)
 	return give_pages(dev, func_id, npages, 0);
 }
 
+enum {
+	MLX5_BLKS_FOR_RECLAIM_PAGES = 12
+};
+
 static int optimal_reclaimed_pages(void)
 {
 	struct mlx5_cmd_prot_block *block;
 	struct mlx5_cmd_layout *lay;
 	int ret;
 
-	ret = (sizeof(lay->in) + sizeof(block->data) -
-	       sizeof(struct mlx5_manage_pages_outbox)) / 8;
+	ret = (sizeof(lay->out) + MLX5_BLKS_FOR_RECLAIM_PAGES * sizeof(block->data) -
+	       sizeof(struct mlx5_manage_pages_outbox)) /
+	       FIELD_SIZEOF(struct mlx5_manage_pages_outbox, pas[0]);
 
 	return ret;
 }
@@ -427,6 +506,7 @@ int mlx5_reclaim_startup_pages(struct mlx5_core_dev *dev)
 void mlx5_pagealloc_init(struct mlx5_core_dev *dev)
 {
 	dev->priv.page_root = RB_ROOT;
+	INIT_LIST_HEAD(&dev->priv.free_list);
 }
 
 void mlx5_pagealloc_cleanup(struct mlx5_core_dev *dev)
diff --git a/drivers/net/ethernet/micrel/ks8842.c b/drivers/net/ethernet/micrel/ks8842.c
index 0951f7aca1ef..822616e3c375 100644
--- a/drivers/net/ethernet/micrel/ks8842.c
+++ b/drivers/net/ethernet/micrel/ks8842.c
@@ -459,8 +459,7 @@ static int ks8842_tx_frame_dma(struct sk_buff *skb, struct net_device *netdev)
 		sg_dma_len(&ctl->sg) += 4 - sg_dma_len(&ctl->sg) % 4;
 
 	ctl->adesc = dmaengine_prep_slave_sg(ctl->chan,
-		&ctl->sg, 1, DMA_MEM_TO_DEV,
-		DMA_PREP_INTERRUPT | DMA_COMPL_SKIP_SRC_UNMAP);
+		&ctl->sg, 1, DMA_MEM_TO_DEV, DMA_PREP_INTERRUPT);
 	if (!ctl->adesc)
 		return NETDEV_TX_BUSY;
 
@@ -571,8 +570,7 @@ static int __ks8842_start_new_rx_dma(struct net_device *netdev)
 		sg_dma_len(sg) = DMA_BUFFER_SIZE;
 
 		ctl->adesc = dmaengine_prep_slave_sg(ctl->chan,
-			sg, 1, DMA_DEV_TO_MEM,
-			DMA_PREP_INTERRUPT | DMA_COMPL_SKIP_SRC_UNMAP);
+			sg, 1, DMA_DEV_TO_MEM, DMA_PREP_INTERRUPT);
 
 		if (!ctl->adesc)
 			goto out;
diff --git a/drivers/net/ethernet/neterion/vxge/vxge-main.c b/drivers/net/ethernet/neterion/vxge/vxge-main.c
index 8614eeb7de81..f9876ea8c8bf 100644
--- a/drivers/net/ethernet/neterion/vxge/vxge-main.c
+++ b/drivers/net/ethernet/neterion/vxge/vxge-main.c
@@ -2072,6 +2072,10 @@ static int vxge_open_vpaths(struct vxgedev *vdev)
 				vdev->config.tx_steering_type;
 			vpath->fifo.ndev = vdev->ndev;
 			vpath->fifo.pdev = vdev->pdev;
+
+			u64_stats_init(&vpath->fifo.stats.syncp);
+			u64_stats_init(&vpath->ring.stats.syncp);
+
 			if (vdev->config.tx_steering_type)
 				vpath->fifo.txq =
 					netdev_get_tx_queue(vdev->ndev, i);
diff --git a/drivers/net/ethernet/nvidia/forcedeth.c b/drivers/net/ethernet/nvidia/forcedeth.c
index 098b96dad66f..2d045be4b5cf 100644
--- a/drivers/net/ethernet/nvidia/forcedeth.c
+++ b/drivers/net/ethernet/nvidia/forcedeth.c
@@ -5619,6 +5619,8 @@ static int nv_probe(struct pci_dev *pci_dev, const struct pci_device_id *id)
 	spin_lock_init(&np->lock);
 	spin_lock_init(&np->hwstats_lock);
 	SET_NETDEV_DEV(dev, &pci_dev->dev);
+	u64_stats_init(&np->swstats_rx_syncp);
+	u64_stats_init(&np->swstats_tx_syncp);
 
 	init_timer(&np->oom_kick);
 	np->oom_kick.data = (unsigned long) dev;
diff --git a/drivers/net/ethernet/oki-semi/pch_gbe/pch_gbe_main.c b/drivers/net/ethernet/oki-semi/pch_gbe/pch_gbe_main.c
index 5a0f04c2c813..27ffe0ebf0a6 100644
--- a/drivers/net/ethernet/oki-semi/pch_gbe/pch_gbe_main.c
+++ b/drivers/net/ethernet/oki-semi/pch_gbe/pch_gbe_main.c
@@ -245,16 +245,8 @@ static int hwtstamp_ioctl(struct net_device *netdev, struct ifreq *ifr, int cmd)
 	/* Get ieee1588's dev information */
 	pdev = adapter->ptp_pdev;
 
-	switch (cfg.tx_type) {
-	case HWTSTAMP_TX_OFF:
-		adapter->hwts_tx_en = 0;
-		break;
-	case HWTSTAMP_TX_ON:
-		adapter->hwts_tx_en = 1;
-		break;
-	default:
+	if (cfg.tx_type != HWTSTAMP_TX_OFF && cfg.tx_type != HWTSTAMP_TX_ON)
 		return -ERANGE;
-	}
 
 	switch (cfg.rx_filter) {
 	case HWTSTAMP_FILTER_NONE:
@@ -284,6 +276,8 @@ static int hwtstamp_ioctl(struct net_device *netdev, struct ifreq *ifr, int cmd)
 		return -ERANGE;
 	}
 
+	adapter->hwts_tx_en = cfg.tx_type == HWTSTAMP_TX_ON;
+
 	/* Clear out any old time stamps. */
 	pch_ch_event_write(pdev, TX_SNAPSHOT_LOCKED | RX_SNAPSHOT_LOCKED);
 
diff --git a/drivers/net/ethernet/qlogic/qlcnic/qlcnic_83xx_hw.c b/drivers/net/ethernet/qlogic/qlcnic/qlcnic_83xx_hw.c
index 09810ddd11ec..b1cb0ffb15c7 100644
--- a/drivers/net/ethernet/qlogic/qlcnic/qlcnic_83xx_hw.c
+++ b/drivers/net/ethernet/qlogic/qlcnic/qlcnic_83xx_hw.c
@@ -1730,7 +1730,7 @@ static void qlcnic_extend_lb_idc_cmpltn_wait(struct qlcnic_adapter *adapter,
 	struct qlcnic_hardware_context *ahw = adapter->ahw;
 	int temp;
 
-	netdev_info(adapter->netdev, "Recieved loopback IDC time extend event for 0x%x seconds\n",
+	netdev_info(adapter->netdev, "Received loopback IDC time extend event for 0x%x seconds\n",
 		    ahw->extend_lb_time);
 	temp = ahw->extend_lb_time * 1000;
 	*max_wait_count += temp / QLC_83XX_LB_MSLEEP_COUNT;
@@ -3537,7 +3537,7 @@ int qlcnic_83xx_resume(struct qlcnic_adapter *adapter)
 
 void qlcnic_83xx_reinit_mbx_work(struct qlcnic_mailbox *mbx)
 {
-	INIT_COMPLETION(mbx->completion);
+	reinit_completion(&mbx->completion);
 	set_bit(QLC_83XX_MBX_READY, &mbx->status);
 }
 
diff --git a/drivers/net/ethernet/realtek/8139too.c b/drivers/net/ethernet/realtek/8139too.c
index 50a92104dd0a..da5972eefdd2 100644
--- a/drivers/net/ethernet/realtek/8139too.c
+++ b/drivers/net/ethernet/realtek/8139too.c
@@ -790,6 +790,9 @@ static struct net_device *rtl8139_init_board(struct pci_dev *pdev)
 
 	pci_set_master (pdev);
 
+	u64_stats_init(&tp->rx_stats.syncp);
+	u64_stats_init(&tp->tx_stats.syncp);
+
 retry:
 	/* PIO bar register comes first. */
 	bar = !use_io;
diff --git a/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c b/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c
index 8d4ccd35a016..8a7a23a84ac5 100644
--- a/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c
+++ b/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c
@@ -435,16 +435,9 @@ static int stmmac_hwtstamp_ioctl(struct net_device *dev, struct ifreq *ifr)
 	if (config.flags)
 		return -EINVAL;
 
-	switch (config.tx_type) {
-	case HWTSTAMP_TX_OFF:
-		priv->hwts_tx_en = 0;
-		break;
-	case HWTSTAMP_TX_ON:
-		priv->hwts_tx_en = 1;
-		break;
-	default:
+	if (config.tx_type != HWTSTAMP_TX_OFF &&
+	    config.tx_type != HWTSTAMP_TX_ON)
 		return -ERANGE;
-	}
 
 	if (priv->adv_ts) {
 		switch (config.rx_filter) {
@@ -576,6 +569,7 @@ static int stmmac_hwtstamp_ioctl(struct net_device *dev, struct ifreq *ifr)
 		}
 	}
 	priv->hwts_rx_en = ((config.rx_filter == HWTSTAMP_FILTER_NONE) ? 0 : 1);
+	priv->hwts_tx_en = config.tx_type == HWTSTAMP_TX_ON;
 
 	if (!priv->hwts_tx_en && !priv->hwts_rx_en)
 		priv->hw->ptp->config_hw_tstamping(priv->ioaddr, 0);
diff --git a/drivers/net/ethernet/ti/cpsw.c b/drivers/net/ethernet/ti/cpsw.c
index 90d41d26ec6d..7536a4c01293 100644
--- a/drivers/net/ethernet/ti/cpsw.c
+++ b/drivers/net/ethernet/ti/cpsw.c
@@ -967,14 +967,19 @@ static inline void cpsw_add_dual_emac_def_ale_entries(
 		priv->host_port, ALE_VLAN, slave->port_vlan);
 }
 
-static void cpsw_slave_open(struct cpsw_slave *slave, struct cpsw_priv *priv)
+static void soft_reset_slave(struct cpsw_slave *slave)
 {
 	char name[32];
-	u32 slave_port;
-
-	sprintf(name, "slave-%d", slave->slave_num);
 
+	snprintf(name, sizeof(name), "slave-%d", slave->slave_num);
 	soft_reset(name, &slave->sliver->soft_reset);
+}
+
+static void cpsw_slave_open(struct cpsw_slave *slave, struct cpsw_priv *priv)
+{
+	u32 slave_port;
+
+	soft_reset_slave(slave);
 
 	/* setup priority mapping */
 	__raw_writel(RX_PRIORITY_MAPPING, &slave->sliver->rx_pri_map);
@@ -1323,6 +1328,10 @@ static int cpsw_hwtstamp_ioctl(struct net_device *dev, struct ifreq *ifr)
 	struct cpts *cpts = priv->cpts;
 	struct hwtstamp_config cfg;
 
+	if (priv->version != CPSW_VERSION_1 &&
+	    priv->version != CPSW_VERSION_2)
+		return -EOPNOTSUPP;
+
 	if (copy_from_user(&cfg, ifr->ifr_data, sizeof(cfg)))
 		return -EFAULT;
 
@@ -1330,16 +1339,8 @@ static int cpsw_hwtstamp_ioctl(struct net_device *dev, struct ifreq *ifr)
 	if (cfg.flags)
 		return -EINVAL;
 
-	switch (cfg.tx_type) {
-	case HWTSTAMP_TX_OFF:
-		cpts->tx_enable = 0;
-		break;
-	case HWTSTAMP_TX_ON:
-		cpts->tx_enable = 1;
-		break;
-	default:
+	if (cfg.tx_type != HWTSTAMP_TX_OFF && cfg.tx_type != HWTSTAMP_TX_ON)
 		return -ERANGE;
-	}
 
 	switch (cfg.rx_filter) {
 	case HWTSTAMP_FILTER_NONE:
@@ -1366,6 +1367,8 @@ static int cpsw_hwtstamp_ioctl(struct net_device *dev, struct ifreq *ifr)
 		return -ERANGE;
 	}
 
+	cpts->tx_enable = cfg.tx_type == HWTSTAMP_TX_ON;
+
 	switch (priv->version) {
 	case CPSW_VERSION_1:
 		cpsw_hwtstamp_v1(priv);
@@ -1374,7 +1377,7 @@ static int cpsw_hwtstamp_ioctl(struct net_device *dev, struct ifreq *ifr)
 		cpsw_hwtstamp_v2(priv);
 		break;
 	default:
-		return -ENOTSUPP;
+		WARN_ON(1);
 	}
 
 	return copy_to_user(ifr->ifr_data, &cfg, sizeof(cfg)) ? -EFAULT : 0;
@@ -2173,8 +2176,9 @@ static int cpsw_suspend(struct device *dev)
 
 	if (netif_running(ndev))
 		cpsw_ndo_stop(ndev);
-	soft_reset("sliver 0", &priv->slaves[0].sliver->soft_reset);
-	soft_reset("sliver 1", &priv->slaves[1].sliver->soft_reset);
+
+	for_each_slave(priv, soft_reset_slave);
+
 	pm_runtime_put_sync(&pdev->dev);
 
 	/* Select sleep pin state */
diff --git a/drivers/net/ethernet/tile/tilepro.c b/drivers/net/ethernet/tile/tilepro.c
index 106be47716e7..edb2e12a0fe2 100644
--- a/drivers/net/ethernet/tile/tilepro.c
+++ b/drivers/net/ethernet/tile/tilepro.c
@@ -1008,6 +1008,8 @@ static void tile_net_register(void *dev_ptr)
 	info->egress_timer.data = (long)info;
 	info->egress_timer.function = tile_net_handle_egress_timer;
 
+	u64_stats_init(&info->stats.syncp);
+
 	priv->cpu[my_cpu] = info;
 
 	/*
diff --git a/drivers/net/ethernet/via/via-rhine.c b/drivers/net/ethernet/via/via-rhine.c
index 4a7293ed95e9..cce6c4bc556a 100644
--- a/drivers/net/ethernet/via/via-rhine.c
+++ b/drivers/net/ethernet/via/via-rhine.c
@@ -987,6 +987,9 @@ static int rhine_init_one(struct pci_dev *pdev, const struct pci_device_id *ent)
 
 	rp->base = ioaddr;
 
+	u64_stats_init(&rp->tx_stats.syncp);
+	u64_stats_init(&rp->rx_stats.syncp);
+
 	/* Get chip registers into a sane state */
 	rhine_power_init(dev);
 	rhine_hw_init(dev, pioaddr);
diff --git a/drivers/net/ethernet/xscale/ixp4xx_eth.c b/drivers/net/ethernet/xscale/ixp4xx_eth.c
index e78802e75ea6..bcc224a83734 100644
--- a/drivers/net/ethernet/xscale/ixp4xx_eth.c
+++ b/drivers/net/ethernet/xscale/ixp4xx_eth.c
@@ -389,16 +389,8 @@ static int hwtstamp_ioctl(struct net_device *netdev, struct ifreq *ifr, int cmd)
 	ch = PORT2CHANNEL(port);
 	regs = (struct ixp46x_ts_regs __iomem *) IXP4XX_TIMESYNC_BASE_VIRT;
 
-	switch (cfg.tx_type) {
-	case HWTSTAMP_TX_OFF:
-		port->hwts_tx_en = 0;
-		break;
-	case HWTSTAMP_TX_ON:
-		port->hwts_tx_en = 1;
-		break;
-	default:
+	if (cfg.tx_type != HWTSTAMP_TX_OFF && cfg.tx_type != HWTSTAMP_TX_ON)
 		return -ERANGE;
-	}
 
 	switch (cfg.rx_filter) {
 	case HWTSTAMP_FILTER_NONE:
@@ -416,6 +408,8 @@ static int hwtstamp_ioctl(struct net_device *netdev, struct ifreq *ifr, int cmd)
 		return -ERANGE;
 	}
 
+	port->hwts_tx_en = cfg.tx_type == HWTSTAMP_TX_ON;
+
 	/* Clear out any old time stamps. */
 	__raw_writel(TX_SNAPSHOT_LOCKED | RX_SNAPSHOT_LOCKED,
 		     &regs->channel[ch].ch_event);
diff --git a/drivers/net/ieee802154/at86rf230.c b/drivers/net/ieee802154/at86rf230.c
index 6f10b4964726..2cbe1c249996 100644
--- a/drivers/net/ieee802154/at86rf230.c
+++ b/drivers/net/ieee802154/at86rf230.c
@@ -561,7 +561,7 @@ at86rf230_xmit(struct ieee802154_dev *dev, struct sk_buff *skb)
 
 	spin_lock_irqsave(&lp->lock, flags);
 	lp->is_tx = 1;
-	INIT_COMPLETION(lp->tx_complete);
+	reinit_completion(&lp->tx_complete);
 	spin_unlock_irqrestore(&lp->lock, flags);
 
 	rc = at86rf230_write_fbuf(lp, skb->data, skb->len);
diff --git a/drivers/net/ieee802154/mrf24j40.c b/drivers/net/ieee802154/mrf24j40.c
index 0632d34905c7..c6e46d6e9f75 100644
--- a/drivers/net/ieee802154/mrf24j40.c
+++ b/drivers/net/ieee802154/mrf24j40.c
@@ -343,7 +343,7 @@ static int mrf24j40_tx(struct ieee802154_dev *dev, struct sk_buff *skb)
 	if (ret)
 		goto err;
 
-	INIT_COMPLETION(devrec->tx_complete);
+	reinit_completion(&devrec->tx_complete);
 
 	/* Set TXNTRIG bit of TXNCON to send packet */
 	ret = read_short_reg(devrec, REG_TXNCON, &val);
diff --git a/drivers/net/ifb.c b/drivers/net/ifb.c
index a3bed28197d2..c14d39bf32d0 100644
--- a/drivers/net/ifb.c
+++ b/drivers/net/ifb.c
@@ -265,6 +265,7 @@ MODULE_PARM_DESC(numifbs, "Number of ifb devices");
 static int __init ifb_init_one(int index)
 {
 	struct net_device *dev_ifb;
+	struct ifb_private *dp;
 	int err;
 
 	dev_ifb = alloc_netdev(sizeof(struct ifb_private),
@@ -273,6 +274,10 @@ static int __init ifb_init_one(int index)
 	if (!dev_ifb)
 		return -ENOMEM;
 
+	dp = netdev_priv(dev_ifb);
+	u64_stats_init(&dp->rsync);
+	u64_stats_init(&dp->tsync);
+
 	dev_ifb->rtnl_link_ops = &ifb_link_ops;
 	err = register_netdevice(dev_ifb);
 	if (err < 0)
diff --git a/drivers/net/irda/ali-ircc.c b/drivers/net/irda/ali-ircc.c
index 7bbd318bc93e..befa45f809c3 100644
--- a/drivers/net/irda/ali-ircc.c
+++ b/drivers/net/irda/ali-ircc.c
@@ -627,7 +627,7 @@ static int ali_ircc_setup(chipio_t *info)
 /*
  * Function ali_ircc_read_dongle_id (int index, info)
  *
- * Try to read dongle indentification. This procedure needs to be executed
+ * Try to read dongle identification. This procedure needs to be executed
  * once after power-on/reset. It also needs to be used whenever you suspect
  * that the user may have plugged/unplugged the IrDA Dongle.
  */
diff --git a/drivers/net/irda/nsc-ircc.c b/drivers/net/irda/nsc-ircc.c
index ceeb53737f86..66bc03bdb138 100644
--- a/drivers/net/irda/nsc-ircc.c
+++ b/drivers/net/irda/nsc-ircc.c
@@ -1035,7 +1035,7 @@ static int nsc_ircc_setup(chipio_t *info)
 /*
  * Function nsc_ircc_read_dongle_id (void)
  *
- * Try to read dongle indentification. This procedure needs to be executed
+ * Try to read dongle identification. This procedure needs to be executed
  * once after power-on/reset. It also needs to be used whenever you suspect
  * that the user may have plugged/unplugged the IrDA Dongle.
  */
diff --git a/drivers/net/loopback.c b/drivers/net/loopback.c
index a17d85a331f1..ac24c27b4b2d 100644
--- a/drivers/net/loopback.c
+++ b/drivers/net/loopback.c
@@ -137,10 +137,16 @@ static const struct ethtool_ops loopback_ethtool_ops = {
 
 static int loopback_dev_init(struct net_device *dev)
 {
+	int i;
 	dev->lstats = alloc_percpu(struct pcpu_lstats);
 	if (!dev->lstats)
 		return -ENOMEM;
 
+	for_each_possible_cpu(i) {
+		struct pcpu_lstats *lb_stats;
+		lb_stats = per_cpu_ptr(dev->lstats, i);
+		u64_stats_init(&lb_stats->syncp);
+	}
 	return 0;
 }
 
diff --git a/drivers/net/macvlan.c b/drivers/net/macvlan.c
index af4aaa5893ff..acf93798dc67 100644
--- a/drivers/net/macvlan.c
+++ b/drivers/net/macvlan.c
@@ -534,6 +534,7 @@ static int macvlan_init(struct net_device *dev)
 {
 	struct macvlan_dev *vlan = netdev_priv(dev);
 	const struct net_device *lowerdev = vlan->lowerdev;
+	int i;
 
 	dev->state		= (dev->state & ~MACVLAN_STATE_MASK) |
 				  (lowerdev->state & MACVLAN_STATE_MASK);
@@ -549,6 +550,12 @@ static int macvlan_init(struct net_device *dev)
 	if (!vlan->pcpu_stats)
 		return -ENOMEM;
 
+	for_each_possible_cpu(i) {
+		struct macvlan_pcpu_stats *mvlstats;
+		mvlstats = per_cpu_ptr(vlan->pcpu_stats, i);
+		u64_stats_init(&mvlstats->syncp);
+	}
+
 	return 0;
 }
 
diff --git a/drivers/net/macvtap.c b/drivers/net/macvtap.c
index 9dccb1edfd2a..dc76670c2f2a 100644
--- a/drivers/net/macvtap.c
+++ b/drivers/net/macvtap.c
@@ -628,6 +628,7 @@ static ssize_t macvtap_get_user(struct macvtap_queue *q, struct msghdr *m,
 				const struct iovec *iv, unsigned long total_len,
 				size_t count, int noblock)
 {
+	int good_linear = SKB_MAX_HEAD(NET_IP_ALIGN);
 	struct sk_buff *skb;
 	struct macvlan_dev *vlan;
 	unsigned long len = total_len;
@@ -670,6 +671,8 @@ static ssize_t macvtap_get_user(struct macvtap_queue *q, struct msghdr *m,
 
 	if (m && m->msg_control && sock_flag(&q->sk, SOCK_ZEROCOPY)) {
 		copylen = vnet_hdr.hdr_len ? vnet_hdr.hdr_len : GOODCOPY_LEN;
+		if (copylen > good_linear)
+			copylen = good_linear;
 		linear = copylen;
 		if (iov_pages(iv, vnet_hdr_len + copylen, count)
 		    <= MAX_SKB_FRAGS)
@@ -678,7 +681,10 @@ static ssize_t macvtap_get_user(struct macvtap_queue *q, struct msghdr *m,
 
 	if (!zerocopy) {
 		copylen = len;
-		linear = vnet_hdr.hdr_len;
+		if (vnet_hdr.hdr_len > good_linear)
+			linear = good_linear;
+		else
+			linear = vnet_hdr.hdr_len;
 	}
 
 	skb = macvtap_alloc_skb(&q->sk, NET_IP_ALIGN, copylen,
diff --git a/drivers/net/nlmon.c b/drivers/net/nlmon.c
index b57ce5f48962..d2bb12bfabd5 100644
--- a/drivers/net/nlmon.c
+++ b/drivers/net/nlmon.c
@@ -47,8 +47,16 @@ static int nlmon_change_mtu(struct net_device *dev, int new_mtu)
 
 static int nlmon_dev_init(struct net_device *dev)
 {
+	int i;
+
 	dev->lstats = alloc_percpu(struct pcpu_lstats);
 
+	for_each_possible_cpu(i) {
+		struct pcpu_lstats *nlmstats;
+		nlmstats = per_cpu_ptr(dev->lstats, i);
+		u64_stats_init(&nlmstats->syncp);
+	}
+
 	return dev->lstats == NULL ? -ENOMEM : 0;
 }
 
diff --git a/drivers/net/team/team.c b/drivers/net/team/team.c
index 50e43e64d51d..34b0de09d881 100644
--- a/drivers/net/team/team.c
+++ b/drivers/net/team/team.c
@@ -1540,6 +1540,12 @@ static int team_init(struct net_device *dev)
 	if (!team->pcpu_stats)
 		return -ENOMEM;
 
+	for_each_possible_cpu(i) {
+		struct team_pcpu_stats *team_stats;
+		team_stats = per_cpu_ptr(team->pcpu_stats, i);
+		u64_stats_init(&team_stats->syncp);
+	}
+
 	for (i = 0; i < TEAM_PORT_HASHENTRIES; i++)
 		INIT_HLIST_HEAD(&team->en_port_hlist[i]);
 	INIT_LIST_HEAD(&team->port_list);
@@ -2644,7 +2650,7 @@ static int team_nl_cmd_port_list_get(struct sk_buff *skb,
 	return err;
 }
 
-static struct genl_ops team_nl_ops[] = {
+static const struct genl_ops team_nl_ops[] = {
 	{
 		.cmd = TEAM_CMD_NOOP,
 		.doit = team_nl_cmd_noop,
@@ -2670,15 +2676,15 @@ static struct genl_ops team_nl_ops[] = {
 	},
 };
 
-static struct genl_multicast_group team_change_event_mcgrp = {
-	.name = TEAM_GENL_CHANGE_EVENT_MC_GRP_NAME,
+static const struct genl_multicast_group team_nl_mcgrps[] = {
+	{ .name = TEAM_GENL_CHANGE_EVENT_MC_GRP_NAME, },
 };
 
 static int team_nl_send_multicast(struct sk_buff *skb,
 				  struct team *team, u32 portid)
 {
-	return genlmsg_multicast_netns(dev_net(team->dev), skb, 0,
-				       team_change_event_mcgrp.id, GFP_KERNEL);
+	return genlmsg_multicast_netns(&team_nl_family, dev_net(team->dev),
+				       skb, 0, 0, GFP_KERNEL);
 }
 
 static int team_nl_send_event_options_get(struct team *team,
@@ -2697,23 +2703,8 @@ static int team_nl_send_event_port_get(struct team *team,
 
 static int team_nl_init(void)
 {
-	int err;
-
-	err = genl_register_family_with_ops(&team_nl_family, team_nl_ops,
-					    ARRAY_SIZE(team_nl_ops));
-	if (err)
-		return err;
-
-	err = genl_register_mc_group(&team_nl_family, &team_change_event_mcgrp);
-	if (err)
-		goto err_change_event_grp_reg;
-
-	return 0;
-
-err_change_event_grp_reg:
-	genl_unregister_family(&team_nl_family);
-
-	return err;
+	return genl_register_family_with_ops_groups(&team_nl_family, team_nl_ops,
+						    team_nl_mcgrps);
 }
 
 static void team_nl_fini(void)
diff --git a/drivers/net/team/team_mode_loadbalance.c b/drivers/net/team/team_mode_loadbalance.c
index 829a9cd2b4da..d671fc3ac5ac 100644
--- a/drivers/net/team/team_mode_loadbalance.c
+++ b/drivers/net/team/team_mode_loadbalance.c
@@ -570,7 +570,7 @@ static int lb_init(struct team *team)
 {
 	struct lb_priv *lb_priv = get_lb_priv(team);
 	lb_select_tx_port_func_t *func;
-	int err;
+	int i, err;
 
 	/* set default tx port selector */
 	func = lb_select_tx_port_get_func("hash");
@@ -588,6 +588,13 @@ static int lb_init(struct team *team)
 		goto err_alloc_pcpu_stats;
 	}
 
+	for_each_possible_cpu(i) {
+		struct lb_pcpu_stats *team_lb_stats;
+		team_lb_stats = per_cpu_ptr(lb_priv->pcpu_stats, i);
+		u64_stats_init(&team_lb_stats->syncp);
+	}
+
+
 	INIT_DELAYED_WORK(&lb_priv->ex->stats.refresh_dw, lb_stats_refresh);
 
 	err = team_options_register(team, lb_options, ARRAY_SIZE(lb_options));
diff --git a/drivers/net/tun.c b/drivers/net/tun.c
index 7cb105c103fe..782e38bfc1ee 100644
--- a/drivers/net/tun.c
+++ b/drivers/net/tun.c
@@ -981,6 +981,7 @@ static ssize_t tun_get_user(struct tun_struct *tun, struct tun_file *tfile,
 	struct sk_buff *skb;
 	size_t len = total_len, align = NET_SKB_PAD, linear;
 	struct virtio_net_hdr gso = { 0 };
+	int good_linear;
 	int offset = 0;
 	int copylen;
 	bool zerocopy = false;
@@ -1021,12 +1022,16 @@ static ssize_t tun_get_user(struct tun_struct *tun, struct tun_file *tfile,
 			return -EINVAL;
 	}
 
+	good_linear = SKB_MAX_HEAD(align);
+
 	if (msg_control) {
 		/* There are 256 bytes to be copied in skb, so there is
 		 * enough room for skb expand head in case it is used.
 		 * The rest of the buffer is mapped from userspace.
 		 */
 		copylen = gso.hdr_len ? gso.hdr_len : GOODCOPY_LEN;
+		if (copylen > good_linear)
+			copylen = good_linear;
 		linear = copylen;
 		if (iov_pages(iv, offset + copylen, count) <= MAX_SKB_FRAGS)
 			zerocopy = true;
@@ -1034,7 +1039,10 @@ static ssize_t tun_get_user(struct tun_struct *tun, struct tun_file *tfile,
 
 	if (!zerocopy) {
 		copylen = len;
-		linear = gso.hdr_len;
+		if (gso.hdr_len > good_linear)
+			linear = good_linear;
+		else
+			linear = gso.hdr_len;
 	}
 
 	skb = tun_alloc_skb(tfile, align, copylen, linear, noblock);
diff --git a/drivers/net/usb/cdc_ncm.c b/drivers/net/usb/cdc_ncm.c
index f74786aa37be..e15ec2b12035 100644
--- a/drivers/net/usb/cdc_ncm.c
+++ b/drivers/net/usb/cdc_ncm.c
@@ -66,7 +66,7 @@ static void cdc_ncm_tx_timeout_start(struct cdc_ncm_ctx *ctx);
 static enum hrtimer_restart cdc_ncm_tx_timer_cb(struct hrtimer *hr_timer);
 static struct usb_driver cdc_ncm_driver;
 
-static u8 cdc_ncm_setup(struct usbnet *dev)
+static int cdc_ncm_setup(struct usbnet *dev)
 {
 	struct cdc_ncm_ctx *ctx = (struct cdc_ncm_ctx *)dev->data[0];
 	struct usb_cdc_ncm_ntb_parameters ncm_parm;
diff --git a/drivers/net/usb/usbnet.c b/drivers/net/usb/usbnet.c
index 90a429b7ebad..8494bb53ebdc 100644
--- a/drivers/net/usb/usbnet.c
+++ b/drivers/net/usb/usbnet.c
@@ -204,9 +204,6 @@ static void intr_complete (struct urb *urb)
 		break;
 	}
 
-	if (!netif_running (dev->net))
-		return;
-
 	status = usb_submit_urb (urb, GFP_ATOMIC);
 	if (status != 0)
 		netif_err(dev, timer, dev->net,
diff --git a/drivers/net/veth.c b/drivers/net/veth.c
index b24db7acbf12..2ec2041b62d4 100644
--- a/drivers/net/veth.c
+++ b/drivers/net/veth.c
@@ -235,10 +235,18 @@ static int veth_change_mtu(struct net_device *dev, int new_mtu)
 
 static int veth_dev_init(struct net_device *dev)
 {
+	int i;
+
 	dev->vstats = alloc_percpu(struct pcpu_vstats);
 	if (!dev->vstats)
 		return -ENOMEM;
 
+	for_each_possible_cpu(i) {
+		struct pcpu_vstats *veth_stats;
+		veth_stats = per_cpu_ptr(dev->vstats, i);
+		u64_stats_init(&veth_stats->syncp);
+	}
+
 	return 0;
 }
 
diff --git a/drivers/net/virtio_net.c b/drivers/net/virtio_net.c
index 01f4eb5c8b78..7bab4de658a9 100644
--- a/drivers/net/virtio_net.c
+++ b/drivers/net/virtio_net.c
@@ -36,7 +36,10 @@ module_param(csum, bool, 0444);
 module_param(gso, bool, 0444);
 
 /* FIXME: MTU in config. */
-#define MAX_PACKET_LEN (ETH_HLEN + VLAN_HLEN + ETH_DATA_LEN)
+#define GOOD_PACKET_LEN (ETH_HLEN + VLAN_HLEN + ETH_DATA_LEN)
+#define MERGE_BUFFER_LEN (ALIGN(GOOD_PACKET_LEN + \
+                                sizeof(struct virtio_net_hdr_mrg_rxbuf), \
+                                L1_CACHE_BYTES))
 #define GOOD_COPY_LEN	128
 
 #define VIRTNET_DRIVER_VERSION "1.0.0"
@@ -314,10 +317,10 @@ static int receive_mergeable(struct receive_queue *rq, struct sk_buff *head_skb)
 			head_skb->dev->stats.rx_length_errors++;
 			return -EINVAL;
 		}
-		if (unlikely(len > MAX_PACKET_LEN)) {
+		if (unlikely(len > MERGE_BUFFER_LEN)) {
 			pr_debug("%s: rx error: merge buffer too long\n",
 				 head_skb->dev->name);
-			len = MAX_PACKET_LEN;
+			len = MERGE_BUFFER_LEN;
 		}
 		if (unlikely(num_skb_frags == MAX_SKB_FRAGS)) {
 			struct sk_buff *nskb = alloc_skb(0, GFP_ATOMIC);
@@ -336,18 +339,17 @@ static int receive_mergeable(struct receive_queue *rq, struct sk_buff *head_skb)
 		if (curr_skb != head_skb) {
 			head_skb->data_len += len;
 			head_skb->len += len;
-			head_skb->truesize += MAX_PACKET_LEN;
+			head_skb->truesize += MERGE_BUFFER_LEN;
 		}
 		page = virt_to_head_page(buf);
 		offset = buf - (char *)page_address(page);
 		if (skb_can_coalesce(curr_skb, num_skb_frags, page, offset)) {
 			put_page(page);
 			skb_coalesce_rx_frag(curr_skb, num_skb_frags - 1,
-					     len, MAX_PACKET_LEN);
+					     len, MERGE_BUFFER_LEN);
 		} else {
 			skb_add_rx_frag(curr_skb, num_skb_frags, page,
-					offset, len,
-					MAX_PACKET_LEN);
+					offset, len, MERGE_BUFFER_LEN);
 		}
 		--rq->num;
 	}
@@ -383,7 +385,7 @@ static void receive_buf(struct receive_queue *rq, void *buf, unsigned int len)
 		struct page *page = virt_to_head_page(buf);
 		skb = page_to_skb(rq, page,
 				  (char *)buf - (char *)page_address(page),
-				  len, MAX_PACKET_LEN);
+				  len, MERGE_BUFFER_LEN);
 		if (unlikely(!skb)) {
 			dev->stats.rx_dropped++;
 			put_page(page);
@@ -471,11 +473,11 @@ static int add_recvbuf_small(struct receive_queue *rq, gfp_t gfp)
 	struct skb_vnet_hdr *hdr;
 	int err;
 
-	skb = __netdev_alloc_skb_ip_align(vi->dev, MAX_PACKET_LEN, gfp);
+	skb = __netdev_alloc_skb_ip_align(vi->dev, GOOD_PACKET_LEN, gfp);
 	if (unlikely(!skb))
 		return -ENOMEM;
 
-	skb_put(skb, MAX_PACKET_LEN);
+	skb_put(skb, GOOD_PACKET_LEN);
 
 	hdr = skb_vnet_hdr(skb);
 	sg_set_buf(rq->sg, &hdr->hdr, sizeof hdr->hdr);
@@ -542,20 +544,20 @@ static int add_recvbuf_mergeable(struct receive_queue *rq, gfp_t gfp)
 	int err;
 
 	if (gfp & __GFP_WAIT) {
-		if (skb_page_frag_refill(MAX_PACKET_LEN, &vi->alloc_frag,
+		if (skb_page_frag_refill(MERGE_BUFFER_LEN, &vi->alloc_frag,
 					 gfp)) {
 			buf = (char *)page_address(vi->alloc_frag.page) +
 			      vi->alloc_frag.offset;
 			get_page(vi->alloc_frag.page);
-			vi->alloc_frag.offset += MAX_PACKET_LEN;
+			vi->alloc_frag.offset += MERGE_BUFFER_LEN;
 		}
 	} else {
-		buf = netdev_alloc_frag(MAX_PACKET_LEN);
+		buf = netdev_alloc_frag(MERGE_BUFFER_LEN);
 	}
 	if (!buf)
 		return -ENOMEM;
 
-	sg_init_one(rq->sg, buf, MAX_PACKET_LEN);
+	sg_init_one(rq->sg, buf, MERGE_BUFFER_LEN);
 	err = virtqueue_add_inbuf(rq->vq, rq->sg, 1, buf, gfp);
 	if (err < 0)
 		put_page(virt_to_head_page(buf));
@@ -591,7 +593,8 @@ static bool try_fill_recv(struct receive_queue *rq, gfp_t gfp)
 	} while (rq->vq->num_free);
 	if (unlikely(rq->num > rq->max))
 		rq->max = rq->num;
-	virtqueue_kick(rq->vq);
+	if (unlikely(!virtqueue_kick(rq->vq)))
+		return false;
 	return !oom;
 }
 
@@ -797,7 +800,7 @@ static netdev_tx_t start_xmit(struct sk_buff *skb, struct net_device *dev)
 	err = xmit_skb(sq, skb);
 
 	/* This should not happen! */
-	if (unlikely(err)) {
+	if (unlikely(err) || unlikely(!virtqueue_kick(sq->vq))) {
 		dev->stats.tx_fifo_errors++;
 		if (net_ratelimit())
 			dev_warn(&dev->dev,
@@ -806,7 +809,6 @@ static netdev_tx_t start_xmit(struct sk_buff *skb, struct net_device *dev)
 		kfree_skb(skb);
 		return NETDEV_TX_OK;
 	}
-	virtqueue_kick(sq->vq);
 
 	/* Don't wait up for transmitted skbs to be freed. */
 	skb_orphan(skb);
@@ -865,12 +867,14 @@ static bool virtnet_send_command(struct virtnet_info *vi, u8 class, u8 cmd,
 	BUG_ON(virtqueue_add_sgs(vi->cvq, sgs, out_num, in_num, vi, GFP_ATOMIC)
 	       < 0);
 
-	virtqueue_kick(vi->cvq);
+	if (unlikely(!virtqueue_kick(vi->cvq)))
+		return status == VIRTIO_NET_OK;
 
 	/* Spin for a response, the kick causes an ioport write, trapping
 	 * into the hypervisor, so the request should be handled immediately.
 	 */
-	while (!virtqueue_get_buf(vi->cvq, &tmp))
+	while (!virtqueue_get_buf(vi->cvq, &tmp) &&
+	       !virtqueue_is_broken(vi->cvq))
 		cpu_relax();
 
 	return status == VIRTIO_NET_OK;
@@ -898,8 +902,13 @@ static int virtnet_set_mac_address(struct net_device *dev, void *p)
 			return -EINVAL;
 		}
 	} else if (virtio_has_feature(vdev, VIRTIO_NET_F_MAC)) {
-		vdev->config->set(vdev, offsetof(struct virtio_net_config, mac),
-				  addr->sa_data, dev->addr_len);
+		unsigned int i;
+
+		/* Naturally, this has an atomicity problem. */
+		for (i = 0; i < dev->addr_len; i++)
+			virtio_cwrite8(vdev,
+				       offsetof(struct virtio_net_config, mac) +
+				       i, addr->sa_data[i]);
 	}
 
 	eth_commit_mac_addr_change(dev, p);
@@ -1281,9 +1290,8 @@ static void virtnet_config_changed_work(struct work_struct *work)
 	if (!vi->config_enable)
 		goto done;
 
-	if (virtio_config_val(vi->vdev, VIRTIO_NET_F_STATUS,
-			      offsetof(struct virtio_net_config, status),
-			      &v) < 0)
+	if (virtio_cread_feature(vi->vdev, VIRTIO_NET_F_STATUS,
+				 struct virtio_net_config, status, &v) < 0)
 		goto done;
 
 	if (v & VIRTIO_NET_S_ANNOUNCE) {
@@ -1507,9 +1515,9 @@ static int virtnet_probe(struct virtio_device *vdev)
 	u16 max_queue_pairs;
 
 	/* Find if host supports multiqueue virtio_net device */
-	err = virtio_config_val(vdev, VIRTIO_NET_F_MQ,
-				offsetof(struct virtio_net_config,
-				max_virtqueue_pairs), &max_queue_pairs);
+	err = virtio_cread_feature(vdev, VIRTIO_NET_F_MQ,
+				   struct virtio_net_config,
+				   max_virtqueue_pairs, &max_queue_pairs);
 
 	/* We need at least 2 queue's */
 	if (err || max_queue_pairs < VIRTIO_NET_CTRL_MQ_VQ_PAIRS_MIN ||
@@ -1561,9 +1569,11 @@ static int virtnet_probe(struct virtio_device *vdev)
 	dev->vlan_features = dev->features;
 
 	/* Configuration may specify what MAC to use.  Otherwise random. */
-	if (virtio_config_val_len(vdev, VIRTIO_NET_F_MAC,
-				  offsetof(struct virtio_net_config, mac),
-				  dev->dev_addr, dev->addr_len) < 0)
+	if (virtio_has_feature(vdev, VIRTIO_NET_F_MAC))
+		virtio_cread_bytes(vdev,
+				   offsetof(struct virtio_net_config, mac),
+				   dev->dev_addr, dev->addr_len);
+	else
 		eth_hw_addr_random(dev);
 
 	/* Set up our device-specific information */
@@ -1576,6 +1586,13 @@ static int virtnet_probe(struct virtio_device *vdev)
 	if (vi->stats == NULL)
 		goto free;
 
+	for_each_possible_cpu(i) {
+		struct virtnet_stats *virtnet_stats;
+		virtnet_stats = per_cpu_ptr(vi->stats, i);
+		u64_stats_init(&virtnet_stats->tx_syncp);
+		u64_stats_init(&virtnet_stats->rx_syncp);
+	}
+
 	mutex_init(&vi->config_lock);
 	vi->config_enable = true;
 	INIT_WORK(&vi->config_work, virtnet_config_changed_work);
@@ -1604,8 +1621,8 @@ static int virtnet_probe(struct virtio_device *vdev)
 	if (err)
 		goto free_stats;
 
-	netif_set_real_num_tx_queues(dev, 1);
-	netif_set_real_num_rx_queues(dev, 1);
+	netif_set_real_num_tx_queues(dev, vi->curr_queue_pairs);
+	netif_set_real_num_rx_queues(dev, vi->curr_queue_pairs);
 
 	err = register_netdev(dev);
 	if (err) {
@@ -1697,7 +1714,7 @@ static void virtnet_remove(struct virtio_device *vdev)
 	free_netdev(vi->dev);
 }
 
-#ifdef CONFIG_PM
+#ifdef CONFIG_PM_SLEEP
 static int virtnet_freeze(struct virtio_device *vdev)
 {
 	struct virtnet_info *vi = vdev->priv;
@@ -1788,7 +1805,7 @@ static struct virtio_driver virtio_net_driver = {
 	.probe =	virtnet_probe,
 	.remove =	virtnet_remove,
 	.config_changed = virtnet_config_changed,
-#ifdef CONFIG_PM
+#ifdef CONFIG_PM_SLEEP
 	.freeze =	virtnet_freeze,
 	.restore =	virtnet_restore,
 #endif
diff --git a/drivers/net/vxlan.c b/drivers/net/vxlan.c
index 78df8f39e57c..0358c07f7669 100644
--- a/drivers/net/vxlan.c
+++ b/drivers/net/vxlan.c
@@ -1880,11 +1880,19 @@ static int vxlan_init(struct net_device *dev)
 	struct vxlan_dev *vxlan = netdev_priv(dev);
 	struct vxlan_net *vn = net_generic(dev_net(dev), vxlan_net_id);
 	struct vxlan_sock *vs;
+	int i;
 
 	dev->tstats = alloc_percpu(struct pcpu_tstats);
 	if (!dev->tstats)
 		return -ENOMEM;
 
+	for_each_possible_cpu(i) {
+		struct pcpu_tstats *vxlan_stats;
+		vxlan_stats = per_cpu_ptr(dev->tstats, i);
+		u64_stats_init(&vxlan_stats->syncp);
+	}
+
+
 	spin_lock(&vn->sock_lock);
 	vs = vxlan_find_sock(dev_net(dev), vxlan->dst_port);
 	if (vs) {
diff --git a/drivers/net/wireless/ath/ath10k/htc.c b/drivers/net/wireless/ath/ath10k/htc.c
index 3118d7506734..edae50b52806 100644
--- a/drivers/net/wireless/ath/ath10k/htc.c
+++ b/drivers/net/wireless/ath/ath10k/htc.c
@@ -534,7 +534,7 @@ int ath10k_htc_wait_target(struct ath10k_htc *htc)
 	u16 credit_count;
 	u16 credit_size;
 
-	INIT_COMPLETION(htc->ctl_resp);
+	reinit_completion(&htc->ctl_resp);
 
 	status = ath10k_hif_start(htc->ar);
 	if (status) {
@@ -669,7 +669,7 @@ int ath10k_htc_connect_service(struct ath10k_htc *htc,
 	req_msg->flags = __cpu_to_le16(flags);
 	req_msg->service_id = __cpu_to_le16(conn_req->service_id);
 
-	INIT_COMPLETION(htc->ctl_resp);
+	reinit_completion(&htc->ctl_resp);
 
 	status = ath10k_htc_send(htc, ATH10K_HTC_EP_0, skb);
 	if (status) {
diff --git a/drivers/net/wireless/ath/ath10k/mac.c b/drivers/net/wireless/ath/ath10k/mac.c
index 0b1cc516e778..97ac8c87cba2 100644
--- a/drivers/net/wireless/ath/ath10k/mac.c
+++ b/drivers/net/wireless/ath/ath10k/mac.c
@@ -92,7 +92,7 @@ static int ath10k_install_key(struct ath10k_vif *arvif,
 
 	lockdep_assert_held(&ar->conf_mutex);
 
-	INIT_COMPLETION(ar->install_key_done);
+	reinit_completion(&ar->install_key_done);
 
 	ret = ath10k_send_key(arvif, key, cmd, macaddr);
 	if (ret)
@@ -438,7 +438,7 @@ static int ath10k_vdev_start(struct ath10k_vif *arvif)
 
 	lockdep_assert_held(&ar->conf_mutex);
 
-	INIT_COMPLETION(ar->vdev_setup_done);
+	reinit_completion(&ar->vdev_setup_done);
 
 	arg.vdev_id = arvif->vdev_id;
 	arg.dtim_period = arvif->dtim_period;
@@ -491,7 +491,7 @@ static int ath10k_vdev_stop(struct ath10k_vif *arvif)
 
 	lockdep_assert_held(&ar->conf_mutex);
 
-	INIT_COMPLETION(ar->vdev_setup_done);
+	reinit_completion(&ar->vdev_setup_done);
 
 	ret = ath10k_wmi_vdev_stop(ar, arvif->vdev_id);
 	if (ret) {
@@ -1666,7 +1666,7 @@ void ath10k_offchan_tx_work(struct work_struct *work)
 		}
 
 		spin_lock_bh(&ar->data_lock);
-		INIT_COMPLETION(ar->offchan_tx_completed);
+		reinit_completion(&ar->offchan_tx_completed);
 		ar->offchan_tx_skb = skb;
 		spin_unlock_bh(&ar->data_lock);
 
@@ -2476,8 +2476,8 @@ static int ath10k_hw_scan(struct ieee80211_hw *hw,
 		goto exit;
 	}
 
-	INIT_COMPLETION(ar->scan.started);
-	INIT_COMPLETION(ar->scan.completed);
+	reinit_completion(&ar->scan.started);
+	reinit_completion(&ar->scan.completed);
 	ar->scan.in_progress = true;
 	ar->scan.aborting = false;
 	ar->scan.is_roc = false;
@@ -2832,9 +2832,9 @@ static int ath10k_remain_on_channel(struct ieee80211_hw *hw,
 		goto exit;
 	}
 
-	INIT_COMPLETION(ar->scan.started);
-	INIT_COMPLETION(ar->scan.completed);
-	INIT_COMPLETION(ar->scan.on_channel);
+	reinit_completion(&ar->scan.started);
+	reinit_completion(&ar->scan.completed);
+	reinit_completion(&ar->scan.on_channel);
 	ar->scan.in_progress = true;
 	ar->scan.aborting = false;
 	ar->scan.is_roc = true;
diff --git a/drivers/net/wireless/ath/ath10k/pci.c b/drivers/net/wireless/ath/ath10k/pci.c
index f8d59c7b9082..9e86a811086f 100644
--- a/drivers/net/wireless/ath/ath10k/pci.c
+++ b/drivers/net/wireless/ath/ath10k/pci.c
@@ -2363,7 +2363,7 @@ static int ath10k_pci_probe(struct pci_dev *pdev,
 		break;
 	default:
 		ret = -ENODEV;
-		ath10k_err("Unkown device ID: %d\n", pci_dev->device);
+		ath10k_err("Unknown device ID: %d\n", pci_dev->device);
 		goto err_ar_pci;
 	}
 
diff --git a/drivers/net/wireless/ath/ath5k/dma.c b/drivers/net/wireless/ath/ath5k/dma.c
index ce86f158423b..ba200b24be64 100644
--- a/drivers/net/wireless/ath/ath5k/dma.c
+++ b/drivers/net/wireless/ath/ath5k/dma.c
@@ -661,7 +661,7 @@ ath5k_hw_get_isr(struct ath5k_hw *ah, enum ath5k_int *interrupt_mask)
 			ah->ah_txq_isr_txok_all |= AR5K_REG_MS(sisr1,
 						AR5K_SISR1_QCU_TXEOL);
 
-		/* Currently this is not much usefull since we treat
+		/* Currently this is not much useful since we treat
 		 * all queues the same way if we get a TXURN (update
 		 * tx trigger level) but we might need it later on*/
 		if (pisr & AR5K_ISR_TXURN)
diff --git a/drivers/net/wireless/ath/ath9k/ar9003_hw.c b/drivers/net/wireless/ath/ath9k/ar9003_hw.c
index b07f164d65cf..20e49095db2a 100644
--- a/drivers/net/wireless/ath/ath9k/ar9003_hw.c
+++ b/drivers/net/wireless/ath/ath9k/ar9003_hw.c
@@ -187,17 +187,17 @@ static void ar9003_hw_init_mode_regs(struct ath_hw *ah)
 		INIT_INI_ARRAY(&ah->iniCckfirJapan2484,
 			       ar9485_1_1_baseband_core_txfir_coeff_japan_2484);
 
-		/* Load PCIE SERDES settings from INI */
-
-		/* Awake Setting */
-
-		INIT_INI_ARRAY(&ah->iniPcieSerdes,
-				ar9485_1_1_pcie_phy_clkreq_disable_L1);
-
-		/* Sleep Setting */
-
-		INIT_INI_ARRAY(&ah->iniPcieSerdesLowPower,
-				ar9485_1_1_pcie_phy_clkreq_disable_L1);
+		if (ah->config.no_pll_pwrsave) {
+			INIT_INI_ARRAY(&ah->iniPcieSerdes,
+				       ar9485_1_1_pcie_phy_clkreq_disable_L1);
+			INIT_INI_ARRAY(&ah->iniPcieSerdesLowPower,
+				       ar9485_1_1_pcie_phy_clkreq_disable_L1);
+		} else {
+			INIT_INI_ARRAY(&ah->iniPcieSerdes,
+				       ar9485_1_1_pll_on_cdr_on_clkreq_disable_L1);
+			INIT_INI_ARRAY(&ah->iniPcieSerdesLowPower,
+				       ar9485_1_1_pll_on_cdr_on_clkreq_disable_L1);
+		}
 	} else if (AR_SREV_9462_21(ah)) {
 		INIT_INI_ARRAY(&ah->iniMac[ATH_INI_CORE],
 			       ar9462_2p1_mac_core);
diff --git a/drivers/net/wireless/ath/ath9k/ar9485_initvals.h b/drivers/net/wireless/ath/ath9k/ar9485_initvals.h
index 6f899c692647..7c1845221e1c 100644
--- a/drivers/net/wireless/ath/ath9k/ar9485_initvals.h
+++ b/drivers/net/wireless/ath/ath9k/ar9485_initvals.h
@@ -32,13 +32,6 @@ static const u32 ar9485_1_1_mac_postamble[][5] = {
 	{0x00008318, 0x00003e80, 0x00007d00, 0x00006880, 0x00003440},
 };
 
-static const u32 ar9485_1_1_pcie_phy_pll_on_clkreq_disable_L1[][2] = {
-	/* Addr      allmodes  */
-	{0x00018c00, 0x18012e5e},
-	{0x00018c04, 0x000801d8},
-	{0x00018c08, 0x0000080c},
-};
-
 static const u32 ar9485Common_wo_xlna_rx_gain_1_1[][2] = {
 	/* Addr      allmodes  */
 	{0x00009e00, 0x037216a0},
@@ -1101,20 +1094,6 @@ static const u32 ar9485_common_rx_gain_1_1[][2] = {
 	{0x0000a1fc, 0x00000296},
 };
 
-static const u32 ar9485_1_1_pcie_phy_pll_on_clkreq_enable_L1[][2] = {
-	/* Addr      allmodes  */
-	{0x00018c00, 0x18052e5e},
-	{0x00018c04, 0x000801d8},
-	{0x00018c08, 0x0000080c},
-};
-
-static const u32 ar9485_1_1_pcie_phy_clkreq_enable_L1[][2] = {
-	/* Addr      allmodes  */
-	{0x00018c00, 0x18053e5e},
-	{0x00018c04, 0x000801d8},
-	{0x00018c08, 0x0000080c},
-};
-
 static const u32 ar9485_1_1_soc_preamble[][2] = {
 	/* Addr      allmodes  */
 	{0x00004014, 0xba280400},
@@ -1173,13 +1152,6 @@ static const u32 ar9485_1_1_baseband_postamble[][5] = {
 	{0x0000be18, 0x00000000, 0x00000000, 0x00000000, 0x00000000},
 };
 
-static const u32 ar9485_1_1_pcie_phy_clkreq_disable_L1[][2] = {
-	/* Addr      allmodes  */
-	{0x00018c00, 0x18013e5e},
-	{0x00018c04, 0x000801d8},
-	{0x00018c08, 0x0000080c},
-};
-
 static const u32 ar9485_1_1_radio_postamble[][2] = {
 	/* Addr      allmodes  */
 	{0x0001609c, 0x0b283f31},
@@ -1358,4 +1330,18 @@ static const u32 ar9485_1_1_baseband_core_txfir_coeff_japan_2484[][2] = {
 	{0x0000a3a0, 0xca9228ee},
 };
 
+static const u32 ar9485_1_1_pcie_phy_clkreq_disable_L1[][2] = {
+	/* Addr      allmodes  */
+	{0x00018c00, 0x18013e5e},
+	{0x00018c04, 0x000801d8},
+	{0x00018c08, 0x0000080c},
+};
+
+static const u32 ar9485_1_1_pll_on_cdr_on_clkreq_disable_L1[][2] = {
+	/* Addr      allmodes  */
+	{0x00018c00, 0x1801265e},
+	{0x00018c04, 0x000801d8},
+	{0x00018c08, 0x0000080c},
+};
+
 #endif /* INITVALS_9485_H */
diff --git a/drivers/net/wireless/ath/ath9k/ath9k.h b/drivers/net/wireless/ath/ath9k/ath9k.h
index e7a38d844a6a..60a5da53668f 100644
--- a/drivers/net/wireless/ath/ath9k/ath9k.h
+++ b/drivers/net/wireless/ath/ath9k/ath9k.h
@@ -632,15 +632,16 @@ void ath_ant_comb_scan(struct ath_softc *sc, struct ath_rx_status *rs);
 /* Main driver core */
 /********************/
 
-#define ATH9K_PCI_CUS198      0x0001
-#define ATH9K_PCI_CUS230      0x0002
-#define ATH9K_PCI_CUS217      0x0004
-#define ATH9K_PCI_CUS252      0x0008
-#define ATH9K_PCI_WOW         0x0010
-#define ATH9K_PCI_BT_ANT_DIV  0x0020
-#define ATH9K_PCI_D3_L1_WAR   0x0040
-#define ATH9K_PCI_AR9565_1ANT 0x0080
-#define ATH9K_PCI_AR9565_2ANT 0x0100
+#define ATH9K_PCI_CUS198          0x0001
+#define ATH9K_PCI_CUS230          0x0002
+#define ATH9K_PCI_CUS217          0x0004
+#define ATH9K_PCI_CUS252          0x0008
+#define ATH9K_PCI_WOW             0x0010
+#define ATH9K_PCI_BT_ANT_DIV      0x0020
+#define ATH9K_PCI_D3_L1_WAR       0x0040
+#define ATH9K_PCI_AR9565_1ANT     0x0080
+#define ATH9K_PCI_AR9565_2ANT     0x0100
+#define ATH9K_PCI_NO_PLL_PWRSAVE  0x0200
 
 /*
  * Default cache line size, in bytes.
diff --git a/drivers/net/wireless/ath/ath9k/dfs_debug.c b/drivers/net/wireless/ath/ath9k/dfs_debug.c
index 90b8342d1ed4..8824610c21fb 100644
--- a/drivers/net/wireless/ath/ath9k/dfs_debug.c
+++ b/drivers/net/wireless/ath/ath9k/dfs_debug.c
@@ -44,14 +44,20 @@ static ssize_t read_file_dfs(struct file *file, char __user *user_buf,
 	if (buf == NULL)
 		return -ENOMEM;
 
-	if (sc->dfs_detector)
-		dfs_pool_stats = sc->dfs_detector->get_stats(sc->dfs_detector);
-
 	len += scnprintf(buf + len, size - len, "DFS support for "
 			 "macVersion = 0x%x, macRev = 0x%x: %s\n",
 			 hw_ver->macVersion, hw_ver->macRev,
 			 (sc->sc_ah->caps.hw_caps & ATH9K_HW_CAP_DFS) ?
 					"enabled" : "disabled");
+
+	if (!sc->dfs_detector) {
+		len += scnprintf(buf + len, size - len,
+				 "DFS detector not enabled\n");
+		goto exit;
+	}
+
+	dfs_pool_stats = sc->dfs_detector->get_stats(sc->dfs_detector);
+
 	len += scnprintf(buf + len, size - len, "Pulse detector statistics:\n");
 	ATH9K_DFS_STAT("pulse events reported   ", pulses_total);
 	ATH9K_DFS_STAT("invalid pulse events    ", pulses_no_dfs);
@@ -76,6 +82,7 @@ static ssize_t read_file_dfs(struct file *file, char __user *user_buf,
 	ATH9K_DFS_POOL_STAT("Seqs. alloc error       ", pseq_alloc_error);
 	ATH9K_DFS_POOL_STAT("Seqs. in use            ", pseq_used);
 
+exit:
 	if (len > size)
 		len = size;
 
diff --git a/drivers/net/wireless/ath/ath9k/hw.h b/drivers/net/wireless/ath/ath9k/hw.h
index 9ea24f1cba73..a2c9a5dbac6b 100644
--- a/drivers/net/wireless/ath/ath9k/hw.h
+++ b/drivers/net/wireless/ath/ath9k/hw.h
@@ -316,6 +316,7 @@ struct ath9k_ops_config {
 	u32 ant_ctrl_comm2g_switch_enable;
 	bool xatten_margin_cfg;
 	bool alt_mingainidx;
+	bool no_pll_pwrsave;
 };
 
 enum ath9k_int {
diff --git a/drivers/net/wireless/ath/ath9k/init.c b/drivers/net/wireless/ath/ath9k/init.c
index d8643ebabd30..710192ed27ed 100644
--- a/drivers/net/wireless/ath/ath9k/init.c
+++ b/drivers/net/wireless/ath/ath9k/init.c
@@ -609,6 +609,11 @@ static void ath9k_init_platform(struct ath_softc *sc)
 		ah->config.pcie_waen = 0x0040473b;
 		ath_info(common, "Enable WAR for ASPM D3/L1\n");
 	}
+
+	if (sc->driver_data & ATH9K_PCI_NO_PLL_PWRSAVE) {
+		ah->config.no_pll_pwrsave = true;
+		ath_info(common, "Disable PLL PowerSave\n");
+	}
 }
 
 static void ath9k_eeprom_request_cb(const struct firmware *eeprom_blob,
@@ -863,8 +868,8 @@ static const struct ieee80211_iface_combination if_comb[] = {
 		.max_interfaces = 1,
 		.num_different_channels = 1,
 		.beacon_int_infra_match = true,
-		.radar_detect_widths =	BIT(NL80211_CHAN_NO_HT) |
-					BIT(NL80211_CHAN_HT20),
+		.radar_detect_widths =	BIT(NL80211_CHAN_WIDTH_20_NOHT) |
+					BIT(NL80211_CHAN_WIDTH_20),
 	}
 };
 
diff --git a/drivers/net/wireless/ath/ath9k/pci.c b/drivers/net/wireless/ath/ath9k/pci.c
index 7e4c2524b630..b5656fce4ff5 100644
--- a/drivers/net/wireless/ath/ath9k/pci.c
+++ b/drivers/net/wireless/ath/ath9k/pci.c
@@ -195,6 +195,93 @@ static DEFINE_PCI_DEVICE_TABLE(ath_pci_id_table) = {
 			 0x3219),
 	  .driver_data = ATH9K_PCI_BT_ANT_DIV },
 
+	/* AR9485 cards with PLL power-save disabled by default. */
+	{ PCI_DEVICE_SUB(PCI_VENDOR_ID_ATHEROS,
+			 0x0032,
+			 PCI_VENDOR_ID_AZWAVE,
+			 0x2C97),
+	  .driver_data = ATH9K_PCI_NO_PLL_PWRSAVE },
+	{ PCI_DEVICE_SUB(PCI_VENDOR_ID_ATHEROS,
+			 0x0032,
+			 PCI_VENDOR_ID_AZWAVE,
+			 0x2100),
+	  .driver_data = ATH9K_PCI_NO_PLL_PWRSAVE },
+	{ PCI_DEVICE_SUB(PCI_VENDOR_ID_ATHEROS,
+			 0x0032,
+			 0x1C56, /* ASKEY */
+			 0x4001),
+	  .driver_data = ATH9K_PCI_NO_PLL_PWRSAVE },
+	{ PCI_DEVICE_SUB(PCI_VENDOR_ID_ATHEROS,
+			 0x0032,
+			 0x11AD, /* LITEON */
+			 0x6627),
+	  .driver_data = ATH9K_PCI_NO_PLL_PWRSAVE },
+	{ PCI_DEVICE_SUB(PCI_VENDOR_ID_ATHEROS,
+			 0x0032,
+			 0x11AD, /* LITEON */
+			 0x6628),
+	  .driver_data = ATH9K_PCI_NO_PLL_PWRSAVE },
+	{ PCI_DEVICE_SUB(PCI_VENDOR_ID_ATHEROS,
+			 0x0032,
+			 PCI_VENDOR_ID_FOXCONN,
+			 0xE04E),
+	  .driver_data = ATH9K_PCI_NO_PLL_PWRSAVE },
+	{ PCI_DEVICE_SUB(PCI_VENDOR_ID_ATHEROS,
+			 0x0032,
+			 PCI_VENDOR_ID_FOXCONN,
+			 0xE04F),
+	  .driver_data = ATH9K_PCI_NO_PLL_PWRSAVE },
+	{ PCI_DEVICE_SUB(PCI_VENDOR_ID_ATHEROS,
+			 0x0032,
+			 0x144F, /* ASKEY */
+			 0x7197),
+	  .driver_data = ATH9K_PCI_NO_PLL_PWRSAVE },
+	{ PCI_DEVICE_SUB(PCI_VENDOR_ID_ATHEROS,
+			 0x0032,
+			 0x1B9A, /* XAVI */
+			 0x2000),
+	  .driver_data = ATH9K_PCI_NO_PLL_PWRSAVE },
+	{ PCI_DEVICE_SUB(PCI_VENDOR_ID_ATHEROS,
+			 0x0032,
+			 0x1B9A, /* XAVI */
+			 0x2001),
+	  .driver_data = ATH9K_PCI_NO_PLL_PWRSAVE },
+	{ PCI_DEVICE_SUB(PCI_VENDOR_ID_ATHEROS,
+			 0x0032,
+			 PCI_VENDOR_ID_AZWAVE,
+			 0x1186),
+	  .driver_data = ATH9K_PCI_NO_PLL_PWRSAVE },
+	{ PCI_DEVICE_SUB(PCI_VENDOR_ID_ATHEROS,
+			 0x0032,
+			 PCI_VENDOR_ID_AZWAVE,
+			 0x1F86),
+	  .driver_data = ATH9K_PCI_NO_PLL_PWRSAVE },
+	{ PCI_DEVICE_SUB(PCI_VENDOR_ID_ATHEROS,
+			 0x0032,
+			 PCI_VENDOR_ID_AZWAVE,
+			 0x1195),
+	  .driver_data = ATH9K_PCI_NO_PLL_PWRSAVE },
+	{ PCI_DEVICE_SUB(PCI_VENDOR_ID_ATHEROS,
+			 0x0032,
+			 PCI_VENDOR_ID_AZWAVE,
+			 0x1F95),
+	  .driver_data = ATH9K_PCI_NO_PLL_PWRSAVE },
+	{ PCI_DEVICE_SUB(PCI_VENDOR_ID_ATHEROS,
+			 0x0032,
+			 0x1B9A, /* XAVI */
+			 0x1C00),
+	  .driver_data = ATH9K_PCI_NO_PLL_PWRSAVE },
+	{ PCI_DEVICE_SUB(PCI_VENDOR_ID_ATHEROS,
+			 0x0032,
+			 0x1B9A, /* XAVI */
+			 0x1C01),
+	  .driver_data = ATH9K_PCI_NO_PLL_PWRSAVE },
+	{ PCI_DEVICE_SUB(PCI_VENDOR_ID_ATHEROS,
+			 0x0032,
+			 PCI_VENDOR_ID_ASUSTEK,
+			 0x850D),
+	  .driver_data = ATH9K_PCI_NO_PLL_PWRSAVE },
+
 	{ PCI_VDEVICE(ATHEROS, 0x0032) }, /* PCI-E  AR9485 */
 	{ PCI_VDEVICE(ATHEROS, 0x0033) }, /* PCI-E  AR9580 */
 
diff --git a/drivers/net/wireless/ath/carl9170/usb.c b/drivers/net/wireless/ath/carl9170/usb.c
index 307bc0ddff99..ca115f33746f 100644
--- a/drivers/net/wireless/ath/carl9170/usb.c
+++ b/drivers/net/wireless/ath/carl9170/usb.c
@@ -773,7 +773,7 @@ void carl9170_usb_stop(struct ar9170 *ar)
 	complete_all(&ar->cmd_wait);
 
 	/* This is required to prevent an early completion on _start */
-	INIT_COMPLETION(ar->cmd_wait);
+	reinit_completion(&ar->cmd_wait);
 
 	/*
 	 * Note:
diff --git a/drivers/net/wireless/ath/wcn36xx/debug.c b/drivers/net/wireless/ath/wcn36xx/debug.c
index 5b84f7ae0b1e..ef44a2da644d 100644
--- a/drivers/net/wireless/ath/wcn36xx/debug.c
+++ b/drivers/net/wireless/ath/wcn36xx/debug.c
@@ -126,7 +126,7 @@ static ssize_t write_file_dump(struct file *file,
 		if (begin == NULL)
 			break;
 
-		if (kstrtoul(begin, 0, (unsigned long *)(arg + i)) != 0)
+		if (kstrtou32(begin, 0, &arg[i]) != 0)
 			break;
 	}
 
diff --git a/drivers/net/wireless/ath/wcn36xx/smd.c b/drivers/net/wireless/ath/wcn36xx/smd.c
index f8c3a10510c2..de9eb2cfbf4b 100644
--- a/drivers/net/wireless/ath/wcn36xx/smd.c
+++ b/drivers/net/wireless/ath/wcn36xx/smd.c
@@ -1286,7 +1286,8 @@ int wcn36xx_smd_send_beacon(struct wcn36xx *wcn, struct ieee80211_vif *vif,
 	} else {
 		wcn36xx_err("Beacon is to big: beacon size=%d\n",
 			      msg_body.beacon_length);
-		return -ENOMEM;
+		ret = -ENOMEM;
+		goto out;
 	}
 	memcpy(msg_body.bssid, vif->addr, ETH_ALEN);
 
@@ -1327,7 +1328,8 @@ int wcn36xx_smd_update_proberesp_tmpl(struct wcn36xx *wcn,
 	if (skb->len > BEACON_TEMPLATE_SIZE) {
 		wcn36xx_warn("probe response template is too big: %d\n",
 			     skb->len);
-		return -E2BIG;
+		ret = -E2BIG;
+		goto out;
 	}
 
 	msg.probe_resp_template_len = skb->len;
@@ -1606,7 +1608,8 @@ int wcn36xx_smd_keep_alive_req(struct wcn36xx *wcn,
 		/* TODO: it also support ARP response type */
 	} else {
 		wcn36xx_warn("unknow keep alive packet type %d\n", packet_type);
-		return -EINVAL;
+		ret = -EINVAL;
+		goto out;
 	}
 
 	PREPARE_HAL_BUF(wcn->hal_buf, msg_body);
diff --git a/drivers/net/wireless/ath/wil6210/main.c b/drivers/net/wireless/ath/wil6210/main.c
index 0a2844c48a60..fd30cddd5882 100644
--- a/drivers/net/wireless/ath/wil6210/main.c
+++ b/drivers/net/wireless/ath/wil6210/main.c
@@ -250,7 +250,7 @@ int wil_reset(struct wil6210_priv *wil)
 
 	/* init after reset */
 	wil->pending_connect_cid = -1;
-	INIT_COMPLETION(wil->wmi_ready);
+	reinit_completion(&wil->wmi_ready);
 
 	/* TODO: release MAC reset */
 	wil6210_enable_irq(wil);
diff --git a/drivers/net/wireless/brcm80211/brcmfmac/p2p.c b/drivers/net/wireless/brcm80211/brcmfmac/p2p.c
index d7a974532909..5b5b952d47b1 100644
--- a/drivers/net/wireless/brcm80211/brcmfmac/p2p.c
+++ b/drivers/net/wireless/brcm80211/brcmfmac/p2p.c
@@ -1148,7 +1148,7 @@ static s32 brcmf_p2p_af_searching_channel(struct brcmf_p2p_info *p2p)
 
 	pri_vif = p2p->bss_idx[P2PAPI_BSSCFG_PRIMARY].vif;
 
-	INIT_COMPLETION(afx_hdl->act_frm_scan);
+	reinit_completion(&afx_hdl->act_frm_scan);
 	set_bit(BRCMF_P2P_STATUS_FINDING_COMMON_CHANNEL, &p2p->status);
 	afx_hdl->is_active = true;
 	afx_hdl->peer_chan = P2P_INVALID_CHANNEL;
@@ -1501,7 +1501,7 @@ static s32 brcmf_p2p_tx_action_frame(struct brcmf_p2p_info *p2p,
 
 	brcmf_dbg(TRACE, "Enter\n");
 
-	INIT_COMPLETION(p2p->send_af_done);
+	reinit_completion(&p2p->send_af_done);
 	clear_bit(BRCMF_P2P_STATUS_ACTION_TX_COMPLETED, &p2p->status);
 	clear_bit(BRCMF_P2P_STATUS_ACTION_TX_NOACK, &p2p->status);
 
diff --git a/drivers/net/wireless/libertas/debugfs.c b/drivers/net/wireless/libertas/debugfs.c
index 668dd27616a0..cc6a0a586f0b 100644
--- a/drivers/net/wireless/libertas/debugfs.c
+++ b/drivers/net/wireless/libertas/debugfs.c
@@ -913,7 +913,10 @@ static ssize_t lbs_debugfs_write(struct file *f, const char __user *buf,
 	char *p2;
 	struct debug_data *d = f->private_data;
 
-	pdata = kmalloc(cnt, GFP_KERNEL);
+	if (cnt == 0)
+		return 0;
+
+	pdata = kmalloc(cnt + 1, GFP_KERNEL);
 	if (pdata == NULL)
 		return 0;
 
@@ -922,6 +925,7 @@ static ssize_t lbs_debugfs_write(struct file *f, const char __user *buf,
 		kfree(pdata);
 		return 0;
 	}
+	pdata[cnt] = '\0';
 
 	p0 = pdata;
 	for (i = 0; i < num_of_items; i++) {
diff --git a/drivers/net/wireless/libertas/if_cs.c b/drivers/net/wireless/libertas/if_cs.c
index ef8c98e21098..f499efc6abcf 100644
--- a/drivers/net/wireless/libertas/if_cs.c
+++ b/drivers/net/wireless/libertas/if_cs.c
@@ -902,6 +902,7 @@ static int if_cs_probe(struct pcmcia_device *p_dev)
 	if (card->model == MODEL_UNKNOWN) {
 		pr_err("unsupported manf_id 0x%04x / card_id 0x%04x\n",
 		       p_dev->manf_id, p_dev->card_id);
+		ret = -ENODEV;
 		goto out2;
 	}
 
diff --git a/drivers/net/wireless/mac80211_hwsim.c b/drivers/net/wireless/mac80211_hwsim.c
index de0df86704e7..9df7bc91a26f 100644
--- a/drivers/net/wireless/mac80211_hwsim.c
+++ b/drivers/net/wireless/mac80211_hwsim.c
@@ -2097,7 +2097,7 @@ out:
 }
 
 /* Generic Netlink operations array */
-static struct genl_ops hwsim_ops[] = {
+static const struct genl_ops hwsim_ops[] = {
 	{
 		.cmd = HWSIM_CMD_REGISTER,
 		.policy = hwsim_genl_policy,
@@ -2148,8 +2148,7 @@ static int hwsim_init_netlink(void)
 
 	printk(KERN_INFO "mac80211_hwsim: initializing netlink\n");
 
-	rc = genl_register_family_with_ops(&hwsim_genl_family,
-		hwsim_ops, ARRAY_SIZE(hwsim_ops));
+	rc = genl_register_family_with_ops(&hwsim_genl_family, hwsim_ops);
 	if (rc)
 		goto failure;
 
diff --git a/drivers/net/wireless/mwifiex/fw.h b/drivers/net/wireless/mwifiex/fw.h
index f80f30b6160e..c8385ec77a86 100644
--- a/drivers/net/wireless/mwifiex/fw.h
+++ b/drivers/net/wireless/mwifiex/fw.h
@@ -1020,8 +1020,8 @@ struct mwifiex_power_group {
 } __packed;
 
 struct mwifiex_types_power_group {
-	u16 type;
-	u16 length;
+	__le16 type;
+	__le16 length;
 } __packed;
 
 struct host_cmd_ds_txpwr_cfg {
diff --git a/drivers/net/wireless/mwifiex/ie.c b/drivers/net/wireless/mwifiex/ie.c
index 220af4fe0fc6..81ac001ee741 100644
--- a/drivers/net/wireless/mwifiex/ie.c
+++ b/drivers/net/wireless/mwifiex/ie.c
@@ -82,7 +82,7 @@ mwifiex_update_autoindex_ies(struct mwifiex_private *priv,
 			     struct mwifiex_ie_list *ie_list)
 {
 	u16 travel_len, index, mask;
-	s16 input_len;
+	s16 input_len, tlv_len;
 	struct mwifiex_ie *ie;
 	u8 *tmp;
 
@@ -91,11 +91,13 @@ mwifiex_update_autoindex_ies(struct mwifiex_private *priv,
 
 	ie_list->len = 0;
 
-	while (input_len > 0) {
+	while (input_len >= sizeof(struct mwifiex_ie_types_header)) {
 		ie = (struct mwifiex_ie *)(((u8 *)ie_list) + travel_len);
-		input_len -= le16_to_cpu(ie->ie_length) + MWIFIEX_IE_HDR_SIZE;
-		travel_len += le16_to_cpu(ie->ie_length) + MWIFIEX_IE_HDR_SIZE;
+		tlv_len = le16_to_cpu(ie->ie_length);
+		travel_len += tlv_len + MWIFIEX_IE_HDR_SIZE;
 
+		if (input_len < tlv_len + MWIFIEX_IE_HDR_SIZE)
+			return -1;
 		index = le16_to_cpu(ie->ie_index);
 		mask = le16_to_cpu(ie->mgmt_subtype_mask);
 
@@ -132,6 +134,7 @@ mwifiex_update_autoindex_ies(struct mwifiex_private *priv,
 		le16_add_cpu(&ie_list->len,
 			     le16_to_cpu(priv->mgmt_ie[index].ie_length) +
 			     MWIFIEX_IE_HDR_SIZE);
+		input_len -= tlv_len + MWIFIEX_IE_HDR_SIZE;
 	}
 
 	if (GET_BSS_ROLE(priv) == MWIFIEX_BSS_ROLE_UAP)
diff --git a/drivers/net/wireless/mwifiex/sdio.c b/drivers/net/wireless/mwifiex/sdio.c
index 1576104e3d95..9bf8898743ab 100644
--- a/drivers/net/wireless/mwifiex/sdio.c
+++ b/drivers/net/wireless/mwifiex/sdio.c
@@ -1029,7 +1029,10 @@ static int mwifiex_decode_rx_packet(struct mwifiex_adapter *adapter,
 				    struct sk_buff *skb, u32 upld_typ)
 {
 	u8 *cmd_buf;
+	__le16 *curr_ptr = (__le16 *)skb->data;
+	u16 pkt_len = le16_to_cpu(*curr_ptr);
 
+	skb_trim(skb, pkt_len);
 	skb_pull(skb, INTF_HEADER_LEN);
 
 	switch (upld_typ) {
diff --git a/drivers/net/wireless/mwifiex/sta_cmd.c b/drivers/net/wireless/mwifiex/sta_cmd.c
index 7d66018a2e33..2181ee283d82 100644
--- a/drivers/net/wireless/mwifiex/sta_cmd.c
+++ b/drivers/net/wireless/mwifiex/sta_cmd.c
@@ -239,14 +239,14 @@ static int mwifiex_cmd_tx_power_cfg(struct host_cmd_ds_command *cmd,
 			memmove(cmd_txp_cfg, txp,
 				sizeof(struct host_cmd_ds_txpwr_cfg) +
 				sizeof(struct mwifiex_types_power_group) +
-				pg_tlv->length);
+				le16_to_cpu(pg_tlv->length));
 
 			pg_tlv = (struct mwifiex_types_power_group *) ((u8 *)
 				  cmd_txp_cfg +
 				  sizeof(struct host_cmd_ds_txpwr_cfg));
 			cmd->size = cpu_to_le16(le16_to_cpu(cmd->size) +
 				  sizeof(struct mwifiex_types_power_group) +
-				  pg_tlv->length);
+				  le16_to_cpu(pg_tlv->length));
 		} else {
 			memmove(cmd_txp_cfg, txp, sizeof(*txp));
 		}
diff --git a/drivers/net/wireless/mwifiex/sta_cmdresp.c b/drivers/net/wireless/mwifiex/sta_cmdresp.c
index 58a6013712d2..2675ca7f8d14 100644
--- a/drivers/net/wireless/mwifiex/sta_cmdresp.c
+++ b/drivers/net/wireless/mwifiex/sta_cmdresp.c
@@ -274,17 +274,20 @@ static int mwifiex_ret_tx_rate_cfg(struct mwifiex_private *priv,
 	struct host_cmd_ds_tx_rate_cfg *rate_cfg = &resp->params.tx_rate_cfg;
 	struct mwifiex_rate_scope *rate_scope;
 	struct mwifiex_ie_types_header *head;
-	u16 tlv, tlv_buf_len;
+	u16 tlv, tlv_buf_len, tlv_buf_left;
 	u8 *tlv_buf;
 	u32 i;
 
-	tlv_buf = ((u8 *)rate_cfg) +
-			sizeof(struct host_cmd_ds_tx_rate_cfg);
-	tlv_buf_len = le16_to_cpu(*(__le16 *) (tlv_buf + sizeof(u16)));
+	tlv_buf = ((u8 *)rate_cfg) + sizeof(struct host_cmd_ds_tx_rate_cfg);
+	tlv_buf_left = le16_to_cpu(resp->size) - S_DS_GEN - sizeof(*rate_cfg);
 
-	while (tlv_buf && tlv_buf_len > 0) {
-		tlv = (*tlv_buf);
-		tlv = tlv | (*(tlv_buf + 1) << 8);
+	while (tlv_buf_left >= sizeof(*head)) {
+		head = (struct mwifiex_ie_types_header *)tlv_buf;
+		tlv = le16_to_cpu(head->type);
+		tlv_buf_len = le16_to_cpu(head->len);
+
+		if (tlv_buf_left < (sizeof(*head) + tlv_buf_len))
+			break;
 
 		switch (tlv) {
 		case TLV_TYPE_RATE_SCOPE:
@@ -304,9 +307,8 @@ static int mwifiex_ret_tx_rate_cfg(struct mwifiex_private *priv,
 			/* Add RATE_DROP tlv here */
 		}
 
-		head = (struct mwifiex_ie_types_header *) tlv_buf;
-		tlv_buf += le16_to_cpu(head->len) + sizeof(*head);
-		tlv_buf_len -= le16_to_cpu(head->len);
+		tlv_buf += (sizeof(*head) + tlv_buf_len);
+		tlv_buf_left -= (sizeof(*head) + tlv_buf_len);
 	}
 
 	priv->is_data_rate_auto = mwifiex_is_rate_auto(priv);
@@ -340,13 +342,17 @@ static int mwifiex_get_power_level(struct mwifiex_private *priv, void *data_buf)
 		((u8 *) data_buf + sizeof(struct host_cmd_ds_txpwr_cfg));
 	pg = (struct mwifiex_power_group *)
 		((u8 *) pg_tlv_hdr + sizeof(struct mwifiex_types_power_group));
-	length = pg_tlv_hdr->length;
-	if (length > 0) {
-		max_power = pg->power_max;
-		min_power = pg->power_min;
-		length -= sizeof(struct mwifiex_power_group);
-	}
-	while (length) {
+	length = le16_to_cpu(pg_tlv_hdr->length);
+
+	/* At least one structure required to update power */
+	if (length < sizeof(struct mwifiex_power_group))
+		return 0;
+
+	max_power = pg->power_max;
+	min_power = pg->power_min;
+	length -= sizeof(struct mwifiex_power_group);
+
+	while (length >= sizeof(struct mwifiex_power_group)) {
 		pg++;
 		if (max_power < pg->power_max)
 			max_power = pg->power_max;
@@ -356,10 +362,8 @@ static int mwifiex_get_power_level(struct mwifiex_private *priv, void *data_buf)
 
 		length -= sizeof(struct mwifiex_power_group);
 	}
-	if (pg_tlv_hdr->length > 0) {
-		priv->min_tx_power_level = (u8) min_power;
-		priv->max_tx_power_level = (u8) max_power;
-	}
+	priv->min_tx_power_level = (u8) min_power;
+	priv->max_tx_power_level = (u8) max_power;
 
 	return 0;
 }
diff --git a/drivers/net/wireless/mwifiex/sta_ioctl.c b/drivers/net/wireless/mwifiex/sta_ioctl.c
index f084412eee0b..c8e029df770e 100644
--- a/drivers/net/wireless/mwifiex/sta_ioctl.c
+++ b/drivers/net/wireless/mwifiex/sta_ioctl.c
@@ -638,8 +638,9 @@ int mwifiex_set_tx_power(struct mwifiex_private *priv,
 		txp_cfg->mode = cpu_to_le32(1);
 		pg_tlv = (struct mwifiex_types_power_group *)
 			 (buf + sizeof(struct host_cmd_ds_txpwr_cfg));
-		pg_tlv->type = TLV_TYPE_POWER_GROUP;
-		pg_tlv->length = 4 * sizeof(struct mwifiex_power_group);
+		pg_tlv->type = cpu_to_le16(TLV_TYPE_POWER_GROUP);
+		pg_tlv->length =
+			cpu_to_le16(4 * sizeof(struct mwifiex_power_group));
 		pg = (struct mwifiex_power_group *)
 		     (buf + sizeof(struct host_cmd_ds_txpwr_cfg)
 		      + sizeof(struct mwifiex_types_power_group));
diff --git a/drivers/net/wireless/mwifiex/uap_txrx.c b/drivers/net/wireless/mwifiex/uap_txrx.c
index 1cfe5a738c47..92f76d655e6c 100644
--- a/drivers/net/wireless/mwifiex/uap_txrx.c
+++ b/drivers/net/wireless/mwifiex/uap_txrx.c
@@ -97,6 +97,7 @@ static void mwifiex_uap_queue_bridged_pkt(struct mwifiex_private *priv,
 	struct mwifiex_txinfo *tx_info;
 	int hdr_chop;
 	struct timeval tv;
+	struct ethhdr *p_ethhdr;
 	u8 rfc1042_eth_hdr[ETH_ALEN] = { 0xaa, 0xaa, 0x03, 0x00, 0x00, 0x00 };
 
 	uap_rx_pd = (struct uap_rxpd *)(skb->data);
@@ -112,14 +113,36 @@ static void mwifiex_uap_queue_bridged_pkt(struct mwifiex_private *priv,
 	}
 
 	if (!memcmp(&rx_pkt_hdr->rfc1042_hdr,
-		    rfc1042_eth_hdr, sizeof(rfc1042_eth_hdr)))
+		    rfc1042_eth_hdr, sizeof(rfc1042_eth_hdr))) {
+		/* Replace the 803 header and rfc1042 header (llc/snap) with
+		 * an Ethernet II header, keep the src/dst and snap_type
+		 * (ethertype).
+		 *
+		 * The firmware only passes up SNAP frames converting all RX
+		 * data from 802.11 to 802.2/LLC/SNAP frames.
+		 *
+		 * To create the Ethernet II, just move the src, dst address
+		 * right before the snap_type.
+		 */
+		p_ethhdr = (struct ethhdr *)
+			((u8 *)(&rx_pkt_hdr->eth803_hdr)
+			 + sizeof(rx_pkt_hdr->eth803_hdr)
+			 + sizeof(rx_pkt_hdr->rfc1042_hdr)
+			 - sizeof(rx_pkt_hdr->eth803_hdr.h_dest)
+			 - sizeof(rx_pkt_hdr->eth803_hdr.h_source)
+			 - sizeof(rx_pkt_hdr->rfc1042_hdr.snap_type));
+		memcpy(p_ethhdr->h_source, rx_pkt_hdr->eth803_hdr.h_source,
+		       sizeof(p_ethhdr->h_source));
+		memcpy(p_ethhdr->h_dest, rx_pkt_hdr->eth803_hdr.h_dest,
+		       sizeof(p_ethhdr->h_dest));
 		/* Chop off the rxpd + the excess memory from
 		 * 802.2/llc/snap header that was removed.
 		 */
-		hdr_chop = (u8 *)eth_hdr - (u8 *)uap_rx_pd;
-	else
+		hdr_chop = (u8 *)p_ethhdr - (u8 *)uap_rx_pd;
+	} else {
 		/* Chop off the rxpd */
 		hdr_chop = (u8 *)&rx_pkt_hdr->eth803_hdr - (u8 *)uap_rx_pd;
+	}
 
 	/* Chop off the leading header bytes so the it points
 	 * to the start of either the reconstructed EthII frame
diff --git a/drivers/net/wireless/mwifiex/wmm.c b/drivers/net/wireless/mwifiex/wmm.c
index 5dd0ccc70b86..13eaeed03898 100644
--- a/drivers/net/wireless/mwifiex/wmm.c
+++ b/drivers/net/wireless/mwifiex/wmm.c
@@ -722,6 +722,9 @@ int mwifiex_ret_wmm_get_status(struct mwifiex_private *priv,
 		tlv_hdr = (struct mwifiex_ie_types_data *) curr;
 		tlv_len = le16_to_cpu(tlv_hdr->header.len);
 
+		if (resp_len < tlv_len + sizeof(tlv_hdr->header))
+			break;
+
 		switch (le16_to_cpu(tlv_hdr->header.type)) {
 		case TLV_TYPE_WMMQSTATUS:
 			tlv_wmm_qstatus =
diff --git a/drivers/net/wireless/prism54/islpci_dev.c b/drivers/net/wireless/prism54/islpci_dev.c
index 41a16d30c79c..e05d9b4c8317 100644
--- a/drivers/net/wireless/prism54/islpci_dev.c
+++ b/drivers/net/wireless/prism54/islpci_dev.c
@@ -811,6 +811,10 @@ static const struct net_device_ops islpci_netdev_ops = {
 	.ndo_validate_addr	= eth_validate_addr,
 };
 
+static struct device_type wlan_type = {
+	.name	= "wlan",
+};
+
 struct net_device *
 islpci_setup(struct pci_dev *pdev)
 {
@@ -821,9 +825,8 @@ islpci_setup(struct pci_dev *pdev)
 		return ndev;
 
 	pci_set_drvdata(pdev, ndev);
-#if defined(SET_NETDEV_DEV)
 	SET_NETDEV_DEV(ndev, &pdev->dev);
-#endif
+	SET_NETDEV_DEVTYPE(ndev, &wlan_type);
 
 	/* setup the structure members */
 	ndev->base_addr = pci_resource_start(pdev, 0);
diff --git a/drivers/net/wireless/rt2x00/rt2800lib.c b/drivers/net/wireless/rt2x00/rt2800lib.c
index c5738f14c4ba..776aff3678ff 100644
--- a/drivers/net/wireless/rt2x00/rt2800lib.c
+++ b/drivers/net/wireless/rt2x00/rt2800lib.c
@@ -2640,7 +2640,7 @@ static void rt2800_config_channel_rf53xx(struct rt2x00_dev *rt2x00dev,
 
 	if (rt2x00_rt(rt2x00dev, RT5392)) {
 		rt2800_rfcsr_read(rt2x00dev, 50, &rfcsr);
-		if (info->default_power1 > POWER_BOUND)
+		if (info->default_power2 > POWER_BOUND)
 			rt2x00_set_field8(&rfcsr, RFCSR50_TX, POWER_BOUND);
 		else
 			rt2x00_set_field8(&rfcsr, RFCSR50_TX,
diff --git a/drivers/net/wireless/rt2x00/rt2800mmio.c b/drivers/net/wireless/rt2x00/rt2800mmio.c
index ae152280e071..a8cc736b5063 100644
--- a/drivers/net/wireless/rt2x00/rt2800mmio.c
+++ b/drivers/net/wireless/rt2x00/rt2800mmio.c
@@ -446,7 +446,7 @@ static void rt2800mmio_txstatus_interrupt(struct rt2x00_dev *rt2x00dev)
 		if (!rt2x00_get_field32(status, TX_STA_FIFO_VALID))
 			break;
 
-		if (!kfifo_put(&rt2x00dev->txstatus_fifo, &status)) {
+		if (!kfifo_put(&rt2x00dev->txstatus_fifo, status)) {
 			rt2x00_warn(rt2x00dev, "TX status FIFO overrun, drop tx status report\n");
 			break;
 		}
diff --git a/drivers/net/wireless/rt2x00/rt2800usb.c b/drivers/net/wireless/rt2x00/rt2800usb.c
index 997df03a0c2e..a81ceb61d746 100644
--- a/drivers/net/wireless/rt2x00/rt2800usb.c
+++ b/drivers/net/wireless/rt2x00/rt2800usb.c
@@ -164,7 +164,7 @@ static bool rt2800usb_tx_sta_fifo_read_completed(struct rt2x00_dev *rt2x00dev,
 
 	valid = rt2x00_get_field32(tx_status, TX_STA_FIFO_VALID);
 	if (valid) {
-		if (!kfifo_put(&rt2x00dev->txstatus_fifo, &tx_status))
+		if (!kfifo_put(&rt2x00dev->txstatus_fifo, tx_status))
 			rt2x00_warn(rt2x00dev, "TX status FIFO overrun\n");
 
 		queue_work(rt2x00dev->workqueue, &rt2x00dev->txdone_work);
diff --git a/drivers/net/wireless/rt2x00/rt2x00lib.h b/drivers/net/wireless/rt2x00/rt2x00lib.h
index a0935987fa3a..7f40ab8e1bd8 100644
--- a/drivers/net/wireless/rt2x00/rt2x00lib.h
+++ b/drivers/net/wireless/rt2x00/rt2x00lib.h
@@ -146,7 +146,7 @@ void rt2x00queue_remove_l2pad(struct sk_buff *skb, unsigned int header_length);
  * @local: frame is not from mac80211
  */
 int rt2x00queue_write_tx_frame(struct data_queue *queue, struct sk_buff *skb,
-			       bool local);
+			       struct ieee80211_sta *sta, bool local);
 
 /**
  * rt2x00queue_update_beacon - Send new beacon from mac80211
diff --git a/drivers/net/wireless/rt2x00/rt2x00mac.c b/drivers/net/wireless/rt2x00/rt2x00mac.c
index 7c157857f5ce..2183e7978399 100644
--- a/drivers/net/wireless/rt2x00/rt2x00mac.c
+++ b/drivers/net/wireless/rt2x00/rt2x00mac.c
@@ -90,7 +90,7 @@ static int rt2x00mac_tx_rts_cts(struct rt2x00_dev *rt2x00dev,
 				  frag_skb->data, data_length, tx_info,
 				  (struct ieee80211_rts *)(skb->data));
 
-	retval = rt2x00queue_write_tx_frame(queue, skb, true);
+	retval = rt2x00queue_write_tx_frame(queue, skb, NULL, true);
 	if (retval) {
 		dev_kfree_skb_any(skb);
 		rt2x00_warn(rt2x00dev, "Failed to send RTS/CTS frame\n");
@@ -151,7 +151,7 @@ void rt2x00mac_tx(struct ieee80211_hw *hw,
 			goto exit_fail;
 	}
 
-	if (unlikely(rt2x00queue_write_tx_frame(queue, skb, false)))
+	if (unlikely(rt2x00queue_write_tx_frame(queue, skb, control->sta, false)))
 		goto exit_fail;
 
 	/*
diff --git a/drivers/net/wireless/rt2x00/rt2x00queue.c b/drivers/net/wireless/rt2x00/rt2x00queue.c
index 50590b1420a5..a5d38e8ad9e4 100644
--- a/drivers/net/wireless/rt2x00/rt2x00queue.c
+++ b/drivers/net/wireless/rt2x00/rt2x00queue.c
@@ -635,7 +635,7 @@ static void rt2x00queue_bar_check(struct queue_entry *entry)
 }
 
 int rt2x00queue_write_tx_frame(struct data_queue *queue, struct sk_buff *skb,
-			       bool local)
+			       struct ieee80211_sta *sta, bool local)
 {
 	struct ieee80211_tx_info *tx_info;
 	struct queue_entry *entry;
@@ -649,7 +649,7 @@ int rt2x00queue_write_tx_frame(struct data_queue *queue, struct sk_buff *skb,
 	 * after that we are free to use the skb->cb array
 	 * for our information.
 	 */
-	rt2x00queue_create_tx_descriptor(queue->rt2x00dev, skb, &txdesc, NULL);
+	rt2x00queue_create_tx_descriptor(queue->rt2x00dev, skb, &txdesc, sta);
 
 	/*
 	 * All information is retrieved from the skb->cb array,
diff --git a/drivers/net/wireless/rtlwifi/base.c b/drivers/net/wireless/rtlwifi/base.c
index 9a78e3daf742..ff784072fb42 100644
--- a/drivers/net/wireless/rtlwifi/base.c
+++ b/drivers/net/wireless/rtlwifi/base.c
@@ -37,6 +37,7 @@
 
 #include <linux/ip.h>
 #include <linux/module.h>
+#include <linux/udp.h>
 
 /*
  *NOTICE!!!: This file will be very big, we should
@@ -1074,64 +1075,52 @@ u8 rtl_is_special_data(struct ieee80211_hw *hw, struct sk_buff *skb, u8 is_tx)
 	if (!ieee80211_is_data(fc))
 		return false;
 
+	ip = (const struct iphdr *)(skb->data + mac_hdr_len +
+				    SNAP_SIZE + PROTOC_TYPE_SIZE);
+	ether_type = be16_to_cpup((__be16 *)
+				  (skb->data + mac_hdr_len + SNAP_SIZE));
 
-	ip = (struct iphdr *)((u8 *) skb->data + mac_hdr_len +
-			      SNAP_SIZE + PROTOC_TYPE_SIZE);
-	ether_type = *(u16 *) ((u8 *) skb->data + mac_hdr_len + SNAP_SIZE);
-	/*	ether_type = ntohs(ether_type); */
-
-	if (ETH_P_IP == ether_type) {
-		if (IPPROTO_UDP == ip->protocol) {
-			struct udphdr *udp = (struct udphdr *)((u8 *) ip +
-							       (ip->ihl << 2));
-			if (((((u8 *) udp)[1] == 68) &&
-			     (((u8 *) udp)[3] == 67)) ||
-			    ((((u8 *) udp)[1] == 67) &&
-			     (((u8 *) udp)[3] == 68))) {
-				/*
-				 * 68 : UDP BOOTP client
-				 * 67 : UDP BOOTP server
-				 */
-				RT_TRACE(rtlpriv, (COMP_SEND | COMP_RECV),
-					 DBG_DMESG, "dhcp %s !!\n",
-					 is_tx ? "Tx" : "Rx");
-
-				if (is_tx) {
-					rtlpriv->enter_ps = false;
-					schedule_work(&rtlpriv->
-						      works.lps_change_work);
-					ppsc->last_delaylps_stamp_jiffies =
-					    jiffies;
-				}
+	switch (ether_type) {
+	case ETH_P_IP: {
+		struct udphdr *udp;
+		u16 src;
+		u16 dst;
 
-				return true;
-			}
-		}
-	} else if (ETH_P_ARP == ether_type) {
-		if (is_tx) {
-			rtlpriv->enter_ps = false;
-			schedule_work(&rtlpriv->works.lps_change_work);
-			ppsc->last_delaylps_stamp_jiffies = jiffies;
-		}
+		if (ip->protocol != IPPROTO_UDP)
+			return false;
+		udp = (struct udphdr *)((u8 *)ip + (ip->ihl << 2));
+		src = be16_to_cpu(udp->source);
+		dst = be16_to_cpu(udp->dest);
 
-		return true;
-	} else if (ETH_P_PAE == ether_type) {
+		/* If this case involves port 68 (UDP BOOTP client) connecting
+		 * with port 67 (UDP BOOTP server), then return true so that
+		 * the lowest speed is used.
+		 */
+		if (!((src == 68 && dst == 67) || (src == 67 && dst == 68)))
+			return false;
+
+		RT_TRACE(rtlpriv, (COMP_SEND | COMP_RECV), DBG_DMESG,
+			 "dhcp %s !!\n", is_tx ? "Tx" : "Rx");
+		break;
+	}
+	case ETH_P_ARP:
+		break;
+	case ETH_P_PAE:
 		RT_TRACE(rtlpriv, (COMP_SEND | COMP_RECV), DBG_DMESG,
 			 "802.1X %s EAPOL pkt!!\n", is_tx ? "Tx" : "Rx");
-
-		if (is_tx) {
-			rtlpriv->enter_ps = false;
-			schedule_work(&rtlpriv->works.lps_change_work);
-			ppsc->last_delaylps_stamp_jiffies = jiffies;
-		}
-
-		return true;
-	} else if (ETH_P_IPV6 == ether_type) {
-		/* IPv6 */
-		return true;
+		break;
+	case ETH_P_IPV6:
+		/* TODO: Is this right? */
+		return false;
+	default:
+		return false;
 	}
-
-	return false;
+	if (is_tx) {
+		rtlpriv->enter_ps = false;
+		schedule_work(&rtlpriv->works.lps_change_work);
+		ppsc->last_delaylps_stamp_jiffies = jiffies;
+	}
+	return true;
 }
 EXPORT_SYMBOL_GPL(rtl_is_special_data);
 
diff --git a/drivers/net/wireless/rtlwifi/efuse.c b/drivers/net/wireless/rtlwifi/efuse.c
index ae13fb94b2e8..2ffc7298f686 100644
--- a/drivers/net/wireless/rtlwifi/efuse.c
+++ b/drivers/net/wireless/rtlwifi/efuse.c
@@ -262,9 +262,9 @@ void read_efuse(struct ieee80211_hw *hw, u16 _offset, u16 _size_byte, u8 *pbuf)
 			    sizeof(u8), GFP_ATOMIC);
 	if (!efuse_tbl)
 		return;
-	efuse_word = kmalloc(EFUSE_MAX_WORD_UNIT * sizeof(u16 *), GFP_ATOMIC);
+	efuse_word = kzalloc(EFUSE_MAX_WORD_UNIT * sizeof(u16 *), GFP_ATOMIC);
 	if (!efuse_word)
-		goto done;
+		goto out;
 	for (i = 0; i < EFUSE_MAX_WORD_UNIT; i++) {
 		efuse_word[i] = kmalloc(efuse_max_section * sizeof(u16),
 					GFP_ATOMIC);
@@ -378,6 +378,7 @@ done:
 	for (i = 0; i < EFUSE_MAX_WORD_UNIT; i++)
 		kfree(efuse_word[i]);
 	kfree(efuse_word);
+out:
 	kfree(efuse_tbl);
 }
 
diff --git a/drivers/net/wireless/rtlwifi/rtl8192cu/trx.c b/drivers/net/wireless/rtlwifi/rtl8192cu/trx.c
index 25e50ffc44ec..b0c346a9e4b8 100644
--- a/drivers/net/wireless/rtlwifi/rtl8192cu/trx.c
+++ b/drivers/net/wireless/rtlwifi/rtl8192cu/trx.c
@@ -349,7 +349,7 @@ bool rtl92cu_rx_query_desc(struct ieee80211_hw *hw,
 						 p_drvinfo);
 	}
 	/*rx_status->qual = stats->signal; */
-	rx_status->signal = stats->rssi + 10;
+	rx_status->signal = stats->recvsignalpower + 10;
 	return true;
 }
 
diff --git a/drivers/net/wireless/rtlwifi/rtl8192de/trx.c b/drivers/net/wireless/rtlwifi/rtl8192de/trx.c
index 945ddecf90c9..0eb0f4ae5920 100644
--- a/drivers/net/wireless/rtlwifi/rtl8192de/trx.c
+++ b/drivers/net/wireless/rtlwifi/rtl8192de/trx.c
@@ -525,7 +525,7 @@ bool rtl92de_rx_query_desc(struct ieee80211_hw *hw,	struct rtl_stats *stats,
 						   p_drvinfo);
 	}
 	/*rx_status->qual = stats->signal; */
-	rx_status->signal = stats->rssi + 10;
+	rx_status->signal = stats->recvsignalpower + 10;
 	return true;
 }
 
diff --git a/drivers/net/wireless/rtlwifi/rtl8192se/rf.c b/drivers/net/wireless/rtlwifi/rtl8192se/rf.c
index 5061f1db3f02..92d38ab3c60e 100644
--- a/drivers/net/wireless/rtlwifi/rtl8192se/rf.c
+++ b/drivers/net/wireless/rtlwifi/rtl8192se/rf.c
@@ -265,7 +265,7 @@ static void _rtl92s_get_txpower_writeval_byregulatory(struct ieee80211_hw *hw,
 				    rtlefuse->pwrgroup_ht40
 				    [RF90_PATH_A][chnl - 1]) {
 					pwrdiff_limit[i] =
-					  rtlefuse->pwrgroup_ht20
+					  rtlefuse->pwrgroup_ht40
 					  [RF90_PATH_A][chnl - 1];
 				}
 			} else {
diff --git a/drivers/net/wireless/rtlwifi/rtl8192se/trx.c b/drivers/net/wireless/rtlwifi/rtl8192se/trx.c
index 222d2e792ca6..27efbcdac6a9 100644
--- a/drivers/net/wireless/rtlwifi/rtl8192se/trx.c
+++ b/drivers/net/wireless/rtlwifi/rtl8192se/trx.c
@@ -329,7 +329,7 @@ bool rtl92se_rx_query_desc(struct ieee80211_hw *hw, struct rtl_stats *stats,
 	}
 
 	/*rx_status->qual = stats->signal; */
-	rx_status->signal = stats->rssi + 10;
+	rx_status->signal = stats->recvsignalpower + 10;
 
 	return true;
 }
diff --git a/drivers/net/wireless/rtlwifi/wifi.h b/drivers/net/wireless/rtlwifi/wifi.h
index d224dc3bb092..0c65386fa30d 100644
--- a/drivers/net/wireless/rtlwifi/wifi.h
+++ b/drivers/net/wireless/rtlwifi/wifi.h
@@ -77,11 +77,7 @@
 #define RTL_SLOT_TIME_9				9
 #define RTL_SLOT_TIME_20			20
 
-/*related with tcp/ip. */
-/*if_ehther.h*/
-#define ETH_P_PAE		0x888E	/*Port Access Entity (IEEE 802.1X) */
-#define ETH_P_IP		0x0800	/*Internet Protocol packet */
-#define ETH_P_ARP		0x0806	/*Address Resolution packet */
+/*related to tcp/ip. */
 #define SNAP_SIZE		6
 #define PROTOC_TYPE_SIZE	2
 
diff --git a/drivers/net/wireless/zd1211rw/zd_usb.c b/drivers/net/wireless/zd1211rw/zd_usb.c
index 7ef0b4a181e1..84d94f572a46 100644
--- a/drivers/net/wireless/zd1211rw/zd_usb.c
+++ b/drivers/net/wireless/zd1211rw/zd_usb.c
@@ -1619,7 +1619,7 @@ static void prepare_read_regs_int(struct zd_usb *usb,
 	atomic_set(&intr->read_regs_enabled, 1);
 	intr->read_regs.req = req;
 	intr->read_regs.req_count = count;
-	INIT_COMPLETION(intr->read_regs.completion);
+	reinit_completion(&intr->read_regs.completion);
 	spin_unlock_irq(&intr->lock);
 }
 
diff --git a/drivers/net/xen-netfront.c b/drivers/net/xen-netfront.c
index dd1011e55cb5..e59acb1daa23 100644
--- a/drivers/net/xen-netfront.c
+++ b/drivers/net/xen-netfront.c
@@ -277,12 +277,13 @@ static void xennet_alloc_rx_buffers(struct net_device *dev)
 		if (!page) {
 			kfree_skb(skb);
 no_skb:
-			/* Any skbuffs queued for refill? Force them out. */
-			if (i != 0)
-				goto refill;
 			/* Could not allocate any skbuffs. Try again later. */
 			mod_timer(&np->rx_refill_timer,
 				  jiffies + (HZ/10));
+
+			/* Any skbuffs queued for refill? Force them out. */
+			if (i != 0)
+				goto refill;
 			break;
 		}
 
@@ -1340,6 +1341,12 @@ static struct net_device *xennet_create_dev(struct xenbus_device *dev)
 	if (np->stats == NULL)
 		goto exit;
 
+	for_each_possible_cpu(i) {
+		struct netfront_stats *xen_nf_stats;
+		xen_nf_stats = per_cpu_ptr(np->stats, i);
+		u64_stats_init(&xen_nf_stats->syncp);
+	}
+
 	/* Initialise tx_skbs as a free chain containing every entry. */
 	np->tx_skb_freelist = 0;
 	for (i = 0; i < NET_TX_RING_SIZE; i++) {
diff --git a/drivers/ntb/ntb_transport.c b/drivers/ntb/ntb_transport.c
index 12a9e83c008b..d0222f13d154 100644
--- a/drivers/ntb/ntb_transport.c
+++ b/drivers/ntb/ntb_transport.c
@@ -1034,10 +1034,9 @@ static void ntb_async_rx(struct ntb_queue_entry *entry, void *offset,
 	struct dma_chan *chan = qp->dma_chan;
 	struct dma_device *device;
 	size_t pay_off, buff_off;
-	dma_addr_t src, dest;
+	struct dmaengine_unmap_data *unmap;
 	dma_cookie_t cookie;
 	void *buf = entry->buf;
-	unsigned long flags;
 
 	entry->len = len;
 
@@ -1045,35 +1044,49 @@ static void ntb_async_rx(struct ntb_queue_entry *entry, void *offset,
 		goto err;
 
 	if (len < copy_bytes) 
-		goto err1;
+		goto err_wait;
 
 	device = chan->device;
 	pay_off = (size_t) offset & ~PAGE_MASK;
 	buff_off = (size_t) buf & ~PAGE_MASK;
 
 	if (!is_dma_copy_aligned(device, pay_off, buff_off, len))
-		goto err1;
+		goto err_wait;
 
-	dest = dma_map_single(device->dev, buf, len, DMA_FROM_DEVICE);
-	if (dma_mapping_error(device->dev, dest))
-		goto err1;
+	unmap = dmaengine_get_unmap_data(device->dev, 2, GFP_NOWAIT);
+	if (!unmap)
+		goto err_wait;
 
-	src = dma_map_single(device->dev, offset, len, DMA_TO_DEVICE);
-	if (dma_mapping_error(device->dev, src))
-		goto err2;
+	unmap->len = len;
+	unmap->addr[0] = dma_map_page(device->dev, virt_to_page(offset),
+				      pay_off, len, DMA_TO_DEVICE);
+	if (dma_mapping_error(device->dev, unmap->addr[0]))
+		goto err_get_unmap;
+
+	unmap->to_cnt = 1;
 
-	flags = DMA_COMPL_DEST_UNMAP_SINGLE | DMA_COMPL_SRC_UNMAP_SINGLE |
-		DMA_PREP_INTERRUPT;
-	txd = device->device_prep_dma_memcpy(chan, dest, src, len, flags);
+	unmap->addr[1] = dma_map_page(device->dev, virt_to_page(buf),
+				      buff_off, len, DMA_FROM_DEVICE);
+	if (dma_mapping_error(device->dev, unmap->addr[1]))
+		goto err_get_unmap;
+
+	unmap->from_cnt = 1;
+
+	txd = device->device_prep_dma_memcpy(chan, unmap->addr[1],
+					     unmap->addr[0], len,
+					     DMA_PREP_INTERRUPT);
 	if (!txd)
-		goto err3;
+		goto err_get_unmap;
 
 	txd->callback = ntb_rx_copy_callback;
 	txd->callback_param = entry;
+	dma_set_unmap(txd, unmap);
 
 	cookie = dmaengine_submit(txd);
 	if (dma_submit_error(cookie))
-		goto err3;
+		goto err_set_unmap;
+
+	dmaengine_unmap_put(unmap);
 
 	qp->last_cookie = cookie;
 
@@ -1081,11 +1094,11 @@ static void ntb_async_rx(struct ntb_queue_entry *entry, void *offset,
 
 	return;
 
-err3:
-	dma_unmap_single(device->dev, src, len, DMA_TO_DEVICE);
-err2:
-	dma_unmap_single(device->dev, dest, len, DMA_FROM_DEVICE);
-err1:
+err_set_unmap:
+	dmaengine_unmap_put(unmap);
+err_get_unmap:
+	dmaengine_unmap_put(unmap);
+err_wait:
 	/* If the callbacks come out of order, the writing of the index to the
 	 * last completed will be out of order.  This may result in the
 	 * receive stalling forever.
@@ -1245,12 +1258,12 @@ static void ntb_async_tx(struct ntb_transport_qp *qp,
 	struct dma_chan *chan = qp->dma_chan;
 	struct dma_device *device;
 	size_t dest_off, buff_off;
-	dma_addr_t src, dest;
+	struct dmaengine_unmap_data *unmap;
+	dma_addr_t dest;
 	dma_cookie_t cookie;
 	void __iomem *offset;
 	size_t len = entry->len;
 	void *buf = entry->buf;
-	unsigned long flags;
 
 	offset = qp->tx_mw + qp->tx_max_frame * qp->tx_index;
 	hdr = offset + qp->tx_max_frame - sizeof(struct ntb_payload_header);
@@ -1273,28 +1286,41 @@ static void ntb_async_tx(struct ntb_transport_qp *qp,
 	if (!is_dma_copy_aligned(device, buff_off, dest_off, len))
 		goto err;
 
-	src = dma_map_single(device->dev, buf, len, DMA_TO_DEVICE);
-	if (dma_mapping_error(device->dev, src))
+	unmap = dmaengine_get_unmap_data(device->dev, 1, GFP_NOWAIT);
+	if (!unmap)
 		goto err;
 
-	flags = DMA_COMPL_SRC_UNMAP_SINGLE | DMA_PREP_INTERRUPT;
-	txd = device->device_prep_dma_memcpy(chan, dest, src, len, flags);
+	unmap->len = len;
+	unmap->addr[0] = dma_map_page(device->dev, virt_to_page(buf),
+				      buff_off, len, DMA_TO_DEVICE);
+	if (dma_mapping_error(device->dev, unmap->addr[0]))
+		goto err_get_unmap;
+
+	unmap->to_cnt = 1;
+
+	txd = device->device_prep_dma_memcpy(chan, dest, unmap->addr[0], len,
+					     DMA_PREP_INTERRUPT);
 	if (!txd)
-		goto err1;
+		goto err_get_unmap;
 
 	txd->callback = ntb_tx_copy_callback;
 	txd->callback_param = entry;
+	dma_set_unmap(txd, unmap);
 
 	cookie = dmaengine_submit(txd);
 	if (dma_submit_error(cookie))
-		goto err1;
+		goto err_set_unmap;
+
+	dmaengine_unmap_put(unmap);
 
 	dma_async_issue_pending(chan);
 	qp->tx_async++;
 
 	return;
-err1:
-	dma_unmap_single(device->dev, src, len, DMA_TO_DEVICE);
+err_set_unmap:
+	dmaengine_unmap_put(unmap);
+err_get_unmap:
+	dmaengine_unmap_put(unmap);
 err:
 	ntb_memcpy_tx(entry, offset);
 	qp->tx_memcpy++;
diff --git a/drivers/parport/Kconfig b/drivers/parport/Kconfig
index dc82ef096f3b..2872ece81f35 100644
--- a/drivers/parport/Kconfig
+++ b/drivers/parport/Kconfig
@@ -31,15 +31,18 @@ menuconfig PARPORT
 
 	  If unsure, say Y.
 
+config ARCH_MIGHT_HAVE_PC_PARPORT
+	bool
+	help
+	  Select this config option from the architecture Kconfig if
+	  the architecture might have PC parallel port hardware.
+
 if PARPORT
 
 config PARPORT_PC
 	tristate "PC-style hardware"
-	depends on (!SPARC64 || PCI) && !SPARC32 && !M32R && !FRV && !S390 && \
-		(!M68K || ISA) && !MN10300 && !AVR32 && !BLACKFIN && \
-		!XTENSA && !CRIS
-
-	---help---
+	depends on ARCH_MIGHT_HAVE_PC_PARPORT
+	help
 	  You should say Y here if you have a PC-style parallel port. All
 	  IBM PC compatible computers and some Alphas have PC-style
 	  parallel ports.  PA-RISC owners should only say Y here if they
diff --git a/drivers/parport/parport_ip32.c b/drivers/parport/parport_ip32.c
index d4716273651e..c864f82bd37d 100644
--- a/drivers/parport/parport_ip32.c
+++ b/drivers/parport/parport_ip32.c
@@ -1331,7 +1331,7 @@ static unsigned int parport_ip32_fwp_wait_interrupt(struct parport *p)
 			break;
 
 		/* Initialize mutex used to take interrupts into account */
-		INIT_COMPLETION(priv->irq_complete);
+		reinit_completion(&priv->irq_complete);
 
 		/* Enable serviceIntr */
 		parport_ip32_frob_econtrol(p, ECR_SERVINTR, 0);
@@ -1446,7 +1446,7 @@ static size_t parport_ip32_fifo_write_block_dma(struct parport *p,
 	priv->irq_mode = PARPORT_IP32_IRQ_HERE;
 
 	parport_ip32_dma_start(DMA_TO_DEVICE, (void *)buf, len);
-	INIT_COMPLETION(priv->irq_complete);
+	reinit_completion(&priv->irq_complete);
 	parport_ip32_frob_econtrol(p, ECR_DMAEN | ECR_SERVINTR, ECR_DMAEN);
 
 	nfault_timeout = min((unsigned long)physport->cad->timeout,
diff --git a/drivers/pci/pcie/aer/aerdrv_core.c b/drivers/pci/pcie/aer/aerdrv_core.c
index 85ca36f2136d..6b3a958e1be6 100644
--- a/drivers/pci/pcie/aer/aerdrv_core.c
+++ b/drivers/pci/pcie/aer/aerdrv_core.c
@@ -574,7 +574,7 @@ void aer_recover_queue(int domain, unsigned int bus, unsigned int devfn,
 	};
 
 	spin_lock_irqsave(&aer_recover_ring_lock, flags);
-	if (kfifo_put(&aer_recover_ring, &entry))
+	if (kfifo_put(&aer_recover_ring, entry))
 		schedule_work(&aer_recover_work);
 	else
 		pr_err("AER recover: Buffer overflow when recovering AER for %04x:%02x:%02x:%x\n",
diff --git a/drivers/pinctrl/pinctrl-single.c b/drivers/pinctrl/pinctrl-single.c
index 0846922b2316..829b98c5c66f 100644
--- a/drivers/pinctrl/pinctrl-single.c
+++ b/drivers/pinctrl/pinctrl-single.c
@@ -1604,6 +1604,9 @@ static inline void pcs_irq_set(struct pcs_soc_data *pcs_soc,
 		pcs->write(mask, pcswi->reg);
 		raw_spin_unlock(&pcs->lock);
 	}
+
+	if (pcs_soc->rearm)
+		pcs_soc->rearm();
 }
 
 /**
@@ -1626,8 +1629,6 @@ static void pcs_irq_unmask(struct irq_data *d)
 	struct pcs_soc_data *pcs_soc = irq_data_get_irq_chip_data(d);
 
 	pcs_irq_set(pcs_soc, d->irq, true);
-	if (pcs_soc->rearm)
-		pcs_soc->rearm();
 }
 
 /**
@@ -1678,11 +1679,6 @@ static int pcs_irq_handle(struct pcs_soc_data *pcs_soc)
 		}
 	}
 
-	/*
-	 * For debugging on omaps, you may want to call pcs_soc->rearm()
-	 * here to see wake-up interrupts during runtime also.
-	 */
-
 	return count;
 }
 
diff --git a/drivers/platform/x86/apple-gmux.c b/drivers/platform/x86/apple-gmux.c
index 5753c82335ad..b9429fbf1cd8 100644
--- a/drivers/platform/x86/apple-gmux.c
+++ b/drivers/platform/x86/apple-gmux.c
@@ -289,7 +289,7 @@ static int gmux_switchto(enum vga_switcheroo_client_id id)
 static int gmux_set_discrete_state(struct apple_gmux_data *gmux_data,
 				   enum vga_switcheroo_state state)
 {
-	INIT_COMPLETION(gmux_data->powerchange_done);
+	reinit_completion(&gmux_data->powerchange_done);
 
 	if (state == VGA_SWITCHEROO_ON) {
 		gmux_write8(gmux_data, GMUX_PORT_DISCRETE_POWER, 1);
diff --git a/drivers/power/Kconfig b/drivers/power/Kconfig
index e6f92b450913..5e2054afe840 100644
--- a/drivers/power/Kconfig
+++ b/drivers/power/Kconfig
@@ -346,6 +346,12 @@ config CHARGER_BQ24190
 	help
 	  Say Y to enable support for the TI BQ24190 battery charger.
 
+config CHARGER_BQ24735
+	tristate "TI BQ24735 battery charger support"
+	depends on I2C && GPIOLIB
+	help
+	  Say Y to enable support for the TI BQ24735 battery charger.
+
 config CHARGER_SMB347
 	tristate "Summit Microelectronics SMB347 Battery Charger"
 	depends on I2C
diff --git a/drivers/power/Makefile b/drivers/power/Makefile
index a4b74177706f..372b4e8ab598 100644
--- a/drivers/power/Makefile
+++ b/drivers/power/Makefile
@@ -52,6 +52,7 @@ obj-$(CONFIG_CHARGER_MAX8997)	+= max8997_charger.o
 obj-$(CONFIG_CHARGER_MAX8998)	+= max8998_charger.o
 obj-$(CONFIG_CHARGER_BQ2415X)	+= bq2415x_charger.o
 obj-$(CONFIG_CHARGER_BQ24190)	+= bq24190_charger.o
+obj-$(CONFIG_CHARGER_BQ24735)	+= bq24735-charger.o
 obj-$(CONFIG_POWER_AVS)		+= avs/
 obj-$(CONFIG_CHARGER_SMB347)	+= smb347-charger.o
 obj-$(CONFIG_CHARGER_TPS65090)	+= tps65090-charger.o
diff --git a/drivers/power/ab8500_charger.c b/drivers/power/ab8500_charger.c
index a4c4a10b3a41..19110aa613a1 100644
--- a/drivers/power/ab8500_charger.c
+++ b/drivers/power/ab8500_charger.c
@@ -766,7 +766,6 @@ static int ab8500_charger_max_usb_curr(struct ab8500_charger *di,
 			ret = -ENXIO;
 			break;
 		}
-		break;
 	case USB_STAT_CARKIT_1:
 	case USB_STAT_CARKIT_2:
 	case USB_STAT_ACA_DOCK_CHARGER:
@@ -1387,8 +1386,12 @@ static int ab8500_charger_ac_en(struct ux500_charger *charger,
 		 * the GPADC module independant of the AB8500 chargers
 		 */
 		if (!di->vddadc_en_ac) {
-			regulator_enable(di->regu);
-			di->vddadc_en_ac = true;
+			ret = regulator_enable(di->regu);
+			if (ret)
+				dev_warn(di->dev,
+					"Failed to enable regulator\n");
+			else
+				di->vddadc_en_ac = true;
 		}
 
 		/* Check if the requested voltage or current is valid */
@@ -1556,8 +1559,12 @@ static int ab8500_charger_usb_en(struct ux500_charger *charger,
 		 * the GPADC module independant of the AB8500 chargers
 		 */
 		if (!di->vddadc_en_usb) {
-			regulator_enable(di->regu);
-			di->vddadc_en_usb = true;
+			ret = regulator_enable(di->regu);
+			if (ret)
+				dev_warn(di->dev,
+					"Failed to enable regulator\n");
+			else
+				di->vddadc_en_usb = true;
 		}
 
 		/* Enable USB charging */
diff --git a/drivers/power/ab8500_fg.c b/drivers/power/ab8500_fg.c
index 754970717c31..3cb4178e397c 100644
--- a/drivers/power/ab8500_fg.c
+++ b/drivers/power/ab8500_fg.c
@@ -574,8 +574,8 @@ int ab8500_fg_inst_curr_start(struct ab8500_fg *di)
 	}
 
 	/* Return and WFI */
-	INIT_COMPLETION(di->ab8500_fg_started);
-	INIT_COMPLETION(di->ab8500_fg_complete);
+	reinit_completion(&di->ab8500_fg_started);
+	reinit_completion(&di->ab8500_fg_complete);
 	enable_irq(di->irq);
 
 	/* Note: cc_lock is still locked */
diff --git a/drivers/power/bq2415x_charger.c b/drivers/power/bq2415x_charger.c
index 0727f9256138..df893dd1447d 100644
--- a/drivers/power/bq2415x_charger.c
+++ b/drivers/power/bq2415x_charger.c
@@ -605,9 +605,13 @@ static int bq2415x_set_battery_regulation_voltage(struct bq2415x_device *bq,
 {
 	int val = (mV/10 - 350) / 2;
 
+	/*
+	 * According to datasheet, maximum battery regulation voltage is
+	 * 4440mV which is b101111 = 47.
+	 */
 	if (val < 0)
 		val = 0;
-	else if (val > 94) /* FIXME: Max is 94 or 122 ? Set max value ? */
+	else if (val > 47)
 		return -EINVAL;
 
 	return bq2415x_i2c_write_mask(bq, BQ2415X_REG_VOLTAGE, val,
diff --git a/drivers/power/bq24735-charger.c b/drivers/power/bq24735-charger.c
new file mode 100644
index 000000000000..d022b823305b
--- /dev/null
+++ b/drivers/power/bq24735-charger.c
@@ -0,0 +1,419 @@
+/*
+ * Battery charger driver for TI BQ24735
+ *
+ * Copyright (c) 2013, NVIDIA CORPORATION.  All rights reserved.
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation;
+ *
+ * This program is distributed in the hope that it will be useful, but WITHOUT
+ * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or
+ * FITNESS FOR A PARTICULAR PURPOSE.  See the GNU General Public License for
+ * more details.
+ *
+ * You should have received a copy of the GNU General Public License along
+ * with this program; if not, write to the Free Software Foundation, Inc.,
+ * 51 Franklin Street, Fifth Floor, Boston, MA  02110-1301, USA.
+ */
+
+#include <linux/err.h>
+#include <linux/gpio.h>
+#include <linux/i2c.h>
+#include <linux/init.h>
+#include <linux/interrupt.h>
+#include <linux/kernel.h>
+#include <linux/module.h>
+#include <linux/of.h>
+#include <linux/of_gpio.h>
+#include <linux/power_supply.h>
+#include <linux/slab.h>
+
+#include <linux/power/bq24735-charger.h>
+
+#define BQ24735_CHG_OPT			0x12
+#define BQ24735_CHG_OPT_CHARGE_DISABLE	(1 << 0)
+#define BQ24735_CHG_OPT_AC_PRESENT	(1 << 4)
+#define BQ24735_CHARGE_CURRENT		0x14
+#define BQ24735_CHARGE_CURRENT_MASK	0x1fc0
+#define BQ24735_CHARGE_VOLTAGE		0x15
+#define BQ24735_CHARGE_VOLTAGE_MASK	0x7ff0
+#define BQ24735_INPUT_CURRENT		0x3f
+#define BQ24735_INPUT_CURRENT_MASK	0x1f80
+#define BQ24735_MANUFACTURER_ID		0xfe
+#define BQ24735_DEVICE_ID		0xff
+
+struct bq24735 {
+	struct power_supply	charger;
+	struct i2c_client	*client;
+	struct bq24735_platform	*pdata;
+};
+
+static inline struct bq24735 *to_bq24735(struct power_supply *psy)
+{
+	return container_of(psy, struct bq24735, charger);
+}
+
+static enum power_supply_property bq24735_charger_properties[] = {
+	POWER_SUPPLY_PROP_ONLINE,
+};
+
+static inline int bq24735_write_word(struct i2c_client *client, u8 reg,
+				     u16 value)
+{
+	return i2c_smbus_write_word_data(client, reg, le16_to_cpu(value));
+}
+
+static inline int bq24735_read_word(struct i2c_client *client, u8 reg)
+{
+	s32 ret = i2c_smbus_read_word_data(client, reg);
+
+	return ret < 0 ? ret : le16_to_cpu(ret);
+}
+
+static int bq24735_update_word(struct i2c_client *client, u8 reg,
+			       u16 mask, u16 value)
+{
+	unsigned int tmp;
+	int ret;
+
+	ret = bq24735_read_word(client, reg);
+	if (ret < 0)
+		return ret;
+
+	tmp = ret & ~mask;
+	tmp |= value & mask;
+
+	return bq24735_write_word(client, reg, tmp);
+}
+
+static inline int bq24735_enable_charging(struct bq24735 *charger)
+{
+	return bq24735_update_word(charger->client, BQ24735_CHG_OPT,
+				   BQ24735_CHG_OPT_CHARGE_DISABLE,
+				   ~BQ24735_CHG_OPT_CHARGE_DISABLE);
+}
+
+static inline int bq24735_disable_charging(struct bq24735 *charger)
+{
+	return bq24735_update_word(charger->client, BQ24735_CHG_OPT,
+				   BQ24735_CHG_OPT_CHARGE_DISABLE,
+				   BQ24735_CHG_OPT_CHARGE_DISABLE);
+}
+
+static int bq24735_config_charger(struct bq24735 *charger)
+{
+	struct bq24735_platform *pdata = charger->pdata;
+	int ret;
+	u16 value;
+
+	if (pdata->charge_current) {
+		value = pdata->charge_current & BQ24735_CHARGE_CURRENT_MASK;
+
+		ret = bq24735_write_word(charger->client,
+					 BQ24735_CHARGE_CURRENT, value);
+		if (ret < 0) {
+			dev_err(&charger->client->dev,
+				"Failed to write charger current : %d\n",
+				ret);
+			return ret;
+		}
+	}
+
+	if (pdata->charge_voltage) {
+		value = pdata->charge_voltage & BQ24735_CHARGE_VOLTAGE_MASK;
+
+		ret = bq24735_write_word(charger->client,
+					 BQ24735_CHARGE_VOLTAGE, value);
+		if (ret < 0) {
+			dev_err(&charger->client->dev,
+				"Failed to write charger voltage : %d\n",
+				ret);
+			return ret;
+		}
+	}
+
+	if (pdata->input_current) {
+		value = pdata->input_current & BQ24735_INPUT_CURRENT_MASK;
+
+		ret = bq24735_write_word(charger->client,
+					 BQ24735_INPUT_CURRENT, value);
+		if (ret < 0) {
+			dev_err(&charger->client->dev,
+				"Failed to write input current : %d\n",
+				ret);
+			return ret;
+		}
+	}
+
+	return 0;
+}
+
+static bool bq24735_charger_is_present(struct bq24735 *charger)
+{
+	struct bq24735_platform *pdata = charger->pdata;
+	int ret;
+
+	if (pdata->status_gpio_valid) {
+		ret = gpio_get_value_cansleep(pdata->status_gpio);
+		return ret ^= pdata->status_gpio_active_low == 0;
+	} else {
+		int ac = 0;
+
+		ac = bq24735_read_word(charger->client, BQ24735_CHG_OPT);
+		if (ac < 0) {
+			dev_err(&charger->client->dev,
+				"Failed to read charger options : %d\n",
+				ac);
+			return false;
+		}
+		return (ac & BQ24735_CHG_OPT_AC_PRESENT) ? true : false;
+	}
+
+	return false;
+}
+
+static irqreturn_t bq24735_charger_isr(int irq, void *devid)
+{
+	struct power_supply *psy = devid;
+	struct bq24735 *charger = to_bq24735(psy);
+
+	if (bq24735_charger_is_present(charger))
+		bq24735_enable_charging(charger);
+	else
+		bq24735_disable_charging(charger);
+
+	power_supply_changed(psy);
+
+	return IRQ_HANDLED;
+}
+
+static int bq24735_charger_get_property(struct power_supply *psy,
+					enum power_supply_property psp,
+					union power_supply_propval *val)
+{
+	struct bq24735 *charger;
+
+	charger = container_of(psy, struct bq24735, charger);
+
+	switch (psp) {
+	case POWER_SUPPLY_PROP_ONLINE:
+		val->intval = bq24735_charger_is_present(charger) ? 1 : 0;
+		break;
+	default:
+		return -EINVAL;
+	}
+
+	return 0;
+}
+
+static struct bq24735_platform *bq24735_parse_dt_data(struct i2c_client *client)
+{
+	struct bq24735_platform *pdata;
+	struct device_node *np = client->dev.of_node;
+	u32 val;
+	int ret;
+	enum of_gpio_flags flags;
+
+	pdata = devm_kzalloc(&client->dev, sizeof(*pdata), GFP_KERNEL);
+	if (!pdata) {
+		dev_err(&client->dev,
+			"Memory alloc for bq24735 pdata failed\n");
+		return NULL;
+	}
+
+	pdata->status_gpio = of_get_named_gpio_flags(np, "ti,ac-detect-gpios",
+						     0, &flags);
+
+	if (flags & OF_GPIO_ACTIVE_LOW)
+		pdata->status_gpio_active_low = 1;
+
+	ret = of_property_read_u32(np, "ti,charge-current", &val);
+	if (!ret)
+		pdata->charge_current = val;
+
+	ret = of_property_read_u32(np, "ti,charge-voltage", &val);
+	if (!ret)
+		pdata->charge_voltage = val;
+
+	ret = of_property_read_u32(np, "ti,input-current", &val);
+	if (!ret)
+		pdata->input_current = val;
+
+	return pdata;
+}
+
+static int bq24735_charger_probe(struct i2c_client *client,
+				 const struct i2c_device_id *id)
+{
+	int ret;
+	struct bq24735 *charger;
+	struct power_supply *supply;
+	char *name;
+
+	charger = devm_kzalloc(&client->dev, sizeof(*charger), GFP_KERNEL);
+	if (!charger)
+		return -ENOMEM;
+
+	charger->pdata = client->dev.platform_data;
+
+	if (IS_ENABLED(CONFIG_OF) && !charger->pdata && client->dev.of_node)
+		charger->pdata = bq24735_parse_dt_data(client);
+
+	if (!charger->pdata) {
+		dev_err(&client->dev, "no platform data provided\n");
+		return -EINVAL;
+	}
+
+	name = (char *)charger->pdata->name;
+	if (!name) {
+		name = kasprintf(GFP_KERNEL, "bq24735@%s",
+				 dev_name(&client->dev));
+		if (!name) {
+			dev_err(&client->dev, "Failed to alloc device name\n");
+			return -ENOMEM;
+		}
+	}
+
+	charger->client = client;
+
+	supply = &charger->charger;
+
+	supply->name = name;
+	supply->type = POWER_SUPPLY_TYPE_MAINS;
+	supply->properties = bq24735_charger_properties;
+	supply->num_properties = ARRAY_SIZE(bq24735_charger_properties);
+	supply->get_property = bq24735_charger_get_property;
+	supply->supplied_to = charger->pdata->supplied_to;
+	supply->num_supplicants = charger->pdata->num_supplicants;
+	supply->of_node = client->dev.of_node;
+
+	i2c_set_clientdata(client, charger);
+
+	ret = bq24735_read_word(client, BQ24735_MANUFACTURER_ID);
+	if (ret < 0) {
+		dev_err(&client->dev, "Failed to read manufacturer id : %d\n",
+			ret);
+		goto err_free_name;
+	} else if (ret != 0x0040) {
+		dev_err(&client->dev,
+			"manufacturer id mismatch. 0x0040 != 0x%04x\n", ret);
+		ret = -ENODEV;
+		goto err_free_name;
+	}
+
+	ret = bq24735_read_word(client, BQ24735_DEVICE_ID);
+	if (ret < 0) {
+		dev_err(&client->dev, "Failed to read device id : %d\n", ret);
+		goto err_free_name;
+	} else if (ret != 0x000B) {
+		dev_err(&client->dev,
+			"device id mismatch. 0x000b != 0x%04x\n", ret);
+		ret = -ENODEV;
+		goto err_free_name;
+	}
+
+	if (gpio_is_valid(charger->pdata->status_gpio)) {
+		ret = devm_gpio_request(&client->dev,
+					charger->pdata->status_gpio,
+					name);
+		if (ret) {
+			dev_err(&client->dev,
+				"Failed GPIO request for GPIO %d: %d\n",
+				charger->pdata->status_gpio, ret);
+		}
+
+		charger->pdata->status_gpio_valid = !ret;
+	}
+
+	ret = bq24735_config_charger(charger);
+	if (ret < 0) {
+		dev_err(&client->dev, "failed in configuring charger");
+		goto err_free_name;
+	}
+
+	/* check for AC adapter presence */
+	if (bq24735_charger_is_present(charger)) {
+		ret = bq24735_enable_charging(charger);
+		if (ret < 0) {
+			dev_err(&client->dev, "Failed to enable charging\n");
+			goto err_free_name;
+		}
+	}
+
+	ret = power_supply_register(&client->dev, supply);
+	if (ret < 0) {
+		dev_err(&client->dev, "Failed to register power supply: %d\n",
+			ret);
+		goto err_free_name;
+	}
+
+	if (client->irq) {
+		ret = devm_request_threaded_irq(&client->dev, client->irq,
+						NULL, bq24735_charger_isr,
+						IRQF_TRIGGER_RISING |
+						IRQF_TRIGGER_FALLING |
+						IRQF_ONESHOT,
+						supply->name, supply);
+		if (ret) {
+			dev_err(&client->dev,
+				"Unable to register IRQ %d err %d\n",
+				client->irq, ret);
+			goto err_unregister_supply;
+		}
+	}
+
+	return 0;
+err_unregister_supply:
+	power_supply_unregister(supply);
+err_free_name:
+	if (name != charger->pdata->name)
+		kfree(name);
+
+	return ret;
+}
+
+static int bq24735_charger_remove(struct i2c_client *client)
+{
+	struct bq24735 *charger = i2c_get_clientdata(client);
+
+	if (charger->client->irq)
+		devm_free_irq(&charger->client->dev, charger->client->irq,
+			      &charger->charger);
+
+	power_supply_unregister(&charger->charger);
+
+	if (charger->charger.name != charger->pdata->name)
+		kfree(charger->charger.name);
+
+	return 0;
+}
+
+static const struct i2c_device_id bq24735_charger_id[] = {
+	{ "bq24735-charger", 0 },
+	{}
+};
+MODULE_DEVICE_TABLE(i2c, bq24735_charger_id);
+
+static const struct of_device_id bq24735_match_ids[] = {
+	{ .compatible = "ti,bq24735", },
+	{ /* end */ }
+};
+MODULE_DEVICE_TABLE(of, bq24735_match_ids);
+
+static struct i2c_driver bq24735_charger_driver = {
+	.driver = {
+		.name = "bq24735-charger",
+		.owner = THIS_MODULE,
+		.of_match_table = bq24735_match_ids,
+	},
+	.probe = bq24735_charger_probe,
+	.remove = bq24735_charger_remove,
+	.id_table = bq24735_charger_id,
+};
+
+module_i2c_driver(bq24735_charger_driver);
+
+MODULE_DESCRIPTION("bq24735 battery charging driver");
+MODULE_AUTHOR("Darbha Sriharsha <dsriharsha@nvidia.com>");
+MODULE_LICENSE("GPL v2");
diff --git a/drivers/power/charger-manager.c b/drivers/power/charger-manager.c
index e30e847600bb..7287c0efd6bf 100644
--- a/drivers/power/charger-manager.c
+++ b/drivers/power/charger-manager.c
@@ -1378,7 +1378,8 @@ static int charger_manager_register_sysfs(struct charger_manager *cm)
 		charger = &desc->charger_regulators[i];
 
 		snprintf(buf, 10, "charger.%d", i);
-		str = kzalloc(sizeof(char) * (strlen(buf) + 1), GFP_KERNEL);
+		str = devm_kzalloc(cm->dev,
+				sizeof(char) * (strlen(buf) + 1), GFP_KERNEL);
 		if (!str) {
 			ret = -ENOMEM;
 			goto err;
@@ -1452,30 +1453,23 @@ static int charger_manager_probe(struct platform_device *pdev)
 			rtc_dev = NULL;
 			dev_err(&pdev->dev, "Cannot get RTC %s\n",
 				g_desc->rtc_name);
-			ret = -ENODEV;
-			goto err_alloc;
+			return -ENODEV;
 		}
 	}
 
 	if (!desc) {
 		dev_err(&pdev->dev, "No platform data (desc) found\n");
-		ret = -ENODEV;
-		goto err_alloc;
+		return -ENODEV;
 	}
 
-	cm = kzalloc(sizeof(struct charger_manager), GFP_KERNEL);
-	if (!cm) {
-		ret = -ENOMEM;
-		goto err_alloc;
-	}
+	cm = devm_kzalloc(&pdev->dev,
+			sizeof(struct charger_manager),	GFP_KERNEL);
+	if (!cm)
+		return -ENOMEM;
 
 	/* Basic Values. Unspecified are Null or 0 */
 	cm->dev = &pdev->dev;
-	cm->desc = kmemdup(desc, sizeof(struct charger_desc), GFP_KERNEL);
-	if (!cm->desc) {
-		ret = -ENOMEM;
-		goto err_alloc_desc;
-	}
+	cm->desc = desc;
 	cm->last_temp_mC = INT_MIN; /* denotes "unmeasured, yet" */
 
 	/*
@@ -1498,27 +1492,23 @@ static int charger_manager_probe(struct platform_device *pdev)
 	}
 
 	if (!desc->charger_regulators || desc->num_charger_regulators < 1) {
-		ret = -EINVAL;
 		dev_err(&pdev->dev, "charger_regulators undefined\n");
-		goto err_no_charger;
+		return -EINVAL;
 	}
 
 	if (!desc->psy_charger_stat || !desc->psy_charger_stat[0]) {
 		dev_err(&pdev->dev, "No power supply defined\n");
-		ret = -EINVAL;
-		goto err_no_charger_stat;
+		return -EINVAL;
 	}
 
 	/* Counting index only */
 	while (desc->psy_charger_stat[i])
 		i++;
 
-	cm->charger_stat = kzalloc(sizeof(struct power_supply *) * (i + 1),
-				   GFP_KERNEL);
-	if (!cm->charger_stat) {
-		ret = -ENOMEM;
-		goto err_no_charger_stat;
-	}
+	cm->charger_stat = devm_kzalloc(&pdev->dev,
+				sizeof(struct power_supply *) * i, GFP_KERNEL);
+	if (!cm->charger_stat)
+		return -ENOMEM;
 
 	for (i = 0; desc->psy_charger_stat[i]; i++) {
 		cm->charger_stat[i] = power_supply_get_by_name(
@@ -1526,8 +1516,7 @@ static int charger_manager_probe(struct platform_device *pdev)
 		if (!cm->charger_stat[i]) {
 			dev_err(&pdev->dev, "Cannot find power supply \"%s\"\n",
 				desc->psy_charger_stat[i]);
-			ret = -ENODEV;
-			goto err_chg_stat;
+			return -ENODEV;
 		}
 	}
 
@@ -1535,21 +1524,18 @@ static int charger_manager_probe(struct platform_device *pdev)
 	if (!cm->fuel_gauge) {
 		dev_err(&pdev->dev, "Cannot find power supply \"%s\"\n",
 			desc->psy_fuel_gauge);
-		ret = -ENODEV;
-		goto err_chg_stat;
+		return -ENODEV;
 	}
 
 	if (desc->polling_interval_ms == 0 ||
 	    msecs_to_jiffies(desc->polling_interval_ms) <= CM_JIFFIES_SMALL) {
 		dev_err(&pdev->dev, "polling_interval_ms is too small\n");
-		ret = -EINVAL;
-		goto err_chg_stat;
+		return -EINVAL;
 	}
 
 	if (!desc->temperature_out_of_range) {
 		dev_err(&pdev->dev, "there is no temperature_out_of_range\n");
-		ret = -EINVAL;
-		goto err_chg_stat;
+		return -EINVAL;
 	}
 
 	if (!desc->charging_max_duration_ms ||
@@ -1570,14 +1556,13 @@ static int charger_manager_probe(struct platform_device *pdev)
 	cm->charger_psy.name = cm->psy_name_buf;
 
 	/* Allocate for psy properties because they may vary */
-	cm->charger_psy.properties = kzalloc(sizeof(enum power_supply_property)
+	cm->charger_psy.properties = devm_kzalloc(&pdev->dev,
+				sizeof(enum power_supply_property)
 				* (ARRAY_SIZE(default_charger_props) +
-				NUM_CHARGER_PSY_OPTIONAL),
-				GFP_KERNEL);
-	if (!cm->charger_psy.properties) {
-		ret = -ENOMEM;
-		goto err_chg_stat;
-	}
+				NUM_CHARGER_PSY_OPTIONAL), GFP_KERNEL);
+	if (!cm->charger_psy.properties)
+		return -ENOMEM;
+
 	memcpy(cm->charger_psy.properties, default_charger_props,
 		sizeof(enum power_supply_property) *
 		ARRAY_SIZE(default_charger_props));
@@ -1614,7 +1599,7 @@ static int charger_manager_probe(struct platform_device *pdev)
 	if (ret) {
 		dev_err(&pdev->dev, "Cannot register charger-manager with name \"%s\"\n",
 			cm->charger_psy.name);
-		goto err_register;
+		return ret;
 	}
 
 	/* Register extcon device for charger cable */
@@ -1655,8 +1640,6 @@ err_reg_sysfs:
 		charger = &desc->charger_regulators[i];
 		sysfs_remove_group(&cm->charger_psy.dev->kobj,
 				&charger->attr_g);
-
-		kfree(charger->attr_g.name);
 	}
 err_reg_extcon:
 	for (i = 0; i < desc->num_charger_regulators; i++) {
@@ -1674,16 +1657,7 @@ err_reg_extcon:
 	}
 
 	power_supply_unregister(&cm->charger_psy);
-err_register:
-	kfree(cm->charger_psy.properties);
-err_chg_stat:
-	kfree(cm->charger_stat);
-err_no_charger_stat:
-err_no_charger:
-	kfree(cm->desc);
-err_alloc_desc:
-	kfree(cm);
-err_alloc:
+
 	return ret;
 }
 
@@ -1718,11 +1692,6 @@ static int charger_manager_remove(struct platform_device *pdev)
 
 	try_charger_enable(cm, false);
 
-	kfree(cm->charger_psy.properties);
-	kfree(cm->charger_stat);
-	kfree(cm->desc);
-	kfree(cm);
-
 	return 0;
 }
 
diff --git a/drivers/power/isp1704_charger.c b/drivers/power/isp1704_charger.c
index fc04d191579b..1bb3a91b1acc 100644
--- a/drivers/power/isp1704_charger.c
+++ b/drivers/power/isp1704_charger.c
@@ -2,6 +2,7 @@
  * ISP1704 USB Charger Detection driver
  *
  * Copyright (C) 2010 Nokia Corporation
+ * Copyright (C) 2012 - 2013 Pali Rohár <pali.rohar@gmail.com>
  *
  * This program is free software; you can redistribute it and/or modify
  * it under the terms of the GNU General Public License as published by
@@ -65,10 +66,6 @@ struct isp1704_charger {
 	unsigned		present:1;
 	unsigned		online:1;
 	unsigned		current_max;
-
-	/* temp storage variables */
-	unsigned long		event;
-	unsigned		max_power;
 };
 
 static inline int isp1704_read(struct isp1704_charger *isp, u32 reg)
@@ -231,56 +228,59 @@ static inline int isp1704_charger_detect(struct isp1704_charger *isp)
 	return ret;
 }
 
+static inline int isp1704_charger_detect_dcp(struct isp1704_charger *isp)
+{
+	if (isp1704_charger_detect(isp) &&
+			isp1704_charger_type(isp) == POWER_SUPPLY_TYPE_USB_DCP)
+		return true;
+	else
+		return false;
+}
+
 static void isp1704_charger_work(struct work_struct *data)
 {
-	int			detect;
-	unsigned long		event;
-	unsigned		power;
 	struct isp1704_charger	*isp =
 		container_of(data, struct isp1704_charger, work);
 	static DEFINE_MUTEX(lock);
 
-	event = isp->event;
-	power = isp->max_power;
-
 	mutex_lock(&lock);
 
-	if (event != USB_EVENT_NONE)
-		isp1704_charger_set_power(isp, 1);
-
-	switch (event) {
+	switch (isp->phy->last_event) {
 	case USB_EVENT_VBUS:
-		isp->online = true;
+		/* do not call wall charger detection more times */
+		if (!isp->present) {
+			isp->online = true;
+			isp->present = 1;
+			isp1704_charger_set_power(isp, 1);
+
+			/* detect wall charger */
+			if (isp1704_charger_detect_dcp(isp)) {
+				isp->psy.type = POWER_SUPPLY_TYPE_USB_DCP;
+				isp->current_max = 1800;
+			} else {
+				isp->psy.type = POWER_SUPPLY_TYPE_USB;
+				isp->current_max = 500;
+			}
 
-		/* detect charger */
-		detect = isp1704_charger_detect(isp);
-
-		if (detect) {
-			isp->present = detect;
-			isp->psy.type = isp1704_charger_type(isp);
+			/* enable data pullups */
+			if (isp->phy->otg->gadget)
+				usb_gadget_connect(isp->phy->otg->gadget);
 		}
 
-		switch (isp->psy.type) {
-		case POWER_SUPPLY_TYPE_USB_DCP:
-			isp->current_max = 1800;
-			break;
-		case POWER_SUPPLY_TYPE_USB_CDP:
+		if (isp->psy.type != POWER_SUPPLY_TYPE_USB_DCP) {
 			/*
 			 * Only 500mA here or high speed chirp
 			 * handshaking may break
 			 */
-			isp->current_max = 500;
-			/* FALLTHROUGH */
-		case POWER_SUPPLY_TYPE_USB:
-		default:
-			/* enable data pullups */
-			if (isp->phy->otg->gadget)
-				usb_gadget_connect(isp->phy->otg->gadget);
+			if (isp->current_max > 500)
+				isp->current_max = 500;
+
+			if (isp->current_max > 100)
+				isp->psy.type = POWER_SUPPLY_TYPE_USB_CDP;
 		}
 		break;
 	case USB_EVENT_NONE:
 		isp->online = false;
-		isp->current_max = 0;
 		isp->present = 0;
 		isp->current_max = 0;
 		isp->psy.type = POWER_SUPPLY_TYPE_USB;
@@ -298,12 +298,6 @@ static void isp1704_charger_work(struct work_struct *data)
 
 		isp1704_charger_set_power(isp, 0);
 		break;
-	case USB_EVENT_ENUMERATED:
-		if (isp->present)
-			isp->current_max = 1800;
-		else
-			isp->current_max = power;
-		break;
 	default:
 		goto out;
 	}
@@ -314,16 +308,11 @@ out:
 }
 
 static int isp1704_notifier_call(struct notifier_block *nb,
-		unsigned long event, void *power)
+		unsigned long val, void *v)
 {
 	struct isp1704_charger *isp =
 		container_of(nb, struct isp1704_charger, nb);
 
-	isp->event = event;
-
-	if (power)
-		isp->max_power = *((unsigned *)power);
-
 	schedule_work(&isp->work);
 
 	return NOTIFY_OK;
@@ -462,13 +451,13 @@ static int isp1704_charger_probe(struct platform_device *pdev)
 	if (isp->phy->otg->gadget)
 		usb_gadget_disconnect(isp->phy->otg->gadget);
 
+	if (isp->phy->last_event == USB_EVENT_NONE)
+		isp1704_charger_set_power(isp, 0);
+
 	/* Detect charger if VBUS is valid (the cable was already plugged). */
-	ret = isp1704_read(isp, ULPI_USB_INT_STS);
-	isp1704_charger_set_power(isp, 0);
-	if ((ret & ULPI_INT_VBUS_VALID) && !isp->phy->otg->default_a) {
-		isp->event = USB_EVENT_VBUS;
+	if (isp->phy->last_event == USB_EVENT_VBUS &&
+			!isp->phy->otg->default_a)
 		schedule_work(&isp->work);
-	}
 
 	return 0;
 fail2:
diff --git a/drivers/power/jz4740-battery.c b/drivers/power/jz4740-battery.c
index d9686aa9270a..6c8931d4ad62 100644
--- a/drivers/power/jz4740-battery.c
+++ b/drivers/power/jz4740-battery.c
@@ -73,7 +73,7 @@ static long jz_battery_read_voltage(struct jz_battery *battery)
 
 	mutex_lock(&battery->lock);
 
-	INIT_COMPLETION(battery->read_completion);
+	reinit_completion(&battery->read_completion);
 
 	enable_irq(battery->irq);
 	battery->cell->enable(battery->pdev);
diff --git a/drivers/power/max17042_battery.c b/drivers/power/max17042_battery.c
index d664ef58afa7..e0b22f9b6fdd 100644
--- a/drivers/power/max17042_battery.c
+++ b/drivers/power/max17042_battery.c
@@ -33,6 +33,7 @@
 #include <linux/power_supply.h>
 #include <linux/power/max17042_battery.h>
 #include <linux/of.h>
+#include <linux/regmap.h>
 
 /* Status register bits */
 #define STATUS_POR_BIT         (1 << 1)
@@ -67,6 +68,7 @@
 
 struct max17042_chip {
 	struct i2c_client *client;
+	struct regmap *regmap;
 	struct power_supply battery;
 	enum max170xx_chip_type chip_type;
 	struct max17042_platform_data *pdata;
@@ -74,35 +76,6 @@ struct max17042_chip {
 	int    init_complete;
 };
 
-static int max17042_write_reg(struct i2c_client *client, u8 reg, u16 value)
-{
-	int ret = i2c_smbus_write_word_data(client, reg, value);
-
-	if (ret < 0)
-		dev_err(&client->dev, "%s: err %d\n", __func__, ret);
-
-	return ret;
-}
-
-static int max17042_read_reg(struct i2c_client *client, u8 reg)
-{
-	int ret = i2c_smbus_read_word_data(client, reg);
-
-	if (ret < 0)
-		dev_err(&client->dev, "%s: err %d\n", __func__, ret);
-
-	return ret;
-}
-
-static void max17042_set_reg(struct i2c_client *client,
-			     struct max17042_reg_data *data, int size)
-{
-	int i;
-
-	for (i = 0; i < size; i++)
-		max17042_write_reg(client, data[i].addr, data[i].data);
-}
-
 static enum power_supply_property max17042_battery_props[] = {
 	POWER_SUPPLY_PROP_PRESENT,
 	POWER_SUPPLY_PROP_CYCLE_COUNT,
@@ -125,96 +98,98 @@ static int max17042_get_property(struct power_supply *psy,
 {
 	struct max17042_chip *chip = container_of(psy,
 				struct max17042_chip, battery);
+	struct regmap *map = chip->regmap;
 	int ret;
+	u32 data;
 
 	if (!chip->init_complete)
 		return -EAGAIN;
 
 	switch (psp) {
 	case POWER_SUPPLY_PROP_PRESENT:
-		ret = max17042_read_reg(chip->client, MAX17042_STATUS);
+		ret = regmap_read(map, MAX17042_STATUS, &data);
 		if (ret < 0)
 			return ret;
 
-		if (ret & MAX17042_STATUS_BattAbsent)
+		if (data & MAX17042_STATUS_BattAbsent)
 			val->intval = 0;
 		else
 			val->intval = 1;
 		break;
 	case POWER_SUPPLY_PROP_CYCLE_COUNT:
-		ret = max17042_read_reg(chip->client, MAX17042_Cycles);
+		ret = regmap_read(map, MAX17042_Cycles, &data);
 		if (ret < 0)
 			return ret;
 
-		val->intval = ret;
+		val->intval = data;
 		break;
 	case POWER_SUPPLY_PROP_VOLTAGE_MAX:
-		ret = max17042_read_reg(chip->client, MAX17042_MinMaxVolt);
+		ret = regmap_read(map, MAX17042_MinMaxVolt, &data);
 		if (ret < 0)
 			return ret;
 
-		val->intval = ret >> 8;
+		val->intval = data >> 8;
 		val->intval *= 20000; /* Units of LSB = 20mV */
 		break;
 	case POWER_SUPPLY_PROP_VOLTAGE_MIN_DESIGN:
 		if (chip->chip_type == MAX17042)
-			ret = max17042_read_reg(chip->client, MAX17042_V_empty);
+			ret = regmap_read(map, MAX17042_V_empty, &data);
 		else
-			ret = max17042_read_reg(chip->client, MAX17047_V_empty);
+			ret = regmap_read(map, MAX17047_V_empty, &data);
 		if (ret < 0)
 			return ret;
 
-		val->intval = ret >> 7;
+		val->intval = data >> 7;
 		val->intval *= 10000; /* Units of LSB = 10mV */
 		break;
 	case POWER_SUPPLY_PROP_VOLTAGE_NOW:
-		ret = max17042_read_reg(chip->client, MAX17042_VCELL);
+		ret = regmap_read(map, MAX17042_VCELL, &data);
 		if (ret < 0)
 			return ret;
 
-		val->intval = ret * 625 / 8;
+		val->intval = data * 625 / 8;
 		break;
 	case POWER_SUPPLY_PROP_VOLTAGE_AVG:
-		ret = max17042_read_reg(chip->client, MAX17042_AvgVCELL);
+		ret = regmap_read(map, MAX17042_AvgVCELL, &data);
 		if (ret < 0)
 			return ret;
 
-		val->intval = ret * 625 / 8;
+		val->intval = data * 625 / 8;
 		break;
 	case POWER_SUPPLY_PROP_VOLTAGE_OCV:
-		ret = max17042_read_reg(chip->client, MAX17042_OCVInternal);
+		ret = regmap_read(map, MAX17042_OCVInternal, &data);
 		if (ret < 0)
 			return ret;
 
-		val->intval = ret * 625 / 8;
+		val->intval = data * 625 / 8;
 		break;
 	case POWER_SUPPLY_PROP_CAPACITY:
-		ret = max17042_read_reg(chip->client, MAX17042_RepSOC);
+		ret = regmap_read(map, MAX17042_RepSOC, &data);
 		if (ret < 0)
 			return ret;
 
-		val->intval = ret >> 8;
+		val->intval = data >> 8;
 		break;
 	case POWER_SUPPLY_PROP_CHARGE_FULL:
-		ret = max17042_read_reg(chip->client, MAX17042_FullCAP);
+		ret = regmap_read(map, MAX17042_FullCAP, &data);
 		if (ret < 0)
 			return ret;
 
-		val->intval = ret * 1000 / 2;
+		val->intval = data * 1000 / 2;
 		break;
 	case POWER_SUPPLY_PROP_CHARGE_COUNTER:
-		ret = max17042_read_reg(chip->client, MAX17042_QH);
+		ret = regmap_read(map, MAX17042_QH, &data);
 		if (ret < 0)
 			return ret;
 
-		val->intval = ret * 1000 / 2;
+		val->intval = data * 1000 / 2;
 		break;
 	case POWER_SUPPLY_PROP_TEMP:
-		ret = max17042_read_reg(chip->client, MAX17042_TEMP);
+		ret = regmap_read(map, MAX17042_TEMP, &data);
 		if (ret < 0)
 			return ret;
 
-		val->intval = ret;
+		val->intval = data;
 		/* The value is signed. */
 		if (val->intval & 0x8000) {
 			val->intval = (0x7fff & ~val->intval) + 1;
@@ -226,11 +201,11 @@ static int max17042_get_property(struct power_supply *psy,
 		break;
 	case POWER_SUPPLY_PROP_CURRENT_NOW:
 		if (chip->pdata->enable_current_sense) {
-			ret = max17042_read_reg(chip->client, MAX17042_Current);
+			ret = regmap_read(map, MAX17042_Current, &data);
 			if (ret < 0)
 				return ret;
 
-			val->intval = ret;
+			val->intval = data;
 			if (val->intval & 0x8000) {
 				/* Negative */
 				val->intval = ~val->intval & 0x7fff;
@@ -244,12 +219,11 @@ static int max17042_get_property(struct power_supply *psy,
 		break;
 	case POWER_SUPPLY_PROP_CURRENT_AVG:
 		if (chip->pdata->enable_current_sense) {
-			ret = max17042_read_reg(chip->client,
-						MAX17042_AvgCurrent);
+			ret = regmap_read(map, MAX17042_AvgCurrent, &data);
 			if (ret < 0)
 				return ret;
 
-			val->intval = ret;
+			val->intval = data;
 			if (val->intval & 0x8000) {
 				/* Negative */
 				val->intval = ~val->intval & 0x7fff;
@@ -267,16 +241,15 @@ static int max17042_get_property(struct power_supply *psy,
 	return 0;
 }
 
-static int max17042_write_verify_reg(struct i2c_client *client,
-				u8 reg, u16 value)
+static int max17042_write_verify_reg(struct regmap *map, u8 reg, u32 value)
 {
 	int retries = 8;
 	int ret;
-	u16 read_value;
+	u32 read_value;
 
 	do {
-		ret = i2c_smbus_write_word_data(client, reg, value);
-		read_value =  max17042_read_reg(client, reg);
+		ret = regmap_write(map, reg, value);
+		regmap_read(map, reg, &read_value);
 		if (read_value != value) {
 			ret = -EIO;
 			retries--;
@@ -284,50 +257,51 @@ static int max17042_write_verify_reg(struct i2c_client *client,
 	} while (retries && read_value != value);
 
 	if (ret < 0)
-		dev_err(&client->dev, "%s: err %d\n", __func__, ret);
+		pr_err("%s: err %d\n", __func__, ret);
 
 	return ret;
 }
 
-static inline void max17042_override_por(
-	struct i2c_client *client, u8 reg, u16 value)
+static inline void max17042_override_por(struct regmap *map,
+					 u8 reg, u16 value)
 {
 	if (value)
-		max17042_write_reg(client, reg, value);
+		regmap_write(map, reg, value);
 }
 
 static inline void max10742_unlock_model(struct max17042_chip *chip)
 {
-	struct i2c_client *client = chip->client;
-	max17042_write_reg(client, MAX17042_MLOCKReg1, MODEL_UNLOCK1);
-	max17042_write_reg(client, MAX17042_MLOCKReg2, MODEL_UNLOCK2);
+	struct regmap *map = chip->regmap;
+	regmap_write(map, MAX17042_MLOCKReg1, MODEL_UNLOCK1);
+	regmap_write(map, MAX17042_MLOCKReg2, MODEL_UNLOCK2);
 }
 
 static inline void max10742_lock_model(struct max17042_chip *chip)
 {
-	struct i2c_client *client = chip->client;
-	max17042_write_reg(client, MAX17042_MLOCKReg1, MODEL_LOCK1);
-	max17042_write_reg(client, MAX17042_MLOCKReg2, MODEL_LOCK2);
+	struct regmap *map = chip->regmap;
+
+	regmap_write(map, MAX17042_MLOCKReg1, MODEL_LOCK1);
+	regmap_write(map, MAX17042_MLOCKReg2, MODEL_LOCK2);
 }
 
 static inline void max17042_write_model_data(struct max17042_chip *chip,
 					u8 addr, int size)
 {
-	struct i2c_client *client = chip->client;
+	struct regmap *map = chip->regmap;
 	int i;
 	for (i = 0; i < size; i++)
-		max17042_write_reg(client, addr + i,
-				chip->pdata->config_data->cell_char_tbl[i]);
+		regmap_write(map, addr + i,
+			chip->pdata->config_data->cell_char_tbl[i]);
 }
 
 static inline void max17042_read_model_data(struct max17042_chip *chip,
-					u8 addr, u16 *data, int size)
+					u8 addr, u32 *data, int size)
 {
-	struct i2c_client *client = chip->client;
+	struct regmap *map = chip->regmap;
 	int i;
 
 	for (i = 0; i < size; i++)
-		data[i] = max17042_read_reg(client, addr + i);
+		regmap_read(map, addr + i, &data[i]);
 }
 
 static inline int max17042_model_data_compare(struct max17042_chip *chip,
@@ -350,7 +324,7 @@ static int max17042_init_model(struct max17042_chip *chip)
 {
 	int ret;
 	int table_size = ARRAY_SIZE(chip->pdata->config_data->cell_char_tbl);
-	u16 *temp_data;
+	u32 *temp_data;
 
 	temp_data = kcalloc(table_size, sizeof(*temp_data), GFP_KERNEL);
 	if (!temp_data)
@@ -365,7 +339,7 @@ static int max17042_init_model(struct max17042_chip *chip)
 	ret = max17042_model_data_compare(
 		chip,
 		chip->pdata->config_data->cell_char_tbl,
-		temp_data,
+		(u16 *)temp_data,
 		table_size);
 
 	max10742_lock_model(chip);
@@ -378,7 +352,7 @@ static int max17042_verify_model_lock(struct max17042_chip *chip)
 {
 	int i;
 	int table_size = ARRAY_SIZE(chip->pdata->config_data->cell_char_tbl);
-	u16 *temp_data;
+	u32 *temp_data;
 	int ret = 0;
 
 	temp_data = kcalloc(table_size, sizeof(*temp_data), GFP_KERNEL);
@@ -398,40 +372,38 @@ static int max17042_verify_model_lock(struct max17042_chip *chip)
 static void max17042_write_config_regs(struct max17042_chip *chip)
 {
 	struct max17042_config_data *config = chip->pdata->config_data;
+	struct regmap *map = chip->regmap;
 
-	max17042_write_reg(chip->client, MAX17042_CONFIG, config->config);
-	max17042_write_reg(chip->client, MAX17042_LearnCFG, config->learn_cfg);
-	max17042_write_reg(chip->client, MAX17042_FilterCFG,
+	regmap_write(map, MAX17042_CONFIG, config->config);
+	regmap_write(map, MAX17042_LearnCFG, config->learn_cfg);
+	regmap_write(map, MAX17042_FilterCFG,
 			config->filter_cfg);
-	max17042_write_reg(chip->client, MAX17042_RelaxCFG, config->relax_cfg);
+	regmap_write(map, MAX17042_RelaxCFG, config->relax_cfg);
 	if (chip->chip_type == MAX17047)
-		max17042_write_reg(chip->client, MAX17047_FullSOCThr,
+		regmap_write(map, MAX17047_FullSOCThr,
 						config->full_soc_thresh);
 }
 
 static void  max17042_write_custom_regs(struct max17042_chip *chip)
 {
 	struct max17042_config_data *config = chip->pdata->config_data;
+	struct regmap *map = chip->regmap;
 
-	max17042_write_verify_reg(chip->client, MAX17042_RCOMP0,
-				config->rcomp0);
-	max17042_write_verify_reg(chip->client, MAX17042_TempCo,
-				config->tcompc0);
-	max17042_write_verify_reg(chip->client, MAX17042_ICHGTerm,
-				config->ichgt_term);
+	max17042_write_verify_reg(map, MAX17042_RCOMP0, config->rcomp0);
+	max17042_write_verify_reg(map, MAX17042_TempCo,	config->tcompc0);
+	max17042_write_verify_reg(map, MAX17042_ICHGTerm, config->ichgt_term);
 	if (chip->chip_type == MAX17042) {
-		max17042_write_reg(chip->client, MAX17042_EmptyTempCo,
-					config->empty_tempco);
-		max17042_write_verify_reg(chip->client, MAX17042_K_empty0,
+		regmap_write(map, MAX17042_EmptyTempCo,	config->empty_tempco);
+		max17042_write_verify_reg(map, MAX17042_K_empty0,
 					config->kempty0);
 	} else {
-		max17042_write_verify_reg(chip->client, MAX17047_QRTbl00,
+		max17042_write_verify_reg(map, MAX17047_QRTbl00,
 						config->qrtbl00);
-		max17042_write_verify_reg(chip->client, MAX17047_QRTbl10,
+		max17042_write_verify_reg(map, MAX17047_QRTbl10,
 						config->qrtbl10);
-		max17042_write_verify_reg(chip->client, MAX17047_QRTbl20,
+		max17042_write_verify_reg(map, MAX17047_QRTbl20,
 						config->qrtbl20);
-		max17042_write_verify_reg(chip->client, MAX17047_QRTbl30,
+		max17042_write_verify_reg(map, MAX17047_QRTbl30,
 						config->qrtbl30);
 	}
 }
@@ -439,58 +411,60 @@ static void  max17042_write_custom_regs(struct max17042_chip *chip)
 static void max17042_update_capacity_regs(struct max17042_chip *chip)
 {
 	struct max17042_config_data *config = chip->pdata->config_data;
+	struct regmap *map = chip->regmap;
 
-	max17042_write_verify_reg(chip->client, MAX17042_FullCAP,
+	max17042_write_verify_reg(map, MAX17042_FullCAP,
 				config->fullcap);
-	max17042_write_reg(chip->client, MAX17042_DesignCap,
-			config->design_cap);
-	max17042_write_verify_reg(chip->client, MAX17042_FullCAPNom,
+	regmap_write(map, MAX17042_DesignCap, config->design_cap);
+	max17042_write_verify_reg(map, MAX17042_FullCAPNom,
 				config->fullcapnom);
 }
 
 static void max17042_reset_vfsoc0_reg(struct max17042_chip *chip)
 {
-	u16 vfSoc;
+	unsigned int vfSoc;
+	struct regmap *map = chip->regmap;
 
-	vfSoc = max17042_read_reg(chip->client, MAX17042_VFSOC);
-	max17042_write_reg(chip->client, MAX17042_VFSOC0Enable, VFSOC0_UNLOCK);
-	max17042_write_verify_reg(chip->client, MAX17042_VFSOC0, vfSoc);
-	max17042_write_reg(chip->client, MAX17042_VFSOC0Enable, VFSOC0_LOCK);
+	regmap_read(map, MAX17042_VFSOC, &vfSoc);
+	regmap_write(map, MAX17042_VFSOC0Enable, VFSOC0_UNLOCK);
+	max17042_write_verify_reg(map, MAX17042_VFSOC0, vfSoc);
+	regmap_write(map, MAX17042_VFSOC0Enable, VFSOC0_LOCK);
 }
 
 static void max17042_load_new_capacity_params(struct max17042_chip *chip)
 {
-	u16 full_cap0, rep_cap, dq_acc, vfSoc;
+	u32 full_cap0, rep_cap, dq_acc, vfSoc;
 	u32 rem_cap;
 
 	struct max17042_config_data *config = chip->pdata->config_data;
+	struct regmap *map = chip->regmap;
 
-	full_cap0 = max17042_read_reg(chip->client, MAX17042_FullCAP0);
-	vfSoc = max17042_read_reg(chip->client, MAX17042_VFSOC);
+	regmap_read(map, MAX17042_FullCAP0, &full_cap0);
+	regmap_read(map, MAX17042_VFSOC, &vfSoc);
 
 	/* fg_vfSoc needs to shifted by 8 bits to get the
 	 * perc in 1% accuracy, to get the right rem_cap multiply
 	 * full_cap0, fg_vfSoc and devide by 100
 	 */
 	rem_cap = ((vfSoc >> 8) * full_cap0) / 100;
-	max17042_write_verify_reg(chip->client, MAX17042_RemCap, (u16)rem_cap);
+	max17042_write_verify_reg(map, MAX17042_RemCap, rem_cap);
 
-	rep_cap = (u16)rem_cap;
-	max17042_write_verify_reg(chip->client, MAX17042_RepCap, rep_cap);
+	rep_cap = rem_cap;
+	max17042_write_verify_reg(map, MAX17042_RepCap, rep_cap);
 
 	/* Write dQ_acc to 200% of Capacity and dP_acc to 200% */
 	dq_acc = config->fullcap / dQ_ACC_DIV;
-	max17042_write_verify_reg(chip->client, MAX17042_dQacc, dq_acc);
-	max17042_write_verify_reg(chip->client, MAX17042_dPacc, dP_ACC_200);
+	max17042_write_verify_reg(map, MAX17042_dQacc, dq_acc);
+	max17042_write_verify_reg(map, MAX17042_dPacc, dP_ACC_200);
 
-	max17042_write_verify_reg(chip->client, MAX17042_FullCAP,
+	max17042_write_verify_reg(map, MAX17042_FullCAP,
 			config->fullcap);
-	max17042_write_reg(chip->client, MAX17042_DesignCap,
+	regmap_write(map, MAX17042_DesignCap,
 			config->design_cap);
-	max17042_write_verify_reg(chip->client, MAX17042_FullCAPNom,
+	max17042_write_verify_reg(map, MAX17042_FullCAPNom,
 			config->fullcapnom);
 	/* Update SOC register with new SOC */
-	max17042_write_reg(chip->client, MAX17042_RepSOC, vfSoc);
+	regmap_write(map, MAX17042_RepSOC, vfSoc);
 }
 
 /*
@@ -500,59 +474,60 @@ static void max17042_load_new_capacity_params(struct max17042_chip *chip)
  */
 static inline void max17042_override_por_values(struct max17042_chip *chip)
 {
-	struct i2c_client *client = chip->client;
+	struct regmap *map = chip->regmap;
 	struct max17042_config_data *config = chip->pdata->config_data;
 
-	max17042_override_por(client, MAX17042_TGAIN, config->tgain);
-	max17042_override_por(client, MAx17042_TOFF, config->toff);
-	max17042_override_por(client, MAX17042_CGAIN, config->cgain);
-	max17042_override_por(client, MAX17042_COFF, config->coff);
-
-	max17042_override_por(client, MAX17042_VALRT_Th, config->valrt_thresh);
-	max17042_override_por(client, MAX17042_TALRT_Th, config->talrt_thresh);
-	max17042_override_por(client, MAX17042_SALRT_Th,
-			config->soc_alrt_thresh);
-	max17042_override_por(client, MAX17042_CONFIG, config->config);
-	max17042_override_por(client, MAX17042_SHDNTIMER, config->shdntimer);
-
-	max17042_override_por(client, MAX17042_DesignCap, config->design_cap);
-	max17042_override_por(client, MAX17042_ICHGTerm, config->ichgt_term);
-
-	max17042_override_por(client, MAX17042_AtRate, config->at_rate);
-	max17042_override_por(client, MAX17042_LearnCFG, config->learn_cfg);
-	max17042_override_por(client, MAX17042_FilterCFG, config->filter_cfg);
-	max17042_override_por(client, MAX17042_RelaxCFG, config->relax_cfg);
-	max17042_override_por(client, MAX17042_MiscCFG, config->misc_cfg);
-	max17042_override_por(client, MAX17042_MaskSOC, config->masksoc);
-
-	max17042_override_por(client, MAX17042_FullCAP, config->fullcap);
-	max17042_override_por(client, MAX17042_FullCAPNom, config->fullcapnom);
+	max17042_override_por(map, MAX17042_TGAIN, config->tgain);
+	max17042_override_por(map, MAx17042_TOFF, config->toff);
+	max17042_override_por(map, MAX17042_CGAIN, config->cgain);
+	max17042_override_por(map, MAX17042_COFF, config->coff);
+
+	max17042_override_por(map, MAX17042_VALRT_Th, config->valrt_thresh);
+	max17042_override_por(map, MAX17042_TALRT_Th, config->talrt_thresh);
+	max17042_override_por(map, MAX17042_SALRT_Th,
+						config->soc_alrt_thresh);
+	max17042_override_por(map, MAX17042_CONFIG, config->config);
+	max17042_override_por(map, MAX17042_SHDNTIMER, config->shdntimer);
+
+	max17042_override_por(map, MAX17042_DesignCap, config->design_cap);
+	max17042_override_por(map, MAX17042_ICHGTerm, config->ichgt_term);
+
+	max17042_override_por(map, MAX17042_AtRate, config->at_rate);
+	max17042_override_por(map, MAX17042_LearnCFG, config->learn_cfg);
+	max17042_override_por(map, MAX17042_FilterCFG, config->filter_cfg);
+	max17042_override_por(map, MAX17042_RelaxCFG, config->relax_cfg);
+	max17042_override_por(map, MAX17042_MiscCFG, config->misc_cfg);
+	max17042_override_por(map, MAX17042_MaskSOC, config->masksoc);
+
+	max17042_override_por(map, MAX17042_FullCAP, config->fullcap);
+	max17042_override_por(map, MAX17042_FullCAPNom, config->fullcapnom);
 	if (chip->chip_type == MAX17042)
-		max17042_override_por(client, MAX17042_SOC_empty,
+		max17042_override_por(map, MAX17042_SOC_empty,
 						config->socempty);
-	max17042_override_por(client, MAX17042_LAvg_empty, config->lavg_empty);
-	max17042_override_por(client, MAX17042_dQacc, config->dqacc);
-	max17042_override_por(client, MAX17042_dPacc, config->dpacc);
+	max17042_override_por(map, MAX17042_LAvg_empty, config->lavg_empty);
+	max17042_override_por(map, MAX17042_dQacc, config->dqacc);
+	max17042_override_por(map, MAX17042_dPacc, config->dpacc);
 
 	if (chip->chip_type == MAX17042)
-		max17042_override_por(client, MAX17042_V_empty, config->vempty);
+		max17042_override_por(map, MAX17042_V_empty, config->vempty);
 	else
-		max17042_override_por(client, MAX17047_V_empty, config->vempty);
-	max17042_override_por(client, MAX17042_TempNom, config->temp_nom);
-	max17042_override_por(client, MAX17042_TempLim, config->temp_lim);
-	max17042_override_por(client, MAX17042_FCTC, config->fctc);
-	max17042_override_por(client, MAX17042_RCOMP0, config->rcomp0);
-	max17042_override_por(client, MAX17042_TempCo, config->tcompc0);
+		max17042_override_por(map, MAX17047_V_empty, config->vempty);
+	max17042_override_por(map, MAX17042_TempNom, config->temp_nom);
+	max17042_override_por(map, MAX17042_TempLim, config->temp_lim);
+	max17042_override_por(map, MAX17042_FCTC, config->fctc);
+	max17042_override_por(map, MAX17042_RCOMP0, config->rcomp0);
+	max17042_override_por(map, MAX17042_TempCo, config->tcompc0);
 	if (chip->chip_type) {
-		max17042_override_por(client, MAX17042_EmptyTempCo,
-					config->empty_tempco);
-		max17042_override_por(client, MAX17042_K_empty0,
-					config->kempty0);
+		max17042_override_por(map, MAX17042_EmptyTempCo,
+						config->empty_tempco);
+		max17042_override_por(map, MAX17042_K_empty0,
+						config->kempty0);
 	}
 }
 
 static int max17042_init_chip(struct max17042_chip *chip)
 {
+	struct regmap *map = chip->regmap;
 	int ret;
 	int val;
 
@@ -597,31 +572,32 @@ static int max17042_init_chip(struct max17042_chip *chip)
 	max17042_load_new_capacity_params(chip);
 
 	/* Init complete, Clear the POR bit */
-	val = max17042_read_reg(chip->client, MAX17042_STATUS);
-	max17042_write_reg(chip->client, MAX17042_STATUS,
-			val & (~STATUS_POR_BIT));
+	regmap_read(map, MAX17042_STATUS, &val);
+	regmap_write(map, MAX17042_STATUS, val & (~STATUS_POR_BIT));
 	return 0;
 }
 
 static void max17042_set_soc_threshold(struct max17042_chip *chip, u16 off)
 {
-	u16 soc, soc_tr;
+	struct regmap *map = chip->regmap;
+	u32 soc, soc_tr;
 
 	/* program interrupt thesholds such that we should
 	 * get interrupt for every 'off' perc change in the soc
 	 */
-	soc = max17042_read_reg(chip->client, MAX17042_RepSOC) >> 8;
+	regmap_read(map, MAX17042_RepSOC, &soc);
+	soc >>= 8;
 	soc_tr = (soc + off) << 8;
 	soc_tr |= (soc - off);
-	max17042_write_reg(chip->client, MAX17042_SALRT_Th, soc_tr);
+	regmap_write(map, MAX17042_SALRT_Th, soc_tr);
 }
 
 static irqreturn_t max17042_thread_handler(int id, void *dev)
 {
 	struct max17042_chip *chip = dev;
-	u16 val;
+	u32 val;
 
-	val = max17042_read_reg(chip->client, MAX17042_STATUS);
+	regmap_read(chip->regmap, MAX17042_STATUS, &val);
 	if ((val & STATUS_INTR_SOCMIN_BIT) ||
 		(val & STATUS_INTR_SOCMAX_BIT)) {
 		dev_info(&chip->client->dev, "SOC threshold INTR\n");
@@ -682,13 +658,20 @@ max17042_get_pdata(struct device *dev)
 }
 #endif
 
+static struct regmap_config max17042_regmap_config = {
+	.reg_bits = 8,
+	.val_bits = 16,
+	.val_format_endian = REGMAP_ENDIAN_NATIVE,
+};
+
 static int max17042_probe(struct i2c_client *client,
 			const struct i2c_device_id *id)
 {
 	struct i2c_adapter *adapter = to_i2c_adapter(client->dev.parent);
 	struct max17042_chip *chip;
 	int ret;
-	int reg;
+	int i;
+	u32 val;
 
 	if (!i2c_check_functionality(adapter, I2C_FUNC_SMBUS_WORD_DATA))
 		return -EIO;
@@ -698,6 +681,12 @@ static int max17042_probe(struct i2c_client *client,
 		return -ENOMEM;
 
 	chip->client = client;
+	chip->regmap = devm_regmap_init_i2c(client, &max17042_regmap_config);
+	if (IS_ERR(chip->regmap)) {
+		dev_err(&client->dev, "Failed to initialize regmap\n");
+		return -EINVAL;
+	}
+
 	chip->pdata = max17042_get_pdata(&client->dev);
 	if (!chip->pdata) {
 		dev_err(&client->dev, "no platform data provided\n");
@@ -706,15 +695,15 @@ static int max17042_probe(struct i2c_client *client,
 
 	i2c_set_clientdata(client, chip);
 
-	ret = max17042_read_reg(chip->client, MAX17042_DevName);
-	if (ret == MAX17042_IC_VERSION) {
+	regmap_read(chip->regmap, MAX17042_DevName, &val);
+	if (val == MAX17042_IC_VERSION) {
 		dev_dbg(&client->dev, "chip type max17042 detected\n");
 		chip->chip_type = MAX17042;
-	} else if (ret == MAX17047_IC_VERSION) {
+	} else if (val == MAX17047_IC_VERSION) {
 		dev_dbg(&client->dev, "chip type max17047/50 detected\n");
 		chip->chip_type = MAX17047;
 	} else {
-		dev_err(&client->dev, "device version mismatch: %x\n", ret);
+		dev_err(&client->dev, "device version mismatch: %x\n", val);
 		return -EIO;
 	}
 
@@ -733,13 +722,15 @@ static int max17042_probe(struct i2c_client *client,
 		chip->pdata->r_sns = MAX17042_DEFAULT_SNS_RESISTOR;
 
 	if (chip->pdata->init_data)
-		max17042_set_reg(client, chip->pdata->init_data,
-				chip->pdata->num_init_data);
+		for (i = 0; i < chip->pdata->num_init_data; i++)
+			regmap_write(chip->regmap,
+					chip->pdata->init_data[i].addr,
+					chip->pdata->init_data[i].data);
 
 	if (!chip->pdata->enable_current_sense) {
-		max17042_write_reg(client, MAX17042_CGAIN, 0x0000);
-		max17042_write_reg(client, MAX17042_MiscCFG, 0x0003);
-		max17042_write_reg(client, MAX17042_LearnCFG, 0x0007);
+		regmap_write(chip->regmap, MAX17042_CGAIN, 0x0000);
+		regmap_write(chip->regmap, MAX17042_MiscCFG, 0x0003);
+		regmap_write(chip->regmap, MAX17042_LearnCFG, 0x0007);
 	}
 
 	ret = power_supply_register(&client->dev, &chip->battery);
@@ -754,9 +745,9 @@ static int max17042_probe(struct i2c_client *client,
 						IRQF_TRIGGER_FALLING,
 						chip->battery.name, chip);
 		if (!ret) {
-			reg =  max17042_read_reg(client, MAX17042_CONFIG);
-			reg |= CONFIG_ALRT_BIT_ENBL;
-			max17042_write_reg(client, MAX17042_CONFIG, reg);
+			regmap_read(chip->regmap, MAX17042_CONFIG, &val);
+			val |= CONFIG_ALRT_BIT_ENBL;
+			regmap_write(chip->regmap, MAX17042_CONFIG, val);
 			max17042_set_soc_threshold(chip, 1);
 		} else {
 			client->irq = 0;
@@ -765,8 +756,8 @@ static int max17042_probe(struct i2c_client *client,
 		}
 	}
 
-	reg = max17042_read_reg(chip->client, MAX17042_STATUS);
-	if (reg & STATUS_POR_BIT) {
+	regmap_read(chip->regmap, MAX17042_STATUS, &val);
+	if (val & STATUS_POR_BIT) {
 		INIT_WORK(&chip->work, max17042_init_worker);
 		schedule_work(&chip->work);
 	} else {
@@ -786,7 +777,7 @@ static int max17042_remove(struct i2c_client *client)
 	return 0;
 }
 
-#ifdef CONFIG_PM
+#ifdef CONFIG_PM_SLEEP
 static int max17042_suspend(struct device *dev)
 {
 	struct max17042_chip *chip = dev_get_drvdata(dev);
@@ -816,17 +807,11 @@ static int max17042_resume(struct device *dev)
 
 	return 0;
 }
-
-static const struct dev_pm_ops max17042_pm_ops = {
-	.suspend	= max17042_suspend,
-	.resume		= max17042_resume,
-};
-
-#define MAX17042_PM_OPS (&max17042_pm_ops)
-#else
-#define MAX17042_PM_OPS NULL
 #endif
 
+static SIMPLE_DEV_PM_OPS(max17042_pm_ops, max17042_suspend,
+			max17042_resume);
+
 #ifdef CONFIG_OF
 static const struct of_device_id max17042_dt_match[] = {
 	{ .compatible = "maxim,max17042" },
@@ -849,7 +834,7 @@ static struct i2c_driver max17042_i2c_driver = {
 	.driver	= {
 		.name	= "max17042",
 		.of_match_table = of_match_ptr(max17042_dt_match),
-		.pm	= MAX17042_PM_OPS,
+		.pm	= &max17042_pm_ops,
 	},
 	.probe		= max17042_probe,
 	.remove		= max17042_remove,
diff --git a/drivers/power/pm2301_charger.c b/drivers/power/pm2301_charger.c
index ffa10ed83eb1..62c15af58c9a 100644
--- a/drivers/power/pm2301_charger.c
+++ b/drivers/power/pm2301_charger.c
@@ -205,7 +205,7 @@ static int pm2xxx_charger_batt_therm_mngt(struct pm2xxx_charger *pm2, int val)
 }
 
 
-int pm2xxx_charger_die_therm_mngt(struct pm2xxx_charger *pm2, int val)
+static int pm2xxx_charger_die_therm_mngt(struct pm2xxx_charger *pm2, int val)
 {
 	queue_work(pm2->charger_wq, &pm2->check_main_thermal_prot_work);
 
@@ -722,8 +722,12 @@ static int pm2xxx_charger_ac_en(struct ux500_charger *charger,
 
 		dev_dbg(pm2->dev, "Enable AC: %dmV %dmA\n", vset, iset);
 		if (!pm2->vddadc_en_ac) {
-			regulator_enable(pm2->regu);
-			pm2->vddadc_en_ac = true;
+			ret = regulator_enable(pm2->regu);
+			if (ret)
+				dev_warn(pm2->dev,
+					"Failed to enable vddadc regulator\n");
+			else
+				pm2->vddadc_en_ac = true;
 		}
 
 		ret = pm2xxx_charging_init(pm2);
@@ -953,37 +957,24 @@ static int  pm2xxx_runtime_suspend(struct device *dev)
 {
 	struct i2c_client *pm2xxx_i2c_client = to_i2c_client(dev);
 	struct pm2xxx_charger *pm2;
-	int ret = 0;
 
 	pm2 = (struct pm2xxx_charger *)i2c_get_clientdata(pm2xxx_i2c_client);
-	if (!pm2) {
-		dev_err(pm2->dev, "no pm2xxx_charger data supplied\n");
-		ret = -EINVAL;
-		return ret;
-	}
-
 	clear_lpn_pin(pm2);
 
-	return ret;
+	return 0;
 }
 
 static int  pm2xxx_runtime_resume(struct device *dev)
 {
 	struct i2c_client *pm2xxx_i2c_client = to_i2c_client(dev);
 	struct pm2xxx_charger *pm2;
-	int ret = 0;
 
 	pm2 = (struct pm2xxx_charger *)i2c_get_clientdata(pm2xxx_i2c_client);
-	if (!pm2) {
-		dev_err(pm2->dev, "no pm2xxx_charger data supplied\n");
-		ret = -EINVAL;
-		return ret;
-	}
 
 	if (gpio_is_valid(pm2->lpn_pin) && gpio_get_value(pm2->lpn_pin) == 0)
 		set_lpn_pin(pm2);
 
-	return ret;
+	return 0;
 }
 
 #endif
diff --git a/drivers/power/tps65090-charger.c b/drivers/power/tps65090-charger.c
index bdd7b9b2546a..8fc9d6df87f6 100644
--- a/drivers/power/tps65090-charger.c
+++ b/drivers/power/tps65090-charger.c
@@ -15,15 +15,17 @@
  * You should have received a copy of the GNU General Public License
  * along with this program.  If not, see <http://www.gnu.org/licenses/>.
  */
+#include <linux/delay.h>
 #include <linux/err.h>
 #include <linux/init.h>
 #include <linux/interrupt.h>
 #include <linux/kernel.h>
 #include <linux/module.h>
-#include <linux/slab.h>
-#include <linux/delay.h>
+#include <linux/of_device.h>
 #include <linux/platform_device.h>
 #include <linux/power_supply.h>
+#include <linux/slab.h>
+
 #include <linux/mfd/tps65090.h>
 
 #define TPS65090_REG_INTR_STS	0x00
@@ -185,10 +187,6 @@ static irqreturn_t tps65090_charger_isr(int irq, void *dev_id)
 	return IRQ_HANDLED;
 }
 
-#if defined(CONFIG_OF)
-
-#include <linux/of_device.h>
-
 static struct tps65090_platform_data *
 		tps65090_parse_dt_charger_data(struct platform_device *pdev)
 {
@@ -210,13 +208,6 @@ static struct tps65090_platform_data *
 	return pdata;
 
 }
-#else
-static struct tps65090_platform_data *
-		tps65090_parse_dt_charger_data(struct platform_device *pdev)
-{
-	return NULL;
-}
-#endif
 
 static int tps65090_charger_probe(struct platform_device *pdev)
 {
@@ -228,7 +219,7 @@ static int tps65090_charger_probe(struct platform_device *pdev)
 
 	pdata = dev_get_platdata(pdev->dev.parent);
 
-	if (!pdata && pdev->dev.of_node)
+	if (IS_ENABLED(CONFIG_OF) && !pdata && pdev->dev.of_node)
 		pdata = tps65090_parse_dt_charger_data(pdev);
 
 	if (!pdata) {
@@ -277,13 +268,13 @@ static int tps65090_charger_probe(struct platform_device *pdev)
 	if (ret) {
 		dev_err(cdata->dev, "Unable to register irq %d err %d\n", irq,
 			ret);
-		goto fail_free_irq;
+		goto fail_unregister_supply;
 	}
 
 	ret = tps65090_config_charger(cdata);
 	if (ret < 0) {
 		dev_err(&pdev->dev, "charger config failed, err %d\n", ret);
-		goto fail_free_irq;
+		goto fail_unregister_supply;
 	}
 
 	/* Check for charger presence */
@@ -292,14 +283,14 @@ static int tps65090_charger_probe(struct platform_device *pdev)
 	if (ret < 0) {
 		dev_err(cdata->dev, "%s(): Error in reading reg 0x%x", __func__,
 			TPS65090_REG_CG_STATUS1);
-		goto fail_free_irq;
+		goto fail_unregister_supply;
 	}
 
 	if (status1 != 0) {
 		ret = tps65090_enable_charging(cdata);
 		if (ret < 0) {
 			dev_err(cdata->dev, "error enabling charger\n");
-			goto fail_free_irq;
+			goto fail_unregister_supply;
 		}
 		cdata->ac_online = 1;
 		power_supply_changed(&cdata->ac);
@@ -307,8 +298,6 @@ static int tps65090_charger_probe(struct platform_device *pdev)
 
 	return 0;
 
-fail_free_irq:
-	devm_free_irq(cdata->dev, irq, cdata);
 fail_unregister_supply:
 	power_supply_unregister(&cdata->ac);
 
@@ -319,7 +308,6 @@ static int tps65090_charger_remove(struct platform_device *pdev)
 {
 	struct tps65090_charger *cdata = platform_get_drvdata(pdev);
 
-	devm_free_irq(cdata->dev, cdata->irq, cdata);
 	power_supply_unregister(&cdata->ac);
 
 	return 0;
diff --git a/drivers/power/twl4030_charger.c b/drivers/power/twl4030_charger.c
index d98abe911e37..f14108844e1a 100644
--- a/drivers/power/twl4030_charger.c
+++ b/drivers/power/twl4030_charger.c
@@ -495,10 +495,38 @@ static enum power_supply_property twl4030_charger_props[] = {
 	POWER_SUPPLY_PROP_CURRENT_NOW,
 };
 
+#ifdef CONFIG_OF
+static const struct twl4030_bci_platform_data *
+twl4030_bci_parse_dt(struct device *dev)
+{
+	struct device_node *np = dev->of_node;
+	struct twl4030_bci_platform_data *pdata;
+	u32 num;
+
+	if (!np)
+		return NULL;
+	pdata = devm_kzalloc(dev, sizeof(*pdata), GFP_KERNEL);
+	if (!pdata)
+		return pdata;
+
+	if (of_property_read_u32(np, "ti,bb-uvolt", &num) == 0)
+		pdata->bb_uvolt = num;
+	if (of_property_read_u32(np, "ti,bb-uamp", &num) == 0)
+		pdata->bb_uamp = num;
+	return pdata;
+}
+#else
+static inline const struct twl4030_bci_platform_data *
+twl4030_bci_parse_dt(struct device *dev)
+{
+	return NULL;
+}
+#endif
+
 static int __init twl4030_bci_probe(struct platform_device *pdev)
 {
 	struct twl4030_bci *bci;
-	struct twl4030_bci_platform_data *pdata = pdev->dev.platform_data;
+	const struct twl4030_bci_platform_data *pdata = pdev->dev.platform_data;
 	int ret;
 	u32 reg;
 
@@ -506,6 +534,9 @@ static int __init twl4030_bci_probe(struct platform_device *pdev)
 	if (bci == NULL)
 		return -ENOMEM;
 
+	if (!pdata)
+		pdata = twl4030_bci_parse_dt(&pdev->dev);
+
 	bci->dev = &pdev->dev;
 	bci->irq_chg = platform_get_irq(pdev, 0);
 	bci->irq_bci = platform_get_irq(pdev, 1);
@@ -581,8 +612,11 @@ static int __init twl4030_bci_probe(struct platform_device *pdev)
 
 	twl4030_charger_enable_ac(true);
 	twl4030_charger_enable_usb(bci, true);
-	twl4030_charger_enable_backup(pdata->bb_uvolt,
-				      pdata->bb_uamp);
+	if (pdata)
+		twl4030_charger_enable_backup(pdata->bb_uvolt,
+					      pdata->bb_uamp);
+	else
+		twl4030_charger_enable_backup(0, 0);
 
 	return 0;
 
@@ -631,10 +665,17 @@ static int __exit twl4030_bci_remove(struct platform_device *pdev)
 	return 0;
 }
 
+static const struct of_device_id twl_bci_of_match[] = {
+	{.compatible = "ti,twl4030-bci", },
+	{ }
+};
+MODULE_DEVICE_TABLE(of, twl_bci_of_match);
+
 static struct platform_driver twl4030_bci_driver = {
 	.driver	= {
 		.name	= "twl4030_bci",
 		.owner	= THIS_MODULE,
+		.of_match_table = of_match_ptr(twl_bci_of_match),
 	},
 	.remove	= __exit_p(twl4030_bci_remove),
 };
diff --git a/drivers/pwm/Kconfig b/drivers/pwm/Kconfig
index 75840b5cea6d..eece329d7872 100644
--- a/drivers/pwm/Kconfig
+++ b/drivers/pwm/Kconfig
@@ -62,6 +62,15 @@ config PWM_BFIN
 	  To compile this driver as a module, choose M here: the module
 	  will be called pwm-bfin.
 
+config PWM_EP93XX
+	tristate "Cirrus Logic EP93xx PWM support"
+	depends on ARCH_EP93XX
+	help
+	  Generic PWM framework driver for Cirrus Logic EP93xx.
+
+	  To compile this driver as a module, choose M here: the module
+	  will be called pwm-ep93xx.
+
 config PWM_IMX
 	tristate "i.MX PWM support"
 	depends on ARCH_MXC
diff --git a/drivers/pwm/Makefile b/drivers/pwm/Makefile
index 77a8c185c5b2..8b754e4dba4a 100644
--- a/drivers/pwm/Makefile
+++ b/drivers/pwm/Makefile
@@ -3,6 +3,7 @@ obj-$(CONFIG_PWM_SYSFS)		+= sysfs.o
 obj-$(CONFIG_PWM_AB8500)	+= pwm-ab8500.o
 obj-$(CONFIG_PWM_ATMEL_TCB)	+= pwm-atmel-tcb.o
 obj-$(CONFIG_PWM_BFIN)		+= pwm-bfin.o
+obj-$(CONFIG_PWM_EP93XX)	+= pwm-ep93xx.o
 obj-$(CONFIG_PWM_IMX)		+= pwm-imx.o
 obj-$(CONFIG_PWM_JZ4740)	+= pwm-jz4740.o
 obj-$(CONFIG_PWM_LPC32XX)	+= pwm-lpc32xx.o
diff --git a/drivers/pwm/pwm-atmel-tcb.c b/drivers/pwm/pwm-atmel-tcb.c
index ba6ce01035e4..f3dcd02390f1 100644
--- a/drivers/pwm/pwm-atmel-tcb.c
+++ b/drivers/pwm/pwm-atmel-tcb.c
@@ -249,6 +249,8 @@ static int atmel_tcb_pwm_enable(struct pwm_chip *chip, struct pwm_device *pwm)
 		}
 	}
 
+	cmr |= (tcbpwm->div & ATMEL_TC_TCCLKS);
+
 	__raw_writel(cmr, regs + ATMEL_TC_REG(group, CMR));
 
 	if (index == 0)
@@ -305,7 +307,7 @@ static int atmel_tcb_pwm_config(struct pwm_chip *chip, struct pwm_device *pwm,
 		i = slowclk;
 		rate = 32768;
 		min = div_u64(NSEC_PER_SEC, rate);
-		max = min << 16;
+		max = min << tc->tcb_config->counter_width;
 
 		/* If period is too big return ERANGE error */
 		if (max < period_ns)
diff --git a/drivers/pwm/pwm-ep93xx.c b/drivers/pwm/pwm-ep93xx.c
new file mode 100644
index 000000000000..33aa4461e1ce
--- /dev/null
+++ b/drivers/pwm/pwm-ep93xx.c
@@ -0,0 +1,230 @@
+/*
+ * PWM framework driver for Cirrus Logic EP93xx
+ *
+ * Copyright (c) 2009        Matthieu Crapet <mcrapet@gmail.com>
+ * Copyright (c) 2009, 2013  H Hartley Sweeten <hsweeten@visionengravers.com>
+ *
+ * EP9301/02 have only one channel:
+ *   platform device ep93xx-pwm.1 - PWMOUT1 (EGPIO14)
+ *
+ * EP9307 has only one channel:
+ *   platform device ep93xx-pwm.0 - PWMOUT
+ *
+ * EP9312/15 have two channels:
+ *   platform device ep93xx-pwm.0 - PWMOUT
+ *   platform device ep93xx-pwm.1 - PWMOUT1 (EGPIO14)
+ *
+ * This program is free software; you can redistribute it and/or
+ * modify it under the terms of the GNU General Public License
+ * as published by the Free Software Foundation; either version
+ * 2 of the License, or (at your option) any later version.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+ * GNU General Public License for more details.
+ */
+
+#include <linux/module.h>
+#include <linux/platform_device.h>
+#include <linux/slab.h>
+#include <linux/clk.h>
+#include <linux/err.h>
+#include <linux/io.h>
+#include <linux/pwm.h>
+
+#include <asm/div64.h>
+
+#include <mach/platform.h>	/* for ep93xx_pwm_{acquire,release}_gpio() */
+
+#define EP93XX_PWMx_TERM_COUNT	0x00
+#define EP93XX_PWMx_DUTY_CYCLE	0x04
+#define EP93XX_PWMx_ENABLE	0x08
+#define EP93XX_PWMx_INVERT	0x0c
+
+struct ep93xx_pwm {
+	void __iomem *base;
+	struct clk *clk;
+	struct pwm_chip chip;
+};
+
+static inline struct ep93xx_pwm *to_ep93xx_pwm(struct pwm_chip *chip)
+{
+	return container_of(chip, struct ep93xx_pwm, chip);
+}
+
+static int ep93xx_pwm_request(struct pwm_chip *chip, struct pwm_device *pwm)
+{
+	struct platform_device *pdev = to_platform_device(chip->dev);
+
+	return ep93xx_pwm_acquire_gpio(pdev);
+}
+
+static void ep93xx_pwm_free(struct pwm_chip *chip, struct pwm_device *pwm)
+{
+	struct platform_device *pdev = to_platform_device(chip->dev);
+
+	ep93xx_pwm_release_gpio(pdev);
+}
+
+static int ep93xx_pwm_config(struct pwm_chip *chip, struct pwm_device *pwm,
+			     int duty_ns, int period_ns)
+{
+	struct ep93xx_pwm *ep93xx_pwm = to_ep93xx_pwm(chip);
+	void __iomem *base = ep93xx_pwm->base;
+	unsigned long long c;
+	unsigned long period_cycles;
+	unsigned long duty_cycles;
+	unsigned long term;
+	int ret = 0;
+
+	/*
+	 * The clock needs to be enabled to access the PWM registers.
+	 * Configuration can be changed at any time.
+	 */
+	if (!test_bit(PWMF_ENABLED, &pwm->flags)) {
+		ret = clk_enable(ep93xx_pwm->clk);
+		if (ret)
+			return ret;
+	}
+
+	c = clk_get_rate(ep93xx_pwm->clk);
+	c *= period_ns;
+	do_div(c, 1000000000);
+	period_cycles = c;
+
+	c = period_cycles;
+	c *= duty_ns;
+	do_div(c, period_ns);
+	duty_cycles = c;
+
+	if (period_cycles < 0x10000 && duty_cycles < 0x10000) {
+		term = readw(base + EP93XX_PWMx_TERM_COUNT);
+
+		/* Order is important if PWM is running */
+		if (period_cycles > term) {
+			writew(period_cycles, base + EP93XX_PWMx_TERM_COUNT);
+			writew(duty_cycles, base + EP93XX_PWMx_DUTY_CYCLE);
+		} else {
+			writew(duty_cycles, base + EP93XX_PWMx_DUTY_CYCLE);
+			writew(period_cycles, base + EP93XX_PWMx_TERM_COUNT);
+		}
+	} else {
+		ret = -EINVAL;
+	}
+
+	if (!test_bit(PWMF_ENABLED, &pwm->flags))
+		clk_disable(ep93xx_pwm->clk);
+
+	return ret;
+}
+
+static int ep93xx_pwm_polarity(struct pwm_chip *chip, struct pwm_device *pwm,
+			       enum pwm_polarity polarity)
+{
+	struct ep93xx_pwm *ep93xx_pwm = to_ep93xx_pwm(chip);
+	int ret;
+
+	/*
+	 * The clock needs to be enabled to access the PWM registers.
+	 * Polarity can only be changed when the PWM is disabled.
+	 */
+	ret = clk_enable(ep93xx_pwm->clk);
+	if (ret)
+		return ret;
+
+	if (polarity == PWM_POLARITY_INVERSED)
+		writew(0x1, ep93xx_pwm->base + EP93XX_PWMx_INVERT);
+	else
+		writew(0x0, ep93xx_pwm->base + EP93XX_PWMx_INVERT);
+
+	clk_disable(ep93xx_pwm->clk);
+
+	return 0;
+}
+
+static int ep93xx_pwm_enable(struct pwm_chip *chip, struct pwm_device *pwm)
+{
+	struct ep93xx_pwm *ep93xx_pwm = to_ep93xx_pwm(chip);
+	int ret;
+
+	ret = clk_enable(ep93xx_pwm->clk);
+	if (ret)
+		return ret;
+
+	writew(0x1, ep93xx_pwm->base + EP93XX_PWMx_ENABLE);
+
+	return 0;
+}
+
+static void ep93xx_pwm_disable(struct pwm_chip *chip, struct pwm_device *pwm)
+{
+	struct ep93xx_pwm *ep93xx_pwm = to_ep93xx_pwm(chip);
+
+	writew(0x0, ep93xx_pwm->base + EP93XX_PWMx_ENABLE);
+	clk_disable(ep93xx_pwm->clk);
+}
+
+static const struct pwm_ops ep93xx_pwm_ops = {
+	.request = ep93xx_pwm_request,
+	.free = ep93xx_pwm_free,
+	.config = ep93xx_pwm_config,
+	.set_polarity = ep93xx_pwm_polarity,
+	.enable = ep93xx_pwm_enable,
+	.disable = ep93xx_pwm_disable,
+	.owner = THIS_MODULE,
+};
+
+static int ep93xx_pwm_probe(struct platform_device *pdev)
+{
+	struct ep93xx_pwm *ep93xx_pwm;
+	struct resource *res;
+	int ret;
+
+	ep93xx_pwm = devm_kzalloc(&pdev->dev, sizeof(*ep93xx_pwm), GFP_KERNEL);
+	if (!ep93xx_pwm)
+		return -ENOMEM;
+
+	res = platform_get_resource(pdev, IORESOURCE_MEM, 0);
+	ep93xx_pwm->base = devm_ioremap_resource(&pdev->dev, res);
+	if (IS_ERR(ep93xx_pwm->base))
+		return PTR_ERR(ep93xx_pwm->base);
+
+	ep93xx_pwm->clk = devm_clk_get(&pdev->dev, "pwm_clk");
+	if (IS_ERR(ep93xx_pwm->clk))
+		return PTR_ERR(ep93xx_pwm->clk);
+
+	ep93xx_pwm->chip.dev = &pdev->dev;
+	ep93xx_pwm->chip.ops = &ep93xx_pwm_ops;
+	ep93xx_pwm->chip.base = -1;
+	ep93xx_pwm->chip.npwm = 1;
+
+	ret = pwmchip_add(&ep93xx_pwm->chip);
+	if (ret < 0)
+		return ret;
+
+	platform_set_drvdata(pdev, ep93xx_pwm);
+	return 0;
+}
+
+static int ep93xx_pwm_remove(struct platform_device *pdev)
+{
+	struct ep93xx_pwm *ep93xx_pwm = platform_get_drvdata(pdev);
+
+	return pwmchip_remove(&ep93xx_pwm->chip);
+}
+
+static struct platform_driver ep93xx_pwm_driver = {
+	.driver = {
+		.name = "ep93xx-pwm",
+	},
+	.probe = ep93xx_pwm_probe,
+	.remove = ep93xx_pwm_remove,
+};
+module_platform_driver(ep93xx_pwm_driver);
+
+MODULE_DESCRIPTION("Cirrus Logic EP93xx PWM driver");
+MODULE_AUTHOR("Matthieu Crapet <mcrapet@gmail.com>, "
+	      "H Hartley Sweeten <hsweeten@visionengravers.com>");
+MODULE_ALIAS("platform:ep93xx-pwm");
+MODULE_LICENSE("GPL");
diff --git a/drivers/pwm/pwm-imx.c b/drivers/pwm/pwm-imx.c
index 2b7c4f88b461..cc4773344874 100644
--- a/drivers/pwm/pwm-imx.c
+++ b/drivers/pwm/pwm-imx.c
@@ -16,6 +16,7 @@
 #include <linux/clk.h>
 #include <linux/io.h>
 #include <linux/pwm.h>
+#include <linux/of.h>
 #include <linux/of_device.h>
 
 /* i.MX1 and i.MX21 share the same PWM function block: */
@@ -296,7 +297,7 @@ static struct platform_driver imx_pwm_driver = {
 	.driver		= {
 		.name	= "imx-pwm",
 		.owner = THIS_MODULE,
-		.of_match_table = of_match_ptr(imx_pwm_dt_ids),
+		.of_match_table = imx_pwm_dt_ids,
 	},
 	.probe		= imx_pwm_probe,
 	.remove		= imx_pwm_remove,
diff --git a/drivers/pwm/pwm-lpc32xx.c b/drivers/pwm/pwm-lpc32xx.c
index efac99e03d57..9dc0f9d42bfa 100644
--- a/drivers/pwm/pwm-lpc32xx.c
+++ b/drivers/pwm/pwm-lpc32xx.c
@@ -169,7 +169,7 @@ static struct platform_driver lpc32xx_pwm_driver = {
 	.driver = {
 		.name = "lpc32xx-pwm",
 		.owner = THIS_MODULE,
-		.of_match_table = of_match_ptr(lpc32xx_pwm_dt_ids),
+		.of_match_table = lpc32xx_pwm_dt_ids,
 	},
 	.probe = lpc32xx_pwm_probe,
 	.remove = lpc32xx_pwm_remove,
diff --git a/drivers/pwm/pwm-mxs.c b/drivers/pwm/pwm-mxs.c
index c2c5a4fd1b96..9475bc7a6f97 100644
--- a/drivers/pwm/pwm-mxs.c
+++ b/drivers/pwm/pwm-mxs.c
@@ -189,7 +189,7 @@ static struct platform_driver mxs_pwm_driver = {
 	.driver = {
 		.name = "mxs-pwm",
 		.owner = THIS_MODULE,
-		.of_match_table = of_match_ptr(mxs_pwm_dt_ids),
+		.of_match_table = mxs_pwm_dt_ids,
 	},
 	.probe = mxs_pwm_probe,
 	.remove = mxs_pwm_remove,
diff --git a/drivers/pwm/pwm-samsung.c b/drivers/pwm/pwm-samsung.c
index fcc8b9adde9f..b59639e0c029 100644
--- a/drivers/pwm/pwm-samsung.c
+++ b/drivers/pwm/pwm-samsung.c
@@ -18,6 +18,7 @@
 #include <linux/io.h>
 #include <linux/kernel.h>
 #include <linux/module.h>
+#include <linux/of.h>
 #include <linux/platform_device.h>
 #include <linux/pwm.h>
 #include <linux/slab.h>
@@ -224,8 +225,8 @@ static int pwm_samsung_request(struct pwm_chip *chip, struct pwm_device *pwm)
 
 static void pwm_samsung_free(struct pwm_chip *chip, struct pwm_device *pwm)
 {
-	pwm_set_chip_data(pwm, NULL);
 	devm_kfree(chip->dev, pwm_get_chip_data(pwm));
+	pwm_set_chip_data(pwm, NULL);
 }
 
 static int pwm_samsung_enable(struct pwm_chip *chip, struct pwm_device *pwm)
diff --git a/drivers/pwm/pwm-tiecap.c b/drivers/pwm/pwm-tiecap.c
index c2e2e5852362..4e5c3d13d4f8 100644
--- a/drivers/pwm/pwm-tiecap.c
+++ b/drivers/pwm/pwm-tiecap.c
@@ -26,7 +26,6 @@
 #include <linux/pm_runtime.h>
 #include <linux/pwm.h>
 #include <linux/of_device.h>
-#include <linux/pinctrl/consumer.h>
 
 #include "pwm-tipwmss.h"
 
@@ -208,11 +207,6 @@ static int ecap_pwm_probe(struct platform_device *pdev)
 	struct clk *clk;
 	struct ecap_pwm_chip *pc;
 	u16 status;
-	struct pinctrl *pinctrl;
-
-	pinctrl = devm_pinctrl_get_select_default(&pdev->dev);
-	if (IS_ERR(pinctrl))
-		dev_warn(&pdev->dev, "unable to select pin group\n");
 
 	pc = devm_kzalloc(&pdev->dev, sizeof(*pc), GFP_KERNEL);
 	if (!pc) {
diff --git a/drivers/pwm/pwm-tiehrpwm.c b/drivers/pwm/pwm-tiehrpwm.c
index 084f55246532..a4d8f519d965 100644
--- a/drivers/pwm/pwm-tiehrpwm.c
+++ b/drivers/pwm/pwm-tiehrpwm.c
@@ -26,7 +26,6 @@
 #include <linux/clk.h>
 #include <linux/pm_runtime.h>
 #include <linux/of_device.h>
-#include <linux/pinctrl/consumer.h>
 
 #include "pwm-tipwmss.h"
 
@@ -439,11 +438,6 @@ static int ehrpwm_pwm_probe(struct platform_device *pdev)
 	struct clk *clk;
 	struct ehrpwm_pwm_chip *pc;
 	u16 status;
-	struct pinctrl *pinctrl;
-
-	pinctrl = devm_pinctrl_get_select_default(&pdev->dev);
-	if (IS_ERR(pinctrl))
-		dev_warn(&pdev->dev, "unable to select pin group\n");
 
 	pc = devm_kzalloc(&pdev->dev, sizeof(*pc), GFP_KERNEL);
 	if (!pc) {
diff --git a/drivers/pwm/pwm-twl-led.c b/drivers/pwm/pwm-twl-led.c
index 29d1bba4804e..b964470025c5 100644
--- a/drivers/pwm/pwm-twl-led.c
+++ b/drivers/pwm/pwm-twl-led.c
@@ -21,6 +21,7 @@
  */
 
 #include <linux/module.h>
+#include <linux/of.h>
 #include <linux/platform_device.h>
 #include <linux/pwm.h>
 #include <linux/i2c/twl.h>
diff --git a/drivers/pwm/pwm-twl.c b/drivers/pwm/pwm-twl.c
index eef910580eae..b99a50e626a6 100644
--- a/drivers/pwm/pwm-twl.c
+++ b/drivers/pwm/pwm-twl.c
@@ -18,6 +18,7 @@
  */
 
 #include <linux/module.h>
+#include <linux/of.h>
 #include <linux/platform_device.h>
 #include <linux/pwm.h>
 #include <linux/i2c/twl.h>
diff --git a/drivers/regulator/tps65910-regulator.c b/drivers/regulator/tps65910-regulator.c
index 23f8d1ce877d..a00132e31ec7 100644
--- a/drivers/regulator/tps65910-regulator.c
+++ b/drivers/regulator/tps65910-regulator.c
@@ -906,7 +906,7 @@ static int tps65910_set_ext_sleep_config(struct tps65910_reg *pmic,
 		}
 		ret = tps65910_reg_write(pmic->mfd, sr_reg_add, 0);
 		if (ret < 0) {
-			dev_err(mfd->dev, "Error in settting sr register\n");
+			dev_err(mfd->dev, "Error in setting sr register\n");
 			return ret;
 		}
 	}
diff --git a/drivers/remoteproc/remoteproc_virtio.c b/drivers/remoteproc/remoteproc_virtio.c
index b09c75c21b60..a34b50690b4e 100644
--- a/drivers/remoteproc/remoteproc_virtio.c
+++ b/drivers/remoteproc/remoteproc_virtio.c
@@ -30,7 +30,7 @@
 #include "remoteproc_internal.h"
 
 /* kick the remote processor, and let it know which virtqueue to poke at */
-static void rproc_virtio_notify(struct virtqueue *vq)
+static bool rproc_virtio_notify(struct virtqueue *vq)
 {
 	struct rproc_vring *rvring = vq->priv;
 	struct rproc *rproc = rvring->rvdev->rproc;
@@ -39,6 +39,7 @@ static void rproc_virtio_notify(struct virtqueue *vq)
 	dev_dbg(&rproc->dev, "kicking vq index: %d\n", notifyid);
 
 	rproc->ops->kick(rproc, notifyid);
+	return true;
 }
 
 /**
diff --git a/drivers/rtc/rtc-hid-sensor-time.c b/drivers/rtc/rtc-hid-sensor-time.c
index 45560ffb038d..965a9da70867 100644
--- a/drivers/rtc/rtc-hid-sensor-time.c
+++ b/drivers/rtc/rtc-hid-sensor-time.c
@@ -209,7 +209,7 @@ static int hid_rtc_read_time(struct device *dev, struct rtc_time *tm)
 		platform_get_drvdata(to_platform_device(dev));
 	int ret;
 
-	INIT_COMPLETION(time_state->comp_last_time);
+	reinit_completion(&time_state->comp_last_time);
 	/* get a report with all values through requesting one value */
 	sensor_hub_input_attr_get_raw_value(time_state->common_attributes.hsdev,
 			HID_USAGE_SENSOR_TIME, hid_time_addresses[0],
@@ -236,7 +236,7 @@ static const struct rtc_class_ops hid_time_rtc_ops = {
 static int hid_time_probe(struct platform_device *pdev)
 {
 	int ret = 0;
-	struct hid_sensor_hub_device *hsdev = pdev->dev.platform_data;
+	struct hid_sensor_hub_device *hsdev = dev_get_platdata(&pdev->dev);
 	struct hid_time_state *time_state = devm_kzalloc(&pdev->dev,
 		sizeof(struct hid_time_state), GFP_KERNEL);
 
@@ -281,11 +281,18 @@ static int hid_time_probe(struct platform_device *pdev)
 		goto err_open;
 	}
 
+	/*
+	 * Enable HID input processing early in order to be able to read the
+	 * clock already in devm_rtc_device_register().
+	 */
+	hid_device_io_start(hsdev->hdev);
+
 	time_state->rtc = devm_rtc_device_register(&pdev->dev,
 					"hid-sensor-time", &hid_time_rtc_ops,
 					THIS_MODULE);
 
 	if (IS_ERR_OR_NULL(time_state->rtc)) {
+		hid_device_io_stop(hsdev->hdev);
 		ret = time_state->rtc ? PTR_ERR(time_state->rtc) : -ENODEV;
 		time_state->rtc = NULL;
 		dev_err(&pdev->dev, "rtc device register failed!\n");
@@ -303,7 +310,7 @@ err_open:
 
 static int hid_time_remove(struct platform_device *pdev)
 {
-	struct hid_sensor_hub_device *hsdev = pdev->dev.platform_data;
+	struct hid_sensor_hub_device *hsdev = dev_get_platdata(&pdev->dev);
 
 	sensor_hub_device_close(hsdev);
 	sensor_hub_remove_callback(hsdev, HID_USAGE_SENSOR_TIME);
diff --git a/drivers/s390/block/scm_blk.c b/drivers/s390/block/scm_blk.c
index 548209a9c43c..d0ab5019d885 100644
--- a/drivers/s390/block/scm_blk.c
+++ b/drivers/s390/block/scm_blk.c
@@ -118,22 +118,6 @@ static void scm_request_done(struct scm_request *scmrq)
 	spin_unlock_irqrestore(&list_lock, flags);
 }
 
-static int scm_open(struct block_device *blkdev, fmode_t mode)
-{
-	return scm_get_ref();
-}
-
-static void scm_release(struct gendisk *gendisk, fmode_t mode)
-{
-	scm_put_ref();
-}
-
-static const struct block_device_operations scm_blk_devops = {
-	.owner = THIS_MODULE,
-	.open = scm_open,
-	.release = scm_release,
-};
-
 static bool scm_permit_request(struct scm_blk_dev *bdev, struct request *req)
 {
 	return rq_data_dir(req) != WRITE || bdev->state != SCM_WR_PROHIBIT;
@@ -256,7 +240,7 @@ static void scm_blk_request(struct request_queue *rq)
 		atomic_inc(&bdev->queued_reqs);
 		blk_start_request(req);
 
-		ret = scm_start_aob(scmrq->aob);
+		ret = eadm_start_aob(scmrq->aob);
 		if (ret) {
 			SCM_LOG(5, "no subchannel");
 			scm_request_requeue(scmrq);
@@ -320,7 +304,7 @@ static void scm_blk_handle_error(struct scm_request *scmrq)
 	}
 
 restart:
-	if (!scm_start_aob(scmrq->aob))
+	if (!eadm_start_aob(scmrq->aob))
 		return;
 
 requeue:
@@ -363,6 +347,10 @@ static void scm_blk_tasklet(struct scm_blk_dev *bdev)
 	blk_run_queue(bdev->rq);
 }
 
+static const struct block_device_operations scm_blk_devops = {
+	.owner = THIS_MODULE,
+};
+
 int scm_blk_dev_setup(struct scm_blk_dev *bdev, struct scm_device *scmdev)
 {
 	struct request_queue *rq;
diff --git a/drivers/s390/block/scm_blk_cluster.c b/drivers/s390/block/scm_blk_cluster.c
index c0d102e3a48b..27f930cd657f 100644
--- a/drivers/s390/block/scm_blk_cluster.c
+++ b/drivers/s390/block/scm_blk_cluster.c
@@ -187,7 +187,7 @@ bool scm_need_cluster_request(struct scm_request *scmrq)
 void scm_initiate_cluster_request(struct scm_request *scmrq)
 {
 	scm_prepare_cluster_request(scmrq);
-	if (scm_start_aob(scmrq->aob))
+	if (eadm_start_aob(scmrq->aob))
 		scm_request_requeue(scmrq);
 }
 
diff --git a/drivers/s390/char/Makefile b/drivers/s390/char/Makefile
index 17821a026c9c..b69ab17f13fa 100644
--- a/drivers/s390/char/Makefile
+++ b/drivers/s390/char/Makefile
@@ -3,7 +3,8 @@
 #
 
 obj-y += ctrlchar.o keyboard.o defkeymap.o sclp.o sclp_rw.o sclp_quiesce.o \
-	 sclp_cmd.o sclp_config.o sclp_cpi_sys.o sclp_ocf.o sclp_ctl.o
+	 sclp_cmd.o sclp_config.o sclp_cpi_sys.o sclp_ocf.o sclp_ctl.o \
+	 sclp_early.o
 
 obj-$(CONFIG_TN3270) += raw3270.o
 obj-$(CONFIG_TN3270_CONSOLE) += con3270.o
diff --git a/drivers/s390/char/fs3270.c b/drivers/s390/char/fs3270.c
index f93cc32eb818..71e974738014 100644
--- a/drivers/s390/char/fs3270.c
+++ b/drivers/s390/char/fs3270.c
@@ -564,6 +564,7 @@ static void __exit
 fs3270_exit(void)
 {
 	raw3270_unregister_notifier(&fs3270_notifier);
+	device_destroy(class3270, MKDEV(IBM_FS3270_MAJOR, 0));
 	__unregister_chrdev(IBM_FS3270_MAJOR, 0, 1, "fs3270");
 }
 
diff --git a/drivers/s390/char/sclp.h b/drivers/s390/char/sclp.h
index 40d1406289ed..6fbe09686d18 100644
--- a/drivers/s390/char/sclp.h
+++ b/drivers/s390/char/sclp.h
@@ -99,6 +99,7 @@ struct init_sccb {
 } __attribute__((packed));
 
 extern u64 sclp_facilities;
+
 #define SCLP_HAS_CHP_INFO	(sclp_facilities & 0x8000000000000000ULL)
 #define SCLP_HAS_CHP_RECONFIG	(sclp_facilities & 0x2000000000000000ULL)
 #define SCLP_HAS_CPU_INFO	(sclp_facilities & 0x0800000000000000ULL)
@@ -179,6 +180,10 @@ void sclp_sdias_exit(void);
 extern int sclp_console_pages;
 extern int sclp_console_drop;
 extern unsigned long sclp_console_full;
+extern u8 sclp_fac84;
+extern unsigned long long sclp_rzm;
+extern unsigned long long sclp_rnmax;
+extern __initdata int sclp_early_read_info_sccb_valid;
 
 /* useful inlines */
 
diff --git a/drivers/s390/char/sclp_cmd.c b/drivers/s390/char/sclp_cmd.c
index 77df9cb00688..eaa21d542c5c 100644
--- a/drivers/s390/char/sclp_cmd.c
+++ b/drivers/s390/char/sclp_cmd.c
@@ -28,168 +28,6 @@
 
 #include "sclp.h"
 
-#define SCLP_CMDW_READ_SCP_INFO		0x00020001
-#define SCLP_CMDW_READ_SCP_INFO_FORCED	0x00120001
-
-struct read_info_sccb {
-	struct	sccb_header header;	/* 0-7 */
-	u16	rnmax;			/* 8-9 */
-	u8	rnsize;			/* 10 */
-	u8	_reserved0[24 - 11];	/* 11-15 */
-	u8	loadparm[8];		/* 24-31 */
-	u8	_reserved1[48 - 32];	/* 32-47 */
-	u64	facilities;		/* 48-55 */
-	u8	_reserved2[84 - 56];	/* 56-83 */
-	u8	fac84;			/* 84 */
-	u8	fac85;			/* 85 */
-	u8	_reserved3[91 - 86];	/* 86-90 */
-	u8	flags;			/* 91 */
-	u8	_reserved4[100 - 92];	/* 92-99 */
-	u32	rnsize2;		/* 100-103 */
-	u64	rnmax2;			/* 104-111 */
-	u8	_reserved5[4096 - 112];	/* 112-4095 */
-} __attribute__((packed, aligned(PAGE_SIZE)));
-
-static struct init_sccb __initdata early_event_mask_sccb __aligned(PAGE_SIZE);
-static struct read_info_sccb __initdata early_read_info_sccb;
-static int __initdata early_read_info_sccb_valid;
-
-u64 sclp_facilities;
-static u8 sclp_fac84;
-static unsigned long long rzm;
-static unsigned long long rnmax;
-
-static int __init sclp_cmd_sync_early(sclp_cmdw_t cmd, void *sccb)
-{
-	int rc;
-
-	__ctl_set_bit(0, 9);
-	rc = sclp_service_call(cmd, sccb);
-	if (rc)
-		goto out;
-	__load_psw_mask(PSW_DEFAULT_KEY | PSW_MASK_BASE | PSW_MASK_EA |
-			PSW_MASK_BA | PSW_MASK_EXT | PSW_MASK_WAIT);
-	local_irq_disable();
-out:
-	/* Contents of the sccb might have changed. */
-	barrier();
-	__ctl_clear_bit(0, 9);
-	return rc;
-}
-
-static void __init sclp_read_info_early(void)
-{
-	int rc;
-	int i;
-	struct read_info_sccb *sccb;
-	sclp_cmdw_t commands[] = {SCLP_CMDW_READ_SCP_INFO_FORCED,
-				  SCLP_CMDW_READ_SCP_INFO};
-
-	sccb = &early_read_info_sccb;
-	for (i = 0; i < ARRAY_SIZE(commands); i++) {
-		do {
-			memset(sccb, 0, sizeof(*sccb));
-			sccb->header.length = sizeof(*sccb);
-			sccb->header.function_code = 0x80;
-			sccb->header.control_mask[2] = 0x80;
-			rc = sclp_cmd_sync_early(commands[i], sccb);
-		} while (rc == -EBUSY);
-
-		if (rc)
-			break;
-		if (sccb->header.response_code == 0x10) {
-			early_read_info_sccb_valid = 1;
-			break;
-		}
-		if (sccb->header.response_code != 0x1f0)
-			break;
-	}
-}
-
-static void __init sclp_event_mask_early(void)
-{
-	struct init_sccb *sccb = &early_event_mask_sccb;
-	int rc;
-
-	do {
-		memset(sccb, 0, sizeof(*sccb));
-		sccb->header.length = sizeof(*sccb);
-		sccb->mask_length = sizeof(sccb_mask_t);
-		rc = sclp_cmd_sync_early(SCLP_CMDW_WRITE_EVENT_MASK, sccb);
-	} while (rc == -EBUSY);
-}
-
-void __init sclp_facilities_detect(void)
-{
-	struct read_info_sccb *sccb;
-
-	sclp_read_info_early();
-	if (!early_read_info_sccb_valid)
-		return;
-
-	sccb = &early_read_info_sccb;
-	sclp_facilities = sccb->facilities;
-	sclp_fac84 = sccb->fac84;
-	if (sccb->fac85 & 0x02)
-		S390_lowcore.machine_flags |= MACHINE_FLAG_ESOP;
-	rnmax = sccb->rnmax ? sccb->rnmax : sccb->rnmax2;
-	rzm = sccb->rnsize ? sccb->rnsize : sccb->rnsize2;
-	rzm <<= 20;
-
-	sclp_event_mask_early();
-}
-
-bool __init sclp_has_linemode(void)
-{
-	struct init_sccb *sccb = &early_event_mask_sccb;
-
-	if (sccb->header.response_code != 0x20)
-		return 0;
-	if (!(sccb->sclp_send_mask & (EVTYP_OPCMD_MASK | EVTYP_PMSGCMD_MASK)))
-		return 0;
-	if (!(sccb->sclp_receive_mask & (EVTYP_MSG_MASK | EVTYP_PMSGCMD_MASK)))
-		return 0;
-	return 1;
-}
-
-bool __init sclp_has_vt220(void)
-{
-	struct init_sccb *sccb = &early_event_mask_sccb;
-
-	if (sccb->header.response_code != 0x20)
-		return 0;
-	if (sccb->sclp_send_mask & EVTYP_VT220MSG_MASK)
-		return 1;
-	return 0;
-}
-
-unsigned long long sclp_get_rnmax(void)
-{
-	return rnmax;
-}
-
-unsigned long long sclp_get_rzm(void)
-{
-	return rzm;
-}
-
-/*
- * This function will be called after sclp_facilities_detect(), which gets
- * called from early.c code. Therefore the sccb should have valid contents.
- */
-void __init sclp_get_ipl_info(struct sclp_ipl_info *info)
-{
-	struct read_info_sccb *sccb;
-
-	if (!early_read_info_sccb_valid)
-		return;
-	sccb = &early_read_info_sccb;
-	info->is_valid = 1;
-	if (sccb->flags & 0x2)
-		info->has_dump = 1;
-	memcpy(&info->loadparm, &sccb->loadparm, LOADPARM_LEN);
-}
-
 static void sclp_sync_callback(struct sclp_req *req, void *data)
 {
 	struct completion *completion = data;
@@ -356,14 +194,14 @@ struct assign_storage_sccb {
 
 int arch_get_memory_phys_device(unsigned long start_pfn)
 {
-	if (!rzm)
+	if (!sclp_rzm)
 		return 0;
-	return PFN_PHYS(start_pfn) >> ilog2(rzm);
+	return PFN_PHYS(start_pfn) >> ilog2(sclp_rzm);
 }
 
 static unsigned long long rn2addr(u16 rn)
 {
-	return (unsigned long long) (rn - 1) * rzm;
+	return (unsigned long long) (rn - 1) * sclp_rzm;
 }
 
 static int do_assign_storage(sclp_cmdw_t cmd, u16 rn)
@@ -404,7 +242,7 @@ static int sclp_assign_storage(u16 rn)
 	if (rc)
 		return rc;
 	start = rn2addr(rn);
-	storage_key_init_range(start, start + rzm);
+	storage_key_init_range(start, start + sclp_rzm);
 	return 0;
 }
 
@@ -462,7 +300,7 @@ static int sclp_mem_change_state(unsigned long start, unsigned long size,
 		istart = rn2addr(incr->rn);
 		if (start + size - 1 < istart)
 			break;
-		if (start > istart + rzm - 1)
+		if (start > istart + sclp_rzm - 1)
 			continue;
 		if (online)
 			rc |= sclp_assign_storage(incr->rn);
@@ -526,7 +364,7 @@ static void __init add_memory_merged(u16 rn)
 	if (!first_rn)
 		goto skip_add;
 	start = rn2addr(first_rn);
-	size = (unsigned long long ) num * rzm;
+	size = (unsigned long long) num * sclp_rzm;
 	if (start >= VMEM_MAX_PHYS)
 		goto skip_add;
 	if (start + size > VMEM_MAX_PHYS)
@@ -574,7 +412,7 @@ static void __init insert_increment(u16 rn, int standby, int assigned)
 	}
 	if (!assigned)
 		new_incr->rn = last_rn + 1;
-	if (new_incr->rn > rnmax) {
+	if (new_incr->rn > sclp_rnmax) {
 		kfree(new_incr);
 		return;
 	}
@@ -617,7 +455,7 @@ static int __init sclp_detect_standby_memory(void)
 
 	if (OLDMEM_BASE) /* No standby memory in kdump mode */
 		return 0;
-	if (!early_read_info_sccb_valid)
+	if (!sclp_early_read_info_sccb_valid)
 		return 0;
 	if ((sclp_facilities & 0xe00000000000ULL) != 0xe00000000000ULL)
 		return 0;
@@ -661,7 +499,7 @@ static int __init sclp_detect_standby_memory(void)
 	}
 	if (rc || list_empty(&sclp_mem_list))
 		goto out;
-	for (i = 1; i <= rnmax - assigned; i++)
+	for (i = 1; i <= sclp_rnmax - assigned; i++)
 		insert_increment(0, 1, 0);
 	rc = register_memory_notifier(&sclp_mem_nb);
 	if (rc)
diff --git a/drivers/s390/char/sclp_early.c b/drivers/s390/char/sclp_early.c
new file mode 100644
index 000000000000..f7aa080e9b28
--- /dev/null
+++ b/drivers/s390/char/sclp_early.c
@@ -0,0 +1,264 @@
+/*
+ * SCLP early driver
+ *
+ * Copyright IBM Corp. 2013
+ */
+
+#define KMSG_COMPONENT "sclp_early"
+#define pr_fmt(fmt) KMSG_COMPONENT ": " fmt
+
+#include <asm/ctl_reg.h>
+#include <asm/sclp.h>
+#include <asm/ipl.h>
+#include "sclp_sdias.h"
+#include "sclp.h"
+
+#define SCLP_CMDW_READ_SCP_INFO		0x00020001
+#define SCLP_CMDW_READ_SCP_INFO_FORCED	0x00120001
+
+struct read_info_sccb {
+	struct	sccb_header header;	/* 0-7 */
+	u16	rnmax;			/* 8-9 */
+	u8	rnsize;			/* 10 */
+	u8	_reserved0[24 - 11];	/* 11-15 */
+	u8	loadparm[8];		/* 24-31 */
+	u8	_reserved1[48 - 32];	/* 32-47 */
+	u64	facilities;		/* 48-55 */
+	u8	_reserved2[84 - 56];	/* 56-83 */
+	u8	fac84;			/* 84 */
+	u8	fac85;			/* 85 */
+	u8	_reserved3[91 - 86];	/* 86-90 */
+	u8	flags;			/* 91 */
+	u8	_reserved4[100 - 92];	/* 92-99 */
+	u32	rnsize2;		/* 100-103 */
+	u64	rnmax2;			/* 104-111 */
+	u8	_reserved5[4096 - 112];	/* 112-4095 */
+} __packed __aligned(PAGE_SIZE);
+
+static __initdata struct init_sccb early_event_mask_sccb __aligned(PAGE_SIZE);
+static __initdata struct read_info_sccb early_read_info_sccb;
+static __initdata char sccb_early[PAGE_SIZE] __aligned(PAGE_SIZE);
+static unsigned long sclp_hsa_size;
+
+__initdata int sclp_early_read_info_sccb_valid;
+u64 sclp_facilities;
+u8 sclp_fac84;
+unsigned long long sclp_rzm;
+unsigned long long sclp_rnmax;
+
+static int __init sclp_cmd_sync_early(sclp_cmdw_t cmd, void *sccb)
+{
+	int rc;
+
+	__ctl_set_bit(0, 9);
+	rc = sclp_service_call(cmd, sccb);
+	if (rc)
+		goto out;
+	__load_psw_mask(PSW_DEFAULT_KEY | PSW_MASK_BASE | PSW_MASK_EA |
+			PSW_MASK_BA | PSW_MASK_EXT | PSW_MASK_WAIT);
+	local_irq_disable();
+out:
+	/* Contents of the sccb might have changed. */
+	barrier();
+	__ctl_clear_bit(0, 9);
+	return rc;
+}
+
+static void __init sclp_read_info_early(void)
+{
+	int rc;
+	int i;
+	struct read_info_sccb *sccb;
+	sclp_cmdw_t commands[] = {SCLP_CMDW_READ_SCP_INFO_FORCED,
+				  SCLP_CMDW_READ_SCP_INFO};
+
+	sccb = &early_read_info_sccb;
+	for (i = 0; i < ARRAY_SIZE(commands); i++) {
+		do {
+			memset(sccb, 0, sizeof(*sccb));
+			sccb->header.length = sizeof(*sccb);
+			sccb->header.function_code = 0x80;
+			sccb->header.control_mask[2] = 0x80;
+			rc = sclp_cmd_sync_early(commands[i], sccb);
+		} while (rc == -EBUSY);
+
+		if (rc)
+			break;
+		if (sccb->header.response_code == 0x10) {
+			sclp_early_read_info_sccb_valid = 1;
+			break;
+		}
+		if (sccb->header.response_code != 0x1f0)
+			break;
+	}
+}
+
+static void __init sclp_facilities_detect(void)
+{
+	struct read_info_sccb *sccb;
+
+	sclp_read_info_early();
+	if (!sclp_early_read_info_sccb_valid)
+		return;
+
+	sccb = &early_read_info_sccb;
+	sclp_facilities = sccb->facilities;
+	sclp_fac84 = sccb->fac84;
+	if (sccb->fac85 & 0x02)
+		S390_lowcore.machine_flags |= MACHINE_FLAG_ESOP;
+	sclp_rnmax = sccb->rnmax ? sccb->rnmax : sccb->rnmax2;
+	sclp_rzm = sccb->rnsize ? sccb->rnsize : sccb->rnsize2;
+	sclp_rzm <<= 20;
+}
+
+bool __init sclp_has_linemode(void)
+{
+	struct init_sccb *sccb = &early_event_mask_sccb;
+
+	if (sccb->header.response_code != 0x20)
+		return 0;
+	if (!(sccb->sclp_send_mask & (EVTYP_OPCMD_MASK | EVTYP_PMSGCMD_MASK)))
+		return 0;
+	if (!(sccb->sclp_receive_mask & (EVTYP_MSG_MASK | EVTYP_PMSGCMD_MASK)))
+		return 0;
+	return 1;
+}
+
+bool __init sclp_has_vt220(void)
+{
+	struct init_sccb *sccb = &early_event_mask_sccb;
+
+	if (sccb->header.response_code != 0x20)
+		return 0;
+	if (sccb->sclp_send_mask & EVTYP_VT220MSG_MASK)
+		return 1;
+	return 0;
+}
+
+unsigned long long sclp_get_rnmax(void)
+{
+	return sclp_rnmax;
+}
+
+unsigned long long sclp_get_rzm(void)
+{
+	return sclp_rzm;
+}
+
+/*
+ * This function will be called after sclp_facilities_detect(), which gets
+ * called from early.c code. Therefore the sccb should have valid contents.
+ */
+void __init sclp_get_ipl_info(struct sclp_ipl_info *info)
+{
+	struct read_info_sccb *sccb;
+
+	if (!sclp_early_read_info_sccb_valid)
+		return;
+	sccb = &early_read_info_sccb;
+	info->is_valid = 1;
+	if (sccb->flags & 0x2)
+		info->has_dump = 1;
+	memcpy(&info->loadparm, &sccb->loadparm, LOADPARM_LEN);
+}
+
+static int __init sclp_cmd_early(sclp_cmdw_t cmd, void *sccb)
+{
+	int rc;
+
+	do {
+		rc = sclp_cmd_sync_early(cmd, sccb);
+	} while (rc == -EBUSY);
+
+	if (rc)
+		return -EIO;
+	if (((struct sccb_header *) sccb)->response_code != 0x0020)
+		return -EIO;
+	return 0;
+}
+
+static void __init sccb_init_eq_size(struct sdias_sccb *sccb)
+{
+	memset(sccb, 0, sizeof(*sccb));
+
+	sccb->hdr.length = sizeof(*sccb);
+	sccb->evbuf.hdr.length = sizeof(struct sdias_evbuf);
+	sccb->evbuf.hdr.type = EVTYP_SDIAS;
+	sccb->evbuf.event_qual = SDIAS_EQ_SIZE;
+	sccb->evbuf.data_id = SDIAS_DI_FCP_DUMP;
+	sccb->evbuf.event_id = 4712;
+	sccb->evbuf.dbs = 1;
+}
+
+static int __init sclp_set_event_mask(unsigned long receive_mask,
+				      unsigned long send_mask)
+{
+	struct init_sccb *sccb = (void *) &sccb_early;
+
+	memset(sccb, 0, sizeof(*sccb));
+	sccb->header.length = sizeof(*sccb);
+	sccb->mask_length = sizeof(sccb_mask_t);
+	sccb->receive_mask = receive_mask;
+	sccb->send_mask = send_mask;
+	return sclp_cmd_early(SCLP_CMDW_WRITE_EVENT_MASK, sccb);
+}
+
+static long __init sclp_hsa_size_init(void)
+{
+	struct sdias_sccb *sccb = (void *) &sccb_early;
+
+	sccb_init_eq_size(sccb);
+	if (sclp_cmd_early(SCLP_CMDW_WRITE_EVENT_DATA, sccb))
+		return -EIO;
+	if (sccb->evbuf.blk_cnt != 0)
+		return (sccb->evbuf.blk_cnt - 1) * PAGE_SIZE;
+	return 0;
+}
+
+static long __init sclp_hsa_copy_wait(void)
+{
+	struct sccb_header *sccb = (void *) &sccb_early;
+
+	memset(sccb, 0, PAGE_SIZE);
+	sccb->length = PAGE_SIZE;
+	if (sclp_cmd_early(SCLP_CMDW_READ_EVENT_DATA, sccb))
+		return -EIO;
+	return (((struct sdias_sccb *) sccb)->evbuf.blk_cnt - 1) * PAGE_SIZE;
+}
+
+unsigned long sclp_get_hsa_size(void)
+{
+	return sclp_hsa_size;
+}
+
+static void __init sclp_hsa_size_detect(void)
+{
+	long size;
+
+	/* First try synchronous interface (LPAR) */
+	if (sclp_set_event_mask(0, 0x40000010))
+		return;
+	size = sclp_hsa_size_init();
+	if (size < 0)
+		return;
+	if (size != 0)
+		goto out;
+	/* Then try asynchronous interface (z/VM) */
+	if (sclp_set_event_mask(0x00000010, 0x40000010))
+		return;
+	size = sclp_hsa_size_init();
+	if (size < 0)
+		return;
+	size = sclp_hsa_copy_wait();
+	if (size < 0)
+		return;
+out:
+	sclp_hsa_size = size;
+}
+
+void __init sclp_early_detect(void)
+{
+	sclp_facilities_detect();
+	sclp_hsa_size_detect();
+	sclp_set_event_mask(0, 0);
+}
diff --git a/drivers/s390/char/sclp_sdias.c b/drivers/s390/char/sclp_sdias.c
index b1032931a1c4..561a0414b352 100644
--- a/drivers/s390/char/sclp_sdias.c
+++ b/drivers/s390/char/sclp_sdias.c
@@ -1,7 +1,7 @@
 /*
- * Sclp "store data in absolut storage"
+ * SCLP "store data in absolute storage"
  *
- * Copyright IBM Corp. 2003, 2007
+ * Copyright IBM Corp. 2003, 2013
  * Author(s): Michael Holzheu
  */
 
@@ -14,6 +14,7 @@
 #include <asm/debug.h>
 #include <asm/ipl.h>
 
+#include "sclp_sdias.h"
 #include "sclp.h"
 #include "sclp_rw.h"
 
@@ -22,46 +23,12 @@
 #define SDIAS_RETRIES 300
 #define SDIAS_SLEEP_TICKS 50
 
-#define EQ_STORE_DATA	0x0
-#define EQ_SIZE		0x1
-#define DI_FCP_DUMP	0x0
-#define ASA_SIZE_32	0x0
-#define ASA_SIZE_64	0x1
-#define EVSTATE_ALL_STORED	0x0
-#define EVSTATE_NO_DATA		0x3
-#define EVSTATE_PART_STORED	0x10
-
 static struct debug_info *sdias_dbf;
 
 static struct sclp_register sclp_sdias_register = {
 	.send_mask = EVTYP_SDIAS_MASK,
 };
 
-struct sdias_evbuf {
-	struct	evbuf_header hdr;
-	u8	event_qual;
-	u8	data_id;
-	u64	reserved2;
-	u32	event_id;
-	u16	reserved3;
-	u8	asa_size;
-	u8	event_status;
-	u32	reserved4;
-	u32	blk_cnt;
-	u64	asa;
-	u32	reserved5;
-	u32	fbn;
-	u32	reserved6;
-	u32	lbn;
-	u16	reserved7;
-	u16	dbs;
-} __attribute__((packed));
-
-struct sdias_sccb {
-	struct sccb_header  hdr;
-	struct sdias_evbuf  evbuf;
-} __attribute__((packed));
-
 static struct sdias_sccb sccb __attribute__((aligned(4096)));
 static struct sdias_evbuf sdias_evbuf;
 
@@ -148,8 +115,8 @@ int sclp_sdias_blk_count(void)
 	sccb.hdr.length = sizeof(sccb);
 	sccb.evbuf.hdr.length = sizeof(struct sdias_evbuf);
 	sccb.evbuf.hdr.type = EVTYP_SDIAS;
-	sccb.evbuf.event_qual = EQ_SIZE;
-	sccb.evbuf.data_id = DI_FCP_DUMP;
+	sccb.evbuf.event_qual = SDIAS_EQ_SIZE;
+	sccb.evbuf.data_id = SDIAS_DI_FCP_DUMP;
 	sccb.evbuf.event_id = 4712;
 	sccb.evbuf.dbs = 1;
 
@@ -208,13 +175,13 @@ int sclp_sdias_copy(void *dest, int start_blk, int nr_blks)
 	sccb.evbuf.hdr.length = sizeof(struct sdias_evbuf);
 	sccb.evbuf.hdr.type = EVTYP_SDIAS;
 	sccb.evbuf.hdr.flags = 0;
-	sccb.evbuf.event_qual = EQ_STORE_DATA;
-	sccb.evbuf.data_id = DI_FCP_DUMP;
+	sccb.evbuf.event_qual = SDIAS_EQ_STORE_DATA;
+	sccb.evbuf.data_id = SDIAS_DI_FCP_DUMP;
 	sccb.evbuf.event_id = 4712;
 #ifdef CONFIG_64BIT
-	sccb.evbuf.asa_size = ASA_SIZE_64;
+	sccb.evbuf.asa_size = SDIAS_ASA_SIZE_64;
 #else
-	sccb.evbuf.asa_size = ASA_SIZE_32;
+	sccb.evbuf.asa_size = SDIAS_ASA_SIZE_32;
 #endif
 	sccb.evbuf.event_status = 0;
 	sccb.evbuf.blk_cnt = nr_blks;
@@ -240,20 +207,19 @@ int sclp_sdias_copy(void *dest, int start_blk, int nr_blks)
 	}
 
 	switch (sdias_evbuf.event_status) {
-		case EVSTATE_ALL_STORED:
-			TRACE("all stored\n");
-			break;
-		case EVSTATE_PART_STORED:
-			TRACE("part stored: %i\n", sdias_evbuf.blk_cnt);
-			break;
-		case EVSTATE_NO_DATA:
-			TRACE("no data\n");
-			/* fall through */
-		default:
-			pr_err("Error from SCLP while copying hsa. "
-			       "Event status = %x\n",
-			       sdias_evbuf.event_status);
-			rc = -EIO;
+	case SDIAS_EVSTATE_ALL_STORED:
+		TRACE("all stored\n");
+		break;
+	case SDIAS_EVSTATE_PART_STORED:
+		TRACE("part stored: %i\n", sdias_evbuf.blk_cnt);
+		break;
+	case SDIAS_EVSTATE_NO_DATA:
+		TRACE("no data\n");
+		/* fall through */
+	default:
+		pr_err("Error from SCLP while copying hsa. Event status = %x\n",
+		       sdias_evbuf.event_status);
+		rc = -EIO;
 	}
 out:
 	mutex_unlock(&sdias_mutex);
diff --git a/drivers/s390/char/sclp_sdias.h b/drivers/s390/char/sclp_sdias.h
new file mode 100644
index 000000000000..f2431c414150
--- /dev/null
+++ b/drivers/s390/char/sclp_sdias.h
@@ -0,0 +1,46 @@
+/*
+ * SCLP "store data in absolute storage"
+ *
+ * Copyright IBM Corp. 2003, 2013
+ */
+
+#ifndef SCLP_SDIAS_H
+#define SCLP_SDIAS_H
+
+#include "sclp.h"
+
+#define SDIAS_EQ_STORE_DATA		0x0
+#define SDIAS_EQ_SIZE			0x1
+#define SDIAS_DI_FCP_DUMP		0x0
+#define SDIAS_ASA_SIZE_32		0x0
+#define SDIAS_ASA_SIZE_64		0x1
+#define SDIAS_EVSTATE_ALL_STORED	0x0
+#define SDIAS_EVSTATE_NO_DATA		0x3
+#define SDIAS_EVSTATE_PART_STORED	0x10
+
+struct sdias_evbuf {
+	struct	evbuf_header hdr;
+	u8	event_qual;
+	u8	data_id;
+	u64	reserved2;
+	u32	event_id;
+	u16	reserved3;
+	u8	asa_size;
+	u8	event_status;
+	u32	reserved4;
+	u32	blk_cnt;
+	u64	asa;
+	u32	reserved5;
+	u32	fbn;
+	u32	reserved6;
+	u32	lbn;
+	u16	reserved7;
+	u16	dbs;
+} __packed;
+
+struct sdias_sccb {
+	struct sccb_header	hdr;
+	struct sdias_evbuf	evbuf;
+} __packed;
+
+#endif /* SCLP_SDIAS_H */
diff --git a/drivers/s390/char/zcore.c b/drivers/s390/char/zcore.c
index ffb1fcf0bf5b..3d8e4d63f514 100644
--- a/drivers/s390/char/zcore.c
+++ b/drivers/s390/char/zcore.c
@@ -328,9 +328,9 @@ static ssize_t zcore_read(struct file *file, char __user *buf, size_t count,
 	mem_offs = 0;
 
 	/* Copy from HSA data */
-	if (*ppos < (ZFCPDUMP_HSA_SIZE + HEADER_SIZE)) {
-		size = min((count - hdr_count), (size_t) (ZFCPDUMP_HSA_SIZE
-			   - mem_start));
+	if (*ppos < sclp_get_hsa_size() + HEADER_SIZE) {
+		size = min((count - hdr_count),
+			   (size_t) (sclp_get_hsa_size() - mem_start));
 		rc = memcpy_hsa_user(buf + hdr_count, mem_start, size);
 		if (rc)
 			goto fail;
@@ -490,7 +490,7 @@ static ssize_t zcore_hsa_read(struct file *filp, char __user *buf,
 	static char str[18];
 
 	if (hsa_available)
-		snprintf(str, sizeof(str), "%lx\n", ZFCPDUMP_HSA_SIZE);
+		snprintf(str, sizeof(str), "%lx\n", sclp_get_hsa_size());
 	else
 		snprintf(str, sizeof(str), "0\n");
 	return simple_read_from_buffer(buf, count, ppos, str, strlen(str));
@@ -584,17 +584,9 @@ static int __init sys_info_init(enum arch_id arch, unsigned long mem_end)
 
 static int __init check_sdias(void)
 {
-	int rc, act_hsa_size;
-
-	rc = sclp_sdias_blk_count();
-	if (rc < 0) {
+	if (!sclp_get_hsa_size()) {
 		TRACE("Could not determine HSA size\n");
-		return rc;
-	}
-	act_hsa_size = (rc - 1) * PAGE_SIZE;
-	if (act_hsa_size < ZFCPDUMP_HSA_SIZE) {
-		TRACE("HSA size too small: %i\n", act_hsa_size);
-		return -EINVAL;
+		return -ENODEV;
 	}
 	return 0;
 }
@@ -662,7 +654,7 @@ static int __init zcore_reipl_init(void)
 	ipl_block = (void *) __get_free_page(GFP_KERNEL);
 	if (!ipl_block)
 		return -ENOMEM;
-	if (ipib_info.ipib < ZFCPDUMP_HSA_SIZE)
+	if (ipib_info.ipib < sclp_get_hsa_size())
 		rc = memcpy_hsa_kernel(ipl_block, ipib_info.ipib, PAGE_SIZE);
 	else
 		rc = memcpy_real(ipl_block, (void *) ipib_info.ipib, PAGE_SIZE);
diff --git a/drivers/s390/cio/eadm_sch.c b/drivers/s390/cio/eadm_sch.c
index aca7bfc113aa..3a2ee4a740b4 100644
--- a/drivers/s390/cio/eadm_sch.c
+++ b/drivers/s390/cio/eadm_sch.c
@@ -190,7 +190,7 @@ static struct subchannel *eadm_get_idle_sch(void)
 	return NULL;
 }
 
-static int eadm_start_aob(struct aob *aob)
+int eadm_start_aob(struct aob *aob)
 {
 	struct eadm_private *private;
 	struct subchannel *sch;
@@ -218,6 +218,7 @@ out_unlock:
 
 	return ret;
 }
+EXPORT_SYMBOL_GPL(eadm_start_aob);
 
 static int eadm_subchannel_probe(struct subchannel *sch)
 {
@@ -380,11 +381,6 @@ static struct css_driver eadm_subchannel_driver = {
 	.restore = eadm_subchannel_restore,
 };
 
-static struct eadm_ops eadm_ops = {
-	.eadm_start = eadm_start_aob,
-	.owner = THIS_MODULE,
-};
-
 static int __init eadm_sch_init(void)
 {
 	int ret;
@@ -404,7 +400,6 @@ static int __init eadm_sch_init(void)
 	if (ret)
 		goto cleanup;
 
-	register_eadm_ops(&eadm_ops);
 	return ret;
 
 cleanup:
@@ -415,7 +410,6 @@ cleanup:
 
 static void __exit eadm_sch_exit(void)
 {
-	unregister_eadm_ops(&eadm_ops);
 	css_driver_unregister(&eadm_subchannel_driver);
 	isc_unregister(EADM_SCH_ISC);
 	debug_unregister(eadm_debug);
diff --git a/drivers/s390/cio/scm.c b/drivers/s390/cio/scm.c
index 46ec25632e8b..15268edc54ae 100644
--- a/drivers/s390/cio/scm.c
+++ b/drivers/s390/cio/scm.c
@@ -15,8 +15,6 @@
 #include "chsc.h"
 
 static struct device *scm_root;
-static struct eadm_ops *eadm_ops;
-static DEFINE_MUTEX(eadm_ops_mutex);
 
 #define to_scm_dev(n) container_of(n, struct scm_device, dev)
 #define	to_scm_drv(d) container_of(d, struct scm_driver, drv)
@@ -73,49 +71,6 @@ void scm_driver_unregister(struct scm_driver *scmdrv)
 }
 EXPORT_SYMBOL_GPL(scm_driver_unregister);
 
-int scm_get_ref(void)
-{
-	int ret = 0;
-
-	mutex_lock(&eadm_ops_mutex);
-	if (!eadm_ops || !try_module_get(eadm_ops->owner))
-		ret = -ENOENT;
-	mutex_unlock(&eadm_ops_mutex);
-
-	return ret;
-}
-EXPORT_SYMBOL_GPL(scm_get_ref);
-
-void scm_put_ref(void)
-{
-	mutex_lock(&eadm_ops_mutex);
-	module_put(eadm_ops->owner);
-	mutex_unlock(&eadm_ops_mutex);
-}
-EXPORT_SYMBOL_GPL(scm_put_ref);
-
-void register_eadm_ops(struct eadm_ops *ops)
-{
-	mutex_lock(&eadm_ops_mutex);
-	eadm_ops = ops;
-	mutex_unlock(&eadm_ops_mutex);
-}
-EXPORT_SYMBOL_GPL(register_eadm_ops);
-
-void unregister_eadm_ops(struct eadm_ops *ops)
-{
-	mutex_lock(&eadm_ops_mutex);
-	eadm_ops = NULL;
-	mutex_unlock(&eadm_ops_mutex);
-}
-EXPORT_SYMBOL_GPL(unregister_eadm_ops);
-
-int scm_start_aob(struct aob *aob)
-{
-	return eadm_ops->eadm_start(aob);
-}
-EXPORT_SYMBOL_GPL(scm_start_aob);
-
 void scm_irq_handler(struct aob *aob, int error)
 {
 	struct aob_rq_header *aobrq = (void *) aob->request.data;
diff --git a/drivers/s390/kvm/kvm_virtio.c b/drivers/s390/kvm/kvm_virtio.c
index af2166fa5159..1abd0db29915 100644
--- a/drivers/s390/kvm/kvm_virtio.c
+++ b/drivers/s390/kvm/kvm_virtio.c
@@ -166,11 +166,15 @@ static void kvm_reset(struct virtio_device *vdev)
  * make a hypercall.  We hand the address  of the virtqueue so the Host
  * knows which virtqueue we're talking about.
  */
-static void kvm_notify(struct virtqueue *vq)
+static bool kvm_notify(struct virtqueue *vq)
 {
+	long rc;
 	struct kvm_vqconfig *config = vq->priv;
 
-	kvm_hypercall1(KVM_S390_VIRTIO_NOTIFY, config->address);
+	rc = kvm_hypercall1(KVM_S390_VIRTIO_NOTIFY, config->address);
+	if (rc < 0)
+		return false;
+	return true;
 }
 
 /*
diff --git a/drivers/s390/kvm/virtio_ccw.c b/drivers/s390/kvm/virtio_ccw.c
index 779dc5136291..d6297176ab85 100644
--- a/drivers/s390/kvm/virtio_ccw.c
+++ b/drivers/s390/kvm/virtio_ccw.c
@@ -162,7 +162,7 @@ static inline long do_kvm_notify(struct subchannel_id schid,
 	return __rc;
 }
 
-static void virtio_ccw_kvm_notify(struct virtqueue *vq)
+static bool virtio_ccw_kvm_notify(struct virtqueue *vq)
 {
 	struct virtio_ccw_vq_info *info = vq->priv;
 	struct virtio_ccw_device *vcdev;
@@ -171,6 +171,9 @@ static void virtio_ccw_kvm_notify(struct virtqueue *vq)
 	vcdev = to_vc_device(info->vq->vdev);
 	ccw_device_get_schid(vcdev->cdev, &schid);
 	info->cookie = do_kvm_notify(schid, vq->index, info->cookie);
+	if (info->cookie < 0)
+		return false;
+	return true;
 }
 
 static int virtio_ccw_read_vq_conf(struct virtio_ccw_device *vcdev,
diff --git a/drivers/s390/scsi/zfcp_dbf.c b/drivers/s390/scsi/zfcp_dbf.c
index 132a905b6bdb..0ca64484cfa3 100644
--- a/drivers/s390/scsi/zfcp_dbf.c
+++ b/drivers/s390/scsi/zfcp_dbf.c
@@ -344,7 +344,7 @@ void zfcp_dbf_san(char *tag, struct zfcp_dbf *dbf, void *data, u8 id, u16 len,
 
 /**
  * zfcp_dbf_san_req - trace event for issued SAN request
- * @tag: indentifier for event
+ * @tag: identifier for event
  * @fsf_req: request containing issued CT data
  * d_id: destination ID
  */
@@ -361,7 +361,7 @@ void zfcp_dbf_san_req(char *tag, struct zfcp_fsf_req *fsf, u32 d_id)
 
 /**
  * zfcp_dbf_san_res - trace event for received SAN request
- * @tag: indentifier for event
+ * @tag: identifier for event
  * @fsf_req: request containing issued CT data
  */
 void zfcp_dbf_san_res(char *tag, struct zfcp_fsf_req *fsf)
@@ -377,7 +377,7 @@ void zfcp_dbf_san_res(char *tag, struct zfcp_fsf_req *fsf)
 
 /**
  * zfcp_dbf_san_in_els - trace event for incoming ELS
- * @tag: indentifier for event
+ * @tag: identifier for event
  * @fsf_req: request containing issued CT data
  */
 void zfcp_dbf_san_in_els(char *tag, struct zfcp_fsf_req *fsf)
diff --git a/drivers/scsi/aacraid/commctrl.c b/drivers/scsi/aacraid/commctrl.c
index d85ac1a9d2c0..fbcd48d0bfc3 100644
--- a/drivers/scsi/aacraid/commctrl.c
+++ b/drivers/scsi/aacraid/commctrl.c
@@ -511,7 +511,8 @@ static int aac_send_raw_srb(struct aac_dev* dev, void __user * arg)
 		goto cleanup;
 	}
 
-	if (fibsize > (dev->max_fib_size - sizeof(struct aac_fibhdr))) {
+	if ((fibsize < (sizeof(struct user_aac_srb) - sizeof(struct user_sgentry))) ||
+	    (fibsize > (dev->max_fib_size - sizeof(struct aac_fibhdr)))) {
 		rcode = -EINVAL;
 		goto cleanup;
 	}
diff --git a/drivers/scsi/arcmsr/arcmsr_hba.c b/drivers/scsi/arcmsr/arcmsr_hba.c
index 33c52bc2c7b4..97fd450aff09 100644
--- a/drivers/scsi/arcmsr/arcmsr_hba.c
+++ b/drivers/scsi/arcmsr/arcmsr_hba.c
@@ -1035,7 +1035,6 @@ static void arcmsr_remove(struct pci_dev *pdev)
 	pci_release_regions(pdev);
 	scsi_host_put(host);
 	pci_disable_device(pdev);
-	pci_set_drvdata(pdev, NULL);
 }
 
 static void arcmsr_shutdown(struct pci_dev *pdev)
diff --git a/drivers/scsi/atp870u.c b/drivers/scsi/atp870u.c
index 15a629d8ed08..a795d81ef875 100644
--- a/drivers/scsi/atp870u.c
+++ b/drivers/scsi/atp870u.c
@@ -3144,8 +3144,6 @@ static void atp870u_remove (struct pci_dev *pdev)
 	atp870u_free_tables(pshost);
 	printk(KERN_INFO "scsi_host_put : %p\n",pshost);
 	scsi_host_put(pshost);
-	printk(KERN_INFO "pci_set_drvdata : %p\n",pdev);
-	pci_set_drvdata(pdev, NULL);	
 }
 MODULE_LICENSE("GPL");
 
diff --git a/drivers/scsi/bfa/bfad.c b/drivers/scsi/bfa/bfad.c
index 7591fa4e28bb..fc80a325a1e6 100644
--- a/drivers/scsi/bfa/bfad.c
+++ b/drivers/scsi/bfa/bfad.c
@@ -804,7 +804,6 @@ bfad_pci_uninit(struct pci_dev *pdev, struct bfad_s *bfad)
 	/* Disable PCIE Advanced Error Recovery (AER) */
 	pci_disable_pcie_error_reporting(pdev);
 	pci_disable_device(pdev);
-	pci_set_drvdata(pdev, NULL);
 }
 
 bfa_status_t
diff --git a/drivers/scsi/bnx2i/bnx2i_hwi.c b/drivers/scsi/bnx2i/bnx2i_hwi.c
index 5be718c241c4..e4cf23df4b4f 100644
--- a/drivers/scsi/bnx2i/bnx2i_hwi.c
+++ b/drivers/scsi/bnx2i/bnx2i_hwi.c
@@ -126,7 +126,7 @@ static void bnx2i_iscsi_license_error(struct bnx2i_hba *hba, u32 error_code)
 
 /**
  * bnx2i_arm_cq_event_coalescing - arms CQ to enable EQ notification
- * @ep:		endpoint (transport indentifier) structure
+ * @ep:		endpoint (transport identifier) structure
  * @action:	action, ARM or DISARM. For now only ARM_CQE is used
  *
  * Arm'ing CQ will enable chip to generate global EQ events inorder to interrupt
@@ -756,7 +756,7 @@ void bnx2i_send_cmd_cleanup_req(struct bnx2i_hba *hba, struct bnx2i_cmd *cmd)
 /**
  * bnx2i_send_conn_destroy - initiates iscsi connection teardown process
  * @hba:	adapter structure pointer
- * @ep:		endpoint (transport indentifier) structure
+ * @ep:		endpoint (transport identifier) structure
  *
  * this routine prepares and posts CONN_OFLD_REQ1/2 KWQE to initiate
  * 	iscsi connection context clean-up process
@@ -791,7 +791,7 @@ int bnx2i_send_conn_destroy(struct bnx2i_hba *hba, struct bnx2i_endpoint *ep)
 /**
  * bnx2i_570x_send_conn_ofld_req - initiates iscsi conn context setup process
  * @hba: 		adapter structure pointer
- * @ep: 		endpoint (transport indentifier) structure
+ * @ep: 		endpoint (transport identifier) structure
  *
  * 5706/5708/5709 specific - prepares and posts CONN_OFLD_REQ1/2 KWQE
  */
@@ -851,7 +851,7 @@ static int bnx2i_570x_send_conn_ofld_req(struct bnx2i_hba *hba,
 /**
  * bnx2i_5771x_send_conn_ofld_req - initiates iscsi connection context creation
  * @hba: 		adapter structure pointer
- * @ep: 		endpoint (transport indentifier) structure
+ * @ep: 		endpoint (transport identifier) structure
  *
  * 57710 specific - prepares and posts CONN_OFLD_REQ1/2 KWQE
  */
@@ -920,7 +920,7 @@ static int bnx2i_5771x_send_conn_ofld_req(struct bnx2i_hba *hba,
  * bnx2i_send_conn_ofld_req - initiates iscsi connection context setup process
  *
  * @hba: 		adapter structure pointer
- * @ep: 		endpoint (transport indentifier) structure
+ * @ep: 		endpoint (transport identifier) structure
  *
  * this routine prepares and posts CONN_OFLD_REQ1/2 KWQE
  */
@@ -939,7 +939,7 @@ int bnx2i_send_conn_ofld_req(struct bnx2i_hba *hba, struct bnx2i_endpoint *ep)
 
 /**
  * setup_qp_page_tables - iscsi QP page table setup function
- * @ep:		endpoint (transport indentifier) structure
+ * @ep:		endpoint (transport identifier) structure
  *
  * Sets up page tables for SQ/RQ/CQ, 1G/sec (5706/5708/5709) devices requires
  * 	64-bit address in big endian format. Whereas 10G/sec (57710) requires
@@ -1046,7 +1046,7 @@ static void setup_qp_page_tables(struct bnx2i_endpoint *ep)
 /**
  * bnx2i_alloc_qp_resc - allocates required resources for QP.
  * @hba:	adapter structure pointer
- * @ep:		endpoint (transport indentifier) structure
+ * @ep:		endpoint (transport identifier) structure
  *
  * Allocate QP (transport layer for iSCSI connection) resources, DMA'able
  *	memory for SQ/RQ/CQ and page tables. EP structure elements such
@@ -1191,7 +1191,7 @@ mem_alloc_err:
 /**
  * bnx2i_free_qp_resc - free memory resources held by QP
  * @hba:	adapter structure pointer
- * @ep:	endpoint (transport indentifier) structure
+ * @ep:	endpoint (transport identifier) structure
  *
  * Free QP resources - SQ/RQ/CQ memory and page tables.
  */
diff --git a/drivers/scsi/bnx2i/bnx2i_iscsi.c b/drivers/scsi/bnx2i/bnx2i_iscsi.c
index fabeb88602ac..854dad7d5b03 100644
--- a/drivers/scsi/bnx2i/bnx2i_iscsi.c
+++ b/drivers/scsi/bnx2i/bnx2i_iscsi.c
@@ -596,7 +596,7 @@ void bnx2i_drop_session(struct iscsi_cls_session *cls_session)
 /**
  * bnx2i_ep_destroy_list_add - add an entry to EP destroy list
  * @hba:	pointer to adapter instance
- * @ep:		pointer to endpoint (transport indentifier) structure
+ * @ep:		pointer to endpoint (transport identifier) structure
  *
  * EP destroy queue manager
  */
@@ -613,7 +613,7 @@ static int bnx2i_ep_destroy_list_add(struct bnx2i_hba *hba,
  * bnx2i_ep_destroy_list_del - add an entry to EP destroy list
  *
  * @hba: 		pointer to adapter instance
- * @ep: 		pointer to endpoint (transport indentifier) structure
+ * @ep: 		pointer to endpoint (transport identifier) structure
  *
  * EP destroy queue manager
  */
@@ -630,7 +630,7 @@ static int bnx2i_ep_destroy_list_del(struct bnx2i_hba *hba,
 /**
  * bnx2i_ep_ofld_list_add - add an entry to ep offload pending list
  * @hba:	pointer to adapter instance
- * @ep:		pointer to endpoint (transport indentifier) structure
+ * @ep:		pointer to endpoint (transport identifier) structure
  *
  * pending conn offload completion queue manager
  */
@@ -646,7 +646,7 @@ static int bnx2i_ep_ofld_list_add(struct bnx2i_hba *hba,
 /**
  * bnx2i_ep_ofld_list_del - add an entry to ep offload pending list
  * @hba: 		pointer to adapter instance
- * @ep: 		pointer to endpoint (transport indentifier) structure
+ * @ep: 		pointer to endpoint (transport identifier) structure
  *
  * pending conn offload completion queue manager
  */
@@ -721,7 +721,7 @@ bnx2i_find_ep_in_destroy_list(struct bnx2i_hba *hba, u32 iscsi_cid)
 /**
  * bnx2i_ep_active_list_add - add an entry to ep active list
  * @hba:	pointer to adapter instance
- * @ep:		pointer to endpoint (transport indentifier) structure
+ * @ep:		pointer to endpoint (transport identifier) structure
  *
  * current active conn queue manager
  */
@@ -737,7 +737,7 @@ static void bnx2i_ep_active_list_add(struct bnx2i_hba *hba,
 /**
  * bnx2i_ep_active_list_del - deletes an entry to ep active list
  * @hba:	pointer to adapter instance
- * @ep:		pointer to endpoint (transport indentifier) structure
+ * @ep:		pointer to endpoint (transport identifier) structure
  *
  * current active conn queue manager
  */
@@ -1695,7 +1695,7 @@ no_nx2_route:
 /**
  * bnx2i_tear_down_conn - tear down iscsi/tcp connection and free resources
  * @hba:	pointer to adapter instance
- * @ep:		endpoint (transport indentifier) structure
+ * @ep:		endpoint (transport identifier) structure
  *
  * destroys cm_sock structure and on chip iscsi context
  */
diff --git a/drivers/scsi/csiostor/csio_init.c b/drivers/scsi/csiostor/csio_init.c
index 00346fe939d5..1aafc331ee63 100644
--- a/drivers/scsi/csiostor/csio_init.c
+++ b/drivers/scsi/csiostor/csio_init.c
@@ -1010,7 +1010,6 @@ err_lnode_exit:
 	csio_hw_stop(hw);
 	spin_unlock_irq(&hw->lock);
 	csio_lnodes_unblock_request(hw);
-	pci_set_drvdata(hw->pdev, NULL);
 	csio_lnodes_exit(hw, 0);
 	csio_hw_free(hw);
 err_pci_exit:
@@ -1044,7 +1043,6 @@ static void csio_remove_one(struct pci_dev *pdev)
 
 	csio_lnodes_exit(hw, 0);
 	csio_hw_free(hw);
-	pci_set_drvdata(pdev, NULL);
 	csio_pci_exit(pdev, &bars);
 }
 
diff --git a/drivers/scsi/dc395x.c b/drivers/scsi/dc395x.c
index 42e8624a9b9a..83d9bf6fa6ca 100644
--- a/drivers/scsi/dc395x.c
+++ b/drivers/scsi/dc395x.c
@@ -4861,7 +4861,6 @@ static void dc395x_remove_one(struct pci_dev *dev)
 	adapter_uninit(acb);
 	pci_disable_device(dev);
 	scsi_host_put(scsi_host);
-	pci_set_drvdata(dev, NULL);
 }
 
 
diff --git a/drivers/scsi/fnic/fnic_main.c b/drivers/scsi/fnic/fnic_main.c
index be09b101b4a1..33e4ec2bfe73 100644
--- a/drivers/scsi/fnic/fnic_main.c
+++ b/drivers/scsi/fnic/fnic_main.c
@@ -1005,7 +1005,6 @@ static void fnic_remove(struct pci_dev *pdev)
 	fnic_iounmap(fnic);
 	pci_release_regions(pdev);
 	pci_disable_device(pdev);
-	pci_set_drvdata(pdev, NULL);
 	scsi_host_put(lp->host);
 }
 
diff --git a/drivers/scsi/gdth.c b/drivers/scsi/gdth.c
index 6d55b4e7e792..ee4fa40a50b1 100644
--- a/drivers/scsi/gdth.c
+++ b/drivers/scsi/gdth.c
@@ -594,8 +594,6 @@ static void gdth_pci_remove_one(struct pci_dev *pdev)
 {
 	gdth_ha_str *ha = pci_get_drvdata(pdev);
 
-	pci_set_drvdata(pdev, NULL);
-
 	list_del(&ha->list);
 	gdth_remove_one(ha);
 
diff --git a/drivers/scsi/hpsa.c b/drivers/scsi/hpsa.c
index fb5a89815150..22f6432eb475 100644
--- a/drivers/scsi/hpsa.c
+++ b/drivers/scsi/hpsa.c
@@ -5017,7 +5017,6 @@ static void hpsa_remove_one(struct pci_dev *pdev)
 	kfree(h->hba_inquiry_data);
 	pci_disable_device(pdev);
 	pci_release_regions(pdev);
-	pci_set_drvdata(pdev, NULL);
 	kfree(h);
 }
 
diff --git a/drivers/scsi/lpfc/lpfc_init.c b/drivers/scsi/lpfc/lpfc_init.c
index ca6bf2af7ce8..68c94cc85c35 100644
--- a/drivers/scsi/lpfc/lpfc_init.c
+++ b/drivers/scsi/lpfc/lpfc_init.c
@@ -4581,8 +4581,6 @@ lpfc_disable_pci_dev(struct lpfc_hba *phba)
 	/* Release PCI resource and disable PCI device */
 	pci_release_selected_regions(pdev, bars);
 	pci_disable_device(pdev);
-	/* Null out PCI private reference to driver */
-	pci_set_drvdata(pdev, NULL);
 
 	return;
 }
@@ -9429,7 +9427,6 @@ lpfc_pci_remove_one_s3(struct pci_dev *pdev)
 	/* Disable interrupt */
 	lpfc_sli_disable_intr(phba);
 
-	pci_set_drvdata(pdev, NULL);
 	scsi_host_put(shost);
 
 	/*
diff --git a/drivers/scsi/megaraid/megaraid_mbox.c b/drivers/scsi/megaraid/megaraid_mbox.c
index 515c9629e9fe..d1a4b82836ea 100644
--- a/drivers/scsi/megaraid/megaraid_mbox.c
+++ b/drivers/scsi/megaraid/megaraid_mbox.c
@@ -534,7 +534,6 @@ megaraid_probe_one(struct pci_dev *pdev, const struct pci_device_id *id)
 	return 0;
 
 out_cmm_unreg:
-	pci_set_drvdata(pdev, NULL);
 	megaraid_cmm_unregister(adapter);
 out_fini_mbox:
 	megaraid_fini_mbox(adapter);
@@ -594,11 +593,6 @@ megaraid_detach_one(struct pci_dev *pdev)
 	// detach from the IO sub-system
 	megaraid_io_detach(adapter);
 
-	// reset the device state in the PCI structure. We check this
-	// condition when we enter here. If the device state is NULL,
-	// that would mean the device has already been removed
-	pci_set_drvdata(pdev, NULL);
-
 	// Unregister from common management module
 	//
 	// FIXME: this must return success or failure for conditions if there
diff --git a/drivers/scsi/megaraid/megaraid_sas_base.c b/drivers/scsi/megaraid/megaraid_sas_base.c
index 2cf9470dd11b..0a743a5d1647 100644
--- a/drivers/scsi/megaraid/megaraid_sas_base.c
+++ b/drivers/scsi/megaraid/megaraid_sas_base.c
@@ -4451,7 +4451,6 @@ retry_irq_register:
 	megasas_mgmt_info.instance[megasas_mgmt_info.max_index] = NULL;
 	megasas_mgmt_info.max_index--;
 
-	pci_set_drvdata(pdev, NULL);
 	instance->instancet->disable_intr(instance);
 	if (instance->msix_vectors)
 		for (i = 0 ; i < instance->msix_vectors; i++)
@@ -4807,8 +4806,6 @@ static void megasas_detach_one(struct pci_dev *pdev)
 		}
 	}
 
-	pci_set_drvdata(instance->pdev, NULL);
-
 	instance->instancet->disable_intr(instance);
 
 	if (instance->msix_vectors)
@@ -4850,8 +4847,6 @@ static void megasas_detach_one(struct pci_dev *pdev)
 				instance->evt_detail, instance->evt_detail_h);
 	scsi_host_put(host);
 
-	pci_set_drvdata(pdev, NULL);
-
 	pci_disable_device(pdev);
 
 	return;
diff --git a/drivers/scsi/mvsas/mv_init.c b/drivers/scsi/mvsas/mv_init.c
index 7b7381d7671f..5ff978be249d 100644
--- a/drivers/scsi/mvsas/mv_init.c
+++ b/drivers/scsi/mvsas/mv_init.c
@@ -657,7 +657,6 @@ static void mvs_pci_remove(struct pci_dev *pdev)
 	tasklet_kill(&((struct mvs_prv_info *)sha->lldd_ha)->mv_tasklet);
 #endif
 
-	pci_set_drvdata(pdev, NULL);
 	sas_unregister_ha(sha);
 	sas_remove_host(mvi->shost);
 	scsi_remove_host(mvi->shost);
diff --git a/drivers/scsi/mvsas/mv_sas.c b/drivers/scsi/mvsas/mv_sas.c
index 6b1b4e91e53f..6c1f223a8e1d 100644
--- a/drivers/scsi/mvsas/mv_sas.c
+++ b/drivers/scsi/mvsas/mv_sas.c
@@ -1411,7 +1411,7 @@ static int mvs_exec_internal_tmf_task(struct domain_device *dev,
 
 		if (res) {
 			del_timer(&task->slow_task->timer);
-			mv_printk("executing internel task failed:%d\n", res);
+			mv_printk("executing internal task failed:%d\n", res);
 			goto ex_err;
 		}
 
diff --git a/drivers/scsi/mvumi.c b/drivers/scsi/mvumi.c
index c3601b57a80c..edbee8dc62c9 100644
--- a/drivers/scsi/mvumi.c
+++ b/drivers/scsi/mvumi.c
@@ -2583,7 +2583,6 @@ static int mvumi_probe_one(struct pci_dev *pdev, const struct pci_device_id *id)
 	return 0;
 
 fail_io_attach:
-	pci_set_drvdata(pdev, NULL);
 	mhba->instancet->disable_intr(mhba);
 	free_irq(mhba->pdev->irq, mhba);
 fail_init_irq:
@@ -2618,7 +2617,6 @@ static void mvumi_detach_one(struct pci_dev *pdev)
 	free_irq(mhba->pdev->irq, mhba);
 	mvumi_release_fw(mhba);
 	scsi_host_put(host);
-	pci_set_drvdata(pdev, NULL);
 	pci_disable_device(pdev);
 	dev_dbg(&pdev->dev, "driver is removed!\n");
 }
diff --git a/drivers/scsi/ncr53c8xx.c b/drivers/scsi/ncr53c8xx.c
index 5982a587babc..7d014b11df62 100644
--- a/drivers/scsi/ncr53c8xx.c
+++ b/drivers/scsi/ncr53c8xx.c
@@ -1615,7 +1615,7 @@ struct ncb {
 	spinlock_t	smp_lock;	/* Lock for SMP threading       */
 
 	/*----------------------------------------------------------------
-	**	Chip and controller indentification.
+	**	Chip and controller identification.
 	**----------------------------------------------------------------
 	*/
 	int		unit;		/* Unit number			*/
diff --git a/drivers/scsi/pm8001/pm8001_init.c b/drivers/scsi/pm8001/pm8001_init.c
index 662bf13c42f0..34f5f5ffef05 100644
--- a/drivers/scsi/pm8001/pm8001_init.c
+++ b/drivers/scsi/pm8001/pm8001_init.c
@@ -909,7 +909,6 @@ static void pm8001_pci_remove(struct pci_dev *pdev)
 	struct pm8001_hba_info *pm8001_ha;
 	int i;
 	pm8001_ha = sha->lldd_ha;
-	pci_set_drvdata(pdev, NULL);
 	sas_unregister_ha(sha);
 	sas_remove_host(pm8001_ha->shost);
 	list_del(&pm8001_ha->list);
diff --git a/drivers/scsi/pmcraid.c b/drivers/scsi/pmcraid.c
index 1eb7b0280a45..bd6f743d87a7 100644
--- a/drivers/scsi/pmcraid.c
+++ b/drivers/scsi/pmcraid.c
@@ -1512,7 +1512,8 @@ static int pmcraid_notify_aen(
 	}
 
 	result =
-		genlmsg_multicast(skb, 0, pmcraid_event_family.id, GFP_ATOMIC);
+		genlmsg_multicast(&pmcraid_event_family, skb, 0,
+				  pmcraid_event_family.id, GFP_ATOMIC);
 
 	/* If there are no listeners, genlmsg_multicast may return non-zero
 	 * value.
@@ -6049,7 +6050,6 @@ out_release_regions:
 
 out_disable_device:
 	atomic_dec(&pmcraid_adapter_count);
-	pci_set_drvdata(pdev, NULL);
 	pci_disable_device(pdev);
 	return -ENODEV;
 }
diff --git a/drivers/scsi/qla2xxx/qla_os.c b/drivers/scsi/qla2xxx/qla_os.c
index bcd57f699ebb..52be35e0300c 100644
--- a/drivers/scsi/qla2xxx/qla_os.c
+++ b/drivers/scsi/qla2xxx/qla_os.c
@@ -3179,7 +3179,6 @@ qla2x00_remove_one(struct pci_dev *pdev)
 	pci_disable_pcie_error_reporting(pdev);
 
 	pci_disable_device(pdev);
-	pci_set_drvdata(pdev, NULL);
 }
 
 static void
diff --git a/drivers/scsi/qla4xxx/ql4_os.c b/drivers/scsi/qla4xxx/ql4_os.c
index 6dc3e99b7f9c..a28d5e624aab 100644
--- a/drivers/scsi/qla4xxx/ql4_os.c
+++ b/drivers/scsi/qla4xxx/ql4_os.c
@@ -7827,7 +7827,6 @@ static void qla4xxx_remove_adapter(struct pci_dev *pdev)
 
 	pci_disable_pcie_error_reporting(pdev);
 	pci_disable_device(pdev);
-	pci_set_drvdata(pdev, NULL);
 }
 
 /**
diff --git a/drivers/scsi/scsi_transport_srp.c b/drivers/scsi/scsi_transport_srp.c
index f379c7f3034c..2700a5a09bd4 100644
--- a/drivers/scsi/scsi_transport_srp.c
+++ b/drivers/scsi/scsi_transport_srp.c
@@ -24,12 +24,15 @@
 #include <linux/err.h>
 #include <linux/slab.h>
 #include <linux/string.h>
+#include <linux/delay.h>
 
 #include <scsi/scsi.h>
+#include <scsi/scsi_cmnd.h>
 #include <scsi/scsi_device.h>
 #include <scsi/scsi_host.h>
 #include <scsi/scsi_transport.h>
 #include <scsi/scsi_transport_srp.h>
+#include "scsi_priv.h"
 #include "scsi_transport_srp_internal.h"
 
 struct srp_host_attrs {
@@ -38,7 +41,7 @@ struct srp_host_attrs {
 #define to_srp_host_attrs(host)	((struct srp_host_attrs *)(host)->shost_data)
 
 #define SRP_HOST_ATTRS 0
-#define SRP_RPORT_ATTRS 3
+#define SRP_RPORT_ATTRS 8
 
 struct srp_internal {
 	struct scsi_transport_template t;
@@ -54,6 +57,36 @@ struct srp_internal {
 
 #define	dev_to_rport(d)	container_of(d, struct srp_rport, dev)
 #define transport_class_to_srp_rport(dev) dev_to_rport((dev)->parent)
+static inline struct Scsi_Host *rport_to_shost(struct srp_rport *r)
+{
+	return dev_to_shost(r->dev.parent);
+}
+
+/**
+ * srp_tmo_valid() - check timeout combination validity
+ *
+ * The combination of the timeout parameters must be such that SCSI commands
+ * are finished in a reasonable time. Hence do not allow the fast I/O fail
+ * timeout to exceed SCSI_DEVICE_BLOCK_MAX_TIMEOUT. Furthermore, these
+ * parameters must be such that multipath can detect failed paths timely.
+ * Hence do not allow all three parameters to be disabled simultaneously.
+ */
+int srp_tmo_valid(int reconnect_delay, int fast_io_fail_tmo, int dev_loss_tmo)
+{
+	if (reconnect_delay < 0 && fast_io_fail_tmo < 0 && dev_loss_tmo < 0)
+		return -EINVAL;
+	if (reconnect_delay == 0)
+		return -EINVAL;
+	if (fast_io_fail_tmo > SCSI_DEVICE_BLOCK_MAX_TIMEOUT)
+		return -EINVAL;
+	if (dev_loss_tmo >= LONG_MAX / HZ)
+		return -EINVAL;
+	if (fast_io_fail_tmo >= 0 && dev_loss_tmo >= 0 &&
+	    fast_io_fail_tmo >= dev_loss_tmo)
+		return -EINVAL;
+	return 0;
+}
+EXPORT_SYMBOL_GPL(srp_tmo_valid);
 
 static int srp_host_setup(struct transport_container *tc, struct device *dev,
 			  struct device *cdev)
@@ -134,10 +167,465 @@ static ssize_t store_srp_rport_delete(struct device *dev,
 
 static DEVICE_ATTR(delete, S_IWUSR, NULL, store_srp_rport_delete);
 
+static ssize_t show_srp_rport_state(struct device *dev,
+				    struct device_attribute *attr,
+				    char *buf)
+{
+	static const char *const state_name[] = {
+		[SRP_RPORT_RUNNING]	= "running",
+		[SRP_RPORT_BLOCKED]	= "blocked",
+		[SRP_RPORT_FAIL_FAST]	= "fail-fast",
+		[SRP_RPORT_LOST]	= "lost",
+	};
+	struct srp_rport *rport = transport_class_to_srp_rport(dev);
+	enum srp_rport_state state = rport->state;
+
+	return sprintf(buf, "%s\n",
+		       (unsigned)state < ARRAY_SIZE(state_name) ?
+		       state_name[state] : "???");
+}
+
+static DEVICE_ATTR(state, S_IRUGO, show_srp_rport_state, NULL);
+
+static ssize_t srp_show_tmo(char *buf, int tmo)
+{
+	return tmo >= 0 ? sprintf(buf, "%d\n", tmo) : sprintf(buf, "off\n");
+}
+
+static int srp_parse_tmo(int *tmo, const char *buf)
+{
+	int res = 0;
+
+	if (strncmp(buf, "off", 3) != 0)
+		res = kstrtoint(buf, 0, tmo);
+	else
+		*tmo = -1;
+
+	return res;
+}
+
+static ssize_t show_reconnect_delay(struct device *dev,
+				    struct device_attribute *attr, char *buf)
+{
+	struct srp_rport *rport = transport_class_to_srp_rport(dev);
+
+	return srp_show_tmo(buf, rport->reconnect_delay);
+}
+
+static ssize_t store_reconnect_delay(struct device *dev,
+				     struct device_attribute *attr,
+				     const char *buf, const size_t count)
+{
+	struct srp_rport *rport = transport_class_to_srp_rport(dev);
+	int res, delay;
+
+	res = srp_parse_tmo(&delay, buf);
+	if (res)
+		goto out;
+	res = srp_tmo_valid(delay, rport->fast_io_fail_tmo,
+			    rport->dev_loss_tmo);
+	if (res)
+		goto out;
+
+	if (rport->reconnect_delay <= 0 && delay > 0 &&
+	    rport->state != SRP_RPORT_RUNNING) {
+		queue_delayed_work(system_long_wq, &rport->reconnect_work,
+				   delay * HZ);
+	} else if (delay <= 0) {
+		cancel_delayed_work(&rport->reconnect_work);
+	}
+	rport->reconnect_delay = delay;
+	res = count;
+
+out:
+	return res;
+}
+
+static DEVICE_ATTR(reconnect_delay, S_IRUGO | S_IWUSR, show_reconnect_delay,
+		   store_reconnect_delay);
+
+static ssize_t show_failed_reconnects(struct device *dev,
+				      struct device_attribute *attr, char *buf)
+{
+	struct srp_rport *rport = transport_class_to_srp_rport(dev);
+
+	return sprintf(buf, "%d\n", rport->failed_reconnects);
+}
+
+static DEVICE_ATTR(failed_reconnects, S_IRUGO, show_failed_reconnects, NULL);
+
+static ssize_t show_srp_rport_fast_io_fail_tmo(struct device *dev,
+					       struct device_attribute *attr,
+					       char *buf)
+{
+	struct srp_rport *rport = transport_class_to_srp_rport(dev);
+
+	return srp_show_tmo(buf, rport->fast_io_fail_tmo);
+}
+
+static ssize_t store_srp_rport_fast_io_fail_tmo(struct device *dev,
+						struct device_attribute *attr,
+						const char *buf, size_t count)
+{
+	struct srp_rport *rport = transport_class_to_srp_rport(dev);
+	int res;
+	int fast_io_fail_tmo;
+
+	res = srp_parse_tmo(&fast_io_fail_tmo, buf);
+	if (res)
+		goto out;
+	res = srp_tmo_valid(rport->reconnect_delay, fast_io_fail_tmo,
+			    rport->dev_loss_tmo);
+	if (res)
+		goto out;
+	rport->fast_io_fail_tmo = fast_io_fail_tmo;
+	res = count;
+
+out:
+	return res;
+}
+
+static DEVICE_ATTR(fast_io_fail_tmo, S_IRUGO | S_IWUSR,
+		   show_srp_rport_fast_io_fail_tmo,
+		   store_srp_rport_fast_io_fail_tmo);
+
+static ssize_t show_srp_rport_dev_loss_tmo(struct device *dev,
+					   struct device_attribute *attr,
+					   char *buf)
+{
+	struct srp_rport *rport = transport_class_to_srp_rport(dev);
+
+	return srp_show_tmo(buf, rport->dev_loss_tmo);
+}
+
+static ssize_t store_srp_rport_dev_loss_tmo(struct device *dev,
+					    struct device_attribute *attr,
+					    const char *buf, size_t count)
+{
+	struct srp_rport *rport = transport_class_to_srp_rport(dev);
+	int res;
+	int dev_loss_tmo;
+
+	res = srp_parse_tmo(&dev_loss_tmo, buf);
+	if (res)
+		goto out;
+	res = srp_tmo_valid(rport->reconnect_delay, rport->fast_io_fail_tmo,
+			    dev_loss_tmo);
+	if (res)
+		goto out;
+	rport->dev_loss_tmo = dev_loss_tmo;
+	res = count;
+
+out:
+	return res;
+}
+
+static DEVICE_ATTR(dev_loss_tmo, S_IRUGO | S_IWUSR,
+		   show_srp_rport_dev_loss_tmo,
+		   store_srp_rport_dev_loss_tmo);
+
+static int srp_rport_set_state(struct srp_rport *rport,
+			       enum srp_rport_state new_state)
+{
+	enum srp_rport_state old_state = rport->state;
+
+	lockdep_assert_held(&rport->mutex);
+
+	switch (new_state) {
+	case SRP_RPORT_RUNNING:
+		switch (old_state) {
+		case SRP_RPORT_LOST:
+			goto invalid;
+		default:
+			break;
+		}
+		break;
+	case SRP_RPORT_BLOCKED:
+		switch (old_state) {
+		case SRP_RPORT_RUNNING:
+			break;
+		default:
+			goto invalid;
+		}
+		break;
+	case SRP_RPORT_FAIL_FAST:
+		switch (old_state) {
+		case SRP_RPORT_LOST:
+			goto invalid;
+		default:
+			break;
+		}
+		break;
+	case SRP_RPORT_LOST:
+		break;
+	}
+	rport->state = new_state;
+	return 0;
+
+invalid:
+	return -EINVAL;
+}
+
+/**
+ * srp_reconnect_work() - reconnect and schedule a new attempt if necessary
+ */
+static void srp_reconnect_work(struct work_struct *work)
+{
+	struct srp_rport *rport = container_of(to_delayed_work(work),
+					struct srp_rport, reconnect_work);
+	struct Scsi_Host *shost = rport_to_shost(rport);
+	int delay, res;
+
+	res = srp_reconnect_rport(rport);
+	if (res != 0) {
+		shost_printk(KERN_ERR, shost,
+			     "reconnect attempt %d failed (%d)\n",
+			     ++rport->failed_reconnects, res);
+		delay = rport->reconnect_delay *
+			min(100, max(1, rport->failed_reconnects - 10));
+		if (delay > 0)
+			queue_delayed_work(system_long_wq,
+					   &rport->reconnect_work, delay * HZ);
+	}
+}
+
+static void __rport_fail_io_fast(struct srp_rport *rport)
+{
+	struct Scsi_Host *shost = rport_to_shost(rport);
+	struct srp_internal *i;
+
+	lockdep_assert_held(&rport->mutex);
+
+	if (srp_rport_set_state(rport, SRP_RPORT_FAIL_FAST))
+		return;
+	scsi_target_unblock(rport->dev.parent, SDEV_TRANSPORT_OFFLINE);
+
+	/* Involve the LLD if possible to terminate all I/O on the rport. */
+	i = to_srp_internal(shost->transportt);
+	if (i->f->terminate_rport_io)
+		i->f->terminate_rport_io(rport);
+}
+
+/**
+ * rport_fast_io_fail_timedout() - fast I/O failure timeout handler
+ */
+static void rport_fast_io_fail_timedout(struct work_struct *work)
+{
+	struct srp_rport *rport = container_of(to_delayed_work(work),
+					struct srp_rport, fast_io_fail_work);
+	struct Scsi_Host *shost = rport_to_shost(rport);
+
+	pr_info("fast_io_fail_tmo expired for SRP %s / %s.\n",
+		dev_name(&rport->dev), dev_name(&shost->shost_gendev));
+
+	mutex_lock(&rport->mutex);
+	if (rport->state == SRP_RPORT_BLOCKED)
+		__rport_fail_io_fast(rport);
+	mutex_unlock(&rport->mutex);
+}
+
+/**
+ * rport_dev_loss_timedout() - device loss timeout handler
+ */
+static void rport_dev_loss_timedout(struct work_struct *work)
+{
+	struct srp_rport *rport = container_of(to_delayed_work(work),
+					struct srp_rport, dev_loss_work);
+	struct Scsi_Host *shost = rport_to_shost(rport);
+	struct srp_internal *i = to_srp_internal(shost->transportt);
+
+	pr_info("dev_loss_tmo expired for SRP %s / %s.\n",
+		dev_name(&rport->dev), dev_name(&shost->shost_gendev));
+
+	mutex_lock(&rport->mutex);
+	WARN_ON(srp_rport_set_state(rport, SRP_RPORT_LOST) != 0);
+	scsi_target_unblock(rport->dev.parent, SDEV_TRANSPORT_OFFLINE);
+	mutex_unlock(&rport->mutex);
+
+	i->f->rport_delete(rport);
+}
+
+static void __srp_start_tl_fail_timers(struct srp_rport *rport)
+{
+	struct Scsi_Host *shost = rport_to_shost(rport);
+	int delay, fast_io_fail_tmo, dev_loss_tmo;
+
+	lockdep_assert_held(&rport->mutex);
+
+	if (!rport->deleted) {
+		delay = rport->reconnect_delay;
+		fast_io_fail_tmo = rport->fast_io_fail_tmo;
+		dev_loss_tmo = rport->dev_loss_tmo;
+		pr_debug("%s current state: %d\n",
+			 dev_name(&shost->shost_gendev), rport->state);
+
+		if (delay > 0)
+			queue_delayed_work(system_long_wq,
+					   &rport->reconnect_work,
+					   1UL * delay * HZ);
+		if (fast_io_fail_tmo >= 0 &&
+		    srp_rport_set_state(rport, SRP_RPORT_BLOCKED) == 0) {
+			pr_debug("%s new state: %d\n",
+				 dev_name(&shost->shost_gendev),
+				 rport->state);
+			scsi_target_block(&shost->shost_gendev);
+			queue_delayed_work(system_long_wq,
+					   &rport->fast_io_fail_work,
+					   1UL * fast_io_fail_tmo * HZ);
+		}
+		if (dev_loss_tmo >= 0)
+			queue_delayed_work(system_long_wq,
+					   &rport->dev_loss_work,
+					   1UL * dev_loss_tmo * HZ);
+	} else {
+		pr_debug("%s has already been deleted\n",
+			 dev_name(&shost->shost_gendev));
+		srp_rport_set_state(rport, SRP_RPORT_FAIL_FAST);
+		scsi_target_unblock(&shost->shost_gendev,
+				    SDEV_TRANSPORT_OFFLINE);
+	}
+}
+
+/**
+ * srp_start_tl_fail_timers() - start the transport layer failure timers
+ *
+ * Start the transport layer fast I/O failure and device loss timers. Do not
+ * modify a timer that was already started.
+ */
+void srp_start_tl_fail_timers(struct srp_rport *rport)
+{
+	mutex_lock(&rport->mutex);
+	__srp_start_tl_fail_timers(rport);
+	mutex_unlock(&rport->mutex);
+}
+EXPORT_SYMBOL(srp_start_tl_fail_timers);
+
+/**
+ * scsi_request_fn_active() - number of kernel threads inside scsi_request_fn()
+ */
+static int scsi_request_fn_active(struct Scsi_Host *shost)
+{
+	struct scsi_device *sdev;
+	struct request_queue *q;
+	int request_fn_active = 0;
+
+	shost_for_each_device(sdev, shost) {
+		q = sdev->request_queue;
+
+		spin_lock_irq(q->queue_lock);
+		request_fn_active += q->request_fn_active;
+		spin_unlock_irq(q->queue_lock);
+	}
+
+	return request_fn_active;
+}
+
+/**
+ * srp_reconnect_rport() - reconnect to an SRP target port
+ *
+ * Blocks SCSI command queueing before invoking reconnect() such that
+ * queuecommand() won't be invoked concurrently with reconnect() from outside
+ * the SCSI EH. This is important since a reconnect() implementation may
+ * reallocate resources needed by queuecommand().
+ *
+ * Notes:
+ * - This function neither waits until outstanding requests have finished nor
+ *   tries to abort these. It is the responsibility of the reconnect()
+ *   function to finish outstanding commands before reconnecting to the target
+ *   port.
+ * - It is the responsibility of the caller to ensure that the resources
+ *   reallocated by the reconnect() function won't be used while this function
+ *   is in progress. One possible strategy is to invoke this function from
+ *   the context of the SCSI EH thread only. Another possible strategy is to
+ *   lock the rport mutex inside each SCSI LLD callback that can be invoked by
+ *   the SCSI EH (the scsi_host_template.eh_*() functions and also the
+ *   scsi_host_template.queuecommand() function).
+ */
+int srp_reconnect_rport(struct srp_rport *rport)
+{
+	struct Scsi_Host *shost = rport_to_shost(rport);
+	struct srp_internal *i = to_srp_internal(shost->transportt);
+	struct scsi_device *sdev;
+	int res;
+
+	pr_debug("SCSI host %s\n", dev_name(&shost->shost_gendev));
+
+	res = mutex_lock_interruptible(&rport->mutex);
+	if (res)
+		goto out;
+	scsi_target_block(&shost->shost_gendev);
+	while (scsi_request_fn_active(shost))
+		msleep(20);
+	res = i->f->reconnect(rport);
+	pr_debug("%s (state %d): transport.reconnect() returned %d\n",
+		 dev_name(&shost->shost_gendev), rport->state, res);
+	if (res == 0) {
+		cancel_delayed_work(&rport->fast_io_fail_work);
+		cancel_delayed_work(&rport->dev_loss_work);
+
+		rport->failed_reconnects = 0;
+		srp_rport_set_state(rport, SRP_RPORT_RUNNING);
+		scsi_target_unblock(&shost->shost_gendev, SDEV_RUNNING);
+		/*
+		 * If the SCSI error handler has offlined one or more devices,
+		 * invoking scsi_target_unblock() won't change the state of
+		 * these devices into running so do that explicitly.
+		 */
+		spin_lock_irq(shost->host_lock);
+		__shost_for_each_device(sdev, shost)
+			if (sdev->sdev_state == SDEV_OFFLINE)
+				sdev->sdev_state = SDEV_RUNNING;
+		spin_unlock_irq(shost->host_lock);
+	} else if (rport->state == SRP_RPORT_RUNNING) {
+		/*
+		 * srp_reconnect_rport() was invoked with fast_io_fail
+		 * off. Mark the port as failed and start the TL failure
+		 * timers if these had not yet been started.
+		 */
+		__rport_fail_io_fast(rport);
+		scsi_target_unblock(&shost->shost_gendev,
+				    SDEV_TRANSPORT_OFFLINE);
+		__srp_start_tl_fail_timers(rport);
+	} else if (rport->state != SRP_RPORT_BLOCKED) {
+		scsi_target_unblock(&shost->shost_gendev,
+				    SDEV_TRANSPORT_OFFLINE);
+	}
+	mutex_unlock(&rport->mutex);
+
+out:
+	return res;
+}
+EXPORT_SYMBOL(srp_reconnect_rport);
+
+/**
+ * srp_timed_out() - SRP transport intercept of the SCSI timeout EH
+ *
+ * If a timeout occurs while an rport is in the blocked state, ask the SCSI
+ * EH to continue waiting (BLK_EH_RESET_TIMER). Otherwise let the SCSI core
+ * handle the timeout (BLK_EH_NOT_HANDLED).
+ *
+ * Note: This function is called from soft-IRQ context and with the request
+ * queue lock held.
+ */
+static enum blk_eh_timer_return srp_timed_out(struct scsi_cmnd *scmd)
+{
+	struct scsi_device *sdev = scmd->device;
+	struct Scsi_Host *shost = sdev->host;
+	struct srp_internal *i = to_srp_internal(shost->transportt);
+
+	pr_debug("timeout for sdev %s\n", dev_name(&sdev->sdev_gendev));
+	return i->f->reset_timer_if_blocked && scsi_device_blocked(sdev) ?
+		BLK_EH_RESET_TIMER : BLK_EH_NOT_HANDLED;
+}
+
 static void srp_rport_release(struct device *dev)
 {
 	struct srp_rport *rport = dev_to_rport(dev);
 
+	cancel_delayed_work_sync(&rport->reconnect_work);
+	cancel_delayed_work_sync(&rport->fast_io_fail_work);
+	cancel_delayed_work_sync(&rport->dev_loss_work);
+
 	put_device(dev->parent);
 	kfree(rport);
 }
@@ -185,6 +673,24 @@ static int srp_host_match(struct attribute_container *cont, struct device *dev)
 }
 
 /**
+ * srp_rport_get() - increment rport reference count
+ */
+void srp_rport_get(struct srp_rport *rport)
+{
+	get_device(&rport->dev);
+}
+EXPORT_SYMBOL(srp_rport_get);
+
+/**
+ * srp_rport_put() - decrement rport reference count
+ */
+void srp_rport_put(struct srp_rport *rport)
+{
+	put_device(&rport->dev);
+}
+EXPORT_SYMBOL(srp_rport_put);
+
+/**
  * srp_rport_add - add a SRP remote port to the device hierarchy
  * @shost:	scsi host the remote port is connected to.
  * @ids:	The port id for the remote port.
@@ -196,12 +702,15 @@ struct srp_rport *srp_rport_add(struct Scsi_Host *shost,
 {
 	struct srp_rport *rport;
 	struct device *parent = &shost->shost_gendev;
+	struct srp_internal *i = to_srp_internal(shost->transportt);
 	int id, ret;
 
 	rport = kzalloc(sizeof(*rport), GFP_KERNEL);
 	if (!rport)
 		return ERR_PTR(-ENOMEM);
 
+	mutex_init(&rport->mutex);
+
 	device_initialize(&rport->dev);
 
 	rport->dev.parent = get_device(parent);
@@ -210,6 +719,17 @@ struct srp_rport *srp_rport_add(struct Scsi_Host *shost,
 	memcpy(rport->port_id, ids->port_id, sizeof(rport->port_id));
 	rport->roles = ids->roles;
 
+	if (i->f->reconnect)
+		rport->reconnect_delay = i->f->reconnect_delay ?
+			*i->f->reconnect_delay : 10;
+	INIT_DELAYED_WORK(&rport->reconnect_work, srp_reconnect_work);
+	rport->fast_io_fail_tmo = i->f->fast_io_fail_tmo ?
+		*i->f->fast_io_fail_tmo : 15;
+	rport->dev_loss_tmo = i->f->dev_loss_tmo ? *i->f->dev_loss_tmo : 60;
+	INIT_DELAYED_WORK(&rport->fast_io_fail_work,
+			  rport_fast_io_fail_timedout);
+	INIT_DELAYED_WORK(&rport->dev_loss_work, rport_dev_loss_timedout);
+
 	id = atomic_inc_return(&to_srp_host_attrs(shost)->next_port_id);
 	dev_set_name(&rport->dev, "port-%d:%d", shost->host_no, id);
 
@@ -259,6 +779,13 @@ void srp_rport_del(struct srp_rport *rport)
 	transport_remove_device(dev);
 	device_del(dev);
 	transport_destroy_device(dev);
+
+	mutex_lock(&rport->mutex);
+	if (rport->state == SRP_RPORT_BLOCKED)
+		__rport_fail_io_fast(rport);
+	rport->deleted = true;
+	mutex_unlock(&rport->mutex);
+
 	put_device(dev);
 }
 EXPORT_SYMBOL_GPL(srp_rport_del);
@@ -310,6 +837,8 @@ srp_attach_transport(struct srp_function_template *ft)
 	if (!i)
 		return NULL;
 
+	i->t.eh_timed_out = srp_timed_out;
+
 	i->t.tsk_mgmt_response = srp_tsk_mgmt_response;
 	i->t.it_nexus_response = srp_it_nexus_response;
 
@@ -327,6 +856,15 @@ srp_attach_transport(struct srp_function_template *ft)
 	count = 0;
 	i->rport_attrs[count++] = &dev_attr_port_id;
 	i->rport_attrs[count++] = &dev_attr_roles;
+	if (ft->has_rport_state) {
+		i->rport_attrs[count++] = &dev_attr_state;
+		i->rport_attrs[count++] = &dev_attr_fast_io_fail_tmo;
+		i->rport_attrs[count++] = &dev_attr_dev_loss_tmo;
+	}
+	if (ft->reconnect) {
+		i->rport_attrs[count++] = &dev_attr_reconnect_delay;
+		i->rport_attrs[count++] = &dev_attr_failed_reconnects;
+	}
 	if (ft->rport_delete)
 		i->rport_attrs[count++] = &dev_attr_delete;
 	i->rport_attrs[count++] = NULL;
diff --git a/drivers/scsi/stex.c b/drivers/scsi/stex.c
index 325c31caa6e0..1aa4befcfbd0 100644
--- a/drivers/scsi/stex.c
+++ b/drivers/scsi/stex.c
@@ -1790,8 +1790,6 @@ static void stex_remove(struct pci_dev *pdev)
 
 	scsi_remove_host(hba->host);
 
-	pci_set_drvdata(pdev, NULL);
-
 	stex_hba_stop(hba);
 
 	stex_hba_free(hba);
diff --git a/drivers/scsi/sym53c8xx_2/sym_glue.h b/drivers/scsi/sym53c8xx_2/sym_glue.h
index b80bf709f104..805369521df8 100644
--- a/drivers/scsi/sym53c8xx_2/sym_glue.h
+++ b/drivers/scsi/sym53c8xx_2/sym_glue.h
@@ -174,7 +174,7 @@ struct sym_slcb {
  */
 struct sym_shcb {
 	/*
-	 *  Chip and controller indentification.
+	 *  Chip and controller identification.
 	 */
 	int		unit;
 	char		inst_name[16];
diff --git a/drivers/scsi/tmscsim.c b/drivers/scsi/tmscsim.c
index 11423615c2ea..b006cf789ba1 100644
--- a/drivers/scsi/tmscsim.c
+++ b/drivers/scsi/tmscsim.c
@@ -2553,7 +2553,6 @@ static void dc390_remove_one(struct pci_dev *dev)
 
 	pci_disable_device(dev);
 	scsi_host_put(scsi_host);
-	pci_set_drvdata(dev, NULL);
 }
 
 static struct pci_device_id tmscsim_pci_tbl[] = {
diff --git a/drivers/scsi/ufs/ufshcd-pci.c b/drivers/scsi/ufs/ufshcd-pci.c
index a823cf44e949..8b9531204c2b 100644
--- a/drivers/scsi/ufs/ufshcd-pci.c
+++ b/drivers/scsi/ufs/ufshcd-pci.c
@@ -132,7 +132,6 @@ static void ufshcd_pci_remove(struct pci_dev *pdev)
 	pm_runtime_forbid(&pdev->dev);
 	pm_runtime_get_noresume(&pdev->dev);
 	ufshcd_remove(hba);
-	pci_set_drvdata(pdev, NULL);
 }
 
 /**
diff --git a/drivers/scsi/virtio_scsi.c b/drivers/scsi/virtio_scsi.c
index 74b88efde6ad..c3173dced870 100644
--- a/drivers/scsi/virtio_scsi.c
+++ b/drivers/scsi/virtio_scsi.c
@@ -224,6 +224,9 @@ static void virtscsi_vq_done(struct virtio_scsi *vscsi,
 		virtqueue_disable_cb(vq);
 		while ((buf = virtqueue_get_buf(vq, &len)) != NULL)
 			fn(vscsi, buf);
+
+		if (unlikely(virtqueue_is_broken(vq)))
+			break;
 	} while (!virtqueue_enable_cb(vq));
 	spin_unlock_irqrestore(&virtscsi_vq->vq_lock, flags);
 }
@@ -710,19 +713,15 @@ static struct scsi_host_template virtscsi_host_template_multi = {
 #define virtscsi_config_get(vdev, fld) \
 	({ \
 		typeof(((struct virtio_scsi_config *)0)->fld) __val; \
-		vdev->config->get(vdev, \
-				  offsetof(struct virtio_scsi_config, fld), \
-				  &__val, sizeof(__val)); \
+		virtio_cread(vdev, struct virtio_scsi_config, fld, &__val); \
 		__val; \
 	})
 
 #define virtscsi_config_set(vdev, fld, val) \
-	(void)({ \
+	do { \
 		typeof(((struct virtio_scsi_config *)0)->fld) __val = (val); \
-		vdev->config->set(vdev, \
-				  offsetof(struct virtio_scsi_config, fld), \
-				  &__val, sizeof(__val)); \
-	})
+		virtio_cwrite(vdev, struct virtio_scsi_config, fld, &__val); \
+	} while(0)
 
 static void __virtscsi_set_affinity(struct virtio_scsi *vscsi, bool affinity)
 {
@@ -954,7 +953,7 @@ static void virtscsi_remove(struct virtio_device *vdev)
 	scsi_host_put(shost);
 }
 
-#ifdef CONFIG_PM
+#ifdef CONFIG_PM_SLEEP
 static int virtscsi_freeze(struct virtio_device *vdev)
 {
 	virtscsi_remove_vqs(vdev);
@@ -988,7 +987,7 @@ static struct virtio_driver virtio_scsi_driver = {
 	.id_table = id_table,
 	.probe = virtscsi_probe,
 	.scan = virtscsi_scan,
-#ifdef CONFIG_PM
+#ifdef CONFIG_PM_SLEEP
 	.freeze = virtscsi_freeze,
 	.restore = virtscsi_restore,
 #endif
diff --git a/drivers/scsi/vmw_pvscsi.c b/drivers/scsi/vmw_pvscsi.c
index 3bfaa66fa0d1..b9755ec0e812 100644
--- a/drivers/scsi/vmw_pvscsi.c
+++ b/drivers/scsi/vmw_pvscsi.c
@@ -1405,7 +1405,6 @@ out_release_resources:
 out_free_host:
 	scsi_host_put(host);
 out_disable_device:
-	pci_set_drvdata(pdev, NULL);
 	pci_disable_device(pdev);
 
 	return error;
@@ -1445,7 +1444,6 @@ static void pvscsi_remove(struct pci_dev *pdev)
 
 	scsi_host_put(host);
 
-	pci_set_drvdata(pdev, NULL);
 	pci_disable_device(pdev);
 }
 
diff --git a/drivers/spi/spi-bcm2835.c b/drivers/spi/spi-bcm2835.c
index 4c332143a310..3ed666fe840a 100644
--- a/drivers/spi/spi-bcm2835.c
+++ b/drivers/spi/spi-bcm2835.c
@@ -217,7 +217,7 @@ static int bcm2835_spi_start_transfer(struct spi_device *spi,
 		cs |= spi->chip_select;
 	}
 
-	INIT_COMPLETION(bs->done);
+	reinit_completion(&bs->done);
 	bs->tx_buf = tfr->tx_buf;
 	bs->rx_buf = tfr->rx_buf;
 	bs->len = tfr->len;
diff --git a/drivers/spi/spi-clps711x.c b/drivers/spi/spi-clps711x.c
index e2a5a426b2ef..6f03d7e6435d 100644
--- a/drivers/spi/spi-clps711x.c
+++ b/drivers/spi/spi-clps711x.c
@@ -105,7 +105,7 @@ static int spi_clps711x_transfer_one_message(struct spi_master *master,
 
 		gpio_set_value(cs, !!(msg->spi->mode & SPI_CS_HIGH));
 
-		INIT_COMPLETION(hw->done);
+		reinit_completion(&hw->done);
 
 		hw->count = 0;
 		hw->len = xfer->len;
diff --git a/drivers/spi/spi-davinci.c b/drivers/spi/spi-davinci.c
index dd72445ba2ea..50b2d88c8190 100644
--- a/drivers/spi/spi-davinci.c
+++ b/drivers/spi/spi-davinci.c
@@ -554,7 +554,7 @@ static int davinci_spi_bufs(struct spi_device *spi, struct spi_transfer *t)
 	clear_io_bits(dspi->base + SPIGCR1, SPIGCR1_POWERDOWN_MASK);
 	set_io_bits(dspi->base + SPIGCR1, SPIGCR1_SPIENA_MASK);
 
-	INIT_COMPLETION(dspi->done);
+	reinit_completion(&dspi->done);
 
 	if (spicfg->io_type == SPI_IO_TYPE_INTR)
 		set_io_bits(dspi->base + SPIINT, SPIINT_MASKINT);
diff --git a/drivers/spi/spi-dw-mid.c b/drivers/spi/spi-dw-mid.c
index b9f0192758d6..6d207afec8cb 100644
--- a/drivers/spi/spi-dw-mid.c
+++ b/drivers/spi/spi-dw-mid.c
@@ -150,7 +150,7 @@ static int mid_spi_dma_transfer(struct dw_spi *dws, int cs_change)
 				&dws->tx_sgl,
 				1,
 				DMA_MEM_TO_DEV,
-				DMA_PREP_INTERRUPT | DMA_COMPL_SKIP_DEST_UNMAP);
+				DMA_PREP_INTERRUPT);
 	txdesc->callback = dw_spi_dma_done;
 	txdesc->callback_param = dws;
 
@@ -173,7 +173,7 @@ static int mid_spi_dma_transfer(struct dw_spi *dws, int cs_change)
 				&dws->rx_sgl,
 				1,
 				DMA_DEV_TO_MEM,
-				DMA_PREP_INTERRUPT | DMA_COMPL_SKIP_DEST_UNMAP);
+				DMA_PREP_INTERRUPT);
 	rxdesc->callback = dw_spi_dma_done;
 	rxdesc->callback_param = dws;
 
diff --git a/drivers/spi/spi-fsl-espi.c b/drivers/spi/spi-fsl-espi.c
index 32200d4f8780..80d8f40f7e05 100644
--- a/drivers/spi/spi-fsl-espi.c
+++ b/drivers/spi/spi-fsl-espi.c
@@ -232,7 +232,7 @@ static int fsl_espi_bufs(struct spi_device *spi, struct spi_transfer *t)
 	mpc8xxx_spi->tx = t->tx_buf;
 	mpc8xxx_spi->rx = t->rx_buf;
 
-	INIT_COMPLETION(mpc8xxx_spi->done);
+	reinit_completion(&mpc8xxx_spi->done);
 
 	/* Set SPCOM[CS] and SPCOM[TRANLEN] field */
 	if ((t->len - 1) > SPCOM_TRANLEN_MAX) {
diff --git a/drivers/spi/spi-fsl-spi.c b/drivers/spi/spi-fsl-spi.c
index 2129fcd1c31b..119f7af94537 100644
--- a/drivers/spi/spi-fsl-spi.c
+++ b/drivers/spi/spi-fsl-spi.c
@@ -339,7 +339,7 @@ static int fsl_spi_bufs(struct spi_device *spi, struct spi_transfer *t,
 	mpc8xxx_spi->tx = t->tx_buf;
 	mpc8xxx_spi->rx = t->rx_buf;
 
-	INIT_COMPLETION(mpc8xxx_spi->done);
+	reinit_completion(&mpc8xxx_spi->done);
 
 	if (mpc8xxx_spi->flags & SPI_CPM_MODE)
 		ret = fsl_spi_cpm_bufs(mpc8xxx_spi, t, is_dma_mapped);
diff --git a/drivers/spi/spi-mpc512x-psc.c b/drivers/spi/spi-mpc512x-psc.c
index 58d5ee0e4443..9602bbd8d7ea 100644
--- a/drivers/spi/spi-mpc512x-psc.c
+++ b/drivers/spi/spi-mpc512x-psc.c
@@ -167,7 +167,7 @@ static int mpc512x_psc_spi_transfer_rxtx(struct spi_device *spi,
 			}
 
 			/* have the ISR trigger when the TX FIFO is empty */
-			INIT_COMPLETION(mps->txisrdone);
+			reinit_completion(&mps->txisrdone);
 			out_be32(&fifo->txisr, MPC512x_PSC_FIFO_EMPTY);
 			out_be32(&fifo->tximr, MPC512x_PSC_FIFO_EMPTY);
 			wait_for_completion(&mps->txisrdone);
diff --git a/drivers/spi/spi-mxs.c b/drivers/spi/spi-mxs.c
index de333059a9a7..73afb56c08cc 100644
--- a/drivers/spi/spi-mxs.c
+++ b/drivers/spi/spi-mxs.c
@@ -202,7 +202,7 @@ static int mxs_spi_txrx_dma(struct mxs_spi *spi,
 	if (!dma_xfer)
 		return -ENOMEM;
 
-	INIT_COMPLETION(spi->c);
+	reinit_completion(&spi->c);
 
 	/* Chip select was already programmed into CTRL0 */
 	ctrl0 = readl(ssp->base + HW_SSP_CTRL0);
diff --git a/drivers/spi/spi-s3c64xx.c b/drivers/spi/spi-s3c64xx.c
index 9e2020df9e0f..4c4b0a1219a7 100644
--- a/drivers/spi/spi-s3c64xx.c
+++ b/drivers/spi/spi-s3c64xx.c
@@ -890,7 +890,7 @@ static int s3c64xx_spi_transfer_one(struct spi_master *master,
 	unsigned long flags;
 	int use_dma;
 
-	INIT_COMPLETION(sdd->xfer_completion);
+		reinit_completion(&sdd->xfer_completion);
 
 	/* Only BPW and Speed may change across transfers */
 	bpw = xfer->bits_per_word;
diff --git a/drivers/spi/spi-sh-msiof.c b/drivers/spi/spi-sh-msiof.c
index 2a95435a6a11..c74298cf70e2 100644
--- a/drivers/spi/spi-sh-msiof.c
+++ b/drivers/spi/spi-sh-msiof.c
@@ -465,7 +465,7 @@ static int sh_msiof_spi_txrx_once(struct sh_msiof_spi_priv *p,
 	ret = ret ? ret : sh_msiof_modify_ctr_wait(p, 0, CTR_TXE);
 
 	/* start by setting frame bit */
-	INIT_COMPLETION(p->done);
+	reinit_completion(&p->done);
 	ret = ret ? ret : sh_msiof_modify_ctr_wait(p, 0, CTR_TFSE);
 	if (ret) {
 		dev_err(&p->pdev->dev, "failed to start hardware\n");
diff --git a/drivers/spi/spi-sirf.c b/drivers/spi/spi-sirf.c
index 592b4aff651f..ed5e501c4652 100644
--- a/drivers/spi/spi-sirf.c
+++ b/drivers/spi/spi-sirf.c
@@ -305,8 +305,8 @@ static int spi_sirfsoc_transfer(struct spi_device *spi, struct spi_transfer *t)
 	sspi->tx = t->tx_buf ? t->tx_buf : sspi->dummypage;
 	sspi->rx = t->rx_buf ? t->rx_buf : sspi->dummypage;
 	sspi->left_tx_word = sspi->left_rx_word = t->len / sspi->word_width;
-	INIT_COMPLETION(sspi->rx_done);
-	INIT_COMPLETION(sspi->tx_done);
+	reinit_completion(&sspi->rx_done);
+	reinit_completion(&sspi->tx_done);
 
 	writel(SIRFSOC_SPI_INT_MASK_ALL, sspi->base + SIRFSOC_SPI_INT_STATUS);
 
diff --git a/drivers/spi/spi-tegra114.c b/drivers/spi/spi-tegra114.c
index 9146bb3c2489..aaecfb3ebf58 100644
--- a/drivers/spi/spi-tegra114.c
+++ b/drivers/spi/spi-tegra114.c
@@ -451,7 +451,7 @@ static void tegra_spi_dma_complete(void *args)
 
 static int tegra_spi_start_tx_dma(struct tegra_spi_data *tspi, int len)
 {
-	INIT_COMPLETION(tspi->tx_dma_complete);
+	reinit_completion(&tspi->tx_dma_complete);
 	tspi->tx_dma_desc = dmaengine_prep_slave_single(tspi->tx_dma_chan,
 				tspi->tx_dma_phys, len, DMA_MEM_TO_DEV,
 				DMA_PREP_INTERRUPT |  DMA_CTRL_ACK);
@@ -470,7 +470,7 @@ static int tegra_spi_start_tx_dma(struct tegra_spi_data *tspi, int len)
 
 static int tegra_spi_start_rx_dma(struct tegra_spi_data *tspi, int len)
 {
-	INIT_COMPLETION(tspi->rx_dma_complete);
+	reinit_completion(&tspi->rx_dma_complete);
 	tspi->rx_dma_desc = dmaengine_prep_slave_single(tspi->rx_dma_chan,
 				tspi->rx_dma_phys, len, DMA_DEV_TO_MEM,
 				DMA_PREP_INTERRUPT |  DMA_CTRL_ACK);
@@ -844,7 +844,7 @@ static int tegra_spi_transfer_one_message(struct spi_master *master,
 	list_for_each_entry(xfer, &msg->transfers, transfer_list) {
 		unsigned long cmd1;
 
-		INIT_COMPLETION(tspi->xfer_completion);
+		reinit_completion(&tspi->xfer_completion);
 
 		cmd1 = tegra_spi_setup_transfer_one(spi, xfer, is_first_msg);
 
diff --git a/drivers/spi/spi-tegra20-sflash.c b/drivers/spi/spi-tegra20-sflash.c
index 79be8ce6a9d1..4dc8e8129459 100644
--- a/drivers/spi/spi-tegra20-sflash.c
+++ b/drivers/spi/spi-tegra20-sflash.c
@@ -339,7 +339,7 @@ static int tegra_sflash_transfer_one_message(struct spi_master *master,
 	msg->actual_length = 0;
 	single_xfer = list_is_singular(&msg->transfers);
 	list_for_each_entry(xfer, &msg->transfers, transfer_list) {
-		INIT_COMPLETION(tsd->xfer_completion);
+		reinit_completion(&tsd->xfer_completion);
 		ret = tegra_sflash_start_transfer_one(spi, xfer,
 					is_first_msg, single_xfer);
 		if (ret < 0) {
diff --git a/drivers/spi/spi-tegra20-slink.c b/drivers/spi/spi-tegra20-slink.c
index af0a67886ae8..e66715ba37ed 100644
--- a/drivers/spi/spi-tegra20-slink.c
+++ b/drivers/spi/spi-tegra20-slink.c
@@ -462,7 +462,7 @@ static void tegra_slink_dma_complete(void *args)
 
 static int tegra_slink_start_tx_dma(struct tegra_slink_data *tspi, int len)
 {
-	INIT_COMPLETION(tspi->tx_dma_complete);
+	reinit_completion(&tspi->tx_dma_complete);
 	tspi->tx_dma_desc = dmaengine_prep_slave_single(tspi->tx_dma_chan,
 				tspi->tx_dma_phys, len, DMA_MEM_TO_DEV,
 				DMA_PREP_INTERRUPT |  DMA_CTRL_ACK);
@@ -481,7 +481,7 @@ static int tegra_slink_start_tx_dma(struct tegra_slink_data *tspi, int len)
 
 static int tegra_slink_start_rx_dma(struct tegra_slink_data *tspi, int len)
 {
-	INIT_COMPLETION(tspi->rx_dma_complete);
+	reinit_completion(&tspi->rx_dma_complete);
 	tspi->rx_dma_desc = dmaengine_prep_slave_single(tspi->rx_dma_chan,
 				tspi->rx_dma_phys, len, DMA_DEV_TO_MEM,
 				DMA_PREP_INTERRUPT |  DMA_CTRL_ACK);
@@ -836,7 +836,7 @@ static int tegra_slink_transfer_one(struct spi_master *master,
 	struct tegra_slink_data *tspi = spi_master_get_devdata(master);
 	int ret;
 
-	INIT_COMPLETION(tspi->xfer_completion);
+	reinit_completion(&tspi->xfer_completion);
 	ret = tegra_slink_start_transfer_one(spi, xfer);
 	if (ret < 0) {
 		dev_err(tspi->dev,
diff --git a/drivers/spi/spi-xilinx.c b/drivers/spi/spi-xilinx.c
index ec3a83f52ea2..6d4ce4615163 100644
--- a/drivers/spi/spi-xilinx.c
+++ b/drivers/spi/spi-xilinx.c
@@ -258,7 +258,7 @@ static int xilinx_spi_txrx_bufs(struct spi_device *spi, struct spi_transfer *t)
 	xspi->tx_ptr = t->tx_buf;
 	xspi->rx_ptr = t->rx_buf;
 	xspi->remaining_bytes = t->len;
-	INIT_COMPLETION(xspi->done);
+	reinit_completion(&xspi->done);
 
 
 	/* Enable the transmit empty interrupt, which we use to determine
diff --git a/drivers/spi/spi.c b/drivers/spi/spi.c
index 8763d93ce2b7..18cc625d887f 100644
--- a/drivers/spi/spi.c
+++ b/drivers/spi/spi.c
@@ -582,7 +582,7 @@ static int spi_transfer_one_message(struct spi_master *master,
 	list_for_each_entry(xfer, &msg->transfers, transfer_list) {
 		trace_spi_transfer_start(msg, xfer);
 
-		INIT_COMPLETION(master->xfer_completion);
+		reinit_completion(&master->xfer_completion);
 
 		ret = master->transfer_one(master, msg->spi, xfer);
 		if (ret < 0) {
diff --git a/drivers/staging/iio/adc/ad7606.h b/drivers/staging/iio/adc/ad7606.h
index 9221a74efd18..93c7299e8353 100644
--- a/drivers/staging/iio/adc/ad7606.h
+++ b/drivers/staging/iio/adc/ad7606.h
@@ -42,7 +42,7 @@ struct ad7606_platform_data {
 
 /**
  * struct ad7606_chip_info - chip specifc information
- * @name:		indentification string for chip
+ * @name:		identification string for chip
  * @int_vref_mv:	the internal reference voltage
  * @channels:		channel specification
  * @num_channels:	number of channels
diff --git a/drivers/staging/iio/adc/mxs-lradc.c b/drivers/staging/iio/adc/mxs-lradc.c
index aeae76b77be5..e2dd7830b320 100644
--- a/drivers/staging/iio/adc/mxs-lradc.c
+++ b/drivers/staging/iio/adc/mxs-lradc.c
@@ -783,7 +783,7 @@ static int mxs_lradc_read_raw(struct iio_dev *iio_dev,
 	if (!ret)
 		return -EBUSY;
 
-	INIT_COMPLETION(lradc->completion);
+	reinit_completion(&lradc->completion);
 
 	/*
 	 * No buffered operation in progress, map the channel and trigger it.
diff --git a/drivers/staging/imx-drm/Kconfig b/drivers/staging/imx-drm/Kconfig
index 394254f7d6b5..5032ff7c2259 100644
--- a/drivers/staging/imx-drm/Kconfig
+++ b/drivers/staging/imx-drm/Kconfig
@@ -1,6 +1,7 @@
 config DRM_IMX
 	tristate "DRM Support for Freescale i.MX"
 	select DRM_KMS_HELPER
+	select DRM_KMS_FB_HELPER
 	select VIDEOMODE_HELPERS
 	select DRM_GEM_CMA_HELPER
 	select DRM_KMS_CMA_HELPER
diff --git a/drivers/staging/imx-drm/imx-drm-core.c b/drivers/staging/imx-drm/imx-drm-core.c
index 3d3a824f6de7..51aa9772f959 100644
--- a/drivers/staging/imx-drm/imx-drm-core.c
+++ b/drivers/staging/imx-drm/imx-drm-core.c
@@ -407,14 +407,14 @@ static int imx_drm_driver_load(struct drm_device *drm, unsigned long flags)
 
 	/*
 	 * enable drm irq mode.
-	 * - with irq_enabled = 1, we can use the vblank feature.
+	 * - with irq_enabled = true, we can use the vblank feature.
 	 *
 	 * P.S. note that we wouldn't use drm irq handler but
 	 *      just specific driver own one instead because
 	 *      drm framework supports only one irq handler and
 	 *      drivers can well take care of their interrupts
 	 */
-	drm->irq_enabled = 1;
+	drm->irq_enabled = true;
 
 	drm_mode_config_init(drm);
 	imx_drm_mode_config_init(drm);
@@ -434,11 +434,11 @@ static int imx_drm_driver_load(struct drm_device *drm, unsigned long flags)
 		goto err_init;
 
 	/*
-	 * with vblank_disable_allowed = 1, vblank interrupt will be disabled
+	 * with vblank_disable_allowed = true, vblank interrupt will be disabled
 	 * by drm timer once a current process gives up ownership of
 	 * vblank event.(after drm_vblank_put function is called)
 	 */
-	imxdrm->drm->vblank_disable_allowed = 1;
+	imxdrm->drm->vblank_disable_allowed = true;
 
 	if (!imx_drm_device_get())
 		ret = -EINVAL;
diff --git a/drivers/staging/media/lirc/TODO b/drivers/staging/media/lirc/TODO
index b6cb593f55c6..cbea5d84fed3 100644
--- a/drivers/staging/media/lirc/TODO
+++ b/drivers/staging/media/lirc/TODO
@@ -2,6 +2,11 @@
   (see drivers/media/IR/mceusb.c vs. lirc_mceusb.c in lirc cvs for an
   example of a previously completed port).
 
