| Commit message (Collapse) | Author | Age | Files | Lines |
|---|
| |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
This is intended for use with NVDIMMs that are physically persistent
(physically like flash) so that they can be used as a cost-effective
RAM replacement. Intel Optane DC persistent memory is one
implementation of this kind of NVDIMM.
Currently, a persistent memory region is "owned" by a device driver,
either the "Direct DAX" or "Filesystem DAX" drivers. These drivers
allow applications to explicitly use persistent memory, generally
by being modified to use special, new libraries. (DIMM-based
persistent memory hardware/software is described in great detail
here: Documentation/nvdimm/nvdimm.txt).
However, this limits persistent memory use to applications which
*have* been modified. To make it more broadly usable, this driver
"hotplugs" memory into the kernel, to be managed and used just like
normal RAM would be.
To make this work, management software must remove the device from
being controlled by the "Device DAX" infrastructure:
echo dax0.0 > /sys/bus/dax/drivers/device_dax/unbind
and then tell the new driver that it can bind to the device:
echo dax0.0 > /sys/bus/dax/drivers/kmem/new_id
After this, there will be a number of new memory sections visible
in sysfs that can be onlined, or that may get onlined by existing
udev-initiated memory hotplug rules.
This rebinding procedure is currently a one-way trip. Once memory
is bound to "kmem", it's there permanently and can not be
unbound and assigned back to device_dax.
The kmem driver will never bind to a dax device unless the device
is *explicitly* bound to the driver. There are two reasons for
this: One, since it is a one-way trip, it can not be undone if
bound incorrectly. Two, the kmem driver destroys data on the
device. Think of if you had good data on a pmem device. It
would be catastrophic if you compile-in "kmem", but leave out
the "device_dax" driver. kmem would take over the device and
write volatile data all over your good data.
This inherits any existing NUMA information for the newly-added
memory from the persistent memory device that came from the
firmware. On Intel platforms, the firmware has guarantees that
require each socket's persistent memory to be in a separate
memory-only NUMA node. That means that this patch is not expected
to create NUMA nodes, but will simply hotplug memory into existing
nodes.
Because NUMA nodes are created, the existing NUMA APIs and tools
are sufficient to create policies for applications or memory areas
to have affinity for or an aversion to using this memory.
There is currently some metadata at the beginning of pmem regions.
The section-size memory hotplug restrictions, plus this small
reserved area can cause the "loss" of a section or two of capacity.
This should be fixable in follow-on patches. But, as a first step,
losing 256MB of memory (worst case) out of hundreds of gigabytes
is a good tradeoff vs. the required code to fix this up precisely.
This calculation is also the reason we export
memory_block_size_bytes().
Signed-off-by: Dave Hansen <dave.hansen@linux.intel.com>
Reviewed-by: Dan Williams <dan.j.williams@intel.com>
Reviewed-by: Keith Busch <keith.busch@intel.com>
Cc: Dave Jiang <dave.jiang@intel.com>
Cc: Ross Zwisler <zwisler@kernel.org>
Cc: Vishal Verma <vishal.l.verma@intel.com>
Cc: Tom Lendacky <thomas.lendacky@amd.com>
Cc: Andrew Morton <akpm@linux-foundation.org>
Cc: Michal Hocko <mhocko@suse.com>
Cc: linux-nvdimm@lists.01.org
Cc: linux-kernel@vger.kernel.org
Cc: linux-mm@kvack.org
Cc: Huang Ying <ying.huang@intel.com>
Cc: Fengguang Wu <fengguang.wu@intel.com>
Cc: Borislav Petkov <bp@suse.de>
Cc: Bjorn Helgaas <bhelgaas@google.com>
Cc: Yaowei Bai <baiyaowei@cmss.chinamobile.com>
Cc: Takashi Iwai <tiwai@suse.de>
Cc: Jerome Glisse <jglisse@redhat.com>
Reviewed-by: Vishal Verma <vishal.l.verma@intel.com>
Signed-off-by: Dan Williams <dan.j.williams@intel.com>
|
| |
|
|
|
|
|
|
|
|
|
|
|
|
| |
On the expectation that some environments may not upgrade libdaxctl
(userspace component that depends on the /sys/class/dax hierarchy),
provide a default / legacy dax_pmem_compat driver. The dax_pmem_compat
driver implements the original /sys/class/dax sysfs layout rather than
/sys/bus/dax. When userspace is upgraded it can blacklist this module
and switch to the dax_pmem driver going forward.
CONFIG_DEV_DAX_PMEM_COMPAT and supporting code will be deleted according
to the dax_pmem entry in Documentation/ABI/obsolete/.
Signed-off-by: Dan Williams <dan.j.williams@intel.com>
|
| |
|
|
|
|
|
|
|
|
|
|
|
|
| |
In support of allowing device-mapper to compile out idle/dead code when
there are no dax providers in the system, introduce the DAX_DRIVER
symbol. This is selected by all leaf drivers that device-mapper might be
layered on top. This allows device-mapper to conditionally 'select DAX'
only when a provider is present.