+- lirc_bt829 uses registers on a Mach64 VT, which has a separate kernel
+  framebuffer driver (atyfb) and userland X driver (mach64).  It can't
+  simply be converted to a normal PCI driver, but ideally it should be
+  coordinated with the other drivers.
+
 Please send patches to:
 Jarod Wilson <jarod@wilsonet.com>
 Greg Kroah-Hartman <greg@kroah.com>
diff --git a/drivers/staging/media/lirc/lirc_bt829.c b/drivers/staging/media/lirc/lirc_bt829.c
index fbbdce4c119f..30edc740ac25 100644
--- a/drivers/staging/media/lirc/lirc_bt829.c
+++ b/drivers/staging/media/lirc/lirc_bt829.c
@@ -63,7 +63,7 @@ static bool debug;
 	} while (0)
 
 static int atir_minor;
-static unsigned long pci_addr_phys;
+static phys_addr_t pci_addr_phys;
 static unsigned char *pci_addr_lin;
 
 static struct lirc_driver atir_driver;
@@ -78,11 +78,11 @@ static struct pci_dev *do_pci_probe(void)
 		pci_addr_phys = 0;
 		if (my_dev->resource[0].flags & IORESOURCE_MEM) {
 			pci_addr_phys = my_dev->resource[0].start;
-			pr_info("memory at 0x%08X\n",
-			       (unsigned int)pci_addr_phys);
+			pr_info("memory at %pa\n", &pci_addr_phys);
 		}
 		if (pci_addr_phys == 0) {
 			pr_err("no memory resource ?\n");
+			pci_dev_put(my_dev);
 			return NULL;
 		}
 	} else {
@@ -120,13 +120,20 @@ static void atir_set_use_dec(void *data)
 int init_module(void)
 {
 	struct pci_dev *pdev;
+	int rc;
 
 	pdev = do_pci_probe();
 	if (pdev == NULL)
 		return -ENODEV;
 
-	if (!atir_init_start())
-		return -ENODEV;
+	rc = pci_enable_device(pdev);
+	if (rc)
+		goto err_put_dev;
+
+	if (!atir_init_start()) {
+		rc = -ENODEV;
+		goto err_disable;
+	}
 
 	strcpy(atir_driver.name, "ATIR");
 	atir_driver.minor       = -1;
@@ -142,17 +149,31 @@ int init_module(void)
 	atir_minor = lirc_register_driver(&atir_driver);
 	if (atir_minor < 0) {
 		pr_err("failed to register driver!\n");
-		return atir_minor;
+		rc = atir_minor;
+		goto err_unmap;
 	}
 	dprintk("driver is registered on minor %d\n", atir_minor);
 
 	return 0;
+
+err_unmap:
+	iounmap(pci_addr_lin);
+err_disable:
+	pci_disable_device(pdev);
+err_put_dev:
+	pci_dev_put(pdev);
+	return rc;
 }
 
 
 void cleanup_module(void)
 {
+	struct pci_dev *pdev = to_pci_dev(atir_driver.dev);
+
 	lirc_unregister_driver(atir_minor);
+	iounmap(pci_addr_lin);
+	pci_disable_device(pdev);
+	pci_dev_put(pdev);
 }
 
 
diff --git a/drivers/staging/media/lirc/lirc_serial.c b/drivers/staging/media/lirc/lirc_serial.c
index f6bc4c91ab35..2e3a98575d47 100644
--- a/drivers/staging/media/lirc/lirc_serial.c
+++ b/drivers/staging/media/lirc/lirc_serial.c
@@ -707,7 +707,8 @@ static irqreturn_t irq_handler(int i, void *blah)
 				pr_warn("ignoring spike: %d %d %lx %lx %lx %lx\n",
 					dcd, sense,
 					tv.tv_sec, lasttv.tv_sec,
-					tv.tv_usec, lasttv.tv_usec);
+					(unsigned long)tv.tv_usec,
+					(unsigned long)lasttv.tv_usec);
 				continue;
 			}
 
@@ -719,7 +720,8 @@ static irqreturn_t irq_handler(int i, void *blah)
 				pr_warn("%d %d %lx %lx %lx %lx\n",
 					dcd, sense,
 					tv.tv_sec, lasttv.tv_sec,
-					tv.tv_usec, lasttv.tv_usec);
+					(unsigned long)tv.tv_usec,
+					(unsigned long)lasttv.tv_usec);
 				data = PULSE_MASK;
 			} else if (deltv > 15) {
 				data = PULSE_MASK; /* really long time */
@@ -728,7 +730,8 @@ static irqreturn_t irq_handler(int i, void *blah)
 					pr_warn("AIEEEE: %d %d %lx %lx %lx %lx\n",
 						dcd, sense,
 						tv.tv_sec, lasttv.tv_sec,
-						tv.tv_usec, lasttv.tv_usec);
+						(unsigned long)tv.tv_usec,
+						(unsigned long)lasttv.tv_usec);
 					/*
 					 * detecting pulse while this
 					 * MUST be a space!
diff --git a/drivers/staging/media/lirc/lirc_zilog.c b/drivers/staging/media/lirc/lirc_zilog.c
index 11d5338b4f2f..0feeaadf29dc 100644
--- a/drivers/staging/media/lirc/lirc_zilog.c
+++ b/drivers/staging/media/lirc/lirc_zilog.c
@@ -61,6 +61,9 @@
 #include <media/lirc_dev.h>
 #include <media/lirc.h>
 
+/* Max transfer size done by I2C transfer functions */
+#define MAX_XFER_SIZE  64
+
 struct IR;
 
 struct IR_rx {
@@ -941,7 +944,14 @@ static ssize_t read(struct file *filep, char *outbuf, size_t n, loff_t *ppos)
 			schedule();
 			set_current_state(TASK_INTERRUPTIBLE);
 		} else {
-			unsigned char buf[rbuf->chunk_size];
+			unsigned char buf[MAX_XFER_SIZE];
+
+			if (rbuf->chunk_size > sizeof(buf)) {
+				zilog_error("chunk_size is too big (%d)!\n",
+					    rbuf->chunk_size);
+				ret = -EINVAL;
+				break;
+			}
 			m = lirc_buffer_read(rbuf, buf);
 			if (m == rbuf->chunk_size) {
 				ret = copy_to_user((void *)outbuf+written, buf,
diff --git a/drivers/staging/media/msi3101/Kconfig b/drivers/staging/media/msi3101/Kconfig
index 76d5bbd4d93c..0c349c8595e4 100644
--- a/drivers/staging/media/msi3101/Kconfig
+++ b/drivers/staging/media/msi3101/Kconfig
@@ -1,4 +1,5 @@
 config USB_MSI3101
 	tristate "Mirics MSi3101 SDR Dongle"
 	depends on USB && VIDEO_DEV && VIDEO_V4L2
-        select VIDEOBUF2_VMALLOC
+	select VIDEOBUF2_CORE
+	select VIDEOBUF2_VMALLOC
diff --git a/drivers/staging/media/solo6x10/solo6x10-disp.c b/drivers/staging/media/solo6x10/solo6x10-disp.c
index 32d9953bc36e..145295a5db72 100644
--- a/drivers/staging/media/solo6x10/solo6x10-disp.c
+++ b/drivers/staging/media/solo6x10/solo6x10-disp.c
@@ -176,18 +176,27 @@ static void solo_vout_config(struct solo_dev *solo_dev)
 static int solo_dma_vin_region(struct solo_dev *solo_dev, u32 off,
 			       u16 val, int reg_size)
 {
-	u16 buf[64];
-	int i;
-	int ret = 0;
+	u16 *buf;
+	const int n = 64, size = n * sizeof(*buf);
+	int i, ret = 0;
+
+	buf = kmalloc(size, GFP_KERNEL);
+	if (!buf)
+		return -ENOMEM;
 
-	for (i = 0; i < sizeof(buf) >> 1; i++)
+	for (i = 0; i < n; i++)
 		buf[i] = cpu_to_le16(val);
 
-	for (i = 0; i < reg_size; i += sizeof(buf))
-		ret |= solo_p2m_dma(solo_dev, 1, buf,
-				    SOLO_MOTION_EXT_ADDR(solo_dev) + off + i,
-				    sizeof(buf), 0, 0);
+	for (i = 0; i < reg_size; i += size) {
+		ret = solo_p2m_dma(solo_dev, 1, buf,
+				   SOLO_MOTION_EXT_ADDR(solo_dev) + off + i,
+				   size, 0, 0);
+
+		if (ret)
+			break;
+	}
 
+	kfree(buf);
 	return ret;
 }
 
diff --git a/drivers/staging/media/solo6x10/solo6x10-p2m.c b/drivers/staging/media/solo6x10/solo6x10-p2m.c
index 333594189b81..7f2f2472655b 100644
--- a/drivers/staging/media/solo6x10/solo6x10-p2m.c
+++ b/drivers/staging/media/solo6x10/solo6x10-p2m.c
@@ -87,7 +87,7 @@ int solo_p2m_dma_desc(struct solo_dev *solo_dev,
 	if (mutex_lock_interruptible(&p2m_dev->mutex))
 		return -EINTR;
 
-	INIT_COMPLETION(p2m_dev->completion);
+	reinit_completion(&p2m_dev->completion);
 	p2m_dev->error = 0;
 
 	if (desc_cnt > 1 && solo_dev->type != SOLO_DEV_6110 && desc_mode) {
diff --git a/drivers/staging/media/solo6x10/solo6x10-v4l2-enc.c b/drivers/staging/media/solo6x10/solo6x10-v4l2-enc.c
index a4c589604b02..d582c5b84c14 100644
--- a/drivers/staging/media/solo6x10/solo6x10-v4l2-enc.c
+++ b/drivers/staging/media/solo6x10/solo6x10-v4l2-enc.c
@@ -95,38 +95,11 @@ static unsigned char vop_6110_pal_cif[] = {
 	0x01, 0x68, 0xce, 0x32, 0x28, 0x00, 0x00, 0x00,
 };
 
-struct vop_header {
-	/* VE_STATUS0 */
-	u32 mpeg_size:20, sad_motion_flag:1, video_motion_flag:1, vop_type:2,
-		channel:5, source_fl:1, interlace:1, progressive:1;
-
-	/* VE_STATUS1 */
-	u32 vsize:8, hsize:8, last_queue:4, nop0:8, scale:4;
-
-	/* VE_STATUS2 */
-	u32 mpeg_off;
-
-	/* VE_STATUS3 */
-	u32 jpeg_off;
-
-	/* VE_STATUS4 */
-	u32 jpeg_size:20, interval:10, nop1:2;
-
-	/* VE_STATUS5/6 */
-	u32 sec, usec;
-
-	/* VE_STATUS7/8/9 */
-	u32 nop2[3];
-
-	/* VE_STATUS10 */
-	u32 mpeg_size_alt:20, nop3:12;
-
-	u32 end_nops[5];
-} __packed;
+typedef __le32 vop_header[16];
 
 struct solo_enc_buf {
 	enum solo_enc_types	type;
-	struct vop_header	*vh;
+	const vop_header	*vh;
 	int			motion;
 };
 
@@ -346,7 +319,7 @@ static int enc_get_mpeg_dma(struct solo_dev *solo_dev, dma_addr_t dma,
 /* Build a descriptor queue out of an SG list and send it to the P2M for
  * processing. */
 static int solo_send_desc(struct solo_enc_dev *solo_enc, int skip,
-			  struct vb2_dma_sg_desc *vbuf, int off, int size,
+			  struct sg_table *vbuf, int off, int size,
 			  unsigned int base, unsigned int base_size)
 {
 	struct solo_dev *solo_dev = solo_enc->solo_dev;
@@ -359,7 +332,7 @@ static int solo_send_desc(struct solo_enc_dev *solo_enc, int skip,
 
 	solo_enc->desc_count = 1;
 
-	for_each_sg(vbuf->sglist, sg, vbuf->num_pages, i) {
+	for_each_sg(vbuf->sgl, sg, vbuf->nents, i) {
 		struct solo_p2m_desc *desc;
 		dma_addr_t dma;
 		int len;
@@ -430,84 +403,145 @@ static int solo_send_desc(struct solo_enc_dev *solo_enc, int skip,
 				 solo_enc->desc_count - 1);
 }
 
+/* Extract values from VOP header - VE_STATUSxx */
+static inline int vop_interlaced(const vop_header *vh)
+{
+	return (__le32_to_cpu((*vh)[0]) >> 30) & 1;
+}
+
+static inline u8 vop_channel(const vop_header *vh)
+{
+	return (__le32_to_cpu((*vh)[0]) >> 24) & 0x1F;
+}
+
+static inline u8 vop_type(const vop_header *vh)
+{
+	return (__le32_to_cpu((*vh)[0]) >> 22) & 3;
+}
+
+static inline u32 vop_mpeg_size(const vop_header *vh)
+{
+	return __le32_to_cpu((*vh)[0]) & 0xFFFFF;
+}
+
+static inline u8 vop_hsize(const vop_header *vh)
+{
+	return (__le32_to_cpu((*vh)[1]) >> 8) & 0xFF;
+}
+
+static inline u8 vop_vsize(const vop_header *vh)
+{
+	return __le32_to_cpu((*vh)[1]) & 0xFF;
+}
+
+static inline u32 vop_mpeg_offset(const vop_header *vh)
+{
+	return __le32_to_cpu((*vh)[2]);
+}
+
+static inline u32 vop_jpeg_offset(const vop_header *vh)
+{
+	return __le32_to_cpu((*vh)[3]);
+}
+
+static inline u32 vop_jpeg_size(const vop_header *vh)
+{
+	return __le32_to_cpu((*vh)[4]) & 0xFFFFF;
+}
+
+static inline u32 vop_sec(const vop_header *vh)
+{
+	return __le32_to_cpu((*vh)[5]);
+}
+
+static inline u32 vop_usec(const vop_header *vh)
+{
+	return __le32_to_cpu((*vh)[6]);
+}
+
 static int solo_fill_jpeg(struct solo_enc_dev *solo_enc,
-		struct vb2_buffer *vb, struct vop_header *vh)
+			  struct vb2_buffer *vb, const vop_header *vh)
 {
 	struct solo_dev *solo_dev = solo_enc->solo_dev;
-	struct vb2_dma_sg_desc *vbuf = vb2_dma_sg_plane_desc(vb, 0);
+	struct sg_table *vbuf = vb2_dma_sg_plane_desc(vb, 0);
 	int frame_size;
 	int ret;
 
 	vb->v4l2_buf.flags |= V4L2_BUF_FLAG_KEYFRAME;
 
-	if (vb2_plane_size(vb, 0) < vh->jpeg_size + solo_enc->jpeg_len)
+	if (vb2_plane_size(vb, 0) < vop_jpeg_size(vh) + solo_enc->jpeg_len)
 		return -EIO;
 
-	sg_copy_from_buffer(vbuf->sglist, vbuf->num_pages,
-			solo_enc->jpeg_header,
-			solo_enc->jpeg_len);
-
-	frame_size = (vh->jpeg_size + solo_enc->jpeg_len + (DMA_ALIGN - 1))
+	frame_size = (vop_jpeg_size(vh) + solo_enc->jpeg_len + (DMA_ALIGN - 1))
 		& ~(DMA_ALIGN - 1);
-	vb2_set_plane_payload(vb, 0, vh->jpeg_size + solo_enc->jpeg_len);
+	vb2_set_plane_payload(vb, 0, vop_jpeg_size(vh) + solo_enc->jpeg_len);
 
-	dma_map_sg(&solo_dev->pdev->dev, vbuf->sglist, vbuf->num_pages,
+	/* may discard all previous data in vbuf->sgl */
+	dma_map_sg(&solo_dev->pdev->dev, vbuf->sgl, vbuf->nents,
 			DMA_FROM_DEVICE);
-	ret = solo_send_desc(solo_enc, solo_enc->jpeg_len, vbuf, vh->jpeg_off,
-			frame_size, SOLO_JPEG_EXT_ADDR(solo_dev),
-			SOLO_JPEG_EXT_SIZE(solo_dev));
-	dma_unmap_sg(&solo_dev->pdev->dev, vbuf->sglist, vbuf->num_pages,
+	ret = solo_send_desc(solo_enc, solo_enc->jpeg_len, vbuf,
+			     vop_jpeg_offset(vh) - SOLO_JPEG_EXT_ADDR(solo_dev),
+			     frame_size, SOLO_JPEG_EXT_ADDR(solo_dev),
+			     SOLO_JPEG_EXT_SIZE(solo_dev));
+	dma_unmap_sg(&solo_dev->pdev->dev, vbuf->sgl, vbuf->nents,
 			DMA_FROM_DEVICE);
+
+	/* add the header only after dma_unmap_sg() */
+	sg_copy_from_buffer(vbuf->sgl, vbuf->nents,
+			    solo_enc->jpeg_header, solo_enc->jpeg_len);
+
 	return ret;
 }
 
 static int solo_fill_mpeg(struct solo_enc_dev *solo_enc,
-		struct vb2_buffer *vb, struct vop_header *vh)
+		struct vb2_buffer *vb, const vop_header *vh)
 {
 	struct solo_dev *solo_dev = solo_enc->solo_dev;
-	struct vb2_dma_sg_desc *vbuf = vb2_dma_sg_plane_desc(vb, 0);
+	struct sg_table *vbuf = vb2_dma_sg_plane_desc(vb, 0);
 	int frame_off, frame_size;
 	int skip = 0;
 	int ret;
 
-	if (vb2_plane_size(vb, 0) < vh->mpeg_size)
+	if (vb2_plane_size(vb, 0) < vop_mpeg_size(vh))
 		return -EIO;
 
 	/* If this is a key frame, add extra header */
-	if (!vh->vop_type) {
-		sg_copy_from_buffer(vbuf->sglist, vbuf->num_pages,
-				solo_enc->vop,
-				solo_enc->vop_len);
-
+	vb->v4l2_buf.flags &= ~(V4L2_BUF_FLAG_KEYFRAME | V4L2_BUF_FLAG_PFRAME | V4L2_BUF_FLAG_BFRAME);
+	if (!vop_type(vh)) {
 		skip = solo_enc->vop_len;
-
 		vb->v4l2_buf.flags |= V4L2_BUF_FLAG_KEYFRAME;
-		vb2_set_plane_payload(vb, 0, vh->mpeg_size + solo_enc->vop_len);
+		vb2_set_plane_payload(vb, 0, vop_mpeg_size(vh) + solo_enc->vop_len);
 	} else {
 		vb->v4l2_buf.flags |= V4L2_BUF_FLAG_PFRAME;
-		vb2_set_plane_payload(vb, 0, vh->mpeg_size);
+		vb2_set_plane_payload(vb, 0, vop_mpeg_size(vh));
 	}
 
 	/* Now get the actual mpeg payload */
-	frame_off = (vh->mpeg_off + sizeof(*vh))
+	frame_off = (vop_mpeg_offset(vh) - SOLO_MP4E_EXT_ADDR(solo_dev) + sizeof(*vh))
 		% SOLO_MP4E_EXT_SIZE(solo_dev);
-	frame_size = (vh->mpeg_size + skip + (DMA_ALIGN - 1))
+	frame_size = (vop_mpeg_size(vh) + skip + (DMA_ALIGN - 1))
 		& ~(DMA_ALIGN - 1);
 
-	dma_map_sg(&solo_dev->pdev->dev, vbuf->sglist, vbuf->num_pages,
+	/* may discard all previous data in vbuf->sgl */
+	dma_map_sg(&solo_dev->pdev->dev, vbuf->sgl, vbuf->nents,
 			DMA_FROM_DEVICE);
 	ret = solo_send_desc(solo_enc, skip, vbuf, frame_off, frame_size,
 			SOLO_MP4E_EXT_ADDR(solo_dev),
 			SOLO_MP4E_EXT_SIZE(solo_dev));
-	dma_unmap_sg(&solo_dev->pdev->dev, vbuf->sglist, vbuf->num_pages,
+	dma_unmap_sg(&solo_dev->pdev->dev, vbuf->sgl, vbuf->nents,
 			DMA_FROM_DEVICE);
+
+	/* add the header only after dma_unmap_sg() */
+	if (!vop_type(vh))
+		sg_copy_from_buffer(vbuf->sgl, vbuf->nents,
+				    solo_enc->vop, solo_enc->vop_len);
 	return ret;
 }
 
 static int solo_enc_fillbuf(struct solo_enc_dev *solo_enc,
 			    struct vb2_buffer *vb, struct solo_enc_buf *enc_buf)
 {
-	struct vop_header *vh = enc_buf->vh;
+	const vop_header *vh = enc_buf->vh;
 	int ret;
 
 	/* Check for motion flags */
@@ -531,8 +565,8 @@ static int solo_enc_fillbuf(struct solo_enc_dev *solo_enc,
 
 	if (!ret) {
 		vb->v4l2_buf.sequence = solo_enc->sequence++;
-		vb->v4l2_buf.timestamp.tv_sec = vh->sec;
-		vb->v4l2_buf.timestamp.tv_usec = vh->usec;
+		vb->v4l2_buf.timestamp.tv_sec = vop_sec(vh);
+		vb->v4l2_buf.timestamp.tv_usec = vop_usec(vh);
 	}
 
 	vb2_buffer_done(vb, ret ? VB2_BUF_STATE_ERROR : VB2_BUF_STATE_DONE);
@@ -605,15 +639,13 @@ static void solo_handle_ring(struct solo_dev *solo_dev)
 
 		/* FAIL... */
 		if (enc_get_mpeg_dma(solo_dev, solo_dev->vh_dma, off,
-				     sizeof(struct vop_header)))
+				     sizeof(vop_header)))
 			continue;
 
-		enc_buf.vh = (struct vop_header *)solo_dev->vh_buf;
-		enc_buf.vh->mpeg_off -= SOLO_MP4E_EXT_ADDR(solo_dev);
-		enc_buf.vh->jpeg_off -= SOLO_JPEG_EXT_ADDR(solo_dev);
+		enc_buf.vh = solo_dev->vh_buf;
 
 		/* Sanity check */
-		if (enc_buf.vh->mpeg_off != off)
+		if (vop_mpeg_offset(enc_buf.vh) != SOLO_MP4E_EXT_ADDR(solo_dev) + off)
 			continue;
 
 		if (solo_motion_detected(solo_enc))
@@ -1329,7 +1361,7 @@ int solo_enc_v4l2_init(struct solo_dev *solo_dev, unsigned nr)
 
 	init_waitqueue_head(&solo_dev->ring_thread_wait);
 
-	solo_dev->vh_size = sizeof(struct vop_header);
+	solo_dev->vh_size = sizeof(vop_header);
 	solo_dev->vh_buf = pci_alloc_consistent(solo_dev->pdev,
 						solo_dev->vh_size,
 						&solo_dev->vh_dma);
diff --git a/drivers/staging/media/solo6x10/solo6x10.h b/drivers/staging/media/solo6x10/solo6x10.h
index 6f91d2e34b2a..f1bbb8cb74e6 100644
--- a/drivers/staging/media/solo6x10/solo6x10.h
+++ b/drivers/staging/media/solo6x10/solo6x10.h
@@ -94,7 +94,6 @@
 #define SOLO_ENC_MODE_HD1		1
 #define SOLO_ENC_MODE_D1		9
 
-#define SOLO_DEFAULT_GOP		30
 #define SOLO_DEFAULT_QP			3
 
 #ifndef V4L2_BUF_FLAG_MOTION_ON
diff --git a/drivers/staging/tidspbridge/core/sync.c b/drivers/staging/tidspbridge/core/sync.c
index 7bb550acaf4a..743ff09d82d2 100644
--- a/drivers/staging/tidspbridge/core/sync.c
+++ b/drivers/staging/tidspbridge/core/sync.c
@@ -72,7 +72,7 @@ int sync_wait_on_multiple_events(struct sync_object **events,
 	spin_lock_bh(&sync_lock);
 	for (i = 0; i < count; i++) {
 		if (completion_done(&events[i]->comp)) {
-			INIT_COMPLETION(events[i]->comp);
+			reinit_completion(&events[i]->comp);
 			*index = i;
 			spin_unlock_bh(&sync_lock);
 			status = 0;
@@ -92,7 +92,7 @@ int sync_wait_on_multiple_events(struct sync_object **events,
 	spin_lock_bh(&sync_lock);
 	for (i = 0; i < count; i++) {
 		if (completion_done(&events[i]->comp)) {
-			INIT_COMPLETION(events[i]->comp);
+			reinit_completion(&events[i]->comp);
 			*index = i;
 			status = 0;
 		}
diff --git a/drivers/staging/tidspbridge/include/dspbridge/sync.h b/drivers/staging/tidspbridge/include/dspbridge/sync.h
index 58a0d5c5543d..fc19b9707087 100644
--- a/drivers/staging/tidspbridge/include/dspbridge/sync.h
+++ b/drivers/staging/tidspbridge/include/dspbridge/sync.h
@@ -59,7 +59,7 @@ static inline void sync_init_event(struct sync_object *event)
 
 static inline void sync_reset_event(struct sync_object *event)
 {
-	INIT_COMPLETION(event->comp);
+	reinit_completion(&event->comp);
 	event->multi_comp = NULL;
 }
 
diff --git a/drivers/staging/tidspbridge/rmgr/drv_interface.c b/drivers/staging/tidspbridge/rmgr/drv_interface.c
index 6d04eb48bfbc..1aa4a3fd0f1b 100644
--- a/drivers/staging/tidspbridge/rmgr/drv_interface.c
+++ b/drivers/staging/tidspbridge/rmgr/drv_interface.c
@@ -332,7 +332,7 @@ static void bridge_recover(struct work_struct *work)
 	struct dev_object *dev;
 	struct cfg_devnode *dev_node;
 	if (atomic_read(&bridge_cref)) {
-		INIT_COMPLETION(bridge_comp);
+		reinit_completion(&bridge_comp);
 		while (!wait_for_completion_timeout(&bridge_comp,
 						msecs_to_jiffies(REC_TIMEOUT)))
 			pr_info("%s:%d handle(s) still opened\n",
@@ -348,7 +348,7 @@ static void bridge_recover(struct work_struct *work)
 
 void bridge_recover_schedule(void)
 {
-	INIT_COMPLETION(bridge_open_comp);
+	reinit_completion(&bridge_open_comp);
 	recover = true;
 	queue_work(bridge_rec_queue, &bridge_recovery_work);
 }
@@ -389,7 +389,7 @@ static int omap3_bridge_startup(struct platform_device *pdev)
 #ifdef CONFIG_TIDSPBRIDGE_RECOVERY
 	bridge_rec_queue = create_workqueue("bridge_rec_queue");
 	INIT_WORK(&bridge_recovery_work, bridge_recover);
-	INIT_COMPLETION(bridge_comp);
+	reinit_completion(&bridge_comp);
 #endif
 
 #ifdef CONFIG_PM
diff --git a/drivers/thermal/Kconfig b/drivers/thermal/Kconfig
index 5ef596765060..f35a1f75b15b 100644
--- a/drivers/thermal/Kconfig
+++ b/drivers/thermal/Kconfig
@@ -56,7 +56,7 @@ config THERMAL_DEFAULT_GOV_USER_SPACE
 	select THERMAL_GOV_USER_SPACE
 	help
 	  Select this if you want to let the user space manage the
-	  lpatform thermals.
+	  platform thermals.
 
 endchoice
 
@@ -69,6 +69,7 @@ config THERMAL_GOV_STEP_WISE
 	bool "Step_wise thermal governor"
 	help
 	  Enable this to manage platform thermals using a simple linear
+	  governor.
 
 config THERMAL_GOV_USER_SPACE
 	bool "User_space thermal governor"
@@ -116,14 +117,14 @@ config SPEAR_THERMAL
 	depends on OF
 	help
 	  Enable this to plug the SPEAr thermal sensor driver into the Linux
-	  thermal framework
+	  thermal framework.
 
 config RCAR_THERMAL
 	tristate "Renesas R-Car thermal driver"
 	depends on ARCH_SHMOBILE
 	help
 	  Enable this to plug the R-Car thermal sensor driver into the Linux
-	  thermal framework
+	  thermal framework.
 
 config KIRKWOOD_THERMAL
 	tristate "Temperature sensor on Marvell Kirkwood SoCs"
diff --git a/drivers/thermal/cpu_cooling.c b/drivers/thermal/cpu_cooling.c
index d17902886c3f..02a46f23d14c 100644
--- a/drivers/thermal/cpu_cooling.c
+++ b/drivers/thermal/cpu_cooling.c
@@ -469,10 +469,10 @@ cpufreq_cooling_register(const struct cpumask *clip_cpus)
 
 	cool_dev = thermal_cooling_device_register(dev_name, cpufreq_dev,
 						   &cpufreq_cooling_ops);
-	if (!cool_dev) {
+	if (IS_ERR(cool_dev)) {
 		release_idr(&cpufreq_idr, cpufreq_dev->id);
 		kfree(cpufreq_dev);
-		return ERR_PTR(-EINVAL);
+		return cool_dev;
 	}
 	cpufreq_dev->cool_dev = cool_dev;
 	cpufreq_dev->cpufreq_state = 0;
diff --git a/drivers/thermal/intel_powerclamp.c b/drivers/thermal/intel_powerclamp.c
index b40b37cd25e0..8f181b3f842b 100644
--- a/drivers/thermal/intel_powerclamp.c
+++ b/drivers/thermal/intel_powerclamp.c
@@ -675,6 +675,11 @@ static const struct x86_cpu_id intel_powerclamp_ids[] = {
 	{ X86_VENDOR_INTEL, 6, 0x2e},
 	{ X86_VENDOR_INTEL, 6, 0x2f},
 	{ X86_VENDOR_INTEL, 6, 0x3a},
+	{ X86_VENDOR_INTEL, 6, 0x3c},
+	{ X86_VENDOR_INTEL, 6, 0x3e},
+	{ X86_VENDOR_INTEL, 6, 0x3f},
+	{ X86_VENDOR_INTEL, 6, 0x45},
+	{ X86_VENDOR_INTEL, 6, 0x46},
 	{}
 };
 MODULE_DEVICE_TABLE(x86cpu, intel_powerclamp_ids);
@@ -758,21 +763,39 @@ static int powerclamp_init(void)
 	/* probe cpu features and ids here */
 	retval = powerclamp_probe();
 	if (retval)
-		return retval;
+		goto exit_free;
+
 	/* set default limit, maybe adjusted during runtime based on feedback */
 	window_size = 2;
 	register_hotcpu_notifier(&powerclamp_cpu_notifier);
+
 	powerclamp_thread = alloc_percpu(struct task_struct *);
+	if (!powerclamp_thread) {
+		retval = -ENOMEM;
+		goto exit_unregister;
+	}
+
 	cooling_dev = thermal_cooling_device_register("intel_powerclamp", NULL,
 						&powerclamp_cooling_ops);
-	if (IS_ERR(cooling_dev))
-		return -ENODEV;
+	if (IS_ERR(cooling_dev)) {
+		retval = -ENODEV;
+		goto exit_free_thread;
+	}
 
 	if (!duration)
 		duration = jiffies_to_msecs(DEFAULT_DURATION_JIFFIES);
+
 	powerclamp_create_debug_files();
 
 	return 0;
+
+exit_free_thread:
+	free_percpu(powerclamp_thread);
+exit_unregister:
+	unregister_hotcpu_notifier(&powerclamp_cpu_notifier);
+exit_free:
+	kfree(cpu_clamping_mask);
+	return retval;
 }
 module_init(powerclamp_init);
 
diff --git a/drivers/thermal/thermal_core.c b/drivers/thermal/thermal_core.c
index 4962a6aaf295..f1d511a9475b 100644
--- a/drivers/thermal/thermal_core.c
+++ b/drivers/thermal/thermal_core.c
@@ -247,10 +247,11 @@ static void bind_cdev(struct thermal_cooling_device *cdev)
 		if (!pos->tzp && !pos->ops->bind)
 			continue;
 
-		if (!pos->tzp && pos->ops->bind) {
+		if (pos->ops->bind) {
 			ret = pos->ops->bind(pos, cdev);
 			if (ret)
 				print_bind_err_msg(pos, cdev, ret);
+			continue;
 		}
 
 		tzp = pos->tzp;
@@ -282,8 +283,8 @@ static void bind_tz(struct thermal_zone_device *tz)
 
 	mutex_lock(&thermal_list_lock);
 
-	/* If there is no platform data, try to use ops->bind */
-	if (!tzp && tz->ops->bind) {
+	/* If there is ops->bind, try to use ops->bind */
+	if (tz->ops->bind) {
 		list_for_each_entry(pos, &thermal_cdev_list, node) {
 			ret = tz->ops->bind(tz, pos);
 			if (ret)
@@ -1038,7 +1039,8 @@ static void thermal_release(struct device *dev)
 		     sizeof("thermal_zone") - 1)) {
 		tz = to_thermal_zone(dev);
 		kfree(tz);
-	} else {
+	} else if(!strncmp(dev_name(dev), "cooling_device",
+			sizeof("cooling_device") - 1)){
 		cdev = to_cooling_device(dev);
 		kfree(cdev);
 	}
@@ -1606,15 +1608,17 @@ exit:
 EXPORT_SYMBOL_GPL(thermal_zone_get_zone_by_name);
 
 #ifdef CONFIG_NET
+static const struct genl_multicast_group thermal_event_mcgrps[] = {
+	{ .name = THERMAL_GENL_MCAST_GROUP_NAME, },
+};
+
 static struct genl_family thermal_event_genl_family = {
 	.id = GENL_ID_GENERATE,
 	.name = THERMAL_GENL_FAMILY_NAME,
 	.version = THERMAL_GENL_VERSION,
 	.maxattr = THERMAL_GENL_ATTR_MAX,
-};
-
-static struct genl_multicast_group thermal_event_mcgrp = {
-	.name = THERMAL_GENL_MCAST_GROUP_NAME,
+	.mcgrps = thermal_event_mcgrps,
+	.n_mcgrps = ARRAY_SIZE(thermal_event_mcgrps),
 };
 
 int thermal_generate_netlink_event(struct thermal_zone_device *tz,
@@ -1675,7 +1679,8 @@ int thermal_generate_netlink_event(struct thermal_zone_device *tz,
 		return result;
 	}
 
-	result = genlmsg_multicast(skb, 0, thermal_event_mcgrp.id, GFP_ATOMIC);
+	result = genlmsg_multicast(&thermal_event_genl_family, skb, 0,
+				   0, GFP_ATOMIC);
 	if (result)
 		dev_err(&tz->device, "Failed to send netlink event:%d", result);
 
@@ -1685,17 +1690,7 @@ EXPORT_SYMBOL_GPL(thermal_generate_netlink_event);
 
 static int genetlink_init(void)
 {
-	int result;
-
-	result = genl_register_family(&thermal_event_genl_family);
-	if (result)
-		return result;
-
-	result = genl_register_mc_group(&thermal_event_genl_family,
-					&thermal_event_mcgrp);
-	if (result)
-		genl_unregister_family(&thermal_event_genl_family);
-	return result;
+	return genl_register_family(&thermal_event_genl_family);
 }
 
 static void genetlink_exit(void)
diff --git a/drivers/tty/hvc/hvc_xen.c b/drivers/tty/hvc/hvc_xen.c
index c193af6a628f..636c9baad7a5 100644
--- a/drivers/tty/hvc/hvc_xen.c
+++ b/drivers/tty/hvc/hvc_xen.c
@@ -183,7 +183,7 @@ static int dom0_write_console(uint32_t vtermno, const char *str, int len)
 {
 	int rc = HYPERVISOR_console_io(CONSOLEIO_write, len, (char *)str);
 	if (rc < 0)
-		return 0;
+		return rc;
 
 	return len;
 }
@@ -642,7 +642,22 @@ struct console xenboot_console = {
 
 void xen_raw_console_write(const char *str)
 {
-	dom0_write_console(0, str, strlen(str));
+	ssize_t len = strlen(str);
+	int rc = 0;
+
+	if (xen_domain()) {
+		rc = dom0_write_console(0, str, len);
+#ifdef CONFIG_X86
+		if (rc == -ENOSYS && xen_hvm_domain())
+			goto outb_print;
+
+	} else if (xen_cpuid_base()) {
+		int i;
+outb_print:
+		for (i = 0; i < len; i++)
+			outb(str[i], 0xe9);
+#endif
+	}
 }
 
 void xen_raw_printk(const char *fmt, ...)
diff --git a/drivers/tty/metag_da.c b/drivers/tty/metag_da.c
index 0e888621f484..7332e2ca4615 100644
--- a/drivers/tty/metag_da.c
+++ b/drivers/tty/metag_da.c
@@ -495,7 +495,7 @@ static int dashtty_write(struct tty_struct *tty, const unsigned char *buf,
 	count = dport->xmit_cnt;
 	/* xmit buffer no longer empty? */
 	if (count)
-		INIT_COMPLETION(dport->xmit_empty);
+		reinit_completion(&dport->xmit_empty);
 	mutex_unlock(&dport->xmit_lock);
 
 	if (total) {
diff --git a/drivers/tty/serial/sh-sci.c b/drivers/tty/serial/sh-sci.c
index 537750261aaa..7d8103cd3e2e 100644
--- a/drivers/tty/serial/sh-sci.c
+++ b/drivers/tty/serial/sh-sci.c
@@ -1433,7 +1433,7 @@ static void work_fn_rx(struct work_struct *work)
 	desc = s->desc_rx[new];
 
 	if (dma_async_is_tx_complete(s->chan_rx, s->active_rx, NULL, NULL) !=
-	    DMA_SUCCESS) {
+	    DMA_COMPLETE) {
 		/* Handle incomplete DMA receive */
 		struct dma_chan *chan = s->chan_rx;
 		struct shdma_desc *sh_desc = container_of(desc,
diff --git a/drivers/usb/c67x00/c67x00-sched.c b/drivers/usb/c67x00/c67x00-sched.c
index aa491627a45b..892cc96466eb 100644
--- a/drivers/usb/c67x00/c67x00-sched.c
+++ b/drivers/usb/c67x00/c67x00-sched.c
@@ -344,7 +344,7 @@ void c67x00_endpoint_disable(struct usb_hcd *hcd, struct usb_host_endpoint *ep)
 		/* it could happen that we reinitialize this completion, while
 		 * somebody was waiting for that completion.  The timeout and
 		 * while loop handle such cases, but this might be improved */
-		INIT_COMPLETION(c67x00->endpoint_disable);
+		reinit_completion(&c67x00->endpoint_disable);
 		c67x00_sched_kick(c67x00);
 		wait_for_completion_timeout(&c67x00->endpoint_disable, 1 * HZ);
 
diff --git a/drivers/usb/core/Kconfig b/drivers/usb/core/Kconfig
index db535b0aa172..fed7f68d025d 100644
--- a/drivers/usb/core/Kconfig
+++ b/drivers/usb/core/Kconfig
@@ -28,7 +28,7 @@ config USB_DEFAULT_PERSIST
 	bool "Enable USB persist by default"
 	default y
 	help
-	  Say N here if you don't want USB power session persistance
+	  Say N here if you don't want USB power session persistence
 	  enabled by default.  If you say N it will make suspended USB
 	  devices that lose power get reenumerated as if they had been
 	  unplugged, causing any mounted filesystems to be lost.  The
diff --git a/drivers/usb/gadget/f_fs.c b/drivers/usb/gadget/f_fs.c
index 44cf775a8627..774e8b89cdb5 100644
--- a/drivers/usb/gadget/f_fs.c
+++ b/drivers/usb/gadget/f_fs.c
@@ -373,7 +373,7 @@ static int __ffs_ep0_queue_wait(struct ffs_data *ffs, char *data, size_t len)
 	if (req->buf == NULL)
 		req->buf = (void *)0xDEADBABE;
 
-	INIT_COMPLETION(ffs->ep0req_completion);
+	reinit_completion(&ffs->ep0req_completion);
 
 	ret = usb_ep_queue(ffs->gadget->ep0, req, GFP_ATOMIC);
 	if (unlikely(ret < 0))
diff --git a/drivers/usb/serial/mos7720.c b/drivers/usb/serial/mos7720.c
index 84657e07dc5d..439c951f261b 100644
--- a/drivers/usb/serial/mos7720.c
+++ b/drivers/usb/serial/mos7720.c
@@ -455,7 +455,7 @@ static int parport_prologue(struct parport *pp)
 		return -1;
 	}
 	mos_parport->msg_pending = true;   /* synch usb call pending */
-	INIT_COMPLETION(mos_parport->syncmsg_compl);
+	reinit_completion(&mos_parport->syncmsg_compl);
 	spin_unlock(&release_lock);
 
 	mutex_lock(&mos_parport->serial->disc_mutex);
diff --git a/drivers/video/68328fb.c b/drivers/video/68328fb.c
index fa44fbed397d..552258c8f99d 100644
--- a/drivers/video/68328fb.c
+++ b/drivers/video/68328fb.c
@@ -478,11 +478,10 @@ int __init mc68x328fb_init(void)
 		return -EINVAL;
 	}
 
-	printk(KERN_INFO
-		"fb%d: %s frame buffer device\n", fb_info.node,	fb_info.fix.id);
-	printk(KERN_INFO
-		"fb%d: %dx%dx%d at 0x%08lx\n", fb_info.node,
-		mc68x328fb_default.xres_virtual, mc68x328fb_default.yres_virtual,
+	fb_info(&fb_info, "%s frame buffer device\n", fb_info.fix.id);
+	fb_info(&fb_info, "%dx%dx%d at 0x%08lx\n",
+		mc68x328fb_default.xres_virtual,
+		mc68x328fb_default.yres_virtual,
 		1 << mc68x328fb_default.bits_per_pixel, videomemory);
 
 	return 0;
diff --git a/drivers/video/Kconfig b/drivers/video/Kconfig
index 84b685f7ab6e..4f2e1b35eb38 100644
--- a/drivers/video/Kconfig
+++ b/drivers/video/Kconfig
@@ -19,10 +19,10 @@ source "drivers/char/agp/Kconfig"
 
 source "drivers/gpu/vga/Kconfig"
 
-source "drivers/gpu/drm/Kconfig"
-
 source "drivers/gpu/host1x/Kconfig"
 
+source "drivers/gpu/drm/Kconfig"
+
 config VGASTATE
        tristate
        default n
@@ -996,6 +996,8 @@ config FB_ATMEL
 	select FB_CFB_FILLRECT
 	select FB_CFB_COPYAREA
 	select FB_CFB_IMAGEBLIT
+	select FB_MODE_HELPERS
+	select VIDEOMODE_HELPERS
 	help
 	  This enables support for the AT91/AT32 LCD Controller.
 
diff --git a/drivers/video/amba-clcd.c b/drivers/video/amba-clcd.c
index afe4702a5528..14d6b3793e0a 100644
--- a/drivers/video/amba-clcd.c
+++ b/drivers/video/amba-clcd.c
@@ -545,7 +545,7 @@ static int clcdfb_register(struct clcd_fb *fb)
 
 static int clcdfb_probe(struct amba_device *dev, const struct amba_id *id)
 {
-	struct clcd_board *board = dev->dev.platform_data;
+	struct clcd_board *board = dev_get_platdata(&dev->dev);
 	struct clcd_fb *fb;
 	int ret;
 
@@ -599,8 +599,6 @@ static int clcdfb_remove(struct amba_device *dev)
 {
 	struct clcd_fb *fb = amba_get_drvdata(dev);
 
-	amba_set_drvdata(dev, NULL);
-
 	clcdfb_disable(fb);
 	unregister_framebuffer(&fb->fb);
 	if (fb->fb.cmap.len)
diff --git a/drivers/video/amifb.c b/drivers/video/amifb.c
index a6780eecff0e..0dac36ce09d6 100644
--- a/drivers/video/amifb.c
+++ b/drivers/video/amifb.c
@@ -3742,13 +3742,12 @@ default_chipset:
 	if (err)
 		goto unset_drvdata;
 
-	printk("fb%d: %s frame buffer device, using %dK of video memory\n",
-	       info->node, info->fix.id, info->fix.smem_len>>10);
+	fb_info(info, "%s frame buffer device, using %dK of video memory\n",
+		info->fix.id, info->fix.smem_len>>10);
 
 	return 0;
 
 unset_drvdata:
-	dev_set_drvdata(&pdev->dev, NULL);
 	fb_dealloc_cmap(&info->cmap);
 free_irq:
 	free_irq(IRQ_AMIGA_COPPER, info->par);
@@ -3768,7 +3767,6 @@ static int __exit amifb_remove(struct platform_device *pdev)
 	struct fb_info *info = dev_get_drvdata(&pdev->dev);
 
 	unregister_framebuffer(info);
-	dev_set_drvdata(&pdev->dev, NULL);
 	fb_dealloc_cmap(&info->cmap);
 	free_irq(IRQ_AMIGA_COPPER, info->par);
 	custom.dmacon = DMAF_ALL | DMAF_MASTER;
diff --git a/drivers/video/arcfb.c b/drivers/video/arcfb.c
index e43401afdd03..1b0b233b8b39 100644
--- a/drivers/video/arcfb.c
+++ b/drivers/video/arcfb.c
@@ -556,9 +556,8 @@ static int arcfb_probe(struct platform_device *dev)
 			goto err1;
 		}
 	}
-	printk(KERN_INFO
-	       "fb%d: Arc frame buffer device, using %dK of video memory\n",
-	       info->node, videomemorysize >> 10);
+	fb_info(info, "Arc frame buffer device, using %dK of video memory\n",
+		videomemorysize >> 10);
 
 	/* this inits the lcd but doesn't clear dirty pixels */
 	for (i = 0; i < num_cols * num_rows; i++) {
@@ -572,8 +571,7 @@ static int arcfb_probe(struct platform_device *dev)
 	/* if we were told to splash the screen, we just clear it */
 	if (!nosplash) {
 		for (i = 0; i < num_cols * num_rows; i++) {
-			printk(KERN_INFO "fb%d: splashing lcd %d\n",
-				info->node, i);
+			fb_info(info, "splashing lcd %d\n", i);
 			ks108_set_start_line(par, i, 0);
 			ks108_clear_lcd(par, i);
 		}
diff --git a/drivers/video/arkfb.c b/drivers/video/arkfb.c
index 94a51f1ef904..a6b29bd4a12a 100644
--- a/drivers/video/arkfb.c
+++ b/drivers/video/arkfb.c
@@ -137,8 +137,7 @@ static void arkfb_settile(struct fb_info *info, struct fb_tilemap *map)
 
 	if ((map->width != 8) || (map->height != 16) ||
 	    (map->depth != 1) || (map->length != 256)) {
-	    	printk(KERN_ERR "fb%d: unsupported font parameters: width %d, "
-		       "height %d, depth %d, length %d\n", info->node,
+		fb_err(info, "unsupported font parameters: width %d, height %d, depth %d, length %d\n",
 		       map->width, map->height, map->depth, map->length);
 		return;
 	}
@@ -517,7 +516,7 @@ static void ark_set_pixclock(struct fb_info *info, u32 pixclock)
 
 	int rv = dac_set_freq(par->dac, 0, 1000000000 / pixclock);
 	if (rv < 0) {
-		printk(KERN_ERR "fb%d: cannot set requested pixclock, keeping old value\n", info->node);
+		fb_err(info, "cannot set requested pixclock, keeping old value\n");
 		return;
 	}
 
@@ -584,7 +583,7 @@ static int arkfb_check_var(struct fb_var_screeninfo *var, struct fb_info *info)
 	rv = svga_match_format (arkfb_formats, var, NULL);
 	if (rv < 0)
 	{
-		printk(KERN_ERR "fb%d: unsupported mode requested\n", info->node);
+		fb_err(info, "unsupported mode requested\n");
 		return rv;
 	}
 
@@ -604,14 +603,15 @@ static int arkfb_check_var(struct fb_var_screeninfo *var, struct fb_info *info)
 	mem = ((var->bits_per_pixel * var->xres_virtual) >> 3) * var->yres_virtual;
 	if (mem > info->screen_size)
 	{
-		printk(KERN_ERR "fb%d: not enough framebuffer memory (%d kB requested , %d kB available)\n", info->node, mem >> 10, (unsigned int) (info->screen_size >> 10));
+		fb_err(info, "not enough framebuffer memory (%d kB requested, %d kB available)\n",
+		       mem >> 10, (unsigned int) (info->screen_size >> 10));
 		return -EINVAL;
 	}
 
 	rv = svga_check_timings (&ark_timing_regs, var, info->node);
 	if (rv < 0)
 	{
-		printk(KERN_ERR "fb%d: invalid timings requested\n", info->node);
+		fb_err(info, "invalid timings requested\n");
 		return rv;
 	}
 
@@ -693,7 +693,7 @@ static int arkfb_set_par(struct fb_info *info)
 	vga_wseq(par->state.vgabase, 0x18, regval);
 
 	/* Set the offset register */
-	pr_debug("fb%d: offset register       : %d\n", info->node, offset_value);
+	fb_dbg(info, "offset register       : %d\n", offset_value);
 	svga_wcrt_multi(par->state.vgabase, ark_offset_regs, offset_value);
 
 	/* fix for hi-res textmode */
@@ -716,7 +716,7 @@ static int arkfb_set_par(struct fb_info *info)
 	/* Set mode-specific register values */
 	switch (mode) {
 	case 0:
-		pr_debug("fb%d: text mode\n", info->node);
+		fb_dbg(info, "text mode\n");
 		svga_set_textmode_vga_regs(par->state.vgabase);
 
 		vga_wseq(par->state.vgabase, 0x11, 0x10); /* basic VGA mode */
@@ -725,7 +725,7 @@ static int arkfb_set_par(struct fb_info *info)
 
 		break;
 	case 1:
-		pr_debug("fb%d: 4 bit pseudocolor\n", info->node);
+		fb_dbg(info, "4 bit pseudocolor\n");
 		vga_wgfx(par->state.vgabase, VGA_GFX_MODE, 0x40);
 
 		vga_wseq(par->state.vgabase, 0x11, 0x10); /* basic VGA mode */
@@ -733,44 +733,44 @@ static int arkfb_set_par(struct fb_info *info)
 		dac_set_mode(par->dac, DAC_PSEUDO8_8);
 		break;
 	case 2:
-		pr_debug("fb%d: 4 bit pseudocolor, planar\n", info->node);
+		fb_dbg(info, "4 bit pseudocolor, planar\n");
 
 		vga_wseq(par->state.vgabase, 0x11, 0x10); /* basic VGA mode */
 		svga_wcrt_mask(par->state.vgabase, 0x46, 0x00, 0x04); /* 8bit pixel path */
 		dac_set_mode(par->dac, DAC_PSEUDO8_8);
 		break;
 	case 3:
-		pr_debug("fb%d: 8 bit pseudocolor\n", info->node);
+		fb_dbg(info, "8 bit pseudocolor\n");
 
 		vga_wseq(par->state.vgabase, 0x11, 0x16); /* 8bpp accel mode */
 
 		if (info->var.pixclock > 20000) {
-			pr_debug("fb%d: not using multiplex\n", info->node);
+			fb_dbg(info, "not using multiplex\n");
 			svga_wcrt_mask(par->state.vgabase, 0x46, 0x00, 0x04); /* 8bit pixel path */
 			dac_set_mode(par->dac, DAC_PSEUDO8_8);
 		} else {
-			pr_debug("fb%d: using multiplex\n", info->node);
+			fb_dbg(info, "using multiplex\n");
 			svga_wcrt_mask(par->state.vgabase, 0x46, 0x04, 0x04); /* 16bit pixel path */
 			dac_set_mode(par->dac, DAC_PSEUDO8_16);
 			hdiv = 2;
 		}
 		break;
 	case 4:
-		pr_debug("fb%d: 5/5/5 truecolor\n", info->node);
+		fb_dbg(info, "5/5/5 truecolor\n");
 
 		vga_wseq(par->state.vgabase, 0x11, 0x1A); /* 16bpp accel mode */
 		svga_wcrt_mask(par->state.vgabase, 0x46, 0x04, 0x04); /* 16bit pixel path */
 		dac_set_mode(par->dac, DAC_RGB1555_16);
 		break;
 	case 5:
-		pr_debug("fb%d: 5/6/5 truecolor\n", info->node);
+		fb_dbg(info, "5/6/5 truecolor\n");
 
 		vga_wseq(par->state.vgabase, 0x11, 0x1A); /* 16bpp accel mode */
 		svga_wcrt_mask(par->state.vgabase, 0x46, 0x04, 0x04); /* 16bit pixel path */
 		dac_set_mode(par->dac, DAC_RGB0565_16);
 		break;
 	case 6:
-		pr_debug("fb%d: 8/8/8 truecolor\n", info->node);
+		fb_dbg(info, "8/8/8 truecolor\n");
 
 		vga_wseq(par->state.vgabase, 0x11, 0x16); /* 8bpp accel mode ??? */
 		svga_wcrt_mask(par->state.vgabase, 0x46, 0x04, 0x04); /* 16bit pixel path */
@@ -779,7 +779,7 @@ static int arkfb_set_par(struct fb_info *info)
 		hdiv = 2;
 		break;
 	case 7:
-		pr_debug("fb%d: 8/8/8/8 truecolor\n", info->node);
+		fb_dbg(info, "8/8/8/8 truecolor\n");
 
 		vga_wseq(par->state.vgabase, 0x11, 0x1E); /* 32bpp accel mode */
 		svga_wcrt_mask(par->state.vgabase, 0x46, 0x04, 0x04); /* 16bit pixel path */
@@ -787,7 +787,7 @@ static int arkfb_set_par(struct fb_info *info)
 		hmul = 2;
 		break;
 	default:
-		printk(KERN_ERR "fb%d: unsupported mode - bug\n", info->node);
+		fb_err(info, "unsupported mode - bug\n");
 		return -EINVAL;
 	}
 
@@ -879,19 +879,19 @@ static int arkfb_blank(int blank_mode, struct fb_info *info)
 
 	switch (blank_mode) {
 	case FB_BLANK_UNBLANK:
-		pr_debug("fb%d: unblank\n", info->node);
+		fb_dbg(info, "unblank\n");
 		svga_wseq_mask(par->state.vgabase, 0x01, 0x00, 0x20);
 		svga_wcrt_mask(par->state.vgabase, 0x17, 0x80, 0x80);
 		break;
 	case FB_BLANK_NORMAL:
-		pr_debug("fb%d: blank\n", info->node);
+		fb_dbg(info, "blank\n");
 		svga_wseq_mask(par->state.vgabase, 0x01, 0x20, 0x20);
 		svga_wcrt_mask(par->state.vgabase, 0x17, 0x80, 0x80);
 		break;
 	case FB_BLANK_POWERDOWN:
 	case FB_BLANK_HSYNC_SUSPEND:
 	case FB_BLANK_VSYNC_SUSPEND:
-		pr_debug("fb%d: sync down\n", info->node);
+		fb_dbg(info, "sync down\n");
 		svga_wseq_mask(par->state.vgabase, 0x01, 0x20, 0x20);
 		svga_wcrt_mask(par->state.vgabase, 0x17, 0x00, 0x80);
 		break;
@@ -1048,12 +1048,12 @@ static int ark_pci_probe(struct pci_dev *dev, const struct pci_device_id *id)
 
 	rc = register_framebuffer(info);
 	if (rc < 0) {
-		dev_err(info->device, "cannot register framebugger\n");
+		dev_err(info->device, "cannot register framebuffer\n");
 		goto err_reg_fb;
 	}
 
-	printk(KERN_INFO "fb%d: %s on %s, %d MB RAM\n", info->node, info->fix.id,
-		 pci_name(dev), info->fix.smem_len >> 20);
+	fb_info(info, "%s on %s, %d MB RAM\n",
+		info->fix.id, pci_name(dev), info->fix.smem_len >> 20);
 
 	/* Record a reference to the driver data */
 	pci_set_drvdata(dev, info);
@@ -1108,7 +1108,6 @@ static void ark_pci_remove(struct pci_dev *dev)
 		pci_release_regions(dev);
 /*		pci_disable_device(dev); */
 
-		pci_set_drvdata(dev, NULL);
 		framebuffer_release(info);
 	}
 }
diff --git a/drivers/video/asiliantfb.c b/drivers/video/asiliantfb.c
index d5a37d62847b..d611f1a1ac53 100644
--- a/drivers/video/asiliantfb.c
+++ b/drivers/video/asiliantfb.c
@@ -527,8 +527,8 @@ static int init_asiliant(struct fb_info *p, unsigned long addr)
 		return err;
 	}
 
-	printk(KERN_INFO "fb%d: Asiliant 69000 frame buffer (%dK RAM detected)\n",
-		p->node, p->fix.smem_len / 1024);
+	fb_info(p, "Asiliant 69000 frame buffer (%dK RAM detected)\n",
+		p->fix.smem_len / 1024);
 
 	writeb(0xff, mmio_base + 0x78c);
 	chips_hw_init(p);
diff --git a/drivers/video/atafb.c b/drivers/video/atafb.c
index 64e41f5448c4..e21d1f58554c 100644
--- a/drivers/video/atafb.c
+++ b/drivers/video/atafb.c
@@ -3246,11 +3246,8 @@ int __init atafb_init(void)
 		return -EINVAL;
 	}
 
-	// FIXME: mode needs setting!
-	//printk("fb%d: %s frame buffer device, using %dK of video memory\n",
-	//       fb_info.node, fb_info.mode->name, screen_len>>10);
-	printk("fb%d: frame buffer device, using %dK of video memory\n",
-	       fb_info.node, screen_len >> 10);
+	fb_info(&fb_info, "frame buffer device, using %dK of video memory\n",
+		screen_len >> 10);
 
 	/* TODO: This driver cannot be unloaded yet */
 	return 0;
diff --git a/drivers/video/atmel_lcdfb.c b/drivers/video/atmel_lcdfb.c
index 088511a58a26..8521051cf946 100644
--- a/drivers/video/atmel_lcdfb.c
+++ b/drivers/video/atmel_lcdfb.c
@@ -20,12 +20,55 @@
 #include <linux/gfp.h>
 #include <linux/module.h>
 #include <linux/platform_data/atmel.h>
+#include <linux/of.h>
+#include <linux/of_device.h>
+#include <linux/of_gpio.h>
+#include <video/of_display_timing.h>
+#include <video/videomode.h>
 
 #include <mach/cpu.h>
 #include <asm/gpio.h>
 
 #include <video/atmel_lcdc.h>
 
+struct atmel_lcdfb_config {
+	bool have_alt_pixclock;
+	bool have_hozval;
+	bool have_intensity_bit;
+};
+
+ /* LCD Controller info data structure, stored in device platform_data */
+struct atmel_lcdfb_info {
+	spinlock_t		lock;
+	struct fb_info		*info;
+	void __iomem		*mmio;
+	int			irq_base;
+	struct work_struct	task;
+
+	unsigned int		smem_len;
+	struct platform_device	*pdev;
+	struct clk		*bus_clk;
+	struct clk		*lcdc_clk;
+
+	struct backlight_device	*backlight;
+	u8			bl_power;
+	u8			saved_lcdcon;
+
+	u32			pseudo_palette[16];
+	bool			have_intensity_bit;
+
+	struct atmel_lcdfb_pdata pdata;
+
+	struct atmel_lcdfb_config *config;
+};
+
+struct atmel_lcdfb_power_ctrl_gpio {
+	int gpio;
+	int active_low;
+
+	struct list_head list;
+};
+
 #define lcdc_readl(sinfo, reg)		__raw_readl((sinfo)->mmio+(reg))
 #define lcdc_writel(sinfo, reg, val)	__raw_writel((val), (sinfo)->mmio+(reg))
 
@@ -34,12 +77,6 @@
 #define ATMEL_LCDC_DMA_BURST_LEN	8	/* words */
 #define ATMEL_LCDC_FIFO_SIZE		512	/* words */
 
-struct atmel_lcdfb_config {
-	bool have_alt_pixclock;
-	bool have_hozval;
-	bool have_intensity_bit;
-};
-
 static struct atmel_lcdfb_config at91sam9261_config = {
 	.have_hozval		= true,
 	.have_intensity_bit	= true,
@@ -248,18 +285,27 @@ static void exit_backlight(struct atmel_lcdfb_info *sinfo)
 
 static void init_contrast(struct atmel_lcdfb_info *sinfo)
 {
+	struct atmel_lcdfb_pdata *pdata = &sinfo->pdata;
+
 	/* contrast pwm can be 'inverted' */
-	if (sinfo->lcdcon_pol_negative)
+	if (pdata->lcdcon_pol_negative)
 			contrast_ctr &= ~(ATMEL_LCDC_POL_POSITIVE);
 
 	/* have some default contrast/backlight settings */
 	lcdc_writel(sinfo, ATMEL_LCDC_CONTRAST_CTR, contrast_ctr);
 	lcdc_writel(sinfo, ATMEL_LCDC_CONTRAST_VAL, ATMEL_LCDC_CVAL_DEFAULT);
 
-	if (sinfo->lcdcon_is_backlight)
+	if (pdata->lcdcon_is_backlight)
 		init_backlight(sinfo);
 }
 
+static inline void atmel_lcdfb_power_control(struct atmel_lcdfb_info *sinfo, int on)
+{
+	struct atmel_lcdfb_pdata *pdata = &sinfo->pdata;
+
+	if (pdata->atmel_lcdfb_power_control)
+		pdata->atmel_lcdfb_power_control(pdata, on);
+}
 
 static struct fb_fix_screeninfo atmel_lcdfb_fix __initdata = {
 	.type		= FB_TYPE_PACKED_PIXELS,
@@ -299,9 +345,11 @@ static unsigned long compute_hozval(struct atmel_lcdfb_info *sinfo,
 
 static void atmel_lcdfb_stop_nowait(struct atmel_lcdfb_info *sinfo)
 {
+	struct atmel_lcdfb_pdata *pdata = &sinfo->pdata;
+
 	/* Turn off the LCD controller and the DMA controller */
 	lcdc_writel(sinfo, ATMEL_LCDC_PWRCON,
-			sinfo->guard_time << ATMEL_LCDC_GUARDT_OFFSET);
+			pdata->guard_time << ATMEL_LCDC_GUARDT_OFFSET);
 
 	/* Wait for the LCDC core to become idle */
 	while (lcdc_readl(sinfo, ATMEL_LCDC_PWRCON) & ATMEL_LCDC_BUSY)
@@ -321,9 +369,11 @@ static void atmel_lcdfb_stop(struct atmel_lcdfb_info *sinfo)
 
 static void atmel_lcdfb_start(struct atmel_lcdfb_info *sinfo)
 {
-	lcdc_writel(sinfo, ATMEL_LCDC_DMACON, sinfo->default_dmacon);
+	struct atmel_lcdfb_pdata *pdata = &sinfo->pdata;
+
+	lcdc_writel(sinfo, ATMEL_LCDC_DMACON, pdata->default_dmacon);
 	lcdc_writel(sinfo, ATMEL_LCDC_PWRCON,
-		(sinfo->guard_time << ATMEL_LCDC_GUARDT_OFFSET)
+		(pdata->guard_time << ATMEL_LCDC_GUARDT_OFFSET)
 		| ATMEL_LCDC_PWR);
 }
 
@@ -424,6 +474,7 @@ static int atmel_lcdfb_check_var(struct fb_var_screeninfo *var,
 {
 	struct device *dev = info->device;
 	struct atmel_lcdfb_info *sinfo = info->par;
+	struct atmel_lcdfb_pdata *pdata = &sinfo->pdata;
 	unsigned long clk_value_khz;
 
 	clk_value_khz = clk_get_rate(sinfo->lcdc_clk) / 1000;
@@ -510,7 +561,7 @@ static int atmel_lcdfb_check_var(struct fb_var_screeninfo *var,
 		else
 			var->green.length = 6;
 
-		if (sinfo->lcd_wiring_mode == ATMEL_LCDC_WIRING_RGB) {
+		if (pdata->lcd_wiring_mode == ATMEL_LCDC_WIRING_RGB) {
 			/* RGB:5X5 mode */
 			var->red.offset = var->green.length + 5;
 			var->blue.offset = 0;
@@ -527,7 +578,7 @@ static int atmel_lcdfb_check_var(struct fb_var_screeninfo *var,
 		var->transp.length = 8;
 		/* fall through */
 	case 24:
-		if (sinfo->lcd_wiring_mode == ATMEL_LCDC_WIRING_RGB) {
+		if (pdata->lcd_wiring_mode == ATMEL_LCDC_WIRING_RGB) {
 			/* RGB:888 mode */
 			var->red.offset = 16;
 			var->blue.offset = 0;
@@ -576,6 +627,7 @@ static void atmel_lcdfb_reset(struct atmel_lcdfb_info *sinfo)
 static int atmel_lcdfb_set_par(struct fb_info *info)
 {
 	struct atmel_lcdfb_info *sinfo = info->par;
+	struct atmel_lcdfb_pdata *pdata = &sinfo->pdata;
 	unsigned long hozval_linesz;
 	unsigned long value;
 	unsigned long clk_value_khz;
@@ -637,7 +689,7 @@ static int atmel_lcdfb_set_par(struct fb_info *info)
 
 
 	/* Initialize control register 2 */
-	value = sinfo->default_lcdcon2;
+	value = pdata->default_lcdcon2;
 
 	if (!(info->var.sync & FB_SYNC_HOR_HIGH_ACT))
 		value |= ATMEL_LCDC_INVLINE_INVERTED;
@@ -741,6 +793,7 @@ static int atmel_lcdfb_setcolreg(unsigned int regno, unsigned int red,
 			     unsigned int transp, struct fb_info *info)
 {
 	struct atmel_lcdfb_info *sinfo = info->par;
+	struct atmel_lcdfb_pdata *pdata = &sinfo->pdata;
 	unsigned int val;
 	u32 *pal;
 	int ret = 1;
@@ -777,8 +830,7 @@ static int atmel_lcdfb_setcolreg(unsigned int regno, unsigned int red,
 				 */
 			} else {
 				/* new style BGR:565 / RGB:565 */
-				if (sinfo->lcd_wiring_mode ==
-				    ATMEL_LCDC_WIRING_RGB) {
+				if (pdata->lcd_wiring_mode == ATMEL_LCDC_WIRING_RGB) {
 					val  = ((blue >> 11) & 0x001f);
 					val |= ((red  >>  0) & 0xf800);
 				} else {
@@ -912,16 +964,187 @@ static void atmel_lcdfb_stop_clock(struct atmel_lcdfb_info *sinfo)
 	clk_disable_unprepare(sinfo->lcdc_clk);
 }
 
+#ifdef CONFIG_OF
+static const struct of_device_id atmel_lcdfb_dt_ids[] = {
+	{ .compatible = "atmel,at91sam9261-lcdc" , .data = &at91sam9261_config, },
+	{ .compatible = "atmel,at91sam9263-lcdc" , .data = &at91sam9263_config, },
+	{ .compatible = "atmel,at91sam9g10-lcdc" , .data = &at91sam9g10_config, },
+	{ .compatible = "atmel,at91sam9g45-lcdc" , .data = &at91sam9g45_config, },
+	{ .compatible = "atmel,at91sam9g45es-lcdc" , .data = &at91sam9g45es_config, },
+	{ .compatible = "atmel,at91sam9rl-lcdc" , .data = &at91sam9rl_config, },
+	{ .compatible = "atmel,at32ap-lcdc" , .data = &at32ap_config, },
+	{ /* sentinel */ }
+};
+
+MODULE_DEVICE_TABLE(of, atmel_lcdfb_dt_ids);
+
+static const char *atmel_lcdfb_wiring_modes[] = {
+	[ATMEL_LCDC_WIRING_BGR]	= "BRG",
+	[ATMEL_LCDC_WIRING_RGB]	= "RGB",
+};
+
+const int atmel_lcdfb_get_of_wiring_modes(struct device_node *np)
+{
+	const char *mode;
+	int err, i;
+
+	err = of_property_read_string(np, "atmel,lcd-wiring-mode", &mode);
+	if (err < 0)
+		return ATMEL_LCDC_WIRING_BGR;
+
+	for (i = 0; i < ARRAY_SIZE(atmel_lcdfb_wiring_modes); i++)
+		if (!strcasecmp(mode, atmel_lcdfb_wiring_modes[i]))
+			return i;
+
+	return -ENODEV;
+}
+
+static void atmel_lcdfb_power_control_gpio(struct atmel_lcdfb_pdata *pdata, int on)
+{
+	struct atmel_lcdfb_power_ctrl_gpio *og;
+
+	list_for_each_entry(og, &pdata->pwr_gpios, list)
+		gpio_set_value(og->gpio, on);
+}
+
+static int atmel_lcdfb_of_init(struct atmel_lcdfb_info *sinfo)
+{
+	struct fb_info *info = sinfo->info;
+	struct atmel_lcdfb_pdata *pdata = &sinfo->pdata;
+	struct fb_var_screeninfo *var = &info->var;
+	struct device *dev = &sinfo->pdev->dev;
+	struct device_node *np =dev->of_node;
+	struct device_node *display_np;
+	struct device_node *timings_np;
+	struct display_timings *timings;
+	enum of_gpio_flags flags;
+	struct atmel_lcdfb_power_ctrl_gpio *og;
+	bool is_gpio_power = false;
+	int ret = -ENOENT;
+	int i, gpio;
+
+	sinfo->config = (struct atmel_lcdfb_config*)
+		of_match_device(atmel_lcdfb_dt_ids, dev)->data;
+
+	display_np = of_parse_phandle(np, "display", 0);
+	if (!display_np) {
+		dev_err(dev, "failed to find display phandle\n");
+		return -ENOENT;
+	}
+
+	ret = of_property_read_u32(display_np, "bits-per-pixel", &var->bits_per_pixel);
+	if (ret < 0) {
+		dev_err(dev, "failed to get property bits-per-pixel\n");
+		goto put_display_node;
+	}
+
+	ret = of_property_read_u32(display_np, "atmel,guard-time", &pdata->guard_time);
+	if (ret < 0) {
+		dev_err(dev, "failed to get property atmel,guard-time\n");
+		goto put_display_node;
+	}
+
+	ret = of_property_read_u32(display_np, "atmel,lcdcon2", &pdata->default_lcdcon2);
+	if (ret < 0) {
+		dev_err(dev, "failed to get property atmel,lcdcon2\n");
+		goto put_display_node;
+	}
+
+	ret = of_property_read_u32(display_np, "atmel,dmacon", &pdata->default_dmacon);
+	if (ret < 0) {
+		dev_err(dev, "failed to get property bits-per-pixel\n");
+		goto put_display_node;
+	}
+
+	ret = -ENOMEM;
+	for (i = 0; i < of_gpio_named_count(display_np, "atmel,power-control-gpio"); i++) {
+		gpio = of_get_named_gpio_flags(display_np, "atmel,power-control-gpio",
+					       i, &flags);
+		if (gpio < 0)
+			continue;
+
+		og = devm_kzalloc(dev, sizeof(*og), GFP_KERNEL);
+		if (!og)
+			goto put_display_node;
+
+		og->gpio = gpio;
+		og->active_low = flags & OF_GPIO_ACTIVE_LOW;
+		is_gpio_power = true;
+		ret = devm_gpio_request(dev, gpio, "lcd-power-control-gpio");
+		if (ret) {
+			dev_err(dev, "request gpio %d failed\n", gpio);
+			goto put_display_node;
+		}
+
+		ret = gpio_direction_output(gpio, og->active_low);
+		if (ret) {
+			dev_err(dev, "set direction output gpio %d failed\n", gpio);
+			goto put_display_node;
+		}
+	}
+
+	if (is_gpio_power)
+		pdata->atmel_lcdfb_power_control = atmel_lcdfb_power_control_gpio;
+
+	ret = atmel_lcdfb_get_of_wiring_modes(display_np);
+	if (ret < 0) {
+		dev_err(dev, "invalid atmel,lcd-wiring-mode\n");
+		goto put_display_node;
+	}
+	pdata->lcd_wiring_mode = ret;
+
+	pdata->lcdcon_is_backlight = of_property_read_bool(display_np, "atmel,lcdcon-backlight");
+
+	timings = of_get_display_timings(display_np);
+	if (!timings) {
+		dev_err(dev, "failed to get display timings\n");
+		goto put_display_node;
+	}
+
+	timings_np = of_find_node_by_name(display_np, "display-timings");
+	if (!timings_np) {
+		dev_err(dev, "failed to find display-timings node\n");
+		goto put_display_node;
+	}
+
+	for (i = 0; i < of_get_child_count(timings_np); i++) {
+		struct videomode vm;
+		struct fb_videomode fb_vm;
+
+		ret = videomode_from_timings(timings, &vm, i);
+		if (ret < 0)
+			goto put_timings_node;
+		ret = fb_videomode_from_videomode(&vm, &fb_vm);
+		if (ret < 0)
+			goto put_timings_node;
+
+		fb_add_videomode(&fb_vm, &info->modelist);
+	}
+
+	return 0;
+
+put_timings_node:
+	of_node_put(timings_np);
+put_display_node:
+	of_node_put(display_np);
+	return ret;
+}
+#else
+static int atmel_lcdfb_of_init(struct atmel_lcdfb_info *sinfo)
+{
+	return 0;
+}
+#endif
 
 static int __init atmel_lcdfb_probe(struct platform_device *pdev)
 {
 	struct device *dev = &pdev->dev;
 	struct fb_info *info;
 	struct atmel_lcdfb_info *sinfo;
-	struct atmel_lcdfb_info *pdata_sinfo;
-	struct fb_videomode fbmode;
+	struct atmel_lcdfb_pdata *pdata = NULL;
 	struct resource *regs = NULL;
 	struct resource *map = NULL;
+	struct fb_modelist *modelist;
 	int ret;
 
 	dev_dbg(dev, "%s BEGIN\n", __func__);
@@ -934,26 +1157,35 @@ static int __init atmel_lcdfb_probe(struct platform_device *pdev)
 	}
 
 	sinfo = info->par;
+	sinfo->pdev = pdev;
+	sinfo->info = info;
+
+	INIT_LIST_HEAD(&info->modelist);
 
-	if (dev->platform_data) {
-		pdata_sinfo = (struct atmel_lcdfb_info *)dev->platform_data;
-		sinfo->default_bpp = pdata_sinfo->default_bpp;
-		sinfo->default_dmacon = pdata_sinfo->default_dmacon;
-		sinfo->default_lcdcon2 = pdata_sinfo->default_lcdcon2;
-		sinfo->default_monspecs = pdata_sinfo->default_monspecs;
-		sinfo->atmel_lcdfb_power_control = pdata_sinfo->atmel_lcdfb_power_control;
-		sinfo->guard_time = pdata_sinfo->guard_time;
-		sinfo->smem_len = pdata_sinfo->smem_len;
-		sinfo->lcdcon_is_backlight = pdata_sinfo->lcdcon_is_backlight;
-		sinfo->lcdcon_pol_negative = pdata_sinfo->lcdcon_pol_negative;
-		sinfo->lcd_wiring_mode = pdata_sinfo->lcd_wiring_mode;
+	if (pdev->dev.of_node) {
+		ret = atmel_lcdfb_of_init(sinfo);
+		if (ret)
+			goto free_info;
+	} else if (dev_get_platdata(dev)) {
+		struct fb_monspecs *monspecs;
+		int i;
+
+		pdata = dev_get_platdata(dev);
+		monspecs = pdata->default_monspecs;
+		sinfo->pdata = *pdata;
+
+		for (i = 0; i < monspecs->modedb_len; i++)
+			fb_add_videomode(&monspecs->modedb[i], &info->modelist);
+
+		sinfo->config = atmel_lcdfb_get_config(pdev);
+
+		info->var.bits_per_pixel = pdata->default_bpp ? pdata->default_bpp : 16;
+		memcpy(&info->monspecs, pdata->default_monspecs, sizeof(info->monspecs));
 	} else {
 		dev_err(dev, "cannot get default configuration\n");
 		goto free_info;
 	}
-	sinfo->info = info;
-	sinfo->pdev = pdev;
-	sinfo->config = atmel_lcdfb_get_config(pdev);
+
 	if (!sinfo->config)
 		goto free_info;
 
@@ -962,7 +1194,6 @@ static int __init atmel_lcdfb_probe(struct platform_device *pdev)
 	info->pseudo_palette = sinfo->pseudo_palette;
 	info->fbops = &atmel_lcdfb_ops;
 
-	memcpy(&info->monspecs, sinfo->default_monspecs, sizeof(info->monspecs));
 	info->fix = atmel_lcdfb_fix;
 
 	/* Enable LCDC Clocks */
@@ -978,14 +1209,11 @@ static int __init atmel_lcdfb_probe(struct platform_device *pdev)
 	}
 	atmel_lcdfb_start_clock(sinfo);
 
-	ret = fb_find_mode(&info->var, info, NULL, info->monspecs.modedb,
-			info->monspecs.modedb_len, info->monspecs.modedb,
-			sinfo->default_bpp);
-	if (!ret) {
-		dev_err(dev, "no suitable video mode found\n");
-		goto stop_clk;
-	}
+	modelist = list_first_entry(&info->modelist,
+			struct fb_modelist, list);
+	fb_videomode_to_var(&info->var, &modelist->mode);
 
+	atmel_lcdfb_check_var(&info->var, info);
 
 	regs = platform_get_resource(pdev, IORESOURCE_MEM, 0);
 	if (!regs) {
@@ -1069,18 +1297,6 @@ static int __init atmel_lcdfb_probe(struct platform_device *pdev)
 		goto unregister_irqs;
 	}
 
-	/*
-	 * This makes sure that our colour bitfield
-	 * descriptors are correctly initialised.
-	 */
-	atmel_lcdfb_check_var(&info->var, info);
-
-	ret = fb_set_var(info, &info->var);
-	if (ret) {
-		dev_warn(dev, "unable to set display parameters\n");
-		goto free_cmap;
-	}
-
 	dev_set_drvdata(dev, info);
 
 	/*
@@ -1092,13 +1308,8 @@ static int __init atmel_lcdfb_probe(struct platform_device *pdev)
 		goto reset_drvdata;
 	}
 
-	/* add selected videomode to modelist */
-	fb_var_to_videomode(&fbmode, &info->var);
-	fb_add_videomode(&fbmode, &info->modelist);
-
 	/* Power up the LCDC screen */
-	if (sinfo->atmel_lcdfb_power_control)
-		sinfo->atmel_lcdfb_power_control(1);
+	atmel_lcdfb_power_control(sinfo, 1);
 
 	dev_info(dev, "fb%d: Atmel LCDC at 0x%08lx (mapped at %p), irq %d\n",
 		       info->node, info->fix.mmio_start, sinfo->mmio, sinfo->irq_base);
@@ -1107,7 +1318,6 @@ static int __init atmel_lcdfb_probe(struct platform_device *pdev)
 
 reset_drvdata:
 	dev_set_drvdata(dev, NULL);
-free_cmap:
 	fb_dealloc_cmap(&info->cmap);
 unregister_irqs:
 	cancel_work_sync(&sinfo->task);
@@ -1143,15 +1353,16 @@ static int __exit atmel_lcdfb_remove(struct platform_device *pdev)
 	struct device *dev = &pdev->dev;
 	struct fb_info *info = dev_get_drvdata(dev);
 	struct atmel_lcdfb_info *sinfo;
+	struct atmel_lcdfb_pdata *pdata;
 
 	if (!info || !info->par)
 		return 0;
 	sinfo = info->par;
+	pdata = &sinfo->pdata;
 
 	cancel_work_sync(&sinfo->task);
 	exit_backlight(sinfo);
-	if (sinfo->atmel_lcdfb_power_control)
-		sinfo->atmel_lcdfb_power_control(0);
+	atmel_lcdfb_power_control(sinfo, 0);
 	unregister_framebuffer(info);
 	atmel_lcdfb_stop_clock(sinfo);
 	clk_put(sinfo->lcdc_clk);
@@ -1167,7 +1378,6 @@ static int __exit atmel_lcdfb_remove(struct platform_device *pdev)
 		atmel_lcdfb_free_video_memory(sinfo);
 	}
 
-	dev_set_drvdata(dev, NULL);
 	framebuffer_release(info);
 
 	return 0;
@@ -1188,9 +1398,7 @@ static int atmel_lcdfb_suspend(struct platform_device *pdev, pm_message_t mesg)
 
 	sinfo->saved_lcdcon = lcdc_readl(sinfo, ATMEL_LCDC_CONTRAST_CTR);
 	lcdc_writel(sinfo, ATMEL_LCDC_CONTRAST_CTR, 0);
-	if (sinfo->atmel_lcdfb_power_control)
-		sinfo->atmel_lcdfb_power_control(0);
-
+	atmel_lcdfb_power_control(sinfo, 0);
 	atmel_lcdfb_stop(sinfo);
 	atmel_lcdfb_stop_clock(sinfo);
 
@@ -1204,8 +1412,7 @@ static int atmel_lcdfb_resume(struct platform_device *pdev)
 
 	atmel_lcdfb_start_clock(sinfo);
 	atmel_lcdfb_start(sinfo);
-	if (sinfo->atmel_lcdfb_power_control)
-		sinfo->atmel_lcdfb_power_control(1);
+	atmel_lcdfb_power_control(sinfo, 1);
 	lcdc_writel(sinfo, ATMEL_LCDC_CONTRAST_CTR, sinfo->saved_lcdcon);
 
 	/* Enable FIFO & DMA errors */
@@ -1228,6 +1435,7 @@ static struct platform_driver atmel_lcdfb_driver = {
 	.driver		= {
 		.name	= "atmel_lcdfb",
 		.owner	= THIS_MODULE,
+		.of_match_table	= of_match_ptr(atmel_lcdfb_dt_ids),
 	},
 };
 
diff --git a/drivers/video/aty/aty128fb.c b/drivers/video/aty/aty128fb.c
index a4dfe8cb0a0a..12ca031877d4 100644
--- a/drivers/video/aty/aty128fb.c
+++ b/drivers/video/aty/aty128fb.c
@@ -413,7 +413,6 @@ struct aty128fb_par {
 	int blitter_may_be_busy;
 	int fifo_slots;                 /* free slots in FIFO (64 max) */
 
-	int	pm_reg;
 	int crt_on, lcd_on;
 	struct pci_dev *pdev;
 	struct fb_info *next;
@@ -2016,7 +2015,6 @@ static int aty128_init(struct pci_dev *pdev, const struct pci_device_id *ent)
 
 	aty128_init_engine(par);
 
-	par->pm_reg = pdev->pm_cap;
 	par->pdev = pdev;
 	par->asleep = 0;
 	par->lock_blank = 0;
@@ -2029,8 +2027,8 @@ static int aty128_init(struct pci_dev *pdev, const struct pci_device_id *ent)
 	if (register_framebuffer(info) < 0)
 		return 0;
 
-	printk(KERN_INFO "fb%d: %s frame buffer device on %s\n",
-	       info->node, info->fix.id, video_card);
+	fb_info(info, "%s frame buffer device on %s\n",
+		info->fix.id, video_card);
 
 	return 1;	/* success! */
 }
@@ -2397,7 +2395,7 @@ static void aty128_set_suspend(struct aty128fb_par *par, int suspend)
 	u32	pmgt;
 	struct pci_dev *pdev = par->pdev;
 
-	if (!par->pm_reg)
+	if (!par->pdev->pm_cap)
 		return;
 		
 	/* Set the chip into the appropriate suspend mode (we use D2,
diff --git a/drivers/video/aty/atyfb_base.c b/drivers/video/aty/atyfb_base.c
index 9b0f12c5c284..28fafbf864a5 100644
--- a/drivers/video/aty/atyfb_base.c
+++ b/drivers/video/aty/atyfb_base.c
@@ -1848,7 +1848,6 @@ static int atyfb_ioctl(struct fb_info *info, u_int cmd, u_long arg)
 
 			return aty_waitforvblank(par, crtc);
 		}
-		break;
 
 #if defined(DEBUG) && defined(CONFIG_FB_ATY_CT)
 	case ATYIO_CLKR:
diff --git a/drivers/video/aty/radeon_base.c b/drivers/video/aty/radeon_base.c
index 1e30b2b3e79f..26d80a4486fb 100644
--- a/drivers/video/aty/radeon_base.c
+++ b/drivers/video/aty/radeon_base.c
@@ -819,11 +819,6 @@ static int radeonfb_check_var (struct fb_var_screeninfo *var, struct fb_info *in
 	if (v.xres_virtual < v.xres)
 		v.xres = v.xres_virtual;
 
-	if (v.xoffset < 0)
-                v.xoffset = 0;
-        if (v.yoffset < 0)
-                v.yoffset = 0;
-         
         if (v.xoffset > v.xres_virtual - v.xres)
                 v.xoffset = v.xres_virtual - v.xres - 1;
                         
diff --git a/drivers/video/aty/radeon_pm.c b/drivers/video/aty/radeon_pm.c
index f7091ece580d..46a12f1a93c3 100644
--- a/drivers/video/aty/radeon_pm.c
+++ b/drivers/video/aty/radeon_pm.c
@@ -1427,6 +1427,8 @@ static void radeon_pm_full_reset_sdram(struct radeonfb_info *rinfo)
 	mdelay( 15);
 }
 
+#if defined(CONFIG_PM)
+#if defined(CONFIG_X86) || defined(CONFIG_PPC_PMAC)
 static void radeon_pm_reset_pad_ctlr_strength(struct radeonfb_info *rinfo)
 {
 	u32 tmp, tmp2;
@@ -1939,9 +1941,10 @@ static void radeon_reinitialize_M10(struct radeonfb_info *rinfo)
 	 */
 	radeon_pm_m10_enable_lvds_spread_spectrum(rinfo);
 }
+#endif
 
 #ifdef CONFIG_PPC_OF
-
+#ifdef CONFIG_PPC_PMAC
 static void radeon_pm_m9p_reconfigure_mc(struct radeonfb_info *rinfo)
 {
 	OUTREG(MC_CNTL, rinfo->save_regs[46]);
@@ -2202,6 +2205,8 @@ static void radeon_reinitialize_M9P(struct radeonfb_info *rinfo)
 	radeon_pm_restore_pixel_pll(rinfo);
 	radeon_pm_m10_enable_lvds_spread_spectrum(rinfo);
 }
+#endif
+#endif
 
 #if 0 /* Not ready yet */
 static void radeon_reinitialize_QW(struct radeonfb_info *rinfo)
@@ -2515,13 +2520,13 @@ static void radeonfb_whack_power_state(struct radeonfb_info *rinfo, pci_power_t
 
 	for (;;) {
 		pci_read_config_word(rinfo->pdev,
-				     rinfo->pm_reg+PCI_PM_CTRL,
+				     rinfo->pdev->pm_cap + PCI_PM_CTRL,
 				     &pwr_cmd);
-		if (pwr_cmd & 2)
+		if (pwr_cmd & state)
 			break;
-		pwr_cmd = (pwr_cmd & ~PCI_PM_CTRL_STATE_MASK) | 2;
+		pwr_cmd = (pwr_cmd & ~PCI_PM_CTRL_STATE_MASK) | state;
 		pci_write_config_word(rinfo->pdev,
-				      rinfo->pm_reg+PCI_PM_CTRL,
+				      rinfo->pdev->pm_cap + PCI_PM_CTRL,
 				      pwr_cmd);
 		msleep(500);
 	}
@@ -2532,7 +2537,7 @@ static void radeon_set_suspend(struct radeonfb_info *rinfo, int suspend)
 {
 	u32 tmp;
 
-	if (!rinfo->pm_reg)
+	if (!rinfo->pdev->pm_cap)
 		return;
 
 	/* Set the chip into appropriate suspend mode (we use D2,
@@ -2804,9 +2809,6 @@ static void radeonfb_early_resume(void *data)
 
 void radeonfb_pm_init(struct radeonfb_info *rinfo, int dynclk, int ignore_devlist, int force_sleep)
 {
-	/* Find PM registers in config space if any*/
-	rinfo->pm_reg = rinfo->pdev->pm_cap;
-
 	/* Enable/Disable dynamic clocks: TODO add sysfs access */
 	if (rinfo->family == CHIP_FAMILY_RS480)
 		rinfo->dynclk = -1;
@@ -2830,7 +2832,7 @@ void radeonfb_pm_init(struct radeonfb_info *rinfo, int dynclk, int ignore_devlis
 	 * reason. --BenH
 	 */
 	if (machine_is(powermac) && rinfo->of_node) {
-		if (rinfo->is_mobility && rinfo->pm_reg &&
+		if (rinfo->is_mobility && rinfo->pdev->pm_cap &&
 		    rinfo->family <= CHIP_FAMILY_RV250)
 			rinfo->pm_mode |= radeon_pm_d2;
 
diff --git a/drivers/video/aty/radeonfb.h b/drivers/video/aty/radeonfb.h
index 7351e66c7f54..cb846044f57c 100644
--- a/drivers/video/aty/radeonfb.h
+++ b/drivers/video/aty/radeonfb.h
@@ -342,7 +342,6 @@ struct radeonfb_info {
 
 	int			mtrr_hdl;
 
-	int			pm_reg;
 	u32			save_regs[100];
 	int			asleep;
 	int			lock_blank;
diff --git a/drivers/video/au1100fb.c b/drivers/video/au1100fb.c
index 22ad85242e5b..372d4aea9d1c 100644
--- a/drivers/video/au1100fb.c
+++ b/drivers/video/au1100fb.c
@@ -564,7 +564,7 @@ int au1100fb_drv_remove(struct platform_device *dev)
 	if (!dev)
 		return -ENODEV;
 
-	fbdev = (struct au1100fb_device *) platform_get_drvdata(dev);
+	fbdev = platform_get_drvdata(dev);
 
 #if !defined(CONFIG_FRAMEBUFFER_CONSOLE) && defined(CONFIG_LOGO)
 	au1100fb_fb_blank(VESA_POWERDOWN, &fbdev->info);
@@ -636,19 +636,7 @@ static struct platform_driver au1100fb_driver = {
 	.suspend	= au1100fb_drv_suspend,
         .resume		= au1100fb_drv_resume,
 };
-
-static int __init au1100fb_load(void)
-{
-	return platform_driver_register(&au1100fb_driver);
-}
-
-static void __exit au1100fb_unload(void)
-{
-	platform_driver_unregister(&au1100fb_driver);
-}
-
-module_init(au1100fb_load);
-module_exit(au1100fb_unload);
+module_platform_driver(au1100fb_driver);
 
 MODULE_DESCRIPTION(DRIVER_DESC);
 MODULE_LICENSE("GPL");
diff --git a/drivers/video/au1200fb.c b/drivers/video/au1200fb.c
index 1d02897d17f2..4cfba78a1458 100644
--- a/drivers/video/au1200fb.c
+++ b/drivers/video/au1200fb.c
@@ -1853,21 +1853,7 @@ static struct platform_driver au1200fb_driver = {
 	.probe		= au1200fb_drv_probe,
 	.remove		= au1200fb_drv_remove,
 };
-
-/*-------------------------------------------------------------------------*/
-
-static int __init au1200fb_init(void)
-{
-	return platform_driver_register(&au1200fb_driver);
-}
-
-static void __exit au1200fb_cleanup(void)
-{
-	platform_driver_unregister(&au1200fb_driver);
-}
-
-module_init(au1200fb_init);
-module_exit(au1200fb_cleanup);
+module_platform_driver(au1200fb_driver);
 
 MODULE_DESCRIPTION(DRIVER_DESC);
 MODULE_LICENSE("GPL");
diff --git a/drivers/video/backlight/l4f00242t03.c b/drivers/video/backlight/l4f00242t03.c
index 923eae2f85f8..b5fc13bc24e7 100644
--- a/drivers/video/backlight/l4f00242t03.c
+++ b/drivers/video/backlight/l4f00242t03.c
@@ -244,7 +244,6 @@ static int l4f00242t03_remove(struct spi_device *spi)
 
 	l4f00242t03_lcd_power_set(priv->ld, FB_BLANK_POWERDOWN);
 	lcd_device_unregister(priv->ld);
-	spi_set_drvdata(spi, NULL);
 
 	return 0;
 }
diff --git a/drivers/video/backlight/pwm_bl.c b/drivers/video/backlight/pwm_bl.c
index 36db5d98dd2f..fb80d68f4d33 100644
--- a/drivers/video/backlight/pwm_bl.c
+++ b/drivers/video/backlight/pwm_bl.c
@@ -10,6 +10,8 @@
  * published by the Free Software Foundation.
  */
 
+#include <linux/gpio.h>
+#include <linux/of_gpio.h>
 #include <linux/module.h>
 #include <linux/kernel.h>
 #include <linux/init.h>
@@ -19,6 +21,7 @@
 #include <linux/err.h>
 #include <linux/pwm.h>
 #include <linux/pwm_backlight.h>
+#include <linux/regulator/consumer.h>
 #include <linux/slab.h>
 
 struct pwm_bl_data {
@@ -27,6 +30,11 @@ struct pwm_bl_data {
 	unsigned int		period;
 	unsigned int		lth_brightness;
 	unsigned int		*levels;
+	bool			enabled;
+	struct regulator	*power_supply;
+	int			enable_gpio;
+	unsigned long		enable_gpio_flags;
+	unsigned int		scale;
 	int			(*notify)(struct device *,
 					  int brightness);
 	void			(*notify_after)(struct device *,
@@ -35,11 +43,65 @@ struct pwm_bl_data {
 	void			(*exit)(struct device *);
 };
 
+static void pwm_backlight_power_on(struct pwm_bl_data *pb, int brightness)
+{
+	int err;
+
+	if (pb->enabled)
+		return;
+
+	err = regulator_enable(pb->power_supply);
+	if (err < 0)
+		dev_err(pb->dev, "failed to enable power supply\n");
+
+	if (gpio_is_valid(pb->enable_gpio)) {
+		if (pb->enable_gpio_flags & PWM_BACKLIGHT_GPIO_ACTIVE_LOW)
+			gpio_set_value(pb->enable_gpio, 0);
+		else
+			gpio_set_value(pb->enable_gpio, 1);
+	}
+
+	pwm_enable(pb->pwm);
+	pb->enabled = true;
+}
+
+static void pwm_backlight_power_off(struct pwm_bl_data *pb)
+{
+	if (!pb->enabled)
+		return;
+
+	pwm_config(pb->pwm, 0, pb->period);
+	pwm_disable(pb->pwm);
+
+	if (gpio_is_valid(pb->enable_gpio)) {
+		if (pb->enable_gpio_flags & PWM_BACKLIGHT_GPIO_ACTIVE_LOW)
+			gpio_set_value(pb->enable_gpio, 1);
+		else
+			gpio_set_value(pb->enable_gpio, 0);
+	}
+
+	regulator_disable(pb->power_supply);
+	pb->enabled = false;
+}
+
+static int compute_duty_cycle(struct pwm_bl_data *pb, int brightness)
+{
+	unsigned int lth = pb->lth_brightness;
+	int duty_cycle;
+
+	if (pb->levels)
+		duty_cycle = pb->levels[brightness];
+	else
+		duty_cycle = brightness;
+
+	return (duty_cycle * (pb->period - lth) / pb->scale) + lth;
+}
+
 static int pwm_backlight_update_status(struct backlight_device *bl)
 {
 	struct pwm_bl_data *pb = bl_get_data(bl);
 	int brightness = bl->props.brightness;
-	int max = bl->props.max_brightness;
+	int duty_cycle;
 
 	if (bl->props.power != FB_BLANK_UNBLANK ||
 	    bl->props.fb_blank != FB_BLANK_UNBLANK ||
@@ -49,24 +111,12 @@ static int pwm_backlight_update_status(struct backlight_device *bl)
 	if (pb->notify)
 		brightness = pb->notify(pb->dev, brightness);
 
-	if (brightness == 0) {
-		pwm_config(pb->pwm, 0, pb->period);
-		pwm_disable(pb->pwm);
-	} else {
-		int duty_cycle;
-
-		if (pb->levels) {
-			duty_cycle = pb->levels[brightness];
-			max = pb->levels[max];
-		} else {
-			duty_cycle = brightness;
-		}
-
-		duty_cycle = pb->lth_brightness +
-		     (duty_cycle * (pb->period - pb->lth_brightness) / max);
+	if (brightness > 0) {
+		duty_cycle = compute_duty_cycle(pb, brightness);
 		pwm_config(pb->pwm, duty_cycle, pb->period);
-		pwm_enable(pb->pwm);
-	}
+		pwm_backlight_power_on(pb, brightness);
+	} else
+		pwm_backlight_power_off(pb);
 
 	if (pb->notify_after)
 		pb->notify_after(pb->dev, brightness);
@@ -98,6 +148,7 @@ static int pwm_backlight_parse_dt(struct device *dev,
 				  struct platform_pwm_backlight_data *data)
 {
 	struct device_node *node = dev->of_node;
+	enum of_gpio_flags flags;
 	struct property *prop;
 	int length;
 	u32 value;
@@ -138,11 +189,13 @@ static int pwm_backlight_parse_dt(struct device *dev,
 		data->max_brightness--;
 	}
 
-	/*
-	 * TODO: Most users of this driver use a number of GPIOs to control
-	 *       backlight power. Support for specifying these needs to be
-	 *       added.
-	 */
+	data->enable_gpio = of_get_named_gpio_flags(node, "enable-gpios", 0,
+						    &flags);
+	if (data->enable_gpio == -EPROBE_DEFER)
+		return -EPROBE_DEFER;
+
+	if (gpio_is_valid(data->enable_gpio) && (flags & OF_GPIO_ACTIVE_LOW))
+		data->enable_gpio_flags |= PWM_BACKLIGHT_GPIO_ACTIVE_LOW;
 
 	return 0;
 }
@@ -168,7 +221,6 @@ static int pwm_backlight_probe(struct platform_device *pdev)
 	struct backlight_properties props;
 	struct backlight_device *bl;
 	struct pwm_bl_data *pb;
-	unsigned int max;
 	int ret;
 
 	if (!data) {
@@ -195,16 +247,46 @@ static int pwm_backlight_probe(struct platform_device *pdev)
 	}
 
 	if (data->levels) {
-		max = data->levels[data->max_brightness];
+		unsigned int i;
+
+		for (i = 0; i <= data->max_brightness; i++)
+			if (data->levels[i] > pb->scale)
+				pb->scale = data->levels[i];
+
 		pb->levels = data->levels;
 	} else
-		max = data->max_brightness;
+		pb->scale = data->max_brightness;
 
+	pb->enable_gpio = data->enable_gpio;
+	pb->enable_gpio_flags = data->enable_gpio_flags;
 	pb->notify = data->notify;
 	pb->notify_after = data->notify_after;
 	pb->check_fb = data->check_fb;
 	pb->exit = data->exit;
 	pb->dev = &pdev->dev;
+	pb->enabled = false;
+
+	if (gpio_is_valid(pb->enable_gpio)) {
+		unsigned long flags;
+
+		if (pb->enable_gpio_flags & PWM_BACKLIGHT_GPIO_ACTIVE_LOW)
+			flags = GPIOF_OUT_INIT_HIGH;
+		else
+			flags = GPIOF_OUT_INIT_LOW;
+
+		ret = gpio_request_one(pb->enable_gpio, flags, "enable");
+		if (ret < 0) {
+			dev_err(&pdev->dev, "failed to request GPIO#%d: %d\n",
+				pb->enable_gpio, ret);
+			goto err_alloc;
+		}
+	}
+
+	pb->power_supply = devm_regulator_get(&pdev->dev, "power");
+	if (IS_ERR(pb->power_supply)) {
+		ret = PTR_ERR(pb->power_supply);
+		goto err_gpio;
+	}
 
 	pb->pwm = devm_pwm_get(&pdev->dev, NULL);
 	if (IS_ERR(pb->pwm)) {
@@ -214,7 +296,7 @@ static int pwm_backlight_probe(struct platform_device *pdev)
 		if (IS_ERR(pb->pwm)) {
 			dev_err(&pdev->dev, "unable to request legacy PWM\n");
 			ret = PTR_ERR(pb->pwm);
-			goto err_alloc;
+			goto err_gpio;
 		}
 	}
 
@@ -229,7 +311,7 @@ static int pwm_backlight_probe(struct platform_device *pdev)
 		pwm_set_period(pb->pwm, data->pwm_period_ns);
 
 	pb->period = pwm_get_period(pb->pwm);
-	pb->lth_brightness = data->lth_brightness * (pb->period / max);
+	pb->lth_brightness = data->lth_brightness * (pb->period / pb->scale);
 
 	memset(&props, 0, sizeof(struct backlight_properties));
 	props.type = BACKLIGHT_RAW;
@@ -239,7 +321,7 @@ static int pwm_backlight_probe(struct platform_device *pdev)
 	if (IS_ERR(bl)) {
 		dev_err(&pdev->dev, "failed to register backlight\n");
 		ret = PTR_ERR(bl);
-		goto err_alloc;
+		goto err_gpio;
 	}
 
 	if (data->dft_brightness > data->max_brightness) {
@@ -255,6 +337,9 @@ static int pwm_backlight_probe(struct platform_device *pdev)
 	platform_set_drvdata(pdev, bl);
 	return 0;
 
+err_gpio:
+	if (gpio_is_valid(pb->enable_gpio))
+		gpio_free(pb->enable_gpio);
 err_alloc:
 	if (data->exit)
 		data->exit(&pdev->dev);
@@ -267,10 +352,11 @@ static int pwm_backlight_remove(struct platform_device *pdev)
 	struct pwm_bl_data *pb = bl_get_data(bl);
 
 	backlight_device_unregister(bl);
-	pwm_config(pb->pwm, 0, pb->period);
-	pwm_disable(pb->pwm);
+	pwm_backlight_power_off(pb);
+
 	if (pb->exit)
 		pb->exit(&pdev->dev);
+
 	return 0;
 }
 
@@ -282,10 +368,12 @@ static int pwm_backlight_suspend(struct device *dev)
 
 	if (pb->notify)
 		pb->notify(pb->dev, 0);
-	pwm_config(pb->pwm, 0, pb->period);
-	pwm_disable(pb->pwm);
+
+	pwm_backlight_power_off(pb);
+
 	if (pb->notify_after)
 		pb->notify_after(pb->dev, 0);
+
 	return 0;
 }
 
@@ -294,12 +382,19 @@ static int pwm_backlight_resume(struct device *dev)
 	struct backlight_device *bl = dev_get_drvdata(dev);
 
 	backlight_update_status(bl);
+
 	return 0;
 }
 #endif
 
-static SIMPLE_DEV_PM_OPS(pwm_backlight_pm_ops, pwm_backlight_suspend,
-			 pwm_backlight_resume);
+static const struct dev_pm_ops pwm_backlight_pm_ops = {
+#ifdef CONFIG_PM_SLEEP
+	.suspend = pwm_backlight_suspend,
+	.resume = pwm_backlight_resume,
+	.poweroff = pwm_backlight_suspend,
+	.restore = pwm_backlight_resume,
+#endif
+};
 
 static struct platform_driver pwm_backlight_driver = {
 	.driver		= {
@@ -317,4 +412,3 @@ module_platform_driver(pwm_backlight_driver);
 MODULE_DESCRIPTION("PWM based Backlight Driver");
 MODULE_LICENSE("GPL");
 MODULE_ALIAS("platform:pwm-backlight");
-
diff --git a/drivers/video/backlight/tosa_lcd.c b/drivers/video/backlight/tosa_lcd.c
index bf081573e5b5..be5d636764bf 100644
--- a/drivers/video/backlight/tosa_lcd.c
+++ b/drivers/video/backlight/tosa_lcd.c
@@ -198,7 +198,7 @@ static int tosa_lcd_probe(struct spi_device *spi)
 	ret = devm_gpio_request_one(&spi->dev, TOSA_GPIO_TG_ON,
 				GPIOF_OUT_INIT_LOW, "tg #pwr");
 	if (ret < 0)
-		goto err_gpio_tg;
+		return ret;
 
 	mdelay(60);
 
@@ -219,8 +219,6 @@ static int tosa_lcd_probe(struct spi_device *spi)
 
 err_register:
 	tosa_lcd_tg_off(data);
-err_gpio_tg:
-	spi_set_drvdata(spi, NULL);
 	return ret;
 }
 
@@ -235,8 +233,6 @@ static int tosa_lcd_remove(struct spi_device *spi)
 
 	tosa_lcd_tg_off(data);
 
-	spi_set_drvdata(spi, NULL);
-
 	return 0;
 }
 
diff --git a/drivers/video/bf54x-lq043fb.c b/drivers/video/bf54x-lq043fb.c
index 87f288bfc58c..42b8f9d11018 100644
--- a/drivers/video/bf54x-lq043fb.c
+++ b/drivers/video/bf54x-lq043fb.c
@@ -761,19 +761,7 @@ static struct platform_driver bfin_bf54x_driver = {
 		   .owner = THIS_MODULE,
 		   },
 };
-
-static int __init bfin_bf54x_driver_init(void)
-{
-	return platform_driver_register(&bfin_bf54x_driver);
-}
-
-static void __exit bfin_bf54x_driver_cleanup(void)
-{
-	platform_driver_unregister(&bfin_bf54x_driver);
-}
+module_platform_driver(bfin_bf54x_driver);
 
 MODULE_DESCRIPTION("Blackfin BF54x TFT LCD Driver");
 MODULE_LICENSE("GPL");
-
-module_init(bfin_bf54x_driver_init);
-module_exit(bfin_bf54x_driver_cleanup);
diff --git a/drivers/video/bfin-t350mcqb-fb.c b/drivers/video/bfin-t350mcqb-fb.c
index 48c0c4e38a62..b5cf1307a3d9 100644
--- a/drivers/video/bfin-t350mcqb-fb.c
+++ b/drivers/video/bfin-t350mcqb-fb.c
@@ -664,19 +664,7 @@ static struct platform_driver bfin_t350mcqb_driver = {
 		   .owner = THIS_MODULE,
 		   },
 };
-
-static int __init bfin_t350mcqb_driver_init(void)
-{
-	return platform_driver_register(&bfin_t350mcqb_driver);
-}
-
-static void __exit bfin_t350mcqb_driver_cleanup(void)
-{
-	platform_driver_unregister(&bfin_t350mcqb_driver);
-}
+module_platform_driver(bfin_t350mcqb_driver);
 
 MODULE_DESCRIPTION("Blackfin TFT LCD Driver");
 MODULE_LICENSE("GPL");
-
-module_init(bfin_t350mcqb_driver_init);
-module_exit(bfin_t350mcqb_driver_cleanup);
diff --git a/drivers/video/broadsheetfb.c b/drivers/video/broadsheetfb.c
index b09701c79432..8556264b16b7 100644
--- a/drivers/video/broadsheetfb.c
+++ b/drivers/video/broadsheetfb.c
@@ -1167,9 +1167,8 @@ static int broadsheetfb_probe(struct platform_device *dev)
 	if (retval < 0)
 		goto err_unreg_fb;
 
-	printk(KERN_INFO
-	       "fb%d: Broadsheet frame buffer, using %dK of video memory\n",
-	       info->node, videomemorysize >> 10);
+	fb_info(info, "Broadsheet frame buffer, using %dK of video memory\n",
+		videomemorysize >> 10);
 
 
 	return 0;
@@ -1217,19 +1216,7 @@ static struct platform_driver broadsheetfb_driver = {
 		.name	= "broadsheetfb",
 	},
 };
-
-static int __init broadsheetfb_init(void)
-{
-	return platform_driver_register(&broadsheetfb_driver);
-}
-
-static void __exit broadsheetfb_exit(void)
-{
-	platform_driver_unregister(&broadsheetfb_driver);
-}
-
-module_init(broadsheetfb_init);
-module_exit(broadsheetfb_exit);
+module_platform_driver(broadsheetfb_driver);
 
 MODULE_DESCRIPTION("fbdev driver for Broadsheet controller");
 MODULE_AUTHOR("Jaya Kumar");
diff --git a/drivers/video/bw2.c b/drivers/video/bw2.c
index 60017fc634b5..bc123d6947a4 100644
--- a/drivers/video/bw2.c
+++ b/drivers/video/bw2.c
@@ -363,8 +363,6 @@ static int bw2_remove(struct platform_device *op)
 
 	framebuffer_release(info);
 
-	dev_set_drvdata(&op->dev, NULL);
-
 	return 0;
 }
 
diff --git a/drivers/video/carminefb.c b/drivers/video/carminefb.c
index 153dd65b0ae8..65f7c15f5fdb 100644
--- a/drivers/video/carminefb.c
+++ b/drivers/video/carminefb.c
@@ -585,8 +585,7 @@ static int alloc_carmine_fb(void __iomem *regs, void __iomem *smem_base,
 	if (ret < 0)
 		goto err_dealloc_cmap;
 
-	printk(KERN_INFO "fb%d: %s frame buffer device\n", info->node,
-			info->fix.id);
+	fb_info(info, "%s frame buffer device\n", info->fix.id);
 
 	*rinfo = info;
 	return 0;
@@ -746,7 +745,6 @@ static void carminefb_remove(struct pci_dev *dev)
 	iounmap(hw->v_regs);
 	release_mem_region(fix.mmio_start, fix.mmio_len);
 
-	pci_set_drvdata(dev, NULL);
 	pci_disable_device(dev);
 	kfree(hw);
 }
diff --git a/drivers/video/cfbimgblt.c b/drivers/video/cfbimgblt.c
index baed57d3cfff..a2bb276a8b24 100644
--- a/drivers/video/cfbimgblt.c
+++ b/drivers/video/cfbimgblt.c
@@ -181,7 +181,7 @@ static inline void slow_imageblit(const struct fb_image *image, struct fb_info *
 			}
 			shift += bpp;
 			shift &= (32 - 1);
-			if (!l) { l = 8; s++; };
+			if (!l) { l = 8; s++; }
 		}
 
 		/* write trailing bits */
diff --git a/drivers/video/cg14.c b/drivers/video/cg14.c
index ed3b8891e006..c79745b136bb 100644
--- a/drivers/video/cg14.c
+++ b/drivers/video/cg14.c
@@ -330,7 +330,7 @@ static int cg14_ioctl(struct fb_info *info, unsigned int cmd, unsigned long arg)
 		default:
 			ret = -ENOSYS;
 			break;
-		};
+		}
 		if (!ret) {
 			sbus_writeb(cur_mode, &regs->mcr);
 			par->mode = mode;
@@ -343,7 +343,7 @@ static int cg14_ioctl(struct fb_info *info, unsigned int cmd, unsigned long arg)
 					  FBTYPE_MDICOLOR, 8,
 					  info->fix.smem_len);
 		break;
-	};
+	}
 
 	return ret;
 }
@@ -583,8 +583,6 @@ static int cg14_remove(struct platform_device *op)
 
 	framebuffer_release(info);
 
-	dev_set_drvdata(&op->dev, NULL);
-
 	return 0;
 }
 
diff --git a/drivers/video/cg3.c b/drivers/video/cg3.c
index 9f63507ded37..64a89d5747ed 100644
--- a/drivers/video/cg3.c
+++ b/drivers/video/cg3.c
@@ -446,8 +446,6 @@ static int cg3_remove(struct platform_device *op)
 
 	framebuffer_release(info);
 
-	dev_set_drvdata(&op->dev, NULL);
-
 	return 0;
 }
 
diff --git a/drivers/video/cg6.c b/drivers/video/cg6.c
index 3545decc7485..70781fea092a 100644
--- a/drivers/video/cg6.c
+++ b/drivers/video/cg6.c
@@ -624,7 +624,7 @@ static void cg6_init_fix(struct fb_info *info, int linebytes)
 	default:
 		cg6_cpu_name = "i386";
 		break;
-	};
+	}
 	if (((conf >> CG6_FHC_REV_SHIFT) & CG6_FHC_REV_MASK) >= 11) {
 		if (info->fix.smem_len <= 0x100000)
 			cg6_card_name = "TGX";
@@ -839,8 +839,6 @@ static int cg6_remove(struct platform_device *op)
 
 	framebuffer_release(info);
 
-	dev_set_drvdata(&op->dev, NULL);
-
 	return 0;
 }
 
diff --git a/drivers/video/cirrusfb.c b/drivers/video/cirrusfb.c
index 97db3ba8f237..5aab9b9dc210 100644
--- a/drivers/video/cirrusfb.c
+++ b/drivers/video/cirrusfb.c
@@ -595,11 +595,6 @@ static int cirrusfb_check_var(struct fb_var_screeninfo *var,
 		return -EINVAL;
 	}
 
-	if (var->xoffset < 0)
-		var->xoffset = 0;
-	if (var->yoffset < 0)
-		var->yoffset = 0;
-
 	/* truncate xoffset and yoffset to maximum if too high */
 	if (var->xoffset > var->xres_virtual - var->xres)
 		var->xoffset = var->xres_virtual - var->xres - 1;
@@ -2159,7 +2154,6 @@ static int cirrusfb_pci_register(struct pci_dev *pdev,
 	if (!ret)
 		return 0;
 
-	pci_set_drvdata(pdev, NULL);
 	iounmap(info->screen_base);
 err_release_legacy:
 	if (release_io_ports)
diff --git a/drivers/video/cobalt_lcdfb.c b/drivers/video/cobalt_lcdfb.c
index a9031498e10c..d5533f4db1cf 100644
--- a/drivers/video/cobalt_lcdfb.c
+++ b/drivers/video/cobalt_lcdfb.c
@@ -368,8 +368,7 @@ static int cobalt_lcdfb_probe(struct platform_device *dev)
 
 	lcd_clear(info);
 
-	printk(KERN_INFO "fb%d: Cobalt server LCD frame buffer device\n",
-		info->node);
+	fb_info(info, "Cobalt server LCD frame buffer device\n");
 
 	return 0;
 }
@@ -395,19 +394,7 @@ static struct platform_driver cobalt_lcdfb_driver = {
 		.owner	= THIS_MODULE,
 	},
 };
-
-static int __init cobalt_lcdfb_init(void)
-{
-	return platform_driver_register(&cobalt_lcdfb_driver);
-}
-
-static void __exit cobalt_lcdfb_exit(void)
-{
-	platform_driver_unregister(&cobalt_lcdfb_driver);
-}
-
-module_init(cobalt_lcdfb_init);
-module_exit(cobalt_lcdfb_exit);
+module_platform_driver(cobalt_lcdfb_driver);
 
 MODULE_LICENSE("GPL v2");
 MODULE_AUTHOR("Yoichi Yuasa");
diff --git a/drivers/video/controlfb.c b/drivers/video/controlfb.c
index 67b77b40aa7f..fdadef979238 100644
--- a/drivers/video/controlfb.c
+++ b/drivers/video/controlfb.c
@@ -471,8 +471,8 @@ try_again:
 	/* Register with fbdev layer */
 	if (register_framebuffer(&p->info) < 0)
 		return -ENXIO;
-	
-	printk(KERN_INFO "fb%d: control display adapter\n", p->info.node);	
+
+	fb_info(&p->info, "control display adapter\n");
 
 	return 0;
 }
diff --git a/drivers/video/cyber2000fb.c b/drivers/video/cyber2000fb.c
index 1c446bc48b42..b0a950f36970 100644
--- a/drivers/video/cyber2000fb.c
+++ b/drivers/video/cyber2000fb.c
@@ -1810,11 +1810,6 @@ static void cyberpro_pci_remove(struct pci_dev *dev)
 		iounmap(cfb->region);
 		cyberpro_free_fb_info(cfb);
 
-		/*
-		 * Ensure that the driver data is no longer
-		 * valid.
-		 */
-		pci_set_drvdata(dev, NULL);
 		if (cfb == int_cfb_info)
 			int_cfb_info = NULL;
 
diff --git a/drivers/video/da8xx-fb.c b/drivers/video/da8xx-fb.c
index e030e17a83f2..a1d74dd11988 100644
--- a/drivers/video/da8xx-fb.c
+++ b/drivers/video/da8xx-fb.c
@@ -129,7 +129,6 @@
 
 #define LCD_NUM_BUFFERS	2
 
-#define WSI_TIMEOUT	50
 #define PALETTE_SIZE	256
 
 #define	CLK_MIN_DIV	2
@@ -1314,7 +1313,7 @@ static struct fb_ops da8xx_fb_ops = {
 
 static struct fb_videomode *da8xx_fb_get_videomode(struct platform_device *dev)
 {
-	struct da8xx_lcdc_platform_data *fb_pdata = dev->dev.platform_data;
+	struct da8xx_lcdc_platform_data *fb_pdata = dev_get_platdata(&dev->dev);
 	struct fb_videomode *lcdc_info;
 	int i;
 
@@ -1336,7 +1335,7 @@ static struct fb_videomode *da8xx_fb_get_videomode(struct platform_device *dev)
 static int fb_probe(struct platform_device *device)
 {
 	struct da8xx_lcdc_platform_data *fb_pdata =
-						device->dev.platform_data;
+						dev_get_platdata(&device->dev);
 	static struct resource *lcdc_regs;
 	struct lcd_ctrl_config *lcd_cfg;
 	struct fb_videomode *lcdc_info;
@@ -1548,7 +1547,7 @@ err_pm_runtime_disable:
 }
 
 #ifdef CONFIG_PM
-struct lcdc_context {
+static struct lcdc_context {
 	u32 clk_enable;
 	u32 ctrl;
 	u32 dma_ctrl;
@@ -1663,19 +1662,7 @@ static struct platform_driver da8xx_fb_driver = {
 		   .owner = THIS_MODULE,
 		   },
 };
-
-static int __init da8xx_fb_init(void)
-{
-	return platform_driver_register(&da8xx_fb_driver);
-}
-
-static void __exit da8xx_fb_cleanup(void)
-{
-	platform_driver_unregister(&da8xx_fb_driver);
-}
-
-module_init(da8xx_fb_init);
-module_exit(da8xx_fb_cleanup);
+module_platform_driver(da8xx_fb_driver);
 
 MODULE_DESCRIPTION("Framebuffer driver for TI da8xx/omap-l1xx");
 MODULE_AUTHOR("Texas Instruments");
diff --git a/drivers/video/efifb.c b/drivers/video/efifb.c
index 7f9ff75d0db2..cd7c0df9f24b 100644
--- a/drivers/video/efifb.c
+++ b/drivers/video/efifb.c
@@ -108,8 +108,8 @@ static int efifb_setup(char *options)
 			if (!*this_opt) continue;
 
 			for (i = 0; i < M_UNKNOWN; i++) {
-				if (!strcmp(this_opt, efifb_dmi_list[i].optname) &&
-				    efifb_dmi_list[i].base != 0) {
+				if (efifb_dmi_list[i].base != 0 &&
+				    !strcmp(this_opt, efifb_dmi_list[i].optname)) {
 					screen_info.lfb_base = efifb_dmi_list[i].base;
 					screen_info.lfb_linelength = efifb_dmi_list[i].stride;
 					screen_info.lfb_width = efifb_dmi_list[i].width;
@@ -322,8 +322,7 @@ static int efifb_probe(struct platform_device *dev)
 		printk(KERN_ERR "efifb: cannot register framebuffer\n");
 		goto err_fb_dealoc;
 	}
-	printk(KERN_INFO "fb%d: %s frame buffer device\n",
-		info->node, info->fix.id);
+	fb_info(info, "%s frame buffer device\n", info->fix.id);
 	return 0;
 
 err_fb_dealoc:
diff --git a/drivers/video/ep93xx-fb.c b/drivers/video/ep93xx-fb.c
index 28a837dfddd1..35a0f533f1a2 100644
--- a/drivers/video/ep93xx-fb.c
+++ b/drivers/video/ep93xx-fb.c
@@ -487,7 +487,7 @@ static void ep93xxfb_dealloc_videomem(struct fb_info *info)
 
 static int ep93xxfb_probe(struct platform_device *pdev)
 {
-	struct ep93xxfb_mach_info *mach_info = pdev->dev.platform_data;
+	struct ep93xxfb_mach_info *mach_info = dev_get_platdata(&pdev->dev);
 	struct fb_info *info;
 	struct ep93xx_fbi *fbi;
 	struct resource *res;
diff --git a/drivers/video/exynos/exynos_mipi_dsi.c b/drivers/video/exynos/exynos_mipi_dsi.c
index 00b3a52c1d68..cee9602f9a7b 100644
--- a/drivers/video/exynos/exynos_mipi_dsi.c
+++ b/drivers/video/exynos/exynos_mipi_dsi.c
@@ -141,7 +141,6 @@ static int exynos_mipi_dsi_early_blank_mode(struct mipi_dsim_device *dsim,
 
 static int exynos_mipi_dsi_blank_mode(struct mipi_dsim_device *dsim, int power)
 {
-	struct platform_device *pdev = to_platform_device(dsim->dev);
 	struct mipi_dsim_lcd_driver *client_drv = dsim->dsim_lcd_drv;
 	struct mipi_dsim_lcd_device *client_dev = dsim->dsim_lcd_dev;
 
diff --git a/drivers/video/exynos/exynos_mipi_dsi_common.c b/drivers/video/exynos/exynos_mipi_dsi_common.c
index 520fc9bd887b..85edabfdef5a 100644
--- a/drivers/video/exynos/exynos_mipi_dsi_common.c
+++ b/drivers/video/exynos/exynos_mipi_dsi_common.c
@@ -220,7 +220,7 @@ int exynos_mipi_dsi_wr_data(struct mipi_dsim_device *dsim, unsigned int data_id,
 	case MIPI_DSI_DCS_LONG_WRITE:
 	{
 		unsigned int size, payload = 0;
-		INIT_COMPLETION(dsim_wr_comp);
+		reinit_completion(&dsim_wr_comp);
 
 		size = data_size * 4;
 
@@ -356,7 +356,7 @@ int exynos_mipi_dsi_rd_data(struct mipi_dsim_device *dsim, unsigned int data_id,
 	msleep(20);
 
 	mutex_lock(&dsim->lock);
-	INIT_COMPLETION(dsim_rd_comp);
+	reinit_completion(&dsim_rd_comp);
 	exynos_mipi_dsi_rd_tx_header(dsim,
 		MIPI_DSI_SET_MAXIMUM_RETURN_PACKET_SIZE, req_size);
 
@@ -376,6 +376,7 @@ int exynos_mipi_dsi_rd_data(struct mipi_dsim_device *dsim, unsigned int data_id,
 			"data id %x is not supported current DSI spec.\n",
 			data_id);
 
+		mutex_unlock(&dsim->lock);
 		return -EINVAL;
 	}
 
@@ -667,7 +668,7 @@ int exynos_mipi_dsi_init_dsim(struct mipi_dsim_device *dsim)
 	default:
 		dev_info(dsim->dev, "data lane is invalid.\n");
 		return -EINVAL;
-	};
+	}
 
 	exynos_mipi_dsi_sw_reset(dsim);
 	exynos_mipi_dsi_func_reset(dsim);
diff --git a/drivers/video/fb-puv3.c b/drivers/video/fb-puv3.c
index 27fc956166fa..6db9ebd042a3 100644
--- a/drivers/video/fb-puv3.c
+++ b/drivers/video/fb-puv3.c
@@ -713,9 +713,8 @@ static int unifb_probe(struct platform_device *dev)
 	platform_set_drvdata(dev, info);
 	platform_device_add_data(dev, unifb_regs, sizeof(u32) * UNIFB_REGS_NUM);
 
-	printk(KERN_INFO
-	       "fb%d: Virtual frame buffer device, using %dM of video memory\n",
-	       info->node, UNIFB_MEMSIZE >> 20);
+	fb_info(info, "Virtual frame buffer device, using %dM of video memory\n",
+		UNIFB_MEMSIZE >> 20);
 	return 0;
 err2:
 	fb_dealloc_cmap(&info->cmap);
diff --git a/drivers/video/fbmem.c b/drivers/video/fbmem.c
index dacaf74256a3..010d19105ebc 100644
--- a/drivers/video/fbmem.c
+++ b/drivers/video/fbmem.c
@@ -1108,14 +1108,16 @@ static long do_fb_ioctl(struct fb_info *info, unsigned int cmd,
 	case FBIOPUT_VSCREENINFO:
 		if (copy_from_user(&var, argp, sizeof(var)))
 			return -EFAULT;
-		if (!lock_fb_info(info))
-			return -ENODEV;
 		console_lock();
+		if (!lock_fb_info(info)) {
+			console_unlock();
+			return -ENODEV;
+		}
 		info->flags |= FBINFO_MISC_USEREVENT;
 		ret = fb_set_var(info, &var);
 		info->flags &= ~FBINFO_MISC_USEREVENT;
-		console_unlock();
 		unlock_fb_info(info);
+		console_unlock();
 		if (!ret && copy_to_user(argp, &var, sizeof(var)))
 			ret = -EFAULT;
 		break;
@@ -1144,12 +1146,14 @@ static long do_fb_ioctl(struct fb_info *info, unsigned int cmd,
 	case FBIOPAN_DISPLAY:
 		if (copy_from_user(&var, argp, sizeof(var)))
 			return -EFAULT;
-		if (!lock_fb_info(info))
-			return -ENODEV;
 		console_lock();
+		if (!lock_fb_info(info)) {
+			console_unlock();
+			return -ENODEV;
+		}
 		ret = fb_pan_display(info, &var);
-		console_unlock();
 		unlock_fb_info(info);
+		console_unlock();
 		if (ret == 0 && copy_to_user(argp, &var, sizeof(var)))
 			return -EFAULT;
 		break;
@@ -1184,23 +1188,27 @@ static long do_fb_ioctl(struct fb_info *info, unsigned int cmd,
 			break;
 		}
 		event.data = &con2fb;
-		if (!lock_fb_info(info))
-			return -ENODEV;
 		console_lock();
+		if (!lock_fb_info(info)) {
+			console_unlock();
+			return -ENODEV;
+		}
 		event.info = info;
 		ret = fb_notifier_call_chain(FB_EVENT_SET_CONSOLE_MAP, &event);
-		console_unlock();
 		unlock_fb_info(info);
+		console_unlock();
 		break;
 	case FBIOBLANK:
-		if (!lock_fb_info(info))
-			return -ENODEV;
 		console_lock();
+		if (!lock_fb_info(info)) {
+			console_unlock();
+			return -ENODEV;
+		}
 		info->flags |= FBINFO_MISC_USEREVENT;
 		ret = fb_blank(info, arg);
 		info->flags &= ~FBINFO_MISC_USEREVENT;
-		console_unlock();
 		unlock_fb_info(info);
+		console_unlock();
 		break;
 	default:
 		if (!lock_fb_info(info))
@@ -1660,12 +1668,15 @@ static int do_register_framebuffer(struct fb_info *fb_info)
 	registered_fb[i] = fb_info;
 
 	event.info = fb_info;
-	if (!lock_fb_info(fb_info))
-		return -ENODEV;
 	console_lock();
+	if (!lock_fb_info(fb_info)) {
+		console_unlock();
+		return -ENODEV;
+	}
+
 	fb_notifier_call_chain(FB_EVENT_FB_REGISTERED, &event);
-	console_unlock();
 	unlock_fb_info(fb_info);
+	console_unlock();
 	return 0;
 }
 
@@ -1678,13 +1689,16 @@ static int do_unregister_framebuffer(struct fb_info *fb_info)
 	if (i < 0 || i >= FB_MAX || registered_fb[i] != fb_info)
 		return -EINVAL;
 
-	if (!lock_fb_info(fb_info))
-		return -ENODEV;
 	console_lock();
+	if (!lock_fb_info(fb_info)) {
+		console_unlock();
+		return -ENODEV;
+	}
+
 	event.info = fb_info;
 	ret = fb_notifier_call_chain(FB_EVENT_FB_UNBIND, &event);
-	console_unlock();
 	unlock_fb_info(fb_info);
+	console_unlock();
 
 	if (ret)
 		return -EINVAL;
diff --git a/drivers/video/fbsysfs.c b/drivers/video/fbsysfs.c
index ef476b02fbe5..53444ac19fe0 100644
--- a/drivers/video/fbsysfs.c
+++ b/drivers/video/fbsysfs.c
@@ -177,9 +177,12 @@ static ssize_t store_modes(struct device *device,
 	if (i * sizeof(struct fb_videomode) != count)
 		return -EINVAL;
 
-	if (!lock_fb_info(fb_info))
-		return -ENODEV;
 	console_lock();
+	if (!lock_fb_info(fb_info)) {
+		console_unlock();
+		return -ENODEV;
+	}
+
 	list_splice(&fb_info->modelist, &old_list);
 	fb_videomode_to_modelist((const struct fb_videomode *)buf, i,
 				 &fb_info->modelist);
@@ -189,8 +192,8 @@ static ssize_t store_modes(struct device *device,
 	} else
 		fb_destroy_modelist(&old_list);
 
-	console_unlock();
 	unlock_fb_info(fb_info);
+	console_unlock();
 
 	return 0;
 }
@@ -404,12 +407,16 @@ static ssize_t store_fbstate(struct device *device,
 
 	state = simple_strtoul(buf, &last, 0);
 
-	if (!lock_fb_info(fb_info))
-		return -ENODEV;
 	console_lock();
+	if (!lock_fb_info(fb_info)) {
+		console_unlock();
+		return -ENODEV;
+	}
+
 	fb_set_suspend(fb_info, (int)state);
-	console_unlock();
+
 	unlock_fb_info(fb_info);
+	console_unlock();
 
 	return count;
 }
diff --git a/drivers/video/ffb.c b/drivers/video/ffb.c
index 6d2744794dd1..4c4ffa61ae26 100644
--- a/drivers/video/ffb.c
+++ b/drivers/video/ffb.c
@@ -1035,8 +1035,6 @@ static int ffb_remove(struct platform_device *op)
 
 	framebuffer_release(info);
 
-	dev_set_drvdata(&op->dev, NULL);
-
 	return 0;
 }
 
diff --git a/drivers/video/fm2fb.c b/drivers/video/fm2fb.c
index c99c9671302b..e69d47af9932 100644
--- a/drivers/video/fm2fb.c
+++ b/drivers/video/fm2fb.c
@@ -289,7 +289,7 @@ static int fm2fb_probe(struct zorro_dev *z, const struct zorro_device_id *id)
 		zorro_release_device(z);
 		return -EINVAL;
 	}
-	printk("fb%d: %s frame buffer device\n", info->node, fb_fix.id);
+	fb_info(info, "%s frame buffer device\n", fb_fix.id);
 	return 0;
 }
 
diff --git a/drivers/video/fsl-diu-fb.c b/drivers/video/fsl-diu-fb.c
index b047ec58ac30..e8758b9c3bcc 100644
--- a/drivers/video/fsl-diu-fb.c
+++ b/drivers/video/fsl-diu-fb.c
@@ -1104,7 +1104,7 @@ static int fsl_diu_cursor(struct fb_info *info, struct fb_cursor *cursor)
 
 		fsl_diu_load_cursor_image(info, image, bg, fg,
 			cursor->image.width, cursor->image.height);
-	};
+	}
 
 	/*
 	 * Show or hide the cursor.  The cursor data is always stored in the
diff --git a/drivers/video/gbefb.c b/drivers/video/gbefb.c
index ceab37020fff..4c7cb368a9dc 100644
--- a/drivers/video/gbefb.c
+++ b/drivers/video/gbefb.c
@@ -1236,9 +1236,9 @@ static int gbefb_probe(struct platform_device *p_dev)
 	platform_set_drvdata(p_dev, info);
 	gbefb_create_sysfs(&p_dev->dev);
 
-	printk(KERN_INFO "fb%d: %s rev %d @ 0x%08x using %dkB memory\n",
-	       info->node, info->fix.id, gbe_revision, (unsigned) GBE_BASE,
-	       gbe_mem_size >> 10);
+	fb_info(info, "%s rev %d @ 0x%08x using %dkB memory\n",
+		info->fix.id, gbe_revision, (unsigned)GBE_BASE,
+		gbe_mem_size >> 10);
 
 	return 0;
 
diff --git a/drivers/video/geode/gx1fb_core.c b/drivers/video/geode/gx1fb_core.c
index ebbaada7b941..2794ba11f332 100644
--- a/drivers/video/geode/gx1fb_core.c
+++ b/drivers/video/geode/gx1fb_core.c
@@ -357,7 +357,7 @@ static int gx1fb_probe(struct pci_dev *pdev, const struct pci_device_id *id)
 		goto err;
 	}
 	pci_set_drvdata(pdev, info);
-	printk(KERN_INFO "fb%d: %s frame buffer device\n", info->node, info->fix.id);
+	fb_info(info, "%s frame buffer device\n", info->fix.id);
 	return 0;
 
   err:
@@ -399,7 +399,6 @@ static void gx1fb_remove(struct pci_dev *pdev)
 	release_mem_region(gx1_gx_base() + 0x8300, 0x100);
 
 	fb_dealloc_cmap(&info->cmap);
-	pci_set_drvdata(pdev, NULL);
 
 	framebuffer_release(info);
 }
diff --git a/drivers/video/geode/gxfb_core.c b/drivers/video/geode/gxfb_core.c
index 19f0c1add747..1790f14bab15 100644
--- a/drivers/video/geode/gxfb_core.c
+++ b/drivers/video/geode/gxfb_core.c
@@ -423,7 +423,7 @@ static int gxfb_probe(struct pci_dev *pdev, const struct pci_device_id *id)
 		goto err;
 	}
 	pci_set_drvdata(pdev, info);
-	printk(KERN_INFO "fb%d: %s frame buffer device\n", info->node, info->fix.id);
+	fb_info(info, "%s frame buffer device\n", info->fix.id);
 	return 0;
 
   err:
@@ -471,7 +471,6 @@ static void gxfb_remove(struct pci_dev *pdev)
 	pci_release_region(pdev, 1);
 
 	fb_dealloc_cmap(&info->cmap);
-	pci_set_drvdata(pdev, NULL);
 
 	framebuffer_release(info);
 }
diff --git a/drivers/video/geode/lxfb_core.c b/drivers/video/geode/lxfb_core.c
index 4dd7b5566962..9e1d19d673a1 100644
--- a/drivers/video/geode/lxfb_core.c
+++ b/drivers/video/geode/lxfb_core.c
@@ -555,8 +555,7 @@ static int lxfb_probe(struct pci_dev *pdev, const struct pci_device_id *id)
 		goto err;
 	}
 	pci_set_drvdata(pdev, info);
-	printk(KERN_INFO "fb%d: %s frame buffer device\n",
-		info->node, info->fix.id);
+	fb_info(info, "%s frame buffer device\n", info->fix.id);
 
 	return 0;
 
@@ -606,7 +605,6 @@ static void lxfb_remove(struct pci_dev *pdev)
 	pci_release_region(pdev, 3);
 
 	fb_dealloc_cmap(&info->cmap);
-	pci_set_drvdata(pdev, NULL);
 	framebuffer_release(info);
 }
 
diff --git a/drivers/video/grvga.c b/drivers/video/grvga.c
index 861109e7de1b..c078701f15f6 100644
--- a/drivers/video/grvga.c
+++ b/drivers/video/grvga.c
@@ -496,7 +496,6 @@ static int grvga_probe(struct platform_device *dev)
 	return 0;
 
 free_mem:
-	dev_set_drvdata(&dev->dev, NULL);
 	if (grvga_fix_addr)
 		iounmap((void *)virtual_start);
 	else
@@ -530,7 +529,6 @@ static int grvga_remove(struct platform_device *device)
 			kfree((void *)info->screen_base);
 
 		framebuffer_release(info);
-		dev_set_drvdata(&device->dev, NULL);
 	}
 
 	return 0;
@@ -557,19 +555,7 @@ static struct platform_driver grvga_driver = {
 	.remove		= grvga_remove,
 };
 
-
-static int __init grvga_init(void)
-{
-	return platform_driver_register(&grvga_driver);
-}
-
-static void __exit grvga_exit(void)
-{
-	platform_driver_unregister(&grvga_driver);
-}
-
-module_init(grvga_init);
-module_exit(grvga_exit);
+module_platform_driver(grvga_driver);
 
 MODULE_LICENSE("GPL");
 MODULE_AUTHOR("Aeroflex Gaisler");
diff --git a/drivers/video/gxt4500.c b/drivers/video/gxt4500.c
index c35663f6a54a..135d78a02588 100644
--- a/drivers/video/gxt4500.c
+++ b/drivers/video/gxt4500.c
@@ -698,8 +698,7 @@ static int gxt4500_probe(struct pci_dev *pdev, const struct pci_device_id *ent)
 		dev_err(&pdev->dev, "gxt4500: cannot register framebuffer\n");
 		goto err_free_cmap;
 	}
-	printk(KERN_INFO "fb%d: %s frame buffer device\n",
-	       info->node, info->fix.id);
+	fb_info(info, "%s frame buffer device\n", info->fix.id);
 
 	return 0;
 
diff --git a/drivers/video/hecubafb.c b/drivers/video/hecubafb.c
index 59d23181fdb0..f64120ec9192 100644
--- a/drivers/video/hecubafb.c
+++ b/drivers/video/hecubafb.c
@@ -261,9 +261,8 @@ static int hecubafb_probe(struct platform_device *dev)
 		goto err_fbreg;
 	platform_set_drvdata(dev, info);
 
-	printk(KERN_INFO
-	       "fb%d: Hecuba frame buffer device, using %dK of video memory\n",
-	       info->node, videomemorysize >> 10);
+	fb_info(info, "Hecuba frame buffer device, using %dK of video memory\n",
+		videomemorysize >> 10);
 
 	/* this inits the dpy */
 	retval = par->board->init(par);
@@ -305,19 +304,7 @@ static struct platform_driver hecubafb_driver = {
 		.name	= "hecubafb",
 	},
 };
-
-static int __init hecubafb_init(void)
-{
-	return platform_driver_register(&hecubafb_driver);
-}
-
-static void __exit hecubafb_exit(void)
-{
-	platform_driver_unregister(&hecubafb_driver);
-}
-
-module_init(hecubafb_init);
-module_exit(hecubafb_exit);
+module_platform_driver(hecubafb_driver);
 
 MODULE_DESCRIPTION("fbdev driver for Hecuba/Apollo controller");
 MODULE_AUTHOR("Jaya Kumar");
diff --git a/drivers/video/hgafb.c b/drivers/video/hgafb.c
index 1e9e2d819d1f..5ff9fe2116a4 100644
--- a/drivers/video/hgafb.c
+++ b/drivers/video/hgafb.c
@@ -586,8 +586,7 @@ static int hgafb_probe(struct platform_device *pdev)
 		return -EINVAL;
 	}
 
-        printk(KERN_INFO "fb%d: %s frame buffer device\n",
-               info->node, info->fix.id);
+	fb_info(info, "%s frame buffer device\n", info->fix.id);
 	platform_set_drvdata(pdev, info);
 	return 0;
 }
diff --git a/drivers/video/hitfb.c b/drivers/video/hitfb.c
index c2414d6ab646..a648d5186c6e 100644
--- a/drivers/video/hitfb.c
+++ b/drivers/video/hitfb.c
@@ -405,8 +405,7 @@ static int hitfb_probe(struct platform_device *dev)
 
 	platform_set_drvdata(dev, info);
 
-	printk(KERN_INFO "fb%d: %s frame buffer device\n",
-	       info->node, info->fix.id);
+	fb_info(info, "%s frame buffer device\n", info->fix.id);
 
 	return 0;
 
diff --git a/drivers/video/hpfb.c b/drivers/video/hpfb.c
index b802f93cef5d..a1b7e5fa9b09 100644
--- a/drivers/video/hpfb.c
+++ b/drivers/video/hpfb.c
@@ -298,8 +298,7 @@ static int hpfb_init_one(unsigned long phys_base, unsigned long virt_base)
 	if (ret < 0)
 		goto dealloc_cmap;
 
-	printk(KERN_INFO "fb%d: %s frame buffer device\n",
-	       fb_info.node, fb_info.fix.id);
+	fb_info(&fb_info, "%s frame buffer device\n", fb_info.fix.id);
 
 	return 0;
 
diff --git a/drivers/video/hyperv_fb.c b/drivers/video/hyperv_fb.c
index 8ac99b87c07e..130708f96430 100644
--- a/drivers/video/hyperv_fb.c
+++ b/drivers/video/hyperv_fb.c
@@ -575,6 +575,10 @@ static int hvfb_setcolreg(unsigned regno, unsigned red, unsigned green,
 	return 0;
 }
 
+static int hvfb_blank(int blank, struct fb_info *info)
+{
+	return 1;	/* get fb_blank to set the colormap to all black */
+}
 
 static struct fb_ops hvfb_ops = {
 	.owner = THIS_MODULE,
@@ -584,6 +588,7 @@ static struct fb_ops hvfb_ops = {
 	.fb_fillrect = cfb_fillrect,
 	.fb_copyarea = cfb_copyarea,
 	.fb_imageblit = cfb_imageblit,
+	.fb_blank = hvfb_blank,
 };
 
 
@@ -795,12 +800,21 @@ static int hvfb_remove(struct hv_device *hdev)
 }
 
 
+static DEFINE_PCI_DEVICE_TABLE(pci_stub_id_table) = {
+	{
+		.vendor      = PCI_VENDOR_ID_MICROSOFT,
+		.device      = PCI_DEVICE_ID_HYPERV_VIDEO,
+	},
+	{ /* end of list */ }
+};
+
 static const struct hv_vmbus_device_id id_table[] = {
 	/* Synthetic Video Device GUID */
 	{HV_SYNTHVID_GUID},
 	{}
 };
 
+MODULE_DEVICE_TABLE(pci, pci_stub_id_table);
 MODULE_DEVICE_TABLE(vmbus, id_table);
 
 static struct hv_driver hvfb_drv = {
@@ -810,14 +824,43 @@ static struct hv_driver hvfb_drv = {
 	.remove = hvfb_remove,
 };
 
+static int hvfb_pci_stub_probe(struct pci_dev *pdev,
+			       const struct pci_device_id *ent)
+{
+	return 0;
+}
+
+static void hvfb_pci_stub_remove(struct pci_dev *pdev)
+{
+}
+
+static struct pci_driver hvfb_pci_stub_driver = {
+	.name =		KBUILD_MODNAME,
+	.id_table =	pci_stub_id_table,
+	.probe =	hvfb_pci_stub_probe,
+	.remove =	hvfb_pci_stub_remove,
+};
 
 static int __init hvfb_drv_init(void)
 {
-	return vmbus_driver_register(&hvfb_drv);
+	int ret;
+
+	ret = vmbus_driver_register(&hvfb_drv);
+	if (ret != 0)
+		return ret;
+
+	ret = pci_register_driver(&hvfb_pci_stub_driver);
+	if (ret != 0) {
+		vmbus_driver_unregister(&hvfb_drv);
+		return ret;
+	}
+
+	return 0;
 }
 
 static void __exit hvfb_drv_exit(void)
 {
+	pci_unregister_driver(&hvfb_pci_stub_driver);
 	vmbus_driver_unregister(&hvfb_drv);
 }
 
diff --git a/drivers/video/i740fb.c b/drivers/video/i740fb.c
index 6c4838818950..ca7c9df193b0 100644
--- a/drivers/video/i740fb.c
+++ b/drivers/video/i740fb.c
@@ -203,8 +203,7 @@ static int i740fb_release(struct fb_info *info, int user)
 
 	mutex_lock(&(par->open_lock));
 	if (par->ref_count == 0) {
-		printk(KERN_ERR "fb%d: release called with zero refcount\n",
-			info->node);
+		fb_err(info, "release called with zero refcount\n");
 		mutex_unlock(&(par->open_lock));
 		return -EINVAL;
 	}
@@ -1067,7 +1066,7 @@ static int i740fb_probe(struct pci_dev *dev, const struct pci_device_id *ent)
 	par->has_sgram = !((tmp & DRAM_RAS_TIMING) ||
 			   (tmp & DRAM_RAS_PRECHARGE));
 
-	printk(KERN_INFO "fb%d: Intel740 on %s, %ld KB %s\n", info->node,
+	fb_info(info, "Intel740 on %s, %ld KB %s\n",
 		pci_name(dev), info->screen_size >> 10,
 		par->has_sgram ? "SGRAM" : "SDRAM");
 
@@ -1143,8 +1142,7 @@ static int i740fb_probe(struct pci_dev *dev, const struct pci_device_id *ent)
 		goto err_reg_framebuffer;
 	}
 
-	printk(KERN_INFO "fb%d: %s frame buffer device\n",
-		info->node, info->fix.id);
+	fb_info(info, "%s frame buffer device\n", info->fix.id);
 	pci_set_drvdata(dev, info);
 #ifdef CONFIG_MTRR
 	if (mtrr) {
@@ -1194,7 +1192,6 @@ static void i740fb_remove(struct pci_dev *dev)
 		pci_iounmap(dev, info->screen_base);
 		pci_release_regions(dev);
 /*		pci_disable_device(dev); */
-		pci_set_drvdata(dev, NULL);
 		framebuffer_release(info);
 	}
 }
diff --git a/drivers/video/i810/i810_main.c b/drivers/video/i810/i810_main.c
index 4ce3438ade6f..038192ac7369 100644
--- a/drivers/video/i810/i810_main.c
+++ b/drivers/video/i810/i810_main.c
@@ -2129,7 +2129,6 @@ static void __exit i810fb_remove_pci(struct pci_dev *dev)
 
 	unregister_framebuffer(info);  
 	i810fb_release_resource(info, par);
-	pci_set_drvdata(dev, NULL);
 	printk("cleanup_module:  unloaded i810 framebuffer device\n");
 }                                                	
 
diff --git a/drivers/video/igafb.c b/drivers/video/igafb.c
index 79cbfa7d1a9b..486f18897414 100644
--- a/drivers/video/igafb.c
+++ b/drivers/video/igafb.c
@@ -360,9 +360,8 @@ static int __init iga_init(struct fb_info *info, struct iga_par *par)
 	if (register_framebuffer(info) < 0)
 		return 0;
 
-	printk("fb%d: %s frame buffer device at 0x%08lx [%dMB VRAM]\n",
-	       info->node, info->fix.id, 
-	       par->frame_buffer_phys, info->fix.smem_len >> 20);
+	fb_info(info, "%s frame buffer device at 0x%08lx [%dMB VRAM]\n",
+		info->fix.id, par->frame_buffer_phys, info->fix.smem_len >> 20);
 
 	iga_blank_border(par); 
 	return 1;
diff --git a/drivers/video/imsttfb.c b/drivers/video/imsttfb.c
index d5220cc90e93..aae10ce74f14 100644
--- a/drivers/video/imsttfb.c
+++ b/drivers/video/imsttfb.c
@@ -1461,8 +1461,8 @@ static void init_imstt(struct fb_info *info)
 	}
 
 	tmp = (read_reg_le32(par->dc_regs, SSTATUS) & 0x0f00) >> 8;
-	printk("fb%u: %s frame buffer; %uMB vram; chip version %u\n",
-		info->node, info->fix.id, info->fix.smem_len >> 20, tmp);
+	fb_info(info, "%s frame buffer; %uMB vram; chip version %u\n",
+		info->fix.id, info->fix.smem_len >> 20, tmp);
 }
 
 static int imsttfb_probe(struct pci_dev *pdev, const struct pci_device_id *ent)
diff --git a/drivers/video/imxfb.c b/drivers/video/imxfb.c
index 38733ac2b698..44ee678481d5 100644
--- a/drivers/video/imxfb.c
+++ b/drivers/video/imxfb.c
@@ -755,7 +755,7 @@ static int imxfb_resume(struct platform_device *dev)
 
 static int imxfb_init_fbinfo(struct platform_device *pdev)
 {
-	struct imx_fb_platform_data *pdata = pdev->dev.platform_data;
+	struct imx_fb_platform_data *pdata = dev_get_platdata(&pdev->dev);
 	struct fb_info *info = dev_get_drvdata(&pdev->dev);
 	struct imxfb_info *fbi = info->par;
 	struct device_node *np;
@@ -877,7 +877,7 @@ static int imxfb_probe(struct platform_device *pdev)
 	if (!res)
 		return -ENODEV;
 
-	pdata = pdev->dev.platform_data;
+	pdata = dev_get_platdata(&pdev->dev);
 
 	info = framebuffer_alloc(sizeof(struct imxfb_info), &pdev->dev);
 	if (!info)
@@ -1066,7 +1066,7 @@ static int imxfb_remove(struct platform_device *pdev)
 #endif
 	unregister_framebuffer(info);
 
-	pdata = pdev->dev.platform_data;
+	pdata = dev_get_platdata(&pdev->dev);
 	if (pdata && pdata->exit)
 		pdata->exit(fbi->pdev);
 
diff --git a/drivers/video/intelfb/intelfbdrv.c b/drivers/video/intelfb/intelfbdrv.c
index 8209e46c5d28..b847d530471a 100644
--- a/drivers/video/intelfb/intelfbdrv.c
+++ b/drivers/video/intelfb/intelfbdrv.c
@@ -931,8 +931,6 @@ static void intelfb_pci_unregister(struct pci_dev *pdev)
 		return;
 
 	cleanup(dinfo);
-
-	pci_set_drvdata(pdev, NULL);
 }
 
 /***************************************************************
diff --git a/drivers/video/jz4740_fb.c b/drivers/video/jz4740_fb.c
index 2c49112fdd6c..87790e9644d0 100644
--- a/drivers/video/jz4740_fb.c
+++ b/drivers/video/jz4740_fb.c
@@ -99,9 +99,9 @@
 #define JZ_LCD_CTRL_BPP_15_16		0x4
 #define JZ_LCD_CTRL_BPP_18_24		0x5
 
-#define JZ_LCD_CMD_SOF_IRQ BIT(15)
-#define JZ_LCD_CMD_EOF_IRQ BIT(16)
-#define JZ_LCD_CMD_ENABLE_PAL BIT(12)
+#define JZ_LCD_CMD_SOF_IRQ BIT(31)
+#define JZ_LCD_CMD_EOF_IRQ BIT(30)
+#define JZ_LCD_CMD_ENABLE_PAL BIT(28)
 
 #define JZ_LCD_SYNC_MASK 0x3ff
 
@@ -471,7 +471,7 @@ static int jzfb_set_par(struct fb_info *info)
 	writel(ctrl, jzfb->base + JZ_REG_LCD_CTRL);
 
 	if (!jzfb->is_enabled)
-		clk_disable(jzfb->ldclk);
+		clk_disable_unprepare(jzfb->ldclk);
 
 	mutex_unlock(&jzfb->lock);
 
@@ -485,7 +485,7 @@ static void jzfb_enable(struct jzfb *jzfb)
 {
 	uint32_t ctrl;
 
-	clk_enable(jzfb->ldclk);
+	clk_prepare_enable(jzfb->ldclk);
 
 	jz_gpio_bulk_resume(jz_lcd_ctrl_pins, jzfb_num_ctrl_pins(jzfb));
 	jz_gpio_bulk_resume(jz_lcd_data_pins, jzfb_num_data_pins(jzfb));
@@ -514,7 +514,7 @@ static void jzfb_disable(struct jzfb *jzfb)
 	jz_gpio_bulk_suspend(jz_lcd_ctrl_pins, jzfb_num_ctrl_pins(jzfb));
 	jz_gpio_bulk_suspend(jz_lcd_data_pins, jzfb_num_data_pins(jzfb));
 
-	clk_disable(jzfb->ldclk);
+	clk_disable_unprepare(jzfb->ldclk);
 }
 
 static int jzfb_blank(int blank_mode, struct fb_info *info)
@@ -693,7 +693,7 @@ static int jzfb_probe(struct platform_device *pdev)
 
 	fb_alloc_cmap(&fb->cmap, 256, 0);
 
-	clk_enable(jzfb->ldclk);
+	clk_prepare_enable(jzfb->ldclk);
 	jzfb->is_enabled = 1;
 
 	writel(jzfb->framedesc->next, jzfb->base + JZ_REG_LCD_DA0);
@@ -763,7 +763,7 @@ static int jzfb_suspend(struct device *dev)
 static int jzfb_resume(struct device *dev)
 {
 	struct jzfb *jzfb = dev_get_drvdata(dev);
-	clk_enable(jzfb->ldclk);
+	clk_prepare_enable(jzfb->ldclk);
 
 	mutex_lock(&jzfb->lock);
 	if (jzfb->is_enabled)
@@ -798,18 +798,7 @@ static struct platform_driver jzfb_driver = {
 		.pm = JZFB_PM_OPS,
 	},
 };
-
-static int __init jzfb_init(void)
-{
-	return platform_driver_register(&jzfb_driver);
-}
-module_init(jzfb_init);
-
-static void __exit jzfb_exit(void)
-{
-	platform_driver_unregister(&jzfb_driver);
-}
-module_exit(jzfb_exit);
+module_platform_driver(jzfb_driver);
 
 MODULE_LICENSE("GPL");
 MODULE_AUTHOR("Lars-Peter Clausen <lars@metafoo.de>");
diff --git a/drivers/video/kyro/fbdev.c b/drivers/video/kyro/fbdev.c
index 6157f74ac600..50c857477e4f 100644
--- a/drivers/video/kyro/fbdev.c
+++ b/drivers/video/kyro/fbdev.c
@@ -623,7 +623,6 @@ static int kyrofb_ioctl(struct fb_info *info,
 				"command instead.\n");
 			return -EINVAL;
 		}
-		break;
 	case KYRO_IOCTL_UVSTRIDE:
 		if (copy_to_user(argp, &deviceInfo.ulOverlayUVStride, sizeof(unsigned long)))
 			return -EFAULT;
@@ -736,10 +735,10 @@ static int kyrofb_probe(struct pci_dev *pdev, const struct pci_device_id *ent)
 	if (register_framebuffer(info) < 0)
 		goto out_unmap;
 
-	printk("fb%d: %s frame buffer device, at %dx%d@%d using %ldk/%ldk of VRAM\n",
-	       info->node, info->fix.id, info->var.xres,
-	       info->var.yres, info->var.bits_per_pixel, size >> 10,
-	       (unsigned long)info->fix.smem_len >> 10);
+	fb_info(info, "%s frame buffer device, at %dx%d@%d using %ldk/%ldk of VRAM\n",
+		info->fix.id,
+		info->var.xres, info->var.yres, info->var.bits_per_pixel,
+		size >> 10, (unsigned long)info->fix.smem_len >> 10);
 
 	pci_set_drvdata(pdev, info);
 
@@ -779,7 +778,6 @@ static void kyrofb_remove(struct pci_dev *pdev)
 #endif
 
 	unregister_framebuffer(info);
-	pci_set_drvdata(pdev, NULL);
 	framebuffer_release(info);
 }
 
diff --git a/drivers/video/leo.c b/drivers/video/leo.c
index b17f5009a436..2c7f7d479fe2 100644
--- a/drivers/video/leo.c
+++ b/drivers/video/leo.c
@@ -469,7 +469,7 @@ static void leo_wid_put(struct fb_info *info, struct fb_wid_list *wl)
 
 		default:
 			continue;
-		};
+		}
 		sbus_writel(0x5800 + j, &lx_krn->krn_type);
 		sbus_writel(wi->wi_values[0], &lx_krn->krn_value);
 	}
@@ -648,8 +648,6 @@ static int leo_remove(struct platform_device *op)
 
 	framebuffer_release(info);
 
-	dev_set_drvdata(&op->dev, NULL);
-
 	return 0;
 }
 
diff --git a/drivers/video/macfb.c b/drivers/video/macfb.c
index fe01add3700e..5bd2eb8d4f39 100644
--- a/drivers/video/macfb.c
+++ b/drivers/video/macfb.c
@@ -913,8 +913,7 @@ static int __init macfb_init(void)
 	if (err)
 		goto fail_dealloc;
 
-	pr_info("fb%d: %s frame buffer device\n",
-	        fb_info.node, fb_info.fix.id);
+	fb_info(&fb_info, "%s frame buffer device\n", fb_info.fix.id);
 
 	return 0;
 
diff --git a/drivers/video/matrox/matroxfb_DAC1064.c b/drivers/video/matrox/matroxfb_DAC1064.c
index 1717623aabc0..a01147fdf270 100644
--- a/drivers/video/matrox/matroxfb_DAC1064.c
+++ b/drivers/video/matrox/matroxfb_DAC1064.c
@@ -494,7 +494,7 @@ static int m1064_compute(void* out, struct my_timming* m) {
 			if (inDAC1064(minfo, M1064_XPIXPLLSTAT) & 0x40)
 				break;
 			udelay(10);
-		};
+		}
 
 		CRITEND
 
@@ -639,7 +639,7 @@ static void MGAG100_progPixClock(const struct matrox_fb_info *minfo, int flags,
 		if (inDAC1064(minfo, M1064_XPIXPLLSTAT) & 0x40)
 			break;
 		udelay(10);
-	};
+	}
 	if (!clk)
 		printk(KERN_ERR "matroxfb: Pixel PLL%c not locked after usual time\n", (reg-M1064_XPIXPLLAM-2)/4 + 'A');
 	selClk = inDAC1064(minfo, M1064_XPIXCLKCTRL) & ~M1064_XPIXCLKCTRL_SRC_MASK;
diff --git a/drivers/video/matrox/matroxfb_Ti3026.c b/drivers/video/matrox/matroxfb_Ti3026.c
index 9a44cec394b5..195ad7cac1ba 100644
--- a/drivers/video/matrox/matroxfb_Ti3026.c
+++ b/drivers/video/matrox/matroxfb_Ti3026.c
@@ -473,7 +473,7 @@ static void ti3026_setMCLK(struct matrox_fb_info *minfo, int fout)
 		if (inTi3026(minfo, TVP3026_XPIXPLLDATA) & 0x40)
 			break;
 		udelay(10);
-	};
+	}
 	if (!tmout)
 		printk(KERN_ERR "matroxfb: Temporary pixel PLL not locked after 5 secs\n");
 
diff --git a/drivers/video/matrox/matroxfb_base.c b/drivers/video/matrox/matroxfb_base.c
index 245652911650..87c64ff4546c 100644
--- a/drivers/video/matrox/matroxfb_base.c
+++ b/drivers/video/matrox/matroxfb_base.c
@@ -1893,14 +1893,12 @@ static int initMatrox2(struct matrox_fb_info *minfo, struct board *b)
 	if (register_framebuffer(&minfo->fbcon) < 0) {
 		goto failVideoIO;
 	}
-	printk("fb%d: %s frame buffer device\n",
-	       minfo->fbcon.node, minfo->fbcon.fix.id);
+	fb_info(&minfo->fbcon, "%s frame buffer device\n", minfo->fbcon.fix.id);
 
 	/* there is no console on this fb... but we have to initialize hardware
 	 * until someone tells me what is proper thing to do */
 	if (!minfo->initialized) {
-		printk(KERN_INFO "fb%d: initializing hardware\n",
-		       minfo->fbcon.node);
+		fb_info(&minfo->fbcon, "initializing hardware\n");
 		/* We have to use FB_ACTIVATE_FORCE, as we had to put vesafb_defined to the fbcon.var
 		 * already before, so register_framebuffer works correctly. */
 		vesafb_defined.activate |= FB_ACTIVATE_FORCE;
diff --git a/drivers/video/matrox/matroxfb_maven.c b/drivers/video/matrox/matroxfb_maven.c
index fd2897455696..ee41a0f276b2 100644
--- a/drivers/video/matrox/matroxfb_maven.c
+++ b/drivers/video/matrox/matroxfb_maven.c
@@ -1295,19 +1295,7 @@ static struct i2c_driver maven_driver={
 	.id_table	= maven_id,
 };
 
-static int __init matroxfb_maven_init(void)
-{
-	return i2c_add_driver(&maven_driver);
-}
-
-static void __exit matroxfb_maven_exit(void)
-{
-	i2c_del_driver(&maven_driver);
-}
-
+module_i2c_driver(maven_driver);
 MODULE_AUTHOR("(c) 1999-2002 Petr Vandrovec <vandrove@vc.cvut.cz>");
 MODULE_DESCRIPTION("Matrox G200/G400 Matrox MGA-TVO driver");
 MODULE_LICENSE("GPL");
-module_init(matroxfb_maven_init);
-module_exit(matroxfb_maven_exit);
-/* we do not have __setup() yet */
diff --git a/drivers/video/mb862xx/mb862xxfbdrv.c b/drivers/video/mb862xx/mb862xxfbdrv.c
index 91c59c9fb082..0cd4c3318511 100644
--- a/drivers/video/mb862xx/mb862xxfbdrv.c
+++ b/drivers/video/mb862xx/mb862xxfbdrv.c
@@ -781,7 +781,6 @@ rel_reg:
 irqdisp:
 	irq_dispose_mapping(par->irq);
 fbrel:
-	dev_set_drvdata(dev, NULL);
 	framebuffer_release(info);
 	return ret;
 }
@@ -814,7 +813,6 @@ static int of_platform_mb862xx_remove(struct platform_device *ofdev)
 	iounmap(par->mmio_base);
 	iounmap(par->fb_base);
 
-	dev_set_drvdata(&ofdev->dev, NULL);
 	release_mem_region(par->res->start, res_size);
 	framebuffer_release(fbi);
 	return 0;
@@ -1157,7 +1155,6 @@ static void mb862xx_pci_remove(struct pci_dev *pdev)
 
 	device_remove_file(&pdev->dev, &dev_attr_dispregs);
 
-	pci_set_drvdata(pdev, NULL);
 	unregister_framebuffer(fbi);
 	fb_dealloc_cmap(&fbi->cmap);
 
diff --git a/drivers/video/mbx/mbxfb.c b/drivers/video/mbx/mbxfb.c
index 0c1a874ffd2b..f0a5392f5fd3 100644
--- a/drivers/video/mbx/mbxfb.c
+++ b/drivers/video/mbx/mbxfb.c
@@ -890,7 +890,7 @@ static int mbxfb_probe(struct platform_device *dev)
 
 	dev_dbg(&dev->dev, "mbxfb_probe\n");
 
-	pdata = dev->dev.platform_data;
+	pdata = dev_get_platdata(&dev->dev);
 	if (!pdata) {
 		dev_err(&dev->dev, "platform data is required\n");
 		return -EINVAL;
@@ -976,7 +976,7 @@ static int mbxfb_probe(struct platform_device *dev)
 
 	platform_set_drvdata(dev, fbi);
 
-	printk(KERN_INFO "fb%d: mbx frame buffer device\n", fbi->node);
+	fb_info(fbi, "mbx frame buffer device\n");
 
 	if (mfbi->platform_probe)
 		mfbi->platform_probe(fbi);
diff --git a/drivers/video/metronomefb.c b/drivers/video/metronomefb.c
index f30150d71be9..195cc2db4c2c 100644
--- a/drivers/video/metronomefb.c
+++ b/drivers/video/metronomefb.c
@@ -690,7 +690,8 @@ static int metronomefb_probe(struct platform_device *dev)
 		goto err_csum_table;
 	}
 
-	if (board->setup_irq(info))
+	retval = board->setup_irq(info);
+	if (retval)
 		goto err_csum_table;
 
 	retval = metronome_init_regs(par);
@@ -769,23 +770,11 @@ static struct platform_driver metronomefb_driver = {
 		.name	= "metronomefb",
 	},
 };
-
-static int __init metronomefb_init(void)
-{
-	return platform_driver_register(&metronomefb_driver);
-}
-
-static void __exit metronomefb_exit(void)
-{
-	platform_driver_unregister(&metronomefb_driver);
-}
+module_platform_driver(metronomefb_driver);
 
 module_param(user_wfm_size, uint, 0);
 MODULE_PARM_DESC(user_wfm_size, "Set custom waveform size");
 
-module_init(metronomefb_init);
-module_exit(metronomefb_exit);
-
 MODULE_DESCRIPTION("fbdev driver for Metronome controller");
 MODULE_AUTHOR("Jaya Kumar");
 MODULE_LICENSE("GPL");
diff --git a/drivers/video/mmp/fb/mmpfb.c b/drivers/video/mmp/fb/mmpfb.c
index 4ab95b8daed3..7ab31eb76a8c 100644
--- a/drivers/video/mmp/fb/mmpfb.c
+++ b/drivers/video/mmp/fb/mmpfb.c
@@ -392,12 +392,29 @@ static int var_update(struct fb_info *info)
 	return 0;
 }
 
+static void mmpfb_set_win(struct fb_info *info)
+{
+	struct mmpfb_info *fbi = info->par;
+	struct fb_var_screeninfo *var = &info->var;
+	struct mmp_win win;
+	u32 stride;
+
+	memset(&win, 0, sizeof(win));
+	win.xsrc = win.xdst = fbi->mode.xres;
+	win.ysrc = win.ydst = fbi->mode.yres;
+	win.pix_fmt = fbi->pix_fmt;
+	stride = pixfmt_to_stride(win.pix_fmt);
+	win.pitch[0] = var->xres_virtual * stride;
+	win.pitch[1] = win.pitch[2] =
+		(stride == 1) ? (var->xres_virtual >> 1) : 0;
+	mmp_overlay_set_win(fbi->overlay, &win);
+}
+
 static int mmpfb_set_par(struct fb_info *info)
 {
 	struct mmpfb_info *fbi = info->par;
 	struct fb_var_screeninfo *var = &info->var;
 	struct mmp_addr addr;
-	struct mmp_win win;
 	struct mmp_mode mode;
 	int ret;
 
@@ -409,11 +426,8 @@ static int mmpfb_set_par(struct fb_info *info)
 	fbmode_to_mmpmode(&mode, &fbi->mode, fbi->output_fmt);
 	mmp_path_set_mode(fbi->path, &mode);
 
-	memset(&win, 0, sizeof(win));
-	win.xsrc = win.xdst = fbi->mode.xres;
-	win.ysrc = win.ydst = fbi->mode.yres;
-	win.pix_fmt = fbi->pix_fmt;
-	mmp_overlay_set_win(fbi->overlay, &win);
+	/* set window related info */
+	mmpfb_set_win(info);
 
 	/* set address always */
 	memset(&addr, 0, sizeof(addr));
@@ -427,16 +441,12 @@ static int mmpfb_set_par(struct fb_info *info)
 static void mmpfb_power(struct mmpfb_info *fbi, int power)
 {
 	struct mmp_addr addr;
-	struct mmp_win win;
 	struct fb_var_screeninfo *var = &fbi->fb_info->var;
 
 	/* for power on, always set address/window again */
 	if (power) {
-		memset(&win, 0, sizeof(win));
-		win.xsrc = win.xdst = fbi->mode.xres;
-		win.ysrc = win.ydst = fbi->mode.yres;
-		win.pix_fmt = fbi->pix_fmt;
-		mmp_overlay_set_win(fbi->overlay, &win);
+		/* set window related info */
+		mmpfb_set_win(fbi->fb_info);
 
 		/* set address always */
 		memset(&addr, 0, sizeof(addr));
diff --git a/drivers/video/mmp/hw/mmp_ctrl.c b/drivers/video/mmp/hw/mmp_ctrl.c
index 6ac755270ab4..8621a9f2bdcc 100644
--- a/drivers/video/mmp/hw/mmp_ctrl.c
+++ b/drivers/video/mmp/hw/mmp_ctrl.c
@@ -53,15 +53,14 @@ static irqreturn_t ctrl_handle_irq(int irq, void *dev_id)
 		tmp = readl_relaxed(ctrl->reg_base + SPU_IRQ_ISR);
 		if (tmp & isr)
 			writel_relaxed(~isr, ctrl->reg_base + SPU_IRQ_ISR);
-	} while ((isr = readl(ctrl->reg_base + SPU_IRQ_ISR)) & imask);
+	} while ((isr = readl_relaxed(ctrl->reg_base + SPU_IRQ_ISR)) & imask);
 
 	return IRQ_HANDLED;
 }
 
 static u32 fmt_to_reg(struct mmp_overlay *overlay, int pix_fmt)
 {
-	u32 link_config = path_to_path_plat(overlay->path)->link_config;
-	u32 rbswap, uvswap = 0, yuvswap = 0,
+	u32 rbswap = 0, uvswap = 0, yuvswap = 0,
 		csc_en = 0, val = 0,
 		vid = overlay_is_vid(overlay);
 
@@ -71,27 +70,23 @@ static u32 fmt_to_reg(struct mmp_overlay *overlay, int pix_fmt)
 	case PIXFMT_RGB888PACK:
 	case PIXFMT_RGB888UNPACK:
 	case PIXFMT_RGBA888:
-		rbswap = !(link_config & 0x1);
+		rbswap = 1;
 		break;
 	case PIXFMT_VYUY:
 	case PIXFMT_YVU422P:
 	case PIXFMT_YVU420P:
-		rbswap = link_config & 0x1;
 		uvswap = 1;
 		break;
 	case PIXFMT_YUYV:
-		rbswap = link_config & 0x1;
 		yuvswap = 1;
 		break;
 	default:
-		rbswap = link_config & 0x1;
 		break;
 	}
 
 	switch (pix_fmt) {
 	case PIXFMT_RGB565:
 	case PIXFMT_BGR565:
-		val = 0;
 		break;
 	case PIXFMT_RGB1555:
 	case PIXFMT_BGR1555:
@@ -147,17 +142,27 @@ static void dmafetch_set_fmt(struct mmp_overlay *overlay)
 static void overlay_set_win(struct mmp_overlay *overlay, struct mmp_win *win)
 {
 	struct lcd_regs *regs = path_regs(overlay->path);
-	u32 pitch;
 
 	/* assert win supported */
 	memcpy(&overlay->win, win, sizeof(struct mmp_win));
 
 	mutex_lock(&overlay->access_ok);
-	pitch = win->xsrc * pixfmt_to_stride(win->pix_fmt);
-	writel_relaxed(pitch, &regs->g_pitch);
-	writel_relaxed((win->ysrc << 16) | win->xsrc, &regs->g_size);
-	writel_relaxed((win->ydst << 16) | win->xdst, &regs->g_size_z);
-	writel_relaxed(0, &regs->g_start);
+
+	if (overlay_is_vid(overlay)) {
+		writel_relaxed(win->pitch[0], &regs->v_pitch_yc);
+		writel_relaxed(win->pitch[2] << 16 |
+				win->pitch[1], &regs->v_pitch_uv);
+
+		writel_relaxed((win->ysrc << 16) | win->xsrc, &regs->v_size);
+		writel_relaxed((win->ydst << 16) | win->xdst, &regs->v_size_z);
+		writel_relaxed(win->ypos << 16 | win->xpos, &regs->v_start);
+	} else {
+		writel_relaxed(win->pitch[0], &regs->g_pitch);
+
+		writel_relaxed((win->ysrc << 16) | win->xsrc, &regs->g_size);
+		writel_relaxed((win->ydst << 16) | win->xdst, &regs->g_size_z);
+		writel_relaxed(win->ypos << 16 | win->xpos, &regs->g_start);
+	}
 
 	dmafetch_set_fmt(overlay);
 	mutex_unlock(&overlay->access_ok);
@@ -239,7 +244,13 @@ static int overlay_set_addr(struct mmp_overlay *overlay, struct mmp_addr *addr)
 
 	/* FIXME: assert addr supported */
 	memcpy(&overlay->addr, addr, sizeof(struct mmp_addr));
-	writel(addr->phys[0], &regs->g_0);
+
+	if (overlay_is_vid(overlay)) {
+		writel_relaxed(addr->phys[0], &regs->v_y0);
+		writel_relaxed(addr->phys[1], &regs->v_u0);
+		writel_relaxed(addr->phys[2], &regs->v_v0);
+	} else
+		writel_relaxed(addr->phys[0], &regs->g_0);
 
 	return overlay->addr.phys[0];
 }
@@ -248,7 +259,8 @@ static void path_set_mode(struct mmp_path *path, struct mmp_mode *mode)
 {
 	struct lcd_regs *regs = path_regs(path);
 	u32 total_x, total_y, vsync_ctrl, tmp, sclk_src, sclk_div,
-		link_config = path_to_path_plat(path)->link_config;
+		link_config = path_to_path_plat(path)->link_config,
+		dsi_rbswap = path_to_path_plat(path)->link_config;
 
 	/* FIXME: assert videomode supported */
 	memcpy(&path->mode, mode, sizeof(struct mmp_mode));
@@ -263,6 +275,12 @@ static void path_set_mode(struct mmp_path *path, struct mmp_mode *mode)
 	tmp |= CFG_DUMB_ENA(1);
 	writel_relaxed(tmp, ctrl_regs(path) + intf_ctrl(path->id));
 
+	/* interface rb_swap setting */
+	tmp = readl_relaxed(ctrl_regs(path) + intf_rbswap_ctrl(path->id)) &
+		(~(CFG_INTFRBSWAP_MASK));
+	tmp |= dsi_rbswap & CFG_INTFRBSWAP_MASK;
+	writel_relaxed(tmp, ctrl_regs(path) + intf_rbswap_ctrl(path->id));
+
 	writel_relaxed((mode->yres << 16) | mode->xres, &regs->screen_active);
 	writel_relaxed((mode->left_margin << 16) | mode->right_margin,
 		&regs->screen_h_porch);
@@ -370,20 +388,12 @@ static void path_set_default(struct mmp_path *path)
 	 * bus arbiter for faster read if not tv path;
 	 * 2.enable horizontal smooth filter;
 	 */
-	if (PATH_PN == path->id) {
-		mask = CFG_GRA_HSMOOTH_MASK | CFG_DMA_HSMOOTH_MASK
-			| CFG_ARBFAST_ENA(1);
-		tmp = readl_relaxed(ctrl_regs(path) + dma_ctrl(0, path->id));
-		tmp |= mask;
-		writel_relaxed(tmp, ctrl_regs(path) + dma_ctrl(0, path->id));
-	} else if (PATH_TV == path->id) {
-		mask = CFG_GRA_HSMOOTH_MASK | CFG_DMA_HSMOOTH_MASK
-			| CFG_ARBFAST_ENA(1);
-		tmp = readl_relaxed(ctrl_regs(path) + dma_ctrl(0, path->id));
-		tmp &= ~mask;
-		tmp |= CFG_GRA_HSMOOTH_MASK | CFG_DMA_HSMOOTH_MASK;
-		writel_relaxed(tmp, ctrl_regs(path) + dma_ctrl(0, path->id));
-	}
+	mask = CFG_GRA_HSMOOTH_MASK | CFG_DMA_HSMOOTH_MASK | CFG_ARBFAST_ENA(1);
+	tmp = readl_relaxed(ctrl_regs(path) + dma_ctrl(0, path->id));
+	tmp |= mask;
+	if (PATH_TV == path->id)
+		tmp &= ~CFG_ARBFAST_ENA(1);
+	writel_relaxed(tmp, ctrl_regs(path) + dma_ctrl(0, path->id));
 }
 
 static int path_init(struct mmphw_path_plat *path_plat,
@@ -419,6 +429,7 @@ static int path_init(struct mmphw_path_plat *path_plat,
 	path_plat->path = path;
 	path_plat->path_config = config->path_config;
 	path_plat->link_config = config->link_config;
+	path_plat->dsi_rbswap = config->dsi_rbswap;
 	path_set_default(path);
 
 	kfree(path_info);
diff --git a/drivers/video/mmp/hw/mmp_ctrl.h b/drivers/video/mmp/hw/mmp_ctrl.h
index edd2002b0e99..53301cfdb1ae 100644
--- a/drivers/video/mmp/hw/mmp_ctrl.h
+++ b/drivers/video/mmp/hw/mmp_ctrl.h
@@ -163,6 +163,8 @@ struct lcd_regs {
 
 #define LCD_SCLK(path) ((PATH_PN == path->id) ? LCD_CFG_SCLK_DIV :\
 	((PATH_TV == path->id) ? LCD_TCLK_DIV : LCD_PN2_SCLK_DIV))
+#define intf_rbswap_ctrl(id)	((id) ? (((id) & 1) ? LCD_TVIF_CTRL : \
+				PN2_IOPAD_CONTROL) : LCD_TOP_CTRL)
 
 /* dither configure */
 #ifdef CONFIG_CPU_PXA988
@@ -615,6 +617,8 @@ struct lcd_regs {
 #define LCD_SPU_DUMB_CTRL			0x01B8
 #define	 CFG_DUMBMODE(mode)			((mode)<<28)
 #define	 CFG_DUMBMODE_MASK			0xF0000000
+#define	 CFG_INTFRBSWAP(mode)			((mode)<<24)
+#define	 CFG_INTFRBSWAP_MASK			0x0F000000
 #define	 CFG_LCDGPIO_O(data)			((data)<<20)
 #define	 CFG_LCDGPIO_O_MASK			0x0FF00000
 #define	 CFG_LCDGPIO_ENA(gpio)			((gpio)<<12)
@@ -1427,6 +1431,7 @@ struct mmphw_path_plat {
 	struct mmp_path *path;
 	u32 path_config;
 	u32 link_config;
+	u32 dsi_rbswap;
 };
 
 /* mmp ctrl describes mmp controller related info */
diff --git a/drivers/video/mx3fb.c b/drivers/video/mx3fb.c
index cfdb380ec81e..804f874d32d3 100644
--- a/drivers/video/mx3fb.c
+++ b/drivers/video/mx3fb.c
@@ -1354,7 +1354,7 @@ static struct fb_info *mx3fb_init_fbinfo(struct device *dev, struct fb_ops *ops)
 static int init_fb_chan(struct mx3fb_data *mx3fb, struct idmac_channel *ichan)
 {
 	struct device *dev = mx3fb->dev;
-	struct mx3fb_platform_data *mx3fb_pdata = dev->platform_data;
+	struct mx3fb_platform_data *mx3fb_pdata = dev_get_platdata(dev);
 	const char *name = mx3fb_pdata->name;
 	unsigned int irq;
 	struct fb_info *fbi;
@@ -1462,7 +1462,7 @@ static bool chan_filter(struct dma_chan *chan, void *arg)
 		return false;
 
 	dev = rq->mx3fb->dev;
-	mx3fb_pdata = dev->platform_data;
+	mx3fb_pdata = dev_get_platdata(dev);
 
 	return rq->id == chan->chan_id &&
 		mx3fb_pdata->dma_dev == chan->device->dev;
diff --git a/drivers/video/neofb.c b/drivers/video/neofb.c
index c172a5281f9e..44f99a60bb9b 100644
--- a/drivers/video/neofb.c
+++ b/drivers/video/neofb.c
@@ -2106,8 +2106,7 @@ static int neofb_probe(struct pci_dev *dev, const struct pci_device_id *id)
 	if (err < 0)
 		goto err_reg_fb;
 
-	printk(KERN_INFO "fb%d: %s frame buffer device\n",
-	       info->node, info->fix.id);
+	fb_info(info, "%s frame buffer device\n", info->fix.id);
 
 	/*
 	 * Our driver data
@@ -2148,12 +2147,6 @@ static void neofb_remove(struct pci_dev *dev)
 		fb_destroy_modedb(info->monspecs.modedb);
 		neo_unmap_mmio(info);
 		neo_free_fb_info(info);
-
-		/*
-		 * Ensure that the driver data is no longer
-		 * valid.
-		 */
-		pci_set_drvdata(dev, NULL);
 	}
 }
 
diff --git a/drivers/video/nuc900fb.c b/drivers/video/nuc900fb.c
index 796e5112ceee..478f9808dee4 100644
--- a/drivers/video/nuc900fb.c
+++ b/drivers/video/nuc900fb.c
@@ -91,7 +91,7 @@ static int nuc900fb_check_var(struct fb_var_screeninfo *var,
 			       struct fb_info *info)
 {
 	struct nuc900fb_info *fbi = info->par;
-	struct nuc900fb_mach_info *mach_info = fbi->dev->platform_data;
+	struct nuc900fb_mach_info *mach_info = dev_get_platdata(fbi->dev);
 	struct nuc900fb_display *display = NULL;
 	struct nuc900fb_display *default_display = mach_info->displays +
 						   mach_info->default_display;
@@ -358,7 +358,7 @@ static inline void modify_gpio(void __iomem *reg,
 static int nuc900fb_init_registers(struct fb_info *info)
 {
 	struct nuc900fb_info *fbi = info->par;
-	struct nuc900fb_mach_info *mach_info = fbi->dev->platform_data;
+	struct nuc900fb_mach_info *mach_info = dev_get_platdata(fbi->dev);
 	void __iomem *regs = fbi->io;
 
 	/*reset the display engine*/
@@ -512,7 +512,7 @@ static int nuc900fb_probe(struct platform_device *pdev)
 	int size;
 
 	dev_dbg(&pdev->dev, "devinit\n");
-	mach_info = pdev->dev.platform_data;
+	mach_info = dev_get_platdata(&pdev->dev);
 	if (mach_info == NULL) {
 		dev_err(&pdev->dev,
 			"no platform data for lcd, cannot attach\n");
@@ -647,8 +647,7 @@ static int nuc900fb_probe(struct platform_device *pdev)
 		goto free_cpufreq;
 	}
 
-	printk(KERN_INFO "fb%d: %s frame buffer device\n",
-		fbinfo->node, fbinfo->fix.id);
+	fb_info(fbinfo, "%s frame buffer device\n", fbinfo->fix.id);
 
 	return 0;
 
diff --git a/drivers/video/nvidia/nv_hw.c b/drivers/video/nvidia/nv_hw.c
index ed20a9871b33..81c80ac3c76f 100644
--- a/drivers/video/nvidia/nv_hw.c
+++ b/drivers/video/nvidia/nv_hw.c
@@ -1300,7 +1300,7 @@ void NVLoadStateExt(struct nvidia_par *par, RIVA_HW_STATE * state)
 					break;
 				default:
 					break;
-				};
+				}
 
 				NV_WR32(par->PGRAPH, 0x0b38, 0x2ffff800);
 				NV_WR32(par->PGRAPH, 0x0b3c, 0x00006000);
diff --git a/drivers/video/offb.c b/drivers/video/offb.c
index 0c4f34311eda..9dbea2223401 100644
--- a/drivers/video/offb.c
+++ b/drivers/video/offb.c
@@ -515,8 +515,7 @@ static void __init offb_init_fb(const char *name, const char *full_name,
 	if (register_framebuffer(info) < 0)
 		goto out_err;
 
-	printk(KERN_INFO "fb%d: Open Firmware frame buffer device on %s\n",
-	       info->node, full_name);
+	fb_info(info, "Open Firmware frame buffer device on %s\n", full_name);
 	return;
 
 out_err:
diff --git a/drivers/video/omap/hwa742.c b/drivers/video/omap/hwa742.c
index f349ee6f0cea..a4ee65b8f918 100644
--- a/drivers/video/omap/hwa742.c
+++ b/drivers/video/omap/hwa742.c
@@ -947,7 +947,7 @@ static int hwa742_init(struct omapfb_device *fbdev, int ext_mode,
 	hwa742.extif = fbdev->ext_if;
 	hwa742.int_ctrl = fbdev->int_ctrl;
 
-	omapfb_conf = fbdev->dev->platform_data;
+	omapfb_conf = dev_get_platdata(fbdev->dev);
 
 	hwa742.sys_ck = clk_get(NULL, "hwa_sys_ck");
 
diff --git a/drivers/video/omap/omapfb_main.c b/drivers/video/omap/omapfb_main.c
index d40612c31a98..e4fc6d9b5371 100644
--- a/drivers/video/omap/omapfb_main.c
+++ b/drivers/video/omap/omapfb_main.c
@@ -1602,7 +1602,7 @@ static int omapfb_find_ctrl(struct omapfb_device *fbdev)
 	char name[17];
 	int i;
 
-	conf = fbdev->dev->platform_data;
+	conf = dev_get_platdata(fbdev->dev);
 
 	fbdev->ctrl = NULL;
 
@@ -1674,7 +1674,7 @@ static int omapfb_do_probe(struct platform_device *pdev,
 		goto cleanup;
 	}
 
-	if (pdev->dev.platform_data == NULL) {
+	if (dev_get_platdata(&pdev->dev) == NULL) {
 		dev_err(&pdev->dev, "missing platform data\n");
 		r = -ENOENT;
 		goto cleanup;
diff --git a/drivers/video/omap2/displays-new/Kconfig b/drivers/video/omap2/displays-new/Kconfig
index 10b25e7cd878..e6cfc38160d3 100644
--- a/drivers/video/omap2/displays-new/Kconfig
+++ b/drivers/video/omap2/displays-new/Kconfig
@@ -57,6 +57,12 @@ config DISPLAY_PANEL_SHARP_LS037V7DW01
         help
           LCD Panel used in TI's SDP3430 and EVM boards
 
+config DISPLAY_PANEL_TPO_TD028TTEC1
+        tristate "TPO TD028TTEC1 LCD Panel"
+        depends on SPI
+        help
+          LCD panel used in Openmoko.
+
 config DISPLAY_PANEL_TPO_TD043MTEA1
         tristate "TPO TD043MTEA1 LCD Panel"
         depends on SPI
diff --git a/drivers/video/omap2/displays-new/Makefile b/drivers/video/omap2/displays-new/Makefile
index 5aeb11b8fcd5..0323a8a1c682 100644
--- a/drivers/video/omap2/displays-new/Makefile
+++ b/drivers/video/omap2/displays-new/Makefile
@@ -8,5 +8,6 @@ obj-$(CONFIG_DISPLAY_PANEL_DSI_CM) += panel-dsi-cm.o
 obj-$(CONFIG_DISPLAY_PANEL_SONY_ACX565AKM) += panel-sony-acx565akm.o
 obj-$(CONFIG_DISPLAY_PANEL_LGPHILIPS_LB035Q02) += panel-lgphilips-lb035q02.o
 obj-$(CONFIG_DISPLAY_PANEL_SHARP_LS037V7DW01) += panel-sharp-ls037v7dw01.o
+obj-$(CONFIG_DISPLAY_PANEL_TPO_TD028TTEC1) += panel-tpo-td028ttec1.o
 obj-$(CONFIG_DISPLAY_PANEL_TPO_TD043MTEA1) += panel-tpo-td043mtea1.o
 obj-$(CONFIG_DISPLAY_PANEL_NEC_NL8048HL11) += panel-nec-nl8048hl11.o
diff --git a/drivers/video/omap2/displays-new/connector-dvi.c b/drivers/video/omap2/displays-new/connector-dvi.c
index 63d88ee6dfe4..b6c50904038e 100644
--- a/drivers/video/omap2/displays-new/connector-dvi.c
+++ b/drivers/video/omap2/displays-new/connector-dvi.c
@@ -262,6 +262,9 @@ static int dvic_probe_pdata(struct platform_device *pdev)
 
 	in = omap_dss_find_output(pdata->source);
 	if (in == NULL) {
+		if (ddata->i2c_adapter)
+			i2c_put_adapter(ddata->i2c_adapter);
+
 		dev_err(&pdev->dev, "Failed to find video source\n");
 		return -EPROBE_DEFER;
 	}
@@ -313,6 +316,10 @@ static int dvic_probe(struct platform_device *pdev)
 
 err_reg:
 	omap_dss_put_device(ddata->in);
+
+	if (ddata->i2c_adapter)
+		i2c_put_adapter(ddata->i2c_adapter);
+
 	return r;
 }
 
diff --git a/drivers/video/omap2/displays-new/encoder-tpd12s015.c b/drivers/video/omap2/displays-new/encoder-tpd12s015.c
index 798ef200b055..d5c936cb217f 100644
--- a/drivers/video/omap2/displays-new/encoder-tpd12s015.c
+++ b/drivers/video/omap2/displays-new/encoder-tpd12s015.c
@@ -69,7 +69,7 @@ static int tpd_connect(struct omap_dss_device *dssdev,
 	dst->src = dssdev;
 	dssdev->dst = dst;
 
-	INIT_COMPLETION(ddata->hpd_completion);
+	reinit_completion(&ddata->hpd_completion);
 
 	gpio_set_value_cansleep(ddata->ct_cp_hpd_gpio, 1);
 	/* DC-DC converter needs at max 300us to get to 90% of 5V */
diff --git a/drivers/video/omap2/displays-new/panel-dsi-cm.c b/drivers/video/omap2/displays-new/panel-dsi-cm.c
index aaaea6469cd9..b7baafe83aa3 100644
--- a/drivers/video/omap2/displays-new/panel-dsi-cm.c
+++ b/drivers/video/omap2/displays-new/panel-dsi-cm.c
@@ -599,7 +599,7 @@ static int dsicm_power_on(struct panel_drv_data *ddata)
 	if (r) {
 		dev_err(&ddata->pdev->dev, "failed to configure DSI pins\n");
 		goto err0;
-	};
+	}
 
 	r = in->ops.dsi->set_config(in, &dsi_config);
 	if (r) {
diff --git a/drivers/video/omap2/displays-new/panel-tpo-td028ttec1.c b/drivers/video/omap2/displays-new/panel-tpo-td028ttec1.c
new file mode 100644
index 000000000000..9a08908fe998
--- /dev/null
+++ b/drivers/video/omap2/displays-new/panel-tpo-td028ttec1.c
@@ -0,0 +1,480 @@
+/*
+ * Toppoly TD028TTEC1 panel support
+ *
+ * Copyright (C) 2008 Nokia Corporation
+ * Author: Tomi Valkeinen <tomi.valkeinen@nokia.com>
+ *
+ * Neo 1973 code (jbt6k74.c):
+ * Copyright (C) 2006-2007 by OpenMoko, Inc.
+ * Author: Harald Welte <laforge@openmoko.org>
+ *
+ * Ported and adapted from Neo 1973 U-Boot by:
+ * H. Nikolaus Schaller <hns@goldelico.com>
+ *
+ * This program is free software; you can redistribute it and/or modify it
+ * under the terms of the GNU General Public License version 2 as published by
+ * the Free Software Foundation.
+ *
+ * This program is distributed in the hope that it will be useful, but WITHOUT
+ * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or
+ * FITNESS FOR A PARTICULAR PURPOSE.  See the GNU General Public License for
+ * more details.
+ *
+ * You should have received a copy of the GNU General Public License along with
+ * this program.  If not, see <http://www.gnu.org/licenses/>.
+ */
+
+#include <linux/module.h>
+#include <linux/delay.h>
+#include <linux/spi/spi.h>
+#include <linux/gpio.h>
+#include <video/omapdss.h>
+#include <video/omap-panel-data.h>
+
+struct panel_drv_data {
+	struct omap_dss_device dssdev;
+	struct omap_dss_device *in;
+
+	int data_lines;
+
+	struct omap_video_timings videomode;
+
+	struct spi_device *spi_dev;
+};
+
+static struct omap_video_timings td028ttec1_panel_timings = {
+	.x_res		= 480,
+	.y_res		= 640,
+	.pixel_clock	= 22153,
+	.hfp		= 24,
+	.hsw		= 8,
+	.hbp		= 8,
+	.vfp		= 4,
+	.vsw		= 2,
+	.vbp		= 2,
+
+	.vsync_level	= OMAPDSS_SIG_ACTIVE_LOW,
+	.hsync_level	= OMAPDSS_SIG_ACTIVE_LOW,
+
+	.data_pclk_edge	= OMAPDSS_DRIVE_SIG_FALLING_EDGE,
+	.de_level	= OMAPDSS_SIG_ACTIVE_HIGH,
+	.sync_pclk_edge	= OMAPDSS_DRIVE_SIG_OPPOSITE_EDGES,
+};
+
+#define JBT_COMMAND	0x000
+#define JBT_DATA	0x100
+
+static int jbt_ret_write_0(struct panel_drv_data *ddata, u8 reg)
+{
+	int rc;
+	u16 tx_buf = JBT_COMMAND | reg;
+
+	rc = spi_write(ddata->spi_dev, (u8 *)&tx_buf,
+			1*sizeof(u16));
+	if (rc != 0)
+		dev_err(&ddata->spi_dev->dev,
+			"jbt_ret_write_0 spi_write ret %d\n", rc);
+
+	return rc;
+}
+
+static int jbt_reg_write_1(struct panel_drv_data *ddata, u8 reg, u8 data)
+{
+	int rc;
+	u16 tx_buf[2];
+
+	tx_buf[0] = JBT_COMMAND | reg;
+	tx_buf[1] = JBT_DATA | data;
+	rc = spi_write(ddata->spi_dev, (u8 *)tx_buf,
+			2*sizeof(u16));
+	if (rc != 0)
+		dev_err(&ddata->spi_dev->dev,
+			"jbt_reg_write_1 spi_write ret %d\n", rc);
+
+	return rc;
+}
+
+static int jbt_reg_write_2(struct panel_drv_data *ddata, u8 reg, u16 data)
+{
+	int rc;
+	u16 tx_buf[3];
+
+	tx_buf[0] = JBT_COMMAND | reg;
+	tx_buf[1] = JBT_DATA | (data >> 8);
+	tx_buf[2] = JBT_DATA | (data & 0xff);
+
+	rc = spi_write(ddata->spi_dev, (u8 *)tx_buf,
+			3*sizeof(u16));
+
+	if (rc != 0)
+		dev_err(&ddata->spi_dev->dev,
+			"jbt_reg_write_2 spi_write ret %d\n", rc);
+
+	return rc;
+}
+
+enum jbt_register {
+	JBT_REG_SLEEP_IN		= 0x10,
+	JBT_REG_SLEEP_OUT		= 0x11,
+
+	JBT_REG_DISPLAY_OFF		= 0x28,
+	JBT_REG_DISPLAY_ON		= 0x29,
+
+	JBT_REG_RGB_FORMAT		= 0x3a,
+	JBT_REG_QUAD_RATE		= 0x3b,
+
+	JBT_REG_POWER_ON_OFF		= 0xb0,
+	JBT_REG_BOOSTER_OP		= 0xb1,
+	JBT_REG_BOOSTER_MODE		= 0xb2,
+	JBT_REG_BOOSTER_FREQ		= 0xb3,
+	JBT_REG_OPAMP_SYSCLK		= 0xb4,
+	JBT_REG_VSC_VOLTAGE		= 0xb5,
+	JBT_REG_VCOM_VOLTAGE		= 0xb6,
+	JBT_REG_EXT_DISPL		= 0xb7,
+	JBT_REG_OUTPUT_CONTROL		= 0xb8,
+	JBT_REG_DCCLK_DCEV		= 0xb9,
+	JBT_REG_DISPLAY_MODE1		= 0xba,
+	JBT_REG_DISPLAY_MODE2		= 0xbb,
+	JBT_REG_DISPLAY_MODE		= 0xbc,
+	JBT_REG_ASW_SLEW		= 0xbd,
+	JBT_REG_DUMMY_DISPLAY		= 0xbe,
+	JBT_REG_DRIVE_SYSTEM		= 0xbf,
+
+	JBT_REG_SLEEP_OUT_FR_A		= 0xc0,
+	JBT_REG_SLEEP_OUT_FR_B		= 0xc1,
+	JBT_REG_SLEEP_OUT_FR_C		= 0xc2,
+	JBT_REG_SLEEP_IN_LCCNT_D	= 0xc3,
+	JBT_REG_SLEEP_IN_LCCNT_E	= 0xc4,
+	JBT_REG_SLEEP_IN_LCCNT_F	= 0xc5,
+	JBT_REG_SLEEP_IN_LCCNT_G	= 0xc6,
+
+	JBT_REG_GAMMA1_FINE_1		= 0xc7,
+	JBT_REG_GAMMA1_FINE_2		= 0xc8,
+	JBT_REG_GAMMA1_INCLINATION	= 0xc9,
+	JBT_REG_GAMMA1_BLUE_OFFSET	= 0xca,
+
+	JBT_REG_BLANK_CONTROL		= 0xcf,
+	JBT_REG_BLANK_TH_TV		= 0xd0,
+	JBT_REG_CKV_ON_OFF		= 0xd1,
+	JBT_REG_CKV_1_2			= 0xd2,
+	JBT_REG_OEV_TIMING		= 0xd3,
+	JBT_REG_ASW_TIMING_1		= 0xd4,
+	JBT_REG_ASW_TIMING_2		= 0xd5,
+
+	JBT_REG_HCLOCK_VGA		= 0xec,
+	JBT_REG_HCLOCK_QVGA		= 0xed,
+};
+
+#define to_panel_data(p) container_of(p, struct panel_drv_data, dssdev)
+
+static int td028ttec1_panel_connect(struct omap_dss_device *dssdev)
+{
+	struct panel_drv_data *ddata = to_panel_data(dssdev);
+	struct omap_dss_device *in = ddata->in;
+	int r;
+
+	if (omapdss_device_is_connected(dssdev))
+		return 0;
+
+	r = in->ops.dpi->connect(in, dssdev);
+	if (r)
+		return r;
+
+	return 0;
+}
+
+static void td028ttec1_panel_disconnect(struct omap_dss_device *dssdev)
+{
+	struct panel_drv_data *ddata = to_panel_data(dssdev);
+	struct omap_dss_device *in = ddata->in;
+
+	if (!omapdss_device_is_connected(dssdev))
+		return;
+
+	in->ops.dpi->disconnect(in, dssdev);
+}
+
+static int td028ttec1_panel_enable(struct omap_dss_device *dssdev)
+{
+	struct panel_drv_data *ddata = to_panel_data(dssdev);
+	struct omap_dss_device *in = ddata->in;
+	int r;
+
+	if (!omapdss_device_is_connected(dssdev))
+		return -ENODEV;
+
+	if (omapdss_device_is_enabled(dssdev))
+		return 0;
+
+	in->ops.dpi->set_data_lines(in, ddata->data_lines);
+	in->ops.dpi->set_timings(in, &ddata->videomode);
+
+	r = in->ops.dpi->enable(in);
+	if (r)
+		return r;
+
+	dev_dbg(dssdev->dev, "td028ttec1_panel_enable() - state %d\n",
+		dssdev->state);
+
+	/* three times command zero */
+	r |= jbt_ret_write_0(ddata, 0x00);
+	usleep_range(1000, 2000);
+	r |= jbt_ret_write_0(ddata, 0x00);
+	usleep_range(1000, 2000);
+	r |= jbt_ret_write_0(ddata, 0x00);
+	usleep_range(1000, 2000);
+
+	if (r) {
+		dev_warn(dssdev->dev, "transfer error\n");
+		goto transfer_err;
+	}
+
+	/* deep standby out */
+	r |= jbt_reg_write_1(ddata, JBT_REG_POWER_ON_OFF, 0x17);
+
+	/* RGB I/F on, RAM write off, QVGA through, SIGCON enable */
+	r |= jbt_reg_write_1(ddata, JBT_REG_DISPLAY_MODE, 0x80);
+
+	/* Quad mode off */
+	r |= jbt_reg_write_1(ddata, JBT_REG_QUAD_RATE, 0x00);
+
+	/* AVDD on, XVDD on */
+	r |= jbt_reg_write_1(ddata, JBT_REG_POWER_ON_OFF, 0x16);
+
+	/* Output control */
+	r |= jbt_reg_write_2(ddata, JBT_REG_OUTPUT_CONTROL, 0xfff9);
+
+	/* Sleep mode off */
+	r |= jbt_ret_write_0(ddata, JBT_REG_SLEEP_OUT);
+
+	/* at this point we have like 50% grey */
+
+	/* initialize register set */
+	r |= jbt_reg_write_1(ddata, JBT_REG_DISPLAY_MODE1, 0x01);
+	r |= jbt_reg_write_1(ddata, JBT_REG_DISPLAY_MODE2, 0x00);
+	r |= jbt_reg_write_1(ddata, JBT_REG_RGB_FORMAT, 0x60);
+	r |= jbt_reg_write_1(ddata, JBT_REG_DRIVE_SYSTEM, 0x10);
+	r |= jbt_reg_write_1(ddata, JBT_REG_BOOSTER_OP, 0x56);
+	r |= jbt_reg_write_1(ddata, JBT_REG_BOOSTER_MODE, 0x33);
+	r |= jbt_reg_write_1(ddata, JBT_REG_BOOSTER_FREQ, 0x11);
+	r |= jbt_reg_write_1(ddata, JBT_REG_BOOSTER_FREQ, 0x11);
+	r |= jbt_reg_write_1(ddata, JBT_REG_OPAMP_SYSCLK, 0x02);
+	r |= jbt_reg_write_1(ddata, JBT_REG_VSC_VOLTAGE, 0x2b);
+	r |= jbt_reg_write_1(ddata, JBT_REG_VCOM_VOLTAGE, 0x40);
+	r |= jbt_reg_write_1(ddata, JBT_REG_EXT_DISPL, 0x03);
+	r |= jbt_reg_write_1(ddata, JBT_REG_DCCLK_DCEV, 0x04);
+	/*
+	 * default of 0x02 in JBT_REG_ASW_SLEW responsible for 72Hz requirement
+	 * to avoid red / blue flicker
+	 */
+	r |= jbt_reg_write_1(ddata, JBT_REG_ASW_SLEW, 0x04);
+	r |= jbt_reg_write_1(ddata, JBT_REG_DUMMY_DISPLAY, 0x00);
+
+	r |= jbt_reg_write_1(ddata, JBT_REG_SLEEP_OUT_FR_A, 0x11);
+	r |= jbt_reg_write_1(ddata, JBT_REG_SLEEP_OUT_FR_B, 0x11);
+	r |= jbt_reg_write_1(ddata, JBT_REG_SLEEP_OUT_FR_C, 0x11);
+	r |= jbt_reg_write_2(ddata, JBT_REG_SLEEP_IN_LCCNT_D, 0x2040);
+	r |= jbt_reg_write_2(ddata, JBT_REG_SLEEP_IN_LCCNT_E, 0x60c0);
+	r |= jbt_reg_write_2(ddata, JBT_REG_SLEEP_IN_LCCNT_F, 0x1020);
+	r |= jbt_reg_write_2(ddata, JBT_REG_SLEEP_IN_LCCNT_G, 0x60c0);
+
+	r |= jbt_reg_write_2(ddata, JBT_REG_GAMMA1_FINE_1, 0x5533);
+	r |= jbt_reg_write_1(ddata, JBT_REG_GAMMA1_FINE_2, 0x00);
+	r |= jbt_reg_write_1(ddata, JBT_REG_GAMMA1_INCLINATION, 0x00);
+	r |= jbt_reg_write_1(ddata, JBT_REG_GAMMA1_BLUE_OFFSET, 0x00);
+
+	r |= jbt_reg_write_2(ddata, JBT_REG_HCLOCK_VGA, 0x1f0);
+	r |= jbt_reg_write_1(ddata, JBT_REG_BLANK_CONTROL, 0x02);
+	r |= jbt_reg_write_2(ddata, JBT_REG_BLANK_TH_TV, 0x0804);
+
+	r |= jbt_reg_write_1(ddata, JBT_REG_CKV_ON_OFF, 0x01);
+	r |= jbt_reg_write_2(ddata, JBT_REG_CKV_1_2, 0x0000);
+
+	r |= jbt_reg_write_2(ddata, JBT_REG_OEV_TIMING, 0x0d0e);
+	r |= jbt_reg_write_2(ddata, JBT_REG_ASW_TIMING_1, 0x11a4);
+	r |= jbt_reg_write_1(ddata, JBT_REG_ASW_TIMING_2, 0x0e);
+
+	r |= jbt_ret_write_0(ddata, JBT_REG_DISPLAY_ON);
+
+	dssdev->state = OMAP_DSS_DISPLAY_ACTIVE;
+
+transfer_err:
+
+	return r ? -EIO : 0;
+}
+
+static void td028ttec1_panel_disable(struct omap_dss_device *dssdev)
+{
+	struct panel_drv_data *ddata = to_panel_data(dssdev);
+	struct omap_dss_device *in = ddata->in;
+
+	if (!omapdss_device_is_enabled(dssdev))
+		return;
+
+	dev_dbg(dssdev->dev, "td028ttec1_panel_disable()\n");
+
+	jbt_ret_write_0(ddata, JBT_REG_DISPLAY_OFF);
+	jbt_reg_write_2(ddata, JBT_REG_OUTPUT_CONTROL, 0x8002);
+	jbt_ret_write_0(ddata, JBT_REG_SLEEP_IN);
+	jbt_reg_write_1(ddata, JBT_REG_POWER_ON_OFF, 0x00);
+
+	in->ops.dpi->disable(in);
+
+	dssdev->state = OMAP_DSS_DISPLAY_DISABLED;
+}
+
+static void td028ttec1_panel_set_timings(struct omap_dss_device *dssdev,
+		struct omap_video_timings *timings)
+{
+	struct panel_drv_data *ddata = to_panel_data(dssdev);
+	struct omap_dss_device *in = ddata->in;
+
+	ddata->videomode = *timings;
+	dssdev->panel.timings = *timings;
+
+	in->ops.dpi->set_timings(in, timings);
+}
+
+static void td028ttec1_panel_get_timings(struct omap_dss_device *dssdev,
+		struct omap_video_timings *timings)
+{
+	struct panel_drv_data *ddata = to_panel_data(dssdev);
+
+	*timings = ddata->videomode;
+}
+
+static int td028ttec1_panel_check_timings(struct omap_dss_device *dssdev,
+		struct omap_video_timings *timings)
+{
+	struct panel_drv_data *ddata = to_panel_data(dssdev);
+	struct omap_dss_device *in = ddata->in;
+
+	return in->ops.dpi->check_timings(in, timings);
+}
+
+static struct omap_dss_driver td028ttec1_ops = {
+	.connect	= td028ttec1_panel_connect,
+	.disconnect	= td028ttec1_panel_disconnect,
+
+	.enable		= td028ttec1_panel_enable,
+	.disable	= td028ttec1_panel_disable,
+
+	.set_timings	= td028ttec1_panel_set_timings,
+	.get_timings	= td028ttec1_panel_get_timings,
+	.check_timings	= td028ttec1_panel_check_timings,
+};
+
+static int td028ttec1_panel_probe_pdata(struct spi_device *spi)
+{
+	const struct panel_tpo_td028ttec1_platform_data *pdata;
+	struct panel_drv_data *ddata = dev_get_drvdata(&spi->dev);
+	struct omap_dss_device *dssdev, *in;
+
+	pdata = dev_get_platdata(&spi->dev);
+
+	in = omap_dss_find_output(pdata->source);
+	if (in == NULL) {
+		dev_err(&spi->dev, "failed to find video source '%s'\n",
+				pdata->source);
+		return -EPROBE_DEFER;
+	}
+
+	ddata->in = in;
+
+	ddata->data_lines = pdata->data_lines;
+
+	dssdev = &ddata->dssdev;
+	dssdev->name = pdata->name;
+
+	return 0;
+}
+
+static int td028ttec1_panel_probe(struct spi_device *spi)
+{
+	struct panel_drv_data *ddata;
+	struct omap_dss_device *dssdev;
+	int r;
+
+	dev_dbg(&spi->dev, "%s\n", __func__);
+
+	spi->bits_per_word = 9;
+	spi->mode = SPI_MODE_3;
+
+	r = spi_setup(spi);
+	if (r < 0) {
+		dev_err(&spi->dev, "spi_setup failed: %d\n", r);
+		return r;
+	}
+
+	ddata = devm_kzalloc(&spi->dev, sizeof(*ddata), GFP_KERNEL);
+	if (ddata == NULL)
+		return -ENOMEM;
+
+	dev_set_drvdata(&spi->dev, ddata);
+
+	ddata->spi_dev = spi;
+
+	if (dev_get_platdata(&spi->dev)) {
+		r = td028ttec1_panel_probe_pdata(spi);
+		if (r)
+			return r;
+	} else {
+		return -ENODEV;
+	}
+
+	ddata->videomode = td028ttec1_panel_timings;
+
+	dssdev = &ddata->dssdev;
+	dssdev->dev = &spi->dev;
+	dssdev->driver = &td028ttec1_ops;
+	dssdev->type = OMAP_DISPLAY_TYPE_DPI;
+	dssdev->owner = THIS_MODULE;
+	dssdev->panel.timings = ddata->videomode;
+	dssdev->phy.dpi.data_lines = ddata->data_lines;
+
+	r = omapdss_register_display(dssdev);
+	if (r) {
+		dev_err(&spi->dev, "Failed to register panel\n");
+		goto err_reg;
+	}
+
+	return 0;
+
+err_reg:
+	omap_dss_put_device(ddata->in);
+	return r;
+}
+
+static int td028ttec1_panel_remove(struct spi_device *spi)
+{
+	struct panel_drv_data *ddata = dev_get_drvdata(&spi->dev);
+	struct omap_dss_device *dssdev = &ddata->dssdev;
+	struct omap_dss_device *in = ddata->in;
+
+	dev_dbg(&ddata->spi_dev->dev, "%s\n", __func__);
+
+	omapdss_unregister_display(dssdev);
+
+	td028ttec1_panel_disable(dssdev);
+	td028ttec1_panel_disconnect(dssdev);
+
+	omap_dss_put_device(in);
+
+	return 0;
+}
+
+static struct spi_driver td028ttec1_spi_driver = {
+	.probe		= td028ttec1_panel_probe,
+	.remove		= td028ttec1_panel_remove,
+
+	.driver         = {
+		.name   = "panel-tpo-td028ttec1",
+		.owner  = THIS_MODULE,
+	},
+};
+
+module_spi_driver(td028ttec1_spi_driver);
+
+MODULE_AUTHOR("H. Nikolaus Schaller <hns@goldelico.com>");
+MODULE_DESCRIPTION("Toppoly TD028TTEC1 panel driver");
+MODULE_LICENSE("GPL");
diff --git a/drivers/video/omap2/dss/Makefile b/drivers/video/omap2/dss/Makefile
index 94832eb06a3d..d3aa91bdd6a8 100644
--- a/drivers/video/omap2/dss/Makefile
+++ b/drivers/video/omap2/dss/Makefile
@@ -10,5 +10,6 @@ omapdss-$(CONFIG_OMAP2_DSS_RFBI) += rfbi.o
 omapdss-$(CONFIG_OMAP2_DSS_VENC) += venc.o
 omapdss-$(CONFIG_OMAP2_DSS_SDI) += sdi.o
 omapdss-$(CONFIG_OMAP2_DSS_DSI) += dsi.o
-omapdss-$(CONFIG_OMAP4_DSS_HDMI) += hdmi.o ti_hdmi_4xxx_ip.o
+omapdss-$(CONFIG_OMAP4_DSS_HDMI) += hdmi4.o hdmi_common.o hdmi_wp.o hdmi_pll.o \
+	hdmi_phy.o hdmi4_core.o
 ccflags-$(CONFIG_OMAP2_DSS_DEBUG) += -DDEBUG
diff --git a/drivers/video/omap2/dss/core.c b/drivers/video/omap2/dss/core.c
index 60d3958d04f7..ffa45c894cd4 100644
--- a/drivers/video/omap2/dss/core.c
+++ b/drivers/video/omap2/dss/core.c
@@ -266,7 +266,7 @@ static int (*dss_output_drv_reg_funcs[])(void) __initdata = {
 	venc_init_platform_driver,
 #endif
 #ifdef CONFIG_OMAP4_DSS_HDMI
-	hdmi_init_platform_driver,
+	hdmi4_init_platform_driver,
 #endif
 };
 
@@ -287,7 +287,7 @@ static void (*dss_output_drv_unreg_funcs[])(void) __exitdata = {
 	venc_uninit_platform_driver,
 #endif
 #ifdef CONFIG_OMAP4_DSS_HDMI
-	hdmi_uninit_platform_driver,
+	hdmi4_uninit_platform_driver,
 #endif
 };
 
diff --git a/drivers/video/omap2/dss/dispc.c b/drivers/video/omap2/dss/dispc.c
index 477975009eee..4ec59ca72e5d 100644
--- a/drivers/video/omap2/dss/dispc.c
+++ b/drivers/video/omap2/dss/dispc.c
@@ -2352,7 +2352,7 @@ int dispc_ovl_check(enum omap_plane plane, enum omap_channel channel,
 {
 	enum omap_overlay_caps caps = dss_feat_get_overlay_caps(plane);
 	bool five_taps = true;
-	bool fieldmode = 0;
+	bool fieldmode = false;
 	u16 in_height = oi->height;
 	u16 in_width = oi->width;
 	bool ilace = timings->interlace;
@@ -2365,7 +2365,7 @@ int dispc_ovl_check(enum omap_plane plane, enum omap_channel channel,
 	out_height = oi->out_height == 0 ? oi->height : oi->out_height;
 
 	if (ilace && oi->height == out_height)
-		fieldmode = 1;
+		fieldmode = true;
 
 	if (ilace) {
 		if (fieldmode)
@@ -2396,7 +2396,7 @@ static int dispc_ovl_setup_common(enum omap_plane plane,
 		bool mem_to_mem)
 {
 	bool five_taps = true;
-	bool fieldmode = 0;
+	bool fieldmode = false;
 	int r, cconv = 0;
 	unsigned offset0, offset1;
 	s32 row_inc;
@@ -2417,7 +2417,7 @@ static int dispc_ovl_setup_common(enum omap_plane plane,
 	out_height = out_height == 0 ? height : out_height;
 
 	if (ilace && height == out_height)
-		fieldmode = 1;
+		fieldmode = true;
 
 	if (ilace) {
 		if (fieldmode)
@@ -2918,7 +2918,7 @@ static void _dispc_mgr_set_lcd_timings(enum omap_channel channel, int hsw,
 		break;
 	default:
 		BUG();
-	};
+	}
 
 	l = dispc_read_reg(DISPC_POL_FREQ(channel));
 	l |= FLD_VAL(onoff, 17, 17);
diff --git a/drivers/video/omap2/dss/display.c b/drivers/video/omap2/dss/display.c
index fafe7c941a60..669a81fdf58e 100644
--- a/drivers/video/omap2/dss/display.c
+++ b/drivers/video/omap2/dss/display.c
@@ -266,7 +266,7 @@ void videomode_to_omap_video_timings(const struct videomode *vm,
 		OMAPDSS_SIG_ACTIVE_LOW;
 	ovt->de_level = vm->flags & DISPLAY_FLAGS_DE_HIGH ?
 		OMAPDSS_SIG_ACTIVE_HIGH :
-		OMAPDSS_SIG_ACTIVE_HIGH;
+		OMAPDSS_SIG_ACTIVE_LOW;
 	ovt->data_pclk_edge = vm->flags & DISPLAY_FLAGS_PIXDATA_POSEDGE ?
 		OMAPDSS_DRIVE_SIG_RISING_EDGE :
 		OMAPDSS_DRIVE_SIG_FALLING_EDGE;
diff --git a/drivers/video/omap2/dss/dsi.c b/drivers/video/omap2/dss/dsi.c
index a598b5812285..6056b27cf73c 100644
--- a/drivers/video/omap2/dss/dsi.c
+++ b/drivers/video/omap2/dss/dsi.c
@@ -312,7 +312,7 @@ struct dsi_data {
 	struct dsi_isr_tables isr_tables_copy;
 
 	int update_channel;
-#ifdef DEBUG
+#ifdef DSI_PERF_MEASURE
 	unsigned update_bytes;
 #endif
 
@@ -334,7 +334,7 @@ struct dsi_data {
 
 	u32		errors;
 	spinlock_t	errors_lock;
-#ifdef DEBUG
+#ifdef DSI_PERF_MEASURE
 	ktime_t perf_setup_time;
 	ktime_t perf_start_time;
 #endif
@@ -373,7 +373,7 @@ struct dsi_packet_sent_handler_data {
 	struct completion *completion;
 };
 
-#ifdef DEBUG
+#ifdef DSI_PERF_MEASURE
 static bool dsi_perf;
 module_param(dsi_perf, bool, 0644);
 #endif
@@ -497,7 +497,7 @@ u8 dsi_get_pixel_size(enum omap_dss_dsi_pixel_format fmt)
 	}
 }
 
-#ifdef DEBUG
+#ifdef DSI_PERF_MEASURE
 static void dsi_perf_mark_setup(struct platform_device *dsidev)
 {
 	struct dsi_data *dsi = dsi_get_dsidrv_data(dsidev);
@@ -4066,7 +4066,7 @@ static int dsi_enable_video_output(struct omap_dss_device *dssdev, int channel)
 		default:
 			r = -EINVAL;
 			goto err_pix_fmt;
-		};
+		}
 
 		dsi_if_enable(dsidev, false);
 		dsi_vc_enable(dsidev, channel, false);
@@ -4277,7 +4277,7 @@ static int dsi_update(struct omap_dss_device *dssdev, int channel,
 	dw = dsi->timings.x_res;
 	dh = dsi->timings.y_res;
 
-#ifdef DEBUG
+#ifdef DSI_PERF_MEASURE
 	dsi->update_bytes = dw * dh *
 		dsi_get_pixel_size(dsi->pix_fmt) / 8;
 #endif
diff --git a/drivers/video/omap2/dss/dss.h b/drivers/video/omap2/dss/dss.h
index e172531d196b..f538e867c0f8 100644
--- a/drivers/video/omap2/dss/dss.h
+++ b/drivers/video/omap2/dss/dss.h
@@ -427,8 +427,8 @@ int venc_init_platform_driver(void) __init;
 void venc_uninit_platform_driver(void) __exit;
 
 /* HDMI */
-int hdmi_init_platform_driver(void) __init;
-void hdmi_uninit_platform_driver(void) __exit;
+int hdmi4_init_platform_driver(void) __init;
+void hdmi4_uninit_platform_driver(void) __exit;
 
 /* RFBI */
 int rfbi_init_platform_driver(void) __init;
diff --git a/drivers/video/omap2/dss/dss_features.c b/drivers/video/omap2/dss/dss_features.c
index b9cfebb378a2..f8fd6dbacabc 100644
--- a/drivers/video/omap2/dss/dss_features.c
+++ b/drivers/video/omap2/dss/dss_features.c
@@ -789,50 +789,6 @@ static const struct omap_dss_features omap5_dss_features = {
 	.burst_size_unit = 16,
 };
 
-#if defined(CONFIG_OMAP4_DSS_HDMI)
-/* HDMI OMAP4 Functions*/
-static const struct ti_hdmi_ip_ops omap4_hdmi_functions = {
-
-	.video_configure	=	ti_hdmi_4xxx_basic_configure,
-	.phy_enable		=	ti_hdmi_4xxx_phy_enable,
-	.phy_disable		=	ti_hdmi_4xxx_phy_disable,
-	.read_edid		=	ti_hdmi_4xxx_read_edid,
-	.pll_enable		=	ti_hdmi_4xxx_pll_enable,
-	.pll_disable		=	ti_hdmi_4xxx_pll_disable,
-	.video_enable		=	ti_hdmi_4xxx_wp_video_start,
-	.video_disable		=	ti_hdmi_4xxx_wp_video_stop,
-	.dump_wrapper		=	ti_hdmi_4xxx_wp_dump,
-	.dump_core		=	ti_hdmi_4xxx_core_dump,
-	.dump_pll		=	ti_hdmi_4xxx_pll_dump,
-	.dump_phy		=	ti_hdmi_4xxx_phy_dump,
-#if defined(CONFIG_OMAP4_DSS_HDMI_AUDIO)
-	.audio_enable		=       ti_hdmi_4xxx_wp_audio_enable,
-	.audio_disable		=       ti_hdmi_4xxx_wp_audio_disable,
-	.audio_start		=       ti_hdmi_4xxx_audio_start,
-	.audio_stop		=       ti_hdmi_4xxx_audio_stop,
-	.audio_config		=	ti_hdmi_4xxx_audio_config,
-	.audio_get_dma_port	=	ti_hdmi_4xxx_audio_get_dma_port,
-#endif
-
-};
-
-void dss_init_hdmi_ip_ops(struct hdmi_ip_data *ip_data,
-		enum omapdss_version version)
-{
-	switch (version) {
-	case OMAPDSS_VER_OMAP4430_ES1:
-	case OMAPDSS_VER_OMAP4430_ES2:
-	case OMAPDSS_VER_OMAP4:
-		ip_data->ops = &omap4_hdmi_functions;
-		break;
-	default:
-		ip_data->ops = NULL;
-	}
-
-	WARN_ON(ip_data->ops == NULL);
-}
-#endif
-
 /* Functions returning values related to a DSS feature */
 int dss_feat_get_num_mgrs(void)
 {
diff --git a/drivers/video/omap2/dss/dss_features.h b/drivers/video/omap2/dss/dss_features.h
index 489b9bec4a6d..10b0556e1352 100644
--- a/drivers/video/omap2/dss/dss_features.h
+++ b/drivers/video/omap2/dss/dss_features.h
@@ -20,10 +20,6 @@
 #ifndef __OMAP2_DSS_FEATURES_H
 #define __OMAP2_DSS_FEATURES_H
 
-#if defined(CONFIG_OMAP4_DSS_HDMI)
-#include "ti_hdmi.h"
-#endif
-
 #define MAX_DSS_MANAGERS	4
 #define MAX_DSS_OVERLAYS	4
 #define MAX_DSS_LCD_MANAGERS	3
@@ -117,8 +113,4 @@ bool dss_feat_rotation_type_supported(enum omap_dss_rotation_type rot_type);
 bool dss_has_feature(enum dss_feat_id id);
 void dss_feat_get_reg_field(enum dss_feat_reg_field id, u8 *start, u8 *end);
 void dss_features_init(enum omapdss_version version);
-#if defined(CONFIG_OMAP4_DSS_HDMI)
-void dss_init_hdmi_ip_ops(struct hdmi_ip_data *ip_data,
-		enum omapdss_version version);
-#endif
 #endif
diff --git a/drivers/video/omap2/dss/hdmi.c b/drivers/video/omap2/dss/hdmi.c
deleted file mode 100644
index 82a964074993..000000000000
--- a/drivers/video/omap2/dss/hdmi.c
+++ /dev/null
@@ -1,1184 +0,0 @@
-/*
- * hdmi.c
- *
- * HDMI interface DSS driver setting for TI's OMAP4 family of processor.
- * Copyright (C) 2010-2011 Texas Instruments Incorporated - http://www.ti.com/
- * Authors: Yong Zhi
- *	Mythri pk <mythripk@ti.com>
- *
- * This program is free software; you can redistribute it and/or modify it
- * under the terms of the GNU General Public License version 2 as published by
- * the Free Software Foundation.
- *
- * This program is distributed in the hope that it will be useful, but WITHOUT
- * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or
- * FITNESS FOR A PARTICULAR PURPOSE.  See the GNU General Public License for
- * more details.
- *
- * You should have received a copy of the GNU General Public License along with
- * this program.  If not, see <http://www.gnu.org/licenses/>.
- */
-
-#define DSS_SUBSYS_NAME "HDMI"
-
-#include <linux/kernel.h>
-#include <linux/module.h>
-#include <linux/err.h>
-#include <linux/io.h>
-#include <linux/interrupt.h>
-#include <linux/mutex.h>
-#include <linux/delay.h>
-#include <linux/string.h>
-#include <linux/platform_device.h>
-#include <linux/pm_runtime.h>
-#include <linux/clk.h>
-#include <linux/gpio.h>
-#include <linux/regulator/consumer.h>
-#include <video/omapdss.h>
-
-#include "ti_hdmi.h"
-#include "dss.h"
-#include "dss_features.h"
-
-#define HDMI_WP			0x0
-#define HDMI_CORE_SYS		0x400
-#define HDMI_CORE_AV		0x900
-#define HDMI_PLLCTRL		0x200
-#define HDMI_PHY		0x300
-
-/* HDMI EDID Length move this */
-#define HDMI_EDID_MAX_LENGTH			256
-#define EDID_TIMING_DESCRIPTOR_SIZE		0x12
-#define EDID_DESCRIPTOR_BLOCK0_ADDRESS		0x36
-#define EDID_DESCRIPTOR_BLOCK1_ADDRESS		0x80
-#define EDID_SIZE_BLOCK0_TIMING_DESCRIPTOR	4
-#define EDID_SIZE_BLOCK1_TIMING_DESCRIPTOR	4
-
-#define HDMI_DEFAULT_REGN 16
-#define HDMI_DEFAULT_REGM2 1
-
-static struct {
-	struct mutex lock;
-	struct platform_device *pdev;
-
-	struct hdmi_ip_data ip_data;
-
-	struct clk *sys_clk;
-	struct regulator *vdda_hdmi_dac_reg;
-
-	bool core_enabled;
-
-	struct omap_dss_device output;
-} hdmi;
-
-/*
- * Logic for the below structure :
- * user enters the CEA or VESA timings by specifying the HDMI/DVI code.
- * There is a correspondence between CEA/VESA timing and code, please
- * refer to section 6.3 in HDMI 1.3 specification for timing code.
- *
- * In the below structure, cea_vesa_timings corresponds to all OMAP4
- * supported CEA and VESA timing values.code_cea corresponds to the CEA
- * code, It is used to get the timing from cea_vesa_timing array.Similarly
- * with code_vesa. Code_index is used for back mapping, that is once EDID
- * is read from the TV, EDID is parsed to find the timing values and then
- * map it to corresponding CEA or VESA index.
- */
-
-static const struct hdmi_config cea_timings[] = {
-	{
-		{ 640, 480, 25200, 96, 16, 48, 2, 10, 33,
-			OMAPDSS_SIG_ACTIVE_LOW, OMAPDSS_SIG_ACTIVE_LOW,
-			false, },
-		{ 1, HDMI_HDMI },
-	},
-	{
-		{ 720, 480, 27027, 62, 16, 60, 6, 9, 30,
-			OMAPDSS_SIG_ACTIVE_LOW, OMAPDSS_SIG_ACTIVE_LOW,
-			false, },
-		{ 2, HDMI_HDMI },
-	},
-	{
-		{ 1280, 720, 74250, 40, 110, 220, 5, 5, 20,
-			OMAPDSS_SIG_ACTIVE_HIGH, OMAPDSS_SIG_ACTIVE_HIGH,
-			false, },
-		{ 4, HDMI_HDMI },
-	},
-	{
-		{ 1920, 540, 74250, 44, 88, 148, 5, 2, 15,
-			OMAPDSS_SIG_ACTIVE_HIGH, OMAPDSS_SIG_ACTIVE_HIGH,
-			true, },
-		{ 5, HDMI_HDMI },
-	},
-	{
-		{ 1440, 240, 27027, 124, 38, 114, 3, 4, 15,
-			OMAPDSS_SIG_ACTIVE_LOW, OMAPDSS_SIG_ACTIVE_LOW,
-			true, },
-		{ 6, HDMI_HDMI },
-	},
-	{
-		{ 1920, 1080, 148500, 44, 88, 148, 5, 4, 36,
-			OMAPDSS_SIG_ACTIVE_HIGH, OMAPDSS_SIG_ACTIVE_HIGH,
-			false, },
-		{ 16, HDMI_HDMI },
-	},
-	{
-		{ 720, 576, 27000, 64, 12, 68, 5, 5, 39,
-			OMAPDSS_SIG_ACTIVE_LOW, OMAPDSS_SIG_ACTIVE_LOW,
-			false, },
-		{ 17, HDMI_HDMI },
-	},
-	{
-		{ 1280, 720, 74250, 40, 440, 220, 5, 5, 20,
-			OMAPDSS_SIG_ACTIVE_HIGH, OMAPDSS_SIG_ACTIVE_HIGH,
-			false, },
-		{ 19, HDMI_HDMI },
-	},
-	{
-		{ 1920, 540, 74250, 44, 528, 148, 5, 2, 15,
-			OMAPDSS_SIG_ACTIVE_HIGH, OMAPDSS_SIG_ACTIVE_HIGH,
-			true, },
-		{ 20, HDMI_HDMI },
-	},
-	{
-		{ 1440, 288, 27000, 126, 24, 138, 3, 2, 19,
-			OMAPDSS_SIG_ACTIVE_LOW, OMAPDSS_SIG_ACTIVE_LOW,
-			true, },
-		{ 21, HDMI_HDMI },
-	},
-	{
-		{ 1440, 576, 54000, 128, 24, 136, 5, 5, 39,
-			OMAPDSS_SIG_ACTIVE_LOW, OMAPDSS_SIG_ACTIVE_LOW,
-			false, },
-		{ 29, HDMI_HDMI },
-	},
-	{
-		{ 1920, 1080, 148500, 44, 528, 148, 5, 4, 36,
-			OMAPDSS_SIG_ACTIVE_HIGH, OMAPDSS_SIG_ACTIVE_HIGH,
-			false, },
-		{ 31, HDMI_HDMI },
-	},
-	{
-		{ 1920, 1080, 74250, 44, 638, 148, 5, 4, 36,
-			OMAPDSS_SIG_ACTIVE_HIGH, OMAPDSS_SIG_ACTIVE_HIGH,
-			false, },
-		{ 32, HDMI_HDMI },
-	},
-	{
-		{ 2880, 480, 108108, 248, 64, 240, 6, 9, 30,
-			OMAPDSS_SIG_ACTIVE_LOW, OMAPDSS_SIG_ACTIVE_LOW,
-			false, },
-		{ 35, HDMI_HDMI },
-	},
-	{
-		{ 2880, 576, 108000, 256, 48, 272, 5, 5, 39,
-			OMAPDSS_SIG_ACTIVE_LOW, OMAPDSS_SIG_ACTIVE_LOW,
-			false, },
-		{ 37, HDMI_HDMI },
-	},
-};
-
-static const struct hdmi_config vesa_timings[] = {
-/* VESA From Here */
-	{
-		{ 640, 480, 25175, 96, 16, 48, 2, 11, 31,
-			OMAPDSS_SIG_ACTIVE_LOW, OMAPDSS_SIG_ACTIVE_LOW,
-			false, },
-		{ 4, HDMI_DVI },
-	},
-	{
-		{ 800, 600, 40000, 128, 40, 88, 4, 1, 23,
-			OMAPDSS_SIG_ACTIVE_HIGH, OMAPDSS_SIG_ACTIVE_HIGH,
-			false, },
-		{ 9, HDMI_DVI },
-	},
-	{
-		{ 848, 480, 33750, 112, 16, 112, 8, 6, 23,
-			OMAPDSS_SIG_ACTIVE_HIGH, OMAPDSS_SIG_ACTIVE_HIGH,
-			false, },
-		{ 0xE, HDMI_DVI },
-	},
-	{
-		{ 1280, 768, 79500, 128, 64, 192, 7, 3, 20,
-			OMAPDSS_SIG_ACTIVE_HIGH, OMAPDSS_SIG_ACTIVE_LOW,
-			false, },
-		{ 0x17, HDMI_DVI },
-	},
-	{
-		{ 1280, 800, 83500, 128, 72, 200, 6, 3, 22,
-			OMAPDSS_SIG_ACTIVE_HIGH, OMAPDSS_SIG_ACTIVE_LOW,
-			false, },
-		{ 0x1C, HDMI_DVI },
-	},
-	{
-		{ 1360, 768, 85500, 112, 64, 256, 6, 3, 18,
-			OMAPDSS_SIG_ACTIVE_HIGH, OMAPDSS_SIG_ACTIVE_HIGH,
-			false, },
-		{ 0x27, HDMI_DVI },
-	},
-	{
-		{ 1280, 960, 108000, 112, 96, 312, 3, 1, 36,
-			OMAPDSS_SIG_ACTIVE_HIGH, OMAPDSS_SIG_ACTIVE_HIGH,
-			false, },
-		{ 0x20, HDMI_DVI },
-	},
-	{
-		{ 1280, 1024, 108000, 112, 48, 248, 3, 1, 38,
-			OMAPDSS_SIG_ACTIVE_HIGH, OMAPDSS_SIG_ACTIVE_HIGH,
-			false, },
-		{ 0x23, HDMI_DVI },
-	},
-	{
-		{ 1024, 768, 65000, 136, 24, 160, 6, 3, 29,
-			OMAPDSS_SIG_ACTIVE_LOW, OMAPDSS_SIG_ACTIVE_LOW,
-			false, },
-		{ 0x10, HDMI_DVI },
-	},
-	{
-		{ 1400, 1050, 121750, 144, 88, 232, 4, 3, 32,
-			OMAPDSS_SIG_ACTIVE_HIGH, OMAPDSS_SIG_ACTIVE_LOW,
-			false, },
-		{ 0x2A, HDMI_DVI },
-	},
-	{
-		{ 1440, 900, 106500, 152, 80, 232, 6, 3, 25,
-			OMAPDSS_SIG_ACTIVE_HIGH, OMAPDSS_SIG_ACTIVE_LOW,
-			false, },
-		{ 0x2F, HDMI_DVI },
-	},
-	{
-		{ 1680, 1050, 146250, 176 , 104, 280, 6, 3, 30,
-			OMAPDSS_SIG_ACTIVE_HIGH, OMAPDSS_SIG_ACTIVE_LOW,
-			false, },
-		{ 0x3A, HDMI_DVI },
-	},
-	{
-		{ 1366, 768, 85500, 143, 70, 213, 3, 3, 24,
-			OMAPDSS_SIG_ACTIVE_HIGH, OMAPDSS_SIG_ACTIVE_HIGH,
-			false, },
-		{ 0x51, HDMI_DVI },
-	},
-	{
-		{ 1920, 1080, 148500, 44, 148, 80, 5, 4, 36,
-			OMAPDSS_SIG_ACTIVE_HIGH, OMAPDSS_SIG_ACTIVE_HIGH,
-			false, },
-		{ 0x52, HDMI_DVI },
-	},
-	{
-		{ 1280, 768, 68250, 32, 48, 80, 7, 3, 12,
-			OMAPDSS_SIG_ACTIVE_LOW, OMAPDSS_SIG_ACTIVE_HIGH,
-			false, },
-		{ 0x16, HDMI_DVI },
-	},
-	{
-		{ 1400, 1050, 101000, 32, 48, 80, 4, 3, 23,
-			OMAPDSS_SIG_ACTIVE_LOW, OMAPDSS_SIG_ACTIVE_HIGH,
-			false, },
-		{ 0x29, HDMI_DVI },
-	},
-	{
-		{ 1680, 1050, 119000, 32, 48, 80, 6, 3, 21,
-			OMAPDSS_SIG_ACTIVE_LOW, OMAPDSS_SIG_ACTIVE_HIGH,
-			false, },
-		{ 0x39, HDMI_DVI },
-	},
-	{
-		{ 1280, 800, 79500, 32, 48, 80, 6, 3, 14,
-			OMAPDSS_SIG_ACTIVE_LOW, OMAPDSS_SIG_ACTIVE_HIGH,
-			false, },
-		{ 0x1B, HDMI_DVI },
-	},
-	{
-		{ 1280, 720, 74250, 40, 110, 220, 5, 5, 20,
-			OMAPDSS_SIG_ACTIVE_HIGH, OMAPDSS_SIG_ACTIVE_HIGH,
-			false, },
-		{ 0x55, HDMI_DVI },
-	},
-	{
-		{ 1920, 1200, 154000, 32, 48, 80, 6, 3, 26,
-			OMAPDSS_SIG_ACTIVE_LOW, OMAPDSS_SIG_ACTIVE_HIGH,
-			false, },
-		{ 0x44, HDMI_DVI },
-	},
-};
-
-static int hdmi_runtime_get(void)
-{
-	int r;
-
-	DSSDBG("hdmi_runtime_get\n");
-
-	r = pm_runtime_get_sync(&hdmi.pdev->dev);
-	WARN_ON(r < 0);
-	if (r < 0)
-		return r;
-
-	return 0;
-}
-
-static void hdmi_runtime_put(void)
-{
-	int r;
-
-	DSSDBG("hdmi_runtime_put\n");
-
-	r = pm_runtime_put_sync(&hdmi.pdev->dev);
-	WARN_ON(r < 0 && r != -ENOSYS);
-}
-
-static int hdmi_init_regulator(void)
-{
-	struct regulator *reg;
-
-	if (hdmi.vdda_hdmi_dac_reg != NULL)
-		return 0;
-
-	reg = devm_regulator_get(&hdmi.pdev->dev, "vdda_hdmi_dac");
-
-	/* DT HACK: try VDAC to make omapdss work for o4 sdp/panda */
-	if (IS_ERR(reg))
-		reg = devm_regulator_get(&hdmi.pdev->dev, "VDAC");
-
-	if (IS_ERR(reg)) {
-		DSSERR("can't get VDDA_HDMI_DAC regulator\n");
-		return PTR_ERR(reg);
-	}
-
-	hdmi.vdda_hdmi_dac_reg = reg;
-
-	return 0;
-}
-
-static const struct hdmi_config *hdmi_find_timing(
-					const struct hdmi_config *timings_arr,
-					int len)
-{
-	int i;
-
-	for (i = 0; i < len; i++) {
-		if (timings_arr[i].cm.code == hdmi.ip_data.cfg.cm.code)
-			return &timings_arr[i];
-	}
-	return NULL;
-}
-
-static const struct hdmi_config *hdmi_get_timings(void)
-{
-       const struct hdmi_config *arr;
-       int len;
-
-       if (hdmi.ip_data.cfg.cm.mode == HDMI_DVI) {
-               arr = vesa_timings;
-               len = ARRAY_SIZE(vesa_timings);
-       } else {
-               arr = cea_timings;
-               len = ARRAY_SIZE(cea_timings);
-       }
-
-       return hdmi_find_timing(arr, len);
-}
-
-static bool hdmi_timings_compare(struct omap_video_timings *timing1,
-				const struct omap_video_timings *timing2)
-{
-	int timing1_vsync, timing1_hsync, timing2_vsync, timing2_hsync;
-
-	if ((DIV_ROUND_CLOSEST(timing2->pixel_clock, 1000) ==
-			DIV_ROUND_CLOSEST(timing1->pixel_clock, 1000)) &&
-		(timing2->x_res == timing1->x_res) &&
-		(timing2->y_res == timing1->y_res)) {
-
-		timing2_hsync = timing2->hfp + timing2->hsw + timing2->hbp;
-		timing1_hsync = timing1->hfp + timing1->hsw + timing1->hbp;
-		timing2_vsync = timing2->vfp + timing2->vsw + timing2->vbp;
-		timing1_vsync = timing2->vfp + timing2->vsw + timing2->vbp;
-
-		DSSDBG("timing1_hsync = %d timing1_vsync = %d"\
-			"timing2_hsync = %d timing2_vsync = %d\n",
-			timing1_hsync, timing1_vsync,
-			timing2_hsync, timing2_vsync);
-
-		if ((timing1_hsync == timing2_hsync) &&
-			(timing1_vsync == timing2_vsync)) {
-			return true;
-		}
-	}
-	return false;
-}
-
-static struct hdmi_cm hdmi_get_code(struct omap_video_timings *timing)
-{
-	int i;
-	struct hdmi_cm cm = {-1};
-	DSSDBG("hdmi_get_code\n");
-
-	for (i = 0; i < ARRAY_SIZE(cea_timings); i++) {
-		if (hdmi_timings_compare(timing, &cea_timings[i].timings)) {
-			cm = cea_timings[i].cm;
-			goto end;
-		}
-	}
-	for (i = 0; i < ARRAY_SIZE(vesa_timings); i++) {
-		if (hdmi_timings_compare(timing, &vesa_timings[i].timings)) {
-			cm = vesa_timings[i].cm;
-			goto end;
-		}
-	}
-
-end:	return cm;
-
-}
-
-static void hdmi_compute_pll(struct omap_dss_device *dssdev, int phy,
-		struct hdmi_pll_info *pi)
-{
-	unsigned long clkin, refclk;
-	u32 mf;
-
-	clkin = clk_get_rate(hdmi.sys_clk) / 10000;
-	/*
-	 * Input clock is predivided by N + 1
-	 * out put of which is reference clk
-	 */
-
-	pi->regn = HDMI_DEFAULT_REGN;
-
-	refclk = clkin / pi->regn;
-
-	pi->regm2 = HDMI_DEFAULT_REGM2;
-
-	/*
-	 * multiplier is pixel_clk/ref_clk
-	 * Multiplying by 100 to avoid fractional part removal
-	 */
-	pi->regm = phy * pi->regm2 / refclk;
-
-	/*
-	 * fractional multiplier is remainder of the difference between
-	 * multiplier and actual phy(required pixel clock thus should be
-	 * multiplied by 2^18(262144) divided by the reference clock
-	 */
-	mf = (phy - pi->regm / pi->regm2 * refclk) * 262144;
-	pi->regmf = pi->regm2 * mf / refclk;
-
-	/*
-	 * Dcofreq should be set to 1 if required pixel clock
-	 * is greater than 1000MHz
-	 */
-	pi->dcofreq = phy > 1000 * 100;
-	pi->regsd = ((pi->regm * clkin / 10) / (pi->regn * 250) + 5) / 10;
-
-	/* Set the reference clock to sysclk reference */
-	pi->refsel = HDMI_REFSEL_SYSCLK;
-
-	DSSDBG("M = %d Mf = %d\n", pi->regm, pi->regmf);
-	DSSDBG("range = %d sd = %d\n", pi->dcofreq, pi->regsd);
-}
-
-static int hdmi_power_on_core(struct omap_dss_device *dssdev)
-{
-	int r;
-
-	r = regulator_enable(hdmi.vdda_hdmi_dac_reg);
-	if (r)
-		return r;
-
-	r = hdmi_runtime_get();
-	if (r)
-		goto err_runtime_get;
-
-	/* Make selection of HDMI in DSS */
-	dss_select_hdmi_venc_clk_source(DSS_HDMI_M_PCLK);
-
-	hdmi.core_enabled = true;
-
-	return 0;
-
-err_runtime_get:
-	regulator_disable(hdmi.vdda_hdmi_dac_reg);
-
-	return r;
-}
-
-static void hdmi_power_off_core(struct omap_dss_device *dssdev)
-{
-	hdmi.core_enabled = false;
-
-	hdmi_runtime_put();
-	regulator_disable(hdmi.vdda_hdmi_dac_reg);
-}
-
-static int hdmi_power_on_full(struct omap_dss_device *dssdev)
-{
-	int r;
-	struct omap_video_timings *p;
-	struct omap_overlay_manager *mgr = hdmi.output.manager;
-	unsigned long phy;
-
-	r = hdmi_power_on_core(dssdev);
-	if (r)
-		return r;
-
-	dss_mgr_disable(mgr);
-
-	p = &hdmi.ip_data.cfg.timings;
-
-	DSSDBG("hdmi_power_on x_res= %d y_res = %d\n", p->x_res, p->y_res);
-
-	phy = p->pixel_clock;
-
-	hdmi_compute_pll(dssdev, phy, &hdmi.ip_data.pll_data);
-
-	hdmi.ip_data.ops->video_disable(&hdmi.ip_data);
-
-	/* config the PLL and PHY hdmi_set_pll_pwrfirst */
-	r = hdmi.ip_data.ops->pll_enable(&hdmi.ip_data);
-	if (r) {
-		DSSDBG("Failed to lock PLL\n");
-		goto err_pll_enable;
-	}
-
-	r = hdmi.ip_data.ops->phy_enable(&hdmi.ip_data);
-	if (r) {
-		DSSDBG("Failed to start PHY\n");
-		goto err_phy_enable;
-	}
-
-	hdmi.ip_data.ops->video_configure(&hdmi.ip_data);
-
-	/* bypass TV gamma table */
-	dispc_enable_gamma_table(0);
-
-	/* tv size */
-	dss_mgr_set_timings(mgr, p);
-
-	r = hdmi.ip_data.ops->video_enable(&hdmi.ip_data);
-	if (r)
-		goto err_vid_enable;
-
-	r = dss_mgr_enable(mgr);
-	if (r)
-		goto err_mgr_enable;
-
-	return 0;
-
-err_mgr_enable:
-	hdmi.ip_data.ops->video_disable(&hdmi.ip_data);
-err_vid_enable:
-	hdmi.ip_data.ops->phy_disable(&hdmi.ip_data);
-err_phy_enable:
-	hdmi.ip_data.ops->pll_disable(&hdmi.ip_data);
-err_pll_enable:
-	hdmi_power_off_core(dssdev);
-	return -EIO;
-}
-
-static void hdmi_power_off_full(struct omap_dss_device *dssdev)
-{
-	struct omap_overlay_manager *mgr = hdmi.output.manager;
-
-	dss_mgr_disable(mgr);
-
-	hdmi.ip_data.ops->video_disable(&hdmi.ip_data);
-	hdmi.ip_data.ops->phy_disable(&hdmi.ip_data);
-	hdmi.ip_data.ops->pll_disable(&hdmi.ip_data);
-
-	hdmi_power_off_core(dssdev);
-}
-
-static int hdmi_display_check_timing(struct omap_dss_device *dssdev,
-					struct omap_video_timings *timings)
-{
-	struct hdmi_cm cm;
-
-	cm = hdmi_get_code(timings);
-	if (cm.code == -1) {
-		return -EINVAL;
-	}
-
-	return 0;
-
-}
-
-static void hdmi_display_set_timing(struct omap_dss_device *dssdev,
-		struct omap_video_timings *timings)
-{
-	struct hdmi_cm cm;
-	const struct hdmi_config *t;
-
-	mutex_lock(&hdmi.lock);
-
-	cm = hdmi_get_code(timings);
-	hdmi.ip_data.cfg.cm = cm;
-
-	t = hdmi_get_timings();
-	if (t != NULL) {
-		hdmi.ip_data.cfg = *t;
-
-		dispc_set_tv_pclk(t->timings.pixel_clock * 1000);
-	}
-
-	mutex_unlock(&hdmi.lock);
-}
-
-static void hdmi_display_get_timings(struct omap_dss_device *dssdev,
-		struct omap_video_timings *timings)
-{
-	const struct hdmi_config *cfg;
-
-	cfg = hdmi_get_timings();
-	if (cfg == NULL)
-		cfg = &vesa_timings[0];
-
-	memcpy(timings, &cfg->timings, sizeof(cfg->timings));
-}
-
-static void hdmi_dump_regs(struct seq_file *s)
-{
-	mutex_lock(&hdmi.lock);
-
-	if (hdmi_runtime_get()) {
-		mutex_unlock(&hdmi.lock);
-		return;
-	}
-
-	hdmi.ip_data.ops->dump_wrapper(&hdmi.ip_data, s);
-	hdmi.ip_data.ops->dump_pll(&hdmi.ip_data, s);
-	hdmi.ip_data.ops->dump_phy(&hdmi.ip_data, s);
-	hdmi.ip_data.ops->dump_core(&hdmi.ip_data, s);
-
-	hdmi_runtime_put();
-	mutex_unlock(&hdmi.lock);
-}
-
-static int read_edid(u8 *buf, int len)
-{
-	int r;
-
-	mutex_lock(&hdmi.lock);
-
-	r = hdmi_runtime_get();
-	BUG_ON(r);
-
-	r = hdmi.ip_data.ops->read_edid(&hdmi.ip_data, buf, len);
-
-	hdmi_runtime_put();
-	mutex_unlock(&hdmi.lock);
-
-	return r;
-}
-
-static int hdmi_display_enable(struct omap_dss_device *dssdev)
-{
-	struct omap_dss_device *out = &hdmi.output;
-	int r = 0;
-
-	DSSDBG("ENTER hdmi_display_enable\n");
-
-	mutex_lock(&hdmi.lock);
-
-	if (out == NULL || out->manager == NULL) {
-		DSSERR("failed to enable display: no output/manager\n");
-		r = -ENODEV;
-		goto err0;
-	}
-
-	r = hdmi_power_on_full(dssdev);
-	if (r) {
-		DSSERR("failed to power on device\n");
-		goto err0;
-	}
-
-	mutex_unlock(&hdmi.lock);
-	return 0;
-
-err0:
-	mutex_unlock(&hdmi.lock);
-	return r;
-}
-
-static void hdmi_display_disable(struct omap_dss_device *dssdev)
-{
-	DSSDBG("Enter hdmi_display_disable\n");
-
-	mutex_lock(&hdmi.lock);
-
-	hdmi_power_off_full(dssdev);
-
-	mutex_unlock(&hdmi.lock);
-}
-
-static int hdmi_core_enable(struct omap_dss_device *dssdev)
-{
-	int r = 0;
-
-	DSSDBG("ENTER omapdss_hdmi_core_enable\n");
-
-	mutex_lock(&hdmi.lock);
-
-	r = hdmi_power_on_core(dssdev);
-	if (r) {
-		DSSERR("failed to power on device\n");
-		goto err0;
-	}
-
-	mutex_unlock(&hdmi.lock);
-	return 0;
-
-err0:
-	mutex_unlock(&hdmi.lock);
-	return r;
-}
-
-static void hdmi_core_disable(struct omap_dss_device *dssdev)
-{
-	DSSDBG("Enter omapdss_hdmi_core_disable\n");
-
-	mutex_lock(&hdmi.lock);
-
-	hdmi_power_off_core(dssdev);
-
-	mutex_unlock(&hdmi.lock);
-}
-
-static int hdmi_get_clocks(struct platform_device *pdev)
-{
-	struct clk *clk;
-
-	clk = devm_clk_get(&pdev->dev, "sys_clk");
-	if (IS_ERR(clk)) {
-		DSSERR("can't get sys_clk\n");
-		return PTR_ERR(clk);
-	}
-
-	hdmi.sys_clk = clk;
-
-	return 0;
-}
-
-#if defined(CONFIG_OMAP4_DSS_HDMI_AUDIO)
-int hdmi_compute_acr(u32 sample_freq, u32 *n, u32 *cts)
-{
-	u32 deep_color;
-	bool deep_color_correct = false;
-	u32 pclk = hdmi.ip_data.cfg.timings.pixel_clock;
-
-	if (n == NULL || cts == NULL)
-		return -EINVAL;
-
-	/* TODO: When implemented, query deep color mode here. */
-	deep_color = 100;
-
-	/*
-	 * When using deep color, the default N value (as in the HDMI
-	 * specification) yields to an non-integer CTS. Hence, we
-	 * modify it while keeping the restrictions described in
-	 * section 7.2.1 of the HDMI 1.4a specification.
-	 */
-	switch (sample_freq) {
-	case 32000:
-	case 48000:
-	case 96000:
-	case 192000:
-		if (deep_color == 125)
-			if (pclk == 27027 || pclk == 74250)
-				deep_color_correct = true;
-		if (deep_color == 150)
-			if (pclk == 27027)
-				deep_color_correct = true;
-		break;
-	case 44100:
-	case 88200:
-	case 176400:
-		if (deep_color == 125)
-			if (pclk == 27027)
-				deep_color_correct = true;
-		break;
-	default:
-		return -EINVAL;
-	}
-
-	if (deep_color_correct) {
-		switch (sample_freq) {
-		case 32000:
-			*n = 8192;
-			break;
-		case 44100:
-			*n = 12544;
-			break;
-		case 48000:
-			*n = 8192;
-			break;
-		case 88200:
-			*n = 25088;
-			break;
-		case 96000:
-			*n = 16384;
-			break;
-		case 176400:
-			*n = 50176;
-			break;
-		case 192000:
-			*n = 32768;
-			break;
-		default:
-			return -EINVAL;
-		}
-	} else {
-		switch (sample_freq) {
-		case 32000:
-			*n = 4096;
-			break;
-		case 44100:
-			*n = 6272;
-			break;
-		case 48000:
-			*n = 6144;
-			break;
-		case 88200:
-			*n = 12544;
-			break;
-		case 96000:
-			*n = 12288;
-			break;
-		case 176400:
-			*n = 25088;
-			break;
-		case 192000:
-			*n = 24576;
-			break;
-		default:
-			return -EINVAL;
-		}
-	}
-	/* Calculate CTS. See HDMI 1.3a or 1.4a specifications */
-	*cts = pclk * (*n / 128) * deep_color / (sample_freq / 10);
-
-	return 0;
-}
-
-static bool hdmi_mode_has_audio(void)
-{
-	if (hdmi.ip_data.cfg.cm.mode == HDMI_HDMI)
-		return true;
-	else
-		return false;
-}
-
-#endif
-
-static int hdmi_connect(struct omap_dss_device *dssdev,
-		struct omap_dss_device *dst)
-{
-	struct omap_overlay_manager *mgr;
-	int r;
-
-	dss_init_hdmi_ip_ops(&hdmi.ip_data, omapdss_get_version());
-
-	r = hdmi_init_regulator();
-	if (r)
-		return r;
-
-	mgr = omap_dss_get_overlay_manager(dssdev->dispc_channel);
-	if (!mgr)
-		return -ENODEV;
-
-	r = dss_mgr_connect(mgr, dssdev);
-	if (r)
-		return r;
-
-	r = omapdss_output_set_device(dssdev, dst);
-	if (r) {
-		DSSERR("failed to connect output to new device: %s\n",
-				dst->name);
-		dss_mgr_disconnect(mgr, dssdev);
-		return r;
-	}
-
-	return 0;
-}
-
-static void hdmi_disconnect(struct omap_dss_device *dssdev,
-		struct omap_dss_device *dst)
-{
-	WARN_ON(dst != dssdev->dst);
-
-	if (dst != dssdev->dst)
-		return;
-
-	omapdss_output_unset_device(dssdev);
-
-	if (dssdev->manager)
-		dss_mgr_disconnect(dssdev->manager, dssdev);
-}
-
-static int hdmi_read_edid(struct omap_dss_device *dssdev,
-		u8 *edid, int len)
-{
-	bool need_enable;
-	int r;
-
-	need_enable = hdmi.core_enabled == false;
-
-	if (need_enable) {
-		r = hdmi_core_enable(dssdev);
-		if (r)
-			return r;
-	}
-
-	r = read_edid(edid, len);
-
-	if (need_enable)
-		hdmi_core_disable(dssdev);
-
-	return r;
-}
-
-#if defined(CONFIG_OMAP4_DSS_HDMI_AUDIO)
-static int hdmi_audio_enable(struct omap_dss_device *dssdev)
-{
-	int r;
-
-	mutex_lock(&hdmi.lock);
-
-	if (!hdmi_mode_has_audio()) {
-		r = -EPERM;
-		goto err;
-	}
-
-
-	r = hdmi.ip_data.ops->audio_enable(&hdmi.ip_data);
-	if (r)
-		goto err;
-
-	mutex_unlock(&hdmi.lock);
-	return 0;
-
-err:
-	mutex_unlock(&hdmi.lock);
-	return r;
-}
-
-static void hdmi_audio_disable(struct omap_dss_device *dssdev)
-{
-	hdmi.ip_data.ops->audio_disable(&hdmi.ip_data);
-}
-
-static int hdmi_audio_start(struct omap_dss_device *dssdev)
-{
-	return hdmi.ip_data.ops->audio_start(&hdmi.ip_data);
-}
-
-static void hdmi_audio_stop(struct omap_dss_device *dssdev)
-{
-	hdmi.ip_data.ops->audio_stop(&hdmi.ip_data);
-}
-
-static bool hdmi_audio_supported(struct omap_dss_device *dssdev)
-{
-	bool r;
-
-	mutex_lock(&hdmi.lock);
-
-	r = hdmi_mode_has_audio();
-
-	mutex_unlock(&hdmi.lock);
-	return r;
-}
-
-static int hdmi_audio_config(struct omap_dss_device *dssdev,
-		struct omap_dss_audio *audio)
-{
-	int r;
-
-	mutex_lock(&hdmi.lock);
-
-	if (!hdmi_mode_has_audio()) {
-		r = -EPERM;
-		goto err;
-	}
-
-	r = hdmi.ip_data.ops->audio_config(&hdmi.ip_data, audio);
-	if (r)
-		goto err;
-
-	mutex_unlock(&hdmi.lock);
-	return 0;
-
-err:
-	mutex_unlock(&hdmi.lock);
-	return r;
-}
-#else
-static int hdmi_audio_enable(struct omap_dss_device *dssdev)
-{
-	return -EPERM;
-}
-
-static void hdmi_audio_disable(struct omap_dss_device *dssdev)
-{
-}
-
-static int hdmi_audio_start(struct omap_dss_device *dssdev)
-{
-	return -EPERM;
-}
-
-static void hdmi_audio_stop(struct omap_dss_device *dssdev)
-{
-}
-
-static bool hdmi_audio_supported(struct omap_dss_device *dssdev)
-{
-	return false;
-}
-
-static int hdmi_audio_config(struct omap_dss_device *dssdev,
-		struct omap_dss_audio *audio)
-{
-	return -EPERM;
-}
-#endif
-
-static const struct omapdss_hdmi_ops hdmi_ops = {
-	.connect		= hdmi_connect,
-	.disconnect		= hdmi_disconnect,
-
-	.enable			= hdmi_display_enable,
-	.disable		= hdmi_display_disable,
-
-	.check_timings		= hdmi_display_check_timing,
-	.set_timings		= hdmi_display_set_timing,
-	.get_timings		= hdmi_display_get_timings,
-
-	.read_edid		= hdmi_read_edid,
-
-	.audio_enable		= hdmi_audio_enable,
-	.audio_disable		= hdmi_audio_disable,
-	.audio_start		= hdmi_audio_start,
-	.audio_stop		= hdmi_audio_stop,
-	.audio_supported	= hdmi_audio_supported,
-	.audio_config		= hdmi_audio_config,
-};
-
-static void hdmi_init_output(struct platform_device *pdev)
-{
-	struct omap_dss_device *out = &hdmi.output;
-
-	out->dev = &pdev->dev;
-	out->id = OMAP_DSS_OUTPUT_HDMI;
-	out->output_type = OMAP_DISPLAY_TYPE_HDMI;
-	out->name = "hdmi.0";
-	out->dispc_channel = OMAP_DSS_CHANNEL_DIGIT;
-	out->ops.hdmi = &hdmi_ops;
-	out->owner = THIS_MODULE;
-
-	omapdss_register_output(out);
-}
-
-static void __exit hdmi_uninit_output(struct platform_device *pdev)
-{
-	struct omap_dss_device *out = &hdmi.output;
-
-	omapdss_unregister_output(out);
-}
-
-/* HDMI HW IP initialisation */
-static int omapdss_hdmihw_probe(struct platform_device *pdev)
-{
-	struct resource *res;
-	int r;
-
-	hdmi.pdev = pdev;
-
-	mutex_init(&hdmi.lock);
-	mutex_init(&hdmi.ip_data.lock);
-
-	res = platform_get_resource(hdmi.pdev, IORESOURCE_MEM, 0);
-
-	/* Base address taken from platform */
-	hdmi.ip_data.base_wp = devm_ioremap_resource(&pdev->dev, res);
-	if (IS_ERR(hdmi.ip_data.base_wp))
-		return PTR_ERR(hdmi.ip_data.base_wp);
-
-	hdmi.ip_data.irq = platform_get_irq(pdev, 0);
-	if (hdmi.ip_data.irq < 0) {
-		DSSERR("platform_get_irq failed\n");
-		return -ENODEV;
-	}
-
-	r = hdmi_get_clocks(pdev);
-	if (r) {
-		DSSERR("can't get clocks\n");
-		return r;
-	}
-
-	pm_runtime_enable(&pdev->dev);
-
-	hdmi.ip_data.core_sys_offset = HDMI_CORE_SYS;
-	hdmi.ip_data.core_av_offset = HDMI_CORE_AV;
-	hdmi.ip_data.pll_offset = HDMI_PLLCTRL;
-	hdmi.ip_data.phy_offset = HDMI_PHY;
-
-	hdmi_init_output(pdev);
-
-	dss_debugfs_create_file("hdmi", hdmi_dump_regs);
-
-	return 0;
-}
-
-static int __exit omapdss_hdmihw_remove(struct platform_device *pdev)
-{
-	hdmi_uninit_output(pdev);
-
-	pm_runtime_disable(&pdev->dev);
-
-	return 0;
-}
-
-static int hdmi_runtime_suspend(struct device *dev)
-{
-	clk_disable_unprepare(hdmi.sys_clk);
-
-	dispc_runtime_put();
-
-	return 0;
-}
-
-static int hdmi_runtime_resume(struct device *dev)
-{
-	int r;
-
-	r = dispc_runtime_get();
-	if (r < 0)
-		return r;
-
-	clk_prepare_enable(hdmi.sys_clk);
-
-	return 0;
-}
-
-static const struct dev_pm_ops hdmi_pm_ops = {
-	.runtime_suspend = hdmi_runtime_suspend,
-	.runtime_resume = hdmi_runtime_resume,
-};
-
-static struct platform_driver omapdss_hdmihw_driver = {
-	.probe		= omapdss_hdmihw_probe,
-	.remove         = __exit_p(omapdss_hdmihw_remove),
-	.driver         = {
-		.name   = "omapdss_hdmi",
-		.owner  = THIS_MODULE,
-		.pm	= &hdmi_pm_ops,
-	},
-};
-
-int __init hdmi_init_platform_driver(void)
-{
-	return platform_driver_register(&omapdss_hdmihw_driver);
-}
-
-void __exit hdmi_uninit_platform_driver(void)
-{
-	platform_driver_unregister(&omapdss_hdmihw_driver);
-}
diff --git a/drivers/video/omap2/dss/hdmi.h b/drivers/video/omap2/dss/hdmi.h
new file mode 100644
index 000000000000..b0493768a5d7
--- /dev/null
+++ b/drivers/video/omap2/dss/hdmi.h
@@ -0,0 +1,444 @@
+/*
+ * HDMI driver definition for TI OMAP4 Processor.
+ *
+ * Copyright (C) 2010-2011 Texas Instruments Incorporated - http://www.ti.com/
+ *
+ * This program is free software; you can redistribute it and/or modify it
+ * under the terms of the GNU General Public License version 2 as published by
+ * the Free Software Foundation.
+ *
+ * This program is distributed in the hope that it will be useful, but WITHOUT
+ * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or
+ * FITNESS FOR A PARTICULAR PURPOSE.  See the GNU General Public License for
+ * more details.
+ *
+ * You should have received a copy of the GNU General Public License along with
+ * this program.  If not, see <http://www.gnu.org/licenses/>.
+ */
+
+#ifndef _HDMI_H
+#define _HDMI_H
+
+#include <linux/delay.h>
+#include <linux/io.h>
+#include <linux/platform_device.h>
+#include <video/omapdss.h>
+
+#include "dss.h"
+
+/* HDMI Wrapper */
+
+#define HDMI_WP_REVISION			0x0
+#define HDMI_WP_SYSCONFIG			0x10
+#define HDMI_WP_IRQSTATUS_RAW			0x24
+#define HDMI_WP_IRQSTATUS			0x28
+#define HDMI_WP_IRQENABLE_SET			0x2C
+#define HDMI_WP_IRQENABLE_CLR			0x30
+#define HDMI_WP_IRQWAKEEN			0x34
+#define HDMI_WP_PWR_CTRL			0x40
+#define HDMI_WP_DEBOUNCE			0x44
+#define HDMI_WP_VIDEO_CFG			0x50
+#define HDMI_WP_VIDEO_SIZE			0x60
+#define HDMI_WP_VIDEO_TIMING_H			0x68
+#define HDMI_WP_VIDEO_TIMING_V			0x6C
+#define HDMI_WP_WP_CLK				0x70
+#define HDMI_WP_AUDIO_CFG			0x80
+#define HDMI_WP_AUDIO_CFG2			0x84
+#define HDMI_WP_AUDIO_CTRL			0x88
+#define HDMI_WP_AUDIO_DATA			0x8C
+
+/* HDMI WP IRQ flags */
+
+#define HDMI_IRQ_OCP_TIMEOUT			(1 << 4)
+#define HDMI_IRQ_AUDIO_FIFO_UNDERFLOW		(1 << 8)
+#define HDMI_IRQ_AUDIO_FIFO_OVERFLOW		(1 << 9)
+#define HDMI_IRQ_AUDIO_FIFO_SAMPLE_REQ		(1 << 10)
+#define HDMI_IRQ_VIDEO_VSYNC			(1 << 16)
+#define HDMI_IRQ_VIDEO_FRAME_DONE		(1 << 17)
+#define HDMI_IRQ_PHY_LINE5V_ASSERT		(1 << 24)
+#define HDMI_IRQ_LINK_CONNECT			(1 << 25)
+#define HDMI_IRQ_LINK_DISCONNECT		(1 << 26)
+#define HDMI_IRQ_PLL_LOCK			(1 << 29)
+#define HDMI_IRQ_PLL_UNLOCK			(1 << 30)
+#define HDMI_IRQ_PLL_RECAL			(1 << 31)
+
+/* HDMI PLL */
+
+#define PLLCTRL_PLL_CONTROL			0x0
+#define PLLCTRL_PLL_STATUS			0x4
+#define PLLCTRL_PLL_GO				0x8
+#define PLLCTRL_CFG1				0xC
+#define PLLCTRL_CFG2				0x10
+#define PLLCTRL_CFG3				0x14
+#define PLLCTRL_SSC_CFG1			0x18
+#define PLLCTRL_SSC_CFG2			0x1C
+#define PLLCTRL_CFG4				0x20
+
+/* HDMI PHY */
+
+#define HDMI_TXPHY_TX_CTRL			0x0
+#define HDMI_TXPHY_DIGITAL_CTRL			0x4
+#define HDMI_TXPHY_POWER_CTRL			0x8
+#define HDMI_TXPHY_PAD_CFG_CTRL			0xC
+
+enum hdmi_pll_pwr {
+	HDMI_PLLPWRCMD_ALLOFF = 0,
+	HDMI_PLLPWRCMD_PLLONLY = 1,
+	HDMI_PLLPWRCMD_BOTHON_ALLCLKS = 2,
+	HDMI_PLLPWRCMD_BOTHON_NOPHYCLK = 3
+};
+
+enum hdmi_phy_pwr {
+	HDMI_PHYPWRCMD_OFF = 0,
+	HDMI_PHYPWRCMD_LDOON = 1,
+	HDMI_PHYPWRCMD_TXON = 2
+};
+
+enum hdmi_core_hdmi_dvi {
+	HDMI_DVI = 0,
+	HDMI_HDMI = 1
+};
+
+enum hdmi_clk_refsel {
+	HDMI_REFSEL_PCLK = 0,
+	HDMI_REFSEL_REF1 = 1,
+	HDMI_REFSEL_REF2 = 2,
+	HDMI_REFSEL_SYSCLK = 3
+};
+
+enum hdmi_packing_mode {
+	HDMI_PACK_10b_RGB_YUV444 = 0,
+	HDMI_PACK_24b_RGB_YUV444_YUV422 = 1,
+	HDMI_PACK_20b_YUV422 = 2,
+	HDMI_PACK_ALREADYPACKED = 7
+};
+
+enum hdmi_stereo_channels {
+	HDMI_AUDIO_STEREO_NOCHANNELS = 0,
+	HDMI_AUDIO_STEREO_ONECHANNEL = 1,
+	HDMI_AUDIO_STEREO_TWOCHANNELS = 2,
+	HDMI_AUDIO_STEREO_THREECHANNELS = 3,
+	HDMI_AUDIO_STEREO_FOURCHANNELS = 4
+};
+
+enum hdmi_audio_type {
+	HDMI_AUDIO_TYPE_LPCM = 0,
+	HDMI_AUDIO_TYPE_IEC = 1
+};
+
+enum hdmi_audio_justify {
+	HDMI_AUDIO_JUSTIFY_LEFT = 0,
+	HDMI_AUDIO_JUSTIFY_RIGHT = 1
+};
+
+enum hdmi_audio_sample_order {
+	HDMI_AUDIO_SAMPLE_RIGHT_FIRST = 0,
+	HDMI_AUDIO_SAMPLE_LEFT_FIRST = 1
+};
+
+enum hdmi_audio_samples_perword {
+	HDMI_AUDIO_ONEWORD_ONESAMPLE = 0,
+	HDMI_AUDIO_ONEWORD_TWOSAMPLES = 1
+};
+
+enum hdmi_audio_sample_size {
+	HDMI_AUDIO_SAMPLE_16BITS = 0,
+	HDMI_AUDIO_SAMPLE_24BITS = 1
+};
+
+enum hdmi_audio_transf_mode {
+	HDMI_AUDIO_TRANSF_DMA = 0,
+	HDMI_AUDIO_TRANSF_IRQ = 1
+};
+
+enum hdmi_audio_blk_strt_end_sig {
+	HDMI_AUDIO_BLOCK_SIG_STARTEND_ON = 0,
+	HDMI_AUDIO_BLOCK_SIG_STARTEND_OFF = 1
+};
+
+enum hdmi_core_audio_layout {
+	HDMI_AUDIO_LAYOUT_2CH = 0,
+	HDMI_AUDIO_LAYOUT_8CH = 1
+};
+
+enum hdmi_core_cts_mode {
+	HDMI_AUDIO_CTS_MODE_HW = 0,
+	HDMI_AUDIO_CTS_MODE_SW = 1
+};
+
+enum hdmi_audio_mclk_mode {
+	HDMI_AUDIO_MCLK_128FS = 0,
+	HDMI_AUDIO_MCLK_256FS = 1,
+	HDMI_AUDIO_MCLK_384FS = 2,
+	HDMI_AUDIO_MCLK_512FS = 3,
+	HDMI_AUDIO_MCLK_768FS = 4,
+	HDMI_AUDIO_MCLK_1024FS = 5,
+	HDMI_AUDIO_MCLK_1152FS = 6,
+	HDMI_AUDIO_MCLK_192FS = 7
+};
+
+/* INFOFRAME_AVI_ and INFOFRAME_AUDIO_ definitions */
+enum hdmi_core_infoframe {
+	HDMI_INFOFRAME_AVI_DB1Y_RGB = 0,
+	HDMI_INFOFRAME_AVI_DB1Y_YUV422 = 1,
+	HDMI_INFOFRAME_AVI_DB1Y_YUV444 = 2,
+	HDMI_INFOFRAME_AVI_DB1A_ACTIVE_FORMAT_OFF = 0,
+	HDMI_INFOFRAME_AVI_DB1A_ACTIVE_FORMAT_ON =  1,
+	HDMI_INFOFRAME_AVI_DB1B_NO = 0,
+	HDMI_INFOFRAME_AVI_DB1B_VERT = 1,
+	HDMI_INFOFRAME_AVI_DB1B_HORI = 2,
+	HDMI_INFOFRAME_AVI_DB1B_VERTHORI = 3,
+	HDMI_INFOFRAME_AVI_DB1S_0 = 0,
+	HDMI_INFOFRAME_AVI_DB1S_1 = 1,
+	HDMI_INFOFRAME_AVI_DB1S_2 = 2,
+	HDMI_INFOFRAME_AVI_DB2C_NO = 0,
+	HDMI_INFOFRAME_AVI_DB2C_ITU601 = 1,
+	HDMI_INFOFRAME_AVI_DB2C_ITU709 = 2,
+	HDMI_INFOFRAME_AVI_DB2C_EC_EXTENDED = 3,
+	HDMI_INFOFRAME_AVI_DB2M_NO = 0,
+	HDMI_INFOFRAME_AVI_DB2M_43 = 1,
+	HDMI_INFOFRAME_AVI_DB2M_169 = 2,
+	HDMI_INFOFRAME_AVI_DB2R_SAME = 8,
+	HDMI_INFOFRAME_AVI_DB2R_43 = 9,
+	HDMI_INFOFRAME_AVI_DB2R_169 = 10,
+	HDMI_INFOFRAME_AVI_DB2R_149 = 11,
+	HDMI_INFOFRAME_AVI_DB3ITC_NO = 0,
+	HDMI_INFOFRAME_AVI_DB3ITC_YES = 1,
+	HDMI_INFOFRAME_AVI_DB3EC_XVYUV601 = 0,
+	HDMI_INFOFRAME_AVI_DB3EC_XVYUV709 = 1,
+	HDMI_INFOFRAME_AVI_DB3Q_DEFAULT = 0,
+	HDMI_INFOFRAME_AVI_DB3Q_LR = 1,
+	HDMI_INFOFRAME_AVI_DB3Q_FR = 2,
+	HDMI_INFOFRAME_AVI_DB3SC_NO = 0,
+	HDMI_INFOFRAME_AVI_DB3SC_HORI = 1,
+	HDMI_INFOFRAME_AVI_DB3SC_VERT = 2,
+	HDMI_INFOFRAME_AVI_DB3SC_HORIVERT = 3,
+	HDMI_INFOFRAME_AVI_DB5PR_NO = 0,
+	HDMI_INFOFRAME_AVI_DB5PR_2 = 1,
+	HDMI_INFOFRAME_AVI_DB5PR_3 = 2,
+	HDMI_INFOFRAME_AVI_DB5PR_4 = 3,
+	HDMI_INFOFRAME_AVI_DB5PR_5 = 4,
+	HDMI_INFOFRAME_AVI_DB5PR_6 = 5,
+	HDMI_INFOFRAME_AVI_DB5PR_7 = 6,
+	HDMI_INFOFRAME_AVI_DB5PR_8 = 7,
+	HDMI_INFOFRAME_AVI_DB5PR_9 = 8,
+	HDMI_INFOFRAME_AVI_DB5PR_10 = 9,
+};
+
+struct hdmi_cm {
+	int	code;
+	int	mode;
+};
+
+struct hdmi_video_format {
+	enum hdmi_packing_mode	packing_mode;
+	u32			y_res;	/* Line per panel */
+	u32			x_res;	/* pixel per line */
+};
+
+struct hdmi_config {
+	struct omap_video_timings timings;
+	struct hdmi_cm cm;
+};
+
+/* HDMI PLL structure */
+struct hdmi_pll_info {
+	u16 regn;
+	u16 regm;
+	u32 regmf;
+	u16 regm2;
+	u16 regsd;
+	u16 dcofreq;
+	enum hdmi_clk_refsel refsel;
+};
+
+struct hdmi_audio_format {
+	enum hdmi_stereo_channels		stereo_channels;
+	u8					active_chnnls_msk;
+	enum hdmi_audio_type			type;
+	enum hdmi_audio_justify			justification;
+	enum hdmi_audio_sample_order		sample_order;
+	enum hdmi_audio_samples_perword		samples_per_word;
+	enum hdmi_audio_sample_size		sample_size;
+	enum hdmi_audio_blk_strt_end_sig	en_sig_blk_strt_end;
+};
+
+struct hdmi_audio_dma {
+	u8				transfer_size;
+	u8				block_size;
+	enum hdmi_audio_transf_mode	mode;
+	u16				fifo_threshold;
+};
+
+struct hdmi_core_audio_i2s_config {
+	u8 in_length_bits;
+	u8 justification;
+	u8 sck_edge_mode;
+	u8 vbit;
+	u8 direction;
+	u8 shift;
+	u8 active_sds;
+};
+
+struct hdmi_core_audio_config {
+	struct hdmi_core_audio_i2s_config	i2s_cfg;
+	struct snd_aes_iec958			*iec60958_cfg;
+	bool					fs_override;
+	u32					n;
+	u32					cts;
+	u32					aud_par_busclk;
+	enum hdmi_core_audio_layout		layout;
+	enum hdmi_core_cts_mode			cts_mode;
+	bool					use_mclk;
+	enum hdmi_audio_mclk_mode		mclk_mode;
+	bool					en_acr_pkt;
+	bool					en_dsd_audio;
+	bool					en_parallel_aud_input;
+	bool					en_spdif;
+};
+
+/*
+ * Refer to section 8.2 in HDMI 1.3 specification for
+ * details about infoframe databytes
+ */
+struct hdmi_core_infoframe_avi {
+	/* Y0, Y1 rgb,yCbCr */
+	u8	db1_format;
+	/* A0  Active information Present */
+	u8	db1_active_info;
+	/* B0, B1 Bar info data valid */
+	u8	db1_bar_info_dv;
+	/* S0, S1 scan information */
+	u8	db1_scan_info;
+	/* C0, C1 colorimetry */
+	u8	db2_colorimetry;
+	/* M0, M1 Aspect ratio (4:3, 16:9) */
+	u8	db2_aspect_ratio;
+	/* R0...R3 Active format aspect ratio */
+	u8	db2_active_fmt_ar;
+	/* ITC IT content. */
+	u8	db3_itc;
+	/* EC0, EC1, EC2 Extended colorimetry */
+	u8	db3_ec;
+	/* Q1, Q0 Quantization range */
+	u8	db3_q_range;
+	/* SC1, SC0 Non-uniform picture scaling */
+	u8	db3_nup_scaling;
+	/* VIC0..6 Video format identification */
+	u8	db4_videocode;
+	/* PR0..PR3 Pixel repetition factor */
+	u8	db5_pixel_repeat;
+	/* Line number end of top bar */
+	u16	db6_7_line_eoftop;
+	/* Line number start of bottom bar */
+	u16	db8_9_line_sofbottom;
+	/* Pixel number end of left bar */
+	u16	db10_11_pixel_eofleft;
+	/* Pixel number start of right bar */
+	u16	db12_13_pixel_sofright;
+};
+
+struct hdmi_wp_data {
+	void __iomem *base;
+};
+
+struct hdmi_pll_data {
+	void __iomem *base;
+
+	struct hdmi_pll_info info;
+};
+
+struct hdmi_phy_data {
+	void __iomem *base;
+
+	int irq;
+};
+
+struct hdmi_core_data {
+	void __iomem *base;
+
+	struct hdmi_core_infoframe_avi avi_cfg;
+};
+
+static inline void hdmi_write_reg(void __iomem *base_addr, const u16 idx,
+		u32 val)
+{
+	__raw_writel(val, base_addr + idx);
+}
+
+static inline u32 hdmi_read_reg(void __iomem *base_addr, const u16 idx)
+{
+	return __raw_readl(base_addr + idx);
+}
+
+#define REG_FLD_MOD(base, idx, val, start, end) \
+	hdmi_write_reg(base, idx, FLD_MOD(hdmi_read_reg(base, idx),\
+							val, start, end))
+#define REG_GET(base, idx, start, end) \
+	FLD_GET(hdmi_read_reg(base, idx), start, end)
+
+static inline int hdmi_wait_for_bit_change(void __iomem *base_addr,
+		const u16 idx, int b2, int b1, u32 val)
+{
+	u32 t = 0;
+	while (val != REG_GET(base_addr, idx, b2, b1)) {
+		udelay(1);
+		if (t++ > 10000)
+			return !val;
+	}
+	return val;
+}
+
+/* HDMI wrapper funcs */
+int hdmi_wp_video_start(struct hdmi_wp_data *wp);
+void hdmi_wp_video_stop(struct hdmi_wp_data *wp);
+void hdmi_wp_dump(struct hdmi_wp_data *wp, struct seq_file *s);
+u32 hdmi_wp_get_irqstatus(struct hdmi_wp_data *wp);
+void hdmi_wp_set_irqstatus(struct hdmi_wp_data *wp, u32 irqstatus);
+void hdmi_wp_set_irqenable(struct hdmi_wp_data *wp, u32 mask);
+void hdmi_wp_clear_irqenable(struct hdmi_wp_data *wp, u32 mask);
+int hdmi_wp_set_phy_pwr(struct hdmi_wp_data *wp, enum hdmi_phy_pwr val);
+int hdmi_wp_set_pll_pwr(struct hdmi_wp_data *wp, enum hdmi_pll_pwr val);
+void hdmi_wp_video_config_format(struct hdmi_wp_data *wp,
+		struct hdmi_video_format *video_fmt);
+void hdmi_wp_video_config_interface(struct hdmi_wp_data *wp,
+		struct omap_video_timings *timings);
+void hdmi_wp_video_config_timing(struct hdmi_wp_data *wp,
+		struct omap_video_timings *timings);
+void hdmi_wp_init_vid_fmt_timings(struct hdmi_video_format *video_fmt,
+		struct omap_video_timings *timings, struct hdmi_config *param);
+int hdmi_wp_init(struct platform_device *pdev, struct hdmi_wp_data *wp);
+
+/* HDMI PLL funcs */
+int hdmi_pll_enable(struct hdmi_pll_data *pll, struct hdmi_wp_data *wp);
+void hdmi_pll_disable(struct hdmi_pll_data *pll, struct hdmi_wp_data *wp);
+void hdmi_pll_dump(struct hdmi_pll_data *pll, struct seq_file *s);
+void hdmi_pll_compute(struct hdmi_pll_data *pll, unsigned long clkin, int phy);
+int hdmi_pll_init(struct platform_device *pdev, struct hdmi_pll_data *pll);
+
+/* HDMI PHY funcs */
+int hdmi_phy_enable(struct hdmi_phy_data *phy, struct hdmi_wp_data *wp,
+		struct hdmi_config *cfg);
+void hdmi_phy_disable(struct hdmi_phy_data *phy, struct hdmi_wp_data *wp);
+void hdmi_phy_dump(struct hdmi_phy_data *phy, struct seq_file *s);
+int hdmi_phy_init(struct platform_device *pdev, struct hdmi_phy_data *phy);
+
+/* HDMI common funcs */
+const struct hdmi_config *hdmi_default_timing(void);
+const struct hdmi_config *hdmi_get_timings(int mode, int code);
+struct hdmi_cm hdmi_get_code(struct omap_video_timings *timing);
+
+#if defined(CONFIG_OMAP4_DSS_HDMI_AUDIO)
+int hdmi_compute_acr(u32 pclk, u32 sample_freq, u32 *n, u32 *cts);
+int hdmi_wp_audio_enable(struct hdmi_wp_data *wp, bool enable);
+int hdmi_wp_audio_core_req_enable(struct hdmi_wp_data *wp, bool enable);
+void hdmi_wp_audio_config_format(struct hdmi_wp_data *wp,
+		struct hdmi_audio_format *aud_fmt);
+void hdmi_wp_audio_config_dma(struct hdmi_wp_data *wp,
+		struct hdmi_audio_dma *aud_dma);
+static inline bool hdmi_mode_has_audio(int mode)
+{
+	return mode == HDMI_HDMI ? true : false;
+}
+#endif
+#endif
diff --git a/drivers/video/omap2/dss/hdmi4.c b/drivers/video/omap2/dss/hdmi4.c
new file mode 100644
index 000000000000..e14009614338
--- /dev/null
+++ b/drivers/video/omap2/dss/hdmi4.c
@@ -0,0 +1,696 @@
+/*
+ * HDMI interface DSS driver for TI's OMAP4 family of SoCs.
+ * Copyright (C) 2010-2011 Texas Instruments Incorporated - http://www.ti.com/
+ * Authors: Yong Zhi
+ *	Mythri pk <mythripk@ti.com>
+ *
+ * This program is free software; you can redistribute it and/or modify it
+ * under the terms of the GNU General Public License version 2 as published by
+ * the Free Software Foundation.
+ *
+ * This program is distributed in the hope that it will be useful, but WITHOUT
+ * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or
+ * FITNESS FOR A PARTICULAR PURPOSE.  See the GNU General Public License for
+ * more details.
+ *
+ * You should have received a copy of the GNU General Public License along with
+ * this program.  If not, see <http://www.gnu.org/licenses/>.
+ */
+
+#define DSS_SUBSYS_NAME "HDMI"
+
+#include <linux/kernel.h>
+#include <linux/module.h>
+#include <linux/err.h>
+#include <linux/io.h>
+#include <linux/interrupt.h>
+#include <linux/mutex.h>
+#include <linux/delay.h>
+#include <linux/string.h>
+#include <linux/platform_device.h>
+#include <linux/pm_runtime.h>
+#include <linux/clk.h>
+#include <linux/gpio.h>
+#include <linux/regulator/consumer.h>
+#include <video/omapdss.h>
+
+#include "hdmi4_core.h"
+#include "dss.h"
+#include "dss_features.h"
+
+static struct {
+	struct mutex lock;
+	struct platform_device *pdev;
+
+	struct hdmi_wp_data	wp;
+	struct hdmi_pll_data	pll;
+	struct hdmi_phy_data	phy;
+	struct hdmi_core_data	core;
+
+	struct hdmi_config cfg;
+
+	struct clk *sys_clk;
+	struct regulator *vdda_hdmi_dac_reg;
+
+	bool core_enabled;
+
+	struct omap_dss_device output;
+} hdmi;
+
+static int hdmi_runtime_get(void)
+{
+	int r;
+
+	DSSDBG("hdmi_runtime_get\n");
+
+	r = pm_runtime_get_sync(&hdmi.pdev->dev);
+	WARN_ON(r < 0);
+	if (r < 0)
+		return r;
+
+	return 0;
+}
+
+static void hdmi_runtime_put(void)
+{
+	int r;
+
+	DSSDBG("hdmi_runtime_put\n");
+
+	r = pm_runtime_put_sync(&hdmi.pdev->dev);
+	WARN_ON(r < 0 && r != -ENOSYS);
+}
+
+static int hdmi_init_regulator(void)
+{
+	struct regulator *reg;
+
+	if (hdmi.vdda_hdmi_dac_reg != NULL)
+		return 0;
+
+	reg = devm_regulator_get(&hdmi.pdev->dev, "vdda_hdmi_dac");
+
+	/* DT HACK: try VDAC to make omapdss work for o4 sdp/panda */
+	if (IS_ERR(reg))
+		reg = devm_regulator_get(&hdmi.pdev->dev, "VDAC");
+
+	if (IS_ERR(reg)) {
+		DSSERR("can't get VDDA_HDMI_DAC regulator\n");
+		return PTR_ERR(reg);
+	}
+
+	hdmi.vdda_hdmi_dac_reg = reg;
+
+	return 0;
+}
+
+static int hdmi_power_on_core(struct omap_dss_device *dssdev)
+{
+	int r;
+
+	r = regulator_enable(hdmi.vdda_hdmi_dac_reg);
+	if (r)
+		return r;
+
+	r = hdmi_runtime_get();
+	if (r)
+		goto err_runtime_get;
+
+	/* Make selection of HDMI in DSS */
+	dss_select_hdmi_venc_clk_source(DSS_HDMI_M_PCLK);
+
+	hdmi.core_enabled = true;
+
+	return 0;
+
+err_runtime_get:
+	regulator_disable(hdmi.vdda_hdmi_dac_reg);
+
+	return r;
+}
+
+static void hdmi_power_off_core(struct omap_dss_device *dssdev)
+{
+	hdmi.core_enabled = false;
+
+	hdmi_runtime_put();
+	regulator_disable(hdmi.vdda_hdmi_dac_reg);
+}
+
+static int hdmi_power_on_full(struct omap_dss_device *dssdev)
+{
+	int r;
+	struct omap_video_timings *p;
+	struct omap_overlay_manager *mgr = hdmi.output.manager;
+	unsigned long phy;
+
+	r = hdmi_power_on_core(dssdev);
+	if (r)
+		return r;
+
+	dss_mgr_disable(mgr);
+
+	p = &hdmi.cfg.timings;
+
+	DSSDBG("hdmi_power_on x_res= %d y_res = %d\n", p->x_res, p->y_res);
+
+	phy = p->pixel_clock;
+
+	hdmi_pll_compute(&hdmi.pll, clk_get_rate(hdmi.sys_clk), phy);
+
+	hdmi_wp_video_stop(&hdmi.wp);
+
+	/* config the PLL and PHY hdmi_set_pll_pwrfirst */
+	r = hdmi_pll_enable(&hdmi.pll, &hdmi.wp);
+	if (r) {
+		DSSDBG("Failed to lock PLL\n");
+		goto err_pll_enable;
+	}
+
+	r = hdmi_phy_enable(&hdmi.phy, &hdmi.wp, &hdmi.cfg);
+	if (r) {
+		DSSDBG("Failed to start PHY\n");
+		goto err_phy_enable;
+	}
+
+	hdmi4_configure(&hdmi.core, &hdmi.wp, &hdmi.cfg);
+
+	/* bypass TV gamma table */
+	dispc_enable_gamma_table(0);
+
+	/* tv size */
+	dss_mgr_set_timings(mgr, p);
+
+	r = hdmi_wp_video_start(&hdmi.wp);
+	if (r)
+		goto err_vid_enable;
+
+	r = dss_mgr_enable(mgr);
+	if (r)
+		goto err_mgr_enable;
+
+	return 0;
+
+err_mgr_enable:
+	hdmi_wp_video_stop(&hdmi.wp);
+err_vid_enable:
+	hdmi_phy_disable(&hdmi.phy, &hdmi.wp);
+err_phy_enable:
+	hdmi_pll_disable(&hdmi.pll, &hdmi.wp);
+err_pll_enable:
+	hdmi_power_off_core(dssdev);
+	return -EIO;
+}
+
+static void hdmi_power_off_full(struct omap_dss_device *dssdev)
+{
+	struct omap_overlay_manager *mgr = hdmi.output.manager;
+
+	dss_mgr_disable(mgr);
+
+	hdmi_wp_video_stop(&hdmi.wp);
+	hdmi_phy_disable(&hdmi.phy, &hdmi.wp);
+	hdmi_pll_disable(&hdmi.pll, &hdmi.wp);
+
+	hdmi_power_off_core(dssdev);
+}
+
+static int hdmi_display_check_timing(struct omap_dss_device *dssdev,
+					struct omap_video_timings *timings)
+{
+	struct hdmi_cm cm;
+
+	cm = hdmi_get_code(timings);
+	if (cm.code == -1)
+		return -EINVAL;
+
+	return 0;
+
+}
+
+static void hdmi_display_set_timing(struct omap_dss_device *dssdev,
+		struct omap_video_timings *timings)
+{
+	struct hdmi_cm cm;
+	const struct hdmi_config *t;
+
+	mutex_lock(&hdmi.lock);
+
+	cm = hdmi_get_code(timings);
+	hdmi.cfg.cm = cm;
+
+	t = hdmi_get_timings(cm.mode, cm.code);
+	if (t != NULL) {
+		hdmi.cfg = *t;
+
+		dispc_set_tv_pclk(t->timings.pixel_clock * 1000);
+	}
+
+	mutex_unlock(&hdmi.lock);
+}
+
+static void hdmi_display_get_timings(struct omap_dss_device *dssdev,
+		struct omap_video_timings *timings)
+{
+	const struct hdmi_config *cfg;
+	struct hdmi_cm cm = hdmi.cfg.cm;
+
+	cfg = hdmi_get_timings(cm.mode, cm.code);
+	if (cfg == NULL)
+		cfg = hdmi_default_timing();
+
+	memcpy(timings, &cfg->timings, sizeof(cfg->timings));
+}
+
+static void hdmi_dump_regs(struct seq_file *s)
+{
+	mutex_lock(&hdmi.lock);
+
+	if (hdmi_runtime_get()) {
+		mutex_unlock(&hdmi.lock);
+		return;
+	}
+
+	hdmi_wp_dump(&hdmi.wp, s);
+	hdmi_pll_dump(&hdmi.pll, s);
+	hdmi_phy_dump(&hdmi.phy, s);
+	hdmi4_core_dump(&hdmi.core, s);
+
+	hdmi_runtime_put();
+	mutex_unlock(&hdmi.lock);
+}
+
+static int read_edid(u8 *buf, int len)
+{
+	int r;
+
+	mutex_lock(&hdmi.lock);
+
+	r = hdmi_runtime_get();
+	BUG_ON(r);
+
+	r = hdmi4_read_edid(&hdmi.core,  buf, len);
+
+	hdmi_runtime_put();
+	mutex_unlock(&hdmi.lock);
+
+	return r;
+}
+
+static int hdmi_display_enable(struct omap_dss_device *dssdev)
+{
+	struct omap_dss_device *out = &hdmi.output;
+	int r = 0;
+
+	DSSDBG("ENTER hdmi_display_enable\n");
+
+	mutex_lock(&hdmi.lock);
+
+	if (out == NULL || out->manager == NULL) {
+		DSSERR("failed to enable display: no output/manager\n");
+		r = -ENODEV;
+		goto err0;
+	}
+
+	r = hdmi_power_on_full(dssdev);
+	if (r) {
+		DSSERR("failed to power on device\n");
+		goto err0;
+	}
+
+	mutex_unlock(&hdmi.lock);
+	return 0;
+
+err0:
+	mutex_unlock(&hdmi.lock);
+	return r;
+}
+
+static void hdmi_display_disable(struct omap_dss_device *dssdev)
+{
+	DSSDBG("Enter hdmi_display_disable\n");
+
+	mutex_lock(&hdmi.lock);
+
+	hdmi_power_off_full(dssdev);
+
+	mutex_unlock(&hdmi.lock);
+}
+
+static int hdmi_core_enable(struct omap_dss_device *dssdev)
+{
+	int r = 0;
+
+	DSSDBG("ENTER omapdss_hdmi_core_enable\n");
+
+	mutex_lock(&hdmi.lock);
+
+	r = hdmi_power_on_core(dssdev);
+	if (r) {
+		DSSERR("failed to power on device\n");
+		goto err0;
+	}
+
+	mutex_unlock(&hdmi.lock);
+	return 0;
+
+err0:
+	mutex_unlock(&hdmi.lock);
+	return r;
+}
+
+static void hdmi_core_disable(struct omap_dss_device *dssdev)
+{
+	DSSDBG("Enter omapdss_hdmi_core_disable\n");
+
+	mutex_lock(&hdmi.lock);
+
+	hdmi_power_off_core(dssdev);
+
+	mutex_unlock(&hdmi.lock);
+}
+
+static int hdmi_get_clocks(struct platform_device *pdev)
+{
+	struct clk *clk;
+
+	clk = devm_clk_get(&pdev->dev, "sys_clk");
+	if (IS_ERR(clk)) {
+		DSSERR("can't get sys_clk\n");
+		return PTR_ERR(clk);
+	}
+
+	hdmi.sys_clk = clk;
+
+	return 0;
+}
+
+static int hdmi_connect(struct omap_dss_device *dssdev,
+		struct omap_dss_device *dst)
+{
+	struct omap_overlay_manager *mgr;
+	int r;
+
+	r = hdmi_init_regulator();
+	if (r)
+		return r;
+
+	mgr = omap_dss_get_overlay_manager(dssdev->dispc_channel);
+	if (!mgr)
+		return -ENODEV;
+
+	r = dss_mgr_connect(mgr, dssdev);
+	if (r)
+		return r;
+
+	r = omapdss_output_set_device(dssdev, dst);
+	if (r) {
+		DSSERR("failed to connect output to new device: %s\n",
+				dst->name);
+		dss_mgr_disconnect(mgr, dssdev);
+		return r;
+	}
+
+	return 0;
+}
+
+static void hdmi_disconnect(struct omap_dss_device *dssdev,
+		struct omap_dss_device *dst)
+{
+	WARN_ON(dst != dssdev->dst);
+
+	if (dst != dssdev->dst)
+		return;
+
+	omapdss_output_unset_device(dssdev);
+
+	if (dssdev->manager)
+		dss_mgr_disconnect(dssdev->manager, dssdev);
+}
+
+static int hdmi_read_edid(struct omap_dss_device *dssdev,
+		u8 *edid, int len)
+{
+	bool need_enable;
+	int r;
+
+	need_enable = hdmi.core_enabled == false;
+
+	if (need_enable) {
+		r = hdmi_core_enable(dssdev);
+		if (r)
+			return r;
+	}
+
+	r = read_edid(edid, len);
+
+	if (need_enable)
+		hdmi_core_disable(dssdev);
+
+	return r;
+}
+
+#if defined(CONFIG_OMAP4_DSS_HDMI_AUDIO)
+static int hdmi_audio_enable(struct omap_dss_device *dssdev)
+{
+	int r;
+
+	mutex_lock(&hdmi.lock);
+
+	if (!hdmi_mode_has_audio(hdmi.cfg.cm.mode)) {
+		r = -EPERM;
+		goto err;
+	}
+
+	r = hdmi_wp_audio_enable(&hdmi.wp, true);
+	if (r)
+		goto err;
+
+	mutex_unlock(&hdmi.lock);
+	return 0;
+
+err:
+	mutex_unlock(&hdmi.lock);
+	return r;
+}
+
+static void hdmi_audio_disable(struct omap_dss_device *dssdev)
+{
+	hdmi_wp_audio_enable(&hdmi.wp, false);
+}
+
+static int hdmi_audio_start(struct omap_dss_device *dssdev)
+{
+	return hdmi4_audio_start(&hdmi.core, &hdmi.wp);
+}
+
+static void hdmi_audio_stop(struct omap_dss_device *dssdev)
+{
+	hdmi4_audio_stop(&hdmi.core, &hdmi.wp);
+}
+
+static bool hdmi_audio_supported(struct omap_dss_device *dssdev)
+{
+	bool r;
+
+	mutex_lock(&hdmi.lock);
+
+	r = hdmi_mode_has_audio(hdmi.cfg.cm.mode);
+
+	mutex_unlock(&hdmi.lock);
+	return r;
+}
+
+static int hdmi_audio_config(struct omap_dss_device *dssdev,
+		struct omap_dss_audio *audio)
+{
+	int r;
+	u32 pclk = hdmi.cfg.timings.pixel_clock;
+
+	mutex_lock(&hdmi.lock);
+
+	if (!hdmi_mode_has_audio(hdmi.cfg.cm.mode)) {
+		r = -EPERM;
+		goto err;
+	}
+
+	r = hdmi4_audio_config(&hdmi.core, &hdmi.wp, audio, pclk);
+	if (r)
+		goto err;
+
+	mutex_unlock(&hdmi.lock);
+	return 0;
+
+err:
+	mutex_unlock(&hdmi.lock);
+	return r;
+}
+#else
+static int hdmi_audio_enable(struct omap_dss_device *dssdev)
+{
+	return -EPERM;
+}
+
+static void hdmi_audio_disable(struct omap_dss_device *dssdev)
+{
+}
+
+static int hdmi_audio_start(struct omap_dss_device *dssdev)
+{
+	return -EPERM;
+}
+
+static void hdmi_audio_stop(struct omap_dss_device *dssdev)
+{
+}
+
+static bool hdmi_audio_supported(struct omap_dss_device *dssdev)
+{
+	return false;
+}
+
+static int hdmi_audio_config(struct omap_dss_device *dssdev,
+		struct omap_dss_audio *audio)
+{
+	return -EPERM;
+}
+#endif
+
+static const struct omapdss_hdmi_ops hdmi_ops = {
+	.connect		= hdmi_connect,
+	.disconnect		= hdmi_disconnect,
+
+	.enable			= hdmi_display_enable,
+	.disable		= hdmi_display_disable,
+
+	.check_timings		= hdmi_display_check_timing,
+	.set_timings		= hdmi_display_set_timing,
+	.get_timings		= hdmi_display_get_timings,
+
+	.read_edid		= hdmi_read_edid,
+
+	.audio_enable		= hdmi_audio_enable,
+	.audio_disable		= hdmi_audio_disable,
+	.audio_start		= hdmi_audio_start,
+	.audio_stop		= hdmi_audio_stop,
+	.audio_supported	= hdmi_audio_supported,
+	.audio_config		= hdmi_audio_config,
+};
+
+static void hdmi_init_output(struct platform_device *pdev)
+{
+	struct omap_dss_device *out = &hdmi.output;
+
+	out->dev = &pdev->dev;
+	out->id = OMAP_DSS_OUTPUT_HDMI;
+	out->output_type = OMAP_DISPLAY_TYPE_HDMI;
+	out->name = "hdmi.0";
+	out->dispc_channel = OMAP_DSS_CHANNEL_DIGIT;
+	out->ops.hdmi = &hdmi_ops;
+	out->owner = THIS_MODULE;
+
+	omapdss_register_output(out);
+}
+
+static void __exit hdmi_uninit_output(struct platform_device *pdev)
+{
+	struct omap_dss_device *out = &hdmi.output;
+
+	omapdss_unregister_output(out);
+}
+
+/* HDMI HW IP initialisation */
+static int omapdss_hdmihw_probe(struct platform_device *pdev)
+{
+	int r;
+
+	hdmi.pdev = pdev;
+
+	mutex_init(&hdmi.lock);
+
+	r = hdmi_wp_init(pdev, &hdmi.wp);
+	if (r)
+		return r;
+
+	r = hdmi_pll_init(pdev, &hdmi.pll);
+	if (r)
+		return r;
+
+	r = hdmi_phy_init(pdev, &hdmi.phy);
+	if (r)
+		return r;
+
+	r = hdmi4_core_init(pdev, &hdmi.core);
+	if (r)
+		return r;
+
+	r = hdmi_get_clocks(pdev);
+	if (r) {
+		DSSERR("can't get clocks\n");
+		return r;
+	}
+
+	pm_runtime_enable(&pdev->dev);
+
+	hdmi_init_output(pdev);
+
+	dss_debugfs_create_file("hdmi", hdmi_dump_regs);
+
+	return 0;
+}
+
+static int __exit omapdss_hdmihw_remove(struct platform_device *pdev)
+{
+	hdmi_uninit_output(pdev);
+
+	pm_runtime_disable(&pdev->dev);
+
+	return 0;
+}
+
+static int hdmi_runtime_suspend(struct device *dev)
+{
+	clk_disable_unprepare(hdmi.sys_clk);
+
+	dispc_runtime_put();
+
+	return 0;
+}
+
+static int hdmi_runtime_resume(struct device *dev)
+{
+	int r;
+
+	r = dispc_runtime_get();
+	if (r < 0)
+		return r;
+
+	clk_prepare_enable(hdmi.sys_clk);
+
+	return 0;
+}
+
+static const struct dev_pm_ops hdmi_pm_ops = {
+	.runtime_suspend = hdmi_runtime_suspend,
+	.runtime_resume = hdmi_runtime_resume,
+};
+
+static struct platform_driver omapdss_hdmihw_driver = {
+	.probe		= omapdss_hdmihw_probe,
+	.remove         = __exit_p(omapdss_hdmihw_remove),
+	.driver         = {
+		.name   = "omapdss_hdmi",
+		.owner  = THIS_MODULE,
+		.pm	= &hdmi_pm_ops,
+	},
+};
+
+int __init hdmi4_init_platform_driver(void)
+{
+	return platform_driver_register(&omapdss_hdmihw_driver);
+}
+
+void __exit hdmi4_uninit_platform_driver(void)
+{
+	platform_driver_unregister(&omapdss_hdmihw_driver);
+}
diff --git a/drivers/video/omap2/dss/ti_hdmi_4xxx_ip.c b/drivers/video/omap2/dss/hdmi4_core.c
index 3dfe00956a4f..5dd5e5489b41 100644
--- a/drivers/video/omap2/dss/ti_hdmi_4xxx_ip.c
+++ b/drivers/video/omap2/dss/hdmi4_core.c
@@ -26,6 +26,7 @@
 #include <linux/interrupt.h>
 #include <linux/mutex.h>
 #include <linux/delay.h>
+#include <linux/platform_device.h>
 #include <linux/string.h>
 #include <linux/seq_file.h>
 #if defined(CONFIG_OMAP4_DSS_HDMI_AUDIO)
@@ -33,304 +34,19 @@
 #include <sound/asoundef.h>
 #endif
 
-#include "ti_hdmi_4xxx_ip.h"
-#include "dss.h"
+#include "hdmi4_core.h"
 #include "dss_features.h"
 
-#define HDMI_IRQ_LINK_CONNECT		(1 << 25)
-#define HDMI_IRQ_LINK_DISCONNECT	(1 << 26)
+#define HDMI_CORE_AV		0x500
 
-static inline void hdmi_write_reg(void __iomem *base_addr,
-				const u16 idx, u32 val)
+static inline void __iomem *hdmi_av_base(struct hdmi_core_data *core)
 {
-	__raw_writel(val, base_addr + idx);
+	return core->base + HDMI_CORE_AV;
 }
 
-static inline u32 hdmi_read_reg(void __iomem *base_addr,
-				const u16 idx)
+static int hdmi_core_ddc_init(struct hdmi_core_data *core)
 {
-	return __raw_readl(base_addr + idx);
-}
-
-static inline void __iomem *hdmi_wp_base(struct hdmi_ip_data *ip_data)
-{
-	return ip_data->base_wp;
-}
-
-static inline void __iomem *hdmi_phy_base(struct hdmi_ip_data *ip_data)
-{
-	return ip_data->base_wp + ip_data->phy_offset;
-}
-
-static inline void __iomem *hdmi_pll_base(struct hdmi_ip_data *ip_data)
-{
-	return ip_data->base_wp + ip_data->pll_offset;
-}
-
-static inline void __iomem *hdmi_av_base(struct hdmi_ip_data *ip_data)
-{
-	return ip_data->base_wp + ip_data->core_av_offset;
-}
-
-static inline void __iomem *hdmi_core_sys_base(struct hdmi_ip_data *ip_data)
-{
-	return ip_data->base_wp + ip_data->core_sys_offset;
-}
-
-static inline int hdmi_wait_for_bit_change(void __iomem *base_addr,
-				const u16 idx,
-				int b2, int b1, u32 val)
-{
-	u32 t = 0;
-	while (val != REG_GET(base_addr, idx, b2, b1)) {
-		udelay(1);
-		if (t++ > 10000)
-			return !val;
-	}
-	return val;
-}
-
-static int hdmi_pll_init(struct hdmi_ip_data *ip_data)
-{
-	u32 r;
-	void __iomem *pll_base = hdmi_pll_base(ip_data);
-	struct hdmi_pll_info *fmt = &ip_data->pll_data;
-
-	/* PLL start always use manual mode */
-	REG_FLD_MOD(pll_base, PLLCTRL_PLL_CONTROL, 0x0, 0, 0);
-
-	r = hdmi_read_reg(pll_base, PLLCTRL_CFG1);
-	r = FLD_MOD(r, fmt->regm, 20, 9); /* CFG1_PLL_REGM */
-	r = FLD_MOD(r, fmt->regn - 1, 8, 1);  /* CFG1_PLL_REGN */
-
-	hdmi_write_reg(pll_base, PLLCTRL_CFG1, r);
-
-	r = hdmi_read_reg(pll_base, PLLCTRL_CFG2);
-
-	r = FLD_MOD(r, 0x0, 12, 12); /* PLL_HIGHFREQ divide by 2 */
-	r = FLD_MOD(r, 0x1, 13, 13); /* PLL_REFEN */
-	r = FLD_MOD(r, 0x0, 14, 14); /* PHY_CLKINEN de-assert during locking */
-	r = FLD_MOD(r, fmt->refsel, 22, 21); /* REFSEL */
-
-	if (fmt->dcofreq) {
-		/* divider programming for frequency beyond 1000Mhz */
-		REG_FLD_MOD(pll_base, PLLCTRL_CFG3, fmt->regsd, 17, 10);
-		r = FLD_MOD(r, 0x4, 3, 1); /* 1000MHz and 2000MHz */
-	} else {
-		r = FLD_MOD(r, 0x2, 3, 1); /* 500MHz and 1000MHz */
-	}
-
-	hdmi_write_reg(pll_base, PLLCTRL_CFG2, r);
-
-	r = hdmi_read_reg(pll_base, PLLCTRL_CFG4);
-	r = FLD_MOD(r, fmt->regm2, 24, 18);
-	r = FLD_MOD(r, fmt->regmf, 17, 0);
-
-	hdmi_write_reg(pll_base, PLLCTRL_CFG4, r);
-
-	/* go now */
-	REG_FLD_MOD(pll_base, PLLCTRL_PLL_GO, 0x1, 0, 0);
-
-	/* wait for bit change */
-	if (hdmi_wait_for_bit_change(pll_base, PLLCTRL_PLL_GO,
-							0, 0, 1) != 1) {
-		pr_err("PLL GO bit not set\n");
-		return -ETIMEDOUT;
-	}
-
-	/* Wait till the lock bit is set in PLL status */
-	if (hdmi_wait_for_bit_change(pll_base,
-				PLLCTRL_PLL_STATUS, 1, 1, 1) != 1) {
-		pr_err("cannot lock PLL\n");
-		pr_err("CFG1 0x%x\n",
-			hdmi_read_reg(pll_base, PLLCTRL_CFG1));
-		pr_err("CFG2 0x%x\n",
-			hdmi_read_reg(pll_base, PLLCTRL_CFG2));
-		pr_err("CFG4 0x%x\n",
-			hdmi_read_reg(pll_base, PLLCTRL_CFG4));
-		return -ETIMEDOUT;
-	}
-
-	pr_debug("PLL locked!\n");
-
-	return 0;
-}
-
-/* PHY_PWR_CMD */
-static int hdmi_set_phy_pwr(struct hdmi_ip_data *ip_data, enum hdmi_phy_pwr val)
-{
-	/* Return if already the state */
-	if (REG_GET(hdmi_wp_base(ip_data), HDMI_WP_PWR_CTRL, 5, 4) == val)
-		return 0;
-
-	/* Command for power control of HDMI PHY */
-	REG_FLD_MOD(hdmi_wp_base(ip_data), HDMI_WP_PWR_CTRL, val, 7, 6);
-
-	/* Status of the power control of HDMI PHY */
-	if (hdmi_wait_for_bit_change(hdmi_wp_base(ip_data),
-				HDMI_WP_PWR_CTRL, 5, 4, val) != val) {
-		pr_err("Failed to set PHY power mode to %d\n", val);
-		return -ETIMEDOUT;
-	}
-
-	return 0;
-}
-
-/* PLL_PWR_CMD */
-static int hdmi_set_pll_pwr(struct hdmi_ip_data *ip_data, enum hdmi_pll_pwr val)
-{
-	/* Command for power control of HDMI PLL */
-	REG_FLD_MOD(hdmi_wp_base(ip_data), HDMI_WP_PWR_CTRL, val, 3, 2);
-
-	/* wait till PHY_PWR_STATUS is set */
-	if (hdmi_wait_for_bit_change(hdmi_wp_base(ip_data), HDMI_WP_PWR_CTRL,
-						1, 0, val) != val) {
-		pr_err("Failed to set PLL_PWR_STATUS\n");
-		return -ETIMEDOUT;
-	}
-
-	return 0;
-}
-
-static int hdmi_pll_reset(struct hdmi_ip_data *ip_data)
-{
-	/* SYSRESET  controlled by power FSM */
-	REG_FLD_MOD(hdmi_pll_base(ip_data), PLLCTRL_PLL_CONTROL, 0x0, 3, 3);
-
-	/* READ 0x0 reset is in progress */
-	if (hdmi_wait_for_bit_change(hdmi_pll_base(ip_data),
-				PLLCTRL_PLL_STATUS, 0, 0, 1) != 1) {
-		pr_err("Failed to sysreset PLL\n");
-		return -ETIMEDOUT;
-	}
-
-	return 0;
-}
-
-int ti_hdmi_4xxx_pll_enable(struct hdmi_ip_data *ip_data)
-{
-	u16 r = 0;
-
-	r = hdmi_set_pll_pwr(ip_data, HDMI_PLLPWRCMD_ALLOFF);
-	if (r)
-		return r;
-
-	r = hdmi_set_pll_pwr(ip_data, HDMI_PLLPWRCMD_BOTHON_ALLCLKS);
-	if (r)
-		return r;
-
-	r = hdmi_pll_reset(ip_data);
-	if (r)
-		return r;
-
-	r = hdmi_pll_init(ip_data);
-	if (r)
-		return r;
-
-	return 0;
-}
-
-void ti_hdmi_4xxx_pll_disable(struct hdmi_ip_data *ip_data)
-{
-	hdmi_set_pll_pwr(ip_data, HDMI_PLLPWRCMD_ALLOFF);
-}
-
-static irqreturn_t hdmi_irq_handler(int irq, void *data)
-{
-	struct hdmi_ip_data *ip_data = data;
-	void __iomem *wp_base = hdmi_wp_base(ip_data);
-	u32 irqstatus;
-
-	irqstatus = hdmi_read_reg(wp_base, HDMI_WP_IRQSTATUS);
-	hdmi_write_reg(wp_base, HDMI_WP_IRQSTATUS, irqstatus);
-	/* flush posted write */
-	hdmi_read_reg(wp_base, HDMI_WP_IRQSTATUS);
-
-	if ((irqstatus & HDMI_IRQ_LINK_CONNECT) &&
-			irqstatus & HDMI_IRQ_LINK_DISCONNECT) {
-		/*
-		 * If we get both connect and disconnect interrupts at the same
-		 * time, turn off the PHY, clear interrupts, and restart, which
-		 * raises connect interrupt if a cable is connected, or nothing
-		 * if cable is not connected.
-		 */
-		hdmi_set_phy_pwr(ip_data, HDMI_PHYPWRCMD_OFF);
-
-		hdmi_write_reg(wp_base, HDMI_WP_IRQSTATUS,
-			HDMI_IRQ_LINK_CONNECT | HDMI_IRQ_LINK_DISCONNECT);
-		/* flush posted write */
-		hdmi_read_reg(wp_base, HDMI_WP_IRQSTATUS);
-
-		hdmi_set_phy_pwr(ip_data, HDMI_PHYPWRCMD_LDOON);
-	} else if (irqstatus & HDMI_IRQ_LINK_CONNECT) {
-		hdmi_set_phy_pwr(ip_data, HDMI_PHYPWRCMD_TXON);
-	} else if (irqstatus & HDMI_IRQ_LINK_DISCONNECT) {
-		hdmi_set_phy_pwr(ip_data, HDMI_PHYPWRCMD_LDOON);
-	}
-
-	return IRQ_HANDLED;
-}
-
-int ti_hdmi_4xxx_phy_enable(struct hdmi_ip_data *ip_data)
-{
-	u16 r = 0;
-	void __iomem *phy_base = hdmi_phy_base(ip_data);
-
-	hdmi_write_reg(hdmi_wp_base(ip_data), HDMI_WP_IRQENABLE_CLR,
-			0xffffffff);
-
-	hdmi_write_reg(hdmi_wp_base(ip_data), HDMI_WP_IRQSTATUS,
-			HDMI_IRQ_LINK_CONNECT | HDMI_IRQ_LINK_DISCONNECT);
-
-	r = hdmi_set_phy_pwr(ip_data, HDMI_PHYPWRCMD_LDOON);
-	if (r)
-		return r;
-
-	/*
-	 * Read address 0 in order to get the SCP reset done completed
-	 * Dummy access performed to make sure reset is done
-	 */
-	hdmi_read_reg(phy_base, HDMI_TXPHY_TX_CTRL);
-
-	/*
-	 * Write to phy address 0 to configure the clock
-	 * use HFBITCLK write HDMI_TXPHY_TX_CONTROL_FREQOUT field
-	 */
-	REG_FLD_MOD(phy_base, HDMI_TXPHY_TX_CTRL, 0x1, 31, 30);
-
-	/* Write to phy address 1 to start HDMI line (TXVALID and TMDSCLKEN) */
-	hdmi_write_reg(phy_base, HDMI_TXPHY_DIGITAL_CTRL, 0xF0000000);
-
-	/* Setup max LDO voltage */
-	REG_FLD_MOD(phy_base, HDMI_TXPHY_POWER_CTRL, 0xB, 3, 0);
-
-	/* Write to phy address 3 to change the polarity control */
-	REG_FLD_MOD(phy_base, HDMI_TXPHY_PAD_CFG_CTRL, 0x1, 27, 27);
-
-	r = request_threaded_irq(ip_data->irq, NULL, hdmi_irq_handler,
-				 IRQF_ONESHOT, "OMAP HDMI", ip_data);
-	if (r) {
-		DSSERR("HDMI IRQ request failed\n");
-		hdmi_set_phy_pwr(ip_data, HDMI_PHYPWRCMD_OFF);
-		return r;
-	}
-
-	hdmi_write_reg(hdmi_wp_base(ip_data), HDMI_WP_IRQENABLE_SET,
-			HDMI_IRQ_LINK_CONNECT | HDMI_IRQ_LINK_DISCONNECT);
-
-	return 0;
-}
-
-void ti_hdmi_4xxx_phy_disable(struct hdmi_ip_data *ip_data)
-{
-	free_irq(ip_data->irq, ip_data);
-
-	hdmi_set_phy_pwr(ip_data, HDMI_PHYPWRCMD_OFF);
-}
-
-static int hdmi_core_ddc_init(struct hdmi_ip_data *ip_data)
-{
-	void __iomem *base = hdmi_core_sys_base(ip_data);
+	void __iomem *base = core->base;
 
 	/* Turn on CLK for DDC */
 	REG_FLD_MOD(base, HDMI_CORE_AV_DPD, 0x7, 2, 0);
@@ -370,10 +86,10 @@ static int hdmi_core_ddc_init(struct hdmi_ip_data *ip_data)
 	return 0;
 }
 
-static int hdmi_core_ddc_edid(struct hdmi_ip_data *ip_data,
+static int hdmi_core_ddc_edid(struct hdmi_core_data *core,
 		u8 *pedid, int ext)
 {
-	void __iomem *base = hdmi_core_sys_base(ip_data);
+	void __iomem *base = core->base;
 	u32 i;
 	char checksum;
 	u32 offset = 0;
@@ -452,26 +168,25 @@ static int hdmi_core_ddc_edid(struct hdmi_ip_data *ip_data,
 	return 0;
 }
 
-int ti_hdmi_4xxx_read_edid(struct hdmi_ip_data *ip_data,
-				u8 *edid, int len)
+int hdmi4_read_edid(struct hdmi_core_data *core, u8 *edid, int len)
 {
 	int r, l;
 
 	if (len < 128)
 		return -EINVAL;
 
-	r = hdmi_core_ddc_init(ip_data);
+	r = hdmi_core_ddc_init(core);
 	if (r)
 		return r;
 
-	r = hdmi_core_ddc_edid(ip_data, edid, 0);
+	r = hdmi_core_ddc_edid(core, edid, 0);
 	if (r)
 		return r;
 
 	l = 128;
 
 	if (len >= 128 * 2 && edid[0x7e] > 0) {
-		r = hdmi_core_ddc_edid(ip_data, edid + 0x80, 1);
+		r = hdmi_core_ddc_edid(core, edid + 0x80, 1);
 		if (r)
 			return r;
 		l += 128;
@@ -508,7 +223,7 @@ static void hdmi_core_init(struct hdmi_core_video_config *video_cfg,
 	avi_cfg->db3_nup_scaling = 0;
 	avi_cfg->db4_videocode = 0;
 	avi_cfg->db5_pixel_repeat = 0;
-	avi_cfg->db6_7_line_eoftop = 0 ;
+	avi_cfg->db6_7_line_eoftop = 0;
 	avi_cfg->db8_9_line_sofbottom = 0;
 	avi_cfg->db10_11_pixel_eofleft = 0;
 	avi_cfg->db12_13_pixel_sofright = 0;
@@ -524,38 +239,39 @@ static void hdmi_core_init(struct hdmi_core_video_config *video_cfg,
 	repeat_cfg->generic_pkt_repeat = 0;
 }
 
-static void hdmi_core_powerdown_disable(struct hdmi_ip_data *ip_data)
+static void hdmi_core_powerdown_disable(struct hdmi_core_data *core)
 {
 	pr_debug("Enter hdmi_core_powerdown_disable\n");
-	REG_FLD_MOD(hdmi_core_sys_base(ip_data), HDMI_CORE_CTRL1, 0x0, 0, 0);
+	REG_FLD_MOD(core->base, HDMI_CORE_SYS_SYS_CTRL1, 0x0, 0, 0);
 }
 
-static void hdmi_core_swreset_release(struct hdmi_ip_data *ip_data)
+static void hdmi_core_swreset_release(struct hdmi_core_data *core)
 {
 	pr_debug("Enter hdmi_core_swreset_release\n");
-	REG_FLD_MOD(hdmi_core_sys_base(ip_data), HDMI_CORE_SYS_SRST, 0x0, 0, 0);
+	REG_FLD_MOD(core->base, HDMI_CORE_SYS_SRST, 0x0, 0, 0);
 }
 
-static void hdmi_core_swreset_assert(struct hdmi_ip_data *ip_data)
+static void hdmi_core_swreset_assert(struct hdmi_core_data *core)
 {
 	pr_debug("Enter hdmi_core_swreset_assert\n");
-	REG_FLD_MOD(hdmi_core_sys_base(ip_data), HDMI_CORE_SYS_SRST, 0x1, 0, 0);
+	REG_FLD_MOD(core->base, HDMI_CORE_SYS_SRST, 0x1, 0, 0);
 }
 
 /* HDMI_CORE_VIDEO_CONFIG */
-static void hdmi_core_video_config(struct hdmi_ip_data *ip_data,
+static void hdmi_core_video_config(struct hdmi_core_data *core,
 				struct hdmi_core_video_config *cfg)
 {
 	u32 r = 0;
-	void __iomem *core_sys_base = hdmi_core_sys_base(ip_data);
+	void __iomem *core_sys_base = core->base;
+	void __iomem *core_av_base = hdmi_av_base(core);
 
 	/* sys_ctrl1 default configuration not tunable */
-	r = hdmi_read_reg(core_sys_base, HDMI_CORE_CTRL1);
-	r = FLD_MOD(r, HDMI_CORE_CTRL1_VEN_FOLLOWVSYNC, 5, 5);
-	r = FLD_MOD(r, HDMI_CORE_CTRL1_HEN_FOLLOWHSYNC, 4, 4);
-	r = FLD_MOD(r, HDMI_CORE_CTRL1_BSEL_24BITBUS, 2, 2);
-	r = FLD_MOD(r, HDMI_CORE_CTRL1_EDGE_RISINGEDGE, 1, 1);
-	hdmi_write_reg(core_sys_base, HDMI_CORE_CTRL1, r);
+	r = hdmi_read_reg(core_sys_base, HDMI_CORE_SYS_SYS_CTRL1);
+	r = FLD_MOD(r, HDMI_CORE_SYS_SYS_CTRL1_VEN_FOLLOWVSYNC, 5, 5);
+	r = FLD_MOD(r, HDMI_CORE_SYS_SYS_CTRL1_HEN_FOLLOWHSYNC, 4, 4);
+	r = FLD_MOD(r, HDMI_CORE_SYS_SYS_CTRL1_BSEL_24BITBUS, 2, 2);
+	r = FLD_MOD(r, HDMI_CORE_SYS_SYS_CTRL1_EDGE_RISINGEDGE, 1, 1);
+	hdmi_write_reg(core_sys_base, HDMI_CORE_SYS_SYS_CTRL1, r);
 
 	REG_FLD_MOD(core_sys_base,
 			HDMI_CORE_SYS_VID_ACEN, cfg->ip_bus_width, 7, 6);
@@ -574,23 +290,23 @@ static void hdmi_core_video_config(struct hdmi_ip_data *ip_data,
 	hdmi_write_reg(core_sys_base, HDMI_CORE_SYS_VID_MODE, r);
 
 	/* HDMI_Ctrl */
-	r = hdmi_read_reg(hdmi_av_base(ip_data), HDMI_CORE_AV_HDMI_CTRL);
+	r = hdmi_read_reg(core_av_base, HDMI_CORE_AV_HDMI_CTRL);
 	r = FLD_MOD(r, cfg->deep_color_pkt, 6, 6);
 	r = FLD_MOD(r, cfg->pkt_mode, 5, 3);
 	r = FLD_MOD(r, cfg->hdmi_dvi, 0, 0);
-	hdmi_write_reg(hdmi_av_base(ip_data), HDMI_CORE_AV_HDMI_CTRL, r);
+	hdmi_write_reg(core_av_base, HDMI_CORE_AV_HDMI_CTRL, r);
 
 	/* TMDS_CTRL */
 	REG_FLD_MOD(core_sys_base,
 			HDMI_CORE_SYS_TMDS_CTRL, cfg->tclk_sel_clkmult, 6, 5);
 }
 
-static void hdmi_core_aux_infoframe_avi_config(struct hdmi_ip_data *ip_data)
+static void hdmi_core_aux_infoframe_avi_config(struct hdmi_core_data *core)
 {
 	u32 val;
 	char sum = 0, checksum = 0;
-	void __iomem *av_base = hdmi_av_base(ip_data);
-	struct hdmi_core_infoframe_avi info_avi = ip_data->avi_cfg;
+	void __iomem *av_base = hdmi_av_base(core);
+	struct hdmi_core_infoframe_avi info_avi = core->avi_cfg;
 
 	sum += 0x82 + 0x002 + 0x00D;
 	hdmi_write_reg(av_base, HDMI_CORE_AV_AVI_TYPE, 0x082);
@@ -661,160 +377,64 @@ static void hdmi_core_aux_infoframe_avi_config(struct hdmi_ip_data *ip_data)
 	hdmi_write_reg(av_base, HDMI_CORE_AV_AVI_CHSUM, checksum);
 }
 
-static void hdmi_core_av_packet_config(struct hdmi_ip_data *ip_data,
+static void hdmi_core_av_packet_config(struct hdmi_core_data *core,
 		struct hdmi_core_packet_enable_repeat repeat_cfg)
 {
 	/* enable/repeat the infoframe */
-	hdmi_write_reg(hdmi_av_base(ip_data), HDMI_CORE_AV_PB_CTRL1,
+	hdmi_write_reg(hdmi_av_base(core), HDMI_CORE_AV_PB_CTRL1,
 		(repeat_cfg.audio_pkt << 5) |
 		(repeat_cfg.audio_pkt_repeat << 4) |
 		(repeat_cfg.avi_infoframe << 1) |
 		(repeat_cfg.avi_infoframe_repeat));
 
 	/* enable/repeat the packet */
-	hdmi_write_reg(hdmi_av_base(ip_data), HDMI_CORE_AV_PB_CTRL2,
+	hdmi_write_reg(hdmi_av_base(core), HDMI_CORE_AV_PB_CTRL2,
 		(repeat_cfg.gen_cntrl_pkt << 3) |
 		(repeat_cfg.gen_cntrl_pkt_repeat << 2) |
 		(repeat_cfg.generic_pkt << 1) |
 		(repeat_cfg.generic_pkt_repeat));
 }
 
-static void hdmi_wp_init(struct omap_video_timings *timings,
-			struct hdmi_video_format *video_fmt)
-{
-	pr_debug("Enter hdmi_wp_init\n");
-
-	timings->hbp = 0;
-	timings->hfp = 0;
-	timings->hsw = 0;
-	timings->vbp = 0;
-	timings->vfp = 0;
-	timings->vsw = 0;
-
-	video_fmt->packing_mode = HDMI_PACK_10b_RGB_YUV444;
-	video_fmt->y_res = 0;
-	video_fmt->x_res = 0;
-
-}
-
-int ti_hdmi_4xxx_wp_video_start(struct hdmi_ip_data *ip_data)
-{
-	REG_FLD_MOD(hdmi_wp_base(ip_data), HDMI_WP_VIDEO_CFG, true, 31, 31);
-	return 0;
-}
-
-void ti_hdmi_4xxx_wp_video_stop(struct hdmi_ip_data *ip_data)
-{
-	REG_FLD_MOD(hdmi_wp_base(ip_data), HDMI_WP_VIDEO_CFG, false, 31, 31);
-}
-
-static void hdmi_wp_video_init_format(struct hdmi_video_format *video_fmt,
-	struct omap_video_timings *timings, struct hdmi_config *param)
-{
-	pr_debug("Enter hdmi_wp_video_init_format\n");
-
-	video_fmt->y_res = param->timings.y_res;
-	video_fmt->x_res = param->timings.x_res;
-
-	timings->hbp = param->timings.hbp;
-	timings->hfp = param->timings.hfp;
-	timings->hsw = param->timings.hsw;
-	timings->vbp = param->timings.vbp;
-	timings->vfp = param->timings.vfp;
-	timings->vsw = param->timings.vsw;
-}
-
-static void hdmi_wp_video_config_format(struct hdmi_ip_data *ip_data,
-		struct hdmi_video_format *video_fmt)
-{
-	u32 l = 0;
-
-	REG_FLD_MOD(hdmi_wp_base(ip_data), HDMI_WP_VIDEO_CFG,
-			video_fmt->packing_mode, 10, 8);
-
-	l |= FLD_VAL(video_fmt->y_res, 31, 16);
-	l |= FLD_VAL(video_fmt->x_res, 15, 0);
-	hdmi_write_reg(hdmi_wp_base(ip_data), HDMI_WP_VIDEO_SIZE, l);
-}
-
-static void hdmi_wp_video_config_interface(struct hdmi_ip_data *ip_data)
-{
-	u32 r;
-	bool vsync_pol, hsync_pol;
-	pr_debug("Enter hdmi_wp_video_config_interface\n");
-
-	vsync_pol = ip_data->cfg.timings.vsync_level == OMAPDSS_SIG_ACTIVE_HIGH;
-	hsync_pol = ip_data->cfg.timings.hsync_level == OMAPDSS_SIG_ACTIVE_HIGH;
-
-	r = hdmi_read_reg(hdmi_wp_base(ip_data), HDMI_WP_VIDEO_CFG);
-	r = FLD_MOD(r, vsync_pol, 7, 7);
-	r = FLD_MOD(r, hsync_pol, 6, 6);
-	r = FLD_MOD(r, ip_data->cfg.timings.interlace, 3, 3);
-	r = FLD_MOD(r, 1, 1, 0); /* HDMI_TIMING_MASTER_24BIT */
-	hdmi_write_reg(hdmi_wp_base(ip_data), HDMI_WP_VIDEO_CFG, r);
-}
-
-static void hdmi_wp_video_config_timing(struct hdmi_ip_data *ip_data,
-		struct omap_video_timings *timings)
-{
-	u32 timing_h = 0;
-	u32 timing_v = 0;
-
-	pr_debug("Enter hdmi_wp_video_config_timing\n");
-
-	timing_h |= FLD_VAL(timings->hbp, 31, 20);
-	timing_h |= FLD_VAL(timings->hfp, 19, 8);
-	timing_h |= FLD_VAL(timings->hsw, 7, 0);
-	hdmi_write_reg(hdmi_wp_base(ip_data), HDMI_WP_VIDEO_TIMING_H, timing_h);
-
-	timing_v |= FLD_VAL(timings->vbp, 31, 20);
-	timing_v |= FLD_VAL(timings->vfp, 19, 8);
-	timing_v |= FLD_VAL(timings->vsw, 7, 0);
-	hdmi_write_reg(hdmi_wp_base(ip_data), HDMI_WP_VIDEO_TIMING_V, timing_v);
-}
-
-void ti_hdmi_4xxx_basic_configure(struct hdmi_ip_data *ip_data)
+void hdmi4_configure(struct hdmi_core_data *core,
+	struct hdmi_wp_data *wp, struct hdmi_config *cfg)
 {
 	/* HDMI */
 	struct omap_video_timings video_timing;
 	struct hdmi_video_format video_format;
 	/* HDMI core */
-	struct hdmi_core_infoframe_avi *avi_cfg = &ip_data->avi_cfg;
+	struct hdmi_core_infoframe_avi *avi_cfg = &core->avi_cfg;
 	struct hdmi_core_video_config v_core_cfg;
 	struct hdmi_core_packet_enable_repeat repeat_cfg;
-	struct hdmi_config *cfg = &ip_data->cfg;
-
-	hdmi_wp_init(&video_timing, &video_format);
 
 	hdmi_core_init(&v_core_cfg, avi_cfg, &repeat_cfg);
 
-	hdmi_wp_video_init_format(&video_format, &video_timing, cfg);
+	hdmi_wp_init_vid_fmt_timings(&video_format, &video_timing, cfg);
 
-	hdmi_wp_video_config_timing(ip_data, &video_timing);
+	hdmi_wp_video_config_timing(wp, &video_timing);
 
 	/* video config */
 	video_format.packing_mode = HDMI_PACK_24b_RGB_YUV444_YUV422;
 
-	hdmi_wp_video_config_format(ip_data, &video_format);
+	hdmi_wp_video_config_format(wp, &video_format);
 
-	hdmi_wp_video_config_interface(ip_data);
+	hdmi_wp_video_config_interface(wp, &video_timing);
 
 	/*
 	 * configure core video part
 	 * set software reset in the core
 	 */
-	hdmi_core_swreset_assert(ip_data);
+	hdmi_core_swreset_assert(core);
 
 	/* power down off */
-	hdmi_core_powerdown_disable(ip_data);
+	hdmi_core_powerdown_disable(core);
 
 	v_core_cfg.pkt_mode = HDMI_PACKETMODE24BITPERPIXEL;
 	v_core_cfg.hdmi_dvi = cfg->cm.mode;
 
-	hdmi_core_video_config(ip_data, &v_core_cfg);
+	hdmi_core_video_config(core, &v_core_cfg);
 
 	/* release software reset in the core */
-	hdmi_core_swreset_release(ip_data);
+	hdmi_core_swreset_release(core);
 
 	/*
 	 * configure packet
@@ -839,7 +459,7 @@ void ti_hdmi_4xxx_basic_configure(struct hdmi_ip_data *ip_data)
 	avi_cfg->db10_11_pixel_eofleft = 0;
 	avi_cfg->db12_13_pixel_sofright = 0;
 
-	hdmi_core_aux_infoframe_avi_config(ip_data);
+	hdmi_core_aux_infoframe_avi_config(core);
 
 	/* enable/repeat the infoframe */
 	repeat_cfg.avi_infoframe = HDMI_PACKETENABLE;
@@ -847,65 +467,30 @@ void ti_hdmi_4xxx_basic_configure(struct hdmi_ip_data *ip_data)
 	/* wakeup */
 	repeat_cfg.audio_pkt = HDMI_PACKETENABLE;
 	repeat_cfg.audio_pkt_repeat = HDMI_PACKETREPEATON;
-	hdmi_core_av_packet_config(ip_data, repeat_cfg);
-}
-
-void ti_hdmi_4xxx_wp_dump(struct hdmi_ip_data *ip_data, struct seq_file *s)
-{
-#define DUMPREG(r) seq_printf(s, "%-35s %08x\n", #r,\
-		hdmi_read_reg(hdmi_wp_base(ip_data), r))
-
-	DUMPREG(HDMI_WP_REVISION);
-	DUMPREG(HDMI_WP_SYSCONFIG);
-	DUMPREG(HDMI_WP_IRQSTATUS_RAW);
-	DUMPREG(HDMI_WP_IRQSTATUS);
-	DUMPREG(HDMI_WP_PWR_CTRL);
-	DUMPREG(HDMI_WP_IRQENABLE_SET);
-	DUMPREG(HDMI_WP_VIDEO_CFG);
-	DUMPREG(HDMI_WP_VIDEO_SIZE);
-	DUMPREG(HDMI_WP_VIDEO_TIMING_H);
-	DUMPREG(HDMI_WP_VIDEO_TIMING_V);
-	DUMPREG(HDMI_WP_WP_CLK);
-	DUMPREG(HDMI_WP_AUDIO_CFG);
-	DUMPREG(HDMI_WP_AUDIO_CFG2);
-	DUMPREG(HDMI_WP_AUDIO_CTRL);
-	DUMPREG(HDMI_WP_AUDIO_DATA);
-}
-
-void ti_hdmi_4xxx_pll_dump(struct hdmi_ip_data *ip_data, struct seq_file *s)
-{
-#define DUMPPLL(r) seq_printf(s, "%-35s %08x\n", #r,\
-		hdmi_read_reg(hdmi_pll_base(ip_data), r))
-
-	DUMPPLL(PLLCTRL_PLL_CONTROL);
-	DUMPPLL(PLLCTRL_PLL_STATUS);
-	DUMPPLL(PLLCTRL_PLL_GO);
-	DUMPPLL(PLLCTRL_CFG1);
-	DUMPPLL(PLLCTRL_CFG2);
-	DUMPPLL(PLLCTRL_CFG3);
-	DUMPPLL(PLLCTRL_CFG4);
+	hdmi_core_av_packet_config(core, repeat_cfg);
 }
 
-void ti_hdmi_4xxx_core_dump(struct hdmi_ip_data *ip_data, struct seq_file *s)
+void hdmi4_core_dump(struct hdmi_core_data *core, struct seq_file *s)
 {
 	int i;
 
 #define CORE_REG(i, name) name(i)
 #define DUMPCORE(r) seq_printf(s, "%-35s %08x\n", #r,\
-		hdmi_read_reg(hdmi_core_sys_base(ip_data), r))
+		hdmi_read_reg(core->base, r))
 #define DUMPCOREAV(r) seq_printf(s, "%-35s %08x\n", #r,\
-		hdmi_read_reg(hdmi_av_base(ip_data), r))
+		hdmi_read_reg(hdmi_av_base(core), r))
 #define DUMPCOREAV2(i, r) seq_printf(s, "%s[%d]%*s %08x\n", #r, i, \
 		(i < 10) ? 32 - (int)strlen(#r) : 31 - (int)strlen(#r), " ", \
-		hdmi_read_reg(hdmi_av_base(ip_data), CORE_REG(i, r)))
+		hdmi_read_reg(hdmi_av_base(core), CORE_REG(i, r)))
 
 	DUMPCORE(HDMI_CORE_SYS_VND_IDL);
 	DUMPCORE(HDMI_CORE_SYS_DEV_IDL);
 	DUMPCORE(HDMI_CORE_SYS_DEV_IDH);
 	DUMPCORE(HDMI_CORE_SYS_DEV_REV);
 	DUMPCORE(HDMI_CORE_SYS_SRST);
-	DUMPCORE(HDMI_CORE_CTRL1);
+	DUMPCORE(HDMI_CORE_SYS_SYS_CTRL1);
 	DUMPCORE(HDMI_CORE_SYS_SYS_STAT);
+	DUMPCORE(HDMI_CORE_SYS_SYS_CTRL3);
 	DUMPCORE(HDMI_CORE_SYS_DE_DLY);
 	DUMPCORE(HDMI_CORE_SYS_DE_CTRL);
 	DUMPCORE(HDMI_CORE_SYS_DE_TOP);
@@ -913,14 +498,58 @@ void ti_hdmi_4xxx_core_dump(struct hdmi_ip_data *ip_data, struct seq_file *s)
 	DUMPCORE(HDMI_CORE_SYS_DE_CNTH);
 	DUMPCORE(HDMI_CORE_SYS_DE_LINL);
 	DUMPCORE(HDMI_CORE_SYS_DE_LINH_1);
+	DUMPCORE(HDMI_CORE_SYS_HRES_L);
+	DUMPCORE(HDMI_CORE_SYS_HRES_H);
+	DUMPCORE(HDMI_CORE_SYS_VRES_L);
+	DUMPCORE(HDMI_CORE_SYS_VRES_H);
+	DUMPCORE(HDMI_CORE_SYS_IADJUST);
+	DUMPCORE(HDMI_CORE_SYS_POLDETECT);
+	DUMPCORE(HDMI_CORE_SYS_HWIDTH1);
+	DUMPCORE(HDMI_CORE_SYS_HWIDTH2);
+	DUMPCORE(HDMI_CORE_SYS_VWIDTH);
+	DUMPCORE(HDMI_CORE_SYS_VID_CTRL);
 	DUMPCORE(HDMI_CORE_SYS_VID_ACEN);
 	DUMPCORE(HDMI_CORE_SYS_VID_MODE);
+	DUMPCORE(HDMI_CORE_SYS_VID_BLANK1);
+	DUMPCORE(HDMI_CORE_SYS_VID_BLANK3);
+	DUMPCORE(HDMI_CORE_SYS_VID_BLANK1);
+	DUMPCORE(HDMI_CORE_SYS_DC_HEADER);
+	DUMPCORE(HDMI_CORE_SYS_VID_DITHER);
+	DUMPCORE(HDMI_CORE_SYS_RGB2XVYCC_CT);
+	DUMPCORE(HDMI_CORE_SYS_R2Y_COEFF_LOW);
+	DUMPCORE(HDMI_CORE_SYS_R2Y_COEFF_UP);
+	DUMPCORE(HDMI_CORE_SYS_G2Y_COEFF_LOW);
+	DUMPCORE(HDMI_CORE_SYS_G2Y_COEFF_UP);
+	DUMPCORE(HDMI_CORE_SYS_B2Y_COEFF_LOW);
+	DUMPCORE(HDMI_CORE_SYS_B2Y_COEFF_UP);
+	DUMPCORE(HDMI_CORE_SYS_R2CB_COEFF_LOW);
+	DUMPCORE(HDMI_CORE_SYS_R2CB_COEFF_UP);
+	DUMPCORE(HDMI_CORE_SYS_G2CB_COEFF_LOW);
+	DUMPCORE(HDMI_CORE_SYS_G2CB_COEFF_UP);
+	DUMPCORE(HDMI_CORE_SYS_B2CB_COEFF_LOW);
+	DUMPCORE(HDMI_CORE_SYS_B2CB_COEFF_UP);
+	DUMPCORE(HDMI_CORE_SYS_R2CR_COEFF_LOW);
+	DUMPCORE(HDMI_CORE_SYS_R2CR_COEFF_UP);
+	DUMPCORE(HDMI_CORE_SYS_G2CR_COEFF_LOW);
+	DUMPCORE(HDMI_CORE_SYS_G2CR_COEFF_UP);
+	DUMPCORE(HDMI_CORE_SYS_B2CR_COEFF_LOW);
+	DUMPCORE(HDMI_CORE_SYS_B2CR_COEFF_UP);
+	DUMPCORE(HDMI_CORE_SYS_RGB_OFFSET_LOW);
+	DUMPCORE(HDMI_CORE_SYS_RGB_OFFSET_UP);
+	DUMPCORE(HDMI_CORE_SYS_Y_OFFSET_LOW);
+	DUMPCORE(HDMI_CORE_SYS_Y_OFFSET_UP);
+	DUMPCORE(HDMI_CORE_SYS_CBCR_OFFSET_LOW);
+	DUMPCORE(HDMI_CORE_SYS_CBCR_OFFSET_UP);
 	DUMPCORE(HDMI_CORE_SYS_INTR_STATE);
 	DUMPCORE(HDMI_CORE_SYS_INTR1);
 	DUMPCORE(HDMI_CORE_SYS_INTR2);
 	DUMPCORE(HDMI_CORE_SYS_INTR3);
 	DUMPCORE(HDMI_CORE_SYS_INTR4);
-	DUMPCORE(HDMI_CORE_SYS_UMASK1);
+	DUMPCORE(HDMI_CORE_SYS_INTR_UNMASK1);
+	DUMPCORE(HDMI_CORE_SYS_INTR_UNMASK2);
+	DUMPCORE(HDMI_CORE_SYS_INTR_UNMASK3);
+	DUMPCORE(HDMI_CORE_SYS_INTR_UNMASK4);
+	DUMPCORE(HDMI_CORE_SYS_INTR_CTRL);
 	DUMPCORE(HDMI_CORE_SYS_TMDS_CTRL);
 
 	DUMPCORE(HDMI_CORE_DDC_ADDR);
@@ -1009,60 +638,12 @@ void ti_hdmi_4xxx_core_dump(struct hdmi_ip_data *ip_data, struct seq_file *s)
 	DUMPCOREAV(HDMI_CORE_AV_CEC_ADDR_ID);
 }
 
-void ti_hdmi_4xxx_phy_dump(struct hdmi_ip_data *ip_data, struct seq_file *s)
-{
-#define DUMPPHY(r) seq_printf(s, "%-35s %08x\n", #r,\
-		hdmi_read_reg(hdmi_phy_base(ip_data), r))
-
-	DUMPPHY(HDMI_TXPHY_TX_CTRL);
-	DUMPPHY(HDMI_TXPHY_DIGITAL_CTRL);
-	DUMPPHY(HDMI_TXPHY_POWER_CTRL);
-	DUMPPHY(HDMI_TXPHY_PAD_CFG_CTRL);
-}
-
 #if defined(CONFIG_OMAP4_DSS_HDMI_AUDIO)
-static void ti_hdmi_4xxx_wp_audio_config_format(struct hdmi_ip_data *ip_data,
-					struct hdmi_audio_format *aud_fmt)
-{
-	u32 r;
-
-	DSSDBG("Enter hdmi_wp_audio_config_format\n");
-
-	r = hdmi_read_reg(hdmi_wp_base(ip_data), HDMI_WP_AUDIO_CFG);
-	r = FLD_MOD(r, aud_fmt->stereo_channels, 26, 24);
-	r = FLD_MOD(r, aud_fmt->active_chnnls_msk, 23, 16);
-	r = FLD_MOD(r, aud_fmt->en_sig_blk_strt_end, 5, 5);
-	r = FLD_MOD(r, aud_fmt->type, 4, 4);
-	r = FLD_MOD(r, aud_fmt->justification, 3, 3);
-	r = FLD_MOD(r, aud_fmt->sample_order, 2, 2);
-	r = FLD_MOD(r, aud_fmt->samples_per_word, 1, 1);
-	r = FLD_MOD(r, aud_fmt->sample_size, 0, 0);
-	hdmi_write_reg(hdmi_wp_base(ip_data), HDMI_WP_AUDIO_CFG, r);
-}
-
-static void ti_hdmi_4xxx_wp_audio_config_dma(struct hdmi_ip_data *ip_data,
-					struct hdmi_audio_dma *aud_dma)
-{
-	u32 r;
-
-	DSSDBG("Enter hdmi_wp_audio_config_dma\n");
-
-	r = hdmi_read_reg(hdmi_wp_base(ip_data), HDMI_WP_AUDIO_CFG2);
-	r = FLD_MOD(r, aud_dma->transfer_size, 15, 8);
-	r = FLD_MOD(r, aud_dma->block_size, 7, 0);
-	hdmi_write_reg(hdmi_wp_base(ip_data), HDMI_WP_AUDIO_CFG2, r);
-
-	r = hdmi_read_reg(hdmi_wp_base(ip_data), HDMI_WP_AUDIO_CTRL);
-	r = FLD_MOD(r, aud_dma->mode, 9, 9);
-	r = FLD_MOD(r, aud_dma->fifo_threshold, 8, 0);
-	hdmi_write_reg(hdmi_wp_base(ip_data), HDMI_WP_AUDIO_CTRL, r);
-}
-
-static void ti_hdmi_4xxx_core_audio_config(struct hdmi_ip_data *ip_data,
+static void hdmi_core_audio_config(struct hdmi_core_data *core,
 					struct hdmi_core_audio_config *cfg)
 {
 	u32 r;
-	void __iomem *av_base = hdmi_av_base(ip_data);
+	void __iomem *av_base = hdmi_av_base(core);
 
 	/*
 	 * Parameters for generation of Audio Clock Recovery packets
@@ -1157,11 +738,11 @@ static void ti_hdmi_4xxx_core_audio_config(struct hdmi_ip_data *ip_data,
 	REG_FLD_MOD(av_base, HDMI_CORE_AV_SWAP_I2S, 1, 5, 5);
 }
 
-static void ti_hdmi_4xxx_core_audio_infoframe_cfg(struct hdmi_ip_data *ip_data,
+static void hdmi_core_audio_infoframe_cfg(struct hdmi_core_data *core,
 		struct snd_cea_861_aud_if *info_aud)
 {
 	u8 sum = 0, checksum = 0;
-	void __iomem *av_base = hdmi_av_base(ip_data);
+	void __iomem *av_base = hdmi_av_base(core);
 
 	/*
 	 * Set audio info frame type, version and length as
@@ -1207,20 +788,20 @@ static void ti_hdmi_4xxx_core_audio_infoframe_cfg(struct hdmi_ip_data *ip_data,
 	 */
 }
 
-int ti_hdmi_4xxx_audio_config(struct hdmi_ip_data *ip_data,
-		struct omap_dss_audio *audio)
+int hdmi4_audio_config(struct hdmi_core_data *core, struct hdmi_wp_data *wp,
+		struct omap_dss_audio *audio, u32 pclk)
 {
 	struct hdmi_audio_format audio_format;
 	struct hdmi_audio_dma audio_dma;
-	struct hdmi_core_audio_config core;
+	struct hdmi_core_audio_config acore;
 	int err, n, cts, channel_count;
 	unsigned int fs_nr;
 	bool word_length_16b = false;
 
-	if (!audio || !audio->iec || !audio->cea || !ip_data)
+	if (!audio || !audio->iec || !audio->cea || !core)
 		return -EINVAL;
 
-	core.iec60958_cfg = audio->iec;
+	acore.iec60958_cfg = audio->iec;
 	/*
 	 * In the IEC-60958 status word, check if the audio sample word length
 	 * is 16-bit as several optimizations can be performed in such case.
@@ -1231,22 +812,22 @@ int ti_hdmi_4xxx_audio_config(struct hdmi_ip_data *ip_data,
 
 	/* I2S configuration. See Phillips' specification */
 	if (word_length_16b)
-		core.i2s_cfg.justification = HDMI_AUDIO_JUSTIFY_LEFT;
+		acore.i2s_cfg.justification = HDMI_AUDIO_JUSTIFY_LEFT;
 	else
-		core.i2s_cfg.justification = HDMI_AUDIO_JUSTIFY_RIGHT;
+		acore.i2s_cfg.justification = HDMI_AUDIO_JUSTIFY_RIGHT;
 	/*
 	 * The I2S input word length is twice the lenght given in the IEC-60958
 	 * status word. If the word size is greater than
 	 * 20 bits, increment by one.
 	 */
-	core.i2s_cfg.in_length_bits = audio->iec->status[4]
+	acore.i2s_cfg.in_length_bits = audio->iec->status[4]
 		& IEC958_AES4_CON_WORDLEN;
 	if (audio->iec->status[4] & IEC958_AES4_CON_MAX_WORDLEN_24)
-		core.i2s_cfg.in_length_bits++;
-	core.i2s_cfg.sck_edge_mode = HDMI_AUDIO_I2S_SCK_EDGE_RISING;
-	core.i2s_cfg.vbit = HDMI_AUDIO_I2S_VBIT_FOR_PCM;
-	core.i2s_cfg.direction = HDMI_AUDIO_I2S_MSB_SHIFTED_FIRST;
-	core.i2s_cfg.shift = HDMI_AUDIO_I2S_FIRST_BIT_SHIFT;
+		acore.i2s_cfg.in_length_bits++;
+	acore.i2s_cfg.sck_edge_mode = HDMI_AUDIO_I2S_SCK_EDGE_RISING;
+	acore.i2s_cfg.vbit = HDMI_AUDIO_I2S_VBIT_FOR_PCM;
+	acore.i2s_cfg.direction = HDMI_AUDIO_I2S_MSB_SHIFTED_FIRST;
+	acore.i2s_cfg.shift = HDMI_AUDIO_I2S_FIRST_BIT_SHIFT;
 
 	/* convert sample frequency to a number */
 	switch (audio->iec->status[3] & IEC958_AES3_CON_FS) {
@@ -1275,23 +856,23 @@ int ti_hdmi_4xxx_audio_config(struct hdmi_ip_data *ip_data,
 		return -EINVAL;
 	}
 
-	err = hdmi_compute_acr(fs_nr, &n, &cts);
+	err = hdmi_compute_acr(pclk, fs_nr, &n, &cts);
 
 	/* Audio clock regeneration settings */
-	core.n = n;
-	core.cts = cts;
+	acore.n = n;
+	acore.cts = cts;
 	if (dss_has_feature(FEAT_HDMI_CTS_SWMODE)) {
-		core.aud_par_busclk = 0;
-		core.cts_mode = HDMI_AUDIO_CTS_MODE_SW;
-		core.use_mclk = dss_has_feature(FEAT_HDMI_AUDIO_USE_MCLK);
+		acore.aud_par_busclk = 0;
+		acore.cts_mode = HDMI_AUDIO_CTS_MODE_SW;
+		acore.use_mclk = dss_has_feature(FEAT_HDMI_AUDIO_USE_MCLK);
 	} else {
-		core.aud_par_busclk = (((128 * 31) - 1) << 8);
-		core.cts_mode = HDMI_AUDIO_CTS_MODE_HW;
-		core.use_mclk = true;
+		acore.aud_par_busclk = (((128 * 31) - 1) << 8);
+		acore.cts_mode = HDMI_AUDIO_CTS_MODE_HW;
+		acore.use_mclk = true;
 	}
 
-	if (core.use_mclk)
-		core.mclk_mode = HDMI_AUDIO_MCLK_128FS;
+	if (acore.use_mclk)
+		acore.mclk_mode = HDMI_AUDIO_MCLK_128FS;
 
 	/* Audio channels settings */
 	channel_count = (audio->cea->db1_ct_cc &
@@ -1329,25 +910,25 @@ int ti_hdmi_4xxx_audio_config(struct hdmi_ip_data *ip_data,
 	 */
 	if (channel_count == 2) {
 		audio_format.stereo_channels = HDMI_AUDIO_STEREO_ONECHANNEL;
-		core.i2s_cfg.active_sds = HDMI_AUDIO_I2S_SD0_EN;
-		core.layout = HDMI_AUDIO_LAYOUT_2CH;
+		acore.i2s_cfg.active_sds = HDMI_AUDIO_I2S_SD0_EN;
+		acore.layout = HDMI_AUDIO_LAYOUT_2CH;
 	} else {
 		audio_format.stereo_channels = HDMI_AUDIO_STEREO_FOURCHANNELS;
-		core.i2s_cfg.active_sds = HDMI_AUDIO_I2S_SD0_EN |
+		acore.i2s_cfg.active_sds = HDMI_AUDIO_I2S_SD0_EN |
 				HDMI_AUDIO_I2S_SD1_EN | HDMI_AUDIO_I2S_SD2_EN |
 				HDMI_AUDIO_I2S_SD3_EN;
-		core.layout = HDMI_AUDIO_LAYOUT_8CH;
+		acore.layout = HDMI_AUDIO_LAYOUT_8CH;
 	}
 
-	core.en_spdif = false;
+	acore.en_spdif = false;
 	/* use sample frequency from channel status word */
-	core.fs_override = true;
+	acore.fs_override = true;
 	/* enable ACR packets */
-	core.en_acr_pkt = true;
+	acore.en_acr_pkt = true;
 	/* disable direct streaming digital audio */
-	core.en_dsd_audio = false;
+	acore.en_dsd_audio = false;
 	/* use parallel audio interface */
-	core.en_parallel_aud_input = true;
+	acore.en_parallel_aud_input = true;
 
 	/* DMA settings */
 	if (word_length_16b)
@@ -1374,49 +955,37 @@ int ti_hdmi_4xxx_audio_config(struct hdmi_ip_data *ip_data,
 	audio_format.en_sig_blk_strt_end = HDMI_AUDIO_BLOCK_SIG_STARTEND_ON;
 
 	/* configure DMA and audio FIFO format*/
-	ti_hdmi_4xxx_wp_audio_config_dma(ip_data, &audio_dma);
-	ti_hdmi_4xxx_wp_audio_config_format(ip_data, &audio_format);
+	hdmi_wp_audio_config_dma(wp, &audio_dma);
+	hdmi_wp_audio_config_format(wp, &audio_format);
 
 	/* configure the core*/
-	ti_hdmi_4xxx_core_audio_config(ip_data, &core);
+	hdmi_core_audio_config(core, &acore);
 
 	/* configure CEA 861 audio infoframe*/
-	ti_hdmi_4xxx_core_audio_infoframe_cfg(ip_data, audio->cea);
+	hdmi_core_audio_infoframe_cfg(core, audio->cea);
 
 	return 0;
 }
 
-int ti_hdmi_4xxx_wp_audio_enable(struct hdmi_ip_data *ip_data)
+int hdmi4_audio_start(struct hdmi_core_data *core, struct hdmi_wp_data *wp)
 {
-	REG_FLD_MOD(hdmi_wp_base(ip_data),
-		    HDMI_WP_AUDIO_CTRL, true, 31, 31);
-	return 0;
-}
+	REG_FLD_MOD(hdmi_av_base(core),
+		    HDMI_CORE_AV_AUD_MODE, true, 0, 0);
 
-void ti_hdmi_4xxx_wp_audio_disable(struct hdmi_ip_data *ip_data)
-{
-	REG_FLD_MOD(hdmi_wp_base(ip_data),
-		    HDMI_WP_AUDIO_CTRL, false, 31, 31);
-}
+	hdmi_wp_audio_core_req_enable(wp, true);
 
-int ti_hdmi_4xxx_audio_start(struct hdmi_ip_data *ip_data)
-{
-	REG_FLD_MOD(hdmi_av_base(ip_data),
-		    HDMI_CORE_AV_AUD_MODE, true, 0, 0);
-	REG_FLD_MOD(hdmi_wp_base(ip_data),
-		    HDMI_WP_AUDIO_CTRL, true, 30, 30);
 	return 0;
 }
 
-void ti_hdmi_4xxx_audio_stop(struct hdmi_ip_data *ip_data)
+void hdmi4_audio_stop(struct hdmi_core_data *core, struct hdmi_wp_data *wp)
 {
-	REG_FLD_MOD(hdmi_av_base(ip_data),
+	REG_FLD_MOD(hdmi_av_base(core),
 		    HDMI_CORE_AV_AUD_MODE, false, 0, 0);
-	REG_FLD_MOD(hdmi_wp_base(ip_data),
-		    HDMI_WP_AUDIO_CTRL, false, 30, 30);
+
+	hdmi_wp_audio_core_req_enable(wp, false);
 }
 
-int ti_hdmi_4xxx_audio_get_dma_port(u32 *offset, u32 *size)
+int hdmi4_audio_get_dma_port(u32 *offset, u32 *size)
 {
 	if (!offset || !size)
 		return -EINVAL;
@@ -1424,4 +993,42 @@ int ti_hdmi_4xxx_audio_get_dma_port(u32 *offset, u32 *size)
 	*size = 4;
 	return 0;
 }
+
 #endif
+
+#define CORE_OFFSET		0x400
+#define CORE_SIZE		0xc00
+
+int hdmi4_core_init(struct platform_device *pdev, struct hdmi_core_data *core)
+{
+	struct resource *res;
+	struct resource temp_res;
+
+	res = platform_get_resource_byname(pdev, IORESOURCE_MEM, "hdmi_core");
+	if (!res) {
+		DSSDBG("can't get CORE mem resource by name\n");
+		/*
+		 * if hwmod/DT doesn't have the memory resource information
+		 * split into HDMI sub blocks by name, we try again by getting
+		 * the platform's first resource. this code will be removed when
+		 * the driver can get the mem resources by name
+		 */
+		res = platform_get_resource(pdev, IORESOURCE_MEM, 0);
+		if (!res) {
+			DSSERR("can't get CORE mem resource\n");
+			return -EINVAL;
+		}
+
+		temp_res.start = res->start + CORE_OFFSET;
+		temp_res.end = temp_res.start + CORE_SIZE - 1;
+		res = &temp_res;
+	}
+
+	core->base = devm_ioremap(&pdev->dev, res->start, resource_size(res));
+	if (!core->base) {
+		DSSERR("can't ioremap CORE\n");
+		return -ENOMEM;
+	}
+
+	return 0;
+}
diff --git a/drivers/video/omap2/dss/ti_hdmi_4xxx_ip.h b/drivers/video/omap2/dss/hdmi4_core.h
index 6ef2f929a76d..bb646896fa82 100644
--- a/drivers/video/omap2/dss/ti_hdmi_4xxx_ip.h
+++ b/drivers/video/omap2/dss/hdmi4_core.h
@@ -1,7 +1,5 @@
 /*
- * ti_hdmi_4xxx_ip.h
- *
- * HDMI header definition for DM81xx, DM38xx, TI OMAP4 etc processors.
+ * HDMI header definition for OMAP4 HDMI core IP
  *
  * Copyright (C) 2010-2011 Texas Instruments Incorporated - http://www.ti.com/
  *
@@ -18,41 +16,22 @@
  * this program.  If not, see <http://www.gnu.org/licenses/>.
  */
 
-#ifndef _HDMI_TI_4xxx_H_
-#define _HDMI_TI_4xxx_H_
-
-#include <linux/string.h>
-#include <video/omapdss.h>
-#include "ti_hdmi.h"
-
-/* HDMI Wrapper */
+#ifndef _HDMI4_CORE_H_
+#define _HDMI4_CORE_H_
 
-#define HDMI_WP_REVISION			0x0
-#define HDMI_WP_SYSCONFIG			0x10
-#define HDMI_WP_IRQSTATUS_RAW			0x24
-#define HDMI_WP_IRQSTATUS			0x28
-#define HDMI_WP_PWR_CTRL			0x40
-#define HDMI_WP_IRQENABLE_SET			0x2C
-#define HDMI_WP_IRQENABLE_CLR			0x30
-#define HDMI_WP_VIDEO_CFG			0x50
-#define HDMI_WP_VIDEO_SIZE			0x60
-#define HDMI_WP_VIDEO_TIMING_H			0x68
-#define HDMI_WP_VIDEO_TIMING_V			0x6C
-#define HDMI_WP_WP_CLK				0x70
-#define HDMI_WP_AUDIO_CFG			0x80
-#define HDMI_WP_AUDIO_CFG2			0x84
-#define HDMI_WP_AUDIO_CTRL			0x88
-#define HDMI_WP_AUDIO_DATA			0x8C
+#include "hdmi.h"
 
-/* HDMI IP Core System */
+/* OMAP4 HDMI IP Core System */
 
 #define HDMI_CORE_SYS_VND_IDL			0x0
 #define HDMI_CORE_SYS_DEV_IDL			0x8
 #define HDMI_CORE_SYS_DEV_IDH			0xC
 #define HDMI_CORE_SYS_DEV_REV			0x10
 #define HDMI_CORE_SYS_SRST			0x14
-#define HDMI_CORE_CTRL1				0x20
+#define HDMI_CORE_SYS_SYS_CTRL1			0x20
 #define HDMI_CORE_SYS_SYS_STAT			0x24
+#define HDMI_CORE_SYS_SYS_CTRL3			0x28
+#define HDMI_CORE_SYS_DCTL			0x34
 #define HDMI_CORE_SYS_DE_DLY			0xC8
 #define HDMI_CORE_SYS_DE_CTRL			0xCC
 #define HDMI_CORE_SYS_DE_TOP			0xD0
@@ -60,20 +39,65 @@
 #define HDMI_CORE_SYS_DE_CNTH			0xDC
 #define HDMI_CORE_SYS_DE_LINL			0xE0
 #define HDMI_CORE_SYS_DE_LINH_1			0xE4
+#define HDMI_CORE_SYS_HRES_L			0xE8
+#define HDMI_CORE_SYS_HRES_H			0xEC
+#define HDMI_CORE_SYS_VRES_L			0xF0
+#define HDMI_CORE_SYS_VRES_H			0xF4
+#define HDMI_CORE_SYS_IADJUST			0xF8
+#define HDMI_CORE_SYS_POLDETECT			0xFC
+#define HDMI_CORE_SYS_HWIDTH1			0x110
+#define HDMI_CORE_SYS_HWIDTH2			0x114
+#define HDMI_CORE_SYS_VWIDTH			0x11C
+#define HDMI_CORE_SYS_VID_CTRL			0x120
 #define HDMI_CORE_SYS_VID_ACEN			0x124
 #define HDMI_CORE_SYS_VID_MODE			0x128
+#define HDMI_CORE_SYS_VID_BLANK1		0x12C
+#define HDMI_CORE_SYS_VID_BLANK2		0x130
+#define HDMI_CORE_SYS_VID_BLANK3		0x134
+#define HDMI_CORE_SYS_DC_HEADER			0x138
+#define HDMI_CORE_SYS_VID_DITHER		0x13C
+#define HDMI_CORE_SYS_RGB2XVYCC_CT		0x140
+#define HDMI_CORE_SYS_R2Y_COEFF_LOW		0x144
+#define HDMI_CORE_SYS_R2Y_COEFF_UP		0x148
+#define HDMI_CORE_SYS_G2Y_COEFF_LOW		0x14C
+#define HDMI_CORE_SYS_G2Y_COEFF_UP		0x150
+#define HDMI_CORE_SYS_B2Y_COEFF_LOW		0x154
+#define HDMI_CORE_SYS_B2Y_COEFF_UP		0x158
+#define HDMI_CORE_SYS_R2CB_COEFF_LOW		0x15C
+#define HDMI_CORE_SYS_R2CB_COEFF_UP		0x160
+#define HDMI_CORE_SYS_G2CB_COEFF_LOW		0x164
+#define HDMI_CORE_SYS_G2CB_COEFF_UP		0x168
+#define HDMI_CORE_SYS_B2CB_COEFF_LOW		0x16C
+#define HDMI_CORE_SYS_B2CB_COEFF_UP		0x170
+#define HDMI_CORE_SYS_R2CR_COEFF_LOW		0x174
+#define HDMI_CORE_SYS_R2CR_COEFF_UP		0x178
+#define HDMI_CORE_SYS_G2CR_COEFF_LOW		0x17C
+#define HDMI_CORE_SYS_G2CR_COEFF_UP		0x180
+#define HDMI_CORE_SYS_B2CR_COEFF_LOW		0x184
+#define HDMI_CORE_SYS_B2CR_COEFF_UP		0x188
+#define HDMI_CORE_SYS_RGB_OFFSET_LOW		0x18C
+#define HDMI_CORE_SYS_RGB_OFFSET_UP		0x190
+#define HDMI_CORE_SYS_Y_OFFSET_LOW		0x194
+#define HDMI_CORE_SYS_Y_OFFSET_UP		0x198
+#define HDMI_CORE_SYS_CBCR_OFFSET_LOW		0x19C
+#define HDMI_CORE_SYS_CBCR_OFFSET_UP		0x1A0
 #define HDMI_CORE_SYS_INTR_STATE		0x1C0
 #define HDMI_CORE_SYS_INTR1			0x1C4
 #define HDMI_CORE_SYS_INTR2			0x1C8
 #define HDMI_CORE_SYS_INTR3			0x1CC
 #define HDMI_CORE_SYS_INTR4			0x1D0
-#define HDMI_CORE_SYS_UMASK1			0x1D4
+#define HDMI_CORE_SYS_INTR_UNMASK1		0x1D4
+#define HDMI_CORE_SYS_INTR_UNMASK2		0x1D8
+#define HDMI_CORE_SYS_INTR_UNMASK3		0x1DC
+#define HDMI_CORE_SYS_INTR_UNMASK4		0x1E0
+#define HDMI_CORE_SYS_INTR_CTRL			0x1E4
 #define HDMI_CORE_SYS_TMDS_CTRL			0x208
 
-#define HDMI_CORE_CTRL1_VEN_FOLLOWVSYNC	0x1
-#define HDMI_CORE_CTRL1_HEN_FOLLOWHSYNC	0x1
-#define HDMI_CORE_CTRL1_BSEL_24BITBUS	0x1
-#define HDMI_CORE_CTRL1_EDGE_RISINGEDGE	0x1
+/* value definitions for HDMI_CORE_SYS_SYS_CTRL1 fields */
+#define HDMI_CORE_SYS_SYS_CTRL1_VEN_FOLLOWVSYNC	0x1
+#define HDMI_CORE_SYS_SYS_CTRL1_HEN_FOLLOWHSYNC	0x1
+#define HDMI_CORE_SYS_SYS_CTRL1_BSEL_24BITBUS	0x1
+#define HDMI_CORE_SYS_SYS_CTRL1_EDGE_RISINGEDGE	0x1
 
 /* HDMI DDC E-DID */
 #define HDMI_CORE_DDC_ADDR			0x3B4
@@ -158,35 +182,6 @@
 #define HDMI_CORE_AV_GEN_DBYTE_NELEMS		31
 #define HDMI_CORE_AV_GEN2_DBYTE_NELEMS		31
 
-/* PLL */
-
-#define PLLCTRL_PLL_CONTROL			0x0
-#define PLLCTRL_PLL_STATUS			0x4
-#define PLLCTRL_PLL_GO				0x8
-#define PLLCTRL_CFG1				0xC
-#define PLLCTRL_CFG2				0x10
-#define PLLCTRL_CFG3				0x14
-#define PLLCTRL_CFG4				0x20
-
-/* HDMI PHY */
-
-#define HDMI_TXPHY_TX_CTRL			0x0
-#define HDMI_TXPHY_DIGITAL_CTRL			0x4
-#define HDMI_TXPHY_POWER_CTRL			0x8
-#define HDMI_TXPHY_PAD_CFG_CTRL			0xC
-
-#define REG_FLD_MOD(base, idx, val, start, end) \
-	hdmi_write_reg(base, idx, FLD_MOD(hdmi_read_reg(base, idx),\
-							val, start, end))
-#define REG_GET(base, idx, start, end) \
-	FLD_GET(hdmi_read_reg(base, idx), start, end)
-
-enum hdmi_phy_pwr {
-	HDMI_PHYPWRCMD_OFF = 0,
-	HDMI_PHYPWRCMD_LDOON = 1,
-	HDMI_PHYPWRCMD_TXON = 2
-};
-
 enum hdmi_core_inputbus_width {
 	HDMI_INPUT_8BIT = 0,
 	HDMI_INPUT_10BIT = 1,
@@ -229,114 +224,6 @@ enum hdmi_core_packet_ctrl {
 	HDMI_PACKETREPEATOFF = 0
 };
 
-/* INFOFRAME_AVI_ and INFOFRAME_AUDIO_ definitions */
-enum hdmi_core_infoframe {
-	HDMI_INFOFRAME_AVI_DB1Y_RGB = 0,
-	HDMI_INFOFRAME_AVI_DB1Y_YUV422 = 1,
-	HDMI_INFOFRAME_AVI_DB1Y_YUV444 = 2,
-	HDMI_INFOFRAME_AVI_DB1A_ACTIVE_FORMAT_OFF = 0,
-	HDMI_INFOFRAME_AVI_DB1A_ACTIVE_FORMAT_ON =  1,
-	HDMI_INFOFRAME_AVI_DB1B_NO = 0,
-	HDMI_INFOFRAME_AVI_DB1B_VERT = 1,
-	HDMI_INFOFRAME_AVI_DB1B_HORI = 2,
-	HDMI_INFOFRAME_AVI_DB1B_VERTHORI = 3,
-	HDMI_INFOFRAME_AVI_DB1S_0 = 0,
-	HDMI_INFOFRAME_AVI_DB1S_1 = 1,
-	HDMI_INFOFRAME_AVI_DB1S_2 = 2,
-	HDMI_INFOFRAME_AVI_DB2C_NO = 0,
-	HDMI_INFOFRAME_AVI_DB2C_ITU601 = 1,
-	HDMI_INFOFRAME_AVI_DB2C_ITU709 = 2,
-	HDMI_INFOFRAME_AVI_DB2C_EC_EXTENDED = 3,
-	HDMI_INFOFRAME_AVI_DB2M_NO = 0,
-	HDMI_INFOFRAME_AVI_DB2M_43 = 1,
-	HDMI_INFOFRAME_AVI_DB2M_169 = 2,
-	HDMI_INFOFRAME_AVI_DB2R_SAME = 8,
-	HDMI_INFOFRAME_AVI_DB2R_43 = 9,
-	HDMI_INFOFRAME_AVI_DB2R_169 = 10,
-	HDMI_INFOFRAME_AVI_DB2R_149 = 11,
-	HDMI_INFOFRAME_AVI_DB3ITC_NO = 0,
-	HDMI_INFOFRAME_AVI_DB3ITC_YES = 1,
-	HDMI_INFOFRAME_AVI_DB3EC_XVYUV601 = 0,
-	HDMI_INFOFRAME_AVI_DB3EC_XVYUV709 = 1,
-	HDMI_INFOFRAME_AVI_DB3Q_DEFAULT = 0,
-	HDMI_INFOFRAME_AVI_DB3Q_LR = 1,
-	HDMI_INFOFRAME_AVI_DB3Q_FR = 2,
-	HDMI_INFOFRAME_AVI_DB3SC_NO = 0,
-	HDMI_INFOFRAME_AVI_DB3SC_HORI = 1,
-	HDMI_INFOFRAME_AVI_DB3SC_VERT = 2,
-	HDMI_INFOFRAME_AVI_DB3SC_HORIVERT = 3,
-	HDMI_INFOFRAME_AVI_DB5PR_NO = 0,
-	HDMI_INFOFRAME_AVI_DB5PR_2 = 1,
-	HDMI_INFOFRAME_AVI_DB5PR_3 = 2,
-	HDMI_INFOFRAME_AVI_DB5PR_4 = 3,
-	HDMI_INFOFRAME_AVI_DB5PR_5 = 4,
-	HDMI_INFOFRAME_AVI_DB5PR_6 = 5,
-	HDMI_INFOFRAME_AVI_DB5PR_7 = 6,
-	HDMI_INFOFRAME_AVI_DB5PR_8 = 7,
-	HDMI_INFOFRAME_AVI_DB5PR_9 = 8,
-	HDMI_INFOFRAME_AVI_DB5PR_10 = 9,
-};
-
-enum hdmi_packing_mode {
-	HDMI_PACK_10b_RGB_YUV444 = 0,
-	HDMI_PACK_24b_RGB_YUV444_YUV422 = 1,
-	HDMI_PACK_20b_YUV422 = 2,
-	HDMI_PACK_ALREADYPACKED = 7
-};
-
-enum hdmi_core_audio_layout {
-	HDMI_AUDIO_LAYOUT_2CH = 0,
-	HDMI_AUDIO_LAYOUT_8CH = 1
-};
-
-enum hdmi_core_cts_mode {
-	HDMI_AUDIO_CTS_MODE_HW = 0,
-	HDMI_AUDIO_CTS_MODE_SW = 1
-};
-
-enum hdmi_stereo_channels {
-	HDMI_AUDIO_STEREO_NOCHANNELS = 0,
-	HDMI_AUDIO_STEREO_ONECHANNEL = 1,
-	HDMI_AUDIO_STEREO_TWOCHANNELS = 2,
-	HDMI_AUDIO_STEREO_THREECHANNELS = 3,
-	HDMI_AUDIO_STEREO_FOURCHANNELS = 4
-};
-
-enum hdmi_audio_type {
-	HDMI_AUDIO_TYPE_LPCM = 0,
-	HDMI_AUDIO_TYPE_IEC = 1
-};
-
-enum hdmi_audio_justify {
-	HDMI_AUDIO_JUSTIFY_LEFT = 0,
-	HDMI_AUDIO_JUSTIFY_RIGHT = 1
-};
-
-enum hdmi_audio_sample_order {
-	HDMI_AUDIO_SAMPLE_RIGHT_FIRST = 0,
-	HDMI_AUDIO_SAMPLE_LEFT_FIRST = 1
-};
-
-enum hdmi_audio_samples_perword {
-	HDMI_AUDIO_ONEWORD_ONESAMPLE = 0,
-	HDMI_AUDIO_ONEWORD_TWOSAMPLES = 1
-};
-
-enum hdmi_audio_sample_size {
-	HDMI_AUDIO_SAMPLE_16BITS = 0,
-	HDMI_AUDIO_SAMPLE_24BITS = 1
-};
-
-enum hdmi_audio_transf_mode {
-	HDMI_AUDIO_TRANSF_DMA = 0,
-	HDMI_AUDIO_TRANSF_IRQ = 1
-};
-
-enum hdmi_audio_blk_strt_end_sig {
-	HDMI_AUDIO_BLOCK_SIG_STARTEND_ON = 0,
-	HDMI_AUDIO_BLOCK_SIG_STARTEND_OFF = 1
-};
-
 enum hdmi_audio_i2s_config {
 	HDMI_AUDIO_I2S_MSB_SHIFTED_FIRST = 0,
 	HDMI_AUDIO_I2S_LSB_SHIFTED_FIRST = 1,
@@ -352,17 +239,6 @@ enum hdmi_audio_i2s_config {
 	HDMI_AUDIO_I2S_SD3_EN = 1 << 3,
 };
 
-enum hdmi_audio_mclk_mode {
-	HDMI_AUDIO_MCLK_128FS = 0,
-	HDMI_AUDIO_MCLK_256FS = 1,
-	HDMI_AUDIO_MCLK_384FS = 2,
-	HDMI_AUDIO_MCLK_512FS = 3,
-	HDMI_AUDIO_MCLK_768FS = 4,
-	HDMI_AUDIO_MCLK_1024FS = 5,
-	HDMI_AUDIO_MCLK_1152FS = 6,
-	HDMI_AUDIO_MCLK_192FS = 7
-};
-
 struct hdmi_core_video_config {
 	enum hdmi_core_inputbus_width	ip_bus_width;
 	enum hdmi_core_dither_trunc	op_dither_truc;
@@ -383,55 +259,18 @@ struct hdmi_core_packet_enable_repeat {
 	u32	generic_pkt_repeat;
 };
 
-struct hdmi_video_format {
-	enum hdmi_packing_mode	packing_mode;
-	u32			y_res;	/* Line per panel */
-	u32			x_res;	/* pixel per line */
-};
-
-struct hdmi_audio_format {
-	enum hdmi_stereo_channels		stereo_channels;
-	u8					active_chnnls_msk;
-	enum hdmi_audio_type			type;
-	enum hdmi_audio_justify			justification;
-	enum hdmi_audio_sample_order		sample_order;
-	enum hdmi_audio_samples_perword		samples_per_word;
-	enum hdmi_audio_sample_size		sample_size;
-	enum hdmi_audio_blk_strt_end_sig	en_sig_blk_strt_end;
-};
-
-struct hdmi_audio_dma {
-	u8				transfer_size;
-	u8				block_size;
-	enum hdmi_audio_transf_mode	mode;
-	u16				fifo_threshold;
-};
-
-struct hdmi_core_audio_i2s_config {
-	u8 in_length_bits;
-	u8 justification;
-	u8 sck_edge_mode;
-	u8 vbit;
-	u8 direction;
-	u8 shift;
-	u8 active_sds;
-};
-
-struct hdmi_core_audio_config {
-	struct hdmi_core_audio_i2s_config	i2s_cfg;
-	struct snd_aes_iec958			*iec60958_cfg;
-	bool					fs_override;
-	u32					n;
-	u32					cts;
-	u32					aud_par_busclk;
-	enum hdmi_core_audio_layout		layout;
-	enum hdmi_core_cts_mode			cts_mode;
-	bool					use_mclk;
-	enum hdmi_audio_mclk_mode		mclk_mode;
-	bool					en_acr_pkt;
-	bool					en_dsd_audio;
-	bool					en_parallel_aud_input;
-	bool					en_spdif;
-};
+int hdmi4_read_edid(struct hdmi_core_data *core, u8 *edid, int len);
+void hdmi4_configure(struct hdmi_core_data *core, struct hdmi_wp_data *wp,
+		struct hdmi_config *cfg);
+void hdmi4_core_dump(struct hdmi_core_data *core, struct seq_file *s);
+int hdmi4_core_init(struct platform_device *pdev, struct hdmi_core_data *core);
+
+#if defined(CONFIG_OMAP4_DSS_HDMI_AUDIO)
+int hdmi4_audio_start(struct hdmi_core_data *core, struct hdmi_wp_data *wp);
+void hdmi4_audio_stop(struct hdmi_core_data *core, struct hdmi_wp_data *wp);
+int hdmi4_audio_config(struct hdmi_core_data *core, struct hdmi_wp_data *wp,
+		struct omap_dss_audio *audio, u32 pclk);
+int hdmi4_audio_get_dma_port(u32 *offset, u32 *size);
+#endif
 
 #endif
diff --git a/drivers/video/omap2/dss/hdmi_common.c b/drivers/video/omap2/dss/hdmi_common.c
new file mode 100644
index 000000000000..5586aaad9d63
--- /dev/null
+++ b/drivers/video/omap2/dss/hdmi_common.c
@@ -0,0 +1,423 @@
+
+/*
+ * Logic for the below structure :
+ * user enters the CEA or VESA timings by specifying the HDMI/DVI code.
+ * There is a correspondence between CEA/VESA timing and code, please
+ * refer to section 6.3 in HDMI 1.3 specification for timing code.
+ *
+ * In the below structure, cea_vesa_timings corresponds to all OMAP4
+ * supported CEA and VESA timing values.code_cea corresponds to the CEA
+ * code, It is used to get the timing from cea_vesa_timing array.Similarly
+ * with code_vesa. Code_index is used for back mapping, that is once EDID
+ * is read from the TV, EDID is parsed to find the timing values and then
+ * map it to corresponding CEA or VESA index.
+ */
+
+#include <linux/kernel.h>
+#include <linux/err.h>
+#include <video/omapdss.h>
+
+#include "hdmi.h"
+
+static const struct hdmi_config cea_timings[] = {
+	{
+		{ 640, 480, 25200, 96, 16, 48, 2, 10, 33,
+			OMAPDSS_SIG_ACTIVE_LOW, OMAPDSS_SIG_ACTIVE_LOW,
+			false, },
+		{ 1, HDMI_HDMI },
+	},
+	{
+		{ 720, 480, 27027, 62, 16, 60, 6, 9, 30,
+			OMAPDSS_SIG_ACTIVE_LOW, OMAPDSS_SIG_ACTIVE_LOW,
+			false, },
+		{ 2, HDMI_HDMI },
+	},
+	{
+		{ 1280, 720, 74250, 40, 110, 220, 5, 5, 20,
+			OMAPDSS_SIG_ACTIVE_HIGH, OMAPDSS_SIG_ACTIVE_HIGH,
+			false, },
+		{ 4, HDMI_HDMI },
+	},
+	{
+		{ 1920, 540, 74250, 44, 88, 148, 5, 2, 15,
+			OMAPDSS_SIG_ACTIVE_HIGH, OMAPDSS_SIG_ACTIVE_HIGH,
+			true, },
+		{ 5, HDMI_HDMI },
+	},
+	{
+		{ 1440, 240, 27027, 124, 38, 114, 3, 4, 15,
+			OMAPDSS_SIG_ACTIVE_LOW, OMAPDSS_SIG_ACTIVE_LOW,
+			true, },
+		{ 6, HDMI_HDMI },
+	},
+	{
+		{ 1920, 1080, 148500, 44, 88, 148, 5, 4, 36,
+			OMAPDSS_SIG_ACTIVE_HIGH, OMAPDSS_SIG_ACTIVE_HIGH,
+			false, },
+		{ 16, HDMI_HDMI },
+	},
+	{
+		{ 720, 576, 27000, 64, 12, 68, 5, 5, 39,
+			OMAPDSS_SIG_ACTIVE_LOW, OMAPDSS_SIG_ACTIVE_LOW,
+			false, },
+		{ 17, HDMI_HDMI },
+	},
+	{
+		{ 1280, 720, 74250, 40, 440, 220, 5, 5, 20,
+			OMAPDSS_SIG_ACTIVE_HIGH, OMAPDSS_SIG_ACTIVE_HIGH,
+			false, },
+		{ 19, HDMI_HDMI },
+	},
+	{
+		{ 1920, 540, 74250, 44, 528, 148, 5, 2, 15,
+			OMAPDSS_SIG_ACTIVE_HIGH, OMAPDSS_SIG_ACTIVE_HIGH,
+			true, },
+		{ 20, HDMI_HDMI },
+	},
+	{
+		{ 1440, 288, 27000, 126, 24, 138, 3, 2, 19,
+			OMAPDSS_SIG_ACTIVE_LOW, OMAPDSS_SIG_ACTIVE_LOW,
+			true, },
+		{ 21, HDMI_HDMI },
+	},
+	{
+		{ 1440, 576, 54000, 128, 24, 136, 5, 5, 39,
+			OMAPDSS_SIG_ACTIVE_LOW, OMAPDSS_SIG_ACTIVE_LOW,
+			false, },
+		{ 29, HDMI_HDMI },
+	},
+	{
+		{ 1920, 1080, 148500, 44, 528, 148, 5, 4, 36,
+			OMAPDSS_SIG_ACTIVE_HIGH, OMAPDSS_SIG_ACTIVE_HIGH,
+			false, },
+		{ 31, HDMI_HDMI },
+	},
+	{
+		{ 1920, 1080, 74250, 44, 638, 148, 5, 4, 36,
+			OMAPDSS_SIG_ACTIVE_HIGH, OMAPDSS_SIG_ACTIVE_HIGH,
+			false, },
+		{ 32, HDMI_HDMI },
+	},
+	{
+		{ 2880, 480, 108108, 248, 64, 240, 6, 9, 30,
+			OMAPDSS_SIG_ACTIVE_LOW, OMAPDSS_SIG_ACTIVE_LOW,
+			false, },
+		{ 35, HDMI_HDMI },
+	},
+	{
+		{ 2880, 576, 108000, 256, 48, 272, 5, 5, 39,
+			OMAPDSS_SIG_ACTIVE_LOW, OMAPDSS_SIG_ACTIVE_LOW,
+			false, },
+		{ 37, HDMI_HDMI },
+	},
+};
+
+static const struct hdmi_config vesa_timings[] = {
+/* VESA From Here */
+	{
+		{ 640, 480, 25175, 96, 16, 48, 2, 11, 31,
+			OMAPDSS_SIG_ACTIVE_LOW, OMAPDSS_SIG_ACTIVE_LOW,
+			false, },
+		{ 4, HDMI_DVI },
+	},
+	{
+		{ 800, 600, 40000, 128, 40, 88, 4, 1, 23,
+			OMAPDSS_SIG_ACTIVE_HIGH, OMAPDSS_SIG_ACTIVE_HIGH,
+			false, },
+		{ 9, HDMI_DVI },
+	},
+	{
+		{ 848, 480, 33750, 112, 16, 112, 8, 6, 23,
+			OMAPDSS_SIG_ACTIVE_HIGH, OMAPDSS_SIG_ACTIVE_HIGH,
+			false, },
+		{ 0xE, HDMI_DVI },
+	},
+	{
+		{ 1280, 768, 79500, 128, 64, 192, 7, 3, 20,
+			OMAPDSS_SIG_ACTIVE_HIGH, OMAPDSS_SIG_ACTIVE_LOW,
+			false, },
+		{ 0x17, HDMI_DVI },
+	},
+	{
+		{ 1280, 800, 83500, 128, 72, 200, 6, 3, 22,
+			OMAPDSS_SIG_ACTIVE_HIGH, OMAPDSS_SIG_ACTIVE_LOW,
+			false, },
+		{ 0x1C, HDMI_DVI },
+	},
+	{
+		{ 1360, 768, 85500, 112, 64, 256, 6, 3, 18,
+			OMAPDSS_SIG_ACTIVE_HIGH, OMAPDSS_SIG_ACTIVE_HIGH,
+			false, },
+		{ 0x27, HDMI_DVI },
+	},
+	{
+		{ 1280, 960, 108000, 112, 96, 312, 3, 1, 36,
+			OMAPDSS_SIG_ACTIVE_HIGH, OMAPDSS_SIG_ACTIVE_HIGH,
+			false, },
+		{ 0x20, HDMI_DVI },
+	},
+	{
+		{ 1280, 1024, 108000, 112, 48, 248, 3, 1, 38,
+			OMAPDSS_SIG_ACTIVE_HIGH, OMAPDSS_SIG_ACTIVE_HIGH,
+			false, },
+		{ 0x23, HDMI_DVI },
+	},
+	{
+		{ 1024, 768, 65000, 136, 24, 160, 6, 3, 29,
+			OMAPDSS_SIG_ACTIVE_LOW, OMAPDSS_SIG_ACTIVE_LOW,
+			false, },
+		{ 0x10, HDMI_DVI },
+	},
+	{
+		{ 1400, 1050, 121750, 144, 88, 232, 4, 3, 32,
+			OMAPDSS_SIG_ACTIVE_HIGH, OMAPDSS_SIG_ACTIVE_LOW,
+			false, },
+		{ 0x2A, HDMI_DVI },
+	},
+	{
+		{ 1440, 900, 106500, 152, 80, 232, 6, 3, 25,
+			OMAPDSS_SIG_ACTIVE_HIGH, OMAPDSS_SIG_ACTIVE_LOW,
+			false, },
+		{ 0x2F, HDMI_DVI },
+	},
+	{
+		{ 1680, 1050, 146250, 176 , 104, 280, 6, 3, 30,
+			OMAPDSS_SIG_ACTIVE_HIGH, OMAPDSS_SIG_ACTIVE_LOW,
+			false, },
+		{ 0x3A, HDMI_DVI },
+	},
+	{
+		{ 1366, 768, 85500, 143, 70, 213, 3, 3, 24,
+			OMAPDSS_SIG_ACTIVE_HIGH, OMAPDSS_SIG_ACTIVE_HIGH,
+			false, },
+		{ 0x51, HDMI_DVI },
+	},
+	{
+		{ 1920, 1080, 148500, 44, 148, 80, 5, 4, 36,
+			OMAPDSS_SIG_ACTIVE_HIGH, OMAPDSS_SIG_ACTIVE_HIGH,
+			false, },
+		{ 0x52, HDMI_DVI },
+	},
+	{
+		{ 1280, 768, 68250, 32, 48, 80, 7, 3, 12,
+			OMAPDSS_SIG_ACTIVE_LOW, OMAPDSS_SIG_ACTIVE_HIGH,
+			false, },
+		{ 0x16, HDMI_DVI },
+	},
+	{
+		{ 1400, 1050, 101000, 32, 48, 80, 4, 3, 23,
+			OMAPDSS_SIG_ACTIVE_LOW, OMAPDSS_SIG_ACTIVE_HIGH,
+			false, },
+		{ 0x29, HDMI_DVI },
+	},
+	{
+		{ 1680, 1050, 119000, 32, 48, 80, 6, 3, 21,
+			OMAPDSS_SIG_ACTIVE_LOW, OMAPDSS_SIG_ACTIVE_HIGH,
+			false, },
+		{ 0x39, HDMI_DVI },
+	},
+	{
+		{ 1280, 800, 79500, 32, 48, 80, 6, 3, 14,
+			OMAPDSS_SIG_ACTIVE_LOW, OMAPDSS_SIG_ACTIVE_HIGH,
+			false, },
+		{ 0x1B, HDMI_DVI },
+	},
+	{
+		{ 1280, 720, 74250, 40, 110, 220, 5, 5, 20,
+			OMAPDSS_SIG_ACTIVE_HIGH, OMAPDSS_SIG_ACTIVE_HIGH,
+			false, },
+		{ 0x55, HDMI_DVI },
+	},
+	{
+		{ 1920, 1200, 154000, 32, 48, 80, 6, 3, 26,
+			OMAPDSS_SIG_ACTIVE_LOW, OMAPDSS_SIG_ACTIVE_HIGH,
+			false, },
+		{ 0x44, HDMI_DVI },
+	},
+};
+
+const struct hdmi_config *hdmi_default_timing(void)
+{
+	return &vesa_timings[0];
+}
+
+static const struct hdmi_config *hdmi_find_timing(int code,
+			const struct hdmi_config *timings_arr, int len)
+{
+	int i;
+
+	for (i = 0; i < len; i++) {
+		if (timings_arr[i].cm.code == code)
+			return &timings_arr[i];
+	}
+
+	return NULL;
+}
+
+const struct hdmi_config *hdmi_get_timings(int mode, int code)
+{
+	const struct hdmi_config *arr;
+	int len;
+
+	if (mode == HDMI_DVI) {
+		arr = vesa_timings;
+		len = ARRAY_SIZE(vesa_timings);
+	} else {
+		arr = cea_timings;
+		len = ARRAY_SIZE(cea_timings);
+	}
+
+	return hdmi_find_timing(code, arr, len);
+}
+
+static bool hdmi_timings_compare(struct omap_video_timings *timing1,
+			const struct omap_video_timings *timing2)
+{
+	int timing1_vsync, timing1_hsync, timing2_vsync, timing2_hsync;
+
+	if ((DIV_ROUND_CLOSEST(timing2->pixel_clock, 1000) ==
+			DIV_ROUND_CLOSEST(timing1->pixel_clock, 1000)) &&
+		(timing2->x_res == timing1->x_res) &&
+		(timing2->y_res == timing1->y_res)) {
+
+		timing2_hsync = timing2->hfp + timing2->hsw + timing2->hbp;
+		timing1_hsync = timing1->hfp + timing1->hsw + timing1->hbp;
+		timing2_vsync = timing2->vfp + timing2->vsw + timing2->vbp;
+		timing1_vsync = timing1->vfp + timing1->vsw + timing1->vbp;
+
+		DSSDBG("timing1_hsync = %d timing1_vsync = %d"\
+			"timing2_hsync = %d timing2_vsync = %d\n",
+			timing1_hsync, timing1_vsync,
+			timing2_hsync, timing2_vsync);
+
+		if ((timing1_hsync == timing2_hsync) &&
+			(timing1_vsync == timing2_vsync)) {
+			return true;
+		}
+	}
+	return false;
+}
+
+struct hdmi_cm hdmi_get_code(struct omap_video_timings *timing)
+{
+	int i;
+	struct hdmi_cm cm = {-1};
+	DSSDBG("hdmi_get_code\n");
+
+	for (i = 0; i < ARRAY_SIZE(cea_timings); i++) {
+		if (hdmi_timings_compare(timing, &cea_timings[i].timings)) {
+			cm = cea_timings[i].cm;
+			goto end;
+		}
+	}
+	for (i = 0; i < ARRAY_SIZE(vesa_timings); i++) {
+		if (hdmi_timings_compare(timing, &vesa_timings[i].timings)) {
+			cm = vesa_timings[i].cm;
+			goto end;
+		}
+	}
+
+end:
+	return cm;
+}
+
+#if defined(CONFIG_OMAP4_DSS_HDMI_AUDIO)
+int hdmi_compute_acr(u32 pclk, u32 sample_freq, u32 *n, u32 *cts)
+{
+	u32 deep_color;
+	bool deep_color_correct = false;
+
+	if (n == NULL || cts == NULL)
+		return -EINVAL;
+
+	/* TODO: When implemented, query deep color mode here. */
+	deep_color = 100;
+
+	/*
+	 * When using deep color, the default N value (as in the HDMI
+	 * specification) yields to an non-integer CTS. Hence, we
+	 * modify it while keeping the restrictions described in
+	 * section 7.2.1 of the HDMI 1.4a specification.
+	 */
+	switch (sample_freq) {
+	case 32000:
+	case 48000:
+	case 96000:
+	case 192000:
+		if (deep_color == 125)
+			if (pclk == 27027 || pclk == 74250)
+				deep_color_correct = true;
+		if (deep_color == 150)
+			if (pclk == 27027)
+				deep_color_correct = true;
+		break;
+	case 44100:
+	case 88200:
+	case 176400:
+		if (deep_color == 125)
+			if (pclk == 27027)
+				deep_color_correct = true;
+		break;
+	default:
+		return -EINVAL;
+	}
+
+	if (deep_color_correct) {
+		switch (sample_freq) {
+		case 32000:
+			*n = 8192;
+			break;
+		case 44100:
+			*n = 12544;
+			break;
+		case 48000:
+			*n = 8192;
+			break;
+		case 88200:
+			*n = 25088;
+			break;
+		case 96000:
+			*n = 16384;
+			break;
+		case 176400:
+			*n = 50176;
+			break;
+		case 192000:
+			*n = 32768;
+			break;
+		default:
+			return -EINVAL;
+		}
+	} else {
+		switch (sample_freq) {
+		case 32000:
+			*n = 4096;
+			break;
+		case 44100:
+			*n = 6272;
+			break;
+		case 48000:
+			*n = 6144;
+			break;
+		case 88200:
+			*n = 12544;
+			break;
+		case 96000:
+			*n = 12288;
+			break;
+		case 176400:
+			*n = 25088;
+			break;
+		case 192000:
+			*n = 24576;
+			break;
+		default:
+			return -EINVAL;
+		}
+	}
+	/* Calculate CTS. See HDMI 1.3a or 1.4a specifications */
+	*cts = pclk * (*n / 128) * deep_color / (sample_freq / 10);
+
+	return 0;
+}
+#endif
diff --git a/drivers/video/omap2/dss/hdmi_phy.c b/drivers/video/omap2/dss/hdmi_phy.c
new file mode 100644
index 000000000000..45acb997ac00
--- /dev/null
+++ b/drivers/video/omap2/dss/hdmi_phy.c
@@ -0,0 +1,160 @@
+/*
+ * HDMI PHY
+ *
+ * Copyright (C) 2013 Texas Instruments Incorporated
+ *
+ * This program is free software; you can redistribute it and/or modify it
+ * under the terms of the GNU General Public License version 2 as published by
+ * the Free Software Foundation.
+ */
+
+#include <linux/kernel.h>
+#include <linux/err.h>
+#include <linux/io.h>
+#include <linux/platform_device.h>
+#include <video/omapdss.h>
+
+#include "dss.h"
+#include "hdmi.h"
+
+void hdmi_phy_dump(struct hdmi_phy_data *phy, struct seq_file *s)
+{
+#define DUMPPHY(r) seq_printf(s, "%-35s %08x\n", #r,\
+		hdmi_read_reg(phy->base, r))
+
+	DUMPPHY(HDMI_TXPHY_TX_CTRL);
+	DUMPPHY(HDMI_TXPHY_DIGITAL_CTRL);
+	DUMPPHY(HDMI_TXPHY_POWER_CTRL);
+	DUMPPHY(HDMI_TXPHY_PAD_CFG_CTRL);
+}
+
+static irqreturn_t hdmi_irq_handler(int irq, void *data)
+{
+	struct hdmi_wp_data *wp = data;
+	u32 irqstatus;
+
+	irqstatus = hdmi_wp_get_irqstatus(wp);
+	hdmi_wp_set_irqstatus(wp, irqstatus);
+
+	if ((irqstatus & HDMI_IRQ_LINK_CONNECT) &&
+			irqstatus & HDMI_IRQ_LINK_DISCONNECT) {
+		/*
+		 * If we get both connect and disconnect interrupts at the same
+		 * time, turn off the PHY, clear interrupts, and restart, which
+		 * raises connect interrupt if a cable is connected, or nothing
+		 * if cable is not connected.
+		 */
+		hdmi_wp_set_phy_pwr(wp, HDMI_PHYPWRCMD_OFF);
+
+		hdmi_wp_set_irqstatus(wp, HDMI_IRQ_LINK_CONNECT |
+				HDMI_IRQ_LINK_DISCONNECT);
+
+		hdmi_wp_set_phy_pwr(wp, HDMI_PHYPWRCMD_LDOON);
+	} else if (irqstatus & HDMI_IRQ_LINK_CONNECT) {
+		hdmi_wp_set_phy_pwr(wp, HDMI_PHYPWRCMD_TXON);
+	} else if (irqstatus & HDMI_IRQ_LINK_DISCONNECT) {
+		hdmi_wp_set_phy_pwr(wp, HDMI_PHYPWRCMD_LDOON);
+	}
+
+	return IRQ_HANDLED;
+}
+
+int hdmi_phy_enable(struct hdmi_phy_data *phy, struct hdmi_wp_data *wp,
+			struct hdmi_config *cfg)
+{
+	u16 r = 0;
+	u32 irqstatus;
+
+	hdmi_wp_clear_irqenable(wp, 0xffffffff);
+
+	irqstatus = hdmi_wp_get_irqstatus(wp);
+	hdmi_wp_set_irqstatus(wp, irqstatus);
+
+	r = hdmi_wp_set_phy_pwr(wp, HDMI_PHYPWRCMD_LDOON);
+	if (r)
+		return r;
+
+	/*
+	 * Read address 0 in order to get the SCP reset done completed
+	 * Dummy access performed to make sure reset is done
+	 */
+	hdmi_read_reg(phy->base, HDMI_TXPHY_TX_CTRL);
+
+	/*
+	 * Write to phy address 0 to configure the clock
+	 * use HFBITCLK write HDMI_TXPHY_TX_CONTROL_FREQOUT field
+	 */
+	REG_FLD_MOD(phy->base, HDMI_TXPHY_TX_CTRL, 0x1, 31, 30);
+
+	/* Write to phy address 1 to start HDMI line (TXVALID and TMDSCLKEN) */
+	hdmi_write_reg(phy->base, HDMI_TXPHY_DIGITAL_CTRL, 0xF0000000);
+
+	/* Setup max LDO voltage */
+	REG_FLD_MOD(phy->base, HDMI_TXPHY_POWER_CTRL, 0xB, 3, 0);
+
+	/* Write to phy address 3 to change the polarity control */
+	REG_FLD_MOD(phy->base, HDMI_TXPHY_PAD_CFG_CTRL, 0x1, 27, 27);
+
+	r = request_threaded_irq(phy->irq, NULL, hdmi_irq_handler,
+				IRQF_ONESHOT, "OMAP HDMI", wp);
+	if (r) {
+		DSSERR("HDMI IRQ request failed\n");
+		hdmi_wp_set_phy_pwr(wp, HDMI_PHYPWRCMD_OFF);
+		return r;
+	}
+
+	hdmi_wp_set_irqenable(wp,
+		HDMI_IRQ_LINK_CONNECT | HDMI_IRQ_LINK_DISCONNECT);
+
+	return 0;
+}
+
+void hdmi_phy_disable(struct hdmi_phy_data *phy, struct hdmi_wp_data *wp)
+{
+	free_irq(phy->irq, wp);
+
+	hdmi_wp_set_phy_pwr(wp, HDMI_PHYPWRCMD_OFF);
+}
+
+#define PHY_OFFSET	0x300
+#define PHY_SIZE	0x100
+
+int hdmi_phy_init(struct platform_device *pdev, struct hdmi_phy_data *phy)
+{
+	struct resource *res;
+	struct resource temp_res;
+
+	res = platform_get_resource_byname(pdev, IORESOURCE_MEM, "hdmi_txphy");
+	if (!res) {
+		DSSDBG("can't get PHY mem resource by name\n");
+		/*
+		 * if hwmod/DT doesn't have the memory resource information
+		 * split into HDMI sub blocks by name, we try again by getting
+		 * the platform's first resource. this code will be removed when
+		 * the driver can get the mem resources by name
+		 */
+		res = platform_get_resource(pdev, IORESOURCE_MEM, 0);
+		if (!res) {
+			DSSERR("can't get PHY mem resource\n");
+			return -EINVAL;
+		}
+
+		temp_res.start = res->start + PHY_OFFSET;
+		temp_res.end = temp_res.start + PHY_SIZE - 1;
+		res = &temp_res;
+	}
+
+	phy->base = devm_ioremap(&pdev->dev, res->start, resource_size(res));
+	if (!phy->base) {
+		DSSERR("can't ioremap TX PHY\n");
+		return -ENOMEM;
+	}
+
+	phy->irq = platform_get_irq(pdev, 0);
+	if (phy->irq < 0) {
+		DSSERR("platform_get_irq failed\n");
+		return -ENODEV;
+	}
+
+	return 0;
+}
diff --git a/drivers/video/omap2/dss/hdmi_pll.c b/drivers/video/omap2/dss/hdmi_pll.c
new file mode 100644
index 000000000000..d3e6e78c0082
--- /dev/null
+++ b/drivers/video/omap2/dss/hdmi_pll.c
@@ -0,0 +1,230 @@
+/*
+ * HDMI PLL
+ *
+ * Copyright (C) 2013 Texas Instruments Incorporated
+ *
+ * This program is free software; you can redistribute it and/or modify it
+ * under the terms of the GNU General Public License version 2 as published by
+ * the Free Software Foundation.
+ */
+
+#include <linux/kernel.h>
+#include <linux/module.h>
+#include <linux/err.h>
+#include <linux/io.h>
+#include <linux/platform_device.h>
+#include <video/omapdss.h>
+
+#include "dss.h"
+#include "hdmi.h"
+
+#define HDMI_DEFAULT_REGN 16
+#define HDMI_DEFAULT_REGM2 1
+
+void hdmi_pll_dump(struct hdmi_pll_data *pll, struct seq_file *s)
+{
+#define DUMPPLL(r) seq_printf(s, "%-35s %08x\n", #r,\
+		hdmi_read_reg(pll->base, r))
+
+	DUMPPLL(PLLCTRL_PLL_CONTROL);
+	DUMPPLL(PLLCTRL_PLL_STATUS);
+	DUMPPLL(PLLCTRL_PLL_GO);
+	DUMPPLL(PLLCTRL_CFG1);
+	DUMPPLL(PLLCTRL_CFG2);
+	DUMPPLL(PLLCTRL_CFG3);
+	DUMPPLL(PLLCTRL_SSC_CFG1);
+	DUMPPLL(PLLCTRL_SSC_CFG2);
+	DUMPPLL(PLLCTRL_CFG4);
+}
+
+void hdmi_pll_compute(struct hdmi_pll_data *pll, unsigned long clkin, int phy)
+{
+	struct hdmi_pll_info *pi = &pll->info;
+	unsigned long refclk;
+	u32 mf;
+
+	/* use our funky units */
+	clkin /= 10000;
+
+	/*
+	 * Input clock is predivided by N + 1
+	 * out put of which is reference clk
+	 */
+
+	pi->regn = HDMI_DEFAULT_REGN;
+
+	refclk = clkin / pi->regn;
+
+	pi->regm2 = HDMI_DEFAULT_REGM2;
+
+	/*
+	 * multiplier is pixel_clk/ref_clk
+	 * Multiplying by 100 to avoid fractional part removal
+	 */
+	pi->regm = phy * pi->regm2 / refclk;
+
+	/*
+	 * fractional multiplier is remainder of the difference between
+	 * multiplier and actual phy(required pixel clock thus should be
+	 * multiplied by 2^18(262144) divided by the reference clock
+	 */
+	mf = (phy - pi->regm / pi->regm2 * refclk) * 262144;
+	pi->regmf = pi->regm2 * mf / refclk;
+
+	/*
+	 * Dcofreq should be set to 1 if required pixel clock
+	 * is greater than 1000MHz
+	 */
+	pi->dcofreq = phy > 1000 * 100;
+	pi->regsd = ((pi->regm * clkin / 10) / (pi->regn * 250) + 5) / 10;
+
+	/* Set the reference clock to sysclk reference */
+	pi->refsel = HDMI_REFSEL_SYSCLK;
+
+	DSSDBG("M = %d Mf = %d\n", pi->regm, pi->regmf);
+	DSSDBG("range = %d sd = %d\n", pi->dcofreq, pi->regsd);
+}
+
+
+static int hdmi_pll_config(struct hdmi_pll_data *pll)
+{
+	u32 r;
+	struct hdmi_pll_info *fmt = &pll->info;
+
+	/* PLL start always use manual mode */
+	REG_FLD_MOD(pll->base, PLLCTRL_PLL_CONTROL, 0x0, 0, 0);
+
+	r = hdmi_read_reg(pll->base, PLLCTRL_CFG1);
+	r = FLD_MOD(r, fmt->regm, 20, 9);	/* CFG1_PLL_REGM */
+	r = FLD_MOD(r, fmt->regn - 1, 8, 1);	/* CFG1_PLL_REGN */
+	hdmi_write_reg(pll->base, PLLCTRL_CFG1, r);
+
+	r = hdmi_read_reg(pll->base, PLLCTRL_CFG2);
+
+	r = FLD_MOD(r, 0x0, 12, 12);	/* PLL_HIGHFREQ divide by 2 */
+	r = FLD_MOD(r, 0x1, 13, 13);	/* PLL_REFEN */
+	r = FLD_MOD(r, 0x0, 14, 14);	/* PHY_CLKINEN de-assert during locking */
+	r = FLD_MOD(r, fmt->refsel, 22, 21);	/* REFSEL */
+
+	if (fmt->dcofreq) {
+		/* divider programming for frequency beyond 1000Mhz */
+		REG_FLD_MOD(pll->base, PLLCTRL_CFG3, fmt->regsd, 17, 10);
+		r = FLD_MOD(r, 0x4, 3, 1);	/* 1000MHz and 2000MHz */
+	} else {
+		r = FLD_MOD(r, 0x2, 3, 1);	/* 500MHz and 1000MHz */
+	}
+
+	hdmi_write_reg(pll->base, PLLCTRL_CFG2, r);
+
+	r = hdmi_read_reg(pll->base, PLLCTRL_CFG4);
+	r = FLD_MOD(r, fmt->regm2, 24, 18);
+	r = FLD_MOD(r, fmt->regmf, 17, 0);
+	hdmi_write_reg(pll->base, PLLCTRL_CFG4, r);
+
+	/* go now */
+	REG_FLD_MOD(pll->base, PLLCTRL_PLL_GO, 0x1, 0, 0);
+
+	/* wait for bit change */
+	if (hdmi_wait_for_bit_change(pll->base, PLLCTRL_PLL_GO,
+			0, 0, 1) != 1) {
+		pr_err("PLL GO bit not set\n");
+		return -ETIMEDOUT;
+	}
+
+	/* Wait till the lock bit is set in PLL status */
+	if (hdmi_wait_for_bit_change(pll->base,
+			PLLCTRL_PLL_STATUS, 1, 1, 1) != 1) {
+		pr_err("cannot lock PLL\n");
+		pr_err("CFG1 0x%x\n",
+			hdmi_read_reg(pll->base, PLLCTRL_CFG1));
+		pr_err("CFG2 0x%x\n",
+			hdmi_read_reg(pll->base, PLLCTRL_CFG2));
+		pr_err("CFG4 0x%x\n",
+			hdmi_read_reg(pll->base, PLLCTRL_CFG4));
+		return -ETIMEDOUT;
+	}
+
+	pr_debug("PLL locked!\n");
+
+	return 0;
+}
+
+static int hdmi_pll_reset(struct hdmi_pll_data *pll)
+{
+	/* SYSRESET  controlled by power FSM */
+	REG_FLD_MOD(pll->base, PLLCTRL_PLL_CONTROL, 0x0, 3, 3);
+
+	/* READ 0x0 reset is in progress */
+	if (hdmi_wait_for_bit_change(pll->base, PLLCTRL_PLL_STATUS, 0, 0, 1)
+			!= 1) {
+		pr_err("Failed to sysreset PLL\n");
+		return -ETIMEDOUT;
+	}
+
+	return 0;
+}
+
+int hdmi_pll_enable(struct hdmi_pll_data *pll, struct hdmi_wp_data *wp)
+{
+	u16 r = 0;
+
+	r = hdmi_wp_set_pll_pwr(wp, HDMI_PLLPWRCMD_ALLOFF);
+	if (r)
+		return r;
+
+	r = hdmi_wp_set_pll_pwr(wp, HDMI_PLLPWRCMD_BOTHON_ALLCLKS);
+	if (r)
+		return r;
+
+	r = hdmi_pll_reset(pll);
+	if (r)
+		return r;
+
+	r = hdmi_pll_config(pll);
+	if (r)
+		return r;
+
+	return 0;
+}
+
+void hdmi_pll_disable(struct hdmi_pll_data *pll, struct hdmi_wp_data *wp)
+{
+	hdmi_wp_set_pll_pwr(wp, HDMI_PLLPWRCMD_ALLOFF);
+}
+
+#define PLL_OFFSET	0x200
+#define PLL_SIZE	0x100
+
+int hdmi_pll_init(struct platform_device *pdev, struct hdmi_pll_data *pll)
+{
+	struct resource *res;
+	struct resource temp_res;
+
+	res = platform_get_resource_byname(pdev, IORESOURCE_MEM, "hdmi_pllctrl");
+	if (!res) {
+		DSSDBG("can't get PLL mem resource by name\n");
+		/*
+		 * if hwmod/DT doesn't have the memory resource information
+		 * split into HDMI sub blocks by name, we try again by getting
+		 * the platform's first resource. this code will be removed when
+		 * the driver can get the mem resources by name
+		 */
+		res = platform_get_resource(pdev, IORESOURCE_MEM, 0);
+		if (!res) {
+			DSSERR("can't get PLL mem resource\n");
+			return -EINVAL;
+		}
+
+		temp_res.start = res->start + PLL_OFFSET;
+		temp_res.end = temp_res.start + PLL_SIZE - 1;
+		res = &temp_res;
+	}
+
+	pll->base = devm_ioremap(&pdev->dev, res->start, resource_size(res));
+	if (!pll->base) {
+		DSSERR("can't ioremap PLLCTRL\n");
+		return -ENOMEM;
+	}
+
+	return 0;
+}
diff --git a/drivers/video/omap2/dss/hdmi_wp.c b/drivers/video/omap2/dss/hdmi_wp.c
new file mode 100644
index 000000000000..8151d8969a6e
--- /dev/null
+++ b/drivers/video/omap2/dss/hdmi_wp.c
@@ -0,0 +1,271 @@
+/*
+ * HDMI wrapper
+ *
+ * Copyright (C) 2013 Texas Instruments Incorporated
+ *
+ * This program is free software; you can redistribute it and/or modify it
+ * under the terms of the GNU General Public License version 2 as published by
+ * the Free Software Foundation.
+ */
+
+#include <linux/kernel.h>
+#include <linux/err.h>
+#include <linux/io.h>
+#include <linux/platform_device.h>
+#include <video/omapdss.h>
+
+#include "dss.h"
+#include "hdmi.h"
+
+void hdmi_wp_dump(struct hdmi_wp_data *wp, struct seq_file *s)
+{
+#define DUMPREG(r) seq_printf(s, "%-35s %08x\n", #r, hdmi_read_reg(wp->base, r))
+
+	DUMPREG(HDMI_WP_REVISION);
+	DUMPREG(HDMI_WP_SYSCONFIG);
+	DUMPREG(HDMI_WP_IRQSTATUS_RAW);
+	DUMPREG(HDMI_WP_IRQSTATUS);
+	DUMPREG(HDMI_WP_IRQENABLE_SET);
+	DUMPREG(HDMI_WP_IRQENABLE_CLR);
+	DUMPREG(HDMI_WP_IRQWAKEEN);
+	DUMPREG(HDMI_WP_PWR_CTRL);
+	DUMPREG(HDMI_WP_DEBOUNCE);
+	DUMPREG(HDMI_WP_VIDEO_CFG);
+	DUMPREG(HDMI_WP_VIDEO_SIZE);
+	DUMPREG(HDMI_WP_VIDEO_TIMING_H);
+	DUMPREG(HDMI_WP_VIDEO_TIMING_V);
+	DUMPREG(HDMI_WP_WP_CLK);
+	DUMPREG(HDMI_WP_AUDIO_CFG);
+	DUMPREG(HDMI_WP_AUDIO_CFG2);
+	DUMPREG(HDMI_WP_AUDIO_CTRL);
+	DUMPREG(HDMI_WP_AUDIO_DATA);
+}
+
+u32 hdmi_wp_get_irqstatus(struct hdmi_wp_data *wp)
+{
+	return hdmi_read_reg(wp->base, HDMI_WP_IRQSTATUS);
+}
+
+void hdmi_wp_set_irqstatus(struct hdmi_wp_data *wp, u32 irqstatus)
+{
+	hdmi_write_reg(wp->base, HDMI_WP_IRQSTATUS, irqstatus);
+	/* flush posted write */
+	hdmi_read_reg(wp->base, HDMI_WP_IRQSTATUS);
+}
+
+void hdmi_wp_set_irqenable(struct hdmi_wp_data *wp, u32 mask)
+{
+	hdmi_write_reg(wp->base, HDMI_WP_IRQENABLE_SET, mask);
+}
+
+void hdmi_wp_clear_irqenable(struct hdmi_wp_data *wp, u32 mask)
+{
+	hdmi_write_reg(wp->base, HDMI_WP_IRQENABLE_CLR, mask);
+}
+
+/* PHY_PWR_CMD */
+int hdmi_wp_set_phy_pwr(struct hdmi_wp_data *wp, enum hdmi_phy_pwr val)
+{
+	/* Return if already the state */
+	if (REG_GET(wp->base, HDMI_WP_PWR_CTRL, 5, 4) == val)
+		return 0;
+
+	/* Command for power control of HDMI PHY */
+	REG_FLD_MOD(wp->base, HDMI_WP_PWR_CTRL, val, 7, 6);
+
+	/* Status of the power control of HDMI PHY */
+	if (hdmi_wait_for_bit_change(wp->base, HDMI_WP_PWR_CTRL, 5, 4, val)
+			!= val) {
+		pr_err("Failed to set PHY power mode to %d\n", val);
+		return -ETIMEDOUT;
+	}
+
+	return 0;
+}
+
+/* PLL_PWR_CMD */
+int hdmi_wp_set_pll_pwr(struct hdmi_wp_data *wp, enum hdmi_pll_pwr val)
+{
+	/* Command for power control of HDMI PLL */
+	REG_FLD_MOD(wp->base, HDMI_WP_PWR_CTRL, val, 3, 2);
+
+	/* wait till PHY_PWR_STATUS is set */
+	if (hdmi_wait_for_bit_change(wp->base, HDMI_WP_PWR_CTRL, 1, 0, val)
+			!= val) {
+		pr_err("Failed to set PLL_PWR_STATUS\n");
+		return -ETIMEDOUT;
+	}
+
+	return 0;
+}
+
+int hdmi_wp_video_start(struct hdmi_wp_data *wp)
+{
+	REG_FLD_MOD(wp->base, HDMI_WP_VIDEO_CFG, true, 31, 31);
+
+	return 0;
+}
+
+void hdmi_wp_video_stop(struct hdmi_wp_data *wp)
+{
+	REG_FLD_MOD(wp->base, HDMI_WP_VIDEO_CFG, false, 31, 31);
+}
+
+void hdmi_wp_video_config_format(struct hdmi_wp_data *wp,
+		struct hdmi_video_format *video_fmt)
+{
+	u32 l = 0;
+
+	REG_FLD_MOD(wp->base, HDMI_WP_VIDEO_CFG, video_fmt->packing_mode,
+		10, 8);
+
+	l |= FLD_VAL(video_fmt->y_res, 31, 16);
+	l |= FLD_VAL(video_fmt->x_res, 15, 0);
+	hdmi_write_reg(wp->base, HDMI_WP_VIDEO_SIZE, l);
+}
+
+void hdmi_wp_video_config_interface(struct hdmi_wp_data *wp,
+		struct omap_video_timings *timings)
+{
+	u32 r;
+	bool vsync_pol, hsync_pol;
+	pr_debug("Enter hdmi_wp_video_config_interface\n");
+
+	vsync_pol = timings->vsync_level == OMAPDSS_SIG_ACTIVE_HIGH;
+	hsync_pol = timings->hsync_level == OMAPDSS_SIG_ACTIVE_HIGH;
+
+	r = hdmi_read_reg(wp->base, HDMI_WP_VIDEO_CFG);
+	r = FLD_MOD(r, vsync_pol, 7, 7);
+	r = FLD_MOD(r, hsync_pol, 6, 6);
+	r = FLD_MOD(r, timings->interlace, 3, 3);
+	r = FLD_MOD(r, 1, 1, 0); /* HDMI_TIMING_MASTER_24BIT */
+	hdmi_write_reg(wp->base, HDMI_WP_VIDEO_CFG, r);
+}
+
+void hdmi_wp_video_config_timing(struct hdmi_wp_data *wp,
+		struct omap_video_timings *timings)
+{
+	u32 timing_h = 0;
+	u32 timing_v = 0;
+
+	pr_debug("Enter hdmi_wp_video_config_timing\n");
+
+	timing_h |= FLD_VAL(timings->hbp, 31, 20);
+	timing_h |= FLD_VAL(timings->hfp, 19, 8);
+	timing_h |= FLD_VAL(timings->hsw, 7, 0);
+	hdmi_write_reg(wp->base, HDMI_WP_VIDEO_TIMING_H, timing_h);
+
+	timing_v |= FLD_VAL(timings->vbp, 31, 20);
+	timing_v |= FLD_VAL(timings->vfp, 19, 8);
+	timing_v |= FLD_VAL(timings->vsw, 7, 0);
+	hdmi_write_reg(wp->base, HDMI_WP_VIDEO_TIMING_V, timing_v);
+}
+
+void hdmi_wp_init_vid_fmt_timings(struct hdmi_video_format *video_fmt,
+		struct omap_video_timings *timings, struct hdmi_config *param)
+{
+	pr_debug("Enter hdmi_wp_video_init_format\n");
+
+	video_fmt->packing_mode = HDMI_PACK_10b_RGB_YUV444;
+	video_fmt->y_res = param->timings.y_res;
+	video_fmt->x_res = param->timings.x_res;
+
+	timings->hbp = param->timings.hbp;
+	timings->hfp = param->timings.hfp;
+	timings->hsw = param->timings.hsw;
+	timings->vbp = param->timings.vbp;
+	timings->vfp = param->timings.vfp;
+	timings->vsw = param->timings.vsw;
+	timings->vsync_level = param->timings.vsync_level;
+	timings->hsync_level = param->timings.hsync_level;
+	timings->interlace = param->timings.interlace;
+}
+
+#if defined(CONFIG_OMAP4_DSS_HDMI_AUDIO)
+void hdmi_wp_audio_config_format(struct hdmi_wp_data *wp,
+		struct hdmi_audio_format *aud_fmt)
+{
+	u32 r;
+
+	DSSDBG("Enter hdmi_wp_audio_config_format\n");
+
+	r = hdmi_read_reg(wp->base, HDMI_WP_AUDIO_CFG);
+	r = FLD_MOD(r, aud_fmt->stereo_channels, 26, 24);
+	r = FLD_MOD(r, aud_fmt->active_chnnls_msk, 23, 16);
+	r = FLD_MOD(r, aud_fmt->en_sig_blk_strt_end, 5, 5);
+	r = FLD_MOD(r, aud_fmt->type, 4, 4);
+	r = FLD_MOD(r, aud_fmt->justification, 3, 3);
+	r = FLD_MOD(r, aud_fmt->sample_order, 2, 2);
+	r = FLD_MOD(r, aud_fmt->samples_per_word, 1, 1);
+	r = FLD_MOD(r, aud_fmt->sample_size, 0, 0);
+	hdmi_write_reg(wp->base, HDMI_WP_AUDIO_CFG, r);
+}
+
+void hdmi_wp_audio_config_dma(struct hdmi_wp_data *wp,
+		struct hdmi_audio_dma *aud_dma)
+{
+	u32 r;
+
+	DSSDBG("Enter hdmi_wp_audio_config_dma\n");
+
+	r = hdmi_read_reg(wp->base, HDMI_WP_AUDIO_CFG2);
+	r = FLD_MOD(r, aud_dma->transfer_size, 15, 8);
+	r = FLD_MOD(r, aud_dma->block_size, 7, 0);
+	hdmi_write_reg(wp->base, HDMI_WP_AUDIO_CFG2, r);
+
+	r = hdmi_read_reg(wp->base, HDMI_WP_AUDIO_CTRL);
+	r = FLD_MOD(r, aud_dma->mode, 9, 9);
+	r = FLD_MOD(r, aud_dma->fifo_threshold, 8, 0);
+	hdmi_write_reg(wp->base, HDMI_WP_AUDIO_CTRL, r);
+}
+
+int hdmi_wp_audio_enable(struct hdmi_wp_data *wp, bool enable)
+{
+	REG_FLD_MOD(wp->base, HDMI_WP_AUDIO_CTRL, enable, 31, 31);
+
+	return 0;
+}
+
+int hdmi_wp_audio_core_req_enable(struct hdmi_wp_data *wp, bool enable)
+{
+	REG_FLD_MOD(wp->base, HDMI_WP_AUDIO_CTRL, enable, 30, 30);
+
+	return 0;
+}
+#endif
+
+#define WP_SIZE	0x200
+
+int hdmi_wp_init(struct platform_device *pdev, struct hdmi_wp_data *wp)
+{
+	struct resource *res;
+	struct resource temp_res;
+
+	res = platform_get_resource_byname(pdev, IORESOURCE_MEM, "hdmi_wp");
+	if (!res) {
+		DSSDBG("can't get WP mem resource by name\n");
+		/*
+		 * if hwmod/DT doesn't have the memory resource information
+		 * split into HDMI sub blocks by name, we try again by getting
+		 * the platform's first resource. this code will be removed when
+		 * the driver can get the mem resources by name
+		 */
+		res = platform_get_resource(pdev, IORESOURCE_MEM, 0);
+		if (!res) {
+			DSSERR("can't get WP mem resource\n");
+			return -EINVAL;
+		}
+
+		temp_res.start = res->start;
+		temp_res.end = temp_res.start + WP_SIZE - 1;
+		res = &temp_res;
+	}
+
+	wp->base = devm_ioremap(&pdev->dev, res->start, resource_size(res));
+	if (!wp->base) {
+		DSSERR("can't ioremap HDMI WP\n");
+		return -ENOMEM;
+	}
+
+	return 0;
+}
diff --git a/drivers/video/omap2/dss/ti_hdmi.h b/drivers/video/omap2/dss/ti_hdmi.h
deleted file mode 100644
index 45215f44617c..000000000000
--- a/drivers/video/omap2/dss/ti_hdmi.h
+++ /dev/null
@@ -1,187 +0,0 @@
-/*
- * ti_hdmi.h
- *
- * HDMI driver definition for TI OMAP4, DM81xx, DM38xx  Processor.
- *
- * Copyright (C) 2010-2011 Texas Instruments Incorporated - http://www.ti.com/
- *
- * This program is free software; you can redistribute it and/or modify it
- * under the terms of the GNU General Public License version 2 as published by
- * the Free Software Foundation.
- *
- * This program is distributed in the hope that it will be useful, but WITHOUT
- * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or
- * FITNESS FOR A PARTICULAR PURPOSE.  See the GNU General Public License for
- * more details.
- *
- * You should have received a copy of the GNU General Public License along with
- * this program.  If not, see <http://www.gnu.org/licenses/>.
- */
-
-#ifndef _TI_HDMI_H
-#define _TI_HDMI_H
-
-struct hdmi_ip_data;
-
-enum hdmi_pll_pwr {
-	HDMI_PLLPWRCMD_ALLOFF = 0,
-	HDMI_PLLPWRCMD_PLLONLY = 1,
-	HDMI_PLLPWRCMD_BOTHON_ALLCLKS = 2,
-	HDMI_PLLPWRCMD_BOTHON_NOPHYCLK = 3
-};
-
-enum hdmi_core_hdmi_dvi {
-	HDMI_DVI = 0,
-	HDMI_HDMI = 1
-};
-
-enum hdmi_clk_refsel {
-	HDMI_REFSEL_PCLK = 0,
-	HDMI_REFSEL_REF1 = 1,
-	HDMI_REFSEL_REF2 = 2,
-	HDMI_REFSEL_SYSCLK = 3
-};
-
-struct hdmi_cm {
-	int	code;
-	int	mode;
-};
-
-struct hdmi_config {
-	struct omap_video_timings timings;
-	struct hdmi_cm cm;
-};
-
-/* HDMI PLL structure */
-struct hdmi_pll_info {
-	u16 regn;
-	u16 regm;
-	u32 regmf;
-	u16 regm2;
-	u16 regsd;
-	u16 dcofreq;
-	enum hdmi_clk_refsel refsel;
-};
-
-struct ti_hdmi_ip_ops {
-
-	void (*video_configure)(struct hdmi_ip_data *ip_data);
-
-	int (*phy_enable)(struct hdmi_ip_data *ip_data);
-
-	void (*phy_disable)(struct hdmi_ip_data *ip_data);
-
-	int (*read_edid)(struct hdmi_ip_data *ip_data, u8 *edid, int len);
-
-	int (*pll_enable)(struct hdmi_ip_data *ip_data);
-
-	void (*pll_disable)(struct hdmi_ip_data *ip_data);
-
-	int (*video_enable)(struct hdmi_ip_data *ip_data);
-
-	void (*video_disable)(struct hdmi_ip_data *ip_data);
-
-	void (*dump_wrapper)(struct hdmi_ip_data *ip_data, struct seq_file *s);
-
-	void (*dump_core)(struct hdmi_ip_data *ip_data, struct seq_file *s);
-
-	void (*dump_pll)(struct hdmi_ip_data *ip_data, struct seq_file *s);
-
-	void (*dump_phy)(struct hdmi_ip_data *ip_data, struct seq_file *s);
-
-#if defined(CONFIG_OMAP4_DSS_HDMI_AUDIO)
-	int (*audio_enable)(struct hdmi_ip_data *ip_data);
-
-	void (*audio_disable)(struct hdmi_ip_data *ip_data);
-
-	int (*audio_start)(struct hdmi_ip_data *ip_data);
-
-	void (*audio_stop)(struct hdmi_ip_data *ip_data);
-
-	int (*audio_config)(struct hdmi_ip_data *ip_data,
-		struct omap_dss_audio *audio);
-
-	int (*audio_get_dma_port)(u32 *offset, u32 *size);
-#endif
-
-};
-
-/*
- * Refer to section 8.2 in HDMI 1.3 specification for
- * details about infoframe databytes
- */
-struct hdmi_core_infoframe_avi {
-	/* Y0, Y1 rgb,yCbCr */
-	u8	db1_format;
-	/* A0  Active information Present */
-	u8	db1_active_info;
-	/* B0, B1 Bar info data valid */
-	u8	db1_bar_info_dv;
-	/* S0, S1 scan information */
-	u8	db1_scan_info;
-	/* C0, C1 colorimetry */
-	u8	db2_colorimetry;
-	/* M0, M1 Aspect ratio (4:3, 16:9) */
-	u8	db2_aspect_ratio;
-	/* R0...R3 Active format aspect ratio */
-	u8	db2_active_fmt_ar;
-	/* ITC IT content. */
-	u8	db3_itc;
-	/* EC0, EC1, EC2 Extended colorimetry */
-	u8	db3_ec;
-	/* Q1, Q0 Quantization range */
-	u8	db3_q_range;
-	/* SC1, SC0 Non-uniform picture scaling */
-	u8	db3_nup_scaling;
-	/* VIC0..6 Video format identification */
-	u8	db4_videocode;
-	/* PR0..PR3 Pixel repetition factor */
-	u8	db5_pixel_repeat;
-	/* Line number end of top bar */
-	u16	db6_7_line_eoftop;
-	/* Line number start of bottom bar */
-	u16	db8_9_line_sofbottom;
-	/* Pixel number end of left bar */
-	u16	db10_11_pixel_eofleft;
-	/* Pixel number start of right bar */
-	u16	db12_13_pixel_sofright;
-};
-
-struct hdmi_ip_data {
-	void __iomem	*base_wp;	/* HDMI wrapper */
-	unsigned long	core_sys_offset;
-	unsigned long	core_av_offset;
-	unsigned long	pll_offset;
-	unsigned long	phy_offset;
-	int		irq;
-	const struct ti_hdmi_ip_ops *ops;
-	struct hdmi_config cfg;
-	struct hdmi_pll_info pll_data;
-	struct hdmi_core_infoframe_avi avi_cfg;
-
-	/* ti_hdmi_4xxx_ip private data. These should be in a separate struct */
-	struct mutex lock;
-};
-int ti_hdmi_4xxx_phy_enable(struct hdmi_ip_data *ip_data);
-void ti_hdmi_4xxx_phy_disable(struct hdmi_ip_data *ip_data);
-int ti_hdmi_4xxx_read_edid(struct hdmi_ip_data *ip_data, u8 *edid, int len);
-int ti_hdmi_4xxx_wp_video_start(struct hdmi_ip_data *ip_data);
-void ti_hdmi_4xxx_wp_video_stop(struct hdmi_ip_data *ip_data);
-int ti_hdmi_4xxx_pll_enable(struct hdmi_ip_data *ip_data);
-void ti_hdmi_4xxx_pll_disable(struct hdmi_ip_data *ip_data);
-void ti_hdmi_4xxx_basic_configure(struct hdmi_ip_data *ip_data);
-void ti_hdmi_4xxx_wp_dump(struct hdmi_ip_data *ip_data, struct seq_file *s);
-void ti_hdmi_4xxx_pll_dump(struct hdmi_ip_data *ip_data, struct seq_file *s);
-void ti_hdmi_4xxx_core_dump(struct hdmi_ip_data *ip_data, struct seq_file *s);
-void ti_hdmi_4xxx_phy_dump(struct hdmi_ip_data *ip_data, struct seq_file *s);
-#if defined(CONFIG_OMAP4_DSS_HDMI_AUDIO)
-int hdmi_compute_acr(u32 sample_freq, u32 *n, u32 *cts);
-int ti_hdmi_4xxx_wp_audio_enable(struct hdmi_ip_data *ip_data);
-void ti_hdmi_4xxx_wp_audio_disable(struct hdmi_ip_data *ip_data);
-int ti_hdmi_4xxx_audio_start(struct hdmi_ip_data *ip_data);
-void ti_hdmi_4xxx_audio_stop(struct hdmi_ip_data *ip_data);
-int ti_hdmi_4xxx_audio_config(struct hdmi_ip_data *ip_data,
-		struct omap_dss_audio *audio);
-int ti_hdmi_4xxx_audio_get_dma_port(u32 *offset, u32 *size);
-#endif
-#endif
diff --git a/drivers/video/p9100.c b/drivers/video/p9100.c
index 4b23af6e5c28..367cea8f43f3 100644
--- a/drivers/video/p9100.c
+++ b/drivers/video/p9100.c
@@ -339,8 +339,6 @@ static int p9100_remove(struct platform_device *op)
 
 	framebuffer_release(info);
 
-	dev_set_drvdata(&op->dev, NULL);
-
 	return 0;
 }
 
diff --git a/drivers/video/platinumfb.c b/drivers/video/platinumfb.c
index 3d86bac62d3e..4c9299576827 100644
--- a/drivers/video/platinumfb.c
+++ b/drivers/video/platinumfb.c
@@ -403,7 +403,7 @@ try_again:
 	if (rc < 0)
 		return rc;
 
-	printk(KERN_INFO "fb%d: Apple Platinum frame buffer device\n", info->node);
+	fb_info(info, "Apple Platinum frame buffer device\n");
 
 	return 0;
 }
@@ -639,7 +639,6 @@ static int platinumfb_probe(struct platform_device* odev)
 		iounmap(pinfo->frame_buffer);
 		iounmap(pinfo->platinum_regs);
 		iounmap(pinfo->cmap_regs);
-		dev_set_drvdata(&odev->dev, NULL);
 		framebuffer_release(info);
 	}
 
diff --git a/drivers/video/pm2fb.c b/drivers/video/pm2fb.c
index 81354eeab021..3b85b647bc10 100644
--- a/drivers/video/pm2fb.c
+++ b/drivers/video/pm2fb.c
@@ -1694,8 +1694,8 @@ static int pm2fb_probe(struct pci_dev *pdev, const struct pci_device_id *id)
 	if (retval < 0)
 		goto err_exit_all;
 
-	printk(KERN_INFO "fb%d: %s frame buffer device, memory = %dK.\n",
-	       info->node, info->fix.id, pm2fb_fix.smem_len / 1024);
+	fb_info(info, "%s frame buffer device, memory = %dK\n",
+		info->fix.id, pm2fb_fix.smem_len / 1024);
 
 	/*
 	 * Our driver data
@@ -1744,7 +1744,6 @@ static void pm2fb_remove(struct pci_dev *pdev)
 	iounmap(par->v_regs);
 	release_mem_region(fix->mmio_start, fix->mmio_len);
 
-	pci_set_drvdata(pdev, NULL);
 	fb_dealloc_cmap(&info->cmap);
 	kfree(info->pixmap.addr);
 	framebuffer_release(info);
diff --git a/drivers/video/pm3fb.c b/drivers/video/pm3fb.c
index 7718faa4a73b..4bf3273d0433 100644
--- a/drivers/video/pm3fb.c
+++ b/drivers/video/pm3fb.c
@@ -1445,8 +1445,7 @@ static int pm3fb_probe(struct pci_dev *dev, const struct pci_device_id *ent)
 		retval = -EINVAL;
 		goto err_exit_all;
 	}
-	printk(KERN_INFO "fb%d: %s frame buffer device\n", info->node,
-	   info->fix.id);
+	fb_info(info, "%s frame buffer device\n", info->fix.id);
 	pci_set_drvdata(dev, info);
 	return 0;
 
@@ -1489,7 +1488,6 @@ static void pm3fb_remove(struct pci_dev *dev)
 		iounmap(par->v_regs);
 		release_mem_region(fix->mmio_start, fix->mmio_len);
 
-		pci_set_drvdata(dev, NULL);
 		kfree(info->pixmap.addr);
 		framebuffer_release(info);
 	}
diff --git a/drivers/video/pmag-ba-fb.c b/drivers/video/pmag-ba-fb.c
index d1e46cedb1f7..914a52ba8477 100644
--- a/drivers/video/pmag-ba-fb.c
+++ b/drivers/video/pmag-ba-fb.c
@@ -212,8 +212,8 @@ static int pmagbafb_probe(struct device *dev)
 
 	get_device(dev);
 
-	pr_info("fb%d: %s frame buffer device at %s\n",
-		info->node, info->fix.id, dev_name(dev));
+	fb_info(info, "%s frame buffer device at %s\n",
+		info->fix.id, dev_name(dev));
 
 	return 0;
 
diff --git a/drivers/video/pmagb-b-fb.c b/drivers/video/pmagb-b-fb.c
index 0e1317400328..0822b6f8dddc 100644
--- a/drivers/video/pmagb-b-fb.c
+++ b/drivers/video/pmagb-b-fb.c
@@ -328,11 +328,10 @@ static int pmagbbfb_probe(struct device *dev)
 	snprintf(freq1, sizeof(freq1), "%u.%03uMHz",
 		 par->osc1 / 1000, par->osc1 % 1000);
 
-	pr_info("fb%d: %s frame buffer device at %s\n",
-		info->node, info->fix.id, dev_name(dev));
-	pr_info("fb%d: Osc0: %s, Osc1: %s, Osc%u selected\n",
-		info->node, freq0, par->osc1 ? freq1 : "disabled",
-		par->osc1 != 0);
+	fb_info(info, "%s frame buffer device at %s\n",
+		info->fix.id, dev_name(dev));
+	fb_info(info, "Osc0: %s, Osc1: %s, Osc%u selected\n",
+		freq0, par->osc1 ? freq1 : "disabled", par->osc1 != 0);
 
 	return 0;
 
diff --git a/drivers/video/pvr2fb.c b/drivers/video/pvr2fb.c
index df07860563e6..167cffff3d4e 100644
--- a/drivers/video/pvr2fb.c
+++ b/drivers/video/pvr2fb.c
@@ -817,24 +817,25 @@ static int pvr2fb_common_init(void)
 
 	rev = fb_readl(par->mmio_base + 0x04);
 
-	printk("fb%d: %s (rev %ld.%ld) frame buffer device, using %ldk/%ldk of video memory\n",
-	       fb_info->node, fb_info->fix.id, (rev >> 4) & 0x0f, rev & 0x0f,
-	       modememused >> 10, (unsigned long)(fb_info->fix.smem_len >> 10));
-	printk("fb%d: Mode %dx%d-%d pitch = %ld cable: %s video output: %s\n",
-	       fb_info->node, fb_info->var.xres, fb_info->var.yres,
-	       fb_info->var.bits_per_pixel,
-	       get_line_length(fb_info->var.xres, fb_info->var.bits_per_pixel),
-	       (char *)pvr2_get_param(cables, NULL, cable_type, 3),
-	       (char *)pvr2_get_param(outputs, NULL, video_output, 3));
+	fb_info(fb_info, "%s (rev %ld.%ld) frame buffer device, using %ldk/%ldk of video memory\n",
+		fb_info->fix.id, (rev >> 4) & 0x0f, rev & 0x0f,
+		modememused >> 10,
+		(unsigned long)(fb_info->fix.smem_len >> 10));
+	fb_info(fb_info, "Mode %dx%d-%d pitch = %ld cable: %s video output: %s\n",
+		fb_info->var.xres, fb_info->var.yres,
+		fb_info->var.bits_per_pixel,
+		get_line_length(fb_info->var.xres, fb_info->var.bits_per_pixel),
+		(char *)pvr2_get_param(cables, NULL, cable_type, 3),
+		(char *)pvr2_get_param(outputs, NULL, video_output, 3));
 
 #ifdef CONFIG_SH_STORE_QUEUES
-	printk(KERN_NOTICE "fb%d: registering with SQ API\n", fb_info->node);
+	fb_notice(fb_info, "registering with SQ API\n");
 
 	pvr2fb_map = sq_remap(fb_info->fix.smem_start, fb_info->fix.smem_len,
 			      fb_info->fix.id, PAGE_SHARED);
 
-	printk(KERN_NOTICE "fb%d: Mapped video memory to SQ addr 0x%lx\n",
-	       fb_info->node, pvr2fb_map);
+	fb_notice(fb_info, "Mapped video memory to SQ addr 0x%lx\n",
+		  pvr2fb_map);
 #endif
 
 	return 0;
diff --git a/drivers/video/pxa168fb.c b/drivers/video/pxa168fb.c
index aa9bd1f76d60..c95b9e46d48f 100644
--- a/drivers/video/pxa168fb.c
+++ b/drivers/video/pxa168fb.c
@@ -364,7 +364,7 @@ static void set_graphics_start(struct fb_info *info, int xoffset, int yoffset)
 static void set_dumb_panel_control(struct fb_info *info)
 {
 	struct pxa168fb_info *fbi = info->par;
-	struct pxa168fb_mach_info *mi = fbi->dev->platform_data;
+	struct pxa168fb_mach_info *mi = dev_get_platdata(fbi->dev);
 	u32 x;
 
 	/*
@@ -407,7 +407,7 @@ static int pxa168fb_set_par(struct fb_info *info)
 	u32 x;
 	struct pxa168fb_mach_info *mi;
 
-	mi = fbi->dev->platform_data;
+	mi = dev_get_platdata(fbi->dev);
 
 	/*
 	 * Set additional mode info.
@@ -609,7 +609,7 @@ static int pxa168fb_probe(struct platform_device *pdev)
 	struct clk *clk;
 	int irq, ret;
 
-	mi = pdev->dev.platform_data;
+	mi = dev_get_platdata(&pdev->dev);
 	if (mi == NULL) {
 		dev_err(&pdev->dev, "no platform data defined\n");
 		return -EINVAL;
diff --git a/drivers/video/pxafb.c b/drivers/video/pxafb.c
index eca2de45f7a6..1ecd9cec2921 100644
--- a/drivers/video/pxafb.c
+++ b/drivers/video/pxafb.c
@@ -457,7 +457,7 @@ static int pxafb_adjust_timing(struct pxafb_info *fbi,
 static int pxafb_check_var(struct fb_var_screeninfo *var, struct fb_info *info)
 {
 	struct pxafb_info *fbi = (struct pxafb_info *)info;
-	struct pxafb_mach_info *inf = fbi->dev->platform_data;
+	struct pxafb_mach_info *inf = dev_get_platdata(fbi->dev);
 	int err;
 
 	if (inf->fixed_modes) {
@@ -1230,7 +1230,7 @@ static unsigned int __smart_timing(unsigned time_ns, unsigned long lcd_clk)
 static void setup_smart_timing(struct pxafb_info *fbi,
 				struct fb_var_screeninfo *var)
 {
-	struct pxafb_mach_info *inf = fbi->dev->platform_data;
+	struct pxafb_mach_info *inf = dev_get_platdata(fbi->dev);
 	struct pxafb_mode_info *mode = &inf->modes[0];
 	unsigned long lclk = clk_get_rate(fbi->clk);
 	unsigned t1, t2, t3, t4;
@@ -1258,14 +1258,14 @@ static void setup_smart_timing(struct pxafb_info *fbi,
 static int pxafb_smart_thread(void *arg)
 {
 	struct pxafb_info *fbi = arg;
-	struct pxafb_mach_info *inf = fbi->dev->platform_data;
+	struct pxafb_mach_info *inf = dev_get_platdata(fbi->dev);
 
 	if (!inf->smart_update) {
 		pr_err("%s: not properly initialized, thread terminated\n",
 				__func__);
 		return -EINVAL;
 	}
-	inf = fbi->dev->platform_data;
+	inf = dev_get_platdata(fbi->dev);
 
 	pr_debug("%s(): task starting\n", __func__);
 
@@ -1793,7 +1793,7 @@ static struct pxafb_info *pxafb_init_fbinfo(struct device *dev)
 {
 	struct pxafb_info *fbi;
 	void *addr;
-	struct pxafb_mach_info *inf = dev->platform_data;
+	struct pxafb_mach_info *inf = dev_get_platdata(dev);
 
 	/* Alloc the pxafb_info and pseudo_palette in one step */
 	fbi = kmalloc(sizeof(struct pxafb_info) + sizeof(u32) * 16, GFP_KERNEL);
@@ -1855,7 +1855,7 @@ static struct pxafb_info *pxafb_init_fbinfo(struct device *dev)
 #ifdef CONFIG_FB_PXA_PARAMETERS
 static int parse_opt_mode(struct device *dev, const char *this_opt)
 {
-	struct pxafb_mach_info *inf = dev->platform_data;
+	struct pxafb_mach_info *inf = dev_get_platdata(dev);
 
 	const char *name = this_opt+5;
 	unsigned int namelen = strlen(name);
@@ -1914,7 +1914,7 @@ done:
 
 static int parse_opt(struct device *dev, char *this_opt)
 {
-	struct pxafb_mach_info *inf = dev->platform_data;
+	struct pxafb_mach_info *inf = dev_get_platdata(dev);
 	struct pxafb_mode_info *mode = &inf->modes[0];
 	char s[64];
 
@@ -2102,7 +2102,7 @@ static int pxafb_probe(struct platform_device *dev)
 
 	dev_dbg(&dev->dev, "pxafb_probe\n");
 
-	inf = dev->dev.platform_data;
+	inf = dev_get_platdata(&dev->dev);
 	ret = -ENOMEM;
 	fbi = NULL;
 	if (!inf)
diff --git a/drivers/video/q40fb.c b/drivers/video/q40fb.c
index d44c7351de0f..7487f76f6275 100644
--- a/drivers/video/q40fb.c
+++ b/drivers/video/q40fb.c
@@ -119,8 +119,7 @@ static int q40fb_probe(struct platform_device *dev)
 		return -EINVAL;
 	}
 
-        printk(KERN_INFO "fb%d: Q40 frame buffer alive and kicking !\n",
-	       info->node);
+	fb_info(info, "Q40 frame buffer alive and kicking !\n");
 	return 0;
 }
 
diff --git a/drivers/video/riva/fbdev.c b/drivers/video/riva/fbdev.c
index 9536715b5a1b..a5514acd2ac6 100644
--- a/drivers/video/riva/fbdev.c
+++ b/drivers/video/riva/fbdev.c
@@ -1185,11 +1185,6 @@ static int rivafb_check_var(struct fb_var_screeninfo *var, struct fb_info *info)
 	if (rivafb_do_maximize(info, var, nom, den) < 0)
 		return -EINVAL;
 
-	if (var->xoffset < 0)
-		var->xoffset = 0;
-	if (var->yoffset < 0)
-		var->yoffset = 0;
-
 	/* truncate xoffset and yoffset to maximum if too high */
 	if (var->xoffset > var->xres_virtual - var->xres)
 		var->xoffset = var->xres_virtual - var->xres - 1;
diff --git a/drivers/video/s1d13xxxfb.c b/drivers/video/s1d13xxxfb.c
index 05c2dc3d4bc0..83433cb0dfba 100644
--- a/drivers/video/s1d13xxxfb.c
+++ b/drivers/video/s1d13xxxfb.c
@@ -777,8 +777,8 @@ static int s1d13xxxfb_probe(struct platform_device *pdev)
 	printk(KERN_INFO "Epson S1D13XXX FB Driver\n");
 
 	/* enable platform-dependent hardware glue, if any */
-	if (pdev->dev.platform_data)
-		pdata = pdev->dev.platform_data;
+	if (dev_get_platdata(&pdev->dev))
+		pdata = dev_get_platdata(&pdev->dev);
 
 	if (pdata && pdata->platform_init_video)
 		pdata->platform_init_video();
@@ -901,8 +901,7 @@ static int s1d13xxxfb_probe(struct platform_device *pdev)
 		goto bail;
 	}
 
-	printk(KERN_INFO "fb%d: %s frame buffer device\n",
-	       info->node, info->fix.id);
+	fb_info(info, "%s frame buffer device\n", info->fix.id);
 
 	return 0;
 
@@ -923,8 +922,8 @@ static int s1d13xxxfb_suspend(struct platform_device *dev, pm_message_t state)
 	lcd_enable(s1dfb, 0);
 	crt_enable(s1dfb, 0);
 
-	if (dev->dev.platform_data)
-		pdata = dev->dev.platform_data;
+	if (dev_get_platdata(&dev->dev))
+		pdata = dev_get_platdata(&dev->dev);
 
 #if 0
 	if (!s1dfb->disp_save)
@@ -973,8 +972,8 @@ static int s1d13xxxfb_resume(struct platform_device *dev)
 	while ((s1d13xxxfb_readreg(s1dfb, S1DREG_PS_STATUS) & 0x01))
 		udelay(10);
 
-	if (dev->dev.platform_data)
-		pdata = dev->dev.platform_data;
+	if (dev_get_platdata(&dev->dev))
+		pdata = dev_get_platdata(&dev->dev);
 
 	if (s1dfb->regs_save) {
 		/* will write RO regs, *should* get away with it :) */
diff --git a/drivers/video/s3c-fb.c b/drivers/video/s3c-fb.c
index 2e7991c7ca08..62acae2694a9 100644
--- a/drivers/video/s3c-fb.c
+++ b/drivers/video/s3c-fb.c
@@ -1378,7 +1378,7 @@ static int s3c_fb_probe(struct platform_device *pdev)
 		return -EINVAL;
 	}
 
-	pd = pdev->dev.platform_data;
+	pd = dev_get_platdata(&pdev->dev);
 	if (!pd) {
 		dev_err(dev, "no platform data specified\n");
 		return -EINVAL;
diff --git a/drivers/video/s3c2410fb.c b/drivers/video/s3c2410fb.c
index 21a32adbb8ea..81af5a63e9e1 100644
--- a/drivers/video/s3c2410fb.c
+++ b/drivers/video/s3c2410fb.c
@@ -123,7 +123,7 @@ static int s3c2410fb_check_var(struct fb_var_screeninfo *var,
 			       struct fb_info *info)
 {
 	struct s3c2410fb_info *fbi = info->par;
-	struct s3c2410fb_mach_info *mach_info = fbi->dev->platform_data;
+	struct s3c2410fb_mach_info *mach_info = dev_get_platdata(fbi->dev);
 	struct s3c2410fb_display *display = NULL;
 	struct s3c2410fb_display *default_display = mach_info->displays +
 						    mach_info->default_display;
@@ -686,7 +686,7 @@ static inline void modify_gpio(void __iomem *reg,
 static int s3c2410fb_init_registers(struct fb_info *info)
 {
 	struct s3c2410fb_info *fbi = info->par;
-	struct s3c2410fb_mach_info *mach_info = fbi->dev->platform_data;
+	struct s3c2410fb_mach_info *mach_info = dev_get_platdata(fbi->dev);
 	unsigned long flags;
 	void __iomem *regs = fbi->io;
 	void __iomem *tpal;
@@ -833,7 +833,7 @@ static int s3c24xxfb_probe(struct platform_device *pdev,
 	int size;
 	u32 lcdcon1;
 
-	mach_info = pdev->dev.platform_data;
+	mach_info = dev_get_platdata(&pdev->dev);
 	if (mach_info == NULL) {
 		dev_err(&pdev->dev,
 			"no platform data for lcd, cannot attach\n");
diff --git a/drivers/video/s3fb.c b/drivers/video/s3fb.c
index d838ba829459..968b2997175a 100644
--- a/drivers/video/s3fb.c
+++ b/drivers/video/s3fb.c
@@ -306,8 +306,8 @@ static void s3fb_settile_fast(struct fb_info *info, struct fb_tilemap *map)
 
 	if ((map->width != 8) || (map->height != 16) ||
 	    (map->depth != 1) || (map->length != 256)) {
-	    	printk(KERN_ERR "fb%d: unsupported font parameters: width %d, height %d, depth %d, length %d\n",
-			info->node, map->width, map->height, map->depth, map->length);
+		fb_err(info, "unsupported font parameters: width %d, height %d, depth %d, length %d\n",
+		       map->width, map->height, map->depth, map->length);
 		return;
 	}
 
@@ -476,7 +476,7 @@ static void s3_set_pixclock(struct fb_info *info, u32 pixclock)
 	rv = svga_compute_pll((par->chip == CHIP_365_TRIO3D) ? &s3_trio3d_pll : &s3_pll,
 			      1000000000 / pixclock, &m, &n, &r, info->node);
 	if (rv < 0) {
-		printk(KERN_ERR "fb%d: cannot set requested pixclock, keeping old value\n", info->node);
+		fb_err(info, "cannot set requested pixclock, keeping old value\n");
 		return;
 	}
 
@@ -569,7 +569,7 @@ static int s3fb_check_var(struct fb_var_screeninfo *var, struct fb_info *info)
 		rv = -EINVAL;
 
 	if (rv < 0) {
-		printk(KERN_ERR "fb%d: unsupported mode requested\n", info->node);
+		fb_err(info, "unsupported mode requested\n");
 		return rv;
 	}
 
@@ -587,22 +587,21 @@ static int s3fb_check_var(struct fb_var_screeninfo *var, struct fb_info *info)
 	/* Check whether have enough memory */
 	mem = ((var->bits_per_pixel * var->xres_virtual) >> 3) * var->yres_virtual;
 	if (mem > info->screen_size) {
-		printk(KERN_ERR "fb%d: not enough framebuffer memory (%d kB requested , %d kB available)\n",
-			info->node, mem >> 10, (unsigned int) (info->screen_size >> 10));
+		fb_err(info, "not enough framebuffer memory (%d kB requested , %u kB available)\n",
+		       mem >> 10, (unsigned int) (info->screen_size >> 10));
 		return -EINVAL;
 	}
 
 	rv = svga_check_timings (&s3_timing_regs, var, info->node);
 	if (rv < 0) {
-		printk(KERN_ERR "fb%d: invalid timings requested\n", info->node);
+		fb_err(info, "invalid timings requested\n");
 		return rv;
 	}
 
 	rv = svga_compute_pll(&s3_pll, PICOS2KHZ(var->pixclock), &m, &n, &r,
 				info->node);
 	if (rv < 0) {
-		printk(KERN_ERR "fb%d: invalid pixclock value requested\n",
-			info->node);
+		fb_err(info, "invalid pixclock value requested\n");
 		return rv;
 	}
 
@@ -686,7 +685,7 @@ static int s3fb_set_par(struct fb_info *info)
 
 
 	/* Set the offset register */
-	pr_debug("fb%d: offset register       : %d\n", info->node, offset_value);
+	fb_dbg(info, "offset register       : %d\n", offset_value);
 	svga_wcrt_multi(par->state.vgabase, s3_offset_regs, offset_value);
 
 	if (par->chip != CHIP_357_VIRGE_GX2 &&
@@ -769,7 +768,7 @@ static int s3fb_set_par(struct fb_info *info)
 	/* Set mode-specific register values */
 	switch (mode) {
 	case 0:
-		pr_debug("fb%d: text mode\n", info->node);
+		fb_dbg(info, "text mode\n");
 		svga_set_textmode_vga_regs(par->state.vgabase);
 
 		/* Set additional registers like in 8-bit mode */
@@ -780,12 +779,12 @@ static int s3fb_set_par(struct fb_info *info)
 		svga_wcrt_mask(par->state.vgabase, 0x3A, 0x00, 0x30);
 
 		if (fasttext) {
-			pr_debug("fb%d: high speed text mode set\n", info->node);
+			fb_dbg(info, "high speed text mode set\n");
 			svga_wcrt_mask(par->state.vgabase, 0x31, 0x40, 0x40);
 		}
 		break;
 	case 1:
-		pr_debug("fb%d: 4 bit pseudocolor\n", info->node);
+		fb_dbg(info, "4 bit pseudocolor\n");
 		vga_wgfx(par->state.vgabase, VGA_GFX_MODE, 0x40);
 
 		/* Set additional registers like in 8-bit mode */
@@ -796,7 +795,7 @@ static int s3fb_set_par(struct fb_info *info)
 		svga_wcrt_mask(par->state.vgabase, 0x3A, 0x00, 0x30);
 		break;
 	case 2:
-		pr_debug("fb%d: 4 bit pseudocolor, planar\n", info->node);
+		fb_dbg(info, "4 bit pseudocolor, planar\n");
 
 		/* Set additional registers like in 8-bit mode */
 		svga_wcrt_mask(par->state.vgabase, 0x50, 0x00, 0x30);
@@ -806,7 +805,7 @@ static int s3fb_set_par(struct fb_info *info)
 		svga_wcrt_mask(par->state.vgabase, 0x3A, 0x00, 0x30);
 		break;
 	case 3:
-		pr_debug("fb%d: 8 bit pseudocolor\n", info->node);
+		fb_dbg(info, "8 bit pseudocolor\n");
 		svga_wcrt_mask(par->state.vgabase, 0x50, 0x00, 0x30);
 		if (info->var.pixclock > 20000 ||
 		    par->chip == CHIP_357_VIRGE_GX2 ||
@@ -822,7 +821,7 @@ static int s3fb_set_par(struct fb_info *info)
 		}
 		break;
 	case 4:
-		pr_debug("fb%d: 5/5/5 truecolor\n", info->node);
+		fb_dbg(info, "5/5/5 truecolor\n");
 		if (par->chip == CHIP_988_VIRGE_VX) {
 			if (info->var.pixclock > 20000)
 				svga_wcrt_mask(par->state.vgabase, 0x67, 0x20, 0xF0);
@@ -850,7 +849,7 @@ static int s3fb_set_par(struct fb_info *info)
 		}
 		break;
 	case 5:
-		pr_debug("fb%d: 5/6/5 truecolor\n", info->node);
+		fb_dbg(info, "5/6/5 truecolor\n");
 		if (par->chip == CHIP_988_VIRGE_VX) {
 			if (info->var.pixclock > 20000)
 				svga_wcrt_mask(par->state.vgabase, 0x67, 0x40, 0xF0);
@@ -879,16 +878,16 @@ static int s3fb_set_par(struct fb_info *info)
 		break;
 	case 6:
 		/* VIRGE VX case */
-		pr_debug("fb%d: 8/8/8 truecolor\n", info->node);
+		fb_dbg(info, "8/8/8 truecolor\n");
 		svga_wcrt_mask(par->state.vgabase, 0x67, 0xD0, 0xF0);
 		break;
 	case 7:
-		pr_debug("fb%d: 8/8/8/8 truecolor\n", info->node);
+		fb_dbg(info, "8/8/8/8 truecolor\n");
 		svga_wcrt_mask(par->state.vgabase, 0x50, 0x30, 0x30);
 		svga_wcrt_mask(par->state.vgabase, 0x67, 0xD0, 0xF0);
 		break;
 	default:
-		printk(KERN_ERR "fb%d: unsupported mode - bug\n", info->node);
+		fb_err(info, "unsupported mode - bug\n");
 		return -EINVAL;
 	}
 
@@ -991,27 +990,27 @@ static int s3fb_blank(int blank_mode, struct fb_info *info)
 
 	switch (blank_mode) {
 	case FB_BLANK_UNBLANK:
-		pr_debug("fb%d: unblank\n", info->node);
+		fb_dbg(info, "unblank\n");
 		svga_wcrt_mask(par->state.vgabase, 0x56, 0x00, 0x06);
 		svga_wseq_mask(par->state.vgabase, 0x01, 0x00, 0x20);
 		break;
 	case FB_BLANK_NORMAL:
-		pr_debug("fb%d: blank\n", info->node);
+		fb_dbg(info, "blank\n");
 		svga_wcrt_mask(par->state.vgabase, 0x56, 0x00, 0x06);
 		svga_wseq_mask(par->state.vgabase, 0x01, 0x20, 0x20);
 		break;
 	case FB_BLANK_HSYNC_SUSPEND:
-		pr_debug("fb%d: hsync\n", info->node);
+		fb_dbg(info, "hsync\n");
 		svga_wcrt_mask(par->state.vgabase, 0x56, 0x02, 0x06);
 		svga_wseq_mask(par->state.vgabase, 0x01, 0x20, 0x20);
 		break;
 	case FB_BLANK_VSYNC_SUSPEND:
-		pr_debug("fb%d: vsync\n", info->node);
+		fb_dbg(info, "vsync\n");
 		svga_wcrt_mask(par->state.vgabase, 0x56, 0x04, 0x06);
 		svga_wseq_mask(par->state.vgabase, 0x01, 0x20, 0x20);
 		break;
 	case FB_BLANK_POWERDOWN:
-		pr_debug("fb%d: sync down\n", info->node);
+		fb_dbg(info, "sync down\n");
 		svga_wcrt_mask(par->state.vgabase, 0x56, 0x06, 0x06);
 		svga_wseq_mask(par->state.vgabase, 0x01, 0x20, 0x20);
 		break;
@@ -1352,13 +1351,16 @@ static int s3_pci_probe(struct pci_dev *dev, const struct pci_device_id *id)
 		goto err_reg_fb;
 	}
 
-	printk(KERN_INFO "fb%d: %s on %s, %d MB RAM, %d MHz MCLK\n", info->node, info->fix.id,
-		 pci_name(dev), info->fix.smem_len >> 20, (par->mclk_freq + 500) / 1000);
+	fb_info(info, "%s on %s, %d MB RAM, %d MHz MCLK\n",
+		info->fix.id, pci_name(dev),
+		info->fix.smem_len >> 20, (par->mclk_freq + 500) / 1000);
 
 	if (par->chip == CHIP_UNKNOWN)
-		printk(KERN_INFO "fb%d: unknown chip, CR2D=%x, CR2E=%x, CRT2F=%x, CRT30=%x\n",
-			info->node, vga_rcrt(par->state.vgabase, 0x2d), vga_rcrt(par->state.vgabase, 0x2e),
-			vga_rcrt(par->state.vgabase, 0x2f), vga_rcrt(par->state.vgabase, 0x30));
+		fb_info(info, "unknown chip, CR2D=%x, CR2E=%x, CRT2F=%x, CRT30=%x\n",
+			vga_rcrt(par->state.vgabase, 0x2d),
+			vga_rcrt(par->state.vgabase, 0x2e),
+			vga_rcrt(par->state.vgabase, 0x2f),
+			vga_rcrt(par->state.vgabase, 0x30));
 
 	/* Record a reference to the driver data */
 	pci_set_drvdata(dev, info);
@@ -1424,7 +1426,6 @@ static void s3_pci_remove(struct pci_dev *dev)
 		pci_release_regions(dev);
 /*		pci_disable_device(dev); */
 
-		pci_set_drvdata(dev, NULL);
 		framebuffer_release(info);
 	}
 }
diff --git a/drivers/video/sa1100fb.c b/drivers/video/sa1100fb.c
index de76da0c6429..580c444ec301 100644
--- a/drivers/video/sa1100fb.c
+++ b/drivers/video/sa1100fb.c
@@ -1116,7 +1116,7 @@ static struct fb_monspecs monspecs = {
 
 static struct sa1100fb_info *sa1100fb_init_fbinfo(struct device *dev)
 {
-	struct sa1100fb_mach_info *inf = dev->platform_data;
+	struct sa1100fb_mach_info *inf = dev_get_platdata(dev);
 	struct sa1100fb_info *fbi;
 	unsigned i;
 
@@ -1201,7 +1201,7 @@ static int sa1100fb_probe(struct platform_device *pdev)
 	struct resource *res;
 	int ret, irq;
 
-	if (!pdev->dev.platform_data) {
+	if (!dev_get_platdata(&pdev->dev)) {
 		dev_err(&pdev->dev, "no platform LCD data\n");
 		return -EINVAL;
 	}
diff --git a/drivers/video/savage/savagefb_driver.c b/drivers/video/savage/savagefb_driver.c
index 741b2395d01e..4dbf45f3b21a 100644
--- a/drivers/video/savage/savagefb_driver.c
+++ b/drivers/video/savage/savagefb_driver.c
@@ -2362,12 +2362,6 @@ static void savagefb_remove(struct pci_dev *dev)
 		kfree(info->pixmap.addr);
 		pci_release_regions(dev);
 		framebuffer_release(info);
-
-		/*
-		 * Ensure that the driver data is no longer
-		 * valid.
-		 */
-		pci_set_drvdata(dev, NULL);
 	}
 }
 
diff --git a/drivers/video/sbuslib.c b/drivers/video/sbuslib.c
index 296afae442f4..a350209ffbd3 100644
--- a/drivers/video/sbuslib.c
+++ b/drivers/video/sbuslib.c
@@ -186,7 +186,7 @@ int sbusfb_ioctl_helper(unsigned long cmd, unsigned long arg,
 	}
 	default:
 		return -EINVAL;
-	};
+	}
 }
 EXPORT_SYMBOL(sbusfb_ioctl_helper);
 
diff --git a/drivers/video/sgivwfb.c b/drivers/video/sgivwfb.c
index a9ac3ce2d0e9..bc74d0408998 100644
--- a/drivers/video/sgivwfb.c
+++ b/drivers/video/sgivwfb.c
@@ -803,8 +803,8 @@ static int sgivwfb_probe(struct platform_device *dev)
 
 	platform_set_drvdata(dev, info);
 
-	printk(KERN_INFO "fb%d: SGI DBE frame buffer device, using %ldK of video memory at %#lx\n",      
-		info->node, sgivwfb_mem_size >> 10, sgivwfb_mem_phys);
+	fb_info(info, "SGI DBE frame buffer device, using %ldK of video memory at %#lx\n",
+		sgivwfb_mem_size >> 10, sgivwfb_mem_phys);
 	return 0;
 
 fail_register_framebuffer:
diff --git a/drivers/video/sh_mobile_hdmi.c b/drivers/video/sh_mobile_hdmi.c
index bfe4728480fd..9a33ee0413fb 100644
--- a/drivers/video/sh_mobile_hdmi.c
+++ b/drivers/video/sh_mobile_hdmi.c
@@ -498,7 +498,7 @@ static void sh_hdmi_video_config(struct sh_hdmi *hdmi)
 static void sh_hdmi_audio_config(struct sh_hdmi *hdmi)
 {
 	u8 data;
-	struct sh_mobile_hdmi_info *pdata = hdmi->dev->platform_data;
+	struct sh_mobile_hdmi_info *pdata = dev_get_platdata(hdmi->dev);
 
 	/*
 	 * [7:4] L/R data swap control
@@ -815,7 +815,7 @@ static unsigned long sh_hdmi_rate_error(struct sh_hdmi *hdmi,
 		unsigned long *hdmi_rate, unsigned long *parent_rate)
 {
 	unsigned long target = PICOS2KHZ(mode->pixclock) * 1000, rate_error;
-	struct sh_mobile_hdmi_info *pdata = hdmi->dev->platform_data;
+	struct sh_mobile_hdmi_info *pdata = dev_get_platdata(hdmi->dev);
 
 	*hdmi_rate = clk_round_rate(hdmi->hdmi_clk, target);
 	if ((long)*hdmi_rate < 0)
@@ -1271,7 +1271,7 @@ static void sh_hdmi_htop1_init(struct sh_hdmi *hdmi)
 
 static int __init sh_hdmi_probe(struct platform_device *pdev)
 {
-	struct sh_mobile_hdmi_info *pdata = pdev->dev.platform_data;
+	struct sh_mobile_hdmi_info *pdata = dev_get_platdata(&pdev->dev);
 	struct resource *res = platform_get_resource(pdev, IORESOURCE_MEM, 0);
 	struct resource *htop1_res;
 	int irq = platform_get_irq(pdev, 0), ret;
@@ -1290,7 +1290,7 @@ static int __init sh_hdmi_probe(struct platform_device *pdev)
 		}
 	}
 
-	hdmi = kzalloc(sizeof(*hdmi), GFP_KERNEL);
+	hdmi = devm_kzalloc(&pdev->dev, sizeof(*hdmi), GFP_KERNEL);
 	if (!hdmi) {
 		dev_err(&pdev->dev, "Cannot allocate device data\n");
 		return -ENOMEM;
@@ -1304,7 +1304,7 @@ static int __init sh_hdmi_probe(struct platform_device *pdev)
 	if (IS_ERR(hdmi->hdmi_clk)) {
 		ret = PTR_ERR(hdmi->hdmi_clk);
 		dev_err(&pdev->dev, "Unable to get clock: %d\n", ret);
-		goto egetclk;
+		return ret;
 	}
 
 	/* select register access functions */
@@ -1326,7 +1326,7 @@ static int __init sh_hdmi_probe(struct platform_device *pdev)
 		goto erate;
 	}
 
-	ret = clk_enable(hdmi->hdmi_clk);
+	ret = clk_prepare_enable(hdmi->hdmi_clk);
 	if (ret < 0) {
 		dev_err(hdmi->dev, "Cannot enable clock: %d\n", ret);
 		goto erate;
@@ -1404,11 +1404,9 @@ emap_htop1:
 emap:
 	release_mem_region(res->start, resource_size(res));
 ereqreg:
-	clk_disable(hdmi->hdmi_clk);
+	clk_disable_unprepare(hdmi->hdmi_clk);
 erate:
 	clk_put(hdmi->hdmi_clk);
-egetclk:
-	kfree(hdmi);
 
 	return ret;
 }
@@ -1427,13 +1425,12 @@ static int __exit sh_hdmi_remove(struct platform_device *pdev)
 	cancel_delayed_work_sync(&hdmi->edid_work);
 	pm_runtime_put(&pdev->dev);
 	pm_runtime_disable(&pdev->dev);
-	clk_disable(hdmi->hdmi_clk);
+	clk_disable_unprepare(hdmi->hdmi_clk);
 	clk_put(hdmi->hdmi_clk);
 	if (hdmi->htop1)
 		iounmap(hdmi->htop1);
 	iounmap(hdmi->base);
 	release_mem_region(res->start, resource_size(res));
-	kfree(hdmi);
 
 	return 0;
 }
diff --git a/drivers/video/sh_mobile_lcdcfb.c b/drivers/video/sh_mobile_lcdcfb.c
index 0264704a52be..ab85ad6c25ec 100644
--- a/drivers/video/sh_mobile_lcdcfb.c
+++ b/drivers/video/sh_mobile_lcdcfb.c
@@ -344,7 +344,7 @@ static void sh_mobile_lcdc_clk_on(struct sh_mobile_lcdc_priv *priv)
 {
 	if (atomic_inc_and_test(&priv->hw_usecnt)) {
 		if (priv->dot_clk)
-			clk_enable(priv->dot_clk);
+			clk_prepare_enable(priv->dot_clk);
 		pm_runtime_get_sync(priv->dev);
 		if (priv->meram_dev && priv->meram_dev->pdev)
 			pm_runtime_get_sync(&priv->meram_dev->pdev->dev);
@@ -358,7 +358,7 @@ static void sh_mobile_lcdc_clk_off(struct sh_mobile_lcdc_priv *priv)
 			pm_runtime_put_sync(&priv->meram_dev->pdev->dev);
 		pm_runtime_put(priv->dev);
 		if (priv->dot_clk)
-			clk_disable(priv->dot_clk);
+			clk_disable_unprepare(priv->dot_clk);
 	}
 }
 
@@ -574,8 +574,9 @@ static int sh_mobile_lcdc_display_notify(struct sh_mobile_lcdc_chan *ch,
 	switch (event) {
 	case SH_MOBILE_LCDC_EVENT_DISPLAY_CONNECT:
 		/* HDMI plug in */
+		console_lock();
 		if (lock_fb_info(info)) {
-			console_lock();
+
 
 			ch->display.width = monspec->max_x * 10;
 			ch->display.height = monspec->max_y * 10;
@@ -594,19 +595,20 @@ static int sh_mobile_lcdc_display_notify(struct sh_mobile_lcdc_chan *ch,
 				fb_set_suspend(info, 0);
 			}
 
-			console_unlock();
+
 			unlock_fb_info(info);
 		}
+		console_unlock();
 		break;
 
 	case SH_MOBILE_LCDC_EVENT_DISPLAY_DISCONNECT:
 		/* HDMI disconnect */
+		console_lock();
 		if (lock_fb_info(info)) {
-			console_lock();
 			fb_set_suspend(info, 1);
-			console_unlock();
 			unlock_fb_info(info);
 		}
+		console_unlock();
 		break;
 
 	case SH_MOBILE_LCDC_EVENT_DISPLAY_MODE:
diff --git a/drivers/video/simplefb.c b/drivers/video/simplefb.c
index 8d7810613058..210f3a02121a 100644
--- a/drivers/video/simplefb.c
+++ b/drivers/video/simplefb.c
@@ -66,8 +66,15 @@ static int simplefb_setcolreg(u_int regno, u_int red, u_int green, u_int blue,
 	return 0;
 }
 
+static void simplefb_destroy(struct fb_info *info)
+{
+	if (info->screen_base)
+		iounmap(info->screen_base);
+}
+
 static struct fb_ops simplefb_ops = {
 	.owner		= THIS_MODULE,
+	.fb_destroy	= simplefb_destroy,
 	.fb_setcolreg	= simplefb_setcolreg,
 	.fb_fillrect	= cfb_fillrect,
 	.fb_copyarea	= cfb_copyarea,
@@ -132,7 +139,7 @@ static int simplefb_parse_dt(struct platform_device *pdev,
 static int simplefb_parse_pd(struct platform_device *pdev,
 			     struct simplefb_params *params)
 {
-	struct simplefb_platform_data *pd = pdev->dev.platform_data;
+	struct simplefb_platform_data *pd = dev_get_platdata(&pdev->dev);
 	int i;
 
 	params->width = pd->width;
@@ -167,7 +174,7 @@ static int simplefb_probe(struct platform_device *pdev)
 		return -ENODEV;
 
 	ret = -ENODEV;
-	if (pdev->dev.platform_data)
+	if (dev_get_platdata(&pdev->dev))
 		ret = simplefb_parse_pd(pdev, &params);
 	else if (pdev->dev.of_node)
 		ret = simplefb_parse_dt(pdev, &params);
@@ -212,17 +219,26 @@ static int simplefb_probe(struct platform_device *pdev)
 
 	info->fbops = &simplefb_ops;
 	info->flags = FBINFO_DEFAULT | FBINFO_MISC_FIRMWARE;
-	info->screen_base = devm_ioremap(&pdev->dev, info->fix.smem_start,
-					 info->fix.smem_len);
+	info->screen_base = ioremap_wc(info->fix.smem_start,
+				       info->fix.smem_len);
 	if (!info->screen_base) {
 		framebuffer_release(info);
 		return -ENODEV;
 	}
 	info->pseudo_palette = (void *)(info + 1);
 
+	dev_info(&pdev->dev, "framebuffer at 0x%lx, 0x%x bytes, mapped to 0x%p\n",
+			     info->fix.smem_start, info->fix.smem_len,
+			     info->screen_base);
+	dev_info(&pdev->dev, "format=%s, mode=%dx%dx%d, linelength=%d\n",
+			     params.format->name,
+			     info->var.xres, info->var.yres,
+			     info->var.bits_per_pixel, info->fix.line_length);
+
 	ret = register_framebuffer(info);
 	if (ret < 0) {
 		dev_err(&pdev->dev, "Unable to register simplefb: %d\n", ret);
+		iounmap(info->screen_base);
 		framebuffer_release(info);
 		return ret;
 	}
diff --git a/drivers/video/sis/sis_main.c b/drivers/video/sis/sis_main.c
index 977e27927a21..22ad028bf123 100644
--- a/drivers/video/sis/sis_main.c
+++ b/drivers/video/sis/sis_main.c
@@ -5994,7 +5994,6 @@ static int sisfb_probe(struct pci_dev *pdev, const struct pci_device_id *ent)
 	if(!ivideo->sisvga_enabled) {
 		if(pci_enable_device(pdev)) {
 			if(ivideo->nbridge) pci_dev_put(ivideo->nbridge);
-			pci_set_drvdata(pdev, NULL);
 			framebuffer_release(sis_fb_info);
 			return -EIO;
 		}
@@ -6211,7 +6210,6 @@ error_3:	vfree(ivideo->bios_abase);
 			pci_dev_put(ivideo->lpcdev);
 		if(ivideo->nbridge)
 			pci_dev_put(ivideo->nbridge);
-		pci_set_drvdata(pdev, NULL);
 		if(!ivideo->sisvga_enabled)
 			pci_disable_device(pdev);
 		framebuffer_release(sis_fb_info);
@@ -6480,8 +6478,8 @@ error_3:	vfree(ivideo->bios_abase);
 									"disabled");
 
 
-		printk(KERN_INFO "fb%d: %s frame buffer device version %d.%d.%d\n",
-			sis_fb_info->node, ivideo->myid, VER_MAJOR, VER_MINOR, VER_LEVEL);
+		fb_info(sis_fb_info, "%s frame buffer device version %d.%d.%d\n",
+			ivideo->myid, VER_MAJOR, VER_MINOR, VER_LEVEL);
 
 		printk(KERN_INFO "sisfb: Copyright (C) 2001-2005 Thomas Winischhofer\n");
 
@@ -6523,8 +6521,6 @@ static void sisfb_remove(struct pci_dev *pdev)
 		mtrr_del(ivideo->mtrr, ivideo->video_base, ivideo->video_size);
 #endif
 
-	pci_set_drvdata(pdev, NULL);
-
 	/* If device was disabled when starting, disable
 	 * it when quitting.
 	 */
diff --git a/drivers/video/skeletonfb.c b/drivers/video/skeletonfb.c
index 2d4694c6b9e0..fefde7c6add7 100644
--- a/drivers/video/skeletonfb.c
+++ b/drivers/video/skeletonfb.c
@@ -824,8 +824,7 @@ static int xxxfb_probe(struct pci_dev *dev, const struct pci_device_id *ent)
 	fb_dealloc_cmap(&info->cmap);
 	return -EINVAL;
     }
-    printk(KERN_INFO "fb%d: %s frame buffer device\n", info->node,
-	   info->fix.id);
+    fb_info(info, "%s frame buffer device\n", info->fix.id);
     pci_set_drvdata(dev, info); /* or platform_set_drvdata(pdev, info) */
     return 0;
 }
diff --git a/drivers/video/smscufx.c b/drivers/video/smscufx.c
index e188ada2ffd1..d513ed6a49f2 100644
--- a/drivers/video/smscufx.c
+++ b/drivers/video/smscufx.c
@@ -1147,7 +1147,7 @@ static void ufx_free_framebuffer_work(struct work_struct *work)
 
 	fb_destroy_modelist(&info->modelist);
 
-	dev->info = 0;
+	dev->info = NULL;
 
 	/* Assume info structure is freed after this point */
 	framebuffer_release(info);
diff --git a/drivers/video/ssd1307fb.c b/drivers/video/ssd1307fb.c
index 44967c8fef2b..f4daa59f0a80 100644
--- a/drivers/video/ssd1307fb.c
+++ b/drivers/video/ssd1307fb.c
@@ -569,7 +569,7 @@ static struct i2c_driver ssd1307fb_driver = {
 	.id_table = ssd1307fb_i2c_id,
 	.driver = {
 		.name = "ssd1307fb",
-		.of_match_table = of_match_ptr(ssd1307fb_of_match),
+		.of_match_table = ssd1307fb_of_match,
 		.owner = THIS_MODULE,
 	},
 };
diff --git a/drivers/video/sstfb.c b/drivers/video/sstfb.c
index 9c00026e3ae2..f0cb279ef333 100644
--- a/drivers/video/sstfb.c
+++ b/drivers/video/sstfb.c
@@ -706,10 +706,10 @@ static void sstfb_setvgapass( struct fb_info *info, int enable )
 	fbiinit0 = sst_read (FBIINIT0);
 	if (par->vgapass) {
 		sst_write(FBIINIT0, fbiinit0 & ~DIS_VGA_PASSTHROUGH);
-		printk(KERN_INFO "fb%d: Enabling VGA pass-through\n", info->node );
+		fb_info(info, "Enabling VGA pass-through\n");
 	} else {
 		sst_write(FBIINIT0, fbiinit0 | DIS_VGA_PASSTHROUGH);
-		printk(KERN_INFO "fb%d: Disabling VGA pass-through\n", info->node );
+		fb_info(info, "Disabling VGA pass-through\n");
 	}
 	pci_write_config_dword(sst_dev, PCI_INIT_ENABLE, tmp);
 }
@@ -1437,8 +1437,8 @@ static int sstfb_probe(struct pci_dev *pdev, const struct pci_device_id *id)
 		printk(KERN_WARNING "sstfb: can't create sysfs entry.\n");
 
 
-	printk(KERN_INFO "fb%d: %s frame buffer device at 0x%p\n",
-	       info->node, fix->id, info->screen_base);
+	fb_info(info, "%s frame buffer device at 0x%p\n",
+		fix->id, info->screen_base);
 
 	return 0;
 
diff --git a/drivers/video/stifb.c b/drivers/video/stifb.c
index 019a1feef995..cfe8a2f905c5 100644
--- a/drivers/video/stifb.c
+++ b/drivers/video/stifb.c
@@ -1283,9 +1283,7 @@ static int __init stifb_init_fb(struct sti_struct *sti, int bpp_pref)
 
 	sti->info = info; /* save for unregister_framebuffer() */
 
-	printk(KERN_INFO 
-	    "fb%d: %s %dx%d-%d frame buffer device, %s, id: %04x, mmio: 0x%04lx\n",
-		fb->info.node, 
+	fb_info(&fb->info, "%s %dx%d-%d frame buffer device, %s, id: %04x, mmio: 0x%04lx\n",
 		fix->id,
 		var->xres, 
 		var->yres,
diff --git a/drivers/video/sunxvr1000.c b/drivers/video/sunxvr1000.c
index cc6f48bba36b..58241b47a96d 100644
--- a/drivers/video/sunxvr1000.c
+++ b/drivers/video/sunxvr1000.c
@@ -186,8 +186,6 @@ static int gfb_remove(struct platform_device *op)
 
         framebuffer_release(info);
 
-	dev_set_drvdata(&op->dev, NULL);
-
 	return 0;
 }
 
diff --git a/drivers/video/svgalib.c b/drivers/video/svgalib.c
index 33df9ec91795..9e01322fabe3 100644
--- a/drivers/video/svgalib.c
+++ b/drivers/video/svgalib.c
@@ -198,8 +198,8 @@ void svga_settile(struct fb_info *info, struct fb_tilemap *map)
 
 	if ((map->width != 8) || (map->height != 16) ||
 	    (map->depth != 1) || (map->length != 256)) {
-	    	printk(KERN_ERR "fb%d: unsupported font parameters: width %d, height %d, depth %d, length %d\n",
-			info->node, map->width, map->height, map->depth, map->length);
+		fb_err(info, "unsupported font parameters: width %d, height %d, depth %d, length %d\n",
+		       map->width, map->height, map->depth, map->length);
 		return;
 	}
 
diff --git a/drivers/video/sysimgblt.c b/drivers/video/sysimgblt.c
index 186c6f607be2..a4d05b1b17d7 100644
--- a/drivers/video/sysimgblt.c
+++ b/drivers/video/sysimgblt.c
@@ -152,7 +152,7 @@ static void slow_imageblit(const struct fb_image *image, struct fb_info *p,
 			}
 			shift += bpp;
 			shift &= (32 - 1);
-			if (!l) { l = 8; s++; };
+			if (!l) { l = 8; s++; }
 		}
 
 		/* write trailing bits */
diff --git a/drivers/video/tcx.c b/drivers/video/tcx.c
index c000852500aa..7fb2d696fac7 100644
--- a/drivers/video/tcx.c
+++ b/drivers/video/tcx.c
@@ -232,7 +232,7 @@ tcx_blank(int blank, struct fb_info *info)
 
 	case FB_BLANK_POWERDOWN: /* Poweroff */
 		break;
-	};
+	}
 
 	sbus_writel(val, &thc->thc_misc);
 
@@ -434,7 +434,7 @@ static int tcx_probe(struct platform_device *op)
 		default:
 			j = i;
 			break;
-		};
+		}
 		par->mmap_map[i].poff = op->resource[j].start;
 	}
 
@@ -498,8 +498,6 @@ static int tcx_remove(struct platform_device *op)
 
 	framebuffer_release(info);
 
-	dev_set_drvdata(&op->dev, NULL);
-
 	return 0;
 }
 
diff --git a/drivers/video/tdfxfb.c b/drivers/video/tdfxfb.c
index 64bc28ba4037..f761fe375f5b 100644
--- a/drivers/video/tdfxfb.c
+++ b/drivers/video/tdfxfb.c
@@ -1646,7 +1646,6 @@ static void tdfxfb_remove(struct pci_dev *pdev)
 			   pci_resource_len(pdev, 1));
 	release_mem_region(pci_resource_start(pdev, 0),
 			   pci_resource_len(pdev, 0));
-	pci_set_drvdata(pdev, NULL);
 	fb_dealloc_cmap(&info->cmap);
 	framebuffer_release(info);
 }
diff --git a/drivers/video/tgafb.c b/drivers/video/tgafb.c
index c9c8e5a1fdee..f28674fea909 100644
--- a/drivers/video/tgafb.c
+++ b/drivers/video/tgafb.c
@@ -1671,8 +1671,8 @@ static int tgafb_register(struct device *dev)
 	if (tga_bus_tc)
 		pr_info("tgafb: SFB+ detected, rev=0x%02x\n",
 			par->tga_chip_rev);
-	pr_info("fb%d: %s frame buffer device at 0x%lx\n",
-		info->node, info->fix.id, (long)bar0_start);
+	fb_info(info, "%s frame buffer device at 0x%lx\n",
+		info->fix.id, (long)bar0_start);
 
 	return 0;
 
diff --git a/drivers/video/tmiofb.c b/drivers/video/tmiofb.c
index deb8733f3c70..7fb4e321a431 100644
--- a/drivers/video/tmiofb.c
+++ b/drivers/video/tmiofb.c
@@ -250,7 +250,7 @@ static irqreturn_t tmiofb_irq(int irq, void *__info)
  */
 static int tmiofb_hw_stop(struct platform_device *dev)
 {
-	struct tmio_fb_data *data = dev->dev.platform_data;
+	struct tmio_fb_data *data = dev_get_platdata(&dev->dev);
 	struct fb_info *info = platform_get_drvdata(dev);
 	struct tmiofb_par *par = info->par;
 
@@ -311,7 +311,7 @@ static int tmiofb_hw_init(struct platform_device *dev)
  */
 static void tmiofb_hw_mode(struct platform_device *dev)
 {
-	struct tmio_fb_data *data = dev->dev.platform_data;
+	struct tmio_fb_data *data = dev_get_platdata(&dev->dev);
 	struct fb_info *info = platform_get_drvdata(dev);
 	struct fb_videomode *mode = info->mode;
 	struct tmiofb_par *par = info->par;
@@ -557,7 +557,7 @@ static int tmiofb_ioctl(struct fb_info *fbi,
 static struct fb_videomode *
 tmiofb_find_mode(struct fb_info *info, struct fb_var_screeninfo *var)
 {
-	struct tmio_fb_data *data = info->device->platform_data;
+	struct tmio_fb_data *data = dev_get_platdata(info->device);
 	struct fb_videomode *best = NULL;
 	int i;
 
@@ -577,7 +577,7 @@ static int tmiofb_check_var(struct fb_var_screeninfo *var, struct fb_info *info)
 {
 
 	struct fb_videomode *mode;
-	struct tmio_fb_data *data = info->device->platform_data;
+	struct tmio_fb_data *data = dev_get_platdata(info->device);
 
 	mode = tmiofb_find_mode(info, var);
 	if (!mode || var->bits_per_pixel > 16)
@@ -678,7 +678,7 @@ static struct fb_ops tmiofb_ops = {
 static int tmiofb_probe(struct platform_device *dev)
 {
 	const struct mfd_cell *cell = mfd_get_cell(dev);
-	struct tmio_fb_data *data = dev->dev.platform_data;
+	struct tmio_fb_data *data = dev_get_platdata(&dev->dev);
 	struct resource *ccr = platform_get_resource(dev, IORESOURCE_MEM, 1);
 	struct resource *lcr = platform_get_resource(dev, IORESOURCE_MEM, 0);
 	struct resource *vram = platform_get_resource(dev, IORESOURCE_MEM, 2);
@@ -781,8 +781,7 @@ static int tmiofb_probe(struct platform_device *dev)
 	if (retval < 0)
 		goto err_register_framebuffer;
 
-	printk(KERN_INFO "fb%d: %s frame buffer device\n",
-				info->node, info->fix.id);
+	fb_info(info, "%s frame buffer device\n", info->fix.id);
 
 	return 0;
 
diff --git a/drivers/video/tridentfb.c b/drivers/video/tridentfb.c
index ab57d387d6b5..7ed9a227f5ea 100644
--- a/drivers/video/tridentfb.c
+++ b/drivers/video/tridentfb.c
@@ -1553,7 +1553,6 @@ static void trident_pci_remove(struct pci_dev *dev)
 	iounmap(info->screen_base);
 	release_mem_region(tridentfb_fix.smem_start, tridentfb_fix.smem_len);
 	release_mem_region(tridentfb_fix.mmio_start, tridentfb_fix.mmio_len);
-	pci_set_drvdata(dev, NULL);
 	kfree(info->pixmap.addr);
 	fb_dealloc_cmap(&info->cmap);
 	framebuffer_release(info);
diff --git a/drivers/video/udlfb.c b/drivers/video/udlfb.c
index d2e5bc3cf969..025f14e30eed 100644
--- a/drivers/video/udlfb.c
+++ b/drivers/video/udlfb.c
@@ -1166,7 +1166,7 @@ static int dlfb_realloc_framebuffer(struct dlfb_data *dev, struct fb_info *info)
 	int new_len;
 	unsigned char *old_fb = info->screen_base;
 	unsigned char *new_fb;
-	unsigned char *new_back = 0;
+	unsigned char *new_back = NULL;
 
 	pr_warn("Reallocating framebuffer. Addresses will change!\n");
 
diff --git a/drivers/video/uvesafb.c b/drivers/video/uvesafb.c
index 7aec6f39fdd5..256fba7f4641 100644
--- a/drivers/video/uvesafb.c
+++ b/drivers/video/uvesafb.c
@@ -233,8 +233,7 @@ out:
 static void uvesafb_free(struct uvesafb_ktask *task)
 {
 	if (task) {
-		if (task->done)
-			kfree(task->done);
+		kfree(task->done);
 		kfree(task);
 	}
 }
@@ -1332,8 +1331,8 @@ setmode:
 				FB_VISUAL_PSEUDOCOLOR : FB_VISUAL_TRUECOLOR;
 	info->fix.line_length = mode->bytes_per_scan_line;
 
-out:	if (crtc != NULL)
-		kfree(crtc);
+out:
+	kfree(crtc);
 	uvesafb_free(task);
 
 	return err;
@@ -1771,13 +1770,11 @@ static int uvesafb_probe(struct platform_device *dev)
 			"using %dk, total %dk\n", info->fix.smem_start,
 			info->screen_base, info->fix.smem_len/1024,
 			par->vbe_ib.total_memory * 64);
-	printk(KERN_INFO "fb%d: %s frame buffer device\n", info->node,
-			info->fix.id);
+	fb_info(info, "%s frame buffer device\n", info->fix.id);
 
 	err = sysfs_create_group(&dev->dev.kobj, &uvesafb_dev_attgrp);
 	if (err != 0)
-		printk(KERN_WARNING "fb%d: failed to register attributes\n",
-			info->node);
+		fb_warn(info, "failed to register attributes\n");
 
 	return 0;
 
@@ -1793,8 +1790,7 @@ out_mode:
 	fb_destroy_modedb(info->monspecs.modedb);
 	fb_dealloc_cmap(&info->cmap);
 out:
-	if (par->vbe_modes)
-		kfree(par->vbe_modes);
+	kfree(par->vbe_modes);
 
 	framebuffer_release(info);
 	return err;
@@ -1817,12 +1813,9 @@ static int uvesafb_remove(struct platform_device *dev)
 		fb_dealloc_cmap(&info->cmap);
 
 		if (par) {
-			if (par->vbe_modes)
-				kfree(par->vbe_modes);
-			if (par->vbe_state_orig)
-				kfree(par->vbe_state_orig);
-			if (par->vbe_state_saved)
-				kfree(par->vbe_state_saved);
+			kfree(par->vbe_modes);
+			kfree(par->vbe_state_orig);
+			kfree(par->vbe_state_saved);
 		}
 
 		framebuffer_release(info);
diff --git a/drivers/video/valkyriefb.c b/drivers/video/valkyriefb.c
index 3f5a041601da..e287ebc47817 100644
--- a/drivers/video/valkyriefb.c
+++ b/drivers/video/valkyriefb.c
@@ -392,7 +392,7 @@ int __init valkyriefb_init(void)
 	if ((err = register_framebuffer(&p->info)) != 0)
 		goto out_cmap_free;
 
-	printk(KERN_INFO "fb%d: valkyrie frame buffer device\n", p->info.node);
+	fb_info(&p->info, "valkyrie frame buffer device\n");
 	return 0;
 
  out_cmap_free:
diff --git a/drivers/video/vesafb.c b/drivers/video/vesafb.c
index bd83233ec227..1c7da3b098d6 100644
--- a/drivers/video/vesafb.c
+++ b/drivers/video/vesafb.c
@@ -489,8 +489,7 @@ static int vesafb_probe(struct platform_device *dev)
 		fb_dealloc_cmap(&info->cmap);
 		goto err;
 	}
-	printk(KERN_INFO "fb%d: %s frame buffer device\n",
-	       info->node, info->fix.id);
+	fb_info(info, "%s frame buffer device\n", info->fix.id);
 	return 0;
 err:
 	if (info->screen_base)
diff --git a/drivers/video/vfb.c b/drivers/video/vfb.c
index ee5985efa15c..70a897b1e458 100644
--- a/drivers/video/vfb.c
+++ b/drivers/video/vfb.c
@@ -390,9 +390,8 @@ static int vfb_pan_display(struct fb_var_screeninfo *var,
 			   struct fb_info *info)
 {
 	if (var->vmode & FB_VMODE_YWRAP) {
-		if (var->yoffset < 0
-		    || var->yoffset >= info->var.yres_virtual
-		    || var->xoffset)
+		if (var->yoffset >= info->var.yres_virtual ||
+		    var->xoffset)
 			return -EINVAL;
 	} else {
 		if (var->xoffset + info->var.xres > info->var.xres_virtual ||
@@ -527,9 +526,8 @@ static int vfb_probe(struct platform_device *dev)
 		goto err2;
 	platform_set_drvdata(dev, info);
 
-	printk(KERN_INFO
-	       "fb%d: Virtual frame buffer device, using %ldK of video memory\n",
-	       info->node, videomemorysize >> 10);
+	fb_info(info, "Virtual frame buffer device, using %ldK of video memory\n",
+		videomemorysize >> 10);
 	return 0;
 err2:
 	fb_dealloc_cmap(&info->cmap);
diff --git a/drivers/video/vga16fb.c b/drivers/video/vga16fb.c
index 2827333703d9..283d335a759f 100644
--- a/drivers/video/vga16fb.c
+++ b/drivers/video/vga16fb.c
@@ -1377,8 +1377,7 @@ static int vga16fb_probe(struct platform_device *dev)
 		goto err_check_var;
 	}
 
-	printk(KERN_INFO "fb%d: %s frame buffer device\n",
-	       info->node, info->fix.id);
+	fb_info(info, "%s frame buffer device\n", info->fix.id);
 	platform_set_drvdata(dev, info);
 
 	return 0;
diff --git a/drivers/video/vt8500lcdfb.c b/drivers/video/vt8500lcdfb.c
index 897484903c30..b30e5a439d1f 100644
--- a/drivers/video/vt8500lcdfb.c
+++ b/drivers/video/vt8500lcdfb.c
@@ -365,7 +365,7 @@ static int vt8500lcd_probe(struct platform_device *pdev)
 	if (!fb_mem_virt) {
 		pr_err("%s: Failed to allocate framebuffer\n", __func__);
 		return -ENOMEM;
-	};
+	}
 
 	fbi->fb.fix.smem_start	= fb_mem_phys;
 	fbi->fb.fix.smem_len	= fb_mem_len;
diff --git a/drivers/video/vt8623fb.c b/drivers/video/vt8623fb.c
index e9557fa014ee..8bc6e0958a09 100644
--- a/drivers/video/vt8623fb.c
+++ b/drivers/video/vt8623fb.c
@@ -266,7 +266,7 @@ static void vt8623_set_pixclock(struct fb_info *info, u32 pixclock)
 
 	rv = svga_compute_pll(&vt8623_pll, 1000000000 / pixclock, &m, &n, &r, info->node);
 	if (rv < 0) {
-		printk(KERN_ERR "fb%d: cannot set requested pixclock, keeping old value\n", info->node);
+		fb_err(info, "cannot set requested pixclock, keeping old value\n");
 		return;
 	}
 
@@ -335,7 +335,7 @@ static int vt8623fb_check_var(struct fb_var_screeninfo *var, struct fb_info *inf
 	rv = svga_match_format (vt8623fb_formats, var, NULL);
 	if (rv < 0)
 	{
-		printk(KERN_ERR "fb%d: unsupported mode requested\n", info->node);
+		fb_err(info, "unsupported mode requested\n");
 		return rv;
 	}
 
@@ -354,21 +354,23 @@ static int vt8623fb_check_var(struct fb_var_screeninfo *var, struct fb_info *inf
 	mem = ((var->bits_per_pixel * var->xres_virtual) >> 3) * var->yres_virtual;
 	if (mem > info->screen_size)
 	{
-		printk(KERN_ERR "fb%d: not enough framebuffer memory (%d kB requested , %d kB available)\n", info->node, mem >> 10, (unsigned int) (info->screen_size >> 10));
+		fb_err(info, "not enough framebuffer memory (%d kB requested, %d kB available)\n",
+		       mem >> 10, (unsigned int) (info->screen_size >> 10));
 		return -EINVAL;
 	}
 
 	/* Text mode is limited to 256 kB of memory */
 	if ((var->bits_per_pixel == 0) && (mem > (256*1024)))
 	{
-		printk(KERN_ERR "fb%d: text framebuffer size too large (%d kB requested, 256 kB possible)\n", info->node, mem >> 10);
+		fb_err(info, "text framebuffer size too large (%d kB requested, 256 kB possible)\n",
+		       mem >> 10);
 		return -EINVAL;
 	}
 
 	rv = svga_check_timings (&vt8623_timing_regs, var, info->node);
 	if (rv < 0)
 	{
-		printk(KERN_ERR "fb%d: invalid timings requested\n", info->node);
+		fb_err(info, "invalid timings requested\n");
 		return rv;
 	}
 
@@ -474,32 +476,32 @@ static int vt8623fb_set_par(struct fb_info *info)
 	mode = svga_match_format(vt8623fb_formats, &(info->var), &(info->fix));
 	switch (mode) {
 	case 0:
-		pr_debug("fb%d: text mode\n", info->node);
+		fb_dbg(info, "text mode\n");
 		svga_set_textmode_vga_regs(par->state.vgabase);
 		svga_wseq_mask(par->state.vgabase, 0x15, 0x00, 0xFE);
 		svga_wcrt_mask(par->state.vgabase, 0x11, 0x60, 0x70);
 		break;
 	case 1:
-		pr_debug("fb%d: 4 bit pseudocolor\n", info->node);
+		fb_dbg(info, "4 bit pseudocolor\n");
 		vga_wgfx(par->state.vgabase, VGA_GFX_MODE, 0x40);
 		svga_wseq_mask(par->state.vgabase, 0x15, 0x20, 0xFE);
 		svga_wcrt_mask(par->state.vgabase, 0x11, 0x00, 0x70);
 		break;
 	case 2:
-		pr_debug("fb%d: 4 bit pseudocolor, planar\n", info->node);
+		fb_dbg(info, "4 bit pseudocolor, planar\n");
 		svga_wseq_mask(par->state.vgabase, 0x15, 0x00, 0xFE);
 		svga_wcrt_mask(par->state.vgabase, 0x11, 0x00, 0x70);
 		break;
 	case 3:
-		pr_debug("fb%d: 8 bit pseudocolor\n", info->node);
+		fb_dbg(info, "8 bit pseudocolor\n");
 		svga_wseq_mask(par->state.vgabase, 0x15, 0x22, 0xFE);
 		break;
 	case 4:
-		pr_debug("fb%d: 5/6/5 truecolor\n", info->node);
+		fb_dbg(info, "5/6/5 truecolor\n");
 		svga_wseq_mask(par->state.vgabase, 0x15, 0xB6, 0xFE);
 		break;
 	case 5:
-		pr_debug("fb%d: 8/8/8 truecolor\n", info->node);
+		fb_dbg(info, "8/8/8 truecolor\n");
 		svga_wseq_mask(par->state.vgabase, 0x15, 0xAE, 0xFE);
 		break;
 	default:
@@ -584,27 +586,27 @@ static int vt8623fb_blank(int blank_mode, struct fb_info *info)
 
 	switch (blank_mode) {
 	case FB_BLANK_UNBLANK:
-		pr_debug("fb%d: unblank\n", info->node);
+		fb_dbg(info, "unblank\n");
 		svga_wcrt_mask(par->state.vgabase, 0x36, 0x00, 0x30);
 		svga_wseq_mask(par->state.vgabase, 0x01, 0x00, 0x20);
 		break;
 	case FB_BLANK_NORMAL:
-		pr_debug("fb%d: blank\n", info->node);
+		fb_dbg(info, "blank\n");
 		svga_wcrt_mask(par->state.vgabase, 0x36, 0x00, 0x30);
 		svga_wseq_mask(par->state.vgabase, 0x01, 0x20, 0x20);
 		break;
 	case FB_BLANK_HSYNC_SUSPEND:
-		pr_debug("fb%d: DPMS standby (hsync off)\n", info->node);
+		fb_dbg(info, "DPMS standby (hsync off)\n");
 		svga_wcrt_mask(par->state.vgabase, 0x36, 0x10, 0x30);
 		svga_wseq_mask(par->state.vgabase, 0x01, 0x20, 0x20);
 		break;
 	case FB_BLANK_VSYNC_SUSPEND:
-		pr_debug("fb%d: DPMS suspend (vsync off)\n", info->node);
+		fb_dbg(info, "DPMS suspend (vsync off)\n");
 		svga_wcrt_mask(par->state.vgabase, 0x36, 0x20, 0x30);
 		svga_wseq_mask(par->state.vgabase, 0x01, 0x20, 0x20);
 		break;
 	case FB_BLANK_POWERDOWN:
-		pr_debug("fb%d: DPMS off (no sync)\n", info->node);
+		fb_dbg(info, "DPMS off (no sync)\n");
 		svga_wcrt_mask(par->state.vgabase, 0x36, 0x30, 0x30);
 		svga_wseq_mask(par->state.vgabase, 0x01, 0x20, 0x20);
 		break;
@@ -769,12 +771,12 @@ static int vt8623_pci_probe(struct pci_dev *dev, const struct pci_device_id *id)
 
 	rc = register_framebuffer(info);
 	if (rc < 0) {
-		dev_err(info->device, "cannot register framebugger\n");
+		dev_err(info->device, "cannot register framebuffer\n");
 		goto err_reg_fb;
 	}
 
-	printk(KERN_INFO "fb%d: %s on %s, %d MB RAM\n", info->node, info->fix.id,
-		 pci_name(dev), info->fix.smem_len >> 20);
+	fb_info(info, "%s on %s, %d MB RAM\n",
+		info->fix.id, pci_name(dev), info->fix.smem_len >> 20);
 
 	/* Record a reference to the driver data */
 	pci_set_drvdata(dev, info);
@@ -829,7 +831,6 @@ static void vt8623_pci_remove(struct pci_dev *dev)
 		pci_release_regions(dev);
 /*		pci_disable_device(dev); */
 
-		pci_set_drvdata(dev, NULL);
 		framebuffer_release(info);
 	}
 }
diff --git a/drivers/video/w100fb.c b/drivers/video/w100fb.c
index 7a299e951f75..10951c82f6ed 100644
--- a/drivers/video/w100fb.c
+++ b/drivers/video/w100fb.c
@@ -680,7 +680,7 @@ int w100fb_probe(struct platform_device *pdev)
 	par = info->par;
 	platform_set_drvdata(pdev, info);
 
-	inf = pdev->dev.platform_data;
+	inf = dev_get_platdata(&pdev->dev);
 	par->chip_id = chip_id;
 	par->mach = inf;
 	par->fastpll_mode = 0;
@@ -761,10 +761,9 @@ int w100fb_probe(struct platform_device *pdev)
 	err |= device_create_file(&pdev->dev, &dev_attr_flip);
 
 	if (err != 0)
-		printk(KERN_WARNING "fb%d: failed to register attributes (%d)\n",
-				info->node, err);
+		fb_warn(info, "failed to register attributes (%d)\n", err);
 
-	printk(KERN_INFO "fb%d: %s frame buffer device\n", info->node, info->fix.id);
+	fb_info(info, "%s frame buffer device\n", info->fix.id);
 	return 0;
 out:
 	if (info) {
diff --git a/drivers/video/wm8505fb.c b/drivers/video/wm8505fb.c
index 3072f30cad19..537d199612af 100644
--- a/drivers/video/wm8505fb.c
+++ b/drivers/video/wm8505fb.c
@@ -372,14 +372,12 @@ static int wm8505fb_probe(struct platform_device *pdev)
 	}
 
 	ret = device_create_file(&pdev->dev, &dev_attr_contrast);
-	if (ret < 0) {
-		printk(KERN_WARNING "fb%d: failed to register attributes (%d)\n",
-			fbi->fb.node, ret);
-	}
+	if (ret < 0)
+		fb_warn(&fbi->fb, "failed to register attributes (%d)\n", ret);
 
-	printk(KERN_INFO "fb%d: %s frame buffer at 0x%lx-0x%lx\n",
-	       fbi->fb.node, fbi->fb.fix.id, fbi->fb.fix.smem_start,
-	       fbi->fb.fix.smem_start + fbi->fb.fix.smem_len - 1);
+	fb_info(&fbi->fb, "%s frame buffer at 0x%lx-0x%lx\n",
+		fbi->fb.fix.id, fbi->fb.fix.smem_start,
+		fbi->fb.fix.smem_start + fbi->fb.fix.smem_len - 1);
 
 	return 0;
 }
@@ -411,7 +409,7 @@ static struct platform_driver wm8505fb_driver = {
 	.driver		= {
 		.owner	= THIS_MODULE,
 		.name	= DRIVER_NAME,
-		.of_match_table = of_match_ptr(wmt_dt_ids),
+		.of_match_table = wmt_dt_ids,
 	},
 };
 
diff --git a/drivers/video/wmt_ge_rops.c b/drivers/video/wmt_ge_rops.c
index 4aaeb18223bc..b0a9f34b2e01 100644
--- a/drivers/video/wmt_ge_rops.c
+++ b/drivers/video/wmt_ge_rops.c
@@ -169,13 +169,13 @@ static struct platform_driver wmt_ge_rops_driver = {
 	.driver		= {
 		.owner	= THIS_MODULE,
 		.name	= "wmt_ge_rops",
-		.of_match_table = of_match_ptr(wmt_dt_ids),
+		.of_match_table = wmt_dt_ids,
 	},
 };
 
 module_platform_driver(wmt_ge_rops_driver);
 
-MODULE_AUTHOR("Alexey Charkov <alchark@gmail.com");
+MODULE_AUTHOR("Alexey Charkov <alchark@gmail.com>");
 MODULE_DESCRIPTION("Accelerators for raster operations using "
 		   "WonderMedia Graphics Engine");
 MODULE_LICENSE("GPL v2");
diff --git a/drivers/video/xilinxfb.c b/drivers/video/xilinxfb.c
index 84c664ea8eb9..6ff1a91e9dfd 100644
--- a/drivers/video/xilinxfb.c
+++ b/drivers/video/xilinxfb.c
@@ -260,10 +260,9 @@ static int xilinxfb_assign(struct platform_device *pdev,
 
 		res = platform_get_resource(pdev, IORESOURCE_MEM, 0);
 		drvdata->regs = devm_ioremap_resource(&pdev->dev, res);
-		if (IS_ERR(drvdata->regs)) {
-			rc = PTR_ERR(drvdata->regs);
-			goto err_region;
-		}
+		if (IS_ERR(drvdata->regs))
+			return PTR_ERR(drvdata->regs);
+
 		drvdata->regs_phys = res->start;
 	}
 
@@ -279,11 +278,7 @@ static int xilinxfb_assign(struct platform_device *pdev,
 
 	if (!drvdata->fb_virt) {
 		dev_err(dev, "Could not allocate frame buffer memory\n");
-		rc = -ENOMEM;
-		if (drvdata->flags & BUS_ACCESS_FLAG)
-			goto err_fbmem;
-		else
-			goto err_region;
+		return -ENOMEM;
 	}
 
 	/* Clear (turn to black) the framebuffer */
@@ -363,14 +358,6 @@ err_cmap:
 	/* Turn off the display */
 	xilinx_fb_out32(drvdata, REG_CTRL, 0);
 
-err_fbmem:
-	if (drvdata->flags & BUS_ACCESS_FLAG)
-		devm_iounmap(dev, drvdata->regs);
-
-err_region:
-	kfree(drvdata);
-	dev_set_drvdata(dev, NULL);
-
 	return rc;
 }
 
@@ -395,17 +382,12 @@ static int xilinxfb_release(struct device *dev)
 	/* Turn off the display */
 	xilinx_fb_out32(drvdata, REG_CTRL, 0);
 
-	/* Release the resources, as allocated based on interface */
-	if (drvdata->flags & BUS_ACCESS_FLAG)
-		devm_iounmap(dev, drvdata->regs);
 #ifdef CONFIG_PPC_DCR
-	else
+	/* Release the resources, as allocated based on interface */
+	if (!(drvdata->flags & BUS_ACCESS_FLAG))
 		dcr_unmap(drvdata->dcr_host, drvdata->dcr_len);
 #endif
 
-	kfree(drvdata);
-	dev_set_drvdata(dev, NULL);
-
 	return 0;
 }
 
@@ -413,7 +395,7 @@ static int xilinxfb_release(struct device *dev)
  * OF bus binding
  */
 
-static int xilinxfb_of_probe(struct platform_device *op)
+static int xilinxfb_of_probe(struct platform_device *pdev)
 {
 	const u32 *prop;
 	u32 tft_access = 0;
@@ -425,17 +407,15 @@ static int xilinxfb_of_probe(struct platform_device *op)
 	pdata = xilinx_fb_default_pdata;
 
 	/* Allocate the driver data region */
-	drvdata = kzalloc(sizeof(*drvdata), GFP_KERNEL);
-	if (!drvdata) {
-		dev_err(&op->dev, "Couldn't allocate device private record\n");
+	drvdata = devm_kzalloc(&pdev->dev, sizeof(*drvdata), GFP_KERNEL);
+	if (!drvdata)
 		return -ENOMEM;
-	}
 
 	/*
 	 * To check whether the core is connected directly to DCR or BUS
 	 * interface and initialize the tft_access accordingly.
 	 */
-	of_property_read_u32(op->dev.of_node, "xlnx,dcr-splb-slave-if",
+	of_property_read_u32(pdev->dev.of_node, "xlnx,dcr-splb-slave-if",
 			     &tft_access);
 
 	/*
@@ -448,40 +428,39 @@ static int xilinxfb_of_probe(struct platform_device *op)
 #ifdef CONFIG_PPC_DCR
 	else {
 		int start;
-		start = dcr_resource_start(op->dev.of_node, 0);
-		drvdata->dcr_len = dcr_resource_len(op->dev.of_node, 0);
-		drvdata->dcr_host = dcr_map(op->dev.of_node, start, drvdata->dcr_len);
+		start = dcr_resource_start(pdev->dev.of_node, 0);
+		drvdata->dcr_len = dcr_resource_len(pdev->dev.of_node, 0);
+		drvdata->dcr_host = dcr_map(pdev->dev.of_node, start, drvdata->dcr_len);
 		if (!DCR_MAP_OK(drvdata->dcr_host)) {
-			dev_err(&op->dev, "invalid DCR address\n");
-			kfree(drvdata);
+			dev_err(&pdev->dev, "invalid DCR address\n");
 			return -ENODEV;
 		}
 	}
 #endif
 
-	prop = of_get_property(op->dev.of_node, "phys-size", &size);
+	prop = of_get_property(pdev->dev.of_node, "phys-size", &size);
 	if ((prop) && (size >= sizeof(u32)*2)) {
 		pdata.screen_width_mm = prop[0];
 		pdata.screen_height_mm = prop[1];
 	}
 
-	prop = of_get_property(op->dev.of_node, "resolution", &size);
+	prop = of_get_property(pdev->dev.of_node, "resolution", &size);
 	if ((prop) && (size >= sizeof(u32)*2)) {
 		pdata.xres = prop[0];
 		pdata.yres = prop[1];
 	}
 
-	prop = of_get_property(op->dev.of_node, "virtual-resolution", &size);
+	prop = of_get_property(pdev->dev.of_node, "virtual-resolution", &size);
 	if ((prop) && (size >= sizeof(u32)*2)) {
 		pdata.xvirt = prop[0];
 		pdata.yvirt = prop[1];
 	}
 
-	if (of_find_property(op->dev.of_node, "rotate-display", NULL))
+	if (of_find_property(pdev->dev.of_node, "rotate-display", NULL))
 		pdata.rotate_screen = 1;
 
-	dev_set_drvdata(&op->dev, drvdata);
-	return xilinxfb_assign(op, drvdata, &pdata);
+	dev_set_drvdata(&pdev->dev, drvdata);
+	return xilinxfb_assign(pdev, drvdata, &pdata);
 }
 
 static int xilinxfb_of_remove(struct platform_device *op)
diff --git a/drivers/virtio/virtio_balloon.c b/drivers/virtio/virtio_balloon.c
index 1f572c00a1be..c444654fc33f 100644
--- a/drivers/virtio/virtio_balloon.c
+++ b/drivers/virtio/virtio_balloon.c
@@ -275,9 +275,8 @@ static inline s64 towards_target(struct virtio_balloon *vb)
 	__le32 v;
 	s64 target;
 
-	vb->vdev->config->get(vb->vdev,
-			      offsetof(struct virtio_balloon_config, num_pages),
-			      &v, sizeof(v));
+	virtio_cread(vb->vdev, struct virtio_balloon_config, num_pages, &v);
+
 	target = le32_to_cpu(v);
 	return target - vb->num_pages;
 }
@@ -286,9 +285,8 @@ static void update_balloon_size(struct virtio_balloon *vb)
 {
 	__le32 actual = cpu_to_le32(vb->num_pages);
 
-	vb->vdev->config->set(vb->vdev,
-			      offsetof(struct virtio_balloon_config, actual),
-			      &actual, sizeof(actual));
+	virtio_cwrite(vb->vdev, struct virtio_balloon_config, num_pages,
+		      &actual);
 }
 
 static int balloon(void *_vballoon)
@@ -513,7 +511,7 @@ static void virtballoon_remove(struct virtio_device *vdev)
 	kfree(vb);
 }
 
-#ifdef CONFIG_PM
+#ifdef CONFIG_PM_SLEEP
 static int virtballoon_freeze(struct virtio_device *vdev)
 {
 	struct virtio_balloon *vb = vdev->priv;
@@ -556,7 +554,7 @@ static struct virtio_driver virtio_balloon_driver = {
 	.probe =	virtballoon_probe,
 	.remove =	virtballoon_remove,
 	.config_changed = virtballoon_changed,
-#ifdef CONFIG_PM
+#ifdef CONFIG_PM_SLEEP
 	.freeze	=	virtballoon_freeze,
 	.restore =	virtballoon_restore,
 #endif
diff --git a/drivers/virtio/virtio_mmio.c b/drivers/virtio/virtio_mmio.c
index 1ba0d6831015..c600ccfd6922 100644
--- a/drivers/virtio/virtio_mmio.c
+++ b/drivers/virtio/virtio_mmio.c
@@ -219,13 +219,14 @@ static void vm_reset(struct virtio_device *vdev)
 /* Transport interface */
 
 /* the notify function used when creating a virt queue */
-static void vm_notify(struct virtqueue *vq)
+static bool vm_notify(struct virtqueue *vq)
 {
 	struct virtio_mmio_device *vm_dev = to_virtio_mmio_device(vq->vdev);
 
 	/* We write the queue's selector into the notification register to
 	 * signal the other end */
 	writel(vq->index, vm_dev->base + VIRTIO_MMIO_QUEUE_NOTIFY);
+	return true;
 }
 
 /* Notify all virtqueues on an interrupt. */
@@ -470,7 +471,7 @@ static int virtio_mmio_probe(struct platform_device *pdev)
 
 	/* Check magic value */
 	magic = readl(vm_dev->base + VIRTIO_MMIO_MAGIC_VALUE);
-	if (memcmp(&magic, "virt", 4) != 0) {
+	if (magic != ('v' | 'i' << 8 | 'r' << 16 | 't' << 24)) {
 		dev_warn(&pdev->dev, "Wrong magic value 0x%08lx!\n", magic);
 		return -ENODEV;
 	}
diff --git a/drivers/virtio/virtio_pci.c b/drivers/virtio/virtio_pci.c
index 98917fc872a4..a37c69941d30 100644
--- a/drivers/virtio/virtio_pci.c
+++ b/drivers/virtio/virtio_pci.c
@@ -197,13 +197,14 @@ static void vp_reset(struct virtio_device *vdev)
 }
 
 /* the notify function used when creating a virt queue */
-static void vp_notify(struct virtqueue *vq)
+static bool vp_notify(struct virtqueue *vq)
 {
 	struct virtio_pci_device *vp_dev = to_vp_device(vq->vdev);
 
 	/* we write the queue's selector into the notification register to
 	 * signal the other end */
 	iowrite16(vq->index, vp_dev->ioaddr + VIRTIO_PCI_QUEUE_NOTIFY);
+	return true;
 }
 
 /* Handle a configuration change: Tell driver if it wants to know. */
diff --git a/drivers/virtio/virtio_ring.c b/drivers/virtio/virtio_ring.c
index 6b4a4db4404d..28b5338fff71 100644
--- a/drivers/virtio/virtio_ring.c
+++ b/drivers/virtio/virtio_ring.c
@@ -81,7 +81,7 @@ struct vring_virtqueue
 	u16 last_used_idx;
 
 	/* How to notify other side. FIXME: commonalize hcalls! */
-	void (*notify)(struct virtqueue *vq);
+	bool (*notify)(struct virtqueue *vq);
 
 #ifdef DEBUG
 	/* They're supposed to lock for us. */
@@ -173,6 +173,8 @@ static inline int vring_add_indirect(struct vring_virtqueue *vq,
 	head = vq->free_head;
 	vq->vring.desc[head].flags = VRING_DESC_F_INDIRECT;
 	vq->vring.desc[head].addr = virt_to_phys(desc);
+	/* kmemleak gives a false positive, as it's hidden by virt_to_phys */
+	kmemleak_ignore(desc);
 	vq->vring.desc[head].len = i * sizeof(struct vring_desc);
 
 	/* Update free pointer */
@@ -428,13 +430,22 @@ EXPORT_SYMBOL_GPL(virtqueue_kick_prepare);
  * @vq: the struct virtqueue
  *
  * This does not need to be serialized.
+ *
+ * Returns false if host notify failed or queue is broken, otherwise true.
  */
-void virtqueue_notify(struct virtqueue *_vq)
+bool virtqueue_notify(struct virtqueue *_vq)
 {
 	struct vring_virtqueue *vq = to_vvq(_vq);
 
+	if (unlikely(vq->broken))
+		return false;
+
 	/* Prod other side to tell it about changes. */
-	vq->notify(_vq);
+	if (!vq->notify(_vq)) {
+		vq->broken = true;
+		return false;
+	}
+	return true;
 }
 EXPORT_SYMBOL_GPL(virtqueue_notify);
 
@@ -447,11 +458,14 @@ EXPORT_SYMBOL_GPL(virtqueue_notify);
  *
  * Caller must ensure we don't call this with other virtqueue
  * operations at the same time (except where noted).
+ *
+ * Returns false if kick failed, otherwise true.
  */
-void virtqueue_kick(struct virtqueue *vq)
+bool virtqueue_kick(struct virtqueue *vq)
 {
 	if (virtqueue_kick_prepare(vq))
-		virtqueue_notify(vq);
+		return virtqueue_notify(vq);
+	return true;
 }
 EXPORT_SYMBOL_GPL(virtqueue_kick);
 
@@ -742,7 +756,7 @@ struct virtqueue *vring_new_virtqueue(unsigned int index,
 				      struct virtio_device *vdev,
 				      bool weak_barriers,
 				      void *pages,
-				      void (*notify)(struct virtqueue *),
+				      bool (*notify)(struct virtqueue *),
 				      void (*callback)(struct virtqueue *),
 				      const char *name)
 {
@@ -837,4 +851,12 @@ unsigned int virtqueue_get_vring_size(struct virtqueue *_vq)
 }
 EXPORT_SYMBOL_GPL(virtqueue_get_vring_size);
 
+bool virtqueue_is_broken(struct virtqueue *_vq)
+{
+	struct vring_virtqueue *vq = to_vvq(_vq);
+
+	return vq->broken;
+}
+EXPORT_SYMBOL_GPL(virtqueue_is_broken);
+
 MODULE_LICENSE("GPL");
diff --git a/drivers/w1/masters/w1-gpio.c b/drivers/w1/masters/w1-gpio.c
index 264ad1c583ab..e36b18b2817b 100644
--- a/drivers/w1/masters/w1-gpio.c
+++ b/drivers/w1/masters/w1-gpio.c
@@ -56,7 +56,7 @@ MODULE_DEVICE_TABLE(of, w1_gpio_dt_ids);
 
 static int w1_gpio_probe_dt(struct platform_device *pdev)
 {
-	struct w1_gpio_platform_data *pdata = pdev->dev.platform_data;
+	struct w1_gpio_platform_data *pdata = dev_get_platdata(&pdev->dev);
 	struct device_node *np = pdev->dev.of_node;
 	int gpio;
 
@@ -92,7 +92,7 @@ static int w1_gpio_probe(struct platform_device *pdev)
 		}
 	}
 
-	pdata = pdev->dev.platform_data;
+	pdata = dev_get_platdata(&pdev->dev);
 
 	if (!pdata) {
 		dev_err(&pdev->dev, "No configuration data\n");
@@ -154,7 +154,7 @@ static int w1_gpio_probe(struct platform_device *pdev)
 static int w1_gpio_remove(struct platform_device *pdev)
 {
 	struct w1_bus_master *master = platform_get_drvdata(pdev);
-	struct w1_gpio_platform_data *pdata = pdev->dev.platform_data;
+	struct w1_gpio_platform_data *pdata = dev_get_platdata(&pdev->dev);
 
 	if (pdata->enable_external_pullup)
 		pdata->enable_external_pullup(0);
@@ -171,7 +171,7 @@ static int w1_gpio_remove(struct platform_device *pdev)
 
 static int w1_gpio_suspend(struct platform_device *pdev, pm_message_t state)
 {
-	struct w1_gpio_platform_data *pdata = pdev->dev.platform_data;
+	struct w1_gpio_platform_data *pdata = dev_get_platdata(&pdev->dev);
 
 	if (pdata->enable_external_pullup)
 		pdata->enable_external_pullup(0);
@@ -181,7 +181,7 @@ static int w1_gpio_suspend(struct platform_device *pdev, pm_message_t state)
 
 static int w1_gpio_resume(struct platform_device *pdev)
 {
-	struct w1_gpio_platform_data *pdata = pdev->dev.platform_data;
+	struct w1_gpio_platform_data *pdata = dev_get_platdata(&pdev->dev);
 
 	if (pdata->enable_external_pullup)
 		pdata->enable_external_pullup(1);
diff --git a/drivers/watchdog/Kconfig b/drivers/watchdog/Kconfig
index 6df632e0bb55..5be6e919f785 100644
--- a/drivers/watchdog/Kconfig
+++ b/drivers/watchdog/Kconfig
@@ -392,6 +392,25 @@ config RETU_WATCHDOG
 	  To compile this driver as a module, choose M here: the
 	  module will be called retu_wdt.
 
+config MOXART_WDT
+	tristate "MOXART watchdog"
+	depends on ARCH_MOXART
+	help
+	  Say Y here to include Watchdog timer support for the watchdog
+	  existing on the MOXA ART SoC series platforms.
+
+	  To compile this driver as a module, choose M here: the
+	  module will be called moxart_wdt.
+
+config SIRFSOC_WATCHDOG
+	tristate "SiRFSOC watchdog"
+	depends on ARCH_SIRF
+	select WATCHDOG_CORE
+	default y
+	help
+	  Support for CSR SiRFprimaII and SiRFatlasVI watchdog. When
+	  the watchdog triggers the system will be reset.
+
 # AVR32 Architecture
 
 config AT32AP700X_WDT
@@ -866,6 +885,7 @@ config VIA_WDT
 config W83627HF_WDT
 	tristate "W83627HF/W83627DHG Watchdog Timer"
 	depends on X86
+	select WATCHDOG_CORE
 	---help---
 	  This is the driver for the hardware watchdog on the W83627HF chipset
 	  as used in Advantech PC-9578 and Tyan S2721-533 motherboards
@@ -1125,6 +1145,13 @@ config LANTIQ_WDT
 	help
 	  Hardware driver for the Lantiq SoC Watchdog Timer.
 
+config RALINK_WDT
+	tristate "Ralink SoC watchdog"
+	select WATCHDOG_CORE
+	depends on RALINK
+	help
+	  Hardware driver for the Ralink SoC Watchdog Timer.
+
 # PARISC Architecture
 
 # POWERPC Architecture
diff --git a/drivers/watchdog/Makefile b/drivers/watchdog/Makefile
index 8c7b8bcbbdc5..91bd95a64baf 100644
--- a/drivers/watchdog/Makefile
+++ b/drivers/watchdog/Makefile
@@ -55,6 +55,8 @@ obj-$(CONFIG_IMX2_WDT) += imx2_wdt.o
 obj-$(CONFIG_UX500_WATCHDOG) += ux500_wdt.o
 obj-$(CONFIG_RETU_WATCHDOG) += retu_wdt.o
 obj-$(CONFIG_BCM2835_WDT) += bcm2835_wdt.o
+obj-$(CONFIG_MOXART_WDT) += moxart_wdt.o
+obj-$(CONFIG_SIRFSOC_WATCHDOG) += sirfsoc_wdt.o
 
 # AVR32 Architecture
 obj-$(CONFIG_AT32AP700X_WDT) += at32ap700x_wdt.o
@@ -134,6 +136,7 @@ obj-$(CONFIG_TXX9_WDT) += txx9wdt.o
 obj-$(CONFIG_OCTEON_WDT) += octeon-wdt.o
 octeon-wdt-y := octeon-wdt-main.o octeon-wdt-nmi.o
 obj-$(CONFIG_LANTIQ_WDT) += lantiq_wdt.o
+obj-$(CONFIG_RALINK_WDT) += rt2880_wdt.o
 
 # PARISC Architecture
 
diff --git a/drivers/watchdog/acquirewdt.c b/drivers/watchdog/acquirewdt.c
index 24a517777fa0..5cf1621def9c 100644
--- a/drivers/watchdog/acquirewdt.c
+++ b/drivers/watchdog/acquirewdt.c
@@ -60,8 +60,7 @@
 #include <linux/types.h>		/* For standard types (like size_t) */
 #include <linux/errno.h>		/* For the -ENODEV/... values */
 #include <linux/kernel.h>		/* For printk/panic/... */
-#include <linux/miscdevice.h>		/* For MODULE_ALIAS_MISCDEV
-							(WATCHDOG_MINOR) */
+#include <linux/miscdevice.h>		/* For struct miscdevice */
 #include <linux/watchdog.h>		/* For the watchdog specific items */
 #include <linux/fs.h>			/* For file operations */
 #include <linux/ioport.h>		/* For io-port access */
@@ -337,4 +336,3 @@ module_exit(acq_exit);
 MODULE_AUTHOR("David Woodhouse");
 MODULE_DESCRIPTION("Acquire Inc. Single Board Computer Watchdog Timer driver");
 MODULE_LICENSE("GPL");
-MODULE_ALIAS_MISCDEV(WATCHDOG_MINOR);
diff --git a/drivers/watchdog/advantechwdt.c b/drivers/watchdog/advantechwdt.c
index cc6702fc5268..a8961addc59c 100644
--- a/drivers/watchdog/advantechwdt.c
+++ b/drivers/watchdog/advantechwdt.c
@@ -345,4 +345,3 @@ module_exit(advwdt_exit);
 MODULE_LICENSE("GPL");
 MODULE_AUTHOR("Marek Michalkiewicz <marekm@linux.org.pl>");
 MODULE_DESCRIPTION("Advantech Single Board Computer WDT driver");
-MODULE_ALIAS_MISCDEV(WATCHDOG_MINOR);
diff --git a/drivers/watchdog/alim1535_wdt.c b/drivers/watchdog/alim1535_wdt.c
index 41b84936a521..fbb7b94cabfd 100644
--- a/drivers/watchdog/alim1535_wdt.c
+++ b/drivers/watchdog/alim1535_wdt.c
@@ -452,4 +452,3 @@ module_exit(watchdog_exit);
 MODULE_AUTHOR("Alan Cox");
 MODULE_DESCRIPTION("ALi M1535 PMU Watchdog Timer driver");
 MODULE_LICENSE("GPL");
-MODULE_ALIAS_MISCDEV(WATCHDOG_MINOR);
diff --git a/drivers/watchdog/alim7101_wdt.c b/drivers/watchdog/alim7101_wdt.c
index 5eee55012e33..12f0b762b528 100644
--- a/drivers/watchdog/alim7101_wdt.c
+++ b/drivers/watchdog/alim7101_wdt.c
@@ -425,4 +425,3 @@ MODULE_DEVICE_TABLE(pci, alim7101_pci_tbl);
 MODULE_AUTHOR("Steve Hill");
 MODULE_DESCRIPTION("ALi M7101 PMU Computer Watchdog Timer driver");
 MODULE_LICENSE("GPL");
-MODULE_ALIAS_MISCDEV(WATCHDOG_MINOR);
diff --git a/drivers/watchdog/ar7_wdt.c b/drivers/watchdog/ar7_wdt.c
index b3709f9cf5be..3a996576343a 100644
--- a/drivers/watchdog/ar7_wdt.c
+++ b/drivers/watchdog/ar7_wdt.c
@@ -46,7 +46,6 @@
 MODULE_AUTHOR("Nicolas Thill <nico@openwrt.org>");
 MODULE_DESCRIPTION(LONGNAME);
 MODULE_LICENSE("GPL");
-MODULE_ALIAS_MISCDEV(WATCHDOG_MINOR);
 
 static int margin = 60;
 module_param(margin, int, 0);
diff --git a/drivers/watchdog/at32ap700x_wdt.c b/drivers/watchdog/at32ap700x_wdt.c
index b178e717ef09..afe7d17e6776 100644
--- a/drivers/watchdog/at32ap700x_wdt.c
+++ b/drivers/watchdog/at32ap700x_wdt.c
@@ -434,4 +434,3 @@ module_platform_driver_probe(at32_wdt_driver, at32_wdt_probe);
 MODULE_AUTHOR("Hans-Christian Egtvedt <egtvedt@samfundet.no>");
 MODULE_DESCRIPTION("Watchdog driver for Atmel AT32AP700X");
 MODULE_LICENSE("GPL");
-MODULE_ALIAS_MISCDEV(WATCHDOG_MINOR);
diff --git a/drivers/watchdog/at91rm9200_wdt.c b/drivers/watchdog/at91rm9200_wdt.c
index 1c75260b987c..dee6cc21d270 100644
--- a/drivers/watchdog/at91rm9200_wdt.c
+++ b/drivers/watchdog/at91rm9200_wdt.c
@@ -269,7 +269,7 @@ static struct platform_driver at91wdt_driver = {
 	.driver		= {
 		.name	= "at91_wdt",
 		.owner	= THIS_MODULE,
-		.of_match_table = of_match_ptr(at91_wdt_dt_ids),
+		.of_match_table = at91_wdt_dt_ids,
 	},
 };
 
@@ -297,5 +297,4 @@ module_exit(at91_wdt_exit);
 MODULE_AUTHOR("Andrew Victor");
 MODULE_DESCRIPTION("Watchdog driver for Atmel AT91RM9200");
 MODULE_LICENSE("GPL");
-MODULE_ALIAS_MISCDEV(WATCHDOG_MINOR);
 MODULE_ALIAS("platform:at91_wdt");
diff --git a/drivers/watchdog/ath79_wdt.c b/drivers/watchdog/ath79_wdt.c
index 37cb09b27b63..9fa1f69dac13 100644
--- a/drivers/watchdog/ath79_wdt.c
+++ b/drivers/watchdog/ath79_wdt.c
@@ -329,4 +329,3 @@ MODULE_AUTHOR("Gabor Juhos <juhosg@openwrt.org");
 MODULE_AUTHOR("Imre Kaloz <kaloz@openwrt.org");
 MODULE_LICENSE("GPL v2");
 MODULE_ALIAS("platform:" DRIVER_NAME);
-MODULE_ALIAS_MISCDEV(WATCHDOG_MINOR);
diff --git a/drivers/watchdog/bcm2835_wdt.c b/drivers/watchdog/bcm2835_wdt.c
index 61566fc47f84..a6a2cebb2587 100644
--- a/drivers/watchdog/bcm2835_wdt.c
+++ b/drivers/watchdog/bcm2835_wdt.c
@@ -186,4 +186,3 @@ MODULE_PARM_DESC(nowayout, "Watchdog cannot be stopped once started (default="
 MODULE_AUTHOR("Lubomir Rintel <lkundrak@v3.sk>");
 MODULE_DESCRIPTION("Driver for Broadcom BCM2835 watchdog timer");
 MODULE_LICENSE("GPL");
-MODULE_ALIAS_MISCDEV(WATCHDOG_MINOR);
diff --git a/drivers/watchdog/bcm63xx_wdt.c b/drivers/watchdog/bcm63xx_wdt.c
index a14a58d9d110..4eb188b87f8e 100644
--- a/drivers/watchdog/bcm63xx_wdt.c
+++ b/drivers/watchdog/bcm63xx_wdt.c
@@ -317,5 +317,4 @@ MODULE_AUTHOR("Miguel Gaio <miguel.gaio@efixo.com>");
 MODULE_AUTHOR("Florian Fainelli <florian@openwrt.org>");
 MODULE_DESCRIPTION("Driver for the Broadcom BCM63xx SoC watchdog");
 MODULE_LICENSE("GPL");
-MODULE_ALIAS_MISCDEV(WATCHDOG_MINOR);
 MODULE_ALIAS("platform:bcm63xx-wdt");
diff --git a/drivers/watchdog/bfin_wdt.c b/drivers/watchdog/bfin_wdt.c
index 5d36d6fb4969..a3b6a5b30f9f 100644
--- a/drivers/watchdog/bfin_wdt.c
+++ b/drivers/watchdog/bfin_wdt.c
@@ -465,7 +465,6 @@ module_exit(bfin_wdt_exit);
 MODULE_AUTHOR("Michele d'Amico, Mike Frysinger <vapier@gentoo.org>");
 MODULE_DESCRIPTION("Blackfin Watchdog Device Driver");
 MODULE_LICENSE("GPL");
-MODULE_ALIAS_MISCDEV(WATCHDOG_MINOR);
 
 module_param(timeout, uint, 0);
 MODULE_PARM_DESC(timeout,
diff --git a/drivers/watchdog/cpu5wdt.c b/drivers/watchdog/cpu5wdt.c
index f270bb7bc456..f7ae49edb518 100644
--- a/drivers/watchdog/cpu5wdt.c
+++ b/drivers/watchdog/cpu5wdt.c
@@ -289,7 +289,6 @@ MODULE_AUTHOR("Heiko Ronsdorf <hero@ihg.uni-duisburg.de>");
 MODULE_DESCRIPTION("sma cpu5 watchdog driver");
 MODULE_SUPPORTED_DEVICE("sma cpu5 watchdog");
 MODULE_LICENSE("GPL");
-MODULE_ALIAS_MISCDEV(WATCHDOG_MINOR);
 
 module_param(port, int, 0);
 MODULE_PARM_DESC(port, "base address of watchdog card, default is 0x91");
diff --git a/drivers/watchdog/davinci_wdt.c b/drivers/watchdog/davinci_wdt.c
index bead7740c86a..dd625cca1ae5 100644
--- a/drivers/watchdog/davinci_wdt.c
+++ b/drivers/watchdog/davinci_wdt.c
@@ -267,5 +267,4 @@ MODULE_PARM_DESC(heartbeat,
 		 __MODULE_STRING(DEFAULT_HEARTBEAT));
 
 MODULE_LICENSE("GPL");
-MODULE_ALIAS_MISCDEV(WATCHDOG_MINOR);
 MODULE_ALIAS("platform:watchdog");
diff --git a/drivers/watchdog/dw_wdt.c b/drivers/watchdog/dw_wdt.c
index e621098bf663..a46f5c7ee7ff 100644
--- a/drivers/watchdog/dw_wdt.c
+++ b/drivers/watchdog/dw_wdt.c
@@ -29,6 +29,7 @@
 #include <linux/miscdevice.h>
 #include <linux/module.h>
 #include <linux/moduleparam.h>
+#include <linux/of.h>
 #include <linux/pm.h>
 #include <linux/platform_device.h>
 #include <linux/spinlock.h>
@@ -203,12 +204,12 @@ static long dw_wdt_ioctl(struct file *filp, unsigned int cmd, unsigned long arg)
 
 	switch (cmd) {
 	case WDIOC_GETSUPPORT:
-		return copy_to_user((struct watchdog_info *)arg, &dw_wdt_ident,
+		return copy_to_user((void __user *)arg, &dw_wdt_ident,
 				    sizeof(dw_wdt_ident)) ? -EFAULT : 0;
 
 	case WDIOC_GETSTATUS:
 	case WDIOC_GETBOOTSTATUS:
-		return put_user(0, (int *)arg);
+		return put_user(0, (int __user *)arg);
 
 	case WDIOC_KEEPALIVE:
 		dw_wdt_set_next_heartbeat();
@@ -252,17 +253,17 @@ static int dw_wdt_release(struct inode *inode, struct file *filp)
 	return 0;
 }
 
-#ifdef CONFIG_PM
+#ifdef CONFIG_PM_SLEEP
 static int dw_wdt_suspend(struct device *dev)
 {
-	clk_disable(dw_wdt.clk);
+	clk_disable_unprepare(dw_wdt.clk);
 
 	return 0;
 }
 
 static int dw_wdt_resume(struct device *dev)
 {
-	int err = clk_enable(dw_wdt.clk);
+	int err = clk_prepare_enable(dw_wdt.clk);
 
 	if (err)
 		return err;
@@ -271,12 +272,9 @@ static int dw_wdt_resume(struct device *dev)
 
 	return 0;
 }
+#endif /* CONFIG_PM_SLEEP */
 
-static const struct dev_pm_ops dw_wdt_pm_ops = {
-	.suspend	= dw_wdt_suspend,
-	.resume		= dw_wdt_resume,
-};
-#endif /* CONFIG_PM */
+static SIMPLE_DEV_PM_OPS(dw_wdt_pm_ops, dw_wdt_suspend, dw_wdt_resume);
 
 static const struct file_operations wdt_fops = {
 	.owner		= THIS_MODULE,
@@ -309,7 +307,7 @@ static int dw_wdt_drv_probe(struct platform_device *pdev)
 	if (IS_ERR(dw_wdt.clk))
 		return PTR_ERR(dw_wdt.clk);
 
-	ret = clk_enable(dw_wdt.clk);
+	ret = clk_prepare_enable(dw_wdt.clk);
 	if (ret)
 		return ret;
 
@@ -326,7 +324,7 @@ static int dw_wdt_drv_probe(struct platform_device *pdev)
 	return 0;
 
 out_disable_clk:
-	clk_disable(dw_wdt.clk);
+	clk_disable_unprepare(dw_wdt.clk);
 
 	return ret;
 }
@@ -335,20 +333,27 @@ static int dw_wdt_drv_remove(struct platform_device *pdev)
 {
 	misc_deregister(&dw_wdt_miscdev);
 
-	clk_disable(dw_wdt.clk);
+	clk_disable_unprepare(dw_wdt.clk);
 
 	return 0;
 }
 
+#ifdef CONFIG_OF
+static const struct of_device_id dw_wdt_of_match[] = {
+	{ .compatible = "snps,dw-wdt", },
+	{ /* sentinel */ }
+};
+MODULE_DEVICE_TABLE(of, dw_wdt_of_match);
+#endif
+
 static struct platform_driver dw_wdt_driver = {
 	.probe		= dw_wdt_drv_probe,
 	.remove		= dw_wdt_drv_remove,
 	.driver		= {
 		.name	= "dw_wdt",
 		.owner	= THIS_MODULE,
-#ifdef CONFIG_PM
+		.of_match_table = of_match_ptr(dw_wdt_of_match),
 		.pm	= &dw_wdt_pm_ops,
-#endif /* CONFIG_PM */
 	},
 };
 
@@ -357,4 +362,3 @@ module_platform_driver(dw_wdt_driver);
 MODULE_AUTHOR("Jamie Iles");
 MODULE_DESCRIPTION("Synopsys DesignWare Watchdog Driver");
 MODULE_LICENSE("GPL");
-MODULE_ALIAS_MISCDEV(WATCHDOG_MINOR);
diff --git a/drivers/watchdog/ep93xx_wdt.c b/drivers/watchdog/ep93xx_wdt.c
index e0574844c313..833e81311848 100644
--- a/drivers/watchdog/ep93xx_wdt.c
+++ b/drivers/watchdog/ep93xx_wdt.c
@@ -179,4 +179,3 @@ MODULE_AUTHOR("Ray Lehtiniemi <rayl@mail.com>,"
 MODULE_DESCRIPTION("EP93xx Watchdog");
 MODULE_LICENSE("GPL");
 MODULE_VERSION(WDT_VERSION);
-MODULE_ALIAS_MISCDEV(WATCHDOG_MINOR);
diff --git a/drivers/watchdog/eurotechwdt.c b/drivers/watchdog/eurotechwdt.c
index cd31b8a2a729..23ee53240c4c 100644
--- a/drivers/watchdog/eurotechwdt.c
+++ b/drivers/watchdog/eurotechwdt.c
@@ -477,4 +477,3 @@ module_exit(eurwdt_exit);
 MODULE_AUTHOR("Rodolfo Giometti");
 MODULE_DESCRIPTION("Driver for Eurotech CPU-1220/1410 on board watchdog");
 MODULE_LICENSE("GPL");
-MODULE_ALIAS_MISCDEV(WATCHDOG_MINOR);
diff --git a/drivers/watchdog/gef_wdt.c b/drivers/watchdog/gef_wdt.c
index 3755833430dc..25beb30878d7 100644
--- a/drivers/watchdog/gef_wdt.c
+++ b/drivers/watchdog/gef_wdt.c
@@ -331,5 +331,4 @@ module_exit(gef_wdt_exit);
 MODULE_AUTHOR("Martyn Welch <martyn.welch@ge.com>");
 MODULE_DESCRIPTION("GE watchdog driver");
 MODULE_LICENSE("GPL");
-MODULE_ALIAS_MISCDEV(WATCHDOG_MINOR);
 MODULE_ALIAS("platform:gef_wdt");
diff --git a/drivers/watchdog/geodewdt.c b/drivers/watchdog/geodewdt.c
index fcd599d4e225..4a6ae84b42bc 100644
--- a/drivers/watchdog/geodewdt.c
+++ b/drivers/watchdog/geodewdt.c
@@ -297,4 +297,3 @@ module_exit(geodewdt_exit);
 MODULE_AUTHOR("Advanced Micro Devices, Inc");
 MODULE_DESCRIPTION("Geode GX/LX Watchdog Driver");
 MODULE_LICENSE("GPL");
-MODULE_ALIAS_MISCDEV(WATCHDOG_MINOR);
diff --git a/drivers/watchdog/hpwdt.c b/drivers/watchdog/hpwdt.c
index 19f3c3fc65f4..45b979d9dd13 100644
--- a/drivers/watchdog/hpwdt.c
+++ b/drivers/watchdog/hpwdt.c
@@ -881,7 +881,6 @@ MODULE_AUTHOR("Tom Mingarelli");
 MODULE_DESCRIPTION("hp watchdog driver");
 MODULE_LICENSE("GPL");
 MODULE_VERSION(HPWDT_VERSION);
-MODULE_ALIAS_MISCDEV(WATCHDOG_MINOR);
 
 module_param(soft_margin, int, 0);
 MODULE_PARM_DESC(soft_margin, "Watchdog timeout in seconds");
diff --git a/drivers/watchdog/i6300esb.c b/drivers/watchdog/i6300esb.c
index 2b2ea13d03ea..a72fe9361ddf 100644
--- a/drivers/watchdog/i6300esb.c
+++ b/drivers/watchdog/i6300esb.c
@@ -497,4 +497,3 @@ module_pci_driver(esb_driver);
 MODULE_AUTHOR("Ross Biro and David Härdeman");
 MODULE_DESCRIPTION("Watchdog driver for Intel 6300ESB chipsets");
 MODULE_LICENSE("GPL");
-MODULE_ALIAS_MISCDEV(WATCHDOG_MINOR);
diff --git a/drivers/watchdog/iTCO_wdt.c b/drivers/watchdog/iTCO_wdt.c
index 6130321da387..04f8af65acfd 100644
--- a/drivers/watchdog/iTCO_wdt.c
+++ b/drivers/watchdog/iTCO_wdt.c
@@ -56,8 +56,6 @@
 #include <linux/types.h>		/* For standard types (like size_t) */
 #include <linux/errno.h>		/* For the -ENODEV/... values */
 #include <linux/kernel.h>		/* For printk/panic/... */
-#include <linux/miscdevice.h>		/* For MODULE_ALIAS_MISCDEV
-							(WATCHDOG_MINOR) */
 #include <linux/watchdog.h>		/* For the watchdog specific items */
 #include <linux/init.h>			/* For __init/__exit/... */
 #include <linux/fs.h>			/* For file operations */
@@ -394,7 +392,7 @@ static int iTCO_wdt_probe(struct platform_device *dev)
 {
 	int ret = -ENODEV;
 	unsigned long val32;
-	struct lpc_ich_info *ich_info = dev->dev.platform_data;
+	struct lpc_ich_info *ich_info = dev_get_platdata(&dev->dev);
 
 	if (!ich_info)
 		goto out;
@@ -582,5 +580,4 @@ MODULE_AUTHOR("Wim Van Sebroeck <wim@iguana.be>");
 MODULE_DESCRIPTION("Intel TCO WatchDog Timer Driver");
 MODULE_VERSION(DRV_VERSION);
 MODULE_LICENSE("GPL");
-MODULE_ALIAS_MISCDEV(WATCHDOG_MINOR);
 MODULE_ALIAS("platform:" DRV_NAME);
diff --git a/drivers/watchdog/ib700wdt.c b/drivers/watchdog/ib700wdt.c
index eb6b5cc98ec6..7ae36690c449 100644
--- a/drivers/watchdog/ib700wdt.c
+++ b/drivers/watchdog/ib700wdt.c
@@ -382,6 +382,5 @@ module_exit(ibwdt_exit);
 MODULE_AUTHOR("Charles Howes <chowes@vsol.net>");
 MODULE_DESCRIPTION("IB700 SBC watchdog driver");
 MODULE_LICENSE("GPL");
-MODULE_ALIAS_MISCDEV(WATCHDOG_MINOR);
 
 /* end of ib700wdt.c */
diff --git a/drivers/watchdog/ibmasr.c b/drivers/watchdog/ibmasr.c
index bc3fb8fe89ab..db0a34460e57 100644
--- a/drivers/watchdog/ibmasr.c
+++ b/drivers/watchdog/ibmasr.c
@@ -419,4 +419,3 @@ MODULE_PARM_DESC(nowayout,
 MODULE_DESCRIPTION("IBM Automatic Server Restart driver");
 MODULE_AUTHOR("Andrey Panin");
 MODULE_LICENSE("GPL");
-MODULE_ALIAS_MISCDEV(WATCHDOG_MINOR);
diff --git a/drivers/watchdog/ie6xx_wdt.c b/drivers/watchdog/ie6xx_wdt.c
index e24ef6a6e064..70a240297c6d 100644
--- a/drivers/watchdog/ie6xx_wdt.c
+++ b/drivers/watchdog/ie6xx_wdt.c
@@ -344,5 +344,4 @@ module_exit(ie6xx_wdt_exit);
 MODULE_AUTHOR("Alexander Stein <alexander.stein@systec-electronic.com>");
 MODULE_DESCRIPTION("Intel Atom E6xx Watchdog Device Driver");
 MODULE_LICENSE("GPL");
-MODULE_ALIAS_MISCDEV(WATCHDOG_MINOR);
 MODULE_ALIAS("platform:" DRIVER_NAME);
diff --git a/drivers/watchdog/imx2_wdt.c b/drivers/watchdog/imx2_wdt.c
index 693ac3f4de5a..b4786bccc42c 100644
--- a/drivers/watchdog/imx2_wdt.c
+++ b/drivers/watchdog/imx2_wdt.c
@@ -322,6 +322,7 @@ static const struct of_device_id imx2_wdt_dt_ids[] = {
 	{ .compatible = "fsl,imx21-wdt", },
 	{ /* sentinel */ }
 };
+MODULE_DEVICE_TABLE(of, imx2_wdt_dt_ids);
 
 static struct platform_driver imx2_wdt_driver = {
 	.remove		= __exit_p(imx2_wdt_remove),
@@ -338,5 +339,4 @@ module_platform_driver_probe(imx2_wdt_driver, imx2_wdt_probe);
 MODULE_AUTHOR("Wolfram Sang");
 MODULE_DESCRIPTION("Watchdog driver for IMX2 and later");
 MODULE_LICENSE("GPL v2");
-MODULE_ALIAS_MISCDEV(WATCHDOG_MINOR);
 MODULE_ALIAS("platform:" DRIVER_NAME);
diff --git a/drivers/watchdog/indydog.c b/drivers/watchdog/indydog.c
index 6d90f7a2ce22..1b5c25a47b87 100644
--- a/drivers/watchdog/indydog.c
+++ b/drivers/watchdog/indydog.c
@@ -214,4 +214,3 @@ module_exit(watchdog_exit);
 MODULE_AUTHOR("Guido Guenther <agx@sigxcpu.org>");
 MODULE_DESCRIPTION("Hardware Watchdog Device for SGI IP22");
 MODULE_LICENSE("GPL");
-MODULE_ALIAS_MISCDEV(WATCHDOG_MINOR);
diff --git a/drivers/watchdog/intel_scu_watchdog.c b/drivers/watchdog/intel_scu_watchdog.c
index 8ced25613956..e13e65e996aa 100644
--- a/drivers/watchdog/intel_scu_watchdog.c
+++ b/drivers/watchdog/intel_scu_watchdog.c
@@ -564,5 +564,4 @@ module_exit(intel_scu_watchdog_exit);
 MODULE_AUTHOR("Intel Corporation");
 MODULE_DESCRIPTION("Intel SCU Watchdog Device Driver");
 MODULE_LICENSE("GPL");
-MODULE_ALIAS_MISCDEV(WATCHDOG_MINOR);
 MODULE_VERSION(WDT_VER);
diff --git a/drivers/watchdog/iop_wdt.c b/drivers/watchdog/iop_wdt.c
index d964faf1a250..b16013ffacc2 100644
--- a/drivers/watchdog/iop_wdt.c
+++ b/drivers/watchdog/iop_wdt.c
@@ -259,4 +259,3 @@ MODULE_PARM_DESC(nowayout, "Watchdog cannot be stopped once started");
 MODULE_AUTHOR("Curt E Bruns <curt.e.bruns@intel.com>");
 MODULE_DESCRIPTION("iop watchdog timer driver");
 MODULE_LICENSE("GPL");
-MODULE_ALIAS_MISCDEV(WATCHDOG_MINOR);
diff --git a/drivers/watchdog/it8712f_wdt.c b/drivers/watchdog/it8712f_wdt.c
index f4cce6d66a55..41b3979a9d87 100644
--- a/drivers/watchdog/it8712f_wdt.c
+++ b/drivers/watchdog/it8712f_wdt.c
@@ -41,7 +41,6 @@
 MODULE_AUTHOR("Jorge Boncompte - DTI2 <jorge@dti2.net>");
 MODULE_DESCRIPTION("IT8712F Watchdog Driver");
 MODULE_LICENSE("GPL");
-MODULE_ALIAS_MISCDEV(WATCHDOG_MINOR);
 
 static int max_units = 255;
 static int margin = 60;		/* in seconds */
diff --git a/drivers/watchdog/it87_wdt.c b/drivers/watchdog/it87_wdt.c
index d3dcc6988b5f..e2bba68ae71e 100644
--- a/drivers/watchdog/it87_wdt.c
+++ b/drivers/watchdog/it87_wdt.c
@@ -772,4 +772,3 @@ module_exit(it87_wdt_exit);
 MODULE_AUTHOR("Oliver Schuster");
 MODULE_DESCRIPTION("Hardware Watchdog Device Driver for IT87xx EC-LPC I/O");
 MODULE_LICENSE("GPL");
-MODULE_ALIAS_MISCDEV(WATCHDOG_MINOR);
diff --git a/drivers/watchdog/ixp4xx_wdt.c b/drivers/watchdog/ixp4xx_wdt.c
index 5580b4fff7fe..f20cc53ff719 100644
--- a/drivers/watchdog/ixp4xx_wdt.c
+++ b/drivers/watchdog/ixp4xx_wdt.c
@@ -208,5 +208,3 @@ module_param(nowayout, bool, 0);
 MODULE_PARM_DESC(nowayout, "Watchdog cannot be stopped once started");
 
 MODULE_LICENSE("GPL");
-MODULE_ALIAS_MISCDEV(WATCHDOG_MINOR);
-
diff --git a/drivers/watchdog/jz4740_wdt.c b/drivers/watchdog/jz4740_wdt.c
index d1afdf684c18..2de486a7eea1 100644
--- a/drivers/watchdog/jz4740_wdt.c
+++ b/drivers/watchdog/jz4740_wdt.c
@@ -222,5 +222,4 @@ module_platform_driver(jz4740_wdt_driver);
 MODULE_AUTHOR("Paul Cercueil <paul@crapouillou.net>");
 MODULE_DESCRIPTION("jz4740 Watchdog Driver");
 MODULE_LICENSE("GPL");
-MODULE_ALIAS_MISCDEV(WATCHDOG_MINOR);
 MODULE_ALIAS("platform:jz4740-wdt");
diff --git a/drivers/watchdog/kempld_wdt.c b/drivers/watchdog/kempld_wdt.c
index 5c3d4df63e68..a1a3638c579c 100644
--- a/drivers/watchdog/kempld_wdt.c
+++ b/drivers/watchdog/kempld_wdt.c
@@ -67,7 +67,7 @@ enum {
 	PRESCALER_12,
 };
 
-const u32 kempld_prescaler[] = {
+static const u32 kempld_prescaler[] = {
 	[PRESCALER_21] = (1 << 21) - 1,
 	[PRESCALER_17] = (1 << 17) - 1,
 	[PRESCALER_12] = (1 << 12) - 1,
@@ -361,7 +361,7 @@ static long kempld_wdt_ioctl(struct watchdog_device *wdd, unsigned int cmd,
 		ret = kempld_wdt_keepalive(wdd);
 		break;
 	case WDIOC_GETPRETIMEOUT:
-		ret = put_user(wdt_data->pretimeout, (int *)arg);
+		ret = put_user(wdt_data->pretimeout, (int __user *)arg);
 		break;
 	}
 
@@ -578,4 +578,3 @@ module_platform_driver(kempld_wdt_driver);
 MODULE_DESCRIPTION("KEM PLD Watchdog Driver");
 MODULE_AUTHOR("Michael Brunner <michael.brunner@kontron.com>");
 MODULE_LICENSE("GPL");
-MODULE_ALIAS_MISCDEV(WATCHDOG_MINOR);
diff --git a/drivers/watchdog/ks8695_wdt.c b/drivers/watchdog/ks8695_wdt.c
index dce9ecffd44a..40ca5594a336 100644
--- a/drivers/watchdog/ks8695_wdt.c
+++ b/drivers/watchdog/ks8695_wdt.c
@@ -323,5 +323,4 @@ module_exit(ks8695_wdt_exit);
 MODULE_AUTHOR("Andrew Victor");
 MODULE_DESCRIPTION("Watchdog driver for KS8695");
 MODULE_LICENSE("GPL");
-MODULE_ALIAS_MISCDEV(WATCHDOG_MINOR);
 MODULE_ALIAS("platform:ks8695_wdt");
diff --git a/drivers/watchdog/lantiq_wdt.c b/drivers/watchdog/lantiq_wdt.c
index 088fd0c9d888..3b3148c764a3 100644
--- a/drivers/watchdog/lantiq_wdt.c
+++ b/drivers/watchdog/lantiq_wdt.c
@@ -249,4 +249,3 @@ MODULE_PARM_DESC(nowayout, "Watchdog cannot be stopped once started");
 MODULE_AUTHOR("John Crispin <blogic@openwrt.org>");
 MODULE_DESCRIPTION("Lantiq SoC Watchdog");
 MODULE_LICENSE("GPL");
-MODULE_ALIAS_MISCDEV(WATCHDOG_MINOR);
diff --git a/drivers/watchdog/m54xx_wdt.c b/drivers/watchdog/m54xx_wdt.c
index 173494a681e6..da6fa2b68074 100644
--- a/drivers/watchdog/m54xx_wdt.c
+++ b/drivers/watchdog/m54xx_wdt.c
@@ -223,4 +223,3 @@ module_param(nowayout, bool, 0);
 MODULE_PARM_DESC(nowayout, "Watchdog cannot be stopped once started");
 
 MODULE_LICENSE("GPL");
-MODULE_ALIAS_MISCDEV(WATCHDOG_MINOR);
diff --git a/drivers/watchdog/machzwd.c b/drivers/watchdog/machzwd.c
index bf84f788e592..9826b59ef734 100644
--- a/drivers/watchdog/machzwd.c
+++ b/drivers/watchdog/machzwd.c
@@ -92,7 +92,6 @@ static unsigned short zf_readw(unsigned char port)
 MODULE_AUTHOR("Fernando Fuganti <fuganti@conectiva.com.br>");
 MODULE_DESCRIPTION("MachZ ZF-Logic Watchdog driver");
 MODULE_LICENSE("GPL");
-MODULE_ALIAS_MISCDEV(WATCHDOG_MINOR);
 
 static bool nowayout = WATCHDOG_NOWAYOUT;
 module_param(nowayout, bool, 0);
diff --git a/drivers/watchdog/max63xx_wdt.c b/drivers/watchdog/max63xx_wdt.c
index cc9d328086ed..6d4f3998e1f6 100644
--- a/drivers/watchdog/max63xx_wdt.c
+++ b/drivers/watchdog/max63xx_wdt.c
@@ -258,4 +258,3 @@ MODULE_PARM_DESC(nodelay,
 		 "(max6373/74 only, default=0)");
 
 MODULE_LICENSE("GPL");
-MODULE_ALIAS_MISCDEV(WATCHDOG_MINOR);
diff --git a/drivers/watchdog/mixcomwd.c b/drivers/watchdog/mixcomwd.c
index 97d62ee50341..be86ea359eee 100644
--- a/drivers/watchdog/mixcomwd.c
+++ b/drivers/watchdog/mixcomwd.c
@@ -315,4 +315,3 @@ MODULE_AUTHOR("Gergely Madarasz <gorgo@itc.hu>");
 MODULE_DESCRIPTION("MixCom Watchdog driver");
 MODULE_VERSION(VERSION);
 MODULE_LICENSE("GPL");
-MODULE_ALIAS_MISCDEV(WATCHDOG_MINOR);
diff --git a/drivers/watchdog/moxart_wdt.c b/drivers/watchdog/moxart_wdt.c
new file mode 100644
index 000000000000..4166e4d116a8
--- /dev/null
+++ b/drivers/watchdog/moxart_wdt.c
@@ -0,0 +1,165 @@
+/*
+ * MOXA ART SoCs watchdog driver.
+ *
+ * Copyright (C) 2013 Jonas Jensen
+ *
+ * Jonas Jensen <jonas.jensen@gmail.com>
+ *
+ * This file is licensed under the terms of the GNU General Public
+ * License version 2.  This program is licensed "as is" without any
+ * warranty of any kind, whether express or implied.
+ */
+
+#include <linux/clk.h>
+#include <linux/io.h>
+#include <linux/module.h>
+#include <linux/err.h>
+#include <linux/kernel.h>
+#include <linux/platform_device.h>
+#include <linux/watchdog.h>
+#include <linux/moduleparam.h>
+
+#define REG_COUNT			0x4
+#define REG_MODE			0x8
+#define REG_ENABLE			0xC
+
+struct moxart_wdt_dev {
+	struct watchdog_device dev;
+	void __iomem *base;
+	unsigned int clock_frequency;
+};
+
+static int heartbeat;
+
+static int moxart_wdt_stop(struct watchdog_device *wdt_dev)
+{
+	struct moxart_wdt_dev *moxart_wdt = watchdog_get_drvdata(wdt_dev);
+
+	writel(0, moxart_wdt->base + REG_ENABLE);
+
+	return 0;
+}
+
+static int moxart_wdt_start(struct watchdog_device *wdt_dev)
+{
+	struct moxart_wdt_dev *moxart_wdt = watchdog_get_drvdata(wdt_dev);
+
+	writel(moxart_wdt->clock_frequency * wdt_dev->timeout,
+	       moxart_wdt->base + REG_COUNT);
+	writel(0x5ab9, moxart_wdt->base + REG_MODE);
+	writel(0x03, moxart_wdt->base + REG_ENABLE);
+
+	return 0;
+}
+
+static int moxart_wdt_set_timeout(struct watchdog_device *wdt_dev,
+				  unsigned int timeout)
+{
+	wdt_dev->timeout = timeout;
+
+	return 0;
+}
+
+static const struct watchdog_info moxart_wdt_info = {
+	.identity       = "moxart-wdt",
+	.options        = WDIOF_SETTIMEOUT | WDIOF_KEEPALIVEPING |
+			  WDIOF_MAGICCLOSE,
+};
+
+static const struct watchdog_ops moxart_wdt_ops = {
+	.owner          = THIS_MODULE,
+	.start          = moxart_wdt_start,
+	.stop           = moxart_wdt_stop,
+	.set_timeout    = moxart_wdt_set_timeout,
+};
+
+static int moxart_wdt_probe(struct platform_device *pdev)
+{
+	struct moxart_wdt_dev *moxart_wdt;
+	struct device *dev = &pdev->dev;
+	struct device_node *node = dev->of_node;
+	struct resource *res;
+	struct clk *clk;
+	int err;
+	unsigned int max_timeout;
+	bool nowayout = WATCHDOG_NOWAYOUT;
+
+	moxart_wdt = devm_kzalloc(dev, sizeof(*moxart_wdt), GFP_KERNEL);
+	if (!moxart_wdt)
+		return -ENOMEM;
+
+	platform_set_drvdata(pdev, moxart_wdt);
+
+	res = platform_get_resource(pdev, IORESOURCE_MEM, 0);
+	moxart_wdt->base = devm_ioremap_resource(dev, res);
+	if (IS_ERR(moxart_wdt->base))
+		return PTR_ERR(moxart_wdt->base);
+
+	clk = of_clk_get(node, 0);
+	if (IS_ERR(clk)) {
+		pr_err("%s: of_clk_get failed\n", __func__);
+		return PTR_ERR(clk);
+	}
+
+	moxart_wdt->clock_frequency = clk_get_rate(clk);
+	if (moxart_wdt->clock_frequency == 0) {
+		pr_err("%s: incorrect clock frequency\n", __func__);
+		return -EINVAL;
+	}
+
+	max_timeout = UINT_MAX / moxart_wdt->clock_frequency;
+
+	moxart_wdt->dev.info = &moxart_wdt_info;
+	moxart_wdt->dev.ops = &moxart_wdt_ops;
+	moxart_wdt->dev.timeout = max_timeout;
+	moxart_wdt->dev.min_timeout = 1;
+	moxart_wdt->dev.max_timeout = max_timeout;
+	moxart_wdt->dev.parent = dev;
+
+	watchdog_init_timeout(&moxart_wdt->dev, heartbeat, dev);
+	watchdog_set_nowayout(&moxart_wdt->dev, nowayout);
+
+	watchdog_set_drvdata(&moxart_wdt->dev, moxart_wdt);
+
+	err = watchdog_register_device(&moxart_wdt->dev);
+	if (err)
+		return err;
+
+	dev_dbg(dev, "Watchdog enabled (heartbeat=%d sec, nowayout=%d)\n",
+		moxart_wdt->dev.timeout, nowayout);
+
+	return 0;
+}
+
+static int moxart_wdt_remove(struct platform_device *pdev)
+{
+	struct moxart_wdt_dev *moxart_wdt = platform_get_drvdata(pdev);
+
+	moxart_wdt_stop(&moxart_wdt->dev);
+	watchdog_unregister_device(&moxart_wdt->dev);
+
+	return 0;
+}
+
+static const struct of_device_id moxart_watchdog_match[] = {
+	{ .compatible = "moxa,moxart-watchdog" },
+	{ },
+};
+
+static struct platform_driver moxart_wdt_driver = {
+	.probe      = moxart_wdt_probe,
+	.remove     = moxart_wdt_remove,
+	.driver     = {
+		.name		= "moxart-watchdog",
+		.owner		= THIS_MODULE,
+		.of_match_table	= moxart_watchdog_match,
+	},
+};
+module_platform_driver(moxart_wdt_driver);
+
+module_param(heartbeat, int, 0);
+MODULE_PARM_DESC(heartbeat, "Watchdog heartbeat in seconds");
+
+MODULE_DESCRIPTION("MOXART watchdog driver");
+MODULE_LICENSE("GPL");
+MODULE_AUTHOR("Jonas Jensen <jonas.jensen@gmail.com>");
diff --git a/drivers/watchdog/mpc8xxx_wdt.c b/drivers/watchdog/mpc8xxx_wdt.c
index d0ebebae607c..d82152077fd9 100644
--- a/drivers/watchdog/mpc8xxx_wdt.c
+++ b/drivers/watchdog/mpc8xxx_wdt.c
@@ -330,4 +330,3 @@ MODULE_AUTHOR("Dave Updegraff, Kumar Gala");
 MODULE_DESCRIPTION("Driver for watchdog timer in MPC8xx/MPC83xx/MPC86xx "
 		   "uProcessors");
 MODULE_LICENSE("GPL");
-MODULE_ALIAS_MISCDEV(WATCHDOG_MINOR);
diff --git a/drivers/watchdog/mtx-1_wdt.c b/drivers/watchdog/mtx-1_wdt.c
index b4341110ad4f..edb31ffd7927 100644
--- a/drivers/watchdog/mtx-1_wdt.c
+++ b/drivers/watchdog/mtx-1_wdt.c
@@ -257,5 +257,4 @@ module_platform_driver(mtx1_wdt_driver);
 MODULE_AUTHOR("Michael Stickel, Florian Fainelli");
 MODULE_DESCRIPTION("Driver for the MTX-1 watchdog");
 MODULE_LICENSE("GPL");
-MODULE_ALIAS_MISCDEV(WATCHDOG_MINOR);
 MODULE_ALIAS("platform:mtx1-wdt");
diff --git a/drivers/watchdog/mv64x60_wdt.c b/drivers/watchdog/mv64x60_wdt.c
index e4cf98019265..f9fa58409396 100644
--- a/drivers/watchdog/mv64x60_wdt.c
+++ b/drivers/watchdog/mv64x60_wdt.c
@@ -255,7 +255,7 @@ static struct miscdevice mv64x60_wdt_miscdev = {
 
 static int mv64x60_wdt_probe(struct platform_device *dev)
 {
-	struct mv64x60_wdt_pdata *pdata = dev->dev.platform_data;
+	struct mv64x60_wdt_pdata *pdata = dev_get_platdata(&dev->dev);
 	struct resource *r;
 	int timeout = 10;
 
@@ -323,5 +323,4 @@ module_exit(mv64x60_wdt_exit);
 MODULE_AUTHOR("James Chapman <jchapman@katalix.com>");
 MODULE_DESCRIPTION("MV64x60 watchdog driver");
 MODULE_LICENSE("GPL");
-MODULE_ALIAS_MISCDEV(WATCHDOG_MINOR);
 MODULE_ALIAS("platform:" MV64x60_WDT_NAME);
diff --git a/drivers/watchdog/nuc900_wdt.c b/drivers/watchdog/nuc900_wdt.c
index b15b6efd91a1..a0d893b0930e 100644
--- a/drivers/watchdog/nuc900_wdt.c
+++ b/drivers/watchdog/nuc900_wdt.c
@@ -307,5 +307,4 @@ module_platform_driver(nuc900wdt_driver);
 MODULE_AUTHOR("Wan ZongShun <mcuos.com@gmail.com>");
 MODULE_DESCRIPTION("Watchdog driver for NUC900");
 MODULE_LICENSE("GPL");
-MODULE_ALIAS_MISCDEV(WATCHDOG_MINOR);
 MODULE_ALIAS("platform:nuc900-wdt");
diff --git a/drivers/watchdog/nv_tco.c b/drivers/watchdog/nv_tco.c
index 59cf19eeea07..231e5b9d5c8e 100644
--- a/drivers/watchdog/nv_tco.c
+++ b/drivers/watchdog/nv_tco.c
@@ -513,4 +513,3 @@ module_exit(nv_tco_cleanup_module);
 MODULE_AUTHOR("Mike Waychison");
 MODULE_DESCRIPTION("TCO timer driver for NV chipsets");
 MODULE_LICENSE("GPL");
-MODULE_ALIAS_MISCDEV(WATCHDOG_MINOR);
diff --git a/drivers/watchdog/of_xilinx_wdt.c b/drivers/watchdog/of_xilinx_wdt.c
index 4dd281f2c33f..fb57103c8ebc 100644
--- a/drivers/watchdog/of_xilinx_wdt.c
+++ b/drivers/watchdog/of_xilinx_wdt.c
@@ -405,4 +405,3 @@ module_platform_driver(xwdt_driver);
 MODULE_AUTHOR("Alejandro Cabrera <aldaya@gmail.com>");
 MODULE_DESCRIPTION("Xilinx Watchdog driver");
 MODULE_LICENSE("GPL v2");
-MODULE_ALIAS_MISCDEV(WATCHDOG_MINOR);
diff --git a/drivers/watchdog/omap_wdt.c b/drivers/watchdog/omap_wdt.c
index af88ffd1068f..09cf0135e8ac 100644
--- a/drivers/watchdog/omap_wdt.c
+++ b/drivers/watchdog/omap_wdt.c
@@ -68,14 +68,14 @@ static void omap_wdt_reload(struct omap_wdt_dev *wdev)
 	void __iomem    *base = wdev->base;
 
 	/* wait for posted write to complete */
-	while ((__raw_readl(base + OMAP_WATCHDOG_WPS)) & 0x08)
+	while ((readl_relaxed(base + OMAP_WATCHDOG_WPS)) & 0x08)
 		cpu_relax();
 
 	wdev->wdt_trgr_pattern = ~wdev->wdt_trgr_pattern;
-	__raw_writel(wdev->wdt_trgr_pattern, (base + OMAP_WATCHDOG_TGR));
+	writel_relaxed(wdev->wdt_trgr_pattern, (base + OMAP_WATCHDOG_TGR));
 
 	/* wait for posted write to complete */
-	while ((__raw_readl(base + OMAP_WATCHDOG_WPS)) & 0x08)
+	while ((readl_relaxed(base + OMAP_WATCHDOG_WPS)) & 0x08)
 		cpu_relax();
 	/* reloaded WCRR from WLDR */
 }
@@ -85,12 +85,12 @@ static void omap_wdt_enable(struct omap_wdt_dev *wdev)
 	void __iomem *base = wdev->base;
 
 	/* Sequence to enable the watchdog */
-	__raw_writel(0xBBBB, base + OMAP_WATCHDOG_SPR);
-	while ((__raw_readl(base + OMAP_WATCHDOG_WPS)) & 0x10)
+	writel_relaxed(0xBBBB, base + OMAP_WATCHDOG_SPR);
+	while ((readl_relaxed(base + OMAP_WATCHDOG_WPS)) & 0x10)
 		cpu_relax();
 
-	__raw_writel(0x4444, base + OMAP_WATCHDOG_SPR);
-	while ((__raw_readl(base + OMAP_WATCHDOG_WPS)) & 0x10)
+	writel_relaxed(0x4444, base + OMAP_WATCHDOG_SPR);
+	while ((readl_relaxed(base + OMAP_WATCHDOG_WPS)) & 0x10)
 		cpu_relax();
 }
 
@@ -99,12 +99,12 @@ static void omap_wdt_disable(struct omap_wdt_dev *wdev)
 	void __iomem *base = wdev->base;
 
 	/* sequence required to disable watchdog */
-	__raw_writel(0xAAAA, base + OMAP_WATCHDOG_SPR);	/* TIMER_MODE */
-	while (__raw_readl(base + OMAP_WATCHDOG_WPS) & 0x10)
+	writel_relaxed(0xAAAA, base + OMAP_WATCHDOG_SPR);	/* TIMER_MODE */
+	while (readl_relaxed(base + OMAP_WATCHDOG_WPS) & 0x10)
 		cpu_relax();
 
-	__raw_writel(0x5555, base + OMAP_WATCHDOG_SPR);	/* TIMER_MODE */
-	while (__raw_readl(base + OMAP_WATCHDOG_WPS) & 0x10)
+	writel_relaxed(0x5555, base + OMAP_WATCHDOG_SPR);	/* TIMER_MODE */
+	while (readl_relaxed(base + OMAP_WATCHDOG_WPS) & 0x10)
 		cpu_relax();
 }
 
@@ -115,11 +115,11 @@ static void omap_wdt_set_timer(struct omap_wdt_dev *wdev,
 	void __iomem *base = wdev->base;
 
 	/* just count up at 32 KHz */
-	while (__raw_readl(base + OMAP_WATCHDOG_WPS) & 0x04)
+	while (readl_relaxed(base + OMAP_WATCHDOG_WPS) & 0x04)
 		cpu_relax();
 
-	__raw_writel(pre_margin, base + OMAP_WATCHDOG_LDR);
-	while (__raw_readl(base + OMAP_WATCHDOG_WPS) & 0x04)
+	writel_relaxed(pre_margin, base + OMAP_WATCHDOG_LDR);
+	while (readl_relaxed(base + OMAP_WATCHDOG_WPS) & 0x04)
 		cpu_relax();
 }
 
@@ -135,11 +135,11 @@ static int omap_wdt_start(struct watchdog_device *wdog)
 	pm_runtime_get_sync(wdev->dev);
 
 	/* initialize prescaler */
-	while (__raw_readl(base + OMAP_WATCHDOG_WPS) & 0x01)
+	while (readl_relaxed(base + OMAP_WATCHDOG_WPS) & 0x01)
 		cpu_relax();
 
-	__raw_writel((1 << 5) | (PTV << 2), base + OMAP_WATCHDOG_CNTRL);
-	while (__raw_readl(base + OMAP_WATCHDOG_WPS) & 0x01)
+	writel_relaxed((1 << 5) | (PTV << 2), base + OMAP_WATCHDOG_CNTRL);
+	while (readl_relaxed(base + OMAP_WATCHDOG_WPS) & 0x01)
 		cpu_relax();
 
 	omap_wdt_set_timer(wdev, wdog->timeout);
@@ -205,7 +205,7 @@ static const struct watchdog_ops omap_wdt_ops = {
 
 static int omap_wdt_probe(struct platform_device *pdev)
 {
-	struct omap_wd_timer_platform_data *pdata = pdev->dev.platform_data;
+	struct omap_wd_timer_platform_data *pdata = dev_get_platdata(&pdev->dev);
 	struct watchdog_device *omap_wdt;
 	struct resource *res, *mem;
 	struct omap_wdt_dev *wdev;
@@ -275,7 +275,7 @@ static int omap_wdt_probe(struct platform_device *pdev)
 	}
 
 	pr_info("OMAP Watchdog Timer Rev 0x%02x: initial timeout %d sec\n",
-		__raw_readl(wdev->base + OMAP_WATCHDOG_REV) & 0xFF,
+		readl_relaxed(wdev->base + OMAP_WATCHDOG_REV) & 0xFF,
 		omap_wdt->timeout);
 
 	pm_runtime_put_sync(wdev->dev);
diff --git a/drivers/watchdog/orion_wdt.c b/drivers/watchdog/orion_wdt.c
index 4ea5fcccac02..44edca66d564 100644
--- a/drivers/watchdog/orion_wdt.c
+++ b/drivers/watchdog/orion_wdt.c
@@ -207,7 +207,7 @@ static struct platform_driver orion_wdt_driver = {
 	.driver		= {
 		.owner	= THIS_MODULE,
 		.name	= "orion_wdt",
-		.of_match_table = of_match_ptr(orion_wdt_of_match_table),
+		.of_match_table = orion_wdt_of_match_table,
 	},
 };
 
@@ -225,4 +225,3 @@ MODULE_PARM_DESC(nowayout, "Watchdog cannot be stopped once started (default="
 
 MODULE_LICENSE("GPL");
 MODULE_ALIAS("platform:orion_wdt");
-MODULE_ALIAS_MISCDEV(WATCHDOG_MINOR);
diff --git a/drivers/watchdog/pc87413_wdt.c b/drivers/watchdog/pc87413_wdt.c
index 5afb89b48650..5211d56b3681 100644
--- a/drivers/watchdog/pc87413_wdt.c
+++ b/drivers/watchdog/pc87413_wdt.c
@@ -580,8 +580,6 @@ MODULE_AUTHOR("Sven Anders <anders@anduras.de>, "
 MODULE_DESCRIPTION("PC87413 WDT driver");
 MODULE_LICENSE("GPL");
 
-MODULE_ALIAS_MISCDEV(WATCHDOG_MINOR);
-
 module_param(io, int, 0);
 MODULE_PARM_DESC(io, MODNAME " I/O port (default: "
 					__MODULE_STRING(IO_DEFAULT) ").");
diff --git a/drivers/watchdog/pcwd.c b/drivers/watchdog/pcwd.c
index 33e49a7f889f..e936f15dc7c7 100644
--- a/drivers/watchdog/pcwd.c
+++ b/drivers/watchdog/pcwd.c
@@ -61,7 +61,7 @@
 #include <linux/delay.h>	/* For mdelay function */
 #include <linux/timer.h>	/* For timer related operations */
 #include <linux/jiffies.h>	/* For jiffies stuff */
-#include <linux/miscdevice.h>	/* For MODULE_ALIAS_MISCDEV(WATCHDOG_MINOR) */
+#include <linux/miscdevice.h>	/* For struct miscdevice */
 #include <linux/watchdog.h>	/* For the watchdog specific items */
 #include <linux/reboot.h>	/* For kernel_power_off() */
 #include <linux/init.h>		/* For __init/__exit/... */
@@ -1011,5 +1011,3 @@ MODULE_AUTHOR("Ken Hollis <kenji@bitgate.com>, "
 MODULE_DESCRIPTION("Berkshire ISA-PC Watchdog driver");
 MODULE_VERSION(WATCHDOG_VERSION);
 MODULE_LICENSE("GPL");
-MODULE_ALIAS_MISCDEV(WATCHDOG_MINOR);
-MODULE_ALIAS_MISCDEV(TEMP_MINOR);
diff --git a/drivers/watchdog/pcwd_pci.c b/drivers/watchdog/pcwd_pci.c
index 7890f84edf76..b4864f254b48 100644
--- a/drivers/watchdog/pcwd_pci.c
+++ b/drivers/watchdog/pcwd_pci.c
@@ -40,7 +40,7 @@
 #include <linux/errno.h>	/* For the -ENODEV/... values */
 #include <linux/kernel.h>	/* For printk/panic/... */
 #include <linux/delay.h>	/* For mdelay function */
-#include <linux/miscdevice.h>	/* For MODULE_ALIAS_MISCDEV(WATCHDOG_MINOR) */
+#include <linux/miscdevice.h>	/* For struct miscdevice */
 #include <linux/watchdog.h>	/* For the watchdog specific items */
 #include <linux/notifier.h>	/* For notifier support */
 #include <linux/reboot.h>	/* For reboot_notifier stuff */
@@ -820,5 +820,3 @@ module_pci_driver(pcipcwd_driver);
 MODULE_AUTHOR("Wim Van Sebroeck <wim@iguana.be>");
 MODULE_DESCRIPTION("Berkshire PCI-PC Watchdog driver");
 MODULE_LICENSE("GPL");
-MODULE_ALIAS_MISCDEV(WATCHDOG_MINOR);
-MODULE_ALIAS_MISCDEV(TEMP_MINOR);
diff --git a/drivers/watchdog/pcwd_usb.c b/drivers/watchdog/pcwd_usb.c
index 7b14d1847927..b731b5d129be 100644
--- a/drivers/watchdog/pcwd_usb.c
+++ b/drivers/watchdog/pcwd_usb.c
@@ -32,7 +32,7 @@
 #include <linux/errno.h>	/* For the -ENODEV/... values */
 #include <linux/kernel.h>	/* For printk/panic/... */
 #include <linux/delay.h>	/* For mdelay function */
-#include <linux/miscdevice.h>	/* For MODULE_ALIAS_MISCDEV(WATCHDOG_MINOR) */
+#include <linux/miscdevice.h>	/* For struct miscdevice */
 #include <linux/watchdog.h>	/* For the watchdog specific items */
 #include <linux/notifier.h>	/* For notifier support */
 #include <linux/reboot.h>	/* For reboot_notifier stuff */
@@ -72,8 +72,6 @@ do {							\
 MODULE_AUTHOR(DRIVER_AUTHOR);
 MODULE_DESCRIPTION(DRIVER_DESC);
 MODULE_LICENSE(DRIVER_LICENSE);
-MODULE_ALIAS_MISCDEV(WATCHDOG_MINOR);
-MODULE_ALIAS_MISCDEV(TEMP_MINOR);
 
 /* Module Parameters */
 module_param(debug, int, 0);
@@ -235,13 +233,17 @@ static int usb_pcwd_send_command(struct usb_pcwd_private *usb_pcwd,
 		unsigned char cmd, unsigned char *msb, unsigned char *lsb)
 {
 	int got_response, count;
-	unsigned char buf[6];
+	unsigned char *buf;
 
 	/* We will not send any commands if the USB PCWD device does
 	 * not exist */
 	if ((!usb_pcwd) || (!usb_pcwd->exists))
 		return -1;
 
+	buf = kmalloc(6, GFP_KERNEL);
+	if (buf == NULL)
+		return 0;
+
 	/* The USB PC Watchdog uses a 6 byte report format.
 	 * The board currently uses only 3 of the six bytes of the report. */
 	buf[0] = cmd;			/* Byte 0 = CMD */
@@ -256,8 +258,8 @@ static int usb_pcwd_send_command(struct usb_pcwd_private *usb_pcwd,
 
 	if (usb_control_msg(usb_pcwd->udev, usb_sndctrlpipe(usb_pcwd->udev, 0),
 			HID_REQ_SET_REPORT, HID_DT_REPORT,
-			0x0200, usb_pcwd->interface_number, buf, sizeof(buf),
-			USB_COMMAND_TIMEOUT) != sizeof(buf)) {
+			0x0200, usb_pcwd->interface_number, buf, 6,
+			USB_COMMAND_TIMEOUT) != 6) {
 		dbg("usb_pcwd_send_command: error in usb_control_msg for "
 				"cmd 0x%x 0x%x 0x%x\n", cmd, *msb, *lsb);
 	}
@@ -277,6 +279,8 @@ static int usb_pcwd_send_command(struct usb_pcwd_private *usb_pcwd,
 		*lsb = usb_pcwd->cmd_data_lsb;
 	}
 
+	kfree(buf);
+
 	return got_response;
 }
 
diff --git a/drivers/watchdog/pika_wdt.c b/drivers/watchdog/pika_wdt.c
index 329bc60ad7a2..0cdfee266690 100644
--- a/drivers/watchdog/pika_wdt.c
+++ b/drivers/watchdog/pika_wdt.c
@@ -299,5 +299,3 @@ module_exit(pikawdt_exit);
 MODULE_AUTHOR("Sean MacLennan <smaclennan@pikatech.com>");
 MODULE_DESCRIPTION("PIKA FPGA based Watchdog Timer");
 MODULE_LICENSE("GPL");
-MODULE_ALIAS_MISCDEV(WATCHDOG_MINOR);
-
diff --git a/drivers/watchdog/pnx4008_wdt.c b/drivers/watchdog/pnx4008_wdt.c
index b30bd430f591..1bdcc313e1d9 100644
--- a/drivers/watchdog/pnx4008_wdt.c
+++ b/drivers/watchdog/pnx4008_wdt.c
@@ -233,5 +233,4 @@ MODULE_PARM_DESC(nowayout,
 		 "Set to 1 to keep watchdog running after device release");
 
 MODULE_LICENSE("GPL");
-MODULE_ALIAS_MISCDEV(WATCHDOG_MINOR);
 MODULE_ALIAS("platform:pnx4008-watchdog");
diff --git a/drivers/watchdog/pnx833x_wdt.c b/drivers/watchdog/pnx833x_wdt.c
index 1b62a7dfcc95..882fdcb46ad1 100644
--- a/drivers/watchdog/pnx833x_wdt.c
+++ b/drivers/watchdog/pnx833x_wdt.c
@@ -278,4 +278,3 @@ module_exit(watchdog_exit);
 MODULE_AUTHOR("Daniel Laird/Andre McCurdy");
 MODULE_DESCRIPTION("Hardware Watchdog Device for PNX833x");
 MODULE_LICENSE("GPL");
-MODULE_ALIAS_MISCDEV(WATCHDOG_MINOR);
diff --git a/drivers/watchdog/rc32434_wdt.c b/drivers/watchdog/rc32434_wdt.c
index 9cf6bc7a234f..71e78ef4b736 100644
--- a/drivers/watchdog/rc32434_wdt.c
+++ b/drivers/watchdog/rc32434_wdt.c
@@ -25,8 +25,7 @@
 #include <linux/errno.h>		/* For the -ENODEV/... values */
 #include <linux/kernel.h>		/* For printk/panic/... */
 #include <linux/fs.h>			/* For file operations */
-#include <linux/miscdevice.h>		/* For MODULE_ALIAS_MISCDEV
-							(WATCHDOG_MINOR) */
+#include <linux/miscdevice.h>		/* For struct miscdevice */
 #include <linux/watchdog.h>		/* For the watchdog specific items */
 #include <linux/init.h>			/* For __init/__exit/... */
 #include <linux/platform_device.h>	/* For platform_driver framework */
@@ -329,4 +328,3 @@ MODULE_AUTHOR("Ondrej Zajicek <santiago@crfreenet.org>,"
 		"Florian Fainelli <florian@openwrt.org>");
 MODULE_DESCRIPTION("Driver for the IDT RC32434 SoC watchdog");
 MODULE_LICENSE("GPL");
-MODULE_ALIAS_MISCDEV(WATCHDOG_MINOR);
diff --git a/drivers/watchdog/rdc321x_wdt.c b/drivers/watchdog/rdc321x_wdt.c
index b0f116c2fd53..082d06262959 100644
--- a/drivers/watchdog/rdc321x_wdt.c
+++ b/drivers/watchdog/rdc321x_wdt.c
@@ -231,7 +231,7 @@ static int rdc321x_wdt_probe(struct platform_device *pdev)
 	struct resource *r;
 	struct rdc321x_wdt_pdata *pdata;
 
-	pdata = pdev->dev.platform_data;
+	pdata = dev_get_platdata(&pdev->dev);
 	if (!pdata) {
 		dev_err(&pdev->dev, "no platform data supplied\n");
 		return -ENODEV;
@@ -298,4 +298,3 @@ module_platform_driver(rdc321x_wdt_driver);
 MODULE_AUTHOR("Florian Fainelli <florian@openwrt.org>");
 MODULE_DESCRIPTION("RDC321x watchdog driver");
 MODULE_LICENSE("GPL");
-MODULE_ALIAS_MISCDEV(WATCHDOG_MINOR);
diff --git a/drivers/watchdog/rt2880_wdt.c b/drivers/watchdog/rt2880_wdt.c
new file mode 100644
index 000000000000..53d37fea183e
--- /dev/null
+++ b/drivers/watchdog/rt2880_wdt.c
@@ -0,0 +1,207 @@
+/*
+ * Ralink RT288x/RT3xxx/MT76xx built-in hardware watchdog timer
+ *
+ * Copyright (C) 2011 Gabor Juhos <juhosg@openwrt.org>
+ * Copyright (C) 2013 John Crispin <blogic@openwrt.org>
+ *
+ * This driver was based on: drivers/watchdog/softdog.c
+ *
+ * This program is free software; you can redistribute it and/or modify it
+ * under the terms of the GNU General Public License version 2 as published
+ * by the Free Software Foundation.
+ */
+
+#include <linux/clk.h>
+#include <linux/reset.h>
+#include <linux/module.h>
+#include <linux/kernel.h>
+#include <linux/watchdog.h>
+#include <linux/miscdevice.h>
+#include <linux/moduleparam.h>
+#include <linux/platform_device.h>
+
+#include <asm/mach-ralink/ralink_regs.h>
+
+#define SYSC_RSTSTAT			0x38
+#define WDT_RST_CAUSE			BIT(1)
+
+#define RALINK_WDT_TIMEOUT		30
+#define RALINK_WDT_PRESCALE		65536
+
+#define TIMER_REG_TMR1LOAD		0x00
+#define TIMER_REG_TMR1CTL		0x08
+
+#define TMRSTAT_TMR1RST			BIT(5)
+
+#define TMR1CTL_ENABLE			BIT(7)
+#define TMR1CTL_MODE_SHIFT		4
+#define TMR1CTL_MODE_MASK		0x3
+#define TMR1CTL_MODE_FREE_RUNNING	0x0
+#define TMR1CTL_MODE_PERIODIC		0x1
+#define TMR1CTL_MODE_TIMEOUT		0x2
+#define TMR1CTL_MODE_WDT		0x3
+#define TMR1CTL_PRESCALE_MASK		0xf
+#define TMR1CTL_PRESCALE_65536		0xf
+
+static struct clk *rt288x_wdt_clk;
+static unsigned long rt288x_wdt_freq;
+static void __iomem *rt288x_wdt_base;
+
+static bool nowayout = WATCHDOG_NOWAYOUT;
+module_param(nowayout, bool, 0);
+MODULE_PARM_DESC(nowayout,
+		"Watchdog cannot be stopped once started (default="
+		__MODULE_STRING(WATCHDOG_NOWAYOUT) ")");
+
+static inline void rt_wdt_w32(unsigned reg, u32 val)
+{
+	iowrite32(val, rt288x_wdt_base + reg);
+}
+
+static inline u32 rt_wdt_r32(unsigned reg)
+{
+	return ioread32(rt288x_wdt_base + reg);
+}
+
+static int rt288x_wdt_ping(struct watchdog_device *w)
+{
+	rt_wdt_w32(TIMER_REG_TMR1LOAD, w->timeout * rt288x_wdt_freq);
+
+	return 0;
+}
+
+static int rt288x_wdt_start(struct watchdog_device *w)
+{
+	u32 t;
+
+	t = rt_wdt_r32(TIMER_REG_TMR1CTL);
+	t &= ~(TMR1CTL_MODE_MASK << TMR1CTL_MODE_SHIFT |
+		TMR1CTL_PRESCALE_MASK);
+	t |= (TMR1CTL_MODE_WDT << TMR1CTL_MODE_SHIFT |
+		TMR1CTL_PRESCALE_65536);
+	rt_wdt_w32(TIMER_REG_TMR1CTL, t);
+
+	rt288x_wdt_ping(w);
+
+	t = rt_wdt_r32(TIMER_REG_TMR1CTL);
+	t |= TMR1CTL_ENABLE;
+	rt_wdt_w32(TIMER_REG_TMR1CTL, t);
+
+	return 0;
+}
+
+static int rt288x_wdt_stop(struct watchdog_device *w)
+{
+	u32 t;
+
+	rt288x_wdt_ping(w);
+
+	t = rt_wdt_r32(TIMER_REG_TMR1CTL);
+	t &= ~TMR1CTL_ENABLE;
+	rt_wdt_w32(TIMER_REG_TMR1CTL, t);
+
+	return 0;
+}
+
+static int rt288x_wdt_set_timeout(struct watchdog_device *w, unsigned int t)
+{
+	w->timeout = t;
+	rt288x_wdt_ping(w);
+
+	return 0;
+}
+
+static int rt288x_wdt_bootcause(void)
+{
+	if (rt_sysc_r32(SYSC_RSTSTAT) & WDT_RST_CAUSE)
+		return WDIOF_CARDRESET;
+
+	return 0;
+}
+
+static struct watchdog_info rt288x_wdt_info = {
+	.identity = "Ralink Watchdog",
+	.options = WDIOF_SETTIMEOUT | WDIOF_KEEPALIVEPING | WDIOF_MAGICCLOSE,
+};
+
+static struct watchdog_ops rt288x_wdt_ops = {
+	.owner = THIS_MODULE,
+	.start = rt288x_wdt_start,
+	.stop = rt288x_wdt_stop,
+	.ping = rt288x_wdt_ping,
+	.set_timeout = rt288x_wdt_set_timeout,
+};
+
+static struct watchdog_device rt288x_wdt_dev = {
+	.info = &rt288x_wdt_info,
+	.ops = &rt288x_wdt_ops,
+	.min_timeout = 1,
+};
+
+static int rt288x_wdt_probe(struct platform_device *pdev)
+{
+	struct resource *res;
+	int ret;
+
+	res = platform_get_resource(pdev, IORESOURCE_MEM, 0);
+	rt288x_wdt_base = devm_ioremap_resource(&pdev->dev, res);
+	if (IS_ERR(rt288x_wdt_base))
+		return PTR_ERR(rt288x_wdt_base);
+
+	rt288x_wdt_clk = devm_clk_get(&pdev->dev, NULL);
+	if (IS_ERR(rt288x_wdt_clk))
+		return PTR_ERR(rt288x_wdt_clk);
+
+	device_reset(&pdev->dev);
+
+	rt288x_wdt_freq = clk_get_rate(rt288x_wdt_clk) / RALINK_WDT_PRESCALE;
+
+	rt288x_wdt_dev.dev = &pdev->dev;
+	rt288x_wdt_dev.bootstatus = rt288x_wdt_bootcause();
+
+	rt288x_wdt_dev.max_timeout = (0xfffful / rt288x_wdt_freq);
+	rt288x_wdt_dev.timeout = rt288x_wdt_dev.max_timeout;
+
+	watchdog_set_nowayout(&rt288x_wdt_dev, nowayout);
+
+	ret = watchdog_register_device(&rt288x_wdt_dev);
+	if (!ret)
+		dev_info(&pdev->dev, "Initialized\n");
+
+	return 0;
+}
+
+static int rt288x_wdt_remove(struct platform_device *pdev)
+{
+	watchdog_unregister_device(&rt288x_wdt_dev);
+
+	return 0;
+}
+
+static void rt288x_wdt_shutdown(struct platform_device *pdev)
+{
+	rt288x_wdt_stop(&rt288x_wdt_dev);
+}
+
+static const struct of_device_id rt288x_wdt_match[] = {
+	{ .compatible = "ralink,rt2880-wdt" },
+	{},
+};
+MODULE_DEVICE_TABLE(of, rt288x_wdt_match);
+
+static struct platform_driver rt288x_wdt_driver = {
+	.probe		= rt288x_wdt_probe,
+	.remove		= rt288x_wdt_remove,
+	.shutdown	= rt288x_wdt_shutdown,
+	.driver		= {
+		.name		= KBUILD_MODNAME,
+		.owner		= THIS_MODULE,
+		.of_match_table	= rt288x_wdt_match,
+	},
+};
+
+module_platform_driver(rt288x_wdt_driver);
+
+MODULE_DESCRIPTION("MediaTek/Ralink RT288x/RT3xxx hardware watchdog driver");
+MODULE_AUTHOR("Gabor Juhos <juhosg@openwrt.org");
+MODULE_LICENSE("GPL v2");
diff --git a/drivers/watchdog/s3c2410_wdt.c b/drivers/watchdog/s3c2410_wdt.c
index 23aad7c6bf5d..7d8fd041ee25 100644
--- a/drivers/watchdog/s3c2410_wdt.c
+++ b/drivers/watchdog/s3c2410_wdt.c
@@ -29,7 +29,6 @@
 #include <linux/moduleparam.h>
 #include <linux/types.h>
 #include <linux/timer.h>
-#include <linux/miscdevice.h> /* for MODULE_ALIAS_MISCDEV */
 #include <linux/watchdog.h>
 #include <linux/init.h>
 #include <linux/platform_device.h>
@@ -539,5 +538,4 @@ MODULE_AUTHOR("Ben Dooks <ben@simtec.co.uk>, "
 	      "Dimitry Andric <dimitry.andric@tomtom.com>");
 MODULE_DESCRIPTION("S3C2410 Watchdog Device Driver");
 MODULE_LICENSE("GPL");
-MODULE_ALIAS_MISCDEV(WATCHDOG_MINOR);
 MODULE_ALIAS("platform:s3c2410-wdt");
diff --git a/drivers/watchdog/sa1100_wdt.c b/drivers/watchdog/sa1100_wdt.c
index ccd6b29e21bf..e1d39a1e9628 100644
--- a/drivers/watchdog/sa1100_wdt.c
+++ b/drivers/watchdog/sa1100_wdt.c
@@ -193,4 +193,3 @@ module_param(margin, int, 0);
 MODULE_PARM_DESC(margin, "Watchdog margin in seconds (default 60s)");
 
 MODULE_LICENSE("GPL");
-MODULE_ALIAS_MISCDEV(WATCHDOG_MINOR);
diff --git a/drivers/watchdog/sb_wdog.c b/drivers/watchdog/sb_wdog.c
index ea5d84a1fdad..3abae50773b8 100644
--- a/drivers/watchdog/sb_wdog.c
+++ b/drivers/watchdog/sb_wdog.c
@@ -341,7 +341,6 @@ MODULE_PARM_DESC(timeout,
       "Watchdog timeout in microseconds (max/default 8388607 or 8.3ish secs)");
 
 MODULE_LICENSE("GPL");
-MODULE_ALIAS_MISCDEV(WATCHDOG_MINOR);
 
 /*
  * example code that can be put in a platform code area to utilize the
diff --git a/drivers/watchdog/sbc60xxwdt.c b/drivers/watchdog/sbc60xxwdt.c
index 63632ec87c7e..2eef58a0cf05 100644
--- a/drivers/watchdog/sbc60xxwdt.c
+++ b/drivers/watchdog/sbc60xxwdt.c
@@ -387,4 +387,3 @@ module_exit(sbc60xxwdt_unload);
 MODULE_AUTHOR("Jakob Oestergaard <jakob@unthought.net>");
 MODULE_DESCRIPTION("60xx Single Board Computer Watchdog Timer driver");
 MODULE_LICENSE("GPL");
-MODULE_ALIAS_MISCDEV(WATCHDOG_MINOR);
diff --git a/drivers/watchdog/sbc7240_wdt.c b/drivers/watchdog/sbc7240_wdt.c
index 719edc8fdeb3..5f268add17ce 100644
--- a/drivers/watchdog/sbc7240_wdt.c
+++ b/drivers/watchdog/sbc7240_wdt.c
@@ -309,5 +309,3 @@ MODULE_AUTHOR("Gilles Gigan");
 MODULE_DESCRIPTION("Watchdog device driver for single board"
 		   " computers EPIC Nano 7240 from iEi");
 MODULE_LICENSE("GPL");
-MODULE_ALIAS_MISCDEV(WATCHDOG_MINOR);
-
diff --git a/drivers/watchdog/sbc8360.c b/drivers/watchdog/sbc8360.c
index d4781e05f017..da60560ca446 100644
--- a/drivers/watchdog/sbc8360.c
+++ b/drivers/watchdog/sbc8360.c
@@ -404,6 +404,5 @@ MODULE_AUTHOR("Ian E. Morgan <imorgan@webcon.ca>");
 MODULE_DESCRIPTION("SBC8360 watchdog driver");
 MODULE_LICENSE("GPL");
 MODULE_VERSION("1.01");
-MODULE_ALIAS_MISCDEV(WATCHDOG_MINOR);
 
 /* end of sbc8360.c */
diff --git a/drivers/watchdog/sbc_epx_c3.c b/drivers/watchdog/sbc_epx_c3.c
index 0c3e9f66ef77..a1c502e0d8ec 100644
--- a/drivers/watchdog/sbc_epx_c3.c
+++ b/drivers/watchdog/sbc_epx_c3.c
@@ -220,4 +220,3 @@ MODULE_DESCRIPTION("Hardware Watchdog Device for Winsystems EPX-C3 SBC.  "
 	"so only use it if you are *sure* you are running on this specific "
 	"SBC system from Winsystems!  It writes to IO ports 0x1ee and 0x1ef!");
 MODULE_LICENSE("GPL");
-MODULE_ALIAS_MISCDEV(WATCHDOG_MINOR);
diff --git a/drivers/watchdog/sbc_fitpc2_wdt.c b/drivers/watchdog/sbc_fitpc2_wdt.c
index 90d5527ca886..a517d8bae757 100644
--- a/drivers/watchdog/sbc_fitpc2_wdt.c
+++ b/drivers/watchdog/sbc_fitpc2_wdt.c
@@ -263,5 +263,3 @@ module_param(nowayout, bool, 0);
 MODULE_PARM_DESC(nowayout, "Watchdog cannot be stopped once started");
 
 MODULE_LICENSE("GPL");
-MODULE_ALIAS_MISCDEV(WATCHDOG_MINOR);
-
diff --git a/drivers/watchdog/sc1200wdt.c b/drivers/watchdog/sc1200wdt.c
index 3fb83b0c28c2..3b9fff9dcf65 100644
--- a/drivers/watchdog/sc1200wdt.c
+++ b/drivers/watchdog/sc1200wdt.c
@@ -476,4 +476,3 @@ MODULE_AUTHOR("Zwane Mwaikambo <zwane@commfireservices.com>");
 MODULE_DESCRIPTION(
 	"Driver for National Semiconductor PC87307/PC97307 watchdog component");
 MODULE_LICENSE("GPL");
-MODULE_ALIAS_MISCDEV(WATCHDOG_MINOR);
diff --git a/drivers/watchdog/sc520_wdt.c b/drivers/watchdog/sc520_wdt.c
index 707e027e5002..f353e18b1a82 100644
--- a/drivers/watchdog/sc520_wdt.c
+++ b/drivers/watchdog/sc520_wdt.c
@@ -433,4 +433,3 @@ MODULE_AUTHOR("Scott and Bill Jennings");
 MODULE_DESCRIPTION(
 	"Driver for watchdog timer in AMD \"Elan\" SC520 uProcessor");
 MODULE_LICENSE("GPL");
-MODULE_ALIAS_MISCDEV(WATCHDOG_MINOR);
diff --git a/drivers/watchdog/sch311x_wdt.c b/drivers/watchdog/sch311x_wdt.c
index af7b136b1874..b96127ea3de1 100644
--- a/drivers/watchdog/sch311x_wdt.c
+++ b/drivers/watchdog/sch311x_wdt.c
@@ -26,8 +26,7 @@
 #include <linux/types.h>		/* For standard types (like size_t) */
 #include <linux/errno.h>		/* For the -ENODEV/... values */
 #include <linux/kernel.h>		/* For printk/... */
-#include <linux/miscdevice.h>		/* For MODULE_ALIAS_MISCDEV
-							(WATCHDOG_MINOR) */
+#include <linux/miscdevice.h>		/* For struct miscdevice */
 #include <linux/watchdog.h>		/* For the watchdog specific items */
 #include <linux/init.h>			/* For __init/__exit/... */
 #include <linux/fs.h>			/* For file operations */
@@ -545,5 +544,3 @@ module_exit(sch311x_wdt_exit);
 MODULE_AUTHOR("Wim Van Sebroeck <wim@iguana.be>");
 MODULE_DESCRIPTION("SMSC SCH311x WatchDog Timer Driver");
 MODULE_LICENSE("GPL");
-MODULE_ALIAS_MISCDEV(WATCHDOG_MINOR);
-
diff --git a/drivers/watchdog/scx200_wdt.c b/drivers/watchdog/scx200_wdt.c
index 8ae7c282d465..836377cf9271 100644
--- a/drivers/watchdog/scx200_wdt.c
+++ b/drivers/watchdog/scx200_wdt.c
@@ -37,7 +37,6 @@
 MODULE_AUTHOR("Christer Weinigel <wingel@nano-system.com>");
 MODULE_DESCRIPTION("NatSemi SCx200 Watchdog Driver");
 MODULE_LICENSE("GPL");
-MODULE_ALIAS_MISCDEV(WATCHDOG_MINOR);
 
 static int margin = 60;		/* in seconds */
 module_param(margin, int, 0);
diff --git a/drivers/watchdog/shwdt.c b/drivers/watchdog/shwdt.c
index 5bca79457768..f9b8e06f3558 100644
--- a/drivers/watchdog/shwdt.c
+++ b/drivers/watchdog/shwdt.c
@@ -343,7 +343,6 @@ MODULE_AUTHOR("Paul Mundt <lethal@linux-sh.org>");
 MODULE_DESCRIPTION("SuperH watchdog driver");
 MODULE_LICENSE("GPL");
 MODULE_ALIAS("platform:" DRV_NAME);
-MODULE_ALIAS_MISCDEV(WATCHDOG_MINOR);
 
 module_param(clock_division_ratio, int, 0);
 MODULE_PARM_DESC(clock_division_ratio,
diff --git a/drivers/watchdog/sirfsoc_wdt.c b/drivers/watchdog/sirfsoc_wdt.c
new file mode 100644
index 000000000000..ced3edc95957
--- /dev/null
+++ b/drivers/watchdog/sirfsoc_wdt.c
@@ -0,0 +1,226 @@
+/*
+ * Watchdog driver for CSR SiRFprimaII and SiRFatlasVI
+ *
+ * Copyright (c) 2013 Cambridge Silicon Radio Limited, a CSR plc group company.
+ *
+ * Licensed under GPLv2 or later.
+ */
+
+#include <linux/module.h>
+#include <linux/watchdog.h>
+#include <linux/platform_device.h>
+#include <linux/moduleparam.h>
+#include <linux/of.h>
+#include <linux/io.h>
+#include <linux/uaccess.h>
+
+#define CLOCK_FREQ	1000000
+
+#define SIRFSOC_TIMER_COUNTER_LO	0x0000
+#define SIRFSOC_TIMER_MATCH_0		0x0008
+#define SIRFSOC_TIMER_INT_EN		0x0024
+#define SIRFSOC_TIMER_WATCHDOG_EN	0x0028
+#define SIRFSOC_TIMER_LATCH		0x0030
+#define SIRFSOC_TIMER_LATCHED_LO	0x0034
+
+#define SIRFSOC_TIMER_WDT_INDEX		5
+
+#define SIRFSOC_WDT_MIN_TIMEOUT		30		/* 30 secs */
+#define SIRFSOC_WDT_MAX_TIMEOUT		(10 * 60)	/* 10 mins */
+#define SIRFSOC_WDT_DEFAULT_TIMEOUT	30		/* 30 secs */
+
+static unsigned int timeout = SIRFSOC_WDT_DEFAULT_TIMEOUT;
+static bool nowayout = WATCHDOG_NOWAYOUT;
+
+module_param(timeout, uint, 0);
+module_param(nowayout, bool, 0);
+
+MODULE_PARM_DESC(timeout, "Default watchdog timeout (in seconds)");
+MODULE_PARM_DESC(nowayout, "Watchdog cannot be stopped once started (default="
+			__MODULE_STRING(WATCHDOG_NOWAYOUT) ")");
+
+static unsigned int sirfsoc_wdt_gettimeleft(struct watchdog_device *wdd)
+{
+	u32 counter, match;
+	void __iomem *wdt_base;
+	int time_left;
+
+	wdt_base = watchdog_get_drvdata(wdd);
+	counter = readl(wdt_base + SIRFSOC_TIMER_COUNTER_LO);
+	match = readl(wdt_base +
+		SIRFSOC_TIMER_MATCH_0 + (SIRFSOC_TIMER_WDT_INDEX << 2));
+
+	time_left = match - counter;
+
+	return time_left / CLOCK_FREQ;
+}
+
+static int sirfsoc_wdt_updatetimeout(struct watchdog_device *wdd)
+{
+	u32 counter, timeout_ticks;
+	void __iomem *wdt_base;
+
+	timeout_ticks = wdd->timeout * CLOCK_FREQ;
+	wdt_base = watchdog_get_drvdata(wdd);
+
+	/* Enable the latch before reading the LATCH_LO register */
+	writel(1, wdt_base + SIRFSOC_TIMER_LATCH);
+
+	/* Set the TO value */
+	counter = readl(wdt_base + SIRFSOC_TIMER_LATCHED_LO);
+
+	counter += timeout_ticks;
+
+	writel(counter, wdt_base +
+		SIRFSOC_TIMER_MATCH_0 + (SIRFSOC_TIMER_WDT_INDEX << 2));
+
+	return 0;
+}
+
+static int sirfsoc_wdt_enable(struct watchdog_device *wdd)
+{
+	void __iomem *wdt_base = watchdog_get_drvdata(wdd);
+	sirfsoc_wdt_updatetimeout(wdd);
+
+	/*
+	 * NOTE: If interrupt is not enabled
+	 * then WD-Reset doesn't get generated at all.
+	 */
+	writel(readl(wdt_base + SIRFSOC_TIMER_INT_EN)
+		| (1 << SIRFSOC_TIMER_WDT_INDEX),
+		wdt_base + SIRFSOC_TIMER_INT_EN);
+	writel(1, wdt_base + SIRFSOC_TIMER_WATCHDOG_EN);
+
+	return 0;
+}
+
+static int sirfsoc_wdt_disable(struct watchdog_device *wdd)
+{
+	void __iomem *wdt_base = watchdog_get_drvdata(wdd);
+
+	writel(0, wdt_base + SIRFSOC_TIMER_WATCHDOG_EN);
+	writel(readl(wdt_base + SIRFSOC_TIMER_INT_EN)
+		& (~(1 << SIRFSOC_TIMER_WDT_INDEX)),
+		wdt_base + SIRFSOC_TIMER_INT_EN);
+
+	return 0;
+}
+
+static int sirfsoc_wdt_settimeout(struct watchdog_device *wdd, unsigned int to)
+{
+	wdd->timeout = to;
+	sirfsoc_wdt_updatetimeout(wdd);
+
+	return 0;
+}
+
+#define OPTIONS (WDIOF_SETTIMEOUT | WDIOF_KEEPALIVEPING | WDIOF_MAGICCLOSE)
+
+static const struct watchdog_info sirfsoc_wdt_ident = {
+	.options          =     OPTIONS,
+	.firmware_version =	0,
+	.identity         =	"SiRFSOC Watchdog",
+};
+
+static struct watchdog_ops sirfsoc_wdt_ops = {
+	.owner = THIS_MODULE,
+	.start = sirfsoc_wdt_enable,
+	.stop = sirfsoc_wdt_disable,
+	.get_timeleft = sirfsoc_wdt_gettimeleft,
+	.ping = sirfsoc_wdt_updatetimeout,
+	.set_timeout = sirfsoc_wdt_settimeout,
+};
+
+static struct watchdog_device sirfsoc_wdd = {
+	.info = &sirfsoc_wdt_ident,
+	.ops = &sirfsoc_wdt_ops,
+	.timeout = SIRFSOC_WDT_DEFAULT_TIMEOUT,
+	.min_timeout = SIRFSOC_WDT_MIN_TIMEOUT,
+	.max_timeout = SIRFSOC_WDT_MAX_TIMEOUT,
+};
+
+static int sirfsoc_wdt_probe(struct platform_device *pdev)
+{
+	struct resource *res;
+	int ret;
+	void __iomem *base;
+
+	res = platform_get_resource(pdev, IORESOURCE_MEM, 0);
+	base = devm_ioremap_resource(&pdev->dev, res);
+	if (IS_ERR(base))
+		return PTR_ERR(base);
+
+	watchdog_set_drvdata(&sirfsoc_wdd, base);
+
+	watchdog_init_timeout(&sirfsoc_wdd, timeout, &pdev->dev);
+	watchdog_set_nowayout(&sirfsoc_wdd, nowayout);
+
+	ret = watchdog_register_device(&sirfsoc_wdd);
+	if (ret)
+		return ret;
+
+	platform_set_drvdata(pdev, &sirfsoc_wdd);
+
+	return 0;
+}
+
+static void sirfsoc_wdt_shutdown(struct platform_device *pdev)
+{
+	struct watchdog_device *wdd = platform_get_drvdata(pdev);
+
+	sirfsoc_wdt_disable(wdd);
+}
+
+static int sirfsoc_wdt_remove(struct platform_device *pdev)
+{
+	sirfsoc_wdt_shutdown(pdev);
+	return 0;
+}
+
+#ifdef	CONFIG_PM_SLEEP
+static int sirfsoc_wdt_suspend(struct device *dev)
+{
+	return 0;
+}
+
+static int sirfsoc_wdt_resume(struct device *dev)
+{
+	struct watchdog_device *wdd = dev_get_drvdata(dev);
+
+	/*
+	 * NOTE: Since timer controller registers settings are saved
+	 * and restored back by the timer-prima2.c, so we need not
+	 * update WD settings except refreshing timeout.
+	 */
+	sirfsoc_wdt_updatetimeout(wdd);
+
+	return 0;
+}
+#endif
+
+static SIMPLE_DEV_PM_OPS(sirfsoc_wdt_pm_ops,
+		sirfsoc_wdt_suspend, sirfsoc_wdt_resume);
+
+static const struct of_device_id sirfsoc_wdt_of_match[] = {
+	{ .compatible = "sirf,prima2-tick"},
+	{},
+};
+MODULE_DEVICE_TABLE(of, sirfsoc_wdt_of_match);
+
+static struct platform_driver sirfsoc_wdt_driver = {
+	.driver = {
+		.name = "sirfsoc-wdt",
+		.owner = THIS_MODULE,
+		.pm = &sirfsoc_wdt_pm_ops,
+		.of_match_table	= of_match_ptr(sirfsoc_wdt_of_match),
+	},
+	.probe = sirfsoc_wdt_probe,
+	.remove = sirfsoc_wdt_remove,
+	.shutdown = sirfsoc_wdt_shutdown,
+};
+module_platform_driver(sirfsoc_wdt_driver);
+
+MODULE_DESCRIPTION("SiRF SoC watchdog driver");
+MODULE_AUTHOR("Xianglong Du <Xianglong.Du@csr.com>");
+MODULE_LICENSE("GPL v2");
+MODULE_ALIAS("platform:sirfsoc-wdt");
diff --git a/drivers/watchdog/smsc37b787_wdt.c b/drivers/watchdog/smsc37b787_wdt.c
index 6d665f9c1d58..445ea1ad1fa9 100644
--- a/drivers/watchdog/smsc37b787_wdt.c
+++ b/drivers/watchdog/smsc37b787_wdt.c
@@ -603,8 +603,6 @@ MODULE_DESCRIPTION("Driver for SMsC 37B787 watchdog component (Version "
 								VERSION ")");
 MODULE_LICENSE("GPL");
 
-MODULE_ALIAS_MISCDEV(WATCHDOG_MINOR);
-
 #ifdef SMSC_SUPPORT_MINUTES
 module_param(unit, int, 0);
 MODULE_PARM_DESC(unit,
diff --git a/drivers/watchdog/softdog.c b/drivers/watchdog/softdog.c
index b68b1e519d53..ef2638fee4a8 100644
--- a/drivers/watchdog/softdog.c
+++ b/drivers/watchdog/softdog.c
@@ -207,4 +207,3 @@ module_exit(watchdog_exit);
 MODULE_AUTHOR("Alan Cox");
 MODULE_DESCRIPTION("Software Watchdog Device Driver");
 MODULE_LICENSE("GPL");
-MODULE_ALIAS_MISCDEV(WATCHDOG_MINOR);
diff --git a/drivers/watchdog/sp5100_tco.c b/drivers/watchdog/sp5100_tco.c
index 0e9d8c479c35..ce63a1bbf395 100644
--- a/drivers/watchdog/sp5100_tco.c
+++ b/drivers/watchdog/sp5100_tco.c
@@ -580,4 +580,3 @@ module_exit(sp5100_tco_cleanup_module);
 MODULE_AUTHOR("Priyanka Gupta");
 MODULE_DESCRIPTION("TCO timer driver for SP5100/SB800 chipset");
 MODULE_LICENSE("GPL");
-MODULE_ALIAS_MISCDEV(WATCHDOG_MINOR);
diff --git a/drivers/watchdog/sp805_wdt.c b/drivers/watchdog/sp805_wdt.c
index 58df98aec122..3f786ce0a6f2 100644
--- a/drivers/watchdog/sp805_wdt.c
+++ b/drivers/watchdog/sp805_wdt.c
@@ -268,7 +268,6 @@ static int sp805_wdt_remove(struct amba_device *adev)
 	struct sp805_wdt *wdt = amba_get_drvdata(adev);
 
 	watchdog_unregister_device(&wdt->wdd);
-	amba_set_drvdata(adev, NULL);
 	watchdog_set_drvdata(&wdt->wdd, NULL);
 
 	return 0;
diff --git a/drivers/watchdog/stmp3xxx_rtc_wdt.c b/drivers/watchdog/stmp3xxx_rtc_wdt.c
index c97e98dcde62..d667f6b51d35 100644
--- a/drivers/watchdog/stmp3xxx_rtc_wdt.c
+++ b/drivers/watchdog/stmp3xxx_rtc_wdt.c
@@ -30,7 +30,7 @@ MODULE_PARM_DESC(heartbeat, "Watchdog heartbeat period in seconds from 1 to "
 static int wdt_start(struct watchdog_device *wdd)
 {
 	struct device *dev = watchdog_get_drvdata(wdd);
-	struct stmp3xxx_wdt_pdata *pdata = dev->platform_data;
+	struct stmp3xxx_wdt_pdata *pdata = dev_get_platdata(dev);
 
 	pdata->wdt_set_timeout(dev->parent, wdd->timeout * WDOG_TICK_RATE);
 	return 0;
@@ -39,7 +39,7 @@ static int wdt_start(struct watchdog_device *wdd)
 static int wdt_stop(struct watchdog_device *wdd)
 {
 	struct device *dev = watchdog_get_drvdata(wdd);
-	struct stmp3xxx_wdt_pdata *pdata = dev->platform_data;
+	struct stmp3xxx_wdt_pdata *pdata = dev_get_platdata(dev);
 
 	pdata->wdt_set_timeout(dev->parent, 0);
 	return 0;
@@ -108,4 +108,3 @@ module_platform_driver(stmp3xxx_wdt_driver);
 MODULE_DESCRIPTION("STMP3XXX RTC Watchdog Driver");
 MODULE_LICENSE("GPL v2");
 MODULE_AUTHOR("Wolfram Sang <w.sang@pengutronix.de>");
-MODULE_ALIAS_MISCDEV(WATCHDOG_MINOR);
diff --git a/drivers/watchdog/sunxi_wdt.c b/drivers/watchdog/sunxi_wdt.c
index f6caa77151c7..76332d893e12 100644
--- a/drivers/watchdog/sunxi_wdt.c
+++ b/drivers/watchdog/sunxi_wdt.c
@@ -217,7 +217,7 @@ static struct platform_driver sunxi_wdt_driver = {
 	.driver		= {
 		.owner		= THIS_MODULE,
 		.name		= DRV_NAME,
-		.of_match_table	= of_match_ptr(sunxi_wdt_dt_ids)
+		.of_match_table	= sunxi_wdt_dt_ids,
 	},
 };
 
diff --git a/drivers/watchdog/ts72xx_wdt.c b/drivers/watchdog/ts72xx_wdt.c
index c9b0c627fe7e..09d4831aa61f 100644
--- a/drivers/watchdog/ts72xx_wdt.c
+++ b/drivers/watchdog/ts72xx_wdt.c
@@ -192,7 +192,7 @@ static int ts72xx_wdt_open(struct inode *inode, struct file *file)
 		dev_err(&wdt->pdev->dev,
 			"failed to convert timeout (%d) to register value\n",
 			timeout);
-		return -EINVAL;
+		return regval;
 	}
 
 	if (mutex_lock_interruptible(&wdt->lock))
@@ -305,7 +305,8 @@ static long ts72xx_wdt_ioctl(struct file *file, unsigned int cmd,
 
 	switch (cmd) {
 	case WDIOC_GETSUPPORT:
-		error = copy_to_user(argp, &winfo, sizeof(winfo));
+		if (copy_to_user(argp, &winfo, sizeof(winfo)))
+			error = -EFAULT;
 		break;
 
 	case WDIOC_GETSTATUS:
@@ -320,10 +321,9 @@ static long ts72xx_wdt_ioctl(struct file *file, unsigned int cmd,
 	case WDIOC_SETOPTIONS: {
 		int options;
 
-		if (get_user(options, p)) {
-			error = -EFAULT;
+		error = get_user(options, p);
+		if (error)
 			break;
-		}
 
 		error = -EINVAL;
 
@@ -341,30 +341,26 @@ static long ts72xx_wdt_ioctl(struct file *file, unsigned int cmd,
 
 	case WDIOC_SETTIMEOUT: {
 		int new_timeout;
+		int regval;
 
-		if (get_user(new_timeout, p)) {
-			error = -EFAULT;
-		} else {
-			int regval;
-
-			regval = timeout_to_regval(new_timeout);
-			if (regval < 0) {
-				error = -EINVAL;
-			} else {
-				ts72xx_wdt_stop(wdt);
-				wdt->regval = regval;
-				ts72xx_wdt_start(wdt);
-			}
-		}
+		error = get_user(new_timeout, p);
 		if (error)
 			break;
 
+		regval = timeout_to_regval(new_timeout);
+		if (regval < 0) {
+			error = regval;
+			break;
+		}
+		ts72xx_wdt_stop(wdt);
+		wdt->regval = regval;
+		ts72xx_wdt_start(wdt);
+
 		/*FALLTHROUGH*/
 	}
 
 	case WDIOC_GETTIMEOUT:
-		if (put_user(regval_to_timeout(wdt->regval), p))
-			error = -EFAULT;
+		error = put_user(regval_to_timeout(wdt->regval), p);
 		break;
 
 	default:
diff --git a/drivers/watchdog/txx9wdt.c b/drivers/watchdog/txx9wdt.c
index 88f23c5cfddb..0fd0e8ae62a8 100644
--- a/drivers/watchdog/txx9wdt.c
+++ b/drivers/watchdog/txx9wdt.c
@@ -176,5 +176,4 @@ module_platform_driver_probe(txx9wdt_driver, txx9wdt_probe);
 
 MODULE_DESCRIPTION("TXx9 Watchdog Driver");
 MODULE_LICENSE("GPL");
-MODULE_ALIAS_MISCDEV(WATCHDOG_MINOR);
 MODULE_ALIAS("platform:txx9wdt");
diff --git a/drivers/watchdog/ux500_wdt.c b/drivers/watchdog/ux500_wdt.c
index a614d84121c3..e029b5768f2c 100644
--- a/drivers/watchdog/ux500_wdt.c
+++ b/drivers/watchdog/ux500_wdt.c
@@ -88,7 +88,7 @@ static struct watchdog_device ux500_wdt = {
 static int ux500_wdt_probe(struct platform_device *pdev)
 {
 	int ret;
-	struct ux500_wdt_data *pdata = pdev->dev.platform_data;
+	struct ux500_wdt_data *pdata = dev_get_platdata(&pdev->dev);
 
 	if (pdata) {
 		if (pdata->timeout > 0)
@@ -167,5 +167,4 @@ module_platform_driver(ux500_wdt_driver);
 MODULE_AUTHOR("Jonas Aaberg <jonas.aberg@stericsson.com>");
 MODULE_DESCRIPTION("Ux500 Watchdog Driver");
 MODULE_LICENSE("GPL");
-MODULE_ALIAS_MISCDEV(WATCHDOG_MINOR);
 MODULE_ALIAS("platform:ux500_wdt");
diff --git a/drivers/watchdog/w83627hf_wdt.c b/drivers/watchdog/w83627hf_wdt.c
index 92f1326f0cfc..e24b21082874 100644
--- a/drivers/watchdog/w83627hf_wdt.c
+++ b/drivers/watchdog/w83627hf_wdt.c
@@ -1,6 +1,9 @@
 /*
  *	w83627hf/thf WDT driver
  *
+ *	(c) Copyright 2013 Guenter Roeck
+ *		converted to watchdog infrastructure
+ *
  *	(c) Copyright 2007 Vlad Drukker <vlad@storewiz.com>
  *		added support for W83627THF.
  *
@@ -31,31 +34,22 @@
 #include <linux/module.h>
 #include <linux/moduleparam.h>
 #include <linux/types.h>
-#include <linux/miscdevice.h>
 #include <linux/watchdog.h>
-#include <linux/fs.h>
 #include <linux/ioport.h>
 #include <linux/notifier.h>
 #include <linux/reboot.h>
 #include <linux/init.h>
-#include <linux/spinlock.h>
 #include <linux/io.h>
-#include <linux/uaccess.h>
-
 
 #define WATCHDOG_NAME "w83627hf/thf/hg/dhg WDT"
 #define WATCHDOG_TIMEOUT 60		/* 60 sec default timeout */
 
-static unsigned long wdt_is_open;
-static char expect_close;
-static DEFINE_SPINLOCK(io_lock);
-
 /* You must set this - there is no sane way to probe for this board. */
 static int wdt_io = 0x2E;
 module_param(wdt_io, int, 0);
 MODULE_PARM_DESC(wdt_io, "w83627hf/thf WDT io port (default 0x2E)");
 
-static int timeout = WATCHDOG_TIMEOUT;	/* in seconds */
+static int timeout;			/* in seconds */
 module_param(timeout, int, 0);
 MODULE_PARM_DESC(timeout,
 		"Watchdog timeout in seconds. 1 <= timeout <= 255, default="
@@ -76,236 +70,147 @@ MODULE_PARM_DESC(nowayout,
 							(same as EFER) */
 #define WDT_EFDR (WDT_EFIR+1) /* Extended Function Data Register */
 
-static void w83627hf_select_wd_register(void)
+#define W83627HF_LD_WDT		0x08
+
+static void superio_outb(int reg, int val)
 {
-	unsigned char c;
+	outb(reg, WDT_EFER);
+	outb(val, WDT_EFDR);
+}
+
+static inline int superio_inb(int reg)
+{
+	outb(reg, WDT_EFER);
+	return inb(WDT_EFDR);
+}
+
+static int superio_enter(void)
+{
+	if (!request_muxed_region(wdt_io, 2, WATCHDOG_NAME))
+		return -EBUSY;
+
 	outb_p(0x87, WDT_EFER); /* Enter extended function mode */
 	outb_p(0x87, WDT_EFER); /* Again according to manual */
 
-	outb(0x20, WDT_EFER);	/* check chip version	*/
-	c = inb(WDT_EFDR);
-	if (c == 0x82) {	/* W83627THF		*/
-		outb_p(0x2b, WDT_EFER); /* select GPIO3 */
-		c = ((inb_p(WDT_EFDR) & 0xf7) | 0x04); /* select WDT0 */
-		outb_p(0x2b, WDT_EFER);
-		outb_p(c, WDT_EFDR);	/* set GPIO3 to WDT0 */
-	} else if (c == 0x88 || c == 0xa0) {	/* W83627EHF / W83627DHG */
-		outb_p(0x2d, WDT_EFER); /* select GPIO5 */
-		c = inb_p(WDT_EFDR) & ~0x01; /* PIN77 -> WDT0# */
-		outb_p(0x2d, WDT_EFER);
-		outb_p(c, WDT_EFDR); /* set GPIO5 to WDT0 */
-	}
+	return 0;
+}
 
-	outb_p(0x07, WDT_EFER); /* point to logical device number reg */
-	outb_p(0x08, WDT_EFDR); /* select logical device 8 (GPIO2) */
-	outb_p(0x30, WDT_EFER); /* select CR30 */
-	outb_p(0x01, WDT_EFDR); /* set bit 0 to activate GPIO2 */
+static void superio_select(int ld)
+{
+	superio_outb(0x07, ld);
 }
 
-static void w83627hf_unselect_wd_register(void)
+static void superio_exit(void)
 {
 	outb_p(0xAA, WDT_EFER); /* Leave extended function mode */
+	release_region(wdt_io, 2);
 }
 
 /* tyan motherboards seem to set F5 to 0x4C ?
  * So explicitly init to appropriate value. */
 
-static void w83627hf_init(void)
+static int w83627hf_init(struct watchdog_device *wdog)
 {
+	int ret;
 	unsigned char t;
 
-	w83627hf_select_wd_register();
+	ret = superio_enter();
+	if (ret)
+		return ret;
+
+	superio_select(W83627HF_LD_WDT);
+	t = superio_inb(0x20);	/* check chip version	*/
+	if (t == 0x82) {	/* W83627THF		*/
+		t = (superio_inb(0x2b) & 0xf7);
+		superio_outb(0x2b, t | 0x04); /* set GPIO3 to WDT0 */
+	} else if (t == 0x88 || t == 0xa0) {	/* W83627EHF / W83627DHG */
+		t = superio_inb(0x2d);
+		superio_outb(0x2d, t & ~0x01);	/* set GPIO5 to WDT0 */
+	}
+
+	/* set CR30 bit 0 to activate GPIO2 */
+	t = superio_inb(0x30);
+	if (!(t & 0x01))
+		superio_outb(0x30, t | 0x01);
 
-	outb_p(0xF6, WDT_EFER); /* Select CRF6 */
-	t = inb_p(WDT_EFDR);      /* read CRF6 */
+	t = superio_inb(0xF6);
 	if (t != 0) {
 		pr_info("Watchdog already running. Resetting timeout to %d sec\n",
-			timeout);
-		outb_p(timeout, WDT_EFDR);    /* Write back to CRF6 */
+			wdog->timeout);
+		superio_outb(0xF6, wdog->timeout);
 	}
 
-	outb_p(0xF5, WDT_EFER); /* Select CRF5 */
-	t = inb_p(WDT_EFDR);      /* read CRF5 */
-	t &= ~0x0C;               /* set second mode & disable keyboard
-				    turning off watchdog */
-	t |= 0x02;		  /* enable the WDTO# output low pulse
-				    to the KBRST# pin (PIN60) */
-	outb_p(t, WDT_EFDR);    /* Write back to CRF5 */
-
-	outb_p(0xF7, WDT_EFER); /* Select CRF7 */
-	t = inb_p(WDT_EFDR);      /* read CRF7 */
-	t &= ~0xC0;               /* disable keyboard & mouse turning off
-				    watchdog */
-	outb_p(t, WDT_EFDR);    /* Write back to CRF7 */
-
-	w83627hf_unselect_wd_register();
-}
-
-static void wdt_set_time(int timeout)
-{
-	spin_lock(&io_lock);
-
-	w83627hf_select_wd_register();
+	/* set second mode & disable keyboard turning off watchdog */
+	t = superio_inb(0xF5) & ~0x0C;
+	/* enable the WDTO# output low pulse to the KBRST# pin */
+	t |= 0x02;
+	superio_outb(0xF5, t);
 
-	outb_p(0xF6, WDT_EFER);    /* Select CRF6 */
-	outb_p(timeout, WDT_EFDR); /* Write Timeout counter to CRF6 */
+	/* disable keyboard & mouse turning off watchdog */
+	t = superio_inb(0xF7) & ~0xC0;
+	superio_outb(0xF7, t);
 
-	w83627hf_unselect_wd_register();
+	superio_exit();
 
-	spin_unlock(&io_lock);
-}
-
-static int wdt_ping(void)
-{
-	wdt_set_time(timeout);
 	return 0;
 }
 
-static int wdt_disable(void)
+static int wdt_set_time(unsigned int timeout)
 {
-	wdt_set_time(0);
-	return 0;
-}
+	int ret;
+
+	ret = superio_enter();
+	if (ret)
+		return ret;
+
+	superio_select(W83627HF_LD_WDT);
+	superio_outb(0xF6, timeout);
+	superio_exit();
 
-static int wdt_set_heartbeat(int t)
-{
-	if (t < 1 || t > 255)
-		return -EINVAL;
-	timeout = t;
 	return 0;
 }
 
-static int wdt_get_time(void)
+static int wdt_start(struct watchdog_device *wdog)
 {
-	int timeleft;
-
-	spin_lock(&io_lock);
-
-	w83627hf_select_wd_register();
-
-	outb_p(0xF6, WDT_EFER);    /* Select CRF6 */
-	timeleft = inb_p(WDT_EFDR); /* Read Timeout counter to CRF6 */
-
-	w83627hf_unselect_wd_register();
-
-	spin_unlock(&io_lock);
-
-	return timeleft;
+	return wdt_set_time(wdog->timeout);
 }
 
-static ssize_t wdt_write(struct file *file, const char __user *buf,
-						size_t count, loff_t *ppos)
+static int wdt_stop(struct watchdog_device *wdog)
 {
-	if (count) {
-		if (!nowayout) {
-			size_t i;
-
-			expect_close = 0;
-
-			for (i = 0; i != count; i++) {
-				char c;
-				if (get_user(c, buf + i))
-					return -EFAULT;
-				if (c == 'V')
-					expect_close = 42;
-			}
-		}
-		wdt_ping();
-	}
-	return count;
+	return wdt_set_time(0);
 }
 
-static long wdt_ioctl(struct file *file, unsigned int cmd, unsigned long arg)
+static int wdt_set_timeout(struct watchdog_device *wdog, unsigned int timeout)
 {
-	void __user *argp = (void __user *)arg;
-	int __user *p = argp;
-	int timeval;
-	static const struct watchdog_info ident = {
-		.options = WDIOF_KEEPALIVEPING | WDIOF_SETTIMEOUT |
-							WDIOF_MAGICCLOSE,
-		.firmware_version = 1,
-		.identity = "W83627HF WDT",
-	};
-
-	switch (cmd) {
-	case WDIOC_GETSUPPORT:
-		if (copy_to_user(argp, &ident, sizeof(ident)))
-			return -EFAULT;
-		break;
-	case WDIOC_GETSTATUS:
-	case WDIOC_GETBOOTSTATUS:
-		return put_user(0, p);
-	case WDIOC_SETOPTIONS:
-	{
-		int options, retval = -EINVAL;
-
-		if (get_user(options, p))
-			return -EFAULT;
-		if (options & WDIOS_DISABLECARD) {
-			wdt_disable();
-			retval = 0;
-		}
-		if (options & WDIOS_ENABLECARD) {
-			wdt_ping();
-			retval = 0;
-		}
-		return retval;
-	}
-	case WDIOC_KEEPALIVE:
-		wdt_ping();
-		break;
-	case WDIOC_SETTIMEOUT:
-		if (get_user(timeval, p))
-			return -EFAULT;
-		if (wdt_set_heartbeat(timeval))
-			return -EINVAL;
-		wdt_ping();
-		/* Fall */
-	case WDIOC_GETTIMEOUT:
-		return put_user(timeout, p);
-	case WDIOC_GETTIMELEFT:
-		timeval = wdt_get_time();
-		return put_user(timeval, p);
-	default:
-		return -ENOTTY;
-	}
+	wdog->timeout = timeout;
+
 	return 0;
 }
 
-static int wdt_open(struct inode *inode, struct file *file)
+static unsigned int wdt_get_time(struct watchdog_device *wdog)
 {
-	if (test_and_set_bit(0, &wdt_is_open))
-		return -EBUSY;
-	/*
-	 *	Activate
-	 */
+	unsigned int timeleft;
+	int ret;
 
-	wdt_ping();
-	return nonseekable_open(inode, file);
-}
+	ret = superio_enter();
+	if (ret)
+		return 0;
 
-static int wdt_close(struct inode *inode, struct file *file)
-{
-	if (expect_close == 42)
-		wdt_disable();
-	else {
-		pr_crit("Unexpected close, not stopping watchdog!\n");
-		wdt_ping();
-	}
-	expect_close = 0;
-	clear_bit(0, &wdt_is_open);
-	return 0;
+	superio_select(W83627HF_LD_WDT);
+	timeleft = superio_inb(0xF6);
+	superio_exit();
+
+	return timeleft;
 }
 
 /*
  *	Notifier for system down
  */
-
 static int wdt_notify_sys(struct notifier_block *this, unsigned long code,
 	void *unused)
 {
 	if (code == SYS_DOWN || code == SYS_HALT)
-		wdt_disable();	/* Turn the WDT off */
+		wdt_set_time(0);	/* Turn the WDT off */
 
 	return NOTIFY_DONE;
 }
@@ -314,19 +219,25 @@ static int wdt_notify_sys(struct notifier_block *this, unsigned long code,
  *	Kernel Interfaces
  */
 
-static const struct file_operations wdt_fops = {
-	.owner		= THIS_MODULE,
-	.llseek		= no_llseek,
-	.write		= wdt_write,
-	.unlocked_ioctl	= wdt_ioctl,
-	.open		= wdt_open,
-	.release	= wdt_close,
+static struct watchdog_info wdt_info = {
+	.options = WDIOF_SETTIMEOUT | WDIOF_KEEPALIVEPING | WDIOF_MAGICCLOSE,
+	.identity = "W83627HF Watchdog",
 };
 
-static struct miscdevice wdt_miscdev = {
-	.minor = WATCHDOG_MINOR,
-	.name = "watchdog",
-	.fops = &wdt_fops,
+static struct watchdog_ops wdt_ops = {
+	.owner = THIS_MODULE,
+	.start = wdt_start,
+	.stop = wdt_stop,
+	.set_timeout = wdt_set_timeout,
+	.get_timeleft = wdt_get_time,
+};
+
+static struct watchdog_device wdt_dev = {
+	.info = &wdt_info,
+	.ops = &wdt_ops,
+	.timeout = WATCHDOG_TIMEOUT,
+	.min_timeout = 1,
+	.max_timeout = 255,
 };
 
 /*
@@ -344,50 +255,39 @@ static int __init wdt_init(void)
 
 	pr_info("WDT driver for the Winbond(TM) W83627HF/THF/HG/DHG Super I/O chip initialising\n");
 
-	if (wdt_set_heartbeat(timeout)) {
-		wdt_set_heartbeat(WATCHDOG_TIMEOUT);
-		pr_info("timeout value must be 1 <= timeout <= 255, using %d\n",
-			WATCHDOG_TIMEOUT);
-	}
+	watchdog_init_timeout(&wdt_dev, timeout, NULL);
+	watchdog_set_nowayout(&wdt_dev, nowayout);
 
-	if (!request_region(wdt_io, 1, WATCHDOG_NAME)) {
-		pr_err("I/O address 0x%04x already in use\n", wdt_io);
-		ret = -EIO;
-		goto out;
+	ret = w83627hf_init(&wdt_dev);
+	if (ret) {
+		pr_err("failed to initialize watchdog (err=%d)\n", ret);
+		return ret;
 	}
 
-	w83627hf_init();
-
 	ret = register_reboot_notifier(&wdt_notifier);
 	if (ret != 0) {
 		pr_err("cannot register reboot notifier (err=%d)\n", ret);
-		goto unreg_regions;
+		return ret;
 	}
 
-	ret = misc_register(&wdt_miscdev);
-	if (ret != 0) {
-		pr_err("cannot register miscdev on minor=%d (err=%d)\n",
-		       WATCHDOG_MINOR, ret);
+	ret = watchdog_register_device(&wdt_dev);
+	if (ret)
 		goto unreg_reboot;
-	}
 
 	pr_info("initialized. timeout=%d sec (nowayout=%d)\n",
-		timeout, nowayout);
+		wdt_dev.timeout, nowayout);
 
-out:
 	return ret;
+
 unreg_reboot:
 	unregister_reboot_notifier(&wdt_notifier);
-unreg_regions:
-	release_region(wdt_io, 1);
-	goto out;
+	return ret;
 }
 
 static void __exit wdt_exit(void)
 {
-	misc_deregister(&wdt_miscdev);
+	watchdog_unregister_device(&wdt_dev);
 	unregister_reboot_notifier(&wdt_notifier);
-	release_region(wdt_io, 1);
 }
 
 module_init(wdt_init);
@@ -396,4 +296,3 @@ module_exit(wdt_exit);
 MODULE_LICENSE("GPL");
 MODULE_AUTHOR("Pádraig  Brady <P@draigBrady.com>");
 MODULE_DESCRIPTION("w83627hf/thf WDT driver");
-MODULE_ALIAS_MISCDEV(WATCHDOG_MINOR);
diff --git a/drivers/watchdog/w83697hf_wdt.c b/drivers/watchdog/w83697hf_wdt.c
index cd9f3c1e1af4..aaf2995d37f4 100644
--- a/drivers/watchdog/w83697hf_wdt.c
+++ b/drivers/watchdog/w83697hf_wdt.c
@@ -458,4 +458,3 @@ MODULE_LICENSE("GPL");
 MODULE_AUTHOR("Marcus Junker <junker@anduras.de>, "
 		"Samuel Tardieu <sam@rfc1149.net>");
 MODULE_DESCRIPTION("w83697hf/hg WDT driver");
-MODULE_ALIAS_MISCDEV(WATCHDOG_MINOR);
diff --git a/drivers/watchdog/w83697ug_wdt.c b/drivers/watchdog/w83697ug_wdt.c
index 274be0bfaf24..ff58cb74671f 100644
--- a/drivers/watchdog/w83697ug_wdt.c
+++ b/drivers/watchdog/w83697ug_wdt.c
@@ -395,4 +395,3 @@ module_exit(wdt_exit);
 MODULE_LICENSE("GPL");
 MODULE_AUTHOR("Flemming Frandsen <ff@nrvissing.net>");
 MODULE_DESCRIPTION("w83697ug/uf WDT driver");
-MODULE_ALIAS_MISCDEV(WATCHDOG_MINOR);
diff --git a/drivers/watchdog/w83877f_wdt.c b/drivers/watchdog/w83877f_wdt.c
index 7874ae06232b..f0483c75ed32 100644
--- a/drivers/watchdog/w83877f_wdt.c
+++ b/drivers/watchdog/w83877f_wdt.c
@@ -406,4 +406,3 @@ module_exit(w83877f_wdt_unload);
 MODULE_AUTHOR("Scott and Bill Jennings");
 MODULE_DESCRIPTION("Driver for watchdog timer in w83877f chip");
 MODULE_LICENSE("GPL");
-MODULE_ALIAS_MISCDEV(WATCHDOG_MINOR);
diff --git a/drivers/watchdog/w83977f_wdt.c b/drivers/watchdog/w83977f_wdt.c
index 5d2c902825c2..91bf55a20024 100644
--- a/drivers/watchdog/w83977f_wdt.c
+++ b/drivers/watchdog/w83977f_wdt.c
@@ -527,4 +527,3 @@ module_exit(w83977f_wdt_exit);
 MODULE_AUTHOR("Jose Goncalves <jose.goncalves@inov.pt>");
 MODULE_DESCRIPTION("Driver for watchdog timer in W83977F I/O chip");
 MODULE_LICENSE("GPL");
-MODULE_ALIAS_MISCDEV(WATCHDOG_MINOR);
diff --git a/drivers/watchdog/wafer5823wdt.c b/drivers/watchdog/wafer5823wdt.c
index 25aba6e00a23..db0da7ea4fd8 100644
--- a/drivers/watchdog/wafer5823wdt.c
+++ b/drivers/watchdog/wafer5823wdt.c
@@ -322,6 +322,5 @@ module_exit(wafwdt_exit);
 MODULE_AUTHOR("Justin Cormack");
 MODULE_DESCRIPTION("ICP Wafer 5823 Single Board Computer WDT driver");
 MODULE_LICENSE("GPL");
-MODULE_ALIAS_MISCDEV(WATCHDOG_MINOR);
 
 /* end of wafer5823wdt.c */
diff --git a/drivers/watchdog/watchdog_core.c b/drivers/watchdog/watchdog_core.c
index 05d18b4c661b..461336c4519f 100644
--- a/drivers/watchdog/watchdog_core.c
+++ b/drivers/watchdog/watchdog_core.c
@@ -77,7 +77,7 @@ int watchdog_init_timeout(struct watchdog_device *wdd,
 
 	watchdog_check_min_max_timeout(wdd);
 
-	/* try to get the tiemout module parameter first */
+	/* try to get the timeout module parameter first */
 	if (!watchdog_timeout_invalid(wdd, timeout_parm)) {
 		wdd->timeout = timeout_parm;
 		return ret;
diff --git a/drivers/watchdog/wdrtas.c b/drivers/watchdog/wdrtas.c
index 3045debd5411..0240c60d14e3 100644
--- a/drivers/watchdog/wdrtas.c
+++ b/drivers/watchdog/wdrtas.c
@@ -48,8 +48,6 @@
 MODULE_AUTHOR("Utz Bacher <utz.bacher@de.ibm.com>");
 MODULE_DESCRIPTION("RTAS watchdog driver");
 MODULE_LICENSE("GPL");
-MODULE_ALIAS_MISCDEV(WATCHDOG_MINOR);
-MODULE_ALIAS_MISCDEV(TEMP_MINOR);
 
 static bool wdrtas_nowayout = WATCHDOG_NOWAYOUT;
 static atomic_t wdrtas_miscdev_open = ATOMIC_INIT(0);
diff --git a/drivers/watchdog/wdt.c b/drivers/watchdog/wdt.c
index ee4333c01109..e0206b5b7d89 100644
--- a/drivers/watchdog/wdt.c
+++ b/drivers/watchdog/wdt.c
@@ -664,6 +664,4 @@ module_exit(wdt_exit);
 
 MODULE_AUTHOR("Alan Cox");
 MODULE_DESCRIPTION("Driver for ISA ICS watchdog cards (WDT500/501)");
-MODULE_ALIAS_MISCDEV(WATCHDOG_MINOR);
-MODULE_ALIAS_MISCDEV(TEMP_MINOR);
 MODULE_LICENSE("GPL");
diff --git a/drivers/watchdog/wdt285.c b/drivers/watchdog/wdt285.c
index 5eec74053882..7355ddd0b207 100644
--- a/drivers/watchdog/wdt285.c
+++ b/drivers/watchdog/wdt285.c
@@ -224,7 +224,6 @@ static void __exit footbridge_watchdog_exit(void)
 MODULE_AUTHOR("Phil Blundell <pb@nexus.co.uk>");
 MODULE_DESCRIPTION("Footbridge watchdog driver");
 MODULE_LICENSE("GPL");
-MODULE_ALIAS_MISCDEV(WATCHDOG_MINOR);
 
 module_param(soft_margin, int, 0);
 MODULE_PARM_DESC(soft_margin, "Watchdog timeout in seconds");
diff --git a/drivers/watchdog/wdt977.c b/drivers/watchdog/wdt977.c
index 65a402344933..a8e6f87f60c9 100644
--- a/drivers/watchdog/wdt977.c
+++ b/drivers/watchdog/wdt977.c
@@ -507,4 +507,3 @@ module_exit(wd977_exit);
 MODULE_AUTHOR("Woody Suwalski <woodys@xandros.com>");
 MODULE_DESCRIPTION("W83977AF Watchdog driver");
 MODULE_LICENSE("GPL");
-MODULE_ALIAS_MISCDEV(WATCHDOG_MINOR);
diff --git a/drivers/watchdog/wdt_pci.c b/drivers/watchdog/wdt_pci.c
index 36a54c0e32dd..ee89ba4dea63 100644
--- a/drivers/watchdog/wdt_pci.c
+++ b/drivers/watchdog/wdt_pci.c
@@ -744,5 +744,3 @@ module_pci_driver(wdtpci_driver);
 MODULE_AUTHOR("JP Nollmann, Alan Cox");
 MODULE_DESCRIPTION("Driver for the ICS PCI-WDT500/501 watchdog cards");
 MODULE_LICENSE("GPL");
-MODULE_ALIAS_MISCDEV(WATCHDOG_MINOR);
-MODULE_ALIAS_MISCDEV(TEMP_MINOR);
diff --git a/drivers/watchdog/wm831x_wdt.c b/drivers/watchdog/wm831x_wdt.c
index d4e47eda4182..e243bd01c774 100644
--- a/drivers/watchdog/wm831x_wdt.c
+++ b/drivers/watchdog/wm831x_wdt.c
@@ -184,7 +184,7 @@ static const struct watchdog_ops wm831x_wdt_ops = {
 static int wm831x_wdt_probe(struct platform_device *pdev)
 {
 	struct wm831x *wm831x = dev_get_drvdata(pdev->dev.parent);
-	struct wm831x_pdata *chip_pdata;
+	struct wm831x_pdata *chip_pdata = dev_get_platdata(pdev->dev.parent);
 	struct wm831x_watchdog_pdata *pdata;
 	struct wm831x_wdt_drvdata *driver_data;
 	struct watchdog_device *wm831x_wdt;
@@ -231,12 +231,10 @@ static int wm831x_wdt_probe(struct platform_device *pdev)
 		wm831x_wdt->timeout = wm831x_wdt_cfgs[i].time;
 
 	/* Apply any configuration */
-	if (pdev->dev.parent->platform_data) {
-		chip_pdata = pdev->dev.parent->platform_data;
+	if (chip_pdata)
 		pdata = chip_pdata->watchdog;
-	} else {
+	else
 		pdata = NULL;
-	}
 
 	if (pdata) {
 		reg &= ~(WM831X_WDOG_SECACT_MASK | WM831X_WDOG_PRIMACT_MASK |
diff --git a/drivers/watchdog/xen_wdt.c b/drivers/watchdog/xen_wdt.c
index 92ad33d0cb71..7a42dffd39e5 100644
--- a/drivers/watchdog/xen_wdt.c
+++ b/drivers/watchdog/xen_wdt.c
@@ -362,4 +362,3 @@ MODULE_AUTHOR("Jan Beulich <jbeulich@novell.com>");
 MODULE_DESCRIPTION("Xen WatchDog Timer Driver");
 MODULE_VERSION(DRV_VERSION);
 MODULE_LICENSE("GPL");
-MODULE_ALIAS_MISCDEV(WATCHDOG_MINOR);
diff --git a/drivers/xen/Kconfig b/drivers/xen/Kconfig
index 23eae5cb69c2..c794ea182140 100644
--- a/drivers/xen/Kconfig
+++ b/drivers/xen/Kconfig
@@ -140,7 +140,6 @@ config XEN_GRANT_DEV_ALLOC
 
 config SWIOTLB_XEN
 	def_bool y
-	depends on PCI && X86
 	select SWIOTLB
 
 config XEN_TMEM
diff --git a/drivers/xen/balloon.c b/drivers/xen/balloon.c
index b232908a6192..55ea73f7c70b 100644
--- a/drivers/xen/balloon.c
+++ b/drivers/xen/balloon.c
@@ -596,7 +596,7 @@ static void __init balloon_add_region(unsigned long start_pfn,
 	}
 }
 
-static int __cpuinit balloon_cpu_notify(struct notifier_block *self,
+static int balloon_cpu_notify(struct notifier_block *self,
 				    unsigned long action, void *hcpu)
 {
 	int cpu = (long)hcpu;
@@ -616,7 +616,7 @@ static int __cpuinit balloon_cpu_notify(struct notifier_block *self,
 	return NOTIFY_OK;
 }
 
-static struct notifier_block balloon_cpu_notifier __cpuinitdata = {
+static struct notifier_block balloon_cpu_notifier = {
 	.notifier_call	= balloon_cpu_notify,
 };
 
@@ -641,7 +641,7 @@ static int __init balloon_init(void)
 
 	balloon_stats.current_pages = xen_pv_domain()
 		? min(xen_start_info->nr_pages - xen_released_pages, max_pfn)
-		: max_pfn;
+		: get_num_physpages();
 	balloon_stats.target_pages  = balloon_stats.current_pages;
 	balloon_stats.balloon_low   = 0;
 	balloon_stats.balloon_high  = 0;
diff --git a/drivers/xen/evtchn.c b/drivers/xen/evtchn.c
index 8b3a69a06c39..5de2063e16d3 100644
--- a/drivers/xen/evtchn.c
+++ b/drivers/xen/evtchn.c
@@ -305,7 +305,7 @@ static int evtchn_bind_to_user(struct per_user_data *u, int port)
 	if (rc < 0)
 		goto err;
 
-	rc = bind_evtchn_to_irqhandler(port, evtchn_interrupt, IRQF_DISABLED,
+	rc = bind_evtchn_to_irqhandler(port, evtchn_interrupt, 0,
 				       u->name, evtchn);
 	if (rc < 0)
 		goto err;
diff --git a/drivers/xen/grant-table.c b/drivers/xen/grant-table.c
index c4d2298893b1..62ccf5424ba8 100644
--- a/drivers/xen/grant-table.c
+++ b/drivers/xen/grant-table.c
@@ -49,6 +49,7 @@
 #include <xen/grant_table.h>
 #include <xen/interface/memory.h>
 #include <xen/hvc-console.h>
+#include <xen/swiotlb-xen.h>
 #include <asm/xen/hypercall.h>
 #include <asm/xen/interface.h>
 
@@ -898,8 +899,16 @@ int gnttab_map_refs(struct gnttab_map_grant_ref *map_ops,
 			gnttab_retry_eagain_gop(GNTTABOP_map_grant_ref, map_ops + i,
 						&map_ops[i].status, __func__);
 
-	if (xen_feature(XENFEAT_auto_translated_physmap))
+	/* this is basically a nop on x86 */
+	if (xen_feature(XENFEAT_auto_translated_physmap)) {
+		for (i = 0; i < count; i++) {
+			if (map_ops[i].status)
+				continue;
+			set_phys_to_machine(map_ops[i].host_addr >> PAGE_SHIFT,
+					map_ops[i].dev_bus_addr >> PAGE_SHIFT);
+		}
 		return ret;
+	}
 
 	if (!in_interrupt() && paravirt_get_lazy_mode() == PARAVIRT_LAZY_NONE) {
 		arch_enter_lazy_mmu_mode();
@@ -942,8 +951,14 @@ int gnttab_unmap_refs(struct gnttab_unmap_grant_ref *unmap_ops,
 	if (ret)
 		return ret;
 
-	if (xen_feature(XENFEAT_auto_translated_physmap))
+	/* this is basically a nop on x86 */
+	if (xen_feature(XENFEAT_auto_translated_physmap)) {
+		for (i = 0; i < count; i++) {
+			set_phys_to_machine(unmap_ops[i].host_addr >> PAGE_SHIFT,
+					INVALID_P2M_ENTRY);
+		}
 		return ret;
+	}
 
 	if (!in_interrupt() && paravirt_get_lazy_mode() == PARAVIRT_LAZY_NONE) {
 		arch_enter_lazy_mmu_mode();
diff --git a/drivers/xen/pci.c b/drivers/xen/pci.c
index cc86b267056f..188825122aae 100644
--- a/drivers/xen/pci.c
+++ b/drivers/xen/pci.c
@@ -26,6 +26,7 @@
 #include <asm/xen/hypervisor.h>
 #include <asm/xen/hypercall.h>
 #include "../pci/pci.h"
+#include <asm/pci_x86.h>
 
 static bool __read_mostly pci_seg_supported = true;
 
@@ -192,3 +193,49 @@ static int __init register_xen_pci_notifier(void)
 }
 
 arch_initcall(register_xen_pci_notifier);
+
+#ifdef CONFIG_PCI_MMCONFIG
+static int __init xen_mcfg_late(void)
+{
+	struct pci_mmcfg_region *cfg;
+	int rc;
+
+	if (!xen_initial_domain())
+		return 0;
+
+	if ((pci_probe & PCI_PROBE_MMCONF) == 0)
+		return 0;
+
+	if (list_empty(&pci_mmcfg_list))
+		return 0;
+
+	/* Check whether they are in the right area. */
+	list_for_each_entry(cfg, &pci_mmcfg_list, list) {
+		struct physdev_pci_mmcfg_reserved r;
+
+		r.address = cfg->address;
+		r.segment = cfg->segment;
+		r.start_bus = cfg->start_bus;
+		r.end_bus = cfg->end_bus;
+		r.flags = XEN_PCI_MMCFG_RESERVED;
+
+		rc = HYPERVISOR_physdev_op(PHYSDEVOP_pci_mmcfg_reserved, &r);
+		switch (rc) {
+		case 0:
+		case -ENOSYS:
+			continue;
+
+		default:
+			pr_warn("Failed to report MMCONFIG reservation"
+				" state for %s to hypervisor"
+				" (%d)\n",
+				cfg->name, rc);
+		}
+	}
+	return 0;
+}
+/*
+ * Needs to be done after acpi_init which are subsys_initcall.
+ */
+subsys_initcall_sync(xen_mcfg_late);
+#endif
diff --git a/drivers/xen/platform-pci.c b/drivers/xen/platform-pci.c
index 99db9e1eb8ba..2f3528e93cb9 100644
--- a/drivers/xen/platform-pci.c
+++ b/drivers/xen/platform-pci.c
@@ -84,7 +84,7 @@ static irqreturn_t do_hvm_evtchn_intr(int irq, void *dev_id)
 static int xen_allocate_irq(struct pci_dev *pdev)
 {
 	return request_irq(pdev->irq, do_hvm_evtchn_intr,
-			IRQF_DISABLED | IRQF_NOBALANCING | IRQF_TRIGGER_RISING,
+			IRQF_NOBALANCING | IRQF_TRIGGER_RISING,
 			"xen-platform-pci", pdev);
 }
 
diff --git a/drivers/xen/swiotlb-xen.c b/drivers/xen/swiotlb-xen.c
index 1b2277c311d2..a224bc74b6b9 100644
--- a/drivers/xen/swiotlb-xen.c
+++ b/drivers/xen/swiotlb-xen.c
@@ -42,12 +42,31 @@
 #include <xen/page.h>
 #include <xen/xen-ops.h>
 #include <xen/hvc-console.h>
+
+#include <asm/dma-mapping.h>
+#include <asm/xen/page-coherent.h>
+
+#include <trace/events/swiotlb.h>
 /*
  * Used to do a quick range check in swiotlb_tbl_unmap_single and
  * swiotlb_tbl_sync_single_*, to see if the memory was in fact allocated by this
  * API.
  */
 
+#ifndef CONFIG_X86
+static unsigned long dma_alloc_coherent_mask(struct device *dev,
+					    gfp_t gfp)
+{
+	unsigned long dma_mask = 0;
+
+	dma_mask = dev->coherent_dma_mask;
+	if (!dma_mask)
+		dma_mask = (gfp & GFP_DMA) ? DMA_BIT_MASK(24) : DMA_BIT_MASK(32);
+
+	return dma_mask;
+}
+#endif
+
 static char *xen_io_tlb_start, *xen_io_tlb_end;
 static unsigned long xen_io_tlb_nslabs;
 /*
@@ -56,17 +75,17 @@ static unsigned long xen_io_tlb_nslabs;
 
 static u64 start_dma_addr;
 
-static dma_addr_t xen_phys_to_bus(phys_addr_t paddr)
+static inline dma_addr_t xen_phys_to_bus(phys_addr_t paddr)
 {
 	return phys_to_machine(XPADDR(paddr)).maddr;
 }
 
-static phys_addr_t xen_bus_to_phys(dma_addr_t baddr)
+static inline phys_addr_t xen_bus_to_phys(dma_addr_t baddr)
 {
 	return machine_to_phys(XMADDR(baddr)).paddr;
 }
 
-static dma_addr_t xen_virt_to_bus(void *address)
+static inline dma_addr_t xen_virt_to_bus(void *address)
 {
 	return xen_phys_to_bus(virt_to_phys(address));
 }
@@ -89,7 +108,7 @@ static int check_pages_physically_contiguous(unsigned long pfn,
 	return 1;
 }
 
-static int range_straddles_page_boundary(phys_addr_t p, size_t size)
+static inline int range_straddles_page_boundary(phys_addr_t p, size_t size)
 {
 	unsigned long pfn = PFN_DOWN(p);
 	unsigned int offset = p & ~PAGE_MASK;
@@ -126,6 +145,8 @@ xen_swiotlb_fixup(void *buf, size_t size, unsigned long nslabs)
 {
 	int i, rc;
 	int dma_bits;
+	dma_addr_t dma_handle;
+	phys_addr_t p = virt_to_phys(buf);
 
 	dma_bits = get_order(IO_TLB_SEGSIZE << IO_TLB_SHIFT) + PAGE_SHIFT;
 
@@ -135,9 +156,9 @@ xen_swiotlb_fixup(void *buf, size_t size, unsigned long nslabs)
 
 		do {
 			rc = xen_create_contiguous_region(
-				(unsigned long)buf + (i << IO_TLB_SHIFT),
+				p + (i << IO_TLB_SHIFT),
 				get_order(slabs << IO_TLB_SHIFT),
-				dma_bits);
+				dma_bits, &dma_handle);
 		} while (rc && dma_bits++ < max_dma_bits);
 		if (rc)
 			return rc;
@@ -263,7 +284,6 @@ xen_swiotlb_alloc_coherent(struct device *hwdev, size_t size,
 	void *ret;
 	int order = get_order(size);
 	u64 dma_mask = DMA_BIT_MASK(32);
-	unsigned long vstart;
 	phys_addr_t phys;
 	dma_addr_t dev_addr;
 
@@ -278,8 +298,12 @@ xen_swiotlb_alloc_coherent(struct device *hwdev, size_t size,
 	if (dma_alloc_from_coherent(hwdev, size, dma_handle, &ret))
 		return ret;
 
-	vstart = __get_free_pages(flags, order);
-	ret = (void *)vstart;
+	/* On ARM this function returns an ioremap'ped virtual address for
+	 * which virt_to_phys doesn't return the corresponding physical
+	 * address. In fact on ARM virt_to_phys only works for kernel direct
+	 * mapped RAM memory. Also see comment below.
+	 */
+	ret = xen_alloc_coherent_pages(hwdev, size, dma_handle, flags, attrs);
 
 	if (!ret)
 		return ret;
@@ -287,18 +311,21 @@ xen_swiotlb_alloc_coherent(struct device *hwdev, size_t size,
 	if (hwdev && hwdev->coherent_dma_mask)
 		dma_mask = dma_alloc_coherent_mask(hwdev, flags);
 
-	phys = virt_to_phys(ret);
+	/* At this point dma_handle is the physical address, next we are
+	 * going to set it to the machine address.
+	 * Do not use virt_to_phys(ret) because on ARM it doesn't correspond
+	 * to *dma_handle. */
+	phys = *dma_handle;
 	dev_addr = xen_phys_to_bus(phys);
 	if (((dev_addr + size - 1 <= dma_mask)) &&
 	    !range_straddles_page_boundary(phys, size))
 		*dma_handle = dev_addr;
 	else {
-		if (xen_create_contiguous_region(vstart, order,
-						 fls64(dma_mask)) != 0) {
-			free_pages(vstart, order);
+		if (xen_create_contiguous_region(phys, order,
+						 fls64(dma_mask), dma_handle) != 0) {
+			xen_free_coherent_pages(hwdev, size, ret, (dma_addr_t)phys, attrs);
 			return NULL;
 		}
-		*dma_handle = virt_to_machine(ret).maddr;
 	}
 	memset(ret, 0, size);
 	return ret;
@@ -319,13 +346,15 @@ xen_swiotlb_free_coherent(struct device *hwdev, size_t size, void *vaddr,
 	if (hwdev && hwdev->coherent_dma_mask)
 		dma_mask = hwdev->coherent_dma_mask;
 
-	phys = virt_to_phys(vaddr);
+	/* do not use virt_to_phys because on ARM it doesn't return you the
+	 * physical address */
+	phys = xen_bus_to_phys(dev_addr);
 
 	if (((dev_addr + size - 1 > dma_mask)) ||
 	    range_straddles_page_boundary(phys, size))
-		xen_destroy_contiguous_region((unsigned long)vaddr, order);
+		xen_destroy_contiguous_region(phys, order);
 
-	free_pages((unsigned long)vaddr, order);
+	xen_free_coherent_pages(hwdev, size, vaddr, (dma_addr_t)phys, attrs);
 }
 EXPORT_SYMBOL_GPL(xen_swiotlb_free_coherent);
 
@@ -352,16 +381,25 @@ dma_addr_t xen_swiotlb_map_page(struct device *dev, struct page *page,
 	 * buffering it.
 	 */
 	if (dma_capable(dev, dev_addr, size) &&
-	    !range_straddles_page_boundary(phys, size) && !swiotlb_force)
+	    !range_straddles_page_boundary(phys, size) && !swiotlb_force) {
+		/* we are not interested in the dma_addr returned by
+		 * xen_dma_map_page, only in the potential cache flushes executed
+		 * by the function. */
+		xen_dma_map_page(dev, page, offset, size, dir, attrs);
 		return dev_addr;
+	}
 
 	/*
 	 * Oh well, have to allocate and map a bounce buffer.
 	 */
+	trace_swiotlb_bounced(dev, dev_addr, size, swiotlb_force);
+
 	map = swiotlb_tbl_map_single(dev, start_dma_addr, phys, size, dir);
 	if (map == SWIOTLB_MAP_ERROR)
 		return DMA_ERROR_CODE;
 
+	xen_dma_map_page(dev, pfn_to_page(map >> PAGE_SHIFT),
+					map & ~PAGE_MASK, size, dir, attrs);
 	dev_addr = xen_phys_to_bus(map);
 
 	/*
@@ -384,12 +422,15 @@ EXPORT_SYMBOL_GPL(xen_swiotlb_map_page);
  * whatever the device wrote there.
  */
 static void xen_unmap_single(struct device *hwdev, dma_addr_t dev_addr,
-			     size_t size, enum dma_data_direction dir)
+			     size_t size, enum dma_data_direction dir,
+				 struct dma_attrs *attrs)
 {
 	phys_addr_t paddr = xen_bus_to_phys(dev_addr);
 
 	BUG_ON(dir == DMA_NONE);
 
+	xen_dma_unmap_page(hwdev, paddr, size, dir, attrs);
+
 	/* NOTE: We use dev_addr here, not paddr! */
 	if (is_xen_swiotlb_buffer(dev_addr)) {
 		swiotlb_tbl_unmap_single(hwdev, paddr, size, dir);
@@ -412,7 +453,7 @@ void xen_swiotlb_unmap_page(struct device *hwdev, dma_addr_t dev_addr,
 			    size_t size, enum dma_data_direction dir,
 			    struct dma_attrs *attrs)
 {
-	xen_unmap_single(hwdev, dev_addr, size, dir);
+	xen_unmap_single(hwdev, dev_addr, size, dir, attrs);
 }
 EXPORT_SYMBOL_GPL(xen_swiotlb_unmap_page);
 
@@ -435,11 +476,15 @@ xen_swiotlb_sync_single(struct device *hwdev, dma_addr_t dev_addr,
 
 	BUG_ON(dir == DMA_NONE);
 
+	if (target == SYNC_FOR_CPU)
+		xen_dma_sync_single_for_cpu(hwdev, paddr, size, dir);
+
 	/* NOTE: We use dev_addr here, not paddr! */
-	if (is_xen_swiotlb_buffer(dev_addr)) {
+	if (is_xen_swiotlb_buffer(dev_addr))
 		swiotlb_tbl_sync_single(hwdev, paddr, size, dir, target);
-		return;
-	}
+
+	if (target == SYNC_FOR_DEVICE)
+		xen_dma_sync_single_for_cpu(hwdev, paddr, size, dir);
 
 	if (dir != DMA_FROM_DEVICE)
 		return;
@@ -502,16 +547,26 @@ xen_swiotlb_map_sg_attrs(struct device *hwdev, struct scatterlist *sgl,
 								 sg->length,
 								 dir);
 			if (map == SWIOTLB_MAP_ERROR) {
+				dev_warn(hwdev, "swiotlb buffer is full\n");
 				/* Don't panic here, we expect map_sg users
 				   to do proper error handling. */
 				xen_swiotlb_unmap_sg_attrs(hwdev, sgl, i, dir,
 							   attrs);
 				sg_dma_len(sgl) = 0;
-				return DMA_ERROR_CODE;
+				return 0;
 			}
 			sg->dma_address = xen_phys_to_bus(map);
-		} else
+		} else {
+			/* we are not interested in the dma_addr returned by
+			 * xen_dma_map_page, only in the potential cache flushes executed
+			 * by the function. */
+			xen_dma_map_page(hwdev, pfn_to_page(paddr >> PAGE_SHIFT),
+						paddr & ~PAGE_MASK,
+						sg->length,
+						dir,
+						attrs);
 			sg->dma_address = dev_addr;
+		}
 		sg_dma_len(sg) = sg->length;
 	}
 	return nelems;
@@ -533,7 +588,7 @@ xen_swiotlb_unmap_sg_attrs(struct device *hwdev, struct scatterlist *sgl,
 	BUG_ON(dir == DMA_NONE);
 
 	for_each_sg(sgl, sg, nelems, i)
-		xen_unmap_single(hwdev, sg->dma_address, sg_dma_len(sg), dir);
+		xen_unmap_single(hwdev, sg->dma_address, sg_dma_len(sg), dir, attrs);
 
 }
 EXPORT_SYMBOL_GPL(xen_swiotlb_unmap_sg_attrs);
@@ -593,3 +648,15 @@ xen_swiotlb_dma_supported(struct device *hwdev, u64 mask)
 	return xen_virt_to_bus(xen_io_tlb_end - 1) <= mask;
 }
 EXPORT_SYMBOL_GPL(xen_swiotlb_dma_supported);
+
+int
+xen_swiotlb_set_dma_mask(struct device *dev, u64 dma_mask)
+{
+	if (!dev->dma_mask || !xen_swiotlb_dma_supported(dev, dma_mask))
+		return -EIO;
+
+	*dev->dma_mask = dma_mask;
+
+	return 0;
+}
+EXPORT_SYMBOL_GPL(xen_swiotlb_set_dma_mask);
diff --git a/fs/bio.c b/fs/bio.c
index 2bdb4e25ee77..33d79a4eb92d 100644
--- a/fs/bio.c
+++ b/fs/bio.c
@@ -601,7 +601,7 @@ EXPORT_SYMBOL(bio_get_nr_vecs);
 
 static int __bio_add_page(struct request_queue *q, struct bio *bio, struct page
 			  *page, unsigned int len, unsigned int offset,
-			  unsigned short max_sectors)
+			  unsigned int max_sectors)
 {
 	int retried_segments = 0;
 	struct bio_vec *bvec;
diff --git a/fs/btrfs/Kconfig b/fs/btrfs/Kconfig
index 398cbd517be2..f9d5094e1029 100644
--- a/fs/btrfs/Kconfig
+++ b/fs/btrfs/Kconfig
@@ -59,7 +59,8 @@ config BTRFS_FS_RUN_SANITY_TESTS
 	help
 	  This will run some basic sanity tests on the free space cache
 	  code to make sure it is acting as it should.  These are mostly
-	  regression tests and are only really interesting to btrfs devlopers.
+	  regression tests and are only really interesting to btrfs
+	  developers.
 
 	  If unsure, say N.
 
diff --git a/fs/btrfs/Makefile b/fs/btrfs/Makefile
index a91a6a355cc5..1a44e42d602a 100644
--- a/fs/btrfs/Makefile
+++ b/fs/btrfs/Makefile
@@ -14,4 +14,6 @@ btrfs-y += super.o ctree.o extent-tree.o print-tree.o root-tree.o dir-item.o \
 btrfs-$(CONFIG_BTRFS_FS_POSIX_ACL) += acl.o
 btrfs-$(CONFIG_BTRFS_FS_CHECK_INTEGRITY) += check-integrity.o
 
-btrfs-$(CONFIG_BTRFS_FS_RUN_SANITY_TESTS) += tests/free-space-tests.o
+btrfs-$(CONFIG_BTRFS_FS_RUN_SANITY_TESTS) += tests/free-space-tests.o \
+	tests/extent-buffer-tests.o tests/btrfs-tests.o \
+	tests/extent-io-tests.o tests/inode-tests.o
diff --git a/fs/btrfs/acl.c b/fs/btrfs/acl.c
index e15d2b0d8d3b..0890c83643e9 100644
--- a/fs/btrfs/acl.c
+++ b/fs/btrfs/acl.c
@@ -229,7 +229,7 @@ int btrfs_init_acl(struct btrfs_trans_handle *trans,
 		if (ret > 0) {
 			/* we need an acl */
 			ret = btrfs_set_acl(trans, inode, acl, ACL_TYPE_ACCESS);
-		} else {
+		} else if (ret < 0) {
 			cache_no_acl(inode);
 		}
 	} else {
diff --git a/fs/btrfs/async-thread.c b/fs/btrfs/async-thread.c
index 08cc08f037a6..8aec751fa464 100644
--- a/fs/btrfs/async-thread.c
+++ b/fs/btrfs/async-thread.c
@@ -262,7 +262,7 @@ static struct btrfs_work *get_next_work(struct btrfs_worker_thread *worker,
 	struct btrfs_work *work = NULL;
 	struct list_head *cur = NULL;
 
-	if(!list_empty(prio_head))
+	if (!list_empty(prio_head))
 		cur = prio_head->next;
 
 	smp_mb();
diff --git a/fs/btrfs/backref.c b/fs/btrfs/backref.c
index 0552a599b28f..3775947429b2 100644
--- a/fs/btrfs/backref.c
+++ b/fs/btrfs/backref.c
@@ -185,6 +185,9 @@ static int __add_prelim_ref(struct list_head *head, u64 root_id,
 {
 	struct __prelim_ref *ref;
 
+	if (root_id == BTRFS_DATA_RELOC_TREE_OBJECTID)
+		return 0;
+
 	ref = kmem_cache_alloc(btrfs_prelim_ref_cache, gfp_mask);
 	if (!ref)
 		return -ENOMEM;
@@ -323,8 +326,7 @@ static int __resolve_indirect_ref(struct btrfs_fs_info *fs_info,
 
 	eb = path->nodes[level];
 	while (!eb) {
-		if (!level) {
-			WARN_ON(1);
+		if (WARN_ON(!level)) {
 			ret = 1;
 			goto out;
 		}
@@ -1619,7 +1621,7 @@ static int iterate_inode_refs(u64 inum, struct btrfs_root *fs_root,
 		btrfs_set_lock_blocking_rw(eb, BTRFS_READ_LOCK);
 		btrfs_release_path(path);
 
-		item = btrfs_item_nr(eb, slot);
+		item = btrfs_item_nr(slot);
 		iref = btrfs_item_ptr(eb, slot, struct btrfs_inode_ref);
 
 		for (cur = 0; cur < btrfs_item_size(eb, item); cur += len) {
diff --git a/fs/btrfs/btrfs_inode.h b/fs/btrfs/btrfs_inode.h
index 71f074e1870b..ac0b39db27d1 100644
--- a/fs/btrfs/btrfs_inode.h
+++ b/fs/btrfs/btrfs_inode.h
@@ -19,6 +19,7 @@
 #ifndef __BTRFS_I__
 #define __BTRFS_I__
 
+#include <linux/hash.h>
 #include "extent_map.h"
 #include "extent_io.h"
 #include "ordered-data.h"
@@ -179,6 +180,25 @@ static inline struct btrfs_inode *BTRFS_I(struct inode *inode)
 	return container_of(inode, struct btrfs_inode, vfs_inode);
 }
 
+static inline unsigned long btrfs_inode_hash(u64 objectid,
+					     const struct btrfs_root *root)
+{
+	u64 h = objectid ^ (root->objectid * GOLDEN_RATIO_PRIME);
+
+#if BITS_PER_LONG == 32
+	h = (h >> 32) ^ (h & 0xffffffff);
+#endif
+
+	return (unsigned long)h;
+}
+
+static inline void btrfs_insert_inode_hash(struct inode *inode)
+{
+	unsigned long h = btrfs_inode_hash(inode->i_ino, BTRFS_I(inode)->root);
+
+	__insert_inode_hash(inode, h);
+}
+
 static inline u64 btrfs_ino(struct inode *inode)
 {
 	u64 ino = BTRFS_I(inode)->location.objectid;
diff --git a/fs/btrfs/check-integrity.c b/fs/btrfs/check-integrity.c
index 1c47be187240..e0aab4456974 100644
--- a/fs/btrfs/check-integrity.c
+++ b/fs/btrfs/check-integrity.c
@@ -1038,7 +1038,7 @@ leaf_item_out_of_bounce_error:
 						     disk_item_offset,
 						     sizeof(struct btrfs_item));
 			item_offset = btrfs_stack_item_offset(&disk_item);
-			item_size = btrfs_stack_item_offset(&disk_item);
+			item_size = btrfs_stack_item_size(&disk_item);
 			disk_key = &disk_item.key;
 			type = btrfs_disk_key_type(disk_key);
 
@@ -1900,7 +1900,9 @@ again:
 							       dev_state,
 							       dev_bytenr);
 			}
-			if (block->logical_bytenr != bytenr) {
+			if (block->logical_bytenr != bytenr &&
+			    !(!block->is_metadata &&
+			      block->logical_bytenr == 0))
 				printk(KERN_INFO
 				       "Written block @%llu (%s/%llu/%d)"
 				       " found in hash table, %c,"
@@ -1910,15 +1912,14 @@ again:
 				       block->mirror_num,
 				       btrfsic_get_block_type(state, block),
 				       block->logical_bytenr);
-				block->logical_bytenr = bytenr;
-			} else if (state->print_mask &
-				   BTRFSIC_PRINT_MASK_VERBOSE)
+			else if (state->print_mask & BTRFSIC_PRINT_MASK_VERBOSE)
 				printk(KERN_INFO
 				       "Written block @%llu (%s/%llu/%d)"
 				       " found in hash table, %c.\n",
 				       bytenr, dev_state->name, dev_bytenr,
 				       block->mirror_num,
 				       btrfsic_get_block_type(state, block));
+			block->logical_bytenr = bytenr;
 		} else {
 			if (num_pages * PAGE_CACHE_SIZE <
 			    state->datablock_size) {
@@ -2463,10 +2464,8 @@ static int btrfsic_process_written_superblock(
 		}
 	}
 
-	if (-1 == btrfsic_check_all_ref_blocks(state, superblock, 0)) {
-		WARN_ON(1);
+	if (WARN_ON(-1 == btrfsic_check_all_ref_blocks(state, superblock, 0)))
 		btrfsic_dump_tree(state);
-	}
 
 	return 0;
 }
@@ -2906,7 +2905,7 @@ static void btrfsic_cmp_log_and_dev_bytenr(struct btrfsic_state *state,
 		btrfsic_release_block_ctx(&block_ctx);
 	}
 
-	if (!match) {
+	if (WARN_ON(!match)) {
 		printk(KERN_INFO "btrfs: attempt to write M-block which contains logical bytenr that doesn't map to dev+physical bytenr of submit_bio,"
 		       " buffer->log_bytenr=%llu, submit_bio(bdev=%s,"
 		       " phys_bytenr=%llu)!\n",
@@ -2923,7 +2922,6 @@ static void btrfsic_cmp_log_and_dev_bytenr(struct btrfsic_state *state,
 			       bytenr, block_ctx.dev->name,
 			       block_ctx.dev_bytenr, mirror_num);
 		}
-		WARN_ON(1);
 	}
 }
 
diff --git a/fs/btrfs/compat.h b/fs/btrfs/compat.h
deleted file mode 100644
index 7c4503ef6efd..000000000000
--- a/fs/btrfs/compat.h
+++ /dev/null
@@ -1,7 +0,0 @@
-#ifndef _COMPAT_H_
-#define _COMPAT_H_
-
-#define btrfs_drop_nlink(inode) drop_nlink(inode)
-#define btrfs_inc_nlink(inode)	inc_nlink(inode)
-
-#endif /* _COMPAT_H_ */
diff --git a/fs/btrfs/compression.c b/fs/btrfs/compression.c
index 6aad98cb343f..1499b27b4186 100644
--- a/fs/btrfs/compression.c
+++ b/fs/btrfs/compression.c
@@ -32,7 +32,6 @@
 #include <linux/writeback.h>
 #include <linux/bit_spinlock.h>
 #include <linux/slab.h>
-#include "compat.h"
 #include "ctree.h"
 #include "disk-io.h"
 #include "transaction.h"
@@ -360,7 +359,7 @@ int btrfs_submit_compressed_write(struct inode *inode, u64 start,
 	bdev = BTRFS_I(inode)->root->fs_info->fs_devices->latest_bdev;
 
 	bio = compressed_bio_alloc(bdev, first_byte, GFP_NOFS);
-	if(!bio) {
+	if (!bio) {
 		kfree(cb);
 		return -ENOMEM;
 	}
diff --git a/fs/btrfs/ctree.c b/fs/btrfs/ctree.c
index 61b5bcd57b7e..316136bd6dd7 100644
--- a/fs/btrfs/ctree.c
+++ b/fs/btrfs/ctree.c
@@ -274,7 +274,7 @@ int btrfs_copy_root(struct btrfs_trans_handle *trans,
 	else
 		btrfs_set_header_owner(cow, new_root_objectid);
 
-	write_extent_buffer(cow, root->fs_info->fsid, btrfs_header_fsid(cow),
+	write_extent_buffer(cow, root->fs_info->fsid, btrfs_header_fsid(),
 			    BTRFS_FSID_SIZE);
 
 	WARN_ON(btrfs_header_generation(buf) > trans->transid);
@@ -996,7 +996,7 @@ static noinline int __btrfs_cow_block(struct btrfs_trans_handle *trans,
 	else
 		btrfs_set_header_owner(cow, root->root_key.objectid);
 
-	write_extent_buffer(cow, root->fs_info->fsid, btrfs_header_fsid(cow),
+	write_extent_buffer(cow, root->fs_info->fsid, btrfs_header_fsid(),
 			    BTRFS_FSID_SIZE);
 
 	ret = update_ref_for_cow(trans, root, buf, cow, &last_ref);
@@ -1285,11 +1285,10 @@ get_old_root(struct btrfs_root *root, u64 time_seq)
 		free_extent_buffer(eb_root);
 		blocksize = btrfs_level_size(root, old_root->level);
 		old = read_tree_block(root, logical, blocksize, 0);
-		if (!old || !extent_buffer_uptodate(old)) {
+		if (WARN_ON(!old || !extent_buffer_uptodate(old))) {
 			free_extent_buffer(old);
 			pr_warn("btrfs: failed to read tree block %llu from get_old_root\n",
 				logical);
-			WARN_ON(1);
 		} else {
 			eb = btrfs_clone_extent_buffer(old);
 			free_extent_buffer(old);
@@ -2758,7 +2757,7 @@ int btrfs_search_old_slot(struct btrfs_root *root, struct btrfs_key *key,
 	int level;
 	int lowest_unlock = 1;
 	u8 lowest_level = 0;
-	int prev_cmp;
+	int prev_cmp = -1;
 
 	lowest_level = p->lowest_level;
 	WARN_ON(p->nodes[0] != NULL);
@@ -2769,7 +2768,6 @@ int btrfs_search_old_slot(struct btrfs_root *root, struct btrfs_key *key,
 	}
 
 again:
-	prev_cmp = -1;
 	b = get_old_root(root, time_seq);
 	level = btrfs_header_level(b);
 	p->locks[level] = BTRFS_READ_LOCK;
@@ -2787,6 +2785,11 @@ again:
 		 */
 		btrfs_unlock_up_safe(p, level + 1);
 
+		/*
+		 * Since we can unwind eb's we want to do a real search every
+		 * time.
+		 */
+		prev_cmp = -1;
 		ret = key_search(b, key, level, &prev_cmp, &slot);
 
 		if (level != 0) {
@@ -3148,7 +3151,7 @@ static noinline int insert_new_root(struct btrfs_trans_handle *trans,
 	btrfs_set_header_backref_rev(c, BTRFS_MIXED_BACKREF_REV);
 	btrfs_set_header_owner(c, root->root_key.objectid);
 
-	write_extent_buffer(c, root->fs_info->fsid, btrfs_header_fsid(c),
+	write_extent_buffer(c, root->fs_info->fsid, btrfs_header_fsid(),
 			    BTRFS_FSID_SIZE);
 
 	write_extent_buffer(c, root->fs_info->chunk_tree_uuid,
@@ -3287,7 +3290,7 @@ static noinline int split_node(struct btrfs_trans_handle *trans,
 	btrfs_set_header_backref_rev(split, BTRFS_MIXED_BACKREF_REV);
 	btrfs_set_header_owner(split, root->root_key.objectid);
 	write_extent_buffer(split, root->fs_info->fsid,
-			    btrfs_header_fsid(split), BTRFS_FSID_SIZE);
+			    btrfs_header_fsid(), BTRFS_FSID_SIZE);
 	write_extent_buffer(split, root->fs_info->chunk_tree_uuid,
 			    btrfs_header_chunk_tree_uuid(split),
 			    BTRFS_UUID_SIZE);
@@ -3337,8 +3340,8 @@ static int leaf_space_used(struct extent_buffer *l, int start, int nr)
 	if (!nr)
 		return 0;
 	btrfs_init_map_token(&token);
-	start_item = btrfs_item_nr(l, start);
-	end_item = btrfs_item_nr(l, end);
+	start_item = btrfs_item_nr(start);
+	end_item = btrfs_item_nr(end);
 	data_len = btrfs_token_item_offset(l, start_item, &token) +
 		btrfs_token_item_size(l, start_item, &token);
 	data_len = data_len - btrfs_token_item_offset(l, end_item, &token);
@@ -3406,7 +3409,7 @@ static noinline int __push_leaf_right(struct btrfs_trans_handle *trans,
 	slot = path->slots[1];
 	i = left_nritems - 1;
 	while (i >= nr) {
-		item = btrfs_item_nr(left, i);
+		item = btrfs_item_nr(i);
 
 		if (!empty && push_items > 0) {
 			if (path->slots[0] > i)
@@ -3470,7 +3473,7 @@ static noinline int __push_leaf_right(struct btrfs_trans_handle *trans,
 	btrfs_set_header_nritems(right, right_nritems);
 	push_space = BTRFS_LEAF_DATA_SIZE(root);
 	for (i = 0; i < right_nritems; i++) {
-		item = btrfs_item_nr(right, i);
+		item = btrfs_item_nr(i);
 		push_space -= btrfs_token_item_size(right, item, &token);
 		btrfs_set_token_item_offset(right, item, push_space, &token);
 	}
@@ -3612,7 +3615,7 @@ static noinline int __push_leaf_left(struct btrfs_trans_handle *trans,
 		nr = min(right_nritems - 1, max_slot);
 
 	for (i = 0; i < nr; i++) {
-		item = btrfs_item_nr(right, i);
+		item = btrfs_item_nr(i);
 
 		if (!empty && push_items > 0) {
 			if (path->slots[0] < i)
@@ -3639,8 +3642,7 @@ static noinline int __push_leaf_left(struct btrfs_trans_handle *trans,
 		ret = 1;
 		goto out;
 	}
-	if (!empty && push_items == btrfs_header_nritems(right))
-		WARN_ON(1);
+	WARN_ON(!empty && push_items == btrfs_header_nritems(right));
 
 	/* push data from right to left */
 	copy_extent_buffer(left, right,
@@ -3663,7 +3665,7 @@ static noinline int __push_leaf_left(struct btrfs_trans_handle *trans,
 	for (i = old_left_nritems; i < old_left_nritems + push_items; i++) {
 		u32 ioff;
 
-		item = btrfs_item_nr(left, i);
+		item = btrfs_item_nr(i);
 
 		ioff = btrfs_token_item_offset(left, item, &token);
 		btrfs_set_token_item_offset(left, item,
@@ -3694,7 +3696,7 @@ static noinline int __push_leaf_left(struct btrfs_trans_handle *trans,
 	btrfs_set_header_nritems(right, right_nritems);
 	push_space = BTRFS_LEAF_DATA_SIZE(root);
 	for (i = 0; i < right_nritems; i++) {
-		item = btrfs_item_nr(right, i);
+		item = btrfs_item_nr(i);
 
 		push_space = push_space - btrfs_token_item_size(right,
 								item, &token);
@@ -3835,7 +3837,7 @@ static noinline void copy_for_split(struct btrfs_trans_handle *trans,
 		      btrfs_item_end_nr(l, mid);
 
 	for (i = 0; i < nritems; i++) {
-		struct btrfs_item *item = btrfs_item_nr(right, i);
+		struct btrfs_item *item = btrfs_item_nr(i);
 		u32 ioff;
 
 		ioff = btrfs_token_item_offset(right, item, &token);
@@ -4016,7 +4018,7 @@ again:
 				    data_size > BTRFS_LEAF_DATA_SIZE(root)) {
 					if (data_size && !tried_avoid_double)
 						goto push_for_double;
-					split = 2 ;
+					split = 2;
 				}
 			}
 		}
@@ -4042,7 +4044,7 @@ again:
 	btrfs_set_header_owner(right, root->root_key.objectid);
 	btrfs_set_header_level(right, 0);
 	write_extent_buffer(right, root->fs_info->fsid,
-			    btrfs_header_fsid(right), BTRFS_FSID_SIZE);
+			    btrfs_header_fsid(), BTRFS_FSID_SIZE);
 
 	write_extent_buffer(right, root->fs_info->chunk_tree_uuid,
 			    btrfs_header_chunk_tree_uuid(right),
@@ -4177,7 +4179,7 @@ static noinline int split_item(struct btrfs_trans_handle *trans,
 
 	btrfs_set_path_blocking(path);
 
-	item = btrfs_item_nr(leaf, path->slots[0]);
+	item = btrfs_item_nr(path->slots[0]);
 	orig_offset = btrfs_item_offset(leaf, item);
 	item_size = btrfs_item_size(leaf, item);
 
@@ -4200,7 +4202,7 @@ static noinline int split_item(struct btrfs_trans_handle *trans,
 	btrfs_cpu_key_to_disk(&disk_key, new_key);
 	btrfs_set_item_key(leaf, &disk_key, slot);
 
-	new_item = btrfs_item_nr(leaf, slot);
+	new_item = btrfs_item_nr(slot);
 
 	btrfs_set_item_offset(leaf, new_item, orig_offset);
 	btrfs_set_item_size(leaf, new_item, item_size - split_offset);
@@ -4339,7 +4341,7 @@ void btrfs_truncate_item(struct btrfs_root *root, struct btrfs_path *path,
 	/* first correct the data pointers */
 	for (i = slot; i < nritems; i++) {
 		u32 ioff;
-		item = btrfs_item_nr(leaf, i);
+		item = btrfs_item_nr(i);
 
 		ioff = btrfs_token_item_offset(leaf, item, &token);
 		btrfs_set_token_item_offset(leaf, item,
@@ -4387,7 +4389,7 @@ void btrfs_truncate_item(struct btrfs_root *root, struct btrfs_path *path,
 			fixup_low_keys(root, path, &disk_key, 1);
 	}
 
-	item = btrfs_item_nr(leaf, slot);
+	item = btrfs_item_nr(slot);
 	btrfs_set_item_size(leaf, item, new_size);
 	btrfs_mark_buffer_dirty(leaf);
 
@@ -4441,7 +4443,7 @@ void btrfs_extend_item(struct btrfs_root *root, struct btrfs_path *path,
 	/* first correct the data pointers */
 	for (i = slot; i < nritems; i++) {
 		u32 ioff;
-		item = btrfs_item_nr(leaf, i);
+		item = btrfs_item_nr(i);
 
 		ioff = btrfs_token_item_offset(leaf, item, &token);
 		btrfs_set_token_item_offset(leaf, item,
@@ -4455,7 +4457,7 @@ void btrfs_extend_item(struct btrfs_root *root, struct btrfs_path *path,
 
 	data_end = old_data;
 	old_size = btrfs_item_size_nr(leaf, slot);
-	item = btrfs_item_nr(leaf, slot);
+	item = btrfs_item_nr(slot);
 	btrfs_set_item_size(leaf, item, old_size + data_size);
 	btrfs_mark_buffer_dirty(leaf);
 
@@ -4514,7 +4516,7 @@ void setup_items_for_insert(struct btrfs_root *root, struct btrfs_path *path,
 		for (i = slot; i < nritems; i++) {
 			u32 ioff;
 
-			item = btrfs_item_nr(leaf, i);
+			item = btrfs_item_nr( i);
 			ioff = btrfs_token_item_offset(leaf, item, &token);
 			btrfs_set_token_item_offset(leaf, item,
 						    ioff - total_data, &token);
@@ -4535,7 +4537,7 @@ void setup_items_for_insert(struct btrfs_root *root, struct btrfs_path *path,
 	for (i = 0; i < nr; i++) {
 		btrfs_cpu_key_to_disk(&disk_key, cpu_key + i);
 		btrfs_set_item_key(leaf, &disk_key, slot + i);
-		item = btrfs_item_nr(leaf, slot + i);
+		item = btrfs_item_nr(slot + i);
 		btrfs_set_token_item_offset(leaf, item,
 					    data_end - data_size[i], &token);
 		data_end -= data_size[i];
@@ -4730,7 +4732,7 @@ int btrfs_del_items(struct btrfs_trans_handle *trans, struct btrfs_root *root,
 		for (i = slot + nr; i < nritems; i++) {
 			u32 ioff;
 
-			item = btrfs_item_nr(leaf, i);
+			item = btrfs_item_nr(i);
 			ioff = btrfs_token_item_offset(leaf, item, &token);
 			btrfs_set_token_item_offset(leaf, item,
 						    ioff + dsize, &token);
@@ -4823,14 +4825,18 @@ static int btrfs_prev_leaf(struct btrfs_root *root, struct btrfs_path *path)
 
 	btrfs_item_key_to_cpu(path->nodes[0], &key, 0);
 
-	if (key.offset > 0)
+	if (key.offset > 0) {
 		key.offset--;
-	else if (key.type > 0)
+	} else if (key.type > 0) {
 		key.type--;
-	else if (key.objectid > 0)
+		key.offset = (u64)-1;
+	} else if (key.objectid > 0) {
 		key.objectid--;
-	else
+		key.type = (u8)-1;
+		key.offset = (u64)-1;
+	} else {
 		return 1;
+	}
 
 	btrfs_release_path(path);
 	ret = btrfs_search_slot(NULL, root, &key, path, 0, 0);
@@ -4866,7 +4872,6 @@ static int btrfs_prev_leaf(struct btrfs_root *root, struct btrfs_path *path)
  * was nothing in the tree that matched the search criteria.
  */
 int btrfs_search_forward(struct btrfs_root *root, struct btrfs_key *min_key,
-			 struct btrfs_key *max_key,
 			 struct btrfs_path *path,
 			 u64 min_trans)
 {
@@ -4911,10 +4916,8 @@ again:
 		 * If it is too old, old, skip to the next one.
 		 */
 		while (slot < nritems) {
-			u64 blockptr;
 			u64 gen;
 
-			blockptr = btrfs_node_blockptr(cur, slot);
 			gen = btrfs_node_ptr_generation(cur, slot);
 			if (gen < min_trans) {
 				slot++;
diff --git a/fs/btrfs/ctree.h b/fs/btrfs/ctree.h
index 0506f40ede83..f9aeb2759a64 100644
--- a/fs/btrfs/ctree.h
+++ b/fs/btrfs/ctree.h
@@ -47,6 +47,12 @@ extern struct kmem_cache *btrfs_path_cachep;
 extern struct kmem_cache *btrfs_free_space_cachep;
 struct btrfs_ordered_sum;
 
+#ifdef CONFIG_BTRFS_FS_RUN_SANITY_TESTS
+#define STATIC noinline
+#else
+#define STATIC static noinline
+#endif
+
 #define BTRFS_MAGIC 0x4D5F53665248425FULL /* ascii _BHRfS_M, no null */
 
 #define BTRFS_MAX_MIRRORS 3
@@ -1580,7 +1586,6 @@ struct btrfs_fs_info {
 	atomic_t scrubs_paused;
 	atomic_t scrub_cancel_req;
 	wait_queue_head_t scrub_pause_wait;
-	struct rw_semaphore scrub_super_lock;
 	int scrub_workers_refcnt;
 	struct btrfs_workers scrub_workers;
 	struct btrfs_workers scrub_wr_completion_workers;
@@ -1724,7 +1729,9 @@ struct btrfs_root {
 	int ref_cows;
 	int track_dirty;
 	int in_radix;
-
+#ifdef CONFIG_BTRFS_FS_RUN_SANITY_TESTS
+	int dummy_root;
+#endif
 	u64 defrag_trans_start;
 	struct btrfs_key defrag_progress;
 	struct btrfs_key defrag_max;
@@ -2461,8 +2468,7 @@ static inline unsigned long btrfs_item_nr_offset(int nr)
 		sizeof(struct btrfs_item) * nr;
 }
 
-static inline struct btrfs_item *btrfs_item_nr(struct extent_buffer *eb,
-					       int nr)
+static inline struct btrfs_item *btrfs_item_nr(int nr)
 {
 	return (struct btrfs_item *)btrfs_item_nr_offset(nr);
 }
@@ -2475,30 +2481,30 @@ static inline u32 btrfs_item_end(struct extent_buffer *eb,
 
 static inline u32 btrfs_item_end_nr(struct extent_buffer *eb, int nr)
 {
-	return btrfs_item_end(eb, btrfs_item_nr(eb, nr));
+	return btrfs_item_end(eb, btrfs_item_nr(nr));
 }
 
 static inline u32 btrfs_item_offset_nr(struct extent_buffer *eb, int nr)
 {
-	return btrfs_item_offset(eb, btrfs_item_nr(eb, nr));
+	return btrfs_item_offset(eb, btrfs_item_nr(nr));
 }
 
 static inline u32 btrfs_item_size_nr(struct extent_buffer *eb, int nr)
 {
-	return btrfs_item_size(eb, btrfs_item_nr(eb, nr));
+	return btrfs_item_size(eb, btrfs_item_nr(nr));
 }
 
 static inline void btrfs_item_key(struct extent_buffer *eb,
 			   struct btrfs_disk_key *disk_key, int nr)
 {
-	struct btrfs_item *item = btrfs_item_nr(eb, nr);
+	struct btrfs_item *item = btrfs_item_nr(nr);
 	read_eb_member(eb, item, struct btrfs_item, key, disk_key);
 }
 
 static inline void btrfs_set_item_key(struct extent_buffer *eb,
 			       struct btrfs_disk_key *disk_key, int nr)
 {
-	struct btrfs_item *item = btrfs_item_nr(eb, nr);
+	struct btrfs_item *item = btrfs_item_nr(nr);
 	write_eb_member(eb, item, struct btrfs_item, key, disk_key);
 }
 
@@ -2666,7 +2672,7 @@ static inline void btrfs_set_header_backref_rev(struct extent_buffer *eb,
 	btrfs_set_header_flags(eb, flags);
 }
 
-static inline unsigned long btrfs_header_fsid(struct extent_buffer *eb)
+static inline unsigned long btrfs_header_fsid(void)
 {
 	return offsetof(struct btrfs_header, fsid);
 }
@@ -3105,11 +3111,6 @@ static inline u32 btrfs_level_size(struct btrfs_root *root, int level)
 	((unsigned long)(btrfs_leaf_data(leaf) + \
 	btrfs_item_offset_nr(leaf, slot)))
 
-static inline struct dentry *fdentry(struct file *file)
-{
-	return file->f_path.dentry;
-}
-
 static inline bool btrfs_mixed_space_info(struct btrfs_space_info *space_info)
 {
 	return ((space_info->flags & BTRFS_BLOCK_GROUP_METADATA) &&
@@ -3308,7 +3309,6 @@ int btrfs_find_next_key(struct btrfs_root *root, struct btrfs_path *path,
 			struct btrfs_key *key, int lowest_level,
 			u64 min_trans);
 int btrfs_search_forward(struct btrfs_root *root, struct btrfs_key *min_key,
-			 struct btrfs_key *max_key,
 			 struct btrfs_path *path,
 			 u64 min_trans);
 enum btrfs_compare_tree_result {
@@ -3675,8 +3675,7 @@ int btrfs_truncate_inode_items(struct btrfs_trans_handle *trans,
 			       u32 min_type);
 
 int btrfs_start_delalloc_inodes(struct btrfs_root *root, int delay_iput);
-int btrfs_start_all_delalloc_inodes(struct btrfs_fs_info *fs_info,
-				    int delay_iput);
+int btrfs_start_delalloc_roots(struct btrfs_fs_info *fs_info, int delay_iput);
 int btrfs_set_extent_delalloc(struct inode *inode, u64 start, u64 end,
 			      struct extent_state **cached_state);
 int btrfs_create_subvol_root(struct btrfs_trans_handle *trans,
@@ -3944,9 +3943,7 @@ int btrfs_scrub_dev(struct btrfs_fs_info *fs_info, u64 devid, u64 start,
 		    u64 end, struct btrfs_scrub_progress *progress,
 		    int readonly, int is_dev_replace);
 void btrfs_scrub_pause(struct btrfs_root *root);
-void btrfs_scrub_pause_super(struct btrfs_root *root);
 void btrfs_scrub_continue(struct btrfs_root *root);
-void btrfs_scrub_continue_super(struct btrfs_root *root);
 int btrfs_scrub_cancel(struct btrfs_fs_info *info);
 int btrfs_scrub_cancel_dev(struct btrfs_fs_info *info,
 			   struct btrfs_device *dev);
@@ -4028,5 +4025,9 @@ static inline int btrfs_defrag_cancelled(struct btrfs_fs_info *fs_info)
 	return signal_pending(current);
 }
 
+/* Sanity test specific functions */
+#ifdef CONFIG_BTRFS_FS_RUN_SANITY_TESTS
+void btrfs_test_destroy_inode(struct inode *inode);
+#endif
 
 #endif
diff --git a/fs/btrfs/delayed-inode.c b/fs/btrfs/delayed-inode.c
index cbd9523ad09c..8d292fbae659 100644
--- a/fs/btrfs/delayed-inode.c
+++ b/fs/btrfs/delayed-inode.c
@@ -108,8 +108,8 @@ static struct btrfs_delayed_node *btrfs_get_delayed_node(struct inode *inode)
 			return node;
 		}
 		btrfs_inode->delayed_node = node;
-		atomic_inc(&node->refs);	/* can be accessed */
-		atomic_inc(&node->refs);	/* cached in the inode */
+		/* can be accessed and cached in the inode */
+		atomic_add(2, &node->refs);
 		spin_unlock(&root->inode_lock);
 		return node;
 	}
@@ -138,8 +138,8 @@ again:
 		return ERR_PTR(-ENOMEM);
 	btrfs_init_delayed_node(node, root, ino);
 
-	atomic_inc(&node->refs);	/* cached in the btrfs inode */
-	atomic_inc(&node->refs);	/* can be accessed */
+	/* cached in the btrfs inode and can be accessed */
+	atomic_add(2, &node->refs);
 
 	ret = radix_tree_preload(GFP_NOFS & ~__GFP_HIGHMEM);
 	if (ret) {
@@ -649,14 +649,13 @@ static int btrfs_delayed_inode_reserve_metadata(
 			goto out;
 
 		ret = btrfs_block_rsv_migrate(src_rsv, dst_rsv, num_bytes);
-		if (!ret)
+		if (!WARN_ON(ret))
 			goto out;
 
 		/*
 		 * Ok this is a problem, let's just steal from the global rsv
 		 * since this really shouldn't happen that often.
 		 */
-		WARN_ON(1);
 		ret = btrfs_block_rsv_migrate(&root->fs_info->global_block_rsv,
 					      dst_rsv, num_bytes);
 		goto out;
@@ -771,13 +770,13 @@ static int btrfs_batch_insert_items(struct btrfs_root *root,
 	 */
 	btrfs_set_path_blocking(path);
 
-	keys = kmalloc(sizeof(struct btrfs_key) * nitems, GFP_NOFS);
+	keys = kmalloc_array(nitems, sizeof(struct btrfs_key), GFP_NOFS);
 	if (!keys) {
 		ret = -ENOMEM;
 		goto out;
 	}
 
-	data_size = kmalloc(sizeof(u32) * nitems, GFP_NOFS);
+	data_size = kmalloc_array(nitems, sizeof(u32), GFP_NOFS);
 	if (!data_size) {
 		ret = -ENOMEM;
 		goto error;
@@ -1174,8 +1173,10 @@ int btrfs_commit_inode_delayed_items(struct btrfs_trans_handle *trans,
 	mutex_unlock(&delayed_node->mutex);
 
 	path = btrfs_alloc_path();
-	if (!path)
+	if (!path) {
+		btrfs_release_delayed_node(delayed_node);
 		return -ENOMEM;
+	}
 	path->leave_spinning = 1;
 
 	block_rsv = trans->block_rsv;
diff --git a/fs/btrfs/dev-replace.c b/fs/btrfs/dev-replace.c
index 9efb94e95858..342f9fd411e3 100644
--- a/fs/btrfs/dev-replace.c
+++ b/fs/btrfs/dev-replace.c
@@ -26,7 +26,6 @@
 #include <linux/kthread.h>
 #include <linux/math64.h>
 #include <asm/div64.h>
-#include "compat.h"
 #include "ctree.h"
 #include "extent_map.h"
 #include "disk-io.h"
@@ -38,7 +37,6 @@
 #include "rcu-string.h"
 #include "dev-replace.h"
 
-static u64 btrfs_get_seconds_since_1970(void);
 static int btrfs_dev_replace_finishing(struct btrfs_fs_info *fs_info,
 				       int scrub_ret);
 static void btrfs_dev_replace_update_device_in_mapping_tree(
@@ -296,13 +294,6 @@ void btrfs_after_dev_replace_commit(struct btrfs_fs_info *fs_info)
 		dev_replace->cursor_left_last_write_of_item;
 }
 
-static u64 btrfs_get_seconds_since_1970(void)
-{
-	struct timespec t = CURRENT_TIME_SEC;
-
-	return t.tv_sec;
-}
-
 int btrfs_dev_replace_start(struct btrfs_root *root,
 			    struct btrfs_ioctl_dev_replace_args *args)
 {
@@ -390,7 +381,7 @@ int btrfs_dev_replace_start(struct btrfs_root *root,
 	 * go to the tgtdev as well (refer to btrfs_map_block()).
 	 */
 	dev_replace->replace_state = BTRFS_IOCTL_DEV_REPLACE_STATE_STARTED;
-	dev_replace->time_started = btrfs_get_seconds_since_1970();
+	dev_replace->time_started = get_seconds();
 	dev_replace->cursor_left = 0;
 	dev_replace->committed_cursor_left = 0;
 	dev_replace->cursor_left_last_write_of_item = 0;
@@ -400,7 +391,7 @@ int btrfs_dev_replace_start(struct btrfs_root *root,
 	args->result = BTRFS_IOCTL_DEV_REPLACE_RESULT_NO_ERROR;
 	btrfs_dev_replace_unlock(dev_replace);
 
-	btrfs_wait_all_ordered_extents(root->fs_info);
+	btrfs_wait_ordered_roots(root->fs_info, -1);
 
 	/* force writing the updated state information to disk */
 	trans = btrfs_start_transaction(root, 0);
@@ -470,12 +461,12 @@ static int btrfs_dev_replace_finishing(struct btrfs_fs_info *fs_info,
 	 * flush all outstanding I/O and inode extent mappings before the
 	 * copy operation is declared as being finished
 	 */
-	ret = btrfs_start_all_delalloc_inodes(root->fs_info, 0);
+	ret = btrfs_start_delalloc_roots(root->fs_info, 0);
 	if (ret) {
 		mutex_unlock(&dev_replace->lock_finishing_cancel_unmount);
 		return ret;
 	}
-	btrfs_wait_all_ordered_extents(root->fs_info);
+	btrfs_wait_ordered_roots(root->fs_info, -1);
 
 	trans = btrfs_start_transaction(root, 0);
 	if (IS_ERR(trans)) {
@@ -493,7 +484,7 @@ static int btrfs_dev_replace_finishing(struct btrfs_fs_info *fs_info,
 			  : BTRFS_IOCTL_DEV_REPLACE_STATE_FINISHED;
 	dev_replace->tgtdev = NULL;
 	dev_replace->srcdev = NULL;
-	dev_replace->time_stopped = btrfs_get_seconds_since_1970();
+	dev_replace->time_stopped = get_seconds();
 	dev_replace->item_needs_writeback = 1;
 
 	if (scrub_ret) {
@@ -650,6 +641,9 @@ static u64 __btrfs_dev_replace_cancel(struct btrfs_fs_info *fs_info)
 	u64 result;
 	int ret;
 
+	if (fs_info->sb->s_flags & MS_RDONLY)
+		return -EROFS;
+
 	mutex_lock(&dev_replace->lock_finishing_cancel_unmount);
 	btrfs_dev_replace_lock(dev_replace);
 	switch (dev_replace->replace_state) {
@@ -668,7 +662,7 @@ static u64 __btrfs_dev_replace_cancel(struct btrfs_fs_info *fs_info)
 		break;
 	}
 	dev_replace->replace_state = BTRFS_IOCTL_DEV_REPLACE_STATE_CANCELED;
-	dev_replace->time_stopped = btrfs_get_seconds_since_1970();
+	dev_replace->time_stopped = get_seconds();
 	dev_replace->item_needs_writeback = 1;
 	btrfs_dev_replace_unlock(dev_replace);
 	btrfs_scrub_cancel(fs_info);
@@ -703,7 +697,7 @@ void btrfs_dev_replace_suspend_for_unmount(struct btrfs_fs_info *fs_info)
 	case BTRFS_IOCTL_DEV_REPLACE_STATE_STARTED:
 		dev_replace->replace_state =
 			BTRFS_IOCTL_DEV_REPLACE_STATE_SUSPENDED;
-		dev_replace->time_stopped = btrfs_get_seconds_since_1970();
+		dev_replace->time_stopped = get_seconds();
 		dev_replace->item_needs_writeback = 1;
 		pr_info("btrfs: suspending dev_replace for unmount\n");
 		break;
diff --git a/fs/btrfs/dir-item.c b/fs/btrfs/dir-item.c
index 79e594e341c7..c031ea3fd70f 100644
--- a/fs/btrfs/dir-item.c
+++ b/fs/btrfs/dir-item.c
@@ -58,7 +58,7 @@ static struct btrfs_dir_item *insert_with_overflow(struct btrfs_trans_handle
 		return ERR_PTR(ret);
 	WARN_ON(ret > 0);
 	leaf = path->nodes[0];
-	item = btrfs_item_nr(leaf, path->slots[0]);
+	item = btrfs_item_nr(path->slots[0]);
 	ptr = btrfs_item_ptr(leaf, path->slots[0], char);
 	BUG_ON(data_size > btrfs_item_size(leaf, item));
 	ptr += btrfs_item_size(leaf, item) - data_size;
@@ -474,8 +474,10 @@ int verify_dir_item(struct btrfs_root *root,
 	}
 
 	/* BTRFS_MAX_XATTR_SIZE is the same for all dir items */
-	if (btrfs_dir_data_len(leaf, dir_item) > BTRFS_MAX_XATTR_SIZE(root)) {
-		printk(KERN_CRIT "btrfs: invalid dir item data len: %u\n",
+	if ((btrfs_dir_data_len(leaf, dir_item) +
+	     btrfs_dir_name_len(leaf, dir_item)) > BTRFS_MAX_XATTR_SIZE(root)) {
+		printk(KERN_CRIT "btrfs: invalid dir item name + data len: %u + %u\n",
+		       (unsigned)btrfs_dir_name_len(leaf, dir_item),
 		       (unsigned)btrfs_dir_data_len(leaf, dir_item));
 		return 1;
 	}
diff --git a/fs/btrfs/disk-io.c b/fs/btrfs/disk-io.c
index 62176ad89846..4c4ed0bb3da1 100644
--- a/fs/btrfs/disk-io.c
+++ b/fs/btrfs/disk-io.c
@@ -33,7 +33,6 @@
 #include <linux/uuid.h>
 #include <linux/semaphore.h>
 #include <asm/unaligned.h>
-#include "compat.h"
 #include "ctree.h"
 #include "disk-io.h"
 #include "transaction.h"
@@ -64,7 +63,6 @@ static void btrfs_destroy_ordered_operations(struct btrfs_transaction *t,
 static void btrfs_destroy_ordered_extents(struct btrfs_root *root);
 static int btrfs_destroy_delayed_refs(struct btrfs_transaction *trans,
 				      struct btrfs_root *root);
-static void btrfs_evict_pending_snapshots(struct btrfs_transaction *t);
 static void btrfs_destroy_delalloc_inodes(struct btrfs_root *root);
 static int btrfs_destroy_marked_extents(struct btrfs_root *root,
 					struct extent_io_tree *dirty_pages,
@@ -477,14 +475,8 @@ static int csum_dirty_buffer(struct btrfs_root *root, struct page *page)
 	if (page != eb->pages[0])
 		return 0;
 	found_start = btrfs_header_bytenr(eb);
-	if (found_start != start) {
-		WARN_ON(1);
+	if (WARN_ON(found_start != start || !PageUptodate(page)))
 		return 0;
-	}
-	if (!PageUptodate(page)) {
-		WARN_ON(1);
-		return 0;
-	}
 	csum_tree_block(root, eb, 0);
 	return 0;
 }
@@ -496,7 +488,7 @@ static int check_tree_block_fsid(struct btrfs_root *root,
 	u8 fsid[BTRFS_UUID_SIZE];
 	int ret = 1;
 
-	read_extent_buffer(eb, fsid, btrfs_header_fsid(eb), BTRFS_FSID_SIZE);
+	read_extent_buffer(eb, fsid, btrfs_header_fsid(), BTRFS_FSID_SIZE);
 	while (fs_devices) {
 		if (!memcmp(fsid, fs_devices->fsid, BTRFS_FSID_SIZE)) {
 			ret = 0;
@@ -1105,8 +1097,7 @@ struct extent_buffer *btrfs_find_tree_block(struct btrfs_root *root,
 {
 	struct inode *btree_inode = root->fs_info->btree_inode;
 	struct extent_buffer *eb;
-	eb = find_extent_buffer(&BTRFS_I(btree_inode)->io_tree,
-				bytenr, blocksize);
+	eb = find_extent_buffer(&BTRFS_I(btree_inode)->io_tree, bytenr);
 	return eb;
 }
 
@@ -1229,14 +1220,18 @@ static void __setup_root(u32 nodesize, u32 leafsize, u32 sectorsize,
 	atomic_set(&root->refs, 1);
 	root->log_transid = 0;
 	root->last_log_commit = 0;
-	extent_io_tree_init(&root->dirty_log_pages,
-			     fs_info->btree_inode->i_mapping);
+	if (fs_info)
+		extent_io_tree_init(&root->dirty_log_pages,
+				     fs_info->btree_inode->i_mapping);
 
 	memset(&root->root_key, 0, sizeof(root->root_key));
 	memset(&root->root_item, 0, sizeof(root->root_item));
 	memset(&root->defrag_progress, 0, sizeof(root->defrag_progress));
 	memset(&root->root_kobj, 0, sizeof(root->root_kobj));
-	root->defrag_trans_start = fs_info->generation;
+	if (fs_info)
+		root->defrag_trans_start = fs_info->generation;
+	else
+		root->defrag_trans_start = 0;
 	init_completion(&root->kobj_unregister);
 	root->defrag_running = 0;
 	root->root_key.objectid = objectid;
@@ -1253,6 +1248,22 @@ static struct btrfs_root *btrfs_alloc_root(struct btrfs_fs_info *fs_info)
 	return root;
 }
 
+#ifdef CONFIG_BTRFS_FS_RUN_SANITY_TESTS
+/* Should only be used by the testing infrastructure */
+struct btrfs_root *btrfs_alloc_dummy_root(void)
+{
+	struct btrfs_root *root;
+
+	root = btrfs_alloc_root(NULL);
+	if (!root)
+		return ERR_PTR(-ENOMEM);
+	__setup_root(4096, 4096, 4096, 4096, root, NULL, 1);
+	root->dummy_root = 1;
+
+	return root;
+}
+#endif
+
 struct btrfs_root *btrfs_create_tree(struct btrfs_trans_handle *trans,
 				     struct btrfs_fs_info *fs_info,
 				     u64 objectid)
@@ -1292,7 +1303,7 @@ struct btrfs_root *btrfs_create_tree(struct btrfs_trans_handle *trans,
 	btrfs_set_header_owner(leaf, objectid);
 	root->node = leaf;
 
-	write_extent_buffer(leaf, fs_info->fsid, btrfs_header_fsid(leaf),
+	write_extent_buffer(leaf, fs_info->fsid, btrfs_header_fsid(),
 			    BTRFS_FSID_SIZE);
 	write_extent_buffer(leaf, fs_info->chunk_tree_uuid,
 			    btrfs_header_chunk_tree_uuid(leaf),
@@ -1379,7 +1390,7 @@ static struct btrfs_root *alloc_log_tree(struct btrfs_trans_handle *trans,
 	root->node = leaf;
 
 	write_extent_buffer(root->node, root->fs_info->fsid,
-			    btrfs_header_fsid(root->node), BTRFS_FSID_SIZE);
+			    btrfs_header_fsid(), BTRFS_FSID_SIZE);
 	btrfs_mark_buffer_dirty(root->node);
 	btrfs_tree_unlock(root->node);
 	return root;
@@ -1780,6 +1791,9 @@ sleep:
 		wake_up_process(root->fs_info->cleaner_kthread);
 		mutex_unlock(&root->fs_info->transaction_kthread_mutex);
 
+		if (unlikely(test_bit(BTRFS_FS_STATE_ERROR,
+				      &root->fs_info->fs_state)))
+			btrfs_cleanup_transaction(root);
 		if (!try_to_freeze()) {
 			set_current_state(TASK_INTERRUPTIBLE);
 			if (!kthread_should_stop() &&
@@ -2013,50 +2027,28 @@ static void btrfs_stop_all_workers(struct btrfs_fs_info *fs_info)
 	btrfs_stop_workers(&fs_info->qgroup_rescan_workers);
 }
 
+static void free_root_extent_buffers(struct btrfs_root *root)
+{
+	if (root) {
+		free_extent_buffer(root->node);
+		free_extent_buffer(root->commit_root);
+		root->node = NULL;
+		root->commit_root = NULL;
+	}
+}
+
 /* helper to cleanup tree roots */
 static void free_root_pointers(struct btrfs_fs_info *info, int chunk_root)
 {
-	free_extent_buffer(info->tree_root->node);
-	free_extent_buffer(info->tree_root->commit_root);
-	info->tree_root->node = NULL;
-	info->tree_root->commit_root = NULL;
-
-	if (info->dev_root) {
-		free_extent_buffer(info->dev_root->node);
-		free_extent_buffer(info->dev_root->commit_root);
-		info->dev_root->node = NULL;
-		info->dev_root->commit_root = NULL;
-	}
-	if (info->extent_root) {
-		free_extent_buffer(info->extent_root->node);
-		free_extent_buffer(info->extent_root->commit_root);
-		info->extent_root->node = NULL;
-		info->extent_root->commit_root = NULL;
-	}
-	if (info->csum_root) {
-		free_extent_buffer(info->csum_root->node);
-		free_extent_buffer(info->csum_root->commit_root);
-		info->csum_root->node = NULL;
-		info->csum_root->commit_root = NULL;
-	}
-	if (info->quota_root) {
-		free_extent_buffer(info->quota_root->node);
-		free_extent_buffer(info->quota_root->commit_root);
-		info->quota_root->node = NULL;
-		info->quota_root->commit_root = NULL;
-	}
-	if (info->uuid_root) {
-		free_extent_buffer(info->uuid_root->node);
-		free_extent_buffer(info->uuid_root->commit_root);
-		info->uuid_root->node = NULL;
-		info->uuid_root->commit_root = NULL;
-	}
-	if (chunk_root) {
-		free_extent_buffer(info->chunk_root->node);
-		free_extent_buffer(info->chunk_root->commit_root);
-		info->chunk_root->node = NULL;
-		info->chunk_root->commit_root = NULL;
-	}
+	free_root_extent_buffers(info->tree_root);
+
+	free_root_extent_buffers(info->dev_root);
+	free_root_extent_buffers(info->extent_root);
+	free_root_extent_buffers(info->csum_root);
+	free_root_extent_buffers(info->quota_root);
+	free_root_extent_buffers(info->uuid_root);
+	if (chunk_root)
+		free_root_extent_buffers(info->chunk_root);
 }
 
 static void del_fs_roots(struct btrfs_fs_info *fs_info)
@@ -2230,7 +2222,6 @@ int open_ctree(struct super_block *sb,
 	atomic_set(&fs_info->scrubs_paused, 0);
 	atomic_set(&fs_info->scrub_cancel_req, 0);
 	init_waitqueue_head(&fs_info->scrub_pause_wait);
-	init_rwsem(&fs_info->scrub_super_lock);
 	fs_info->scrub_workers_refcnt = 0;
 #ifdef CONFIG_BTRFS_FS_CHECK_INTEGRITY
 	fs_info->check_integrity_print_mask = 0;
@@ -2272,7 +2263,7 @@ int open_ctree(struct super_block *sb,
 	       sizeof(struct btrfs_key));
 	set_bit(BTRFS_INODE_DUMMY,
 		&BTRFS_I(fs_info->btree_inode)->runtime_flags);
-	insert_inode_hash(fs_info->btree_inode);
+	btrfs_insert_inode_hash(fs_info->btree_inode);
 
 	spin_lock_init(&fs_info->block_group_cache_lock);
 	fs_info->block_group_cache_tree = RB_ROOT;
@@ -2670,6 +2661,7 @@ retry_root_backup:
 
 	btrfs_set_root_node(&tree_root->root_item, tree_root->node);
 	tree_root->commit_root = btrfs_root_node(tree_root);
+	btrfs_set_root_refs(&tree_root->root_item, 1);
 
 	location.objectid = BTRFS_EXTENT_TREE_OBJECTID;
 	location.type = BTRFS_ROOT_ITEM_KEY;
@@ -3448,10 +3440,7 @@ static int write_all_supers(struct btrfs_root *root, int max_mirrors)
 int write_ctree_super(struct btrfs_trans_handle *trans,
 		      struct btrfs_root *root, int max_mirrors)
 {
-	int ret;
-
-	ret = write_all_supers(root, max_mirrors);
-	return ret;
+	return write_all_supers(root, max_mirrors);
 }
 
 /* Drop a fs root from the radix tree and free it. */
@@ -3614,12 +3603,12 @@ int close_ctree(struct btrfs_root *root)
 		       percpu_counter_sum(&fs_info->delalloc_bytes));
 	}
 
+	del_fs_roots(fs_info);
+
 	btrfs_free_block_groups(fs_info);
 
 	btrfs_stop_all_workers(fs_info);
 
-	del_fs_roots(fs_info);
-
 	free_root_pointers(fs_info, 1);
 
 	iput(fs_info->btree_inode);
@@ -3669,10 +3658,20 @@ int btrfs_set_buffer_uptodate(struct extent_buffer *buf)
 
 void btrfs_mark_buffer_dirty(struct extent_buffer *buf)
 {
-	struct btrfs_root *root = BTRFS_I(buf->pages[0]->mapping->host)->root;
+	struct btrfs_root *root;
 	u64 transid = btrfs_header_generation(buf);
 	int was_dirty;
 
+#ifdef CONFIG_BTRFS_FS_RUN_SANITY_TESTS
+	/*
+	 * This is a fast path so only do this check if we have sanity tests
+	 * enabled.  Normal people shouldn't be marking dummy buffers as dirty
+	 * outside of the sanity tests.
+	 */
+	if (unlikely(test_bit(EXTENT_BUFFER_DUMMY, &buf->bflags)))
+		return;
+#endif
+	root = BTRFS_I(buf->pages[0]->mapping->host)->root;
 	btrfs_assert_tree_locked(buf);
 	if (transid != root->fs_info->generation)
 		WARN(1, KERN_CRIT "btrfs transid mismatch buffer %llu, "
@@ -3802,7 +3801,8 @@ static void btrfs_destroy_all_ordered_extents(struct btrfs_fs_info *fs_info)
 	while (!list_empty(&splice)) {
 		root = list_first_entry(&splice, struct btrfs_root,
 					ordered_root);
-		list_del_init(&root->ordered_root);
+		list_move_tail(&root->ordered_root,
+			       &fs_info->ordered_roots);
 
 		btrfs_destroy_ordered_extents(root);
 
@@ -3880,24 +3880,6 @@ static int btrfs_destroy_delayed_refs(struct btrfs_transaction *trans,
 	return ret;
 }
 
-static void btrfs_evict_pending_snapshots(struct btrfs_transaction *t)
-{
-	struct btrfs_pending_snapshot *snapshot;
-	struct list_head splice;
-
-	INIT_LIST_HEAD(&splice);
-
-	list_splice_init(&t->pending_snapshots, &splice);
-
-	while (!list_empty(&splice)) {
-		snapshot = list_entry(splice.next,
-				      struct btrfs_pending_snapshot,
-				      list);
-		snapshot->error = -ECANCELED;
-		list_del_init(&snapshot->list);
-	}
-}
-
 static void btrfs_destroy_delalloc_inodes(struct btrfs_root *root)
 {
 	struct btrfs_inode *btrfs_inode;
@@ -4027,15 +4009,13 @@ again:
 void btrfs_cleanup_one_transaction(struct btrfs_transaction *cur_trans,
 				   struct btrfs_root *root)
 {
+	btrfs_destroy_ordered_operations(cur_trans, root);
+
 	btrfs_destroy_delayed_refs(cur_trans, root);
-	btrfs_block_rsv_release(root, &root->fs_info->trans_block_rsv,
-				cur_trans->dirty_pages.dirty_bytes);
 
 	cur_trans->state = TRANS_STATE_COMMIT_START;
 	wake_up(&root->fs_info->transaction_blocked_wait);
 
-	btrfs_evict_pending_snapshots(cur_trans);
-
 	cur_trans->state = TRANS_STATE_UNBLOCKED;
 	wake_up(&root->fs_info->transaction_wait);
 
@@ -4059,63 +4039,51 @@ void btrfs_cleanup_one_transaction(struct btrfs_transaction *cur_trans,
 static int btrfs_cleanup_transaction(struct btrfs_root *root)
 {
 	struct btrfs_transaction *t;
-	LIST_HEAD(list);
 
 	mutex_lock(&root->fs_info->transaction_kthread_mutex);
 
 	spin_lock(&root->fs_info->trans_lock);
-	list_splice_init(&root->fs_info->trans_list, &list);
-	root->fs_info->running_transaction = NULL;
-	spin_unlock(&root->fs_info->trans_lock);
-
-	while (!list_empty(&list)) {
-		t = list_entry(list.next, struct btrfs_transaction, list);
-
-		btrfs_destroy_ordered_operations(t, root);
-
-		btrfs_destroy_all_ordered_extents(root->fs_info);
-
-		btrfs_destroy_delayed_refs(t, root);
-
-		/*
-		 *  FIXME: cleanup wait for commit
-		 *  We needn't acquire the lock here, because we are during
-		 *  the umount, there is no other task which will change it.
-		 */
-		t->state = TRANS_STATE_COMMIT_START;
-		smp_mb();
-		if (waitqueue_active(&root->fs_info->transaction_blocked_wait))
-			wake_up(&root->fs_info->transaction_blocked_wait);
-
-		btrfs_evict_pending_snapshots(t);
-
-		t->state = TRANS_STATE_UNBLOCKED;
-		smp_mb();
-		if (waitqueue_active(&root->fs_info->transaction_wait))
-			wake_up(&root->fs_info->transaction_wait);
-
-		btrfs_destroy_delayed_inodes(root);
-		btrfs_assert_delayed_root_empty(root);
-
-		btrfs_destroy_all_delalloc_inodes(root->fs_info);
-
-		btrfs_destroy_marked_extents(root, &t->dirty_pages,
-					     EXTENT_DIRTY);
-
-		btrfs_destroy_pinned_extent(root,
-					    root->fs_info->pinned_extents);
-
-		t->state = TRANS_STATE_COMPLETED;
-		smp_mb();
-		if (waitqueue_active(&t->commit_wait))
-			wake_up(&t->commit_wait);
+	while (!list_empty(&root->fs_info->trans_list)) {
+		t = list_first_entry(&root->fs_info->trans_list,
+				     struct btrfs_transaction, list);
+		if (t->state >= TRANS_STATE_COMMIT_START) {
+			atomic_inc(&t->use_count);
+			spin_unlock(&root->fs_info->trans_lock);
+			btrfs_wait_for_commit(root, t->transid);
+			btrfs_put_transaction(t);
+			spin_lock(&root->fs_info->trans_lock);
+			continue;
+		}
+		if (t == root->fs_info->running_transaction) {
+			t->state = TRANS_STATE_COMMIT_DOING;
+			spin_unlock(&root->fs_info->trans_lock);
+			/*
+			 * We wait for 0 num_writers since we don't hold a trans
+			 * handle open currently for this transaction.
+			 */
+			wait_event(t->writer_wait,
+				   atomic_read(&t->num_writers) == 0);
+		} else {
+			spin_unlock(&root->fs_info->trans_lock);
+		}
+		btrfs_cleanup_one_transaction(t, root);
 
-		atomic_set(&t->use_count, 0);
+		spin_lock(&root->fs_info->trans_lock);
+		if (t == root->fs_info->running_transaction)
+			root->fs_info->running_transaction = NULL;
 		list_del_init(&t->list);
-		memset(t, 0, sizeof(*t));
-		kmem_cache_free(btrfs_transaction_cachep, t);
-	}
+		spin_unlock(&root->fs_info->trans_lock);
 
+		btrfs_put_transaction(t);
+		trace_btrfs_transaction_commit(root);
+		spin_lock(&root->fs_info->trans_lock);
+	}
+	spin_unlock(&root->fs_info->trans_lock);
+	btrfs_destroy_all_ordered_extents(root->fs_info);
+	btrfs_destroy_delayed_inodes(root);
+	btrfs_assert_delayed_root_empty(root);
+	btrfs_destroy_pinned_extent(root, root->fs_info->pinned_extents);
+	btrfs_destroy_all_delalloc_inodes(root->fs_info);
 	mutex_unlock(&root->fs_info->transaction_kthread_mutex);
 
 	return 0;
diff --git a/fs/btrfs/disk-io.h b/fs/btrfs/disk-io.h
index 5ce2a7da8b11..53059df350f8 100644
--- a/fs/btrfs/disk-io.h
+++ b/fs/btrfs/disk-io.h
@@ -86,6 +86,10 @@ void btrfs_drop_and_free_fs_root(struct btrfs_fs_info *fs_info,
 				 struct btrfs_root *root);
 void btrfs_free_fs_root(struct btrfs_root *root);
 
+#ifdef CONFIG_BTRFS_FS_RUN_SANITY_TESTS
+struct btrfs_root *btrfs_alloc_dummy_root(void);
+#endif
+
 /*
  * This function is used to grab the root, and avoid it is freed when we
  * access it. But it doesn't ensure that the tree is not dropped.
diff --git a/fs/btrfs/export.c b/fs/btrfs/export.c
index 4b8691607373..41422a3de8ed 100644
--- a/fs/btrfs/export.c
+++ b/fs/btrfs/export.c
@@ -5,7 +5,6 @@
 #include "btrfs_inode.h"
 #include "print-tree.h"
 #include "export.h"
-#include "compat.h"
 
 #define BTRFS_FID_SIZE_NON_CONNECTABLE (offsetof(struct btrfs_fid, \
 						 parent_objectid) / 4)
diff --git a/fs/btrfs/extent-tree.c b/fs/btrfs/extent-tree.c
index d58bef130a41..45d98d01028f 100644
--- a/fs/btrfs/extent-tree.c
+++ b/fs/btrfs/extent-tree.c
@@ -25,7 +25,6 @@
 #include <linux/slab.h>
 #include <linux/ratelimit.h>
 #include <linux/percpu_counter.h>
-#include "compat.h"
 #include "hash.h"
 #include "ctree.h"
 #include "disk-io.h"
@@ -1551,9 +1550,8 @@ again:
 	if (ret && !insert) {
 		err = -ENOENT;
 		goto out;
-	} else if (ret) {
+	} else if (WARN_ON(ret)) {
 		err = -EIO;
-		WARN_ON(1);
 		goto out;
 	}
 
@@ -1979,7 +1977,6 @@ static int __btrfs_inc_extent_ref(struct btrfs_trans_handle *trans,
 	struct btrfs_extent_item *item;
 	u64 refs;
 	int ret;
-	int err = 0;
 
 	path = btrfs_alloc_path();
 	if (!path)
@@ -1992,14 +1989,9 @@ static int __btrfs_inc_extent_ref(struct btrfs_trans_handle *trans,
 					   path, bytenr, num_bytes, parent,
 					   root_objectid, owner, offset,
 					   refs_to_add, extent_op);
-	if (ret == 0)
+	if (ret != -EAGAIN)
 		goto out;
 
-	if (ret != -EAGAIN) {
-		err = ret;
-		goto out;
-	}
-
 	leaf = path->nodes[0];
 	item = btrfs_item_ptr(leaf, path->slots[0], struct btrfs_extent_item);
 	refs = btrfs_extent_refs(leaf, item);
@@ -2021,7 +2013,7 @@ static int __btrfs_inc_extent_ref(struct btrfs_trans_handle *trans,
 		btrfs_abort_transaction(trans, root, ret);
 out:
 	btrfs_free_path(path);
-	return err;
+	return ret;
 }
 
 static int run_delayed_data_ref(struct btrfs_trans_handle *trans,
@@ -2137,15 +2129,28 @@ again:
 	}
 	if (ret > 0) {
 		if (metadata) {
-			btrfs_release_path(path);
-			metadata = 0;
+			if (path->slots[0] > 0) {
+				path->slots[0]--;
+				btrfs_item_key_to_cpu(path->nodes[0], &key,
+						      path->slots[0]);
+				if (key.objectid == node->bytenr &&
+				    key.type == BTRFS_EXTENT_ITEM_KEY &&
+				    key.offset == node->num_bytes)
+					ret = 0;
+			}
+			if (ret > 0) {
+				btrfs_release_path(path);
+				metadata = 0;
 
-			key.offset = node->num_bytes;
-			key.type = BTRFS_EXTENT_ITEM_KEY;
-			goto again;
+				key.objectid = node->bytenr;
+				key.offset = node->num_bytes;
+				key.type = BTRFS_EXTENT_ITEM_KEY;
+				goto again;
+			}
+		} else {
+			err = -EIO;
+			goto out;
 		}
-		err = -EIO;
-		goto out;
 	}
 
 	leaf = path->nodes[0];
@@ -2234,8 +2239,12 @@ static int run_one_delayed_ref(struct btrfs_trans_handle *trans,
 {
 	int ret = 0;
 
-	if (trans->aborted)
+	if (trans->aborted) {
+		if (insert_reserved)
+			btrfs_pin_extent(root, node->bytenr,
+					 node->num_bytes, 1);
 		return 0;
+	}
 
 	if (btrfs_delayed_ref_is_head(node)) {
 		struct btrfs_delayed_ref_head *head;
@@ -2411,6 +2420,14 @@ static noinline int run_clustered_refs(struct btrfs_trans_handle *trans,
 				btrfs_free_delayed_extent_op(extent_op);
 
 				if (ret) {
+					/*
+					 * Need to reset must_insert_reserved if
+					 * there was an error so the abort stuff
+					 * can cleanup the reserved space
+					 * properly.
+					 */
+					if (must_insert_reserved)
+						locked_ref->must_insert_reserved = 1;
 					btrfs_debug(fs_info, "run_delayed_extent_op returned %d", ret);
 					spin_lock(&delayed_refs->lock);
 					btrfs_delayed_ref_unlock(locked_ref);
@@ -3197,8 +3214,7 @@ again:
 		if (ret)
 			goto out_put;
 
-		ret = btrfs_truncate_free_space_cache(root, trans, path,
-						      inode);
+		ret = btrfs_truncate_free_space_cache(root, trans, inode);
 		if (ret)
 			goto out_put;
 	}
@@ -3318,10 +3334,9 @@ again:
 		last = cache->key.objectid + cache->key.offset;
 
 		err = write_one_cache_group(trans, root, path, cache);
+		btrfs_put_block_group(cache);
 		if (err) /* File system offline */
 			goto out;
-
-		btrfs_put_block_group(cache);
 	}
 
 	while (1) {
@@ -3605,10 +3620,9 @@ int btrfs_check_data_free_space(struct inode *inode, u64 bytes)
 	/* make sure bytes are sectorsize aligned */
 	bytes = ALIGN(bytes, root->sectorsize);
 
-	if (root == root->fs_info->tree_root ||
-	    BTRFS_I(inode)->location.objectid == BTRFS_FREE_INO_OBJECTID) {
-		alloc_chunk = 0;
+	if (btrfs_is_free_space_inode(inode)) {
 		committed = 1;
+		ASSERT(current->journal_info);
 	}
 
 	data_sinfo = fs_info->data_sinfo;
@@ -3636,6 +3650,16 @@ again:
 			spin_unlock(&data_sinfo->lock);
 alloc:
 			alloc_target = btrfs_get_alloc_profile(root, 1);
+			/*
+			 * It is ugly that we don't call nolock join
+			 * transaction for the free space inode case here.
+			 * But it is safe because we only do the data space
+			 * reservation for the free space cache in the
+			 * transaction context, the common join transaction
+			 * just increase the counter of the current transaction
+			 * handler, doesn't try to acquire the trans_lock of
+			 * the fs.
+			 */
 			trans = btrfs_join_transaction(root);
 			if (IS_ERR(trans))
 				return PTR_ERR(trans);
@@ -3681,6 +3705,9 @@ commit_trans:
 			goto again;
 		}
 
+		trace_btrfs_space_reservation(root->fs_info,
+					      "space_info:enospc",
+					      data_sinfo->flags, bytes, 1);
 		return -ENOSPC;
 	}
 	data_sinfo->bytes_may_use += bytes;
@@ -3989,12 +4016,26 @@ static void btrfs_writeback_inodes_sb_nr(struct btrfs_root *root,
 		 * the filesystem is readonly(all dirty pages are written to
 		 * the disk).
 		 */
-		btrfs_start_all_delalloc_inodes(root->fs_info, 0);
+		btrfs_start_delalloc_roots(root->fs_info, 0);
 		if (!current->journal_info)
-			btrfs_wait_all_ordered_extents(root->fs_info);
+			btrfs_wait_ordered_roots(root->fs_info, -1);
 	}
 }
 
+static inline int calc_reclaim_items_nr(struct btrfs_root *root, u64 to_reclaim)
+{
+	u64 bytes;
+	int nr;
+
+	bytes = btrfs_calc_trans_metadata_size(root, 1);
+	nr = (int)div64_u64(to_reclaim, bytes);
+	if (!nr)
+		nr = 1;
+	return nr;
+}
+
+#define EXTENT_SIZE_PER_ITEM	(256 * 1024)
+
 /*
  * shrink metadata reservation for delalloc
  */
@@ -4007,24 +4048,30 @@ static void shrink_delalloc(struct btrfs_root *root, u64 to_reclaim, u64 orig,
 	u64 delalloc_bytes;
 	u64 max_reclaim;
 	long time_left;
-	unsigned long nr_pages = (2 * 1024 * 1024) >> PAGE_CACHE_SHIFT;
-	int loops = 0;
+	unsigned long nr_pages;
+	int loops;
+	int items;
 	enum btrfs_reserve_flush_enum flush;
 
+	/* Calc the number of the pages we need flush for space reservation */
+	items = calc_reclaim_items_nr(root, to_reclaim);
+	to_reclaim = items * EXTENT_SIZE_PER_ITEM;
+
 	trans = (struct btrfs_trans_handle *)current->journal_info;
 	block_rsv = &root->fs_info->delalloc_block_rsv;
 	space_info = block_rsv->space_info;
 
-	smp_mb();
 	delalloc_bytes = percpu_counter_sum_positive(
 						&root->fs_info->delalloc_bytes);
 	if (delalloc_bytes == 0) {
 		if (trans)
 			return;
-		btrfs_wait_all_ordered_extents(root->fs_info);
+		if (wait_ordered)
+			btrfs_wait_ordered_roots(root->fs_info, items);
 		return;
 	}
 
+	loops = 0;
 	while (delalloc_bytes && loops < 3) {
 		max_reclaim = min(delalloc_bytes, to_reclaim);
 		nr_pages = max_reclaim >> PAGE_CACHE_SHIFT;
@@ -4033,9 +4080,19 @@ static void shrink_delalloc(struct btrfs_root *root, u64 to_reclaim, u64 orig,
 		 * We need to wait for the async pages to actually start before
 		 * we do anything.
 		 */
-		wait_event(root->fs_info->async_submit_wait,
-			   !atomic_read(&root->fs_info->async_delalloc_pages));
+		max_reclaim = atomic_read(&root->fs_info->async_delalloc_pages);
+		if (!max_reclaim)
+			goto skip_async;
+
+		if (max_reclaim <= nr_pages)
+			max_reclaim = 0;
+		else
+			max_reclaim -= nr_pages;
 
+		wait_event(root->fs_info->async_submit_wait,
+			   atomic_read(&root->fs_info->async_delalloc_pages) <=
+			   (int)max_reclaim);
+skip_async:
 		if (!trans)
 			flush = BTRFS_RESERVE_FLUSH_ALL;
 		else
@@ -4049,13 +4106,12 @@ static void shrink_delalloc(struct btrfs_root *root, u64 to_reclaim, u64 orig,
 
 		loops++;
 		if (wait_ordered && !trans) {
-			btrfs_wait_all_ordered_extents(root->fs_info);
+			btrfs_wait_ordered_roots(root->fs_info, items);
 		} else {
 			time_left = schedule_timeout_killable(1);
 			if (time_left)
 				break;
 		}
-		smp_mb();
 		delalloc_bytes = percpu_counter_sum_positive(
 						&root->fs_info->delalloc_bytes);
 	}
@@ -4140,16 +4196,11 @@ static int flush_space(struct btrfs_root *root,
 	switch (state) {
 	case FLUSH_DELAYED_ITEMS_NR:
 	case FLUSH_DELAYED_ITEMS:
-		if (state == FLUSH_DELAYED_ITEMS_NR) {
-			u64 bytes = btrfs_calc_trans_metadata_size(root, 1);
-
-			nr = (int)div64_u64(num_bytes, bytes);
-			if (!nr)
-				nr = 1;
-			nr *= 2;
-		} else {
+		if (state == FLUSH_DELAYED_ITEMS_NR)
+			nr = calc_reclaim_items_nr(root, num_bytes) * 2;
+		else
 			nr = -1;
-		}
+
 		trans = btrfs_join_transaction(root);
 		if (IS_ERR(trans)) {
 			ret = PTR_ERR(trans);
@@ -4332,6 +4383,10 @@ out:
 		    !block_rsv_use_bytes(global_rsv, orig_bytes))
 			ret = 0;
 	}
+	if (ret == -ENOSPC)
+		trace_btrfs_space_reservation(root->fs_info,
+					      "space_info:enospc",
+					      space_info->flags, orig_bytes, 1);
 	if (flushing) {
 		spin_lock(&space_info->lock);
 		space_info->flush = 0;
@@ -4986,7 +5041,7 @@ int btrfs_delalloc_reserve_metadata(struct inode *inode, u64 num_bytes)
 		mutex_unlock(&BTRFS_I(inode)->delalloc_mutex);
 
 	if (to_reserve)
-		trace_btrfs_space_reservation(root->fs_info,"delalloc",
+		trace_btrfs_space_reservation(root->fs_info, "delalloc",
 					      btrfs_ino(inode), to_reserve, 1);
 	block_rsv_add_bytes(block_rsv, to_reserve, 1);
 
@@ -5264,6 +5319,8 @@ static int pin_down_extent(struct btrfs_root *root,
 
 	set_extent_dirty(root->fs_info->pinned_extents, bytenr,
 			 bytenr + num_bytes - 1, GFP_NOFS | __GFP_NOFAIL);
+	if (reserved)
+		trace_btrfs_reserved_extent_free(root, bytenr, num_bytes);
 	return 0;
 }
 
@@ -5718,9 +5775,8 @@ static int __btrfs_free_extent(struct btrfs_trans_handle *trans,
 			}
 			extent_slot = path->slots[0];
 		}
-	} else if (ret == -ENOENT) {
+	} else if (WARN_ON(ret == -ENOENT)) {
 		btrfs_print_leaf(extent_root, path->nodes[0]);
-		WARN_ON(1);
 		btrfs_err(info,
 			"unable to find ref byte nr %llu parent %llu root %llu  owner %llu offset %llu",
 			bytenr, parent, root_objectid, owner_objectid,
@@ -5967,6 +6023,7 @@ void btrfs_free_tree_block(struct btrfs_trans_handle *trans,
 
 		btrfs_add_free_space(cache, buf->start, buf->len);
 		btrfs_update_reserved_bytes(cache, buf->len, RESERVE_FREE);
+		trace_btrfs_reserved_extent_free(root, buf->start, buf->len);
 		pin = 0;
 	}
 out:
@@ -6594,8 +6651,6 @@ again:
 		}
 	}
 
-	trace_btrfs_reserved_extent_alloc(root, ins->objectid, ins->offset);
-
 	return ret;
 }
 
@@ -6707,6 +6762,7 @@ static int alloc_reserved_file_extent(struct btrfs_trans_handle *trans,
 			ins->objectid, ins->offset);
 		BUG();
 	}
+	trace_btrfs_reserved_extent_alloc(root, ins->objectid, ins->offset);
 	return ret;
 }
 
@@ -6731,13 +6787,18 @@ static int alloc_reserved_tree_block(struct btrfs_trans_handle *trans,
 		size += sizeof(*block_info);
 
 	path = btrfs_alloc_path();
-	if (!path)
+	if (!path) {
+		btrfs_free_and_pin_reserved_extent(root, ins->objectid,
+						   root->leafsize);
 		return -ENOMEM;
+	}
 
 	path->leave_spinning = 1;
 	ret = btrfs_insert_empty_item(trans, fs_info->extent_root, path,
 				      ins, size);
 	if (ret) {
+		btrfs_free_and_pin_reserved_extent(root, ins->objectid,
+						   root->leafsize);
 		btrfs_free_path(path);
 		return ret;
 	}
@@ -6779,6 +6840,8 @@ static int alloc_reserved_tree_block(struct btrfs_trans_handle *trans,
 			ins->objectid, ins->offset);
 		BUG();
 	}
+
+	trace_btrfs_reserved_extent_alloc(root, ins->objectid, root->leafsize);
 	return ret;
 }
 
@@ -7983,7 +8046,7 @@ u64 btrfs_account_ro_block_groups_free_space(struct btrfs_space_info *sinfo)
 
 	spin_lock(&sinfo->lock);
 
-	for(i = 0; i < BTRFS_NR_RAID_TYPES; i++)
+	for (i = 0; i < BTRFS_NR_RAID_TYPES; i++)
 		if (!list_empty(&sinfo->block_groups[i]))
 			free_bytes += __btrfs_get_ro_block_group_free_space(
 						&sinfo->block_groups[i]);
@@ -8271,15 +8334,14 @@ int btrfs_free_block_groups(struct btrfs_fs_info *info)
 
 	release_global_block_rsv(info);
 
-	while(!list_empty(&info->space_info)) {
+	while (!list_empty(&info->space_info)) {
 		space_info = list_entry(info->space_info.next,
 					struct btrfs_space_info,
 					list);
 		if (btrfs_test_opt(info->tree_root, ENOSPC_DEBUG)) {
-			if (space_info->bytes_pinned > 0 ||
+			if (WARN_ON(space_info->bytes_pinned > 0 ||
 			    space_info->bytes_reserved > 0 ||
-			    space_info->bytes_may_use > 0) {
-				WARN_ON(1);
+			    space_info->bytes_may_use > 0)) {
 				dump_space_info(space_info, 0, 0);
 			}
 		}
diff --git a/fs/btrfs/extent_io.c b/fs/btrfs/extent_io.c
index 51731b76900d..856bc2b2192c 100644
--- a/fs/btrfs/extent_io.c
+++ b/fs/btrfs/extent_io.c
@@ -13,13 +13,13 @@
 #include <linux/cleancache.h>
 #include "extent_io.h"
 #include "extent_map.h"
-#include "compat.h"
 #include "ctree.h"
 #include "btrfs_inode.h"
 #include "volumes.h"
 #include "check-integrity.h"
 #include "locking.h"
 #include "rcu-string.h"
+#include "backref.h"
 
 static struct kmem_cache *extent_state_cache;
 static struct kmem_cache *extent_buffer_cache;
@@ -1597,11 +1597,10 @@ done:
  *
  * 1 is returned if we find something, 0 if nothing was in the tree
  */
-static noinline u64 find_lock_delalloc_range(struct inode *inode,
-					     struct extent_io_tree *tree,
-					     struct page *locked_page,
-					     u64 *start, u64 *end,
-					     u64 max_bytes)
+STATIC u64 find_lock_delalloc_range(struct inode *inode,
+				    struct extent_io_tree *tree,
+				    struct page *locked_page, u64 *start,
+				    u64 *end, u64 max_bytes)
 {
 	u64 delalloc_start;
 	u64 delalloc_end;
@@ -1740,10 +1739,8 @@ u64 count_range_bits(struct extent_io_tree *tree,
 	u64 last = 0;
 	int found = 0;
 
-	if (search_end <= cur_start) {
-		WARN_ON(1);
+	if (WARN_ON(search_end <= cur_start))
 		return 0;
-	}
 
 	spin_lock(&tree->lock);
 	if (cur_start == 0 && bits == EXTENT_DIRTY) {
@@ -3569,9 +3566,8 @@ retry:
 			 * but no sense in crashing the users box for something
 			 * we can survive anyway.
 			 */
-			if (!eb) {
+			if (WARN_ON(!eb)) {
 				spin_unlock(&mapping->private_lock);
-				WARN_ON(1);
 				continue;
 			}
 
@@ -4038,7 +4034,7 @@ static struct extent_map *get_extent_skip_holes(struct inode *inode,
 	if (offset >= last)
 		return NULL;
 
-	while(1) {
+	while (1) {
 		len = last - offset;
 		if (len == 0)
 			break;
@@ -4062,6 +4058,19 @@ static struct extent_map *get_extent_skip_holes(struct inode *inode,
 	return NULL;
 }
 
+static noinline int count_ext_ref(u64 inum, u64 offset, u64 root_id, void *ctx)
+{
+	unsigned long cnt = *((unsigned long *)ctx);
+
+	cnt++;
+	*((unsigned long *)ctx) = cnt;
+
+	/* Now we're sure that the extent is shared. */
+	if (cnt > 1)
+		return 1;
+	return 0;
+}
+
 int extent_fiemap(struct inode *inode, struct fiemap_extent_info *fieinfo,
 		__u64 start, __u64 len, get_extent_t *get_extent)
 {
@@ -4128,7 +4137,7 @@ int extent_fiemap(struct inode *inode, struct fiemap_extent_info *fieinfo,
 		last = found_key.offset;
 		last_for_get_extent = last + 1;
 	}
-	btrfs_free_path(path);
+	btrfs_release_path(path);
 
 	/*
 	 * we might have some extents allocated but more delalloc past those
@@ -4198,7 +4207,24 @@ int extent_fiemap(struct inode *inode, struct fiemap_extent_info *fieinfo,
 			flags |= (FIEMAP_EXTENT_DELALLOC |
 				  FIEMAP_EXTENT_UNKNOWN);
 		} else {
+			unsigned long ref_cnt = 0;
+
 			disko = em->block_start + offset_in_extent;
+
+			/*
+			 * As btrfs supports shared space, this information
+			 * can be exported to userspace tools via
+			 * flag FIEMAP_EXTENT_SHARED.
+			 */
+			ret = iterate_inodes_from_logical(
+					em->block_start,
+					BTRFS_I(inode)->root->fs_info,
+					path, count_ext_ref, &ref_cnt);
+			if (ret < 0 && ret != -ENOENT)
+				goto out_free;
+
+			if (ref_cnt > 1)
+				flags |= FIEMAP_EXTENT_SHARED;
 		}
 		if (test_bit(EXTENT_FLAG_COMPRESSED, &em->flags))
 			flags |= FIEMAP_EXTENT_ENCODED;
@@ -4230,6 +4256,7 @@ int extent_fiemap(struct inode *inode, struct fiemap_extent_info *fieinfo,
 out_free:
 	free_extent_map(em);
 out:
+	btrfs_free_path(path);
 	unlock_extent_cached(&BTRFS_I(inode)->io_tree, start, start + len - 1,
 			     &cached_state, GFP_NOFS);
 	return ret;
@@ -4455,6 +4482,23 @@ static void mark_extent_buffer_accessed(struct extent_buffer *eb)
 	}
 }
 
+struct extent_buffer *find_extent_buffer(struct extent_io_tree *tree,
+					 		u64 start)
+{
+	struct extent_buffer *eb;
+
+	rcu_read_lock();
+	eb = radix_tree_lookup(&tree->buffer, start >> PAGE_CACHE_SHIFT);
+	if (eb && atomic_inc_not_zero(&eb->refs)) {
+		rcu_read_unlock();
+		mark_extent_buffer_accessed(eb);
+		return eb;
+	}
+	rcu_read_unlock();
+
+	return NULL;
+}
+
 struct extent_buffer *alloc_extent_buffer(struct extent_io_tree *tree,
 					  u64 start, unsigned long len)
 {
@@ -4468,14 +4512,10 @@ struct extent_buffer *alloc_extent_buffer(struct extent_io_tree *tree,
 	int uptodate = 1;
 	int ret;
 
-	rcu_read_lock();
-	eb = radix_tree_lookup(&tree->buffer, start >> PAGE_CACHE_SHIFT);
-	if (eb && atomic_inc_not_zero(&eb->refs)) {
-		rcu_read_unlock();
-		mark_extent_buffer_accessed(eb);
+
+	eb = find_extent_buffer(tree, start);
+	if (eb)
 		return eb;
-	}
-	rcu_read_unlock();
 
 	eb = __alloc_extent_buffer(tree, start, len, GFP_NOFS);
 	if (!eb)
@@ -4534,24 +4574,17 @@ again:
 
 	spin_lock(&tree->buffer_lock);
 	ret = radix_tree_insert(&tree->buffer, start >> PAGE_CACHE_SHIFT, eb);
+	spin_unlock(&tree->buffer_lock);
+	radix_tree_preload_end();
 	if (ret == -EEXIST) {
-		exists = radix_tree_lookup(&tree->buffer,
-						start >> PAGE_CACHE_SHIFT);
-		if (!atomic_inc_not_zero(&exists->refs)) {
-			spin_unlock(&tree->buffer_lock);
-			radix_tree_preload_end();
-			exists = NULL;
+		exists = find_extent_buffer(tree, start);
+		if (exists)
+			goto free_eb;
+		else
 			goto again;
-		}
-		spin_unlock(&tree->buffer_lock);
-		radix_tree_preload_end();
-		mark_extent_buffer_accessed(exists);
-		goto free_eb;
 	}
 	/* add one reference for the tree */
 	check_buffer_tree_ref(eb);
-	spin_unlock(&tree->buffer_lock);
-	radix_tree_preload_end();
 
 	/*
 	 * there is a race where release page may have
@@ -4582,23 +4615,6 @@ free_eb:
 	return exists;
 }
 
-struct extent_buffer *find_extent_buffer(struct extent_io_tree *tree,
-					 u64 start, unsigned long len)
-{
-	struct extent_buffer *eb;
-
-	rcu_read_lock();
-	eb = radix_tree_lookup(&tree->buffer, start >> PAGE_CACHE_SHIFT);
-	if (eb && atomic_inc_not_zero(&eb->refs)) {
-		rcu_read_unlock();
-		mark_extent_buffer_accessed(eb);
-		return eb;
-	}
-	rcu_read_unlock();
-
-	return NULL;
-}
-
 static inline void btrfs_release_extent_buffer_rcu(struct rcu_head *head)
 {
 	struct extent_buffer *eb =
@@ -5062,23 +5078,6 @@ void copy_extent_buffer(struct extent_buffer *dst, struct extent_buffer *src,
 	}
 }
 
-static void move_pages(struct page *dst_page, struct page *src_page,
-		       unsigned long dst_off, unsigned long src_off,
-		       unsigned long len)
-{
-	char *dst_kaddr = page_address(dst_page);
-	if (dst_page == src_page) {
-		memmove(dst_kaddr + dst_off, dst_kaddr + src_off, len);
-	} else {
-		char *src_kaddr = page_address(src_page);
-		char *p = dst_kaddr + dst_off + len;
-		char *s = src_kaddr + src_off + len;
-
-		while (len--)
-			*--p = *--s;
-	}
-}
-
 static inline bool areas_overlap(unsigned long src, unsigned long dst, unsigned long len)
 {
 	unsigned long distance = (src > dst) ? src - dst : dst - src;
@@ -5189,7 +5188,7 @@ void memmove_extent_buffer(struct extent_buffer *dst, unsigned long dst_offset,
 
 		cur = min_t(unsigned long, len, src_off_in_page + 1);
 		cur = min(cur, dst_off_in_page + 1);
-		move_pages(extent_buffer_page(dst, dst_i),
+		copy_pages(extent_buffer_page(dst, dst_i),
 			   extent_buffer_page(dst, src_i),
 			   dst_off_in_page - cur + 1,
 			   src_off_in_page - cur + 1, cur);
diff --git a/fs/btrfs/extent_io.h b/fs/btrfs/extent_io.h
index 6dbc645f1f3d..19620c58f096 100644
--- a/fs/btrfs/extent_io.h
+++ b/fs/btrfs/extent_io.h
@@ -271,7 +271,7 @@ struct extent_buffer *alloc_extent_buffer(struct extent_io_tree *tree,
 struct extent_buffer *alloc_dummy_extent_buffer(u64 start, unsigned long len);
 struct extent_buffer *btrfs_clone_extent_buffer(struct extent_buffer *src);
 struct extent_buffer *find_extent_buffer(struct extent_io_tree *tree,
-					 u64 start, unsigned long len);
+					 u64 start);
 void free_extent_buffer(struct extent_buffer *eb);
 void free_extent_buffer_stale(struct extent_buffer *eb);
 #define WAIT_NONE	0
@@ -345,4 +345,10 @@ int repair_io_failure(struct btrfs_fs_info *fs_info, u64 start,
 int end_extent_writepage(struct page *page, int err, u64 start, u64 end);
 int repair_eb_io_failure(struct btrfs_root *root, struct extent_buffer *eb,
 			 int mirror_num);
+#ifdef CONFIG_BTRFS_FS_RUN_SANITY_TESTS
+noinline u64 find_lock_delalloc_range(struct inode *inode,
+				      struct extent_io_tree *tree,
+				      struct page *locked_page, u64 *start,
+				      u64 *end, u64 max_bytes);
+#endif
 #endif
diff --git a/fs/btrfs/extent_map.h b/fs/btrfs/extent_map.h
index 61adc44b7805..93fba716d7f8 100644
--- a/fs/btrfs/extent_map.h
+++ b/fs/btrfs/extent_map.h
@@ -3,10 +3,10 @@
 
 #include <linux/rbtree.h>
 
-#define EXTENT_MAP_LAST_BYTE (u64)-4
-#define EXTENT_MAP_HOLE (u64)-3
-#define EXTENT_MAP_INLINE (u64)-2
-#define EXTENT_MAP_DELALLOC (u64)-1
+#define EXTENT_MAP_LAST_BYTE ((u64)-4)
+#define EXTENT_MAP_HOLE ((u64)-3)
+#define EXTENT_MAP_INLINE ((u64)-2)
+#define EXTENT_MAP_DELALLOC ((u64)-1)
 
 /* bits for the flags field */
 #define EXTENT_FLAG_PINNED 0 /* this entry not yet on disk, don't free it */
diff --git a/fs/btrfs/file-item.c b/fs/btrfs/file-item.c
index 4f53159bdb9d..6f3848860283 100644
--- a/fs/btrfs/file-item.c
+++ b/fs/btrfs/file-item.c
@@ -329,6 +329,9 @@ int btrfs_lookup_csums_range(struct btrfs_root *root, u64 start, u64 end,
 	u64 csum_end;
 	u16 csum_size = btrfs_super_csum_size(root->fs_info->super_copy);
 
+	ASSERT(start == ALIGN(start, root->sectorsize) &&
+	       (end + 1) == ALIGN(end + 1, root->sectorsize));
+
 	path = btrfs_alloc_path();
 	if (!path)
 		return -ENOMEM;
@@ -846,10 +849,8 @@ insert:
 	path->leave_spinning = 0;
 	if (ret < 0)
 		goto fail_unlock;
-	if (ret != 0) {
-		WARN_ON(1);
+	if (WARN_ON(ret != 0))
 		goto fail_unlock;
-	}
 	leaf = path->nodes[0];
 csum:
 	item = btrfs_item_ptr(leaf, path->slots[0], struct btrfs_csum_item);
diff --git a/fs/btrfs/file.c b/fs/btrfs/file.c
index 72da4df53c9a..82d0342763c5 100644
--- a/fs/btrfs/file.c
+++ b/fs/btrfs/file.c
@@ -39,7 +39,6 @@
 #include "print-tree.h"
 #include "tree-log.h"
 #include "locking.h"
-#include "compat.h"
 #include "volumes.h"
 
 static struct kmem_cache *btrfs_inode_defrag_cachep;
@@ -370,7 +369,7 @@ int btrfs_run_defrag_inodes(struct btrfs_fs_info *fs_info)
 	u64 root_objectid = 0;
 
 	atomic_inc(&fs_info->defrag_running);
-	while(1) {
+	while (1) {
 		/* Pause the auto defragger. */
 		if (test_bit(BTRFS_FS_STATE_REMOUNTING,
 			     &fs_info->fs_state))
@@ -1281,6 +1280,7 @@ again:
 		}
 		wait_on_page_writeback(pages[i]);
 	}
+	faili = num_pages - 1;
 	err = 0;
 	if (start_pos < inode->i_size) {
 		struct btrfs_ordered_extent *ordered;
@@ -1299,8 +1299,10 @@ again:
 				unlock_page(pages[i]);
 				page_cache_release(pages[i]);
 			}
-			btrfs_wait_ordered_range(inode, start_pos,
-						 last_pos - start_pos);
+			err = btrfs_wait_ordered_range(inode, start_pos,
+						       last_pos - start_pos);
+			if (err)
+				goto fail;
 			goto again;
 		}
 		if (ordered)
@@ -1809,8 +1811,13 @@ int btrfs_sync_file(struct file *file, loff_t start, loff_t end, int datasync)
 	atomic_inc(&root->log_batch);
 	full_sync = test_bit(BTRFS_INODE_NEEDS_FULL_SYNC,
 			     &BTRFS_I(inode)->runtime_flags);
-	if (full_sync)
-		btrfs_wait_ordered_range(inode, start, end - start + 1);
+	if (full_sync) {
+		ret = btrfs_wait_ordered_range(inode, start, end - start + 1);
+		if (ret) {
+			mutex_unlock(&inode->i_mutex);
+			goto out;
+		}
+	}
 	atomic_inc(&root->log_batch);
 
 	/*
@@ -1876,27 +1883,20 @@ int btrfs_sync_file(struct file *file, loff_t start, loff_t end, int datasync)
 	mutex_unlock(&inode->i_mutex);
 
 	if (ret != BTRFS_NO_LOG_SYNC) {
-		if (ret > 0) {
-			/*
-			 * If we didn't already wait for ordered extents we need
-			 * to do that now.
-			 */
-			if (!full_sync)
-				btrfs_wait_ordered_range(inode, start,
-							 end - start + 1);
-			ret = btrfs_commit_transaction(trans, root);
-		} else {
+		if (!ret) {
 			ret = btrfs_sync_log(trans, root);
-			if (ret == 0) {
+			if (!ret) {
 				ret = btrfs_end_transaction(trans, root);
-			} else {
-				if (!full_sync)
-					btrfs_wait_ordered_range(inode, start,
-								 end -
-								 start + 1);
-				ret = btrfs_commit_transaction(trans, root);
+				goto out;
 			}
 		}
+		if (!full_sync) {
+			ret = btrfs_wait_ordered_range(inode, start,
+						       end - start + 1);
+			if (ret)
+				goto out;
+		}
+		ret = btrfs_commit_transaction(trans, root);
 	} else {
 		ret = btrfs_end_transaction(trans, root);
 	}
@@ -2067,7 +2067,9 @@ static int btrfs_punch_hole(struct inode *inode, loff_t offset, loff_t len)
 	bool same_page = ((offset >> PAGE_CACHE_SHIFT) ==
 			  ((offset + len - 1) >> PAGE_CACHE_SHIFT));
 
-	btrfs_wait_ordered_range(inode, offset, len);
+	ret = btrfs_wait_ordered_range(inode, offset, len);
+	if (ret)
+		return ret;
 
 	mutex_lock(&inode->i_mutex);
 	/*
@@ -2136,8 +2138,12 @@ static int btrfs_punch_hole(struct inode *inode, loff_t offset, loff_t len)
 			btrfs_put_ordered_extent(ordered);
 		unlock_extent_cached(&BTRFS_I(inode)->io_tree, lockstart,
 				     lockend, &cached_state, GFP_NOFS);
-		btrfs_wait_ordered_range(inode, lockstart,
-					 lockend - lockstart + 1);
+		ret = btrfs_wait_ordered_range(inode, lockstart,
+					       lockend - lockstart + 1);
+		if (ret) {
+			mutex_unlock(&inode->i_mutex);
+			return ret;
+		}
 	}
 
 	path = btrfs_alloc_path();
@@ -2308,7 +2314,10 @@ static long btrfs_fallocate(struct file *file, int mode,
 	 * wait for ordered IO before we have any locks.  We'll loop again
 	 * below with the locks held.
 	 */
-	btrfs_wait_ordered_range(inode, alloc_start, alloc_end - alloc_start);
+	ret = btrfs_wait_ordered_range(inode, alloc_start,
+				       alloc_end - alloc_start);
+	if (ret)
+		goto out;
 
 	locked_end = alloc_end - 1;
 	while (1) {
@@ -2332,8 +2341,10 @@ static long btrfs_fallocate(struct file *file, int mode,
 			 * we can't wait on the range with the transaction
 			 * running or with the extent lock held
 			 */
-			btrfs_wait_ordered_range(inode, alloc_start,
-						 alloc_end - alloc_start);
+			ret = btrfs_wait_ordered_range(inode, alloc_start,
+						       alloc_end - alloc_start);
+			if (ret)
+				goto out;
 		} else {
 			if (ordered)
 				btrfs_put_ordered_extent(ordered);
@@ -2405,14 +2416,12 @@ out_reserve_fail:
 static int find_desired_extent(struct inode *inode, loff_t *offset, int whence)
 {
 	struct btrfs_root *root = BTRFS_I(inode)->root;
-	struct extent_map *em;
+	struct extent_map *em = NULL;
 	struct extent_state *cached_state = NULL;
 	u64 lockstart = *offset;
 	u64 lockend = i_size_read(inode);
 	u64 start = *offset;
-	u64 orig_start = *offset;
 	u64 len = i_size_read(inode);
-	u64 last_end = 0;
 	int ret = 0;
 
 	lockend = max_t(u64, root->sectorsize, lockend);
@@ -2429,89 +2438,35 @@ static int find_desired_extent(struct inode *inode, loff_t *offset, int whence)
 	lock_extent_bits(&BTRFS_I(inode)->io_tree, lockstart, lockend, 0,
 			 &cached_state);
 
-	/*
-	 * Delalloc is such a pain.  If we have a hole and we have pending
-	 * delalloc for a portion of the hole we will get back a hole that
-	 * exists for the entire range since it hasn't been actually written
-	 * yet.  So to take care of this case we need to look for an extent just
-	 * before the position we want in case there is outstanding delalloc
-	 * going on here.
-	 */
-	if (whence == SEEK_HOLE && start != 0) {
-		if (start <= root->sectorsize)
-			em = btrfs_get_extent_fiemap(inode, NULL, 0, 0,
-						     root->sectorsize, 0);
-		else
-			em = btrfs_get_extent_fiemap(inode, NULL, 0,
-						     start - root->sectorsize,
-						     root->sectorsize, 0);
-		if (IS_ERR(em)) {
-			ret = PTR_ERR(em);
-			goto out;
-		}
-		last_end = em->start + em->len;
-		if (em->block_start == EXTENT_MAP_DELALLOC)
-			last_end = min_t(u64, last_end, inode->i_size);
-		free_extent_map(em);
-	}
-
-	while (1) {
+	while (start < inode->i_size) {
 		em = btrfs_get_extent_fiemap(inode, NULL, 0, start, len, 0);
 		if (IS_ERR(em)) {
 			ret = PTR_ERR(em);
+			em = NULL;
 			break;
 		}
 
-		if (em->block_start == EXTENT_MAP_HOLE) {
-			if (test_bit(EXTENT_FLAG_VACANCY, &em->flags)) {
-				if (last_end <= orig_start) {
-					free_extent_map(em);
-					ret = -ENXIO;
-					break;
-				}
-			}
-
-			if (whence == SEEK_HOLE) {
-				*offset = start;
-				free_extent_map(em);
-				break;
-			}
-		} else {
-			if (whence == SEEK_DATA) {
-				if (em->block_start == EXTENT_MAP_DELALLOC) {
-					if (start >= inode->i_size) {
-						free_extent_map(em);
-						ret = -ENXIO;
-						break;
-					}
-				}
-
-				if (!test_bit(EXTENT_FLAG_PREALLOC,
-					      &em->flags)) {
-					*offset = start;
-					free_extent_map(em);
-					break;
-				}
-			}
-		}
+		if (whence == SEEK_HOLE &&
+		    (em->block_start == EXTENT_MAP_HOLE ||
+		     test_bit(EXTENT_FLAG_PREALLOC, &em->flags)))
+			break;
+		else if (whence == SEEK_DATA &&
+			   (em->block_start != EXTENT_MAP_HOLE &&
+			    !test_bit(EXTENT_FLAG_PREALLOC, &em->flags)))
+			break;
 
 		start = em->start + em->len;
-		last_end = em->start + em->len;
-
-		if (em->block_start == EXTENT_MAP_DELALLOC)
-			last_end = min_t(u64, last_end, inode->i_size);
-
-		if (test_bit(EXTENT_FLAG_VACANCY, &em->flags)) {
-			free_extent_map(em);
-			ret = -ENXIO;
-			break;
-		}
 		free_extent_map(em);
+		em = NULL;
 		cond_resched();
 	}
-	if (!ret)
-		*offset = min(*offset, inode->i_size);
-out:
+	free_extent_map(em);
+	if (!ret) {
+		if (whence == SEEK_DATA && start >= inode->i_size)
+			ret = -ENXIO;
+		else
+			*offset = min_t(loff_t, start, inode->i_size);
+	}
 	unlock_extent_cached(&BTRFS_I(inode)->io_tree, lockstart, lockend,
 			     &cached_state, GFP_NOFS);
 	return ret;
diff --git a/fs/btrfs/free-space-cache.c b/fs/btrfs/free-space-cache.c
index b4f9904c4c6b..057be95b1e1e 100644
--- a/fs/btrfs/free-space-cache.c
+++ b/fs/btrfs/free-space-cache.c
@@ -218,7 +218,6 @@ int btrfs_check_trunc_cache_free_space(struct btrfs_root *root,
 
 int btrfs_truncate_free_space_cache(struct btrfs_root *root,
 				    struct btrfs_trans_handle *trans,
-				    struct btrfs_path *path,
 				    struct inode *inode)
 {
 	int ret = 0;
@@ -1009,8 +1008,13 @@ static int __btrfs_write_out_cache(struct btrfs_root *root, struct inode *inode,
 	if (ret)
 		goto out;
 
-
-	btrfs_wait_ordered_range(inode, 0, (u64)-1);
+	ret = btrfs_wait_ordered_range(inode, 0, (u64)-1);
+	if (ret) {
+		clear_extent_bit(&BTRFS_I(inode)->io_tree, 0, inode->i_size - 1,
+				 EXTENT_DIRTY | EXTENT_DELALLOC, 0, 0, NULL,
+				 GFP_NOFS);
+		goto out;
+	}
 
 	key.objectid = BTRFS_FREE_SPACE_OBJECTID;
 	key.offset = offset;
@@ -2276,7 +2280,7 @@ u64 btrfs_alloc_from_cluster(struct btrfs_block_group_cache *block_group,
 		goto out;
 
 	entry = rb_entry(node, struct btrfs_free_space, offset_index);
-	while(1) {
+	while (1) {
 		if (entry->bytes < bytes && entry->bytes > *max_extent_size)
 			*max_extent_size = entry->bytes;
 
@@ -2967,19 +2971,15 @@ out:
 
 int btrfs_write_out_ino_cache(struct btrfs_root *root,
 			      struct btrfs_trans_handle *trans,
-			      struct btrfs_path *path)
+			      struct btrfs_path *path,
+			      struct inode *inode)
 {
 	struct btrfs_free_space_ctl *ctl = root->free_ino_ctl;
-	struct inode *inode;
 	int ret;
 
 	if (!btrfs_test_opt(root, INODE_MAP_CACHE))
 		return 0;
 
-	inode = lookup_free_ino_inode(root, path);
-	if (IS_ERR(inode))
-		return 0;
-
 	ret = __btrfs_write_out_cache(root, inode, ctl, NULL, trans, path, 0);
 	if (ret) {
 		btrfs_delalloc_release_metadata(inode, inode->i_size);
@@ -2990,7 +2990,6 @@ int btrfs_write_out_ino_cache(struct btrfs_root *root,
 #endif
 	}
 
-	iput(inode);
 	return ret;
 }
 
diff --git a/fs/btrfs/free-space-cache.h b/fs/btrfs/free-space-cache.h
index e737f92cf6d0..0cf4977ef70d 100644
--- a/fs/btrfs/free-space-cache.h
+++ b/fs/btrfs/free-space-cache.h
@@ -58,7 +58,6 @@ int btrfs_check_trunc_cache_free_space(struct btrfs_root *root,
 				       struct btrfs_block_rsv *rsv);
 int btrfs_truncate_free_space_cache(struct btrfs_root *root,
 				    struct btrfs_trans_handle *trans,
-				    struct btrfs_path *path,
 				    struct inode *inode);
 int load_free_space_cache(struct btrfs_fs_info *fs_info,
 			  struct btrfs_block_group_cache *block_group);
@@ -76,7 +75,8 @@ int load_free_ino_cache(struct btrfs_fs_info *fs_info,
 			struct btrfs_root *root);
 int btrfs_write_out_ino_cache(struct btrfs_root *root,
 			      struct btrfs_trans_handle *trans,
-			      struct btrfs_path *path);
+			      struct btrfs_path *path,
+			      struct inode *inode);
 
 void btrfs_init_free_space_ctl(struct btrfs_block_group_cache *block_group);
 int __btrfs_add_free_space(struct btrfs_free_space_ctl *ctl,
diff --git a/fs/btrfs/inode-item.c b/fs/btrfs/inode-item.c
index e0b7034d6343..ec82fae07097 100644
--- a/fs/btrfs/inode-item.c
+++ b/fs/btrfs/inode-item.c
@@ -369,7 +369,7 @@ static int btrfs_insert_inode_extref(struct btrfs_trans_handle *trans,
 		goto out;
 
 	leaf = path->nodes[0];
-	item = btrfs_item_nr(leaf, path->slots[0]);
+	item = btrfs_item_nr(path->slots[0]);
 	ptr = (unsigned long)btrfs_item_ptr(leaf, path->slots[0], char);
 	ptr += btrfs_item_size(leaf, item) - ins_len;
 	extref = (struct btrfs_inode_extref *)ptr;
diff --git a/fs/btrfs/inode-map.c b/fs/btrfs/inode-map.c
index 2c66ddbbe670..ab485e57b6fe 100644
--- a/fs/btrfs/inode-map.c
+++ b/fs/btrfs/inode-map.c
@@ -78,10 +78,8 @@ again:
 			    btrfs_transaction_in_commit(fs_info)) {
 				leaf = path->nodes[0];
 
-				if (btrfs_header_nritems(leaf) == 0) {
-					WARN_ON(1);
+				if (WARN_ON(btrfs_header_nritems(leaf) == 0))
 					break;
-				}
 
 				/*
 				 * Save the key so we can advances forward
@@ -237,7 +235,7 @@ again:
 		start_caching(root);
 
 		if (objectid <= root->cache_progress ||
-		    objectid > root->highest_objectid)
+		    objectid >= root->highest_objectid)
 			__btrfs_add_free_space(ctl, objectid, 1);
 		else
 			__btrfs_add_free_space(pinned, objectid, 1);
@@ -412,8 +410,7 @@ int btrfs_save_ino_cache(struct btrfs_root *root,
 		return 0;
 
 	/* Don't save inode cache if we are deleting this root */
-	if (btrfs_root_refs(&root->root_item) == 0 &&
-	    root != root->fs_info->tree_root)
+	if (btrfs_root_refs(&root->root_item) == 0)
 		return 0;
 
 	if (!btrfs_test_opt(root, INODE_MAP_CACHE))
@@ -467,7 +464,7 @@ again:
 	}
 
 	if (i_size_read(inode) > 0) {
-		ret = btrfs_truncate_free_space_cache(root, trans, path, inode);
+		ret = btrfs_truncate_free_space_cache(root, trans, inode);
 		if (ret) {
 			if (ret != -ENOSPC)
 				btrfs_abort_transaction(trans, root, ret);
@@ -504,7 +501,7 @@ again:
 	}
 	btrfs_free_reserved_data_space(inode, prealloc);
 
-	ret = btrfs_write_out_ino_cache(root, trans, path);
+	ret = btrfs_write_out_ino_cache(root, trans, path, inode);
 out_put:
 	iput(inode);
 out_release:
diff --git a/fs/btrfs/inode.c b/fs/btrfs/inode.c
index 51e3afa78354..da8d2f696ac5 100644
--- a/fs/btrfs/inode.c
+++ b/fs/btrfs/inode.c
@@ -43,7 +43,6 @@
 #include <linux/btrfs.h>
 #include <linux/blkdev.h>
 #include <linux/posix_acl_xattr.h>
-#include "compat.h"
 #include "ctree.h"
 #include "disk-io.h"
 #include "transaction.h"
@@ -844,7 +843,10 @@ static noinline int cow_file_range(struct inode *inode,
 	struct extent_map_tree *em_tree = &BTRFS_I(inode)->extent_tree;
 	int ret = 0;
 
-	BUG_ON(btrfs_is_free_space_inode(inode));
+	if (btrfs_is_free_space_inode(inode)) {
+		WARN_ON_ONCE(1);
+		return -EINVAL;
+	}
 
 	num_bytes = ALIGN(end - start + 1, blocksize);
 	num_bytes = max(blocksize,  num_bytes);
@@ -1178,10 +1180,8 @@ static noinline int run_delalloc_nocow(struct inode *inode,
 	while (1) {
 		ret = btrfs_lookup_file_extent(trans, root, path, ino,
 					       cur_offset, 0);
-		if (ret < 0) {
-			btrfs_abort_transaction(trans, root, ret);
+		if (ret < 0)
 			goto error;
-		}
 		if (ret > 0 && path->slots[0] > 0 && check_prev) {
 			leaf = path->nodes[0];
 			btrfs_item_key_to_cpu(leaf, &found_key,
@@ -1195,10 +1195,8 @@ next_slot:
 		leaf = path->nodes[0];
 		if (path->slots[0] >= btrfs_header_nritems(leaf)) {
 			ret = btrfs_next_leaf(root, path);
-			if (ret < 0) {
-				btrfs_abort_transaction(trans, root, ret);
+			if (ret < 0)
 				goto error;
-			}
 			if (ret > 0)
 				break;
 			leaf = path->nodes[0];
@@ -1289,10 +1287,8 @@ out_check:
 			ret = cow_file_range(inode, locked_page,
 					     cow_start, found_key.offset - 1,
 					     page_started, nr_written, 1);
-			if (ret) {
-				btrfs_abort_transaction(trans, root, ret);
+			if (ret)
 				goto error;
-			}
 			cow_start = (u64)-1;
 		}
 
@@ -1339,10 +1335,8 @@ out_check:
 		    BTRFS_DATA_RELOC_TREE_OBJECTID) {
 			ret = btrfs_reloc_clone_csums(inode, cur_offset,
 						      num_bytes);
-			if (ret) {
-				btrfs_abort_transaction(trans, root, ret);
+			if (ret)
 				goto error;
-			}
 		}
 
 		extent_clear_unlock_delalloc(inode, cur_offset,
@@ -1364,10 +1358,8 @@ out_check:
 	if (cow_start != (u64)-1) {
 		ret = cow_file_range(inode, locked_page, cow_start, end,
 				     page_started, nr_written, 1);
-		if (ret) {
-			btrfs_abort_transaction(trans, root, ret);
+		if (ret)
 			goto error;
-		}
 	}
 
 error:
@@ -1551,7 +1543,13 @@ static void btrfs_clear_bit_hook(struct inode *inode,
 			spin_unlock(&BTRFS_I(inode)->lock);
 		}
 
-		if (*bits & EXTENT_DO_ACCOUNTING)
+		/*
+		 * We don't reserve metadata space for space cache inodes so we
+		 * don't need to call dellalloc_release_metadata if there is an
+		 * error.
+		 */
+		if (*bits & EXTENT_DO_ACCOUNTING &&
+		    root != root->fs_info->tree_root)
 			btrfs_delalloc_release_metadata(inode, len);
 
 		if (root->root_key.objectid != BTRFS_DATA_RELOC_TREE_OBJECTID
@@ -2041,10 +2039,8 @@ static noinline int record_one_backref(u64 inum, u64 offset, u64 root_id,
 		key.offset = offset;
 
 	ret = btrfs_search_slot(NULL, root, &key, path, 0, 0);
-	if (ret < 0) {
-		WARN_ON(1);
+	if (WARN_ON(ret < 0))
 		return ret;
-	}
 	ret = 0;
 
 	while (1) {
@@ -2367,10 +2363,23 @@ out_unlock:
 	return ret;
 }
 
+static void free_sa_defrag_extent(struct new_sa_defrag_extent *new)
+{
+	struct old_sa_defrag_extent *old, *tmp;
+
+	if (!new)
+		return;
+
+	list_for_each_entry_safe(old, tmp, &new->head, list) {
+		list_del(&old->list);
+		kfree(old);
+	}
+	kfree(new);
+}
+
 static void relink_file_extents(struct new_sa_defrag_extent *new)
 {
 	struct btrfs_path *path;
-	struct old_sa_defrag_extent *old, *tmp;
 	struct sa_defrag_extent_backref *backref;
 	struct sa_defrag_extent_backref *prev = NULL;
 	struct inode *inode;
@@ -2413,16 +2422,11 @@ static void relink_file_extents(struct new_sa_defrag_extent *new)
 	kfree(prev);
 
 	btrfs_free_path(path);
-
-	list_for_each_entry_safe(old, tmp, &new->head, list) {
-		list_del(&old->list);
-		kfree(old);
-	}
 out:
+	free_sa_defrag_extent(new);
+
 	atomic_dec(&root->fs_info->defrag_running);
 	wake_up(&root->fs_info->transaction_wait);
-
-	kfree(new);
 }
 
 static struct new_sa_defrag_extent *
@@ -2432,7 +2436,7 @@ record_old_file_extents(struct inode *inode,
 	struct btrfs_root *root = BTRFS_I(inode)->root;
 	struct btrfs_path *path;
 	struct btrfs_key key;
-	struct old_sa_defrag_extent *old, *tmp;
+	struct old_sa_defrag_extent *old;
 	struct new_sa_defrag_extent *new;
 	int ret;
 
@@ -2480,7 +2484,7 @@ record_old_file_extents(struct inode *inode,
 		if (slot >= btrfs_header_nritems(l)) {
 			ret = btrfs_next_leaf(root, path);
 			if (ret < 0)
-				goto out_free_list;
+				goto out_free_path;
 			else if (ret > 0)
 				break;
 			continue;
@@ -2509,7 +2513,7 @@ record_old_file_extents(struct inode *inode,
 
 		old = kmalloc(sizeof(*old), GFP_NOFS);
 		if (!old)
-			goto out_free_list;
+			goto out_free_path;
 
 		offset = max(new->file_pos, key.offset);
 		end = min(new->file_pos + new->len, key.offset + num_bytes);
@@ -2531,15 +2535,10 @@ next:
 
 	return new;
 
-out_free_list:
-	list_for_each_entry_safe(old, tmp, &new->head, list) {
-		list_del(&old->list);
-		kfree(old);
-	}
 out_free_path:
 	btrfs_free_path(path);
 out_kfree:
-	kfree(new);
+	free_sa_defrag_extent(new);
 	return NULL;
 }
 
@@ -2710,8 +2709,14 @@ out:
 	btrfs_remove_ordered_extent(inode, ordered_extent);
 
 	/* for snapshot-aware defrag */
-	if (new)
-		relink_file_extents(new);
+	if (new) {
+		if (ret) {
+			free_sa_defrag_extent(new);
+			atomic_dec(&root->fs_info->defrag_running);
+		} else {
+			relink_file_extents(new);
+		}
+	}
 
 	/* once for us */
 	btrfs_put_ordered_extent(ordered_extent);
@@ -2969,6 +2974,7 @@ int btrfs_orphan_add(struct btrfs_trans_handle *trans, struct inode *inode)
 	if (insert >= 1) {
 		ret = btrfs_insert_orphan_item(trans, root, btrfs_ino(inode));
 		if (ret) {
+			atomic_dec(&root->orphan_inodes);
 			if (reserve) {
 				clear_bit(BTRFS_INODE_ORPHAN_META_RESERVED,
 					  &BTRFS_I(inode)->runtime_flags);
@@ -3018,14 +3024,16 @@ static int btrfs_orphan_del(struct btrfs_trans_handle *trans,
 		release_rsv = 1;
 	spin_unlock(&root->orphan_lock);
 
-	if (trans && delete_item)
-		ret = btrfs_del_orphan_item(trans, root, btrfs_ino(inode));
-
-	if (release_rsv) {
-		btrfs_orphan_release_metadata(inode);
+	if (delete_item) {
 		atomic_dec(&root->orphan_inodes);
+		if (trans)
+			ret = btrfs_del_orphan_item(trans, root,
+						    btrfs_ino(inode));
 	}
 
+	if (release_rsv)
+		btrfs_orphan_release_metadata(inode);
+
 	return ret;
 }
 
@@ -3172,8 +3180,7 @@ int btrfs_orphan_cleanup(struct btrfs_root *root)
 
 		/* if we have links, this was a truncate, lets do that */
 		if (inode->i_nlink) {
-			if (!S_ISREG(inode->i_mode)) {
-				WARN_ON(1);
+			if (WARN_ON(!S_ISREG(inode->i_mode))) {
 				iput(inode);
 				continue;
 			}
@@ -3636,7 +3643,7 @@ int btrfs_unlink_inode(struct btrfs_trans_handle *trans,
 	int ret;
 	ret = __btrfs_unlink_inode(trans, root, dir, inode, name, name_len);
 	if (!ret) {
-		btrfs_drop_nlink(inode);
+		drop_nlink(inode);
 		ret = btrfs_update_inode(trans, root, inode);
 	}
 	return ret;
@@ -4230,15 +4237,16 @@ int btrfs_cont_expand(struct inode *inode, loff_t oldsize, loff_t size)
 
 	while (1) {
 		struct btrfs_ordered_extent *ordered;
-		btrfs_wait_ordered_range(inode, hole_start,
-					 block_end - hole_start);
+
 		lock_extent_bits(io_tree, hole_start, block_end - 1, 0,
 				 &cached_state);
-		ordered = btrfs_lookup_ordered_extent(inode, hole_start);
+		ordered = btrfs_lookup_ordered_range(inode, hole_start,
+						     block_end - hole_start);
 		if (!ordered)
 			break;
 		unlock_extent_cached(io_tree, hole_start, block_end - 1,
 				     &cached_state, GFP_NOFS);
+		btrfs_start_ordered_extent(inode, ordered, 1);
 		btrfs_put_ordered_extent(ordered);
 	}
 
@@ -4472,8 +4480,10 @@ void btrfs_evict_inode(struct inode *inode)
 	trace_btrfs_inode_evict(inode);
 
 	truncate_inode_pages(&inode->i_data, 0);
-	if (inode->i_nlink && (btrfs_root_refs(&root->root_item) != 0 ||
-			       btrfs_is_free_space_inode(inode)))
+	if (inode->i_nlink &&
+	    ((btrfs_root_refs(&root->root_item) != 0 &&
+	      root->root_key.objectid != BTRFS_ROOT_TREE_OBJECTID) ||
+	     btrfs_is_free_space_inode(inode)))
 		goto no_delete;
 
 	if (is_bad_inode(inode)) {
@@ -4490,7 +4500,8 @@ void btrfs_evict_inode(struct inode *inode)
 	}
 
 	if (inode->i_nlink > 0) {
-		BUG_ON(btrfs_root_refs(&root->root_item) != 0);
+		BUG_ON(btrfs_root_refs(&root->root_item) != 0 &&
+		       root->root_key.objectid != BTRFS_ROOT_TREE_OBJECTID);
 		goto no_delete;
 	}
 
@@ -4731,14 +4742,7 @@ static void inode_tree_del(struct inode *inode)
 	}
 	spin_unlock(&root->inode_lock);
 
-	/*
-	 * Free space cache has inodes in the tree root, but the tree root has a
-	 * root_refs of 0, so this could end up dropping the tree root as a
-	 * snapshot, so we need the extra !root->fs_info->tree_root check to
-	 * make sure we don't drop it.
-	 */
-	if (empty && btrfs_root_refs(&root->root_item) == 0 &&
-	    root != root->fs_info->tree_root) {
+	if (empty && btrfs_root_refs(&root->root_item) == 0) {
 		synchronize_srcu(&root->fs_info->subvol_srcu);
 		spin_lock(&root->inode_lock);
 		empty = RB_EMPTY_ROOT(&root->inode_tree);
@@ -4831,10 +4835,12 @@ static struct inode *btrfs_iget_locked(struct super_block *s,
 {
 	struct inode *inode;
 	struct btrfs_iget_args args;
+	unsigned long hashval = btrfs_inode_hash(objectid, root);
+
 	args.ino = objectid;
 	args.root = root;
 
-	inode = iget5_locked(s, objectid, btrfs_find_actor,
+	inode = iget5_locked(s, hashval, btrfs_find_actor,
 			     btrfs_init_locked_inode,
 			     (void *)&args);
 	return inode;
@@ -5048,7 +5054,7 @@ static int btrfs_real_readdir(struct file *file, struct dir_context *ctx)
 			continue;
 		}
 
-		item = btrfs_item_nr(leaf, slot);
+		item = btrfs_item_nr(slot);
 		btrfs_item_key_to_cpu(leaf, &found_key, slot);
 
 		if (found_key.objectid != key.objectid)
@@ -5454,7 +5460,7 @@ static struct inode *btrfs_new_inode(struct btrfs_trans_handle *trans,
 				BTRFS_INODE_NODATASUM;
 	}
 
-	insert_inode_hash(inode);
+	btrfs_insert_inode_hash(inode);
 	inode_tree_add(inode);
 
 	trace_btrfs_inode_new(inode);
@@ -5730,7 +5736,7 @@ static int btrfs_link(struct dentry *old_dentry, struct inode *dir,
 		goto fail;
 	}
 
-	btrfs_inc_nlink(inode);
+	inc_nlink(inode);
 	inode_inc_iversion(inode);
 	inode->i_ctime = CURRENT_TIME;
 	ihold(inode);
@@ -5860,7 +5866,7 @@ static noinline int uncompress_inline(struct btrfs_path *path,
 	compress_type = btrfs_file_extent_compression(leaf, item);
 	max_size = btrfs_file_extent_ram_bytes(leaf, item);
 	inline_size = btrfs_file_extent_inline_item_len(leaf,
-					btrfs_item_nr(leaf, path->slots[0]));
+					btrfs_item_nr(path->slots[0]));
 	tmp = kmalloc(inline_size, GFP_NOFS);
 	if (!tmp)
 		return -ENOMEM;
@@ -5974,7 +5980,14 @@ again:
 	found_type = btrfs_key_type(&found_key);
 	if (found_key.objectid != objectid ||
 	    found_type != BTRFS_EXTENT_DATA_KEY) {
-		goto not_found;
+		/*
+		 * If we backup past the first extent we want to move forward
+		 * and see if there is an extent in front of us, otherwise we'll
+		 * say there is a hole for our whole search range which can
+		 * cause problems.
+		 */
+		extent_end = start;
+		goto next;
 	}
 
 	found_type = btrfs_file_extent_type(leaf, item);
@@ -5989,7 +6002,7 @@ again:
 		size = btrfs_file_extent_inline_len(leaf, item);
 		extent_end = ALIGN(extent_start + size, root->sectorsize);
 	}
-
+next:
 	if (start >= extent_end) {
 		path->slots[0]++;
 		if (path->slots[0] >= btrfs_header_nritems(leaf)) {
@@ -6249,7 +6262,7 @@ struct extent_map *btrfs_get_extent_fiemap(struct inode *inode, struct page *pag
 	/* adjust the range_start to make sure it doesn't
 	 * go backwards from the start they passed in
 	 */
-	range_start = max(start,range_start);
+	range_start = max(start, range_start);
 	found = found_end - range_start;
 
 	if (found > 0) {
@@ -7053,7 +7066,7 @@ static int btrfs_submit_direct_hook(int rw, struct btrfs_dio_private *dip,
 			}
 		} else {
 			submit_len += bvec->bv_len;
-			nr_pages ++;
+			nr_pages++;
 			bvec++;
 		}
 	}
@@ -7222,7 +7235,9 @@ static ssize_t btrfs_direct_IO(int rw, struct kiocb *iocb,
 	 * outstanding dirty pages are on disk.
 	 */
 	count = iov_length(iov, nr_segs);
-	btrfs_wait_ordered_range(inode, offset, count);
+	ret = btrfs_wait_ordered_range(inode, offset, count);
+	if (ret)
+		return ret;
 
 	if (rw & WRITE) {
 		/*
@@ -7563,7 +7578,10 @@ static int btrfs_truncate(struct inode *inode)
 	u64 mask = root->sectorsize - 1;
 	u64 min_size = btrfs_calc_trunc_metadata_size(root, 1);
 
-	btrfs_wait_ordered_range(inode, inode->i_size & (~mask), (u64)-1);
+	ret = btrfs_wait_ordered_range(inode, inode->i_size & (~mask),
+				       (u64)-1);
+	if (ret)
+		return ret;
 
 	/*
 	 * Yes ladies and gentelment, this is indeed ugly.  The fact is we have
@@ -7787,6 +7805,14 @@ struct inode *btrfs_alloc_inode(struct super_block *sb)
 	return inode;
 }
 
+#ifdef CONFIG_BTRFS_FS_RUN_SANITY_TESTS
+void btrfs_test_destroy_inode(struct inode *inode)
+{
+	btrfs_drop_extent_cache(inode, 0, (u64)-1, 0);
+	kmem_cache_free(btrfs_inode_cachep, BTRFS_I(inode));
+}
+#endif
+
 static void btrfs_i_callback(struct rcu_head *head)
 {
 	struct inode *inode = container_of(head, struct inode, i_rcu);
@@ -7857,8 +7883,7 @@ int btrfs_drop_inode(struct inode *inode)
 		return 1;
 
 	/* the snap/subvol tree is on deleting */
-	if (btrfs_root_refs(&root->root_item) == 0 &&
-	    root != root->fs_info->tree_root)
+	if (btrfs_root_refs(&root->root_item) == 0)
 		return 1;
 	else
 		return generic_drop_inode(inode);
@@ -7995,8 +8020,7 @@ static int btrfs_rename(struct inode *old_dir, struct dentry *old_dentry,
 		if (ret == -EEXIST) {
 			/* we shouldn't get
 			 * eexist without a new_inode */
-			if (!new_inode) {
-				WARN_ON(1);
+			if (WARN_ON(!new_inode)) {
 				return ret;
 			}
 		} else {
@@ -8144,18 +8168,24 @@ out_notrans:
 static void btrfs_run_delalloc_work(struct btrfs_work *work)
 {
 	struct btrfs_delalloc_work *delalloc_work;
+	struct inode *inode;
 
 	delalloc_work = container_of(work, struct btrfs_delalloc_work,
 				     work);
-	if (delalloc_work->wait)
-		btrfs_wait_ordered_range(delalloc_work->inode, 0, (u64)-1);
-	else
-		filemap_flush(delalloc_work->inode->i_mapping);
+	inode = delalloc_work->inode;
+	if (delalloc_work->wait) {
+		btrfs_wait_ordered_range(inode, 0, (u64)-1);
+	} else {
+		filemap_flush(inode->i_mapping);
+		if (test_bit(BTRFS_INODE_HAS_ASYNC_EXTENT,
+			     &BTRFS_I(inode)->runtime_flags))
+			filemap_flush(inode->i_mapping);
+	}
 
 	if (delalloc_work->delay_iput)
-		btrfs_add_delayed_iput(delalloc_work->inode);
+		btrfs_add_delayed_iput(inode);
 	else
-		iput(delalloc_work->inode);
+		iput(inode);
 	complete(&delalloc_work->completion);
 }
 
@@ -8276,8 +8306,7 @@ int btrfs_start_delalloc_inodes(struct btrfs_root *root, int delay_iput)
 	return ret;
 }
 
-int btrfs_start_all_delalloc_inodes(struct btrfs_fs_info *fs_info,
-				    int delay_iput)
+int btrfs_start_delalloc_roots(struct btrfs_fs_info *fs_info, int delay_iput)
 {
 	struct btrfs_root *root;
 	struct list_head splice;
@@ -8337,14 +8366,14 @@ static int btrfs_symlink(struct inode *dir, struct dentry *dentry,
 	int err;
 	int drop_inode = 0;
 	u64 objectid;
-	u64 index = 0 ;
+	u64 index = 0;
 	int name_len;
 	int datasize;
 	unsigned long ptr;
 	struct btrfs_file_extent_item *ei;
 	struct extent_buffer *leaf;
 
-	name_len = strlen(symname) + 1;
+	name_len = strlen(symname);
 	if (name_len > BTRFS_MAX_INLINE_DATA_SIZE(root))
 		return -ENAMETOOLONG;
 
@@ -8432,7 +8461,7 @@ static int btrfs_symlink(struct inode *dir, struct dentry *dentry,
 	inode->i_mapping->a_ops = &btrfs_symlink_aops;
 	inode->i_mapping->backing_dev_info = &root->fs_info->bdi;
 	inode_set_bytes(inode, name_len);
-	btrfs_i_size_write(inode, name_len - 1);
+	btrfs_i_size_write(inode, name_len);
 	err = btrfs_update_inode(trans, root, inode);
 	if (err)
 		drop_inode = 1;
@@ -8491,6 +8520,8 @@ static int __btrfs_prealloc_file_range(struct inode *inode, int mode,
 						  ins.offset, 0, 0, 0,
 						  BTRFS_FILE_EXTENT_PREALLOC);
 		if (ret) {
+			btrfs_free_reserved_extent(root, ins.objectid,
+						   ins.offset);
 			btrfs_abort_transaction(trans, root, ret);
 			if (own_trans)
 				btrfs_end_transaction(trans, root);
diff --git a/fs/btrfs/ioctl.c b/fs/btrfs/ioctl.c
index 9d46f60cb943..a111622598b0 100644
--- a/fs/btrfs/ioctl.c
+++ b/fs/btrfs/ioctl.c
@@ -44,7 +44,6 @@
 #include <linux/uuid.h>
 #include <linux/btrfs.h>
 #include <linux/uaccess.h>
-#include "compat.h"
 #include "ctree.h"
 #include "disk-io.h"
 #include "transaction.h"
@@ -321,7 +320,7 @@ static int btrfs_ioctl_getversion(struct file *file, int __user *arg)
 
 static noinline int btrfs_ioctl_fitrim(struct file *file, void __user *arg)
 {
-	struct btrfs_fs_info *fs_info = btrfs_sb(fdentry(file)->d_sb);
+	struct btrfs_fs_info *fs_info = btrfs_sb(file_inode(file)->i_sb);
 	struct btrfs_device *device;
 	struct request_queue *q;
 	struct fstrim_range range;
@@ -369,9 +368,13 @@ static noinline int btrfs_ioctl_fitrim(struct file *file, void __user *arg)
 
 int btrfs_is_empty_uuid(u8 *uuid)
 {
-	static char empty_uuid[BTRFS_UUID_SIZE] = {0};
+	int i;
 
-	return !memcmp(uuid, empty_uuid, BTRFS_UUID_SIZE);
+	for (i = 0; i < BTRFS_UUID_SIZE; i++) {
+		if (uuid[i])
+			return 0;
+	}
+	return 1;
 }
 
 static noinline int create_subvol(struct inode *dir,
@@ -436,7 +439,7 @@ static noinline int create_subvol(struct inode *dir,
 	btrfs_set_header_backref_rev(leaf, BTRFS_MIXED_BACKREF_REV);
 	btrfs_set_header_owner(leaf, objectid);
 
-	write_extent_buffer(leaf, root->fs_info->fsid, btrfs_header_fsid(leaf),
+	write_extent_buffer(leaf, root->fs_info->fsid, btrfs_header_fsid(),
 			    BTRFS_FSID_SIZE);
 	write_extent_buffer(leaf, root->fs_info->chunk_tree_uuid,
 			    btrfs_header_chunk_tree_uuid(leaf),
@@ -574,7 +577,7 @@ static int create_snapshot(struct btrfs_root *root, struct inode *dir,
 	if (ret)
 		return ret;
 
-	btrfs_wait_ordered_extents(root);
+	btrfs_wait_ordered_extents(root, -1);
 
 	pending_snapshot = kzalloc(sizeof(*pending_snapshot), GFP_NOFS);
 	if (!pending_snapshot)
@@ -688,7 +691,7 @@ static inline int btrfs_check_sticky(struct inode *dir, struct inode *inode)
  *     nfs_async_unlink().
  */
 
-static int btrfs_may_delete(struct inode *dir,struct dentry *victim,int isdir)
+static int btrfs_may_delete(struct inode *dir, struct dentry *victim, int isdir)
 {
 	int error;
 
@@ -842,7 +845,6 @@ static int find_new_extents(struct btrfs_root *root,
 {
 	struct btrfs_path *path;
 	struct btrfs_key min_key;
-	struct btrfs_key max_key;
 	struct extent_buffer *leaf;
 	struct btrfs_file_extent_item *extent;
 	int type;
@@ -857,15 +859,10 @@ static int find_new_extents(struct btrfs_root *root,
 	min_key.type = BTRFS_EXTENT_DATA_KEY;
 	min_key.offset = *off;
 
-	max_key.objectid = ino;
-	max_key.type = (u8)-1;
-	max_key.offset = (u64)-1;
-
 	path->keep_locks = 1;
 
-	while(1) {
-		ret = btrfs_search_forward(root, &min_key, &max_key,
-					   path, newer_than);
+	while (1) {
+		ret = btrfs_search_forward(root, &min_key, path, newer_than);
 		if (ret != 0)
 			goto none;
 		if (min_key.objectid != ino)
@@ -1206,7 +1203,7 @@ int btrfs_defrag_file(struct inode *inode, struct file *file,
 		ra = &file->f_ra;
 	}
 
-	pages = kmalloc(sizeof(struct page *) * max_cluster,
+	pages = kmalloc_array(max_cluster, sizeof(struct page *),
 			GFP_NOFS);
 	if (!pages) {
 		ret = -ENOMEM;
@@ -1893,7 +1890,6 @@ static noinline int search_ioctl(struct inode *inode,
 {
 	struct btrfs_root *root;
 	struct btrfs_key key;
-	struct btrfs_key max_key;
 	struct btrfs_path *path;
 	struct btrfs_ioctl_search_key *sk = &args->key;
 	struct btrfs_fs_info *info = BTRFS_I(inode)->root->fs_info;
@@ -1925,15 +1921,10 @@ static noinline int search_ioctl(struct inode *inode,
 	key.type = sk->min_type;
 	key.offset = sk->min_offset;
 
-	max_key.objectid = sk->max_objectid;
-	max_key.type = sk->max_type;
-	max_key.offset = sk->max_offset;
-
 	path->keep_locks = 1;
 
-	while(1) {
-		ret = btrfs_search_forward(root, &key, &max_key, path,
-					   sk->min_transid);
+	while (1) {
+		ret = btrfs_search_forward(root, &key, path, sk->min_transid);
 		if (ret != 0) {
 			if (ret > 0)
 				ret = 0;
@@ -2018,7 +2009,7 @@ static noinline int btrfs_search_path_in_tree(struct btrfs_fs_info *info,
 	key.type = BTRFS_INODE_REF_KEY;
 	key.offset = (u64)-1;
 
-	while(1) {
+	while (1) {
 		ret = btrfs_search_slot(NULL, root, &key, path, 0, 0);
 		if (ret < 0)
 			goto out;
@@ -2047,7 +2038,7 @@ static noinline int btrfs_search_path_in_tree(struct btrfs_fs_info *info,
 		}
 
 		*(ptr + len) = '/';
-		read_extent_buffer(l, ptr,(unsigned long)(iref + 1), len);
+		read_extent_buffer(l, ptr, (unsigned long)(iref + 1), len);
 
 		if (key.offset == BTRFS_FIRST_FREE_OBJECTID)
 			break;
@@ -2058,7 +2049,7 @@ static noinline int btrfs_search_path_in_tree(struct btrfs_fs_info *info,
 		dirid = key.objectid;
 	}
 	memmove(name, ptr, total_len);
-	name[total_len]='\0';
+	name[total_len] = '\0';
 	ret = 0;
 out:
 	btrfs_free_path(path);
@@ -2098,7 +2089,7 @@ static noinline int btrfs_ioctl_ino_lookup(struct file *file,
 static noinline int btrfs_ioctl_snap_destroy(struct file *file,
 					     void __user *arg)
 {
-	struct dentry *parent = fdentry(file);
+	struct dentry *parent = file->f_path.dentry;
 	struct dentry *dentry;
 	struct inode *dir = parent->d_inode;
 	struct inode *inode;
@@ -2144,7 +2135,7 @@ static noinline int btrfs_ioctl_snap_destroy(struct file *file,
 
 	inode = dentry->d_inode;
 	dest = BTRFS_I(inode)->root;
-	if (!capable(CAP_SYS_ADMIN)){
+	if (!capable(CAP_SYS_ADMIN)) {
 		/*
 		 * Regular user.  Only allow this with a special mount
 		 * option, when the user has write+exec access to the
@@ -2727,15 +2718,10 @@ static long btrfs_ioctl_file_extent_same(struct file *file,
 	size = sizeof(tmp) +
 		tmp.dest_count * sizeof(struct btrfs_ioctl_same_extent_info);
 
-	same = kmalloc(size, GFP_NOFS);
-	if (!same) {
-		ret = -EFAULT;
-		goto out;
-	}
+	same = memdup_user((struct btrfs_ioctl_same_args __user *)argp, size);
 
-	if (copy_from_user(same,
-			   (struct btrfs_ioctl_same_args __user *)argp, size)) {
-		ret = -EFAULT;
+	if (IS_ERR(same)) {
+		ret = PTR_ERR(same);
 		goto out;
 	}
 
@@ -3119,7 +3105,7 @@ out:
 static noinline long btrfs_ioctl_clone(struct file *file, unsigned long srcfd,
 				       u64 off, u64 olen, u64 destoff)
 {
-	struct inode *inode = fdentry(file)->d_inode;
+	struct inode *inode = file_inode(file);
 	struct btrfs_root *root = BTRFS_I(inode)->root;
 	struct fd src_file;
 	struct inode *src;
@@ -3679,9 +3665,10 @@ static long btrfs_ioctl_dev_replace(struct btrfs_root *root, void __user *arg)
 
 	switch (p->cmd) {
 	case BTRFS_IOCTL_DEV_REPLACE_CMD_START:
-		if (root->fs_info->sb->s_flags & MS_RDONLY)
-			return -EROFS;
-
+		if (root->fs_info->sb->s_flags & MS_RDONLY) {
+			ret = -EROFS;
+			goto out;
+		}
 		if (atomic_xchg(
 			&root->fs_info->mutually_exclusive_operation_running,
 			1)) {
@@ -3707,7 +3694,7 @@ static long btrfs_ioctl_dev_replace(struct btrfs_root *root, void __user *arg)
 
 	if (copy_to_user(arg, p, sizeof(*p)))
 		ret = -EFAULT;
-
+out:
 	kfree(p);
 	return ret;
 }
@@ -4317,7 +4304,7 @@ static long btrfs_ioctl_quota_rescan_status(struct file *file, void __user *arg)
 
 static long btrfs_ioctl_quota_rescan_wait(struct file *file, void __user *arg)
 {
-	struct btrfs_root *root = BTRFS_I(fdentry(file)->d_inode)->root;
+	struct btrfs_root *root = BTRFS_I(file_inode(file))->root;
 
 	if (!capable(CAP_SYS_ADMIN))
 		return -EPERM;
@@ -4557,9 +4544,15 @@ long btrfs_ioctl(struct file *file, unsigned int
 		return btrfs_ioctl_logical_to_ino(root, argp);
 	case BTRFS_IOC_SPACE_INFO:
 		return btrfs_ioctl_space_info(root, argp);
-	case BTRFS_IOC_SYNC:
-		btrfs_sync_fs(file->f_dentry->d_sb, 1);
-		return 0;
+	case BTRFS_IOC_SYNC: {
+		int ret;
+
+		ret = btrfs_start_delalloc_roots(root->fs_info, 0);
+		if (ret)
+			return ret;
+		ret = btrfs_sync_fs(file->f_dentry->d_sb, 1);
+		return ret;
+	}
 	case BTRFS_IOC_START_SYNC:
 		return btrfs_ioctl_start_sync(root, argp);
 	case BTRFS_IOC_WAIT_SYNC:
diff --git a/fs/btrfs/ordered-data.c b/fs/btrfs/ordered-data.c
index c702cb62f78a..25a8f3812f14 100644
--- a/fs/btrfs/ordered-data.c
+++ b/fs/btrfs/ordered-data.c
@@ -537,7 +537,9 @@ void btrfs_remove_ordered_extent(struct inode *inode,
 	 */
 	if (RB_EMPTY_ROOT(&tree->tree) &&
 	    !mapping_tagged(inode->i_mapping, PAGECACHE_TAG_DIRTY)) {
+		spin_lock(&root->fs_info->ordered_root_lock);
 		list_del_init(&BTRFS_I(inode)->ordered_operations);
+		spin_unlock(&root->fs_info->ordered_root_lock);
 	}
 
 	if (!root->nr_ordered_extents) {
@@ -563,10 +565,11 @@ static void btrfs_run_ordered_extent_work(struct btrfs_work *work)
  * wait for all the ordered extents in a root.  This is done when balancing
  * space between drives.
  */
-void btrfs_wait_ordered_extents(struct btrfs_root *root)
+int btrfs_wait_ordered_extents(struct btrfs_root *root, int nr)
 {
 	struct list_head splice, works;
 	struct btrfs_ordered_extent *ordered, *next;
+	int count = 0;
 
 	INIT_LIST_HEAD(&splice);
 	INIT_LIST_HEAD(&works);
@@ -574,7 +577,7 @@ void btrfs_wait_ordered_extents(struct btrfs_root *root)
 	mutex_lock(&root->fs_info->ordered_operations_mutex);
 	spin_lock(&root->ordered_extent_lock);
 	list_splice_init(&root->ordered_extents, &splice);
-	while (!list_empty(&splice)) {
+	while (!list_empty(&splice) && nr) {
 		ordered = list_first_entry(&splice, struct btrfs_ordered_extent,
 					   root_extent_list);
 		list_move_tail(&ordered->root_extent_list,
@@ -589,7 +592,11 @@ void btrfs_wait_ordered_extents(struct btrfs_root *root)
 
 		cond_resched();
 		spin_lock(&root->ordered_extent_lock);
+		if (nr != -1)
+			nr--;
+		count++;
 	}
+	list_splice_tail(&splice, &root->ordered_extents);
 	spin_unlock(&root->ordered_extent_lock);
 
 	list_for_each_entry_safe(ordered, next, &works, work_list) {
@@ -599,18 +606,21 @@ void btrfs_wait_ordered_extents(struct btrfs_root *root)
 		cond_resched();
 	}
 	mutex_unlock(&root->fs_info->ordered_operations_mutex);
+
+	return count;
 }
 
-void btrfs_wait_all_ordered_extents(struct btrfs_fs_info *fs_info)
+void btrfs_wait_ordered_roots(struct btrfs_fs_info *fs_info, int nr)
 {
 	struct btrfs_root *root;
 	struct list_head splice;
+	int done;
 
 	INIT_LIST_HEAD(&splice);
 
 	spin_lock(&fs_info->ordered_root_lock);
 	list_splice_init(&fs_info->ordered_roots, &splice);
-	while (!list_empty(&splice)) {
+	while (!list_empty(&splice) && nr) {
 		root = list_first_entry(&splice, struct btrfs_root,
 					ordered_root);
 		root = btrfs_grab_fs_root(root);
@@ -619,10 +629,14 @@ void btrfs_wait_all_ordered_extents(struct btrfs_fs_info *fs_info)
 			       &fs_info->ordered_roots);
 		spin_unlock(&fs_info->ordered_root_lock);
 