Cc: Martin Schwidefsky <schwidefsky@de.ibm.com>
Cc: Heiko Carstens <heiko.carstens@de.ibm.com>
Reported-by: Bart Van Assche <Bart.VanAssche@wdc.com>
Reviewed-by: Mike Snitzer <snitzer@redhat.com>
Signed-off-by: Dan Williams <dan.j.williams@intel.com>
|
| |
|
|
|
|
|
|
|
|
|
|
|
| |
There is no point to ask how many device-dax instances the kernel should
support. Since we are already using a dynamic major number, just allow
the max number of minors by default and be done. This also fixes the
fact that the proposed max for the NR_DEV_DAX range was larger than what
could be supported by alloc_chrdev_region().
Fixes: ba09c01d2fa8 ("dax: convert to the cdev api")
Reported-by: Geert Uytterhoeven <geert@linux-m68k.org>
Tested-by: Geert Uytterhoeven <geert@linux-m68k.org>
Signed-off-by: Dan Williams <dan.j.williams@intel.com>
|
| |
|
|
|
|
|
|
|
| |
ERROR: "devm_create_dev_dax" [drivers/dax/dax_pmem.ko] undefined!
ERROR: "alloc_dax_region" [drivers/dax/dax_pmem.ko] undefined!
ERROR: "dax_region_put" [drivers/dax/dax_pmem.ko] undefined!
Signed-off-by: Mike Galbraith <efault@gmx.de>
Signed-off-by: Dan Williams <dan.j.williams@intel.com>
|
| |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
We want dax capable drivers to be able to publish a set of dax
operations [1]. However, we do not want to further abuse block_devices
to advertise these operations. Instead we will attach these operations
to a dax device and add a lookup mechanism to go from block device path
to a dax device. A dax capable driver like pmem or brd is responsible
for registering a dax device, alongside a block device, and then a dax
capable filesystem is responsible for retrieving the dax device by path
name if it wants to call dax_operations.
For now, we refactor the dax pseudo-fs to be a generic facility, rather
than an implementation detail, of the device-dax use case. Where a "dax
device" is just an inode + dax infrastructure, and "Device DAX" is a
mapping service layered on top of that base 'struct dax_device'.
"Filesystem DAX" is then a mapping service that layers a filesystem on
top of that same base device. Filesystem DAX is associated with a
block_device for now, but perhaps directly to a dax device in the
future, or for new pmem-only filesystems.
[1]: https://lkml.org/lkml/2017/1/19/880
Suggested-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Dan Williams <dan.j.williams@intel.com>
|
| |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
The following warning triggers with a new unit test that stresses the
device-dax interface.
===============================
[ ERR: suspicious RCU usage. ]
4.11.0-rc4+ #1049 Tainted: G O
-------------------------------
./include/linux/rcupdate.h:521 Illegal context switch in RCU read-side critical section!
other info that might help us debug this:
rcu_scheduler_active = 2, debug_locks = 0
2 locks held by fio/9070:
#0: (&mm->mmap_sem){++++++}, at: [<ffffffff8d0739d7>] __do_page_fault+0x167/0x4f0
#1: (rcu_read_lock){......}, at: [<ffffffffc03fbd02>] dax_dev_huge_fault+0x32/0x620 [dax]
Call Trace:
dump_stack+0x86/0xc3
lockdep_rcu_suspicious+0xd7/0x110
___might_sleep+0xac/0x250
__might_sleep+0x4a/0x80
__alloc_pages_nodemask+0x23a/0x360
alloc_pages_current+0xa1/0x1f0
pte_alloc_one+0x17/0x80
__pte_alloc+0x1e/0x120
__get_locked_pte+0x1bf/0x1d0
insert_pfn.isra.70+0x3a/0x100
? lookup_memtype+0xa6/0xd0
vm_insert_mixed+0x64/0x90
dax_dev_huge_fault+0x520/0x620 [dax]
? dax_dev_huge_fault+0x32/0x620 [dax]
dax_dev_fault+0x10/0x20 [dax]
__do_fault+0x1e/0x140
__handle_mm_fault+0x9af/0x10d0
handle_mm_fault+0x16d/0x370
? handle_mm_fault+0x47/0x370
__do_page_fault+0x28c/0x4f0
trace_do_page_fault+0x58/0x2a0
do_async_page_fault+0x1a/0xa0
async_page_fault+0x28/0x30
Inserting a page table entry may trigger an allocation while we are
holding a read lock to keep the device instance alive for the duration
of the fault. Use srcu for this keep-alive protection.