-		btrfs_wait_ordered_extents(root);
+		done = btrfs_wait_ordered_extents(root, nr);
 		btrfs_put_fs_root(root);
 
 		spin_lock(&fs_info->ordered_root_lock);
+		if (nr != -1) {
+			nr -= done;
+			WARN_ON(nr < 0);
+		}
 	}
 	spin_unlock(&fs_info->ordered_root_lock);
 }
@@ -734,8 +748,9 @@ void btrfs_start_ordered_extent(struct inode *inode,
 /*
  * Used to wait on ordered extents across a large range of bytes.
  */
-void btrfs_wait_ordered_range(struct inode *inode, u64 start, u64 len)
+int btrfs_wait_ordered_range(struct inode *inode, u64 start, u64 len)
 {
+	int ret = 0;
 	u64 end;
 	u64 orig_end;
 	struct btrfs_ordered_extent *ordered;
@@ -751,8 +766,9 @@ void btrfs_wait_ordered_range(struct inode *inode, u64 start, u64 len)
 	/* start IO across the range first to instantiate any delalloc
 	 * extents
 	 */
-	filemap_fdatawrite_range(inode->i_mapping, start, orig_end);
-
+	ret = filemap_fdatawrite_range(inode->i_mapping, start, orig_end);
+	if (ret)
+		return ret;
 	/*
 	 * So with compression we will find and lock a dirty page and clear the
 	 * first one as dirty, setup an async extent, and immediately return
@@ -768,10 +784,15 @@ void btrfs_wait_ordered_range(struct inode *inode, u64 start, u64 len)
 	 * right and you are wrong.
 	 */
 	if (test_bit(BTRFS_INODE_HAS_ASYNC_EXTENT,
-		     &BTRFS_I(inode)->runtime_flags))
-		filemap_fdatawrite_range(inode->i_mapping, start, orig_end);
-
-	filemap_fdatawait_range(inode->i_mapping, start, orig_end);
+		     &BTRFS_I(inode)->runtime_flags)) {
+		ret = filemap_fdatawrite_range(inode->i_mapping, start,
+					       orig_end);
+		if (ret)
+			return ret;
+	}
+	ret = filemap_fdatawait_range(inode->i_mapping, start, orig_end);
+	if (ret)
+		return ret;
 
 	end = orig_end;
 	while (1) {
@@ -788,11 +809,14 @@ void btrfs_wait_ordered_range(struct inode *inode, u64 start, u64 len)
 		}
 		btrfs_start_ordered_extent(inode, ordered, 1);
 		end = ordered->file_offset;
+		if (test_bit(BTRFS_ORDERED_IOERR, &ordered->flags))
+			ret = -EIO;
 		btrfs_put_ordered_extent(ordered);
-		if (end == 0 || end == start)
+		if (ret || end == 0 || end == start)
 			break;
 		end--;
 	}
+	return ret;
 }
 
 /*
@@ -1076,7 +1100,7 @@ void btrfs_add_ordered_operation(struct btrfs_trans_handle *trans,
 	 * if this file hasn't been changed since the last transaction
 	 * commit, we can safely return without doing anything
 	 */
-	if (last_mod < root->fs_info->last_trans_committed)
+	if (last_mod <= root->fs_info->last_trans_committed)
 		return;
 
 	spin_lock(&root->fs_info->ordered_root_lock);
diff --git a/fs/btrfs/ordered-data.h b/fs/btrfs/ordered-data.h
index 0c0b35612d7a..9b0450f7ac20 100644
--- a/fs/btrfs/ordered-data.h
+++ b/fs/btrfs/ordered-data.h
@@ -180,7 +180,7 @@ struct btrfs_ordered_extent *btrfs_lookup_ordered_extent(struct inode *inode,
 							 u64 file_offset);
 void btrfs_start_ordered_extent(struct inode *inode,
 				struct btrfs_ordered_extent *entry, int wait);
-void btrfs_wait_ordered_range(struct inode *inode, u64 start, u64 len);
+int btrfs_wait_ordered_range(struct inode *inode, u64 start, u64 len);
 struct btrfs_ordered_extent *
 btrfs_lookup_first_ordered_extent(struct inode * inode, u64 file_offset);
 struct btrfs_ordered_extent *btrfs_lookup_ordered_range(struct inode *inode,
@@ -195,8 +195,8 @@ int btrfs_run_ordered_operations(struct btrfs_trans_handle *trans,
 void btrfs_add_ordered_operation(struct btrfs_trans_handle *trans,
 				 struct btrfs_root *root,
 				 struct inode *inode);
-void btrfs_wait_ordered_extents(struct btrfs_root *root);
-void btrfs_wait_all_ordered_extents(struct btrfs_fs_info *fs_info);
+int btrfs_wait_ordered_extents(struct btrfs_root *root, int nr);
+void btrfs_wait_ordered_roots(struct btrfs_fs_info *fs_info, int nr);
 void btrfs_get_logged_extents(struct btrfs_root *log, struct inode *inode);
 void btrfs_wait_logged_extents(struct btrfs_root *log, u64 transid);
 void btrfs_free_logged_extents(struct btrfs_root *log, u64 transid);
diff --git a/fs/btrfs/print-tree.c b/fs/btrfs/print-tree.c
index 0088bedc8631..417053b17181 100644
--- a/fs/btrfs/print-tree.c
+++ b/fs/btrfs/print-tree.c
@@ -193,7 +193,7 @@ void btrfs_print_leaf(struct btrfs_root *root, struct extent_buffer *l)
 	btrfs_info(root->fs_info, "leaf %llu total ptrs %d free space %d",
 		   btrfs_header_bytenr(l), nr, btrfs_leaf_free_space(root, l));
 	for (i = 0 ; i < nr ; i++) {
-		item = btrfs_item_nr(l, i);
+		item = btrfs_item_nr(i);
 		btrfs_item_key_to_cpu(l, &key, i);
 		type = btrfs_key_type(&key);
 		printk(KERN_INFO "\titem %d key (%llu %u %llu) itemoff %d "
diff --git a/fs/btrfs/raid56.c b/fs/btrfs/raid56.c
index d0ecfbd9cc9f..24ac21840a9a 100644
--- a/fs/btrfs/raid56.c
+++ b/fs/btrfs/raid56.c
@@ -33,7 +33,6 @@
 #include <linux/raid/xor.h>
 #include <linux/vmalloc.h>
 #include <asm/div64.h>
-#include "compat.h"
 #include "ctree.h"
 #include "extent_map.h"
 #include "disk-io.h"
diff --git a/fs/btrfs/relocation.c b/fs/btrfs/relocation.c
index 4a355726151e..ce459a7cb16d 100644
--- a/fs/btrfs/relocation.c
+++ b/fs/btrfs/relocation.c
@@ -1383,6 +1383,7 @@ int btrfs_init_reloc_root(struct btrfs_trans_handle *trans,
 {
 	struct btrfs_root *reloc_root;
 	struct reloc_control *rc = root->fs_info->reloc_ctl;
+	struct btrfs_block_rsv *rsv;
 	int clear_rsv = 0;
 	int ret;
 
@@ -1396,13 +1397,14 @@ int btrfs_init_reloc_root(struct btrfs_trans_handle *trans,
 	    root->root_key.objectid == BTRFS_TREE_RELOC_OBJECTID)
 		return 0;
 
-	if (!trans->block_rsv) {
+	if (!trans->reloc_reserved) {
+		rsv = trans->block_rsv;
 		trans->block_rsv = rc->block_rsv;
 		clear_rsv = 1;
 	}
 	reloc_root = create_reloc_root(trans, root, root->root_key.objectid);
 	if (clear_rsv)
-		trans->block_rsv = NULL;
+		trans->block_rsv = rsv;
 
 	ret = __add_reloc_root(reloc_root);
 	BUG_ON(ret < 0);
@@ -1775,8 +1777,7 @@ again:
 			new_ptr_gen = 0;
 		}
 
-		if (new_bytenr > 0 && new_bytenr == old_bytenr) {
-			WARN_ON(1);
+		if (WARN_ON(new_bytenr > 0 && new_bytenr == old_bytenr)) {
 			ret = level;
 			break;
 		}
@@ -2058,7 +2059,7 @@ static noinline_for_stack int merge_reloc_root(struct reloc_control *rc,
 	LIST_HEAD(inode_list);
 	struct btrfs_key key;
 	struct btrfs_key next_key;
-	struct btrfs_trans_handle *trans;
+	struct btrfs_trans_handle *trans = NULL;
 	struct btrfs_root *reloc_root;
 	struct btrfs_root_item *root_item;
 	struct btrfs_path *path;
@@ -2107,18 +2108,19 @@ static noinline_for_stack int merge_reloc_root(struct reloc_control *rc,
 	memset(&next_key, 0, sizeof(next_key));
 
 	while (1) {
-		trans = btrfs_start_transaction(root, 0);
-		BUG_ON(IS_ERR(trans));
-		trans->block_rsv = rc->block_rsv;
-
 		ret = btrfs_block_rsv_refill(root, rc->block_rsv, min_reserved,
 					     BTRFS_RESERVE_FLUSH_ALL);
 		if (ret) {
-			BUG_ON(ret != -EAGAIN);
-			ret = btrfs_commit_transaction(trans, root);
-			BUG_ON(ret);
-			continue;
+			err = ret;
+			goto out;
 		}
+		trans = btrfs_start_transaction(root, 0);
+		if (IS_ERR(trans)) {
+			err = PTR_ERR(trans);
+			trans = NULL;
+			goto out;
+		}
+		trans->block_rsv = rc->block_rsv;
 
 		replaced = 0;
 		max_level = level;
@@ -2164,6 +2166,7 @@ static noinline_for_stack int merge_reloc_root(struct reloc_control *rc,
 		root_item->drop_level = level;
 
 		btrfs_end_transaction_throttle(trans, root);
+		trans = NULL;
 
 		btrfs_btree_balance_dirty(root);
 
@@ -2192,7 +2195,8 @@ out:
 		btrfs_update_reloc_root(trans, root);
 	}
 
-	btrfs_end_transaction_throttle(trans, root);
+	if (trans)
+		btrfs_end_transaction_throttle(trans, root);
 
 	btrfs_btree_balance_dirty(root);
 
@@ -3258,7 +3262,7 @@ static int add_tree_block(struct reloc_control *rc,
 	struct rb_node *rb_node;
 	u32 item_size;
 	int level = -1;
-	int generation;
+	u64 generation;
 
 	eb =  path->nodes[0];
 	item_size = btrfs_item_size_nr(eb, path->slots[0]);
@@ -3407,7 +3411,6 @@ static int delete_block_group_cache(struct btrfs_fs_info *fs_info,
 				    struct inode *inode, u64 ino)
 {
 	struct btrfs_key key;
-	struct btrfs_path *path;
 	struct btrfs_root *root = fs_info->tree_root;
 	struct btrfs_trans_handle *trans;
 	int ret = 0;
@@ -3432,22 +3435,14 @@ truncate:
 	if (ret)
 		goto out;
 
-	path = btrfs_alloc_path();
-	if (!path) {
-		ret = -ENOMEM;
-		goto out;
-	}
-
 	trans = btrfs_join_transaction(root);
 	if (IS_ERR(trans)) {
-		btrfs_free_path(path);
 		ret = PTR_ERR(trans);
 		goto out;
 	}
 
-	ret = btrfs_truncate_free_space_cache(root, trans, path, inode);
+	ret = btrfs_truncate_free_space_cache(root, trans, inode);
 
-	btrfs_free_path(path);
 	btrfs_end_transaction(trans, root);
 	btrfs_btree_balance_dirty(root);
 out:
@@ -3549,10 +3544,8 @@ static int find_data_references(struct reloc_control *rc,
 				err = ret;
 				goto out;
 			}
-			if (ret > 0) {
-				WARN_ON(1);
+			if (WARN_ON(ret > 0))
 				goto out;
-			}
 
 			leaf = path->nodes[0];
 			nritems = btrfs_header_nritems(leaf);
@@ -3572,11 +3565,9 @@ static int find_data_references(struct reloc_control *rc,
 		}
 
 		btrfs_item_key_to_cpu(leaf, &key, path->slots[0]);
-		if (key.objectid != ref_objectid ||
-		    key.type != BTRFS_EXTENT_DATA_KEY) {
-			WARN_ON(1);
+		if (WARN_ON(key.objectid != ref_objectid ||
+		    key.type != BTRFS_EXTENT_DATA_KEY))
 			break;
-		}
 
 		fi = btrfs_item_ptr(leaf, path->slots[0],
 				    struct btrfs_file_extent_item);
@@ -4001,16 +3992,6 @@ restart:
 			}
 		}
 
-		ret = btrfs_block_rsv_check(rc->extent_root, rc->block_rsv, 5);
-		if (ret < 0) {
-			if (ret != -ENOSPC) {
-				err = ret;
-				WARN_ON(1);
-				break;
-			}
-			rc->commit_transaction = 1;
-		}
-
 		if (rc->commit_transaction) {
 			rc->commit_transaction = 0;
 			ret = btrfs_commit_transaction(trans, rc->extent_root);
@@ -4241,12 +4222,12 @@ int btrfs_relocate_block_group(struct btrfs_root *extent_root, u64 group_start)
 	printk(KERN_INFO "btrfs: relocating block group %llu flags %llu\n",
 	       rc->block_group->key.objectid, rc->block_group->flags);
 
-	ret = btrfs_start_all_delalloc_inodes(fs_info, 0);
+	ret = btrfs_start_delalloc_roots(fs_info, 0);
 	if (ret < 0) {
 		err = ret;
 		goto out;
 	}
-	btrfs_wait_all_ordered_extents(fs_info);
+	btrfs_wait_ordered_roots(fs_info, -1);
 
 	while (1) {
 		mutex_lock(&fs_info->cleaner_mutex);
@@ -4264,7 +4245,12 @@ int btrfs_relocate_block_group(struct btrfs_root *extent_root, u64 group_start)
 			rc->extents_found);
 
 		if (rc->stage == MOVE_DATA_EXTENTS && rc->found_file_extent) {
-			btrfs_wait_ordered_range(rc->data_inode, 0, (u64)-1);
+			ret = btrfs_wait_ordered_range(rc->data_inode, 0,
+						       (u64)-1);
+			if (ret) {
+				err = ret;
+				goto out;
+			}
 			invalidate_mapping_pages(rc->data_inode->i_mapping,
 						 0, -1);
 			rc->stage = UPDATE_DATA_PTRS;
@@ -4481,6 +4467,7 @@ int btrfs_reloc_clone_csums(struct inode *inode, u64 file_pos, u64 len)
 	struct btrfs_root *root = BTRFS_I(inode)->root;
 	int ret;
 	u64 disk_bytenr;
+	u64 new_bytenr;
 	LIST_HEAD(list);
 
 	ordered = btrfs_lookup_ordered_extent(inode, file_pos);
@@ -4492,13 +4479,24 @@ int btrfs_reloc_clone_csums(struct inode *inode, u64 file_pos, u64 len)
 	if (ret)
 		goto out;
 
-	disk_bytenr = ordered->start;
 	while (!list_empty(&list)) {
 		sums = list_entry(list.next, struct btrfs_ordered_sum, list);
 		list_del_init(&sums->list);
 
-		sums->bytenr = disk_bytenr;
-		disk_bytenr += sums->len;
+		/*
+		 * We need to offset the new_bytenr based on where the csum is.
+		 * We need to do this because we will read in entire prealloc
+		 * extents but we may have written to say the middle of the
+		 * prealloc extent, so we need to make sure the csum goes with
+		 * the right disk offset.
+		 *
+		 * We can do this because the data reloc inode refers strictly
+		 * to the on disk bytes, so we don't have to worry about
+		 * disk_len vs real len like with real inodes since it's all
+		 * disk length.
+		 */
+		new_bytenr = ordered->start + (sums->bytenr - disk_bytenr);
+		sums->bytenr = new_bytenr;
 
 		btrfs_add_ordered_sum(inode, ordered, sums);
 	}
diff --git a/fs/btrfs/scrub.c b/fs/btrfs/scrub.c
index a18e0e23f6a6..2544805544f0 100644
--- a/fs/btrfs/scrub.c
+++ b/fs/btrfs/scrub.c
@@ -2717,8 +2717,6 @@ int scrub_enumerate_chunks(struct scrub_ctx *sctx,
 		mutex_unlock(&fs_info->scrub_lock);
 		wake_up(&fs_info->scrub_pause_wait);
 
-		dev_replace->cursor_left = dev_replace->cursor_right;
-		dev_replace->item_needs_writeback = 1;
 		btrfs_put_block_group(cache);
 		if (ret)
 			break;
@@ -2732,6 +2730,9 @@ int scrub_enumerate_chunks(struct scrub_ctx *sctx,
 			break;
 		}
 
+		dev_replace->cursor_left = dev_replace->cursor_right;
+		dev_replace->item_needs_writeback = 1;
+
 		key.offset = found_key.offset + length;
 		btrfs_release_path(path);
 	}
@@ -2783,7 +2784,6 @@ static noinline_for_stack int scrub_workers_get(struct btrfs_fs_info *fs_info,
 {
 	int ret = 0;
 
-	mutex_lock(&fs_info->scrub_lock);
 	if (fs_info->scrub_workers_refcnt == 0) {
 		if (is_dev_replace)
 			btrfs_init_workers(&fs_info->scrub_workers, "scrub", 1,
@@ -2813,21 +2813,17 @@ static noinline_for_stack int scrub_workers_get(struct btrfs_fs_info *fs_info,
 	}
 	++fs_info->scrub_workers_refcnt;
 out:
-	mutex_unlock(&fs_info->scrub_lock);
-
 	return ret;
 }
 
 static noinline_for_stack void scrub_workers_put(struct btrfs_fs_info *fs_info)
 {
-	mutex_lock(&fs_info->scrub_lock);
 	if (--fs_info->scrub_workers_refcnt == 0) {
 		btrfs_stop_workers(&fs_info->scrub_workers);
 		btrfs_stop_workers(&fs_info->scrub_wr_completion_workers);
 		btrfs_stop_workers(&fs_info->scrub_nocow_workers);
 	}
 	WARN_ON(fs_info->scrub_workers_refcnt < 0);
-	mutex_unlock(&fs_info->scrub_lock);
 }
 
 int btrfs_scrub_dev(struct btrfs_fs_info *fs_info, u64 devid, u64 start,
@@ -2888,23 +2884,18 @@ int btrfs_scrub_dev(struct btrfs_fs_info *fs_info, u64 devid, u64 start,
 		return -EINVAL;
 	}
 
-	ret = scrub_workers_get(fs_info, is_dev_replace);
-	if (ret)
-		return ret;
 
 	mutex_lock(&fs_info->fs_devices->device_list_mutex);
 	dev = btrfs_find_device(fs_info, devid, NULL, NULL);
 	if (!dev || (dev->missing && !is_dev_replace)) {
 		mutex_unlock(&fs_info->fs_devices->device_list_mutex);
-		scrub_workers_put(fs_info);
 		return -ENODEV;
 	}
-	mutex_lock(&fs_info->scrub_lock);
 
+	mutex_lock(&fs_info->scrub_lock);
 	if (!dev->in_fs_metadata || dev->is_tgtdev_for_dev_replace) {
 		mutex_unlock(&fs_info->scrub_lock);
 		mutex_unlock(&fs_info->fs_devices->device_list_mutex);
-		scrub_workers_put(fs_info);
 		return -EIO;
 	}
 
@@ -2915,10 +2906,17 @@ int btrfs_scrub_dev(struct btrfs_fs_info *fs_info, u64 devid, u64 start,
 		btrfs_dev_replace_unlock(&fs_info->dev_replace);
 		mutex_unlock(&fs_info->scrub_lock);
 		mutex_unlock(&fs_info->fs_devices->device_list_mutex);
-		scrub_workers_put(fs_info);
 		return -EINPROGRESS;
 	}
 	btrfs_dev_replace_unlock(&fs_info->dev_replace);
+
+	ret = scrub_workers_get(fs_info, is_dev_replace);
+	if (ret) {
+		mutex_unlock(&fs_info->scrub_lock);
+		mutex_unlock(&fs_info->fs_devices->device_list_mutex);
+		return ret;
+	}
+
 	sctx = scrub_setup_ctx(dev, is_dev_replace);
 	if (IS_ERR(sctx)) {
 		mutex_unlock(&fs_info->scrub_lock);
@@ -2931,13 +2929,15 @@ int btrfs_scrub_dev(struct btrfs_fs_info *fs_info, u64 devid, u64 start,
 
 	atomic_inc(&fs_info->scrubs_running);
 	mutex_unlock(&fs_info->scrub_lock);
-	mutex_unlock(&fs_info->fs_devices->device_list_mutex);
 
 	if (!is_dev_replace) {
-		down_read(&fs_info->scrub_super_lock);
+		/*
+		 * by holding device list mutex, we can
+		 * kick off writing super in log tree sync.
+		 */
 		ret = scrub_supers(sctx, dev);
-		up_read(&fs_info->scrub_super_lock);
 	}
+	mutex_unlock(&fs_info->fs_devices->device_list_mutex);
 
 	if (!ret)
 		ret = scrub_enumerate_chunks(sctx, dev, start, end,
@@ -2954,10 +2954,10 @@ int btrfs_scrub_dev(struct btrfs_fs_info *fs_info, u64 devid, u64 start,
 
 	mutex_lock(&fs_info->scrub_lock);
 	dev->scrub_device = NULL;
+	scrub_workers_put(fs_info);
 	mutex_unlock(&fs_info->scrub_lock);
 
 	scrub_free_ctx(sctx);
-	scrub_workers_put(fs_info);
 
 	return ret;
 }
@@ -2987,16 +2987,6 @@ void btrfs_scrub_continue(struct btrfs_root *root)
 	wake_up(&fs_info->scrub_pause_wait);
 }
 
-void btrfs_scrub_pause_super(struct btrfs_root *root)
-{
-	down_write(&root->fs_info->scrub_super_lock);
-}
-
-void btrfs_scrub_continue_super(struct btrfs_root *root)
-{
-	up_write(&root->fs_info->scrub_super_lock);
-}
-
 int btrfs_scrub_cancel(struct btrfs_fs_info *fs_info)
 {
 	mutex_lock(&fs_info->scrub_lock);
diff --git a/fs/btrfs/send.c b/fs/btrfs/send.c
index e46e0ed74925..6837fe87f3a6 100644
--- a/fs/btrfs/send.c
+++ b/fs/btrfs/send.c
@@ -121,7 +121,6 @@ struct send_ctx {
 	struct list_head name_cache_list;
 	int name_cache_size;
 
-	struct file *cur_inode_filp;
 	char *read_buf;
 };
 
@@ -565,10 +564,8 @@ static int begin_cmd(struct send_ctx *sctx, int cmd)
 {
 	struct btrfs_cmd_header *hdr;
 
-	if (!sctx->send_buf) {
-		WARN_ON(1);
+	if (WARN_ON(!sctx->send_buf))
 		return -EINVAL;
-	}
 
 	BUG_ON(sctx->send_size);
 
@@ -791,7 +788,7 @@ static int iterate_inode_ref(struct btrfs_root *root, struct btrfs_path *path,
 	if (found_key->type == BTRFS_INODE_REF_KEY) {
 		ptr = (unsigned long)btrfs_item_ptr(eb, slot,
 						    struct btrfs_inode_ref);
-		item = btrfs_item_nr(eb, slot);
+		item = btrfs_item_nr(slot);
 		total = btrfs_item_size(eb, item);
 		elem_size = sizeof(*iref);
 	} else {
@@ -905,7 +902,7 @@ static int iterate_dir_item(struct btrfs_root *root, struct btrfs_path *path,
 
 	eb = path->nodes[0];
 	slot = path->slots[0];
-	item = btrfs_item_nr(eb, slot);
+	item = btrfs_item_nr(slot);
 	di = btrfs_item_ptr(eb, slot, struct btrfs_dir_item);
 	cur = 0;
 	len = 0;
@@ -2120,77 +2117,6 @@ out:
 }
 
 /*
- * Called for regular files when sending extents data. Opens a struct file
- * to read from the file.
- */
-static int open_cur_inode_file(struct send_ctx *sctx)
-{
-	int ret = 0;
-	struct btrfs_key key;
-	struct path path;
-	struct inode *inode;
-	struct dentry *dentry;
-	struct file *filp;
-	int new = 0;
-
-	if (sctx->cur_inode_filp)
-		goto out;
-
-	key.objectid = sctx->cur_ino;
-	key.type = BTRFS_INODE_ITEM_KEY;
-	key.offset = 0;
-
-	inode = btrfs_iget(sctx->send_root->fs_info->sb, &key, sctx->send_root,
-			&new);
-	if (IS_ERR(inode)) {
-		ret = PTR_ERR(inode);
-		goto out;
-	}
-
-	dentry = d_obtain_alias(inode);
-	inode = NULL;
-	if (IS_ERR(dentry)) {
-		ret = PTR_ERR(dentry);
-		goto out;
-	}
-
-	path.mnt = sctx->mnt;
-	path.dentry = dentry;
-	filp = dentry_open(&path, O_RDONLY | O_LARGEFILE, current_cred());
-	dput(dentry);
-	dentry = NULL;
-	if (IS_ERR(filp)) {
-		ret = PTR_ERR(filp);
-		goto out;
-	}
-	sctx->cur_inode_filp = filp;
-
-out:
-	/*
-	 * no xxxput required here as every vfs op
-	 * does it by itself on failure
-	 */
-	return ret;
-}
-
-/*
- * Closes the struct file that was created in open_cur_inode_file
- */
-static int close_cur_inode_file(struct send_ctx *sctx)
-{
-	int ret = 0;
-
-	if (!sctx->cur_inode_filp)
-		goto out;
-
-	ret = filp_close(sctx->cur_inode_filp, NULL);
-	sctx->cur_inode_filp = NULL;
-
-out:
-	return ret;
-}
-
-/*
  * Sends a BTRFS_SEND_C_SUBVOL command/item to userspace
  */
 static int send_subvol_begin(struct send_ctx *sctx)
@@ -3622,6 +3548,72 @@ out:
 	return ret;
 }
 
+static ssize_t fill_read_buf(struct send_ctx *sctx, u64 offset, u32 len)
+{
+	struct btrfs_root *root = sctx->send_root;
+	struct btrfs_fs_info *fs_info = root->fs_info;
+	struct inode *inode;
+	struct page *page;
+	char *addr;
+	struct btrfs_key key;
+	pgoff_t index = offset >> PAGE_CACHE_SHIFT;
+	pgoff_t last_index;
+	unsigned pg_offset = offset & ~PAGE_CACHE_MASK;
+	ssize_t ret = 0;
+
+	key.objectid = sctx->cur_ino;
+	key.type = BTRFS_INODE_ITEM_KEY;
+	key.offset = 0;
+
+	inode = btrfs_iget(fs_info->sb, &key, root, NULL);
+	if (IS_ERR(inode))
+		return PTR_ERR(inode);
+
+	if (offset + len > i_size_read(inode)) {
+		if (offset > i_size_read(inode))
+			len = 0;
+		else
+			len = offset - i_size_read(inode);
+	}
+	if (len == 0)
+		goto out;
+
+	last_index = (offset + len - 1) >> PAGE_CACHE_SHIFT;
+	while (index <= last_index) {
+		unsigned cur_len = min_t(unsigned, len,
+					 PAGE_CACHE_SIZE - pg_offset);
+		page = find_or_create_page(inode->i_mapping, index, GFP_NOFS);
+		if (!page) {
+			ret = -ENOMEM;
+			break;
+		}
+
+		if (!PageUptodate(page)) {
+			btrfs_readpage(NULL, page);
+			lock_page(page);
+			if (!PageUptodate(page)) {
+				unlock_page(page);
+				page_cache_release(page);
+				ret = -EIO;
+				break;
+			}
+		}
+
+		addr = kmap(page);
+		memcpy(sctx->read_buf + ret, addr + pg_offset, cur_len);
+		kunmap(page);
+		unlock_page(page);
+		page_cache_release(page);
+		index++;
+		pg_offset = 0;
+		len -= cur_len;
+		ret += cur_len;
+	}
+out:
+	iput(inode);
+	return ret;
+}
+
 /*
  * Read some bytes from the current inode/file and send a write command to
  * user space.
@@ -3630,35 +3622,20 @@ static int send_write(struct send_ctx *sctx, u64 offset, u32 len)
 {
 	int ret = 0;
 	struct fs_path *p;
-	loff_t pos = offset;
-	int num_read = 0;
-	mm_segment_t old_fs;
+	ssize_t num_read = 0;
 
 	p = fs_path_alloc();
 	if (!p)
 		return -ENOMEM;
 
-	/*
-	 * vfs normally only accepts user space buffers for security reasons.
-	 * we only read from the file and also only provide the read_buf buffer
-	 * to vfs. As this buffer does not come from a user space call, it's
-	 * ok to temporary allow kernel space buffers.
-	 */
-	old_fs = get_fs();
-	set_fs(KERNEL_DS);
-
 verbose_printk("btrfs: send_write offset=%llu, len=%d\n", offset, len);
 
-	ret = open_cur_inode_file(sctx);
-	if (ret < 0)
-		goto out;
-
-	ret = vfs_read(sctx->cur_inode_filp, sctx->read_buf, len, &pos);
-	if (ret < 0)
-		goto out;
-	num_read = ret;
-	if (!num_read)
+	num_read = fill_read_buf(sctx, offset, len);
+	if (num_read <= 0) {
+		if (num_read < 0)
+			ret = num_read;
 		goto out;
+	}
 
 	ret = begin_cmd(sctx, BTRFS_SEND_C_WRITE);
 	if (ret < 0)
@@ -3677,7 +3654,6 @@ verbose_printk("btrfs: send_write offset=%llu, len=%d\n", offset, len);
 tlv_put_failure:
 out:
 	fs_path_free(p);
-	set_fs(old_fs);
 	if (ret < 0)
 		return ret;
 	return num_read;
@@ -3926,16 +3902,16 @@ static int is_extent_unchanged(struct send_ctx *sctx,
 	while (key.offset < ekey->offset + left_len) {
 		ei = btrfs_item_ptr(eb, slot, struct btrfs_file_extent_item);
 		right_type = btrfs_file_extent_type(eb, ei);
-		right_disknr = btrfs_file_extent_disk_bytenr(eb, ei);
-		right_len = btrfs_file_extent_num_bytes(eb, ei);
-		right_offset = btrfs_file_extent_offset(eb, ei);
-		right_gen = btrfs_file_extent_generation(eb, ei);
-
 		if (right_type != BTRFS_FILE_EXTENT_REG) {
 			ret = 0;
 			goto out;
 		}
 
+		right_disknr = btrfs_file_extent_disk_bytenr(eb, ei);
+		right_len = btrfs_file_extent_num_bytes(eb, ei);
+		right_offset = btrfs_file_extent_offset(eb, ei);
+		right_gen = btrfs_file_extent_generation(eb, ei);
+
 		/*
 		 * Are we at extent 8? If yes, we know the extent is changed.
 		 * This may only happen on the first iteration.
@@ -4222,10 +4198,6 @@ static int changed_inode(struct send_ctx *sctx,
 	u64 left_gen = 0;
 	u64 right_gen = 0;
 
-	ret = close_cur_inode_file(sctx);
-	if (ret < 0)
-		goto out;
-
 	sctx->cur_ino = key->objectid;
 	sctx->cur_inode_new_gen = 0;
 
@@ -4686,11 +4658,6 @@ static int send_subvol(struct send_ctx *sctx)
 	}
 
 out:
-	if (!ret)
-		ret = close_cur_inode_file(sctx);
-	else
-		close_cur_inode_file(sctx);
-
 	free_recorded_refs(sctx);
 	return ret;
 }
diff --git a/fs/btrfs/super.c b/fs/btrfs/super.c
index e913328d0f2a..2d8ac1bf0cf9 100644
--- a/fs/btrfs/super.c
+++ b/fs/btrfs/super.c
@@ -42,7 +42,6 @@
 #include <linux/cleancache.h>
 #include <linux/ratelimit.h>
 #include <linux/btrfs.h>
-#include "compat.h"
 #include "delayed-inode.h"
 #include "ctree.h"
 #include "disk-io.h"
@@ -921,7 +920,7 @@ int btrfs_sync_fs(struct super_block *sb, int wait)
 		return 0;
 	}
 
-	btrfs_wait_all_ordered_extents(fs_info);
+	btrfs_wait_ordered_roots(fs_info, -1);
 
 	trans = btrfs_attach_transaction_barrier(root);
 	if (IS_ERR(trans)) {
@@ -1330,6 +1329,12 @@ static int btrfs_remount(struct super_block *sb, int *flags, char *data)
 		 * this also happens on 'umount -rf' or on shutdown, when
 		 * the filesystem is busy.
 		 */
+
+		/* wait for the uuid_scan task to finish */
+		down(&fs_info->uuid_tree_rescan_sem);
+		/* avoid complains from lockdep et al. */
+		up(&fs_info->uuid_tree_rescan_sem);
+
 		sb->s_flags |= MS_RDONLY;
 
 		btrfs_dev_replace_suspend_for_unmount(fs_info);
@@ -1465,7 +1470,7 @@ static int btrfs_calc_avail_data_space(struct btrfs_root *root, u64 *free_bytes)
 	nr_devices = fs_info->fs_devices->open_devices;
 	BUG_ON(!nr_devices);
 
-	devices_info = kmalloc(sizeof(*devices_info) * nr_devices,
+	devices_info = kmalloc_array(nr_devices, sizeof(*devices_info),
 			       GFP_NOFS);
 	if (!devices_info)
 		return -ENOMEM;
@@ -1789,7 +1794,25 @@ static void btrfs_print_info(void)
 
 static int btrfs_run_sanity_tests(void)
 {
-	return btrfs_test_free_space_cache();
+	int ret;
+
+	ret = btrfs_init_test_fs();
+	if (ret)
+		return ret;
+
+	ret = btrfs_test_free_space_cache();
+	if (ret)
+		goto out;
+	ret = btrfs_test_extent_buffer_operations();
+	if (ret)
+		goto out;
+	ret = btrfs_test_extent_io();
+	if (ret)
+		goto out;
+	ret = btrfs_test_inodes();
+out:
+	btrfs_destroy_test_fs();
+	return ret;
 }
 
 static int __init init_btrfs_fs(void)
diff --git a/fs/btrfs/tests/btrfs-tests.c b/fs/btrfs/tests/btrfs-tests.c
new file mode 100644
index 000000000000..757ef00a75a4
--- /dev/null
+++ b/fs/btrfs/tests/btrfs-tests.c
@@ -0,0 +1,74 @@
+/*
+ * Copyright (C) 2013 Fusion IO.  All rights reserved.
+ *
+ * This program is free software; you can redistribute it and/or
+ * modify it under the terms of the GNU General Public
+ * License v2 as published by the Free Software Foundation.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the GNU
+ * General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public
+ * License along with this program; if not, write to the
+ * Free Software Foundation, Inc., 59 Temple Place - Suite 330,
+ * Boston, MA 021110-1307, USA.
+ */
+
+#include <linux/fs.h>
+#include <linux/mount.h>
+#include <linux/magic.h>
+#include "btrfs-tests.h"
+#include "../ctree.h"
+
+static struct vfsmount *test_mnt = NULL;
+
+static const struct super_operations btrfs_test_super_ops = {
+	.alloc_inode	= btrfs_alloc_inode,
+	.destroy_inode	= btrfs_test_destroy_inode,
+};
+
+static struct dentry *btrfs_test_mount(struct file_system_type *fs_type,
+				       int flags, const char *dev_name,
+				       void *data)
+{
+	return mount_pseudo(fs_type, "btrfs_test:", &btrfs_test_super_ops,
+			    NULL, BTRFS_TEST_MAGIC);
+}
+
+static struct file_system_type test_type = {
+	.name		= "btrfs_test_fs",
+	.mount		= btrfs_test_mount,
+	.kill_sb	= kill_anon_super,
+};
+
+struct inode *btrfs_new_test_inode(void)
+{
+	return new_inode(test_mnt->mnt_sb);
+}
+
+int btrfs_init_test_fs(void)
+{
+	int ret;
+
+	ret = register_filesystem(&test_type);
+	if (ret) {
+		printk(KERN_ERR "btrfs: cannot register test file system\n");
+		return ret;
+	}
+
+	test_mnt = kern_mount(&test_type);
+	if (IS_ERR(test_mnt)) {
+		printk(KERN_ERR "btrfs: cannot mount test file system\n");
+		unregister_filesystem(&test_type);
+		return ret;
+	}
+	return 0;
+}
+
+void btrfs_destroy_test_fs(void)
+{
+	kern_unmount(test_mnt);
+	unregister_filesystem(&test_type);
+}
diff --git a/fs/btrfs/tests/btrfs-tests.h b/fs/btrfs/tests/btrfs-tests.h
index 580877625776..b353bc806ca0 100644
--- a/fs/btrfs/tests/btrfs-tests.h
+++ b/fs/btrfs/tests/btrfs-tests.h
@@ -24,11 +24,36 @@
 #define test_msg(fmt, ...) pr_info("btrfs: selftest: " fmt, ##__VA_ARGS__)
 
 int btrfs_test_free_space_cache(void);
+int btrfs_test_extent_buffer_operations(void);
+int btrfs_test_extent_io(void);
+int btrfs_test_inodes(void);
+int btrfs_init_test_fs(void);
+void btrfs_destroy_test_fs(void);
+struct inode *btrfs_new_test_inode(void);
 #else
 static inline int btrfs_test_free_space_cache(void)
 {
 	return 0;
 }
+static inline int btrfs_test_extent_buffer_operations(void)
+{
+	return 0;
+}
+static inline int btrfs_init_test_fs(void)
+{
+	return 0;
+}
+static inline void btrfs_destroy_test_fs(void)
+{
+}
+static inline int btrfs_test_extent_io(void)
+{
+	return 0;
+}
+static inline int btrfs_test_inodes(void)
+{
+	return 0;
+}
 #endif
 
 #endif
diff --git a/fs/btrfs/tests/extent-buffer-tests.c b/fs/btrfs/tests/extent-buffer-tests.c
new file mode 100644
index 000000000000..cc286ce97d1e
--- /dev/null
+++ b/fs/btrfs/tests/extent-buffer-tests.c
@@ -0,0 +1,229 @@
+/*
+ * Copyright (C) 2013 Fusion IO.  All rights reserved.
+ *
+ * This program is free software; you can redistribute it and/or
+ * modify it under the terms of the GNU General Public
+ * License v2 as published by the Free Software Foundation.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the GNU
+ * General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public
+ * License along with this program; if not, write to the
+ * Free Software Foundation, Inc., 59 Temple Place - Suite 330,
+ * Boston, MA 021110-1307, USA.
+ */
+
+#include <linux/slab.h>
+#include "btrfs-tests.h"
+#include "../ctree.h"
+#include "../extent_io.h"
+#include "../disk-io.h"
+
+static int test_btrfs_split_item(void)
+{
+	struct btrfs_path *path;
+	struct btrfs_root *root;
+	struct extent_buffer *eb;
+	struct btrfs_item *item;
+	char *value = "mary had a little lamb";
+	char *split1 = "mary had a little";
+	char *split2 = " lamb";
+	char *split3 = "mary";
+	char *split4 = " had a little";
+	char buf[32];
+	struct btrfs_key key;
+	u32 value_len = strlen(value);
+	int ret = 0;
+
+	test_msg("Running btrfs_split_item tests\n");
+
+	root = btrfs_alloc_dummy_root();
+	if (IS_ERR(root)) {
+		test_msg("Could not allocate root\n");
+		return PTR_ERR(root);
+	}
+
+	path = btrfs_alloc_path();
+	if (!path) {
+		test_msg("Could not allocate path\n");
+		kfree(root);
+		return -ENOMEM;
+	}
+
+	path->nodes[0] = eb = alloc_dummy_extent_buffer(0, 4096);
+	if (!eb) {
+		test_msg("Could not allocate dummy buffer\n");
+		ret = -ENOMEM;
+		goto out;
+	}
+	path->slots[0] = 0;
+
+	key.objectid = 0;
+	key.type = BTRFS_EXTENT_CSUM_KEY;
+	key.offset = 0;
+
+	setup_items_for_insert(root, path, &key, &value_len, value_len,
+			       value_len + sizeof(struct btrfs_item), 1);
+	item = btrfs_item_nr(0);
+	write_extent_buffer(eb, value, btrfs_item_ptr_offset(eb, 0),
+			    value_len);
+
+	key.offset = 3;
+
+	/*
+	 * Passing NULL trans here should be safe because we have plenty of
+	 * space in this leaf to split the item without having to split the
+	 * leaf.
+	 */
+	ret = btrfs_split_item(NULL, root, path, &key, 17);
+	if (ret) {
+		test_msg("Split item failed %d\n", ret);
+		goto out;
+	}
+
+	/*
+	 * Read the first slot, it should have the original key and contain only
+	 * 'mary had a little'
+	 */
+	btrfs_item_key_to_cpu(eb, &key, 0);
+	if (key.objectid != 0 || key.type != BTRFS_EXTENT_CSUM_KEY ||
+	    key.offset != 0) {
+		test_msg("Invalid key at slot 0\n");
+		ret = -EINVAL;
+		goto out;
+	}
+
+	item = btrfs_item_nr(0);
+	if (btrfs_item_size(eb, item) != strlen(split1)) {
+		test_msg("Invalid len in the first split\n");
+		ret = -EINVAL;
+		goto out;
+	}
+
+	read_extent_buffer(eb, buf, btrfs_item_ptr_offset(eb, 0),
+			   strlen(split1));
+	if (memcmp(buf, split1, strlen(split1))) {
+		test_msg("Data in the buffer doesn't match what it should "
+			 "in the first split have='%.*s' want '%s'\n",
+			 (int)strlen(split1), buf, split1);
+		ret = -EINVAL;
+		goto out;
+	}
+
+	btrfs_item_key_to_cpu(eb, &key, 1);
+	if (key.objectid != 0 || key.type != BTRFS_EXTENT_CSUM_KEY ||
+	    key.offset != 3) {
+		test_msg("Invalid key at slot 1\n");
+		ret = -EINVAL;
+		goto out;
+	}
+
+	item = btrfs_item_nr(1);
+	if (btrfs_item_size(eb, item) != strlen(split2)) {
+		test_msg("Invalid len in the second split\n");
+		ret = -EINVAL;
+		goto out;
+	}
+
+	read_extent_buffer(eb, buf, btrfs_item_ptr_offset(eb, 1),
+			   strlen(split2));
+	if (memcmp(buf, split2, strlen(split2))) {
+		test_msg("Data in the buffer doesn't match what it should "
+			 "in the second split\n");
+		ret = -EINVAL;
+		goto out;
+	}
+
+	key.offset = 1;
+	/* Do it again so we test memmoving the other items in the leaf */
+	ret = btrfs_split_item(NULL, root, path, &key, 4);
+	if (ret) {
+		test_msg("Second split item failed %d\n", ret);
+		goto out;
+	}
+
+	btrfs_item_key_to_cpu(eb, &key, 0);
+	if (key.objectid != 0 || key.type != BTRFS_EXTENT_CSUM_KEY ||
+	    key.offset != 0) {
+		test_msg("Invalid key at slot 0\n");
+		ret = -EINVAL;
+		goto out;
+	}
+
+	item = btrfs_item_nr(0);
+	if (btrfs_item_size(eb, item) != strlen(split3)) {
+		test_msg("Invalid len in the first split\n");
+		ret = -EINVAL;
+		goto out;
+	}
+
+	read_extent_buffer(eb, buf, btrfs_item_ptr_offset(eb, 0),
+			   strlen(split3));
+	if (memcmp(buf, split3, strlen(split3))) {
+		test_msg("Data in the buffer doesn't match what it should "
+			 "in the third split");
+		ret = -EINVAL;
+		goto out;
+	}
+
+	btrfs_item_key_to_cpu(eb, &key, 1);
+	if (key.objectid != 0 || key.type != BTRFS_EXTENT_CSUM_KEY ||
+	    key.offset != 1) {
+		test_msg("Invalid key at slot 1\n");
+		ret = -EINVAL;
+		goto out;
+	}
+
+	item = btrfs_item_nr(1);
+	if (btrfs_item_size(eb, item) != strlen(split4)) {
+		test_msg("Invalid len in the second split\n");
+		ret = -EINVAL;
+		goto out;
+	}
+
+	read_extent_buffer(eb, buf, btrfs_item_ptr_offset(eb, 1),
+			   strlen(split4));
+	if (memcmp(buf, split4, strlen(split4))) {
+		test_msg("Data in the buffer doesn't match what it should "
+			 "in the fourth split\n");
+		ret = -EINVAL;
+		goto out;
+	}
+
+	btrfs_item_key_to_cpu(eb, &key, 2);
+	if (key.objectid != 0 || key.type != BTRFS_EXTENT_CSUM_KEY ||
+	    key.offset != 3) {
+		test_msg("Invalid key at slot 2\n");
+		ret = -EINVAL;
+		goto out;
+	}
+
+	item = btrfs_item_nr(2);
+	if (btrfs_item_size(eb, item) != strlen(split2)) {
+		test_msg("Invalid len in the second split\n");
+		ret = -EINVAL;
+		goto out;
+	}
+
+	read_extent_buffer(eb, buf, btrfs_item_ptr_offset(eb, 2),
+			   strlen(split2));
+	if (memcmp(buf, split2, strlen(split2))) {
+		test_msg("Data in the buffer doesn't match what it should "
+			 "in the last chunk\n");
+		ret = -EINVAL;
+		goto out;
+	}
+out:
+	btrfs_free_path(path);
+	kfree(root);
+	return ret;
+}
+
+int btrfs_test_extent_buffer_operations(void)
+{
+	test_msg("Running extent buffer operation tests");
+	return test_btrfs_split_item();
+}
diff --git a/fs/btrfs/tests/extent-io-tests.c b/fs/btrfs/tests/extent-io-tests.c
new file mode 100644
index 000000000000..7e99c2f98dd0
--- /dev/null
+++ b/fs/btrfs/tests/extent-io-tests.c
@@ -0,0 +1,276 @@
+/*
+ * Copyright (C) 2013 Fusion IO.  All rights reserved.
+ *
+ * This program is free software; you can redistribute it and/or
+ * modify it under the terms of the GNU General Public
+ * License v2 as published by the Free Software Foundation.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the GNU
+ * General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public
+ * License along with this program; if not, write to the
+ * Free Software Foundation, Inc., 59 Temple Place - Suite 330,
+ * Boston, MA 021110-1307, USA.
+ */
+
+#include <linux/pagemap.h>
+#include <linux/sched.h>
+#include "btrfs-tests.h"
+#include "../extent_io.h"
+
+#define PROCESS_UNLOCK		(1 << 0)
+#define PROCESS_RELEASE		(1 << 1)
+#define PROCESS_TEST_LOCKED	(1 << 2)
+
+static noinline int process_page_range(struct inode *inode, u64 start, u64 end,
+				       unsigned long flags)
+{
+	int ret;
+	struct page *pages[16];
+	unsigned long index = start >> PAGE_CACHE_SHIFT;
+	unsigned long end_index = end >> PAGE_CACHE_SHIFT;
+	unsigned long nr_pages = end_index - index + 1;
+	int i;
+	int count = 0;
+	int loops = 0;
+
+	while (nr_pages > 0) {
+		ret = find_get_pages_contig(inode->i_mapping, index,
+				     min_t(unsigned long, nr_pages,
+				     ARRAY_SIZE(pages)), pages);
+		for (i = 0; i < ret; i++) {
+			if (flags & PROCESS_TEST_LOCKED &&
+			    !PageLocked(pages[i]))
+				count++;
+			if (flags & PROCESS_UNLOCK && PageLocked(pages[i]))
+				unlock_page(pages[i]);
+			page_cache_release(pages[i]);
+			if (flags & PROCESS_RELEASE)
+				page_cache_release(pages[i]);
+		}
+		nr_pages -= ret;
+		index += ret;
+		cond_resched();
+		loops++;
+		if (loops > 100000) {
+			printk(KERN_ERR "stuck in a loop, start %Lu, end %Lu, nr_pages %lu, ret %d\n", start, end, nr_pages, ret);
+			break;
+		}
+	}
+	return count;
+}
+
+static int test_find_delalloc(void)
+{
+	struct inode *inode;
+	struct extent_io_tree tmp;
+	struct page *page;
+	struct page *locked_page = NULL;
+	unsigned long index = 0;
+	u64 total_dirty = 256 * 1024 * 1024;
+	u64 max_bytes = 128 * 1024 * 1024;
+	u64 start, end, test_start;
+	u64 found;
+	int ret = -EINVAL;
+
+	inode = btrfs_new_test_inode();
+	if (!inode) {
+		test_msg("Failed to allocate test inode\n");
+		return -ENOMEM;
+	}
+
+	extent_io_tree_init(&tmp, &inode->i_data);
+
+	/*
+	 * First go through and create and mark all of our pages dirty, we pin
+	 * everything to make sure our pages don't get evicted and screw up our
+	 * test.
+	 */
+	for (index = 0; index < (total_dirty >> PAGE_CACHE_SHIFT); index++) {
+		page = find_or_create_page(inode->i_mapping, index, GFP_NOFS);
+		if (!page) {
+			test_msg("Failed to allocate test page\n");
+			ret = -ENOMEM;
+			goto out;
+		}
+		SetPageDirty(page);
+		if (index) {
+			unlock_page(page);
+		} else {
+			page_cache_get(page);
+			locked_page = page;
+		}
+	}
+
+	/* Test this scenario
+	 * |--- delalloc ---|
+	 * |---  search  ---|
+	 */
+	set_extent_delalloc(&tmp, 0, 4095, NULL, GFP_NOFS);
+	start = 0;
+	end = 0;
+	found = find_lock_delalloc_range(inode, &tmp, locked_page, &start,
+					 &end, max_bytes);
+	if (!found) {
+		test_msg("Should have found at least one delalloc\n");
+		goto out_bits;
+	}
+	if (start != 0 || end != 4095) {
+		test_msg("Expected start 0 end 4095, got start %Lu end %Lu\n",
+			 start, end);
+		goto out_bits;
+	}
+	unlock_extent(&tmp, start, end);
+	unlock_page(locked_page);
+	page_cache_release(locked_page);
+
+	/*
+	 * Test this scenario
+	 *
+	 * |--- delalloc ---|
+	 *           |--- search ---|
+	 */
+	test_start = 64 * 1024 * 1024;
+	locked_page = find_lock_page(inode->i_mapping,
+				     test_start >> PAGE_CACHE_SHIFT);
+	if (!locked_page) {
+		test_msg("Couldn't find the locked page\n");
+		goto out_bits;
+	}
+	set_extent_delalloc(&tmp, 4096, max_bytes - 1, NULL, GFP_NOFS);
+	start = test_start;
+	end = 0;
+	found = find_lock_delalloc_range(inode, &tmp, locked_page, &start,
+					 &end, max_bytes);
+	if (!found) {
+		test_msg("Couldn't find delalloc in our range\n");
+		goto out_bits;
+	}
+	if (start != test_start || end != max_bytes - 1) {
+		test_msg("Expected start %Lu end %Lu, got start %Lu, end "
+			 "%Lu\n", test_start, max_bytes - 1, start, end);
+		goto out_bits;
+	}
+	if (process_page_range(inode, start, end,
+			       PROCESS_TEST_LOCKED | PROCESS_UNLOCK)) {
+		test_msg("There were unlocked pages in the range\n");
+		goto out_bits;
+	}
+	unlock_extent(&tmp, start, end);
+	/* locked_page was unlocked above */
+	page_cache_release(locked_page);
+
+	/*
+	 * Test this scenario
+	 * |--- delalloc ---|
+	 *                    |--- search ---|
+	 */
+	test_start = max_bytes + 4096;
+	locked_page = find_lock_page(inode->i_mapping, test_start >>
+				     PAGE_CACHE_SHIFT);
+	if (!locked_page) {
+		test_msg("Could'nt find the locked page\n");
+		goto out_bits;
+	}
+	start = test_start;
+	end = 0;
+	found = find_lock_delalloc_range(inode, &tmp, locked_page, &start,
+					 &end, max_bytes);
+	if (found) {
+		test_msg("Found range when we shouldn't have\n");
+		goto out_bits;
+	}
+	if (end != (u64)-1) {
+		test_msg("Did not return the proper end offset\n");
+		goto out_bits;
+	}
+
+	/*
+	 * Test this scenario
+	 * [------- delalloc -------|
+	 * [max_bytes]|-- search--|
+	 *
+	 * We are re-using our test_start from above since it works out well.
+	 */
+	set_extent_delalloc(&tmp, max_bytes, total_dirty - 1, NULL, GFP_NOFS);
+	start = test_start;
+	end = 0;
+	found = find_lock_delalloc_range(inode, &tmp, locked_page, &start,
+					 &end, max_bytes);
+	if (!found) {
+		test_msg("Didn't find our range\n");
+		goto out_bits;
+	}
+	if (start != test_start || end != total_dirty - 1) {
+		test_msg("Expected start %Lu end %Lu, got start %Lu end %Lu\n",
+			 test_start, total_dirty - 1, start, end);
+		goto out_bits;
+	}
+	if (process_page_range(inode, start, end,
+			       PROCESS_TEST_LOCKED | PROCESS_UNLOCK)) {
+		test_msg("Pages in range were not all locked\n");
+		goto out_bits;
+	}
+	unlock_extent(&tmp, start, end);
+
+	/*
+	 * Now to test where we run into a page that is no longer dirty in the
+	 * range we want to find.
+	 */
+	page = find_get_page(inode->i_mapping, (max_bytes + (1 * 1024 * 1024))
+			     >> PAGE_CACHE_SHIFT);
+	if (!page) {
+		test_msg("Couldn't find our page\n");
+		goto out_bits;
+	}
+	ClearPageDirty(page);
+	page_cache_release(page);
+
+	/* We unlocked it in the previous test */
+	lock_page(locked_page);
+	start = test_start;
+	end = 0;
+	/*
+	 * Currently if we fail to find dirty pages in the delalloc range we
+	 * will adjust max_bytes down to PAGE_CACHE_SIZE and then re-search.  If
+	 * this changes at any point in the future we will need to fix this
+	 * tests expected behavior.
+	 */
+	found = find_lock_delalloc_range(inode, &tmp, locked_page, &start,
+					 &end, max_bytes);
+	if (!found) {
+		test_msg("Didn't find our range\n");
+		goto out_bits;
+	}
+	if (start != test_start && end != test_start + PAGE_CACHE_SIZE - 1) {
+		test_msg("Expected start %Lu end %Lu, got start %Lu end %Lu\n",
+			 test_start, test_start + PAGE_CACHE_SIZE - 1, start,
+			 end);
+		goto out_bits;
+	}
+	if (process_page_range(inode, start, end, PROCESS_TEST_LOCKED |
+			       PROCESS_UNLOCK)) {
+		test_msg("Pages in range were not all locked\n");
+		goto out_bits;
+	}
+	ret = 0;
+out_bits:
+	clear_extent_bits(&tmp, 0, total_dirty - 1,
+			  (unsigned long)-1, GFP_NOFS);
+out:
+	if (locked_page)
+		page_cache_release(locked_page);
+	process_page_range(inode, 0, total_dirty - 1,
+			   PROCESS_UNLOCK | PROCESS_RELEASE);
+	iput(inode);
+	return ret;
+}
+
+int btrfs_test_extent_io(void)
+{
+	test_msg("Running find delalloc tests\n");
+	return test_find_delalloc();
+}
diff --git a/fs/btrfs/tests/inode-tests.c b/fs/btrfs/tests/inode-tests.c
new file mode 100644
index 000000000000..397d1f99a8eb
--- /dev/null
+++ b/fs/btrfs/tests/inode-tests.c
@@ -0,0 +1,955 @@
+/*
+ * Copyright (C) 2013 Fusion IO.  All rights reserved.
+ *
+ * This program is free software; you can redistribute it and/or
+ * modify it under the terms of the GNU General Public
+ * License v2 as published by the Free Software Foundation.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the GNU
+ * General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public
+ * License along with this program; if not, write to the
+ * Free Software Foundation, Inc., 59 Temple Place - Suite 330,
+ * Boston, MA 021110-1307, USA.
+ */
+
+#include "btrfs-tests.h"
+#include "../ctree.h"
+#include "../btrfs_inode.h"
+#include "../disk-io.h"
+#include "../extent_io.h"
+#include "../volumes.h"
+
+static struct btrfs_fs_info *alloc_dummy_fs_info(void)
+{
+	struct btrfs_fs_info *fs_info = kzalloc(sizeof(struct btrfs_fs_info),
+						GFP_NOFS);
+	if (!fs_info)
+		return fs_info;
+	fs_info->fs_devices = kzalloc(sizeof(struct btrfs_fs_devices),
+				      GFP_NOFS);
+	if (!fs_info->fs_devices) {
+		kfree(fs_info);
+		return NULL;
+	}
+	return fs_info;
+}
+static void free_dummy_root(struct btrfs_root *root)
+{
+	if (!root)
+		return;
+	if (root->fs_info) {
+		kfree(root->fs_info->fs_devices);
+		kfree(root->fs_info);
+	}
+	if (root->node)
+		free_extent_buffer(root->node);
+	kfree(root);
+}
+
+static void insert_extent(struct btrfs_root *root, u64 start, u64 len,
+			  u64 ram_bytes, u64 offset, u64 disk_bytenr,
+			  u64 disk_len, u32 type, u8 compression, int slot)
+{
+	struct btrfs_path path;
+	struct btrfs_file_extent_item *fi;
+	struct extent_buffer *leaf = root->node;
+	struct btrfs_key key;
+	u32 value_len = sizeof(struct btrfs_file_extent_item);
+
+	if (type == BTRFS_FILE_EXTENT_INLINE)
+		value_len += len;
+	memset(&path, 0, sizeof(path));
+
+	path.nodes[0] = leaf;
+	path.slots[0] = slot;
+
+	key.objectid = BTRFS_FIRST_FREE_OBJECTID;
+	key.type = BTRFS_EXTENT_DATA_KEY;
+	key.offset = start;
+
+	setup_items_for_insert(root, &path, &key, &value_len, value_len,
+			       value_len + sizeof(struct btrfs_item), 1);
+	fi = btrfs_item_ptr(leaf, slot, struct btrfs_file_extent_item);
+	btrfs_set_file_extent_generation(leaf, fi, 1);
+	btrfs_set_file_extent_type(leaf, fi, type);
+	btrfs_set_file_extent_disk_bytenr(leaf, fi, disk_bytenr);
+	btrfs_set_file_extent_disk_num_bytes(leaf, fi, disk_len);
+	btrfs_set_file_extent_offset(leaf, fi, offset);
+	btrfs_set_file_extent_num_bytes(leaf, fi, len);
+	btrfs_set_file_extent_ram_bytes(leaf, fi, ram_bytes);
+	btrfs_set_file_extent_compression(leaf, fi, compression);
+	btrfs_set_file_extent_encryption(leaf, fi, 0);
+	btrfs_set_file_extent_other_encoding(leaf, fi, 0);
+}
+
+static void insert_inode_item_key(struct btrfs_root *root)
+{
+	struct btrfs_path path;
+	struct extent_buffer *leaf = root->node;
+	struct btrfs_key key;
+	u32 value_len = 0;
+
+	memset(&path, 0, sizeof(path));
+
+	path.nodes[0] = leaf;
+	path.slots[0] = 0;
+
+	key.objectid = BTRFS_INODE_ITEM_KEY;
+	key.type = BTRFS_INODE_ITEM_KEY;
+	key.offset = 0;
+
+	setup_items_for_insert(root, &path, &key, &value_len, value_len,
+			       value_len + sizeof(struct btrfs_item), 1);
+}
+
+/*
+ * Build the most complicated map of extents the earth has ever seen.  We want
+ * this so we can test all of the corner cases of btrfs_get_extent.  Here is a
+ * diagram of how the extents will look though this may not be possible we still
+ * want to make sure everything acts normally (the last number is not inclusive)
+ *
+ * [0 - 5][5 -  6][6 - 10][10 - 4096][  4096 - 8192 ][8192 - 12288]
+ * [hole ][inline][ hole ][ regular ][regular1 split][    hole    ]
+ *
+ * [ 12288 - 20480][20480 - 24576][  24576 - 28672  ][28672 - 36864][36864 - 45056]
+ * [regular1 split][   prealloc1 ][prealloc1 written][   prealloc1 ][ compressed  ]
+ *
+ * [45056 - 49152][49152-53248][53248-61440][61440-65536][     65536+81920   ]
+ * [ compressed1 ][  regular  ][compressed1][  regular  ][ hole but no extent]
+ *
+ * [81920-86016]
+ * [  regular  ]
+ */
+static void setup_file_extents(struct btrfs_root *root)
+{
+	int slot = 0;
+	u64 disk_bytenr = 1 * 1024 * 1024;
+	u64 offset = 0;
+
+	/* First we want a hole */
+	insert_extent(root, offset, 5, 5, 0, 0, 0, BTRFS_FILE_EXTENT_REG, 0,
+		      slot);
+	slot++;
+	offset += 5;
+
+	/*
+	 * Now we want an inline extent, I don't think this is possible but hey
+	 * why not?  Also keep in mind if we have an inline extent it counts as
+	 * the whole first page.  If we were to expand it we would have to cow
+	 * and we wouldn't have an inline extent anymore.
+	 */
+	insert_extent(root, offset, 1, 1, 0, 0, 0, BTRFS_FILE_EXTENT_INLINE, 0,
+		      slot);
+	slot++;
+	offset = 4096;
+
+	/* Now another hole */
+	insert_extent(root, offset, 4, 4, 0, 0, 0, BTRFS_FILE_EXTENT_REG, 0,
+		      slot);
+	slot++;
+	offset += 4;
+
+	/* Now for a regular extent */
+	insert_extent(root, offset, 4095, 4095, 0, disk_bytenr, 4096,
+		      BTRFS_FILE_EXTENT_REG, 0, slot);
+	slot++;
+	disk_bytenr += 4096;
+	offset += 4095;
+
+	/*
+	 * Now for 3 extents that were split from a hole punch so we test
+	 * offsets properly.
+	 */
+	insert_extent(root, offset, 4096, 16384, 0, disk_bytenr, 16384,
+		      BTRFS_FILE_EXTENT_REG, 0, slot);
+	slot++;
+	offset += 4096;
+	insert_extent(root, offset, 4096, 4096, 0, 0, 0, BTRFS_FILE_EXTENT_REG,
+		      0, slot);
+	slot++;
+	offset += 4096;
+	insert_extent(root, offset, 8192, 16384, 8192, disk_bytenr, 16384,
+		      BTRFS_FILE_EXTENT_REG, 0, slot);
+	slot++;
+	offset += 8192;
+	disk_bytenr += 16384;
+
+	/* Now for a unwritten prealloc extent */
+	insert_extent(root, offset, 4096, 4096, 0, disk_bytenr, 4096,
+		      BTRFS_FILE_EXTENT_PREALLOC, 0, slot);
+	slot++;
+	offset += 4096;
+
+	/*
+	 * We want to jack up disk_bytenr a little more so the em stuff doesn't
+	 * merge our records.
+	 */
+	disk_bytenr += 8192;
+
+	/*
+	 * Now for a partially written prealloc extent, basically the same as
+	 * the hole punch example above.  Ram_bytes never changes when you mark
+	 * extents written btw.
+	 */
+	insert_extent(root, offset, 4096, 16384, 0, disk_bytenr, 16384,
+		      BTRFS_FILE_EXTENT_PREALLOC, 0, slot);
+	slot++;
+	offset += 4096;
+	insert_extent(root, offset, 4096, 16384, 4096, disk_bytenr, 16384,
+		      BTRFS_FILE_EXTENT_REG, 0, slot);
+	slot++;
+	offset += 4096;
+	insert_extent(root, offset, 8192, 16384, 8192, disk_bytenr, 16384,
+		      BTRFS_FILE_EXTENT_PREALLOC, 0, slot);
+	slot++;
+	offset += 8192;
+	disk_bytenr += 16384;
+
+	/* Now a normal compressed extent */
+	insert_extent(root, offset, 8192, 8192, 0, disk_bytenr, 4096,
+		      BTRFS_FILE_EXTENT_REG, BTRFS_COMPRESS_ZLIB, slot);
+	slot++;
+	offset += 8192;
+	/* No merges */
+	disk_bytenr += 8192;
+
+	/* Now a split compressed extent */
+	insert_extent(root, offset, 4096, 16384, 0, disk_bytenr, 4096,
+		      BTRFS_FILE_EXTENT_REG, BTRFS_COMPRESS_ZLIB, slot);
+	slot++;
+	offset += 4096;
+	insert_extent(root, offset, 4096, 4096, 0, disk_bytenr + 4096, 4096,
+		      BTRFS_FILE_EXTENT_REG, 0, slot);
+	slot++;
+	offset += 4096;
+	insert_extent(root, offset, 8192, 16384, 8192, disk_bytenr, 4096,
+		      BTRFS_FILE_EXTENT_REG, BTRFS_COMPRESS_ZLIB, slot);
+	slot++;
+	offset += 8192;
+	disk_bytenr += 8192;
+
+	/* Now extents that have a hole but no hole extent */
+	insert_extent(root, offset, 4096, 4096, 0, disk_bytenr, 4096,
+		      BTRFS_FILE_EXTENT_REG, 0, slot);
+	slot++;
+	offset += 16384;
+	disk_bytenr += 4096;
+	insert_extent(root, offset, 4096, 4096, 0, disk_bytenr, 4096,
+		      BTRFS_FILE_EXTENT_REG, 0, slot);
+}
+
+static unsigned long prealloc_only = 0;
+static unsigned long compressed_only = 0;
+static unsigned long vacancy_only = 0;
+
+static noinline int test_btrfs_get_extent(void)
+{
+	struct inode *inode = NULL;
+	struct btrfs_root *root = NULL;
+	struct extent_map *em = NULL;
+	u64 orig_start;
+	u64 disk_bytenr;
+	u64 offset;
+	int ret = -ENOMEM;
+
+	inode = btrfs_new_test_inode();
+	if (!inode) {
+		test_msg("Couldn't allocate inode\n");
+		return ret;
+	}
+
+	BTRFS_I(inode)->location.type = BTRFS_INODE_ITEM_KEY;
+	BTRFS_I(inode)->location.objectid = BTRFS_FIRST_FREE_OBJECTID;
+	BTRFS_I(inode)->location.offset = 0;
+
+	root = btrfs_alloc_dummy_root();
+	if (IS_ERR(root)) {
+		test_msg("Couldn't allocate root\n");
+		goto out;
+	}
+
+	/*
+	 * We do this since btrfs_get_extent wants to assign em->bdev to
+	 * root->fs_info->fs_devices->latest_bdev.
+	 */
+	root->fs_info = alloc_dummy_fs_info();
+	if (!root->fs_info) {
+		test_msg("Couldn't allocate dummy fs info\n");
+		goto out;
+	}
+
+	root->node = alloc_dummy_extent_buffer(0, 4096);
+	if (!root->node) {
+		test_msg("Couldn't allocate dummy buffer\n");
+		goto out;
+	}
+
+	/*
+	 * We will just free a dummy node if it's ref count is 2 so we need an
+	 * extra ref so our searches don't accidently release our page.
+	 */
+	extent_buffer_get(root->node);
+	btrfs_set_header_nritems(root->node, 0);
+	btrfs_set_header_level(root->node, 0);
+	ret = -EINVAL;
+
+	/* First with no extents */
+	BTRFS_I(inode)->root = root;
+	em = btrfs_get_extent(inode, NULL, 0, 0, 4096, 0);
+	if (IS_ERR(em)) {
+		em = NULL;
+		test_msg("Got an error when we shouldn't have\n");
+		goto out;
+	}
+	if (em->block_start != EXTENT_MAP_HOLE) {
+		test_msg("Expected a hole, got %llu\n", em->block_start);
+		goto out;
+	}
+	if (!test_bit(EXTENT_FLAG_VACANCY, &em->flags)) {
+		test_msg("Vacancy flag wasn't set properly\n");
+		goto out;
+	}
+	free_extent_map(em);
+	btrfs_drop_extent_cache(inode, 0, (u64)-1, 0);
+
+	/*
+	 * All of the magic numbers are based on the mapping setup in
+	 * setup_file_extents, so if you change anything there you need to
+	 * update the comment and update the expected values below.
+	 */
+	setup_file_extents(root);
+
+	em = btrfs_get_extent(inode, NULL, 0, 0, (u64)-1, 0);
+	if (IS_ERR(em)) {
+		test_msg("Got an error when we shouldn't have\n");
+		goto out;
+	}
+	if (em->block_start != EXTENT_MAP_HOLE) {
+		test_msg("Expected a hole, got %llu\n", em->block_start);
+		goto out;
+	}
+	if (em->start != 0 || em->len != 5) {
+		test_msg("Unexpected extent wanted start 0 len 5, got start "
+			 "%llu len %llu\n", em->start, em->len);
+		goto out;
+	}
+	if (em->flags != 0) {
+		test_msg("Unexpected flags set, want 0 have %lu\n", em->flags);
+		goto out;
+	}
+	offset = em->start + em->len;
+	free_extent_map(em);
+
+	em = btrfs_get_extent(inode, NULL, 0, offset, 4096, 0);
+	if (IS_ERR(em)) {
+		test_msg("Got an error when we shouldn't have\n");
+		goto out;
+	}
+	if (em->block_start != EXTENT_MAP_INLINE) {
+		test_msg("Expected an inline, got %llu\n", em->block_start);
+		goto out;
+	}
+	if (em->start != offset || em->len != 4091) {
+		test_msg("Unexpected extent wanted start %llu len 1, got start "
+			 "%llu len %llu\n", offset, em->start, em->len);
+		goto out;
+	}
+	if (em->flags != 0) {
+		test_msg("Unexpected flags set, want 0 have %lu\n", em->flags);
+		goto out;
+	}
+	/*
+	 * We don't test anything else for inline since it doesn't get set
+	 * unless we have a page for it to write into.  Maybe we should change
+	 * this?
+	 */
+	offset = em->start + em->len;
+	free_extent_map(em);
+
+	em = btrfs_get_extent(inode, NULL, 0, offset, 4096, 0);
+	if (IS_ERR(em)) {
+		test_msg("Got an error when we shouldn't have\n");
+		goto out;
+	}
+	if (em->block_start != EXTENT_MAP_HOLE) {
+		test_msg("Expected a hole, got %llu\n", em->block_start);
+		goto out;
+	}
+	if (em->start != offset || em->len != 4) {
+		test_msg("Unexpected extent wanted start %llu len 4, got start "
+			 "%llu len %llu\n", offset, em->start, em->len);
+		goto out;
+	}
+	if (em->flags != 0) {
+		test_msg("Unexpected flags set, want 0 have %lu\n", em->flags);
+		goto out;
+	}
+	offset = em->start + em->len;
+	free_extent_map(em);
+
+	/* Regular extent */
+	em = btrfs_get_extent(inode, NULL, 0, offset, 4096, 0);
+	if (IS_ERR(em)) {
+		test_msg("Got an error when we shouldn't have\n");
+		goto out;
+	}
+	if (em->block_start >= EXTENT_MAP_LAST_BYTE) {
+		test_msg("Expected a real extent, got %llu\n", em->block_start);
+		goto out;
+	}
+	if (em->start != offset || em->len != 4095) {
+		test_msg("Unexpected extent wanted start %llu len 4095, got "
+			 "start %llu len %llu\n", offset, em->start, em->len);
+		goto out;
+	}
+	if (em->flags != 0) {
+		test_msg("Unexpected flags set, want 0 have %lu\n", em->flags);
+		goto out;
+	}
+	if (em->orig_start != em->start) {
+		test_msg("Wrong orig offset, want %llu, have %llu\n", em->start,
+			 em->orig_start);
+		goto out;
+	}
+	offset = em->start + em->len;
+	free_extent_map(em);
+
+	/* The next 3 are split extents */
+	em = btrfs_get_extent(inode, NULL, 0, offset, 4096, 0);
+	if (IS_ERR(em)) {
+		test_msg("Got an error when we shouldn't have\n");
+		goto out;
+	}
+	if (em->block_start >= EXTENT_MAP_LAST_BYTE) {
+		test_msg("Expected a real extent, got %llu\n", em->block_start);
+		goto out;
+	}
+	if (em->start != offset || em->len != 4096) {
+		test_msg("Unexpected extent wanted start %llu len 4096, got "
+			 "start %llu len %llu\n", offset, em->start, em->len);
+		goto out;
+	}
+	if (em->flags != 0) {
+		test_msg("Unexpected flags set, want 0 have %lu\n", em->flags);
+		goto out;
+	}
+	if (em->orig_start != em->start) {
+		test_msg("Wrong orig offset, want %llu, have %llu\n", em->start,
+			 em->orig_start);
+		goto out;
+	}
+	disk_bytenr = em->block_start;
+	orig_start = em->start;
+	offset = em->start + em->len;
+	free_extent_map(em);
+
+	em = btrfs_get_extent(inode, NULL, 0, offset, 4096, 0);
+	if (IS_ERR(em)) {
+		test_msg("Got an error when we shouldn't have\n");
+		goto out;
+	}
+	if (em->block_start != EXTENT_MAP_HOLE) {
+		test_msg("Expected a hole, got %llu\n", em->block_start);
+		goto out;
+	}
+	if (em->start != offset || em->len != 4096) {
+		test_msg("Unexpected extent wanted start %llu len 4096, got "
+			 "start %llu len %llu\n", offset, em->start, em->len);
+		goto out;
+	}
+	if (em->flags != 0) {
+		test_msg("Unexpected flags set, want 0 have %lu\n", em->flags);
+		goto out;
+	}
+	offset = em->start + em->len;
+	free_extent_map(em);
+
+	em = btrfs_get_extent(inode, NULL, 0, offset, 4096, 0);
+	if (IS_ERR(em)) {
+		test_msg("Got an error when we shouldn't have\n");
+		goto out;
+	}
+	if (em->block_start >= EXTENT_MAP_LAST_BYTE) {
+		test_msg("Expected a real extent, got %llu\n", em->block_start);
+		goto out;
+	}
+	if (em->start != offset || em->len != 8192) {
+		test_msg("Unexpected extent wanted start %llu len 8192, got "
+			 "start %llu len %llu\n", offset, em->start, em->len);
+		goto out;
+	}
+	if (em->flags != 0) {
+		test_msg("Unexpected flags set, want 0 have %lu\n", em->flags);
+		goto out;
+	}
+	if (em->orig_start != orig_start) {
+		test_msg("Wrong orig offset, want %llu, have %llu\n",
+			 orig_start, em->orig_start);
+		goto out;
+	}
+	disk_bytenr += (em->start - orig_start);
+	if (em->block_start != disk_bytenr) {
+		test_msg("Wrong block start, want %llu, have %llu\n",
+			 disk_bytenr, em->block_start);
+		goto out;
+	}
+	offset = em->start + em->len;
+	free_extent_map(em);
+
+	/* Prealloc extent */
+	em = btrfs_get_extent(inode, NULL, 0, offset, 4096, 0);
+	if (IS_ERR(em)) {
+		test_msg("Got an error when we shouldn't have\n");
+		goto out;
+	}
+	if (em->block_start >= EXTENT_MAP_LAST_BYTE) {
+		test_msg("Expected a real extent, got %llu\n", em->block_start);
+		goto out;
+	}
+	if (em->start != offset || em->len != 4096) {
+		test_msg("Unexpected extent wanted start %llu len 4096, got "
+			 "start %llu len %llu\n", offset, em->start, em->len);
+		goto out;
+	}
+	if (em->flags != prealloc_only) {
+		test_msg("Unexpected flags set, want %lu have %lu\n",
+			 prealloc_only, em->flags);
+		goto out;
+	}
+	if (em->orig_start != em->start) {
+		test_msg("Wrong orig offset, want %llu, have %llu\n", em->start,
+			 em->orig_start);
+		goto out;
+	}
+	offset = em->start + em->len;
+	free_extent_map(em);
+
+	/* The next 3 are a half written prealloc extent */
+	em = btrfs_get_extent(inode, NULL, 0, offset, 4096, 0);
+	if (IS_ERR(em)) {
+		test_msg("Got an error when we shouldn't have\n");
+		goto out;
+	}
+	if (em->block_start >= EXTENT_MAP_LAST_BYTE) {
+		test_msg("Expected a real extent, got %llu\n", em->block_start);
+		goto out;
+	}
+	if (em->start != offset || em->len != 4096) {
+		test_msg("Unexpected extent wanted start %llu len 4096, got "
+			 "start %llu len %llu\n", offset, em->start, em->len);
+		goto out;
+	}
+	if (em->flags != prealloc_only) {
+		test_msg("Unexpected flags set, want %lu have %lu\n",
+			 prealloc_only, em->flags);
+		goto out;
+	}
+	if (em->orig_start != em->start) {
+		test_msg("Wrong orig offset, want %llu, have %llu\n", em->start,
+			 em->orig_start);
+		goto out;
+	}
+	disk_bytenr = em->block_start;
+	orig_start = em->start;
+	offset = em->start + em->len;
+	free_extent_map(em);
+
+	em = btrfs_get_extent(inode, NULL, 0, offset, 4096, 0);
+	if (IS_ERR(em)) {
+		test_msg("Got an error when we shouldn't have\n");
+		goto out;
+	}
+	if (em->block_start >= EXTENT_MAP_HOLE) {
+		test_msg("Expected a real extent, got %llu\n", em->block_start);
+		goto out;
+	}
+	if (em->start != offset || em->len != 4096) {
+		test_msg("Unexpected extent wanted start %llu len 4096, got "
+			 "start %llu len %llu\n", offset, em->start, em->len);
+		goto out;
+	}
+	if (em->flags != 0) {
+		test_msg("Unexpected flags set, want 0 have %lu\n", em->flags);
+		goto out;
+	}
+	if (em->orig_start != orig_start) {
+		test_msg("Unexpected orig offset, wanted %llu, have %llu\n",
+			 orig_start, em->orig_start);
+		goto out;
+	}
+	if (em->block_start != (disk_bytenr + (em->start - em->orig_start))) {
+		test_msg("Unexpected block start, wanted %llu, have %llu\n",
+			 disk_bytenr + (em->start - em->orig_start),
+			 em->block_start);
+		goto out;
+	}
+	offset = em->start + em->len;
+	free_extent_map(em);
+
+	em = btrfs_get_extent(inode, NULL, 0, offset, 4096, 0);
+	if (IS_ERR(em)) {
+		test_msg("Got an error when we shouldn't have\n");
+		goto out;
+	}
+	if (em->block_start >= EXTENT_MAP_LAST_BYTE) {
+		test_msg("Expected a real extent, got %llu\n", em->block_start);
+		goto out;
+	}
+	if (em->start != offset || em->len != 8192) {
+		test_msg("Unexpected extent wanted start %llu len 8192, got "
+			 "start %llu len %llu\n", offset, em->start, em->len);
+		goto out;
+	}
+	if (em->flags != prealloc_only) {
+		test_msg("Unexpected flags set, want %lu have %lu\n",
+			 prealloc_only, em->flags);
+		goto out;
+	}
+	if (em->orig_start != orig_start) {
+		test_msg("Wrong orig offset, want %llu, have %llu\n", orig_start,
+			 em->orig_start);
+		goto out;
+	}
+	if (em->block_start != (disk_bytenr + (em->start - em->orig_start))) {
+		test_msg("Unexpected block start, wanted %llu, have %llu\n",
+			 disk_bytenr + (em->start - em->orig_start),
+			 em->block_start);
+		goto out;
+	}
+	offset = em->start + em->len;
+	free_extent_map(em);
+
+	/* Now for the compressed extent */
+	em = btrfs_get_extent(inode, NULL, 0, offset, 4096, 0);
+	if (IS_ERR(em)) {
+		test_msg("Got an error when we shouldn't have\n");
+		goto out;
+	}
+	if (em->block_start >= EXTENT_MAP_LAST_BYTE) {
+		test_msg("Expected a real extent, got %llu\n", em->block_start);
+		goto out;
+	}
+	if (em->start != offset || em->len != 8192) {
+		test_msg("Unexpected extent wanted start %llu len 8192, got "
+			 "start %llu len %llu\n", offset, em->start, em->len);
+		goto out;
+	}
+	if (em->flags != compressed_only) {
+		test_msg("Unexpected flags set, want %lu have %lu\n",
+			 compressed_only, em->flags);
+		goto out;
+	}
+	if (em->orig_start != em->start) {
+		test_msg("Wrong orig offset, want %llu, have %llu\n",
+			 em->start, em->orig_start);
+		goto out;
+	}
+	if (em->compress_type != BTRFS_COMPRESS_ZLIB) {
+		test_msg("Unexpected compress type, wanted %d, got %d\n",
+			 BTRFS_COMPRESS_ZLIB, em->compress_type);
+		goto out;
+	}
+	offset = em->start + em->len;
+	free_extent_map(em);
+
+	/* Split compressed extent */
+	em = btrfs_get_extent(inode, NULL, 0, offset, 4096, 0);
+	if (IS_ERR(em)) {
+		test_msg("Got an error when we shouldn't have\n");
+		goto out;
+	}
+	if (em->block_start >= EXTENT_MAP_LAST_BYTE) {
+		test_msg("Expected a real extent, got %llu\n", em->block_start);
+		goto out;
+	}
+	if (em->start != offset || em->len != 4096) {
+		test_msg("Unexpected extent wanted start %llu len 4096, got "
+			 "start %llu len %llu\n", offset, em->start, em->len);
+		goto out;
+	}
+	if (em->flags != compressed_only) {
+		test_msg("Unexpected flags set, want %lu have %lu\n",
+			 compressed_only, em->flags);
+		goto out;
+	}
+	if (em->orig_start != em->start) {
+		test_msg("Wrong orig offset, want %llu, have %llu\n",
+			 em->start, em->orig_start);
+		goto out;
+	}
+	if (em->compress_type != BTRFS_COMPRESS_ZLIB) {
+		test_msg("Unexpected compress type, wanted %d, got %d\n",
+			 BTRFS_COMPRESS_ZLIB, em->compress_type);
+		goto out;
+	}
+	disk_bytenr = em->block_start;
+	orig_start = em->start;
+	offset = em->start + em->len;
+	free_extent_map(em);
+
+	em = btrfs_get_extent(inode, NULL, 0, offset, 4096, 0);
+	if (IS_ERR(em)) {
+		test_msg("Got an error when we shouldn't have\n");
+		goto out;
+	}
+	if (em->block_start >= EXTENT_MAP_LAST_BYTE) {
+		test_msg("Expected a real extent, got %llu\n", em->block_start);
+		goto out;
+	}
+	if (em->start != offset || em->len != 4096) {
+		test_msg("Unexpected extent wanted start %llu len 4096, got "
+			 "start %llu len %llu\n", offset, em->start, em->len);
+		goto out;
+	}
+	if (em->flags != 0) {
+		test_msg("Unexpected flags set, want 0 have %lu\n", em->flags);
+		goto out;
+	}
+	if (em->orig_start != em->start) {
+		test_msg("Wrong orig offset, want %llu, have %llu\n", em->start,
+			 em->orig_start);
+		goto out;
+	}
+	offset = em->start + em->len;
+	free_extent_map(em);
+
+	em = btrfs_get_extent(inode, NULL, 0, offset, 4096, 0);
+	if (IS_ERR(em)) {
+		test_msg("Got an error when we shouldn't have\n");
+		goto out;
+	}
+	if (em->block_start != disk_bytenr) {
+		test_msg("Block start does not match, want %llu got %llu\n",
+			 disk_bytenr, em->block_start);
+		goto out;
+	}
+	if (em->start != offset || em->len != 8192) {
+		test_msg("Unexpected extent wanted start %llu len 8192, got "
+			 "start %llu len %llu\n", offset, em->start, em->len);
+		goto out;
+	}
+	if (em->flags != compressed_only) {
+		test_msg("Unexpected flags set, want %lu have %lu\n",
+			 compressed_only, em->flags);
+		goto out;
+	}
+	if (em->orig_start != orig_start) {
+		test_msg("Wrong orig offset, want %llu, have %llu\n",
+			 em->start, orig_start);
+		goto out;
+	}
+	if (em->compress_type != BTRFS_COMPRESS_ZLIB) {
+		test_msg("Unexpected compress type, wanted %d, got %d\n",
+			 BTRFS_COMPRESS_ZLIB, em->compress_type);
+		goto out;
+	}
+	offset = em->start + em->len;
+	free_extent_map(em);
+
+	/* A hole between regular extents but no hole extent */
+	em = btrfs_get_extent(inode, NULL, 0, offset + 6, 4096, 0);
+	if (IS_ERR(em)) {
+		test_msg("Got an error when we shouldn't have\n");
+		goto out;
+	}
+	if (em->block_start >= EXTENT_MAP_LAST_BYTE) {
+		test_msg("Expected a real extent, got %llu\n", em->block_start);
+		goto out;
+	}
+	if (em->start != offset || em->len != 4096) {
+		test_msg("Unexpected extent wanted start %llu len 4096, got "
+			 "start %llu len %llu\n", offset, em->start, em->len);
+		goto out;
+	}
+	if (em->flags != 0) {
+		test_msg("Unexpected flags set, want 0 have %lu\n", em->flags);
+		goto out;
+	}
+	if (em->orig_start != em->start) {
+		test_msg("Wrong orig offset, want %llu, have %llu\n", em->start,
+			 em->orig_start);
+		goto out;
+	}
+	offset = em->start + em->len;
+	free_extent_map(em);
+
+	em = btrfs_get_extent(inode, NULL, 0, offset, 4096 * 1024, 0);
+	if (IS_ERR(em)) {
+		test_msg("Got an error when we shouldn't have\n");
+		goto out;
+	}
+	if (em->block_start != EXTENT_MAP_HOLE) {
+		test_msg("Expected a hole extent, got %llu\n", em->block_start);
+		goto out;
+	}
+	/*
+	 * Currently we just return a length that we requested rather than the
+	 * length of the actual hole, if this changes we'll have to change this
+	 * test.
+	 */
+	if (em->start != offset || em->len != 12288) {
+		test_msg("Unexpected extent wanted start %llu len 12288, got "
+			 "start %llu len %llu\n", offset, em->start, em->len);
+		goto out;
+	}
+	if (em->flags != vacancy_only) {
+		test_msg("Unexpected flags set, want %lu have %lu\n",
+			 vacancy_only, em->flags);
+		goto out;
+	}
+	if (em->orig_start != em->start) {
+		test_msg("Wrong orig offset, want %llu, have %llu\n", em->start,
+			 em->orig_start);
+		goto out;
+	}
+	offset = em->start + em->len;
+	free_extent_map(em);
+
+	em = btrfs_get_extent(inode, NULL, 0, offset, 4096, 0);
+	if (IS_ERR(em)) {
+		test_msg("Got an error when we shouldn't have\n");
+		goto out;
+	}
+	if (em->block_start >= EXTENT_MAP_LAST_BYTE) {
+		test_msg("Expected a real extent, got %llu\n", em->block_start);
+		goto out;
+	}
+	if (em->start != offset || em->len != 4096) {
+		test_msg("Unexpected extent wanted start %llu len 4096, got "
+			 "start %llu len %llu\n", offset, em->start, em->len);
+		goto out;
+	}
+	if (em->flags != 0) {
+		test_msg("Unexpected flags set, want 0 have %lu\n", em->flags);
+		goto out;
+	}
+	if (em->orig_start != em->start) {
+		test_msg("Wrong orig offset, want %llu, have %llu\n", em->start,
+			 em->orig_start);
+		goto out;
+	}
+	ret = 0;
+out:
+	if (!IS_ERR(em))
+		free_extent_map(em);
+	iput(inode);
+	free_dummy_root(root);
+	return ret;
+}
+
+static int test_hole_first(void)
+{
+	struct inode *inode = NULL;
+	struct btrfs_root *root = NULL;
+	struct extent_map *em = NULL;
+	int ret = -ENOMEM;
+
+	inode = btrfs_new_test_inode();
+	if (!inode) {
+		test_msg("Couldn't allocate inode\n");
+		return ret;
+	}
+
+	BTRFS_I(inode)->location.type = BTRFS_INODE_ITEM_KEY;
+	BTRFS_I(inode)->location.objectid = BTRFS_FIRST_FREE_OBJECTID;
+	BTRFS_I(inode)->location.offset = 0;
+
+	root = btrfs_alloc_dummy_root();
+	if (IS_ERR(root)) {
+		test_msg("Couldn't allocate root\n");
+		goto out;
+	}
+
+	root->fs_info = alloc_dummy_fs_info();
+	if (!root->fs_info) {
+		test_msg("Couldn't allocate dummy fs info\n");
+		goto out;
+	}
+
+	root->node = alloc_dummy_extent_buffer(0, 4096);
+	if (!root->node) {
+		test_msg("Couldn't allocate dummy buffer\n");
+		goto out;
+	}
+
+	extent_buffer_get(root->node);
+	btrfs_set_header_nritems(root->node, 0);
+	btrfs_set_header_level(root->node, 0);
+	BTRFS_I(inode)->root = root;
+	ret = -EINVAL;
+
+	/*
+	 * Need a blank inode item here just so we don't confuse
+	 * btrfs_get_extent.
+	 */
+	insert_inode_item_key(root);
+	insert_extent(root, 4096, 4096, 4096, 0, 4096, 4096,
+		      BTRFS_FILE_EXTENT_REG, 0, 1);
+	em = btrfs_get_extent(inode, NULL, 0, 0, 8192, 0);
+	if (IS_ERR(em)) {
+		test_msg("Got an error when we shouldn't have\n");
+		goto out;
+	}
+	if (em->block_start != EXTENT_MAP_HOLE) {
+		test_msg("Expected a hole, got %llu\n", em->block_start);
+		goto out;
+	}
+	if (em->start != 0 || em->len != 4096) {
+		test_msg("Unexpected extent wanted start 0 len 4096, got start "
+			 "%llu len %llu\n", em->start, em->len);
+		goto out;
+	}
+	if (em->flags != vacancy_only) {
+		test_msg("Wrong flags, wanted %lu, have %lu\n", vacancy_only,
+			 em->flags);
+		goto out;
+	}
+	free_extent_map(em);
+
+	em = btrfs_get_extent(inode, NULL, 0, 4096, 8192, 0);
+	if (IS_ERR(em)) {
+		test_msg("Got an error when we shouldn't have\n");
+		goto out;
+	}
+	if (em->block_start != 4096) {
+		test_msg("Expected a real extent, got %llu\n", em->block_start);
+		goto out;
+	}
+	if (em->start != 4096 || em->len != 4096) {
+		test_msg("Unexpected extent wanted start 4096 len 4096, got "
+			 "start %llu len %llu\n", em->start, em->len);
+		goto out;
+	}
+	if (em->flags != 0) {
+		test_msg("Unexpected flags set, wanted 0 got %lu\n",
+			 em->flags);
+		goto out;
+	}
+	ret = 0;
+out:
+	if (!IS_ERR(em))
+		free_extent_map(em);
+	iput(inode);
+	free_dummy_root(root);
+	return ret;
+}
+
+int btrfs_test_inodes(void)
+{
+	int ret;
+
+	set_bit(EXTENT_FLAG_COMPRESSED, &compressed_only);
+	set_bit(EXTENT_FLAG_VACANCY, &vacancy_only);
+	set_bit(EXTENT_FLAG_PREALLOC, &prealloc_only);
+
+	test_msg("Running btrfs_get_extent tests\n");
+	ret = test_btrfs_get_extent();
+	if (ret)
+		return ret;
+	test_msg("Running hole first btrfs_get_extent test\n");
+	return test_hole_first();
+}
diff --git a/fs/btrfs/transaction.c b/fs/btrfs/transaction.c
index 8c81bdc1ef9b..57c16b46afbd 100644
--- a/fs/btrfs/transaction.c
+++ b/fs/btrfs/transaction.c
@@ -57,7 +57,7 @@ static unsigned int btrfs_blocked_trans_types[TRANS_STATE_MAX] = {
 					   __TRANS_JOIN_NOLOCK),
 };
 
-static void put_transaction(struct btrfs_transaction *transaction)
+void btrfs_put_transaction(struct btrfs_transaction *transaction)
 {
 	WARN_ON(atomic_read(&transaction->use_count) == 0);
 	if (atomic_dec_and_test(&transaction->use_count)) {
@@ -332,7 +332,7 @@ static void wait_current_trans(struct btrfs_root *root)
 		wait_event(root->fs_info->transaction_wait,
 			   cur_trans->state >= TRANS_STATE_UNBLOCKED ||
 			   cur_trans->aborted);
-		put_transaction(cur_trans);
+		btrfs_put_transaction(cur_trans);
 	} else {
 		spin_unlock(&root->fs_info->trans_lock);
 	}
@@ -353,6 +353,17 @@ static int may_wait_transaction(struct btrfs_root *root, int type)
 	return 0;
 }
 
+static inline bool need_reserve_reloc_root(struct btrfs_root *root)
+{
+	if (!root->fs_info->reloc_ctl ||
+	    !root->ref_cows ||
+	    root->root_key.objectid == BTRFS_TREE_RELOC_OBJECTID ||
+	    root->reloc_root)
+		return false;
+
+	return true;
+}
+
 static struct btrfs_trans_handle *
 start_transaction(struct btrfs_root *root, u64 num_items, unsigned int type,
 		  enum btrfs_reserve_flush_enum flush)
@@ -360,8 +371,9 @@ start_transaction(struct btrfs_root *root, u64 num_items, unsigned int type,
 	struct btrfs_trans_handle *h;
 	struct btrfs_transaction *cur_trans;
 	u64 num_bytes = 0;
-	int ret;
 	u64 qgroup_reserved = 0;
+	bool reloc_reserved = false;
+	int ret;
 
 	if (test_bit(BTRFS_FS_STATE_ERROR, &root->fs_info->fs_state))
 		return ERR_PTR(-EROFS);
@@ -390,6 +402,14 @@ start_transaction(struct btrfs_root *root, u64 num_items, unsigned int type,
 		}
 
 		num_bytes = btrfs_calc_trans_metadata_size(root, num_items);
+		/*
+		 * Do the reservation for the relocation root creation
+		 */
+		if (unlikely(need_reserve_reloc_root(root))) {
+			num_bytes += root->nodesize;
+			reloc_reserved = true;
+		}
+
 		ret = btrfs_block_rsv_add(root,
 					  &root->fs_info->trans_block_rsv,
 					  num_bytes, flush);
@@ -451,6 +471,7 @@ again:
 	h->delayed_ref_elem.seq = 0;
 	h->type = type;
 	h->allocating_chunk = false;
+	h->reloc_reserved = false;
 	INIT_LIST_HEAD(&h->qgroup_ref_list);
 	INIT_LIST_HEAD(&h->new_bgs);
 
@@ -466,6 +487,7 @@ again:
 					      h->transid, num_bytes, 1);
 		h->block_rsv = &root->fs_info->trans_block_rsv;
 		h->bytes_reserved = num_bytes;
+		h->reloc_reserved = reloc_reserved;
 	}
 	h->qgroup_reserved = qgroup_reserved;
 
@@ -610,7 +632,7 @@ int btrfs_wait_for_commit(struct btrfs_root *root, u64 transid)
 	}
 
 	wait_for_commit(root, cur_trans);
-	put_transaction(cur_trans);
+	btrfs_put_transaction(cur_trans);
 out:
 	return ret;
 }
@@ -735,7 +757,7 @@ static int __btrfs_end_transaction(struct btrfs_trans_handle *trans,
 	smp_mb();
 	if (waitqueue_active(&cur_trans->writer_wait))
 		wake_up(&cur_trans->writer_wait);
-	put_transaction(cur_trans);
+	btrfs_put_transaction(cur_trans);
 
 	if (current->journal_info == trans)
 		current->journal_info = NULL;
@@ -744,8 +766,10 @@ static int __btrfs_end_transaction(struct btrfs_trans_handle *trans,
 		btrfs_run_delayed_iputs(root);
 
 	if (trans->aborted ||
-	    test_bit(BTRFS_FS_STATE_ERROR, &root->fs_info->fs_state))
+	    test_bit(BTRFS_FS_STATE_ERROR, &root->fs_info->fs_state)) {
+		wake_up_process(info->transaction_kthread);
 		err = -EIO;
+	}
 	assert_qgroups_uptodate(trans);
 
 	kmem_cache_free(btrfs_trans_handle_cachep, trans);
@@ -948,16 +972,19 @@ static noinline int commit_cowonly_roots(struct btrfs_trans_handle *trans,
 		return ret;
 
 	ret = btrfs_run_dev_stats(trans, root->fs_info);
-	WARN_ON(ret);
+	if (ret)
+		return ret;
 	ret = btrfs_run_dev_replace(trans, root->fs_info);
-	WARN_ON(ret);
-
+	if (ret)
+		return ret;
 	ret = btrfs_run_qgroups(trans, root->fs_info);
-	BUG_ON(ret);
+	if (ret)
+		return ret;
 
 	/* run_qgroups might have added some more refs */
 	ret = btrfs_run_delayed_refs(trans, root, (unsigned long)-1);
-	BUG_ON(ret);
+	if (ret)
+		return ret;
 
 	while (!list_empty(&fs_info->dirty_cowonly_roots)) {
 		next = fs_info->dirty_cowonly_roots.next;
@@ -1510,7 +1537,7 @@ int btrfs_commit_transaction_async(struct btrfs_trans_handle *trans,
 	if (current->journal_info == trans)
 		current->journal_info = NULL;
 
-	put_transaction(cur_trans);
+	btrfs_put_transaction(cur_trans);
 	return 0;
 }
 
@@ -1552,8 +1579,10 @@ static void cleanup_transaction(struct btrfs_trans_handle *trans,
 		root->fs_info->running_transaction = NULL;
 	spin_unlock(&root->fs_info->trans_lock);
 
-	put_transaction(cur_trans);
-	put_transaction(cur_trans);
+	if (trans->type & __TRANS_FREEZABLE)
+		sb_end_intwrite(root->fs_info->sb);
+	btrfs_put_transaction(cur_trans);
+	btrfs_put_transaction(cur_trans);
 
 	trace_btrfs_transaction_commit(root);
 
@@ -1571,15 +1600,19 @@ static int btrfs_flush_all_pending_stuffs(struct btrfs_trans_handle *trans,
 	int ret;
 
 	ret = btrfs_run_delayed_items(trans, root);
-	if (ret)
-		return ret;
-
 	/*
 	 * running the delayed items may have added new refs. account
 	 * them now so that they hinder processing of more delayed refs
 	 * as little as possible.
 	 */
-	btrfs_delayed_refs_qgroup_accounting(trans, root->fs_info);
+	if (ret) {
+		btrfs_delayed_refs_qgroup_accounting(trans, root->fs_info);
+		return ret;
+	}
+
+	ret = btrfs_delayed_refs_qgroup_accounting(trans, root->fs_info);
+	if (ret)
+		return ret;
 
 	/*
 	 * rename don't use btrfs_join_transaction, so, once we
@@ -1596,14 +1629,14 @@ static int btrfs_flush_all_pending_stuffs(struct btrfs_trans_handle *trans,
 static inline int btrfs_start_delalloc_flush(struct btrfs_fs_info *fs_info)
 {
 	if (btrfs_test_opt(fs_info->tree_root, FLUSHONCOMMIT))
-		return btrfs_start_all_delalloc_inodes(fs_info, 1);
+		return btrfs_start_delalloc_roots(fs_info, 1);
 	return 0;
 }
 
 static inline void btrfs_wait_delalloc_flush(struct btrfs_fs_info *fs_info)
 {
 	if (btrfs_test_opt(fs_info->tree_root, FLUSHONCOMMIT))
-		btrfs_wait_all_ordered_extents(fs_info);
+		btrfs_wait_ordered_roots(fs_info, -1);
 }
 
 int btrfs_commit_transaction(struct btrfs_trans_handle *trans,
@@ -1669,7 +1702,7 @@ int btrfs_commit_transaction(struct btrfs_trans_handle *trans,
 
 		wait_for_commit(root, cur_trans);
 
-		put_transaction(cur_trans);
+		btrfs_put_transaction(cur_trans);
 
 		return ret;
 	}
@@ -1686,7 +1719,7 @@ int btrfs_commit_transaction(struct btrfs_trans_handle *trans,
 
 			wait_for_commit(root, prev_trans);
 
-			put_transaction(prev_trans);
+			btrfs_put_transaction(prev_trans);
 		} else {
 			spin_unlock(&root->fs_info->trans_lock);
 		}
@@ -1885,8 +1918,8 @@ int btrfs_commit_transaction(struct btrfs_trans_handle *trans,
 	list_del_init(&cur_trans->list);
 	spin_unlock(&root->fs_info->trans_lock);
 
-	put_transaction(cur_trans);
-	put_transaction(cur_trans);
+	btrfs_put_transaction(cur_trans);
+	btrfs_put_transaction(cur_trans);
 
 	if (trans->type & __TRANS_FREEZABLE)
 		sb_end_intwrite(root->fs_info->sb);
diff --git a/fs/btrfs/transaction.h b/fs/btrfs/transaction.h
index 5c2af8491621..7657d115067d 100644
--- a/fs/btrfs/transaction.h
+++ b/fs/btrfs/transaction.h
@@ -92,6 +92,7 @@ struct btrfs_trans_handle {
 	short aborted;
 	short adding_csums;
 	bool allocating_chunk;
+	bool reloc_reserved;
 	unsigned int type;
 	/*
 	 * this root is only needed to validate that the root passed to
@@ -166,4 +167,5 @@ int btrfs_wait_marked_extents(struct btrfs_root *root,
 				struct extent_io_tree *dirty_pages, int mark);
 int btrfs_transaction_blocked(struct btrfs_fs_info *info);
 int btrfs_transaction_in_commit(struct btrfs_fs_info *info);
+void btrfs_put_transaction(struct btrfs_transaction *transaction);
 #endif
diff --git a/fs/btrfs/tree-defrag.c b/fs/btrfs/tree-defrag.c
index 94e05c1f118a..76928ca97741 100644
--- a/fs/btrfs/tree-defrag.c
+++ b/fs/btrfs/tree-defrag.c
@@ -37,7 +37,6 @@ int btrfs_defrag_leaves(struct btrfs_trans_handle *trans,
 	int ret = 0;
 	int wret;
 	int level;
-	int is_extent = 0;
 	int next_key_ret = 0;
 	u64 last_ret = 0;
 	u64 min_trans = 0;
@@ -50,7 +49,7 @@ int btrfs_defrag_leaves(struct btrfs_trans_handle *trans,
 		goto out;
 	}
 
-	if (root->ref_cows == 0 && !is_extent)
+	if (root->ref_cows == 0)
 		goto out;
 
 	if (btrfs_test_opt(root, SSD))
@@ -85,7 +84,7 @@ int btrfs_defrag_leaves(struct btrfs_trans_handle *trans,
 
 	path->keep_locks = 1;
 
-	ret = btrfs_search_forward(root, &key, NULL, path, min_trans);
+	ret = btrfs_search_forward(root, &key, path, min_trans);
 	if (ret < 0)
 		goto out;
 	if (ret > 0) {
diff --git a/fs/btrfs/tree-log.c b/fs/btrfs/tree-log.c
index 79f057c0619a..744553c83fe2 100644
--- a/fs/btrfs/tree-log.c
+++ b/fs/btrfs/tree-log.c
@@ -26,7 +26,6 @@
 #include "locking.h"
 #include "print-tree.h"
 #include "backref.h"
-#include "compat.h"
 #include "tree-log.h"
 #include "hash.h"
 
@@ -936,7 +935,7 @@ again:
 					    parent_objectid,
 					    victim_name,
 					    victim_name_len)) {
-				btrfs_inc_nlink(inode);
+				inc_nlink(inode);
 				btrfs_release_path(path);
 
 				ret = btrfs_unlink_inode(trans, root, dir,
@@ -1006,7 +1005,7 @@ again:
 				victim_parent = read_one_inode(root,
 							       parent_objectid);
 				if (victim_parent) {
-					btrfs_inc_nlink(inode);
+					inc_nlink(inode);
 					btrfs_release_path(path);
 
 					ret = btrfs_unlink_inode(trans, root,
@@ -1113,11 +1112,11 @@ static noinline int add_inode_ref(struct btrfs_trans_handle *trans,
 				  struct extent_buffer *eb, int slot,
 				  struct btrfs_key *key)
 {
-	struct inode *dir;
-	struct inode *inode;
+	struct inode *dir = NULL;
+	struct inode *inode = NULL;
 	unsigned long ref_ptr;
 	unsigned long ref_end;
-	char *name;
+	char *name = NULL;
 	int namelen;
 	int ret;
 	int search_done = 0;
@@ -1150,13 +1149,15 @@ static noinline int add_inode_ref(struct btrfs_trans_handle *trans,
 	 * care of the rest
 	 */
 	dir = read_one_inode(root, parent_objectid);
-	if (!dir)
-		return -ENOENT;
+	if (!dir) {
+		ret = -ENOENT;
+		goto out;
+	}
 
 	inode = read_one_inode(root, inode_objectid);
 	if (!inode) {
-		iput(dir);
-		return -EIO;
+		ret = -EIO;
+		goto out;
 	}
 
 	while (ref_ptr < ref_end) {
@@ -1169,14 +1170,16 @@ static noinline int add_inode_ref(struct btrfs_trans_handle *trans,
 			 */
 			if (!dir)
 				dir = read_one_inode(root, parent_objectid);
-			if (!dir)
-				return -ENOENT;
+			if (!dir) {
+				ret = -ENOENT;
+				goto out;
+			}
 		} else {
 			ret = ref_get_fields(eb, ref_ptr, &namelen, &name,
 					     &ref_index);
 		}
 		if (ret)
-			return ret;
+			goto out;
 
 		/* if we already have a perfect match, we're done */
 		if (!inode_in_dir(root, path, btrfs_ino(dir), btrfs_ino(inode),
@@ -1196,12 +1199,11 @@ static noinline int add_inode_ref(struct btrfs_trans_handle *trans,
 						      parent_objectid,
 						      ref_index, name, namelen,
 						      &search_done);
-				if (ret == 1) {
-					ret = 0;
+				if (ret) {
+					if (ret == 1)
+						ret = 0;
 					goto out;
 				}
-				if (ret)
-					goto out;
 			}
 
 			/* insert our name */
@@ -1215,6 +1217,7 @@ static noinline int add_inode_ref(struct btrfs_trans_handle *trans,
 
 		ref_ptr = (unsigned long)(ref_ptr + ref_struct_size) + namelen;
 		kfree(name);
+		name = NULL;
 		if (log_ref_ver) {
 			iput(dir);
 			dir = NULL;
@@ -1225,6 +1228,7 @@ static noinline int add_inode_ref(struct btrfs_trans_handle *trans,
 	ret = overwrite_item(trans, root, path, eb, slot, key);
 out:
 	btrfs_release_path(path);
+	kfree(name);
 	iput(dir);
 	iput(inode);
 	return ret;
@@ -1307,6 +1311,7 @@ static int count_inode_refs(struct btrfs_root *root,
 				break;
 			path->slots[0]--;
 		}
+process_slot:
 		btrfs_item_key_to_cpu(path->nodes[0], &key,
 				      path->slots[0]);
 		if (key.objectid != ino ||
@@ -1327,6 +1332,10 @@ static int count_inode_refs(struct btrfs_root *root,
 
 		if (key.offset == 0)
 			break;
+		if (path->slots[0] > 0) {
+			path->slots[0]--;
+			goto process_slot;
+		}
 		key.offset--;
 		btrfs_release_path(path);
 	}
@@ -1480,7 +1489,7 @@ static noinline int link_to_fixup_dir(struct btrfs_trans_handle *trans,
 		if (!inode->i_nlink)
 			set_nlink(inode, 1);
 		else
-			btrfs_inc_nlink(inode);
+			inc_nlink(inode);
 		ret = btrfs_update_inode(trans, root, inode);
 	} else if (ret == -EEXIST) {
 		ret = 0;
@@ -1823,7 +1832,7 @@ again:
 						     dir_key->offset,
 						     name, name_len, 0);
 		}
-		if (IS_ERR_OR_NULL(log_di)) {
+		if (!log_di || (IS_ERR(log_di) && PTR_ERR(log_di) == -ENOENT)) {
 			btrfs_dir_item_key_to_cpu(eb, di, &location);
 			btrfs_release_path(path);
 			btrfs_release_path(log_path);
@@ -1841,7 +1850,7 @@ again:
 				goto out;
 			}
 
-			btrfs_inc_nlink(inode);
+			inc_nlink(inode);
 			ret = btrfs_unlink_inode(trans, root, dir, inode,
 						 name, name_len);
 			if (!ret)
@@ -1860,6 +1869,9 @@ again:
 				goto again;
 			ret = 0;
 			goto out;
+		} else if (IS_ERR(log_di)) {
+			kfree(name);
+			return PTR_ERR(log_di);
 		}
 		btrfs_release_path(log_path);
 		kfree(name);
@@ -2118,8 +2130,7 @@ static noinline int walk_down_log_tree(struct btrfs_trans_handle *trans,
 		WARN_ON(*level >= BTRFS_MAX_LEVEL);
 		cur = path->nodes[*level];
 
-		if (btrfs_header_level(cur) != *level)
-			WARN_ON(1);
+		WARN_ON(btrfs_header_level(cur) != *level);
 
 		if (path->slots[*level] >=
 		    btrfs_header_nritems(cur))
@@ -2151,11 +2162,13 @@ static noinline int walk_down_log_tree(struct btrfs_trans_handle *trans,
 					return ret;
 				}
 
-				btrfs_tree_lock(next);
-				btrfs_set_lock_blocking(next);
-				clean_tree_block(trans, root, next);
-				btrfs_wait_tree_block_writeback(next);
-				btrfs_tree_unlock(next);
+				if (trans) {
+					btrfs_tree_lock(next);
+					btrfs_set_lock_blocking(next);
+					clean_tree_block(trans, root, next);
+					btrfs_wait_tree_block_writeback(next);
+					btrfs_tree_unlock(next);
+				}
 
 				WARN_ON(root_owner !=
 					BTRFS_TREE_LOG_OBJECTID);
@@ -2227,11 +2240,13 @@ static noinline int walk_up_log_tree(struct btrfs_trans_handle *trans,
 
 				next = path->nodes[*level];
 
-				btrfs_tree_lock(next);
-				btrfs_set_lock_blocking(next);
-				clean_tree_block(trans, root, next);
-				btrfs_wait_tree_block_writeback(next);
-				btrfs_tree_unlock(next);
+				if (trans) {
+					btrfs_tree_lock(next);
+					btrfs_set_lock_blocking(next);
+					clean_tree_block(trans, root, next);
+					btrfs_wait_tree_block_writeback(next);
+					btrfs_tree_unlock(next);
+				}
 
 				WARN_ON(root_owner != BTRFS_TREE_LOG_OBJECTID);
 				ret = btrfs_free_and_pin_reserved_extent(root,
@@ -2301,11 +2316,13 @@ static int walk_log_tree(struct btrfs_trans_handle *trans,
 
 			next = path->nodes[orig_level];
 
-			btrfs_tree_lock(next);
-			btrfs_set_lock_blocking(next);
-			clean_tree_block(trans, log, next);
-			btrfs_wait_tree_block_writeback(next);
-			btrfs_tree_unlock(next);
+			if (trans) {
+				btrfs_tree_lock(next);
+				btrfs_set_lock_blocking(next);
+				clean_tree_block(trans, log, next);
+				btrfs_wait_tree_block_writeback(next);
+				btrfs_tree_unlock(next);
+			}
 
 			WARN_ON(log->root_key.objectid !=
 				BTRFS_TREE_LOG_OBJECTID);
@@ -2571,9 +2588,7 @@ int btrfs_sync_log(struct btrfs_trans_handle *trans,
 	 * the running transaction open, so a full commit can't hop
 	 * in and cause problems either.
 	 */
-	btrfs_scrub_pause_super(root);
 	ret = write_ctree_super(trans, root->fs_info->tree_root, 1);
-	btrfs_scrub_continue_super(root);
 	if (ret) {
 		btrfs_abort_transaction(trans, root, ret);
 		goto out_wake_log_root;
@@ -2608,13 +2623,10 @@ static void free_log_tree(struct btrfs_trans_handle *trans,
 		.process_func = process_one_buffer
 	};
 
-	if (trans) {
-		ret = walk_log_tree(trans, log, &wc);
-
-		/* I don't think this can happen but just in case */
-		if (ret)
-			btrfs_abort_transaction(trans, log, ret);
-	}
+	ret = walk_log_tree(trans, log, &wc);
+	/* I don't think this can happen but just in case */
+	if (ret)
+		btrfs_abort_transaction(trans, log, ret);
 
 	while (1) {
 		ret = find_first_extent_bit(&log->dirty_log_pages,
@@ -2867,7 +2879,6 @@ static noinline int log_dir_items(struct btrfs_trans_handle *trans,
 			  u64 min_offset, u64 *last_offset_ret)
 {
 	struct btrfs_key min_key;
-	struct btrfs_key max_key;
 	struct btrfs_root *log = root->log_root;
 	struct extent_buffer *src;
 	int err = 0;
@@ -2879,9 +2890,6 @@ static noinline int log_dir_items(struct btrfs_trans_handle *trans,
 	u64 ino = btrfs_ino(inode);
 
 	log = root->log_root;
-	max_key.objectid = ino;
-	max_key.offset = (u64)-1;
-	max_key.type = key_type;
 
 	min_key.objectid = ino;
 	min_key.type = key_type;
@@ -2889,8 +2897,7 @@ static noinline int log_dir_items(struct btrfs_trans_handle *trans,
 
 	path->keep_locks = 1;
 
-	ret = btrfs_search_forward(root, &min_key, &max_key,
-				   path, trans->transid);
+	ret = btrfs_search_forward(root, &min_key, path, trans->transid);
 
 	/*
 	 * we didn't find anything from this transaction, see if there
@@ -2943,10 +2950,8 @@ static noinline int log_dir_items(struct btrfs_trans_handle *trans,
 
 	/* find the first key from this transaction again */
 	ret = btrfs_search_slot(NULL, root, &min_key, path, 0, 0);
-	if (ret != 0) {
-		WARN_ON(1);
+	if (WARN_ON(ret != 0))
 		goto done;
-	}
 
 	/*
 	 * we have a block from this transaction, log every item in it
@@ -3172,11 +3177,10 @@ static int log_inode_item(struct btrfs_trans_handle *trans,
 			  struct inode *inode)
 {
 	struct btrfs_inode_item *inode_item;
-	struct btrfs_key key;
 	int ret;
 
-	memcpy(&key, &BTRFS_I(inode)->location, sizeof(key));
-	ret = btrfs_insert_empty_item(trans, log, path, &key,
+	ret = btrfs_insert_empty_item(trans, log, path,
+				      &BTRFS_I(inode)->location,
 				      sizeof(*inode_item));
 	if (ret && ret != -EEXIST)
 		return ret;
@@ -3375,7 +3379,7 @@ static int log_one_extent(struct btrfs_trans_handle *trans,
 		btrfs_set_token_file_extent_type(leaf, fi,
 						 BTRFS_FILE_EXTENT_REG,
 						 &token);
-		if (em->block_start == 0)
+		if (em->block_start == EXTENT_MAP_HOLE)
 			skip_csum = true;
 	}
 
@@ -3417,11 +3421,6 @@ static int log_one_extent(struct btrfs_trans_handle *trans,
 	if (skip_csum)
 		return 0;
 
-	if (em->compress_type) {
-		csum_offset = 0;
-		csum_len = block_len;
-	}
-
 	/*
 	 * First check and see if our csums are on our outstanding ordered
 	 * extents.
@@ -3505,8 +3504,13 @@ unlocked:
 	if (!mod_len || ret)
 		return ret;
 
-	csum_offset = mod_start - em->start;
-	csum_len = mod_len;
+	if (em->compress_type) {
+		csum_offset = 0;
+		csum_len = block_len;
+	} else {
+		csum_offset = mod_start - em->start;
+		csum_len = mod_len;
+	}
 
 	/* block start is already adjusted for the file extent offset. */
 	ret = btrfs_lookup_csums_range(log->fs_info->csum_root,
@@ -3719,7 +3723,7 @@ static int btrfs_log_inode(struct btrfs_trans_handle *trans,
 
 	while (1) {
 		ins_nr = 0;
-		ret = btrfs_search_forward(root, &min_key, &max_key,
+		ret = btrfs_search_forward(root, &min_key,
 					   path, trans->transid);
 		if (ret != 0)
 			break;
@@ -3769,14 +3773,14 @@ next_slot:
 		}
 		btrfs_release_path(path);
 
-		if (min_key.offset < (u64)-1)
+		if (min_key.offset < (u64)-1) {
 			min_key.offset++;
-		else if (min_key.type < (u8)-1)
+		} else if (min_key.type < max_key.type) {
 			min_key.type++;
-		else if (min_key.objectid < (u64)-1)
-			min_key.objectid++;
-		else
+			min_key.offset = 0;
+		} else {
 			break;
+		}
 	}
 	if (ins_nr) {
 		ret = copy_items(trans, inode, dst_path, src, ins_start_slot,
diff --git a/fs/btrfs/uuid-tree.c b/fs/btrfs/uuid-tree.c
index dd0dea3766f7..fbda90004fe9 100644
--- a/fs/btrfs/uuid-tree.c
+++ b/fs/btrfs/uuid-tree.c
@@ -260,7 +260,6 @@ int btrfs_uuid_tree_iterate(struct btrfs_fs_info *fs_info,
 {
 	struct btrfs_root *root = fs_info->uuid_root;
 	struct btrfs_key key;
-	struct btrfs_key max_key;
 	struct btrfs_path *path;
 	int ret = 0;
 	struct extent_buffer *leaf;
@@ -277,13 +276,10 @@ int btrfs_uuid_tree_iterate(struct btrfs_fs_info *fs_info,
 	key.objectid = 0;
 	key.type = 0;
 	key.offset = 0;
-	max_key.objectid = (u64)-1;
-	max_key.type = (u8)-1;
-	max_key.offset = (u64)-1;
 
 again_search_slot:
 	path->keep_locks = 1;
-	ret = btrfs_search_forward(root, &key, &max_key, path, 0);
+	ret = btrfs_search_forward(root, &key, path, 0);
 	if (ret) {
 		if (ret > 0)
 			ret = 0;
diff --git a/fs/btrfs/volumes.c b/fs/btrfs/volumes.c
index 043b215769c2..0db637097862 100644
--- a/fs/btrfs/volumes.c
+++ b/fs/btrfs/volumes.c
@@ -28,7 +28,6 @@
 #include <linux/raid/pq.h>
 #include <linux/semaphore.h>
 #include <asm/div64.h>
-#include "compat.h"
 #include "ctree.h"
 #include "extent_map.h"
 #include "disk-io.h"
@@ -666,7 +665,8 @@ static int __btrfs_close_devices(struct btrfs_fs_devices *fs_devices)
 		if (device->bdev)
 			fs_devices->open_devices--;
 
-		if (device->writeable && !device->is_tgtdev_for_dev_replace) {
+		if (device->writeable &&
+		    device->devid != BTRFS_DEV_REPLACE_DEVID) {
 			list_del_init(&device->dev_alloc_list);
 			fs_devices->rw_devices--;
 		}
@@ -2041,6 +2041,7 @@ int btrfs_init_new_device(struct btrfs_root *root, char *device_path)
 	device->in_fs_metadata = 1;
 	device->is_tgtdev_for_dev_replace = 0;
 	device->mode = FMODE_EXCL;
+	device->dev_stats_valid = 1;
 	set_blocksize(device->bdev, 4096);
 
 	if (seeding_dev) {
@@ -2208,6 +2209,7 @@ int btrfs_init_dev_replace_tgtdev(struct btrfs_root *root, char *device_path,
 	device->in_fs_metadata = 1;
 	device->is_tgtdev_for_dev_replace = 1;
 	device->mode = FMODE_EXCL;
+	device->dev_stats_valid = 1;
 	set_blocksize(device->bdev, 4096);
 	device->fs_devices = fs_info->fs_devices;
 	list_add(&device->dev_list, &fs_info->fs_devices->devices);
@@ -2550,8 +2552,7 @@ again:
 		failed = 0;
 		retried = true;
 		goto again;
-	} else if (failed && retried) {
-		WARN_ON(1);
+	} else if (WARN_ON(failed && retried)) {
 		ret = -ENOSPC;
 	}
 error:
@@ -3423,6 +3424,9 @@ int btrfs_pause_balance(struct btrfs_fs_info *fs_info)
 
 int btrfs_cancel_balance(struct btrfs_fs_info *fs_info)
 {
+	if (fs_info->sb->s_flags & MS_RDONLY)
+		return -EROFS;
+
 	mutex_lock(&fs_info->balance_mutex);
 	if (!fs_info->balance_ctl) {
 		mutex_unlock(&fs_info->balance_mutex);
@@ -3488,7 +3492,7 @@ static int btrfs_uuid_scan_kthread(void *data)
 	path->keep_locks = 1;
 
 	while (1) {
-		ret = btrfs_search_forward(root, &key, &max_key, path, 0);
+		ret = btrfs_search_forward(root, &key, path, 0);
 		if (ret) {
 			if (ret > 0)
 				ret = 0;
@@ -4488,6 +4492,7 @@ int btrfs_num_copies(struct btrfs_fs_info *fs_info, u64 logical, u64 len)
 		btrfs_crit(fs_info, "Invalid mapping for %Lu-%Lu, got "
 			    "%Lu-%Lu\n", logical, logical+len, em->start,
 			    em->start + em->len);
+		free_extent_map(em);
 		return 1;
 	}
 
@@ -4668,6 +4673,7 @@ static int __btrfs_map_block(struct btrfs_fs_info *fs_info, int rw,
 		btrfs_crit(fs_info, "found a bad mapping, wanted %Lu, "
 			   "found %Lu-%Lu\n", logical, em->start,
 			   em->start + em->len);
+		free_extent_map(em);
 		return -EINVAL;
 	}
 
@@ -4895,7 +4901,7 @@ static int __btrfs_map_block(struct btrfs_fs_info *fs_info, int rw,
 			num_stripes = map->num_stripes;
 			max_errors = nr_parity_stripes(map);
 
-			raid_map = kmalloc(sizeof(u64) * num_stripes,
+			raid_map = kmalloc_array(num_stripes, sizeof(u64),
 					   GFP_NOFS);
 			if (!raid_map) {
 				ret = -ENOMEM;
@@ -5395,10 +5401,8 @@ static int bio_size_ok(struct block_device *bdev, struct bio *bio,
 		.bi_rw = bio->bi_rw,
 	};
 
-	if (bio->bi_vcnt == 0) {
-		WARN_ON(1);
+	if (WARN_ON(bio->bi_vcnt == 0))
 		return 1;
-	}
 
 	prev = &bio->bi_io_vec[bio->bi_vcnt - 1];
 	if (bio_sectors(bio) > max_sectors)
@@ -5631,10 +5635,8 @@ struct btrfs_device *btrfs_alloc_device(struct btrfs_fs_info *fs_info,
 	struct btrfs_device *dev;
 	u64 tmp;
 
-	if (!devid && !fs_info) {
-		WARN_ON(1);
+	if (WARN_ON(!devid && !fs_info))
 		return ERR_PTR(-EINVAL);
-	}
 
 	dev = __alloc_device();
 	if (IS_ERR(dev))
diff --git a/fs/btrfs/volumes.h b/fs/btrfs/volumes.h
index b72f540c8b29..8b3cd142b373 100644
--- a/fs/btrfs/volumes.h
+++ b/fs/btrfs/volumes.h
@@ -43,9 +43,8 @@ struct btrfs_device {
 	/* WRITE_SYNC bios */
 	struct btrfs_pending_bios pending_sync_bios;
 
-	int running_pending;
 	u64 generation;
-
+	int running_pending;
 	int writeable;
 	int in_fs_metadata;
 	int missing;
@@ -53,11 +52,11 @@ struct btrfs_device {
 	int is_tgtdev_for_dev_replace;
 
 	spinlock_t io_lock;
+	/* the mode sent to blkdev_get */
+	fmode_t mode;
 
 	struct block_device *bdev;
 
-	/* the mode sent to blkdev_get */
-	fmode_t mode;
 
 	struct rcu_string *name;
 
@@ -78,16 +77,21 @@ struct btrfs_device {
 
 	/* optimal io width for this device */
 	u32 io_width;
+	/* type and info about this device */
+	u64 type;
 
 	/* minimal io size for this device */
 	u32 sector_size;
 
-	/* type and info about this device */
-	u64 type;
 
 	/* physical drive uuid (or lvm uuid) */
 	u8 uuid[BTRFS_UUID_SIZE];
 
+	/* for sending down flush barriers */
+	int nobarriers;
+	struct bio *flush_bio;
+	struct completion flush_wait;
+
 	/* per-device scrub information */
 	struct scrub_ctx *scrub_device;
 
@@ -103,10 +107,6 @@ struct btrfs_device {
 	struct radix_tree_root reada_zones;
 	struct radix_tree_root reada_extents;
 
-	/* for sending down flush barriers */
-	struct bio *flush_bio;
-	struct completion flush_wait;
-	int nobarriers;
 
 	/* disk I/O failure stats. For detailed description refer to
 	 * enum btrfs_dev_stat_values in ioctl.h */
@@ -132,7 +132,9 @@ struct btrfs_fs_devices {
 
 	/* all of the devices in the FS, protected by a mutex
 	 * so we can safely walk it to write out the supers without
-	 * worrying about add/remove by the multi-device code
+	 * worrying about add/remove by the multi-device code.
+	 * Scrubbing super can kick off supers writing by holding
+	 * this mutex lock.
 	 */
 	struct mutex device_list_mutex;
 	struct list_head devices;
diff --git a/fs/cifs/cifsencrypt.c b/fs/cifs/cifsencrypt.c
index fc6f4f3a1a9d..4934347321d3 100644
--- a/fs/cifs/cifsencrypt.c
+++ b/fs/cifs/cifsencrypt.c
@@ -548,7 +548,13 @@ static int
 CalcNTLMv2_response(const struct cifs_ses *ses, char *ntlmv2_hash)
 {
 	int rc;
-	unsigned int offset = CIFS_SESS_KEY_SIZE + 8;
+	struct ntlmv2_resp *ntlmv2 = (struct ntlmv2_resp *)
+	    (ses->auth_key.response + CIFS_SESS_KEY_SIZE);
+	unsigned int hash_len;
+
+	/* The MD5 hash starts at challenge_key.key */
+	hash_len = ses->auth_key.len - (CIFS_SESS_KEY_SIZE +
+		offsetof(struct ntlmv2_resp, challenge.key[0]));
 
 	if (!ses->server->secmech.sdeschmacmd5) {
 		cifs_dbg(VFS, "%s: can't generate ntlmv2 hash\n", __func__);
@@ -556,7 +562,7 @@ CalcNTLMv2_response(const struct cifs_ses *ses, char *ntlmv2_hash)
 	}
 
 	rc = crypto_shash_setkey(ses->server->secmech.hmacmd5,
-				ntlmv2_hash, CIFS_HMAC_MD5_HASH_SIZE);
+				 ntlmv2_hash, CIFS_HMAC_MD5_HASH_SIZE);
 	if (rc) {
 		cifs_dbg(VFS, "%s: Could not set NTLMV2 Hash as a key\n",
 			 __func__);
@@ -570,20 +576,21 @@ CalcNTLMv2_response(const struct cifs_ses *ses, char *ntlmv2_hash)
 	}
 
 	if (ses->server->negflavor == CIFS_NEGFLAVOR_EXTENDED)
-		memcpy(ses->auth_key.response + offset,
-			ses->ntlmssp->cryptkey, CIFS_SERVER_CHALLENGE_SIZE);
+		memcpy(ntlmv2->challenge.key,
+		       ses->ntlmssp->cryptkey, CIFS_SERVER_CHALLENGE_SIZE);
 	else
-		memcpy(ses->auth_key.response + offset,
-			ses->server->cryptkey, CIFS_SERVER_CHALLENGE_SIZE);
+		memcpy(ntlmv2->challenge.key,
+		       ses->server->cryptkey, CIFS_SERVER_CHALLENGE_SIZE);
 	rc = crypto_shash_update(&ses->server->secmech.sdeschmacmd5->shash,
-		ses->auth_key.response + offset, ses->auth_key.len - offset);
+				 ntlmv2->challenge.key, hash_len);
 	if (rc) {
 		cifs_dbg(VFS, "%s: Could not update with response\n", __func__);
 		return rc;
 	}
 
+	/* Note that the MD5 digest over writes anon.challenge_key.key */
 	rc = crypto_shash_final(&ses->server->secmech.sdeschmacmd5->shash,
-		ses->auth_key.response + CIFS_SESS_KEY_SIZE);
+				ntlmv2->ntlmv2_hash);
 	if (rc)
 		cifs_dbg(VFS, "%s: Could not generate md5 hash\n", __func__);
 
@@ -627,7 +634,7 @@ setup_ntlmv2_rsp(struct cifs_ses *ses, const struct nls_table *nls_cp)
 	int rc;
 	int baselen;
 	unsigned int tilen;
-	struct ntlmv2_resp *buf;
+	struct ntlmv2_resp *ntlmv2;
 	char ntlmv2_hash[16];
 	unsigned char *tiblob = NULL; /* target info blob */
 
@@ -660,13 +667,14 @@ setup_ntlmv2_rsp(struct cifs_ses *ses, const struct nls_table *nls_cp)
 	}
 	ses->auth_key.len += baselen;
 
-	buf = (struct ntlmv2_resp *)
+	ntlmv2 = (struct ntlmv2_resp *)
 			(ses->auth_key.response + CIFS_SESS_KEY_SIZE);
-	buf->blob_signature = cpu_to_le32(0x00000101);
-	buf->reserved = 0;
-	buf->time = cpu_to_le64(cifs_UnixTimeToNT(CURRENT_TIME));
-	get_random_bytes(&buf->client_chal, sizeof(buf->client_chal));
-	buf->reserved2 = 0;
+	ntlmv2->blob_signature = cpu_to_le32(0x00000101);
+	ntlmv2->reserved = 0;
+	/* Must be within 5 minutes of the server */
+	ntlmv2->time = cpu_to_le64(cifs_UnixTimeToNT(CURRENT_TIME));
+	get_random_bytes(&ntlmv2->client_chal, sizeof(ntlmv2->client_chal));
+	ntlmv2->reserved2 = 0;
 
 	memcpy(ses->auth_key.response + baselen, tiblob, tilen);
 
@@ -706,7 +714,7 @@ setup_ntlmv2_rsp(struct cifs_ses *ses, const struct nls_table *nls_cp)
 	}
 
 	rc = crypto_shash_update(&ses->server->secmech.sdeschmacmd5->shash,
-		ses->auth_key.response + CIFS_SESS_KEY_SIZE,
+		ntlmv2->ntlmv2_hash,
 		CIFS_HMAC_MD5_HASH_SIZE);
 	if (rc) {
 		cifs_dbg(VFS, "%s: Could not update with response\n", __func__);
diff --git a/fs/cifs/cifsglob.h b/fs/cifs/cifsglob.h
index 26b1c1dc93f6..d9ea7ada1378 100644
--- a/fs/cifs/cifsglob.h
+++ b/fs/cifs/cifsglob.h
@@ -261,7 +261,7 @@ struct smb_version_operations {
 	/* query path data from the server */
 	int (*query_path_info)(const unsigned int, struct cifs_tcon *,
 			       struct cifs_sb_info *, const char *,
-			       FILE_ALL_INFO *, bool *);
+			       FILE_ALL_INFO *, bool *, bool *);
 	/* query file data from the server */
 	int (*query_file_info)(const unsigned int, struct cifs_tcon *,
 			       struct cifs_fid *, FILE_ALL_INFO *);
@@ -381,6 +381,9 @@ struct smb_version_operations {
 	char * (*create_lease_buf)(u8 *, u8);
 	/* parse lease context buffer and return oplock/epoch info */
 	__u8 (*parse_lease_buf)(void *, unsigned int *);
+	int (*clone_range)(const unsigned int, struct cifsFileInfo *src_file,
+			struct cifsFileInfo *target_file, u64 src_off, u64 len,
+			u64 dest_off);
 };
 
 struct smb_version_values {
@@ -855,6 +858,9 @@ struct cifs_tcon {
 	__le64 vol_create_time;
 	__u32 ss_flags;		/* sector size flags */
 	__u32 perf_sector_size; /* best sector size for perf */
+	__u32 max_chunks;
+	__u32 max_bytes_chunk;
+	__u32 max_bytes_copy;
 #endif /* CONFIG_CIFS_SMB2 */
 #ifdef CONFIG_CIFS_FSCACHE
 	u64 resource_id;		/* server resource id */
diff --git a/fs/cifs/cifspdu.h b/fs/cifs/cifspdu.h
index 9e5ee34de986..33df36ef9d52 100644
--- a/fs/cifs/cifspdu.h
+++ b/fs/cifs/cifspdu.h
@@ -697,7 +697,13 @@ struct ntlmssp2_name {
 } __attribute__((packed));
 
 struct ntlmv2_resp {
-	char ntlmv2_hash[CIFS_ENCPWD_SIZE];
+	union {
+	    char ntlmv2_hash[CIFS_ENCPWD_SIZE];
+	    struct {
+		__u8 reserved[8];
+		__u8 key[CIFS_SERVER_CHALLENGE_SIZE];
+	    } __attribute__((packed)) challenge;
+	} __attribute__((packed));
 	__le32 blob_signature;
 	__u32  reserved;
 	__le64  time;
diff --git a/fs/cifs/cifssmb.c b/fs/cifs/cifssmb.c
index 93b29474714a..124aa0230c1b 100644
--- a/fs/cifs/cifssmb.c
+++ b/fs/cifs/cifssmb.c
@@ -3369,11 +3369,13 @@ static __u16 ACL_to_cifs_posix(char *parm_data, const char *pACL,
 		return 0;
 	}
 	cifs_acl->version = cpu_to_le16(1);
-	if (acl_type == ACL_TYPE_ACCESS)
+	if (acl_type == ACL_TYPE_ACCESS) {
 		cifs_acl->access_entry_count = cpu_to_le16(count);
-	else if (acl_type == ACL_TYPE_DEFAULT)
+		cifs_acl->default_entry_count = __constant_cpu_to_le16(0xFFFF);
+	} else if (acl_type == ACL_TYPE_DEFAULT) {
 		cifs_acl->default_entry_count = cpu_to_le16(count);
-	else {
+		cifs_acl->access_entry_count = __constant_cpu_to_le16(0xFFFF);
+	} else {
 		cifs_dbg(FYI, "unknown ACL type %d\n", acl_type);
 		return 0;
 	}
diff --git a/fs/cifs/dir.c b/fs/cifs/dir.c
index 5384c2a640ca..11ff5f116b20 100644
--- a/fs/cifs/dir.c
+++ b/fs/cifs/dir.c
@@ -756,7 +756,7 @@ cifs_lookup(struct inode *parent_dir_inode, struct dentry *direntry,
 	/*	if it was once a directory (but how can we tell?) we could do
 		shrink_dcache_parent(direntry); */
 	} else if (rc != -EACCES) {
-		cifs_dbg(VFS, "Unexpected lookup error %d\n", rc);
+		cifs_dbg(FYI, "Unexpected lookup error %d\n", rc);
 		/* We special case check for Access Denied - since that
 		is a common return code */
 	}
diff --git a/fs/cifs/file.c b/fs/cifs/file.c
index 7ddddf2e2504..5a5a87240fe2 100644
--- a/fs/cifs/file.c
+++ b/fs/cifs/file.c
@@ -3663,6 +3663,27 @@ void cifs_oplock_break(struct work_struct *work)
 	}
 }
 
+/*
+ * The presence of cifs_direct_io() in the address space ops vector
+ * allowes open() O_DIRECT flags which would have failed otherwise.
+ *
+ * In the non-cached mode (mount with cache=none), we shunt off direct read and write requests
+ * so this method should never be called.
+ *
+ * Direct IO is not yet supported in the cached mode. 
+ */
+static ssize_t
+cifs_direct_io(int rw, struct kiocb *iocb, const struct iovec *iov,
+               loff_t pos, unsigned long nr_segs)
+{
+        /*
+         * FIXME
+         * Eventually need to support direct IO for non forcedirectio mounts
+         */
+        return -EINVAL;
+}
+
+
 const struct address_space_operations cifs_addr_ops = {
 	.readpage = cifs_readpage,
 	.readpages = cifs_readpages,
@@ -3672,6 +3693,7 @@ const struct address_space_operations cifs_addr_ops = {
 	.write_end = cifs_write_end,
 	.set_page_dirty = __set_page_dirty_nobuffers,
 	.releasepage = cifs_release_page,
+	.direct_IO = cifs_direct_io,
 	.invalidatepage = cifs_invalidate_page,
 	.launder_page = cifs_launder_page,
 };
diff --git a/fs/cifs/inode.c b/fs/cifs/inode.c
index 867b7cdc794a..36f9ebb93ceb 100644
--- a/fs/cifs/inode.c
+++ b/fs/cifs/inode.c
@@ -542,7 +542,8 @@ static int cifs_sfu_mode(struct cifs_fattr *fattr, const unsigned char *path,
 /* Fill a cifs_fattr struct with info from FILE_ALL_INFO */
 static void
 cifs_all_info_to_fattr(struct cifs_fattr *fattr, FILE_ALL_INFO *info,
-		       struct cifs_sb_info *cifs_sb, bool adjust_tz)
+		       struct cifs_sb_info *cifs_sb, bool adjust_tz,
+		       bool symlink)
 {
 	struct cifs_tcon *tcon = cifs_sb_master_tcon(cifs_sb);
 
@@ -569,7 +570,11 @@ cifs_all_info_to_fattr(struct cifs_fattr *fattr, FILE_ALL_INFO *info,
 	fattr->cf_createtime = le64_to_cpu(info->CreationTime);
 
 	fattr->cf_nlink = le32_to_cpu(info->NumberOfLinks);
-	if (fattr->cf_cifsattrs & ATTR_DIRECTORY) {
+
+	if (symlink) {
+		fattr->cf_mode = S_IFLNK;
+		fattr->cf_dtype = DT_LNK;
+	} else if (fattr->cf_cifsattrs & ATTR_DIRECTORY) {
 		fattr->cf_mode = S_IFDIR | cifs_sb->mnt_dir_mode;
 		fattr->cf_dtype = DT_DIR;
 		/*
@@ -578,10 +583,6 @@ cifs_all_info_to_fattr(struct cifs_fattr *fattr, FILE_ALL_INFO *info,
 		 */
 		if (!tcon->unix_ext)
 			fattr->cf_flags |= CIFS_FATTR_UNKNOWN_NLINK;
-	} else if (fattr->cf_cifsattrs & ATTR_REPARSE) {
-		fattr->cf_mode = S_IFLNK;
-		fattr->cf_dtype = DT_LNK;
-		fattr->cf_nlink = le32_to_cpu(info->NumberOfLinks);
 	} else {
 		fattr->cf_mode = S_IFREG | cifs_sb->mnt_file_mode;
 		fattr->cf_dtype = DT_REG;
@@ -626,7 +627,8 @@ cifs_get_file_info(struct file *filp)
 	rc = server->ops->query_file_info(xid, tcon, &cfile->fid, &find_data);
 	switch (rc) {
 	case 0:
-		cifs_all_info_to_fattr(&fattr, &find_data, cifs_sb, false);
+		cifs_all_info_to_fattr(&fattr, &find_data, cifs_sb, false,
+				       false);
 		break;
 	case -EREMOTE:
 		cifs_create_dfs_fattr(&fattr, inode->i_sb);
@@ -673,6 +675,7 @@ cifs_get_inode_info(struct inode **inode, const char *full_path,
 	bool adjust_tz = false;
 	struct cifs_fattr fattr;
 	struct cifs_search_info *srchinf = NULL;
+	bool symlink = false;
 
 	tlink = cifs_sb_tlink(cifs_sb);
 	if (IS_ERR(tlink))
@@ -702,12 +705,12 @@ cifs_get_inode_info(struct inode **inode, const char *full_path,
 		}
 		data = (FILE_ALL_INFO *)buf;
 		rc = server->ops->query_path_info(xid, tcon, cifs_sb, full_path,
-						  data, &adjust_tz);
+						  data, &adjust_tz, &symlink);
 	}
 
 	if (!rc) {
-		cifs_all_info_to_fattr(&fattr, (FILE_ALL_INFO *)data, cifs_sb,
-				       adjust_tz);
+		cifs_all_info_to_fattr(&fattr, data, cifs_sb, adjust_tz,
+				       symlink);
 	} else if (rc == -EREMOTE) {
 		cifs_create_dfs_fattr(&fattr, sb);
 		rc = 0;
diff --git a/fs/cifs/ioctl.c b/fs/cifs/ioctl.c
index ba54bf6ab116..409b45eefe70 100644
--- a/fs/cifs/ioctl.c
+++ b/fs/cifs/ioctl.c
@@ -22,12 +22,120 @@
  */
 
 #include <linux/fs.h>
+#include <linux/file.h>
+#include <linux/mount.h>
+#include <linux/mm.h>
+#include <linux/pagemap.h>
+#include <linux/btrfs.h>
 #include "cifspdu.h"
 #include "cifsglob.h"
 #include "cifsproto.h"
 #include "cifs_debug.h"
 #include "cifsfs.h"
 
+static long cifs_ioctl_clone(unsigned int xid, struct file *dst_file,
+			unsigned long srcfd, u64 off, u64 len, u64 destoff)
+{
+	int rc;
+	struct cifsFileInfo *smb_file_target = dst_file->private_data;
+	struct inode *target_inode = file_inode(dst_file);
+	struct cifs_tcon *target_tcon;
+	struct fd src_file;
+	struct cifsFileInfo *smb_file_src;
+	struct inode *src_inode;
+	struct cifs_tcon *src_tcon;
+
+	cifs_dbg(FYI, "ioctl clone range\n");
+	/* the destination must be opened for writing */
+	if (!(dst_file->f_mode & FMODE_WRITE)) {
+		cifs_dbg(FYI, "file target not open for write\n");
+		return -EINVAL;
+	}
+
+	/* check if target volume is readonly and take reference */
+	rc = mnt_want_write_file(dst_file);
+	if (rc) {
+		cifs_dbg(FYI, "mnt_want_write failed with rc %d\n", rc);
+		return rc;
+	}
+
+	src_file = fdget(srcfd);
+	if (!src_file.file) {
+		rc = -EBADF;
+		goto out_drop_write;
+	}
+
+	if ((!src_file.file->private_data) || (!dst_file->private_data)) {
+		rc = -EBADF;
+		cifs_dbg(VFS, "missing cifsFileInfo on copy range src file\n");
+		goto out_fput;
+	}
+
+	rc = -EXDEV;
+	smb_file_target = dst_file->private_data;
+	smb_file_src = src_file.file->private_data;
+	src_tcon = tlink_tcon(smb_file_src->tlink);
+	target_tcon = tlink_tcon(smb_file_target->tlink);
+
+	/* check if source and target are on same tree connection */
+	if (src_tcon != target_tcon) {
+		cifs_dbg(VFS, "file copy src and target on different volume\n");
+		goto out_fput;
+	}
+
+	src_inode = src_file.file->f_dentry->d_inode;
+
+	/*
+	 * Note: cifs case is easier than btrfs since server responsible for
+	 * checks for proper open modes and file type and if it wants
+	 * server could even support copy of range where source = target
+	 */
+
+	/* so we do not deadlock racing two ioctls on same files */
+	if (target_inode < src_inode) {
+		mutex_lock_nested(&target_inode->i_mutex, I_MUTEX_PARENT);
+		mutex_lock_nested(&src_inode->i_mutex, I_MUTEX_CHILD);
+	} else {
+		mutex_lock_nested(&src_inode->i_mutex, I_MUTEX_PARENT);
+		mutex_lock_nested(&target_inode->i_mutex, I_MUTEX_CHILD);
+	}
+
+	/* determine range to clone */
+	rc = -EINVAL;
+	if (off + len > src_inode->i_size || off + len < off)
+		goto out_unlock;
+	if (len == 0)
+		len = src_inode->i_size - off;
+
+	cifs_dbg(FYI, "about to flush pages\n");
+	/* should we flush first and last page first */
+	truncate_inode_pages_range(&target_inode->i_data, destoff,
+				   PAGE_CACHE_ALIGN(destoff + len)-1);
+
+	if (target_tcon->ses->server->ops->clone_range)
+		rc = target_tcon->ses->server->ops->clone_range(xid,
+			smb_file_src, smb_file_target, off, len, destoff);
+
+	/* force revalidate of size and timestamps of target file now
+	   that target is updated on the server */
+	CIFS_I(target_inode)->time = 0;
+out_unlock:
+	/* although unlocking in the reverse order from locking is not
+	   strictly necessary here it is a little cleaner to be consistent */
+	if (target_inode < src_inode) {
+		mutex_unlock(&src_inode->i_mutex);
+		mutex_unlock(&target_inode->i_mutex);
+	} else {
+		mutex_unlock(&target_inode->i_mutex);
+		mutex_unlock(&src_inode->i_mutex);
+	}
+out_fput:
+	fdput(src_file);
+out_drop_write:
+	mnt_drop_write_file(dst_file);
+	return rc;
+}
+
 long cifs_ioctl(struct file *filep, unsigned int command, unsigned long arg)
 {
 	struct inode *inode = file_inode(filep);
@@ -105,6 +213,9 @@ long cifs_ioctl(struct file *filep, unsigned int command, unsigned long arg)
 				cifs_dbg(FYI, "set compress flag rc %d\n", rc);
 			}
 			break;
+		case BTRFS_IOC_CLONE:
+			rc = cifs_ioctl_clone(xid, filep, arg, 0, 0, 0);
+			break;
 		default:
 			cifs_dbg(FYI, "unsupported ioctl\n");
 			break;
diff --git a/fs/cifs/netmisc.c b/fs/cifs/netmisc.c
index 651a5279607b..049884552e76 100644
--- a/fs/cifs/netmisc.c
+++ b/fs/cifs/netmisc.c
@@ -51,7 +51,7 @@ static const struct smb_to_posix_error mapping_table_ERRDOS[] = {
 	{ERRnoaccess, -EACCES},
 	{ERRbadfid, -EBADF},
 	{ERRbadmcb, -EIO},
-	{ERRnomem, -ENOMEM},
+	{ERRnomem, -EREMOTEIO},
 	{ERRbadmem, -EFAULT},
 	{ERRbadenv, -EFAULT},
 	{ERRbadformat, -EINVAL},
diff --git a/fs/cifs/readdir.c b/fs/cifs/readdir.c
index 53a75f3d0179..5940ecabbe6a 100644
--- a/fs/cifs/readdir.c
+++ b/fs/cifs/readdir.c
@@ -134,22 +134,6 @@ out:
 	dput(dentry);
 }
 
-/*
- * Is it possible that this directory might turn out to be a DFS referral
- * once we go to try and use it?
- */
-static bool
-cifs_dfs_is_possible(struct cifs_sb_info *cifs_sb)
-{
-#ifdef CONFIG_CIFS_DFS_UPCALL
-	struct cifs_tcon *tcon = cifs_sb_master_tcon(cifs_sb);
-
-	if (tcon->Flags & SMB_SHARE_IS_IN_DFS)
-		return true;
-#endif
-	return false;
-}
-
 static void
 cifs_fill_common_info(struct cifs_fattr *fattr, struct cifs_sb_info *cifs_sb)
 {
@@ -159,27 +143,19 @@ cifs_fill_common_info(struct cifs_fattr *fattr, struct cifs_sb_info *cifs_sb)
 	if (fattr->cf_cifsattrs & ATTR_DIRECTORY) {
 		fattr->cf_mode = S_IFDIR | cifs_sb->mnt_dir_mode;
 		fattr->cf_dtype = DT_DIR;
-		/*
-		 * Windows CIFS servers generally make DFS referrals look
-		 * like directories in FIND_* responses with the reparse
-		 * attribute flag also set (since DFS junctions are
-		 * reparse points). We must revalidate at least these
-		 * directory inodes before trying to use them (if
-		 * they are DFS we will get PATH_NOT_COVERED back
-		 * when queried directly and can then try to connect
-		 * to the DFS target)
-		 */
-		if (cifs_dfs_is_possible(cifs_sb) &&
-		    (fattr->cf_cifsattrs & ATTR_REPARSE))
-			fattr->cf_flags |= CIFS_FATTR_NEED_REVAL;
-	} else if (fattr->cf_cifsattrs & ATTR_REPARSE) {
-		fattr->cf_mode = S_IFLNK;
-		fattr->cf_dtype = DT_LNK;
 	} else {
 		fattr->cf_mode = S_IFREG | cifs_sb->mnt_file_mode;
 		fattr->cf_dtype = DT_REG;
 	}
 
+	/*
+	 * We need to revalidate it further to make a decision about whether it
+	 * is a symbolic link, DFS referral or a reparse point with a direct
+	 * access like junctions, deduplicated files, NFS symlinks.
+	 */
+	if (fattr->cf_cifsattrs & ATTR_REPARSE)
+		fattr->cf_flags |= CIFS_FATTR_NEED_REVAL;
+
 	/* non-unix readdir doesn't provide nlink */
 	fattr->cf_flags |= CIFS_FATTR_UNKNOWN_NLINK;
 
diff --git a/fs/cifs/smb1ops.c b/fs/cifs/smb1ops.c
index 384cffe42850..5f5ba0dc2ee1 100644
--- a/fs/cifs/smb1ops.c
+++ b/fs/cifs/smb1ops.c
@@ -534,10 +534,12 @@ cifs_is_path_accessible(const unsigned int xid, struct cifs_tcon *tcon,
 static int
 cifs_query_path_info(const unsigned int xid, struct cifs_tcon *tcon,
 		     struct cifs_sb_info *cifs_sb, const char *full_path,
-		     FILE_ALL_INFO *data, bool *adjustTZ)
+		     FILE_ALL_INFO *data, bool *adjustTZ, bool *symlink)
 {
 	int rc;
 
+	*symlink = false;
+
 	/* could do find first instead but this returns more info */
 	rc = CIFSSMBQPathInfo(xid, tcon, full_path, data, 0 /* not legacy */,
 			      cifs_sb->local_nls, cifs_sb->mnt_cifs_flags &
@@ -554,6 +556,23 @@ cifs_query_path_info(const unsigned int xid, struct cifs_tcon *tcon,
 						CIFS_MOUNT_MAP_SPECIAL_CHR);
 		*adjustTZ = true;
 	}
+
+	if (!rc && (le32_to_cpu(data->Attributes) & ATTR_REPARSE)) {
+		int tmprc;
+		int oplock = 0;
+		__u16 netfid;
+
+		/* Need to check if this is a symbolic link or not */
+		tmprc = CIFSSMBOpen(xid, tcon, full_path, FILE_OPEN,
+				    FILE_READ_ATTRIBUTES, 0, &netfid, &oplock,
+				    NULL, cifs_sb->local_nls,
+			cifs_sb->mnt_cifs_flags & CIFS_MOUNT_MAP_SPECIAL_CHR);
+		if (tmprc == -EOPNOTSUPP)
+			*symlink = true;
+		else
+			CIFSSMBClose(xid, tcon, netfid);
+	}
+
 	return rc;
 }
 
diff --git a/fs/cifs/smb2inode.c b/fs/cifs/smb2inode.c
index 78ff88c467b9..84c012a6aba0 100644
--- a/fs/cifs/smb2inode.c
+++ b/fs/cifs/smb2inode.c
@@ -123,12 +123,13 @@ move_smb2_info_to_cifs(FILE_ALL_INFO *dst, struct smb2_file_all_info *src)
 int
 smb2_query_path_info(const unsigned int xid, struct cifs_tcon *tcon,
 		     struct cifs_sb_info *cifs_sb, const char *full_path,
-		     FILE_ALL_INFO *data, bool *adjust_tz)
+		     FILE_ALL_INFO *data, bool *adjust_tz, bool *symlink)
 {
 	int rc;
 	struct smb2_file_all_info *smb2_data;
 
 	*adjust_tz = false;
+	*symlink = false;
 
 	smb2_data = kzalloc(sizeof(struct smb2_file_all_info) + MAX_NAME * 2,
 			    GFP_KERNEL);
@@ -136,9 +137,16 @@ smb2_query_path_info(const unsigned int xid, struct cifs_tcon *tcon,
 		return -ENOMEM;
 
 	rc = smb2_open_op_close(xid, tcon, cifs_sb, full_path,
-				FILE_READ_ATTRIBUTES, FILE_OPEN,
-				OPEN_REPARSE_POINT, smb2_data,
-				SMB2_OP_QUERY_INFO);
+				FILE_READ_ATTRIBUTES, FILE_OPEN, 0,
+				smb2_data, SMB2_OP_QUERY_INFO);
+	if (rc == -EOPNOTSUPP) {
+		*symlink = true;
+		/* Failed on a symbolic link - query a reparse point info */
+		rc = smb2_open_op_close(xid, tcon, cifs_sb, full_path,
+					FILE_READ_ATTRIBUTES, FILE_OPEN,
+					OPEN_REPARSE_POINT, smb2_data,
+					SMB2_OP_QUERY_INFO);
+	}
 	if (rc)
 		goto out;
 
diff --git a/fs/cifs/smb2maperror.c b/fs/cifs/smb2maperror.c
index 7c2f45c06fc2..94bd4fbb13d3 100644
--- a/fs/cifs/smb2maperror.c
+++ b/fs/cifs/smb2maperror.c
@@ -306,7 +306,7 @@ static const struct status_to_posix_error smb2_error_map_table[] = {
 	{STATUS_NONEXISTENT_SECTOR, -EIO, "STATUS_NONEXISTENT_SECTOR"},
 	{STATUS_MORE_PROCESSING_REQUIRED, -EIO,
 	"STATUS_MORE_PROCESSING_REQUIRED"},
-	{STATUS_NO_MEMORY, -ENOMEM, "STATUS_NO_MEMORY"},
+	{STATUS_NO_MEMORY, -EREMOTEIO, "STATUS_NO_MEMORY"},
 	{STATUS_CONFLICTING_ADDRESSES, -EADDRINUSE,
 	"STATUS_CONFLICTING_ADDRESSES"},
 	{STATUS_NOT_MAPPED_VIEW, -EIO, "STATUS_NOT_MAPPED_VIEW"},
diff --git a/fs/cifs/smb2ops.c b/fs/cifs/smb2ops.c
index c571be8cb76e..11dde4b24f8a 100644
--- a/fs/cifs/smb2ops.c
+++ b/fs/cifs/smb2ops.c
@@ -494,6 +494,85 @@ smb2_close_file(const unsigned int xid, struct cifs_tcon *tcon,
 }
 
 static int
+SMB2_request_res_key(const unsigned int xid, struct cifs_tcon *tcon,
+		     u64 persistent_fid, u64 volatile_fid,
+		     struct copychunk_ioctl *pcchunk)
+{
+	int rc;
+	unsigned int ret_data_len;
+	struct resume_key_req *res_key;
+
+	rc = SMB2_ioctl(xid, tcon, persistent_fid, volatile_fid,
+			FSCTL_SRV_REQUEST_RESUME_KEY, true /* is_fsctl */,
+			NULL, 0 /* no input */,
+			(char **)&res_key, &ret_data_len);
+
+	if (rc) {
+		cifs_dbg(VFS, "refcpy ioctl error %d getting resume key\n", rc);
+		goto req_res_key_exit;
+	}
+	if (ret_data_len < sizeof(struct resume_key_req)) {
+		cifs_dbg(VFS, "Invalid refcopy resume key length\n");
+		rc = -EINVAL;
+		goto req_res_key_exit;
+	}
+	memcpy(pcchunk->SourceKey, res_key->ResumeKey, COPY_CHUNK_RES_KEY_SIZE);
+
+req_res_key_exit:
+	kfree(res_key);
+	return rc;
+}
+
+static int
+smb2_clone_range(const unsigned int xid,
+			struct cifsFileInfo *srcfile,
+			struct cifsFileInfo *trgtfile, u64 src_off,
+			u64 len, u64 dest_off)
+{
+	int rc;
+	unsigned int ret_data_len;
+	struct copychunk_ioctl *pcchunk;
+	char *retbuf = NULL;
+
+	pcchunk = kmalloc(sizeof(struct copychunk_ioctl), GFP_KERNEL);
+
+	if (pcchunk == NULL)
+		return -ENOMEM;
+
+	cifs_dbg(FYI, "in smb2_clone_range - about to call request res key\n");
+	/* Request a key from the server to identify the source of the copy */
+	rc = SMB2_request_res_key(xid, tlink_tcon(srcfile->tlink),
+				srcfile->fid.persistent_fid,
+				srcfile->fid.volatile_fid, pcchunk);
+
+	/* Note: request_res_key sets res_key null only if rc !=0 */
+	if (rc)
+		return rc;
+
+	/* For now array only one chunk long, will make more flexible later */
+	pcchunk->ChunkCount = __constant_cpu_to_le32(1);
+	pcchunk->Reserved = 0;
+	pcchunk->SourceOffset = cpu_to_le64(src_off);
+	pcchunk->TargetOffset = cpu_to_le64(dest_off);
+	pcchunk->Length = cpu_to_le32(len);
+	pcchunk->Reserved2 = 0;
+
+	/* Request that server copy to target from src file identified by key */
+	rc = SMB2_ioctl(xid, tlink_tcon(trgtfile->tlink),
+			trgtfile->fid.persistent_fid,
+			trgtfile->fid.volatile_fid, FSCTL_SRV_COPYCHUNK_WRITE,
+			true /* is_fsctl */, (char *)pcchunk,
+			sizeof(struct copychunk_ioctl),	&retbuf, &ret_data_len);
+
+	/* BB need to special case rc = EINVAL to alter chunk size */
+
+	cifs_dbg(FYI, "rc %d data length out %d\n", rc, ret_data_len);
+
+	kfree(pcchunk);
+	return rc;
+}
+
+static int
 smb2_flush_file(const unsigned int xid, struct cifs_tcon *tcon,
 		struct cifs_fid *fid)
 {
@@ -1017,6 +1096,7 @@ struct smb_version_operations smb20_operations = {
 	.set_oplock_level = smb2_set_oplock_level,
 	.create_lease_buf = smb2_create_lease_buf,
 	.parse_lease_buf = smb2_parse_lease_buf,
+	.clone_range = smb2_clone_range,
 };
 
 struct smb_version_operations smb21_operations = {
@@ -1090,6 +1170,7 @@ struct smb_version_operations smb21_operations = {
 	.set_oplock_level = smb21_set_oplock_level,
 	.create_lease_buf = smb2_create_lease_buf,
 	.parse_lease_buf = smb2_parse_lease_buf,
+	.clone_range = smb2_clone_range,
 };
 
 struct smb_version_operations smb30_operations = {
@@ -1165,6 +1246,7 @@ struct smb_version_operations smb30_operations = {
 	.set_oplock_level = smb3_set_oplock_level,
 	.create_lease_buf = smb3_create_lease_buf,
 	.parse_lease_buf = smb3_parse_lease_buf,
+	.clone_range = smb2_clone_range,
 };
 
 struct smb_version_values smb20_values = {
diff --git a/fs/cifs/smb2pdu.c b/fs/cifs/smb2pdu.c
index 8ab05b0d6778..d65270c290a1 100644
--- a/fs/cifs/smb2pdu.c
+++ b/fs/cifs/smb2pdu.c
@@ -630,6 +630,8 @@ ssetup_ntlmssp_authenticate:
 		goto ssetup_exit;
 
 	ses->session_flags = le16_to_cpu(rsp->SessionFlags);
+	if (ses->session_flags & SMB2_SESSION_FLAG_ENCRYPT_DATA)
+		cifs_dbg(VFS, "SMB3 encryption not supported yet\n");
 ssetup_exit:
 	free_rsp_buf(resp_buftype, rsp);
 
@@ -717,6 +719,14 @@ static inline void cifs_stats_fail_inc(struct cifs_tcon *tcon, uint16_t code)
 
 #define MAX_SHARENAME_LENGTH (255 /* server */ + 80 /* share */ + 1 /* NULL */)
 
+/* These are similar values to what Windows uses */
+static inline void init_copy_chunk_defaults(struct cifs_tcon *tcon)
+{
+	tcon->max_chunks = 256;
+	tcon->max_bytes_chunk = 1048576;
+	tcon->max_bytes_copy = 16777216;
+}
+
 int
 SMB2_tcon(const unsigned int xid, struct cifs_ses *ses, const char *tree,
 	  struct cifs_tcon *tcon, const struct nls_table *cp)
@@ -818,7 +828,7 @@ SMB2_tcon(const unsigned int xid, struct cifs_ses *ses, const char *tree,
 	if ((rsp->Capabilities & SMB2_SHARE_CAP_DFS) &&
 	    ((tcon->share_flags & SHI1005_FLAGS_DFS) == 0))
 		cifs_dbg(VFS, "DFS capability contradicts DFS flag\n");
-
+	init_copy_chunk_defaults(tcon);
 tcon_exit:
 	free_rsp_buf(resp_buftype, rsp);
 	kfree(unc_path);
diff --git a/fs/cifs/smb2pdu.h b/fs/cifs/smb2pdu.h
index 6183b1b7550f..f88320bbb477 100644
--- a/fs/cifs/smb2pdu.h
+++ b/fs/cifs/smb2pdu.h
@@ -122,6 +122,23 @@ struct smb2_pdu {
 	__le16 StructureSize2; /* size of wct area (varies, request specific) */
 } __packed;
 
+struct smb2_transform_hdr {
+	__be32 smb2_buf_length;	/* big endian on wire */
+				/* length is only two or three bytes - with
+				 one or two byte type preceding it that MBZ */
+	__u8   ProtocolId[4];	/* 0xFD 'S' 'M' 'B' */
+	__u8   Signature[16];
+	__u8   Nonce[11];
+	__u8   Reserved[5];
+	__le32 OriginalMessageSize;
+	__u16  Reserved1;
+	__le16 EncryptionAlgorithm;
+	__u64  SessionId;
+} __packed;
+
+/* Encryption Algorithms */
+#define SMB2_ENCRYPTION_AES128_CCM	__constant_cpu_to_le16(0x0001)
+
 /*
  *	SMB2 flag definitions
  */
@@ -237,6 +254,7 @@ struct smb2_sess_setup_req {
 /* Currently defined SessionFlags */
 #define SMB2_SESSION_FLAG_IS_GUEST	0x0001
 #define SMB2_SESSION_FLAG_IS_NULL	0x0002
+#define SMB2_SESSION_FLAG_ENCRYPT_DATA	0x0004
 struct smb2_sess_setup_rsp {
 	struct smb2_hdr hdr;
 	__le16 StructureSize; /* Must be 9 */
@@ -534,9 +552,16 @@ struct create_durable {
 	} Data;
 } __packed;
 
+#define COPY_CHUNK_RES_KEY_SIZE	24
+struct resume_key_req {
+	char ResumeKey[COPY_CHUNK_RES_KEY_SIZE];
+	__le32	ContextLength;	/* MBZ */
+	char	Context[0];	/* ignored, Windows sets to 4 bytes of zero */
+} __packed;
+
 /* this goes in the ioctl buffer when doing a copychunk request */
 struct copychunk_ioctl {
-	char SourceKey[24];
+	char SourceKey[COPY_CHUNK_RES_KEY_SIZE];
 	__le32 ChunkCount; /* we are only sending 1 */
 	__le32 Reserved;
 	/* array will only be one chunk long for us */
@@ -546,6 +571,12 @@ struct copychunk_ioctl {
 	__u32 Reserved2;
 } __packed;
 
+struct copychunk_ioctl_rsp {
+	__le32 ChunksWritten;
+	__le32 ChunkBytesWritten;
+	__le32 TotalBytesWritten;
+} __packed;
+
 /* Response and Request are the same format */
 struct validate_negotiate_info {
 	__le32 Capabilities;
diff --git a/fs/cifs/smb2proto.h b/fs/cifs/smb2proto.h
index 313813e4c19b..b4eea105b08c 100644
--- a/fs/cifs/smb2proto.h
+++ b/fs/cifs/smb2proto.h
@@ -61,7 +61,7 @@ extern void move_smb2_info_to_cifs(FILE_ALL_INFO *dst,
 extern int smb2_query_path_info(const unsigned int xid, struct cifs_tcon *tcon,
 				struct cifs_sb_info *cifs_sb,
 				const char *full_path, FILE_ALL_INFO *data,
-				bool *adjust_tz);
+				bool *adjust_tz, bool *symlink);
 extern int smb2_set_path_size(const unsigned int xid, struct cifs_tcon *tcon,
 			      const char *full_path, __u64 size,
 			      struct cifs_sb_info *cifs_sb, bool set_alloc);
diff --git a/fs/dcache.c b/fs/dcache.c
index a9dd384c5e80..0a38ef8d7f00 100644
--- a/fs/dcache.c
+++ b/fs/dcache.c
@@ -2606,7 +2606,7 @@ static void __d_move(struct dentry * dentry, struct dentry * target)
 	dentry_lock_for_move(dentry, target);
 
 	write_seqcount_begin(&dentry->d_seq);
-	write_seqcount_begin(&target->d_seq);
+	write_seqcount_begin_nested(&target->d_seq, DENTRY_D_LOCK_NESTED);
 
 	/* __d_drop does write_seqcount_barrier, but they're OK to nest. */
 
@@ -2738,7 +2738,7 @@ static void __d_materialise_dentry(struct dentry *dentry, struct dentry *anon)
 	dentry_lock_for_move(anon, dentry);
 
 	write_seqcount_begin(&dentry->d_seq);
-	write_seqcount_begin(&anon->d_seq);
+	write_seqcount_begin_nested(&anon->d_seq, DENTRY_D_LOCK_NESTED);
 
 	dparent = dentry->d_parent;
 
diff --git a/fs/dlm/netlink.c b/fs/dlm/netlink.c
index 60a327863b11..e7cfbaf8d0e2 100644
--- a/fs/dlm/netlink.c
+++ b/fs/dlm/netlink.c
@@ -74,14 +74,16 @@ static int user_cmd(struct sk_buff *skb, struct genl_info *info)
 	return 0;
 }
 
-static struct genl_ops dlm_nl_ops = {
-	.cmd		= DLM_CMD_HELLO,
-	.doit		= user_cmd,
+static struct genl_ops dlm_nl_ops[] = {
+	{
+		.cmd	= DLM_CMD_HELLO,
+		.doit	= user_cmd,
+	},
 };
 
 int __init dlm_netlink_init(void)
 {
-	return genl_register_family_with_ops(&family, &dlm_nl_ops, 1);
+	return genl_register_family_with_ops(&family, dlm_nl_ops);
 }
 
 void dlm_netlink_exit(void)
diff --git a/fs/ecryptfs/crypto.c b/fs/ecryptfs/crypto.c
index 000eae2782b6..2f6735dbf1a9 100644
--- a/fs/ecryptfs/crypto.c
+++ b/fs/ecryptfs/crypto.c
@@ -392,7 +392,7 @@ static int crypt_scatterlist(struct ecryptfs_crypt_stat *crypt_stat,
 
 		wait_for_completion(&ecr->completion);
 		rc = ecr->rc;
-		INIT_COMPLETION(ecr->completion);
+		reinit_completion(&ecr->completion);
 	}
 out:
 	ablkcipher_request_free(req);
diff --git a/fs/ext4/balloc.c b/fs/ext4/balloc.c
index dc5d572ebd6a..6ea7b1436bbc 100644
--- a/fs/ext4/balloc.c
+++ b/fs/ext4/balloc.c
@@ -640,6 +640,7 @@ ext4_fsblk_t ext4_count_free_clusters(struct super_block *sb)
 	struct ext4_group_desc *gdp;
 	ext4_group_t i;
 	ext4_group_t ngroups = ext4_get_groups_count(sb);
+	struct ext4_group_info *grp;
 #ifdef EXT4FS_DEBUG
 	struct ext4_super_block *es;
 	ext4_fsblk_t bitmap_count;
@@ -655,7 +656,11 @@ ext4_fsblk_t ext4_count_free_clusters(struct super_block *sb)
 		gdp = ext4_get_group_desc(sb, i, NULL);
 		if (!gdp)
 			continue;
-		desc_count += ext4_free_group_clusters(sb, gdp);
+		grp = NULL;
+		if (EXT4_SB(sb)->s_group_info)
+			grp = ext4_get_group_info(sb, i);
+		if (!grp || !EXT4_MB_GRP_BBITMAP_CORRUPT(grp))
+			desc_count += ext4_free_group_clusters(sb, gdp);
 		brelse(bitmap_bh);
 		bitmap_bh = ext4_read_block_bitmap(sb, i);
 		if (bitmap_bh == NULL)
@@ -679,7 +684,11 @@ ext4_fsblk_t ext4_count_free_clusters(struct super_block *sb)
 		gdp = ext4_get_group_desc(sb, i, NULL);
 		if (!gdp)
 			continue;
-		desc_count += ext4_free_group_clusters(sb, gdp);
+		grp = NULL;
+		if (EXT4_SB(sb)->s_group_info)
+			grp = ext4_get_group_info(sb, i);
+		if (!grp || !EXT4_MB_GRP_BBITMAP_CORRUPT(grp))
+			desc_count += ext4_free_group_clusters(sb, gdp);
 	}
 
 	return desc_count;
diff --git a/fs/ext4/ext4.h b/fs/ext4/ext4.h
index d01d62315f7e..e6185031c1cc 100644
--- a/fs/ext4/ext4.h
+++ b/fs/ext4/ext4.h
@@ -29,6 +29,7 @@
 #include <linux/wait.h>
 #include <linux/blockgroup_lock.h>
 #include <linux/percpu_counter.h>
+#include <linux/ratelimit.h>
 #include <crypto/hash.h>
 #ifdef __KERNEL__
 #include <linux/compat.h>
@@ -1314,6 +1315,11 @@ struct ext4_sb_info {
 	unsigned long s_es_last_sorted;
 	struct percpu_counter s_extent_cache_cnt;
 	spinlock_t s_es_lru_lock ____cacheline_aligned_in_smp;
+
+	/* Ratelimit ext4 messages. */
+	struct ratelimit_state s_err_ratelimit_state;
+	struct ratelimit_state s_warning_ratelimit_state;
+	struct ratelimit_state s_msg_ratelimit_state;
 };
 
 static inline struct ext4_sb_info *EXT4_SB(struct super_block *sb)
@@ -1396,7 +1402,18 @@ static inline void ext4_clear_inode_##name(struct inode *inode, int bit) \
 	clear_bit(bit + (offset), &EXT4_I(inode)->i_##field);		\
 }
 
+/* Add these declarations here only so that these functions can be
+ * found by name.  Otherwise, they are very hard to locate. */
+static inline int ext4_test_inode_flag(struct inode *inode, int bit);
+static inline void ext4_set_inode_flag(struct inode *inode, int bit);
+static inline void ext4_clear_inode_flag(struct inode *inode, int bit);
 EXT4_INODE_BIT_FNS(flag, flags, 0)
+
+/* Add these declarations here only so that these functions can be
+ * found by name.  Otherwise, they are very hard to locate. */
+static inline int ext4_test_inode_state(struct inode *inode, int bit);
+static inline void ext4_set_inode_state(struct inode *inode, int bit);
+static inline void ext4_clear_inode_state(struct inode *inode, int bit);
 #if (BITS_PER_LONG < 64)
 EXT4_INODE_BIT_FNS(state, state_flags, 0)
 
diff --git a/fs/ext4/extents.c b/fs/ext4/extents.c
index 54d52afcdb19..35f65cf4f318 100644
--- a/fs/ext4/extents.c
+++ b/fs/ext4/extents.c
@@ -1666,7 +1666,7 @@ int
 ext4_can_extents_be_merged(struct inode *inode, struct ext4_extent *ex1,
 				struct ext4_extent *ex2)
 {
-	unsigned short ext1_ee_len, ext2_ee_len, max_len;
+	unsigned short ext1_ee_len, ext2_ee_len;
 
 	/*
 	 * Make sure that both extents are initialized. We don't merge
@@ -1677,11 +1677,6 @@ ext4_can_extents_be_merged(struct inode *inode, struct ext4_extent *ex1,
 	if (ext4_ext_is_uninitialized(ex1) || ext4_ext_is_uninitialized(ex2))
 		return 0;
 
-	if (ext4_ext_is_uninitialized(ex1))
-		max_len = EXT_UNINIT_MAX_LEN;
-	else
-		max_len = EXT_INIT_MAX_LEN;
-
 	ext1_ee_len = ext4_ext_get_actual_len(ex1);
 	ext2_ee_len = ext4_ext_get_actual_len(ex2);
 
@@ -1694,7 +1689,7 @@ ext4_can_extents_be_merged(struct inode *inode, struct ext4_extent *ex1,
 	 * as an RO_COMPAT feature, refuse to merge to extents if
 	 * this can result in the top bit of ee_len being set.
 	 */
-	if (ext1_ee_len + ext2_ee_len > max_len)
+	if (ext1_ee_len + ext2_ee_len > EXT_INIT_MAX_LEN)
 		return 0;
 #ifdef AGGRESSIVE_TEST
 	if (ext1_ee_len >= 4)
@@ -1720,7 +1715,6 @@ static int ext4_ext_try_to_merge_right(struct inode *inode,
 	struct ext4_extent_header *eh;
 	unsigned int depth, len;
 	int merge_done = 0;
-	int uninitialized = 0;
 
 	depth = ext_depth(inode);
 	BUG_ON(path[depth].p_hdr == NULL);
@@ -1730,12 +1724,8 @@ static int ext4_ext_try_to_merge_right(struct inode *inode,
 		if (!ext4_can_extents_be_merged(inode, ex, ex + 1))
 			break;
 		/* merge with next extent! */
-		if (ext4_ext_is_uninitialized(ex))
-			uninitialized = 1;
 		ex->ee_len = cpu_to_le16(ext4_ext_get_actual_len(ex)
 				+ ext4_ext_get_actual_len(ex + 1));
-		if (uninitialized)
-			ext4_ext_mark_uninitialized(ex);
 
 		if (ex + 1 < EXT_LAST_EXTENT(eh)) {
 			len = (EXT_LAST_EXTENT(eh) - ex - 1)
@@ -1890,7 +1880,6 @@ int ext4_ext_insert_extent(handle_t *handle, struct inode *inode,
 	struct ext4_ext_path *npath = NULL;
 	int depth, len, err;
 	ext4_lblk_t next;
-	unsigned uninitialized = 0;
 	int mb_flags = 0;
 
 	if (unlikely(ext4_ext_get_actual_len(newext) == 0)) {
@@ -1942,18 +1931,8 @@ int ext4_ext_insert_extent(handle_t *handle, struct inode *inode,
 			if (err)
 				return err;
 
-			/*
-			 * ext4_can_extents_be_merged should have checked
-			 * that either both extents are uninitialized, or
-			 * both aren't. Thus we need to check only one of
-			 * them here.
-			 */
-			if (ext4_ext_is_uninitialized(ex))
-				uninitialized = 1;
 			ex->ee_len = cpu_to_le16(ext4_ext_get_actual_len(ex)
 					+ ext4_ext_get_actual_len(newext));
-			if (uninitialized)
-				ext4_ext_mark_uninitialized(ex);
 			eh = path[depth].p_hdr;
 			nearex = ex;
 			goto merge;
@@ -1976,20 +1955,10 @@ prepend:
 			if (err)
 				return err;
 
-			/*
-			 * ext4_can_extents_be_merged should have checked
-			 * that either both extents are uninitialized, or
-			 * both aren't. Thus we need to check only one of
-			 * them here.
-			 */
-			if (ext4_ext_is_uninitialized(ex))
-				uninitialized = 1;
 			ex->ee_block = newext->ee_block;
 			ext4_ext_store_pblock(ex, ext4_ext_pblock(newext));
 			ex->ee_len = cpu_to_le16(ext4_ext_get_actual_len(ex)
 					+ ext4_ext_get_actual_len(newext));
-			if (uninitialized)
-				ext4_ext_mark_uninitialized(ex);
 			eh = path[depth].p_hdr;
 			nearex = ex;
 			goto merge;
diff --git a/fs/ext4/ialloc.c b/fs/ext4/ialloc.c
index 137193ff389b..0ee59a6644e2 100644
--- a/fs/ext4/ialloc.c
+++ b/fs/ext4/ialloc.c
@@ -432,7 +432,7 @@ static int find_group_orlov(struct super_block *sb, struct inode *parent,
 			ext4fs_dirhash(qstr->name, qstr->len, &hinfo);
 			grp = hinfo.hash;
 		} else
-			get_random_bytes(&grp, sizeof(grp));
+			grp = prandom_u32();
 		parent_group = (unsigned)grp % ngroups;
 		for (i = 0; i < ngroups; i++) {
 			g = (parent_group + i) % ngroups;
diff --git a/fs/ext4/inline.c b/fs/ext4/inline.c
index d9ecbf1113a7..bae987549dc3 100644
--- a/fs/ext4/inline.c
+++ b/fs/ext4/inline.c
@@ -994,11 +994,9 @@ static int ext4_add_dirent_to_inline(handle_t *handle,
 	struct inode	*dir = dentry->d_parent->d_inode;
 	const char	*name = dentry->d_name.name;
 	int		namelen = dentry->d_name.len;
-	unsigned short	reclen;
 	int		err;
 	struct ext4_dir_entry_2 *de;
 
-	reclen = EXT4_DIR_REC_LEN(namelen);
 	err = ext4_find_dest_de(dir, inode, iloc->bh,
 				inline_start, inline_size,
 				name, namelen, &de);
@@ -1442,6 +1440,7 @@ int ext4_read_inline_dir(struct file *file,
 	if (ret < 0)
 		goto out;
 
+	ret = 0;
 	sb = inode->i_sb;
 	parent_ino = le32_to_cpu(((struct ext4_dir_entry_2 *)dir_buf)->inode);
 	offset = ctx->pos;
diff --git a/fs/ext4/inode.c b/fs/ext4/inode.c
index e274e9c1171f..075763474118 100644
--- a/fs/ext4/inode.c
+++ b/fs/ext4/inode.c
@@ -2178,6 +2178,9 @@ static int mpage_map_one_extent(handle_t *handle, struct mpage_da_data *mpd)
  *
  * @handle - handle for journal operations
  * @mpd - extent to map
+ * @give_up_on_write - we set this to true iff there is a fatal error and there
+ *                     is no hope of writing the data. The caller should discard
+ *                     dirty pages to avoid infinite loops.
  *
  * The function maps extent starting at mpd->lblk of length mpd->len. If it is
  * delayed, blocks are allocated, if it is unwritten, we may need to convert
@@ -2295,6 +2298,7 @@ static int mpage_prepare_extent_to_map(struct mpage_da_data *mpd)
 	struct address_space *mapping = mpd->inode->i_mapping;
 	struct pagevec pvec;
 	unsigned int nr_pages;
+	long left = mpd->wbc->nr_to_write;
 	pgoff_t index = mpd->first_page;
 	pgoff_t end = mpd->last_page;
 	int tag;
@@ -2330,6 +2334,17 @@ static int mpage_prepare_extent_to_map(struct mpage_da_data *mpd)
 			if (page->index > end)
 				goto out;
 
+			/*
+			 * Accumulated enough dirty pages? This doesn't apply
+			 * to WB_SYNC_ALL mode. For integrity sync we have to
+			 * keep going because someone may be concurrently
+			 * dirtying pages, and we might have synced a lot of
+			 * newly appeared dirty pages, but have not synced all
+			 * of the old dirty pages.
+			 */
+			if (mpd->wbc->sync_mode == WB_SYNC_NONE && left <= 0)
+				goto out;
+
 			/* If we can't merge this page, we are done. */
 			if (mpd->map.m_len > 0 && mpd->next_page != page->index)
 				goto out;
@@ -2364,19 +2379,7 @@ static int mpage_prepare_extent_to_map(struct mpage_da_data *mpd)
 			if (err <= 0)
 				goto out;
 			err = 0;
-
-			/*
-			 * Accumulated enough dirty pages? This doesn't apply
-			 * to WB_SYNC_ALL mode. For integrity sync we have to
-			 * keep going because someone may be concurrently
-			 * dirtying pages, and we might have synced a lot of
-			 * newly appeared dirty pages, but have not synced all
-			 * of the old dirty pages.
-			 */
-			if (mpd->wbc->sync_mode == WB_SYNC_NONE &&
-			    mpd->next_page - mpd->first_page >=
-							mpd->wbc->nr_to_write)
-				goto out;
+			left--;
 		}
 		pagevec_release(&pvec);
 		cond_resched();
@@ -2420,16 +2423,15 @@ static int ext4_writepages(struct address_space *mapping,
 	 * because that could violate lock ordering on umount
 	 */
 	if (!mapping->nrpages || !mapping_tagged(mapping, PAGECACHE_TAG_DIRTY))
-		return 0;
+		goto out_writepages;
 
 	if (ext4_should_journal_data(inode)) {
 		struct blk_plug plug;
-		int ret;
 
 		blk_start_plug(&plug);
 		ret = write_cache_pages(mapping, wbc, __writepage, mapping);
 		blk_finish_plug(&plug);
-		return ret;
+		goto out_writepages;
 	}
 
 	/*
@@ -2442,8 +2444,10 @@ static int ext4_writepages(struct address_space *mapping,
 	 * *never* be called, so if that ever happens, we would want
 	 * the stack trace.
 	 */
-	if (unlikely(sbi->s_mount_flags & EXT4_MF_FS_ABORTED))
-		return -EROFS;
+	if (unlikely(sbi->s_mount_flags & EXT4_MF_FS_ABORTED)) {
+		ret = -EROFS;
+		goto out_writepages;
+	}
 
 	if (ext4_should_dioread_nolock(inode)) {
 		/*
@@ -4690,6 +4694,15 @@ int ext4_getattr(struct vfsmount *mnt, struct dentry *dentry,
 	generic_fillattr(inode, stat);
 
 	/*
+	 * If there is inline data in the inode, the inode will normally not
+	 * have data blocks allocated (it may have an external xattr block).
+	 * Report at least one sector for such files, so tools like tar, rsync,
+	 * others doen't incorrectly think the file is completely sparse.
+	 */
+	if (unlikely(ext4_has_inline_data(inode)))
+		stat->blocks += (stat->size + 511) >> 9;
+
+	/*
 	 * We can't update i_blocks if the block allocation is delayed
 	 * otherwise in the case of system crash before the real block
 	 * allocation is done, we will have i_blocks inconsistent with
@@ -4700,9 +4713,8 @@ int ext4_getattr(struct vfsmount *mnt, struct dentry *dentry,
 	 * blocks for this file.
 	 */
 	delalloc_blocks = EXT4_C2B(EXT4_SB(inode->i_sb),
-				EXT4_I(inode)->i_reserved_data_blocks);
-
-	stat->blocks += delalloc_blocks << (inode->i_sb->s_blocksize_bits-9);
+				   EXT4_I(inode)->i_reserved_data_blocks);
+	stat->blocks += delalloc_blocks << (inode->i_sb->s_blocksize_bits - 9);
 	return 0;
 }
 
diff --git a/fs/ext4/mballoc.c b/fs/ext4/mballoc.c
index a41e3ba8cfaa..4d113efa024c 100644
--- a/fs/ext4/mballoc.c
+++ b/fs/ext4/mballoc.c
@@ -4794,8 +4794,8 @@ do_more:
 					 " group:%d block:%d count:%lu failed"
 					 " with %d", block_group, bit, count,
 					 err);
-		}
-
+		} else
+			EXT4_MB_GRP_CLEAR_TRIMMED(e4b.bd_info);
 
 		ext4_lock_group(sb, block_group);
 		mb_clear_bits(bitmap_bh->b_data, bit, count_clusters);
diff --git a/fs/ext4/mmp.c b/fs/ext4/mmp.c
index 214461e42a05..04434ad3e8e0 100644
--- a/fs/ext4/mmp.c
+++ b/fs/ext4/mmp.c
@@ -259,7 +259,7 @@ static unsigned int mmp_new_seq(void)
 	u32 new_seq;
 
 	do {
-		get_random_bytes(&new_seq, sizeof(u32));
+		new_seq = prandom_u32();
 	} while (new_seq > EXT4_MMP_SEQ_MAX);
 
 	return new_seq;
diff --git a/fs/ext4/page-io.c b/fs/ext4/page-io.c
index d7d0c7b46ed4..d488f80ee32d 100644
--- a/fs/ext4/page-io.c
+++ b/fs/ext4/page-io.c
@@ -197,14 +197,15 @@ static void dump_completed_IO(struct inode *inode, struct list_head *head)
 static void ext4_add_complete_io(ext4_io_end_t *io_end)
 {
 	struct ext4_inode_info *ei = EXT4_I(io_end->inode);
+	struct ext4_sb_info *sbi = EXT4_SB(io_end->inode->i_sb);
 	struct workqueue_struct *wq;
 	unsigned long flags;
 
 	/* Only reserved conversions from writeback should enter here */
 	WARN_ON(!(io_end->flag & EXT4_IO_END_UNWRITTEN));
-	WARN_ON(!io_end->handle);
+	WARN_ON(!io_end->handle && sbi->s_journal);
 	spin_lock_irqsave(&ei->i_completed_io_lock, flags);
-	wq = EXT4_SB(io_end->inode->i_sb)->rsv_conversion_wq;
+	wq = sbi->rsv_conversion_wq;
 	if (list_empty(&ei->i_rsv_conversion_list))
 		queue_work(wq, &ei->i_rsv_conversion_work);
 	list_add_tail(&io_end->list, &ei->i_rsv_conversion_list);
diff --git a/fs/ext4/super.c b/fs/ext4/super.c
index 2c2e6cbc6bed..c977f4e4e63b 100644
--- a/fs/ext4/super.c
+++ b/fs/ext4/super.c
@@ -411,20 +411,26 @@ static void ext4_handle_error(struct super_block *sb)
 			sb->s_id);
 }
 
+#define ext4_error_ratelimit(sb)					\
+		___ratelimit(&(EXT4_SB(sb)->s_err_ratelimit_state),	\
+			     "EXT4-fs error")
+
 void __ext4_error(struct super_block *sb, const char *function,
 		  unsigned int line, const char *fmt, ...)
 {
 	struct va_format vaf;
 	va_list args;
 
-	va_start(args, fmt);
-	vaf.fmt = fmt;
-	vaf.va = &args;
-	printk(KERN_CRIT "EXT4-fs error (device %s): %s:%d: comm %s: %pV\n",
-	       sb->s_id, function, line, current->comm, &vaf);
-	va_end(args);
+	if (ext4_error_ratelimit(sb)) {
+		va_start(args, fmt);
+		vaf.fmt = fmt;
+		vaf.va = &args;
+		printk(KERN_CRIT
+		       "EXT4-fs error (device %s): %s:%d: comm %s: %pV\n",
+		       sb->s_id, function, line, current->comm, &vaf);
+		va_end(args);
+	}
 	save_error_info(sb, function, line);
-
 	ext4_handle_error(sb);
 }
 
@@ -438,22 +444,23 @@ void __ext4_error_inode(struct inode *inode, const char *function,
 
 	es->s_last_error_ino = cpu_to_le32(inode->i_ino);
 	es->s_last_error_block = cpu_to_le64(block);
+	if (ext4_error_ratelimit(inode->i_sb)) {
+		va_start(args, fmt);
+		vaf.fmt = fmt;
+		vaf.va = &args;
+		if (block)
+			printk(KERN_CRIT "EXT4-fs error (device %s): %s:%d: "
+			       "inode #%lu: block %llu: comm %s: %pV\n",
+			       inode->i_sb->s_id, function, line, inode->i_ino,
+			       block, current->comm, &vaf);
+		else
+			printk(KERN_CRIT "EXT4-fs error (device %s): %s:%d: "
+			       "inode #%lu: comm %s: %pV\n",
+			       inode->i_sb->s_id, function, line, inode->i_ino,
+			       current->comm, &vaf);
+		va_end(args);
+	}
 	save_error_info(inode->i_sb, function, line);
-	va_start(args, fmt);
-	vaf.fmt = fmt;
-	vaf.va = &args;
-	if (block)
-		printk(KERN_CRIT "EXT4-fs error (device %s): %s:%d: "
-		       "inode #%lu: block %llu: comm %s: %pV\n",
-		       inode->i_sb->s_id, function, line, inode->i_ino,
-		       block, current->comm, &vaf);
-	else
-		printk(KERN_CRIT "EXT4-fs error (device %s): %s:%d: "
-		       "inode #%lu: comm %s: %pV\n",
-		       inode->i_sb->s_id, function, line, inode->i_ino,
-		       current->comm, &vaf);
-	va_end(args);
-
 	ext4_handle_error(inode->i_sb);
 }
 
@@ -469,27 +476,28 @@ void __ext4_error_file(struct file *file, const char *function,
 
 	es = EXT4_SB(inode->i_sb)->s_es;
 	es->s_last_error_ino = cpu_to_le32(inode->i_ino);
+	if (ext4_error_ratelimit(inode->i_sb)) {
+		path = d_path(&(file->f_path), pathname, sizeof(pathname));
+		if (IS_ERR(path))
+			path = "(unknown)";
+		va_start(args, fmt);
+		vaf.fmt = fmt;
+		vaf.va = &args;
+		if (block)
+			printk(KERN_CRIT
+			       "EXT4-fs error (device %s): %s:%d: inode #%lu: "
+			       "block %llu: comm %s: path %s: %pV\n",
+			       inode->i_sb->s_id, function, line, inode->i_ino,
+			       block, current->comm, path, &vaf);
+		else
+			printk(KERN_CRIT
+			       "EXT4-fs error (device %s): %s:%d: inode #%lu: "
+			       "comm %s: path %s: %pV\n",
+			       inode->i_sb->s_id, function, line, inode->i_ino,
+			       current->comm, path, &vaf);
+		va_end(args);
+	}
 	save_error_info(inode->i_sb, function, line);
-	path = d_path(&(file->f_path), pathname, sizeof(pathname));
-	if (IS_ERR(path))
-		path = "(unknown)";
-	va_start(args, fmt);
-	vaf.fmt = fmt;
-	vaf.va = &args;
-	if (block)
-		printk(KERN_CRIT
-		       "EXT4-fs error (device %s): %s:%d: inode #%lu: "
-		       "block %llu: comm %s: path %s: %pV\n",
-		       inode->i_sb->s_id, function, line, inode->i_ino,
-		       block, current->comm, path, &vaf);
-	else
-		printk(KERN_CRIT
-		       "EXT4-fs error (device %s): %s:%d: inode #%lu: "
-		       "comm %s: path %s: %pV\n",
-		       inode->i_sb->s_id, function, line, inode->i_ino,
-		       current->comm, path, &vaf);
-	va_end(args);
-
 	ext4_handle_error(inode->i_sb);
 }
 
@@ -543,11 +551,13 @@ void __ext4_std_error(struct super_block *sb, const char *function,
 	    (sb->s_flags & MS_RDONLY))
 		return;
 
-	errstr = ext4_decode_error(sb, errno, nbuf);
-	printk(KERN_CRIT "EXT4-fs error (device %s) in %s:%d: %s\n",
-	       sb->s_id, function, line, errstr);
-	save_error_info(sb, function, line);
+	if (ext4_error_ratelimit(sb)) {
+		errstr = ext4_decode_error(sb, errno, nbuf);
+		printk(KERN_CRIT "EXT4-fs error (device %s) in %s:%d: %s\n",
+		       sb->s_id, function, line, errstr);
+	}
 
+	save_error_info(sb, function, line);
 	ext4_handle_error(sb);
 }
 
@@ -597,6 +607,9 @@ void __ext4_msg(struct super_block *sb,
 	struct va_format vaf;
 	va_list args;
 
+	if (!___ratelimit(&(EXT4_SB(sb)->s_msg_ratelimit_state), "EXT4-fs"))
+		return;
+
 	va_start(args, fmt);
 	vaf.fmt = fmt;
 	vaf.va = &args;
@@ -610,6 +623,10 @@ void __ext4_warning(struct super_block *sb, const char *function,
 	struct va_format vaf;
 	va_list args;
 
+	if (!___ratelimit(&(EXT4_SB(sb)->s_warning_ratelimit_state),
+			  "EXT4-fs warning"))
+		return;
+
 	va_start(args, fmt);
 	vaf.fmt = fmt;
 	vaf.va = &args;
@@ -633,18 +650,20 @@ __acquires(bitlock)
 	es->s_last_error_block = cpu_to_le64(block);
 	__save_error_info(sb, function, line);
 
-	va_start(args, fmt);
-
-	vaf.fmt = fmt;
-	vaf.va = &args;
-	printk(KERN_CRIT "EXT4-fs error (device %s): %s:%d: group %u, ",
-	       sb->s_id, function, line, grp);
-	if (ino)
-		printk(KERN_CONT "inode %lu: ", ino);
-	if (block)
-		printk(KERN_CONT "block %llu:", (unsigned long long) block);
-	printk(KERN_CONT "%pV\n", &vaf);
-	va_end(args);
+	if (ext4_error_ratelimit(sb)) {
+		va_start(args, fmt);
+		vaf.fmt = fmt;
+		vaf.va = &args;
+		printk(KERN_CRIT "EXT4-fs error (device %s): %s:%d: group %u, ",
+		       sb->s_id, function, line, grp);
+		if (ino)
+			printk(KERN_CONT "inode %lu: ", ino);
+		if (block)
+			printk(KERN_CONT "block %llu:",
+			       (unsigned long long) block);
+		printk(KERN_CONT "%pV\n", &vaf);
+		va_end(args);
+	}
 
 	if (test_opt(sb, ERRORS_CONT)) {
 		ext4_commit_super(sb, 0);
@@ -2606,6 +2625,12 @@ EXT4_RW_ATTR_SBI_UI(mb_group_prealloc, s_mb_group_prealloc);
 EXT4_DEPRECATED_ATTR(max_writeback_mb_bump, 128);
 EXT4_RW_ATTR_SBI_UI(extent_max_zeroout_kb, s_extent_max_zeroout_kb);
 EXT4_ATTR(trigger_fs_error, 0200, NULL, trigger_test_error);
+EXT4_RW_ATTR_SBI_UI(err_ratelimit_interval_ms, s_err_ratelimit_state.interval);
+EXT4_RW_ATTR_SBI_UI(err_ratelimit_burst, s_err_ratelimit_state.burst);
+EXT4_RW_ATTR_SBI_UI(warning_ratelimit_interval_ms, s_warning_ratelimit_state.interval);
+EXT4_RW_ATTR_SBI_UI(warning_ratelimit_burst, s_warning_ratelimit_state.burst);
+EXT4_RW_ATTR_SBI_UI(msg_ratelimit_interval_ms, s_msg_ratelimit_state.interval);
+EXT4_RW_ATTR_SBI_UI(msg_ratelimit_burst, s_msg_ratelimit_state.burst);
 
 static struct attribute *ext4_attrs[] = {
 	ATTR_LIST(delayed_allocation_blocks),
@@ -2623,6 +2648,12 @@ static struct attribute *ext4_attrs[] = {
 	ATTR_LIST(max_writeback_mb_bump),
 	ATTR_LIST(extent_max_zeroout_kb),
 	ATTR_LIST(trigger_fs_error),
+	ATTR_LIST(err_ratelimit_interval_ms),
+	ATTR_LIST(err_ratelimit_burst),
+	ATTR_LIST(warning_ratelimit_interval_ms),
+	ATTR_LIST(warning_ratelimit_burst),
+	ATTR_LIST(msg_ratelimit_interval_ms),
+	ATTR_LIST(msg_ratelimit_burst),
 	NULL,
 };
 
@@ -3037,7 +3068,6 @@ static struct ext4_li_request *ext4_li_request_new(struct super_block *sb,
 {
 	struct ext4_sb_info *sbi = EXT4_SB(sb);
 	struct ext4_li_request *elr;
-	unsigned long rnd;
 
 	elr = kzalloc(sizeof(*elr), GFP_KERNEL);
 	if (!elr)
@@ -3052,10 +3082,8 @@ static struct ext4_li_request *ext4_li_request_new(struct super_block *sb,
 	 * spread the inode table initialization requests
 	 * better.
 	 */
-	get_random_bytes(&rnd, sizeof(rnd));
-	elr->lr_next_sched = jiffies + (unsigned long)rnd %
-			     (EXT4_DEF_LI_MAX_START_DELAY * HZ);
-
+	elr->lr_next_sched = jiffies + (prandom_u32() %
+				(EXT4_DEF_LI_MAX_START_DELAY * HZ));
 	return elr;
 }
 
@@ -4118,6 +4146,11 @@ no_journal:
 	if (es->s_error_count)
 		mod_timer(&sbi->s_err_report, jiffies + 300*HZ); /* 5 minutes */
 
+	/* Enable message ratelimiting. Default is 10 messages per 5 secs. */
+	ratelimit_state_init(&sbi->s_err_ratelimit_state, 5 * HZ, 10);
+	ratelimit_state_init(&sbi->s_warning_ratelimit_state, 5 * HZ, 10);
+	ratelimit_state_init(&sbi->s_msg_ratelimit_state, 5 * HZ, 10);
+
 	kfree(orig_data);
 	return 0;
 
diff --git a/fs/ext4/xattr.c b/fs/ext4/xattr.c
index 03e9bebba198..1423c4816a47 100644
--- a/fs/ext4/xattr.c
+++ b/fs/ext4/xattr.c
@@ -1352,6 +1352,7 @@ retry:
 					new_extra_isize = s_min_extra_isize;
 					kfree(is); is = NULL;
 					kfree(bs); bs = NULL;
+					brelse(bh);
 					goto retry;
 				}
 				error = -1;
diff --git a/fs/fs_struct.c b/fs/fs_struct.c
index d8ac61d0c932..7dca743b2ce1 100644
--- a/fs/fs_struct.c
+++ b/fs/fs_struct.c
@@ -161,6 +161,6 @@ EXPORT_SYMBOL(current_umask);
 struct fs_struct init_fs = {
 	.users		= 1,
 	.lock		= __SPIN_LOCK_UNLOCKED(init_fs.lock),
-	.seq		= SEQCNT_ZERO,
+	.seq		= SEQCNT_ZERO(init_fs.seq),
 	.umask		= 0022,
 };
diff --git a/fs/hfsplus/xattr.c b/fs/hfsplus/xattr.c
index efc85b1377cc..3c6136f98c73 100644
--- a/fs/hfsplus/xattr.c
+++ b/fs/hfsplus/xattr.c
@@ -129,7 +129,7 @@ static int can_set_xattr(struct inode *inode, const char *name,
 
 static void hfsplus_init_header_node(struct inode *attr_file,
 					u32 clump_size,
-					char *buf, size_t node_size)
+					char *buf, u16 node_size)
 {
 	struct hfs_bnode_desc *desc;
 	struct hfs_btree_header_rec *head;
@@ -139,8 +139,9 @@ static void hfsplus_init_header_node(struct inode *attr_file,
 	char *bmp;
 	u32 used_nodes;
 	u32 used_bmp_bytes;
+	loff_t tmp;
 
-	hfs_dbg(ATTR_MOD, "init_hdr_attr_file: clump %u, node_size %zu\n",
+	hfs_dbg(ATTR_MOD, "init_hdr_attr_file: clump %u, node_size %u\n",
 				clump_size, node_size);
 
 	/* The end of the node contains list of record offsets */
@@ -154,7 +155,9 @@ static void hfsplus_init_header_node(struct inode *attr_file,
 
 	head = (struct hfs_btree_header_rec *)(buf + offset);
 	head->node_size = cpu_to_be16(node_size);
-	head->node_count = cpu_to_be32(i_size_read(attr_file) / node_size);
+	tmp = i_size_read(attr_file);
+	do_div(tmp, node_size);
+	head->node_count = cpu_to_be32(tmp);
 	head->free_nodes = cpu_to_be32(be32_to_cpu(head->node_count) - 1);
 	head->clump_size = cpu_to_be32(clump_size);
 	head->attributes |= cpu_to_be32(HFS_TREE_BIGKEYS | HFS_TREE_VARIDXKEYS);
diff --git a/fs/nfs/Kconfig b/fs/nfs/Kconfig
index 38c1768b4142..3dece03f2fc8 100644
--- a/fs/nfs/Kconfig
+++ b/fs/nfs/Kconfig
@@ -116,17 +116,17 @@ config NFS_V4_2
 config PNFS_FILE_LAYOUT
 	tristate
 	depends on NFS_V4_1
-	default m
+	default NFS_V4
 
 config PNFS_BLOCK
 	tristate
 	depends on NFS_V4_1 && BLK_DEV_DM
-	default m
+	default NFS_V4
 
 config PNFS_OBJLAYOUT
 	tristate
 	depends on NFS_V4_1 && SCSI_OSD_ULD
-	default m
+	default NFS_V4
 
 config NFS_V4_1_IMPLEMENTATION_ID_DOMAIN
 	string "NFSv4.1 Implementation ID Domain"
diff --git a/fs/nfs/nfs4state.c b/fs/nfs/nfs4state.c
index c8e729deb4f7..059c01b67a71 100644
--- a/fs/nfs/nfs4state.c
+++ b/fs/nfs/nfs4state.c
@@ -244,7 +244,7 @@ static int nfs4_drain_slot_tbl(struct nfs4_slot_table *tbl)
 	set_bit(NFS4_SLOT_TBL_DRAINING, &tbl->slot_tbl_state);
 	spin_lock(&tbl->slot_tbl_lock);
 	if (tbl->highest_used_slotid != NFS4_NO_SLOT) {
-		INIT_COMPLETION(tbl->complete);
+		reinit_completion(&tbl->complete);
 		spin_unlock(&tbl->slot_tbl_lock);
 		return wait_for_completion_interruptible(&tbl->complete);
 	}
@@ -2093,10 +2093,15 @@ again:
 			nfs4_root_machine_cred(clp);
 			goto again;
 		}
-		if (i > 2)
+		if (clnt->cl_auth->au_flavor == RPC_AUTH_UNIX)
 			break;
 	case -NFS4ERR_CLID_INUSE:
 	case -NFS4ERR_WRONGSEC:
+		/* No point in retrying if we already used RPC_AUTH_UNIX */
+		if (clnt->cl_auth->au_flavor == RPC_AUTH_UNIX) {
+			status = -EPERM;
+			break;
+		}
 		clnt = rpc_clone_client_set_auth(clnt, RPC_AUTH_UNIX);
 		if (IS_ERR(clnt)) {
 			status = PTR_ERR(clnt);
diff --git a/fs/nfs/super.c b/fs/nfs/super.c
index 317d6fc2160e..910ed906eb82 100644
--- a/fs/nfs/super.c
+++ b/fs/nfs/super.c
@@ -1614,7 +1614,7 @@ static int nfs_parse_mount_options(char *raw,
 		goto out_minorversion_mismatch;
 
 	if (mnt->options & NFS_OPTION_MIGRATION &&
-	    mnt->version != 4 && mnt->minorversion != 0)
+	    (mnt->version != 4 || mnt->minorversion != 0))
 		goto out_migration_misuse;
 
 	/*
diff --git a/fs/nfsd/Kconfig b/fs/nfsd/Kconfig
index dc8f1ef665ce..f994e750e0d1 100644
--- a/fs/nfsd/Kconfig
+++ b/fs/nfsd/Kconfig
@@ -95,7 +95,7 @@ config NFSD_V4_SECURITY_LABEL
 	Smack policies on NFSv4 files, say N.
 
 	WARNING: there is still a chance of backwards-incompatible protocol changes.
-	For now we recommend "Y" only for developers and testers."
+	For now we recommend "Y" only for developers and testers.
 
 config NFSD_FAULT_INJECTION
 	bool "NFS server manual fault injection"
diff --git a/fs/nfsd/export.c b/fs/nfsd/export.c
index 5f38ea36e266..8513c598fabf 100644
--- a/fs/nfsd/export.c
+++ b/fs/nfsd/export.c
@@ -536,16 +536,12 @@ static int svc_export_parse(struct cache_detail *cd, char *mesg, int mlen)
 		if (err)
 			goto out3;
 		exp.ex_anon_uid= make_kuid(&init_user_ns, an_int);
-		if (!uid_valid(exp.ex_anon_uid))
-			goto out3;
 
 		/* anon gid */
 		err = get_int(&mesg, &an_int);
 		if (err)
 			goto out3;
 		exp.ex_anon_gid= make_kgid(&init_user_ns, an_int);
-		if (!gid_valid(exp.ex_anon_gid))
-			goto out3;
 
 		/* fsid */
 		err = get_int(&mesg, &an_int);
@@ -583,6 +579,26 @@ static int svc_export_parse(struct cache_detail *cd, char *mesg, int mlen)
 				   exp.ex_uuid);
 		if (err)
 			goto out4;
+		/*
+		 * No point caching this if it would immediately expire.
+		 * Also, this protects exportfs's dummy export from the
+		 * anon_uid/anon_gid checks:
+		 */
+		if (exp.h.expiry_time < seconds_since_boot())
+			goto out4;
+		/*
+		 * For some reason exportfs has been passing down an
+		 * invalid (-1) uid & gid on the "dummy" export which it
+		 * uses to test export support.  To make sure exportfs
+		 * sees errors from check_export we therefore need to
+		 * delay these checks till after check_export:
+		 */
+		err = -EINVAL;
+		if (!uid_valid(exp.ex_anon_uid))
+			goto out4;
+		if (!gid_valid(exp.ex_anon_gid))
+			goto out4;
+		err = 0;
 	}
 
 	expp = svc_export_lookup(&exp);
diff --git a/fs/nfsd/nfs4state.c b/fs/nfsd/nfs4state.c
index f36a30a9f2d1..105d6fa7c514 100644
--- a/fs/nfsd/nfs4state.c
+++ b/fs/nfsd/nfs4state.c
@@ -402,11 +402,16 @@ static void remove_stid(struct nfs4_stid *s)
 	idr_remove(stateids, s->sc_stateid.si_opaque.so_id);
 }
 
+static void nfs4_free_stid(struct kmem_cache *slab, struct nfs4_stid *s)
+{
+	kmem_cache_free(slab, s);
+}
+
 void
 nfs4_put_delegation(struct nfs4_delegation *dp)
 {
 	if (atomic_dec_and_test(&dp->dl_count)) {
-		kmem_cache_free(deleg_slab, dp);
+		nfs4_free_stid(deleg_slab, &dp->dl_stid);
 		num_delegations--;
 	}
 }
@@ -610,7 +615,7 @@ static void close_generic_stateid(struct nfs4_ol_stateid *stp)
 static void free_generic_stateid(struct nfs4_ol_stateid *stp)
 {
 	remove_stid(&stp->st_stid);
-	kmem_cache_free(stateid_slab, stp);
+	nfs4_free_stid(stateid_slab, &stp->st_stid);
 }
 
 static void release_lock_stateid(struct nfs4_ol_stateid *stp)
@@ -668,7 +673,6 @@ static void unhash_open_stateid(struct nfs4_ol_stateid *stp)
 static void release_open_stateid(struct nfs4_ol_stateid *stp)
 {
 	unhash_open_stateid(stp);
-	unhash_stid(&stp->st_stid);
 	free_generic_stateid(stp);
 }
 
@@ -690,7 +694,6 @@ static void release_last_closed_stateid(struct nfs4_openowner *oo)
 	struct nfs4_ol_stateid *s = oo->oo_last_closed_stid;
 
 	if (s) {
-		unhash_stid(&s->st_stid);
 		free_generic_stateid(s);
 		oo->oo_last_closed_stid = NULL;
 	}
@@ -1127,6 +1130,11 @@ destroy_client(struct nfs4_client *clp)
 		dp = list_entry(reaplist.next, struct nfs4_delegation, dl_recall_lru);
 		destroy_delegation(dp);
 	}
+	list_splice_init(&clp->cl_revoked, &reaplist);
+	while (!list_empty(&reaplist)) {
+		dp = list_entry(reaplist.next, struct nfs4_delegation, dl_recall_lru);
+		destroy_revoked_delegation(dp);
+	}
 	while (!list_empty(&clp->cl_openowners)) {
 		oo = list_entry(clp->cl_openowners.next, struct nfs4_openowner, oo_perclient);
 		release_openowner(oo);
@@ -3154,7 +3162,7 @@ nfs4_open_delegation(struct net *net, struct svc_fh *fh,
 	open->op_delegate_type = NFS4_OPEN_DELEGATE_READ;
 	return;
 out_free:
-	unhash_stid(&dp->dl_stid);
+	remove_stid(&dp->dl_stid);
 	nfs4_put_delegation(dp);
 out_no_deleg:
 	open->op_delegate_type = NFS4_OPEN_DELEGATE_NONE;
@@ -3995,10 +4003,9 @@ nfsd4_close(struct svc_rqst *rqstp, struct nfsd4_compound_state *cstate,
 
 	nfsd4_close_open_stateid(stp);
 
-	if (cstate->minorversion) {
-		unhash_stid(&stp->st_stid);
+	if (cstate->minorversion)
 		free_generic_stateid(stp);
-	} else
+	else
 		oo->oo_last_closed_stid = stp;
 
 	if (list_empty(&oo->oo_owner.so_stateids)) {
@@ -5119,7 +5126,6 @@ out_recovery:
 	return ret;
 }
 
-/* should be called with the state lock held */
 void
 nfs4_state_shutdown_net(struct net *net)
 {
@@ -5130,6 +5136,7 @@ nfs4_state_shutdown_net(struct net *net)
 	cancel_delayed_work_sync(&nn->laundromat_work);
 	locks_end_grace(&nn->nfsd4_manager);
 
+	nfs4_lock_state();
 	INIT_LIST_HEAD(&reaplist);
 	spin_lock(&recall_lock);
 	list_for_each_safe(pos, next, &nn->del_recall_lru) {
@@ -5144,6 +5151,7 @@ nfs4_state_shutdown_net(struct net *net)
 
 	nfsd4_client_tracking_exit(net);
 	nfs4_state_destroy_net(net);
+	nfs4_unlock_state();
 }
 
 void
diff --git a/fs/nfsd/nfs4xdr.c b/fs/nfsd/nfs4xdr.c
index d9454fe5653f..088de1355e93 100644
--- a/fs/nfsd/nfs4xdr.c
+++ b/fs/nfsd/nfs4xdr.c
@@ -411,6 +411,7 @@ nfsd4_decode_fattr(struct nfsd4_compoundargs *argp, u32 *bmval,
 		label->data = kzalloc(dummy32 + 1, GFP_KERNEL);
 		if (!label->data)
 			return nfserr_jukebox;
+		label->len = dummy32;
 		defer_free(argp, kfree, label->data);
 		memcpy(label->data, buf, dummy32);
 	}
@@ -945,13 +946,16 @@ static __be32
 nfsd4_decode_open_confirm(struct nfsd4_compoundargs *argp, struct nfsd4_open_confirm *open_conf)
 {
 	DECODE_HEAD;
-		    
+
+	if (argp->minorversion >= 1)
+		return nfserr_notsupp;
+
 	status = nfsd4_decode_stateid(argp, &open_conf->oc_req_stateid);
 	if (status)
 		return status;
 	READ_BUF(4);
 	READ32(open_conf->oc_seqid);
-						        
+
 	DECODE_TAIL;
 }
 
@@ -991,6 +995,14 @@ nfsd4_decode_putfh(struct nfsd4_compoundargs *argp, struct nfsd4_putfh *putfh)
 }
 
 static __be32
+nfsd4_decode_putpubfh(struct nfsd4_compoundargs *argp, void *p)
+{
+	if (argp->minorversion == 0)
+		return nfs_ok;
+	return nfserr_notsupp;
+}
+
+static __be32
 nfsd4_decode_read(struct nfsd4_compoundargs *argp, struct nfsd4_read *read)
 {
 	DECODE_HEAD;
@@ -1061,6 +1073,9 @@ nfsd4_decode_renew(struct nfsd4_compoundargs *argp, clientid_t *clientid)
 {
 	DECODE_HEAD;
 
+	if (argp->minorversion >= 1)
+		return nfserr_notsupp;
+
 	READ_BUF(sizeof(clientid_t));
 	COPYMEM(clientid, sizeof(clientid_t));
 
@@ -1111,6 +1126,9 @@ nfsd4_decode_setclientid(struct nfsd4_compoundargs *argp, struct nfsd4_setclient
 {
 	DECODE_HEAD;
 
+	if (argp->minorversion >= 1)
+		return nfserr_notsupp;
+
 	READ_BUF(NFS4_VERIFIER_SIZE);
 	COPYMEM(setclientid->se_verf.data, NFS4_VERIFIER_SIZE);
 
@@ -1137,6 +1155,9 @@ nfsd4_decode_setclientid_confirm(struct nfsd4_compoundargs *argp, struct nfsd4_s
 {
 	DECODE_HEAD;
 
+	if (argp->minorversion >= 1)
+		return nfserr_notsupp;
+
 	READ_BUF(8 + NFS4_VERIFIER_SIZE);
 	COPYMEM(&scd_c->sc_clientid, 8);
 	COPYMEM(&scd_c->sc_confirm, NFS4_VERIFIER_SIZE);
@@ -1220,6 +1241,9 @@ nfsd4_decode_release_lockowner(struct nfsd4_compoundargs *argp, struct nfsd4_rel
 {
 	DECODE_HEAD;
 
+	if (argp->minorversion >= 1)
+		return nfserr_notsupp;
+
 	READ_BUF(12);
 	COPYMEM(&rlockowner->rl_clientid, sizeof(clientid_t));
 	READ32(rlockowner->rl_owner.len);
@@ -1519,7 +1543,7 @@ static nfsd4_dec nfsd4_dec_ops[] = {
 	[OP_OPEN_CONFIRM]	= (nfsd4_dec)nfsd4_decode_open_confirm,
 	[OP_OPEN_DOWNGRADE]	= (nfsd4_dec)nfsd4_decode_open_downgrade,
 	[OP_PUTFH]		= (nfsd4_dec)nfsd4_decode_putfh,
-	[OP_PUTPUBFH]		= (nfsd4_dec)nfsd4_decode_noop,
+	[OP_PUTPUBFH]		= (nfsd4_dec)nfsd4_decode_putpubfh,
 	[OP_PUTROOTFH]		= (nfsd4_dec)nfsd4_decode_noop,
 	[OP_READ]		= (nfsd4_dec)nfsd4_decode_read,
 	[OP_READDIR]		= (nfsd4_dec)nfsd4_decode_readdir,
@@ -1536,46 +1560,6 @@ static nfsd4_dec nfsd4_dec_ops[] = {
 	[OP_VERIFY]		= (nfsd4_dec)nfsd4_decode_verify,
 	[OP_WRITE]		= (nfsd4_dec)nfsd4_decode_write,
 	[OP_RELEASE_LOCKOWNER]	= (nfsd4_dec)nfsd4_decode_release_lockowner,
-};
-
-static nfsd4_dec nfsd41_dec_ops[] = {
-	[OP_ACCESS]		= (nfsd4_dec)nfsd4_decode_access,
-	[OP_CLOSE]		= (nfsd4_dec)nfsd4_decode_close,
-	[OP_COMMIT]		= (nfsd4_dec)nfsd4_decode_commit,
-	[OP_CREATE]		= (nfsd4_dec)nfsd4_decode_create,
-	[OP_DELEGPURGE]		= (nfsd4_dec)nfsd4_decode_notsupp,
-	[OP_DELEGRETURN]	= (nfsd4_dec)nfsd4_decode_delegreturn,
-	[OP_GETATTR]		= (nfsd4_dec)nfsd4_decode_getattr,
-	[OP_GETFH]		= (nfsd4_dec)nfsd4_decode_noop,
-	[OP_LINK]		= (nfsd4_dec)nfsd4_decode_link,
-	[OP_LOCK]		= (nfsd4_dec)nfsd4_decode_lock,
-	[OP_LOCKT]		= (nfsd4_dec)nfsd4_decode_lockt,
-	[OP_LOCKU]		= (nfsd4_dec)nfsd4_decode_locku,
-	[OP_LOOKUP]		= (nfsd4_dec)nfsd4_decode_lookup,
-	[OP_LOOKUPP]		= (nfsd4_dec)nfsd4_decode_noop,
-	[OP_NVERIFY]		= (nfsd4_dec)nfsd4_decode_verify,
-	[OP_OPEN]		= (nfsd4_dec)nfsd4_decode_open,
-	[OP_OPENATTR]		= (nfsd4_dec)nfsd4_decode_notsupp,
-	[OP_OPEN_CONFIRM]	= (nfsd4_dec)nfsd4_decode_notsupp,
-	[OP_OPEN_DOWNGRADE]	= (nfsd4_dec)nfsd4_decode_open_downgrade,
-	[OP_PUTFH]		= (nfsd4_dec)nfsd4_decode_putfh,
-	[OP_PUTPUBFH]		= (nfsd4_dec)nfsd4_decode_notsupp,
-	[OP_PUTROOTFH]		= (nfsd4_dec)nfsd4_decode_noop,
-	[OP_READ]		= (nfsd4_dec)nfsd4_decode_read,
-	[OP_READDIR]		= (nfsd4_dec)nfsd4_decode_readdir,
-	[OP_READLINK]		= (nfsd4_dec)nfsd4_decode_noop,
-	[OP_REMOVE]		= (nfsd4_dec)nfsd4_decode_remove,
-	[OP_RENAME]		= (nfsd4_dec)nfsd4_decode_rename,
-	[OP_RENEW]		= (nfsd4_dec)nfsd4_decode_notsupp,
-	[OP_RESTOREFH]		= (nfsd4_dec)nfsd4_decode_noop,
-	[OP_SAVEFH]		= (nfsd4_dec)nfsd4_decode_noop,
-	[OP_SECINFO]		= (nfsd4_dec)nfsd4_decode_secinfo,
-	[OP_SETATTR]		= (nfsd4_dec)nfsd4_decode_setattr,
-	[OP_SETCLIENTID]	= (nfsd4_dec)nfsd4_decode_notsupp,
-	[OP_SETCLIENTID_CONFIRM]= (nfsd4_dec)nfsd4_decode_notsupp,
-	[OP_VERIFY]		= (nfsd4_dec)nfsd4_decode_verify,
-	[OP_WRITE]		= (nfsd4_dec)nfsd4_decode_write,
-	[OP_RELEASE_LOCKOWNER]	= (nfsd4_dec)nfsd4_decode_notsupp,
 
 	/* new operations for NFSv4.1 */
 	[OP_BACKCHANNEL_CTL]	= (nfsd4_dec)nfsd4_decode_backchannel_ctl,
@@ -1599,24 +1583,53 @@ static nfsd4_dec nfsd41_dec_ops[] = {
 	[OP_RECLAIM_COMPLETE]	= (nfsd4_dec)nfsd4_decode_reclaim_complete,
 };
 
-struct nfsd4_minorversion_ops {
-	nfsd4_dec *decoders;
-	int nops;
-};
+static inline bool
+nfsd4_opnum_in_range(struct nfsd4_compoundargs *argp, struct nfsd4_op *op)
+{
+	if (op->opnum < FIRST_NFS4_OP)
+		return false;
+	else if (argp->minorversion == 0 && op->opnum > LAST_NFS40_OP)
+		return false;
+	else if (argp->minorversion == 1 && op->opnum > LAST_NFS41_OP)
+		return false;
+	else if (argp->minorversion == 2 && op->opnum > LAST_NFS42_OP)
+		return false;
+	return true;
+}
 
-static struct nfsd4_minorversion_ops nfsd4_minorversion[] = {
-	[0] = { nfsd4_dec_ops, ARRAY_SIZE(nfsd4_dec_ops) },
-	[1] = { nfsd41_dec_ops, ARRAY_SIZE(nfsd41_dec_ops) },
-	[2] = { nfsd41_dec_ops, ARRAY_SIZE(nfsd41_dec_ops) },
-};
+/*
+ * Return a rough estimate of the maximum possible reply size.  Note the
+ * estimate includes rpc headers so is meant to be passed to
+ * svc_reserve, not svc_reserve_auth.
+ *
+ * Also note the current compound encoding permits only one operation to
+ * use pages beyond the first one, so the maximum possible length is the
+ * maximum over these values, not the sum.
+ */
+static int nfsd4_max_reply(u32 opnum)
+{
+	switch (opnum) {
+	case OP_READLINK:
+	case OP_READDIR:
+		/*
+		 * Both of these ops take a single page for data and put
+		 * the head and tail in another page:
+		 */
+		return 2 * PAGE_SIZE;
+	case OP_READ:
+		return INT_MAX;
+	default:
+		return PAGE_SIZE;
+	}
+}
 
 static __be32
 nfsd4_decode_compound(struct nfsd4_compoundargs *argp)
 {
 	DECODE_HEAD;
 	struct nfsd4_op *op;
-	struct nfsd4_minorversion_ops *ops;
 	bool cachethis = false;
+	int max_reply = PAGE_SIZE;
 	int i;
 
 	READ_BUF(4);
@@ -1640,10 +1653,9 @@ nfsd4_decode_compound(struct nfsd4_compoundargs *argp)
 		}
 	}
 
-	if (argp->minorversion >= ARRAY_SIZE(nfsd4_minorversion))
+	if (argp->minorversion > NFSD_SUPPORTED_MINOR_VERSION)
 		argp->opcnt = 0;
 
-	ops = &nfsd4_minorversion[argp->minorversion];
 	for (i = 0; i < argp->opcnt; i++) {
 		op = &argp->ops[i];
 		op->replay = NULL;
@@ -1651,8 +1663,8 @@ nfsd4_decode_compound(struct nfsd4_compoundargs *argp)
 		READ_BUF(4);
 		READ32(op->opnum);
 
-		if (op->opnum >= FIRST_NFS4_OP && op->opnum <= LAST_NFS4_OP)
-			op->status = ops->decoders[op->opnum](argp, &op->u);
+		if (nfsd4_opnum_in_range(argp, op))
+			op->status = nfsd4_dec_ops[op->opnum](argp, &op->u);
 		else {
 			op->opnum = OP_ILLEGAL;
 			op->status = nfserr_op_illegal;
@@ -1667,10 +1679,14 @@ nfsd4_decode_compound(struct nfsd4_compoundargs *argp)
 		 * op in the compound wants to be cached:
 		 */
 		cachethis |= nfsd4_cache_this_op(op);
+
+		max_reply = max(max_reply, nfsd4_max_reply(op->opnum));
 	}
 	/* Sessions make the DRC unnecessary: */
 	if (argp->minorversion)
 		cachethis = false;
+	if (max_reply != INT_MAX)
+		svc_reserve(argp->rqstp, max_reply);
 	argp->rqstp->rq_cachetype = cachethis ? RC_REPLBUFF : RC_NOCACHE;
 
 	DECODE_TAIL;
@@ -2375,7 +2391,7 @@ out_acl:
 	if (bmval0 & FATTR4_WORD0_MAXFILESIZE) {
 		if ((buflen -= 8) < 0)
 			goto out_resource;
-		WRITE64(~(u64)0);
+		WRITE64(exp->ex_path.mnt->mnt_sb->s_maxbytes);
 	}
 	if (bmval0 & FATTR4_WORD0_MAXLINK) {
 		if ((buflen -= 4) < 0)
diff --git a/fs/nfsd/nfsfh.c b/fs/nfsd/nfsfh.c
index 3d0e15ae6f72..3c37b160dcad 100644
--- a/fs/nfsd/nfsfh.c
+++ b/fs/nfsd/nfsfh.c
@@ -598,22 +598,20 @@ fh_update(struct svc_fh *fhp)
 		_fh_update_old(dentry, fhp->fh_export, &fhp->fh_handle);
 	} else {
 		if (fhp->fh_handle.fh_fileid_type != FILEID_ROOT)
-			goto out;
+			return 0;
 
 		_fh_update(fhp, fhp->fh_export, dentry);
 		if (fhp->fh_handle.fh_fileid_type == FILEID_INVALID)
 			return nfserr_opnotsupp;
 	}
-out:
 	return 0;
-
 out_bad:
 	printk(KERN_ERR "fh_update: fh not verified!\n");
-	goto out;
+	return nfserr_serverfault;
 out_negative:
 	printk(KERN_ERR "fh_update: %pd2 still negative!\n",
 		dentry);
-	goto out;
+	return nfserr_serverfault;
 }
 
 /*
diff --git a/fs/ocfs2/dlmglue.c b/fs/ocfs2/dlmglue.c
index 3a44a648dae7..3407b2c62b21 100644
--- a/fs/ocfs2/dlmglue.c
+++ b/fs/ocfs2/dlmglue.c
@@ -1304,7 +1304,7 @@ static int ocfs2_wait_for_mask(struct ocfs2_mask_waiter *mw)
 {
 	wait_for_completion(&mw->mw_complete);
 	/* Re-arm the completion in case we want to wait on it again */
-	INIT_COMPLETION(mw->mw_complete);
+	reinit_completion(&mw->mw_complete);
 	return mw->mw_status;
 }
 
@@ -1355,7 +1355,7 @@ static int ocfs2_wait_for_mask_interruptible(struct ocfs2_mask_waiter *mw,
 	else
 		ret = mw->mw_status;
 	/* Re-arm the completion in case we want to wait on it again */
-	INIT_COMPLETION(mw->mw_complete);
+	reinit_completion(&mw->mw_complete);
 	return ret;
 }
 
diff --git a/fs/proc/consoles.c b/fs/proc/consoles.c
index b701eaa482bf..51942d5abcec 100644
--- a/fs/proc/consoles.c
+++ b/fs/proc/consoles.c
@@ -29,7 +29,6 @@ static int show_console_dev(struct seq_file *m, void *v)
 	char flags[ARRAY_SIZE(con_flags) + 1];
 	struct console *con = v;
 	unsigned int a;
-	int len;
 	dev_t dev = 0;
 
 	if (con->device) {
@@ -47,11 +46,10 @@ static int show_console_dev(struct seq_file *m, void *v)
 			con_flags[a].name : ' ';
 	flags[a] = 0;
 
-	seq_printf(m, "%s%d%n", con->name, con->index, &len);
-	len = 21 - len;
-	if (len < 1)
-		len = 1;
-	seq_printf(m, "%*c%c%c%c (%s)", len, ' ', con->read ? 'R' : '-',
+	seq_setwidth(m, 21 - 1);
+	seq_printf(m, "%s%d", con->name, con->index);
+	seq_pad(m, ' ');
+	seq_printf(m, "%c%c%c (%s)", con->read ? 'R' : '-',
 			con->write ? 'W' : '-', con->unblank ? 'U' : '-',
 			flags);
 	if (dev)
diff --git a/fs/proc/meminfo.c b/fs/proc/meminfo.c
index c805d5b69ba1..a77d2b299199 100644
--- a/fs/proc/meminfo.c
+++ b/fs/proc/meminfo.c
@@ -1,8 +1,8 @@
 #include <linux/fs.h>
-#include <linux/hugetlb.h>
 #include <linux/init.h>
 #include <linux/kernel.h>
 #include <linux/mm.h>
+#include <linux/hugetlb.h>
 #include <linux/mman.h>
 #include <linux/mmzone.h>
 #include <linux/proc_fs.h>
diff --git a/fs/proc/nommu.c b/fs/proc/nommu.c
index ccfd99bd1c5a..5f9bc8a746c9 100644
--- a/fs/proc/nommu.c
+++ b/fs/proc/nommu.c
@@ -39,7 +39,7 @@ static int nommu_region_show(struct seq_file *m, struct vm_region *region)
 	unsigned long ino = 0;
 	struct file *file;
 	dev_t dev = 0;
-	int flags, len;
+	int flags;
 
 	flags = region->vm_flags;
 	file = region->vm_file;
@@ -50,8 +50,9 @@ static int nommu_region_show(struct seq_file *m, struct vm_region *region)
 		ino = inode->i_ino;
 	}
 
+	seq_setwidth(m, 25 + sizeof(void *) * 6 - 1);
 	seq_printf(m,
-		   "%08lx-%08lx %c%c%c%c %08llx %02x:%02x %lu %n",
+		   "%08lx-%08lx %c%c%c%c %08llx %02x:%02x %lu ",
 		   region->vm_start,
 		   region->vm_end,
 		   flags & VM_READ ? 'r' : '-',
@@ -59,13 +60,10 @@ static int nommu_region_show(struct seq_file *m, struct vm_region *region)
 		   flags & VM_EXEC ? 'x' : '-',
 		   flags & VM_MAYSHARE ? flags & VM_SHARED ? 'S' : 's' : 'p',
 		   ((loff_t)region->vm_pgoff) << PAGE_SHIFT,
-		   MAJOR(dev), MINOR(dev), ino, &len);
+		   MAJOR(dev), MINOR(dev), ino);
 
 	if (file) {
-		len = 25 + sizeof(void *) * 6 - len;
-		if (len < 1)
-			len = 1;
-		seq_printf(m, "%*c", len, ' ');
+		seq_pad(m, ' ');
 		seq_path(m, &file->f_path, "");
 	}
 
diff --git a/fs/proc/task_mmu.c b/fs/proc/task_mmu.c
index abbe825d20ff..fb52b548080d 100644
--- a/fs/proc/task_mmu.c
+++ b/fs/proc/task_mmu.c
@@ -62,7 +62,8 @@ void task_mem(struct seq_file *m, struct mm_struct *mm)
 		total_rss << (PAGE_SHIFT-10),
 		data << (PAGE_SHIFT-10),
 		mm->stack_vm << (PAGE_SHIFT-10), text, lib,
-		(PTRS_PER_PTE*sizeof(pte_t)*mm->nr_ptes) >> 10,
+		(PTRS_PER_PTE * sizeof(pte_t) *
+		 atomic_long_read(&mm->nr_ptes)) >> 10,
 		swap << (PAGE_SHIFT-10));
 }
 
@@ -83,14 +84,6 @@ unsigned long task_statm(struct mm_struct *mm,
 	return mm->total_vm;
 }
 
-static void pad_len_spaces(struct seq_file *m, int len)
-{
-	len = 25 + sizeof(void*) * 6 - len;
-	if (len < 1)
-		len = 1;
-	seq_printf(m, "%*c", len, ' ');
-}
-
 #ifdef CONFIG_NUMA
 /*
  * These functions are for numa_maps but called in generic **maps seq_file
@@ -268,7 +261,6 @@ show_map_vma(struct seq_file *m, struct vm_area_struct *vma, int is_pid)
 	unsigned long long pgoff = 0;
 	unsigned long start, end;
 	dev_t dev = 0;
-	int len;
 	const char *name = NULL;
 
 	if (file) {
@@ -286,7 +278,8 @@ show_map_vma(struct seq_file *m, struct vm_area_struct *vma, int is_pid)
 	if (stack_guard_page_end(vma, end))
 		end -= PAGE_SIZE;
 
-	seq_printf(m, "%08lx-%08lx %c%c%c%c %08llx %02x:%02x %lu %n",
+	seq_setwidth(m, 25 + sizeof(void *) * 6 - 1);
+	seq_printf(m, "%08lx-%08lx %c%c%c%c %08llx %02x:%02x %lu ",
 			start,
 			end,
 			flags & VM_READ ? 'r' : '-',
@@ -294,14 +287,14 @@ show_map_vma(struct seq_file *m, struct vm_area_struct *vma, int is_pid)
 			flags & VM_EXEC ? 'x' : '-',
 			flags & VM_MAYSHARE ? 's' : 'p',
 			pgoff,
-			MAJOR(dev), MINOR(dev), ino, &len);
+			MAJOR(dev), MINOR(dev), ino);
 
 	/*
 	 * Print the dentry name for named mappings, and a
 	 * special [heap] marker for the heap:
 	 */
 	if (file) {
-		pad_len_spaces(m, len);
+		seq_pad(m, ' ');
 		seq_path(m, &file->f_path, "\n");
 		goto done;
 	}
@@ -333,7 +326,7 @@ show_map_vma(struct seq_file *m, struct vm_area_struct *vma, int is_pid)
 				name = "[stack]";
 			} else {
 				/* Thread stack in /proc/PID/maps */
-				pad_len_spaces(m, len);
+				seq_pad(m, ' ');
 				seq_printf(m, "[stack:%d]", tid);
 			}
 		}
@@ -341,7 +334,7 @@ show_map_vma(struct seq_file *m, struct vm_area_struct *vma, int is_pid)
 
 done:
 	if (name) {
-		pad_len_spaces(m, len);
+		seq_pad(m, ' ');
 		seq_puts(m, name);
 	}
 	seq_putc(m, '\n');
@@ -505,9 +498,9 @@ static int smaps_pte_range(pmd_t *pmd, unsigned long addr, unsigned long end,
 	pte_t *pte;
 	spinlock_t *ptl;
 
-	if (pmd_trans_huge_lock(pmd, vma) == 1) {
+	if (pmd_trans_huge_lock(pmd, vma, &ptl) == 1) {
 		smaps_pte_entry(*(pte_t *)pmd, addr, HPAGE_PMD_SIZE, walk);
-		spin_unlock(&walk->mm->page_table_lock);
+		spin_unlock(ptl);
 		mss->anonymous_thp += HPAGE_PMD_SIZE;
 		return 0;
 	}
@@ -998,13 +991,14 @@ static int pagemap_pte_range(pmd_t *pmd, unsigned long addr, unsigned long end,
 {
 	struct vm_area_struct *vma;
 	struct pagemapread *pm = walk->private;
+	spinlock_t *ptl;
 	pte_t *pte;
 	int err = 0;
 	pagemap_entry_t pme = make_pme(PM_NOT_PRESENT(pm->v2));
 
 	/* find the first VMA at or above 'addr' */
 	vma = find_vma(walk->mm, addr);
-	if (vma && pmd_trans_huge_lock(pmd, vma) == 1) {
+	if (vma && pmd_trans_huge_lock(pmd, vma, &ptl) == 1) {
 		int pmd_flags2;
 
 		if ((vma->vm_flags & VM_SOFTDIRTY) || pmd_soft_dirty(*pmd))
@@ -1022,7 +1016,7 @@ static int pagemap_pte_range(pmd_t *pmd, unsigned long addr, unsigned long end,
 			if (err)
 				break;
 		}
-		spin_unlock(&walk->mm->page_table_lock);
+		spin_unlock(ptl);
 		return err;
 	}
 
@@ -1324,7 +1318,7 @@ static int gather_pte_stats(pmd_t *pmd, unsigned long addr,
 
 	md = walk->private;
 
-	if (pmd_trans_huge_lock(pmd, md->vma) == 1) {
+	if (pmd_trans_huge_lock(pmd, md->vma, &ptl) == 1) {
 		pte_t huge_pte = *(pte_t *)pmd;
 		struct page *page;
 
@@ -1332,7 +1326,7 @@ static int gather_pte_stats(pmd_t *pmd, unsigned long addr,
 		if (page)
 			gather_stats(page, md, pte_dirty(huge_pte),
 				     HPAGE_PMD_SIZE/PAGE_SIZE);
-		spin_unlock(&walk->mm->page_table_lock);
+		spin_unlock(ptl);
 		return 0;
 	}
 
diff --git a/fs/proc/task_nommu.c b/fs/proc/task_nommu.c
index 56123a6f462e..678455d2d683 100644
--- a/fs/proc/task_nommu.c
+++ b/fs/proc/task_nommu.c
@@ -123,14 +123,6 @@ unsigned long task_statm(struct mm_struct *mm,
 	return size;
 }
 
-static void pad_len_spaces(struct seq_file *m, int len)
-{
-	len = 25 + sizeof(void*) * 6 - len;
-	if (len < 1)
-		len = 1;
-	seq_printf(m, "%*c", len, ' ');
-}
-
 /*
  * display a single VMA to a sequenced file
  */
@@ -142,7 +134,7 @@ static int nommu_vma_show(struct seq_file *m, struct vm_area_struct *vma,
 	unsigned long ino = 0;
 	struct file *file;
 	dev_t dev = 0;
-	int flags, len;
+	int flags;
 	unsigned long long pgoff = 0;
 
 	flags = vma->vm_flags;
@@ -155,8 +147,9 @@ static int nommu_vma_show(struct seq_file *m, struct vm_area_struct *vma,
 		pgoff = (loff_t)vma->vm_pgoff << PAGE_SHIFT;
 	}
 
+	seq_setwidth(m, 25 + sizeof(void *) * 6 - 1);
 	seq_printf(m,
-		   "%08lx-%08lx %c%c%c%c %08llx %02x:%02x %lu %n",
+		   "%08lx-%08lx %c%c%c%c %08llx %02x:%02x %lu ",
 		   vma->vm_start,
 		   vma->vm_end,
 		   flags & VM_READ ? 'r' : '-',
@@ -164,16 +157,16 @@ static int nommu_vma_show(struct seq_file *m, struct vm_area_struct *vma,
 		   flags & VM_EXEC ? 'x' : '-',
 		   flags & VM_MAYSHARE ? flags & VM_SHARED ? 'S' : 's' : 'p',
 		   pgoff,
-		   MAJOR(dev), MINOR(dev), ino, &len);
+		   MAJOR(dev), MINOR(dev), ino);
 
 	if (file) {
-		pad_len_spaces(m, len);
+		seq_pad(m, ' ');
 		seq_path(m, &file->f_path, "");
 	} else if (mm) {
 		pid_t tid = vm_is_stack(priv->task, vma, is_pid);
 
 		if (tid != 0) {
-			pad_len_spaces(m, len);
+			seq_pad(m, ' ');
 			/*
 			 * Thread stack in /proc/PID/task/TID/maps or
 			 * the main process stack.
diff --git a/fs/quota/netlink.c b/fs/quota/netlink.c
index 16e8abb7709b..72d29177998e 100644
--- a/fs/quota/netlink.c
+++ b/fs/quota/netlink.c
@@ -9,13 +9,25 @@
 #include <net/netlink.h>
 #include <net/genetlink.h>
 
+static const struct genl_multicast_group quota_mcgrps[] = {
+	{ .name = "events", },
+};
+
 /* Netlink family structure for quota */
 static struct genl_family quota_genl_family = {
-	.id = GENL_ID_GENERATE,
+	/*
+	 * Needed due to multicast group ID abuse - old code assumed
+	 * the family ID was also a valid multicast group ID (which
+	 * isn't true) and userspace might thus rely on it. Assign a
+	 * static ID for this group to make dealing with that easier.
+	 */
+	.id = GENL_ID_VFS_DQUOT,
 	.hdrsize = 0,
 	.name = "VFS_DQUOT",
 	.version = 1,
 	.maxattr = QUOTA_NL_A_MAX,
+	.mcgrps = quota_mcgrps,
+	.n_mcgrps = ARRAY_SIZE(quota_mcgrps),
 };
 
 /**
@@ -78,7 +90,7 @@ void quota_send_warning(struct kqid qid, dev_t dev,
 		goto attr_err_out;
 	genlmsg_end(skb, msg_head);
 
-	genlmsg_multicast(skb, 0, quota_genl_family.id, GFP_NOFS);
+	genlmsg_multicast(&quota_genl_family, skb, 0, 0, GFP_NOFS);
 	return;
 attr_err_out:
 	printk(KERN_ERR "VFS: Not enough space to compose quota message!\n");
diff --git a/fs/seq_file.c b/fs/seq_file.c
index a290157265ef..1d641bb108d2 100644
--- a/fs/seq_file.c
+++ b/fs/seq_file.c
@@ -136,6 +136,7 @@ static int traverse(struct seq_file *m, loff_t offset)
 Eoverflow:
 	m->op->stop(m, p);
 	kfree(m->buf);
+	m->count = 0;
 	m->buf = kmalloc(m->size <<= 1, GFP_KERNEL);
 	return !m->buf ? -ENOMEM : -EAGAIN;
 }
@@ -232,10 +233,10 @@ ssize_t seq_read(struct file *file, char __user *buf, size_t size, loff_t *ppos)
 			goto Fill;
 		m->op->stop(m, p);
 		kfree(m->buf);
+		m->count = 0;
 		m->buf = kmalloc(m->size <<= 1, GFP_KERNEL);
 		if (!m->buf)
 			goto Enomem;
-		m->count = 0;
 		m->version = 0;
 		pos = m->index;
 		p = m->op->start(m, &pos);
@@ -766,6 +767,21 @@ int seq_write(struct seq_file *seq, const void *data, size_t len)
 }
 EXPORT_SYMBOL(seq_write);
 
+/**
+ * seq_pad - write padding spaces to buffer
+ * @m: seq_file identifying the buffer to which data should be written
+ * @c: the byte to append after padding if non-zero
+ */
+void seq_pad(struct seq_file *m, char c)
+{
+	int size = m->pad_until - m->count;
+	if (size > 0)
+		seq_printf(m, "%*s", size, "");
+	if (c)
+		seq_putc(m, c);
+}
+EXPORT_SYMBOL(seq_pad);
+
 struct list_head *seq_list_start(struct list_head *head, loff_t pos)
 {
 	struct list_head *lh;
diff --git a/fs/xfs/Makefile b/fs/xfs/Makefile
index 0719e4db93f2..c21f43506661 100644
--- a/fs/xfs/Makefile
+++ b/fs/xfs/Makefile
@@ -66,12 +66,14 @@ xfs-y				+= xfs_alloc.o \
 				   xfs_bmap_btree.o \
 				   xfs_btree.o \
 				   xfs_da_btree.o \
+				   xfs_da_format.o \
 				   xfs_dir2.o \
 				   xfs_dir2_block.o \
 				   xfs_dir2_data.o \
 				   xfs_dir2_leaf.o \
 				   xfs_dir2_node.o \
 				   xfs_dir2_sf.o \
+				   xfs_dquot_buf.o \
 				   xfs_ialloc.o \
 				   xfs_ialloc_btree.o \
 				   xfs_icreate_item.o \
@@ -103,7 +105,11 @@ xfs-$(CONFIG_XFS_QUOTA)		+= xfs_dquot.o \
 				   xfs_qm_bhv.o \
 				   xfs_qm.o \
 				   xfs_quotaops.o
-xfs-$(CONFIG_XFS_RT)		+= xfs_rtalloc.o
+
+# xfs_rtbitmap is shared with libxfs
+xfs-$(CONFIG_XFS_RT)		+= xfs_rtalloc.o \
+				   xfs_rtbitmap.o
+
 xfs-$(CONFIG_XFS_POSIX_ACL)	+= xfs_acl.o
 xfs-$(CONFIG_PROC_FS)		+= xfs_stats.o
 xfs-$(CONFIG_SYSCTL)		+= xfs_sysctl.o
diff --git a/fs/xfs/kmem.c b/fs/xfs/kmem.c
index a02cfb9e3bce..66a36befc5c0 100644
--- a/fs/xfs/kmem.c
+++ b/fs/xfs/kmem.c
@@ -63,17 +63,6 @@ kmem_alloc(size_t size, xfs_km_flags_t flags)
 }
 
 void *
-kmem_zalloc(size_t size, xfs_km_flags_t flags)
-{
-	void	*ptr;
-
-	ptr = kmem_alloc(size, flags);
-	if (ptr)
-		memset((char *)ptr, 0, (int)size);
-	return ptr;
-}
-
-void *
 kmem_zalloc_large(size_t size, xfs_km_flags_t flags)
 {
 	void	*ptr;
@@ -128,14 +117,3 @@ kmem_zone_alloc(kmem_zone_t *zone, xfs_km_flags_t flags)
 		congestion_wait(BLK_RW_ASYNC, HZ/50);
 	} while (1);
 }
-
-void *
-kmem_zone_zalloc(kmem_zone_t *zone, xfs_km_flags_t flags)
-{
-	void	*ptr;
-
-	ptr = kmem_zone_alloc(zone, flags);
-	if (ptr)
-		memset((char *)ptr, 0, kmem_cache_size(zone));
-	return ptr;
-}
diff --git a/fs/xfs/kmem.h b/fs/xfs/kmem.h
index 3a7371cab508..64db0e53edea 100644
--- a/fs/xfs/kmem.h
+++ b/fs/xfs/kmem.h
@@ -32,6 +32,7 @@ typedef unsigned __bitwise xfs_km_flags_t;
 #define KM_NOSLEEP	((__force xfs_km_flags_t)0x0002u)
 #define KM_NOFS		((__force xfs_km_flags_t)0x0004u)
 #define KM_MAYFAIL	((__force xfs_km_flags_t)0x0008u)
+#define KM_ZERO		((__force xfs_km_flags_t)0x0010u)
 
 /*
  * We use a special process flag to avoid recursive callbacks into
@@ -43,7 +44,7 @@ kmem_flags_convert(xfs_km_flags_t flags)
 {
 	gfp_t	lflags;
 
-	BUG_ON(flags & ~(KM_SLEEP|KM_NOSLEEP|KM_NOFS|KM_MAYFAIL));
+	BUG_ON(flags & ~(KM_SLEEP|KM_NOSLEEP|KM_NOFS|KM_MAYFAIL|KM_ZERO));
 
 	if (flags & KM_NOSLEEP) {
 		lflags = GFP_ATOMIC | __GFP_NOWARN;
@@ -52,11 +53,14 @@ kmem_flags_convert(xfs_km_flags_t flags)
 		if ((current->flags & PF_FSTRANS) || (flags & KM_NOFS))
 			lflags &= ~__GFP_FS;
 	}
+
+	if (flags & KM_ZERO)
+		lflags |= __GFP_ZERO;
+
 	return lflags;
 }
 
 extern void *kmem_alloc(size_t, xfs_km_flags_t);
-extern void *kmem_zalloc(size_t, xfs_km_flags_t);
 extern void *kmem_zalloc_large(size_t size, xfs_km_flags_t);
 extern void *kmem_realloc(const void *, size_t, size_t, xfs_km_flags_t);
 extern void  kmem_free(const void *);
@@ -64,6 +68,12 @@ extern void  kmem_free(const void *);
 
 extern void *kmem_zalloc_greedy(size_t *, size_t, size_t);
 
+static inline void *
+kmem_zalloc(size_t size, xfs_km_flags_t flags)
+{
+	return kmem_alloc(size, flags | KM_ZERO);
+}
+
 /*
  * Zone interfaces
  */
@@ -102,6 +112,11 @@ kmem_zone_destroy(kmem_zone_t *zone)
 }
 
 extern void *kmem_zone_alloc(kmem_zone_t *, xfs_km_flags_t);
-extern void *kmem_zone_zalloc(kmem_zone_t *, xfs_km_flags_t);
+
+static inline void *
+kmem_zone_zalloc(kmem_zone_t *zone, xfs_km_flags_t flags)
+{
+	return kmem_zone_alloc(zone, flags | KM_ZERO);
+}
 
 #endif /* __XFS_SUPPORT_KMEM_H__ */
diff --git a/fs/xfs/xfs_acl.c b/fs/xfs/xfs_acl.c
index 0e2f37efedd0..370eb3e121d1 100644
--- a/fs/xfs/xfs_acl.c
+++ b/fs/xfs/xfs_acl.c
@@ -16,15 +16,15 @@
  * Inc.,  51 Franklin St, Fifth Floor, Boston, MA  02110-1301  USA
  */
 #include "xfs.h"
+#include "xfs_format.h"
 #include "xfs_log_format.h"
 #include "xfs_trans_resv.h"
-#include "xfs_acl.h"
-#include "xfs_attr.h"
-#include "xfs_bmap_btree.h"
-#include "xfs_inode.h"
 #include "xfs_ag.h"
 #include "xfs_sb.h"
 #include "xfs_mount.h"
+#include "xfs_inode.h"
+#include "xfs_acl.h"
+#include "xfs_attr.h"
 #include "xfs_trace.h"
 #include <linux/slab.h>
 #include <linux/xattr.h>
diff --git a/fs/xfs/xfs_ag.h b/fs/xfs/xfs_ag.h
index 1cb740afd674..3fc109819c34 100644
--- a/fs/xfs/xfs_ag.h
+++ b/fs/xfs/xfs_ag.h
@@ -128,8 +128,6 @@ typedef struct xfs_agf {
 extern int xfs_read_agf(struct xfs_mount *mp, struct xfs_trans *tp,
 			xfs_agnumber_t agno, int flags, struct xfs_buf **bpp);
 
-extern const struct xfs_buf_ops xfs_agf_buf_ops;
-
 /*
  * Size of the unlinked inode hash table in the agi.
  */
@@ -191,8 +189,6 @@ typedef struct xfs_agi {
 extern int xfs_read_agi(struct xfs_mount *mp, struct xfs_trans *tp,
 				xfs_agnumber_t agno, struct xfs_buf **bpp);
 
-extern const struct xfs_buf_ops xfs_agi_buf_ops;
-
 /*
  * The third a.g. block contains the a.g. freelist, an array
  * of block pointers to blocks owned by the allocation btree code.
diff --git a/fs/xfs/xfs_alloc.c b/fs/xfs/xfs_alloc.c
index 5a1393f5e020..9eab2dfdcbb5 100644
--- a/fs/xfs/xfs_alloc.c
+++ b/fs/xfs/xfs_alloc.c
@@ -17,25 +17,25 @@
  */
 #include "xfs.h"
 #include "xfs_fs.h"
-#include "xfs_types.h"
+#include "xfs_format.h"
+#include "xfs_log_format.h"
+#include "xfs_shared.h"
+#include "xfs_trans_resv.h"
 #include "xfs_bit.h"
-#include "xfs_log.h"
-#include "xfs_trans.h"
 #include "xfs_sb.h"
 #include "xfs_ag.h"
 #include "xfs_mount.h"
-#include "xfs_bmap_btree.h"
-#include "xfs_alloc_btree.h"
-#include "xfs_ialloc_btree.h"
-#include "xfs_dinode.h"
 #include "xfs_inode.h"
 #include "xfs_btree.h"
+#include "xfs_alloc_btree.h"
 #include "xfs_alloc.h"
 #include "xfs_extent_busy.h"
 #include "xfs_error.h"
 #include "xfs_cksum.h"
 #include "xfs_trace.h"
+#include "xfs_trans.h"
 #include "xfs_buf_item.h"
+#include "xfs_log.h"
 
 struct workqueue_struct *xfs_alloc_wq;
 
@@ -2294,6 +2294,8 @@ xfs_read_agf(
 {
 	int		error;
 
+	trace_xfs_read_agf(mp, agno);
+
 	ASSERT(agno != NULLAGNUMBER);
 	error = xfs_trans_read_buf(
 			mp, tp, mp->m_ddev_targp,
@@ -2324,8 +2326,9 @@ xfs_alloc_read_agf(
 	struct xfs_perag	*pag;		/* per allocation group data */
 	int			error;
 
-	ASSERT(agno != NULLAGNUMBER);
+	trace_xfs_alloc_read_agf(mp, agno);
 
+	ASSERT(agno != NULLAGNUMBER);
 	error = xfs_read_agf(mp, tp, agno,
 			(flags & XFS_ALLOC_FLAG_TRYLOCK) ? XBF_TRYLOCK : 0,
 			bpp);
diff --git a/fs/xfs/xfs_alloc.h b/fs/xfs/xfs_alloc.h
index 99d0a6101558..feacb061bab7 100644
--- a/fs/xfs/xfs_alloc.h
+++ b/fs/xfs/xfs_alloc.h
@@ -231,7 +231,4 @@ xfs_alloc_get_rec(
 	xfs_extlen_t		*len,	/* output: length of extent */
 	int			*stat);	/* output: success/failure */
 
-extern const struct xfs_buf_ops xfs_agf_buf_ops;
-extern const struct xfs_buf_ops xfs_agfl_buf_ops;
-
 #endif	/* __XFS_ALLOC_H__ */
diff --git a/fs/xfs/xfs_alloc_btree.c b/fs/xfs/xfs_alloc_btree.c
index cafc90251d19..13085429e523 100644
--- a/fs/xfs/xfs_alloc_btree.c
+++ b/fs/xfs/xfs_alloc_btree.c
@@ -17,23 +17,21 @@
  */
 #include "xfs.h"
 #include "xfs_fs.h"
-#include "xfs_types.h"
-#include "xfs_log.h"
-#include "xfs_trans.h"
+#include "xfs_shared.h"
+#include "xfs_format.h"
+#include "xfs_log_format.h"
+#include "xfs_trans_resv.h"
 #include "xfs_sb.h"
 #include "xfs_ag.h"
 #include "xfs_mount.h"
-#include "xfs_bmap_btree.h"
-#include "xfs_alloc_btree.h"
-#include "xfs_ialloc_btree.h"
-#include "xfs_dinode.h"
-#include "xfs_inode.h"
 #include "xfs_btree.h"
+#include "xfs_alloc_btree.h"
 #include "xfs_alloc.h"
 #include "xfs_extent_busy.h"
 #include "xfs_error.h"
 #include "xfs_trace.h"
 #include "xfs_cksum.h"
+#include "xfs_trans.h"
 
 
 STATIC struct xfs_btree_cur *
diff --git a/fs/xfs/xfs_alloc_btree.h b/fs/xfs/xfs_alloc_btree.h
index e3a3f7424192..45e189e7e81c 100644
--- a/fs/xfs/xfs_alloc_btree.h
+++ b/fs/xfs/xfs_alloc_btree.h
@@ -27,39 +27,6 @@ struct xfs_btree_cur;
 struct xfs_mount;
 
 /*
- * There are two on-disk btrees, one sorted by blockno and one sorted
- * by blockcount and blockno.  All blocks look the same to make the code
- * simpler; if we have time later, we'll make the optimizations.
- */
-#define	XFS_ABTB_MAGIC		0x41425442	/* 'ABTB' for bno tree */
-#define	XFS_ABTB_CRC_MAGIC	0x41423342	/* 'AB3B' */
-#define	XFS_ABTC_MAGIC		0x41425443	/* 'ABTC' for cnt tree */
-#define	XFS_ABTC_CRC_MAGIC	0x41423343	/* 'AB3C' */
-
-/*
- * Data record/key structure
- */
-typedef struct xfs_alloc_rec {
-	__be32		ar_startblock;	/* starting block number */
-	__be32		ar_blockcount;	/* count of free blocks */
-} xfs_alloc_rec_t, xfs_alloc_key_t;
-
-typedef struct xfs_alloc_rec_incore {
-	xfs_agblock_t	ar_startblock;	/* starting block number */
-	xfs_extlen_t	ar_blockcount;	/* count of free blocks */
-} xfs_alloc_rec_incore_t;
-
-/* btree pointer type */
-typedef __be32 xfs_alloc_ptr_t;
-
-/*
- * Block numbers in the AG:
- * SB is sector 0, AGF is sector 1, AGI is sector 2, AGFL is sector 3.
- */
-#define	XFS_BNO_BLOCK(mp)	((xfs_agblock_t)(XFS_AGFL_BLOCK(mp) + 1))
-#define	XFS_CNT_BLOCK(mp)	((xfs_agblock_t)(XFS_BNO_BLOCK(mp) + 1))
-
-/*
  * Btree block header size depends on a superblock flag.
  */
 #define XFS_ALLOC_BLOCK_LEN(mp) \
@@ -95,6 +62,4 @@ extern struct xfs_btree_cur *xfs_allocbt_init_cursor(struct xfs_mount *,
 		xfs_agnumber_t, xfs_btnum_t);
 extern int xfs_allocbt_maxrecs(struct xfs_mount *, int, int);
 
-extern const struct xfs_buf_ops xfs_allocbt_buf_ops;
-
 #endif	/* __XFS_ALLOC_BTREE_H__ */
diff --git a/fs/xfs/xfs_aops.c b/fs/xfs/xfs_aops.c
index e51e581454e9..71c8c9d2b882 100644
--- a/fs/xfs/xfs_aops.c
+++ b/fs/xfs/xfs_aops.c
@@ -16,14 +16,15 @@
  * Inc.,  51 Franklin St, Fifth Floor, Boston, MA  02110-1301  USA
  */
 #include "xfs.h"
-#include "xfs_log.h"
+#include "xfs_shared.h"
+#include "xfs_format.h"
+#include "xfs_log_format.h"
+#include "xfs_trans_resv.h"
 #include "xfs_sb.h"
 #include "xfs_ag.h"
-#include "xfs_trans.h"
 #include "xfs_mount.h"
-#include "xfs_bmap_btree.h"
-#include "xfs_dinode.h"
 #include "xfs_inode.h"
+#include "xfs_trans.h"
 #include "xfs_inode_item.h"
 #include "xfs_alloc.h"
 #include "xfs_error.h"
@@ -31,6 +32,8 @@
 #include "xfs_trace.h"
 #include "xfs_bmap.h"
 #include "xfs_bmap_util.h"
+#include "xfs_bmap_btree.h"
+#include "xfs_dinode.h"
 #include <linux/aio.h>
 #include <linux/gfp.h>
 #include <linux/mpage.h>
@@ -333,7 +336,7 @@ xfs_map_blocks(
 
 	if (type == XFS_IO_DELALLOC &&
 	    (!nimaps || isnullstartblock(imap->br_startblock))) {
-		error = xfs_iomap_write_allocate(ip, offset, count, imap);
+		error = xfs_iomap_write_allocate(ip, offset, imap);
 		if (!error)
 			trace_xfs_map_blocks_alloc(ip, offset, count, type, imap);
 		return -XFS_ERROR(error);
@@ -1569,8 +1572,7 @@ xfs_vm_write_begin(
 
 	ASSERT(len <= PAGE_CACHE_SIZE);
 
-	page = grab_cache_page_write_begin(mapping, index,
-					   flags | AOP_FLAG_NOFS);
+	page = grab_cache_page_write_begin(mapping, index, flags);
 	if (!page)
 		return -ENOMEM;
 
diff --git a/fs/xfs/xfs_attr.c b/fs/xfs/xfs_attr.c
index ddcf2267ffa6..b86127072ac3 100644
--- a/fs/xfs/xfs_attr.c
+++ b/fs/xfs/xfs_attr.c
@@ -17,23 +17,24 @@
  */
 #include "xfs.h"
 #include "xfs_fs.h"
+#include "xfs_shared.h"
 #include "xfs_format.h"
+#include "xfs_log_format.h"
+#include "xfs_trans_resv.h"
 #include "xfs_bit.h"
-#include "xfs_log.h"
-#include "xfs_trans.h"
-#include "xfs_trans_priv.h"
 #include "xfs_sb.h"
 #include "xfs_ag.h"
 #include "xfs_mount.h"
+#include "xfs_da_format.h"
 #include "xfs_da_btree.h"
-#include "xfs_bmap_btree.h"
 #include "xfs_attr_sf.h"
-#include "xfs_dinode.h"
 #include "xfs_inode.h"
 #include "xfs_alloc.h"
+#include "xfs_trans.h"
 #include "xfs_inode_item.h"
 #include "xfs_bmap.h"
 #include "xfs_bmap_util.h"
+#include "xfs_bmap_btree.h"
 #include "xfs_attr.h"
 #include "xfs_attr_leaf.h"
 #include "xfs_attr_remote.h"
@@ -41,6 +42,7 @@
 #include "xfs_quota.h"
 #include "xfs_trans_space.h"
 #include "xfs_trace.h"
+#include "xfs_dinode.h"
 
 /*
  * xfs_attr.c
diff --git a/fs/xfs/xfs_attr_inactive.c b/fs/xfs/xfs_attr_inactive.c
index bb24b07cbedb..09480c57f069 100644
--- a/fs/xfs/xfs_attr_inactive.c
+++ b/fs/xfs/xfs_attr_inactive.c
@@ -18,22 +18,20 @@
  */
 #include "xfs.h"
 #include "xfs_fs.h"
+#include "xfs_shared.h"
 #include "xfs_format.h"
+#include "xfs_log_format.h"
+#include "xfs_trans_resv.h"
 #include "xfs_bit.h"
-#include "xfs_log.h"
-#include "xfs_trans.h"
 #include "xfs_sb.h"
 #include "xfs_ag.h"
 #include "xfs_mount.h"
+#include "xfs_da_format.h"
 #include "xfs_da_btree.h"
-#include "xfs_bmap_btree.h"
-#include "xfs_alloc_btree.h"
-#include "xfs_ialloc_btree.h"
+#include "xfs_inode.h"
 #include "xfs_alloc.h"
-#include "xfs_btree.h"
 #include "xfs_attr_remote.h"
-#include "xfs_dinode.h"
-#include "xfs_inode.h"
+#include "xfs_trans.h"
 #include "xfs_inode_item.h"
 #include "xfs_bmap.h"
 #include "xfs_attr.h"
@@ -41,7 +39,8 @@
 #include "xfs_error.h"
 #include "xfs_quota.h"
 #include "xfs_trace.h"
-#include "xfs_trans_priv.h"
+#include "xfs_dinode.h"
+#include "xfs_dir2.h"
 
 /*
  * Look at all the extents for this logical region,
@@ -232,13 +231,13 @@ xfs_attr3_node_inactive(
 	}
 
 	node = bp->b_addr;
-	xfs_da3_node_hdr_from_disk(&ichdr, node);
+	dp->d_ops->node_hdr_from_disk(&ichdr, node);
 	parent_blkno = bp->b_bn;
 	if (!ichdr.count) {
 		xfs_trans_brelse(*trans, bp);
 		return 0;
 	}
-	btree = xfs_da3_node_tree_p(node);
+	btree = dp->d_ops->node_tree_p(node);
 	child_fsb = be32_to_cpu(btree[0].before);
 	xfs_trans_brelse(*trans, bp);	/* no locks for later trans */
 
diff --git a/fs/xfs/xfs_attr_leaf.c b/fs/xfs/xfs_attr_leaf.c
index 86db20a9cc02..7b126f46a2f9 100644
--- a/fs/xfs/xfs_attr_leaf.c
+++ b/fs/xfs/xfs_attr_leaf.c
@@ -18,32 +18,31 @@
  */
 #include "xfs.h"
 #include "xfs_fs.h"
-#include "xfs_types.h"
+#include "xfs_shared.h"
+#include "xfs_format.h"
+#include "xfs_log_format.h"
+#include "xfs_trans_resv.h"
 #include "xfs_bit.h"
-#include "xfs_log.h"
-#include "xfs_trans.h"
-#include "xfs_trans_priv.h"
 #include "xfs_sb.h"
 #include "xfs_ag.h"
 #include "xfs_mount.h"
+#include "xfs_da_format.h"
 #include "xfs_da_btree.h"
-#include "xfs_bmap_btree.h"
-#include "xfs_alloc_btree.h"
-#include "xfs_ialloc_btree.h"
-#include "xfs_alloc.h"
-#include "xfs_btree.h"
-#include "xfs_attr_sf.h"
-#include "xfs_attr_remote.h"
-#include "xfs_dinode.h"
 #include "xfs_inode.h"
+#include "xfs_trans.h"
 #include "xfs_inode_item.h"
+#include "xfs_bmap_btree.h"
 #include "xfs_bmap.h"
+#include "xfs_attr_sf.h"
+#include "xfs_attr_remote.h"
 #include "xfs_attr.h"
 #include "xfs_attr_leaf.h"
 #include "xfs_error.h"
 #include "xfs_trace.h"
 #include "xfs_buf_item.h"
 #include "xfs_cksum.h"
+#include "xfs_dinode.h"
+#include "xfs_dir2.h"
 
 
 /*
@@ -918,8 +917,8 @@ xfs_attr3_leaf_to_node(
 	if (error)
 		goto out;
 	node = bp1->b_addr;
-	xfs_da3_node_hdr_from_disk(&icnodehdr, node);
-	btree = xfs_da3_node_tree_p(node);
+	dp->d_ops->node_hdr_from_disk(&icnodehdr, node);
+	btree = dp->d_ops->node_tree_p(node);
 
 	leaf = bp2->b_addr;
 	xfs_attr3_leaf_hdr_from_disk(&icleafhdr, leaf);
@@ -929,7 +928,7 @@ xfs_attr3_leaf_to_node(
 	btree[0].hashval = entries[icleafhdr.count - 1].hashval;
 	btree[0].before = cpu_to_be32(blkno);
 	icnodehdr.count = 1;
-	xfs_da3_node_hdr_to_disk(node, &icnodehdr);
+	dp->d_ops->node_hdr_to_disk(node, &icnodehdr);
 	xfs_trans_log_buf(args->trans, bp1, 0, XFS_LBSIZE(mp) - 1);
 	error = 0;
 out:
diff --git a/fs/xfs/xfs_attr_leaf.h b/fs/xfs/xfs_attr_leaf.h
index c1022138c7e6..3ec5ec0b8678 100644
--- a/fs/xfs/xfs_attr_leaf.h
+++ b/fs/xfs/xfs_attr_leaf.h
@@ -19,16 +19,6 @@
 #ifndef __XFS_ATTR_LEAF_H__
 #define	__XFS_ATTR_LEAF_H__
 
-/*
- * Attribute storage layout, internal structure, access macros, etc.
- *
- * Attribute lists are structured around Btrees where all the data
- * elements are in the leaf nodes.  Attribute names are hashed into an int,
- * then that int is used as the index into the Btree.  Since the hashval
- * of an attribute name may not be unique, we may have duplicate keys.  The
- * internal links in the Btree are logical block offsets into the file.
- */
-
 struct attrlist;
 struct attrlist_cursor_kern;
 struct xfs_attr_list_context;
@@ -38,226 +28,6 @@ struct xfs_da_state_blk;
 struct xfs_inode;
 struct xfs_trans;
 
-/*========================================================================
- * Attribute structure when equal to XFS_LBSIZE(mp) bytes.
- *========================================================================*/
-
-/*
- * This is the structure of the leaf nodes in the Btree.
- *
- * Struct leaf_entry's are packed from the top.  Name/values grow from the
- * bottom but are not packed.  The freemap contains run-length-encoded entries
- * for the free bytes after the leaf_entry's, but only the N largest such,
- * smaller runs are dropped.  When the freemap doesn't show enough space
- * for an allocation, we compact the name/value area and try again.  If we
- * still don't have enough space, then we have to split the block.  The
- * name/value structs (both local and remote versions) must be 32bit aligned.
- *
- * Since we have duplicate hash keys, for each key that matches, compare
- * the actual name string.  The root and intermediate node search always
- * takes the first-in-the-block key match found, so we should only have
- * to work "forw"ard.  If none matches, continue with the "forw"ard leaf
- * nodes until the hash key changes or the attribute name is found.
- *
- * We store the fact that an attribute is a ROOT/USER/SECURE attribute in
- * the leaf_entry.  The namespaces are independent only because we also look
- * at the namespace bit when we are looking for a matching attribute name.
- *
- * We also store an "incomplete" bit in the leaf_entry.  It shows that an
- * attribute is in the middle of being created and should not be shown to
- * the user if we crash during the time that the bit is set.  We clear the
- * bit when we have finished setting up the attribute.  We do this because
- * we cannot create some large attributes inside a single transaction, and we
- * need some indication that we weren't finished if we crash in the middle.
- */
-#define XFS_ATTR_LEAF_MAPSIZE	3	/* how many freespace slots */
-
-typedef struct xfs_attr_leaf_map {	/* RLE map of free bytes */
-	__be16	base;			  /* base of free region */
-	__be16	size;			  /* length of free region */
-} xfs_attr_leaf_map_t;
-
-typedef struct xfs_attr_leaf_hdr {	/* constant-structure header block */
-	xfs_da_blkinfo_t info;		/* block type, links, etc. */
-	__be16	count;			/* count of active leaf_entry's */
-	__be16	usedbytes;		/* num bytes of names/values stored */
-	__be16	firstused;		/* first used byte in name area */
-	__u8	holes;			/* != 0 if blk needs compaction */
-	__u8	pad1;
-	xfs_attr_leaf_map_t freemap[XFS_ATTR_LEAF_MAPSIZE];
-					/* N largest free regions */
-} xfs_attr_leaf_hdr_t;
-
-typedef struct xfs_attr_leaf_entry {	/* sorted on key, not name */
-	__be32	hashval;		/* hash value of name */
-	__be16	nameidx;		/* index into buffer of name/value */
-	__u8	flags;			/* LOCAL/ROOT/SECURE/INCOMPLETE flag */
-	__u8	pad2;			/* unused pad byte */
-} xfs_attr_leaf_entry_t;
-
-typedef struct xfs_attr_leaf_name_local {
-	__be16	valuelen;		/* number of bytes in value */
-	__u8	namelen;		/* length of name bytes */
-	__u8	nameval[1];		/* name/value bytes */
-} xfs_attr_leaf_name_local_t;
-
-typedef struct xfs_attr_leaf_name_remote {
-	__be32	valueblk;		/* block number of value bytes */
-	__be32	valuelen;		/* number of bytes in value */
-	__u8	namelen;		/* length of name bytes */
-	__u8	name[1];		/* name bytes */
-} xfs_attr_leaf_name_remote_t;
-
-typedef struct xfs_attr_leafblock {
-	xfs_attr_leaf_hdr_t	hdr;	/* constant-structure header block */
-	xfs_attr_leaf_entry_t	entries[1];	/* sorted on key, not name */
-	xfs_attr_leaf_name_local_t namelist;	/* grows from bottom of buf */
-	xfs_attr_leaf_name_remote_t valuelist;	/* grows from bottom of buf */
-} xfs_attr_leafblock_t;
-
-/*
- * CRC enabled leaf structures. Called "version 3" structures to match the
- * version number of the directory and dablk structures for this feature, and
- * attr2 is already taken by the variable inode attribute fork size feature.
- */
-struct xfs_attr3_leaf_hdr {
-	struct xfs_da3_blkinfo	info;
-	__be16			count;
-	__be16			usedbytes;
-	__be16			firstused;
-	__u8			holes;
-	__u8			pad1;
-	struct xfs_attr_leaf_map freemap[XFS_ATTR_LEAF_MAPSIZE];
-	__be32			pad2;		/* 64 bit alignment */
-};
-
-#define XFS_ATTR3_LEAF_CRC_OFF	(offsetof(struct xfs_attr3_leaf_hdr, info.crc))
-
-struct xfs_attr3_leafblock {
-	struct xfs_attr3_leaf_hdr	hdr;
-	struct xfs_attr_leaf_entry	entries[1];
-
-	/*
-	 * The rest of the block contains the following structures after the
-	 * leaf entries, growing from the bottom up. The variables are never
-	 * referenced, the locations accessed purely from helper functions.
-	 *
-	 * struct xfs_attr_leaf_name_local
-	 * struct xfs_attr_leaf_name_remote
-	 */
-};
-
-/*
- * incore, neutral version of the attribute leaf header
- */
-struct xfs_attr3_icleaf_hdr {
-	__uint32_t	forw;
-	__uint32_t	back;
-	__uint16_t	magic;
-	__uint16_t	count;
-	__uint16_t	usedbytes;
-	__uint16_t	firstused;
-	__u8		holes;
-	struct {
-		__uint16_t	base;
-		__uint16_t	size;
-	} freemap[XFS_ATTR_LEAF_MAPSIZE];
-};
-
-/*
- * Flags used in the leaf_entry[i].flags field.
- * NOTE: the INCOMPLETE bit must not collide with the flags bits specified
- * on the system call, they are "or"ed together for various operations.
- */
-#define	XFS_ATTR_LOCAL_BIT	0	/* attr is stored locally */
-#define	XFS_ATTR_ROOT_BIT	1	/* limit access to trusted attrs */
-#define	XFS_ATTR_SECURE_BIT	2	/* limit access to secure attrs */
-#define	XFS_ATTR_INCOMPLETE_BIT	7	/* attr in middle of create/delete */
-#define XFS_ATTR_LOCAL		(1 << XFS_ATTR_LOCAL_BIT)
-#define XFS_ATTR_ROOT		(1 << XFS_ATTR_ROOT_BIT)
-#define XFS_ATTR_SECURE		(1 << XFS_ATTR_SECURE_BIT)
-#define XFS_ATTR_INCOMPLETE	(1 << XFS_ATTR_INCOMPLETE_BIT)
-
-/*
- * Conversion macros for converting namespace bits from argument flags
- * to ondisk flags.
- */
-#define XFS_ATTR_NSP_ARGS_MASK		(ATTR_ROOT | ATTR_SECURE)
-#define XFS_ATTR_NSP_ONDISK_MASK	(XFS_ATTR_ROOT | XFS_ATTR_SECURE)
-#define XFS_ATTR_NSP_ONDISK(flags)	((flags) & XFS_ATTR_NSP_ONDISK_MASK)
-#define XFS_ATTR_NSP_ARGS(flags)	((flags) & XFS_ATTR_NSP_ARGS_MASK)
-#define XFS_ATTR_NSP_ARGS_TO_ONDISK(x)	(((x) & ATTR_ROOT ? XFS_ATTR_ROOT : 0) |\
-					 ((x) & ATTR_SECURE ? XFS_ATTR_SECURE : 0))
-#define XFS_ATTR_NSP_ONDISK_TO_ARGS(x)	(((x) & XFS_ATTR_ROOT ? ATTR_ROOT : 0) |\
-					 ((x) & XFS_ATTR_SECURE ? ATTR_SECURE : 0))
-
-/*
- * Alignment for namelist and valuelist entries (since they are mixed
- * there can be only one alignment value)
- */
-#define	XFS_ATTR_LEAF_NAME_ALIGN	((uint)sizeof(xfs_dablk_t))
-
-static inline int
-xfs_attr3_leaf_hdr_size(struct xfs_attr_leafblock *leafp)
-{
-	if (leafp->hdr.info.magic == cpu_to_be16(XFS_ATTR3_LEAF_MAGIC))
-		return sizeof(struct xfs_attr3_leaf_hdr);
-	return sizeof(struct xfs_attr_leaf_hdr);
-}
-
-static inline struct xfs_attr_leaf_entry *
-xfs_attr3_leaf_entryp(xfs_attr_leafblock_t *leafp)
-{
-	if (leafp->hdr.info.magic == cpu_to_be16(XFS_ATTR3_LEAF_MAGIC))
-		return &((struct xfs_attr3_leafblock *)leafp)->entries[0];
-	return &leafp->entries[0];
-}
-
-/*
- * Cast typed pointers for "local" and "remote" name/value structs.
- */
-static inline char *
-xfs_attr3_leaf_name(xfs_attr_leafblock_t *leafp, int idx)
-{
-	struct xfs_attr_leaf_entry *entries = xfs_attr3_leaf_entryp(leafp);
-
-	return &((char *)leafp)[be16_to_cpu(entries[idx].nameidx)];
-}
-
-static inline xfs_attr_leaf_name_remote_t *
-xfs_attr3_leaf_name_remote(xfs_attr_leafblock_t *leafp, int idx)
-{
-	return (xfs_attr_leaf_name_remote_t *)xfs_attr3_leaf_name(leafp, idx);
-}
-
-static inline xfs_attr_leaf_name_local_t *
-xfs_attr3_leaf_name_local(xfs_attr_leafblock_t *leafp, int idx)
-{
-	return (xfs_attr_leaf_name_local_t *)xfs_attr3_leaf_name(leafp, idx);
-}
-
-/*
- * Calculate total bytes used (including trailing pad for alignment) for
- * a "local" name/value structure, a "remote" name/value structure, and
- * a pointer which might be either.
- */
-static inline int xfs_attr_leaf_entsize_remote(int nlen)
-{
-	return ((uint)sizeof(xfs_attr_leaf_name_remote_t) - 1 + (nlen) + \
-		XFS_ATTR_LEAF_NAME_ALIGN - 1) & ~(XFS_ATTR_LEAF_NAME_ALIGN - 1);
-}
-
-static inline int xfs_attr_leaf_entsize_local(int nlen, int vlen)
-{
-	return ((uint)sizeof(xfs_attr_leaf_name_local_t) - 1 + (nlen) + (vlen) +
-		XFS_ATTR_LEAF_NAME_ALIGN - 1) & ~(XFS_ATTR_LEAF_NAME_ALIGN - 1);
-}
-
-static inline int xfs_attr_leaf_entsize_local_max(int bsize)
-{
-	return (((bsize) >> 1) + ((bsize) >> 2));
-}
-
 /*
  * Used to keep a list of "remote value" extents when unlinking an inode.
  */
@@ -336,6 +106,4 @@ void	xfs_attr3_leaf_hdr_from_disk(struct xfs_attr3_icleaf_hdr *to,
 void	xfs_attr3_leaf_hdr_to_disk(struct xfs_attr_leafblock *to,
 				   struct xfs_attr3_icleaf_hdr *from);
 
-extern const struct xfs_buf_ops xfs_attr3_leaf_buf_ops;
-
 #endif	/* __XFS_ATTR_LEAF_H__ */
diff --git a/fs/xfs/xfs_attr_list.c b/fs/xfs/xfs_attr_list.c
index cbc80d485177..2d174b128153 100644
--- a/fs/xfs/xfs_attr_list.c
+++ b/fs/xfs/xfs_attr_list.c
@@ -18,31 +18,29 @@
  */
 #include "xfs.h"
 #include "xfs_fs.h"
-#include "xfs_types.h"
+#include "xfs_format.h"
+#include "xfs_log_format.h"
+#include "xfs_trans_resv.h"
 #include "xfs_bit.h"
-#include "xfs_log.h"
-#include "xfs_trans.h"
 #include "xfs_sb.h"
 #include "xfs_ag.h"
 #include "xfs_mount.h"
+#include "xfs_da_format.h"
 #include "xfs_da_btree.h"
-#include "xfs_bmap_btree.h"
-#include "xfs_alloc_btree.h"
-#include "xfs_ialloc_btree.h"
-#include "xfs_alloc.h"
-#include "xfs_btree.h"
-#include "xfs_attr_sf.h"
-#include "xfs_attr_remote.h"
-#include "xfs_dinode.h"
 #include "xfs_inode.h"
+#include "xfs_trans.h"
 #include "xfs_inode_item.h"
 #include "xfs_bmap.h"
 #include "xfs_attr.h"
+#include "xfs_attr_sf.h"
+#include "xfs_attr_remote.h"
 #include "xfs_attr_leaf.h"
 #include "xfs_error.h"
 #include "xfs_trace.h"
 #include "xfs_buf_item.h"
 #include "xfs_cksum.h"
+#include "xfs_dinode.h"
+#include "xfs_dir2.h"
 
 STATIC int
 xfs_attr_shortform_compare(const void *a, const void *b)
@@ -229,6 +227,7 @@ xfs_attr_node_list(xfs_attr_list_context_t *context)
 	struct xfs_da_node_entry *btree;
 	int error, i;
 	struct xfs_buf *bp;
+	struct xfs_inode	*dp = context->dp;
 
 	trace_xfs_attr_node_list(context);
 
@@ -242,7 +241,7 @@ xfs_attr_node_list(xfs_attr_list_context_t *context)
 	 */
 	bp = NULL;
 	if (cursor->blkno > 0) {
-		error = xfs_da3_node_read(NULL, context->dp, cursor->blkno, -1,
+		error = xfs_da3_node_read(NULL, dp, cursor->blkno, -1,
 					      &bp, XFS_ATTR_FORK);
 		if ((error != 0) && (error != EFSCORRUPTED))
 			return(error);
@@ -292,7 +291,7 @@ xfs_attr_node_list(xfs_attr_list_context_t *context)
 		for (;;) {
 			__uint16_t magic;
 
-			error = xfs_da3_node_read(NULL, context->dp,
+			error = xfs_da3_node_read(NULL, dp,
 						      cursor->blkno, -1, &bp,
 						      XFS_ATTR_FORK);
 			if (error)
@@ -312,8 +311,8 @@ xfs_attr_node_list(xfs_attr_list_context_t *context)
 				return XFS_ERROR(EFSCORRUPTED);
 			}
 
-			xfs_da3_node_hdr_from_disk(&nodehdr, node);
-			btree = xfs_da3_node_tree_p(node);
+			dp->d_ops->node_hdr_from_disk(&nodehdr, node);
+			btree = dp->d_ops->node_tree_p(node);
 			for (i = 0; i < nodehdr.count; btree++, i++) {
 				if (cursor->hashval
 						<= be32_to_cpu(btree->hashval)) {
@@ -349,8 +348,7 @@ xfs_attr_node_list(xfs_attr_list_context_t *context)
 			break;
 		cursor->blkno = leafhdr.forw;
 		xfs_trans_brelse(NULL, bp);
-		error = xfs_attr3_leaf_read(NULL, context->dp, cursor->blkno, -1,
-					   &bp);
+		error = xfs_attr3_leaf_read(NULL, dp, cursor->blkno, -1, &bp);
 		if (error)
 			return error;
 	}
diff --git a/fs/xfs/xfs_attr_remote.c b/fs/xfs/xfs_attr_remote.c
index 712a502de619..739e0a52deda 100644
--- a/fs/xfs/xfs_attr_remote.c
+++ b/fs/xfs/xfs_attr_remote.c
@@ -18,20 +18,19 @@
  */
 #include "xfs.h"
 #include "xfs_fs.h"
-#include "xfs_types.h"
+#include "xfs_shared.h"
+#include "xfs_format.h"
+#include "xfs_log_format.h"
+#include "xfs_trans_resv.h"
 #include "xfs_bit.h"
-#include "xfs_log.h"
-#include "xfs_trans.h"
-#include "xfs_trans_priv.h"
 #include "xfs_sb.h"
 #include "xfs_ag.h"
 #include "xfs_mount.h"
-#include "xfs_error.h"
+#include "xfs_da_format.h"
 #include "xfs_da_btree.h"
-#include "xfs_bmap_btree.h"
-#include "xfs_dinode.h"
 #include "xfs_inode.h"
 #include "xfs_alloc.h"
+#include "xfs_trans.h"
 #include "xfs_inode_item.h"
 #include "xfs_bmap.h"
 #include "xfs_bmap_util.h"
@@ -42,6 +41,7 @@
 #include "xfs_trace.h"
 #include "xfs_cksum.h"
 #include "xfs_buf_item.h"
+#include "xfs_error.h"
 
 #define ATTR_RMTVALUE_MAPSIZE	1	/* # of map entries at once */
 
diff --git a/fs/xfs/xfs_attr_remote.h b/fs/xfs/xfs_attr_remote.h
index 92a8fd7977cc..5a9acfa156d7 100644
--- a/fs/xfs/xfs_attr_remote.h
+++ b/fs/xfs/xfs_attr_remote.h
@@ -18,35 +18,6 @@
 #ifndef __XFS_ATTR_REMOTE_H__
 #define	__XFS_ATTR_REMOTE_H__
 
-#define XFS_ATTR3_RMT_MAGIC	0x5841524d	/* XARM */
-
-/*
- * There is one of these headers per filesystem block in a remote attribute.
- * This is done to ensure there is a 1:1 mapping between the attribute value
- * length and the number of blocks needed to store the attribute. This makes the
- * verification of a buffer a little more complex, but greatly simplifies the
- * allocation, reading and writing of these attributes as we don't have to guess
- * the number of blocks needed to store the attribute data.
- */
-struct xfs_attr3_rmt_hdr {
-	__be32	rm_magic;
-	__be32	rm_offset;
-	__be32	rm_bytes;
-	__be32	rm_crc;
-	uuid_t	rm_uuid;
-	__be64	rm_owner;
-	__be64	rm_blkno;
-	__be64	rm_lsn;
-};
-
-#define XFS_ATTR3_RMT_CRC_OFF	offsetof(struct xfs_attr3_rmt_hdr, rm_crc)
-
-#define XFS_ATTR3_RMT_BUF_SPACE(mp, bufsize)	\
-	((bufsize) - (xfs_sb_version_hascrc(&(mp)->m_sb) ? \
-			sizeof(struct xfs_attr3_rmt_hdr) : 0))
-
-extern const struct xfs_buf_ops xfs_attr3_rmt_buf_ops;
-
 int xfs_attr3_rmt_blocks(struct xfs_mount *mp, int attrlen);
 
 int xfs_attr_rmtval_get(struct xfs_da_args *args);
diff --git a/fs/xfs/xfs_bit.c b/fs/xfs/xfs_bit.c
index 48228848f5ae..0e8885a59646 100644
--- a/fs/xfs/xfs_bit.c
+++ b/fs/xfs/xfs_bit.c
@@ -16,10 +16,8 @@
  * Inc.,  51 Franklin St, Fifth Floor, Boston, MA  02110-1301  USA
  */
 #include "xfs.h"
+#include "xfs_log_format.h"
 #include "xfs_bit.h"
-#include "xfs_log.h"
-#include "xfs_trans.h"
-#include "xfs_buf_item.h"
 
 /*
  * XFS bit manipulation routines, used in non-realtime code.
diff --git a/fs/xfs/xfs_bmap.c b/fs/xfs/xfs_bmap.c
index f47e65c30be6..1c02da8bb7df 100644
--- a/fs/xfs/xfs_bmap.c
+++ b/fs/xfs/xfs_bmap.c
@@ -17,39 +17,37 @@
  */
 #include "xfs.h"
 #include "xfs_fs.h"
+#include "xfs_shared.h"
 #include "xfs_format.h"
+#include "xfs_log_format.h"
+#include "xfs_trans_resv.h"
 #include "xfs_bit.h"
-#include "xfs_log.h"
 #include "xfs_inum.h"
-#include "xfs_trans.h"
 #include "xfs_sb.h"
 #include "xfs_ag.h"
 #include "xfs_mount.h"
+#include "xfs_da_format.h"
 #include "xfs_da_btree.h"
-#include "xfs_dir2_format.h"
 #include "xfs_dir2.h"
-#include "xfs_bmap_btree.h"
-#include "xfs_alloc_btree.h"
-#include "xfs_ialloc_btree.h"
-#include "xfs_dinode.h"
 #include "xfs_inode.h"
 #include "xfs_btree.h"
-#include "xfs_mount.h"
-#include "xfs_itable.h"
+#include "xfs_trans.h"
 #include "xfs_inode_item.h"
 #include "xfs_extfree_item.h"
 #include "xfs_alloc.h"
 #include "xfs_bmap.h"
 #include "xfs_bmap_util.h"
+#include "xfs_bmap_btree.h"
 #include "xfs_rtalloc.h"
 #include "xfs_error.h"
-#include "xfs_attr_leaf.h"
 #include "xfs_quota.h"
 #include "xfs_trans_space.h"
 #include "xfs_buf_item.h"
-#include "xfs_filestream.h"
 #include "xfs_trace.h"
 #include "xfs_symlink.h"
+#include "xfs_attr_leaf.h"
+#include "xfs_dinode.h"
+#include "xfs_filestream.h"
 
 
 kmem_zone_t		*xfs_bmap_free_item_zone;
@@ -1482,7 +1480,7 @@ xfs_bmap_search_extents(
 		xfs_alert_tag(ip->i_mount, XFS_PTAG_FSBLOCK_ZERO,
 				"Access to block zero in inode %llu "
 				"start_block: %llx start_off: %llx "
-				"blkcnt: %llx extent-state: %x lastx: %x\n",
+				"blkcnt: %llx extent-state: %x lastx: %x",
 			(unsigned long long)ip->i_ino,
 			(unsigned long long)gotp->br_startblock,
 			(unsigned long long)gotp->br_startoff,
diff --git a/fs/xfs/xfs_bmap_btree.c b/fs/xfs/xfs_bmap_btree.c
index bb8de8e399c4..706bc3f777cb 100644
--- a/fs/xfs/xfs_bmap_btree.c
+++ b/fs/xfs/xfs_bmap_btree.c
@@ -17,27 +17,26 @@
  */
 #include "xfs.h"
 #include "xfs_fs.h"
+#include "xfs_shared.h"
 #include "xfs_format.h"
+#include "xfs_log_format.h"
+#include "xfs_trans_resv.h"
 #include "xfs_bit.h"
-#include "xfs_log.h"
-#include "xfs_trans.h"
 #include "xfs_sb.h"
 #include "xfs_ag.h"
 #include "xfs_mount.h"
-#include "xfs_bmap_btree.h"
-#include "xfs_alloc_btree.h"
-#include "xfs_ialloc_btree.h"
-#include "xfs_dinode.h"
 #include "xfs_inode.h"
+#include "xfs_trans.h"
 #include "xfs_inode_item.h"
 #include "xfs_alloc.h"
 #include "xfs_btree.h"
-#include "xfs_itable.h"
+#include "xfs_bmap_btree.h"
 #include "xfs_bmap.h"
 #include "xfs_error.h"
 #include "xfs_quota.h"
 #include "xfs_trace.h"
 #include "xfs_cksum.h"
+#include "xfs_dinode.h"
 
 /*
  * Determine the extent state.
diff --git a/fs/xfs/xfs_bmap_btree.h b/fs/xfs/xfs_bmap_btree.h
index e367461a638e..6e42e1e50b89 100644
--- a/fs/xfs/xfs_bmap_btree.h
+++ b/fs/xfs/xfs_bmap_btree.h
@@ -18,9 +18,6 @@
 #ifndef __XFS_BMAP_BTREE_H__
 #define __XFS_BMAP_BTREE_H__
 
-#define XFS_BMAP_MAGIC		0x424d4150	/* 'BMAP' */
-#define XFS_BMAP_CRC_MAGIC	0x424d4133	/* 'BMA3' */
-
 struct xfs_btree_cur;
 struct xfs_btree_block;
 struct xfs_mount;
@@ -28,85 +25,6 @@ struct xfs_inode;
 struct xfs_trans;
 
 /*
- * Bmap root header, on-disk form only.
- */
-typedef struct xfs_bmdr_block {
-	__be16		bb_level;	/* 0 is a leaf */
-	__be16		bb_numrecs;	/* current # of data records */
-} xfs_bmdr_block_t;
-
-/*
- * Bmap btree record and extent descriptor.
- *  l0:63 is an extent flag (value 1 indicates non-normal).
- *  l0:9-62 are startoff.
- *  l0:0-8 and l1:21-63 are startblock.
- *  l1:0-20 are blockcount.
- */
-#define BMBT_EXNTFLAG_BITLEN	1
-#define BMBT_STARTOFF_BITLEN	54
-#define BMBT_STARTBLOCK_BITLEN	52
-#define BMBT_BLOCKCOUNT_BITLEN	21
-
-typedef struct xfs_bmbt_rec {
-	__be64			l0, l1;
-} xfs_bmbt_rec_t;
-
-typedef __uint64_t	xfs_bmbt_rec_base_t;	/* use this for casts */
-typedef xfs_bmbt_rec_t xfs_bmdr_rec_t;
-
-typedef struct xfs_bmbt_rec_host {
-	__uint64_t		l0, l1;
-} xfs_bmbt_rec_host_t;
-
-/*
- * Values and macros for delayed-allocation startblock fields.
- */
-#define STARTBLOCKVALBITS	17
-#define STARTBLOCKMASKBITS	(15 + XFS_BIG_BLKNOS * 20)
-#define DSTARTBLOCKMASKBITS	(15 + 20)
-#define STARTBLOCKMASK		\
-	(((((xfs_fsblock_t)1) << STARTBLOCKMASKBITS) - 1) << STARTBLOCKVALBITS)
-#define DSTARTBLOCKMASK		\
-	(((((xfs_dfsbno_t)1) << DSTARTBLOCKMASKBITS) - 1) << STARTBLOCKVALBITS)
-
-static inline int isnullstartblock(xfs_fsblock_t x)
-{
-	return ((x) & STARTBLOCKMASK) == STARTBLOCKMASK;
-}
-
-static inline int isnulldstartblock(xfs_dfsbno_t x)
-{
-	return ((x) & DSTARTBLOCKMASK) == DSTARTBLOCKMASK;
-}
-
-static inline xfs_fsblock_t nullstartblock(int k)
-{
-	ASSERT(k < (1 << STARTBLOCKVALBITS));
-	return STARTBLOCKMASK | (k);
-}
-
-static inline xfs_filblks_t startblockval(xfs_fsblock_t x)
-{
-	return (xfs_filblks_t)((x) & ~STARTBLOCKMASK);
-}
-
-/*
- * Possible extent formats.
- */
-typedef enum {
-	XFS_EXTFMT_NOSTATE = 0,
-	XFS_EXTFMT_HASSTATE
-} xfs_exntfmt_t;
-
-/*
- * Possible extent states.
- */
-typedef enum {
-	XFS_EXT_NORM, XFS_EXT_UNWRITTEN,
-	XFS_EXT_DMAPI_OFFLINE, XFS_EXT_INVALID
-} xfs_exntst_t;
-
-/*
  * Extent state and extent format macros.
  */
 #define XFS_EXTFMT_INODE(x)	\
@@ -115,27 +33,6 @@ typedef enum {
 #define ISUNWRITTEN(x)	((x)->br_state == XFS_EXT_UNWRITTEN)
 
 /*
- * Incore version of above.
- */
-typedef struct xfs_bmbt_irec
-{
-	xfs_fileoff_t	br_startoff;	/* starting file offset */
-	xfs_fsblock_t	br_startblock;	/* starting block number */
-	xfs_filblks_t	br_blockcount;	/* number of blocks */
-	xfs_exntst_t	br_state;	/* extent state */
-} xfs_bmbt_irec_t;
-
-/*
- * Key structure for non-leaf levels of the tree.
- */
-typedef struct xfs_bmbt_key {
-	__be64		br_startoff;	/* starting file offset */
-} xfs_bmbt_key_t, xfs_bmdr_key_t;
-
-/* btree pointer type */
-typedef __be64 xfs_bmbt_ptr_t, xfs_bmdr_ptr_t;
-
-/*
  * Btree block header size depends on a superblock flag.
  */
 #define XFS_BMBT_BLOCK_LEN(mp) \
@@ -243,6 +140,4 @@ extern int xfs_bmbt_change_owner(struct xfs_trans *tp, struct xfs_inode *ip,
 extern struct xfs_btree_cur *xfs_bmbt_init_cursor(struct xfs_mount *,
 		struct xfs_trans *, struct xfs_inode *, int);
 
-extern const struct xfs_buf_ops xfs_bmbt_buf_ops;
-
 #endif	/* __XFS_BMAP_BTREE_H__ */
diff --git a/fs/xfs/xfs_bmap_util.c b/fs/xfs/xfs_bmap_util.c
index 97f952caea74..5887e41c0323 100644
--- a/fs/xfs/xfs_bmap_util.c
+++ b/fs/xfs/xfs_bmap_util.c
@@ -18,31 +18,31 @@
  */
 #include "xfs.h"
 #include "xfs_fs.h"
+#include "xfs_shared.h"
 #include "xfs_format.h"
+#include "xfs_log_format.h"
+#include "xfs_trans_resv.h"
 #include "xfs_bit.h"
-#include "xfs_log.h"
-#include "xfs_inum.h"
-#include "xfs_trans.h"
 #include "xfs_sb.h"
 #include "xfs_ag.h"
 #include "xfs_mount.h"
-#include "xfs_da_btree.h"
-#include "xfs_bmap_btree.h"
-#include "xfs_alloc_btree.h"
-#include "xfs_ialloc_btree.h"
-#include "xfs_dinode.h"
+#include "xfs_da_format.h"
 #include "xfs_inode.h"
 #include "xfs_btree.h"
+#include "xfs_trans.h"
 #include "xfs_extfree_item.h"
 #include "xfs_alloc.h"
 #include "xfs_bmap.h"
 #include "xfs_bmap_util.h"
+#include "xfs_bmap_btree.h"
 #include "xfs_rtalloc.h"
 #include "xfs_error.h"
 #include "xfs_quota.h"
 #include "xfs_trans_space.h"
 #include "xfs_trace.h"
 #include "xfs_icache.h"
+#include "xfs_log.h"
+#include "xfs_dinode.h"
 
 /* Kernel only BMAP related definitions and functions */
 
@@ -965,32 +965,12 @@ xfs_free_eofblocks(
 	return error;
 }
 
-/*
- * xfs_alloc_file_space()
- *      This routine allocates disk space for the given file.
- *
- *	If alloc_type == 0, this request is for an ALLOCSP type
- *	request which will change the file size.  In this case, no
- *	DMAPI event will be generated by the call.  A TRUNCATE event
- *	will be generated later by xfs_setattr.
- *
- *	If alloc_type != 0, this request is for a RESVSP type
- *	request, and a DMAPI DM_EVENT_WRITE will be generated if the
- *	lower block boundary byte address is less than the file's
- *	length.
- *
- * RETURNS:
- *       0 on success
- *      errno on error
- *
- */
-STATIC int
+int
 xfs_alloc_file_space(
-	xfs_inode_t		*ip,
+	struct xfs_inode	*ip,
 	xfs_off_t		offset,
 	xfs_off_t		len,
-	int			alloc_type,
-	int			attr_flags)
+	int			alloc_type)
 {
 	xfs_mount_t		*mp = ip->i_mount;
 	xfs_off_t		count;
@@ -1232,24 +1212,11 @@ xfs_zero_remaining_bytes(
 	return error;
 }
 
-/*
- * xfs_free_file_space()
- *      This routine frees disk space for the given file.
- *
- *	This routine is only called by xfs_change_file_space
- *	for an UNRESVSP type call.
- *
- * RETURNS:
- *       0 on success
- *      errno on error
- *
- */
-STATIC int
+int
 xfs_free_file_space(
-	xfs_inode_t		*ip,
+	struct xfs_inode	*ip,
 	xfs_off_t		offset,
-	xfs_off_t		len,
-	int			attr_flags)
+	xfs_off_t		len)
 {
 	int			committed;
 	int			done;
@@ -1267,7 +1234,6 @@ xfs_free_file_space(
 	int			rt;
 	xfs_fileoff_t		startoffset_fsb;
 	xfs_trans_t		*tp;
-	int			need_iolock = 1;
 
 	mp = ip->i_mount;
 
@@ -1284,20 +1250,15 @@ xfs_free_file_space(
 	startoffset_fsb	= XFS_B_TO_FSB(mp, offset);
 	endoffset_fsb = XFS_B_TO_FSBT(mp, offset + len);
 
-	if (attr_flags & XFS_ATTR_NOLOCK)
-		need_iolock = 0;
-	if (need_iolock) {
-		xfs_ilock(ip, XFS_IOLOCK_EXCL);
-		/* wait for the completion of any pending DIOs */
-		inode_dio_wait(VFS_I(ip));
-	}
+	/* wait for the completion of any pending DIOs */
+	inode_dio_wait(VFS_I(ip));
 
 	rounding = max_t(xfs_off_t, 1 << mp->m_sb.sb_blocklog, PAGE_CACHE_SIZE);
 	ioffset = offset & ~(rounding - 1);
 	error = -filemap_write_and_wait_range(VFS_I(ip)->i_mapping,
 					      ioffset, -1);
 	if (error)
-		goto out_unlock_iolock;
+		goto out;
 	truncate_pagecache_range(VFS_I(ip), ioffset, -1);
 
 	/*
@@ -1311,7 +1272,7 @@ xfs_free_file_space(
 		error = xfs_bmapi_read(ip, startoffset_fsb, 1,
 					&imap, &nimap, 0);
 		if (error)
-			goto out_unlock_iolock;
+			goto out;
 		ASSERT(nimap == 0 || nimap == 1);
 		if (nimap && imap.br_startblock != HOLESTARTBLOCK) {
 			xfs_daddr_t	block;
@@ -1326,7 +1287,7 @@ xfs_free_file_space(
 		error = xfs_bmapi_read(ip, endoffset_fsb - 1, 1,
 					&imap, &nimap, 0);
 		if (error)
-			goto out_unlock_iolock;
+			goto out;
 		ASSERT(nimap == 0 || nimap == 1);
 		if (nimap && imap.br_startblock != HOLESTARTBLOCK) {
 			ASSERT(imap.br_startblock != DELAYSTARTBLOCK);
@@ -1412,27 +1373,23 @@ xfs_free_file_space(
 		xfs_iunlock(ip, XFS_ILOCK_EXCL);
 	}
 
- out_unlock_iolock:
-	if (need_iolock)
-		xfs_iunlock(ip, XFS_IOLOCK_EXCL);
+ out:
 	return error;
 
  error0:
 	xfs_bmap_cancel(&free_list);
  error1:
 	xfs_trans_cancel(tp, XFS_TRANS_RELEASE_LOG_RES | XFS_TRANS_ABORT);
-	xfs_iunlock(ip, need_iolock ? (XFS_ILOCK_EXCL | XFS_IOLOCK_EXCL) :
-		    XFS_ILOCK_EXCL);
-	return error;
+	xfs_iunlock(ip, XFS_ILOCK_EXCL);
+	goto out;
 }
 
 
-STATIC int
+int
 xfs_zero_file_space(
 	struct xfs_inode	*ip,
 	xfs_off_t		offset,
-	xfs_off_t		len,
-	int			attr_flags)
+	xfs_off_t		len)
 {
 	struct xfs_mount	*mp = ip->i_mount;
 	uint			granularity;
@@ -1453,9 +1410,6 @@ xfs_zero_file_space(
 	ASSERT(start_boundary >= offset);
 	ASSERT(end_boundary <= offset + len);
 
-	if (!(attr_flags & XFS_ATTR_NOLOCK))
-		xfs_ilock(ip, XFS_IOLOCK_EXCL);
-
 	if (start_boundary < end_boundary - 1) {
 		/* punch out the page cache over the conversion range */
 		truncate_pagecache_range(VFS_I(ip), start_boundary,
@@ -1463,16 +1417,16 @@ xfs_zero_file_space(
 		/* convert the blocks */
 		error = xfs_alloc_file_space(ip, start_boundary,
 					end_boundary - start_boundary - 1,
-					XFS_BMAPI_PREALLOC | XFS_BMAPI_CONVERT,
-					attr_flags);
+					XFS_BMAPI_PREALLOC | XFS_BMAPI_CONVERT);
 		if (error)
-			goto out_unlock;
+			goto out;
 
 		/* We've handled the interior of the range, now for the edges */
-		if (start_boundary != offset)
+		if (start_boundary != offset) {
 			error = xfs_iozero(ip, offset, start_boundary - offset);
-		if (error)
-			goto out_unlock;
+			if (error)
+				goto out;
+		}
 
 		if (end_boundary != offset + len)
 			error = xfs_iozero(ip, end_boundary,
@@ -1486,197 +1440,12 @@ xfs_zero_file_space(
 		error = xfs_iozero(ip, offset, len);
 	}
 
-out_unlock:
-	if (!(attr_flags & XFS_ATTR_NOLOCK))
-		xfs_iunlock(ip, XFS_IOLOCK_EXCL);
+out:
 	return error;
 
 }
 
 /*
- * xfs_change_file_space()
- *      This routine allocates or frees disk space for the given file.
- *      The user specified parameters are checked for alignment and size
- *      limitations.
- *
- * RETURNS:
- *       0 on success
- *      errno on error
- *
- */
-int
-xfs_change_file_space(
-	xfs_inode_t	*ip,
-	int		cmd,
-	xfs_flock64_t	*bf,
-	xfs_off_t	offset,
-	int		attr_flags)
-{
-	xfs_mount_t	*mp = ip->i_mount;
-	int		clrprealloc;
-	int		error;
-	xfs_fsize_t	fsize;
-	int		setprealloc;
-	xfs_off_t	startoffset;
-	xfs_trans_t	*tp;
-	struct iattr	iattr;
-
-	if (!S_ISREG(ip->i_d.di_mode))
-		return XFS_ERROR(EINVAL);
-
-	switch (bf->l_whence) {
-	case 0: /*SEEK_SET*/
-		break;
-	case 1: /*SEEK_CUR*/
-		bf->l_start += offset;
-		break;
-	case 2: /*SEEK_END*/
-		bf->l_start += XFS_ISIZE(ip);
-		break;
-	default:
-		return XFS_ERROR(EINVAL);
-	}
-
-	/*
-	 * length of <= 0 for resv/unresv/zero is invalid.  length for
-	 * alloc/free is ignored completely and we have no idea what userspace
-	 * might have set it to, so set it to zero to allow range
-	 * checks to pass.
-	 */
-	switch (cmd) {
-	case XFS_IOC_ZERO_RANGE:
-	case XFS_IOC_RESVSP:
-	case XFS_IOC_RESVSP64:
-	case XFS_IOC_UNRESVSP:
-	case XFS_IOC_UNRESVSP64:
-		if (bf->l_len <= 0)
-			return XFS_ERROR(EINVAL);
-		break;
-	default:
-		bf->l_len = 0;
-		break;
-	}
-
-	if (bf->l_start < 0 ||
-	    bf->l_start > mp->m_super->s_maxbytes ||
-	    bf->l_start + bf->l_len < 0 ||
-	    bf->l_start + bf->l_len >= mp->m_super->s_maxbytes)
-		return XFS_ERROR(EINVAL);
-
-	bf->l_whence = 0;
-
-	startoffset = bf->l_start;
-	fsize = XFS_ISIZE(ip);
-
-	setprealloc = clrprealloc = 0;
-	switch (cmd) {
-	case XFS_IOC_ZERO_RANGE:
-		error = xfs_zero_file_space(ip, startoffset, bf->l_len,
-						attr_flags);
-		if (error)
-			return error;
-		setprealloc = 1;
-		break;
-
-	case XFS_IOC_RESVSP:
-	case XFS_IOC_RESVSP64:
-		error = xfs_alloc_file_space(ip, startoffset, bf->l_len,
-						XFS_BMAPI_PREALLOC, attr_flags);
-		if (error)
-			return error;
-		setprealloc = 1;
-		break;
-
-	case XFS_IOC_UNRESVSP:
-	case XFS_IOC_UNRESVSP64:
-		if ((error = xfs_free_file_space(ip, startoffset, bf->l_len,
-								attr_flags)))
-			return error;
-		break;
-
-	case XFS_IOC_ALLOCSP:
-	case XFS_IOC_ALLOCSP64:
-	case XFS_IOC_FREESP:
-	case XFS_IOC_FREESP64:
-		/*
-		 * These operations actually do IO when extending the file, but
-		 * the allocation is done seperately to the zeroing that is
-		 * done. This set of operations need to be serialised against
-		 * other IO operations, such as truncate and buffered IO. We
-		 * need to take the IOLOCK here to serialise the allocation and
-		 * zeroing IO to prevent other IOLOCK holders (e.g. getbmap,
-		 * truncate, direct IO) from racing against the transient
-		 * allocated but not written state we can have here.
-		 */
-		xfs_ilock(ip, XFS_IOLOCK_EXCL);
-		if (startoffset > fsize) {
-			error = xfs_alloc_file_space(ip, fsize,
-					startoffset - fsize, 0,
-					attr_flags | XFS_ATTR_NOLOCK);
-			if (error) {
-				xfs_iunlock(ip, XFS_IOLOCK_EXCL);
-				break;
-			}
-		}
-
-		iattr.ia_valid = ATTR_SIZE;
-		iattr.ia_size = startoffset;
-
-		error = xfs_setattr_size(ip, &iattr,
-					 attr_flags | XFS_ATTR_NOLOCK);
-		xfs_iunlock(ip, XFS_IOLOCK_EXCL);
-
-		if (error)
-			return error;
-
-		clrprealloc = 1;
-		break;
-
-	default:
-		ASSERT(0);
-		return XFS_ERROR(EINVAL);
-	}
-
-	/*
-	 * update the inode timestamp, mode, and prealloc flag bits
-	 */
-	tp = xfs_trans_alloc(mp, XFS_TRANS_WRITEID);
-	error = xfs_trans_reserve(tp, &M_RES(mp)->tr_writeid, 0, 0);
-	if (error) {
-		xfs_trans_cancel(tp, 0);
-		return error;
-	}
-
-	xfs_ilock(ip, XFS_ILOCK_EXCL);
-	xfs_trans_ijoin(tp, ip, XFS_ILOCK_EXCL);
-
-	if ((attr_flags & XFS_ATTR_DMI) == 0) {
-		ip->i_d.di_mode &= ~S_ISUID;
-
-		/*
-		 * Note that we don't have to worry about mandatory
-		 * file locking being disabled here because we only
-		 * clear the S_ISGID bit if the Group execute bit is
-		 * on, but if it was on then mandatory locking wouldn't
-		 * have been enabled.
-		 */
-		if (ip->i_d.di_mode & S_IXGRP)
-			ip->i_d.di_mode &= ~S_ISGID;
-
-		xfs_trans_ichgtime(tp, ip, XFS_ICHGTIME_MOD | XFS_ICHGTIME_CHG);
-	}
-	if (setprealloc)
-		ip->i_d.di_flags |= XFS_DIFLAG_PREALLOC;
-	else if (clrprealloc)
-		ip->i_d.di_flags &= ~XFS_DIFLAG_PREALLOC;
-
-	xfs_trans_log_inode(tp, ip, XFS_ILOG_CORE);
-	if (attr_flags & XFS_ATTR_SYNC)
-		xfs_trans_set_sync(tp);
-	return xfs_trans_commit(tp, 0);
-}
-
-/*
  * We need to check that the format of the data fork in the temporary inode is
  * valid for the target inode before doing the swap. This is not a problem with
  * attr1 because of the fixed fork offset, but attr2 has a dynamically sized
diff --git a/fs/xfs/xfs_bmap_util.h b/fs/xfs/xfs_bmap_util.h
index 061260946f7a..900747b25772 100644
--- a/fs/xfs/xfs_bmap_util.h
+++ b/fs/xfs/xfs_bmap_util.h
@@ -93,9 +93,12 @@ int	xfs_bmap_last_extent(struct xfs_trans *tp, struct xfs_inode *ip,
 			     int *is_empty);
 
 /* preallocation and hole punch interface */
-int	xfs_change_file_space(struct xfs_inode *ip, int cmd,
-			      xfs_flock64_t *bf, xfs_off_t offset,
-			      int attr_flags);
+int	xfs_alloc_file_space(struct xfs_inode *ip, xfs_off_t offset,
+			     xfs_off_t len, int alloc_type);
+int	xfs_free_file_space(struct xfs_inode *ip, xfs_off_t offset,
+			    xfs_off_t len);
+int	xfs_zero_file_space(struct xfs_inode *ip, xfs_off_t offset,
+			    xfs_off_t len);
 
 /* EOF block manipulation functions */
 bool	xfs_can_free_eofblocks(struct xfs_inode *ip, bool force);
diff --git a/fs/xfs/xfs_btree.c b/fs/xfs/xfs_btree.c
index 5690e102243d..9adaae4f3e2f 100644
--- a/fs/xfs/xfs_btree.c
+++ b/fs/xfs/xfs_btree.c
@@ -17,18 +17,16 @@
  */
 #include "xfs.h"
 #include "xfs_fs.h"
-#include "xfs_types.h"
+#include "xfs_shared.h"
+#include "xfs_format.h"
+#include "xfs_log_format.h"
+#include "xfs_trans_resv.h"
 #include "xfs_bit.h"
-#include "xfs_log.h"
-#include "xfs_trans.h"
 #include "xfs_sb.h"
 #include "xfs_ag.h"
 #include "xfs_mount.h"
-#include "xfs_bmap_btree.h"
-#include "xfs_alloc_btree.h"
-#include "xfs_ialloc_btree.h"
-#include "xfs_dinode.h"
 #include "xfs_inode.h"
+#include "xfs_trans.h"
 #include "xfs_inode_item.h"
 #include "xfs_buf_item.h"
 #include "xfs_btree.h"
diff --git a/fs/xfs/xfs_btree.h b/fs/xfs/xfs_btree.h
index 06729b67ad58..91e34f21bace 100644
--- a/fs/xfs/xfs_btree.h
+++ b/fs/xfs/xfs_btree.h
@@ -27,73 +27,6 @@ struct xfs_trans;
 extern kmem_zone_t	*xfs_btree_cur_zone;
 
 /*
- * This nonsense is to make -wlint happy.
- */
-#define	XFS_LOOKUP_EQ	((xfs_lookup_t)XFS_LOOKUP_EQi)
-#define	XFS_LOOKUP_LE	((xfs_lookup_t)XFS_LOOKUP_LEi)
-#define	XFS_LOOKUP_GE	((xfs_lookup_t)XFS_LOOKUP_GEi)
-
-#define	XFS_BTNUM_BNO	((xfs_btnum_t)XFS_BTNUM_BNOi)
-#define	XFS_BTNUM_CNT	((xfs_btnum_t)XFS_BTNUM_CNTi)
-#define	XFS_BTNUM_BMAP	((xfs_btnum_t)XFS_BTNUM_BMAPi)
-#define	XFS_BTNUM_INO	((xfs_btnum_t)XFS_BTNUM_INOi)
-
-/*
- * Generic btree header.
- *
- * This is a combination of the actual format used on disk for short and long
- * format btrees.  The first three fields are shared by both format, but the
- * pointers are different and should be used with care.
- *
- * To get the size of the actual short or long form headers please use the size
- * macros below.  Never use sizeof(xfs_btree_block).
- *
- * The blkno, crc, lsn, owner and uuid fields are only available in filesystems
- * with the crc feature bit, and all accesses to them must be conditional on
- * that flag.
- */
-struct xfs_btree_block {
-	__be32		bb_magic;	/* magic number for block type */
-	__be16		bb_level;	/* 0 is a leaf */
-	__be16		bb_numrecs;	/* current # of data records */
-	union {
-		struct {
-			__be32		bb_leftsib;
-			__be32		bb_rightsib;
-
-			__be64		bb_blkno;
-			__be64		bb_lsn;
-			uuid_t		bb_uuid;
-			__be32		bb_owner;
-			__le32		bb_crc;
-		} s;			/* short form pointers */
-		struct	{
-			__be64		bb_leftsib;
-			__be64		bb_rightsib;
-
-			__be64		bb_blkno;
-			__be64		bb_lsn;
-			uuid_t		bb_uuid;
-			__be64		bb_owner;
-			__le32		bb_crc;
-			__be32		bb_pad; /* padding for alignment */
-		} l;			/* long form pointers */
-	} bb_u;				/* rest */
-};
-
-#define XFS_BTREE_SBLOCK_LEN	16	/* size of a short form block */
-#define XFS_BTREE_LBLOCK_LEN	24	/* size of a long form block */
-
-/* sizes of CRC enabled btree blocks */
-#define XFS_BTREE_SBLOCK_CRC_LEN	(XFS_BTREE_SBLOCK_LEN + 40)
-#define XFS_BTREE_LBLOCK_CRC_LEN	(XFS_BTREE_LBLOCK_LEN + 48)
-
-#define XFS_BTREE_SBLOCK_CRC_OFF \
-	offsetof(struct xfs_btree_block, bb_u.s.bb_crc)
-#define XFS_BTREE_LBLOCK_CRC_OFF \
-	offsetof(struct xfs_btree_block, bb_u.l.bb_crc)
-
-/*
  * Generic key, ptr and record wrapper structures.
  *
  * These are disk format structures, and are converted where necessary
@@ -119,6 +52,18 @@ union xfs_btree_rec {
 };
 
 /*
+ * This nonsense is to make -wlint happy.
+ */
+#define	XFS_LOOKUP_EQ	((xfs_lookup_t)XFS_LOOKUP_EQi)
+#define	XFS_LOOKUP_LE	((xfs_lookup_t)XFS_LOOKUP_LEi)
+#define	XFS_LOOKUP_GE	((xfs_lookup_t)XFS_LOOKUP_GEi)
+
+#define	XFS_BTNUM_BNO	((xfs_btnum_t)XFS_BTNUM_BNOi)
+#define	XFS_BTNUM_CNT	((xfs_btnum_t)XFS_BTNUM_CNTi)
+#define	XFS_BTNUM_BMAP	((xfs_btnum_t)XFS_BTNUM_BMAPi)
+#define	XFS_BTNUM_INO	((xfs_btnum_t)XFS_BTNUM_INOi)
+
+/*
  * For logging record fields.
  */
 #define	XFS_BB_MAGIC		(1 << 0)
diff --git a/fs/xfs/xfs_buf.c b/fs/xfs/xfs_buf.c
index 263470075ea2..c7f0b77dcb00 100644
--- a/fs/xfs/xfs_buf.c
+++ b/fs/xfs/xfs_buf.c
@@ -34,12 +34,13 @@
 #include <linux/backing-dev.h>
 #include <linux/freezer.h>
 
-#include "xfs_sb.h"
+#include "xfs_log_format.h"
 #include "xfs_trans_resv.h"
-#include "xfs_log.h"
+#include "xfs_sb.h"
 #include "xfs_ag.h"
 #include "xfs_mount.h"
 #include "xfs_trace.h"
+#include "xfs_log.h"
 
 static kmem_zone_t *xfs_buf_zone;
 
@@ -590,7 +591,7 @@ found:
 		error = _xfs_buf_map_pages(bp, flags);
 		if (unlikely(error)) {
 			xfs_warn(target->bt_mount,
-				"%s: failed to map pages\n", __func__);
+				"%s: failed to map pagesn", __func__);
 			xfs_buf_relse(bp);
 			return NULL;
 		}
@@ -809,7 +810,7 @@ xfs_buf_get_uncached(
 	error = _xfs_buf_map_pages(bp, 0);
 	if (unlikely(error)) {
 		xfs_warn(target->bt_mount,
-			"%s: failed to map pages\n", __func__);
+			"%s: failed to map pages", __func__);
 		goto fail_free_mem;
 	}
 
@@ -1618,7 +1619,7 @@ xfs_setsize_buftarg_flags(
 		bdevname(btp->bt_bdev, name);
 
 		xfs_warn(btp->bt_mount,
-			"Cannot set_blocksize to %u on device %s\n",
+			"Cannot set_blocksize to %u on device %s",
 			sectorsize, name);
 		return EINVAL;
 	}
diff --git a/fs/xfs/xfs_buf_item.c b/fs/xfs/xfs_buf_item.c
index f1d85cfc0a54..a64f67ba25d3 100644
--- a/fs/xfs/xfs_buf_item.c
+++ b/fs/xfs/xfs_buf_item.c
@@ -17,17 +17,18 @@
  */
 #include "xfs.h"
 #include "xfs_fs.h"
-#include "xfs_types.h"
+#include "xfs_log_format.h"
+#include "xfs_trans_resv.h"
 #include "xfs_bit.h"
-#include "xfs_log.h"
-#include "xfs_trans.h"
 #include "xfs_sb.h"
 #include "xfs_ag.h"
 #include "xfs_mount.h"
+#include "xfs_trans.h"
 #include "xfs_buf_item.h"
 #include "xfs_trans_priv.h"
 #include "xfs_error.h"
 #include "xfs_trace.h"
+#include "xfs_log.h"
 
 
 kmem_zone_t	*xfs_buf_item_zone;
@@ -808,7 +809,7 @@ xfs_buf_item_init(
  * Mark bytes first through last inclusive as dirty in the buf
  * item's bitmap.
  */
-void
+static void
 xfs_buf_item_log_segment(
 	struct xfs_buf_log_item	*bip,
 	uint			first,
diff --git a/fs/xfs/xfs_buf_item.h b/fs/xfs/xfs_buf_item.h
index db6371087fe8..3f3455a41510 100644
--- a/fs/xfs/xfs_buf_item.h
+++ b/fs/xfs/xfs_buf_item.h
@@ -71,10 +71,6 @@ void	xfs_buf_attach_iodone(struct xfs_buf *,
 void	xfs_buf_iodone_callbacks(struct xfs_buf *);
 void	xfs_buf_iodone(struct xfs_buf *, struct xfs_log_item *);
 
-void	xfs_trans_buf_set_type(struct xfs_trans *, struct xfs_buf *,
-			       enum xfs_blft);
-void	xfs_trans_buf_copy_type(struct xfs_buf *dst_bp, struct xfs_buf *src_bp);
-
 extern kmem_zone_t	*xfs_buf_item_zone;
 
 #endif	/* __XFS_BUF_ITEM_H__ */
diff --git a/fs/xfs/xfs_da_btree.c b/fs/xfs/xfs_da_btree.c
index 20bf8e8002d6..796272a2e129 100644
--- a/fs/xfs/xfs_da_btree.c
+++ b/fs/xfs/xfs_da_btree.c
@@ -18,20 +18,20 @@
  */
 #include "xfs.h"
 #include "xfs_fs.h"
-#include "xfs_types.h"
+#include "xfs_shared.h"
+#include "xfs_format.h"
+#include "xfs_log_format.h"
+#include "xfs_trans_resv.h"
 #include "xfs_bit.h"
-#include "xfs_log.h"
-#include "xfs_trans.h"
 #include "xfs_sb.h"
 #include "xfs_ag.h"
 #include "xfs_mount.h"
+#include "xfs_da_format.h"
 #include "xfs_da_btree.h"
-#include "xfs_bmap_btree.h"
-#include "xfs_dir2_format.h"
 #include "xfs_dir2.h"
 #include "xfs_dir2_priv.h"
-#include "xfs_dinode.h"
 #include "xfs_inode.h"
+#include "xfs_trans.h"
 #include "xfs_inode_item.h"
 #include "xfs_alloc.h"
 #include "xfs_bmap.h"
@@ -129,56 +129,6 @@ xfs_da_state_free(xfs_da_state_t *state)
 	kmem_zone_free(xfs_da_state_zone, state);
 }
 
-void
-xfs_da3_node_hdr_from_disk(
-	struct xfs_da3_icnode_hdr	*to,
-	struct xfs_da_intnode		*from)
-{
-	ASSERT(from->hdr.info.magic == cpu_to_be16(XFS_DA_NODE_MAGIC) ||
-	       from->hdr.info.magic == cpu_to_be16(XFS_DA3_NODE_MAGIC));
-
-	if (from->hdr.info.magic == cpu_to_be16(XFS_DA3_NODE_MAGIC)) {
-		struct xfs_da3_node_hdr *hdr3 = (struct xfs_da3_node_hdr *)from;
-
-		to->forw = be32_to_cpu(hdr3->info.hdr.forw);
-		to->back = be32_to_cpu(hdr3->info.hdr.back);
-		to->magic = be16_to_cpu(hdr3->info.hdr.magic);
-		to->count = be16_to_cpu(hdr3->__count);
-		to->level = be16_to_cpu(hdr3->__level);
-		return;
-	}
-	to->forw = be32_to_cpu(from->hdr.info.forw);
-	to->back = be32_to_cpu(from->hdr.info.back);
-	to->magic = be16_to_cpu(from->hdr.info.magic);
-	to->count = be16_to_cpu(from->hdr.__count);
-	to->level = be16_to_cpu(from->hdr.__level);
-}
-
-void
-xfs_da3_node_hdr_to_disk(
-	struct xfs_da_intnode		*to,
-	struct xfs_da3_icnode_hdr	*from)
-{
-	ASSERT(from->magic == XFS_DA_NODE_MAGIC ||
-	       from->magic == XFS_DA3_NODE_MAGIC);
-
-	if (from->magic == XFS_DA3_NODE_MAGIC) {
-		struct xfs_da3_node_hdr *hdr3 = (struct xfs_da3_node_hdr *)to;
-
-		hdr3->info.hdr.forw = cpu_to_be32(from->forw);
-		hdr3->info.hdr.back = cpu_to_be32(from->back);
-		hdr3->info.hdr.magic = cpu_to_be16(from->magic);
-		hdr3->__count = cpu_to_be16(from->count);
-		hdr3->__level = cpu_to_be16(from->level);
-		return;
-	}
-	to->hdr.info.forw = cpu_to_be32(from->forw);
-	to->hdr.info.back = cpu_to_be32(from->back);
-	to->hdr.info.magic = cpu_to_be16(from->magic);
-	to->hdr.__count = cpu_to_be16(from->count);
-	to->hdr.__level = cpu_to_be16(from->level);
-}
-
 static bool
 xfs_da3_node_verify(
 	struct xfs_buf		*bp)
@@ -186,8 +136,11 @@ xfs_da3_node_verify(
 	struct xfs_mount	*mp = bp->b_target->bt_mount;
 	struct xfs_da_intnode	*hdr = bp->b_addr;
 	struct xfs_da3_icnode_hdr ichdr;
+	const struct xfs_dir_ops *ops;
 
-	xfs_da3_node_hdr_from_disk(&ichdr, hdr);
+	ops = xfs_dir_get_ops(mp, NULL);
+
+	ops->node_hdr_from_disk(&ichdr, hdr);
 
 	if (xfs_sb_version_hascrc(&mp->m_sb)) {
 		struct xfs_da3_node_hdr *hdr3 = bp->b_addr;
@@ -354,11 +307,12 @@ xfs_da3_node_create(
 	struct xfs_da3_icnode_hdr ichdr = {0};
 	struct xfs_buf		*bp;
 	int			error;
+	struct xfs_inode	*dp = args->dp;
 
 	trace_xfs_da_node_create(args);
 	ASSERT(level <= XFS_DA_NODE_MAXDEPTH);
 
-	error = xfs_da_get_buf(tp, args->dp, blkno, -1, &bp, whichfork);
+	error = xfs_da_get_buf(tp, dp, blkno, -1, &bp, whichfork);
 	if (error)
 		return(error);
 	bp->b_ops = &xfs_da3_node_buf_ops;
@@ -377,9 +331,9 @@ xfs_da3_node_create(
 	}
 	ichdr.level = level;
 
-	xfs_da3_node_hdr_to_disk(node, &ichdr);
+	dp->d_ops->node_hdr_to_disk(node, &ichdr);
 	xfs_trans_log_buf(tp, bp,
-		XFS_DA_LOGRANGE(node, &node->hdr, xfs_da3_node_hdr_size(node)));
+		XFS_DA_LOGRANGE(node, &node->hdr, dp->d_ops->node_hdr_size));
 
 	*bpp = bp;
 	return(0);
@@ -589,8 +543,8 @@ xfs_da3_root_split(
 	    oldroot->hdr.info.magic == cpu_to_be16(XFS_DA3_NODE_MAGIC)) {
 		struct xfs_da3_icnode_hdr nodehdr;
 
-		xfs_da3_node_hdr_from_disk(&nodehdr, oldroot);
-		btree = xfs_da3_node_tree_p(oldroot);
+		dp->d_ops->node_hdr_from_disk(&nodehdr, oldroot);
+		btree = dp->d_ops->node_tree_p(oldroot);
 		size = (int)((char *)&btree[nodehdr.count] - (char *)oldroot);
 		level = nodehdr.level;
 
@@ -604,8 +558,8 @@ xfs_da3_root_split(
 		struct xfs_dir2_leaf_entry *ents;
 
 		leaf = (xfs_dir2_leaf_t *)oldroot;
-		xfs_dir3_leaf_hdr_from_disk(&leafhdr, leaf);
-		ents = xfs_dir3_leaf_ents_p(leaf);
+		dp->d_ops->leaf_hdr_from_disk(&leafhdr, leaf);
+		ents = dp->d_ops->leaf_ents_p(leaf);
 
 		ASSERT(leafhdr.magic == XFS_DIR2_LEAFN_MAGIC ||
 		       leafhdr.magic == XFS_DIR3_LEAFN_MAGIC);
@@ -649,14 +603,14 @@ xfs_da3_root_split(
 		return error;
 
 	node = bp->b_addr;
-	xfs_da3_node_hdr_from_disk(&nodehdr, node);
-	btree = xfs_da3_node_tree_p(node);
+	dp->d_ops->node_hdr_from_disk(&nodehdr, node);
+	btree = dp->d_ops->node_tree_p(node);
 	btree[0].hashval = cpu_to_be32(blk1->hashval);
 	btree[0].before = cpu_to_be32(blk1->blkno);
 	btree[1].hashval = cpu_to_be32(blk2->hashval);
 	btree[1].before = cpu_to_be32(blk2->blkno);
 	nodehdr.count = 2;
-	xfs_da3_node_hdr_to_disk(node, &nodehdr);
+	dp->d_ops->node_hdr_to_disk(node, &nodehdr);
 
 #ifdef DEBUG
 	if (oldroot->hdr.info.magic == cpu_to_be16(XFS_DIR2_LEAFN_MAGIC) ||
@@ -693,11 +647,12 @@ xfs_da3_node_split(
 	int			newcount;
 	int			error;
 	int			useextra;
+	struct xfs_inode	*dp = state->args->dp;
 
 	trace_xfs_da_node_split(state->args);
 
 	node = oldblk->bp->b_addr;
-	xfs_da3_node_hdr_from_disk(&nodehdr, node);
+	dp->d_ops->node_hdr_from_disk(&nodehdr, node);
 
 	/*
 	 * With V2 dirs the extra block is data or freespace.
@@ -744,7 +699,7 @@ xfs_da3_node_split(
 	 * If we had double-split op below us, then add the extra block too.
 	 */
 	node = oldblk->bp->b_addr;
-	xfs_da3_node_hdr_from_disk(&nodehdr, node);
+	dp->d_ops->node_hdr_from_disk(&nodehdr, node);
 	if (oldblk->index <= nodehdr.count) {
 		oldblk->index++;
 		xfs_da3_node_add(state, oldblk, addblk);
@@ -793,15 +748,16 @@ xfs_da3_node_rebalance(
 	int			count;
 	int			tmp;
 	int			swap = 0;
+	struct xfs_inode	*dp = state->args->dp;
 
 	trace_xfs_da_node_rebalance(state->args);
 
 	node1 = blk1->bp->b_addr;
 	node2 = blk2->bp->b_addr;
-	xfs_da3_node_hdr_from_disk(&nodehdr1, node1);
-	xfs_da3_node_hdr_from_disk(&nodehdr2, node2);
-	btree1 = xfs_da3_node_tree_p(node1);
-	btree2 = xfs_da3_node_tree_p(node2);
+	dp->d_ops->node_hdr_from_disk(&nodehdr1, node1);
+	dp->d_ops->node_hdr_from_disk(&nodehdr2, node2);
+	btree1 = dp->d_ops->node_tree_p(node1);
+	btree2 = dp->d_ops->node_tree_p(node2);
 
 	/*
 	 * Figure out how many entries need to move, and in which direction.
@@ -814,10 +770,10 @@ xfs_da3_node_rebalance(
 		tmpnode = node1;
 		node1 = node2;
 		node2 = tmpnode;
-		xfs_da3_node_hdr_from_disk(&nodehdr1, node1);
-		xfs_da3_node_hdr_from_disk(&nodehdr2, node2);
-		btree1 = xfs_da3_node_tree_p(node1);
-		btree2 = xfs_da3_node_tree_p(node2);
+		dp->d_ops->node_hdr_from_disk(&nodehdr1, node1);
+		dp->d_ops->node_hdr_from_disk(&nodehdr2, node2);
+		btree1 = dp->d_ops->node_tree_p(node1);
+		btree2 = dp->d_ops->node_tree_p(node2);
 		swap = 1;
 	}
 
@@ -879,15 +835,14 @@ xfs_da3_node_rebalance(
 	/*
 	 * Log header of node 1 and all current bits of node 2.
 	 */
-	xfs_da3_node_hdr_to_disk(node1, &nodehdr1);
+	dp->d_ops->node_hdr_to_disk(node1, &nodehdr1);
 	xfs_trans_log_buf(tp, blk1->bp,
-		XFS_DA_LOGRANGE(node1, &node1->hdr,
-				xfs_da3_node_hdr_size(node1)));
+		XFS_DA_LOGRANGE(node1, &node1->hdr, dp->d_ops->node_hdr_size));
 
-	xfs_da3_node_hdr_to_disk(node2, &nodehdr2);
+	dp->d_ops->node_hdr_to_disk(node2, &nodehdr2);
 	xfs_trans_log_buf(tp, blk2->bp,
 		XFS_DA_LOGRANGE(node2, &node2->hdr,
-				xfs_da3_node_hdr_size(node2) +
+				dp->d_ops->node_hdr_size +
 				(sizeof(btree2[0]) * nodehdr2.count)));
 
 	/*
@@ -897,10 +852,10 @@ xfs_da3_node_rebalance(
 	if (swap) {
 		node1 = blk1->bp->b_addr;
 		node2 = blk2->bp->b_addr;
-		xfs_da3_node_hdr_from_disk(&nodehdr1, node1);
-		xfs_da3_node_hdr_from_disk(&nodehdr2, node2);
-		btree1 = xfs_da3_node_tree_p(node1);
-		btree2 = xfs_da3_node_tree_p(node2);
+		dp->d_ops->node_hdr_from_disk(&nodehdr1, node1);
+		dp->d_ops->node_hdr_from_disk(&nodehdr2, node2);
+		btree1 = dp->d_ops->node_tree_p(node1);
+		btree2 = dp->d_ops->node_tree_p(node2);
 	}
 	blk1->hashval = be32_to_cpu(btree1[nodehdr1.count - 1].hashval);
 	blk2->hashval = be32_to_cpu(btree2[nodehdr2.count - 1].hashval);
@@ -927,12 +882,13 @@ xfs_da3_node_add(
 	struct xfs_da3_icnode_hdr nodehdr;
 	struct xfs_da_node_entry *btree;
 	int			tmp;
+	struct xfs_inode	*dp = state->args->dp;
 
 	trace_xfs_da_node_add(state->args);
 
 	node = oldblk->bp->b_addr;
-	xfs_da3_node_hdr_from_disk(&nodehdr, node);
-	btree = xfs_da3_node_tree_p(node);
+	dp->d_ops->node_hdr_from_disk(&nodehdr, node);
+	btree = dp->d_ops->node_tree_p(node);
 
 	ASSERT(oldblk->index >= 0 && oldblk->index <= nodehdr.count);
 	ASSERT(newblk->blkno != 0);
@@ -955,9 +911,9 @@ xfs_da3_node_add(
 				tmp + sizeof(*btree)));
 
 	nodehdr.count += 1;
-	xfs_da3_node_hdr_to_disk(node, &nodehdr);
+	dp->d_ops->node_hdr_to_disk(node, &nodehdr);
 	xfs_trans_log_buf(state->args->trans, oldblk->bp,
-		XFS_DA_LOGRANGE(node, &node->hdr, xfs_da3_node_hdr_size(node)));
+		XFS_DA_LOGRANGE(node, &node->hdr, dp->d_ops->node_hdr_size));
 
 	/*
 	 * Copy the last hash value from the oldblk to propagate upwards.
@@ -1094,6 +1050,7 @@ xfs_da3_root_join(
 	struct xfs_da3_icnode_hdr oldroothdr;
 	struct xfs_da_node_entry *btree;
 	int			error;
+	struct xfs_inode	*dp = state->args->dp;
 
 	trace_xfs_da_root_join(state->args);
 
@@ -1101,7 +1058,7 @@ xfs_da3_root_join(
 
 	args = state->args;
 	oldroot = root_blk->bp->b_addr;
-	xfs_da3_node_hdr_from_disk(&oldroothdr, oldroot);
+	dp->d_ops->node_hdr_from_disk(&oldroothdr, oldroot);
 	ASSERT(oldroothdr.forw == 0);
 	ASSERT(oldroothdr.back == 0);
 
@@ -1115,10 +1072,10 @@ xfs_da3_root_join(
 	 * Read in the (only) child block, then copy those bytes into
 	 * the root block's buffer and free the original child block.
 	 */
-	btree = xfs_da3_node_tree_p(oldroot);
+	btree = dp->d_ops->node_tree_p(oldroot);
 	child = be32_to_cpu(btree[0].before);
 	ASSERT(child != 0);
-	error = xfs_da3_node_read(args->trans, args->dp, child, -1, &bp,
+	error = xfs_da3_node_read(args->trans, dp, child, -1, &bp,
 					     args->whichfork);
 	if (error)
 		return error;
@@ -1168,6 +1125,7 @@ xfs_da3_node_toosmall(
 	int			error;
 	int			retval;
 	int			i;
+	struct xfs_inode	*dp = state->args->dp;
 
 	trace_xfs_da_node_toosmall(state->args);
 
@@ -1179,7 +1137,7 @@ xfs_da3_node_toosmall(
 	blk = &state->path.blk[ state->path.active-1 ];
 	info = blk->bp->b_addr;
 	node = (xfs_da_intnode_t *)info;
-	xfs_da3_node_hdr_from_disk(&nodehdr, node);
+	dp->d_ops->node_hdr_from_disk(&nodehdr, node);
 	if (nodehdr.count > (state->node_ents >> 1)) {
 		*action = 0;	/* blk over 50%, don't try to join */
 		return(0);	/* blk over 50%, don't try to join */
@@ -1231,13 +1189,13 @@ xfs_da3_node_toosmall(
 			blkno = nodehdr.back;
 		if (blkno == 0)
 			continue;
-		error = xfs_da3_node_read(state->args->trans, state->args->dp,
+		error = xfs_da3_node_read(state->args->trans, dp,
 					blkno, -1, &bp, state->args->whichfork);
 		if (error)
 			return(error);
 
 		node = bp->b_addr;
-		xfs_da3_node_hdr_from_disk(&thdr, node);
+		dp->d_ops->node_hdr_from_disk(&thdr, node);
 		xfs_trans_brelse(state->args->trans, bp);
 
 		if (count - thdr.count >= 0)
@@ -1275,6 +1233,7 @@ xfs_da3_node_toosmall(
  */
 STATIC uint
 xfs_da3_node_lasthash(
+	struct xfs_inode	*dp,
 	struct xfs_buf		*bp,
 	int			*count)
 {
@@ -1283,12 +1242,12 @@ xfs_da3_node_lasthash(
 	struct xfs_da3_icnode_hdr nodehdr;
 
 	node = bp->b_addr;
-	xfs_da3_node_hdr_from_disk(&nodehdr, node);
+	dp->d_ops->node_hdr_from_disk(&nodehdr, node);
 	if (count)
 		*count = nodehdr.count;
 	if (!nodehdr.count)
 		return 0;
-	btree = xfs_da3_node_tree_p(node);
+	btree = dp->d_ops->node_tree_p(node);
 	return be32_to_cpu(btree[nodehdr.count - 1].hashval);
 }
 
@@ -1307,6 +1266,7 @@ xfs_da3_fixhashpath(
 	xfs_dahash_t		lasthash=0;
 	int			level;
 	int			count;
+	struct xfs_inode	*dp = state->args->dp;
 
 	trace_xfs_da_fixhashpath(state->args);
 
@@ -1319,12 +1279,12 @@ xfs_da3_fixhashpath(
 			return;
 		break;
 	case XFS_DIR2_LEAFN_MAGIC:
-		lasthash = xfs_dir2_leafn_lasthash(blk->bp, &count);
+		lasthash = xfs_dir2_leafn_lasthash(dp, blk->bp, &count);
 		if (count == 0)
 			return;
 		break;
 	case XFS_DA_NODE_MAGIC:
-		lasthash = xfs_da3_node_lasthash(blk->bp, &count);
+		lasthash = xfs_da3_node_lasthash(dp, blk->bp, &count);
 		if (count == 0)
 			return;
 		break;
@@ -1333,8 +1293,8 @@ xfs_da3_fixhashpath(
 		struct xfs_da3_icnode_hdr nodehdr;
 
 		node = blk->bp->b_addr;
-		xfs_da3_node_hdr_from_disk(&nodehdr, node);
-		btree = xfs_da3_node_tree_p(node);
+		dp->d_ops->node_hdr_from_disk(&nodehdr, node);
+		btree = dp->d_ops->node_tree_p(node);
 		if (be32_to_cpu(btree->hashval) == lasthash)
 			break;
 		blk->hashval = lasthash;
@@ -1360,11 +1320,12 @@ xfs_da3_node_remove(
 	struct xfs_da_node_entry *btree;
 	int			index;
 	int			tmp;
+	struct xfs_inode	*dp = state->args->dp;
 
 	trace_xfs_da_node_remove(state->args);
 
 	node = drop_blk->bp->b_addr;
-	xfs_da3_node_hdr_from_disk(&nodehdr, node);
+	dp->d_ops->node_hdr_from_disk(&nodehdr, node);
 	ASSERT(drop_blk->index < nodehdr.count);
 	ASSERT(drop_blk->index >= 0);
 
@@ -1372,7 +1333,7 @@ xfs_da3_node_remove(
 	 * Copy over the offending entry, or just zero it out.
 	 */
 	index = drop_blk->index;
-	btree = xfs_da3_node_tree_p(node);
+	btree = dp->d_ops->node_tree_p(node);
 	if (index < nodehdr.count - 1) {
 		tmp  = nodehdr.count - index - 1;
 		tmp *= (uint)sizeof(xfs_da_node_entry_t);
@@ -1385,9 +1346,9 @@ xfs_da3_node_remove(
 	xfs_trans_log_buf(state->args->trans, drop_blk->bp,
 	    XFS_DA_LOGRANGE(node, &btree[index], sizeof(btree[index])));
 	nodehdr.count -= 1;
-	xfs_da3_node_hdr_to_disk(node, &nodehdr);
+	dp->d_ops->node_hdr_to_disk(node, &nodehdr);
 	xfs_trans_log_buf(state->args->trans, drop_blk->bp,
-	    XFS_DA_LOGRANGE(node, &node->hdr, xfs_da3_node_hdr_size(node)));
+	    XFS_DA_LOGRANGE(node, &node->hdr, dp->d_ops->node_hdr_size));
 
 	/*
 	 * Copy the last hash value from the block to propagate upwards.
@@ -1414,15 +1375,16 @@ xfs_da3_node_unbalance(
 	struct xfs_trans	*tp;
 	int			sindex;
 	int			tmp;
+	struct xfs_inode	*dp = state->args->dp;
 
 	trace_xfs_da_node_unbalance(state->args);
 
 	drop_node = drop_blk->bp->b_addr;
 	save_node = save_blk->bp->b_addr;
-	xfs_da3_node_hdr_from_disk(&drop_hdr, drop_node);
-	xfs_da3_node_hdr_from_disk(&save_hdr, save_node);
-	drop_btree = xfs_da3_node_tree_p(drop_node);
-	save_btree = xfs_da3_node_tree_p(save_node);
+	dp->d_ops->node_hdr_from_disk(&drop_hdr, drop_node);
+	dp->d_ops->node_hdr_from_disk(&save_hdr, save_node);
+	drop_btree = dp->d_ops->node_tree_p(drop_node);
+	save_btree = dp->d_ops->node_tree_p(save_node);
 	tp = state->args->trans;
 
 	/*
@@ -1456,10 +1418,10 @@ xfs_da3_node_unbalance(
 	memcpy(&save_btree[sindex], &drop_btree[0], tmp);
 	save_hdr.count += drop_hdr.count;
 
-	xfs_da3_node_hdr_to_disk(save_node, &save_hdr);
+	dp->d_ops->node_hdr_to_disk(save_node, &save_hdr);
 	xfs_trans_log_buf(tp, save_blk->bp,
 		XFS_DA_LOGRANGE(save_node, &save_node->hdr,
-				xfs_da3_node_hdr_size(save_node)));
+				dp->d_ops->node_hdr_size));
 
 	/*
 	 * Save the last hashval in the remaining block for upward propagation.
@@ -1501,6 +1463,7 @@ xfs_da3_node_lookup_int(
 	int			max;
 	int			error;
 	int			retval;
+	struct xfs_inode	*dp = state->args->dp;
 
 	args = state->args;
 
@@ -1536,7 +1499,8 @@ xfs_da3_node_lookup_int(
 		if (blk->magic == XFS_DIR2_LEAFN_MAGIC ||
 		    blk->magic == XFS_DIR3_LEAFN_MAGIC) {
 			blk->magic = XFS_DIR2_LEAFN_MAGIC;
-			blk->hashval = xfs_dir2_leafn_lasthash(blk->bp, NULL);
+			blk->hashval = xfs_dir2_leafn_lasthash(args->dp,
+							       blk->bp, NULL);
 			break;
 		}
 
@@ -1547,8 +1511,8 @@ xfs_da3_node_lookup_int(
 		 * Search an intermediate node for a match.
 		 */
 		node = blk->bp->b_addr;
-		xfs_da3_node_hdr_from_disk(&nodehdr, node);
-		btree = xfs_da3_node_tree_p(node);
+		dp->d_ops->node_hdr_from_disk(&nodehdr, node);
+		btree = dp->d_ops->node_tree_p(node);
 
 		max = nodehdr.count;
 		blk->hashval = be32_to_cpu(btree[max - 1].hashval);
@@ -1643,6 +1607,7 @@ xfs_da3_node_lookup_int(
  */
 STATIC int
 xfs_da3_node_order(
+	struct xfs_inode *dp,
 	struct xfs_buf	*node1_bp,
 	struct xfs_buf	*node2_bp)
 {
@@ -1655,10 +1620,10 @@ xfs_da3_node_order(
 
 	node1 = node1_bp->b_addr;
 	node2 = node2_bp->b_addr;
-	xfs_da3_node_hdr_from_disk(&node1hdr, node1);
-	xfs_da3_node_hdr_from_disk(&node2hdr, node2);
-	btree1 = xfs_da3_node_tree_p(node1);
-	btree2 = xfs_da3_node_tree_p(node2);
+	dp->d_ops->node_hdr_from_disk(&node1hdr, node1);
+	dp->d_ops->node_hdr_from_disk(&node2hdr, node2);
+	btree1 = dp->d_ops->node_tree_p(node1);
+	btree2 = dp->d_ops->node_tree_p(node2);
 
 	if (node1hdr.count > 0 && node2hdr.count > 0 &&
 	    ((be32_to_cpu(btree2[0].hashval) < be32_to_cpu(btree1[0].hashval)) ||
@@ -1685,6 +1650,7 @@ xfs_da3_blk_link(
 	struct xfs_buf		*bp;
 	int			before = 0;
 	int			error;
+	struct xfs_inode	*dp = state->args->dp;
 
 	/*
 	 * Set up environment.
@@ -1702,10 +1668,10 @@ xfs_da3_blk_link(
 		before = xfs_attr_leaf_order(old_blk->bp, new_blk->bp);
 		break;
 	case XFS_DIR2_LEAFN_MAGIC:
-		before = xfs_dir2_leafn_order(old_blk->bp, new_blk->bp);
+		before = xfs_dir2_leafn_order(dp, old_blk->bp, new_blk->bp);
 		break;
 	case XFS_DA_NODE_MAGIC:
-		before = xfs_da3_node_order(old_blk->bp, new_blk->bp);
+		before = xfs_da3_node_order(dp, old_blk->bp, new_blk->bp);
 		break;
 	}
 
@@ -1720,7 +1686,7 @@ xfs_da3_blk_link(
 		new_info->forw = cpu_to_be32(old_blk->blkno);
 		new_info->back = old_info->back;
 		if (old_info->back) {
-			error = xfs_da3_node_read(args->trans, args->dp,
+			error = xfs_da3_node_read(args->trans, dp,
 						be32_to_cpu(old_info->back),
 						-1, &bp, args->whichfork);
 			if (error)
@@ -1741,7 +1707,7 @@ xfs_da3_blk_link(
 		new_info->forw = old_info->forw;
 		new_info->back = cpu_to_be32(old_blk->blkno);
 		if (old_info->forw) {
-			error = xfs_da3_node_read(args->trans, args->dp,
+			error = xfs_da3_node_read(args->trans, dp,
 						be32_to_cpu(old_info->forw),
 						-1, &bp, args->whichfork);
 			if (error)
@@ -1861,6 +1827,7 @@ xfs_da3_path_shift(
 	xfs_dablk_t		blkno = 0;
 	int			level;
 	int			error;
+	struct xfs_inode	*dp = state->args->dp;
 
 	trace_xfs_da_path_shift(state->args);
 
@@ -1876,8 +1843,8 @@ xfs_da3_path_shift(
 	level = (path->active-1) - 1;	/* skip bottom layer in path */
 	for (blk = &path->blk[level]; level >= 0; blk--, level--) {
 		node = blk->bp->b_addr;
-		xfs_da3_node_hdr_from_disk(&nodehdr, node);
-		btree = xfs_da3_node_tree_p(node);
+		dp->d_ops->node_hdr_from_disk(&nodehdr, node);
+		btree = dp->d_ops->node_tree_p(node);
 
 		if (forward && (blk->index < nodehdr.count - 1)) {
 			blk->index++;
@@ -1911,7 +1878,7 @@ xfs_da3_path_shift(
 		 * Read the next child block.
 		 */
 		blk->blkno = blkno;
-		error = xfs_da3_node_read(args->trans, args->dp, blkno, -1,
+		error = xfs_da3_node_read(args->trans, dp, blkno, -1,
 					&blk->bp, args->whichfork);
 		if (error)
 			return(error);
@@ -1933,8 +1900,8 @@ xfs_da3_path_shift(
 		case XFS_DA3_NODE_MAGIC:
 			blk->magic = XFS_DA_NODE_MAGIC;
 			node = (xfs_da_intnode_t *)info;
-			xfs_da3_node_hdr_from_disk(&nodehdr, node);
-			btree = xfs_da3_node_tree_p(node);
+			dp->d_ops->node_hdr_from_disk(&nodehdr, node);
+			btree = dp->d_ops->node_tree_p(node);
 			blk->hashval = be32_to_cpu(btree[nodehdr.count - 1].hashval);
 			if (forward)
 				blk->index = 0;
@@ -1947,16 +1914,15 @@ xfs_da3_path_shift(
 			blk->magic = XFS_ATTR_LEAF_MAGIC;
 			ASSERT(level == path->active-1);
 			blk->index = 0;
-			blk->hashval = xfs_attr_leaf_lasthash(blk->bp,
-							      NULL);
+			blk->hashval = xfs_attr_leaf_lasthash(blk->bp, NULL);
 			break;
 		case XFS_DIR2_LEAFN_MAGIC:
 		case XFS_DIR3_LEAFN_MAGIC:
 			blk->magic = XFS_DIR2_LEAFN_MAGIC;
 			ASSERT(level == path->active-1);
 			blk->index = 0;
-			blk->hashval = xfs_dir2_leafn_lasthash(blk->bp,
-							       NULL);
+			blk->hashval = xfs_dir2_leafn_lasthash(args->dp,
+							       blk->bp, NULL);
 			break;
 		default:
 			ASSERT(0);
@@ -2163,7 +2129,7 @@ xfs_da3_swap_lastblock(
 	struct xfs_dir2_leaf	*dead_leaf2;
 	struct xfs_da_node_entry *btree;
 	struct xfs_da3_icnode_hdr par_hdr;
-	struct xfs_inode	*ip;
+	struct xfs_inode	*dp;
 	struct xfs_trans	*tp;
 	struct xfs_mount	*mp;
 	struct xfs_buf		*dead_buf;
@@ -2187,12 +2153,12 @@ xfs_da3_swap_lastblock(
 	dead_buf = *dead_bufp;
 	dead_blkno = *dead_blknop;
 	tp = args->trans;
-	ip = args->dp;
+	dp = args->dp;
 	w = args->whichfork;
 	ASSERT(w == XFS_DATA_FORK);
-	mp = ip->i_mount;
+	mp = dp->i_mount;
 	lastoff = mp->m_dirfreeblk;
-	error = xfs_bmap_last_before(tp, ip, &lastoff, w);
+	error = xfs_bmap_last_before(tp, dp, &lastoff, w);
 	if (error)
 		return error;
 	if (unlikely(lastoff == 0)) {
@@ -2204,7 +2170,7 @@ xfs_da3_swap_lastblock(
 	 * Read the last block in the btree space.
 	 */
 	last_blkno = (xfs_dablk_t)lastoff - mp->m_dirblkfsbs;
-	error = xfs_da3_node_read(tp, ip, last_blkno, -1, &last_buf, w);
+	error = xfs_da3_node_read(tp, dp, last_blkno, -1, &last_buf, w);
 	if (error)
 		return error;
 	/*
@@ -2222,16 +2188,16 @@ xfs_da3_swap_lastblock(
 		struct xfs_dir2_leaf_entry *ents;
 
 		dead_leaf2 = (xfs_dir2_leaf_t *)dead_info;
-		xfs_dir3_leaf_hdr_from_disk(&leafhdr, dead_leaf2);
-		ents = xfs_dir3_leaf_ents_p(dead_leaf2);
+		dp->d_ops->leaf_hdr_from_disk(&leafhdr, dead_leaf2);
+		ents = dp->d_ops->leaf_ents_p(dead_leaf2);
 		dead_level = 0;
 		dead_hash = be32_to_cpu(ents[leafhdr.count - 1].hashval);
 	} else {
 		struct xfs_da3_icnode_hdr deadhdr;
 
 		dead_node = (xfs_da_intnode_t *)dead_info;
-		xfs_da3_node_hdr_from_disk(&deadhdr, dead_node);
-		btree = xfs_da3_node_tree_p(dead_node);
+		dp->d_ops->node_hdr_from_disk(&deadhdr, dead_node);
+		btree = dp->d_ops->node_tree_p(dead_node);
 		dead_level = deadhdr.level;
 		dead_hash = be32_to_cpu(btree[deadhdr.count - 1].hashval);
 	}
@@ -2240,7 +2206,7 @@ xfs_da3_swap_lastblock(
 	 * If the moved block has a left sibling, fix up the pointers.
 	 */
 	if ((sib_blkno = be32_to_cpu(dead_info->back))) {
-		error = xfs_da3_node_read(tp, ip, sib_blkno, -1, &sib_buf, w);
+		error = xfs_da3_node_read(tp, dp, sib_blkno, -1, &sib_buf, w);
 		if (error)
 			goto done;
 		sib_info = sib_buf->b_addr;
@@ -2262,7 +2228,7 @@ xfs_da3_swap_lastblock(
 	 * If the moved block has a right sibling, fix up the pointers.
 	 */
 	if ((sib_blkno = be32_to_cpu(dead_info->forw))) {
-		error = xfs_da3_node_read(tp, ip, sib_blkno, -1, &sib_buf, w);
+		error = xfs_da3_node_read(tp, dp, sib_blkno, -1, &sib_buf, w);
 		if (error)
 			goto done;
 		sib_info = sib_buf->b_addr;
@@ -2286,11 +2252,11 @@ xfs_da3_swap_lastblock(
 	 * Walk down the tree looking for the parent of the moved block.
 	 */
 	for (;;) {
-		error = xfs_da3_node_read(tp, ip, par_blkno, -1, &par_buf, w);
+		error = xfs_da3_node_read(tp, dp, par_blkno, -1, &par_buf, w);
 		if (error)
 			goto done;
 		par_node = par_buf->b_addr;
-		xfs_da3_node_hdr_from_disk(&par_hdr, par_node);
+		dp->d_ops->node_hdr_from_disk(&par_hdr, par_node);
 		if (level >= 0 && level != par_hdr.level + 1) {
 			XFS_ERROR_REPORT("xfs_da_swap_lastblock(4)",
 					 XFS_ERRLEVEL_LOW, mp);
@@ -2298,7 +2264,7 @@ xfs_da3_swap_lastblock(
 			goto done;
 		}
 		level = par_hdr.level;
-		btree = xfs_da3_node_tree_p(par_node);
+		btree = dp->d_ops->node_tree_p(par_node);
 		for (entno = 0;
 		     entno < par_hdr.count &&
 		     be32_to_cpu(btree[entno].hashval) < dead_hash;
@@ -2337,18 +2303,18 @@ xfs_da3_swap_lastblock(
 			error = XFS_ERROR(EFSCORRUPTED);
 			goto done;
 		}
-		error = xfs_da3_node_read(tp, ip, par_blkno, -1, &par_buf, w);
+		error = xfs_da3_node_read(tp, dp, par_blkno, -1, &par_buf, w);
 		if (error)
 			goto done;
 		par_node = par_buf->b_addr;
-		xfs_da3_node_hdr_from_disk(&par_hdr, par_node);
+		dp->d_ops->node_hdr_from_disk(&par_hdr, par_node);
 		if (par_hdr.level != level) {
 			XFS_ERROR_REPORT("xfs_da_swap_lastblock(7)",
 					 XFS_ERRLEVEL_LOW, mp);
 			error = XFS_ERROR(EFSCORRUPTED);
 			goto done;
 		}
-		btree = xfs_da3_node_tree_p(par_node);
+		btree = dp->d_ops->node_tree_p(par_node);
 		entno = 0;
 	}
 	/*
diff --git a/fs/xfs/xfs_da_btree.h b/fs/xfs/xfs_da_btree.h
index b1f267995dea..6e95ea79f5d7 100644
--- a/fs/xfs/xfs_da_btree.h
+++ b/fs/xfs/xfs_da_btree.h
@@ -23,146 +23,7 @@ struct xfs_bmap_free;
 struct xfs_inode;
 struct xfs_trans;
 struct zone;
-
-/*========================================================================
- * Directory Structure when greater than XFS_LBSIZE(mp) bytes.
- *========================================================================*/
-
-/*
- * This structure is common to both leaf nodes and non-leaf nodes in the Btree.
- *
- * It is used to manage a doubly linked list of all blocks at the same
- * level in the Btree, and to identify which type of block this is.
- */
-#define XFS_DA_NODE_MAGIC	0xfebe	/* magic number: non-leaf blocks */
-#define XFS_ATTR_LEAF_MAGIC	0xfbee	/* magic number: attribute leaf blks */
-#define	XFS_DIR2_LEAF1_MAGIC	0xd2f1	/* magic number: v2 dirlf single blks */
-#define	XFS_DIR2_LEAFN_MAGIC	0xd2ff	/* magic number: v2 dirlf multi blks */
-
-typedef struct xfs_da_blkinfo {
-	__be32		forw;			/* previous block in list */
-	__be32		back;			/* following block in list */
-	__be16		magic;			/* validity check on block */
-	__be16		pad;			/* unused */
-} xfs_da_blkinfo_t;
-
-/*
- * CRC enabled directory structure types
- *
- * The headers change size for the additional verification information, but
- * otherwise the tree layouts and contents are unchanged. Hence the da btree
- * code can use the struct xfs_da_blkinfo for manipulating the tree links and
- * magic numbers without modification for both v2 and v3 nodes.
- */
-#define XFS_DA3_NODE_MAGIC	0x3ebe	/* magic number: non-leaf blocks */
-#define XFS_ATTR3_LEAF_MAGIC	0x3bee	/* magic number: attribute leaf blks */
-#define	XFS_DIR3_LEAF1_MAGIC	0x3df1	/* magic number: v2 dirlf single blks */
-#define	XFS_DIR3_LEAFN_MAGIC	0x3dff	/* magic number: v2 dirlf multi blks */
-
-struct xfs_da3_blkinfo {
-	/*
-	 * the node link manipulation code relies on the fact that the first
-	 * element of this structure is the struct xfs_da_blkinfo so it can
-	 * ignore the differences in the rest of the structures.
-	 */
-	struct xfs_da_blkinfo	hdr;
-	__be32			crc;	/* CRC of block */
-	__be64			blkno;	/* first block of the buffer */
-	__be64			lsn;	/* sequence number of last write */
-	uuid_t			uuid;	/* filesystem we belong to */
-	__be64			owner;	/* inode that owns the block */
-};
-
-/*
- * This is the structure of the root and intermediate nodes in the Btree.
- * The leaf nodes are defined above.
- *
- * Entries are not packed.
- *
- * Since we have duplicate keys, use a binary search but always follow
- * all match in the block, not just the first match found.
- */
-#define	XFS_DA_NODE_MAXDEPTH	5	/* max depth of Btree */
-
-typedef struct xfs_da_node_hdr {
-	struct xfs_da_blkinfo	info;	/* block type, links, etc. */
-	__be16			__count; /* count of active entries */
-	__be16			__level; /* level above leaves (leaf == 0) */
-} xfs_da_node_hdr_t;
-
-struct xfs_da3_node_hdr {
-	struct xfs_da3_blkinfo	info;	/* block type, links, etc. */
-	__be16			__count; /* count of active entries */
-	__be16			__level; /* level above leaves (leaf == 0) */
-	__be32			__pad32;
-};
-
-#define XFS_DA3_NODE_CRC_OFF	(offsetof(struct xfs_da3_node_hdr, info.crc))
-
-typedef struct xfs_da_node_entry {
-	__be32	hashval;	/* hash value for this descendant */
-	__be32	before;		/* Btree block before this key */
-} xfs_da_node_entry_t;
-
-typedef struct xfs_da_intnode {
-	struct xfs_da_node_hdr	hdr;
-	struct xfs_da_node_entry __btree[];
-} xfs_da_intnode_t;
-
-struct xfs_da3_intnode {
-	struct xfs_da3_node_hdr	hdr;
-	struct xfs_da_node_entry __btree[];
-};
-
-/*
- * In-core version of the node header to abstract the differences in the v2 and
- * v3 disk format of the headers. Callers need to convert to/from disk format as
- * appropriate.
- */
-struct xfs_da3_icnode_hdr {
-	__uint32_t	forw;
-	__uint32_t	back;
-	__uint16_t	magic;
-	__uint16_t	count;
-	__uint16_t	level;
-};
-
-extern void xfs_da3_node_hdr_from_disk(struct xfs_da3_icnode_hdr *to,
-				       struct xfs_da_intnode *from);
-extern void xfs_da3_node_hdr_to_disk(struct xfs_da_intnode *to,
-				     struct xfs_da3_icnode_hdr *from);
-
-static inline int
-__xfs_da3_node_hdr_size(bool v3)
-{
-	if (v3)
-		return sizeof(struct xfs_da3_node_hdr);
-	return sizeof(struct xfs_da_node_hdr);
-}
-static inline int
-xfs_da3_node_hdr_size(struct xfs_da_intnode *dap)
-{
-	bool	v3 = dap->hdr.info.magic == cpu_to_be16(XFS_DA3_NODE_MAGIC);
-
-	return __xfs_da3_node_hdr_size(v3);
-}
-
-static inline struct xfs_da_node_entry *
-xfs_da3_node_tree_p(struct xfs_da_intnode *dap)
-{
-	if (dap->hdr.info.magic == cpu_to_be16(XFS_DA3_NODE_MAGIC)) {
-		struct xfs_da3_intnode *dap3 = (struct xfs_da3_intnode *)dap;
-		return dap3->__btree;
-	}
-	return dap->__btree;
-}
-
-extern void xfs_da3_intnode_from_disk(struct xfs_da3_icnode_hdr *to,
-				      struct xfs_da_intnode *from);
-extern void xfs_da3_intnode_to_disk(struct xfs_da_intnode *to,
-				    struct xfs_da3_icnode_hdr *from);
-
-#define	XFS_LBSIZE(mp)	(mp)->m_sb.sb_blocksize
+struct xfs_dir_ops;
 
 /*========================================================================
  * Btree searching and modification structure definitions.
@@ -309,8 +170,6 @@ int	xfs_da3_node_read(struct xfs_trans *tp, struct xfs_inode *dp,
 			 xfs_dablk_t bno, xfs_daddr_t mappedbno,
 			 struct xfs_buf **bpp, int which_fork);
 