Fixes: dee410792419 ("/dev/dax, core: file operations and dax-mmap")
Cc: <stable@vger.kernel.org>
Signed-off-by: Dan Williams <dan.j.williams@intel.com>
|
| |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
A bugfix just tried to address a randconfig build problem and introduced
a variant of the same problem: with CONFIG_LIBNVDIMM=y and
CONFIG_NVDIMM_DAX=m, the nvdimm module now fails to link:
drivers/nvdimm/built-in.o: In function `to_nd_device_type':
bus.c:(.text+0x1b5d): undefined reference to `is_nd_dax'
drivers/nvdimm/built-in.o: In function `nd_region_notify_driver_action.constprop.2':
region_devs.c:(.text+0x6b6c): undefined reference to `is_nd_dax'
region_devs.c:(.text+0x6b8c): undefined reference to `to_nd_dax'
drivers/nvdimm/built-in.o: In function `nd_region_probe':
region.c:(.text+0x70f3): undefined reference to `nd_dax_create'
drivers/nvdimm/built-in.o: In function `mode_show':
namespace_devs.c:(.text+0xa196): undefined reference to `is_nd_dax'
drivers/nvdimm/built-in.o: In function `nvdimm_namespace_common_probe':
(.text+0xa55f): undefined reference to `is_nd_dax'
drivers/nvdimm/built-in.o: In function `nvdimm_namespace_common_probe':
(.text+0xa56e): undefined reference to `to_nd_dax'
This reverts the earlier fix, making NVDIMM_DAX a 'bool' option again
as it should be (it gets linked into the libnvdimm module). To fix
the original problem, I'm adding a dependency on LIBNVDIMM to
DEV_DAX_PMEM, which ensures we can't have that one built-in if the
rest is a module.
Fixes: 4e65e9381c7a ("/dev/dax: fix Kconfig dependency build breakage")
Signed-off-by: Arnd Bergmann <arnd@arndb.de>
Reviewed-by: Ross Zwisler <ross.zwisler@linux.intel.com>
Signed-off-by: Dan Williams <dan.j.williams@intel.com>
|
| |
|
|
|
|
|
|
|
|
|
|
|
|
| |
A goal of the device-DAX interface is to be able to support many
exclusive allocations (partitions) of performance / feature
differentiated memory. This count may exceed the default minors limit
of 256.
As a result of switching to an embedded cdev the inode-to-dax_dev
conversion is simplified, as well as reference counting which can switch
to the cdev kobject lifetime.
Cc: Al Viro <viro@zeniv.linux.org.uk>
Signed-off-by: Dan Williams <dan.j.williams@intel.com>
|
| |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
The "Device DAX" core enables dax mappings of performance / feature
differentiated memory. An open mapping or file handle keeps the backing
struct device live, but new mappings are only possible while the device
is enabled. Faults are handled under rcu_read_lock to synchronize
with the enabled state of the device.
Similar to the filesystem-dax case the backing memory may optionally
have struct page entries. However, unlike fs-dax there is no support
for private mappings, or mappings that are not backed by media (see
use of zero-page in fs-dax).
Mappings are always guaranteed to match the alignment of the dax_region.
If the dax_region is configured to have a 2MB alignment, all mappings
are guaranteed to be backed by a pmd entry. Contrast this determinism
with the fs-dax case where pmd mappings are opportunistic. If userspace
attempts to force a misaligned mapping, the driver will fail the mmap
attempt. See dax_dev_check_vma() for other scenarios that are rejected,
like MAP_PRIVATE mappings.
Cc: Hannes Reinecke <hare@suse.de>
Cc: Jeff Moyer <jmoyer@redhat.com>
Cc: Christoph Hellwig <hch@lst.de>
Cc: Andrew Morton <akpm@linux-foundation.org>
Cc: Dave Hansen <dave.hansen@linux.intel.com>
Cc: Ross Zwisler <ross.zwisler@linux.intel.com>
Acked-by: "Paul E. McKenney" <paulmck@linux.vnet.ibm.com>
Reviewed-by: Johannes Thumshirn <jthumshirn@suse.de>
Signed-off-by: Dan Williams <dan.j.williams@intel.com>
|
|
|
Device DAX is the device-centric analogue of Filesystem DAX
(CONFIG_FS_DAX). It allows memory ranges to be allocated and mapped
without need of an intervening file system. Device DAX is strict,
precise and predictable. Specifically this interface:
1/ Guarantees fault granularity with respect to a given page size (pte,
pmd, or pud) set at configuration time.
2/ Enforces deterministic behavior by being strict about what fault
scenarios are supported.
For example, by forcing MADV_DONTFORK semantics and omitting MAP_PRIVATE
support device-dax guarantees that a mapping always behaves/performs the
same once established. It is the "what you see is what you get" access
mechanism to differentiated memory vs filesystem DAX which has
filesystem specific implementation semantics.
Persistent memory is the first target, but the mechanism is also
targeted for exclusive allocations of performance differentiated memory
ranges.
This commit is limited to the base device driver infrastructure to
associate a dax device with pmem range.
Cc: Jeff Moyer <jmoyer@redhat.com>
Cc: Christoph Hellwig <hch@lst.de>
Cc: Andrew Morton <akpm@linux-foundation.org>
Cc: Dave Hansen <dave.hansen@linux.intel.com>
Cc: Ross Zwisler <ross.zwisler@linux.intel.com>
Reviewed-by: Johannes Thumshirn <jthumshirn@suse.de>
Signed-off-by: Dan Williams <dan.j.williams@intel.com>
|