-extern const struct xfs_buf_ops xfs_da3_node_buf_ops;
-
 /*
  * Utility routines.
  */
diff --git a/fs/xfs/xfs_da_format.c b/fs/xfs/xfs_da_format.c
new file mode 100644
index 000000000000..e6c83e1fbc8a
--- /dev/null
+++ b/fs/xfs/xfs_da_format.c
@@ -0,0 +1,907 @@
+/*
+ * Copyright (c) 2000,2002,2005 Silicon Graphics, Inc.
+ * Copyright (c) 2013 Red Hat, Inc.
+ * All Rights Reserved.
+ *
+ * This program is free software; you can redistribute it and/or
+ * modify it under the terms of the GNU General Public License as
+ * published by the Free Software Foundation.
+ *
+ * This program is distributed in the hope that it would be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+ * GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program; if not, write the Free Software Foundation,
+ * Inc.,  51 Franklin St, Fifth Floor, Boston, MA  02110-1301  USA
+ */
+#include "xfs.h"
+#include "xfs_fs.h"
+#include "xfs_shared.h"
+#include "xfs_format.h"
+#include "xfs_log_format.h"
+#include "xfs_trans_resv.h"
+#include "xfs_sb.h"
+#include "xfs_ag.h"
+#include "xfs_mount.h"
+#include "xfs_da_format.h"
+#include "xfs_inode.h"
+#include "xfs_dir2.h"
+
+/*
+ * Shortform directory ops
+ */
+static int
+xfs_dir2_sf_entsize(
+	struct xfs_dir2_sf_hdr	*hdr,
+	int			len)
+{
+	int count = sizeof(struct xfs_dir2_sf_entry);	/* namelen + offset */
+
+	count += len;					/* name */
+	count += hdr->i8count ? sizeof(xfs_dir2_ino8_t) :
+				sizeof(xfs_dir2_ino4_t); /* ino # */
+	return count;
+}
+
+static int
+xfs_dir3_sf_entsize(
+	struct xfs_dir2_sf_hdr	*hdr,
+	int			len)
+{
+	return xfs_dir2_sf_entsize(hdr, len) + sizeof(__uint8_t);
+}
+
+static struct xfs_dir2_sf_entry *
+xfs_dir2_sf_nextentry(
+	struct xfs_dir2_sf_hdr	*hdr,
+	struct xfs_dir2_sf_entry *sfep)
+{
+	return (struct xfs_dir2_sf_entry *)
+		((char *)sfep + xfs_dir2_sf_entsize(hdr, sfep->namelen));
+}
+
+static struct xfs_dir2_sf_entry *
+xfs_dir3_sf_nextentry(
+	struct xfs_dir2_sf_hdr	*hdr,
+	struct xfs_dir2_sf_entry *sfep)
+{
+	return (struct xfs_dir2_sf_entry *)
+		((char *)sfep + xfs_dir3_sf_entsize(hdr, sfep->namelen));
+}
+
+
+/*
+ * For filetype enabled shortform directories, the file type field is stored at
+ * the end of the name.  Because it's only a single byte, endian conversion is
+ * not necessary. For non-filetype enable directories, the type is always
+ * unknown and we never store the value.
+ */
+static __uint8_t
+xfs_dir2_sfe_get_ftype(
+	struct xfs_dir2_sf_entry *sfep)
+{
+	return XFS_DIR3_FT_UNKNOWN;
+}
+
+static void
+xfs_dir2_sfe_put_ftype(
+	struct xfs_dir2_sf_entry *sfep,
+	__uint8_t		ftype)
+{
+	ASSERT(ftype < XFS_DIR3_FT_MAX);
+}
+
+static __uint8_t
+xfs_dir3_sfe_get_ftype(
+	struct xfs_dir2_sf_entry *sfep)
+{
+	__uint8_t	ftype;
+
+	ftype = sfep->name[sfep->namelen];
+	if (ftype >= XFS_DIR3_FT_MAX)
+		return XFS_DIR3_FT_UNKNOWN;
+	return ftype;
+}
+
+static void
+xfs_dir3_sfe_put_ftype(
+	struct xfs_dir2_sf_entry *sfep,
+	__uint8_t		ftype)
+{
+	ASSERT(ftype < XFS_DIR3_FT_MAX);
+
+	sfep->name[sfep->namelen] = ftype;
+}
+
+/*
+ * Inode numbers in short-form directories can come in two versions,
+ * either 4 bytes or 8 bytes wide.  These helpers deal with the
+ * two forms transparently by looking at the headers i8count field.
+ *
+ * For 64-bit inode number the most significant byte must be zero.
+ */
+static xfs_ino_t
+xfs_dir2_sf_get_ino(
+	struct xfs_dir2_sf_hdr	*hdr,
+	xfs_dir2_inou_t		*from)
+{
+	if (hdr->i8count)
+		return get_unaligned_be64(&from->i8.i) & 0x00ffffffffffffffULL;
+	else
+		return get_unaligned_be32(&from->i4.i);
+}
+
+static void
+xfs_dir2_sf_put_ino(
+	struct xfs_dir2_sf_hdr	*hdr,
+	xfs_dir2_inou_t		*to,
+	xfs_ino_t		ino)
+{
+	ASSERT((ino & 0xff00000000000000ULL) == 0);
+
+	if (hdr->i8count)
+		put_unaligned_be64(ino, &to->i8.i);
+	else
+		put_unaligned_be32(ino, &to->i4.i);
+}
+
+static xfs_ino_t
+xfs_dir2_sf_get_parent_ino(
+	struct xfs_dir2_sf_hdr	*hdr)
+{
+	return xfs_dir2_sf_get_ino(hdr, &hdr->parent);
+}
+
+static void
+xfs_dir2_sf_put_parent_ino(
+	struct xfs_dir2_sf_hdr	*hdr,
+	xfs_ino_t		ino)
+{
+	xfs_dir2_sf_put_ino(hdr, &hdr->parent, ino);
+}
+
+/*
+ * In short-form directory entries the inode numbers are stored at variable
+ * offset behind the entry name. If the entry stores a filetype value, then it
+ * sits between the name and the inode number. Hence the inode numbers may only
+ * be accessed through the helpers below.
+ */
+static xfs_ino_t
+xfs_dir2_sfe_get_ino(
+	struct xfs_dir2_sf_hdr	*hdr,
+	struct xfs_dir2_sf_entry *sfep)
+{
+	return xfs_dir2_sf_get_ino(hdr,
+				(xfs_dir2_inou_t *)&sfep->name[sfep->namelen]);
+}
+
+static void
+xfs_dir2_sfe_put_ino(
+	struct xfs_dir2_sf_hdr	*hdr,
+	struct xfs_dir2_sf_entry *sfep,
+	xfs_ino_t		ino)
+{
+	xfs_dir2_sf_put_ino(hdr,
+			    (xfs_dir2_inou_t *)&sfep->name[sfep->namelen], ino);
+}
+
+static xfs_ino_t
+xfs_dir3_sfe_get_ino(
+	struct xfs_dir2_sf_hdr	*hdr,
+	struct xfs_dir2_sf_entry *sfep)
+{
+	return xfs_dir2_sf_get_ino(hdr,
+			(xfs_dir2_inou_t *)&sfep->name[sfep->namelen + 1]);
+}
+
+static void
+xfs_dir3_sfe_put_ino(
+	struct xfs_dir2_sf_hdr	*hdr,
+	struct xfs_dir2_sf_entry *sfep,
+	xfs_ino_t		ino)
+{
+	xfs_dir2_sf_put_ino(hdr,
+			(xfs_dir2_inou_t *)&sfep->name[sfep->namelen + 1], ino);
+}
+
+
+/*
+ * Directory data block operations
+ */
+
+/*
+ * For special situations, the dirent size ends up fixed because we always know
+ * what the size of the entry is. That's true for the "." and "..", and
+ * therefore we know that they are a fixed size and hence their offsets are
+ * constant, as is the first entry.
+ *
+ * Hence, this calculation is written as a macro to be able to be calculated at
+ * compile time and so certain offsets can be calculated directly in the
+ * structure initaliser via the macro. There are two macros - one for dirents
+ * with ftype and without so there are no unresolvable conditionals in the
+ * calculations. We also use round_up() as XFS_DIR2_DATA_ALIGN is always a power
+ * of 2 and the compiler doesn't reject it (unlike roundup()).
+ */
+#define XFS_DIR2_DATA_ENTSIZE(n)					\
+	round_up((offsetof(struct xfs_dir2_data_entry, name[0]) + (n) +	\
+		 sizeof(xfs_dir2_data_off_t)), XFS_DIR2_DATA_ALIGN)
+
+#define XFS_DIR3_DATA_ENTSIZE(n)					\
+	round_up((offsetof(struct xfs_dir2_data_entry, name[0]) + (n) +	\
+		 sizeof(xfs_dir2_data_off_t) + sizeof(__uint8_t)),	\
+		XFS_DIR2_DATA_ALIGN)
+
+static int
+xfs_dir2_data_entsize(
+	int			n)
+{
+	return XFS_DIR2_DATA_ENTSIZE(n);
+}
+
+static int
+xfs_dir3_data_entsize(
+	int			n)
+{
+	return XFS_DIR3_DATA_ENTSIZE(n);
+}
+
+static __uint8_t
+xfs_dir2_data_get_ftype(
+	struct xfs_dir2_data_entry *dep)
+{
+	return XFS_DIR3_FT_UNKNOWN;
+}
+
+static void
+xfs_dir2_data_put_ftype(
+	struct xfs_dir2_data_entry *dep,
+	__uint8_t		ftype)
+{
+	ASSERT(ftype < XFS_DIR3_FT_MAX);
+}
+
+static __uint8_t
+xfs_dir3_data_get_ftype(
+	struct xfs_dir2_data_entry *dep)
+{
+	__uint8_t	ftype = dep->name[dep->namelen];
+
+	ASSERT(ftype < XFS_DIR3_FT_MAX);
+	if (ftype >= XFS_DIR3_FT_MAX)
+		return XFS_DIR3_FT_UNKNOWN;
+	return ftype;
+}
+
+static void
+xfs_dir3_data_put_ftype(
+	struct xfs_dir2_data_entry *dep,
+	__uint8_t		type)
+{
+	ASSERT(type < XFS_DIR3_FT_MAX);
+	ASSERT(dep->namelen != 0);
+
+	dep->name[dep->namelen] = type;
+}
+
+/*
+ * Pointer to an entry's tag word.
+ */
+static __be16 *
+xfs_dir2_data_entry_tag_p(
+	struct xfs_dir2_data_entry *dep)
+{
+	return (__be16 *)((char *)dep +
+		xfs_dir2_data_entsize(dep->namelen) - sizeof(__be16));
+}
+
+static __be16 *
+xfs_dir3_data_entry_tag_p(
+	struct xfs_dir2_data_entry *dep)
+{
+	return (__be16 *)((char *)dep +
+		xfs_dir3_data_entsize(dep->namelen) - sizeof(__be16));
+}
+
+/*
+ * location of . and .. in data space (always block 0)
+ */
+static struct xfs_dir2_data_entry *
+xfs_dir2_data_dot_entry_p(
+	struct xfs_dir2_data_hdr *hdr)
+{
+	return (struct xfs_dir2_data_entry *)
+		((char *)hdr + sizeof(struct xfs_dir2_data_hdr));
+}
+
+static struct xfs_dir2_data_entry *
+xfs_dir2_data_dotdot_entry_p(
+	struct xfs_dir2_data_hdr *hdr)
+{
+	return (struct xfs_dir2_data_entry *)
+		((char *)hdr + sizeof(struct xfs_dir2_data_hdr) +
+				XFS_DIR2_DATA_ENTSIZE(1));
+}
+
+static struct xfs_dir2_data_entry *
+xfs_dir2_data_first_entry_p(
+	struct xfs_dir2_data_hdr *hdr)
+{
+	return (struct xfs_dir2_data_entry *)
+		((char *)hdr + sizeof(struct xfs_dir2_data_hdr) +
+				XFS_DIR2_DATA_ENTSIZE(1) +
+				XFS_DIR2_DATA_ENTSIZE(2));
+}
+
+static struct xfs_dir2_data_entry *
+xfs_dir2_ftype_data_dotdot_entry_p(
+	struct xfs_dir2_data_hdr *hdr)
+{
+	return (struct xfs_dir2_data_entry *)
+		((char *)hdr + sizeof(struct xfs_dir2_data_hdr) +
+				XFS_DIR3_DATA_ENTSIZE(1));
+}
+
+static struct xfs_dir2_data_entry *
+xfs_dir2_ftype_data_first_entry_p(
+	struct xfs_dir2_data_hdr *hdr)
+{
+	return (struct xfs_dir2_data_entry *)
+		((char *)hdr + sizeof(struct xfs_dir2_data_hdr) +
+				XFS_DIR3_DATA_ENTSIZE(1) +
+				XFS_DIR3_DATA_ENTSIZE(2));
+}
+
+static struct xfs_dir2_data_entry *
+xfs_dir3_data_dot_entry_p(
+	struct xfs_dir2_data_hdr *hdr)
+{
+	return (struct xfs_dir2_data_entry *)
+		((char *)hdr + sizeof(struct xfs_dir3_data_hdr));
+}
+
+static struct xfs_dir2_data_entry *
+xfs_dir3_data_dotdot_entry_p(
+	struct xfs_dir2_data_hdr *hdr)
+{
+	return (struct xfs_dir2_data_entry *)
+		((char *)hdr + sizeof(struct xfs_dir3_data_hdr) +
+				XFS_DIR3_DATA_ENTSIZE(1));
+}
+
+static struct xfs_dir2_data_entry *
+xfs_dir3_data_first_entry_p(
+	struct xfs_dir2_data_hdr *hdr)
+{
+	return (struct xfs_dir2_data_entry *)
+		((char *)hdr + sizeof(struct xfs_dir3_data_hdr) +
+				XFS_DIR3_DATA_ENTSIZE(1) +
+				XFS_DIR3_DATA_ENTSIZE(2));
+}
+
+static struct xfs_dir2_data_free *
+xfs_dir2_data_bestfree_p(struct xfs_dir2_data_hdr *hdr)
+{
+	return hdr->bestfree;
+}
+
+static struct xfs_dir2_data_free *
+xfs_dir3_data_bestfree_p(struct xfs_dir2_data_hdr *hdr)
+{
+	return ((struct xfs_dir3_data_hdr *)hdr)->best_free;
+}
+
+static struct xfs_dir2_data_entry *
+xfs_dir2_data_entry_p(struct xfs_dir2_data_hdr *hdr)
+{
+	return (struct xfs_dir2_data_entry *)
+		((char *)hdr + sizeof(struct xfs_dir2_data_hdr));
+}
+
+static struct xfs_dir2_data_unused *
+xfs_dir2_data_unused_p(struct xfs_dir2_data_hdr *hdr)
+{
+	return (struct xfs_dir2_data_unused *)
+		((char *)hdr + sizeof(struct xfs_dir2_data_hdr));
+}
+
+static struct xfs_dir2_data_entry *
+xfs_dir3_data_entry_p(struct xfs_dir2_data_hdr *hdr)
+{
+	return (struct xfs_dir2_data_entry *)
+		((char *)hdr + sizeof(struct xfs_dir3_data_hdr));
+}
+
+static struct xfs_dir2_data_unused *
+xfs_dir3_data_unused_p(struct xfs_dir2_data_hdr *hdr)
+{
+	return (struct xfs_dir2_data_unused *)
+		((char *)hdr + sizeof(struct xfs_dir3_data_hdr));
+}
+
+
+/*
+ * Directory Leaf block operations
+ */
+static int
+xfs_dir2_max_leaf_ents(struct xfs_mount *mp)
+{
+	return (mp->m_dirblksize - sizeof(struct xfs_dir2_leaf_hdr)) /
+		(uint)sizeof(struct xfs_dir2_leaf_entry);
+}
+
+static struct xfs_dir2_leaf_entry *
+xfs_dir2_leaf_ents_p(struct xfs_dir2_leaf *lp)
+{
+	return lp->__ents;
+}
+
+static int
+xfs_dir3_max_leaf_ents(struct xfs_mount *mp)
+{
+	return (mp->m_dirblksize - sizeof(struct xfs_dir3_leaf_hdr)) /
+		(uint)sizeof(struct xfs_dir2_leaf_entry);
+}
+
+static struct xfs_dir2_leaf_entry *
+xfs_dir3_leaf_ents_p(struct xfs_dir2_leaf *lp)
+{
+	return ((struct xfs_dir3_leaf *)lp)->__ents;
+}
+
+static void
+xfs_dir2_leaf_hdr_from_disk(
+	struct xfs_dir3_icleaf_hdr	*to,
+	struct xfs_dir2_leaf		*from)
+{
+	to->forw = be32_to_cpu(from->hdr.info.forw);
+	to->back = be32_to_cpu(from->hdr.info.back);
+	to->magic = be16_to_cpu(from->hdr.info.magic);
+	to->count = be16_to_cpu(from->hdr.count);
+	to->stale = be16_to_cpu(from->hdr.stale);
+
+	ASSERT(to->magic == XFS_DIR2_LEAF1_MAGIC ||
+	       to->magic == XFS_DIR2_LEAFN_MAGIC);
+}
+
+static void
+xfs_dir2_leaf_hdr_to_disk(
+	struct xfs_dir2_leaf		*to,
+	struct xfs_dir3_icleaf_hdr	*from)
+{
+	ASSERT(from->magic == XFS_DIR2_LEAF1_MAGIC ||
+	       from->magic == XFS_DIR2_LEAFN_MAGIC);
+
+	to->hdr.info.forw = cpu_to_be32(from->forw);
+	to->hdr.info.back = cpu_to_be32(from->back);
+	to->hdr.info.magic = cpu_to_be16(from->magic);
+	to->hdr.count = cpu_to_be16(from->count);
+	to->hdr.stale = cpu_to_be16(from->stale);
+}
+
+static void
+xfs_dir3_leaf_hdr_from_disk(
+	struct xfs_dir3_icleaf_hdr	*to,
+	struct xfs_dir2_leaf		*from)
+{
+	struct xfs_dir3_leaf_hdr *hdr3 = (struct xfs_dir3_leaf_hdr *)from;
+
+	to->forw = be32_to_cpu(hdr3->info.hdr.forw);
+	to->back = be32_to_cpu(hdr3->info.hdr.back);
+	to->magic = be16_to_cpu(hdr3->info.hdr.magic);
+	to->count = be16_to_cpu(hdr3->count);
+	to->stale = be16_to_cpu(hdr3->stale);
+
+	ASSERT(to->magic == XFS_DIR3_LEAF1_MAGIC ||
+	       to->magic == XFS_DIR3_LEAFN_MAGIC);
+}
+
+static void
+xfs_dir3_leaf_hdr_to_disk(
+	struct xfs_dir2_leaf		*to,
+	struct xfs_dir3_icleaf_hdr	*from)
+{
+	struct xfs_dir3_leaf_hdr *hdr3 = (struct xfs_dir3_leaf_hdr *)to;
+
+	ASSERT(from->magic == XFS_DIR3_LEAF1_MAGIC ||
+	       from->magic == XFS_DIR3_LEAFN_MAGIC);
+
+	hdr3->info.hdr.forw = cpu_to_be32(from->forw);
+	hdr3->info.hdr.back = cpu_to_be32(from->back);
+	hdr3->info.hdr.magic = cpu_to_be16(from->magic);
+	hdr3->count = cpu_to_be16(from->count);
+	hdr3->stale = cpu_to_be16(from->stale);
+}
+
+
+/*
+ * Directory/Attribute Node block operations
+ */
+static struct xfs_da_node_entry *
+xfs_da2_node_tree_p(struct xfs_da_intnode *dap)
+{
+	return dap->__btree;
+}
+
+static struct xfs_da_node_entry *
+xfs_da3_node_tree_p(struct xfs_da_intnode *dap)
+{
+	return ((struct xfs_da3_intnode *)dap)->__btree;
+}
+
+static void
+xfs_da2_node_hdr_from_disk(
+	struct xfs_da3_icnode_hdr	*to,
+	struct xfs_da_intnode		*from)
+{
+	ASSERT(from->hdr.info.magic == cpu_to_be16(XFS_DA_NODE_MAGIC));
+	to->forw = be32_to_cpu(from->hdr.info.forw);
+	to->back = be32_to_cpu(from->hdr.info.back);
+	to->magic = be16_to_cpu(from->hdr.info.magic);
+	to->count = be16_to_cpu(from->hdr.__count);
+	to->level = be16_to_cpu(from->hdr.__level);
+}
+
+static void
+xfs_da2_node_hdr_to_disk(
+	struct xfs_da_intnode		*to,
+	struct xfs_da3_icnode_hdr	*from)
+{
+	ASSERT(from->magic == XFS_DA_NODE_MAGIC);
+	to->hdr.info.forw = cpu_to_be32(from->forw);
+	to->hdr.info.back = cpu_to_be32(from->back);
+	to->hdr.info.magic = cpu_to_be16(from->magic);
+	to->hdr.__count = cpu_to_be16(from->count);
+	to->hdr.__level = cpu_to_be16(from->level);
+}
+
+static void
+xfs_da3_node_hdr_from_disk(
+	struct xfs_da3_icnode_hdr	*to,
+	struct xfs_da_intnode		*from)
+{
+	struct xfs_da3_node_hdr *hdr3 = (struct xfs_da3_node_hdr *)from;
+
+	ASSERT(from->hdr.info.magic == cpu_to_be16(XFS_DA3_NODE_MAGIC));
+	to->forw = be32_to_cpu(hdr3->info.hdr.forw);
+	to->back = be32_to_cpu(hdr3->info.hdr.back);
+	to->magic = be16_to_cpu(hdr3->info.hdr.magic);
+	to->count = be16_to_cpu(hdr3->__count);
+	to->level = be16_to_cpu(hdr3->__level);
+}
+
+static void
+xfs_da3_node_hdr_to_disk(
+	struct xfs_da_intnode		*to,
+	struct xfs_da3_icnode_hdr	*from)
+{
+	struct xfs_da3_node_hdr *hdr3 = (struct xfs_da3_node_hdr *)to;
+
+	ASSERT(from->magic == XFS_DA3_NODE_MAGIC);
+	hdr3->info.hdr.forw = cpu_to_be32(from->forw);
+	hdr3->info.hdr.back = cpu_to_be32(from->back);
+	hdr3->info.hdr.magic = cpu_to_be16(from->magic);
+	hdr3->__count = cpu_to_be16(from->count);
+	hdr3->__level = cpu_to_be16(from->level);
+}
+
+
+/*
+ * Directory free space block operations
+ */
+static int
+xfs_dir2_free_max_bests(struct xfs_mount *mp)
+{
+	return (mp->m_dirblksize - sizeof(struct xfs_dir2_free_hdr)) /
+		sizeof(xfs_dir2_data_off_t);
+}
+
+static __be16 *
+xfs_dir2_free_bests_p(struct xfs_dir2_free *free)
+{
+	return (__be16 *)((char *)free + sizeof(struct xfs_dir2_free_hdr));
+}
+
+/*
+ * Convert data space db to the corresponding free db.
+ */
+static xfs_dir2_db_t
+xfs_dir2_db_to_fdb(struct xfs_mount *mp, xfs_dir2_db_t db)
+{
+	return XFS_DIR2_FREE_FIRSTDB(mp) + db / xfs_dir2_free_max_bests(mp);
+}
+
+/*
+ * Convert data space db to the corresponding index in a free db.
+ */
+static int
+xfs_dir2_db_to_fdindex(struct xfs_mount *mp, xfs_dir2_db_t db)
+{
+	return db % xfs_dir2_free_max_bests(mp);
+}
+
+static int
+xfs_dir3_free_max_bests(struct xfs_mount *mp)
+{
+	return (mp->m_dirblksize - sizeof(struct xfs_dir3_free_hdr)) /
+		sizeof(xfs_dir2_data_off_t);
+}
+
+static __be16 *
+xfs_dir3_free_bests_p(struct xfs_dir2_free *free)
+{
+	return (__be16 *)((char *)free + sizeof(struct xfs_dir3_free_hdr));
+}
+
+/*
+ * Convert data space db to the corresponding free db.
+ */
+static xfs_dir2_db_t
+xfs_dir3_db_to_fdb(struct xfs_mount *mp, xfs_dir2_db_t db)
+{
+	return XFS_DIR2_FREE_FIRSTDB(mp) + db / xfs_dir3_free_max_bests(mp);
+}
+
+/*
+ * Convert data space db to the corresponding index in a free db.
+ */
+static int
+xfs_dir3_db_to_fdindex(struct xfs_mount *mp, xfs_dir2_db_t db)
+{
+	return db % xfs_dir3_free_max_bests(mp);
+}
+
+static void
+xfs_dir2_free_hdr_from_disk(
+	struct xfs_dir3_icfree_hdr	*to,
+	struct xfs_dir2_free		*from)
+{
+	to->magic = be32_to_cpu(from->hdr.magic);
+	to->firstdb = be32_to_cpu(from->hdr.firstdb);
+	to->nvalid = be32_to_cpu(from->hdr.nvalid);
+	to->nused = be32_to_cpu(from->hdr.nused);
+	ASSERT(to->magic == XFS_DIR2_FREE_MAGIC);
+}
+
+static void
+xfs_dir2_free_hdr_to_disk(
+	struct xfs_dir2_free		*to,
+	struct xfs_dir3_icfree_hdr	*from)
+{
+	ASSERT(from->magic == XFS_DIR2_FREE_MAGIC);
+
+	to->hdr.magic = cpu_to_be32(from->magic);
+	to->hdr.firstdb = cpu_to_be32(from->firstdb);
+	to->hdr.nvalid = cpu_to_be32(from->nvalid);
+	to->hdr.nused = cpu_to_be32(from->nused);
+}
+
+static void
+xfs_dir3_free_hdr_from_disk(
+	struct xfs_dir3_icfree_hdr	*to,
+	struct xfs_dir2_free		*from)
+{
+	struct xfs_dir3_free_hdr *hdr3 = (struct xfs_dir3_free_hdr *)from;
+
+	to->magic = be32_to_cpu(hdr3->hdr.magic);
+	to->firstdb = be32_to_cpu(hdr3->firstdb);
+	to->nvalid = be32_to_cpu(hdr3->nvalid);
+	to->nused = be32_to_cpu(hdr3->nused);
+
+	ASSERT(to->magic == XFS_DIR3_FREE_MAGIC);
+}
+
+static void
+xfs_dir3_free_hdr_to_disk(
+	struct xfs_dir2_free		*to,
+	struct xfs_dir3_icfree_hdr	*from)
+{
+	struct xfs_dir3_free_hdr *hdr3 = (struct xfs_dir3_free_hdr *)to;
+
+	ASSERT(from->magic == XFS_DIR3_FREE_MAGIC);
+
+	hdr3->hdr.magic = cpu_to_be32(from->magic);
+	hdr3->firstdb = cpu_to_be32(from->firstdb);
+	hdr3->nvalid = cpu_to_be32(from->nvalid);
+	hdr3->nused = cpu_to_be32(from->nused);
+}
+
+static const struct xfs_dir_ops xfs_dir2_ops = {
+	.sf_entsize = xfs_dir2_sf_entsize,
+	.sf_nextentry = xfs_dir2_sf_nextentry,
+	.sf_get_ftype = xfs_dir2_sfe_get_ftype,
+	.sf_put_ftype = xfs_dir2_sfe_put_ftype,
+	.sf_get_ino = xfs_dir2_sfe_get_ino,
+	.sf_put_ino = xfs_dir2_sfe_put_ino,
+	.sf_get_parent_ino = xfs_dir2_sf_get_parent_ino,
+	.sf_put_parent_ino = xfs_dir2_sf_put_parent_ino,
+
+	.data_entsize = xfs_dir2_data_entsize,
+	.data_get_ftype = xfs_dir2_data_get_ftype,
+	.data_put_ftype = xfs_dir2_data_put_ftype,
+	.data_entry_tag_p = xfs_dir2_data_entry_tag_p,
+	.data_bestfree_p = xfs_dir2_data_bestfree_p,
+
+	.data_dot_offset = sizeof(struct xfs_dir2_data_hdr),
+	.data_dotdot_offset = sizeof(struct xfs_dir2_data_hdr) +
+				XFS_DIR2_DATA_ENTSIZE(1),
+	.data_first_offset =  sizeof(struct xfs_dir2_data_hdr) +
+				XFS_DIR2_DATA_ENTSIZE(1) +
+				XFS_DIR2_DATA_ENTSIZE(2),
+	.data_entry_offset = sizeof(struct xfs_dir2_data_hdr),
+
+	.data_dot_entry_p = xfs_dir2_data_dot_entry_p,
+	.data_dotdot_entry_p = xfs_dir2_data_dotdot_entry_p,
+	.data_first_entry_p = xfs_dir2_data_first_entry_p,
+	.data_entry_p = xfs_dir2_data_entry_p,
+	.data_unused_p = xfs_dir2_data_unused_p,
+
+	.leaf_hdr_size = sizeof(struct xfs_dir2_leaf_hdr),
+	.leaf_hdr_to_disk = xfs_dir2_leaf_hdr_to_disk,
+	.leaf_hdr_from_disk = xfs_dir2_leaf_hdr_from_disk,
+	.leaf_max_ents = xfs_dir2_max_leaf_ents,
+	.leaf_ents_p = xfs_dir2_leaf_ents_p,
+
+	.node_hdr_size = sizeof(struct xfs_da_node_hdr),
+	.node_hdr_to_disk = xfs_da2_node_hdr_to_disk,
+	.node_hdr_from_disk = xfs_da2_node_hdr_from_disk,
+	.node_tree_p = xfs_da2_node_tree_p,
+
+	.free_hdr_size = sizeof(struct xfs_dir2_free_hdr),
+	.free_hdr_to_disk = xfs_dir2_free_hdr_to_disk,
+	.free_hdr_from_disk = xfs_dir2_free_hdr_from_disk,
+	.free_max_bests = xfs_dir2_free_max_bests,
+	.free_bests_p = xfs_dir2_free_bests_p,
+	.db_to_fdb = xfs_dir2_db_to_fdb,
+	.db_to_fdindex = xfs_dir2_db_to_fdindex,
+};
+
+static const struct xfs_dir_ops xfs_dir2_ftype_ops = {
+	.sf_entsize = xfs_dir3_sf_entsize,
+	.sf_nextentry = xfs_dir3_sf_nextentry,
+	.sf_get_ftype = xfs_dir3_sfe_get_ftype,
+	.sf_put_ftype = xfs_dir3_sfe_put_ftype,
+	.sf_get_ino = xfs_dir3_sfe_get_ino,
+	.sf_put_ino = xfs_dir3_sfe_put_ino,
+	.sf_get_parent_ino = xfs_dir2_sf_get_parent_ino,
+	.sf_put_parent_ino = xfs_dir2_sf_put_parent_ino,
+
+	.data_entsize = xfs_dir3_data_entsize,
+	.data_get_ftype = xfs_dir3_data_get_ftype,
+	.data_put_ftype = xfs_dir3_data_put_ftype,
+	.data_entry_tag_p = xfs_dir3_data_entry_tag_p,
+	.data_bestfree_p = xfs_dir2_data_bestfree_p,
+
+	.data_dot_offset = sizeof(struct xfs_dir2_data_hdr),
+	.data_dotdot_offset = sizeof(struct xfs_dir2_data_hdr) +
+				XFS_DIR3_DATA_ENTSIZE(1),
+	.data_first_offset =  sizeof(struct xfs_dir2_data_hdr) +
+				XFS_DIR3_DATA_ENTSIZE(1) +
+				XFS_DIR3_DATA_ENTSIZE(2),
+	.data_entry_offset = sizeof(struct xfs_dir2_data_hdr),
+
+	.data_dot_entry_p = xfs_dir2_data_dot_entry_p,
+	.data_dotdot_entry_p = xfs_dir2_ftype_data_dotdot_entry_p,
+	.data_first_entry_p = xfs_dir2_ftype_data_first_entry_p,
+	.data_entry_p = xfs_dir2_data_entry_p,
+	.data_unused_p = xfs_dir2_data_unused_p,
+
+	.leaf_hdr_size = sizeof(struct xfs_dir2_leaf_hdr),
+	.leaf_hdr_to_disk = xfs_dir2_leaf_hdr_to_disk,
+	.leaf_hdr_from_disk = xfs_dir2_leaf_hdr_from_disk,
+	.leaf_max_ents = xfs_dir2_max_leaf_ents,
+	.leaf_ents_p = xfs_dir2_leaf_ents_p,
+
+	.node_hdr_size = sizeof(struct xfs_da_node_hdr),
+	.node_hdr_to_disk = xfs_da2_node_hdr_to_disk,
+	.node_hdr_from_disk = xfs_da2_node_hdr_from_disk,
+	.node_tree_p = xfs_da2_node_tree_p,
+
+	.free_hdr_size = sizeof(struct xfs_dir2_free_hdr),
+	.free_hdr_to_disk = xfs_dir2_free_hdr_to_disk,
+	.free_hdr_from_disk = xfs_dir2_free_hdr_from_disk,
+	.free_max_bests = xfs_dir2_free_max_bests,
+	.free_bests_p = xfs_dir2_free_bests_p,
+	.db_to_fdb = xfs_dir2_db_to_fdb,
+	.db_to_fdindex = xfs_dir2_db_to_fdindex,
+};
+
+static const struct xfs_dir_ops xfs_dir3_ops = {
+	.sf_entsize = xfs_dir3_sf_entsize,
+	.sf_nextentry = xfs_dir3_sf_nextentry,
+	.sf_get_ftype = xfs_dir3_sfe_get_ftype,
+	.sf_put_ftype = xfs_dir3_sfe_put_ftype,
+	.sf_get_ino = xfs_dir3_sfe_get_ino,
+	.sf_put_ino = xfs_dir3_sfe_put_ino,
+	.sf_get_parent_ino = xfs_dir2_sf_get_parent_ino,
+	.sf_put_parent_ino = xfs_dir2_sf_put_parent_ino,
+
+	.data_entsize = xfs_dir3_data_entsize,
+	.data_get_ftype = xfs_dir3_data_get_ftype,
+	.data_put_ftype = xfs_dir3_data_put_ftype,
+	.data_entry_tag_p = xfs_dir3_data_entry_tag_p,
+	.data_bestfree_p = xfs_dir3_data_bestfree_p,
+
+	.data_dot_offset = sizeof(struct xfs_dir3_data_hdr),
+	.data_dotdot_offset = sizeof(struct xfs_dir3_data_hdr) +
+				XFS_DIR3_DATA_ENTSIZE(1),
+	.data_first_offset =  sizeof(struct xfs_dir3_data_hdr) +
+				XFS_DIR3_DATA_ENTSIZE(1) +
+				XFS_DIR3_DATA_ENTSIZE(2),
+	.data_entry_offset = sizeof(struct xfs_dir3_data_hdr),
+
+	.data_dot_entry_p = xfs_dir3_data_dot_entry_p,
+	.data_dotdot_entry_p = xfs_dir3_data_dotdot_entry_p,
+	.data_first_entry_p = xfs_dir3_data_first_entry_p,
+	.data_entry_p = xfs_dir3_data_entry_p,
+	.data_unused_p = xfs_dir3_data_unused_p,
+
+	.leaf_hdr_size = sizeof(struct xfs_dir3_leaf_hdr),
+	.leaf_hdr_to_disk = xfs_dir3_leaf_hdr_to_disk,
+	.leaf_hdr_from_disk = xfs_dir3_leaf_hdr_from_disk,
+	.leaf_max_ents = xfs_dir3_max_leaf_ents,
+	.leaf_ents_p = xfs_dir3_leaf_ents_p,
+
+	.node_hdr_size = sizeof(struct xfs_da3_node_hdr),
+	.node_hdr_to_disk = xfs_da3_node_hdr_to_disk,
+	.node_hdr_from_disk = xfs_da3_node_hdr_from_disk,
+	.node_tree_p = xfs_da3_node_tree_p,
+
+	.free_hdr_size = sizeof(struct xfs_dir3_free_hdr),
+	.free_hdr_to_disk = xfs_dir3_free_hdr_to_disk,
+	.free_hdr_from_disk = xfs_dir3_free_hdr_from_disk,
+	.free_max_bests = xfs_dir3_free_max_bests,
+	.free_bests_p = xfs_dir3_free_bests_p,
+	.db_to_fdb = xfs_dir3_db_to_fdb,
+	.db_to_fdindex = xfs_dir3_db_to_fdindex,
+};
+
+static const struct xfs_dir_ops xfs_dir2_nondir_ops = {
+	.node_hdr_size = sizeof(struct xfs_da_node_hdr),
+	.node_hdr_to_disk = xfs_da2_node_hdr_to_disk,
+	.node_hdr_from_disk = xfs_da2_node_hdr_from_disk,
+	.node_tree_p = xfs_da2_node_tree_p,
+};
+
+static const struct xfs_dir_ops xfs_dir3_nondir_ops = {
+	.node_hdr_size = sizeof(struct xfs_da3_node_hdr),
+	.node_hdr_to_disk = xfs_da3_node_hdr_to_disk,
+	.node_hdr_from_disk = xfs_da3_node_hdr_from_disk,
+	.node_tree_p = xfs_da3_node_tree_p,
+};
+
+/*
+ * Return the ops structure according to the current config.  If we are passed
+ * an inode, then that overrides the default config we use which is based on
+ * feature bits.
+ */
+const struct xfs_dir_ops *
+xfs_dir_get_ops(
+	struct xfs_mount	*mp,
+	struct xfs_inode	*dp)
+{
+	if (dp)
+		return dp->d_ops;
+	if (mp->m_dir_inode_ops)
+		return mp->m_dir_inode_ops;
+	if (xfs_sb_version_hascrc(&mp->m_sb))
+		return &xfs_dir3_ops;
+	if (xfs_sb_version_hasftype(&mp->m_sb))
+		return &xfs_dir2_ftype_ops;
+	return &xfs_dir2_ops;
+}
+
+const struct xfs_dir_ops *
+xfs_nondir_get_ops(
+	struct xfs_mount	*mp,
+	struct xfs_inode	*dp)
+{
+	if (dp)
+		return dp->d_ops;
+	if (mp->m_nondir_inode_ops)
+		return mp->m_nondir_inode_ops;
+	if (xfs_sb_version_hascrc(&mp->m_sb))
+		return &xfs_dir3_nondir_ops;
+	return &xfs_dir2_nondir_ops;
+}
diff --git a/fs/xfs/xfs_dir2_format.h b/fs/xfs/xfs_da_format.h
index 9cf67381adf6..a19d3f8f639c 100644
--- a/fs/xfs/xfs_dir2_format.h
+++ b/fs/xfs/xfs_da_format.h
@@ -16,8 +16,113 @@
  * along with this program; if not, write the Free Software Foundation,
  * Inc.,  51 Franklin St, Fifth Floor, Boston, MA  02110-1301  USA
  */
-#ifndef __XFS_DIR2_FORMAT_H__
-#define __XFS_DIR2_FORMAT_H__
+#ifndef __XFS_DA_FORMAT_H__
+#define __XFS_DA_FORMAT_H__
+
+/*========================================================================
+ * Directory Structure when greater than XFS_LBSIZE(mp) bytes.
+ *========================================================================*/
+
+/*
+ * This structure is common to both leaf nodes and non-leaf nodes in the Btree.
+ *
+ * It is used to manage a doubly linked list of all blocks at the same
+ * level in the Btree, and to identify which type of block this is.
+ */
+#define XFS_DA_NODE_MAGIC	0xfebe	/* magic number: non-leaf blocks */
+#define XFS_ATTR_LEAF_MAGIC	0xfbee	/* magic number: attribute leaf blks */
+#define	XFS_DIR2_LEAF1_MAGIC	0xd2f1	/* magic number: v2 dirlf single blks */
+#define	XFS_DIR2_LEAFN_MAGIC	0xd2ff	/* magic number: v2 dirlf multi blks */
+
+typedef struct xfs_da_blkinfo {
+	__be32		forw;			/* previous block in list */
+	__be32		back;			/* following block in list */
+	__be16		magic;			/* validity check on block */
+	__be16		pad;			/* unused */
+} xfs_da_blkinfo_t;
+
+/*
+ * CRC enabled directory structure types
+ *
+ * The headers change size for the additional verification information, but
+ * otherwise the tree layouts and contents are unchanged. Hence the da btree
+ * code can use the struct xfs_da_blkinfo for manipulating the tree links and
+ * magic numbers without modification for both v2 and v3 nodes.
+ */
+#define XFS_DA3_NODE_MAGIC	0x3ebe	/* magic number: non-leaf blocks */
+#define XFS_ATTR3_LEAF_MAGIC	0x3bee	/* magic number: attribute leaf blks */
+#define	XFS_DIR3_LEAF1_MAGIC	0x3df1	/* magic number: v2 dirlf single blks */
+#define	XFS_DIR3_LEAFN_MAGIC	0x3dff	/* magic number: v2 dirlf multi blks */
+
+struct xfs_da3_blkinfo {
+	/*
+	 * the node link manipulation code relies on the fact that the first
+	 * element of this structure is the struct xfs_da_blkinfo so it can
+	 * ignore the differences in the rest of the structures.
+	 */
+	struct xfs_da_blkinfo	hdr;
+	__be32			crc;	/* CRC of block */
+	__be64			blkno;	/* first block of the buffer */
+	__be64			lsn;	/* sequence number of last write */
+	uuid_t			uuid;	/* filesystem we belong to */
+	__be64			owner;	/* inode that owns the block */
+};
+
+/*
+ * This is the structure of the root and intermediate nodes in the Btree.
+ * The leaf nodes are defined above.
+ *
+ * Entries are not packed.
+ *
+ * Since we have duplicate keys, use a binary search but always follow
+ * all match in the block, not just the first match found.
+ */
+#define	XFS_DA_NODE_MAXDEPTH	5	/* max depth of Btree */
+
+typedef struct xfs_da_node_hdr {
+	struct xfs_da_blkinfo	info;	/* block type, links, etc. */
+	__be16			__count; /* count of active entries */
+	__be16			__level; /* level above leaves (leaf == 0) */
+} xfs_da_node_hdr_t;
+
+struct xfs_da3_node_hdr {
+	struct xfs_da3_blkinfo	info;	/* block type, links, etc. */
+	__be16			__count; /* count of active entries */
+	__be16			__level; /* level above leaves (leaf == 0) */
+	__be32			__pad32;
+};
+
+#define XFS_DA3_NODE_CRC_OFF	(offsetof(struct xfs_da3_node_hdr, info.crc))
+
+typedef struct xfs_da_node_entry {
+	__be32	hashval;	/* hash value for this descendant */
+	__be32	before;		/* Btree block before this key */
+} xfs_da_node_entry_t;
+
+typedef struct xfs_da_intnode {
+	struct xfs_da_node_hdr	hdr;
+	struct xfs_da_node_entry __btree[];
+} xfs_da_intnode_t;
+
+struct xfs_da3_intnode {
+	struct xfs_da3_node_hdr	hdr;
+	struct xfs_da_node_entry __btree[];
+};
+
+/*
+ * In-core version of the node header to abstract the differences in the v2 and
+ * v3 disk format of the headers. Callers need to convert to/from disk format as
+ * appropriate.
+ */
+struct xfs_da3_icnode_hdr {
+	__uint32_t	forw;
+	__uint32_t	back;
+	__uint16_t	magic;
+	__uint16_t	count;
+	__uint16_t	level;
+};
+
+#define	XFS_LBSIZE(mp)	(mp)->m_sb.sb_blocksize
 
 /*
  * Directory version 2.
@@ -189,79 +294,6 @@ xfs_dir2_sf_firstentry(struct xfs_dir2_sf_hdr *hdr)
 		((char *)hdr + xfs_dir2_sf_hdr_size(hdr->i8count));
 }
 
-static inline int
-xfs_dir3_sf_entsize(
-	struct xfs_mount	*mp,
-	struct xfs_dir2_sf_hdr	*hdr,
-	int			len)
-{
-	int count = sizeof(struct xfs_dir2_sf_entry); 	/* namelen + offset */
-
-	count += len;					/* name */
-	count += hdr->i8count ? sizeof(xfs_dir2_ino8_t) :
-				sizeof(xfs_dir2_ino4_t); /* ino # */
-	if (xfs_sb_version_hasftype(&mp->m_sb))
-		count += sizeof(__uint8_t);		/* file type */
-	return count;
-}
-
-static inline struct xfs_dir2_sf_entry *
-xfs_dir3_sf_nextentry(
-	struct xfs_mount	*mp,
-	struct xfs_dir2_sf_hdr	*hdr,
-	struct xfs_dir2_sf_entry *sfep)
-{
-	return (struct xfs_dir2_sf_entry *)
-		((char *)sfep + xfs_dir3_sf_entsize(mp, hdr, sfep->namelen));
-}
-
-/*
- * in dir3 shortform directories, the file type field is stored at a variable
- * offset after the inode number. Because it's only a single byte, endian
- * conversion is not necessary.
- */
-static inline __uint8_t *
-xfs_dir3_sfe_ftypep(
-	struct xfs_dir2_sf_hdr	*hdr,
-	struct xfs_dir2_sf_entry *sfep)
-{
-	return (__uint8_t *)&sfep->name[sfep->namelen];
-}
-
-static inline __uint8_t
-xfs_dir3_sfe_get_ftype(
-	struct xfs_mount	*mp,
-	struct xfs_dir2_sf_hdr	*hdr,
-	struct xfs_dir2_sf_entry *sfep)
-{
-	__uint8_t	*ftp;
-
-	if (!xfs_sb_version_hasftype(&mp->m_sb))
-		return XFS_DIR3_FT_UNKNOWN;
-
-	ftp = xfs_dir3_sfe_ftypep(hdr, sfep);
-	if (*ftp >= XFS_DIR3_FT_MAX)
-		return XFS_DIR3_FT_UNKNOWN;
-	return *ftp;
-}
-
-static inline void
-xfs_dir3_sfe_put_ftype(
-	struct xfs_mount	*mp,
-	struct xfs_dir2_sf_hdr	*hdr,
-	struct xfs_dir2_sf_entry *sfep,
-	__uint8_t		ftype)
-{
-	__uint8_t	*ftp;
-
-	ASSERT(ftype < XFS_DIR3_FT_MAX);
-
-	if (!xfs_sb_version_hasftype(&mp->m_sb))
-		return;
-	ftp = xfs_dir3_sfe_ftypep(hdr, sfep);
-	*ftp = ftype;
-}
-
 /*
  * Data block structures.
  *
@@ -345,17 +377,6 @@ struct xfs_dir3_data_hdr {
 
 #define XFS_DIR3_DATA_CRC_OFF  offsetof(struct xfs_dir3_data_hdr, hdr.crc)
 
-static inline struct xfs_dir2_data_free *
-xfs_dir3_data_bestfree_p(struct xfs_dir2_data_hdr *hdr)
-{
-	if (hdr->magic == cpu_to_be32(XFS_DIR3_DATA_MAGIC) ||
-	    hdr->magic == cpu_to_be32(XFS_DIR3_BLOCK_MAGIC)) {
-		struct xfs_dir3_data_hdr *hdr3 = (struct xfs_dir3_data_hdr *)hdr;
-		return hdr3->best_free;
-	}
-	return hdr->bestfree;
-}
-
 /*
  * Active entry in a data block.
  *
@@ -389,72 +410,6 @@ typedef struct xfs_dir2_data_unused {
 } xfs_dir2_data_unused_t;
 
 /*
- * Size of a data entry.
- */
-static inline int
-__xfs_dir3_data_entsize(
-	bool	ftype,
-	int	n)
-{
-	int	size = offsetof(struct xfs_dir2_data_entry, name[0]);
-
-	size += n;
-	size += sizeof(xfs_dir2_data_off_t);
-	if (ftype)
-		size += sizeof(__uint8_t);
-	return roundup(size, XFS_DIR2_DATA_ALIGN);
-}
-static inline int
-xfs_dir3_data_entsize(
-	struct xfs_mount	*mp,
-	int			n)
-{
-	bool ftype = xfs_sb_version_hasftype(&mp->m_sb) ? true : false;
-	return __xfs_dir3_data_entsize(ftype, n);
-}
-
-static inline __uint8_t
-xfs_dir3_dirent_get_ftype(
-	struct xfs_mount	*mp,
-	struct xfs_dir2_data_entry *dep)
-{
-	if (xfs_sb_version_hasftype(&mp->m_sb)) {
-		__uint8_t	type = dep->name[dep->namelen];
-
-		ASSERT(type < XFS_DIR3_FT_MAX);
-		if (type < XFS_DIR3_FT_MAX)
-			return type;
-
-	}
-	return XFS_DIR3_FT_UNKNOWN;
-}
-
-static inline void
-xfs_dir3_dirent_put_ftype(
-	struct xfs_mount	*mp,
-	struct xfs_dir2_data_entry *dep,
-	__uint8_t		type)
-{
-	ASSERT(type < XFS_DIR3_FT_MAX);
-	ASSERT(dep->namelen != 0);
-
-	if (xfs_sb_version_hasftype(&mp->m_sb))
-		dep->name[dep->namelen] = type;
-}
-
-/*
- * Pointer to an entry's tag word.
- */
-static inline __be16 *
-xfs_dir3_data_entry_tag_p(
-	struct xfs_mount	*mp,
-	struct xfs_dir2_data_entry *dep)
-{
-	return (__be16 *)((char *)dep +
-		xfs_dir3_data_entsize(mp, dep->namelen) - sizeof(__be16));
-}
-
-/*
  * Pointer to a freespace's tag word.
  */
 static inline __be16 *
@@ -464,93 +419,6 @@ xfs_dir2_data_unused_tag_p(struct xfs_dir2_data_unused *dup)
 			be16_to_cpu(dup->length) - sizeof(__be16));
 }
 
-static inline size_t
-xfs_dir3_data_hdr_size(bool dir3)
-{
-	if (dir3)
-		return sizeof(struct xfs_dir3_data_hdr);
-	return sizeof(struct xfs_dir2_data_hdr);
-}
-
-static inline size_t
-xfs_dir3_data_entry_offset(struct xfs_dir2_data_hdr *hdr)
-{
-	bool dir3 = hdr->magic == cpu_to_be32(XFS_DIR3_DATA_MAGIC) ||
-		    hdr->magic == cpu_to_be32(XFS_DIR3_BLOCK_MAGIC);
-	return xfs_dir3_data_hdr_size(dir3);
-}
-
-static inline struct xfs_dir2_data_entry *
-xfs_dir3_data_entry_p(struct xfs_dir2_data_hdr *hdr)
-{
-	return (struct xfs_dir2_data_entry *)
-		((char *)hdr + xfs_dir3_data_entry_offset(hdr));
-}
-
-static inline struct xfs_dir2_data_unused *
-xfs_dir3_data_unused_p(struct xfs_dir2_data_hdr *hdr)
-{
-	return (struct xfs_dir2_data_unused *)
-		((char *)hdr + xfs_dir3_data_entry_offset(hdr));
-}
-
-/*
- * Offsets of . and .. in data space (always block 0)
- *
- * XXX: there is scope for significant optimisation of the logic here. Right
- * now we are checking for "dir3 format" over and over again. Ideally we should
- * only do it once for each operation.
- */
-static inline xfs_dir2_data_aoff_t
-xfs_dir3_data_dot_offset(struct xfs_mount *mp)
-{
-	return xfs_dir3_data_hdr_size(xfs_sb_version_hascrc(&mp->m_sb));
-}
-
-static inline xfs_dir2_data_aoff_t
-xfs_dir3_data_dotdot_offset(struct xfs_mount *mp)
-{
-	return xfs_dir3_data_dot_offset(mp) +
-		xfs_dir3_data_entsize(mp, 1);
-}
-
-static inline xfs_dir2_data_aoff_t
-xfs_dir3_data_first_offset(struct xfs_mount *mp)
-{
-	return xfs_dir3_data_dotdot_offset(mp) +
-		xfs_dir3_data_entsize(mp, 2);
-}
-
-/*
- * location of . and .. in data space (always block 0)
- */
-static inline struct xfs_dir2_data_entry *
-xfs_dir3_data_dot_entry_p(
-	struct xfs_mount	*mp,
-	struct xfs_dir2_data_hdr *hdr)
-{
-	return (struct xfs_dir2_data_entry *)
-		((char *)hdr + xfs_dir3_data_dot_offset(mp));
-}
-
-static inline struct xfs_dir2_data_entry *
-xfs_dir3_data_dotdot_entry_p(
-	struct xfs_mount	*mp,
-	struct xfs_dir2_data_hdr *hdr)
-{
-	return (struct xfs_dir2_data_entry *)
-		((char *)hdr + xfs_dir3_data_dotdot_offset(mp));
-}
-
-static inline struct xfs_dir2_data_entry *
-xfs_dir3_data_first_entry_p(
-	struct xfs_mount	*mp,
-	struct xfs_dir2_data_hdr *hdr)
-{
-	return (struct xfs_dir2_data_entry *)
-		((char *)hdr + xfs_dir3_data_first_offset(mp));
-}
-
 /*
  * Leaf block structures.
  *
@@ -645,39 +513,6 @@ struct xfs_dir3_leaf {
 
 #define XFS_DIR3_LEAF_CRC_OFF  offsetof(struct xfs_dir3_leaf_hdr, info.crc)
 
-extern void xfs_dir3_leaf_hdr_from_disk(struct xfs_dir3_icleaf_hdr *to,
-					struct xfs_dir2_leaf *from);
-
-static inline int
-xfs_dir3_leaf_hdr_size(struct xfs_dir2_leaf *lp)
-{
-	if (lp->hdr.info.magic == cpu_to_be16(XFS_DIR3_LEAF1_MAGIC) ||
-	    lp->hdr.info.magic == cpu_to_be16(XFS_DIR3_LEAFN_MAGIC))
-		return sizeof(struct xfs_dir3_leaf_hdr);
-	return sizeof(struct xfs_dir2_leaf_hdr);
-}
-
-static inline int
-xfs_dir3_max_leaf_ents(struct xfs_mount *mp, struct xfs_dir2_leaf *lp)
-{
-	return (mp->m_dirblksize - xfs_dir3_leaf_hdr_size(lp)) /
-		(uint)sizeof(struct xfs_dir2_leaf_entry);
-}
-
-/*
- * Get address of the bestcount field in the single-leaf block.
- */
-static inline struct xfs_dir2_leaf_entry *
-xfs_dir3_leaf_ents_p(struct xfs_dir2_leaf *lp)
-{
-	if (lp->hdr.info.magic == cpu_to_be16(XFS_DIR3_LEAF1_MAGIC) ||
-	    lp->hdr.info.magic == cpu_to_be16(XFS_DIR3_LEAFN_MAGIC)) {
-		struct xfs_dir3_leaf *lp3 = (struct xfs_dir3_leaf *)lp;
-		return lp3->__ents;
-	}
-	return lp->__ents;
-}
-
 /*
  * Get address of the bestcount field in the single-leaf block.
  */
@@ -869,48 +704,6 @@ struct xfs_dir3_icfree_hdr {
 
 };
 
-void xfs_dir3_free_hdr_from_disk(struct xfs_dir3_icfree_hdr *to,
-				 struct xfs_dir2_free *from);
-
-static inline int
-xfs_dir3_free_hdr_size(struct xfs_mount *mp)
-{
-	if (xfs_sb_version_hascrc(&mp->m_sb))
-		return sizeof(struct xfs_dir3_free_hdr);
-	return sizeof(struct xfs_dir2_free_hdr);
-}
-
-static inline int
-xfs_dir3_free_max_bests(struct xfs_mount *mp)
-{
-	return (mp->m_dirblksize - xfs_dir3_free_hdr_size(mp)) /
-		sizeof(xfs_dir2_data_off_t);
-}
-
-static inline __be16 *
-xfs_dir3_free_bests_p(struct xfs_mount *mp, struct xfs_dir2_free *free)
-{
-	return (__be16 *)((char *)free + xfs_dir3_free_hdr_size(mp));
-}
-
-/*
- * Convert data space db to the corresponding free db.
- */
-static inline xfs_dir2_db_t
-xfs_dir2_db_to_fdb(struct xfs_mount *mp, xfs_dir2_db_t db)
-{
-	return XFS_DIR2_FREE_FIRSTDB(mp) + db / xfs_dir3_free_max_bests(mp);
-}
-
-/*
- * Convert data space db to the corresponding index in a free db.
- */
-static inline int
-xfs_dir2_db_to_fdindex(struct xfs_mount *mp, xfs_dir2_db_t db)
-{
-	return db % xfs_dir3_free_max_bests(mp);
-}
-
 /*
  * Single block format.
  *
@@ -961,4 +754,262 @@ xfs_dir2_block_leaf_p(struct xfs_dir2_block_tail *btp)
 	return ((struct xfs_dir2_leaf_entry *)btp) - be32_to_cpu(btp->count);
 }
 
-#endif /* __XFS_DIR2_FORMAT_H__ */
+
+/*
+ * Attribute storage layout
+ *
+ * Attribute lists are structured around Btrees where all the data
+ * elements are in the leaf nodes.  Attribute names are hashed into an int,
+ * then that int is used as the index into the Btree.  Since the hashval
+ * of an attribute name may not be unique, we may have duplicate keys.  The
+ * internal links in the Btree are logical block offsets into the file.
+ *
+ *========================================================================
+ * Attribute structure when equal to XFS_LBSIZE(mp) bytes.
+ *========================================================================
+ *
+ * Struct leaf_entry's are packed from the top.  Name/values grow from the
+ * bottom but are not packed.  The freemap contains run-length-encoded entries
+ * for the free bytes after the leaf_entry's, but only the N largest such,
+ * smaller runs are dropped.  When the freemap doesn't show enough space
+ * for an allocation, we compact the name/value area and try again.  If we
+ * still don't have enough space, then we have to split the block.  The
+ * name/value structs (both local and remote versions) must be 32bit aligned.
+ *
+ * Since we have duplicate hash keys, for each key that matches, compare
+ * the actual name string.  The root and intermediate node search always
+ * takes the first-in-the-block key match found, so we should only have
+ * to work "forw"ard.  If none matches, continue with the "forw"ard leaf
+ * nodes until the hash key changes or the attribute name is found.
+ *
+ * We store the fact that an attribute is a ROOT/USER/SECURE attribute in
+ * the leaf_entry.  The namespaces are independent only because we also look
+ * at the namespace bit when we are looking for a matching attribute name.
+ *
+ * We also store an "incomplete" bit in the leaf_entry.  It shows that an
+ * attribute is in the middle of being created and should not be shown to
+ * the user if we crash during the time that the bit is set.  We clear the
+ * bit when we have finished setting up the attribute.  We do this because
+ * we cannot create some large attributes inside a single transaction, and we
+ * need some indication that we weren't finished if we crash in the middle.
+ */
+#define XFS_ATTR_LEAF_MAPSIZE	3	/* how many freespace slots */
+
+typedef struct xfs_attr_leaf_map {	/* RLE map of free bytes */
+	__be16	base;			  /* base of free region */
+	__be16	size;			  /* length of free region */
+} xfs_attr_leaf_map_t;
+
+typedef struct xfs_attr_leaf_hdr {	/* constant-structure header block */
+	xfs_da_blkinfo_t info;		/* block type, links, etc. */
+	__be16	count;			/* count of active leaf_entry's */
+	__be16	usedbytes;		/* num bytes of names/values stored */
+	__be16	firstused;		/* first used byte in name area */
+	__u8	holes;			/* != 0 if blk needs compaction */
+	__u8	pad1;
+	xfs_attr_leaf_map_t freemap[XFS_ATTR_LEAF_MAPSIZE];
+					/* N largest free regions */
+} xfs_attr_leaf_hdr_t;
+
+typedef struct xfs_attr_leaf_entry {	/* sorted on key, not name */
+	__be32	hashval;		/* hash value of name */
+	__be16	nameidx;		/* index into buffer of name/value */
+	__u8	flags;			/* LOCAL/ROOT/SECURE/INCOMPLETE flag */
+	__u8	pad2;			/* unused pad byte */
+} xfs_attr_leaf_entry_t;
+
+typedef struct xfs_attr_leaf_name_local {
+	__be16	valuelen;		/* number of bytes in value */
+	__u8	namelen;		/* length of name bytes */
+	__u8	nameval[1];		/* name/value bytes */
+} xfs_attr_leaf_name_local_t;
+
+typedef struct xfs_attr_leaf_name_remote {
+	__be32	valueblk;		/* block number of value bytes */
+	__be32	valuelen;		/* number of bytes in value */
+	__u8	namelen;		/* length of name bytes */
+	__u8	name[1];		/* name bytes */
+} xfs_attr_leaf_name_remote_t;
+
+typedef struct xfs_attr_leafblock {
+	xfs_attr_leaf_hdr_t	hdr;	/* constant-structure header block */
+	xfs_attr_leaf_entry_t	entries[1];	/* sorted on key, not name */
+	xfs_attr_leaf_name_local_t namelist;	/* grows from bottom of buf */
+	xfs_attr_leaf_name_remote_t valuelist;	/* grows from bottom of buf */
+} xfs_attr_leafblock_t;
+
+/*
+ * CRC enabled leaf structures. Called "version 3" structures to match the
+ * version number of the directory and dablk structures for this feature, and
+ * attr2 is already taken by the variable inode attribute fork size feature.
+ */
+struct xfs_attr3_leaf_hdr {
+	struct xfs_da3_blkinfo	info;
+	__be16			count;
+	__be16			usedbytes;
+	__be16			firstused;
+	__u8			holes;
+	__u8			pad1;
+	struct xfs_attr_leaf_map freemap[XFS_ATTR_LEAF_MAPSIZE];
+	__be32			pad2;		/* 64 bit alignment */
+};
+
+#define XFS_ATTR3_LEAF_CRC_OFF	(offsetof(struct xfs_attr3_leaf_hdr, info.crc))
+
+struct xfs_attr3_leafblock {
+	struct xfs_attr3_leaf_hdr	hdr;
+	struct xfs_attr_leaf_entry	entries[1];
+
+	/*
+	 * The rest of the block contains the following structures after the
+	 * leaf entries, growing from the bottom up. The variables are never
+	 * referenced, the locations accessed purely from helper functions.
+	 *
+	 * struct xfs_attr_leaf_name_local
+	 * struct xfs_attr_leaf_name_remote
+	 */
+};
+
+/*
+ * incore, neutral version of the attribute leaf header
+ */
+struct xfs_attr3_icleaf_hdr {
+	__uint32_t	forw;
+	__uint32_t	back;
+	__uint16_t	magic;
+	__uint16_t	count;
+	__uint16_t	usedbytes;
+	__uint16_t	firstused;
+	__u8		holes;
+	struct {
+		__uint16_t	base;
+		__uint16_t	size;
+	} freemap[XFS_ATTR_LEAF_MAPSIZE];
+};
+
+/*
+ * Flags used in the leaf_entry[i].flags field.
+ * NOTE: the INCOMPLETE bit must not collide with the flags bits specified
+ * on the system call, they are "or"ed together for various operations.
+ */
+#define	XFS_ATTR_LOCAL_BIT	0	/* attr is stored locally */
+#define	XFS_ATTR_ROOT_BIT	1	/* limit access to trusted attrs */
+#define	XFS_ATTR_SECURE_BIT	2	/* limit access to secure attrs */
+#define	XFS_ATTR_INCOMPLETE_BIT	7	/* attr in middle of create/delete */
+#define XFS_ATTR_LOCAL		(1 << XFS_ATTR_LOCAL_BIT)
+#define XFS_ATTR_ROOT		(1 << XFS_ATTR_ROOT_BIT)
+#define XFS_ATTR_SECURE		(1 << XFS_ATTR_SECURE_BIT)
+#define XFS_ATTR_INCOMPLETE	(1 << XFS_ATTR_INCOMPLETE_BIT)
+
+/*
+ * Conversion macros for converting namespace bits from argument flags
+ * to ondisk flags.
+ */
+#define XFS_ATTR_NSP_ARGS_MASK		(ATTR_ROOT | ATTR_SECURE)
+#define XFS_ATTR_NSP_ONDISK_MASK	(XFS_ATTR_ROOT | XFS_ATTR_SECURE)
+#define XFS_ATTR_NSP_ONDISK(flags)	((flags) & XFS_ATTR_NSP_ONDISK_MASK)
+#define XFS_ATTR_NSP_ARGS(flags)	((flags) & XFS_ATTR_NSP_ARGS_MASK)
+#define XFS_ATTR_NSP_ARGS_TO_ONDISK(x)	(((x) & ATTR_ROOT ? XFS_ATTR_ROOT : 0) |\
+					 ((x) & ATTR_SECURE ? XFS_ATTR_SECURE : 0))
+#define XFS_ATTR_NSP_ONDISK_TO_ARGS(x)	(((x) & XFS_ATTR_ROOT ? ATTR_ROOT : 0) |\
+					 ((x) & XFS_ATTR_SECURE ? ATTR_SECURE : 0))
+
+/*
+ * Alignment for namelist and valuelist entries (since they are mixed
+ * there can be only one alignment value)
+ */
+#define	XFS_ATTR_LEAF_NAME_ALIGN	((uint)sizeof(xfs_dablk_t))
+
+static inline int
+xfs_attr3_leaf_hdr_size(struct xfs_attr_leafblock *leafp)
+{
+	if (leafp->hdr.info.magic == cpu_to_be16(XFS_ATTR3_LEAF_MAGIC))
+		return sizeof(struct xfs_attr3_leaf_hdr);
+	return sizeof(struct xfs_attr_leaf_hdr);
+}
+
+static inline struct xfs_attr_leaf_entry *
+xfs_attr3_leaf_entryp(xfs_attr_leafblock_t *leafp)
+{
+	if (leafp->hdr.info.magic == cpu_to_be16(XFS_ATTR3_LEAF_MAGIC))
+		return &((struct xfs_attr3_leafblock *)leafp)->entries[0];
+	return &leafp->entries[0];
+}
+
+/*
+ * Cast typed pointers for "local" and "remote" name/value structs.
+ */
+static inline char *
+xfs_attr3_leaf_name(xfs_attr_leafblock_t *leafp, int idx)
+{
+	struct xfs_attr_leaf_entry *entries = xfs_attr3_leaf_entryp(leafp);
+
+	return &((char *)leafp)[be16_to_cpu(entries[idx].nameidx)];
+}
+
+static inline xfs_attr_leaf_name_remote_t *
+xfs_attr3_leaf_name_remote(xfs_attr_leafblock_t *leafp, int idx)
+{
+	return (xfs_attr_leaf_name_remote_t *)xfs_attr3_leaf_name(leafp, idx);
+}
+
+static inline xfs_attr_leaf_name_local_t *
+xfs_attr3_leaf_name_local(xfs_attr_leafblock_t *leafp, int idx)
+{
+	return (xfs_attr_leaf_name_local_t *)xfs_attr3_leaf_name(leafp, idx);
+}
+
+/*
+ * Calculate total bytes used (including trailing pad for alignment) for
+ * a "local" name/value structure, a "remote" name/value structure, and
+ * a pointer which might be either.
+ */
+static inline int xfs_attr_leaf_entsize_remote(int nlen)
+{
+	return ((uint)sizeof(xfs_attr_leaf_name_remote_t) - 1 + (nlen) + \
+		XFS_ATTR_LEAF_NAME_ALIGN - 1) & ~(XFS_ATTR_LEAF_NAME_ALIGN - 1);
+}
+
+static inline int xfs_attr_leaf_entsize_local(int nlen, int vlen)
+{
+	return ((uint)sizeof(xfs_attr_leaf_name_local_t) - 1 + (nlen) + (vlen) +
+		XFS_ATTR_LEAF_NAME_ALIGN - 1) & ~(XFS_ATTR_LEAF_NAME_ALIGN - 1);
+}
+
+static inline int xfs_attr_leaf_entsize_local_max(int bsize)
+{
+	return (((bsize) >> 1) + ((bsize) >> 2));
+}
+
+
+
+/*
+ * Remote attribute block format definition
+ *
+ * There is one of these headers per filesystem block in a remote attribute.
+ * This is done to ensure there is a 1:1 mapping between the attribute value
+ * length and the number of blocks needed to store the attribute. This makes the
+ * verification of a buffer a little more complex, but greatly simplifies the
+ * allocation, reading and writing of these attributes as we don't have to guess
+ * the number of blocks needed to store the attribute data.
+ */
+#define XFS_ATTR3_RMT_MAGIC	0x5841524d	/* XARM */
+
+struct xfs_attr3_rmt_hdr {
+	__be32	rm_magic;
+	__be32	rm_offset;
+	__be32	rm_bytes;
+	__be32	rm_crc;
+	uuid_t	rm_uuid;
+	__be64	rm_owner;
+	__be64	rm_blkno;
+	__be64	rm_lsn;
+};
+
+#define XFS_ATTR3_RMT_CRC_OFF	offsetof(struct xfs_attr3_rmt_hdr, rm_crc)
+
+#define XFS_ATTR3_RMT_BUF_SPACE(mp, bufsize)	\
+	((bufsize) - (xfs_sb_version_hascrc(&(mp)->m_sb) ? \
+			sizeof(struct xfs_attr3_rmt_hdr) : 0))
+
+#endif /* __XFS_DA_FORMAT_H__ */
diff --git a/fs/xfs/xfs_dir2.c b/fs/xfs/xfs_dir2.c
index edf203ab50af..ce16ef02997a 100644
--- a/fs/xfs/xfs_dir2.c
+++ b/fs/xfs/xfs_dir2.c
@@ -17,25 +17,24 @@
  */
 #include "xfs.h"
 #include "xfs_fs.h"
-#include "xfs_types.h"
-#include "xfs_log.h"
+#include "xfs_format.h"
+#include "xfs_log_format.h"
+#include "xfs_trans_resv.h"
 #include "xfs_inum.h"
-#include "xfs_trans.h"
 #include "xfs_sb.h"
 #include "xfs_ag.h"
 #include "xfs_mount.h"
+#include "xfs_da_format.h"
 #include "xfs_da_btree.h"
-#include "xfs_bmap_btree.h"
-#include "xfs_alloc_btree.h"
-#include "xfs_dinode.h"
 #include "xfs_inode.h"
+#include "xfs_trans.h"
 #include "xfs_inode_item.h"
 #include "xfs_bmap.h"
-#include "xfs_dir2_format.h"
 #include "xfs_dir2.h"
 #include "xfs_dir2_priv.h"
 #include "xfs_error.h"
 #include "xfs_trace.h"
+#include "xfs_dinode.h"
 
 struct xfs_name xfs_name_dotdot = { (unsigned char *)"..", 2, XFS_DIR3_FT_DIR };
 
@@ -96,13 +95,17 @@ xfs_dir_mount(
 	ASSERT(xfs_sb_version_hasdirv2(&mp->m_sb));
 	ASSERT((1 << (mp->m_sb.sb_blocklog + mp->m_sb.sb_dirblklog)) <=
 	       XFS_MAX_BLOCKSIZE);
+
+	mp->m_dir_inode_ops = xfs_dir_get_ops(mp, NULL);
+	mp->m_nondir_inode_ops = xfs_nondir_get_ops(mp, NULL);
+
 	mp->m_dirblksize = 1 << (mp->m_sb.sb_blocklog + mp->m_sb.sb_dirblklog);
 	mp->m_dirblkfsbs = 1 << mp->m_sb.sb_dirblklog;
 	mp->m_dirdatablk = xfs_dir2_db_to_da(mp, XFS_DIR2_DATA_FIRSTDB(mp));
 	mp->m_dirleafblk = xfs_dir2_db_to_da(mp, XFS_DIR2_LEAF_FIRSTDB(mp));
 	mp->m_dirfreeblk = xfs_dir2_db_to_da(mp, XFS_DIR2_FREE_FIRSTDB(mp));
 
-	nodehdr_size = __xfs_da3_node_hdr_size(xfs_sb_version_hascrc(&mp->m_sb));
+	nodehdr_size = mp->m_dir_inode_ops->node_hdr_size;
 	mp->m_attr_node_ents = (mp->m_sb.sb_blocksize - nodehdr_size) /
 				(uint)sizeof(xfs_da_node_entry_t);
 	mp->m_dir_node_ents = (mp->m_dirblksize - nodehdr_size) /
@@ -113,6 +116,7 @@ xfs_dir_mount(
 		mp->m_dirnameops = &xfs_ascii_ci_nameops;
 	else
 		mp->m_dirnameops = &xfs_default_nameops;
+
 }
 
 /*
diff --git a/fs/xfs/xfs_dir2.h b/fs/xfs/xfs_dir2.h
index 9910401327d4..cec70e0781ab 100644
--- a/fs/xfs/xfs_dir2.h
+++ b/fs/xfs/xfs_dir2.h
@@ -32,6 +32,83 @@ struct xfs_dir2_data_unused;
 extern struct xfs_name	xfs_name_dotdot;
 
 /*
+ * directory operations vector for encode/decode routines
+ */
+struct xfs_dir_ops {
+	int	(*sf_entsize)(struct xfs_dir2_sf_hdr *hdr, int len);
+	struct xfs_dir2_sf_entry *
+		(*sf_nextentry)(struct xfs_dir2_sf_hdr *hdr,
+				struct xfs_dir2_sf_entry *sfep);
+	__uint8_t (*sf_get_ftype)(struct xfs_dir2_sf_entry *sfep);
+	void	(*sf_put_ftype)(struct xfs_dir2_sf_entry *sfep,
+				__uint8_t ftype);
+	xfs_ino_t (*sf_get_ino)(struct xfs_dir2_sf_hdr *hdr,
+				struct xfs_dir2_sf_entry *sfep);
+	void	(*sf_put_ino)(struct xfs_dir2_sf_hdr *hdr,
+			      struct xfs_dir2_sf_entry *sfep,
+			      xfs_ino_t ino);
+	xfs_ino_t (*sf_get_parent_ino)(struct xfs_dir2_sf_hdr *hdr);
+	void	(*sf_put_parent_ino)(struct xfs_dir2_sf_hdr *hdr,
+				     xfs_ino_t ino);
+
+	int	(*data_entsize)(int len);
+	__uint8_t (*data_get_ftype)(struct xfs_dir2_data_entry *dep);
+	void	(*data_put_ftype)(struct xfs_dir2_data_entry *dep,
+				__uint8_t ftype);
+	__be16 * (*data_entry_tag_p)(struct xfs_dir2_data_entry *dep);
+	struct xfs_dir2_data_free *
+		(*data_bestfree_p)(struct xfs_dir2_data_hdr *hdr);
+
+	xfs_dir2_data_aoff_t data_dot_offset;
+	xfs_dir2_data_aoff_t data_dotdot_offset;
+	xfs_dir2_data_aoff_t data_first_offset;
+	size_t	data_entry_offset;
+
+	struct xfs_dir2_data_entry *
+		(*data_dot_entry_p)(struct xfs_dir2_data_hdr *hdr);
+	struct xfs_dir2_data_entry *
+		(*data_dotdot_entry_p)(struct xfs_dir2_data_hdr *hdr);
+	struct xfs_dir2_data_entry *
+		(*data_first_entry_p)(struct xfs_dir2_data_hdr *hdr);
+	struct xfs_dir2_data_entry *
+		(*data_entry_p)(struct xfs_dir2_data_hdr *hdr);
+	struct xfs_dir2_data_unused *
+		(*data_unused_p)(struct xfs_dir2_data_hdr *hdr);
+
+	int	leaf_hdr_size;
+	void	(*leaf_hdr_to_disk)(struct xfs_dir2_leaf *to,
+				    struct xfs_dir3_icleaf_hdr *from);
+	void	(*leaf_hdr_from_disk)(struct xfs_dir3_icleaf_hdr *to,
+				      struct xfs_dir2_leaf *from);
+	int	(*leaf_max_ents)(struct xfs_mount *mp);
+	struct xfs_dir2_leaf_entry *
+		(*leaf_ents_p)(struct xfs_dir2_leaf *lp);
+
+	int	node_hdr_size;
+	void	(*node_hdr_to_disk)(struct xfs_da_intnode *to,
+				    struct xfs_da3_icnode_hdr *from);
+	void	(*node_hdr_from_disk)(struct xfs_da3_icnode_hdr *to,
+				      struct xfs_da_intnode *from);
+	struct xfs_da_node_entry *
+		(*node_tree_p)(struct xfs_da_intnode *dap);
+
+	int	free_hdr_size;
+	void	(*free_hdr_to_disk)(struct xfs_dir2_free *to,
+				    struct xfs_dir3_icfree_hdr *from);
+	void	(*free_hdr_from_disk)(struct xfs_dir3_icfree_hdr *to,
+				      struct xfs_dir2_free *from);
+	int	(*free_max_bests)(struct xfs_mount *mp);
+	__be16 * (*free_bests_p)(struct xfs_dir2_free *free);
+	xfs_dir2_db_t (*db_to_fdb)(struct xfs_mount *mp, xfs_dir2_db_t db);
+	int	(*db_to_fdindex)(struct xfs_mount *mp, xfs_dir2_db_t db);
+};
+
+extern const struct xfs_dir_ops *
+	xfs_dir_get_ops(struct xfs_mount *mp, struct xfs_inode *dp);
+extern const struct xfs_dir_ops *
+	xfs_nondir_get_ops(struct xfs_mount *mp, struct xfs_inode *dp);
+
+/*
  * Generic directory interface routines
  */
 extern void xfs_dir_startup(void);
@@ -65,37 +142,30 @@ extern int xfs_dir2_sf_to_block(struct xfs_da_args *args);
 /*
  * Interface routines used by userspace utilities
  */
-extern xfs_ino_t xfs_dir2_sf_get_parent_ino(struct xfs_dir2_sf_hdr *sfp);
-extern void xfs_dir2_sf_put_parent_ino(struct xfs_dir2_sf_hdr *sfp,
-		xfs_ino_t ino);
-extern xfs_ino_t xfs_dir3_sfe_get_ino(struct xfs_mount *mp,
-		struct xfs_dir2_sf_hdr *sfp, struct xfs_dir2_sf_entry *sfep);
-extern void xfs_dir3_sfe_put_ino(struct xfs_mount *mp,
-		struct xfs_dir2_sf_hdr *hdr, struct xfs_dir2_sf_entry *sfep,
-		xfs_ino_t ino);
-
 extern int xfs_dir2_isblock(struct xfs_trans *tp, struct xfs_inode *dp, int *r);
 extern int xfs_dir2_isleaf(struct xfs_trans *tp, struct xfs_inode *dp, int *r);
 extern int xfs_dir2_shrink_inode(struct xfs_da_args *args, xfs_dir2_db_t db,
 				struct xfs_buf *bp);
 
-extern void xfs_dir2_data_freescan(struct xfs_mount *mp,
+extern void xfs_dir2_data_freescan(struct xfs_inode *dp,
 		struct xfs_dir2_data_hdr *hdr, int *loghead);
-extern void xfs_dir2_data_log_entry(struct xfs_trans *tp, struct xfs_buf *bp,
-		struct xfs_dir2_data_entry *dep);
-extern void xfs_dir2_data_log_header(struct xfs_trans *tp,
+extern void xfs_dir2_data_log_entry(struct xfs_trans *tp, struct xfs_inode *dp,
+		struct xfs_buf *bp, struct xfs_dir2_data_entry *dep);
+extern void xfs_dir2_data_log_header(struct xfs_trans *tp, struct xfs_inode *dp,
 		struct xfs_buf *bp);
 extern void xfs_dir2_data_log_unused(struct xfs_trans *tp, struct xfs_buf *bp,
 		struct xfs_dir2_data_unused *dup);
-extern void xfs_dir2_data_make_free(struct xfs_trans *tp, struct xfs_buf *bp,
+extern void xfs_dir2_data_make_free(struct xfs_trans *tp, struct xfs_inode *dp,
+		struct xfs_buf *bp, xfs_dir2_data_aoff_t offset,
+		xfs_dir2_data_aoff_t len, int *needlogp, int *needscanp);
+extern void xfs_dir2_data_use_free(struct xfs_trans *tp, struct xfs_inode *dp,
+		struct xfs_buf *bp, struct xfs_dir2_data_unused *dup,
 		xfs_dir2_data_aoff_t offset, xfs_dir2_data_aoff_t len,
 		int *needlogp, int *needscanp);
-extern void xfs_dir2_data_use_free(struct xfs_trans *tp, struct xfs_buf *bp,
-		struct xfs_dir2_data_unused *dup, xfs_dir2_data_aoff_t offset,
-		xfs_dir2_data_aoff_t len, int *needlogp, int *needscanp);
 
 extern struct xfs_dir2_data_free *xfs_dir2_data_freefind(
-		struct xfs_dir2_data_hdr *hdr, struct xfs_dir2_data_unused *dup);
+		struct xfs_dir2_data_hdr *hdr, struct xfs_dir2_data_free *bf,
+		struct xfs_dir2_data_unused *dup);
 
 extern const struct xfs_buf_ops xfs_dir3_block_buf_ops;
 extern const struct xfs_buf_ops xfs_dir3_leafn_buf_ops;
diff --git a/fs/xfs/xfs_dir2_block.c b/fs/xfs/xfs_dir2_block.c
index 12dad188939d..90cdbf4b5f19 100644
--- a/fs/xfs/xfs_dir2_block.c
+++ b/fs/xfs/xfs_dir2_block.c
@@ -18,25 +18,25 @@
  */
 #include "xfs.h"
 #include "xfs_fs.h"
-#include "xfs_types.h"
-#include "xfs_log.h"
-#include "xfs_trans.h"
+#include "xfs_format.h"
+#include "xfs_log_format.h"
+#include "xfs_trans_resv.h"
 #include "xfs_sb.h"
 #include "xfs_ag.h"
 #include "xfs_mount.h"
+#include "xfs_da_format.h"
 #include "xfs_da_btree.h"
-#include "xfs_bmap_btree.h"
-#include "xfs_dinode.h"
 #include "xfs_inode.h"
+#include "xfs_trans.h"
 #include "xfs_inode_item.h"
 #include "xfs_bmap.h"
 #include "xfs_buf_item.h"
-#include "xfs_dir2_format.h"
 #include "xfs_dir2.h"
 #include "xfs_dir2_priv.h"
 #include "xfs_error.h"
 #include "xfs_trace.h"
 #include "xfs_cksum.h"
+#include "xfs_dinode.h"
 
 /*
  * Local function prototypes.
@@ -168,6 +168,7 @@ xfs_dir3_block_init(
 
 static void
 xfs_dir2_block_need_space(
+	struct xfs_inode		*dp,
 	struct xfs_dir2_data_hdr	*hdr,
 	struct xfs_dir2_block_tail	*btp,
 	struct xfs_dir2_leaf_entry	*blp,
@@ -183,7 +184,7 @@ xfs_dir2_block_need_space(
 	struct xfs_dir2_data_unused	*enddup = NULL;
 
 	*compact = 0;
-	bf = xfs_dir3_data_bestfree_p(hdr);
+	bf = dp->d_ops->data_bestfree_p(hdr);
 
 	/*
 	 * If there are stale entries we'll use one for the leaf.
@@ -280,6 +281,7 @@ out:
 static void
 xfs_dir2_block_compact(
 	struct xfs_trans		*tp,
+	struct xfs_inode		*dp,
 	struct xfs_buf			*bp,
 	struct xfs_dir2_data_hdr	*hdr,
 	struct xfs_dir2_block_tail	*btp,
@@ -312,7 +314,7 @@ xfs_dir2_block_compact(
 	*lfloglow = toidx + 1 - (be32_to_cpu(btp->stale) - 1);
 	*lfloghigh -= be32_to_cpu(btp->stale) - 1;
 	be32_add_cpu(&btp->count, -(be32_to_cpu(btp->stale) - 1));
-	xfs_dir2_data_make_free(tp, bp,
+	xfs_dir2_data_make_free(tp, dp, bp,
 		(xfs_dir2_data_aoff_t)((char *)blp - (char *)hdr),
 		(xfs_dir2_data_aoff_t)((be32_to_cpu(btp->stale) - 1) * sizeof(*blp)),
 		needlog, &needscan);
@@ -323,7 +325,7 @@ xfs_dir2_block_compact(
 	 * This needs to happen before the next call to use_free.
 	 */
 	if (needscan)
-		xfs_dir2_data_freescan(tp->t_mountp, hdr, needlog);
+		xfs_dir2_data_freescan(dp, hdr, needlog);
 }
 
 /*
@@ -369,7 +371,7 @@ xfs_dir2_block_addname(
 	if (error)
 		return error;
 
-	len = xfs_dir3_data_entsize(mp, args->namelen);
+	len = dp->d_ops->data_entsize(args->namelen);
 
 	/*
 	 * Set up pointers to parts of the block.
@@ -382,7 +384,7 @@ xfs_dir2_block_addname(
 	 * Find out if we can reuse stale entries or whether we need extra
 	 * space for entry and new leaf.
 	 */
-	xfs_dir2_block_need_space(hdr, btp, blp, &tagp, &dup,
+	xfs_dir2_block_need_space(dp, hdr, btp, blp, &tagp, &dup,
 				  &enddup, &compact, len);
 
 	/*
@@ -418,7 +420,7 @@ xfs_dir2_block_addname(
 	 * If need to compact the leaf entries, do it now.
 	 */
 	if (compact) {
-		xfs_dir2_block_compact(tp, bp, hdr, btp, blp, &needlog,
+		xfs_dir2_block_compact(tp, dp, bp, hdr, btp, blp, &needlog,
 				      &lfloghigh, &lfloglow);
 		/* recalculate blp post-compaction */
 		blp = xfs_dir2_block_leaf_p(btp);
@@ -453,7 +455,7 @@ xfs_dir2_block_addname(
 		/*
 		 * Mark the space needed for the new leaf entry, now in use.
 		 */
-		xfs_dir2_data_use_free(tp, bp, enddup,
+		xfs_dir2_data_use_free(tp, dp, bp, enddup,
 			(xfs_dir2_data_aoff_t)
 			((char *)enddup - (char *)hdr + be16_to_cpu(enddup->length) -
 			 sizeof(*blp)),
@@ -468,7 +470,7 @@ xfs_dir2_block_addname(
 		 * This needs to happen before the next call to use_free.
 		 */
 		if (needscan) {
-			xfs_dir2_data_freescan(mp, hdr, &needlog);
+			xfs_dir2_data_freescan(dp, hdr, &needlog);
 			needscan = 0;
 		}
 		/*
@@ -540,7 +542,7 @@ xfs_dir2_block_addname(
 	/*
 	 * Mark space for the data entry used.
 	 */
-	xfs_dir2_data_use_free(tp, bp, dup,
+	xfs_dir2_data_use_free(tp, dp, bp, dup,
 		(xfs_dir2_data_aoff_t)((char *)dup - (char *)hdr),
 		(xfs_dir2_data_aoff_t)len, &needlog, &needscan);
 	/*
@@ -549,18 +551,18 @@ xfs_dir2_block_addname(
 	dep->inumber = cpu_to_be64(args->inumber);
 	dep->namelen = args->namelen;
 	memcpy(dep->name, args->name, args->namelen);
-	xfs_dir3_dirent_put_ftype(mp, dep, args->filetype);
-	tagp = xfs_dir3_data_entry_tag_p(mp, dep);
+	dp->d_ops->data_put_ftype(dep, args->filetype);
+	tagp = dp->d_ops->data_entry_tag_p(dep);
 	*tagp = cpu_to_be16((char *)dep - (char *)hdr);
 	/*
 	 * Clean up the bestfree array and log the header, tail, and entry.
 	 */
 	if (needscan)
-		xfs_dir2_data_freescan(mp, hdr, &needlog);
+		xfs_dir2_data_freescan(dp, hdr, &needlog);
 	if (needlog)
-		xfs_dir2_data_log_header(tp, bp);
+		xfs_dir2_data_log_header(tp, dp, bp);
 	xfs_dir2_block_log_tail(tp, bp);
-	xfs_dir2_data_log_entry(tp, bp, dep);
+	xfs_dir2_data_log_entry(tp, dp, bp, dep);
 	xfs_dir3_data_check(dp, bp);
 	return 0;
 }
@@ -642,7 +644,7 @@ xfs_dir2_block_lookup(
 	 * Fill in inode number, CI name if appropriate, release the block.
 	 */
 	args->inumber = be64_to_cpu(dep->inumber);
-	args->filetype = xfs_dir3_dirent_get_ftype(mp, dep);
+	args->filetype = dp->d_ops->data_get_ftype(dep);
 	error = xfs_dir_cilookup_result(args, dep->name, dep->namelen);
 	xfs_trans_brelse(args->trans, bp);
 	return XFS_ERROR(error);
@@ -799,9 +801,9 @@ xfs_dir2_block_removename(
 	 * Mark the data entry's space free.
 	 */
 	needlog = needscan = 0;
-	xfs_dir2_data_make_free(tp, bp,
+	xfs_dir2_data_make_free(tp, dp, bp,
 		(xfs_dir2_data_aoff_t)((char *)dep - (char *)hdr),
-		xfs_dir3_data_entsize(mp, dep->namelen), &needlog, &needscan);
+		dp->d_ops->data_entsize(dep->namelen), &needlog, &needscan);
 	/*
 	 * Fix up the block tail.
 	 */
@@ -816,9 +818,9 @@ xfs_dir2_block_removename(
 	 * Fix up bestfree, log the header if necessary.
 	 */
 	if (needscan)
-		xfs_dir2_data_freescan(mp, hdr, &needlog);
+		xfs_dir2_data_freescan(dp, hdr, &needlog);
 	if (needlog)
-		xfs_dir2_data_log_header(tp, bp);
+		xfs_dir2_data_log_header(tp, dp, bp);
 	xfs_dir3_data_check(dp, bp);
 	/*
 	 * See if the size as a shortform is good enough.
@@ -875,8 +877,8 @@ xfs_dir2_block_replace(
 	 * Change the inode number to the new value.
 	 */
 	dep->inumber = cpu_to_be64(args->inumber);
-	xfs_dir3_dirent_put_ftype(mp, dep, args->filetype);
-	xfs_dir2_data_log_entry(args->trans, bp, dep);
+	dp->d_ops->data_put_ftype(dep, args->filetype);
+	xfs_dir2_data_log_entry(args->trans, dp, bp, dep);
 	xfs_dir3_data_check(dp, bp);
 	return 0;
 }
@@ -934,8 +936,8 @@ xfs_dir2_leaf_to_block(
 	tp = args->trans;
 	mp = dp->i_mount;
 	leaf = lbp->b_addr;
-	xfs_dir3_leaf_hdr_from_disk(&leafhdr, leaf);
-	ents = xfs_dir3_leaf_ents_p(leaf);
+	dp->d_ops->leaf_hdr_from_disk(&leafhdr, leaf);
+	ents = dp->d_ops->leaf_ents_p(leaf);
 	ltp = xfs_dir2_leaf_tail_p(mp, leaf);
 
 	ASSERT(leafhdr.magic == XFS_DIR2_LEAF1_MAGIC ||
@@ -949,7 +951,7 @@ xfs_dir2_leaf_to_block(
 	while (dp->i_d.di_size > mp->m_dirblksize) {
 		int hdrsz;
 
-		hdrsz = xfs_dir3_data_hdr_size(xfs_sb_version_hascrc(&mp->m_sb));
+		hdrsz = dp->d_ops->data_entry_offset;
 		bestsp = xfs_dir2_leaf_bests_p(ltp);
 		if (be16_to_cpu(bestsp[be32_to_cpu(ltp->bestcount) - 1]) ==
 					    mp->m_dirblksize - hdrsz) {
@@ -999,7 +1001,7 @@ xfs_dir2_leaf_to_block(
 	/*
 	 * Use up the space at the end of the block (blp/btp).
 	 */
-	xfs_dir2_data_use_free(tp, dbp, dup, mp->m_dirblksize - size, size,
+	xfs_dir2_data_use_free(tp, dp, dbp, dup, mp->m_dirblksize - size, size,
 		&needlog, &needscan);
 	/*
 	 * Initialize the block tail.
@@ -1023,9 +1025,9 @@ xfs_dir2_leaf_to_block(
 	 * Scan the bestfree if we need it and log the data block header.
 	 */
 	if (needscan)
-		xfs_dir2_data_freescan(mp, hdr, &needlog);
+		xfs_dir2_data_freescan(dp, hdr, &needlog);
 	if (needlog)
-		xfs_dir2_data_log_header(tp, dbp);
+		xfs_dir2_data_log_header(tp, dp, dbp);
 	/*
 	 * Pitch the old leaf block.
 	 */
@@ -1136,9 +1138,9 @@ xfs_dir2_sf_to_block(
 	 * The whole thing is initialized to free by the init routine.
 	 * Say we're using the leaf and tail area.
 	 */
-	dup = xfs_dir3_data_unused_p(hdr);
+	dup = dp->d_ops->data_unused_p(hdr);
 	needlog = needscan = 0;
-	xfs_dir2_data_use_free(tp, bp, dup, mp->m_dirblksize - i, i, &needlog,
+	xfs_dir2_data_use_free(tp, dp, bp, dup, mp->m_dirblksize - i, i, &needlog,
 		&needscan);
 	ASSERT(needscan == 0);
 	/*
@@ -1152,38 +1154,38 @@ xfs_dir2_sf_to_block(
 	/*
 	 * Remove the freespace, we'll manage it.
 	 */
-	xfs_dir2_data_use_free(tp, bp, dup,
+	xfs_dir2_data_use_free(tp, dp, bp, dup,
 		(xfs_dir2_data_aoff_t)((char *)dup - (char *)hdr),
 		be16_to_cpu(dup->length), &needlog, &needscan);
 	/*
 	 * Create entry for .
 	 */
-	dep = xfs_dir3_data_dot_entry_p(mp, hdr);
+	dep = dp->d_ops->data_dot_entry_p(hdr);
 	dep->inumber = cpu_to_be64(dp->i_ino);
 	dep->namelen = 1;
 	dep->name[0] = '.';
-	xfs_dir3_dirent_put_ftype(mp, dep, XFS_DIR3_FT_DIR);
-	tagp = xfs_dir3_data_entry_tag_p(mp, dep);
+	dp->d_ops->data_put_ftype(dep, XFS_DIR3_FT_DIR);
+	tagp = dp->d_ops->data_entry_tag_p(dep);
 	*tagp = cpu_to_be16((char *)dep - (char *)hdr);
-	xfs_dir2_data_log_entry(tp, bp, dep);
+	xfs_dir2_data_log_entry(tp, dp, bp, dep);
 	blp[0].hashval = cpu_to_be32(xfs_dir_hash_dot);
 	blp[0].address = cpu_to_be32(xfs_dir2_byte_to_dataptr(mp,
 				(char *)dep - (char *)hdr));
 	/*
 	 * Create entry for ..
 	 */
-	dep = xfs_dir3_data_dotdot_entry_p(mp, hdr);
-	dep->inumber = cpu_to_be64(xfs_dir2_sf_get_parent_ino(sfp));
+	dep = dp->d_ops->data_dotdot_entry_p(hdr);
+	dep->inumber = cpu_to_be64(dp->d_ops->sf_get_parent_ino(sfp));
 	dep->namelen = 2;
 	dep->name[0] = dep->name[1] = '.';
-	xfs_dir3_dirent_put_ftype(mp, dep, XFS_DIR3_FT_DIR);
-	tagp = xfs_dir3_data_entry_tag_p(mp, dep);
+	dp->d_ops->data_put_ftype(dep, XFS_DIR3_FT_DIR);
+	tagp = dp->d_ops->data_entry_tag_p(dep);
 	*tagp = cpu_to_be16((char *)dep - (char *)hdr);
-	xfs_dir2_data_log_entry(tp, bp, dep);
+	xfs_dir2_data_log_entry(tp, dp, bp, dep);
 	blp[1].hashval = cpu_to_be32(xfs_dir_hash_dotdot);
 	blp[1].address = cpu_to_be32(xfs_dir2_byte_to_dataptr(mp,
 				(char *)dep - (char *)hdr));
-	offset = xfs_dir3_data_first_offset(mp);
+	offset = dp->d_ops->data_first_offset;
 	/*
 	 * Loop over existing entries, stuff them in.
 	 */
@@ -1214,7 +1216,9 @@ xfs_dir2_sf_to_block(
 			*xfs_dir2_data_unused_tag_p(dup) = cpu_to_be16(
 				((char *)dup - (char *)hdr));
 			xfs_dir2_data_log_unused(tp, bp, dup);
-			xfs_dir2_data_freeinsert(hdr, dup, &dummy);
+			xfs_dir2_data_freeinsert(hdr,
+						 dp->d_ops->data_bestfree_p(hdr),
+						 dup, &dummy);
 			offset += be16_to_cpu(dup->length);
 			continue;
 		}
@@ -1222,14 +1226,13 @@ xfs_dir2_sf_to_block(
 		 * Copy a real entry.
 		 */
 		dep = (xfs_dir2_data_entry_t *)((char *)hdr + newoffset);
-		dep->inumber = cpu_to_be64(xfs_dir3_sfe_get_ino(mp, sfp, sfep));
+		dep->inumber = cpu_to_be64(dp->d_ops->sf_get_ino(sfp, sfep));
 		dep->namelen = sfep->namelen;
-		xfs_dir3_dirent_put_ftype(mp, dep,
-					xfs_dir3_sfe_get_ftype(mp, sfp, sfep));
+		dp->d_ops->data_put_ftype(dep, dp->d_ops->sf_get_ftype(sfep));
 		memcpy(dep->name, sfep->name, dep->namelen);
-		tagp = xfs_dir3_data_entry_tag_p(mp, dep);
+		tagp = dp->d_ops->data_entry_tag_p(dep);
 		*tagp = cpu_to_be16((char *)dep - (char *)hdr);
-		xfs_dir2_data_log_entry(tp, bp, dep);
+		xfs_dir2_data_log_entry(tp, dp, bp, dep);
 		name.name = sfep->name;
 		name.len = sfep->namelen;
 		blp[2 + i].hashval = cpu_to_be32(mp->m_dirnameops->
@@ -1240,7 +1243,7 @@ xfs_dir2_sf_to_block(
 		if (++i == sfp->count)
 			sfep = NULL;
 		else
-			sfep = xfs_dir3_sf_nextentry(mp, sfp, sfep);
+			sfep = dp->d_ops->sf_nextentry(sfp, sfep);
 	}
 	/* Done with the temporary buffer */
 	kmem_free(sfp);
diff --git a/fs/xfs/xfs_dir2_data.c b/fs/xfs/xfs_dir2_data.c
index 47e1326c169a..70acff4ee173 100644
--- a/fs/xfs/xfs_dir2_data.c
+++ b/fs/xfs/xfs_dir2_data.c
@@ -18,20 +18,19 @@
  */
 #include "xfs.h"
 #include "xfs_fs.h"
-#include "xfs_types.h"
-#include "xfs_log.h"
-#include "xfs_trans.h"
+#include "xfs_format.h"
+#include "xfs_log_format.h"
+#include "xfs_trans_resv.h"
 #include "xfs_sb.h"
 #include "xfs_ag.h"
 #include "xfs_mount.h"
+#include "xfs_da_format.h"
 #include "xfs_da_btree.h"
-#include "xfs_bmap_btree.h"
-#include "xfs_dinode.h"
 #include "xfs_inode.h"
-#include "xfs_dir2_format.h"
 #include "xfs_dir2.h"
 #include "xfs_dir2_priv.h"
 #include "xfs_error.h"
+#include "xfs_trans.h"
 #include "xfs_buf_item.h"
 #include "xfs_cksum.h"
 
@@ -63,11 +62,18 @@ __xfs_dir3_data_check(
 	char			*p;		/* current data position */
 	int			stale;		/* count of stale leaves */
 	struct xfs_name		name;
+	const struct xfs_dir_ops *ops;
 
 	mp = bp->b_target->bt_mount;
+
+	/*
+	 * We can be passed a null dp here from a verifier, so we need to go the
+	 * hard way to get them.
+	 */
+	ops = xfs_dir_get_ops(mp, dp);
+
 	hdr = bp->b_addr;
-	bf = xfs_dir3_data_bestfree_p(hdr);
-	p = (char *)xfs_dir3_data_entry_p(hdr);
+	p = (char *)ops->data_entry_p(hdr);
 
 	switch (hdr->magic) {
 	case cpu_to_be32(XFS_DIR3_BLOCK_MAGIC):
@@ -75,6 +81,16 @@ __xfs_dir3_data_check(
 		btp = xfs_dir2_block_tail_p(mp, hdr);
 		lep = xfs_dir2_block_leaf_p(btp);
 		endp = (char *)lep;
+
+		/*
+		 * The number of leaf entries is limited by the size of the
+		 * block and the amount of space used by the data entries.
+		 * We don't know how much space is used by the data entries yet,
+		 * so just ensure that the count falls somewhere inside the
+		 * block right now.
+		 */
+		XFS_WANT_CORRUPTED_RETURN(be32_to_cpu(btp->count) <
+			((char *)btp - p) / sizeof(struct xfs_dir2_leaf_entry));
 		break;
 	case cpu_to_be32(XFS_DIR3_DATA_MAGIC):
 	case cpu_to_be32(XFS_DIR2_DATA_MAGIC):
@@ -85,10 +101,11 @@ __xfs_dir3_data_check(
 		return EFSCORRUPTED;
 	}
 
-	count = lastfree = freeseen = 0;
 	/*
 	 * Account for zero bestfree entries.
 	 */
+	bf = ops->data_bestfree_p(hdr);
+	count = lastfree = freeseen = 0;
 	if (!bf[0].length) {
 		XFS_WANT_CORRUPTED_RETURN(!bf[0].offset);
 		freeseen |= 1 << 0;
@@ -121,7 +138,7 @@ __xfs_dir3_data_check(
 			XFS_WANT_CORRUPTED_RETURN(
 				be16_to_cpu(*xfs_dir2_data_unused_tag_p(dup)) ==
 					       (char *)dup - (char *)hdr);
-			dfp = xfs_dir2_data_freefind(hdr, dup);
+			dfp = xfs_dir2_data_freefind(hdr, bf, dup);
 			if (dfp) {
 				i = (int)(dfp - bf);
 				XFS_WANT_CORRUPTED_RETURN(
@@ -147,10 +164,10 @@ __xfs_dir3_data_check(
 		XFS_WANT_CORRUPTED_RETURN(
 			!xfs_dir_ino_validate(mp, be64_to_cpu(dep->inumber)));
 		XFS_WANT_CORRUPTED_RETURN(
-			be16_to_cpu(*xfs_dir3_data_entry_tag_p(mp, dep)) ==
+			be16_to_cpu(*ops->data_entry_tag_p(dep)) ==
 					       (char *)dep - (char *)hdr);
 		XFS_WANT_CORRUPTED_RETURN(
-			xfs_dir3_dirent_get_ftype(mp, dep) < XFS_DIR3_FT_MAX);
+				ops->data_get_ftype(dep) < XFS_DIR3_FT_MAX);
 		count++;
 		lastfree = 0;
 		if (hdr->magic == cpu_to_be32(XFS_DIR2_BLOCK_MAGIC) ||
@@ -168,7 +185,7 @@ __xfs_dir3_data_check(
 			}
 			XFS_WANT_CORRUPTED_RETURN(i < be32_to_cpu(btp->count));
 		}
-		p += xfs_dir3_data_entsize(mp, dep->namelen);
+		p += ops->data_entsize(dep->namelen);
 	}
 	/*
 	 * Need to have seen all the entries and all the bestfree slots.
@@ -327,19 +344,18 @@ xfs_dir3_data_readahead(
  */
 xfs_dir2_data_free_t *
 xfs_dir2_data_freefind(
-	xfs_dir2_data_hdr_t	*hdr,		/* data block */
-	xfs_dir2_data_unused_t	*dup)		/* data unused entry */
+	struct xfs_dir2_data_hdr *hdr,		/* data block header */
+	struct xfs_dir2_data_free *bf,		/* bestfree table pointer */
+	struct xfs_dir2_data_unused *dup)	/* unused space */
 {
 	xfs_dir2_data_free_t	*dfp;		/* bestfree entry */
 	xfs_dir2_data_aoff_t	off;		/* offset value needed */
-	struct xfs_dir2_data_free *bf;
 #ifdef DEBUG
 	int			matched;	/* matched the value */
 	int			seenzero;	/* saw a 0 bestfree entry */
 #endif
 
 	off = (xfs_dir2_data_aoff_t)((char *)dup - (char *)hdr);
-	bf = xfs_dir3_data_bestfree_p(hdr);
 
 #ifdef DEBUG
 	/*
@@ -399,11 +415,11 @@ xfs_dir2_data_freefind(
  */
 xfs_dir2_data_free_t *				/* entry inserted */
 xfs_dir2_data_freeinsert(
-	xfs_dir2_data_hdr_t	*hdr,		/* data block pointer */
-	xfs_dir2_data_unused_t	*dup,		/* unused space */
+	struct xfs_dir2_data_hdr *hdr,		/* data block pointer */
+	struct xfs_dir2_data_free *dfp,		/* bestfree table pointer */
+	struct xfs_dir2_data_unused *dup,	/* unused space */
 	int			*loghead)	/* log the data header (out) */
 {
-	xfs_dir2_data_free_t	*dfp;		/* bestfree table pointer */
 	xfs_dir2_data_free_t	new;		/* new bestfree entry */
 
 	ASSERT(hdr->magic == cpu_to_be32(XFS_DIR2_DATA_MAGIC) ||
@@ -411,7 +427,6 @@ xfs_dir2_data_freeinsert(
 	       hdr->magic == cpu_to_be32(XFS_DIR3_DATA_MAGIC) ||
 	       hdr->magic == cpu_to_be32(XFS_DIR3_BLOCK_MAGIC));
 
-	dfp = xfs_dir3_data_bestfree_p(hdr);
 	new.length = dup->length;
 	new.offset = cpu_to_be16((char *)dup - (char *)hdr);
 
@@ -444,11 +459,11 @@ xfs_dir2_data_freeinsert(
  */
 STATIC void
 xfs_dir2_data_freeremove(
-	xfs_dir2_data_hdr_t	*hdr,		/* data block header */
-	xfs_dir2_data_free_t	*dfp,		/* bestfree entry pointer */
+	struct xfs_dir2_data_hdr *hdr,		/* data block header */
+	struct xfs_dir2_data_free *bf,		/* bestfree table pointer */
+	struct xfs_dir2_data_free *dfp,		/* bestfree entry pointer */
 	int			*loghead)	/* out: log data header */
 {
-	struct xfs_dir2_data_free *bf;
 
 	ASSERT(hdr->magic == cpu_to_be32(XFS_DIR2_DATA_MAGIC) ||
 	       hdr->magic == cpu_to_be32(XFS_DIR2_BLOCK_MAGIC) ||
@@ -458,7 +473,6 @@ xfs_dir2_data_freeremove(
 	/*
 	 * It's the first entry, slide the next 2 up.
 	 */
-	bf = xfs_dir3_data_bestfree_p(hdr);
 	if (dfp == &bf[0]) {
 		bf[0] = bf[1];
 		bf[1] = bf[2];
@@ -486,9 +500,9 @@ xfs_dir2_data_freeremove(
  */
 void
 xfs_dir2_data_freescan(
-	xfs_mount_t		*mp,		/* filesystem mount point */
-	xfs_dir2_data_hdr_t	*hdr,		/* data block header */
-	int			*loghead)	/* out: log data header */
+	struct xfs_inode	*dp,
+	struct xfs_dir2_data_hdr *hdr,
+	int			*loghead)
 {
 	xfs_dir2_block_tail_t	*btp;		/* block tail */
 	xfs_dir2_data_entry_t	*dep;		/* active data entry */
@@ -505,19 +519,19 @@ xfs_dir2_data_freescan(
 	/*
 	 * Start by clearing the table.
 	 */
-	bf = xfs_dir3_data_bestfree_p(hdr);
+	bf = dp->d_ops->data_bestfree_p(hdr);
 	memset(bf, 0, sizeof(*bf) * XFS_DIR2_DATA_FD_COUNT);
 	*loghead = 1;
 	/*
 	 * Set up pointers.
 	 */
-	p = (char *)xfs_dir3_data_entry_p(hdr);
+	p = (char *)dp->d_ops->data_entry_p(hdr);
 	if (hdr->magic == cpu_to_be32(XFS_DIR2_BLOCK_MAGIC) ||
 	    hdr->magic == cpu_to_be32(XFS_DIR3_BLOCK_MAGIC)) {
-		btp = xfs_dir2_block_tail_p(mp, hdr);
+		btp = xfs_dir2_block_tail_p(dp->i_mount, hdr);
 		endp = (char *)xfs_dir2_block_leaf_p(btp);
 	} else
-		endp = (char *)hdr + mp->m_dirblksize;
+		endp = (char *)hdr + dp->i_mount->m_dirblksize;
 	/*
 	 * Loop over the block's entries.
 	 */
@@ -529,7 +543,7 @@ xfs_dir2_data_freescan(
 		if (be16_to_cpu(dup->freetag) == XFS_DIR2_DATA_FREE_TAG) {
 			ASSERT((char *)dup - (char *)hdr ==
 			       be16_to_cpu(*xfs_dir2_data_unused_tag_p(dup)));
-			xfs_dir2_data_freeinsert(hdr, dup, loghead);
+			xfs_dir2_data_freeinsert(hdr, bf, dup, loghead);
 			p += be16_to_cpu(dup->length);
 		}
 		/*
@@ -538,8 +552,8 @@ xfs_dir2_data_freescan(
 		else {
 			dep = (xfs_dir2_data_entry_t *)p;
 			ASSERT((char *)dep - (char *)hdr ==
-			       be16_to_cpu(*xfs_dir3_data_entry_tag_p(mp, dep)));
-			p += xfs_dir3_data_entsize(mp, dep->namelen);
+			       be16_to_cpu(*dp->d_ops->data_entry_tag_p(dep)));
+			p += dp->d_ops->data_entsize(dep->namelen);
 		}
 	}
 }
@@ -594,8 +608,8 @@ xfs_dir3_data_init(
 	} else
 		hdr->magic = cpu_to_be32(XFS_DIR2_DATA_MAGIC);
 
-	bf = xfs_dir3_data_bestfree_p(hdr);
-	bf[0].offset = cpu_to_be16(xfs_dir3_data_entry_offset(hdr));
+	bf = dp->d_ops->data_bestfree_p(hdr);
+	bf[0].offset = cpu_to_be16(dp->d_ops->data_entry_offset);
 	for (i = 1; i < XFS_DIR2_DATA_FD_COUNT; i++) {
 		bf[i].length = 0;
 		bf[i].offset = 0;
@@ -604,17 +618,17 @@ xfs_dir3_data_init(
 	/*
 	 * Set up an unused entry for the block's body.
 	 */
-	dup = xfs_dir3_data_unused_p(hdr);
+	dup = dp->d_ops->data_unused_p(hdr);
 	dup->freetag = cpu_to_be16(XFS_DIR2_DATA_FREE_TAG);
 
-	t = mp->m_dirblksize - (uint)xfs_dir3_data_entry_offset(hdr);
+	t = mp->m_dirblksize - (uint)dp->d_ops->data_entry_offset;
 	bf[0].length = cpu_to_be16(t);
 	dup->length = cpu_to_be16(t);
 	*xfs_dir2_data_unused_tag_p(dup) = cpu_to_be16((char *)dup - (char *)hdr);
 	/*
 	 * Log it and return it.
 	 */
-	xfs_dir2_data_log_header(tp, bp);
+	xfs_dir2_data_log_header(tp, dp, bp);
 	xfs_dir2_data_log_unused(tp, bp, dup);
 	*bpp = bp;
 	return 0;
@@ -626,11 +640,11 @@ xfs_dir3_data_init(
 void
 xfs_dir2_data_log_entry(
 	struct xfs_trans	*tp,
+	struct xfs_inode	*dp,
 	struct xfs_buf		*bp,
 	xfs_dir2_data_entry_t	*dep)		/* data entry pointer */
 {
 	struct xfs_dir2_data_hdr *hdr = bp->b_addr;
-	struct xfs_mount	*mp = tp->t_mountp;
 
 	ASSERT(hdr->magic == cpu_to_be32(XFS_DIR2_DATA_MAGIC) ||
 	       hdr->magic == cpu_to_be32(XFS_DIR3_DATA_MAGIC) ||
@@ -638,7 +652,7 @@ xfs_dir2_data_log_entry(
 	       hdr->magic == cpu_to_be32(XFS_DIR3_BLOCK_MAGIC));
 
 	xfs_trans_log_buf(tp, bp, (uint)((char *)dep - (char *)hdr),
-		(uint)((char *)(xfs_dir3_data_entry_tag_p(mp, dep) + 1) -
+		(uint)((char *)(dp->d_ops->data_entry_tag_p(dep) + 1) -
 		       (char *)hdr - 1));
 }
 
@@ -648,16 +662,19 @@ xfs_dir2_data_log_entry(
 void
 xfs_dir2_data_log_header(
 	struct xfs_trans	*tp,
+	struct xfs_inode	*dp,
 	struct xfs_buf		*bp)
 {
-	xfs_dir2_data_hdr_t	*hdr = bp->b_addr;
+#ifdef DEBUG
+	struct xfs_dir2_data_hdr *hdr = bp->b_addr;
 
 	ASSERT(hdr->magic == cpu_to_be32(XFS_DIR2_DATA_MAGIC) ||
 	       hdr->magic == cpu_to_be32(XFS_DIR3_DATA_MAGIC) ||
 	       hdr->magic == cpu_to_be32(XFS_DIR2_BLOCK_MAGIC) ||
 	       hdr->magic == cpu_to_be32(XFS_DIR3_BLOCK_MAGIC));
+#endif
 
-	xfs_trans_log_buf(tp, bp, 0, xfs_dir3_data_entry_offset(hdr) - 1);
+	xfs_trans_log_buf(tp, bp, 0, dp->d_ops->data_entry_offset - 1);
 }
 
 /*
@@ -698,6 +715,7 @@ xfs_dir2_data_log_unused(
 void
 xfs_dir2_data_make_free(
 	struct xfs_trans	*tp,
+	struct xfs_inode	*dp,
 	struct xfs_buf		*bp,
 	xfs_dir2_data_aoff_t	offset,		/* starting byte offset */
 	xfs_dir2_data_aoff_t	len,		/* length in bytes */
@@ -735,7 +753,7 @@ xfs_dir2_data_make_free(
 	 * If this isn't the start of the block, then back up to
 	 * the previous entry and see if it's free.
 	 */
-	if (offset > xfs_dir3_data_entry_offset(hdr)) {
+	if (offset > dp->d_ops->data_entry_offset) {
 		__be16			*tagp;	/* tag just before us */
 
 		tagp = (__be16 *)((char *)hdr + offset) - 1;
@@ -761,15 +779,15 @@ xfs_dir2_data_make_free(
 	 * Previous and following entries are both free,
 	 * merge everything into a single free entry.
 	 */
-	bf = xfs_dir3_data_bestfree_p(hdr);
+	bf = dp->d_ops->data_bestfree_p(hdr);
 	if (prevdup && postdup) {
 		xfs_dir2_data_free_t	*dfp2;	/* another bestfree pointer */
 
 		/*
 		 * See if prevdup and/or postdup are in bestfree table.
 		 */
-		dfp = xfs_dir2_data_freefind(hdr, prevdup);
-		dfp2 = xfs_dir2_data_freefind(hdr, postdup);
+		dfp = xfs_dir2_data_freefind(hdr, bf, prevdup);
+		dfp2 = xfs_dir2_data_freefind(hdr, bf, postdup);
 		/*
 		 * We need a rescan unless there are exactly 2 free entries
 		 * namely our two.  Then we know what's happening, otherwise
@@ -797,12 +815,13 @@ xfs_dir2_data_make_free(
 				ASSERT(dfp2 == dfp);
 				dfp2 = &bf[1];
 			}
-			xfs_dir2_data_freeremove(hdr, dfp2, needlogp);
-			xfs_dir2_data_freeremove(hdr, dfp, needlogp);
+			xfs_dir2_data_freeremove(hdr, bf, dfp2, needlogp);
+			xfs_dir2_data_freeremove(hdr, bf, dfp, needlogp);
 			/*
 			 * Now insert the new entry.
 			 */
-			dfp = xfs_dir2_data_freeinsert(hdr, prevdup, needlogp);
+			dfp = xfs_dir2_data_freeinsert(hdr, bf, prevdup,
+						       needlogp);
 			ASSERT(dfp == &bf[0]);
 			ASSERT(dfp->length == prevdup->length);
 			ASSERT(!dfp[1].length);
@@ -813,7 +832,7 @@ xfs_dir2_data_make_free(
 	 * The entry before us is free, merge with it.
 	 */
 	else if (prevdup) {
-		dfp = xfs_dir2_data_freefind(hdr, prevdup);
+		dfp = xfs_dir2_data_freefind(hdr, bf, prevdup);
 		be16_add_cpu(&prevdup->length, len);
 		*xfs_dir2_data_unused_tag_p(prevdup) =
 			cpu_to_be16((char *)prevdup - (char *)hdr);
@@ -824,8 +843,8 @@ xfs_dir2_data_make_free(
 		 * the old one and add the new one.
 		 */
 		if (dfp) {
-			xfs_dir2_data_freeremove(hdr, dfp, needlogp);
-			xfs_dir2_data_freeinsert(hdr, prevdup, needlogp);
+			xfs_dir2_data_freeremove(hdr, bf, dfp, needlogp);
+			xfs_dir2_data_freeinsert(hdr, bf, prevdup, needlogp);
 		}
 		/*
 		 * Otherwise we need a scan if the new entry is big enough.
@@ -839,7 +858,7 @@ xfs_dir2_data_make_free(
 	 * The following entry is free, merge with it.
 	 */
 	else if (postdup) {
-		dfp = xfs_dir2_data_freefind(hdr, postdup);
+		dfp = xfs_dir2_data_freefind(hdr, bf, postdup);
 		newdup = (xfs_dir2_data_unused_t *)((char *)hdr + offset);
 		newdup->freetag = cpu_to_be16(XFS_DIR2_DATA_FREE_TAG);
 		newdup->length = cpu_to_be16(len + be16_to_cpu(postdup->length));
@@ -852,8 +871,8 @@ xfs_dir2_data_make_free(
 		 * the old one and add the new one.
 		 */
 		if (dfp) {
-			xfs_dir2_data_freeremove(hdr, dfp, needlogp);
-			xfs_dir2_data_freeinsert(hdr, newdup, needlogp);
+			xfs_dir2_data_freeremove(hdr, bf, dfp, needlogp);
+			xfs_dir2_data_freeinsert(hdr, bf, newdup, needlogp);
 		}
 		/*
 		 * Otherwise we need a scan if the new entry is big enough.
@@ -873,7 +892,7 @@ xfs_dir2_data_make_free(
 		*xfs_dir2_data_unused_tag_p(newdup) =
 			cpu_to_be16((char *)newdup - (char *)hdr);
 		xfs_dir2_data_log_unused(tp, bp, newdup);
-		xfs_dir2_data_freeinsert(hdr, newdup, needlogp);
+		xfs_dir2_data_freeinsert(hdr, bf, newdup, needlogp);
 	}
 	*needscanp = needscan;
 }
@@ -884,6 +903,7 @@ xfs_dir2_data_make_free(
 void
 xfs_dir2_data_use_free(
 	struct xfs_trans	*tp,
+	struct xfs_inode	*dp,
 	struct xfs_buf		*bp,
 	xfs_dir2_data_unused_t	*dup,		/* unused entry */
 	xfs_dir2_data_aoff_t	offset,		/* starting offset to use */
@@ -913,9 +933,9 @@ xfs_dir2_data_use_free(
 	/*
 	 * Look up the entry in the bestfree table.
 	 */
-	dfp = xfs_dir2_data_freefind(hdr, dup);
 	oldlen = be16_to_cpu(dup->length);
-	bf = xfs_dir3_data_bestfree_p(hdr);
+	bf = dp->d_ops->data_bestfree_p(hdr);
+	dfp = xfs_dir2_data_freefind(hdr, bf, dup);
 	ASSERT(dfp || oldlen <= be16_to_cpu(bf[2].length));
 	/*
 	 * Check for alignment with front and back of the entry.
@@ -932,7 +952,8 @@ xfs_dir2_data_use_free(
 		if (dfp) {
 			needscan = (bf[2].offset != 0);
 			if (!needscan)
-				xfs_dir2_data_freeremove(hdr, dfp, needlogp);
+				xfs_dir2_data_freeremove(hdr, bf, dfp,
+							 needlogp);
 		}
 	}
 	/*
@@ -950,8 +971,9 @@ xfs_dir2_data_use_free(
 		 * If it was in the table, remove it and add the new one.
 		 */
 		if (dfp) {
-			xfs_dir2_data_freeremove(hdr, dfp, needlogp);
-			dfp = xfs_dir2_data_freeinsert(hdr, newdup, needlogp);
+			xfs_dir2_data_freeremove(hdr, bf, dfp, needlogp);
+			dfp = xfs_dir2_data_freeinsert(hdr, bf, newdup,
+						       needlogp);
 			ASSERT(dfp != NULL);
 			ASSERT(dfp->length == newdup->length);
 			ASSERT(be16_to_cpu(dfp->offset) == (char *)newdup - (char *)hdr);
@@ -977,8 +999,9 @@ xfs_dir2_data_use_free(
 		 * If it was in the table, remove it and add the new one.
 		 */
 		if (dfp) {
-			xfs_dir2_data_freeremove(hdr, dfp, needlogp);
-			dfp = xfs_dir2_data_freeinsert(hdr, newdup, needlogp);
+			xfs_dir2_data_freeremove(hdr, bf, dfp, needlogp);
+			dfp = xfs_dir2_data_freeinsert(hdr, bf, newdup,
+						       needlogp);
 			ASSERT(dfp != NULL);
 			ASSERT(dfp->length == newdup->length);
 			ASSERT(be16_to_cpu(dfp->offset) == (char *)newdup - (char *)hdr);
@@ -1017,9 +1040,11 @@ xfs_dir2_data_use_free(
 		if (dfp) {
 			needscan = (bf[2].length != 0);
 			if (!needscan) {
-				xfs_dir2_data_freeremove(hdr, dfp, needlogp);
-				xfs_dir2_data_freeinsert(hdr, newdup, needlogp);
-				xfs_dir2_data_freeinsert(hdr, newdup2,
+				xfs_dir2_data_freeremove(hdr, bf, dfp,
+							 needlogp);
+				xfs_dir2_data_freeinsert(hdr, bf, newdup,
+							 needlogp);
+				xfs_dir2_data_freeinsert(hdr, bf, newdup2,
 							 needlogp);
 			}
 		}
diff --git a/fs/xfs/xfs_dir2_leaf.c b/fs/xfs/xfs_dir2_leaf.c
index 1021c8356d08..ae47ec6e16c4 100644
--- a/fs/xfs/xfs_dir2_leaf.c
+++ b/fs/xfs/xfs_dir2_leaf.c
@@ -18,23 +18,21 @@
  */
 #include "xfs.h"
 #include "xfs_fs.h"
-#include "xfs_types.h"
-#include "xfs_bit.h"
-#include "xfs_log.h"
-#include "xfs_trans.h"
+#include "xfs_format.h"
+#include "xfs_log_format.h"
+#include "xfs_trans_resv.h"
 #include "xfs_sb.h"
 #include "xfs_ag.h"
 #include "xfs_mount.h"
+#include "xfs_da_format.h"
 #include "xfs_da_btree.h"
-#include "xfs_bmap_btree.h"
-#include "xfs_dinode.h"
 #include "xfs_inode.h"
 #include "xfs_bmap.h"
-#include "xfs_dir2_format.h"
 #include "xfs_dir2.h"
 #include "xfs_dir2_priv.h"
 #include "xfs_error.h"
 #include "xfs_trace.h"
+#include "xfs_trans.h"
 #include "xfs_buf_item.h"
 #include "xfs_cksum.h"
 
@@ -52,21 +50,21 @@ static void xfs_dir3_leaf_log_tail(struct xfs_trans *tp, struct xfs_buf *bp);
  * Pop an assert if something is wrong.
  */
 #ifdef DEBUG
-#define	xfs_dir3_leaf_check(mp, bp) \
+#define	xfs_dir3_leaf_check(dp, bp) \
 do { \
-	if (!xfs_dir3_leaf1_check((mp), (bp))) \
+	if (!xfs_dir3_leaf1_check((dp), (bp))) \
 		ASSERT(0); \
 } while (0);
 
 STATIC bool
 xfs_dir3_leaf1_check(
-	struct xfs_mount	*mp,
+	struct xfs_inode	*dp,
 	struct xfs_buf		*bp)
 {
 	struct xfs_dir2_leaf	*leaf = bp->b_addr;
 	struct xfs_dir3_icleaf_hdr leafhdr;
 
-	xfs_dir3_leaf_hdr_from_disk(&leafhdr, leaf);
+	dp->d_ops->leaf_hdr_from_disk(&leafhdr, leaf);
 
 	if (leafhdr.magic == XFS_DIR3_LEAF1_MAGIC) {
 		struct xfs_dir3_leaf_hdr *leaf3 = bp->b_addr;
@@ -75,71 +73,16 @@ xfs_dir3_leaf1_check(
 	} else if (leafhdr.magic != XFS_DIR2_LEAF1_MAGIC)
 		return false;
 
-	return xfs_dir3_leaf_check_int(mp, &leafhdr, leaf);
+	return xfs_dir3_leaf_check_int(dp->i_mount, dp, &leafhdr, leaf);
 }
 #else
-#define	xfs_dir3_leaf_check(mp, bp)
+#define	xfs_dir3_leaf_check(dp, bp)
 #endif
 
-void
-xfs_dir3_leaf_hdr_from_disk(
-	struct xfs_dir3_icleaf_hdr	*to,
-	struct xfs_dir2_leaf		*from)
-{
-	if (from->hdr.info.magic == cpu_to_be16(XFS_DIR2_LEAF1_MAGIC) ||
-	    from->hdr.info.magic == cpu_to_be16(XFS_DIR2_LEAFN_MAGIC)) {
-		to->forw = be32_to_cpu(from->hdr.info.forw);
-		to->back = be32_to_cpu(from->hdr.info.back);
-		to->magic = be16_to_cpu(from->hdr.info.magic);
-		to->count = be16_to_cpu(from->hdr.count);
-		to->stale = be16_to_cpu(from->hdr.stale);
-	} else {
-		struct xfs_dir3_leaf_hdr *hdr3 = (struct xfs_dir3_leaf_hdr *)from;
-
-		to->forw = be32_to_cpu(hdr3->info.hdr.forw);
-		to->back = be32_to_cpu(hdr3->info.hdr.back);
-		to->magic = be16_to_cpu(hdr3->info.hdr.magic);
-		to->count = be16_to_cpu(hdr3->count);
-		to->stale = be16_to_cpu(hdr3->stale);
-	}
-
-	ASSERT(to->magic == XFS_DIR2_LEAF1_MAGIC ||
-	       to->magic == XFS_DIR3_LEAF1_MAGIC ||
-	       to->magic == XFS_DIR2_LEAFN_MAGIC ||
-	       to->magic == XFS_DIR3_LEAFN_MAGIC);
-}
-
-void
-xfs_dir3_leaf_hdr_to_disk(
-	struct xfs_dir2_leaf		*to,
-	struct xfs_dir3_icleaf_hdr	*from)
-{
-	ASSERT(from->magic == XFS_DIR2_LEAF1_MAGIC ||
-	       from->magic == XFS_DIR3_LEAF1_MAGIC ||
-	       from->magic == XFS_DIR2_LEAFN_MAGIC ||
-	       from->magic == XFS_DIR3_LEAFN_MAGIC);
-
-	if (from->magic == XFS_DIR2_LEAF1_MAGIC ||
-	    from->magic == XFS_DIR2_LEAFN_MAGIC) {
-		to->hdr.info.forw = cpu_to_be32(from->forw);
-		to->hdr.info.back = cpu_to_be32(from->back);
-		to->hdr.info.magic = cpu_to_be16(from->magic);
-		to->hdr.count = cpu_to_be16(from->count);
-		to->hdr.stale = cpu_to_be16(from->stale);
-	} else {
-		struct xfs_dir3_leaf_hdr *hdr3 = (struct xfs_dir3_leaf_hdr *)to;
-
-		hdr3->info.hdr.forw = cpu_to_be32(from->forw);
-		hdr3->info.hdr.back = cpu_to_be32(from->back);
-		hdr3->info.hdr.magic = cpu_to_be16(from->magic);
-		hdr3->count = cpu_to_be16(from->count);
-		hdr3->stale = cpu_to_be16(from->stale);
-	}
-}
-
 bool
 xfs_dir3_leaf_check_int(
 	struct xfs_mount	*mp,
+	struct xfs_inode	*dp,
 	struct xfs_dir3_icleaf_hdr *hdr,
 	struct xfs_dir2_leaf	*leaf)
 {
@@ -147,8 +90,21 @@ xfs_dir3_leaf_check_int(
 	xfs_dir2_leaf_tail_t	*ltp;
 	int			stale;
 	int			i;
+	const struct xfs_dir_ops *ops;
+	struct xfs_dir3_icleaf_hdr leafhdr;
 
-	ents = xfs_dir3_leaf_ents_p(leaf);
+	/*
+	 * we can be passed a null dp here from a verifier, so we need to go the
+	 * hard way to get them.
+	 */
+	ops = xfs_dir_get_ops(mp, dp);
+
+	if (!hdr) {
+		ops->leaf_hdr_from_disk(&leafhdr, leaf);
+		hdr = &leafhdr;
+	}
+
+	ents = ops->leaf_ents_p(leaf);
 	ltp = xfs_dir2_leaf_tail_p(mp, leaf);
 
 	/*
@@ -156,7 +112,7 @@ xfs_dir3_leaf_check_int(
 	 * Should factor in the size of the bests table as well.
 	 * We can deduce a value for that from di_size.
 	 */
-	if (hdr->count > xfs_dir3_max_leaf_ents(mp, leaf))
+	if (hdr->count > ops->leaf_max_ents(mp))
 		return false;
 
 	/* Leaves and bests don't overlap in leaf format. */
@@ -192,7 +148,6 @@ xfs_dir3_leaf_verify(
 {
 	struct xfs_mount	*mp = bp->b_target->bt_mount;
 	struct xfs_dir2_leaf	*leaf = bp->b_addr;
-	struct xfs_dir3_icleaf_hdr leafhdr;
 
 	ASSERT(magic == XFS_DIR2_LEAF1_MAGIC || magic == XFS_DIR2_LEAFN_MAGIC);
 
@@ -214,8 +169,7 @@ xfs_dir3_leaf_verify(
 			return false;
 	}
 
-	xfs_dir3_leaf_hdr_from_disk(&leafhdr, leaf);
-	return xfs_dir3_leaf_check_int(mp, &leafhdr, leaf);
+	return xfs_dir3_leaf_check_int(mp, NULL, NULL, leaf);
 }
 
 static void
@@ -401,7 +355,7 @@ xfs_dir3_leaf_get_buf(
 		return error;
 
 	xfs_dir3_leaf_init(mp, tp, bp, dp->i_ino, magic);
-	xfs_dir3_leaf_log_header(tp, bp);
+	xfs_dir3_leaf_log_header(tp, dp, bp);
 	if (magic == XFS_DIR2_LEAF1_MAGIC)
 		xfs_dir3_leaf_log_tail(tp, bp);
 	*bpp = bp;
@@ -462,31 +416,31 @@ xfs_dir2_block_to_leaf(
 	xfs_dir3_data_check(dp, dbp);
 	btp = xfs_dir2_block_tail_p(mp, hdr);
 	blp = xfs_dir2_block_leaf_p(btp);
-	bf = xfs_dir3_data_bestfree_p(hdr);
-	ents = xfs_dir3_leaf_ents_p(leaf);
+	bf = dp->d_ops->data_bestfree_p(hdr);
+	ents = dp->d_ops->leaf_ents_p(leaf);
 
 	/*
 	 * Set the counts in the leaf header.
 	 */
-	xfs_dir3_leaf_hdr_from_disk(&leafhdr, leaf);
+	dp->d_ops->leaf_hdr_from_disk(&leafhdr, leaf);
 	leafhdr.count = be32_to_cpu(btp->count);
 	leafhdr.stale = be32_to_cpu(btp->stale);
-	xfs_dir3_leaf_hdr_to_disk(leaf, &leafhdr);
-	xfs_dir3_leaf_log_header(tp, lbp);
+	dp->d_ops->leaf_hdr_to_disk(leaf, &leafhdr);
+	xfs_dir3_leaf_log_header(tp, dp, lbp);
 
 	/*
 	 * Could compact these but I think we always do the conversion
 	 * after squeezing out stale entries.
 	 */
 	memcpy(ents, blp, be32_to_cpu(btp->count) * sizeof(xfs_dir2_leaf_entry_t));
-	xfs_dir3_leaf_log_ents(tp, lbp, 0, leafhdr.count - 1);
+	xfs_dir3_leaf_log_ents(tp, dp, lbp, 0, leafhdr.count - 1);
 	needscan = 0;
 	needlog = 1;
 	/*
 	 * Make the space formerly occupied by the leaf entries and block
 	 * tail be free.
 	 */
-	xfs_dir2_data_make_free(tp, dbp,
+	xfs_dir2_data_make_free(tp, dp, dbp,
 		(xfs_dir2_data_aoff_t)((char *)blp - (char *)hdr),
 		(xfs_dir2_data_aoff_t)((char *)hdr + mp->m_dirblksize -
 				       (char *)blp),
@@ -502,7 +456,7 @@ xfs_dir2_block_to_leaf(
 		hdr->magic = cpu_to_be32(XFS_DIR3_DATA_MAGIC);
 
 	if (needscan)
-		xfs_dir2_data_freescan(mp, hdr, &needlog);
+		xfs_dir2_data_freescan(dp, hdr, &needlog);
 	/*
 	 * Set up leaf tail and bests table.
 	 */
@@ -514,8 +468,8 @@ xfs_dir2_block_to_leaf(
 	 * Log the data header and leaf bests table.
 	 */
 	if (needlog)
-		xfs_dir2_data_log_header(tp, dbp);
-	xfs_dir3_leaf_check(mp, lbp);
+		xfs_dir2_data_log_header(tp, dp, dbp);
+	xfs_dir3_leaf_check(dp, lbp);
 	xfs_dir3_data_check(dp, dbp);
 	xfs_dir3_leaf_log_bests(tp, lbp, 0, 0);
 	return 0;
@@ -699,10 +653,10 @@ xfs_dir2_leaf_addname(
 	index = xfs_dir2_leaf_search_hash(args, lbp);
 	leaf = lbp->b_addr;
 	ltp = xfs_dir2_leaf_tail_p(mp, leaf);
-	ents = xfs_dir3_leaf_ents_p(leaf);
-	xfs_dir3_leaf_hdr_from_disk(&leafhdr, leaf);
+	ents = dp->d_ops->leaf_ents_p(leaf);
+	dp->d_ops->leaf_hdr_from_disk(&leafhdr, leaf);
 	bestsp = xfs_dir2_leaf_bests_p(ltp);
-	length = xfs_dir3_data_entsize(mp, args->namelen);
+	length = dp->d_ops->data_entsize(args->namelen);
 
 	/*
 	 * See if there are any entries with the same hash value
@@ -864,7 +818,7 @@ xfs_dir2_leaf_addname(
 		else
 			xfs_dir3_leaf_log_bests(tp, lbp, use_block, use_block);
 		hdr = dbp->b_addr;
-		bf = xfs_dir3_data_bestfree_p(hdr);
+		bf = dp->d_ops->data_bestfree_p(hdr);
 		bestsp[use_block] = bf[0].length;
 		grown = 1;
 	} else {
@@ -880,7 +834,7 @@ xfs_dir2_leaf_addname(
 			return error;
 		}
 		hdr = dbp->b_addr;
-		bf = xfs_dir3_data_bestfree_p(hdr);
+		bf = dp->d_ops->data_bestfree_p(hdr);
 		grown = 0;
 	}
 	/*
@@ -893,7 +847,7 @@ xfs_dir2_leaf_addname(
 	/*
 	 * Mark the initial part of our freespace in use for the new entry.
 	 */
-	xfs_dir2_data_use_free(tp, dbp, dup,
+	xfs_dir2_data_use_free(tp, dp, dbp, dup,
 		(xfs_dir2_data_aoff_t)((char *)dup - (char *)hdr), length,
 		&needlog, &needscan);
 	/*
@@ -903,20 +857,20 @@ xfs_dir2_leaf_addname(
 	dep->inumber = cpu_to_be64(args->inumber);
 	dep->namelen = args->namelen;
 	memcpy(dep->name, args->name, dep->namelen);
-	xfs_dir3_dirent_put_ftype(mp, dep, args->filetype);
-	tagp = xfs_dir3_data_entry_tag_p(mp, dep);
+	dp->d_ops->data_put_ftype(dep, args->filetype);
+	tagp = dp->d_ops->data_entry_tag_p(dep);
 	*tagp = cpu_to_be16((char *)dep - (char *)hdr);
 	/*
 	 * Need to scan fix up the bestfree table.
 	 */
 	if (needscan)
-		xfs_dir2_data_freescan(mp, hdr, &needlog);
+		xfs_dir2_data_freescan(dp, hdr, &needlog);
 	/*
 	 * Need to log the data block's header.
 	 */
 	if (needlog)
-		xfs_dir2_data_log_header(tp, dbp);
-	xfs_dir2_data_log_entry(tp, dbp, dep);
+		xfs_dir2_data_log_header(tp, dp, dbp);
+	xfs_dir2_data_log_entry(tp, dp, dbp, dep);
 	/*
 	 * If the bests table needs to be changed, do it.
 	 * Log the change unless we've already done that.
@@ -939,10 +893,10 @@ xfs_dir2_leaf_addname(
 	/*
 	 * Log the leaf fields and give up the buffers.
 	 */
-	xfs_dir3_leaf_hdr_to_disk(leaf, &leafhdr);
-	xfs_dir3_leaf_log_header(tp, lbp);
-	xfs_dir3_leaf_log_ents(tp, lbp, lfloglow, lfloghigh);
-	xfs_dir3_leaf_check(mp, lbp);
+	dp->d_ops->leaf_hdr_to_disk(leaf, &leafhdr);
+	xfs_dir3_leaf_log_header(tp, dp, lbp);
+	xfs_dir3_leaf_log_ents(tp, dp, lbp, lfloglow, lfloghigh);
+	xfs_dir3_leaf_check(dp, lbp);
 	xfs_dir3_data_check(dp, dbp);
 	return 0;
 }
@@ -962,6 +916,7 @@ xfs_dir3_leaf_compact(
 	int		loglow;		/* first leaf entry to log */
 	int		to;		/* target leaf index */
 	struct xfs_dir2_leaf_entry *ents;
+	struct xfs_inode *dp = args->dp;
 
 	leaf = bp->b_addr;
 	if (!leafhdr->stale)
@@ -970,7 +925,7 @@ xfs_dir3_leaf_compact(
 	/*
 	 * Compress out the stale entries in place.
 	 */
-	ents = xfs_dir3_leaf_ents_p(leaf);
+	ents = dp->d_ops->leaf_ents_p(leaf);
 	for (from = to = 0, loglow = -1; from < leafhdr->count; from++) {
 		if (ents[from].address == cpu_to_be32(XFS_DIR2_NULL_DATAPTR))
 			continue;
@@ -991,10 +946,10 @@ xfs_dir3_leaf_compact(
 	leafhdr->count -= leafhdr->stale;
 	leafhdr->stale = 0;
 
-	xfs_dir3_leaf_hdr_to_disk(leaf, leafhdr);
-	xfs_dir3_leaf_log_header(args->trans, bp);
+	dp->d_ops->leaf_hdr_to_disk(leaf, leafhdr);
+	xfs_dir3_leaf_log_header(args->trans, dp, bp);
 	if (loglow != -1)
-		xfs_dir3_leaf_log_ents(args->trans, bp, loglow, to - 1);
+		xfs_dir3_leaf_log_ents(args->trans, dp, bp, loglow, to - 1);
 }
 
 /*
@@ -1121,10 +1076,11 @@ xfs_dir3_leaf_log_bests(
  */
 void
 xfs_dir3_leaf_log_ents(
-	xfs_trans_t		*tp,		/* transaction pointer */
-	struct xfs_buf		*bp,		/* leaf buffer */
-	int			first,		/* first entry to log */
-	int			last)		/* last entry to log */
+	struct xfs_trans	*tp,
+	struct xfs_inode	*dp,
+	struct xfs_buf		*bp,
+	int			first,
+	int			last)
 {
 	xfs_dir2_leaf_entry_t	*firstlep;	/* pointer to first entry */
 	xfs_dir2_leaf_entry_t	*lastlep;	/* pointer to last entry */
@@ -1136,7 +1092,7 @@ xfs_dir3_leaf_log_ents(
 	       leaf->hdr.info.magic == cpu_to_be16(XFS_DIR2_LEAFN_MAGIC) ||
 	       leaf->hdr.info.magic == cpu_to_be16(XFS_DIR3_LEAFN_MAGIC));
 
-	ents = xfs_dir3_leaf_ents_p(leaf);
+	ents = dp->d_ops->leaf_ents_p(leaf);
 	firstlep = &ents[first];
 	lastlep = &ents[last];
 	xfs_trans_log_buf(tp, bp, (uint)((char *)firstlep - (char *)leaf),
@@ -1149,6 +1105,7 @@ xfs_dir3_leaf_log_ents(
 void
 xfs_dir3_leaf_log_header(
 	struct xfs_trans	*tp,
+	struct xfs_inode	*dp,
 	struct xfs_buf		*bp)
 {
 	struct xfs_dir2_leaf	*leaf = bp->b_addr;
@@ -1159,7 +1116,7 @@ xfs_dir3_leaf_log_header(
 	       leaf->hdr.info.magic == cpu_to_be16(XFS_DIR3_LEAFN_MAGIC));
 
 	xfs_trans_log_buf(tp, bp, (uint)((char *)&leaf->hdr - (char *)leaf),
-			  xfs_dir3_leaf_hdr_size(leaf) - 1);
+			  dp->d_ops->leaf_hdr_size - 1);
 }
 
 /*
@@ -1214,9 +1171,9 @@ xfs_dir2_leaf_lookup(
 	}
 	tp = args->trans;
 	dp = args->dp;
-	xfs_dir3_leaf_check(dp->i_mount, lbp);
+	xfs_dir3_leaf_check(dp, lbp);
 	leaf = lbp->b_addr;
-	ents = xfs_dir3_leaf_ents_p(leaf);
+	ents = dp->d_ops->leaf_ents_p(leaf);
 	/*
 	 * Get to the leaf entry and contained data entry address.
 	 */
@@ -1232,7 +1189,7 @@ xfs_dir2_leaf_lookup(
 	 * Return the found inode number & CI name if appropriate
 	 */
 	args->inumber = be64_to_cpu(dep->inumber);
-	args->filetype = xfs_dir3_dirent_get_ftype(dp->i_mount, dep);
+	args->filetype = dp->d_ops->data_get_ftype(dep);
 	error = xfs_dir_cilookup_result(args, dep->name, dep->namelen);
 	xfs_trans_brelse(tp, dbp);
 	xfs_trans_brelse(tp, lbp);
@@ -1279,9 +1236,9 @@ xfs_dir2_leaf_lookup_int(
 
 	*lbpp = lbp;
 	leaf = lbp->b_addr;
-	xfs_dir3_leaf_check(mp, lbp);
-	ents = xfs_dir3_leaf_ents_p(leaf);
-	xfs_dir3_leaf_hdr_from_disk(&leafhdr, leaf);
+	xfs_dir3_leaf_check(dp, lbp);
+	ents = dp->d_ops->leaf_ents_p(leaf);
+	dp->d_ops->leaf_hdr_from_disk(&leafhdr, leaf);
 
 	/*
 	 * Look for the first leaf entry with our hash value.
@@ -1415,9 +1372,9 @@ xfs_dir2_leaf_removename(
 	leaf = lbp->b_addr;
 	hdr = dbp->b_addr;
 	xfs_dir3_data_check(dp, dbp);
-	bf = xfs_dir3_data_bestfree_p(hdr);
-	xfs_dir3_leaf_hdr_from_disk(&leafhdr, leaf);
-	ents = xfs_dir3_leaf_ents_p(leaf);
+	bf = dp->d_ops->data_bestfree_p(hdr);
+	dp->d_ops->leaf_hdr_from_disk(&leafhdr, leaf);
+	ents = dp->d_ops->leaf_ents_p(leaf);
 	/*
 	 * Point to the leaf entry, use that to point to the data entry.
 	 */
@@ -1433,27 +1390,27 @@ xfs_dir2_leaf_removename(
 	/*
 	 * Mark the former data entry unused.
 	 */
-	xfs_dir2_data_make_free(tp, dbp,
+	xfs_dir2_data_make_free(tp, dp, dbp,
 		(xfs_dir2_data_aoff_t)((char *)dep - (char *)hdr),
-		xfs_dir3_data_entsize(mp, dep->namelen), &needlog, &needscan);
+		dp->d_ops->data_entsize(dep->namelen), &needlog, &needscan);
 	/*
 	 * We just mark the leaf entry stale by putting a null in it.
 	 */
 	leafhdr.stale++;
-	xfs_dir3_leaf_hdr_to_disk(leaf, &leafhdr);
-	xfs_dir3_leaf_log_header(tp, lbp);
+	dp->d_ops->leaf_hdr_to_disk(leaf, &leafhdr);
+	xfs_dir3_leaf_log_header(tp, dp, lbp);
 
 	lep->address = cpu_to_be32(XFS_DIR2_NULL_DATAPTR);
-	xfs_dir3_leaf_log_ents(tp, lbp, index, index);
+	xfs_dir3_leaf_log_ents(tp, dp, lbp, index, index);
 
 	/*
 	 * Scan the freespace in the data block again if necessary,
 	 * log the data block header if necessary.
 	 */
 	if (needscan)
-		xfs_dir2_data_freescan(mp, hdr, &needlog);
+		xfs_dir2_data_freescan(dp, hdr, &needlog);
 	if (needlog)
-		xfs_dir2_data_log_header(tp, dbp);
+		xfs_dir2_data_log_header(tp, dp, dbp);
 	/*
 	 * If the longest freespace in the data block has changed,
 	 * put the new value in the bests table and log that.
@@ -1467,7 +1424,7 @@ xfs_dir2_leaf_removename(
 	 * If the data block is now empty then get rid of the data block.
 	 */
 	if (be16_to_cpu(bf[0].length) ==
-			mp->m_dirblksize - xfs_dir3_data_entry_offset(hdr)) {
+			mp->m_dirblksize - dp->d_ops->data_entry_offset) {
 		ASSERT(db != mp->m_dirdatablk);
 		if ((error = xfs_dir2_shrink_inode(args, db, dbp))) {
 			/*
@@ -1478,7 +1435,7 @@ xfs_dir2_leaf_removename(
 			 */
 			if (error == ENOSPC && args->total == 0)
 				error = 0;
-			xfs_dir3_leaf_check(mp, lbp);
+			xfs_dir3_leaf_check(dp, lbp);
 			return error;
 		}
 		dbp = NULL;
@@ -1512,7 +1469,7 @@ xfs_dir2_leaf_removename(
 	else if (db != mp->m_dirdatablk)
 		dbp = NULL;
 
-	xfs_dir3_leaf_check(mp, lbp);
+	xfs_dir3_leaf_check(dp, lbp);
 	/*
 	 * See if we can convert to block form.
 	 */
@@ -1547,7 +1504,7 @@ xfs_dir2_leaf_replace(
 	}
 	dp = args->dp;
 	leaf = lbp->b_addr;
-	ents = xfs_dir3_leaf_ents_p(leaf);
+	ents = dp->d_ops->leaf_ents_p(leaf);
 	/*
 	 * Point to the leaf entry, get data address from it.
 	 */
@@ -1563,10 +1520,10 @@ xfs_dir2_leaf_replace(
 	 * Put the new inode number in, log it.
 	 */
 	dep->inumber = cpu_to_be64(args->inumber);
-	xfs_dir3_dirent_put_ftype(dp->i_mount, dep, args->filetype);
+	dp->d_ops->data_put_ftype(dep, args->filetype);
 	tp = args->trans;
-	xfs_dir2_data_log_entry(tp, dbp, dep);
-	xfs_dir3_leaf_check(dp->i_mount, lbp);
+	xfs_dir2_data_log_entry(tp, dp, dbp, dep);
+	xfs_dir3_leaf_check(dp, lbp);
 	xfs_trans_brelse(tp, lbp);
 	return 0;
 }
@@ -1592,8 +1549,8 @@ xfs_dir2_leaf_search_hash(
 	struct xfs_dir3_icleaf_hdr leafhdr;
 
 	leaf = lbp->b_addr;
-	ents = xfs_dir3_leaf_ents_p(leaf);
-	xfs_dir3_leaf_hdr_from_disk(&leafhdr, leaf);
+	ents = args->dp->d_ops->leaf_ents_p(leaf);
+	args->dp->d_ops->leaf_hdr_from_disk(&leafhdr, leaf);
 
 	/*
 	 * Note, the table cannot be empty, so we have to go through the loop.
@@ -1661,12 +1618,12 @@ xfs_dir2_leaf_trim_data(
 #ifdef DEBUG
 {
 	struct xfs_dir2_data_hdr *hdr = dbp->b_addr;
-	struct xfs_dir2_data_free *bf = xfs_dir3_data_bestfree_p(hdr);
+	struct xfs_dir2_data_free *bf = dp->d_ops->data_bestfree_p(hdr);
 
 	ASSERT(hdr->magic == cpu_to_be32(XFS_DIR2_DATA_MAGIC) ||
 	       hdr->magic == cpu_to_be32(XFS_DIR3_DATA_MAGIC));
 	ASSERT(be16_to_cpu(bf[0].length) ==
-	       mp->m_dirblksize - xfs_dir3_data_entry_offset(hdr));
+	       mp->m_dirblksize - dp->d_ops->data_entry_offset);
 	ASSERT(db == be32_to_cpu(ltp->bestcount) - 1);
 }
 #endif
@@ -1782,7 +1739,7 @@ xfs_dir2_node_to_leaf(
 		return 0;
 	lbp = state->path.blk[0].bp;
 	leaf = lbp->b_addr;
-	xfs_dir3_leaf_hdr_from_disk(&leafhdr, leaf);
+	dp->d_ops->leaf_hdr_from_disk(&leafhdr, leaf);
 
 	ASSERT(leafhdr.magic == XFS_DIR2_LEAFN_MAGIC ||
 	       leafhdr.magic == XFS_DIR3_LEAFN_MAGIC);
@@ -1794,7 +1751,7 @@ xfs_dir2_node_to_leaf(
 	if (error)
 		return error;
 	free = fbp->b_addr;
-	xfs_dir3_free_hdr_from_disk(&freehdr, free);
+	dp->d_ops->free_hdr_from_disk(&freehdr, free);
 
 	ASSERT(!freehdr.firstdb);
 
@@ -1828,14 +1785,14 @@ xfs_dir2_node_to_leaf(
 	/*
 	 * Set up the leaf bests table.
 	 */
-	memcpy(xfs_dir2_leaf_bests_p(ltp), xfs_dir3_free_bests_p(mp, free),
+	memcpy(xfs_dir2_leaf_bests_p(ltp), dp->d_ops->free_bests_p(free),
 		freehdr.nvalid * sizeof(xfs_dir2_data_off_t));
 
-	xfs_dir3_leaf_hdr_to_disk(leaf, &leafhdr);
-	xfs_dir3_leaf_log_header(tp, lbp);
+	dp->d_ops->leaf_hdr_to_disk(leaf, &leafhdr);
+	xfs_dir3_leaf_log_header(tp, dp, lbp);
 	xfs_dir3_leaf_log_bests(tp, lbp, 0, be32_to_cpu(ltp->bestcount) - 1);
 	xfs_dir3_leaf_log_tail(tp, lbp);
-	xfs_dir3_leaf_check(mp, lbp);
+	xfs_dir3_leaf_check(dp, lbp);
 
 	/*
 	 * Get rid of the freespace block.
diff --git a/fs/xfs/xfs_dir2_node.c b/fs/xfs/xfs_dir2_node.c
index 4c3dba7ffb74..56369d4509d5 100644
--- a/fs/xfs/xfs_dir2_node.c
+++ b/fs/xfs/xfs_dir2_node.c
@@ -18,22 +18,21 @@
  */
 #include "xfs.h"
 #include "xfs_fs.h"
-#include "xfs_types.h"
-#include "xfs_log.h"
-#include "xfs_trans.h"
+#include "xfs_format.h"
+#include "xfs_log_format.h"
+#include "xfs_trans_resv.h"
 #include "xfs_sb.h"
 #include "xfs_ag.h"
 #include "xfs_mount.h"
+#include "xfs_da_format.h"
 #include "xfs_da_btree.h"
-#include "xfs_bmap_btree.h"
-#include "xfs_dinode.h"
 #include "xfs_inode.h"
 #include "xfs_bmap.h"
-#include "xfs_dir2_format.h"
 #include "xfs_dir2.h"
 #include "xfs_dir2_priv.h"
 #include "xfs_error.h"
 #include "xfs_trace.h"
+#include "xfs_trans.h"
 #include "xfs_buf_item.h"
 #include "xfs_cksum.h"
 
@@ -55,21 +54,21 @@ static int xfs_dir2_node_addname_int(xfs_da_args_t *args,
  * Check internal consistency of a leafn block.
  */
 #ifdef DEBUG
-#define	xfs_dir3_leaf_check(mp, bp) \
+#define	xfs_dir3_leaf_check(dp, bp) \
 do { \
-	if (!xfs_dir3_leafn_check((mp), (bp))) \
+	if (!xfs_dir3_leafn_check((dp), (bp))) \
 		ASSERT(0); \
 } while (0);
 
 static bool
 xfs_dir3_leafn_check(
-	struct xfs_mount	*mp,
+	struct xfs_inode	*dp,
 	struct xfs_buf		*bp)
 {
 	struct xfs_dir2_leaf	*leaf = bp->b_addr;
 	struct xfs_dir3_icleaf_hdr leafhdr;
 
-	xfs_dir3_leaf_hdr_from_disk(&leafhdr, leaf);
+	dp->d_ops->leaf_hdr_from_disk(&leafhdr, leaf);
 
 	if (leafhdr.magic == XFS_DIR3_LEAFN_MAGIC) {
 		struct xfs_dir3_leaf_hdr *leaf3 = bp->b_addr;
@@ -78,10 +77,10 @@ xfs_dir3_leafn_check(
 	} else if (leafhdr.magic != XFS_DIR2_LEAFN_MAGIC)
 		return false;
 
-	return xfs_dir3_leaf_check_int(mp, &leafhdr, leaf);
+	return xfs_dir3_leaf_check_int(dp->i_mount, dp, &leafhdr, leaf);
 }
 #else
-#define	xfs_dir3_leaf_check(mp, bp)
+#define	xfs_dir3_leaf_check(dp, bp)
 #endif
 
 static bool
@@ -193,53 +192,6 @@ xfs_dir2_free_try_read(
 	return __xfs_dir3_free_read(tp, dp, fbno, -2, bpp);
 }
 
-
-void
-xfs_dir3_free_hdr_from_disk(
-	struct xfs_dir3_icfree_hdr	*to,
-	struct xfs_dir2_free		*from)
-{
-	if (from->hdr.magic == cpu_to_be32(XFS_DIR2_FREE_MAGIC)) {
-		to->magic = be32_to_cpu(from->hdr.magic);
-		to->firstdb = be32_to_cpu(from->hdr.firstdb);
-		to->nvalid = be32_to_cpu(from->hdr.nvalid);
-		to->nused = be32_to_cpu(from->hdr.nused);
-	} else {
-		struct xfs_dir3_free_hdr *hdr3 = (struct xfs_dir3_free_hdr *)from;
-
-		to->magic = be32_to_cpu(hdr3->hdr.magic);
-		to->firstdb = be32_to_cpu(hdr3->firstdb);
-		to->nvalid = be32_to_cpu(hdr3->nvalid);
-		to->nused = be32_to_cpu(hdr3->nused);
-	}
-
-	ASSERT(to->magic == XFS_DIR2_FREE_MAGIC ||
-	       to->magic == XFS_DIR3_FREE_MAGIC);
-}
-
-static void
-xfs_dir3_free_hdr_to_disk(
-	struct xfs_dir2_free		*to,
-	struct xfs_dir3_icfree_hdr	*from)
-{
-	ASSERT(from->magic == XFS_DIR2_FREE_MAGIC ||
-	       from->magic == XFS_DIR3_FREE_MAGIC);
-
-	if (from->magic == XFS_DIR2_FREE_MAGIC) {
-		to->hdr.magic = cpu_to_be32(from->magic);
-		to->hdr.firstdb = cpu_to_be32(from->firstdb);
-		to->hdr.nvalid = cpu_to_be32(from->nvalid);
-		to->hdr.nused = cpu_to_be32(from->nused);
-	} else {
-		struct xfs_dir3_free_hdr *hdr3 = (struct xfs_dir3_free_hdr *)to;
-
-		hdr3->hdr.magic = cpu_to_be32(from->magic);
-		hdr3->firstdb = cpu_to_be32(from->firstdb);
-		hdr3->nvalid = cpu_to_be32(from->nvalid);
-		hdr3->nused = cpu_to_be32(from->nused);
-	}
-}
-
 static int
 xfs_dir3_free_get_buf(
 	struct xfs_trans	*tp,
@@ -277,7 +229,7 @@ xfs_dir3_free_get_buf(
 		uuid_copy(&hdr3->hdr.uuid, &mp->m_sb.sb_uuid);
 	} else
 		hdr.magic = XFS_DIR2_FREE_MAGIC;
-	xfs_dir3_free_hdr_to_disk(bp->b_addr, &hdr);
+	dp->d_ops->free_hdr_to_disk(bp->b_addr, &hdr);
 	*bpp = bp;
 	return 0;
 }
@@ -288,6 +240,7 @@ xfs_dir3_free_get_buf(
 STATIC void
 xfs_dir2_free_log_bests(
 	struct xfs_trans	*tp,
+	struct xfs_inode	*dp,
 	struct xfs_buf		*bp,
 	int			first,		/* first entry to log */
 	int			last)		/* last entry to log */
@@ -296,7 +249,7 @@ xfs_dir2_free_log_bests(
 	__be16			*bests;
 
 	free = bp->b_addr;
-	bests = xfs_dir3_free_bests_p(tp->t_mountp, free);
+	bests = dp->d_ops->free_bests_p(free);
 	ASSERT(free->hdr.magic == cpu_to_be32(XFS_DIR2_FREE_MAGIC) ||
 	       free->hdr.magic == cpu_to_be32(XFS_DIR3_FREE_MAGIC));
 	xfs_trans_log_buf(tp, bp,
@@ -311,6 +264,7 @@ xfs_dir2_free_log_bests(
 static void
 xfs_dir2_free_log_header(
 	struct xfs_trans	*tp,
+	struct xfs_inode	*dp,
 	struct xfs_buf		*bp)
 {
 #ifdef DEBUG
@@ -320,7 +274,7 @@ xfs_dir2_free_log_header(
 	ASSERT(free->hdr.magic == cpu_to_be32(XFS_DIR2_FREE_MAGIC) ||
 	       free->hdr.magic == cpu_to_be32(XFS_DIR3_FREE_MAGIC));
 #endif
-	xfs_trans_log_buf(tp, bp, 0, xfs_dir3_free_hdr_size(tp->t_mountp) - 1);
+	xfs_trans_log_buf(tp, bp, 0, dp->d_ops->free_hdr_size - 1);
 }
 
 /*
@@ -369,7 +323,7 @@ xfs_dir2_leaf_to_node(
 		return error;
 
 	free = fbp->b_addr;
-	xfs_dir3_free_hdr_from_disk(&freehdr, free);
+	dp->d_ops->free_hdr_from_disk(&freehdr, free);
 	leaf = lbp->b_addr;
 	ltp = xfs_dir2_leaf_tail_p(mp, leaf);
 	ASSERT(be32_to_cpu(ltp->bestcount) <=
@@ -380,7 +334,7 @@ xfs_dir2_leaf_to_node(
 	 * Count active entries.
 	 */
 	from = xfs_dir2_leaf_bests_p(ltp);
-	to = xfs_dir3_free_bests_p(mp, free);
+	to = dp->d_ops->free_bests_p(free);
 	for (i = n = 0; i < be32_to_cpu(ltp->bestcount); i++, from++, to++) {
 		if ((off = be16_to_cpu(*from)) != NULLDATAOFF)
 			n++;
@@ -393,9 +347,9 @@ xfs_dir2_leaf_to_node(
 	freehdr.nused = n;
 	freehdr.nvalid = be32_to_cpu(ltp->bestcount);
 
-	xfs_dir3_free_hdr_to_disk(fbp->b_addr, &freehdr);
-	xfs_dir2_free_log_bests(tp, fbp, 0, freehdr.nvalid - 1);
-	xfs_dir2_free_log_header(tp, fbp);
+	dp->d_ops->free_hdr_to_disk(fbp->b_addr, &freehdr);
+	xfs_dir2_free_log_bests(tp, dp, fbp, 0, freehdr.nvalid - 1);
+	xfs_dir2_free_log_header(tp, dp, fbp);
 
 	/*
 	 * Converting the leaf to a leafnode is just a matter of changing the
@@ -409,8 +363,8 @@ xfs_dir2_leaf_to_node(
 		leaf->hdr.info.magic = cpu_to_be16(XFS_DIR3_LEAFN_MAGIC);
 	lbp->b_ops = &xfs_dir3_leafn_buf_ops;
 	xfs_trans_buf_set_type(tp, lbp, XFS_BLFT_DIR_LEAFN_BUF);
-	xfs_dir3_leaf_log_header(tp, lbp);
-	xfs_dir3_leaf_check(mp, lbp);
+	xfs_dir3_leaf_log_header(tp, dp, lbp);
+	xfs_dir3_leaf_check(dp, lbp);
 	return 0;
 }
 
@@ -443,8 +397,8 @@ xfs_dir2_leafn_add(
 	mp = dp->i_mount;
 	tp = args->trans;
 	leaf = bp->b_addr;
-	xfs_dir3_leaf_hdr_from_disk(&leafhdr, leaf);
-	ents = xfs_dir3_leaf_ents_p(leaf);
+	dp->d_ops->leaf_hdr_from_disk(&leafhdr, leaf);
+	ents = dp->d_ops->leaf_ents_p(leaf);
 
 	/*
 	 * Quick check just to make sure we are not going to index
@@ -460,7 +414,7 @@ xfs_dir2_leafn_add(
 	 * a compact.
 	 */
 
-	if (leafhdr.count == xfs_dir3_max_leaf_ents(mp, leaf)) {
+	if (leafhdr.count == dp->d_ops->leaf_max_ents(mp)) {
 		if (!leafhdr.stale)
 			return XFS_ERROR(ENOSPC);
 		compact = leafhdr.stale > 1;
@@ -498,30 +452,30 @@ xfs_dir2_leafn_add(
 	lep->address = cpu_to_be32(xfs_dir2_db_off_to_dataptr(mp,
 				args->blkno, args->index));
 
-	xfs_dir3_leaf_hdr_to_disk(leaf, &leafhdr);
-	xfs_dir3_leaf_log_header(tp, bp);
-	xfs_dir3_leaf_log_ents(tp, bp, lfloglow, lfloghigh);
-	xfs_dir3_leaf_check(mp, bp);
+	dp->d_ops->leaf_hdr_to_disk(leaf, &leafhdr);
+	xfs_dir3_leaf_log_header(tp, dp, bp);
+	xfs_dir3_leaf_log_ents(tp, dp, bp, lfloglow, lfloghigh);
+	xfs_dir3_leaf_check(dp, bp);
 	return 0;
 }
 
 #ifdef DEBUG
 static void
 xfs_dir2_free_hdr_check(
-	struct xfs_mount *mp,
+	struct xfs_inode *dp,
 	struct xfs_buf	*bp,
 	xfs_dir2_db_t	db)
 {
 	struct xfs_dir3_icfree_hdr hdr;
 
-	xfs_dir3_free_hdr_from_disk(&hdr, bp->b_addr);
+	dp->d_ops->free_hdr_from_disk(&hdr, bp->b_addr);
 
-	ASSERT((hdr.firstdb % xfs_dir3_free_max_bests(mp)) == 0);
+	ASSERT((hdr.firstdb % dp->d_ops->free_max_bests(dp->i_mount)) == 0);
 	ASSERT(hdr.firstdb <= db);
 	ASSERT(db < hdr.firstdb + hdr.nvalid);
 }
 #else
-#define xfs_dir2_free_hdr_check(mp, dp, db)
+#define xfs_dir2_free_hdr_check(dp, bp, db)
 #endif	/* DEBUG */
 
 /*
@@ -530,6 +484,7 @@ xfs_dir2_free_hdr_check(
  */
 xfs_dahash_t					/* hash value */
 xfs_dir2_leafn_lasthash(
+	struct xfs_inode *dp,
 	struct xfs_buf	*bp,			/* leaf buffer */
 	int		*count)			/* count of entries in leaf */
 {
@@ -537,7 +492,7 @@ xfs_dir2_leafn_lasthash(
 	struct xfs_dir2_leaf_entry *ents;
 	struct xfs_dir3_icleaf_hdr leafhdr;
 
-	xfs_dir3_leaf_hdr_from_disk(&leafhdr, leaf);
+	dp->d_ops->leaf_hdr_from_disk(&leafhdr, leaf);
 
 	ASSERT(leafhdr.magic == XFS_DIR2_LEAFN_MAGIC ||
 	       leafhdr.magic == XFS_DIR3_LEAFN_MAGIC);
@@ -547,7 +502,7 @@ xfs_dir2_leafn_lasthash(
 	if (!leafhdr.count)
 		return 0;
 
-	ents = xfs_dir3_leaf_ents_p(leaf);
+	ents = dp->d_ops->leaf_ents_p(leaf);
 	return be32_to_cpu(ents[leafhdr.count - 1].hashval);
 }
 
@@ -584,10 +539,10 @@ xfs_dir2_leafn_lookup_for_addname(
 	tp = args->trans;
 	mp = dp->i_mount;
 	leaf = bp->b_addr;
-	xfs_dir3_leaf_hdr_from_disk(&leafhdr, leaf);
-	ents = xfs_dir3_leaf_ents_p(leaf);
+	dp->d_ops->leaf_hdr_from_disk(&leafhdr, leaf);
+	ents = dp->d_ops->leaf_ents_p(leaf);
 
-	xfs_dir3_leaf_check(mp, bp);
+	xfs_dir3_leaf_check(dp, bp);
 	ASSERT(leafhdr.count > 0);
 
 	/*
@@ -605,7 +560,7 @@ xfs_dir2_leafn_lookup_for_addname(
 		ASSERT(free->hdr.magic == cpu_to_be32(XFS_DIR2_FREE_MAGIC) ||
 		       free->hdr.magic == cpu_to_be32(XFS_DIR3_FREE_MAGIC));
 	}
-	length = xfs_dir3_data_entsize(mp, args->namelen);
+	length = dp->d_ops->data_entsize(args->namelen);
 	/*
 	 * Loop over leaf entries with the right hash value.
 	 */
@@ -637,7 +592,7 @@ xfs_dir2_leafn_lookup_for_addname(
 			 * Convert the data block to the free block
 			 * holding its freespace information.
 			 */
-			newfdb = xfs_dir2_db_to_fdb(mp, newdb);
+			newfdb = dp->d_ops->db_to_fdb(mp, newdb);
 			/*
 			 * If it's not the one we have in hand, read it in.
 			 */
@@ -655,16 +610,16 @@ xfs_dir2_leafn_lookup_for_addname(
 					return error;
 				free = curbp->b_addr;
 
-				xfs_dir2_free_hdr_check(mp, curbp, curdb);
+				xfs_dir2_free_hdr_check(dp, curbp, curdb);
 			}
 			/*
 			 * Get the index for our entry.
 			 */
-			fi = xfs_dir2_db_to_fdindex(mp, curdb);
+			fi = dp->d_ops->db_to_fdindex(mp, curdb);
 			/*
 			 * If it has room, return it.
 			 */
-			bests = xfs_dir3_free_bests_p(mp, free);
+			bests = dp->d_ops->free_bests_p(free);
 			if (unlikely(bests[fi] == cpu_to_be16(NULLDATAOFF))) {
 				XFS_ERROR_REPORT("xfs_dir2_leafn_lookup_int",
 							XFS_ERRLEVEL_LOW, mp);
@@ -734,10 +689,10 @@ xfs_dir2_leafn_lookup_for_entry(
 	tp = args->trans;
 	mp = dp->i_mount;
 	leaf = bp->b_addr;
-	xfs_dir3_leaf_hdr_from_disk(&leafhdr, leaf);
-	ents = xfs_dir3_leaf_ents_p(leaf);
+	dp->d_ops->leaf_hdr_from_disk(&leafhdr, leaf);
+	ents = dp->d_ops->leaf_ents_p(leaf);
 
-	xfs_dir3_leaf_check(mp, bp);
+	xfs_dir3_leaf_check(dp, bp);
 	ASSERT(leafhdr.count > 0);
 
 	/*
@@ -816,7 +771,7 @@ xfs_dir2_leafn_lookup_for_entry(
 				xfs_trans_brelse(tp, state->extrablk.bp);
 			args->cmpresult = cmp;
 			args->inumber = be64_to_cpu(dep->inumber);
-			args->filetype = xfs_dir3_dirent_get_ftype(mp, dep);
+			args->filetype = dp->d_ops->data_get_ftype(dep);
 			*indexp = index;
 			state->extravalid = 1;
 			state->extrablk.bp = curbp;
@@ -907,7 +862,7 @@ xfs_dir3_leafn_moveents(
 	if (start_d < dhdr->count) {
 		memmove(&dents[start_d + count], &dents[start_d],
 			(dhdr->count - start_d) * sizeof(xfs_dir2_leaf_entry_t));
-		xfs_dir3_leaf_log_ents(tp, bp_d, start_d + count,
+		xfs_dir3_leaf_log_ents(tp, args->dp, bp_d, start_d + count,
 				       count + dhdr->count - 1);
 	}
 	/*
@@ -929,7 +884,8 @@ xfs_dir3_leafn_moveents(
 	 */
 	memcpy(&dents[start_d], &sents[start_s],
 		count * sizeof(xfs_dir2_leaf_entry_t));
-	xfs_dir3_leaf_log_ents(tp, bp_d, start_d, start_d + count - 1);
+	xfs_dir3_leaf_log_ents(tp, args->dp, bp_d,
+			       start_d, start_d + count - 1);
 
 	/*
 	 * If there are source entries after the ones we copied,
@@ -938,7 +894,8 @@ xfs_dir3_leafn_moveents(
 	if (start_s + count < shdr->count) {
 		memmove(&sents[start_s], &sents[start_s + count],
 			count * sizeof(xfs_dir2_leaf_entry_t));
-		xfs_dir3_leaf_log_ents(tp, bp_s, start_s, start_s + count - 1);
+		xfs_dir3_leaf_log_ents(tp, args->dp, bp_s,
+				       start_s, start_s + count - 1);
 	}
 
 	/*
@@ -956,6 +913,7 @@ xfs_dir3_leafn_moveents(
  */
 int						/* sort order */
 xfs_dir2_leafn_order(
+	struct xfs_inode	*dp,
 	struct xfs_buf		*leaf1_bp,		/* leaf1 buffer */
 	struct xfs_buf		*leaf2_bp)		/* leaf2 buffer */
 {
@@ -966,10 +924,10 @@ xfs_dir2_leafn_order(
 	struct xfs_dir3_icleaf_hdr hdr1;
 	struct xfs_dir3_icleaf_hdr hdr2;
 
-	xfs_dir3_leaf_hdr_from_disk(&hdr1, leaf1);
-	xfs_dir3_leaf_hdr_from_disk(&hdr2, leaf2);
-	ents1 = xfs_dir3_leaf_ents_p(leaf1);
-	ents2 = xfs_dir3_leaf_ents_p(leaf2);
+	dp->d_ops->leaf_hdr_from_disk(&hdr1, leaf1);
+	dp->d_ops->leaf_hdr_from_disk(&hdr2, leaf2);
+	ents1 = dp->d_ops->leaf_ents_p(leaf1);
+	ents2 = dp->d_ops->leaf_ents_p(leaf2);
 
 	if (hdr1.count > 0 && hdr2.count > 0 &&
 	    (be32_to_cpu(ents2[0].hashval) < be32_to_cpu(ents1[0].hashval) ||
@@ -1007,12 +965,13 @@ xfs_dir2_leafn_rebalance(
 	struct xfs_dir2_leaf_entry *ents2;
 	struct xfs_dir3_icleaf_hdr hdr1;
 	struct xfs_dir3_icleaf_hdr hdr2;
+	struct xfs_inode	*dp = state->args->dp;
 
 	args = state->args;
 	/*
 	 * If the block order is wrong, swap the arguments.
 	 */
-	if ((swap = xfs_dir2_leafn_order(blk1->bp, blk2->bp))) {
+	if ((swap = xfs_dir2_leafn_order(dp, blk1->bp, blk2->bp))) {
 		xfs_da_state_blk_t	*tmp;	/* temp for block swap */
 
 		tmp = blk1;
@@ -1021,10 +980,10 @@ xfs_dir2_leafn_rebalance(
 	}
 	leaf1 = blk1->bp->b_addr;
 	leaf2 = blk2->bp->b_addr;
-	xfs_dir3_leaf_hdr_from_disk(&hdr1, leaf1);
-	xfs_dir3_leaf_hdr_from_disk(&hdr2, leaf2);
-	ents1 = xfs_dir3_leaf_ents_p(leaf1);
-	ents2 = xfs_dir3_leaf_ents_p(leaf2);
+	dp->d_ops->leaf_hdr_from_disk(&hdr1, leaf1);
+	dp->d_ops->leaf_hdr_from_disk(&hdr2, leaf2);
+	ents1 = dp->d_ops->leaf_ents_p(leaf1);
+	ents2 = dp->d_ops->leaf_ents_p(leaf2);
 
 	oldsum = hdr1.count + hdr2.count;
 #if defined(DEBUG) || defined(XFS_WARN)
@@ -1070,13 +1029,13 @@ xfs_dir2_leafn_rebalance(
 	ASSERT(hdr1.stale + hdr2.stale == oldstale);
 
 	/* log the changes made when moving the entries */
-	xfs_dir3_leaf_hdr_to_disk(leaf1, &hdr1);
-	xfs_dir3_leaf_hdr_to_disk(leaf2, &hdr2);
-	xfs_dir3_leaf_log_header(args->trans, blk1->bp);
-	xfs_dir3_leaf_log_header(args->trans, blk2->bp);
+	dp->d_ops->leaf_hdr_to_disk(leaf1, &hdr1);
+	dp->d_ops->leaf_hdr_to_disk(leaf2, &hdr2);
+	xfs_dir3_leaf_log_header(args->trans, dp, blk1->bp);
+	xfs_dir3_leaf_log_header(args->trans, dp, blk2->bp);
 
-	xfs_dir3_leaf_check(args->dp->i_mount, blk1->bp);
-	xfs_dir3_leaf_check(args->dp->i_mount, blk2->bp);
+	xfs_dir3_leaf_check(dp, blk1->bp);
+	xfs_dir3_leaf_check(dp, blk2->bp);
 
 	/*
 	 * Mark whether we're inserting into the old or new leaf.
@@ -1097,11 +1056,11 @@ xfs_dir2_leafn_rebalance(
 	 * Finally sanity check just to make sure we are not returning a
 	 * negative index
 	 */
-	if(blk2->index < 0) {
+	if (blk2->index < 0) {
 		state->inleaf = 1;
 		blk2->index = 0;
-		xfs_alert(args->dp->i_mount,
-	"%s: picked the wrong leaf? reverting original leaf: blk1->index %d\n",
+		xfs_alert(dp->i_mount,
+	"%s: picked the wrong leaf? reverting original leaf: blk1->index %d",
 			__func__, blk1->index);
 	}
 }
@@ -1120,17 +1079,17 @@ xfs_dir3_data_block_free(
 	int			logfree = 0;
 	__be16			*bests;
 	struct xfs_dir3_icfree_hdr freehdr;
+	struct xfs_inode	*dp = args->dp;
 
-	xfs_dir3_free_hdr_from_disk(&freehdr, free);
-
-	bests = xfs_dir3_free_bests_p(tp->t_mountp, free);
+	dp->d_ops->free_hdr_from_disk(&freehdr, free);
+	bests = dp->d_ops->free_bests_p(free);
 	if (hdr) {
 		/*
 		 * Data block is not empty, just set the free entry to the new
 		 * value.
 		 */
 		bests[findex] = cpu_to_be16(longest);
-		xfs_dir2_free_log_bests(tp, fbp, findex, findex);
+		xfs_dir2_free_log_bests(tp, dp, fbp, findex, findex);
 		return 0;
 	}
 
@@ -1157,8 +1116,8 @@ xfs_dir3_data_block_free(
 		logfree = 1;
 	}
 
-	xfs_dir3_free_hdr_to_disk(free, &freehdr);
-	xfs_dir2_free_log_header(tp, fbp);
+	dp->d_ops->free_hdr_to_disk(free, &freehdr);
+	xfs_dir2_free_log_header(tp, dp, fbp);
 
 	/*
 	 * If there are no useful entries left in the block, get rid of the
@@ -1182,7 +1141,7 @@ xfs_dir3_data_block_free(
 
 	/* Log the free entry that changed, unless we got rid of it.  */
 	if (logfree)
-		xfs_dir2_free_log_bests(tp, fbp, findex, findex);
+		xfs_dir2_free_log_bests(tp, dp, fbp, findex, findex);
 	return 0;
 }
 
@@ -1222,8 +1181,8 @@ xfs_dir2_leafn_remove(
 	tp = args->trans;
 	mp = dp->i_mount;
 	leaf = bp->b_addr;
-	xfs_dir3_leaf_hdr_from_disk(&leafhdr, leaf);
-	ents = xfs_dir3_leaf_ents_p(leaf);
+	dp->d_ops->leaf_hdr_from_disk(&leafhdr, leaf);
+	ents = dp->d_ops->leaf_ents_p(leaf);
 
 	/*
 	 * Point to the entry we're removing.
@@ -1243,11 +1202,11 @@ xfs_dir2_leafn_remove(
 	 * Log the leaf block changes.
 	 */
 	leafhdr.stale++;
-	xfs_dir3_leaf_hdr_to_disk(leaf, &leafhdr);
-	xfs_dir3_leaf_log_header(tp, bp);
+	dp->d_ops->leaf_hdr_to_disk(leaf, &leafhdr);
+	xfs_dir3_leaf_log_header(tp, dp, bp);
 
 	lep->address = cpu_to_be32(XFS_DIR2_NULL_DATAPTR);
-	xfs_dir3_leaf_log_ents(tp, bp, index, index);
+	xfs_dir3_leaf_log_ents(tp, dp, bp, index, index);
 
 	/*
 	 * Make the data entry free.  Keep track of the longest freespace
@@ -1256,19 +1215,19 @@ xfs_dir2_leafn_remove(
 	dbp = dblk->bp;
 	hdr = dbp->b_addr;
 	dep = (xfs_dir2_data_entry_t *)((char *)hdr + off);
-	bf = xfs_dir3_data_bestfree_p(hdr);
+	bf = dp->d_ops->data_bestfree_p(hdr);
 	longest = be16_to_cpu(bf[0].length);
 	needlog = needscan = 0;
-	xfs_dir2_data_make_free(tp, dbp, off,
-		xfs_dir3_data_entsize(mp, dep->namelen), &needlog, &needscan);
+	xfs_dir2_data_make_free(tp, dp, dbp, off,
+		dp->d_ops->data_entsize(dep->namelen), &needlog, &needscan);
 	/*
 	 * Rescan the data block freespaces for bestfree.
 	 * Log the data block header if needed.
 	 */
 	if (needscan)
-		xfs_dir2_data_freescan(mp, hdr, &needlog);
+		xfs_dir2_data_freescan(dp, hdr, &needlog);
 	if (needlog)
-		xfs_dir2_data_log_header(tp, dbp);
+		xfs_dir2_data_log_header(tp, dp, dbp);
 	xfs_dir3_data_check(dp, dbp);
 	/*
 	 * If the longest data block freespace changes, need to update
@@ -1285,7 +1244,7 @@ xfs_dir2_leafn_remove(
 		 * Convert the data block number to a free block,
 		 * read in the free block.
 		 */
-		fdb = xfs_dir2_db_to_fdb(mp, db);
+		fdb = dp->d_ops->db_to_fdb(mp, db);
 		error = xfs_dir2_free_read(tp, dp, xfs_dir2_db_to_da(mp, fdb),
 					   &fbp);
 		if (error)
@@ -1294,22 +1253,22 @@ xfs_dir2_leafn_remove(
 #ifdef DEBUG
 	{
 		struct xfs_dir3_icfree_hdr freehdr;
-		xfs_dir3_free_hdr_from_disk(&freehdr, free);
-		ASSERT(freehdr.firstdb == xfs_dir3_free_max_bests(mp) *
+		dp->d_ops->free_hdr_from_disk(&freehdr, free);
+		ASSERT(freehdr.firstdb == dp->d_ops->free_max_bests(mp) *
 					  (fdb - XFS_DIR2_FREE_FIRSTDB(mp)));
 	}
 #endif
 		/*
 		 * Calculate which entry we need to fix.
 		 */
-		findex = xfs_dir2_db_to_fdindex(mp, db);
+		findex = dp->d_ops->db_to_fdindex(mp, db);
 		longest = be16_to_cpu(bf[0].length);
 		/*
 		 * If the data block is now empty we can get rid of it
 		 * (usually).
 		 */
 		if (longest == mp->m_dirblksize -
-			       xfs_dir3_data_entry_offset(hdr)) {
+			       dp->d_ops->data_entry_offset) {
 			/*
 			 * Try to punch out the data block.
 			 */
@@ -1336,12 +1295,12 @@ xfs_dir2_leafn_remove(
 			return error;
 	}
 
-	xfs_dir3_leaf_check(mp, bp);
+	xfs_dir3_leaf_check(dp, bp);
 	/*
 	 * Return indication of whether this leaf block is empty enough
 	 * to justify trying to join it with a neighbor.
 	 */
-	*rval = (xfs_dir3_leaf_hdr_size(leaf) +
+	*rval = (dp->d_ops->leaf_hdr_size +
 		 (uint)sizeof(ents[0]) * (leafhdr.count - leafhdr.stale)) <
 		mp->m_dir_magicpct;
 	return 0;
@@ -1360,13 +1319,14 @@ xfs_dir2_leafn_split(
 	xfs_dablk_t		blkno;		/* new leaf block number */
 	int			error;		/* error return value */
 	xfs_mount_t		*mp;		/* filesystem mount point */
+	struct xfs_inode	*dp;
 
 	/*
 	 * Allocate space for a new leaf node.
 	 */
 	args = state->args;
-	mp = args->dp->i_mount;
-	ASSERT(args != NULL);
+	dp = args->dp;
+	mp = dp->i_mount;
 	ASSERT(oldblk->magic == XFS_DIR2_LEAFN_MAGIC);
 	error = xfs_da_grow_inode(args, &blkno);
 	if (error) {
@@ -1401,10 +1361,10 @@ xfs_dir2_leafn_split(
 	/*
 	 * Update last hashval in each block since we added the name.
 	 */
-	oldblk->hashval = xfs_dir2_leafn_lasthash(oldblk->bp, NULL);
-	newblk->hashval = xfs_dir2_leafn_lasthash(newblk->bp, NULL);
-	xfs_dir3_leaf_check(mp, oldblk->bp);
-	xfs_dir3_leaf_check(mp, newblk->bp);
+	oldblk->hashval = xfs_dir2_leafn_lasthash(dp, oldblk->bp, NULL);
+	newblk->hashval = xfs_dir2_leafn_lasthash(dp, newblk->bp, NULL);
+	xfs_dir3_leaf_check(dp, oldblk->bp);
+	xfs_dir3_leaf_check(dp, newblk->bp);
 	return error;
 }
 
@@ -1434,6 +1394,7 @@ xfs_dir2_leafn_toosmall(
 	int			rval;		/* result from path_shift */
 	struct xfs_dir3_icleaf_hdr leafhdr;
 	struct xfs_dir2_leaf_entry *ents;
+	struct xfs_inode	*dp = state->args->dp;
 
 	/*
 	 * Check for the degenerate case of the block being over 50% full.
@@ -1442,12 +1403,12 @@ xfs_dir2_leafn_toosmall(
 	 */
 	blk = &state->path.blk[state->path.active - 1];
 	leaf = blk->bp->b_addr;
-	xfs_dir3_leaf_hdr_from_disk(&leafhdr, leaf);
-	ents = xfs_dir3_leaf_ents_p(leaf);
-	xfs_dir3_leaf_check(state->args->dp->i_mount, blk->bp);
+	dp->d_ops->leaf_hdr_from_disk(&leafhdr, leaf);
+	ents = dp->d_ops->leaf_ents_p(leaf);
+	xfs_dir3_leaf_check(dp, blk->bp);
 
 	count = leafhdr.count - leafhdr.stale;
-	bytes = xfs_dir3_leaf_hdr_size(leaf) + count * sizeof(ents[0]);
+	bytes = dp->d_ops->leaf_hdr_size + count * sizeof(ents[0]);
 	if (bytes > (state->blocksize >> 1)) {
 		/*
 		 * Blk over 50%, don't try to join.
@@ -1492,7 +1453,7 @@ xfs_dir2_leafn_toosmall(
 		/*
 		 * Read the sibling leaf block.
 		 */
-		error = xfs_dir3_leafn_read(state->args->trans, state->args->dp,
+		error = xfs_dir3_leafn_read(state->args->trans, dp,
 					    blkno, -1, &bp);
 		if (error)
 			return error;
@@ -1504,8 +1465,8 @@ xfs_dir2_leafn_toosmall(
 		bytes = state->blocksize - (state->blocksize >> 2);
 
 		leaf = bp->b_addr;
-		xfs_dir3_leaf_hdr_from_disk(&hdr2, leaf);
-		ents = xfs_dir3_leaf_ents_p(leaf);
+		dp->d_ops->leaf_hdr_from_disk(&hdr2, leaf);
+		ents = dp->d_ops->leaf_ents_p(leaf);
 		count += hdr2.count - hdr2.stale;
 		bytes -= count * sizeof(ents[0]);
 
@@ -1559,6 +1520,7 @@ xfs_dir2_leafn_unbalance(
 	struct xfs_dir3_icleaf_hdr drophdr;
 	struct xfs_dir2_leaf_entry *sents;
 	struct xfs_dir2_leaf_entry *dents;
+	struct xfs_inode	*dp = state->args->dp;
 
 	args = state->args;
 	ASSERT(drop_blk->magic == XFS_DIR2_LEAFN_MAGIC);
@@ -1566,10 +1528,10 @@ xfs_dir2_leafn_unbalance(
 	drop_leaf = drop_blk->bp->b_addr;
 	save_leaf = save_blk->bp->b_addr;
 
-	xfs_dir3_leaf_hdr_from_disk(&savehdr, save_leaf);
-	xfs_dir3_leaf_hdr_from_disk(&drophdr, drop_leaf);
-	sents = xfs_dir3_leaf_ents_p(save_leaf);
-	dents = xfs_dir3_leaf_ents_p(drop_leaf);
+	dp->d_ops->leaf_hdr_from_disk(&savehdr, save_leaf);
+	dp->d_ops->leaf_hdr_from_disk(&drophdr, drop_leaf);
+	sents = dp->d_ops->leaf_ents_p(save_leaf);
+	dents = dp->d_ops->leaf_ents_p(drop_leaf);
 
 	/*
 	 * If there are any stale leaf entries, take this opportunity
@@ -1584,7 +1546,7 @@ xfs_dir2_leafn_unbalance(
 	 * Move the entries from drop to the appropriate end of save.
 	 */
 	drop_blk->hashval = be32_to_cpu(dents[drophdr.count - 1].hashval);
-	if (xfs_dir2_leafn_order(save_blk->bp, drop_blk->bp))
+	if (xfs_dir2_leafn_order(dp, save_blk->bp, drop_blk->bp))
 		xfs_dir3_leafn_moveents(args, drop_blk->bp, &drophdr, dents, 0,
 					save_blk->bp, &savehdr, sents, 0,
 					drophdr.count);
@@ -1595,13 +1557,13 @@ xfs_dir2_leafn_unbalance(
 	save_blk->hashval = be32_to_cpu(sents[savehdr.count - 1].hashval);
 
 	/* log the changes made when moving the entries */
-	xfs_dir3_leaf_hdr_to_disk(save_leaf, &savehdr);
-	xfs_dir3_leaf_hdr_to_disk(drop_leaf, &drophdr);
-	xfs_dir3_leaf_log_header(args->trans, save_blk->bp);
-	xfs_dir3_leaf_log_header(args->trans, drop_blk->bp);
+	dp->d_ops->leaf_hdr_to_disk(save_leaf, &savehdr);
+	dp->d_ops->leaf_hdr_to_disk(drop_leaf, &drophdr);
+	xfs_dir3_leaf_log_header(args->trans, dp, save_blk->bp);
+	xfs_dir3_leaf_log_header(args->trans, dp, drop_blk->bp);
 
-	xfs_dir3_leaf_check(args->dp->i_mount, save_blk->bp);
-	xfs_dir3_leaf_check(args->dp->i_mount, drop_blk->bp);
+	xfs_dir3_leaf_check(dp, save_blk->bp);
+	xfs_dir3_leaf_check(dp, drop_blk->bp);
 }
 
 /*
@@ -1712,7 +1674,7 @@ xfs_dir2_node_addname_int(
 	dp = args->dp;
 	mp = dp->i_mount;
 	tp = args->trans;
-	length = xfs_dir3_data_entsize(mp, args->namelen);
+	length = dp->d_ops->data_entsize(args->namelen);
 	/*
 	 * If we came in with a freespace block that means that lookup
 	 * found an entry with our hash value.  This is the freespace
@@ -1726,8 +1688,8 @@ xfs_dir2_node_addname_int(
 		ifbno = fblk->blkno;
 		free = fbp->b_addr;
 		findex = fblk->index;
-		bests = xfs_dir3_free_bests_p(mp, free);
-		xfs_dir3_free_hdr_from_disk(&freehdr, free);
+		bests = dp->d_ops->free_bests_p(free);
+		dp->d_ops->free_hdr_from_disk(&freehdr, free);
 
 		/*
 		 * This means the free entry showed that the data block had
@@ -1819,8 +1781,8 @@ xfs_dir2_node_addname_int(
 		 * and the freehdr are actually initialised if they are placed
 		 * there, so we have to do it here to avoid warnings. Blech.
 		 */
-		bests = xfs_dir3_free_bests_p(mp, free);
-		xfs_dir3_free_hdr_from_disk(&freehdr, free);
+		bests = dp->d_ops->free_bests_p(free);
+		dp->d_ops->free_hdr_from_disk(&freehdr, free);
 		if (be16_to_cpu(bests[findex]) != NULLDATAOFF &&
 		    be16_to_cpu(bests[findex]) >= length)
 			dbno = freehdr.firstdb + findex;
@@ -1871,7 +1833,7 @@ xfs_dir2_node_addname_int(
 		 * Get the freespace block corresponding to the data block
 		 * that was just allocated.
 		 */
-		fbno = xfs_dir2_db_to_fdb(mp, dbno);
+		fbno = dp->d_ops->db_to_fdb(mp, dbno);
 		error = xfs_dir2_free_try_read(tp, dp,
 					       xfs_dir2_db_to_da(mp, fbno),
 					       &fbp);
@@ -1888,12 +1850,12 @@ xfs_dir2_node_addname_int(
 			if (error)
 				return error;
 
-			if (unlikely(xfs_dir2_db_to_fdb(mp, dbno) != fbno)) {
+			if (unlikely(dp->d_ops->db_to_fdb(mp, dbno) != fbno)) {
 				xfs_alert(mp,
 			"%s: dir ino %llu needed freesp block %lld for\n"
 			"  data block %lld, got %lld ifbno %llu lastfbno %d",
 					__func__, (unsigned long long)dp->i_ino,
-					(long long)xfs_dir2_db_to_fdb(mp, dbno),
+					(long long)dp->d_ops->db_to_fdb(mp, dbno),
 					(long long)dbno, (long long)fbno,
 					(unsigned long long)ifbno, lastfbno);
 				if (fblk) {
@@ -1918,30 +1880,30 @@ xfs_dir2_node_addname_int(
 			if (error)
 				return error;
 			free = fbp->b_addr;
-			bests = xfs_dir3_free_bests_p(mp, free);
-			xfs_dir3_free_hdr_from_disk(&freehdr, free);
+			bests = dp->d_ops->free_bests_p(free);
+			dp->d_ops->free_hdr_from_disk(&freehdr, free);
 
 			/*
 			 * Remember the first slot as our empty slot.
 			 */
 			freehdr.firstdb = (fbno - XFS_DIR2_FREE_FIRSTDB(mp)) *
-					xfs_dir3_free_max_bests(mp);
+					dp->d_ops->free_max_bests(mp);
 		} else {
 			free = fbp->b_addr;
-			bests = xfs_dir3_free_bests_p(mp, free);
-			xfs_dir3_free_hdr_from_disk(&freehdr, free);
+			bests = dp->d_ops->free_bests_p(free);
+			dp->d_ops->free_hdr_from_disk(&freehdr, free);
 		}
 
 		/*
 		 * Set the freespace block index from the data block number.
 		 */
-		findex = xfs_dir2_db_to_fdindex(mp, dbno);
+		findex = dp->d_ops->db_to_fdindex(mp, dbno);
 		/*
 		 * If it's after the end of the current entries in the
 		 * freespace block, extend that table.
 		 */
 		if (findex >= freehdr.nvalid) {
-			ASSERT(findex < xfs_dir3_free_max_bests(mp));
+			ASSERT(findex < dp->d_ops->free_max_bests(mp));
 			freehdr.nvalid = findex + 1;
 			/*
 			 * Tag new entry so nused will go up.
@@ -1954,8 +1916,8 @@ xfs_dir2_node_addname_int(
 		 */
 		if (bests[findex] == cpu_to_be16(NULLDATAOFF)) {
 			freehdr.nused++;
-			xfs_dir3_free_hdr_to_disk(fbp->b_addr, &freehdr);
-			xfs_dir2_free_log_header(tp, fbp);
+			dp->d_ops->free_hdr_to_disk(fbp->b_addr, &freehdr);
+			xfs_dir2_free_log_header(tp, dp, fbp);
 		}
 		/*
 		 * Update the real value in the table.
@@ -1963,7 +1925,7 @@ xfs_dir2_node_addname_int(
 		 * change again.
 		 */
 		hdr = dbp->b_addr;
-		bf = xfs_dir3_data_bestfree_p(hdr);
+		bf = dp->d_ops->data_bestfree_p(hdr);
 		bests[findex] = bf[0].length;
 		logfree = 1;
 	}
@@ -1985,7 +1947,7 @@ xfs_dir2_node_addname_int(
 		if (error)
 			return error;
 		hdr = dbp->b_addr;
-		bf = xfs_dir3_data_bestfree_p(hdr);
+		bf = dp->d_ops->data_bestfree_p(hdr);
 		logfree = 0;
 	}
 	ASSERT(be16_to_cpu(bf[0].length) >= length);
@@ -1998,7 +1960,7 @@ xfs_dir2_node_addname_int(
 	/*
 	 * Mark the first part of the unused space, inuse for us.
 	 */
-	xfs_dir2_data_use_free(tp, dbp, dup,
+	xfs_dir2_data_use_free(tp, dp, dbp, dup,
 		(xfs_dir2_data_aoff_t)((char *)dup - (char *)hdr), length,
 		&needlog, &needscan);
 	/*
@@ -2008,24 +1970,24 @@ xfs_dir2_node_addname_int(
 	dep->inumber = cpu_to_be64(args->inumber);
 	dep->namelen = args->namelen;
 	memcpy(dep->name, args->name, dep->namelen);
-	xfs_dir3_dirent_put_ftype(mp, dep, args->filetype);
-	tagp = xfs_dir3_data_entry_tag_p(mp, dep);
+	dp->d_ops->data_put_ftype(dep, args->filetype);
+	tagp = dp->d_ops->data_entry_tag_p(dep);
 	*tagp = cpu_to_be16((char *)dep - (char *)hdr);
-	xfs_dir2_data_log_entry(tp, dbp, dep);
+	xfs_dir2_data_log_entry(tp, dp, dbp, dep);
 	/*
 	 * Rescan the block for bestfree if needed.
 	 */
 	if (needscan)
-		xfs_dir2_data_freescan(mp, hdr, &needlog);
+		xfs_dir2_data_freescan(dp, hdr, &needlog);
 	/*
 	 * Log the data block header if needed.
 	 */
 	if (needlog)
-		xfs_dir2_data_log_header(tp, dbp);
+		xfs_dir2_data_log_header(tp, dp, dbp);
 	/*
 	 * If the freespace entry is now wrong, update it.
 	 */
-	bests = xfs_dir3_free_bests_p(mp, free); /* gcc is so stupid */
+	bests = dp->d_ops->free_bests_p(free); /* gcc is so stupid */
 	if (be16_to_cpu(bests[findex]) != be16_to_cpu(bf[0].length)) {
 		bests[findex] = bf[0].length;
 		logfree = 1;
@@ -2034,7 +1996,7 @@ xfs_dir2_node_addname_int(
 	 * Log the freespace entry if needed.
 	 */
 	if (logfree)
-		xfs_dir2_free_log_bests(tp, fbp, findex, findex);
+		xfs_dir2_free_log_bests(tp, dp, fbp, findex, findex);
 	/*
 	 * Return the data block and offset in args, then drop the data block.
 	 */
@@ -2212,7 +2174,7 @@ xfs_dir2_node_replace(
 		blk = &state->path.blk[state->path.active - 1];
 		ASSERT(blk->magic == XFS_DIR2_LEAFN_MAGIC);
 		leaf = blk->bp->b_addr;
-		ents = xfs_dir3_leaf_ents_p(leaf);
+		ents = args->dp->d_ops->leaf_ents_p(leaf);
 		lep = &ents[blk->index];
 		ASSERT(state->extravalid);
 		/*
@@ -2229,8 +2191,9 @@ xfs_dir2_node_replace(
 		 * Fill in the new inode number and log the entry.
 		 */
 		dep->inumber = cpu_to_be64(inum);
-		xfs_dir3_dirent_put_ftype(state->mp, dep, args->filetype);
-		xfs_dir2_data_log_entry(args->trans, state->extrablk.bp, dep);
+		args->dp->d_ops->data_put_ftype(dep, args->filetype);
+		xfs_dir2_data_log_entry(args->trans, args->dp,
+					state->extrablk.bp, dep);
 		rval = 0;
 	}
 	/*
@@ -2285,7 +2248,7 @@ xfs_dir2_node_trim_free(
 	if (!bp)
 		return 0;
 	free = bp->b_addr;
-	xfs_dir3_free_hdr_from_disk(&freehdr, free);
+	dp->d_ops->free_hdr_from_disk(&freehdr, free);
 
 	/*
 	 * If there are used entries, there's nothing to do.
diff --git a/fs/xfs/xfs_dir2_priv.h b/fs/xfs/xfs_dir2_priv.h
index 1bad84c40829..8b9d2281f85b 100644
--- a/fs/xfs/xfs_dir2_priv.h
+++ b/fs/xfs/xfs_dir2_priv.h
@@ -59,7 +59,8 @@ extern int xfs_dir3_data_readahead(struct xfs_trans *tp, struct xfs_inode *dp,
 
 extern struct xfs_dir2_data_free *
 xfs_dir2_data_freeinsert(struct xfs_dir2_data_hdr *hdr,
-		struct xfs_dir2_data_unused *dup, int *loghead);
+		struct xfs_dir2_data_free *bf, struct xfs_dir2_data_unused *dup,
+		int *loghead);
 extern int xfs_dir3_data_init(struct xfs_da_args *args, xfs_dir2_db_t blkno,
 		struct xfs_buf **bpp);
 
@@ -76,9 +77,9 @@ extern void xfs_dir3_leaf_compact_x1(struct xfs_dir3_icleaf_hdr *leafhdr,
 		int *lowstalep, int *highstalep, int *lowlogp, int *highlogp);
 extern int xfs_dir3_leaf_get_buf(struct xfs_da_args *args, xfs_dir2_db_t bno,
 		struct xfs_buf **bpp, __uint16_t magic);
-extern void xfs_dir3_leaf_log_ents(struct xfs_trans *tp, struct xfs_buf *bp,
-		int first, int last);
-extern void xfs_dir3_leaf_log_header(struct xfs_trans *tp,
+extern void xfs_dir3_leaf_log_ents(struct xfs_trans *tp, struct xfs_inode *dp,
+		struct xfs_buf *bp, int first, int last);
+extern void xfs_dir3_leaf_log_header(struct xfs_trans *tp, struct xfs_inode *dp,
 		struct xfs_buf *bp);
 extern int xfs_dir2_leaf_lookup(struct xfs_da_args *args);
 extern int xfs_dir2_leaf_removename(struct xfs_da_args *args);
@@ -93,21 +94,18 @@ xfs_dir3_leaf_find_entry(struct xfs_dir3_icleaf_hdr *leafhdr,
 		int lowstale, int highstale, int *lfloglow, int *lfloghigh);
 extern int xfs_dir2_node_to_leaf(struct xfs_da_state *state);
 
-extern void xfs_dir3_leaf_hdr_from_disk(struct xfs_dir3_icleaf_hdr *to,
-		struct xfs_dir2_leaf *from);
-extern void xfs_dir3_leaf_hdr_to_disk(struct xfs_dir2_leaf *to,
-		struct xfs_dir3_icleaf_hdr *from);
-extern bool xfs_dir3_leaf_check_int(struct xfs_mount *mp,
+extern bool xfs_dir3_leaf_check_int(struct xfs_mount *mp, struct xfs_inode *dp,
 		struct xfs_dir3_icleaf_hdr *hdr, struct xfs_dir2_leaf *leaf);
 
 /* xfs_dir2_node.c */
 extern int xfs_dir2_leaf_to_node(struct xfs_da_args *args,
 		struct xfs_buf *lbp);
-extern xfs_dahash_t xfs_dir2_leafn_lasthash(struct xfs_buf *bp, int *count);
+extern xfs_dahash_t xfs_dir2_leafn_lasthash(struct xfs_inode *dp,
+		struct xfs_buf *bp, int *count);
 extern int xfs_dir2_leafn_lookup_int(struct xfs_buf *bp,
 		struct xfs_da_args *args, int *indexp,
 		struct xfs_da_state *state);
-extern int xfs_dir2_leafn_order(struct xfs_buf *leaf1_bp,
+extern int xfs_dir2_leafn_order(struct xfs_inode *dp, struct xfs_buf *leaf1_bp,
 		struct xfs_buf *leaf2_bp);
 extern int xfs_dir2_leafn_split(struct xfs_da_state *state,
 	struct xfs_da_state_blk *oldblk, struct xfs_da_state_blk *newblk);
diff --git a/fs/xfs/xfs_dir2_readdir.c b/fs/xfs/xfs_dir2_readdir.c
index 8f84153e98a8..c4e50c6ed584 100644
--- a/fs/xfs/xfs_dir2_readdir.c
+++ b/fs/xfs/xfs_dir2_readdir.c
@@ -18,23 +18,23 @@
  */
 #include "xfs.h"
 #include "xfs_fs.h"
-#include "xfs_types.h"
+#include "xfs_format.h"
+#include "xfs_log_format.h"
+#include "xfs_trans_resv.h"
 #include "xfs_bit.h"
-#include "xfs_log.h"
-#include "xfs_trans.h"
 #include "xfs_sb.h"
 #include "xfs_ag.h"
 #include "xfs_mount.h"
+#include "xfs_da_format.h"
 #include "xfs_da_btree.h"
-#include "xfs_bmap_btree.h"
-#include "xfs_dinode.h"
 #include "xfs_inode.h"
-#include "xfs_dir2_format.h"
 #include "xfs_dir2.h"
 #include "xfs_dir2_priv.h"
 #include "xfs_error.h"
 #include "xfs_trace.h"
 #include "xfs_bmap.h"
+#include "xfs_trans.h"
+#include "xfs_dinode.h"
 
 /*
  * Directory file type support functions
@@ -119,9 +119,9 @@ xfs_dir2_sf_getdents(
 	 * mp->m_dirdatablk.
 	 */
 	dot_offset = xfs_dir2_db_off_to_dataptr(mp, mp->m_dirdatablk,
-					     xfs_dir3_data_dot_offset(mp));
+						dp->d_ops->data_dot_offset);
 	dotdot_offset = xfs_dir2_db_off_to_dataptr(mp, mp->m_dirdatablk,
-						xfs_dir3_data_dotdot_offset(mp));
+						dp->d_ops->data_dotdot_offset);
 
 	/*
 	 * Put . entry unless we're starting past it.
@@ -136,7 +136,7 @@ xfs_dir2_sf_getdents(
 	 * Put .. entry unless we're starting past it.
 	 */
 	if (ctx->pos <= dotdot_offset) {
-		ino = xfs_dir2_sf_get_parent_ino(sfp);
+		ino = dp->d_ops->sf_get_parent_ino(sfp);
 		ctx->pos = dotdot_offset & 0x7fffffff;
 		if (!dir_emit(ctx, "..", 2, ino, DT_DIR))
 			return 0;
@@ -153,17 +153,17 @@ xfs_dir2_sf_getdents(
 				xfs_dir2_sf_get_offset(sfep));
 
 		if (ctx->pos > off) {
-			sfep = xfs_dir3_sf_nextentry(mp, sfp, sfep);
+			sfep = dp->d_ops->sf_nextentry(sfp, sfep);
 			continue;
 		}
 
-		ino = xfs_dir3_sfe_get_ino(mp, sfp, sfep);
-		filetype = xfs_dir3_sfe_get_ftype(mp, sfp, sfep);
+		ino = dp->d_ops->sf_get_ino(sfp, sfep);
+		filetype = dp->d_ops->sf_get_ftype(sfep);
 		ctx->pos = off & 0x7fffffff;
 		if (!dir_emit(ctx, (char *)sfep->name, sfep->namelen, ino,
 			    xfs_dir3_get_dtype(mp, filetype)))
 			return 0;
-		sfep = xfs_dir3_sf_nextentry(mp, sfp, sfep);
+		sfep = dp->d_ops->sf_nextentry(sfp, sfep);
 	}
 
 	ctx->pos = xfs_dir2_db_off_to_dataptr(mp, mp->m_dirdatablk + 1, 0) &
@@ -213,7 +213,7 @@ xfs_dir2_block_getdents(
 	 * Set up values for the loop.
 	 */
 	btp = xfs_dir2_block_tail_p(mp, hdr);
-	ptr = (char *)xfs_dir3_data_entry_p(hdr);
+	ptr = (char *)dp->d_ops->data_entry_p(hdr);
 	endptr = (char *)xfs_dir2_block_leaf_p(btp);
 
 	/*
@@ -237,7 +237,7 @@ xfs_dir2_block_getdents(
 		/*
 		 * Bump pointer for the next iteration.
 		 */
-		ptr += xfs_dir3_data_entsize(mp, dep->namelen);
+		ptr += dp->d_ops->data_entsize(dep->namelen);
 		/*
 		 * The entry is before the desired starting point, skip it.
 		 */
@@ -248,7 +248,7 @@ xfs_dir2_block_getdents(
 					    (char *)dep - (char *)hdr);
 
 		ctx->pos = cook & 0x7fffffff;
-		filetype = xfs_dir3_dirent_get_ftype(mp, dep);
+		filetype = dp->d_ops->data_get_ftype(dep);
 		/*
 		 * If it didn't fit, set the final offset to here & return.
 		 */
@@ -578,13 +578,13 @@ xfs_dir2_leaf_getdents(
 			/*
 			 * Find our position in the block.
 			 */
-			ptr = (char *)xfs_dir3_data_entry_p(hdr);
+			ptr = (char *)dp->d_ops->data_entry_p(hdr);
 			byteoff = xfs_dir2_byte_to_off(mp, curoff);
 			/*
 			 * Skip past the header.
 			 */
 			if (byteoff == 0)
-				curoff += xfs_dir3_data_entry_offset(hdr);
+				curoff += dp->d_ops->data_entry_offset;
 			/*
 			 * Skip past entries until we reach our offset.
 			 */
@@ -601,7 +601,7 @@ xfs_dir2_leaf_getdents(
 					}
 					dep = (xfs_dir2_data_entry_t *)ptr;
 					length =
-					   xfs_dir3_data_entsize(mp, dep->namelen);
+					   dp->d_ops->data_entsize(dep->namelen);
 					ptr += length;
 				}
 				/*
@@ -632,8 +632,8 @@ xfs_dir2_leaf_getdents(
 		}
 
 		dep = (xfs_dir2_data_entry_t *)ptr;
-		length = xfs_dir3_data_entsize(mp, dep->namelen);
-		filetype = xfs_dir3_dirent_get_ftype(mp, dep);
+		length = dp->d_ops->data_entsize(dep->namelen);
+		filetype = dp->d_ops->data_get_ftype(dep);
 
 		ctx->pos = xfs_dir2_byte_to_dataptr(mp, curoff) & 0x7fffffff;
 		if (!dir_emit(ctx, (char *)dep->name, dep->namelen,
diff --git a/fs/xfs/xfs_dir2_sf.c b/fs/xfs/xfs_dir2_sf.c
index 3ef6d402084c..aafc6e46cb58 100644
--- a/fs/xfs/xfs_dir2_sf.c
+++ b/fs/xfs/xfs_dir2_sf.c
@@ -17,22 +17,22 @@
  */
 #include "xfs.h"
 #include "xfs_fs.h"
-#include "xfs_types.h"
-#include "xfs_log.h"
-#include "xfs_trans.h"
+#include "xfs_format.h"
+#include "xfs_log_format.h"
+#include "xfs_trans_resv.h"
 #include "xfs_sb.h"
 #include "xfs_ag.h"
 #include "xfs_mount.h"
+#include "xfs_da_format.h"
 #include "xfs_da_btree.h"
-#include "xfs_bmap_btree.h"
-#include "xfs_dinode.h"
 #include "xfs_inode.h"
+#include "xfs_trans.h"
 #include "xfs_inode_item.h"
 #include "xfs_error.h"
-#include "xfs_dir2_format.h"
 #include "xfs_dir2.h"
 #include "xfs_dir2_priv.h"
 #include "xfs_trace.h"
+#include "xfs_dinode.h"
 
 /*
  * Prototypes for internal functions.
@@ -57,89 +57,6 @@ static void xfs_dir2_sf_toino8(xfs_da_args_t *args);
 #endif /* XFS_BIG_INUMS */
 
 /*
- * Inode numbers in short-form directories can come in two versions,
- * either 4 bytes or 8 bytes wide.  These helpers deal with the
- * two forms transparently by looking at the headers i8count field.
- *
- * For 64-bit inode number the most significant byte must be zero.
- */
-static xfs_ino_t
-xfs_dir2_sf_get_ino(
-	struct xfs_dir2_sf_hdr	*hdr,
-	xfs_dir2_inou_t		*from)
-{
-	if (hdr->i8count)
-		return get_unaligned_be64(&from->i8.i) & 0x00ffffffffffffffULL;
-	else
-		return get_unaligned_be32(&from->i4.i);
-}
-
-static void
-xfs_dir2_sf_put_ino(
-	struct xfs_dir2_sf_hdr	*hdr,
-	xfs_dir2_inou_t		*to,
-	xfs_ino_t		ino)
-{
-	ASSERT((ino & 0xff00000000000000ULL) == 0);
-
-	if (hdr->i8count)
-		put_unaligned_be64(ino, &to->i8.i);
-	else
-		put_unaligned_be32(ino, &to->i4.i);
-}
-
-xfs_ino_t
-xfs_dir2_sf_get_parent_ino(
-	struct xfs_dir2_sf_hdr	*hdr)
-{
-	return xfs_dir2_sf_get_ino(hdr, &hdr->parent);
-}
-
-void
-xfs_dir2_sf_put_parent_ino(
-	struct xfs_dir2_sf_hdr	*hdr,
-	xfs_ino_t		ino)
-{
-	xfs_dir2_sf_put_ino(hdr, &hdr->parent, ino);
-}
-
-/*
- * In short-form directory entries the inode numbers are stored at variable
- * offset behind the entry name. If the entry stores a filetype value, then it
- * sits between the name and the inode number. Hence the inode numbers may only
- * be accessed through the helpers below.
- */
-static xfs_dir2_inou_t *
-xfs_dir3_sfe_inop(
-	struct xfs_mount	*mp,
-	struct xfs_dir2_sf_entry *sfep)
-{
-	__uint8_t	*ptr = &sfep->name[sfep->namelen];
-	if (xfs_sb_version_hasftype(&mp->m_sb))
-		ptr++;
-	return (xfs_dir2_inou_t *)ptr;
-}
-
-xfs_ino_t
-xfs_dir3_sfe_get_ino(
-	struct xfs_mount	*mp,
-	struct xfs_dir2_sf_hdr	*hdr,
-	struct xfs_dir2_sf_entry *sfep)
-{
-	return xfs_dir2_sf_get_ino(hdr, xfs_dir3_sfe_inop(mp, sfep));
-}
-
-void
-xfs_dir3_sfe_put_ino(
-	struct xfs_mount	*mp,
-	struct xfs_dir2_sf_hdr	*hdr,
-	struct xfs_dir2_sf_entry *sfep,
-	xfs_ino_t		ino)
-{
-	xfs_dir2_sf_put_ino(hdr, xfs_dir3_sfe_inop(mp, sfep), ino);
-}
-
-/*
  * Given a block directory (dp/block), calculate its size as a shortform (sf)
  * directory and a header for the sf directory, if it will fit it the
  * space currently present in the inode.  If it won't fit, the output
@@ -226,7 +143,7 @@ xfs_dir2_block_sfsize(
 	 */
 	sfhp->count = count;
 	sfhp->i8count = i8count;
-	xfs_dir2_sf_put_parent_ino(sfhp, parent);
+	dp->d_ops->sf_put_parent_ino(sfhp, parent);
 	return size;
 }
 
@@ -293,7 +210,7 @@ xfs_dir2_block_to_sf(
 	 * Set up to loop over the block's entries.
 	 */
 	btp = xfs_dir2_block_tail_p(mp, hdr);
-	ptr = (char *)xfs_dir3_data_entry_p(hdr);
+	ptr = (char *)dp->d_ops->data_entry_p(hdr);
 	endptr = (char *)xfs_dir2_block_leaf_p(btp);
 	sfep = xfs_dir2_sf_firstentry(sfp);
 	/*
@@ -321,7 +238,7 @@ xfs_dir2_block_to_sf(
 		else if (dep->namelen == 2 &&
 			 dep->name[0] == '.' && dep->name[1] == '.')
 			ASSERT(be64_to_cpu(dep->inumber) ==
-			       xfs_dir2_sf_get_parent_ino(sfp));
+			       dp->d_ops->sf_get_parent_ino(sfp));
 		/*
 		 * Normal entry, copy it into shortform.
 		 */
@@ -331,14 +248,14 @@ xfs_dir2_block_to_sf(
 				(xfs_dir2_data_aoff_t)
 				((char *)dep - (char *)hdr));
 			memcpy(sfep->name, dep->name, dep->namelen);
-			xfs_dir3_sfe_put_ino(mp, sfp, sfep,
-					     be64_to_cpu(dep->inumber));
-			xfs_dir3_sfe_put_ftype(mp, sfp, sfep,
-					xfs_dir3_dirent_get_ftype(mp, dep));
+			dp->d_ops->sf_put_ino(sfp, sfep,
+					      be64_to_cpu(dep->inumber));
+			dp->d_ops->sf_put_ftype(sfep,
+					dp->d_ops->data_get_ftype(dep));
 
-			sfep = xfs_dir3_sf_nextentry(mp, sfp, sfep);
+			sfep = dp->d_ops->sf_nextentry(sfp, sfep);
 		}
-		ptr += xfs_dir3_data_entsize(mp, dep->namelen);
+		ptr += dp->d_ops->data_entsize(dep->namelen);
 	}
 	ASSERT((char *)sfep - (char *)sfp == size);
 	xfs_dir2_sf_check(args);
@@ -389,7 +306,7 @@ xfs_dir2_sf_addname(
 	/*
 	 * Compute entry (and change in) size.
 	 */
-	add_entsize = xfs_dir3_sf_entsize(dp->i_mount, sfp, args->namelen);
+	add_entsize = dp->d_ops->sf_entsize(sfp, args->namelen);
 	incr_isize = add_entsize;
 	objchange = 0;
 #if XFS_BIG_INUMS
@@ -483,8 +400,7 @@ xfs_dir2_sf_addname_easy(
 	/*
 	 * Grow the in-inode space.
 	 */
-	xfs_idata_realloc(dp,
-			  xfs_dir3_sf_entsize(dp->i_mount, sfp, args->namelen),
+	xfs_idata_realloc(dp, dp->d_ops->sf_entsize(sfp, args->namelen),
 			  XFS_DATA_FORK);
 	/*
 	 * Need to set up again due to realloc of the inode data.
@@ -497,8 +413,8 @@ xfs_dir2_sf_addname_easy(
 	sfep->namelen = args->namelen;
 	xfs_dir2_sf_put_offset(sfep, offset);
 	memcpy(sfep->name, args->name, sfep->namelen);
-	xfs_dir3_sfe_put_ino(dp->i_mount, sfp, sfep, args->inumber);
-	xfs_dir3_sfe_put_ftype(dp->i_mount, sfp, sfep, args->filetype);
+	dp->d_ops->sf_put_ino(sfp, sfep, args->inumber);
+	dp->d_ops->sf_put_ftype(sfep, args->filetype);
 
 	/*
 	 * Update the header and inode.
@@ -557,13 +473,13 @@ xfs_dir2_sf_addname_hard(
 	 * to insert the new entry.
 	 * If it's going to end up at the end then oldsfep will point there.
 	 */
-	for (offset = xfs_dir3_data_first_offset(mp),
+	for (offset = dp->d_ops->data_first_offset,
 	      oldsfep = xfs_dir2_sf_firstentry(oldsfp),
-	      add_datasize = xfs_dir3_data_entsize(mp, args->namelen),
+	      add_datasize = dp->d_ops->data_entsize(args->namelen),
 	      eof = (char *)oldsfep == &buf[old_isize];
 	     !eof;
-	     offset = new_offset + xfs_dir3_data_entsize(mp, oldsfep->namelen),
-	      oldsfep = xfs_dir3_sf_nextentry(mp, oldsfp, oldsfep),
+	     offset = new_offset + dp->d_ops->data_entsize(oldsfep->namelen),
+	      oldsfep = dp->d_ops->sf_nextentry(oldsfp, oldsfep),
 	      eof = (char *)oldsfep == &buf[old_isize]) {
 		new_offset = xfs_dir2_sf_get_offset(oldsfep);
 		if (offset + add_datasize <= new_offset)
@@ -592,8 +508,8 @@ xfs_dir2_sf_addname_hard(
 	sfep->namelen = args->namelen;
 	xfs_dir2_sf_put_offset(sfep, offset);
 	memcpy(sfep->name, args->name, sfep->namelen);
-	xfs_dir3_sfe_put_ino(mp, sfp, sfep, args->inumber);
-	xfs_dir3_sfe_put_ftype(mp, sfp, sfep, args->filetype);
+	dp->d_ops->sf_put_ino(sfp, sfep, args->inumber);
+	dp->d_ops->sf_put_ftype(sfep, args->filetype);
 	sfp->count++;
 #if XFS_BIG_INUMS
 	if (args->inumber > XFS_DIR2_MAX_SHORT_INUM && !objchange)
@@ -603,7 +519,7 @@ xfs_dir2_sf_addname_hard(
 	 * If there's more left to copy, do that.
 	 */
 	if (!eof) {
-		sfep = xfs_dir3_sf_nextentry(mp, sfp, sfep);
+		sfep = dp->d_ops->sf_nextentry(sfp, sfep);
 		memcpy(sfep, oldsfep, old_isize - nbytes);
 	}
 	kmem_free(buf);
@@ -639,8 +555,8 @@ xfs_dir2_sf_addname_pick(
 	mp = dp->i_mount;
 
 	sfp = (xfs_dir2_sf_hdr_t *)dp->i_df.if_u1.if_data;
-	size = xfs_dir3_data_entsize(mp, args->namelen);
-	offset = xfs_dir3_data_first_offset(mp);
+	size = dp->d_ops->data_entsize(args->namelen);
+	offset = dp->d_ops->data_first_offset;
 	sfep = xfs_dir2_sf_firstentry(sfp);
 	holefit = 0;
 	/*
@@ -652,8 +568,8 @@ xfs_dir2_sf_addname_pick(
 		if (!holefit)
 			holefit = offset + size <= xfs_dir2_sf_get_offset(sfep);
 		offset = xfs_dir2_sf_get_offset(sfep) +
-			 xfs_dir3_data_entsize(mp, sfep->namelen);
-		sfep = xfs_dir3_sf_nextentry(mp, sfp, sfep);
+			 dp->d_ops->data_entsize(sfep->namelen);
+		sfep = dp->d_ops->sf_nextentry(sfp, sfep);
 	}
 	/*
 	 * Calculate data bytes used excluding the new entry, if this
@@ -713,21 +629,20 @@ xfs_dir2_sf_check(
 	mp = dp->i_mount;
 
 	sfp = (xfs_dir2_sf_hdr_t *)dp->i_df.if_u1.if_data;
-	offset = xfs_dir3_data_first_offset(mp);
-	ino = xfs_dir2_sf_get_parent_ino(sfp);
+	offset = dp->d_ops->data_first_offset;
+	ino = dp->d_ops->sf_get_parent_ino(sfp);
 	i8count = ino > XFS_DIR2_MAX_SHORT_INUM;
 
 	for (i = 0, sfep = xfs_dir2_sf_firstentry(sfp);
 	     i < sfp->count;
-	     i++, sfep = xfs_dir3_sf_nextentry(mp, sfp, sfep)) {
+	     i++, sfep = dp->d_ops->sf_nextentry(sfp, sfep)) {
 		ASSERT(xfs_dir2_sf_get_offset(sfep) >= offset);
-		ino = xfs_dir3_sfe_get_ino(mp, sfp, sfep);
+		ino = dp->d_ops->sf_get_ino(sfp, sfep);
 		i8count += ino > XFS_DIR2_MAX_SHORT_INUM;
 		offset =
 			xfs_dir2_sf_get_offset(sfep) +
-			xfs_dir3_data_entsize(mp, sfep->namelen);
-		ASSERT(xfs_dir3_sfe_get_ftype(mp, sfp, sfep) <
-							XFS_DIR3_FT_MAX);
+			dp->d_ops->data_entsize(sfep->namelen);
+		ASSERT(dp->d_ops->sf_get_ftype(sfep) < XFS_DIR3_FT_MAX);
 	}
 	ASSERT(i8count == sfp->i8count);
 	ASSERT(XFS_BIG_INUMS || i8count == 0);
@@ -783,7 +698,7 @@ xfs_dir2_sf_create(
 	/*
 	 * Now can put in the inode number, since i8count is set.
 	 */
-	xfs_dir2_sf_put_parent_ino(sfp, pino);
+	dp->d_ops->sf_put_parent_ino(sfp, pino);
 	sfp->count = 0;
 	dp->i_d.di_size = size;
 	xfs_dir2_sf_check(args);
@@ -838,7 +753,7 @@ xfs_dir2_sf_lookup(
 	 */
 	if (args->namelen == 2 &&
 	    args->name[0] == '.' && args->name[1] == '.') {
-		args->inumber = xfs_dir2_sf_get_parent_ino(sfp);
+		args->inumber = dp->d_ops->sf_get_parent_ino(sfp);
 		args->cmpresult = XFS_CMP_EXACT;
 		args->filetype = XFS_DIR3_FT_DIR;
 		return XFS_ERROR(EEXIST);
@@ -848,7 +763,7 @@ xfs_dir2_sf_lookup(
 	 */
 	ci_sfep = NULL;
 	for (i = 0, sfep = xfs_dir2_sf_firstentry(sfp); i < sfp->count;
-	     i++, sfep = xfs_dir3_sf_nextentry(dp->i_mount, sfp, sfep)) {
+	     i++, sfep = dp->d_ops->sf_nextentry(sfp, sfep)) {
 		/*
 		 * Compare name and if it's an exact match, return the inode
 		 * number. If it's the first case-insensitive match, store the
@@ -858,10 +773,8 @@ xfs_dir2_sf_lookup(
 								sfep->namelen);
 		if (cmp != XFS_CMP_DIFFERENT && cmp != args->cmpresult) {
 			args->cmpresult = cmp;
-			args->inumber = xfs_dir3_sfe_get_ino(dp->i_mount,
-							     sfp, sfep);
-			args->filetype = xfs_dir3_sfe_get_ftype(dp->i_mount,
-								sfp, sfep);
+			args->inumber = dp->d_ops->sf_get_ino(sfp, sfep);
+			args->filetype = dp->d_ops->sf_get_ftype(sfep);
 			if (cmp == XFS_CMP_EXACT)
 				return XFS_ERROR(EEXIST);
 			ci_sfep = sfep;
@@ -917,10 +830,10 @@ xfs_dir2_sf_removename(
 	 * Find the one we're deleting.
 	 */
 	for (i = 0, sfep = xfs_dir2_sf_firstentry(sfp); i < sfp->count;
-	     i++, sfep = xfs_dir3_sf_nextentry(dp->i_mount, sfp, sfep)) {
+	     i++, sfep = dp->d_ops->sf_nextentry(sfp, sfep)) {
 		if (xfs_da_compname(args, sfep->name, sfep->namelen) ==
 								XFS_CMP_EXACT) {
-			ASSERT(xfs_dir3_sfe_get_ino(dp->i_mount, sfp, sfep) ==
+			ASSERT(dp->d_ops->sf_get_ino(sfp, sfep) ==
 			       args->inumber);
 			break;
 		}
@@ -934,7 +847,7 @@ xfs_dir2_sf_removename(
 	 * Calculate sizes.
 	 */
 	byteoff = (int)((char *)sfep - (char *)sfp);
-	entsize = xfs_dir3_sf_entsize(dp->i_mount, sfp, args->namelen);
+	entsize = dp->d_ops->sf_entsize(sfp, args->namelen);
 	newsize = oldsize - entsize;
 	/*
 	 * Copy the part if any after the removed entry, sliding it down.
@@ -1041,28 +954,25 @@ xfs_dir2_sf_replace(
 	if (args->namelen == 2 &&
 	    args->name[0] == '.' && args->name[1] == '.') {
 #if XFS_BIG_INUMS || defined(DEBUG)
-		ino = xfs_dir2_sf_get_parent_ino(sfp);
+		ino = dp->d_ops->sf_get_parent_ino(sfp);
 		ASSERT(args->inumber != ino);
 #endif
-		xfs_dir2_sf_put_parent_ino(sfp, args->inumber);
+		dp->d_ops->sf_put_parent_ino(sfp, args->inumber);
 	}
 	/*
 	 * Normal entry, look for the name.
 	 */
 	else {
 		for (i = 0, sfep = xfs_dir2_sf_firstentry(sfp); i < sfp->count;
-		     i++, sfep = xfs_dir3_sf_nextentry(dp->i_mount, sfp, sfep)) {
+		     i++, sfep = dp->d_ops->sf_nextentry(sfp, sfep)) {
 			if (xfs_da_compname(args, sfep->name, sfep->namelen) ==
 								XFS_CMP_EXACT) {
 #if XFS_BIG_INUMS || defined(DEBUG)
-				ino = xfs_dir3_sfe_get_ino(dp->i_mount,
-							   sfp, sfep);
+				ino = dp->d_ops->sf_get_ino(sfp, sfep);
 				ASSERT(args->inumber != ino);
 #endif
-				xfs_dir3_sfe_put_ino(dp->i_mount, sfp, sfep,
-						     args->inumber);
-				xfs_dir3_sfe_put_ftype(dp->i_mount, sfp, sfep,
-						       args->filetype);
+				dp->d_ops->sf_put_ino(sfp, sfep, args->inumber);
+				dp->d_ops->sf_put_ftype(sfep, args->filetype);
 				break;
 			}
 		}
@@ -1165,22 +1075,21 @@ xfs_dir2_sf_toino4(
 	 */
 	sfp->count = oldsfp->count;
 	sfp->i8count = 0;
-	xfs_dir2_sf_put_parent_ino(sfp, xfs_dir2_sf_get_parent_ino(oldsfp));
+	dp->d_ops->sf_put_parent_ino(sfp, dp->d_ops->sf_get_parent_ino(oldsfp));
 	/*
 	 * Copy the entries field by field.
 	 */
 	for (i = 0, sfep = xfs_dir2_sf_firstentry(sfp),
 		    oldsfep = xfs_dir2_sf_firstentry(oldsfp);
 	     i < sfp->count;
-	     i++, sfep = xfs_dir3_sf_nextentry(mp, sfp, sfep),
-		  oldsfep = xfs_dir3_sf_nextentry(mp, oldsfp, oldsfep)) {
+	     i++, sfep = dp->d_ops->sf_nextentry(sfp, sfep),
+		  oldsfep = dp->d_ops->sf_nextentry(oldsfp, oldsfep)) {
 		sfep->namelen = oldsfep->namelen;
 		sfep->offset = oldsfep->offset;
 		memcpy(sfep->name, oldsfep->name, sfep->namelen);
-		xfs_dir3_sfe_put_ino(mp, sfp, sfep,
-			xfs_dir3_sfe_get_ino(mp, oldsfp, oldsfep));
-		xfs_dir3_sfe_put_ftype(mp, sfp, sfep,
-			xfs_dir3_sfe_get_ftype(mp, oldsfp, oldsfep));
+		dp->d_ops->sf_put_ino(sfp, sfep,
+				      dp->d_ops->sf_get_ino(oldsfp, oldsfep));
+		dp->d_ops->sf_put_ftype(sfep, dp->d_ops->sf_get_ftype(oldsfep));
 	}
 	/*
 	 * Clean up the inode.
@@ -1244,22 +1153,21 @@ xfs_dir2_sf_toino8(
 	 */
 	sfp->count = oldsfp->count;
 	sfp->i8count = 1;
-	xfs_dir2_sf_put_parent_ino(sfp, xfs_dir2_sf_get_parent_ino(oldsfp));
+	dp->d_ops->sf_put_parent_ino(sfp, dp->d_ops->sf_get_parent_ino(oldsfp));
 	/*
 	 * Copy the entries field by field.
 	 */
 	for (i = 0, sfep = xfs_dir2_sf_firstentry(sfp),
 		    oldsfep = xfs_dir2_sf_firstentry(oldsfp);
 	     i < sfp->count;
-	     i++, sfep = xfs_dir3_sf_nextentry(mp, sfp, sfep),
-		  oldsfep = xfs_dir3_sf_nextentry(mp, oldsfp, oldsfep)) {
+	     i++, sfep = dp->d_ops->sf_nextentry(sfp, sfep),
+		  oldsfep = dp->d_ops->sf_nextentry(oldsfp, oldsfep)) {
 		sfep->namelen = oldsfep->namelen;
 		sfep->offset = oldsfep->offset;
 		memcpy(sfep->name, oldsfep->name, sfep->namelen);
-		xfs_dir3_sfe_put_ino(mp, sfp, sfep,
-			xfs_dir3_sfe_get_ino(mp, oldsfp, oldsfep));
-		xfs_dir3_sfe_put_ftype(mp, sfp, sfep,
-			xfs_dir3_sfe_get_ftype(mp, oldsfp, oldsfep));
+		dp->d_ops->sf_put_ino(sfp, sfep,
+				      dp->d_ops->sf_get_ino(oldsfp, oldsfep));
+		dp->d_ops->sf_put_ftype(sfep, dp->d_ops->sf_get_ftype(oldsfep));
 	}
 	/*
 	 * Clean up the inode.
diff --git a/fs/xfs/xfs_discard.c b/fs/xfs/xfs_discard.c
index 45560ee1a4ba..8367d6dc18c9 100644
--- a/fs/xfs/xfs_discard.c
+++ b/fs/xfs/xfs_discard.c
@@ -17,22 +17,21 @@
  */
 #include "xfs.h"
 #include "xfs_format.h"
-#include "xfs_log.h"
-#include "xfs_trans.h"
+#include "xfs_log_format.h"
+#include "xfs_trans_resv.h"
 #include "xfs_sb.h"
 #include "xfs_ag.h"
 #include "xfs_mount.h"
 #include "xfs_quota.h"
-#include "xfs_alloc_btree.h"
-#include "xfs_bmap_btree.h"
-#include "xfs_ialloc_btree.h"
-#include "xfs_btree.h"
 #include "xfs_inode.h"
+#include "xfs_btree.h"
+#include "xfs_alloc_btree.h"
 #include "xfs_alloc.h"
 #include "xfs_error.h"
 #include "xfs_extent_busy.h"
 #include "xfs_discard.h"
 #include "xfs_trace.h"
+#include "xfs_log.h"
 
 STATIC int
 xfs_trim_extents(
diff --git a/fs/xfs/xfs_dquot.c b/fs/xfs/xfs_dquot.c
index 1ee776d477c3..6b1e695caf0e 100644
--- a/fs/xfs/xfs_dquot.c
+++ b/fs/xfs/xfs_dquot.c
@@ -18,28 +18,28 @@
 #include "xfs.h"
 #include "xfs_fs.h"
 #include "xfs_format.h"
+#include "xfs_log_format.h"
+#include "xfs_shared.h"
+#include "xfs_trans_resv.h"
 #include "xfs_bit.h"
-#include "xfs_log.h"
-#include "xfs_trans.h"
 #include "xfs_sb.h"
 #include "xfs_ag.h"
-#include "xfs_alloc.h"
-#include "xfs_quota.h"
 #include "xfs_mount.h"
-#include "xfs_bmap_btree.h"
 #include "xfs_inode.h"
 #include "xfs_bmap.h"
 #include "xfs_bmap_util.h"
-#include "xfs_rtalloc.h"
+#include "xfs_alloc.h"
+#include "xfs_quota.h"
 #include "xfs_error.h"
-#include "xfs_itable.h"
-#include "xfs_attr.h"
+#include "xfs_trans.h"
 #include "xfs_buf_item.h"
 #include "xfs_trans_space.h"
 #include "xfs_trans_priv.h"
 #include "xfs_qm.h"
 #include "xfs_cksum.h"
 #include "xfs_trace.h"
+#include "xfs_log.h"
+#include "xfs_bmap_btree.h"
 
 /*
  * Lock order:
@@ -292,118 +292,6 @@ xfs_dquot_set_prealloc_limits(struct xfs_dquot *dqp)
 	dqp->q_low_space[XFS_QLOWSP_5_PCNT] = space * 5;
 }
 
-STATIC bool
-xfs_dquot_buf_verify_crc(
-	struct xfs_mount	*mp,
-	struct xfs_buf		*bp)
-{
-	struct xfs_dqblk	*d = (struct xfs_dqblk *)bp->b_addr;
-	int			ndquots;
-	int			i;
-
-	if (!xfs_sb_version_hascrc(&mp->m_sb))
-		return true;
-
-	/*
-	 * if we are in log recovery, the quota subsystem has not been
-	 * initialised so we have no quotainfo structure. In that case, we need
-	 * to manually calculate the number of dquots in the buffer.
-	 */
-	if (mp->m_quotainfo)
-		ndquots = mp->m_quotainfo->qi_dqperchunk;
-	else
-		ndquots = xfs_qm_calc_dquots_per_chunk(mp,
-					XFS_BB_TO_FSB(mp, bp->b_length));
-
-	for (i = 0; i < ndquots; i++, d++) {
-		if (!xfs_verify_cksum((char *)d, sizeof(struct xfs_dqblk),
-				 XFS_DQUOT_CRC_OFF))
-			return false;
-		if (!uuid_equal(&d->dd_uuid, &mp->m_sb.sb_uuid))
-			return false;
-	}
-	return true;
-}
-
-STATIC bool
-xfs_dquot_buf_verify(
-	struct xfs_mount	*mp,
-	struct xfs_buf		*bp)
-{
-	struct xfs_dqblk	*d = (struct xfs_dqblk *)bp->b_addr;
-	xfs_dqid_t		id = 0;
-	int			ndquots;
-	int			i;
-
-	/*
-	 * if we are in log recovery, the quota subsystem has not been
-	 * initialised so we have no quotainfo structure. In that case, we need
-	 * to manually calculate the number of dquots in the buffer.
-	 */
-	if (mp->m_quotainfo)
-		ndquots = mp->m_quotainfo->qi_dqperchunk;
-	else
-		ndquots = xfs_qm_calc_dquots_per_chunk(mp, bp->b_length);
-
-	/*
-	 * On the first read of the buffer, verify that each dquot is valid.
-	 * We don't know what the id of the dquot is supposed to be, just that
-	 * they should be increasing monotonically within the buffer. If the
-	 * first id is corrupt, then it will fail on the second dquot in the
-	 * buffer so corruptions could point to the wrong dquot in this case.
-	 */
-	for (i = 0; i < ndquots; i++) {
-		struct xfs_disk_dquot	*ddq;
-		int			error;
-
-		ddq = &d[i].dd_diskdq;
-
-		if (i == 0)
-			id = be32_to_cpu(ddq->d_id);
-
-		error = xfs_qm_dqcheck(mp, ddq, id + i, 0, XFS_QMOPT_DOWARN,
-				       "xfs_dquot_buf_verify");
-		if (error)
-			return false;
-	}
-	return true;
-}
-
-static void
-xfs_dquot_buf_read_verify(
-	struct xfs_buf	*bp)
-{
-	struct xfs_mount	*mp = bp->b_target->bt_mount;
-
-	if (!xfs_dquot_buf_verify_crc(mp, bp) || !xfs_dquot_buf_verify(mp, bp)) {
-		XFS_CORRUPTION_ERROR(__func__, XFS_ERRLEVEL_LOW, mp, bp->b_addr);
-		xfs_buf_ioerror(bp, EFSCORRUPTED);
-	}
-}
-
-/*
- * we don't calculate the CRC here as that is done when the dquot is flushed to
- * the buffer after the update is done. This ensures that the dquot in the
- * buffer always has an up-to-date CRC value.
- */
-void
-xfs_dquot_buf_write_verify(
-	struct xfs_buf	*bp)
-{
-	struct xfs_mount	*mp = bp->b_target->bt_mount;
-
-	if (!xfs_dquot_buf_verify(mp, bp)) {
-		XFS_CORRUPTION_ERROR(__func__, XFS_ERRLEVEL_LOW, mp, bp->b_addr);
-		xfs_buf_ioerror(bp, EFSCORRUPTED);
-		return;
-	}
-}
-
-const struct xfs_buf_ops xfs_dquot_buf_ops = {
-	.verify_read = xfs_dquot_buf_read_verify,
-	.verify_write = xfs_dquot_buf_write_verify,
-};
-
 /*
  * Allocate a block and fill it with dquots.
  * This is called when the bmapi finds a hole.
@@ -514,6 +402,7 @@ xfs_qm_dqalloc(
 
 	return (error);
 }
+
 STATIC int
 xfs_qm_dqrepair(
 	struct xfs_mount	*mp,
@@ -547,7 +436,7 @@ xfs_qm_dqrepair(
 	/* Do the actual repair of dquots in this buffer */
 	for (i = 0; i < mp->m_quotainfo->qi_dqperchunk; i++) {
 		ddq = &d[i].dd_diskdq;
-		error = xfs_qm_dqcheck(mp, ddq, firstid + i,
+		error = xfs_dqcheck(mp, ddq, firstid + i,
 				       dqp->dq_flags & XFS_DQ_ALLTYPES,
 				       XFS_QMOPT_DQREPAIR, "xfs_qm_dqrepair");
 		if (error) {
@@ -1133,7 +1022,7 @@ xfs_qm_dqflush(
 	/*
 	 * A simple sanity check in case we got a corrupted dquot..
 	 */
-	error = xfs_qm_dqcheck(mp, &dqp->q_core, be32_to_cpu(ddqp->d_id), 0,
+	error = xfs_dqcheck(mp, &dqp->q_core, be32_to_cpu(ddqp->d_id), 0,
 			   XFS_QMOPT_DOWARN, "dqflush (incore copy)");
 	if (error) {
 		xfs_buf_relse(bp);
diff --git a/fs/xfs/xfs_dquot.h b/fs/xfs/xfs_dquot.h
index 55abbca2883d..d22ed0053c32 100644
--- a/fs/xfs/xfs_dquot.h
+++ b/fs/xfs/xfs_dquot.h
@@ -172,6 +172,4 @@ static inline struct xfs_dquot *xfs_qm_dqhold(struct xfs_dquot *dqp)
 	return dqp;
 }
 
-extern const struct xfs_buf_ops xfs_dquot_buf_ops;
-
 #endif /* __XFS_DQUOT_H__ */
diff --git a/fs/xfs/xfs_dquot_buf.c b/fs/xfs/xfs_dquot_buf.c
new file mode 100644
index 000000000000..d401457d2f25
--- /dev/null
+++ b/fs/xfs/xfs_dquot_buf.c
@@ -0,0 +1,288 @@
+/*
+ * Copyright (c) 2000-2006 Silicon Graphics, Inc.
+ * Copyright (c) 2013 Red Hat, Inc.
+ * All Rights Reserved.
+ *
+ * This program is free software; you can redistribute it and/or
+ * modify it under the terms of the GNU General Public License as
+ * published by the Free Software Foundation.
+ *
+ * This program is distributed in the hope that it would be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+ * GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program; if not, write the Free Software Foundation,
+ * Inc.,  51 Franklin St, Fifth Floor, Boston, MA  02110-1301  USA
+ */
+#include "xfs.h"
+#include "xfs_fs.h"
+#include "xfs_shared.h"
+#include "xfs_format.h"
+#include "xfs_log_format.h"
+#include "xfs_trans_resv.h"
+#include "xfs_sb.h"
+#include "xfs_ag.h"
+#include "xfs_mount.h"
+#include "xfs_inode.h"
+#include "xfs_quota.h"
+#include "xfs_trans.h"
+#include "xfs_qm.h"
+#include "xfs_error.h"
+#include "xfs_cksum.h"
+#include "xfs_trace.h"
+
+int
+xfs_calc_dquots_per_chunk(
+	struct xfs_mount	*mp,
+	unsigned int		nbblks)	/* basic block units */
+{
+	unsigned int	ndquots;
+
+	ASSERT(nbblks > 0);
+	ndquots = BBTOB(nbblks);
+	do_div(ndquots, sizeof(xfs_dqblk_t));
+
+	return ndquots;
+}
+
+/*
+ * Do some primitive error checking on ondisk dquot data structures.
+ */
+int
+xfs_dqcheck(
+	struct xfs_mount *mp,
+	xfs_disk_dquot_t *ddq,
+	xfs_dqid_t	 id,
+	uint		 type,	  /* used only when IO_dorepair is true */
+	uint		 flags,
+	char		 *str)
+{
+	xfs_dqblk_t	 *d = (xfs_dqblk_t *)ddq;
+	int		errs = 0;
+
+	/*
+	 * We can encounter an uninitialized dquot buffer for 2 reasons:
+	 * 1. If we crash while deleting the quotainode(s), and those blks got
+	 *    used for user data. This is because we take the path of regular
+	 *    file deletion; however, the size field of quotainodes is never
+	 *    updated, so all the tricks that we play in itruncate_finish
+	 *    don't quite matter.
+	 *
+	 * 2. We don't play the quota buffers when there's a quotaoff logitem.
+	 *    But the allocation will be replayed so we'll end up with an
+	 *    uninitialized quota block.
+	 *
+	 * This is all fine; things are still consistent, and we haven't lost
+	 * any quota information. Just don't complain about bad dquot blks.
+	 */
+	if (ddq->d_magic != cpu_to_be16(XFS_DQUOT_MAGIC)) {
+		if (flags & XFS_QMOPT_DOWARN)
+			xfs_alert(mp,
+			"%s : XFS dquot ID 0x%x, magic 0x%x != 0x%x",
+			str, id, be16_to_cpu(ddq->d_magic), XFS_DQUOT_MAGIC);
+		errs++;
+	}
+	if (ddq->d_version != XFS_DQUOT_VERSION) {
+		if (flags & XFS_QMOPT_DOWARN)
+			xfs_alert(mp,
+			"%s : XFS dquot ID 0x%x, version 0x%x != 0x%x",
+			str, id, ddq->d_version, XFS_DQUOT_VERSION);
+		errs++;
+	}
+
+	if (ddq->d_flags != XFS_DQ_USER &&
+	    ddq->d_flags != XFS_DQ_PROJ &&
+	    ddq->d_flags != XFS_DQ_GROUP) {
+		if (flags & XFS_QMOPT_DOWARN)
+			xfs_alert(mp,
+			"%s : XFS dquot ID 0x%x, unknown flags 0x%x",
+			str, id, ddq->d_flags);
+		errs++;
+	}
+
+	if (id != -1 && id != be32_to_cpu(ddq->d_id)) {
+		if (flags & XFS_QMOPT_DOWARN)
+			xfs_alert(mp,
+			"%s : ondisk-dquot 0x%p, ID mismatch: "
+			"0x%x expected, found id 0x%x",
+			str, ddq, id, be32_to_cpu(ddq->d_id));
+		errs++;
+	}
+
+	if (!errs && ddq->d_id) {
+		if (ddq->d_blk_softlimit &&
+		    be64_to_cpu(ddq->d_bcount) >
+				be64_to_cpu(ddq->d_blk_softlimit)) {
+			if (!ddq->d_btimer) {
+				if (flags & XFS_QMOPT_DOWARN)
+					xfs_alert(mp,
+			"%s : Dquot ID 0x%x (0x%p) BLK TIMER NOT STARTED",
+					str, (int)be32_to_cpu(ddq->d_id), ddq);
+				errs++;
+			}
+		}
+		if (ddq->d_ino_softlimit &&
+		    be64_to_cpu(ddq->d_icount) >
+				be64_to_cpu(ddq->d_ino_softlimit)) {
+			if (!ddq->d_itimer) {
+				if (flags & XFS_QMOPT_DOWARN)
+					xfs_alert(mp,
+			"%s : Dquot ID 0x%x (0x%p) INODE TIMER NOT STARTED",
+					str, (int)be32_to_cpu(ddq->d_id), ddq);
+				errs++;
+			}
+		}
+		if (ddq->d_rtb_softlimit &&
+		    be64_to_cpu(ddq->d_rtbcount) >
+				be64_to_cpu(ddq->d_rtb_softlimit)) {
+			if (!ddq->d_rtbtimer) {
+				if (flags & XFS_QMOPT_DOWARN)
+					xfs_alert(mp,
+			"%s : Dquot ID 0x%x (0x%p) RTBLK TIMER NOT STARTED",
+					str, (int)be32_to_cpu(ddq->d_id), ddq);
+				errs++;
+			}
+		}
+	}
+
+	if (!errs || !(flags & XFS_QMOPT_DQREPAIR))
+		return errs;
+
+	if (flags & XFS_QMOPT_DOWARN)
+		xfs_notice(mp, "Re-initializing dquot ID 0x%x", id);
+
+	/*
+	 * Typically, a repair is only requested by quotacheck.
+	 */
+	ASSERT(id != -1);
+	ASSERT(flags & XFS_QMOPT_DQREPAIR);
+	memset(d, 0, sizeof(xfs_dqblk_t));
+
+	d->dd_diskdq.d_magic = cpu_to_be16(XFS_DQUOT_MAGIC);
+	d->dd_diskdq.d_version = XFS_DQUOT_VERSION;
+	d->dd_diskdq.d_flags = type;
+	d->dd_diskdq.d_id = cpu_to_be32(id);
+
+	if (xfs_sb_version_hascrc(&mp->m_sb)) {
+		uuid_copy(&d->dd_uuid, &mp->m_sb.sb_uuid);
+		xfs_update_cksum((char *)d, sizeof(struct xfs_dqblk),
+				 XFS_DQUOT_CRC_OFF);
+	}
+
+	return errs;
+}
+
+STATIC bool
+xfs_dquot_buf_verify_crc(
+	struct xfs_mount	*mp,
+	struct xfs_buf		*bp)
+{
+	struct xfs_dqblk	*d = (struct xfs_dqblk *)bp->b_addr;
+	int			ndquots;
+	int			i;
+
+	if (!xfs_sb_version_hascrc(&mp->m_sb))
+		return true;
+
+	/*
+	 * if we are in log recovery, the quota subsystem has not been
+	 * initialised so we have no quotainfo structure. In that case, we need
+	 * to manually calculate the number of dquots in the buffer.
+	 */
+	if (mp->m_quotainfo)
+		ndquots = mp->m_quotainfo->qi_dqperchunk;
+	else
+		ndquots = xfs_calc_dquots_per_chunk(mp,
+					XFS_BB_TO_FSB(mp, bp->b_length));
+
+	for (i = 0; i < ndquots; i++, d++) {
+		if (!xfs_verify_cksum((char *)d, sizeof(struct xfs_dqblk),
+				 XFS_DQUOT_CRC_OFF))
+			return false;
+		if (!uuid_equal(&d->dd_uuid, &mp->m_sb.sb_uuid))
+			return false;
+	}
+	return true;
+}
+
+STATIC bool
+xfs_dquot_buf_verify(
+	struct xfs_mount	*mp,
+	struct xfs_buf		*bp)
+{
+	struct xfs_dqblk	*d = (struct xfs_dqblk *)bp->b_addr;
+	xfs_dqid_t		id = 0;
+	int			ndquots;
+	int			i;
+
+	/*
+	 * if we are in log recovery, the quota subsystem has not been
+	 * initialised so we have no quotainfo structure. In that case, we need
+	 * to manually calculate the number of dquots in the buffer.
+	 */
+	if (mp->m_quotainfo)
+		ndquots = mp->m_quotainfo->qi_dqperchunk;
+	else
+		ndquots = xfs_calc_dquots_per_chunk(mp, bp->b_length);
+
+	/*
+	 * On the first read of the buffer, verify that each dquot is valid.
+	 * We don't know what the id of the dquot is supposed to be, just that
+	 * they should be increasing monotonically within the buffer. If the
+	 * first id is corrupt, then it will fail on the second dquot in the
+	 * buffer so corruptions could point to the wrong dquot in this case.
+	 */
+	for (i = 0; i < ndquots; i++) {
+		struct xfs_disk_dquot	*ddq;
+		int			error;
+
+		ddq = &d[i].dd_diskdq;
+
+		if (i == 0)
+			id = be32_to_cpu(ddq->d_id);
+
+		error = xfs_dqcheck(mp, ddq, id + i, 0, XFS_QMOPT_DOWARN,
+				       "xfs_dquot_buf_verify");
+		if (error)
+			return false;
+	}
+	return true;
+}
+
+static void
+xfs_dquot_buf_read_verify(
+	struct xfs_buf	*bp)
+{
+	struct xfs_mount	*mp = bp->b_target->bt_mount;
+
+	if (!xfs_dquot_buf_verify_crc(mp, bp) || !xfs_dquot_buf_verify(mp, bp)) {
+		XFS_CORRUPTION_ERROR(__func__, XFS_ERRLEVEL_LOW, mp, bp->b_addr);
+		xfs_buf_ioerror(bp, EFSCORRUPTED);
+	}
+}
+
+/*
+ * we don't calculate the CRC here as that is done when the dquot is flushed to
+ * the buffer after the update is done. This ensures that the dquot in the
+ * buffer always has an up-to-date CRC value.
+ */
+static void
+xfs_dquot_buf_write_verify(
+	struct xfs_buf	*bp)
+{
+	struct xfs_mount	*mp = bp->b_target->bt_mount;
+
+	if (!xfs_dquot_buf_verify(mp, bp)) {
+		XFS_CORRUPTION_ERROR(__func__, XFS_ERRLEVEL_LOW, mp, bp->b_addr);
+		xfs_buf_ioerror(bp, EFSCORRUPTED);
+		return;
+	}
+}
+
+const struct xfs_buf_ops xfs_dquot_buf_ops = {
+	.verify_read = xfs_dquot_buf_read_verify,
+	.verify_write = xfs_dquot_buf_write_verify,
+};
+
diff --git a/fs/xfs/xfs_dquot_item.c b/fs/xfs/xfs_dquot_item.c
index e838d84b4e85..92e5f62eefc6 100644
--- a/fs/xfs/xfs_dquot_item.c
+++ b/fs/xfs/xfs_dquot_item.c
@@ -18,23 +18,19 @@
 #include "xfs.h"
 #include "xfs_fs.h"
 #include "xfs_format.h"
-#include "xfs_log.h"
-#include "xfs_trans.h"
+#include "xfs_log_format.h"
+#include "xfs_trans_resv.h"
 #include "xfs_sb.h"
 #include "xfs_ag.h"
-#include "xfs_alloc.h"
-#include "xfs_quota.h"
 #include "xfs_mount.h"
-#include "xfs_bmap_btree.h"
 #include "xfs_inode.h"
-#include "xfs_bmap.h"
-#include "xfs_rtalloc.h"
+#include "xfs_quota.h"
 #include "xfs_error.h"
-#include "xfs_itable.h"
-#include "xfs_attr.h"
+#include "xfs_trans.h"
 #include "xfs_buf_item.h"
 #include "xfs_trans_priv.h"
 #include "xfs_qm.h"
+#include "xfs_log.h"
 
 static inline struct xfs_dq_logitem *DQUOT_ITEM(struct xfs_log_item *lip)
 {
diff --git a/fs/xfs/xfs_error.c b/fs/xfs/xfs_error.c
index 1123d93ff795..9995b807d627 100644
--- a/fs/xfs/xfs_error.c
+++ b/fs/xfs/xfs_error.c
@@ -16,16 +16,13 @@
  * Inc.,  51 Franklin St, Fifth Floor, Boston, MA  02110-1301  USA
  */
 #include "xfs.h"
+#include "xfs_format.h"
 #include "xfs_fs.h"
-#include "xfs_types.h"
-#include "xfs_log.h"
-#include "xfs_trans.h"
+#include "xfs_log_format.h"
+#include "xfs_trans_resv.h"
 #include "xfs_sb.h"
 #include "xfs_ag.h"
 #include "xfs_mount.h"
-#include "xfs_bmap_btree.h"
-#include "xfs_dinode.h"
-#include "xfs_inode.h"
 #include "xfs_error.h"
 
 #ifdef DEBUG
@@ -159,7 +156,7 @@ xfs_error_report(
 {
 	if (level <= xfs_error_level) {
 		xfs_alert_tag(mp, XFS_PTAG_ERROR_REPORT,
-		"Internal error %s at line %d of file %s.  Caller 0x%p\n",
+		"Internal error %s at line %d of file %s.  Caller 0x%p",
 			    tag, linenum, filename, ra);
 
 		xfs_stack_trace();
diff --git a/fs/xfs/xfs_export.c b/fs/xfs/xfs_export.c
index 066df425c14f..1399e187d425 100644
--- a/fs/xfs/xfs_export.c
+++ b/fs/xfs/xfs_export.c
@@ -16,21 +16,21 @@
  * Inc.,  51 Franklin St, Fifth Floor, Boston, MA  02110-1301  USA
  */
 #include "xfs.h"
-#include "xfs_types.h"
-#include "xfs_log.h"
-#include "xfs_trans.h"
+#include "xfs_format.h"
+#include "xfs_log_format.h"
+#include "xfs_trans_resv.h"
 #include "xfs_sb.h"
 #include "xfs_ag.h"
 #include "xfs_mount.h"
-#include "xfs_da_btree.h"
-#include "xfs_dir2_format.h"
+#include "xfs_da_format.h"
 #include "xfs_dir2.h"
 #include "xfs_export.h"
-#include "xfs_bmap_btree.h"
 #include "xfs_inode.h"
+#include "xfs_trans.h"
 #include "xfs_inode_item.h"
 #include "xfs_trace.h"
 #include "xfs_icache.h"
+#include "xfs_log.h"
 
 /*
  * Note that we only accept fileids which are long enough rather than allow
diff --git a/fs/xfs/xfs_extent_busy.c b/fs/xfs/xfs_extent_busy.c
index e43708e2f080..fd22f69049d4 100644
--- a/fs/xfs/xfs_extent_busy.c
+++ b/fs/xfs/xfs_extent_busy.c
@@ -19,17 +19,18 @@
  */
 #include "xfs.h"
 #include "xfs_fs.h"
-#include "xfs_types.h"
-#include "xfs_log.h"
-#include "xfs_trans.h"
+#include "xfs_format.h"
+#include "xfs_log_format.h"
+#include "xfs_shared.h"
+#include "xfs_trans_resv.h"
 #include "xfs_sb.h"
 #include "xfs_ag.h"
 #include "xfs_mount.h"
-#include "xfs_bmap_btree.h"
 #include "xfs_alloc.h"
-#include "xfs_inode.h"
 #include "xfs_extent_busy.h"
 #include "xfs_trace.h"
+#include "xfs_trans.h"
+#include "xfs_log.h"
 
 void
 xfs_extent_busy_insert(
diff --git a/fs/xfs/xfs_extent_busy.h b/fs/xfs/xfs_extent_busy.h
index 985412d65ba5..bfff284d2dcc 100644
--- a/fs/xfs/xfs_extent_busy.h
+++ b/fs/xfs/xfs_extent_busy.h
@@ -20,6 +20,10 @@
 #ifndef __XFS_EXTENT_BUSY_H__
 #define	__XFS_EXTENT_BUSY_H__
 
+struct xfs_mount;
+struct xfs_trans;
+struct xfs_alloc_arg;
+
 /*
  * Busy block/extent entry.  Indexed by a rbtree in perag to mark blocks that
  * have been freed but whose transactions aren't committed to disk yet.
diff --git a/fs/xfs/xfs_extfree_item.c b/fs/xfs/xfs_extfree_item.c
index dc53e8febbbe..3680d04f973f 100644
--- a/fs/xfs/xfs_extfree_item.c
+++ b/fs/xfs/xfs_extfree_item.c
@@ -17,14 +17,14 @@
  */
 #include "xfs.h"
 #include "xfs_fs.h"
-#include "xfs_types.h"
-#include "xfs_log.h"
-#include "xfs_trans.h"
-#include "xfs_buf_item.h"
+#include "xfs_log_format.h"
+#include "xfs_trans_resv.h"
 #include "xfs_sb.h"
 #include "xfs_ag.h"
 #include "xfs_mount.h"
+#include "xfs_trans.h"
 #include "xfs_trans_priv.h"
+#include "xfs_buf_item.h"
 #include "xfs_extfree_item.h"
 
 
diff --git a/fs/xfs/xfs_file.c b/fs/xfs/xfs_file.c
index 4c749ab543d0..52c91e143725 100644
--- a/fs/xfs/xfs_file.c
+++ b/fs/xfs/xfs_file.c
@@ -17,25 +17,27 @@
  */
 #include "xfs.h"
 #include "xfs_fs.h"
-#include "xfs_log.h"
+#include "xfs_shared.h"
+#include "xfs_format.h"
+#include "xfs_log_format.h"
+#include "xfs_trans_resv.h"
 #include "xfs_sb.h"
 #include "xfs_ag.h"
-#include "xfs_trans.h"
 #include "xfs_mount.h"
-#include "xfs_bmap_btree.h"
-#include "xfs_alloc.h"
-#include "xfs_dinode.h"
+#include "xfs_da_format.h"
+#include "xfs_da_btree.h"
 #include "xfs_inode.h"
+#include "xfs_trans.h"
 #include "xfs_inode_item.h"
 #include "xfs_bmap.h"
 #include "xfs_bmap_util.h"
 #include "xfs_error.h"
-#include "xfs_da_btree.h"
-#include "xfs_dir2_format.h"
 #include "xfs_dir2.h"
 #include "xfs_dir2_priv.h"
 #include "xfs_ioctl.h"
 #include "xfs_trace.h"
+#include "xfs_log.h"
+#include "xfs_dinode.h"
 
 #include <linux/aio.h>
 #include <linux/dcache.h>
@@ -805,44 +807,64 @@ out:
 
 STATIC long
 xfs_file_fallocate(
-	struct file	*file,
-	int		mode,
-	loff_t		offset,
-	loff_t		len)
+	struct file		*file,
+	int			mode,
+	loff_t			offset,
+	loff_t			len)
 {
-	struct inode	*inode = file_inode(file);
-	long		error;
-	loff_t		new_size = 0;
-	xfs_flock64_t	bf;
-	xfs_inode_t	*ip = XFS_I(inode);
-	int		cmd = XFS_IOC_RESVSP;
-	int		attr_flags = XFS_ATTR_NOLOCK;
+	struct inode		*inode = file_inode(file);
+	struct xfs_inode	*ip = XFS_I(inode);
+	struct xfs_trans	*tp;
+	long			error;
+	loff_t			new_size = 0;
 
+	if (!S_ISREG(inode->i_mode))
+		return -EINVAL;
 	if (mode & ~(FALLOC_FL_KEEP_SIZE | FALLOC_FL_PUNCH_HOLE))
 		return -EOPNOTSUPP;
 
-	bf.l_whence = 0;
-	bf.l_start = offset;
-	bf.l_len = len;
-
 	xfs_ilock(ip, XFS_IOLOCK_EXCL);
+	if (mode & FALLOC_FL_PUNCH_HOLE) {
+		error = xfs_free_file_space(ip, offset, len);
+		if (error)
+			goto out_unlock;
+	} else {
+		if (!(mode & FALLOC_FL_KEEP_SIZE) &&
+		    offset + len > i_size_read(inode)) {
+			new_size = offset + len;
+			error = -inode_newsize_ok(inode, new_size);
+			if (error)
+				goto out_unlock;
+		}
 
-	if (mode & FALLOC_FL_PUNCH_HOLE)
-		cmd = XFS_IOC_UNRESVSP;
-
-	/* check the new inode size is valid before allocating */
-	if (!(mode & FALLOC_FL_KEEP_SIZE) &&
-	    offset + len > i_size_read(inode)) {
-		new_size = offset + len;
-		error = inode_newsize_ok(inode, new_size);
+		error = xfs_alloc_file_space(ip, offset, len,
+					     XFS_BMAPI_PREALLOC);
 		if (error)
 			goto out_unlock;
 	}
 
-	if (file->f_flags & O_DSYNC)
-		attr_flags |= XFS_ATTR_SYNC;
+	tp = xfs_trans_alloc(ip->i_mount, XFS_TRANS_WRITEID);
+	error = xfs_trans_reserve(tp, &M_RES(ip->i_mount)->tr_writeid, 0, 0);
+	if (error) {
+		xfs_trans_cancel(tp, 0);
+		goto out_unlock;
+	}
+
+	xfs_ilock(ip, XFS_ILOCK_EXCL);
+	xfs_trans_ijoin(tp, ip, XFS_ILOCK_EXCL);
+	ip->i_d.di_mode &= ~S_ISUID;
+	if (ip->i_d.di_mode & S_IXGRP)
+		ip->i_d.di_mode &= ~S_ISGID;
+
+	if (!(mode & FALLOC_FL_PUNCH_HOLE))
+		ip->i_d.di_flags |= XFS_DIFLAG_PREALLOC;
 
-	error = -xfs_change_file_space(ip, cmd, &bf, 0, attr_flags);
+	xfs_trans_ichgtime(tp, ip, XFS_ICHGTIME_MOD | XFS_ICHGTIME_CHG);
+	xfs_trans_log_inode(tp, ip, XFS_ILOG_CORE);
+
+	if (file->f_flags & O_DSYNC)
+		xfs_trans_set_sync(tp);
+	error = xfs_trans_commit(tp, 0);
 	if (error)
 		goto out_unlock;
 
@@ -852,12 +874,12 @@ xfs_file_fallocate(
 
 		iattr.ia_valid = ATTR_SIZE;
 		iattr.ia_size = new_size;
-		error = -xfs_setattr_size(ip, &iattr, XFS_ATTR_NOLOCK);
+		error = xfs_setattr_size(ip, &iattr);
 	}
 
 out_unlock:
 	xfs_iunlock(ip, XFS_IOLOCK_EXCL);
-	return error;
+	return -error;
 }
 
 
diff --git a/fs/xfs/xfs_filestream.c b/fs/xfs/xfs_filestream.c
index ce78e654d37b..12b6e7701985 100644
--- a/fs/xfs/xfs_filestream.c
+++ b/fs/xfs/xfs_filestream.c
@@ -16,19 +16,19 @@
  * Inc.,  51 Franklin St, Fifth Floor, Boston, MA  02110-1301  USA
  */
 #include "xfs.h"
-#include "xfs_log.h"
-#include "xfs_bmap_btree.h"
-#include "xfs_inum.h"
-#include "xfs_dinode.h"
-#include "xfs_inode.h"
+#include "xfs_format.h"
+#include "xfs_log_format.h"
+#include "xfs_trans_resv.h"
 #include "xfs_ag.h"
-#include "xfs_trans.h"
 #include "xfs_sb.h"
 #include "xfs_mount.h"
+#include "xfs_inum.h"
+#include "xfs_inode.h"
 #include "xfs_bmap.h"
 #include "xfs_bmap_util.h"
 #include "xfs_alloc.h"
 #include "xfs_mru_cache.h"
+#include "xfs_dinode.h"
 #include "xfs_filestream.h"
 #include "xfs_trace.h"
 
diff --git a/fs/xfs/xfs_format.h b/fs/xfs/xfs_format.h
index 35c08ff54ca0..b6ab5a3cfa12 100644
--- a/fs/xfs/xfs_format.h
+++ b/fs/xfs/xfs_format.h
@@ -156,14 +156,259 @@ struct xfs_dsymlink_hdr {
 	((bufsize) - (xfs_sb_version_hascrc(&(mp)->m_sb) ? \
 			sizeof(struct xfs_dsymlink_hdr) : 0))
 
-int xfs_symlink_blocks(struct xfs_mount *mp, int pathlen);
-int xfs_symlink_hdr_set(struct xfs_mount *mp, xfs_ino_t ino, uint32_t offset,
-			uint32_t size, struct xfs_buf *bp);
-bool xfs_symlink_hdr_ok(struct xfs_mount *mp, xfs_ino_t ino, uint32_t offset,
-			uint32_t size, struct xfs_buf *bp);
-void xfs_symlink_local_to_remote(struct xfs_trans *tp, struct xfs_buf *bp,
-				 struct xfs_inode *ip, struct xfs_ifork *ifp);
-
-extern const struct xfs_buf_ops xfs_symlink_buf_ops;
+
+/*
+ * Allocation Btree format definitions
+ *
+ * There are two on-disk btrees, one sorted by blockno and one sorted
+ * by blockcount and blockno.  All blocks look the same to make the code
+ * simpler; if we have time later, we'll make the optimizations.
+ */
+#define	XFS_ABTB_MAGIC		0x41425442	/* 'ABTB' for bno tree */
+#define	XFS_ABTB_CRC_MAGIC	0x41423342	/* 'AB3B' */
+#define	XFS_ABTC_MAGIC		0x41425443	/* 'ABTC' for cnt tree */
+#define	XFS_ABTC_CRC_MAGIC	0x41423343	/* 'AB3C' */
+
+/*
+ * Data record/key structure
+ */
+typedef struct xfs_alloc_rec {
+	__be32		ar_startblock;	/* starting block number */
+	__be32		ar_blockcount;	/* count of free blocks */
+} xfs_alloc_rec_t, xfs_alloc_key_t;
+
+typedef struct xfs_alloc_rec_incore {
+	xfs_agblock_t	ar_startblock;	/* starting block number */
+	xfs_extlen_t	ar_blockcount;	/* count of free blocks */
+} xfs_alloc_rec_incore_t;
+
+/* btree pointer type */
+typedef __be32 xfs_alloc_ptr_t;
+
+/*
+ * Block numbers in the AG:
+ * SB is sector 0, AGF is sector 1, AGI is sector 2, AGFL is sector 3.
+ */
+#define	XFS_BNO_BLOCK(mp)	((xfs_agblock_t)(XFS_AGFL_BLOCK(mp) + 1))
+#define	XFS_CNT_BLOCK(mp)	((xfs_agblock_t)(XFS_BNO_BLOCK(mp) + 1))
+
+
+/*
+ * Inode Allocation Btree format definitions
+ *
+ * There is a btree for the inode map per allocation group.
+ */
+#define	XFS_IBT_MAGIC		0x49414254	/* 'IABT' */
+#define	XFS_IBT_CRC_MAGIC	0x49414233	/* 'IAB3' */
+
+typedef	__uint64_t	xfs_inofree_t;
+#define	XFS_INODES_PER_CHUNK		(NBBY * sizeof(xfs_inofree_t))
+#define	XFS_INODES_PER_CHUNK_LOG	(XFS_NBBYLOG + 3)
+#define	XFS_INOBT_ALL_FREE		((xfs_inofree_t)-1)
+#define	XFS_INOBT_MASK(i)		((xfs_inofree_t)1 << (i))
+
+static inline xfs_inofree_t xfs_inobt_maskn(int i, int n)
+{
+	return ((n >= XFS_INODES_PER_CHUNK ? 0 : XFS_INOBT_MASK(n)) - 1) << i;
+}
+
+/*
+ * Data record structure
+ */
+typedef struct xfs_inobt_rec {
+	__be32		ir_startino;	/* starting inode number */
+	__be32		ir_freecount;	/* count of free inodes (set bits) */
+	__be64		ir_free;	/* free inode mask */
+} xfs_inobt_rec_t;
+
+typedef struct xfs_inobt_rec_incore {
+	xfs_agino_t	ir_startino;	/* starting inode number */
+	__int32_t	ir_freecount;	/* count of free inodes (set bits) */
+	xfs_inofree_t	ir_free;	/* free inode mask */
+} xfs_inobt_rec_incore_t;
+
+
+/*
+ * Key structure
+ */
+typedef struct xfs_inobt_key {
+	__be32		ir_startino;	/* starting inode number */
+} xfs_inobt_key_t;
+
+/* btree pointer type */
+typedef __be32 xfs_inobt_ptr_t;
+
+/*
+ * block numbers in the AG.
+ */
+#define	XFS_IBT_BLOCK(mp)		((xfs_agblock_t)(XFS_CNT_BLOCK(mp) + 1))
+#define	XFS_PREALLOC_BLOCKS(mp)		((xfs_agblock_t)(XFS_IBT_BLOCK(mp) + 1))
+
+
+
+/*
+ * BMAP Btree format definitions
+ *
+ * This includes both the root block definition that sits inside an inode fork
+ * and the record/pointer formats for the leaf/node in the blocks.
+ */
+#define XFS_BMAP_MAGIC		0x424d4150	/* 'BMAP' */
+#define XFS_BMAP_CRC_MAGIC	0x424d4133	/* 'BMA3' */
+
+/*
+ * Bmap root header, on-disk form only.
+ */
+typedef struct xfs_bmdr_block {
+	__be16		bb_level;	/* 0 is a leaf */
+	__be16		bb_numrecs;	/* current # of data records */
+} xfs_bmdr_block_t;
+
+/*
+ * Bmap btree record and extent descriptor.
+ *  l0:63 is an extent flag (value 1 indicates non-normal).
+ *  l0:9-62 are startoff.
+ *  l0:0-8 and l1:21-63 are startblock.
+ *  l1:0-20 are blockcount.
+ */
+#define BMBT_EXNTFLAG_BITLEN	1
+#define BMBT_STARTOFF_BITLEN	54
+#define BMBT_STARTBLOCK_BITLEN	52
+#define BMBT_BLOCKCOUNT_BITLEN	21
+
+typedef struct xfs_bmbt_rec {
+	__be64			l0, l1;
+} xfs_bmbt_rec_t;
+
+typedef __uint64_t	xfs_bmbt_rec_base_t;	/* use this for casts */
+typedef xfs_bmbt_rec_t xfs_bmdr_rec_t;
+
+typedef struct xfs_bmbt_rec_host {
+	__uint64_t		l0, l1;
+} xfs_bmbt_rec_host_t;
+
+/*
+ * Values and macros for delayed-allocation startblock fields.
+ */
+#define STARTBLOCKVALBITS	17
+#define STARTBLOCKMASKBITS	(15 + XFS_BIG_BLKNOS * 20)
+#define DSTARTBLOCKMASKBITS	(15 + 20)
+#define STARTBLOCKMASK		\
+	(((((xfs_fsblock_t)1) << STARTBLOCKMASKBITS) - 1) << STARTBLOCKVALBITS)
+#define DSTARTBLOCKMASK		\
+	(((((xfs_dfsbno_t)1) << DSTARTBLOCKMASKBITS) - 1) << STARTBLOCKVALBITS)
+
+static inline int isnullstartblock(xfs_fsblock_t x)
+{
+	return ((x) & STARTBLOCKMASK) == STARTBLOCKMASK;
+}
+
+static inline int isnulldstartblock(xfs_dfsbno_t x)
+{
+	return ((x) & DSTARTBLOCKMASK) == DSTARTBLOCKMASK;
+}
+
+static inline xfs_fsblock_t nullstartblock(int k)
+{
+	ASSERT(k < (1 << STARTBLOCKVALBITS));
+	return STARTBLOCKMASK | (k);
+}
+
+static inline xfs_filblks_t startblockval(xfs_fsblock_t x)
+{
+	return (xfs_filblks_t)((x) & ~STARTBLOCKMASK);
+}
+
+/*
+ * Possible extent formats.
+ */
+typedef enum {
+	XFS_EXTFMT_NOSTATE = 0,
+	XFS_EXTFMT_HASSTATE
+} xfs_exntfmt_t;
+
+/*
+ * Possible extent states.
+ */
+typedef enum {
+	XFS_EXT_NORM, XFS_EXT_UNWRITTEN,
+	XFS_EXT_DMAPI_OFFLINE, XFS_EXT_INVALID
+} xfs_exntst_t;
+
+/*
+ * Incore version of above.
+ */
+typedef struct xfs_bmbt_irec
+{
+	xfs_fileoff_t	br_startoff;	/* starting file offset */
+	xfs_fsblock_t	br_startblock;	/* starting block number */
+	xfs_filblks_t	br_blockcount;	/* number of blocks */
+	xfs_exntst_t	br_state;	/* extent state */
+} xfs_bmbt_irec_t;
+
+/*
+ * Key structure for non-leaf levels of the tree.
+ */
+typedef struct xfs_bmbt_key {
+	__be64		br_startoff;	/* starting file offset */
+} xfs_bmbt_key_t, xfs_bmdr_key_t;
+
+/* btree pointer type */
+typedef __be64 xfs_bmbt_ptr_t, xfs_bmdr_ptr_t;
+
+
+/*
+ * Generic Btree block format definitions
+ *
+ * This is a combination of the actual format used on disk for short and long
+ * format btrees.  The first three fields are shared by both format, but the
+ * pointers are different and should be used with care.
+ *
+ * To get the size of the actual short or long form headers please use the size
+ * macros below.  Never use sizeof(xfs_btree_block).
+ *
+ * The blkno, crc, lsn, owner and uuid fields are only available in filesystems
+ * with the crc feature bit, and all accesses to them must be conditional on
+ * that flag.
+ */
+struct xfs_btree_block {
+	__be32		bb_magic;	/* magic number for block type */
+	__be16		bb_level;	/* 0 is a leaf */
+	__be16		bb_numrecs;	/* current # of data records */
+	union {
+		struct {
+			__be32		bb_leftsib;
+			__be32		bb_rightsib;
+
+			__be64		bb_blkno;
+			__be64		bb_lsn;
+			uuid_t		bb_uuid;
+			__be32		bb_owner;
+			__le32		bb_crc;
+		} s;			/* short form pointers */
+		struct	{
+			__be64		bb_leftsib;
+			__be64		bb_rightsib;
+
+			__be64		bb_blkno;
+			__be64		bb_lsn;
+			uuid_t		bb_uuid;
+			__be64		bb_owner;
+			__le32		bb_crc;
+			__be32		bb_pad; /* padding for alignment */
+		} l;			/* long form pointers */
+	} bb_u;				/* rest */
+};
+
+#define XFS_BTREE_SBLOCK_LEN	16	/* size of a short form block */
+#define XFS_BTREE_LBLOCK_LEN	24	/* size of a long form block */
+
+/* sizes of CRC enabled btree blocks */
+#define XFS_BTREE_SBLOCK_CRC_LEN	(XFS_BTREE_SBLOCK_LEN + 40)
+#define XFS_BTREE_LBLOCK_CRC_LEN	(XFS_BTREE_LBLOCK_LEN + 48)
+
+#define XFS_BTREE_SBLOCK_CRC_OFF \
+	offsetof(struct xfs_btree_block, bb_u.s.bb_crc)
+#define XFS_BTREE_LBLOCK_CRC_OFF \
+	offsetof(struct xfs_btree_block, bb_u.l.bb_crc)
 
 #endif /* __XFS_FORMAT_H__ */
diff --git a/fs/xfs/xfs_fs.h b/fs/xfs/xfs_fs.h
index 18272c766a50..c5fc116dfaa3 100644
--- a/fs/xfs/xfs_fs.h
+++ b/fs/xfs/xfs_fs.h
@@ -233,11 +233,11 @@ typedef struct xfs_fsop_resblks {
 #define XFS_FSOP_GEOM_FLAGS_LOGV2	0x0100	/* log format version 2	*/
 #define XFS_FSOP_GEOM_FLAGS_SECTOR	0x0200	/* sector sizes >1BB	*/
 #define XFS_FSOP_GEOM_FLAGS_ATTR2	0x0400	/* inline attributes rework */
-#define XFS_FSOP_GEOM_FLAGS_PROJID32	0x0800  /* 32-bit project IDs	*/
+#define XFS_FSOP_GEOM_FLAGS_PROJID32	0x0800	/* 32-bit project IDs	*/
 #define XFS_FSOP_GEOM_FLAGS_DIRV2CI	0x1000	/* ASCII only CI names	*/
 #define XFS_FSOP_GEOM_FLAGS_LAZYSB	0x4000	/* lazy superblock counters */
 #define XFS_FSOP_GEOM_FLAGS_V5SB	0x8000	/* version 5 superblock */
-
+#define XFS_FSOP_GEOM_FLAGS_FTYPE	0x10000	/* inode directory types */
 
 /*
  * Minimum and maximum sizes need for growth checks.
diff --git a/fs/xfs/xfs_fsops.c b/fs/xfs/xfs_fsops.c
index e64ee5288b86..a6e54b3319bd 100644
--- a/fs/xfs/xfs_fsops.c
+++ b/fs/xfs/xfs_fsops.c
@@ -17,28 +17,29 @@
  */
 #include "xfs.h"
 #include "xfs_fs.h"
-#include "xfs_types.h"
-#include "xfs_log.h"
-#include "xfs_trans.h"
+#include "xfs_shared.h"
+#include "xfs_format.h"
+#include "xfs_log_format.h"
+#include "xfs_trans_resv.h"
 #include "xfs_sb.h"
 #include "xfs_ag.h"
 #include "xfs_mount.h"
-#include "xfs_bmap_btree.h"
-#include "xfs_alloc_btree.h"
-#include "xfs_ialloc_btree.h"
-#include "xfs_dinode.h"
 #include "xfs_inode.h"
+#include "xfs_trans.h"
 #include "xfs_inode_item.h"
-#include "xfs_btree.h"
 #include "xfs_error.h"
+#include "xfs_btree.h"
+#include "xfs_alloc_btree.h"
 #include "xfs_alloc.h"
 #include "xfs_ialloc.h"
 #include "xfs_fsops.h"
 #include "xfs_itable.h"
 #include "xfs_trans_space.h"
 #include "xfs_rtalloc.h"
-#include "xfs_filestream.h"
 #include "xfs_trace.h"
+#include "xfs_log.h"
+#include "xfs_dinode.h"
+#include "xfs_filestream.h"
 
 /*
  * File system operations
@@ -101,7 +102,9 @@ xfs_fs_geometry(
 			(xfs_sb_version_hasprojid32bit(&mp->m_sb) ?
 				XFS_FSOP_GEOM_FLAGS_PROJID32 : 0) |
 			(xfs_sb_version_hascrc(&mp->m_sb) ?
-				XFS_FSOP_GEOM_FLAGS_V5SB : 0);
+				XFS_FSOP_GEOM_FLAGS_V5SB : 0) |
+			(xfs_sb_version_hasftype(&mp->m_sb) ?
+				XFS_FSOP_GEOM_FLAGS_FTYPE : 0);
 		geo->logsectsize = xfs_sb_version_hassector(&mp->m_sb) ?
 				mp->m_sb.sb_logsectsize : BBSIZE;
 		geo->rtsectsize = mp->m_sb.sb_blocksize;
@@ -153,7 +156,7 @@ xfs_growfs_data_private(
 	xfs_buf_t		*bp;
 	int			bucket;
 	int			dpct;
-	int			error;
+	int			error, saved_error = 0;
 	xfs_agnumber_t		nagcount;
 	xfs_agnumber_t		nagimax = 0;
 	xfs_rfsblock_t		nb, nb_mod;
@@ -496,29 +499,33 @@ xfs_growfs_data_private(
 				error = ENOMEM;
 		}
 
+		/*
+		 * If we get an error reading or writing alternate superblocks,
+		 * continue.  xfs_repair chooses the "best" superblock based
+		 * on most matches; if we break early, we'll leave more
+		 * superblocks un-updated than updated, and xfs_repair may
+		 * pick them over the properly-updated primary.
+		 */
 		if (error) {
 			xfs_warn(mp,
 		"error %d reading secondary superblock for ag %d",
 				error, agno);
-			break;
+			saved_error = error;
+			continue;
 		}
 		xfs_sb_to_disk(XFS_BUF_TO_SBP(bp), &mp->m_sb, XFS_SB_ALL_BITS);
 
-		/*
-		 * If we get an error writing out the alternate superblocks,
-		 * just issue a warning and continue.  The real work is
-		 * already done and committed.
-		 */
 		error = xfs_bwrite(bp);
 		xfs_buf_relse(bp);
 		if (error) {
 			xfs_warn(mp,
 		"write error %d updating secondary superblock for ag %d",
 				error, agno);
-			break; /* no point in continuing */
+			saved_error = error;
+			continue;
 		}
 	}
-	return error;
+	return saved_error ? saved_error : error;
 
  error0:
 	xfs_trans_cancel(tp, XFS_TRANS_ABORT);
diff --git a/fs/xfs/xfs_ialloc.c b/fs/xfs/xfs_ialloc.c
index ccf2fb143962..e87719c5bebe 100644
--- a/fs/xfs/xfs_ialloc.c
+++ b/fs/xfs/xfs_ialloc.c
@@ -17,29 +17,30 @@
  */
 #include "xfs.h"
 #include "xfs_fs.h"
-#include "xfs_types.h"
+#include "xfs_shared.h"
+#include "xfs_format.h"
+#include "xfs_log_format.h"
+#include "xfs_trans_resv.h"
 #include "xfs_bit.h"
-#include "xfs_log.h"
 #include "xfs_inum.h"
-#include "xfs_trans.h"
 #include "xfs_sb.h"
 #include "xfs_ag.h"
 #include "xfs_mount.h"
-#include "xfs_bmap_btree.h"
-#include "xfs_alloc_btree.h"
-#include "xfs_ialloc_btree.h"
-#include "xfs_dinode.h"
 #include "xfs_inode.h"
 #include "xfs_btree.h"
 #include "xfs_ialloc.h"
+#include "xfs_ialloc_btree.h"
 #include "xfs_alloc.h"
 #include "xfs_rtalloc.h"
 #include "xfs_error.h"
 #include "xfs_bmap.h"
 #include "xfs_cksum.h"
+#include "xfs_trans.h"
 #include "xfs_buf_item.h"
 #include "xfs_icreate_item.h"
 #include "xfs_icache.h"
+#include "xfs_dinode.h"
+#include "xfs_trace.h"
 
 
 /*
@@ -1627,8 +1628,9 @@ xfs_read_agi(
 {
 	int			error;
 
-	ASSERT(agno != NULLAGNUMBER);
+	trace_xfs_read_agi(mp, agno);
 
+	ASSERT(agno != NULLAGNUMBER);
 	error = xfs_trans_read_buf(mp, tp, mp->m_ddev_targp,
 			XFS_AG_DADDR(mp, agno, XFS_AGI_DADDR(mp)),
 			XFS_FSS_TO_BB(mp, 1), 0, bpp, &xfs_agi_buf_ops);
@@ -1651,6 +1653,8 @@ xfs_ialloc_read_agi(
 	struct xfs_perag	*pag;	/* per allocation group data */
 	int			error;
 
+	trace_xfs_ialloc_read_agi(mp, agno);
+
 	error = xfs_read_agi(mp, tp, agno, bpp);
 	if (error)
 		return error;
diff --git a/fs/xfs/xfs_ialloc.h b/fs/xfs/xfs_ialloc.h
index 68c07320f096..a8f76a5ff418 100644
--- a/fs/xfs/xfs_ialloc.h
+++ b/fs/xfs/xfs_ialloc.h
@@ -23,6 +23,7 @@ struct xfs_dinode;
 struct xfs_imap;
 struct xfs_mount;
 struct xfs_trans;
+struct xfs_btree_cur;
 
 /*
  * Allocation parameters for inode allocation.
@@ -42,7 +43,7 @@ struct xfs_trans;
 static inline struct xfs_dinode *
 xfs_make_iptr(struct xfs_mount *mp, struct xfs_buf *b, int o)
 {
-	return (xfs_dinode_t *)
+	return (struct xfs_dinode *)
 		(xfs_buf_offset(b, o << (mp)->m_sb.sb_inodelog));
 }
 
@@ -158,6 +159,4 @@ int xfs_ialloc_inode_init(struct xfs_mount *mp, struct xfs_trans *tp,
 			  xfs_agnumber_t agno, xfs_agblock_t agbno,
 			  xfs_agblock_t length, unsigned int gen);
 
-extern const struct xfs_buf_ops xfs_agi_buf_ops;
-
 #endif	/* __XFS_IALLOC_H__ */
diff --git a/fs/xfs/xfs_ialloc_btree.c b/fs/xfs/xfs_ialloc_btree.c
index 5448eb6b8c12..c8fa5bbb36de 100644
--- a/fs/xfs/xfs_ialloc_btree.c
+++ b/fs/xfs/xfs_ialloc_btree.c
@@ -17,24 +17,23 @@
  */
 #include "xfs.h"
 #include "xfs_fs.h"
-#include "xfs_types.h"
+#include "xfs_shared.h"
+#include "xfs_format.h"
+#include "xfs_log_format.h"
+#include "xfs_trans_resv.h"
 #include "xfs_bit.h"
-#include "xfs_log.h"
-#include "xfs_trans.h"
 #include "xfs_sb.h"
 #include "xfs_ag.h"
 #include "xfs_mount.h"
-#include "xfs_bmap_btree.h"
-#include "xfs_alloc_btree.h"
-#include "xfs_ialloc_btree.h"
-#include "xfs_dinode.h"
 #include "xfs_inode.h"
 #include "xfs_btree.h"
 #include "xfs_ialloc.h"
+#include "xfs_ialloc_btree.h"
 #include "xfs_alloc.h"
 #include "xfs_error.h"
 #include "xfs_trace.h"
 #include "xfs_cksum.h"
+#include "xfs_trans.h"
 
 
 STATIC int
diff --git a/fs/xfs/xfs_ialloc_btree.h b/fs/xfs/xfs_ialloc_btree.h
index 3ac36b7642e9..f38b22011c4e 100644
--- a/fs/xfs/xfs_ialloc_btree.h
+++ b/fs/xfs/xfs_ialloc_btree.h
@@ -27,55 +27,6 @@ struct xfs_btree_cur;
 struct xfs_mount;
 
 /*
- * There is a btree for the inode map per allocation group.
- */
-#define	XFS_IBT_MAGIC		0x49414254	/* 'IABT' */
-#define	XFS_IBT_CRC_MAGIC	0x49414233	/* 'IAB3' */
-
-typedef	__uint64_t	xfs_inofree_t;
-#define	XFS_INODES_PER_CHUNK		(NBBY * sizeof(xfs_inofree_t))
-#define	XFS_INODES_PER_CHUNK_LOG	(XFS_NBBYLOG + 3)
-#define	XFS_INOBT_ALL_FREE		((xfs_inofree_t)-1)
-#define	XFS_INOBT_MASK(i)		((xfs_inofree_t)1 << (i))
-
-static inline xfs_inofree_t xfs_inobt_maskn(int i, int n)
-{
-	return ((n >= XFS_INODES_PER_CHUNK ? 0 : XFS_INOBT_MASK(n)) - 1) << i;
-}
-
-/*
- * Data record structure
- */
-typedef struct xfs_inobt_rec {
-	__be32		ir_startino;	/* starting inode number */
-	__be32		ir_freecount;	/* count of free inodes (set bits) */
-	__be64		ir_free;	/* free inode mask */
-} xfs_inobt_rec_t;
-
-typedef struct xfs_inobt_rec_incore {
-	xfs_agino_t	ir_startino;	/* starting inode number */
-	__int32_t	ir_freecount;	/* count of free inodes (set bits) */
-	xfs_inofree_t	ir_free;	/* free inode mask */
-} xfs_inobt_rec_incore_t;
-
-
-/*
- * Key structure
- */
-typedef struct xfs_inobt_key {
-	__be32		ir_startino;	/* starting inode number */
-} xfs_inobt_key_t;
-
-/* btree pointer type */
-typedef __be32 xfs_inobt_ptr_t;
-
-/*
- * block numbers in the AG.
- */
-#define	XFS_IBT_BLOCK(mp)		((xfs_agblock_t)(XFS_CNT_BLOCK(mp) + 1))
-#define	XFS_PREALLOC_BLOCKS(mp)		((xfs_agblock_t)(XFS_IBT_BLOCK(mp) + 1))
-
-/*
  * Btree block header size depends on a superblock flag.
  */
 #define XFS_INOBT_BLOCK_LEN(mp) \
@@ -110,6 +61,4 @@ extern struct xfs_btree_cur *xfs_inobt_init_cursor(struct xfs_mount *,
 		struct xfs_trans *, struct xfs_buf *, xfs_agnumber_t);
 extern int xfs_inobt_maxrecs(struct xfs_mount *, int, int);
 
-extern const struct xfs_buf_ops xfs_inobt_buf_ops;
-
 #endif	/* __XFS_IALLOC_BTREE_H__ */
diff --git a/fs/xfs/xfs_icache.c b/fs/xfs/xfs_icache.c
index 474807a401c8..98d35244eecc 100644
--- a/fs/xfs/xfs_icache.c
+++ b/fs/xfs/xfs_icache.c
@@ -18,24 +18,19 @@
 #include "xfs.h"
 #include "xfs_fs.h"
 #include "xfs_format.h"
-#include "xfs_types.h"
-#include "xfs_log.h"
-#include "xfs_log_priv.h"
+#include "xfs_log_format.h"
+#include "xfs_trans_resv.h"
 #include "xfs_inum.h"
-#include "xfs_trans.h"
-#include "xfs_trans_priv.h"
 #include "xfs_sb.h"
 #include "xfs_ag.h"
 #include "xfs_mount.h"
-#include "xfs_bmap_btree.h"
 #include "xfs_inode.h"
-#include "xfs_dinode.h"
 #include "xfs_error.h"
-#include "xfs_filestream.h"
+#include "xfs_trans.h"
+#include "xfs_trans_priv.h"
 #include "xfs_inode_item.h"
 #include "xfs_quota.h"
 #include "xfs_trace.h"
-#include "xfs_fsops.h"
 #include "xfs_icache.h"
 #include "xfs_bmap_util.h"
 
@@ -500,11 +495,6 @@ xfs_inode_ag_walk_grab(
 	if (!igrab(inode))
 		return ENOENT;
 
-	if (is_bad_inode(inode)) {
-		IRELE(ip);
-		return ENOENT;
-	}
-
 	/* inode is valid */
 	return 0;
 
@@ -918,8 +908,6 @@ restart:
 		xfs_iflock(ip);
 	}
 
-	if (is_bad_inode(VFS_I(ip)))
-		goto reclaim;
 	if (XFS_FORCED_SHUTDOWN(ip->i_mount)) {
 		xfs_iunpin_wait(ip);
 		xfs_iflush_abort(ip, false);
diff --git a/fs/xfs/xfs_icreate_item.c b/fs/xfs/xfs_icreate_item.c
index 5a5a593994d4..d2eaccfa73f4 100644
--- a/fs/xfs/xfs_icreate_item.c
+++ b/fs/xfs/xfs_icreate_item.c
@@ -17,13 +17,14 @@
  */
 #include "xfs.h"
 #include "xfs_fs.h"
-#include "xfs_types.h"
+#include "xfs_shared.h"
+#include "xfs_log_format.h"
+#include "xfs_trans_resv.h"
 #include "xfs_bit.h"
-#include "xfs_log.h"
-#include "xfs_trans.h"
 #include "xfs_sb.h"
 #include "xfs_ag.h"
 #include "xfs_mount.h"
+#include "xfs_trans.h"
 #include "xfs_trans_priv.h"
 #include "xfs_error.h"
 #include "xfs_icreate_item.h"
diff --git a/fs/xfs/xfs_inode.c b/fs/xfs/xfs_inode.c
index e3d75385aa76..001aa893ed59 100644
--- a/fs/xfs/xfs_inode.c
+++ b/fs/xfs/xfs_inode.c
@@ -19,39 +19,38 @@
 
 #include "xfs.h"
 #include "xfs_fs.h"
+#include "xfs_shared.h"
 #include "xfs_format.h"
-#include "xfs_log.h"
+#include "xfs_log_format.h"
+#include "xfs_trans_resv.h"
 #include "xfs_inum.h"
-#include "xfs_trans.h"
-#include "xfs_trans_space.h"
-#include "xfs_trans_priv.h"
 #include "xfs_sb.h"
 #include "xfs_ag.h"
 #include "xfs_mount.h"
+#include "xfs_inode.h"
+#include "xfs_da_format.h"
 #include "xfs_da_btree.h"
-#include "xfs_dir2_format.h"
 #include "xfs_dir2.h"
-#include "xfs_bmap_btree.h"
-#include "xfs_alloc_btree.h"
-#include "xfs_ialloc_btree.h"
 #include "xfs_attr_sf.h"
 #include "xfs_attr.h"
-#include "xfs_dinode.h"
-#include "xfs_inode.h"
+#include "xfs_trans_space.h"
+#include "xfs_trans.h"
 #include "xfs_buf_item.h"
 #include "xfs_inode_item.h"
-#include "xfs_btree.h"
-#include "xfs_alloc.h"
 #include "xfs_ialloc.h"
 #include "xfs_bmap.h"
 #include "xfs_bmap_util.h"
 #include "xfs_error.h"
 #include "xfs_quota.h"
+#include "xfs_dinode.h"
 #include "xfs_filestream.h"
 #include "xfs_cksum.h"
 #include "xfs_trace.h"
 #include "xfs_icache.h"
 #include "xfs_symlink.h"
+#include "xfs_trans_priv.h"
+#include "xfs_log.h"
+#include "xfs_bmap_btree.h"
 
 kmem_zone_t *xfs_inode_zone;
 
@@ -1663,6 +1662,126 @@ xfs_release(
 }
 
 /*
+ * xfs_inactive_truncate
+ *
+ * Called to perform a truncate when an inode becomes unlinked.
+ */
+STATIC int
+xfs_inactive_truncate(
+	struct xfs_inode *ip)
+{
+	struct xfs_mount	*mp = ip->i_mount;
+	struct xfs_trans	*tp;
+	int			error;
+
+	tp = xfs_trans_alloc(mp, XFS_TRANS_INACTIVE);
+	error = xfs_trans_reserve(tp, &M_RES(mp)->tr_itruncate, 0, 0);
+	if (error) {
+		ASSERT(XFS_FORCED_SHUTDOWN(mp));
+		xfs_trans_cancel(tp, 0);
+		return error;
+	}
+
+	xfs_ilock(ip, XFS_ILOCK_EXCL);
+	xfs_trans_ijoin(tp, ip, 0);
+
+	/*
+	 * Log the inode size first to prevent stale data exposure in the event
+	 * of a system crash before the truncate completes. See the related
+	 * comment in xfs_setattr_size() for details.
+	 */
+	ip->i_d.di_size = 0;
+	xfs_trans_log_inode(tp, ip, XFS_ILOG_CORE);
+
+	error = xfs_itruncate_extents(&tp, ip, XFS_DATA_FORK, 0);
+	if (error)
+		goto error_trans_cancel;
+
+	ASSERT(ip->i_d.di_nextents == 0);
+
+	error = xfs_trans_commit(tp, XFS_TRANS_RELEASE_LOG_RES);
+	if (error)
+		goto error_unlock;
+
+	xfs_iunlock(ip, XFS_ILOCK_EXCL);
+	return 0;
+
+error_trans_cancel:
+	xfs_trans_cancel(tp, XFS_TRANS_RELEASE_LOG_RES | XFS_TRANS_ABORT);
+error_unlock:
+	xfs_iunlock(ip, XFS_ILOCK_EXCL);
+	return error;
+}
+
+/*
+ * xfs_inactive_ifree()
+ *
+ * Perform the inode free when an inode is unlinked.
+ */
+STATIC int
+xfs_inactive_ifree(
+	struct xfs_inode *ip)
+{
+	xfs_bmap_free_t		free_list;
+	xfs_fsblock_t		first_block;
+	int			committed;
+	struct xfs_mount	*mp = ip->i_mount;
+	struct xfs_trans	*tp;
+	int			error;
+
+	tp = xfs_trans_alloc(mp, XFS_TRANS_INACTIVE);
+	error = xfs_trans_reserve(tp, &M_RES(mp)->tr_ifree, 0, 0);
+	if (error) {
+		ASSERT(XFS_FORCED_SHUTDOWN(mp));
+		xfs_trans_cancel(tp, XFS_TRANS_RELEASE_LOG_RES);
+		return error;
+	}
+
+	xfs_ilock(ip, XFS_ILOCK_EXCL);
+	xfs_trans_ijoin(tp, ip, 0);
+
+	xfs_bmap_init(&free_list, &first_block);
+	error = xfs_ifree(tp, ip, &free_list);
+	if (error) {
+		/*
+		 * If we fail to free the inode, shut down.  The cancel
+		 * might do that, we need to make sure.  Otherwise the
+		 * inode might be lost for a long time or forever.
+		 */
+		if (!XFS_FORCED_SHUTDOWN(mp)) {
+			xfs_notice(mp, "%s: xfs_ifree returned error %d",
+				__func__, error);
+			xfs_force_shutdown(mp, SHUTDOWN_META_IO_ERROR);
+		}
+		xfs_trans_cancel(tp, XFS_TRANS_RELEASE_LOG_RES|XFS_TRANS_ABORT);
+		xfs_iunlock(ip, XFS_ILOCK_EXCL);
+		return error;
+	}
+
+	/*
+	 * Credit the quota account(s). The inode is gone.
+	 */
+	xfs_trans_mod_dquot_byino(tp, ip, XFS_TRANS_DQ_ICOUNT, -1);
+
+	/*
+	 * Just ignore errors at this point.  There is nothing we can
+	 * do except to try to keep going. Make sure it's not a silent
+	 * error.
+	 */
+	error = xfs_bmap_finish(&tp,  &free_list, &committed);
+	if (error)
+		xfs_notice(mp, "%s: xfs_bmap_finish returned error %d",
+			__func__, error);
+	error = xfs_trans_commit(tp, XFS_TRANS_RELEASE_LOG_RES);
+	if (error)
+		xfs_notice(mp, "%s: xfs_trans_commit returned error %d",
+			__func__, error);
+
+	xfs_iunlock(ip, XFS_ILOCK_EXCL);
+	return 0;
+}
+
+/*
  * xfs_inactive
  *
  * This is called when the vnode reference count for the vnode
@@ -1670,16 +1789,11 @@ xfs_release(
  * now be truncated.  Also, we clear all of the read-ahead state
  * kept for the inode here since the file is now closed.
  */
-int
+void
 xfs_inactive(
 	xfs_inode_t	*ip)
 {
-	xfs_bmap_free_t		free_list;
-	xfs_fsblock_t		first_block;
-	int			committed;
-	struct xfs_trans	*tp;
 	struct xfs_mount	*mp;
-	struct xfs_trans_res	*resp;
 	int			error;
 	int			truncate = 0;
 
@@ -1687,19 +1801,17 @@ xfs_inactive(
 	 * If the inode is already free, then there can be nothing
 	 * to clean up here.
 	 */
-	if (ip->i_d.di_mode == 0 || is_bad_inode(VFS_I(ip))) {
+	if (ip->i_d.di_mode == 0) {
 		ASSERT(ip->i_df.if_real_bytes == 0);
 		ASSERT(ip->i_df.if_broot_bytes == 0);
-		return VN_INACTIVE_CACHE;
+		return;
 	}
 
 	mp = ip->i_mount;
 
-	error = 0;
-
 	/* If this is a read-only mount, don't do this (would generate I/O) */
 	if (mp->m_flags & XFS_MOUNT_RDONLY)
-		goto out;
+		return;
 
 	if (ip->i_d.di_nlink != 0) {
 		/*
@@ -1707,12 +1819,10 @@ xfs_inactive(
 		 * cache. Post-eof blocks must be freed, lest we end up with
 		 * broken free space accounting.
 		 */
-		if (xfs_can_free_eofblocks(ip, true)) {
-			error = xfs_free_eofblocks(mp, ip, false);
-			if (error)
-				return VN_INACTIVE_CACHE;
-		}
-		goto out;
+		if (xfs_can_free_eofblocks(ip, true))
+			xfs_free_eofblocks(mp, ip, false);
+
+		return;
 	}
 
 	if (S_ISREG(ip->i_d.di_mode) &&
@@ -1722,36 +1832,14 @@ xfs_inactive(
 
 	error = xfs_qm_dqattach(ip, 0);
 	if (error)
-		return VN_INACTIVE_CACHE;
+		return;
 
-	tp = xfs_trans_alloc(mp, XFS_TRANS_INACTIVE);
-	resp = (truncate || S_ISLNK(ip->i_d.di_mode)) ?
-		&M_RES(mp)->tr_itruncate : &M_RES(mp)->tr_ifree;
-
-	error = xfs_trans_reserve(tp, resp, 0, 0);
-	if (error) {
-		ASSERT(XFS_FORCED_SHUTDOWN(mp));
-		xfs_trans_cancel(tp, 0);
-		return VN_INACTIVE_CACHE;
-	}
-
-	xfs_ilock(ip, XFS_ILOCK_EXCL);
-	xfs_trans_ijoin(tp, ip, 0);
-
-	if (S_ISLNK(ip->i_d.di_mode)) {
-		error = xfs_inactive_symlink(ip, &tp);
-		if (error)
-			goto out_cancel;
-	} else if (truncate) {
-		ip->i_d.di_size = 0;
-		xfs_trans_log_inode(tp, ip, XFS_ILOG_CORE);
-
-		error = xfs_itruncate_extents(&tp, ip, XFS_DATA_FORK, 0);
-		if (error)
-			goto out_cancel;
-
-		ASSERT(ip->i_d.di_nextents == 0);
-	}
+	if (S_ISLNK(ip->i_d.di_mode))
+		error = xfs_inactive_symlink(ip);
+	else if (truncate)
+		error = xfs_inactive_truncate(ip);
+	if (error)
+		return;
 
 	/*
 	 * If there are attributes associated with the file then blow them away
@@ -1762,25 +1850,9 @@ xfs_inactive(
 	if (ip->i_d.di_anextents > 0) {
 		ASSERT(ip->i_d.di_forkoff != 0);
 
-		error = xfs_trans_commit(tp, XFS_TRANS_RELEASE_LOG_RES);
-		if (error)
-			goto out_unlock;
-
-		xfs_iunlock(ip, XFS_ILOCK_EXCL);
-
 		error = xfs_attr_inactive(ip);
 		if (error)
-			goto out;
-
-		tp = xfs_trans_alloc(mp, XFS_TRANS_INACTIVE);
-		error = xfs_trans_reserve(tp, &M_RES(mp)->tr_ifree, 0, 0);
-		if (error) {
-			xfs_trans_cancel(tp, 0);
-			goto out;
-		}
-
-		xfs_ilock(ip, XFS_ILOCK_EXCL);
-		xfs_trans_ijoin(tp, ip, 0);
+			return;
 	}
 
 	if (ip->i_afp)
@@ -1791,52 +1863,14 @@ xfs_inactive(
 	/*
 	 * Free the inode.
 	 */
-	xfs_bmap_init(&free_list, &first_block);
-	error = xfs_ifree(tp, ip, &free_list);
-	if (error) {
-		/*
-		 * If we fail to free the inode, shut down.  The cancel
-		 * might do that, we need to make sure.  Otherwise the
-		 * inode might be lost for a long time or forever.
-		 */
-		if (!XFS_FORCED_SHUTDOWN(mp)) {
-			xfs_notice(mp, "%s: xfs_ifree returned error %d",
-				__func__, error);
-			xfs_force_shutdown(mp, SHUTDOWN_META_IO_ERROR);
-		}
-		xfs_trans_cancel(tp, XFS_TRANS_RELEASE_LOG_RES|XFS_TRANS_ABORT);
-	} else {
-		/*
-		 * Credit the quota account(s). The inode is gone.
-		 */
-		xfs_trans_mod_dquot_byino(tp, ip, XFS_TRANS_DQ_ICOUNT, -1);
-
-		/*
-		 * Just ignore errors at this point.  There is nothing we can
-		 * do except to try to keep going. Make sure it's not a silent
-		 * error.
-		 */
-		error = xfs_bmap_finish(&tp,  &free_list, &committed);
-		if (error)
-			xfs_notice(mp, "%s: xfs_bmap_finish returned error %d",
-				__func__, error);
-		error = xfs_trans_commit(tp, XFS_TRANS_RELEASE_LOG_RES);
-		if (error)
-			xfs_notice(mp, "%s: xfs_trans_commit returned error %d",
-				__func__, error);
-	}
+	error = xfs_inactive_ifree(ip);
+	if (error)
+		return;
 
 	/*
 	 * Release the dquots held by inode, if any.
 	 */
 	xfs_qm_dqdetach(ip);
-out_unlock:
-	xfs_iunlock(ip, XFS_ILOCK_EXCL);
-out:
-	return VN_INACTIVE_CACHE;
-out_cancel:
-	xfs_trans_cancel(tp, XFS_TRANS_RELEASE_LOG_RES | XFS_TRANS_ABORT);
-	goto out_unlock;
 }
 
 /*
@@ -2370,6 +2404,33 @@ xfs_iunpin_wait(
 		__xfs_iunpin_wait(ip);
 }
 
+/*
+ * Removing an inode from the namespace involves removing the directory entry
+ * and dropping the link count on the inode. Removing the directory entry can
+ * result in locking an AGF (directory blocks were freed) and removing a link
+ * count can result in placing the inode on an unlinked list which results in
+ * locking an AGI.
+ *
+ * The big problem here is that we have an ordering constraint on AGF and AGI
+ * locking - inode allocation locks the AGI, then can allocate a new extent for
+ * new inodes, locking the AGF after the AGI. Similarly, freeing the inode
+ * removes the inode from the unlinked list, requiring that we lock the AGI
+ * first, and then freeing the inode can result in an inode chunk being freed
+ * and hence freeing disk space requiring that we lock an AGF.
+ *
+ * Hence the ordering that is imposed by other parts of the code is AGI before
+ * AGF. This means we cannot remove the directory entry before we drop the inode
+ * reference count and put it on the unlinked list as this results in a lock
+ * order of AGF then AGI, and this can deadlock against inode allocation and
+ * freeing. Therefore we must drop the link counts before we remove the
+ * directory entry.
+ *
+ * This is still safe from a transactional point of view - it is not until we
+ * get to xfs_bmap_finish() that we have the possibility of multiple
+ * transactions in this operation. Hence as long as we remove the directory
+ * entry and drop the link count in the first transaction of the remove
+ * operation, there are no transactional constraints on the ordering here.
+ */
 int
 xfs_remove(
 	xfs_inode_t             *dp,
@@ -2439,6 +2500,7 @@ xfs_remove(
 	/*
 	 * If we're removing a directory perform some additional validation.
 	 */
+	cancel_flags |= XFS_TRANS_ABORT;
 	if (is_dir) {
 		ASSERT(ip->i_d.di_nlink >= 2);
 		if (ip->i_d.di_nlink != 2) {
@@ -2449,31 +2511,16 @@ xfs_remove(
 			error = XFS_ERROR(ENOTEMPTY);
 			goto out_trans_cancel;
 		}
-	}
 
-	xfs_bmap_init(&free_list, &first_block);
-	error = xfs_dir_removename(tp, dp, name, ip->i_ino,
-					&first_block, &free_list, resblks);
-	if (error) {
-		ASSERT(error != ENOENT);
-		goto out_bmap_cancel;
-	}
-	xfs_trans_ichgtime(tp, dp, XFS_ICHGTIME_MOD | XFS_ICHGTIME_CHG);
-
-	if (is_dir) {
-		/*
-		 * Drop the link from ip's "..".
-		 */
+		/* Drop the link from ip's "..".  */
 		error = xfs_droplink(tp, dp);
 		if (error)
-			goto out_bmap_cancel;
+			goto out_trans_cancel;
 
-		/*
-		 * Drop the "." link from ip to self.
-		 */
+		/* Drop the "." link from ip to self.  */
 		error = xfs_droplink(tp, ip);
 		if (error)
-			goto out_bmap_cancel;
+			goto out_trans_cancel;
 	} else {
 		/*
 		 * When removing a non-directory we need to log the parent
@@ -2482,20 +2529,24 @@ xfs_remove(
 		 */
 		xfs_trans_log_inode(tp, dp, XFS_ILOG_CORE);
 	}
+	xfs_trans_ichgtime(tp, dp, XFS_ICHGTIME_MOD | XFS_ICHGTIME_CHG);
 
-	/*
-	 * Drop the link from dp to ip.
-	 */
+	/* Drop the link from dp to ip. */
 	error = xfs_droplink(tp, ip);
 	if (error)
-		goto out_bmap_cancel;
+		goto out_trans_cancel;
 
-	/*
-	 * Determine if this is the last link while
-	 * we are in the transaction.
-	 */
+	/* Determine if this is the last link while the inode is locked */
 	link_zero = (ip->i_d.di_nlink == 0);
 
+	xfs_bmap_init(&free_list, &first_block);
+	error = xfs_dir_removename(tp, dp, name, ip->i_ino,
+					&first_block, &free_list, resblks);
+	if (error) {
+		ASSERT(error != ENOENT);
+		goto out_bmap_cancel;
+	}
+
 	/*
 	 * If this is a synchronous mount, make sure that the
 	 * remove transaction goes to disk before returning to
@@ -2525,7 +2576,6 @@ xfs_remove(
 
  out_bmap_cancel:
 	xfs_bmap_cancel(&free_list);
-	cancel_flags |= XFS_TRANS_ABORT;
  out_trans_cancel:
 	xfs_trans_cancel(tp, cancel_flags);
  std_return:
diff --git a/fs/xfs/xfs_inode.h b/fs/xfs/xfs_inode.h
index 4a91358c1470..9e6efccbae04 100644
--- a/fs/xfs/xfs_inode.h
+++ b/fs/xfs/xfs_inode.h
@@ -24,7 +24,6 @@
 /*
  * Kernel only inode definitions
  */
-
 struct xfs_dinode;
 struct xfs_inode;
 struct xfs_buf;
@@ -50,6 +49,9 @@ typedef struct xfs_inode {
 	xfs_ifork_t		*i_afp;		/* attribute fork pointer */
 	xfs_ifork_t		i_df;		/* data fork */
 
+	/* operations vectors */
+	const struct xfs_dir_ops *d_ops;		/* directory ops vector */
+
 	/* Transaction and locking information. */
 	struct xfs_inode_log_item *i_itemp;	/* logging information */
 	mrlock_t		i_lock;		/* inode lock */
@@ -316,7 +318,7 @@ static inline int xfs_isiflocked(struct xfs_inode *ip)
 
 
 int		xfs_release(struct xfs_inode *ip);
-int		xfs_inactive(struct xfs_inode *ip);
+void		xfs_inactive(struct xfs_inode *ip);
 int		xfs_lookup(struct xfs_inode *dp, struct xfs_name *name,
 			   struct xfs_inode **ipp, struct xfs_name *ci_name);
 int		xfs_create(struct xfs_inode *dp, struct xfs_name *name,
diff --git a/fs/xfs/xfs_inode_buf.c b/fs/xfs/xfs_inode_buf.c
index 63382d37f565..4fc9f39dd89e 100644
--- a/fs/xfs/xfs_inode_buf.c
+++ b/fs/xfs/xfs_inode_buf.c
@@ -17,20 +17,20 @@
  */
 #include "xfs.h"
 #include "xfs_fs.h"
+#include "xfs_shared.h"
 #include "xfs_format.h"
-#include "xfs_log.h"
-#include "xfs_trans.h"
+#include "xfs_log_format.h"
+#include "xfs_trans_resv.h"
 #include "xfs_sb.h"
 #include "xfs_ag.h"
 #include "xfs_mount.h"
-#include "xfs_bmap_btree.h"
-#include "xfs_ialloc_btree.h"
-#include "xfs_dinode.h"
 #include "xfs_inode.h"
 #include "xfs_error.h"
 #include "xfs_cksum.h"
 #include "xfs_icache.h"
+#include "xfs_trans.h"
 #include "xfs_ialloc.h"
+#include "xfs_dinode.h"
 
 /*
  * Check that none of the inode's in the buffer have a next
diff --git a/fs/xfs/xfs_inode_buf.h b/fs/xfs/xfs_inode_buf.h
index abba0ae8cf2d..9308c47f2a52 100644
--- a/fs/xfs/xfs_inode_buf.h
+++ b/fs/xfs/xfs_inode_buf.h
@@ -47,7 +47,4 @@ void	xfs_inobp_check(struct xfs_mount *, struct xfs_buf *);
 #define	xfs_inobp_check(mp, bp)
 #endif /* DEBUG */
 
-extern const struct xfs_buf_ops xfs_inode_buf_ops;
-extern const struct xfs_buf_ops xfs_inode_buf_ra_ops;
-
 #endif	/* __XFS_INODE_BUF_H__ */
diff --git a/fs/xfs/xfs_inode_fork.c b/fs/xfs/xfs_inode_fork.c
index 02f1083955bb..cfee14a83cfe 100644
--- a/fs/xfs/xfs_inode_fork.c
+++ b/fs/xfs/xfs_inode_fork.c
@@ -20,31 +20,21 @@
 #include "xfs.h"
 #include "xfs_fs.h"
 #include "xfs_format.h"
-#include "xfs_log.h"
+#include "xfs_log_format.h"
+#include "xfs_trans_resv.h"
 #include "xfs_inum.h"
-#include "xfs_trans.h"
-#include "xfs_trans_priv.h"
 #include "xfs_sb.h"
 #include "xfs_ag.h"
 #include "xfs_mount.h"
-#include "xfs_bmap_btree.h"
-#include "xfs_alloc_btree.h"
-#include "xfs_ialloc_btree.h"
-#include "xfs_attr_sf.h"
-#include "xfs_dinode.h"
 #include "xfs_inode.h"
-#include "xfs_buf_item.h"
+#include "xfs_trans.h"
 #include "xfs_inode_item.h"
-#include "xfs_btree.h"
-#include "xfs_alloc.h"
-#include "xfs_ialloc.h"
+#include "xfs_bmap_btree.h"
 #include "xfs_bmap.h"
 #include "xfs_error.h"
-#include "xfs_quota.h"
-#include "xfs_filestream.h"
-#include "xfs_cksum.h"
 #include "xfs_trace.h"
-#include "xfs_icache.h"
+#include "xfs_attr_sf.h"
+#include "xfs_dinode.h"
 
 kmem_zone_t *xfs_ifork_zone;
 
@@ -1031,15 +1021,14 @@ xfs_iext_add(
 		 * the next index needed in the indirection array.
 		 */
 		else {
-			int	count = ext_diff;
+			uint	count = ext_diff;
 
 			while (count) {
 				erp = xfs_iext_irec_new(ifp, erp_idx);
-				erp->er_extcount = count;
-				count -= MIN(count, (int)XFS_LINEAR_EXTS);
-				if (count) {
+				erp->er_extcount = min(count, XFS_LINEAR_EXTS);
+				count -= erp->er_extcount;
+				if (count)
 					erp_idx++;
-				}
 			}
 		}
 	}
@@ -1359,7 +1348,7 @@ xfs_iext_remove_indirect(
 void
 xfs_iext_realloc_direct(
 	xfs_ifork_t	*ifp,		/* inode fork pointer */
-	int		new_size)	/* new size of extents */
+	int		new_size)	/* new size of extents after adding */
 {
 	int		rnew_size;	/* real new size of extents */
 
@@ -1397,13 +1386,8 @@ xfs_iext_realloc_direct(
 				rnew_size - ifp->if_real_bytes);
 		}
 	}
-	/*
-	 * Switch from the inline extent buffer to a direct
-	 * extent list. Be sure to include the inline extent
-	 * bytes in new_size.
-	 */
+	/* Switch from the inline extent buffer to a direct extent list */
 	else {
-		new_size += ifp->if_bytes;
 		if (!is_power_of_2(new_size)) {
 			rnew_size = roundup_pow_of_two(new_size);
 		}
diff --git a/fs/xfs/xfs_inode_fork.h b/fs/xfs/xfs_inode_fork.h
index 28661a0d9058..eb329a1ea888 100644
--- a/fs/xfs/xfs_inode_fork.h
+++ b/fs/xfs/xfs_inode_fork.h
@@ -19,6 +19,7 @@
 #define	__XFS_INODE_FORK_H__
 
 struct xfs_inode_log_item;
+struct xfs_dinode;
 
 /*
  * The following xfs_ext_irec_t struct introduces a second (top) level
diff --git a/fs/xfs/xfs_inode_item.c b/fs/xfs/xfs_inode_item.c
index 378081109844..7c0d391f9a6e 100644
--- a/fs/xfs/xfs_inode_item.c
+++ b/fs/xfs/xfs_inode_item.c
@@ -17,19 +17,19 @@
  */
 #include "xfs.h"
 #include "xfs_fs.h"
-#include "xfs_types.h"
-#include "xfs_log.h"
-#include "xfs_trans.h"
+#include "xfs_format.h"
+#include "xfs_log_format.h"
+#include "xfs_trans_resv.h"
 #include "xfs_sb.h"
 #include "xfs_ag.h"
 #include "xfs_mount.h"
-#include "xfs_trans_priv.h"
-#include "xfs_bmap_btree.h"
-#include "xfs_dinode.h"
 #include "xfs_inode.h"
+#include "xfs_trans.h"
 #include "xfs_inode_item.h"
 #include "xfs_error.h"
 #include "xfs_trace.h"
+#include "xfs_trans_priv.h"
+#include "xfs_dinode.h"
 
 
 kmem_zone_t	*xfs_ili_zone;		/* inode log item zone */
diff --git a/fs/xfs/xfs_ioctl.c b/fs/xfs/xfs_ioctl.c
index 668e8f4ccf5e..4d613401a5e0 100644
--- a/fs/xfs/xfs_ioctl.c
+++ b/fs/xfs/xfs_ioctl.c
@@ -17,32 +17,31 @@
  */
 #include "xfs.h"
 #include "xfs_fs.h"
+#include "xfs_shared.h"
 #include "xfs_format.h"
-#include "xfs_log.h"
-#include "xfs_trans.h"
+#include "xfs_log_format.h"
+#include "xfs_trans_resv.h"
 #include "xfs_sb.h"
 #include "xfs_ag.h"
-#include "xfs_alloc.h"
 #include "xfs_mount.h"
-#include "xfs_bmap_btree.h"
-#include "xfs_dinode.h"
 #include "xfs_inode.h"
 #include "xfs_ioctl.h"
+#include "xfs_alloc.h"
 #include "xfs_rtalloc.h"
 #include "xfs_itable.h"
 #include "xfs_error.h"
 #include "xfs_attr.h"
 #include "xfs_bmap.h"
 #include "xfs_bmap_util.h"
-#include "xfs_buf_item.h"
 #include "xfs_fsops.h"
 #include "xfs_discard.h"
 #include "xfs_quota.h"
-#include "xfs_inode_item.h"
 #include "xfs_export.h"
 #include "xfs_trace.h"
 #include "xfs_icache.h"
 #include "xfs_symlink.h"
+#include "xfs_dinode.h"
+#include "xfs_trans.h"
 
 #include <linux/capability.h>
 #include <linux/dcache.h>
@@ -641,7 +640,11 @@ xfs_ioc_space(
 	unsigned int		cmd,
 	xfs_flock64_t		*bf)
 {
-	int			attr_flags = 0;
+	struct xfs_mount	*mp = ip->i_mount;
+	struct xfs_trans	*tp;
+	struct iattr		iattr;
+	bool			setprealloc = false;
+	bool			clrprealloc = false;
 	int			error;
 
 	/*
@@ -661,19 +664,128 @@ xfs_ioc_space(
 	if (!S_ISREG(inode->i_mode))
 		return -XFS_ERROR(EINVAL);
 
-	if (filp->f_flags & (O_NDELAY|O_NONBLOCK))
-		attr_flags |= XFS_ATTR_NONBLOCK;
+	error = mnt_want_write_file(filp);
+	if (error)
+		return error;
 
-	if (filp->f_flags & O_DSYNC)
-		attr_flags |= XFS_ATTR_SYNC;
+	xfs_ilock(ip, XFS_IOLOCK_EXCL);
+
+	switch (bf->l_whence) {
+	case 0: /*SEEK_SET*/
+		break;
+	case 1: /*SEEK_CUR*/
+		bf->l_start += filp->f_pos;
+		break;
+	case 2: /*SEEK_END*/
+		bf->l_start += XFS_ISIZE(ip);
+		break;
+	default:
+		error = XFS_ERROR(EINVAL);
+		goto out_unlock;
+	}
 
-	if (ioflags & IO_INVIS)
-		attr_flags |= XFS_ATTR_DMI;
+	/*
+	 * length of <= 0 for resv/unresv/zero is invalid.  length for
+	 * alloc/free is ignored completely and we have no idea what userspace
+	 * might have set it to, so set it to zero to allow range
+	 * checks to pass.
+	 */
+	switch (cmd) {
+	case XFS_IOC_ZERO_RANGE:
+	case XFS_IOC_RESVSP:
+	case XFS_IOC_RESVSP64:
+	case XFS_IOC_UNRESVSP:
+	case XFS_IOC_UNRESVSP64:
+		if (bf->l_len <= 0) {
+			error = XFS_ERROR(EINVAL);
+			goto out_unlock;
+		}
+		break;
+	default:
+		bf->l_len = 0;
+		break;
+	}
+
+	if (bf->l_start < 0 ||
+	    bf->l_start > mp->m_super->s_maxbytes ||
+	    bf->l_start + bf->l_len < 0 ||
+	    bf->l_start + bf->l_len >= mp->m_super->s_maxbytes) {
+		error = XFS_ERROR(EINVAL);
+		goto out_unlock;
+	}
+
+	switch (cmd) {
+	case XFS_IOC_ZERO_RANGE:
+		error = xfs_zero_file_space(ip, bf->l_start, bf->l_len);
+		if (!error)
+			setprealloc = true;
+		break;
+	case XFS_IOC_RESVSP:
+	case XFS_IOC_RESVSP64:
+		error = xfs_alloc_file_space(ip, bf->l_start, bf->l_len,
+						XFS_BMAPI_PREALLOC);
+		if (!error)
+			setprealloc = true;
+		break;
+	case XFS_IOC_UNRESVSP:
+	case XFS_IOC_UNRESVSP64:
+		error = xfs_free_file_space(ip, bf->l_start, bf->l_len);
+		break;
+	case XFS_IOC_ALLOCSP:
+	case XFS_IOC_ALLOCSP64:
+	case XFS_IOC_FREESP:
+	case XFS_IOC_FREESP64:
+		if (bf->l_start > XFS_ISIZE(ip)) {
+			error = xfs_alloc_file_space(ip, XFS_ISIZE(ip),
+					bf->l_start - XFS_ISIZE(ip), 0);
+			if (error)
+				goto out_unlock;
+		}
+
+		iattr.ia_valid = ATTR_SIZE;
+		iattr.ia_size = bf->l_start;
+
+		error = xfs_setattr_size(ip, &iattr);
+		if (!error)
+			clrprealloc = true;
+		break;
+	default:
+		ASSERT(0);
+		error = XFS_ERROR(EINVAL);
+	}
 
-	error = mnt_want_write_file(filp);
 	if (error)
-		return error;
-	error = xfs_change_file_space(ip, cmd, bf, filp->f_pos, attr_flags);
+		goto out_unlock;
+
+	tp = xfs_trans_alloc(mp, XFS_TRANS_WRITEID);
+	error = xfs_trans_reserve(tp, &M_RES(mp)->tr_writeid, 0, 0);
+	if (error) {
+		xfs_trans_cancel(tp, 0);
+		goto out_unlock;
+	}
+
+	xfs_ilock(ip, XFS_ILOCK_EXCL);
+	xfs_trans_ijoin(tp, ip, XFS_ILOCK_EXCL);
+
+	if (!(ioflags & IO_INVIS)) {
+		ip->i_d.di_mode &= ~S_ISUID;
+		if (ip->i_d.di_mode & S_IXGRP)
+			ip->i_d.di_mode &= ~S_ISGID;
+		xfs_trans_ichgtime(tp, ip, XFS_ICHGTIME_MOD | XFS_ICHGTIME_CHG);
+	}
+
+	if (setprealloc)
+		ip->i_d.di_flags |= XFS_DIFLAG_PREALLOC;
+	else if (clrprealloc)
+		ip->i_d.di_flags &= ~XFS_DIFLAG_PREALLOC;
+
+	xfs_trans_log_inode(tp, ip, XFS_ILOG_CORE);
+	if (filp->f_flags & O_DSYNC)
+		xfs_trans_set_sync(tp);
+	error = xfs_trans_commit(tp, 0);
+
+out_unlock:
+	xfs_iunlock(ip, XFS_IOLOCK_EXCL);
 	mnt_drop_write_file(filp);
 	return -error;
 }
diff --git a/fs/xfs/xfs_ioctl32.c b/fs/xfs/xfs_ioctl32.c
index f671f7e472ac..e8fb1231db81 100644
--- a/fs/xfs/xfs_ioctl32.c
+++ b/fs/xfs/xfs_ioctl32.c
@@ -22,14 +22,13 @@
 #include <asm/uaccess.h>
 #include "xfs.h"
 #include "xfs_fs.h"
-#include "xfs_log.h"
-#include "xfs_trans.h"
+#include "xfs_format.h"
+#include "xfs_log_format.h"
+#include "xfs_trans_resv.h"
 #include "xfs_sb.h"
 #include "xfs_ag.h"
 #include "xfs_mount.h"
-#include "xfs_bmap_btree.h"
 #include "xfs_vnode.h"
-#include "xfs_dinode.h"
 #include "xfs_inode.h"
 #include "xfs_itable.h"
 #include "xfs_error.h"
diff --git a/fs/xfs/xfs_iomap.c b/fs/xfs/xfs_iomap.c
index 8d4d49b6fbf3..22d1cbea283d 100644
--- a/fs/xfs/xfs_iomap.c
+++ b/fs/xfs/xfs_iomap.c
@@ -17,34 +17,28 @@
  */
 #include "xfs.h"
 #include "xfs_fs.h"
+#include "xfs_shared.h"
 #include "xfs_format.h"
-#include "xfs_log.h"
-#include "xfs_trans.h"
+#include "xfs_log_format.h"
+#include "xfs_trans_resv.h"
 #include "xfs_sb.h"
 #include "xfs_ag.h"
-#include "xfs_alloc.h"
-#include "xfs_quota.h"
 #include "xfs_mount.h"
-#include "xfs_bmap_btree.h"
-#include "xfs_alloc_btree.h"
-#include "xfs_ialloc_btree.h"
-#include "xfs_dinode.h"
 #include "xfs_inode.h"
-#include "xfs_inode_item.h"
 #include "xfs_btree.h"
+#include "xfs_bmap_btree.h"
 #include "xfs_bmap.h"
 #include "xfs_bmap_util.h"
-#include "xfs_rtalloc.h"
 #include "xfs_error.h"
-#include "xfs_itable.h"
-#include "xfs_attr.h"
-#include "xfs_buf_item.h"
+#include "xfs_trans.h"
 #include "xfs_trans_space.h"
 #include "xfs_iomap.h"
 #include "xfs_trace.h"
 #include "xfs_icache.h"
+#include "xfs_quota.h"
 #include "xfs_dquot_item.h"
 #include "xfs_dquot.h"
+#include "xfs_dinode.h"
 
 
 #define XFS_WRITEIO_ALIGN(mp,off)	(((off) >> mp->m_writeio_log) \
@@ -110,7 +104,7 @@ xfs_alert_fsblock_zero(
 	xfs_alert_tag(ip->i_mount, XFS_PTAG_FSBLOCK_ZERO,
 			"Access to block zero in inode %llu "
 			"start_block: %llx start_off: %llx "
-			"blkcnt: %llx extent-state: %x\n",
+			"blkcnt: %llx extent-state: %x",
 		(unsigned long long)ip->i_ino,
 		(unsigned long long)imap->br_startblock,
 		(unsigned long long)imap->br_startoff,
@@ -655,7 +649,6 @@ int
 xfs_iomap_write_allocate(
 	xfs_inode_t	*ip,
 	xfs_off_t	offset,
-	size_t		count,
 	xfs_bmbt_irec_t *imap)
 {
 	xfs_mount_t	*mp = ip->i_mount;
diff --git a/fs/xfs/xfs_iomap.h b/fs/xfs/xfs_iomap.h
index 80615760959a..411fbb8919ef 100644
--- a/fs/xfs/xfs_iomap.h
+++ b/fs/xfs/xfs_iomap.h
@@ -21,12 +21,12 @@
 struct xfs_inode;
 struct xfs_bmbt_irec;
 
-extern int xfs_iomap_write_direct(struct xfs_inode *, xfs_off_t, size_t,
+int xfs_iomap_write_direct(struct xfs_inode *, xfs_off_t, size_t,
 			struct xfs_bmbt_irec *, int);
-extern int xfs_iomap_write_delay(struct xfs_inode *, xfs_off_t, size_t,
+int xfs_iomap_write_delay(struct xfs_inode *, xfs_off_t, size_t,
 			struct xfs_bmbt_irec *);
-extern int xfs_iomap_write_allocate(struct xfs_inode *, xfs_off_t, size_t,
+int xfs_iomap_write_allocate(struct xfs_inode *, xfs_off_t,
 			struct xfs_bmbt_irec *);
-extern int xfs_iomap_write_unwritten(struct xfs_inode *, xfs_off_t, size_t);
+int xfs_iomap_write_unwritten(struct xfs_inode *, xfs_off_t, size_t);
 
 #endif /* __XFS_IOMAP_H__*/
diff --git a/fs/xfs/xfs_iops.c b/fs/xfs/xfs_iops.c
index 2b8952d9199b..27e0e544e963 100644
--- a/fs/xfs/xfs_iops.c
+++ b/fs/xfs/xfs_iops.c
@@ -17,32 +17,28 @@
  */
 #include "xfs.h"
 #include "xfs_fs.h"
+#include "xfs_shared.h"
 #include "xfs_format.h"
-#include "xfs_acl.h"
-#include "xfs_log.h"
-#include "xfs_trans.h"
+#include "xfs_log_format.h"
+#include "xfs_trans_resv.h"
 #include "xfs_sb.h"
 #include "xfs_ag.h"
-#include "xfs_alloc.h"
-#include "xfs_quota.h"
 #include "xfs_mount.h"
-#include "xfs_bmap_btree.h"
-#include "xfs_dinode.h"
+#include "xfs_da_format.h"
 #include "xfs_inode.h"
 #include "xfs_bmap.h"
 #include "xfs_bmap_util.h"
-#include "xfs_rtalloc.h"
+#include "xfs_acl.h"
+#include "xfs_quota.h"
 #include "xfs_error.h"
-#include "xfs_itable.h"
 #include "xfs_attr.h"
-#include "xfs_buf_item.h"
-#include "xfs_inode_item.h"
+#include "xfs_trans.h"
 #include "xfs_trace.h"
 #include "xfs_icache.h"
 #include "xfs_symlink.h"
 #include "xfs_da_btree.h"
-#include "xfs_dir2_format.h"
 #include "xfs_dir2_priv.h"
+#include "xfs_dinode.h"
 
 #include <linux/capability.h>
 #include <linux/xattr.h>
@@ -709,8 +705,7 @@ out_dqrele:
 int
 xfs_setattr_size(
 	struct xfs_inode	*ip,
-	struct iattr		*iattr,
-	int			flags)
+	struct iattr		*iattr)
 {
 	struct xfs_mount	*mp = ip->i_mount;
 	struct inode		*inode = VFS_I(ip);
@@ -733,15 +728,11 @@ xfs_setattr_size(
 	if (error)
 		return XFS_ERROR(error);
 
+	ASSERT(xfs_isilocked(ip, XFS_IOLOCK_EXCL));
 	ASSERT(S_ISREG(ip->i_d.di_mode));
 	ASSERT((mask & (ATTR_UID|ATTR_GID|ATTR_ATIME|ATTR_ATIME_SET|
 			ATTR_MTIME_SET|ATTR_KILL_PRIV|ATTR_TIMES_SET)) == 0);
 
-	if (!(flags & XFS_ATTR_NOLOCK)) {
-		lock_flags |= XFS_IOLOCK_EXCL;
-		xfs_ilock(ip, lock_flags);
-	}
-
 	oldsize = inode->i_size;
 	newsize = iattr->ia_size;
 
@@ -750,12 +741,11 @@ xfs_setattr_size(
 	 */
 	if (newsize == 0 && oldsize == 0 && ip->i_d.di_nextents == 0) {
 		if (!(mask & (ATTR_CTIME|ATTR_MTIME)))
-			goto out_unlock;
+			return 0;
 
 		/*
 		 * Use the regular setattr path to update the timestamps.
 		 */
-		xfs_iunlock(ip, lock_flags);
 		iattr->ia_valid &= ~ATTR_SIZE;
 		return xfs_setattr_nonsize(ip, iattr, 0);
 	}
@@ -765,7 +755,7 @@ xfs_setattr_size(
 	 */
 	error = xfs_qm_dqattach(ip, 0);
 	if (error)
-		goto out_unlock;
+		return error;
 
 	/*
 	 * Now we can make the changes.  Before we join the inode to the
@@ -783,7 +773,7 @@ xfs_setattr_size(
 		 */
 		error = xfs_zero_eof(ip, newsize, oldsize);
 		if (error)
-			goto out_unlock;
+			return error;
 	}
 
 	/*
@@ -802,7 +792,7 @@ xfs_setattr_size(
 		error = -filemap_write_and_wait_range(VFS_I(ip)->i_mapping,
 						      ip->i_d.di_size, newsize);
 		if (error)
-			goto out_unlock;
+			return error;
 	}
 
 	/*
@@ -812,7 +802,7 @@ xfs_setattr_size(
 
 	error = -block_truncate_page(inode->i_mapping, newsize, xfs_get_blocks);
 	if (error)
-		goto out_unlock;
+		return error;
 
 	tp = xfs_trans_alloc(mp, XFS_TRANS_SETATTR_SIZE);
 	error = xfs_trans_reserve(tp, &M_RES(mp)->tr_itruncate, 0, 0);
@@ -916,12 +906,21 @@ out_trans_cancel:
 
 STATIC int
 xfs_vn_setattr(
-	struct dentry	*dentry,
-	struct iattr	*iattr)
+	struct dentry		*dentry,
+	struct iattr		*iattr)
 {
-	if (iattr->ia_valid & ATTR_SIZE)
-		return -xfs_setattr_size(XFS_I(dentry->d_inode), iattr, 0);
-	return -xfs_setattr_nonsize(XFS_I(dentry->d_inode), iattr, 0);
+	struct xfs_inode	*ip = XFS_I(dentry->d_inode);
+	int			error;
+
+	if (iattr->ia_valid & ATTR_SIZE) {
+		xfs_ilock(ip, XFS_IOLOCK_EXCL);
+		error = xfs_setattr_size(ip, iattr);
+		xfs_iunlock(ip, XFS_IOLOCK_EXCL);
+	} else {
+		error = xfs_setattr_nonsize(ip, iattr, 0);
+	}
+
+	return -error;
 }
 
 STATIC int
@@ -1169,6 +1168,7 @@ xfs_setup_inode(
 	struct xfs_inode	*ip)
 {
 	struct inode		*inode = &ip->i_vnode;
+	gfp_t			gfp_mask;
 
 	inode->i_ino = ip->i_ino;
 	inode->i_state = I_NEW;
@@ -1204,6 +1204,7 @@ xfs_setup_inode(
 	inode->i_ctime.tv_nsec	= ip->i_d.di_ctime.t_nsec;
 	xfs_diflags_to_iflags(inode, ip);
 
+	ip->d_ops = ip->i_mount->m_nondir_inode_ops;
 	switch (inode->i_mode & S_IFMT) {
 	case S_IFREG:
 		inode->i_op = &xfs_inode_operations;
@@ -1216,6 +1217,7 @@ xfs_setup_inode(
 		else
 			inode->i_op = &xfs_dir_inode_operations;
 		inode->i_fop = &xfs_dir_file_operations;
+		ip->d_ops = ip->i_mount->m_dir_inode_ops;
 		break;
 	case S_IFLNK:
 		inode->i_op = &xfs_symlink_inode_operations;
@@ -1229,6 +1231,14 @@ xfs_setup_inode(
 	}
 
 	/*
+	 * Ensure all page cache allocations are done from GFP_NOFS context to
+	 * prevent direct reclaim recursion back into the filesystem and blowing
+	 * stacks or deadlocking.
+	 */
+	gfp_mask = mapping_gfp_mask(inode->i_mapping);
+	mapping_set_gfp_mask(inode->i_mapping, (gfp_mask & ~(__GFP_FS)));
+
+	/*
 	 * If there is no attribute fork no ACL can exist on this inode,
 	 * and it can't have any file capabilities attached to it either.
 	 */
diff --git a/fs/xfs/xfs_iops.h b/fs/xfs/xfs_iops.h
index d81fb41205ec..d2c5057b5cc4 100644
--- a/fs/xfs/xfs_iops.h
+++ b/fs/xfs/xfs_iops.h
@@ -30,14 +30,10 @@ extern void xfs_setup_inode(struct xfs_inode *);
 /*
  * Internal setattr interfaces.
  */
-#define	XFS_ATTR_DMI		0x01	/* invocation from a DMI function */
-#define	XFS_ATTR_NONBLOCK	0x02	/* return EAGAIN if op would block */
-#define XFS_ATTR_NOLOCK		0x04	/* Don't grab any conflicting locks */
-#define XFS_ATTR_NOACL		0x08	/* Don't call xfs_acl_chmod */
-#define XFS_ATTR_SYNC		0x10	/* synchronous operation required */
+#define XFS_ATTR_NOACL		0x01	/* Don't call xfs_acl_chmod */
 
 extern int xfs_setattr_nonsize(struct xfs_inode *ip, struct iattr *vap,
 			       int flags);
-extern int xfs_setattr_size(struct xfs_inode *ip, struct iattr *vap, int flags);
+extern int xfs_setattr_size(struct xfs_inode *ip, struct iattr *vap);
 
 #endif /* __XFS_IOPS_H__ */
diff --git a/fs/xfs/xfs_itable.c b/fs/xfs/xfs_itable.c
index 084b3e1741fd..c237ad15d500 100644
--- a/fs/xfs/xfs_itable.c
+++ b/fs/xfs/xfs_itable.c
@@ -17,24 +17,23 @@
  */
 #include "xfs.h"
 #include "xfs_fs.h"
-#include "xfs_types.h"
-#include "xfs_log.h"
+#include "xfs_shared.h"
+#include "xfs_format.h"
+#include "xfs_log_format.h"
+#include "xfs_trans_resv.h"
 #include "xfs_inum.h"
-#include "xfs_trans.h"
 #include "xfs_sb.h"
 #include "xfs_ag.h"
 #include "xfs_mount.h"
-#include "xfs_bmap_btree.h"
-#include "xfs_alloc_btree.h"
-#include "xfs_ialloc_btree.h"
-#include "xfs_dinode.h"
 #include "xfs_inode.h"
+#include "xfs_btree.h"
 #include "xfs_ialloc.h"
+#include "xfs_ialloc_btree.h"
 #include "xfs_itable.h"
 #include "xfs_error.h"
-#include "xfs_btree.h"
 #include "xfs_trace.h"
 #include "xfs_icache.h"
+#include "xfs_dinode.h"
 
 STATIC int
 xfs_internal_inum(
diff --git a/fs/xfs/xfs_log.c b/fs/xfs/xfs_log.c
index a2dea108071a..8497a00e399d 100644
--- a/fs/xfs/xfs_log.c
+++ b/fs/xfs/xfs_log.c
@@ -17,21 +17,19 @@
  */
 #include "xfs.h"
 #include "xfs_fs.h"
-#include "xfs_types.h"
-#include "xfs_log.h"
-#include "xfs_trans.h"
+#include "xfs_shared.h"
+#include "xfs_format.h"
+#include "xfs_log_format.h"
+#include "xfs_trans_resv.h"
 #include "xfs_sb.h"
 #include "xfs_ag.h"
 #include "xfs_mount.h"
 #include "xfs_error.h"
+#include "xfs_trans.h"
+#include "xfs_trans_priv.h"
+#include "xfs_log.h"
 #include "xfs_log_priv.h"
-#include "xfs_buf_item.h"
-#include "xfs_bmap_btree.h"
-#include "xfs_alloc_btree.h"
-#include "xfs_ialloc_btree.h"
 #include "xfs_log_recover.h"
-#include "xfs_trans_priv.h"
-#include "xfs_dinode.h"
 #include "xfs_inode.h"
 #include "xfs_trace.h"
 #include "xfs_fsops.h"
@@ -1000,27 +998,34 @@ xfs_log_space_wake(
 }
 
 /*
- * Determine if we have a transaction that has gone to disk
- * that needs to be covered. To begin the transition to the idle state
- * firstly the log needs to be idle (no AIL and nothing in the iclogs).
- * If we are then in a state where covering is needed, the caller is informed
- * that dummy transactions are required to move the log into the idle state.
+ * Determine if we have a transaction that has gone to disk that needs to be
+ * covered. To begin the transition to the idle state firstly the log needs to
+ * be idle. That means the CIL, the AIL and the iclogs needs to be empty before
+ * we start attempting to cover the log.
  *
- * Because this is called as part of the sync process, we should also indicate
- * that dummy transactions should be issued in anything but the covered or
- * idle states. This ensures that the log tail is accurately reflected in
- * the log at the end of the sync, hence if a crash occurrs avoids replay
- * of transactions where the metadata is already on disk.
+ * Only if we are then in a state where covering is needed, the caller is
+ * informed that dummy transactions are required to move the log into the idle
+ * state.
+ *
+ * If there are any items in the AIl or CIL, then we do not want to attempt to
+ * cover the log as we may be in a situation where there isn't log space
+ * available to run a dummy transaction and this can lead to deadlocks when the
+ * tail of the log is pinned by an item that is modified in the CIL.  Hence
+ * there's no point in running a dummy transaction at this point because we
+ * can't start trying to idle the log until both the CIL and AIL are empty.
  */
 int
 xfs_log_need_covered(xfs_mount_t *mp)
 {
-	int		needed = 0;
 	struct xlog	*log = mp->m_log;
+	int		needed = 0;
 
 	if (!xfs_fs_writable(mp))
 		return 0;
 
+	if (!xlog_cil_empty(log))
+		return 0;
+
 	spin_lock(&log->l_icloglock);
 	switch (log->l_covered_state) {
 	case XLOG_STATE_COVER_DONE:
@@ -1029,14 +1034,17 @@ xfs_log_need_covered(xfs_mount_t *mp)
 		break;
 	case XLOG_STATE_COVER_NEED:
 	case XLOG_STATE_COVER_NEED2:
-		if (!xfs_ail_min_lsn(log->l_ailp) &&
-		    xlog_iclogs_empty(log)) {
-			if (log->l_covered_state == XLOG_STATE_COVER_NEED)
-				log->l_covered_state = XLOG_STATE_COVER_DONE;
-			else
-				log->l_covered_state = XLOG_STATE_COVER_DONE2;
-		}
-		/* FALLTHRU */
+		if (xfs_ail_min_lsn(log->l_ailp))
+			break;
+		if (!xlog_iclogs_empty(log))
+			break;
+
+		needed = 1;
+		if (log->l_covered_state == XLOG_STATE_COVER_NEED)
+			log->l_covered_state = XLOG_STATE_COVER_DONE;
+		else
+			log->l_covered_state = XLOG_STATE_COVER_DONE2;
+		break;
 	default:
 		needed = 1;
 		break;
@@ -1068,6 +1076,7 @@ xlog_assign_tail_lsn_locked(
 		tail_lsn = lip->li_lsn;
 	else
 		tail_lsn = atomic64_read(&log->l_last_sync_lsn);
+	trace_xfs_log_assign_tail_lsn(log, tail_lsn);
 	atomic64_set(&log->l_tail_lsn, tail_lsn);
 	return tail_lsn;
 }
@@ -1979,7 +1988,7 @@ xlog_print_tic_res(
 
 	for (i = 0; i < ticket->t_res_num; i++) {
 		uint r_type = ticket->t_res_arr[i].r_type;
-		xfs_warn(mp, "region[%u]: %s - %u bytes\n", i,
+		xfs_warn(mp, "region[%u]: %s - %u bytes", i,
 			    ((r_type <= 0 || r_type > XLOG_REG_TYPE_MAX) ?
 			    "bad-rtype" : res_type_str[r_type-1]),
 			    ticket->t_res_arr[i].r_len);
@@ -3702,11 +3711,9 @@ xlog_verify_iclog(
 	/* check validity of iclog pointers */
 	spin_lock(&log->l_icloglock);
 	icptr = log->l_iclog;
-	for (i=0; i < log->l_iclog_bufs; i++) {
-		if (icptr == NULL)
-			xfs_emerg(log->l_mp, "%s: invalid ptr", __func__);
-		icptr = icptr->ic_next;
-	}
+	for (i = 0; i < log->l_iclog_bufs; i++, icptr = icptr->ic_next)
+		ASSERT(icptr);
+
 	if (icptr != log->l_iclog)
 		xfs_emerg(log->l_mp, "%s: corrupt iclog ring", __func__);
 	spin_unlock(&log->l_icloglock);
diff --git a/fs/xfs/xfs_log.h b/fs/xfs/xfs_log.h
index 1c458487f000..e148719e0a5d 100644
--- a/fs/xfs/xfs_log.h
+++ b/fs/xfs/xfs_log.h
@@ -18,8 +18,6 @@
 #ifndef	__XFS_LOG_H__
 #define __XFS_LOG_H__
 
-#include "xfs_log_format.h"
-
 struct xfs_log_vec {
 	struct xfs_log_vec	*lv_next;	/* next lv in build list */
 	int			lv_niovecs;	/* number of iovecs in lv */
@@ -82,11 +80,7 @@ struct xlog_ticket;
 struct xfs_log_item;
 struct xfs_item_ops;
 struct xfs_trans;
-
-void	xfs_log_item_init(struct xfs_mount	*mp,
-			struct xfs_log_item	*item,
-			int			type,
-			const struct xfs_item_ops *ops);
+struct xfs_log_callback;
 
 xfs_lsn_t xfs_log_done(struct xfs_mount *mp,
 		       struct xlog_ticket *ticket,
@@ -114,7 +108,7 @@ xfs_lsn_t xlog_assign_tail_lsn_locked(struct xfs_mount *mp);
 void	  xfs_log_space_wake(struct xfs_mount *mp);
 int	  xfs_log_notify(struct xfs_mount	*mp,
 			 struct xlog_in_core	*iclog,
-			 xfs_log_callback_t	*callback_entry);
+			 struct xfs_log_callback *callback_entry);
 int	  xfs_log_release_iclog(struct xfs_mount *mp,
 			 struct xlog_in_core	 *iclog);
 int	  xfs_log_reserve(struct xfs_mount *mp,
diff --git a/fs/xfs/xfs_log_cil.c b/fs/xfs/xfs_log_cil.c
index cfe97973ba36..5eb51fc5eb84 100644
--- a/fs/xfs/xfs_log_cil.c
+++ b/fs/xfs/xfs_log_cil.c
@@ -17,11 +17,9 @@
 
 #include "xfs.h"
 #include "xfs_fs.h"
-#include "xfs_types.h"
-#include "xfs_log.h"
-#include "xfs_trans.h"
-#include "xfs_trans_priv.h"
-#include "xfs_log_priv.h"
+#include "xfs_log_format.h"
+#include "xfs_shared.h"
+#include "xfs_trans_resv.h"
 #include "xfs_sb.h"
 #include "xfs_ag.h"
 #include "xfs_mount.h"
@@ -29,6 +27,10 @@
 #include "xfs_alloc.h"
 #include "xfs_extent_busy.h"
 #include "xfs_discard.h"
+#include "xfs_trans.h"
+#include "xfs_trans_priv.h"
+#include "xfs_log.h"
+#include "xfs_log_priv.h"
 
 /*
  * Allocate a new ticket. Failing to get a new ticket makes it really hard to
@@ -711,6 +713,20 @@ xlog_cil_push_foreground(
 	xlog_cil_push(log);
 }
 
+bool
+xlog_cil_empty(
+	struct xlog	*log)
+{
+	struct xfs_cil	*cil = log->l_cilp;
+	bool		empty = false;
+
+	spin_lock(&cil->xc_push_lock);
+	if (list_empty(&cil->xc_cil))
+		empty = true;
+	spin_unlock(&cil->xc_push_lock);
+	return empty;
+}
+
 /*
  * Commit a transaction with the given vector to the Committed Item List.
  *
diff --git a/fs/xfs/xfs_log_format.h b/fs/xfs/xfs_log_format.h
index ca7e28a8ed31..f0969c77bdbe 100644
--- a/fs/xfs/xfs_log_format.h
+++ b/fs/xfs/xfs_log_format.h
@@ -234,178 +234,6 @@ typedef struct xfs_trans_header {
 	{ XFS_LI_ICREATE,	"XFS_LI_ICREATE" }
 
 /*
- * Transaction types.  Used to distinguish types of buffers.
- */
-#define XFS_TRANS_SETATTR_NOT_SIZE	1
-#define XFS_TRANS_SETATTR_SIZE		2
-#define XFS_TRANS_INACTIVE		3
-#define XFS_TRANS_CREATE		4
-#define XFS_TRANS_CREATE_TRUNC		5
-#define XFS_TRANS_TRUNCATE_FILE		6
-#define XFS_TRANS_REMOVE		7
-#define XFS_TRANS_LINK			8
-#define XFS_TRANS_RENAME		9
-#define XFS_TRANS_MKDIR			10
-#define XFS_TRANS_RMDIR			11
-#define XFS_TRANS_SYMLINK		12
-#define XFS_TRANS_SET_DMATTRS		13
-#define XFS_TRANS_GROWFS		14
-#define XFS_TRANS_STRAT_WRITE		15
-#define XFS_TRANS_DIOSTRAT		16
-/* 17 was XFS_TRANS_WRITE_SYNC */
-#define	XFS_TRANS_WRITEID		18
-#define	XFS_TRANS_ADDAFORK		19
-#define	XFS_TRANS_ATTRINVAL		20
-#define	XFS_TRANS_ATRUNCATE		21
-#define	XFS_TRANS_ATTR_SET		22
-#define	XFS_TRANS_ATTR_RM		23
-#define	XFS_TRANS_ATTR_FLAG		24
-#define	XFS_TRANS_CLEAR_AGI_BUCKET	25
-#define XFS_TRANS_QM_SBCHANGE		26
-/*
- * Dummy entries since we use the transaction type to index into the
- * trans_type[] in xlog_recover_print_trans_head()
- */
-#define XFS_TRANS_DUMMY1		27
-#define XFS_TRANS_DUMMY2		28
-#define XFS_TRANS_QM_QUOTAOFF		29
-#define XFS_TRANS_QM_DQALLOC		30
-#define XFS_TRANS_QM_SETQLIM		31
-#define XFS_TRANS_QM_DQCLUSTER		32
-#define XFS_TRANS_QM_QINOCREATE		33
-#define XFS_TRANS_QM_QUOTAOFF_END	34
-#define XFS_TRANS_SB_UNIT		35
-#define XFS_TRANS_FSYNC_TS		36
-#define	XFS_TRANS_GROWFSRT_ALLOC	37
-#define	XFS_TRANS_GROWFSRT_ZERO		38
-#define	XFS_TRANS_GROWFSRT_FREE		39
-#define	XFS_TRANS_SWAPEXT		40
-#define	XFS_TRANS_SB_COUNT		41
-#define	XFS_TRANS_CHECKPOINT		42
-#define	XFS_TRANS_ICREATE		43
-#define	XFS_TRANS_TYPE_MAX		43
-/* new transaction types need to be reflected in xfs_logprint(8) */
-
-#define XFS_TRANS_TYPES \
-	{ XFS_TRANS_SETATTR_NOT_SIZE,	"SETATTR_NOT_SIZE" }, \
-	{ XFS_TRANS_SETATTR_SIZE,	"SETATTR_SIZE" }, \
-	{ XFS_TRANS_INACTIVE,		"INACTIVE" }, \
-	{ XFS_TRANS_CREATE,		"CREATE" }, \
-	{ XFS_TRANS_CREATE_TRUNC,	"CREATE_TRUNC" }, \
-	{ XFS_TRANS_TRUNCATE_FILE,	"TRUNCATE_FILE" }, \
-	{ XFS_TRANS_REMOVE,		"REMOVE" }, \
-	{ XFS_TRANS_LINK,		"LINK" }, \
-	{ XFS_TRANS_RENAME,		"RENAME" }, \
-	{ XFS_TRANS_MKDIR,		"MKDIR" }, \
-	{ XFS_TRANS_RMDIR,		"RMDIR" }, \
-	{ XFS_TRANS_SYMLINK,		"SYMLINK" }, \
-	{ XFS_TRANS_SET_DMATTRS,	"SET_DMATTRS" }, \
-	{ XFS_TRANS_GROWFS,		"GROWFS" }, \
-	{ XFS_TRANS_STRAT_WRITE,	"STRAT_WRITE" }, \
-	{ XFS_TRANS_DIOSTRAT,		"DIOSTRAT" }, \
-	{ XFS_TRANS_WRITEID,		"WRITEID" }, \
-	{ XFS_TRANS_ADDAFORK,		"ADDAFORK" }, \
-	{ XFS_TRANS_ATTRINVAL,		"ATTRINVAL" }, \
-	{ XFS_TRANS_ATRUNCATE,		"ATRUNCATE" }, \
-	{ XFS_TRANS_ATTR_SET,		"ATTR_SET" }, \
-	{ XFS_TRANS_ATTR_RM,		"ATTR_RM" }, \
-	{ XFS_TRANS_ATTR_FLAG,		"ATTR_FLAG" }, \
-	{ XFS_TRANS_CLEAR_AGI_BUCKET,	"CLEAR_AGI_BUCKET" }, \
-	{ XFS_TRANS_QM_SBCHANGE,	"QM_SBCHANGE" }, \
-	{ XFS_TRANS_QM_QUOTAOFF,	"QM_QUOTAOFF" }, \
-	{ XFS_TRANS_QM_DQALLOC,		"QM_DQALLOC" }, \
-	{ XFS_TRANS_QM_SETQLIM,		"QM_SETQLIM" }, \
-	{ XFS_TRANS_QM_DQCLUSTER,	"QM_DQCLUSTER" }, \
-	{ XFS_TRANS_QM_QINOCREATE,	"QM_QINOCREATE" }, \
-	{ XFS_TRANS_QM_QUOTAOFF_END,	"QM_QOFF_END" }, \
-	{ XFS_TRANS_SB_UNIT,		"SB_UNIT" }, \
-	{ XFS_TRANS_FSYNC_TS,		"FSYNC_TS" }, \
-	{ XFS_TRANS_GROWFSRT_ALLOC,	"GROWFSRT_ALLOC" }, \
-	{ XFS_TRANS_GROWFSRT_ZERO,	"GROWFSRT_ZERO" }, \
-	{ XFS_TRANS_GROWFSRT_FREE,	"GROWFSRT_FREE" }, \
-	{ XFS_TRANS_SWAPEXT,		"SWAPEXT" }, \
-	{ XFS_TRANS_SB_COUNT,		"SB_COUNT" }, \
-	{ XFS_TRANS_CHECKPOINT,		"CHECKPOINT" }, \
-	{ XFS_TRANS_DUMMY1,		"DUMMY1" }, \
-	{ XFS_TRANS_DUMMY2,		"DUMMY2" }, \
-	{ XLOG_UNMOUNT_REC_TYPE,	"UNMOUNT" }
-
-/*
- * This structure is used to track log items associated with
- * a transaction.  It points to the log item and keeps some
- * flags to track the state of the log item.  It also tracks
- * the amount of space needed to log the item it describes
- * once we get to commit processing (see xfs_trans_commit()).
- */
-struct xfs_log_item_desc {
-	struct xfs_log_item	*lid_item;
-	struct list_head	lid_trans;
-	unsigned char		lid_flags;
-};
-
-#define XFS_LID_DIRTY		0x1
-
-/*
- * Values for t_flags.
- */
-#define	XFS_TRANS_DIRTY		0x01	/* something needs to be logged */
-#define	XFS_TRANS_SB_DIRTY	0x02	/* superblock is modified */
-#define	XFS_TRANS_PERM_LOG_RES	0x04	/* xact took a permanent log res */
-#define	XFS_TRANS_SYNC		0x08	/* make commit synchronous */
-#define XFS_TRANS_DQ_DIRTY	0x10	/* at least one dquot in trx dirty */
-#define XFS_TRANS_RESERVE	0x20    /* OK to use reserved data blocks */
-#define XFS_TRANS_FREEZE_PROT	0x40	/* Transaction has elevated writer
-					   count in superblock */
-
-/*
- * Values for call flags parameter.
- */
-#define	XFS_TRANS_RELEASE_LOG_RES	0x4
-#define	XFS_TRANS_ABORT			0x8
-
-/*
- * Field values for xfs_trans_mod_sb.
- */
-#define	XFS_TRANS_SB_ICOUNT		0x00000001
-#define	XFS_TRANS_SB_IFREE		0x00000002
-#define	XFS_TRANS_SB_FDBLOCKS		0x00000004
-#define	XFS_TRANS_SB_RES_FDBLOCKS	0x00000008
-#define	XFS_TRANS_SB_FREXTENTS		0x00000010
-#define	XFS_TRANS_SB_RES_FREXTENTS	0x00000020
-#define	XFS_TRANS_SB_DBLOCKS		0x00000040
-#define	XFS_TRANS_SB_AGCOUNT		0x00000080
-#define	XFS_TRANS_SB_IMAXPCT		0x00000100
-#define	XFS_TRANS_SB_REXTSIZE		0x00000200
-#define	XFS_TRANS_SB_RBMBLOCKS		0x00000400
-#define	XFS_TRANS_SB_RBLOCKS		0x00000800
-#define	XFS_TRANS_SB_REXTENTS		0x00001000
-#define	XFS_TRANS_SB_REXTSLOG		0x00002000
-
-/*
- * Here we centralize the specification of XFS meta-data buffer
- * reference count values.  This determine how hard the buffer
- * cache tries to hold onto the buffer.
- */
-#define	XFS_AGF_REF		4
-#define	XFS_AGI_REF		4
-#define	XFS_AGFL_REF		3
-#define	XFS_INO_BTREE_REF	3
-#define	XFS_ALLOC_BTREE_REF	2
-#define	XFS_BMAP_BTREE_REF	2
-#define	XFS_DIR_BTREE_REF	2
-#define	XFS_INO_REF		2
-#define	XFS_ATTR_BTREE_REF	1
-#define	XFS_DQUOT_REF		1
-
-/*
- * Flags for xfs_trans_ichgtime().
- */
-#define	XFS_ICHGTIME_MOD	0x1	/* data fork modification timestamp */
-#define	XFS_ICHGTIME_CHG	0x2	/* inode field change timestamp */
-#define	XFS_ICHGTIME_CREATE	0x4	/* inode create timestamp */
-
-
-/*
  * Inode Log Item Format definitions.
  *
  * This is the structure used to lay out an inode log item in the
@@ -797,7 +625,6 @@ typedef struct xfs_qoff_logformat {
 	char			qf_pad[12];	/* padding for future */
 } xfs_qoff_logformat_t;
 
-
 /*
  * Disk quotas status in m_qflags, and also sb_qflags. 16 bits.
  */
@@ -849,8 +676,4 @@ struct xfs_icreate_log {
 	__be32		icl_gen;	/* inode generation number to use */
 };
 
-int	xfs_log_calc_unit_res(struct xfs_mount *mp, int unit_bytes);
-int	xfs_log_calc_minimum_size(struct xfs_mount *);
-
-
 #endif /* __XFS_LOG_FORMAT_H__ */
diff --git a/fs/xfs/xfs_log_priv.h b/fs/xfs/xfs_log_priv.h
index 136654b9400d..9bc403a9e54f 100644
--- a/fs/xfs/xfs_log_priv.h
+++ b/fs/xfs/xfs_log_priv.h
@@ -22,6 +22,7 @@ struct xfs_buf;
 struct xlog;
 struct xlog_ticket;
 struct xfs_mount;
+struct xfs_log_callback;
 
 /*
  * Flags for log structure
@@ -227,8 +228,8 @@ typedef struct xlog_in_core {
 
 	/* Callback structures need their own cacheline */
 	spinlock_t		ic_callback_lock ____cacheline_aligned_in_smp;
-	xfs_log_callback_t	*ic_callback;
-	xfs_log_callback_t	**ic_callback_tail;
+	struct xfs_log_callback	*ic_callback;
+	struct xfs_log_callback	**ic_callback_tail;
 
 	/* reference counts need their own cacheline */
 	atomic_t		ic_refcnt ____cacheline_aligned_in_smp;
@@ -254,7 +255,7 @@ struct xfs_cil_ctx {
 	int			space_used;	/* aggregate size of regions */
 	struct list_head	busy_extents;	/* busy extents in chkpt */
 	struct xfs_log_vec	*lv_chain;	/* logvecs being pushed */
-	xfs_log_callback_t	log_cb;		/* completion callback hook. */
+	struct xfs_log_callback	log_cb;		/* completion callback hook. */
 	struct list_head	committing;	/* ctx committing list */
 };
 
@@ -514,12 +515,10 @@ xlog_assign_grant_head(atomic64_t *head, int cycle, int space)
 /*
  * Committed Item List interfaces
  */
-int
-xlog_cil_init(struct xlog *log);
-void
-xlog_cil_init_post_recovery(struct xlog *log);
-void
-xlog_cil_destroy(struct xlog *log);
+int	xlog_cil_init(struct xlog *log);
+void	xlog_cil_init_post_recovery(struct xlog *log);
+void	xlog_cil_destroy(struct xlog *log);
+bool	xlog_cil_empty(struct xlog *log);
 
 /*
  * CIL force routines
diff --git a/fs/xfs/xfs_log_recover.c b/fs/xfs/xfs_log_recover.c
index 39797490a1f1..b6b669df40f3 100644
--- a/fs/xfs/xfs_log_recover.c
+++ b/fs/xfs/xfs_log_recover.c
@@ -17,42 +17,34 @@
  */
 #include "xfs.h"
 #include "xfs_fs.h"
+#include "xfs_shared.h"
 #include "xfs_format.h"
+#include "xfs_log_format.h"
+#include "xfs_trans_resv.h"
 #include "xfs_bit.h"
-#include "xfs_log.h"
 #include "xfs_inum.h"
-#include "xfs_trans.h"
 #include "xfs_sb.h"
 #include "xfs_ag.h"
 #include "xfs_mount.h"
-#include "xfs_error.h"
-#include "xfs_bmap_btree.h"
-#include "xfs_alloc_btree.h"
-#include "xfs_ialloc_btree.h"
-#include "xfs_btree.h"
-#include "xfs_dinode.h"
+#include "xfs_da_format.h"
 #include "xfs_inode.h"
-#include "xfs_inode_item.h"
-#include "xfs_alloc.h"
-#include "xfs_ialloc.h"
+#include "xfs_trans.h"
+#include "xfs_log.h"
 #include "xfs_log_priv.h"
-#include "xfs_buf_item.h"
 #include "xfs_log_recover.h"
+#include "xfs_inode_item.h"
 #include "xfs_extfree_item.h"
 #include "xfs_trans_priv.h"
+#include "xfs_alloc.h"
+#include "xfs_ialloc.h"
 #include "xfs_quota.h"
 #include "xfs_cksum.h"
 #include "xfs_trace.h"
 #include "xfs_icache.h"
-#include "xfs_icreate_item.h"
-
-/* Need all the magic numbers and buffer ops structures from these headers */
-#include "xfs_symlink.h"
-#include "xfs_da_btree.h"
-#include "xfs_dir2_format.h"
+#include "xfs_bmap_btree.h"
+#include "xfs_dinode.h"
+#include "xfs_error.h"
 #include "xfs_dir2.h"
-#include "xfs_attr_leaf.h"
-#include "xfs_attr_remote.h"
 
 #define BLK_AVG(blk1, blk2)	((blk1+blk2) >> 1)
 
@@ -305,9 +297,9 @@ xlog_header_check_dump(
 	xfs_mount_t		*mp,
 	xlog_rec_header_t	*head)
 {
-	xfs_debug(mp, "%s:  SB : uuid = %pU, fmt = %d\n",
+	xfs_debug(mp, "%s:  SB : uuid = %pU, fmt = %d",
 		__func__, &mp->m_sb.sb_uuid, XLOG_FMT);
-	xfs_debug(mp, "    log : uuid = %pU, fmt = %d\n",
+	xfs_debug(mp, "    log : uuid = %pU, fmt = %d",
 		&head->h_fs_uuid, be32_to_cpu(head->h_fmt));
 }
 #else
@@ -2362,7 +2354,7 @@ xlog_recover_do_reg_buffer(
 					item->ri_buf[i].i_len, __func__);
 				goto next;
 			}
-			error = xfs_qm_dqcheck(mp, item->ri_buf[i].i_addr,
+			error = xfs_dqcheck(mp, item->ri_buf[i].i_addr,
 					       -1, 0, XFS_QMOPT_DOWARN,
 					       "dquot_buf_recover");
 			if (error)
@@ -2394,133 +2386,6 @@ xlog_recover_do_reg_buffer(
 }
 
 /*
- * Do some primitive error checking on ondisk dquot data structures.
- */
-int
-xfs_qm_dqcheck(
-	struct xfs_mount *mp,
-	xfs_disk_dquot_t *ddq,
-	xfs_dqid_t	 id,
-	uint		 type,	  /* used only when IO_dorepair is true */
-	uint		 flags,
-	char		 *str)
-{
-	xfs_dqblk_t	 *d = (xfs_dqblk_t *)ddq;
-	int		errs = 0;
-
-	/*
-	 * We can encounter an uninitialized dquot buffer for 2 reasons:
-	 * 1. If we crash while deleting the quotainode(s), and those blks got
-	 *    used for user data. This is because we take the path of regular
-	 *    file deletion; however, the size field of quotainodes is never
-	 *    updated, so all the tricks that we play in itruncate_finish
-	 *    don't quite matter.
-	 *
-	 * 2. We don't play the quota buffers when there's a quotaoff logitem.
-	 *    But the allocation will be replayed so we'll end up with an
-	 *    uninitialized quota block.
-	 *
-	 * This is all fine; things are still consistent, and we haven't lost
-	 * any quota information. Just don't complain about bad dquot blks.
-	 */
-	if (ddq->d_magic != cpu_to_be16(XFS_DQUOT_MAGIC)) {
-		if (flags & XFS_QMOPT_DOWARN)
-			xfs_alert(mp,
-			"%s : XFS dquot ID 0x%x, magic 0x%x != 0x%x",
-			str, id, be16_to_cpu(ddq->d_magic), XFS_DQUOT_MAGIC);
-		errs++;
-	}
-	if (ddq->d_version != XFS_DQUOT_VERSION) {
-		if (flags & XFS_QMOPT_DOWARN)
-			xfs_alert(mp,
-			"%s : XFS dquot ID 0x%x, version 0x%x != 0x%x",
-			str, id, ddq->d_version, XFS_DQUOT_VERSION);
-		errs++;
-	}
-
-	if (ddq->d_flags != XFS_DQ_USER &&
-	    ddq->d_flags != XFS_DQ_PROJ &&
-	    ddq->d_flags != XFS_DQ_GROUP) {
-		if (flags & XFS_QMOPT_DOWARN)
-			xfs_alert(mp,
-			"%s : XFS dquot ID 0x%x, unknown flags 0x%x",
-			str, id, ddq->d_flags);
-		errs++;
-	}
-
-	if (id != -1 && id != be32_to_cpu(ddq->d_id)) {
-		if (flags & XFS_QMOPT_DOWARN)
-			xfs_alert(mp,
-			"%s : ondisk-dquot 0x%p, ID mismatch: "
-			"0x%x expected, found id 0x%x",
-			str, ddq, id, be32_to_cpu(ddq->d_id));
-		errs++;
-	}
-
-	if (!errs && ddq->d_id) {
-		if (ddq->d_blk_softlimit &&
-		    be64_to_cpu(ddq->d_bcount) >
-				be64_to_cpu(ddq->d_blk_softlimit)) {
-			if (!ddq->d_btimer) {
-				if (flags & XFS_QMOPT_DOWARN)
-					xfs_alert(mp,
-			"%s : Dquot ID 0x%x (0x%p) BLK TIMER NOT STARTED",
-					str, (int)be32_to_cpu(ddq->d_id), ddq);
-				errs++;
-			}
-		}
-		if (ddq->d_ino_softlimit &&
-		    be64_to_cpu(ddq->d_icount) >
-				be64_to_cpu(ddq->d_ino_softlimit)) {
-			if (!ddq->d_itimer) {
-				if (flags & XFS_QMOPT_DOWARN)
-					xfs_alert(mp,
-			"%s : Dquot ID 0x%x (0x%p) INODE TIMER NOT STARTED",
-					str, (int)be32_to_cpu(ddq->d_id), ddq);
-				errs++;
-			}
-		}
-		if (ddq->d_rtb_softlimit &&
-		    be64_to_cpu(ddq->d_rtbcount) >
-				be64_to_cpu(ddq->d_rtb_softlimit)) {
-			if (!ddq->d_rtbtimer) {
-				if (flags & XFS_QMOPT_DOWARN)
-					xfs_alert(mp,
-			"%s : Dquot ID 0x%x (0x%p) RTBLK TIMER NOT STARTED",
-					str, (int)be32_to_cpu(ddq->d_id), ddq);
-				errs++;
-			}
-		}
-	}
-
-	if (!errs || !(flags & XFS_QMOPT_DQREPAIR))
-		return errs;
-
-	if (flags & XFS_QMOPT_DOWARN)
-		xfs_notice(mp, "Re-initializing dquot ID 0x%x", id);
-
-	/*
-	 * Typically, a repair is only requested by quotacheck.
-	 */
-	ASSERT(id != -1);
-	ASSERT(flags & XFS_QMOPT_DQREPAIR);
-	memset(d, 0, sizeof(xfs_dqblk_t));
-
-	d->dd_diskdq.d_magic = cpu_to_be16(XFS_DQUOT_MAGIC);
-	d->dd_diskdq.d_version = XFS_DQUOT_VERSION;
-	d->dd_diskdq.d_flags = type;
-	d->dd_diskdq.d_id = cpu_to_be32(id);
-
-	if (xfs_sb_version_hascrc(&mp->m_sb)) {
-		uuid_copy(&d->dd_uuid, &mp->m_sb.sb_uuid);
-		xfs_update_cksum((char *)d, sizeof(struct xfs_dqblk),
-				 XFS_DQUOT_CRC_OFF);
-	}
-
-	return errs;
-}
-
-/*
  * Perform a dquot buffer recovery.
  * Simple algorithm: if we have found a QUOTAOFF log item of the same type
  * (ie. USR or GRP), then just toss this buffer away; don't recover it.
@@ -3125,7 +2990,7 @@ xlog_recover_dquot_pass2(
 	 */
 	dq_f = item->ri_buf[0].i_addr;
 	ASSERT(dq_f);
-	error = xfs_qm_dqcheck(mp, recddq, dq_f->qlf_id, 0, XFS_QMOPT_DOWARN,
+	error = xfs_dqcheck(mp, recddq, dq_f->qlf_id, 0, XFS_QMOPT_DOWARN,
 			   "xlog_recover_dquot_pass2 (log copy)");
 	if (error)
 		return XFS_ERROR(EIO);
@@ -3145,7 +3010,7 @@ xlog_recover_dquot_pass2(
 	 * was among a chunk of dquots created earlier, and we did some
 	 * minimal initialization then.
 	 */
-	error = xfs_qm_dqcheck(mp, ddq, dq_f->qlf_id, 0, XFS_QMOPT_DOWARN,
+	error = xfs_dqcheck(mp, ddq, dq_f->qlf_id, 0, XFS_QMOPT_DOWARN,
 			   "xlog_recover_dquot_pass2");
 	if (error) {
 		xfs_buf_relse(bp);
@@ -4077,7 +3942,7 @@ xlog_unpack_data_crc(
 	if (crc != rhead->h_crc) {
 		if (rhead->h_crc || xfs_sb_version_hascrc(&log->l_mp->m_sb)) {
 			xfs_alert(log->l_mp,
-		"log record CRC mismatch: found 0x%x, expected 0x%x.\n",
+		"log record CRC mismatch: found 0x%x, expected 0x%x.",
 					le32_to_cpu(rhead->h_crc),
 					le32_to_cpu(crc));
 			xfs_hex_dump(dp, 32);
diff --git a/fs/xfs/xfs_log_rlimit.c b/fs/xfs/xfs_log_rlimit.c
index bbcec0bbc12d..2af1a0a4d0f1 100644
--- a/fs/xfs/xfs_log_rlimit.c
+++ b/fs/xfs/xfs_log_rlimit.c
@@ -17,16 +17,19 @@
  */
 #include "xfs.h"
 #include "xfs_fs.h"
-#include "xfs_log.h"
-#include "xfs_trans.h"
+#include "xfs_shared.h"
+#include "xfs_format.h"
+#include "xfs_log_format.h"
+#include "xfs_trans_resv.h"
 #include "xfs_ag.h"
 #include "xfs_sb.h"
 #include "xfs_mount.h"
+#include "xfs_da_format.h"
 #include "xfs_trans_space.h"
-#include "xfs_bmap_btree.h"
 #include "xfs_inode.h"
 #include "xfs_da_btree.h"
 #include "xfs_attr_leaf.h"
+#include "xfs_bmap_btree.h"
 
 /*
  * Calculate the maximum length in bytes that would be required for a local
diff --git a/fs/xfs/xfs_message.c b/fs/xfs/xfs_message.c
index 9163dc140532..63ca2f0420b1 100644
--- a/fs/xfs/xfs_message.c
+++ b/fs/xfs/xfs_message.c
@@ -17,9 +17,8 @@
 
 #include "xfs.h"
 #include "xfs_fs.h"
-#include "xfs_types.h"
-#include "xfs_log.h"
-#include "xfs_trans.h"
+#include "xfs_log_format.h"
+#include "xfs_trans_resv.h"
 #include "xfs_sb.h"
 #include "xfs_ag.h"
 #include "xfs_mount.h"
diff --git a/fs/xfs/xfs_mount.c b/fs/xfs/xfs_mount.c
index 5dcc68019d1b..da88f167af78 100644
--- a/fs/xfs/xfs_mount.c
+++ b/fs/xfs/xfs_mount.c
@@ -17,35 +17,30 @@
  */
 #include "xfs.h"
 #include "xfs_fs.h"
+#include "xfs_shared.h"
 #include "xfs_format.h"
+#include "xfs_log_format.h"
+#include "xfs_trans_resv.h"
 #include "xfs_bit.h"
-#include "xfs_log.h"
 #include "xfs_inum.h"
-#include "xfs_trans.h"
-#include "xfs_trans_priv.h"
 #include "xfs_sb.h"
 #include "xfs_ag.h"
 #include "xfs_mount.h"
-#include "xfs_da_btree.h"
-#include "xfs_dir2_format.h"
-#include "xfs_dir2.h"
-#include "xfs_bmap_btree.h"
-#include "xfs_alloc_btree.h"
-#include "xfs_ialloc_btree.h"
-#include "xfs_dinode.h"
+#include "xfs_da_format.h"
 #include "xfs_inode.h"
-#include "xfs_btree.h"
+#include "xfs_dir2.h"
 #include "xfs_ialloc.h"
 #include "xfs_alloc.h"
 #include "xfs_rtalloc.h"
 #include "xfs_bmap.h"
+#include "xfs_trans.h"
+#include "xfs_trans_priv.h"
+#include "xfs_log.h"
 #include "xfs_error.h"
 #include "xfs_quota.h"
 #include "xfs_fsops.h"
 #include "xfs_trace.h"
 #include "xfs_icache.h"
-#include "xfs_cksum.h"
-#include "xfs_buf_item.h"
 
 
 #ifdef HAVE_PERCPU_SB
diff --git a/fs/xfs/xfs_mount.h b/fs/xfs/xfs_mount.h
index 1fa0584b5627..1d8101a10d8e 100644
--- a/fs/xfs/xfs_mount.h
+++ b/fs/xfs/xfs_mount.h
@@ -26,6 +26,7 @@ struct xfs_mru_cache;
 struct xfs_nameops;
 struct xfs_ail;
 struct xfs_quotainfo;
+struct xfs_dir_ops;
 
 #ifdef HAVE_PERCPU_SB
 
@@ -148,6 +149,8 @@ typedef struct xfs_mount {
 	int			m_dir_magicpct;	/* 37% of the dir blocksize */
 	__uint8_t		m_sectbb_log;	/* sectlog - BBSHIFT */
 	const struct xfs_nameops *m_dirnameops;	/* vector of dir name ops */
+	const struct xfs_dir_ops *m_dir_inode_ops; /* vector of dir inode ops */
+	const struct xfs_dir_ops *m_nondir_inode_ops; /* !dir inode ops */
 	int			m_dirblksize;	/* directory block sz--bytes */
 	int			m_dirblkfsbs;	/* directory block sz--fsbs */
 	xfs_dablk_t		m_dirdatablk;	/* blockno of dir data v2 */
diff --git a/fs/xfs/xfs_qm.c b/fs/xfs/xfs_qm.c
index 3e6c2e6c9cd2..14a4996cfec6 100644
--- a/fs/xfs/xfs_qm.c
+++ b/fs/xfs/xfs_qm.c
@@ -17,31 +17,28 @@
  */
 #include "xfs.h"
 #include "xfs_fs.h"
+#include "xfs_shared.h"
 #include "xfs_format.h"
+#include "xfs_log_format.h"
+#include "xfs_trans_resv.h"
 #include "xfs_bit.h"
-#include "xfs_log.h"
-#include "xfs_trans.h"
 #include "xfs_sb.h"
 #include "xfs_ag.h"
-#include "xfs_alloc.h"
-#include "xfs_quota.h"
 #include "xfs_mount.h"
-#include "xfs_bmap_btree.h"
-#include "xfs_ialloc_btree.h"
-#include "xfs_dinode.h"
 #include "xfs_inode.h"
 #include "xfs_ialloc.h"
 #include "xfs_itable.h"
-#include "xfs_rtalloc.h"
+#include "xfs_quota.h"
 #include "xfs_error.h"
 #include "xfs_bmap.h"
-#include "xfs_attr.h"
-#include "xfs_buf_item.h"
+#include "xfs_bmap_btree.h"
+#include "xfs_trans.h"
 #include "xfs_trans_space.h"
 #include "xfs_qm.h"
 #include "xfs_trace.h"
 #include "xfs_icache.h"
 #include "xfs_cksum.h"
+#include "xfs_dinode.h"
 
 /*
  * The global quota manager. There is only one of these for the entire
@@ -664,20 +661,6 @@ xfs_qm_dqdetach(
 	}
 }
 
-int
-xfs_qm_calc_dquots_per_chunk(
-	struct xfs_mount	*mp,
-	unsigned int		nbblks)	/* basic block units */
-{
-	unsigned int	ndquots;
-
-	ASSERT(nbblks > 0);
-	ndquots = BBTOB(nbblks);
-	do_div(ndquots, sizeof(xfs_dqblk_t));
-
-	return ndquots;
-}
-
 struct xfs_qm_isolate {
 	struct list_head	buffers;
 	struct list_head	dispose;
@@ -858,7 +841,7 @@ xfs_qm_init_quotainfo(
 
 	/* Precalc some constants */
 	qinf->qi_dqchunklen = XFS_FSB_TO_BB(mp, XFS_DQUOT_CLUSTER_SIZE_FSB);
-	qinf->qi_dqperchunk = xfs_qm_calc_dquots_per_chunk(mp,
+	qinf->qi_dqperchunk = xfs_calc_dquots_per_chunk(mp,
 							qinf->qi_dqchunklen);
 
 	mp->m_qflags |= (mp->m_sb.sb_qflags & XFS_ALL_QUOTA_CHKD);
@@ -1092,10 +1075,10 @@ xfs_qm_reset_dqcounts(
 		/*
 		 * Do a sanity check, and if needed, repair the dqblk. Don't
 		 * output any warnings because it's perfectly possible to
-		 * find uninitialised dquot blks. See comment in xfs_qm_dqcheck.
+		 * find uninitialised dquot blks. See comment in xfs_dqcheck.
 		 */
-		(void) xfs_qm_dqcheck(mp, ddq, id+j, type, XFS_QMOPT_DQREPAIR,
-				      "xfs_quotacheck");
+		xfs_dqcheck(mp, ddq, id+j, type, XFS_QMOPT_DQREPAIR,
+			    "xfs_quotacheck");
 		ddq->d_bcount = 0;
 		ddq->d_icount = 0;
 		ddq->d_rtbcount = 0;
diff --git a/fs/xfs/xfs_qm.h b/fs/xfs/xfs_qm.h
index 2b602df9c242..a788b66a5cb1 100644
--- a/fs/xfs/xfs_qm.h
+++ b/fs/xfs/xfs_qm.h
@@ -103,8 +103,6 @@ xfs_dq_to_quota_inode(struct xfs_dquot *dqp)
 	return NULL;
 }
 
-extern int	xfs_qm_calc_dquots_per_chunk(struct xfs_mount *mp,
-					     unsigned int nbblks);
 extern void	xfs_trans_mod_dquot(struct xfs_trans *,
 					struct xfs_dquot *, uint, long);
 extern int	xfs_trans_reserve_quota_bydquots(struct xfs_trans *,
diff --git a/fs/xfs/xfs_qm_bhv.c b/fs/xfs/xfs_qm_bhv.c
index 3af50ccdfac1..e9be63abd8d2 100644
--- a/fs/xfs/xfs_qm_bhv.c
+++ b/fs/xfs/xfs_qm_bhv.c
@@ -18,21 +18,15 @@
 #include "xfs.h"
 #include "xfs_fs.h"
 #include "xfs_format.h"
-#include "xfs_log.h"
-#include "xfs_trans.h"
+#include "xfs_log_format.h"
+#include "xfs_trans_resv.h"
 #include "xfs_sb.h"
 #include "xfs_ag.h"
-#include "xfs_alloc.h"
 #include "xfs_quota.h"
 #include "xfs_mount.h"
-#include "xfs_bmap_btree.h"
 #include "xfs_inode.h"
-#include "xfs_itable.h"
-#include "xfs_bmap.h"
-#include "xfs_rtalloc.h"
 #include "xfs_error.h"
-#include "xfs_attr.h"
-#include "xfs_buf_item.h"
+#include "xfs_trans.h"
 #include "xfs_qm.h"
 
 
diff --git a/fs/xfs/xfs_qm_syscalls.c b/fs/xfs/xfs_qm_syscalls.c
index 8174aad0b388..437c9198031a 100644
--- a/fs/xfs/xfs_qm_syscalls.c
+++ b/fs/xfs/xfs_qm_syscalls.c
@@ -20,24 +20,18 @@
 
 #include "xfs.h"
 #include "xfs_fs.h"
+#include "xfs_shared.h"
 #include "xfs_format.h"
+#include "xfs_log_format.h"
+#include "xfs_trans_resv.h"
 #include "xfs_bit.h"
-#include "xfs_log.h"
-#include "xfs_trans.h"
 #include "xfs_sb.h"
 #include "xfs_ag.h"
-#include "xfs_alloc.h"
-#include "xfs_quota.h"
 #include "xfs_mount.h"
-#include "xfs_bmap_btree.h"
 #include "xfs_inode.h"
-#include "xfs_inode_item.h"
-#include "xfs_itable.h"
-#include "xfs_bmap.h"
-#include "xfs_rtalloc.h"
+#include "xfs_trans.h"
 #include "xfs_error.h"
-#include "xfs_attr.h"
-#include "xfs_buf_item.h"
+#include "xfs_quota.h"
 #include "xfs_qm.h"
 #include "xfs_trace.h"
 #include "xfs_icache.h"
@@ -287,7 +281,7 @@ xfs_qm_scall_trunc_qfiles(
 	int		error = 0, error2 = 0;
 
 	if (!xfs_sb_version_hasquota(&mp->m_sb) || flags == 0) {
-		xfs_debug(mp, "%s: flags=%x m_qflags=%x\n",
+		xfs_debug(mp, "%s: flags=%x m_qflags=%x",
 			__func__, flags, mp->m_qflags);
 		return XFS_ERROR(EINVAL);
 	}
@@ -325,7 +319,7 @@ xfs_qm_scall_quotaon(
 	sbflags = 0;
 
 	if (flags == 0) {
-		xfs_debug(mp, "%s: zero flags, m_qflags=%x\n",
+		xfs_debug(mp, "%s: zero flags, m_qflags=%x",
 			__func__, mp->m_qflags);
 		return XFS_ERROR(EINVAL);
 	}
@@ -348,7 +342,7 @@ xfs_qm_scall_quotaon(
 	     (mp->m_sb.sb_qflags & XFS_PQUOTA_ACCT) == 0 &&
 	     (flags & XFS_PQUOTA_ENFD))) {
 		xfs_debug(mp,
-			"%s: Can't enforce without acct, flags=%x sbflags=%x\n",
+			"%s: Can't enforce without acct, flags=%x sbflags=%x",
 			__func__, flags, mp->m_sb.sb_qflags);
 		return XFS_ERROR(EINVAL);
 	}
@@ -648,7 +642,7 @@ xfs_qm_scall_setqlim(
 			q->qi_bsoftlimit = soft;
 		}
 	} else {
-		xfs_debug(mp, "blkhard %Ld < blksoft %Ld\n", hard, soft);
+		xfs_debug(mp, "blkhard %Ld < blksoft %Ld", hard, soft);
 	}
 	hard = (newlim->d_fieldmask & FS_DQ_RTBHARD) ?
 		(xfs_qcnt_t) XFS_BB_TO_FSB(mp, newlim->d_rtb_hardlimit) :
@@ -664,7 +658,7 @@ xfs_qm_scall_setqlim(
 			q->qi_rtbsoftlimit = soft;
 		}
 	} else {
-		xfs_debug(mp, "rtbhard %Ld < rtbsoft %Ld\n", hard, soft);
+		xfs_debug(mp, "rtbhard %Ld < rtbsoft %Ld", hard, soft);
 	}
 
 	hard = (newlim->d_fieldmask & FS_DQ_IHARD) ?
@@ -681,7 +675,7 @@ xfs_qm_scall_setqlim(
 			q->qi_isoftlimit = soft;
 		}
 	} else {
-		xfs_debug(mp, "ihard %Ld < isoft %Ld\n", hard, soft);
+		xfs_debug(mp, "ihard %Ld < isoft %Ld", hard, soft);
 	}
 
 	/*
diff --git a/fs/xfs/xfs_quota.h b/fs/xfs/xfs_quota.h
index e7d84d2d8683..5376dd406ba2 100644
--- a/fs/xfs/xfs_quota.h
+++ b/fs/xfs/xfs_quota.h
@@ -150,10 +150,6 @@ static inline int xfs_trans_reserve_quota_bydquots(struct xfs_trans *tp,
 	xfs_trans_reserve_quota_bydquots(tp, mp, ud, gd, pd, nb, ni, \
 				f | XFS_QMOPT_RES_REGBLKS)
 
-extern int xfs_qm_dqcheck(struct xfs_mount *, xfs_disk_dquot_t *,
-				xfs_dqid_t, uint, uint, char *);
 extern int xfs_mount_reset_sbqflags(struct xfs_mount *);
 
-extern const struct xfs_buf_ops xfs_dquot_buf_ops;
-
 #endif	/* __XFS_QUOTA_H__ */
diff --git a/fs/xfs/xfs_quota_defs.h b/fs/xfs/xfs_quota_defs.h
index e6b0d6e1f4f2..b3b2b1065c0f 100644
--- a/fs/xfs/xfs_quota_defs.h
+++ b/fs/xfs/xfs_quota_defs.h
@@ -154,4 +154,8 @@ typedef __uint16_t	xfs_qwarncnt_t;
 		(XFS_QMOPT_UQUOTA | XFS_QMOPT_PQUOTA | XFS_QMOPT_GQUOTA)
 #define XFS_QMOPT_RESBLK_MASK	(XFS_QMOPT_RES_REGBLKS | XFS_QMOPT_RES_RTBLKS)
 
+extern int xfs_dqcheck(struct xfs_mount *mp, xfs_disk_dquot_t *ddq,
+		       xfs_dqid_t id, uint type, uint flags, char *str);
+extern int xfs_calc_dquots_per_chunk(struct xfs_mount *mp, unsigned int nbblks);
+
 #endif	/* __XFS_QUOTA_H__ */
diff --git a/fs/xfs/xfs_quotaops.c b/fs/xfs/xfs_quotaops.c
index 1326d81596c2..af33cafe69b6 100644
--- a/fs/xfs/xfs_quotaops.c
+++ b/fs/xfs/xfs_quotaops.c
@@ -17,15 +17,14 @@
  */
 #include "xfs.h"
 #include "xfs_format.h"
+#include "xfs_log_format.h"
 #include "xfs_trans_resv.h"
-#include "xfs_log.h"
 #include "xfs_sb.h"
 #include "xfs_ag.h"
 #include "xfs_mount.h"
+#include "xfs_inode.h"
 #include "xfs_quota.h"
 #include "xfs_trans.h"
-#include "xfs_bmap_btree.h"
-#include "xfs_inode.h"
 #include "xfs_qm.h"
 #include <linux/quota.h>
 
diff --git a/fs/xfs/xfs_rtalloc.c b/fs/xfs/xfs_rtalloc.c
index 6f9e63c9fc26..a6a76b2b6a85 100644
--- a/fs/xfs/xfs_rtalloc.c
+++ b/fs/xfs/xfs_rtalloc.c
@@ -17,172 +17,260 @@
  */
 #include "xfs.h"
 #include "xfs_fs.h"
+#include "xfs_shared.h"
 #include "xfs_format.h"
+#include "xfs_log_format.h"
+#include "xfs_trans_resv.h"
 #include "xfs_bit.h"
-#include "xfs_log.h"
-#include "xfs_trans.h"
 #include "xfs_sb.h"
 #include "xfs_ag.h"
 #include "xfs_mount.h"
-#include "xfs_bmap_btree.h"
-#include "xfs_dinode.h"
 #include "xfs_inode.h"
-#include "xfs_alloc.h"
 #include "xfs_bmap.h"
 #include "xfs_bmap_util.h"
-#include "xfs_rtalloc.h"
-#include "xfs_fsops.h"
+#include "xfs_bmap_btree.h"
+#include "xfs_alloc.h"
 #include "xfs_error.h"
-#include "xfs_inode_item.h"
+#include "xfs_trans.h"
 #include "xfs_trans_space.h"
 #include "xfs_trace.h"
 #include "xfs_buf.h"
 #include "xfs_icache.h"
+#include "xfs_dinode.h"
+#include "xfs_rtalloc.h"
 
 
 /*
- * Prototypes for internal functions.
+ * Read and return the summary information for a given extent size,
+ * bitmap block combination.
+ * Keeps track of a current summary block, so we don't keep reading
+ * it from the buffer cache.
  */
+STATIC int				/* error */
+xfs_rtget_summary(
+	xfs_mount_t	*mp,		/* file system mount structure */
+	xfs_trans_t	*tp,		/* transaction pointer */
+	int		log,		/* log2 of extent size */
+	xfs_rtblock_t	bbno,		/* bitmap block number */
+	xfs_buf_t	**rbpp,		/* in/out: summary block buffer */
+	xfs_fsblock_t	*rsb,		/* in/out: summary block number */
+	xfs_suminfo_t	*sum)		/* out: summary info for this block */
+{
+	xfs_buf_t	*bp;		/* buffer for summary block */
+	int		error;		/* error value */
+	xfs_fsblock_t	sb;		/* summary fsblock */
+	int		so;		/* index into the summary file */
+	xfs_suminfo_t	*sp;		/* pointer to returned data */
 
+	/*
+	 * Compute entry number in the summary file.
+	 */
+	so = XFS_SUMOFFS(mp, log, bbno);
+	/*
+	 * Compute the block number in the summary file.
+	 */
+	sb = XFS_SUMOFFSTOBLOCK(mp, so);
+	/*
+	 * If we have an old buffer, and the block number matches, use that.
+	 */
+	if (rbpp && *rbpp && *rsb == sb)
+		bp = *rbpp;
+	/*
+	 * Otherwise we have to get the buffer.
+	 */
+	else {
+		/*
+		 * If there was an old one, get rid of it first.
+		 */
+		if (rbpp && *rbpp)
+			xfs_trans_brelse(tp, *rbpp);
+		error = xfs_rtbuf_get(mp, tp, sb, 1, &bp);
+		if (error) {
+			return error;
+		}
+		/*
+		 * Remember this buffer and block for the next call.
+		 */
+		if (rbpp) {
+			*rbpp = bp;
+			*rsb = sb;
+		}
+	}
+	/*
+	 * Point to the summary information & copy it out.
+	 */
+	sp = XFS_SUMPTR(mp, bp, so);
+	*sum = *sp;
+	/*
+	 * Drop the buffer if we're not asked to remember it.
+	 */
+	if (!rbpp)
+		xfs_trans_brelse(tp, bp);
+	return 0;
+}
 
-STATIC int xfs_rtallocate_range(xfs_mount_t *, xfs_trans_t *, xfs_rtblock_t,
-		xfs_extlen_t, xfs_buf_t **, xfs_fsblock_t *);
-STATIC int xfs_rtany_summary(xfs_mount_t *, xfs_trans_t *, int, int,
-		xfs_rtblock_t, xfs_buf_t **, xfs_fsblock_t *, int *);
-STATIC int xfs_rtcheck_range(xfs_mount_t *, xfs_trans_t *, xfs_rtblock_t,
-		xfs_extlen_t, int, xfs_rtblock_t *, int *);
-STATIC int xfs_rtfind_back(xfs_mount_t *, xfs_trans_t *, xfs_rtblock_t,
-		xfs_rtblock_t, xfs_rtblock_t *);
-STATIC int xfs_rtfind_forw(xfs_mount_t *, xfs_trans_t *, xfs_rtblock_t,
-		xfs_rtblock_t, xfs_rtblock_t *);
-STATIC int xfs_rtget_summary( xfs_mount_t *, xfs_trans_t *, int,
-		xfs_rtblock_t, xfs_buf_t **, xfs_fsblock_t *, xfs_suminfo_t *);
-STATIC int xfs_rtmodify_range(xfs_mount_t *, xfs_trans_t *, xfs_rtblock_t,
-		xfs_extlen_t, int);
-STATIC int xfs_rtmodify_summary(xfs_mount_t *, xfs_trans_t *, int,
-		xfs_rtblock_t, int, xfs_buf_t **, xfs_fsblock_t *);
-
-/*
- * Internal functions.
- */
 
 /*
- * Allocate space to the bitmap or summary file, and zero it, for growfs.
+ * Return whether there are any free extents in the size range given
+ * by low and high, for the bitmap block bbno.
  */
 STATIC int				/* error */
-xfs_growfs_rt_alloc(
-	xfs_mount_t	*mp,		/* file system mount point */
-	xfs_extlen_t	oblocks,	/* old count of blocks */
-	xfs_extlen_t	nblocks,	/* new count of blocks */
-	xfs_inode_t	*ip)		/* inode (bitmap/summary) */
+xfs_rtany_summary(
+	xfs_mount_t	*mp,		/* file system mount structure */
+	xfs_trans_t	*tp,		/* transaction pointer */
+	int		low,		/* low log2 extent size */
+	int		high,		/* high log2 extent size */
+	xfs_rtblock_t	bbno,		/* bitmap block number */
+	xfs_buf_t	**rbpp,		/* in/out: summary block buffer */
+	xfs_fsblock_t	*rsb,		/* in/out: summary block number */
+	int		*stat)		/* out: any good extents here? */
 {
-	xfs_fileoff_t	bno;		/* block number in file */
-	xfs_buf_t	*bp;		/* temporary buffer for zeroing */
-	int		committed;	/* transaction committed flag */
-	xfs_daddr_t	d;		/* disk block address */
-	int		error;		/* error return value */
-	xfs_fsblock_t	firstblock;	/* first block allocated in xaction */
-	xfs_bmap_free_t	flist;		/* list of freed blocks */
-	xfs_fsblock_t	fsbno;		/* filesystem block for bno */
-	xfs_bmbt_irec_t	map;		/* block map output */
-	int		nmap;		/* number of block maps */
-	int		resblks;	/* space reservation */
+	int		error;		/* error value */
+	int		log;		/* loop counter, log2 of ext. size */
+	xfs_suminfo_t	sum;		/* summary data */
 
 	/*
-	 * Allocate space to the file, as necessary.
+	 * Loop over logs of extent sizes.  Order is irrelevant.
 	 */
-	while (oblocks < nblocks) {
-		int		cancelflags = 0;
-		xfs_trans_t	*tp;
-
-		tp = xfs_trans_alloc(mp, XFS_TRANS_GROWFSRT_ALLOC);
-		resblks = XFS_GROWFSRT_SPACE_RES(mp, nblocks - oblocks);
+	for (log = low; log <= high; log++) {
 		/*
-		 * Reserve space & log for one extent added to the file.
+		 * Get one summary datum.
 		 */
-		error = xfs_trans_reserve(tp, &M_RES(mp)->tr_growdata,
-					  resblks, 0);
-		if (error)
-			goto error_cancel;
-		cancelflags = XFS_TRANS_RELEASE_LOG_RES;
+		error = xfs_rtget_summary(mp, tp, log, bbno, rbpp, rsb, &sum);
+		if (error) {
+			return error;
+		}
 		/*
-		 * Lock the inode.
+		 * If there are any, return success.
 		 */
-		xfs_ilock(ip, XFS_ILOCK_EXCL);
-		xfs_trans_ijoin(tp, ip, XFS_ILOCK_EXCL);
+		if (sum) {
+			*stat = 1;
+			return 0;
+		}
+	}
+	/*
+	 * Found nothing, return failure.
+	 */
+	*stat = 0;
+	return 0;
+}
 
-		xfs_bmap_init(&flist, &firstblock);
-		/*
-		 * Allocate blocks to the bitmap file.
-		 */
-		nmap = 1;
-		cancelflags |= XFS_TRANS_ABORT;
-		error = xfs_bmapi_write(tp, ip, oblocks, nblocks - oblocks,
-					XFS_BMAPI_METADATA, &firstblock,
-					resblks, &map, &nmap, &flist);
-		if (!error && nmap < 1)
-			error = XFS_ERROR(ENOSPC);
-		if (error)
-			goto error_cancel;
-		/*
-		 * Free any blocks freed up in the transaction, then commit.
-		 */
-		error = xfs_bmap_finish(&tp, &flist, &committed);
-		if (error)
-			goto error_cancel;
-		error = xfs_trans_commit(tp, XFS_TRANS_RELEASE_LOG_RES);
-		if (error)
-			goto error;
-		/*
-		 * Now we need to clear the allocated blocks.
-		 * Do this one block per transaction, to keep it simple.
-		 */
-		cancelflags = 0;
-		for (bno = map.br_startoff, fsbno = map.br_startblock;
-		     bno < map.br_startoff + map.br_blockcount;
-		     bno++, fsbno++) {
-			tp = xfs_trans_alloc(mp, XFS_TRANS_GROWFSRT_ZERO);
-			/*
-			 * Reserve log for one block zeroing.
-			 */
-			error = xfs_trans_reserve(tp, &M_RES(mp)->tr_growrtzero,
-						  0, 0);
+
+/*
+ * Copy and transform the summary file, given the old and new
+ * parameters in the mount structures.
+ */
+STATIC int				/* error */
+xfs_rtcopy_summary(
+	xfs_mount_t	*omp,		/* old file system mount point */
+	xfs_mount_t	*nmp,		/* new file system mount point */
+	xfs_trans_t	*tp)		/* transaction pointer */
+{
+	xfs_rtblock_t	bbno;		/* bitmap block number */
+	xfs_buf_t	*bp;		/* summary buffer */
+	int		error;		/* error return value */
+	int		log;		/* summary level number (log length) */
+	xfs_suminfo_t	sum;		/* summary data */
+	xfs_fsblock_t	sumbno;		/* summary block number */
+
+	bp = NULL;
+	for (log = omp->m_rsumlevels - 1; log >= 0; log--) {
+		for (bbno = omp->m_sb.sb_rbmblocks - 1;
+		     (xfs_srtblock_t)bbno >= 0;
+		     bbno--) {
+			error = xfs_rtget_summary(omp, tp, log, bbno, &bp,
+				&sumbno, &sum);
 			if (error)
-				goto error_cancel;
-			/*
-			 * Lock the bitmap inode.
-			 */
-			xfs_ilock(ip, XFS_ILOCK_EXCL);
-			xfs_trans_ijoin(tp, ip, XFS_ILOCK_EXCL);
-			/*
-			 * Get a buffer for the block.
-			 */
-			d = XFS_FSB_TO_DADDR(mp, fsbno);
-			bp = xfs_trans_get_buf(tp, mp->m_ddev_targp, d,
-				mp->m_bsize, 0);
-			if (bp == NULL) {
-				error = XFS_ERROR(EIO);
-error_cancel:
-				xfs_trans_cancel(tp, cancelflags);
-				goto error;
-			}
-			memset(bp->b_addr, 0, mp->m_sb.sb_blocksize);
-			xfs_trans_log_buf(tp, bp, 0, mp->m_sb.sb_blocksize - 1);
-			/*
-			 * Commit the transaction.
-			 */
-			error = xfs_trans_commit(tp, 0);
+				return error;
+			if (sum == 0)
+				continue;
+			error = xfs_rtmodify_summary(omp, tp, log, bbno, -sum,
+				&bp, &sumbno);
 			if (error)
-				goto error;
+				return error;
+			error = xfs_rtmodify_summary(nmp, tp, log, bbno, sum,
+				&bp, &sumbno);
+			if (error)
+				return error;
+			ASSERT(sum > 0);
 		}
-		/*
-		 * Go on to the next extent, if any.
-		 */
-		oblocks = map.br_startoff + map.br_blockcount;
 	}
 	return 0;
+}
+/*
+ * Mark an extent specified by start and len allocated.
+ * Updates all the summary information as well as the bitmap.
+ */
+STATIC int				/* error */
+xfs_rtallocate_range(
+	xfs_mount_t	*mp,		/* file system mount point */
+	xfs_trans_t	*tp,		/* transaction pointer */
+	xfs_rtblock_t	start,		/* start block to allocate */
+	xfs_extlen_t	len,		/* length to allocate */
+	xfs_buf_t	**rbpp,		/* in/out: summary block buffer */
+	xfs_fsblock_t	*rsb)		/* in/out: summary block number */
+{
+	xfs_rtblock_t	end;		/* end of the allocated extent */
+	int		error;		/* error value */
+	xfs_rtblock_t	postblock = 0;	/* first block allocated > end */
+	xfs_rtblock_t	preblock = 0;	/* first block allocated < start */
 
-error:
+	end = start + len - 1;
+	/*
+	 * Assume we're allocating out of the middle of a free extent.
+	 * We need to find the beginning and end of the extent so we can
+	 * properly update the summary.
+	 */
+	error = xfs_rtfind_back(mp, tp, start, 0, &preblock);
+	if (error) {
+		return error;
+	}
+	/*
+	 * Find the next allocated block (end of free extent).
+	 */
+	error = xfs_rtfind_forw(mp, tp, end, mp->m_sb.sb_rextents - 1,
+		&postblock);
+	if (error) {
+		return error;
+	}
+	/*
+	 * Decrement the summary information corresponding to the entire
+	 * (old) free extent.
+	 */
+	error = xfs_rtmodify_summary(mp, tp,
+		XFS_RTBLOCKLOG(postblock + 1 - preblock),
+		XFS_BITTOBLOCK(mp, preblock), -1, rbpp, rsb);
+	if (error) {
+		return error;
+	}
+	/*
+	 * If there are blocks not being allocated at the front of the
+	 * old extent, add summary data for them to be free.
+	 */
+	if (preblock < start) {
+		error = xfs_rtmodify_summary(mp, tp,
+			XFS_RTBLOCKLOG(start - preblock),
+			XFS_BITTOBLOCK(mp, preblock), 1, rbpp, rsb);
+		if (error) {
+			return error;
+		}
+	}
+	/*
+	 * If there are blocks not being allocated at the end of the
+	 * old extent, add summary data for them to be free.
+	 */
+	if (postblock > end) {
+		error = xfs_rtmodify_summary(mp, tp,
+			XFS_RTBLOCKLOG(postblock - end),
+			XFS_BITTOBLOCK(mp, end + 1), 1, rbpp, rsb);
+		if (error) {
+			return error;
+		}
+	}
+	/*
+	 * Modify the bitmap to mark this extent allocated.
+	 */
+	error = xfs_rtmodify_range(mp, tp, start, len, 0);
 	return error;
 }
 
@@ -721,1112 +809,126 @@ xfs_rtallocate_extent_size(
 }
 
 /*
- * Mark an extent specified by start and len allocated.
- * Updates all the summary information as well as the bitmap.
+ * Allocate space to the bitmap or summary file, and zero it, for growfs.
  */
 STATIC int				/* error */
-xfs_rtallocate_range(
+xfs_growfs_rt_alloc(
 	xfs_mount_t	*mp,		/* file system mount point */
-	xfs_trans_t	*tp,		/* transaction pointer */
-	xfs_rtblock_t	start,		/* start block to allocate */
-	xfs_extlen_t	len,		/* length to allocate */
-	xfs_buf_t	**rbpp,		/* in/out: summary block buffer */
-	xfs_fsblock_t	*rsb)		/* in/out: summary block number */
+	xfs_extlen_t	oblocks,	/* old count of blocks */
+	xfs_extlen_t	nblocks,	/* new count of blocks */
+	xfs_inode_t	*ip)		/* inode (bitmap/summary) */
 {
-	xfs_rtblock_t	end;		/* end of the allocated extent */
-	int		error;		/* error value */
-	xfs_rtblock_t	postblock = 0;	/* first block allocated > end */
-	xfs_rtblock_t	preblock = 0;	/* first block allocated < start */
+	xfs_fileoff_t	bno;		/* block number in file */
+	xfs_buf_t	*bp;		/* temporary buffer for zeroing */
+	int		committed;	/* transaction committed flag */
+	xfs_daddr_t	d;		/* disk block address */
+	int		error;		/* error return value */
+	xfs_fsblock_t	firstblock;	/* first block allocated in xaction */
+	xfs_bmap_free_t	flist;		/* list of freed blocks */
+	xfs_fsblock_t	fsbno;		/* filesystem block for bno */
+	xfs_bmbt_irec_t	map;		/* block map output */
+	int		nmap;		/* number of block maps */
+	int		resblks;	/* space reservation */
 
-	end = start + len - 1;
-	/*
-	 * Assume we're allocating out of the middle of a free extent.
-	 * We need to find the beginning and end of the extent so we can
-	 * properly update the summary.
-	 */
-	error = xfs_rtfind_back(mp, tp, start, 0, &preblock);
-	if (error) {
-		return error;
-	}
-	/*
-	 * Find the next allocated block (end of free extent).
-	 */
-	error = xfs_rtfind_forw(mp, tp, end, mp->m_sb.sb_rextents - 1,
-		&postblock);
-	if (error) {
-		return error;
-	}
-	/*
-	 * Decrement the summary information corresponding to the entire
-	 * (old) free extent.
-	 */
-	error = xfs_rtmodify_summary(mp, tp,
-		XFS_RTBLOCKLOG(postblock + 1 - preblock),
-		XFS_BITTOBLOCK(mp, preblock), -1, rbpp, rsb);
-	if (error) {
-		return error;
-	}
-	/*
-	 * If there are blocks not being allocated at the front of the
-	 * old extent, add summary data for them to be free.
-	 */
-	if (preblock < start) {
-		error = xfs_rtmodify_summary(mp, tp,
-			XFS_RTBLOCKLOG(start - preblock),
-			XFS_BITTOBLOCK(mp, preblock), 1, rbpp, rsb);
-		if (error) {
-			return error;
-		}
-	}
-	/*
-	 * If there are blocks not being allocated at the end of the
-	 * old extent, add summary data for them to be free.
-	 */
-	if (postblock > end) {
-		error = xfs_rtmodify_summary(mp, tp,
-			XFS_RTBLOCKLOG(postblock - end),
-			XFS_BITTOBLOCK(mp, end + 1), 1, rbpp, rsb);
-		if (error) {
-			return error;
-		}
-	}
 	/*
-	 * Modify the bitmap to mark this extent allocated.
+	 * Allocate space to the file, as necessary.
 	 */
-	error = xfs_rtmodify_range(mp, tp, start, len, 0);
-	return error;
-}
-
-/*
- * Return whether there are any free extents in the size range given
- * by low and high, for the bitmap block bbno.
- */
-STATIC int				/* error */
-xfs_rtany_summary(
-	xfs_mount_t	*mp,		/* file system mount structure */
-	xfs_trans_t	*tp,		/* transaction pointer */
-	int		low,		/* low log2 extent size */
-	int		high,		/* high log2 extent size */
-	xfs_rtblock_t	bbno,		/* bitmap block number */
-	xfs_buf_t	**rbpp,		/* in/out: summary block buffer */
-	xfs_fsblock_t	*rsb,		/* in/out: summary block number */
-	int		*stat)		/* out: any good extents here? */
-{
-	int		error;		/* error value */
-	int		log;		/* loop counter, log2 of ext. size */
-	xfs_suminfo_t	sum;		/* summary data */
+	while (oblocks < nblocks) {
+		int		cancelflags = 0;
+		xfs_trans_t	*tp;
 
-	/*
-	 * Loop over logs of extent sizes.  Order is irrelevant.
-	 */
-	for (log = low; log <= high; log++) {
+		tp = xfs_trans_alloc(mp, XFS_TRANS_GROWFSRT_ALLOC);
+		resblks = XFS_GROWFSRT_SPACE_RES(mp, nblocks - oblocks);
 		/*
-		 * Get one summary datum.
+		 * Reserve space & log for one extent added to the file.
 		 */
-		error = xfs_rtget_summary(mp, tp, log, bbno, rbpp, rsb, &sum);
-		if (error) {
-			return error;
-		}
+		error = xfs_trans_reserve(tp, &M_RES(mp)->tr_growdata,
+					  resblks, 0);
+		if (error)
+			goto error_cancel;
+		cancelflags = XFS_TRANS_RELEASE_LOG_RES;
 		/*
-		 * If there are any, return success.
+		 * Lock the inode.
 		 */
-		if (sum) {
-			*stat = 1;
-			return 0;
-		}
-	}
-	/*
-	 * Found nothing, return failure.
-	 */
-	*stat = 0;
-	return 0;
-}
-
-/*
- * Get a buffer for the bitmap or summary file block specified.
- * The buffer is returned read and locked.
- */
-STATIC int				/* error */
-xfs_rtbuf_get(
-	xfs_mount_t	*mp,		/* file system mount structure */
-	xfs_trans_t	*tp,		/* transaction pointer */
-	xfs_rtblock_t	block,		/* block number in bitmap or summary */
-	int		issum,		/* is summary not bitmap */
-	xfs_buf_t	**bpp)		/* output: buffer for the block */
-{
-	xfs_buf_t	*bp;		/* block buffer, result */
-	xfs_inode_t	*ip;		/* bitmap or summary inode */
-	xfs_bmbt_irec_t	map;
-	int		nmap = 1;
-	int		error;		/* error value */
-
-	ip = issum ? mp->m_rsumip : mp->m_rbmip;
-
-	error = xfs_bmapi_read(ip, block, 1, &map, &nmap, XFS_DATA_FORK);
-	if (error)
-		return error;
-
-	ASSERT(map.br_startblock != NULLFSBLOCK);
-	error = xfs_trans_read_buf(mp, tp, mp->m_ddev_targp,
-				   XFS_FSB_TO_DADDR(mp, map.br_startblock),
-				   mp->m_bsize, 0, &bp, NULL);
-	if (error)
-		return error;
-	ASSERT(!xfs_buf_geterror(bp));
-	*bpp = bp;
-	return 0;
-}
-
-#ifdef DEBUG
-/*
- * Check that the given extent (block range) is allocated already.
- */
-STATIC int				/* error */
-xfs_rtcheck_alloc_range(
-	xfs_mount_t	*mp,		/* file system mount point */
-	xfs_trans_t	*tp,		/* transaction pointer */
-	xfs_rtblock_t	bno,		/* starting block number of extent */
-	xfs_extlen_t	len,		/* length of extent */
-	int		*stat)		/* out: 1 for allocated, 0 for not */
-{
-	xfs_rtblock_t	new;		/* dummy for xfs_rtcheck_range */
-
-	return xfs_rtcheck_range(mp, tp, bno, len, 0, &new, stat);
-}
-#endif
-
-/*
- * Check that the given range is either all allocated (val = 0) or
- * all free (val = 1).
- */
-STATIC int				/* error */
-xfs_rtcheck_range(
-	xfs_mount_t	*mp,		/* file system mount point */
-	xfs_trans_t	*tp,		/* transaction pointer */
-	xfs_rtblock_t	start,		/* starting block number of extent */
-	xfs_extlen_t	len,		/* length of extent */
-	int		val,		/* 1 for free, 0 for allocated */
-	xfs_rtblock_t	*new,		/* out: first block not matching */
-	int		*stat)		/* out: 1 for matches, 0 for not */
-{
-	xfs_rtword_t	*b;		/* current word in buffer */
-	int		bit;		/* bit number in the word */
-	xfs_rtblock_t	block;		/* bitmap block number */
-	xfs_buf_t	*bp;		/* buf for the block */
-	xfs_rtword_t	*bufp;		/* starting word in buffer */
-	int		error;		/* error value */
-	xfs_rtblock_t	i;		/* current bit number rel. to start */
-	xfs_rtblock_t	lastbit;	/* last useful bit in word */
-	xfs_rtword_t	mask;		/* mask of relevant bits for value */
-	xfs_rtword_t	wdiff;		/* difference from wanted value */
-	int		word;		/* word number in the buffer */
+		xfs_ilock(ip, XFS_ILOCK_EXCL);
+		xfs_trans_ijoin(tp, ip, XFS_ILOCK_EXCL);
 
-	/*
-	 * Compute starting bitmap block number
-	 */
-	block = XFS_BITTOBLOCK(mp, start);
-	/*
-	 * Read the bitmap block.
-	 */
-	error = xfs_rtbuf_get(mp, tp, block, 0, &bp);
-	if (error) {
-		return error;
-	}
-	bufp = bp->b_addr;
-	/*
-	 * Compute the starting word's address, and starting bit.
-	 */
-	word = XFS_BITTOWORD(mp, start);
-	b = &bufp[word];
-	bit = (int)(start & (XFS_NBWORD - 1));
-	/*
-	 * 0 (allocated) => all zero's; 1 (free) => all one's.
-	 */
-	val = -val;
-	/*
-	 * If not starting on a word boundary, deal with the first
-	 * (partial) word.
-	 */
-	if (bit) {
-		/*
-		 * Compute first bit not examined.
-		 */
-		lastbit = XFS_RTMIN(bit + len, XFS_NBWORD);
-		/*
-		 * Mask of relevant bits.
-		 */
-		mask = (((xfs_rtword_t)1 << (lastbit - bit)) - 1) << bit;
-		/*
-		 * Compute difference between actual and desired value.
-		 */
-		if ((wdiff = (*b ^ val) & mask)) {
-			/*
-			 * Different, compute first wrong bit and return.
-			 */
-			xfs_trans_brelse(tp, bp);
-			i = XFS_RTLOBIT(wdiff) - bit;
-			*new = start + i;
-			*stat = 0;
-			return 0;
-		}
-		i = lastbit - bit;
-		/*
-		 * Go on to next block if that's where the next word is
-		 * and we need the next word.
-		 */
-		if (++word == XFS_BLOCKWSIZE(mp) && i < len) {
-			/*
-			 * If done with this block, get the next one.
-			 */
-			xfs_trans_brelse(tp, bp);
-			error = xfs_rtbuf_get(mp, tp, ++block, 0, &bp);
-			if (error) {
-				return error;
-			}
-			b = bufp = bp->b_addr;
-			word = 0;
-		} else {
-			/*
-			 * Go on to the next word in the buffer.
-			 */
-			b++;
-		}
-	} else {
-		/*
-		 * Starting on a word boundary, no partial word.
-		 */
-		i = 0;
-	}
-	/*
-	 * Loop over whole words in buffers.  When we use up one buffer
-	 * we move on to the next one.
-	 */
-	while (len - i >= XFS_NBWORD) {
-		/*
-		 * Compute difference between actual and desired value.
-		 */
-		if ((wdiff = *b ^ val)) {
-			/*
-			 * Different, compute first wrong bit and return.
-			 */
-			xfs_trans_brelse(tp, bp);
-			i += XFS_RTLOBIT(wdiff);
-			*new = start + i;
-			*stat = 0;
-			return 0;
-		}
-		i += XFS_NBWORD;
+		xfs_bmap_init(&flist, &firstblock);
 		/*
-		 * Go on to next block if that's where the next word is
-		 * and we need the next word.
+		 * Allocate blocks to the bitmap file.
 		 */
-		if (++word == XFS_BLOCKWSIZE(mp) && i < len) {
-			/*
-			 * If done with this block, get the next one.
-			 */
-			xfs_trans_brelse(tp, bp);
-			error = xfs_rtbuf_get(mp, tp, ++block, 0, &bp);
-			if (error) {
-				return error;
-			}
-			b = bufp = bp->b_addr;
-			word = 0;
-		} else {
-			/*
-			 * Go on to the next word in the buffer.
-			 */
-			b++;
-		}
-	}
-	/*
-	 * If not ending on a word boundary, deal with the last
-	 * (partial) word.
-	 */
-	if ((lastbit = len - i)) {
+		nmap = 1;
+		cancelflags |= XFS_TRANS_ABORT;
+		error = xfs_bmapi_write(tp, ip, oblocks, nblocks - oblocks,
+					XFS_BMAPI_METADATA, &firstblock,
+					resblks, &map, &nmap, &flist);
+		if (!error && nmap < 1)
+			error = XFS_ERROR(ENOSPC);
+		if (error)
+			goto error_cancel;
 		/*
-		 * Mask of relevant bits.
+		 * Free any blocks freed up in the transaction, then commit.
 		 */
-		mask = ((xfs_rtword_t)1 << lastbit) - 1;
+		error = xfs_bmap_finish(&tp, &flist, &committed);
+		if (error)
+			goto error_cancel;
+		error = xfs_trans_commit(tp, XFS_TRANS_RELEASE_LOG_RES);
+		if (error)
+			goto error;
 		/*
-		 * Compute difference between actual and desired value.
+		 * Now we need to clear the allocated blocks.
+		 * Do this one block per transaction, to keep it simple.
 		 */
-		if ((wdiff = (*b ^ val) & mask)) {
+		cancelflags = 0;
+		for (bno = map.br_startoff, fsbno = map.br_startblock;
+		     bno < map.br_startoff + map.br_blockcount;
+		     bno++, fsbno++) {
+			tp = xfs_trans_alloc(mp, XFS_TRANS_GROWFSRT_ZERO);
 			/*
-			 * Different, compute first wrong bit and return.
+			 * Reserve log for one block zeroing.
 			 */
-			xfs_trans_brelse(tp, bp);
-			i += XFS_RTLOBIT(wdiff);
-			*new = start + i;
-			*stat = 0;
-			return 0;
-		} else
-			i = len;
-	}
-	/*
-	 * Successful, return.
-	 */
-	xfs_trans_brelse(tp, bp);
-	*new = start + i;
-	*stat = 1;
-	return 0;
-}
-
-/*
- * Copy and transform the summary file, given the old and new
- * parameters in the mount structures.
- */
-STATIC int				/* error */
-xfs_rtcopy_summary(
-	xfs_mount_t	*omp,		/* old file system mount point */
-	xfs_mount_t	*nmp,		/* new file system mount point */
-	xfs_trans_t	*tp)		/* transaction pointer */
-{
-	xfs_rtblock_t	bbno;		/* bitmap block number */
-	xfs_buf_t	*bp;		/* summary buffer */
-	int		error;		/* error return value */
-	int		log;		/* summary level number (log length) */
-	xfs_suminfo_t	sum;		/* summary data */
-	xfs_fsblock_t	sumbno;		/* summary block number */
-
-	bp = NULL;
-	for (log = omp->m_rsumlevels - 1; log >= 0; log--) {
-		for (bbno = omp->m_sb.sb_rbmblocks - 1;
-		     (xfs_srtblock_t)bbno >= 0;
-		     bbno--) {
-			error = xfs_rtget_summary(omp, tp, log, bbno, &bp,
-				&sumbno, &sum);
-			if (error)
-				return error;
-			if (sum == 0)
-				continue;
-			error = xfs_rtmodify_summary(omp, tp, log, bbno, -sum,
-				&bp, &sumbno);
-			if (error)
-				return error;
-			error = xfs_rtmodify_summary(nmp, tp, log, bbno, sum,
-				&bp, &sumbno);
+			error = xfs_trans_reserve(tp, &M_RES(mp)->tr_growrtzero,
+						  0, 0);
 			if (error)
-				return error;
-			ASSERT(sum > 0);
-		}
-	}
-	return 0;
-}
-
-/*
- * Searching backward from start to limit, find the first block whose
- * allocated/free state is different from start's.
- */
-STATIC int				/* error */
-xfs_rtfind_back(
-	xfs_mount_t	*mp,		/* file system mount point */
-	xfs_trans_t	*tp,		/* transaction pointer */
-	xfs_rtblock_t	start,		/* starting block to look at */
-	xfs_rtblock_t	limit,		/* last block to look at */
-	xfs_rtblock_t	*rtblock)	/* out: start block found */
-{
-	xfs_rtword_t	*b;		/* current word in buffer */
-	int		bit;		/* bit number in the word */
-	xfs_rtblock_t	block;		/* bitmap block number */
-	xfs_buf_t	*bp;		/* buf for the block */
-	xfs_rtword_t	*bufp;		/* starting word in buffer */
-	int		error;		/* error value */
-	xfs_rtblock_t	firstbit;	/* first useful bit in the word */
-	xfs_rtblock_t	i;		/* current bit number rel. to start */
-	xfs_rtblock_t	len;		/* length of inspected area */
-	xfs_rtword_t	mask;		/* mask of relevant bits for value */
-	xfs_rtword_t	want;		/* mask for "good" values */
-	xfs_rtword_t	wdiff;		/* difference from wanted value */
-	int		word;		/* word number in the buffer */
-
-	/*
-	 * Compute and read in starting bitmap block for starting block.
-	 */
-	block = XFS_BITTOBLOCK(mp, start);
-	error = xfs_rtbuf_get(mp, tp, block, 0, &bp);
-	if (error) {
-		return error;
-	}
-	bufp = bp->b_addr;
-	/*
-	 * Get the first word's index & point to it.
-	 */
-	word = XFS_BITTOWORD(mp, start);
-	b = &bufp[word];
-	bit = (int)(start & (XFS_NBWORD - 1));
-	len = start - limit + 1;
-	/*
-	 * Compute match value, based on the bit at start: if 1 (free)
-	 * then all-ones, else all-zeroes.
-	 */
-	want = (*b & ((xfs_rtword_t)1 << bit)) ? -1 : 0;
-	/*
-	 * If the starting position is not word-aligned, deal with the
-	 * partial word.
-	 */
-	if (bit < XFS_NBWORD - 1) {
-		/*
-		 * Calculate first (leftmost) bit number to look at,
-		 * and mask for all the relevant bits in this word.
-		 */
-		firstbit = XFS_RTMAX((xfs_srtblock_t)(bit - len + 1), 0);
-		mask = (((xfs_rtword_t)1 << (bit - firstbit + 1)) - 1) <<
-			firstbit;
-		/*
-		 * Calculate the difference between the value there
-		 * and what we're looking for.
-		 */
-		if ((wdiff = (*b ^ want) & mask)) {
-			/*
-			 * Different.  Mark where we are and return.
-			 */
-			xfs_trans_brelse(tp, bp);
-			i = bit - XFS_RTHIBIT(wdiff);
-			*rtblock = start - i + 1;
-			return 0;
-		}
-		i = bit - firstbit + 1;
-		/*
-		 * Go on to previous block if that's where the previous word is
-		 * and we need the previous word.
-		 */
-		if (--word == -1 && i < len) {
-			/*
-			 * If done with this block, get the previous one.
-			 */
-			xfs_trans_brelse(tp, bp);
-			error = xfs_rtbuf_get(mp, tp, --block, 0, &bp);
-			if (error) {
-				return error;
-			}
-			bufp = bp->b_addr;
-			word = XFS_BLOCKWMASK(mp);
-			b = &bufp[word];
-		} else {
-			/*
-			 * Go on to the previous word in the buffer.
-			 */
-			b--;
-		}
-	} else {
-		/*
-		 * Starting on a word boundary, no partial word.
-		 */
-		i = 0;
-	}
-	/*
-	 * Loop over whole words in buffers.  When we use up one buffer
-	 * we move on to the previous one.
-	 */
-	while (len - i >= XFS_NBWORD) {
-		/*
-		 * Compute difference between actual and desired value.
-		 */
-		if ((wdiff = *b ^ want)) {
-			/*
-			 * Different, mark where we are and return.
-			 */
-			xfs_trans_brelse(tp, bp);
-			i += XFS_NBWORD - 1 - XFS_RTHIBIT(wdiff);
-			*rtblock = start - i + 1;
-			return 0;
-		}
-		i += XFS_NBWORD;
-		/*
-		 * Go on to previous block if that's where the previous word is
-		 * and we need the previous word.
-		 */
-		if (--word == -1 && i < len) {
-			/*
-			 * If done with this block, get the previous one.
-			 */
-			xfs_trans_brelse(tp, bp);
-			error = xfs_rtbuf_get(mp, tp, --block, 0, &bp);
-			if (error) {
-				return error;
-			}
-			bufp = bp->b_addr;
-			word = XFS_BLOCKWMASK(mp);
-			b = &bufp[word];
-		} else {
-			/*
-			 * Go on to the previous word in the buffer.
-			 */
-			b--;
-		}
-	}
-	/*
-	 * If not ending on a word boundary, deal with the last
-	 * (partial) word.
-	 */
-	if (len - i) {
-		/*
-		 * Calculate first (leftmost) bit number to look at,
-		 * and mask for all the relevant bits in this word.
-		 */
-		firstbit = XFS_NBWORD - (len - i);
-		mask = (((xfs_rtword_t)1 << (len - i)) - 1) << firstbit;
-		/*
-		 * Compute difference between actual and desired value.
-		 */
-		if ((wdiff = (*b ^ want) & mask)) {
-			/*
-			 * Different, mark where we are and return.
-			 */
-			xfs_trans_brelse(tp, bp);
-			i += XFS_NBWORD - 1 - XFS_RTHIBIT(wdiff);
-			*rtblock = start - i + 1;
-			return 0;
-		} else
-			i = len;
-	}
-	/*
-	 * No match, return that we scanned the whole area.
-	 */
-	xfs_trans_brelse(tp, bp);
-	*rtblock = start - i + 1;
-	return 0;
-}
-
-/*
- * Searching forward from start to limit, find the first block whose
- * allocated/free state is different from start's.
- */
-STATIC int				/* error */
-xfs_rtfind_forw(
-	xfs_mount_t	*mp,		/* file system mount point */
-	xfs_trans_t	*tp,		/* transaction pointer */
-	xfs_rtblock_t	start,		/* starting block to look at */
-	xfs_rtblock_t	limit,		/* last block to look at */
-	xfs_rtblock_t	*rtblock)	/* out: start block found */
-{
-	xfs_rtword_t	*b;		/* current word in buffer */
-	int		bit;		/* bit number in the word */
-	xfs_rtblock_t	block;		/* bitmap block number */
-	xfs_buf_t	*bp;		/* buf for the block */
-	xfs_rtword_t	*bufp;		/* starting word in buffer */
-	int		error;		/* error value */
-	xfs_rtblock_t	i;		/* current bit number rel. to start */
-	xfs_rtblock_t	lastbit;	/* last useful bit in the word */
-	xfs_rtblock_t	len;		/* length of inspected area */
-	xfs_rtword_t	mask;		/* mask of relevant bits for value */
-	xfs_rtword_t	want;		/* mask for "good" values */
-	xfs_rtword_t	wdiff;		/* difference from wanted value */
-	int		word;		/* word number in the buffer */
-
-	/*
-	 * Compute and read in starting bitmap block for starting block.
-	 */
-	block = XFS_BITTOBLOCK(mp, start);
-	error = xfs_rtbuf_get(mp, tp, block, 0, &bp);
-	if (error) {
-		return error;
-	}
-	bufp = bp->b_addr;
-	/*
-	 * Get the first word's index & point to it.
-	 */
-	word = XFS_BITTOWORD(mp, start);
-	b = &bufp[word];
-	bit = (int)(start & (XFS_NBWORD - 1));
-	len = limit - start + 1;
-	/*
-	 * Compute match value, based on the bit at start: if 1 (free)
-	 * then all-ones, else all-zeroes.
-	 */
-	want = (*b & ((xfs_rtword_t)1 << bit)) ? -1 : 0;
-	/*
-	 * If the starting position is not word-aligned, deal with the
-	 * partial word.
-	 */
-	if (bit) {
-		/*
-		 * Calculate last (rightmost) bit number to look at,
-		 * and mask for all the relevant bits in this word.
-		 */
-		lastbit = XFS_RTMIN(bit + len, XFS_NBWORD);
-		mask = (((xfs_rtword_t)1 << (lastbit - bit)) - 1) << bit;
-		/*
-		 * Calculate the difference between the value there
-		 * and what we're looking for.
-		 */
-		if ((wdiff = (*b ^ want) & mask)) {
-			/*
-			 * Different.  Mark where we are and return.
-			 */
-			xfs_trans_brelse(tp, bp);
-			i = XFS_RTLOBIT(wdiff) - bit;
-			*rtblock = start + i - 1;
-			return 0;
-		}
-		i = lastbit - bit;
-		/*
-		 * Go on to next block if that's where the next word is
-		 * and we need the next word.
-		 */
-		if (++word == XFS_BLOCKWSIZE(mp) && i < len) {
-			/*
-			 * If done with this block, get the previous one.
-			 */
-			xfs_trans_brelse(tp, bp);
-			error = xfs_rtbuf_get(mp, tp, ++block, 0, &bp);
-			if (error) {
-				return error;
-			}
-			b = bufp = bp->b_addr;
-			word = 0;
-		} else {
-			/*
-			 * Go on to the previous word in the buffer.
-			 */
-			b++;
-		}
-	} else {
-		/*
-		 * Starting on a word boundary, no partial word.
-		 */
-		i = 0;
-	}
-	/*
-	 * Loop over whole words in buffers.  When we use up one buffer
-	 * we move on to the next one.
-	 */
-	while (len - i >= XFS_NBWORD) {
-		/*
-		 * Compute difference between actual and desired value.
-		 */
-		if ((wdiff = *b ^ want)) {
+				goto error_cancel;
 			/*
-			 * Different, mark where we are and return.
+			 * Lock the bitmap inode.
 			 */
-			xfs_trans_brelse(tp, bp);
-			i += XFS_RTLOBIT(wdiff);
-			*rtblock = start + i - 1;
-			return 0;
-		}
-		i += XFS_NBWORD;
-		/*
-		 * Go on to next block if that's where the next word is
-		 * and we need the next word.
-		 */
-		if (++word == XFS_BLOCKWSIZE(mp) && i < len) {
+			xfs_ilock(ip, XFS_ILOCK_EXCL);
+			xfs_trans_ijoin(tp, ip, XFS_ILOCK_EXCL);
 			/*
-			 * If done with this block, get the next one.
+			 * Get a buffer for the block.
 			 */
-			xfs_trans_brelse(tp, bp);
-			error = xfs_rtbuf_get(mp, tp, ++block, 0, &bp);
-			if (error) {
-				return error;
+			d = XFS_FSB_TO_DADDR(mp, fsbno);
+			bp = xfs_trans_get_buf(tp, mp->m_ddev_targp, d,
+				mp->m_bsize, 0);
+			if (bp == NULL) {
+				error = XFS_ERROR(EIO);
+error_cancel:
+				xfs_trans_cancel(tp, cancelflags);
+				goto error;
 			}
-			b = bufp = bp->b_addr;
-			word = 0;
-		} else {
+			memset(bp->b_addr, 0, mp->m_sb.sb_blocksize);
+			xfs_trans_log_buf(tp, bp, 0, mp->m_sb.sb_blocksize - 1);
 			/*
-			 * Go on to the next word in the buffer.
+			 * Commit the transaction.
 			 */
-			b++;
+			error = xfs_trans_commit(tp, 0);
+			if (error)
+				goto error;
 		}
-	}
-	/*
-	 * If not ending on a word boundary, deal with the last
-	 * (partial) word.
-	 */
-	if ((lastbit = len - i)) {
-		/*
-		 * Calculate mask for all the relevant bits in this word.
-		 */
-		mask = ((xfs_rtword_t)1 << lastbit) - 1;
 		/*
-		 * Compute difference between actual and desired value.
+		 * Go on to the next extent, if any.
 		 */
-		if ((wdiff = (*b ^ want) & mask)) {
-			/*
-			 * Different, mark where we are and return.
-			 */
-			xfs_trans_brelse(tp, bp);
-			i += XFS_RTLOBIT(wdiff);
-			*rtblock = start + i - 1;
-			return 0;
-		} else
-			i = len;
+		oblocks = map.br_startoff + map.br_blockcount;
 	}
-	/*
-	 * No match, return that we scanned the whole area.
-	 */
-	xfs_trans_brelse(tp, bp);
-	*rtblock = start + i - 1;
 	return 0;
-}
 
-/*
- * Mark an extent specified by start and len freed.
- * Updates all the summary information as well as the bitmap.
- */
-STATIC int				/* error */
-xfs_rtfree_range(
-	xfs_mount_t	*mp,		/* file system mount point */
-	xfs_trans_t	*tp,		/* transaction pointer */
-	xfs_rtblock_t	start,		/* starting block to free */
-	xfs_extlen_t	len,		/* length to free */
-	xfs_buf_t	**rbpp,		/* in/out: summary block buffer */
-	xfs_fsblock_t	*rsb)		/* in/out: summary block number */
-{
-	xfs_rtblock_t	end;		/* end of the freed extent */
-	int		error;		/* error value */
-	xfs_rtblock_t	postblock;	/* first block freed > end */
-	xfs_rtblock_t	preblock;	/* first block freed < start */
-
-	end = start + len - 1;
-	/*
-	 * Modify the bitmap to mark this extent freed.
-	 */
-	error = xfs_rtmodify_range(mp, tp, start, len, 1);
-	if (error) {
-		return error;
-	}
-	/*
-	 * Assume we're freeing out of the middle of an allocated extent.
-	 * We need to find the beginning and end of the extent so we can
-	 * properly update the summary.
-	 */
-	error = xfs_rtfind_back(mp, tp, start, 0, &preblock);
-	if (error) {
-		return error;
-	}
-	/*
-	 * Find the next allocated block (end of allocated extent).
-	 */
-	error = xfs_rtfind_forw(mp, tp, end, mp->m_sb.sb_rextents - 1,
-		&postblock);
-	if (error)
-		return error;
-	/*
-	 * If there are blocks not being freed at the front of the
-	 * old extent, add summary data for them to be allocated.
-	 */
-	if (preblock < start) {
-		error = xfs_rtmodify_summary(mp, tp,
-			XFS_RTBLOCKLOG(start - preblock),
-			XFS_BITTOBLOCK(mp, preblock), -1, rbpp, rsb);
-		if (error) {
-			return error;
-		}
-	}
-	/*
-	 * If there are blocks not being freed at the end of the
-	 * old extent, add summary data for them to be allocated.
-	 */
-	if (postblock > end) {
-		error = xfs_rtmodify_summary(mp, tp,
-			XFS_RTBLOCKLOG(postblock - end),
-			XFS_BITTOBLOCK(mp, end + 1), -1, rbpp, rsb);
-		if (error) {
-			return error;
-		}
-	}
-	/*
-	 * Increment the summary information corresponding to the entire
-	 * (new) free extent.
-	 */
-	error = xfs_rtmodify_summary(mp, tp,
-		XFS_RTBLOCKLOG(postblock + 1 - preblock),
-		XFS_BITTOBLOCK(mp, preblock), 1, rbpp, rsb);
+error:
 	return error;
 }
 
 /*
- * Read and return the summary information for a given extent size,
- * bitmap block combination.
- * Keeps track of a current summary block, so we don't keep reading
- * it from the buffer cache.
- */
-STATIC int				/* error */
-xfs_rtget_summary(
-	xfs_mount_t	*mp,		/* file system mount structure */
-	xfs_trans_t	*tp,		/* transaction pointer */
-	int		log,		/* log2 of extent size */
-	xfs_rtblock_t	bbno,		/* bitmap block number */
-	xfs_buf_t	**rbpp,		/* in/out: summary block buffer */
-	xfs_fsblock_t	*rsb,		/* in/out: summary block number */
-	xfs_suminfo_t	*sum)		/* out: summary info for this block */
-{
-	xfs_buf_t	*bp;		/* buffer for summary block */
-	int		error;		/* error value */
-	xfs_fsblock_t	sb;		/* summary fsblock */
-	int		so;		/* index into the summary file */
-	xfs_suminfo_t	*sp;		/* pointer to returned data */
-
-	/*
-	 * Compute entry number in the summary file.
-	 */
-	so = XFS_SUMOFFS(mp, log, bbno);
-	/*
-	 * Compute the block number in the summary file.
-	 */
-	sb = XFS_SUMOFFSTOBLOCK(mp, so);
-	/*
-	 * If we have an old buffer, and the block number matches, use that.
-	 */
-	if (rbpp && *rbpp && *rsb == sb)
-		bp = *rbpp;
-	/*
-	 * Otherwise we have to get the buffer.
-	 */
-	else {
-		/*
-		 * If there was an old one, get rid of it first.
-		 */
-		if (rbpp && *rbpp)
-			xfs_trans_brelse(tp, *rbpp);
-		error = xfs_rtbuf_get(mp, tp, sb, 1, &bp);
-		if (error) {
-			return error;
-		}
-		/*
-		 * Remember this buffer and block for the next call.
-		 */
-		if (rbpp) {
-			*rbpp = bp;
-			*rsb = sb;
-		}
-	}
-	/*
-	 * Point to the summary information & copy it out.
-	 */
-	sp = XFS_SUMPTR(mp, bp, so);
-	*sum = *sp;
-	/*
-	 * Drop the buffer if we're not asked to remember it.
-	 */
-	if (!rbpp)
-		xfs_trans_brelse(tp, bp);
-	return 0;
-}
-
-/*
- * Set the given range of bitmap bits to the given value.
- * Do whatever I/O and logging is required.
- */
-STATIC int				/* error */
-xfs_rtmodify_range(
-	xfs_mount_t	*mp,		/* file system mount point */
-	xfs_trans_t	*tp,		/* transaction pointer */
-	xfs_rtblock_t	start,		/* starting block to modify */
-	xfs_extlen_t	len,		/* length of extent to modify */
-	int		val)		/* 1 for free, 0 for allocated */
-{
-	xfs_rtword_t	*b;		/* current word in buffer */
-	int		bit;		/* bit number in the word */
-	xfs_rtblock_t	block;		/* bitmap block number */
-	xfs_buf_t	*bp;		/* buf for the block */
-	xfs_rtword_t	*bufp;		/* starting word in buffer */
-	int		error;		/* error value */
-	xfs_rtword_t	*first;		/* first used word in the buffer */
-	int		i;		/* current bit number rel. to start */
-	int		lastbit;	/* last useful bit in word */
-	xfs_rtword_t	mask;		/* mask o frelevant bits for value */
-	int		word;		/* word number in the buffer */
-
-	/*
-	 * Compute starting bitmap block number.
-	 */
-	block = XFS_BITTOBLOCK(mp, start);
-	/*
-	 * Read the bitmap block, and point to its data.
-	 */
-	error = xfs_rtbuf_get(mp, tp, block, 0, &bp);
-	if (error) {
-		return error;
-	}
-	bufp = bp->b_addr;
-	/*
-	 * Compute the starting word's address, and starting bit.
-	 */
-	word = XFS_BITTOWORD(mp, start);
-	first = b = &bufp[word];
-	bit = (int)(start & (XFS_NBWORD - 1));
-	/*
-	 * 0 (allocated) => all zeroes; 1 (free) => all ones.
-	 */
-	val = -val;
-	/*
-	 * If not starting on a word boundary, deal with the first
-	 * (partial) word.
-	 */
-	if (bit) {
-		/*
-		 * Compute first bit not changed and mask of relevant bits.
-		 */
-		lastbit = XFS_RTMIN(bit + len, XFS_NBWORD);
-		mask = (((xfs_rtword_t)1 << (lastbit - bit)) - 1) << bit;
-		/*
-		 * Set/clear the active bits.
-		 */
-		if (val)
-			*b |= mask;
-		else
-			*b &= ~mask;
-		i = lastbit - bit;
-		/*
-		 * Go on to the next block if that's where the next word is
-		 * and we need the next word.
-		 */
-		if (++word == XFS_BLOCKWSIZE(mp) && i < len) {
-			/*
-			 * Log the changed part of this block.
-			 * Get the next one.
-			 */
-			xfs_trans_log_buf(tp, bp,
-				(uint)((char *)first - (char *)bufp),
-				(uint)((char *)b - (char *)bufp));
-			error = xfs_rtbuf_get(mp, tp, ++block, 0, &bp);
-			if (error) {
-				return error;
-			}
-			first = b = bufp = bp->b_addr;
-			word = 0;
-		} else {
-			/*
-			 * Go on to the next word in the buffer
-			 */
-			b++;
-		}
-	} else {
-		/*
-		 * Starting on a word boundary, no partial word.
-		 */
-		i = 0;
-	}
-	/*
-	 * Loop over whole words in buffers.  When we use up one buffer
-	 * we move on to the next one.
-	 */
-	while (len - i >= XFS_NBWORD) {
-		/*
-		 * Set the word value correctly.
-		 */
-		*b = val;
-		i += XFS_NBWORD;
-		/*
-		 * Go on to the next block if that's where the next word is
-		 * and we need the next word.
-		 */
-		if (++word == XFS_BLOCKWSIZE(mp) && i < len) {
-			/*
-			 * Log the changed part of this block.
-			 * Get the next one.
-			 */
-			xfs_trans_log_buf(tp, bp,
-				(uint)((char *)first - (char *)bufp),
-				(uint)((char *)b - (char *)bufp));
-			error = xfs_rtbuf_get(mp, tp, ++block, 0, &bp);
-			if (error) {
-				return error;
-			}
-			first = b = bufp = bp->b_addr;
-			word = 0;
-		} else {
-			/*
-			 * Go on to the next word in the buffer
-			 */
-			b++;
-		}
-	}
-	/*
-	 * If not ending on a word boundary, deal with the last
-	 * (partial) word.
-	 */
-	if ((lastbit = len - i)) {
-		/*
-		 * Compute a mask of relevant bits.
-		 */
-		bit = 0;
-		mask = ((xfs_rtword_t)1 << lastbit) - 1;
-		/*
-		 * Set/clear the active bits.
-		 */
-		if (val)
-			*b |= mask;
-		else
-			*b &= ~mask;
-		b++;
-	}
-	/*
-	 * Log any remaining changed bytes.
-	 */
-	if (b > first)
-		xfs_trans_log_buf(tp, bp, (uint)((char *)first - (char *)bufp),
-			(uint)((char *)b - (char *)bufp - 1));
-	return 0;
-}
-
-/*
- * Read and modify the summary information for a given extent size,
- * bitmap block combination.
- * Keeps track of a current summary block, so we don't keep reading
- * it from the buffer cache.
- */
-STATIC int				/* error */
-xfs_rtmodify_summary(
-	xfs_mount_t	*mp,		/* file system mount point */
-	xfs_trans_t	*tp,		/* transaction pointer */
-	int		log,		/* log2 of extent size */
-	xfs_rtblock_t	bbno,		/* bitmap block number */
-	int		delta,		/* change to make to summary info */
-	xfs_buf_t	**rbpp,		/* in/out: summary block buffer */
-	xfs_fsblock_t	*rsb)		/* in/out: summary block number */
-{
-	xfs_buf_t	*bp;		/* buffer for the summary block */
-	int		error;		/* error value */
-	xfs_fsblock_t	sb;		/* summary fsblock */
-	int		so;		/* index into the summary file */
-	xfs_suminfo_t	*sp;		/* pointer to returned data */
-
-	/*
-	 * Compute entry number in the summary file.
-	 */
-	so = XFS_SUMOFFS(mp, log, bbno);
-	/*
-	 * Compute the block number in the summary file.
-	 */
-	sb = XFS_SUMOFFSTOBLOCK(mp, so);
-	/*
-	 * If we have an old buffer, and the block number matches, use that.
-	 */
-	if (rbpp && *rbpp && *rsb == sb)
-		bp = *rbpp;
-	/*
-	 * Otherwise we have to get the buffer.
-	 */
-	else {
-		/*
-		 * If there was an old one, get rid of it first.
-		 */
-		if (rbpp && *rbpp)
-			xfs_trans_brelse(tp, *rbpp);
-		error = xfs_rtbuf_get(mp, tp, sb, 1, &bp);
-		if (error) {
-			return error;
-		}
-		/*
-		 * Remember this buffer and block for the next call.
-		 */
-		if (rbpp) {
-			*rbpp = bp;
-			*rsb = sb;
-		}
-	}
-	/*
-	 * Point to the summary information, modify and log it.
-	 */
-	sp = XFS_SUMPTR(mp, bp, so);
-	*sp += delta;
-	xfs_trans_log_buf(tp, bp, (uint)((char *)sp - (char *)bp->b_addr),
-		(uint)((char *)sp - (char *)bp->b_addr + sizeof(*sp) - 1));
-	return 0;
-}
-
-/*
  * Visible (exported) functions.
  */
 
@@ -2129,66 +1231,6 @@ xfs_rtallocate_extent(
 }
 
 /*
- * Free an extent in the realtime subvolume.  Length is expressed in
- * realtime extents, as is the block number.
- */
-int					/* error */
-xfs_rtfree_extent(
-	xfs_trans_t	*tp,		/* transaction pointer */
-	xfs_rtblock_t	bno,		/* starting block number to free */
-	xfs_extlen_t	len)		/* length of extent freed */
-{
-	int		error;		/* error value */
-	xfs_mount_t	*mp;		/* file system mount structure */
-	xfs_fsblock_t	sb;		/* summary file block number */
-	xfs_buf_t	*sumbp;		/* summary file block buffer */
-
-	mp = tp->t_mountp;
-
-	ASSERT(mp->m_rbmip->i_itemp != NULL);
-	ASSERT(xfs_isilocked(mp->m_rbmip, XFS_ILOCK_EXCL));
-
-#ifdef DEBUG
-	/*
-	 * Check to see that this whole range is currently allocated.
-	 */
-	{
-		int	stat;		/* result from checking range */
-
-		error = xfs_rtcheck_alloc_range(mp, tp, bno, len, &stat);
-		if (error) {
-			return error;
-		}
-		ASSERT(stat);
-	}
-#endif
-	sumbp = NULL;
-	/*
-	 * Free the range of realtime blocks.
-	 */
-	error = xfs_rtfree_range(mp, tp, bno, len, &sumbp, &sb);
-	if (error) {
-		return error;
-	}
-	/*
-	 * Mark more blocks free in the superblock.
-	 */
-	xfs_trans_mod_sb(tp, XFS_TRANS_SB_FREXTENTS, (long)len);
-	/*
-	 * If we've now freed all the blocks, reset the file sequence
-	 * number to 0.
-	 */
-	if (tp->t_frextents_delta + mp->m_sb.sb_frextents ==
-	    mp->m_sb.sb_rextents) {
-		if (!(mp->m_rbmip->i_d.di_flags & XFS_DIFLAG_NEWRTBM))
-			mp->m_rbmip->i_d.di_flags |= XFS_DIFLAG_NEWRTBM;
-		*(__uint64_t *)&mp->m_rbmip->i_d.di_atime = 0;
-		xfs_trans_log_inode(tp, mp->m_rbmip, XFS_ILOG_CORE);
-	}
-	return 0;
-}
-
-/*
  * Initialize realtime fields in the mount structure.
  */
 int				/* error */
diff --git a/fs/xfs/xfs_rtalloc.h b/fs/xfs/xfs_rtalloc.h
index b2a1a24c0e2f..752b63d10300 100644
--- a/fs/xfs/xfs_rtalloc.h
+++ b/fs/xfs/xfs_rtalloc.h
@@ -95,6 +95,30 @@ xfs_growfs_rt(
 	struct xfs_mount	*mp,	/* file system mount structure */
 	xfs_growfs_rt_t		*in);	/* user supplied growfs struct */
 
+/*
+ * From xfs_rtbitmap.c
+ */
+int xfs_rtbuf_get(struct xfs_mount *mp, struct xfs_trans *tp,
+		  xfs_rtblock_t block, int issum, struct xfs_buf **bpp);
+int xfs_rtcheck_range(struct xfs_mount *mp, struct xfs_trans *tp,
+		      xfs_rtblock_t start, xfs_extlen_t len, int val,
+		      xfs_rtblock_t *new, int *stat);
+int xfs_rtfind_back(struct xfs_mount *mp, struct xfs_trans *tp,
+		    xfs_rtblock_t start, xfs_rtblock_t limit,
+		    xfs_rtblock_t *rtblock);
+int xfs_rtfind_forw(struct xfs_mount *mp, struct xfs_trans *tp,
+		    xfs_rtblock_t start, xfs_rtblock_t limit,
+		    xfs_rtblock_t *rtblock);
+int xfs_rtmodify_range(struct xfs_mount *mp, struct xfs_trans *tp,
+		       xfs_rtblock_t start, xfs_extlen_t len, int val);
+int xfs_rtmodify_summary(struct xfs_mount *mp, struct xfs_trans *tp, int log,
+			 xfs_rtblock_t bbno, int delta, xfs_buf_t **rbpp,
+			 xfs_fsblock_t *rsb);
+int xfs_rtfree_range(struct xfs_mount *mp, struct xfs_trans *tp,
+		     xfs_rtblock_t start, xfs_extlen_t len,
+		     struct xfs_buf **rbpp, xfs_fsblock_t *rsb);
+
+
 #else
 # define xfs_rtallocate_extent(t,b,min,max,l,a,f,p,rb)  (ENOSYS)
 # define xfs_rtfree_extent(t,b,l)                       (ENOSYS)
diff --git a/fs/xfs/xfs_rtbitmap.c b/fs/xfs/xfs_rtbitmap.c
new file mode 100644
index 000000000000..b1f2fe8af4a8
--- /dev/null
+++ b/fs/xfs/xfs_rtbitmap.c
@@ -0,0 +1,974 @@
+/*
+ * Copyright (c) 2000-2005 Silicon Graphics, Inc.
+ * All Rights Reserved.
+ *
+ * This program is free software; you can redistribute it and/or
+ * modify it under the terms of the GNU General Public License as
+ * published by the Free Software Foundation.
+ *
+ * This program is distributed in the hope that it would be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+ * GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program; if not, write the Free Software Foundation,
+ * Inc.,  51 Franklin St, Fifth Floor, Boston, MA  02110-1301  USA
+ */
+#include "xfs.h"
+#include "xfs_fs.h"
+#include "xfs_shared.h"
+#include "xfs_format.h"
+#include "xfs_log_format.h"
+#include "xfs_trans_resv.h"
+#include "xfs_bit.h"
+#include "xfs_sb.h"
+#include "xfs_ag.h"
+#include "xfs_mount.h"
+#include "xfs_inode.h"
+#include "xfs_bmap.h"
+#include "xfs_bmap_util.h"
+#include "xfs_bmap_btree.h"
+#include "xfs_alloc.h"
+#include "xfs_error.h"
+#include "xfs_trans.h"
+#include "xfs_trans_space.h"
+#include "xfs_trace.h"
+#include "xfs_buf.h"
+#include "xfs_icache.h"
+#include "xfs_dinode.h"
+#include "xfs_rtalloc.h"
+
+
+/*
+ * Realtime allocator bitmap functions shared with userspace.
+ */
+
+/*
+ * Get a buffer for the bitmap or summary file block specified.
+ * The buffer is returned read and locked.
+ */
+int
+xfs_rtbuf_get(
+	xfs_mount_t	*mp,		/* file system mount structure */
+	xfs_trans_t	*tp,		/* transaction pointer */
+	xfs_rtblock_t	block,		/* block number in bitmap or summary */
+	int		issum,		/* is summary not bitmap */
+	xfs_buf_t	**bpp)		/* output: buffer for the block */
+{
+	xfs_buf_t	*bp;		/* block buffer, result */
+	xfs_inode_t	*ip;		/* bitmap or summary inode */
+	xfs_bmbt_irec_t	map;
+	int		nmap = 1;
+	int		error;		/* error value */
+
+	ip = issum ? mp->m_rsumip : mp->m_rbmip;
+
+	error = xfs_bmapi_read(ip, block, 1, &map, &nmap, XFS_DATA_FORK);
+	if (error)
+		return error;
+
+	ASSERT(map.br_startblock != NULLFSBLOCK);
+	error = xfs_trans_read_buf(mp, tp, mp->m_ddev_targp,
+				   XFS_FSB_TO_DADDR(mp, map.br_startblock),
+				   mp->m_bsize, 0, &bp, NULL);
+	if (error)
+		return error;
+	ASSERT(!xfs_buf_geterror(bp));
+	*bpp = bp;
+	return 0;
+}
+
+/*
+ * Searching backward from start to limit, find the first block whose
+ * allocated/free state is different from start's.
+ */
+int
+xfs_rtfind_back(
+	xfs_mount_t	*mp,		/* file system mount point */
+	xfs_trans_t	*tp,		/* transaction pointer */
+	xfs_rtblock_t	start,		/* starting block to look at */
+	xfs_rtblock_t	limit,		/* last block to look at */
+	xfs_rtblock_t	*rtblock)	/* out: start block found */
+{
+	xfs_rtword_t	*b;		/* current word in buffer */
+	int		bit;		/* bit number in the word */
+	xfs_rtblock_t	block;		/* bitmap block number */
+	xfs_buf_t	*bp;		/* buf for the block */
+	xfs_rtword_t	*bufp;		/* starting word in buffer */
+	int		error;		/* error value */
+	xfs_rtblock_t	firstbit;	/* first useful bit in the word */
+	xfs_rtblock_t	i;		/* current bit number rel. to start */
+	xfs_rtblock_t	len;		/* length of inspected area */
+	xfs_rtword_t	mask;		/* mask of relevant bits for value */
+	xfs_rtword_t	want;		/* mask for "good" values */
+	xfs_rtword_t	wdiff;		/* difference from wanted value */
+	int		word;		/* word number in the buffer */
+
+	/*
+	 * Compute and read in starting bitmap block for starting block.
+	 */
+	block = XFS_BITTOBLOCK(mp, start);
+	error = xfs_rtbuf_get(mp, tp, block, 0, &bp);
+	if (error) {
+		return error;
+	}
+	bufp = bp->b_addr;
+	/*
+	 * Get the first word's index & point to it.
+	 */
+	word = XFS_BITTOWORD(mp, start);
+	b = &bufp[word];
+	bit = (int)(start & (XFS_NBWORD - 1));
+	len = start - limit + 1;
+	/*
+	 * Compute match value, based on the bit at start: if 1 (free)
+	 * then all-ones, else all-zeroes.
+	 */
+	want = (*b & ((xfs_rtword_t)1 << bit)) ? -1 : 0;
+	/*
+	 * If the starting position is not word-aligned, deal with the
+	 * partial word.
+	 */
+	if (bit < XFS_NBWORD - 1) {
+		/*
+		 * Calculate first (leftmost) bit number to look at,
+		 * and mask for all the relevant bits in this word.
+		 */
+		firstbit = XFS_RTMAX((xfs_srtblock_t)(bit - len + 1), 0);
+		mask = (((xfs_rtword_t)1 << (bit - firstbit + 1)) - 1) <<
+			firstbit;
+		/*
+		 * Calculate the difference between the value there
+		 * and what we're looking for.
+		 */
+		if ((wdiff = (*b ^ want) & mask)) {
+			/*
+			 * Different.  Mark where we are and return.
+			 */
+			xfs_trans_brelse(tp, bp);
+			i = bit - XFS_RTHIBIT(wdiff);
+			*rtblock = start - i + 1;
+			return 0;
+		}
+		i = bit - firstbit + 1;
+		/*
+		 * Go on to previous block if that's where the previous word is
+		 * and we need the previous word.
+		 */
+		if (--word == -1 && i < len) {
+			/*
+			 * If done with this block, get the previous one.
+			 */
+			xfs_trans_brelse(tp, bp);
+			error = xfs_rtbuf_get(mp, tp, --block, 0, &bp);
+			if (error) {
+				return error;
+			}
+			bufp = bp->b_addr;
+			word = XFS_BLOCKWMASK(mp);
+			b = &bufp[word];
+		} else {
+			/*
+			 * Go on to the previous word in the buffer.
+			 */
+			b--;
+		}
+	} else {
+		/*
+		 * Starting on a word boundary, no partial word.
+		 */
+		i = 0;
+	}
+	/*
+	 * Loop over whole words in buffers.  When we use up one buffer
+	 * we move on to the previous one.
+	 */
+	while (len - i >= XFS_NBWORD) {
+		/*
+		 * Compute difference between actual and desired value.
+		 */
+		if ((wdiff = *b ^ want)) {
+			/*
+			 * Different, mark where we are and return.
+			 */
+			xfs_trans_brelse(tp, bp);
+			i += XFS_NBWORD - 1 - XFS_RTHIBIT(wdiff);
+			*rtblock = start - i + 1;
+			return 0;
+		}
+		i += XFS_NBWORD;
+		/*
+		 * Go on to previous block if that's where the previous word is
+		 * and we need the previous word.
+		 */
+		if (--word == -1 && i < len) {
+			/*
+			 * If done with this block, get the previous one.
+			 */
+			xfs_trans_brelse(tp, bp);
+			error = xfs_rtbuf_get(mp, tp, --block, 0, &bp);
+			if (error) {
+				return error;
+			}
+			bufp = bp->b_addr;
+			word = XFS_BLOCKWMASK(mp);
+			b = &bufp[word];
+		} else {
+			/*
+			 * Go on to the previous word in the buffer.
+			 */
+			b--;
+		}
+	}
+	/*
+	 * If not ending on a word boundary, deal with the last
+	 * (partial) word.
+	 */
+	if (len - i) {
+		/*
+		 * Calculate first (leftmost) bit number to look at,
+		 * and mask for all the relevant bits in this word.
+		 */
+		firstbit = XFS_NBWORD - (len - i);
+		mask = (((xfs_rtword_t)1 << (len - i)) - 1) << firstbit;
+		/*
+		 * Compute difference between actual and desired value.
+		 */
+		if ((wdiff = (*b ^ want) & mask)) {
+			/*
+			 * Different, mark where we are and return.
+			 */
+			xfs_trans_brelse(tp, bp);
+			i += XFS_NBWORD - 1 - XFS_RTHIBIT(wdiff);
+			*rtblock = start - i + 1;
+			return 0;
+		} else
+			i = len;
+	}
+	/*
+	 * No match, return that we scanned the whole area.
+	 */
+	xfs_trans_brelse(tp, bp);
+	*rtblock = start - i + 1;
+	return 0;
+}
+
+/*
+ * Searching forward from start to limit, find the first block whose
+ * allocated/free state is different from start's.
+ */
+int
+xfs_rtfind_forw(
+	xfs_mount_t	*mp,		/* file system mount point */
+	xfs_trans_t	*tp,		/* transaction pointer */
+	xfs_rtblock_t	start,		/* starting block to look at */
+	xfs_rtblock_t	limit,		/* last block to look at */
+	xfs_rtblock_t	*rtblock)	/* out: start block found */
+{
+	xfs_rtword_t	*b;		/* current word in buffer */
+	int		bit;		/* bit number in the word */
+	xfs_rtblock_t	block;		/* bitmap block number */
+	xfs_buf_t	*bp;		/* buf for the block */
+	xfs_rtword_t	*bufp;		/* starting word in buffer */
+	int		error;		/* error value */
+	xfs_rtblock_t	i;		/* current bit number rel. to start */
+	xfs_rtblock_t	lastbit;	/* last useful bit in the word */
+	xfs_rtblock_t	len;		/* length of inspected area */
+	xfs_rtword_t	mask;		/* mask of relevant bits for value */
+	xfs_rtword_t	want;		/* mask for "good" values */
+	xfs_rtword_t	wdiff;		/* difference from wanted value */
+	int		word;		/* word number in the buffer */
+
+	/*
+	 * Compute and read in starting bitmap block for starting block.
+	 */
+	block = XFS_BITTOBLOCK(mp, start);
+	error = xfs_rtbuf_get(mp, tp, block, 0, &bp);
+	if (error) {
+		return error;
+	}
+	bufp = bp->b_addr;
+	/*
+	 * Get the first word's index & point to it.
+	 */
+	word = XFS_BITTOWORD(mp, start);
+	b = &bufp[word];
+	bit = (int)(start & (XFS_NBWORD - 1));
+	len = limit - start + 1;
+	/*
+	 * Compute match value, based on the bit at start: if 1 (free)
+	 * then all-ones, else all-zeroes.
+	 */
+	want = (*b & ((xfs_rtword_t)1 << bit)) ? -1 : 0;
+	/*
+	 * If the starting position is not word-aligned, deal with the
+	 * partial word.
+	 */
+	if (bit) {
+		/*
+		 * Calculate last (rightmost) bit number to look at,
+		 * and mask for all the relevant bits in this word.
+		 */
+		lastbit = XFS_RTMIN(bit + len, XFS_NBWORD);
+		mask = (((xfs_rtword_t)1 << (lastbit - bit)) - 1) << bit;
+		/*
+		 * Calculate the difference between the value there
+		 * and what we're looking for.
+		 */
+		if ((wdiff = (*b ^ want) & mask)) {
+			/*
+			 * Different.  Mark where we are and return.
+			 */
+			xfs_trans_brelse(tp, bp);
+			i = XFS_RTLOBIT(wdiff) - bit;
+			*rtblock = start + i - 1;
+			return 0;
+		}
+		i = lastbit - bit;
+		/*
+		 * Go on to next block if that's where the next word is
+		 * and we need the next word.
+		 */
+		if (++word == XFS_BLOCKWSIZE(mp) && i < len) {
+			/*
+			 * If done with this block, get the previous one.
+			 */
+			xfs_trans_brelse(tp, bp);
+			error = xfs_rtbuf_get(mp, tp, ++block, 0, &bp);
+			if (error) {
+				return error;
+			}
+			b = bufp = bp->b_addr;
+			word = 0;
+		} else {
+			/*
+			 * Go on to the previous word in the buffer.
+			 */
+			b++;
+		}
+	} else {
+		/*
+		 * Starting on a word boundary, no partial word.
+		 */
+		i = 0;
+	}
+	/*
+	 * Loop over whole words in buffers.  When we use up one buffer
+	 * we move on to the next one.
+	 */
+	while (len - i >= XFS_NBWORD) {
+		/*
+		 * Compute difference between actual and desired value.
+		 */
+		if ((wdiff = *b ^ want)) {
+			/*
+			 * Different, mark where we are and return.
+			 */
+			xfs_trans_brelse(tp, bp);
+			i += XFS_RTLOBIT(wdiff);
+			*rtblock = start + i - 1;
+			return 0;
+		}
+		i += XFS_NBWORD;
+		/*
+		 * Go on to next block if that's where the next word is
+		 * and we need the next word.
+		 */
+		if (++word == XFS_BLOCKWSIZE(mp) && i < len) {
+			/*
+			 * If done with this block, get the next one.
+			 */
+			xfs_trans_brelse(tp, bp);
+			error = xfs_rtbuf_get(mp, tp, ++block, 0, &bp);
+			if (error) {
+				return error;
+			}
+			b = bufp = bp->b_addr;
+			word = 0;
+		} else {
+			/*
+			 * Go on to the next word in the buffer.
+			 */
+			b++;
+		}
+	}
+	/*
+	 * If not ending on a word boundary, deal with the last
+	 * (partial) word.
+	 */
+	if ((lastbit = len - i)) {
+		/*
+		 * Calculate mask for all the relevant bits in this word.
+		 */
+		mask = ((xfs_rtword_t)1 << lastbit) - 1;
+		/*
+		 * Compute difference between actual and desired value.
+		 */
+		if ((wdiff = (*b ^ want) & mask)) {
+			/*
+			 * Different, mark where we are and return.
+			 */
+			xfs_trans_brelse(tp, bp);
+			i += XFS_RTLOBIT(wdiff);
+			*rtblock = start + i - 1;
+			return 0;
+		} else
+			i = len;
+	}
+	/*
+	 * No match, return that we scanned the whole area.
+	 */
+	xfs_trans_brelse(tp, bp);
+	*rtblock = start + i - 1;
+	return 0;
+}
+
+/*
+ * Read and modify the summary information for a given extent size,
+ * bitmap block combination.
+ * Keeps track of a current summary block, so we don't keep reading
+ * it from the buffer cache.
+ */
+int
+xfs_rtmodify_summary(
+	xfs_mount_t	*mp,		/* file system mount point */
+	xfs_trans_t	*tp,		/* transaction pointer */
+	int		log,		/* log2 of extent size */
+	xfs_rtblock_t	bbno,		/* bitmap block number */
+	int		delta,		/* change to make to summary info */
+	xfs_buf_t	**rbpp,		/* in/out: summary block buffer */
+	xfs_fsblock_t	*rsb)		/* in/out: summary block number */
+{
+	xfs_buf_t	*bp;		/* buffer for the summary block */
+	int		error;		/* error value */
+	xfs_fsblock_t	sb;		/* summary fsblock */
+	int		so;		/* index into the summary file */
+	xfs_suminfo_t	*sp;		/* pointer to returned data */
+
+	/*
+	 * Compute entry number in the summary file.
+	 */
+	so = XFS_SUMOFFS(mp, log, bbno);
+	/*
+	 * Compute the block number in the summary file.
+	 */
+	sb = XFS_SUMOFFSTOBLOCK(mp, so);
+	/*
+	 * If we have an old buffer, and the block number matches, use that.
+	 */
+	if (rbpp && *rbpp && *rsb == sb)
+		bp = *rbpp;
+	/*
+	 * Otherwise we have to get the buffer.
+	 */
+	else {
+		/*
+		 * If there was an old one, get rid of it first.
+		 */
+		if (rbpp && *rbpp)
+			xfs_trans_brelse(tp, *rbpp);
+		error = xfs_rtbuf_get(mp, tp, sb, 1, &bp);
+		if (error) {
+			return error;
+		}
+		/*
+		 * Remember this buffer and block for the next call.
+		 */
+		if (rbpp) {
+			*rbpp = bp;
+			*rsb = sb;
+		}
+	}
+	/*
+	 * Point to the summary information, modify and log it.
+	 */
+	sp = XFS_SUMPTR(mp, bp, so);
+	*sp += delta;
+	xfs_trans_log_buf(tp, bp, (uint)((char *)sp - (char *)bp->b_addr),
+		(uint)((char *)sp - (char *)bp->b_addr + sizeof(*sp) - 1));
+	return 0;
+}
+
+/*
+ * Set the given range of bitmap bits to the given value.
+ * Do whatever I/O and logging is required.
+ */
+int
+xfs_rtmodify_range(
+	xfs_mount_t	*mp,		/* file system mount point */
+	xfs_trans_t	*tp,		/* transaction pointer */
+	xfs_rtblock_t	start,		/* starting block to modify */
+	xfs_extlen_t	len,		/* length of extent to modify */
+	int		val)		/* 1 for free, 0 for allocated */
+{
+	xfs_rtword_t	*b;		/* current word in buffer */
+	int		bit;		/* bit number in the word */
+	xfs_rtblock_t	block;		/* bitmap block number */
+	xfs_buf_t	*bp;		/* buf for the block */
+	xfs_rtword_t	*bufp;		/* starting word in buffer */
+	int		error;		/* error value */
+	xfs_rtword_t	*first;		/* first used word in the buffer */
+	int		i;		/* current bit number rel. to start */
+	int		lastbit;	/* last useful bit in word */
+	xfs_rtword_t	mask;		/* mask o frelevant bits for value */
+	int		word;		/* word number in the buffer */
+
+	/*
+	 * Compute starting bitmap block number.
+	 */
+	block = XFS_BITTOBLOCK(mp, start);
+	/*
+	 * Read the bitmap block, and point to its data.
+	 */
+	error = xfs_rtbuf_get(mp, tp, block, 0, &bp);
+	if (error) {
+		return error;
+	}
+	bufp = bp->b_addr;
+	/*
+	 * Compute the starting word's address, and starting bit.
+	 */
+	word = XFS_BITTOWORD(mp, start);
+	first = b = &bufp[word];
+	bit = (int)(start & (XFS_NBWORD - 1));
+	/*
+	 * 0 (allocated) => all zeroes; 1 (free) => all ones.
+	 */
+	val = -val;
+	/*
+	 * If not starting on a word boundary, deal with the first
+	 * (partial) word.
+	 */
+	if (bit) {
+		/*
+		 * Compute first bit not changed and mask of relevant bits.
+		 */
+		lastbit = XFS_RTMIN(bit + len, XFS_NBWORD);
+		mask = (((xfs_rtword_t)1 << (lastbit - bit)) - 1) << bit;
+		/*
+		 * Set/clear the active bits.
+		 */
+		if (val)
+			*b |= mask;
+		else
+			*b &= ~mask;
+		i = lastbit - bit;
+		/*
+		 * Go on to the next block if that's where the next word is
+		 * and we need the next word.
+		 */
+		if (++word == XFS_BLOCKWSIZE(mp) && i < len) {
+			/*
+			 * Log the changed part of this block.
+			 * Get the next one.
+			 */
+			xfs_trans_log_buf(tp, bp,
+				(uint)((char *)first - (char *)bufp),
+				(uint)((char *)b - (char *)bufp));
+			error = xfs_rtbuf_get(mp, tp, ++block, 0, &bp);
+			if (error) {
+				return error;
+			}
+			first = b = bufp = bp->b_addr;
+			word = 0;
+		} else {
+			/*
+			 * Go on to the next word in the buffer
+			 */
+			b++;
+		}
+	} else {
+		/*
+		 * Starting on a word boundary, no partial word.
+		 */
+		i = 0;
+	}
+	/*
+	 * Loop over whole words in buffers.  When we use up one buffer
+	 * we move on to the next one.
+	 */
+	while (len - i >= XFS_NBWORD) {
+		/*
+		 * Set the word value correctly.
+		 */
+		*b = val;
+		i += XFS_NBWORD;
+		/*
+		 * Go on to the next block if that's where the next word is
+		 * and we need the next word.
+		 */
+		if (++word == XFS_BLOCKWSIZE(mp) && i < len) {
+			/*
+			 * Log the changed part of this block.
+			 * Get the next one.
+			 */
+			xfs_trans_log_buf(tp, bp,
+				(uint)((char *)first - (char *)bufp),
+				(uint)((char *)b - (char *)bufp));
+			error = xfs_rtbuf_get(mp, tp, ++block, 0, &bp);
+			if (error) {
+				return error;
+			}
+			first = b = bufp = bp->b_addr;
+			word = 0;
+		} else {
+			/*
+			 * Go on to the next word in the buffer
+			 */
+			b++;
+		}
+	}
+	/*
+	 * If not ending on a word boundary, deal with the last
+	 * (partial) word.
+	 */
+	if ((lastbit = len - i)) {
+		/*
+		 * Compute a mask of relevant bits.
+		 */
+		bit = 0;
+		mask = ((xfs_rtword_t)1 << lastbit) - 1;
+		/*
+		 * Set/clear the active bits.
+		 */
+		if (val)
+			*b |= mask;
+		else
+			*b &= ~mask;
+		b++;
+	}
+	/*
+	 * Log any remaining changed bytes.
+	 */
+	if (b > first)
+		xfs_trans_log_buf(tp, bp, (uint)((char *)first - (char *)bufp),
+			(uint)((char *)b - (char *)bufp - 1));
+	return 0;
+}
+
+/*
+ * Mark an extent specified by start and len freed.
+ * Updates all the summary information as well as the bitmap.
+ */
+int
+xfs_rtfree_range(
+	xfs_mount_t	*mp,		/* file system mount point */
+	xfs_trans_t	*tp,		/* transaction pointer */
+	xfs_rtblock_t	start,		/* starting block to free */
+	xfs_extlen_t	len,		/* length to free */
+	xfs_buf_t	**rbpp,		/* in/out: summary block buffer */
+	xfs_fsblock_t	*rsb)		/* in/out: summary block number */
+{
+	xfs_rtblock_t	end;		/* end of the freed extent */
+	int		error;		/* error value */
+	xfs_rtblock_t	postblock;	/* first block freed > end */
+	xfs_rtblock_t	preblock;	/* first block freed < start */
+
+	end = start + len - 1;
+	/*
+	 * Modify the bitmap to mark this extent freed.
+	 */
+	error = xfs_rtmodify_range(mp, tp, start, len, 1);
+	if (error) {
+		return error;
+	}
+	/*
+	 * Assume we're freeing out of the middle of an allocated extent.
+	 * We need to find the beginning and end of the extent so we can
+	 * properly update the summary.
+	 */
+	error = xfs_rtfind_back(mp, tp, start, 0, &preblock);
+	if (error) {
+		return error;
+	}
+	/*
+	 * Find the next allocated block (end of allocated extent).
+	 */
+	error = xfs_rtfind_forw(mp, tp, end, mp->m_sb.sb_rextents - 1,
+		&postblock);
+	if (error)
+		return error;
+	/*
+	 * If there are blocks not being freed at the front of the
+	 * old extent, add summary data for them to be allocated.
+	 */
+	if (preblock < start) {
+		error = xfs_rtmodify_summary(mp, tp,
+			XFS_RTBLOCKLOG(start - preblock),
+			XFS_BITTOBLOCK(mp, preblock), -1, rbpp, rsb);
+		if (error) {
+			return error;
+		}
+	}
+	/*
+	 * If there are blocks not being freed at the end of the
+	 * old extent, add summary data for them to be allocated.
+	 */
+	if (postblock > end) {
+		error = xfs_rtmodify_summary(mp, tp,
+			XFS_RTBLOCKLOG(postblock - end),
+			XFS_BITTOBLOCK(mp, end + 1), -1, rbpp, rsb);
+		if (error) {
+			return error;
+		}
+	}
+	/*
+	 * Increment the summary information corresponding to the entire
+	 * (new) free extent.
+	 */
+	error = xfs_rtmodify_summary(mp, tp,
+		XFS_RTBLOCKLOG(postblock + 1 - preblock),
+		XFS_BITTOBLOCK(mp, preblock), 1, rbpp, rsb);
+	return error;
+}
+
+/*
+ * Check that the given range is either all allocated (val = 0) or
+ * all free (val = 1).
+ */
+int
+xfs_rtcheck_range(
+	xfs_mount_t	*mp,		/* file system mount point */
+	xfs_trans_t	*tp,		/* transaction pointer */
+	xfs_rtblock_t	start,		/* starting block number of extent */
+	xfs_extlen_t	len,		/* length of extent */
+	int		val,		/* 1 for free, 0 for allocated */
+	xfs_rtblock_t	*new,		/* out: first block not matching */
+	int		*stat)		/* out: 1 for matches, 0 for not */
+{
+	xfs_rtword_t	*b;		/* current word in buffer */
+	int		bit;		/* bit number in the word */
+	xfs_rtblock_t	block;		/* bitmap block number */
+	xfs_buf_t	*bp;		/* buf for the block */
+	xfs_rtword_t	*bufp;		/* starting word in buffer */
+	int		error;		/* error value */
+	xfs_rtblock_t	i;		/* current bit number rel. to start */
+	xfs_rtblock_t	lastbit;	/* last useful bit in word */
+	xfs_rtword_t	mask;		/* mask of relevant bits for value */
+	xfs_rtword_t	wdiff;		/* difference from wanted value */
+	int		word;		/* word number in the buffer */
+
+	/*
+	 * Compute starting bitmap block number
+	 */
+	block = XFS_BITTOBLOCK(mp, start);
+	/*
+	 * Read the bitmap block.
+	 */
+	error = xfs_rtbuf_get(mp, tp, block, 0, &bp);
+	if (error) {
+		return error;
+	}
+	bufp = bp->b_addr;
+	/*
+	 * Compute the starting word's address, and starting bit.
+	 */
+	word = XFS_BITTOWORD(mp, start);
+	b = &bufp[word];
+	bit = (int)(start & (XFS_NBWORD - 1));
+	/*
+	 * 0 (allocated) => all zero's; 1 (free) => all one's.
+	 */
+	val = -val;
+	/*
+	 * If not starting on a word boundary, deal with the first
+	 * (partial) word.
+	 */
+	if (bit) {
+		/*
+		 * Compute first bit not examined.
+		 */
+		lastbit = XFS_RTMIN(bit + len, XFS_NBWORD);
+		/*
+		 * Mask of relevant bits.
+		 */
+		mask = (((xfs_rtword_t)1 << (lastbit - bit)) - 1) << bit;
+		/*
+		 * Compute difference between actual and desired value.
+		 */
+		if ((wdiff = (*b ^ val) & mask)) {
+			/*
+			 * Different, compute first wrong bit and return.
+			 */
+			xfs_trans_brelse(tp, bp);
+			i = XFS_RTLOBIT(wdiff) - bit;
+			*new = start + i;
+			*stat = 0;
+			return 0;
+		}
+		i = lastbit - bit;
+		/*
+		 * Go on to next block if that's where the next word is
+		 * and we need the next word.
+		 */
+		if (++word == XFS_BLOCKWSIZE(mp) && i < len) {
+			/*
+			 * If done with this block, get the next one.
+			 */
+			xfs_trans_brelse(tp, bp);
+			error = xfs_rtbuf_get(mp, tp, ++block, 0, &bp);
+			if (error) {
+				return error;
+			}
+			b = bufp = bp->b_addr;
+			word = 0;
+		} else {
+			/*
+			 * Go on to the next word in the buffer.
+			 */
+			b++;
+		}
+	} else {
+		/*
+		 * Starting on a word boundary, no partial word.
+		 */
+		i = 0;
+	}
+	/*
+	 * Loop over whole words in buffers.  When we use up one buffer
+	 * we move on to the next one.
+	 */
+	while (len - i >= XFS_NBWORD) {
+		/*
+		 * Compute difference between actual and desired value.
+		 */
+		if ((wdiff = *b ^ val)) {
+			/*
+			 * Different, compute first wrong bit and return.
+			 */
+			xfs_trans_brelse(tp, bp);
+			i += XFS_RTLOBIT(wdiff);
+			*new = start + i;
+			*stat = 0;
+			return 0;
+		}
+		i += XFS_NBWORD;
+		/*
+		 * Go on to next block if that's where the next word is
+		 * and we need the next word.
+		 */
+		if (++word == XFS_BLOCKWSIZE(mp) && i < len) {
+			/*
+			 * If done with this block, get the next one.
+			 */
+			xfs_trans_brelse(tp, bp);
+			error = xfs_rtbuf_get(mp, tp, ++block, 0, &bp);
+			if (error) {
+				return error;
+			}
+			b = bufp = bp->b_addr;
+			word = 0;
+		} else {
+			/*
+			 * Go on to the next word in the buffer.
+			 */
+			b++;
+		}
+	}
+	/*
+	 * If not ending on a word boundary, deal with the last
+	 * (partial) word.
+	 */
+	if ((lastbit = len - i)) {
+		/*
+		 * Mask of relevant bits.
+		 */
+		mask = ((xfs_rtword_t)1 << lastbit) - 1;
+		/*
+		 * Compute difference between actual and desired value.
+		 */
+		if ((wdiff = (*b ^ val) & mask)) {
+			/*
+			 * Different, compute first wrong bit and return.
+			 */
+			xfs_trans_brelse(tp, bp);
+			i += XFS_RTLOBIT(wdiff);
+			*new = start + i;
+			*stat = 0;
+			return 0;
+		} else
+			i = len;
+	}
+	/*
+	 * Successful, return.
+	 */
+	xfs_trans_brelse(tp, bp);
+	*new = start + i;
+	*stat = 1;
+	return 0;
+}
+
+#ifdef DEBUG
+/*
+ * Check that the given extent (block range) is allocated already.
+ */
+STATIC int				/* error */
+xfs_rtcheck_alloc_range(
+	xfs_mount_t	*mp,		/* file system mount point */
+	xfs_trans_t	*tp,		/* transaction pointer */
+	xfs_rtblock_t	bno,		/* starting block number of extent */
+	xfs_extlen_t	len)		/* length of extent */
+{
+	xfs_rtblock_t	new;		/* dummy for xfs_rtcheck_range */
+	int		stat;
+	int		error;
+
+	error = xfs_rtcheck_range(mp, tp, bno, len, 0, &new, &stat);
+	if (error)
+		return error;
+	ASSERT(stat);
+	return 0;
+}
+#else
+#define xfs_rtcheck_alloc_range(m,t,b,l)	(0)
+#endif
+/*
+ * Free an extent in the realtime subvolume.  Length is expressed in
+ * realtime extents, as is the block number.
+ */
+int					/* error */
+xfs_rtfree_extent(
+	xfs_trans_t	*tp,		/* transaction pointer */
+	xfs_rtblock_t	bno,		/* starting block number to free */
+	xfs_extlen_t	len)		/* length of extent freed */
+{
+	int		error;		/* error value */
+	xfs_mount_t	*mp;		/* file system mount structure */
+	xfs_fsblock_t	sb;		/* summary file block number */
+	xfs_buf_t	*sumbp = NULL;	/* summary file block buffer */
+
+	mp = tp->t_mountp;
+
+	ASSERT(mp->m_rbmip->i_itemp != NULL);
+	ASSERT(xfs_isilocked(mp->m_rbmip, XFS_ILOCK_EXCL));
+
+	error = xfs_rtcheck_alloc_range(mp, tp, bno, len);
+	if (error)
+		return error;
+
+	/*
+	 * Free the range of realtime blocks.
+	 */
+	error = xfs_rtfree_range(mp, tp, bno, len, &sumbp, &sb);
+	if (error) {
+		return error;
+	}
+	/*
+	 * Mark more blocks free in the superblock.
+	 */
+	xfs_trans_mod_sb(tp, XFS_TRANS_SB_FREXTENTS, (long)len);
+	/*
+	 * If we've now freed all the blocks, reset the file sequence
+	 * number to 0.
+	 */
+	if (tp->t_frextents_delta + mp->m_sb.sb_frextents ==
+	    mp->m_sb.sb_rextents) {
+		if (!(mp->m_rbmip->i_d.di_flags & XFS_DIFLAG_NEWRTBM))
+			mp->m_rbmip->i_d.di_flags |= XFS_DIFLAG_NEWRTBM;
+		*(__uint64_t *)&mp->m_rbmip->i_d.di_atime = 0;
+		xfs_trans_log_inode(tp, mp->m_rbmip, XFS_ILOG_CORE);
+	}
+	return 0;
+}
+
diff --git a/fs/xfs/xfs_sb.c b/fs/xfs/xfs_sb.c
index a5b59d92eb70..b7c9aea77f8f 100644
--- a/fs/xfs/xfs_sb.c
+++ b/fs/xfs/xfs_sb.c
@@ -17,34 +17,26 @@
  */
 #include "xfs.h"
 #include "xfs_fs.h"
+#include "xfs_shared.h"
 #include "xfs_format.h"
+#include "xfs_log_format.h"
+#include "xfs_trans_resv.h"
 #include "xfs_bit.h"
-#include "xfs_log.h"
-#include "xfs_inum.h"
-#include "xfs_trans.h"
-#include "xfs_trans_priv.h"
 #include "xfs_sb.h"
 #include "xfs_ag.h"
 #include "xfs_mount.h"
-#include "xfs_da_btree.h"
-#include "xfs_dir2_format.h"
-#include "xfs_dir2.h"
-#include "xfs_bmap_btree.h"
-#include "xfs_alloc_btree.h"
-#include "xfs_ialloc_btree.h"
-#include "xfs_dinode.h"
 #include "xfs_inode.h"
-#include "xfs_btree.h"
 #include "xfs_ialloc.h"
 #include "xfs_alloc.h"
-#include "xfs_rtalloc.h"
-#include "xfs_bmap.h"
 #include "xfs_error.h"
-#include "xfs_quota.h"
-#include "xfs_fsops.h"
 #include "xfs_trace.h"
 #include "xfs_cksum.h"
+#include "xfs_trans.h"
 #include "xfs_buf_item.h"
+#include "xfs_dinode.h"
+#include "xfs_bmap_btree.h"
+#include "xfs_alloc_btree.h"
+#include "xfs_ialloc_btree.h"
 
 /*
  * Physical superblock buffer manipulations. Shared with libxfs in userspace.
@@ -249,13 +241,13 @@ xfs_mount_validate_sb(
 	if (xfs_sb_version_has_pquotino(sbp)) {
 		if (sbp->sb_qflags & (XFS_OQUOTA_ENFD | XFS_OQUOTA_CHKD)) {
 			xfs_notice(mp,
-			   "Version 5 of Super block has XFS_OQUOTA bits.\n");
+			   "Version 5 of Super block has XFS_OQUOTA bits.");
 			return XFS_ERROR(EFSCORRUPTED);
 		}
 	} else if (sbp->sb_qflags & (XFS_PQUOTA_ENFD | XFS_GQUOTA_ENFD |
 				XFS_PQUOTA_CHKD | XFS_GQUOTA_CHKD)) {
 			xfs_notice(mp,
-"Superblock earlier than Version 5 has XFS_[PQ]UOTA_{ENFD|CHKD} bits.\n");
+"Superblock earlier than Version 5 has XFS_[PQ]UOTA_{ENFD|CHKD} bits.");
 			return XFS_ERROR(EFSCORRUPTED);
 	}
 
@@ -596,6 +588,11 @@ xfs_sb_verify(
  * single bit error could clear the feature bit and unused parts of the
  * superblock are supposed to be zero. Hence a non-null crc field indicates that
  * we've potentially lost a feature bit and we should check it anyway.
+ *
+ * However, past bugs (i.e. in growfs) left non-zeroed regions beyond the
+ * last field in V4 secondary superblocks.  So for secondary superblocks,
+ * we are more forgiving, and ignore CRC failures if the primary doesn't
+ * indicate that the fs version is V5.
  */
 static void
 xfs_sb_read_verify(
@@ -616,16 +613,21 @@ xfs_sb_read_verify(
 
 		if (!xfs_verify_cksum(bp->b_addr, be16_to_cpu(dsb->sb_sectsize),
 				      offsetof(struct xfs_sb, sb_crc))) {
-			error = EFSCORRUPTED;
-			goto out_error;
+			/* Only fail bad secondaries on a known V5 filesystem */
+			if (bp->b_bn != XFS_SB_DADDR &&
+			    xfs_sb_version_hascrc(&mp->m_sb)) {
+				error = EFSCORRUPTED;
+				goto out_error;
+			}
 		}
 	}
 	error = xfs_sb_verify(bp, true);
 
 out_error:
 	if (error) {
-		XFS_CORRUPTION_ERROR(__func__, XFS_ERRLEVEL_LOW,
-				     mp, bp->b_addr);
+		if (error != EWRONGFS)
+			XFS_CORRUPTION_ERROR(__func__, XFS_ERRLEVEL_LOW,
+					     mp, bp->b_addr);
 		xfs_buf_ioerror(bp, error);
 	}
 }
diff --git a/fs/xfs/xfs_sb.h b/fs/xfs/xfs_sb.h
index 6835b44f850e..35061d4b614c 100644
--- a/fs/xfs/xfs_sb.h
+++ b/fs/xfs/xfs_sb.h
@@ -699,7 +699,4 @@ extern void	xfs_sb_from_disk(struct xfs_sb *, struct xfs_dsb *);
 extern void	xfs_sb_to_disk(struct xfs_dsb *, struct xfs_sb *, __int64_t);
 extern void	xfs_sb_quota_from_disk(struct xfs_sb *sbp);
 
-extern const struct xfs_buf_ops xfs_sb_buf_ops;
-extern const struct xfs_buf_ops xfs_sb_quiet_buf_ops;
-
 #endif	/* __XFS_SB_H__ */
diff --git a/fs/xfs/xfs_shared.h b/fs/xfs/xfs_shared.h
new file mode 100644
index 000000000000..8c5035a13df1
--- /dev/null
+++ b/fs/xfs/xfs_shared.h
@@ -0,0 +1,244 @@
+/*
+ * Copyright (c) 2000-2005 Silicon Graphics, Inc.
+ * Copyright (c) 2013 Red Hat, Inc.
+ * All Rights Reserved.
+ *
+ * This program is free software; you can redistribute it and/or
+ * modify it under the terms of the GNU General Public License as
+ * published by the Free Software Foundation.
+ *
+ * This program is distributed in the hope that it would be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+ * GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program; if not, write the Free Software Foundation,
+ * Inc.,  51 Franklin St, Fifth Floor, Boston, MA  02110-1301  USA
+ */
+#ifndef __XFS_SHARED_H__
+#define __XFS_SHARED_H__
+
+/*
+ * Definitions shared between kernel and userspace that don't fit into any other
+ * header file that is shared with userspace.
+ */
+struct xfs_ifork;
+struct xfs_buf;
+struct xfs_buf_ops;
+struct xfs_mount;
+struct xfs_trans;
+struct xfs_inode;
+
+/*
+ * Buffer verifier operations are widely used, including userspace tools
+ */
+extern const struct xfs_buf_ops xfs_agf_buf_ops;
+extern const struct xfs_buf_ops xfs_agi_buf_ops;
+extern const struct xfs_buf_ops xfs_agf_buf_ops;
+extern const struct xfs_buf_ops xfs_agfl_buf_ops;
+extern const struct xfs_buf_ops xfs_allocbt_buf_ops;
+extern const struct xfs_buf_ops xfs_attr3_leaf_buf_ops;
+extern const struct xfs_buf_ops xfs_attr3_rmt_buf_ops;
+extern const struct xfs_buf_ops xfs_bmbt_buf_ops;
+extern const struct xfs_buf_ops xfs_da3_node_buf_ops;
+extern const struct xfs_buf_ops xfs_dquot_buf_ops;
+extern const struct xfs_buf_ops xfs_symlink_buf_ops;
+extern const struct xfs_buf_ops xfs_agi_buf_ops;
+extern const struct xfs_buf_ops xfs_inobt_buf_ops;
+extern const struct xfs_buf_ops xfs_inode_buf_ops;
+extern const struct xfs_buf_ops xfs_inode_buf_ra_ops;
+extern const struct xfs_buf_ops xfs_dquot_buf_ops;
+extern const struct xfs_buf_ops xfs_sb_buf_ops;
+extern const struct xfs_buf_ops xfs_sb_quiet_buf_ops;
+extern const struct xfs_buf_ops xfs_symlink_buf_ops;
+
+/*
+ * Transaction types.  Used to distinguish types of buffers. These never reach
+ * the log.
+ */
+#define XFS_TRANS_SETATTR_NOT_SIZE	1
+#define XFS_TRANS_SETATTR_SIZE		2
+#define XFS_TRANS_INACTIVE		3
+#define XFS_TRANS_CREATE		4
+#define XFS_TRANS_CREATE_TRUNC		5
+#define XFS_TRANS_TRUNCATE_FILE		6
+#define XFS_TRANS_REMOVE		7
+#define XFS_TRANS_LINK			8
+#define XFS_TRANS_RENAME		9
+#define XFS_TRANS_MKDIR			10
+#define XFS_TRANS_RMDIR			11
+#define XFS_TRANS_SYMLINK		12
+#define XFS_TRANS_SET_DMATTRS		13
+#define XFS_TRANS_GROWFS		14
+#define XFS_TRANS_STRAT_WRITE		15
+#define XFS_TRANS_DIOSTRAT		16
+/* 17 was XFS_TRANS_WRITE_SYNC */
+#define	XFS_TRANS_WRITEID		18
+#define	XFS_TRANS_ADDAFORK		19
+#define	XFS_TRANS_ATTRINVAL		20
+#define	XFS_TRANS_ATRUNCATE		21
+#define	XFS_TRANS_ATTR_SET		22
+#define	XFS_TRANS_ATTR_RM		23
+#define	XFS_TRANS_ATTR_FLAG		24
+#define	XFS_TRANS_CLEAR_AGI_BUCKET	25
+#define XFS_TRANS_QM_SBCHANGE		26
+/*
+ * Dummy entries since we use the transaction type to index into the
+ * trans_type[] in xlog_recover_print_trans_head()
+ */
+#define XFS_TRANS_DUMMY1		27
+#define XFS_TRANS_DUMMY2		28
+#define XFS_TRANS_QM_QUOTAOFF		29
+#define XFS_TRANS_QM_DQALLOC		30
+#define XFS_TRANS_QM_SETQLIM		31
+#define XFS_TRANS_QM_DQCLUSTER		32
+#define XFS_TRANS_QM_QINOCREATE		33
+#define XFS_TRANS_QM_QUOTAOFF_END	34
+#define XFS_TRANS_SB_UNIT		35
+#define XFS_TRANS_FSYNC_TS		36
+#define	XFS_TRANS_GROWFSRT_ALLOC	37
+#define	XFS_TRANS_GROWFSRT_ZERO		38
+#define	XFS_TRANS_GROWFSRT_FREE		39
+#define	XFS_TRANS_SWAPEXT		40
+#define	XFS_TRANS_SB_COUNT		41
+#define	XFS_TRANS_CHECKPOINT		42
+#define	XFS_TRANS_ICREATE		43
+#define	XFS_TRANS_TYPE_MAX		43
+/* new transaction types need to be reflected in xfs_logprint(8) */
+
+#define XFS_TRANS_TYPES \
+	{ XFS_TRANS_SETATTR_NOT_SIZE,	"SETATTR_NOT_SIZE" }, \
+	{ XFS_TRANS_SETATTR_SIZE,	"SETATTR_SIZE" }, \
+	{ XFS_TRANS_INACTIVE,		"INACTIVE" }, \
+	{ XFS_TRANS_CREATE,		"CREATE" }, \
+	{ XFS_TRANS_CREATE_TRUNC,	"CREATE_TRUNC" }, \
+	{ XFS_TRANS_TRUNCATE_FILE,	"TRUNCATE_FILE" }, \
+	{ XFS_TRANS_REMOVE,		"REMOVE" }, \
+	{ XFS_TRANS_LINK,		"LINK" }, \
+	{ XFS_TRANS_RENAME,		"RENAME" }, \
+	{ XFS_TRANS_MKDIR,		"MKDIR" }, \
+	{ XFS_TRANS_RMDIR,		"RMDIR" }, \
+	{ XFS_TRANS_SYMLINK,		"SYMLINK" }, \
+	{ XFS_TRANS_SET_DMATTRS,	"SET_DMATTRS" }, \
+	{ XFS_TRANS_GROWFS,		"GROWFS" }, \
+	{ XFS_TRANS_STRAT_WRITE,	"STRAT_WRITE" }, \
+	{ XFS_TRANS_DIOSTRAT,		"DIOSTRAT" }, \
+	{ XFS_TRANS_WRITEID,		"WRITEID" }, \
+	{ XFS_TRANS_ADDAFORK,		"ADDAFORK" }, \
+	{ XFS_TRANS_ATTRINVAL,		"ATTRINVAL" }, \
+	{ XFS_TRANS_ATRUNCATE,		"ATRUNCATE" }, \
+	{ XFS_TRANS_ATTR_SET,		"ATTR_SET" }, \
+	{ XFS_TRANS_ATTR_RM,		"ATTR_RM" }, \
+	{ XFS_TRANS_ATTR_FLAG,		"ATTR_FLAG" }, \
+	{ XFS_TRANS_CLEAR_AGI_BUCKET,	"CLEAR_AGI_BUCKET" }, \
+	{ XFS_TRANS_QM_SBCHANGE,	"QM_SBCHANGE" }, \
+	{ XFS_TRANS_QM_QUOTAOFF,	"QM_QUOTAOFF" }, \
+	{ XFS_TRANS_QM_DQALLOC,		"QM_DQALLOC" }, \
+	{ XFS_TRANS_QM_SETQLIM,		"QM_SETQLIM" }, \
+	{ XFS_TRANS_QM_DQCLUSTER,	"QM_DQCLUSTER" }, \
+	{ XFS_TRANS_QM_QINOCREATE,	"QM_QINOCREATE" }, \
+	{ XFS_TRANS_QM_QUOTAOFF_END,	"QM_QOFF_END" }, \
+	{ XFS_TRANS_SB_UNIT,		"SB_UNIT" }, \
+	{ XFS_TRANS_FSYNC_TS,		"FSYNC_TS" }, \
+	{ XFS_TRANS_GROWFSRT_ALLOC,	"GROWFSRT_ALLOC" }, \
+	{ XFS_TRANS_GROWFSRT_ZERO,	"GROWFSRT_ZERO" }, \
+	{ XFS_TRANS_GROWFSRT_FREE,	"GROWFSRT_FREE" }, \
+	{ XFS_TRANS_SWAPEXT,		"SWAPEXT" }, \
+	{ XFS_TRANS_SB_COUNT,		"SB_COUNT" }, \
+	{ XFS_TRANS_CHECKPOINT,		"CHECKPOINT" }, \
+	{ XFS_TRANS_DUMMY1,		"DUMMY1" }, \
+	{ XFS_TRANS_DUMMY2,		"DUMMY2" }, \
+	{ XLOG_UNMOUNT_REC_TYPE,	"UNMOUNT" }
+
+/*
+ * This structure is used to track log items associated with
+ * a transaction.  It points to the log item and keeps some
+ * flags to track the state of the log item.  It also tracks
+ * the amount of space needed to log the item it describes
+ * once we get to commit processing (see xfs_trans_commit()).
+ */
+struct xfs_log_item_desc {
+	struct xfs_log_item	*lid_item;
+	struct list_head	lid_trans;
+	unsigned char		lid_flags;
+};
+
+#define XFS_LID_DIRTY		0x1
+
+/* log size calculation functions */
+int	xfs_log_calc_unit_res(struct xfs_mount *mp, int unit_bytes);
+int	xfs_log_calc_minimum_size(struct xfs_mount *);
+
+
+/*
+ * Values for t_flags.
+ */
+#define	XFS_TRANS_DIRTY		0x01	/* something needs to be logged */
+#define	XFS_TRANS_SB_DIRTY	0x02	/* superblock is modified */
+#define	XFS_TRANS_PERM_LOG_RES	0x04	/* xact took a permanent log res */
+#define	XFS_TRANS_SYNC		0x08	/* make commit synchronous */
+#define XFS_TRANS_DQ_DIRTY	0x10	/* at least one dquot in trx dirty */
+#define XFS_TRANS_RESERVE	0x20    /* OK to use reserved data blocks */
+#define XFS_TRANS_FREEZE_PROT	0x40	/* Transaction has elevated writer
+					   count in superblock */
+/*
+ * Values for call flags parameter.
+ */
+#define	XFS_TRANS_RELEASE_LOG_RES	0x4
+#define	XFS_TRANS_ABORT			0x8
+
+/*
+ * Field values for xfs_trans_mod_sb.
+ */
+#define	XFS_TRANS_SB_ICOUNT		0x00000001
+#define	XFS_TRANS_SB_IFREE		0x00000002
+#define	XFS_TRANS_SB_FDBLOCKS		0x00000004
+#define	XFS_TRANS_SB_RES_FDBLOCKS	0x00000008
+#define	XFS_TRANS_SB_FREXTENTS		0x00000010
+#define	XFS_TRANS_SB_RES_FREXTENTS	0x00000020
+#define	XFS_TRANS_SB_DBLOCKS		0x00000040
+#define	XFS_TRANS_SB_AGCOUNT		0x00000080
+#define	XFS_TRANS_SB_IMAXPCT		0x00000100
+#define	XFS_TRANS_SB_REXTSIZE		0x00000200
+#define	XFS_TRANS_SB_RBMBLOCKS		0x00000400
+#define	XFS_TRANS_SB_RBLOCKS		0x00000800
+#define	XFS_TRANS_SB_REXTENTS		0x00001000
+#define	XFS_TRANS_SB_REXTSLOG		0x00002000
+
+/*
+ * Here we centralize the specification of XFS meta-data buffer reference count
+ * values.  This determines how hard the buffer cache tries to hold onto the
+ * buffer.
+ */
+#define	XFS_AGF_REF		4
+#define	XFS_AGI_REF		4
+#define	XFS_AGFL_REF		3
+#define	XFS_INO_BTREE_REF	3
+#define	XFS_ALLOC_BTREE_REF	2
+#define	XFS_BMAP_BTREE_REF	2
+#define	XFS_DIR_BTREE_REF	2
+#define	XFS_INO_REF		2
+#define	XFS_ATTR_BTREE_REF	1
+#define	XFS_DQUOT_REF		1
+
+/*
+ * Flags for xfs_trans_ichgtime().
+ */
+#define	XFS_ICHGTIME_MOD	0x1	/* data fork modification timestamp */
+#define	XFS_ICHGTIME_CHG	0x2	/* inode field change timestamp */
+#define	XFS_ICHGTIME_CREATE	0x4	/* inode create timestamp */
+
+
+/*
+ * Symlink decoding/encoding functions
+ */
+int xfs_symlink_blocks(struct xfs_mount *mp, int pathlen);
+int xfs_symlink_hdr_set(struct xfs_mount *mp, xfs_ino_t ino, uint32_t offset,
+			uint32_t size, struct xfs_buf *bp);
+bool xfs_symlink_hdr_ok(struct xfs_mount *mp, xfs_ino_t ino, uint32_t offset,
+			uint32_t size, struct xfs_buf *bp);
+void xfs_symlink_local_to_remote(struct xfs_trans *tp, struct xfs_buf *bp,
+				 struct xfs_inode *ip, struct xfs_ifork *ifp);
+
+#endif /* __XFS_SHARED_H__ */
diff --git a/fs/xfs/xfs_super.c b/fs/xfs/xfs_super.c
index 8968f5036fa1..f317488263dd 100644
--- a/fs/xfs/xfs_super.c
+++ b/fs/xfs/xfs_super.c
@@ -17,34 +17,26 @@
  */
 
 #include "xfs.h"
+#include "xfs_shared.h"
 #include "xfs_format.h"
-#include "xfs_log.h"
+#include "xfs_log_format.h"
+#include "xfs_trans_resv.h"
 #include "xfs_inum.h"
-#include "xfs_trans.h"
 #include "xfs_sb.h"
 #include "xfs_ag.h"
-#include "xfs_alloc.h"
-#include "xfs_quota.h"
 #include "xfs_mount.h"
-#include "xfs_bmap_btree.h"
-#include "xfs_alloc_btree.h"
-#include "xfs_ialloc_btree.h"
-#include "xfs_dinode.h"
+#include "xfs_da_format.h"
 #include "xfs_inode.h"
 #include "xfs_btree.h"
-#include "xfs_ialloc.h"
 #include "xfs_bmap.h"
-#include "xfs_rtalloc.h"
+#include "xfs_alloc.h"
 #include "xfs_error.h"
-#include "xfs_itable.h"
 #include "xfs_fsops.h"
-#include "xfs_attr.h"
+#include "xfs_trans.h"
 #include "xfs_buf_item.h"
+#include "xfs_log.h"
 #include "xfs_log_priv.h"
-#include "xfs_trans_priv.h"
-#include "xfs_filestream.h"
 #include "xfs_da_btree.h"
-#include "xfs_dir2_format.h"
 #include "xfs_dir2.h"
 #include "xfs_extfree_item.h"
 #include "xfs_mru_cache.h"
@@ -52,6 +44,9 @@
 #include "xfs_icache.h"
 #include "xfs_trace.h"
 #include "xfs_icreate_item.h"
+#include "xfs_dinode.h"
+#include "xfs_filestream.h"
+#include "xfs_quota.h"
 
 #include <linux/namei.h>
 #include <linux/init.h>
@@ -946,10 +941,6 @@ xfs_fs_destroy_inode(
 
 	XFS_STATS_INC(vn_reclaim);
 
-	/* bad inode, get out here ASAP */
-	if (is_bad_inode(inode))
-		goto out_reclaim;
-
 	ASSERT(XFS_FORCED_SHUTDOWN(ip->i_mount) || ip->i_delayed_blks == 0);
 
 	/*
@@ -965,7 +956,6 @@ xfs_fs_destroy_inode(
 	 * this more efficiently than we can here, so simply let background
 	 * reclaim tear down all inodes.
 	 */
-out_reclaim:
 	xfs_inode_set_reclaim_tag(ip);
 }
 
@@ -1165,7 +1155,7 @@ xfs_restore_resvblks(struct xfs_mount *mp)
  * Note: xfs_log_quiesce() stops background log work - the callers must ensure
  * it is started again when appropriate.
  */
-void
+static void
 xfs_quiesce_attr(
 	struct xfs_mount	*mp)
 {
@@ -1246,7 +1236,7 @@ xfs_fs_remount(
 			 */
 #if 0
 			xfs_info(mp,
-		"mount option \"%s\" not supported for remount\n", p);
+		"mount option \"%s\" not supported for remount", p);
 			return -EINVAL;
 #else
 			break;
@@ -1491,10 +1481,6 @@ xfs_fs_fill_super(
 		error = ENOENT;
 		goto out_unmount;
 	}
-	if (is_bad_inode(root)) {
-		error = EINVAL;
-		goto out_unmount;
-	}
 	sb->s_root = d_make_root(root);
 	if (!sb->s_root) {
 		error = ENOMEM;
diff --git a/fs/xfs/xfs_symlink.c b/fs/xfs/xfs_symlink.c
index f622a97a7e33..14e58f2c96bd 100644
--- a/fs/xfs/xfs_symlink.c
+++ b/fs/xfs/xfs_symlink.c
@@ -17,31 +17,31 @@
  * Inc.,  51 Franklin St, Fifth Floor, Boston, MA  02110-1301  USA
  */
 #include "xfs.h"
+#include "xfs_shared.h"
 #include "xfs_fs.h"
 #include "xfs_format.h"
+#include "xfs_log_format.h"
+#include "xfs_trans_resv.h"
 #include "xfs_bit.h"
-#include "xfs_log.h"
-#include "xfs_trans.h"
 #include "xfs_sb.h"
 #include "xfs_ag.h"
 #include "xfs_mount.h"
-#include "xfs_da_btree.h"
-#include "xfs_dir2_format.h"
+#include "xfs_da_format.h"
 #include "xfs_dir2.h"
-#include "xfs_bmap_btree.h"
-#include "xfs_ialloc_btree.h"
-#include "xfs_dinode.h"
 #include "xfs_inode.h"
 #include "xfs_ialloc.h"
 #include "xfs_alloc.h"
 #include "xfs_bmap.h"
+#include "xfs_bmap_btree.h"
 #include "xfs_bmap_util.h"
 #include "xfs_error.h"
 #include "xfs_quota.h"
 #include "xfs_trans_space.h"
 #include "xfs_trace.h"
 #include "xfs_symlink.h"
-#include "xfs_buf_item.h"
+#include "xfs_trans.h"
+#include "xfs_log.h"
+#include "xfs_dinode.h"
 
 /* ----- Kernel only functions below ----- */
 STATIC int
@@ -424,8 +424,7 @@ xfs_symlink(
  */
 STATIC int
 xfs_inactive_symlink_rmt(
-	xfs_inode_t	*ip,
-	xfs_trans_t	**tpp)
+	struct xfs_inode *ip)
 {
 	xfs_buf_t	*bp;
 	int		committed;
@@ -437,11 +436,9 @@ xfs_inactive_symlink_rmt(
 	xfs_mount_t	*mp;
 	xfs_bmbt_irec_t	mval[XFS_SYMLINK_MAPS];
 	int		nmaps;
-	xfs_trans_t	*ntp;
 	int		size;
 	xfs_trans_t	*tp;
 
-	tp = *tpp;
 	mp = ip->i_mount;
 	ASSERT(ip->i_df.if_flags & XFS_IFEXTENTS);
 	/*
@@ -453,6 +450,16 @@ xfs_inactive_symlink_rmt(
 	 */
 	ASSERT(ip->i_d.di_nextents > 0 && ip->i_d.di_nextents <= 2);
 
+	tp = xfs_trans_alloc(mp, XFS_TRANS_INACTIVE);
+	error = xfs_trans_reserve(tp, &M_RES(mp)->tr_itruncate, 0, 0);
+	if (error) {
+		xfs_trans_cancel(tp, 0);
+		return error;
+	}
+
+	xfs_ilock(ip, XFS_ILOCK_EXCL);
+	xfs_trans_ijoin(tp, ip, 0);
+
 	/*
 	 * Lock the inode, fix the size, and join it to the transaction.
 	 * Hold it so in the normal path, we still have it locked for
@@ -471,7 +478,7 @@ xfs_inactive_symlink_rmt(
 	error = xfs_bmapi_read(ip, 0, xfs_symlink_blocks(mp, size),
 				mval, &nmaps, 0);
 	if (error)
-		goto error0;
+		goto error_trans_cancel;
 	/*
 	 * Invalidate the block(s). No validation is done.
 	 */
@@ -481,22 +488,24 @@ xfs_inactive_symlink_rmt(
 			XFS_FSB_TO_BB(mp, mval[i].br_blockcount), 0);
 		if (!bp) {
 			error = ENOMEM;
-			goto error1;
+			goto error_bmap_cancel;
 		}
 		xfs_trans_binval(tp, bp);
 	}
 	/*
 	 * Unmap the dead block(s) to the free_list.
 	 */
-	if ((error = xfs_bunmapi(tp, ip, 0, size, XFS_BMAPI_METADATA, nmaps,
-			&first_block, &free_list, &done)))
-		goto error1;
+	error = xfs_bunmapi(tp, ip, 0, size, XFS_BMAPI_METADATA, nmaps,
+			    &first_block, &free_list, &done);
+	if (error)
+		goto error_bmap_cancel;
 	ASSERT(done);
 	/*
 	 * Commit the first transaction.  This logs the EFI and the inode.
 	 */
-	if ((error = xfs_bmap_finish(&tp, &free_list, &committed)))
-		goto error1;
+	error = xfs_bmap_finish(&tp, &free_list, &committed);
+	if (error)
+		goto error_bmap_cancel;
 	/*
 	 * The transaction must have been committed, since there were
 	 * actually extents freed by xfs_bunmapi.  See xfs_bmap_finish.
@@ -511,26 +520,13 @@ xfs_inactive_symlink_rmt(
 	xfs_trans_ijoin(tp, ip, 0);
 	xfs_trans_log_inode(tp, ip, XFS_ILOG_CORE);
 	/*
-	 * Get a new, empty transaction to return to our caller.
-	 */
-	ntp = xfs_trans_dup(tp);
-	/*
 	 * Commit the transaction containing extent freeing and EFDs.
-	 * If we get an error on the commit here or on the reserve below,
-	 * we need to unlock the inode since the new transaction doesn't
-	 * have the inode attached.
 	 */
-	error = xfs_trans_commit(tp, 0);
-	tp = ntp;
+	error = xfs_trans_commit(tp, XFS_TRANS_RELEASE_LOG_RES);
 	if (error) {
 		ASSERT(XFS_FORCED_SHUTDOWN(mp));
-		goto error0;
+		goto error_unlock;
 	}
-	/*
-	 * transaction commit worked ok so we can drop the extra ticket
-	 * reference that we gained in xfs_trans_dup()
-	 */
-	xfs_log_ticket_put(tp->t_ticket);
 
 	/*
 	 * Remove the memory for extent descriptions (just bookkeeping).
@@ -538,23 +534,16 @@ xfs_inactive_symlink_rmt(
 	if (ip->i_df.if_bytes)
 		xfs_idata_realloc(ip, -ip->i_df.if_bytes, XFS_DATA_FORK);
 	ASSERT(ip->i_df.if_bytes == 0);
-	/*
-	 * Put an itruncate log reservation in the new transaction
-	 * for our caller.
-	 */
-	error = xfs_trans_reserve(tp, &M_RES(mp)->tr_itruncate, 0, 0);
-	if (error) {
-		ASSERT(XFS_FORCED_SHUTDOWN(mp));
-		goto error0;
-	}
 
-	xfs_trans_ijoin(tp, ip, 0);
-	*tpp = tp;
+	xfs_iunlock(ip, XFS_ILOCK_EXCL);
 	return 0;
 
- error1:
+error_bmap_cancel:
 	xfs_bmap_cancel(&free_list);
- error0:
+error_trans_cancel:
+	xfs_trans_cancel(tp, XFS_TRANS_RELEASE_LOG_RES | XFS_TRANS_ABORT);
+error_unlock:
+	xfs_iunlock(ip, XFS_ILOCK_EXCL);
 	return error;
 }
 
@@ -563,41 +552,46 @@ xfs_inactive_symlink_rmt(
  */
 int
 xfs_inactive_symlink(
-	struct xfs_inode	*ip,
-	struct xfs_trans	**tp)
+	struct xfs_inode	*ip)
 {
 	struct xfs_mount	*mp = ip->i_mount;
 	int			pathlen;
 
 	trace_xfs_inactive_symlink(ip);
 
-	ASSERT(xfs_isilocked(ip, XFS_ILOCK_EXCL));
-
 	if (XFS_FORCED_SHUTDOWN(mp))
 		return XFS_ERROR(EIO);
 
+	xfs_ilock(ip, XFS_ILOCK_EXCL);
+
 	/*
 	 * Zero length symlinks _can_ exist.
 	 */
 	pathlen = (int)ip->i_d.di_size;
-	if (!pathlen)
+	if (!pathlen) {
+		xfs_iunlock(ip, XFS_ILOCK_EXCL);
 		return 0;
+	}
 
 	if (pathlen < 0 || pathlen > MAXPATHLEN) {
 		xfs_alert(mp, "%s: inode (0x%llx) bad symlink length (%d)",
 			 __func__, (unsigned long long)ip->i_ino, pathlen);
+		xfs_iunlock(ip, XFS_ILOCK_EXCL);
 		ASSERT(0);
 		return XFS_ERROR(EFSCORRUPTED);
 	}
 
 	if (ip->i_df.if_flags & XFS_IFINLINE) {
-		if (ip->i_df.if_bytes > 0)
+		if (ip->i_df.if_bytes > 0) 
 			xfs_idata_realloc(ip, -(ip->i_df.if_bytes),
 					  XFS_DATA_FORK);
+		xfs_iunlock(ip, XFS_ILOCK_EXCL);
 		ASSERT(ip->i_df.if_bytes == 0);
 		return 0;
 	}
 
+	xfs_iunlock(ip, XFS_ILOCK_EXCL);
+
 	/* remove the remote symlink */
-	return xfs_inactive_symlink_rmt(ip, tp);
+	return xfs_inactive_symlink_rmt(ip);
 }
diff --git a/fs/xfs/xfs_symlink.h b/fs/xfs/xfs_symlink.h
index 99338ba666ac..e75245d09116 100644
--- a/fs/xfs/xfs_symlink.h
+++ b/fs/xfs/xfs_symlink.h
@@ -22,6 +22,6 @@
 int xfs_symlink(struct xfs_inode *dp, struct xfs_name *link_name,
 		const char *target_path, umode_t mode, struct xfs_inode **ipp);
 int xfs_readlink(struct xfs_inode *ip, char *link);
-int xfs_inactive_symlink(struct xfs_inode *ip, struct xfs_trans **tpp);
+int xfs_inactive_symlink(struct xfs_inode *ip);
 
 #endif /* __XFS_SYMLINK_H */
diff --git a/fs/xfs/xfs_symlink_remote.c b/fs/xfs/xfs_symlink_remote.c
index 01c85e3f6470..bf59a2b45f8c 100644
--- a/fs/xfs/xfs_symlink_remote.c
+++ b/fs/xfs/xfs_symlink_remote.c
@@ -19,8 +19,9 @@
 #include "xfs.h"
 #include "xfs_fs.h"
 #include "xfs_format.h"
-#include "xfs_log.h"
-#include "xfs_trans.h"
+#include "xfs_log_format.h"
+#include "xfs_shared.h"
+#include "xfs_trans_resv.h"
 #include "xfs_ag.h"
 #include "xfs_sb.h"
 #include "xfs_mount.h"
@@ -30,6 +31,7 @@
 #include "xfs_trace.h"
 #include "xfs_symlink.h"
 #include "xfs_cksum.h"
+#include "xfs_trans.h"
 #include "xfs_buf_item.h"
 
 
diff --git a/fs/xfs/xfs_trace.c b/fs/xfs/xfs_trace.c
index 5d7b3e40705f..dee3279c095e 100644
--- a/fs/xfs/xfs_trace.c
+++ b/fs/xfs/xfs_trace.c
@@ -17,19 +17,16 @@
  */
 #include "xfs.h"
 #include "xfs_fs.h"
-#include "xfs_types.h"
+#include "xfs_shared.h"
 #include "xfs_format.h"
-#include "xfs_log.h"
-#include "xfs_trans.h"
+#include "xfs_log_format.h"
+#include "xfs_trans_resv.h"
 #include "xfs_sb.h"
 #include "xfs_ag.h"
-#include "xfs_bmap_btree.h"
-#include "xfs_alloc_btree.h"
-#include "xfs_ialloc_btree.h"
-#include "xfs_dinode.h"
+#include "xfs_mount.h"
+#include "xfs_da_format.h"
 #include "xfs_inode.h"
 #include "xfs_btree.h"
-#include "xfs_mount.h"
 #include "xfs_da_btree.h"
 #include "xfs_ialloc.h"
 #include "xfs_itable.h"
@@ -37,6 +34,8 @@
 #include "xfs_bmap.h"
 #include "xfs_attr.h"
 #include "xfs_attr_leaf.h"
+#include "xfs_trans.h"
+#include "xfs_log.h"
 #include "xfs_log_priv.h"
 #include "xfs_buf_item.h"
 #include "xfs_quota.h"
@@ -46,6 +45,7 @@
 #include "xfs_dquot.h"
 #include "xfs_log_recover.h"
 #include "xfs_inode_item.h"
+#include "xfs_bmap_btree.h"
 
 /*
  * We include this last to have the helpers above available for the trace
diff --git a/fs/xfs/xfs_trace.h b/fs/xfs/xfs_trace.h
index 47910e638c18..425dfa45b9a0 100644
--- a/fs/xfs/xfs_trace.h
+++ b/fs/xfs/xfs_trace.h
@@ -31,8 +31,8 @@ struct xfs_da_args;
 struct xfs_da_node_entry;
 struct xfs_dquot;
 struct xfs_log_item;
-struct xlog_ticket;
 struct xlog;
+struct xlog_ticket;
 struct xlog_recover;
 struct xlog_recover_item;
 struct xfs_buf_log_format;
@@ -135,6 +135,31 @@ DEFINE_PERAG_REF_EVENT(xfs_perag_clear_reclaim);
 DEFINE_PERAG_REF_EVENT(xfs_perag_set_eofblocks);
 DEFINE_PERAG_REF_EVENT(xfs_perag_clear_eofblocks);
 
+DECLARE_EVENT_CLASS(xfs_ag_class,
+	TP_PROTO(struct xfs_mount *mp, xfs_agnumber_t agno),
+	TP_ARGS(mp, agno),
+	TP_STRUCT__entry(
+		__field(dev_t, dev)
+		__field(xfs_agnumber_t, agno)
+	),
+	TP_fast_assign(
+		__entry->dev = mp->m_super->s_dev;
+		__entry->agno = agno;
+	),
+	TP_printk("dev %d:%d agno %u",
+		  MAJOR(__entry->dev), MINOR(__entry->dev),
+		  __entry->agno)
+);
+#define DEFINE_AG_EVENT(name)	\
+DEFINE_EVENT(xfs_ag_class, name,	\
+	TP_PROTO(struct xfs_mount *mp, xfs_agnumber_t agno),	\
+	TP_ARGS(mp, agno))
+
+DEFINE_AG_EVENT(xfs_read_agf);
+DEFINE_AG_EVENT(xfs_alloc_read_agf);
+DEFINE_AG_EVENT(xfs_read_agi);
+DEFINE_AG_EVENT(xfs_ialloc_read_agi);
+
 TRACE_EVENT(xfs_attr_list_node_descend,
 	TP_PROTO(struct xfs_attr_list_context *ctx,
 		 struct xfs_da_node_entry *btree),
@@ -938,6 +963,63 @@ DEFINE_LOG_ITEM_EVENT(xfs_ail_pinned);
 DEFINE_LOG_ITEM_EVENT(xfs_ail_locked);
 DEFINE_LOG_ITEM_EVENT(xfs_ail_flushing);
 
+DECLARE_EVENT_CLASS(xfs_ail_class,
+	TP_PROTO(struct xfs_log_item *lip, xfs_lsn_t old_lsn, xfs_lsn_t new_lsn),
+	TP_ARGS(lip, old_lsn, new_lsn),
+	TP_STRUCT__entry(
+		__field(dev_t, dev)
+		__field(void *, lip)
+		__field(uint, type)
+		__field(uint, flags)
+		__field(xfs_lsn_t, old_lsn)
+		__field(xfs_lsn_t, new_lsn)
+	),
+	TP_fast_assign(
+		__entry->dev = lip->li_mountp->m_super->s_dev;
+		__entry->lip = lip;
+		__entry->type = lip->li_type;
+		__entry->flags = lip->li_flags;
+		__entry->old_lsn = old_lsn;
+		__entry->new_lsn = new_lsn;
+	),
+	TP_printk("dev %d:%d lip 0x%p old lsn %d/%d new lsn %d/%d type %s flags %s",
+		  MAJOR(__entry->dev), MINOR(__entry->dev),
+		  __entry->lip,
+		  CYCLE_LSN(__entry->old_lsn), BLOCK_LSN(__entry->old_lsn),
+		  CYCLE_LSN(__entry->new_lsn), BLOCK_LSN(__entry->new_lsn),
+		  __print_symbolic(__entry->type, XFS_LI_TYPE_DESC),
+		  __print_flags(__entry->flags, "|", XFS_LI_FLAGS))
+)
+
+#define DEFINE_AIL_EVENT(name) \
+DEFINE_EVENT(xfs_ail_class, name, \
+	TP_PROTO(struct xfs_log_item *lip, xfs_lsn_t old_lsn, xfs_lsn_t new_lsn), \
+	TP_ARGS(lip, old_lsn, new_lsn))
+DEFINE_AIL_EVENT(xfs_ail_insert);
+DEFINE_AIL_EVENT(xfs_ail_move);
+DEFINE_AIL_EVENT(xfs_ail_delete);
+
+TRACE_EVENT(xfs_log_assign_tail_lsn,
+	TP_PROTO(struct xlog *log, xfs_lsn_t new_lsn),
+	TP_ARGS(log, new_lsn),
+	TP_STRUCT__entry(
+		__field(dev_t, dev)
+		__field(xfs_lsn_t, new_lsn)
+		__field(xfs_lsn_t, old_lsn)
+		__field(xfs_lsn_t, last_sync_lsn)
+	),
+	TP_fast_assign(
+		__entry->dev = log->l_mp->m_super->s_dev;
+		__entry->new_lsn = new_lsn;
+		__entry->old_lsn = atomic64_read(&log->l_tail_lsn);
+		__entry->last_sync_lsn = atomic64_read(&log->l_last_sync_lsn);
+	),
+	TP_printk("dev %d:%d new tail lsn %d/%d, old lsn %d/%d, last sync %d/%d",
+		  MAJOR(__entry->dev), MINOR(__entry->dev),
+		  CYCLE_LSN(__entry->new_lsn), BLOCK_LSN(__entry->new_lsn),
+		  CYCLE_LSN(__entry->old_lsn), BLOCK_LSN(__entry->old_lsn),
+		  CYCLE_LSN(__entry->last_sync_lsn), BLOCK_LSN(__entry->last_sync_lsn))
+)
 
 DECLARE_EVENT_CLASS(xfs_file_class,
 	TP_PROTO(struct xfs_inode *ip, size_t count, loff_t offset, int flags),
diff --git a/fs/xfs/xfs_trans.c b/fs/xfs/xfs_trans.c
index 5411e01ab452..c812c5c060de 100644
--- a/fs/xfs/xfs_trans.c
+++ b/fs/xfs/xfs_trans.c
@@ -18,32 +18,21 @@
  */
 #include "xfs.h"
 #include "xfs_fs.h"
+#include "xfs_shared.h"
 #include "xfs_format.h"
-#include "xfs_log.h"
-#include "xfs_trans.h"
+#include "xfs_log_format.h"
+#include "xfs_trans_resv.h"
 #include "xfs_sb.h"
 #include "xfs_ag.h"
 #include "xfs_mount.h"
-#include "xfs_error.h"
-#include "xfs_da_btree.h"
-#include "xfs_bmap_btree.h"
-#include "xfs_alloc_btree.h"
-#include "xfs_ialloc_btree.h"
-#include "xfs_dinode.h"
 #include "xfs_inode.h"
-#include "xfs_btree.h"
-#include "xfs_ialloc.h"
-#include "xfs_alloc.h"
 #include "xfs_extent_busy.h"
-#include "xfs_bmap.h"
 #include "xfs_quota.h"
-#include "xfs_qm.h"
+#include "xfs_trans.h"
 #include "xfs_trans_priv.h"
-#include "xfs_trans_space.h"
-#include "xfs_inode_item.h"
-#include "xfs_log_priv.h"
-#include "xfs_buf_item.h"
+#include "xfs_log.h"
 #include "xfs_trace.h"
+#include "xfs_error.h"
 
 kmem_zone_t	*xfs_trans_zone;
 kmem_zone_t	*xfs_log_item_desc_zone;
diff --git a/fs/xfs/xfs_trans.h b/fs/xfs/xfs_trans.h
index 09cf40b89e8c..9b96d35e483d 100644
--- a/fs/xfs/xfs_trans.h
+++ b/fs/xfs/xfs_trans.h
@@ -18,10 +18,6 @@
 #ifndef	__XFS_TRANS_H__
 #define	__XFS_TRANS_H__
 
-struct xfs_log_item;
-
-#include "xfs_trans_resv.h"
-
 /* kernel only transaction subsystem defines */
 
 struct xfs_buf;
@@ -77,6 +73,9 @@ struct xfs_item_ops {
 	void (*iop_committing)(xfs_log_item_t *, xfs_lsn_t);
 };
 
+void	xfs_log_item_init(struct xfs_mount *mp, struct xfs_log_item *item,
+			  int type, const struct xfs_item_ops *ops);
+
 /*
  * Return values for the iop_push() routines.
  */
@@ -85,18 +84,12 @@ struct xfs_item_ops {
 #define XFS_ITEM_LOCKED		2
 #define XFS_ITEM_FLUSHING	3
 
-/*
- * This is the type of function which can be given to xfs_trans_callback()
- * to be called upon the transaction's commit to disk.
- */
-typedef void (*xfs_trans_callback_t)(struct xfs_trans *, void *);
 
 /*
  * This is the structure maintained for every active transaction.
  */
 typedef struct xfs_trans {
 	unsigned int		t_magic;	/* magic number */
-	xfs_log_callback_t	t_logcb;	/* log callback struct */
 	unsigned int		t_type;		/* transaction type */
 	unsigned int		t_log_res;	/* amt of log space resvd */
 	unsigned int		t_log_count;	/* count for perm log res */
@@ -132,7 +125,6 @@ typedef struct xfs_trans {
 	int64_t			t_rextents_delta;/* superblocks rextents chg */
 	int64_t			t_rextslog_delta;/* superblocks rextslog chg */
 	struct list_head	t_items;	/* log item descriptors */
-	xfs_trans_header_t	t_header;	/* header for in-log trans */
 	struct list_head	t_busy;		/* list of busy extents */
 	unsigned long		t_pflags;	/* saved process flags state */
 } xfs_trans_t;
@@ -237,10 +229,16 @@ void		xfs_trans_log_efd_extent(xfs_trans_t *,
 					 xfs_fsblock_t,
 					 xfs_extlen_t);
 int		xfs_trans_commit(xfs_trans_t *, uint flags);
+int		xfs_trans_roll(struct xfs_trans **, struct xfs_inode *);
 void		xfs_trans_cancel(xfs_trans_t *, int);
 int		xfs_trans_ail_init(struct xfs_mount *);
 void		xfs_trans_ail_destroy(struct xfs_mount *);
 
+void		xfs_trans_buf_set_type(struct xfs_trans *, struct xfs_buf *,
+				       enum xfs_blft);
+void		xfs_trans_buf_copy_type(struct xfs_buf *dst_bp,
+					struct xfs_buf *src_bp);
+
 extern kmem_zone_t	*xfs_trans_zone;
 extern kmem_zone_t	*xfs_log_item_desc_zone;
 
diff --git a/fs/xfs/xfs_trans_ail.c b/fs/xfs/xfs_trans_ail.c
index 21c6d7ddbc06..a7287354e535 100644
--- a/fs/xfs/xfs_trans_ail.c
+++ b/fs/xfs/xfs_trans_ail.c
@@ -18,15 +18,16 @@
  */
 #include "xfs.h"
 #include "xfs_fs.h"
-#include "xfs_types.h"
-#include "xfs_log.h"
-#include "xfs_trans.h"
+#include "xfs_log_format.h"
+#include "xfs_trans_resv.h"
 #include "xfs_sb.h"
 #include "xfs_ag.h"
 #include "xfs_mount.h"
+#include "xfs_trans.h"
 #include "xfs_trans_priv.h"
 #include "xfs_trace.h"
 #include "xfs_error.h"
+#include "xfs_log.h"
 
 #ifdef DEBUG
 /*
@@ -658,11 +659,13 @@ xfs_trans_ail_update_bulk(
 			if (XFS_LSN_CMP(lsn, lip->li_lsn) <= 0)
 				continue;
 
+			trace_xfs_ail_move(lip, lip->li_lsn, lsn);
 			xfs_ail_delete(ailp, lip);
 			if (mlip == lip)
 				mlip_changed = 1;
 		} else {
 			lip->li_flags |= XFS_LI_IN_AIL;
+			trace_xfs_ail_insert(lip, 0, lsn);
 		}
 		lip->li_lsn = lsn;
 		list_add(&lip->li_ail, &tmp);
@@ -731,6 +734,7 @@ xfs_trans_ail_delete_bulk(
 			return;
 		}
 
+		trace_xfs_ail_delete(lip, mlip->li_lsn, lip->li_lsn);
 		xfs_ail_delete(ailp, lip);
 		lip->li_flags &= ~XFS_LI_IN_AIL;
 		lip->li_lsn = 0;
diff --git a/fs/xfs/xfs_trans_buf.c b/fs/xfs/xfs_trans_buf.c
index 8c75b8f67270..c035d11b7734 100644
--- a/fs/xfs/xfs_trans_buf.c
+++ b/fs/xfs/xfs_trans_buf.c
@@ -17,17 +17,15 @@
  */
 #include "xfs.h"
 #include "xfs_fs.h"
-#include "xfs_types.h"
-#include "xfs_log.h"
-#include "xfs_trans.h"
+#include "xfs_shared.h"
+#include "xfs_format.h"
+#include "xfs_log_format.h"
+#include "xfs_trans_resv.h"
 #include "xfs_sb.h"
 #include "xfs_ag.h"
 #include "xfs_mount.h"
-#include "xfs_bmap_btree.h"
-#include "xfs_alloc_btree.h"
-#include "xfs_ialloc_btree.h"
-#include "xfs_dinode.h"
 #include "xfs_inode.h"
+#include "xfs_trans.h"
 #include "xfs_buf_item.h"
 #include "xfs_trans_priv.h"
 #include "xfs_error.h"
diff --git a/fs/xfs/xfs_trans_dquot.c b/fs/xfs/xfs_trans_dquot.c
index 54ee3c5dee76..cd2a10e15d3a 100644
--- a/fs/xfs/xfs_trans_dquot.c
+++ b/fs/xfs/xfs_trans_dquot.c
@@ -17,23 +17,18 @@
  */
 #include "xfs.h"
 #include "xfs_fs.h"
+#include "xfs_shared.h"
 #include "xfs_format.h"
-#include "xfs_log.h"
-#include "xfs_trans.h"
+#include "xfs_log_format.h"
+#include "xfs_trans_resv.h"
 #include "xfs_sb.h"
 #include "xfs_ag.h"
-#include "xfs_alloc.h"
-#include "xfs_quota.h"
 #include "xfs_mount.h"
-#include "xfs_bmap_btree.h"
 #include "xfs_inode.h"
-#include "xfs_itable.h"
-#include "xfs_bmap.h"
-#include "xfs_rtalloc.h"
 #include "xfs_error.h"
-#include "xfs_attr.h"
-#include "xfs_buf_item.h"
+#include "xfs_trans.h"
 #include "xfs_trans_priv.h"
+#include "xfs_quota.h"
 #include "xfs_qm.h"
 
 STATIC void	xfs_trans_alloc_dqinfo(xfs_trans_t *);
diff --git a/fs/xfs/xfs_trans_extfree.c b/fs/xfs/xfs_trans_extfree.c
index 8d71b16eccae..47978ba89dae 100644
--- a/fs/xfs/xfs_trans_extfree.c
+++ b/fs/xfs/xfs_trans_extfree.c
@@ -17,12 +17,13 @@
  */
 #include "xfs.h"
 #include "xfs_fs.h"
-#include "xfs_types.h"
-#include "xfs_log.h"
-#include "xfs_trans.h"
+#include "xfs_shared.h"
+#include "xfs_log_format.h"
+#include "xfs_trans_resv.h"
 #include "xfs_sb.h"
 #include "xfs_ag.h"
 #include "xfs_mount.h"
+#include "xfs_trans.h"
 #include "xfs_trans_priv.h"
 #include "xfs_extfree_item.h"
 
diff --git a/fs/xfs/xfs_trans_inode.c b/fs/xfs/xfs_trans_inode.c
index 53dfe46f3680..1bba7f60d94c 100644
--- a/fs/xfs/xfs_trans_inode.c
+++ b/fs/xfs/xfs_trans_inode.c
@@ -17,18 +17,15 @@
  */
 #include "xfs.h"
 #include "xfs_fs.h"
-#include "xfs_types.h"
-#include "xfs_log.h"
-#include "xfs_trans.h"
+#include "xfs_shared.h"
+#include "xfs_format.h"
+#include "xfs_log_format.h"
+#include "xfs_trans_resv.h"
 #include "xfs_sb.h"
 #include "xfs_ag.h"
 #include "xfs_mount.h"
-#include "xfs_bmap_btree.h"
-#include "xfs_alloc_btree.h"
-#include "xfs_ialloc_btree.h"
-#include "xfs_dinode.h"
 #include "xfs_inode.h"
-#include "xfs_btree.h"
+#include "xfs_trans.h"
 #include "xfs_trans_priv.h"
 #include "xfs_inode_item.h"
 #include "xfs_trace.h"
diff --git a/fs/xfs/xfs_trans_priv.h b/fs/xfs/xfs_trans_priv.h
index c52def0b441c..12e86af9d9b9 100644
--- a/fs/xfs/xfs_trans_priv.h
+++ b/fs/xfs/xfs_trans_priv.h
@@ -27,7 +27,6 @@ struct xfs_log_vec;
 
 
 void	xfs_trans_init(struct xfs_mount *);
-int	xfs_trans_roll(struct xfs_trans **, struct xfs_inode *);
 void	xfs_trans_add_item(struct xfs_trans *, struct xfs_log_item *);
 void	xfs_trans_del_item(struct xfs_log_item *);
 void	xfs_trans_free_items(struct xfs_trans *tp, xfs_lsn_t commit_lsn,
diff --git a/fs/xfs/xfs_trans_resv.c b/fs/xfs/xfs_trans_resv.c
index a65a3cc40610..d53d9f0627a7 100644
--- a/fs/xfs/xfs_trans_resv.c
+++ b/fs/xfs/xfs_trans_resv.c
@@ -18,27 +18,19 @@
  */
 #include "xfs.h"
 #include "xfs_fs.h"
+#include "xfs_shared.h"
 #include "xfs_format.h"
-#include "xfs_log.h"
+#include "xfs_log_format.h"
 #include "xfs_trans_resv.h"
-#include "xfs_trans.h"
 #include "xfs_sb.h"
 #include "xfs_ag.h"
 #include "xfs_mount.h"
-#include "xfs_error.h"
-#include "xfs_da_btree.h"
-#include "xfs_bmap_btree.h"
-#include "xfs_alloc_btree.h"
-#include "xfs_ialloc_btree.h"
-#include "xfs_dinode.h"
+#include "xfs_da_format.h"
 #include "xfs_inode.h"
-#include "xfs_btree.h"
+#include "xfs_bmap_btree.h"
 #include "xfs_ialloc.h"
-#include "xfs_alloc.h"
-#include "xfs_extent_busy.h"
-#include "xfs_bmap.h"
-#include "xfs_bmap_util.h"
 #include "xfs_quota.h"
+#include "xfs_trans.h"
 #include "xfs_qm.h"
 #include "xfs_trans_space.h"
 #include "xfs_trace.h"
diff --git a/fs/xfs/xfs_vnode.h b/fs/xfs/xfs_vnode.h
index db14d0c08682..3e8e797c6d11 100644
--- a/fs/xfs/xfs_vnode.h
+++ b/fs/xfs/xfs_vnode.h
@@ -25,14 +25,6 @@ struct xfs_inode;
 struct attrlist_cursor_kern;
 
 /*
- * Return values for xfs_inactive.  A return value of
- * VN_INACTIVE_NOCACHE implies that the file system behavior
- * has disassociated its state and bhv_desc_t from the vnode.
- */
-#define	VN_INACTIVE_CACHE	0
-#define	VN_INACTIVE_NOCACHE	1
-
-/*
  * Flags for read/write calls - same values as IRIX
  */
 #define IO_ISDIRECT	0x00004		/* bypass page cache */
diff --git a/fs/xfs/xfs_xattr.c b/fs/xfs/xfs_xattr.c
index e01f35ea76ba..9d479073ba41 100644
--- a/fs/xfs/xfs_xattr.c
+++ b/fs/xfs/xfs_xattr.c
@@ -17,9 +17,13 @@
  */
 
 #include "xfs.h"
+#include "xfs_format.h"
 #include "xfs_log_format.h"
-#include "xfs_da_btree.h"
-#include "xfs_bmap_btree.h"
+#include "xfs_trans_resv.h"
+#include "xfs_sb.h"
+#include "xfs_ag.h"
+#include "xfs_mount.h"
+#include "xfs_da_format.h"
 #include "xfs_inode.h"
 #include "xfs_attr.h"
 #include "xfs_attr_leaf.h"
diff --git a/include/asm-generic/memory_model.h b/include/asm-generic/memory_model.h
index aea9e45efce6..14909b0b9cae 100644
--- a/include/asm-generic/memory_model.h
+++ b/include/asm-generic/memory_model.h
@@ -53,7 +53,7 @@
 
 #elif defined(CONFIG_SPARSEMEM)
 /*
- * Note: section's mem_map is encorded to reflect its start_pfn.
+ * Note: section's mem_map is encoded to reflect its start_pfn.
  * section[i].section_mem_map == mem_map's address - start_pfn;
  */
 #define __page_to_pfn(pg)					\
diff --git a/include/asm-generic/vmlinux.lds.h b/include/asm-generic/vmlinux.lds.h
index 83e2c31e8b00..bc2121fa9132 100644
--- a/include/asm-generic/vmlinux.lds.h
+++ b/include/asm-generic/vmlinux.lds.h
@@ -473,6 +473,7 @@
 #define KERNEL_CTORS()	. = ALIGN(8);			   \
 			VMLINUX_SYMBOL(__ctors_start) = .; \
 			*(.ctors)			   \
+			*(.init_array)			   \
 			VMLINUX_SYMBOL(__ctors_end) = .;
 #else
 #define KERNEL_CTORS()
diff --git a/include/drm/drmP.h b/include/drm/drmP.h
index b46fb45f2cca..1d4a920ef7ff 100644
--- a/include/drm/drmP.h
+++ b/include/drm/drmP.h
@@ -150,6 +150,7 @@ int drm_err(const char *func, const char *format, ...);
 #define DRIVER_BUS_PCI 0x1
 #define DRIVER_BUS_PLATFORM 0x2
 #define DRIVER_BUS_USB 0x3
+#define DRIVER_BUS_HOST1X 0x4
 
 /***********************************************************************/
 /** \name Begin the DRM... */
@@ -412,7 +413,12 @@ struct drm_prime_file_private {
 
 /** File private data */
 struct drm_file {
-	int authenticated;
+	unsigned always_authenticated :1;
+	unsigned authenticated :1;
+	unsigned is_master :1; /* this file private is a master for a minor */
+	/* true when the client has asked us to expose stereo 3D mode flags */
+	unsigned stereo_allowed :1;
+
 	struct pid *pid;
 	kuid_t uid;
 	drm_magic_t magic;
@@ -429,10 +435,8 @@ struct drm_file {
 	struct file *filp;
 	void *driver_priv;
 
-	int is_master; /* this file private is a master for a minor */
 	struct drm_master *master; /* master this node is currently associated with
 				      N.B. not always minor->master */
-
 	/**
 	 * fbs - List of framebuffers associated with this file.
 	 *
@@ -667,8 +671,6 @@ struct drm_gem_object {
 	uint32_t pending_read_domains;
 	uint32_t pending_write_domain;
 
-	void *driver_private;
-
 	/**
 	 * dma_buf - dma buf associated with this GEM object
 	 *
@@ -834,12 +836,17 @@ struct drm_driver {
 	/**
 	 * Called by vblank timestamping code.
 	 *
-	 * Return the current display scanout position from a crtc.
+	 * Return the current display scanout position from a crtc, and an
+	 * optional accurate ktime_get timestamp of when position was measured.
 	 *
 	 * \param dev  DRM device.
 	 * \param crtc Id of the crtc to query.
 	 * \param *vpos Target location for current vertical scanout position.
 	 * \param *hpos Target location for current horizontal scanout position.
+	 * \param *stime Target location for timestamp taken immediately before
+	 *               scanout position query. Can be NULL to skip timestamp.
+	 * \param *etime Target location for timestamp taken immediately after
+	 *               scanout position query. Can be NULL to skip timestamp.
 	 *
 	 * Returns vpos as a positive number while in active scanout area.
 	 * Returns vpos as a negative number inside vblank, counting the number
@@ -856,7 +863,8 @@ struct drm_driver {
 	 *
 	 */
 	int (*get_scanout_position) (struct drm_device *dev, int crtc,
-				     int *vpos, int *hpos);
+				     int *vpos, int *hpos, ktime_t *stime,
+				     ktime_t *etime);
 
 	/**
 	 * Called by \c drm_get_last_vbltimestamp. Should return a precise
@@ -922,7 +930,6 @@ struct drm_driver {
 	 *
 	 * Returns 0 on success.
 	 */
-	int (*gem_init_object) (struct drm_gem_object *obj);
 	void (*gem_free_object) (struct drm_gem_object *obj);
 	int (*gem_open_object) (struct drm_gem_object *, struct drm_file *);
 	void (*gem_close_object) (struct drm_gem_object *, struct drm_file *);
@@ -997,27 +1004,6 @@ struct drm_driver {
 #define DRM_MINOR_CONTROL 2
 #define DRM_MINOR_RENDER 3
 
-
-/**
- * debugfs node list. This structure represents a debugfs file to
- * be created by the drm core
- */
-struct drm_debugfs_list {
-	const char *name; /** file name */
-	int (*show)(struct seq_file*, void*); /** show callback */
-	u32 driver_features; /**< Required driver features for this entry */
-};
-
-/**
- * debugfs node structure. This structure represents a debugfs file.
- */
-struct drm_debugfs_node {
-	struct list_head list;
-	struct drm_minor *minor;
-	struct drm_debugfs_list *debugfs_ent;
-	struct dentry *dent;
-};
-
 /**
  * Info file list entry. This structure represents a debugfs or proc file to
  * be created by the drm core
@@ -1046,7 +1032,7 @@ struct drm_minor {
 	int index;			/**< Minor device number */
 	int type;                       /**< Control or render */
 	dev_t device;			/**< Device number for mknod */
-	struct device kdev;		/**< Linux device */
+	struct device *kdev;		/**< Linux device */
 	struct drm_device *dev;
 
 	struct dentry *debugfs_root;
@@ -1081,6 +1067,19 @@ struct drm_pending_vblank_event {
 	struct drm_event_vblank event;
 };
 
+struct drm_vblank_crtc {
+	wait_queue_head_t queue;	/**< VBLANK wait queue */
+	struct timeval time[DRM_VBLANKTIME_RBSIZE];	/**< timestamp of current count */
+	atomic_t count;			/**< number of VBLANK interrupts */
+	atomic_t refcount;		/* number of users of vblank interruptsper crtc */
+	u32 last;			/* protected by dev->vbl_lock, used */
+					/* for wraparound handling */
+	u32 last_wait;			/* Last vblank seqno waited per CRTC */
+	unsigned int inmodeset;		/* Display driver is setting mode */
+	bool enabled;			/* so we don't call enable more than
+					   once per disable */
+};
+
 /**
  * DRM device structure. This structure represent a complete card that
  * may contain multiple heads.
@@ -1105,25 +1104,16 @@ struct drm_device {
 	atomic_t buf_alloc;		/**< Buffer allocation in progress */
 	/*@} */
 
-	/** \name Performance counters */
-	/*@{ */
-	unsigned long counters;
-	enum drm_stat_type types[15];
-	atomic_t counts[15];
-	/*@} */
-
 	struct list_head filelist;
 
 	/** \name Memory management */
 	/*@{ */
 	struct list_head maplist;	/**< Linked list of regions */
-	int map_count;			/**< Number of mappable regions */
 	struct drm_open_hash map_hash;	/**< User token hash table for maps */
 
 	/** \name Context handle management */
 	/*@{ */
 	struct list_head ctxlist;	/**< Linked list of context handles */
-	int ctx_count;			/**< Number of context handles */
 	struct mutex ctxlist_mutex;	/**< For ctxlist */
 
 	struct idr ctx_idr;
@@ -1139,12 +1129,11 @@ struct drm_device {
 
 	/** \name Context support */
 	/*@{ */
-	int irq_enabled;		/**< True if irq handler is enabled */
+	bool irq_enabled;		/**< True if irq handler is enabled */
 	__volatile__ long context_flag;	/**< Context swapping flag */
 	int last_context;		/**< Last current context */
 	/*@} */
 
-	struct work_struct work;
 	/** \name VBLANK IRQ support */
 	/*@{ */
 
@@ -1154,20 +1143,13 @@ struct drm_device {
 	 * Once the modeset ioctl *has* been called though, we can safely
 	 * disable them when unused.
 	 */
-	int vblank_disable_allowed;
+	bool vblank_disable_allowed;
+
+	/* array of size num_crtcs */
+	struct drm_vblank_crtc *vblank;
 
-	wait_queue_head_t *vbl_queue;   /**< VBLANK wait queue */
-	atomic_t *_vblank_count;        /**< number of VBLANK interrupts (driver must alloc the right number of counters) */
-	struct timeval *_vblank_time;   /**< timestamp of current vblank_count (drivers must alloc right number of fields) */
 	spinlock_t vblank_time_lock;    /**< Protects vblank count and time updates during vblank enable/disable */
 	spinlock_t vbl_lock;
-	atomic_t *vblank_refcount;      /* number of users of vblank interruptsper crtc */
-	u32 *last_vblank;               /* protected by dev->vbl_lock, used */
-					/* for wraparound handling */
-	int *vblank_enabled;            /* so we don't call enable more than
-					   once per disable */
-	int *vblank_inmodeset;          /* Display driver is setting mode */
-	u32 *last_vblank_wait;		/* Last vblank seqno waited per CRTC */
 	struct timer_list vblank_disable_timer;
 
 	u32 max_vblank_count;           /**< size of vblank counter register */
@@ -1184,8 +1166,6 @@ struct drm_device {
 
 	struct device *dev;             /**< Device structure */
 	struct pci_dev *pdev;		/**< PCI device structure */
-	int pci_vendor;			/**< PCI vendor id */
-	int pci_device;			/**< PCI device id */
 #ifdef __alpha__
 	struct pci_controller *hose;
 #endif
@@ -1303,6 +1283,8 @@ extern int drm_getstats(struct drm_device *dev, void *data,
 			struct drm_file *file_priv);
 extern int drm_getcap(struct drm_device *dev, void *data,
 		      struct drm_file *file_priv);
+extern int drm_setclientcap(struct drm_device *dev, void *data,
+			    struct drm_file *file_priv);
 extern int drm_setversion(struct drm_device *dev, void *data,
 			  struct drm_file *file_priv);
 extern int drm_noop(struct drm_device *dev, void *data,
@@ -1454,7 +1436,6 @@ extern struct drm_master *drm_master_get(struct drm_master *master);
 extern void drm_master_put(struct drm_master **master);
 
 extern void drm_put_dev(struct drm_device *dev);
-extern int drm_put_minor(struct drm_minor **minor);
 extern void drm_unplug_dev(struct drm_device *dev);
 extern unsigned int drm_debug;
 extern unsigned int drm_rnodes;
@@ -1474,10 +1455,11 @@ extern struct drm_local_map *drm_getsarea(struct drm_device *dev);
 #if defined(CONFIG_DEBUG_FS)
 extern int drm_debugfs_init(struct drm_minor *minor, int minor_id,
 			    struct dentry *root);
-extern int drm_debugfs_create_files(struct drm_info_list *files, int count,
-				    struct dentry *root, struct drm_minor *minor);
-extern int drm_debugfs_remove_files(struct drm_info_list *files, int count,
-                                    struct drm_minor *minor);
+extern int drm_debugfs_create_files(const struct drm_info_list *files,
+				    int count, struct dentry *root,
+				    struct drm_minor *minor);
+extern int drm_debugfs_remove_files(const struct drm_info_list *files,
+				    int count, struct drm_minor *minor);
 extern int drm_debugfs_cleanup(struct drm_minor *minor);
 #endif
 
@@ -1556,8 +1538,6 @@ int drm_gem_init(struct drm_device *dev);
 void drm_gem_destroy(struct drm_device *dev);
 void drm_gem_object_release(struct drm_gem_object *obj);
 void drm_gem_object_free(struct kref *kref);
-struct drm_gem_object *drm_gem_object_alloc(struct drm_device *dev,
-					    size_t size);
 int drm_gem_object_init(struct drm_device *dev,
 			struct drm_gem_object *obj, size_t size);
 void drm_gem_private_object_init(struct drm_device *dev,
@@ -1645,10 +1625,11 @@ static __inline__ void drm_core_dropmap(struct drm_local_map *map)
 
 #include <drm/drm_mem_util.h>
 
-extern int drm_fill_in_dev(struct drm_device *dev,
-			   const struct pci_device_id *ent,
-			   struct drm_driver *driver);
-int drm_get_minor(struct drm_device *dev, struct drm_minor **minor, int type);
+struct drm_device *drm_dev_alloc(struct drm_driver *driver,
+				 struct device *parent);
+void drm_dev_free(struct drm_device *dev);
+int drm_dev_register(struct drm_device *dev, unsigned long flags);
+void drm_dev_unregister(struct drm_device *dev);
 /*@}*/
 
 /* PCI section */
diff --git a/include/drm/drm_crtc.h b/include/drm/drm_crtc.h
index 24f499569a2f..f32c5cd51f41 100644
--- a/include/drm/drm_crtc.h
+++ b/include/drm/drm_crtc.h
@@ -108,6 +108,7 @@ enum drm_mode_status {
     MODE_ONE_HEIGHT,    /* only one height is supported */
     MODE_ONE_SIZE,      /* only one resolution is supported */
     MODE_NO_REDUCED,    /* monitor doesn't accept reduced blanking */
+    MODE_NO_STEREO,	/* stereo modes not supported */
     MODE_UNVERIFIED = -3, /* mode needs to reverified */
     MODE_BAD = -2,	/* unspecified reason */
     MODE_ERROR	= -1	/* error condition */
@@ -124,7 +125,10 @@ enum drm_mode_status {
 	.vscan = (vs), .flags = (f), \
 	.base.type = DRM_MODE_OBJECT_MODE
 
-#define CRTC_INTERLACE_HALVE_V 0x1 /* halve V values for interlacing */
+#define CRTC_INTERLACE_HALVE_V	(1 << 0) /* halve V values for interlacing */
+#define CRTC_STEREO_DOUBLE	(1 << 1) /* adjust timings for stereo modes */
+
+#define DRM_MODE_FLAG_3D_MAX	DRM_MODE_FLAG_3D_SIDE_BY_SIDE_HALF
 
 struct drm_display_mode {
 	/* Header */
@@ -155,8 +159,7 @@ struct drm_display_mode {
 	int height_mm;
 
 	/* Actual mode we give to hw */
-	int clock_index;
-	int synth_clock;
+	int crtc_clock;		/* in KHz */
 	int crtc_hdisplay;
 	int crtc_hblank_start;
 	int crtc_hblank_end;
@@ -180,6 +183,11 @@ struct drm_display_mode {
 	int hsync;		/* in kHz */
 };
 
+static inline bool drm_mode_is_stereo(const struct drm_display_mode *mode)
+{
+	return mode->flags & DRM_MODE_FLAG_3D_MASK;
+}
+
 enum drm_connector_status {
 	connector_status_connected = 1,
 	connector_status_disconnected = 2,
@@ -587,7 +595,7 @@ enum drm_connector_force {
  */
 struct drm_connector {
 	struct drm_device *dev;
-	struct device kdev;
+	struct device *kdev;
 	struct device_attribute *attr;
 	struct list_head head;
 
@@ -597,6 +605,7 @@ struct drm_connector {
 	int connector_type_id;
 	bool interlace_allowed;
 	bool doublescan_allowed;
+	bool stereo_allowed;
 	struct list_head modes; /* list of modes on this connector */
 
 	enum drm_connector_status status;
@@ -964,6 +973,7 @@ extern int drm_mode_group_init_legacy_group(struct drm_device *dev, struct drm_m
 extern bool drm_probe_ddc(struct i2c_adapter *adapter);
 extern struct edid *drm_get_edid(struct drm_connector *connector,
 				 struct i2c_adapter *adapter);
+extern struct edid *drm_edid_duplicate(const struct edid *edid);
 extern int drm_add_edid_modes(struct drm_connector *connector, struct edid *edid);
 extern void drm_mode_probed_add(struct drm_connector *connector, struct drm_display_mode *mode);
 extern void drm_mode_copy(struct drm_display_mode *dst, const struct drm_display_mode *src);
@@ -975,7 +985,7 @@ extern void drm_mode_config_reset(struct drm_device *dev);
 extern void drm_mode_config_cleanup(struct drm_device *dev);
 extern void drm_mode_set_name(struct drm_display_mode *mode);
 extern bool drm_mode_equal(const struct drm_display_mode *mode1, const struct drm_display_mode *mode2);
-extern bool drm_mode_equal_no_clocks(const struct drm_display_mode *mode1, const struct drm_display_mode *mode2);
+extern bool drm_mode_equal_no_clocks_no_stereo(const struct drm_display_mode *mode1, const struct drm_display_mode *mode2);
 extern int drm_mode_width(const struct drm_display_mode *mode);
 extern int drm_mode_height(const struct drm_display_mode *mode);
 
@@ -1108,6 +1118,8 @@ extern struct drm_display_mode *drm_gtf_mode_complex(struct drm_device *dev,
 				int GTF_2C, int GTF_K, int GTF_2J);
 extern int drm_add_modes_noedid(struct drm_connector *connector,
 				int hdisplay, int vdisplay);
+extern void drm_set_preferred_mode(struct drm_connector *connector,
+				   int hpref, int vpref);
 
 extern int drm_edid_header_is_valid(const u8 *raw_edid);
 extern bool drm_edid_block_valid(u8 *raw_edid, int block, bool print_bad_edid);
@@ -1135,4 +1147,21 @@ extern int drm_format_horz_chroma_subsampling(uint32_t format);
 extern int drm_format_vert_chroma_subsampling(uint32_t format);
 extern const char *drm_get_format_name(uint32_t format);
 
+/* Helpers */
+static inline struct drm_crtc *drm_crtc_find(struct drm_device *dev,
+	uint32_t id)
+{
+	struct drm_mode_object *mo;
+	mo = drm_mode_object_find(dev, id, DRM_MODE_OBJECT_CRTC);
+	return mo ? obj_to_crtc(mo) : NULL;
+}
+
+static inline struct drm_encoder *drm_encoder_find(struct drm_device *dev,
+	uint32_t id)
+{
+	struct drm_mode_object *mo;
+	mo = drm_mode_object_find(dev, id, DRM_MODE_OBJECT_ENCODER);
+	return mo ? obj_to_encoder(mo) : NULL;
+}
+
 #endif /* __DRM_CRTC_H__ */
diff --git a/include/drm/drm_crtc_helper.h b/include/drm/drm_crtc_helper.h
index f43d556bf40b..ef6ad3a8e58e 100644
--- a/include/drm/drm_crtc_helper.h
+++ b/include/drm/drm_crtc_helper.h
@@ -163,7 +163,7 @@ static inline void drm_connector_helper_add(struct drm_connector *connector,
 extern int drm_helper_resume_force_mode(struct drm_device *dev);
 extern void drm_kms_helper_poll_init(struct drm_device *dev);
 extern void drm_kms_helper_poll_fini(struct drm_device *dev);
-extern void drm_helper_hpd_irq_event(struct drm_device *dev);
+extern bool drm_helper_hpd_irq_event(struct drm_device *dev);
 extern void drm_kms_helper_hotplug_event(struct drm_device *dev);
 
 extern void drm_kms_helper_poll_disable(struct drm_device *dev);
diff --git a/include/drm/drm_dp_helper.h b/include/drm/drm_dp_helper.h
index ae8dbfb1207c..a92c3754e3bb 100644
--- a/include/drm/drm_dp_helper.h
+++ b/include/drm/drm_dp_helper.h
@@ -77,10 +77,10 @@
 #define DP_DOWNSTREAMPORT_PRESENT           0x005
 # define DP_DWN_STRM_PORT_PRESENT           (1 << 0)
 # define DP_DWN_STRM_PORT_TYPE_MASK         0x06
-/* 00b = DisplayPort */
-/* 01b = Analog */
-/* 10b = TMDS or HDMI */
-/* 11b = Other */
+# define DP_DWN_STRM_PORT_TYPE_DP           (0 << 1)
+# define DP_DWN_STRM_PORT_TYPE_ANALOG       (1 << 1)
+# define DP_DWN_STRM_PORT_TYPE_TMDS         (2 << 1)
+# define DP_DWN_STRM_PORT_TYPE_OTHER        (3 << 1)
 # define DP_FORMAT_CONVERSION               (1 << 3)
 # define DP_DETAILED_CAP_INFO_AVAILABLE	    (1 << 4) /* DPI */
 
@@ -333,20 +333,20 @@ i2c_dp_aux_add_bus(struct i2c_adapter *adapter);
 
 
 #define DP_LINK_STATUS_SIZE	   6
-bool drm_dp_channel_eq_ok(u8 link_status[DP_LINK_STATUS_SIZE],
+bool drm_dp_channel_eq_ok(const u8 link_status[DP_LINK_STATUS_SIZE],
 			  int lane_count);
-bool drm_dp_clock_recovery_ok(u8 link_status[DP_LINK_STATUS_SIZE],
+bool drm_dp_clock_recovery_ok(const u8 link_status[DP_LINK_STATUS_SIZE],
 			      int lane_count);
-u8 drm_dp_get_adjust_request_voltage(u8 link_status[DP_LINK_STATUS_SIZE],
+u8 drm_dp_get_adjust_request_voltage(const u8 link_status[DP_LINK_STATUS_SIZE],
 				     int lane);
-u8 drm_dp_get_adjust_request_pre_emphasis(u8 link_status[DP_LINK_STATUS_SIZE],
+u8 drm_dp_get_adjust_request_pre_emphasis(const u8 link_status[DP_LINK_STATUS_SIZE],
 					  int lane);
 
 #define DP_RECEIVER_CAP_SIZE		0xf
 #define EDP_PSR_RECEIVER_CAP_SIZE	2
 
-void drm_dp_link_train_clock_recovery_delay(u8 dpcd[DP_RECEIVER_CAP_SIZE]);
-void drm_dp_link_train_channel_eq_delay(u8 dpcd[DP_RECEIVER_CAP_SIZE]);
+void drm_dp_link_train_clock_recovery_delay(const u8 dpcd[DP_RECEIVER_CAP_SIZE]);
+void drm_dp_link_train_channel_eq_delay(const u8 dpcd[DP_RECEIVER_CAP_SIZE]);
 
 u8 drm_dp_link_rate_to_bw_code(int link_rate);
 int drm_dp_bw_code_to_link_rate(u8 link_bw);
@@ -379,15 +379,22 @@ struct edp_vsc_psr {
 #define EDP_VSC_PSR_CRC_VALUES_VALID	(1<<2)
 
 static inline int
-drm_dp_max_link_rate(u8 dpcd[DP_RECEIVER_CAP_SIZE])
+drm_dp_max_link_rate(const u8 dpcd[DP_RECEIVER_CAP_SIZE])
 {
 	return drm_dp_bw_code_to_link_rate(dpcd[DP_MAX_LINK_RATE]);
 }
 
 static inline u8
-drm_dp_max_lane_count(u8 dpcd[DP_RECEIVER_CAP_SIZE])
+drm_dp_max_lane_count(const u8 dpcd[DP_RECEIVER_CAP_SIZE])
 {
 	return dpcd[DP_MAX_LANE_COUNT] & DP_MAX_LANE_COUNT_MASK;
 }
 
+static inline bool
+drm_dp_enhanced_frame_cap(const u8 dpcd[DP_RECEIVER_CAP_SIZE])
+{
+	return dpcd[DP_DPCD_REV] >= 0x11 &&
+		(dpcd[DP_MAX_LANE_COUNT] & DP_ENHANCED_FRAME_CAP);
+}
+
 #endif /* _DRM_DP_HELPER_H_ */
diff --git a/include/drm/drm_pciids.h b/include/drm/drm_pciids.h
index 3d79e513c0b3..87578c109e48 100644
--- a/include/drm/drm_pciids.h
+++ b/include/drm/drm_pciids.h
@@ -261,6 +261,18 @@
 	{0x1002, 0x679B, PCI_ANY_ID, PCI_ANY_ID, 0, 0, CHIP_TAHITI|RADEON_NEW_MEMMAP}, \
 	{0x1002, 0x679E, PCI_ANY_ID, PCI_ANY_ID, 0, 0, CHIP_TAHITI|RADEON_NEW_MEMMAP}, \
 	{0x1002, 0x679F, PCI_ANY_ID, PCI_ANY_ID, 0, 0, CHIP_TAHITI|RADEON_NEW_MEMMAP}, \
+	{0x1002, 0x67A0, PCI_ANY_ID, PCI_ANY_ID, 0, 0, CHIP_HAWAII|RADEON_NEW_MEMMAP}, \
+	{0x1002, 0x67A1, PCI_ANY_ID, PCI_ANY_ID, 0, 0, CHIP_HAWAII|RADEON_NEW_MEMMAP}, \
+	{0x1002, 0x67A2, PCI_ANY_ID, PCI_ANY_ID, 0, 0, CHIP_HAWAII|RADEON_NEW_MEMMAP}, \
+	{0x1002, 0x67A8, PCI_ANY_ID, PCI_ANY_ID, 0, 0, CHIP_HAWAII|RADEON_NEW_MEMMAP}, \
+	{0x1002, 0x67A9, PCI_ANY_ID, PCI_ANY_ID, 0, 0, CHIP_HAWAII|RADEON_NEW_MEMMAP}, \
+	{0x1002, 0x67AA, PCI_ANY_ID, PCI_ANY_ID, 0, 0, CHIP_HAWAII|RADEON_NEW_MEMMAP}, \
+	{0x1002, 0x67B0, PCI_ANY_ID, PCI_ANY_ID, 0, 0, CHIP_HAWAII|RADEON_NEW_MEMMAP}, \
+	{0x1002, 0x67B1, PCI_ANY_ID, PCI_ANY_ID, 0, 0, CHIP_HAWAII|RADEON_NEW_MEMMAP}, \
+	{0x1002, 0x67B8, PCI_ANY_ID, PCI_ANY_ID, 0, 0, CHIP_HAWAII|RADEON_NEW_MEMMAP}, \
+	{0x1002, 0x67B9, PCI_ANY_ID, PCI_ANY_ID, 0, 0, CHIP_HAWAII|RADEON_NEW_MEMMAP}, \
+	{0x1002, 0x67BA, PCI_ANY_ID, PCI_ANY_ID, 0, 0, CHIP_HAWAII|RADEON_NEW_MEMMAP}, \
+	{0x1002, 0x67BE, PCI_ANY_ID, PCI_ANY_ID, 0, 0, CHIP_HAWAII|RADEON_NEW_MEMMAP}, \
 	{0x1002, 0x6800, PCI_ANY_ID, PCI_ANY_ID, 0, 0, CHIP_PITCAIRN|RADEON_IS_MOBILITY|RADEON_NEW_MEMMAP}, \
 	{0x1002, 0x6801, PCI_ANY_ID, PCI_ANY_ID, 0, 0, CHIP_PITCAIRN|RADEON_IS_MOBILITY|RADEON_NEW_MEMMAP}, \
 	{0x1002, 0x6802, PCI_ANY_ID, PCI_ANY_ID, 0, 0, CHIP_PITCAIRN|RADEON_IS_MOBILITY|RADEON_NEW_MEMMAP}, \
diff --git a/include/drm/i915_drm.h b/include/drm/i915_drm.h
index 3abfa6ea226e..97d5497debc1 100644
--- a/include/drm/i915_drm.h
+++ b/include/drm/i915_drm.h
@@ -49,6 +49,10 @@ extern bool i915_gpu_turbo_disable(void);
 #define    SNB_GMCH_GGMS_MASK	0x3
 #define    SNB_GMCH_GMS_SHIFT   3 /* Graphics Mode Select */
 #define    SNB_GMCH_GMS_MASK    0x1f
+#define    BDW_GMCH_GGMS_SHIFT	6
+#define    BDW_GMCH_GGMS_MASK	0x3
+#define    BDW_GMCH_GMS_SHIFT   8
+#define    BDW_GMCH_GMS_MASK    0xff
 
 #define I830_GMCH_CTRL			0x52
 
diff --git a/include/drm/i915_pciids.h b/include/drm/i915_pciids.h
index 8a10f5c354e6..940ece4934ba 100644
--- a/include/drm/i915_pciids.h
+++ b/include/drm/i915_pciids.h
@@ -208,4 +208,29 @@
 #define INTEL_VLV_D_IDS(info) \
 	INTEL_VGA_DEVICE(0x0155, info)
 
+#define _INTEL_BDW_M(gt, id, info) \
+	INTEL_VGA_DEVICE((((gt) - 1) << 4) | (id), info)
+#define _INTEL_BDW_D(gt, id, info) \
+	INTEL_VGA_DEVICE((((gt) - 1) << 4) | (id), info)
+
+#define _INTEL_BDW_M_IDS(gt, info) \
+	_INTEL_BDW_M(gt, 0x1602, info), /* ULT */ \
+	_INTEL_BDW_M(gt, 0x1606, info), /* ULT */ \
+	_INTEL_BDW_M(gt, 0x160B, info), /* Iris */ \
+	_INTEL_BDW_M(gt, 0x160E, info) /* ULX */
+
+#define _INTEL_BDW_D_IDS(gt, info) \
+	_INTEL_BDW_D(gt, 0x160A, info), /* Server */ \
+	_INTEL_BDW_D(gt, 0x160D, info) /* Workstation */
+
+#define INTEL_BDW_M_IDS(info) \
+	_INTEL_BDW_M_IDS(1, info), \
+	_INTEL_BDW_M_IDS(2, info), \
+	_INTEL_BDW_M_IDS(3, info)
+
+#define INTEL_BDW_D_IDS(info) \
+	_INTEL_BDW_D_IDS(1, info), \
+	_INTEL_BDW_D_IDS(2, info), \
+	_INTEL_BDW_D_IDS(3, info)
+
 #endif /* _I915_PCIIDS_H */
diff --git a/include/drm/ttm/ttm_page_alloc.h b/include/drm/ttm/ttm_page_alloc.h
index 706b962c6467..d1f61bfe0ebe 100644
--- a/include/drm/ttm/ttm_page_alloc.h
+++ b/include/drm/ttm/ttm_page_alloc.h
@@ -62,7 +62,7 @@ extern void ttm_pool_unpopulate(struct ttm_tt *ttm);
 extern int ttm_page_alloc_debugfs(struct seq_file *m, void *data);
 
 
-#ifdef CONFIG_SWIOTLB
+#if defined(CONFIG_SWIOTLB) || defined(CONFIG_INTEL_IOMMU)
 /**
  * Initialize pool allocator.
  */
@@ -94,6 +94,15 @@ static inline int ttm_dma_page_alloc_debugfs(struct seq_file *m, void *data)
 {
 	return 0;
 }
+static inline int ttm_dma_populate(struct ttm_dma_tt *ttm_dma,
+				   struct device *dev)
+{
+	return -ENOMEM;
+}
+static inline void ttm_dma_unpopulate(struct ttm_dma_tt *ttm_dma,
+				      struct device *dev)
+{
+}
 #endif
 
 #endif
diff --git a/include/dt-bindings/mfd/as3722.h b/include/dt-bindings/mfd/as3722.h
new file mode 100644
index 000000000000..0e692562d77b
--- /dev/null
+++ b/include/dt-bindings/mfd/as3722.h
@@ -0,0 +1,52 @@
+/*
+ * This header provides macros for ams AS3722 device bindings.
+ *
+ * Copyright (c) 2013, NVIDIA Corporation.
+ *
+ * Author: Laxman Dewangan <ldewangan@nvidia.com>
+ *
+ */
+
+#ifndef __DT_BINDINGS_AS3722_H__
+#define __DT_BINDINGS_AS3722_H__
+
+/* External control pins */
+#define AS3722_EXT_CONTROL_PIN_ENABLE1 1
+#define AS3722_EXT_CONTROL_PIN_ENABLE2 2
+#define AS3722_EXT_CONTROL_PIN_ENABLE2 3
+
+/* Interrupt numbers for AS3722 */
+#define AS3722_IRQ_LID			0
+#define AS3722_IRQ_ACOK			1
+#define AS3722_IRQ_ENABLE1		2
+#define AS3722_IRQ_OCCUR_ALARM_SD0	3
+#define AS3722_IRQ_ONKEY_LONG_PRESS	4
+#define AS3722_IRQ_ONKEY		5
+#define AS3722_IRQ_OVTMP		6
+#define AS3722_IRQ_LOWBAT		7
+#define AS3722_IRQ_SD0_LV		8
+#define AS3722_IRQ_SD1_LV		9
+#define AS3722_IRQ_SD2_LV		10
+#define AS3722_IRQ_PWM1_OV_PROT		11
+#define AS3722_IRQ_PWM2_OV_PROT		12
+#define AS3722_IRQ_ENABLE2		13
+#define AS3722_IRQ_SD6_LV		14
+#define AS3722_IRQ_RTC_REP		15
+#define AS3722_IRQ_RTC_ALARM		16
+#define AS3722_IRQ_GPIO1		17
+#define AS3722_IRQ_GPIO2		18
+#define AS3722_IRQ_GPIO3		19
+#define AS3722_IRQ_GPIO4		20
+#define AS3722_IRQ_GPIO5		21
+#define AS3722_IRQ_WATCHDOG		22
+#define AS3722_IRQ_ENABLE3		23
+#define AS3722_IRQ_TEMP_SD0_SHUTDOWN	24
+#define AS3722_IRQ_TEMP_SD1_SHUTDOWN	25
+#define AS3722_IRQ_TEMP_SD2_SHUTDOWN	26
+#define AS3722_IRQ_TEMP_SD0_ALARM	27
+#define AS3722_IRQ_TEMP_SD1_ALARM	28
+#define AS3722_IRQ_TEMP_SD6_ALARM	29
+#define AS3722_IRQ_OCCUR_ALARM_SD6	30
+#define AS3722_IRQ_ADC			31
+
+#endif /* __DT_BINDINGS_AS3722_H__ */
diff --git a/include/linux/amba/serial.h b/include/linux/amba/serial.h
index 62d9303c2837..0ddb5c02ad8b 100644
--- a/include/linux/amba/serial.h
+++ b/include/linux/amba/serial.h
@@ -40,7 +40,7 @@
 #define UART010_LCRL		0x10	/* Line control register, low byte. */
 #define UART010_CR		0x14	/* Control register. */
 #define UART01x_FR		0x18	/* Flag register (Read only). */
-#define UART010_IIR		0x1C	/* Interrupt indentification register (Read). */
+#define UART010_IIR		0x1C	/* Interrupt identification register (Read). */
 #define UART010_ICR		0x1C	/* Interrupt clear register (Write). */
 #define ST_UART011_LCRH_RX	0x1C    /* Rx line control register. */
 #define UART01x_ILPR		0x20	/* IrDA low power counter register. */
diff --git a/include/linux/blkdev.h b/include/linux/blkdev.h
index f26ec20f6354..1b135d49b279 100644
--- a/include/linux/blkdev.h
+++ b/include/linux/blkdev.h
@@ -505,6 +505,9 @@ struct request_queue {
 				 (1 << QUEUE_FLAG_SAME_COMP)	|	\
 				 (1 << QUEUE_FLAG_ADD_RANDOM))
 
+#define QUEUE_FLAG_MQ_DEFAULT	((1 << QUEUE_FLAG_IO_STAT) |		\
+				 (1 << QUEUE_FLAG_SAME_COMP))
+
 static inline void queue_lockdep_assert_held(struct request_queue *q)
 {
 	if (q->queue_lock)
diff --git a/include/linux/cmdline-parser.h b/include/linux/cmdline-parser.h
index 98e892ef6d5a..a0f9280421ec 100644
--- a/include/linux/cmdline-parser.h
+++ b/include/linux/cmdline-parser.h
@@ -8,6 +8,8 @@
 #define CMDLINEPARSEH
 
 #include <linux/blkdev.h>
+#include <linux/fs.h>
+#include <linux/slab.h>
 
 /* partition flags */
 #define PF_RDONLY                   0x01 /* Device is read only */
diff --git a/include/linux/completion.h b/include/linux/completion.h
index 22c33e35bcb2..5d5aaae3af43 100644
--- a/include/linux/completion.h
+++ b/include/linux/completion.h
@@ -19,8 +19,8 @@
  *
  * See also:  complete(), wait_for_completion() (and friends _timeout,
  * _interruptible, _interruptible_timeout, and _killable), init_completion(),
- * and macros DECLARE_COMPLETION(), DECLARE_COMPLETION_ONSTACK(), and
- * INIT_COMPLETION().
+ * reinit_completion(), and macros DECLARE_COMPLETION(),
+ * DECLARE_COMPLETION_ONSTACK().
  */
 struct completion {
 	unsigned int done;
@@ -65,7 +65,7 @@ struct completion {
 
 /**
  * init_completion - Initialize a dynamically allocated completion
- * @x:  completion structure that is to be initialized
+ * @x:  pointer to completion structure that is to be initialized
  *
  * This inline function will initialize a dynamically created completion
  * structure.
@@ -76,6 +76,18 @@ static inline void init_completion(struct completion *x)
 	init_waitqueue_head(&x->wait);
 }
 
+/**
+ * reinit_completion - reinitialize a completion structure
+ * @x:  pointer to completion structure that is to be reinitialized
+ *
+ * This inline function should be used to reinitialize a completion structure so it can
+ * be reused. This is especially important after complete_all() is used.
+ */
+static inline void reinit_completion(struct completion *x)
+{
+	x->done = 0;
+}
+
 extern void wait_for_completion(struct completion *);
 extern void wait_for_completion_io(struct completion *);
 extern int wait_for_completion_interruptible(struct completion *x);
@@ -94,14 +106,4 @@ extern bool completion_done(struct completion *x);
 extern void complete(struct completion *);
 extern void complete_all(struct completion *);
 
-/**
- * INIT_COMPLETION - reinitialize a completion structure
- * @x:  completion structure to be reinitialized
- *
- * This macro should be used to reinitialize a completion structure so it can
- * be reused. This is especially important after complete_all() is used.
- */
-#define INIT_COMPLETION(x)	((x).done = 0)
-
-
 #endif
diff --git a/include/linux/cpufreq.h b/include/linux/cpufreq.h
index 5bd6ab9b0c27..dc196bbcf227 100644
--- a/include/linux/cpufreq.h
+++ b/include/linux/cpufreq.h
@@ -107,8 +107,16 @@ struct cpufreq_policy {
 #define CPUFREQ_SHARED_TYPE_ALL	 (2) /* All dependent CPUs should set freq */
 #define CPUFREQ_SHARED_TYPE_ANY	 (3) /* Freq can be set from any dependent CPU*/
 
+#ifdef CONFIG_CPU_FREQ
 struct cpufreq_policy *cpufreq_cpu_get(unsigned int cpu);
 void cpufreq_cpu_put(struct cpufreq_policy *policy);
+#else
+static inline struct cpufreq_policy *cpufreq_cpu_get(unsigned int cpu)
+{
+	return NULL;
+}
+static inline void cpufreq_cpu_put(struct cpufreq_policy *policy) { }
+#endif
 
 static inline bool policy_is_shared(struct cpufreq_policy *policy)
 {
diff --git a/include/linux/cpuset.h b/include/linux/cpuset.h
index cc1b01cf2035..3fe661fe96d1 100644
--- a/include/linux/cpuset.h
+++ b/include/linux/cpuset.h
@@ -110,10 +110,14 @@ static inline bool put_mems_allowed(unsigned int seq)
 
 static inline void set_mems_allowed(nodemask_t nodemask)
 {
+	unsigned long flags;
+
 	task_lock(current);
+	local_irq_save(flags);
 	write_seqcount_begin(&current->mems_allowed_seq);
 	current->mems_allowed = nodemask;
 	write_seqcount_end(&current->mems_allowed_seq);
+	local_irq_restore(flags);
 	task_unlock(current);
 }
 
diff --git a/include/linux/devfreq.h b/include/linux/devfreq.h
index 7a7cc74d7f27..d48dc00232a4 100644
--- a/include/linux/devfreq.h
+++ b/include/linux/devfreq.h
@@ -168,7 +168,7 @@ struct devfreq {
 	unsigned long max_freq;
 	bool stop_polling;
 
-	/* information for device freqeuncy transition */
+	/* information for device frequency transition */
 	unsigned int total_trans;
 	unsigned int *trans_table;
 	unsigned long *time_in_state;
diff --git a/include/linux/dmaengine.h b/include/linux/dmaengine.h
index 0bc727534108..41cf0c399288 100644
--- a/include/linux/dmaengine.h
+++ b/include/linux/dmaengine.h
@@ -45,13 +45,13 @@ static inline int dma_submit_error(dma_cookie_t cookie)
 
 /**
  * enum dma_status - DMA transaction status
- * @DMA_SUCCESS: transaction completed successfully
+ * @DMA_COMPLETE: transaction completed
  * @DMA_IN_PROGRESS: transaction not yet processed
  * @DMA_PAUSED: transaction is paused
  * @DMA_ERROR: transaction failed
  */
 enum dma_status {
-	DMA_SUCCESS,
+	DMA_COMPLETE,
 	DMA_IN_PROGRESS,
 	DMA_PAUSED,
 	DMA_ERROR,
@@ -171,12 +171,6 @@ struct dma_interleaved_template {
  * @DMA_CTRL_ACK - if clear, the descriptor cannot be reused until the client
  *  acknowledges receipt, i.e. has has a chance to establish any dependency
  *  chains
- * @DMA_COMPL_SKIP_SRC_UNMAP - set to disable dma-unmapping the source buffer(s)
- * @DMA_COMPL_SKIP_DEST_UNMAP - set to disable dma-unmapping the destination(s)
- * @DMA_COMPL_SRC_UNMAP_SINGLE - set to do the source dma-unmapping as single
- * 	(if not set, do the source dma-unmapping as page)
- * @DMA_COMPL_DEST_UNMAP_SINGLE - set to do the destination dma-unmapping as single
- * 	(if not set, do the destination dma-unmapping as page)
  * @DMA_PREP_PQ_DISABLE_P - prevent generation of P while generating Q
  * @DMA_PREP_PQ_DISABLE_Q - prevent generation of Q while generating P
  * @DMA_PREP_CONTINUE - indicate to a driver that it is reusing buffers as
@@ -188,14 +182,10 @@ struct dma_interleaved_template {
 enum dma_ctrl_flags {
 	DMA_PREP_INTERRUPT = (1 << 0),
 	DMA_CTRL_ACK = (1 << 1),
-	DMA_COMPL_SKIP_SRC_UNMAP = (1 << 2),
-	DMA_COMPL_SKIP_DEST_UNMAP = (1 << 3),
-	DMA_COMPL_SRC_UNMAP_SINGLE = (1 << 4),
-	DMA_COMPL_DEST_UNMAP_SINGLE = (1 << 5),
-	DMA_PREP_PQ_DISABLE_P = (1 << 6),
-	DMA_PREP_PQ_DISABLE_Q = (1 << 7),
-	DMA_PREP_CONTINUE = (1 << 8),
-	DMA_PREP_FENCE = (1 << 9),
+	DMA_PREP_PQ_DISABLE_P = (1 << 2),
+	DMA_PREP_PQ_DISABLE_Q = (1 << 3),
+	DMA_PREP_CONTINUE = (1 << 4),
+	DMA_PREP_FENCE = (1 << 5),
 };
 
 /**
@@ -413,6 +403,17 @@ void dma_chan_cleanup(struct kref *kref);
 typedef bool (*dma_filter_fn)(struct dma_chan *chan, void *filter_param);
 
 typedef void (*dma_async_tx_callback)(void *dma_async_param);
+
+struct dmaengine_unmap_data {
+	u8 to_cnt;
+	u8 from_cnt;
+	u8 bidi_cnt;
+	struct device *dev;
+	struct kref kref;
+	size_t len;
+	dma_addr_t addr[0];
+};
+
 /**
  * struct dma_async_tx_descriptor - async transaction descriptor
  * ---dma generic offload fields---
@@ -438,6 +439,7 @@ struct dma_async_tx_descriptor {
 	dma_cookie_t (*tx_submit)(struct dma_async_tx_descriptor *tx);
 	dma_async_tx_callback callback;
 	void *callback_param;
+	struct dmaengine_unmap_data *unmap;
 #ifdef CONFIG_ASYNC_TX_ENABLE_CHANNEL_SWITCH
 	struct dma_async_tx_descriptor *next;
 	struct dma_async_tx_descriptor *parent;
@@ -445,6 +447,40 @@ struct dma_async_tx_descriptor {
 #endif
 };
 
+#ifdef CONFIG_DMA_ENGINE
+static inline void dma_set_unmap(struct dma_async_tx_descriptor *tx,
+				 struct dmaengine_unmap_data *unmap)
+{
+	kref_get(&unmap->kref);
+	tx->unmap = unmap;
+}
+
+struct dmaengine_unmap_data *
+dmaengine_get_unmap_data(struct device *dev, int nr, gfp_t flags);
+void dmaengine_unmap_put(struct dmaengine_unmap_data *unmap);
+#else
+static inline void dma_set_unmap(struct dma_async_tx_descriptor *tx,
+				 struct dmaengine_unmap_data *unmap)
+{
+}
+static inline struct dmaengine_unmap_data *
+dmaengine_get_unmap_data(struct device *dev, int nr, gfp_t flags)
+{
+	return NULL;
+}
+static inline void dmaengine_unmap_put(struct dmaengine_unmap_data *unmap)
+{
+}
+#endif
+
+static inline void dma_descriptor_unmap(struct dma_async_tx_descriptor *tx)
+{
+	if (tx->unmap) {
+		dmaengine_unmap_put(tx->unmap);
+		tx->unmap = NULL;
+	}
+}
+
 #ifndef CONFIG_ASYNC_TX_ENABLE_CHANNEL_SWITCH
 static inline void txd_lock(struct dma_async_tx_descriptor *txd)
 {
@@ -979,10 +1015,10 @@ static inline enum dma_status dma_async_is_complete(dma_cookie_t cookie,
 {
 	if (last_complete <= last_used) {
 		if ((cookie <= last_complete) || (cookie > last_used))
-			return DMA_SUCCESS;
+			return DMA_COMPLETE;
 	} else {
 		if ((cookie <= last_complete) && (cookie > last_used))
-			return DMA_SUCCESS;
+			return DMA_COMPLETE;
 	}
 	return DMA_IN_PROGRESS;
 }
@@ -1013,11 +1049,11 @@ static inline struct dma_chan *dma_find_channel(enum dma_transaction_type tx_typ
 }
 static inline enum dma_status dma_sync_wait(struct dma_chan *chan, dma_cookie_t cookie)
 {
-	return DMA_SUCCESS;
+	return DMA_COMPLETE;
 }
 static inline enum dma_status dma_wait_for_async_tx(struct dma_async_tx_descriptor *tx)
 {
-	return DMA_SUCCESS;
+	return DMA_COMPLETE;
 }
 static inline void dma_issue_pending_all(void)
 {
diff --git a/include/linux/export.h b/include/linux/export.h
index 412cd509effe..3f2793d51899 100644
--- a/include/linux/export.h
+++ b/include/linux/export.h
@@ -43,7 +43,7 @@ extern struct module __this_module;
 /* Mark the CRC weak since genksyms apparently decides not to
  * generate a checksums for some symbols */
 #define __CRC_SYMBOL(sym, sec)					\
-	extern void *__crc_##sym __attribute__((weak));		\
+	extern __visible void *__crc_##sym __attribute__((weak));		\
 	static const unsigned long __kcrctab_##sym		\
 	__used							\
 	__attribute__((section("___kcrctab" sec "+" #sym), unused))	\
@@ -59,7 +59,7 @@ extern struct module __this_module;
 	static const char __kstrtab_##sym[]			\
 	__attribute__((section("__ksymtab_strings"), aligned(1))) \
 	= VMLINUX_SYMBOL_STR(sym);				\
-	static const struct kernel_symbol __ksymtab_##sym	\
+	__visible const struct kernel_symbol __ksymtab_##sym	\
 	__used							\
 	__attribute__((section("___ksymtab" sec "+" #sym), unused))	\
 	= { (unsigned long)&sym, __kstrtab_##sym }
diff --git a/include/linux/fb.h b/include/linux/fb.h
index ffac70aab3e9..70c4836e4a9f 100644
--- a/include/linux/fb.h
+++ b/include/linux/fb.h
@@ -792,4 +792,16 @@ extern int fb_find_mode(struct fb_var_screeninfo *var,
 			const struct fb_videomode *default_mode,
 			unsigned int default_bpp);
 
+/* Convenience logging macros */
+#define fb_err(fb_info, fmt, ...)					\
+	pr_err("fb%d: " fmt, (fb_info)->node, ##__VA_ARGS__)
+#define fb_notice(info, fmt, ...)					\
+	pr_notice("fb%d: " fmt, (fb_info)->node, ##__VA_ARGS__)
+#define fb_warn(fb_info, fmt, ...)					\
+	pr_warn("fb%d: " fmt, (fb_info)->node, ##__VA_ARGS__)
+#define fb_info(fb_info, fmt, ...)					\
+	pr_info("fb%d: " fmt, (fb_info)->node, ##__VA_ARGS__)
+#define fb_dbg(fb_info, fmt, ...)					\
+	pr_debug("fb%d: " fmt, (fb_info)->node, ##__VA_ARGS__)
+
 #endif /* _LINUX_FB_H */
diff --git a/include/linux/ftrace.h b/include/linux/ftrace.h
index 9f15c0064c50..31ea4b428360 100644
--- a/include/linux/ftrace.h
+++ b/include/linux/ftrace.h
@@ -533,11 +533,11 @@ static inline int ftrace_force_update(void) { return 0; }
 static inline void ftrace_disable_daemon(void) { }
 static inline void ftrace_enable_daemon(void) { }
 static inline void ftrace_release_mod(struct module *mod) {}
-static inline int register_ftrace_command(struct ftrace_func_command *cmd)
+static inline __init int register_ftrace_command(struct ftrace_func_command *cmd)
 {
 	return -EINVAL;
 }
-static inline int unregister_ftrace_command(char *cmd_name)
+static inline __init int unregister_ftrace_command(char *cmd_name)
 {
 	return -EINVAL;
 }
@@ -721,6 +721,7 @@ ftrace_push_return_trace(unsigned long ret, unsigned long func, int *depth,
 extern char __irqentry_text_start[];
 extern char __irqentry_text_end[];
 
+#define FTRACE_NOTRACE_DEPTH 65536
 #define FTRACE_RETFUNC_DEPTH 50
 #define FTRACE_RETSTACK_ALLOC_SIZE 32
 extern int register_ftrace_graph(trace_func_graph_ret_t retfunc,
diff --git a/include/linux/ftrace_event.h b/include/linux/ftrace_event.h
index 5eaa746735ff..9abbe630c456 100644
--- a/include/linux/ftrace_event.h
+++ b/include/linux/ftrace_event.h
@@ -202,6 +202,7 @@ enum {
 	TRACE_EVENT_FL_NO_SET_FILTER_BIT,
 	TRACE_EVENT_FL_IGNORE_ENABLE_BIT,
 	TRACE_EVENT_FL_WAS_ENABLED_BIT,
+	TRACE_EVENT_FL_USE_CALL_FILTER_BIT,
 };
 
 /*
@@ -213,6 +214,7 @@ enum {
  *  WAS_ENABLED   - Set and stays set when an event was ever enabled
  *                    (used for module unloading, if a module event is enabled,
  *                     it is best to clear the buffers that used it).
+ *  USE_CALL_FILTER - For ftrace internal events, don't use file filter
  */
 enum {
 	TRACE_EVENT_FL_FILTERED		= (1 << TRACE_EVENT_FL_FILTERED_BIT),
@@ -220,6 +222,7 @@ enum {
 	TRACE_EVENT_FL_NO_SET_FILTER	= (1 << TRACE_EVENT_FL_NO_SET_FILTER_BIT),
 	TRACE_EVENT_FL_IGNORE_ENABLE	= (1 << TRACE_EVENT_FL_IGNORE_ENABLE_BIT),
 	TRACE_EVENT_FL_WAS_ENABLED	= (1 << TRACE_EVENT_FL_WAS_ENABLED_BIT),
+	TRACE_EVENT_FL_USE_CALL_FILTER	= (1 << TRACE_EVENT_FL_USE_CALL_FILTER_BIT),
 };
 
 struct ftrace_event_call {
@@ -238,6 +241,7 @@ struct ftrace_event_call {
 	 *   bit 2:		failed to apply filter
 	 *   bit 3:		ftrace internal event (do not enable)
 	 *   bit 4:		Event was enabled by module
+	 *   bit 5:		use call filter rather than file filter
 	 */
 	int			flags; /* static flags of different events */
 
@@ -253,6 +257,8 @@ struct ftrace_subsystem_dir;
 enum {
 	FTRACE_EVENT_FL_ENABLED_BIT,
 	FTRACE_EVENT_FL_RECORDED_CMD_BIT,
+	FTRACE_EVENT_FL_FILTERED_BIT,
+	FTRACE_EVENT_FL_NO_SET_FILTER_BIT,
 	FTRACE_EVENT_FL_SOFT_MODE_BIT,
 	FTRACE_EVENT_FL_SOFT_DISABLED_BIT,
 };
@@ -261,6 +267,8 @@ enum {
  * Ftrace event file flags:
  *  ENABLED	  - The event is enabled
  *  RECORDED_CMD  - The comms should be recorded at sched_switch
+ *  FILTERED	  - The event has a filter attached
+ *  NO_SET_FILTER - Set when filter has error and is to be ignored
  *  SOFT_MODE     - The event is enabled/disabled by SOFT_DISABLED
  *  SOFT_DISABLED - When set, do not trace the event (even though its
  *                   tracepoint may be enabled)
@@ -268,6 +276,8 @@ enum {
 enum {
 	FTRACE_EVENT_FL_ENABLED		= (1 << FTRACE_EVENT_FL_ENABLED_BIT),
 	FTRACE_EVENT_FL_RECORDED_CMD	= (1 << FTRACE_EVENT_FL_RECORDED_CMD_BIT),
+	FTRACE_EVENT_FL_FILTERED	= (1 << FTRACE_EVENT_FL_FILTERED_BIT),
+	FTRACE_EVENT_FL_NO_SET_FILTER	= (1 << FTRACE_EVENT_FL_NO_SET_FILTER_BIT),
 	FTRACE_EVENT_FL_SOFT_MODE	= (1 << FTRACE_EVENT_FL_SOFT_MODE_BIT),
 	FTRACE_EVENT_FL_SOFT_DISABLED	= (1 << FTRACE_EVENT_FL_SOFT_DISABLED_BIT),
 };
@@ -275,6 +285,7 @@ enum {
 struct ftrace_event_file {
 	struct list_head		list;
 	struct ftrace_event_call	*event_call;
+	struct event_filter		*filter;
 	struct dentry			*dir;
 	struct trace_array		*tr;
 	struct ftrace_subsystem_dir	*system;
@@ -310,12 +321,16 @@ struct ftrace_event_file {
 
 #define MAX_FILTER_STR_VAL	256	/* Should handle KSYM_SYMBOL_LEN */
 
-extern void destroy_preds(struct ftrace_event_call *call);
+extern void destroy_preds(struct ftrace_event_file *file);
+extern void destroy_call_preds(struct ftrace_event_call *call);
 extern int filter_match_preds(struct event_filter *filter, void *rec);
-extern int filter_current_check_discard(struct ring_buffer *buffer,
-					struct ftrace_event_call *call,
-					void *rec,
-					struct ring_buffer_event *event);
+
+extern int filter_check_discard(struct ftrace_event_file *file, void *rec,
+				struct ring_buffer *buffer,
+				struct ring_buffer_event *event);
+extern int call_filter_check_discard(struct ftrace_event_call *call, void *rec,
+				     struct ring_buffer *buffer,
+				     struct ring_buffer_event *event);
 
 enum {
 	FILTER_OTHER = 0,
diff --git a/include/linux/genl_magic_func.h b/include/linux/genl_magic_func.h
index 023bc346b877..c0894dd8827b 100644
--- a/include/linux/genl_magic_func.h
+++ b/include/linux/genl_magic_func.h
@@ -273,49 +273,40 @@ static struct genl_family ZZZ_genl_family __read_mostly = {
  * Magic: define multicast groups
  * Magic: define multicast group registration helper
  */
+#define ZZZ_genl_mcgrps		CONCAT_(GENL_MAGIC_FAMILY, _genl_mcgrps)
+static const struct genl_multicast_group ZZZ_genl_mcgrps[] = {
+#undef GENL_mc_group
+#define GENL_mc_group(group) { .name = #group, },
+#include GENL_MAGIC_INCLUDE_FILE
+};
+
+enum CONCAT_(GENL_MAGIC_FAMILY, group_ids) {
+#undef GENL_mc_group
+#define GENL_mc_group(group) CONCAT_(GENL_MAGIC_FAMILY, _group_ ## group),
+#include GENL_MAGIC_INCLUDE_FILE
+};
+
 #undef GENL_mc_group
 #define GENL_mc_group(group)						\
-static struct genl_multicast_group					\
-CONCAT_(GENL_MAGIC_FAMILY, _mcg_ ## group) __read_mostly = {		\
-	.name = #group,							\
-};									\
 static int CONCAT_(GENL_MAGIC_FAMILY, _genl_multicast_ ## group)(	\
 	struct sk_buff *skb, gfp_t flags)				\
 {									\
 	unsigned int group_id =						\
-		CONCAT_(GENL_MAGIC_FAMILY, _mcg_ ## group).id;	\
-	if (!group_id)							\
-		return -EINVAL;						\
-	return genlmsg_multicast(skb, 0, group_id, flags);		\
+		CONCAT_(GENL_MAGIC_FAMILY, _group_ ## group);		\
+	return genlmsg_multicast(&ZZZ_genl_family, skb, 0,		\
+				 group_id, flags);			\
 }
 
 #include GENL_MAGIC_INCLUDE_FILE
 
-int CONCAT_(GENL_MAGIC_FAMILY, _genl_register)(void)
-{
-	int err = genl_register_family_with_ops(&ZZZ_genl_family,
-		ZZZ_genl_ops, ARRAY_SIZE(ZZZ_genl_ops));
-	if (err)
-		return err;
-#undef GENL_mc_group
-#define GENL_mc_group(group)						\
-	err = genl_register_mc_group(&ZZZ_genl_family,			\
-		&CONCAT_(GENL_MAGIC_FAMILY, _mcg_ ## group));		\
-	if (err)							\
-		goto fail;						\
-	else								\
-		pr_info("%s: mcg %s: %u\n", #group,			\
-			__stringify(GENL_MAGIC_FAMILY),			\
-			CONCAT_(GENL_MAGIC_FAMILY, _mcg_ ## group).id);
-
-#include GENL_MAGIC_INCLUDE_FILE
-
 #undef GENL_mc_group
 #define GENL_mc_group(group)
-	return 0;
-fail:
-	genl_unregister_family(&ZZZ_genl_family);
-	return err;
+
+int CONCAT_(GENL_MAGIC_FAMILY, _genl_register)(void)
+{
+	return genl_register_family_with_ops_groups(&ZZZ_genl_family,	\
+						    ZZZ_genl_ops,	\
+						    ZZZ_genl_mcgrps);
 }
 
 void CONCAT_(GENL_MAGIC_FAMILY, _genl_unregister)(void)
diff --git a/include/linux/host1x.h b/include/linux/host1x.h
new file mode 100644
index 000000000000..f5b9b87ac9a9
--- /dev/null
+++ b/include/linux/host1x.h
@@ -0,0 +1,284 @@
+/*
+ * Copyright (c) 2009-2013, NVIDIA Corporation. All rights reserved.
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation; either version 2 of the License, or
+ * (at your option) any later version.
+ *
+ * This program is distributed in the hope that it will be useful, but WITHOUT
+ * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or
+ * FITNESS FOR A PARTICULAR PURPOSE.  See the GNU General Public License for
+ * more details.
+ *
+ * You should have received a copy of the GNU General Public License along
+ * with this program; if not, write to the Free Software Foundation, Inc.,
+ * 51 Franklin Street, Fifth Floor, Boston, MA  02110-1301, USA.
+ */
+
+#ifndef __LINUX_HOST1X_H
+#define __LINUX_HOST1X_H
+
+#include <linux/device.h>
+#include <linux/types.h>
+
+enum host1x_class {
+	HOST1X_CLASS_HOST1X = 0x1,
+	HOST1X_CLASS_GR2D = 0x51,
+	HOST1X_CLASS_GR2D_SB = 0x52,
+	HOST1X_CLASS_GR3D = 0x60,
+};
+
+struct host1x_client;
+
+struct host1x_client_ops {
+	int (*init)(struct host1x_client *client);
+	int (*exit)(struct host1x_client *client);
+};
+
+struct host1x_client {
+	struct list_head list;
+	struct device *parent;
+	struct device *dev;
+
+	const struct host1x_client_ops *ops;
+
+	enum host1x_class class;
+	struct host1x_channel *channel;
+
+	struct host1x_syncpt **syncpts;
+	unsigned int num_syncpts;
+};
+
+/*
+ * host1x buffer objects
+ */
+
+struct host1x_bo;
+struct sg_table;
+
+struct host1x_bo_ops {
+	struct host1x_bo *(*get)(struct host1x_bo *bo);
+	void (*put)(struct host1x_bo *bo);
+	dma_addr_t (*pin)(struct host1x_bo *bo, struct sg_table **sgt);
+	void (*unpin)(struct host1x_bo *bo, struct sg_table *sgt);
+	void *(*mmap)(struct host1x_bo *bo);
+	void (*munmap)(struct host1x_bo *bo, void *addr);
+	void *(*kmap)(struct host1x_bo *bo, unsigned int pagenum);
+	void (*kunmap)(struct host1x_bo *bo, unsigned int pagenum, void *addr);
+};
+
+struct host1x_bo {
+	const struct host1x_bo_ops *ops;
+};
+
+static inline void host1x_bo_init(struct host1x_bo *bo,
+				  const struct host1x_bo_ops *ops)
+{
+	bo->ops = ops;
+}
+
+static inline struct host1x_bo *host1x_bo_get(struct host1x_bo *bo)
+{
+	return bo->ops->get(bo);
+}
+
+static inline void host1x_bo_put(struct host1x_bo *bo)
+{
+	bo->ops->put(bo);
+}
+
+static inline dma_addr_t host1x_bo_pin(struct host1x_bo *bo,
+				       struct sg_table **sgt)
+{
+	return bo->ops->pin(bo, sgt);
+}
+
+static inline void host1x_bo_unpin(struct host1x_bo *bo, struct sg_table *sgt)
+{
+	bo->ops->unpin(bo, sgt);
+}
+
+static inline void *host1x_bo_mmap(struct host1x_bo *bo)
+{
+	return bo->ops->mmap(bo);
+}
+
+static inline void host1x_bo_munmap(struct host1x_bo *bo, void *addr)
+{
+	bo->ops->munmap(bo, addr);
+}
+
+static inline void *host1x_bo_kmap(struct host1x_bo *bo, unsigned int pagenum)
+{
+	return bo->ops->kmap(bo, pagenum);
+}
+
+static inline void host1x_bo_kunmap(struct host1x_bo *bo,
+				    unsigned int pagenum, void *addr)
+{
+	bo->ops->kunmap(bo, pagenum, addr);
+}
+
+/*
+ * host1x syncpoints
+ */
+
+#define HOST1X_SYNCPT_CLIENT_MANAGED	(1 << 0)
+#define HOST1X_SYNCPT_HAS_BASE		(1 << 1)
+
+struct host1x_syncpt_base;
+struct host1x_syncpt;
+struct host1x;
+
+struct host1x_syncpt *host1x_syncpt_get(struct host1x *host, u32 id);
+u32 host1x_syncpt_id(struct host1x_syncpt *sp);
+u32 host1x_syncpt_read_min(struct host1x_syncpt *sp);
+u32 host1x_syncpt_read_max(struct host1x_syncpt *sp);
+int host1x_syncpt_incr(struct host1x_syncpt *sp);
+int host1x_syncpt_wait(struct host1x_syncpt *sp, u32 thresh, long timeout,
+		       u32 *value);
+struct host1x_syncpt *host1x_syncpt_request(struct device *dev,
+					    unsigned long flags);
+void host1x_syncpt_free(struct host1x_syncpt *sp);
+
+struct host1x_syncpt_base *host1x_syncpt_get_base(struct host1x_syncpt *sp);
+u32 host1x_syncpt_base_id(struct host1x_syncpt_base *base);
+
+/*
+ * host1x channel
+ */
+
+struct host1x_channel;
+struct host1x_job;
+
+struct host1x_channel *host1x_channel_request(struct device *dev);
+void host1x_channel_free(struct host1x_channel *channel);
+struct host1x_channel *host1x_channel_get(struct host1x_channel *channel);
+void host1x_channel_put(struct host1x_channel *channel);
+int host1x_job_submit(struct host1x_job *job);
+
+/*
+ * host1x job
+ */
+
+struct host1x_reloc {
+	struct host1x_bo *cmdbuf;
+	u32 cmdbuf_offset;
+	struct host1x_bo *target;
+	u32 target_offset;
+	u32 shift;
+	u32 pad;
+};
+
+struct host1x_job {
+	/* When refcount goes to zero, job can be freed */
+	struct kref ref;
+
+	/* List entry */
+	struct list_head list;
+
+	/* Channel where job is submitted to */
+	struct host1x_channel *channel;
+
+	u32 client;
+
+	/* Gathers and their memory */
+	struct host1x_job_gather *gathers;
+	unsigned int num_gathers;
+
+	/* Wait checks to be processed at submit time */
+	struct host1x_waitchk *waitchk;
+	unsigned int num_waitchk;
+	u32 waitchk_mask;
+
+	/* Array of handles to be pinned & unpinned */
+	struct host1x_reloc *relocarray;
+	unsigned int num_relocs;
+	struct host1x_job_unpin_data *unpins;
+	unsigned int num_unpins;
+
+	dma_addr_t *addr_phys;
+	dma_addr_t *gather_addr_phys;
+	dma_addr_t *reloc_addr_phys;
+
+	/* Sync point id, number of increments and end related to the submit */
+	u32 syncpt_id;
+	u32 syncpt_incrs;
+	u32 syncpt_end;
+
+	/* Maximum time to wait for this job */
+	unsigned int timeout;
+
+	/* Index and number of slots used in the push buffer */
+	unsigned int first_get;
+	unsigned int num_slots;
+
+	/* Copy of gathers */
+	size_t gather_copy_size;
+	dma_addr_t gather_copy;
+	u8 *gather_copy_mapped;
+
+	/* Check if register is marked as an address reg */
+	int (*is_addr_reg)(struct device *dev, u32 reg, u32 class);
+
+	/* Request a SETCLASS to this class */
+	u32 class;
+
+	/* Add a channel wait for previous ops to complete */
+	bool serialize;
+};
+
+struct host1x_job *host1x_job_alloc(struct host1x_channel *ch,
+				    u32 num_cmdbufs, u32 num_relocs,
+				    u32 num_waitchks);
+void host1x_job_add_gather(struct host1x_job *job, struct host1x_bo *mem_id,
+			   u32 words, u32 offset);
+struct host1x_job *host1x_job_get(struct host1x_job *job);
+void host1x_job_put(struct host1x_job *job);
+int host1x_job_pin(struct host1x_job *job, struct device *dev);
+void host1x_job_unpin(struct host1x_job *job);
+
+/*
+ * subdevice probe infrastructure
+ */
+
+struct host1x_device;
+
+struct host1x_driver {
+	const struct of_device_id *subdevs;
+	struct list_head list;
+	const char *name;
+
+	int (*probe)(struct host1x_device *device);
+	int (*remove)(struct host1x_device *device);
+};
+
+int host1x_driver_register(struct host1x_driver *driver);
+void host1x_driver_unregister(struct host1x_driver *driver);
+
+struct host1x_device {
+	struct host1x_driver *driver;
+	struct list_head list;
+	struct device dev;
+
+	struct mutex subdevs_lock;
+	struct list_head subdevs;
+	struct list_head active;
+
+	struct mutex clients_lock;
+	struct list_head clients;
+};
+
+static inline struct host1x_device *to_host1x_device(struct device *dev)
+{
+	return container_of(dev, struct host1x_device, dev);
+}
+
+int host1x_device_init(struct host1x_device *device);
+int host1x_device_exit(struct host1x_device *device);
+
+int host1x_client_register(struct host1x_client *client);
+int host1x_client_unregister(struct host1x_client *client);
+
+#endif
diff --git a/include/linux/huge_mm.h b/include/linux/huge_mm.h
index 3935428c57cf..91672e2deec3 100644
--- a/include/linux/huge_mm.h
+++ b/include/linux/huge_mm.h
@@ -54,7 +54,8 @@ enum page_check_address_pmd_flag {
 extern pmd_t *page_check_address_pmd(struct page *page,
 				     struct mm_struct *mm,
 				     unsigned long address,
-				     enum page_check_address_pmd_flag flag);
+				     enum page_check_address_pmd_flag flag,
+				     spinlock_t **ptl);
 
 #define HPAGE_PMD_ORDER (HPAGE_PMD_SHIFT-PAGE_SHIFT)
 #define HPAGE_PMD_NR (1<<HPAGE_PMD_ORDER)
@@ -129,15 +130,15 @@ extern void __vma_adjust_trans_huge(struct vm_area_struct *vma,
 				    unsigned long start,
 				    unsigned long end,
 				    long adjust_next);
-extern int __pmd_trans_huge_lock(pmd_t *pmd,
-				 struct vm_area_struct *vma);
+extern int __pmd_trans_huge_lock(pmd_t *pmd, struct vm_area_struct *vma,
+		spinlock_t **ptl);
 /* mmap_sem must be held on entry */
-static inline int pmd_trans_huge_lock(pmd_t *pmd,
-				      struct vm_area_struct *vma)
+static inline int pmd_trans_huge_lock(pmd_t *pmd, struct vm_area_struct *vma,
+		spinlock_t **ptl)
 {
 	VM_BUG_ON(!rwsem_is_locked(&vma->vm_mm->mmap_sem));
 	if (pmd_trans_huge(*pmd))
-		return __pmd_trans_huge_lock(pmd, vma);
+		return __pmd_trans_huge_lock(pmd, vma, ptl);
 	else
 		return 0;
 }
@@ -215,8 +216,8 @@ static inline void vma_adjust_trans_huge(struct vm_area_struct *vma,
 					 long adjust_next)
 {
 }
-static inline int pmd_trans_huge_lock(pmd_t *pmd,
-				      struct vm_area_struct *vma)
+static inline int pmd_trans_huge_lock(pmd_t *pmd, struct vm_area_struct *vma,
+		spinlock_t **ptl)
 {
 	return 0;
 }
diff --git a/include/linux/hugetlb.h b/include/linux/hugetlb.h
index 0393270466c3..acd2010328f3 100644
--- a/include/linux/hugetlb.h
+++ b/include/linux/hugetlb.h
@@ -392,6 +392,15 @@ static inline int hugepage_migration_support(struct hstate *h)
 	return pmd_huge_support() && (huge_page_shift(h) == PMD_SHIFT);
 }
 
+static inline spinlock_t *huge_pte_lockptr(struct hstate *h,
+					   struct mm_struct *mm, pte_t *pte)
+{
+	if (huge_page_size(h) == PMD_SIZE)
+		return pmd_lockptr(mm, (pmd_t *) pte);
+	VM_BUG_ON(huge_page_size(h) == PAGE_SIZE);
+	return &mm->page_table_lock;
+}
+
 #else	/* CONFIG_HUGETLB_PAGE */
 struct hstate {};
 #define alloc_huge_page_node(h, nid) NULL
@@ -401,6 +410,7 @@ struct hstate {};
 #define hstate_sizelog(s) NULL
 #define hstate_vma(v) NULL
 #define hstate_inode(i) NULL
+#define page_hstate(page) NULL
 #define huge_page_size(h) PAGE_SIZE
 #define huge_page_mask(h) PAGE_MASK
 #define vma_kernel_pagesize(v) PAGE_SIZE
@@ -421,6 +431,22 @@ static inline pgoff_t basepage_index(struct page *page)
 #define dissolve_free_huge_pages(s, e)	do {} while (0)
 #define pmd_huge_support()	0
 #define hugepage_migration_support(h)	0
+
+static inline spinlock_t *huge_pte_lockptr(struct hstate *h,
+					   struct mm_struct *mm, pte_t *pte)
+{
+	return &mm->page_table_lock;
+}
 #endif	/* CONFIG_HUGETLB_PAGE */
 
+static inline spinlock_t *huge_pte_lock(struct hstate *h,
+					struct mm_struct *mm, pte_t *pte)
+{
+	spinlock_t *ptl;
+
+	ptl = huge_pte_lockptr(h, mm, pte);
+	spin_lock(ptl);
+	return ptl;
+}
+
 #endif /* _LINUX_HUGETLB_H */
diff --git a/include/linux/i2c.h b/include/linux/i2c.h
index 2ab11dc38077..eff50e062be8 100644
--- a/include/linux/i2c.h
+++ b/include/linux/i2c.h
@@ -205,7 +205,6 @@ struct i2c_driver {
  * @name: Indicates the type of the device, usually a chip name that's
  *	generic enough to hide second-sourcing and compatible revisions.
  * @adapter: manages the bus segment hosting this I2C device
- * @driver: device's driver, hence pointer to access routines
  * @dev: Driver model device node for the slave.
  * @irq: indicates the IRQ generated by this device (if any)
  * @detected: member of an i2c_driver.clients list or i2c-core's
@@ -222,7 +221,6 @@ struct i2c_client {
 					/* _LOWER_ 7 bits		*/
 	char name[I2C_NAME_SIZE];
 	struct i2c_adapter *adapter;	/* the adapter we sit on	*/
-	struct i2c_driver *driver;	/* and our access routines	*/
 	struct device dev;		/* the device structure		*/
 	int irq;			/* irq issued by device		*/
 	struct list_head detected;
diff --git a/include/linux/if_macvlan.h b/include/linux/if_macvlan.h
index c2702856295e..84ba5ac39e03 100644
--- a/include/linux/if_macvlan.h
+++ b/include/linux/if_macvlan.h
@@ -119,4 +119,21 @@ extern int macvlan_link_register(struct rtnl_link_ops *ops);
 extern netdev_tx_t macvlan_start_xmit(struct sk_buff *skb,
 				      struct net_device *dev);
 
+#if IS_ENABLED(CONFIG_MACVLAN)
+static inline struct net_device *
+macvlan_dev_real_dev(const struct net_device *dev)
+{
+	struct macvlan_dev *macvlan = netdev_priv(dev);
+
+	return macvlan->lowerdev;
+}
+#else
+static inline struct net_device *
+macvlan_dev_real_dev(const struct net_device *dev)
+{
+	BUG();
+	return NULL;
+}
+#endif
+
 #endif /* _LINUX_IF_MACVLAN_H */
diff --git a/include/linux/init_task.h b/include/linux/init_task.h
index 5cd0f0949927..b0ed422e4e4a 100644
--- a/include/linux/init_task.h
+++ b/include/linux/init_task.h
@@ -32,10 +32,10 @@ extern struct fs_struct init_fs;
 #endif
 
 #ifdef CONFIG_CPUSETS
-#define INIT_CPUSET_SEQ							\
-	.mems_allowed_seq = SEQCNT_ZERO,
+#define INIT_CPUSET_SEQ(tsk)							\
+	.mems_allowed_seq = SEQCNT_ZERO(tsk.mems_allowed_seq),
 #else
-#define INIT_CPUSET_SEQ
+#define INIT_CPUSET_SEQ(tsk)
 #endif
 
 #define INIT_SIGNALS(sig) {						\
@@ -220,7 +220,7 @@ extern struct task_group root_task_group;
 	INIT_FTRACE_GRAPH						\
 	INIT_TRACE_RECURSION						\
 	INIT_TASK_RCU_PREEMPT(tsk)					\
-	INIT_CPUSET_SEQ							\
+	INIT_CPUSET_SEQ(tsk)						\
 	INIT_VTIME(tsk)							\
 }
 
diff --git a/include/linux/interrupt.h b/include/linux/interrupt.h
index c9e831dc80bc..db43b58a3355 100644
--- a/include/linux/interrupt.h
+++ b/include/linux/interrupt.h
@@ -11,8 +11,6 @@
 #include <linux/irqnr.h>
 #include <linux/hardirq.h>
 #include <linux/irqflags.h>
-#include <linux/smp.h>
-#include <linux/percpu.h>
 #include <linux/hrtimer.h>
 #include <linux/kref.h>
 #include <linux/workqueue.h>
@@ -392,15 +390,6 @@ extern void __raise_softirq_irqoff(unsigned int nr);
 extern void raise_softirq_irqoff(unsigned int nr);
 extern void raise_softirq(unsigned int nr);
 
-/* This is the worklist that queues up per-cpu softirq work.
- *
- * send_remote_sendirq() adds work to these lists, and
- * the softirq handler itself dequeues from them.  The queues
- * are protected by disabling local cpu interrupts and they must
- * only be accessed by the local cpu that they are for.
- */
-DECLARE_PER_CPU(struct list_head [NR_SOFTIRQS], softirq_work_list);
-
 DECLARE_PER_CPU(struct task_struct *, ksoftirqd);
 
 static inline struct task_struct *this_cpu_ksoftirqd(void)
@@ -408,17 +397,6 @@ static inline struct task_struct *this_cpu_ksoftirqd(void)
 	return this_cpu_read(ksoftirqd);
 }
 
-/* Try to send a softirq to a remote cpu.  If this cannot be done, the
- * work will be queued to the local cpu.
- */
-extern void send_remote_softirq(struct call_single_data *cp, int cpu, int softirq);
-
-/* Like send_remote_softirq(), but the caller must disable local cpu interrupts
- * and compute the current cpu, passed in as 'this_cpu'.
- */
-extern void __send_remote_softirq(struct call_single_data *cp, int cpu,
-				  int this_cpu, int softirq);
-
 /* Tasklets --- multithreaded analogue of BHs.
 
    Main feature differing them of generic softirqs: tasklet
diff --git a/include/linux/iommu.h b/include/linux/iommu.h
index 7ea319e95b47..a444c790fa72 100644
--- a/include/linux/iommu.h
+++ b/include/linux/iommu.h
@@ -22,6 +22,7 @@
 #include <linux/errno.h>
 #include <linux/err.h>
 #include <linux/types.h>
+#include <trace/events/iommu.h>
 
 #define IOMMU_READ	(1)
 #define IOMMU_WRITE	(2)
@@ -227,6 +228,7 @@ static inline int report_iommu_fault(struct iommu_domain *domain,
 		ret = domain->handler(domain, dev, iova, flags,
 						domain->handler_token);
 
+	trace_io_page_fault(dev, iova, flags);
 	return ret;
 }
 
diff --git a/include/linux/irq.h b/include/linux/irq.h
index 56bb0dc8b7d4..7dc10036eff5 100644
--- a/include/linux/irq.h
+++ b/include/linux/irq.h
@@ -70,6 +70,9 @@ typedef	void (*irq_preflow_handler_t)(struct irq_data *data);
  * IRQ_MOVE_PCNTXT		- Interrupt can be migrated from process context
  * IRQ_NESTED_TRHEAD		- Interrupt nests into another thread
  * IRQ_PER_CPU_DEVID		- Dev_id is a per-cpu variable
+ * IRQ_IS_POLLED		- Always polled by another interrupt. Exclude
+ *				  it from the spurious interrupt detection
+ *				  mechanism and from core side polling.
  */
 enum {
 	IRQ_TYPE_NONE		= 0x00000000,
@@ -94,12 +97,14 @@ enum {
 	IRQ_NESTED_THREAD	= (1 << 15),
 	IRQ_NOTHREAD		= (1 << 16),
 	IRQ_PER_CPU_DEVID	= (1 << 17),
+	IRQ_IS_POLLED		= (1 << 18),
 };
 
 #define IRQF_MODIFY_MASK	\
 	(IRQ_TYPE_SENSE_MASK | IRQ_NOPROBE | IRQ_NOREQUEST | \
 	 IRQ_NOAUTOEN | IRQ_MOVE_PCNTXT | IRQ_LEVEL | IRQ_NO_BALANCING | \
-	 IRQ_PER_CPU | IRQ_NESTED_THREAD | IRQ_NOTHREAD | IRQ_PER_CPU_DEVID)
+	 IRQ_PER_CPU | IRQ_NESTED_THREAD | IRQ_NOTHREAD | IRQ_PER_CPU_DEVID | \
+	 IRQ_IS_POLLED)
 
 #define IRQ_NO_BALANCING_MASK	(IRQ_PER_CPU | IRQ_NO_BALANCING)
 
diff --git a/include/linux/kernel.h b/include/linux/kernel.h
index 672ddc4de4af..d4e98d13eff4 100644
--- a/include/linux/kernel.h
+++ b/include/linux/kernel.h
@@ -501,7 +501,6 @@ void tracing_snapshot_alloc(void);
 
 extern void tracing_start(void);
 extern void tracing_stop(void);
-extern void ftrace_off_permanent(void);
 
 static inline __printf(1, 2)
 void ____trace_printk_check_format(const char *fmt, ...)
@@ -639,7 +638,6 @@ extern void ftrace_dump(enum ftrace_dump_mode oops_dump_mode);
 #else
 static inline void tracing_start(void) { }
 static inline void tracing_stop(void) { }
-static inline void ftrace_off_permanent(void) { }
 static inline void trace_dump_stack(int skip) { }
 
 static inline void tracing_on(void) { }
diff --git a/include/linux/kfifo.h b/include/linux/kfifo.h
index 10308c6a3d1c..552d51efb429 100644
--- a/include/linux/kfifo.h
+++ b/include/linux/kfifo.h
@@ -1,7 +1,7 @@
 /*
  * A generic kernel FIFO implementation
  *
- * Copyright (C) 2009/2010 Stefani Seibold <stefani@seibold.net>
+ * Copyright (C) 2013 Stefani Seibold <stefani@seibold.net>
  *
  * This program is free software; you can redistribute it and/or modify
  * it under the terms of the GNU General Public License as published by
@@ -67,9 +67,10 @@ struct __kfifo {
 	union { \
 		struct __kfifo	kfifo; \
 		datatype	*type; \
+		const datatype	*const_type; \
 		char		(*rectype)[recsize]; \
 		ptrtype		*ptr; \
-		const ptrtype	*ptr_const; \
+		ptrtype const	*ptr_const; \
 	}
 
 #define __STRUCT_KFIFO(type, size, recsize, ptrtype) \
@@ -386,16 +387,12 @@ __kfifo_int_must_check_helper( \
 #define	kfifo_put(fifo, val) \
 ({ \
 	typeof((fifo) + 1) __tmp = (fifo); \
-	typeof((val) + 1) __val = (val); \
+	typeof(*__tmp->const_type) __val = (val); \
 	unsigned int __ret; \
-	const size_t __recsize = sizeof(*__tmp->rectype); \
+	size_t __recsize = sizeof(*__tmp->rectype); \
 	struct __kfifo *__kfifo = &__tmp->kfifo; \
-	if (0) { \
-		typeof(__tmp->ptr_const) __dummy __attribute__ ((unused)); \
-		__dummy = (typeof(__val))NULL; \
-	} \
 	if (__recsize) \
-		__ret = __kfifo_in_r(__kfifo, __val, sizeof(*__val), \
+		__ret = __kfifo_in_r(__kfifo, &__val, sizeof(__val), \
 			__recsize); \
 	else { \
 		__ret = !kfifo_is_full(__tmp); \
@@ -404,7 +401,7 @@ __kfifo_int_must_check_helper( \
 			((typeof(__tmp->type))__kfifo->data) : \
 			(__tmp->buf) \
 			)[__kfifo->in & __tmp->kfifo.mask] = \
-				*(typeof(__tmp->type))__val; \
+				(typeof(*__tmp->type))__val; \
 			smp_wmb(); \
 			__kfifo->in++; \
 		} \
@@ -415,7 +412,7 @@ __kfifo_int_must_check_helper( \
 /**
  * kfifo_get - get data from the fifo
  * @fifo: address of the fifo to be used
- * @val: the var where to store the data to be added
+ * @val: address where to store the data
  *
  * This macro reads the data from the fifo.
  * It returns 0 if the fifo was empty. Otherwise it returns the number
@@ -428,12 +425,10 @@ __kfifo_int_must_check_helper( \
 __kfifo_uint_must_check_helper( \
 ({ \
 	typeof((fifo) + 1) __tmp = (fifo); \
-	typeof((val) + 1) __val = (val); \
+	typeof(__tmp->ptr) __val = (val); \
 	unsigned int __ret; \
 	const size_t __recsize = sizeof(*__tmp->rectype); \
 	struct __kfifo *__kfifo = &__tmp->kfifo; \
-	if (0) \
-		__val = (typeof(__tmp->ptr))0; \
 	if (__recsize) \
 		__ret = __kfifo_out_r(__kfifo, __val, sizeof(*__val), \
 			__recsize); \
@@ -456,7 +451,7 @@ __kfifo_uint_must_check_helper( \
 /**
  * kfifo_peek - get data from the fifo without removing
  * @fifo: address of the fifo to be used
- * @val: the var where to store the data to be added
+ * @val: address where to store the data
  *
  * This reads the data from the fifo without removing it from the fifo.
  * It returns 0 if the fifo was empty. Otherwise it returns the number
@@ -469,12 +464,10 @@ __kfifo_uint_must_check_helper( \
 __kfifo_uint_must_check_helper( \
 ({ \
 	typeof((fifo) + 1) __tmp = (fifo); \
-	typeof((val) + 1) __val = (val); \
+	typeof(__tmp->ptr) __val = (val); \
 	unsigned int __ret; \
 	const size_t __recsize = sizeof(*__tmp->rectype); \
 	struct __kfifo *__kfifo = &__tmp->kfifo; \
-	if (0) \
-		__val = (typeof(__tmp->ptr))NULL; \
 	if (__recsize) \
 		__ret = __kfifo_out_peek_r(__kfifo, __val, sizeof(*__val), \
 			__recsize); \
@@ -508,14 +501,10 @@ __kfifo_uint_must_check_helper( \
 #define	kfifo_in(fifo, buf, n) \
 ({ \
 	typeof((fifo) + 1) __tmp = (fifo); \
-	typeof((buf) + 1) __buf = (buf); \
+	typeof(__tmp->ptr_const) __buf = (buf); \
 	unsigned long __n = (n); \
 	const size_t __recsize = sizeof(*__tmp->rectype); \
 	struct __kfifo *__kfifo = &__tmp->kfifo; \
-	if (0) { \
-		typeof(__tmp->ptr_const) __dummy __attribute__ ((unused)); \
-		__dummy = (typeof(__buf))NULL; \
-	} \
 	(__recsize) ?\
 	__kfifo_in_r(__kfifo, __buf, __n, __recsize) : \
 	__kfifo_in(__kfifo, __buf, __n); \
@@ -561,14 +550,10 @@ __kfifo_uint_must_check_helper( \
 __kfifo_uint_must_check_helper( \
 ({ \
 	typeof((fifo) + 1) __tmp = (fifo); \
-	typeof((buf) + 1) __buf = (buf); \
+	typeof(__tmp->ptr) __buf = (buf); \
 	unsigned long __n = (n); \
 	const size_t __recsize = sizeof(*__tmp->rectype); \
 	struct __kfifo *__kfifo = &__tmp->kfifo; \
-	if (0) { \
-		typeof(__tmp->ptr) __dummy = NULL; \
-		__buf = __dummy; \
-	} \
 	(__recsize) ?\
 	__kfifo_out_r(__kfifo, __buf, __n, __recsize) : \
 	__kfifo_out(__kfifo, __buf, __n); \
@@ -773,14 +758,10 @@ __kfifo_uint_must_check_helper( \
 __kfifo_uint_must_check_helper( \
 ({ \
 	typeof((fifo) + 1) __tmp = (fifo); \
-	typeof((buf) + 1) __buf = (buf); \
+	typeof(__tmp->ptr) __buf = (buf); \
 	unsigned long __n = (n); \
 	const size_t __recsize = sizeof(*__tmp->rectype); \
 	struct __kfifo *__kfifo = &__tmp->kfifo; \
-	if (0) { \
-		typeof(__tmp->ptr) __dummy __attribute__ ((unused)) = NULL; \
-		__buf = __dummy; \
-	} \
 	(__recsize) ? \
 	__kfifo_out_peek_r(__kfifo, __buf, __n, __recsize) : \
 	__kfifo_out_peek(__kfifo, __buf, __n); \
diff --git a/include/linux/kvm_host.h b/include/linux/kvm_host.h
index 0fbbc7aa02cb..9523d2ad7535 100644
--- a/include/linux/kvm_host.h
+++ b/include/linux/kvm_host.h
@@ -142,7 +142,7 @@ struct kvm;
 struct kvm_vcpu;
 extern struct kmem_cache *kvm_vcpu_cache;
 
-extern raw_spinlock_t kvm_lock;
+extern spinlock_t kvm_lock;
 extern struct list_head vm_list;
 
 struct kvm_io_range {
@@ -189,8 +189,7 @@ struct kvm_async_pf {
 	gva_t gva;
 	unsigned long addr;
 	struct kvm_arch_async_pf arch;
-	struct page *page;
-	bool done;
+	bool   wakeup_all;
 };
 
 void kvm_clear_async_pf_completion_queue(struct kvm_vcpu *vcpu);
@@ -508,9 +507,10 @@ int kvm_set_memory_region(struct kvm *kvm,
 			  struct kvm_userspace_memory_region *mem);
 int __kvm_set_memory_region(struct kvm *kvm,
 			    struct kvm_userspace_memory_region *mem);
-void kvm_arch_free_memslot(struct kvm_memory_slot *free,
+void kvm_arch_free_memslot(struct kvm *kvm, struct kvm_memory_slot *free,
 			   struct kvm_memory_slot *dont);
-int kvm_arch_create_memslot(struct kvm_memory_slot *slot, unsigned long npages);
+int kvm_arch_create_memslot(struct kvm *kvm, struct kvm_memory_slot *slot,
+			    unsigned long npages);
 void kvm_arch_memslots_updated(struct kvm *kvm);
 int kvm_arch_prepare_memory_region(struct kvm *kvm,
 				struct kvm_memory_slot *memslot,
@@ -671,6 +671,25 @@ static inline void kvm_arch_free_vm(struct kvm *kvm)
 }
 #endif
 
+#ifdef __KVM_HAVE_ARCH_NONCOHERENT_DMA
+void kvm_arch_register_noncoherent_dma(struct kvm *kvm);
+void kvm_arch_unregister_noncoherent_dma(struct kvm *kvm);
+bool kvm_arch_has_noncoherent_dma(struct kvm *kvm);
+#else
+static inline void kvm_arch_register_noncoherent_dma(struct kvm *kvm)
+{
+}
+
+static inline void kvm_arch_unregister_noncoherent_dma(struct kvm *kvm)
+{
+}
+
+static inline bool kvm_arch_has_noncoherent_dma(struct kvm *kvm)
+{
+	return false;
+}
+#endif
+
 static inline wait_queue_head_t *kvm_arch_vcpu_wq(struct kvm_vcpu *vcpu)
 {
 #ifdef __KVM_HAVE_ARCH_WQP
@@ -747,9 +766,6 @@ void kvm_unregister_irq_ack_notifier(struct kvm *kvm,
 int kvm_request_irq_source_id(struct kvm *kvm);
 void kvm_free_irq_source_id(struct kvm *kvm, int irq_source_id);
 
-/* For vcpu->arch.iommu_flags */
-#define KVM_IOMMU_CACHE_COHERENCY	0x1
-
 #ifdef CONFIG_KVM_DEVICE_ASSIGNMENT
 int kvm_iommu_map_pages(struct kvm *kvm, struct kvm_memory_slot *slot);
 void kvm_iommu_unmap_pages(struct kvm *kvm, struct kvm_memory_slot *slot);
@@ -789,7 +805,7 @@ static inline void kvm_guest_enter(void)
 
 	/* KVM does not hold any references to rcu protected data when it
 	 * switches CPU into a guest mode. In fact switching to a guest mode
-	 * is very similar to exiting to userspase from rcu point of view. In
+	 * is very similar to exiting to userspace from rcu point of view. In
 	 * addition CPU may stay in a guest mode for quite a long time (up to
 	 * one time slice). Lets treat guest mode as quiescent state, just like
 	 * we do with user-mode execution.
@@ -842,13 +858,6 @@ static inline int memslot_id(struct kvm *kvm, gfn_t gfn)
 	return gfn_to_memslot(kvm, gfn)->id;
 }
 
-static inline gfn_t gfn_to_index(gfn_t gfn, gfn_t base_gfn, int level)
-{
-	/* KVM_HPAGE_GFN_SHIFT(PT_PAGE_TABLE_LEVEL) must be 0. */
-	return (gfn >> KVM_HPAGE_GFN_SHIFT(level)) -
-		(base_gfn >> KVM_HPAGE_GFN_SHIFT(level));
-}
-
 static inline gfn_t
 hva_to_gfn_memslot(unsigned long hva, struct kvm_memory_slot *slot)
 {
@@ -1066,6 +1075,7 @@ struct kvm_device *kvm_device_from_filp(struct file *filp);
 
 extern struct kvm_device_ops kvm_mpic_ops;
 extern struct kvm_device_ops kvm_xics_ops;
+extern struct kvm_device_ops kvm_vfio_ops;
 
 #ifdef CONFIG_HAVE_KVM_CPU_RELAX_INTERCEPT
 
diff --git a/include/linux/llist.h b/include/linux/llist.h
index 8828a78dec9a..fbf10a0bc095 100644
--- a/include/linux/llist.h
+++ b/include/linux/llist.h
@@ -195,4 +195,6 @@ static inline struct llist_node *llist_del_all(struct llist_head *head)
 
 extern struct llist_node *llist_del_first(struct llist_head *head);
 
+struct llist_node *llist_reverse_order(struct llist_node *head);
+
 #endif /* LLIST_H */
diff --git a/include/linux/lockdep.h b/include/linux/lockdep.h
index cfc2f119779a..92b1bfc5da60 100644
--- a/include/linux/lockdep.h
+++ b/include/linux/lockdep.h
@@ -497,6 +497,10 @@ static inline void print_irqtrace_events(struct task_struct *curr)
 #define rwlock_acquire_read(l, s, t, i)		lock_acquire_shared_recursive(l, s, t, NULL, i)
 #define rwlock_release(l, n, i)			lock_release(l, n, i)
 
+#define seqcount_acquire(l, s, t, i)		lock_acquire_exclusive(l, s, t, NULL, i)
+#define seqcount_acquire_read(l, s, t, i)	lock_acquire_shared_recursive(l, s, t, NULL, i)
+#define seqcount_release(l, n, i)		lock_release(l, n, i)
+
 #define mutex_acquire(l, s, t, i)		lock_acquire_exclusive(l, s, t, NULL, i)
 #define mutex_acquire_nest(l, s, t, n, i)	lock_acquire_exclusive(l, s, t, n, i)
 #define mutex_release(l, n, i)			lock_release(l, n, i)
@@ -504,11 +508,11 @@ static inline void print_irqtrace_events(struct task_struct *curr)
 #define rwsem_acquire(l, s, t, i)		lock_acquire_exclusive(l, s, t, NULL, i)
 #define rwsem_acquire_nest(l, s, t, n, i)	lock_acquire_exclusive(l, s, t, n, i)
 #define rwsem_acquire_read(l, s, t, i)		lock_acquire_shared(l, s, t, NULL, i)
-# define rwsem_release(l, n, i)			lock_release(l, n, i)
+#define rwsem_release(l, n, i)			lock_release(l, n, i)
 
 #define lock_map_acquire(l)			lock_acquire_exclusive(l, 0, 0, NULL, _THIS_IP_)
 #define lock_map_acquire_read(l)		lock_acquire_shared_recursive(l, 0, 0, NULL, _THIS_IP_)
-# define lock_map_release(l)			lock_release(l, 1, _THIS_IP_)
+#define lock_map_release(l)			lock_release(l, 1, _THIS_IP_)
 
 #ifdef CONFIG_PROVE_LOCKING
 # define might_lock(lock) 						\
diff --git a/include/linux/lockref.h b/include/linux/lockref.h
index 13dfd36a3294..c8929c3832db 100644
--- a/include/linux/lockref.h
+++ b/include/linux/lockref.h
@@ -15,10 +15,15 @@
  */
 
 #include <linux/spinlock.h>
+#include <generated/bounds.h>
+
+#define USE_CMPXCHG_LOCKREF \
+	(IS_ENABLED(CONFIG_ARCH_USE_CMPXCHG_LOCKREF) && \
+	 IS_ENABLED(CONFIG_SMP) && !BLOATED_SPINLOCKS)
 
 struct lockref {
 	union {
-#ifdef CONFIG_CMPXCHG_LOCKREF
+#if USE_CMPXCHG_LOCKREF
 		aligned_u64 lock_count;
 #endif
 		struct {
diff --git a/include/linux/mfd/arizona/registers.h b/include/linux/mfd/arizona/registers.h
index 4706d3d46e56..cb49417f8ba9 100644
--- a/include/linux/mfd/arizona/registers.h
+++ b/include/linux/mfd/arizona/registers.h
@@ -1908,7 +1908,7 @@
 #define ARIZONA_FLL2_SYNC_GAIN_MASK              0x003c  /* FLL2_SYNC_GAIN */
 #define ARIZONA_FLL2_SYNC_GAIN_SHIFT                  2  /* FLL2_SYNC_GAIN */
 #define ARIZONA_FLL2_SYNC_GAIN_WIDTH                  4  /* FLL2_SYNC_GAIN */
-#define ARIZONA_FLL2_SYNC_BW_MASK                0x0001  /* FLL2_SYNC_BW */
+#define ARIZONA_FLL2_SYNC_BW                     0x0001  /* FLL2_SYNC_BW */
 #define ARIZONA_FLL2_SYNC_BW_MASK                0x0001  /* FLL2_SYNC_BW */
 #define ARIZONA_FLL2_SYNC_BW_SHIFT                    0  /* FLL2_SYNC_BW */
 #define ARIZONA_FLL2_SYNC_BW_WIDTH                    1  /* FLL2_SYNC_BW */
diff --git a/include/linux/mfd/as3722.h b/include/linux/mfd/as3722.h
new file mode 100644
index 000000000000..16bf8a0dcd97
--- /dev/null
+++ b/include/linux/mfd/as3722.h
@@ -0,0 +1,423 @@
+/*
+ * as3722 definitions
+ *
+ * Copyright (C) 2013 ams
+ * Copyright (c) 2013, NVIDIA Corporation. All rights reserved.
+ *
+ * Author: Florian Lobmaier <florian.lobmaier@ams.com>
+ * Author: Laxman Dewangan <ldewangan@nvidia.com>
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation; either version 2 of the License, or
+ * (at your option) any later version.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+ * GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program; if not, write to the Free Software
+ * Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA  02111-1307 USA
+ *
+ */
+
+#ifndef __LINUX_MFD_AS3722_H__
+#define __LINUX_MFD_AS3722_H__
+
+#include <linux/regmap.h>
+
+/* AS3722 registers */
+#define AS3722_SD0_VOLTAGE_REG				0x00
+#define AS3722_SD1_VOLTAGE_REG				0x01
+#define AS3722_SD2_VOLTAGE_REG				0x02
+#define AS3722_SD3_VOLTAGE_REG				0x03
+#define AS3722_SD4_VOLTAGE_REG				0x04
+#define AS3722_SD5_VOLTAGE_REG				0x05
+#define AS3722_SD6_VOLTAGE_REG				0x06
+#define AS3722_GPIO0_CONTROL_REG			0x08
+#define AS3722_GPIO1_CONTROL_REG			0x09
+#define AS3722_GPIO2_CONTROL_REG			0x0A
+#define AS3722_GPIO3_CONTROL_REG			0x0B
+#define AS3722_GPIO4_CONTROL_REG			0x0C
+#define AS3722_GPIO5_CONTROL_REG			0x0D
+#define AS3722_GPIO6_CONTROL_REG			0x0E
+#define AS3722_GPIO7_CONTROL_REG			0x0F
+#define AS3722_LDO0_VOLTAGE_REG				0x10
+#define AS3722_LDO1_VOLTAGE_REG				0x11
+#define AS3722_LDO2_VOLTAGE_REG				0x12
+#define AS3722_LDO3_VOLTAGE_REG				0x13
+#define AS3722_LDO4_VOLTAGE_REG				0x14
+#define AS3722_LDO5_VOLTAGE_REG				0x15
+#define AS3722_LDO6_VOLTAGE_REG				0x16
+#define AS3722_LDO7_VOLTAGE_REG				0x17
+#define AS3722_LDO9_VOLTAGE_REG				0x19
+#define AS3722_LDO10_VOLTAGE_REG			0x1A
+#define AS3722_LDO11_VOLTAGE_REG			0x1B
+#define AS3722_GPIO_DEB1_REG				0x1E
+#define AS3722_GPIO_DEB2_REG				0x1F
+#define AS3722_GPIO_SIGNAL_OUT_REG			0x20
+#define AS3722_GPIO_SIGNAL_IN_REG			0x21
+#define AS3722_REG_SEQU_MOD1_REG			0x22
+#define AS3722_REG_SEQU_MOD2_REG			0x23
+#define AS3722_REG_SEQU_MOD3_REG			0x24
+#define AS3722_SD_PHSW_CTRL_REG				0x27
+#define AS3722_SD_PHSW_STATUS				0x28
+#define AS3722_SD0_CONTROL_REG				0x29
+#define AS3722_SD1_CONTROL_REG				0x2A
+#define AS3722_SDmph_CONTROL_REG			0x2B
+#define AS3722_SD23_CONTROL_REG				0x2C
+#define AS3722_SD4_CONTROL_REG				0x2D
+#define AS3722_SD5_CONTROL_REG				0x2E
+#define AS3722_SD6_CONTROL_REG				0x2F
+#define AS3722_SD_DVM_REG				0x30
+#define AS3722_RESET_REASON_REG				0x31
+#define AS3722_BATTERY_VOLTAGE_MONITOR_REG		0x32
+#define AS3722_STARTUP_CONTROL_REG			0x33
+#define AS3722_RESET_TIMER_REG				0x34
+#define AS3722_REFERENCE_CONTROL_REG			0x35
+#define AS3722_RESET_CONTROL_REG			0x36
+#define AS3722_OVER_TEMP_CONTROL_REG			0x37
+#define AS3722_WATCHDOG_CONTROL_REG			0x38
+#define AS3722_REG_STANDBY_MOD1_REG			0x39
+#define AS3722_REG_STANDBY_MOD2_REG			0x3A
+#define AS3722_REG_STANDBY_MOD3_REG			0x3B
+#define AS3722_ENABLE_CTRL1_REG				0x3C
+#define AS3722_ENABLE_CTRL2_REG				0x3D
+#define AS3722_ENABLE_CTRL3_REG				0x3E
+#define AS3722_ENABLE_CTRL4_REG				0x3F
+#define AS3722_ENABLE_CTRL5_REG				0x40
+#define AS3722_PWM_CONTROL_L_REG			0x41
+#define AS3722_PWM_CONTROL_H_REG			0x42
+#define AS3722_WATCHDOG_TIMER_REG			0x46
+#define AS3722_WATCHDOG_SOFTWARE_SIGNAL_REG		0x48
+#define AS3722_IOVOLTAGE_REG				0x49
+#define AS3722_BATTERY_VOLTAGE_MONITOR2_REG		0x4A
+#define AS3722_SD_CONTROL_REG				0x4D
+#define AS3722_LDOCONTROL0_REG				0x4E
+#define AS3722_LDOCONTROL1_REG				0x4F
+#define AS3722_SD0_PROTECT_REG				0x50
+#define AS3722_SD6_PROTECT_REG				0x51
+#define AS3722_PWM_VCONTROL1_REG			0x52
+#define AS3722_PWM_VCONTROL2_REG			0x53
+#define AS3722_PWM_VCONTROL3_REG			0x54
+#define AS3722_PWM_VCONTROL4_REG			0x55
+#define AS3722_BB_CHARGER_REG				0x57
+#define AS3722_CTRL_SEQU1_REG				0x58
+#define AS3722_CTRL_SEQU2_REG				0x59
+#define AS3722_OVCURRENT_REG				0x5A
+#define AS3722_OVCURRENT_DEB_REG			0x5B
+#define AS3722_SDLV_DEB_REG				0x5C
+#define AS3722_OC_PG_CTRL_REG				0x5D
+#define AS3722_OC_PG_CTRL2_REG				0x5E
+#define AS3722_CTRL_STATUS				0x5F
+#define AS3722_RTC_CONTROL_REG				0x60
+#define AS3722_RTC_SECOND_REG				0x61
+#define AS3722_RTC_MINUTE_REG				0x62
+#define AS3722_RTC_HOUR_REG				0x63
+#define AS3722_RTC_DAY_REG				0x64
+#define AS3722_RTC_MONTH_REG				0x65
+#define AS3722_RTC_YEAR_REG				0x66
+#define AS3722_RTC_ALARM_SECOND_REG			0x67
+#define AS3722_RTC_ALARM_MINUTE_REG			0x68
+#define AS3722_RTC_ALARM_HOUR_REG			0x69
+#define AS3722_RTC_ALARM_DAY_REG			0x6A
+#define AS3722_RTC_ALARM_MONTH_REG			0x6B
+#define AS3722_RTC_ALARM_YEAR_REG			0x6C
+#define AS3722_SRAM_REG					0x6D
+#define AS3722_RTC_ACCESS_REG				0x6F
+#define AS3722_RTC_STATUS_REG				0x73
+#define AS3722_INTERRUPT_MASK1_REG			0x74
+#define AS3722_INTERRUPT_MASK2_REG			0x75
+#define AS3722_INTERRUPT_MASK3_REG			0x76
+#define AS3722_INTERRUPT_MASK4_REG			0x77
+#define AS3722_INTERRUPT_STATUS1_REG			0x78
+#define AS3722_INTERRUPT_STATUS2_REG			0x79
+#define AS3722_INTERRUPT_STATUS3_REG			0x7A
+#define AS3722_INTERRUPT_STATUS4_REG			0x7B
+#define AS3722_TEMP_STATUS_REG				0x7D
+#define AS3722_ADC0_CONTROL_REG				0x80
+#define AS3722_ADC1_CONTROL_REG				0x81
+#define AS3722_ADC0_MSB_RESULT_REG			0x82
+#define AS3722_ADC0_LSB_RESULT_REG			0x83
+#define AS3722_ADC1_MSB_RESULT_REG			0x84
+#define AS3722_ADC1_LSB_RESULT_REG			0x85
+#define AS3722_ADC1_THRESHOLD_HI_MSB_REG		0x86
+#define AS3722_ADC1_THRESHOLD_HI_LSB_REG		0x87
+#define AS3722_ADC1_THRESHOLD_LO_MSB_REG		0x88
+#define AS3722_ADC1_THRESHOLD_LO_LSB_REG		0x89
+#define AS3722_ADC_CONFIGURATION_REG			0x8A
+#define AS3722_ASIC_ID1_REG				0x90
+#define AS3722_ASIC_ID2_REG				0x91
+#define AS3722_LOCK_REG					0x9E
+#define AS3722_MAX_REGISTER				0xF4
+
+#define AS3722_SD0_EXT_ENABLE_MASK			0x03
+#define AS3722_SD1_EXT_ENABLE_MASK			0x0C
+#define AS3722_SD2_EXT_ENABLE_MASK			0x30
+#define AS3722_SD3_EXT_ENABLE_MASK			0xC0
+#define AS3722_SD4_EXT_ENABLE_MASK			0x03
+#define AS3722_SD5_EXT_ENABLE_MASK			0x0C
+#define AS3722_SD6_EXT_ENABLE_MASK			0x30
+#define AS3722_LDO0_EXT_ENABLE_MASK			0x03
+#define AS3722_LDO1_EXT_ENABLE_MASK			0x0C
+#define AS3722_LDO2_EXT_ENABLE_MASK			0x30
+#define AS3722_LDO3_EXT_ENABLE_MASK			0xC0
+#define AS3722_LDO4_EXT_ENABLE_MASK			0x03
+#define AS3722_LDO5_EXT_ENABLE_MASK			0x0C
+#define AS3722_LDO6_EXT_ENABLE_MASK			0x30
+#define AS3722_LDO7_EXT_ENABLE_MASK			0xC0
+#define AS3722_LDO9_EXT_ENABLE_MASK			0x0C
+#define AS3722_LDO10_EXT_ENABLE_MASK			0x30
+#define AS3722_LDO11_EXT_ENABLE_MASK			0xC0
+
+#define AS3722_OVCURRENT_SD0_ALARM_MASK			0x07
+#define AS3722_OVCURRENT_SD0_ALARM_SHIFT		0x01
+#define AS3722_OVCURRENT_SD0_TRIP_MASK			0x18
+#define AS3722_OVCURRENT_SD0_TRIP_SHIFT			0x03
+#define AS3722_OVCURRENT_SD1_TRIP_MASK			0x60
+#define AS3722_OVCURRENT_SD1_TRIP_SHIFT			0x05
+
+#define AS3722_OVCURRENT_SD6_ALARM_MASK			0x07
+#define AS3722_OVCURRENT_SD6_ALARM_SHIFT		0x01
+#define AS3722_OVCURRENT_SD6_TRIP_MASK			0x18
+#define AS3722_OVCURRENT_SD6_TRIP_SHIFT			0x03
+
+/* AS3722 register bits and bit masks */
+#define AS3722_LDO_ILIMIT_MASK				BIT(7)
+#define AS3722_LDO_ILIMIT_BIT				BIT(7)
+#define AS3722_LDO0_VSEL_MASK				0x1F
+#define AS3722_LDO0_VSEL_MIN				0x01
+#define AS3722_LDO0_VSEL_MAX				0x12
+#define AS3722_LDO0_NUM_VOLT				0x12
+#define AS3722_LDO3_VSEL_MASK				0x3F
+#define AS3722_LDO3_VSEL_MIN				0x01
+#define AS3722_LDO3_VSEL_MAX				0x2D
+#define AS3722_LDO3_NUM_VOLT				0x2D
+#define AS3722_LDO_VSEL_MASK				0x7F
+#define AS3722_LDO_VSEL_MIN				0x01
+#define AS3722_LDO_VSEL_MAX				0x7F
+#define AS3722_LDO_VSEL_DNU_MIN				0x25
+#define AS3722_LDO_VSEL_DNU_MAX				0x3F
+#define AS3722_LDO_NUM_VOLT				0x80
+
+#define AS3722_LDO0_CTRL				BIT(0)
+#define AS3722_LDO1_CTRL				BIT(1)
+#define AS3722_LDO2_CTRL				BIT(2)
+#define AS3722_LDO3_CTRL				BIT(3)
+#define AS3722_LDO4_CTRL				BIT(4)
+#define AS3722_LDO5_CTRL				BIT(5)
+#define AS3722_LDO6_CTRL				BIT(6)
+#define AS3722_LDO7_CTRL				BIT(7)
+#define AS3722_LDO9_CTRL				BIT(1)
+#define AS3722_LDO10_CTRL				BIT(2)
+#define AS3722_LDO11_CTRL				BIT(3)
+
+#define AS3722_LDO3_MODE_MASK				(3 << 6)
+#define AS3722_LDO3_MODE_VAL(n)				(((n) & 0x3) << 6)
+#define AS3722_LDO3_MODE_PMOS				AS3722_LDO3_MODE_VAL(0)
+#define AS3722_LDO3_MODE_PMOS_TRACKING			AS3722_LDO3_MODE_VAL(1)
+#define AS3722_LDO3_MODE_NMOS				AS3722_LDO3_MODE_VAL(2)
+#define AS3722_LDO3_MODE_SWITCH				AS3722_LDO3_MODE_VAL(3)
+
+#define AS3722_SD_VSEL_MASK				0x7F
+#define AS3722_SD0_VSEL_MIN				0x01
+#define AS3722_SD0_VSEL_MAX				0x5A
+#define AS3722_SD2_VSEL_MIN				0x01
+#define AS3722_SD2_VSEL_MAX				0x7F
+
+#define AS3722_SDn_CTRL(n)				BIT(n)
+
+#define AS3722_SD0_MODE_FAST				BIT(4)
+#define AS3722_SD1_MODE_FAST				BIT(4)
+#define AS3722_SD2_MODE_FAST				BIT(2)
+#define AS3722_SD3_MODE_FAST				BIT(6)
+#define AS3722_SD4_MODE_FAST				BIT(2)
+#define AS3722_SD5_MODE_FAST				BIT(2)
+#define AS3722_SD6_MODE_FAST				BIT(4)
+
+#define AS3722_POWER_OFF				BIT(1)
+
+#define AS3722_INTERRUPT_MASK1_LID			BIT(0)
+#define AS3722_INTERRUPT_MASK1_ACOK			BIT(1)
+#define AS3722_INTERRUPT_MASK1_ENABLE1			BIT(2)
+#define AS3722_INTERRUPT_MASK1_OCURR_ALARM_SD0		BIT(3)
+#define AS3722_INTERRUPT_MASK1_ONKEY_LONG		BIT(4)
+#define AS3722_INTERRUPT_MASK1_ONKEY			BIT(5)
+#define AS3722_INTERRUPT_MASK1_OVTMP			BIT(6)
+#define AS3722_INTERRUPT_MASK1_LOWBAT			BIT(7)
+
+#define AS3722_INTERRUPT_MASK2_SD0_LV			BIT(0)
+#define AS3722_INTERRUPT_MASK2_SD1_LV			BIT(1)
+#define AS3722_INTERRUPT_MASK2_SD2345_LV		BIT(2)
+#define AS3722_INTERRUPT_MASK2_PWM1_OV_PROT		BIT(3)
+#define AS3722_INTERRUPT_MASK2_PWM2_OV_PROT		BIT(4)
+#define AS3722_INTERRUPT_MASK2_ENABLE2			BIT(5)
+#define AS3722_INTERRUPT_MASK2_SD6_LV			BIT(6)
+#define AS3722_INTERRUPT_MASK2_RTC_REP			BIT(7)
+
+#define AS3722_INTERRUPT_MASK3_RTC_ALARM		BIT(0)
+#define AS3722_INTERRUPT_MASK3_GPIO1			BIT(1)
+#define AS3722_INTERRUPT_MASK3_GPIO2			BIT(2)
+#define AS3722_INTERRUPT_MASK3_GPIO3			BIT(3)
+#define AS3722_INTERRUPT_MASK3_GPIO4			BIT(4)
+#define AS3722_INTERRUPT_MASK3_GPIO5			BIT(5)
+#define AS3722_INTERRUPT_MASK3_WATCHDOG			BIT(6)
+#define AS3722_INTERRUPT_MASK3_ENABLE3			BIT(7)
+
+#define AS3722_INTERRUPT_MASK4_TEMP_SD0_SHUTDOWN	BIT(0)
+#define AS3722_INTERRUPT_MASK4_TEMP_SD1_SHUTDOWN	BIT(1)
+#define AS3722_INTERRUPT_MASK4_TEMP_SD6_SHUTDOWN	BIT(2)
+#define AS3722_INTERRUPT_MASK4_TEMP_SD0_ALARM		BIT(3)
+#define AS3722_INTERRUPT_MASK4_TEMP_SD1_ALARM		BIT(4)
+#define AS3722_INTERRUPT_MASK4_TEMP_SD6_ALARM		BIT(5)
+#define AS3722_INTERRUPT_MASK4_OCCUR_ALARM_SD6		BIT(6)
+#define AS3722_INTERRUPT_MASK4_ADC			BIT(7)
+
+#define AS3722_ADC1_INTERVAL_TIME			BIT(0)
+#define AS3722_ADC1_INT_MODE_ON				BIT(1)
+#define AS3722_ADC_BUF_ON				BIT(2)
+#define AS3722_ADC1_LOW_VOLTAGE_RANGE			BIT(5)
+#define AS3722_ADC1_INTEVAL_SCAN			BIT(6)
+#define AS3722_ADC1_INT_MASK				BIT(7)
+
+#define AS3722_ADC_MSB_VAL_MASK				0x7F
+#define AS3722_ADC_LSB_VAL_MASK				0x07
+
+#define AS3722_ADC0_CONV_START				BIT(7)
+#define AS3722_ADC0_CONV_NOTREADY			BIT(7)
+#define AS3722_ADC0_SOURCE_SELECT_MASK			0x1F
+
+#define AS3722_ADC1_CONV_START				BIT(7)
+#define AS3722_ADC1_CONV_NOTREADY			BIT(7)
+#define AS3722_ADC1_SOURCE_SELECT_MASK			0x1F
+
+/* GPIO modes */
+#define AS3722_GPIO_MODE_MASK				0x07
+#define AS3722_GPIO_MODE_INPUT				0x00
+#define AS3722_GPIO_MODE_OUTPUT_VDDH			0x01
+#define AS3722_GPIO_MODE_IO_OPEN_DRAIN			0x02
+#define AS3722_GPIO_MODE_ADC_IN				0x03
+#define AS3722_GPIO_MODE_INPUT_PULL_UP			0x04
+#define AS3722_GPIO_MODE_INPUT_PULL_DOWN		0x05
+#define AS3722_GPIO_MODE_IO_OPEN_DRAIN_PULL_UP		0x06
+#define AS3722_GPIO_MODE_OUTPUT_VDDL			0x07
+#define AS3722_GPIO_MODE_VAL(n)			((n) & AS3722_GPIO_MODE_MASK)
+
+#define AS3722_GPIO_INV					BIT(7)
+#define AS3722_GPIO_IOSF_MASK				0x78
+#define AS3722_GPIO_IOSF_VAL(n)				(((n) & 0xF) << 3)
+#define AS3722_GPIO_IOSF_NORMAL				AS3722_GPIO_IOSF_VAL(0)
+#define AS3722_GPIO_IOSF_INTERRUPT_OUT			AS3722_GPIO_IOSF_VAL(1)
+#define AS3722_GPIO_IOSF_VSUP_LOW_OUT			AS3722_GPIO_IOSF_VAL(2)
+#define AS3722_GPIO_IOSF_GPIO_INTERRUPT_IN		AS3722_GPIO_IOSF_VAL(3)
+#define AS3722_GPIO_IOSF_ISINK_PWM_IN			AS3722_GPIO_IOSF_VAL(4)
+#define AS3722_GPIO_IOSF_VOLTAGE_STBY			AS3722_GPIO_IOSF_VAL(5)
+#define AS3722_GPIO_IOSF_PWR_GOOD_OUT			AS3722_GPIO_IOSF_VAL(7)
+#define AS3722_GPIO_IOSF_Q32K_OUT			AS3722_GPIO_IOSF_VAL(8)
+#define AS3722_GPIO_IOSF_WATCHDOG_IN			AS3722_GPIO_IOSF_VAL(9)
+#define AS3722_GPIO_IOSF_SOFT_RESET_IN			AS3722_GPIO_IOSF_VAL(11)
+#define AS3722_GPIO_IOSF_PWM_OUT			AS3722_GPIO_IOSF_VAL(12)
+#define AS3722_GPIO_IOSF_VSUP_LOW_DEB_OUT		AS3722_GPIO_IOSF_VAL(13)
+#define AS3722_GPIO_IOSF_SD6_LOW_VOLT_LOW		AS3722_GPIO_IOSF_VAL(14)
+
+#define AS3722_GPIOn_SIGNAL(n)				BIT(n)
+#define AS3722_GPIOn_CONTROL_REG(n)		(AS3722_GPIO0_CONTROL_REG + n)
+#define AS3722_I2C_PULL_UP				BIT(4)
+#define AS3722_INT_PULL_UP				BIT(5)
+
+#define AS3722_RTC_REP_WAKEUP_EN			BIT(0)
+#define AS3722_RTC_ALARM_WAKEUP_EN			BIT(1)
+#define AS3722_RTC_ON					BIT(2)
+#define AS3722_RTC_IRQMODE				BIT(3)
+#define AS3722_RTC_CLK32K_OUT_EN			BIT(5)
+
+#define AS3722_WATCHDOG_TIMER_MAX			0x7F
+#define AS3722_WATCHDOG_ON				BIT(0)
+#define AS3722_WATCHDOG_SW_SIG				BIT(0)
+
+#define AS3722_EXT_CONTROL_ENABLE1			0x1
+#define AS3722_EXT_CONTROL_ENABLE2			0x2
+#define AS3722_EXT_CONTROL_ENABLE3			0x3
+
+/* Interrupt IDs */
+enum as3722_irq {
+	AS3722_IRQ_LID,
+	AS3722_IRQ_ACOK,
+	AS3722_IRQ_ENABLE1,
+	AS3722_IRQ_OCCUR_ALARM_SD0,
+	AS3722_IRQ_ONKEY_LONG_PRESS,
+	AS3722_IRQ_ONKEY,
+	AS3722_IRQ_OVTMP,
+	AS3722_IRQ_LOWBAT,
+	AS3722_IRQ_SD0_LV,
+	AS3722_IRQ_SD1_LV,
+	AS3722_IRQ_SD2_LV,
+	AS3722_IRQ_PWM1_OV_PROT,
+	AS3722_IRQ_PWM2_OV_PROT,
+	AS3722_IRQ_ENABLE2,
+	AS3722_IRQ_SD6_LV,
+	AS3722_IRQ_RTC_REP,
+	AS3722_IRQ_RTC_ALARM,
+	AS3722_IRQ_GPIO1,
+	AS3722_IRQ_GPIO2,
+	AS3722_IRQ_GPIO3,
+	AS3722_IRQ_GPIO4,
+	AS3722_IRQ_GPIO5,
+	AS3722_IRQ_WATCHDOG,
+	AS3722_IRQ_ENABLE3,
+	AS3722_IRQ_TEMP_SD0_SHUTDOWN,
+	AS3722_IRQ_TEMP_SD1_SHUTDOWN,
+	AS3722_IRQ_TEMP_SD2_SHUTDOWN,
+	AS3722_IRQ_TEMP_SD0_ALARM,
+	AS3722_IRQ_TEMP_SD1_ALARM,
+	AS3722_IRQ_TEMP_SD6_ALARM,
+	AS3722_IRQ_OCCUR_ALARM_SD6,
+	AS3722_IRQ_ADC,
+	AS3722_IRQ_MAX,
+};
+
+struct as3722 {
+	struct device *dev;
+	struct regmap *regmap;
+	int chip_irq;
+	unsigned long irq_flags;
+	bool en_intern_int_pullup;
+	bool en_intern_i2c_pullup;
+	struct regmap_irq_chip_data *irq_data;
+};
+
+static inline int as3722_read(struct as3722 *as3722, u32 reg, u32 *dest)
+{
+	return regmap_read(as3722->regmap, reg, dest);
+}
+
+static inline int as3722_write(struct as3722 *as3722, u32 reg, u32 value)
+{
+	return regmap_write(as3722->regmap, reg, value);
+}
+
+static inline int as3722_block_read(struct as3722 *as3722, u32 reg,
+		int count, u8 *buf)
+{
+	return regmap_bulk_read(as3722->regmap, reg, buf, count);
+}
+
+static inline int as3722_block_write(struct as3722 *as3722, u32 reg,
+		int count, u8 *data)
+{
+	return regmap_bulk_write(as3722->regmap, reg, data, count);
+}
+
+static inline int as3722_update_bits(struct as3722 *as3722, u32 reg,
+		u32 mask, u8 val)
+{
+	return regmap_update_bits(as3722->regmap, reg, mask, val);
+}
+
+static inline int as3722_irq_get_virq(struct as3722 *as3722, int irq)
+{
+	return regmap_irq_get_virq(as3722->irq_data, irq);
+}
+#endif /* __LINUX_MFD_AS3722_H__ */
diff --git a/include/linux/mfd/core.h b/include/linux/mfd/core.h
index 7314fc4e6d25..bdba8c61207b 100644
--- a/include/linux/mfd/core.h
+++ b/include/linux/mfd/core.h
@@ -104,7 +104,7 @@ static inline const struct mfd_cell *mfd_get_cell(struct platform_device *pdev)
 }
 
 extern int mfd_add_devices(struct device *parent, int id,
-			   struct mfd_cell *cells, int n_devs,
+			   const struct mfd_cell *cells, int n_devs,
 			   struct resource *mem_base,
 			   int irq_base, struct irq_domain *irq_domain);
 
diff --git a/include/linux/mfd/da9052/da9052.h b/include/linux/mfd/da9052/da9052.h
index 786d02eb79d2..21e21b81cc75 100644
--- a/include/linux/mfd/da9052/da9052.h
+++ b/include/linux/mfd/da9052/da9052.h
@@ -148,10 +148,15 @@ static inline int da9052_group_read(struct da9052 *da9052, unsigned char reg,
 				     unsigned reg_cnt, unsigned char *val)
 {
 	int ret;
+	unsigned int tmp;
+	int i;
 
-	ret = regmap_bulk_read(da9052->regmap, reg, val, reg_cnt);
-	if (ret < 0)
-		return ret;
+	for (i = 0; i < reg_cnt; i++) {
+		ret = regmap_read(da9052->regmap, reg + i, &tmp);
+		val[i] = (unsigned char)tmp;
+		if (ret < 0)
+			return ret;
+	}
 
 	if (da9052->fix_io) {
 		ret = da9052->fix_io(da9052, reg);
@@ -166,10 +171,13 @@ static inline int da9052_group_write(struct da9052 *da9052, unsigned char reg,
 				      unsigned reg_cnt, unsigned char *val)
 {
 	int ret;
+	int i;
 
-	ret = regmap_raw_write(da9052->regmap, reg, val, reg_cnt);
-	if (ret < 0)
-		return ret;
+	for (i = 0; i < reg_cnt; i++) {
+		ret = regmap_write(da9052->regmap, reg + i, val[i]);
+		if (ret < 0)
+			return ret;
+	}
 
 	if (da9052->fix_io) {
 		ret = da9052->fix_io(da9052, reg);
diff --git a/include/linux/mfd/max77693-private.h b/include/linux/mfd/max77693-private.h
index 244fb0d51589..3e050b933dd0 100644
--- a/include/linux/mfd/max77693-private.h
+++ b/include/linux/mfd/max77693-private.h
@@ -323,7 +323,6 @@ struct max77693_dev {
 
 	int irq;
 	int irq_gpio;
-	bool wakeup;
 	struct mutex irqlock;
 	int irq_masks_cur[MAX77693_IRQ_GROUP_NR];
 	int irq_masks_cache[MAX77693_IRQ_GROUP_NR];
diff --git a/include/linux/mfd/max77693.h b/include/linux/mfd/max77693.h
index 676f0f388992..3f3dc45f93ee 100644
--- a/include/linux/mfd/max77693.h
+++ b/include/linux/mfd/max77693.h
@@ -64,8 +64,6 @@ struct max77693_muic_platform_data {
 };
 
 struct max77693_platform_data {
-	int wakeup;
-
 	/* regulator data */
 	struct max77693_regulator_data *regulators;
 	int num_regulators;
diff --git a/include/linux/mfd/rtsx_pci.h b/include/linux/mfd/rtsx_pci.h
index d1382dfbeff0..0ce772105508 100644
--- a/include/linux/mfd/rtsx_pci.h
+++ b/include/linux/mfd/rtsx_pci.h
@@ -756,6 +756,59 @@
 #define PCR_SETTING_REG2		0x814
 #define PCR_SETTING_REG3		0x747
 
+/* Phy bits */
+#define PHY_PCR_FORCE_CODE			0xB000
+#define PHY_PCR_OOBS_CALI_50			0x0800
+#define PHY_PCR_OOBS_VCM_08			0x0200
+#define PHY_PCR_OOBS_SEN_90			0x0040
+#define PHY_PCR_RSSI_EN				0x0002
+
+#define PHY_RCR1_ADP_TIME			0x0100
+#define PHY_RCR1_VCO_COARSE			0x001F
+
+#define PHY_RCR2_EMPHASE_EN			0x8000
+#define PHY_RCR2_NADJR				0x4000
+#define PHY_RCR2_CDR_CP_10			0x0400
+#define PHY_RCR2_CDR_SR_2			0x0100
+#define PHY_RCR2_FREQSEL_12			0x0040
+#define PHY_RCR2_CPADJEN			0x0020
+#define PHY_RCR2_CDR_SC_8			0x0008
+#define PHY_RCR2_CALIB_LATE			0x0002
+
+#define PHY_RDR_RXDSEL_1_9			0x4000
+
+#define PHY_TUNE_TUNEREF_1_0			0x4000
+#define PHY_TUNE_VBGSEL_1252			0x0C00
+#define PHY_TUNE_SDBUS_33			0x0200
+#define PHY_TUNE_TUNED18			0x01C0
+#define PHY_TUNE_TUNED12			0X0020
+
+#define PHY_BPCR_IBRXSEL			0x0400
+#define PHY_BPCR_IBTXSEL			0x0100
+#define PHY_BPCR_IB_FILTER			0x0080
+#define PHY_BPCR_CMIRROR_EN			0x0040
+
+#define PHY_REG_REV_RESV			0xE000
+#define PHY_REG_REV_RXIDLE_LATCHED		0x1000
+#define PHY_REG_REV_P1_EN			0x0800
+#define PHY_REG_REV_RXIDLE_EN			0x0400
+#define PHY_REG_REV_CLKREQ_DLY_TIMER_1_0	0x0040
+#define PHY_REG_REV_STOP_CLKRD			0x0020
+#define PHY_REG_REV_RX_PWST			0x0008
+#define PHY_REG_REV_STOP_CLKWR			0x0004
+
+#define PHY_FLD3_TIMER_4			0x7800
+#define PHY_FLD3_TIMER_6			0x00E0
+#define PHY_FLD3_RXDELINK			0x0004
+
+#define PHY_FLD4_FLDEN_SEL			0x4000
+#define PHY_FLD4_REQ_REF			0x2000
+#define PHY_FLD4_RXAMP_OFF			0x1000
+#define PHY_FLD4_REQ_ADDA			0x0800
+#define PHY_FLD4_BER_COUNT			0x00E0
+#define PHY_FLD4_BER_TIMER			0x000A
+#define PHY_FLD4_BER_CHK_EN			0x0001
+
 #define rtsx_pci_init_cmd(pcr)		((pcr)->ci = 0)
 
 struct rtsx_pcr;
diff --git a/include/linux/mfd/si476x-core.h b/include/linux/mfd/si476x-core.h
index ba89b94e4a56..674b45d5a757 100644
--- a/include/linux/mfd/si476x-core.h
+++ b/include/linux/mfd/si476x-core.h
@@ -316,7 +316,7 @@ enum si476x_smoothmetrics {
  * response to 'FM_RD_STATUS' command
  * @rdstpptyint: Traffic program flag(TP) and/or program type(PTY)
  * code has changed.
- * @rdspiint: Program indentifiaction(PI) code has changed.
+ * @rdspiint: Program identification(PI) code has changed.
  * @rdssyncint: RDS synchronization has changed.
  * @rdsfifoint: RDS was received and the RDS FIFO has at least
  * 'FM_RDS_INTERRUPT_FIFO_COUNT' elements in it.
diff --git a/include/linux/mfd/stw481x.h b/include/linux/mfd/stw481x.h
new file mode 100644
index 000000000000..eda121556e5d
--- /dev/null
+++ b/include/linux/mfd/stw481x.h
@@ -0,0 +1,56 @@
+/*
+ * Copyright (C) 2011 ST-Ericsson SA
+ * Written on behalf of Linaro for ST-Ericsson
+ *
+ * Author: Linus Walleij <linus.walleij@linaro.org>
+ *
+ * License terms: GNU General Public License (GPL) version 2
+ */
+#ifndef MFD_STW481X_H
+#define MFD_STW481X_H
+
+#include <linux/i2c.h>
+#include <linux/regulator/machine.h>
+#include <linux/regmap.h>
+#include <linux/bitops.h>
+
+/* These registers are accessed from more than one driver */
+#define STW_CONF1			0x11U
+#define STW_CONF1_PDN_VMMC		0x01U
+#define STW_CONF1_VMMC_MASK		0x0eU
+#define STW_CONF1_VMMC_1_8V		0x02U
+#define STW_CONF1_VMMC_2_85V		0x04U
+#define STW_CONF1_VMMC_3V		0x06U
+#define STW_CONF1_VMMC_1_85V		0x08U
+#define STW_CONF1_VMMC_2_6V		0x0aU
+#define STW_CONF1_VMMC_2_7V		0x0cU
+#define STW_CONF1_VMMC_3_3V		0x0eU
+#define STW_CONF1_MMC_LS_STATUS		0x10U
+#define STW_PCTL_REG_LO			0x1eU
+#define STW_PCTL_REG_HI			0x1fU
+#define STW_CONF1_V_MONITORING		0x20U
+#define STW_CONF1_IT_WARN		0x40U
+#define STW_CONF1_PDN_VAUX		0x80U
+#define STW_CONF2			0x20U
+#define STW_CONF2_MASK_TWARN		0x01U
+#define STW_CONF2_VMMC_EXT		0x02U
+#define STW_CONF2_MASK_IT_WAKE_UP	0x04U
+#define STW_CONF2_GPO1			0x08U
+#define STW_CONF2_GPO2			0x10U
+#define STW_VCORE_SLEEP			0x21U
+
+/**
+ * struct stw481x - state holder for the Stw481x drivers
+ * @mutex: mutex to serialize I2C accesses
+ * @i2c_client: corresponding I2C client
+ * @regulator: regulator device for regulator children
+ * @map: regmap handle to access device registers
+ */
+struct stw481x {
+	struct mutex		lock;
+	struct i2c_client	*client;
+	struct regulator_dev	*vmmc_regulator;
+	struct regmap		*map;
+};
+
+#endif
diff --git a/include/linux/mfd/syscon.h b/include/linux/mfd/syscon.h
index b473577f36db..8789fa3c7fd9 100644
--- a/include/linux/mfd/syscon.h
+++ b/include/linux/mfd/syscon.h
@@ -17,10 +17,35 @@
 
 struct device_node;
 
+#ifdef CONFIG_MFD_SYSCON
 extern struct regmap *syscon_node_to_regmap(struct device_node *np);
 extern struct regmap *syscon_regmap_lookup_by_compatible(const char *s);
 extern struct regmap *syscon_regmap_lookup_by_pdevname(const char *s);
 extern struct regmap *syscon_regmap_lookup_by_phandle(
 					struct device_node *np,
 					const char *property);
+#else
+static inline struct regmap *syscon_node_to_regmap(struct device_node *np)
+{
+	return ERR_PTR(-ENOSYS);
+}
+
+static inline struct regmap *syscon_regmap_lookup_by_compatible(const char *s)
+{
+	return ERR_PTR(-ENOSYS);
+}
+
+static inline struct regmap *syscon_regmap_lookup_by_pdevname(const char *s)
+{
+	return ERR_PTR(-ENOSYS);
+}
+
+static inline struct regmap *syscon_regmap_lookup_by_phandle(
+					struct device_node *np,
+					const char *property)
+{
+	return ERR_PTR(-ENOSYS);
+}
+#endif
+
 #endif /* __LINUX_MFD_SYSCON_H__ */
diff --git a/include/linux/mfd/ti_am335x_tscadc.h b/include/linux/mfd/ti_am335x_tscadc.h
index 08cce7f96ab9..d498d98f0c2c 100644
--- a/include/linux/mfd/ti_am335x_tscadc.h
+++ b/include/linux/mfd/ti_am335x_tscadc.h
@@ -134,13 +134,18 @@
 #define FIFO1_THRESHOLD		19
 
 /*
-* ADC runs at 3MHz, and it takes
-* 15 cycles to latch one data output.
-* Hence the idle time for ADC to
-* process one sample data would be
-* around 5 micro seconds.
-*/
-#define IDLE_TIMEOUT 5 /* microsec */
+ * time in us for processing a single channel, calculated as follows:
+ *
+ * num cycles = open delay + (sample delay + conv time) * averaging
+ *
+ * num cycles: 152 + (1 + 13) * 16 = 376
+ *
+ * clock frequency: 26MHz / 8 = 3.25MHz
+ * clock period: 1 / 3.25MHz = 308ns
+ *
+ * processing time: 376 * 308ns = 116us
+ */
+#define IDLE_TIMEOUT 116 /* microsec */
 
 #define TSCADC_CELLS		2
 
@@ -155,6 +160,7 @@ struct ti_tscadc_dev {
 	struct mfd_cell cells[TSCADC_CELLS];
 	u32 reg_se_cache;
 	spinlock_t reg_lock;
+	unsigned int clk_div;
 
 	/* tsc device */
 	struct titsc *tsc;
diff --git a/include/linux/mfd/wm8994/core.h b/include/linux/mfd/wm8994/core.h
index 40854ac0ba3d..eefafa62d304 100644
--- a/include/linux/mfd/wm8994/core.h
+++ b/include/linux/mfd/wm8994/core.h
@@ -56,8 +56,6 @@ struct irq_domain;
 #define WM8994_IRQ_GPIO(x) (x + WM8994_IRQ_TEMP_WARN)
 
 struct wm8994 {
-	struct mutex irq_lock;
-
 	struct wm8994_pdata pdata;
 
 	enum wm8994_type type;
@@ -85,16 +83,43 @@ struct wm8994 {
 };
 
 /* Device I/O API */
-int wm8994_reg_read(struct wm8994 *wm8994, unsigned short reg);
-int wm8994_reg_write(struct wm8994 *wm8994, unsigned short reg,
-		 unsigned short val);
-int wm8994_set_bits(struct wm8994 *wm8994, unsigned short reg,
-		    unsigned short mask, unsigned short val);
-int wm8994_bulk_read(struct wm8994 *wm8994, unsigned short reg,
-		     int count, u16 *buf);
-int wm8994_bulk_write(struct wm8994 *wm8994, unsigned short reg,
-		     int count, const u16 *buf);
 
+static inline int wm8994_reg_read(struct wm8994 *wm8994, unsigned short reg)
+{
+	unsigned int val;
+	int ret;
+
+	ret = regmap_read(wm8994->regmap, reg, &val);
+
+	if (ret < 0)
+		return ret;
+	else
+		return val;
+}
+
+static inline int wm8994_reg_write(struct wm8994 *wm8994, unsigned short reg,
+				   unsigned short val)
+{
+	return regmap_write(wm8994->regmap, reg, val);
+}
+
+static inline int wm8994_bulk_read(struct wm8994 *wm8994, unsigned short reg,
+				   int count, u16 *buf)
+{
+	return regmap_bulk_read(wm8994->regmap, reg, buf, count);
+}
+
+static inline int wm8994_bulk_write(struct wm8994 *wm8994, unsigned short reg,
+				    int count, const u16 *buf)
+{
+	return regmap_raw_write(wm8994->regmap, reg, buf, count * sizeof(u16));
+}
+
+static inline int wm8994_set_bits(struct wm8994 *wm8994, unsigned short reg,
+		    unsigned short mask, unsigned short val)
+{
+	return regmap_update_bits(wm8994->regmap, reg, mask, val);
+}
 
 /* Helper to save on boilerplate */
 static inline int wm8994_request_irq(struct wm8994 *wm8994, int irq,
diff --git a/include/linux/mlx5/device.h b/include/linux/mlx5/device.h
index 5eb4e31af22b..da78875807fc 100644
--- a/include/linux/mlx5/device.h
+++ b/include/linux/mlx5/device.h
@@ -230,6 +230,15 @@ enum {
 	MLX5_MAX_PAGE_SHIFT		= 31
 };
 
+enum {
+	MLX5_ADAPTER_PAGE_SHIFT		= 12
+};
+
+enum {
+	MLX5_CAP_OFF_DCT		= 41,
+	MLX5_CAP_OFF_CMDIF_CSUM		= 46,
+};
+
 struct mlx5_inbox_hdr {
 	__be16		opcode;
 	u8		rsvd[4];
@@ -319,9 +328,9 @@ struct mlx5_hca_cap {
 	u8	rsvd25[42];
 	__be16  log_uar_page_sz;
 	u8	rsvd26[28];
-	u8	log_msx_atomic_size_qp;
+	u8	log_max_atomic_size_qp;
 	u8	rsvd27[2];
-	u8	log_msx_atomic_size_dc;
+	u8	log_max_atomic_size_dc;
 	u8	rsvd28[76];
 };
 
diff --git a/include/linux/mlx5/driver.h b/include/linux/mlx5/driver.h
index 6b8c496572c8..554548cd3dd4 100644
--- a/include/linux/mlx5/driver.h
+++ b/include/linux/mlx5/driver.h
@@ -483,6 +483,7 @@ struct mlx5_priv {
 	struct rb_root		page_root;
 	int			fw_pages;
 	int			reg_pages;
+	struct list_head	free_list;
 
 	struct mlx5_core_health health;
 
@@ -557,9 +558,11 @@ typedef void (*mlx5_cmd_cbk_t)(int status, void *context);
 struct mlx5_cmd_work_ent {
 	struct mlx5_cmd_msg    *in;
 	struct mlx5_cmd_msg    *out;
+	void		       *uout;
+	int			uout_size;
 	mlx5_cmd_cbk_t		callback;
 	void		       *context;
-	int idx;
+	int			idx;
 	struct completion	done;
 	struct mlx5_cmd        *cmd;
 	struct work_struct	work;
@@ -570,6 +573,7 @@ struct mlx5_cmd_work_ent {
 	u8			token;
 	struct timespec		ts1;
 	struct timespec		ts2;
+	u16			op;
 };
 
 struct mlx5_pas {
@@ -653,6 +657,9 @@ void mlx5_cmd_use_polling(struct mlx5_core_dev *dev);
 int mlx5_cmd_status_to_err(struct mlx5_outbox_hdr *hdr);
 int mlx5_cmd_exec(struct mlx5_core_dev *dev, void *in, int in_size, void *out,
 		  int out_size);
+int mlx5_cmd_exec_cb(struct mlx5_core_dev *dev, void *in, int in_size,
+		     void *out, int out_size, mlx5_cmd_cbk_t callback,
+		     void *context);
 int mlx5_cmd_alloc_uar(struct mlx5_core_dev *dev, u32 *uarn);
 int mlx5_cmd_free_uar(struct mlx5_core_dev *dev, u32 uarn);
 int mlx5_alloc_uuars(struct mlx5_core_dev *dev, struct mlx5_uuar_info *uuari);
@@ -676,7 +683,9 @@ int mlx5_core_query_srq(struct mlx5_core_dev *dev, struct mlx5_core_srq *srq,
 int mlx5_core_arm_srq(struct mlx5_core_dev *dev, struct mlx5_core_srq *srq,
 		      u16 lwm, int is_srq);
 int mlx5_core_create_mkey(struct mlx5_core_dev *dev, struct mlx5_core_mr *mr,
-			  struct mlx5_create_mkey_mbox_in *in, int inlen);
+			  struct mlx5_create_mkey_mbox_in *in, int inlen,
+			  mlx5_cmd_cbk_t callback, void *context,
+			  struct mlx5_create_mkey_mbox_out *out);
 int mlx5_core_destroy_mkey(struct mlx5_core_dev *dev, struct mlx5_core_mr *mr);
 int mlx5_core_query_mkey(struct mlx5_core_dev *dev, struct mlx5_core_mr *mr,
 			 struct mlx5_query_mkey_mbox_out *out, int outlen);
@@ -745,6 +754,11 @@ static inline u32 mlx5_idx_to_mkey(u32 mkey_idx)
 	return mkey_idx << 8;
 }
 
+static inline u8 mlx5_mkey_variant(u32 mkey)
+{
+	return mkey & 0xff;
+}
+
 enum {
 	MLX5_PROF_MASK_QP_SIZE		= (u64)1 << 0,
 	MLX5_PROF_MASK_MR_CACHE		= (u64)1 << 1,
diff --git a/include/linux/mm.h b/include/linux/mm.h
index 42a35d94b82c..0548eb201e05 100644
--- a/include/linux/mm.h
+++ b/include/linux/mm.h
@@ -1316,32 +1316,85 @@ static inline pmd_t *pmd_alloc(struct mm_struct *mm, pud_t *pud, unsigned long a
 }
 #endif /* CONFIG_MMU && !__ARCH_HAS_4LEVEL_HACK */
 
-#if USE_SPLIT_PTLOCKS
-/*
- * We tuck a spinlock to guard each pagetable page into its struct page,
- * at page->private, with BUILD_BUG_ON to make sure that this will not
- * overflow into the next struct page (as it might with DEBUG_SPINLOCK).
- * When freeing, reset page->mapping so free_pages_check won't complain.
- */
-#define __pte_lockptr(page)	&((page)->ptl)
-#define pte_lock_init(_page)	do {					\
-	spin_lock_init(__pte_lockptr(_page));				\
-} while (0)
-#define pte_lock_deinit(page)	((page)->mapping = NULL)
-#define pte_lockptr(mm, pmd)	({(void)(mm); __pte_lockptr(pmd_page(*(pmd)));})
-#else	/* !USE_SPLIT_PTLOCKS */
+#if USE_SPLIT_PTE_PTLOCKS
+#if BLOATED_SPINLOCKS
+void __init ptlock_cache_init(void);
+extern bool ptlock_alloc(struct page *page);
+extern void ptlock_free(struct page *page);
+
+static inline spinlock_t *ptlock_ptr(struct page *page)
+{
+	return page->ptl;
+}
+#else /* BLOATED_SPINLOCKS */
+static inline void ptlock_cache_init(void) {}
+static inline bool ptlock_alloc(struct page *page)
+{
+	return true;
+}
+
+static inline void ptlock_free(struct page *page)
+{
+}
+
+static inline spinlock_t *ptlock_ptr(struct page *page)
+{
+	return &page->ptl;
+}
+#endif /* BLOATED_SPINLOCKS */
+
+static inline spinlock_t *pte_lockptr(struct mm_struct *mm, pmd_t *pmd)
+{
+	return ptlock_ptr(pmd_page(*pmd));
+}
+
+static inline bool ptlock_init(struct page *page)
+{
+	/*
+	 * prep_new_page() initialize page->private (and therefore page->ptl)
+	 * with 0. Make sure nobody took it in use in between.
+	 *
+	 * It can happen if arch try to use slab for page table allocation:
+	 * slab code uses page->slab_cache and page->first_page (for tail
+	 * pages), which share storage with page->ptl.
+	 */
+	VM_BUG_ON(*(unsigned long *)&page->ptl);
+	if (!ptlock_alloc(page))
+		return false;
+	spin_lock_init(ptlock_ptr(page));
+	return true;
+}
+
+/* Reset page->mapping so free_pages_check won't complain. */
+static inline void pte_lock_deinit(struct page *page)
+{
+	page->mapping = NULL;
+	ptlock_free(page);
+}
+
+#else	/* !USE_SPLIT_PTE_PTLOCKS */
 /*
  * We use mm->page_table_lock to guard all pagetable pages of the mm.
  */
-#define pte_lock_init(page)	do {} while (0)
-#define pte_lock_deinit(page)	do {} while (0)
-#define pte_lockptr(mm, pmd)	({(void)(pmd); &(mm)->page_table_lock;})
-#endif /* USE_SPLIT_PTLOCKS */
+static inline spinlock_t *pte_lockptr(struct mm_struct *mm, pmd_t *pmd)
+{
+	return &mm->page_table_lock;
+}
+static inline void ptlock_cache_init(void) {}
+static inline bool ptlock_init(struct page *page) { return true; }
+static inline void pte_lock_deinit(struct page *page) {}
+#endif /* USE_SPLIT_PTE_PTLOCKS */
+
+static inline void pgtable_init(void)
+{
+	ptlock_cache_init();
+	pgtable_cache_init();
+}
 
-static inline void pgtable_page_ctor(struct page *page)
+static inline bool pgtable_page_ctor(struct page *page)
 {
-	pte_lock_init(page);
 	inc_zone_page_state(page, NR_PAGETABLE);
+	return ptlock_init(page);
 }
 
 static inline void pgtable_page_dtor(struct page *page)
@@ -1378,6 +1431,52 @@ static inline void pgtable_page_dtor(struct page *page)
 	((unlikely(pmd_none(*(pmd))) && __pte_alloc_kernel(pmd, address))? \
 		NULL: pte_offset_kernel(pmd, address))
 
+#if USE_SPLIT_PMD_PTLOCKS
+
+static inline spinlock_t *pmd_lockptr(struct mm_struct *mm, pmd_t *pmd)
+{
+	return ptlock_ptr(virt_to_page(pmd));
+}
+
+static inline bool pgtable_pmd_page_ctor(struct page *page)
+{
+#ifdef CONFIG_TRANSPARENT_HUGEPAGE
+	page->pmd_huge_pte = NULL;
+#endif
+	return ptlock_init(page);
+}
+
+static inline void pgtable_pmd_page_dtor(struct page *page)
+{
+#ifdef CONFIG_TRANSPARENT_HUGEPAGE
+	VM_BUG_ON(page->pmd_huge_pte);
+#endif
+	ptlock_free(page);
+}
+
+#define pmd_huge_pte(mm, pmd) (virt_to_page(pmd)->pmd_huge_pte)
+
+#else
+
+static inline spinlock_t *pmd_lockptr(struct mm_struct *mm, pmd_t *pmd)
+{
+	return &mm->page_table_lock;
+}
+
+static inline bool pgtable_pmd_page_ctor(struct page *page) { return true; }
+static inline void pgtable_pmd_page_dtor(struct page *page) {}
+
+#define pmd_huge_pte(mm, pmd) ((mm)->pmd_huge_pte)
+
+#endif
+
+static inline spinlock_t *pmd_lock(struct mm_struct *mm, pmd_t *pmd)
+{
+	spinlock_t *ptl = pmd_lockptr(mm, pmd);
+	spin_lock(ptl);
+	return ptl;
+}
+
 extern void free_area_init(unsigned long * zones_size);
 extern void free_area_init_node(int nid, unsigned long * zones_size,
 		unsigned long zone_start_pfn, unsigned long *zholes_size);
diff --git a/include/linux/mm_types.h b/include/linux/mm_types.h
index a3198e5aaf4e..10f5a7272b80 100644
--- a/include/linux/mm_types.h
+++ b/include/linux/mm_types.h
@@ -23,7 +23,9 @@
 
 struct address_space;
 
-#define USE_SPLIT_PTLOCKS	(NR_CPUS >= CONFIG_SPLIT_PTLOCK_CPUS)
+#define USE_SPLIT_PTE_PTLOCKS	(NR_CPUS >= CONFIG_SPLIT_PTLOCK_CPUS)
+#define USE_SPLIT_PMD_PTLOCKS	(USE_SPLIT_PTE_PTLOCKS && \
+		IS_ENABLED(CONFIG_ARCH_ENABLE_SPLIT_PMD_PTLOCK))
 
 /*
  * Each physical page in the system has a struct page associated with
@@ -63,6 +65,9 @@ struct page {
 						 * this page is only used to
 						 * free other pages.
 						 */
+#if defined(CONFIG_TRANSPARENT_HUGEPAGE) && USE_SPLIT_PMD_PTLOCKS
+		pgtable_t pmd_huge_pte; /* protected by page->ptl */
+#endif
 		};
 
 		union {
@@ -141,9 +146,13 @@ struct page {
 						 * indicates order in the buddy
 						 * system if PG_buddy is set.
 						 */
-#if USE_SPLIT_PTLOCKS
+#if USE_SPLIT_PTE_PTLOCKS
+#if BLOATED_SPINLOCKS
+		spinlock_t *ptl;
+#else
 		spinlock_t ptl;
 #endif
+#endif
 		struct kmem_cache *slab_cache;	/* SL[AU]B: Pointer to slab */
 		struct page *first_page;	/* Compound tail pages */
 	};
@@ -309,14 +318,14 @@ enum {
 	NR_MM_COUNTERS
 };
 
-#if USE_SPLIT_PTLOCKS && defined(CONFIG_MMU)
+#if USE_SPLIT_PTE_PTLOCKS && defined(CONFIG_MMU)
 #define SPLIT_RSS_COUNTING
 /* per-thread cached information, */
 struct task_rss_stat {
 	int events;	/* for synchronization threshold */
 	int count[NR_MM_COUNTERS];
 };
-#endif /* USE_SPLIT_PTLOCKS */
+#endif /* USE_SPLIT_PTE_PTLOCKS */
 
 struct mm_rss_stat {
 	atomic_long_t count[NR_MM_COUNTERS];
@@ -339,6 +348,7 @@ struct mm_struct {
 	pgd_t * pgd;
 	atomic_t mm_users;			/* How many users with user space? */
 	atomic_t mm_count;			/* How many references to "struct mm_struct" (users count as 1) */
+	atomic_long_t nr_ptes;			/* Page table pages */
 	int map_count;				/* number of VMAs */
 
 	spinlock_t page_table_lock;		/* Protects page tables and some counters */
@@ -360,7 +370,6 @@ struct mm_struct {
 	unsigned long exec_vm;		/* VM_EXEC & ~VM_WRITE */
 	unsigned long stack_vm;		/* VM_GROWSUP/DOWN */
 	unsigned long def_flags;
-	unsigned long nr_ptes;		/* Page table pages */
 	unsigned long start_code, end_code, start_data, end_data;
 	unsigned long start_brk, brk, start_stack;
 	unsigned long arg_start, arg_end, env_start, env_end;
@@ -406,7 +415,7 @@ struct mm_struct {
 #ifdef CONFIG_MMU_NOTIFIER
 	struct mmu_notifier_mm *mmu_notifier_mm;
 #endif
-#ifdef CONFIG_TRANSPARENT_HUGEPAGE
+#if defined(CONFIG_TRANSPARENT_HUGEPAGE) && !USE_SPLIT_PMD_PTLOCKS
 	pgtable_t pmd_huge_pte; /* protected by page_table_lock */
 #endif
 #ifdef CONFIG_CPUMASK_OFFSTACK
diff --git a/include/linux/mmc/card.h b/include/linux/mmc/card.h
index 842de3e21e70..176fdf824b14 100644
--- a/include/linux/mmc/card.h
+++ b/include/linux/mmc/card.h
@@ -240,6 +240,7 @@ struct mmc_part {
 struct mmc_card {
 	struct mmc_host		*host;		/* the host this device belongs to */
 	struct device		dev;		/* the device */
+	u32			ocr;		/* the current OCR setting */
 	unsigned int		rca;		/* relative card address of device */
 	unsigned int		type;		/* card type */
 #define MMC_TYPE_MMC		0		/* MMC card */
@@ -257,6 +258,7 @@ struct mmc_card {
 #define MMC_CARD_REMOVED	(1<<7)		/* card has been removed */
 #define MMC_STATE_HIGHSPEED_200	(1<<8)		/* card is in HS200 mode */
 #define MMC_STATE_DOING_BKOPS	(1<<10)		/* card is doing BKOPS */
+#define MMC_STATE_SUSPENDED	(1<<11)		/* card is suspended */
 	unsigned int		quirks; 	/* card quirks */
 #define MMC_QUIRK_LENIENT_FN0	(1<<0)		/* allow SDIO FN0 writes outside of the VS CCCR range */
 #define MMC_QUIRK_BLKSZ_FOR_BYTE_MODE (1<<1)	/* use func->cur_blksize */
@@ -420,10 +422,10 @@ static inline void __maybe_unused remove_quirk(struct mmc_card *card, int data)
 #define mmc_card_blockaddr(c)	((c)->state & MMC_STATE_BLOCKADDR)
 #define mmc_card_ddr_mode(c)	((c)->state & MMC_STATE_HIGHSPEED_DDR)
 #define mmc_card_uhs(c)		((c)->state & MMC_STATE_ULTRAHIGHSPEED)
-#define mmc_sd_card_uhs(c)	((c)->state & MMC_STATE_ULTRAHIGHSPEED)
 #define mmc_card_ext_capacity(c) ((c)->state & MMC_CARD_SDXC)
 #define mmc_card_removed(c)	((c) && ((c)->state & MMC_CARD_REMOVED))
 #define mmc_card_doing_bkops(c)	((c)->state & MMC_STATE_DOING_BKOPS)
+#define mmc_card_suspended(c)	((c)->state & MMC_STATE_SUSPENDED)
 
 #define mmc_card_set_present(c)	((c)->state |= MMC_STATE_PRESENT)
 #define mmc_card_set_readonly(c) ((c)->state |= MMC_STATE_READONLY)
@@ -432,11 +434,12 @@ static inline void __maybe_unused remove_quirk(struct mmc_card *card, int data)
 #define mmc_card_set_blockaddr(c) ((c)->state |= MMC_STATE_BLOCKADDR)
 #define mmc_card_set_ddr_mode(c) ((c)->state |= MMC_STATE_HIGHSPEED_DDR)
 #define mmc_card_set_uhs(c) ((c)->state |= MMC_STATE_ULTRAHIGHSPEED)
-#define mmc_sd_card_set_uhs(c) ((c)->state |= MMC_STATE_ULTRAHIGHSPEED)
 #define mmc_card_set_ext_capacity(c) ((c)->state |= MMC_CARD_SDXC)
 #define mmc_card_set_removed(c) ((c)->state |= MMC_CARD_REMOVED)
 #define mmc_card_set_doing_bkops(c)	((c)->state |= MMC_STATE_DOING_BKOPS)
 #define mmc_card_clr_doing_bkops(c)	((c)->state &= ~MMC_STATE_DOING_BKOPS)
+#define mmc_card_set_suspended(c) ((c)->state |= MMC_STATE_SUSPENDED)
+#define mmc_card_clr_suspended(c) ((c)->state &= ~MMC_STATE_SUSPENDED)
 
 /*
  * Quirk add/remove for MMC products.
diff --git a/include/linux/mmc/core.h b/include/linux/mmc/core.h
index da51bec578c3..87079fc38011 100644
--- a/include/linux/mmc/core.h
+++ b/include/linux/mmc/core.h
@@ -151,7 +151,8 @@ extern int mmc_app_cmd(struct mmc_host *, struct mmc_card *);
 extern int mmc_wait_for_app_cmd(struct mmc_host *, struct mmc_card *,
 	struct mmc_command *, int);
 extern void mmc_start_bkops(struct mmc_card *card, bool from_exception);
-extern int __mmc_switch(struct mmc_card *, u8, u8, u8, unsigned int, bool);
+extern int __mmc_switch(struct mmc_card *, u8, u8, u8, unsigned int, bool,
+			bool);
 extern int mmc_switch(struct mmc_card *, u8, u8, u8, unsigned int);
 extern int mmc_send_ext_csd(struct mmc_card *card, u8 *ext_csd);
 
@@ -188,7 +189,6 @@ extern unsigned int mmc_align_data_size(struct mmc_card *, unsigned int);
 
 extern int __mmc_claim_host(struct mmc_host *host, atomic_t *abort);
 extern void mmc_release_host(struct mmc_host *host);
-extern int mmc_try_claim_host(struct mmc_host *host);
 
 extern void mmc_get_card(struct mmc_card *card);
 extern void mmc_put_card(struct mmc_card *card);
diff --git a/include/linux/mmc/dw_mmc.h b/include/linux/mmc/dw_mmc.h
index 198f0fa44e9f..6ce7d2cd3c7a 100644
--- a/include/linux/mmc/dw_mmc.h
+++ b/include/linux/mmc/dw_mmc.h
@@ -15,6 +15,7 @@
 #define LINUX_MMC_DW_MMC_H
 
 #include <linux/scatterlist.h>
+#include <linux/mmc/core.h>
 
 #define MAX_MCI_SLOTS	2
 
@@ -129,6 +130,9 @@ struct dw_mci {
 	struct mmc_request	*mrq;
 	struct mmc_command	*cmd;
 	struct mmc_data		*data;
+	struct mmc_command	stop_abort;
+	unsigned int		prev_blksz;
+	unsigned char		timing;
 	struct workqueue_struct	*card_workqueue;
 
 	/* DMA interface members*/
diff --git a/include/linux/mmc/host.h b/include/linux/mmc/host.h
index 3b0c33ae13e1..99f5709ac343 100644
--- a/include/linux/mmc/host.h
+++ b/include/linux/mmc/host.h
@@ -254,6 +254,7 @@ struct mmc_host {
 #define MMC_CAP_UHS_SDR50	(1 << 17)	/* Host supports UHS SDR50 mode */
 #define MMC_CAP_UHS_SDR104	(1 << 18)	/* Host supports UHS SDR104 mode */
 #define MMC_CAP_UHS_DDR50	(1 << 19)	/* Host supports UHS DDR50 mode */
+#define MMC_CAP_RUNTIME_RESUME	(1 << 20)	/* Resume at runtime_resume. */
 #define MMC_CAP_DRIVER_TYPE_A	(1 << 23)	/* Host supports Driver Type A */
 #define MMC_CAP_DRIVER_TYPE_C	(1 << 24)	/* Host supports Driver Type C */
 #define MMC_CAP_DRIVER_TYPE_D	(1 << 25)	/* Host supports Driver Type D */
@@ -309,7 +310,6 @@ struct mmc_host {
 	spinlock_t		lock;		/* lock for claim and bus ops */
 
 	struct mmc_ios		ios;		/* current io bus settings */
-	u32			ocr;		/* the current OCR setting */
 
 	/* group bitfields together to minimize padding */
 	unsigned int		use_spi_crc:1;
@@ -382,9 +382,6 @@ static inline void *mmc_priv(struct mmc_host *host)
 #define mmc_classdev(x)	(&(x)->class_dev)
 #define mmc_hostname(x)	(dev_name(&(x)->class_dev))
 
-int mmc_suspend_host(struct mmc_host *);
-int mmc_resume_host(struct mmc_host *);
-
 int mmc_power_save_host(struct mmc_host *host);
 int mmc_power_restore_host(struct mmc_host *host);
 
diff --git a/include/linux/module.h b/include/linux/module.h
index 05f2447f8c15..15cd6b1b211e 100644
--- a/include/linux/module.h
+++ b/include/linux/module.h
@@ -367,9 +367,6 @@ struct module
 	/* What modules do I depend on? */
 	struct list_head target_list;
 
-	/* Who is waiting for us to be unloaded */
-	struct task_struct *waiter;
-
 	/* Destruction function. */
 	void (*exit)(void);
 
diff --git a/include/linux/mutex.h b/include/linux/mutex.h
index bab49da8a0f0..d3181936c138 100644
--- a/include/linux/mutex.h
+++ b/include/linux/mutex.h
@@ -131,7 +131,7 @@ static inline int mutex_is_locked(struct mutex *lock)
 }
 
 /*
- * See kernel/mutex.c for detailed documentation of these APIs.
+ * See kernel/locking/mutex.c for detailed documentation of these APIs.
  * Also see Documentation/mutex-design.txt.
  */
 #ifdef CONFIG_DEBUG_LOCK_ALLOC
diff --git a/include/linux/netdevice.h b/include/linux/netdevice.h
index 8b3de7cc2ffc..7f0ed423a360 100644
--- a/include/linux/netdevice.h
+++ b/include/linux/netdevice.h
@@ -1587,7 +1587,7 @@ static inline void *netdev_priv(const struct net_device *dev)
 #define SET_NETDEV_DEV(net, pdev)	((net)->dev.parent = (pdev))
 
 /* Set the sysfs device type for the network logical device to allow
- * fin grained indentification of different network device types. For
+ * fine-grained identification of different network device types. For
  * example Ethernet, Wirelss LAN, Bluetooth, WiMAX etc.
  */
 #define SET_NETDEV_DEVTYPE(net, devtype)	((net)->dev.type = (devtype))
diff --git a/include/linux/nfs4.h b/include/linux/nfs4.h
index c6f41b616965..c1637062c1ce 100644
--- a/include/linux/nfs4.h
+++ b/include/linux/nfs4.h
@@ -118,6 +118,9 @@ Needs to be updated if more operations are defined in future.*/
 
 #define FIRST_NFS4_OP	OP_ACCESS
 #define LAST_NFS4_OP 	OP_RECLAIM_COMPLETE
+#define LAST_NFS40_OP	OP_RELEASE_LOCKOWNER
+#define LAST_NFS41_OP	OP_RECLAIM_COMPLETE
+#define LAST_NFS42_OP	OP_RECLAIM_COMPLETE
 
 enum nfsstat4 {
 	NFS4_OK = 0,
diff --git a/include/linux/i2c/at24.h b/include/linux/platform_data/at24.h
index 285025a9cdc9..c42aa89d34ee 100644
--- a/include/linux/i2c/at24.h
+++ b/include/linux/platform_data/at24.h
@@ -28,7 +28,7 @@
  *
  * void get_mac_addr(struct memory_accessor *mem_acc, void *context)
  * {
- * 	u8 *mac_addr = ethernet_pdata->mac_addr;
+ *	u8 *mac_addr = ethernet_pdata->mac_addr;
  *	off_t offset = context;
  *
  *	// Read MAC addr from EEPROM
diff --git a/include/linux/platform_data/edma.h b/include/linux/platform_data/edma.h
index 179fb91bb5f2..f50821cb64be 100644
--- a/include/linux/platform_data/edma.h
+++ b/include/linux/platform_data/edma.h
@@ -67,10 +67,10 @@ struct edmacc_param {
 #define ITCCHEN		BIT(23)
 
 /*ch_status paramater of callback function possible values*/
-#define DMA_COMPLETE 1
-#define DMA_CC_ERROR 2
-#define DMA_TC1_ERROR 3
-#define DMA_TC2_ERROR 4
+#define EDMA_DMA_COMPLETE 1
+#define EDMA_DMA_CC_ERROR 2
+#define EDMA_DMA_TC1_ERROR 3
+#define EDMA_DMA_TC2_ERROR 4
 
 enum address_mode {
 	INCR = 0,
diff --git a/include/linux/platform_data/mmc-esdhc-imx.h b/include/linux/platform_data/mmc-esdhc-imx.h
index d44912d81578..75f70f6ac137 100644
--- a/include/linux/platform_data/mmc-esdhc-imx.h
+++ b/include/linux/platform_data/mmc-esdhc-imx.h
@@ -10,6 +10,8 @@
 #ifndef __ASM_ARCH_IMX_ESDHC_H
 #define __ASM_ARCH_IMX_ESDHC_H
 
+#include <linux/types.h>
+
 enum wp_types {
 	ESDHC_WP_NONE,		/* no WP, neither controller nor gpio */
 	ESDHC_WP_CONTROLLER,	/* mmc controller internal WP */
@@ -32,6 +34,7 @@ enum cd_types {
  * @cd_gpio:	gpio for card_detect interrupt
  * @wp_type:	type of write_protect method (see wp_types enum above)
  * @cd_type:	type of card_detect method (see cd_types enum above)
+ * @support_vsel:  indicate it supports 1.8v switching
  */
 
 struct esdhc_platform_data {
@@ -41,5 +44,7 @@ struct esdhc_platform_data {
 	enum cd_types cd_type;
 	int max_bus_width;
 	unsigned int f_max;
+	bool support_vsel;
+	unsigned int delay_line;
 };
 #endif /* __ASM_ARCH_IMX_ESDHC_H */
diff --git a/include/linux/platform_data/zforce_ts.h b/include/linux/platform_data/zforce_ts.h
new file mode 100644
index 000000000000..0472ab2f6ede
--- /dev/null
+++ b/include/linux/platform_data/zforce_ts.h
@@ -0,0 +1,26 @@
+/* drivers/input/touchscreen/zforce.c
+ *
+ * Copyright (C) 2012-2013 MundoReader S.L.
+ *
+ * This software is licensed under the terms of the GNU General Public
+ * License version 2, as published by the Free Software Foundation, and
+ * may be copied, distributed, and modified under those terms.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+ * GNU General Public License for more details.
+ */
+
+#ifndef _LINUX_INPUT_ZFORCE_TS_H
+#define _LINUX_INPUT_ZFORCE_TS_H
+
+struct zforce_ts_platdata {
+	int gpio_int;
+	int gpio_rst;
+
+	unsigned int x_max;
+	unsigned int y_max;
+};
+
+#endif /* _LINUX_INPUT_ZFORCE_TS_H */
diff --git a/include/linux/power/bq24735-charger.h b/include/linux/power/bq24735-charger.h
new file mode 100644
index 000000000000..f536164a6069
--- /dev/null
+++ b/include/linux/power/bq24735-charger.h
@@ -0,0 +1,39 @@
+/*
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation; either version 2 of the License, or
+ * (at your option) any later version.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+ * GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program; if not, write to the Free Software
+ * Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA  02111-1307  USA
+ */
+
+#ifndef __CHARGER_BQ24735_H_
+#define __CHARGER_BQ24735_H_
+
+#include <linux/types.h>
+#include <linux/power_supply.h>
+
+struct bq24735_platform {
+	uint32_t charge_current;
+	uint32_t charge_voltage;
+	uint32_t input_current;
+
+	const char *name;
+
+	int status_gpio;
+	int status_gpio_active_low;
+	bool status_gpio_valid;
+
+	char **supplied_to;
+	size_t num_supplicants;
+};
+
+#endif /* __CHARGER_BQ24735_H_ */
diff --git a/include/linux/preempt_mask.h b/include/linux/preempt_mask.h
index 931bc616219f..d169820203dd 100644
--- a/include/linux/preempt_mask.h
+++ b/include/linux/preempt_mask.h
@@ -11,36 +11,23 @@
  * - bits 0-7 are the preemption count (max preemption depth: 256)
  * - bits 8-15 are the softirq count (max # of softirqs: 256)
  *
- * The hardirq count can in theory reach the same as NR_IRQS.
- * In reality, the number of nested IRQS is limited to the stack
- * size as well. For archs with over 1000 IRQS it is not practical
- * to expect that they will all nest. We give a max of 10 bits for
- * hardirq nesting. An arch may choose to give less than 10 bits.
- * m68k expects it to be 8.
+ * The hardirq count could in theory be the same as the number of
+ * interrupts in the system, but we run all interrupt handlers with
+ * interrupts disabled, so we cannot have nesting interrupts. Though
+ * there are a few palaeontologic drivers which reenable interrupts in
+ * the handler, so we need more than one bit here.
  *
- * - bits 16-25 are the hardirq count (max # of nested hardirqs: 1024)
- * - bit 26 is the NMI_MASK
- * - bit 27 is the PREEMPT_ACTIVE flag
- *
- * PREEMPT_MASK: 0x000000ff
- * SOFTIRQ_MASK: 0x0000ff00
- * HARDIRQ_MASK: 0x03ff0000
- *     NMI_MASK: 0x04000000
+ * PREEMPT_MASK:	0x000000ff
+ * SOFTIRQ_MASK:	0x0000ff00
+ * HARDIRQ_MASK:	0x000f0000
+ *     NMI_MASK:	0x00100000
+ * PREEMPT_ACTIVE:	0x00200000
  */
 #define PREEMPT_BITS	8
 #define SOFTIRQ_BITS	8
+#define HARDIRQ_BITS	4
 #define NMI_BITS	1
 
-#define MAX_HARDIRQ_BITS 10
-
-#ifndef HARDIRQ_BITS
-# define HARDIRQ_BITS	MAX_HARDIRQ_BITS
-#endif
-
-#if HARDIRQ_BITS > MAX_HARDIRQ_BITS
-#error HARDIRQ_BITS too high!
-#endif
-
 #define PREEMPT_SHIFT	0
 #define SOFTIRQ_SHIFT	(PREEMPT_SHIFT + PREEMPT_BITS)
 #define HARDIRQ_SHIFT	(SOFTIRQ_SHIFT + SOFTIRQ_BITS)
@@ -60,15 +47,9 @@
 
 #define SOFTIRQ_DISABLE_OFFSET	(2 * SOFTIRQ_OFFSET)
 
-#ifndef PREEMPT_ACTIVE
 #define PREEMPT_ACTIVE_BITS	1
 #define PREEMPT_ACTIVE_SHIFT	(NMI_SHIFT + NMI_BITS)
 #define PREEMPT_ACTIVE	(__IRQ_MASK(PREEMPT_ACTIVE_BITS) << PREEMPT_ACTIVE_SHIFT)
-#endif
-
-#if PREEMPT_ACTIVE < (1 << (NMI_SHIFT + NMI_BITS))
-#error PREEMPT_ACTIVE is too low!
-#endif
 
 #define hardirq_count()	(preempt_count() & HARDIRQ_MASK)
 #define softirq_count()	(preempt_count() & SOFTIRQ_MASK)
diff --git a/include/linux/pwm_backlight.h b/include/linux/pwm_backlight.h
index 56f4a866539a..2de2e275b2cb 100644
--- a/include/linux/pwm_backlight.h
+++ b/include/linux/pwm_backlight.h
@@ -6,6 +6,9 @@
 
 #include <linux/backlight.h>
 
+/* TODO: convert to gpiod_*() API once it has been merged */
+#define PWM_BACKLIGHT_GPIO_ACTIVE_LOW	(1 << 0)
+
 struct platform_pwm_backlight_data {
 	int pwm_id;
 	unsigned int max_brightness;
@@ -13,6 +16,8 @@ struct platform_pwm_backlight_data {
 	unsigned int lth_brightness;
 	unsigned int pwm_period_ns;
 	unsigned int *levels;
+	int enable_gpio;
+	unsigned long enable_gpio_flags;
 	int (*init)(struct device *dev);
 	int (*notify)(struct device *dev, int brightness);
 	void (*notify_after)(struct device *dev, int brightness);
diff --git a/include/linux/sched.h b/include/linux/sched.h
index f7efc8604652..7e35d4b9e14a 100644
--- a/include/linux/sched.h
+++ b/include/linux/sched.h
@@ -22,7 +22,7 @@ struct sched_param {
 #include <linux/errno.h>
 #include <linux/nodemask.h>
 #include <linux/mm_types.h>
-#include <linux/preempt.h>
+#include <linux/preempt_mask.h>
 
 #include <asm/page.h>
 #include <asm/ptrace.h>
@@ -286,6 +286,14 @@ static inline void lockup_detector_init(void)
 }
 #endif
 
+#ifdef CONFIG_DETECT_HUNG_TASK
+void reset_hung_task_detector(void);
+#else
+static inline void reset_hung_task_detector(void)
+{
+}
+#endif
+
 /* Attach to any functions which should be ignored in wchan output. */
 #define __sched		__attribute__((__section__(".sched.text")))
 
diff --git a/include/linux/sched/sysctl.h b/include/linux/sched/sysctl.h
index 10d16c4fbe89..41467f8ff8ec 100644
--- a/include/linux/sched/sysctl.h
+++ b/include/linux/sched/sysctl.h
@@ -2,8 +2,8 @@
 #define _SCHED_SYSCTL_H
 
 #ifdef CONFIG_DETECT_HUNG_TASK
+extern int	     sysctl_hung_task_check_count;
 extern unsigned int  sysctl_hung_task_panic;
-extern unsigned long sysctl_hung_task_check_count;
 extern unsigned long sysctl_hung_task_timeout_secs;
 extern unsigned long sysctl_hung_task_warnings;
 extern int proc_dohung_task_timeout_secs(struct ctl_table *table, int write,
diff --git a/include/linux/seq_file.h b/include/linux/seq_file.h
index 4e32edc8f506..52e0097f61f0 100644
--- a/include/linux/seq_file.h
+++ b/include/linux/seq_file.h
@@ -20,6 +20,7 @@ struct seq_file {
 	size_t size;
 	size_t from;
 	size_t count;
+	size_t pad_until;
 	loff_t index;
 	loff_t read_pos;
 	u64 version;
@@ -79,6 +80,20 @@ static inline void seq_commit(struct seq_file *m, int num)
 	}
 }
 
+/**
+ * seq_setwidth - set padding width
+ * @m: the seq_file handle
+ * @size: the max number of bytes to pad.
+ *
+ * Call seq_setwidth() for setting max width, then call seq_printf() etc. and
+ * finally call seq_pad() to pad the remaining bytes.
+ */
+static inline void seq_setwidth(struct seq_file *m, size_t size)
+{
+	m->pad_until = m->count + size;
+}
+void seq_pad(struct seq_file *m, char c);
+
 char *mangle_path(char *s, const char *p, const char *esc);
 int seq_open(struct file *, const struct seq_operations *);
 ssize_t seq_read(struct file *, char __user *, size_t, loff_t *);
diff --git a/include/linux/seqlock.h b/include/linux/seqlock.h
index 21a209336e79..1e8a8b6e837d 100644
--- a/include/linux/seqlock.h
+++ b/include/linux/seqlock.h
@@ -34,6 +34,7 @@
 
 #include <linux/spinlock.h>
 #include <linux/preempt.h>
+#include <linux/lockdep.h>
 #include <asm/processor.h>
 
 /*
@@ -44,10 +45,50 @@
  */
 typedef struct seqcount {
 	unsigned sequence;
+#ifdef CONFIG_DEBUG_LOCK_ALLOC
+	struct lockdep_map dep_map;
+#endif
 } seqcount_t;
 
-#define SEQCNT_ZERO { 0 }
-#define seqcount_init(x)	do { *(x) = (seqcount_t) SEQCNT_ZERO; } while (0)
+static inline void __seqcount_init(seqcount_t *s, const char *name,
+					  struct lock_class_key *key)
+{
+	/*
+	 * Make sure we are not reinitializing a held lock:
+	 */
+	lockdep_init_map(&s->dep_map, name, key, 0);
+	s->sequence = 0;
+}
+
+#ifdef CONFIG_DEBUG_LOCK_ALLOC
+# define SEQCOUNT_DEP_MAP_INIT(lockname) \
+		.dep_map = { .name = #lockname } \
+
+# define seqcount_init(s)				\
+	do {						\
+		static struct lock_class_key __key;	\
+		__seqcount_init((s), #s, &__key);	\
+	} while (0)
+
+static inline void seqcount_lockdep_reader_access(const seqcount_t *s)
+{
+	seqcount_t *l = (seqcount_t *)s;
+	unsigned long flags;
+
+	local_irq_save(flags);
+	seqcount_acquire_read(&l->dep_map, 0, 0, _RET_IP_);
+	seqcount_release(&l->dep_map, 1, _RET_IP_);
+	local_irq_restore(flags);
+}
+
+#else
+# define SEQCOUNT_DEP_MAP_INIT(lockname)
+# define seqcount_init(s) __seqcount_init(s, NULL, NULL)
+# define seqcount_lockdep_reader_access(x)
+#endif
+
+#define SEQCNT_ZERO(lockname) { .sequence = 0, SEQCOUNT_DEP_MAP_INIT(lockname)}
+
 
 /**
  * __read_seqcount_begin - begin a seq-read critical section (without barrier)
@@ -76,6 +117,22 @@ repeat:
 }
 
 /**
+ * read_seqcount_begin_no_lockdep - start seq-read critical section w/o lockdep
+ * @s: pointer to seqcount_t
+ * Returns: count to be passed to read_seqcount_retry
+ *
+ * read_seqcount_begin_no_lockdep opens a read critical section of the given
+ * seqcount, but without any lockdep checking. Validity of the critical
+ * section is tested by checking read_seqcount_retry function.
+ */
+static inline unsigned read_seqcount_begin_no_lockdep(const seqcount_t *s)
+{
+	unsigned ret = __read_seqcount_begin(s);
+	smp_rmb();
+	return ret;
+}
+
+/**
  * read_seqcount_begin - begin a seq-read critical section
  * @s: pointer to seqcount_t
  * Returns: count to be passed to read_seqcount_retry
@@ -86,9 +143,8 @@ repeat:
  */
 static inline unsigned read_seqcount_begin(const seqcount_t *s)
 {
-	unsigned ret = __read_seqcount_begin(s);
-	smp_rmb();
-	return ret;
+	seqcount_lockdep_reader_access(s);
+	return read_seqcount_begin_no_lockdep(s);
 }
 
 /**
@@ -108,6 +164,8 @@ static inline unsigned read_seqcount_begin(const seqcount_t *s)
 static inline unsigned raw_seqcount_begin(const seqcount_t *s)
 {
 	unsigned ret = ACCESS_ONCE(s->sequence);
+
+	seqcount_lockdep_reader_access(s);
 	smp_rmb();
 	return ret & ~1;
 }
@@ -152,14 +210,21 @@ static inline int read_seqcount_retry(const seqcount_t *s, unsigned start)
  * Sequence counter only version assumes that callers are using their
  * own mutexing.
  */
-static inline void write_seqcount_begin(seqcount_t *s)
+static inline void write_seqcount_begin_nested(seqcount_t *s, int subclass)
 {
 	s->sequence++;
 	smp_wmb();
+	seqcount_acquire(&s->dep_map, subclass, 0, _RET_IP_);
+}
+
+static inline void write_seqcount_begin(seqcount_t *s)
+{
+	write_seqcount_begin_nested(s, 0);
 }
 
 static inline void write_seqcount_end(seqcount_t *s)
 {
+	seqcount_release(&s->dep_map, 1, _RET_IP_);
 	smp_wmb();
 	s->sequence++;
 }
@@ -188,7 +253,7 @@ typedef struct {
  */
 #define __SEQLOCK_UNLOCKED(lockname)			\
 	{						\
-		.seqcount = SEQCNT_ZERO,		\
+		.seqcount = SEQCNT_ZERO(lockname),	\
 		.lock =	__SPIN_LOCK_UNLOCKED(lockname)	\
 	}
 
diff --git a/include/linux/skbuff.h b/include/linux/skbuff.h
index 215b5ea1cb30..bec1cc7d5e3c 100644
--- a/include/linux/skbuff.h
+++ b/include/linux/skbuff.h
@@ -2263,24 +2263,6 @@ static inline void skb_postpull_rcsum(struct sk_buff *skb,
 
 unsigned char *skb_pull_rcsum(struct sk_buff *skb, unsigned int len);
 
-/**
- *	pskb_trim_rcsum - trim received skb and update checksum
- *	@skb: buffer to trim
- *	@len: new length
- *
- *	This is exactly the same as pskb_trim except that it ensures the
- *	checksum of received packets are still valid after the operation.
- */
-
-static inline int pskb_trim_rcsum(struct sk_buff *skb, unsigned int len)
-{
-	if (likely(len >= skb->len))
-		return 0;
-	if (skb->ip_summed == CHECKSUM_COMPLETE)
-		skb->ip_summed = CHECKSUM_NONE;
-	return __pskb_trim(skb, len);
-}
-
 #define skb_queue_walk(queue, skb) \
 		for (skb = (queue)->next;					\
 		     skb != (struct sk_buff *)(queue);				\
@@ -2378,6 +2360,27 @@ __wsum __skb_checksum(const struct sk_buff *skb, int offset, int len,
 __wsum skb_checksum(const struct sk_buff *skb, int offset, int len,
 		    __wsum csum);
 
+/**
+ *	pskb_trim_rcsum - trim received skb and update checksum
+ *	@skb: buffer to trim
+ *	@len: new length
+ *
+ *	This is exactly the same as pskb_trim except that it ensures the
+ *	checksum of received packets are still valid after the operation.
+ */
+
+static inline int pskb_trim_rcsum(struct sk_buff *skb, unsigned int len)
+{
+	if (likely(len >= skb->len))
+		return 0;
+	if (skb->ip_summed == CHECKSUM_COMPLETE) {
+		__wsum adj = skb_checksum(skb, len, skb->len - len, 0);
+
+		skb->csum = csum_sub(skb->csum, adj);
+	}
+	return __pskb_trim(skb, len);
+}
+
 static inline void *skb_header_pointer(const struct sk_buff *skb, int offset,
 				       int len, void *buffer)
 {
diff --git a/include/linux/smp.h b/include/linux/smp.h
index 731f5237d5f4..5da22ee42e16 100644
--- a/include/linux/smp.h
+++ b/include/linux/smp.h
@@ -49,6 +49,9 @@ void on_each_cpu_cond(bool (*cond_func)(int cpu, void *info),
 		smp_call_func_t func, void *info, bool wait,
 		gfp_t gfp_flags);
 
+void __smp_call_function_single(int cpuid, struct call_single_data *data,
+				int wait);
+
 #ifdef CONFIG_SMP
 
 #include <linux/preempt.h>
@@ -95,9 +98,6 @@ int smp_call_function(smp_call_func_t func, void *info, int wait);
 void smp_call_function_many(const struct cpumask *mask,
 			    smp_call_func_t func, void *info, bool wait);
 
-void __smp_call_function_single(int cpuid, struct call_single_data *data,
-				int wait);
-
 int smp_call_function_any(const struct cpumask *mask,
 			  smp_call_func_t func, void *info, int wait);
 
@@ -106,14 +106,10 @@ void kick_all_cpus_sync(void);
 /*
  * Generic and arch helpers
  */
-#ifdef CONFIG_USE_GENERIC_SMP_HELPERS
 void __init call_function_init(void);
 void generic_smp_call_function_single_interrupt(void);
 #define generic_smp_call_function_interrupt \
 	generic_smp_call_function_single_interrupt
-#else
-static inline void call_function_init(void) { }
-#endif
 
 /*
  * Mark the boot cpu "online" so that it can call console drivers in
@@ -155,12 +151,6 @@ smp_call_function_any(const struct cpumask *mask, smp_call_func_t func,
 
 static inline void kick_all_cpus_sync(void) {  }
 
-static inline void __smp_call_function_single(int cpuid,
-		struct call_single_data *data, int wait)
-{
-	on_each_cpu(data->func, data->info, wait);
-}
-
 #endif /* !SMP */
 
 /*
diff --git a/include/linux/srcu.h b/include/linux/srcu.h
index c114614ed172..9b058eecd403 100644
--- a/include/linux/srcu.h
+++ b/include/linux/srcu.h
@@ -237,4 +237,18 @@ static inline void srcu_read_unlock(struct srcu_struct *sp, int idx)
 	__srcu_read_unlock(sp, idx);
 }
 
+/**
+ * smp_mb__after_srcu_read_unlock - ensure full ordering after srcu_read_unlock
+ *
+ * Converts the preceding srcu_read_unlock into a two-way memory barrier.
+ *
+ * Call this after srcu_read_unlock, to guarantee that all memory operations
+ * that occur after smp_mb__after_srcu_read_unlock will appear to happen after
+ * the preceding srcu_read_unlock.
+ */
+static inline void smp_mb__after_srcu_read_unlock(void)
+{
+	/* __srcu_read_unlock has smp_mb() internally so nothing to do here. */
+}
+
 #endif
diff --git a/include/linux/swapops.h b/include/linux/swapops.h
index 8d4fa82bfb91..c0f75261a728 100644
--- a/include/linux/swapops.h
+++ b/include/linux/swapops.h
@@ -139,7 +139,8 @@ static inline void make_migration_entry_read(swp_entry_t *entry)
 
 extern void migration_entry_wait(struct mm_struct *mm, pmd_t *pmd,
 					unsigned long address);
-extern void migration_entry_wait_huge(struct mm_struct *mm, pte_t *pte);
+extern void migration_entry_wait_huge(struct vm_area_struct *vma,
+		struct mm_struct *mm, pte_t *pte);
 #else
 
 #define make_migration_entry(page, write) swp_entry(0, 0)
@@ -151,8 +152,8 @@ static inline int is_migration_entry(swp_entry_t swp)
 static inline void make_migration_entry_read(swp_entry_t *entryp) { }
 static inline void migration_entry_wait(struct mm_struct *mm, pmd_t *pmd,
 					 unsigned long address) { }
-static inline void migration_entry_wait_huge(struct mm_struct *mm,
-					pte_t *pte) { }
+static inline void migration_entry_wait_huge(struct vm_area_struct *vma,
+		struct mm_struct *mm, pte_t *pte) { }
 static inline int is_write_migration_entry(swp_entry_t entry)
 {
 	return 0;
diff --git a/include/linux/syscalls.h b/include/linux/syscalls.h
index c27f846f6b71..94273bbe6050 100644
--- a/include/linux/syscalls.h
+++ b/include/linux/syscalls.h
@@ -120,7 +120,7 @@ extern struct trace_event_functions exit_syscall_print_funcs;
 		.class			= &event_class_syscall_enter,	\
 		.event.funcs            = &enter_syscall_print_funcs,	\
 		.data			= (void *)&__syscall_meta_##sname,\
-		.flags			= TRACE_EVENT_FL_CAP_ANY,	\
+		.flags                  = TRACE_EVENT_FL_CAP_ANY,	\
 	};								\
 	static struct ftrace_event_call __used				\
 	  __attribute__((section("_ftrace_events")))			\
@@ -134,7 +134,7 @@ extern struct trace_event_functions exit_syscall_print_funcs;
 		.class			= &event_class_syscall_exit,	\
 		.event.funcs		= &exit_syscall_print_funcs,	\
 		.data			= (void *)&__syscall_meta_##sname,\
-		.flags			= TRACE_EVENT_FL_CAP_ANY,	\
+		.flags                  = TRACE_EVENT_FL_CAP_ANY,	\
 	};								\
 	static struct ftrace_event_call __used				\
 	  __attribute__((section("_ftrace_events")))			\
diff --git a/include/linux/u64_stats_sync.h b/include/linux/u64_stats_sync.h
index 8da8c4e87da3..7bfabd20204c 100644
--- a/include/linux/u64_stats_sync.h
+++ b/include/linux/u64_stats_sync.h
@@ -67,6 +67,13 @@ struct u64_stats_sync {
 #endif
 };
 
+
+#if BITS_PER_LONG == 32 && defined(CONFIG_SMP)
+# define u64_stats_init(syncp)	seqcount_init(syncp.seq)
+#else
+# define u64_stats_init(syncp)	do { } while (0)
+#endif
+
 static inline void u64_stats_update_begin(struct u64_stats_sync *syncp)
 {
 #if BITS_PER_LONG==32 && defined(CONFIG_SMP)
diff --git a/include/linux/virtio.h b/include/linux/virtio.h
index 36d36cc89329..e4abb84199be 100644
--- a/include/linux/virtio.h
+++ b/include/linux/virtio.h
@@ -51,11 +51,11 @@ int virtqueue_add_sgs(struct virtqueue *vq,
 		      void *data,
 		      gfp_t gfp);
 
-void virtqueue_kick(struct virtqueue *vq);
+bool virtqueue_kick(struct virtqueue *vq);
 
 bool virtqueue_kick_prepare(struct virtqueue *vq);
 
-void virtqueue_notify(struct virtqueue *vq);
+bool virtqueue_notify(struct virtqueue *vq);
 
 void *virtqueue_get_buf(struct virtqueue *vq, unsigned int *len);
 
@@ -73,6 +73,8 @@ void *virtqueue_detach_unused_buf(struct virtqueue *vq);
 
 unsigned int virtqueue_get_vring_size(struct virtqueue *vq);
 
+bool virtqueue_is_broken(struct virtqueue *vq);
+
 /**
  * virtio_device - representation of a device using virtio
  * @index: unique position on the virtio bus
diff --git a/include/linux/virtio_config.h b/include/linux/virtio_config.h
index 29b9104232b4..e8f8f71e843c 100644
--- a/include/linux/virtio_config.h
+++ b/include/linux/virtio_config.h
@@ -96,33 +96,6 @@ static inline bool virtio_has_feature(const struct virtio_device *vdev,
 	return test_bit(fbit, vdev->features);
 }
 
-/**
- * virtio_config_val - look for a feature and get a virtio config entry.
- * @vdev: the virtio device
- * @fbit: the feature bit
- * @offset: the type to search for.
- * @v: a pointer to the value to fill in.
- *
- * The return value is -ENOENT if the feature doesn't exist.  Otherwise
- * the config value is copied into whatever is pointed to by v. */
-#define virtio_config_val(vdev, fbit, offset, v) \
-	virtio_config_buf((vdev), (fbit), (offset), (v), sizeof(*v))
-
-#define virtio_config_val_len(vdev, fbit, offset, v, len) \
-	virtio_config_buf((vdev), (fbit), (offset), (v), (len))
-
-static inline int virtio_config_buf(struct virtio_device *vdev,
-				    unsigned int fbit,
-				    unsigned int offset,
-				    void *buf, unsigned len)
-{
-	if (!virtio_has_feature(vdev, fbit))
-		return -ENOENT;
-
-	vdev->config->get(vdev, offset, buf, len);
-	return 0;
-}
-
 static inline
 struct virtqueue *virtio_find_single_vq(struct virtio_device *vdev,
 					vq_callback_t *c, const char *n)
@@ -162,5 +135,139 @@ int virtqueue_set_affinity(struct virtqueue *vq, int cpu)
 	return 0;
 }
 
+/* Config space accessors. */
+#define virtio_cread(vdev, structname, member, ptr)			\
+	do {								\
+		/* Must match the member's type, and be integer */	\
+		if (!typecheck(typeof((((structname*)0)->member)), *(ptr))) \
+			(*ptr) = 1;					\
+									\
+		switch (sizeof(*ptr)) {					\
+		case 1:							\
+			*(ptr) = virtio_cread8(vdev,			\
+					       offsetof(structname, member)); \
+			break;						\
+		case 2:							\
+			*(ptr) = virtio_cread16(vdev,			\
+						offsetof(structname, member)); \
+			break;						\
+		case 4:							\
+			*(ptr) = virtio_cread32(vdev,			\
+						offsetof(structname, member)); \
+			break;						\
+		case 8:							\
+			*(ptr) = virtio_cread64(vdev,			\
+						offsetof(structname, member)); \
+			break;						\
+		default:						\
+			BUG();						\
+		}							\
+	} while(0)
+
+/* Config space accessors. */
+#define virtio_cwrite(vdev, structname, member, ptr)			\
+	do {								\
+		/* Must match the member's type, and be integer */	\
+		if (!typecheck(typeof((((structname*)0)->member)), *(ptr))) \
+			BUG_ON((*ptr) == 1);				\
+									\
+		switch (sizeof(*ptr)) {					\
+		case 1:							\
+			virtio_cwrite8(vdev,				\
+				       offsetof(structname, member),	\
+				       *(ptr));				\
+			break;						\
+		case 2:							\
+			virtio_cwrite16(vdev,				\
+					offsetof(structname, member),	\
+					*(ptr));			\
+			break;						\
+		case 4:							\
+			virtio_cwrite32(vdev,				\
+					offsetof(structname, member),	\
+					*(ptr));			\
+			break;						\
+		case 8:							\
+			virtio_cwrite64(vdev,				\
+					offsetof(structname, member),	\
+					*(ptr));			\
+			break;						\
+		default:						\
+			BUG();						\
+		}							\
+	} while(0)
+
+static inline u8 virtio_cread8(struct virtio_device *vdev, unsigned int offset)
+{
+	u8 ret;
+	vdev->config->get(vdev, offset, &ret, sizeof(ret));
+	return ret;
+}
+
+static inline void virtio_cread_bytes(struct virtio_device *vdev,
+				      unsigned int offset,
+				      void *buf, size_t len)
+{
+	vdev->config->get(vdev, offset, buf, len);
+}
+
+static inline void virtio_cwrite8(struct virtio_device *vdev,
+				  unsigned int offset, u8 val)
+{
+	vdev->config->set(vdev, offset, &val, sizeof(val));
+}
+
+static inline u16 virtio_cread16(struct virtio_device *vdev,
+				 unsigned int offset)
+{
+	u16 ret;
+	vdev->config->get(vdev, offset, &ret, sizeof(ret));
+	return ret;
+}
+
+static inline void virtio_cwrite16(struct virtio_device *vdev,
+				   unsigned int offset, u16 val)
+{
+	vdev->config->set(vdev, offset, &val, sizeof(val));
+}
+
+static inline u32 virtio_cread32(struct virtio_device *vdev,
+				 unsigned int offset)
+{
+	u32 ret;
+	vdev->config->get(vdev, offset, &ret, sizeof(ret));
+	return ret;
+}
+
+static inline void virtio_cwrite32(struct virtio_device *vdev,
+				   unsigned int offset, u32 val)
+{
+	vdev->config->set(vdev, offset, &val, sizeof(val));
+}
+
+static inline u64 virtio_cread64(struct virtio_device *vdev,
+				 unsigned int offset)
+{
+	u64 ret;
+	vdev->config->get(vdev, offset, &ret, sizeof(ret));
+	return ret;
+}
+
+static inline void virtio_cwrite64(struct virtio_device *vdev,
+				   unsigned int offset, u64 val)
+{
+	vdev->config->set(vdev, offset, &val, sizeof(val));
+}
+
+/* Conditional config space accessors. */
+#define virtio_cread_feature(vdev, fbit, structname, member, ptr)	\
+	({								\
+		int _r = 0;						\
+		if (!virtio_has_feature(vdev, fbit))			\
+			_r = -ENOENT;					\
+		else							\
+			virtio_cread((vdev), structname, member, ptr);	\
+		_r;							\
+	})
 
 #endif /* _LINUX_VIRTIO_CONFIG_H */
diff --git a/include/linux/virtio_ring.h b/include/linux/virtio_ring.h
index b300787af8e0..67e06fe18c03 100644
--- a/include/linux/virtio_ring.h
+++ b/include/linux/virtio_ring.h
@@ -71,7 +71,7 @@ struct virtqueue *vring_new_virtqueue(unsigned int index,
 				      struct virtio_device *vdev,
 				      bool weak_barriers,
 				      void *pages,
-				      void (*notify)(struct virtqueue *vq),
+				      bool (*notify)(struct virtqueue *vq),
 				      void (*callback)(struct virtqueue *vq),
 				      const char *name);
 void vring_del_virtqueue(struct virtqueue *vq);
diff --git a/include/linux/wait.h b/include/linux/wait.h
index 61939ba30aa0..eaa00b10abaa 100644
--- a/include/linux/wait.h
+++ b/include/linux/wait.h
@@ -278,6 +278,31 @@ do {									\
 	__ret;								\
 })
 
+#define __wait_event_cmd(wq, condition, cmd1, cmd2)			\
+	(void)___wait_event(wq, condition, TASK_UNINTERRUPTIBLE, 0, 0,	\
+			    cmd1; schedule(); cmd2)
+
+/**
+ * wait_event_cmd - sleep until a condition gets true
+ * @wq: the waitqueue to wait on
+ * @condition: a C expression for the event to wait for
+ * cmd1: the command will be executed before sleep
+ * cmd2: the command will be executed after sleep
+ *
+ * The process is put to sleep (TASK_UNINTERRUPTIBLE) until the
+ * @condition evaluates to true. The @condition is checked each time
+ * the waitqueue @wq is woken up.
+ *
+ * wake_up() has to be called after changing any variable that could
+ * change the result of the wait condition.
+ */
+#define wait_event_cmd(wq, condition, cmd1, cmd2)			\
+do {									\
+	if (condition)							\
+		break;							\
+	__wait_event_cmd(wq, condition, cmd1, cmd2);			\
+} while (0)
+
 #define __wait_event_interruptible(wq, condition)			\
 	___wait_event(wq, condition, TASK_INTERRUPTIBLE, 0, 0,		\
 		      schedule())
diff --git a/include/media/lm3560.h b/include/media/lm3560.h
new file mode 100644
index 000000000000..46670706d6f8
--- /dev/null
+++ b/include/media/lm3560.h
@@ -0,0 +1,97 @@
+/*
+ * include/media/lm3560.h
+ *
+ * Copyright (C) 2013 Texas Instruments
+ *
+ * Contact: Daniel Jeong <gshark.jeong@gmail.com>
+ *			Ldd-Mlp <ldd-mlp@list.ti.com>
+ *
+ * This program is free software; you can redistribute it and/or
+ * modify it under the terms of the GNU General Public License
+ * version 2 as published by the Free Software Foundation.
+ *
+ * This program is distributed in the hope that it will be useful, but
+ * WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the GNU
+ * General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program; if not, write to the Free Software
+ * Foundation, Inc., 51 Franklin St, Fifth Floor, Boston, MA
+ * 02110-1301 USA
+ *
+ */
+
+#ifndef __LM3560_H__
+#define __LM3560_H__
+
+#include <media/v4l2-subdev.h>
+
+#define LM3560_NAME	"lm3560"
+#define LM3560_I2C_ADDR	(0x53)
+
+/*  FLASH Brightness
+ *	min 62500uA, step 62500uA, max 1000000uA
+ */
+#define LM3560_FLASH_BRT_MIN 62500
+#define LM3560_FLASH_BRT_STEP 62500
+#define LM3560_FLASH_BRT_MAX 1000000
+#define LM3560_FLASH_BRT_uA_TO_REG(a)	\
+	((a) < LM3560_FLASH_BRT_MIN ? 0 :	\
+	 (((a) - LM3560_FLASH_BRT_MIN) / LM3560_FLASH_BRT_STEP))
+#define LM3560_FLASH_BRT_REG_TO_uA(a)		\
+	((a) * LM3560_FLASH_BRT_STEP + LM3560_FLASH_BRT_MIN)
+
+/*  FLASH TIMEOUT DURATION
+ *	min 32ms, step 32ms, max 1024ms
+ */
+#define LM3560_FLASH_TOUT_MIN 32
+#define LM3560_FLASH_TOUT_STEP 32
+#define LM3560_FLASH_TOUT_MAX 1024
+#define LM3560_FLASH_TOUT_ms_TO_REG(a)	\
+	((a) < LM3560_FLASH_TOUT_MIN ? 0 :	\
+	 (((a) - LM3560_FLASH_TOUT_MIN) / LM3560_FLASH_TOUT_STEP))
+#define LM3560_FLASH_TOUT_REG_TO_ms(a)		\
+	((a) * LM3560_FLASH_TOUT_STEP + LM3560_FLASH_TOUT_MIN)
+
+/*  TORCH BRT
+ *	min 31250uA, step 31250uA, max 250000uA
+ */
+#define LM3560_TORCH_BRT_MIN 31250
+#define LM3560_TORCH_BRT_STEP 31250
+#define LM3560_TORCH_BRT_MAX 250000
+#define LM3560_TORCH_BRT_uA_TO_REG(a)	\
+	((a) < LM3560_TORCH_BRT_MIN ? 0 :	\
+	 (((a) - LM3560_TORCH_BRT_MIN) / LM3560_TORCH_BRT_STEP))
+#define LM3560_TORCH_BRT_REG_TO_uA(a)		\
+	((a) * LM3560_TORCH_BRT_STEP + LM3560_TORCH_BRT_MIN)
+
+enum lm3560_led_id {
+	LM3560_LED0 = 0,
+	LM3560_LED1,
+	LM3560_LED_MAX
+};
+
+enum lm3560_peak_current {
+	LM3560_PEAK_1600mA = 0x00,
+	LM3560_PEAK_2300mA = 0x20,
+	LM3560_PEAK_3000mA = 0x40,
+	LM3560_PEAK_3600mA = 0x60
+};
+
+/* struct lm3560_platform_data
+ *
+ * @peak :  peak current
+ * @max_flash_timeout: flash timeout
+ * @max_flash_brt: flash mode led brightness
+ * @max_torch_brt: torch mode led brightness
+ */
+struct lm3560_platform_data {
+	enum lm3560_peak_current peak;
+
+	u32 max_flash_timeout;
+	u32 max_flash_brt[LM3560_LED_MAX];
+	u32 max_torch_brt[LM3560_LED_MAX];
+};
+
+#endif /* __LM3560_H__ */
diff --git a/include/media/soc_camera.h b/include/media/soc_camera.h
index 34d2414f2b8c..865246b00127 100644
--- a/include/media/soc_camera.h
+++ b/include/media/soc_camera.h
@@ -146,9 +146,14 @@ struct soc_camera_subdev_desc {
 	/* sensor driver private platform data */
 	void *drv_priv;
 
-	/* Optional regulators that have to be managed on power on/off events */
-	struct regulator_bulk_data *regulators;
-	int num_regulators;
+	/*
+	 * Set unbalanced_power to true to deal with legacy drivers, failing to
+	 * balance their calls to subdevice's .s_power() method. clock_state is
+	 * then used internally by helper functions, it shouldn't be touched by
+	 * drivers or the platform code.
+	 */
+	bool unbalanced_power;
+	unsigned long clock_state;
 
 	/* Optional callbacks to power on or off and reset the sensor */
 	int (*power)(struct device *, int);
@@ -162,6 +167,9 @@ struct soc_camera_subdev_desc {
 	int (*set_bus_param)(struct soc_camera_subdev_desc *, unsigned long flags);
 	unsigned long (*query_bus_param)(struct soc_camera_subdev_desc *);
 	void (*free_bus)(struct soc_camera_subdev_desc *);
+
+	/* Optional regulators that have to be managed on power on/off events */
+	struct v4l2_subdev_platform_data sd_pdata;
 };
 
 struct soc_camera_host_desc {
@@ -202,9 +210,10 @@ struct soc_camera_link {
 
 	void *priv;
 
-	/* Optional regulators that have to be managed on power on/off events */
-	struct regulator_bulk_data *regulators;
-	int num_regulators;
+	/* Set by platforms to handle misbehaving drivers */
+	bool unbalanced_power;
+	/* Used by soc-camera helper functions */
+	unsigned long clock_state;
 
 	/* Optional callbacks to power on or off and reset the sensor */
 	int (*power)(struct device *, int);
@@ -218,6 +227,12 @@ struct soc_camera_link {
 	unsigned long (*query_bus_param)(struct soc_camera_link *);
 	void (*free_bus)(struct soc_camera_link *);
 
+	/* Optional regulators that have to be managed on power on/off events */
+	struct regulator_bulk_data *regulators;
+	int num_regulators;
+
+	void *host_priv;
+
 	/*
 	 * Host part - keep at bottom and compatible to
 	 * struct soc_camera_host_desc
diff --git a/include/media/v4l2-clk.h b/include/media/v4l2-clk.h
index 0503a90b48bb..0b36cc138304 100644
--- a/include/media/v4l2-clk.h
+++ b/include/media/v4l2-clk.h
@@ -15,6 +15,7 @@
 #define MEDIA_V4L2_CLK_H
 
 #include <linux/atomic.h>
+#include <linux/export.h>
 #include <linux/list.h>
 #include <linux/mutex.h>
 
@@ -51,4 +52,20 @@ void v4l2_clk_disable(struct v4l2_clk *clk);
 unsigned long v4l2_clk_get_rate(struct v4l2_clk *clk);
 int v4l2_clk_set_rate(struct v4l2_clk *clk, unsigned long rate);
 
+struct module;
+
+struct v4l2_clk *__v4l2_clk_register_fixed(const char *dev_id,
+		const char *id, unsigned long rate, struct module *owner);
+void v4l2_clk_unregister_fixed(struct v4l2_clk *clk);
+
+static inline struct v4l2_clk *v4l2_clk_register_fixed(const char *dev_id,
+							const char *id,
+							unsigned long rate)
+{
+	return __v4l2_clk_register_fixed(dev_id, id, rate, THIS_MODULE);
+}
+
+#define v4l2_clk_name_i2c(name, size, adap, client) snprintf(name, size, \
+			  "%d-%04x", adap, client)
+
 #endif
diff --git a/include/media/v4l2-common.h b/include/media/v4l2-common.h
index 16550c439008..48f974866f13 100644
--- a/include/media/v4l2-common.h
+++ b/include/media/v4l2-common.h
@@ -35,7 +35,7 @@
 	printk(level "%s %d-%04x: " fmt, name, i2c_adapter_id(adapter), addr , ## arg)
 
 #define v4l_client_printk(level, client, fmt, arg...)			    \
-	v4l_printk(level, (client)->driver->driver.name, (client)->adapter, \
+	v4l_printk(level, (client)->dev.driver->name, (client)->adapter, \
 		   (client)->addr, fmt , ## arg)
 
 #define v4l_err(client, fmt, arg...) \
@@ -86,7 +86,7 @@ int v4l2_ctrl_check(struct v4l2_ext_control *ctrl, struct v4l2_queryctrl *qctrl,
 		const char * const *menu_items);
 const char *v4l2_ctrl_get_name(u32 id);
 const char * const *v4l2_ctrl_get_menu(u32 id);
-const s64 const *v4l2_ctrl_get_int_menu(u32 id, u32 *len);
+const s64 *v4l2_ctrl_get_int_menu(u32 id, u32 *len);
 int v4l2_ctrl_query_fill(struct v4l2_queryctrl *qctrl, s32 min, s32 max, s32 step, s32 def);
 int v4l2_ctrl_query_menu(struct v4l2_querymenu *qmenu,
 		struct v4l2_queryctrl *qctrl, const char * const *menu_items);
diff --git a/include/media/v4l2-ctrls.h b/include/media/v4l2-ctrls.h
index 47ada23345a1..16f7f2606516 100644
--- a/include/media/v4l2-ctrls.h
+++ b/include/media/v4l2-ctrls.h
@@ -571,7 +571,7 @@ static inline void v4l2_ctrl_lock(struct v4l2_ctrl *ctrl)
 	mutex_lock(ctrl->handler->lock);
 }
 
-/** v4l2_ctrl_lock() - Helper function to unlock the handler
+/** v4l2_ctrl_unlock() - Helper function to unlock the handler
   * associated with the control.
   * @ctrl:	The control to unlock.
   */
diff --git a/include/media/v4l2-fh.h b/include/media/v4l2-fh.h
index a62ee18cb7b7..528cdaf622e1 100644
--- a/include/media/v4l2-fh.h
+++ b/include/media/v4l2-fh.h
@@ -26,7 +26,9 @@
 #ifndef V4L2_FH_H
 #define V4L2_FH_H
 
+#include <linux/fs.h>
 #include <linux/list.h>
+#include <linux/videodev2.h>
 
 struct video_device;
 struct v4l2_ctrl_handler;
diff --git a/include/media/v4l2-subdev.h b/include/media/v4l2-subdev.h
index bfda0fe9aeb0..d67210a37ef3 100644
--- a/include/media/v4l2-subdev.h
+++ b/include/media/v4l2-subdev.h
@@ -559,6 +559,17 @@ struct v4l2_subdev_internal_ops {
 /* Set this flag if this subdev generates events. */
 #define V4L2_SUBDEV_FL_HAS_EVENTS		(1U << 3)
 
+struct regulator_bulk_data;
+
+struct v4l2_subdev_platform_data {
+	/* Optional regulators uset to power on/off the subdevice */
+	struct regulator_bulk_data *regulators;
+	int num_regulators;
+
+	/* Per-subdevice data, specific for a certain video host device */
+	void *host_priv;
+};
+
 /* Each instance of a subdev driver should create this struct, either
    stand-alone or embedded in a larger struct.
  */
@@ -592,6 +603,8 @@ struct v4l2_subdev {
 	struct v4l2_async_subdev *asd;
 	/* Pointer to the managing notifier. */
 	struct v4l2_async_notifier *notifier;
+	/* common part of subdevice platform data */
+	struct v4l2_subdev_platform_data *pdata;
 };
 
 #define media_entity_to_v4l2_subdev(ent) \
@@ -622,13 +635,13 @@ struct v4l2_subdev_fh {
 	v4l2_subdev_get_try_##fun_name(struct v4l2_subdev_fh *fh,	\
 				       unsigned int pad)		\
 	{								\
-		BUG_ON(unlikely(pad >= vdev_to_v4l2_subdev(		\
-					fh->vfh.vdev)->entity.num_pads)); \
+		BUG_ON(pad >= vdev_to_v4l2_subdev(			\
+					fh->vfh.vdev)->entity.num_pads); \
 		return &fh->pad[pad].field_name;			\
 	}
 
 __V4L2_SUBDEV_MK_GET_TRY(v4l2_mbus_framefmt, format, try_fmt)
-__V4L2_SUBDEV_MK_GET_TRY(v4l2_rect, crop, try_compose)
+__V4L2_SUBDEV_MK_GET_TRY(v4l2_rect, crop, try_crop)
 __V4L2_SUBDEV_MK_GET_TRY(v4l2_rect, compose, try_compose)
 #endif
 
diff --git a/include/media/videobuf2-core.h b/include/media/videobuf2-core.h
index 6781258d0b67..bd8218b15009 100644
--- a/include/media/videobuf2-core.h
+++ b/include/media/videobuf2-core.h
@@ -391,7 +391,7 @@ unsigned long vb2_get_unmapped_area(struct vb2_queue *q,
 unsigned int vb2_poll(struct vb2_queue *q, struct file *file, poll_table *wait);
 size_t vb2_read(struct vb2_queue *q, char __user *data, size_t count,
 		loff_t *ppos, int nonblock);
-size_t vb2_write(struct vb2_queue *q, char __user *data, size_t count,
+size_t vb2_write(struct vb2_queue *q, const char __user *data, size_t count,
 		loff_t *ppos, int nonblock);
 
 /**
@@ -491,7 +491,7 @@ int vb2_ioctl_expbuf(struct file *file, void *priv,
 
 int vb2_fop_mmap(struct file *file, struct vm_area_struct *vma);
 int vb2_fop_release(struct file *file);
-ssize_t vb2_fop_write(struct file *file, char __user *buf,
+ssize_t vb2_fop_write(struct file *file, const char __user *buf,
 		size_t count, loff_t *ppos);
 ssize_t vb2_fop_read(struct file *file, char __user *buf,
 		size_t count, loff_t *ppos);
diff --git a/include/media/videobuf2-dma-sg.h b/include/media/videobuf2-dma-sg.h
index 0038526b8ef7..7b89852779af 100644
--- a/include/media/videobuf2-dma-sg.h
+++ b/include/media/videobuf2-dma-sg.h
@@ -15,16 +15,10 @@
 
 #include <media/videobuf2-core.h>
 
-struct vb2_dma_sg_desc {
-	unsigned long		size;
-	unsigned int		num_pages;
-	struct scatterlist	*sglist;
-};
-
-static inline struct vb2_dma_sg_desc *vb2_dma_sg_plane_desc(
+static inline struct sg_table *vb2_dma_sg_plane_desc(
 		struct vb2_buffer *vb, unsigned int plane_no)
 {
-	return (struct vb2_dma_sg_desc *)vb2_plane_cookie(vb, plane_no);
+	return (struct sg_table *)vb2_plane_cookie(vb, plane_no);
 }
 
 extern const struct vb2_mem_ops vb2_dma_sg_memops;
diff --git a/include/net/bluetooth/l2cap.h b/include/net/bluetooth/l2cap.h
index 51329905bfaa..c853b16de4ef 100644
--- a/include/net/bluetooth/l2cap.h
+++ b/include/net/bluetooth/l2cap.h
@@ -240,7 +240,7 @@ struct l2cap_conn_rsp {
 #define L2CAP_PSM_RFCOMM	0x0003
 #define L2CAP_PSM_3DSP		0x0021
 
-/* channel indentifier */
+/* channel identifier */
 #define L2CAP_CID_SIGNALING	0x0001
 #define L2CAP_CID_CONN_LESS	0x0002
 #define L2CAP_CID_A2MP		0x0003
diff --git a/include/net/genetlink.h b/include/net/genetlink.h
index 9b787b62cf16..ace4abf118d7 100644
--- a/include/net/genetlink.h
+++ b/include/net/genetlink.h
@@ -10,16 +10,9 @@
 /**
  * struct genl_multicast_group - generic netlink multicast group
  * @name: name of the multicast group, names are per-family
- * @id: multicast group ID, assigned by the core, to use with
- *      genlmsg_multicast().
- * @list: list entry for linking
- * @family: pointer to family, need not be set before registering
  */
 struct genl_multicast_group {
-	struct genl_family	*family;	/* private */
-	struct list_head	list;		/* private */
 	char			name[GENL_NAMSIZ];
-	u32			id;
 };
 
 struct genl_ops;
@@ -39,9 +32,12 @@ struct genl_info;
  * @post_doit: called after an operation's doit callback, it may
  *	undo operations done by pre_doit, for example release locks
  * @attrbuf: buffer to store parsed attributes
- * @ops_list: list of all assigned operations
  * @family_list: family list
- * @mcast_groups: multicast groups list
+ * @mcgrps: multicast groups used by this family (private)
+ * @n_mcgrps: number of multicast groups (private)
+ * @mcgrp_offset: starting number of multicast group IDs in this family
+ * @ops: the operations supported by this family (private)
+ * @n_ops: number of operations supported by this family (private)
  */
 struct genl_family {
 	unsigned int		id;
@@ -51,16 +47,19 @@ struct genl_family {
 	unsigned int		maxattr;
 	bool			netnsok;
 	bool			parallel_ops;
-	int			(*pre_doit)(struct genl_ops *ops,
+	int			(*pre_doit)(const struct genl_ops *ops,
 					    struct sk_buff *skb,
 					    struct genl_info *info);
-	void			(*post_doit)(struct genl_ops *ops,
+	void			(*post_doit)(const struct genl_ops *ops,
 					     struct sk_buff *skb,
 					     struct genl_info *info);
 	struct nlattr **	attrbuf;	/* private */
-	struct list_head	ops_list;	/* private */
+	const struct genl_ops *	ops;		/* private */
+	const struct genl_multicast_group *mcgrps; /* private */
+	unsigned int		n_ops;		/* private */
+	unsigned int		n_mcgrps;	/* private */
+	unsigned int		mcgrp_offset;	/* private */
 	struct list_head	family_list;	/* private */
-	struct list_head	mcast_groups;	/* private */
 	struct module		*module;
 };
 
@@ -110,16 +109,15 @@ static inline void genl_info_net_set(struct genl_info *info, struct net *net)
  * @ops_list: operations list
  */
 struct genl_ops {
-	u8			cmd;
-	u8			internal_flags;
-	unsigned int		flags;
 	const struct nla_policy	*policy;
 	int		       (*doit)(struct sk_buff *skb,
 				       struct genl_info *info);
 	int		       (*dumpit)(struct sk_buff *skb,
 					 struct netlink_callback *cb);
 	int		       (*done)(struct netlink_callback *cb);
-	struct list_head	ops_list;
+	u8			cmd;
+	u8			internal_flags;
+	u8			flags;
 };
 
 int __genl_register_family(struct genl_family *family);
@@ -130,24 +128,53 @@ static inline int genl_register_family(struct genl_family *family)
 	return __genl_register_family(family);
 }
 
-int __genl_register_family_with_ops(struct genl_family *family,
-				    struct genl_ops *ops, size_t n_ops);
-
-static inline int genl_register_family_with_ops(struct genl_family *family,
-	struct genl_ops *ops, size_t n_ops)
+/**
+ * genl_register_family_with_ops - register a generic netlink family with ops
+ * @family: generic netlink family
+ * @ops: operations to be registered
+ * @n_ops: number of elements to register
+ *
+ * Registers the specified family and operations from the specified table.
+ * Only one family may be registered with the same family name or identifier.
+ *
+ * The family id may equal GENL_ID_GENERATE causing an unique id to
+ * be automatically generated and assigned.
+ *
+ * Either a doit or dumpit callback must be specified for every registered
+ * operation or the function will fail. Only one operation structure per
+ * command identifier may be registered.
+ *
+ * See include/net/genetlink.h for more documenation on the operations
+ * structure.
+ *
+ * Return 0 on success or a negative error code.
+ */
+static inline int
+_genl_register_family_with_ops_grps(struct genl_family *family,
+				    const struct genl_ops *ops, size_t n_ops,
+				    const struct genl_multicast_group *mcgrps,
+				    size_t n_mcgrps)
 {
 	family->module = THIS_MODULE;
-	return __genl_register_family_with_ops(family, ops, n_ops);
+	family->ops = ops;
+	family->n_ops = n_ops;
+	family->mcgrps = mcgrps;
+	family->n_mcgrps = n_mcgrps;
+	return __genl_register_family(family);
 }
 
+#define genl_register_family_with_ops(family, ops)			\
+	_genl_register_family_with_ops_grps((family),			\
+					    (ops), ARRAY_SIZE(ops),	\
+					    NULL, 0)
+#define genl_register_family_with_ops_groups(family, ops, grps)	\
+	_genl_register_family_with_ops_grps((family),			\
+					    (ops), ARRAY_SIZE(ops),	\
+					    (grps), ARRAY_SIZE(grps))
+
 int genl_unregister_family(struct genl_family *family);
-int genl_register_ops(struct genl_family *, struct genl_ops *ops);
-int genl_unregister_ops(struct genl_family *, struct genl_ops *ops);
-int genl_register_mc_group(struct genl_family *family,
-			   struct genl_multicast_group *grp);
-void genl_unregister_mc_group(struct genl_family *family,
-			      struct genl_multicast_group *grp);
-void genl_notify(struct sk_buff *skb, struct net *net, u32 portid,
+void genl_notify(struct genl_family *family,
+		 struct sk_buff *skb, struct net *net, u32 portid,
 		 u32 group, struct nlmsghdr *nlh, gfp_t flags);
 
 void *genlmsg_put(struct sk_buff *skb, u32 portid, u32 seq,
@@ -227,41 +254,54 @@ static inline void genlmsg_cancel(struct sk_buff *skb, void *hdr)
 
 /**
  * genlmsg_multicast_netns - multicast a netlink message to a specific netns
+ * @family: the generic netlink family
  * @net: the net namespace
  * @skb: netlink message as socket buffer
  * @portid: own netlink portid to avoid sending to yourself
- * @group: multicast group id
+ * @group: offset of multicast group in groups array
  * @flags: allocation flags
  */
-static inline int genlmsg_multicast_netns(struct net *net, struct sk_buff *skb,
+static inline int genlmsg_multicast_netns(struct genl_family *family,
+					  struct net *net, struct sk_buff *skb,
 					  u32 portid, unsigned int group, gfp_t flags)
 {
+	if (group >= family->n_mcgrps)
+		return -EINVAL;
+	group = family->mcgrp_offset + group;
 	return nlmsg_multicast(net->genl_sock, skb, portid, group, flags);
 }
 
 /**
  * genlmsg_multicast - multicast a netlink message to the default netns
+ * @family: the generic netlink family
  * @skb: netlink message as socket buffer
  * @portid: own netlink portid to avoid sending to yourself
- * @group: multicast group id
+ * @group: offset of multicast group in groups array
  * @flags: allocation flags
  */
-static inline int genlmsg_multicast(struct sk_buff *skb, u32 portid,
+static inline int genlmsg_multicast(struct genl_family *family,
+				    struct sk_buff *skb, u32 portid,
 				    unsigned int group, gfp_t flags)
 {
-	return genlmsg_multicast_netns(&init_net, skb, portid, group, flags);
+	if (group >= family->n_mcgrps)
+		return -EINVAL;
+	group = family->mcgrp_offset + group;
+	return genlmsg_multicast_netns(family, &init_net, skb,
+				       portid, group, flags);
 }
 
 /**
  * genlmsg_multicast_allns - multicast a netlink message to all net namespaces
+ * @family: the generic netlink family
  * @skb: netlink message as socket buffer
  * @portid: own netlink portid to avoid sending to yourself
- * @group: multicast group id
+ * @group: offset of multicast group in groups array
  * @flags: allocation flags
  *
  * This function must hold the RTNL or rcu_read_lock().
  */
-int genlmsg_multicast_allns(struct sk_buff *skb, u32 portid,
+int genlmsg_multicast_allns(struct genl_family *family,
+			    struct sk_buff *skb, u32 portid,
 			    unsigned int group, gfp_t flags);
 
 /**
@@ -332,5 +372,22 @@ static inline struct sk_buff *genlmsg_new(size_t payload, gfp_t flags)
 	return nlmsg_new(genlmsg_total_size(payload), flags);
 }
 
+/**
+ * genl_set_err - report error to genetlink broadcast listeners
+ * @family: the generic netlink family
+ * @net: the network namespace to report the error to
+ * @portid: the PORTID of a process that we want to skip (if any)
+ * @group: the broadcast group that will notice the error
+ * 	(this is the offset of the multicast group in the groups array)
+ * @code: error code, must be negative (as usual in kernelspace)
+ *
+ * This function returns the number of broadcast listeners that have set the
+ * NETLINK_RECV_NO_ENOBUFS socket option.
+ */
+static inline int genl_set_err(struct genl_family *family, struct net *net,
+			       u32 portid, u32 group, int code)
+{
+	return netlink_set_err(net->genl_sock, portid, group, code);
+}
 
 #endif	/* __NET_GENERIC_NETLINK_H */
diff --git a/include/rdma/ib_verbs.h b/include/rdma/ib_verbs.h
index e393171e2fac..979874c627ee 100644
--- a/include/rdma/ib_verbs.h
+++ b/include/rdma/ib_verbs.h
@@ -67,12 +67,14 @@ enum rdma_node_type {
 	RDMA_NODE_IB_CA 	= 1,
 	RDMA_NODE_IB_SWITCH,
 	RDMA_NODE_IB_ROUTER,
-	RDMA_NODE_RNIC
+	RDMA_NODE_RNIC,
+	RDMA_NODE_USNIC,
 };
 
 enum rdma_transport_type {
 	RDMA_TRANSPORT_IB,
-	RDMA_TRANSPORT_IWARP
+	RDMA_TRANSPORT_IWARP,
+	RDMA_TRANSPORT_USNIC
 };
 
 enum rdma_transport_type
@@ -1436,6 +1438,7 @@ struct ib_device {
 
 	int			     uverbs_abi_ver;
 	u64			     uverbs_cmd_mask;
+	u64			     uverbs_ex_cmd_mask;
 
 	char			     node_desc[64];
 	__be64			     node_guid;
@@ -2384,4 +2387,17 @@ struct ib_flow *ib_create_flow(struct ib_qp *qp,
 			       struct ib_flow_attr *flow_attr, int domain);
 int ib_destroy_flow(struct ib_flow *flow_id);
 
+static inline int ib_check_mr_access(int flags)
+{
+	/*
+	 * Local write permission is required if remote write or
+	 * remote atomic permission is also requested.
+	 */
+	if (flags & (IB_ACCESS_REMOTE_ATOMIC | IB_ACCESS_REMOTE_WRITE) &&
+	    !(flags & IB_ACCESS_LOCAL_WRITE))
+		return -EINVAL;
+
+	return 0;
+}
+
 #endif /* IB_VERBS_H */
diff --git a/include/scsi/scsi_transport_srp.h b/include/scsi/scsi_transport_srp.h
index ff0f04ac91aa..4ebf6913b7b2 100644
--- a/include/scsi/scsi_transport_srp.h
+++ b/include/scsi/scsi_transport_srp.h
@@ -13,6 +13,27 @@ struct srp_rport_identifiers {
 	u8 roles;
 };
 
+/**
+ * enum srp_rport_state - SRP transport layer state
+ * @SRP_RPORT_RUNNING:   Transport layer operational.
+ * @SRP_RPORT_BLOCKED:   Transport layer not operational; fast I/O fail timer
+ *                       is running and I/O has been blocked.
+ * @SRP_RPORT_FAIL_FAST: Fast I/O fail timer has expired; fail I/O fast.
+ * @SRP_RPORT_LOST:      Device loss timer has expired; port is being removed.
+ */
+enum srp_rport_state {
+	SRP_RPORT_RUNNING,
+	SRP_RPORT_BLOCKED,
+	SRP_RPORT_FAIL_FAST,
+	SRP_RPORT_LOST,
+};
+
+/**
+ * struct srp_rport
+ * @lld_data: LLD private data.
+ * @mutex:    Protects against concurrent rport reconnect / fast_io_fail /
+ *   dev_loss_tmo activity.
+ */
 struct srp_rport {
 	/* for initiator and target drivers */
 
@@ -23,11 +44,43 @@ struct srp_rport {
 
 	/* for initiator drivers */
 
-	void *lld_data;	/* LLD private data */
+	void			*lld_data;
+
+	struct mutex		mutex;
+	enum srp_rport_state	state;
+	bool			deleted;
+	int			reconnect_delay;
+	int			failed_reconnects;
+	struct delayed_work	reconnect_work;
+	int			fast_io_fail_tmo;
+	int			dev_loss_tmo;
+	struct delayed_work	fast_io_fail_work;
+	struct delayed_work	dev_loss_work;
 };
 
+/**
+ * struct srp_function_template
+ * @has_rport_state: Whether or not to create the state, fast_io_fail_tmo and
+ *     dev_loss_tmo sysfs attribute for an rport.
+ * @reset_timer_if_blocked: Whether or srp_timed_out() should reset the command
+ *     timer if the device on which it has been queued is blocked.
+ * @reconnect_delay: If not NULL, points to the default reconnect_delay value.
+ * @fast_io_fail_tmo: If not NULL, points to the default fast_io_fail_tmo value.
+ * @dev_loss_tmo: If not NULL, points to the default dev_loss_tmo value.
+ * @reconnect: Callback function for reconnecting to the target. See also
+ *     srp_reconnect_rport().
+ * @terminate_rport_io: Callback function for terminating all outstanding I/O
+ *     requests for an rport.
+ */
 struct srp_function_template {
 	/* for initiator drivers */
+	bool has_rport_state;
+	bool reset_timer_if_blocked;
+	int *reconnect_delay;
+	int *fast_io_fail_tmo;
+	int *dev_loss_tmo;
+	int (*reconnect)(struct srp_rport *rport);
+	void (*terminate_rport_io)(struct srp_rport *rport);
 	void (*rport_delete)(struct srp_rport *rport);
 	/* for target drivers */
 	int (* tsk_mgmt_response)(struct Scsi_Host *, u64, u64, int);
@@ -38,10 +91,36 @@ extern struct scsi_transport_template *
 srp_attach_transport(struct srp_function_template *);
 extern void srp_release_transport(struct scsi_transport_template *);
 
+extern void srp_rport_get(struct srp_rport *rport);
+extern void srp_rport_put(struct srp_rport *rport);
 extern struct srp_rport *srp_rport_add(struct Scsi_Host *,
 				       struct srp_rport_identifiers *);
 extern void srp_rport_del(struct srp_rport *);
-
+extern int srp_tmo_valid(int reconnect_delay, int fast_io_fail_tmo,
+			 int dev_loss_tmo);
+extern int srp_reconnect_rport(struct srp_rport *rport);
+extern void srp_start_tl_fail_timers(struct srp_rport *rport);
 extern void srp_remove_host(struct Scsi_Host *);
 
+/**
+ * srp_chkready() - evaluate the transport layer state before I/O
+ *
+ * Returns a SCSI result code that can be returned by the LLD queuecommand()
+ * implementation. The role of this function is similar to that of
+ * fc_remote_port_chkready().
+ */
+static inline int srp_chkready(struct srp_rport *rport)
+{
+	switch (rport->state) {
+	case SRP_RPORT_RUNNING:
+	case SRP_RPORT_BLOCKED:
+	default:
+		return 0;
+	case SRP_RPORT_FAIL_FAST:
+		return DID_TRANSPORT_FAILFAST << 16;
+	case SRP_RPORT_LOST:
+		return DID_NO_CONNECT << 16;
+	}
+}
+
 #endif
diff --git a/include/trace/events/bcache.h b/include/trace/events/bcache.h
index 5ebda976ea93..e2b9576d00e2 100644
--- a/include/trace/events/bcache.h
+++ b/include/trace/events/bcache.h
@@ -6,11 +6,9 @@
 
 #include <linux/tracepoint.h>
 
-struct search;
-
 DECLARE_EVENT_CLASS(bcache_request,
-	TP_PROTO(struct search *s, struct bio *bio),
-	TP_ARGS(s, bio),
+	TP_PROTO(struct bcache_device *d, struct bio *bio),
+	TP_ARGS(d, bio),
 
 	TP_STRUCT__entry(
 		__field(dev_t,		dev			)
@@ -24,8 +22,8 @@ DECLARE_EVENT_CLASS(bcache_request,
 
 	TP_fast_assign(
 		__entry->dev		= bio->bi_bdev->bd_dev;
-		__entry->orig_major	= s->d->disk->major;
-		__entry->orig_minor	= s->d->disk->first_minor;
+		__entry->orig_major	= d->disk->major;
+		__entry->orig_minor	= d->disk->first_minor;
 		__entry->sector		= bio->bi_sector;
 		__entry->orig_sector	= bio->bi_sector - 16;
 		__entry->nr_sector	= bio->bi_size >> 9;
@@ -79,13 +77,13 @@ DECLARE_EVENT_CLASS(btree_node,
 /* request.c */
 
 DEFINE_EVENT(bcache_request, bcache_request_start,
-	TP_PROTO(struct search *s, struct bio *bio),
-	TP_ARGS(s, bio)
+	TP_PROTO(struct bcache_device *d, struct bio *bio),
+	TP_ARGS(d, bio)
 );
 
 DEFINE_EVENT(bcache_request, bcache_request_end,
-	TP_PROTO(struct search *s, struct bio *bio),
-	TP_ARGS(s, bio)
+	TP_PROTO(struct bcache_device *d, struct bio *bio),
+	TP_ARGS(d, bio)
 );
 
 DECLARE_EVENT_CLASS(bcache_bio,
@@ -370,6 +368,35 @@ DEFINE_EVENT(btree_node, bcache_btree_set_root,
 	TP_ARGS(b)
 );
 
+TRACE_EVENT(bcache_keyscan,
+	TP_PROTO(unsigned nr_found,
+		 unsigned start_inode, uint64_t start_offset,
+		 unsigned end_inode, uint64_t end_offset),
+	TP_ARGS(nr_found,
+		start_inode, start_offset,
+		end_inode, end_offset),
+
+	TP_STRUCT__entry(
+		__field(__u32,	nr_found			)
+		__field(__u32,	start_inode			)
+		__field(__u64,	start_offset			)
+		__field(__u32,	end_inode			)
+		__field(__u64,	end_offset			)
+	),
+
+	TP_fast_assign(
+		__entry->nr_found	= nr_found;
+		__entry->start_inode	= start_inode;
+		__entry->start_offset	= start_offset;
+		__entry->end_inode	= end_inode;
+		__entry->end_offset	= end_offset;
+	),
+
+	TP_printk("found %u keys from %u:%llu to %u:%llu", __entry->nr_found,
+		  __entry->start_inode, __entry->start_offset,
+		  __entry->end_inode, __entry->end_offset)
+);
+
 /* Allocator */
 
 TRACE_EVENT(bcache_alloc_invalidate,
diff --git a/include/trace/events/iommu.h b/include/trace/events/iommu.h
new file mode 100644
index 000000000000..a8f5c32d174b
--- /dev/null
+++ b/include/trace/events/iommu.h
@@ -0,0 +1,162 @@
+/*
+ * iommu trace points
+ *
+ * Copyright (C) 2013 Shuah Khan <shuah.kh@samsung.com>
+ *
+ */
+#undef TRACE_SYSTEM
+#define TRACE_SYSTEM iommu
+
+#if !defined(_TRACE_IOMMU_H) || defined(TRACE_HEADER_MULTI_READ)
+#define _TRACE_IOMMU_H
+
+#include <linux/tracepoint.h>
+#include <linux/pci.h>
+
+struct device;
+
+DECLARE_EVENT_CLASS(iommu_group_event,
+
+	TP_PROTO(int group_id, struct device *dev),
+
+	TP_ARGS(group_id, dev),
+
+	TP_STRUCT__entry(
+		__field(int, gid)
+		__string(device, dev_name(dev))
+	),
+
+	TP_fast_assign(
+		__entry->gid = group_id;
+		__assign_str(device, dev_name(dev));
+	),
+
+	TP_printk("IOMMU: groupID=%d device=%s",
+			__entry->gid, __get_str(device)
+	)
+);
+
+DEFINE_EVENT(iommu_group_event, add_device_to_group,
+
+	TP_PROTO(int group_id, struct device *dev),
+
+	TP_ARGS(group_id, dev)
+
+);
+
+DEFINE_EVENT(iommu_group_event, remove_device_from_group,
+
+	TP_PROTO(int group_id, struct device *dev),
+
+	TP_ARGS(group_id, dev)
+);
+
+DECLARE_EVENT_CLASS(iommu_device_event,
+
+	TP_PROTO(struct device *dev),
+
+	TP_ARGS(dev),
+
+	TP_STRUCT__entry(
+		__string(device, dev_name(dev))
+	),
+
+	TP_fast_assign(
+		__assign_str(device, dev_name(dev));
+	),
+
+	TP_printk("IOMMU: device=%s", __get_str(device)
+	)
+);
+
+DEFINE_EVENT(iommu_device_event, attach_device_to_domain,
+
+	TP_PROTO(struct device *dev),
+
+	TP_ARGS(dev)
+);
+
+DEFINE_EVENT(iommu_device_event, detach_device_from_domain,
+
+	TP_PROTO(struct device *dev),
+
+	TP_ARGS(dev)
+);
+
+DECLARE_EVENT_CLASS(iommu_map_unmap,
+
+	TP_PROTO(unsigned long iova, phys_addr_t paddr, size_t size),
+
+	TP_ARGS(iova, paddr, size),
+
+	TP_STRUCT__entry(
+		__field(u64, iova)
+		__field(u64, paddr)
+		__field(int, size)
+	),
+
+	TP_fast_assign(
+		__entry->iova = iova;
+		__entry->paddr = paddr;
+		__entry->size = size;
+	),
+
+	TP_printk("IOMMU: iova=0x%016llx paddr=0x%016llx size=0x%x",
+			__entry->iova, __entry->paddr, __entry->size
+	)
+);
+
+DEFINE_EVENT(iommu_map_unmap, map,
+
+	TP_PROTO(unsigned long iova, phys_addr_t paddr, size_t size),
+
+	TP_ARGS(iova, paddr, size)
+);
+
+DEFINE_EVENT_PRINT(iommu_map_unmap, unmap,
+
+	TP_PROTO(unsigned long iova, phys_addr_t paddr, size_t size),
+
+	TP_ARGS(iova, paddr, size),
+
+	TP_printk("IOMMU: iova=0x%016llx size=0x%x",
+			__entry->iova, __entry->size
+	)
+);
+
+DECLARE_EVENT_CLASS(iommu_error,
+
+	TP_PROTO(struct device *dev, unsigned long iova, int flags),
+
+	TP_ARGS(dev, iova, flags),
+
+	TP_STRUCT__entry(
+		__string(device, dev_name(dev))
+		__string(driver, dev_driver_string(dev))
+		__field(u64, iova)
+		__field(int, flags)
+	),
+
+	TP_fast_assign(
+		__assign_str(device, dev_name(dev));
+		__assign_str(driver, dev_driver_string(dev));
+		__entry->iova = iova;
+		__entry->flags = flags;
+	),
+
+	TP_printk("IOMMU:%s %s iova=0x%016llx flags=0x%04x",
+			__get_str(driver), __get_str(device),
+			__entry->iova, __entry->flags
+	)
+);
+
+DEFINE_EVENT(iommu_error, io_page_fault,
+
+	TP_PROTO(struct device *dev, unsigned long iova, int flags),
+
+	TP_ARGS(dev, iova, flags)
+);
+#endif /* _TRACE_IOMMU_H */
+
+/* This part must be outside protection */
+#include <trace/define_trace.h>
diff --git a/include/trace/events/kvm.h b/include/trace/events/kvm.h
index 7005d1109ec9..131a0bda7aec 100644
--- a/include/trace/events/kvm.h
+++ b/include/trace/events/kvm.h
@@ -296,23 +296,21 @@ DEFINE_EVENT(kvm_async_pf_nopresent_ready, kvm_async_pf_ready,
 
 TRACE_EVENT(
 	kvm_async_pf_completed,
-	TP_PROTO(unsigned long address, struct page *page, u64 gva),
-	TP_ARGS(address, page, gva),
+	TP_PROTO(unsigned long address, u64 gva),
+	TP_ARGS(address, gva),
 
 	TP_STRUCT__entry(
 		__field(unsigned long, address)
-		__field(pfn_t, pfn)
 		__field(u64, gva)
 		),
 
 	TP_fast_assign(
 		__entry->address = address;
-		__entry->pfn = page ? page_to_pfn(page) : 0;
 		__entry->gva = gva;
 		),
 
-	TP_printk("gva %#llx address %#lx pfn %#llx",  __entry->gva,
-		  __entry->address, __entry->pfn)
+	TP_printk("gva %#llx address %#lx",  __entry->gva,
+		  __entry->address)
 );
 
 #endif
diff --git a/include/trace/events/random.h b/include/trace/events/random.h
index 422df19de732..805af6db41cc 100644
--- a/include/trace/events/random.h
+++ b/include/trace/events/random.h
@@ -7,6 +7,25 @@
 #include <linux/writeback.h>
 #include <linux/tracepoint.h>
 
+TRACE_EVENT(add_device_randomness,
+	TP_PROTO(int bytes, unsigned long IP),
+
+	TP_ARGS(bytes, IP),
+
+	TP_STRUCT__entry(
+		__field(	  int,	bytes			)
+		__field(unsigned long,	IP			)
+	),
+
+	TP_fast_assign(
+		__entry->bytes		= bytes;
+		__entry->IP		= IP;
+	),
+
+	TP_printk("bytes %d caller %pF",
+		__entry->bytes, (void *)__entry->IP)
+);
+
 DECLARE_EVENT_CLASS(random__mix_pool_bytes,
 	TP_PROTO(const char *pool_name, int bytes, unsigned long IP),
 
@@ -68,7 +87,112 @@ TRACE_EVENT(credit_entropy_bits,
 		  (void *)__entry->IP)
 );
 
-TRACE_EVENT(get_random_bytes,
+TRACE_EVENT(push_to_pool,
+	TP_PROTO(const char *pool_name, int pool_bits, int input_bits),
+
+	TP_ARGS(pool_name, pool_bits, input_bits),
+
+	TP_STRUCT__entry(
+		__field( const char *,	pool_name		)
+		__field(	  int,	pool_bits		)
+		__field(	  int,	input_bits		)
+	),
+
+	TP_fast_assign(
+		__entry->pool_name	= pool_name;
+		__entry->pool_bits	= pool_bits;
+		__entry->input_bits	= input_bits;
+	),
+
+	TP_printk("%s: pool_bits %d input_pool_bits %d",
+		  __entry->pool_name, __entry->pool_bits,
+		  __entry->input_bits)
+);
+
+TRACE_EVENT(debit_entropy,
+	TP_PROTO(const char *pool_name, int debit_bits),
+
+	TP_ARGS(pool_name, debit_bits),
+
+	TP_STRUCT__entry(
+		__field( const char *,	pool_name		)
+		__field(	  int,	debit_bits		)
+	),
+
+	TP_fast_assign(
+		__entry->pool_name	= pool_name;
+		__entry->debit_bits	= debit_bits;
+	),
+
+	TP_printk("%s: debit_bits %d", __entry->pool_name,
+		  __entry->debit_bits)
+);
+
+TRACE_EVENT(add_input_randomness,
+	TP_PROTO(int input_bits),
+
+	TP_ARGS(input_bits),
+
+	TP_STRUCT__entry(
+		__field(	  int,	input_bits		)
+	),
+
+	TP_fast_assign(
+		__entry->input_bits	= input_bits;
+	),
+
+	TP_printk("input_pool_bits %d", __entry->input_bits)
+);
+
+TRACE_EVENT(add_disk_randomness,
+	TP_PROTO(dev_t dev, int input_bits),
+
+	TP_ARGS(dev, input_bits),
+
+	TP_STRUCT__entry(
+		__field(	dev_t,	dev			)
+		__field(	  int,	input_bits		)
+	),
+
+	TP_fast_assign(
+		__entry->dev		= dev;
+		__entry->input_bits	= input_bits;
+	),
+
+	TP_printk("dev %d,%d input_pool_bits %d", MAJOR(__entry->dev),
+		  MINOR(__entry->dev), __entry->input_bits)
+);
+
+TRACE_EVENT(xfer_secondary_pool,
+	TP_PROTO(const char *pool_name, int xfer_bits, int request_bits,
+		 int pool_entropy, int input_entropy),
+
+	TP_ARGS(pool_name, xfer_bits, request_bits, pool_entropy,
+		input_entropy),
+
+	TP_STRUCT__entry(
+		__field( const char *,	pool_name		)
+		__field(	  int,	xfer_bits		)
+		__field(	  int,	request_bits		)
+		__field(	  int,	pool_entropy		)
+		__field(	  int,	input_entropy		)
+	),
+
+	TP_fast_assign(
+		__entry->pool_name	= pool_name;
+		__entry->xfer_bits	= xfer_bits;
+		__entry->request_bits	= request_bits;
+		__entry->pool_entropy	= pool_entropy;
+		__entry->input_entropy	= input_entropy;
+	),
+
+	TP_printk("pool %s xfer_bits %d request_bits %d pool_entropy %d "
+		  "input_entropy %d", __entry->pool_name, __entry->xfer_bits,
+		  __entry->request_bits, __entry->pool_entropy,
+		  __entry->input_entropy)
+);
+
+DECLARE_EVENT_CLASS(random__get_random_bytes,
 	TP_PROTO(int nbytes, unsigned long IP),
 
 	TP_ARGS(nbytes, IP),
@@ -86,6 +210,18 @@ TRACE_EVENT(get_random_bytes,
 	TP_printk("nbytes %d caller %pF", __entry->nbytes, (void *)__entry->IP)
 );
 
+DEFINE_EVENT(random__get_random_bytes, get_random_bytes,
+	TP_PROTO(int nbytes, unsigned long IP),
+
+	TP_ARGS(nbytes, IP)
+);
+
+DEFINE_EVENT(random__get_random_bytes, get_random_bytes_arch,
+	TP_PROTO(int nbytes, unsigned long IP),
+
+	TP_ARGS(nbytes, IP)
+);
+
 DECLARE_EVENT_CLASS(random__extract_entropy,
 	TP_PROTO(const char *pool_name, int nbytes, int entropy_count,
 		 unsigned long IP),
@@ -126,7 +262,52 @@ DEFINE_EVENT(random__extract_entropy, extract_entropy_user,
 	TP_ARGS(pool_name, nbytes, entropy_count, IP)
 );
 
+TRACE_EVENT(random_read,
+	TP_PROTO(int got_bits, int need_bits, int pool_left, int input_left),
+
+	TP_ARGS(got_bits, need_bits, pool_left, input_left),
+
+	TP_STRUCT__entry(
+		__field(	  int,	got_bits		)
+		__field(	  int,	need_bits		)
+		__field(	  int,	pool_left		)
+		__field(	  int,	input_left		)
+	),
+
+	TP_fast_assign(
+		__entry->got_bits	= got_bits;
+		__entry->need_bits	= need_bits;
+		__entry->pool_left	= pool_left;
+		__entry->input_left	= input_left;
+	),
+
+	TP_printk("got_bits %d still_needed_bits %d "
+		  "blocking_pool_entropy_left %d input_entropy_left %d",
+		  __entry->got_bits, __entry->got_bits, __entry->pool_left,
+		  __entry->input_left)
+);
+
+TRACE_EVENT(urandom_read,
+	TP_PROTO(int got_bits, int pool_left, int input_left),
+
+	TP_ARGS(got_bits, pool_left, input_left),
+
+	TP_STRUCT__entry(
+		__field(	  int,	got_bits		)
+		__field(	  int,	pool_left		)
+		__field(	  int,	input_left		)
+	),
+
+	TP_fast_assign(
+		__entry->got_bits	= got_bits;
+		__entry->pool_left	= pool_left;
+		__entry->input_left	= input_left;
+	),
 
+	TP_printk("got_bits %d nonblocking_pool_entropy_left %d "
+		  "input_entropy_left %d", __entry->got_bits,
+		  __entry->pool_left, __entry->input_left)
+);
 
 #endif /* _TRACE_RANDOM_H */
 
diff --git a/include/trace/events/sched.h b/include/trace/events/sched.h
index 613381bcde40..04c308413a5d 100644
--- a/include/trace/events/sched.h
+++ b/include/trace/events/sched.h
@@ -424,6 +424,25 @@ TRACE_EVENT(sched_pi_setprio,
 			__entry->oldprio, __entry->newprio)
 );
 
+#ifdef CONFIG_DETECT_HUNG_TASK
+TRACE_EVENT(sched_process_hang,
+	TP_PROTO(struct task_struct *tsk),
+	TP_ARGS(tsk),
+
+	TP_STRUCT__entry(
+		__array( char,	comm,	TASK_COMM_LEN	)
+		__field( pid_t,	pid			)
+	),
+
+	TP_fast_assign(
+		memcpy(__entry->comm, tsk->comm, TASK_COMM_LEN);
+		__entry->pid = tsk->pid;
+	),
+
+	TP_printk("comm=%s pid=%d", __entry->comm, __entry->pid)
+);
+#endif /* CONFIG_DETECT_HUNG_TASK */
+
 #endif /* _TRACE_SCHED_H */
 
 /* This part must be outside protection */
diff --git a/include/trace/events/swiotlb.h b/include/trace/events/swiotlb.h
new file mode 100644
index 000000000000..7ea4c5e7c448
--- /dev/null
+++ b/include/trace/events/swiotlb.h
@@ -0,0 +1,46 @@
+#undef TRACE_SYSTEM
+#define TRACE_SYSTEM swiotlb
+
+#if !defined(_TRACE_SWIOTLB_H) || defined(TRACE_HEADER_MULTI_READ)
+#define _TRACE_SWIOTLB_H
+
+#include <linux/tracepoint.h>
+
+TRACE_EVENT(swiotlb_bounced,
+
+	TP_PROTO(struct device *dev,
+		 dma_addr_t dev_addr,
+		 size_t size,
+		 int swiotlb_force),
+
+	TP_ARGS(dev, dev_addr, size, swiotlb_force),
+
+	TP_STRUCT__entry(
+		__string(	dev_name,	dev_name(dev)	)
+		__field(	u64,	dma_mask		)
+		__field(	dma_addr_t,	dev_addr	)
+		__field(	size_t,	size			)
+		__field(	int,	swiotlb_force		)
+	),
+
+	TP_fast_assign(
+		__assign_str(dev_name, dev_name(dev));
+		__entry->dma_mask = (dev->dma_mask ? *dev->dma_mask : 0);
+		__entry->dev_addr = dev_addr;
+		__entry->size = size;
+		__entry->swiotlb_force = swiotlb_force;
+	),
+
+	TP_printk("dev_name: %s dma_mask=%llx dev_addr=%llx "
+		"size=%zu %s",
+		__get_str(dev_name),
+		__entry->dma_mask,
+		(unsigned long long)__entry->dev_addr,
+		__entry->size,
+		__entry->swiotlb_force ? "swiotlb_force" : "" )
+);
+
+#endif /*  _TRACE_SWIOTLB_H */
+
+/* This part must be outside protection */
+#include <trace/define_trace.h>
diff --git a/include/trace/ftrace.h b/include/trace/ftrace.h
index 5c7ab17cbb02..52594b20179e 100644
--- a/include/trace/ftrace.h
+++ b/include/trace/ftrace.h
@@ -437,9 +437,8 @@ static inline notrace int ftrace_get_offsets_##call(			\
  *	{ <assign>; }  <-- Here we assign the entries by the __field and
  *			   __array macros.
  *
- *	if (!filter_current_check_discard(buffer, event_call, entry, event))
- *		trace_nowake_buffer_unlock_commit(buffer,
- *						   event, irq_flags, pc);
+ *	if (!filter_check_discard(ftrace_file, entry, buffer, event))
+ *		trace_buffer_unlock_commit(buffer, event, irq_flags, pc);
  * }
  *
  * static struct trace_event ftrace_event_type_<call> = {
@@ -553,7 +552,7 @@ ftrace_raw_event_##call(void *__data, proto)				\
 									\
 	{ assign; }							\
 									\
-	if (!filter_current_check_discard(buffer, event_call, entry, event)) \
+	if (!filter_check_discard(ftrace_file, entry, buffer, event))	\
 		trace_buffer_unlock_commit(buffer, event, irq_flags, pc); \
 }
 /*
diff --git a/include/uapi/drm/armada_drm.h b/include/uapi/drm/armada_drm.h
new file mode 100644
index 000000000000..8dec3fdc99c7
--- /dev/null
+++ b/include/uapi/drm/armada_drm.h
@@ -0,0 +1,45 @@
+/*
+ * Copyright (C) 2012 Russell King
+ *  With inspiration from the i915 driver
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ */
+#ifndef DRM_ARMADA_IOCTL_H
+#define DRM_ARMADA_IOCTL_H
+
+#define DRM_ARMADA_GEM_CREATE		0x00
+#define DRM_ARMADA_GEM_MMAP		0x02
+#define DRM_ARMADA_GEM_PWRITE		0x03
+
+#define ARMADA_IOCTL(dir, name, str) \
+	DRM_##dir(DRM_COMMAND_BASE + DRM_ARMADA_##name, struct drm_armada_##str)
+
+struct drm_armada_gem_create {
+	uint32_t handle;
+	uint32_t size;
+};
+#define DRM_IOCTL_ARMADA_GEM_CREATE \
+	ARMADA_IOCTL(IOWR, GEM_CREATE, gem_create)
+
+struct drm_armada_gem_mmap {
+	uint32_t handle;
+	uint32_t pad;
+	uint64_t offset;
+	uint64_t size;
+	uint64_t addr;
+};
+#define DRM_IOCTL_ARMADA_GEM_MMAP \
+	ARMADA_IOCTL(IOWR, GEM_MMAP, gem_mmap)
+
+struct drm_armada_gem_pwrite {
+	uint64_t ptr;
+	uint32_t handle;
+	uint32_t offset;
+	uint32_t size;
+};
+#define DRM_IOCTL_ARMADA_GEM_PWRITE \
+	ARMADA_IOCTL(IOW, GEM_PWRITE, gem_pwrite)
+
+#endif
diff --git a/include/uapi/drm/drm.h b/include/uapi/drm/drm.h
index ece867889cc7..9b24d65fed72 100644
--- a/include/uapi/drm/drm.h
+++ b/include/uapi/drm/drm.h
@@ -611,12 +611,37 @@ struct drm_gem_open {
 	__u64 size;
 };
 
+#define DRM_CAP_DUMB_BUFFER		0x1
+#define DRM_CAP_VBLANK_HIGH_CRTC	0x2
+#define DRM_CAP_DUMB_PREFERRED_DEPTH	0x3
+#define DRM_CAP_DUMB_PREFER_SHADOW	0x4
+#define DRM_CAP_PRIME			0x5
+#define  DRM_PRIME_CAP_IMPORT		0x1
+#define  DRM_PRIME_CAP_EXPORT		0x2
+#define DRM_CAP_TIMESTAMP_MONOTONIC	0x6
+#define DRM_CAP_ASYNC_PAGE_FLIP		0x7
+
 /** DRM_IOCTL_GET_CAP ioctl argument type */
 struct drm_get_cap {
 	__u64 capability;
 	__u64 value;
 };
 
+/**
+ * DRM_CLIENT_CAP_STEREO_3D
+ *
+ * if set to 1, the DRM core will expose the stereo 3D capabilities of the
+ * monitor by advertising the supported 3D layouts in the flags of struct
+ * drm_mode_modeinfo.
+ */
+#define DRM_CLIENT_CAP_STEREO_3D	1
+
+/** DRM_IOCTL_SET_CLIENT_CAP ioctl argument type */
+struct drm_set_client_cap {
+	__u64 capability;
+	__u64 value;
+};
+
 #define DRM_CLOEXEC O_CLOEXEC
 struct drm_prime_handle {
 	__u32 handle;
@@ -649,6 +674,7 @@ struct drm_prime_handle {
 #define DRM_IOCTL_GEM_FLINK		DRM_IOWR(0x0a, struct drm_gem_flink)
 #define DRM_IOCTL_GEM_OPEN		DRM_IOWR(0x0b, struct drm_gem_open)
 #define DRM_IOCTL_GET_CAP		DRM_IOWR(0x0c, struct drm_get_cap)
+#define DRM_IOCTL_SET_CLIENT_CAP	DRM_IOW( 0x0d, struct drm_set_client_cap)
 
 #define DRM_IOCTL_SET_UNIQUE		DRM_IOW( 0x10, struct drm_unique)
 #define DRM_IOCTL_AUTH_MAGIC		DRM_IOW( 0x11, struct drm_auth)
@@ -774,17 +800,6 @@ struct drm_event_vblank {
 	__u32 reserved;
 };
 
-#define DRM_CAP_DUMB_BUFFER 0x1
-#define DRM_CAP_VBLANK_HIGH_CRTC 0x2
-#define DRM_CAP_DUMB_PREFERRED_DEPTH 0x3
-#define DRM_CAP_DUMB_PREFER_SHADOW 0x4
-#define DRM_CAP_PRIME 0x5
-#define DRM_CAP_TIMESTAMP_MONOTONIC 0x6
-#define DRM_CAP_ASYNC_PAGE_FLIP 0x7
-
-#define DRM_PRIME_CAP_IMPORT 0x1
-#define DRM_PRIME_CAP_EXPORT 0x2
-
 /* typedef area */
 #ifndef __KERNEL__
 typedef struct drm_clip_rect drm_clip_rect_t;
diff --git a/include/uapi/drm/drm_mode.h b/include/uapi/drm/drm_mode.h
index 28acbaf4a81e..f104c2603ebe 100644
--- a/include/uapi/drm/drm_mode.h
+++ b/include/uapi/drm/drm_mode.h
@@ -44,20 +44,35 @@
 
 /* Video mode flags */
 /* bit compatible with the xorg definitions. */
-#define DRM_MODE_FLAG_PHSYNC	(1<<0)
-#define DRM_MODE_FLAG_NHSYNC	(1<<1)
-#define DRM_MODE_FLAG_PVSYNC	(1<<2)
-#define DRM_MODE_FLAG_NVSYNC	(1<<3)
-#define DRM_MODE_FLAG_INTERLACE	(1<<4)
-#define DRM_MODE_FLAG_DBLSCAN	(1<<5)
-#define DRM_MODE_FLAG_CSYNC	(1<<6)
-#define DRM_MODE_FLAG_PCSYNC	(1<<7)
-#define DRM_MODE_FLAG_NCSYNC	(1<<8)
-#define DRM_MODE_FLAG_HSKEW	(1<<9) /* hskew provided */
-#define DRM_MODE_FLAG_BCAST	(1<<10)
-#define DRM_MODE_FLAG_PIXMUX	(1<<11)
-#define DRM_MODE_FLAG_DBLCLK	(1<<12)
-#define DRM_MODE_FLAG_CLKDIV2	(1<<13)
+#define DRM_MODE_FLAG_PHSYNC			(1<<0)
+#define DRM_MODE_FLAG_NHSYNC			(1<<1)
+#define DRM_MODE_FLAG_PVSYNC			(1<<2)
+#define DRM_MODE_FLAG_NVSYNC			(1<<3)
+#define DRM_MODE_FLAG_INTERLACE			(1<<4)
+#define DRM_MODE_FLAG_DBLSCAN			(1<<5)
+#define DRM_MODE_FLAG_CSYNC			(1<<6)
+#define DRM_MODE_FLAG_PCSYNC			(1<<7)
+#define DRM_MODE_FLAG_NCSYNC			(1<<8)
+#define DRM_MODE_FLAG_HSKEW			(1<<9) /* hskew provided */
+#define DRM_MODE_FLAG_BCAST			(1<<10)
+#define DRM_MODE_FLAG_PIXMUX			(1<<11)
+#define DRM_MODE_FLAG_DBLCLK			(1<<12)
+#define DRM_MODE_FLAG_CLKDIV2			(1<<13)
+ /*
+  * When adding a new stereo mode don't forget to adjust DRM_MODE_FLAGS_3D_MAX
+  * (define not exposed to user space).
+  */
+#define DRM_MODE_FLAG_3D_MASK			(0x1f<<14)
+#define  DRM_MODE_FLAG_3D_NONE			(0<<14)
+#define  DRM_MODE_FLAG_3D_FRAME_PACKING		(1<<14)
+#define  DRM_MODE_FLAG_3D_FIELD_ALTERNATIVE	(2<<14)
+#define  DRM_MODE_FLAG_3D_LINE_ALTERNATIVE	(3<<14)
+#define  DRM_MODE_FLAG_3D_SIDE_BY_SIDE_FULL	(4<<14)
+#define  DRM_MODE_FLAG_3D_L_DEPTH		(5<<14)
+#define  DRM_MODE_FLAG_3D_L_DEPTH_GFX_GFX_DEPTH	(6<<14)
+#define  DRM_MODE_FLAG_3D_TOP_AND_BOTTOM	(7<<14)
+#define  DRM_MODE_FLAG_3D_SIDE_BY_SIDE_HALF	(8<<14)
+
 
 /* DPMS flags */
 /* bit compatible with the xorg definitions. */
@@ -165,6 +180,7 @@ struct drm_mode_get_plane_res {
 #define DRM_MODE_ENCODER_LVDS	3
 #define DRM_MODE_ENCODER_TVDAC	4
 #define DRM_MODE_ENCODER_VIRTUAL 5
+#define DRM_MODE_ENCODER_DSI	6
 
 struct drm_mode_get_encoder {
 	__u32 encoder_id;
@@ -203,6 +219,7 @@ struct drm_mode_get_encoder {
 #define DRM_MODE_CONNECTOR_TV		13
 #define DRM_MODE_CONNECTOR_eDP		14
 #define DRM_MODE_CONNECTOR_VIRTUAL      15
+#define DRM_MODE_CONNECTOR_DSI		16
 
 struct drm_mode_get_connector {
 
diff --git a/include/uapi/drm/i915_drm.h b/include/uapi/drm/i915_drm.h
index 55bb5729bd78..3a4e97bd8607 100644
--- a/include/uapi/drm/i915_drm.h
+++ b/include/uapi/drm/i915_drm.h
@@ -38,10 +38,10 @@
  *
  * I915_L3_PARITY_UEVENT - Generated when the driver receives a parity mismatch
  *	event from the gpu l3 cache. Additional information supplied is ROW,
- *	BANK, SUBBANK of the affected cacheline. Userspace should keep track of
- *	these events and if a specific cache-line seems to have a persistent
- *	error remap it with the l3 remapping tool supplied in intel-gpu-tools.
- *	The value supplied with the event is always 1.
+ *	BANK, SUBBANK, SLICE of the affected cacheline. Userspace should keep
+ *	track of these events and if a specific cache-line seems to have a
+ *	persistent error remap it with the l3 remapping tool supplied in
+ *	intel-gpu-tools.  The value supplied with the event is always 1.
  *
  * I915_ERROR_UEVENT - Generated upon error detection, currently only via
  *	hangcheck. The error detection event is a good indicator of when things
diff --git a/include/uapi/drm/tegra_drm.h b/include/uapi/drm/tegra_drm.h
index 73bde4eaf16c..5e1ab552cbed 100644
--- a/include/uapi/drm/tegra_drm.h
+++ b/include/uapi/drm/tegra_drm.h
@@ -19,6 +19,9 @@
 
 #include <drm/drm.h>
 
+#define DRM_TEGRA_GEM_CREATE_TILED     (1 << 0)
+#define DRM_TEGRA_GEM_CREATE_BOTTOM_UP (1 << 1)
+
 struct drm_tegra_gem_create {
 	__u64 size;
 	__u32 flags;
@@ -65,6 +68,12 @@ struct drm_tegra_get_syncpt {
 	__u32 id;
 };
 
+struct drm_tegra_get_syncpt_base {
+	__u64 context;
+	__u32 syncpt;
+	__u32 id;
+};
+
 struct drm_tegra_syncpt {
 	__u32 id;
 	__u32 incrs;
@@ -115,15 +124,16 @@ struct drm_tegra_submit {
 	__u32 reserved[5];	/* future expansion */
 };
 
-#define DRM_TEGRA_GEM_CREATE	0x00
-#define DRM_TEGRA_GEM_MMAP	0x01
-#define DRM_TEGRA_SYNCPT_READ	0x02
-#define DRM_TEGRA_SYNCPT_INCR	0x03
-#define DRM_TEGRA_SYNCPT_WAIT	0x04
-#define DRM_TEGRA_OPEN_CHANNEL	0x05
-#define DRM_TEGRA_CLOSE_CHANNEL	0x06
-#define DRM_TEGRA_GET_SYNCPT	0x07
-#define DRM_TEGRA_SUBMIT	0x08
+#define DRM_TEGRA_GEM_CREATE		0x00
+#define DRM_TEGRA_GEM_MMAP		0x01
+#define DRM_TEGRA_SYNCPT_READ		0x02
+#define DRM_TEGRA_SYNCPT_INCR		0x03
+#define DRM_TEGRA_SYNCPT_WAIT		0x04
+#define DRM_TEGRA_OPEN_CHANNEL		0x05
+#define DRM_TEGRA_CLOSE_CHANNEL		0x06
+#define DRM_TEGRA_GET_SYNCPT		0x07
+#define DRM_TEGRA_SUBMIT		0x08
+#define DRM_TEGRA_GET_SYNCPT_BASE	0x09
 
 #define DRM_IOCTL_TEGRA_GEM_CREATE DRM_IOWR(DRM_COMMAND_BASE + DRM_TEGRA_GEM_CREATE, struct drm_tegra_gem_create)
 #define DRM_IOCTL_TEGRA_GEM_MMAP DRM_IOWR(DRM_COMMAND_BASE + DRM_TEGRA_GEM_MMAP, struct drm_tegra_gem_mmap)
@@ -134,5 +144,6 @@ struct drm_tegra_submit {
 #define DRM_IOCTL_TEGRA_CLOSE_CHANNEL DRM_IOWR(DRM_COMMAND_BASE + DRM_TEGRA_CLOSE_CHANNEL, struct drm_tegra_open_channel)
 #define DRM_IOCTL_TEGRA_GET_SYNCPT DRM_IOWR(DRM_COMMAND_BASE + DRM_TEGRA_GET_SYNCPT, struct drm_tegra_get_syncpt)
 #define DRM_IOCTL_TEGRA_SUBMIT DRM_IOWR(DRM_COMMAND_BASE + DRM_TEGRA_SUBMIT, struct drm_tegra_submit)
+#define DRM_IOCTL_TEGRA_GET_SYNCPT_BASE DRM_IOWR(DRM_COMMAND_BASE + DRM_TEGRA_GET_SYNCPT_BASE, struct drm_tegra_get_syncpt_base)
 
 #endif
diff --git a/include/uapi/linux/bcache.h b/include/uapi/linux/bcache.h
new file mode 100644
index 000000000000..164a7e263988
--- /dev/null
+++ b/include/uapi/linux/bcache.h
@@ -0,0 +1,373 @@
+#ifndef _LINUX_BCACHE_H
+#define _LINUX_BCACHE_H
+
+/*
+ * Bcache on disk data structures
+ */
+
+#include <asm/types.h>
+
+#define BITMASK(name, type, field, offset, size)		\
+static inline __u64 name(const type *k)				\
+{ return (k->field >> offset) & ~(~0ULL << size); }		\
+								\
+static inline void SET_##name(type *k, __u64 v)			\
+{								\
+	k->field &= ~(~(~0ULL << size) << offset);		\
+	k->field |= (v & ~(~0ULL << size)) << offset;		\
+}
+
+/* Btree keys - all units are in sectors */
+
+struct bkey {
+	__u64	high;
+	__u64	low;
+	__u64	ptr[];
+};
+
+#define KEY_FIELD(name, field, offset, size)				\
+	BITMASK(name, struct bkey, field, offset, size)
+
+#define PTR_FIELD(name, offset, size)					\
+static inline __u64 name(const struct bkey *k, unsigned i)		\
+{ return (k->ptr[i] >> offset) & ~(~0ULL << size); }			\
+									\
+static inline void SET_##name(struct bkey *k, unsigned i, __u64 v)	\
+{									\
+	k->ptr[i] &= ~(~(~0ULL << size) << offset);			\
+	k->ptr[i] |= (v & ~(~0ULL << size)) << offset;			\
+}
+
+#define KEY_SIZE_BITS		16
+
+KEY_FIELD(KEY_PTRS,	high, 60, 3)
+KEY_FIELD(HEADER_SIZE,	high, 58, 2)
+KEY_FIELD(KEY_CSUM,	high, 56, 2)
+KEY_FIELD(KEY_PINNED,	high, 55, 1)
+KEY_FIELD(KEY_DIRTY,	high, 36, 1)
+
+KEY_FIELD(KEY_SIZE,	high, 20, KEY_SIZE_BITS)
+KEY_FIELD(KEY_INODE,	high, 0,  20)
+
+/* Next time I change the on disk format, KEY_OFFSET() won't be 64 bits */
+
+static inline __u64 KEY_OFFSET(const struct bkey *k)
+{
+	return k->low;
+}
+
+static inline void SET_KEY_OFFSET(struct bkey *k, __u64 v)
+{
+	k->low = v;
+}
+
+/*
+ * The high bit being set is a relic from when we used it to do binary
+ * searches - it told you where a key started. It's not used anymore,
+ * and can probably be safely dropped.
+ */
+#define KEY(inode, offset, size)					\
+((struct bkey) {							\
+	.high = (1ULL << 63) | ((__u64) (size) << 20) | (inode),	\
+	.low = (offset)							\
+})
+
+#define ZERO_KEY			KEY(0, 0, 0)
+
+#define MAX_KEY_INODE			(~(~0 << 20))
+#define MAX_KEY_OFFSET			(~0ULL >> 1)
+#define MAX_KEY				KEY(MAX_KEY_INODE, MAX_KEY_OFFSET, 0)
+
+#define KEY_START(k)			(KEY_OFFSET(k) - KEY_SIZE(k))
+#define START_KEY(k)			KEY(KEY_INODE(k), KEY_START(k), 0)
+
+#define PTR_DEV_BITS			12
+
+PTR_FIELD(PTR_DEV,			51, PTR_DEV_BITS)
+PTR_FIELD(PTR_OFFSET,			8,  43)
+PTR_FIELD(PTR_GEN,			0,  8)
+
+#define PTR_CHECK_DEV			((1 << PTR_DEV_BITS) - 1)
+
+#define PTR(gen, offset, dev)						\
+	((((__u64) dev) << 51) | ((__u64) offset) << 8 | gen)
+
+/* Bkey utility code */
+
+static inline unsigned long bkey_u64s(const struct bkey *k)
+{
+	return (sizeof(struct bkey) / sizeof(__u64)) + KEY_PTRS(k);
+}
+
+static inline unsigned long bkey_bytes(const struct bkey *k)
+{
+	return bkey_u64s(k) * sizeof(__u64);
+}
+
+#define bkey_copy(_dest, _src)	memcpy(_dest, _src, bkey_bytes(_src))
+
+static inline void bkey_copy_key(struct bkey *dest, const struct bkey *src)
+{
+	SET_KEY_INODE(dest, KEY_INODE(src));
+	SET_KEY_OFFSET(dest, KEY_OFFSET(src));
+}
+
+static inline struct bkey *bkey_next(const struct bkey *k)
+{
+	__u64 *d = (void *) k;
+	return (struct bkey *) (d + bkey_u64s(k));
+}
+
+static inline struct bkey *bkey_last(const struct bkey *k, unsigned nr_keys)
+{
+	__u64 *d = (void *) k;
+	return (struct bkey *) (d + nr_keys);
+}
+/* Enough for a key with 6 pointers */
+#define BKEY_PAD		8
+
+#define BKEY_PADDED(key)					\
+	union { struct bkey key; __u64 key ## _pad[BKEY_PAD]; }
+
+/* Superblock */
+
+/* Version 0: Cache device
+ * Version 1: Backing device
+ * Version 2: Seed pointer into btree node checksum
+ * Version 3: Cache device with new UUID format
+ * Version 4: Backing device with data offset
+ */
+#define BCACHE_SB_VERSION_CDEV		0
+#define BCACHE_SB_VERSION_BDEV		1
+#define BCACHE_SB_VERSION_CDEV_WITH_UUID 3
+#define BCACHE_SB_VERSION_BDEV_WITH_OFFSET 4
+#define BCACHE_SB_MAX_VERSION		4
+
+#define SB_SECTOR			8
+#define SB_SIZE				4096
+#define SB_LABEL_SIZE			32
+#define SB_JOURNAL_BUCKETS		256U
+/* SB_JOURNAL_BUCKETS must be divisible by BITS_PER_LONG */
+#define MAX_CACHES_PER_SET		8
+
+#define BDEV_DATA_START_DEFAULT		16	/* sectors */
+
+struct cache_sb {
+	__u64			csum;
+	__u64			offset;	/* sector where this sb was written */
+	__u64			version;
+
+	__u8			magic[16];
+
+	__u8			uuid[16];
+	union {
+		__u8		set_uuid[16];
+		__u64		set_magic;
+	};
+	__u8			label[SB_LABEL_SIZE];
+
+	__u64			flags;
+	__u64			seq;
+	__u64			pad[8];
+
+	union {
+	struct {
+		/* Cache devices */
+		__u64		nbuckets;	/* device size */
+
+		__u16		block_size;	/* sectors */
+		__u16		bucket_size;	/* sectors */
+
+		__u16		nr_in_set;
+		__u16		nr_this_dev;
+	};
+	struct {
+		/* Backing devices */
+		__u64		data_offset;
+
+		/*
+		 * block_size from the cache device section is still used by
+		 * backing devices, so don't add anything here until we fix
+		 * things to not need it for backing devices anymore
+		 */
+	};
+	};
+
+	__u32			last_mount;	/* time_t */
+
+	__u16			first_bucket;
+	union {
+		__u16		njournal_buckets;
+		__u16		keys;
+	};
+	__u64			d[SB_JOURNAL_BUCKETS];	/* journal buckets */
+};
+
+static inline _Bool SB_IS_BDEV(const struct cache_sb *sb)
+{
+	return sb->version == BCACHE_SB_VERSION_BDEV
+		|| sb->version == BCACHE_SB_VERSION_BDEV_WITH_OFFSET;
+}
+
+BITMASK(CACHE_SYNC,			struct cache_sb, flags, 0, 1);
+BITMASK(CACHE_DISCARD,			struct cache_sb, flags, 1, 1);
+BITMASK(CACHE_REPLACEMENT,		struct cache_sb, flags, 2, 3);
+#define CACHE_REPLACEMENT_LRU		0U
+#define CACHE_REPLACEMENT_FIFO		1U
+#define CACHE_REPLACEMENT_RANDOM	2U
+
+BITMASK(BDEV_CACHE_MODE,		struct cache_sb, flags, 0, 4);
+#define CACHE_MODE_WRITETHROUGH		0U
+#define CACHE_MODE_WRITEBACK		1U
+#define CACHE_MODE_WRITEAROUND		2U
+#define CACHE_MODE_NONE			3U
+BITMASK(BDEV_STATE,			struct cache_sb, flags, 61, 2);
+#define BDEV_STATE_NONE			0U
+#define BDEV_STATE_CLEAN		1U
+#define BDEV_STATE_DIRTY		2U
+#define BDEV_STATE_STALE		3U
+
+/*
+ * Magic numbers
+ *
+ * The various other data structures have their own magic numbers, which are
+ * xored with the first part of the cache set's UUID
+ */
+
+#define JSET_MAGIC			0x245235c1a3625032ULL
+#define PSET_MAGIC			0x6750e15f87337f91ULL
+#define BSET_MAGIC			0x90135c78b99e07f5ULL
+
+static inline __u64 jset_magic(struct cache_sb *sb)
+{
+	return sb->set_magic ^ JSET_MAGIC;
+}
+
+static inline __u64 pset_magic(struct cache_sb *sb)
+{
+	return sb->set_magic ^ PSET_MAGIC;
+}
+
+static inline __u64 bset_magic(struct cache_sb *sb)
+{
+	return sb->set_magic ^ BSET_MAGIC;
+}
+
+/*
+ * Journal
+ *
+ * On disk format for a journal entry:
+ * seq is monotonically increasing; every journal entry has its own unique
+ * sequence number.
+ *
+ * last_seq is the oldest journal entry that still has keys the btree hasn't
+ * flushed to disk yet.
+ *
+ * version is for on disk format changes.
+ */
+
+#define BCACHE_JSET_VERSION_UUIDv1	1
+#define BCACHE_JSET_VERSION_UUID	1	/* Always latest UUID format */
+#define BCACHE_JSET_VERSION		1
+
+struct jset {
+	__u64			csum;
+	__u64			magic;
+	__u64			seq;
+	__u32			version;
+	__u32			keys;
+
+	__u64			last_seq;
+
+	BKEY_PADDED(uuid_bucket);
+	BKEY_PADDED(btree_root);
+	__u16			btree_level;
+	__u16			pad[3];
+
+	__u64			prio_bucket[MAX_CACHES_PER_SET];
+
+	union {
+		struct bkey	start[0];
+		__u64		d[0];
+	};
+};
+
+/* Bucket prios/gens */
+
+struct prio_set {
+	__u64			csum;
+	__u64			magic;
+	__u64			seq;
+	__u32			version;
+	__u32			pad;
+
+	__u64			next_bucket;
+
+	struct bucket_disk {
+		__u16		prio;
+		__u8		gen;
+	} __attribute((packed)) data[];
+};
+
+/* UUIDS - per backing device/flash only volume metadata */
+
+struct uuid_entry {
+	union {
+		struct {
+			__u8	uuid[16];
+			__u8	label[32];
+			__u32	first_reg;
+			__u32	last_reg;
+			__u32	invalidated;
+
+			__u32	flags;
+			/* Size of flash only volumes */
+			__u64	sectors;
+		};
+
+		__u8		pad[128];
+	};
+};
+
+BITMASK(UUID_FLASH_ONLY,	struct uuid_entry, flags, 0, 1);
+
+/* Btree nodes */
+
+/* Version 1: Seed pointer into btree node checksum
+ */
+#define BCACHE_BSET_CSUM		1
+#define BCACHE_BSET_VERSION		1
+
+/*
+ * Btree nodes
+ *
+ * On disk a btree node is a list/log of these; within each set the keys are
+ * sorted
+ */
+struct bset {
+	__u64			csum;
+	__u64			magic;
+	__u64			seq;
+	__u32			version;
+	__u32			keys;
+
+	union {
+		struct bkey	start[0];
+		__u64		d[0];
+	};
+};
+
+/* OBSOLETE */
+
+/* UUIDS - per backing device/flash only volume metadata */
+
+struct uuid_entry_v0 {
+	__u8		uuid[16];
+	__u8		label[32];
+	__u32		first_reg;
+	__u32		last_reg;
+	__u32		invalidated;
+	__u32		pad;
+};
+
+#endif /* _LINUX_BCACHE_H */
diff --git a/include/uapi/linux/genetlink.h b/include/uapi/linux/genetlink.h
index c880a417d8a9..1af72d8228e0 100644
--- a/include/uapi/linux/genetlink.h
+++ b/include/uapi/linux/genetlink.h
@@ -27,6 +27,7 @@ struct genlmsghdr {
  */
 #define GENL_ID_GENERATE	0
 #define GENL_ID_CTRL		NLMSG_MIN_TYPE
+#define GENL_ID_VFS_DQUOT	(NLMSG_MIN_TYPE + 1)
 
 /**************************************************************************
  * Controller
diff --git a/include/uapi/linux/kvm.h b/include/uapi/linux/kvm.h
index 99c25338ede8..902f12461873 100644
--- a/include/uapi/linux/kvm.h
+++ b/include/uapi/linux/kvm.h
@@ -518,6 +518,10 @@ struct kvm_ppc_smmu_info {
 /* machine type bits, to be used as argument to KVM_CREATE_VM */
 #define KVM_VM_S390_UCONTROL	1
 
+/* on ppc, 0 indicate default, 1 should force HV and 2 PR */
+#define KVM_VM_PPC_HV 1
+#define KVM_VM_PPC_PR 2
+
 #define KVM_S390_SIE_PAGE_OFFSET 1
 
 /*
@@ -541,6 +545,7 @@ struct kvm_ppc_smmu_info {
 #define KVM_TRACE_ENABLE          __KVM_DEPRECATED_MAIN_W_0x06
 #define KVM_TRACE_PAUSE           __KVM_DEPRECATED_MAIN_0x07
 #define KVM_TRACE_DISABLE         __KVM_DEPRECATED_MAIN_0x08
+#define KVM_GET_EMULATED_CPUID	  _IOWR(KVMIO, 0x09, struct kvm_cpuid2)
 
 /*
  * Extension capability list.
@@ -668,6 +673,7 @@ struct kvm_ppc_smmu_info {
 #define KVM_CAP_IRQ_XICS 92
 #define KVM_CAP_ARM_EL1_32BIT 93
 #define KVM_CAP_SPAPR_MULTITCE 94
+#define KVM_CAP_EXT_EMUL_CPUID 95
 
 #ifdef KVM_CAP_IRQ_ROUTING
 
@@ -843,6 +849,10 @@ struct kvm_device_attr {
 #define KVM_DEV_TYPE_FSL_MPIC_20	1
 #define KVM_DEV_TYPE_FSL_MPIC_42	2
 #define KVM_DEV_TYPE_XICS		3
+#define KVM_DEV_TYPE_VFIO		4
+#define  KVM_DEV_VFIO_GROUP			1
+#define   KVM_DEV_VFIO_GROUP_ADD			1
+#define   KVM_DEV_VFIO_GROUP_DEL			2
 
 /*
  * ioctls for VM fds
@@ -1012,6 +1022,7 @@ struct kvm_s390_ucas_mapping {
 /* VM is being stopped by host */
 #define KVM_KVMCLOCK_CTRL	  _IO(KVMIO,   0xad)
 #define KVM_ARM_VCPU_INIT	  _IOW(KVMIO,  0xae, struct kvm_vcpu_init)
+#define KVM_ARM_PREFERRED_TARGET  _IOR(KVMIO,  0xaf, struct kvm_vcpu_init)
 #define KVM_GET_REG_LIST	  _IOWR(KVMIO, 0xb0, struct kvm_reg_list)
 
 #define KVM_DEV_ASSIGN_ENABLE_IOMMU	(1 << 0)
diff --git a/include/uapi/linux/magic.h b/include/uapi/linux/magic.h
index 2944278a8ba7..77c60311a6c6 100644
--- a/include/uapi/linux/magic.h
+++ b/include/uapi/linux/magic.h
@@ -71,6 +71,6 @@
 #define USBDEVICE_SUPER_MAGIC	0x9fa2
 #define MTD_INODE_FS_MAGIC      0x11307854
 #define ANON_INODE_FS_MAGIC	0x09041934
-
+#define BTRFS_TEST_MAGIC	0x73727279
 
 #endif /* __LINUX_MAGIC_H__ */
diff --git a/include/uapi/linux/pkt_sched.h b/include/uapi/linux/pkt_sched.h
index 307f293477e8..a806687ad98f 100644
--- a/include/uapi/linux/pkt_sched.h
+++ b/include/uapi/linux/pkt_sched.h
@@ -763,13 +763,14 @@ enum {
 
 	TCA_FQ_RATE_ENABLE,	/* enable/disable rate limiting */
 
-	TCA_FQ_FLOW_DEFAULT_RATE,/* for sockets with unspecified sk_rate,
-				  * use the following rate
-				  */
+	TCA_FQ_FLOW_DEFAULT_RATE,/* obsolete, do not use */
 
 	TCA_FQ_FLOW_MAX_RATE,	/* per flow max rate */
 
 	TCA_FQ_BUCKETS_LOG,	/* log2(number of buckets) */
+
+	TCA_FQ_FLOW_REFILL_DELAY,	/* flow credit refill delay in usec */
+
 	__TCA_FQ_MAX
 };
 
diff --git a/include/uapi/linux/raid/md_p.h b/include/uapi/linux/raid/md_p.h
index fe1a5406d4d9..f7cf7f351144 100644
--- a/include/uapi/linux/raid/md_p.h
+++ b/include/uapi/linux/raid/md_p.h
@@ -16,6 +16,7 @@
 #define _MD_P_H
 
 #include <linux/types.h>
+#include <asm/byteorder.h>
 
 /*
  * RAID superblock.
diff --git a/include/uapi/linux/v4l2-controls.h b/include/uapi/linux/v4l2-controls.h
index 083bb5a5aae2..1666aabbbb86 100644
--- a/include/uapi/linux/v4l2-controls.h
+++ b/include/uapi/linux/v4l2-controls.h
@@ -160,6 +160,10 @@ enum v4l2_colorfx {
  * of controls. Total of 16 controls is reserved for this driver */
 #define V4L2_CID_USER_SI476X_BASE		(V4L2_CID_USER_BASE + 0x1040)
 
+/* The base for the TI VPE driver controls. Total of 16 controls is reserved for
+ * this driver */
+#define V4L2_CID_USER_TI_VPE_BASE		(V4L2_CID_USER_BASE + 0x1050)
+
 /* MPEG-class control IDs */
 /* The MPEG controls are applicable to all codec controls
  * and the 'MPEG' part of the define is historical */
diff --git a/include/uapi/rdma/ib_user_verbs.h b/include/uapi/rdma/ib_user_verbs.h
index e3ddd86c90a6..cbfdd4ca9510 100644
--- a/include/uapi/rdma/ib_user_verbs.h
+++ b/include/uapi/rdma/ib_user_verbs.h
@@ -87,10 +87,11 @@ enum {
 	IB_USER_VERBS_CMD_CLOSE_XRCD,
 	IB_USER_VERBS_CMD_CREATE_XSRQ,
 	IB_USER_VERBS_CMD_OPEN_QP,
-#ifdef CONFIG_INFINIBAND_EXPERIMENTAL_UVERBS_FLOW_STEERING
-	IB_USER_VERBS_CMD_CREATE_FLOW = IB_USER_VERBS_CMD_THRESHOLD,
-	IB_USER_VERBS_CMD_DESTROY_FLOW
-#endif /* CONFIG_INFINIBAND_EXPERIMENTAL_UVERBS_FLOW_STEERING */
+};
+
+enum {
+	IB_USER_VERBS_EX_CMD_CREATE_FLOW = IB_USER_VERBS_CMD_THRESHOLD,
+	IB_USER_VERBS_EX_CMD_DESTROY_FLOW
 };
 
 /*
@@ -122,22 +123,24 @@ struct ib_uverbs_comp_event_desc {
  * the rest of the command struct based on these value.
  */
 
+#define IB_USER_VERBS_CMD_COMMAND_MASK 0xff
+#define IB_USER_VERBS_CMD_FLAGS_MASK 0xff000000u
+#define IB_USER_VERBS_CMD_FLAGS_SHIFT 24
+
+#define IB_USER_VERBS_CMD_FLAG_EXTENDED 0x80
+
 struct ib_uverbs_cmd_hdr {
 	__u32 command;
 	__u16 in_words;
 	__u16 out_words;
 };
 
-#ifdef CONFIG_INFINIBAND_EXPERIMENTAL_UVERBS_FLOW_STEERING
-struct ib_uverbs_cmd_hdr_ex {
-	__u32 command;
-	__u16 in_words;
-	__u16 out_words;
+struct ib_uverbs_ex_cmd_hdr {
+	__u64 response;
 	__u16 provider_in_words;
 	__u16 provider_out_words;
 	__u32 cmd_hdr_reserved;
 };
-#endif /* CONFIG_INFINIBAND_EXPERIMENTAL_UVERBS_FLOW_STEERING */
 
 struct ib_uverbs_get_context {
 	__u64 response;
@@ -700,62 +703,71 @@ struct ib_uverbs_detach_mcast {
 	__u64 driver_data[0];
 };
 
-#ifdef CONFIG_INFINIBAND_EXPERIMENTAL_UVERBS_FLOW_STEERING
-struct ib_kern_eth_filter {
+struct ib_uverbs_flow_spec_hdr {
+	__u32 type;
+	__u16 size;
+	__u16 reserved;
+	/* followed by flow_spec */
+	__u64 flow_spec_data[0];
+};
+
+struct ib_uverbs_flow_eth_filter {
 	__u8  dst_mac[6];
 	__u8  src_mac[6];
 	__be16 ether_type;
 	__be16 vlan_tag;
 };
 
-struct ib_kern_spec_eth {
-	__u32  type;
-	__u16  size;
-	__u16  reserved;
-	struct ib_kern_eth_filter val;
-	struct ib_kern_eth_filter mask;
+struct ib_uverbs_flow_spec_eth {
+	union {
+		struct ib_uverbs_flow_spec_hdr hdr;
+		struct {
+			__u32 type;
+			__u16 size;
+			__u16 reserved;
+		};
+	};
+	struct ib_uverbs_flow_eth_filter val;
+	struct ib_uverbs_flow_eth_filter mask;
 };
 
-struct ib_kern_ipv4_filter {
+struct ib_uverbs_flow_ipv4_filter {
 	__be32 src_ip;
 	__be32 dst_ip;
 };
 
-struct ib_kern_spec_ipv4 {
-	__u32  type;
-	__u16  size;
-	__u16  reserved;
-	struct ib_kern_ipv4_filter val;
-	struct ib_kern_ipv4_filter mask;
+struct ib_uverbs_flow_spec_ipv4 {
+	union {
+		struct ib_uverbs_flow_spec_hdr hdr;
+		struct {
+			__u32 type;
+			__u16 size;
+			__u16 reserved;
+		};
+	};
+	struct ib_uverbs_flow_ipv4_filter val;
+	struct ib_uverbs_flow_ipv4_filter mask;
 };
 
-struct ib_kern_tcp_udp_filter {
+struct ib_uverbs_flow_tcp_udp_filter {
 	__be16 dst_port;
 	__be16 src_port;
 };
 
-struct ib_kern_spec_tcp_udp {
-	__u32  type;
-	__u16  size;
-	__u16  reserved;
-	struct ib_kern_tcp_udp_filter val;
-	struct ib_kern_tcp_udp_filter mask;
-};
-
-struct ib_kern_spec {
+struct ib_uverbs_flow_spec_tcp_udp {
 	union {
+		struct ib_uverbs_flow_spec_hdr hdr;
 		struct {
 			__u32 type;
 			__u16 size;
 			__u16 reserved;
 		};
-		struct ib_kern_spec_eth	    eth;
-		struct ib_kern_spec_ipv4    ipv4;
-		struct ib_kern_spec_tcp_udp tcp_udp;
 	};
+	struct ib_uverbs_flow_tcp_udp_filter val;
+	struct ib_uverbs_flow_tcp_udp_filter mask;
 };
 
-struct ib_kern_flow_attr {
+struct ib_uverbs_flow_attr {
 	__u32 type;
 	__u16 size;
 	__u16 priority;
@@ -767,13 +779,13 @@ struct ib_kern_flow_attr {
 	 * struct ib_flow_spec_xxx
 	 * struct ib_flow_spec_yyy
 	 */
+	struct ib_uverbs_flow_spec_hdr flow_specs[0];
 };
 
 struct ib_uverbs_create_flow  {
 	__u32 comp_mask;
-	__u64 response;
 	__u32 qp_handle;
-	struct ib_kern_flow_attr flow_attr;
+	struct ib_uverbs_flow_attr flow_attr;
 };
 
 struct ib_uverbs_create_flow_resp {
@@ -785,7 +797,6 @@ struct ib_uverbs_destroy_flow  {
 	__u32 comp_mask;
 	__u32 flow_handle;
 };
-#endif /* CONFIG_INFINIBAND_EXPERIMENTAL_UVERBS_FLOW_STEERING */
 
 struct ib_uverbs_create_srq {
 	__u64 response;
diff --git a/include/video/atmel_lcdc.h b/include/video/atmel_lcdc.h
index 0f5a2fc69af9..c79f38131926 100644
--- a/include/video/atmel_lcdc.h
+++ b/include/video/atmel_lcdc.h
@@ -31,39 +31,20 @@
 #define ATMEL_LCDC_WIRING_BGR	0
 #define ATMEL_LCDC_WIRING_RGB	1
 
-struct atmel_lcdfb_config;
 
  /* LCD Controller info data structure, stored in device platform_data */
-struct atmel_lcdfb_info {
-	spinlock_t		lock;
-	struct fb_info		*info;
-	void __iomem		*mmio;
-	int			irq_base;
-	struct work_struct	task;
-
+struct atmel_lcdfb_pdata {
 	unsigned int		guard_time;
-	unsigned int 		smem_len;
-	struct platform_device	*pdev;
-	struct clk		*bus_clk;
-	struct clk		*lcdc_clk;
-
-#ifdef CONFIG_BACKLIGHT_ATMEL_LCDC
-	struct backlight_device	*backlight;
-	u8			bl_power;
-#endif
 	bool			lcdcon_is_backlight;
 	bool			lcdcon_pol_negative;
-	u8			saved_lcdcon;
-
 	u8			default_bpp;
 	u8			lcd_wiring_mode;
 	unsigned int		default_lcdcon2;
 	unsigned int		default_dmacon;
-	void (*atmel_lcdfb_power_control)(int on);
+	void (*atmel_lcdfb_power_control)(struct atmel_lcdfb_pdata *pdata, int on);
 	struct fb_monspecs	*default_monspecs;
-	u32			pseudo_palette[16];
 
-	struct atmel_lcdfb_config *config;
+	struct list_head	pwr_gpios;
 };
 
 #define ATMEL_LCDC_DMABADDR1	0x00
diff --git a/include/video/mmp_disp.h b/include/video/mmp_disp.h
index b9dd1fbb0082..9fd9398368d5 100644
--- a/include/video/mmp_disp.h
+++ b/include/video/mmp_disp.h
@@ -91,6 +91,11 @@ struct mmp_win {
 	u16	up_crop;
 	u16	bottom_crop;
 	int	pix_fmt;
+	/*
+	 * pitch[0]: graphics/video layer line length or y pitch
+	 * pitch[1]/pitch[2]: video u/v pitch if non-zero
+	 */
+	u32	pitch[3];
 };
 
 struct mmp_addr {
@@ -334,6 +339,7 @@ struct mmp_mach_path_config {
 	int output_type;
 	u32 path_config;
 	u32 link_config;
+	u32 dsi_rbswap;
 };
 
 struct mmp_mach_plat_info {
diff --git a/include/video/omap-panel-data.h b/include/video/omap-panel-data.h
index f7ac8d972af0..69279c013ac4 100644
--- a/include/video/omap-panel-data.h
+++ b/include/video/omap-panel-data.h
@@ -238,4 +238,17 @@ struct panel_nec_nl8048hl11_platform_data {
 	int qvga_gpio;
 };
 
+/**
+ * panel-tpo-td028ttec1 platform data
+ * @name: name for display entity
+ * @source: name of the display entity used as a video source
+ * @data_lines: number of DPI datalines
+ */
+struct panel_tpo_td028ttec1_platform_data {
+	const char *name;
+	const char *source;
+
+	int data_lines;
+};
+
 #endif /* __OMAP_PANEL_DATA_H */
diff --git a/include/xen/interface/physdev.h b/include/xen/interface/physdev.h
index 7000bb1f6e96..42721d13a106 100644
--- a/include/xen/interface/physdev.h
+++ b/include/xen/interface/physdev.h
@@ -231,6 +231,17 @@ struct physdev_get_free_pirq {
 #define XEN_PCI_DEV_VIRTFN             0x2
 #define XEN_PCI_DEV_PXM                0x4
 
+#define XEN_PCI_MMCFG_RESERVED         0x1
+
+#define PHYSDEVOP_pci_mmcfg_reserved    24
+struct physdev_pci_mmcfg_reserved {
+    uint64_t address;
+    uint16_t segment;
+    uint8_t start_bus;
+    uint8_t end_bus;
+    uint32_t flags;
+};
+
 #define PHYSDEVOP_pci_device_add        25
 struct physdev_pci_device_add {
     /* IN */
diff --git a/include/xen/swiotlb-xen.h b/include/xen/swiotlb-xen.h
index de8bcc641c49..8b2eb93ae8ba 100644
--- a/include/xen/swiotlb-xen.h
+++ b/include/xen/swiotlb-xen.h
@@ -1,6 +1,7 @@
 #ifndef __LINUX_SWIOTLB_XEN_H
 #define __LINUX_SWIOTLB_XEN_H
 
+#include <linux/dma-direction.h>
 #include <linux/swiotlb.h>
 
 extern int xen_swiotlb_init(int verbose, bool early);
@@ -55,4 +56,6 @@ xen_swiotlb_dma_mapping_error(struct device *hwdev, dma_addr_t dma_addr);
 extern int
 xen_swiotlb_dma_supported(struct device *hwdev, u64 mask);
 
+extern int
+xen_swiotlb_set_dma_mask(struct device *dev, u64 dma_mask);
 #endif /* __LINUX_SWIOTLB_XEN_H */
diff --git a/include/xen/xen-ops.h b/include/xen/xen-ops.h
index d6fe062cad6b..fb2ea8f26552 100644
--- a/include/xen/xen-ops.h
+++ b/include/xen/xen-ops.h
@@ -19,10 +19,11 @@ void xen_arch_resume(void);
 int xen_setup_shutdown_event(void);
 
 extern unsigned long *xen_contiguous_bitmap;
-int xen_create_contiguous_region(unsigned long vstart, unsigned int order,
-				unsigned int address_bits);
+int xen_create_contiguous_region(phys_addr_t pstart, unsigned int order,
+				unsigned int address_bits,
+				dma_addr_t *dma_handle);
 
-void xen_destroy_contiguous_region(unsigned long vstart, unsigned int order);
+void xen_destroy_contiguous_region(phys_addr_t pstart, unsigned int order);
 
 struct vm_area_struct;
 int xen_remap_domain_mfn_range(struct vm_area_struct *vma,
diff --git a/init/Kconfig b/init/Kconfig
index 5496f307988e..3fc8a2f2fac4 100644
--- a/init/Kconfig
+++ b/init/Kconfig
@@ -118,6 +118,7 @@ config HAVE_KERNEL_LZ4
 choice
 	prompt "Kernel compression mode"
 	default KERNEL_GZIP
+	depends on HAVE_KERNEL_GZIP || HAVE_KERNEL_BZIP2 || HAVE_KERNEL_LZMA || HAVE_KERNEL_XZ || HAVE_KERNEL_LZO || HAVE_KERNEL_LZ4
 	help
 	  The linux kernel is a kind of self-extracting executable.
 	  Several compression algorithms are available, which differ
@@ -136,13 +137,6 @@ choice
 
 	  If in doubt, select 'gzip'
 
-config KERNEL_UNCOMPRESSED
-	bool "No compression"
-	help
-	  No compression at all. The kernel is huge but the compression and
-	  decompression times are zero.
-	  This is usually not what you want.
-
 config KERNEL_GZIP
 	bool "Gzip"
 	depends on HAVE_KERNEL_GZIP
@@ -851,7 +845,7 @@ config NUMA_BALANCING_DEFAULT_ENABLED
 	default y
 	depends on NUMA_BALANCING
 	help
-	  If set, autonumic NUMA balancing will be enabled if running on a NUMA
+	  If set, automatic NUMA balancing will be enabled if running on a NUMA
 	  machine.
 
 config NUMA_BALANCING
@@ -862,7 +856,7 @@ config NUMA_BALANCING
 	help
 	  This option adds support for automatic NUMA aware memory/task placement.
 	  The mechanism is quite primitive and is based on migrating memory when
-	  it is references to the node the task is running on.
+	  it has references to the node the task is running on.
 
 	  This system will be inactive on UMA systems.
 
diff --git a/init/main.c b/init/main.c
index 6ad1a533a8c7..01573fdfa186 100644
--- a/init/main.c
+++ b/init/main.c
@@ -131,6 +131,8 @@ char __initdata boot_command_line[COMMAND_LINE_SIZE];
 char *saved_command_line;
 /* Command line for parameter parsing */
 static char *static_command_line;
+/* Command line for per-initcall parameter parsing */
+static char *initcall_command_line;
 
 static char *execute_command;
 static char *ramdisk_execute_command;
@@ -354,6 +356,7 @@ static inline void smp_prepare_cpus(unsigned int maxcpus) { }
 static void __init setup_command_line(char *command_line)
 {
 	saved_command_line = alloc_bootmem(strlen (boot_command_line)+1);
+	initcall_command_line = alloc_bootmem(strlen (boot_command_line)+1);
 	static_command_line = alloc_bootmem(strlen (command_line)+1);
 	strcpy (saved_command_line, boot_command_line);
 	strcpy (static_command_line, command_line);
@@ -473,7 +476,7 @@ static void __init mm_init(void)
 	mem_init();
 	kmem_cache_init();
 	percpu_init_late();
-	pgtable_cache_init();
+	pgtable_init();
 	vmalloc_init();
 }
 
@@ -751,9 +754,9 @@ static void __init do_initcall_level(int level)
 	extern const struct kernel_param __start___param[], __stop___param[];
 	initcall_t *fn;
 
-	strcpy(static_command_line, saved_command_line);
+	strcpy(initcall_command_line, saved_command_line);
 	parse_args(initcall_level_names[level],
-		   static_command_line, __start___param,
+		   initcall_command_line, __start___param,
 		   __stop___param - __start___param,
 		   level, level,
 		   &repair_env_string);
diff --git a/kernel/Kconfig.hz b/kernel/Kconfig.hz
index 94fabd534b03..2a202a846757 100644
--- a/kernel/Kconfig.hz
+++ b/kernel/Kconfig.hz
@@ -55,4 +55,4 @@ config HZ
 	default 1000 if HZ_1000
 
 config SCHED_HRTICK
-	def_bool HIGH_RES_TIMERS && (!SMP || USE_GENERIC_SMP_HELPERS)
+	def_bool HIGH_RES_TIMERS
diff --git a/kernel/Makefile b/kernel/Makefile
index a4d1aa8da9bc..09a9c94f42bd 100644
--- a/kernel/Makefile
+++ b/kernel/Makefile
@@ -7,22 +7,19 @@ obj-y     = fork.o exec_domain.o panic.o \
 	    sysctl.o sysctl_binary.o capability.o ptrace.o timer.o user.o \
 	    signal.o sys.o kmod.o workqueue.o pid.o task_work.o \
 	    extable.o params.o posix-timers.o \
-	    kthread.o sys_ni.o posix-cpu-timers.o mutex.o \
-	    hrtimer.o rwsem.o nsproxy.o semaphore.o \
+	    kthread.o sys_ni.o posix-cpu-timers.o \
+	    hrtimer.o nsproxy.o \
 	    notifier.o ksysfs.o cred.o reboot.o \
-	    async.o range.o groups.o lglock.o smpboot.o
+	    async.o range.o groups.o smpboot.o
 
 ifdef CONFIG_FUNCTION_TRACER
 # Do not trace debug files and internal ftrace files
-CFLAGS_REMOVE_lockdep.o = -pg
-CFLAGS_REMOVE_lockdep_proc.o = -pg
-CFLAGS_REMOVE_mutex-debug.o = -pg
-CFLAGS_REMOVE_rtmutex-debug.o = -pg
 CFLAGS_REMOVE_cgroup-debug.o = -pg
 CFLAGS_REMOVE_irq_work.o = -pg
 endif
 
 obj-y += sched/
+obj-y += locking/
 obj-y += power/
 obj-y += printk/
 obj-y += cpu/
@@ -34,26 +31,15 @@ obj-$(CONFIG_FREEZER) += freezer.o
 obj-$(CONFIG_PROFILING) += profile.o
 obj-$(CONFIG_STACKTRACE) += stacktrace.o
 obj-y += time/
-obj-$(CONFIG_DEBUG_MUTEXES) += mutex-debug.o
-obj-$(CONFIG_LOCKDEP) += lockdep.o
-ifeq ($(CONFIG_PROC_FS),y)
-obj-$(CONFIG_LOCKDEP) += lockdep_proc.o
-endif
 obj-$(CONFIG_FUTEX) += futex.o
 ifeq ($(CONFIG_COMPAT),y)
 obj-$(CONFIG_FUTEX) += futex_compat.o
 endif
-obj-$(CONFIG_RT_MUTEXES) += rtmutex.o
-obj-$(CONFIG_DEBUG_RT_MUTEXES) += rtmutex-debug.o
-obj-$(CONFIG_RT_MUTEX_TESTER) += rtmutex-tester.o
 obj-$(CONFIG_GENERIC_ISA_DMA) += dma.o
 obj-$(CONFIG_SMP) += smp.o
 ifneq ($(CONFIG_SMP),y)
 obj-y += up.o
 endif
-obj-$(CONFIG_SMP) += spinlock.o
-obj-$(CONFIG_DEBUG_SPINLOCK) += spinlock.o
-obj-$(CONFIG_PROVE_LOCKING) += spinlock.o
 obj-$(CONFIG_UID16) += uid16.o
 obj-$(CONFIG_MODULES) += module.o
 obj-$(CONFIG_MODULE_SIG) += module_signing.o modsign_pubkey.o modsign_certificate.o
diff --git a/kernel/bounds.c b/kernel/bounds.c
index e8ca97b5c386..5253204afdca 100644
--- a/kernel/bounds.c
+++ b/kernel/bounds.c
@@ -11,6 +11,7 @@
 #include <linux/kbuild.h>
 #include <linux/page_cgroup.h>
 #include <linux/log2.h>
+#include <linux/spinlock_types.h>
 
 void foo(void)
 {
@@ -21,5 +22,6 @@ void foo(void)
 #ifdef CONFIG_SMP
 	DEFINE(NR_CPUS_BITS, ilog2(CONFIG_NR_CPUS));
 #endif
+	DEFINE(BLOATED_SPINLOCKS, sizeof(spinlock_t) > sizeof(int));
 	/* End of constants */
 }
diff --git a/kernel/cpu.c b/kernel/cpu.c
index 973d034acf84..deff2e693766 100644
--- a/kernel/cpu.c
+++ b/kernel/cpu.c
@@ -306,7 +306,6 @@ static int __ref _cpu_down(unsigned int cpu, int tasks_frozen)
 				__func__, cpu);
 		goto out_release;
 	}
-	smpboot_park_threads(cpu);
 
 	/*
 	 * By now we've cleared cpu_active_mask, wait for all preempt-disabled
@@ -315,12 +314,16 @@ static int __ref _cpu_down(unsigned int cpu, int tasks_frozen)
 	 *
 	 * For CONFIG_PREEMPT we have preemptible RCU and its sync_rcu() might
 	 * not imply sync_sched(), so explicitly call both.
+	 *
+	 * Do sync before park smpboot threads to take care the rcu boost case.
 	 */
 #ifdef CONFIG_PREEMPT
 	synchronize_sched();
 #endif
 	synchronize_rcu();
 
+	smpboot_park_threads(cpu);
+
 	/*
 	 * So now all preempt/rcu users must observe !cpu_active().
 	 */
diff --git a/kernel/fork.c b/kernel/fork.c
index f6d11fc67f72..728d5be9548c 100644
--- a/kernel/fork.c
+++ b/kernel/fork.c
@@ -532,7 +532,7 @@ static struct mm_struct *mm_init(struct mm_struct *mm, struct task_struct *p)
 	mm->flags = (current->mm) ?
 		(current->mm->flags & MMF_INIT_MASK) : default_dump_filter;
 	mm->core_state = NULL;
-	mm->nr_ptes = 0;
+	atomic_long_set(&mm->nr_ptes, 0);
 	memset(&mm->rss_stat, 0, sizeof(mm->rss_stat));
 	spin_lock_init(&mm->page_table_lock);
 	mm_init_aio(mm);
@@ -560,7 +560,7 @@ static void check_mm(struct mm_struct *mm)
 					  "mm:%p idx:%d val:%ld\n", mm, i, x);
 	}
 
-#ifdef CONFIG_TRANSPARENT_HUGEPAGE
+#if defined(CONFIG_TRANSPARENT_HUGEPAGE) && !USE_SPLIT_PMD_PTLOCKS
 	VM_BUG_ON(mm->pmd_huge_pte);
 #endif
 }
@@ -814,7 +814,7 @@ struct mm_struct *dup_mm(struct task_struct *tsk)
 	memcpy(mm, oldmm, sizeof(*mm));
 	mm_init_cpumask(mm);
 
-#ifdef CONFIG_TRANSPARENT_HUGEPAGE
+#if defined(CONFIG_TRANSPARENT_HUGEPAGE) && !USE_SPLIT_PMD_PTLOCKS
 	mm->pmd_huge_pte = NULL;
 #endif
 	if (!mm_init(mm, tsk))
diff --git a/kernel/futex.c b/kernel/futex.c
index c3a1a55a5214..80ba086f021d 100644
--- a/kernel/futex.c
+++ b/kernel/futex.c
@@ -66,7 +66,7 @@
 
 #include <asm/futex.h>
 
-#include "rtmutex_common.h"
+#include "locking/rtmutex_common.h"
 
 int __read_mostly futex_cmpxchg_enabled;
 
diff --git a/kernel/hung_task.c b/kernel/hung_task.c
index 3e97fb126e6b..9328b80eaf14 100644
--- a/kernel/hung_task.c
+++ b/kernel/hung_task.c
@@ -16,11 +16,12 @@
 #include <linux/export.h>
 #include <linux/sysctl.h>
 #include <linux/utsname.h>
+#include <trace/events/sched.h>
 
 /*
  * The number of tasks checked:
  */
-unsigned long __read_mostly sysctl_hung_task_check_count = PID_MAX_LIMIT;
+int __read_mostly sysctl_hung_task_check_count = PID_MAX_LIMIT;
 
 /*
  * Limit number of tasks checked in a batch.
@@ -92,6 +93,9 @@ static void check_hung_task(struct task_struct *t, unsigned long timeout)
 		t->last_switch_count = switch_count;
 		return;
 	}
+
+	trace_sched_process_hang(t);
+
 	if (!sysctl_hung_task_warnings)
 		return;
 	sysctl_hung_task_warnings--;
@@ -203,6 +207,14 @@ int proc_dohung_task_timeout_secs(struct ctl_table *table, int write,
 	return ret;
 }
 
+static atomic_t reset_hung_task = ATOMIC_INIT(0);
+
+void reset_hung_task_detector(void)
+{
+	atomic_set(&reset_hung_task, 1);
+}
+EXPORT_SYMBOL_GPL(reset_hung_task_detector);
+
 /*
  * kthread which checks for tasks stuck in D state
  */
@@ -216,6 +228,9 @@ static int watchdog(void *dummy)
 		while (schedule_timeout_interruptible(timeout_jiffies(timeout)))
 			timeout = sysctl_hung_task_timeout_secs;
 
+		if (atomic_xchg(&reset_hung_task, 0))
+			continue;
+
 		check_hung_uninterruptible_tasks(timeout);
 	}
 
diff --git a/kernel/irq/chip.c b/kernel/irq/chip.c
index a3bb14fbe5c6..dc04c166c54d 100644
--- a/kernel/irq/chip.c
+++ b/kernel/irq/chip.c
@@ -214,7 +214,7 @@ void irq_enable(struct irq_desc *desc)
 }
 
 /**
- * irq_disable - Mark interupt disabled
+ * irq_disable - Mark interrupt disabled
  * @desc:	irq descriptor which should be disabled
  *
  * If the chip does not implement the irq_disable callback, we
diff --git a/kernel/irq/manage.c b/kernel/irq/manage.c
index 3e59f951d42f..481a13c43b17 100644
--- a/kernel/irq/manage.c
+++ b/kernel/irq/manage.c
@@ -786,7 +786,7 @@ irq_forced_thread_fn(struct irq_desc *desc, struct irqaction *action)
 }
 
 /*
- * Interrupts explicitely requested as threaded interupts want to be
+ * Interrupts explicitly requested as threaded interrupts want to be
  * preemtible - many of them need to sleep and wait for slow busses to
  * complete.
  */
diff --git a/kernel/irq/settings.h b/kernel/irq/settings.h
index 1162f1030f18..3320b84cc60f 100644
--- a/kernel/irq/settings.h
+++ b/kernel/irq/settings.h
@@ -14,6 +14,7 @@ enum {
 	_IRQ_NO_BALANCING	= IRQ_NO_BALANCING,
 	_IRQ_NESTED_THREAD	= IRQ_NESTED_THREAD,
 	_IRQ_PER_CPU_DEVID	= IRQ_PER_CPU_DEVID,
+	_IRQ_IS_POLLED		= IRQ_IS_POLLED,
 	_IRQF_MODIFY_MASK	= IRQF_MODIFY_MASK,
 };
 
@@ -26,6 +27,7 @@ enum {
 #define IRQ_NOAUTOEN		GOT_YOU_MORON
 #define IRQ_NESTED_THREAD	GOT_YOU_MORON
 #define IRQ_PER_CPU_DEVID	GOT_YOU_MORON
+#define IRQ_IS_POLLED		GOT_YOU_MORON
 #undef IRQF_MODIFY_MASK
 #define IRQF_MODIFY_MASK	GOT_YOU_MORON
 
@@ -147,3 +149,8 @@ static inline bool irq_settings_is_nested_thread(struct irq_desc *desc)
 {
 	return desc->status_use_accessors & _IRQ_NESTED_THREAD;
 }
+
+static inline bool irq_settings_is_polled(struct irq_desc *desc)
+{
+	return desc->status_use_accessors & _IRQ_IS_POLLED;
+}
diff --git a/kernel/irq/spurious.c b/kernel/irq/spurious.c
index 7b5f012bde9d..a1d8cc63b56e 100644
--- a/kernel/irq/spurious.c
+++ b/kernel/irq/spurious.c
@@ -67,8 +67,13 @@ static int try_one_irq(int irq, struct irq_desc *desc, bool force)
 
 	raw_spin_lock(&desc->lock);
 
-	/* PER_CPU and nested thread interrupts are never polled */
-	if (irq_settings_is_per_cpu(desc) || irq_settings_is_nested_thread(desc))
+	/*
+	 * PER_CPU, nested thread interrupts and interrupts explicitely
+	 * marked polled are excluded from polling.
+	 */
+	if (irq_settings_is_per_cpu(desc) ||
+	    irq_settings_is_nested_thread(desc) ||
+	    irq_settings_is_polled(desc))
 		goto out;
 
 	/*
@@ -268,7 +273,8 @@ try_misrouted_irq(unsigned int irq, struct irq_desc *desc,
 void note_interrupt(unsigned int irq, struct irq_desc *desc,
 		    irqreturn_t action_ret)
 {
-	if (desc->istate & IRQS_POLL_INPROGRESS)
+	if (desc->istate & IRQS_POLL_INPROGRESS ||
+	    irq_settings_is_polled(desc))
 		return;
 
 	/* we get here again via the threaded handler */
diff --git a/kernel/kexec.c b/kernel/kexec.c
index 2a74f307c5ec..490afc03627e 100644
--- a/kernel/kexec.c
+++ b/kernel/kexec.c
@@ -921,7 +921,7 @@ static int kimage_load_segment(struct kimage *image,
  *   reinitialize them.
  *
  * - A machine specific part that includes the syscall number
- *   and the copies the image to it's final destination.  And
+ *   and then copies the image to it's final destination.  And
  *   jumps into the image at entry.
  *
  * kexec does not sync, or unmount filesystems so if you need
diff --git a/kernel/locking/Makefile b/kernel/locking/Makefile
new file mode 100644
index 000000000000..baab8e5e7f66
--- /dev/null
+++ b/kernel/locking/Makefile
@@ -0,0 +1,25 @@
+
+obj-y += mutex.o semaphore.o rwsem.o lglock.o
+
+ifdef CONFIG_FUNCTION_TRACER
+CFLAGS_REMOVE_lockdep.o = -pg
+CFLAGS_REMOVE_lockdep_proc.o = -pg
+CFLAGS_REMOVE_mutex-debug.o = -pg
+CFLAGS_REMOVE_rtmutex-debug.o = -pg
+endif
+
+obj-$(CONFIG_DEBUG_MUTEXES) += mutex-debug.o
+obj-$(CONFIG_LOCKDEP) += lockdep.o
+ifeq ($(CONFIG_PROC_FS),y)
+obj-$(CONFIG_LOCKDEP) += lockdep_proc.o
+endif
+obj-$(CONFIG_SMP) += spinlock.o
+obj-$(CONFIG_PROVE_LOCKING) += spinlock.o
+obj-$(CONFIG_RT_MUTEXES) += rtmutex.o
+obj-$(CONFIG_DEBUG_RT_MUTEXES) += rtmutex-debug.o
+obj-$(CONFIG_RT_MUTEX_TESTER) += rtmutex-tester.o
+obj-$(CONFIG_DEBUG_SPINLOCK) += spinlock.o
+obj-$(CONFIG_DEBUG_SPINLOCK) += spinlock_debug.o
+obj-$(CONFIG_RWSEM_GENERIC_SPINLOCK) += rwsem-spinlock.o
+obj-$(CONFIG_RWSEM_XCHGADD_ALGORITHM) += rwsem-xadd.o
+obj-$(CONFIG_PERCPU_RWSEM) += percpu-rwsem.o
diff --git a/kernel/lglock.c b/kernel/locking/lglock.c
index 86ae2aebf004..86ae2aebf004 100644
--- a/kernel/lglock.c
+++ b/kernel/locking/lglock.c
diff --git a/kernel/lockdep.c b/kernel/locking/lockdep.c
index 4e8e14c34e42..576ba756a32d 100644
--- a/kernel/lockdep.c
+++ b/kernel/locking/lockdep.c
@@ -1232,7 +1232,7 @@ static int noop_count(struct lock_list *entry, void *data)
 	return 0;
 }
 
-unsigned long __lockdep_count_forward_deps(struct lock_list *this)
+static unsigned long __lockdep_count_forward_deps(struct lock_list *this)
 {
 	unsigned long  count = 0;
 	struct lock_list *uninitialized_var(target_entry);
@@ -1258,7 +1258,7 @@ unsigned long lockdep_count_forward_deps(struct lock_class *class)
 	return ret;
 }
 
-unsigned long __lockdep_count_backward_deps(struct lock_list *this)
+static unsigned long __lockdep_count_backward_deps(struct lock_list *this)
 {
 	unsigned long  count = 0;
 	struct lock_list *uninitialized_var(target_entry);
diff --git a/kernel/lockdep_internals.h b/kernel/locking/lockdep_internals.h
index 4f560cfedc8f..4f560cfedc8f 100644
--- a/kernel/lockdep_internals.h
+++ b/kernel/locking/lockdep_internals.h
diff --git a/kernel/lockdep_proc.c b/kernel/locking/lockdep_proc.c
index b2c71c5873e4..ef43ac4bafb5 100644
--- a/kernel/lockdep_proc.c
+++ b/kernel/locking/lockdep_proc.c
@@ -421,6 +421,7 @@ static void seq_lock_time(struct seq_file *m, struct lock_time *lt)
 	seq_time(m, lt->min);
 	seq_time(m, lt->max);
 	seq_time(m, lt->total);
+	seq_time(m, lt->nr ? div_s64(lt->total, lt->nr) : 0);
 }
 
 static void seq_stats(struct seq_file *m, struct lock_stat_data *data)
@@ -518,20 +519,20 @@ static void seq_stats(struct seq_file *m, struct lock_stat_data *data)
 	}
 	if (i) {
 		seq_puts(m, "\n");
-		seq_line(m, '.', 0, 40 + 1 + 10 * (14 + 1));
+		seq_line(m, '.', 0, 40 + 1 + 12 * (14 + 1));
 		seq_puts(m, "\n");
 	}
 }
 
 static void seq_header(struct seq_file *m)
 {
-	seq_printf(m, "lock_stat version 0.3\n");
+	seq_puts(m, "lock_stat version 0.4\n");
 
 	if (unlikely(!debug_locks))
 		seq_printf(m, "*WARNING* lock debugging disabled!! - possibly due to a lockdep warning\n");
 
-	seq_line(m, '-', 0, 40 + 1 + 10 * (14 + 1));
-	seq_printf(m, "%40s %14s %14s %14s %14s %14s %14s %14s %14s "
+	seq_line(m, '-', 0, 40 + 1 + 12 * (14 + 1));
+	seq_printf(m, "%40s %14s %14s %14s %14s %14s %14s %14s %14s %14s %14s "
 			"%14s %14s\n",
 			"class name",
 			"con-bounces",
@@ -539,12 +540,14 @@ static void seq_header(struct seq_file *m)
 			"waittime-min",
 			"waittime-max",
 			"waittime-total",
+			"waittime-avg",
 			"acq-bounces",
 			"acquisitions",
 			"holdtime-min",
 			"holdtime-max",
-			"holdtime-total");
-	seq_line(m, '-', 0, 40 + 1 + 10 * (14 + 1));
+			"holdtime-total",
+			"holdtime-avg");
+	seq_line(m, '-', 0, 40 + 1 + 12 * (14 + 1));
 	seq_printf(m, "\n");
 }
 
diff --git a/kernel/lockdep_states.h b/kernel/locking/lockdep_states.h
index 995b0cc2b84c..995b0cc2b84c 100644
--- a/kernel/lockdep_states.h
+++ b/kernel/locking/lockdep_states.h
diff --git a/kernel/mutex-debug.c b/kernel/locking/mutex-debug.c
index 7e3443fe1f48..7e3443fe1f48 100644
--- a/kernel/mutex-debug.c
+++ b/kernel/locking/mutex-debug.c
diff --git a/kernel/mutex-debug.h b/kernel/locking/mutex-debug.h
index 0799fd3e4cfa..0799fd3e4cfa 100644
--- a/kernel/mutex-debug.h
+++ b/kernel/locking/mutex-debug.h
diff --git a/kernel/mutex.c b/kernel/locking/mutex.c
index d24105b1b794..4dd6e4c219de 100644
--- a/kernel/mutex.c
+++ b/kernel/locking/mutex.c
@@ -1,5 +1,5 @@
 /*
- * kernel/mutex.c
+ * kernel/locking/mutex.c
  *
  * Mutexes: blocking mutual exclusion locks
  *
diff --git a/kernel/mutex.h b/kernel/locking/mutex.h
index 4115fbf83b12..4115fbf83b12 100644
--- a/kernel/mutex.h
+++ b/kernel/locking/mutex.h
diff --git a/lib/percpu-rwsem.c b/kernel/locking/percpu-rwsem.c
index 652a8ee8efe9..652a8ee8efe9 100644
--- a/lib/percpu-rwsem.c
+++ b/kernel/locking/percpu-rwsem.c
diff --git a/kernel/rtmutex-debug.c b/kernel/locking/rtmutex-debug.c
index 13b243a323fa..13b243a323fa 100644
--- a/kernel/rtmutex-debug.c
+++ b/kernel/locking/rtmutex-debug.c
diff --git a/kernel/rtmutex-debug.h b/kernel/locking/rtmutex-debug.h
index 14193d596d78..14193d596d78 100644
--- a/kernel/rtmutex-debug.h
+++ b/kernel/locking/rtmutex-debug.h
diff --git a/kernel/rtmutex-tester.c b/kernel/locking/rtmutex-tester.c
index 1d96dd0d93c1..1d96dd0d93c1 100644
--- a/kernel/rtmutex-tester.c
+++ b/kernel/locking/rtmutex-tester.c
diff --git a/kernel/rtmutex.c b/kernel/locking/rtmutex.c
index 0dd6aec1cb6a..0dd6aec1cb6a 100644
--- a/kernel/rtmutex.c
+++ b/kernel/locking/rtmutex.c
diff --git a/kernel/rtmutex.h b/kernel/locking/rtmutex.h
index a1a1dd06421d..a1a1dd06421d 100644
--- a/kernel/rtmutex.h
+++ b/kernel/locking/rtmutex.h
diff --git a/kernel/rtmutex_common.h b/kernel/locking/rtmutex_common.h
index 53a66c85261b..53a66c85261b 100644
--- a/kernel/rtmutex_common.h
+++ b/kernel/locking/rtmutex_common.h
diff --git a/lib/rwsem-spinlock.c b/kernel/locking/rwsem-spinlock.c
index 9be8a9144978..9be8a9144978 100644
--- a/lib/rwsem-spinlock.c
+++ b/kernel/locking/rwsem-spinlock.c
diff --git a/lib/rwsem.c b/kernel/locking/rwsem-xadd.c
index 19c5fa95e0b4..19c5fa95e0b4 100644
--- a/lib/rwsem.c
+++ b/kernel/locking/rwsem-xadd.c
diff --git a/kernel/rwsem.c b/kernel/locking/rwsem.c
index cfff1435bdfb..cfff1435bdfb 100644
--- a/kernel/rwsem.c
+++ b/kernel/locking/rwsem.c
diff --git a/kernel/semaphore.c b/kernel/locking/semaphore.c
index 6815171a4fff..6815171a4fff 100644
--- a/kernel/semaphore.c
+++ b/kernel/locking/semaphore.c
diff --git a/kernel/spinlock.c b/kernel/locking/spinlock.c
index 4b082b5cac9e..4b082b5cac9e 100644
--- a/kernel/spinlock.c
+++ b/kernel/locking/spinlock.c
diff --git a/lib/spinlock_debug.c b/kernel/locking/spinlock_debug.c
index 0374a596cffa..0374a596cffa 100644
--- a/lib/spinlock_debug.c
+++ b/kernel/locking/spinlock_debug.c
diff --git a/kernel/module.c b/kernel/module.c
index af5ebd21d77b..f5a3b1e8ec51 100644
--- a/kernel/module.c
+++ b/kernel/module.c
@@ -641,8 +641,6 @@ static int module_unload_init(struct module *mod)
 
 	/* Hold reference count during initialization. */
 	__this_cpu_write(mod->refptr->incs, 1);
-	/* Backwards compatibility macros put refcount during init. */
-	mod->waiter = current;
 
 	return 0;
 }
@@ -768,16 +766,9 @@ static int __try_stop_module(void *_sref)
 
 static int try_stop_module(struct module *mod, int flags, int *forced)
 {
-	if (flags & O_NONBLOCK) {
-		struct stopref sref = { mod, flags, forced };
+	struct stopref sref = { mod, flags, forced };
 
-		return stop_machine(__try_stop_module, &sref, NULL);
-	} else {
-		/* We don't need to stop the machine for this. */
-		mod->state = MODULE_STATE_GOING;
-		synchronize_sched();
-		return 0;
-	}
+	return stop_machine(__try_stop_module, &sref, NULL);
 }
 
 unsigned long module_refcount(struct module *mod)
@@ -810,21 +801,6 @@ EXPORT_SYMBOL(module_refcount);
 /* This exists whether we can unload or not */
 static void free_module(struct module *mod);
 
-static void wait_for_zero_refcount(struct module *mod)
-{
-	/* Since we might sleep for some time, release the mutex first */
-	mutex_unlock(&module_mutex);
-	for (;;) {
-		pr_debug("Looking at refcount...\n");
-		set_current_state(TASK_UNINTERRUPTIBLE);
-		if (module_refcount(mod) == 0)
-			break;
-		schedule();
-	}
-	current->state = TASK_RUNNING;
-	mutex_lock(&module_mutex);
-}
-
 SYSCALL_DEFINE2(delete_module, const char __user *, name_user,
 		unsigned int, flags)
 {
@@ -839,6 +815,11 @@ SYSCALL_DEFINE2(delete_module, const char __user *, name_user,
 		return -EFAULT;
 	name[MODULE_NAME_LEN-1] = '\0';
 
+	if (!(flags & O_NONBLOCK)) {
+		printk(KERN_WARNING
+		       "waiting module removal not supported: please upgrade");
+	}
+
 	if (mutex_lock_interruptible(&module_mutex) != 0)
 		return -EINTR;
 
@@ -856,8 +837,7 @@ SYSCALL_DEFINE2(delete_module, const char __user *, name_user,
 
 	/* Doing init or already dying? */
 	if (mod->state != MODULE_STATE_LIVE) {
-		/* FIXME: if (force), slam module count and wake up
-                   waiter --RR */
+		/* FIXME: if (force), slam module count damn the torpedoes */
 		pr_debug("%s already dying\n", mod->name);
 		ret = -EBUSY;
 		goto out;
@@ -873,18 +853,11 @@ SYSCALL_DEFINE2(delete_module, const char __user *, name_user,
 		}
 	}
 
-	/* Set this up before setting mod->state */
-	mod->waiter = current;
-
 	/* Stop the machine so refcounts can't move and disable module. */
 	ret = try_stop_module(mod, flags, &forced);
 	if (ret != 0)
 		goto out;
 
-	/* Never wait if forced. */
-	if (!forced && module_refcount(mod) != 0)
-		wait_for_zero_refcount(mod);
-
 	mutex_unlock(&module_mutex);
 	/* Final destruction now no one is using it. */
 	if (mod->exit != NULL)
@@ -1002,9 +975,6 @@ void module_put(struct module *module)
 		__this_cpu_inc(module->refptr->decs);
 
 		trace_module_put(module, _RET_IP_);
-		/* Maybe they're waiting for us to drop reference? */
-		if (unlikely(!module_is_live(module)))
-			wake_up_process(module->waiter);
 		preempt_enable();
 	}
 }
@@ -2728,7 +2698,7 @@ static int check_modinfo(struct module *mod, struct load_info *info, int flags)
 	return 0;
 }
 
-static void find_module_sections(struct module *mod, struct load_info *info)
+static int find_module_sections(struct module *mod, struct load_info *info)
 {
 	mod->kp = section_objs(info, "__param",
 			       sizeof(*mod->kp), &mod->num_kp);
@@ -2758,6 +2728,18 @@ static void find_module_sections(struct module *mod, struct load_info *info)
 #ifdef CONFIG_CONSTRUCTORS
 	mod->ctors = section_objs(info, ".ctors",
 				  sizeof(*mod->ctors), &mod->num_ctors);
+	if (!mod->ctors)
+		mod->ctors = section_objs(info, ".init_array",
+				sizeof(*mod->ctors), &mod->num_ctors);
+	else if (find_sec(info, ".init_array")) {
+		/*
+		 * This shouldn't happen with same compiler and binutils
+		 * building all parts of the module.
+		 */
+		printk(KERN_WARNING "%s: has both .ctors and .init_array.\n",
+		       mod->name);
+		return -EINVAL;
+	}
 #endif
 
 #ifdef CONFIG_TRACEPOINTS
@@ -2795,6 +2777,8 @@ static void find_module_sections(struct module *mod, struct load_info *info)
 
 	info->debug = section_objs(info, "__verbose",
 				   sizeof(*info->debug), &info->num_debug);
+
+	return 0;
 }
 
 static int move_module(struct module *mod, struct load_info *info)
@@ -3248,7 +3232,9 @@ static int load_module(struct load_info *info, const char __user *uargs,
 
 	/* Now we've got everything in the final locations, we can
 	 * find optional sections. */
-	find_module_sections(mod, info);
+	err = find_module_sections(mod, info);
+	if (err)
+		goto free_unload;
 
 	err = check_module_license_and_versions(mod);
 	if (err)
diff --git a/kernel/rcu/tiny.c b/kernel/rcu/tiny.c
index 0c9a934cfec1..1254f312d024 100644
--- a/kernel/rcu/tiny.c
+++ b/kernel/rcu/tiny.c
@@ -181,7 +181,7 @@ EXPORT_SYMBOL_GPL(rcu_irq_enter);
 /*
  * Test whether RCU thinks that the current CPU is idle.
  */
-bool __rcu_is_watching(void)
+bool notrace __rcu_is_watching(void)
 {
 	return rcu_dynticks_nesting;
 }
diff --git a/kernel/rcu/tree.c b/kernel/rcu/tree.c
index 4c06ddfea7cd..dd081987a8ec 100644
--- a/kernel/rcu/tree.c
+++ b/kernel/rcu/tree.c
@@ -664,7 +664,7 @@ void rcu_nmi_exit(void)
  * rcu_is_watching(), the caller of __rcu_is_watching() must have at
  * least disabled preemption.
  */
-bool __rcu_is_watching(void)
+bool notrace __rcu_is_watching(void)
 {
 	return atomic_read(this_cpu_ptr(&rcu_dynticks.dynticks)) & 0x1;
 }
@@ -675,7 +675,7 @@ bool __rcu_is_watching(void)
  * If the current CPU is in its idle loop and is neither in an interrupt
  * or NMI handler, return true.
  */
-bool rcu_is_watching(void)
+bool notrace rcu_is_watching(void)
 {
 	int ret;
 
diff --git a/kernel/rcu/tree_plugin.h b/kernel/rcu/tree_plugin.h
index 3822ac0c4b27..6abb03dff5c0 100644
--- a/kernel/rcu/tree_plugin.h
+++ b/kernel/rcu/tree_plugin.h
@@ -1133,7 +1133,7 @@ void exit_rcu(void)
 
 #ifdef CONFIG_RCU_BOOST
 
-#include "../rtmutex_common.h"
+#include "../locking/rtmutex_common.h"
 
 #ifdef CONFIG_RCU_TRACE
 
diff --git a/kernel/sched/core.c b/kernel/sched/core.c
index 1deccd78be98..c1808606ee5f 100644
--- a/kernel/sched/core.c
+++ b/kernel/sched/core.c
@@ -2253,6 +2253,20 @@ unsigned long long task_sched_runtime(struct task_struct *p)
 	struct rq *rq;
 	u64 ns = 0;
 
+#if defined(CONFIG_64BIT) && defined(CONFIG_SMP)
+	/*
+	 * 64-bit doesn't need locks to atomically read a 64bit value.
+	 * So we have a optimization chance when the task's delta_exec is 0.
+	 * Reading ->on_cpu is racy, but this is ok.
+	 *
+	 * If we race with it leaving cpu, we'll take a lock. So we're correct.
+	 * If we race with it entering cpu, unaccounted time is 0. This is
+	 * indistinguishable from the read occurring a few cycles earlier.
+	 */
+	if (!p->on_cpu)
+		return p->se.sum_exec_runtime;
+#endif
+
 	rq = task_rq_lock(p, &flags);
 	ns = p->se.sum_exec_runtime + do_task_delta_exec(p, rq);
 	task_rq_unlock(rq, p, &flags);
diff --git a/kernel/sched/fair.c b/kernel/sched/fair.c
index df77c605c7a6..e8b652ebe027 100644
--- a/kernel/sched/fair.c
+++ b/kernel/sched/fair.c
@@ -1000,7 +1000,7 @@ struct numa_stats {
  */
 static void update_numa_stats(struct numa_stats *ns, int nid)
 {
-	int cpu;
+	int cpu, cpus = 0;
 
 	memset(ns, 0, sizeof(*ns));
 	for_each_cpu(cpu, cpumask_of_node(nid)) {
@@ -1009,8 +1009,21 @@ static void update_numa_stats(struct numa_stats *ns, int nid)
 		ns->nr_running += rq->nr_running;
 		ns->load += weighted_cpuload(cpu);
 		ns->power += power_of(cpu);
+
+		cpus++;
 	}
 
+	/*
+	 * If we raced with hotplug and there are no CPUs left in our mask
+	 * the @ns structure is NULL'ed and task_numa_compare() will
+	 * not find this node attractive.
+	 *
+	 * We'll either bail at !has_capacity, or we'll detect a huge imbalance
+	 * and bail there.
+	 */
+	if (!cpus)
+		return;
+
 	ns->load = (ns->load * SCHED_POWER_SCALE) / ns->power;
 	ns->capacity = DIV_ROUND_CLOSEST(ns->power, SCHED_POWER_SCALE);
 	ns->has_capacity = (ns->nr_running < ns->capacity);
@@ -1201,9 +1214,21 @@ static int task_numa_migrate(struct task_struct *p)
 	 */
 	rcu_read_lock();
 	sd = rcu_dereference(per_cpu(sd_numa, env.src_cpu));
-	env.imbalance_pct = 100 + (sd->imbalance_pct - 100) / 2;
+	if (sd)
+		env.imbalance_pct = 100 + (sd->imbalance_pct - 100) / 2;
 	rcu_read_unlock();
 
+	/*
+	 * Cpusets can break the scheduler domain tree into smaller
+	 * balance domains, some of which do not cross NUMA boundaries.
+	 * Tasks that are "trapped" in such domains cannot be migrated
+	 * elsewhere, so there is no point in (re)trying.
+	 */
+	if (unlikely(!sd)) {
+		p->numa_preferred_nid = cpu_to_node(task_cpu(p));
+		return -EINVAL;
+	}
+
 	taskweight = task_weight(p, env.src_nid);
 	groupweight = group_weight(p, env.src_nid);
 	update_numa_stats(&env.src_stats, env.src_nid);
@@ -2153,7 +2178,7 @@ static inline void __update_tg_runnable_avg(struct sched_avg *sa,
 	long contrib;
 
 	/* The fraction of a cpu used by this cfs_rq */
-	contrib = div_u64(sa->runnable_avg_sum << NICE_0_SHIFT,
+	contrib = div_u64((u64)sa->runnable_avg_sum << NICE_0_SHIFT,
 			  sa->runnable_avg_period + 1);
 	contrib -= cfs_rq->tg_runnable_contrib;
 
diff --git a/kernel/smp.c b/kernel/smp.c
index 46116100f0ee..bd9f94028838 100644
--- a/kernel/smp.c
+++ b/kernel/smp.c
@@ -15,7 +15,6 @@
 
 #include "smpboot.h"
 
-#ifdef CONFIG_USE_GENERIC_SMP_HELPERS
 enum {
 	CSD_FLAG_LOCK		= 0x01,
 	CSD_FLAG_WAIT		= 0x02,
@@ -140,8 +139,7 @@ static void csd_unlock(struct call_single_data *csd)
  * for execution on the given CPU. data must already have
  * ->func, ->info, and ->flags set.
  */
-static
-void generic_exec_single(int cpu, struct call_single_data *csd, int wait)
+static void generic_exec_single(int cpu, struct call_single_data *csd, int wait)
 {
 	struct call_single_queue *dst = &per_cpu(call_single_queue, cpu);
 	unsigned long flags;
@@ -464,7 +462,6 @@ int smp_call_function(smp_call_func_t func, void *info, int wait)
 	return 0;
 }
 EXPORT_SYMBOL(smp_call_function);
-#endif /* USE_GENERIC_SMP_HELPERS */
 
 /* Setup configured maximum number of CPUs to activate */
 unsigned int setup_max_cpus = NR_CPUS;
diff --git a/kernel/softirq.c b/kernel/softirq.c
index b24988353458..11025ccc06dd 100644
--- a/kernel/softirq.c
+++ b/kernel/softirq.c
@@ -6,8 +6,6 @@
  *	Distribute under GPLv2.
  *
  *	Rewritten. Old one was good in 2.2, but in 2.3 it was immoral. --ANK (990903)
- *
- *	Remote softirq infrastructure is by Jens Axboe.
  */
 
 #include <linux/export.h>
@@ -627,146 +625,17 @@ void tasklet_hrtimer_init(struct tasklet_hrtimer *ttimer,
 }
 EXPORT_SYMBOL_GPL(tasklet_hrtimer_init);
 
-/*
- * Remote softirq bits
- */
-
-DEFINE_PER_CPU(struct list_head [NR_SOFTIRQS], softirq_work_list);
-EXPORT_PER_CPU_SYMBOL(softirq_work_list);
-
-static void __local_trigger(struct call_single_data *cp, int softirq)
-{
-	struct list_head *head = &__get_cpu_var(softirq_work_list[softirq]);
-
-	list_add_tail(&cp->list, head);
-
-	/* Trigger the softirq only if the list was previously empty.  */
-	if (head->next == &cp->list)
-		raise_softirq_irqoff(softirq);
-}
-
-#ifdef CONFIG_USE_GENERIC_SMP_HELPERS
-static void remote_softirq_receive(void *data)
-{
-	struct call_single_data *cp = data;
-	unsigned long flags;
-	int softirq;
-
-	softirq = *(int *)cp->info;
-	local_irq_save(flags);
-	__local_trigger(cp, softirq);
-	local_irq_restore(flags);
-}
-
-static int __try_remote_softirq(struct call_single_data *cp, int cpu, int softirq)
-{
-	if (cpu_online(cpu)) {
-		cp->func = remote_softirq_receive;
-		cp->info = &softirq;
-		cp->flags = 0;
-
-		__smp_call_function_single(cpu, cp, 0);
-		return 0;
-	}
-	return 1;
-}
-#else /* CONFIG_USE_GENERIC_SMP_HELPERS */
-static int __try_remote_softirq(struct call_single_data *cp, int cpu, int softirq)
-{
-	return 1;
-}
-#endif
-
-/**
- * __send_remote_softirq - try to schedule softirq work on a remote cpu
- * @cp: private SMP call function data area
- * @cpu: the remote cpu
- * @this_cpu: the currently executing cpu
- * @softirq: the softirq for the work
- *
- * Attempt to schedule softirq work on a remote cpu.  If this cannot be
- * done, the work is instead queued up on the local cpu.
- *
- * Interrupts must be disabled.
- */
-void __send_remote_softirq(struct call_single_data *cp, int cpu, int this_cpu, int softirq)
-{
-	if (cpu == this_cpu || __try_remote_softirq(cp, cpu, softirq))
-		__local_trigger(cp, softirq);
-}
-EXPORT_SYMBOL(__send_remote_softirq);
-
-/**
- * send_remote_softirq - try to schedule softirq work on a remote cpu
- * @cp: private SMP call function data area
- * @cpu: the remote cpu
- * @softirq: the softirq for the work
- *
- * Like __send_remote_softirq except that disabling interrupts and
- * computing the current cpu is done for the caller.
- */
-void send_remote_softirq(struct call_single_data *cp, int cpu, int softirq)
-{
-	unsigned long flags;
-	int this_cpu;
-
-	local_irq_save(flags);
-	this_cpu = smp_processor_id();
-	__send_remote_softirq(cp, cpu, this_cpu, softirq);
-	local_irq_restore(flags);
-}
-EXPORT_SYMBOL(send_remote_softirq);
-
-static int remote_softirq_cpu_notify(struct notifier_block *self,
-					       unsigned long action, void *hcpu)
-{
-	/*
-	 * If a CPU goes away, splice its entries to the current CPU
-	 * and trigger a run of the softirq
-	 */
-	if (action == CPU_DEAD || action == CPU_DEAD_FROZEN) {
-		int cpu = (unsigned long) hcpu;
-		int i;
-
-		local_irq_disable();
-		for (i = 0; i < NR_SOFTIRQS; i++) {
-			struct list_head *head = &per_cpu(softirq_work_list[i], cpu);
-			struct list_head *local_head;
-
-			if (list_empty(head))
-				continue;
-
-			local_head = &__get_cpu_var(softirq_work_list[i]);
-			list_splice_init(head, local_head);
-			raise_softirq_irqoff(i);
-		}
-		local_irq_enable();
-	}
-
-	return NOTIFY_OK;
-}
-
-static struct notifier_block remote_softirq_cpu_notifier = {
-	.notifier_call	= remote_softirq_cpu_notify,
-};
-
 void __init softirq_init(void)
 {
 	int cpu;
 
 	for_each_possible_cpu(cpu) {
-		int i;
-
 		per_cpu(tasklet_vec, cpu).tail =
 			&per_cpu(tasklet_vec, cpu).head;
 		per_cpu(tasklet_hi_vec, cpu).tail =
 			&per_cpu(tasklet_hi_vec, cpu).head;
-		for (i = 0; i < NR_SOFTIRQS; i++)
-			INIT_LIST_HEAD(&per_cpu(softirq_work_list[i], cpu));
 	}
 
-	register_hotcpu_notifier(&remote_softirq_cpu_notifier);
-
 	open_softirq(TASKLET_SOFTIRQ, tasklet_action);
 	open_softirq(HI_SOFTIRQ, tasklet_hi_action);
 }
diff --git a/kernel/sysctl.c b/kernel/sysctl.c
index d37d9dd8f463..34a604726d0b 100644
--- a/kernel/sysctl.c
+++ b/kernel/sysctl.c
@@ -969,9 +969,10 @@ static struct ctl_table kern_table[] = {
 	{
 		.procname	= "hung_task_check_count",
 		.data		= &sysctl_hung_task_check_count,
-		.maxlen		= sizeof(unsigned long),
+		.maxlen		= sizeof(int),
 		.mode		= 0644,
-		.proc_handler	= proc_doulongvec_minmax,
+		.proc_handler	= proc_dointvec_minmax,
+		.extra1		= &zero,
 	},
 	{
 		.procname	= "hung_task_timeout_secs",
diff --git a/kernel/taskstats.c b/kernel/taskstats.c
index 9f4618eb51c8..13d2f7cd65db 100644
--- a/kernel/taskstats.c
+++ b/kernel/taskstats.c
@@ -673,17 +673,18 @@ err:
 	nlmsg_free(rep_skb);
 }
 
-static struct genl_ops taskstats_ops = {
-	.cmd		= TASKSTATS_CMD_GET,
-	.doit		= taskstats_user_cmd,
-	.policy		= taskstats_cmd_get_policy,
-	.flags		= GENL_ADMIN_PERM,
-};
-
-static struct genl_ops cgroupstats_ops = {
-	.cmd		= CGROUPSTATS_CMD_GET,
-	.doit		= cgroupstats_user_cmd,
-	.policy		= cgroupstats_cmd_get_policy,
+static const struct genl_ops taskstats_ops[] = {
+	{
+		.cmd		= TASKSTATS_CMD_GET,
+		.doit		= taskstats_user_cmd,
+		.policy		= taskstats_cmd_get_policy,
+		.flags		= GENL_ADMIN_PERM,
+	},
+	{
+		.cmd		= CGROUPSTATS_CMD_GET,
+		.doit		= cgroupstats_user_cmd,
+		.policy		= cgroupstats_cmd_get_policy,
+	},
 };
 
 /* Needed early in initialization */
@@ -702,26 +703,13 @@ static int __init taskstats_init(void)
 {
 	int rc;
 
-	rc = genl_register_family(&family);
+	rc = genl_register_family_with_ops(&family, taskstats_ops);
 	if (rc)
 		return rc;
 
-	rc = genl_register_ops(&family, &taskstats_ops);
-	if (rc < 0)
-		goto err;
-
-	rc = genl_register_ops(&family, &cgroupstats_ops);
-	if (rc < 0)
-		goto err_cgroup_ops;
-
 	family_registered = 1;
 	pr_info("registered taskstats version %d\n", TASKSTATS_GENL_VERSION);
 	return 0;
-err_cgroup_ops:
-	genl_unregister_ops(&family, &taskstats_ops);
-err:
-	genl_unregister_family(&family);
-	return rc;
 }
 
 /*
diff --git a/kernel/trace/ftrace.c b/kernel/trace/ftrace.c
index 03cf44ac54d3..22fa55696760 100644
--- a/kernel/trace/ftrace.c
+++ b/kernel/trace/ftrace.c
@@ -3307,7 +3307,11 @@ void unregister_ftrace_function_probe_all(char *glob)
 static LIST_HEAD(ftrace_commands);
 static DEFINE_MUTEX(ftrace_cmd_mutex);
 
-int register_ftrace_command(struct ftrace_func_command *cmd)
+/*
+ * Currently we only register ftrace commands from __init, so mark this
+ * __init too.
+ */
+__init int register_ftrace_command(struct ftrace_func_command *cmd)
 {
 	struct ftrace_func_command *p;
 	int ret = 0;
@@ -3326,7 +3330,11 @@ int register_ftrace_command(struct ftrace_func_command *cmd)
 	return ret;
 }
 
-int unregister_ftrace_command(struct ftrace_func_command *cmd)
+/*
+ * Currently we only unregister ftrace commands from __init, so mark
+ * this __init too.
+ */
+__init int unregister_ftrace_command(struct ftrace_func_command *cmd)
 {
 	struct ftrace_func_command *p, *n;
 	int ret = -ENODEV;
@@ -3641,7 +3649,7 @@ __setup("ftrace_filter=", set_ftrace_filter);
 
 #ifdef CONFIG_FUNCTION_GRAPH_TRACER
 static char ftrace_graph_buf[FTRACE_FILTER_SIZE] __initdata;
-static int ftrace_set_func(unsigned long *array, int *idx, char *buffer);
+static int ftrace_set_func(unsigned long *array, int *idx, int size, char *buffer);
 
 static int __init set_graph_function(char *str)
 {
@@ -3659,7 +3667,7 @@ static void __init set_ftrace_early_graph(char *buf)
 		func = strsep(&buf, ",");
 		/* we allow only one expression at a time */
 		ret = ftrace_set_func(ftrace_graph_funcs, &ftrace_graph_count,
-				      func);
+				      FTRACE_GRAPH_MAX_FUNCS, func);
 		if (ret)
 			printk(KERN_DEBUG "ftrace: function %s not "
 					  "traceable\n", func);
@@ -3776,15 +3784,25 @@ static const struct file_operations ftrace_notrace_fops = {
 static DEFINE_MUTEX(graph_lock);
 
 int ftrace_graph_count;
-int ftrace_graph_filter_enabled;
+int ftrace_graph_notrace_count;
 unsigned long ftrace_graph_funcs[FTRACE_GRAPH_MAX_FUNCS] __read_mostly;
+unsigned long ftrace_graph_notrace_funcs[FTRACE_GRAPH_MAX_FUNCS] __read_mostly;
+
+struct ftrace_graph_data {
+	unsigned long *table;
+	size_t size;
+	int *count;
+	const struct seq_operations *seq_ops;
+};
 
 static void *
 __g_next(struct seq_file *m, loff_t *pos)
 {
-	if (*pos >= ftrace_graph_count)
+	struct ftrace_graph_data *fgd = m->private;
+
+	if (*pos >= *fgd->count)
 		return NULL;
-	return &ftrace_graph_funcs[*pos];
+	return &fgd->table[*pos];
 }
 
 static void *
@@ -3796,10 +3814,12 @@ g_next(struct seq_file *m, void *v, loff_t *pos)
 
 static void *g_start(struct seq_file *m, loff_t *pos)
 {
+	struct ftrace_graph_data *fgd = m->private;
+
 	mutex_lock(&graph_lock);
 
 	/* Nothing, tell g_show to print all functions are enabled */
-	if (!ftrace_graph_filter_enabled && !*pos)
+	if (!*fgd->count && !*pos)
 		return (void *)1;
 
 	return __g_next(m, pos);
@@ -3835,38 +3855,88 @@ static const struct seq_operations ftrace_graph_seq_ops = {
 };
 
 static int
-ftrace_graph_open(struct inode *inode, struct file *file)
+__ftrace_graph_open(struct inode *inode, struct file *file,
+		    struct ftrace_graph_data *fgd)
 {
 	int ret = 0;
 
-	if (unlikely(ftrace_disabled))
-		return -ENODEV;
-
 	mutex_lock(&graph_lock);
 	if ((file->f_mode & FMODE_WRITE) &&
 	    (file->f_flags & O_TRUNC)) {
-		ftrace_graph_filter_enabled = 0;
-		ftrace_graph_count = 0;
-		memset(ftrace_graph_funcs, 0, sizeof(ftrace_graph_funcs));
+		*fgd->count = 0;
+		memset(fgd->table, 0, fgd->size * sizeof(*fgd->table));
 	}
 	mutex_unlock(&graph_lock);
 
-	if (file->f_mode & FMODE_READ)
-		ret = seq_open(file, &ftrace_graph_seq_ops);
+	if (file->f_mode & FMODE_READ) {
+		ret = seq_open(file, fgd->seq_ops);
+		if (!ret) {
+			struct seq_file *m = file->private_data;
+			m->private = fgd;
+		}
+	} else
+		file->private_data = fgd;
 
 	return ret;
 }
 
 static int
+ftrace_graph_open(struct inode *inode, struct file *file)
+{
+	struct ftrace_graph_data *fgd;
+
+	if (unlikely(ftrace_disabled))
+		return -ENODEV;
+
+	fgd = kmalloc(sizeof(*fgd), GFP_KERNEL);
+	if (fgd == NULL)
+		return -ENOMEM;
+
+	fgd->table = ftrace_graph_funcs;
+	fgd->size = FTRACE_GRAPH_MAX_FUNCS;
+	fgd->count = &ftrace_graph_count;
+	fgd->seq_ops = &ftrace_graph_seq_ops;
+
+	return __ftrace_graph_open(inode, file, fgd);
+}
+
+static int
+ftrace_graph_notrace_open(struct inode *inode, struct file *file)
+{
+	struct ftrace_graph_data *fgd;
+
+	if (unlikely(ftrace_disabled))
+		return -ENODEV;
+
+	fgd = kmalloc(sizeof(*fgd), GFP_KERNEL);
+	if (fgd == NULL)
+		return -ENOMEM;
+
+	fgd->table = ftrace_graph_notrace_funcs;
+	fgd->size = FTRACE_GRAPH_MAX_FUNCS;
+	fgd->count = &ftrace_graph_notrace_count;
+	fgd->seq_ops = &ftrace_graph_seq_ops;
+
+	return __ftrace_graph_open(inode, file, fgd);
+}
+
+static int
 ftrace_graph_release(struct inode *inode, struct file *file)
 {
-	if (file->f_mode & FMODE_READ)
+	if (file->f_mode & FMODE_READ) {
+		struct seq_file *m = file->private_data;
+
+		kfree(m->private);
 		seq_release(inode, file);
+	} else {
+		kfree(file->private_data);
+	}
+
 	return 0;
 }
 
 static int
-ftrace_set_func(unsigned long *array, int *idx, char *buffer)
+ftrace_set_func(unsigned long *array, int *idx, int size, char *buffer)
 {
 	struct dyn_ftrace *rec;
 	struct ftrace_page *pg;
@@ -3879,7 +3949,7 @@ ftrace_set_func(unsigned long *array, int *idx, char *buffer)
 
 	/* decode regex */
 	type = filter_parse_regex(buffer, strlen(buffer), &search, &not);
-	if (!not && *idx >= FTRACE_GRAPH_MAX_FUNCS)
+	if (!not && *idx >= size)
 		return -EBUSY;
 
 	search_len = strlen(search);
@@ -3907,7 +3977,7 @@ ftrace_set_func(unsigned long *array, int *idx, char *buffer)
 				fail = 0;
 				if (!exists) {
 					array[(*idx)++] = rec->ip;
-					if (*idx >= FTRACE_GRAPH_MAX_FUNCS)
+					if (*idx >= size)
 						goto out;
 				}
 			} else {
@@ -3925,8 +3995,6 @@ out:
 	if (fail)
 		return -EINVAL;
 
-	ftrace_graph_filter_enabled = !!(*idx);
-
 	return 0;
 }
 
@@ -3935,36 +4003,33 @@ ftrace_graph_write(struct file *file, const char __user *ubuf,
 		   size_t cnt, loff_t *ppos)
 {
 	struct trace_parser parser;
-	ssize_t read, ret;
+	ssize_t read, ret = 0;
+	struct ftrace_graph_data *fgd = file->private_data;
 
 	if (!cnt)
 		return 0;
 
-	mutex_lock(&graph_lock);
-
-	if (trace_parser_get_init(&parser, FTRACE_BUFF_MAX)) {
-		ret = -ENOMEM;
-		goto out_unlock;
-	}
+	if (trace_parser_get_init(&parser, FTRACE_BUFF_MAX))
+		return -ENOMEM;
 
 	read = trace_get_user(&parser, ubuf, cnt, ppos);
 
 	if (read >= 0 && trace_parser_loaded((&parser))) {
 		parser.buffer[parser.idx] = 0;
 
+		mutex_lock(&graph_lock);
+
 		/* we allow only one expression at a time */
-		ret = ftrace_set_func(ftrace_graph_funcs, &ftrace_graph_count,
-					parser.buffer);
-		if (ret)
-			goto out_free;
+		ret = ftrace_set_func(fgd->table, fgd->count, fgd->size,
+				      parser.buffer);
+
+		mutex_unlock(&graph_lock);
 	}
 
-	ret = read;
+	if (!ret)
+		ret = read;
 
-out_free:
 	trace_parser_put(&parser);
-out_unlock:
-	mutex_unlock(&graph_lock);
 
 	return ret;
 }
@@ -3976,6 +4041,14 @@ static const struct file_operations ftrace_graph_fops = {
 	.llseek		= ftrace_filter_lseek,
 	.release	= ftrace_graph_release,
 };
+
+static const struct file_operations ftrace_graph_notrace_fops = {
+	.open		= ftrace_graph_notrace_open,
+	.read		= seq_read,
+	.write		= ftrace_graph_write,
+	.llseek		= ftrace_filter_lseek,
+	.release	= ftrace_graph_release,
+};
 #endif /* CONFIG_FUNCTION_GRAPH_TRACER */
 
 static __init int ftrace_init_dyn_debugfs(struct dentry *d_tracer)
@@ -3997,6 +4070,9 @@ static __init int ftrace_init_dyn_debugfs(struct dentry *d_tracer)
 	trace_create_file("set_graph_function", 0444, d_tracer,
 				    NULL,
 				    &ftrace_graph_fops);
+	trace_create_file("set_graph_notrace", 0444, d_tracer,
+				    NULL,
+				    &ftrace_graph_notrace_fops);
 #endif /* CONFIG_FUNCTION_GRAPH_TRACER */
 
 	return 0;
@@ -4320,12 +4396,21 @@ ftrace_ops_control_func(unsigned long ip, unsigned long parent_ip,
 	 */
 	preempt_disable_notrace();
 	trace_recursion_set(TRACE_CONTROL_BIT);
+
+	/*
+	 * Control funcs (perf) uses RCU. Only trace if
+	 * RCU is currently active.
+	 */
+	if (!rcu_is_watching())
+		goto out;
+
 	do_for_each_ftrace_op(op, ftrace_control_list) {
 		if (!(op->flags & FTRACE_OPS_FL_STUB) &&
 		    !ftrace_function_local_disabled(op) &&
 		    ftrace_ops_test(op, ip, regs))
 			op->func(ip, parent_ip, op, regs);
 	} while_for_each_ftrace_op(op);
+ out:
 	trace_recursion_clear(TRACE_CONTROL_BIT);
 	preempt_enable_notrace();
 }
diff --git a/kernel/trace/trace.c b/kernel/trace/trace.c
index d9fea7dfd5d3..9d20cd9743ef 100644
--- a/kernel/trace/trace.c
+++ b/kernel/trace/trace.c
@@ -235,13 +235,33 @@ void trace_array_put(struct trace_array *this_tr)
 	mutex_unlock(&trace_types_lock);
 }
 
-int filter_current_check_discard(struct ring_buffer *buffer,
-				 struct ftrace_event_call *call, void *rec,
-				 struct ring_buffer_event *event)
+int filter_check_discard(struct ftrace_event_file *file, void *rec,
+			 struct ring_buffer *buffer,
+			 struct ring_buffer_event *event)
 {
-	return filter_check_discard(call, rec, buffer, event);
+	if (unlikely(file->flags & FTRACE_EVENT_FL_FILTERED) &&
+	    !filter_match_preds(file->filter, rec)) {
+		ring_buffer_discard_commit(buffer, event);
+		return 1;
+	}
+
+	return 0;
+}
+EXPORT_SYMBOL_GPL(filter_check_discard);
+
+int call_filter_check_discard(struct ftrace_event_call *call, void *rec,
+			      struct ring_buffer *buffer,
+			      struct ring_buffer_event *event)
+{
+	if (unlikely(call->flags & TRACE_EVENT_FL_FILTERED) &&
+	    !filter_match_preds(call->filter, rec)) {
+		ring_buffer_discard_commit(buffer, event);
+		return 1;
+	}
+
+	return 0;
 }
-EXPORT_SYMBOL_GPL(filter_current_check_discard);
+EXPORT_SYMBOL_GPL(call_filter_check_discard);
 
 cycle_t buffer_ftrace_now(struct trace_buffer *buf, int cpu)
 {
@@ -843,9 +863,12 @@ int trace_get_user(struct trace_parser *parser, const char __user *ubuf,
 	if (isspace(ch)) {
 		parser->buffer[parser->idx] = 0;
 		parser->cont = false;
-	} else {
+	} else if (parser->idx < parser->size - 1) {
 		parser->cont = true;
 		parser->buffer[parser->idx++] = ch;
+	} else {
+		ret = -EINVAL;
+		goto out;
 	}
 
 	*ppos += read;
@@ -1261,21 +1284,6 @@ int is_tracing_stopped(void)
 }
 
 /**
- * ftrace_off_permanent - disable all ftrace code permanently
- *
- * This should only be called when a serious anomally has
- * been detected.  This will turn off the function tracing,
- * ring buffers, and other tracing utilites. It takes no
- * locks and can be called from any context.
- */
-void ftrace_off_permanent(void)
-{
-	tracing_disabled = 1;
-	ftrace_stop();
-	tracing_off_permanent();
-}
-
-/**
  * tracing_start - quick start of the tracer
  *
  * If tracing is enabled but was stopped by tracing_stop,
@@ -1631,7 +1639,7 @@ trace_function(struct trace_array *tr,
 	entry->ip			= ip;
 	entry->parent_ip		= parent_ip;
 
-	if (!filter_check_discard(call, entry, buffer, event))
+	if (!call_filter_check_discard(call, entry, buffer, event))
 		__buffer_unlock_commit(buffer, event);
 }
 
@@ -1715,7 +1723,7 @@ static void __ftrace_trace_stack(struct ring_buffer *buffer,
 
 	entry->size = trace.nr_entries;
 
-	if (!filter_check_discard(call, entry, buffer, event))
+	if (!call_filter_check_discard(call, entry, buffer, event))
 		__buffer_unlock_commit(buffer, event);
 
  out:
@@ -1817,7 +1825,7 @@ ftrace_trace_userstack(struct ring_buffer *buffer, unsigned long flags, int pc)
 	trace.entries		= entry->caller;
 
 	save_stack_trace_user(&trace);
-	if (!filter_check_discard(call, entry, buffer, event))
+	if (!call_filter_check_discard(call, entry, buffer, event))
 		__buffer_unlock_commit(buffer, event);
 
  out_drop_count:
@@ -2009,7 +2017,7 @@ int trace_vbprintk(unsigned long ip, const char *fmt, va_list args)
 	entry->fmt			= fmt;
 
 	memcpy(entry->buf, tbuffer, sizeof(u32) * len);
-	if (!filter_check_discard(call, entry, buffer, event)) {
+	if (!call_filter_check_discard(call, entry, buffer, event)) {
 		__buffer_unlock_commit(buffer, event);
 		ftrace_trace_stack(buffer, flags, 6, pc);
 	}
@@ -2064,7 +2072,7 @@ __trace_array_vprintk(struct ring_buffer *buffer,
 
 	memcpy(&entry->buf, tbuffer, len);
 	entry->buf[len] = '\0';
-	if (!filter_check_discard(call, entry, buffer, event)) {
+	if (!call_filter_check_discard(call, entry, buffer, event)) {
 		__buffer_unlock_commit(buffer, event);
 		ftrace_trace_stack(buffer, flags, 6, pc);
 	}
@@ -2761,7 +2769,7 @@ static void show_snapshot_main_help(struct seq_file *m)
 	seq_printf(m, "# echo 0 > snapshot : Clears and frees snapshot buffer\n");
 	seq_printf(m, "# echo 1 > snapshot : Allocates snapshot buffer, if not already allocated.\n");
 	seq_printf(m, "#                      Takes a snapshot of the main buffer.\n");
-	seq_printf(m, "# echo 2 > snapshot : Clears snapshot buffer (but does not allocate)\n");
+	seq_printf(m, "# echo 2 > snapshot : Clears snapshot buffer (but does not allocate or free)\n");
 	seq_printf(m, "#                      (Doesn't have to be '2' works with any number that\n");
 	seq_printf(m, "#                       is not a '0' or '1')\n");
 }
@@ -2965,6 +2973,11 @@ int tracing_open_generic(struct inode *inode, struct file *filp)
 	return 0;
 }
 
+bool tracing_is_disabled(void)
+{
+	return (tracing_disabled) ? true: false;
+}
+
 /*
  * Open and update trace_array ref count.
  * Must have the current trace_array passed to it.
@@ -5455,12 +5468,12 @@ static struct ftrace_func_command ftrace_snapshot_cmd = {
 	.func			= ftrace_trace_snapshot_callback,
 };
 
-static int register_snapshot_cmd(void)
+static __init int register_snapshot_cmd(void)
 {
 	return register_ftrace_command(&ftrace_snapshot_cmd);
 }
 #else
-static inline int register_snapshot_cmd(void) { return 0; }
+static inline __init int register_snapshot_cmd(void) { return 0; }
 #endif /* defined(CONFIG_TRACER_SNAPSHOT) && defined(CONFIG_DYNAMIC_FTRACE) */
 
 struct dentry *tracing_init_dentry_tr(struct trace_array *tr)
@@ -6254,6 +6267,17 @@ void trace_init_global_iter(struct trace_iterator *iter)
 	iter->trace = iter->tr->current_trace;
 	iter->cpu_file = RING_BUFFER_ALL_CPUS;
 	iter->trace_buffer = &global_trace.trace_buffer;
+
+	if (iter->trace && iter->trace->open)
+		iter->trace->open(iter);
+
+	/* Annotate start of buffers if we had overruns */
+	if (ring_buffer_overruns(iter->trace_buffer->buffer))
+		iter->iter_flags |= TRACE_FILE_ANNOTATE;
+
+	/* Output in nanoseconds only if we are using a clock in nanoseconds. */
+	if (trace_clocks[iter->tr->clock_id].in_ns)
+		iter->iter_flags |= TRACE_FILE_TIME_IN_NS;
 }
 
 void ftrace_dump(enum ftrace_dump_mode oops_dump_mode)
diff --git a/kernel/trace/trace.h b/kernel/trace/trace.h
index 73d08aa25b55..ea189e027b80 100644
--- a/kernel/trace/trace.h
+++ b/kernel/trace/trace.h
@@ -193,8 +193,8 @@ struct trace_array {
 #ifdef CONFIG_FTRACE_SYSCALLS
 	int			sys_refcount_enter;
 	int			sys_refcount_exit;
-	DECLARE_BITMAP(enabled_enter_syscalls, NR_syscalls);
-	DECLARE_BITMAP(enabled_exit_syscalls, NR_syscalls);
+	struct ftrace_event_file __rcu *enter_syscall_files[NR_syscalls];
+	struct ftrace_event_file __rcu *exit_syscall_files[NR_syscalls];
 #endif
 	int			stop_count;
 	int			clock_id;
@@ -515,6 +515,7 @@ void tracing_reset_online_cpus(struct trace_buffer *buf);
 void tracing_reset_current(int cpu);
 void tracing_reset_all_online_cpus(void);
 int tracing_open_generic(struct inode *inode, struct file *filp);
+bool tracing_is_disabled(void);
 struct dentry *trace_create_file(const char *name,
 				 umode_t mode,
 				 struct dentry *parent,
@@ -712,6 +713,8 @@ extern unsigned long trace_flags;
 #define TRACE_GRAPH_PRINT_PROC          0x8
 #define TRACE_GRAPH_PRINT_DURATION      0x10
 #define TRACE_GRAPH_PRINT_ABS_TIME      0x20
+#define TRACE_GRAPH_PRINT_FILL_SHIFT	28
+#define TRACE_GRAPH_PRINT_FILL_MASK	(0x3 << TRACE_GRAPH_PRINT_FILL_SHIFT)
 
 extern enum print_line_t
 print_graph_function_flags(struct trace_iterator *iter, u32 flags);
@@ -731,15 +734,16 @@ extern void __trace_graph_return(struct trace_array *tr,
 #ifdef CONFIG_DYNAMIC_FTRACE
 /* TODO: make this variable */
 #define FTRACE_GRAPH_MAX_FUNCS		32
-extern int ftrace_graph_filter_enabled;
 extern int ftrace_graph_count;
 extern unsigned long ftrace_graph_funcs[FTRACE_GRAPH_MAX_FUNCS];
+extern int ftrace_graph_notrace_count;
+extern unsigned long ftrace_graph_notrace_funcs[FTRACE_GRAPH_MAX_FUNCS];
 
 static inline int ftrace_graph_addr(unsigned long addr)
 {
 	int i;
 
-	if (!ftrace_graph_filter_enabled)
+	if (!ftrace_graph_count)
 		return 1;
 
 	for (i = 0; i < ftrace_graph_count; i++) {
@@ -759,11 +763,31 @@ static inline int ftrace_graph_addr(unsigned long addr)
 
 	return 0;
 }
+
+static inline int ftrace_graph_notrace_addr(unsigned long addr)
+{
+	int i;
+
+	if (!ftrace_graph_notrace_count)
+		return 0;
+
+	for (i = 0; i < ftrace_graph_notrace_count; i++) {
+		if (addr == ftrace_graph_notrace_funcs[i])
+			return 1;
+	}
+
+	return 0;
+}
 #else
 static inline int ftrace_graph_addr(unsigned long addr)
 {
 	return 1;
 }
+
+static inline int ftrace_graph_notrace_addr(unsigned long addr)
+{
+	return 0;
+}
 #endif /* CONFIG_DYNAMIC_FTRACE */
 #else /* CONFIG_FUNCTION_GRAPH_TRACER */
 static inline enum print_line_t
@@ -987,9 +1011,9 @@ struct filter_pred {
 
 extern enum regex_type
 filter_parse_regex(char *buff, int len, char **search, int *not);
-extern void print_event_filter(struct ftrace_event_call *call,
+extern void print_event_filter(struct ftrace_event_file *file,
 			       struct trace_seq *s);
-extern int apply_event_filter(struct ftrace_event_call *call,
+extern int apply_event_filter(struct ftrace_event_file *file,
 			      char *filter_string);
 extern int apply_subsystem_event_filter(struct ftrace_subsystem_dir *dir,
 					char *filter_string);
@@ -1000,20 +1024,6 @@ extern int filter_assign_type(const char *type);
 struct ftrace_event_field *
 trace_find_event_field(struct ftrace_event_call *call, char *name);
 
-static inline int
-filter_check_discard(struct ftrace_event_call *call, void *rec,
-		     struct ring_buffer *buffer,
-		     struct ring_buffer_event *event)
-{
-	if (unlikely(call->flags & TRACE_EVENT_FL_FILTERED) &&
-	    !filter_match_preds(call->filter, rec)) {
-		ring_buffer_discard_commit(buffer, event);
-		return 1;
-	}
-
-	return 0;
-}
-
 extern void trace_event_enable_cmd_record(bool enable);
 extern int event_trace_add_tracer(struct dentry *parent, struct trace_array *tr);
 extern int event_trace_del_tracer(struct trace_array *tr);
diff --git a/kernel/trace/trace_branch.c b/kernel/trace/trace_branch.c
index d594da0dc03c..697fb9bac8f0 100644
--- a/kernel/trace/trace_branch.c
+++ b/kernel/trace/trace_branch.c
@@ -78,7 +78,7 @@ probe_likely_condition(struct ftrace_branch_data *f, int val, int expect)
 	entry->line = f->line;
 	entry->correct = val == expect;
 
-	if (!filter_check_discard(call, entry, buffer, event))
+	if (!call_filter_check_discard(call, entry, buffer, event))
 		__buffer_unlock_commit(buffer, event);
 
  out:
diff --git a/kernel/trace/trace_events.c b/kernel/trace/trace_events.c
index 368a4d50cc30..f919a2e21bf3 100644
--- a/kernel/trace/trace_events.c
+++ b/kernel/trace/trace_events.c
@@ -989,7 +989,7 @@ static ssize_t
 event_filter_read(struct file *filp, char __user *ubuf, size_t cnt,
 		  loff_t *ppos)
 {
-	struct ftrace_event_call *call;
+	struct ftrace_event_file *file;
 	struct trace_seq *s;
 	int r = -ENODEV;
 
@@ -1004,12 +1004,12 @@ event_filter_read(struct file *filp, char __user *ubuf, size_t cnt,
 	trace_seq_init(s);
 
 	mutex_lock(&event_mutex);
-	call = event_file_data(filp);
-	if (call)
-		print_event_filter(call, s);
+	file = event_file_data(filp);
+	if (file)
+		print_event_filter(file, s);
 	mutex_unlock(&event_mutex);
 
-	if (call)
+	if (file)
 		r = simple_read_from_buffer(ubuf, cnt, ppos, s->buffer, s->len);
 
 	kfree(s);
@@ -1021,7 +1021,7 @@ static ssize_t
 event_filter_write(struct file *filp, const char __user *ubuf, size_t cnt,
 		   loff_t *ppos)
 {
-	struct ftrace_event_call *call;
+	struct ftrace_event_file *file;
 	char *buf;
 	int err = -ENODEV;
 
@@ -1039,9 +1039,9 @@ event_filter_write(struct file *filp, const char __user *ubuf, size_t cnt,
 	buf[cnt] = '\0';
 
 	mutex_lock(&event_mutex);
-	call = event_file_data(filp);
-	if (call)
-		err = apply_event_filter(call, buf);
+	file = event_file_data(filp);
+	if (file)
+		err = apply_event_filter(file, buf);
 	mutex_unlock(&event_mutex);
 
 	free_page((unsigned long) buf);
@@ -1062,6 +1062,9 @@ static int subsystem_open(struct inode *inode, struct file *filp)
 	struct trace_array *tr;
 	int ret;
 
+	if (tracing_is_disabled())
+		return -ENODEV;
+
 	/* Make sure the system still exists */
 	mutex_lock(&trace_types_lock);
 	mutex_lock(&event_mutex);
@@ -1108,6 +1111,9 @@ static int system_tr_open(struct inode *inode, struct file *filp)
 	struct trace_array *tr = inode->i_private;
 	int ret;
 
+	if (tracing_is_disabled())
+		return -ENODEV;
+
 	if (trace_array_get(tr) < 0)
 		return -ENODEV;
 
@@ -1124,11 +1130,12 @@ static int system_tr_open(struct inode *inode, struct file *filp)
 	if (ret < 0) {
 		trace_array_put(tr);
 		kfree(dir);
+		return ret;
 	}
 
 	filp->private_data = dir;
 
-	return ret;
+	return 0;
 }
 
 static int subsystem_release(struct inode *inode, struct file *file)
@@ -1539,7 +1546,7 @@ event_create_dir(struct dentry *parent, struct ftrace_event_file *file)
 			return -1;
 		}
 	}
-	trace_create_file("filter", 0644, file->dir, call,
+	trace_create_file("filter", 0644, file->dir, file,
 			  &ftrace_event_filter_fops);
 
 	trace_create_file("format", 0444, file->dir, call,
@@ -1577,6 +1584,7 @@ static void event_remove(struct ftrace_event_call *call)
 		if (file->event_call != call)
 			continue;
 		ftrace_event_enable_disable(file, 0);
+		destroy_preds(file);
 		/*
 		 * The do_for_each_event_file() is
 		 * a double loop. After finding the call for this
@@ -1700,7 +1708,7 @@ static void __trace_remove_event_call(struct ftrace_event_call *call)
 {
 	event_remove(call);
 	trace_destroy_fields(call);
-	destroy_preds(call);
+	destroy_call_preds(call);
 }
 
 static int probe_remove_event_call(struct ftrace_event_call *call)
diff --git a/kernel/trace/trace_events_filter.c b/kernel/trace/trace_events_filter.c
index 97daa8cf958d..2468f56dc5db 100644
--- a/kernel/trace/trace_events_filter.c
+++ b/kernel/trace/trace_events_filter.c
@@ -637,10 +637,18 @@ static void append_filter_err(struct filter_parse_state *ps,
 	free_page((unsigned long) buf);
 }
 
+static inline struct event_filter *event_filter(struct ftrace_event_file *file)
+{
+	if (file->event_call->flags & TRACE_EVENT_FL_USE_CALL_FILTER)
+		return file->event_call->filter;
+	else
+		return file->filter;
+}
+
 /* caller must hold event_mutex */
-void print_event_filter(struct ftrace_event_call *call, struct trace_seq *s)
+void print_event_filter(struct ftrace_event_file *file, struct trace_seq *s)
 {
-	struct event_filter *filter = call->filter;
+	struct event_filter *filter = event_filter(file);
 
 	if (filter && filter->filter_string)
 		trace_seq_printf(s, "%s\n", filter->filter_string);
@@ -766,11 +774,21 @@ static void __free_preds(struct event_filter *filter)
 	filter->n_preds = 0;
 }
 
-static void filter_disable(struct ftrace_event_call *call)
+static void call_filter_disable(struct ftrace_event_call *call)
 {
 	call->flags &= ~TRACE_EVENT_FL_FILTERED;
 }
 
+static void filter_disable(struct ftrace_event_file *file)
+{
+	struct ftrace_event_call *call = file->event_call;
+
+	if (call->flags & TRACE_EVENT_FL_USE_CALL_FILTER)
+		call_filter_disable(call);
+	else
+		file->flags &= ~FTRACE_EVENT_FL_FILTERED;
+}
+
 static void __free_filter(struct event_filter *filter)
 {
 	if (!filter)
@@ -781,16 +799,30 @@ static void __free_filter(struct event_filter *filter)
 	kfree(filter);
 }
 
+void destroy_call_preds(struct ftrace_event_call *call)
+{
+	__free_filter(call->filter);
+	call->filter = NULL;
+}
+
+static void destroy_file_preds(struct ftrace_event_file *file)
+{
+	__free_filter(file->filter);
+	file->filter = NULL;
+}
+
 /*
- * Called when destroying the ftrace_event_call.
- * The call is being freed, so we do not need to worry about
- * the call being currently used. This is for module code removing
+ * Called when destroying the ftrace_event_file.
+ * The file is being freed, so we do not need to worry about
+ * the file being currently used. This is for module code removing
  * the tracepoints from within it.
  */
-void destroy_preds(struct ftrace_event_call *call)
+void destroy_preds(struct ftrace_event_file *file)
 {
-	__free_filter(call->filter);
-	call->filter = NULL;
+	if (file->event_call->flags & TRACE_EVENT_FL_USE_CALL_FILTER)
+		destroy_call_preds(file->event_call);
+	else
+		destroy_file_preds(file);
 }
 
 static struct event_filter *__alloc_filter(void)
@@ -825,28 +857,56 @@ static int __alloc_preds(struct event_filter *filter, int n_preds)
 	return 0;
 }
 
-static void filter_free_subsystem_preds(struct event_subsystem *system)
+static inline void __remove_filter(struct ftrace_event_file *file)
 {
+	struct ftrace_event_call *call = file->event_call;
+
+	filter_disable(file);
+	if (call->flags & TRACE_EVENT_FL_USE_CALL_FILTER)
+		remove_filter_string(call->filter);
+	else
+		remove_filter_string(file->filter);
+}
+
+static void filter_free_subsystem_preds(struct event_subsystem *system,
+					struct trace_array *tr)
+{
+	struct ftrace_event_file *file;
 	struct ftrace_event_call *call;
 
-	list_for_each_entry(call, &ftrace_events, list) {
+	list_for_each_entry(file, &tr->events, list) {
+		call = file->event_call;
 		if (strcmp(call->class->system, system->name) != 0)
 			continue;
 
-		filter_disable(call);
-		remove_filter_string(call->filter);
+		__remove_filter(file);
 	}
 }
 
-static void filter_free_subsystem_filters(struct event_subsystem *system)
+static inline void __free_subsystem_filter(struct ftrace_event_file *file)
 {
+	struct ftrace_event_call *call = file->event_call;
+
+	if (call->flags & TRACE_EVENT_FL_USE_CALL_FILTER) {
+		__free_filter(call->filter);
+		call->filter = NULL;
+	} else {
+		__free_filter(file->filter);
+		file->filter = NULL;
+	}
+}
+
+static void filter_free_subsystem_filters(struct event_subsystem *system,
+					  struct trace_array *tr)
+{
+	struct ftrace_event_file *file;
 	struct ftrace_event_call *call;
 
-	list_for_each_entry(call, &ftrace_events, list) {
+	list_for_each_entry(file, &tr->events, list) {
+		call = file->event_call;
 		if (strcmp(call->class->system, system->name) != 0)
 			continue;
-		__free_filter(call->filter);
-		call->filter = NULL;
+		__free_subsystem_filter(file);
 	}
 }
 
@@ -1617,15 +1677,85 @@ fail:
 	return err;
 }
 
+static inline void event_set_filtered_flag(struct ftrace_event_file *file)
+{
+	struct ftrace_event_call *call = file->event_call;
+
+	if (call->flags & TRACE_EVENT_FL_USE_CALL_FILTER)
+		call->flags |= TRACE_EVENT_FL_FILTERED;
+	else
+		file->flags |= FTRACE_EVENT_FL_FILTERED;
+}
+
+static inline void event_set_filter(struct ftrace_event_file *file,
+				    struct event_filter *filter)
+{
+	struct ftrace_event_call *call = file->event_call;
+
+	if (call->flags & TRACE_EVENT_FL_USE_CALL_FILTER)
+		rcu_assign_pointer(call->filter, filter);
+	else
+		rcu_assign_pointer(file->filter, filter);
+}
+
+static inline void event_clear_filter(struct ftrace_event_file *file)
+{
+	struct ftrace_event_call *call = file->event_call;
+
+	if (call->flags & TRACE_EVENT_FL_USE_CALL_FILTER)
+		RCU_INIT_POINTER(call->filter, NULL);
+	else
+		RCU_INIT_POINTER(file->filter, NULL);
+}
+
+static inline void
+event_set_no_set_filter_flag(struct ftrace_event_file *file)
+{
+	struct ftrace_event_call *call = file->event_call;
+
+	if (call->flags & TRACE_EVENT_FL_USE_CALL_FILTER)
+		call->flags |= TRACE_EVENT_FL_NO_SET_FILTER;
+	else
+		file->flags |= FTRACE_EVENT_FL_NO_SET_FILTER;
+}
+
+static inline void
+event_clear_no_set_filter_flag(struct ftrace_event_file *file)
+{
+	struct ftrace_event_call *call = file->event_call;
+
+	if (call->flags & TRACE_EVENT_FL_USE_CALL_FILTER)
+		call->flags &= ~TRACE_EVENT_FL_NO_SET_FILTER;
+	else
+		file->flags &= ~FTRACE_EVENT_FL_NO_SET_FILTER;
+}
+
+static inline bool
+event_no_set_filter_flag(struct ftrace_event_file *file)
+{
+	struct ftrace_event_call *call = file->event_call;
+
+	if (file->flags & FTRACE_EVENT_FL_NO_SET_FILTER)
+		return true;
+
+	if ((call->flags & TRACE_EVENT_FL_USE_CALL_FILTER) &&
+	    (call->flags & TRACE_EVENT_FL_NO_SET_FILTER))
+		return true;
+
+	return false;
+}
+
 struct filter_list {
 	struct list_head	list;
 	struct event_filter	*filter;
 };
 
 static int replace_system_preds(struct event_subsystem *system,
+				struct trace_array *tr,
 				struct filter_parse_state *ps,
 				char *filter_string)
 {
+	struct ftrace_event_file *file;
 	struct ftrace_event_call *call;
 	struct filter_list *filter_item;
 	struct filter_list *tmp;
@@ -1633,8 +1763,8 @@ static int replace_system_preds(struct event_subsystem *system,
 	bool fail = true;
 	int err;
 
-	list_for_each_entry(call, &ftrace_events, list) {
-
+	list_for_each_entry(file, &tr->events, list) {
+		call = file->event_call;
 		if (strcmp(call->class->system, system->name) != 0)
 			continue;
 
@@ -1644,18 +1774,20 @@ static int replace_system_preds(struct event_subsystem *system,
 		 */
 		err = replace_preds(call, NULL, ps, filter_string, true);
 		if (err)
-			call->flags |= TRACE_EVENT_FL_NO_SET_FILTER;
+			event_set_no_set_filter_flag(file);
 		else
-			call->flags &= ~TRACE_EVENT_FL_NO_SET_FILTER;
+			event_clear_no_set_filter_flag(file);
 	}
 
-	list_for_each_entry(call, &ftrace_events, list) {
+	list_for_each_entry(file, &tr->events, list) {
 		struct event_filter *filter;
 
+		call = file->event_call;
+
 		if (strcmp(call->class->system, system->name) != 0)
 			continue;
 
-		if (call->flags & TRACE_EVENT_FL_NO_SET_FILTER)
+		if (event_no_set_filter_flag(file))
 			continue;
 
 		filter_item = kzalloc(sizeof(*filter_item), GFP_KERNEL);
@@ -1676,17 +1808,17 @@ static int replace_system_preds(struct event_subsystem *system,
 
 		err = replace_preds(call, filter, ps, filter_string, false);
 		if (err) {
-			filter_disable(call);
+			filter_disable(file);
 			parse_error(ps, FILT_ERR_BAD_SUBSYS_FILTER, 0);
 			append_filter_err(ps, filter);
 		} else
-			call->flags |= TRACE_EVENT_FL_FILTERED;
+			event_set_filtered_flag(file);
 		/*
 		 * Regardless of if this returned an error, we still
 		 * replace the filter for the call.
 		 */
-		filter = call->filter;
-		rcu_assign_pointer(call->filter, filter_item->filter);
+		filter = event_filter(file);
+		event_set_filter(file, filter_item->filter);
 		filter_item->filter = filter;
 
 		fail = false;
@@ -1816,6 +1948,7 @@ static int create_filter(struct ftrace_event_call *call,
  * and always remembers @filter_str.
  */
 static int create_system_filter(struct event_subsystem *system,
+				struct trace_array *tr,
 				char *filter_str, struct event_filter **filterp)
 {
 	struct event_filter *filter = NULL;
@@ -1824,7 +1957,7 @@ static int create_system_filter(struct event_subsystem *system,
 
 	err = create_filter_start(filter_str, true, &ps, &filter);
 	if (!err) {
-		err = replace_system_preds(system, ps, filter_str);
+		err = replace_system_preds(system, tr, ps, filter_str);
 		if (!err) {
 			/* System filters just show a default message */
 			kfree(filter->filter_string);
@@ -1840,20 +1973,25 @@ static int create_system_filter(struct event_subsystem *system,
 }
 
 /* caller must hold event_mutex */
-int apply_event_filter(struct ftrace_event_call *call, char *filter_string)
+int apply_event_filter(struct ftrace_event_file *file, char *filter_string)
 {
+	struct ftrace_event_call *call = file->event_call;
 	struct event_filter *filter;
 	int err;
 
 	if (!strcmp(strstrip(filter_string), "0")) {
-		filter_disable(call);
-		filter = call->filter;
+		filter_disable(file);
+		filter = event_filter(file);
+
 		if (!filter)
 			return 0;
-		RCU_INIT_POINTER(call->filter, NULL);
+
+		event_clear_filter(file);
+
 		/* Make sure the filter is not being used */
 		synchronize_sched();
 		__free_filter(filter);
+
 		return 0;
 	}
 
@@ -1866,14 +2004,15 @@ int apply_event_filter(struct ftrace_event_call *call, char *filter_string)
 	 * string
 	 */
 	if (filter) {
-		struct event_filter *tmp = call->filter;
+		struct event_filter *tmp;
 
+		tmp = event_filter(file);
 		if (!err)
-			call->flags |= TRACE_EVENT_FL_FILTERED;
+			event_set_filtered_flag(file);
 		else
-			filter_disable(call);
+			filter_disable(file);
 
-		rcu_assign_pointer(call->filter, filter);
+		event_set_filter(file, filter);
 
 		if (tmp) {
 			/* Make sure the call is done with the filter */
@@ -1889,6 +2028,7 @@ int apply_subsystem_event_filter(struct ftrace_subsystem_dir *dir,
 				 char *filter_string)
 {
 	struct event_subsystem *system = dir->subsystem;
+	struct trace_array *tr = dir->tr;
 	struct event_filter *filter;
 	int err = 0;
 
@@ -1901,18 +2041,18 @@ int apply_subsystem_event_filter(struct ftrace_subsystem_dir *dir,
 	}
 
 	if (!strcmp(strstrip(filter_string), "0")) {
-		filter_free_subsystem_preds(system);
+		filter_free_subsystem_preds(system, tr);
 		remove_filter_string(system->filter);
 		filter = system->filter;
 		system->filter = NULL;
 		/* Ensure all filters are no longer used */
 		synchronize_sched();
-		filter_free_subsystem_filters(system);
+		filter_free_subsystem_filters(system, tr);
 		__free_filter(filter);
 		goto out_unlock;
 	}
 
-	err = create_system_filter(system, filter_string, &filter);
+	err = create_system_filter(system, tr, filter_string, &filter);
 	if (filter) {
 		/*
 		 * No event actually uses the system filter
diff --git a/kernel/trace/trace_export.c b/kernel/trace/trace_export.c
index d21a74670088..7c3e3e72e2b6 100644
--- a/kernel/trace/trace_export.c
+++ b/kernel/trace/trace_export.c
@@ -180,7 +180,7 @@ struct ftrace_event_call __used event_##call = {			\
 	.event.type		= etype,				\
 	.class			= &event_class_ftrace_##call,		\
 	.print_fmt		= print,				\
-	.flags			= TRACE_EVENT_FL_IGNORE_ENABLE,		\
+	.flags			= TRACE_EVENT_FL_IGNORE_ENABLE | TRACE_EVENT_FL_USE_CALL_FILTER, \
 };									\
 struct ftrace_event_call __used						\
 __attribute__((section("_ftrace_events"))) *__event_##call = &event_##call;
diff --git a/kernel/trace/trace_functions_graph.c b/kernel/trace/trace_functions_graph.c
index b5c09242683d..0b99120d395c 100644
--- a/kernel/trace/trace_functions_graph.c
+++ b/kernel/trace/trace_functions_graph.c
@@ -82,9 +82,9 @@ static struct trace_array *graph_array;
  * to fill in space into DURATION column.
  */
 enum {
-	DURATION_FILL_FULL  = -1,
-	DURATION_FILL_START = -2,
-	DURATION_FILL_END   = -3,
+	FLAGS_FILL_FULL  = 1 << TRACE_GRAPH_PRINT_FILL_SHIFT,
+	FLAGS_FILL_START = 2 << TRACE_GRAPH_PRINT_FILL_SHIFT,
+	FLAGS_FILL_END   = 3 << TRACE_GRAPH_PRINT_FILL_SHIFT,
 };
 
 static enum print_line_t
@@ -114,16 +114,37 @@ ftrace_push_return_trace(unsigned long ret, unsigned long func, int *depth,
 		return -EBUSY;
 	}
 
+	/*
+	 * The curr_ret_stack is an index to ftrace return stack of
+	 * current task.  Its value should be in [0, FTRACE_RETFUNC_
+	 * DEPTH) when the function graph tracer is used.  To support
+	 * filtering out specific functions, it makes the index
+	 * negative by subtracting huge value (FTRACE_NOTRACE_DEPTH)
+	 * so when it sees a negative index the ftrace will ignore
+	 * the record.  And the index gets recovered when returning
+	 * from the filtered function by adding the FTRACE_NOTRACE_
+	 * DEPTH and then it'll continue to record functions normally.
+	 *
+	 * The curr_ret_stack is initialized to -1 and get increased
+	 * in this function.  So it can be less than -1 only if it was
+	 * filtered out via ftrace_graph_notrace_addr() which can be
+	 * set from set_graph_notrace file in debugfs by user.
+	 */
+	if (current->curr_ret_stack < -1)
+		return -EBUSY;
+
 	calltime = trace_clock_local();
 
 	index = ++current->curr_ret_stack;
+	if (ftrace_graph_notrace_addr(func))
+		current->curr_ret_stack -= FTRACE_NOTRACE_DEPTH;
 	barrier();
 	current->ret_stack[index].ret = ret;
 	current->ret_stack[index].func = func;
 	current->ret_stack[index].calltime = calltime;
 	current->ret_stack[index].subtime = 0;
 	current->ret_stack[index].fp = frame_pointer;
-	*depth = index;
+	*depth = current->curr_ret_stack;
 
 	return 0;
 }
@@ -137,7 +158,17 @@ ftrace_pop_return_trace(struct ftrace_graph_ret *trace, unsigned long *ret,
 
 	index = current->curr_ret_stack;
 
-	if (unlikely(index < 0)) {
+	/*
+	 * A negative index here means that it's just returned from a
+	 * notrace'd function.  Recover index to get an original
+	 * return address.  See ftrace_push_return_trace().
+	 *
+	 * TODO: Need to check whether the stack gets corrupted.
+	 */
+	if (index < 0)
+		index += FTRACE_NOTRACE_DEPTH;
+
+	if (unlikely(index < 0 || index >= FTRACE_RETFUNC_DEPTH)) {
 		ftrace_graph_stop();
 		WARN_ON(1);
 		/* Might as well panic, otherwise we have no where to go */
@@ -193,6 +224,15 @@ unsigned long ftrace_return_to_handler(unsigned long frame_pointer)
 	trace.rettime = trace_clock_local();
 	barrier();
 	current->curr_ret_stack--;
+	/*
+	 * The curr_ret_stack can be less than -1 only if it was
+	 * filtered out and it's about to return from the function.
+	 * Recover the index and continue to trace normal functions.
+	 */
+	if (current->curr_ret_stack < -1) {
+		current->curr_ret_stack += FTRACE_NOTRACE_DEPTH;
+		return ret;
+	}
 
 	/*
 	 * The trace should run after decrementing the ret counter
@@ -230,7 +270,7 @@ int __trace_graph_entry(struct trace_array *tr,
 		return 0;
 	entry	= ring_buffer_event_data(event);
 	entry->graph_ent			= *trace;
-	if (!filter_current_check_discard(buffer, call, entry, event))
+	if (!call_filter_check_discard(call, entry, buffer, event))
 		__buffer_unlock_commit(buffer, event);
 
 	return 1;
@@ -259,10 +299,20 @@ int trace_graph_entry(struct ftrace_graph_ent *trace)
 
 	/* trace it when it is-nested-in or is a function enabled. */
 	if ((!(trace->depth || ftrace_graph_addr(trace->func)) ||
-	     ftrace_graph_ignore_irqs()) ||
+	     ftrace_graph_ignore_irqs()) || (trace->depth < 0) ||
 	    (max_depth && trace->depth >= max_depth))
 		return 0;
 
+	/*
+	 * Do not trace a function if it's filtered by set_graph_notrace.
+	 * Make the index of ret stack negative to indicate that it should
+	 * ignore further functions.  But it needs its own ret stack entry
+	 * to recover the original index in order to continue tracing after
+	 * returning from the function.
+	 */
+	if (ftrace_graph_notrace_addr(trace->func))
+		return 1;
+
 	local_irq_save(flags);
 	cpu = raw_smp_processor_id();
 	data = per_cpu_ptr(tr->trace_buffer.data, cpu);
@@ -335,7 +385,7 @@ void __trace_graph_return(struct trace_array *tr,
 		return;
 	entry	= ring_buffer_event_data(event);
 	entry->ret				= *trace;
-	if (!filter_current_check_discard(buffer, call, entry, event))
+	if (!call_filter_check_discard(call, entry, buffer, event))
 		__buffer_unlock_commit(buffer, event);
 }
 
@@ -652,7 +702,7 @@ print_graph_irq(struct trace_iterator *iter, unsigned long addr,
 	}
 
 	/* No overhead */
-	ret = print_graph_duration(DURATION_FILL_START, s, flags);
+	ret = print_graph_duration(0, s, flags | FLAGS_FILL_START);
 	if (ret != TRACE_TYPE_HANDLED)
 		return ret;
 
@@ -664,7 +714,7 @@ print_graph_irq(struct trace_iterator *iter, unsigned long addr,
 	if (!ret)
 		return TRACE_TYPE_PARTIAL_LINE;
 
-	ret = print_graph_duration(DURATION_FILL_END, s, flags);
+	ret = print_graph_duration(0, s, flags | FLAGS_FILL_END);
 	if (ret != TRACE_TYPE_HANDLED)
 		return ret;
 
@@ -729,14 +779,14 @@ print_graph_duration(unsigned long long duration, struct trace_seq *s,
 			return TRACE_TYPE_HANDLED;
 
 	/* No real adata, just filling the column with spaces */
-	switch (duration) {
-	case DURATION_FILL_FULL:
+	switch (flags & TRACE_GRAPH_PRINT_FILL_MASK) {
+	case FLAGS_FILL_FULL:
 		ret = trace_seq_puts(s, "              |  ");
 		return ret ? TRACE_TYPE_HANDLED : TRACE_TYPE_PARTIAL_LINE;
-	case DURATION_FILL_START:
+	case FLAGS_FILL_START:
 		ret = trace_seq_puts(s, "  ");
 		return ret ? TRACE_TYPE_HANDLED : TRACE_TYPE_PARTIAL_LINE;
-	case DURATION_FILL_END:
+	case FLAGS_FILL_END:
 		ret = trace_seq_puts(s, " |");
 		return ret ? TRACE_TYPE_HANDLED : TRACE_TYPE_PARTIAL_LINE;
 	}
@@ -852,7 +902,7 @@ print_graph_entry_nested(struct trace_iterator *iter,
 	}
 
 	/* No time */
-	ret = print_graph_duration(DURATION_FILL_FULL, s, flags);
+	ret = print_graph_duration(0, s, flags | FLAGS_FILL_FULL);
 	if (ret != TRACE_TYPE_HANDLED)
 		return ret;
 
@@ -1172,7 +1222,7 @@ print_graph_comment(struct trace_seq *s, struct trace_entry *ent,
 		return TRACE_TYPE_PARTIAL_LINE;
 
 	/* No time */
-	ret = print_graph_duration(DURATION_FILL_FULL, s, flags);
+	ret = print_graph_duration(0, s, flags | FLAGS_FILL_FULL);
 	if (ret != TRACE_TYPE_HANDLED)
 		return ret;
 
diff --git a/kernel/trace/trace_kprobe.c b/kernel/trace/trace_kprobe.c
index 243f6834d026..dae9541ada9e 100644
--- a/kernel/trace/trace_kprobe.c
+++ b/kernel/trace/trace_kprobe.c
@@ -835,7 +835,7 @@ __kprobe_trace_func(struct trace_probe *tp, struct pt_regs *regs,
 	entry->ip = (unsigned long)tp->rp.kp.addr;
 	store_trace_args(sizeof(*entry), tp, regs, (u8 *)&entry[1], dsize);
 
-	if (!filter_current_check_discard(buffer, call, entry, event))
+	if (!filter_check_discard(ftrace_file, entry, buffer, event))
 		trace_buffer_unlock_commit_regs(buffer, event,
 						irq_flags, pc, regs);
 }
@@ -884,7 +884,7 @@ __kretprobe_trace_func(struct trace_probe *tp, struct kretprobe_instance *ri,
 	entry->ret_ip = (unsigned long)ri->ret_addr;
 	store_trace_args(sizeof(*entry), tp, regs, (u8 *)&entry[1], dsize);
 
-	if (!filter_current_check_discard(buffer, call, entry, event))
+	if (!filter_check_discard(ftrace_file, entry, buffer, event))
 		trace_buffer_unlock_commit_regs(buffer, event,
 						irq_flags, pc, regs);
 }
diff --git a/kernel/trace/trace_mmiotrace.c b/kernel/trace/trace_mmiotrace.c
index b3dcfb2f0fef..0abd9b863474 100644
--- a/kernel/trace/trace_mmiotrace.c
+++ b/kernel/trace/trace_mmiotrace.c
@@ -323,7 +323,7 @@ static void __trace_mmiotrace_rw(struct trace_array *tr,
 	entry	= ring_buffer_event_data(event);
 	entry->rw			= *rw;
 
-	if (!filter_check_discard(call, entry, buffer, event))
+	if (!call_filter_check_discard(call, entry, buffer, event))
 		trace_buffer_unlock_commit(buffer, event, 0, pc);
 }
 
@@ -353,7 +353,7 @@ static void __trace_mmiotrace_map(struct trace_array *tr,
 	entry	= ring_buffer_event_data(event);
 	entry->map			= *map;
 
-	if (!filter_check_discard(call, entry, buffer, event))
+	if (!call_filter_check_discard(call, entry, buffer, event))
 		trace_buffer_unlock_commit(buffer, event, 0, pc);
 }
 
diff --git a/kernel/trace/trace_sched_switch.c b/kernel/trace/trace_sched_switch.c
index 4e98e3b257a3..3f34dc9b40f3 100644
--- a/kernel/trace/trace_sched_switch.c
+++ b/kernel/trace/trace_sched_switch.c
@@ -45,7 +45,7 @@ tracing_sched_switch_trace(struct trace_array *tr,
 	entry->next_state		= next->state;
 	entry->next_cpu	= task_cpu(next);
 
-	if (!filter_check_discard(call, entry, buffer, event))
+	if (!call_filter_check_discard(call, entry, buffer, event))
 		trace_buffer_unlock_commit(buffer, event, flags, pc);
 }
 
@@ -101,7 +101,7 @@ tracing_sched_wakeup_trace(struct trace_array *tr,
 	entry->next_state		= wakee->state;
 	entry->next_cpu			= task_cpu(wakee);
 
-	if (!filter_check_discard(call, entry, buffer, event))
+	if (!call_filter_check_discard(call, entry, buffer, event))
 		trace_buffer_unlock_commit(buffer, event, flags, pc);
 }
 
diff --git a/kernel/trace/trace_stat.c b/kernel/trace/trace_stat.c
index 847f88a6194b..7af67360b330 100644
--- a/kernel/trace/trace_stat.c
+++ b/kernel/trace/trace_stat.c
@@ -43,46 +43,15 @@ static DEFINE_MUTEX(all_stat_sessions_mutex);
 /* The root directory for all stat files */
 static struct dentry		*stat_dir;
 
-/*
- * Iterate through the rbtree using a post order traversal path
- * to release the next node.
- * It won't necessary release one at each iteration
- * but it will at least advance closer to the next one
- * to be released.
- */
-static struct rb_node *release_next(struct tracer_stat *ts,
-				    struct rb_node *node)
+static void __reset_stat_session(struct stat_session *session)
 {
-	struct stat_node *snode;
-	struct rb_node *parent = rb_parent(node);
-
-	if (node->rb_left)
-		return node->rb_left;
-	else if (node->rb_right)
-		return node->rb_right;
-	else {
-		if (!parent)
-			;
-		else if (parent->rb_left == node)
-			parent->rb_left = NULL;
-		else
-			parent->rb_right = NULL;
+	struct stat_node *snode, *n;
 
-		snode = container_of(node, struct stat_node, node);
-		if (ts->stat_release)
-			ts->stat_release(snode->stat);
+	rbtree_postorder_for_each_entry_safe(snode, n, &session->stat_root, node) {
+		if (session->ts->stat_release)
+			session->ts->stat_release(snode->stat);
 		kfree(snode);
-
-		return parent;
 	}
-}
-
-static void __reset_stat_session(struct stat_session *session)
-{
-	struct rb_node *node = session->stat_root.rb_node;
-
-	while (node)
-		node = release_next(session->ts, node);
 
 	session->stat_root = RB_ROOT;
 }
diff --git a/kernel/trace/trace_syscalls.c b/kernel/trace/trace_syscalls.c
index 559329d9bd2f..e4b6d11bdf78 100644
--- a/kernel/trace/trace_syscalls.c
+++ b/kernel/trace/trace_syscalls.c
@@ -302,6 +302,7 @@ static int __init syscall_exit_define_fields(struct ftrace_event_call *call)
 static void ftrace_syscall_enter(void *data, struct pt_regs *regs, long id)
 {
 	struct trace_array *tr = data;
+	struct ftrace_event_file *ftrace_file;
 	struct syscall_trace_enter *entry;
 	struct syscall_metadata *sys_data;
 	struct ring_buffer_event *event;
@@ -314,7 +315,13 @@ static void ftrace_syscall_enter(void *data, struct pt_regs *regs, long id)
 	syscall_nr = trace_get_syscall_nr(current, regs);
 	if (syscall_nr < 0)
 		return;
-	if (!test_bit(syscall_nr, tr->enabled_enter_syscalls))
+
+	/* Here we're inside tp handler's rcu_read_lock_sched (__DO_TRACE) */
+	ftrace_file = rcu_dereference_sched(tr->enter_syscall_files[syscall_nr]);
+	if (!ftrace_file)
+		return;
+
+	if (test_bit(FTRACE_EVENT_FL_SOFT_DISABLED_BIT, &ftrace_file->flags))
 		return;
 
 	sys_data = syscall_nr_to_meta(syscall_nr);
@@ -336,8 +343,7 @@ static void ftrace_syscall_enter(void *data, struct pt_regs *regs, long id)
 	entry->nr = syscall_nr;
 	syscall_get_arguments(current, regs, 0, sys_data->nb_args, entry->args);
 
-	if (!filter_current_check_discard(buffer, sys_data->enter_event,
-					  entry, event))
+	if (!filter_check_discard(ftrace_file, entry, buffer, event))
 		trace_current_buffer_unlock_commit(buffer, event,
 						   irq_flags, pc);
 }
@@ -345,6 +351,7 @@ static void ftrace_syscall_enter(void *data, struct pt_regs *regs, long id)
 static void ftrace_syscall_exit(void *data, struct pt_regs *regs, long ret)
 {
 	struct trace_array *tr = data;
+	struct ftrace_event_file *ftrace_file;
 	struct syscall_trace_exit *entry;
 	struct syscall_metadata *sys_data;
 	struct ring_buffer_event *event;
@@ -356,7 +363,13 @@ static void ftrace_syscall_exit(void *data, struct pt_regs *regs, long ret)
 	syscall_nr = trace_get_syscall_nr(current, regs);
 	if (syscall_nr < 0)
 		return;
-	if (!test_bit(syscall_nr, tr->enabled_exit_syscalls))
+
+	/* Here we're inside tp handler's rcu_read_lock_sched (__DO_TRACE()) */
+	ftrace_file = rcu_dereference_sched(tr->exit_syscall_files[syscall_nr]);
+	if (!ftrace_file)
+		return;
+
+	if (test_bit(FTRACE_EVENT_FL_SOFT_DISABLED_BIT, &ftrace_file->flags))
 		return;
 
 	sys_data = syscall_nr_to_meta(syscall_nr);
@@ -377,8 +390,7 @@ static void ftrace_syscall_exit(void *data, struct pt_regs *regs, long ret)
 	entry->nr = syscall_nr;
 	entry->ret = syscall_get_return_value(current, regs);
 
-	if (!filter_current_check_discard(buffer, sys_data->exit_event,
-					  entry, event))
+	if (!filter_check_discard(ftrace_file, entry, buffer, event))
 		trace_current_buffer_unlock_commit(buffer, event,
 						   irq_flags, pc);
 }
@@ -397,7 +409,7 @@ static int reg_event_syscall_enter(struct ftrace_event_file *file,
 	if (!tr->sys_refcount_enter)
 		ret = register_trace_sys_enter(ftrace_syscall_enter, tr);
 	if (!ret) {
-		set_bit(num, tr->enabled_enter_syscalls);
+		rcu_assign_pointer(tr->enter_syscall_files[num], file);
 		tr->sys_refcount_enter++;
 	}
 	mutex_unlock(&syscall_trace_lock);
@@ -415,10 +427,15 @@ static void unreg_event_syscall_enter(struct ftrace_event_file *file,
 		return;
 	mutex_lock(&syscall_trace_lock);
 	tr->sys_refcount_enter--;
-	clear_bit(num, tr->enabled_enter_syscalls);
+	rcu_assign_pointer(tr->enter_syscall_files[num], NULL);
 	if (!tr->sys_refcount_enter)
 		unregister_trace_sys_enter(ftrace_syscall_enter, tr);
 	mutex_unlock(&syscall_trace_lock);
+	/*
+	 * Callers expect the event to be completely disabled on
+	 * return, so wait for current handlers to finish.
+	 */
+	synchronize_sched();
 }
 
 static int reg_event_syscall_exit(struct ftrace_event_file *file,
@@ -435,7 +452,7 @@ static int reg_event_syscall_exit(struct ftrace_event_file *file,
 	if (!tr->sys_refcount_exit)
 		ret = register_trace_sys_exit(ftrace_syscall_exit, tr);
 	if (!ret) {
-		set_bit(num, tr->enabled_exit_syscalls);
+		rcu_assign_pointer(tr->exit_syscall_files[num], file);
 		tr->sys_refcount_exit++;
 	}
 	mutex_unlock(&syscall_trace_lock);
@@ -453,10 +470,15 @@ static void unreg_event_syscall_exit(struct ftrace_event_file *file,
 		return;
 	mutex_lock(&syscall_trace_lock);
 	tr->sys_refcount_exit--;
-	clear_bit(num, tr->enabled_exit_syscalls);
+	rcu_assign_pointer(tr->exit_syscall_files[num], NULL);
 	if (!tr->sys_refcount_exit)
 		unregister_trace_sys_exit(ftrace_syscall_exit, tr);
 	mutex_unlock(&syscall_trace_lock);
+	/*
+	 * Callers expect the event to be completely disabled on
+	 * return, so wait for current handlers to finish.
+	 */
+	synchronize_sched();
 }
 
 static int __init init_syscall_trace(struct ftrace_event_call *call)
diff --git a/kernel/trace/trace_uprobe.c b/kernel/trace/trace_uprobe.c
index 272261b5f94f..b6dcc42ef7f5 100644
--- a/kernel/trace/trace_uprobe.c
+++ b/kernel/trace/trace_uprobe.c
@@ -128,6 +128,7 @@ alloc_trace_uprobe(const char *group, const char *event, int nargs, bool is_ret)
 	if (is_ret)
 		tu->consumer.ret_handler = uretprobe_dispatcher;
 	init_trace_uprobe_filter(&tu->filter);
+	tu->call.flags |= TRACE_EVENT_FL_USE_CALL_FILTER;
 	return tu;
 
 error:
@@ -561,7 +562,7 @@ static void uprobe_trace_print(struct trace_uprobe *tu,
 	for (i = 0; i < tu->nr_args; i++)
 		call_fetch(&tu->args[i].fetch, regs, data + tu->args[i].offset);
 
-	if (!filter_current_check_discard(buffer, call, entry, event))
+	if (!call_filter_check_discard(call, entry, buffer, event))
 		trace_buffer_unlock_commit(buffer, event, 0, 0);
 }
 
diff --git a/kernel/up.c b/kernel/up.c
index 630d72bf7e41..509403e3fbc6 100644
--- a/kernel/up.c
+++ b/kernel/up.c
@@ -22,6 +22,17 @@ int smp_call_function_single(int cpu, void (*func) (void *info), void *info,
 }
 EXPORT_SYMBOL(smp_call_function_single);
 
+void __smp_call_function_single(int cpu, struct call_single_data *csd,
+				int wait)
+{
+	unsigned long flags;
+
+	local_irq_save(flags);
+	csd->func(csd->info);
+	local_irq_restore(flags);
+}
+EXPORT_SYMBOL(__smp_call_function_single);
+
 int on_each_cpu(smp_call_func_t func, void *info, int wait)
 {
 	unsigned long flags;
diff --git a/lib/Kconfig b/lib/Kconfig
index 75485e163ca3..06dc74200a51 100644
--- a/lib/Kconfig
+++ b/lib/Kconfig
@@ -51,13 +51,6 @@ config PERCPU_RWSEM
 config ARCH_USE_CMPXCHG_LOCKREF
 	bool
 
-config CMPXCHG_LOCKREF
-	def_bool y if ARCH_USE_CMPXCHG_LOCKREF
-	depends on SMP
-	depends on !GENERIC_LOCKBREAK
-	depends on !DEBUG_SPINLOCK
-	depends on !DEBUG_LOCK_ALLOC
-
 config CRC_CCITT
 	tristate "CRC-CCITT functions"
 	help
diff --git a/lib/Makefile b/lib/Makefile
index bb016e116ba4..d480a8c92385 100644
--- a/lib/Makefile
+++ b/lib/Makefile
@@ -42,10 +42,6 @@ obj-$(CONFIG_GENERIC_PCI_IOMAP) += pci_iomap.o
 obj-$(CONFIG_HAS_IOMEM) += iomap_copy.o devres.o
 obj-$(CONFIG_CHECK_SIGNATURE) += check_signature.o
 obj-$(CONFIG_DEBUG_LOCKING_API_SELFTESTS) += locking-selftest.o
-obj-$(CONFIG_DEBUG_SPINLOCK) += spinlock_debug.o
-lib-$(CONFIG_RWSEM_GENERIC_SPINLOCK) += rwsem-spinlock.o
-lib-$(CONFIG_RWSEM_XCHGADD_ALGORITHM) += rwsem.o
-lib-$(CONFIG_PERCPU_RWSEM) += percpu-rwsem.o
 
 CFLAGS_hweight.o = $(subst $(quote),,$(CONFIG_ARCH_HWEIGHT_CFLAGS))
 obj-$(CONFIG_GENERIC_HWEIGHT) += hweight.o
diff --git a/lib/kfifo.c b/lib/kfifo.c
index 7b7f83027b7b..d79b9d222065 100644
--- a/lib/kfifo.c
+++ b/lib/kfifo.c
@@ -215,7 +215,7 @@ static unsigned long kfifo_copy_from_user(struct __kfifo *fifo,
 	 * incrementing the fifo->in index counter
 	 */
 	smp_wmb();
-	*copied = len - ret;
+	*copied = len - ret * esize;
 	/* return the number of elements which are not copied */
 	return ret;
 }
@@ -275,7 +275,7 @@ static unsigned long kfifo_copy_to_user(struct __kfifo *fifo, void __user *to,
 	 * incrementing the fifo->out index counter
 	 */
 	smp_wmb();
-	*copied = len - ret;
+	*copied = len - ret * esize;
 	/* return the number of elements which are not copied */
 	return ret;
 }
diff --git a/lib/llist.c b/lib/llist.c
index 4a70d120138c..f76196d07409 100644
--- a/lib/llist.c
+++ b/lib/llist.c
@@ -81,3 +81,25 @@ struct llist_node *llist_del_first(struct llist_head *head)
 	return entry;
 }
 EXPORT_SYMBOL_GPL(llist_del_first);
+
+/**
+ * llist_reverse_order - reverse order of a llist chain
+ * @head:	first item of the list to be reversed
+ *
+ * Reverse the order of a chain of llist entries and return the
+ * new first entry.
+ */
+struct llist_node *llist_reverse_order(struct llist_node *head)
+{
+	struct llist_node *new_head = NULL;
+
+	while (head) {
+		struct llist_node *tmp = head;
+		head = head->next;
+		tmp->next = new_head;
+		new_head = tmp;
+	}
+
+	return new_head;
+}
+EXPORT_SYMBOL_GPL(llist_reverse_order);
diff --git a/lib/lockref.c b/lib/lockref.c
index af6e95d0bed6..d2b123f8456b 100644
--- a/lib/lockref.c
+++ b/lib/lockref.c
@@ -1,7 +1,7 @@
 #include <linux/export.h>
 #include <linux/lockref.h>
 
-#ifdef CONFIG_CMPXCHG_LOCKREF
+#if USE_CMPXCHG_LOCKREF
 
 /*
  * Allow weakly-ordered memory architectures to provide barrier-less
diff --git a/lib/random32.c b/lib/random32.c
index 82da4f4c3489..1e5b2df44291 100644
--- a/lib/random32.c
+++ b/lib/random32.c
@@ -214,18 +214,22 @@ static DEFINE_TIMER(seed_timer, __prandom_timer, 0, 0);
 static void __prandom_timer(unsigned long dontcare)
 {
 	u32 entropy;
+	unsigned long expires;
 
 	get_random_bytes(&entropy, sizeof(entropy));
 	prandom_seed(entropy);
+
 	/* reseed every ~60 seconds, in [40 .. 80) interval with slack */
-	seed_timer.expires = jiffies + (40 * HZ + (prandom_u32() % (40 * HZ)));
+	expires = 40 + (prandom_u32() % 40);
+	seed_timer.expires = jiffies + msecs_to_jiffies(expires * MSEC_PER_SEC);
+
 	add_timer(&seed_timer);
 }
 
-static void prandom_start_seed_timer(void)
+static void __init __prandom_start_seed_timer(void)
 {
 	set_timer_slack(&seed_timer, HZ);
-	seed_timer.expires = jiffies + 40 * HZ;
+	seed_timer.expires = jiffies + msecs_to_jiffies(40 * MSEC_PER_SEC);
 	add_timer(&seed_timer);
 }
 
@@ -270,7 +274,7 @@ void prandom_reseed_late(void)
 static int __init prandom_reseed(void)
 {
 	__prandom_reseed(false);
-	prandom_start_seed_timer();
+	__prandom_start_seed_timer();
 	return 0;
 }
 late_initcall(prandom_reseed);
diff --git a/lib/swiotlb.c b/lib/swiotlb.c
index 4e8686c7e5a4..e4399fa65ad6 100644
--- a/lib/swiotlb.c
+++ b/lib/swiotlb.c
@@ -38,6 +38,9 @@
 #include <linux/bootmem.h>
 #include <linux/iommu-helper.h>
 
+#define CREATE_TRACE_POINTS
+#include <trace/events/swiotlb.h>
+
 #define OFFSET(val,align) ((unsigned long)	\
 	                   ( (val) & ( (align) - 1)))
 
@@ -502,6 +505,7 @@ phys_addr_t swiotlb_tbl_map_single(struct device *hwdev,
 
 not_found:
 	spin_unlock_irqrestore(&io_tlb_lock, flags);
+	dev_warn(hwdev, "swiotlb buffer is full\n");
 	return SWIOTLB_MAP_ERROR;
 found:
 	spin_unlock_irqrestore(&io_tlb_lock, flags);
@@ -726,6 +730,8 @@ dma_addr_t swiotlb_map_page(struct device *dev, struct page *page,
 	if (dma_capable(dev, dev_addr, size) && !swiotlb_force)
 		return dev_addr;
 
+	trace_swiotlb_bounced(dev, dev_addr, size, swiotlb_force);
+
 	/* Oh well, have to allocate and map a bounce buffer. */
 	map = map_single(dev, phys, size, dir);
 	if (map == SWIOTLB_MAP_ERROR) {
diff --git a/lib/vsprintf.c b/lib/vsprintf.c
index 48586ac3a62e..10909c571494 100644
--- a/lib/vsprintf.c
+++ b/lib/vsprintf.c
@@ -1712,18 +1712,16 @@ int vsnprintf(char *buf, size_t size, const char *fmt, va_list args)
 			break;
 
 		case FORMAT_TYPE_NRCHARS: {
-			u8 qualifier = spec.qualifier;
+			/*
+			 * Since %n poses a greater security risk than
+			 * utility, ignore %n and skip its argument.
+			 */
+			void *skip_arg;
 
-			if (qualifier == 'l') {
-				long *ip = va_arg(args, long *);
-				*ip = (str - buf);
-			} else if (_tolower(qualifier) == 'z') {
-				size_t *ip = va_arg(args, size_t *);
-				*ip = (str - buf);
-			} else {
-				int *ip = va_arg(args, int *);
-				*ip = (str - buf);
-			}
+			WARN_ONCE(1, "Please remove ignored %%n in '%s'\n",
+					old_fmt);
+
+			skip_arg = va_arg(args, void *);
 			break;
 		}
 
diff --git a/mm/Kconfig b/mm/Kconfig
index 3f4ffda152bb..eb69f352401d 100644
--- a/mm/Kconfig
+++ b/mm/Kconfig
@@ -20,7 +20,7 @@ config FLATMEM_MANUAL
 
 	  Some users of more advanced features like NUMA and
 	  memory hotplug may have different options here.
-	  DISCONTIGMEM is an more mature, better tested system,
+	  DISCONTIGMEM is a more mature, better tested system,
 	  but is incompatible with memory hotplug and may suffer
 	  decreased performance over SPARSEMEM.  If unsure between
 	  "Sparse Memory" and "Discontiguous Memory", choose
@@ -218,9 +218,11 @@ config SPLIT_PTLOCK_CPUS
 	int
 	default "999999" if ARM && !CPU_CACHE_VIPT
 	default "999999" if PARISC && !PA20
-	default "999999" if DEBUG_SPINLOCK || DEBUG_LOCK_ALLOC
 	default "4"
 
+config ARCH_ENABLE_SPLIT_PMD_PTLOCK
+	boolean
+
 #
 # support for memory balloon compaction
 config BALLOON_COMPACTION
diff --git a/mm/filemap.c b/mm/filemap.c
index ae4846ff4849..b7749a92021c 100644
--- a/mm/filemap.c
+++ b/mm/filemap.c
@@ -1090,7 +1090,6 @@ static void shrink_readahead_size_eio(struct file *filp,
  * @filp:	the file to read
  * @ppos:	current file position
  * @desc:	read_descriptor
- * @actor:	read method
  *
  * This is a generic file read routine, and uses the
  * mapping->a_ops->readpage() function for the actual low-level stuff.
@@ -1099,7 +1098,7 @@ static void shrink_readahead_size_eio(struct file *filp,
  * of the logic when it comes to error handling etc.
  */
 static void do_generic_file_read(struct file *filp, loff_t *ppos,
-		read_descriptor_t *desc, read_actor_t actor)
+		read_descriptor_t *desc)
 {
 	struct address_space *mapping = filp->f_mapping;
 	struct inode *inode = mapping->host;
@@ -1200,13 +1199,14 @@ page_ok:
 		 * Ok, we have the page, and it's up-to-date, so
 		 * now we can copy it to user space...
 		 *
-		 * The actor routine returns how many bytes were actually used..
+		 * The file_read_actor routine returns how many bytes were
+		 * actually used..
 		 * NOTE! This may not be the same as how much of a user buffer
 		 * we filled up (we may be padding etc), so we can only update
 		 * "pos" here (the actor routine has to update the user buffer
 		 * pointers and the remaining count).
 		 */
-		ret = actor(desc, page, offset, nr);
+		ret = file_read_actor(desc, page, offset, nr);
 		offset += ret;
 		index += offset >> PAGE_CACHE_SHIFT;
 		offset &= ~PAGE_CACHE_MASK;
@@ -1479,7 +1479,7 @@ generic_file_aio_read(struct kiocb *iocb, const struct iovec *iov,
 		if (desc.count == 0)
 			continue;
 		desc.error = 0;
-		do_generic_file_read(filp, ppos, &desc, file_read_actor);
+		do_generic_file_read(filp, ppos, &desc);
 		retval += desc.written;
 		if (desc.error) {
 			retval = retval ?: desc.error;
diff --git a/mm/filemap_xip.c b/mm/filemap_xip.c
index 28fe26b64f8a..d8d9fe3f685c 100644
--- a/mm/filemap_xip.c
+++ b/mm/filemap_xip.c
@@ -26,7 +26,7 @@
  * of ZERO_PAGE(), such as /dev/zero
  */
 static DEFINE_MUTEX(xip_sparse_mutex);
-static seqcount_t xip_sparse_seq = SEQCNT_ZERO;
+static seqcount_t xip_sparse_seq = SEQCNT_ZERO(xip_sparse_seq);
 static struct page *__xip_sparse_page;
 
 /* called under xip_sparse_mutex */
diff --git a/mm/huge_memory.c b/mm/huge_memory.c
index 0556c6a44959..bccd5a628ea6 100644
--- a/mm/huge_memory.c
+++ b/mm/huge_memory.c
@@ -710,6 +710,7 @@ static int __do_huge_pmd_anonymous_page(struct mm_struct *mm,
 					struct page *page)
 {
 	pgtable_t pgtable;
+	spinlock_t *ptl;
 
 	VM_BUG_ON(!PageCompound(page));
 	pgtable = pte_alloc_one(mm, haddr);
@@ -724,9 +725,9 @@ static int __do_huge_pmd_anonymous_page(struct mm_struct *mm,
 	 */
 	__SetPageUptodate(page);
 
-	spin_lock(&mm->page_table_lock);
+	ptl = pmd_lock(mm, pmd);
 	if (unlikely(!pmd_none(*pmd))) {
-		spin_unlock(&mm->page_table_lock);
+		spin_unlock(ptl);
 		mem_cgroup_uncharge_page(page);
 		put_page(page);
 		pte_free(mm, pgtable);
@@ -738,8 +739,8 @@ static int __do_huge_pmd_anonymous_page(struct mm_struct *mm,
 		pgtable_trans_huge_deposit(mm, pmd, pgtable);
 		set_pmd_at(mm, haddr, pmd, entry);
 		add_mm_counter(mm, MM_ANONPAGES, HPAGE_PMD_NR);
-		mm->nr_ptes++;
-		spin_unlock(&mm->page_table_lock);
+		atomic_long_inc(&mm->nr_ptes);
+		spin_unlock(ptl);
 	}
 
 	return 0;
@@ -759,6 +760,7 @@ static inline struct page *alloc_hugepage_vma(int defrag,
 			       HPAGE_PMD_ORDER, vma, haddr, nd);
 }
 
+/* Caller must hold page table lock. */
 static bool set_huge_zero_page(pgtable_t pgtable, struct mm_struct *mm,
 		struct vm_area_struct *vma, unsigned long haddr, pmd_t *pmd,
 		struct page *zero_page)
@@ -771,7 +773,7 @@ static bool set_huge_zero_page(pgtable_t pgtable, struct mm_struct *mm,
 	entry = pmd_mkhuge(entry);
 	pgtable_trans_huge_deposit(mm, pmd, pgtable);
 	set_pmd_at(mm, haddr, pmd, entry);
-	mm->nr_ptes++;
+	atomic_long_inc(&mm->nr_ptes);
 	return true;
 }
 
@@ -790,6 +792,7 @@ int do_huge_pmd_anonymous_page(struct mm_struct *mm, struct vm_area_struct *vma,
 		return VM_FAULT_OOM;
 	if (!(flags & FAULT_FLAG_WRITE) &&
 			transparent_hugepage_use_zero_page()) {
+		spinlock_t *ptl;
 		pgtable_t pgtable;
 		struct page *zero_page;
 		bool set;
@@ -802,10 +805,10 @@ int do_huge_pmd_anonymous_page(struct mm_struct *mm, struct vm_area_struct *vma,
 			count_vm_event(THP_FAULT_FALLBACK);
 			return VM_FAULT_FALLBACK;
 		}
-		spin_lock(&mm->page_table_lock);
+		ptl = pmd_lock(mm, pmd);
 		set = set_huge_zero_page(pgtable, mm, vma, haddr, pmd,
 				zero_page);
-		spin_unlock(&mm->page_table_lock);
+		spin_unlock(ptl);
 		if (!set) {
 			pte_free(mm, pgtable);
 			put_huge_zero_page();
@@ -838,6 +841,7 @@ int copy_huge_pmd(struct mm_struct *dst_mm, struct mm_struct *src_mm,
 		  pmd_t *dst_pmd, pmd_t *src_pmd, unsigned long addr,
 		  struct vm_area_struct *vma)
 {
+	spinlock_t *dst_ptl, *src_ptl;
 	struct page *src_page;
 	pmd_t pmd;
 	pgtable_t pgtable;
@@ -848,8 +852,9 @@ int copy_huge_pmd(struct mm_struct *dst_mm, struct mm_struct *src_mm,
 	if (unlikely(!pgtable))
 		goto out;
 
-	spin_lock(&dst_mm->page_table_lock);
-	spin_lock_nested(&src_mm->page_table_lock, SINGLE_DEPTH_NESTING);
+	dst_ptl = pmd_lock(dst_mm, dst_pmd);
+	src_ptl = pmd_lockptr(src_mm, src_pmd);
+	spin_lock_nested(src_ptl, SINGLE_DEPTH_NESTING);
 
 	ret = -EAGAIN;
 	pmd = *src_pmd;
@@ -858,7 +863,7 @@ int copy_huge_pmd(struct mm_struct *dst_mm, struct mm_struct *src_mm,
 		goto out_unlock;
 	}
 	/*
-	 * mm->page_table_lock is enough to be sure that huge zero pmd is not
+	 * When page table lock is held, the huge zero pmd should not be
 	 * under splitting since we don't split the page itself, only pmd to
 	 * a page table.
 	 */
@@ -879,8 +884,8 @@ int copy_huge_pmd(struct mm_struct *dst_mm, struct mm_struct *src_mm,
 	}
 	if (unlikely(pmd_trans_splitting(pmd))) {
 		/* split huge page running from under us */
-		spin_unlock(&src_mm->page_table_lock);
-		spin_unlock(&dst_mm->page_table_lock);
+		spin_unlock(src_ptl);
+		spin_unlock(dst_ptl);
 		pte_free(dst_mm, pgtable);
 
 		wait_split_huge_page(vma->anon_vma, src_pmd); /* src_vma */
@@ -896,12 +901,12 @@ int copy_huge_pmd(struct mm_struct *dst_mm, struct mm_struct *src_mm,
 	pmd = pmd_mkold(pmd_wrprotect(pmd));
 	pgtable_trans_huge_deposit(dst_mm, dst_pmd, pgtable);
 	set_pmd_at(dst_mm, addr, dst_pmd, pmd);
-	dst_mm->nr_ptes++;
+	atomic_long_inc(&dst_mm->nr_ptes);
 
 	ret = 0;
 out_unlock:
-	spin_unlock(&src_mm->page_table_lock);
-	spin_unlock(&dst_mm->page_table_lock);
+	spin_unlock(src_ptl);
+	spin_unlock(dst_ptl);
 out:
 	return ret;
 }
@@ -912,10 +917,11 @@ void huge_pmd_set_accessed(struct mm_struct *mm,
 			   pmd_t *pmd, pmd_t orig_pmd,
 			   int dirty)
 {
+	spinlock_t *ptl;
 	pmd_t entry;
 	unsigned long haddr;
 
-	spin_lock(&mm->page_table_lock);
+	ptl = pmd_lock(mm, pmd);
 	if (unlikely(!pmd_same(*pmd, orig_pmd)))
 		goto unlock;
 
@@ -925,13 +931,14 @@ void huge_pmd_set_accessed(struct mm_struct *mm,
 		update_mmu_cache_pmd(vma, address, pmd);
 
 unlock:
-	spin_unlock(&mm->page_table_lock);
+	spin_unlock(ptl);
 }
 
 static int do_huge_pmd_wp_zero_page_fallback(struct mm_struct *mm,
 		struct vm_area_struct *vma, unsigned long address,
 		pmd_t *pmd, pmd_t orig_pmd, unsigned long haddr)
 {
+	spinlock_t *ptl;
 	pgtable_t pgtable;
 	pmd_t _pmd;
 	struct page *page;
@@ -958,7 +965,7 @@ static int do_huge_pmd_wp_zero_page_fallback(struct mm_struct *mm,
 	mmun_end   = haddr + HPAGE_PMD_SIZE;
 	mmu_notifier_invalidate_range_start(mm, mmun_start, mmun_end);
 
-	spin_lock(&mm->page_table_lock);
+	ptl = pmd_lock(mm, pmd);
 	if (unlikely(!pmd_same(*pmd, orig_pmd)))
 		goto out_free_page;
 
@@ -985,7 +992,7 @@ static int do_huge_pmd_wp_zero_page_fallback(struct mm_struct *mm,
 	}
 	smp_wmb(); /* make pte visible before pmd */
 	pmd_populate(mm, pmd, pgtable);
-	spin_unlock(&mm->page_table_lock);
+	spin_unlock(ptl);
 	put_huge_zero_page();
 	inc_mm_counter(mm, MM_ANONPAGES);
 
@@ -995,7 +1002,7 @@ static int do_huge_pmd_wp_zero_page_fallback(struct mm_struct *mm,
 out:
 	return ret;
 out_free_page:
-	spin_unlock(&mm->page_table_lock);
+	spin_unlock(ptl);
 	mmu_notifier_invalidate_range_end(mm, mmun_start, mmun_end);
 	mem_cgroup_uncharge_page(page);
 	put_page(page);
@@ -1009,6 +1016,7 @@ static int do_huge_pmd_wp_page_fallback(struct mm_struct *mm,
 					struct page *page,
 					unsigned long haddr)
 {
+	spinlock_t *ptl;
 	pgtable_t pgtable;
 	pmd_t _pmd;
 	int ret = 0, i;
@@ -1055,7 +1063,7 @@ static int do_huge_pmd_wp_page_fallback(struct mm_struct *mm,
 	mmun_end   = haddr + HPAGE_PMD_SIZE;
 	mmu_notifier_invalidate_range_start(mm, mmun_start, mmun_end);
 
-	spin_lock(&mm->page_table_lock);
+	ptl = pmd_lock(mm, pmd);
 	if (unlikely(!pmd_same(*pmd, orig_pmd)))
 		goto out_free_pages;
 	VM_BUG_ON(!PageHead(page));
@@ -1081,7 +1089,7 @@ static int do_huge_pmd_wp_page_fallback(struct mm_struct *mm,
 	smp_wmb(); /* make pte visible before pmd */
 	pmd_populate(mm, pmd, pgtable);
 	page_remove_rmap(page);
-	spin_unlock(&mm->page_table_lock);
+	spin_unlock(ptl);
 
 	mmu_notifier_invalidate_range_end(mm, mmun_start, mmun_end);
 
@@ -1092,7 +1100,7 @@ out:
 	return ret;
 
 out_free_pages:
-	spin_unlock(&mm->page_table_lock);
+	spin_unlock(ptl);
 	mmu_notifier_invalidate_range_end(mm, mmun_start, mmun_end);
 	mem_cgroup_uncharge_start();
 	for (i = 0; i < HPAGE_PMD_NR; i++) {
@@ -1107,17 +1115,19 @@ out_free_pages:
 int do_huge_pmd_wp_page(struct mm_struct *mm, struct vm_area_struct *vma,
 			unsigned long address, pmd_t *pmd, pmd_t orig_pmd)
 {
+	spinlock_t *ptl;
 	int ret = 0;
 	struct page *page = NULL, *new_page;
 	unsigned long haddr;
 	unsigned long mmun_start;	/* For mmu_notifiers */
 	unsigned long mmun_end;		/* For mmu_notifiers */
 
+	ptl = pmd_lockptr(mm, pmd);
 	VM_BUG_ON(!vma->anon_vma);
 	haddr = address & HPAGE_PMD_MASK;
 	if (is_huge_zero_pmd(orig_pmd))
 		goto alloc;
-	spin_lock(&mm->page_table_lock);
+	spin_lock(ptl);
 	if (unlikely(!pmd_same(*pmd, orig_pmd)))
 		goto out_unlock;
 
@@ -1133,7 +1143,7 @@ int do_huge_pmd_wp_page(struct mm_struct *mm, struct vm_area_struct *vma,
 		goto out_unlock;
 	}
 	get_page(page);
-	spin_unlock(&mm->page_table_lock);
+	spin_unlock(ptl);
 alloc:
 	if (transparent_hugepage_enabled(vma) &&
 	    !transparent_hugepage_debug_cow())
@@ -1180,11 +1190,11 @@ alloc:
 	mmun_end   = haddr + HPAGE_PMD_SIZE;
 	mmu_notifier_invalidate_range_start(mm, mmun_start, mmun_end);
 
-	spin_lock(&mm->page_table_lock);
+	spin_lock(ptl);
 	if (page)
 		put_page(page);
 	if (unlikely(!pmd_same(*pmd, orig_pmd))) {
-		spin_unlock(&mm->page_table_lock);
+		spin_unlock(ptl);
 		mem_cgroup_uncharge_page(new_page);
 		put_page(new_page);
 		goto out_mn;
@@ -1206,13 +1216,13 @@ alloc:
 		}
 		ret |= VM_FAULT_WRITE;
 	}
-	spin_unlock(&mm->page_table_lock);
+	spin_unlock(ptl);
 out_mn:
 	mmu_notifier_invalidate_range_end(mm, mmun_start, mmun_end);
 out:
 	return ret;
 out_unlock:
-	spin_unlock(&mm->page_table_lock);
+	spin_unlock(ptl);
 	return ret;
 }
 
@@ -1224,7 +1234,7 @@ struct page *follow_trans_huge_pmd(struct vm_area_struct *vma,
 	struct mm_struct *mm = vma->vm_mm;
 	struct page *page = NULL;
 
-	assert_spin_locked(&mm->page_table_lock);
+	assert_spin_locked(pmd_lockptr(mm, pmd));
 
 	if (flags & FOLL_WRITE && !pmd_write(*pmd))
 		goto out;
@@ -1271,6 +1281,7 @@ out:
 int do_huge_pmd_numa_page(struct mm_struct *mm, struct vm_area_struct *vma,
 				unsigned long addr, pmd_t pmd, pmd_t *pmdp)
 {
+	spinlock_t *ptl;
 	struct anon_vma *anon_vma = NULL;
 	struct page *page;
 	unsigned long haddr = addr & HPAGE_PMD_MASK;
@@ -1280,7 +1291,7 @@ int do_huge_pmd_numa_page(struct mm_struct *mm, struct vm_area_struct *vma,
 	bool migrated = false;
 	int flags = 0;
 
-	spin_lock(&mm->page_table_lock);
+	ptl = pmd_lock(mm, pmdp);
 	if (unlikely(!pmd_same(pmd, *pmdp)))
 		goto out_unlock;
 
@@ -1318,7 +1329,7 @@ int do_huge_pmd_numa_page(struct mm_struct *mm, struct vm_area_struct *vma,
 		 * relock and check_same as the page may no longer be mapped.
 		 * As the fault is being retried, do not account for it.
 		 */
-		spin_unlock(&mm->page_table_lock);
+		spin_unlock(ptl);
 		wait_on_page_locked(page);
 		page_nid = -1;
 		goto out;
@@ -1326,13 +1337,13 @@ int do_huge_pmd_numa_page(struct mm_struct *mm, struct vm_area_struct *vma,
 
 	/* Page is misplaced, serialise migrations and parallel THP splits */
 	get_page(page);
-	spin_unlock(&mm->page_table_lock);
+	spin_unlock(ptl);
 	if (!page_locked)
 		lock_page(page);
 	anon_vma = page_lock_anon_vma_read(page);
 
 	/* Confirm the PMD did not change while page_table_lock was released */
-	spin_lock(&mm->page_table_lock);
+	spin_lock(ptl);
 	if (unlikely(!pmd_same(pmd, *pmdp))) {
 		unlock_page(page);
 		put_page(page);
@@ -1344,7 +1355,7 @@ int do_huge_pmd_numa_page(struct mm_struct *mm, struct vm_area_struct *vma,
 	 * Migrate the THP to the requested node, returns with page unlocked
 	 * and pmd_numa cleared.
 	 */
-	spin_unlock(&mm->page_table_lock);
+	spin_unlock(ptl);
 	migrated = migrate_misplaced_transhuge_page(mm, vma,
 				pmdp, pmd, addr, page, target_nid);
 	if (migrated) {
@@ -1361,7 +1372,7 @@ clear_pmdnuma:
 	update_mmu_cache_pmd(vma, addr, pmdp);
 	unlock_page(page);
 out_unlock:
-	spin_unlock(&mm->page_table_lock);
+	spin_unlock(ptl);
 
 out:
 	if (anon_vma)
@@ -1376,9 +1387,10 @@ out:
 int zap_huge_pmd(struct mmu_gather *tlb, struct vm_area_struct *vma,
 		 pmd_t *pmd, unsigned long addr)
 {
+	spinlock_t *ptl;
 	int ret = 0;
 
-	if (__pmd_trans_huge_lock(pmd, vma) == 1) {
+	if (__pmd_trans_huge_lock(pmd, vma, &ptl) == 1) {
 		struct page *page;
 		pgtable_t pgtable;
 		pmd_t orig_pmd;
@@ -1392,8 +1404,8 @@ int zap_huge_pmd(struct mmu_gather *tlb, struct vm_area_struct *vma,
 		tlb_remove_pmd_tlb_entry(tlb, pmd, addr);
 		pgtable = pgtable_trans_huge_withdraw(tlb->mm, pmd);
 		if (is_huge_zero_pmd(orig_pmd)) {
-			tlb->mm->nr_ptes--;
-			spin_unlock(&tlb->mm->page_table_lock);
+			atomic_long_dec(&tlb->mm->nr_ptes);
+			spin_unlock(ptl);
 			put_huge_zero_page();
 		} else {
 			page = pmd_page(orig_pmd);
@@ -1401,8 +1413,8 @@ int zap_huge_pmd(struct mmu_gather *tlb, struct vm_area_struct *vma,
 			VM_BUG_ON(page_mapcount(page) < 0);
 			add_mm_counter(tlb->mm, MM_ANONPAGES, -HPAGE_PMD_NR);
 			VM_BUG_ON(!PageHead(page));
-			tlb->mm->nr_ptes--;
-			spin_unlock(&tlb->mm->page_table_lock);
+			atomic_long_dec(&tlb->mm->nr_ptes);
+			spin_unlock(ptl);
 			tlb_remove_page(tlb, page);
 		}
 		pte_free(tlb->mm, pgtable);
@@ -1415,14 +1427,15 @@ int mincore_huge_pmd(struct vm_area_struct *vma, pmd_t *pmd,
 		unsigned long addr, unsigned long end,
 		unsigned char *vec)
 {
+	spinlock_t *ptl;
 	int ret = 0;
 
-	if (__pmd_trans_huge_lock(pmd, vma) == 1) {
+	if (__pmd_trans_huge_lock(pmd, vma, &ptl) == 1) {
 		/*
 		 * All logical pages in the range are present
 		 * if backed by a huge page.
 		 */
-		spin_unlock(&vma->vm_mm->page_table_lock);
+		spin_unlock(ptl);
 		memset(vec, 1, (end - addr) >> PAGE_SHIFT);
 		ret = 1;
 	}
@@ -1435,6 +1448,7 @@ int move_huge_pmd(struct vm_area_struct *vma, struct vm_area_struct *new_vma,
 		  unsigned long new_addr, unsigned long old_end,
 		  pmd_t *old_pmd, pmd_t *new_pmd)
 {
+	spinlock_t *old_ptl, *new_ptl;
 	int ret = 0;
 	pmd_t pmd;
 
@@ -1455,12 +1469,21 @@ int move_huge_pmd(struct vm_area_struct *vma, struct vm_area_struct *new_vma,
 		goto out;
 	}
 
-	ret = __pmd_trans_huge_lock(old_pmd, vma);
+	/*
+	 * We don't have to worry about the ordering of src and dst
+	 * ptlocks because exclusive mmap_sem prevents deadlock.
+	 */
+	ret = __pmd_trans_huge_lock(old_pmd, vma, &old_ptl);
 	if (ret == 1) {
+		new_ptl = pmd_lockptr(mm, new_pmd);
+		if (new_ptl != old_ptl)
+			spin_lock_nested(new_ptl, SINGLE_DEPTH_NESTING);
 		pmd = pmdp_get_and_clear(mm, old_addr, old_pmd);
 		VM_BUG_ON(!pmd_none(*new_pmd));
 		set_pmd_at(mm, new_addr, new_pmd, pmd_mksoft_dirty(pmd));
-		spin_unlock(&mm->page_table_lock);
+		if (new_ptl != old_ptl)
+			spin_unlock(new_ptl);
+		spin_unlock(old_ptl);
 	}
 out:
 	return ret;
@@ -1476,9 +1499,10 @@ int change_huge_pmd(struct vm_area_struct *vma, pmd_t *pmd,
 		unsigned long addr, pgprot_t newprot, int prot_numa)
 {
 	struct mm_struct *mm = vma->vm_mm;
+	spinlock_t *ptl;
 	int ret = 0;
 
-	if (__pmd_trans_huge_lock(pmd, vma) == 1) {
+	if (__pmd_trans_huge_lock(pmd, vma, &ptl) == 1) {
 		pmd_t entry;
 		ret = 1;
 		if (!prot_numa) {
@@ -1507,7 +1531,7 @@ int change_huge_pmd(struct vm_area_struct *vma, pmd_t *pmd,
 		if (ret == HPAGE_PMD_NR)
 			set_pmd_at(mm, addr, pmd, entry);
 
-		spin_unlock(&vma->vm_mm->page_table_lock);
+		spin_unlock(ptl);
 	}
 
 	return ret;
@@ -1520,12 +1544,13 @@ int change_huge_pmd(struct vm_area_struct *vma, pmd_t *pmd,
  * Note that if it returns 1, this routine returns without unlocking page
  * table locks. So callers must unlock them.
  */
-int __pmd_trans_huge_lock(pmd_t *pmd, struct vm_area_struct *vma)
+int __pmd_trans_huge_lock(pmd_t *pmd, struct vm_area_struct *vma,
+		spinlock_t **ptl)
 {
-	spin_lock(&vma->vm_mm->page_table_lock);
+	*ptl = pmd_lock(vma->vm_mm, pmd);
 	if (likely(pmd_trans_huge(*pmd))) {
 		if (unlikely(pmd_trans_splitting(*pmd))) {
-			spin_unlock(&vma->vm_mm->page_table_lock);
+			spin_unlock(*ptl);
 			wait_split_huge_page(vma->anon_vma, pmd);
 			return -1;
 		} else {
@@ -1534,27 +1559,37 @@ int __pmd_trans_huge_lock(pmd_t *pmd, struct vm_area_struct *vma)
 			return 1;
 		}
 	}
-	spin_unlock(&vma->vm_mm->page_table_lock);
+	spin_unlock(*ptl);
 	return 0;
 }
 
+/*
+ * This function returns whether a given @page is mapped onto the @address
+ * in the virtual space of @mm.
+ *
+ * When it's true, this function returns *pmd with holding the page table lock
+ * and passing it back to the caller via @ptl.
+ * If it's false, returns NULL without holding the page table lock.
+ */
 pmd_t *page_check_address_pmd(struct page *page,
 			      struct mm_struct *mm,
 			      unsigned long address,
-			      enum page_check_address_pmd_flag flag)
+			      enum page_check_address_pmd_flag flag,
+			      spinlock_t **ptl)
 {
-	pmd_t *pmd, *ret = NULL;
+	pmd_t *pmd;
 
 	if (address & ~HPAGE_PMD_MASK)
-		goto out;
+		return NULL;
 
 	pmd = mm_find_pmd(mm, address);
 	if (!pmd)
-		goto out;
+		return NULL;
+	*ptl = pmd_lock(mm, pmd);
 	if (pmd_none(*pmd))
-		goto out;
+		goto unlock;
 	if (pmd_page(*pmd) != page)
-		goto out;
+		goto unlock;
 	/*
 	 * split_vma() may create temporary aliased mappings. There is
 	 * no risk as long as all huge pmd are found and have their
@@ -1564,14 +1599,15 @@ pmd_t *page_check_address_pmd(struct page *page,
 	 */
 	if (flag == PAGE_CHECK_ADDRESS_PMD_NOTSPLITTING_FLAG &&
 	    pmd_trans_splitting(*pmd))
-		goto out;
+		goto unlock;
 	if (pmd_trans_huge(*pmd)) {
 		VM_BUG_ON(flag == PAGE_CHECK_ADDRESS_PMD_SPLITTING_FLAG &&
 			  !pmd_trans_splitting(*pmd));
-		ret = pmd;
+		return pmd;
 	}
-out:
-	return ret;
+unlock:
+	spin_unlock(*ptl);
+	return NULL;
 }
 
 static int __split_huge_page_splitting(struct page *page,
@@ -1579,6 +1615,7 @@ static int __split_huge_page_splitting(struct page *page,
 				       unsigned long address)
 {
 	struct mm_struct *mm = vma->vm_mm;
+	spinlock_t *ptl;
 	pmd_t *pmd;
 	int ret = 0;
 	/* For mmu_notifiers */
@@ -1586,9 +1623,8 @@ static int __split_huge_page_splitting(struct page *page,
 	const unsigned long mmun_end   = address + HPAGE_PMD_SIZE;
 
 	mmu_notifier_invalidate_range_start(mm, mmun_start, mmun_end);
-	spin_lock(&mm->page_table_lock);
 	pmd = page_check_address_pmd(page, mm, address,
-				     PAGE_CHECK_ADDRESS_PMD_NOTSPLITTING_FLAG);
+			PAGE_CHECK_ADDRESS_PMD_NOTSPLITTING_FLAG, &ptl);
 	if (pmd) {
 		/*
 		 * We can't temporarily set the pmd to null in order
@@ -1599,8 +1635,8 @@ static int __split_huge_page_splitting(struct page *page,
 		 */
 		pmdp_splitting_flush(vma, address, pmd);
 		ret = 1;
+		spin_unlock(ptl);
 	}
-	spin_unlock(&mm->page_table_lock);
 	mmu_notifier_invalidate_range_end(mm, mmun_start, mmun_end);
 
 	return ret;
@@ -1731,14 +1767,14 @@ static int __split_huge_page_map(struct page *page,
 				 unsigned long address)
 {
 	struct mm_struct *mm = vma->vm_mm;
+	spinlock_t *ptl;
 	pmd_t *pmd, _pmd;
 	int ret = 0, i;
 	pgtable_t pgtable;
 	unsigned long haddr;
 
-	spin_lock(&mm->page_table_lock);
 	pmd = page_check_address_pmd(page, mm, address,
-				     PAGE_CHECK_ADDRESS_PMD_SPLITTING_FLAG);
+			PAGE_CHECK_ADDRESS_PMD_SPLITTING_FLAG, &ptl);
 	if (pmd) {
 		pgtable = pgtable_trans_huge_withdraw(mm, pmd);
 		pmd_populate(mm, &_pmd, pgtable);
@@ -1793,8 +1829,8 @@ static int __split_huge_page_map(struct page *page,
 		pmdp_invalidate(vma, address, pmd);
 		pmd_populate(mm, pmd, pgtable);
 		ret = 1;
+		spin_unlock(ptl);
 	}
-	spin_unlock(&mm->page_table_lock);
 
 	return ret;
 }
@@ -2346,7 +2382,7 @@ static void collapse_huge_page(struct mm_struct *mm,
 	pte_t *pte;
 	pgtable_t pgtable;
 	struct page *new_page;
-	spinlock_t *ptl;
+	spinlock_t *pmd_ptl, *pte_ptl;
 	int isolated;
 	unsigned long hstart, hend;
 	unsigned long mmun_start;	/* For mmu_notifiers */
@@ -2389,12 +2425,12 @@ static void collapse_huge_page(struct mm_struct *mm,
 	anon_vma_lock_write(vma->anon_vma);
 
 	pte = pte_offset_map(pmd, address);
-	ptl = pte_lockptr(mm, pmd);
+	pte_ptl = pte_lockptr(mm, pmd);
 
 	mmun_start = address;
 	mmun_end   = address + HPAGE_PMD_SIZE;
 	mmu_notifier_invalidate_range_start(mm, mmun_start, mmun_end);
-	spin_lock(&mm->page_table_lock); /* probably unnecessary */
+	pmd_ptl = pmd_lock(mm, pmd); /* probably unnecessary */
 	/*
 	 * After this gup_fast can't run anymore. This also removes
 	 * any huge TLB entry from the CPU so we won't allow
@@ -2402,16 +2438,16 @@ static void collapse_huge_page(struct mm_struct *mm,
 	 * to avoid the risk of CPU bugs in that area.
 	 */
 	_pmd = pmdp_clear_flush(vma, address, pmd);
-	spin_unlock(&mm->page_table_lock);
+	spin_unlock(pmd_ptl);
 	mmu_notifier_invalidate_range_end(mm, mmun_start, mmun_end);
 
-	spin_lock(ptl);
+	spin_lock(pte_ptl);
 	isolated = __collapse_huge_page_isolate(vma, address, pte);
-	spin_unlock(ptl);
+	spin_unlock(pte_ptl);
 
 	if (unlikely(!isolated)) {
 		pte_unmap(pte);
-		spin_lock(&mm->page_table_lock);
+		spin_lock(pmd_ptl);
 		BUG_ON(!pmd_none(*pmd));
 		/*
 		 * We can only use set_pmd_at when establishing
@@ -2419,7 +2455,7 @@ static void collapse_huge_page(struct mm_struct *mm,
 		 * points to regular pagetables. Use pmd_populate for that
 		 */
 		pmd_populate(mm, pmd, pmd_pgtable(_pmd));
-		spin_unlock(&mm->page_table_lock);
+		spin_unlock(pmd_ptl);
 		anon_vma_unlock_write(vma->anon_vma);
 		goto out;
 	}
@@ -2430,7 +2466,7 @@ static void collapse_huge_page(struct mm_struct *mm,
 	 */
 	anon_vma_unlock_write(vma->anon_vma);
 
-	__collapse_huge_page_copy(pte, new_page, vma, address, ptl);
+	__collapse_huge_page_copy(pte, new_page, vma, address, pte_ptl);
 	pte_unmap(pte);
 	__SetPageUptodate(new_page);
 	pgtable = pmd_pgtable(_pmd);
@@ -2445,13 +2481,13 @@ static void collapse_huge_page(struct mm_struct *mm,
 	 */
 	smp_wmb();
 
-	spin_lock(&mm->page_table_lock);
+	spin_lock(pmd_ptl);
 	BUG_ON(!pmd_none(*pmd));
 	page_add_new_anon_rmap(new_page, vma, address);
 	pgtable_trans_huge_deposit(mm, pmd, pgtable);
 	set_pmd_at(mm, address, pmd, _pmd);
 	update_mmu_cache_pmd(vma, address, pmd);
-	spin_unlock(&mm->page_table_lock);
+	spin_unlock(pmd_ptl);
 
 	*hpage = NULL;
 
@@ -2780,6 +2816,7 @@ static void __split_huge_zero_page_pmd(struct vm_area_struct *vma,
 void __split_huge_page_pmd(struct vm_area_struct *vma, unsigned long address,
 		pmd_t *pmd)
 {
+	spinlock_t *ptl;
 	struct page *page;
 	struct mm_struct *mm = vma->vm_mm;
 	unsigned long haddr = address & HPAGE_PMD_MASK;
@@ -2792,22 +2829,22 @@ void __split_huge_page_pmd(struct vm_area_struct *vma, unsigned long address,
 	mmun_end   = haddr + HPAGE_PMD_SIZE;
 again:
 	mmu_notifier_invalidate_range_start(mm, mmun_start, mmun_end);
-	spin_lock(&mm->page_table_lock);
+	ptl = pmd_lock(mm, pmd);
 	if (unlikely(!pmd_trans_huge(*pmd))) {
-		spin_unlock(&mm->page_table_lock);
+		spin_unlock(ptl);
 		mmu_notifier_invalidate_range_end(mm, mmun_start, mmun_end);
 		return;
 	}
 	if (is_huge_zero_pmd(*pmd)) {
 		__split_huge_zero_page_pmd(vma, haddr, pmd);
-		spin_unlock(&mm->page_table_lock);
+		spin_unlock(ptl);
 		mmu_notifier_invalidate_range_end(mm, mmun_start, mmun_end);
 		return;
 	}
 	page = pmd_page(*pmd);
 	VM_BUG_ON(!page_count(page));
 	get_page(page);
-	spin_unlock(&mm->page_table_lock);
+	spin_unlock(ptl);
 	mmu_notifier_invalidate_range_end(mm, mmun_start, mmun_end);
 
 	split_huge_page(page);
diff --git a/mm/hugetlb.c b/mm/hugetlb.c
index 0b7656e804d1..7d57af21f49e 100644
--- a/mm/hugetlb.c
+++ b/mm/hugetlb.c
@@ -2376,6 +2376,7 @@ int copy_hugetlb_page_range(struct mm_struct *dst, struct mm_struct *src,
 	cow = (vma->vm_flags & (VM_SHARED | VM_MAYWRITE)) == VM_MAYWRITE;
 
 	for (addr = vma->vm_start; addr < vma->vm_end; addr += sz) {
+		spinlock_t *src_ptl, *dst_ptl;
 		src_pte = huge_pte_offset(src, addr);
 		if (!src_pte)
 			continue;
@@ -2387,8 +2388,9 @@ int copy_hugetlb_page_range(struct mm_struct *dst, struct mm_struct *src,
 		if (dst_pte == src_pte)
 			continue;
 
-		spin_lock(&dst->page_table_lock);
-		spin_lock_nested(&src->page_table_lock, SINGLE_DEPTH_NESTING);
+		dst_ptl = huge_pte_lock(h, dst, dst_pte);
+		src_ptl = huge_pte_lockptr(h, src, src_pte);
+		spin_lock_nested(src_ptl, SINGLE_DEPTH_NESTING);
 		if (!huge_pte_none(huge_ptep_get(src_pte))) {
 			if (cow)
 				huge_ptep_set_wrprotect(src, addr, src_pte);
@@ -2398,8 +2400,8 @@ int copy_hugetlb_page_range(struct mm_struct *dst, struct mm_struct *src,
 			page_dup_rmap(ptepage);
 			set_huge_pte_at(dst, addr, dst_pte, entry);
 		}
-		spin_unlock(&src->page_table_lock);
-		spin_unlock(&dst->page_table_lock);
+		spin_unlock(src_ptl);
+		spin_unlock(dst_ptl);
 	}
 	return 0;
 
@@ -2442,6 +2444,7 @@ void __unmap_hugepage_range(struct mmu_gather *tlb, struct vm_area_struct *vma,
 	unsigned long address;
 	pte_t *ptep;
 	pte_t pte;
+	spinlock_t *ptl;
 	struct page *page;
 	struct hstate *h = hstate_vma(vma);
 	unsigned long sz = huge_page_size(h);
@@ -2455,25 +2458,25 @@ void __unmap_hugepage_range(struct mmu_gather *tlb, struct vm_area_struct *vma,
 	tlb_start_vma(tlb, vma);
 	mmu_notifier_invalidate_range_start(mm, mmun_start, mmun_end);
 again:
-	spin_lock(&mm->page_table_lock);
 	for (address = start; address < end; address += sz) {
 		ptep = huge_pte_offset(mm, address);
 		if (!ptep)
 			continue;
 
+		ptl = huge_pte_lock(h, mm, ptep);
 		if (huge_pmd_unshare(mm, &address, ptep))
-			continue;
+			goto unlock;
 
 		pte = huge_ptep_get(ptep);
 		if (huge_pte_none(pte))
-			continue;
+			goto unlock;
 
 		/*
 		 * HWPoisoned hugepage is already unmapped and dropped reference
 		 */
 		if (unlikely(is_hugetlb_entry_hwpoisoned(pte))) {
 			huge_pte_clear(mm, address, ptep);
-			continue;
+			goto unlock;
 		}
 
 		page = pte_page(pte);
@@ -2484,7 +2487,7 @@ again:
 		 */
 		if (ref_page) {
 			if (page != ref_page)
-				continue;
+				goto unlock;
 
 			/*
 			 * Mark the VMA as having unmapped its page so that
@@ -2501,13 +2504,18 @@ again:
 
 		page_remove_rmap(page);
 		force_flush = !__tlb_remove_page(tlb, page);
-		if (force_flush)
+		if (force_flush) {
+			spin_unlock(ptl);
 			break;
+		}
 		/* Bail out after unmapping reference page if supplied */
-		if (ref_page)
+		if (ref_page) {
+			spin_unlock(ptl);
 			break;
+		}
+unlock:
+		spin_unlock(ptl);
 	}
-	spin_unlock(&mm->page_table_lock);
 	/*
 	 * mmu_gather ran out of room to batch pages, we break out of
 	 * the PTE lock to avoid doing the potential expensive TLB invalidate
@@ -2613,7 +2621,7 @@ static int unmap_ref_private(struct mm_struct *mm, struct vm_area_struct *vma,
  */
 static int hugetlb_cow(struct mm_struct *mm, struct vm_area_struct *vma,
 			unsigned long address, pte_t *ptep, pte_t pte,
-			struct page *pagecache_page)
+			struct page *pagecache_page, spinlock_t *ptl)
 {
 	struct hstate *h = hstate_vma(vma);
 	struct page *old_page, *new_page;
@@ -2647,8 +2655,8 @@ retry_avoidcopy:
 
 	page_cache_get(old_page);
 
-	/* Drop page_table_lock as buddy allocator may be called */
-	spin_unlock(&mm->page_table_lock);
+	/* Drop page table lock as buddy allocator may be called */
+	spin_unlock(ptl);
 	new_page = alloc_huge_page(vma, address, outside_reserve);
 
 	if (IS_ERR(new_page)) {
@@ -2666,13 +2674,13 @@ retry_avoidcopy:
 			BUG_ON(huge_pte_none(pte));
 			if (unmap_ref_private(mm, vma, old_page, address)) {
 				BUG_ON(huge_pte_none(pte));
-				spin_lock(&mm->page_table_lock);
+				spin_lock(ptl);
 				ptep = huge_pte_offset(mm, address & huge_page_mask(h));
 				if (likely(pte_same(huge_ptep_get(ptep), pte)))
 					goto retry_avoidcopy;
 				/*
-				 * race occurs while re-acquiring page_table_lock, and
-				 * our job is done.
+				 * race occurs while re-acquiring page table
+				 * lock, and our job is done.
 				 */
 				return 0;
 			}
@@ -2680,7 +2688,7 @@ retry_avoidcopy:
 		}
 
 		/* Caller expects lock to be held */
-		spin_lock(&mm->page_table_lock);
+		spin_lock(ptl);
 		if (err == -ENOMEM)
 			return VM_FAULT_OOM;
 		else
@@ -2695,7 +2703,7 @@ retry_avoidcopy:
 		page_cache_release(new_page);
 		page_cache_release(old_page);
 		/* Caller expects lock to be held */
-		spin_lock(&mm->page_table_lock);
+		spin_lock(ptl);
 		return VM_FAULT_OOM;
 	}
 
@@ -2707,10 +2715,10 @@ retry_avoidcopy:
 	mmun_end = mmun_start + huge_page_size(h);
 	mmu_notifier_invalidate_range_start(mm, mmun_start, mmun_end);
 	/*
-	 * Retake the page_table_lock to check for racing updates
+	 * Retake the page table lock to check for racing updates
 	 * before the page tables are altered
 	 */
-	spin_lock(&mm->page_table_lock);
+	spin_lock(ptl);
 	ptep = huge_pte_offset(mm, address & huge_page_mask(h));
 	if (likely(pte_same(huge_ptep_get(ptep), pte))) {
 		ClearPagePrivate(new_page);
@@ -2724,13 +2732,13 @@ retry_avoidcopy:
 		/* Make the old page be freed below */
 		new_page = old_page;
 	}
-	spin_unlock(&mm->page_table_lock);
+	spin_unlock(ptl);
 	mmu_notifier_invalidate_range_end(mm, mmun_start, mmun_end);
 	page_cache_release(new_page);
 	page_cache_release(old_page);
 
 	/* Caller expects lock to be held */
-	spin_lock(&mm->page_table_lock);
+	spin_lock(ptl);
 	return 0;
 }
 
@@ -2778,6 +2786,7 @@ static int hugetlb_no_page(struct mm_struct *mm, struct vm_area_struct *vma,
 	struct page *page;
 	struct address_space *mapping;
 	pte_t new_pte;
+	spinlock_t *ptl;
 
 	/*
 	 * Currently, we are forced to kill the process in the event the
@@ -2864,7 +2873,8 @@ retry:
 			goto backout_unlocked;
 		}
 
-	spin_lock(&mm->page_table_lock);
+	ptl = huge_pte_lockptr(h, mm, ptep);
+	spin_lock(ptl);
 	size = i_size_read(mapping->host) >> huge_page_shift(h);
 	if (idx >= size)
 		goto backout;
@@ -2885,16 +2895,16 @@ retry:
 
 	if ((flags & FAULT_FLAG_WRITE) && !(vma->vm_flags & VM_SHARED)) {
 		/* Optimization, do the COW without a second fault */
-		ret = hugetlb_cow(mm, vma, address, ptep, new_pte, page);
+		ret = hugetlb_cow(mm, vma, address, ptep, new_pte, page, ptl);
 	}
 
-	spin_unlock(&mm->page_table_lock);
+	spin_unlock(ptl);
 	unlock_page(page);
 out:
 	return ret;
 
 backout:
-	spin_unlock(&mm->page_table_lock);
+	spin_unlock(ptl);
 backout_unlocked:
 	unlock_page(page);
 	put_page(page);
@@ -2906,6 +2916,7 @@ int hugetlb_fault(struct mm_struct *mm, struct vm_area_struct *vma,
 {
 	pte_t *ptep;
 	pte_t entry;
+	spinlock_t *ptl;
 	int ret;
 	struct page *page = NULL;
 	struct page *pagecache_page = NULL;
@@ -2918,7 +2929,7 @@ int hugetlb_fault(struct mm_struct *mm, struct vm_area_struct *vma,
 	if (ptep) {
 		entry = huge_ptep_get(ptep);
 		if (unlikely(is_hugetlb_entry_migration(entry))) {
-			migration_entry_wait_huge(mm, ptep);
+			migration_entry_wait_huge(vma, mm, ptep);
 			return 0;
 		} else if (unlikely(is_hugetlb_entry_hwpoisoned(entry)))
 			return VM_FAULT_HWPOISON_LARGE |
@@ -2974,17 +2985,18 @@ int hugetlb_fault(struct mm_struct *mm, struct vm_area_struct *vma,
 	if (page != pagecache_page)
 		lock_page(page);
 
-	spin_lock(&mm->page_table_lock);
+	ptl = huge_pte_lockptr(h, mm, ptep);
+	spin_lock(ptl);
 	/* Check for a racing update before calling hugetlb_cow */
 	if (unlikely(!pte_same(entry, huge_ptep_get(ptep))))
-		goto out_page_table_lock;
+		goto out_ptl;
 
 
 	if (flags & FAULT_FLAG_WRITE) {
 		if (!huge_pte_write(entry)) {
 			ret = hugetlb_cow(mm, vma, address, ptep, entry,
-							pagecache_page);
-			goto out_page_table_lock;
+					pagecache_page, ptl);
+			goto out_ptl;
 		}
 		entry = huge_pte_mkdirty(entry);
 	}
@@ -2993,8 +3005,8 @@ int hugetlb_fault(struct mm_struct *mm, struct vm_area_struct *vma,
 						flags & FAULT_FLAG_WRITE))
 		update_mmu_cache(vma, address, ptep);
 
-out_page_table_lock:
-	spin_unlock(&mm->page_table_lock);
+out_ptl:
+	spin_unlock(ptl);
 
 	if (pagecache_page) {
 		unlock_page(pagecache_page);
@@ -3020,9 +3032,9 @@ long follow_hugetlb_page(struct mm_struct *mm, struct vm_area_struct *vma,
 	unsigned long remainder = *nr_pages;
 	struct hstate *h = hstate_vma(vma);
 
-	spin_lock(&mm->page_table_lock);
 	while (vaddr < vma->vm_end && remainder) {
 		pte_t *pte;
+		spinlock_t *ptl = NULL;
 		int absent;
 		struct page *page;
 
@@ -3030,8 +3042,12 @@ long follow_hugetlb_page(struct mm_struct *mm, struct vm_area_struct *vma,
 		 * Some archs (sparc64, sh*) have multiple pte_ts to
 		 * each hugepage.  We have to make sure we get the
 		 * first, for the page indexing below to work.
+		 *
+		 * Note that page table lock is not held when pte is null.
 		 */
 		pte = huge_pte_offset(mm, vaddr & huge_page_mask(h));
+		if (pte)
+			ptl = huge_pte_lock(h, mm, pte);
 		absent = !pte || huge_pte_none(huge_ptep_get(pte));
 
 		/*
@@ -3043,6 +3059,8 @@ long follow_hugetlb_page(struct mm_struct *mm, struct vm_area_struct *vma,
 		 */
 		if (absent && (flags & FOLL_DUMP) &&
 		    !hugetlbfs_pagecache_present(h, vma, vaddr)) {
+			if (pte)
+				spin_unlock(ptl);
 			remainder = 0;
 			break;
 		}
@@ -3062,10 +3080,10 @@ long follow_hugetlb_page(struct mm_struct *mm, struct vm_area_struct *vma,
 		      !huge_pte_write(huge_ptep_get(pte)))) {
 			int ret;
 
-			spin_unlock(&mm->page_table_lock);
+			if (pte)
+				spin_unlock(ptl);
 			ret = hugetlb_fault(mm, vma, vaddr,
 				(flags & FOLL_WRITE) ? FAULT_FLAG_WRITE : 0);
-			spin_lock(&mm->page_table_lock);
 			if (!(ret & VM_FAULT_ERROR))
 				continue;
 
@@ -3096,8 +3114,8 @@ same_page:
 			 */
 			goto same_page;
 		}
+		spin_unlock(ptl);
 	}
-	spin_unlock(&mm->page_table_lock);
 	*nr_pages = remainder;
 	*position = vaddr;
 
@@ -3118,13 +3136,15 @@ unsigned long hugetlb_change_protection(struct vm_area_struct *vma,
 	flush_cache_range(vma, address, end);
 
 	mutex_lock(&vma->vm_file->f_mapping->i_mmap_mutex);
-	spin_lock(&mm->page_table_lock);
 	for (; address < end; address += huge_page_size(h)) {
+		spinlock_t *ptl;
 		ptep = huge_pte_offset(mm, address);
 		if (!ptep)
 			continue;
+		ptl = huge_pte_lock(h, mm, ptep);
 		if (huge_pmd_unshare(mm, &address, ptep)) {
 			pages++;
+			spin_unlock(ptl);
 			continue;
 		}
 		if (!huge_pte_none(huge_ptep_get(ptep))) {
@@ -3134,8 +3154,8 @@ unsigned long hugetlb_change_protection(struct vm_area_struct *vma,
 			set_huge_pte_at(mm, address, ptep, pte);
 			pages++;
 		}
+		spin_unlock(ptl);
 	}
-	spin_unlock(&mm->page_table_lock);
 	/*
 	 * Must flush TLB before releasing i_mmap_mutex: x86's huge_pmd_unshare
 	 * may have cleared our pud entry and done put_page on the page table:
@@ -3298,6 +3318,7 @@ pte_t *huge_pmd_share(struct mm_struct *mm, unsigned long addr, pud_t *pud)
 	unsigned long saddr;
 	pte_t *spte = NULL;
 	pte_t *pte;
+	spinlock_t *ptl;
 
 	if (!vma_shareable(vma, addr))
 		return (pte_t *)pmd_alloc(mm, pud, addr);
@@ -3320,13 +3341,14 @@ pte_t *huge_pmd_share(struct mm_struct *mm, unsigned long addr, pud_t *pud)
 	if (!spte)
 		goto out;
 
-	spin_lock(&mm->page_table_lock);
+	ptl = huge_pte_lockptr(hstate_vma(vma), mm, spte);
+	spin_lock(ptl);
 	if (pud_none(*pud))
 		pud_populate(mm, pud,
 				(pmd_t *)((unsigned long)spte & PAGE_MASK));
 	else
 		put_page(virt_to_page(spte));
-	spin_unlock(&mm->page_table_lock);
+	spin_unlock(ptl);
 out:
 	pte = (pte_t *)pmd_alloc(mm, pud, addr);
 	mutex_unlock(&mapping->i_mmap_mutex);
@@ -3340,7 +3362,7 @@ out:
  * indicated by page_count > 1, unmap is achieved by clearing pud and
  * decrementing the ref count. If count == 1, the pte page is not shared.
  *
- * called with vma->vm_mm->page_table_lock held.
+ * called with page table lock held.
  *
  * returns: 1 successfully unmapped a shared pte page
  *	    0 the underlying pte page is not shared, or it is the last user
diff --git a/mm/memcontrol.c b/mm/memcontrol.c
index e3cd40b2d5d9..f1a0ae6e11b8 100644
--- a/mm/memcontrol.c
+++ b/mm/memcontrol.c
@@ -6605,10 +6605,10 @@ static int mem_cgroup_count_precharge_pte_range(pmd_t *pmd,
 	pte_t *pte;
 	spinlock_t *ptl;
 
-	if (pmd_trans_huge_lock(pmd, vma) == 1) {
+	if (pmd_trans_huge_lock(pmd, vma, &ptl) == 1) {
 		if (get_mctgt_type_thp(vma, addr, *pmd, NULL) == MC_TARGET_PAGE)
 			mc.precharge += HPAGE_PMD_NR;
-		spin_unlock(&vma->vm_mm->page_table_lock);
+		spin_unlock(ptl);
 		return 0;
 	}
 
@@ -6797,9 +6797,9 @@ static int mem_cgroup_move_charge_pte_range(pmd_t *pmd,
 	 *    to be unlocked in __split_huge_page_splitting(), where the main
 	 *    part of thp split is not executed yet.
 	 */
-	if (pmd_trans_huge_lock(pmd, vma) == 1) {
+	if (pmd_trans_huge_lock(pmd, vma, &ptl) == 1) {
 		if (mc.precharge < HPAGE_PMD_NR) {
-			spin_unlock(&vma->vm_mm->page_table_lock);
+			spin_unlock(ptl);
 			return 0;
 		}
 		target_type = get_mctgt_type_thp(vma, addr, *pmd, &target);
@@ -6816,7 +6816,7 @@ static int mem_cgroup_move_charge_pte_range(pmd_t *pmd,
 			}
 			put_page(page);
 		}
-		spin_unlock(&vma->vm_mm->page_table_lock);
+		spin_unlock(ptl);
 		return 0;
 	}
 
diff --git a/mm/memory-failure.c b/mm/memory-failure.c
index f9d78ec7831f..b7c171602ba1 100644
--- a/mm/memory-failure.c
+++ b/mm/memory-failure.c
@@ -1269,7 +1269,7 @@ void memory_failure_queue(unsigned long pfn, int trapno, int flags)
 
 	mf_cpu = &get_cpu_var(memory_failure_cpu);
 	spin_lock_irqsave(&mf_cpu->lock, proc_flags);
-	if (kfifo_put(&mf_cpu->fifo, &entry))
+	if (kfifo_put(&mf_cpu->fifo, entry))
 		schedule_work_on(smp_processor_id(), &mf_cpu->work);
 	else
 		pr_err("Memory failure: buffer overflow when queuing memory failure at %#lx\n",
diff --git a/mm/memory.c b/mm/memory.c
index bf8665849a5f..0409e8f43fa0 100644
--- a/mm/memory.c
+++ b/mm/memory.c
@@ -382,7 +382,7 @@ static void free_pte_range(struct mmu_gather *tlb, pmd_t *pmd,
 	pgtable_t token = pmd_pgtable(*pmd);
 	pmd_clear(pmd);
 	pte_free_tlb(tlb, token, addr);
-	tlb->mm->nr_ptes--;
+	atomic_long_dec(&tlb->mm->nr_ptes);
 }
 
 static inline void free_pmd_range(struct mmu_gather *tlb, pud_t *pud,
@@ -550,6 +550,7 @@ void free_pgtables(struct mmu_gather *tlb, struct vm_area_struct *vma,
 int __pte_alloc(struct mm_struct *mm, struct vm_area_struct *vma,
 		pmd_t *pmd, unsigned long address)
 {
+	spinlock_t *ptl;
 	pgtable_t new = pte_alloc_one(mm, address);
 	int wait_split_huge_page;
 	if (!new)
@@ -570,15 +571,15 @@ int __pte_alloc(struct mm_struct *mm, struct vm_area_struct *vma,
 	 */
 	smp_wmb(); /* Could be smp_wmb__xxx(before|after)_spin_lock */
 
-	spin_lock(&mm->page_table_lock);
+	ptl = pmd_lock(mm, pmd);
 	wait_split_huge_page = 0;
 	if (likely(pmd_none(*pmd))) {	/* Has another populated it ? */
-		mm->nr_ptes++;
+		atomic_long_inc(&mm->nr_ptes);
 		pmd_populate(mm, pmd, new);
 		new = NULL;
 	} else if (unlikely(pmd_trans_splitting(*pmd)))
 		wait_split_huge_page = 1;
-	spin_unlock(&mm->page_table_lock);
+	spin_unlock(ptl);
 	if (new)
 		pte_free(mm, new);
 	if (wait_split_huge_page)
@@ -1516,20 +1517,20 @@ struct page *follow_page_mask(struct vm_area_struct *vma,
 			split_huge_page_pmd(vma, address, pmd);
 			goto split_fallthrough;
 		}
-		spin_lock(&mm->page_table_lock);
+		ptl = pmd_lock(mm, pmd);
 		if (likely(pmd_trans_huge(*pmd))) {
 			if (unlikely(pmd_trans_splitting(*pmd))) {
-				spin_unlock(&mm->page_table_lock);
+				spin_unlock(ptl);
 				wait_split_huge_page(vma->anon_vma, pmd);
 			} else {
 				page = follow_trans_huge_pmd(vma, address,
 							     pmd, flags);
-				spin_unlock(&mm->page_table_lock);
+				spin_unlock(ptl);
 				*page_mask = HPAGE_PMD_NR - 1;
 				goto out;
 			}
 		} else
-			spin_unlock(&mm->page_table_lock);
+			spin_unlock(ptl);
 		/* fall through */
 	}
 split_fallthrough:
@@ -4269,3 +4270,28 @@ void copy_user_huge_page(struct page *dst, struct page *src,
 	}
 }
 #endif /* CONFIG_TRANSPARENT_HUGEPAGE || CONFIG_HUGETLBFS */
+
+#if USE_SPLIT_PTE_PTLOCKS && BLOATED_SPINLOCKS
+static struct kmem_cache *page_ptl_cachep;
+void __init ptlock_cache_init(void)
+{
+	page_ptl_cachep = kmem_cache_create("page->ptl", sizeof(spinlock_t), 0,
+			SLAB_PANIC, NULL);
+}
+
+bool ptlock_alloc(struct page *page)
+{
+	spinlock_t *ptl;
+
+	ptl = kmalloc(sizeof(spinlock_t), GFP_KERNEL);
+	if (!ptl)
+		return false;
+	page->ptl = ptl;
+	return true;
+}
+
+void ptlock_free(struct page *page)
+{
+	kfree(page->ptl);
+}
+#endif
diff --git a/mm/mempolicy.c b/mm/mempolicy.c
index 4cc19f6ab6c6..c4403cdf3433 100644
--- a/mm/mempolicy.c
+++ b/mm/mempolicy.c
@@ -525,8 +525,9 @@ static void queue_pages_hugetlb_pmd_range(struct vm_area_struct *vma,
 #ifdef CONFIG_HUGETLB_PAGE
 	int nid;
 	struct page *page;
+	spinlock_t *ptl;
 
-	spin_lock(&vma->vm_mm->page_table_lock);
+	ptl = huge_pte_lock(hstate_vma(vma), vma->vm_mm, (pte_t *)pmd);
 	page = pte_page(huge_ptep_get((pte_t *)pmd));
 	nid = page_to_nid(page);
 	if (node_isset(nid, *nodes) == !!(flags & MPOL_MF_INVERT))
@@ -536,7 +537,7 @@ static void queue_pages_hugetlb_pmd_range(struct vm_area_struct *vma,
 	    (flags & MPOL_MF_MOVE && page_mapcount(page) == 1))
 		isolate_huge_page(page, private);
 unlock:
-	spin_unlock(&vma->vm_mm->page_table_lock);
+	spin_unlock(ptl);
 #else
 	BUG();
 #endif
diff --git a/mm/migrate.c b/mm/migrate.c
index dfc8300ecbb2..316e720a2023 100644
--- a/mm/migrate.c
+++ b/mm/migrate.c
@@ -130,7 +130,7 @@ static int remove_migration_pte(struct page *new, struct vm_area_struct *vma,
 		ptep = huge_pte_offset(mm, addr);
 		if (!ptep)
 			goto out;
-		ptl = &mm->page_table_lock;
+		ptl = huge_pte_lockptr(hstate_vma(vma), mm, ptep);
 	} else {
 		pmd = mm_find_pmd(mm, addr);
 		if (!pmd)
@@ -249,9 +249,10 @@ void migration_entry_wait(struct mm_struct *mm, pmd_t *pmd,
 	__migration_entry_wait(mm, ptep, ptl);
 }
 
-void migration_entry_wait_huge(struct mm_struct *mm, pte_t *pte)
+void migration_entry_wait_huge(struct vm_area_struct *vma,
+		struct mm_struct *mm, pte_t *pte)
 {
-	spinlock_t *ptl = &(mm)->page_table_lock;
+	spinlock_t *ptl = huge_pte_lockptr(hstate_vma(vma), mm, pte);
 	__migration_entry_wait(mm, pte, ptl);
 }
 
@@ -1666,6 +1667,7 @@ int migrate_misplaced_transhuge_page(struct mm_struct *mm,
 				unsigned long address,
 				struct page *page, int node)
 {
+	spinlock_t *ptl;
 	unsigned long haddr = address & HPAGE_PMD_MASK;
 	pg_data_t *pgdat = NODE_DATA(node);
 	int isolated = 0;
@@ -1705,9 +1707,9 @@ int migrate_misplaced_transhuge_page(struct mm_struct *mm,
 	WARN_ON(PageLRU(new_page));
 
 	/* Recheck the target PMD */
-	spin_lock(&mm->page_table_lock);
+	ptl = pmd_lock(mm, pmd);
 	if (unlikely(!pmd_same(*pmd, entry))) {
-		spin_unlock(&mm->page_table_lock);
+		spin_unlock(ptl);
 
 		/* Reverse changes made by migrate_page_copy() */
 		if (TestClearPageActive(new_page))
@@ -1752,7 +1754,7 @@ int migrate_misplaced_transhuge_page(struct mm_struct *mm,
 	 * before it's fully transferred to the new page.
 	 */
 	mem_cgroup_end_migration(memcg, page, new_page, true);
-	spin_unlock(&mm->page_table_lock);
+	spin_unlock(ptl);
 
 	unlock_page(new_page);
 	unlock_page(page);
diff --git a/mm/mmap.c b/mm/mmap.c
index 5a6baddde15d..834b2d785f1e 100644
--- a/mm/mmap.c
+++ b/mm/mmap.c
@@ -2724,7 +2724,8 @@ void exit_mmap(struct mm_struct *mm)
 	}
 	vm_unacct_memory(nr_accounted);
 
-	WARN_ON(mm->nr_ptes > (FIRST_USER_ADDRESS+PMD_SIZE-1)>>PMD_SHIFT);
+	WARN_ON(atomic_long_read(&mm->nr_ptes) >
+			(FIRST_USER_ADDRESS+PMD_SIZE-1)>>PMD_SHIFT);
 }
 
 /* Insert vm structure into process list sorted by address
diff --git a/mm/oom_kill.c b/mm/oom_kill.c
index 6738c47f1f72..1e4a600a6163 100644
--- a/mm/oom_kill.c
+++ b/mm/oom_kill.c
@@ -161,7 +161,7 @@ unsigned long oom_badness(struct task_struct *p, struct mem_cgroup *memcg,
 	 * The baseline for the badness score is the proportion of RAM that each
 	 * task's rss, pagetable and swap space use.
 	 */
-	points = get_mm_rss(p->mm) + p->mm->nr_ptes +
+	points = get_mm_rss(p->mm) + atomic_long_read(&p->mm->nr_ptes) +
 		 get_mm_counter(p->mm, MM_SWAPENTS);
 	task_unlock(p);
 
@@ -364,10 +364,10 @@ static void dump_tasks(const struct mem_cgroup *memcg, const nodemask_t *nodemas
 			continue;
 		}
 
-		pr_info("[%5d] %5d %5d %8lu %8lu %7lu %8lu         %5hd %s\n",
+		pr_info("[%5d] %5d %5d %8lu %8lu %7ld %8lu         %5hd %s\n",
 			task->pid, from_kuid(&init_user_ns, task_uid(task)),
 			task->tgid, task->mm->total_vm, get_mm_rss(task->mm),
-			task->mm->nr_ptes,
+			atomic_long_read(&task->mm->nr_ptes),
 			get_mm_counter(task->mm, MM_SWAPENTS),
 			task->signal->oom_score_adj, task->comm);
 		task_unlock(task);
diff --git a/mm/pgtable-generic.c b/mm/pgtable-generic.c
index 3929a40bd6c0..cbb38545d9d6 100644
--- a/mm/pgtable-generic.c
+++ b/mm/pgtable-generic.c
@@ -151,14 +151,14 @@ void pmdp_splitting_flush(struct vm_area_struct *vma, unsigned long address,
 void pgtable_trans_huge_deposit(struct mm_struct *mm, pmd_t *pmdp,
 				pgtable_t pgtable)
 {
-	assert_spin_locked(&mm->page_table_lock);
+	assert_spin_locked(pmd_lockptr(mm, pmdp));
 
 	/* FIFO */
-	if (!mm->pmd_huge_pte)
+	if (!pmd_huge_pte(mm, pmdp))
 		INIT_LIST_HEAD(&pgtable->lru);
 	else
-		list_add(&pgtable->lru, &mm->pmd_huge_pte->lru);
-	mm->pmd_huge_pte = pgtable;
+		list_add(&pgtable->lru, &pmd_huge_pte(mm, pmdp)->lru);
+	pmd_huge_pte(mm, pmdp) = pgtable;
 }
 #endif /* CONFIG_TRANSPARENT_HUGEPAGE */
 #endif
@@ -170,14 +170,14 @@ pgtable_t pgtable_trans_huge_withdraw(struct mm_struct *mm, pmd_t *pmdp)
 {
 	pgtable_t pgtable;
 
-	assert_spin_locked(&mm->page_table_lock);
+	assert_spin_locked(pmd_lockptr(mm, pmdp));
 
 	/* FIFO */
-	pgtable = mm->pmd_huge_pte;
+	pgtable = pmd_huge_pte(mm, pmdp);
 	if (list_empty(&pgtable->lru))
-		mm->pmd_huge_pte = NULL;
+		pmd_huge_pte(mm, pmdp) = NULL;
 	else {
-		mm->pmd_huge_pte = list_entry(pgtable->lru.next,
+		pmd_huge_pte(mm, pmdp) = list_entry(pgtable->lru.next,
 					      struct page, lru);
 		list_del(&pgtable->lru);
 	}
diff --git a/mm/rmap.c b/mm/rmap.c
index fd3ee7a54a13..55c8b8dc9ffb 100644
--- a/mm/rmap.c
+++ b/mm/rmap.c
@@ -601,7 +601,7 @@ pte_t *__page_check_address(struct page *page, struct mm_struct *mm,
 
 	if (unlikely(PageHuge(page))) {
 		pte = huge_pte_offset(mm, address);
-		ptl = &mm->page_table_lock;
+		ptl = huge_pte_lockptr(page_hstate(page), mm, pte);
 		goto check;
 	}
 
@@ -665,25 +665,23 @@ int page_referenced_one(struct page *page, struct vm_area_struct *vma,
 			unsigned long *vm_flags)
 {
 	struct mm_struct *mm = vma->vm_mm;
+	spinlock_t *ptl;
 	int referenced = 0;
 
 	if (unlikely(PageTransHuge(page))) {
 		pmd_t *pmd;
 
-		spin_lock(&mm->page_table_lock);
 		/*
 		 * rmap might return false positives; we must filter
 		 * these out using page_check_address_pmd().
 		 */
 		pmd = page_check_address_pmd(page, mm, address,
-					     PAGE_CHECK_ADDRESS_PMD_FLAG);
-		if (!pmd) {
-			spin_unlock(&mm->page_table_lock);
+					     PAGE_CHECK_ADDRESS_PMD_FLAG, &ptl);
+		if (!pmd)
 			goto out;
-		}
 
 		if (vma->vm_flags & VM_LOCKED) {
-			spin_unlock(&mm->page_table_lock);
+			spin_unlock(ptl);
 			*mapcount = 0;	/* break early from loop */
 			*vm_flags |= VM_LOCKED;
 			goto out;
@@ -692,10 +690,9 @@ int page_referenced_one(struct page *page, struct vm_area_struct *vma,
 		/* go ahead even if the pmd is pmd_trans_splitting() */
 		if (pmdp_clear_flush_young_notify(vma, address, pmd))
 			referenced++;
-		spin_unlock(&mm->page_table_lock);
+		spin_unlock(ptl);
 	} else {
 		pte_t *pte;
-		spinlock_t *ptl;
 
 		/*
 		 * rmap might return false positives; we must filter
diff --git a/mm/slub.c b/mm/slub.c
index 92737a0b787b..7e8bd8d828bc 100644
--- a/mm/slub.c
+++ b/mm/slub.c
@@ -955,7 +955,7 @@ static inline void slab_free_hook(struct kmem_cache *s, void *x)
 	kmemleak_free_recursive(x, s->flags);
 
 	/*
-	 * Trouble is that we may no longer disable interupts in the fast path
+	 * Trouble is that we may no longer disable interrupts in the fast path
 	 * So in order to make the debug calls that expect irqs to be
 	 * disabled we need to disable interrupts temporarily.
 	 */
diff --git a/net/8021q/vlan_dev.c b/net/8021q/vlan_dev.c
index 8db1b985dbf1..762896ebfcf5 100644
--- a/net/8021q/vlan_dev.c
+++ b/net/8021q/vlan_dev.c
@@ -539,7 +539,7 @@ static const struct net_device_ops vlan_netdev_ops;
 static int vlan_dev_init(struct net_device *dev)
 {
 	struct net_device *real_dev = vlan_dev_priv(dev)->real_dev;
-	int subclass = 0;
+	int subclass = 0, i;
 
 	netif_carrier_off(dev);
 
@@ -593,6 +593,13 @@ static int vlan_dev_init(struct net_device *dev)
 	if (!vlan_dev_priv(dev)->vlan_pcpu_stats)
 		return -ENOMEM;
 
+	for_each_possible_cpu(i) {
+		struct vlan_pcpu_stats *vlan_stat;
+		vlan_stat = per_cpu_ptr(vlan_dev_priv(dev)->vlan_pcpu_stats, i);
+		u64_stats_init(&vlan_stat->syncp);
+	}
+
+
 	return 0;
 }
 
diff --git a/net/9p/trans_virtio.c b/net/9p/trans_virtio.c
index 990afab2be1b..9c5a1aa34d12 100644
--- a/net/9p/trans_virtio.c
+++ b/net/9p/trans_virtio.c
@@ -544,9 +544,7 @@ static int p9_virtio_probe(struct virtio_device *vdev)
 
 	chan->inuse = false;
 	if (virtio_has_feature(vdev, VIRTIO_9P_MOUNT_TAG)) {
-		vdev->config->get(vdev,
-				offsetof(struct virtio_9p_config, tag_len),
-				&tag_len, sizeof(tag_len));
+		virtio_cread(vdev, struct virtio_9p_config, tag_len, &tag_len);
 	} else {
 		err = -EINVAL;
 		goto out_free_vq;
@@ -556,8 +554,9 @@ static int p9_virtio_probe(struct virtio_device *vdev)
 		err = -ENOMEM;
 		goto out_free_vq;
 	}
-	vdev->config->get(vdev, offsetof(struct virtio_9p_config, tag),
-			tag, tag_len);
+
+	virtio_cread_bytes(vdev, offsetof(struct virtio_9p_config, tag),
+			   tag, tag_len);
 	chan->tag = tag;
 	chan->tag_len = tag_len;
 	err = sysfs_create_file(&(vdev->dev.kobj), &dev_attr_mount_tag.attr);
diff --git a/net/bridge/br_device.c b/net/bridge/br_device.c
index e6b7fecb3af1..f00cfd2a0143 100644
--- a/net/bridge/br_device.c
+++ b/net/bridge/br_device.c
@@ -88,11 +88,18 @@ out:
 static int br_dev_init(struct net_device *dev)
 {
 	struct net_bridge *br = netdev_priv(dev);
+	int i;
 
 	br->stats = alloc_percpu(struct br_cpu_netstats);
 	if (!br->stats)
 		return -ENOMEM;
 
+	for_each_possible_cpu(i) {
+		struct br_cpu_netstats *br_dev_stats;
+		br_dev_stats = per_cpu_ptr(br->stats, i);
+		u64_stats_init(&br_dev_stats->syncp);
+	}
+
 	return 0;
 }
 
diff --git a/net/bridge/br_if.c b/net/bridge/br_if.c
index c41d5fbb91d0..6e6194fcd88e 100644
--- a/net/bridge/br_if.c
+++ b/net/bridge/br_if.c
@@ -172,6 +172,7 @@ void br_dev_delete(struct net_device *dev, struct list_head *head)
 		del_nbp(p);
 	}
 
+	br_vlan_flush(br);
 	del_timer_sync(&br->gc_timer);
 
 	br_sysfs_delbr(br->dev);
diff --git a/net/bridge/br_vlan.c b/net/bridge/br_vlan.c
index 53f0990eab58..af5ebd18d705 100644
--- a/net/bridge/br_vlan.c
+++ b/net/bridge/br_vlan.c
@@ -34,7 +34,6 @@ static void __vlan_add_flags(struct net_port_vlans *v, u16 vid, u16 flags)
 
 static int __vlan_add(struct net_port_vlans *v, u16 vid, u16 flags)
 {
-	const struct net_device_ops *ops;
 	struct net_bridge_port *p = NULL;
 	struct net_bridge *br;
 	struct net_device *dev;
@@ -53,17 +52,15 @@ static int __vlan_add(struct net_port_vlans *v, u16 vid, u16 flags)
 		br = v->parent.br;
 		dev = br->dev;
 	}
-	ops = dev->netdev_ops;
 
-	if (p && (dev->features & NETIF_F_HW_VLAN_CTAG_FILTER)) {
+	if (p) {
 		/* Add VLAN to the device filter if it is supported.
 		 * Stricly speaking, this is not necessary now, since
 		 * devices are made promiscuous by the bridge, but if
 		 * that ever changes this code will allow tagged
 		 * traffic to enter the bridge.
 		 */
-		err = ops->ndo_vlan_rx_add_vid(dev, htons(ETH_P_8021Q),
-					       vid);
+		err = vlan_vid_add(dev, htons(ETH_P_8021Q), vid);
 		if (err)
 			return err;
 	}
@@ -82,8 +79,8 @@ static int __vlan_add(struct net_port_vlans *v, u16 vid, u16 flags)
 	return 0;
 
 out_filt:
-	if (p && (dev->features & NETIF_F_HW_VLAN_CTAG_FILTER))
-		ops->ndo_vlan_rx_kill_vid(dev, htons(ETH_P_8021Q), vid);
+	if (p)
+		vlan_vid_del(dev, htons(ETH_P_8021Q), vid);
 	return err;
 }
 
@@ -95,13 +92,8 @@ static int __vlan_del(struct net_port_vlans *v, u16 vid)
 	__vlan_delete_pvid(v, vid);
 	clear_bit(vid, v->untagged_bitmap);
 
-	if (v->port_idx) {
-		struct net_device *dev = v->parent.port->dev;
-		const struct net_device_ops *ops = dev->netdev_ops;
-
-		if (dev->features & NETIF_F_HW_VLAN_CTAG_FILTER)
-			ops->ndo_vlan_rx_kill_vid(dev, htons(ETH_P_8021Q), vid);
-	}
+	if (v->port_idx)
+		vlan_vid_del(v->parent.port->dev, htons(ETH_P_8021Q), vid);
 
 	clear_bit(vid, v->vlan_bitmap);
 	v->num_vlans--;
@@ -398,6 +390,7 @@ int nbp_vlan_delete(struct net_bridge_port *port, u16 vid)
 void nbp_vlan_flush(struct net_bridge_port *port)
 {
 	struct net_port_vlans *pv;
+	u16 vid;
 
 	ASSERT_RTNL();
 
@@ -405,6 +398,9 @@ void nbp_vlan_flush(struct net_bridge_port *port)
 	if (!pv)
 		return;
 
+	for_each_set_bit(vid, pv->vlan_bitmap, VLAN_N_VID)
+		vlan_vid_del(port->dev, htons(ETH_P_8021Q), vid);
+
 	__vlan_flush(pv);
 }
 
diff --git a/net/can/af_can.c b/net/can/af_can.c
index 3ab8dd2e1282..d249874a366d 100644
--- a/net/can/af_can.c
+++ b/net/can/af_can.c
@@ -420,7 +420,7 @@ static struct hlist_head *find_rcv_list(canid_t *can_id, canid_t *mask,
  * @mask: CAN mask (see description)
  * @func: callback function on filter match
  * @data: returned parameter for callback function
- * @ident: string for calling module indentification
+ * @ident: string for calling module identification
  *
  * Description:
  *  Invokes the callback function with the received sk_buff and the given
diff --git a/net/core/dev.c b/net/core/dev.c
index 8ffc52e01ece..7e00a7342ee6 100644
--- a/net/core/dev.c
+++ b/net/core/dev.c
@@ -131,6 +131,7 @@
 #include <linux/static_key.h>
 #include <linux/hashtable.h>
 #include <linux/vmalloc.h>
+#include <linux/if_macvlan.h>
 
 #include "net-sysfs.h"
 
@@ -1424,6 +1425,10 @@ void dev_disable_lro(struct net_device *dev)
 	if (is_vlan_dev(dev))
 		dev = vlan_dev_real_dev(dev);
 
+	/* the same for macvlan devices */
+	if (netif_is_macvlan(dev))
+		dev = macvlan_dev_real_dev(dev);
+
 	dev->wanted_features &= ~NETIF_F_LRO;
 	netdev_update_features(dev);
 
@@ -1690,13 +1695,9 @@ int dev_forward_skb(struct net_device *dev, struct sk_buff *skb)
 		kfree_skb(skb);
 		return NET_RX_DROP;
 	}
-	skb->protocol = eth_type_trans(skb, dev);
 
-	/* eth_type_trans() can set pkt_type.
-	 * call skb_scrub_packet() after it to clear pkt_type _after_ calling
-	 * eth_type_trans().
-	 */
 	skb_scrub_packet(skb, true);
+	skb->protocol = eth_type_trans(skb, dev);
 
 	return netif_rx(skb);
 }
diff --git a/net/core/drop_monitor.c b/net/core/drop_monitor.c
index 5e78d44333b9..95897183226e 100644
--- a/net/core/drop_monitor.c
+++ b/net/core/drop_monitor.c
@@ -106,6 +106,10 @@ static struct sk_buff *reset_per_cpu_data(struct per_cpu_dm_data *data)
 	return skb;
 }
 
+static struct genl_multicast_group dropmon_mcgrps[] = {
+	{ .name = "events", },
+};
+
 static void send_dm_alert(struct work_struct *work)
 {
 	struct sk_buff *skb;
@@ -116,7 +120,8 @@ static void send_dm_alert(struct work_struct *work)
 	skb = reset_per_cpu_data(data);
 
 	if (skb)
-		genlmsg_multicast(skb, 0, NET_DM_GRP_ALERT, GFP_KERNEL);
+		genlmsg_multicast(&net_drop_monitor_family, skb, 0,
+				  0, GFP_KERNEL);
 }
 
 /*
@@ -333,7 +338,7 @@ out:
 	return NOTIFY_DONE;
 }
 
-static struct genl_ops dropmon_ops[] = {
+static const struct genl_ops dropmon_ops[] = {
 	{
 		.cmd = NET_DM_CMD_CONFIG,
 		.doit = net_dm_cmd_config,
@@ -364,13 +369,13 @@ static int __init init_net_drop_monitor(void)
 		return -ENOSPC;
 	}
 
-	rc = genl_register_family_with_ops(&net_drop_monitor_family,
-					   dropmon_ops,
-					   ARRAY_SIZE(dropmon_ops));
+	rc = genl_register_family_with_ops_groups(&net_drop_monitor_family,
+						  dropmon_ops, dropmon_mcgrps);
 	if (rc) {
 		pr_err("Could not create drop monitor netlink family\n");
 		return rc;
 	}
+	WARN_ON(net_drop_monitor_family.mcgrp_offset != NET_DM_GRP_ALERT);
 
 	rc = register_netdevice_notifier(&dropmon_net_notifier);
 	if (rc < 0) {
diff --git a/net/hsr/hsr_netlink.c b/net/hsr/hsr_netlink.c
index 4e66bf61f585..5325af85eea6 100644
--- a/net/hsr/hsr_netlink.c
+++ b/net/hsr/hsr_netlink.c
@@ -90,8 +90,8 @@ static struct genl_family hsr_genl_family = {
 	.maxattr = HSR_A_MAX,
 };
 
-static struct genl_multicast_group hsr_network_genl_mcgrp = {
-	.name = "hsr-network",
+static const struct genl_multicast_group hsr_mcgrps[] = {
+	{ .name = "hsr-network", },
 };
 
 
@@ -129,7 +129,7 @@ void hsr_nl_ringerror(struct hsr_priv *hsr_priv, unsigned char addr[ETH_ALEN],
 		goto nla_put_failure;
 
 	genlmsg_end(skb, msg_head);
-	genlmsg_multicast(skb, 0, hsr_network_genl_mcgrp.id, GFP_ATOMIC);
+	genlmsg_multicast(&hsr_genl_family, skb, 0, 0, GFP_ATOMIC);
 
 	return;
 
@@ -163,7 +163,7 @@ void hsr_nl_nodedown(struct hsr_priv *hsr_priv, unsigned char addr[ETH_ALEN])
 		goto nla_put_failure;
 
 	genlmsg_end(skb, msg_head);
-	genlmsg_multicast(skb, 0, hsr_network_genl_mcgrp.id, GFP_ATOMIC);
+	genlmsg_multicast(&hsr_genl_family, skb, 0, 0, GFP_ATOMIC);
 
 	return;
 
@@ -249,7 +249,7 @@ static int hsr_get_node_status(struct sk_buff *skb_in, struct genl_info *info)
 			&hsr_node_if2_age,
 			&hsr_node_if2_seq);
 	if (res < 0)
-		goto fail;
+		goto nla_put_failure;
 
 	res = nla_put(skb_out, HSR_A_NODE_ADDR, ETH_ALEN,
 					nla_data(info->attrs[HSR_A_NODE_ADDR]));
@@ -306,15 +306,6 @@ fail:
 	return res;
 }
 
-static struct genl_ops hsr_ops_get_node_status = {
-	.cmd = HSR_C_GET_NODE_STATUS,
-	.flags = 0,
-	.policy = hsr_genl_policy,
-	.doit = hsr_get_node_status,
-	.dumpit = NULL,
-};
-
-
 /* Get a list of MacAddressA of all nodes known to this node (other than self).
  */
 static int hsr_get_node_list(struct sk_buff *skb_in, struct genl_info *info)
@@ -398,12 +389,21 @@ fail:
 }
 
 
-static struct genl_ops hsr_ops_get_node_list = {
-	.cmd = HSR_C_GET_NODE_LIST,
-	.flags = 0,
-	.policy = hsr_genl_policy,
-	.doit = hsr_get_node_list,
-	.dumpit = NULL,
+static const struct genl_ops hsr_ops[] = {
+	{
+		.cmd = HSR_C_GET_NODE_STATUS,
+		.flags = 0,
+		.policy = hsr_genl_policy,
+		.doit = hsr_get_node_status,
+		.dumpit = NULL,
+	},
+	{
+		.cmd = HSR_C_GET_NODE_LIST,
+		.flags = 0,
+		.policy = hsr_genl_policy,
+		.doit = hsr_get_node_list,
+		.dumpit = NULL,
+	},
 };
 
 int __init hsr_netlink_init(void)
@@ -414,30 +414,13 @@ int __init hsr_netlink_init(void)
 	if (rc)
 		goto fail_rtnl_link_register;
 
-	rc = genl_register_family(&hsr_genl_family);
+	rc = genl_register_family_with_ops_groups(&hsr_genl_family, hsr_ops,
+						  hsr_mcgrps);
 	if (rc)
 		goto fail_genl_register_family;
 
-	rc = genl_register_ops(&hsr_genl_family, &hsr_ops_get_node_status);
-	if (rc)
-		goto fail_genl_register_ops;
-
-	rc = genl_register_ops(&hsr_genl_family, &hsr_ops_get_node_list);
-	if (rc)
-		goto fail_genl_register_ops_node_list;
-
-	rc = genl_register_mc_group(&hsr_genl_family, &hsr_network_genl_mcgrp);
-	if (rc)
-		goto fail_genl_register_mc_group;
-
 	return 0;
 
-fail_genl_register_mc_group:
-	genl_unregister_ops(&hsr_genl_family, &hsr_ops_get_node_list);
-fail_genl_register_ops_node_list:
-	genl_unregister_ops(&hsr_genl_family, &hsr_ops_get_node_status);
-fail_genl_register_ops:
-	genl_unregister_family(&hsr_genl_family);
 fail_genl_register_family:
 	rtnl_link_unregister(&hsr_link_ops);
 fail_rtnl_link_register:
@@ -447,10 +430,7 @@ fail_rtnl_link_register:
 
 void __exit hsr_netlink_exit(void)
 {
-	genl_unregister_mc_group(&hsr_genl_family, &hsr_network_genl_mcgrp);
-	genl_unregister_ops(&hsr_genl_family, &hsr_ops_get_node_status);
 	genl_unregister_family(&hsr_genl_family);
-
 	rtnl_link_unregister(&hsr_link_ops);
 }
 
diff --git a/net/ieee802154/6lowpan.c b/net/ieee802154/6lowpan.c
index 426b5df1c98f..459e200c08a4 100644
--- a/net/ieee802154/6lowpan.c
+++ b/net/ieee802154/6lowpan.c
@@ -956,7 +956,7 @@ lowpan_process_data(struct sk_buff *skb)
 	 * Traffic class carried in-line
 	 * ECN + DSCP (1 byte), Flow Label is elided
 	 */
-	case 1: /* 10b */
+	case 2: /* 10b */
 		if (lowpan_fetch_skb_u8(skb, &tmp))
 			goto drop;
 
@@ -967,7 +967,7 @@ lowpan_process_data(struct sk_buff *skb)
 	 * Flow Label carried in-line
 	 * ECN + 2-bit Pad + Flow Label (3 bytes), DSCP is elided
 	 */
-	case 2: /* 01b */
+	case 1: /* 01b */
 		if (lowpan_fetch_skb_u8(skb, &tmp))
 			goto drop;
 
diff --git a/net/ieee802154/dgram.c b/net/ieee802154/dgram.c
index 581a59504bd5..1865fdf5a5a5 100644
--- a/net/ieee802154/dgram.c
+++ b/net/ieee802154/dgram.c
@@ -315,9 +315,8 @@ static int dgram_recvmsg(struct kiocb *iocb, struct sock *sk,
 	if (saddr) {
 		saddr->family = AF_IEEE802154;
 		saddr->addr = mac_cb(skb)->sa;
-	}
-	if (addr_len)
 		*addr_len = sizeof(*saddr);
+	}
 
 	if (flags & MSG_TRUNC)
 		copied = skb->len;
diff --git a/net/ieee802154/ieee802154.h b/net/ieee802154/ieee802154.h
index aadec428e6ec..cee4425b9956 100644
--- a/net/ieee802154/ieee802154.h
+++ b/net/ieee802154/ieee802154.h
@@ -47,7 +47,24 @@ struct sk_buff *ieee802154_nl_new_reply(struct genl_info *info,
 int ieee802154_nl_reply(struct sk_buff *msg, struct genl_info *info);
 
 extern struct genl_family nl802154_family;
-int nl802154_mac_register(void);
-int nl802154_phy_register(void);
+
+/* genetlink ops/groups */
+int ieee802154_list_phy(struct sk_buff *skb, struct genl_info *info);
+int ieee802154_dump_phy(struct sk_buff *skb, struct netlink_callback *cb);
+int ieee802154_add_iface(struct sk_buff *skb, struct genl_info *info);
+int ieee802154_del_iface(struct sk_buff *skb, struct genl_info *info);
+
+enum ieee802154_mcgrp_ids {
+	IEEE802154_COORD_MCGRP,
+	IEEE802154_BEACON_MCGRP,
+};
+
+int ieee802154_associate_req(struct sk_buff *skb, struct genl_info *info);
+int ieee802154_associate_resp(struct sk_buff *skb, struct genl_info *info);
+int ieee802154_disassociate_req(struct sk_buff *skb, struct genl_info *info);
+int ieee802154_scan_req(struct sk_buff *skb, struct genl_info *info);
+int ieee802154_start_req(struct sk_buff *skb, struct genl_info *info);
+int ieee802154_list_iface(struct sk_buff *skb, struct genl_info *info);
+int ieee802154_dump_iface(struct sk_buff *skb, struct netlink_callback *cb);
 
 #endif
diff --git a/net/ieee802154/netlink.c b/net/ieee802154/netlink.c
index 7e49bbcc6967..43f1b2bf469f 100644
--- a/net/ieee802154/netlink.c
+++ b/net/ieee802154/netlink.c
@@ -70,7 +70,7 @@ int ieee802154_nl_mcast(struct sk_buff *msg, unsigned int group)
 	if (genlmsg_end(msg, hdr) < 0)
 		goto out;
 
-	return genlmsg_multicast(msg, 0, group, GFP_ATOMIC);
+	return genlmsg_multicast(&nl802154_family, msg, 0, group, GFP_ATOMIC);
 out:
 	nlmsg_free(msg);
 	return -ENOBUFS;
@@ -109,31 +109,36 @@ out:
 	return -ENOBUFS;
 }
 
-int __init ieee802154_nl_init(void)
-{
-	int rc;
-
-	rc = genl_register_family(&nl802154_family);
-	if (rc)
-		goto fail;
-
-	rc = nl802154_mac_register();
-	if (rc)
-		goto fail;
+static const struct genl_ops ieee8021154_ops[] = {
+	/* see nl-phy.c */
+	IEEE802154_DUMP(IEEE802154_LIST_PHY, ieee802154_list_phy,
+			ieee802154_dump_phy),
+	IEEE802154_OP(IEEE802154_ADD_IFACE, ieee802154_add_iface),
+	IEEE802154_OP(IEEE802154_DEL_IFACE, ieee802154_del_iface),
+	/* see nl-mac.c */
+	IEEE802154_OP(IEEE802154_ASSOCIATE_REQ, ieee802154_associate_req),
+	IEEE802154_OP(IEEE802154_ASSOCIATE_RESP, ieee802154_associate_resp),
+	IEEE802154_OP(IEEE802154_DISASSOCIATE_REQ, ieee802154_disassociate_req),
+	IEEE802154_OP(IEEE802154_SCAN_REQ, ieee802154_scan_req),
+	IEEE802154_OP(IEEE802154_START_REQ, ieee802154_start_req),
+	IEEE802154_DUMP(IEEE802154_LIST_IFACE, ieee802154_list_iface,
+			ieee802154_dump_iface),
+};
 
-	rc = nl802154_phy_register();
-	if (rc)
-		goto fail;
+static const struct genl_multicast_group ieee802154_mcgrps[] = {
+	[IEEE802154_COORD_MCGRP] = { .name = IEEE802154_MCAST_COORD_NAME, },
+	[IEEE802154_BEACON_MCGRP] = { .name = IEEE802154_MCAST_BEACON_NAME, },
+};
 
-	return 0;
 
-fail:
-	genl_unregister_family(&nl802154_family);
-	return rc;
+int __init ieee802154_nl_init(void)
+{
+	return genl_register_family_with_ops_groups(&nl802154_family,
+						    ieee8021154_ops,
+						    ieee802154_mcgrps);
 }
 
 void __exit ieee802154_nl_exit(void)
 {
 	genl_unregister_family(&nl802154_family);
 }
-
diff --git a/net/ieee802154/nl-mac.c b/net/ieee802154/nl-mac.c
index b0bdd8c51e9c..ba5c1e002f37 100644
--- a/net/ieee802154/nl-mac.c
+++ b/net/ieee802154/nl-mac.c
@@ -39,14 +39,6 @@
 
 #include "ieee802154.h"
 
-static struct genl_multicast_group ieee802154_coord_mcgrp = {
-	.name		= IEEE802154_MCAST_COORD_NAME,
-};
-
-static struct genl_multicast_group ieee802154_beacon_mcgrp = {
-	.name		= IEEE802154_MCAST_BEACON_NAME,
-};
-
 int ieee802154_nl_assoc_indic(struct net_device *dev,
 		struct ieee802154_addr *addr, u8 cap)
 {
@@ -72,7 +64,7 @@ int ieee802154_nl_assoc_indic(struct net_device *dev,
 	    nla_put_u8(msg, IEEE802154_ATTR_CAPABILITY, cap))
 		goto nla_put_failure;
 
-	return ieee802154_nl_mcast(msg, ieee802154_coord_mcgrp.id);
+	return ieee802154_nl_mcast(msg, IEEE802154_COORD_MCGRP);
 
 nla_put_failure:
 	nlmsg_free(msg);
@@ -98,7 +90,7 @@ int ieee802154_nl_assoc_confirm(struct net_device *dev, u16 short_addr,
 	    nla_put_u16(msg, IEEE802154_ATTR_SHORT_ADDR, short_addr) ||
 	    nla_put_u8(msg, IEEE802154_ATTR_STATUS, status))
 		goto nla_put_failure;
-	return ieee802154_nl_mcast(msg, ieee802154_coord_mcgrp.id);
+	return ieee802154_nl_mcast(msg, IEEE802154_COORD_MCGRP);
 
 nla_put_failure:
 	nlmsg_free(msg);
@@ -133,7 +125,7 @@ int ieee802154_nl_disassoc_indic(struct net_device *dev,
 	}
 	if (nla_put_u8(msg, IEEE802154_ATTR_REASON, reason))
 		goto nla_put_failure;
-	return ieee802154_nl_mcast(msg, ieee802154_coord_mcgrp.id);
+	return ieee802154_nl_mcast(msg, IEEE802154_COORD_MCGRP);
 
 nla_put_failure:
 	nlmsg_free(msg);
@@ -157,7 +149,7 @@ int ieee802154_nl_disassoc_confirm(struct net_device *dev, u8 status)
 		    dev->dev_addr) ||
 	    nla_put_u8(msg, IEEE802154_ATTR_STATUS, status))
 		goto nla_put_failure;
-	return ieee802154_nl_mcast(msg, ieee802154_coord_mcgrp.id);
+	return ieee802154_nl_mcast(msg, IEEE802154_COORD_MCGRP);
 
 nla_put_failure:
 	nlmsg_free(msg);
@@ -183,7 +175,7 @@ int ieee802154_nl_beacon_indic(struct net_device *dev,
 	    nla_put_u16(msg, IEEE802154_ATTR_COORD_SHORT_ADDR, coord_addr) ||
 	    nla_put_u16(msg, IEEE802154_ATTR_COORD_PAN_ID, panid))
 		goto nla_put_failure;
-	return ieee802154_nl_mcast(msg, ieee802154_coord_mcgrp.id);
+	return ieee802154_nl_mcast(msg, IEEE802154_COORD_MCGRP);
 
 nla_put_failure:
 	nlmsg_free(msg);
@@ -214,7 +206,7 @@ int ieee802154_nl_scan_confirm(struct net_device *dev,
 	    (edl &&
 	     nla_put(msg, IEEE802154_ATTR_ED_LIST, 27, edl)))
 		goto nla_put_failure;
-	return ieee802154_nl_mcast(msg, ieee802154_coord_mcgrp.id);
+	return ieee802154_nl_mcast(msg, IEEE802154_COORD_MCGRP);
 
 nla_put_failure:
 	nlmsg_free(msg);
@@ -238,7 +230,7 @@ int ieee802154_nl_start_confirm(struct net_device *dev, u8 status)
 		    dev->dev_addr) ||
 	    nla_put_u8(msg, IEEE802154_ATTR_STATUS, status))
 		goto nla_put_failure;
-	return ieee802154_nl_mcast(msg, ieee802154_coord_mcgrp.id);
+	return ieee802154_nl_mcast(msg, IEEE802154_COORD_MCGRP);
 
 nla_put_failure:
 	nlmsg_free(msg);
@@ -309,8 +301,7 @@ static struct net_device *ieee802154_nl_get_dev(struct genl_info *info)
 	return dev;
 }
 
-static int ieee802154_associate_req(struct sk_buff *skb,
-		struct genl_info *info)
+int ieee802154_associate_req(struct sk_buff *skb, struct genl_info *info)
 {
 	struct net_device *dev;
 	struct ieee802154_addr addr;
@@ -357,8 +348,7 @@ out:
 	return ret;
 }
 
-static int ieee802154_associate_resp(struct sk_buff *skb,
-		struct genl_info *info)
+int ieee802154_associate_resp(struct sk_buff *skb, struct genl_info *info)
 {
 	struct net_device *dev;
 	struct ieee802154_addr addr;
@@ -390,8 +380,7 @@ out:
 	return ret;
 }
 
-static int ieee802154_disassociate_req(struct sk_buff *skb,
-		struct genl_info *info)
+int ieee802154_disassociate_req(struct sk_buff *skb, struct genl_info *info)
 {
 	struct net_device *dev;
 	struct ieee802154_addr addr;
@@ -433,7 +422,7 @@ out:
  * PAN_coordinator, battery_life_extension = 0,
  * coord_realignment = 0, security_enable = 0
 */
-static int ieee802154_start_req(struct sk_buff *skb, struct genl_info *info)
+int ieee802154_start_req(struct sk_buff *skb, struct genl_info *info)
 {
 	struct net_device *dev;
 	struct ieee802154_addr addr;
@@ -492,7 +481,7 @@ out:
 	return ret;
 }
 
-static int ieee802154_scan_req(struct sk_buff *skb, struct genl_info *info)
+int ieee802154_scan_req(struct sk_buff *skb, struct genl_info *info)
 {
 	struct net_device *dev;
 	int ret = -EOPNOTSUPP;
@@ -530,8 +519,7 @@ out:
 	return ret;
 }
 
-static int ieee802154_list_iface(struct sk_buff *skb,
-	struct genl_info *info)
+int ieee802154_list_iface(struct sk_buff *skb, struct genl_info *info)
 {
 	/* Request for interface name, index, type, IEEE address,
 	   PAN Id, short address */
@@ -565,8 +553,7 @@ out_dev:
 
 }
 
-static int ieee802154_dump_iface(struct sk_buff *skb,
-	struct netlink_callback *cb)
+int ieee802154_dump_iface(struct sk_buff *skb, struct netlink_callback *cb)
 {
 	struct net *net = sock_net(skb->sk);
 	struct net_device *dev;
@@ -590,41 +577,3 @@ cont:
 
 	return skb->len;
 }
-
-static struct genl_ops ieee802154_coordinator_ops[] = {
-	IEEE802154_OP(IEEE802154_ASSOCIATE_REQ, ieee802154_associate_req),
-	IEEE802154_OP(IEEE802154_ASSOCIATE_RESP, ieee802154_associate_resp),
-	IEEE802154_OP(IEEE802154_DISASSOCIATE_REQ, ieee802154_disassociate_req),
-	IEEE802154_OP(IEEE802154_SCAN_REQ, ieee802154_scan_req),
-	IEEE802154_OP(IEEE802154_START_REQ, ieee802154_start_req),
-	IEEE802154_DUMP(IEEE802154_LIST_IFACE, ieee802154_list_iface,
-							ieee802154_dump_iface),
-};
-
-/*
- * No need to unregister as family unregistration will do it.
- */
-int nl802154_mac_register(void)
-{
-	int i;
-	int rc;
-
-	rc = genl_register_mc_group(&nl802154_family,
-			&ieee802154_coord_mcgrp);
-	if (rc)
-		return rc;
-
-	rc = genl_register_mc_group(&nl802154_family,
-			&ieee802154_beacon_mcgrp);
-	if (rc)
-		return rc;
-
-	for (i = 0; i < ARRAY_SIZE(ieee802154_coordinator_ops); i++) {
-		rc = genl_register_ops(&nl802154_family,
-				&ieee802154_coordinator_ops[i]);
-		if (rc)
-			return rc;
-	}
-
-	return 0;
-}
diff --git a/net/ieee802154/nl-phy.c b/net/ieee802154/nl-phy.c
index 22b1a7058fd3..d08c7a43dcd1 100644
--- a/net/ieee802154/nl-phy.c
+++ b/net/ieee802154/nl-phy.c
@@ -77,8 +77,7 @@ out:
 	return -EMSGSIZE;
 }
 
-static int ieee802154_list_phy(struct sk_buff *skb,
-	struct genl_info *info)
+int ieee802154_list_phy(struct sk_buff *skb, struct genl_info *info)
 {
 	/* Request for interface name, index, type, IEEE address,
 	   PAN Id, short address */
@@ -151,8 +150,7 @@ static int ieee802154_dump_phy_iter(struct wpan_phy *phy, void *_data)
 	return 0;
 }
 
-static int ieee802154_dump_phy(struct sk_buff *skb,
-	struct netlink_callback *cb)
+int ieee802154_dump_phy(struct sk_buff *skb, struct netlink_callback *cb)
 {
 	struct dump_phy_data data = {
 		.cb = cb,
@@ -170,8 +168,7 @@ static int ieee802154_dump_phy(struct sk_buff *skb,
 	return skb->len;
 }
 
-static int ieee802154_add_iface(struct sk_buff *skb,
-		struct genl_info *info)
+int ieee802154_add_iface(struct sk_buff *skb, struct genl_info *info)
 {
 	struct sk_buff *msg;
 	struct wpan_phy *phy;
@@ -273,8 +270,7 @@ out_dev:
 	return rc;
 }
 
-static int ieee802154_del_iface(struct sk_buff *skb,
-		struct genl_info *info)
+int ieee802154_del_iface(struct sk_buff *skb, struct genl_info *info)
 {
 	struct sk_buff *msg;
 	struct wpan_phy *phy;
@@ -356,28 +352,3 @@ out_dev:
 
 	return rc;
 }
-
-static struct genl_ops ieee802154_phy_ops[] = {
-	IEEE802154_DUMP(IEEE802154_LIST_PHY, ieee802154_list_phy,
-							ieee802154_dump_phy),
-	IEEE802154_OP(IEEE802154_ADD_IFACE, ieee802154_add_iface),
-	IEEE802154_OP(IEEE802154_DEL_IFACE, ieee802154_del_iface),
-};
-
-/*
- * No need to unregister as family unregistration will do it.
- */
-int nl802154_phy_register(void)
-{
-	int i;
-	int rc;
-
-	for (i = 0; i < ARRAY_SIZE(ieee802154_phy_ops); i++) {
-		rc = genl_register_ops(&nl802154_family,
-				&ieee802154_phy_ops[i]);
-		if (rc)
-			return rc;
-	}
-
-	return 0;
-}
diff --git a/net/ipv4/af_inet.c b/net/ipv4/af_inet.c
index 68af9aac91d0..70011e029ac1 100644
--- a/net/ipv4/af_inet.c
+++ b/net/ipv4/af_inet.c
@@ -1503,6 +1503,7 @@ int snmp_mib_init(void __percpu *ptr[2], size_t mibsize, size_t align)
 	ptr[0] = __alloc_percpu(mibsize, align);
 	if (!ptr[0])
 		return -ENOMEM;
+
 #if SNMP_ARRAY_SZ == 2
 	ptr[1] = __alloc_percpu(mibsize, align);
 	if (!ptr[1]) {
@@ -1547,6 +1548,8 @@ static const struct net_protocol icmp_protocol = {
 
 static __net_init int ipv4_mib_init_net(struct net *net)
 {
+	int i;
+
 	if (snmp_mib_init((void __percpu **)net->mib.tcp_statistics,
 			  sizeof(struct tcp_mib),
 			  __alignof__(struct tcp_mib)) < 0)
@@ -1555,6 +1558,17 @@ static __net_init int ipv4_mib_init_net(struct net *net)
 			  sizeof(struct ipstats_mib),
 			  __alignof__(struct ipstats_mib)) < 0)
 		goto err_ip_mib;
+
+	for_each_possible_cpu(i) {
+		struct ipstats_mib *af_inet_stats;
+		af_inet_stats = per_cpu_ptr(net->mib.ip_statistics[0], i);
+		u64_stats_init(&af_inet_stats->syncp);
+#if SNMP_ARRAY_SZ == 2
+		af_inet_stats = per_cpu_ptr(net->mib.ip_statistics[1], i);
+		u64_stats_init(&af_inet_stats->syncp);
+#endif
+	}
+
 	if (snmp_mib_init((void __percpu **)net->mib.net_statistics,
 			  sizeof(struct linux_mib),
 			  __alignof__(struct linux_mib)) < 0)
diff --git a/net/ipv4/datagram.c b/net/ipv4/datagram.c
index b28e863fe0a7..19e36376d2a0 100644
--- a/net/ipv4/datagram.c
+++ b/net/ipv4/datagram.c
@@ -57,7 +57,7 @@ int ip4_datagram_connect(struct sock *sk, struct sockaddr *uaddr, int addr_len)
 	if (IS_ERR(rt)) {
 		err = PTR_ERR(rt);
 		if (err == -ENETUNREACH)
-			IP_INC_STATS_BH(sock_net(sk), IPSTATS_MIB_OUTNOROUTES);
+			IP_INC_STATS(sock_net(sk), IPSTATS_MIB_OUTNOROUTES);
 		goto out;
 	}
 
diff --git a/net/ipv4/fib_trie.c b/net/ipv4/fib_trie.c
index ec9a9ef4ce50..5afeb5aa4c7c 100644
--- a/net/ipv4/fib_trie.c
+++ b/net/ipv4/fib_trie.c
@@ -2523,16 +2523,17 @@ static int fib_route_seq_show(struct seq_file *seq, void *v)
 		list_for_each_entry_rcu(fa, &li->falh, fa_list) {
 			const struct fib_info *fi = fa->fa_info;
 			unsigned int flags = fib_flag_trans(fa->fa_type, mask, fi);
-			int len;
 
 			if (fa->fa_type == RTN_BROADCAST
 			    || fa->fa_type == RTN_MULTICAST)
 				continue;
 
+			seq_setwidth(seq, 127);
+
 			if (fi)
 				seq_printf(seq,
 					 "%s\t%08X\t%08X\t%04X\t%d\t%u\t"
-					 "%d\t%08X\t%d\t%u\t%u%n",
+					 "%d\t%08X\t%d\t%u\t%u",
 					 fi->fib_dev ? fi->fib_dev->name : "*",
 					 prefix,
 					 fi->fib_nh->nh_gw, flags, 0, 0,
@@ -2541,15 +2542,15 @@ static int fib_route_seq_show(struct seq_file *seq, void *v)
 					 (fi->fib_advmss ?
 					  fi->fib_advmss + 40 : 0),
 					 fi->fib_window,
-					 fi->fib_rtt >> 3, &len);
+					 fi->fib_rtt >> 3);
 			else
 				seq_printf(seq,
 					 "*\t%08X\t%08X\t%04X\t%d\t%u\t"
-					 "%d\t%08X\t%d\t%u\t%u%n",
+					 "%d\t%08X\t%d\t%u\t%u",
 					 prefix, 0, flags, 0, 0, 0,
-					 mask, 0, 0, 0, &len);
+					 mask, 0, 0, 0);
 
-			seq_printf(seq, "%*s\n", 127 - len, "");
+			seq_pad(seq, '\n');
 		}
 	}
 
diff --git a/net/ipv4/ip_tunnel.c b/net/ipv4/ip_tunnel.c
index 63a6d6d6b875..90ff9570d7d4 100644
--- a/net/ipv4/ip_tunnel.c
+++ b/net/ipv4/ip_tunnel.c
@@ -454,6 +454,8 @@ int ip_tunnel_rcv(struct ip_tunnel *tunnel, struct sk_buff *skb,
 	tstats->rx_bytes += skb->len;
 	u64_stats_update_end(&tstats->syncp);
 
+	skb_scrub_packet(skb, !net_eq(tunnel->net, dev_net(tunnel->dev)));
+
 	if (tunnel->dev->type == ARPHRD_ETHER) {
 		skb->protocol = eth_type_trans(skb, tunnel->dev);
 		skb_postpull_rcsum(skb, eth_hdr(skb), ETH_HLEN);
@@ -461,8 +463,6 @@ int ip_tunnel_rcv(struct ip_tunnel *tunnel, struct sk_buff *skb,
 		skb->dev = tunnel->dev;
 	}
 
-	skb_scrub_packet(skb, !net_eq(tunnel->net, dev_net(tunnel->dev)));
-
 	gro_cells_receive(&tunnel->gro_cells, skb);
 	return 0;
 
@@ -976,13 +976,19 @@ int ip_tunnel_init(struct net_device *dev)
 {
 	struct ip_tunnel *tunnel = netdev_priv(dev);
 	struct iphdr *iph = &tunnel->parms.iph;
-	int err;
+	int i, err;
 
 	dev->destructor	= ip_tunnel_dev_free;
 	dev->tstats = alloc_percpu(struct pcpu_tstats);
 	if (!dev->tstats)
 		return -ENOMEM;
 
+	for_each_possible_cpu(i) {
+		struct pcpu_tstats *ipt_stats;
+		ipt_stats = per_cpu_ptr(dev->tstats, i);
+		u64_stats_init(&ipt_stats->syncp);
+	}
+
 	err = gro_cells_init(&tunnel->gro_cells, dev);
 	if (err) {
 		free_percpu(dev->tstats);
diff --git a/net/ipv4/ip_vti.c b/net/ipv4/ip_vti.c
index 5d9c845d288a..52b802a0cd8c 100644
--- a/net/ipv4/ip_vti.c
+++ b/net/ipv4/ip_vti.c
@@ -126,6 +126,7 @@ static netdev_tx_t vti_tunnel_xmit(struct sk_buff *skb, struct net_device *dev)
 	if (!rt->dst.xfrm ||
 	    rt->dst.xfrm->props.mode != XFRM_MODE_TUNNEL) {
 		dev->stats.tx_carrier_errors++;
+		ip_rt_put(rt);
 		goto tx_error_icmp;
 	}
 	tdev = rt->dst.dev;
diff --git a/net/ipv4/ping.c b/net/ipv4/ping.c
index 9afbdb19f4a2..876c6ca2d8f9 100644
--- a/net/ipv4/ping.c
+++ b/net/ipv4/ping.c
@@ -830,8 +830,6 @@ int ping_recvmsg(struct kiocb *iocb, struct sock *sk, struct msghdr *msg,
 {
 	struct inet_sock *isk = inet_sk(sk);
 	int family = sk->sk_family;
-	struct sockaddr_in *sin;
-	struct sockaddr_in6 *sin6;
 	struct sk_buff *skb;
 	int copied, err;
 
@@ -841,13 +839,6 @@ int ping_recvmsg(struct kiocb *iocb, struct sock *sk, struct msghdr *msg,
 	if (flags & MSG_OOB)
 		goto out;
 
-	if (addr_len) {
-		if (family == AF_INET)
-			*addr_len = sizeof(*sin);
-		else if (family == AF_INET6 && addr_len)
-			*addr_len = sizeof(*sin6);
-	}
-
 	if (flags & MSG_ERRQUEUE) {
 		if (family == AF_INET) {
 			return ip_recv_error(sk, msg, len);
@@ -877,11 +868,15 @@ int ping_recvmsg(struct kiocb *iocb, struct sock *sk, struct msghdr *msg,
 
 	/* Copy the address and add cmsg data. */
 	if (family == AF_INET) {
-		sin = (struct sockaddr_in *) msg->msg_name;
-		sin->sin_family = AF_INET;
-		sin->sin_port = 0 /* skb->h.uh->source */;
-		sin->sin_addr.s_addr = ip_hdr(skb)->saddr;
-		memset(sin->sin_zero, 0, sizeof(sin->sin_zero));
+		struct sockaddr_in *sin = (struct sockaddr_in *)msg->msg_name;
+
+		if (sin) {
+			sin->sin_family = AF_INET;
+			sin->sin_port = 0 /* skb->h.uh->source */;
+			sin->sin_addr.s_addr = ip_hdr(skb)->saddr;
+			memset(sin->sin_zero, 0, sizeof(sin->sin_zero));
+			*addr_len = sizeof(*sin);
+		}
 
 		if (isk->cmsg_flags)
 			ip_cmsg_recv(msg, skb);
@@ -890,17 +885,21 @@ int ping_recvmsg(struct kiocb *iocb, struct sock *sk, struct msghdr *msg,
 	} else if (family == AF_INET6) {
 		struct ipv6_pinfo *np = inet6_sk(sk);
 		struct ipv6hdr *ip6 = ipv6_hdr(skb);
-		sin6 = (struct sockaddr_in6 *) msg->msg_name;
-		sin6->sin6_family = AF_INET6;
-		sin6->sin6_port = 0;
-		sin6->sin6_addr = ip6->saddr;
-
-		sin6->sin6_flowinfo = 0;
-		if (np->sndflow)
-			sin6->sin6_flowinfo = ip6_flowinfo(ip6);
-
-		sin6->sin6_scope_id = ipv6_iface_scope_id(&sin6->sin6_addr,
-							  IP6CB(skb)->iif);
+		struct sockaddr_in6 *sin6 =
+			(struct sockaddr_in6 *)msg->msg_name;
+
+		if (sin6) {
+			sin6->sin6_family = AF_INET6;
+			sin6->sin6_port = 0;
+			sin6->sin6_addr = ip6->saddr;
+			sin6->sin6_flowinfo = 0;
+			if (np->sndflow)
+				sin6->sin6_flowinfo = ip6_flowinfo(ip6);
+			sin6->sin6_scope_id =
+				ipv6_iface_scope_id(&sin6->sin6_addr,
+						    IP6CB(skb)->iif);
+			*addr_len = sizeof(*sin6);
+		}
 
 		if (inet6_sk(sk)->rxopt.all)
 			pingv6_ops.ip6_datagram_recv_ctl(sk, msg, skb);
@@ -1076,7 +1075,7 @@ void ping_seq_stop(struct seq_file *seq, void *v)
 EXPORT_SYMBOL_GPL(ping_seq_stop);
 
 static void ping_v4_format_sock(struct sock *sp, struct seq_file *f,
-		int bucket, int *len)
+		int bucket)
 {
 	struct inet_sock *inet = inet_sk(sp);
 	__be32 dest = inet->inet_daddr;
@@ -1085,7 +1084,7 @@ static void ping_v4_format_sock(struct sock *sp, struct seq_file *f,
 	__u16 srcp = ntohs(inet->inet_sport);
 
 	seq_printf(f, "%5d: %08X:%04X %08X:%04X"
-		" %02X %08X:%08X %02X:%08lX %08X %5u %8d %lu %d %pK %d%n",
+		" %02X %08X:%08X %02X:%08lX %08X %5u %8d %lu %d %pK %d",
 		bucket, src, srcp, dest, destp, sp->sk_state,
 		sk_wmem_alloc_get(sp),
 		sk_rmem_alloc_get(sp),
@@ -1093,23 +1092,22 @@ static void ping_v4_format_sock(struct sock *sp, struct seq_file *f,
 		from_kuid_munged(seq_user_ns(f), sock_i_uid(sp)),
 		0, sock_i_ino(sp),
 		atomic_read(&sp->sk_refcnt), sp,
-		atomic_read(&sp->sk_drops), len);
+		atomic_read(&sp->sk_drops));
 }
 
 static int ping_v4_seq_show(struct seq_file *seq, void *v)
 {
+	seq_setwidth(seq, 127);
 	if (v == SEQ_START_TOKEN)
-		seq_printf(seq, "%-127s\n",
-			   "  sl  local_address rem_address   st tx_queue "
+		seq_puts(seq, "  sl  local_address rem_address   st tx_queue "
 			   "rx_queue tr tm->when retrnsmt   uid  timeout "
 			   "inode ref pointer drops");
 	else {
 		struct ping_iter_state *state = seq->private;
-		int len;
 
-		ping_v4_format_sock(v, seq, state->bucket, &len);
-		seq_printf(seq, "%*s\n", 127 - len, "");
+		ping_v4_format_sock(v, seq, state->bucket);
 	}
+	seq_pad(seq, '\n');
 	return 0;
 }
 
diff --git a/net/ipv4/raw.c b/net/ipv4/raw.c
index 41e1d2845c8f..5cb8ddb505ee 100644
--- a/net/ipv4/raw.c
+++ b/net/ipv4/raw.c
@@ -696,9 +696,6 @@ static int raw_recvmsg(struct kiocb *iocb, struct sock *sk, struct msghdr *msg,
 	if (flags & MSG_OOB)
 		goto out;
 
-	if (addr_len)
-		*addr_len = sizeof(*sin);
-
 	if (flags & MSG_ERRQUEUE) {
 		err = ip_recv_error(sk, msg, len);
 		goto out;
@@ -726,6 +723,7 @@ static int raw_recvmsg(struct kiocb *iocb, struct sock *sk, struct msghdr *msg,
 		sin->sin_addr.s_addr = ip_hdr(skb)->saddr;
 		sin->sin_port = 0;
 		memset(&sin->sin_zero, 0, sizeof(sin->sin_zero));
+		*addr_len = sizeof(*sin);
 	}
 	if (inet->cmsg_flags)
 		ip_cmsg_recv(msg, skb);
diff --git a/net/ipv4/tcp.c b/net/ipv4/tcp.c
index 8e8529d3c8c9..c4638e6f0238 100644
--- a/net/ipv4/tcp.c
+++ b/net/ipv4/tcp.c
@@ -808,12 +808,6 @@ static unsigned int tcp_xmit_size_goal(struct sock *sk, u32 mss_now,
 		xmit_size_goal = min_t(u32, gso_size,
 				       sk->sk_gso_max_size - 1 - hlen);
 
-		/* TSQ : try to have at least two segments in flight
-		 * (one in NIC TX ring, another in Qdisc)
-		 */
-		xmit_size_goal = min_t(u32, xmit_size_goal,
-				       sysctl_tcp_limit_output_bytes >> 1);
-
 		xmit_size_goal = tcp_bound_to_half_wnd(tp, xmit_size_goal);
 
 		/* We try hard to avoid divides here */
@@ -1431,7 +1425,7 @@ static void tcp_service_net_dma(struct sock *sk, bool wait)
 	do {
 		if (dma_async_is_tx_complete(tp->ucopy.dma_chan,
 					      last_issued, &done,
-					      &used) == DMA_SUCCESS) {
+					      &used) == DMA_COMPLETE) {
 			/* Safe to free early-copied skbs now */
 			__skb_queue_purge(&sk->sk_async_wait_queue);
 			break;
@@ -1439,7 +1433,7 @@ static void tcp_service_net_dma(struct sock *sk, bool wait)
 			struct sk_buff *skb;
 			while ((skb = skb_peek(&sk->sk_async_wait_queue)) &&
 			       (dma_async_is_complete(skb->dma_cookie, done,
-						      used) == DMA_SUCCESS)) {
+						      used) == DMA_COMPLETE)) {
 				__skb_dequeue(&sk->sk_async_wait_queue);
 				kfree_skb(skb);
 			}
diff --git a/net/ipv4/tcp_ipv4.c b/net/ipv4/tcp_ipv4.c
index 14bba8a1c5a7..59a6f8b90cd9 100644
--- a/net/ipv4/tcp_ipv4.c
+++ b/net/ipv4/tcp_ipv4.c
@@ -2541,13 +2541,13 @@ void tcp_proc_unregister(struct net *net, struct tcp_seq_afinfo *afinfo)
 EXPORT_SYMBOL(tcp_proc_unregister);
 
 static void get_openreq4(const struct sock *sk, const struct request_sock *req,
-			 struct seq_file *f, int i, kuid_t uid, int *len)
+			 struct seq_file *f, int i, kuid_t uid)
 {
 	const struct inet_request_sock *ireq = inet_rsk(req);
 	long delta = req->expires - jiffies;
 
 	seq_printf(f, "%4d: %08X:%04X %08X:%04X"
-		" %02X %08X:%08X %02X:%08lX %08X %5u %8d %u %d %pK%n",
+		" %02X %08X:%08X %02X:%08lX %08X %5u %8d %u %d %pK",
 		i,
 		ireq->ir_loc_addr,
 		ntohs(inet_sk(sk)->inet_sport),
@@ -2562,11 +2562,10 @@ static void get_openreq4(const struct sock *sk, const struct request_sock *req,
 		0,  /* non standard timer */
 		0, /* open_requests have no inode */
 		atomic_read(&sk->sk_refcnt),
-		req,
-		len);
+		req);
 }
 
-static void get_tcp4_sock(struct sock *sk, struct seq_file *f, int i, int *len)
+static void get_tcp4_sock(struct sock *sk, struct seq_file *f, int i)
 {
 	int timer_active;
 	unsigned long timer_expires;
@@ -2605,7 +2604,7 @@ static void get_tcp4_sock(struct sock *sk, struct seq_file *f, int i, int *len)
 		rx_queue = max_t(int, tp->rcv_nxt - tp->copied_seq, 0);
 
 	seq_printf(f, "%4d: %08X:%04X %08X:%04X %02X %08X:%08X %02X:%08lX "
-			"%08X %5u %8d %lu %d %pK %lu %lu %u %u %d%n",
+			"%08X %5u %8d %lu %d %pK %lu %lu %u %u %d",
 		i, src, srcp, dest, destp, sk->sk_state,
 		tp->write_seq - tp->snd_una,
 		rx_queue,
@@ -2622,12 +2621,11 @@ static void get_tcp4_sock(struct sock *sk, struct seq_file *f, int i, int *len)
 		tp->snd_cwnd,
 		sk->sk_state == TCP_LISTEN ?
 		    (fastopenq ? fastopenq->max_qlen : 0) :
-		    (tcp_in_initial_slowstart(tp) ? -1 : tp->snd_ssthresh),
-		len);
+		    (tcp_in_initial_slowstart(tp) ? -1 : tp->snd_ssthresh));
 }
 
 static void get_timewait4_sock(const struct inet_timewait_sock *tw,
-			       struct seq_file *f, int i, int *len)
+			       struct seq_file *f, int i)
 {
 	__be32 dest, src;
 	__u16 destp, srcp;
@@ -2639,10 +2637,10 @@ static void get_timewait4_sock(const struct inet_timewait_sock *tw,
 	srcp  = ntohs(tw->tw_sport);
 
 	seq_printf(f, "%4d: %08X:%04X %08X:%04X"
-		" %02X %08X:%08X %02X:%08lX %08X %5d %8d %d %d %pK%n",
+		" %02X %08X:%08X %02X:%08lX %08X %5d %8d %d %d %pK",
 		i, src, srcp, dest, destp, tw->tw_substate, 0, 0,
 		3, jiffies_delta_to_clock_t(delta), 0, 0, 0, 0,
-		atomic_read(&tw->tw_refcnt), tw, len);
+		atomic_read(&tw->tw_refcnt), tw);
 }
 
 #define TMPSZ 150
@@ -2651,11 +2649,10 @@ static int tcp4_seq_show(struct seq_file *seq, void *v)
 {
 	struct tcp_iter_state *st;
 	struct sock *sk = v;
-	int len;
 
+	seq_setwidth(seq, TMPSZ - 1);
 	if (v == SEQ_START_TOKEN) {
-		seq_printf(seq, "%-*s\n", TMPSZ - 1,
-			   "  sl  local_address rem_address   st tx_queue "
+		seq_puts(seq, "  sl  local_address rem_address   st tx_queue "
 			   "rx_queue tr tm->when retrnsmt   uid  timeout "
 			   "inode");
 		goto out;
@@ -2666,16 +2663,16 @@ static int tcp4_seq_show(struct seq_file *seq, void *v)
 	case TCP_SEQ_STATE_LISTENING:
 	case TCP_SEQ_STATE_ESTABLISHED:
 		if (sk->sk_state == TCP_TIME_WAIT)
-			get_timewait4_sock(v, seq, st->num, &len);
+			get_timewait4_sock(v, seq, st->num);
 		else
-			get_tcp4_sock(v, seq, st->num, &len);
+			get_tcp4_sock(v, seq, st->num);
 		break;
 	case TCP_SEQ_STATE_OPENREQ:
-		get_openreq4(st->syn_wait_sk, v, seq, st->num, st->uid, &len);
+		get_openreq4(st->syn_wait_sk, v, seq, st->num, st->uid);
 		break;
 	}
-	seq_printf(seq, "%*s\n", TMPSZ - 1 - len, "");
 out:
+	seq_pad(seq, '\n');
 	return 0;
 }
 
diff --git a/net/ipv4/tcp_metrics.c b/net/ipv4/tcp_metrics.c
index 2ab09cbae74d..06493736fbc8 100644
--- a/net/ipv4/tcp_metrics.c
+++ b/net/ipv4/tcp_metrics.c
@@ -663,10 +663,13 @@ void tcp_fastopen_cache_get(struct sock *sk, u16 *mss,
 void tcp_fastopen_cache_set(struct sock *sk, u16 mss,
 			    struct tcp_fastopen_cookie *cookie, bool syn_lost)
 {
+	struct dst_entry *dst = __sk_dst_get(sk);
 	struct tcp_metrics_block *tm;
 
+	if (!dst)
+		return;
 	rcu_read_lock();
-	tm = tcp_get_metrics(sk, __sk_dst_get(sk), true);
+	tm = tcp_get_metrics(sk, dst, true);
 	if (tm) {
 		struct tcp_fastopen_metrics *tfom = &tm->tcpm_fastopen;
 
@@ -988,7 +991,7 @@ static int tcp_metrics_nl_cmd_del(struct sk_buff *skb, struct genl_info *info)
 	return 0;
 }
 
-static struct genl_ops tcp_metrics_nl_ops[] = {
+static const struct genl_ops tcp_metrics_nl_ops[] = {
 	{
 		.cmd = TCP_METRICS_CMD_GET,
 		.doit = tcp_metrics_nl_cmd_get,
@@ -1079,8 +1082,7 @@ void __init tcp_metrics_init(void)
 	if (ret < 0)
 		goto cleanup;
 	ret = genl_register_family_with_ops(&tcp_metrics_nl_family,
-					    tcp_metrics_nl_ops,
-					    ARRAY_SIZE(tcp_metrics_nl_ops));
+					    tcp_metrics_nl_ops);
 	if (ret < 0)
 		goto cleanup_subsys;
 	return;
diff --git a/net/ipv4/tcp_output.c b/net/ipv4/tcp_output.c
index 672854664ff5..7820f3a7dd70 100644
--- a/net/ipv4/tcp_output.c
+++ b/net/ipv4/tcp_output.c
@@ -1875,8 +1875,12 @@ static bool tcp_write_xmit(struct sock *sk, unsigned int mss_now, int nonagle,
 		 *  - better RTT estimation and ACK scheduling
 		 *  - faster recovery
 		 *  - high rates
+		 * Alas, some drivers / subsystems require a fair amount
+		 * of queued bytes to ensure line rate.
+		 * One example is wifi aggregation (802.11 AMPDU)
 		 */
-		limit = max(skb->truesize, sk->sk_pacing_rate >> 10);
+		limit = max_t(unsigned int, sysctl_tcp_limit_output_bytes,
+			      sk->sk_pacing_rate >> 10);
 
 		if (atomic_read(&sk->sk_wmem_alloc) > limit) {
 			set_bit(TSQ_THROTTLED, &tp->tsq_flags);
@@ -3093,7 +3097,6 @@ void tcp_send_window_probe(struct sock *sk)
 {
 	if (sk->sk_state == TCP_ESTABLISHED) {
 		tcp_sk(sk)->snd_wl1 = tcp_sk(sk)->rcv_nxt - 1;
-		tcp_sk(sk)->snd_nxt = tcp_sk(sk)->write_seq;
 		tcp_xmit_probe_skb(sk, 0);
 	}
 }
diff --git a/net/ipv4/udp.c b/net/ipv4/udp.c
index 89909dd730dd..5944d7d668dd 100644
--- a/net/ipv4/udp.c
+++ b/net/ipv4/udp.c
@@ -1235,12 +1235,6 @@ int udp_recvmsg(struct kiocb *iocb, struct sock *sk, struct msghdr *msg,
 	int is_udplite = IS_UDPLITE(sk);
 	bool slow;
 
-	/*
-	 *	Check any passed addresses
-	 */
-	if (addr_len)
-		*addr_len = sizeof(*sin);
-
 	if (flags & MSG_ERRQUEUE)
 		return ip_recv_error(sk, msg, len);
 
@@ -1302,6 +1296,7 @@ try_again:
 		sin->sin_port = udp_hdr(skb)->source;
 		sin->sin_addr.s_addr = ip_hdr(skb)->saddr;
 		memset(sin->sin_zero, 0, sizeof(sin->sin_zero));
+		*addr_len = sizeof(*sin);
 	}
 	if (inet->cmsg_flags)
 		ip_cmsg_recv(msg, skb);
@@ -2331,7 +2326,7 @@ EXPORT_SYMBOL(udp_proc_unregister);
 
 /* ------------------------------------------------------------------------ */
 static void udp4_format_sock(struct sock *sp, struct seq_file *f,
-		int bucket, int *len)
+		int bucket)
 {
 	struct inet_sock *inet = inet_sk(sp);
 	__be32 dest = inet->inet_daddr;
@@ -2340,7 +2335,7 @@ static void udp4_format_sock(struct sock *sp, struct seq_file *f,
 	__u16 srcp	  = ntohs(inet->inet_sport);
 
 	seq_printf(f, "%5d: %08X:%04X %08X:%04X"
-		" %02X %08X:%08X %02X:%08lX %08X %5u %8d %lu %d %pK %d%n",
+		" %02X %08X:%08X %02X:%08lX %08X %5u %8d %lu %d %pK %d",
 		bucket, src, srcp, dest, destp, sp->sk_state,
 		sk_wmem_alloc_get(sp),
 		sk_rmem_alloc_get(sp),
@@ -2348,23 +2343,22 @@ static void udp4_format_sock(struct sock *sp, struct seq_file *f,
 		from_kuid_munged(seq_user_ns(f), sock_i_uid(sp)),
 		0, sock_i_ino(sp),
 		atomic_read(&sp->sk_refcnt), sp,
-		atomic_read(&sp->sk_drops), len);
+		atomic_read(&sp->sk_drops));
 }
 
 int udp4_seq_show(struct seq_file *seq, void *v)
 {
+	seq_setwidth(seq, 127);
 	if (v == SEQ_START_TOKEN)
-		seq_printf(seq, "%-127s\n",
-			   "  sl  local_address rem_address   st tx_queue "
+		seq_puts(seq, "  sl  local_address rem_address   st tx_queue "
 			   "rx_queue tr tm->when retrnsmt   uid  timeout "
 			   "inode ref pointer drops");
 	else {
 		struct udp_iter_state *state = seq->private;
-		int len;
 
-		udp4_format_sock(v, seq, state->bucket, &len);
-		seq_printf(seq, "%*s\n", 127 - len, "");
+		udp4_format_sock(v, seq, state->bucket);
 	}
+	seq_pad(seq, '\n');
 	return 0;
 }
 
diff --git a/net/ipv6/addrconf.c b/net/ipv6/addrconf.c
index 542d09561ed6..12c97d8aa6bb 100644
--- a/net/ipv6/addrconf.c
+++ b/net/ipv6/addrconf.c
@@ -271,10 +271,24 @@ static void addrconf_mod_dad_timer(struct inet6_ifaddr *ifp,
 
 static int snmp6_alloc_dev(struct inet6_dev *idev)
 {
+	int i;
+
 	if (snmp_mib_init((void __percpu **)idev->stats.ipv6,
 			  sizeof(struct ipstats_mib),
 			  __alignof__(struct ipstats_mib)) < 0)
 		goto err_ip;
+
+	for_each_possible_cpu(i) {
+		struct ipstats_mib *addrconf_stats;
+		addrconf_stats = per_cpu_ptr(idev->stats.ipv6[0], i);
+		u64_stats_init(&addrconf_stats->syncp);
+#if SNMP_ARRAY_SZ == 2
+		addrconf_stats = per_cpu_ptr(idev->stats.ipv6[1], i);
+		u64_stats_init(&addrconf_stats->syncp);
+#endif
+	}
+
+
 	idev->stats.icmpv6dev = kzalloc(sizeof(struct icmpv6_mib_device),
 					GFP_KERNEL);
 	if (!idev->stats.icmpv6dev)
@@ -1982,23 +1996,6 @@ static void addrconf_add_mroute(struct net_device *dev)
 	ip6_route_add(&cfg);
 }
 
-#if IS_ENABLED(CONFIG_IPV6_SIT)
-static void sit_route_add(struct net_device *dev)
-{
-	struct fib6_config cfg = {
-		.fc_table = RT6_TABLE_MAIN,
-		.fc_metric = IP6_RT_PRIO_ADDRCONF,
-		.fc_ifindex = dev->ifindex,
-		.fc_dst_len = 96,
-		.fc_flags = RTF_UP | RTF_NONEXTHOP,
-		.fc_nlinfo.nl_net = dev_net(dev),
-	};
-
-	/* prefix length - 96 bits "::d.d.d.d" */
-	ip6_route_add(&cfg);
-}
-#endif
-
 static struct inet6_dev *addrconf_add_dev(struct net_device *dev)
 {
 	struct inet6_dev *idev;
@@ -2528,7 +2525,8 @@ static void sit_add_v4_addrs(struct inet6_dev *idev)
 	struct in6_addr addr;
 	struct net_device *dev;
 	struct net *net = dev_net(idev->dev);
-	int scope;
+	int scope, plen;
+	u32 pflags = 0;
 
 	ASSERT_RTNL();
 
@@ -2538,12 +2536,16 @@ static void sit_add_v4_addrs(struct inet6_dev *idev)
 	if (idev->dev->flags&IFF_POINTOPOINT) {
 		addr.s6_addr32[0] = htonl(0xfe800000);
 		scope = IFA_LINK;
+		plen = 64;
 	} else {
 		scope = IPV6_ADDR_COMPATv4;
+		plen = 96;
+		pflags |= RTF_NONEXTHOP;
 	}
 
 	if (addr.s6_addr32[3]) {
-		add_addr(idev, &addr, 128, scope);
+		add_addr(idev, &addr, plen, scope);
+		addrconf_prefix_route(&addr, plen, idev->dev, 0, pflags);
 		return;
 	}
 
@@ -2555,7 +2557,6 @@ static void sit_add_v4_addrs(struct inet6_dev *idev)
 			int flag = scope;
 
 			for (ifa = in_dev->ifa_list; ifa; ifa = ifa->ifa_next) {
-				int plen;
 
 				addr.s6_addr32[3] = ifa->ifa_local;
 
@@ -2566,12 +2567,10 @@ static void sit_add_v4_addrs(struct inet6_dev *idev)
 						continue;
 					flag |= IFA_HOST;
 				}
-				if (idev->dev->flags&IFF_POINTOPOINT)
-					plen = 64;
-				else
-					plen = 96;
 
 				add_addr(idev, &addr, plen, flag);
+				addrconf_prefix_route(&addr, plen, idev->dev, 0,
+						      pflags);
 			}
 		}
 	}
@@ -2697,7 +2696,6 @@ static void addrconf_sit_config(struct net_device *dev)
 		struct in6_addr addr;
 
 		ipv6_addr_set(&addr,  htonl(0xFE800000), 0, 0, 0);
-		addrconf_prefix_route(&addr, 64, dev, 0, 0);
 		if (!ipv6_generate_eui64(addr.s6_addr + 8, dev))
 			addrconf_add_linklocal(idev, &addr);
 		return;
@@ -2707,8 +2705,6 @@ static void addrconf_sit_config(struct net_device *dev)
 
 	if (dev->flags&IFF_POINTOPOINT)
 		addrconf_add_mroute(dev);
-	else
-		sit_route_add(dev);
 }
 #endif
 
@@ -2726,8 +2722,6 @@ static void addrconf_gre_config(struct net_device *dev)
 	}
 
 	ipv6_addr_set(&addr,  htonl(0xFE800000), 0, 0, 0);
-	addrconf_prefix_route(&addr, 64, dev, 0, 0);
-
 	if (!ipv6_generate_eui64(addr.s6_addr + 8, dev))
 		addrconf_add_linklocal(idev, &addr);
 }
diff --git a/net/ipv6/af_inet6.c b/net/ipv6/af_inet6.c
index 6468bda1f2b9..4fbdb7046d28 100644
--- a/net/ipv6/af_inet6.c
+++ b/net/ipv6/af_inet6.c
@@ -714,6 +714,8 @@ static void ipv6_packet_cleanup(void)
 
 static int __net_init ipv6_init_mibs(struct net *net)
 {
+	int i;
+
 	if (snmp_mib_init((void __percpu **)net->mib.udp_stats_in6,
 			  sizeof(struct udp_mib),
 			  __alignof__(struct udp_mib)) < 0)
@@ -726,6 +728,18 @@ static int __net_init ipv6_init_mibs(struct net *net)
 			  sizeof(struct ipstats_mib),
 			  __alignof__(struct ipstats_mib)) < 0)
 		goto err_ip_mib;
+
+	for_each_possible_cpu(i) {
+		struct ipstats_mib *af_inet6_stats;
+		af_inet6_stats = per_cpu_ptr(net->mib.ipv6_statistics[0], i);
+		u64_stats_init(&af_inet6_stats->syncp);
+#if SNMP_ARRAY_SZ == 2
+		af_inet6_stats = per_cpu_ptr(net->mib.ipv6_statistics[1], i);
+		u64_stats_init(&af_inet6_stats->syncp);
+#endif
+	}
+
+
 	if (snmp_mib_init((void __percpu **)net->mib.icmpv6_statistics,
 			  sizeof(struct icmpv6_mib),
 			  __alignof__(struct icmpv6_mib)) < 0)
@@ -958,10 +972,10 @@ out:
 
 #ifdef CONFIG_SYSCTL
 sysctl_fail:
-	ipv6_packet_cleanup();
+	pingv6_exit();
 #endif
 pingv6_fail:
-	pingv6_exit();
+	ipv6_packet_cleanup();
 ipv6_packet_fail:
 	tcpv6_exit();
 tcpv6_fail:
diff --git a/net/ipv6/ip6_gre.c b/net/ipv6/ip6_gre.c
index bf4a9a084de5..8acb28621f9c 100644
--- a/net/ipv6/ip6_gre.c
+++ b/net/ipv6/ip6_gre.c
@@ -1252,6 +1252,7 @@ static void ip6gre_tunnel_setup(struct net_device *dev)
 static int ip6gre_tunnel_init(struct net_device *dev)
 {
 	struct ip6_tnl *tunnel;
+	int i;
 
 	tunnel = netdev_priv(dev);
 
@@ -1269,6 +1270,13 @@ static int ip6gre_tunnel_init(struct net_device *dev)
 	if (!dev->tstats)
 		return -ENOMEM;
 
+	for_each_possible_cpu(i) {
+		struct pcpu_tstats *ip6gre_tunnel_stats;
+		ip6gre_tunnel_stats = per_cpu_ptr(dev->tstats, i);
+		u64_stats_init(&ip6gre_tunnel_stats->syncp);
+	}
+
+
 	return 0;
 }
 
@@ -1449,6 +1457,7 @@ static void ip6gre_netlink_parms(struct nlattr *data[],
 static int ip6gre_tap_init(struct net_device *dev)
 {
 	struct ip6_tnl *tunnel;
+	int i;
 
 	tunnel = netdev_priv(dev);
 
@@ -1462,6 +1471,12 @@ static int ip6gre_tap_init(struct net_device *dev)
 	if (!dev->tstats)
 		return -ENOMEM;
 
+	for_each_possible_cpu(i) {
+		struct pcpu_tstats *ip6gre_tap_stats;
+		ip6gre_tap_stats = per_cpu_ptr(dev->tstats, i);
+		u64_stats_init(&ip6gre_tap_stats->syncp);
+	}
+
 	return 0;
 }
 
diff --git a/net/ipv6/ip6_output.c b/net/ipv6/ip6_output.c
index 5e31a909a2b0..59df872e2f4d 100644
--- a/net/ipv6/ip6_output.c
+++ b/net/ipv6/ip6_output.c
@@ -910,7 +910,7 @@ static int ip6_dst_lookup_tail(struct sock *sk,
 
 out_err_release:
 	if (err == -ENETUNREACH)
-		IP6_INC_STATS_BH(net, NULL, IPSTATS_MIB_OUTNOROUTES);
+		IP6_INC_STATS(net, NULL, IPSTATS_MIB_OUTNOROUTES);
 	dst_release(*dst);
 	*dst = NULL;
 	return err;
diff --git a/net/ipv6/ip6_tunnel.c b/net/ipv6/ip6_tunnel.c
index 583b77e2f69b..d6062325db08 100644
--- a/net/ipv6/ip6_tunnel.c
+++ b/net/ipv6/ip6_tunnel.c
@@ -1494,12 +1494,19 @@ static inline int
 ip6_tnl_dev_init_gen(struct net_device *dev)
 {
 	struct ip6_tnl *t = netdev_priv(dev);
+	int i;
 
 	t->dev = dev;
 	t->net = dev_net(dev);
 	dev->tstats = alloc_percpu(struct pcpu_tstats);
 	if (!dev->tstats)
 		return -ENOMEM;
+
+	for_each_possible_cpu(i) {
+		struct pcpu_tstats *ip6_tnl_stats;
+		ip6_tnl_stats = per_cpu_ptr(dev->tstats, i);
+		u64_stats_init(&ip6_tnl_stats->syncp);
+	}
 	return 0;
 }
 
@@ -1635,6 +1642,15 @@ static int ip6_tnl_changelink(struct net_device *dev, struct nlattr *tb[],
 	return ip6_tnl_update(t, &p);
 }
 
+static void ip6_tnl_dellink(struct net_device *dev, struct list_head *head)
+{
+	struct net *net = dev_net(dev);
+	struct ip6_tnl_net *ip6n = net_generic(net, ip6_tnl_net_id);
+
+	if (dev != ip6n->fb_tnl_dev)
+		unregister_netdevice_queue(dev, head);
+}
+
 static size_t ip6_tnl_get_size(const struct net_device *dev)
 {
 	return
@@ -1699,6 +1715,7 @@ static struct rtnl_link_ops ip6_link_ops __read_mostly = {
 	.validate	= ip6_tnl_validate,
 	.newlink	= ip6_tnl_newlink,
 	.changelink	= ip6_tnl_changelink,
+	.dellink	= ip6_tnl_dellink,
 	.get_size	= ip6_tnl_get_size,
 	.fill_info	= ip6_tnl_fill_info,
 };
@@ -1715,9 +1732,9 @@ static struct xfrm6_tunnel ip6ip6_handler __read_mostly = {
 	.priority	=	1,
 };
 
-static void __net_exit ip6_tnl_destroy_tunnels(struct ip6_tnl_net *ip6n)
+static void __net_exit ip6_tnl_destroy_tunnels(struct net *net)
 {
-	struct net *net = dev_net(ip6n->fb_tnl_dev);
+	struct ip6_tnl_net *ip6n = net_generic(net, ip6_tnl_net_id);
 	struct net_device *dev, *aux;
 	int h;
 	struct ip6_tnl *t;
@@ -1785,10 +1802,8 @@ err_alloc_dev:
 
 static void __net_exit ip6_tnl_exit_net(struct net *net)
 {
-	struct ip6_tnl_net *ip6n = net_generic(net, ip6_tnl_net_id);
-
 	rtnl_lock();
-	ip6_tnl_destroy_tunnels(ip6n);
+	ip6_tnl_destroy_tunnels(net);
 	rtnl_unlock();
 }
 
diff --git a/net/ipv6/ndisc.c b/net/ipv6/ndisc.c
index f8a55ff1971b..3512177deb4d 100644
--- a/net/ipv6/ndisc.c
+++ b/net/ipv6/ndisc.c
@@ -1726,8 +1726,8 @@ int __init ndisc_init(void)
 				    &ndisc_ifinfo_sysctl_change);
 	if (err)
 		goto out_unregister_pernet;
-#endif
 out:
+#endif
 	return err;
 
 #ifdef CONFIG_SYSCTL
diff --git a/net/ipv6/raw.c b/net/ipv6/raw.c
index 3c00842b0079..e24ff1df0401 100644
--- a/net/ipv6/raw.c
+++ b/net/ipv6/raw.c
@@ -465,9 +465,6 @@ static int rawv6_recvmsg(struct kiocb *iocb, struct sock *sk,
 	if (flags & MSG_OOB)
 		return -EOPNOTSUPP;
 
-	if (addr_len)
-		*addr_len=sizeof(*sin6);
-
 	if (flags & MSG_ERRQUEUE)
 		return ipv6_recv_error(sk, msg, len);
 
@@ -506,6 +503,7 @@ static int rawv6_recvmsg(struct kiocb *iocb, struct sock *sk,
 		sin6->sin6_flowinfo = 0;
 		sin6->sin6_scope_id = ipv6_iface_scope_id(&sin6->sin6_addr,
 							  IP6CB(skb)->iif);
+		*addr_len = sizeof(*sin6);
 	}
 
 	sock_recv_ts_and_drops(msg, sk, skb);
diff --git a/net/ipv6/sit.c b/net/ipv6/sit.c
index 3a9038dd818d..1b4a4a953675 100644
--- a/net/ipv6/sit.c
+++ b/net/ipv6/sit.c
@@ -1320,6 +1320,7 @@ static void ipip6_tunnel_setup(struct net_device *dev)
 static int ipip6_tunnel_init(struct net_device *dev)
 {
 	struct ip_tunnel *tunnel = netdev_priv(dev);
+	int i;
 
 	tunnel->dev = dev;
 	tunnel->net = dev_net(dev);
@@ -1332,6 +1333,12 @@ static int ipip6_tunnel_init(struct net_device *dev)
 	if (!dev->tstats)
 		return -ENOMEM;
 
+	for_each_possible_cpu(i) {
+		struct pcpu_tstats *ipip6_tunnel_stats;
+		ipip6_tunnel_stats = per_cpu_ptr(dev->tstats, i);
+		u64_stats_init(&ipip6_tunnel_stats->syncp);
+	}
+
 	return 0;
 }
 
@@ -1341,6 +1348,7 @@ static int __net_init ipip6_fb_tunnel_init(struct net_device *dev)
 	struct iphdr *iph = &tunnel->parms.iph;
 	struct net *net = dev_net(dev);
 	struct sit_net *sitn = net_generic(net, sit_net_id);
+	int i;
 
 	tunnel->dev = dev;
 	tunnel->net = dev_net(dev);
@@ -1354,6 +1362,13 @@ static int __net_init ipip6_fb_tunnel_init(struct net_device *dev)
 	dev->tstats = alloc_percpu(struct pcpu_tstats);
 	if (!dev->tstats)
 		return -ENOMEM;
+
+	for_each_possible_cpu(i) {
+		struct pcpu_tstats *ipip6_fb_stats;
+		ipip6_fb_stats = per_cpu_ptr(dev->tstats, i);
+		u64_stats_init(&ipip6_fb_stats->syncp);
+	}
+
 	dev_hold(dev);
 	rcu_assign_pointer(sitn->tunnels_wc[0], tunnel);
 	return 0;
@@ -1604,6 +1619,15 @@ static const struct nla_policy ipip6_policy[IFLA_IPTUN_MAX + 1] = {
 #endif
 };
 
+static void ipip6_dellink(struct net_device *dev, struct list_head *head)
+{
+	struct net *net = dev_net(dev);
+	struct sit_net *sitn = net_generic(net, sit_net_id);
+
+	if (dev != sitn->fb_tunnel_dev)
+		unregister_netdevice_queue(dev, head);
+}
+
 static struct rtnl_link_ops sit_link_ops __read_mostly = {
 	.kind		= "sit",
 	.maxtype	= IFLA_IPTUN_MAX,
@@ -1615,6 +1639,7 @@ static struct rtnl_link_ops sit_link_ops __read_mostly = {
 	.changelink	= ipip6_changelink,
 	.get_size	= ipip6_get_size,
 	.fill_info	= ipip6_fill_info,
+	.dellink	= ipip6_dellink,
 };
 
 static struct xfrm_tunnel sit_handler __read_mostly = {
@@ -1629,9 +1654,10 @@ static struct xfrm_tunnel ipip_handler __read_mostly = {
 	.priority	=	2,
 };
 
-static void __net_exit sit_destroy_tunnels(struct sit_net *sitn, struct list_head *head)
+static void __net_exit sit_destroy_tunnels(struct net *net,
+					   struct list_head *head)
 {
-	struct net *net = dev_net(sitn->fb_tunnel_dev);
+	struct sit_net *sitn = net_generic(net, sit_net_id);
 	struct net_device *dev, *aux;
 	int prio;
 
@@ -1706,11 +1732,10 @@ err_alloc_dev:
 
 static void __net_exit sit_exit_net(struct net *net)
 {
-	struct sit_net *sitn = net_generic(net, sit_net_id);
 	LIST_HEAD(list);
 
 	rtnl_lock();
-	sit_destroy_tunnels(sitn, &list);
+	sit_destroy_tunnels(net, &list);
 	unregister_netdevice_many(&list);
 	rtnl_unlock();
 }
diff --git a/net/ipv6/udp.c b/net/ipv6/udp.c
index f3893e897f72..81eb8cf8389b 100644
--- a/net/ipv6/udp.c
+++ b/net/ipv6/udp.c
@@ -392,9 +392,6 @@ int udpv6_recvmsg(struct kiocb *iocb, struct sock *sk,
 	int is_udp4;
 	bool slow;
 
-	if (addr_len)
-		*addr_len = sizeof(struct sockaddr_in6);
-
 	if (flags & MSG_ERRQUEUE)
 		return ipv6_recv_error(sk, msg, len);
 
@@ -480,7 +477,7 @@ try_again:
 				ipv6_iface_scope_id(&sin6->sin6_addr,
 						    IP6CB(skb)->iif);
 		}
-
+		*addr_len = sizeof(*sin6);
 	}
 	if (is_udp4) {
 		if (inet->cmsg_flags)
diff --git a/net/irda/irnetlink.c b/net/irda/irnetlink.c
index c32971269280..a37b81fe0479 100644
--- a/net/irda/irnetlink.c
+++ b/net/irda/irnetlink.c
@@ -131,7 +131,7 @@ static const struct nla_policy irda_nl_policy[IRDA_NL_ATTR_MAX + 1] = {
 	[IRDA_NL_ATTR_MODE] = { .type = NLA_U32 },
 };
 
-static struct genl_ops irda_nl_ops[] = {
+static const struct genl_ops irda_nl_ops[] = {
 	{
 		.cmd = IRDA_NL_CMD_SET_MODE,
 		.doit = irda_nl_set_mode,
@@ -149,8 +149,7 @@ static struct genl_ops irda_nl_ops[] = {
 
 int irda_nl_register(void)
 {
-	return genl_register_family_with_ops(&irda_nl_family,
-		irda_nl_ops, ARRAY_SIZE(irda_nl_ops));
+	return genl_register_family_with_ops(&irda_nl_family, irda_nl_ops);
 }
 
 void irda_nl_unregister(void)
diff --git a/net/l2tp/l2tp_ip.c b/net/l2tp/l2tp_ip.c
index 571db8dd2292..da1a1cee1a08 100644
--- a/net/l2tp/l2tp_ip.c
+++ b/net/l2tp/l2tp_ip.c
@@ -518,9 +518,6 @@ static int l2tp_ip_recvmsg(struct kiocb *iocb, struct sock *sk, struct msghdr *m
 	if (flags & MSG_OOB)
 		goto out;
 
-	if (addr_len)
-		*addr_len = sizeof(*sin);
-
 	skb = skb_recv_datagram(sk, flags, noblock, &err);
 	if (!skb)
 		goto out;
@@ -543,6 +540,7 @@ static int l2tp_ip_recvmsg(struct kiocb *iocb, struct sock *sk, struct msghdr *m
 		sin->sin_addr.s_addr = ip_hdr(skb)->saddr;
 		sin->sin_port = 0;
 		memset(&sin->sin_zero, 0, sizeof(sin->sin_zero));
+		*addr_len = sizeof(*sin);
 	}
 	if (inet->cmsg_flags)
 		ip_cmsg_recv(msg, skb);
diff --git a/net/l2tp/l2tp_netlink.c b/net/l2tp/l2tp_netlink.c
index be446d517bc9..4cfd722e9153 100644
--- a/net/l2tp/l2tp_netlink.c
+++ b/net/l2tp/l2tp_netlink.c
@@ -793,7 +793,7 @@ static struct nla_policy l2tp_nl_policy[L2TP_ATTR_MAX + 1] = {
 	},
 };
 
-static struct genl_ops l2tp_nl_ops[] = {
+static const struct genl_ops l2tp_nl_ops[] = {
 	{
 		.cmd = L2TP_CMD_NOOP,
 		.doit = l2tp_nl_cmd_noop,
@@ -887,13 +887,8 @@ EXPORT_SYMBOL_GPL(l2tp_nl_unregister_ops);
 
 static int l2tp_nl_init(void)
 {
-	int err;
-
 	pr_info("L2TP netlink interface\n");
-	err = genl_register_family_with_ops(&l2tp_nl_family, l2tp_nl_ops,
-					    ARRAY_SIZE(l2tp_nl_ops));
-
-	return err;
+	return genl_register_family_with_ops(&l2tp_nl_family, l2tp_nl_ops);
 }
 
 static void l2tp_nl_cleanup(void)
diff --git a/net/netfilter/ipvs/ip_vs_ctl.c b/net/netfilter/ipvs/ip_vs_ctl.c
index 62786a495cea..35be035ee0ce 100644
--- a/net/netfilter/ipvs/ip_vs_ctl.c
+++ b/net/netfilter/ipvs/ip_vs_ctl.c
@@ -842,7 +842,7 @@ ip_vs_new_dest(struct ip_vs_service *svc, struct ip_vs_dest_user_kern *udest,
 	       struct ip_vs_dest **dest_p)
 {
 	struct ip_vs_dest *dest;
-	unsigned int atype;
+	unsigned int atype, i;
 
 	EnterFunction(2);
 
@@ -869,6 +869,12 @@ ip_vs_new_dest(struct ip_vs_service *svc, struct ip_vs_dest_user_kern *udest,
 	if (!dest->stats.cpustats)
 		goto err_alloc;
 
+	for_each_possible_cpu(i) {
+		struct ip_vs_cpu_stats *ip_vs_dest_stats;
+		ip_vs_dest_stats = per_cpu_ptr(dest->stats.cpustats, i);
+		u64_stats_init(&ip_vs_dest_stats->syncp);
+	}
+
 	dest->af = svc->af;
 	dest->protocol = svc->protocol;
 	dest->vaddr = svc->addr;
@@ -1134,7 +1140,7 @@ static int
 ip_vs_add_service(struct net *net, struct ip_vs_service_user_kern *u,
 		  struct ip_vs_service **svc_p)
 {
-	int ret = 0;
+	int ret = 0, i;
 	struct ip_vs_scheduler *sched = NULL;
 	struct ip_vs_pe *pe = NULL;
 	struct ip_vs_service *svc = NULL;
@@ -1184,6 +1190,13 @@ ip_vs_add_service(struct net *net, struct ip_vs_service_user_kern *u,
 		goto out_err;
 	}
 
+	for_each_possible_cpu(i) {
+		struct ip_vs_cpu_stats *ip_vs_stats;
+		ip_vs_stats = per_cpu_ptr(svc->stats.cpustats, i);
+		u64_stats_init(&ip_vs_stats->syncp);
+	}
+
+
 	/* I'm the first user of the service */
 	atomic_set(&svc->refcnt, 0);
 
@@ -3567,7 +3580,7 @@ out:
 }
 
 
-static struct genl_ops ip_vs_genl_ops[] __read_mostly = {
+static const struct genl_ops ip_vs_genl_ops[] __read_mostly = {
 	{
 		.cmd	= IPVS_CMD_NEW_SERVICE,
 		.flags	= GENL_ADMIN_PERM,
@@ -3666,7 +3679,7 @@ static struct genl_ops ip_vs_genl_ops[] __read_mostly = {
 static int __init ip_vs_genl_register(void)
 {
 	return genl_register_family_with_ops(&ip_vs_genl_family,
-		ip_vs_genl_ops, ARRAY_SIZE(ip_vs_genl_ops));
+					     ip_vs_genl_ops);
 }
 
 static void ip_vs_genl_unregister(void)
@@ -3780,7 +3793,7 @@ static struct notifier_block ip_vs_dst_notifier = {
 
 int __net_init ip_vs_control_net_init(struct net *net)
 {
-	int idx;
+	int i, idx;
 	struct netns_ipvs *ipvs = net_ipvs(net);
 
 	/* Initialize rs_table */
@@ -3799,6 +3812,12 @@ int __net_init ip_vs_control_net_init(struct net *net)
 	if (!ipvs->tot_stats.cpustats)
 		return -ENOMEM;
 
+	for_each_possible_cpu(i) {
+		struct ip_vs_cpu_stats *ipvs_tot_stats;
+		ipvs_tot_stats = per_cpu_ptr(ipvs->tot_stats.cpustats, i);
+		u64_stats_init(&ipvs_tot_stats->syncp);
+	}
+
 	spin_lock_init(&ipvs->tot_stats.lock);
 
 	proc_create("ip_vs", 0, net->proc_net, &ip_vs_info_fops);
diff --git a/net/netfilter/xt_set.c b/net/netfilter/xt_set.c
index e7c4e0e01ff5..80c2e2d603e0 100644
--- a/net/netfilter/xt_set.c
+++ b/net/netfilter/xt_set.c
@@ -84,7 +84,7 @@ set_match_v0_checkentry(const struct xt_mtchk_param *par)
 	index = ip_set_nfnl_get_byindex(par->net, info->match_set.index);
 
 	if (index == IPSET_INVALID_ID) {
-		pr_warning("Cannot find set indentified by id %u to match\n",
+		pr_warning("Cannot find set identified by id %u to match\n",
 			   info->match_set.index);
 		return -ENOENT;
 	}
@@ -134,7 +134,7 @@ set_match_v1_checkentry(const struct xt_mtchk_param *par)
 	index = ip_set_nfnl_get_byindex(par->net, info->match_set.index);
 
 	if (index == IPSET_INVALID_ID) {
-		pr_warning("Cannot find set indentified by id %u to match\n",
+		pr_warning("Cannot find set identified by id %u to match\n",
 			   info->match_set.index);
 		return -ENOENT;
 	}
diff --git a/net/netlabel/netlabel_cipso_v4.c b/net/netlabel/netlabel_cipso_v4.c
index a1100640495d..69345cebe3a3 100644
--- a/net/netlabel/netlabel_cipso_v4.c
+++ b/net/netlabel/netlabel_cipso_v4.c
@@ -737,7 +737,7 @@ static int netlbl_cipsov4_remove(struct sk_buff *skb, struct genl_info *info)
  * NetLabel Generic NETLINK Command Definitions
  */
 
-static struct genl_ops netlbl_cipsov4_ops[] = {
+static const struct genl_ops netlbl_cipsov4_ops[] = {
 	{
 	.cmd = NLBL_CIPSOV4_C_ADD,
 	.flags = GENL_ADMIN_PERM,
@@ -783,5 +783,5 @@ static struct genl_ops netlbl_cipsov4_ops[] = {
 int __init netlbl_cipsov4_genl_init(void)
 {
 	return genl_register_family_with_ops(&netlbl_cipsov4_gnl_family,
-		netlbl_cipsov4_ops, ARRAY_SIZE(netlbl_cipsov4_ops));
+					     netlbl_cipsov4_ops);
 }
diff --git a/net/netlabel/netlabel_mgmt.c b/net/netlabel/netlabel_mgmt.c
index dd1c37d7acbc..8ef83ee97c6a 100644
--- a/net/netlabel/netlabel_mgmt.c
+++ b/net/netlabel/netlabel_mgmt.c
@@ -705,7 +705,7 @@ version_failure:
  * NetLabel Generic NETLINK Command Definitions
  */
 
-static struct genl_ops netlbl_mgmt_genl_ops[] = {
+static const struct genl_ops netlbl_mgmt_genl_ops[] = {
 	{
 	.cmd = NLBL_MGMT_C_ADD,
 	.flags = GENL_ADMIN_PERM,
@@ -779,5 +779,5 @@ static struct genl_ops netlbl_mgmt_genl_ops[] = {
 int __init netlbl_mgmt_genl_init(void)
 {
 	return genl_register_family_with_ops(&netlbl_mgmt_gnl_family,
-		netlbl_mgmt_genl_ops, ARRAY_SIZE(netlbl_mgmt_genl_ops));
+					     netlbl_mgmt_genl_ops);
 }
diff --git a/net/netlabel/netlabel_unlabeled.c b/net/netlabel/netlabel_unlabeled.c
index 8f0897407a2c..43817d73ccf9 100644
--- a/net/netlabel/netlabel_unlabeled.c
+++ b/net/netlabel/netlabel_unlabeled.c
@@ -1323,7 +1323,7 @@ unlabel_staticlistdef_return:
  * NetLabel Generic NETLINK Command Definitions
  */
 
-static struct genl_ops netlbl_unlabel_genl_ops[] = {
+static const struct genl_ops netlbl_unlabel_genl_ops[] = {
 	{
 	.cmd = NLBL_UNLABEL_C_STATICADD,
 	.flags = GENL_ADMIN_PERM,
@@ -1397,7 +1397,7 @@ static struct genl_ops netlbl_unlabel_genl_ops[] = {
 int __init netlbl_unlabel_genl_init(void)
 {
 	return genl_register_family_with_ops(&netlbl_unlabel_gnl_family,
-		netlbl_unlabel_genl_ops, ARRAY_SIZE(netlbl_unlabel_genl_ops));
+					     netlbl_unlabel_genl_ops);
 }
 
 /*
diff --git a/net/netlink/af_netlink.c b/net/netlink/af_netlink.c
index 8df7f64c6db3..f0176e1a5a81 100644
--- a/net/netlink/af_netlink.c
+++ b/net/netlink/af_netlink.c
@@ -2017,7 +2017,7 @@ out:
  * netlink_set_err - report error to broadcast listeners
  * @ssk: the kernel netlink socket, as returned by netlink_kernel_create()
  * @portid: the PORTID of a process that we want to skip (if any)
- * @groups: the broadcast group that will notice the error
+ * @group: the broadcast group that will notice the error
  * @code: error code, must be negative (as usual in kernelspace)
  *
  * This function returns the number of broadcast listeners that have set the
diff --git a/net/netlink/genetlink.c b/net/netlink/genetlink.c
index 0c741cec4d0d..7dbc4f732c75 100644
--- a/net/netlink/genetlink.c
+++ b/net/netlink/genetlink.c
@@ -65,12 +65,24 @@ static struct list_head family_ht[GENL_FAM_TAB_SIZE];
  * To avoid an allocation at boot of just one unsigned long,
  * declare it global instead.
  * Bit 0 is marked as already used since group 0 is invalid.
+ * Bit 1 is marked as already used since the drop-monitor code
+ * abuses the API and thinks it can statically use group 1.
+ * That group will typically conflict with other groups that
+ * any proper users use.
+ * Bit 16 is marked as used since it's used for generic netlink
+ * and the code no longer marks pre-reserved IDs as used.
+ * Bit 17 is marked as already used since the VFS quota code
+ * also abused this API and relied on family == group ID, we
+ * cater to that by giving it a static family and group ID.
  */
-static unsigned long mc_group_start = 0x1;
+static unsigned long mc_group_start = 0x3 | BIT(GENL_ID_CTRL) |
+				      BIT(GENL_ID_VFS_DQUOT);
 static unsigned long *mc_groups = &mc_group_start;
 static unsigned long mc_groups_longs = 1;
 
-static int genl_ctrl_event(int event, void *data);
+static int genl_ctrl_event(int event, struct genl_family *family,
+			   const struct genl_multicast_group *grp,
+			   int grp_id);
 
 static inline unsigned int genl_family_hash(unsigned int id)
 {
@@ -106,13 +118,13 @@ static struct genl_family *genl_family_find_byname(char *name)
 	return NULL;
 }
 
-static struct genl_ops *genl_get_cmd(u8 cmd, struct genl_family *family)
+static const struct genl_ops *genl_get_cmd(u8 cmd, struct genl_family *family)
 {
-	struct genl_ops *ops;
+	int i;
 
-	list_for_each_entry(ops, &family->ops_list, ops_list)
-		if (ops->cmd == cmd)
-			return ops;
+	for (i = 0; i < family->n_ops; i++)
+		if (family->ops[i].cmd == cmd)
+			return &family->ops[i];
 
 	return NULL;
 }
@@ -126,7 +138,8 @@ static u16 genl_generate_id(void)
 	int i;
 
 	for (i = 0; i <= GENL_MAX_ID - GENL_MIN_ID; i++) {
-		if (!genl_family_find_byid(id_gen_idx))
+		if (id_gen_idx != GENL_ID_VFS_DQUOT &&
+		    !genl_family_find_byid(id_gen_idx))
 			return id_gen_idx;
 		if (++id_gen_idx > GENL_MAX_ID)
 			id_gen_idx = GENL_MIN_ID;
@@ -135,62 +148,110 @@ static u16 genl_generate_id(void)
 	return 0;
 }
 
-static struct genl_multicast_group notify_grp;
-
-/**
- * genl_register_mc_group - register a multicast group
- *
- * Registers the specified multicast group and notifies userspace
- * about the new group.
- *
- * Returns 0 on success or a negative error code.
- *
- * @family: The generic netlink family the group shall be registered for.
- * @grp: The group to register, must have a name.
- */
-int genl_register_mc_group(struct genl_family *family,
-			   struct genl_multicast_group *grp)
+static int genl_allocate_reserve_groups(int n_groups, int *first_id)
 {
-	int id;
 	unsigned long *new_groups;
-	int err = 0;
+	int start = 0;
+	int i;
+	int id;
+	bool fits;
+
+	do {
+		if (start == 0)
+			id = find_first_zero_bit(mc_groups,
+						 mc_groups_longs *
+						 BITS_PER_LONG);
+		else
+			id = find_next_zero_bit(mc_groups,
+						mc_groups_longs * BITS_PER_LONG,
+						start);
+
+		fits = true;
+		for (i = id;
+		     i < min_t(int, id + n_groups,
+			       mc_groups_longs * BITS_PER_LONG);
+		     i++) {
+			if (test_bit(i, mc_groups)) {
+				start = i;
+				fits = false;
+				break;
+			}
+		}
 
-	BUG_ON(grp->name[0] == '\0');
-	BUG_ON(memchr(grp->name, '\0', GENL_NAMSIZ) == NULL);
+		if (id >= mc_groups_longs * BITS_PER_LONG) {
+			unsigned long new_longs = mc_groups_longs +
+						  BITS_TO_LONGS(n_groups);
+			size_t nlen = new_longs * sizeof(unsigned long);
+
+			if (mc_groups == &mc_group_start) {
+				new_groups = kzalloc(nlen, GFP_KERNEL);
+				if (!new_groups)
+					return -ENOMEM;
+				mc_groups = new_groups;
+				*mc_groups = mc_group_start;
+			} else {
+				new_groups = krealloc(mc_groups, nlen,
+						      GFP_KERNEL);
+				if (!new_groups)
+					return -ENOMEM;
+				mc_groups = new_groups;
+				for (i = 0; i < BITS_TO_LONGS(n_groups); i++)
+					mc_groups[mc_groups_longs + i] = 0;
+			}
+			mc_groups_longs = new_longs;
+		}
+	} while (!fits);
 
-	genl_lock_all();
+	for (i = id; i < id + n_groups; i++)
+		set_bit(i, mc_groups);
+	*first_id = id;
+	return 0;
+}
 
-	/* special-case our own group */
-	if (grp == &notify_grp)
-		id = GENL_ID_CTRL;
-	else
-		id = find_first_zero_bit(mc_groups,
-					 mc_groups_longs * BITS_PER_LONG);
+static struct genl_family genl_ctrl;
 
+static int genl_validate_assign_mc_groups(struct genl_family *family)
+{
+	int first_id;
+	int n_groups = family->n_mcgrps;
+	int err, i;
+	bool groups_allocated = false;
 
-	if (id >= mc_groups_longs * BITS_PER_LONG) {
-		size_t nlen = (mc_groups_longs + 1) * sizeof(unsigned long);
+	if (!n_groups)
+		return 0;
 
-		if (mc_groups == &mc_group_start) {
-			new_groups = kzalloc(nlen, GFP_KERNEL);
-			if (!new_groups) {
-				err = -ENOMEM;
-				goto out;
-			}
-			mc_groups = new_groups;
-			*mc_groups = mc_group_start;
-		} else {
-			new_groups = krealloc(mc_groups, nlen, GFP_KERNEL);
-			if (!new_groups) {
-				err = -ENOMEM;
-				goto out;
-			}
-			mc_groups = new_groups;
-			mc_groups[mc_groups_longs] = 0;
-		}
-		mc_groups_longs++;
+	for (i = 0; i < n_groups; i++) {
+		const struct genl_multicast_group *grp = &family->mcgrps[i];
+
+		if (WARN_ON(grp->name[0] == '\0'))
+			return -EINVAL;
+		if (WARN_ON(memchr(grp->name, '\0', GENL_NAMSIZ) == NULL))
+			return -EINVAL;
+	}
+
+	/* special-case our own group and hacks */
+	if (family == &genl_ctrl) {
+		first_id = GENL_ID_CTRL;
+		BUG_ON(n_groups != 1);
+	} else if (strcmp(family->name, "NET_DM") == 0) {
+		first_id = 1;
+		BUG_ON(n_groups != 1);
+	} else if (strcmp(family->name, "VFS_DQUOT") == 0) {
+		first_id = GENL_ID_VFS_DQUOT;
+		BUG_ON(n_groups != 1);
+	} else {
+		groups_allocated = true;
+		err = genl_allocate_reserve_groups(n_groups, &first_id);
+		if (err)
+			return err;
 	}
 
+	family->mcgrp_offset = first_id;
+
+	/* if still initializing, can't and don't need to to realloc bitmaps */
+	if (!init_net.genl_sock)
+		return 0;
+
 	if (family->netnsok) {
 		struct net *net;
 
@@ -206,9 +267,7 @@ int genl_register_mc_group(struct genl_family *family,
 				 * number of _possible_ groups has been
 				 * increased on some sockets which is ok.
 				 */
-				rcu_read_unlock();
-				netlink_table_ungrab();
-				goto out;
+				break;
 			}
 		}
 		rcu_read_unlock();
@@ -216,152 +275,67 @@ int genl_register_mc_group(struct genl_family *family,
 	} else {
 		err = netlink_change_ngroups(init_net.genl_sock,
 					     mc_groups_longs * BITS_PER_LONG);
-		if (err)
-			goto out;
 	}
 
-	grp->id = id;
-	set_bit(id, mc_groups);
-	list_add_tail(&grp->list, &family->mcast_groups);
-	grp->family = family;
+	if (groups_allocated && err) {
+		for (i = 0; i < family->n_mcgrps; i++)
+			clear_bit(family->mcgrp_offset + i, mc_groups);
+	}
 
-	genl_ctrl_event(CTRL_CMD_NEWMCAST_GRP, grp);
- out:
-	genl_unlock_all();
 	return err;
 }
-EXPORT_SYMBOL(genl_register_mc_group);
 
-static void __genl_unregister_mc_group(struct genl_family *family,
-				       struct genl_multicast_group *grp)
+static void genl_unregister_mc_groups(struct genl_family *family)
 {
 	struct net *net;
-	BUG_ON(grp->family != family);
+	int i;
 
 	netlink_table_grab();
 	rcu_read_lock();
-	for_each_net_rcu(net)
-		__netlink_clear_multicast_users(net->genl_sock, grp->id);
+	for_each_net_rcu(net) {
+		for (i = 0; i < family->n_mcgrps; i++)
+			__netlink_clear_multicast_users(
+				net->genl_sock, family->mcgrp_offset + i);
+	}
 	rcu_read_unlock();
 	netlink_table_ungrab();
 
-	clear_bit(grp->id, mc_groups);
-	list_del(&grp->list);
-	genl_ctrl_event(CTRL_CMD_DELMCAST_GRP, grp);
-	grp->id = 0;
-	grp->family = NULL;
-}
+	for (i = 0; i < family->n_mcgrps; i++) {
+		int grp_id = family->mcgrp_offset + i;
 
-/**
- * genl_unregister_mc_group - unregister a multicast group
- *
- * Unregisters the specified multicast group and notifies userspace
- * about it. All current listeners on the group are removed.
- *
- * Note: It is not necessary to unregister all multicast groups before
- *       unregistering the family, unregistering the family will cause
- *       all assigned multicast groups to be unregistered automatically.
- *
- * @family: Generic netlink family the group belongs to.
- * @grp: The group to unregister, must have been registered successfully
- *	 previously.
- */
-void genl_unregister_mc_group(struct genl_family *family,
-			      struct genl_multicast_group *grp)
-{
-	genl_lock_all();
-	__genl_unregister_mc_group(family, grp);
-	genl_unlock_all();
+		if (grp_id != 1)
+			clear_bit(grp_id, mc_groups);
+		genl_ctrl_event(CTRL_CMD_DELMCAST_GRP, family,
+				&family->mcgrps[i], grp_id);
+	}
 }
-EXPORT_SYMBOL(genl_unregister_mc_group);
 
-static void genl_unregister_mc_groups(struct genl_family *family)
+static int genl_validate_ops(struct genl_family *family)
 {
-	struct genl_multicast_group *grp, *tmp;
+	const struct genl_ops *ops = family->ops;
+	unsigned int n_ops = family->n_ops;
+	int i, j;
 
-	list_for_each_entry_safe(grp, tmp, &family->mcast_groups, list)
-		__genl_unregister_mc_group(family, grp);
-}
-
-/**
- * genl_register_ops - register generic netlink operations
- * @family: generic netlink family
- * @ops: operations to be registered
- *
- * Registers the specified operations and assigns them to the specified
- * family. Either a doit or dumpit callback must be specified or the
- * operation will fail. Only one operation structure per command
- * identifier may be registered.
- *
- * See include/net/genetlink.h for more documenation on the operations
- * structure.
- *
- * Returns 0 on success or a negative error code.
- */
-int genl_register_ops(struct genl_family *family, struct genl_ops *ops)
-{
-	int err = -EINVAL;
+	if (WARN_ON(n_ops && !ops))
+		return -EINVAL;
 
-	if (ops->dumpit == NULL && ops->doit == NULL)
-		goto errout;
+	if (!n_ops)
+		return 0;
 
-	if (genl_get_cmd(ops->cmd, family)) {
-		err = -EEXIST;
-		goto errout;
+	for (i = 0; i < n_ops; i++) {
+		if (ops[i].dumpit == NULL && ops[i].doit == NULL)
+			return -EINVAL;
+		for (j = i + 1; j < n_ops; j++)
+			if (ops[i].cmd == ops[j].cmd)
+				return -EINVAL;
 	}
 
-	if (ops->dumpit)
-		ops->flags |= GENL_CMD_CAP_DUMP;
-	if (ops->doit)
-		ops->flags |= GENL_CMD_CAP_DO;
-	if (ops->policy)
-		ops->flags |= GENL_CMD_CAP_HASPOL;
+	/* family is not registered yet, so no locking needed */
+	family->ops = ops;
+	family->n_ops = n_ops;
 
-	genl_lock_all();
-	list_add_tail(&ops->ops_list, &family->ops_list);
-	genl_unlock_all();
-
-	genl_ctrl_event(CTRL_CMD_NEWOPS, ops);
-	err = 0;
-errout:
-	return err;
-}
-EXPORT_SYMBOL(genl_register_ops);
-
-/**
- * genl_unregister_ops - unregister generic netlink operations
- * @family: generic netlink family
- * @ops: operations to be unregistered
- *
- * Unregisters the specified operations and unassigns them from the
- * specified family. The operation blocks until the current message
- * processing has finished and doesn't start again until the
- * unregister process has finished.
- *
- * Note: It is not necessary to unregister all operations before
- *       unregistering the family, unregistering the family will cause
- *       all assigned operations to be unregistered automatically.
- *
- * Returns 0 on success or a negative error code.
- */
-int genl_unregister_ops(struct genl_family *family, struct genl_ops *ops)
-{
-	struct genl_ops *rc;
-
-	genl_lock_all();
-	list_for_each_entry(rc, &family->ops_list, ops_list) {
-		if (rc == ops) {
-			list_del(&ops->ops_list);
-			genl_unlock_all();
-			genl_ctrl_event(CTRL_CMD_DELOPS, ops);
-			return 0;
-		}
-	}
-	genl_unlock_all();
-
-	return -ENOENT;
+	return 0;
 }
-EXPORT_SYMBOL(genl_unregister_ops);
 
 /**
  * __genl_register_family - register a generic netlink family
@@ -372,11 +346,14 @@ EXPORT_SYMBOL(genl_unregister_ops);
  * The family id may equal GENL_ID_GENERATE causing an unique id to
  * be automatically generated and assigned.
  *
+ * The family's ops array must already be assigned, you can use the
+ * genl_register_family_with_ops() helper function.
+ *
  * Return 0 on success or a negative error code.
  */
 int __genl_register_family(struct genl_family *family)
 {
-	int err = -EINVAL;
+	int err = -EINVAL, i;
 
 	if (family->id && family->id < GENL_MIN_ID)
 		goto errout;
@@ -384,8 +361,9 @@ int __genl_register_family(struct genl_family *family)
 	if (family->id > GENL_MAX_ID)
 		goto errout;
 
-	INIT_LIST_HEAD(&family->ops_list);
-	INIT_LIST_HEAD(&family->mcast_groups);
+	err = genl_validate_ops(family);
+	if (err)
+		return err;
 
 	genl_lock_all();
 
@@ -418,10 +396,18 @@ int __genl_register_family(struct genl_family *family)
 	} else
 		family->attrbuf = NULL;
 
+	err = genl_validate_assign_mc_groups(family);
+	if (err)
+		goto errout_locked;
+
 	list_add_tail(&family->family_list, genl_family_chain(family->id));
 	genl_unlock_all();
 
-	genl_ctrl_event(CTRL_CMD_NEWFAMILY, family);
+	/* send all events */
+	genl_ctrl_event(CTRL_CMD_NEWFAMILY, family, NULL, 0);
+	for (i = 0; i < family->n_mcgrps; i++)
+		genl_ctrl_event(CTRL_CMD_NEWMCAST_GRP, family,
+				&family->mcgrps[i], family->mcgrp_offset + i);
 
 	return 0;
 
@@ -433,52 +419,6 @@ errout:
 EXPORT_SYMBOL(__genl_register_family);
 
 /**
- * __genl_register_family_with_ops - register a generic netlink family
- * @family: generic netlink family
- * @ops: operations to be registered
- * @n_ops: number of elements to register
- *
- * Registers the specified family and operations from the specified table.
- * Only one family may be registered with the same family name or identifier.
- *
- * The family id may equal GENL_ID_GENERATE causing an unique id to
- * be automatically generated and assigned.
- *
- * Either a doit or dumpit callback must be specified for every registered
- * operation or the function will fail. Only one operation structure per
- * command identifier may be registered.
- *
- * See include/net/genetlink.h for more documenation on the operations
- * structure.
- *
- * This is equivalent to calling genl_register_family() followed by
- * genl_register_ops() for every operation entry in the table taking
- * care to unregister the family on error path.
- *
- * Return 0 on success or a negative error code.
- */
-int __genl_register_family_with_ops(struct genl_family *family,
-	struct genl_ops *ops, size_t n_ops)
-{
-	int err, i;
-
-	err = __genl_register_family(family);
-	if (err)
-		return err;
-
-	for (i = 0; i < n_ops; ++i, ++ops) {
-		err = genl_register_ops(family, ops);
-		if (err)
-			goto err_out;
-	}
-	return 0;
-err_out:
-	genl_unregister_family(family);
-	return err;
-}
-EXPORT_SYMBOL(__genl_register_family_with_ops);
-
-/**
  * genl_unregister_family - unregister generic netlink family
  * @family: generic netlink family
  *
@@ -499,11 +439,11 @@ int genl_unregister_family(struct genl_family *family)
 			continue;
 
 		list_del(&rc->family_list);
-		INIT_LIST_HEAD(&family->ops_list);
+		family->n_ops = 0;
 		genl_unlock_all();
 
 		kfree(family->attrbuf);
-		genl_ctrl_event(CTRL_CMD_DELFAMILY, family);
+		genl_ctrl_event(CTRL_CMD_DELFAMILY, family, NULL, 0);
 		return 0;
 	}
 
@@ -546,7 +486,8 @@ EXPORT_SYMBOL(genlmsg_put);
 
 static int genl_lock_dumpit(struct sk_buff *skb, struct netlink_callback *cb)
 {
-	struct genl_ops *ops = cb->data;
+	/* our ops are always const - netlink API doesn't propagate that */
+	const struct genl_ops *ops = cb->data;
 	int rc;
 
 	genl_lock();
@@ -557,7 +498,8 @@ static int genl_lock_dumpit(struct sk_buff *skb, struct netlink_callback *cb)
 
 static int genl_lock_done(struct netlink_callback *cb)
 {
-	struct genl_ops *ops = cb->data;
+	/* our ops are always const - netlink API doesn't propagate that */
+	const struct genl_ops *ops = cb->data;
 	int rc = 0;
 
 	if (ops->done) {
@@ -572,7 +514,7 @@ static int genl_family_rcv_msg(struct genl_family *family,
 			       struct sk_buff *skb,
 			       struct nlmsghdr *nlh)
 {
-	struct genl_ops *ops;
+	const struct genl_ops *ops;
 	struct net *net = sock_net(skb->sk);
 	struct genl_info info;
 	struct genlmsghdr *hdr = nlmsg_data(nlh);
@@ -604,7 +546,8 @@ static int genl_family_rcv_msg(struct genl_family *family,
 		if (!family->parallel_ops) {
 			struct netlink_dump_control c = {
 				.module = family->module,
-				.data = ops,
+				/* we have const, but the netlink API doesn't */
+				.data = (void *)ops,
 				.dump = genl_lock_dumpit,
 				.done = genl_lock_done,
 			};
@@ -726,24 +669,32 @@ static int ctrl_fill_info(struct genl_family *family, u32 portid, u32 seq,
 	    nla_put_u32(skb, CTRL_ATTR_MAXATTR, family->maxattr))
 		goto nla_put_failure;
 
-	if (!list_empty(&family->ops_list)) {
+	if (family->n_ops) {
 		struct nlattr *nla_ops;
-		struct genl_ops *ops;
-		int idx = 1;
+		int i;
 
 		nla_ops = nla_nest_start(skb, CTRL_ATTR_OPS);
 		if (nla_ops == NULL)
 			goto nla_put_failure;
 
-		list_for_each_entry(ops, &family->ops_list, ops_list) {
+		for (i = 0; i < family->n_ops; i++) {
 			struct nlattr *nest;
+			const struct genl_ops *ops = &family->ops[i];
+			u32 op_flags = ops->flags;
 
-			nest = nla_nest_start(skb, idx++);
+			if (ops->dumpit)
+				op_flags |= GENL_CMD_CAP_DUMP;
+			if (ops->doit)
+				op_flags |= GENL_CMD_CAP_DO;
+			if (ops->policy)
+				op_flags |= GENL_CMD_CAP_HASPOL;
+
+			nest = nla_nest_start(skb, i + 1);
 			if (nest == NULL)
 				goto nla_put_failure;
 
 			if (nla_put_u32(skb, CTRL_ATTR_OP_ID, ops->cmd) ||
-			    nla_put_u32(skb, CTRL_ATTR_OP_FLAGS, ops->flags))
+			    nla_put_u32(skb, CTRL_ATTR_OP_FLAGS, op_flags))
 				goto nla_put_failure;
 
 			nla_nest_end(skb, nest);
@@ -752,23 +703,26 @@ static int ctrl_fill_info(struct genl_family *family, u32 portid, u32 seq,
 		nla_nest_end(skb, nla_ops);
 	}
 
-	if (!list_empty(&family->mcast_groups)) {
-		struct genl_multicast_group *grp;
+	if (family->n_mcgrps) {
 		struct nlattr *nla_grps;
-		int idx = 1;
+		int i;
 
 		nla_grps = nla_nest_start(skb, CTRL_ATTR_MCAST_GROUPS);
 		if (nla_grps == NULL)
 			goto nla_put_failure;
 
-		list_for_each_entry(grp, &family->mcast_groups, list) {
+		for (i = 0; i < family->n_mcgrps; i++) {
 			struct nlattr *nest;
+			const struct genl_multicast_group *grp;
+
+			grp = &family->mcgrps[i];
 
-			nest = nla_nest_start(skb, idx++);
+			nest = nla_nest_start(skb, i + 1);
 			if (nest == NULL)
 				goto nla_put_failure;
 
-			if (nla_put_u32(skb, CTRL_ATTR_MCAST_GRP_ID, grp->id) ||
+			if (nla_put_u32(skb, CTRL_ATTR_MCAST_GRP_ID,
+					family->mcgrp_offset + i) ||
 			    nla_put_string(skb, CTRL_ATTR_MCAST_GRP_NAME,
 					   grp->name))
 				goto nla_put_failure;
@@ -785,9 +739,10 @@ nla_put_failure:
 	return -EMSGSIZE;
 }
 
-static int ctrl_fill_mcgrp_info(struct genl_multicast_group *grp, u32 portid,
-				u32 seq, u32 flags, struct sk_buff *skb,
-				u8 cmd)
+static int ctrl_fill_mcgrp_info(struct genl_family *family,
+				const struct genl_multicast_group *grp,
+				int grp_id, u32 portid, u32 seq, u32 flags,
+				struct sk_buff *skb, u8 cmd)
 {
 	void *hdr;
 	struct nlattr *nla_grps;
@@ -797,8 +752,8 @@ static int ctrl_fill_mcgrp_info(struct genl_multicast_group *grp, u32 portid,
 	if (hdr == NULL)
 		return -1;
 
-	if (nla_put_string(skb, CTRL_ATTR_FAMILY_NAME, grp->family->name) ||
-	    nla_put_u16(skb, CTRL_ATTR_FAMILY_ID, grp->family->id))
+	if (nla_put_string(skb, CTRL_ATTR_FAMILY_NAME, family->name) ||
+	    nla_put_u16(skb, CTRL_ATTR_FAMILY_ID, family->id))
 		goto nla_put_failure;
 
 	nla_grps = nla_nest_start(skb, CTRL_ATTR_MCAST_GROUPS);
@@ -809,7 +764,7 @@ static int ctrl_fill_mcgrp_info(struct genl_multicast_group *grp, u32 portid,
 	if (nest == NULL)
 		goto nla_put_failure;
 
-	if (nla_put_u32(skb, CTRL_ATTR_MCAST_GRP_ID, grp->id) ||
+	if (nla_put_u32(skb, CTRL_ATTR_MCAST_GRP_ID, grp_id) ||
 	    nla_put_string(skb, CTRL_ATTR_MCAST_GRP_NAME,
 			   grp->name))
 		goto nla_put_failure;
@@ -875,8 +830,10 @@ static struct sk_buff *ctrl_build_family_msg(struct genl_family *family,
 	return skb;
 }
 
-static struct sk_buff *ctrl_build_mcgrp_msg(struct genl_multicast_group *grp,
-					    u32 portid, int seq, u8 cmd)
+static struct sk_buff *
+ctrl_build_mcgrp_msg(struct genl_family *family,
+		     const struct genl_multicast_group *grp,
+		     int grp_id, u32 portid, int seq, u8 cmd)
 {
 	struct sk_buff *skb;
 	int err;
@@ -885,7 +842,8 @@ static struct sk_buff *ctrl_build_mcgrp_msg(struct genl_multicast_group *grp,
 	if (skb == NULL)
 		return ERR_PTR(-ENOBUFS);
 
-	err = ctrl_fill_mcgrp_info(grp, portid, seq, 0, skb, cmd);
+	err = ctrl_fill_mcgrp_info(family, grp, grp_id, portid,
+				   seq, 0, skb, cmd);
 	if (err < 0) {
 		nlmsg_free(skb);
 		return ERR_PTR(err);
@@ -947,11 +905,11 @@ static int ctrl_getfamily(struct sk_buff *skb, struct genl_info *info)
 	return genlmsg_reply(msg, info);
 }
 
-static int genl_ctrl_event(int event, void *data)
+static int genl_ctrl_event(int event, struct genl_family *family,
+			   const struct genl_multicast_group *grp,
+			   int grp_id)
 {
 	struct sk_buff *msg;
-	struct genl_family *family;
-	struct genl_multicast_group *grp;
 
 	/* genl is still initialising */
 	if (!init_net.genl_sock)
@@ -960,14 +918,13 @@ static int genl_ctrl_event(int event, void *data)
 	switch (event) {
 	case CTRL_CMD_NEWFAMILY:
 	case CTRL_CMD_DELFAMILY:
-		family = data;
+		WARN_ON(grp);
 		msg = ctrl_build_family_msg(family, 0, 0, event);
 		break;
 	case CTRL_CMD_NEWMCAST_GRP:
 	case CTRL_CMD_DELMCAST_GRP:
-		grp = data;
-		family = grp->family;
-		msg = ctrl_build_mcgrp_msg(data, 0, 0, event);
+		BUG_ON(!grp);
+		msg = ctrl_build_mcgrp_msg(family, grp, grp_id, 0, 0, event);
 		break;
 	default:
 		return -EINVAL;
@@ -977,26 +934,29 @@ static int genl_ctrl_event(int event, void *data)
 		return PTR_ERR(msg);
 
 	if (!family->netnsok) {
-		genlmsg_multicast_netns(&init_net, msg, 0,
-					GENL_ID_CTRL, GFP_KERNEL);
+		genlmsg_multicast_netns(&genl_ctrl, &init_net, msg, 0,
+					0, GFP_KERNEL);
 	} else {
 		rcu_read_lock();
-		genlmsg_multicast_allns(msg, 0, GENL_ID_CTRL, GFP_ATOMIC);
+		genlmsg_multicast_allns(&genl_ctrl, msg, 0,
+					0, GFP_ATOMIC);
 		rcu_read_unlock();
 	}
 
 	return 0;
 }
 
-static struct genl_ops genl_ctrl_ops = {
-	.cmd		= CTRL_CMD_GETFAMILY,
-	.doit		= ctrl_getfamily,
-	.dumpit		= ctrl_dumpfamily,
-	.policy		= ctrl_policy,
+static struct genl_ops genl_ctrl_ops[] = {
+	{
+		.cmd		= CTRL_CMD_GETFAMILY,
+		.doit		= ctrl_getfamily,
+		.dumpit		= ctrl_dumpfamily,
+		.policy		= ctrl_policy,
+	},
 };
 
-static struct genl_multicast_group notify_grp = {
-	.name		= "notify",
+static struct genl_multicast_group genl_ctrl_groups[] = {
+	{ .name = "notify", },
 };
 
 static int __net_init genl_pernet_init(struct net *net)
@@ -1036,7 +996,8 @@ static int __init genl_init(void)
 	for (i = 0; i < GENL_FAM_TAB_SIZE; i++)
 		INIT_LIST_HEAD(&family_ht[i]);
 
-	err = genl_register_family_with_ops(&genl_ctrl, &genl_ctrl_ops, 1);
+	err = genl_register_family_with_ops_groups(&genl_ctrl, genl_ctrl_ops,
+						   genl_ctrl_groups);
 	if (err < 0)
 		goto problem;
 
@@ -1044,10 +1005,6 @@ static int __init genl_init(void)
 	if (err)
 		goto problem;
 
-	err = genl_register_mc_group(&genl_ctrl, &notify_grp);
-	if (err < 0)
-		goto problem;
-
 	return 0;
 
 problem:
@@ -1085,14 +1042,18 @@ static int genlmsg_mcast(struct sk_buff *skb, u32 portid, unsigned long group,
 	return err;
 }
 
-int genlmsg_multicast_allns(struct sk_buff *skb, u32 portid, unsigned int group,
-			    gfp_t flags)
+int genlmsg_multicast_allns(struct genl_family *family, struct sk_buff *skb,
+			    u32 portid, unsigned int group, gfp_t flags)
 {
+	if (group >= family->n_mcgrps)
+		return -EINVAL;
+	group = family->mcgrp_offset + group;
 	return genlmsg_mcast(skb, portid, group, flags);
 }
 EXPORT_SYMBOL(genlmsg_multicast_allns);
 
-void genl_notify(struct sk_buff *skb, struct net *net, u32 portid, u32 group,
+void genl_notify(struct genl_family *family,
+		 struct sk_buff *skb, struct net *net, u32 portid, u32 group,
 		 struct nlmsghdr *nlh, gfp_t flags)
 {
 	struct sock *sk = net->genl_sock;
@@ -1101,6 +1062,9 @@ void genl_notify(struct sk_buff *skb, struct net *net, u32 portid, u32 group,
 	if (nlh)
 		report = nlmsg_report(nlh);
 
+	if (group >= family->n_mcgrps)
+		return;
+	group = family->mcgrp_offset + group;
 	nlmsg_notify(sk, skb, portid, group, report, flags);
 }
 EXPORT_SYMBOL(genl_notify);
diff --git a/net/nfc/netlink.c b/net/nfc/netlink.c
index 84b7e3ea7b7a..a9b2342d5253 100644
--- a/net/nfc/netlink.c
+++ b/net/nfc/netlink.c
@@ -30,8 +30,8 @@
 #include "nfc.h"
 #include "llcp.h"
 
-static struct genl_multicast_group nfc_genl_event_mcgrp = {
-	.name = NFC_GENL_MCAST_EVENT_NAME,
+static const struct genl_multicast_group nfc_genl_mcgrps[] = {
+	{ .name = NFC_GENL_MCAST_EVENT_NAME, },
 };
 
 static struct genl_family nfc_genl_family = {
@@ -194,7 +194,7 @@ int nfc_genl_targets_found(struct nfc_dev *dev)
 
 	genlmsg_end(msg, hdr);
 
-	return genlmsg_multicast(msg, 0, nfc_genl_event_mcgrp.id, GFP_ATOMIC);
+	return genlmsg_multicast(&nfc_genl_family, msg, 0, 0, GFP_ATOMIC);
 
 nla_put_failure:
 	genlmsg_cancel(msg, hdr);
@@ -223,7 +223,7 @@ int nfc_genl_target_lost(struct nfc_dev *dev, u32 target_idx)
 
 	genlmsg_end(msg, hdr);
 
-	genlmsg_multicast(msg, 0, nfc_genl_event_mcgrp.id, GFP_KERNEL);
+	genlmsg_multicast(&nfc_genl_family, msg, 0, 0, GFP_KERNEL);
 
 	return 0;
 
@@ -255,7 +255,7 @@ int nfc_genl_tm_activated(struct nfc_dev *dev, u32 protocol)
 
 	genlmsg_end(msg, hdr);
 
-	genlmsg_multicast(msg, 0, nfc_genl_event_mcgrp.id, GFP_KERNEL);
+	genlmsg_multicast(&nfc_genl_family, msg, 0, 0, GFP_KERNEL);
 
 	return 0;
 
@@ -285,7 +285,7 @@ int nfc_genl_tm_deactivated(struct nfc_dev *dev)
 
 	genlmsg_end(msg, hdr);
 
-	genlmsg_multicast(msg, 0, nfc_genl_event_mcgrp.id, GFP_KERNEL);
+	genlmsg_multicast(&nfc_genl_family, msg, 0, 0, GFP_KERNEL);
 
 	return 0;
 
@@ -318,7 +318,7 @@ int nfc_genl_device_added(struct nfc_dev *dev)
 
 	genlmsg_end(msg, hdr);
 
-	genlmsg_multicast(msg, 0, nfc_genl_event_mcgrp.id, GFP_KERNEL);
+	genlmsg_multicast(&nfc_genl_family, msg, 0, 0, GFP_KERNEL);
 
 	return 0;
 
@@ -348,7 +348,7 @@ int nfc_genl_device_removed(struct nfc_dev *dev)
 
 	genlmsg_end(msg, hdr);
 
-	genlmsg_multicast(msg, 0, nfc_genl_event_mcgrp.id, GFP_KERNEL);
+	genlmsg_multicast(&nfc_genl_family, msg, 0, 0, GFP_KERNEL);
 
 	return 0;
 
@@ -414,7 +414,7 @@ int nfc_genl_llc_send_sdres(struct nfc_dev *dev, struct hlist_head *sdres_list)
 
 	genlmsg_end(msg, hdr);
 
-	return genlmsg_multicast(msg, 0, nfc_genl_event_mcgrp.id, GFP_ATOMIC);
+	return genlmsg_multicast(&nfc_genl_family, msg, 0, 0, GFP_ATOMIC);
 
 nla_put_failure:
 	genlmsg_cancel(msg, hdr);
@@ -448,7 +448,7 @@ int nfc_genl_se_added(struct nfc_dev *dev, u32 se_idx, u16 type)
 
 	genlmsg_end(msg, hdr);
 
-	genlmsg_multicast(msg, 0, nfc_genl_event_mcgrp.id, GFP_KERNEL);
+	genlmsg_multicast(&nfc_genl_family, msg, 0, 0, GFP_KERNEL);
 
 	return 0;
 
@@ -479,7 +479,7 @@ int nfc_genl_se_removed(struct nfc_dev *dev, u32 se_idx)
 
 	genlmsg_end(msg, hdr);
 
-	genlmsg_multicast(msg, 0, nfc_genl_event_mcgrp.id, GFP_KERNEL);
+	genlmsg_multicast(&nfc_genl_family, msg, 0, 0, GFP_KERNEL);
 
 	return 0;
 
@@ -600,7 +600,7 @@ int nfc_genl_dep_link_up_event(struct nfc_dev *dev, u32 target_idx,
 
 	dev->dep_link_up = true;
 
-	genlmsg_multicast(msg, 0, nfc_genl_event_mcgrp.id, GFP_ATOMIC);
+	genlmsg_multicast(&nfc_genl_family, msg, 0, 0, GFP_ATOMIC);
 
 	return 0;
 
@@ -632,7 +632,7 @@ int nfc_genl_dep_link_down_event(struct nfc_dev *dev)
 
 	genlmsg_end(msg, hdr);
 
-	genlmsg_multicast(msg, 0, nfc_genl_event_mcgrp.id, GFP_ATOMIC);
+	genlmsg_multicast(&nfc_genl_family, msg, 0, 0, GFP_ATOMIC);
 
 	return 0;
 
@@ -1137,7 +1137,7 @@ int nfc_genl_fw_download_done(struct nfc_dev *dev, const char *firmware_name,
 
 	genlmsg_end(msg, hdr);
 
-	genlmsg_multicast(msg, 0, nfc_genl_event_mcgrp.id, GFP_KERNEL);
+	genlmsg_multicast(&nfc_genl_family, msg, 0, 0, GFP_KERNEL);
 
 	return 0;
 
@@ -1308,7 +1308,7 @@ static void se_io_cb(void *context, u8 *apdu, size_t apdu_len, int err)
 
 	genlmsg_end(msg, hdr);
 
-	genlmsg_multicast(msg, 0, nfc_genl_event_mcgrp.id, GFP_KERNEL);
+	genlmsg_multicast(&nfc_genl_family, msg, 0, 0, GFP_KERNEL);
 
 	kfree(ctx);
 
@@ -1364,7 +1364,7 @@ static int nfc_genl_se_io(struct sk_buff *skb, struct genl_info *info)
 	return dev->ops->se_io(dev, se_idx, apdu, apdu_len, se_io_cb, ctx);
 }
 
-static struct genl_ops nfc_genl_ops[] = {
+static const struct genl_ops nfc_genl_ops[] = {
 	{
 		.cmd = NFC_CMD_GET_DEVICE,
 		.doit = nfc_genl_get_device,
@@ -1536,16 +1536,15 @@ int __init nfc_genl_init(void)
 {
 	int rc;
 
-	rc = genl_register_family_with_ops(&nfc_genl_family, nfc_genl_ops,
-					   ARRAY_SIZE(nfc_genl_ops));
+	rc = genl_register_family_with_ops_groups(&nfc_genl_family,
+						  nfc_genl_ops,
+						  nfc_genl_mcgrps);
 	if (rc)
 		return rc;
 
-	rc = genl_register_mc_group(&nfc_genl_family, &nfc_genl_event_mcgrp);
-
 	netlink_register_notifier(&nl_notifier);
 
-	return rc;
+	return 0;
 }
 
 /**
diff --git a/net/openvswitch/datapath.c b/net/openvswitch/datapath.c
index 1408adc2a2a7..6f5e1dd3be2d 100644
--- a/net/openvswitch/datapath.c
+++ b/net/openvswitch/datapath.c
@@ -61,11 +61,11 @@
 
 int ovs_net_id __read_mostly;
 
-static void ovs_notify(struct sk_buff *skb, struct genl_info *info,
-		       struct genl_multicast_group *grp)
+static void ovs_notify(struct genl_family *family,
+		       struct sk_buff *skb, struct genl_info *info)
 {
-	genl_notify(skb, genl_info_net(info), info->snd_portid,
-		    grp->id, info->nlhdr, GFP_KERNEL);
+	genl_notify(family, skb, genl_info_net(info), info->snd_portid,
+		    0, info->nlhdr, GFP_KERNEL);
 }
 
 /**
@@ -557,7 +557,7 @@ static const struct nla_policy packet_policy[OVS_PACKET_ATTR_MAX + 1] = {
 	[OVS_PACKET_ATTR_ACTIONS] = { .type = NLA_NESTED },
 };
 
-static struct genl_ops dp_packet_genl_ops[] = {
+static const struct genl_ops dp_packet_genl_ops[] = {
 	{ .cmd = OVS_PACKET_CMD_EXECUTE,
 	  .flags = GENL_ADMIN_PERM, /* Requires CAP_NET_ADMIN privilege. */
 	  .policy = packet_policy,
@@ -877,10 +877,10 @@ static int ovs_flow_cmd_new_or_set(struct sk_buff *skb, struct genl_info *info)
 	ovs_unlock();
 
 	if (!IS_ERR(reply))
-		ovs_notify(reply, info, &ovs_dp_flow_multicast_group);
+		ovs_notify(&dp_flow_genl_family, reply, info);
 	else
-		netlink_set_err(sock_net(skb->sk)->genl_sock, 0,
-				ovs_dp_flow_multicast_group.id, PTR_ERR(reply));
+		genl_set_err(&dp_flow_genl_family, sock_net(skb->sk), 0,
+			     0, PTR_ERR(reply));
 	return 0;
 
 err_flow_free:
@@ -990,7 +990,7 @@ static int ovs_flow_cmd_del(struct sk_buff *skb, struct genl_info *info)
 	ovs_flow_free(flow, true);
 	ovs_unlock();
 
-	ovs_notify(reply, info, &ovs_dp_flow_multicast_group);
+	ovs_notify(&dp_flow_genl_family, reply, info);
 	return 0;
 unlock:
 	ovs_unlock();
@@ -1034,7 +1034,7 @@ static int ovs_flow_cmd_dump(struct sk_buff *skb, struct netlink_callback *cb)
 	return skb->len;
 }
 
-static struct genl_ops dp_flow_genl_ops[] = {
+static const struct genl_ops dp_flow_genl_ops[] = {
 	{ .cmd = OVS_FLOW_CMD_NEW,
 	  .flags = GENL_ADMIN_PERM, /* Requires CAP_NET_ADMIN privilege. */
 	  .policy = flow_policy,
@@ -1199,6 +1199,12 @@ static int ovs_dp_cmd_new(struct sk_buff *skb, struct genl_info *info)
 		goto err_destroy_table;
 	}
 
+	for_each_possible_cpu(i) {
+		struct dp_stats_percpu *dpath_stats;
+		dpath_stats = per_cpu_ptr(dp->stats_percpu, i);
+		u64_stats_init(&dpath_stats->sync);
+	}
+
 	dp->ports = kmalloc(DP_VPORT_HASH_BUCKETS * sizeof(struct hlist_head),
 			    GFP_KERNEL);
 	if (!dp->ports) {
@@ -1237,7 +1243,7 @@ static int ovs_dp_cmd_new(struct sk_buff *skb, struct genl_info *info)
 
 	ovs_unlock();
 
-	ovs_notify(reply, info, &ovs_dp_datapath_multicast_group);
+	ovs_notify(&dp_datapath_genl_family, reply, info);
 	return 0;
 
 err_destroy_local_port:
@@ -1302,7 +1308,7 @@ static int ovs_dp_cmd_del(struct sk_buff *skb, struct genl_info *info)
 	__dp_destroy(dp);
 	ovs_unlock();
 
-	ovs_notify(reply, info, &ovs_dp_datapath_multicast_group);
+	ovs_notify(&dp_datapath_genl_family, reply, info);
 
 	return 0;
 unlock:
@@ -1326,14 +1332,14 @@ static int ovs_dp_cmd_set(struct sk_buff *skb, struct genl_info *info)
 				      info->snd_seq, OVS_DP_CMD_NEW);
 	if (IS_ERR(reply)) {
 		err = PTR_ERR(reply);
-		netlink_set_err(sock_net(skb->sk)->genl_sock, 0,
-				ovs_dp_datapath_multicast_group.id, err);
+		genl_set_err(&dp_datapath_genl_family, sock_net(skb->sk), 0,
+			     0, err);
 		err = 0;
 		goto unlock;
 	}
 
 	ovs_unlock();
-	ovs_notify(reply, info, &ovs_dp_datapath_multicast_group);
+	ovs_notify(&dp_datapath_genl_family, reply, info);
 
 	return 0;
 unlock:
@@ -1392,7 +1398,7 @@ static int ovs_dp_cmd_dump(struct sk_buff *skb, struct netlink_callback *cb)
 	return skb->len;
 }
 
-static struct genl_ops dp_datapath_genl_ops[] = {
+static const struct genl_ops dp_datapath_genl_ops[] = {
 	{ .cmd = OVS_DP_CMD_NEW,
 	  .flags = GENL_ADMIN_PERM, /* Requires CAP_NET_ADMIN privilege. */
 	  .policy = datapath_policy,
@@ -1425,7 +1431,7 @@ static const struct nla_policy vport_policy[OVS_VPORT_ATTR_MAX + 1] = {
 	[OVS_VPORT_ATTR_OPTIONS] = { .type = NLA_NESTED },
 };
 
-static struct genl_family dp_vport_genl_family = {
+struct genl_family dp_vport_genl_family = {
 	.id = GENL_ID_GENERATE,
 	.hdrsize = sizeof(struct ovs_header),
 	.name = OVS_VPORT_FAMILY,
@@ -1595,7 +1601,7 @@ static int ovs_vport_cmd_new(struct sk_buff *skb, struct genl_info *info)
 		goto exit_unlock;
 	}
 
-	ovs_notify(reply, info, &ovs_dp_vport_multicast_group);
+	ovs_notify(&dp_vport_genl_family, reply, info);
 
 exit_unlock:
 	ovs_unlock();
@@ -1642,7 +1648,7 @@ static int ovs_vport_cmd_set(struct sk_buff *skb, struct genl_info *info)
 	BUG_ON(err < 0);
 
 	ovs_unlock();
-	ovs_notify(reply, info, &ovs_dp_vport_multicast_group);
+	ovs_notify(&dp_vport_genl_family, reply, info);
 	return 0;
 
 exit_free:
@@ -1679,7 +1685,7 @@ static int ovs_vport_cmd_del(struct sk_buff *skb, struct genl_info *info)
 	err = 0;
 	ovs_dp_detach_port(vport);
 
-	ovs_notify(reply, info, &ovs_dp_vport_multicast_group);
+	ovs_notify(&dp_vport_genl_family, reply, info);
 
 exit_unlock:
 	ovs_unlock();
@@ -1753,7 +1759,7 @@ out:
 	return skb->len;
 }
 
-static struct genl_ops dp_vport_genl_ops[] = {
+static const struct genl_ops dp_vport_genl_ops[] = {
 	{ .cmd = OVS_VPORT_CMD_NEW,
 	  .flags = GENL_ADMIN_PERM, /* Requires CAP_NET_ADMIN privilege. */
 	  .policy = vport_policy,
@@ -1779,9 +1785,9 @@ static struct genl_ops dp_vport_genl_ops[] = {
 
 struct genl_family_and_ops {
 	struct genl_family *family;
-	struct genl_ops *ops;
+	const struct genl_ops *ops;
 	int n_ops;
-	struct genl_multicast_group *group;
+	const struct genl_multicast_group *group;
 };
 
 static const struct genl_family_and_ops dp_genl_families[] = {
@@ -1817,17 +1823,14 @@ static int dp_register_genl(void)
 	for (i = 0; i < ARRAY_SIZE(dp_genl_families); i++) {
 		const struct genl_family_and_ops *f = &dp_genl_families[i];
 
-		err = genl_register_family_with_ops(f->family, f->ops,
-						    f->n_ops);
+		f->family->ops = f->ops;
+		f->family->n_ops = f->n_ops;
+		f->family->mcgrps = f->group;
+		f->family->n_mcgrps = f->group ? 1 : 0;
+		err = genl_register_family(f->family);
 		if (err)
 			goto error;
 		n_registered++;
-
-		if (f->group) {
-			err = genl_register_mc_group(f->family, f->group);
-			if (err)
-				goto error;
-		}
 	}
 
 	return 0;
diff --git a/net/openvswitch/datapath.h b/net/openvswitch/datapath.h
index d3d14a58aa91..4067ea41be28 100644
--- a/net/openvswitch/datapath.h
+++ b/net/openvswitch/datapath.h
@@ -177,6 +177,7 @@ static inline struct vport *ovs_vport_ovsl(const struct datapath *dp, int port_n
 }
 
 extern struct notifier_block ovs_dp_device_notifier;
+extern struct genl_family dp_vport_genl_family;
 extern struct genl_multicast_group ovs_dp_vport_multicast_group;
 
 void ovs_dp_process_received_packet(struct vport *, struct sk_buff *);
diff --git a/net/openvswitch/dp_notify.c b/net/openvswitch/dp_notify.c
index 5c2dab276109..2c631fe76be1 100644
--- a/net/openvswitch/dp_notify.c
+++ b/net/openvswitch/dp_notify.c
@@ -34,15 +34,14 @@ static void dp_detach_port_notify(struct vport *vport)
 					  OVS_VPORT_CMD_DEL);
 	ovs_dp_detach_port(vport);
 	if (IS_ERR(notify)) {
-		netlink_set_err(ovs_dp_get_net(dp)->genl_sock, 0,
-				ovs_dp_vport_multicast_group.id,
-				PTR_ERR(notify));
+		genl_set_err(&dp_vport_genl_family, ovs_dp_get_net(dp), 0,
+			     0, PTR_ERR(notify));
 		return;
 	}
 
-	genlmsg_multicast_netns(ovs_dp_get_net(dp), notify, 0,
-				ovs_dp_vport_multicast_group.id,
-				GFP_KERNEL);
+	genlmsg_multicast_netns(&dp_vport_genl_family,
+				ovs_dp_get_net(dp), notify, 0,
+				0, GFP_KERNEL);
 }
 
 void ovs_dp_notify_wq(struct work_struct *work)
diff --git a/net/openvswitch/vport.c b/net/openvswitch/vport.c
index 6f65dbe13812..d830a95f03a4 100644
--- a/net/openvswitch/vport.c
+++ b/net/openvswitch/vport.c
@@ -118,6 +118,7 @@ struct vport *ovs_vport_alloc(int priv_size, const struct vport_ops *ops,
 {
 	struct vport *vport;
 	size_t alloc_size;
+	int i;
 
 	alloc_size = sizeof(struct vport);
 	if (priv_size) {
@@ -141,6 +142,13 @@ struct vport *ovs_vport_alloc(int priv_size, const struct vport_ops *ops,
 		return ERR_PTR(-ENOMEM);
 	}
 
+	for_each_possible_cpu(i) {
+		struct pcpu_tstats *vport_stats;
+		vport_stats = per_cpu_ptr(vport->percpu_stats, i);
+		u64_stats_init(&vport_stats->syncp);
+	}
+
+
 	spin_lock_init(&vport->stats_lock);
 
 	return vport;
diff --git a/net/phonet/datagram.c b/net/phonet/datagram.c
index 12c30f3e643e..38946b26e471 100644
--- a/net/phonet/datagram.c
+++ b/net/phonet/datagram.c
@@ -139,9 +139,6 @@ static int pn_recvmsg(struct kiocb *iocb, struct sock *sk,
 			MSG_CMSG_COMPAT))
 		goto out_nofree;
 
-	if (addr_len)
-		*addr_len = sizeof(sa);
-
 	skb = skb_recv_datagram(sk, flags, noblock, &rval);
 	if (skb == NULL)
 		goto out_nofree;
@@ -162,8 +159,10 @@ static int pn_recvmsg(struct kiocb *iocb, struct sock *sk,
 
 	rval = (flags & MSG_TRUNC) ? skb->len : copylen;
 
-	if (msg->msg_name != NULL)
-		memcpy(msg->msg_name, &sa, sizeof(struct sockaddr_pn));
+	if (msg->msg_name != NULL) {
+		memcpy(msg->msg_name, &sa, sizeof(sa));
+		*addr_len = sizeof(sa);
+	}
 
 out:
 	skb_free_datagram(sk, skb);
diff --git a/net/phonet/socket.c b/net/phonet/socket.c
index 77e38f733496..008214a3d5eb 100644
--- a/net/phonet/socket.c
+++ b/net/phonet/socket.c
@@ -595,26 +595,25 @@ static void pn_sock_seq_stop(struct seq_file *seq, void *v)
 
 static int pn_sock_seq_show(struct seq_file *seq, void *v)
 {
-	int len;
-
+	seq_setwidth(seq, 127);
 	if (v == SEQ_START_TOKEN)
-		seq_printf(seq, "%s%n", "pt  loc  rem rs st tx_queue rx_queue "
-			"  uid inode ref pointer drops", &len);
+		seq_puts(seq, "pt  loc  rem rs st tx_queue rx_queue "
+			"  uid inode ref pointer drops");
 	else {
 		struct sock *sk = v;
 		struct pn_sock *pn = pn_sk(sk);
 
 		seq_printf(seq, "%2d %04X:%04X:%02X %02X %08X:%08X %5d %lu "
-			"%d %pK %d%n",
+			"%d %pK %d",
 			sk->sk_protocol, pn->sobject, pn->dobject,
 			pn->resource, sk->sk_state,
 			sk_wmem_alloc_get(sk), sk_rmem_alloc_get(sk),
 			from_kuid_munged(seq_user_ns(seq), sock_i_uid(sk)),
 			sock_i_ino(sk),
 			atomic_read(&sk->sk_refcnt), sk,
-			atomic_read(&sk->sk_drops), &len);
+			atomic_read(&sk->sk_drops));
 	}
-	seq_printf(seq, "%*s\n", 127 - len, "");
+	seq_pad(seq, '\n');
 	return 0;
 }
 
@@ -785,20 +784,19 @@ static void pn_res_seq_stop(struct seq_file *seq, void *v)
 
 static int pn_res_seq_show(struct seq_file *seq, void *v)
 {
-	int len;
-
+	seq_setwidth(seq, 63);
 	if (v == SEQ_START_TOKEN)
-		seq_printf(seq, "%s%n", "rs   uid inode", &len);
+		seq_puts(seq, "rs   uid inode");
 	else {
 		struct sock **psk = v;
 		struct sock *sk = *psk;
 
-		seq_printf(seq, "%02X %5u %lu%n",
+		seq_printf(seq, "%02X %5u %lu",
 			   (int) (psk - pnres.sk),
 			   from_kuid_munged(seq_user_ns(seq), sock_i_uid(sk)),
-			   sock_i_ino(sk), &len);
+			   sock_i_ino(sk));
 	}
-	seq_printf(seq, "%*s\n", 63 - len, "");
+	seq_pad(seq, '\n');
 	return 0;
 }
 
diff --git a/net/sched/sch_fq.c b/net/sched/sch_fq.c
index fdc041c57853..95d843961907 100644
--- a/net/sched/sch_fq.c
+++ b/net/sched/sch_fq.c
@@ -88,7 +88,7 @@ struct fq_sched_data {
 	struct fq_flow	internal;	/* for non classified or high prio packets */
 	u32		quantum;
 	u32		initial_quantum;
-	u32		flow_default_rate;/* rate per flow : bytes per second */
+	u32		flow_refill_delay;
 	u32		flow_max_rate;	/* optional max rate per flow */
 	u32		flow_plimit;	/* max packets per flow */
 	struct rb_root	*fq_root;
@@ -115,6 +115,7 @@ static struct fq_flow detached, throttled;
 static void fq_flow_set_detached(struct fq_flow *f)
 {
 	f->next = &detached;
+	f->age = jiffies;
 }
 
 static bool fq_flow_is_detached(const struct fq_flow *f)
@@ -209,21 +210,15 @@ static void fq_gc(struct fq_sched_data *q,
 	}
 }
 
-static const u8 prio2band[TC_PRIO_MAX + 1] = {
-	1, 2, 2, 2, 1, 2, 0, 0 , 1, 1, 1, 1, 1, 1, 1, 1
-};
-
 static struct fq_flow *fq_classify(struct sk_buff *skb, struct fq_sched_data *q)
 {
 	struct rb_node **p, *parent;
 	struct sock *sk = skb->sk;
 	struct rb_root *root;
 	struct fq_flow *f;
-	int band;
 
 	/* warning: no starvation prevention... */
-	band = prio2band[skb->priority & TC_PRIO_MAX];
-	if (unlikely(band == 0))
+	if (unlikely((skb->priority & TC_PRIO_MAX) == TC_PRIO_CONTROL))
 		return &q->internal;
 
 	if (unlikely(!sk)) {
@@ -373,17 +368,20 @@ static int fq_enqueue(struct sk_buff *skb, struct Qdisc *sch)
 	}
 
 	f->qlen++;
-	flow_queue_add(f, skb);
 	if (skb_is_retransmit(skb))
 		q->stat_tcp_retrans++;
 	sch->qstats.backlog += qdisc_pkt_len(skb);
 	if (fq_flow_is_detached(f)) {
 		fq_flow_add_tail(&q->new_flows, f);
-		if (q->quantum > f->credit)
-			f->credit = q->quantum;
+		if (time_after(jiffies, f->age + q->flow_refill_delay))
+			f->credit = max_t(u32, f->credit, q->quantum);
 		q->inactive_flows--;
 		qdisc_unthrottled(sch);
 	}
+
+	/* Note: this overwrites f->age */
+	flow_queue_add(f, skb);
+
 	if (unlikely(f == &q->internal)) {
 		q->stat_internal_packets++;
 		qdisc_unthrottled(sch);
@@ -461,7 +459,6 @@ begin:
 			fq_flow_add_tail(&q->old_flows, f);
 		} else {
 			fq_flow_set_detached(f);
-			f->age = jiffies;
 			q->inactive_flows++;
 		}
 		goto begin;
@@ -615,6 +612,7 @@ static const struct nla_policy fq_policy[TCA_FQ_MAX + 1] = {
 	[TCA_FQ_FLOW_DEFAULT_RATE]	= { .type = NLA_U32 },
 	[TCA_FQ_FLOW_MAX_RATE]		= { .type = NLA_U32 },
 	[TCA_FQ_BUCKETS_LOG]		= { .type = NLA_U32 },
+	[TCA_FQ_FLOW_REFILL_DELAY]	= { .type = NLA_U32 },
 };
 
 static int fq_change(struct Qdisc *sch, struct nlattr *opt)
@@ -656,7 +654,8 @@ static int fq_change(struct Qdisc *sch, struct nlattr *opt)
 		q->initial_quantum = nla_get_u32(tb[TCA_FQ_INITIAL_QUANTUM]);
 
 	if (tb[TCA_FQ_FLOW_DEFAULT_RATE])
-		q->flow_default_rate = nla_get_u32(tb[TCA_FQ_FLOW_DEFAULT_RATE]);
+		pr_warn_ratelimited("sch_fq: defrate %u ignored.\n",
+				    nla_get_u32(tb[TCA_FQ_FLOW_DEFAULT_RATE]));
 
 	if (tb[TCA_FQ_FLOW_MAX_RATE])
 		q->flow_max_rate = nla_get_u32(tb[TCA_FQ_FLOW_MAX_RATE]);
@@ -670,6 +669,12 @@ static int fq_change(struct Qdisc *sch, struct nlattr *opt)
 			err = -EINVAL;
 	}
 
+	if (tb[TCA_FQ_FLOW_REFILL_DELAY]) {
+		u32 usecs_delay = nla_get_u32(tb[TCA_FQ_FLOW_REFILL_DELAY]) ;
+
+		q->flow_refill_delay = usecs_to_jiffies(usecs_delay);
+	}
+
 	if (!err)
 		err = fq_resize(q, fq_log);
 
@@ -705,7 +710,7 @@ static int fq_init(struct Qdisc *sch, struct nlattr *opt)
 	q->flow_plimit		= 100;
 	q->quantum		= 2 * psched_mtu(qdisc_dev(sch));
 	q->initial_quantum	= 10 * psched_mtu(qdisc_dev(sch));
-	q->flow_default_rate	= 0;
+	q->flow_refill_delay	= msecs_to_jiffies(40);
 	q->flow_max_rate	= ~0U;
 	q->rate_enable		= 1;
 	q->new_flows.first	= NULL;
@@ -732,15 +737,16 @@ static int fq_dump(struct Qdisc *sch, struct sk_buff *skb)
 	if (opts == NULL)
 		goto nla_put_failure;
 
-	/* TCA_FQ_FLOW_DEFAULT_RATE is not used anymore,
-	 * do not bother giving its value
-	 */
+	/* TCA_FQ_FLOW_DEFAULT_RATE is not used anymore */
+
 	if (nla_put_u32(skb, TCA_FQ_PLIMIT, sch->limit) ||
 	    nla_put_u32(skb, TCA_FQ_FLOW_PLIMIT, q->flow_plimit) ||
 	    nla_put_u32(skb, TCA_FQ_QUANTUM, q->quantum) ||
 	    nla_put_u32(skb, TCA_FQ_INITIAL_QUANTUM, q->initial_quantum) ||
 	    nla_put_u32(skb, TCA_FQ_RATE_ENABLE, q->rate_enable) ||
 	    nla_put_u32(skb, TCA_FQ_FLOW_MAX_RATE, q->flow_max_rate) ||
+	    nla_put_u32(skb, TCA_FQ_FLOW_REFILL_DELAY,
+			jiffies_to_usecs(q->flow_refill_delay)) ||
 	    nla_put_u32(skb, TCA_FQ_BUCKETS_LOG, q->fq_trees_log))
 		goto nla_put_failure;
 
diff --git a/net/sctp/associola.c b/net/sctp/associola.c
index c9b91cb1cb0d..68a27f9796d2 100644
--- a/net/sctp/associola.c
+++ b/net/sctp/associola.c
@@ -907,8 +907,8 @@ void sctp_assoc_control_transport(struct sctp_association *asoc,
 		if (!first || t->last_time_heard > first->last_time_heard) {
 			second = first;
 			first = t;
-		}
-		if (!second || t->last_time_heard > second->last_time_heard)
+		} else if (!second ||
+			   t->last_time_heard > second->last_time_heard)
 			second = t;
 	}
 
@@ -929,6 +929,8 @@ void sctp_assoc_control_transport(struct sctp_association *asoc,
 		first = asoc->peer.primary_path;
 	}
 
+	if (!second)
+		second = first;
 	/* If we failed to find a usable transport, just camp on the
 	 * primary, even if it is inactive.
 	 */
diff --git a/net/sctp/objcnt.c b/net/sctp/objcnt.c
index 5ea573b37648..647396baa56f 100644
--- a/net/sctp/objcnt.c
+++ b/net/sctp/objcnt.c
@@ -79,12 +79,13 @@ static sctp_dbg_objcnt_entry_t sctp_dbg_objcnt[] = {
  */
 static int sctp_objcnt_seq_show(struct seq_file *seq, void *v)
 {
-	int i, len;
+	int i;
 
 	i = (int)*(loff_t *)v;
-	seq_printf(seq, "%s: %d%n", sctp_dbg_objcnt[i].label,
-				atomic_read(sctp_dbg_objcnt[i].counter), &len);
-	seq_printf(seq, "%*s\n", 127 - len, "");
+	seq_setwidth(seq, 127);
+	seq_printf(seq, "%s: %d", sctp_dbg_objcnt[i].label,
+				atomic_read(sctp_dbg_objcnt[i].counter));
+	seq_pad(seq, '\n');
 	return 0;
 }
 
diff --git a/net/sunrpc/auth_gss/gss_krb5_unseal.c b/net/sunrpc/auth_gss/gss_krb5_unseal.c
index 6cd930f3678f..6c981ddc19f8 100644
--- a/net/sunrpc/auth_gss/gss_krb5_unseal.c
+++ b/net/sunrpc/auth_gss/gss_krb5_unseal.c
@@ -150,7 +150,6 @@ gss_verify_mic_v2(struct krb5_ctx *ctx,
 	struct xdr_netobj cksumobj = {.len = sizeof(cksumdata),
 				      .data = cksumdata};
 	s32 now;
-	u64 seqnum;
 	u8 *ptr = read_token->data;
 	u8 *cksumkey;
 	u8 flags;
@@ -197,9 +196,10 @@ gss_verify_mic_v2(struct krb5_ctx *ctx,
 	if (now > ctx->endtime)
 		return GSS_S_CONTEXT_EXPIRED;
 
-	/* do sequencing checks */
-
-	seqnum = be64_to_cpup((__be64 *)ptr + 8);
+	/*
+	 * NOTE: the sequence number at ptr + 8 is skipped, rpcsec_gss
+	 * doesn't want it checked; see page 6 of rfc 2203.
+	 */
 
 	return GSS_S_COMPLETE;
 }
diff --git a/net/sunrpc/auth_gss/gss_krb5_wrap.c b/net/sunrpc/auth_gss/gss_krb5_wrap.c
index 1da52d1406fc..42560e55d978 100644
--- a/net/sunrpc/auth_gss/gss_krb5_wrap.c
+++ b/net/sunrpc/auth_gss/gss_krb5_wrap.c
@@ -489,7 +489,6 @@ static u32
 gss_unwrap_kerberos_v2(struct krb5_ctx *kctx, int offset, struct xdr_buf *buf)
 {
 	s32		now;
-	u64		seqnum;
 	u8		*ptr;
 	u8		flags = 0x00;
 	u16		ec, rrc;
@@ -525,7 +524,10 @@ gss_unwrap_kerberos_v2(struct krb5_ctx *kctx, int offset, struct xdr_buf *buf)
 	ec = be16_to_cpup((__be16 *)(ptr + 4));
 	rrc = be16_to_cpup((__be16 *)(ptr + 6));
 
-	seqnum = be64_to_cpup((__be64 *)(ptr + 8));
+	/*
+	 * NOTE: the sequence number at ptr + 8 is skipped, rpcsec_gss
+	 * doesn't want it checked; see page 6 of rfc 2203.
+	 */
 
 	if (rrc != 0)
 		rotate_left(offset + 16, buf, rrc);
@@ -574,8 +576,8 @@ gss_unwrap_kerberos_v2(struct krb5_ctx *kctx, int offset, struct xdr_buf *buf)
 	buf->head[0].iov_len -= GSS_KRB5_TOK_HDR_LEN + headskip;
 	buf->len -= GSS_KRB5_TOK_HDR_LEN + headskip;
 
-	/* Trim off the checksum blob */
-	xdr_buf_trim(buf, GSS_KRB5_TOK_HDR_LEN + tailskip);
+	/* Trim off the trailing "extra count" and checksum blob */
+	xdr_buf_trim(buf, ec + GSS_KRB5_TOK_HDR_LEN + tailskip);
 	return GSS_S_COMPLETE;
 }
 
diff --git a/net/sunrpc/auth_gss/gss_rpc_upcall.c b/net/sunrpc/auth_gss/gss_rpc_upcall.c
index f1eb0d16666c..458f85e9b0ba 100644
--- a/net/sunrpc/auth_gss/gss_rpc_upcall.c
+++ b/net/sunrpc/auth_gss/gss_rpc_upcall.c
@@ -298,7 +298,8 @@ int gssp_accept_sec_context_upcall(struct net *net,
 	if (res.context_handle) {
 		data->out_handle = rctxh.exported_context_token;
 		data->mech_oid.len = rctxh.mech.len;
-		memcpy(data->mech_oid.data, rctxh.mech.data,
+		if (rctxh.mech.data)
+			memcpy(data->mech_oid.data, rctxh.mech.data,
 						data->mech_oid.len);
 		client_name = rctxh.src_name.display_name;
 	}
diff --git a/net/sunrpc/auth_gss/gss_rpc_xdr.c b/net/sunrpc/auth_gss/gss_rpc_xdr.c
index f0f78c5f1c7d..1ec19f6f0c2b 100644
--- a/net/sunrpc/auth_gss/gss_rpc_xdr.c
+++ b/net/sunrpc/auth_gss/gss_rpc_xdr.c
@@ -559,6 +559,8 @@ static int gssx_enc_cred(struct xdr_stream *xdr,
 
 	/* cred->elements */
 	err = dummy_enc_credel_array(xdr, &cred->elements);
+	if (err)
+		return err;
 
 	/* cred->cred_handle_reference */
 	err = gssx_enc_buffer(xdr, &cred->cred_handle_reference);
@@ -740,22 +742,20 @@ void gssx_enc_accept_sec_context(struct rpc_rqst *req,
 		goto done;
 
 	/* arg->context_handle */
-	if (arg->context_handle) {
+	if (arg->context_handle)
 		err = gssx_enc_ctx(xdr, arg->context_handle);
-		if (err)
-			goto done;
-	} else {
+	else
 		err = gssx_enc_bool(xdr, 0);
-	}
+	if (err)
+		goto done;
 
 	/* arg->cred_handle */
-	if (arg->cred_handle) {
+	if (arg->cred_handle)
 		err = gssx_enc_cred(xdr, arg->cred_handle);
-		if (err)
-			goto done;
-	} else {
+	else
 		err = gssx_enc_bool(xdr, 0);
-	}
+	if (err)
+		goto done;
 
 	/* arg->input_token */
 	err = gssx_enc_in_token(xdr, &arg->input_token);
@@ -763,13 +763,12 @@ void gssx_enc_accept_sec_context(struct rpc_rqst *req,
 		goto done;
 
 	/* arg->input_cb */
-	if (arg->input_cb) {
+	if (arg->input_cb)
 		err = gssx_enc_cb(xdr, arg->input_cb);
-		if (err)
-			goto done;
-	} else {
+	else
 		err = gssx_enc_bool(xdr, 0);
-	}
+	if (err)
+		goto done;
 
 	err = gssx_enc_bool(xdr, arg->ret_deleg_cred);
 	if (err)
diff --git a/net/sunrpc/auth_gss/svcauth_gss.c b/net/sunrpc/auth_gss/svcauth_gss.c
index 09fb638bcaa4..008cdade5aae 100644
--- a/net/sunrpc/auth_gss/svcauth_gss.c
+++ b/net/sunrpc/auth_gss/svcauth_gss.c
@@ -1167,8 +1167,8 @@ static int gss_proxy_save_rsc(struct cache_detail *cd,
 	if (!ud->found_creds) {
 		/* userspace seem buggy, we should always get at least a
 		 * mapping to nobody */
-		dprintk("RPC:       No creds found, marking Negative!\n");
-		set_bit(CACHE_NEGATIVE, &rsci.h.flags);
+		dprintk("RPC:       No creds found!\n");
+		goto out;
 	} else {
 
 		/* steal creds */
diff --git a/net/sunrpc/clnt.c b/net/sunrpc/clnt.c
index dab09dac8fc7..f09b7db2c492 100644
--- a/net/sunrpc/clnt.c
+++ b/net/sunrpc/clnt.c
@@ -750,14 +750,16 @@ EXPORT_SYMBOL_GPL(rpc_shutdown_client);
 /*
  * Free an RPC client
  */
-static void
+static struct rpc_clnt *
 rpc_free_client(struct rpc_clnt *clnt)
 {
+	struct rpc_clnt *parent = NULL;
+
 	dprintk_rcu("RPC:       destroying %s client for %s\n",
 			clnt->cl_program->name,
 			rcu_dereference(clnt->cl_xprt)->servername);
 	if (clnt->cl_parent != clnt)
-		rpc_release_client(clnt->cl_parent);
+		parent = clnt->cl_parent;
 	rpc_clnt_remove_pipedir(clnt);
 	rpc_unregister_client(clnt);
 	rpc_free_iostats(clnt->cl_metrics);
@@ -766,18 +768,17 @@ rpc_free_client(struct rpc_clnt *clnt)
 	rpciod_down();
 	rpc_free_clid(clnt);
 	kfree(clnt);
+	return parent;
 }
 
 /*
  * Free an RPC client
  */
-static void
+static struct rpc_clnt * 
 rpc_free_auth(struct rpc_clnt *clnt)
 {
-	if (clnt->cl_auth == NULL) {
-		rpc_free_client(clnt);
-		return;
-	}
+	if (clnt->cl_auth == NULL)
+		return rpc_free_client(clnt);
 
 	/*
 	 * Note: RPCSEC_GSS may need to send NULL RPC calls in order to
@@ -788,7 +789,8 @@ rpc_free_auth(struct rpc_clnt *clnt)
 	rpcauth_release(clnt->cl_auth);
 	clnt->cl_auth = NULL;
 	if (atomic_dec_and_test(&clnt->cl_count))
-		rpc_free_client(clnt);
+		return rpc_free_client(clnt);
+	return NULL;
 }
 
 /*
@@ -799,10 +801,13 @@ rpc_release_client(struct rpc_clnt *clnt)
 {
 	dprintk("RPC:       rpc_release_client(%p)\n", clnt);
 
-	if (list_empty(&clnt->cl_tasks))
-		wake_up(&destroy_wait);
-	if (atomic_dec_and_test(&clnt->cl_count))
-		rpc_free_auth(clnt);
+	do {
+		if (list_empty(&clnt->cl_tasks))
+			wake_up(&destroy_wait);
+		if (!atomic_dec_and_test(&clnt->cl_count))
+			break;
+		clnt = rpc_free_auth(clnt);
+	} while (clnt != NULL);
 }
 EXPORT_SYMBOL_GPL(rpc_release_client);
 
diff --git a/net/sunrpc/svc.c b/net/sunrpc/svc.c
index b974571126fe..e7fbe368b4a3 100644
--- a/net/sunrpc/svc.c
+++ b/net/sunrpc/svc.c
@@ -1104,8 +1104,6 @@ svc_process_common(struct svc_rqst *rqstp, struct kvec *argv, struct kvec *resv)
 	rqstp->rq_vers = vers = svc_getnl(argv);	/* version number */
 	rqstp->rq_proc = proc = svc_getnl(argv);	/* procedure number */
 
-	progp = serv->sv_program;
-
 	for (progp = serv->sv_program; progp; progp = progp->pg_next)
 		if (prog == progp->pg_prog)
 			break;
diff --git a/net/sunrpc/xprtsock.c b/net/sunrpc/xprtsock.c
index 17c88928b7db..dd9d295813cf 100644
--- a/net/sunrpc/xprtsock.c
+++ b/net/sunrpc/xprtsock.c
@@ -393,8 +393,10 @@ static int xs_send_kvec(struct socket *sock, struct sockaddr *addr, int addrlen,
 	return kernel_sendmsg(sock, &msg, NULL, 0, 0);
 }
 
-static int xs_send_pagedata(struct socket *sock, struct xdr_buf *xdr, unsigned int base, int more)
+static int xs_send_pagedata(struct socket *sock, struct xdr_buf *xdr, unsigned int base, int more, bool zerocopy)
 {
+	ssize_t (*do_sendpage)(struct socket *sock, struct page *page,
+			int offset, size_t size, int flags);
 	struct page **ppage;
 	unsigned int remainder;
 	int err, sent = 0;
@@ -403,6 +405,9 @@ static int xs_send_pagedata(struct socket *sock, struct xdr_buf *xdr, unsigned i
 	base += xdr->page_base;
 	ppage = xdr->pages + (base >> PAGE_SHIFT);
 	base &= ~PAGE_MASK;
+	do_sendpage = sock->ops->sendpage;
+	if (!zerocopy)
+		do_sendpage = sock_no_sendpage;
 	for(;;) {
 		unsigned int len = min_t(unsigned int, PAGE_SIZE - base, remainder);
 		int flags = XS_SENDMSG_FLAGS;
@@ -410,7 +415,7 @@ static int xs_send_pagedata(struct socket *sock, struct xdr_buf *xdr, unsigned i
 		remainder -= len;
 		if (remainder != 0 || more)
 			flags |= MSG_MORE;
-		err = sock->ops->sendpage(sock, *ppage, base, len, flags);
+		err = do_sendpage(sock, *ppage, base, len, flags);
 		if (remainder == 0 || err != len)
 			break;
 		sent += err;
@@ -431,9 +436,10 @@ static int xs_send_pagedata(struct socket *sock, struct xdr_buf *xdr, unsigned i
  * @addrlen: UDP only -- length of destination address
  * @xdr: buffer containing this request
  * @base: starting position in the buffer
+ * @zerocopy: true if it is safe to use sendpage()
  *
  */
-static int xs_sendpages(struct socket *sock, struct sockaddr *addr, int addrlen, struct xdr_buf *xdr, unsigned int base)
+static int xs_sendpages(struct socket *sock, struct sockaddr *addr, int addrlen, struct xdr_buf *xdr, unsigned int base, bool zerocopy)
 {
 	unsigned int remainder = xdr->len - base;
 	int err, sent = 0;
@@ -461,7 +467,7 @@ static int xs_sendpages(struct socket *sock, struct sockaddr *addr, int addrlen,
 	if (base < xdr->page_len) {
 		unsigned int len = xdr->page_len - base;
 		remainder -= len;
-		err = xs_send_pagedata(sock, xdr, base, remainder != 0);
+		err = xs_send_pagedata(sock, xdr, base, remainder != 0, zerocopy);
 		if (remainder == 0 || err != len)
 			goto out;
 		sent += err;
@@ -564,7 +570,7 @@ static int xs_local_send_request(struct rpc_task *task)
 			req->rq_svec->iov_base, req->rq_svec->iov_len);
 
 	status = xs_sendpages(transport->sock, NULL, 0,
-						xdr, req->rq_bytes_sent);
+						xdr, req->rq_bytes_sent, true);
 	dprintk("RPC:       %s(%u) = %d\n",
 			__func__, xdr->len - req->rq_bytes_sent, status);
 	if (likely(status >= 0)) {
@@ -620,7 +626,7 @@ static int xs_udp_send_request(struct rpc_task *task)
 	status = xs_sendpages(transport->sock,
 			      xs_addr(xprt),
 			      xprt->addrlen, xdr,
-			      req->rq_bytes_sent);
+			      req->rq_bytes_sent, true);
 
 	dprintk("RPC:       xs_udp_send_request(%u) = %d\n",
 			xdr->len - req->rq_bytes_sent, status);
@@ -693,6 +699,7 @@ static int xs_tcp_send_request(struct rpc_task *task)
 	struct rpc_xprt *xprt = req->rq_xprt;
 	struct sock_xprt *transport = container_of(xprt, struct sock_xprt, xprt);
 	struct xdr_buf *xdr = &req->rq_snd_buf;
+	bool zerocopy = true;
 	int status;
 
 	xs_encode_stream_record_marker(&req->rq_snd_buf);
@@ -700,13 +707,20 @@ static int xs_tcp_send_request(struct rpc_task *task)
 	xs_pktdump("packet data:",
 				req->rq_svec->iov_base,
 				req->rq_svec->iov_len);
+	/* Don't use zero copy if this is a resend. If the RPC call
+	 * completes while the socket holds a reference to the pages,
+	 * then we may end up resending corrupted data.
+	 */
+	if (task->tk_flags & RPC_TASK_SENT)
+		zerocopy = false;
 
 	/* Continue transmitting the packet/record. We must be careful
 	 * to cope with writespace callbacks arriving _after_ we have
 	 * called sendmsg(). */
 	while (1) {
 		status = xs_sendpages(transport->sock,
-					NULL, 0, xdr, req->rq_bytes_sent);
+					NULL, 0, xdr, req->rq_bytes_sent,
+					zerocopy);
 
 		dprintk("RPC:       xs_tcp_send_request(%u) = %d\n",
 				xdr->len - req->rq_bytes_sent, status);
diff --git a/net/tipc/link.c b/net/tipc/link.c
index cf465d66ccde..69cd9bf3f561 100644
--- a/net/tipc/link.c
+++ b/net/tipc/link.c
@@ -2358,7 +2358,8 @@ int tipc_link_recv_fragment(struct sk_buff **head, struct sk_buff **tail,
 		*head = frag;
 		skb_frag_list_init(*head);
 		return 0;
-	} else if (skb_try_coalesce(*head, frag, &headstolen, &delta)) {
+	} else if (*head &&
+		   skb_try_coalesce(*head, frag, &headstolen, &delta)) {
 		kfree_skb_partial(frag, headstolen);
 	} else {
 		if (!*head)
diff --git a/net/tipc/netlink.c b/net/tipc/netlink.c
index 8bcd4985d0fb..9f72a6376362 100644
--- a/net/tipc/netlink.c
+++ b/net/tipc/netlink.c
@@ -76,9 +76,11 @@ static struct genl_family tipc_genl_family = {
 	.maxattr	= 0,
 };
 
-static struct genl_ops tipc_genl_ops = {
-	.cmd		= TIPC_GENL_CMD,
-	.doit		= handle_cmd,
+static struct genl_ops tipc_genl_ops[] = {
+	{
+		.cmd		= TIPC_GENL_CMD,
+		.doit		= handle_cmd,
+	},
 };
 
 static int tipc_genl_family_registered;
@@ -87,8 +89,7 @@ int tipc_netlink_start(void)
 {
 	int res;
 
-	res = genl_register_family_with_ops(&tipc_genl_family,
-		&tipc_genl_ops, 1);
+	res = genl_register_family_with_ops(&tipc_genl_family, tipc_genl_ops);
 	if (res) {
 		pr_err("Failed to register netlink interface\n");
 		return res;
diff --git a/net/vmw_vsock/Kconfig b/net/vmw_vsock/Kconfig
index b5fa7e40cdcb..14810abedc2e 100644
--- a/net/vmw_vsock/Kconfig
+++ b/net/vmw_vsock/Kconfig
@@ -6,7 +6,7 @@ config VSOCKETS
 	tristate "Virtual Socket protocol"
 	help
 	  Virtual Socket Protocol is a socket protocol similar to TCP/IP
-	  allowing comunication between Virtual Machines and hypervisor
+	  allowing communication between Virtual Machines and hypervisor
 	  or host.
 
 	  You should also select one or more hypervisor-specific transports
diff --git a/net/wimax/op-msg.c b/net/wimax/op-msg.c
index 0694d62e4dbc..c278b3356f75 100644
--- a/net/wimax/op-msg.c
+++ b/net/wimax/op-msg.c
@@ -279,7 +279,7 @@ int wimax_msg_send(struct wimax_dev *wimax_dev, struct sk_buff *skb)
 
 	d_printf(1, dev, "CTX: wimax msg, %zu bytes\n", size);
 	d_dump(2, dev, msg, size);
-	genlmsg_multicast(skb, 0, wimax_gnl_mcg.id, GFP_KERNEL);
+	genlmsg_multicast(&wimax_gnl_family, skb, 0, 0, GFP_KERNEL);
 	d_printf(1, dev, "CTX: genl multicast done\n");
 	return 0;
 }
@@ -321,17 +321,6 @@ int wimax_msg(struct wimax_dev *wimax_dev, const char *pipe_name,
 }
 EXPORT_SYMBOL_GPL(wimax_msg);
 
-
-static const struct nla_policy wimax_gnl_msg_policy[WIMAX_GNL_ATTR_MAX + 1] = {
-	[WIMAX_GNL_MSG_IFIDX] = {
-		.type = NLA_U32,
-	},
-	[WIMAX_GNL_MSG_DATA] = {
-		.type = NLA_UNSPEC,	/* libnl doesn't grok BINARY yet */
-	},
-};
-
-
 /*
  * Relays a message from user space to the driver
  *
@@ -340,7 +329,6 @@ static const struct nla_policy wimax_gnl_msg_policy[WIMAX_GNL_ATTR_MAX + 1] = {
  *
  * This call will block while handling/relaying the message.
  */
-static
 int wimax_gnl_doit_msg_from_user(struct sk_buff *skb, struct genl_info *info)
 {
 	int result, ifindex;
@@ -418,16 +406,3 @@ error_no_wimax_dev:
 	return result;
 }
 
-
-/*
- * Generic Netlink glue
- */
-
-struct genl_ops wimax_gnl_msg_from_user = {
-	.cmd = WIMAX_GNL_OP_MSG_FROM_USER,
-	.flags = GENL_ADMIN_PERM,
-	.policy = wimax_gnl_msg_policy,
-	.doit = wimax_gnl_doit_msg_from_user,
-	.dumpit = NULL,
-};
-
diff --git a/net/wimax/op-reset.c b/net/wimax/op-reset.c
index 7ceffe39d70e..eb4580784d9d 100644
--- a/net/wimax/op-reset.c
+++ b/net/wimax/op-reset.c
@@ -92,13 +92,6 @@ int wimax_reset(struct wimax_dev *wimax_dev)
 EXPORT_SYMBOL(wimax_reset);
 
 
-static const struct nla_policy wimax_gnl_reset_policy[WIMAX_GNL_ATTR_MAX + 1] = {
-	[WIMAX_GNL_RESET_IFIDX] = {
-		.type = NLA_U32,
-	},
-};
-
-
 /*
  * Exporting to user space over generic netlink
  *
@@ -106,7 +99,6 @@ static const struct nla_policy wimax_gnl_reset_policy[WIMAX_GNL_ATTR_MAX + 1] =
  *
  * No attributes.
  */
-static
 int wimax_gnl_doit_reset(struct sk_buff *skb, struct genl_info *info)
 {
 	int result, ifindex;
@@ -130,12 +122,3 @@ error_no_wimax_dev:
 	d_fnend(3, NULL, "(skb %p info %p) = %d\n", skb, info, result);
 	return result;
 }
-
-
-struct genl_ops wimax_gnl_reset = {
-	.cmd = WIMAX_GNL_OP_RESET,
-	.flags = GENL_ADMIN_PERM,
-	.policy = wimax_gnl_reset_policy,
-	.doit = wimax_gnl_doit_reset,
-	.dumpit = NULL,
-};
diff --git a/net/wimax/op-rfkill.c b/net/wimax/op-rfkill.c
index 7ab60babdd22..403078d670a9 100644
--- a/net/wimax/op-rfkill.c
+++ b/net/wimax/op-rfkill.c
@@ -411,17 +411,6 @@ void wimax_rfkill_rm(struct wimax_dev *wimax_dev)
  * just query).
  */
 
-static const struct nla_policy wimax_gnl_rfkill_policy[WIMAX_GNL_ATTR_MAX + 1] = {
-	[WIMAX_GNL_RFKILL_IFIDX] = {
-		.type = NLA_U32,
-	},
-	[WIMAX_GNL_RFKILL_STATE] = {
-		.type = NLA_U32		/* enum wimax_rf_state */
-	},
-};
-
-
-static
 int wimax_gnl_doit_rfkill(struct sk_buff *skb, struct genl_info *info)
 {
 	int result, ifindex;
@@ -457,13 +446,3 @@ error_no_wimax_dev:
 	d_fnend(3, NULL, "(skb %p info %p) = %d\n", skb, info, result);
 	return result;
 }
-
-
-struct genl_ops wimax_gnl_rfkill = {
-	.cmd = WIMAX_GNL_OP_RFKILL,
-	.flags = GENL_ADMIN_PERM,
-	.policy = wimax_gnl_rfkill_policy,
-	.doit = wimax_gnl_doit_rfkill,
-	.dumpit = NULL,
-};
-
diff --git a/net/wimax/op-state-get.c b/net/wimax/op-state-get.c
index aff8776e2d41..995c08c827b5 100644
--- a/net/wimax/op-state-get.c
+++ b/net/wimax/op-state-get.c
@@ -33,13 +33,6 @@
 #include "debug-levels.h"
 
 
-static const struct nla_policy wimax_gnl_state_get_policy[WIMAX_GNL_ATTR_MAX + 1] = {
-	[WIMAX_GNL_STGET_IFIDX] = {
-		.type = NLA_U32,
-	},
-};
-
-
 /*
  * Exporting to user space over generic netlink
  *
@@ -48,7 +41,6 @@ static const struct nla_policy wimax_gnl_state_get_policy[WIMAX_GNL_ATTR_MAX + 1
  *
  * No attributes.
  */
-static
 int wimax_gnl_doit_state_get(struct sk_buff *skb, struct genl_info *info)
 {
 	int result, ifindex;
@@ -72,12 +64,3 @@ error_no_wimax_dev:
 	d_fnend(3, NULL, "(skb %p info %p) = %d\n", skb, info, result);
 	return result;
 }
-
-
-struct genl_ops wimax_gnl_state_get = {
-	.cmd = WIMAX_GNL_OP_STATE_GET,
-	.flags = GENL_ADMIN_PERM,
-	.policy = wimax_gnl_state_get_policy,
-	.doit = wimax_gnl_doit_state_get,
-	.dumpit = NULL,
-};
diff --git a/net/wimax/stack.c b/net/wimax/stack.c
index a6470ac39498..ef2191b969a7 100644
--- a/net/wimax/stack.c
+++ b/net/wimax/stack.c
@@ -116,8 +116,9 @@ struct sk_buff *wimax_gnl_re_state_change_alloc(
 		dev_err(dev, "RE_STCH: can't create message\n");
 		goto error_new;
 	}
-	data = genlmsg_put(report_skb, 0, wimax_gnl_mcg.id, &wimax_gnl_family,
-			   0, WIMAX_GNL_RE_STATE_CHANGE);
+	/* FIXME: sending a group ID as the seq is wrong */
+	data = genlmsg_put(report_skb, 0, wimax_gnl_family.mcgrp_offset,
+			   &wimax_gnl_family, 0, WIMAX_GNL_RE_STATE_CHANGE);
 	if (data == NULL) {
 		dev_err(dev, "RE_STCH: can't put data into message\n");
 		goto error_put;
@@ -177,7 +178,7 @@ int wimax_gnl_re_state_change_send(
 		goto out;
 	}
 	genlmsg_end(report_skb, header);
-	genlmsg_multicast(report_skb, 0, wimax_gnl_mcg.id, GFP_KERNEL);
+	genlmsg_multicast(&wimax_gnl_family, report_skb, 0, 0, GFP_KERNEL);
 out:
 	d_fnend(3, dev, "(wimax_dev %p report_skb %p) = %d\n",
 		wimax_dev, report_skb, result);
@@ -402,22 +403,44 @@ void wimax_dev_init(struct wimax_dev *wimax_dev)
 }
 EXPORT_SYMBOL_GPL(wimax_dev_init);
 
-/*
- * This extern is declared here because it's easier to keep track --
- * both declarations are a list of the same
- */
-extern struct genl_ops
-	wimax_gnl_msg_from_user,
-	wimax_gnl_reset,
-	wimax_gnl_rfkill,
-	wimax_gnl_state_get;
+static const struct nla_policy wimax_gnl_policy[WIMAX_GNL_ATTR_MAX + 1] = {
+	[WIMAX_GNL_RESET_IFIDX] = { .type = NLA_U32, },
+	[WIMAX_GNL_RFKILL_IFIDX] = { .type = NLA_U32, },
+	[WIMAX_GNL_RFKILL_STATE] = {
+		.type = NLA_U32		/* enum wimax_rf_state */
+	},
+	[WIMAX_GNL_STGET_IFIDX] = { .type = NLA_U32, },
+	[WIMAX_GNL_MSG_IFIDX] = { .type = NLA_U32, },
+	[WIMAX_GNL_MSG_DATA] = {
+		.type = NLA_UNSPEC,	/* libnl doesn't grok BINARY yet */
+	},
+};
 
-static
-struct genl_ops *wimax_gnl_ops[] = {
-	&wimax_gnl_msg_from_user,
-	&wimax_gnl_reset,
-	&wimax_gnl_rfkill,
-	&wimax_gnl_state_get,
+static const struct genl_ops wimax_gnl_ops[] = {
+	{
+		.cmd = WIMAX_GNL_OP_MSG_FROM_USER,
+		.flags = GENL_ADMIN_PERM,
+		.policy = wimax_gnl_policy,
+		.doit = wimax_gnl_doit_msg_from_user,
+	},
+	{
+		.cmd = WIMAX_GNL_OP_RESET,
+		.flags = GENL_ADMIN_PERM,
+		.policy = wimax_gnl_policy,
+		.doit = wimax_gnl_doit_reset,
+	},
+	{
+		.cmd = WIMAX_GNL_OP_RFKILL,
+		.flags = GENL_ADMIN_PERM,
+		.policy = wimax_gnl_policy,
+		.doit = wimax_gnl_doit_rfkill,
+	},
+	{
+		.cmd = WIMAX_GNL_OP_STATE_GET,
+		.flags = GENL_ADMIN_PERM,
+		.policy = wimax_gnl_policy,
+		.doit = wimax_gnl_doit_state_get,
+	},
 };
 
 
@@ -557,8 +580,8 @@ struct genl_family wimax_gnl_family = {
 	.maxattr = WIMAX_GNL_ATTR_MAX,
 };
 
-struct genl_multicast_group wimax_gnl_mcg = {
-	.name = "msg",
+static const struct genl_multicast_group wimax_gnl_mcgrps[] = {
+	{ .name = "msg", },
 };
 
 
@@ -567,7 +590,7 @@ struct genl_multicast_group wimax_gnl_mcg = {
 static
 int __init wimax_subsys_init(void)
 {
-	int result, cnt;
+	int result;
 
 	d_fnstart(4, NULL, "()\n");
 	d_parse_params(D_LEVEL, D_LEVEL_SIZE, wimax_debug_params,
@@ -575,37 +598,18 @@ int __init wimax_subsys_init(void)
 
 	snprintf(wimax_gnl_family.name, sizeof(wimax_gnl_family.name),
 		 "WiMAX");
-	result = genl_register_family(&wimax_gnl_family);
+	result = genl_register_family_with_ops_groups(&wimax_gnl_family,
+						      wimax_gnl_ops,
+						      wimax_gnl_mcgrps);
 	if (unlikely(result < 0)) {
 		printk(KERN_ERR "cannot register generic netlink family: %d\n",
 		       result);
 		goto error_register_family;
 	}
 
-	for (cnt = 0; cnt < ARRAY_SIZE(wimax_gnl_ops); cnt++) {
-		result = genl_register_ops(&wimax_gnl_family,
-					   wimax_gnl_ops[cnt]);
-		d_printf(4, NULL, "registering generic netlink op code "
-			 "%u: %d\n", wimax_gnl_ops[cnt]->cmd, result);
-		if (unlikely(result < 0)) {
-			printk(KERN_ERR "cannot register generic netlink op "
-			       "code %u: %d\n",
-			       wimax_gnl_ops[cnt]->cmd, result);
-			goto error_register_ops;
-		}
-	}
-
-	result = genl_register_mc_group(&wimax_gnl_family, &wimax_gnl_mcg);
-	if (result < 0)
-		goto error_mc_group;
 	d_fnend(4, NULL, "() = 0\n");
 	return 0;
 
-error_mc_group:
-error_register_ops:
-	for (cnt--; cnt >= 0; cnt--)
-		genl_unregister_ops(&wimax_gnl_family,
-				    wimax_gnl_ops[cnt]);
 	genl_unregister_family(&wimax_gnl_family);
 error_register_family:
 	d_fnend(4, NULL, "() = %d\n", result);
@@ -619,12 +623,7 @@ module_init(wimax_subsys_init);
 static
 void __exit wimax_subsys_exit(void)
 {
-	int cnt;
 	wimax_id_table_release();
-	genl_unregister_mc_group(&wimax_gnl_family, &wimax_gnl_mcg);
-	for (cnt = ARRAY_SIZE(wimax_gnl_ops) - 1; cnt >= 0; cnt--)
-		genl_unregister_ops(&wimax_gnl_family,
-				    wimax_gnl_ops[cnt]);
 	genl_unregister_family(&wimax_gnl_family);
 }
 module_exit(wimax_subsys_exit);
diff --git a/net/wimax/wimax-internal.h b/net/wimax/wimax-internal.h
index 5dcd9c067bf0..b445b82020a8 100644
--- a/net/wimax/wimax-internal.h
+++ b/net/wimax/wimax-internal.h
@@ -84,8 +84,14 @@ void wimax_id_table_release(void);
 int wimax_rfkill_add(struct wimax_dev *);
 void wimax_rfkill_rm(struct wimax_dev *);
 
+/* generic netlink */
 extern struct genl_family wimax_gnl_family;
-extern struct genl_multicast_group wimax_gnl_mcg;
+
+/* ops */
+int wimax_gnl_doit_msg_from_user(struct sk_buff *skb, struct genl_info *info);
+int wimax_gnl_doit_reset(struct sk_buff *skb, struct genl_info *info);
+int wimax_gnl_doit_rfkill(struct sk_buff *skb, struct genl_info *info);
+int wimax_gnl_doit_state_get(struct sk_buff *skb, struct genl_info *info);
 
 #endif /* #ifdef __KERNEL__ */
 #endif /* #ifndef __WIMAX_INTERNAL_H__ */
diff --git a/net/wireless/nl80211.c b/net/wireless/nl80211.c
index a7f4e7902104..a1eb21073176 100644
--- a/net/wireless/nl80211.c
+++ b/net/wireless/nl80211.c
@@ -30,9 +30,9 @@ static int nl80211_crypto_settings(struct cfg80211_registered_device *rdev,
 				   struct cfg80211_crypto_settings *settings,
 				   int cipher_limit);
 
-static int nl80211_pre_doit(struct genl_ops *ops, struct sk_buff *skb,
+static int nl80211_pre_doit(const struct genl_ops *ops, struct sk_buff *skb,
 			    struct genl_info *info);
-static void nl80211_post_doit(struct genl_ops *ops, struct sk_buff *skb,
+static void nl80211_post_doit(const struct genl_ops *ops, struct sk_buff *skb,
 			      struct genl_info *info);
 
 /* the netlink family */
@@ -47,6 +47,25 @@ static struct genl_family nl80211_fam = {
 	.post_doit = nl80211_post_doit,
 };
 
+/* multicast groups */
+enum nl80211_multicast_groups {
+	NL80211_MCGRP_CONFIG,
+	NL80211_MCGRP_SCAN,
+	NL80211_MCGRP_REGULATORY,
+	NL80211_MCGRP_MLME,
+	NL80211_MCGRP_TESTMODE /* keep last - ifdef! */
+};
+
+static const struct genl_multicast_group nl80211_mcgrps[] = {
+	[NL80211_MCGRP_CONFIG] = { .name = "config", },
+	[NL80211_MCGRP_SCAN] = { .name = "scan", },
+	[NL80211_MCGRP_REGULATORY] = { .name = "regulatory", },
+	[NL80211_MCGRP_MLME] = { .name = "mlme", },
+#ifdef CONFIG_NL80211_TESTMODE
+	[NL80211_MCGRP_TESTMODE] = { .name = "testmode", }
+#endif
+};
+
 /* returns ERR_PTR values */
 static struct wireless_dev *
 __cfg80211_wdev_from_attrs(struct net *netns, struct nlattr **attrs)
@@ -6656,10 +6675,6 @@ static int nl80211_set_mcast_rate(struct sk_buff *skb, struct genl_info *info)
 
 
 #ifdef CONFIG_NL80211_TESTMODE
-static struct genl_multicast_group nl80211_testmode_mcgrp = {
-	.name = "testmode",
-};
-
 static int nl80211_testmode_do(struct sk_buff *skb, struct genl_info *info)
 {
 	struct cfg80211_registered_device *rdev = info->user_ptr[0];
@@ -6868,8 +6883,8 @@ void cfg80211_testmode_event(struct sk_buff *skb, gfp_t gfp)
 
 	nla_nest_end(skb, data);
 	genlmsg_end(skb, hdr);
-	genlmsg_multicast_netns(wiphy_net(&rdev->wiphy), skb, 0,
-				nl80211_testmode_mcgrp.id, gfp);
+	genlmsg_multicast_netns(&nl80211_fam, wiphy_net(&rdev->wiphy), skb, 0,
+				NL80211_MCGRP_TESTMODE, gfp);
 }
 EXPORT_SYMBOL(cfg80211_testmode_event);
 #endif
@@ -8851,7 +8866,7 @@ static int nl80211_crit_protocol_stop(struct sk_buff *skb,
 #define NL80211_FLAG_NEED_WDEV_UP	(NL80211_FLAG_NEED_WDEV |\
 					 NL80211_FLAG_CHECK_NETDEV_UP)
 
-static int nl80211_pre_doit(struct genl_ops *ops, struct sk_buff *skb,
+static int nl80211_pre_doit(const struct genl_ops *ops, struct sk_buff *skb,
 			    struct genl_info *info)
 {
 	struct cfg80211_registered_device *rdev;
@@ -8920,7 +8935,7 @@ static int nl80211_pre_doit(struct genl_ops *ops, struct sk_buff *skb,
 	return 0;
 }
 
-static void nl80211_post_doit(struct genl_ops *ops, struct sk_buff *skb,
+static void nl80211_post_doit(const struct genl_ops *ops, struct sk_buff *skb,
 			      struct genl_info *info)
 {
 	if (info->user_ptr[1]) {
@@ -8937,7 +8952,7 @@ static void nl80211_post_doit(struct genl_ops *ops, struct sk_buff *skb,
 		rtnl_unlock();
 }
 
-static struct genl_ops nl80211_ops[] = {
+static const struct genl_ops nl80211_ops[] = {
 	{
 		.cmd = NL80211_CMD_GET_WIPHY,
 		.doit = nl80211_get_wiphy,
@@ -9566,21 +9581,6 @@ static struct genl_ops nl80211_ops[] = {
 	},
 };
 
-static struct genl_multicast_group nl80211_mlme_mcgrp = {
-	.name = "mlme",
-};
-
-/* multicast groups */
-static struct genl_multicast_group nl80211_config_mcgrp = {
-	.name = "config",
-};
-static struct genl_multicast_group nl80211_scan_mcgrp = {
-	.name = "scan",
-};
-static struct genl_multicast_group nl80211_regulatory_mcgrp = {
-	.name = "regulatory",
-};
-
 /* notification functions */
 
 void nl80211_notify_dev_rename(struct cfg80211_registered_device *rdev)
@@ -9597,8 +9597,8 @@ void nl80211_notify_dev_rename(struct cfg80211_registered_device *rdev)
 		return;
 	}
 
-	genlmsg_multicast_netns(wiphy_net(&rdev->wiphy), msg, 0,
-				nl80211_config_mcgrp.id, GFP_KERNEL);
+	genlmsg_multicast_netns(&nl80211_fam, wiphy_net(&rdev->wiphy), msg, 0,
+				NL80211_MCGRP_CONFIG, GFP_KERNEL);
 }
 
 static int nl80211_add_scan_req(struct sk_buff *msg,
@@ -9707,8 +9707,8 @@ void nl80211_send_scan_start(struct cfg80211_registered_device *rdev,
 		return;
 	}
 
-	genlmsg_multicast_netns(wiphy_net(&rdev->wiphy), msg, 0,
-				nl80211_scan_mcgrp.id, GFP_KERNEL);
+	genlmsg_multicast_netns(&nl80211_fam, wiphy_net(&rdev->wiphy), msg, 0,
+				NL80211_MCGRP_SCAN, GFP_KERNEL);
 }
 
 void nl80211_send_scan_done(struct cfg80211_registered_device *rdev,
@@ -9726,8 +9726,8 @@ void nl80211_send_scan_done(struct cfg80211_registered_device *rdev,
 		return;
 	}
 
-	genlmsg_multicast_netns(wiphy_net(&rdev->wiphy), msg, 0,
-				nl80211_scan_mcgrp.id, GFP_KERNEL);
+	genlmsg_multicast_netns(&nl80211_fam, wiphy_net(&rdev->wiphy), msg, 0,
+				NL80211_MCGRP_SCAN, GFP_KERNEL);
 }
 
 void nl80211_send_scan_aborted(struct cfg80211_registered_device *rdev,
@@ -9745,8 +9745,8 @@ void nl80211_send_scan_aborted(struct cfg80211_registered_device *rdev,
 		return;
 	}
 
-	genlmsg_multicast_netns(wiphy_net(&rdev->wiphy), msg, 0,
-				nl80211_scan_mcgrp.id, GFP_KERNEL);
+	genlmsg_multicast_netns(&nl80211_fam, wiphy_net(&rdev->wiphy), msg, 0,
+				NL80211_MCGRP_SCAN, GFP_KERNEL);
 }
 
 void nl80211_send_sched_scan_results(struct cfg80211_registered_device *rdev,
@@ -9764,8 +9764,8 @@ void nl80211_send_sched_scan_results(struct cfg80211_registered_device *rdev,
 		return;
 	}
 
-	genlmsg_multicast_netns(wiphy_net(&rdev->wiphy), msg, 0,
-				nl80211_scan_mcgrp.id, GFP_KERNEL);
+	genlmsg_multicast_netns(&nl80211_fam, wiphy_net(&rdev->wiphy), msg, 0,
+				NL80211_MCGRP_SCAN, GFP_KERNEL);
 }
 
 void nl80211_send_sched_scan(struct cfg80211_registered_device *rdev,
@@ -9782,8 +9782,8 @@ void nl80211_send_sched_scan(struct cfg80211_registered_device *rdev,
 		return;
 	}
 
-	genlmsg_multicast_netns(wiphy_net(&rdev->wiphy), msg, 0,
-				nl80211_scan_mcgrp.id, GFP_KERNEL);
+	genlmsg_multicast_netns(&nl80211_fam, wiphy_net(&rdev->wiphy), msg, 0,
+				NL80211_MCGRP_SCAN, GFP_KERNEL);
 }
 
 /*
@@ -9837,8 +9837,8 @@ void nl80211_send_reg_change_event(struct regulatory_request *request)
 	genlmsg_end(msg, hdr);
 
 	rcu_read_lock();
-	genlmsg_multicast_allns(msg, 0, nl80211_regulatory_mcgrp.id,
-				GFP_ATOMIC);
+	genlmsg_multicast_allns(&nl80211_fam, msg, 0,
+				NL80211_MCGRP_REGULATORY, GFP_ATOMIC);
 	rcu_read_unlock();
 
 	return;
@@ -9873,8 +9873,8 @@ static void nl80211_send_mlme_event(struct cfg80211_registered_device *rdev,
 
 	genlmsg_end(msg, hdr);
 
-	genlmsg_multicast_netns(wiphy_net(&rdev->wiphy), msg, 0,
-				nl80211_mlme_mcgrp.id, gfp);
+	genlmsg_multicast_netns(&nl80211_fam, wiphy_net(&rdev->wiphy), msg, 0,
+				NL80211_MCGRP_MLME, gfp);
 	return;
 
  nla_put_failure:
@@ -9961,8 +9961,8 @@ static void nl80211_send_mlme_timeout(struct cfg80211_registered_device *rdev,
 
 	genlmsg_end(msg, hdr);
 
-	genlmsg_multicast_netns(wiphy_net(&rdev->wiphy), msg, 0,
-				nl80211_mlme_mcgrp.id, gfp);
+	genlmsg_multicast_netns(&nl80211_fam, wiphy_net(&rdev->wiphy), msg, 0,
+				NL80211_MCGRP_MLME, gfp);
 	return;
 
  nla_put_failure:
@@ -10017,8 +10017,8 @@ void nl80211_send_connect_result(struct cfg80211_registered_device *rdev,
 
 	genlmsg_end(msg, hdr);
 
-	genlmsg_multicast_netns(wiphy_net(&rdev->wiphy), msg, 0,
-				nl80211_mlme_mcgrp.id, gfp);
+	genlmsg_multicast_netns(&nl80211_fam, wiphy_net(&rdev->wiphy), msg, 0,
+				NL80211_MCGRP_MLME, gfp);
 	return;
 
  nla_put_failure:
@@ -10056,8 +10056,8 @@ void nl80211_send_roamed(struct cfg80211_registered_device *rdev,
 
 	genlmsg_end(msg, hdr);
 
-	genlmsg_multicast_netns(wiphy_net(&rdev->wiphy), msg, 0,
-				nl80211_mlme_mcgrp.id, gfp);
+	genlmsg_multicast_netns(&nl80211_fam, wiphy_net(&rdev->wiphy), msg, 0,
+				NL80211_MCGRP_MLME, gfp);
 	return;
 
  nla_put_failure:
@@ -10094,8 +10094,8 @@ void nl80211_send_disconnected(struct cfg80211_registered_device *rdev,
 
 	genlmsg_end(msg, hdr);
 
-	genlmsg_multicast_netns(wiphy_net(&rdev->wiphy), msg, 0,
-				nl80211_mlme_mcgrp.id, GFP_KERNEL);
+	genlmsg_multicast_netns(&nl80211_fam, wiphy_net(&rdev->wiphy), msg, 0,
+				NL80211_MCGRP_MLME, GFP_KERNEL);
 	return;
 
  nla_put_failure:
@@ -10128,8 +10128,8 @@ void nl80211_send_ibss_bssid(struct cfg80211_registered_device *rdev,
 
 	genlmsg_end(msg, hdr);
 
-	genlmsg_multicast_netns(wiphy_net(&rdev->wiphy), msg, 0,
-				nl80211_mlme_mcgrp.id, gfp);
+	genlmsg_multicast_netns(&nl80211_fam, wiphy_net(&rdev->wiphy), msg, 0,
+				NL80211_MCGRP_MLME, gfp);
 	return;
 
  nla_put_failure:
@@ -10169,8 +10169,8 @@ void cfg80211_notify_new_peer_candidate(struct net_device *dev, const u8 *addr,
 
 	genlmsg_end(msg, hdr);
 
-	genlmsg_multicast_netns(wiphy_net(&rdev->wiphy), msg, 0,
-				nl80211_mlme_mcgrp.id, gfp);
+	genlmsg_multicast_netns(&nl80211_fam, wiphy_net(&rdev->wiphy), msg, 0,
+				NL80211_MCGRP_MLME, gfp);
 	return;
 
  nla_put_failure:
@@ -10208,8 +10208,8 @@ void nl80211_michael_mic_failure(struct cfg80211_registered_device *rdev,
 
 	genlmsg_end(msg, hdr);
 
-	genlmsg_multicast_netns(wiphy_net(&rdev->wiphy), msg, 0,
-				nl80211_mlme_mcgrp.id, gfp);
+	genlmsg_multicast_netns(&nl80211_fam, wiphy_net(&rdev->wiphy), msg, 0,
+				NL80211_MCGRP_MLME, gfp);
 	return;
 
  nla_put_failure:
@@ -10261,8 +10261,8 @@ void nl80211_send_beacon_hint_event(struct wiphy *wiphy,
 	genlmsg_end(msg, hdr);
 
 	rcu_read_lock();
-	genlmsg_multicast_allns(msg, 0, nl80211_regulatory_mcgrp.id,
-				GFP_ATOMIC);
+	genlmsg_multicast_allns(&nl80211_fam, msg, 0,
+				NL80211_MCGRP_REGULATORY, GFP_ATOMIC);
 	rcu_read_unlock();
 
 	return;
@@ -10307,8 +10307,8 @@ static void nl80211_send_remain_on_chan_event(
 
 	genlmsg_end(msg, hdr);
 
-	genlmsg_multicast_netns(wiphy_net(&rdev->wiphy), msg, 0,
-				nl80211_mlme_mcgrp.id, gfp);
+	genlmsg_multicast_netns(&nl80211_fam, wiphy_net(&rdev->wiphy), msg, 0,
+				NL80211_MCGRP_MLME, gfp);
 	return;
 
  nla_put_failure:
@@ -10362,8 +10362,8 @@ void cfg80211_new_sta(struct net_device *dev, const u8 *mac_addr,
 		return;
 	}
 
-	genlmsg_multicast_netns(wiphy_net(&rdev->wiphy), msg, 0,
-				nl80211_mlme_mcgrp.id, gfp);
+	genlmsg_multicast_netns(&nl80211_fam, wiphy_net(&rdev->wiphy), msg, 0,
+				NL80211_MCGRP_MLME, gfp);
 }
 EXPORT_SYMBOL(cfg80211_new_sta);
 
@@ -10392,8 +10392,8 @@ void cfg80211_del_sta(struct net_device *dev, const u8 *mac_addr, gfp_t gfp)
 
 	genlmsg_end(msg, hdr);
 
-	genlmsg_multicast_netns(wiphy_net(&rdev->wiphy), msg, 0,
-				nl80211_mlme_mcgrp.id, gfp);
+	genlmsg_multicast_netns(&nl80211_fam, wiphy_net(&rdev->wiphy), msg, 0,
+				NL80211_MCGRP_MLME, gfp);
 	return;
 
  nla_put_failure:
@@ -10428,8 +10428,8 @@ void cfg80211_conn_failed(struct net_device *dev, const u8 *mac_addr,
 
 	genlmsg_end(msg, hdr);
 
-	genlmsg_multicast_netns(wiphy_net(&rdev->wiphy), msg, 0,
-				nl80211_mlme_mcgrp.id, gfp);
+	genlmsg_multicast_netns(&nl80211_fam, wiphy_net(&rdev->wiphy), msg, 0,
+				NL80211_MCGRP_MLME, gfp);
 	return;
 
  nla_put_failure:
@@ -10590,8 +10590,8 @@ void cfg80211_mgmt_tx_status(struct wireless_dev *wdev, u64 cookie,
 
 	genlmsg_end(msg, hdr);
 
-	genlmsg_multicast_netns(wiphy_net(&rdev->wiphy), msg, 0,
-				nl80211_mlme_mcgrp.id, gfp);
+	genlmsg_multicast_netns(&nl80211_fam, wiphy_net(&rdev->wiphy), msg, 0,
+				NL80211_MCGRP_MLME, gfp);
 	return;
 
  nla_put_failure:
@@ -10639,8 +10639,8 @@ void cfg80211_cqm_rssi_notify(struct net_device *dev,
 
 	genlmsg_end(msg, hdr);
 
-	genlmsg_multicast_netns(wiphy_net(&rdev->wiphy), msg, 0,
-				nl80211_mlme_mcgrp.id, gfp);
+	genlmsg_multicast_netns(&nl80211_fam, wiphy_net(&rdev->wiphy), msg, 0,
+				NL80211_MCGRP_MLME, gfp);
 	return;
 
  nla_put_failure:
@@ -10684,8 +10684,8 @@ static void nl80211_gtk_rekey_notify(struct cfg80211_registered_device *rdev,
 
 	genlmsg_end(msg, hdr);
 
-	genlmsg_multicast_netns(wiphy_net(&rdev->wiphy), msg, 0,
-				nl80211_mlme_mcgrp.id, gfp);
+	genlmsg_multicast_netns(&nl80211_fam, wiphy_net(&rdev->wiphy), msg, 0,
+				NL80211_MCGRP_MLME, gfp);
 	return;
 
  nla_put_failure:
@@ -10742,8 +10742,8 @@ nl80211_pmksa_candidate_notify(struct cfg80211_registered_device *rdev,
 
 	genlmsg_end(msg, hdr);
 
-	genlmsg_multicast_netns(wiphy_net(&rdev->wiphy), msg, 0,
-				nl80211_mlme_mcgrp.id, gfp);
+	genlmsg_multicast_netns(&nl80211_fam, wiphy_net(&rdev->wiphy), msg, 0,
+				NL80211_MCGRP_MLME, gfp);
 	return;
 
  nla_put_failure:
@@ -10789,8 +10789,8 @@ static void nl80211_ch_switch_notify(struct cfg80211_registered_device *rdev,
 
 	genlmsg_end(msg, hdr);
 
-	genlmsg_multicast_netns(wiphy_net(&rdev->wiphy), msg, 0,
-				nl80211_mlme_mcgrp.id, gfp);
+	genlmsg_multicast_netns(&nl80211_fam, wiphy_net(&rdev->wiphy), msg, 0,
+				NL80211_MCGRP_MLME, gfp);
 	return;
 
  nla_put_failure:
@@ -10866,8 +10866,8 @@ void cfg80211_cqm_txe_notify(struct net_device *dev,
 
 	genlmsg_end(msg, hdr);
 
-	genlmsg_multicast_netns(wiphy_net(&rdev->wiphy), msg, 0,
-				nl80211_mlme_mcgrp.id, gfp);
+	genlmsg_multicast_netns(&nl80211_fam, wiphy_net(&rdev->wiphy), msg, 0,
+				NL80211_MCGRP_MLME, gfp);
 	return;
 
  nla_put_failure:
@@ -10915,8 +10915,8 @@ nl80211_radar_notify(struct cfg80211_registered_device *rdev,
 
 	genlmsg_end(msg, hdr);
 
-	genlmsg_multicast_netns(wiphy_net(&rdev->wiphy), msg, 0,
-				nl80211_mlme_mcgrp.id, gfp);
+	genlmsg_multicast_netns(&nl80211_fam, wiphy_net(&rdev->wiphy), msg, 0,
+				NL80211_MCGRP_MLME, gfp);
 	return;
 
  nla_put_failure:
@@ -10962,8 +10962,8 @@ void cfg80211_cqm_pktloss_notify(struct net_device *dev,
 
 	genlmsg_end(msg, hdr);
 
-	genlmsg_multicast_netns(wiphy_net(&rdev->wiphy), msg, 0,
-				nl80211_mlme_mcgrp.id, gfp);
+	genlmsg_multicast_netns(&nl80211_fam, wiphy_net(&rdev->wiphy), msg, 0,
+				NL80211_MCGRP_MLME, gfp);
 	return;
 
  nla_put_failure:
@@ -11002,8 +11002,8 @@ void cfg80211_probe_status(struct net_device *dev, const u8 *addr,
 
 	genlmsg_end(msg, hdr);
 
-	genlmsg_multicast_netns(wiphy_net(&rdev->wiphy), msg, 0,
-				nl80211_mlme_mcgrp.id, gfp);
+	genlmsg_multicast_netns(&nl80211_fam, wiphy_net(&rdev->wiphy), msg, 0,
+				NL80211_MCGRP_MLME, gfp);
 	return;
 
  nla_put_failure:
@@ -11154,8 +11154,8 @@ void cfg80211_report_wowlan_wakeup(struct wireless_dev *wdev,
 
 	genlmsg_end(msg, hdr);
 
-	genlmsg_multicast_netns(wiphy_net(&rdev->wiphy), msg, 0,
-				nl80211_mlme_mcgrp.id, gfp);
+	genlmsg_multicast_netns(&nl80211_fam, wiphy_net(&rdev->wiphy), msg, 0,
+				NL80211_MCGRP_MLME, gfp);
 	return;
 
  free_msg:
@@ -11196,8 +11196,8 @@ void cfg80211_tdls_oper_request(struct net_device *dev, const u8 *peer,
 
 	genlmsg_end(msg, hdr);
 
-	genlmsg_multicast_netns(wiphy_net(&rdev->wiphy), msg, 0,
-				nl80211_mlme_mcgrp.id, gfp);
+	genlmsg_multicast_netns(&nl80211_fam, wiphy_net(&rdev->wiphy), msg, 0,
+				NL80211_MCGRP_MLME, gfp);
 	return;
 
  nla_put_failure:
@@ -11279,8 +11279,8 @@ void cfg80211_ft_event(struct net_device *netdev,
 
 	genlmsg_end(msg, hdr);
 
-	genlmsg_multicast_netns(wiphy_net(&rdev->wiphy), msg, 0,
-				nl80211_mlme_mcgrp.id, GFP_KERNEL);
+	genlmsg_multicast_netns(&nl80211_fam, wiphy_net(&rdev->wiphy), msg, 0,
+				NL80211_MCGRP_MLME, GFP_KERNEL);
 }
 EXPORT_SYMBOL(cfg80211_ft_event);
 
@@ -11329,33 +11329,11 @@ int nl80211_init(void)
 {
 	int err;
 
-	err = genl_register_family_with_ops(&nl80211_fam,
-		nl80211_ops, ARRAY_SIZE(nl80211_ops));
+	err = genl_register_family_with_ops_groups(&nl80211_fam, nl80211_ops,
+						   nl80211_mcgrps);
 	if (err)
 		return err;
 
-	err = genl_register_mc_group(&nl80211_fam, &nl80211_config_mcgrp);
-	if (err)
-		goto err_out;
-
-	err = genl_register_mc_group(&nl80211_fam, &nl80211_scan_mcgrp);
-	if (err)
-		goto err_out;
-
-	err = genl_register_mc_group(&nl80211_fam, &nl80211_regulatory_mcgrp);
-	if (err)
-		goto err_out;
-
-	err = genl_register_mc_group(&nl80211_fam, &nl80211_mlme_mcgrp);
-	if (err)
-		goto err_out;
-
-#ifdef CONFIG_NL80211_TESTMODE
-	err = genl_register_mc_group(&nl80211_fam, &nl80211_testmode_mcgrp);
-	if (err)
-		goto err_out;
-#endif
-
 	err = netlink_register_notifier(&nl80211_netlink_notifier);
 	if (err)
 		goto err_out;
diff --git a/samples/kfifo/bytestream-example.c b/samples/kfifo/bytestream-example.c
index cfe40addda76..2fca916d9edf 100644
--- a/samples/kfifo/bytestream-example.c
+++ b/samples/kfifo/bytestream-example.c
@@ -64,7 +64,7 @@ static int __init testfunc(void)
 
 	/* put values into the fifo */
 	for (i = 0; i != 10; i++)
-		kfifo_put(&test, &i);
+		kfifo_put(&test, i);
 
 	/* show the number of used elements */
 	printk(KERN_INFO "fifo len: %u\n", kfifo_len(&test));
@@ -85,7 +85,7 @@ static int __init testfunc(void)
 	kfifo_skip(&test);
 
 	/* put values into the fifo until is full */
-	for (i = 20; kfifo_put(&test, &i); i++)
+	for (i = 20; kfifo_put(&test, i); i++)
 		;
 
 	printk(KERN_INFO "queue len: %u\n", kfifo_len(&test));
diff --git a/samples/kfifo/dma-example.c b/samples/kfifo/dma-example.c
index 06473791c08a..aa243db93f01 100644
--- a/samples/kfifo/dma-example.c
+++ b/samples/kfifo/dma-example.c
@@ -39,7 +39,7 @@ static int __init example_init(void)
 	kfifo_in(&fifo, "test", 4);
 
 	for (i = 0; i != 9; i++)
-		kfifo_put(&fifo, &i);
+		kfifo_put(&fifo, i);
 
 	/* kick away first byte */
 	kfifo_skip(&fifo);
diff --git a/samples/kfifo/inttype-example.c b/samples/kfifo/inttype-example.c
index 6f8e79e76c9e..8dc3c2e7105a 100644
--- a/samples/kfifo/inttype-example.c
+++ b/samples/kfifo/inttype-example.c
@@ -61,7 +61,7 @@ static int __init testfunc(void)
 
 	/* put values into the fifo */
 	for (i = 0; i != 10; i++)
-		kfifo_put(&test, &i);
+		kfifo_put(&test, i);
 
 	/* show the number of used elements */
 	printk(KERN_INFO "fifo len: %u\n", kfifo_len(&test));
@@ -78,7 +78,7 @@ static int __init testfunc(void)
 	kfifo_skip(&test);
 
 	/* put values into the fifo until is full */
-	for (i = 20; kfifo_put(&test, &i); i++)
+	for (i = 20; kfifo_put(&test, i); i++)
 		;
 
 	printk(KERN_INFO "queue len: %u\n", kfifo_len(&test));
diff --git a/scripts/Makefile.modpost b/scripts/Makefile.modpost
index 8dcdca27d836..69f0a1417e9a 100644
--- a/scripts/Makefile.modpost
+++ b/scripts/Makefile.modpost
@@ -79,9 +79,11 @@ modpost = scripts/mod/modpost                    \
  $(if $(CONFIG_DEBUG_SECTION_MISMATCH),,-S)      \
  $(if $(KBUILD_EXTMOD)$(KBUILD_MODPOST_WARN),-w)
 
+MODPOST_OPT=$(subst -i,-n,$(filter -i,$(MAKEFLAGS)))
+
 # We can go over command line length here, so be careful.
 quiet_cmd_modpost = MODPOST $(words $(filter-out vmlinux FORCE, $^)) modules
-      cmd_modpost = $(MODLISTCMD) | sed 's/\.ko$$/.o/' | $(modpost) -s -T -
+      cmd_modpost = $(MODLISTCMD) | sed 's/\.ko$$/.o/' | $(modpost) $(MODPOST_OPT) -s -T -
 
 PHONY += __modpost
 __modpost: $(modules:.ko=.o) FORCE
diff --git a/scripts/bloat-o-meter b/scripts/bloat-o-meter
index 6129020c41a9..549d0ab8c662 100755
--- a/scripts/bloat-o-meter
+++ b/scripts/bloat-o-meter
@@ -19,9 +19,10 @@ def getsizes(file):
         size, type, name = l[:-1].split()
         if type in "tTdDbBrR":
             # strip generated symbols
-            if name[:6] == "__mod_": continue
-            # function names begin with '.' on 64-bit powerpc
-            if "." in name[1:]: name = "static." + name.split(".")[0]
+            if name.startswith("__mod_"): continue
+            if name == "linux_banner": continue
+            # statics and some other optimizations adds random .NUMBER
+            name = re.sub(r'\.[0-9]+', '', name)
             sym[name] = sym.get(name, 0) + int(size, 16)
     return sym
 
diff --git a/scripts/coccinelle/api/devm_request_and_ioremap.cocci b/scripts/coccinelle/api/devm_request_and_ioremap.cocci
deleted file mode 100644
index 562ec88b6352..000000000000
--- a/scripts/coccinelle/api/devm_request_and_ioremap.cocci
+++ /dev/null
@@ -1,105 +0,0 @@
-/// Reimplement a call to devm_request_mem_region followed by a call to ioremap
-/// or ioremap_nocache by a call to devm_request_and_ioremap.
-/// Devm_request_and_ioremap was introduced in
-/// 72f8c0bfa0de64c68ee59f40eb9b2683bffffbb0.  It makes the code much more
-/// concise.
-///
-///
-// Confidence: High
-// Copyright: (C) 2011 Julia Lawall, INRIA/LIP6.  GPLv2.
-// Copyright: (C) 2011 Gilles Muller, INRIA/LiP6.  GPLv2.
-// URL: http://coccinelle.lip6.fr/
-// Comments:
-// Options: --no-includes --include-headers
-
-virtual patch
-virtual org
-virtual report
-virtual context
-
-@nm@
-expression myname;
-identifier i;
-@@
-
-struct platform_driver i = { .driver = { .name = myname } };
-
-@depends on patch@
-expression dev,res,size;
-@@
-
--if (!devm_request_mem_region(dev, res->start, size,
--                              \(res->name\|dev_name(dev)\))) {
--   ...
--   return ...;
--}
-... when != res->start
-(
--devm_ioremap(dev,res->start,size)
-+devm_request_and_ioremap(dev,res)
-|
--devm_ioremap_nocache(dev,res->start,size)
-+devm_request_and_ioremap(dev,res)
-)
-... when any
-    when != res->start
-
-// this rule is separate from the previous one, because a single file can
-// have multiple values of myname
-@depends on patch@
-expression dev,res,size;
-expression nm.myname;
-@@
-
--if (!devm_request_mem_region(dev, res->start, size,myname)) {
--   ...
--   return ...;
--}
-... when != res->start
-(
--devm_ioremap(dev,res->start,size)
-+devm_request_and_ioremap(dev,res)
-|
--devm_ioremap_nocache(dev,res->start,size)
-+devm_request_and_ioremap(dev,res)
-)
-... when any
-    when != res->start
-
-
-@pb depends on org || report || context@
-expression dev,res,size;
-expression nm.myname;
-position p1,p2;
-@@
-
-*if
-  (!devm_request_mem_region@p1(dev, res->start, size,
-                              \(res->name\|dev_name(dev)\|myname\))) {
-   ...
-   return ...;
-}
-... when != res->start
-(
-*devm_ioremap@p2(dev,res->start,size)
-|
-*devm_ioremap_nocache@p2(dev,res->start,size)
-)
-... when any
-    when != res->start
-
-@script:python depends on org@
-p1 << pb.p1;
-p2 << pb.p2;
-@@
-
-cocci.print_main("INFO: replace by devm_request_and_ioremap",p1)
-cocci.print_secs("",p2)
-
-@script:python depends on report@
-p1 << pb.p1;
-p2 << pb.p2;
-@@
-
-msg = "INFO: devm_request_mem_region followed by ioremap on line %s can be replaced by devm_request_and_ioremap" % (p2[0].line)
-coccilib.report.print_report(p1[0],msg)
diff --git a/scripts/kallsyms.c b/scripts/kallsyms.c
index 9a11f9f799f4..10085de886fe 100644
--- a/scripts/kallsyms.c
+++ b/scripts/kallsyms.c
@@ -115,6 +115,12 @@ static int read_symbol(FILE *in, struct sym_entry *s)
 			fprintf(stderr, "Read error or end of file.\n");
 		return -1;
 	}
+	if (strlen(str) > KSYM_NAME_LEN) {
+		fprintf(stderr, "Symbol %s too long for kallsyms (%zu vs %d).\n"
+                                "Please increase KSYM_NAME_LEN both in kernel and kallsyms.c\n",
+			str, strlen(str), KSYM_NAME_LEN);
+		return -1;
+	}
 
 	sym = str;
 	/* skip prefix char */
diff --git a/scripts/kconfig/expr.h b/scripts/kconfig/expr.h
index df198a5f4822..ba663e1dc7e3 100644
--- a/scripts/kconfig/expr.h
+++ b/scripts/kconfig/expr.h
@@ -93,7 +93,7 @@ struct symbol {
 #define SYMBOL_CHOICEVAL  0x0020  /* used as a value in a choice block */
 #define SYMBOL_VALID      0x0080  /* set when symbol.curr is calculated */
 #define SYMBOL_OPTIONAL   0x0100  /* choice is optional - values can be 'n' */
-#define SYMBOL_WRITE      0x0200  /* ? */
+#define SYMBOL_WRITE      0x0200  /* write symbol to file (KCONFIG_CONFIG) */
 #define SYMBOL_CHANGED    0x0400  /* ? */
 #define SYMBOL_AUTO       0x1000  /* value from environment variable */
 #define SYMBOL_CHECKED    0x2000  /* used during dependency checking */
diff --git a/scripts/kconfig/mconf.c b/scripts/kconfig/mconf.c
index 2c3963165a0d..59184bb41ef8 100644
--- a/scripts/kconfig/mconf.c
+++ b/scripts/kconfig/mconf.c
@@ -25,7 +25,7 @@
 static const char mconf_readme[] = N_(
 "Overview\n"
 "--------\n"
-"This interface let you select features and parameters for the build.\n"
+"This interface lets you select features and parameters for the build.\n"
 "Features can either be built-in, modularized, or ignored. Parameters\n"
 "must be entered in as decimal or hexadecimal numbers or text.\n"
 "\n"
@@ -39,15 +39,15 @@ static const char mconf_readme[] = N_(
 "\n"
 "To change any of these features, highlight it with the cursor\n"
 "keys and press <Y> to build it in, <M> to make it a module or\n"
-"<N> to removed it.  You may also press the <Space Bar> to cycle\n"
-"through the available options (ie. Y->N->M->Y).\n"
+"<N> to remove it.  You may also press the <Space Bar> to cycle\n"
+"through the available options (i.e. Y->N->M->Y).\n"
 "\n"
 "Some additional keyboard hints:\n"
 "\n"
 "Menus\n"
 "----------\n"
-"o  Use the Up/Down arrow keys (cursor keys) to highlight the item\n"
-"   you wish to change or submenu wish to select and press <Enter>.\n"
+"o  Use the Up/Down arrow keys (cursor keys) to highlight the item you\n"
+"   wish to change or the submenu you wish to select and press <Enter>.\n"
 "   Submenus are designated by \"--->\", empty ones by \"----\".\n"
 "\n"
 "   Shortcut: Press the option's highlighted letter (hotkey).\n"
@@ -65,7 +65,7 @@ static const char mconf_readme[] = N_(
 "             there is a delayed response which you may find annoying.\n"
 "\n"
 "   Also, the <TAB> and cursor keys will cycle between <Select>,\n"
-"   <Exit> and <Help>.\n"
+"   <Exit>, <Help>, <Save>, and <Load>.\n"
 "\n"
 "o  To get help with an item, use the cursor keys to highlight <Help>\n"
 "   and press <ENTER>.\n"
@@ -105,7 +105,7 @@ static const char mconf_readme[] = N_(
 "Text Box    (Help Window)\n"
 "--------\n"
 "o  Use the cursor keys to scroll up/down/left/right.  The VI editor\n"
-"   keys h,j,k,l function here as do <u>, <d>, <SPACE BAR> and <B> for \n"
+"   keys h,j,k,l function here as do <u>, <d>, <SPACE BAR> and <B> for\n"
 "   those who are familiar with less and lynx.\n"
 "\n"
 "o  Press <E>, <X>, <q>, <Enter> or <Esc><Esc> to exit.\n"
@@ -117,23 +117,21 @@ static const char mconf_readme[] = N_(
 "those who, for various reasons, find it necessary to switch\n"
 "between different configurations.\n"
 "\n"
-"At the end of the main menu you will find two options.  One is\n"
-"for saving the current configuration to a file of your choosing.\n"
-"The other option is for loading a previously saved alternate\n"
-"configuration.\n"
+"The <Save> button will let you save the current configuration to\n"
+"a file of your choosing.  Use the <Load> button to load a previously\n"
+"saved alternate configuration.\n"
 "\n"
-"Even if you don't use alternate configuration files, but you\n"
-"find during a Menuconfig session that you have completely messed\n"
-"up your settings, you may use the \"Load Alternate...\" option to\n"
-"restore your previously saved settings from \".config\" without\n"
-"restarting Menuconfig.\n"
+"Even if you don't use alternate configuration files, but you find\n"
+"during a Menuconfig session that you have completely messed up your\n"
+"settings, you may use the <Load> button to restore your previously\n"
+"saved settings from \".config\" without restarting Menuconfig.\n"
 "\n"
 "Other information\n"
 "-----------------\n"
-"If you use Menuconfig in an XTERM window make sure you have your\n"
-"$TERM variable set to point to a xterm definition which supports color.\n"
-"Otherwise, Menuconfig will look rather bad.  Menuconfig will not\n"
-"display correctly in a RXVT window because rxvt displays only one\n"
+"If you use Menuconfig in an XTERM window, make sure you have your\n"
+"$TERM variable set to point to an xterm definition which supports\n"
+"color.  Otherwise, Menuconfig will look rather bad.  Menuconfig will\n"
+"not display correctly in an RXVT window because rxvt displays only one\n"
 "intensity of color, bright.\n"
 "\n"
 "Menuconfig will display larger menus on screens or xterms which are\n"
@@ -148,8 +146,8 @@ static const char mconf_readme[] = N_(
 "\n"
 "Optional personality available\n"
 "------------------------------\n"
-"If you prefer to have all of the options listed in a single menu, rather\n"
-"than the default multimenu hierarchy, run the menuconfig with\n"
+"If you prefer to have all of the options listed in a single menu,\n"
+"rather than the default multimenu hierarchy, run the menuconfig with\n"
 "MENUCONFIG_MODE environment variable set to single_menu. Example:\n"
 "\n"
 "make MENUCONFIG_MODE=single_menu menuconfig\n"
@@ -172,7 +170,7 @@ static const char mconf_readme[] = N_(
 " mono       => selects colors suitable for monochrome displays\n"
 " blackbg    => selects a color scheme with black background\n"
 " classic    => theme with blue background. The classic look\n"
-" bluetitle  => a LCD friendly version of classic. (default)\n"
+" bluetitle  => an LCD friendly version of classic. (default)\n"
 "\n"),
 menu_instructions[] = N_(
 	"Arrow keys navigate the menu.  "
@@ -238,24 +236,24 @@ search_help[] = N_(
 	"Symbol: FOO [=m]\n"
 	"Type  : tristate\n"
 	"Prompt: Foo bus is used to drive the bar HW\n"
-	"  Defined at drivers/pci/Kconfig:47\n"
-	"  Depends on: X86_LOCAL_APIC && X86_IO_APIC || IA64\n"
 	"  Location:\n"
 	"    -> Bus options (PCI, PCMCIA, EISA, ISA)\n"
 	"      -> PCI support (PCI [=y])\n"
 	"(1)     -> PCI access mode (<choice> [=y])\n"
+	"  Defined at drivers/pci/Kconfig:47\n"
+	"  Depends on: X86_LOCAL_APIC && X86_IO_APIC || IA64\n"
 	"  Selects: LIBCRC32\n"
-	"  Selected by: BAR\n"
+	"  Selected by: BAR [=n]\n"
 	"-----------------------------------------------------------------\n"
 	"o The line 'Type:' shows the type of the configuration option for\n"
 	"  this symbol (boolean, tristate, string, ...)\n"
 	"o The line 'Prompt:' shows the text used in the menu structure for\n"
 	"  this symbol\n"
-	"o The 'Defined at' line tell at what file / line number the symbol\n"
+	"o The 'Defined at' line tells at what file / line number the symbol\n"
 	"  is defined\n"
-	"o The 'Depends on:' line tell what symbols needs to be defined for\n"
+	"o The 'Depends on:' line tells what symbols need to be defined for\n"
 	"  this symbol to be visible in the menu (selectable)\n"
-	"o The 'Location:' lines tell where in the menu structure this symbol\n"
+	"o The 'Location:' lines tells where in the menu structure this symbol\n"
 	"  is located\n"
 	"    A location followed by a [=y] indicates that this is a\n"
 	"    selectable menu item - and the current value is displayed inside\n"
@@ -263,9 +261,9 @@ search_help[] = N_(
 	"    Press the key in the (#) prefix to jump directly to that\n"
 	"    location. You will be returned to the current search results\n"
 	"    after exiting this new menu.\n"
-	"o The 'Selects:' line tell what symbol will be automatically\n"
+	"o The 'Selects:' line tells what symbols will be automatically\n"
 	"  selected if this symbol is selected (y or m)\n"
-	"o The 'Selected by' line tell what symbol has selected this symbol\n"
+	"o The 'Selected by' line tells what symbol has selected this symbol\n"
 	"\n"
 	"Only relevant lines are shown.\n"
 	"\n\n"
diff --git a/scripts/kconfig/menu.c b/scripts/kconfig/menu.c
index c1d53200c306..db1512ae30cc 100644
--- a/scripts/kconfig/menu.c
+++ b/scripts/kconfig/menu.c
@@ -119,9 +119,10 @@ void menu_set_type(int type)
 		sym->type = type;
 		return;
 	}
-	menu_warn(current_entry, "type of '%s' redefined from '%s' to '%s'",
-	    sym->name ? sym->name : "<choice>",
-	    sym_type_name(sym->type), sym_type_name(type));
+	menu_warn(current_entry,
+		"ignoring type redefinition of '%s' from '%s' to '%s'",
+		sym->name ? sym->name : "<choice>",
+		sym_type_name(sym->type), sym_type_name(type));
 }
 
 struct property *menu_add_prop(enum prop_type type, char *prompt, struct expr *expr, struct expr *dep)
@@ -583,7 +584,7 @@ static void get_prompt_str(struct gstr *r, struct property *prop,
 		for (j = 4; --i >= 0; j += 2) {
 			menu = submenu[i];
 			if (head && location && menu == location)
-				jump->offset = r->len - 1;
+				jump->offset = strlen(r->s);
 			str_printf(r, "%*c-> %s", j, ' ',
 				   _(menu_get_prompt(menu)));
 			if (menu->sym) {
@@ -597,7 +598,7 @@ static void get_prompt_str(struct gstr *r, struct property *prop,
 }
 
 /*
- * get peoperty of type P_SYMBOL
+ * get property of type P_SYMBOL
  */
 static struct property *get_symbol_prop(struct symbol *sym)
 {
diff --git a/scripts/kconfig/qconf.cc b/scripts/kconfig/qconf.cc
index 1500c38f0cca..9d3b04b0769c 100644
--- a/scripts/kconfig/qconf.cc
+++ b/scripts/kconfig/qconf.cc
@@ -69,6 +69,11 @@ static inline QString qgettext(const QString& str)
 	return QString::fromLocal8Bit(gettext(str.latin1()));
 }
 
+ConfigSettings::ConfigSettings()
+	: QSettings("kernel.org", "qconf")
+{
+}
+
 /**
  * Reads a list of integer values from the application settings.
  */
diff --git a/scripts/kconfig/qconf.h b/scripts/kconfig/qconf.h
index 3715b3e7212c..bde0c6b6f9e8 100644
--- a/scripts/kconfig/qconf.h
+++ b/scripts/kconfig/qconf.h
@@ -32,6 +32,7 @@ class ConfigMainWindow;
 
 class ConfigSettings : public QSettings {
 public:
+	ConfigSettings();
 	Q3ValueList<int> readSizes(const QString& key, bool *ok);
 	bool writeSizes(const QString& key, const Q3ValueList<int>& value);
 };
diff --git a/scripts/kconfig/symbol.c b/scripts/kconfig/symbol.c
index c9a6775565bf..7caabdb51c64 100644
--- a/scripts/kconfig/symbol.c
+++ b/scripts/kconfig/symbol.c
@@ -1047,7 +1047,7 @@ sym_re_search_free:
  * When we check for recursive dependencies we use a stack to save
  * current state so we can print out relevant info to user.
  * The entries are located on the call stack so no need to free memory.
- * Note inser() remove() must always match to properly clear the stack.
+ * Note insert() remove() must always match to properly clear the stack.
  */
 static struct dep_stack {
 	struct dep_stack *prev, *next;
diff --git a/scripts/kconfig/zconf.l b/scripts/kconfig/zconf.l
index 6555a475453b..1a9f53e535ca 100644
--- a/scripts/kconfig/zconf.l
+++ b/scripts/kconfig/zconf.l
@@ -68,7 +68,6 @@ static void alloc_string(const char *str, int size)
 }
 %}
 
-ws	[ \n\t]
 n	[A-Za-z0-9_]
 
 %%
diff --git a/scripts/kernel-doc b/scripts/kernel-doc
index dbd3e1ebbdad..da058da413e7 100755
--- a/scripts/kernel-doc
+++ b/scripts/kernel-doc
@@ -2128,8 +2128,7 @@ sub dump_function($$) {
 
 	create_parameterlist($args, ',', $file);
     } else {
-	print STDERR "Error(${file}:$.): cannot understand prototype: '$prototype'\n";
-	++$errors;
+	print STDERR "Warning(${file}:$.): cannot understand function prototype: '$prototype'\n";
 	return;
     }
 
diff --git a/scripts/mod/modpost.c b/scripts/mod/modpost.c
index bfcea5d3b27d..17855761e6b7 100644
--- a/scripts/mod/modpost.c
+++ b/scripts/mod/modpost.c
@@ -17,6 +17,7 @@
 #include <string.h>
 #include <limits.h>
 #include <stdbool.h>
+#include <errno.h>
 #include "modpost.h"
 #include "../../include/generated/autoconf.h"
 #include "../../include/linux/license.h"
@@ -37,6 +38,8 @@ static int warn_unresolved = 0;
 /* How a symbol is exported */
 static int sec_mismatch_count = 0;
 static int sec_mismatch_verbose = 1;
+/* ignore missing files */
+static int ignore_missing_files;
 
 enum export {
 	export_plain,      export_unused,     export_gpl,
@@ -161,7 +164,7 @@ struct symbol {
 	unsigned int vmlinux:1;    /* 1 if symbol is defined in vmlinux */
 	unsigned int kernel:1;     /* 1 if symbol is from kernel
 				    *  (only for external modules) **/
-	unsigned int preloaded:1;  /* 1 if symbol from Module.symvers */
+	unsigned int preloaded:1;  /* 1 if symbol from Module.symvers, or crc */
 	enum export  export;       /* Type of export */
 	char name[0];
 };
@@ -329,8 +332,11 @@ static void sym_update_crc(const char *name, struct module *mod,
 {
 	struct symbol *s = find_symbol(name);
 
-	if (!s)
+	if (!s) {
 		s = new_symbol(name, mod, export);
+		/* Don't complain when we find it later. */
+		s->preloaded = 1;
+	}
 	s->crc = crc;
 	s->crc_valid = 1;
 }
@@ -407,6 +413,11 @@ static int parse_elf(struct elf_info *info, const char *filename)
 
 	hdr = grab_file(filename, &info->size);
 	if (!hdr) {
+		if (ignore_missing_files) {
+			fprintf(stderr, "%s: %s (ignored)\n", filename,
+				strerror(errno));
+			return 0;
+		}
 		perror(filename);
 		exit(1);
 	}
@@ -1852,7 +1863,7 @@ static void add_header(struct buffer *b, struct module *mod)
 	buf_printf(b, "\n");
 	buf_printf(b, "MODULE_INFO(vermagic, VERMAGIC_STRING);\n");
 	buf_printf(b, "\n");
-	buf_printf(b, "struct module __this_module\n");
+	buf_printf(b, "__visible struct module __this_module\n");
 	buf_printf(b, "__attribute__((section(\".gnu.linkonce.this_module\"))) = {\n");
 	buf_printf(b, "\t.name = KBUILD_MODNAME,\n");
 	if (mod->has_init)
@@ -2118,7 +2129,7 @@ int main(int argc, char **argv)
 	struct ext_sym_list *extsym_iter;
 	struct ext_sym_list *extsym_start = NULL;
 
-	while ((opt = getopt(argc, argv, "i:I:e:msST:o:awM:K:")) != -1) {
+	while ((opt = getopt(argc, argv, "i:I:e:mnsST:o:awM:K:")) != -1) {
 		switch (opt) {
 		case 'i':
 			kernel_read = optarg;
@@ -2138,6 +2149,9 @@ int main(int argc, char **argv)
 		case 'm':
 			modversions = 1;
 			break;
+		case 'n':
+			ignore_missing_files = 1;
+			break;
 		case 'o':
 			dump_write = optarg;
 			break;
diff --git a/scripts/mod/sumversion.c b/scripts/mod/sumversion.c
index 9dfcd6d988da..deb2994b04c4 100644
--- a/scripts/mod/sumversion.c
+++ b/scripts/mod/sumversion.c
@@ -416,7 +416,7 @@ void get_src_version(const char *modname, char sum[], unsigned sumlen)
 		basename = strrchr(modname, '/') + 1;
 	else
 		basename = modname;
-	sprintf(filelist, "%s/%.*s.mod", modverdir,
+	snprintf(filelist, sizeof(filelist), "%s/%.*s.mod", modverdir,
 		(int) strlen(basename) - 2, basename);
 
 	file = grab_file(filelist, &len);
diff --git a/scripts/recordmcount.pl b/scripts/recordmcount.pl
index a674fd5507c1..d0da66396f62 100755
--- a/scripts/recordmcount.pl
+++ b/scripts/recordmcount.pl
@@ -214,13 +214,13 @@ $local_regex = "^[0-9a-fA-F]+\\s+t\\s+(\\S+)";
 $weak_regex = "^[0-9a-fA-F]+\\s+([wW])\\s+(\\S+)";
 $section_regex = "Disassembly of section\\s+(\\S+):";
 $function_regex = "^([0-9a-fA-F]+)\\s+<(.*?)>:";
-$mcount_regex = "^\\s*([0-9a-fA-F]+):.*\\smcount\$";
+$mcount_regex = "^\\s*([0-9a-fA-F]+):.*\\s(mcount|__fentry__)\$";
 $section_type = '@progbits';
 $mcount_adjust = 0;
 $type = ".long";
 
 if ($arch eq "x86_64") {
-    $mcount_regex = "^\\s*([0-9a-fA-F]+):.*\\smcount([+-]0x[0-9a-zA-Z]+)?\$";
+    $mcount_regex = "^\\s*([0-9a-fA-F]+):.*\\s(mcount|__fentry__)([+-]0x[0-9a-zA-Z]+)?\$";
     $type = ".quad";
     $alignment = 8;
     $mcount_adjust = -1;
diff --git a/scripts/show_delta b/scripts/show_delta
index 17df3051747a..e25732b5d701 100755
--- a/scripts/show_delta
+++ b/scripts/show_delta
@@ -13,7 +13,7 @@ import sys
 import string
 
 def usage():
-	print """usage: show_delta [<options>] <filename>
+	print ("""usage: show_delta [<options>] <filename>
 
 This program parses the output from a set of printk message lines which
 have time data prefixed because the CONFIG_PRINTK_TIME option is set, or
@@ -35,7 +35,7 @@ ex: $ dmesg >timefile
 
 will show times relative to the line in the kernel output
 starting with "NET4".
-"""
+""")
 	sys.exit(1)
 
 # returns a tuple containing the seconds and text for each message line
@@ -94,11 +94,11 @@ def main():
 	try:
 		lines = open(filein,"r").readlines()
 	except:
-		print "Problem opening file: %s" % filein
+		print ("Problem opening file: %s" % filein)
 		sys.exit(1)
 
 	if base_str:
-		print 'base= "%s"' % base_str
+		print ('base= "%s"' % base_str)
 		# assume a numeric base.  If that fails, try searching
 		# for a matching line.
 		try:
@@ -117,13 +117,13 @@ def main():
 					# stop at first match
 					break
 			if not found:
-				print 'Couldn\'t find line matching base pattern "%s"' % base_str
+				print ('Couldn\'t find line matching base pattern "%s"' % base_str)
 				sys.exit(1)
 	else:
 		base_time = 0.0
 
 	for line in lines:
-		print convert_line(line, base_time),
+		print (convert_line(line, base_time),)
 
 main()
 
diff --git a/scripts/tags.sh b/scripts/tags.sh
index 74f02e4dddd2..58c455929091 100755
--- a/scripts/tags.sh
+++ b/scripts/tags.sh
@@ -149,15 +149,16 @@ dogtags()
 exuberant()
 {
 	all_target_sources | xargs $1 -a                        \
-	-I __initdata,__exitdata,__initconst,__devinitdata	\
-	-I __devinitconst,__cpuinitdata,__initdata_memblock	\
-	-I __refdata,__attribute				\
+	-I __initdata,__exitdata,__initconst,			\
+	-I __cpuinitdata,__initdata_memblock			\
+	-I __refdata,__attribute,__maybe_unused,__always_unused \
 	-I __acquires,__releases,__deprecated			\
 	-I __read_mostly,__aligned,____cacheline_aligned        \
 	-I ____cacheline_aligned_in_smp                         \
+	-I __cacheline_aligned,__cacheline_aligned_in_smp	\
 	-I ____cacheline_internodealigned_in_smp                \
 	-I __used,__packed,__packed2__,__must_check,__must_hold	\
-	-I EXPORT_SYMBOL,EXPORT_SYMBOL_GPL                      \
+	-I EXPORT_SYMBOL,EXPORT_SYMBOL_GPL,ACPI_EXPORT_SYMBOL   \
 	-I DEFINE_TRACE,EXPORT_TRACEPOINT_SYMBOL,EXPORT_TRACEPOINT_SYMBOL_GPL \
 	-I static,const						\
 	--extra=+f --c-kinds=+px                                \
diff --git a/sound/aoa/fabrics/layout.c b/sound/aoa/fabrics/layout.c
index 61ab640e195f..9dc5806d23dd 100644
--- a/sound/aoa/fabrics/layout.c
+++ b/sound/aoa/fabrics/layout.c
@@ -644,7 +644,7 @@ static int n##_control_put(struct snd_kcontrol *kcontrol,		\
 			   struct snd_ctl_elem_value *ucontrol)		\
 {									\
 	struct gpio_runtime *gpio = snd_kcontrol_chip(kcontrol);	\
-	if (gpio->methods && gpio->methods->get_##n)			\
+	if (gpio->methods && gpio->methods->set_##n)			\
 		gpio->methods->set_##n(gpio,				\
 			!!ucontrol->value.integer.value[0]);		\
 	return 1;							\
@@ -1135,7 +1135,7 @@ static int aoa_fabric_layout_resume(struct soundbus_dev *sdev)
 {
 	struct layout_dev *ldev = dev_get_drvdata(&sdev->ofdev.dev);
 
-	if (ldev->gpio.methods && ldev->gpio.methods->all_amps_off)
+	if (ldev->gpio.methods && ldev->gpio.methods->all_amps_restore)
 		ldev->gpio.methods->all_amps_restore(&ldev->gpio);
 
 	return 0;
diff --git a/sound/core/compress_offload.c b/sound/core/compress_offload.c
index d9af6387f37c..9d518ac73eea 100644
--- a/sound/core/compress_offload.c
+++ b/sound/core/compress_offload.c
@@ -384,8 +384,7 @@ static unsigned int snd_compr_poll(struct file *f, poll_table *wait)
 		return -EFAULT;
 
 	mutex_lock(&stream->device->lock);
-	if (stream->runtime->state == SNDRV_PCM_STATE_PAUSED ||
-			stream->runtime->state == SNDRV_PCM_STATE_OPEN) {
+	if (stream->runtime->state == SNDRV_PCM_STATE_OPEN) {
 		retval = -EBADFD;
 		goto out;
 	}
diff --git a/sound/core/jack.c b/sound/core/jack.c
index b35fe7345c20..8658578eb584 100644
--- a/sound/core/jack.c
+++ b/sound/core/jack.c
@@ -34,12 +34,12 @@ static int jack_switch_types[SND_JACK_SWITCH_TYPES] = {
 	SW_LINEIN_INSERT,
 };
 
-static int snd_jack_dev_free(struct snd_device *device)
+static int snd_jack_dev_disconnect(struct snd_device *device)
 {
 	struct snd_jack *jack = device->device_data;
 
-	if (jack->private_free)
-		jack->private_free(jack);
+	if (!jack->input_dev)
+		return 0;
 
 	/* If the input device is registered with the input subsystem
 	 * then we need to use a different deallocator. */
@@ -47,6 +47,18 @@ static int snd_jack_dev_free(struct snd_device *device)
 		input_unregister_device(jack->input_dev);
 	else
 		input_free_device(jack->input_dev);
+	jack->input_dev = NULL;
+	return 0;
+}
+
+static int snd_jack_dev_free(struct snd_device *device)
+{
+	struct snd_jack *jack = device->device_data;
+
+	if (jack->private_free)
+		jack->private_free(jack);
+
+	snd_jack_dev_disconnect(device);
 
 	kfree(jack->id);
 	kfree(jack);
@@ -110,6 +122,7 @@ int snd_jack_new(struct snd_card *card, const char *id, int type,
 	static struct snd_device_ops ops = {
 		.dev_free = snd_jack_dev_free,
 		.dev_register = snd_jack_dev_register,
+		.dev_disconnect = snd_jack_dev_disconnect,
 	};
 
 	jack = kzalloc(sizeof(struct snd_jack), GFP_KERNEL);
diff --git a/sound/core/memalloc.c b/sound/core/memalloc.c
index 9d93f02c6285..5e1c7bc73b29 100644
--- a/sound/core/memalloc.c
+++ b/sound/core/memalloc.c
@@ -184,11 +184,7 @@ static void snd_malloc_dev_iram(struct snd_dma_buffer *dmab, size_t size)
 	/* Assign the pool into private_data field */
 	dmab->private_data = pool;
 
-	dmab->area = (void *)gen_pool_alloc(pool, size);
-	if (!dmab->area)
-		return;
-
-	dmab->addr = gen_pool_virt_to_phys(pool, (unsigned long)dmab->area);
+	dmab->area = gen_pool_dma_alloc(pool, size, &dmab->addr);
 }
 
 /**
diff --git a/sound/drivers/pcsp/pcsp.c b/sound/drivers/pcsp/pcsp.c
index f664bae3b9b0..328bd29264ce 100644
--- a/sound/drivers/pcsp/pcsp.c
+++ b/sound/drivers/pcsp/pcsp.c
@@ -188,8 +188,8 @@ static int pcsp_probe(struct platform_device *dev)
 static int pcsp_remove(struct platform_device *dev)
 {
 	struct snd_pcsp *chip = platform_get_drvdata(dev);
-	alsa_card_pcsp_exit(chip);
 	pcspkr_input_remove(chip->input_dev);
+	alsa_card_pcsp_exit(chip);
 	return 0;
 }
 
diff --git a/sound/firewire/dice.c b/sound/firewire/dice.c
index 6feee6614193..57bcd31fcc12 100644
--- a/sound/firewire/dice.c
+++ b/sound/firewire/dice.c
@@ -543,7 +543,7 @@ static int dice_change_rate(struct dice *dice, unsigned int clock_rate)
 	__be32 value;
 	int err;
 
-	INIT_COMPLETION(dice->clock_accepted);
+	reinit_completion(&dice->clock_accepted);
 
 	value = cpu_to_be32(clock_rate | CLOCK_SOURCE_ARX1);
 	err = snd_fw_transaction(dice->unit, TCODE_WRITE_QUADLET_REQUEST,
diff --git a/sound/isa/msnd/msnd_pinnacle.c b/sound/isa/msnd/msnd_pinnacle.c
index 81aeb934261a..0a90bd6ae232 100644
--- a/sound/isa/msnd/msnd_pinnacle.c
+++ b/sound/isa/msnd/msnd_pinnacle.c
@@ -73,9 +73,11 @@
 #ifdef MSND_CLASSIC
 #  include "msnd_classic.h"
 #  define LOGNAME			"msnd_classic"
+#  define DEV_NAME			"msnd-classic"
 #else
 #  include "msnd_pinnacle.h"
 #  define LOGNAME			"snd_msnd_pinnacle"
+#  define DEV_NAME			"msnd-pinnacle"
 #endif
 
 static void set_default_audio_parameters(struct snd_msnd *chip)
@@ -1067,8 +1069,6 @@ static int snd_msnd_isa_remove(struct device *pdev, unsigned int dev)
 	return 0;
 }
 
-#define DEV_NAME "msnd-pinnacle"
-
 static struct isa_driver snd_msnd_driver = {
 	.match		= snd_msnd_isa_match,
 	.probe		= snd_msnd_isa_probe,
diff --git a/sound/isa/wavefront/wavefront_synth.c b/sound/isa/wavefront/wavefront_synth.c
index a2f87f9488ee..e5db001363ee 100644
--- a/sound/isa/wavefront/wavefront_synth.c
+++ b/sound/isa/wavefront/wavefront_synth.c
@@ -1196,7 +1196,7 @@ wavefront_send_multisample (snd_wavefront_t *dev, wavefront_patch_info *header)
 	int num_samples;
 	unsigned char *msample_hdr;
 
-	msample_hdr = kmalloc(sizeof(WF_MSAMPLE_BYTES), GFP_KERNEL);
+	msample_hdr = kmalloc(WF_MSAMPLE_BYTES, GFP_KERNEL);
 	if (! msample_hdr)
 		return -ENOMEM;
 
diff --git a/sound/pci/hda/hda_codec.c b/sound/pci/hda/hda_codec.c
index dd5403d40830..afb90f48867f 100644
--- a/sound/pci/hda/hda_codec.c
+++ b/sound/pci/hda/hda_codec.c
@@ -2579,9 +2579,6 @@ int snd_hda_codec_reset(struct hda_codec *codec)
 	cancel_delayed_work_sync(&codec->jackpoll_work);
 #ifdef CONFIG_PM
 	cancel_delayed_work_sync(&codec->power_work);
-	codec->power_on = 0;
-	codec->power_transition = 0;
-	codec->power_jiffies = jiffies;
 	flush_workqueue(bus->workq);
 #endif
 	snd_hda_ctls_clear(codec);
diff --git a/sound/pci/hda/hda_eld.c b/sound/pci/hda/hda_eld.c
index 32d3e3855a6e..79ca80f6c77a 100644
--- a/sound/pci/hda/hda_eld.c
+++ b/sound/pci/hda/hda_eld.c
@@ -680,7 +680,7 @@ int snd_hdmi_get_eld_ati(struct hda_codec *codec, hda_nid_t nid,
 
 	spkalloc = snd_hda_codec_read(codec, nid, 0, ATI_VERB_GET_SPEAKER_ALLOCATION, 0);
 
-	if (!spkalloc) {
+	if (spkalloc <= 0) {
 		snd_printd(KERN_INFO "HDMI ATI/AMD: no speaker allocation for ELD\n");
 		return -EINVAL;
 	}
@@ -742,6 +742,9 @@ int snd_hdmi_get_eld_ati(struct hda_codec *codec, hda_nid_t nid,
 		snd_hda_codec_write(codec, nid, 0, ATI_VERB_SET_AUDIO_DESCRIPTOR, i << 3);
 		ati_sad = snd_hda_codec_read(codec, nid, 0, ATI_VERB_GET_AUDIO_DESCRIPTOR, 0);
 
+		if (ati_sad <= 0)
+			continue;
+
 		if (ati_sad & ATI_AUDIODESC_RATES) {
 			/* format is supported, copy SAD as-is */
 			buf[pos++] = (ati_sad & 0x0000ff) >> 0;
@@ -765,21 +768,39 @@ int snd_hdmi_get_eld_ati(struct hda_codec *codec, hda_nid_t nid,
 		return -EINVAL;
 	}
 
+	/*
+	 * HDMI VSDB latency format:
+	 * separately for both audio and video:
+	 *  0          field not valid or unknown latency
+	 *  [1..251]   msecs = (x-1)*2  (max 500ms with x = 251 = 0xfb)
+	 *  255        audio/video not supported
+	 *
+	 * HDA latency format:
+	 * single value indicating video latency relative to audio:
+	 *  0          unknown or 0ms
+	 *  [1..250]   msecs = x*2  (max 500ms with x = 250 = 0xfa)
+	 *  [251..255] reserved
+	 */
 	aud_synch = snd_hda_codec_read(codec, nid, 0, ATI_VERB_GET_AUDIO_VIDEO_DELAY, 0);
 	if ((aud_synch & ATI_DELAY_VIDEO_LATENCY) && (aud_synch & ATI_DELAY_AUDIO_LATENCY)) {
-		int video_latency = (aud_synch & ATI_DELAY_VIDEO_LATENCY) - 1;
-		int audio_latency = ((aud_synch & ATI_DELAY_AUDIO_LATENCY) >> 8) - 1;
+		int video_latency_hdmi = (aud_synch & ATI_DELAY_VIDEO_LATENCY);
+		int audio_latency_hdmi = (aud_synch & ATI_DELAY_AUDIO_LATENCY) >> 8;
 
-		if (video_latency > audio_latency)
-			buf[6] = min(video_latency - audio_latency, 0xfa);
+		if (video_latency_hdmi <= 0xfb && audio_latency_hdmi <= 0xfb &&
+		    video_latency_hdmi > audio_latency_hdmi)
+			buf[6] = video_latency_hdmi - audio_latency_hdmi;
+		/* else unknown/invalid or 0ms or video ahead of audio, so use zero */
 	}
 
-	/* Baseline length */
-	buf[2] = pos - 4;
-
 	/* SAD count */
 	buf[5] |= ((pos - ELD_FIXED_BYTES - sink_desc_len) / 3) << 4;
 
+	/* Baseline ELD block length is 4-byte aligned */
+	pos = round_up(pos, 4);
+
+	/* Baseline ELD length (4-byte header is not counted in) */
+	buf[2] = (pos - 4) / 4;
+
 	*eld_size = pos;
 
 	return 0;
diff --git a/sound/pci/hda/hda_generic.c b/sound/pci/hda/hda_generic.c
index 276f6e759bac..3067ed4fe3b2 100644
--- a/sound/pci/hda/hda_generic.c
+++ b/sound/pci/hda/hda_generic.c
@@ -796,10 +796,10 @@ static void set_pin_eapd(struct hda_codec *codec, hda_nid_t pin, bool enable)
 	if (spec->own_eapd_ctl ||
 	    !(snd_hda_query_pin_caps(codec, pin) & AC_PINCAP_EAPD))
 		return;
-	if (codec->inv_eapd)
-		enable = !enable;
 	if (spec->keep_eapd_on && !enable)
 		return;
+	if (codec->inv_eapd)
+		enable = !enable;
 	snd_hda_codec_update_cache(codec, pin, 0,
 				   AC_VERB_SET_EAPD_BTLENABLE,
 				   enable ? 0x02 : 0x00);
diff --git a/sound/pci/hda/patch_analog.c b/sound/pci/hda/patch_analog.c
index 87d2e0335ae4..1a83559f4cbd 100644
--- a/sound/pci/hda/patch_analog.c
+++ b/sound/pci/hda/patch_analog.c
@@ -139,6 +139,18 @@ static int ad198x_suspend(struct hda_codec *codec)
 }
 #endif
 
+/* follow EAPD via vmaster hook */
+static void ad_vmaster_eapd_hook(void *private_data, int enabled)
+{
+	struct hda_codec *codec = private_data;
+	struct ad198x_spec *spec = codec->spec;
+
+	if (!spec->eapd_nid)
+		return;
+	snd_hda_codec_update_cache(codec, spec->eapd_nid, 0,
+				   AC_VERB_SET_EAPD_BTLENABLE,
+				   enabled ? 0x02 : 0x00);
+}
 
 /*
  * Automatic parse of I/O pins from the BIOS configuration
@@ -219,8 +231,14 @@ static int alloc_ad_spec(struct hda_codec *codec)
 static void ad_fixup_inv_jack_detect(struct hda_codec *codec,
 				     const struct hda_fixup *fix, int action)
 {
-	if (action == HDA_FIXUP_ACT_PRE_PROBE)
+	struct ad198x_spec *spec = codec->spec;
+
+	if (action == HDA_FIXUP_ACT_PRE_PROBE) {
 		codec->inv_jack_detect = 1;
+		spec->gen.keep_eapd_on = 1;
+		spec->gen.vmaster_mute.hook = ad_vmaster_eapd_hook;
+		spec->eapd_nid = 0x1b;
+	}
 }
 
 enum {
@@ -465,19 +483,6 @@ static int patch_ad1983(struct hda_codec *codec)
  * AD1981 HD specific
  */
 
-/* follow EAPD via vmaster hook */
-static void ad_vmaster_eapd_hook(void *private_data, int enabled)
-{
-	struct hda_codec *codec = private_data;
-	struct ad198x_spec *spec = codec->spec;
-
-	if (!spec->eapd_nid)
-		return;
-	snd_hda_codec_update_cache(codec, spec->eapd_nid, 0,
-				   AC_VERB_SET_EAPD_BTLENABLE,
-				   enabled ? 0x02 : 0x00);
-}
-
 static void ad1981_fixup_hp_eapd(struct hda_codec *codec,
 				 const struct hda_fixup *fix, int action)
 {
diff --git a/sound/pci/hda/patch_cirrus.c b/sound/pci/hda/patch_cirrus.c
index 072755c8289c..fc492ac24caa 100644
--- a/sound/pci/hda/patch_cirrus.c
+++ b/sound/pci/hda/patch_cirrus.c
@@ -47,6 +47,10 @@ struct cs_spec {
 	unsigned int spdif_present:1;
 	unsigned int sense_b:1;
 	hda_nid_t vendor_nid;
+
+	/* for MBP SPDIF control */
+	int (*spdif_sw_put)(struct snd_kcontrol *kcontrol,
+			    struct snd_ctl_elem_value *ucontrol);
 };
 
 /* available models with CS420x */
@@ -331,10 +335,21 @@ static int cs_init(struct hda_codec *codec)
 	return 0;
 }
 
+static int cs_build_controls(struct hda_codec *codec)
+{
+	int err;
+
+	err = snd_hda_gen_build_controls(codec);
+	if (err < 0)
+		return err;
+	snd_hda_apply_fixup(codec, HDA_FIXUP_ACT_BUILD);
+	return 0;
+}
+
 #define cs_free		snd_hda_gen_free
 
 static const struct hda_codec_ops cs_patch_ops = {
-	.build_controls = snd_hda_gen_build_controls,
+	.build_controls = cs_build_controls,
 	.build_pcms = snd_hda_gen_build_pcms,
 	.init = cs_init,
 	.free = cs_free,
@@ -599,12 +614,14 @@ static int patch_cs420x(struct hda_codec *codec)
 enum {
 	CS4208_MAC_AUTO,
 	CS4208_MBA6,
+	CS4208_MBP11,
 	CS4208_GPIO0,
 };
 
 static const struct hda_model_fixup cs4208_models[] = {
 	{ .id = CS4208_GPIO0, .name = "gpio0" },
 	{ .id = CS4208_MBA6, .name = "mba6" },
+	{ .id = CS4208_MBP11, .name = "mbp11" },
 	{}
 };
 
@@ -615,6 +632,7 @@ static const struct snd_pci_quirk cs4208_fixup_tbl[] = {
 
 /* codec SSID matching */
 static const struct snd_pci_quirk cs4208_mac_fixup_tbl[] = {
+	SND_PCI_QUIRK(0x106b, 0x5e00, "MacBookPro 11,2", CS4208_MBP11),
 	SND_PCI_QUIRK(0x106b, 0x7100, "MacBookAir 6,1", CS4208_MBA6),
 	SND_PCI_QUIRK(0x106b, 0x7200, "MacBookAir 6,2", CS4208_MBA6),
 	{} /* terminator */
@@ -646,6 +664,36 @@ static void cs4208_fixup_mac(struct hda_codec *codec,
 	snd_hda_apply_fixup(codec, action);
 }
 
+static int cs4208_spdif_sw_put(struct snd_kcontrol *kcontrol,
+			       struct snd_ctl_elem_value *ucontrol)
+{
+	struct hda_codec *codec = snd_kcontrol_chip(kcontrol);
+	struct cs_spec *spec = codec->spec;
+	hda_nid_t pin = spec->gen.autocfg.dig_out_pins[0];
+	int pinctl = ucontrol->value.integer.value[0] ? PIN_OUT : 0;
+
+	snd_hda_set_pin_ctl_cache(codec, pin, pinctl);
+	return spec->spdif_sw_put(kcontrol, ucontrol);
+}
+
+/* hook the SPDIF switch */
+static void cs4208_fixup_spdif_switch(struct hda_codec *codec,
+				      const struct hda_fixup *fix, int action)
+{
+	if (action == HDA_FIXUP_ACT_BUILD) {
+		struct cs_spec *spec = codec->spec;
+		struct snd_kcontrol *kctl;
+
+		if (!spec->gen.autocfg.dig_out_pins[0])
+			return;
+		kctl = snd_hda_find_mixer_ctl(codec, "IEC958 Playback Switch");
+		if (!kctl)
+			return;
+		spec->spdif_sw_put = kctl->put;
+		kctl->put = cs4208_spdif_sw_put;
+	}
+}
+
 static const struct hda_fixup cs4208_fixups[] = {
 	[CS4208_MBA6] = {
 		.type = HDA_FIXUP_PINS,
@@ -653,6 +701,12 @@ static const struct hda_fixup cs4208_fixups[] = {
 		.chained = true,
 		.chain_id = CS4208_GPIO0,
 	},
+	[CS4208_MBP11] = {
+		.type = HDA_FIXUP_FUNC,
+		.v.func = cs4208_fixup_spdif_switch,
+		.chained = true,
+		.chain_id = CS4208_GPIO0,
+	},
 	[CS4208_GPIO0] = {
 		.type = HDA_FIXUP_FUNC,
 		.v.func = cs4208_fixup_gpio0,
diff --git a/sound/pci/hda/patch_hdmi.c b/sound/pci/hda/patch_hdmi.c
index e68792311bb2..08407bed093e 100644
--- a/sound/pci/hda/patch_hdmi.c
+++ b/sound/pci/hda/patch_hdmi.c
@@ -763,12 +763,12 @@ static struct channel_map_table map_tables[] = {
 	{ SNDRV_CHMAP_RC,	RC },
 	{ SNDRV_CHMAP_FLC,	FLC },
 	{ SNDRV_CHMAP_FRC,	FRC },
-	{ SNDRV_CHMAP_FLH,	FLH },
-	{ SNDRV_CHMAP_FRH,	FRH },
+	{ SNDRV_CHMAP_TFL,	FLH },
+	{ SNDRV_CHMAP_TFR,	FRH },
 	{ SNDRV_CHMAP_FLW,	FLW },
 	{ SNDRV_CHMAP_FRW,	FRW },
 	{ SNDRV_CHMAP_TC,	TC },
-	{ SNDRV_CHMAP_FCH,	FCH },
+	{ SNDRV_CHMAP_TFC,	FCH },
 	{} /* terminator */
 };
 
@@ -1247,6 +1247,9 @@ static int hdmi_pin_hbr_setup(struct hda_codec *codec, hda_nid_t pin_nid,
 		pinctl = snd_hda_codec_read(codec, pin_nid, 0,
 					    AC_VERB_GET_PIN_WIDGET_CONTROL, 0);
 
+		if (pinctl < 0)
+			return hbr ? -EINVAL : 0;
+
 		new_pinctl = pinctl & ~AC_PINCTL_EPT;
 		if (hbr)
 			new_pinctl |= AC_PINCTL_EPT_HBR;
@@ -3091,7 +3094,7 @@ static int atihdmi_pin_hbr_setup(struct hda_codec *codec, hda_nid_t pin_nid,
 	int hbr_ctl, hbr_ctl_new;
 
 	hbr_ctl = snd_hda_codec_read(codec, pin_nid, 0, ATI_VERB_GET_HBR_CONTROL, 0);
-	if (hbr_ctl & ATI_HBR_CAPABLE) {
+	if (hbr_ctl >= 0 && (hbr_ctl & ATI_HBR_CAPABLE)) {
 		if (hbr)
 			hbr_ctl_new = hbr_ctl | ATI_HBR_ENABLE;
 		else
diff --git a/sound/pci/hda/patch_realtek.c b/sound/pci/hda/patch_realtek.c
index 24d924d563aa..04d1e6be600e 100644
--- a/sound/pci/hda/patch_realtek.c
+++ b/sound/pci/hda/patch_realtek.c
@@ -3344,8 +3344,10 @@ static void alc_update_headset_mode(struct hda_codec *codec)
 	else
 		new_headset_mode = ALC_HEADSET_MODE_HEADPHONE;
 
-	if (new_headset_mode == spec->current_headset_mode)
+	if (new_headset_mode == spec->current_headset_mode) {
+		snd_hda_gen_update_outputs(codec);
 		return;
+	}
 
 	switch (new_headset_mode) {
 	case ALC_HEADSET_MODE_UNPLUGGED:
diff --git a/sound/pci/hda/patch_sigmatel.c b/sound/pci/hda/patch_sigmatel.c
index 69a549a82345..d2cc0041d9d3 100644
--- a/sound/pci/hda/patch_sigmatel.c
+++ b/sound/pci/hda/patch_sigmatel.c
@@ -100,6 +100,7 @@ enum {
 	STAC_92HD83XXX_HEADSET_JACK,
 	STAC_92HD83XXX_HP,
 	STAC_HP_ENVY_BASS,
+	STAC_HP_BNB13_EQ,
 	STAC_92HD83XXX_MODELS
 };
 
@@ -2106,6 +2107,434 @@ static void stac92hd83xxx_fixup_headset_jack(struct hda_codec *codec,
 		spec->headset_jack = 1;
 }
 
+static const struct hda_verb hp_bnb13_eq_verbs[] = {
+	/* 44.1KHz base */
+	{ 0x22, 0x7A6, 0x3E },
+	{ 0x22, 0x7A7, 0x68 },
+	{ 0x22, 0x7A8, 0x17 },
+	{ 0x22, 0x7A9, 0x3E },
+	{ 0x22, 0x7AA, 0x68 },
+	{ 0x22, 0x7AB, 0x17 },
+	{ 0x22, 0x7AC, 0x00 },
+	{ 0x22, 0x7AD, 0x80 },
+	{ 0x22, 0x7A6, 0x83 },
+	{ 0x22, 0x7A7, 0x2F },
+	{ 0x22, 0x7A8, 0xD1 },
+	{ 0x22, 0x7A9, 0x83 },
+	{ 0x22, 0x7AA, 0x2F },
+	{ 0x22, 0x7AB, 0xD1 },
+	{ 0x22, 0x7AC, 0x01 },
+	{ 0x22, 0x7AD, 0x80 },
+	{ 0x22, 0x7A6, 0x3E },
+	{ 0x22, 0x7A7, 0x68 },
+	{ 0x22, 0x7A8, 0x17 },
+	{ 0x22, 0x7A9, 0x3E },
+	{ 0x22, 0x7AA, 0x68 },
+	{ 0x22, 0x7AB, 0x17 },
+	{ 0x22, 0x7AC, 0x02 },
+	{ 0x22, 0x7AD, 0x80 },
+	{ 0x22, 0x7A6, 0x7C },
+	{ 0x22, 0x7A7, 0xC6 },
+	{ 0x22, 0x7A8, 0x0C },
+	{ 0x22, 0x7A9, 0x7C },
+	{ 0x22, 0x7AA, 0xC6 },
+	{ 0x22, 0x7AB, 0x0C },
+	{ 0x22, 0x7AC, 0x03 },
+	{ 0x22, 0x7AD, 0x80 },
+	{ 0x22, 0x7A6, 0xC3 },
+	{ 0x22, 0x7A7, 0x25 },
+	{ 0x22, 0x7A8, 0xAF },
+	{ 0x22, 0x7A9, 0xC3 },
+	{ 0x22, 0x7AA, 0x25 },
+	{ 0x22, 0x7AB, 0xAF },
+	{ 0x22, 0x7AC, 0x04 },
+	{ 0x22, 0x7AD, 0x80 },
+	{ 0x22, 0x7A6, 0x3E },
+	{ 0x22, 0x7A7, 0x85 },
+	{ 0x22, 0x7A8, 0x73 },
+	{ 0x22, 0x7A9, 0x3E },
+	{ 0x22, 0x7AA, 0x85 },
+	{ 0x22, 0x7AB, 0x73 },
+	{ 0x22, 0x7AC, 0x05 },
+	{ 0x22, 0x7AD, 0x80 },
+	{ 0x22, 0x7A6, 0x85 },
+	{ 0x22, 0x7A7, 0x39 },
+	{ 0x22, 0x7A8, 0xC7 },
+	{ 0x22, 0x7A9, 0x85 },
+	{ 0x22, 0x7AA, 0x39 },
+	{ 0x22, 0x7AB, 0xC7 },
+	{ 0x22, 0x7AC, 0x06 },
+	{ 0x22, 0x7AD, 0x80 },
+	{ 0x22, 0x7A6, 0x3C },
+	{ 0x22, 0x7A7, 0x90 },
+	{ 0x22, 0x7A8, 0xB0 },
+	{ 0x22, 0x7A9, 0x3C },
+	{ 0x22, 0x7AA, 0x90 },
+	{ 0x22, 0x7AB, 0xB0 },
+	{ 0x22, 0x7AC, 0x07 },
+	{ 0x22, 0x7AD, 0x80 },
+	{ 0x22, 0x7A6, 0x7A },
+	{ 0x22, 0x7A7, 0xC6 },
+	{ 0x22, 0x7A8, 0x39 },
+	{ 0x22, 0x7A9, 0x7A },
+	{ 0x22, 0x7AA, 0xC6 },
+	{ 0x22, 0x7AB, 0x39 },
+	{ 0x22, 0x7AC, 0x08 },
+	{ 0x22, 0x7AD, 0x80 },
+	{ 0x22, 0x7A6, 0xC4 },
+	{ 0x22, 0x7A7, 0xE9 },
+	{ 0x22, 0x7A8, 0xDC },
+	{ 0x22, 0x7A9, 0xC4 },
+	{ 0x22, 0x7AA, 0xE9 },
+	{ 0x22, 0x7AB, 0xDC },
+	{ 0x22, 0x7AC, 0x09 },
+	{ 0x22, 0x7AD, 0x80 },
+	{ 0x22, 0x7A6, 0x3D },
+	{ 0x22, 0x7A7, 0xE1 },
+	{ 0x22, 0x7A8, 0x0D },
+	{ 0x22, 0x7A9, 0x3D },
+	{ 0x22, 0x7AA, 0xE1 },
+	{ 0x22, 0x7AB, 0x0D },
+	{ 0x22, 0x7AC, 0x0A },
+	{ 0x22, 0x7AD, 0x80 },
+	{ 0x22, 0x7A6, 0x89 },
+	{ 0x22, 0x7A7, 0xB6 },
+	{ 0x22, 0x7A8, 0xEB },
+	{ 0x22, 0x7A9, 0x89 },
+	{ 0x22, 0x7AA, 0xB6 },
+	{ 0x22, 0x7AB, 0xEB },
+	{ 0x22, 0x7AC, 0x0B },
+	{ 0x22, 0x7AD, 0x80 },
+	{ 0x22, 0x7A6, 0x39 },
+	{ 0x22, 0x7A7, 0x9D },
+	{ 0x22, 0x7A8, 0xFE },
+	{ 0x22, 0x7A9, 0x39 },
+	{ 0x22, 0x7AA, 0x9D },
+	{ 0x22, 0x7AB, 0xFE },
+	{ 0x22, 0x7AC, 0x0C },
+	{ 0x22, 0x7AD, 0x80 },
+	{ 0x22, 0x7A6, 0x76 },
+	{ 0x22, 0x7A7, 0x49 },
+	{ 0x22, 0x7A8, 0x15 },
+	{ 0x22, 0x7A9, 0x76 },
+	{ 0x22, 0x7AA, 0x49 },
+	{ 0x22, 0x7AB, 0x15 },
+	{ 0x22, 0x7AC, 0x0D },
+	{ 0x22, 0x7AD, 0x80 },
+	{ 0x22, 0x7A6, 0xC8 },
+	{ 0x22, 0x7A7, 0x80 },
+	{ 0x22, 0x7A8, 0xF5 },
+	{ 0x22, 0x7A9, 0xC8 },
+	{ 0x22, 0x7AA, 0x80 },
+	{ 0x22, 0x7AB, 0xF5 },
+	{ 0x22, 0x7AC, 0x0E },
+	{ 0x22, 0x7AD, 0x80 },
+	{ 0x22, 0x7A6, 0x40 },
+	{ 0x22, 0x7A7, 0x00 },
+	{ 0x22, 0x7A8, 0x00 },
+	{ 0x22, 0x7A9, 0x40 },
+	{ 0x22, 0x7AA, 0x00 },
+	{ 0x22, 0x7AB, 0x00 },
+	{ 0x22, 0x7AC, 0x0F },
+	{ 0x22, 0x7AD, 0x80 },
+	{ 0x22, 0x7A6, 0x90 },
+	{ 0x22, 0x7A7, 0x68 },
+	{ 0x22, 0x7A8, 0xF1 },
+	{ 0x22, 0x7A9, 0x90 },
+	{ 0x22, 0x7AA, 0x68 },
+	{ 0x22, 0x7AB, 0xF1 },
+	{ 0x22, 0x7AC, 0x10 },
+	{ 0x22, 0x7AD, 0x80 },
+	{ 0x22, 0x7A6, 0x34 },
+	{ 0x22, 0x7A7, 0x47 },
+	{ 0x22, 0x7A8, 0x6C },
+	{ 0x22, 0x7A9, 0x34 },
+	{ 0x22, 0x7AA, 0x47 },
+	{ 0x22, 0x7AB, 0x6C },
+	{ 0x22, 0x7AC, 0x11 },
+	{ 0x22, 0x7AD, 0x80 },
+	{ 0x22, 0x7A6, 0x6F },
+	{ 0x22, 0x7A7, 0x97 },
+	{ 0x22, 0x7A8, 0x0F },
+	{ 0x22, 0x7A9, 0x6F },
+	{ 0x22, 0x7AA, 0x97 },
+	{ 0x22, 0x7AB, 0x0F },
+	{ 0x22, 0x7AC, 0x12 },
+	{ 0x22, 0x7AD, 0x80 },
+	{ 0x22, 0x7A6, 0xCB },
+	{ 0x22, 0x7A7, 0xB8 },
+	{ 0x22, 0x7A8, 0x94 },
+	{ 0x22, 0x7A9, 0xCB },
+	{ 0x22, 0x7AA, 0xB8 },
+	{ 0x22, 0x7AB, 0x94 },
+	{ 0x22, 0x7AC, 0x13 },
+	{ 0x22, 0x7AD, 0x80 },
+	{ 0x22, 0x7A6, 0x40 },
+	{ 0x22, 0x7A7, 0x00 },
+	{ 0x22, 0x7A8, 0x00 },
+	{ 0x22, 0x7A9, 0x40 },
+	{ 0x22, 0x7AA, 0x00 },
+	{ 0x22, 0x7AB, 0x00 },
+	{ 0x22, 0x7AC, 0x14 },
+	{ 0x22, 0x7AD, 0x80 },
+	{ 0x22, 0x7A6, 0x95 },
+	{ 0x22, 0x7A7, 0x76 },
+	{ 0x22, 0x7A8, 0x5B },
+	{ 0x22, 0x7A9, 0x95 },
+	{ 0x22, 0x7AA, 0x76 },
+	{ 0x22, 0x7AB, 0x5B },
+	{ 0x22, 0x7AC, 0x15 },
+	{ 0x22, 0x7AD, 0x80 },
+	{ 0x22, 0x7A6, 0x31 },
+	{ 0x22, 0x7A7, 0xAC },
+	{ 0x22, 0x7A8, 0x31 },
+	{ 0x22, 0x7A9, 0x31 },
+	{ 0x22, 0x7AA, 0xAC },
+	{ 0x22, 0x7AB, 0x31 },
+	{ 0x22, 0x7AC, 0x16 },
+	{ 0x22, 0x7AD, 0x80 },
+	{ 0x22, 0x7A6, 0x6A },
+	{ 0x22, 0x7A7, 0x89 },
+	{ 0x22, 0x7A8, 0xA5 },
+	{ 0x22, 0x7A9, 0x6A },
+	{ 0x22, 0x7AA, 0x89 },
+	{ 0x22, 0x7AB, 0xA5 },
+	{ 0x22, 0x7AC, 0x17 },
+	{ 0x22, 0x7AD, 0x80 },
+	{ 0x22, 0x7A6, 0xCE },
+	{ 0x22, 0x7A7, 0x53 },
+	{ 0x22, 0x7A8, 0xCF },
+	{ 0x22, 0x7A9, 0xCE },
+	{ 0x22, 0x7AA, 0x53 },
+	{ 0x22, 0x7AB, 0xCF },
+	{ 0x22, 0x7AC, 0x18 },
+	{ 0x22, 0x7AD, 0x80 },
+	{ 0x22, 0x7A6, 0x40 },
+	{ 0x22, 0x7A7, 0x00 },
+	{ 0x22, 0x7A8, 0x00 },
+	{ 0x22, 0x7A9, 0x40 },
+	{ 0x22, 0x7AA, 0x00 },
+	{ 0x22, 0x7AB, 0x00 },
+	{ 0x22, 0x7AC, 0x19 },
+	{ 0x22, 0x7AD, 0x80 },
+	/* 48KHz base */
+	{ 0x22, 0x7A6, 0x3E },
+	{ 0x22, 0x7A7, 0x88 },
+	{ 0x22, 0x7A8, 0xDC },
+	{ 0x22, 0x7A9, 0x3E },
+	{ 0x22, 0x7AA, 0x88 },
+	{ 0x22, 0x7AB, 0xDC },
+	{ 0x22, 0x7AC, 0x1A },
+	{ 0x22, 0x7AD, 0x80 },
+	{ 0x22, 0x7A6, 0x82 },
+	{ 0x22, 0x7A7, 0xEE },
+	{ 0x22, 0x7A8, 0x46 },
+	{ 0x22, 0x7A9, 0x82 },
+	{ 0x22, 0x7AA, 0xEE },
+	{ 0x22, 0x7AB, 0x46 },
+	{ 0x22, 0x7AC, 0x1B },
+	{ 0x22, 0x7AD, 0x80 },
+	{ 0x22, 0x7A6, 0x3E },
+	{ 0x22, 0x7A7, 0x88 },
+	{ 0x22, 0x7A8, 0xDC },
+	{ 0x22, 0x7A9, 0x3E },
+	{ 0x22, 0x7AA, 0x88 },
+	{ 0x22, 0x7AB, 0xDC },
+	{ 0x22, 0x7AC, 0x1C },
+	{ 0x22, 0x7AD, 0x80 },
+	{ 0x22, 0x7A6, 0x7D },
+	{ 0x22, 0x7A7, 0x09 },
+	{ 0x22, 0x7A8, 0x28 },
+	{ 0x22, 0x7A9, 0x7D },
+	{ 0x22, 0x7AA, 0x09 },
+	{ 0x22, 0x7AB, 0x28 },
+	{ 0x22, 0x7AC, 0x1D },
+	{ 0x22, 0x7AD, 0x80 },
+	{ 0x22, 0x7A6, 0xC2 },
+	{ 0x22, 0x7A7, 0xE5 },
+	{ 0x22, 0x7A8, 0xB4 },
+	{ 0x22, 0x7A9, 0xC2 },
+	{ 0x22, 0x7AA, 0xE5 },
+	{ 0x22, 0x7AB, 0xB4 },
+	{ 0x22, 0x7AC, 0x1E },
+	{ 0x22, 0x7AD, 0x80 },
+	{ 0x22, 0x7A6, 0x3E },
+	{ 0x22, 0x7A7, 0xA3 },
+	{ 0x22, 0x7A8, 0x1F },
+	{ 0x22, 0x7A9, 0x3E },
+	{ 0x22, 0x7AA, 0xA3 },
+	{ 0x22, 0x7AB, 0x1F },
+	{ 0x22, 0x7AC, 0x1F },
+	{ 0x22, 0x7AD, 0x80 },
+	{ 0x22, 0x7A6, 0x84 },
+	{ 0x22, 0x7A7, 0xCA },
+	{ 0x22, 0x7A8, 0xF1 },
+	{ 0x22, 0x7A9, 0x84 },
+	{ 0x22, 0x7AA, 0xCA },
+	{ 0x22, 0x7AB, 0xF1 },
+	{ 0x22, 0x7AC, 0x20 },
+	{ 0x22, 0x7AD, 0x80 },
+	{ 0x22, 0x7A6, 0x3C },
+	{ 0x22, 0x7A7, 0xD5 },
+	{ 0x22, 0x7A8, 0x9C },
+	{ 0x22, 0x7A9, 0x3C },
+	{ 0x22, 0x7AA, 0xD5 },
+	{ 0x22, 0x7AB, 0x9C },
+	{ 0x22, 0x7AC, 0x21 },
+	{ 0x22, 0x7AD, 0x80 },
+	{ 0x22, 0x7A6, 0x7B },
+	{ 0x22, 0x7A7, 0x35 },
+	{ 0x22, 0x7A8, 0x0F },
+	{ 0x22, 0x7A9, 0x7B },
+	{ 0x22, 0x7AA, 0x35 },
+	{ 0x22, 0x7AB, 0x0F },
+	{ 0x22, 0x7AC, 0x22 },
+	{ 0x22, 0x7AD, 0x80 },
+	{ 0x22, 0x7A6, 0xC4 },
+	{ 0x22, 0x7A7, 0x87 },
+	{ 0x22, 0x7A8, 0x45 },
+	{ 0x22, 0x7A9, 0xC4 },
+	{ 0x22, 0x7AA, 0x87 },
+	{ 0x22, 0x7AB, 0x45 },
+	{ 0x22, 0x7AC, 0x23 },
+	{ 0x22, 0x7AD, 0x80 },
+	{ 0x22, 0x7A6, 0x3E },
+	{ 0x22, 0x7A7, 0x0A },
+	{ 0x22, 0x7A8, 0x78 },
+	{ 0x22, 0x7A9, 0x3E },
+	{ 0x22, 0x7AA, 0x0A },
+	{ 0x22, 0x7AB, 0x78 },
+	{ 0x22, 0x7AC, 0x24 },
+	{ 0x22, 0x7AD, 0x80 },
+	{ 0x22, 0x7A6, 0x88 },
+	{ 0x22, 0x7A7, 0xE2 },
+	{ 0x22, 0x7A8, 0x05 },
+	{ 0x22, 0x7A9, 0x88 },
+	{ 0x22, 0x7AA, 0xE2 },
+	{ 0x22, 0x7AB, 0x05 },
+	{ 0x22, 0x7AC, 0x25 },
+	{ 0x22, 0x7AD, 0x80 },
+	{ 0x22, 0x7A6, 0x3A },
+	{ 0x22, 0x7A7, 0x1A },
+	{ 0x22, 0x7A8, 0xA3 },
+	{ 0x22, 0x7A9, 0x3A },
+	{ 0x22, 0x7AA, 0x1A },
+	{ 0x22, 0x7AB, 0xA3 },
+	{ 0x22, 0x7AC, 0x26 },
+	{ 0x22, 0x7AD, 0x80 },
+	{ 0x22, 0x7A6, 0x77 },
+	{ 0x22, 0x7A7, 0x1D },
+	{ 0x22, 0x7A8, 0xFB },
+	{ 0x22, 0x7A9, 0x77 },
+	{ 0x22, 0x7AA, 0x1D },
+	{ 0x22, 0x7AB, 0xFB },
+	{ 0x22, 0x7AC, 0x27 },
+	{ 0x22, 0x7AD, 0x80 },
+	{ 0x22, 0x7A6, 0xC7 },
+	{ 0x22, 0x7A7, 0xDA },
+	{ 0x22, 0x7A8, 0xE5 },
+	{ 0x22, 0x7A9, 0xC7 },
+	{ 0x22, 0x7AA, 0xDA },
+	{ 0x22, 0x7AB, 0xE5 },
+	{ 0x22, 0x7AC, 0x28 },
+	{ 0x22, 0x7AD, 0x80 },
+	{ 0x22, 0x7A6, 0x40 },
+	{ 0x22, 0x7A7, 0x00 },
+	{ 0x22, 0x7A8, 0x00 },
+	{ 0x22, 0x7A9, 0x40 },
+	{ 0x22, 0x7AA, 0x00 },
+	{ 0x22, 0x7AB, 0x00 },
+	{ 0x22, 0x7AC, 0x29 },
+	{ 0x22, 0x7AD, 0x80 },
+	{ 0x22, 0x7A6, 0x8E },
+	{ 0x22, 0x7A7, 0xD7 },
+	{ 0x22, 0x7A8, 0x22 },
+	{ 0x22, 0x7A9, 0x8E },
+	{ 0x22, 0x7AA, 0xD7 },
+	{ 0x22, 0x7AB, 0x22 },
+	{ 0x22, 0x7AC, 0x2A },
+	{ 0x22, 0x7AD, 0x80 },
+	{ 0x22, 0x7A6, 0x35 },
+	{ 0x22, 0x7A7, 0x26 },
+	{ 0x22, 0x7A8, 0xC6 },
+	{ 0x22, 0x7A9, 0x35 },
+	{ 0x22, 0x7AA, 0x26 },
+	{ 0x22, 0x7AB, 0xC6 },
+	{ 0x22, 0x7AC, 0x2B },
+	{ 0x22, 0x7AD, 0x80 },
+	{ 0x22, 0x7A6, 0x71 },
+	{ 0x22, 0x7A7, 0x28 },
+	{ 0x22, 0x7A8, 0xDE },
+	{ 0x22, 0x7A9, 0x71 },
+	{ 0x22, 0x7AA, 0x28 },
+	{ 0x22, 0x7AB, 0xDE },
+	{ 0x22, 0x7AC, 0x2C },
+	{ 0x22, 0x7AD, 0x80 },
+	{ 0x22, 0x7A6, 0xCA },
+	{ 0x22, 0x7A7, 0xD9 },
+	{ 0x22, 0x7A8, 0x3A },
+	{ 0x22, 0x7A9, 0xCA },
+	{ 0x22, 0x7AA, 0xD9 },
+	{ 0x22, 0x7AB, 0x3A },
+	{ 0x22, 0x7AC, 0x2D },
+	{ 0x22, 0x7AD, 0x80 },
+	{ 0x22, 0x7A6, 0x40 },
+	{ 0x22, 0x7A7, 0x00 },
+	{ 0x22, 0x7A8, 0x00 },
+	{ 0x22, 0x7A9, 0x40 },
+	{ 0x22, 0x7AA, 0x00 },
+	{ 0x22, 0x7AB, 0x00 },
+	{ 0x22, 0x7AC, 0x2E },
+	{ 0x22, 0x7AD, 0x80 },
+	{ 0x22, 0x7A6, 0x93 },
+	{ 0x22, 0x7A7, 0x5E },
+	{ 0x22, 0x7A8, 0xD8 },
+	{ 0x22, 0x7A9, 0x93 },
+	{ 0x22, 0x7AA, 0x5E },
+	{ 0x22, 0x7AB, 0xD8 },
+	{ 0x22, 0x7AC, 0x2F },
+	{ 0x22, 0x7AD, 0x80 },
+	{ 0x22, 0x7A6, 0x32 },
+	{ 0x22, 0x7A7, 0xB7 },
+	{ 0x22, 0x7A8, 0xB1 },
+	{ 0x22, 0x7A9, 0x32 },
+	{ 0x22, 0x7AA, 0xB7 },
+	{ 0x22, 0x7AB, 0xB1 },
+	{ 0x22, 0x7AC, 0x30 },
+	{ 0x22, 0x7AD, 0x80 },
+	{ 0x22, 0x7A6, 0x6C },
+	{ 0x22, 0x7A7, 0xA1 },
+	{ 0x22, 0x7A8, 0x28 },
+	{ 0x22, 0x7A9, 0x6C },
+	{ 0x22, 0x7AA, 0xA1 },
+	{ 0x22, 0x7AB, 0x28 },
+	{ 0x22, 0x7AC, 0x31 },
+	{ 0x22, 0x7AD, 0x80 },
+	{ 0x22, 0x7A6, 0xCD },
+	{ 0x22, 0x7A7, 0x48 },
+	{ 0x22, 0x7A8, 0x4F },
+	{ 0x22, 0x7A9, 0xCD },
+	{ 0x22, 0x7AA, 0x48 },
+	{ 0x22, 0x7AB, 0x4F },
+	{ 0x22, 0x7AC, 0x32 },
+	{ 0x22, 0x7AD, 0x80 },
+	{ 0x22, 0x7A6, 0x40 },
+	{ 0x22, 0x7A7, 0x00 },
+	{ 0x22, 0x7A8, 0x00 },
+	{ 0x22, 0x7A9, 0x40 },
+	{ 0x22, 0x7AA, 0x00 },
+	{ 0x22, 0x7AB, 0x00 },
+	{ 0x22, 0x7AC, 0x33 },
+	{ 0x22, 0x7AD, 0x80 },
+	/* common */
+	{ 0x22, 0x782, 0xC1 },
+	{ 0x22, 0x771, 0x2C },
+	{ 0x22, 0x772, 0x2C },
+	{ 0x22, 0x788, 0x04 },
+	{ 0x01, 0x7B0, 0x08 },
+	{}
+};
+
 static const struct hda_fixup stac92hd83xxx_fixups[] = {
 	[STAC_92HD83XXX_REF] = {
 		.type = HDA_FIXUP_PINS,
@@ -2174,6 +2603,12 @@ static const struct hda_fixup stac92hd83xxx_fixups[] = {
 			{}
 		},
 	},
+	[STAC_HP_BNB13_EQ] = {
+		.type = HDA_FIXUP_VERBS,
+		.v.verbs = hp_bnb13_eq_verbs,
+		.chained = true,
+		.chain_id = STAC_92HD83XXX_HP_MIC_LED,
+	},
 };
 
 static const struct hda_model_fixup stac92hd83xxx_models[] = {
@@ -2189,6 +2624,7 @@ static const struct hda_model_fixup stac92hd83xxx_models[] = {
 	{ .id = STAC_92HD83XXX_HP_MIC_LED, .name = "hp-mic-led" },
 	{ .id = STAC_92HD83XXX_HEADSET_JACK, .name = "headset-jack" },
 	{ .id = STAC_HP_ENVY_BASS, .name = "hp-envy-bass" },
+	{ .id = STAC_HP_BNB13_EQ, .name = "hp-bnb13-eq" },
 	{}
 };
 
@@ -2235,7 +2671,101 @@ static const struct snd_pci_quirk stac92hd83xxx_fixup_tbl[] = {
 	SND_PCI_QUIRK(PCI_VENDOR_ID_HP, 0x1888,
 			  "HP Envy Spectre", STAC_HP_ENVY_BASS),
 	SND_PCI_QUIRK(PCI_VENDOR_ID_HP, 0x18df,
-			  "HP Folio", STAC_92HD83XXX_HP_MIC_LED),
+			  "HP Folio", STAC_HP_BNB13_EQ),
+	SND_PCI_QUIRK(PCI_VENDOR_ID_HP, 0x18F8,
+			  "HP bNB13", STAC_HP_BNB13_EQ),
+	SND_PCI_QUIRK(PCI_VENDOR_ID_HP, 0x1909,
+			  "HP bNB13", STAC_HP_BNB13_EQ),
+	SND_PCI_QUIRK(PCI_VENDOR_ID_HP, 0x190A,
+			  "HP bNB13", STAC_HP_BNB13_EQ),
+	SND_PCI_QUIRK(PCI_VENDOR_ID_HP, 0x1940,
+			  "HP bNB13", STAC_HP_BNB13_EQ),
+	SND_PCI_QUIRK(PCI_VENDOR_ID_HP, 0x1941,
+			  "HP bNB13", STAC_HP_BNB13_EQ),
+	SND_PCI_QUIRK(PCI_VENDOR_ID_HP, 0x1942,
+			  "HP bNB13", STAC_HP_BNB13_EQ),
+	SND_PCI_QUIRK(PCI_VENDOR_ID_HP, 0x1943,
+			  "HP bNB13", STAC_HP_BNB13_EQ),
+	SND_PCI_QUIRK(PCI_VENDOR_ID_HP, 0x1944,
+			  "HP bNB13", STAC_HP_BNB13_EQ),
+	SND_PCI_QUIRK(PCI_VENDOR_ID_HP, 0x1945,
+			  "HP bNB13", STAC_HP_BNB13_EQ),
+	SND_PCI_QUIRK(PCI_VENDOR_ID_HP, 0x1946,
+			  "HP bNB13", STAC_HP_BNB13_EQ),
+	SND_PCI_QUIRK(PCI_VENDOR_ID_HP, 0x1948,
+			  "HP bNB13", STAC_HP_BNB13_EQ),
+	SND_PCI_QUIRK(PCI_VENDOR_ID_HP, 0x1949,
+			  "HP bNB13", STAC_HP_BNB13_EQ),
+	SND_PCI_QUIRK(PCI_VENDOR_ID_HP, 0x194A,
+			  "HP bNB13", STAC_HP_BNB13_EQ),
+	SND_PCI_QUIRK(PCI_VENDOR_ID_HP, 0x194B,
+			  "HP bNB13", STAC_HP_BNB13_EQ),
+	SND_PCI_QUIRK(PCI_VENDOR_ID_HP, 0x194C,
+			  "HP bNB13", STAC_HP_BNB13_EQ),
+	SND_PCI_QUIRK(PCI_VENDOR_ID_HP, 0x194E,
+			  "HP bNB13", STAC_HP_BNB13_EQ),
+	SND_PCI_QUIRK(PCI_VENDOR_ID_HP, 0x194F,
+			  "HP bNB13", STAC_HP_BNB13_EQ),
+	SND_PCI_QUIRK(PCI_VENDOR_ID_HP, 0x1950,
+			  "HP bNB13", STAC_HP_BNB13_EQ),
+	SND_PCI_QUIRK(PCI_VENDOR_ID_HP, 0x1951,
+			  "HP bNB13", STAC_HP_BNB13_EQ),
+	SND_PCI_QUIRK(PCI_VENDOR_ID_HP, 0x195A,
+			  "HP bNB13", STAC_HP_BNB13_EQ),
+	SND_PCI_QUIRK(PCI_VENDOR_ID_HP, 0x195B,
+			  "HP bNB13", STAC_HP_BNB13_EQ),
+	SND_PCI_QUIRK(PCI_VENDOR_ID_HP, 0x195C,
+			  "HP bNB13", STAC_HP_BNB13_EQ),
+	SND_PCI_QUIRK(PCI_VENDOR_ID_HP, 0x1991,
+			  "HP bNB13", STAC_HP_BNB13_EQ),
+	SND_PCI_QUIRK(PCI_VENDOR_ID_HP, 0x2103,
+			  "HP bNB13", STAC_HP_BNB13_EQ),
+	SND_PCI_QUIRK(PCI_VENDOR_ID_HP, 0x2104,
+			  "HP bNB13", STAC_HP_BNB13_EQ),
+	SND_PCI_QUIRK(PCI_VENDOR_ID_HP, 0x2105,
+			  "HP bNB13", STAC_HP_BNB13_EQ),
+	SND_PCI_QUIRK(PCI_VENDOR_ID_HP, 0x2106,
+			  "HP bNB13", STAC_HP_BNB13_EQ),
+	SND_PCI_QUIRK(PCI_VENDOR_ID_HP, 0x2107,
+			  "HP bNB13", STAC_HP_BNB13_EQ),
+	SND_PCI_QUIRK(PCI_VENDOR_ID_HP, 0x2108,
+			  "HP bNB13", STAC_HP_BNB13_EQ),
+	SND_PCI_QUIRK(PCI_VENDOR_ID_HP, 0x2109,
+			  "HP bNB13", STAC_HP_BNB13_EQ),
+	SND_PCI_QUIRK(PCI_VENDOR_ID_HP, 0x210A,
+			  "HP bNB13", STAC_HP_BNB13_EQ),
+	SND_PCI_QUIRK(PCI_VENDOR_ID_HP, 0x210B,
+			  "HP bNB13", STAC_HP_BNB13_EQ),
+	SND_PCI_QUIRK(PCI_VENDOR_ID_HP, 0x211C,
+			  "HP bNB13", STAC_HP_BNB13_EQ),
+	SND_PCI_QUIRK(PCI_VENDOR_ID_HP, 0x211D,
+			  "HP bNB13", STAC_HP_BNB13_EQ),
+	SND_PCI_QUIRK(PCI_VENDOR_ID_HP, 0x211E,
+			  "HP bNB13", STAC_HP_BNB13_EQ),
+	SND_PCI_QUIRK(PCI_VENDOR_ID_HP, 0x211F,
+			  "HP bNB13", STAC_HP_BNB13_EQ),
+	SND_PCI_QUIRK(PCI_VENDOR_ID_HP, 0x2120,
+			  "HP bNB13", STAC_HP_BNB13_EQ),
+	SND_PCI_QUIRK(PCI_VENDOR_ID_HP, 0x2121,
+			  "HP bNB13", STAC_HP_BNB13_EQ),
+	SND_PCI_QUIRK(PCI_VENDOR_ID_HP, 0x2122,
+			  "HP bNB13", STAC_HP_BNB13_EQ),
+	SND_PCI_QUIRK(PCI_VENDOR_ID_HP, 0x2123,
+			  "HP bNB13", STAC_HP_BNB13_EQ),
+	SND_PCI_QUIRK(PCI_VENDOR_ID_HP, 0x213E,
+			  "HP bNB13", STAC_HP_BNB13_EQ),
+	SND_PCI_QUIRK(PCI_VENDOR_ID_HP, 0x213F,
+			  "HP bNB13", STAC_HP_BNB13_EQ),
+	SND_PCI_QUIRK(PCI_VENDOR_ID_HP, 0x2140,
+			  "HP bNB13", STAC_HP_BNB13_EQ),
+	SND_PCI_QUIRK(PCI_VENDOR_ID_HP, 0x21B2,
+			  "HP bNB13", STAC_HP_BNB13_EQ),
+	SND_PCI_QUIRK(PCI_VENDOR_ID_HP, 0x21B3,
+			  "HP bNB13", STAC_HP_BNB13_EQ),
+	SND_PCI_QUIRK(PCI_VENDOR_ID_HP, 0x21B5,
+			  "HP bNB13", STAC_HP_BNB13_EQ),
+	SND_PCI_QUIRK(PCI_VENDOR_ID_HP, 0x21B6,
+			  "HP bNB13", STAC_HP_BNB13_EQ),
 	SND_PCI_QUIRK_MASK(PCI_VENDOR_ID_HP, 0xff00, 0x1900,
 			  "HP", STAC_92HD83XXX_HP_MIC_LED),
 	SND_PCI_QUIRK_MASK(PCI_VENDOR_ID_HP, 0xff00, 0x2000,
diff --git a/sound/ppc/keywest.c b/sound/ppc/keywest.c
index 01aecc2b5073..0d1c27e911b8 100644
--- a/sound/ppc/keywest.c
+++ b/sound/ppc/keywest.c
@@ -65,7 +65,7 @@ static int keywest_attach_adapter(struct i2c_adapter *adapter)
 	 * already bound. If not it means binding failed, and then there
 	 * is no point in keeping the device instantiated.
 	 */
-	if (!keywest_ctx->client->driver) {
+	if (!keywest_ctx->client->dev.driver) {
 		i2c_unregister_device(keywest_ctx->client);
 		keywest_ctx->client = NULL;
 		return -ENODEV;
@@ -76,7 +76,7 @@ static int keywest_attach_adapter(struct i2c_adapter *adapter)
 	 * This is safe because i2c-core holds the core_lock mutex for us.
 	 */
 	list_add_tail(&keywest_ctx->client->detected,
-		      &keywest_ctx->client->driver->clients);
+		      &to_i2c_driver(keywest_ctx->client->dev.driver)->clients);
 	return 0;
 }
 
diff --git a/sound/soc/blackfin/bf5xx-i2s.c b/sound/soc/blackfin/bf5xx-i2s.c
index 9a174fc47d39..39d774839b3e 100644
--- a/sound/soc/blackfin/bf5xx-i2s.c
+++ b/sound/soc/blackfin/bf5xx-i2s.c
@@ -121,6 +121,7 @@ static int bf5xx_i2s_hw_params(struct snd_pcm_substream *substream,
 		bf5xx_i2s->tcr2 |= 7;
 		bf5xx_i2s->rcr2 |= 7;
 		sport_handle->wdsize = 1;
+		break;
 	case SNDRV_PCM_FORMAT_S16_LE:
 		bf5xx_i2s->tcr2 |= 15;
 		bf5xx_i2s->rcr2 |= 15;
diff --git a/sound/soc/codecs/alc5632.c b/sound/soc/codecs/alc5632.c
index f2e62e45f912..19e9f222d09c 100644
--- a/sound/soc/codecs/alc5632.c
+++ b/sound/soc/codecs/alc5632.c
@@ -614,7 +614,7 @@ struct _pll_div {
 };
 
 /* Note : pll code from original alc5632 driver. Not sure of how good it is */
-/* usefull only for master mode */
+/* useful only for master mode */
 static const struct _pll_div codec_master_pll_div[] = {
 
 	{  2048000,  8192000,	0x0ea0},
diff --git a/sound/soc/codecs/cs42l52.h b/sound/soc/codecs/cs42l52.h
index 1a9412d86d17..6fb8f00f4191 100644
--- a/sound/soc/codecs/cs42l52.h
+++ b/sound/soc/codecs/cs42l52.h
@@ -179,7 +179,7 @@
 #define CS42L52_MICB_CTL			0x11
 #define	CS42L52_MIC_CTL_MIC_SEL_MASK		0xBF
 #define	CS42L52_MIC_CTL_MIC_SEL_SHIFT		6
-#define CS42L52_MIC_CTL_TYPE_MASK		0xDF
+#define CS42L52_MIC_CTL_TYPE_MASK		0x20
 #define CS42L52_MIC_CTL_TYPE_SHIFT		5
 
 
diff --git a/sound/soc/codecs/wm5102.c b/sound/soc/codecs/wm5102.c
index 8bbddc151aa8..a08e8bf6d07c 100644
--- a/sound/soc/codecs/wm5102.c
+++ b/sound/soc/codecs/wm5102.c
@@ -685,13 +685,13 @@ ARIZONA_MIXER_CONTROLS("EQ2", ARIZONA_EQ2MIX_INPUT_1_SOURCE),
 ARIZONA_MIXER_CONTROLS("EQ3", ARIZONA_EQ3MIX_INPUT_1_SOURCE),
 ARIZONA_MIXER_CONTROLS("EQ4", ARIZONA_EQ4MIX_INPUT_1_SOURCE),
 
-SND_SOC_BYTES_MASK("EQ1 Coefficeints", ARIZONA_EQ1_1, 21,
+SND_SOC_BYTES_MASK("EQ1 Coefficients", ARIZONA_EQ1_1, 21,
 		   ARIZONA_EQ1_ENA_MASK),
-SND_SOC_BYTES_MASK("EQ2 Coefficeints", ARIZONA_EQ2_1, 21,
+SND_SOC_BYTES_MASK("EQ2 Coefficients", ARIZONA_EQ2_1, 21,
 		   ARIZONA_EQ2_ENA_MASK),
-SND_SOC_BYTES_MASK("EQ3 Coefficeints", ARIZONA_EQ3_1, 21,
+SND_SOC_BYTES_MASK("EQ3 Coefficients", ARIZONA_EQ3_1, 21,
 		   ARIZONA_EQ3_ENA_MASK),
-SND_SOC_BYTES_MASK("EQ4 Coefficeints", ARIZONA_EQ4_1, 21,
+SND_SOC_BYTES_MASK("EQ4 Coefficients", ARIZONA_EQ4_1, 21,
 		   ARIZONA_EQ4_ENA_MASK),
 
 SOC_SINGLE_TLV("EQ1 B1 Volume", ARIZONA_EQ1_1, ARIZONA_EQ1_B1_GAIN_SHIFT,
diff --git a/sound/soc/codecs/wm5110.c b/sound/soc/codecs/wm5110.c
index 8c91be5d67e3..f2d1094424b9 100644
--- a/sound/soc/codecs/wm5110.c
+++ b/sound/soc/codecs/wm5110.c
@@ -101,13 +101,13 @@ ARIZONA_MIXER_CONTROLS("EQ2", ARIZONA_EQ2MIX_INPUT_1_SOURCE),
 ARIZONA_MIXER_CONTROLS("EQ3", ARIZONA_EQ3MIX_INPUT_1_SOURCE),
 ARIZONA_MIXER_CONTROLS("EQ4", ARIZONA_EQ4MIX_INPUT_1_SOURCE),
 
-SND_SOC_BYTES_MASK("EQ1 Coefficeints", ARIZONA_EQ1_1, 21,
+SND_SOC_BYTES_MASK("EQ1 Coefficients", ARIZONA_EQ1_1, 21,
 		   ARIZONA_EQ1_ENA_MASK),
-SND_SOC_BYTES_MASK("EQ2 Coefficeints", ARIZONA_EQ2_1, 21,
+SND_SOC_BYTES_MASK("EQ2 Coefficients", ARIZONA_EQ2_1, 21,
 		   ARIZONA_EQ2_ENA_MASK),
-SND_SOC_BYTES_MASK("EQ3 Coefficeints", ARIZONA_EQ3_1, 21,
+SND_SOC_BYTES_MASK("EQ3 Coefficients", ARIZONA_EQ3_1, 21,
 		   ARIZONA_EQ3_ENA_MASK),
-SND_SOC_BYTES_MASK("EQ4 Coefficeints", ARIZONA_EQ4_1, 21,
+SND_SOC_BYTES_MASK("EQ4 Coefficients", ARIZONA_EQ4_1, 21,
 		   ARIZONA_EQ4_ENA_MASK),
 
 SOC_SINGLE_TLV("EQ1 B1 Volume", ARIZONA_EQ1_1, ARIZONA_EQ1_B1_GAIN_SHIFT,
diff --git a/sound/soc/codecs/wm8962.c b/sound/soc/codecs/wm8962.c
index ac1ff9947a90..543c5c2631b6 100644
--- a/sound/soc/codecs/wm8962.c
+++ b/sound/soc/codecs/wm8962.c
@@ -3728,6 +3728,8 @@ static int wm8962_i2c_probe(struct i2c_client *i2c,
 	if (ret < 0)
 		goto err_enable;
 
+	regcache_cache_only(wm8962->regmap, true);
+
 	/* The drivers should power up as needed */
 	regulator_bulk_disable(ARRAY_SIZE(wm8962->supplies), wm8962->supplies);
 
diff --git a/sound/soc/codecs/wm8997.c b/sound/soc/codecs/wm8997.c
index 6ec3de3efa4f..1392bb3c9254 100644
--- a/sound/soc/codecs/wm8997.c
+++ b/sound/soc/codecs/wm8997.c
@@ -170,13 +170,13 @@ ARIZONA_MIXER_CONTROLS("EQ2", ARIZONA_EQ2MIX_INPUT_1_SOURCE),
 ARIZONA_MIXER_CONTROLS("EQ3", ARIZONA_EQ3MIX_INPUT_1_SOURCE),
 ARIZONA_MIXER_CONTROLS("EQ4", ARIZONA_EQ4MIX_INPUT_1_SOURCE),
 
-SND_SOC_BYTES_MASK("EQ1 Coefficeints", ARIZONA_EQ1_1, 21,
+SND_SOC_BYTES_MASK("EQ1 Coefficients", ARIZONA_EQ1_1, 21,
 		   ARIZONA_EQ1_ENA_MASK),
-SND_SOC_BYTES_MASK("EQ2 Coefficeints", ARIZONA_EQ2_1, 21,
+SND_SOC_BYTES_MASK("EQ2 Coefficients", ARIZONA_EQ2_1, 21,
 		   ARIZONA_EQ2_ENA_MASK),
-SND_SOC_BYTES_MASK("EQ3 Coefficeints", ARIZONA_EQ3_1, 21,
+SND_SOC_BYTES_MASK("EQ3 Coefficients", ARIZONA_EQ3_1, 21,
 		   ARIZONA_EQ3_ENA_MASK),
-SND_SOC_BYTES_MASK("EQ4 Coefficeints", ARIZONA_EQ4_1, 21,
+SND_SOC_BYTES_MASK("EQ4 Coefficients", ARIZONA_EQ4_1, 21,
 		   ARIZONA_EQ4_ENA_MASK),
 
 SOC_SINGLE_TLV("EQ1 B1 Volume", ARIZONA_EQ1_1, ARIZONA_EQ1_B1_GAIN_SHIFT,
@@ -887,7 +887,7 @@ static const struct snd_soc_dapm_route wm8997_dapm_routes[] = {
 	ARIZONA_MIXER_ROUTES("Mic Mute Mixer", "Mic"),
 
 	ARIZONA_MUX_ROUTES("ISRC1INT1", "ISRC1INT1"),
-	ARIZONA_MUX_ROUTES("ISRC1INT2", "ISRC2INT2"),
+	ARIZONA_MUX_ROUTES("ISRC1INT2", "ISRC1INT2"),
 
 	ARIZONA_MUX_ROUTES("ISRC1DEC1", "ISRC1DEC1"),
 	ARIZONA_MUX_ROUTES("ISRC1DEC2", "ISRC1DEC2"),
diff --git a/sound/soc/davinci/davinci-pcm.c b/sound/soc/davinci/davinci-pcm.c
index fa64cd85204f..fb5d107f5603 100644
--- a/sound/soc/davinci/davinci-pcm.c
+++ b/sound/soc/davinci/davinci-pcm.c
@@ -238,7 +238,7 @@ static void davinci_pcm_dma_irq(unsigned link, u16 ch_status, void *data)
 	print_buf_info(prtd->ram_channel, "i ram_channel");
 	pr_debug("davinci_pcm: link=%d, status=0x%x\n", link, ch_status);
 
-	if (unlikely(ch_status != DMA_COMPLETE))
+	if (unlikely(ch_status != EDMA_DMA_COMPLETE))
 		return;
 
 	if (snd_pcm_running(substream)) {
diff --git a/sound/soc/fsl/imx-pcm-fiq.c b/sound/soc/fsl/imx-pcm-fiq.c
index 41740e488820..c75d43bb2e92 100644
--- a/sound/soc/fsl/imx-pcm-fiq.c
+++ b/sound/soc/fsl/imx-pcm-fiq.c
@@ -42,7 +42,8 @@ struct imx_pcm_runtime_data {
 	struct hrtimer hrt;
 	int poll_time_ns;
 	struct snd_pcm_substream *substream;
-	atomic_t running;
+	atomic_t playing;
+	atomic_t capturing;
 };
 
 static enum hrtimer_restart snd_hrtimer_callback(struct hrtimer *hrt)
@@ -52,7 +53,7 @@ static enum hrtimer_restart snd_hrtimer_callback(struct hrtimer *hrt)
 	struct snd_pcm_substream *substream = iprtd->substream;
 	struct pt_regs regs;
 
-	if (!atomic_read(&iprtd->running))
+	if (!atomic_read(&iprtd->playing) && !atomic_read(&iprtd->capturing))
 		return HRTIMER_NORESTART;
 
 	get_fiq_regs(&regs);
@@ -106,7 +107,6 @@ static int snd_imx_pcm_prepare(struct snd_pcm_substream *substream)
 	return 0;
 }
 
-static int fiq_enable;
 static int imx_pcm_fiq;
 
 static int snd_imx_pcm_trigger(struct snd_pcm_substream *substream, int cmd)
@@ -118,23 +118,27 @@ static int snd_imx_pcm_trigger(struct snd_pcm_substream *substream, int cmd)
 	case SNDRV_PCM_TRIGGER_START:
 	case SNDRV_PCM_TRIGGER_RESUME:
 	case SNDRV_PCM_TRIGGER_PAUSE_RELEASE:
-		atomic_set(&iprtd->running, 1);
+		if (substream->stream == SNDRV_PCM_STREAM_PLAYBACK)
+			atomic_set(&iprtd->playing, 1);
+		else
+			atomic_set(&iprtd->capturing, 1);
 		hrtimer_start(&iprtd->hrt, ns_to_ktime(iprtd->poll_time_ns),
 		      HRTIMER_MODE_REL);
-		if (++fiq_enable == 1)
-			enable_fiq(imx_pcm_fiq);
-
+		enable_fiq(imx_pcm_fiq);
 		break;
 
 	case SNDRV_PCM_TRIGGER_STOP:
 	case SNDRV_PCM_TRIGGER_SUSPEND:
 	case SNDRV_PCM_TRIGGER_PAUSE_PUSH:
-		atomic_set(&iprtd->running, 0);
-
-		if (--fiq_enable == 0)
+		if (substream->stream == SNDRV_PCM_STREAM_PLAYBACK)
+			atomic_set(&iprtd->playing, 0);
+		else
+			atomic_set(&iprtd->capturing, 0);
+		if (!atomic_read(&iprtd->playing) &&
+				!atomic_read(&iprtd->capturing))
 			disable_fiq(imx_pcm_fiq);
-
 		break;
+
 	default:
 		return -EINVAL;
 	}
@@ -182,7 +186,8 @@ static int snd_imx_open(struct snd_pcm_substream *substream)
 
 	iprtd->substream = substream;
 
-	atomic_set(&iprtd->running, 0);
+	atomic_set(&iprtd->playing, 0);
+	atomic_set(&iprtd->capturing, 0);
 	hrtimer_init(&iprtd->hrt, CLOCK_MONOTONIC, HRTIMER_MODE_REL);
 	iprtd->hrt.function = snd_hrtimer_callback;
 
diff --git a/sound/soc/fsl/imx-wm8962.c b/sound/soc/fsl/imx-wm8962.c
index 361f94f09b11..61e48852b9e8 100644
--- a/sound/soc/fsl/imx-wm8962.c
+++ b/sound/soc/fsl/imx-wm8962.c
@@ -215,7 +215,7 @@ static int imx_wm8962_probe(struct platform_device *pdev)
 		goto fail;
 	}
 	codec_dev = of_find_i2c_device_by_node(codec_np);
-	if (!codec_dev || !codec_dev->driver) {
+	if (!codec_dev || !codec_dev->dev.driver) {
 		dev_err(&pdev->dev, "failed to find codec platform device\n");
 		ret = -EINVAL;
 		goto fail;
diff --git a/sound/soc/samsung/ac97.c b/sound/soc/samsung/ac97.c
index 2acf987844e8..350ba23a9893 100644
--- a/sound/soc/samsung/ac97.c
+++ b/sound/soc/samsung/ac97.c
@@ -74,7 +74,7 @@ static void s3c_ac97_activate(struct snd_ac97 *ac97)
 	if (stat == S3C_AC97_GLBSTAT_MAINSTATE_ACTIVE)
 		return; /* Return if already active */
 
-	INIT_COMPLETION(s3c_ac97.done);
+	reinit_completion(&s3c_ac97.done);
 
 	ac_glbctrl = readl(s3c_ac97.regs + S3C_AC97_GLBCTRL);
 	ac_glbctrl = S3C_AC97_GLBCTRL_ACLINKON;
@@ -103,7 +103,7 @@ static unsigned short s3c_ac97_read(struct snd_ac97 *ac97,
 
 	s3c_ac97_activate(ac97);
 
-	INIT_COMPLETION(s3c_ac97.done);
+	reinit_completion(&s3c_ac97.done);
 
 	ac_codec_cmd = readl(s3c_ac97.regs + S3C_AC97_CODEC_CMD);
 	ac_codec_cmd = S3C_AC97_CODEC_CMD_READ | AC_CMD_ADDR(reg);
@@ -140,7 +140,7 @@ static void s3c_ac97_write(struct snd_ac97 *ac97, unsigned short reg,
 
 	s3c_ac97_activate(ac97);
 
-	INIT_COMPLETION(s3c_ac97.done);
+	reinit_completion(&s3c_ac97.done);
 
 	ac_codec_cmd = readl(s3c_ac97.regs + S3C_AC97_CODEC_CMD);
 	ac_codec_cmd = AC_CMD_ADDR(reg) | AC_CMD_DATA(val);
diff --git a/sound/sparc/cs4231.c b/sound/sparc/cs4231.c
index b47f6fe6277f..dbb1b625eb2f 100644
--- a/sound/sparc/cs4231.c
+++ b/sound/sparc/cs4231.c
@@ -907,19 +907,24 @@ static int snd_cs4231_playback_prepare(struct snd_pcm_substream *substream)
 	struct snd_cs4231 *chip = snd_pcm_substream_chip(substream);
 	struct snd_pcm_runtime *runtime = substream->runtime;
 	unsigned long flags;
+	int ret = 0;
 
 	spin_lock_irqsave(&chip->lock, flags);
 
 	chip->image[CS4231_IFACE_CTRL] &= ~(CS4231_PLAYBACK_ENABLE |
 					    CS4231_PLAYBACK_PIO);
 
-	if (WARN_ON(runtime->period_size > 0xffff + 1))
-		return -EINVAL;
+	if (WARN_ON(runtime->period_size > 0xffff + 1)) {
+		ret = -EINVAL;
+		goto out;
+	}
 
 	chip->p_periods_sent = 0;
+
+out:
 	spin_unlock_irqrestore(&chip->lock, flags);
 
-	return 0;
+	return ret;
 }
 
 static int snd_cs4231_capture_hw_params(struct snd_pcm_substream *substream,
diff --git a/sound/usb/mixer_quirks.c b/sound/usb/mixer_quirks.c
index d42a584cf829..3454262358b3 100644
--- a/sound/usb/mixer_quirks.c
+++ b/sound/usb/mixer_quirks.c
@@ -433,6 +433,89 @@ static void snd_audigy2nx_proc_read(struct snd_info_entry *entry,
 	}
 }
 
+/* EMU0204 */
+static int snd_emu0204_ch_switch_info(struct snd_kcontrol *kcontrol,
+				      struct snd_ctl_elem_info *uinfo)
+{
+	static const char *texts[2] = {"1/2",
+				       "3/4"
+	};
+
+	uinfo->type = SNDRV_CTL_ELEM_TYPE_ENUMERATED;
+	uinfo->count = 1;
+	uinfo->value.enumerated.items = 2;
+	if (uinfo->value.enumerated.item > 1)
+		uinfo->value.enumerated.item = 1;
+	strcpy(uinfo->value.enumerated.name,
+		texts[uinfo->value.enumerated.item]);
+
+	return 0;
+}
+
+static int snd_emu0204_ch_switch_get(struct snd_kcontrol *kcontrol,
+				     struct snd_ctl_elem_value *ucontrol)
+{
+	ucontrol->value.enumerated.item[0] = kcontrol->private_value;
+	return 0;
+}
+
+static int snd_emu0204_ch_switch_put(struct snd_kcontrol *kcontrol,
+				     struct snd_ctl_elem_value *ucontrol)
+{
+	struct usb_mixer_interface *mixer = snd_kcontrol_chip(kcontrol);
+	unsigned int value = ucontrol->value.enumerated.item[0];
+	int err, changed;
+	unsigned char buf[2];
+
+	if (value > 1)
+		return -EINVAL;
+
+	buf[0] = 0x01;
+	buf[1] = value ? 0x02 : 0x01;
+
+	changed = value != kcontrol->private_value;
+	down_read(&mixer->chip->shutdown_rwsem);
+	if (mixer->chip->shutdown) {
+		err = -ENODEV;
+		goto out;
+	}
+	err = snd_usb_ctl_msg(mixer->chip->dev,
+		      usb_sndctrlpipe(mixer->chip->dev, 0), UAC_SET_CUR,
+		      USB_RECIP_INTERFACE | USB_TYPE_CLASS | USB_DIR_OUT,
+		      0x0400, 0x0e00, buf, 2);
+ out:
+	up_read(&mixer->chip->shutdown_rwsem);
+	if (err < 0)
+		return err;
+	kcontrol->private_value = value;
+	return changed;
+}
+
+
+static struct snd_kcontrol_new snd_emu0204_controls[] = {
+	{
+		.iface = SNDRV_CTL_ELEM_IFACE_MIXER,
+		.name = "Front Jack Channels",
+		.info = snd_emu0204_ch_switch_info,
+		.get = snd_emu0204_ch_switch_get,
+		.put = snd_emu0204_ch_switch_put,
+		.private_value = 0,
+	},
+};
+
+static int snd_emu0204_controls_create(struct usb_mixer_interface *mixer)
+{
+	int i, err;
+
+	for (i = 0; i < ARRAY_SIZE(snd_emu0204_controls); ++i) {
+		err = snd_ctl_add(mixer->chip->card,
+			snd_ctl_new1(&snd_emu0204_controls[i], mixer));
+		if (err < 0)
+			return err;
+	}
+
+	return 0;
+}
 /* ASUS Xonar U1 / U3 controls */
 
 static int snd_xonar_u1_switch_get(struct snd_kcontrol *kcontrol,
@@ -1545,6 +1628,13 @@ int snd_usb_mixer_apply_create_quirk(struct usb_mixer_interface *mixer)
 					      snd_audigy2nx_proc_read);
 		break;
 
+	/* EMU0204 */
+	case USB_ID(0x041e, 0x3f19):
+		err = snd_emu0204_controls_create(mixer);
+		if (err < 0)
+			break;
+		break;
+
 	case USB_ID(0x0763, 0x2030): /* M-Audio Fast Track C400 */
 	case USB_ID(0x0763, 0x2031): /* M-Audio Fast Track C400 */
 		err = snd_c400_create_mixer(mixer);
diff --git a/sound/usb/stream.c b/sound/usb/stream.c
index d737d0e6e558..2fb71be5e100 100644
--- a/sound/usb/stream.c
+++ b/sound/usb/stream.c
@@ -273,8 +273,8 @@ static struct snd_pcm_chmap_elem *convert_chmap(int channels, unsigned int bits,
 		SNDRV_CHMAP_TSL,	/* top side left */
 		SNDRV_CHMAP_TSR,	/* top side right */
 		SNDRV_CHMAP_BC,		/* bottom center */
-		SNDRV_CHMAP_BLC,	/* bottom left center */
-		SNDRV_CHMAP_BRC,	/* bottom right center */
+		SNDRV_CHMAP_RLC,	/* back left of center */
+		SNDRV_CHMAP_RRC,	/* back right of center */
 		0 /* terminator */
 	};
 	struct snd_pcm_chmap_elem *chmap;
diff --git a/tools/Makefile b/tools/Makefile
index 41067f304215..a9b02008443c 100644
--- a/tools/Makefile
+++ b/tools/Makefile
@@ -15,6 +15,7 @@ help:
 	@echo '  net        - misc networking tools'
 	@echo '  vm         - misc vm tools'
 	@echo '  x86_energy_perf_policy - Intel energy policy tool'
+	@echo '  tmon       - thermal monitoring and tuning tool'
 	@echo ''
 	@echo 'You can do:'
 	@echo ' $$ make -C tools/ <tool>_install'
@@ -50,6 +51,9 @@ selftests: FORCE
 turbostat x86_energy_perf_policy: FORCE
 	$(call descend,power/x86/$@)
 
+tmon: FORCE
+	$(call descend,thermal/$@)
+
 cpupower_install:
 	$(call descend,power/$(@:_install=),install)
 
@@ -62,9 +66,13 @@ selftests_install:
 turbostat_install x86_energy_perf_policy_install:
 	$(call descend,power/x86/$(@:_install=),install)
 
+tmon_install:
+	$(call descend,thermal/$(@:_install=),install)
+
 install: cgroup_install cpupower_install firewire_install lguest_install \
 		perf_install selftests_install turbostat_install usb_install \
-		virtio_install vm_install net_install x86_energy_perf_policy_install
+		virtio_install vm_install net_install x86_energy_perf_policy_install \
+	tmon
 
 cpupower_clean:
 	$(call descend,power/cpupower,clean)
@@ -84,8 +92,11 @@ selftests_clean:
 turbostat_clean x86_energy_perf_policy_clean:
 	$(call descend,power/x86/$(@:_clean=),clean)
 
+tmon_clean:
+	$(call descend,thermal/tmon,clean)
+
 clean: cgroup_clean cpupower_clean firewire_clean lguest_clean perf_clean \
 		selftests_clean turbostat_clean usb_clean virtio_clean \
-		vm_clean net_clean x86_energy_perf_policy_clean
+		vm_clean net_clean x86_energy_perf_policy_clean tmon_clean
 
 .PHONY: FORCE
diff --git a/tools/lib/traceevent/event-parse.c b/tools/lib/traceevent/event-parse.c
index 8f450adaa9c2..0362d575de7d 100644
--- a/tools/lib/traceevent/event-parse.c
+++ b/tools/lib/traceevent/event-parse.c
@@ -3435,6 +3435,19 @@ eval_num_arg(void *data, int size, struct event_format *event, struct print_arg
 			goto out_warning_op;
 		}
 		break;
+	case PRINT_DYNAMIC_ARRAY:
+		/* Without [], we pass the address to the dynamic data */
+		offset = pevent_read_number(pevent,
+					    data + arg->dynarray.field->offset,
+					    arg->dynarray.field->size);
+		/*
+		 * The actual length of the dynamic array is stored
+		 * in the top half of the field, and the offset
+		 * is in the bottom half of the 32 bit field.
+		 */
+		offset &= 0xffff;
+		val = (unsigned long long)(data + offset);
+		break;
 	default: /* not sure what to do there */
 		return 0;
 	}
diff --git a/tools/perf/Documentation/perf-record.txt b/tools/perf/Documentation/perf-record.txt
index 052f7c4dc00c..43b42c4f4a91 100644
--- a/tools/perf/Documentation/perf-record.txt
+++ b/tools/perf/Documentation/perf-record.txt
@@ -201,6 +201,12 @@ abort events and some memory events in precise mode on modern Intel CPUs.
 --transaction::
 Record transaction flags for transaction related events.
 
+--force-per-cpu::
+Force the use of per-cpu mmaps.  By default, when tasks are specified (i.e. -p,
+-t or -u options) per-thread mmaps are created.  This option overrides that and
+forces per-cpu mmaps.  A side-effect of that is that inheritance is
+automatically enabled.  Add the -i option also to disable inheritance.
+
 SEE ALSO
 --------
 linkperf:perf-stat[1], linkperf:perf-list[1]
diff --git a/tools/perf/Documentation/perf-trace.txt b/tools/perf/Documentation/perf-trace.txt
index 7b0497f95a75..fae38d9a44a4 100644
--- a/tools/perf/Documentation/perf-trace.txt
+++ b/tools/perf/Documentation/perf-trace.txt
@@ -93,9 +93,15 @@ the thread executes on the designated CPUs. Default is to monitor all CPUs.
 --comm::
         Show process COMM right beside its ID, on by default, disable with --no-comm.
 
+-s::
 --summary::
-	Show a summary of syscalls by thread with min, max, and average times (in
-    msec) and relative stddev.
+	Show only a summary of syscalls by thread with min, max, and average times
+    (in msec) and relative stddev.
+
+-S::
+--with-summary::
+	Show all syscalls followed by a summary by thread with min, max, and
+    average times (in msec) and relative stddev.
 
 --tool_stats::
 	Show tool stats such as number of times fd->pathname was discovered thru
diff --git a/tools/perf/builtin-kvm.c b/tools/perf/builtin-kvm.c
index cd9f92078aba..f8bf5f244d77 100644
--- a/tools/perf/builtin-kvm.c
+++ b/tools/perf/builtin-kvm.c
@@ -1510,13 +1510,13 @@ static int kvm_events_live(struct perf_kvm_stat *kvm,
 	/*
 	 * target related setups
 	 */
-	err = perf_target__validate(&kvm->opts.target);
+	err = target__validate(&kvm->opts.target);
 	if (err) {
-		perf_target__strerror(&kvm->opts.target, err, errbuf, BUFSIZ);
+		target__strerror(&kvm->opts.target, err, errbuf, BUFSIZ);
 		ui__warning("%s", errbuf);
 	}
 
-	if (perf_target__none(&kvm->opts.target))
+	if (target__none(&kvm->opts.target))
 		kvm->opts.target.system_wide = true;
 
 
@@ -1544,18 +1544,8 @@ static int kvm_events_live(struct perf_kvm_stat *kvm,
 	}
 	kvm->session->evlist = kvm->evlist;
 	perf_session__set_id_hdr_size(kvm->session);
-
-
-	if (perf_target__has_task(&kvm->opts.target))
-		perf_event__synthesize_thread_map(&kvm->tool,
-						  kvm->evlist->threads,
-						  perf_event__process,
-						  &kvm->session->machines.host);
-	else
-		perf_event__synthesize_threads(&kvm->tool, perf_event__process,
-					       &kvm->session->machines.host);
-
-
+	machine__synthesize_threads(&kvm->session->machines.host, &kvm->opts.target,
+				    kvm->evlist->threads, false);
 	err = kvm_live_open_events(kvm);
 	if (err)
 		goto out;
diff --git a/tools/perf/builtin-probe.c b/tools/perf/builtin-probe.c
index 89acc17cf2a0..6ea9e85bdc00 100644
--- a/tools/perf/builtin-probe.c
+++ b/tools/perf/builtin-probe.c
@@ -325,6 +325,8 @@ int cmd_probe(int argc, const char **argv, const char *prefix __maybe_unused)
 		     opt_set_filter),
 	OPT_CALLBACK('x', "exec", NULL, "executable|path",
 			"target executable name or path", opt_set_target),
+	OPT_BOOLEAN(0, "demangle", &symbol_conf.demangle,
+		    "Disable symbol demangling"),
 	OPT_END()
 	};
 	int ret;
diff --git a/tools/perf/builtin-record.c b/tools/perf/builtin-record.c
index 15280b5e5574..7c8020a32784 100644
--- a/tools/perf/builtin-record.c
+++ b/tools/perf/builtin-record.c
@@ -76,12 +76,12 @@ struct perf_record {
 	long			samples;
 };
 
-static int write_output(struct perf_record *rec, void *buf, size_t size)
+static int do_write_output(struct perf_record *rec, void *buf, size_t size)
 {
 	struct perf_data_file *file = &rec->file;
 
 	while (size) {
-		int ret = write(file->fd, buf, size);
+		ssize_t ret = write(file->fd, buf, size);
 
 		if (ret < 0) {
 			pr_err("failed to write perf data, error: %m\n");
@@ -97,6 +97,11 @@ static int write_output(struct perf_record *rec, void *buf, size_t size)
 	return 0;
 }
 
+static int write_output(struct perf_record *rec, void *buf, size_t size)
+{
+	return do_write_output(rec, buf, size);
+}
+
 static int process_synthesized_event(struct perf_tool *tool,
 				     union perf_event *event,
 				     struct perf_sample *sample __maybe_unused,
@@ -480,16 +485,8 @@ static int __cmd_record(struct perf_record *rec, int argc, const char **argv)
 					 perf_event__synthesize_guest_os, tool);
 	}
 
-	if (perf_target__has_task(&opts->target))
-		err = perf_event__synthesize_thread_map(tool, evsel_list->threads,
-						  process_synthesized_event,
-						  machine);
-	else if (perf_target__has_cpu(&opts->target))
-		err = perf_event__synthesize_threads(tool, process_synthesized_event,
-					       machine);
-	else /* command specified */
-		err = 0;
-
+	err = __machine__synthesize_threads(machine, tool, &opts->target, evsel_list->threads,
+					    process_synthesized_event, opts->sample_address);
 	if (err != 0)
 		goto out_delete_session;
 
@@ -509,7 +506,7 @@ static int __cmd_record(struct perf_record *rec, int argc, const char **argv)
 	 * (apart from group members) have enable_on_exec=1 set,
 	 * so don't spoil it by prematurely enabling them.
 	 */
-	if (!perf_target__none(&opts->target))
+	if (!target__none(&opts->target))
 		perf_evlist__enable(evsel_list);
 
 	/*
@@ -538,7 +535,7 @@ static int __cmd_record(struct perf_record *rec, int argc, const char **argv)
 		 * die with the process and we wait for that. Thus no need to
 		 * disable events in this case.
 		 */
-		if (done && !disabled && !perf_target__none(&opts->target)) {
+		if (done && !disabled && !target__none(&opts->target)) {
 			perf_evlist__disable(evsel_list);
 			disabled = true;
 		}
@@ -891,6 +888,8 @@ const struct option record_options[] = {
 		    "sample by weight (on special events only)"),
 	OPT_BOOLEAN(0, "transaction", &record.opts.sample_transaction,
 		    "sample transaction flags (special events only)"),
+	OPT_BOOLEAN(0, "force-per-cpu", &record.opts.target.force_per_cpu,
+		    "force the use of per-cpu mmaps"),
 	OPT_END()
 };
 
@@ -909,7 +908,7 @@ int cmd_record(int argc, const char **argv, const char *prefix __maybe_unused)
 
 	argc = parse_options(argc, argv, record_options, record_usage,
 			    PARSE_OPT_STOP_AT_NON_OPTION);
-	if (!argc && perf_target__none(&rec->opts.target))
+	if (!argc && target__none(&rec->opts.target))
 		usage_with_options(record_usage, record_options);
 
 	if (nr_cgroups && !rec->opts.target.system_wide) {
@@ -939,17 +938,17 @@ int cmd_record(int argc, const char **argv, const char *prefix __maybe_unused)
 		goto out_symbol_exit;
 	}
 
-	err = perf_target__validate(&rec->opts.target);
+	err = target__validate(&rec->opts.target);
 	if (err) {
-		perf_target__strerror(&rec->opts.target, err, errbuf, BUFSIZ);
+		target__strerror(&rec->opts.target, err, errbuf, BUFSIZ);
 		ui__warning("%s", errbuf);
 	}
 
-	err = perf_target__parse_uid(&rec->opts.target);
+	err = target__parse_uid(&rec->opts.target);
 	if (err) {
 		int saved_errno = errno;
 
-		perf_target__strerror(&rec->opts.target, err, errbuf, BUFSIZ);
+		target__strerror(&rec->opts.target, err, errbuf, BUFSIZ);
 		ui__error("%s", errbuf);
 
 		err = -saved_errno;
diff --git a/tools/perf/builtin-stat.c b/tools/perf/builtin-stat.c
index 0fc1c941a73c..ee0d565f83e3 100644
--- a/tools/perf/builtin-stat.c
+++ b/tools/perf/builtin-stat.c
@@ -108,7 +108,7 @@ enum {
 
 static struct perf_evlist	*evsel_list;
 
-static struct perf_target	target = {
+static struct target target = {
 	.uid	= UINT_MAX,
 };
 
@@ -294,11 +294,10 @@ static int create_perf_stat_counter(struct perf_evsel *evsel)
 
 	attr->inherit = !no_inherit;
 
-	if (perf_target__has_cpu(&target))
+	if (target__has_cpu(&target))
 		return perf_evsel__open_per_cpu(evsel, perf_evsel__cpus(evsel));
 
-	if (!perf_target__has_task(&target) &&
-	    perf_evsel__is_group_leader(evsel)) {
+	if (!target__has_task(&target) && perf_evsel__is_group_leader(evsel)) {
 		attr->disabled = 1;
 		if (!initial_delay)
 			attr->enable_on_exec = 1;
@@ -1236,7 +1235,7 @@ static void print_stat(int argc, const char **argv)
 			fprintf(output, "\'system wide");
 		else if (target.cpu_list)
 			fprintf(output, "\'CPU(s) %s", target.cpu_list);
-		else if (!perf_target__has_task(&target)) {
+		else if (!target__has_task(&target)) {
 			fprintf(output, "\'%s", argv[0]);
 			for (i = 1; i < argc; i++)
 				fprintf(output, " %s", argv[i]);
@@ -1667,7 +1666,7 @@ int cmd_stat(int argc, const char **argv, const char *prefix __maybe_unused)
 	} else if (big_num_opt == 0) /* User passed --no-big-num */
 		big_num = false;
 
-	if (!argc && perf_target__none(&target))
+	if (!argc && target__none(&target))
 		usage_with_options(stat_usage, options);
 
 	if (run_count < 0) {
@@ -1680,8 +1679,8 @@ int cmd_stat(int argc, const char **argv, const char *prefix __maybe_unused)
 	}
 
 	/* no_aggr, cgroup are for system-wide only */
-	if ((aggr_mode != AGGR_GLOBAL || nr_cgroups)
-	     && !perf_target__has_cpu(&target)) {
+	if ((aggr_mode != AGGR_GLOBAL || nr_cgroups) &&
+	    !target__has_cpu(&target)) {
 		fprintf(stderr, "both cgroup and no-aggregation "
 			"modes only available in system-wide mode\n");
 
@@ -1694,14 +1693,14 @@ int cmd_stat(int argc, const char **argv, const char *prefix __maybe_unused)
 	if (add_default_attributes())
 		goto out;
 
-	perf_target__validate(&target);
+	target__validate(&target);
 
 	if (perf_evlist__create_maps(evsel_list, &target) < 0) {
-		if (perf_target__has_task(&target)) {
+		if (target__has_task(&target)) {
 			pr_err("Problems finding threads of monitor\n");
 			parse_options_usage(stat_usage, options, "p", 1);
 			parse_options_usage(NULL, options, "t", 1);
-		} else if (perf_target__has_cpu(&target)) {
+		} else if (target__has_cpu(&target)) {
 			perror("failed to parse CPUs map");
 			parse_options_usage(stat_usage, options, "C", 1);
 			parse_options_usage(NULL, options, "a", 1);
diff --git a/tools/perf/builtin-top.c b/tools/perf/builtin-top.c
index 9acca8856ccb..71e6402729a8 100644
--- a/tools/perf/builtin-top.c
+++ b/tools/perf/builtin-top.c
@@ -950,14 +950,8 @@ static int __cmd_top(struct perf_top *top)
 	if (ret)
 		goto out_delete;
 
-	if (perf_target__has_task(&opts->target))
-		perf_event__synthesize_thread_map(&top->tool, top->evlist->threads,
-						  perf_event__process,
-						  &top->session->machines.host);
-	else
-		perf_event__synthesize_threads(&top->tool, perf_event__process,
-					       &top->session->machines.host);
-
+	machine__synthesize_threads(&top->session->machines.host, &opts->target,
+				    top->evlist->threads, false);
 	ret = perf_top__start_counters(top);
 	if (ret)
 		goto out_delete;
@@ -973,7 +967,7 @@ static int __cmd_top(struct perf_top *top)
 	 * XXX 'top' still doesn't start workloads like record, trace, but should,
 	 * so leave the check here.
 	 */
-        if (!perf_target__none(&opts->target))
+        if (!target__none(&opts->target))
                 perf_evlist__enable(top->evlist);
 
 	/* Wait for a minimal set of events before starting the snapshot */
@@ -1059,7 +1053,7 @@ int cmd_top(int argc, const char **argv, const char *prefix __maybe_unused)
 		.sym_pcnt_filter     = 5,
 	};
 	struct perf_record_opts *opts = &top.record_opts;
-	struct perf_target *target = &opts->target;
+	struct target *target = &opts->target;
 	const struct option options[] = {
 	OPT_CALLBACK('e', "event", &top.evlist, "event",
 		     "event selector. use 'perf list' to list available events",
@@ -1175,24 +1169,24 @@ int cmd_top(int argc, const char **argv, const char *prefix __maybe_unused)
 
 	setup_browser(false);
 
-	status = perf_target__validate(target);
+	status = target__validate(target);
 	if (status) {
-		perf_target__strerror(target, status, errbuf, BUFSIZ);
-		ui__warning("%s", errbuf);
+		target__strerror(target, status, errbuf, BUFSIZ);
+		ui__warning("%s\n", errbuf);
 	}
 
-	status = perf_target__parse_uid(target);
+	status = target__parse_uid(target);
 	if (status) {
 		int saved_errno = errno;
 
-		perf_target__strerror(target, status, errbuf, BUFSIZ);
-		ui__error("%s", errbuf);
+		target__strerror(target, status, errbuf, BUFSIZ);
+		ui__error("%s\n", errbuf);
 
 		status = -saved_errno;
 		goto out_delete_evlist;
 	}
 
-	if (perf_target__none(target))
+	if (target__none(target))
 		target->system_wide = true;
 
 	if (perf_evlist__create_maps(top.evlist, target) < 0)
diff --git a/tools/perf/builtin-trace.c b/tools/perf/builtin-trace.c
index 329b7832b5da..8be17fc462ba 100644
--- a/tools/perf/builtin-trace.c
+++ b/tools/perf/builtin-trace.c
@@ -149,21 +149,32 @@ static void perf_evsel__delete_priv(struct perf_evsel *evsel)
 	perf_evsel__delete(evsel);
 }
 
-static struct perf_evsel *perf_evsel__syscall_newtp(const char *direction,
-						    void *handler, int idx)
+static int perf_evsel__init_syscall_tp(struct perf_evsel *evsel, void *handler)
 {
-	struct perf_evsel *evsel = perf_evsel__newtp("raw_syscalls", direction, idx);
-
-	if (evsel) {
-		evsel->priv = malloc(sizeof(struct syscall_tp));
-
-		if (evsel->priv == NULL)
-			goto out_delete;
-
+	evsel->priv = malloc(sizeof(struct syscall_tp));
+	if (evsel->priv != NULL) {
 		if (perf_evsel__init_sc_tp_uint_field(evsel, id))
 			goto out_delete;
 
 		evsel->handler = handler;
+		return 0;
+	}
+
+	return -ENOMEM;
+
+out_delete:
+	free(evsel->priv);
+	evsel->priv = NULL;
+	return -ENOENT;
+}
+
+static struct perf_evsel *perf_evsel__syscall_newtp(const char *direction, void *handler)
+{
+	struct perf_evsel *evsel = perf_evsel__newtp("raw_syscalls", direction);
+
+	if (evsel) {
+		if (perf_evsel__init_syscall_tp(evsel, handler))
+			goto out_delete;
 	}
 
 	return evsel;
@@ -186,17 +197,16 @@ static int perf_evlist__add_syscall_newtp(struct perf_evlist *evlist,
 					  void *sys_exit_handler)
 {
 	int ret = -1;
-	int idx = evlist->nr_entries;
 	struct perf_evsel *sys_enter, *sys_exit;
 
-	sys_enter = perf_evsel__syscall_newtp("sys_enter", sys_enter_handler, idx++);
+	sys_enter = perf_evsel__syscall_newtp("sys_enter", sys_enter_handler);
 	if (sys_enter == NULL)
 		goto out;
 
 	if (perf_evsel__init_sc_tp_ptr_field(sys_enter, args))
 		goto out_delete_sys_enter;
 
-	sys_exit = perf_evsel__syscall_newtp("sys_exit", sys_exit_handler, idx++);
+	sys_exit = perf_evsel__syscall_newtp("sys_exit", sys_exit_handler);
 	if (sys_exit == NULL)
 		goto out_delete_sys_enter;
 
@@ -953,7 +963,8 @@ static struct syscall_fmt {
 	{ .name	    = "mmap",	    .hexret = true,
 	  .arg_scnprintf = { [0] = SCA_HEX,	  /* addr */
 			     [2] = SCA_MMAP_PROT, /* prot */
-			     [3] = SCA_MMAP_FLAGS, /* flags */ }, },
+			     [3] = SCA_MMAP_FLAGS, /* flags */
+			     [4] = SCA_FD, 	  /* fd */ }, },
 	{ .name	    = "mprotect",   .errmsg = true,
 	  .arg_scnprintf = { [0] = SCA_HEX, /* start */
 			     [2] = SCA_MMAP_PROT, /* prot */ }, },
@@ -1157,6 +1168,7 @@ struct trace {
 	bool			sched;
 	bool			multiple_threads;
 	bool			summary;
+	bool			summary_only;
 	bool			show_comm;
 	bool			show_tool_stats;
 	double			duration_filter;
@@ -1342,15 +1354,8 @@ static int trace__symbols_init(struct trace *trace, struct perf_evlist *evlist)
 	if (trace->host == NULL)
 		return -ENOMEM;
 
-	if (perf_target__has_task(&trace->opts.target)) {
-		err = perf_event__synthesize_thread_map(&trace->tool, evlist->threads,
-							trace__tool_process,
-							trace->host);
-	} else {
-		err = perf_event__synthesize_threads(&trace->tool, trace__tool_process,
-						     trace->host);
-	}
-
+	err = __machine__synthesize_threads(trace->host, &trace->tool, &trace->opts.target,
+					    evlist->threads, trace__tool_process, false);
 	if (err)
 		symbol__exit();
 
@@ -1607,7 +1612,7 @@ static int trace__sys_enter(struct trace *trace, struct perf_evsel *evsel,
 					   args, trace, thread);
 
 	if (!strcmp(sc->name, "exit_group") || !strcmp(sc->name, "exit")) {
-		if (!trace->duration_filter) {
+		if (!trace->duration_filter && !trace->summary_only) {
 			trace__fprintf_entry_head(trace, thread, 1, sample->time, trace->output);
 			fprintf(trace->output, "%-70s\n", ttrace->entry_str);
 		}
@@ -1660,6 +1665,9 @@ static int trace__sys_exit(struct trace *trace, struct perf_evsel *evsel,
 	} else if (trace->duration_filter)
 		goto out;
 
+	if (trace->summary_only)
+		goto out;
+
 	trace__fprintf_entry_head(trace, thread, duration, sample->time, trace->output);
 
 	if (ttrace->entry_pending) {
@@ -1762,16 +1770,6 @@ static int trace__process_sample(struct perf_tool *tool,
 	return err;
 }
 
-static bool
-perf_session__has_tp(struct perf_session *session, const char *name)
-{
-	struct perf_evsel *evsel;
-
-	evsel = perf_evlist__find_tracepoint_by_name(session->evlist, name);
-
-	return evsel != NULL;
-}
-
 static int parse_target_str(struct trace *trace)
 {
 	if (trace->opts.target.pid) {
@@ -1824,8 +1822,7 @@ static size_t trace__fprintf_thread_summary(struct trace *trace, FILE *fp);
 
 static void perf_evlist__add_vfs_getname(struct perf_evlist *evlist)
 {
-	struct perf_evsel *evsel = perf_evsel__newtp("probe", "vfs_getname",
-						     evlist->nr_entries);
+	struct perf_evsel *evsel = perf_evsel__newtp("probe", "vfs_getname");
 	if (evsel == NULL)
 		return;
 
@@ -2009,8 +2006,6 @@ out_error:
 static int trace__replay(struct trace *trace)
 {
 	const struct perf_evsel_str_handler handlers[] = {
-		{ "raw_syscalls:sys_enter",  trace__sys_enter, },
-		{ "raw_syscalls:sys_exit",   trace__sys_exit, },
 		{ "probe:vfs_getname",	     trace__vfs_getname, },
 	};
 	struct perf_data_file file = {
@@ -2018,6 +2013,7 @@ static int trace__replay(struct trace *trace)
 		.mode  = PERF_DATA_MODE_READ,
 	};
 	struct perf_session *session;
+	struct perf_evsel *evsel;
 	int err = -1;
 
 	trace->tool.sample	  = trace__process_sample;
@@ -2049,13 +2045,29 @@ static int trace__replay(struct trace *trace)
 	if (err)
 		goto out;
 
-	if (!perf_session__has_tp(session, "raw_syscalls:sys_enter")) {
-		pr_err("Data file does not have raw_syscalls:sys_enter events\n");
+	evsel = perf_evlist__find_tracepoint_by_name(session->evlist,
+						     "raw_syscalls:sys_enter");
+	if (evsel == NULL) {
+		pr_err("Data file does not have raw_syscalls:sys_enter event\n");
 		goto out;
 	}
 
-	if (!perf_session__has_tp(session, "raw_syscalls:sys_exit")) {
-		pr_err("Data file does not have raw_syscalls:sys_exit events\n");
+	if (perf_evsel__init_syscall_tp(evsel, trace__sys_enter) < 0 ||
+	    perf_evsel__init_sc_tp_ptr_field(evsel, args)) {
+		pr_err("Error during initialize raw_syscalls:sys_enter event\n");
+		goto out;
+	}
+
+	evsel = perf_evlist__find_tracepoint_by_name(session->evlist,
+						     "raw_syscalls:sys_exit");
+	if (evsel == NULL) {
+		pr_err("Data file does not have raw_syscalls:sys_exit event\n");
+		goto out;
+	}
+
+	if (perf_evsel__init_syscall_tp(evsel, trace__sys_exit) < 0 ||
+	    perf_evsel__init_sc_tp_uint_field(evsel, ret)) {
+		pr_err("Error during initialize raw_syscalls:sys_exit event\n");
 		goto out;
 	}
 
@@ -2082,12 +2094,7 @@ static size_t trace__fprintf_threads_header(FILE *fp)
 {
 	size_t printed;
 
-	printed  = fprintf(fp, "\n _____________________________________________________________________________\n");
-	printed += fprintf(fp, " __)    Summary of events    (__\n\n");
-	printed += fprintf(fp, "              [ task - pid ]     [ events ] [ ratio ]  [ runtime ]\n");
-	printed += fprintf(fp, "                                  syscall  count    min     max    avg  stddev\n");
-	printed += fprintf(fp, "                                                   msec    msec   msec     %%\n");
-	printed += fprintf(fp, " _____________________________________________________________________________\n\n");
+	printed  = fprintf(fp, "\n Summary of events:\n\n");
 
 	return printed;
 }
@@ -2105,6 +2112,10 @@ static size_t thread__dump_stats(struct thread_trace *ttrace,
 
 	printed += fprintf(fp, "\n");
 
+	printed += fprintf(fp, "   syscall            calls      min       avg       max      stddev\n");
+	printed += fprintf(fp, "                               (msec)    (msec)    (msec)        (%%)\n");
+	printed += fprintf(fp, "   --------------- -------- --------- --------- ---------     ------\n");
+
 	/* each int_node is a syscall */
 	while (inode) {
 		stats = inode->priv;
@@ -2119,10 +2130,10 @@ static size_t thread__dump_stats(struct thread_trace *ttrace,
 			avg /= NSEC_PER_MSEC;
 
 			sc = &trace->syscalls.table[inode->i];
-			printed += fprintf(fp, "%24s  %14s : ", "", sc->name);
-			printed += fprintf(fp, "%5" PRIu64 "  %8.3f  %8.3f",
-					   n, min, max);
-			printed += fprintf(fp, "  %8.3f  %6.2f\n", avg, pct);
+			printed += fprintf(fp, "   %-15s", sc->name);
+			printed += fprintf(fp, " %8" PRIu64 " %9.3f %9.3f",
+					   n, min, avg);
+			printed += fprintf(fp, " %9.3f %9.2f%%\n", max, pct);
 		}
 
 		inode = intlist__next(inode);
@@ -2163,10 +2174,10 @@ static int trace__fprintf_one_thread(struct thread *thread, void *priv)
 	else if (ratio > 5.0)
 		color = PERF_COLOR_YELLOW;
 
-	printed += color_fprintf(fp, color, "%20s", thread__comm_str(thread));
-	printed += fprintf(fp, " - %-5d :%11lu   [", thread->tid, ttrace->nr_events);
-	printed += color_fprintf(fp, color, "%5.1f%%", ratio);
-	printed += fprintf(fp, " ] %10.3f ms\n", ttrace->runtime_ms);
+	printed += color_fprintf(fp, color, " %s (%d), ", thread__comm_str(thread), thread->tid);
+	printed += fprintf(fp, "%lu events, ", ttrace->nr_events);
+	printed += color_fprintf(fp, color, "%.1f%%", ratio);
+	printed += fprintf(fp, ", %.3f msec\n", ttrace->runtime_ms);
 	printed += thread__dump_stats(ttrace, trace, fp);
 
 	data->printed += printed;
@@ -2275,8 +2286,10 @@ int cmd_trace(int argc, const char **argv, const char *prefix __maybe_unused)
 	OPT_INCR('v', "verbose", &verbose, "be more verbose"),
 	OPT_BOOLEAN('T', "time", &trace.full_time,
 		    "Show full timestamp, not time relative to first start"),
-	OPT_BOOLEAN(0, "summary", &trace.summary,
-		    "Show syscall summary with statistics"),
+	OPT_BOOLEAN('s', "summary", &trace.summary_only,
+		    "Show only syscall summary with statistics"),
+	OPT_BOOLEAN('S', "with-summary", &trace.summary,
+		    "Show all syscalls and summary with statistics"),
 	OPT_END()
 	};
 	int err;
@@ -2287,6 +2300,10 @@ int cmd_trace(int argc, const char **argv, const char *prefix __maybe_unused)
 
 	argc = parse_options(argc, argv, trace_options, trace_usage, 0);
 
+	/* summary_only implies summary option, but don't overwrite summary if set */
+	if (trace.summary_only)
+		trace.summary = trace.summary_only;
+
 	if (output_name != NULL) {
 		err = trace__open_output(&trace, output_name);
 		if (err < 0) {
@@ -2310,21 +2327,21 @@ int cmd_trace(int argc, const char **argv, const char *prefix __maybe_unused)
 		}
 	}
 
-	err = perf_target__validate(&trace.opts.target);
+	err = target__validate(&trace.opts.target);
 	if (err) {
-		perf_target__strerror(&trace.opts.target, err, bf, sizeof(bf));
+		target__strerror(&trace.opts.target, err, bf, sizeof(bf));
 		fprintf(trace.output, "%s", bf);
 		goto out_close;
 	}
 
-	err = perf_target__parse_uid(&trace.opts.target);
+	err = target__parse_uid(&trace.opts.target);
 	if (err) {
-		perf_target__strerror(&trace.opts.target, err, bf, sizeof(bf));
+		target__strerror(&trace.opts.target, err, bf, sizeof(bf));
 		fprintf(trace.output, "%s", bf);
 		goto out_close;
 	}
 
-	if (!argc && perf_target__none(&trace.opts.target))
+	if (!argc && target__none(&trace.opts.target))
 		trace.opts.target.system_wide = true;
 
 	if (input_name)
diff --git a/tools/perf/config/Makefile b/tools/perf/config/Makefile
index f5905f2b197d..f7d11a811c74 100644
--- a/tools/perf/config/Makefile
+++ b/tools/perf/config/Makefile
@@ -142,7 +142,8 @@ CORE_FEATURE_TESTS =			\
 	libunwind			\
 	on-exit				\
 	stackprotector			\
-	stackprotector-all
+	stackprotector-all		\
+	timerfd
 
 #
 # So here we detect whether test-all was rebuilt, to be able
@@ -328,8 +329,14 @@ ifndef NO_LIBUNWIND
     msg := $(warning No libunwind found, disabling post unwind support. Please install libunwind-dev[el] >= 1.1);
     NO_LIBUNWIND := 1
   else
-    ifneq ($(feature-libunwind-debug-frame), 1)
-      msg := $(warning No debug_frame support found in libunwind);
+    ifeq ($(ARCH),arm)
+      $(call feature_check,libunwind-debug-frame)
+      ifneq ($(feature-libunwind-debug-frame), 1)
+        msg := $(warning No debug_frame support found in libunwind);
+        CFLAGS += -DNO_LIBUNWIND_DEBUG_FRAME
+      endif
+    else
+      # non-ARM has no dwarf_find_debug_frame() function:
       CFLAGS += -DNO_LIBUNWIND_DEBUG_FRAME
     endif
   endif
@@ -405,7 +412,6 @@ else
   endif
 endif
 
-$(call feature_check,timerfd)
 ifeq ($(feature-timerfd), 1)
   CFLAGS += -DHAVE_TIMERFD_SUPPORT
 else
diff --git a/tools/perf/config/feature-checks/Makefile b/tools/perf/config/feature-checks/Makefile
index e8e195f49a4e..87e790017c69 100644
--- a/tools/perf/config/feature-checks/Makefile
+++ b/tools/perf/config/feature-checks/Makefile
@@ -76,6 +76,9 @@ test-libnuma:
 test-libunwind:
 	$(BUILD) $(LIBUNWIND_LIBS) -lelf
 
+test-libunwind-debug-frame:
+	$(BUILD) $(LIBUNWIND_LIBS) -lelf
+
 test-libaudit:
 	$(BUILD) -laudit
 
diff --git a/tools/perf/config/feature-checks/test-all.c b/tools/perf/config/feature-checks/test-all.c
index 799865b60772..59e7a705e146 100644
--- a/tools/perf/config/feature-checks/test-all.c
+++ b/tools/perf/config/feature-checks/test-all.c
@@ -49,10 +49,6 @@
 # include "test-libunwind.c"
 #undef main
 
-#define main main_test_libunwind_debug_frame
-# include "test-libunwind-debug-frame.c"
-#undef main
-
 #define main main_test_libaudit
 # include "test-libaudit.c"
 #undef main
diff --git a/tools/perf/perf.h b/tools/perf/perf.h
index 6a587e84fdfe..b079304bd53d 100644
--- a/tools/perf/perf.h
+++ b/tools/perf/perf.h
@@ -248,7 +248,7 @@ enum perf_call_graph_mode {
 };
 
 struct perf_record_opts {
-	struct perf_target target;
+	struct target target;
 	int	     call_graph;
 	bool	     group;
 	bool	     inherit_stat;
diff --git a/tools/perf/tests/code-reading.c b/tools/perf/tests/code-reading.c
index 49ccc3b2995e..85d4919dd623 100644
--- a/tools/perf/tests/code-reading.c
+++ b/tools/perf/tests/code-reading.c
@@ -275,8 +275,19 @@ static int process_event(struct machine *machine, struct perf_evlist *evlist,
 	if (event->header.type == PERF_RECORD_SAMPLE)
 		return process_sample_event(machine, evlist, event, state);
 
-	if (event->header.type < PERF_RECORD_MAX)
-		return machine__process_event(machine, event, NULL);
+	if (event->header.type == PERF_RECORD_THROTTLE ||
+	    event->header.type == PERF_RECORD_UNTHROTTLE)
+		return 0;
+
+	if (event->header.type < PERF_RECORD_MAX) {
+		int ret;
+
+		ret = machine__process_event(machine, event, NULL);
+		if (ret < 0)
+			pr_debug("machine__process_event failed, event type %u\n",
+				 event->header.type);
+		return ret;
+	}
 
 	return 0;
 }
@@ -441,7 +452,7 @@ static int do_test_code_reading(bool try_kcore)
 	}
 
 	ret = perf_event__synthesize_thread_map(NULL, threads,
-						perf_event__process, machine);
+						perf_event__process, machine, false);
 	if (ret < 0) {
 		pr_debug("perf_event__synthesize_thread_map failed\n");
 		goto out_err;
diff --git a/tools/perf/tests/evsel-tp-sched.c b/tools/perf/tests/evsel-tp-sched.c
index 9b98c1554833..4774f7fbb758 100644
--- a/tools/perf/tests/evsel-tp-sched.c
+++ b/tools/perf/tests/evsel-tp-sched.c
@@ -32,7 +32,7 @@ static int perf_evsel__test_field(struct perf_evsel *evsel, const char *name,
 
 int test__perf_evsel__tp_sched_test(void)
 {
-	struct perf_evsel *evsel = perf_evsel__newtp("sched", "sched_switch", 0);
+	struct perf_evsel *evsel = perf_evsel__newtp("sched", "sched_switch");
 	int ret = 0;
 
 	if (evsel == NULL) {
@@ -63,7 +63,7 @@ int test__perf_evsel__tp_sched_test(void)
 
 	perf_evsel__delete(evsel);
 
-	evsel = perf_evsel__newtp("sched", "sched_wakeup", 0);
+	evsel = perf_evsel__newtp("sched", "sched_wakeup");
 
 	if (perf_evsel__test_field(evsel, "comm", 16, true))
 		ret = -1;
diff --git a/tools/perf/tests/mmap-basic.c b/tools/perf/tests/mmap-basic.c
index a7232c204eb9..d64ab79c6d35 100644
--- a/tools/perf/tests/mmap-basic.c
+++ b/tools/perf/tests/mmap-basic.c
@@ -65,7 +65,7 @@ int test__basic_mmap(void)
 		char name[64];
 
 		snprintf(name, sizeof(name), "sys_enter_%s", syscall_names[i]);
-		evsels[i] = perf_evsel__newtp("syscalls", name, i);
+		evsels[i] = perf_evsel__newtp("syscalls", name);
 		if (evsels[i] == NULL) {
 			pr_debug("perf_evsel__new\n");
 			goto out_free_evlist;
diff --git a/tools/perf/tests/open-syscall-all-cpus.c b/tools/perf/tests/open-syscall-all-cpus.c
index b0657a9ccda6..5fecdbd2f5f7 100644
--- a/tools/perf/tests/open-syscall-all-cpus.c
+++ b/tools/perf/tests/open-syscall-all-cpus.c
@@ -26,7 +26,7 @@ int test__open_syscall_event_on_all_cpus(void)
 
 	CPU_ZERO(&cpu_set);
 
-	evsel = perf_evsel__newtp("syscalls", "sys_enter_open", 0);
+	evsel = perf_evsel__newtp("syscalls", "sys_enter_open");
 	if (evsel == NULL) {
 		pr_debug("is debugfs mounted on /sys/kernel/debug?\n");
 		goto out_thread_map_delete;
diff --git a/tools/perf/tests/open-syscall-tp-fields.c b/tools/perf/tests/open-syscall-tp-fields.c
index 524b221b829b..41cc0badb74b 100644
--- a/tools/perf/tests/open-syscall-tp-fields.c
+++ b/tools/perf/tests/open-syscall-tp-fields.c
@@ -27,7 +27,7 @@ int test__syscall_open_tp_fields(void)
 		goto out;
 	}
 
-	evsel = perf_evsel__newtp("syscalls", "sys_enter_open", 0);
+	evsel = perf_evsel__newtp("syscalls", "sys_enter_open");
 	if (evsel == NULL) {
 		pr_debug("%s: perf_evsel__newtp\n", __func__);
 		goto out_delete_evlist;
diff --git a/tools/perf/tests/open-syscall.c b/tools/perf/tests/open-syscall.c
index befc0671f95d..c1dc7d25f38c 100644
--- a/tools/perf/tests/open-syscall.c
+++ b/tools/perf/tests/open-syscall.c
@@ -15,7 +15,7 @@ int test__open_syscall_event(void)
 		return -1;
 	}
 
-	evsel = perf_evsel__newtp("syscalls", "sys_enter_open", 0);
+	evsel = perf_evsel__newtp("syscalls", "sys_enter_open");
 	if (evsel == NULL) {
 		pr_debug("is debugfs mounted on /sys/kernel/debug?\n");
 		goto out_thread_map_delete;
diff --git a/tools/perf/tests/parse-events.c b/tools/perf/tests/parse-events.c
index ef671cd41bb3..3cbd10496087 100644
--- a/tools/perf/tests/parse-events.c
+++ b/tools/perf/tests/parse-events.c
@@ -441,9 +441,8 @@ static int test__checkevent_pmu_name(struct perf_evlist *evlist)
 
 static int test__checkevent_pmu_events(struct perf_evlist *evlist)
 {
-	struct perf_evsel *evsel;
+	struct perf_evsel *evsel = perf_evlist__first(evlist);
 
-	evsel = list_entry(evlist->entries.next, struct perf_evsel, node);
 	TEST_ASSERT_VAL("wrong number of entries", 1 == evlist->nr_entries);
 	TEST_ASSERT_VAL("wrong type", PERF_TYPE_RAW == evsel->attr.type);
 	TEST_ASSERT_VAL("wrong exclude_user",
diff --git a/tools/perf/tests/sw-clock.c b/tools/perf/tests/sw-clock.c
index 6e2b44ec0749..6664a7cd828c 100644
--- a/tools/perf/tests/sw-clock.c
+++ b/tools/perf/tests/sw-clock.c
@@ -9,7 +9,7 @@
 #include "util/cpumap.h"
 #include "util/thread_map.h"
 
-#define NR_LOOPS  1000000
+#define NR_LOOPS  10000000
 
 /*
  * This test will open software clock events (cpu-clock, task-clock)
@@ -34,7 +34,7 @@ static int __test__sw_clock_freq(enum perf_sw_ids clock_id)
 		.freq = 1,
 	};
 
-	attr.sample_freq = 10000;
+	attr.sample_freq = 500;
 
 	evlist = perf_evlist__new();
 	if (evlist == NULL) {
@@ -42,7 +42,7 @@ static int __test__sw_clock_freq(enum perf_sw_ids clock_id)
 		return -1;
 	}
 
-	evsel = perf_evsel__new(&attr, 0);
+	evsel = perf_evsel__new(&attr);
 	if (evsel == NULL) {
 		pr_debug("perf_evsel__new\n");
 		goto out_free_evlist;
@@ -57,7 +57,14 @@ static int __test__sw_clock_freq(enum perf_sw_ids clock_id)
 		goto out_delete_maps;
 	}
 
-	perf_evlist__open(evlist);
+	if (perf_evlist__open(evlist)) {
+		const char *knob = "/proc/sys/kernel/perf_event_max_sample_rate";
+
+		err = -errno;
+		pr_debug("Couldn't open evlist: %s\nHint: check %s, using %" PRIu64 " in this test.\n",
+			 strerror(errno), knob, (u64)attr.sample_freq);
+		goto out_delete_maps;
+	}
 
 	err = perf_evlist__mmap(evlist, 128, true);
 	if (err < 0) {
diff --git a/tools/perf/tests/task-exit.c b/tools/perf/tests/task-exit.c
index c33d95f9559a..d09ab579119e 100644
--- a/tools/perf/tests/task-exit.c
+++ b/tools/perf/tests/task-exit.c
@@ -28,7 +28,7 @@ int test__task_exit(void)
 	union perf_event *event;
 	struct perf_evsel *evsel;
 	struct perf_evlist *evlist;
-	struct perf_target target = {
+	struct target target = {
 		.uid		= UINT_MAX,
 		.uses_mmap	= true,
 	};
diff --git a/tools/perf/ui/browser.c b/tools/perf/ui/browser.c
index bbc782e364b0..cbaa7af45513 100644
--- a/tools/perf/ui/browser.c
+++ b/tools/perf/ui/browser.c
@@ -569,7 +569,7 @@ void ui_browser__argv_seek(struct ui_browser *browser, off_t offset, int whence)
 		browser->top = browser->top + browser->top_idx + offset;
 		break;
 	case SEEK_END:
-		browser->top = browser->top + browser->nr_entries + offset;
+		browser->top = browser->top + browser->nr_entries - 1 + offset;
 		break;
 	default:
 		return;
@@ -680,7 +680,7 @@ static void __ui_browser__line_arrow_down(struct ui_browser *browser,
 	if (end >= browser->top_idx + browser->height)
 		end_row = browser->height - 1;
 	else
-		end_row = end - browser->top_idx;;
+		end_row = end - browser->top_idx;
 
 	ui_browser__gotorc(browser, row, column);
 	SLsmg_draw_vline(end_row - row + 1);
diff --git a/tools/perf/ui/browsers/hists.c b/tools/perf/ui/browsers/hists.c
index 16848bb4c418..a440e03cd8c2 100644
--- a/tools/perf/ui/browsers/hists.c
+++ b/tools/perf/ui/browsers/hists.c
@@ -1847,15 +1847,15 @@ browse_hists:
 			switch (key) {
 			case K_TAB:
 				if (pos->node.next == &evlist->entries)
-					pos = list_entry(evlist->entries.next, struct perf_evsel, node);
+					pos = perf_evlist__first(evlist);
 				else
-					pos = list_entry(pos->node.next, struct perf_evsel, node);
+					pos = perf_evsel__next(pos);
 				goto browse_hists;
 			case K_UNTAB:
 				if (pos->node.prev == &evlist->entries)
-					pos = list_entry(evlist->entries.prev, struct perf_evsel, node);
+					pos = perf_evlist__last(evlist);
 				else
-					pos = list_entry(pos->node.prev, struct perf_evsel, node);
+					pos = perf_evsel__prev(pos);
 				goto browse_hists;
 			case K_ESC:
 				if (!ui_browser__dialog_yesno(&menu->b,
@@ -1943,8 +1943,7 @@ int perf_evlist__tui_browse_hists(struct perf_evlist *evlist, const char *help,
 
 single_entry:
 	if (nr_entries == 1) {
-		struct perf_evsel *first = list_entry(evlist->entries.next,
-						      struct perf_evsel, node);
+		struct perf_evsel *first = perf_evlist__first(evlist);
 		const char *ev_name = perf_evsel__name(first);
 
 		return perf_evsel__hists_browse(first, nr_entries, help,
diff --git a/tools/perf/ui/tui/progress.c b/tools/perf/ui/tui/progress.c
index 3e2d936d7443..c61d14b101e0 100644
--- a/tools/perf/ui/tui/progress.c
+++ b/tools/perf/ui/tui/progress.c
@@ -18,13 +18,14 @@ static void tui_progress__update(struct ui_progress *p)
 	if (p->total == 0)
 		return;
 
-	ui__refresh_dimensions(true);
+	ui__refresh_dimensions(false);
 	pthread_mutex_lock(&ui__lock);
 	y = SLtt_Screen_Rows / 2 - 2;
 	SLsmg_set_color(0);
 	SLsmg_draw_box(y, 0, 3, SLtt_Screen_Cols);
 	SLsmg_gotorc(y++, 1);
 	SLsmg_write_string((char *)p->title);
+	SLsmg_fill_region(y, 1, 1, SLtt_Screen_Cols - 2, ' ');
 	SLsmg_set_color(HE_COLORSET_SELECTED);
 	bar = ((SLtt_Screen_Cols - 2) * p->curr) / p->total;
 	SLsmg_fill_region(y, 1, 1, bar, ' ');
diff --git a/tools/perf/util/event.c b/tools/perf/util/event.c
index ec9ae1114ed4..bb788c109fe6 100644
--- a/tools/perf/util/event.c
+++ b/tools/perf/util/event.c
@@ -170,7 +170,8 @@ static int perf_event__synthesize_mmap_events(struct perf_tool *tool,
 					      union perf_event *event,
 					      pid_t pid, pid_t tgid,
 					      perf_event__handler_t process,
-					      struct machine *machine)
+					      struct machine *machine,
+					      bool mmap_data)
 {
 	char filename[PATH_MAX];
 	FILE *fp;
@@ -188,10 +189,6 @@ static int perf_event__synthesize_mmap_events(struct perf_tool *tool,
 	}
 
 	event->header.type = PERF_RECORD_MMAP;
-	/*
-	 * Just like the kernel, see __perf_event_mmap in kernel/perf_event.c
-	 */
-	event->header.misc = PERF_RECORD_MISC_USER;
 
 	while (1) {
 		char bf[BUFSIZ];
@@ -212,12 +209,22 @@ static int perf_event__synthesize_mmap_events(struct perf_tool *tool,
 		       &event->mmap.start, &event->mmap.len, prot,
 		       &event->mmap.pgoff,
 		       execname);
-
-		if (n != 5)
+		/*
+ 		 * Anon maps don't have the execname.
+ 		 */
+		if (n < 4)
 			continue;
+		/*
+		 * Just like the kernel, see __perf_event_mmap in kernel/perf_event.c
+		 */
+		event->header.misc = PERF_RECORD_MISC_USER;
 
-		if (prot[2] != 'x')
-			continue;
+		if (prot[2] != 'x') {
+			if (!mmap_data || prot[0] != 'r')
+				continue;
+
+			event->header.misc |= PERF_RECORD_MISC_MMAP_DATA;
+		}
 
 		if (!strcmp(execname, ""))
 			strcpy(execname, anonstr);
@@ -304,20 +311,21 @@ static int __event__synthesize_thread(union perf_event *comm_event,
 				      pid_t pid, int full,
 					  perf_event__handler_t process,
 				      struct perf_tool *tool,
-				      struct machine *machine)
+				      struct machine *machine, bool mmap_data)
 {
 	pid_t tgid = perf_event__synthesize_comm(tool, comm_event, pid, full,
 						 process, machine);
 	if (tgid == -1)
 		return -1;
 	return perf_event__synthesize_mmap_events(tool, mmap_event, pid, tgid,
-						  process, machine);
+						  process, machine, mmap_data);
 }
 
 int perf_event__synthesize_thread_map(struct perf_tool *tool,
 				      struct thread_map *threads,
 				      perf_event__handler_t process,
-				      struct machine *machine)
+				      struct machine *machine,
+				      bool mmap_data)
 {
 	union perf_event *comm_event, *mmap_event;
 	int err = -1, thread, j;
@@ -334,7 +342,8 @@ int perf_event__synthesize_thread_map(struct perf_tool *tool,
 	for (thread = 0; thread < threads->nr; ++thread) {
 		if (__event__synthesize_thread(comm_event, mmap_event,
 					       threads->map[thread], 0,
-					       process, tool, machine)) {
+					       process, tool, machine,
+					       mmap_data)) {
 			err = -1;
 			break;
 		}
@@ -356,10 +365,10 @@ int perf_event__synthesize_thread_map(struct perf_tool *tool,
 
 			/* if not, generate events for it */
 			if (need_leader &&
-			    __event__synthesize_thread(comm_event,
-						      mmap_event,
-						      comm_event->comm.pid, 0,
-						      process, tool, machine)) {
+			    __event__synthesize_thread(comm_event, mmap_event,
+						       comm_event->comm.pid, 0,
+						       process, tool, machine,
+						       mmap_data)) {
 				err = -1;
 				break;
 			}
@@ -374,7 +383,7 @@ out:
 
 int perf_event__synthesize_threads(struct perf_tool *tool,
 				   perf_event__handler_t process,
-				   struct machine *machine)
+				   struct machine *machine, bool mmap_data)
 {
 	DIR *proc;
 	struct dirent dirent, *next;
@@ -404,7 +413,7 @@ int perf_event__synthesize_threads(struct perf_tool *tool,
  		 * one thread couldn't be synthesized.
  		 */
 		__event__synthesize_thread(comm_event, mmap_event, pid, 1,
-					   process, tool, machine);
+					   process, tool, machine, mmap_data);
 	}
 
 	err = 0;
@@ -528,19 +537,22 @@ int perf_event__process_lost(struct perf_tool *tool __maybe_unused,
 
 size_t perf_event__fprintf_mmap(union perf_event *event, FILE *fp)
 {
-	return fprintf(fp, " %d/%d: [%#" PRIx64 "(%#" PRIx64 ") @ %#" PRIx64 "]: %s\n",
+	return fprintf(fp, " %d/%d: [%#" PRIx64 "(%#" PRIx64 ") @ %#" PRIx64 "]: %c %s\n",
 		       event->mmap.pid, event->mmap.tid, event->mmap.start,
-		       event->mmap.len, event->mmap.pgoff, event->mmap.filename);
+		       event->mmap.len, event->mmap.pgoff,
+		       (event->header.misc & PERF_RECORD_MISC_MMAP_DATA) ? 'r' : 'x',
+		       event->mmap.filename);
 }
 
 size_t perf_event__fprintf_mmap2(union perf_event *event, FILE *fp)
 {
 	return fprintf(fp, " %d/%d: [%#" PRIx64 "(%#" PRIx64 ") @ %#" PRIx64
-			   " %02x:%02x %"PRIu64" %"PRIu64"]: %s\n",
+			   " %02x:%02x %"PRIu64" %"PRIu64"]: %c %s\n",
 		       event->mmap2.pid, event->mmap2.tid, event->mmap2.start,
 		       event->mmap2.len, event->mmap2.pgoff, event->mmap2.maj,
 		       event->mmap2.min, event->mmap2.ino,
 		       event->mmap2.ino_generation,
+		       (event->header.misc & PERF_RECORD_MISC_MMAP_DATA) ? 'r' : 'x',
 		       event->mmap2.filename);
 }
 
diff --git a/tools/perf/util/event.h b/tools/perf/util/event.h
index f8d70f3003ab..30fec9901e44 100644
--- a/tools/perf/util/event.h
+++ b/tools/perf/util/event.h
@@ -208,10 +208,10 @@ typedef int (*perf_event__handler_t)(struct perf_tool *tool,
 int perf_event__synthesize_thread_map(struct perf_tool *tool,
 				      struct thread_map *threads,
 				      perf_event__handler_t process,
-				      struct machine *machine);
+				      struct machine *machine, bool mmap_data);
 int perf_event__synthesize_threads(struct perf_tool *tool,
 				   perf_event__handler_t process,
-				   struct machine *machine);
+				   struct machine *machine, bool mmap_data);
 int perf_event__synthesize_kernel_mmap(struct perf_tool *tool,
 				       perf_event__handler_t process,
 				       struct machine *machine,
diff --git a/tools/perf/util/evlist.c b/tools/perf/util/evlist.c
index b939221efd8d..bbc746aa5716 100644
--- a/tools/perf/util/evlist.c
+++ b/tools/perf/util/evlist.c
@@ -117,6 +117,8 @@ void perf_evlist__delete(struct perf_evlist *evlist)
 void perf_evlist__add(struct perf_evlist *evlist, struct perf_evsel *entry)
 {
 	list_add_tail(&entry->node, &evlist->entries);
+	entry->idx = evlist->nr_entries;
+
 	if (!evlist->nr_entries++)
 		perf_evlist__set_id_pos(evlist);
 }
@@ -165,7 +167,7 @@ int perf_evlist__add_default(struct perf_evlist *evlist)
 
 	event_attr_init(&attr);
 
-	evsel = perf_evsel__new(&attr, 0);
+	evsel = perf_evsel__new(&attr);
 	if (evsel == NULL)
 		goto error;
 
@@ -190,7 +192,7 @@ static int perf_evlist__add_attrs(struct perf_evlist *evlist,
 	size_t i;
 
 	for (i = 0; i < nr_attrs; i++) {
-		evsel = perf_evsel__new(attrs + i, evlist->nr_entries + i);
+		evsel = perf_evsel__new_idx(attrs + i, evlist->nr_entries + i);
 		if (evsel == NULL)
 			goto out_delete_partial_list;
 		list_add_tail(&evsel->node, &head);
@@ -249,9 +251,8 @@ perf_evlist__find_tracepoint_by_name(struct perf_evlist *evlist,
 int perf_evlist__add_newtp(struct perf_evlist *evlist,
 			   const char *sys, const char *name, void *handler)
 {
-	struct perf_evsel *evsel;
+	struct perf_evsel *evsel = perf_evsel__newtp(sys, name);
 
-	evsel = perf_evsel__newtp(sys, name, evlist->nr_entries);
 	if (evsel == NULL)
 		return -1;
 
@@ -704,12 +705,10 @@ static size_t perf_evlist__mmap_size(unsigned long pages)
 	return (pages + 1) * page_size;
 }
 
-int perf_evlist__parse_mmap_pages(const struct option *opt, const char *str,
-				  int unset __maybe_unused)
+static long parse_pages_arg(const char *str, unsigned long min,
+			    unsigned long max)
 {
-	unsigned int *mmap_pages = opt->value;
 	unsigned long pages, val;
-	size_t size;
 	static struct parse_tag tags[] = {
 		{ .tag  = 'B', .mult = 1       },
 		{ .tag  = 'K', .mult = 1 << 10 },
@@ -718,33 +717,49 @@ int perf_evlist__parse_mmap_pages(const struct option *opt, const char *str,
 		{ .tag  = 0 },
 	};
 
+	if (str == NULL)
+		return -EINVAL;
+
 	val = parse_tag_value(str, tags);
 	if (val != (unsigned long) -1) {
 		/* we got file size value */
 		pages = PERF_ALIGN(val, page_size) / page_size;
-		if (pages < (1UL << 31) && !is_power_of_2(pages)) {
-			pages = next_pow2(pages);
-			pr_info("rounding mmap pages size to %lu (%lu pages)\n",
-				pages * page_size, pages);
-		}
 	} else {
 		/* we got pages count value */
 		char *eptr;
 		pages = strtoul(str, &eptr, 10);
-		if (*eptr != '\0') {
-			pr_err("failed to parse --mmap_pages/-m value\n");
-			return -1;
-		}
+		if (*eptr != '\0')
+			return -EINVAL;
 	}
 
-	if (pages > UINT_MAX || pages > SIZE_MAX / page_size) {
-		pr_err("--mmap_pages/-m value too big\n");
-		return -1;
+	if ((pages == 0) && (min == 0)) {
+		/* leave number of pages at 0 */
+	} else if (pages < (1UL << 31) && !is_power_of_2(pages)) {
+		/* round pages up to next power of 2 */
+		pages = next_pow2(pages);
+		pr_info("rounding mmap pages size to %lu bytes (%lu pages)\n",
+			pages * page_size, pages);
 	}
 
-	size = perf_evlist__mmap_size(pages);
-	if (!size) {
-		pr_err("--mmap_pages/-m value must be a power of two.");
+	if (pages > max)
+		return -EINVAL;
+
+	return pages;
+}
+
+int perf_evlist__parse_mmap_pages(const struct option *opt, const char *str,
+				  int unset __maybe_unused)
+{
+	unsigned int *mmap_pages = opt->value;
+	unsigned long max = UINT_MAX;
+	long pages;
+
+	if (max < SIZE_MAX / page_size)
+		max = SIZE_MAX / page_size;
+
+	pages = parse_pages_arg(str, 1, max);
+	if (pages < 0) {
+		pr_err("Invalid argument for --mmap_pages/-m\n");
 		return -1;
 	}
 
@@ -796,8 +811,7 @@ int perf_evlist__mmap(struct perf_evlist *evlist, unsigned int pages,
 	return perf_evlist__mmap_per_cpu(evlist, prot, mask);
 }
 
-int perf_evlist__create_maps(struct perf_evlist *evlist,
-			     struct perf_target *target)
+int perf_evlist__create_maps(struct perf_evlist *evlist, struct target *target)
 {
 	evlist->threads = thread_map__new_str(target->pid, target->tid,
 					      target->uid);
@@ -805,9 +819,11 @@ int perf_evlist__create_maps(struct perf_evlist *evlist,
 	if (evlist->threads == NULL)
 		return -1;
 
-	if (perf_target__has_task(target))
+	if (target->force_per_cpu)
+		evlist->cpus = cpu_map__new(target->cpu_list);
+	else if (target__has_task(target))
 		evlist->cpus = cpu_map__dummy_new();
-	else if (!perf_target__has_cpu(target) && !target->uses_mmap)
+	else if (!target__has_cpu(target) && !target->uses_mmap)
 		evlist->cpus = cpu_map__dummy_new();
 	else
 		evlist->cpus = cpu_map__new(target->cpu_list);
@@ -1016,8 +1032,7 @@ out_err:
 	return err;
 }
 
-int perf_evlist__prepare_workload(struct perf_evlist *evlist,
-				  struct perf_target *target,
+int perf_evlist__prepare_workload(struct perf_evlist *evlist, struct target *target,
 				  const char *argv[], bool pipe_output,
 				  bool want_signal)
 {
@@ -1069,7 +1084,7 @@ int perf_evlist__prepare_workload(struct perf_evlist *evlist,
 		exit(-1);
 	}
 
-	if (perf_target__none(target))
+	if (target__none(target))
 		evlist->threads->map[0] = evlist->workload.pid;
 
 	close(child_ready_pipe[1]);
@@ -1135,7 +1150,7 @@ size_t perf_evlist__fprintf(struct perf_evlist *evlist, FILE *fp)
 				   perf_evsel__name(evsel));
 	}
 
-	return printed + fprintf(fp, "\n");;
+	return printed + fprintf(fp, "\n");
 }
 
 int perf_evlist__strerror_tp(struct perf_evlist *evlist __maybe_unused,
diff --git a/tools/perf/util/evlist.h b/tools/perf/util/evlist.h
index ecaa582f40e2..649d6ea98a84 100644
--- a/tools/perf/util/evlist.h
+++ b/tools/perf/util/evlist.h
@@ -102,7 +102,7 @@ void perf_evlist__config(struct perf_evlist *evlist,
 int perf_record_opts__config(struct perf_record_opts *opts);
 
 int perf_evlist__prepare_workload(struct perf_evlist *evlist,
-				  struct perf_target *target,
+				  struct target *target,
 				  const char *argv[], bool pipe_output,
 				  bool want_signal);
 int perf_evlist__start_workload(struct perf_evlist *evlist);
@@ -134,8 +134,7 @@ static inline void perf_evlist__set_maps(struct perf_evlist *evlist,
 	evlist->threads	= threads;
 }
 
-int perf_evlist__create_maps(struct perf_evlist *evlist,
-			     struct perf_target *target);
+int perf_evlist__create_maps(struct perf_evlist *evlist, struct target *target);
 void perf_evlist__delete_maps(struct perf_evlist *evlist);
 int perf_evlist__apply_filters(struct perf_evlist *evlist);
 
diff --git a/tools/perf/util/evsel.c b/tools/perf/util/evsel.c
index 5280820ed389..46dd4c2a41ce 100644
--- a/tools/perf/util/evsel.c
+++ b/tools/perf/util/evsel.c
@@ -168,7 +168,7 @@ void perf_evsel__init(struct perf_evsel *evsel,
 	perf_evsel__calc_id_pos(evsel);
 }
 
-struct perf_evsel *perf_evsel__new(struct perf_event_attr *attr, int idx)
+struct perf_evsel *perf_evsel__new_idx(struct perf_event_attr *attr, int idx)
 {
 	struct perf_evsel *evsel = zalloc(sizeof(*evsel));
 
@@ -219,7 +219,7 @@ out:
 	return format;
 }
 
-struct perf_evsel *perf_evsel__newtp(const char *sys, const char *name, int idx)
+struct perf_evsel *perf_evsel__newtp_idx(const char *sys, const char *name, int idx)
 {
 	struct perf_evsel *evsel = zalloc(sizeof(*evsel));
 
@@ -645,7 +645,7 @@ void perf_evsel__config(struct perf_evsel *evsel,
 		}
 	}
 
-	if (perf_target__has_cpu(&opts->target))
+	if (target__has_cpu(&opts->target) || opts->target.force_per_cpu)
 		perf_evsel__set_sample_bit(evsel, CPU);
 
 	if (opts->period)
@@ -653,7 +653,7 @@ void perf_evsel__config(struct perf_evsel *evsel,
 
 	if (!perf_missing_features.sample_id_all &&
 	    (opts->sample_time || !opts->no_inherit ||
-	     perf_target__has_cpu(&opts->target)))
+	     target__has_cpu(&opts->target) || opts->target.force_per_cpu))
 		perf_evsel__set_sample_bit(evsel, TIME);
 
 	if (opts->raw_samples) {
@@ -696,7 +696,7 @@ void perf_evsel__config(struct perf_evsel *evsel,
 	 * Setting enable_on_exec for independent events and
 	 * group leaders for traced executed by perf.
 	 */
-	if (perf_target__none(&opts->target) && perf_evsel__is_group_leader(evsel))
+	if (target__none(&opts->target) && perf_evsel__is_group_leader(evsel))
 		attr->enable_on_exec = 1;
 }
 
@@ -2006,8 +2006,7 @@ bool perf_evsel__fallback(struct perf_evsel *evsel, int err,
 	return false;
 }
 
-int perf_evsel__open_strerror(struct perf_evsel *evsel,
-			      struct perf_target *target,
+int perf_evsel__open_strerror(struct perf_evsel *evsel, struct target *target,
 			      int err, char *msg, size_t size)
 {
 	switch (err) {
diff --git a/tools/perf/util/evsel.h b/tools/perf/util/evsel.h
index 64ec8e1a7a28..1ea7c92e6e33 100644
--- a/tools/perf/util/evsel.h
+++ b/tools/perf/util/evsel.h
@@ -96,8 +96,19 @@ struct thread_map;
 struct perf_evlist;
 struct perf_record_opts;
 
-struct perf_evsel *perf_evsel__new(struct perf_event_attr *attr, int idx);
-struct perf_evsel *perf_evsel__newtp(const char *sys, const char *name, int idx);
+struct perf_evsel *perf_evsel__new_idx(struct perf_event_attr *attr, int idx);
+
+static inline struct perf_evsel *perf_evsel__new(struct perf_event_attr *attr)
+{
+	return perf_evsel__new_idx(attr, 0);
+}
+
+struct perf_evsel *perf_evsel__newtp_idx(const char *sys, const char *name, int idx);
+
+static inline struct perf_evsel *perf_evsel__newtp(const char *sys, const char *name)
+{
+	return perf_evsel__newtp_idx(sys, name, 0);
+}
 
 struct event_format *event_format__new(const char *sys, const char *name);
 
@@ -268,6 +279,11 @@ static inline struct perf_evsel *perf_evsel__next(struct perf_evsel *evsel)
 	return list_entry(evsel->node.next, struct perf_evsel, node);
 }
 
+static inline struct perf_evsel *perf_evsel__prev(struct perf_evsel *evsel)
+{
+	return list_entry(evsel->node.prev, struct perf_evsel, node);
+}
+
 /**
  * perf_evsel__is_group_leader - Return whether given evsel is a leader event
  *
@@ -307,8 +323,7 @@ int perf_evsel__fprintf(struct perf_evsel *evsel,
 
 bool perf_evsel__fallback(struct perf_evsel *evsel, int err,
 			  char *msg, size_t msgsize);
-int perf_evsel__open_strerror(struct perf_evsel *evsel,
-			      struct perf_target *target,
+int perf_evsel__open_strerror(struct perf_evsel *evsel, struct target *target,
 			      int err, char *msg, size_t size);
 
 static inline int perf_evsel__group_idx(struct perf_evsel *evsel)
diff --git a/tools/perf/util/header.c b/tools/perf/util/header.c
index 26d9520a0c1b..369c03648f88 100644
--- a/tools/perf/util/header.c
+++ b/tools/perf/util/header.c
@@ -2797,7 +2797,7 @@ int perf_session__read_header(struct perf_session *session)
 			perf_event__attr_swap(&f_attr.attr);
 
 		tmp = lseek(fd, 0, SEEK_CUR);
-		evsel = perf_evsel__new(&f_attr.attr, i);
+		evsel = perf_evsel__new(&f_attr.attr);
 
 		if (evsel == NULL)
 			goto out_delete_evlist;
@@ -2916,7 +2916,7 @@ int perf_event__process_attr(struct perf_tool *tool __maybe_unused,
 			return -ENOMEM;
 	}
 
-	evsel = perf_evsel__new(&event->attr.attr, evlist->nr_entries);
+	evsel = perf_evsel__new(&event->attr.attr);
 	if (evsel == NULL)
 		return -ENOMEM;
 
diff --git a/tools/perf/util/machine.c b/tools/perf/util/machine.c
index ce034c183a7e..84cdb072ac83 100644
--- a/tools/perf/util/machine.c
+++ b/tools/perf/util/machine.c
@@ -1368,7 +1368,7 @@ int machine__resolve_callchain(struct machine *machine,
 
 	return unwind__get_entries(unwind_entry, &callchain_cursor, machine,
 				   thread, evsel->attr.sample_regs_user,
-				   sample);
+				   sample, max_stack);
 
 }
 
@@ -1394,3 +1394,15 @@ int machine__for_each_thread(struct machine *machine,
 	}
 	return rc;
 }
+
+int __machine__synthesize_threads(struct machine *machine, struct perf_tool *tool,
+				  struct target *target, struct thread_map *threads,
+				  perf_event__handler_t process, bool data_mmap)
+{
+	if (target__has_task(target))
+		return perf_event__synthesize_thread_map(tool, threads, process, machine, data_mmap);
+	else if (target__has_cpu(target))
+		return perf_event__synthesize_threads(tool, process, machine, data_mmap);
+	/* command specified */
+	return 0;
+}
diff --git a/tools/perf/util/machine.h b/tools/perf/util/machine.h
index 2389ba81fafe..477133015440 100644
--- a/tools/perf/util/machine.h
+++ b/tools/perf/util/machine.h
@@ -4,6 +4,7 @@
 #include <sys/types.h>
 #include <linux/rbtree.h>
 #include "map.h"
+#include "event.h"
 
 struct addr_location;
 struct branch_stack;
@@ -178,4 +179,15 @@ int machine__for_each_thread(struct machine *machine,
 			     int (*fn)(struct thread *thread, void *p),
 			     void *priv);
 
+int __machine__synthesize_threads(struct machine *machine, struct perf_tool *tool,
+				  struct target *target, struct thread_map *threads,
+				  perf_event__handler_t process, bool data_mmap);
+static inline
+int machine__synthesize_threads(struct machine *machine, struct target *target,
+				struct thread_map *threads, bool data_mmap)
+{
+	return __machine__synthesize_threads(machine, NULL, target, threads,
+					     perf_event__process, data_mmap);
+}
+
 #endif /* __PERF_MACHINE_H */
diff --git a/tools/perf/util/parse-events.c b/tools/perf/util/parse-events.c
index c90e55cf7e82..6de6f89c2a61 100644
--- a/tools/perf/util/parse-events.c
+++ b/tools/perf/util/parse-events.c
@@ -277,7 +277,7 @@ static int __add_event(struct list_head *list, int *idx,
 
 	event_attr_init(attr);
 
-	evsel = perf_evsel__new(attr, (*idx)++);
+	evsel = perf_evsel__new_idx(attr, (*idx)++);
 	if (!evsel)
 		return -ENOMEM;
 
@@ -378,7 +378,7 @@ static int add_tracepoint(struct list_head *list, int *idx,
 {
 	struct perf_evsel *evsel;
 
-	evsel = perf_evsel__newtp(sys_name, evt_name, (*idx)++);
+	evsel = perf_evsel__newtp_idx(sys_name, evt_name, (*idx)++);
 	if (!evsel)
 		return -ENOMEM;
 
@@ -1097,7 +1097,7 @@ static bool is_event_supported(u8 type, unsigned config)
 		.threads = { 0 },
 	};
 
-	evsel = perf_evsel__new(&attr, 0);
+	evsel = perf_evsel__new(&attr);
 	if (evsel) {
 		ret = perf_evsel__open(evsel, NULL, &tmap.map) >= 0;
 		perf_evsel__delete(evsel);
diff --git a/tools/perf/util/sort.c b/tools/perf/util/sort.c
index 3c1b75c8b9a6..8b0bb1f4494a 100644
--- a/tools/perf/util/sort.c
+++ b/tools/perf/util/sort.c
@@ -1137,6 +1137,8 @@ static void sort_entry__setup_elide(struct sort_entry *se,
 
 void sort__setup_elide(FILE *output)
 {
+	struct sort_entry *se;
+
 	sort_entry__setup_elide(&sort_dso, symbol_conf.dso_list,
 				"dso", output);
 	sort_entry__setup_elide(&sort_comm, symbol_conf.comm_list,
@@ -1172,4 +1174,15 @@ void sort__setup_elide(FILE *output)
 					"snoop", output);
 	}
 
+	/*
+	 * It makes no sense to elide all of sort entries.
+	 * Just revert them to show up again.
+	 */
+	list_for_each_entry(se, &hist_entry__sort_list, list) {
+		if (!se->elide)
+			return;
+	}
+
+	list_for_each_entry(se, &hist_entry__sort_list, list)
+		se->elide = false;
 }
diff --git a/tools/perf/util/target.c b/tools/perf/util/target.c
index 065528b7563e..3c778a07b7cc 100644
--- a/tools/perf/util/target.c
+++ b/tools/perf/util/target.c
@@ -13,9 +13,9 @@
 #include <string.h>
 
 
-enum perf_target_errno perf_target__validate(struct perf_target *target)
+enum target_errno target__validate(struct target *target)
 {
-	enum perf_target_errno ret = PERF_ERRNO_TARGET__SUCCESS;
+	enum target_errno ret = TARGET_ERRNO__SUCCESS;
 
 	if (target->pid)
 		target->tid = target->pid;
@@ -23,42 +23,42 @@ enum perf_target_errno perf_target__validate(struct perf_target *target)
 	/* CPU and PID are mutually exclusive */
 	if (target->tid && target->cpu_list) {
 		target->cpu_list = NULL;
-		if (ret == PERF_ERRNO_TARGET__SUCCESS)
-			ret = PERF_ERRNO_TARGET__PID_OVERRIDE_CPU;
+		if (ret == TARGET_ERRNO__SUCCESS)
+			ret = TARGET_ERRNO__PID_OVERRIDE_CPU;
 	}
 
 	/* UID and PID are mutually exclusive */
 	if (target->tid && target->uid_str) {
 		target->uid_str = NULL;
-		if (ret == PERF_ERRNO_TARGET__SUCCESS)
-			ret = PERF_ERRNO_TARGET__PID_OVERRIDE_UID;
+		if (ret == TARGET_ERRNO__SUCCESS)
+			ret = TARGET_ERRNO__PID_OVERRIDE_UID;
 	}
 
 	/* UID and CPU are mutually exclusive */
 	if (target->uid_str && target->cpu_list) {
 		target->cpu_list = NULL;
-		if (ret == PERF_ERRNO_TARGET__SUCCESS)
-			ret = PERF_ERRNO_TARGET__UID_OVERRIDE_CPU;
+		if (ret == TARGET_ERRNO__SUCCESS)
+			ret = TARGET_ERRNO__UID_OVERRIDE_CPU;
 	}
 
 	/* PID and SYSTEM are mutually exclusive */
 	if (target->tid && target->system_wide) {
 		target->system_wide = false;
-		if (ret == PERF_ERRNO_TARGET__SUCCESS)
-			ret = PERF_ERRNO_TARGET__PID_OVERRIDE_SYSTEM;
+		if (ret == TARGET_ERRNO__SUCCESS)
+			ret = TARGET_ERRNO__PID_OVERRIDE_SYSTEM;
 	}
 
 	/* UID and SYSTEM are mutually exclusive */
 	if (target->uid_str && target->system_wide) {
 		target->system_wide = false;
-		if (ret == PERF_ERRNO_TARGET__SUCCESS)
-			ret = PERF_ERRNO_TARGET__UID_OVERRIDE_SYSTEM;
+		if (ret == TARGET_ERRNO__SUCCESS)
+			ret = TARGET_ERRNO__UID_OVERRIDE_SYSTEM;
 	}
 
 	return ret;
 }
 
-enum perf_target_errno perf_target__parse_uid(struct perf_target *target)
+enum target_errno target__parse_uid(struct target *target)
 {
 	struct passwd pwd, *result;
 	char buf[1024];
@@ -66,7 +66,7 @@ enum perf_target_errno perf_target__parse_uid(struct perf_target *target)
 
 	target->uid = UINT_MAX;
 	if (str == NULL)
-		return PERF_ERRNO_TARGET__SUCCESS;
+		return TARGET_ERRNO__SUCCESS;
 
 	/* Try user name first */
 	getpwnam_r(str, &pwd, buf, sizeof(buf), &result);
@@ -79,22 +79,22 @@ enum perf_target_errno perf_target__parse_uid(struct perf_target *target)
 		int uid = strtol(str, &endptr, 10);
 
 		if (*endptr != '\0')
-			return PERF_ERRNO_TARGET__INVALID_UID;
+			return TARGET_ERRNO__INVALID_UID;
 
 		getpwuid_r(uid, &pwd, buf, sizeof(buf), &result);
 
 		if (result == NULL)
-			return PERF_ERRNO_TARGET__USER_NOT_FOUND;
+			return TARGET_ERRNO__USER_NOT_FOUND;
 	}
 
 	target->uid = result->pw_uid;
-	return PERF_ERRNO_TARGET__SUCCESS;
+	return TARGET_ERRNO__SUCCESS;
 }
 
 /*
- * This must have a same ordering as the enum perf_target_errno.
+ * This must have a same ordering as the enum target_errno.
  */
-static const char *perf_target__error_str[] = {
+static const char *target__error_str[] = {
 	"PID/TID switch overriding CPU",
 	"PID/TID switch overriding UID",
 	"UID switch overriding CPU",
@@ -104,7 +104,7 @@ static const char *perf_target__error_str[] = {
 	"Problems obtaining information for user %s",
 };
 
-int perf_target__strerror(struct perf_target *target, int errnum,
+int target__strerror(struct target *target, int errnum,
 			  char *buf, size_t buflen)
 {
 	int idx;
@@ -124,21 +124,19 @@ int perf_target__strerror(struct perf_target *target, int errnum,
 		return 0;
 	}
 
-	if (errnum <  __PERF_ERRNO_TARGET__START ||
-	    errnum >= __PERF_ERRNO_TARGET__END)
+	if (errnum <  __TARGET_ERRNO__START || errnum >= __TARGET_ERRNO__END)
 		return -1;
 
-	idx = errnum - __PERF_ERRNO_TARGET__START;
-	msg = perf_target__error_str[idx];
+	idx = errnum - __TARGET_ERRNO__START;
+	msg = target__error_str[idx];
 
 	switch (errnum) {
-	case PERF_ERRNO_TARGET__PID_OVERRIDE_CPU
-	 ... PERF_ERRNO_TARGET__UID_OVERRIDE_SYSTEM:
+	case TARGET_ERRNO__PID_OVERRIDE_CPU ... TARGET_ERRNO__UID_OVERRIDE_SYSTEM:
 		snprintf(buf, buflen, "%s", msg);
 		break;
 
-	case PERF_ERRNO_TARGET__INVALID_UID:
-	case PERF_ERRNO_TARGET__USER_NOT_FOUND:
+	case TARGET_ERRNO__INVALID_UID:
+	case TARGET_ERRNO__USER_NOT_FOUND:
 		snprintf(buf, buflen, msg, target->uid_str);
 		break;
 
diff --git a/tools/perf/util/target.h b/tools/perf/util/target.h
index a4be8575fda5..2d0c50690892 100644
--- a/tools/perf/util/target.h
+++ b/tools/perf/util/target.h
@@ -4,7 +4,7 @@
 #include <stdbool.h>
 #include <sys/types.h>
 
-struct perf_target {
+struct target {
 	const char   *pid;
 	const char   *tid;
 	const char   *cpu_list;
@@ -12,10 +12,11 @@ struct perf_target {
 	uid_t	     uid;
 	bool	     system_wide;
 	bool	     uses_mmap;
+	bool	     force_per_cpu;
 };
 
-enum perf_target_errno {
-	PERF_ERRNO_TARGET__SUCCESS		= 0,
+enum target_errno {
+	TARGET_ERRNO__SUCCESS		= 0,
 
 	/*
 	 * Choose an arbitrary negative big number not to clash with standard
@@ -24,42 +25,40 @@ enum perf_target_errno {
 	 *
 	 * http://pubs.opengroup.org/onlinepubs/9699919799/basedefs/errno.h.html
 	 */
-	__PERF_ERRNO_TARGET__START		= -10000,
+	__TARGET_ERRNO__START		= -10000,
 
+	/* for target__validate() */
+	TARGET_ERRNO__PID_OVERRIDE_CPU	= __TARGET_ERRNO__START,
+	TARGET_ERRNO__PID_OVERRIDE_UID,
+	TARGET_ERRNO__UID_OVERRIDE_CPU,
+	TARGET_ERRNO__PID_OVERRIDE_SYSTEM,
+	TARGET_ERRNO__UID_OVERRIDE_SYSTEM,
 
-	/* for perf_target__validate() */
-	PERF_ERRNO_TARGET__PID_OVERRIDE_CPU	= __PERF_ERRNO_TARGET__START,
-	PERF_ERRNO_TARGET__PID_OVERRIDE_UID,
-	PERF_ERRNO_TARGET__UID_OVERRIDE_CPU,
-	PERF_ERRNO_TARGET__PID_OVERRIDE_SYSTEM,
-	PERF_ERRNO_TARGET__UID_OVERRIDE_SYSTEM,
+	/* for target__parse_uid() */
+	TARGET_ERRNO__INVALID_UID,
+	TARGET_ERRNO__USER_NOT_FOUND,
 
-	/* for perf_target__parse_uid() */
-	PERF_ERRNO_TARGET__INVALID_UID,
-	PERF_ERRNO_TARGET__USER_NOT_FOUND,
-
-	__PERF_ERRNO_TARGET__END,
+	__TARGET_ERRNO__END,
 };
 
-enum perf_target_errno perf_target__validate(struct perf_target *target);
-enum perf_target_errno perf_target__parse_uid(struct perf_target *target);
+enum target_errno target__validate(struct target *target);
+enum target_errno target__parse_uid(struct target *target);
 
-int perf_target__strerror(struct perf_target *target, int errnum, char *buf,
-			  size_t buflen);
+int target__strerror(struct target *target, int errnum, char *buf, size_t buflen);
 
-static inline bool perf_target__has_task(struct perf_target *target)
+static inline bool target__has_task(struct target *target)
 {
 	return target->tid || target->pid || target->uid_str;
 }
 
-static inline bool perf_target__has_cpu(struct perf_target *target)
+static inline bool target__has_cpu(struct target *target)
 {
 	return target->system_wide || target->cpu_list;
 }
 
-static inline bool perf_target__none(struct perf_target *target)
+static inline bool target__none(struct target *target)
 {
-	return !perf_target__has_task(target) && !perf_target__has_cpu(target);
+	return !target__has_task(target) && !target__has_cpu(target);
 }
 
 #endif /* _PERF_TARGET_H */
diff --git a/tools/perf/util/top.c b/tools/perf/util/top.c
index f857b51b6bde..ce793c7dd23c 100644
--- a/tools/perf/util/top.c
+++ b/tools/perf/util/top.c
@@ -27,7 +27,7 @@ size_t perf_top__header_snprintf(struct perf_top *top, char *bf, size_t size)
 	float ksamples_per_sec;
 	float esamples_percent;
 	struct perf_record_opts *opts = &top->record_opts;
-	struct perf_target *target = &opts->target;
+	struct target *target = &opts->target;
 	size_t ret = 0;
 
 	if (top->samples) {
diff --git a/tools/perf/util/unwind.c b/tools/perf/util/unwind.c
index 5390d0b8862a..0efd5393de85 100644
--- a/tools/perf/util/unwind.c
+++ b/tools/perf/util/unwind.c
@@ -559,7 +559,7 @@ static unw_accessors_t accessors = {
 };
 
 static int get_entries(struct unwind_info *ui, unwind_entry_cb_t cb,
-		       void *arg)
+		       void *arg, int max_stack)
 {
 	unw_addr_space_t addr_space;
 	unw_cursor_t c;
@@ -575,7 +575,7 @@ static int get_entries(struct unwind_info *ui, unwind_entry_cb_t cb,
 	if (ret)
 		display_error(ret);
 
-	while (!ret && (unw_step(&c) > 0)) {
+	while (!ret && (unw_step(&c) > 0) && max_stack--) {
 		unw_word_t ip;
 
 		unw_get_reg(&c, UNW_REG_IP, &ip);
@@ -588,7 +588,8 @@ static int get_entries(struct unwind_info *ui, unwind_entry_cb_t cb,
 
 int unwind__get_entries(unwind_entry_cb_t cb, void *arg,
 			struct machine *machine, struct thread *thread,
-			u64 sample_uregs, struct perf_sample *data)
+			u64 sample_uregs, struct perf_sample *data,
+			int max_stack)
 {
 	unw_word_t ip;
 	struct unwind_info ui = {
@@ -610,5 +611,5 @@ int unwind__get_entries(unwind_entry_cb_t cb, void *arg,
 	if (ret)
 		return -ENOMEM;
 
-	return get_entries(&ui, cb, arg);
+	return get_entries(&ui, cb, arg, max_stack);
 }
diff --git a/tools/perf/util/unwind.h b/tools/perf/util/unwind.h
index ec0c71a2ca2e..d5966f49e22c 100644
--- a/tools/perf/util/unwind.h
+++ b/tools/perf/util/unwind.h
@@ -18,7 +18,7 @@ int unwind__get_entries(unwind_entry_cb_t cb, void *arg,
 			struct machine *machine,
 			struct thread *thread,
 			u64 sample_uregs,
-			struct perf_sample *data);
+			struct perf_sample *data, int max_stack);
 int unwind__arch_reg_id(int regnum);
 #else
 static inline int
@@ -27,7 +27,8 @@ unwind__get_entries(unwind_entry_cb_t cb __maybe_unused,
 		    struct machine *machine __maybe_unused,
 		    struct thread *thread __maybe_unused,
 		    u64 sample_uregs __maybe_unused,
-		    struct perf_sample *data __maybe_unused)
+		    struct perf_sample *data __maybe_unused,
+		    int max_stack __maybe_unused)
 {
 	return 0;
 }
diff --git a/tools/thermal/tmon/Makefile b/tools/thermal/tmon/Makefile
new file mode 100644
index 000000000000..447321104ec0
--- /dev/null
+++ b/tools/thermal/tmon/Makefile
@@ -0,0 +1,47 @@
+VERSION = 1.0
+
+BINDIR=usr/bin
+WARNFLAGS=-Wall -Wshadow -W -Wformat -Wimplicit-function-declaration -Wimplicit-int
+CFLAGS= -O1 ${WARNFLAGS} -fstack-protector
+CC=gcc
+
+CFLAGS+=-D VERSION=\"$(VERSION)\"
+LDFLAGS+=
+TARGET=tmon
+
+INSTALL_PROGRAM=install -m 755 -p
+DEL_FILE=rm -f
+
+INSTALL_CONFIGFILE=install -m 644 -p
+CONFIG_FILE=
+CONFIG_PATH=
+
+
+OBJS = tmon.o tui.o sysfs.o pid.o
+OBJS +=
+
+tmon: $(OBJS) Makefile tmon.h
+	$(CC) ${CFLAGS} $(LDFLAGS) $(OBJS)  -o $(TARGET) -lm -lpanel -lncursesw  -lpthread
+
+valgrind: tmon
+	 sudo valgrind -v --track-origins=yes --tool=memcheck --leak-check=yes --show-reachable=yes --num-callers=20 --track-fds=yes ./$(TARGET)  1> /dev/null
+
+install:
+	- mkdir -p $(INSTALL_ROOT)/$(BINDIR)
+	- $(INSTALL_PROGRAM) "$(TARGET)" "$(INSTALL_ROOT)/$(BINDIR)/$(TARGET)"
+	- mkdir -p $(INSTALL_ROOT)/$(CONFIG_PATH)
+	- $(INSTALL_CONFIGFILE) "$(CONFIG_FILE)" "$(INSTALL_ROOT)/$(CONFIG_PATH)"
+
+uninstall:
+	$(DEL_FILE) "$(INSTALL_ROOT)/$(BINDIR)/$(TARGET)"
+	$(CONFIG_FILE) "$(CONFIG_PATH)"
+
+
+clean:
+	find . -name "*.o" | xargs $(DEL_FILE)
+	rm -f $(TARGET)
+
+dist:
+	git tag v$(VERSION)
+	git archive --format=tar --prefix="$(TARGET)-$(VERSION)/" v$(VERSION) | \
+		gzip > $(TARGET)-$(VERSION).tar.gz
diff --git a/tools/thermal/tmon/README b/tools/thermal/tmon/README
new file mode 100644
index 000000000000..457949897a8e
--- /dev/null
+++ b/tools/thermal/tmon/README
@@ -0,0 +1,50 @@
+TMON - A Monitoring and Testing Tool for Linux kernel thermal subsystem
+
+Why TMON?
+==========
+Increasingly, Linux is running on thermally constrained devices. The simple
+thermal relationship between processor and fan has become past for modern
+computers.
+
+As hardware vendors cope with the thermal constraints on their products, more
+and more sensors are added, new cooling capabilities are introduced. The
+complexity of the thermal relationship can grow exponentially among cooling
+devices, zones, sensors, and trip points. They can also change dynamically.
+
+To expose such relationship to the userspace, Linux generic thermal layer
+introduced sysfs entry at /sys/class/thermal with a matrix of symbolic
+links, trip point bindings, and device instances. To traverse such
+matrix by hand is not a trivial task. Testing is also difficult in that
+thermal conditions are often exception cases that hard to reach in
+normal operations.
+
+TMON is conceived as a tool to help visualize, tune, and test the
+complex thermal subsystem.
+
+Files
+=====
+	tmon.c : main function for set up and configurations.
+	tui.c : handles ncurses based user interface
+	sysfs.c : access to the generic thermal sysfs
+	pid.c : a proportional-integral-derivative (PID) controller
+	that can be used for thermal relationship training.
+
+Requirements
+============
+Depends on ncurses
+
+Build
+=========
+$ make
+$ sudo ./tmon -h
+Usage: tmon [OPTION...]
+  -c, --control         cooling device in control
+  -d, --daemon          run as daemon, no TUI
+  -l, --log             log data to /var/tmp/tmon.log
+  -h, --help            show this help message
+  -t, --time-interval   set time interval for sampling
+  -v, --version         show version
+  -g, --debug           debug message in syslog
+
+1. For monitoring only:
+$ sudo ./tmon
diff --git a/tools/thermal/tmon/pid.c b/tools/thermal/tmon/pid.c
new file mode 100644
index 000000000000..fd7e9e9d6f4a
--- /dev/null
+++ b/tools/thermal/tmon/pid.c
@@ -0,0 +1,131 @@
+/*
+ * pid.c PID controller for testing cooling devices
+ *
+ *
+ *
+ * Copyright (C) 2012 Intel Corporation. All rights reserved.
+ *
+ * This program is free software; you can redistribute it and/or
+ * modify it under the terms of the GNU General Public License version
+ * 2 or later as published by the Free Software Foundation.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+ * GNU General Public License for more details.
+ *
+ * Author Name Jacob Pan <jacob.jun.pan@linux.intel.com>
+ *
+ */
+
+#include <unistd.h>
+#include <stdio.h>
+#include <stdlib.h>
+#include <string.h>
+#include <stdint.h>
+#include <sys/types.h>
+#include <dirent.h>
+#include <libintl.h>
+#include <ctype.h>
+#include <assert.h>
+#include <time.h>
+#include <limits.h>
+#include <math.h>
+#include <sys/stat.h>
+#include <syslog.h>
+
+#include "tmon.h"
+
+/**************************************************************************
+ * PID (Proportional-Integral-Derivative) controller is commonly used in
+ * linear control system, consider the the process.
+ * G(s) = U(s)/E(s)
+ * kp = proportional gain
+ * ki = integral gain
+ * kd = derivative gain
+ * Ts
+ * We use type C Alan Bradley equation which takes set point off the
+ * output dependency in P and D term.
+ *
+ *   y[k] = y[k-1] - kp*(x[k] - x[k-1]) + Ki*Ts*e[k] - Kd*(x[k]
+ *          - 2*x[k-1]+x[k-2])/Ts
+ *
+ *
+ ***********************************************************************/
+struct pid_params p_param;
+/* cached data from previous loop */
+static double xk_1, xk_2; /* input temperature x[k-#] */
+
+/*
+ * TODO: make PID parameters tuned automatically,
+ * 1. use CPU burn to produce open loop unit step response
+ * 2. calculate PID based on Ziegler-Nichols rule
+ *
+ * add a flag for tuning PID
+ */
+int init_thermal_controller(void)
+{
+	int ret = 0;
+
+	/* init pid params */
+	p_param.ts = ticktime;
+	/* TODO: get it from TUI tuning tab */
+	p_param.kp = .36;
+	p_param.ki = 5.0;
+	p_param.kd = 0.19;
+
+	p_param.t_target = target_temp_user;
+
+	return ret;
+}
+
+void controller_reset(void)
+{
+	/* TODO: relax control data when not over thermal limit */
+	syslog(LOG_DEBUG, "TC inactive, relax p-state\n");
+	p_param.y_k = 0.0;
+	xk_1 = 0.0;
+	xk_2 = 0.0;
+	set_ctrl_state(0);
+}
+
+/* To be called at time interval Ts. Type C PID controller.
+ *    y[k] = y[k-1] - kp*(x[k] - x[k-1]) + Ki*Ts*e[k] - Kd*(x[k]
+ *          - 2*x[k-1]+x[k-2])/Ts
+ * TODO: add low pass filter for D term
+ */
+#define GUARD_BAND (2)
+void controller_handler(const double xk, double *yk)
+{
+	double ek;
+	double p_term, i_term, d_term;
+
+	ek = p_param.t_target - xk; /* error */
+	if (ek >= 3.0) {
+		syslog(LOG_DEBUG, "PID: %3.1f Below set point %3.1f, stop\n",
+			xk, p_param.t_target);
+		controller_reset();
+		*yk = 0.0;
+		return;
+	}
+	/* compute intermediate PID terms */
+	p_term = -p_param.kp * (xk - xk_1);
+	i_term = p_param.kp * p_param.ki * p_param.ts * ek;
+	d_term = -p_param.kp * p_param.kd * (xk - 2 * xk_1 + xk_2) / p_param.ts;
+	/* compute output */
+	*yk += p_term + i_term + d_term;
+	/* update sample data */
+	xk_1 = xk;
+	xk_2 = xk_1;
+
+	/* clamp output adjustment range */
+	if (*yk < -LIMIT_HIGH)
+		*yk = -LIMIT_HIGH;
+	else if (*yk > -LIMIT_LOW)
+		*yk = -LIMIT_LOW;
+
+	p_param.y_k = *yk;
+
+	set_ctrl_state(lround(fabs(p_param.y_k)));
+
+}
diff --git a/tools/thermal/tmon/sysfs.c b/tools/thermal/tmon/sysfs.c
new file mode 100644
index 000000000000..dfe454855cd2
--- /dev/null
+++ b/tools/thermal/tmon/sysfs.c
@@ -0,0 +1,596 @@
+/*
+ * sysfs.c sysfs ABI access functions for TMON program
+ *
+ * Copyright (C) 2013 Intel Corporation. All rights reserved.
+ *
+ * This program is free software; you can redistribute it and/or
+ * modify it under the terms of the GNU General Public License version
+ * 2 or later as published by the Free Software Foundation.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+ * GNU General Public License for more details.
+ *
+ * Author: Jacob Pan <jacob.jun.pan@linux.intel.com>
+ *
+ */
+#include <unistd.h>
+#include <stdio.h>
+#include <stdlib.h>
+#include <string.h>
+#include <stdint.h>
+#include <dirent.h>
+#include <libintl.h>
+#include <ctype.h>
+#include <time.h>
+#include <syslog.h>
+#include <sys/time.h>
+#include <errno.h>
+
+#include "tmon.h"
+
+struct tmon_platform_data ptdata;
+const char *trip_type_name[] = {
+	"critical",
+	"hot",
+	"passive",
+	"active",
+};
+
+int sysfs_set_ulong(char *path, char *filename, unsigned long val)
+{
+	FILE *fd;
+	int ret = -1;
+	char filepath[256];
+
+	snprintf(filepath, 256, "%s/%s", path, filename);
+
+	fd = fopen(filepath, "w");
+	if (!fd) {
+		syslog(LOG_ERR, "Err: open %s: %s\n", __func__, filepath);
+		return ret;
+	}
+	ret = fprintf(fd, "%lu", val);
+	fclose(fd);
+
+	return 0;
+}
+
+/* history of thermal data, used for control algo */
+#define NR_THERMAL_RECORDS 3
+struct thermal_data_record trec[NR_THERMAL_RECORDS];
+int cur_thermal_record; /* index to the trec array */
+
+static int sysfs_get_ulong(char *path, char *filename, unsigned long *p_ulong)
+{
+	FILE *fd;
+	int ret = -1;
+	char filepath[256];
+
+	snprintf(filepath, 256, "%s/%s", path, filename);
+
+	fd = fopen(filepath, "r");
+	if (!fd) {
+		syslog(LOG_ERR, "Err: open %s: %s\n", __func__, filepath);
+		return ret;
+	}
+	ret = fscanf(fd, "%lu", p_ulong);
+	fclose(fd);
+
+	return 0;
+}
+
+static int sysfs_get_string(char *path, char *filename, char *str)
+{
+	FILE *fd;
+	int ret = -1;
+	char filepath[256];
+
+	snprintf(filepath, 256, "%s/%s", path, filename);
+
+	fd = fopen(filepath, "r");
+	if (!fd) {
+		syslog(LOG_ERR, "Err: open %s: %s\n", __func__, filepath);
+		return ret;
+	}
+	ret = fscanf(fd, "%256s", str);
+	fclose(fd);
+
+	return ret;
+}
+
+/* get states of the cooling device instance */
+static int probe_cdev(struct cdev_info *cdi, char *path)
+{
+	sysfs_get_string(path, "type", cdi->type);
+	sysfs_get_ulong(path, "max_state",  &cdi->max_state);
+	sysfs_get_ulong(path, "cur_state", &cdi->cur_state);
+
+	syslog(LOG_INFO, "%s: %s: type %s, max %lu, curr %lu inst %d\n",
+		__func__, path,
+		cdi->type, cdi->max_state, cdi->cur_state, cdi->instance);
+
+	return 0;
+}
+
+static int str_to_trip_type(char *name)
+{
+	int i;
+
+	for (i = 0; i < NR_THERMAL_TRIP_TYPE; i++) {
+		if (!strcmp(name, trip_type_name[i]))
+			return i;
+	}
+
+	return -ENOENT;
+}
+
+/* scan and fill in trip point info for a thermal zone and trip point id */
+static int get_trip_point_data(char *tz_path, int tzid, int tpid)
+{
+	char filename[256];
+	char temp_str[256];
+	int trip_type;
+
+	if (tpid >= MAX_NR_TRIP)
+		return -EINVAL;
+	/* check trip point type */
+	snprintf(filename, sizeof(filename), "trip_point_%d_type", tpid);
+	sysfs_get_string(tz_path, filename, temp_str);
+	trip_type = str_to_trip_type(temp_str);
+	if (trip_type < 0) {
+		syslog(LOG_ERR, "%s:%s no matching type\n", __func__, temp_str);
+		return -ENOENT;
+	}
+	ptdata.tzi[tzid].tp[tpid].type = trip_type;
+	syslog(LOG_INFO, "%s:tz:%d tp:%d:type:%s type id %d\n", __func__, tzid,
+		tpid, temp_str, trip_type);
+
+	/* TODO: check attribute */
+
+	return 0;
+}
+
+/* return instance id for file format such as trip_point_4_temp */
+static int get_instance_id(char *name, int pos, int skip)
+{
+	char *ch;
+	int i = 0;
+
+	ch = strtok(name, "_");
+	while (ch != NULL) {
+		++i;
+		syslog(LOG_INFO, "%s:%s:%s:%d", __func__, name, ch, i);
+		ch = strtok(NULL, "_");
+		if (pos == i)
+			return atol(ch + skip);
+	}
+
+	return -1;
+}
+
+/* Find trip point info of a thermal zone */
+static int find_tzone_tp(char *tz_name, char *d_name, struct tz_info *tzi,
+			int tz_id)
+{
+	int tp_id;
+	unsigned long temp_ulong;
+
+	if (strstr(d_name, "trip_point") &&
+		strstr(d_name, "temp")) {
+		/* check if trip point temp is non-zero
+		 * ignore 0/invalid trip points
+		 */
+		sysfs_get_ulong(tz_name, d_name, &temp_ulong);
+		if (temp_ulong < MAX_TEMP_KC) {
+			tzi->nr_trip_pts++;
+			/* found a valid trip point */
+			tp_id = get_instance_id(d_name, 2, 0);
+			syslog(LOG_DEBUG, "tzone %s trip %d temp %lu tpnode %s",
+				tz_name, tp_id, temp_ulong, d_name);
+			if (tp_id < 0 || tp_id >= MAX_NR_TRIP) {
+				syslog(LOG_ERR, "Failed to find TP inst %s\n",
+					d_name);
+				return -1;
+			}
+			get_trip_point_data(tz_name, tz_id, tp_id);
+			tzi->tp[tp_id].temp = temp_ulong;
+		}
+	}
+
+	return 0;
+}
+
+/* check cooling devices for binding info. */
+static int find_tzone_cdev(struct dirent *nl, char *tz_name,
+			struct tz_info *tzi, int tz_id, int cid)
+{
+	unsigned long trip_instance = 0;
+	char cdev_name_linked[256];
+	char cdev_name[256];
+	char cdev_trip_name[256];
+	int cdev_id;
+
+	if (nl->d_type == DT_LNK) {
+		syslog(LOG_DEBUG, "TZ%d: cdev: %s cid %d\n", tz_id, nl->d_name,
+			cid);
+		tzi->nr_cdev++;
+		if (tzi->nr_cdev > ptdata.nr_cooling_dev) {
+			syslog(LOG_ERR, "Err: Too many cdev? %d\n",
+				tzi->nr_cdev);
+			return -EINVAL;
+		}
+		/* find the link to real cooling device record binding */
+		snprintf(cdev_name, 256, "%s/%s", tz_name, nl->d_name);
+		memset(cdev_name_linked, 0, sizeof(cdev_name_linked));
+		if (readlink(cdev_name, cdev_name_linked,
+				sizeof(cdev_name_linked) - 1) != -1) {
+			cdev_id = get_instance_id(cdev_name_linked, 1,
+						sizeof("device") - 1);
+			syslog(LOG_DEBUG, "cdev %s linked to %s : %d\n",
+				cdev_name, cdev_name_linked, cdev_id);
+			tzi->cdev_binding |= (1 << cdev_id);
+
+			/* find the trip point in which the cdev is binded to
+			 * in this tzone
+			 */
+			snprintf(cdev_trip_name, 256, "%s%s", nl->d_name,
+				"_trip_point");
+			sysfs_get_ulong(tz_name, cdev_trip_name,
+					&trip_instance);
+			/* validate trip point range, e.g. trip could return -1
+			 * when passive is enabled
+			 */
+			if (trip_instance > MAX_NR_TRIP)
+				trip_instance = 0;
+			tzi->trip_binding[cdev_id] |= 1 << trip_instance;
+			syslog(LOG_DEBUG, "cdev %s -> trip:%lu: 0x%lx %d\n",
+				cdev_name, trip_instance,
+				tzi->trip_binding[cdev_id],
+				cdev_id);
+
+
+		}
+		return 0;
+	}
+
+	return -ENODEV;
+}
+
+
+
+/*****************************************************************************
+ * Before calling scan_tzones, thermal sysfs must be probed to determine
+ * the number of thermal zones and cooling devices.
+ * We loop through each thermal zone and fill in tz_info struct, i.e.
+ * ptdata.tzi[]
+root@jacob-chiefriver:~# tree -d /sys/class/thermal/thermal_zone0
+/sys/class/thermal/thermal_zone0
+|-- cdev0 -> ../cooling_device4
+|-- cdev1 -> ../cooling_device3
+|-- cdev10 -> ../cooling_device7
+|-- cdev11 -> ../cooling_device6
+|-- cdev12 -> ../cooling_device5
+|-- cdev2 -> ../cooling_device2
+|-- cdev3 -> ../cooling_device1
+|-- cdev4 -> ../cooling_device0
+|-- cdev5 -> ../cooling_device12
+|-- cdev6 -> ../cooling_device11
+|-- cdev7 -> ../cooling_device10
+|-- cdev8 -> ../cooling_device9
+|-- cdev9 -> ../cooling_device8
+|-- device -> ../../../LNXSYSTM:00/device:62/LNXTHERM:00
+|-- power
+`-- subsystem -> ../../../../class/thermal
+*****************************************************************************/
+static int scan_tzones(void)
+{
+	DIR *dir;
+	struct dirent **namelist;
+	char tz_name[256];
+	int i, j, n, k = 0;
+
+	if (!ptdata.nr_tz_sensor)
+		return -1;
+
+	for (i = 0; i <= ptdata.max_tz_instance; i++) {
+		memset(tz_name, 0, sizeof(tz_name));
+		snprintf(tz_name, 256, "%s/%s%d", THERMAL_SYSFS, TZONE, i);
+
+		dir = opendir(tz_name);
+		if (!dir) {
+			syslog(LOG_INFO, "Thermal zone %s skipped\n", tz_name);
+			continue;
+		}
+		/* keep track of valid tzones */
+		n = scandir(tz_name, &namelist, 0, alphasort);
+		if (n < 0)
+			syslog(LOG_ERR, "scandir failed in %s",  tz_name);
+		else {
+			sysfs_get_string(tz_name, "type", ptdata.tzi[k].type);
+			ptdata.tzi[k].instance = i;
+			/* detect trip points and cdev attached to this tzone */
+			j = 0; /* index for cdev */
+			ptdata.tzi[k].nr_cdev = 0;
+			ptdata.tzi[k].nr_trip_pts = 0;
+			while (n--) {
+				char *temp_str;
+
+				if (find_tzone_tp(tz_name, namelist[n]->d_name,
+							&ptdata.tzi[k], k))
+					break;
+				temp_str = strstr(namelist[n]->d_name, "cdev");
+				if (!temp_str) {
+					free(namelist[n]);
+					continue;
+				}
+				if (!find_tzone_cdev(namelist[n], tz_name,
+							&ptdata.tzi[k], i, j))
+					j++; /* increment cdev index */
+				free(namelist[n]);
+			}
+			free(namelist);
+		}
+		/*TODO: reverse trip points */
+		closedir(dir);
+		syslog(LOG_INFO, "TZ %d has %d cdev\n",	i,
+			ptdata.tzi[k].nr_cdev);
+		k++;
+	}
+
+	return 0;
+}
+
+static int scan_cdevs(void)
+{
+	DIR *dir;
+	struct dirent **namelist;
+	char cdev_name[256];
+	int i, n, k = 0;
+
+	if (!ptdata.nr_cooling_dev) {
+		fprintf(stderr, "No cooling devices found\n");
+		return 0;
+	}
+	for (i = 0; i <= ptdata.max_cdev_instance; i++) {
+		memset(cdev_name, 0, sizeof(cdev_name));
+		snprintf(cdev_name, 256, "%s/%s%d", THERMAL_SYSFS, CDEV, i);
+
+		dir = opendir(cdev_name);
+		if (!dir) {
+			syslog(LOG_INFO, "Cooling dev %s skipped\n", cdev_name);
+			/* there is a gap in cooling device id, check again
+			 * for the same index.
+			 */
+			continue;
+		}
+
+		n = scandir(cdev_name, &namelist, 0, alphasort);
+		if (n < 0)
+			syslog(LOG_ERR, "scandir failed in %s",  cdev_name);
+		else {
+			sysfs_get_string(cdev_name, "type", ptdata.cdi[k].type);
+			ptdata.cdi[k].instance = i;
+			if (strstr(ptdata.cdi[k].type, ctrl_cdev)) {
+				ptdata.cdi[k].flag |= CDEV_FLAG_IN_CONTROL;
+				syslog(LOG_DEBUG, "control cdev id %d\n", i);
+			}
+			while (n--)
+				free(namelist[n]);
+			free(namelist);
+		}
+		closedir(dir);
+		k++;
+	}
+	return 0;
+}
+
+
+int probe_thermal_sysfs(void)
+{
+	DIR *dir;
+	struct dirent **namelist;
+	int n;
+
+	dir = opendir(THERMAL_SYSFS);
+	if (!dir) {
+		fprintf(stderr, "\nNo thermal sysfs, exit\n");
+		return -1;
+	}
+	n = scandir(THERMAL_SYSFS, &namelist, 0, alphasort);
+	if (n < 0)
+		syslog(LOG_ERR, "scandir failed in thermal sysfs");
+	else {
+		/* detect number of thermal zones and cooling devices */
+		while (n--) {
+			int inst;
+
+			if (strstr(namelist[n]->d_name, CDEV)) {
+				inst = get_instance_id(namelist[n]->d_name, 1,
+						sizeof("device") - 1);
+				/* keep track of the max cooling device since
+				 * there may be gaps.
+				 */
+				if (inst > ptdata.max_cdev_instance)
+					ptdata.max_cdev_instance = inst;
+
+				syslog(LOG_DEBUG, "found cdev: %s %d %d\n",
+					namelist[n]->d_name,
+					ptdata.nr_cooling_dev,
+					ptdata.max_cdev_instance);
+				ptdata.nr_cooling_dev++;
+			} else if (strstr(namelist[n]->d_name, TZONE)) {
+				inst = get_instance_id(namelist[n]->d_name, 1,
+						sizeof("zone") - 1);
+				if (inst > ptdata.max_tz_instance)
+					ptdata.max_tz_instance = inst;
+
+				syslog(LOG_DEBUG, "found tzone: %s %d %d\n",
+					namelist[n]->d_name,
+					ptdata.nr_tz_sensor,
+					ptdata.max_tz_instance);
+				ptdata.nr_tz_sensor++;
+			}
+			free(namelist[n]);
+		}
+		free(namelist);
+	}
+	syslog(LOG_INFO, "found %d tzone(s), %d cdev(s), target zone %d\n",
+		ptdata.nr_tz_sensor, ptdata.nr_cooling_dev,
+		target_thermal_zone);
+	closedir(dir);
+
+	if (!ptdata.nr_tz_sensor) {
+		fprintf(stderr, "\nNo thermal zones found, exit\n\n");
+		return -1;
+	}
+
+	ptdata.tzi = calloc(sizeof(struct tz_info), ptdata.max_tz_instance+1);
+	if (!ptdata.tzi) {
+		fprintf(stderr, "Err: allocate tz_info\n");
+		return -1;
+	}
+
+	/* we still show thermal zone information if there is no cdev */
+	if (ptdata.nr_cooling_dev) {
+		ptdata.cdi = calloc(sizeof(struct cdev_info),
+				ptdata.max_cdev_instance + 1);
+		if (!ptdata.cdi) {
+			free(ptdata.tzi);
+			fprintf(stderr, "Err: allocate cdev_info\n");
+			return -1;
+		}
+	}
+
+	/* now probe tzones */
+	if (scan_tzones())
+		return -1;
+	if (scan_cdevs())
+		return -1;
+	return 0;
+}
+
+/* convert sysfs zone instance to zone array index */
+int zone_instance_to_index(int zone_inst)
+{
+	int i;
+
+	for (i = 0; i < ptdata.nr_tz_sensor; i++)
+		if (ptdata.tzi[i].instance == zone_inst)
+			return i;
+	return -ENOENT;
+}
+
+/* read temperature of all thermal zones */
+int update_thermal_data()
+{
+	int i;
+	char tz_name[256];
+	static unsigned long samples;
+
+	if (!ptdata.nr_tz_sensor) {
+		syslog(LOG_ERR, "No thermal zones found!\n");
+		return -1;
+	}
+
+	/* circular buffer for keeping historic data */
+	if (cur_thermal_record >= NR_THERMAL_RECORDS)
+		cur_thermal_record = 0;
+	gettimeofday(&trec[cur_thermal_record].tv, NULL);
+	if (tmon_log) {
+		fprintf(tmon_log, "%lu ", ++samples);
+		fprintf(tmon_log, "%3.1f ", p_param.t_target);
+	}
+	for (i = 0; i < ptdata.nr_tz_sensor; i++) {
+		memset(tz_name, 0, sizeof(tz_name));
+		snprintf(tz_name, 256, "%s/%s%d", THERMAL_SYSFS, TZONE,
+			ptdata.tzi[i].instance);
+		sysfs_get_ulong(tz_name, "temp",
+				&trec[cur_thermal_record].temp[i]);
+		if (tmon_log)
+			fprintf(tmon_log, "%lu ",
+				trec[cur_thermal_record].temp[i]/1000);
+	}
+	for (i = 0; i < ptdata.nr_cooling_dev; i++) {
+		char cdev_name[256];
+		unsigned long val;
+
+		snprintf(cdev_name, 256, "%s/%s%d", THERMAL_SYSFS, CDEV,
+			ptdata.cdi[i].instance);
+		probe_cdev(&ptdata.cdi[i], cdev_name);
+		val = ptdata.cdi[i].cur_state;
+		if (val > 1000000)
+			val = 0;
+		if (tmon_log)
+			fprintf(tmon_log, "%lu ", val);
+	}
+
+	if (tmon_log) {
+		fprintf(tmon_log, "\n");
+		fflush(tmon_log);
+	}
+
+	return 0;
+}
+
+void set_ctrl_state(unsigned long state)
+{
+	char ctrl_cdev_path[256];
+	int i;
+	unsigned long cdev_state;
+
+	if (no_control)
+		return;
+	/* set all ctrl cdev to the same state */
+	for (i = 0; i < ptdata.nr_cooling_dev; i++) {
+		if (ptdata.cdi[i].flag & CDEV_FLAG_IN_CONTROL) {
+			if (ptdata.cdi[i].max_state < 10) {
+				strcpy(ctrl_cdev, "None.");
+				return;
+			}
+			/* scale to percentage of max_state */
+			cdev_state = state * ptdata.cdi[i].max_state/100;
+			syslog(LOG_DEBUG,
+				"ctrl cdev %d set state %lu scaled to %lu\n",
+				ptdata.cdi[i].instance, state, cdev_state);
+			snprintf(ctrl_cdev_path, 256, "%s/%s%d", THERMAL_SYSFS,
+				CDEV, ptdata.cdi[i].instance);
+			syslog(LOG_DEBUG, "ctrl cdev path %s", ctrl_cdev_path);
+			sysfs_set_ulong(ctrl_cdev_path, "cur_state",
+					cdev_state);
+		}
+	}
+}
+
+void get_ctrl_state(unsigned long *state)
+{
+	char ctrl_cdev_path[256];
+	int ctrl_cdev_id = -1;
+	int i;
+
+	/* TODO: take average of all ctrl types. also consider change based on
+	 * uevent. Take the first reading for now.
+	 */
+	for (i = 0; i < ptdata.nr_cooling_dev; i++) {
+		if (ptdata.cdi[i].flag & CDEV_FLAG_IN_CONTROL) {
+			ctrl_cdev_id = ptdata.cdi[i].instance;
+			syslog(LOG_INFO, "ctrl cdev %d get state\n",
+				ptdata.cdi[i].instance);
+			break;
+		}
+	}
+	if (ctrl_cdev_id == -1) {
+		*state = 0;
+		return;
+	}
+	snprintf(ctrl_cdev_path, 256, "%s/%s%d", THERMAL_SYSFS,
+		CDEV, ctrl_cdev_id);
+	sysfs_get_ulong(ctrl_cdev_path, "cur_state", state);
+}
+
+void free_thermal_data(void)
+{
+	free(ptdata.tzi);
+	free(ptdata.cdi);
+}
diff --git a/tools/thermal/tmon/tmon.8 b/tools/thermal/tmon/tmon.8
new file mode 100644
index 000000000000..0be727cb9892
--- /dev/null
+++ b/tools/thermal/tmon/tmon.8
@@ -0,0 +1,142 @@
+.TH TMON 8
+.SH NAME
+\fBtmon\fP - A monitoring and testing tool for Linux kernel thermal subsystem
+
+.SH SYNOPSIS
+.ft B
+.B tmon
+.RB [ Options ]
+.br
+.SH DESCRIPTION
+\fBtmon \fP can be used to visualize thermal relationship and
+real-time thermal data; tune
+and test cooling devices and sensors; collect thermal data for offline
+analysis and plot. \fBtmon\fP must be run as root in order to control device
+states via sysfs.
+.PP
+\fBFunctions\fP
+.PP
+.nf
+1. Thermal relationships:
+- show thermal zone information
+- show cooling device information
+- show trip point binding within each thermal zone
+- show trip point and cooling device instance bindings
+.PP
+2. Real time data display
+- show temperature of all thermal zones w.r.t. its trip points and types
+- show states of all cooling devices
+.PP
+3. Thermal relationship learning and device tuning
+- with a built-in Proportional Integral Derivative (\fBPID\fP)
+controller, user can pair a cooling device to a thermal sensor for
+testing the effectiveness and learn about the thermal distance between the two
+- allow manual control of cooling device states and target temperature
+.PP
+4. Data logging in /var/tmp/tmon.log
+- contains thermal configuration data, i.e. cooling device, thermal
+ zones, and trip points. Can be used for data  collection in remote
+ debugging.
+- log real-time thermal data into space separated format that can be
+ directly consumed by plotting tools such as Rscript.
+
+.SS Options
+.PP
+The \fB-c --control\fP option sets a cooling device type to control temperature
+of a thermal zone
+.PP
+The \fB-d --daemon\fP option runs \fBtmon \fP as daemon without user interface
+.PP
+The \fB-g --debug\fP option allow debug messages to be stored in syslog
+.PP
+The \fB-h --help\fP option shows help message
+.PP
+The \fB-l --log\fP option write data to /var/tmp/tmon.log
+.PP
+The \fB-t --time-interval\fP option sets the polling interval in seconds
+.PP
+The \fB-v --version\fP option shows the version of \fBtmon \fP
+.PP
+The \fB-z --zone\fP option sets the target therma zone instance to be controlled
+.PP
+
+.SH FIELD DESCRIPTIONS
+.nf
+.PP
+\fBP \fP passive cooling trip point type
+\fBA \fP active cooling trip point type (fan)
+\fBC \fP critical trip point type
+\fBA \fP hot trip point type
+\fBkp \fP proportional gain of \fBPID\fP controller
+\fBki \fP integral gain of \fBPID\fP controller
+\fBkd \fP derivative gain of \fBPID\fP controller
+
+.SH REQUIREMENT
+Build depends on ncurses
+.PP
+Runtime depends on window size large enough to show the number of
+devices found on the system.
+
+.PP
+
+.SH INTERACTIVE COMMANDS
+.pp
+.nf
+\fBCtrl-C, q/Q\fP stops \fBtmon\fP
+\fBTAB\fP shows tuning pop up panel, choose a letter to modify
+
+.SH EXAMPLES
+Without any parameters, tmon is in monitoring only mode and refresh
+screen every 1 second.
+.PP
+1. For monitoring only:
+.nf
+$ sudo ./tmon
+
+2. Use Processor cooling device to control thermal zone 0 at default 65C.
+$ sudo ./tmon -c Processor -z 0
+
+3. Use intel_powerclamp(idle injection) cooling device to control thermal zone 1
+$ sudo ./tmon -c intel_powerclamp -z 1
+
+4. Turn on debug and collect data log at /var/tmp/tmon.log
+$ sudo ./tmon -g -l
+
+For example, the log below shows PID controller was adjusting current states
+for all cooling devices with "Processor" type such that thermal zone 0
+can stay below 65 dC.
+
+#---------- THERMAL DATA LOG STARTED -----------
+Samples TargetTemp acpitz0    acpitz1    Fan0 Fan1 Fan2 Fan3 Fan4 Fan5
+Fan6 Fan7 Fan8 Fan9 Processor10 Processor11 Processor12 Processor13
+LCD14 intel_powerclamp15 1 65.0 65 65 0 0 0 0 0 0 0 0 0 0 0 0 0 0 6 0 2
+65.0 66 65 0 0 0 0 0 0 0 0 0 0 4 4 4 4 6 0 3 65.0 60 54 0 0 0 0 0 0 0 0
+0 0 4 4 4 4 6 0 4 65.0 53 53 0 0 0 0 0 0 0 0 0 0 4 4 4 4 6 0
+5 65.0 52 52 0 0 0 0 0 0 0 0 0 0 0 0 0 0 6 0
+6 65.0 53 65 0 0 0 0 0 0 0 0 0 0 0 0 0 0 6 0
+7 65.0 68 70 0 0 0 0 0 0 0 0 0 0 0 0 0 0 6 0
+8 65.0 68 68 0 0 0 0 0 0 0 0 0 0 5 5 5 5 6 0
+9 65.0 68 68 0 0 0 0 0 0 0 0 0 0 6 6 6 6 6 0
+10 65.0 67 67 0 0 0 0 0 0 0 0 0 0 7 7 7 7 6 0
+11 65.0 67 67 0 0 0 0 0 0 0 0 0 0 8 8 8 8 6 0
+12 65.0 67 67 0 0 0 0 0 0 0 0 0 0 8 8 8 8 6 0
+13 65.0 67 67 0 0 0 0 0 0 0 0 0 0 9 9 9 9 6 0
+14 65.0 66 66 0 0 0 0 0 0 0 0 0 0 10 10 10 10 6 0
+15 65.0 66 67 0 0 0 0 0 0 0 0 0 0 10 10 10 10 6 0
+16 65.0 66 66 0 0 0 0 0 0 0 0 0 0 11 11 11 11 6 0
+17 65.0 66 66 0 0 0 0 0 0 0 0 0 0 11 11 11 11 6 0
+18 65.0 64 61 0 0 0 0 0 0 0 0 0 0 11 11 11 11 6 0
+19 65.0 60 59 0 0 0 0 0 0 0 0 0 0 12 12 12 12 6 0
+
+Data can be read directly into an array by an example R-script below:
+
+#!/usr/bin/Rscript
+tdata <- read.table("/var/tmp/tmon.log", header=T, comment.char="#")
+attach(tdata)
+jpeg("tmon.jpg")
+X11()
+g_range <- range(0, intel_powerclamp15, TargetTemp, acpitz0)
+plot( Samples, intel_powerclamp15, col="blue", ylim=g_range, axes=FALSE, ann=FALSE)
+par(new=TRUE)
+lines(TargetTemp, type="o", pch=22, lty=2, col="red")
+dev.off()
diff --git a/tools/thermal/tmon/tmon.c b/tools/thermal/tmon/tmon.c
new file mode 100644
index 000000000000..b30f531173e4
--- /dev/null
+++ b/tools/thermal/tmon/tmon.c
@@ -0,0 +1,352 @@
+/*
+ * tmon.c Thermal Monitor (TMON) main function and entry point
+ *
+ * Copyright (C) 2012 Intel Corporation. All rights reserved.
+ *
+ * This program is free software; you can redistribute it and/or
+ * modify it under the terms of the GNU General Public License version
+ * 2 or later as published by the Free Software Foundation.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+ * GNU General Public License for more details.
+ *
+ * Author: Jacob Pan <jacob.jun.pan@linux.intel.com>
+ *
+ */
+
+#include <getopt.h>
+#include <unistd.h>
+#include <stdio.h>
+#include <stdlib.h>
+#include <string.h>
+#include <sys/types.h>
+#include <sys/stat.h>
+#include <ncurses.h>
+#include <ctype.h>
+#include <time.h>
+#include <signal.h>
+#include <limits.h>
+#include <sys/time.h>
+#include <pthread.h>
+#include <math.h>
+#include <stdarg.h>
+#include <syslog.h>
+
+#include "tmon.h"
+
+unsigned long ticktime = 1; /* seconds */
+unsigned long no_control = 1; /* monitoring only or use cooling device for
+			       * temperature control.
+			       */
+double time_elapsed = 0.0;
+unsigned long target_temp_user = 65; /* can be select by tui later */
+int dialogue_on;
+int tmon_exit;
+static short	daemon_mode;
+static int logging; /* for recording thermal data to a file */
+static int debug_on;
+FILE *tmon_log;
+/*cooling device used for the PID controller */
+char ctrl_cdev[CDEV_NAME_SIZE] = "None";
+int target_thermal_zone; /* user selected target zone instance */
+static void	start_daemon_mode(void);
+
+pthread_t event_tid;
+pthread_mutex_t input_lock;
+void usage()
+{
+	printf("Usage: tmon [OPTION...]\n");
+	printf("  -c, --control         cooling device in control\n");
+	printf("  -d, --daemon          run as daemon, no TUI\n");
+	printf("  -g, --debug           debug message in syslog\n");
+	printf("  -h, --help            show this help message\n");
+	printf("  -l, --log             log data to /var/tmp/tmon.log\n");
+	printf("  -t, --time-interval   sampling time interval, > 1 sec.\n");
+	printf("  -v, --version         show version\n");
+	printf("  -z, --zone            target thermal zone id\n");
+
+	exit(0);
+}
+
+void version()
+{
+	printf("TMON version %s\n", VERSION);
+	exit(EXIT_SUCCESS);
+}
+
+static void tmon_cleanup(void)
+{
+
+	syslog(LOG_INFO, "TMON exit cleanup\n");
+	fflush(stdout);
+	refresh();
+	if (tmon_log)
+		fclose(tmon_log);
+	if (event_tid) {
+		pthread_mutex_lock(&input_lock);
+		pthread_cancel(event_tid);
+		pthread_mutex_unlock(&input_lock);
+		pthread_mutex_destroy(&input_lock);
+	}
+	closelog();
+	/* relax control knobs, undo throttling */
+	set_ctrl_state(0);
+
+	keypad(stdscr, FALSE);
+	echo();
+	nocbreak();
+	close_windows();
+	endwin();
+	free_thermal_data();
+
+	exit(1);
+}
+
+
+static void tmon_sig_handler(int sig)
+{
+	syslog(LOG_INFO, "TMON caught signal %d\n", sig);
+	refresh();
+	switch (sig) {
+	case SIGTERM:
+		printf("sigterm, exit and clean up\n");
+		fflush(stdout);
+		break;
+	case SIGKILL:
+		printf("sigkill, exit and clean up\n");
+		fflush(stdout);
+		break;
+	case SIGINT:
+		printf("ctrl-c, exit and clean up\n");
+		fflush(stdout);
+		break;
+	default:
+		break;
+	}
+	tmon_exit = true;
+}
+
+
+static void start_syslog(void)
+{
+	if (debug_on)
+		setlogmask(LOG_UPTO(LOG_DEBUG));
+	else
+		setlogmask(LOG_UPTO(LOG_ERR));
+	openlog("tmon.log", LOG_CONS | LOG_PID | LOG_NDELAY, LOG_LOCAL0);
+	syslog(LOG_NOTICE, "TMON started by User %d", getuid());
+}
+
+static void prepare_logging(void)
+{
+	int i;
+
+	if (!logging)
+		return;
+	/* open local data log file */
+	tmon_log = fopen(TMON_LOG_FILE, "w+");
+	if (!tmon_log) {
+		syslog(LOG_ERR, "failed to open log file %s\n", TMON_LOG_FILE);
+		return;
+	}
+
+	fprintf(tmon_log, "#----------- THERMAL SYSTEM CONFIG -------------\n");
+	for (i = 0; i < ptdata.nr_tz_sensor; i++) {
+		char binding_str[33]; /* size of long + 1 */
+		int j;
+
+		memset(binding_str, 0, sizeof(binding_str));
+		for (j = 0; j < 32; j++)
+			binding_str[j] = (ptdata.tzi[i].cdev_binding & 1<<j) ?
+				'1' : '0';
+
+		fprintf(tmon_log, "#thermal zone %s%02d cdevs binding: %32s\n",
+			ptdata.tzi[i].type,
+			ptdata.tzi[i].instance,
+			binding_str);
+		for (j = 0; j <	ptdata.tzi[i].nr_trip_pts; j++) {
+			fprintf(tmon_log, "#\tTP%02d type:%s, temp:%lu\n", j,
+				trip_type_name[ptdata.tzi[i].tp[j].type],
+				ptdata.tzi[i].tp[j].temp);
+		}
+
+	}
+
+	for (i = 0; i <	ptdata.nr_cooling_dev; i++)
+		fprintf(tmon_log, "#cooling devices%02d: %s\n",
+			i, ptdata.cdi[i].type);
+
+	fprintf(tmon_log, "#---------- THERMAL DATA LOG STARTED -----------\n");
+	fprintf(tmon_log, "Samples TargetTemp ");
+	for (i = 0; i < ptdata.nr_tz_sensor; i++) {
+		fprintf(tmon_log, "%s%d    ", ptdata.tzi[i].type,
+			ptdata.tzi[i].instance);
+	}
+	for (i = 0; i <	ptdata.nr_cooling_dev; i++)
+		fprintf(tmon_log, "%s%d ", ptdata.cdi[i].type,
+			ptdata.cdi[i].instance);
+
+	fprintf(tmon_log, "\n");
+}
+
+static struct option opts[] = {
+	{ "control", 1, NULL, 'c' },
+	{ "daemon", 0, NULL, 'd' },
+	{ "time-interval", 1, NULL, 't' },
+	{ "log", 0, NULL, 'l' },
+	{ "help", 0, NULL, 'h' },
+	{ "version", 0, NULL, 'v' },
+	{ "debug", 0, NULL, 'g' },
+	{ 0, 0, NULL, 0 }
+};
+
+
+int main(int argc, char **argv)
+{
+	int err = 0;
+	int id2 = 0, c;
+	double yk = 0.0; /* controller output */
+	int target_tz_index;
+
+	if (geteuid() != 0) {
+		printf("TMON needs to be run as root\n");
+		exit(EXIT_FAILURE);
+	}
+
+	while ((c = getopt_long(argc, argv, "c:dlht:vgz:", opts, &id2)) != -1) {
+		switch (c) {
+		case 'c':
+			no_control = 0;
+			strncpy(ctrl_cdev, optarg, CDEV_NAME_SIZE);
+			break;
+		case 'd':
+			start_daemon_mode();
+			printf("Run TMON in daemon mode\n");
+			break;
+		case 't':
+			ticktime = strtod(optarg, NULL);
+			if (ticktime < 1)
+				ticktime = 1;
+			break;
+		case 'l':
+			printf("Logging data to /var/tmp/tmon.log\n");
+			logging = 1;
+			break;
+		case 'h':
+			usage();
+			break;
+		case 'v':
+			version();
+			break;
+		case 'g':
+			debug_on = 1;
+			break;
+		case 'z':
+			target_thermal_zone = strtod(optarg, NULL);
+			break;
+		default:
+			break;
+		}
+	}
+	if (pthread_mutex_init(&input_lock, NULL) != 0) {
+		fprintf(stderr, "\n mutex init failed, exit\n");
+		return 1;
+	}
+	start_syslog();
+	if (signal(SIGINT, tmon_sig_handler) == SIG_ERR)
+		syslog(LOG_DEBUG, "Cannot handle SIGINT\n");
+	if (signal(SIGTERM, tmon_sig_handler) == SIG_ERR)
+		syslog(LOG_DEBUG, "Cannot handle SIGINT\n");
+
+	if (probe_thermal_sysfs()) {
+		pthread_mutex_destroy(&input_lock);
+		closelog();
+		return -1;
+	}
+	initialize_curses();
+	setup_windows();
+	signal(SIGWINCH, resize_handler);
+	show_title_bar();
+	show_sensors_w();
+	show_cooling_device();
+	update_thermal_data();
+	show_data_w();
+	prepare_logging();
+	init_thermal_controller();
+
+	nodelay(stdscr, TRUE);
+	err = pthread_create(&event_tid, NULL, &handle_tui_events, NULL);
+	if (err != 0) {
+		printf("\ncan't create thread :[%s]", strerror(err));
+		tmon_cleanup();
+		exit(EXIT_FAILURE);
+	}
+
+	/* validate range of user selected target zone, default to the first
+	 * instance if out of range
+	 */
+	target_tz_index = zone_instance_to_index(target_thermal_zone);
+	if (target_tz_index < 0) {
+		target_thermal_zone = ptdata.tzi[0].instance;
+		syslog(LOG_ERR, "target zone is not found, default to %d\n",
+			target_thermal_zone);
+	}
+	while (1) {
+		sleep(ticktime);
+		show_title_bar();
+		show_sensors_w();
+		update_thermal_data();
+		if (!dialogue_on) {
+			show_data_w();
+			show_cooling_device();
+		}
+		cur_thermal_record++;
+		time_elapsed += ticktime;
+		controller_handler(trec[0].temp[target_tz_index] / 1000,
+				&yk);
+		trec[0].pid_out_pct = yk;
+		if (!dialogue_on)
+			show_control_w();
+		if (tmon_exit)
+			break;
+	}
+	tmon_cleanup();
+	return 0;
+}
+
+static void start_daemon_mode()
+{
+	daemon_mode = 1;
+	/* fork */
+	pid_t	sid, pid = fork();
+	if (pid < 0) {
+		exit(EXIT_FAILURE);
+	} else if (pid > 0)
+		/* kill parent */
+		exit(EXIT_SUCCESS);
+
+	/* disable TUI, it may not be necessary, but saves some resource */
+	disable_tui();
+
+	/* change the file mode mask */
+	umask(0);
+
+	/* new SID for the daemon process */
+	sid = setsid();
+	if (sid < 0)
+		exit(EXIT_FAILURE);
+
+	/* change working directory */
+	if ((chdir("/")) < 0)
+		exit(EXIT_FAILURE);
+
+
+	sleep(10);
+
+	close(STDIN_FILENO);
+	close(STDOUT_FILENO);
+	close(STDERR_FILENO);
+
+}
diff --git a/tools/thermal/tmon/tmon.h b/tools/thermal/tmon/tmon.h
new file mode 100644
index 000000000000..9e3c49c547ac
--- /dev/null
+++ b/tools/thermal/tmon/tmon.h
@@ -0,0 +1,204 @@
+/*
+ * tmon.h contains data structures and constants used by TMON
+ *
+ * Copyright (C) 2012 Intel Corporation. All rights reserved.
+ *
+ * This program is free software; you can redistribute it and/or
+ * modify it under the terms of the GNU General Public License version
+ * 2 or later as published by the Free Software Foundation.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+ * GNU General Public License for more details.
+ *
+ * Author Name Jacob Pan <jacob.jun.pan@linux.intel.com>
+ *
+ */
+
+#ifndef TMON_H
+#define TMON_H
+
+#define MAX_DISP_TEMP 125
+#define MAX_CTRL_TEMP 105
+#define MIN_CTRL_TEMP 40
+#define MAX_NR_TZONE 16
+#define MAX_NR_CDEV 32
+#define MAX_NR_TRIP 16
+#define MAX_NR_CDEV_TRIP 12 /* number of cooling devices that can bind
+			     * to a thermal zone trip.
+			     */
+#define MAX_TEMP_KC 140000
+/* starting char position to draw sensor data, such as tz names
+ * trip point list, etc.
+ */
+#define DATA_LEFT_ALIGN 10
+#define NR_LINES_TZDATA 1
+#define TMON_LOG_FILE "/var/tmp/tmon.log"
+
+extern unsigned long ticktime;
+extern double time_elapsed;
+extern unsigned long target_temp_user;
+extern int dialogue_on;
+extern char ctrl_cdev[];
+extern pthread_mutex_t input_lock;
+extern int tmon_exit;
+extern int target_thermal_zone;
+/* use fixed size record to simplify data processing and transfer
+ * TBD: more info to be added, e.g. programmable trip point data.
+*/
+struct thermal_data_record {
+	struct timeval tv;
+	unsigned long temp[MAX_NR_TZONE];
+	double pid_out_pct;
+};
+
+struct cdev_info {
+	char type[64];
+	int instance;
+	unsigned long max_state;
+	unsigned long cur_state;
+	unsigned long flag;
+};
+
+enum trip_type {
+	THERMAL_TRIP_CRITICAL,
+	THERMAL_TRIP_HOT,
+	THERMAL_TRIP_PASSIVE,
+	THERMAL_TRIP_ACTIVE,
+	NR_THERMAL_TRIP_TYPE,
+};
+
+struct trip_point {
+	enum trip_type type;
+	unsigned long temp;
+	unsigned long hysteresis;
+	int attribute; /* programmability etc. */
+};
+
+/* thermal zone configuration information, binding with cooling devices could
+ * change at runtime.
+ */
+struct tz_info {
+	char type[256]; /* e.g. acpitz */
+	int instance;
+	int passive; /* active zone has passive node to force passive mode */
+	int nr_cdev; /* number of cooling device binded */
+	int nr_trip_pts;
+	struct trip_point tp[MAX_NR_TRIP];
+	unsigned long cdev_binding; /* bitmap for attached cdevs */
+	/* cdev bind trip points, allow one cdev bind to multiple trips */
+	unsigned long trip_binding[MAX_NR_CDEV];
+};
+
+struct tmon_platform_data {
+	int nr_tz_sensor;
+	int nr_cooling_dev;
+	/* keep track of instance ids since there might be gaps */
+	int max_tz_instance;
+	int max_cdev_instance;
+	struct tz_info *tzi;
+	struct cdev_info *cdi;
+};
+
+struct control_ops {
+	void (*set_ratio)(unsigned long ratio);
+	unsigned long (*get_ratio)(unsigned long ratio);
+
+};
+
+enum cdev_types {
+	CDEV_TYPE_PROC,
+	CDEV_TYPE_FAN,
+	CDEV_TYPE_MEM,
+	CDEV_TYPE_NR,
+};
+
+/* REVISIT: the idea is to group sensors if possible, e.g. on intel mid
+ * we have "skin0", "skin1", "sys", "msicdie"
+ * on DPTF enabled systems, we might have PCH, TSKN, TAMB, etc.
+ */
+enum tzone_types {
+	TZONE_TYPE_ACPI,
+	TZONE_TYPE_PCH,
+	TZONE_TYPE_NR,
+};
+
+/* limit the output of PID controller adjustment */
+#define LIMIT_HIGH (95)
+#define LIMIT_LOW  (2)
+
+struct pid_params {
+	double kp;  /* Controller gain from Dialog Box */
+	double ki;  /* Time-constant for I action from Dialog Box */
+	double kd;  /* Time-constant for D action from Dialog Box */
+	double ts;
+	double k_lpf;
+
+	double t_target;
+	double y_k;
+};
+
+extern int init_thermal_controller(void);
+extern void controller_handler(const double xk, double *yk);
+
+extern struct tmon_platform_data ptdata;
+extern struct pid_params p_param;
+
+extern FILE *tmon_log;
+extern int cur_thermal_record; /* index to the trec array */
+extern struct thermal_data_record trec[];
+extern const char *trip_type_name[];
+extern unsigned long no_control;
+
+extern void initialize_curses(void);
+extern void show_controller_stats(char *line);
+extern void show_title_bar(void);
+extern void setup_windows(void);
+extern void disable_tui(void);
+extern void show_sensors_w(void);
+extern void show_data_w(void);
+extern void write_status_bar(int x, char *line);
+extern void show_control_w();
+
+extern void show_cooling_device(void);
+extern void show_dialogue(void);
+extern int update_thermal_data(void);
+
+extern int probe_thermal_sysfs(void);
+extern void free_thermal_data(void);
+extern	void resize_handler(int sig);
+extern void set_ctrl_state(unsigned long state);
+extern void get_ctrl_state(unsigned long *state);
+extern void *handle_tui_events(void *arg);
+extern int sysfs_set_ulong(char *path, char *filename, unsigned long val);
+extern int zone_instance_to_index(int zone_inst);
+extern void close_windows(void);
+
+#define PT_COLOR_DEFAULT    1
+#define PT_COLOR_HEADER_BAR 2
+#define PT_COLOR_ERROR      3
+#define PT_COLOR_RED        4
+#define PT_COLOR_YELLOW     5
+#define PT_COLOR_GREEN      6
+#define PT_COLOR_BRIGHT     7
+#define PT_COLOR_BLUE	    8
+
+/* each thermal zone uses 12 chars, 8 for name, 2 for instance, 2 space
+ * also used to list trip points in forms of AAAC, which represents
+ * A: Active
+ * C: Critical
+ */
+#define TZONE_RECORD_SIZE 12
+#define TZ_LEFT_ALIGN 32
+#define CDEV_NAME_SIZE 20
+#define CDEV_FLAG_IN_CONTROL (1 << 0)
+
+/* dialogue box starts */
+#define DIAG_X 48
+#define DIAG_Y 8
+#define THERMAL_SYSFS "/sys/class/thermal"
+#define CDEV "cooling_device"
+#define TZONE "thermal_zone"
+#define TDATA_LEFT 16
+#endif /* TMON_H */
diff --git a/tools/thermal/tmon/tui.c b/tools/thermal/tmon/tui.c
new file mode 100644
index 000000000000..89f8ef0e15c8
--- /dev/null
+++ b/tools/thermal/tmon/tui.c
@@ -0,0 +1,638 @@
+/*
+ * tui.c ncurses text user interface for TMON program
+ *
+ * Copyright (C) 2013 Intel Corporation. All rights reserved.
+ *
+ * This program is free software; you can redistribute it and/or
+ * modify it under the terms of the GNU General Public License version
+ * 2 or later as published by the Free Software Foundation.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+ * GNU General Public License for more details.
+ *
+ * Author: Jacob Pan <jacob.jun.pan@linux.intel.com>
+ *
+ */
+
+#include <unistd.h>
+#include <stdio.h>
+#include <stdlib.h>
+#include <string.h>
+#include <stdint.h>
+#include <ncurses.h>
+#include <time.h>
+#include <syslog.h>
+#include <panel.h>
+#include <pthread.h>
+#include <signal.h>
+
+#include "tmon.h"
+
+static PANEL *data_panel;
+static PANEL *dialogue_panel;
+static PANEL *top;
+
+static WINDOW *title_bar_window;
+static WINDOW *tz_sensor_window;
+static WINDOW *cooling_device_window;
+static WINDOW *control_window;
+static WINDOW *status_bar_window;
+static WINDOW *thermal_data_window;
+static WINDOW *dialogue_window;
+
+char status_bar_slots[10][40];
+static void draw_hbar(WINDOW *win, int y, int start, int len,
+		unsigned long pattern, bool end);
+
+static int maxx, maxy;
+static int maxwidth = 200;
+
+#define TITLE_BAR_HIGHT 1
+#define SENSOR_WIN_HIGHT 4 /* one row for tz name, one for trip points */
+
+
+/* daemon mode flag (set by startup parameter -d) */
+static int  tui_disabled;
+
+static void close_panel(PANEL *p)
+{
+	if (p) {
+		del_panel(p);
+		p = NULL;
+	}
+}
+
+static void close_window(WINDOW *win)
+{
+	if (win) {
+		delwin(win);
+		win = NULL;
+	}
+}
+
+void close_windows(void)
+{
+	if (tui_disabled)
+		return;
+	/* must delete panels before their attached windows */
+	if (dialogue_window)
+		close_panel(dialogue_panel);
+	if (cooling_device_window)
+		close_panel(data_panel);
+
+	close_window(title_bar_window);
+	close_window(tz_sensor_window);
+	close_window(status_bar_window);
+	close_window(cooling_device_window);
+	close_window(control_window);
+	close_window(thermal_data_window);
+	close_window(dialogue_window);
+
+}
+
+void write_status_bar(int x, char *line)
+{
+	mvwprintw(status_bar_window, 0, x, "%s", line);
+	wrefresh(status_bar_window);
+}
+
+void setup_windows(void)
+{
+	int y_begin = 1;
+
+	if (tui_disabled)
+		return;
+
+	getmaxyx(stdscr, maxy, maxx);
+	resizeterm(maxy, maxx);
+
+	title_bar_window = subwin(stdscr, TITLE_BAR_HIGHT, maxx, 0, 0);
+	y_begin += TITLE_BAR_HIGHT;
+
+	tz_sensor_window = subwin(stdscr, SENSOR_WIN_HIGHT, maxx, y_begin, 0);
+	y_begin += SENSOR_WIN_HIGHT;
+
+	cooling_device_window = subwin(stdscr, ptdata.nr_cooling_dev + 3, maxx,
+				y_begin, 0);
+	y_begin += ptdata.nr_cooling_dev + 3; /* 2 lines for border */
+	/* two lines to show borders, one line per tz show trip point position
+	 * and value.
+	 * dialogue window is a pop-up, when needed it lays on top of cdev win
+	 */
+
+	dialogue_window = subwin(stdscr, ptdata.nr_cooling_dev+5, maxx-50,
+				DIAG_Y, DIAG_X);
+
+	thermal_data_window = subwin(stdscr, ptdata.nr_tz_sensor *
+				NR_LINES_TZDATA + 3, maxx, y_begin, 0);
+	y_begin += ptdata.nr_tz_sensor * NR_LINES_TZDATA + 3;
+	control_window = subwin(stdscr, 4, maxx, y_begin, 0);
+
+	scrollok(cooling_device_window, TRUE);
+	maxwidth = maxx - 18;
+	status_bar_window = subwin(stdscr, 1, maxx, maxy-1, 0);
+
+	strcpy(status_bar_slots[0], " Ctrl-c - Quit ");
+	strcpy(status_bar_slots[1], " TAB - Tuning ");
+	wmove(status_bar_window, 1, 30);
+
+	/* prepare panels for dialogue, if panel already created then we must
+	 * be doing resizing, so just replace windows with new ones, old ones
+	 * should have been deleted by close_window
+	 */
+	data_panel = new_panel(cooling_device_window);
+	if (!data_panel)
+		syslog(LOG_DEBUG, "No data panel\n");
+	else {
+		if (dialogue_window) {
+			dialogue_panel = new_panel(dialogue_window);
+			if (!dialogue_panel)
+				syslog(LOG_DEBUG, "No dialogue panel\n");
+			else {
+				/* Set up the user pointer to the next panel*/
+				set_panel_userptr(data_panel, dialogue_panel);
+				set_panel_userptr(dialogue_panel, data_panel);
+				top = data_panel;
+			}
+		} else
+			syslog(LOG_INFO, "no dialogue win, term too small\n");
+	}
+	doupdate();
+	werase(stdscr);
+	refresh();
+}
+
+void resize_handler(int sig)
+{
+	/* start over when term gets resized, but first we clean up */
+	close_windows();
+	endwin();
+	refresh();
+	clear();
+	getmaxyx(stdscr, maxy, maxx);  /* get the new screen size */
+	setup_windows();
+	/* rate limit */
+	sleep(1);
+	syslog(LOG_DEBUG, "SIG %d, term resized to %d x %d\n",
+		sig, maxy, maxx);
+	signal(SIGWINCH, resize_handler);
+}
+
+const char cdev_title[] = " COOLING DEVICES ";
+void show_cooling_device(void)
+{
+	int i, j, x, y = 0;
+
+	if (tui_disabled || !cooling_device_window)
+		return;
+
+	werase(cooling_device_window);
+	wattron(cooling_device_window, A_BOLD);
+	mvwprintw(cooling_device_window,  1, 1,
+		"ID  Cooling Dev   Cur    Max   Thermal Zone Binding");
+	wattroff(cooling_device_window, A_BOLD);
+	for (j = 0; j <	ptdata.nr_cooling_dev; j++) {
+		/* draw cooling device list on the left in the order of
+		 * cooling device instances. skip unused idr.
+		 */
+		mvwprintw(cooling_device_window, j + 2, 1,
+			"%02d %12.12s%6d %6d",
+			ptdata.cdi[j].instance,
+			ptdata.cdi[j].type,
+			ptdata.cdi[j].cur_state,
+			ptdata.cdi[j].max_state);
+	}
+
+	/* show cdev binding, y is the global cooling device instance */
+	for (i = 0; i < ptdata.nr_tz_sensor; i++) {
+		int tz_inst = ptdata.tzi[i].instance;
+		for (j = 0; j < ptdata.nr_cooling_dev; j++) {
+			int cdev_inst;
+			y = j;
+			x = tz_inst * TZONE_RECORD_SIZE + TZ_LEFT_ALIGN;
+
+			draw_hbar(cooling_device_window, y+2, x,
+				TZONE_RECORD_SIZE-1, ACS_VLINE, false);
+
+			/* draw a column of spaces to separate thermal zones */
+			mvwprintw(cooling_device_window, y+2, x-1, " ");
+			if (ptdata.tzi[i].cdev_binding) {
+				cdev_inst = ptdata.cdi[j].instance;
+				unsigned long trip_binding =
+					ptdata.tzi[i].trip_binding[cdev_inst];
+				int k = 0; /* per zone trip point id that
+					    * binded to this cdev, one to
+					    * many possible based on the
+					    * binding bitmask.
+					    */
+				syslog(LOG_DEBUG,
+					"bind tz%d cdev%d tp%lx %d cdev%lx\n",
+					i, j, trip_binding, y,
+					ptdata.tzi[i].cdev_binding);
+				/* draw each trip binding for the cdev */
+				while (trip_binding >>= 1) {
+					k++;
+					if (!(trip_binding & 1))
+						continue;
+					/* draw '*' to show binding */
+					mvwprintw(cooling_device_window,
+						y + 2,
+						x + ptdata.tzi[i].nr_trip_pts -
+						k - 1, "*");
+				}
+			}
+		}
+	}
+	/* draw border after data so that border will not be messed up
+	 * even there is not enough space for all the data to be shown
+	 */
+	wborder(cooling_device_window, 0, 0, 0, 0, 0, 0, 0, 0);
+	wattron(cooling_device_window, A_BOLD);
+	mvwprintw(cooling_device_window, 0, maxx/2 - sizeof(cdev_title),
+		cdev_title);
+	wattroff(cooling_device_window, A_BOLD);
+
+	wrefresh(cooling_device_window);
+}
+
+const char DIAG_TITLE[] = "[ TUNABLES ]";
+#define DIAG_DEV_ROWS  5
+void show_dialogue(void)
+{
+	int j, x = 0, y = 0;
+	WINDOW *w = dialogue_window;
+
+	if (tui_disabled || !w)
+		return;
+
+	werase(w);
+	box(w, 0, 0);
+	mvwprintw(w, 0, maxx/4, DIAG_TITLE);
+	/* list all the available tunables */
+	for (j = 0; j <= ptdata.nr_cooling_dev; j++) {
+		y = j % DIAG_DEV_ROWS;
+		if (y == 0 && j != 0)
+			x += 20;
+		if (j == ptdata.nr_cooling_dev)
+			/* save last choice for target temp */
+			mvwprintw(w, y+1, x+1, "%C-%.12s", 'A'+j, "Set Temp");
+		else
+			mvwprintw(w, y+1, x+1, "%C-%.10s-%2d", 'A'+j,
+				ptdata.cdi[j].type, ptdata.cdi[j].instance);
+	}
+	wattron(w, A_BOLD);
+	mvwprintw(w, DIAG_DEV_ROWS+1, 1, "Enter Choice [A-Z]?");
+	wattroff(w, A_BOLD);
+	/* y size of dialogue win is nr cdev + 5, so print legend
+	 * at the bottom line
+	 */
+	mvwprintw(w, ptdata.nr_cooling_dev+3, 1,
+		"Legend: A=Active, P=Passive, C=Critical");
+
+	wrefresh(dialogue_window);
+}
+
+void write_dialogue_win(char *buf, int y, int x)
+{
+	WINDOW *w = dialogue_window;
+
+	mvwprintw(w, y, x, "%s", buf);
+}
+
+const char control_title[] = " CONTROLS ";
+void show_control_w(void)
+{
+	unsigned long state;
+
+	get_ctrl_state(&state);
+
+	if (tui_disabled || !control_window)
+		return;
+
+	werase(control_window);
+	mvwprintw(control_window, 1, 1,
+		"PID gain: kp=%2.2f ki=%2.2f kd=%2.2f Output %2.2f",
+		p_param.kp, p_param.ki, p_param.kd, p_param.y_k);
+
+	mvwprintw(control_window, 2, 1,
+		"Target Temp: %2.1fC, Zone: %d, Control Device: %.12s",
+		p_param.t_target, target_thermal_zone, ctrl_cdev);
+
+	/* draw border last such that everything is within boundary */
+	wborder(control_window, 0, 0, 0, 0, 0, 0, 0, 0);
+	wattron(control_window, A_BOLD);
+	mvwprintw(control_window, 0, maxx/2 - sizeof(control_title),
+		control_title);
+	wattroff(control_window, A_BOLD);
+
+	wrefresh(control_window);
+}
+
+void initialize_curses(void)
+{
+	if (tui_disabled)
+		return;
+
+	initscr();
+	start_color();
+	keypad(stdscr, TRUE);	/* enable keyboard mapping */
+	nonl();			/* tell curses not to do NL->CR/NL on output */
+	cbreak();		/* take input chars one at a time */
+	noecho();		/* dont echo input */
+	curs_set(0);		/* turn off cursor */
+	use_default_colors();
+
+	init_pair(PT_COLOR_DEFAULT, COLOR_WHITE, COLOR_BLACK);
+	init_pair(PT_COLOR_HEADER_BAR, COLOR_BLACK, COLOR_WHITE);
+	init_pair(PT_COLOR_ERROR, COLOR_BLACK, COLOR_RED);
+	init_pair(PT_COLOR_RED, COLOR_WHITE, COLOR_RED);
+	init_pair(PT_COLOR_YELLOW, COLOR_WHITE, COLOR_YELLOW);
+	init_pair(PT_COLOR_GREEN, COLOR_WHITE, COLOR_GREEN);
+	init_pair(PT_COLOR_BLUE, COLOR_WHITE, COLOR_BLUE);
+	init_pair(PT_COLOR_BRIGHT, COLOR_WHITE, COLOR_BLACK);
+
+}
+
+void show_title_bar(void)
+{
+	int i;
+	int x = 0;
+
+	if (tui_disabled || !title_bar_window)
+		return;
+
+	wattrset(title_bar_window, COLOR_PAIR(PT_COLOR_HEADER_BAR));
+	wbkgd(title_bar_window, COLOR_PAIR(PT_COLOR_HEADER_BAR));
+	werase(title_bar_window);
+
+	mvwprintw(title_bar_window, 0, 0,
+		"     TMON v%s", VERSION);
+
+	wrefresh(title_bar_window);
+
+	werase(status_bar_window);
+
+	for (i = 0; i < 10; i++) {
+		if (strlen(status_bar_slots[i]) == 0)
+			continue;
+		wattron(status_bar_window, A_REVERSE);
+		mvwprintw(status_bar_window, 0, x, "%s", status_bar_slots[i]);
+		wattroff(status_bar_window, A_REVERSE);
+		x += strlen(status_bar_slots[i]) + 1;
+	}
+	wrefresh(status_bar_window);
+}
+
+static void handle_input_val(int ch)
+{
+	char buf[32];
+	int val;
+	char path[256];
+	WINDOW *w = dialogue_window;
+
+	echo();
+	keypad(w, TRUE);
+	wgetnstr(w, buf, 31);
+	val = atoi(buf);
+
+	if (ch == ptdata.nr_cooling_dev) {
+		snprintf(buf, 31, "Invalid Temp %d! %d-%d", val,
+			MIN_CTRL_TEMP, MAX_CTRL_TEMP);
+		if (val < MIN_CTRL_TEMP || val > MAX_CTRL_TEMP)
+			write_status_bar(40, buf);
+		else {
+			p_param.t_target = val;
+			snprintf(buf, 31, "Set New Target Temp %d", val);
+			write_status_bar(40, buf);
+		}
+	} else {
+		snprintf(path, 256, "%s/%s%d", THERMAL_SYSFS,
+			CDEV, ptdata.cdi[ch].instance);
+		sysfs_set_ulong(path, "cur_state", val);
+	}
+	noecho();
+	dialogue_on = 0;
+	show_data_w();
+	show_control_w();
+
+	top = (PANEL *)panel_userptr(top);
+	top_panel(top);
+}
+
+static void handle_input_choice(int ch)
+{
+	char buf[48];
+	int base = 0;
+	int cdev_id = 0;
+
+	if ((ch >= 'A' && ch <= 'A' + ptdata.nr_cooling_dev) ||
+		(ch >= 'a' && ch <= 'a' + ptdata.nr_cooling_dev)) {
+		base = (ch < 'a') ? 'A' : 'a';
+		cdev_id = ch - base;
+		if (ptdata.nr_cooling_dev == cdev_id)
+			snprintf(buf, sizeof(buf), "New Target Temp:");
+		else
+			snprintf(buf, sizeof(buf), "New Value for %.10s-%2d: ",
+				ptdata.cdi[cdev_id].type,
+				ptdata.cdi[cdev_id].instance);
+		write_dialogue_win(buf, DIAG_DEV_ROWS+2, 2);
+		handle_input_val(cdev_id);
+	} else {
+		snprintf(buf, sizeof(buf), "Invalid selection %d", ch);
+		write_dialogue_win(buf, 8, 2);
+	}
+}
+
+void *handle_tui_events(void *arg)
+{
+	int ch;
+
+	keypad(cooling_device_window, TRUE);
+	while ((ch = wgetch(cooling_device_window)) != EOF) {
+		if (tmon_exit)
+			break;
+		/* when term size is too small, no dialogue panels are set.
+		 * we need to filter out such cases.
+		 */
+		if (!data_panel || !dialogue_panel ||
+			!cooling_device_window ||
+			!dialogue_window) {
+
+			continue;
+		}
+		pthread_mutex_lock(&input_lock);
+		if (dialogue_on) {
+			handle_input_choice(ch);
+			/* top panel filter */
+			if (ch == 'q' || ch == 'Q')
+				ch = 0;
+		}
+		switch (ch) {
+		case KEY_LEFT:
+			box(cooling_device_window, 10, 0);
+			break;
+		case 9: /* TAB */
+			top = (PANEL *)panel_userptr(top);
+			top_panel(top);
+			if (top == dialogue_panel) {
+				dialogue_on = 1;
+				show_dialogue();
+			} else {
+				dialogue_on = 0;
+				/* force refresh */
+				show_data_w();
+				show_control_w();
+			}
+			break;
+		case 'q':
+		case 'Q':
+			tmon_exit = 1;
+			break;
+		}
+		update_panels();
+		doupdate();
+		pthread_mutex_unlock(&input_lock);
+	}
+
+	if (arg)
+		*(int *)arg = 0; /* make gcc happy */
+
+	return NULL;
+}
+
+/* draw a horizontal bar in given pattern */
+static void draw_hbar(WINDOW *win, int y, int start, int len, unsigned long ptn,
+		bool end)
+{
+	mvwaddch(win, y, start, ptn);
+	whline(win, ptn, len);
+	if (end)
+		mvwaddch(win, y, MAX_DISP_TEMP+TDATA_LEFT, ']');
+}
+
+static char trip_type_to_char(int type)
+{
+	switch (type) {
+	case THERMAL_TRIP_CRITICAL: return 'C';
+	case THERMAL_TRIP_HOT: return 'H';
+	case THERMAL_TRIP_PASSIVE: return 'P';
+	case THERMAL_TRIP_ACTIVE: return 'A';
+	default:
+		return '?';
+	}
+}
+
+/* fill a string with trip point type and value in one line
+ * e.g.      P(56)    C(106)
+ * maintain the distance one degree per char
+ */
+static void draw_tp_line(int tz, int y)
+{
+	int j;
+	int x;
+
+	for (j = 0; j < ptdata.tzi[tz].nr_trip_pts; j++) {
+		x = ptdata.tzi[tz].tp[j].temp / 1000;
+		mvwprintw(thermal_data_window, y + 0, x + TDATA_LEFT,
+			"%c%d", trip_type_to_char(ptdata.tzi[tz].tp[j].type),
+			x);
+		syslog(LOG_INFO, "%s:tz %d tp %d temp = %lu\n", __func__,
+			tz, j, ptdata.tzi[tz].tp[j].temp);
+	}
+}
+
+const char data_win_title[] = " THERMAL DATA ";
+void show_data_w(void)
+{
+	int i;
+
+
+	if (tui_disabled || !thermal_data_window)
+		return;
+
+	werase(thermal_data_window);
+	wattron(thermal_data_window, A_BOLD);
+	mvwprintw(thermal_data_window, 0, maxx/2 - sizeof(data_win_title),
+		data_win_title);
+	wattroff(thermal_data_window, A_BOLD);
+	/* draw a line as ruler */
+	for (i = 10; i < MAX_DISP_TEMP; i += 10)
+		mvwprintw(thermal_data_window, 1, i+TDATA_LEFT, "%2d", i);
+
+	for (i = 0; i < ptdata.nr_tz_sensor; i++) {
+		int temp = trec[cur_thermal_record].temp[i] / 1000;
+		int y = 0;
+
+		y = i * NR_LINES_TZDATA + 2;
+		/* y at tz temp data line */
+		mvwprintw(thermal_data_window, y, 1, "%6.6s%2d:[%3d][",
+			ptdata.tzi[i].type,
+			ptdata.tzi[i].instance, temp);
+		draw_hbar(thermal_data_window, y, TDATA_LEFT, temp, ACS_RARROW,
+			true);
+		draw_tp_line(i, y);
+	}
+	wborder(thermal_data_window, 0, 0, 0, 0, 0, 0, 0, 0);
+	wrefresh(thermal_data_window);
+}
+
+const char tz_title[] = "THERMAL ZONES(SENSORS)";
+
+void show_sensors_w(void)
+{
+	int i, j;
+	char buffer[512];
+
+	if (tui_disabled || !tz_sensor_window)
+		return;
+
+	werase(tz_sensor_window);
+
+	memset(buffer, 0, sizeof(buffer));
+	wattron(tz_sensor_window, A_BOLD);
+	mvwprintw(tz_sensor_window, 1, 1, "Thermal Zones:");
+	wattroff(tz_sensor_window, A_BOLD);
+
+	mvwprintw(tz_sensor_window, 1, TZ_LEFT_ALIGN, "%s", buffer);
+	/* fill trip points for each tzone */
+	wattron(tz_sensor_window, A_BOLD);
+	mvwprintw(tz_sensor_window, 2, 1, "Trip Points:");
+	wattroff(tz_sensor_window, A_BOLD);
+
+	/* draw trip point from low to high for each tz */
+	for (i = 0; i < ptdata.nr_tz_sensor; i++) {
+		int inst = ptdata.tzi[i].instance;
+
+		mvwprintw(tz_sensor_window, 1,
+			TZ_LEFT_ALIGN+TZONE_RECORD_SIZE * inst, "%.9s%02d",
+			ptdata.tzi[i].type, ptdata.tzi[i].instance);
+		for (j = ptdata.tzi[i].nr_trip_pts - 1; j >= 0; j--) {
+			/* loop through all trip points */
+			char type;
+			int tp_pos;
+			/* reverse the order here since trips are sorted
+			 * in ascending order in terms of temperature.
+			 */
+			tp_pos = ptdata.tzi[i].nr_trip_pts - j - 1;
+
+			type = trip_type_to_char(ptdata.tzi[i].tp[j].type);
+			mvwaddch(tz_sensor_window, 2,
+				inst * TZONE_RECORD_SIZE + TZ_LEFT_ALIGN +
+				tp_pos,	type);
+			syslog(LOG_DEBUG, "draw tz %d tp %d ch:%c\n",
+				inst, j, type);
+		}
+	}
+	wborder(tz_sensor_window, 0, 0, 0, 0, 0, 0, 0, 0);
+	wattron(tz_sensor_window, A_BOLD);
+	mvwprintw(tz_sensor_window, 0, maxx/2 - sizeof(tz_title), tz_title);
+	wattroff(tz_sensor_window, A_BOLD);
+	wrefresh(tz_sensor_window);
+}
+
+void disable_tui(void)
+{
+	tui_disabled = 1;
+}
diff --git a/tools/virtio/virtio_test.c b/tools/virtio/virtio_test.c
index da7a19558281..bdb71a26ae35 100644
--- a/tools/virtio/virtio_test.c
+++ b/tools/virtio/virtio_test.c
@@ -41,13 +41,14 @@ struct vdev_info {
 	struct vhost_memory *mem;
 };
 
-void vq_notify(struct virtqueue *vq)
+bool vq_notify(struct virtqueue *vq)
 {
 	struct vq_info *info = vq->priv;
 	unsigned long long v = 1;
 	int r;
 	r = write(info->kick, &v, sizeof v);
 	assert(r == sizeof v);
+	return true;
 }
 
 void vq_callback(struct virtqueue *vq)
@@ -171,7 +172,8 @@ static void run_test(struct vdev_info *dev, struct vq_info *vq,
 							 GFP_ATOMIC);
 				if (likely(r == 0)) {
 					++started;
-					virtqueue_kick(vq->vq);
+					if (unlikely(!virtqueue_kick(vq->vq))
+						r = -1;
 				}
 			} else
 				r = -1;
diff --git a/tools/virtio/vringh_test.c b/tools/virtio/vringh_test.c
index d053ea40c001..14a4f4cab5b9 100644
--- a/tools/virtio/vringh_test.c
+++ b/tools/virtio/vringh_test.c
@@ -22,7 +22,7 @@ static u64 user_addr_offset;
 #define RINGSIZE 256
 #define ALIGN 4096
 
-static void never_notify_host(struct virtqueue *vq)
+static bool never_notify_host(struct virtqueue *vq)
 {
 	abort();
 }
@@ -65,17 +65,22 @@ struct guest_virtio_device {
 	unsigned long notifies;
 };
 
-static void parallel_notify_host(struct virtqueue *vq)
+static bool parallel_notify_host(struct virtqueue *vq)
 {
+	int rc;
 	struct guest_virtio_device *gvdev;
 
 	gvdev = container_of(vq->vdev, struct guest_virtio_device, vdev);
-	write(gvdev->to_host_fd, "", 1);
+	rc = write(gvdev->to_host_fd, "", 1);
+	if (rc < 0)
+		return false;
 	gvdev->notifies++;
+	return true;
 }
 
-static void no_notify_host(struct virtqueue *vq)
+static bool no_notify_host(struct virtqueue *vq)
 {
+	return true;
 }
 
 #define NUM_XFERS (10000000)
diff --git a/virt/kvm/Kconfig b/virt/kvm/Kconfig
index 779262f59e25..fbe1a48bd629 100644
--- a/virt/kvm/Kconfig
+++ b/virt/kvm/Kconfig
@@ -27,3 +27,6 @@ config HAVE_KVM_MSI
 
 config HAVE_KVM_CPU_RELAX_INTERCEPT
        bool
+
+config KVM_VFIO
+       bool
diff --git a/virt/kvm/async_pf.c b/virt/kvm/async_pf.c
index 8a39dda7a325..8631d9c14320 100644
--- a/virt/kvm/async_pf.c
+++ b/virt/kvm/async_pf.c
@@ -56,7 +56,6 @@ void kvm_async_pf_vcpu_init(struct kvm_vcpu *vcpu)
 
 static void async_pf_execute(struct work_struct *work)
 {
-	struct page *page = NULL;
 	struct kvm_async_pf *apf =
 		container_of(work, struct kvm_async_pf, work);
 	struct mm_struct *mm = apf->mm;
@@ -68,14 +67,12 @@ static void async_pf_execute(struct work_struct *work)
 
 	use_mm(mm);
 	down_read(&mm->mmap_sem);
-	get_user_pages(current, mm, addr, 1, 1, 0, &page, NULL);
+	get_user_pages(current, mm, addr, 1, 1, 0, NULL, NULL);
 	up_read(&mm->mmap_sem);
 	unuse_mm(mm);
 
 	spin_lock(&vcpu->async_pf.lock);
 	list_add_tail(&apf->link, &vcpu->async_pf.done);
-	apf->page = page;
-	apf->done = true;
 	spin_unlock(&vcpu->async_pf.lock);
 
 	/*
@@ -83,7 +80,7 @@ static void async_pf_execute(struct work_struct *work)
 	 * this point
 	 */
 
-	trace_kvm_async_pf_completed(addr, page, gva);
+	trace_kvm_async_pf_completed(addr, gva);
 
 	if (waitqueue_active(&vcpu->wq))
 		wake_up_interruptible(&vcpu->wq);
@@ -99,9 +96,8 @@ void kvm_clear_async_pf_completion_queue(struct kvm_vcpu *vcpu)
 		struct kvm_async_pf *work =
 			list_entry(vcpu->async_pf.queue.next,
 				   typeof(*work), queue);
-		cancel_work_sync(&work->work);
 		list_del(&work->queue);
-		if (!work->done) { /* work was canceled */
+		if (cancel_work_sync(&work->work)) {
 			mmdrop(work->mm);
 			kvm_put_kvm(vcpu->kvm); /* == work->vcpu->kvm */
 			kmem_cache_free(async_pf_cache, work);
@@ -114,8 +110,6 @@ void kvm_clear_async_pf_completion_queue(struct kvm_vcpu *vcpu)
 			list_entry(vcpu->async_pf.done.next,
 				   typeof(*work), link);
 		list_del(&work->link);
-		if (!is_error_page(work->page))
-			kvm_release_page_clean(work->page);
 		kmem_cache_free(async_pf_cache, work);
 	}
 	spin_unlock(&vcpu->async_pf.lock);
@@ -135,14 +129,11 @@ void kvm_check_async_pf_completion(struct kvm_vcpu *vcpu)
 		list_del(&work->link);
 		spin_unlock(&vcpu->async_pf.lock);
 
-		if (work->page)
-			kvm_arch_async_page_ready(vcpu, work);
+		kvm_arch_async_page_ready(vcpu, work);
 		kvm_arch_async_page_present(vcpu, work);
 
 		list_del(&work->queue);
 		vcpu->async_pf.queued--;
-		if (!is_error_page(work->page))
-			kvm_release_page_clean(work->page);
 		kmem_cache_free(async_pf_cache, work);
 	}
 }
@@ -165,8 +156,7 @@ int kvm_setup_async_pf(struct kvm_vcpu *vcpu, gva_t gva, gfn_t gfn,
 	if (!work)
 		return 0;
 
-	work->page = NULL;
-	work->done = false;
+	work->wakeup_all = false;
 	work->vcpu = vcpu;
 	work->gva = gva;
 	work->addr = gfn_to_hva(vcpu->kvm, gfn);
@@ -206,7 +196,7 @@ int kvm_async_pf_wakeup_all(struct kvm_vcpu *vcpu)
 	if (!work)
 		return -ENOMEM;
 
-	work->page = KVM_ERR_PTR_BAD_PAGE;
+	work->wakeup_all = true;
 	INIT_LIST_HEAD(&work->queue); /* for list_del to work */
 
 	spin_lock(&vcpu->async_pf.lock);
diff --git a/virt/kvm/iommu.c b/virt/kvm/iommu.c
index 72a130bc448a..0df7d4b34dfe 100644
--- a/virt/kvm/iommu.c
+++ b/virt/kvm/iommu.c
@@ -79,7 +79,7 @@ int kvm_iommu_map_pages(struct kvm *kvm, struct kvm_memory_slot *slot)
 	flags = IOMMU_READ;
 	if (!(slot->flags & KVM_MEM_READONLY))
 		flags |= IOMMU_WRITE;
-	if (kvm->arch.iommu_flags & KVM_IOMMU_CACHE_COHERENCY)
+	if (!kvm->arch.iommu_noncoherent)
 		flags |= IOMMU_CACHE;
 
 
@@ -103,6 +103,10 @@ int kvm_iommu_map_pages(struct kvm *kvm, struct kvm_memory_slot *slot)
 		while ((gfn << PAGE_SHIFT) & (page_size - 1))
 			page_size >>= 1;
 
+		/* Make sure hva is aligned to the page size we want to map */
+		while (__gfn_to_hva_memslot(slot, gfn) & (page_size - 1))
+			page_size >>= 1;
+
 		/*
 		 * Pin all pages we are about to map in memory. This is
 		 * important because we unmap and unpin in 4kb steps later.
@@ -140,6 +144,9 @@ static int kvm_iommu_map_memslots(struct kvm *kvm)
 	struct kvm_memslots *slots;
 	struct kvm_memory_slot *memslot;
 
+	if (kvm->arch.iommu_noncoherent)
+		kvm_arch_register_noncoherent_dma(kvm);
+
 	idx = srcu_read_lock(&kvm->srcu);
 	slots = kvm_memslots(kvm);
 
@@ -158,7 +165,8 @@ int kvm_assign_device(struct kvm *kvm,
 {
 	struct pci_dev *pdev = NULL;
 	struct iommu_domain *domain = kvm->arch.iommu_domain;
-	int r, last_flags;
+	int r;
+	bool noncoherent;
 
 	/* check if iommu exists and in use */
 	if (!domain)
@@ -174,15 +182,13 @@ int kvm_assign_device(struct kvm *kvm,
 		return r;
 	}
 
-	last_flags = kvm->arch.iommu_flags;
-	if (iommu_domain_has_cap(kvm->arch.iommu_domain,
-				 IOMMU_CAP_CACHE_COHERENCY))
-		kvm->arch.iommu_flags |= KVM_IOMMU_CACHE_COHERENCY;
+	noncoherent = !iommu_domain_has_cap(kvm->arch.iommu_domain,
+					    IOMMU_CAP_CACHE_COHERENCY);
 
 	/* Check if need to update IOMMU page table for guest memory */
-	if ((last_flags ^ kvm->arch.iommu_flags) ==
-			KVM_IOMMU_CACHE_COHERENCY) {
+	if (noncoherent != kvm->arch.iommu_noncoherent) {
 		kvm_iommu_unmap_memslots(kvm);
+		kvm->arch.iommu_noncoherent = noncoherent;
 		r = kvm_iommu_map_memslots(kvm);
 		if (r)
 			goto out_unmap;
@@ -190,11 +196,7 @@ int kvm_assign_device(struct kvm *kvm,
 
 	pdev->dev_flags |= PCI_DEV_FLAGS_ASSIGNED;
 
-	printk(KERN_DEBUG "assign device %x:%x:%x.%x\n",
-		assigned_dev->host_segnr,
-		assigned_dev->host_busnr,
-		PCI_SLOT(assigned_dev->host_devfn),
-		PCI_FUNC(assigned_dev->host_devfn));
+	dev_info(&pdev->dev, "kvm assign device\n");
 
 	return 0;
 out_unmap:
@@ -220,11 +222,7 @@ int kvm_deassign_device(struct kvm *kvm,
 
 	pdev->dev_flags &= ~PCI_DEV_FLAGS_ASSIGNED;
 
-	printk(KERN_DEBUG "deassign device %x:%x:%x.%x\n",
-		assigned_dev->host_segnr,
-		assigned_dev->host_busnr,
-		PCI_SLOT(assigned_dev->host_devfn),
-		PCI_FUNC(assigned_dev->host_devfn));
+	dev_info(&pdev->dev, "kvm deassign device\n");
 
 	return 0;
 }
@@ -336,6 +334,9 @@ static int kvm_iommu_unmap_memslots(struct kvm *kvm)
 
 	srcu_read_unlock(&kvm->srcu, idx);
 
+	if (kvm->arch.iommu_noncoherent)
+		kvm_arch_unregister_noncoherent_dma(kvm);
+
 	return 0;
 }
 
@@ -350,6 +351,7 @@ int kvm_iommu_unmap_guest(struct kvm *kvm)
 	mutex_lock(&kvm->slots_lock);
 	kvm_iommu_unmap_memslots(kvm);
 	kvm->arch.iommu_domain = NULL;
+	kvm->arch.iommu_noncoherent = false;
 	mutex_unlock(&kvm->slots_lock);
 
 	iommu_domain_free(domain);
diff --git a/virt/kvm/kvm_main.c b/virt/kvm/kvm_main.c
index 1cf9ccb01013..662f34c3287e 100644
--- a/virt/kvm/kvm_main.c
+++ b/virt/kvm/kvm_main.c
@@ -70,7 +70,8 @@ MODULE_LICENSE("GPL");
  * 		kvm->lock --> kvm->slots_lock --> kvm->irq_lock
  */
 
-DEFINE_RAW_SPINLOCK(kvm_lock);
+DEFINE_SPINLOCK(kvm_lock);
+static DEFINE_RAW_SPINLOCK(kvm_count_lock);
 LIST_HEAD(vm_list);
 
 static cpumask_var_t cpus_hardware_enabled;
@@ -186,6 +187,7 @@ void kvm_flush_remote_tlbs(struct kvm *kvm)
 		++kvm->stat.remote_tlb_flush;
 	cmpxchg(&kvm->tlbs_dirty, dirty_count, 0);
 }
+EXPORT_SYMBOL_GPL(kvm_flush_remote_tlbs);
 
 void kvm_reload_remote_mmus(struct kvm *kvm)
 {
@@ -490,9 +492,9 @@ static struct kvm *kvm_create_vm(unsigned long type)
 	if (r)
 		goto out_err;
 
-	raw_spin_lock(&kvm_lock);
+	spin_lock(&kvm_lock);
 	list_add(&kvm->vm_list, &vm_list);
-	raw_spin_unlock(&kvm_lock);
+	spin_unlock(&kvm_lock);
 
 	return kvm;
 
@@ -540,13 +542,13 @@ static void kvm_destroy_dirty_bitmap(struct kvm_memory_slot *memslot)
 /*
  * Free any memory in @free but not in @dont.
  */
-static void kvm_free_physmem_slot(struct kvm_memory_slot *free,
+static void kvm_free_physmem_slot(struct kvm *kvm, struct kvm_memory_slot *free,
 				  struct kvm_memory_slot *dont)
 {
 	if (!dont || free->dirty_bitmap != dont->dirty_bitmap)
 		kvm_destroy_dirty_bitmap(free);
 
-	kvm_arch_free_memslot(free, dont);
+	kvm_arch_free_memslot(kvm, free, dont);
 
 	free->npages = 0;
 }
@@ -557,7 +559,7 @@ void kvm_free_physmem(struct kvm *kvm)
 	struct kvm_memory_slot *memslot;
 
 	kvm_for_each_memslot(memslot, slots)
-		kvm_free_physmem_slot(memslot, NULL);
+		kvm_free_physmem_slot(kvm, memslot, NULL);
 
 	kfree(kvm->memslots);
 }
@@ -581,9 +583,9 @@ static void kvm_destroy_vm(struct kvm *kvm)
 	struct mm_struct *mm = kvm->mm;
 
 	kvm_arch_sync_events(kvm);
-	raw_spin_lock(&kvm_lock);
+	spin_lock(&kvm_lock);
 	list_del(&kvm->vm_list);
-	raw_spin_unlock(&kvm_lock);
+	spin_unlock(&kvm_lock);
 	kvm_free_irq_routing(kvm);
 	for (i = 0; i < KVM_NR_BUSES; i++)
 		kvm_io_bus_destroy(kvm->buses[i]);
@@ -821,7 +823,7 @@ int __kvm_set_memory_region(struct kvm *kvm,
 	if (change == KVM_MR_CREATE) {
 		new.userspace_addr = mem->userspace_addr;
 
-		if (kvm_arch_create_memslot(&new, npages))
+		if (kvm_arch_create_memslot(kvm, &new, npages))
 			goto out_free;
 	}
 
@@ -872,6 +874,19 @@ int __kvm_set_memory_region(struct kvm *kvm,
 			goto out_free;
 	}
 
+	/* actual memory is freed via old in kvm_free_physmem_slot below */
+	if (change == KVM_MR_DELETE) {
+		new.dirty_bitmap = NULL;
+		memset(&new.arch, 0, sizeof(new.arch));
+	}
+
+	old_memslots = install_new_memslots(kvm, slots, &new);
+
+	kvm_arch_commit_memory_region(kvm, mem, &old, change);
+
+	kvm_free_physmem_slot(kvm, &old, &new);
+	kfree(old_memslots);
+
 	/*
 	 * IOMMU mapping:  New slots need to be mapped.  Old slots need to be
 	 * un-mapped and re-mapped if their base changes.  Since base change
@@ -883,29 +898,15 @@ int __kvm_set_memory_region(struct kvm *kvm,
 	 */
 	if ((change == KVM_MR_CREATE) || (change == KVM_MR_MOVE)) {
 		r = kvm_iommu_map_pages(kvm, &new);
-		if (r)
-			goto out_slots;
-	}
-
-	/* actual memory is freed via old in kvm_free_physmem_slot below */
-	if (change == KVM_MR_DELETE) {
-		new.dirty_bitmap = NULL;
-		memset(&new.arch, 0, sizeof(new.arch));
+		return r;
 	}
 
-	old_memslots = install_new_memslots(kvm, slots, &new);
-
-	kvm_arch_commit_memory_region(kvm, mem, &old, change);
-
-	kvm_free_physmem_slot(&old, &new);
-	kfree(old_memslots);
-
 	return 0;
 
 out_slots:
 	kfree(slots);
 out_free:
-	kvm_free_physmem_slot(&new, &old);
+	kvm_free_physmem_slot(kvm, &new, &old);
 out:
 	return r;
 }
@@ -964,6 +965,7 @@ int kvm_get_dirty_log(struct kvm *kvm,
 out:
 	return r;
 }
+EXPORT_SYMBOL_GPL(kvm_get_dirty_log);
 
 bool kvm_largepages_enabled(void)
 {
@@ -1654,6 +1656,7 @@ void mark_page_dirty(struct kvm *kvm, gfn_t gfn)
 	memslot = gfn_to_memslot(kvm, gfn);
 	mark_page_dirty_in_slot(kvm, memslot, gfn);
 }
+EXPORT_SYMBOL_GPL(mark_page_dirty);
 
 /*
  * The vCPU has executed a HLT instruction with in-kernel mode enabled.
@@ -1679,6 +1682,7 @@ void kvm_vcpu_block(struct kvm_vcpu *vcpu)
 
 	finish_wait(&vcpu->wq, &wait);
 }
+EXPORT_SYMBOL_GPL(kvm_vcpu_block);
 
 #ifndef CONFIG_S390
 /*
@@ -2271,6 +2275,11 @@ static int kvm_ioctl_create_device(struct kvm *kvm,
 		ops = &kvm_xics_ops;
 		break;
 #endif
+#ifdef CONFIG_KVM_VFIO
+	case KVM_DEV_TYPE_VFIO:
+		ops = &kvm_vfio_ops;
+		break;
+#endif
 	default:
 		return -ENODEV;
 	}
@@ -2519,44 +2528,12 @@ out:
 }
 #endif
 
-static int kvm_vm_fault(struct vm_area_struct *vma, struct vm_fault *vmf)
-{
-	struct page *page[1];
-	unsigned long addr;
-	int npages;
-	gfn_t gfn = vmf->pgoff;
-	struct kvm *kvm = vma->vm_file->private_data;
-
-	addr = gfn_to_hva(kvm, gfn);
-	if (kvm_is_error_hva(addr))
-		return VM_FAULT_SIGBUS;
-
-	npages = get_user_pages(current, current->mm, addr, 1, 1, 0, page,
-				NULL);
-	if (unlikely(npages != 1))
-		return VM_FAULT_SIGBUS;
-
-	vmf->page = page[0];
-	return 0;
-}
-
-static const struct vm_operations_struct kvm_vm_vm_ops = {
-	.fault = kvm_vm_fault,
-};
-
-static int kvm_vm_mmap(struct file *file, struct vm_area_struct *vma)
-{
-	vma->vm_ops = &kvm_vm_vm_ops;
-	return 0;
-}
-
 static struct file_operations kvm_vm_fops = {
 	.release        = kvm_vm_release,
 	.unlocked_ioctl = kvm_vm_ioctl,
 #ifdef CONFIG_COMPAT
 	.compat_ioctl   = kvm_vm_compat_ioctl,
 #endif
-	.mmap           = kvm_vm_mmap,
 	.llseek		= noop_llseek,
 };
 
@@ -2683,11 +2660,12 @@ static void hardware_enable_nolock(void *junk)
 	}
 }
 
-static void hardware_enable(void *junk)
+static void hardware_enable(void)
 {
-	raw_spin_lock(&kvm_lock);
-	hardware_enable_nolock(junk);
-	raw_spin_unlock(&kvm_lock);
+	raw_spin_lock(&kvm_count_lock);
+	if (kvm_usage_count)
+		hardware_enable_nolock(NULL);
+	raw_spin_unlock(&kvm_count_lock);
 }
 
 static void hardware_disable_nolock(void *junk)
@@ -2700,11 +2678,12 @@ static void hardware_disable_nolock(void *junk)
 	kvm_arch_hardware_disable(NULL);
 }
 
-static void hardware_disable(void *junk)
+static void hardware_disable(void)
 {
-	raw_spin_lock(&kvm_lock);
-	hardware_disable_nolock(junk);
-	raw_spin_unlock(&kvm_lock);
+	raw_spin_lock(&kvm_count_lock);
+	if (kvm_usage_count)
+		hardware_disable_nolock(NULL);
+	raw_spin_unlock(&kvm_count_lock);
 }
 
 static void hardware_disable_all_nolock(void)
@@ -2718,16 +2697,16 @@ static void hardware_disable_all_nolock(void)
 
 static void hardware_disable_all(void)
 {
-	raw_spin_lock(&kvm_lock);
+	raw_spin_lock(&kvm_count_lock);
 	hardware_disable_all_nolock();
-	raw_spin_unlock(&kvm_lock);
+	raw_spin_unlock(&kvm_count_lock);
 }
 
 static int hardware_enable_all(void)
 {
 	int r = 0;
 
-	raw_spin_lock(&kvm_lock);
+	raw_spin_lock(&kvm_count_lock);
 
 	kvm_usage_count++;
 	if (kvm_usage_count == 1) {
@@ -2740,7 +2719,7 @@ static int hardware_enable_all(void)
 		}
 	}
 
-	raw_spin_unlock(&kvm_lock);
+	raw_spin_unlock(&kvm_count_lock);
 
 	return r;
 }
@@ -2750,20 +2729,17 @@ static int kvm_cpu_hotplug(struct notifier_block *notifier, unsigned long val,
 {
 	int cpu = (long)v;
 
-	if (!kvm_usage_count)
-		return NOTIFY_OK;
-
 	val &= ~CPU_TASKS_FROZEN;
 	switch (val) {
 	case CPU_DYING:
 		printk(KERN_INFO "kvm: disabling virtualization on CPU%d\n",
 		       cpu);
-		hardware_disable(NULL);
+		hardware_disable();
 		break;
 	case CPU_STARTING:
 		printk(KERN_INFO "kvm: enabling virtualization on CPU%d\n",
 		       cpu);
-		hardware_enable(NULL);
+		hardware_enable();
 		break;
 	}
 	return NOTIFY_OK;
@@ -3056,10 +3032,10 @@ static int vm_stat_get(void *_offset, u64 *val)
 	struct kvm *kvm;
 
 	*val = 0;
-	raw_spin_lock(&kvm_lock);
+	spin_lock(&kvm_lock);
 	list_for_each_entry(kvm, &vm_list, vm_list)
 		*val += *(u32 *)((void *)kvm + offset);
-	raw_spin_unlock(&kvm_lock);
+	spin_unlock(&kvm_lock);
 	return 0;
 }
 
@@ -3073,12 +3049,12 @@ static int vcpu_stat_get(void *_offset, u64 *val)
 	int i;
 
 	*val = 0;
-	raw_spin_lock(&kvm_lock);
+	spin_lock(&kvm_lock);
 	list_for_each_entry(kvm, &vm_list, vm_list)
 		kvm_for_each_vcpu(i, vcpu, kvm)
 			*val += *(u32 *)((void *)vcpu + offset);
 
-	raw_spin_unlock(&kvm_lock);
+	spin_unlock(&kvm_lock);
 	return 0;
 }
 
@@ -3133,7 +3109,7 @@ static int kvm_suspend(void)
 static void kvm_resume(void)
 {
 	if (kvm_usage_count) {
-		WARN_ON(raw_spin_is_locked(&kvm_lock));
+		WARN_ON(raw_spin_is_locked(&kvm_count_lock));
 		hardware_enable_nolock(NULL);
 	}
 }
diff --git a/virt/kvm/vfio.c b/virt/kvm/vfio.c
new file mode 100644
index 000000000000..ca4260e35037
--- /dev/null
+++ b/virt/kvm/vfio.c
@@ -0,0 +1,264 @@
+/*
+ * VFIO-KVM bridge pseudo device
+ *
+ * Copyright (C) 2013 Red Hat, Inc.  All rights reserved.
+ *     Author: Alex Williamson <alex.williamson@redhat.com>
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ */
+
+#include <linux/errno.h>
+#include <linux/file.h>
+#include <linux/kvm_host.h>
+#include <linux/list.h>
+#include <linux/module.h>
+#include <linux/mutex.h>
+#include <linux/slab.h>
+#include <linux/uaccess.h>
+#include <linux/vfio.h>
+
+struct kvm_vfio_group {
+	struct list_head node;
+	struct vfio_group *vfio_group;
+};
+
+struct kvm_vfio {
+	struct list_head group_list;
+	struct mutex lock;
+	bool noncoherent;
+};
+
+static struct vfio_group *kvm_vfio_group_get_external_user(struct file *filep)
+{
+	struct vfio_group *vfio_group;
+	struct vfio_group *(*fn)(struct file *);
+
+	fn = symbol_get(vfio_group_get_external_user);
+	if (!fn)
+		return ERR_PTR(-EINVAL);
+
+	vfio_group = fn(filep);
+
+	symbol_put(vfio_group_get_external_user);
+
+	return vfio_group;
+}
+
+static void kvm_vfio_group_put_external_user(struct vfio_group *vfio_group)
+{
+	void (*fn)(struct vfio_group *);
+
+	fn = symbol_get(vfio_group_put_external_user);
+	if (!fn)
+		return;
+
+	fn(vfio_group);
+
+	symbol_put(vfio_group_put_external_user);
+}
+
+/*
+ * Groups can use the same or different IOMMU domains.  If the same then
+ * adding a new group may change the coherency of groups we've previously
+ * been told about.  We don't want to care about any of that so we retest
+ * each group and bail as soon as we find one that's noncoherent.  This
+ * means we only ever [un]register_noncoherent_dma once for the whole device.
+ */
+static void kvm_vfio_update_coherency(struct kvm_device *dev)
+{
+	struct kvm_vfio *kv = dev->private;
+	bool noncoherent = false;
+	struct kvm_vfio_group *kvg;
+
+	mutex_lock(&kv->lock);
+
+	list_for_each_entry(kvg, &kv->group_list, node) {
+		/*
+		 * TODO: We need an interface to check the coherency of
+		 * the IOMMU domain this group is using.  For now, assume
+		 * it's always noncoherent.
+		 */
+		noncoherent = true;
+		break;
+	}
+
+	if (noncoherent != kv->noncoherent) {
+		kv->noncoherent = noncoherent;
+
+		if (kv->noncoherent)
+			kvm_arch_register_noncoherent_dma(dev->kvm);
+		else
+			kvm_arch_unregister_noncoherent_dma(dev->kvm);
+	}
+
+	mutex_unlock(&kv->lock);
+}
+
+static int kvm_vfio_set_group(struct kvm_device *dev, long attr, u64 arg)
+{
+	struct kvm_vfio *kv = dev->private;
+	struct vfio_group *vfio_group;
+	struct kvm_vfio_group *kvg;
+	void __user *argp = (void __user *)arg;
+	struct fd f;
+	int32_t fd;
+	int ret;
+
+	switch (attr) {
+	case KVM_DEV_VFIO_GROUP_ADD:
+		if (get_user(fd, (int32_t __user *)argp))
+			return -EFAULT;
+
+		f = fdget(fd);
+		if (!f.file)
+			return -EBADF;
+
+		vfio_group = kvm_vfio_group_get_external_user(f.file);
+		fdput(f);
+
+		if (IS_ERR(vfio_group))
+			return PTR_ERR(vfio_group);
+
+		mutex_lock(&kv->lock);
+
+		list_for_each_entry(kvg, &kv->group_list, node) {
+			if (kvg->vfio_group == vfio_group) {
+				mutex_unlock(&kv->lock);
+				kvm_vfio_group_put_external_user(vfio_group);
+				return -EEXIST;
+			}
+		}
+
+		kvg = kzalloc(sizeof(*kvg), GFP_KERNEL);
+		if (!kvg) {
+			mutex_unlock(&kv->lock);
+			kvm_vfio_group_put_external_user(vfio_group);
+			return -ENOMEM;
+		}
+
+		list_add_tail(&kvg->node, &kv->group_list);
+		kvg->vfio_group = vfio_group;
+
+		mutex_unlock(&kv->lock);
+
+		kvm_vfio_update_coherency(dev);
+
+		return 0;
+
+	case KVM_DEV_VFIO_GROUP_DEL:
+		if (get_user(fd, (int32_t __user *)argp))
+			return -EFAULT;
+
+		f = fdget(fd);
+		if (!f.file)
+			return -EBADF;
+
+		vfio_group = kvm_vfio_group_get_external_user(f.file);
+		fdput(f);
+
+		if (IS_ERR(vfio_group))
+			return PTR_ERR(vfio_group);
+
+		ret = -ENOENT;
+
+		mutex_lock(&kv->lock);
+
+		list_for_each_entry(kvg, &kv->group_list, node) {
+			if (kvg->vfio_group != vfio_group)
+				continue;
+
+			list_del(&kvg->node);
+			kvm_vfio_group_put_external_user(kvg->vfio_group);
+			kfree(kvg);
+			ret = 0;
+			break;
+		}
+
+		mutex_unlock(&kv->lock);
+
+		kvm_vfio_group_put_external_user(vfio_group);
+
+		kvm_vfio_update_coherency(dev);
+
+		return ret;
+	}
+
+	return -ENXIO;
+}
+
+static int kvm_vfio_set_attr(struct kvm_device *dev,
+			     struct kvm_device_attr *attr)
+{
+	switch (attr->group) {
+	case KVM_DEV_VFIO_GROUP:
+		return kvm_vfio_set_group(dev, attr->attr, attr->addr);
+	}
+
+	return -ENXIO;
+}
+
+static int kvm_vfio_has_attr(struct kvm_device *dev,
+			     struct kvm_device_attr *attr)
+{
+	switch (attr->group) {
+	case KVM_DEV_VFIO_GROUP:
+		switch (attr->attr) {
+		case KVM_DEV_VFIO_GROUP_ADD:
+		case KVM_DEV_VFIO_GROUP_DEL:
+			return 0;
+		}
+
+		break;
+	}
+
+	return -ENXIO;
+}
+
+static void kvm_vfio_destroy(struct kvm_device *dev)
+{
+	struct kvm_vfio *kv = dev->private;
+	struct kvm_vfio_group *kvg, *tmp;
+
+	list_for_each_entry_safe(kvg, tmp, &kv->group_list, node) {
+		kvm_vfio_group_put_external_user(kvg->vfio_group);
+		list_del(&kvg->node);
+		kfree(kvg);
+	}
+
+	kvm_vfio_update_coherency(dev);
+
+	kfree(kv);
+	kfree(dev); /* alloc by kvm_ioctl_create_device, free by .destroy */
+}
+
+static int kvm_vfio_create(struct kvm_device *dev, u32 type)
+{
+	struct kvm_device *tmp;
+	struct kvm_vfio *kv;
+
+	/* Only one VFIO "device" per VM */
+	list_for_each_entry(tmp, &dev->kvm->devices, vm_node)
+		if (tmp->ops == &kvm_vfio_ops)
+			return -EBUSY;
+
+	kv = kzalloc(sizeof(*kv), GFP_KERNEL);
+	if (!kv)
+		return -ENOMEM;
+
+	INIT_LIST_HEAD(&kv->group_list);
+	mutex_init(&kv->lock);
+
+	dev->private = kv;
+
+	return 0;
+}
+
+struct kvm_device_ops kvm_vfio_ops = {
+	.name = "kvm-vfio",
+	.create = kvm_vfio_create,
+	.destroy = kvm_vfio_destroy,
+	.set_attr = kvm_vfio_set_attr,
+	.has_attr = kvm_vfio_has_attr,
+};